international response to infectious salmon anemia: prevention

United StatesDepartment ofAgriculture

Animal andPlant HealthInspectionService

United StatesDepartment ofthe Interior

US GeologicalSurvey

United StatesDepartment ofCommerce

National MarineFisheries Service

TechnicalBulletin No 1902

International Responseto Infectious SalmonAnemia PreventionControl and Eradication

The US Departments of Agriculture (USDA) the Interior (USDI)and Commerce prohibit discrimination in all their programs andactivities on the basis of race color national origin sex religionage disability political beliefs sexual orientation or marital orfamily status (Not all prohibited bases apply to all programs)Persons with disabilities who require alternative means forcommunication of program information (Braille large printaudiotape etc) should contact USDArsquos TARGET Center at (202)720ndash2600 (voice and TDD)

To file a complaint of discrimination write USDA Director Office ofCivil Rights Room 326ndashW Whitten Building 1400 IndependenceAvenue SW Washington DC 20250ndash9410 or call (202) 720ndash5964(voice and TDD) USDA USDI and Commerce are equalopportunity providers and employers

The opinions expressed by individuals in this report do notnecessarily represent the policies of USDA USDI or Commerce

Mention of companies or commercial products does not implyrecommendation or endorsement by USDA USDI or Commerceover others not mentioned The Federal Government neitherguarantees nor warrants the standard of any product mentionedProduct names are mentioned solely to report factually on availabledata and to provide specific information

Photo credits The background illustration on the front cover wassupplied as a photo micrograph by Michael Opitz of the Universityof Maine and is reproduced by permission The line art of salmoncame from the 1973 book ldquoThe Salmon Their Fight for Survivalrdquo byAnthony Netboy Houghton Mifflin Company is the copyright holderon the line art which is reproduced by permission Images insidethe proceedings were supplied by the senior authors For reproduc-tion rights please consult each senior author using contactinformation in appendix 1 ldquoAuthorsrsquo Affiliationsrdquo

Issued April 2003




US GeologicalSurvey





Proceedings of a SymposiumNew Orleans LASeptember 3ndash4 2002

Otis Miller DVM MS andRocco C Cipriano PhDTechnical Coordinators1

1Dr Miller is a senior staff veterinarian and national aquaculture coordinator with the USDAndashAnimaland Plant Health Inspection Servicersquos Veterinary Services in Riverdale MD Dr Cipriano is seniorresearch microbiologist with the USDIndashUS Geological Survey National Fish Health ResearchLaboratory in Kearneysville WV

i

Suggested citation for the book using the bibliographic style of the American NationalStandards Institute

Miller Otis Cipriano Rocco C tech coords 2003 International response to infectioussalmon anemia prevention control and eradication proceedings of a symposium 3ndash4September 2002 New Orleans LA Tech Bull 1902 Washington DC US Department ofAgriculture Animal and Plant Health Inspection Service US Department of the Interior USGeological Survey US Department of Commerce National Marine Fisheries Service 194 p

ii

The suppression of infectious salmon anemia (ISA) is key to protectingaquaculture the fastest growing segment of the US agricultural economyGovernment stewards of the salmon resource of the North Atlantic mustcommunicate and cooperate to eradicate ISA before it becomes endemicSharing the science already learned by researchers in Scandinavia andthe United Kingdom is essential to developing an effective managementstrategy

Secretary of Agriculture Ann Veneman acknowledged the importanceof ISA when she authorized the expenditure of $8 million in fiscal years2002 and 2003 to control and eradicate the disease APHIS receivedpass-through funding for ISA research and control efforts in December2001 and in spring 2002 put into place program standards to eradicatethe disease

In September 2002 the US Department of Agriculturersquos Animal andPlant Health Inspection Service (APHIS) hosted a symposium on ISA withassistance from the US Department of the Interiorrsquos US GeologicalSurvey and the US Department of Commercersquos National Marine Fish-eries Service The American Veterinary Medical Association suppliedorganizational expertise as well This 2-day event was held in connectionwith the International Symposium on Aquatic Animal Health in NewOrleans These symposium proceedings capture not only the formalscientific presentations but also a court reporterrsquos transcript of theinformal presentations at the end of the second day which addressedreal-world considerations for ISA prevention control and eradication

Given the importance of the information shared at the meetingAPHIS staff members have pulled out all the stops to produce this book in8 months Once our supply of free copies is exhausted copies can bepurchased from the US Department of Commercersquos National TechnicalInformation Service 5285 Port Royal Road Springfield VA 22161 Alsoa pdf version will be placed on the Web site of APHISrsquo VeterinaryServices at lthttpwwwaphisusdagovvsaquaculturegt

I hope you find the information presented at the conference as usefulas I did

W Ron DeHavenDeputy Administrator Veterinary ServicesAnimal and Plant Health Inspection ServiceUS Department of AgricultureWashington DC

Letter of Transmittal

iii

On behalf of Secretary of Agriculture Ann Veneman APHIS AdministratorBobby Acord and the Deputy Administrator for Veterinary Services RonDeHaven thanks for your interest in learning more about infectioussalmon anemia (ISA) This book documents the scientific paperspresented at a 2-day symposium held September 3ndash4 2002 in NewOrleans LA during the weeklong meeting of the International Symposiumon Aquatic Animal Health In addition to 18 formal scientific presenta-tions we have captured comments from individuals who spoke brieflyduring an open forum held at the end of day 2 of the symposium Whilethe formal presentations were submitted in advance and put throughrigorous peer review the forum talks were not vetted or corrected otherthan for clarity We took the precaution of using a local court reporter tocapture the forum talks verbatim since those speakers were not requiredto submit manuscripts

The meeting itself was structured to provide an internationalresponse to ISA using the themes of prevention control and eradicationFour moderators introduced the main subject topics and the speakersrsquopresentations in the areas of international applied research responsediagnostic and laboratory response management response and regula-tory response Our speakers provided the most current internationalknowledge on the extent of ISA infection in various countries includingthe number of salmon farms affected depopulation statistics and relatedissues such as indemnification regulation and management Alsocovered were effective and ineffective management procedures projectedoutcomes of procedures in current use and new developments in appliedISA science and research including diagnostics and prevention

Presenters represented five countries Canada Chile NorwayScotland and the United States When the number of speakers reached19 we split the meeting into 2 sessions on successive days The secondday culminated in a panel discussion entitled Practical Future Consider-ations for ISA Prevention Control and Eradication This open forum wasdesigned to allow audience participation and permit ISA experts not onthe official agenda to address topics of interest

It always takes a team of workers to organize and execute a meetinglike ours and to bring its results to printed form afterward We will giveyou more particulars on this process and identify all the major players inthe Acknowledgments section immediately following this Foreword

Should you wish to contact any speakers or forum participants formore details appendixes 1 and 2 provide complete contact information

Foreword

iv

We look forward to working with you in helping the aquacultureindustries and natural-resource agencies manage ISA The symposiumheld in New Orleans was an important national and international step incoming to grips with this significant aquatic animal health threat

Otis Miller Jr DVM MSNational Aquaculture CoordinatorChair ISA Symposium Planning CommitteeUSDAndashAPHISndashVeterinary ServicesRiverdale MD

Rocco C CiprianoSenior Research MicrobiologistUS Department of the InteriorUS Geological SurveyNational Fish Health Research LaboratoryKearneysville WV

v

The Veterinary Services unit of USDArsquos APHIS the US GeologicalSurvey the National Marine Fisheries Service and Interiorrsquos US Fish andWildlife Service at the Federal levelmdashalong with Mainersquos Department ofMarine Resources at the State levelmdashare hard at work to ensure asuccessful prevention control and eradication program to manageinfectious salmon anemia (ISA) in the United States Similar efforts arebeing mounted in Canada Scotland and Norway all of which sufferedISA outbreaks before the disease came to US waters

APHIS supports an integrated approach that uses the expertise of allrelevant Federal agencies States and Canadian Provinces as well asindustry stakeholders in an international partnership for development andimplementation of programs involving aquaculture Because ISArepresents a disease threat to the salmon industry on both sides of theAtlantic integrating the scientific and regulatory response to this diseaseacross national borders is extremely important The creation of the firstinternational symposium on ISA held in the United States is an initial stepin acknowledging that this disease is everybodyrsquos issue We [Otis Millerand Rocco Cipriano] organized the symposium but not in a vacuum Wegratefully acknowledge the expertise and cooperation of the OfficeInternational des Epizooties (Paris) the American Veterinary MedicalAssociation (headquartered in Schaumburg IL) the Canada Departmentof Fisheries and Oceans the Norwegian Animal Health AuthorityScotlandrsquos famous marine research lab in Aberdeen aquaculture units inthe Provincial Governments of New Brunswick and Prince Edward Islandand the Maine Aquaculture Association in supplying experts for thepodium

Rocco Cipriano provided yeoman service in the summer of 2002 insetting up a peer-review process for all the manuscripts formallypresented Keeping track of the peer reviewersrsquo comments and makingsure that speakers took those comments into account in revising theircontributions before the meeting in September was a gargantuan taskSome readers may not understand that Roccorsquos role as a ldquoTechnicalCoordinatorrdquo for the book far exceeds that of an editor He also served asthe official reviewer of the entire text for the US Geological Survey Wecould not have made this proceedings without his help before during andafter the meeting itself

The following individuals also worked on putting together theprogram Peter Merrill DVM MicroTechnologies Inc Richmond ME Jill Rolland fisheries biologist with USDAndashAPHISndashVeterinary ServicesRiverdale MD Alasdair McVicar PhD DFO Aquaculture and Fish Health Ottawa ON

Acknowledgments

vi

David Scarfe DVM PhD assistant director of scientific activities withthe American Veterinary Medical Association Schaumburg IL Kevin Amos PhD national fish health coordinator for the USDepartment of Commercersquos National Marine Fisheries Service OlympiaWA Jim Winton PhD chief of the fish health section at the US GeologicalSurveyrsquos Western Fisheries Research Center Seattle WA Gilles Olivier PhD Canada Department of Fisheries and OceansOttawa ON

The following individuals helped by volunteering to act as moderatorsfor the four subsections of the program David Scarfe DVM PhD Laura Brown PhD group leader of genome sciences NationalResearch Council of Canada Halifax NS Paul J Midtlyng DVM PhD VESO Oslo NO Carey Cunningham PhD Fisheries Research Services MarineLaboratory Aberdeen UK Patricia Barbash fishery biologist USDI US Fish and Wildlife ServiceNortheast Fisheries Research Center Lamar PA

Patricia Barbash and Kevin Amos also provided interagency review of theentire text outside USDA

Behind the scenes the following individuals worked on preparing thebook for publication Janet S Wintermute writereditor USDAndashAPHISndashLegislative and PublicAffairs Riverdale MD Heather Curlett designer USDA Design Division Beltsville MD Jill Rolland and Bronte Williams manuscript traffic and preparationUSDAndashAPHISndashVeterinary Services Riverdale MD Anita McGrady printing specialist USDAndashAPHISndashLegislative and PublicAffairs Washington DC

Their single-minded dedication to this project is the reason APHISwas able to issue the proceedings in 8 months

In closing the fourth International Symposium on Aquatic AnimalHealth was a highly approriate setting for a symposium on ISA If it werenot for the unselfish efforts of the organizer of the symposium on aquaticanimal health Dr Ron Thune as well as the program chairs Drs JohnHawke and Jerome La Peyre APHIS could not have held this meeting

vii

And since money drives all endeavors and bookmaking is no exceptionnow is the time to acknowledge financial contributions from APHIS theNational Marine Fisheries Service and the US Geological Survey

I hope you will find that this book enhances your understanding ofISA and supports you in your commitment to help the scientific communitydeal with it

mdash Otis Miller Jr DVM MSNational Aquaculture CoordinatorUSDAndashAPHISndashVeterinary Services4700 River Road Unit 46Riverdale MD 20737(301) 734ndash6188otismilleraphisusdagov

viii

1 Infectious Salmon Anemia The Current Stateof Our Knowledge

Rocco C Cipriano and Otis Miller Jr

13 Role and Function of the OIE Fish DiseasesCommission in the Field of Aquatic AnimalHealth

Tore Haringstein

International Applied Research Response

25 A Comparative Review of Diagnostic AssaysUsed To Detect Infectious Salmon AnemiaVirus in the United States

Peter L Merrill

39 The Development of Infectious Salmon AnemiaVirus Vaccines in Canada

Frederick SB Kibenge Molly JT KibengeTomy Joseph and John McDougall

51 Epidemiologic Study of Infectious SalmonAnemia in Maine Combining the SubjectiveWith the Objective in Risk Factor Estimation

Lori Gustafson Steve Ellis and Larry Hammell

55 The Epidemiology of Infectious Salmon Anemiain Scotland

Alexander G Murray

International Diagnostic and LaboratoryResponse

63 Improved Diagnosis of Infectious SalmonAnemia Virus by Use of a New Cell Line DerivedFrom Atlantic Salmon Kidney Tissue

Jill B Rolland Deborah A Bouchard and James R Winton

Contents

ix

69 Evaluation of Infectious Salmon AnemiaDiagnostic Tests

Carol A McClure K Larry Hammell Ian R Dohoo Henrik Stryhnand Leighanne J Hawkins

75 Development of a Strain Typing Assay forInfectious Salmon Anemia Virus (ISAv)

Marcia Cook Sherry Vincent Rachael Ritchie and Steve Griffiths

87 The Genetics of Infectious Salmon AnemiaVirus

Carey O Cunningham and Michael Snow

International Management Response

97 Infectious Salmon Anemia in Norway and theFaroe Islands An Industrial Approach

Cato Lyngoslashy

111 The Eradication of an Outbreak of ClinicalInfectious Salmon Anemia From Scotland

Ronald M Stagg

125 Practical Grower Experience in the ProactivePrevention and Control of Infectious SalmonAnemia

Sebastian M Belle

135 Survey of Nonsalmonid Marine Fishes forDetection of Infectious Salmon Anemia Virusand Other Salmonid Pathogens

Sharon A MacLean Deborah A Bouchard and Stephen K Ellis

145 Infectious Salmon Anemia in New BrunswickAn Historical Perspective and Update onControl and Management Practices(1997ndash2002)

Sandi M McGeachy and Mark J Moore

x

International Regulatory Response

155 Experiences With Regulatory Responses toInfectious Salmon Anemia in Norway

Kristin E Thorud and Tore Haringstein

161 Regulatory Aspects of Infectious SalmonAnemia Management in Scotland

Alasdair H McVicar

167 Design and Implementation of an InfectiousSalmon Anemia Program

Otis Miller Jr

175 Presentations From the Open Forum

191 Appendix 1mdashAuthorsrsquo Affiliations

193 Appendix 2mdashAffiliations of the Speakers at theOpen Forum

xi

1

Abstract Infectious salmon anemia (ISA) is a highlyinfectious viral disease that causes acute mortality princi-pally among Atlantic salmon (Salmo salar) The cause ofISA is an orthomyxolike enveloped virus that replicatesthroughout most host tissues including midkidney headkidney liver spleen intestine gills muscle and heart Thevirus is cultured in Atlantic salmon head kidney (SHKndash1)cells in the Chinook salmon (Oncorhynchus tshawytscha)embryo (CHSEndash214) cell line and within the TO cell linedeveloped from head kidney leucocytes Clinical signs ofthe disease may include pale gills ascites liver congestionenlarged spleen petechial hemorrhages within visceral fatcongestion of the gut and severe anemia The disease ispronounced in the marine environment where it is mostoften transmitted by cohabitation with infected live salmoninfected biological materials or contaminated equipmentControl of ship and personnel movements among infectedsites destruction of infected lots and the closure andfallowing of virus-contaminated areas may be used toreduce the likelihood of further spread of the disease

Hosts and Geographic Range

Infectious salmon anemia (ISA) is a highly infectiousdisease of Atlantic salmon (Salmo salar) that wasfirst reported in Norwegian aquaculture facilities Thedisease has since been described among premarketAtlantic salmon in Scotland (Bricknell et al 1998)New Brunswick Canada (Lovely et al 1999 Jones etal 1999a) the United Kingdom (Rodger et al 1999)the Cobscook Bay region of Maine (Bouchard et al2001) and in the Faroe Islands (Anonymous 2000)The virus that causes ISA has also been detectedamong coho salmon (Oncorhynchus kisutch) in Chile(Kibenge et al 2001) In Canada the disease wasfirst characterized as a new condition termedldquohemorrhagic kidney syndromerdquo or HKS (Byrne et al1998) The pathology of HKS was later shown to becaused by ISAv (Bouchard et al 1999 Lovely et al1999) although laboratory confirmation of ISA virus(ISAv) was initially complicated by dual isolation ofthat virus and a nonpathogenic Toga-like virus fromHKS samples (Kibenge et al 2000a)

The rapid invasion of ISAv into three bays withinNew Brunswick and its subsequent spread among 21

Infectious Salmon Anemia The Current Stateof Our Knowledge

Rocco C Cipriano and Otis Miller Jr1

1 Dr Cipriano is with the US Geological Surveyrsquos NationalFish Health Research Laboratory in Kearneysville WV DrMiller is with USDAndashAPHISrsquo Veterinary Services inRiverdale MD

farms (Bouchard et al 1998) indicate the severenature of the threat that ISA represents for Atlanticsalmon aquaculture Furthermore the annual cost ofISA outbreaks among farmed fish in 1999 wasreported in US dollars to be $11 million in Norwayand $14 million in Canada The 1998ndash99 epidemicsin Scotland were valued at a cost of $32 million(Hastings et al 1999) Although epizootics of ISAhave been specifically associated with culturedsalmon (Department of Fisheries and Oceans[DFO]mdashCanada) biologists have also detected thepresence of ISAv among Atlantic salmon populationsthat are wild or have escaped from aquacultureoperations at the Magaguadavic River fish trap (Bayof Fundy NB) In addition to Atlantic and Chinooksalmon the pathogen infects but has not produceddisease in freshwater brown trout (Salmo trutta)(Nylund et al 1995a) sea trout (S trutta) (Nylundand Jakobsen 1995) and rainbow trout(Oncorhynchus mykiss) (Nylund et al 1997)Although the virus has been detected in saithe(Pollachius virens) it is unable to replicate in suchhosts (Raynard et al 2001)

Etiology

The cause of ISA is an enveloped virus 45ndash140 nm indiameter (Dannevig et al 1995b) with a buoyantdensity 118 gmL in sucrose and cesium chloridegradients It shows maximum replication at 15 oC butstrongly reduced replication at 25 oC (Falk et al1997) The virus may be cultured in the SHKndash1 cellderived from Atlantic salmon pronephros cells andproduces variable cytopathic effects (CPE) between3 and 12 days after inoculation (Dannevig et al1995ab Kibenge et al 2000b) The ISAv alsoreplicates and produces CPE within the Atlanticsalmon head kidney (ASK) cell line developed byDevold et al (2000) and the TO cell line developedfrom Atlantic salmon head kidney leukocytes byWergeland and Jakobsen (2001) Some but not allstrains of ISAv will also replicate in Chinook salmon(Oncorhynchus tshawytscha) embryo (CHSEndash214)cells and produce CPE between 4 and 17 days afterinoculation (Kibenge et al 2000b) The virus also

2

International Response to Infectious Salmon AnemiaPrevention Control and Eradication

replicates within but does not produce distinct CPEin the AS cell line (Sommer and Mennen 1997)Growth is inhibited by actinomycin D but not by5ndashbromondash2ndashdeoxyuridine (Sommer and Mennen1997 Falk et al 1997) and the virus is most closelyrelated to other orthomyxoviruses (Mjaaland et al1997 Krossoslashy et al 1999 Sandvik et al 2000)Four major polypeptides are evident with estimatedmolecular sizes of 71 53 43 and 24 kDa (Falk et al1997)

Mjaaland et al (1997) indicated that the totalgenome of ISAv (145 kb) consisted of eightsegments between 1 and 23 kb This geneticanalysis suggests a close relationship between ISAvand other viruses in the family Orthomyxoviridae butthe smallest genomic segment (segment 8) is nothomologous with any other known sequence dataDevelopment of a primer set from this segmenttherefore had significant diagnostic value Krossoslashy etal (1999) further established a relationship betweenISAv and other orthomyxoviruses by examining thehighly conserved orthomyxovirid PB1 proteinencoded by segment 2 Intrafamily geneticcomparisons conducted in this manner showed thatISAv has a distant relationship with other orthomyxo-viruses and is more closely related to the influenzaviruses than to the Thogoto viruses The relationshipto the orthomyxoviruses was further strengthened bythe demonstration by Sandvik et al (2000) that theISAv genomic segments had conserved 3rsquo- and 5rsquoends typical for orthomyxoviruses and the ISAvmRNA has heterologous 5rsquo-endsmdashindicating areplication strategy more related to the influenzaviruses than the Thogoto viruses

On the basis of the Krossoslashy teamrsquos geneticcharacterizations and the psychrophilic nature of thisvirus which potentially restricts its host range topoikilothermic vertebrates those authors proposedthat ISAv be the type species of a new genusAquaorthomyxovirus The name Isavirus has beenproposed by the International Committee onTaxonomy of Viruses version 3 (httpwwwncbinlmnihgovICTVdbIctvindexhtm) for this genus inpreference to Aquaorthomyxovirus suggested byKrossoslashy et al (1999)

Nucleotide variations in segments 2 and 8 wereused to differentiate Scottish isolates of ISAv fromthose of Norwegian or North American origins(Cunningham and Snow 2000 Krossoslashy et al 2001)Despite such differentiation the Scottish isolate wasmore closely related to the Norwegian strain than itwas to the North American strain These results mayinitially have indicated that geographic proximityinfluences the ISAv genotype suggesting that distinctstrains occur on both sides of the Atlantic OceanHowever Ritchie et al (2001a) showed thatnucleotide sequences of an ISAv isolate obtainedfrom Atlantic salmon in Nova Scotia were moresimilar to Norwegian and Scottish strains than toisolates from neighboring New Brunswick Althoughdifferences were not detected in the nucleotidesequences analyzed the Nova Scotian isolate did notcause typical ISA and was therefore considered tobe functionally different from the Scottish andNorwegian isolates On the basis of calculatedevolutionary mutation rates in segment 2 Krossoslashy etal (2001) suggested that Canadian and NorwegianISAv isolates diverged about 100 years ago whichinterestingly enough coincided with anthropogenicmovements of salmonid fishes (particularly viatransport of sea trout) between Europe and NorthAmerica

The virus possesses hemagglutinating as wellas fusion and receptor-destroying activity The latteractivity has been suggested to be caused by anacetylesterase (Falk et al 1997) Devold et al (2001)have shown that the hemagglutinin (HA) genecontains a highly polymorphic region (HPR) whichshows sequence variation where distinct groups ofisolates predominate within certain geographic areasAdditionally sequence analyses have been providedfor segment 3 which encodes for the nucleoprotein(NP) that has an approximate mass of 71 to 72 kd(Snow and Cunningham 2001 Ritchie et al 2001bClouthier et al 2002) for segment 4 which ispurported to be a polymerase encoding gene for theP2 protein (Ritchie et al 2001b Clouthier et al 2002)for segment 6 which encodes a 38ndash42-kd glycosy-lated protein determined to be the HA analogue(Griffiths et al 2001 Clouthier et al 2002 Kibenge et

3

Infectious Salmon AnemiaThe Current State of Our Knowledge

al 2002 Devold et al 2001) and for segment 7which possibly encodes for two matrix genesexpressing the P4P5 proteins (Biering et al 2002)Clouthier et al (2002) have completed a genetic mapfor ISAv and in addition to what has been notedabove these authors have completed analysis ofsegments 1 and 5 Segment 1 has been shown toencode for the P1 protein and segment 5 for the47-kd glycosylated P3 protein suggested to be asurface glycoprotein

Pathogenicity

Considerable viral replication occurs within infectedfish and the virus may become widely disseminatedthroughout most tissues including midkidney headkidney liver spleen intestine gills muscle and heart(Jones et al 1999a Rimstad et al 1999) Danneviget al (1994) suggested that liver cells leukocytesand immature erythrocytes are target cells forreplication of ISAv Further study by Nylund et al(1996) supported the hypothesis that leukocytes maybe target cells for ISAv and showed that it canactually replicate in endothelial cells lining the bloodvessels in the ventricle of the heart in endocardialcells and in polymorphonuclear leukocytes In tissueculture the virus binds to sialic acid residues on thecell surface and fuses with endosomes andliposomes where binding uptake and fusion areenhanced as pH values are decreased from 75 to45 (Eliassen et al 2000)

Clinical signs may be evident 2ndash4 weeksfollowing infection and commonly include pale gillsascites enlargement of the liver and spleenpetechiae in the visceral fat congestion of the gutsevere anemia and mortality (Hovland et al 1994Thorud and Djupvik 1988 Evensen et al 1991)Microscopic pathological changes are commonlycharacterized by renal interstitial hemorrhage andtubular necrosis branchial lamellar and filamentalcongestion congestion of the intestine and pyloriccecae and perivascular inflammation in the liver(Mullins et al 1998b Rimstad et al 1999) Prominentlesions are often reported in the parenchyma andvascular system of the liver where congestion and

degeneration of hepatocytes are often followed byhemorrhagic necrosis in the latter stages of disease(Evensen et al 1991)

Speilberg et al (1995) concluded that lesions inthe liver may not be the sole result of an anemiabecause significant ultrastructural damage hadalready occurred before a decrease in hematocritvalues and any viral-induced disruption of thehepatocytes had been observed These observationssuggest that the lesions in the liver may result froman impeded sinusoidal blood flow that culminates inan ischemic hepatocellular necrosis (Speilberg et al1995) Decreases in hepatic glutathione of up to70 percent observed in diseased fish may affect thecapability of the liver to transform and excretexenobiotics from the body (Hjeltnes et al 1992)

The development of anemia suggests thaterythrocytes may be among the most importanttarget cells of the virus Anemia often developsrather late in the course of infection (Dannevig et al1994) and a leukopoenia is suggested to developconcomitantly with the anemia (Thorud 1991)However Dannevigrsquos team demonstrated that headkidney leukocytes are infected earlier and areprobably more important than erythrocytes inreplication of the virus earlier in the infective processStill fish injected with the ISAv may display asuppressed leukocyte response that does notnecessarily correlate with the development oferythrocytic anemia (Dannevig et al 1994)Suppression of immune function and development ofanemia therefore appear to be independent events

Experimentally elevated plasma cortisolconcentrations have been correlated to the severity ofanemia as measured by hematocrit values (Olsen etal 1992) Plasma lactate may also be elevated indiseased fish (Olsen et al 1992)

Transmission

The disease is pronounced in the marineenvironment where it is most often transmitted bycohabitation with infected live salmon or infectedbiological materials such as animal wastes or

4


discharges from normal aquaculture operationsslaughter facilities (Vagsholm et al 1994) andcontaminated wellboats (Shannon 1998 Murray et al2002) Infected fish may transmit the disease weeksbefore they show apparent signs of infection Thevirus may spread horizontally from fish to fish byshedding of virions from the blood gut contentsurine and epidermal mucus of infected salmon(Totland et al 1996) Moreover fish that surviveepizootics may shed viral particles for more than1 month into the surrounding water (Hjeltnes et al1994) within which the virus is relatively shortlivedand may persist for only about 20 hours at 6 oC andup to 4 days in tissues at the same temperature(Nylund et al 1994b) Consequently infectedbiological materials such as animal wastes ordischarges from aquaculture operations slaughterfacilities (Vagsholm et al 1994) and contaminatedwellboats (Shannon 1998 Murray et al 2002) mayestablish better reservoirs of infection than the watercolumn alone Blood and mucus contain especiallylarge amounts of virus and more effectively transmitthe disease than feces plankton and salmon lice(Rolland and Nylund 1998)

Sea lice of the species Caligus elongatus andLepeophtheirus salmonis however may also beimportant vectors of the virus during epidemics(Nyland et al 1994b) There is no evidence thatscallops cocultured with Atlantic salmon eitheraccumulate the pathogen or transmit the disease(Bjoershol et al 1999) The pathogen can betransmitted to but has not produced disease infreshwater brown trout (Nylund et al 1995a) seatrout (Nylund and Jakobsen 1995) sea-run browntrout (Rolland and Nylund 1999) and rainbow trout(Nylund et al 1997) suggesting that these fish maybecome carriers and serve as potential reservoirs ofinfection (Nylund et al 1997) Although brown troutproduce neutralizing antibodies against ISAv within45 days after primary infection the virus may still bepresent 7 months after infection (Nylund et al1994a)

Clearance of the virus following experimentalinfection progresses at a greater rate in Arctic char

(Salvelinus alpinus) than in rainbow trout and browntrout Thus the potential for char to act as a long-term carrier of ISAv may be less than that of othersalmonids all of which apparently clear viable virusby 40 days following injection (Snow et al 2001)

Horizontal transmission of ISAv in fresh waterhas been achieved experimentally (Brown et al 1998)and occurs rapidly between infected and naive smoltsin fresh water Even under these conditionsasymptomatic smolts may remain infective to naiumlveparr for 18 months after the original challenge(Melville and Griffiths 1999)

Vertical transmission of the virus from parent tooffspring via intraovum infection has not beendemonstrated (Melville and Griffiths 1999) Eventhough it is commonly believed that the virus is nottransmitted vertically ISAv mortality has beenreported among first-feeding fry (Nylund et al 1999)This scenario emphasizes the importance ofscreening brood fish and conducting proper eggdisinfection procedures to reduce contagion in earlylife stages

Diagnostics and Detection

Viral replication and the development of CPE intissue culture are routinely used as the standard bywhich all other diagnostic and detection assays aremeasured As already mentioned the virus can becultured in Atlantic SHKndash1 cells the CHSEndash214 cellline the ASK cell line the TO cell line and the AScell line The focal CPE associated with ISAv growthin CHSEndash214 cells suggest that this cell line couldprovide the foundation for a culture-based diagnosticThe lack of focal CPE has been viewed as adisadvantage associated with SHKndash1 and AS cellsUnfortunately the CHSEndash214 line does not supportthe growth of all ISAv isolates (Kibenge et al 2000b)The inability of some ISAv isolates to replicate inCHSEndash214 cells and the lack of distinct CPE in eitherAS or SHKndash1 complicate effective and consistentcultural detection of this virus Consequently paralleluse of both SHKndash1 or CHSEndash214 cells providesmore sensitive detection of ISAv than use of either

5


cell line alone (Opitz et al 2000) Further develop-ment and greater availability of the ASK and TO celllines may alleviate these problems

Nonculture-based diagnostics that detectISAv include an indirect fluorescent antibody test(Falk and Dannevig 1995a) and a reverse-transcriptasendashpolymerase chain reaction (RTndashPCR)procedure (Mjaaland et al 1997) Furtherconfirmation has been effected by the use of a DNAprobe employing primer sets developed againstsegment 8 of the virus (McBeath et al 2000)

Devold et al (2000) found RTndashPCR to be moresensitive for detection of ISAv among carrier sea troutthan either culture or injection of suspect bloodsamples into naiumlve fish Furthermore RTndashPCRscreens of gill mucus present an accurate andsensitive nonlethal alternative for detection of thevirus from other tissues that require necropsy(Griffiths and Melville 2000)

Production of a monoclonal antibody againstISAv enabled Falk et al (1998) to conduct severalserodiagnostic assays including the enzyme-linkedimmunosorbent assay (ELISA) immunofluorescentantibody staining of virus-infected cell culturesimmunoelectron microscopy of negatively stainedvirus preparations virus neutralization assay andhemagglutination inhibition assay

Atlantic salmon that survive infections with ISAv(Falk and Dannevig 1995b) or are either passively(Falk and Dannevig 1995b) or actively immunized(Brown et al 2001 Jones et al 1999b) against thevirus develop an immune resistance against thispathogen Recently an indirect ELISA assay wasdeveloped to detect antibodies to ISAv in Atlanticsalmon sera (Kibenge et al 2002) In a diagnosticsense this assay can theoretically denote previousexposure to the pathogen among nonvaccinated fishby detection of viral-specific antibodies The assayalso permits titration of ISAv-specific antibodies as aconsequence of vaccination In general the currentnonculture-based methods for routine detection andconfirmation of ISAv in decreasing order of sensitivityand specificity are RTndashPCR antibody ELISAimmunofluorescence and histopathologic

examination (Groman et al 2001) It is important tonote that the phenotypic (Kibenge et al 2000b) andgenomic (Blake et al 1999) differences that existamong ISAv isolates may influence the use ofspecific assays

Management

Because of the acute nature of the disease and aninability to control mortality European EconomicCommunity countries require compulsory slaughterof infected stocks (Hill 1994) Similar eradicationprograms have been enacted in Canada (Mullins1998a) Because the virus is readily transmitted inseawater such dissemination may readilycontaminate culture facilities within 5 to 6 km of aninfected site within a 6- to 12-month period (Eide1992) It is recommended therefore that culturesites be spaced no less than 5ndash6 km apart and thatwastewater from slaughter and processing facilitiesbe thoroughly disinfected (Jarp and Karlsen 1997)

Further contagion may be managed by controlof ship and personnel movements among sitesdestruction of infected lots and the closure andfallowing of contaminated sites (Murray 2001)Iodophor chloraminendashT and chlorine dioxide havebeen shown to be effective topical disinfectantsagainst ISAv when used for a minimum of 5 minutesaccording to manufacturerrsquos instructions (Smail et al2001)

Nylund et al (1995b) have observed greateroverall resistance to ISA among two wild strains ofAtlantic salmon than was noted in a strain used incommercial aquaculture Because all other physicalparameters that may have affected challenge resultswere held constant the observed patterns ofresistance were believed to result from geneticdifferences among the strains Such differencescould theoretically be used to select for resistance toISAv (Dahle et al 1996) The management of ISAvthrough the development of disease-resistant strainsof fish however is not consistent with current controlpractices that involve destruction of infectedpopulations and disinfection of contaminated sites

6


Vaccination

Jones et al (1999b) have shown that vaccination viaintraperitoneal injection of inactivated virus elicitedthe best protection if at least 734 degree-days hadelapsed between vaccination and challenge Thesestudies resulted in the development of a commer-cially licensed autogenous product used withinspecific areas (McDougall et al 2001) Protectionwas significantly improved if the viral antigen wasdelivered in an oil emulsion (Jones et al 1999bBrown et al 2001) Christie et al (2001) alsoindicated that vaccination may produce relativepercent survival values of 90 or higher (54 percentmortality among controls) for 6 months aftervaccination without significant risk of viral transmis-sion by vaccinated fish that may have becomeasymptomatic carriers The latter research was theresult of a consortium of scientists from the Universityof Bergen the National Veterinary Institute andIntervet Norbio and was supported financially by theNorwegian Research Council A multivalent vaccineincluding ISAv infectious pancreatic necrosis virusVibrio anguillarum (two serotypes) V salmonicidaAeromonas salmonicida and Moritella viscosaprepared in a water in oil formulation is projected forcommercial availability within Norwegian and FaroeIsland markets by early 2003 Multivalent vaccinecombinations designed for the Canadian and UnitedKingdom markets are also in development

References Cited

Anonymous 2000 ISA hits the Faroes Fish FarmingInternational 27 47

Biering E Falk K Hoel E Thevarajan J JoerinkM Nylund A Endresen C Krossoslashy B 2002Segment 8 encodes a structural protein of infectioussalmon anaemia virus (ISAV) the co-linear transcriptfrom Segment 7 probably encodes a non-structural orminor structural protein Diseases of AquaticOrganisms 49(2) 117ndash122

Bjoershol B Nordmo R Falk K Mortensen S1999 Cohabitation of Atlantic salmon (Salmo salarL) and scallop (Pecten maximus)mdashchallenge withinfectious salmon anaemia (ISA) virus andAeromonas salmonicida subsp salmonicida InBook of abstracts 12th international pectinidworkshop 5ndash11 May 1999 Bergen NO [Place ofpublication publisherrsquos name and paginationunknown]

Blake S Bouchard D Keleher W Opitz MNicholson B L 1999 Genomic relationships of theNorth American isolate of infectious salmon anemiavirus (ISAV) to the Norwegian strain of ISAVDiseases of Aquatic Organisms 35(2) 139ndash144

Bouchard D A Opitz H M Nicholson B L BlakeS Keleher W R 1998 Diagnosis of two emerginginfections in the Bay of Fundy In Cipriano Rocco Ced Proceedings of the 23d annual eastern fish healthworkshop 31 Marchndash2 April 1998 Plymouth MALeetown WV US Department of the Interior 4

Bouchard D Keleher W Opitz H M Blake SEdwards K C Nicholson B L 1999 Isolation ofinfectious salmon anemia virus (ISAV) from Atlanticsalmon in New Brunswick Canada Diseases ofAquatic Organisms 35 131ndash137

Bouchard D A Brockway K Giray C Keleher WMerrill P L 2001 First report of infectious salmonanaemia (ISA) in the United States Bulletin of theEuropean Association of Fish Pathologists 21 86ndash88

Bricknell I R Bruno D W Cunningham CHastings T S McVicar A H Munro P D RaynardR Stagg R M 1998 Report on the first occurrenceof infectious salmon anemia (ISA) in Atlantic salmonin Scotland United Kingdom [Abstract] In KaneA S Poynton S L eds Proceedings of the thirdinternational symposium on aquatic animal health30 Augustndash3 September 1998 Baltimore MDBaltimore APC Press 132

7


Brown L L MacKinnon A M Ferguson H W1998 Infectious salmon anaemia virus pathologyand transmission in fresh and salt water [Abstract]In Kane A S Poynton S L eds Proceedings ofthe third international symposium on aquatic animalhealth 30 Augustndash3 September 1998 Baltimore MDBaltimore APC Press 133

Brown L L Sperker S A Clouthier S ThorntonJ C 2001 Development of a vaccine againstinfectious salmon anaemia virus (ISAV) Bulletin ofthe Aquaculture Association of Canada 100 4ndash7

Byrne P J MacPhee D D Ostland V E JohnsonG Ferguson H W 1998 Hemorrhagic kidneysyndrome of Atlantic salmon Salmo salar L Journalof Fish Diseases 21 81ndash91

Christie K E Koumans J Villoing S RodsethO M 2001 Vaccination of Atlantic salmon withinactivated ISAv induces high protection against ISAand reduced risk of transmission of ISA [Abstract]In Hiney M H ed Proceedings of the 10th

international conference of the European Associationof Fish Pathologists 9ndash14 September 2001 Dublin(Trinity College) Dublin European Association ofFish Pathologists 0ndash105

Clouthier S C Rector T Brown NEC AndersonE D 2002 Genomic organization of infectioussalmon anaemia virus Journal of General Virology83(2) 421ndash428

Cunningham C O Snow M 2000 Genetic analysisof infectious salmon anaemia virus (ISAV) fromScotland Diseases of Aquatic Organisms 41 1ndash8

Dahle G Hjeltnes B Joslashrstad K E 1996 Infectionof Atlantic salmon sibling groups with infectioussalmon anaemia (ISA) and furunculosis Bulletin ofthe European Association of Fish Pathologists 16192ndash195

Dannevig B H Falk K Sjerve E 1994 Infectivityof internal tissues of Atlantic salmon Salmo salar Lexperimentally infected with the aetiological agent ofinfectious salmon anaemia (ISA) Journal of FishDiseases 17 613ndash622

Dannevig B H Falk K McPress C 1995aProgagation of infectious salmon anaemia (ISA) virusin cell culture Veterinary Research 26 438ndash442

Dannevig B H Falk K Namork E 1995b Isolationof the causal virus of infectious salmon anemia (ISA)in a long-term cell line from Atlantic salmon headkidney Journal of General Virology 76 1353ndash1359

Devold M Krossoslashy B Aspehaug V Nylund A2000 Use of RTndashPCR for diagnosis of infectioussalmon anaemia virus (ISAV) in carrier sea troutSalmo trutta after experimental infection Diseases ofAquatic Organisms 40 9ndash18

Devold M Falk K Dale O B Krossoslashy B BieringE Aspehaug V Nilsen F Nylund A 2001 Strainvariation based on the haemagglutinin gene inNorwegian ISA virus isolates collected from 1987 to2001 indications of recombination Diseases ofAquatic Organisms 47(2) 119ndash128

Eide G W 1992 A retrospective analysis ofoutbreaks of infectious salmon anaemia in Sogn andFjordane in 1985ndash91 [En retrospektiv analyse avILA-utbrudd i Sogn og Fjordane 1985-91] NorskVeterinaertidsskrift 104 915ndash919

Eliassen T M Froslashystad MK Dannevig MJankowska M Brech A Falk K RomoslashrenKGjoslashen T 2000 Initial events in infectious salmonanemia virus infection evidence for the requirementof a low-pH step Journal of Virology 74 218ndash227

Evensen O Thorud K E Olsen Y A 1991 Amorphological study of the gross and lightmicroscopic lesion of infectious salmon anemia inAtlantic salmon (Salmo salar L) Research inVeterinary Science 51 215ndash222

Falk K Dannevig B H 1995a Demonstration ofinfectious salmon anemia (ISA) viral antigens in cellcultures and tissue sections Veterinary Research 26499ndash504

Falk K Dannevig B H 1995b Demonstration of aprotective immune response in infectious salmonanaemia (ISA)-infected Atlantic salmon Salmo salarDiseases of Aquatic Organisms 21 1ndash5

8


Falk K Namork E Rimstad E Mjaaland SDannevig B H 1997 Characterization of infectioussalmon anemia virus an othomyxo-like virus fromAtlantic salmon (Salmo salar L) Journal of Virology71 9016ndash9023

Falk K Namork E Dannevig B H 1998Characterization and applications of a monoclonalantibody against infectious salmon anaemia virusDiseases of Aquatic Organisms 34 77ndash85

Griffiths S Melville K 2000 Non-lethal detection ofISAV in Atlantic salmon by RTndashPCR using serum andmucus samples Bulletin of the European Associationof Fish Pathologists 20 157ndash162

Griffiths S Cook M Mallory B Ritchie R 2001Characterisation of ISAV proteins from cell cultureDiseases of Aquatic Organisms 45(1) 19ndash24

Groman D Mackinley D Jones S 2001 The use ofimmuno-cytochemistry in the confirmation of ISAinfections from fixed tissue impressions andhistologic sections [Abstract] In Hiney M H edIdentification characterization and functional aspectsof the ISA virus structural proteins proceedings ofthe 10th international conference of the EuropeanAssociation of Fish Pathologists 9ndash14 September2001 Dublin (Trinity College) Dublin EuropeanAssociation of Fish Pathologists 0ndash115

Hastings T Olivier G Cusack R BricknellINylund A Binde M Munro P Allan C 1999Infectious salmon anaemia Bulletin of the EuropeanAssociation of Fish Pathologists 19 286ndash288

Hill B J 1994 European community measures forcontrol of fish disease In International symposiumon aquatic animal health program and abstracts4ndash8 September 1994 Seattle WA Davis CAUniversity of California at Davis VIndash5

Hjeltnes B Samuelsen O B Svardal A M 1992Changes in plasma and liver glutathione levels inAtlantic salmon Salmo salar suffering from infectioussalmon anemia (ISA) Diseases of AquaticOrganisms 14 31ndash33

Hjeltnes B Flood P R Totland G K Christie KE Kryvi H 1994 Transmission of infectious salmonanaemia (ISA) through naturally excreted material InInternational symposium on aquatic animal health4ndash8 September 1994 Seattle WA Davis CAUniversity of California at Davis Wndash192

Hovland T Nylund A Watanabe K Endresen C1994 Observation of infectious salmon anaemiavirus in Atlantic salmon Salmo salar L Journal ofFish Diseases 17 291ndash296

Jarp J Karlsen E 1997 Infectious salmon anaemia(ISA) risk factors in sea-cultured Atlantic salmonSalmo salar Diseases of Aquatic Organisms 2879ndash86

Jones SRM MacKinnon A M Groman D B1999a Virulence and pathogenicity of infectioussalmon anemia virus isolated from farmed salmon inAtlantic Canada Journal of Aquatic Animal Health 11400ndash405

Jones SRM Mackinnon A M Salonius K 1999bVaccination of freshwater-reared Atlantic salmonreduces mortality associated with infectious salmonanaemia virus Bulletin of the European Associationof Fish Pathologists 19 98ndash101

Kibenge FSB Whyte S K Hammell K LRainnie D Kibenge M T Martin C K 2000a Adual infection of infectious salmon anaemia (ISA)virus and a Toga-like virus in ISA of Atlantic salmonSalmo salar in New Brunswick Canada Diseases ofAquatic Organisms 42 11ndash15

Kibenge FSB Lyaku J R Rainnie D Hammell KL 2000b Growth of infectious salmon anaemia virusin CHSEndash214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81 143ndash150

Kibenge FSB Gaacuterate O N Johnson GArriagada R Kibenge MJT Wadowska D 2001Isolation and identification of infectious salmonanaemia virus (ISAV) from coho salmon in ChileDiseases of Aquatic Organisms 45 9ndash18

9


Kibenge MJT Opazo B Rojas A H KibengeFSB 2002 Serological evidence of infectioussalmon anaemia virus (ISAV) infection in farmedfishes using an indirect enzyme-linkedimmunosorbent assay (ELISA) Diseases of AquaticOrganisms 51 1ndash11

Krossoslashy B Hordvik I Nilsen F Nylund AEndresen C 1999 The putative polymerasesequence of infectious salmon anemia virus suggestsa new genus within the Orthomyxoviridae Journal ofVirology 73 2136ndash2142

Krossoslashy B Nilsen F Falk K Endresen C NylundA 2001 Phylogenetic analysis of infectious salmonanaemia virus isolates from Norway Canada andScotland Diseases of Aquatic Organisms 44(1) 1ndash6

Lovely J E Dannevig B H Falk K Hutchin LMacKinnon A M Melville K J Rimstad EGriffiths S G 1999 First identification of infectioussalmon anaemia virus in North America withhaemorrhagic kidney syndrome Diseases of AquaticOrganisms 35 145ndash148

McBeath AJA Burr K LndashA Cunningham C O2000 Development and use of a DNA probe forconfirmation of cDNA from infectious salmonanaemia virus (ISAV) in PCR products Bulletin of theEuropean Association of Fish Pathologists 20(4)130ndash134

McDougall J Kibenge FSB Evensen Oslash 2001Vaccination against infectious salmon anemia virus(ISAV) In Hiney M H ed Proceedings of the 10th

international conference of the European Associationof Fish Pathologists 9ndash14 September 2001 Dublin(Trinity College) Dublin European Association ofFish Pathologists Pndash192

Melville K J Griffiths S G 1999 Absence ofvertical transmission of infectious salmon anemiavirus (ISAV) from individually infected Atlantic salmonSalmo salar Diseases of Aquatic Organisms 38231ndash234

Mjaaland S Rimstad E Falk K Dannevig B H1997 Genomic characterization of the virus causinginfectious salmon anemia in Atlantic salmon (Salmosalar L) an orthomyxo-like virus in a teleost Journalof Virology 71 7681ndash7686

Mullins J 1998 Infectious salmon anemiamanagement and control in New Brunswick[Abstract] In 6th annual New England farmed fishhealth workshop 3 April 1998 Eastport ME MarineTechnology Center Orono ME University of MaineCooperative Extension 13

Mullins J E Groman D Wadowska D 1998Infectious salmon anaemia in salt water Atlanticsalmon (Salmo salar L) in New Brunswick CanadaBulletin of the European Association of FishPathologists 18 110ndash114

Murray A G 2001 Modeling epidemic spread of twoemerging diseases Pilchard herpesvirus Australiaand infectious salmon anaemia Scotland [Abstract]In Hiney M H ed Proceedings of the 10th

international conference of the European Associationof Fish Pathologists 9ndash14 September 2001 Dublin(Trinity College) Dublin European Association ofFish Pathologists Ondash10

Murray A G Smith R J Stagg R M 2002Shipping and the spread of infectious salmon anemiain Scottish aquaculture Emerging InfectiousDiseases 8 1ndash5

Nylund A Jakobsen P 1995 Sea trout as a carrierof infectious salmon anemia virus Journal ofFisheries Biology 47 174ndash176

Nylund A Alexandersen S Loslashvik P Jakobsen P1994a The response of brown trout (Salmo trutta L)to repeated challenge with infectious salmonanaemia (ISA) Bulletin of the European Associationof Fish Pathologists 14 167ndash170

Nylund A Hovland T Hodneland K Nilsen FLovik P 1994b Mechanisms for transmission ofinfectious salmon anemia (ISA) Diseases of AquaticOrganisms 19 95ndash100

10


Nylund A Alexandersen S Rolland J BJakobsen P 1995a Infectious salmon anemia virus(ISAV) in brown trout Journal of Aquatic AnimalHealth 7 236ndash240

Nylund A Kevenseth A M Krossoslashy B 1995bSusceptibility of wild salmon (Salmo salar L) toinfectious salmon anaemia (ISA) Bulletin of theEuropean Association of Fish Pathologists 15152ndash155

Nylund A Krossoslashy B Watanabe K Holm J A1996 Target cells for the ISA virus in Atlantic salmon(Salmo salar L) Bulletin of the European Associationof Fish Pathologists 16 68ndash72

Nylund A Kvenseth A M Krossoslashy B HodnelandK 1997 Replication of the infectious salmon anemiavirus (ISAV) in rainbow trout Oncorhynchus mykiss(Walbaum) Journal of Fish Diseases 20 275ndash279

Nylund A Krossoslashy B Devold M Aspehaug VSteine N O Hovland T 1999 Outbreak of ISAduring first feeding of salmon fry (Salmo salar)Bulletin of the European Association of FishPathologists 19 70ndash74

Opitz H M Bouchard D Anderson E Blake SNicholson B Keleher W 2000 A comparison ofmethods for the detection of experimentally inducedsubclinical infectious salmon anaemia in Atlanticsalmon Bulletin of the European Association of FishPathologists 20 12ndash22

Olsen Y A Falk K Reite O B 1992 Cortisol andlactate levels in Atlantic salmon Salmo salardeveloping infectious anaemia (ISA) Diseases ofAquatic Organisms 14 99ndash104

Raynard R S Murray A G Gregory A 2001Infectious salmon anaemia virus in wild fish fromScotland Diseases of Aquatic Organisms 46 93ndash100

Rimstad E Falk K Mikalsen A B Teig A 1999Time course tissue distribution of infectious salmonanaemia virus in experimentally infected Atlanticsalmon Salmo salar Diseases of Aquatic Organisms36 107ndash112

Ritchie R J Cook M Melville K Simard NCusack R Griffiths S 2001a Identification ofinfectious salmon anaemia virus in Atlantic salmonfrom Nova Scotia (Canada) evidence for functionalstrain differences Diseases of Aquatic Organisms44(3) 171ndash178

Ritchie R J Heppell J Cook M B Jones SGriffiths S G 2001b Identification andcharacterization of segments 3 and 4 of the ISAVgenome Virus Genes 22 289ndash297

Ritchie R J Bardiot A Melville K Griffiths SCunningham C O Snow M 2002 Characterisationof genomic segment 7 of infectious slmon anaemiavirus (ISAV) Virus Research 84(1ndash2) 161ndash170

Rodger H D Turnbull T Muir F Millar SRichards R H 1999 Infectious salmon anaemia(ISA) in the United Kingdom Bulletin of the EuropeanAssociation of Fish Pathologists 18 115ndash116

Rolland J B Nylund A 1998 Infectiousness oforganic materials originating in ISA-infected fish andtransmission of the disease via salmon lice(Lepeophtheirus salmonis) Bulletin of the EuropeanAssociation of Fish Pathologists 18 173ndash180

Rolland J B Nylund A 1999 Sea running browntrout carrier and transmitter of the infectious salmonanemia virus (ISAV) Bulletin of the EuropeanAssociation of Fish Pathologists 18 50ndash55

Sandvik T Rimstad E Mjaaland S 2000 The viralRNA 3- and 5-end structure and mRNA transcriptionof infectious salmon anaemia virus resemble those ofinfluenza viruses Archives of Virology 145 1659ndash1669

Shannon D 1998 ISA hits Shetlandmdashminister linksspread with wellboats Fish Farming International25 4

11


Smail D A Grant R Rain N Hastings T S 2001Veridical effects of iodophors chloramine T andchlorine doxide against cultured infectious salmonanaemia (ISA) virus [Abstract] In Hiney M H edProceedings of the 10th international conference ofthe European Association of Fish Pathologists9ndash14 September 2001 Dublin (Trinity College)Dublin European Association of Fish PatholigistsOndash10

Snow M Cunningham C O 2001 Characterizationof the putative nucleoprotein gene of infectioussalmon anemia virus (ISAV) Virus Research 74111ndash118

Snow M Raynard R S Bruno D W 2001Comparative susceptibility of Arctic char (Salvelinusalpinus) rainbow trout (Oncorhynchus mykiss) andbrown trout (Salmo trutta) to the Scottish isolate ofinfectious salmon anaemia virus Aquaculture 19647ndash54

Sommer A I Mennen S 1997 Multiplication andhaemadsorbing activity of infectious salmon anaemiavirus in the established Atlantic salmon cell lineJournal of General Virology 78 1891ndash1895

Speilberg L Evensen Oslash Dannevig B H 1995 Asequential study of the light and electron microscopicliver lesions of infectious anemia in Atlantic salmon(Salmo salar L) Veterinary Pathology 32 466ndash478

Thorud K E 1991 Infectious salmon anemia DScdissertation Olso NO Norwegian College ofVeterinary Medicine

Thorud K E Djupvik H O 1988 Infectious salmonanemia in Atlantic salmon (Salmo salar L) Bulletinof the European Association of Fish Pathologists 8109ndash111

Totland G K Hjeltnes B K Flood P R 1996Transmission of infectious salmon anemia (ISA)through natural secretions and excretions frominfected smolts of Atlantic salmon Salmo salarDiseases of Aquatic Organisms 26 25ndash31

Vagsholm I Djupvik H O Willumsen F V TveitA M Tangen K 1994 Infectious salmon anaemia(ISA) epidemiology in Norway Preventive VeterinaryMedicine 19 277ndash290

Wergeland H I Jakobsen R A 2001 A salmonidcell line (TO) for production of infectious salmonanaemia virus (ISAV) Diseases of AquaticOrganisms 44 183ndash190

12

13

Abstract The Office International des Epizooties (OIE) isthe World Organization for Animal Health it currentlycomprises 162 member countries While representation isusually through the member countriesrsquo chief veterinaryofficers competent authorities other than the nationalveterinary services may be responsible for aquatic animalhealth in some OIE member countries

In 1960 the OIE established the Fish DiseasesCommission (FDC) because of increasing awareness ofthe importance of international trade in fish and otheraquatic animals In 1988 the scope of the FDC wasextended to include diseases and pathogens of molluscsand crustaceans

The expansion of international trade in aquaticanimals and their products has called for appropriate rulesto avoid the risk of spread of communicable diseases

Role and Function of the OIE Fish DiseasesCommission in the Field of Aquatic AnimalHealth

Tore Haringstein1

Introduction

The Office International des Epizooties (OIE WorldAnimal Health Organization)mdashan intergovernmentalorganization headquartered in Parismdashwas created byan international agreement on January 25 1924signed in Paris by 28 countries In May 2002 theOIE totaled 162 member countries worldwideRepresentation in OIE is through national delegatesusually the chief veterinary officer of the countryHowever competent authorities other than thenational veterinary services may be responsible foraquatic animal health in some member countriesand this fact makes it necessary for the veterinaryadministrations and other competent authorities tocooperate for the benefit of aquatic animal health

Structure of the OIE

Each OIE member country appoints a delegate Allthe delegates form the OIE International Committeewhich meets once a year in Paris to hold its generalsession The International Committee is the highestauthority within the OIE

At the general session the member countrieselect the president of the International Committee as

Standardization of aquatic animal health requirements fortrade and harmonization of international aquatic animalhealth regulations are critically important to enable trade tocontinue while maintaining effective national diseasecontrol The international aquatic animal health standardsprepared by the FDC are laid down in two importantdocuments OIE International Aquatic Animal Health Codeand the Diagnostic Manual for Aquatic Animal DiseasesCurrently the Code and Manual specifically deal with 13ldquodiseases notifiable to the OIErdquo and 18 ldquoother significantdiseasesrdquo of aquatic animals The diseases are classifiedinto one of these two lists on the basis of their socioeco-nomic importance geographic range and etiology Infec-tious salmon anemia (ISA) is currently listed under the listof ldquoother significant diseasesrdquo and both the Code andManual have chapters dealing with the disease

1Tore Haringstein is with the National Veterinary Institute inOslo Norway

well as members of the administrative regional andspecialist commissions All these positions are heldfor 3 years

The specialist commissions are composed ofmembers elected by the International Committee andare either OIE delegates or internationally renownedexperts from OIE member countries

Currently the OIE has four specialistcommissions

The Foot and Mouth Disease and Other EpizooticsCommission (created 1946)

The Standards Commission (1949)

The International Animal Health Code Commission(1960) and

The Fish Diseases Commission (FDC) (1960)

The FDC was established to deal specificallywith the increase in fish diseases as aquacultureexpanded worldwide As of 1988 the scope of theFDC was extended to include diseases andpathogens of molluscs and crustaceans as well TheFDC has five members

The OIE has these main objectives

To ensure transparency in the animal healthsituation throughout the world including aquaticanimal health

To collect analyze and disseminate scientificveterinary information

14


To contribute expertise and encourage internationalcoordination in the control of animal diseasesincluding aquatic animal diseases

Within its mandate under the Agreement onSanitary and Phytosanitary (SPS) measures of theWorld Trade Organization (WTO) to safeguard worldtrade by publishing health standards for internationaltrade in animals and animal products and

To improve the legal framework and resources ofveterinary services in the member countries

The principal policy of the OIE is to facilitateinternational trade in animals and animal productsincluding aquatic animals and their products basedon health control and preventative measures Thescope also covers food safety and animal welfare

Over the years the OIE had and still has animportant role to play in establishing a framework thatmay be used for strategic planning anddecisionmaking in OIE member countries

Publication of Standards

The expansion of international trade in animals andanimal products including aquatic animals since the1960s has called for

Appropriate veterinary regulations to avoid the riskof communicable diseases spreading to animals oreven to humans

Standardization of animal health requirementsapplicable to trade to avoid unnecessary hindrancesand

Harmonization of international animal healthregulations (This is critically important to ensuregrowth in international trade while maintainingeffective national disease control)

The International Aquatic Animal Health Code(OIE 2002a) (henceforth referred to as the AquaticCode) drawn up by the FDC meets theserequirements Provisions are given as guidelines forthe preparation of veterinary regulations for importand export This regularly updated collection ofrecommended veterinary requirements for

international trade takes into account specialconditions prevailing in various countries and offersappropriate solutions for each one

The Aquatic Code which was approved for thefirst time by the International Committee in May 1995(Haringstein 1996) currently covers a list of 13 ldquodiseasesnotifiable to the OIErdquo and a list of 18 ldquoother significantdiseasesrdquo of aquatic animals of which the inter-national community needs to be aware Diseaseshave been classified into one of these two lists on thebasis of their socioeconomic importance geographicrange and etiology

Prior to 1995 when the first edition of theAquatic Code and the Diagnostic Manual for AquaticAnimal Diseases (OIE 2000) (henceforth referred toas the Diagnostic Manual) were adopted by the OIEInternational Committee aquatic animal diseaseswere included in the OIE International Animal HealthCode (OIE 2002b) which covers diseases ofterrestrial animals This way of providing recom-mendations for sanitary measures to be applied toaquatic animals held obvious drawbacks A decisionto publish a separate Code and Manual for aquaticanimals was thus initially taken in the late 1980s TheFDC carried out the onerous work of preparing bothtexts with a view to producing a separate set ofdocuments based on the same template as that forterrestrial animals but bearing in mind the specificcriteria needed for aquatic animals In addition tomaterial provided by members of the FDCcontributions have also been made by the OIEInternational Animal Health Code Commission theOIE Standards Commission and scientific experts invarious OIE member countries Assistance wasprovided through comments and information neededas well as through preparation of certain chapters ondiseases for which the FDC members themselves didnot possess the necessary expertise The end resultis thus based on international teamwork

Not all countries will be able to comply witheach of the specifications detailed in the AquaticCode and Diagnostic Manual probably only a fewcountries will have the necessary resources Inaddition many countries may still have concerns that

15

Role and Function of the OIE Fish Diseases Commissionin the Field of Aquatic Animal Health

they would like to see resolved before subscribing toall the principles given in the Aquatic Code namelyzoning principles health certification etc

The Aquatic Code sets out general principlescovering

Definitions

Notification systems

Veterinary ethics and certification for internationaltrade

Import risk analysis

Importexport procedures

Contingency plans

Diseases notifiable to the OIE (fish molluscs andcrustaceans)

Other significant diseases (fish molluscs andcrustaceans)

Health control and hygiene and

Model international aquatic animal healthcertificates approved by the OIE

Currently the FDC is preparing a chapter onfallowing of aquaculture enterprises This chapterhas not yet been approved by the OIE InternationalCommittee

Notifications andEpidemiologic Information

The urgency of dispatching information variesaccording to the nature of the disease The OIE hasdevised a warning system whereby membercountries can take action rapidly should the needarise Countries are required to notify the CentralBureau within 24 hours if there isare

A first occurrence or recurrence of a diseasenotifiable to the OIE if the country or zone waspreviously considered to be free of that particulardisease

The emergence of new important findings or aprovisional diagnosis of diseases notifiable to the OIE

that are of epidemiologic significance to othercountries or

New findings (for disease not notifiable to the OIE)that are likely to have exceptional epidemiologicsignificance to other countries

The OIE immediately dispatches the incomingdata by telex fax or electronic mail directly tomember countries at risk and in weekly announce-ments (in Disease Information) to other countries

In addition to this alert system informationreceived from member countries is distributed on aperiodical basis in the monthly Bulletin the annualWorld Animal Health Yearbooks which provideannual animal health statistics and give data on theoccurrence of diseases in each member country andthe annual Animal Health Status reports for all OIEmember countries

Although the responsibility for aquatic animalhealth in many OIE member countries lies withcompetent authorities other than the veterinaryadministration nevertheless within the OIE systemthe veterinary services have the responsibility fordisease reporting in conjunction with the competentauthority in a given country Thus close cooperationbetween the veterinary services in a given countryand the responsible authority for aquatic animalhealth is strongly needed

Veterinary Ethics andCertification for InternationalTrade

There may be different philosophies and opinions inregard health certification ethics etc in differentcountries but the Aquatic Code describes basicprinciples that should be taken into account to ensurethat trade is unimpeded and that such trade does notconstitute a risk to aquatic animal health Informationon the aquatic animal health situation worldwide isthus important in order to diminish the risk of diseasetransfer through international trade in aquatic animalsand their products Certification must be based onthe strictest possible ethical rules

16


The chapters in this section of the Aquatic Codepresent the general requirements and principles ofcertification to be followed

Import Risk Analysis

Any importation of aquatic animals or relatedproducts may involve a risk of disease transfer to theimporting country The Aquatic Code chapter onimport risk analysis provides an objective anddefensible method for assessing risks associatedwith importation This basis enables the importingand exporting countries to have productivediscussions on problems associated with thepotential risks

Import and Export Procedures

In the context of import and export it isimportant to have general arrangements fortransportation of aquatic animals The chapters inthis section of the Aquatic Code providerecommendations for transport and aquatic animalhealth measures before during and upon arrival forfrontier posts in importing countries as well asmeasures concerning international transfer ofpathological material and biological products

Contingency Plans

A number of diseases pose a threat to aquacultureas well as to wild stocks of aquatic animalsworldwide therefore such diseases may causesignificant losses if introduced into countries wherethey are not established Bearing this in mind alldisease control should be based on a legislativeframework that includes contingency plans Acontingency plan can be defined as an establishedplan that is designed to have a rational approach foractions to be taken if emergency situations occur andin which all types of required actions should havebeen considered and described in advance TheAquatic Code gives guidelines for contingencyplanning

OIE International AquaticAnimal Health Code Listingof Pathogens and Diseases

In the International Animal Health Code notifiablediseases are divided into Lists A and B on the basisof their seriousness In the Aquatic Code asexplained before aquatic animal diseases areclassified as either ldquodiseases notifiable to the OIErdquo orldquoother significant diseasesrdquo

Diseases and pathogens are included in theAquatic Code according to the following basicconsiderations resistance or response to therapygeographic range and socioeconomic importanceThe list of diseases and pathogens considered forinclusion in the Aquatic Code is currently restricted tothose affecting fish molluscs or crustaceans (table1) Proposals for diseases to be listed may comefrom member countries or from the FDC andthorough discussions take place before a newdisease is actually added to either list Similarprocedures are followed for deletion of diseases fromthe lists

Categorization of diseases to be listed is ofcourse open to debate and opinions on whichdiseases should be listed vary greatly due to differentviews on the significance and importance of a givendisease For several years the FDC has beenworking to provide a disease categorization systemthat objectively classifies diseases for listing At theFDC meeting in June 2002 the Commissionproposed a set of criteria suitable for listing aquaticanimal diseases and this has been sent to membercountries for comments (table 2) The comments willbe considered in relation to the OIErsquos work on newprocedures for disease notification for terrestrial andaquatic animal diseases Until the new procedureshave been adopted there will be no changes to thecurrent listed diseases

Infectious salmon anemia (ISA) was firstbrought to the attention of the OIE in the early 1990sbecause there was a need for a common scientificname for the disease Until then the disease wasreferred to by numerous different names in published

17


papers The OIE recognized that disease and namedit ISA in 1990

When preparing the first edition of the AquaticCode the FDC concluded that in addition to the list ofdiseases notifiable to the OIE a ldquowaitingrdquo list shouldbe established for diseases that at a later stageshould either be listed as notifiable or should beremoved from consideration ISA was put on thiswaiting list in 1992 and included in the list of other

Table 1mdashDiseases of aquatic animals

Diseases Notifiable to the OIE

Fish DiseasesEpizootic hematopoietic necrosisInfectious hematopoietic necrosisOncorhynchus masou virus diseaseSpring viremia in carpViral hemorrhagic septicemia

Molluscan DiseasesBonamiosis (Bonamia exitiosus B ostrea Mikrocytos roughley)MSX disease (Haplosporidium nelsoni)Marteiliosis (Marteilia refringens M sydneyi)Mikrocytosis (Mikrocytos mackini)Perkinsosis (Perkinsus marinus P olseniatlanticus)

Crustacean DiseasesTaura syndromeWhite spot diseaseYellowhead disease

Other Significant Diseases in Fish Molluscs and Crustaceans

Fish DiseasesChannel catfish virus diseaseInfectious pancreatic necrosisInfectious salmon anemiaViral encephalopathy and retinopathyRed sea bream iridoviral diseaseWhite sturgeon iridoviral diseaseBacterial kidney disease (Renibacterium salmoninarum)Enteric septicemia (Edwardsiella ictaluri)Piscirickettsiosis (Piscirickettsia salmonis)Epizootic ulcerative syndromeGyrodactylosis (Gyrodactylus salaris)

Molluscan DiseasesSSO disease (Haplosporidium costale)Withering syndrome of abalone (Candidatus Xenohaliotiscaliforniensis)

Crustacean DiseasesCrayfish plague (Aphanomyces astaci)Infectious hypodermal and haematopoietic necrosisSpawner-isolated mortality virusSpherical baculovirosis (Penaeus monodon-type baculovirus)Tetrahedral baculovirosis (Baculovirus penaei)

significant diseases when that list was created in1993 Thus ISA was on the list of other significantdiseases in the 1995 edition of the Aquatic CodeISA was placed on this list and not on the list ofdiseases notifiable to the OIE because at that timeISA was not yet sufficiently defined Its etiology wasnot understood well enough and approved diagnosticmethods were not available On more recentoccasions the question of listing ISA as a notifiabledisease has been raised but it has been decided thatno change in listing should be effected until a newnotification system for aquatic animal diseases hasbeen approved

As stated previously the current notificationsystem requires that new findings of a nonnotifiabledisease such as ISA in aquatic animals shall bereported to the OIE immediately if it is of exceptionalepidemiologic significance ISA falls into thiscategory and occurrence of this disease should bereported to the OIE immediately when ISA isdiagnosed in a country for the first time Table 3 liststhe OIE member countries that have reported ISA orISA virus (ISAv)

Health Control and Hygiene

Health control and hygiene prior to internationaltrade in live aquatic animals and their products is alsoan important issue This section of the Aquatic Codeprovides guidelines for hygienic precautions such asthe destruction of pathogens through disinfectionprocedures in farms producing fish molluscs orcrustaceans This section is currently being updatedfor the next edition of the Aquatic Code Further-more a general chapter on procedures fordestruction of carcasses in connection with anoutbreak of disease is in preparation

Model International Aquatic AnimalHealth Certificates

The Aquatic Code contains five different modelhealth certificates that standardize certificationpaperwork worldwide The certificates arecontinuously being improved following commentsreceived from OIE member countries

18


Table 2mdashProposed criteria for listing and for urgentnotification of aquatic animal diseases (June 2002)listing criteria

Parameters that support a listing Explanatory notes

Consequences Where it occurs the disease has been shown to cause The disease generally leads to losses in susceptible2

(any one suffices) significant production losses due to morbidity1 or species and morbidity or mortality is related primarilymortality on a national or multinational or (zonal regional) to the agent and not to management or environmentallevel factors

The disease has been shown to or is strongly See abovesuspected to negatively affect wild aquatic animalpopulations that are an asset worth protecting

The agent is of public health concern

Spread Infectious etiology of the disease is proven(either of the firsttwo plus the third An infectious agent is strongly associated with the Infectious diseases of unknown etiology can haveand fourth) disease but the etiology is not yet known equally high-risk implications as those diseases where

the infectious etiology is proven While diseaseoccurrence data are gathered research should beconducted to elucidate the etiology of the disease andthe results be made available within a reasonable periodof time

Potential for international spread including via live Under international trading practices the entry andanimals their products and inanimate objects establishment of the disease is likely

Several countries or zones are free of the disease Free countries or zones could still be protected Listingbased on the recommendations of the Code and of diseases that are ubiquitous or extremely widespreadManual would render notification infeasible however individual

countries that run a control program on such a diseasecan demand its listing provided that they haveundertaken a scientific evaluation to support theirrequest Examples may be the protection of broodstockfrom widespread diseases or the protection ofthe last remaining free zones from a widespreaddisease

Diagnosis A repeatable robust means of detection or A diagnosis test should be widely available or hasdiagnosis exists undergone a formal standardization and validation

process using routine field samples (see OIEDiagnostic Manual for Aquatic Animal Diseases)

Urgent Notification

Listed diseases

First occurrence or recurrence of a disease in a country or zone of a country if the country or zone of the country was previously consid-ered to be free of that particular disease

Occurrence in a new host species

New pathogen strains or new disease manifestation

Potential for international spread of the disease

Zoonotic potential

Nonlisted diseases

Emerging disease or pathogenic agent if there are findings that are of epidemiologic significance to other countries

1`ldquoMorbidityrdquo includes for example loss of production due to spawning failure2 ldquoSusceptiblerdquo is not restricted to ldquosusceptible to clinical diseaserdquo but includes ldquosusceptible to covert infectionsrdquo

19


Table 3mdashCountries having reported the detection ofISA andor ISA virus

Year of firstCountry detection Region Species

Canada 1997 (1996) New Brunswick Atlantic salmonNova ScotiaCape BretonIslandMagaguadavicRiver

Chile 2001 Region X and XI Coho salmon

Faroe 2000 Streymoy Atlantic salmonIslands Eysturoy Kunoy[Denmark] Bor oy Su uroy

(10 differentlocations)

Ireland 2002 County Mayo Rainbow troutwestern Ireland

Norway 1984 Several counties Atlantic salmon

United 1998 Scotland Atlantic salmonKingdom including

Shetland Islands

United 2001 Maine Atlantic salmonStates

Diagnostic Manual for AquaticAnimal Diseases

In accordance with the current listed diseases in theAquatic Code the FDC has prepared pertinentupdated chapters in the Diagnostic Manual forAquatic Animal Diseases In addition to updateddiagnostic chapters on the listed diseases theDiagnostic Manual provides a general basis forhealth surveillance or control programs for fishmolluscs and crustaceans Chapters on qualitymanagement in veterinary diagnostic laboratoriesand principles of validation of diagnostic assays forinfectious diseases are also included Theseimportant chapters were originally prepared for theOIE Manual of Standards for Diagnostic Tests andVaccines (for diseases of mammals birds and bees)but because the principles are the same fordiagnostic work in aquatic animals the FDC adaptedthe chapters into the Diagnostic Manual for AquaticAnimal Diseases

The Diagnostic Manual consists of the followingsections

General provisions (definitions qualitymanagement in veterinary diagnostic laboratoriesprinciples of validation of diagnostic assays forinfectious diseases)

Separate diagnostic chapters for diseases offish molluscs and crustaceans and a

List of reference laboratories and collaboratingcenters for diseases of fish molluscs andcrustaceans

A comprehensive approach to health control inaquatic animals requires many elements such as

Assessment and maintenance of health status

Sampling screening and diagnostic methods

Verification of diagnoses

Eradication procedures and

Constraints of restocking in open waters andfarming facilities

Each chapter in the Diagnostic Manual iswritten by one or more distinguished experts in thefield based on the outline given above Thesechapters describe the latest methodology for thediagnosis of the disease in question

The Diagnostic Manual also sets standards forscreening and diagnostic methods for diseaseswhich may be applied in any diagnostic laboratoryworking with diseases of aquatic animals In additionto the more conventional methods for isolation andidentification of a putative disease agent using cellcultures culture media for bacteria and fungi indirectfluorescent antibody test (IFAT) enzyme-linkedimmunosorbent assay (ELISA) bacterial isolationand identification histology immunohistochemistryand standard parasitological methods more recentlydeveloped techniques such as the polymerase chainreaction (PCR) are described for diagnosticpurposes for many of the OIE listed diseases

The FDC is currently amending all thediagnostic chapters in the Diagnostic Manual In thenew diagnostic chapter for ISA the FDC proposes aminimum set of criteria for suspicion and verification

ot ot

20


of the disease These criteria refer to reasonablegrounds for suspecting fish of being infected withISAv and the steps that competent authorities shouldfollow to verify the presence or absence of ISA

The proposed criteria for suspicion follow Thepresence of ISA should be suspected if any of thecriteria in (a) through (e) are met

(a) The presence of postmortem findings consistentwith ISA as described in section 21 of the ISAdiagnostic chapter with or without clinical signs

(b) Isolation and identification of ISAv in cell culturefrom a single sample from any fish on the farm asdescribed in part 22 of the ISA diagnostic chapter

(c) Reasonable evidence of the presence of ISAvfrom laboratory tests such as IFAT (231) and RTndashPCR (232)

(d) The transfer of live fish into a farm where thereare reasonable grounds to suspect that ISA waspresent at the time of the fish transfer or

(e) Where an investigation reveals other substantialepidemiologic links to farms with suspected orconfirmed cases of ISA

If immediate investigations do not confirm or rule outthe presence of ISA suspicion of ISA can officially beruled out when following continued investigationsinvolving at least one clinical inspection per month fora period of 6 months and no further significantevidence for the presence of ISA is obtained

Confirmation of ISA

The presence of ISA is officially confirmed if any oneof these three criteria has been met

Clinical signs and postmortem findings of ISA inaccordance with the criteria described in section211 212 and 213 of the ISA chapter in theDiagnostic Manual are detected and ISAv is detectedby one or more of the following methods

(a) Isolation and identification of ISAv in cell culturefrom at least one sample from any fish on the farm

as described in section 22 of the ISA chapter inthe Diagnostic Manual

(b) Detection of ISAv in tissues or tissuepreparations by means of specific antibodiesagainst ISAv (eg IFAT on kidney imprints) asdescribed in part 231 in the ISA chapter in theDiagnostic Manual or

(c) Detection of ISAv by means of RTndashPCR by themethods described in section 232 of the ISAchapter in the Diagnostic Manual

Isolation and identification of ISAv in two samplesfrom one or more fish at the farm tested on separateoccasions using the method described in section 22of the ISA chapter in Diagnostic Manual

Isolation and identification of ISAv from at least onesample from any fish on the farm using the describedmethod in Diagnostic Manual with corroboratingevidence of ISAv in tissue preparations from any fishon the farm using either IFAT or PCR as described inthe Diagnostic Manual

If the principles given above are approved bythe OIE International Committee during its generalsession in 2003 similar criteria will most likely beprepared in consistent manner for the other OIE-listed diseases

OIE Reference Laboratoriesand Collaborating Center forDiseases of Fish Molluscsand Crustaceans

The Diagnostic Manual also lists the OIE referencelaboratories and collaborating center for aquaticanimal diseases including the name of theresponsible reference expert Currently there are22 reference laboratories for diagnosis controlresearch and training for the OIE-listed diseases inaquatic animals as well as one collaborating centerthat covers more horizontal themes of importancesuch as information on aquatic animal diseases Thereference laboratories share nine objectives

21


Provision of a center of expertise on a disease andstandardization of methodology

Storage and distribution of standard strains anddiagnostic standards antisera antigens and otherreagents

Development of new methods

Collection processing and analysis ofepizootiological data

Provision of consultant assistance to the OIE

Training in specific areas

Organization of scientific meetings on behalf of theOIE

Coordination of collaborative studies and

Publication and dissemination of relevantinformationThe OIE collaborating center forinformation on aquatic animal diseases is located atCenter for Environment Fisheries and AquacultureScience (CEFAS) Weymouth laboratory UK andhas established an international database for aquaticanimal diseases This database will also include amapping facility to display the geographic distributionof OIE-listed aquatic animal diseases

Updates of the InternationalAquatic Animal Health Codeand Diagnostic Manual forAquatic Animal Diseases

Since 1995 the Aquatic Code and Diagnostic Manualhas been updated regularly It has now been decidedthat the Aquatic Code should be published annuallywhile the Diagnostic Manual should be publishedevery 2 years Both publications may be ordereddirectly from the OIE in Paris The current version ofthe Aquatic Code and Diagnostic Manual may alsobe downloaded from the OIE Fish DiseasesCommission Web site (httpwwwoieintfdcengen_fdchtm)

Other Scientific and TechnicalInformation

The OIE Scientific and Technical Review is a thrice-yearly publication that includes comprehensivereviews original articles and communicationsRegional conference proceedings contain the texts ofcommunications presented at conferences of each ofthe five regional commissions Other publicationscover a variety of subjects ranging from brucellosisand orbiviruses to chemotherapy and successfulaquatic animal disease emergency programs

Most of these publications are available inEnglish French and Spanish The Aquatic Code andDiagnostic Manual can be ordered on the FDC Website or downloaded from the page

International Conferences

A number of international conferences on aquaticanimal health matters have been organized orcosponsored by the OIE in cooperation with otherorganizations The following list includes the moreimportant ones

Symposium on Fish Vaccination (Paris 1984)

Symposium on Problems of Chemotherapy inAquaculture From Theory to Reality (Paris 1991)

International Conference on Preventing the Spread ofAquatic Animal Diseases Through International Trade(Paris 1995)

OIE International Conference on Risk Analysis inAquatic Animal Health (Paris 2000)

Liaison With OtherInternational Organizations

The OIE has formal agreements with severalinternational organizations such as the Food andAgriculture Organization of the United Nations theWorld Health Organization the World TradeOrganization (WTO) and the Inter-American Institutefor Co-operation on Agriculture The OIE

22


collaborates with these organizations to ensure thatclose working relationships exist on debated issuesconcerning health controls in international trade TheOIE also enjoys privileged relations with numerousother international organizations Table 4 gives anoverview of organizations with which OIE has formalarrangements

Conclusions

Although many publications deal with the diagnosisand control of aquatic animal diseases (the AmericanFisheries Societyrsquos ldquoBlue Bookrdquo [Amos 1085] theCanadian regulatory manual [Canada Department ofFisheries and Oceans 1977] European UnionCouncil Directives 9167 and 9353 and thepublication listed in the bibliography at Food andAgriculture Organization [2001]) the Aquatic Codeand the Diagnostic Manual so far represent the ldquogoldstandardrdquo as the SPF Agreement of the WTO refersto the OIE as the key organization on the setting ofanimal health standards and guidelines

An advantage of the Aquatic Code is that it isfounded on a worldwide organization that is politicallyindependent and that has a high level of competencein the setting of standards for the control of diseasesof terrestrial as well as aquatic animals

The Aquatic Code itself can ensure that similarstandards (legal ethical etc) are followed incertification Strict ethical and moral standards areprerequisites to control and prevent the spread ofdiseases of aquatic animals through internationaltrade Appropriate certification based on sound anduniform international standards will facilitate tradeand ensure that live aquatic animals and theirproducts are free from disease(s) prevalent in theexporting country

If the principles of the Diagnostic Manual areapplied throughout the world a uniform approach tothe diagnosis of the OIE-listed diseases will beachieved A uniform approach will help importers andexporters to meet the requirements for healthcertification specified in the Aquatic Code and thusfacilitate international trade

References Cited

Amos K H ed 1985 [Blue Book] Procedures forthe detection and identification of certain fishpathogens 3d ed Corvallis OR American FisheriesSociety Fish Health Section 139 p

Canada Department of Fisheries and Oceans 1977Fish health protection regulations manual ofcompliance Misc Spec Publ 31 Ottawa ONCanada Department of Fisheries and Oceans 32 p

Table 4mdashOrganizations cooperating with OIE underformal agreements

Food and Agriculture Organization of the United Nations

World Health Organization

Inter-American Institute for Cooperation on Agriculture

World Trade Organization

Organismo Internacional Regional de Sanidad Agropecuaria

Secretariat of the Pacific Community

Pan American Health Organization


International Bank for Reconstruction and Development and the International Development Association (World Bank)

World Veterinary Association

CAB International on sharing of resources in the Animal Health and Production Compendium

International Federation for Animal Health

Organization of African UnityndashInterafrican Bureau for Animal Resources

23


Council Directive of 28 January 1991 concerning theanimal health conditions governing the placing on themarket of aquaculture animals and products (9167EEC) 18 p

Council Directive 9353EEC of June 1993introducing minimum Community measures for thecontrol of certain fish diseases 11 p

Food and Agriculture Organization 2001 Manualprocedures for the implementation of the Asiaregional technical guidelines on health managementfor the responsible movement of live aquatic animalsFish Tech Pap 4021 [Place of publicationunknown] Food and Agriculture Organization UnitedNations 106 p

Haringstein T 1996 Preparation and application of theInternational Aquatic Animal Health Code andDiagnostic Manual for Aquatic Animal Diseases ofthe Office International des Epizooties RevueScientifique et Technique Office International desEpizooties 15(2) 723ndash731

Office International des Epizooties 2000 Diagnosticmanual for aquatic animal diseases 3d ed ParisOffice International des Epizooties 237 p

Office International des Epizooties 2002aInternational aquatic animal health code 5th edParis Office International des Epizooties 157 p

Office International des Epizooties 2002bInternational animal health code 11th ed ParisOffice International des Epizooties 485 p

24


25

A Comparative Review of Diagnostic Assays Used to DetectInfectious Salmon Anemia Virus in the United States

Abstract Although there are several detection techniquesfor infectious salmon anemia virus (ISAv) none of theseassays has yet been validated by reference authoritiessuch as the Office International des Epizooties (OIE) or theNational Veterinary Services Laboratories in Ames IAEach diagnostic test discussed herein has problems thatconfound straightforward pathogen detection interpretationof results or both Analytic and diagnostic variables ofsensitivity and specificity for ISAv detection assays used inthe United States will be discussed

Introduction

Detection methods for ISAv are most often used invarious combinations to help veterinarians salmonproducers and regulators decide on pathogen ordisease-management strategies that directly orindirectly depend on an assayrsquos reliability factors

Micro Technologies Inc of Richmond ME(later referred to in this paper as ldquothe laboratoryrdquo)has refined and used several ISAv detection assayssince 1998 The laboratory is approved by the USDepartment of Agriculturersquos (USDA) Animal and PlantHealth Inspection Service (APHIS) for the detectionof 19 aquatic pathogens including ISAv Thelaboratory has collected processed tested andarchived many thousands of Atlantic salmon (Salmosalar) samples each year since 1996 for reasons offish health certification facilitation of movement fromone facility to another elective diagnostics andbroodstock management The laboratory alsoparticipates in the USDAndashAPHIS-administeredInfectious Salmon Anemia Program in Maine andprovides monthly surveillance tests at active salmonproduction sites

Using summaries of data collected by thelaboratory since 1998 I will compare ISAv diagnosticassays for various absolute and relative correlationaspects to assess some of the sensitivity andspecificity components of those assays A good dealis at stake in establishing some of these parametersbecause reliable tests for ISAv detection (or for that

A Comparative Review of Diagnostic AssaysUsed To Detect Infectious Salmon AnemiaVirus in the United States

Peter L Merrill1

of any pathogen) are at the heart of risk-assessmentapproaches for aquatic systems This reliabilityapplies to the immediate needs of commercialsalmon producers and agencies concerned with thestatus of feral Atlantic salmon and to the developmentof an approval rating system for farms and zones

Assays for Detecting ISAv

Four diagnostic assays are commonly used fordetection of ISAv in Atlantic salmon and other finfishThese include cell culture reverse transcriptasendashpolymerase chain reaction (RTndashPCR) indirectfluorescent antibody testing (IFAT) and histo-pathology (Bouchard et al 2001 Opitz et al 2000Lovely et al 1999) Electron microscopy whichdemonstrates the presence of viral particles is notpractical under typical diagnostic lab conditions but itis useful as a reference standard An enzyme-linkedimmunosorbent assay (ELISA) for detection of ISAvantibodies has also been developed (Kibenge et al2002 S Clouthier Maine BioTek personalcommunication)

Symptoms Associated WithISA

In addition there is a set of physical criteriaassociated with but by no means pathognomonic forthe clinical manifestation of the eponymous diseaseThese include exophthalmia lethargy darkening inexternal appearance petechial hemorrhaging on theskin and surfaces of internal tissues or organsascites hepatic darkening hepato- or splenomegalyforegut darkening and variably pale gills and heart(Evenson et al 1991) Hematocrits have been foundas low as 10 percent or less in fish with advancedclinical disease (Thorud and Djupvik 1988) While allthese conditions are nonspecific indicators ofdisease in connection with mortality they maycollectively allow for a tentative field diagnosis of ISA

1Peter Merrill is an aquatic species veterinarian with MicroTechnologies Inc in Richmond ME

26


Lab Tests

In North America lethal testing using primarily kidneytissue has been the norm for ISAv detection since thepathogen was first found in New Brunswick Canadain 1996 (Department of Fisheries and Oceans 1978Thoesen 1994 Office International des Epizootes2000) More recently the laboratory hasexperimented with nonlethal testing techniques usingblood samples for cell culture and RTndashPCRSerological tools have been developed to detectantibody (Kibenge et al 2002) which might helpassess or differentiate ISAv antibody levels invaccinated and nonvaccinated fish Severalenvironmentally based assays are also indevelopment by the laboratory to characterizeepizootiological variables involved with transmissionand contagion These assays refine techniques usedin RTndashPCR and cell-culture testing of fish butalternatively use fomites (such as netpen materialsboat hulls and other equipment) parasitic vectorssuch as sea lice (Lepeophtheirus salmonis) possiblesentinel-type species like shellfish sediment andseawater itself

There are a number of factors that potentiallyconfound or limit correlation of results among assaysThe sample choice itself is of prime importance InMaine Atlantic salmon are tested for ISAv for one offive principal reasons to establish or maintain facilitycertification status to transfer fish from one locationto another to screen broodstock to monitor underthe USDAndashAPHIS ISA surveillance program and toelectively diagnose unexplained elevated mortalityOther salmonid or nonsalmonid finfish are tested forISAv on a surveillance basis under State or Federalprograms Objectives for these programs may beentirely different from those among commercialsalmon producers However if ascertaining thepresence or absence of the pathogen is thedeterminant for testing a statistically relevant numberof fish must be tested to maximize the probability ofdetection in a population

Sampling

Sample numbers for many certification programsoften used a test power of 095 and a 5-percentpresumption of pathogen prevalence (Department ofFisheries and Oceans 1978 Thoesen 1994) Thusapproximately 60 fish would be selected from anypopulation of more than 300 individuals Howeverthe viral infection rate might be substantially less than5 percent at the beginning of an epizootic or thevirus might be present in more than 5 percent of apopulation but not have replicated to a detectablethreshold Other factors such as changes in viralinfectivity vaccine status genetic strain susceptibilitynutrition temperature sea lice numbers and priortherapeutic treatments may all affect the relationshipbetween sample selection and diagnostic information(Falk and Dannevig 1995a Totland et al 1996 Opitzet al 2000) Pathogen load in the environment isprobably another important variable (Nylund et al1994) There may be a minimal infectivity thresholdfor ISAv to establish itself in an individual fish or apopulation but this has not been assessed per seand probably depends on many other factors whichthemselves would be difficult or impossible toquantify All of these parameters are inherent but reallimits to the basic sample selection process and aredifferent from (but related to) the diagnostic sensitivityand specificity aspects of the assays themselves

Diagnostic sample selection is often skewed toprovide better diagnostic results by using moribundfish or fish that fail to remain competitive with theircohorts (colloquially referred to as ldquoslinksrdquo orldquopinheadsrdquo) Presumably such fish would more likelybe susceptible to ISAv infection than wouldapparently healthy fish While this is probably true itmight or might not reflect actual ISAv infectiondynamics A coinfection or adverse metaboliccondition might also enhance or reduce theprobability of simultaneous ISAv infection Fish forISAv assays are commonly obtained from salmonnet-pen populations during mortality collection diveswhich occur with varying frequency during theproduction cycle In the absence of moribund fish

27


slow swimmers or pinheads in a population the nextlikely sample choice would be freshly dead fishHowever this term is subjective because the time ofdeath is not easy to verify or visually judge

Within the population subset used for samplingthe type and quantity of target tissues selected forISAv detection (dependent on the particular assay)have not been standardized worldwide The 2000edition of the OIE Diagnostic Manual lists ldquospleenheart liver and preferably kidney tissues fromclinically infected fishrdquo (italics added) as the preferredsample sources for diagnostics For cell-cultureassays the laboratory uses gill lamellae (from severalhundred secondary lamellae from a 100-g fish to adozen or so secondary lamellae from 6-kg fish) and1-cm3-sized pieces of kidney (mid- to posterior) andspleen tissue Tissues from no more than five fishare pooled into a single container to avoid diluting thechance of viral detection Reproductive fluids fromspawning fish eggs and sac-fry are also used assample sources for ISAv tests though there may beinterference problems from cytoxicity in the cell linesused to culture ISAv from such sources (Departmentof Fisheries and Oceans 1978 Thoesen 1994)

Gills are commonly collected for cell culture aspart of certification screens for other pathogens ofregulatory concern Although there appears to besufficient probability that ISAv might be detected froman infected fish with or without the use of gill lamellae(Hovland et al 1994) additional information aboutISAv presence gained from including gill tissue mightoutweigh the ensuing questions of whether the assayis detecting an exogenous or endogenous virion orvirions There have been several instances at thelaboratory where cell culture has detected ISAvwithout concurring detection by simultaneous directtissue RTndashPCR Though this situation has been rareit might be explained if a fish were not in factsystemically infected with ISAv but carrying virus onits surface area (eg gills) Although the exact routeof ISAv infection has not been elucidated it mayinclude entry through the gill lamellae (Totland et al1996) thus the use of gill tissue may be a worthwhileindicator of viral presence if only in an environmentalsense

Tissue-Collection Techniques

Actual collection techniques for sample tissuesused in various ISAv assays may influence resultsCross-contamination of samples from different fishduring collection is always possible and depends onsampler experience transportation time constraintsfatigue or sampling environment Samples aresometimes collected in the field under less-than-optimal weather conditions This may result invariability in the quantity or quality of the tissuessubmitted for assay Although it is impractical toflame-sterilize equipment in the field disinfection ofcollecting equipment (scalpels forceps etc) isessential between samples especially for RTndashPCRassays Utensils or even gloved hands with residualmucus or blood can carry enough infective tissue tocause inadvertent contamination of the assayMinimization or avoidance of contamination can beenhanced by changing scalpel blades and glovesbetween cell-culture pools after separate pensystems have been sampled or after testing differentlots of fish Assiduous cleaning and disinfectionprotocols must be followed to remove extraneousorganic andor infective material between groups ofsamples

The technique of collection is even moreimportant for IFAT Slide impressions should bemade by touching the blotted surface to the slide inone or two nonsmearing motions per impressionarea Excessive kidney material or bloodyimpressions might interfere with antibody bindingThe same piece of tissue should be used for cellculture RTndashPCR and IFAT by trimming small sectionsfor each assay A facet of the piece of kidney tissuethat is used for ISAv RTndashPCR can also be used formaking the IFAT slide impression which mayincrease correlation between those tests

Using Blood Instead of Tissue Samples

Blood from ISAv-positive fish has the potentialto be extremely useful as a nonlethal diagnostic toolpossibly supplanting the use of kidney tissue for ISAvRTndashPCR Blood smears have also been reportedly

28


used as adjunct ISAv assays (Office International desEpizooties 2000) Blood smears are easily made inthe field and can be stained with standard WrightrsquosndashRomanowsky or other commercial stains

Preserving Samples

Sample preservation and transportation to adiagnostic laboratory are important secondary factorsin the optimization of assays Cell-culture samplesare often placed into phosphate-buffered saline(PBS) during collection and sent to a laboratory forfurther processing Samples should be cool (4 degC)during transport to avoid killing the virus whichceases replication at 25 degC (Falk 1997) Thesesamples should be thoroughly homogenized anddiluted in PBS augmented with minimal essentialmedium for culturing the virus within 24 h from thetime of collection Cell lines should be inoculatedwithin 48 h thereafter Experiences at the laboratoryindicate that tissue homogenates may be frozen atndash20 degC or lower for up to 3 months without substantialloss of viral recovery

For RTndashPCR individual kidney tissuesaveraging around 025 cm3 should be placed into a110 (weightvolume) dilution of appropriatepreservative Samples may be left at roomtemperature for up to a week without loss ofsensitivity but preferably should be shipped on iceand stored at 4 degC if they are not to be processedwithin 1 week Samples archived for longer than 1month should be frozen at ndash20 degC

Impression slides for IFAT testing should be air-dried fixed in acetone for 10 min and stored in aslide box at 4 degC during shipment

Samples for histological examination should betrimmed into cassettes and kept in a 110volumevolume dilution of 10-percent neutral bufferedformalin which is changed after 24 h

Environmental samples (eg seawatersediment mussels swabs taken from fomitessurfaces etc) should be placed into appropriateclean containers and kept at 4 degC during transportSea lice may be placed in 95-percent ethanol before

processing and shipped at 4 degC without loss of viralrecovery

Under the USDAndashAPHIS ISA program attemptshave been made to standardize collectionpreservation and shipping processes throughuniformly training collection personnel A USDAndashAPHIS-accredited veterinarian must officially sign forall samples whether they were collected personallyor through a delegate

Sensitivity and Specificity

Even a perfectly sensitive or specific assay assumingone exists could still be unreliable if it is performed ina way that distorts interferes with minimizesartificially increases or entirely prevents the chancesof detection of the pathogen for which it wasdesigned A minimum of false-positive and false-negative test results and a maximum of true-positiveand true-negative test results is the goal of alldiagnostic assays The ratios of those resultscompared to some accepted standard against whichall results are judged are reflected in the sensitivityand specificity determinations of individual orcombined detection assays These determinants arereflected in positive and negative predictive valuesBecause none of the ISAv detection techniques havebeen validated no absolute standard exists Thislaboratory has modified its own protocols on manyoccasions to better optimize conflicting or confusingassay results Some of these modifications areincluded in the review of assay techniques thatfollows

ISAV Cell Culture

Several cell lines are used to culture ISAv TheSHKndash1 cell line (Dannevig et al 1997) the CHSEndash214 cell line (Bouchard et al 1999) the TO cell line(Wergeland and Jakobsen 2001) and more recentlythe ASK cell line (see Jill Rollandrsquos paper in thisbook) have been used to successfully culture ISAvDrawbacks to cell culture include the maintenance ofcell lines the incubation timelag to initial observation

29


of cytopathic effects (as much as to 21 days) theinterpretation of questionable cytopathic effects andthe additional steps involved in confirming cytopathiceffects attributable to ISAv using RTndashPCRconfirmation Nonetheless cell culture for ISAv isgenerally acknowledged as the standard againstwhich other assays are judged The potential forfalse-negative results exists when using any of thesethree lines individually but false positives are fewerwhen used in combination

The laboratory has experienced a loss ofsensitivity of the SHKndash1 cell line to ISAv infection dueto repeated passage For this reason the lab iscurrently evaluating the use of the ASK cell line forpotential principal diagnostic use

Cell-culture practices vary between differentlaboratories and different labs use different cell-culture media and buffers (Eliassen et al 2000Kibenge et al 2000 Griffiths et al 2001 Bouchard etal 1999) There is also a tendency to adjust the pHof the culture media according to personal biasesTime and repeated cell transfers may affect thesusceptibility of a cell line to a particular virus (Wolf1988) This laboratory therefore routinely tests thesusceptibility of its SHKndash1 and CHSEndash214 cell linesto ISAv infection and has found that utilizing culturemedia at a pH of 72 is not only adequate for isolationof the virus but also allows for a broad range of cell-culture susceptibility to other virus isolatesSpecifically for ISAv the laboratory has demonstratedthat the relatively lower pH of 72 has likely added toour success in culturing ISAv on the CHSE cell line(Bouchard et al 1999) Eliassen et al (2000) havealso indicated that ISAv may require a lower pH toinfect SHKndash1 cells

Cytopathic effects observed with ISAv can differin time from inoculation to first observationmorphological changes in the cell culture monolayerandor the extent of cytopathic effects in either theSHKndash1 or CHSEndash214 lines Cell cultures areroutinely incubated for 28 d

ISAv RTndashPCR

A 200-mg kidney sample should be submergedin a minimum of five volumes of RNA preservativeaccording to manufacturersrsquo specifications for 1 weekat 25 ordmC 1 month at 4 ordmC or indefinitely at ndash20 ordmCwithout nucleic acid degradation The tissue isconsidered compromised if it was not placed in RNApreservative directly after sampling from the fish andstored appropriately before and during shipment tothe laboratory

Positive controls of RNA extracted frommidkidney tissue obtained from a confirmed clinicalISAv case or supernatant from an ISAv-positive cellculture are used for each run

A commercial amplification kit is used for RTndashPCR amplification The ISAv 1D2 primer set(Mjaaland et al 1997 Blake et al 1999) is usedprimarily at the laboratory The FAndash3RAndash3 primerset (Devold et al 2000) may be used for confirmationof positive samples A modified primer set has beendeveloped at this laboratory from the ISAv 1D2primers for use with samples showing nonspecificbackground banding patterns This phenomenoncorrelates with sample degradation and commonlyoccurs with kidney samples collected from fish thathave been dead for more than 12 h Comparison ofthe sensitivity of ISAv 1D2 and FAndash3RAndash3 primersets showed no consistent differences between thetwo primer sets

The RTndashPCR products are typicallyelectrophoresed on a 2-percent agarose gel at 60 vfor 80 min along with a 100 base-pair DNA ladderGels are stained for 30 to 40 min and photographedunder ultraviolet illumination Using the ISAv 1D2primer set a 493 base-pair fragment is amplifiedfrom ISAv-positive samples Positive results arereported as an amplified band at the position where a493 base-pair fragment would be expected tomigrate based on the location of the positive controland appropriate DNA size marker bands The primerset FAndash3RAndash3 amplifies a 211 base-pair fragmentfrom ISAv-positive samples Similarly positivesamples are reported as an amplified band at the

30


position where a 211 base-pair fragment would beexpected to migrate based on the location of thepositive control and appropriate DNA size markerbands Negative results are reported as the absenceof an amplified band in the expected region If thereis any question on the size of the fragment thesample is electrophoresed again with weak positivecontrols on either side of the sample for greaterscrutiny

The RTndashPCR assay is prone to carryover orairborne contamination as previously discussedExtreme care is therefore essential in the conduct ofthis test

Both PCR and RTndashPCR detect the nucleic acidof an organism in this case a negative-sense RNAvirus and therefore cannot discern between viablevirus particles and nonviable particles TheoreticallyPCR can detect as little as a single genomictemplate If too much RNA is used in the RTndashPCRreaction multiple banding patterns or a blur may beobserved in the lane following electrophoresismaking it difficult to interpret results Because totalRNA is used in this procedure the viral RNA is alsodiluted to some degree by the cellular RNAmdasha factthat may limit assay sensitivity The absolute analyticsensitivity of this assay has not been determined butin-house laboratory comparisons with cell cultureindicated that RTndashPCR sensitivity was an order ofmagnitude higher than cell culture

The laboratory has also investigated the use ofa nested ISAv RTndashPCR procedure as a techniqueusing a second primer set (constructed of base-pairsequences contained within the first primer) toamplify products of the initial RTndashPCR reactionComparison tests of about 100 tissue samples byboth methods did not increase sensitivity

ISAvndashIFAT

Although in theory ISAvndashIFAT should be bothsensitive and specific (Falk and Dannevig 1995b) itis seemingly the most problematic of the commonlyused assays Sample collection and preservationprocesses have varied in difference to the

standardized protocol described earlier Slides arenot always collected preprocessed or shipped to thelaboratory promptly or in the same way Also thesteps involved in laboratory preparation of thesubmitted slides are numerous and technicallycomplex and therefore become subject to cumulativeartifact Positive and negative control slides areprepared by the above technique for each batch ofIFATs read at the laboratory Positive controls aremade using a 1100 dilution of previously ISAv-inoculated cell supernatants from wells that haveproduced appropriate cytopathic effects Negativecontrols are prepared from uninoculated cell wells

The monoclonal or polyclonal primary andsecondary antibodies may be obtained from severalsources and may differ in the quantity and quality ofbinding and reactivity with viral antigenFluorescence patterns for the same slide themselvesmay be inconsistent when viewed with differentmicroscopes or over time using the samemicroscope Most importantly interpretations of thegradient of fluorescence may vary with personalexperience time number of slides viewed fatigueamount of ambient light and the fluorescingwavelength of the microscope light as it changes overtime Hence a large number of potentiallyconfounding variables are inherent in this assay

The gradient of IFAT scoring from 0 (negative)to a 4+ (strongly positive) is not always a clearcutphenomenon because slides that are 99-percentldquonegativerdquo (ie showing no detectable fluorescentreactivity) may yet have one two or more individualcells showing strong characteristics of positiveantibody response This can result in a ldquosplitrdquodesignation (eg 1+2+ up to 3+4+) or a qualifiedrating (such as ldquonegativemdashtwo hot cells observedrdquo)The most difficult distinction is whether to ascribe a2+ rating or a 3+ rating to borderline cases in thosecategories because a 2+ rating is considerednegative overall and a 3+ positive overall Thegradient of variation as well as the absolute gradientof effect can be continuous or discontinuous withinan individual impression between two impressionson the same slide or between two or more slides

31


from the same kidney sample Due to the poorreproducibility of fluorescent effects using black-and-white photography visual images cannot be includedin this document but the laboratory is in the processof preparing a photographic manual of the ISAvndashIFATfluorescent spectrum for in-house and proficiency-testing use

At the laboratory only experienced personnelare used for IFAT reading and at least two viewersare involved in all questionable cases before a finalrating is given With the weight of many thousands ofindividual ISAv assays performed over a 5-yearperiod using different batteries of assays it has beenthe labrsquos experience that IFAT is highly prone to falsepositives and false negatives alike True positivesand true negatives however correlate well withresults by other assays This correlation has beencasually observed to occur with increasing length oftime after infection The OIE 2000 Diagnostic Manual(2000) lists IFAT as a confirmatory assay among fishexhibiting ldquopathological signsrdquo Used in such amanner IFAT results have reportedly correlated wellin the field Nonetheless it is the consensus at thislaboratory that IFAT has limited value as aconfirmatory tool While from a surveillanceperspective it may be better to err conservatively incases where farms or zones have previously testednegative for ISA mixed diagnostic results (such asnegative RTndashPCR tests with accompanying positiveIFATs for the same fish) can confuse salmonproducers and regulators This may cause extralabor at considerable expense for additional analysisIn areas of Canada where ISAv-positive cages orfarms have been found as few as two positive IFATsper cage have been the sole reason to depopulateproduction fish This threshold has been modifiedrecently to ensure that a total of four positive testsmust be found in any cage before depopulation isundertaken Elective action may be taken at lowerthresholds

Histology

Histology is useful as a confirmatory assay afterinfection has caused tissue pathology Because

ldquohealthrdquo and ldquodiseaserdquo are not true states of being butrather points along a continuum there is a gradient ofchange in each tissue or organ system affected byISAv infection that taken as a whole isrepresentative of the syndrome In early infectionfocal congestion and dilatation of hepatic sinusoidsmay be evident followed by rupture of sinusoidalendothelium and erythrocytes apparent within thespace of Disse (Office International des Epizooties2000) In later stages of disease lesions includeareas of multifocal hepatic congestion hemorrhageandor necrosis that may become confluent Thisprocess leads to a ldquozonalrdquo appearance withhepatocellular areas around large veins remainingrelatively intact In the spleen moderate-to-severesinusoidal congestion and erythrophagia have beenreported In kidneys lesions are characterized byacute tubular necrosis with eosinophilic casting andoften substantial interstitial congestion andhemorrhage (Evenson et al 1991 Falk and Dannevig1995a Lovely et al 1999 Bouchard et al 2001)Histology is not always used as a confirming toolbecause of the time involved in processing therelative slowness of reading and overall costsNonetheless characteristic lesions correlate well withan assay like ISAvndashIFAT

Environmental Testing

Environmental samples routinely tested for ISAvat this laboratory include seawater cage and boatsurface swabs suspended and bottom sediment andinvertebrates (eg sea lice [Lepeophtheirussalmonis] and mussels [Mytilus edulis])

Seawater is filtered through arrays of glass fiberand electronegative filters with manipulation of thepH during various steps in order to capture any virusparticles that may be present (Abbaszadegan et al1999 Gilgen et al 1997) Ten L of seawater canreasonably be reduced to a 20-mL concentratewhich is used to inoculate cell cultures or is assayedby RTndashPCR The method has been successful indetecting ISAv by both assays in control samples andby one or both assays in samples not only fromsalmon production sites experiencing clinical ISA but

32


also from sites with fish testing negative for ISAvunder the surveillance monitoring program as well

Potentially infective fomite surfaces such asharvest boat decks and hulls (Murray et al 2001) aresampled by swabbing predetermined areas withsterile sponges Swab samples are stored in 90-percent ethanol concentrated through spin columnsand extracted using methods similar to those used fortissues before they are assayed by RTndashPCR Swabsamples to be used for inoculating cell cultures aremaintained in phosphate-buffered saline andprocessed via routine viral culture procedures SomeISAv nucleic acid has been detected at the laboratoryby RTndashPCR from contaminated sea cages and fromboat bottoms using this technique and confirmedthrough DNA sequencing Parallel detection of ISAvin swab samples by viral culture has not beenobserved in all samples This may be due to theabsence of viable virus particles in the presence ofviral RNA

A virus like ISAv may have potentially multiplecoinfection factors that include a variable period ofincubation an unknown in vivo infectivity thresholdvariable and poorly characterized immunologicfactors and variable mortality thus it might be difficultto establish what the reference standards (infectionor disease) should be Once that has beenestablished diagnostic assays may be furtherevaluated

An ISAv assay may at once be accurate andunbiased without being precise sensitive yet notspecific or the converse An assay can also beperfect in all internal and external parameters but beso expensive time consuming or technically difficultto perform that it cannot be employed The laboratoryassesses these parameters when developing assays

In a practical sense analytic sensitivity refers tothe ability of an assay to detect small quantities ofwhat it was designed to detect Analytic specificity issimilarly used to define how selective an assay is fordetection of a particular pathogen Diagnosticsensitivity and specificity are just as important to alaboratory (or researcher or regulator) from astatistical perspective Diagnostic sensitivity is

characterized as a function of the number of positivetests it gives in terms of true positives and falsenegatives (ldquoTruerdquo and ldquofalserdquo refer to whateverstandard is elected against which test results arecompared) In this sense diagnostic specificityreflects the numbers of false results an assay givesin terms of true negatives and false positives

Predictive Value of Tests

Both the diagnostic sensitivity (DndashSN) andspecificity (DndashSP) of an assay or combination ofassays are integral components of the predictivevalue of those assays The positive predictive valuerepresents the probability that test subjects withpositive test results actually have the pathogen ordisease being assayed The negative predictivevalue of an assay is the probability that test subjectswith negative test results are actually free of thepathogen or disease being assayed Predictivevalues of both types may then be used to establishprevalence determinations in populations The so-called apparent prevalence can be calculated as thesum of true positives and false negatives divided bythe total number of all test results The ldquotruerdquoprevalence rate can be calculated as the number oftrue positives divided by the total number of all testresults Thus it is apparent that prevalencecomputations of either type often needed toformulate or evaluate disease control programsnecessarily relate to sensitivity and specificity valuesfor the particular assays that are used (Thrusfield1997)

Where there are unequivocal diagnosticmethods to prove or disprove test results (egmacroscopic pathogens such as Myxoboluscerebralis spores that can be visualized easily andquantified) sensitivity and specificity can beaccurately computed In the case of ISAv and othersubmicroscopic organisms sensitivity and specificityvalues are more easily estimated or expressed asprobabilities It is important to note again that thereare many potentially confounding variables that mightaffect the determination of an assayrsquos diagnosticsensitivity and specificity such as temporal variations

33


in the infective process metabolic dysfunction (arealistic concern in anadromous finfish raised infreshwater culture conditions before transfer tosaltwater production locations) cross-reactivityfactors for chemical components used in the testnonspecific inhibitors or agglutinins coinfectiontoxins immune suppression factors and blockingantibodies

Comparing Tests

Calculations using data from various studiesperformed at the laboratory under a variety ofsubmission types are presented to compare thesensitivity and specificity of various ISAv assays

Data presented in table 1 below compare ISAvRTndashPCR results using kidney tissue and blood withresults achieved through cell culture on the SHKndash1cell line for the same samples

Comparative DndashSN and DndashSP were calculatedusing standard formulae (Thrusfield 1997) assumingthat a kidney tissue-based cell culture is the goldstandard for comparison (DndashSN and DndashSP = 100)Thus kidney-tissue-based RTndashPCR results in aDndashSN of 096 and a DndashSP is 097 For blood-basedRTndashPCR DndashSN calculates as 100 and DndashSPas 061

An inference from these data suggests thatusing blood as a sample tissue for ISAv detection viacell culture or RTndashPCR is as sensitive as usingkidney tissue At a DndashSN of 100 blood is apparentlyslightly more sensitive a sample source than kidneytissue but somewhat less specific at 061 (comparedto a specificity of 097 for kidney tissue) Blood

therefore generated more ldquofalserdquo PCR-positives thandid kidney tissue

Using the same fish this time comparing IFATprocedures using kidney and blood respectively assample sources DndashSN and DndashSP for IFAT cansimilarly be computed When cell culture of kidneytissue was used as the arbitrary standard forcomparison of either sample source the number oftrue and false positives and negatives can becalculated (table 2)

From these data DndashSN for IFAT using kidneymaterial calculated as 047 and DndashSP as 100however using blood as sample source DndashSN was004 while DndashSP remained 100 The IFAT test usingeither kidney or blood smears as a sample sourcewas much less reliable for ISAv detection becausethe calculated sensitivity was considerably lower thanthat of ISAv RTndashPCR or cell culture Interestinglyspecificity for either type of IFAT was quite high in thisstudy (100) as was the positive predictive value Infield use however this level of specificity may beoffset by the low sensitivity of the assay Becauseassays with low sensitivities produce high numbers offalse negatives this would not be a desirable attributeof a test designed to detect and eliminate infectedanimals from the population

Micro Technologiesrsquo Database

From January through August of 2002 the laboratoryhas developed a database of diagnostic informationaccrued from 1053 Atlantic salmon originatingamong marine production sites in Maine Under theongoing USDAndashAPHIS-sponsored ISAv surveillance

Table 1mdashISAv RTndashPCR sensitivity and specificitycomparisons using kidney and blood from 58 fishas sample source

Tissue test result Cell culture + Cell culture ndash

Kidney RTndashPCR test + 25 1

Kidney RTndashPCR test ndash 1 31

Blood RTndashPCR test + 25 13

Blood RTndashPCR test ndash 0 20

Table 2mdashISAv IFAT sensitivity and specificitycomparisons using kidney and blood from 58 fishas sample sources


Kidney IFAT test + 8 0

Kidney IFAT test ndash 17 33

Blood IFAT test + 1 0

Blood IFAT test ndash 24 33

34


program ISAv RTndashPCR and IFAT were used to assaythose samples (table 3) Cell culture was also usedto retest samples based on initially positive resultsfrom either IFAT or RTndashPCR For this program anyIFAT rating of 2+ necessitated retesting even thougha 2+ IFAT is ordinarily considered a negative reactionFish were blindly submitted for testing from sites ofunknown ISAv status

Eight positive ISAv RTndashPCR results and 171non-zero-graded IFATs were obtained some resultswere from retesting of sites with positive testsFluorescent antibody results produced 111 sampleswith a 1+ rating 45 with 2+ rating and 15 with a 3+rating No 4+ IFAT results were observed Each ofthe 8 ISAv-positive RTndashPCR results and 11 of the 153+ IFAT ratings came from a site with subsequentlyconfirmed ISAv infection Excluding that site theremaining 160 non-zero-graded IFATs were notsupported by RTndashPCR results Such disagreementmay reflect variables associated with sensitivity orspecificity of the assays viability of the pathogens orother unknown factors

Results From a Two-Lab Study

In late 2001 a study examining several comparativediagnostic parameters was undertaken between 2labs using a total of 60 Atlantic salmon exposedeither naturally in the field (and logically throughsubsequent cohabitation in the lab tanks) orexperimentally exposed to ISAv along with 2negative controls Assays included ISAv RTndashPCRusing blood and kidney tissue in addition to virusisolation using SHKndash1 and CHSEndash214 cell lines onindividual fish pools Hanksrsquo balanced salt solutionwas used as a transport medium for this study as acomparative sample preservative

Although 62 percent (37 of 60) of fish selectedfor inclusion in this study from the field had relevantclinical signs fish not demonstrating clinical signsalso tested ISAv-positive by various assaysInspection of the PCR testing results indicated that100 percent of the fish were infected with or carryingISAv (table 4) There was good interlaboratorycorrelation for total ISAv RTndashPCR results using eitherblood or kidney tissue Excellent correlation alsoexisted between blood and kidney as sample tissuefor the ISAv RTndashPCR assay One immediatelyapparent difference in results is for virus isolationusing SHKndash1 cells where one lab failed to cultureany virus from more than 60 samples that had testedoverwhelmingly positive through ISAv RTndashPCR Theother laboratory cultured ISAv from the populationwith both cell lines although at the success rates of

Table 4mdashComparative diagnostic parameterassessments for ISAv tests 2001

RTndashPCR RTndashPCR SHK VI+ CHSE VI+ of Clinical (kidney) (blood) kidneyblood kidneyblood

Fish origin fish signs+ Lab 1 Lab 2 Lab 1 Lab 2 Lab 1 Lab 2 Lab 1 Lab 2

Neg control 2 0 0 0 0 0 00 00 Na 00

Naturally exposed 20 18 19 20 20 20 00 127 Na 41

Experimentally infected 40 19 40 40 40 37 00 1818 Na 912

Totals 62 37 59 60 60 57 00 3025 Na 1313

Lab 1 did not perform culture using CHSEndash214 cells

Table 3mdashComparative surveillance testing results forISAv from 1053 Atlantic salmon

RTndashPCR results ISAv IFAT resultsNeg Pos 0 1+ 2+ 3+

1045 8 882 111 45 15

35


50 percent v 42 percent for SHKndash1 cells using kidneyand blood respectively as sources for viral isolationOnly one laboratory used the CHSEndash214 line with aculture rate of 22 percent for both kidney and bloodFrom these data blood appears to be somewhat lesssensitive a tissue choice for cell culture than kidneyresulting in a 17 percent lower culture rate withSHKndash1 cell culture The CHSEndash214 line overall hada 56 percent lower successful culture rate comparedto SHK cells Based on this apparent difference theSHKndash1 cell line was more sensitive to ISAv infectionthan the CHSEndash214 line although historically thelaboratory has consistently cultured ISAvsuccessfully using both lines The use of both celllines for concurrent cell culture assays for ISAv isrecommended to increase the overall sensitivity ofthe test Both labs produced similar results onnegative control samples

IFAT Produces VariableResults

Other interlab exercises have been performed atperiodic intervals and the usual variant amonglaboratories was results produced for IFAT ratingsThis is particularly true at the lower end of thegradient where labs may disagree on whatspecifically constitutes a 1+ or 2+ rating While stillin the negative category there is a substantialqualitative and quantitative difference between theextremes of the negative range In part asmentioned in earlier sections there may bedifferences in collection preservation preparation orinterpretation One other possible factor affectinginterpretation is the number of fields read per slideWhile in theory the entire slide is scanned during theevaluation process in practice fewer than thepotential total number of fields may actually be readdepending on the size of the impression number ofslides to read time constraints etc Thisphenomenon has been noted many times at ourlaboratory and a logical conclusion that may bedrawn is that the overall negativity of a slide may be afunction of the time taken for reading and the numberof fields viewed

A true ISAv assay validation study for any assayusing the OIE-recommended number of 2000 testanimals has yet to be published but will be anecessary component of rational ISAvISAmanagement approaches internationally AlthoughRTndashPCR appears to be at least as diagnosticallysensitive as cell culture for viral detection thesignificance of the results from RTndashPCR is debatedThere is apparently enough variation betweenavailable cell lines used for virus isolation that eachshould be validated in its own right Fluorescentantibody tests appear to be useful as a detection toolat later stages of infection but that is mainly ananecdotal conclusion without supporting publishedevidence Experiences over the past 3 years at thislaboratory have demonstrated that IFAT resultsalternately correlate and disagree not only withresults from other assays but within a singlediagnostic submission as well

Standards Are Needed

Reference institutions and resource agencies need toprovide the standardizing framework for both theavailable and developing ISAV detection assays Thedetermination of a gold standard with acceptablelevels of sensitivity both generally and for particularISAv assays must be defined This determinationdepends upon the nature of the testing programbeing utilized If the goal of the program is to detectand eliminate ISAv-infected fish a highly sensitiveand fairly specific test is needed Such an assaywould have relatively few false negatives but producesome false positives Alternatively if the goal is toconfirm the results of another assay a very specifictest with reasonable sensitivity would suffice to avoidfalse positives The degree of acceptable levels ofboth sensitivity and specificity will play a deciding rolein these respects Sample size also is a determinantin the reliability aspect of the diagnostic equationbecause at the group or population level sensitivityand specificity are influenced by sample size Withlow prevalence levels of a pathogen as may be thecase with initial ISAv infection in a population even

36


very reliable tests with high sensitivity and specificityhave a relatively low predictive value Becauseprevalence can change over time the anticipatedpredictive value of an assay should be periodicallyreviewed in context to the situation as it becomesbetter characterized statistically

Acknowledgments

The author is indebted to Ms Deborah Bouchard andDr Cem Giray of Micro Technologies Inc for theirassistance with portions of this manuscript Thanksalso to Dr Rocco Cipriano of the National Fish HealthResearch Laboratory Kearneysville WV Dr HMichael Opitz of the University of Maine Orono andDr Beverly Schmitt of APHISrsquo National VeterinaryServices Laboratories in Ames IA for editorialsuggestions

References Cited

Abbaszadegan M Stewart P LeChevallier M1999 A strategy for detection of viruses ingroundwater by PCR Applied and EnvironmentalMicrobiology 65(2) 444ndash449

Blake S Bouchard D Keleher W Opitz MNicholson B L 1999 Genomic relationships of theNorth American isolate of infectious salmon anemiavirus (ISAv) to the Norwegian strain of ISAv Diseasesof Aquatic Organisms 35 139ndash144

Bouchard D Brockway K Giray C Keleher WMerrill P 2001 First report of infectious salmonanemia in the United States Bulletin of the EuropeanAssociation of Fish Pathologists 21(1) 5ndash8

Bouchard D Keleher W Opitz H M Blake SEdwards K Nicholson B 1999 Isolation ofinfectious salmon anemia virus (ISAV) from Atlanticsalmon in New Brunswick Canada Diseases ofAquatic Organisms 35 131ndash137

Dannevig B Brudeseth B Gjoen T Rode EWergeland H I Evensen Oslash Press C M 1997Characterisation of a long-term cell line (SHKndash1)developed from the head kidney of Atlantic salmon(Salmo salar L) Fish and Shellfish Immunology 7(4)213ndash226

Department of Fisheries and Oceans 1978 Fishhealth protection regulations manual of complianceMisc Spec Publ 31 rev Ottawa ON Department ofFisheries and Oceans 42 p

Devold M Krossoslashyl B Aspehaug V Nylund A2000 Use of RTndashPCR for diagnosis of infectioussalmon anemia virus (ISAv) in carrier sea trout Salmotrutta after experimental infection Diseases ofAquatic Organisms 40 9ndash18

Eliassen T Froystad M Dannevig B JankowskaM Falk K Romoren K Gjoen T 2000 Initialevents in infectious salmon anemia virus infectionevidence for the requirement of a low-pH stepJournal of Virology 74(1) 218ndash227

Evensen Oslash Thorud K Olsen Y 1991 Amorphological study of the gross and lightmicroscopic lesions of infectious salmon anemia inAtlantic salmon (Salmo salar) Research in VeterinaryScience 51 215ndash222

Falk K 1997 Properties and cell culture of ISA virusIn Workshop on infectious salmon anemiaproceedings 26 November 1997 St Andrews NBFredericton NB New Brunswick Department ofFisheries and Aquaculture 13ndash14

Falk K Dannevig B 1995a Demonstration ofinfectious salmon anemia (ISA) viral antigens in cellcultures and tissue sections Veterinary Research 26499ndash504

Falk K Dannevig B 1995b Demonstration of aprotective immune response in infectious salmonanemia (ISA)-infected Atlantic salmon Salmo salarDiseases of Aquatic Organisms 21 1ndash5

37


Falk K Press C M Landsverk T Dannevig B1995 Spleen and kidney of Atlantic salmon (Salmosalar L) show histochemical changes early in thecourse of experimentally induced infectious salmonanemia (ISA) Veterinary Immunology andImmunopathology 49 115ndash126

Gilgen M Germann D Luthy J Hubner P 1997Three-step isolation method for sensitive detection ofenterovirus rotavirus hepatitis A virus and smallround structured viruses in water samplesInternational Journal of Food Microbiology 37 189ndash199

Griffiths S Cook M Mallory B Ritchie R 2001Characterisation of ISAV proteins from cell cultureDiseases of Aquatic Organisms 45 19ndash24

Hovland T Nylund A Watanabe K Endresen C1994 Observation of infectious salmon anemia inAtlantic salmon Salmo salar L Journal of FishDiseases 17 291ndash296

Kibenge F Lyaku J Rainnie D Hammell K L2000 Growth of infectious salmon anemia virus inCHSEndash214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81 143ndash150

Kibenge M Oprazo B Rojas A Kibenge F 2002Serological evidence of infectious salmon anemiavirus (ISAV) infection in farmed fishes using anindirect enzyme-linked immunosorbent assay(ELISA) Diseases of Aquatic Organisms 51 1ndash11

Lovely J Dannevig B Falk K Hutchin LMacKinnon A Melville K Rimstad E Griffiths S1999 First identification of infectious salmon anemiavirus in North America with hemorrhagic kidneysyndrome Diseases of Aquatic Organisms 35 145ndash148

Mjaaland S Rimstad E Falk K Dannevig B H1997 Genomic characterization of the virus causinginfectious salmon anemia in Atlantic salmon (Salmosalar L) an Orthomyxo-like virus in a teleost Journalof Virology 71 7681ndash7686

Murray A G Smith R Stagg R 2001 Shippingand the spread of infectious salmon anemia inScotland Emerging Infectious Diseases 8(1) 1ndash4

Nylund A Hovland T Nilsen F Lovik P 1994Mechanisms for transmission of infectious salmonanemia (ISA) Diseases of Aquatic Organisms 1995ndash100

Office International des Epizooties 2000 Diagnosticmanual for aquatic animal diseases 3d ed ParisOffice International des Epizooties Chap 224

Opitz H M Bouchard D Anderson E Blake SNicholson B Keleher W 2000 A comparison ofmethods for the detection of experimentally inducedsub clinical infectious salmon anemia in Atlanticsalmon Bulletin of the European Association of FishPathologists 20(1) 12ndash22

Thoesen John C ed 1994 Suggested proceduresfor the detection and identification of certain finfishand shellfish pathogens 4th ed version 1Fairhaven NJ American Fisheries Society FishHealth Section [No pagination]

Thorud K Djupvik H O 1988 Infectious anemia inAtlantic salmon (Salmo salar L) Bulletin of theEuropean Association of Fish Pathologists 8 109ndash111

Thrusfield M 1997 Veterinary epidemiology 2d edOxford UK Blackwell Science Ltd Chap 17 496 p

Totland G Hjeltnes B Flood P 1996 Transmissionof infectious salmon anemia (ISA) through naturalsecretions and excretions from infected smolts ofAtlantic salmon Salmo salar during theirpresymptomatic phase Diseases of AquaticOrganisms 26 25ndash31

Wergeland H Jakobsen R 2001 A salmonid cellline (TO) for production of infectious salmon anemiaDiseases of Aquatic Organisms 44 183ndash190

Wolf K 1988 Fish viruses and fish viral diseasesIthaca NY Cornell University Press 476 p

38


39

The Development of Infectious Salmon AnemiaVirus Vaccines in Canada


Frederick SB Kibenge Molly JT Kibenge Tomy Josephand John McDougall1

Introduction

Infectious salmon anemia (ISA) is an emerginghighly fatal viral disease of marine-farmed Atlanticsalmon (Salmo salar) that is negatively affecting thesalmon-farming industry in an increasing number ofcountries particularly in the Northern HemisphereThe ISA virus (ISAv) has caused disease in Norwaysince 1984 (Jarp and Karlsen 1997) and the diseasewas probably present in Norwegian salmon farms asearly as 1977 or 1978 (Devold et al 2001) AlthoughISA outbreaks in New Brunswick Canada have beenrecognized since 1996 (Bryne et al 1998 Mullins etal 1998 Lovely et al 1999) there is anecdotalevidence that the virus was present in the Bay ofFundy by 1995 ISA outbreaks have occurred in theFaroe Islands Denmark (Anonymous 2000) and inMaine USA since 2000 (Bouchard et al 2001)Clinical disease also occurred in Scotland in 1998and 1999 (Rodger et al 1998 Murray this volume)and in Nova Scotia in 2000 (Ritchie et al 2001) InAugust 2002 ISAv was also isolated for the first timein Ireland from fish on a rainbow trout farm in Clew

Abstract Infectious salmon anemia virus (ISAv) iscurrently one of the most important viral pathogensthreatening commercial aquaculture in the NorthernHemisphere Morphological biochemical and replicationproperties which are similar to those of the influenzaviruses indicate that ISAv is a member of theOrthomyxoviridae Comparison of ISAv proteins with otherorthomyxoviruses revealed low amino acid identity values(between lt13 percent and lt25 percent) supporting theproposal to assign ISAv to a new genus Isavirus within theOrthomyxoviridae Infectious salmon anemia in the Bay ofFundy New Brunswick is now a managed diseasefollowing the compensation scheme agreed to by theCanadian Federal Government in 1999 and the varioussteps taken by the industry These steps include theadoption of stringent husbandry practices an ISA surveil-lance program depopulation of affected sites and vaccina-

tion However the ISAv vaccines currently used are not100-percent protective Immunized fish do not clear thevirus and they may become carriers which makes controlby vaccination incompatible with depopulation controlmethods There is still an absolute need to match the ISAvvaccine composition to current viruses and to improve onthe immunogenicity of ISAv vaccines Studies on morethan 160 ISAv isolates have established the existence oftwo hemagglutinin (HA) subtypes of ISAv one NorthAmerican and one European and four distinct neuramini-dase (NA) subtypes indicating the existence of up to eightdifferent combinations of HA and NA subtypes amongisolates These observations are discussed in relation tonew vaccine developments for ISA An enzyme-linkedimmunosorbent assay (ELISA) that detects fish antibodiesto ISAv was also developed and can be used to assessvaccine efficacy

Bay County Mayo without evidence of a clinicaloutbreak (Anonymous 2002) Although there is stillcontroversy about the occurrence of ISAv in Chile[Pedro Smith informed us in New Orleans that thenonstandard illness in Chilean coho salmon was notactually ISA despite appearances] laboratory dataon tissues and sera from farmed coho salmon(Oncorhynchus kisutch) support its existence in thatcountry as well (Kibenge et al 2001a 2002)

Clinical disease in farmed Atlantic salmon ischaracterized by high mortality exophthalmia severeanemia ascites hemorrhagic liver necrosis andrenal interstitial hemorrhage and tubular necrosisCurrently each country has its own procedures inplace to deal with ISA outbreaks The mostdangerous infections are the subclinical ones thatmaintain the virus within known geographic areas Itis now suggested that the disease is caused by novelvirulent strains of ISAv that may emerge frombackground benign infections in the wild fishery andthen adapt to intensive aquaculture practices (Murrayet al 2002) The virus appears to be well establishedin the wild fishery (Raynard et al 2001) and cannottherefore be eradicated Thus the aquacultureindustry will have to learn to manage it In NewBrunswick ISA is now a managed disease followingthe compensation scheme agreed to by the CanadianFederal Government in 1999 and the various stepstaken by the industry including stringent husbandrypractices and vaccination

1 Drs Frederick Kibenge and Tomy Joseph are with theAtlantic Veterinary College University of Prince EdwardIsland in Charlottetown PE Dr Molly Kibenge workedthere during preparation of this manuscript but is now withthe Department of Fisheries and Oceans in Nanaimo BCDr John McDougall is with Alpharma AS in Oslo Norway

40


The properties of ISAv indicate strongly that thevirus belongs to the family Orthomyxoviridae and issimilar to influenza viruses Given the high mutabilityof these viruses in their surface glycoproteinshemagglutinin (HA) and neuraminidase (NA) whichare the most important targets for the host immunesystem it is absolutely essential that the ISAvvaccine composition be matched to the actualcirculating viruses In humans new strains ofinfluenza A virus arise every 1ndash2 years as a result ofselected point mutations in these two proteinsmdashaphenomenon known as antigenic drift Sometimesan exchange of the HA or NA gene segment or bothwith a strain of influenza A virus of another subtypeoccurs (a phenomenon known as antigenic shift) andthis may result in an influenza pandemicConsequently human influenza vaccines are updatedyearly to induce immune responses specific for theprevalent strains These vaccines are efficaciouswhen the HA of the vaccine matches that of thecirculating virus but not when it differs as a result ofantigenic drift or shift

This paper discusses the current ISA vaccineeffort in Canada and highlights important challengesrelated to ISAv strain variation immunogenicity andpublic education These concepts are important tounderstand (1) why disease control by vaccination isincompatible with current depopulation controlmethods (2) what new vaccine developments havebeen achieved and (3) what indicators of ISAvvaccination are desired for vaccines to havesubstantial preventative value

Incidence of ISA in NewBrunswick Canada

Initial diagnosis of ISA in the Bay of Fundy wasdelayed because of the preponderance of lesions inthe kidney (rather than the liver) of affected fish andthe disease was initially termed hemorrhagic kidneysyndrome (HKS) by Byrne et al (1998) Laboratoryconfirmation of ISAv in HKS was also initially delayedbecause of the presence of a novel Toga-like virus(that was cytopathic in CHSEndash214 cells) in mixed

infection with ISAv (Kibenge et al 2000a) It was notuntil Canadian laboratories had access to the SHKndash1cell line that confirmation of the etiological role ofISAv in HKS became possible (Lovely et al 1999)Moreover the New Brunswick ISAv isolates showedphenotypic variation in that some isolates werecytopathic on CHSEndash214 cells (Bouchard et al1999) whereas others were not (Kibenge et al2000b) Until August 1997 only the premarketclasses of fish (gt 1 kg) were identified with HKSSince August 1997 mortality associated with thesyndrome has been reported in the smolt year-class(about 300ndash400 g) As of August 2002 ISAoutbreaks in the Bay of Fundy (fig 1) have cost theindustry in excess of $50 million Canadian

25

20

15

10

5

01995 1996

Atlantic salmon year-class1997 1998 1999 2000 2001

Num

ber

of IS

A-a

ffect

ed fa

rms

Figure 1mdashEstimated number of farms affected with ISA in NewBrunswick by August 2002 [Data provided by Dr SandiMcGeachy New Brunswick Department of Agriculture Fisheriesand Aquaculture]

41


Current ISAv Vaccine Usage inNew Brunswick

Increased survival of fresh-water-reared Atlanticsalmon vaccinated with inactivated whole ISAvemulsified with mineral oil and then experimentallyinfected with ISAv has been reported (Brown et al2000 Jones et al 1999b) This is the basis for thecurrent ISAv vaccines used in Canada However ithad been recognized long before this that theimmune response in Atlantic salmon did not providefull protection against the disease for mortality couldbe produced both in fish that had previouslyrecovered from experimental ISAv infection and thosethat were passively immunized with serum from fishthat had recovered from ISA (Falk and Dannevig1995) The first commercially available ISAv vaccinewas an autogenous product using the ISAv isolateNBISA01 (Jones et al 1999ab) All 1999 fishstocked in the New Brunswick parts of the Bay ofFundy were vaccinated with this vaccine Thisvaccine has since been licensed in Canada and the

Table 1mdashPotential outcomes and indicators ofvaccination against infectious salmon anemiavirus (ISAv)

Outcome Indicators

1 Sterile immunity bull No virus detected using the mostsensitive RTndashPCR assay

bull No antibodies to ISAvnonstructural proteins1

bull No clinical disease

2 Transient subclinical bull No virus transmission to2 infection naiumlve fish

bull RTndashPCR positivebull Antibodies to ISAv nonstructural

proteins presentbull No clinical disease

3 Controlled subclinical bull Virus isolation positive3 infection (virus carrier)

bull Antibodies to ISAv nonstructuralproteins present

bull No or reduced mortality (mildclinical disease)

4 Enhanced clinical bull Very high mortality4 infection (severe clinical disease)

1 If salmon were vaccinated with inactivated whole ISAv

United States as Forte V1 by Norvatis (Aqua Health)Since 2001 a second inactivated ISAv vaccine(Brown et al 2000) has also been marketed in NewBrunswick For all intents and purposes practicallyall farmed Atlantic salmon currently stocked in theBay of Fundy are vaccinated against ISA

Several potential outcomes might result fromvaccination against ISAv (table 1) Vaccination willonly have substantial preventive value if the ISAvvaccines induce broadly neutralizing antibodies andthe immunized fish do not become virus carriers The1999 autogenous vaccine was not completelyeffective because more than 1 million of thevaccinated fish had to be removed early after someof them tested positive for ISAv In addition fish atseveral new sites became infected However ISAoutbreaks have continued to occur even when thelicensed ISAv vaccines which have controlledantigen content and long-term stability have beenused (McGeachy 2001) A major challenge withcurrent ISAv vaccine use in Canada is that the fieldperformance of these vaccines cannot be adequatelyevaluated because of the control strategy wherebyany cage with in excess of 005 percent mortality perweek is depopulated Surveys for ISAv in Atlanticsalmon by reverse-transcriptasendashpolymerase chainreaction (RTndashPCR) on different fish farms havedisclosed that virus was present in fish withoutclinical signs of disease from some cages (McClureand Hammell this volume) This fact suggests thatalthough early harvest of cages involved withoutbreaks removes a source of the virus theunaffected cages that remain on a positive farm maybecome a reservoir of the virus However thesenonclinical cages may be the ones in which the ISAvvaccine is protective against clinical disease (table 1outcome 3) Thus the immunity engendered bythese vaccines is neither 100 percent protective norsterile

Researchers in a recent laboratory study(McDougall et al 2001) observed that vaccineefficacy correlated with protection from mortality(clinical disease) and elimination of virus inchallenged fish but surviving fish tested positive forthe virus

42


The Concept of StrainVariation in ISAv

The morphological biochemical and replicationproperties of ISAv indicate strongly that this virus is amember of the Orthomyxoviridae family (Kim andLeong 1999) and it is similar to influenza viruses(fig 2) The single-stranded RNA genome of ISAvcomprises eight segments of negative polarityranging in size from 10 to 24 kb with a totalmolecular size of approximately 143 kb (Clouthier etal 2002) Comparison of the ISAv proteins withthose of other orthomyxoviruses revealed low amino

acid identity values that were between lt13 and lt25percent (Kibenge et al 2001b Krossoslashy et al 1999Ritchie et al 2002 Snow and Cunningham 2001)These results support the proposal to assign ISAv toa new fifth genus Isavirus within the Orthomyxo-viridae family (Anonymous 2001) The concept ofISAv strain variation has gained acceptance amongresearchers since Blake et al (1999) first reportedgenomic sequence data showing significantdifferences between Canadian and Norwegianisolates Our laboratory has studied isolates fromdifferent geographic regions and provided the firstdirect proof that such isolates also show phenotypic

Figure 2mdashComponents of the influenza virion The numbers inbrackets refer to RNA genomic segments encoding thecorresponding viral proteins in influenza A virus

Envelope

Components of influenza virion

Ribonucleoprotein complexNP (5) Nucleoprotein encapsidation of viral genomic RNA viral replicationPB1 PB2 amp PA (2 1 amp 3) protein subunits of RNA polymeraseNSndash(8) Nonstructural protein inhibits interferon response apoptosis inhibitspolyadenylation and nuclear export of cellular mRNA

Hemagglutinin (HA) (4) Receptor bindingmembrane fusion tissue tropism viral spreadpathogenecity immune response 15 subtypes

M2 (7) Ion channel activity

M1 (7) matrix protein highly conservedinteracts with other proteins

NS2 (8) Nuclear export protein

Neurominidase (NA) (6) cleaves virus receptorsimmune response apoptosis 9 subtypes

43


European HA subtype

North AmericanHA subtype

Can-14

Can-8

Can-1

Can-10

Can-11

Can-9

Can-15

Can-3

Can-4

Can-6

Can-12

Can-5

Can-7

Can-1

Can-2

Can-13

Nor-5

Nor-6

Nor-1

Nor-4

Nor-3

Scot-2

Scot-1

Nor-2

01

Figure 3mdashPhylogenetic tree showing the relationship betweendifferent ISAv isolates the two HA subtypes one North Americanand one European are evident For a more detailed analysis referto Kibenge et al (2001b)

variation (Kibenge et al 2000b 2001b) It is possiblethat the virus is mutating in the wild fishery and thestrain variation that was detected (Blake et al 1999Devold et al 2001 Kibenge et al 2000b 2001b) maybe a reflection of the wide range of natural marinefish hosts

A previous study of 32 ISAv isolates (Kibenge etal 2001b) identified two HA subtypes one NorthAmerican and one European on the basis ofneutralization with rabbit antisera to whole virus andby sequence analysis of the HA gene (fig 3)

Recently we compared the NA gene sequence (ISAvRNA segment 5) among 30 ISAv isolates within the 2HA subtypes Our results revealed up to four distinctNA subtypes indicating the existence of eightdifferent combinations of HA and NA subtypesamong ISAv isolates (table 2) Six of the HANAsubtypes were identified among North Americanisolates and four were identified among Europeanisolates Each geographic region had onepredominant HANA subtype These data havesignificant implications with regard to ISAv vaccines

44


used in the Bay of Fundy because they show (1) thatvirus-neutralizing antibodies to ISAv are subtypespecific (2) that both HA subtypes are present inNew Brunswick and (3) that there may be moreantigenic variation if the NA variation is consideredThus candidate vaccines that induce broadlyneutralizing antibodies are needed to provideimproved protection against ISAv infection clinicaldisease or both

Immunogenicity of ISAv

Most commercially available viral vaccines for fishcontain inactivated whole virus as antigenDevelopments in protein expression technology andvaccinology have made it possible to combine safetybenefits of subunit (SU) vaccines with theadvantages of live vaccines Such benefits includethe use of lower antigen mass elicitation of long-lasting immune memory and stimulation of mucosalimmune responses as well as effective cell-mediatedimmune responses by using naked DNA vaccinesThese new approaches to disease prevention inaquaculture represent a radical change in the waythat antigens are delivered DNA immunizationinvolves the direct introduction of plasmid DNAencoding an antigenic protein which is thenexpressed within the cells of the host animal Thisexpression leads to the development of surprisinglystrong immune responses involving both humoral andcellular immunity

Successful DNA vaccinations of fish againstinfectious hematopoietic necrosis virus (IHNv) andviral hemorrhagic septicemia virus (VHSv) havealready been reported (Boudinot et al 1998 Corbeil

et al 2000) The nucleotide sequence information ofall the RNA segments of ISAv and the putativeproteins that are encoded have been published(Clouthier et al 2002) These developments haveopened the way to production of SU and DNAvaccines against ISAv Recombinant monomeric HAcandidate vaccines are bound to elicit neutralizingantibodies that are subtype specific Thus to induceantibodies with broad activity either mixtures ofrecombinant proteins or highly conserved domains ofviral proteins may be necessary In mice vaccinesinducing antibodies to the highly conservedextracellular domain of the M2 protein conferredprotection to influenza A virus infection (Neirynck etal 1999) DNA vaccines for influenza virus showedthat coimmunization with influenza HA + NP DNAsenhanced protective immunity (Pertmer et al 2000)Immunization of mice with a conventional inactivatedtrivalent influenza vaccine supplemented by two NAsubunits of different subtypes was efficacious inprotecting against homotypic and heterotypicinfluenza viral challenges (Johansson et al 2002)However vaccination of pigs with a DNA constructexpressing an influenza virus M2ndashnucleoproteinfusion protein exacerbated clinical signs of diseaseafter challenge with influenza A virus (Heinen et al2002)

Recently we have discovered a uniquephenomenon of ISAv infection in the macrophage-likecell lines SHKndash1 and TO that suggests thatantibodies to some ISAv antigens may be harmfulbecause they facilitate internalization of antibody-bound virus and growth in the cells (fig 4) Thisphenomenon was not seen when we used the TOcells with infectious pancreatic necrosis virus (IPNv)

Table 2mdashHemagglutinin (HA) and neuraminidase (NA)subtypes of ISAv from North American and Europeansources

HA and NA Type1 H1N1 H1N2 H1N3 H1N4 H2N1 H2N2 H2N3 H2N4

Number of ISAv isolates 11 2 3 2 2 5 1 4

1 H1 corresponds to the North American HA subtype H2 corresponds to the European HA subtype N1ndashN4 correspond to the NA (RNA segment 5)genotypes

45


Virus

VirusVirus Antibody

Virus receptor

Fc receptor

Cell

Cell

Virus attachment is followed byvirus-receptor-mediatedendocytosis in all permissivecells as described in Eliassenet al (2000)

After the virus is bound byspecific antibody endocytosisin fish macrophages is mediatedby Fc receptors

Figure 4mdashIllustration of proposed mechanisms by which ISAv mayinfect fish macrophages

Table 3ndashVirus neutralization against ISAv andinfectious pancreatic necrosis virus (IPNv)in different fish cell lines

Rabbit RabbitVirus anti-ISAv anti-IPNv

Cell line (strain) serum titer1 serum titer1

CHSEndash214 ISAv (NBISA01) 640 (640) NDIPNv (FVX73) ND 10260

SHKndash1 ISAv (NBISA01) 20 NDIPNv (FVX73) ND ND

TO ISAv (NBISA01) 20 (960) NDIPNv (FVX73) ND 10260

1 Virus neutralizing (VN) antibody titer expressed as the reciprocalof the highest dilution of serum to completely neutralize 100TCID50 of virus The number in brackets is the VN antibody titerin the presence of staphylococcal Protein A ND denotes notdone

and rabbit antiserum to IPNv (table 3) Thus harmfulISAv antigens would need to be identified andexcluded from ISAv vaccines in order not only toavoid exacerbating clinical disease (table 1 outcome4) but also to achieve sterile immunity in fish (table 1outcome 1)

Current vaccine delivery by injection is also notsuitable for dosing large numbers of fish in a verycost-effective manner There is therefore a need todevelop live recombinant vectors for gene delivery tofish organs such as the gills the skin or the lateralline and the gut Such organs are believed to beimportant in the antigen uptake and in the inductionof immunity during vaccination by immersion (Tatneret al 1984) However the potential for environmentalcontamination dictates that efforts focus ondevelopment of recombinant viral vectors with verylimited or no ability to replicate in fish cells

46


One significant advance in the field of ISAvvaccine development would be determination of theimmune correlatesmdashthe type magnitude breadthandor location of immune responsesmdashassociatedwith protection against infection and clinical diseaseAtlantic salmon antibody response to ISAv studied byWestern blotting revealed that the antibodies boundexclusively to the viral nucleoprotein (Falk and Dale1998) Although previous studies reported increasedresistance of Atlantic salmon to ISAv upon reinfectionor after passive immunization with serum from fishthat had recovered from ISA (Falk and Dannevig1995) or following vaccination with inactivated virus(Brown et al 2000 Jones et al 1999b) antibodylevels in such fish were not determined To datethere has been no report on laboratory assessmentsof ISAv-vaccine-induced immune responses

Fish serology could be important for assessingISAv vaccine efficacy Our research has establisheda protocol for detecting fish antibody to ISAv usingELISA with purified whole virus as the coatingantigen and monoclonal antibodies that detect fishimmunoglobulins bound to the antigen on the ELISAplate According to this ELISA system farmed fishpresented two different types of antibody responsesto ISAv Naturally infected Atlantic salmon carryingISAv that was detected by RTndashPCR had a specificantibody response to ISAv suggestive of recentinfection Those fish that were virus negative by RTndashPCR had an elevated nonspecific antibody reactivitypossibly suggestive of chronic infection or resistanceto ISAv (table 4) Sera from experimental fishcollected up to 6 weeks after infection with ISAv donot show the elevated nonspecific antibody reactivity

Public Perception of NewVaccine Developments

The fish vaccine industry is developing live vaccinesDNA vaccines and SU vaccines as alternativeapproaches to vaccinate against fish viral diseaseThe industry is also pushing for greater acceptanceof these methods of vaccine delivery Howeverproblems arise from various regulatory authorities

with all three of these approaches In addition thepublic the regulatory authorities and politicians mustbe educated concerning such deliveries Forexample in several countries a DNA-vaccinated fishis classified as a genetically modified organism(GMO) even though there is no evidence that it isgenetically modified Public perceptions do not favorproduction of a GMO salmon If the public will not eatthe salmon the fish farmers cannot use this vaccineThus there is an absolute need to educate thepublic the regulatory authorities and the politiciansconcerning new vaccine modalities

Conclusion

ISA is now a managed disease in CanadaPractically all fish stocked in the Bay of Fundy arevaccinated with inactivated whole ISAv antigensClinical disease occurs albeit at lower levels than in1999 and virus is present even in apparently healthyfish Thus the immunity engendered by the vaccinesis neither 100 percent protective nor sterile Becauseof the possible multiplicity of the ISAv HANAsubtypes in existence vaccines that induce broadlyneutralizing antibodies are needed to provideimproved protection against infection and clinical

Table 4mdashInfectious salmon anemia virus RTndashPCR andantibody ELISA results in Atlantic salmon(Kibenge et al 2002)

ELISA OD405 onELISA OD405 on SHKndash1 cellular

Fish number1 RTndashPCR2 ISAv antigens antigens

F01 ndash 3035 plusmn 0027 (ndash) 039 plusmn 0028


F03 + 03 plusmn0021 (+) 0025 plusmn 0004

F04 + 214 plusmn 03 (+) 00


EF06 + 031 plusmn 0016 (+) 0061 plusmn 0038

EF07 ndash 0009 plusmn 0001 (ndash) 002 plusmn 0002

1 F01ndashF05 were samples from farmed Atlantic salmon EF06 and EF07were positive and negative control experimental fish respectively

2 RTndashPCR-assayed kidney tissues3 Parentheses denote ELISA negative (ndash) and positive (+) results

47


disease Harmful antigens need to be identified andexcluded from ISAv vaccines to achieve sterileimmunity in fish The public the regulatoryauthorities and politicians need to be educated aboutthe new vaccine developments such as live(recombinant viral vector) vaccines DNA vaccinesand SU vaccines One achievement that wouldsignificantly advance the field of ISAv vaccinedevelopment is the elucidation of immune correlatesthat are associated with protection against infectionand clinical disease A newly developed antibodyELISA shows two different types of antibodyresponses to ISAv by farmed fish a specific antibodyresponse in virus-positive fish and a nonspecificantibody reactivity in virus-negative fish that may beresistant to ISAv

References Cited

Anonymous 2000 ISA hits the Faroes Fish FarmingInternational 27(5) 47

Anonymous 2001 Orthomyxoviridae Study Groupproposal to International Committee on Taxonomy ofViruses 10 September 2001 (httpwwwdanforthcenterorgiltabictvnetasp_MainPageasp)

Anonymous 2002 Salmon virus detected in ClewBay fish farm The Irish Times wwwirelandcomAugust 12 2002

Blake S Bouchard D Keleher W Opitz MNicholson B L 1999 Genomic relationships of theNorth American isolate of infectious salmon anemiavirus (ISAV) to the Norweigian strain of ISAVDiseases of Aquatic Organisms 35 139ndash144


Bouchard D Brockway K Giray C Keleher WMerrill P L 2001 First report of infectious salmonanemia (ISA) in the United States Bulletin of theEuropean Association of Fish Pathologists 21 86ndash88

Boudinot P Blanco M de Kinkelin P BenmansourA 1998 Combined DNA immunization with theglycoprotein gene of viral hemorrhagic septicemiavirus and infectious hematopoietic necrosis virusinduces double-specific protective immunity andnonspecific response in rainbow trout Virology 249297ndash306

Brown L L Sperker S A Clouthier S Thornton JC 2000 Development of a vaccine against infectioussalmon anaemia virus (ISAV) Bulletin of the AquaticAssociation of Canada 100 4ndash7


Clouthier S C Rector T Brown N E AndersonE D 2002 Genomic organization of infectioussalmon anaemia virus Journal of General Virology83 421ndash428

Corbeil S Kurath G LaPatra S E 2000 Fish DNAvaccine against infectious hematopoietic necrosisvirus efficacy of various routes of immunisation Fishamp Shellfish Immunology 10 711ndash723

Devold M Falk K Dale O B Krossoslashy B BieringE Aspehaug V Nilsen F Nylund A 2001 Strainvariation based on the haemagglutinin gene inNorwegian ISA virus isolates collected from 1987 to2001 indications of recombination Diseases ofAquatic Organisms 47 119ndash128

Eliassen T M Froystad M K Dannevig B HJankowska M Brech A Falk K Romoren KGjoslashen T 2000 Initial events in infectious salmonanaemia virus infection evidence for the requirementof a low-pH step Journal of Virology 74 218ndash27

48


Falk K Dale O B 1999 Demonstration andcharacterization of the antibody response to ISA virusin Atlantic salmon In The fourth internationalsymposium on viruses of lower vertebrates 12ndash15May 1998 Weymouth UK [Place of publicationpublisher and pagination unknown]

Falk K Dannevig B H 1995 Demonstration of aprotective immune response in infectious salmonanaemia (ISA)-infected Atlantic salmon Salmo salarDiseases of Aquatic Organisms 21 1ndash5

Heinen P P Rijsewijk F A de Boer-Luijtze E ABianchi ATJ 2002 Vaccination of pigs with a DNAconstruct expressing an influenza virus M2-nucleoprotein fusion protein exacerbates diseaseafter challenge with influenza A virus Journal ofGeneral Virology 83 1851ndash1859


Johansson B E Pokorny B A Tiso V A 2002Supplementation of conventional trivalent influenzavaccine with purified viral N1 and N2 neuraminidasesinduces a balanced immune response withoutantigenic competition Vaccine 20 1670ndash1674

Jones SRM MacKinnon A M Groman D B1999a Virulence and pathogenicity of infectioussalmon anaemia virus isolated from farmed salmon inAtlantic Canada Journal of Aquatic Animal Health 11400ndash405


Kibenge FSB Whyte S K Hammell K LRainnie D Kibenge M T Martin C K A dualinfection of infectious salmon anaemia (ISA) virusand a Toga-like virus in ISA of Atlantic salmon Salmosalar in New Brunswick Canada 2000a Diseases ofAquatic Organisms 42 11ndash15

Kibenge FSB Lyaku J Rainnie D Hammell K L2000b Growth of infectious salmon anaemia virus inCHSE-214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81 143ndash150

Kibenge FSB Gaacuterate O N Johnson GArriagada R Kibenge MJT Wadowska D 2001aIsolation and identification of infectious salmonanaemia virus (ISAV) from coho salmon in ChileDiseases of Aquatic Organisms 45 9ndash18

Kibenge FSB Kibenge MJT McKenna P KStothard P Marshall R Cusack R R McGeachyS 2001b Antigenic variation among isolates ofinfectious salmon anaemia virus correlates withgenetic variation of the viral haemagglutinin geneJournal of General Virology 82 2869ndash2879


Kim C H Leong JndashA 1999 Fish viruses InGranoff A Webster R G eds Encyclopedia ofvirology 2d ed New York Academic Press 558ndash568

Krossoslashy B Hordvik I Nilsen F Nylund AEndresen C 1999 The putative polymerasesequence of infectious salmon anaemia virussuggests a new genus within the OrthomyxoviridaeJournal of Virology 73 2136ndash2142


McDougall J Kibenge FSB Evensen Oslash 2001Vaccination against infectious salmon anaemia virus(ISAV) In Hiney Maura ed Proceedings of theEuropean Association of Fish Pathologists 10ndash14September 2001 Dublin Galway City IR EuropeanAssociation of Fish Pathologists P-192

49


McGeachy S M Hawkins L 2001 Infectioussalmon anaemia update New Brunswick In OpitzM ed Proceedings 9th annual New England farmedfish health management workshop April 2001Machias ME Orono ME Animal VeterinaryAquaculture Extension Office University of Maine[Pagination unknown]


Murray A G Smith R J Stagg R M 2002Shipping and the spread of infectious salmonanaemia in Scottish aquaculture Emerging InfectiousDiseases 8 1ndash5

Neirynck S Deroo T Saelens X VanlandschootP Jou W M Fiers W 1999 A universal influenza Avaccine based on the extracellular domain of the M2protein Nature Medicine 5 1157ndash1163

Pertmer T M Oran A E Moser J M MadorinC A Robinson H L 2000 DNA vaccines forinfluenza virus differential effects of maternalantibody on immune responses to hemagglutinin andnucleoprotein Journal of Virology 74 7787ndash7793

Raynard R S Murray A G Gregory A 2001Infectious salmon anaemia virus in wild fish fromScotland Diseases of Aquatic Organisms 4693ndash100

Ritchie R J Cook M Melville K Simard NCusack R Griffiths S 2001 Identification ofinfectious salmon anaemia virus in Atlantic salmonfrom Nova Scotia (Canada) evidence for functionalstrain differences Diseases of Aquatic Organisms44 171ndash178

Ritchie R J Bardiot A Melville K Griffiths SCunningham C O Snow M 2002 Identification andcharacterisation of the genomic segment 7 of theinfectious salmon anaemia virus genome VirusResearch 84 161ndash170


Snow M Cunningham C O 2001 Characterisationof the putative nucleoprotein gene of infectioussalmon anaemia virus (ISAV) Virus Research 74111ndash118

Tatner M F Johnson C M Horne M T 1984 Thetissue locatization of Aeromonas salmonicida inrainbow trout Salmo gairdneri Richardson followingthree methods of administration Journal of FishBiology 25 95ndash108

50


51

Epidemiologic Study of Infectious Salmon Anemia in MaineCombining the Subjective With the Objective in Risk Factor Estimation

Epidemiologic Study of Infectious SalmonAnemia in Maine Combining the SubjectiveWith the Objective in Risk Factor Estimation

Lori Gustafson Steve Ellis and Larry Hammell1

1 Lori Gustafson and Steve Ellis are with USDAndashAPHISndashVS in Eastport ME Larry Hammell is with the AtlanticVeterinary College of the University of Prince EdwardIsland in Charlottetown PE

Introduction

The US Department of Agriculturersquos (USDA) Animaland Plant Health Inspection Service (APHIS)currently oversees surveillance for infectious salmonanemia virus (ISAv) in farmed salmon operationsthroughout eastern Maine As a part of this programUSDA has requested the collection and analysis ofsurveillance data for epidemiologic assessmentConsequently we are assimilating and organizingdata to identify key risk factors that may help preventor mitigate future ISA outbreaks However recentlitigation and allegations that the salmon industry iscurrently experiencing have caused great concernover the possible repercus-sions of the furthersharing of proprietary data In this paper we outlinesome of the steps that we are taking to ensureaccurate and complete reporting without causing theindustry further vulnerability

The Problem

Epidemiologic studies of ISA in Norway NewBrunswick Canada and Scotland (Vagsholm et al1994 Jarp and Karlsen 1997 Hammell and Dohoo1999 Stagg et al 2001) target certain managementpractices such as shared equipment multigenera-tional farms farmed fish movements and stockingdensity as potential intervention points for disease

Abstract Infectious salmon anemia (ISA) was firstofficially documented in Maine in February 2001 Thedisease crippled Mainersquos salmon aquaculture industryleading to the mandatory depopulation of 16 million fish inthe winter of 2001ndash02 We provide a preliminary report ofour efforts to track the spread of the virus and identify riskfactors associated with this epidemic We focus on severalpoints of discussion including(1) industryrsquos hesitancy to share production records owingto concerns over public access of the data

(2) our attempts to employ subjective probability estimationtechniques to circumvent these confidentiality issues(3) our plans to follow this subjective process with atargeted analysis of coded empirical data and(4) the additional benefit of opinion-based assessments intheir relative ease at bridging political boundariesThis last point is of great importance to the ISA programbecause Maine and New Brunswick waters are primarilyseparated by management and political rather than spatialdimensions

control Experience gained from the history ofdisease in these countries has helped guide efforts tocontrol ISA in Maine However many questionsregarding disease transmission and risk remainunanswered

ISA reports in Cobscook and Passamaquoddybays do not track a clear pattern but rather movefrom site to site and even pen to pen in seeminglyrandom fashion Many salmon farms in Maine keptdetailed production records throughout the epidemicand anecdotal accounts of the months precedingsalmon depopulation in Maine are packed withextremely interesting and potentially very usefulinformation Epidemiologic evaluation of the ISAoutbreak in Mainersquos Cobscook and Passamaquoddybays could help further define how ISAvdisseminates why certain pens sites and fish aremore vulnerable to infection than others how topredict farms at greatest risk and how to mitigateconsequences of future outbreaks

We had initially hoped to survey farms ineastern Maine on management practices sitelocations environmental characteristics and ISAdisease incidences and impacts However we havefound that despite the inherent value of the datagenerated during the epidemic answers to ourquestions are extremely difficult to obtain Initialinquiries into the willingness of companies andveterinarians to participate in an epidemiologicanalysis of retrospective data on ISA met with variedlevels of enthusiasm Most support the effort but arevery hesitant to offer direct access to empirical data

52


The reasons for this guarded response areclear Authors Gustafson and Ellis represent aFederal agency and therefore this work is technicallyopen to public scrutiny in the United States throughthe Freedom of Information Act Clearly much of theinformation that we are requesting is proprietaryThis is of particular concern to the industry in light ofthe recent litigations against salmon companies inMaine one in which salmon farms were held inviolation of the Clean Water Act followed soonthereafter by another in which a farm was found guiltyof delaying ISA reports to the Department of MarineResources The acquisition and inappropriate use ofepidemiologic information by interest groups lackingexpertise in aquatic animal health could seriouslydamage legitimate disease-control effortsConsequently we have had to change our focus andapproach though not necessarily our goals

Potential Solutions

We see two adjoining avenues for obtaining sensitiveinformation without threatening company security ortrust (1) develop a subjective database in theabsence of empirical data and (2) code a limited setof empirical data to ensure confidentiality

Opinion-Based DataWe are employing subjective probability

estimation techniques to substitute highly sensitiveempirical data with opinion-based informationSubjective databases have been used successfully inhealth care systems to guide resource allocation ormedical intervention when empirical data are eitherlimited or inaccessible (Von Winterfeldt and Edwards1986 Gustafson et al 1992 Bravington 1996) Anadded benefit to opinion-based surveys is thatgeopolitical boundaries are less of an obstacle Thisis an important corollary for ISA analyses becauseMaine shares much of its water climate industryaffiliations and pathogens with neighboring NewBrunswick

We constructed an interview composed ofabout 45 minutesrsquo worth of questions on perceivedrisks of different management practices and environ-mental conditions on ISA occurrence We arequerying a panel of experts that includes veteri-narians site and general managers regulators andresearchers with direct personal experience in the2001ndash02 outbreaks We ask that they base answersto our questions solely on their experience ratherthan on information gleaned from the literature orscientific meetings Questions are asked in a mannerthat provides numerical estimates for the relativeimportance of perceived risks (Von Winterfeldt andEdwards 1986 Gustafson et al 1992)

Our numerical estimates are based onpercentages of a set of sites which in the expertrsquosexperience are known to have particularcharacteristics We ask them to think about 10 ISA-positive sites and 10 ISA-negative sites from acertain region Of these sites we ask them Howmany would you classify as strong current sites (ormoist feed sites or multigeneration sites etc) Theanswers form a likelihood ratio expressing a measureor degree of perceived relative risk For example theanswers that two ISA-positive sites and four ISA-negative sites use moist feed would form a likelihoodratio of 24 in favor of the apparent protective effect ofmoist feed

An iterative interview process (Gustafson et al1992) is followed wherein experts are asked to reviewand defend any unusual answers relative to theaveraged scores of the other participants Note-worthy position statements are circulated to allparticipants who are given a chance to respond oredit their original answers Final answers then arebased on personal reflection followed by grouprevision and confirmation Group statistics arereported as medians and quartiles of the revisedlikelihood ratios and represent the relative magnitudeand variability of perceived risks In this way expertscan discuss their experiences in generalities ratherthan specifics and still address pertinent questionsabout farm-to-farm transmission

53


From this process we will gain a scaled-backset of risk factors deemed important by the majorityof those surveyed The results of this subjectiveprocess incorporating Canadian and US expertisecan provide practical guidance for future diseasecontrol and international conciliation efforts We alsohope that joint agreement on the probable import-ance of a few select factors in the dissemination ofthe 2001ndash02 outbreak will encourage all involved topursue empirical confirmation of the findings

Coded Empirical Data

We hope to confirm some of the findings fromour opinion-based research with an evaluation of aselect portion of empirical data focusing on justthose factors specified as high priority by the expertpanel Although disclosure of site locations wouldreveal company identity other data can be coded bya third party for site names and sequenced by datesof clinical infection Risk factors that do not requirespatial analysis can be evaluated in this fashion Bylimiting the volume of data required to a short list ofselect factors and by blinding USDA to the mostsensitive information empirical data collection effortswill not be quite as intrusive

Conclusion

By using a combination of subjective and codedempirical data we hope to access industry data thatwould otherwise be considered too sensitive to shareInformation obtained through these two techniquesmay help guide management and regulatorydecisions until further and possibly less confidentialempirical databases are established

Acknowledgments

We thank the following individuals for their ideas andsupport in the initial design and organization of this

project Chris Bartlett Austin Dinsmore DavidGustafson Myron Kebus Dan MacPhee CarolMcClure Mark Moore Shirley RoachndashAlbert andDave Starling

References Cited

Bravington M V 1996 An appraisal of Bayesiansynthesis with suggested modifications anddiagnostics Report to the International WhalingCommission 46 531ndash540

Gustafson D H CatsndashBaril W L Alemi F 1992Systems to support health policy analysis Ann ArborMI Health Administration Press 431 p

Hammell K L Dohoo I R 1999 Epidemiologicstudy of risk factors for mortalities in Atlantic salmonin areas of New Brunswick affected withhaemorrhagic kidney syndromendashinfectious salmonanaemia Atlantic Veterinary College University ofPrince Edward Island Charlottetown PE [Submittedto the] Department of Fisheries and Oceans 31March 1999 52 p

Jarp J Karlsen E 1997 Infectious salmon anaemiarisk factors in sea-cultured Atlantic salmon Salmosalar Diseases of Aquatic Organisms 28 79ndash86

Stagg R M Bruno D W Cunningham C ORaynard R S Munro P D Murray A G AllanCET Smail D A McVicar A H Hastings T S2001 Epizootiological investigations into an outbreakof infectious salmon anaemia (ISA) in Scotland FRSMarine Lab Rep 1301 Aberdeen UK FisheriesResearch Services Marine Laboratory 60 p

Vagsholm I Djupvik H O Willumsen F V TveitA M Tangen K 1994 Infectious salmon anaemiaepidemiology in Norway Preventive VeterinaryMedicine 19 277ndash290

Von Winterfeldt D Edwards W 1986 Decisionanalysis and behavioral research Cambridge UKCambridge University Press 604 p

54


55

The Epidemiology of Infectious Salmon Anemia in Scotland

The Epidemiology of Infectious SalmonAnemia in Scotland

Alexander G Murray1

ISA in Scotland EpidemicSpread

ISA is a List I notifiable disease within the EuropeanUnion (EU) and a compulsory eradication policy is inforce In May 1998 this orthomyxoviral disease wasdetected at an Atlantic salmon (Salmo salar) farm inLoch Nevis on the west coast of Scotland (Rodger etal 1998) The disease rapidly spread to marinesalmon farms around Scotland and by August it waspresent at sites along 650 km of the Scottish coastfrom Argyll to the Shetland Islands (Stagg et al2001) The analyzed viral isolates were geneticallyidentical in viral segments 2 and 8 (Cunningham andSnow 2000 and their paper in this volume)Outbreaks were also linked epidemiologically(Scottish Executive Fisheries Research Services2000 Murray et al 2002) as described in this paperThe genetic and epidemiologic evidence means thata virulent pathogen was spread between sites (asopposed to a latent pathogen being triggered byenvironmental factors) The disease is also spreadas an infection in other countries such as Norway(Jarp 1999) This paper discusses the epidemiologybehind the spread of ISA in Scotland and itsimplications for disease control measures

Abstract Infectious salmon anemia (ISA) was confirmedin May 1998 in farmed Atlantic salmon at Loch Nevis on thewest coast of Scotland The disease rapidly spread fromArgyll to Shetland which represents almost the entireregion over which marine salmon farming occurs inScotland However outbreaks were scattered throughoutthis area 36 farm sites were suspected of which 11 wereconfirmed out of a total of 343 active sites Isolates of theISA virus (ISAv) were genetically identical thereforeoutbreaks were linked by spread and were not due to somewidespread environmental factor The pattern of widely butrelatively lightly scattered outbreaks is not that expectedfrom spread through the environment Instead it reflected

patterns of movements within the aquaculture industrymdashinparticular movements of wellboats that serviced theindustry It can be concluded that movements of live fish inwellboats spread ISA over a large area and that wellboatscollecting harvests explained most of the remaining spreadThe relationships between the number of vessel visits forthese purposes and the suspicion or confirmation of ISAwere statistically significant (p lt 0002) There was noevidence of spread through the environment but suchcontagion may have been significant where sites werelocated within a few kilometers of each other In suchinstances analysis of wellboat movements could notresolve the specific route of ISA contagion

1 Alexander Murray is with the Fisheries ResearchServicesrsquo Marine Laboratory Aberdeen UK

Basic Theory of EpidemicDisease

The basic spread of an epidemic disease may bedescribed by dividing the population of hosts (N) intothree classes (Anderson 1981 Reno 1998)susceptible (S) infected (I) and removed (Q whichmay represent resistant or dead hosts) For adisease with turnover that is rapid relative to that ofthe host population (births and deaths not due to thedisease) the dynamics of infection can be simplifiedto fit the following equations as described by Murrayet al 2001a

dSdt = ndashβSI (1)

dIdt = βSI ndash mI (2)

dQdt = mI (3)

Parameters are the rate of transmission (β) andof removal (m) of infection (cure or death of host)From equation 2 the disease will have a rate ofspread (new cases per case removed) equivalent to

R0 = βSm (4)

If R0 gt 1 removal of cases of infection is lessthan the rate at which new cases occur and anepidemic will progress (Anderson 1981) Thus R0depends on the parameters β or m and the variable S(Reno 1998) This analysis illustrates the basicfactors behind disease transmission and it is notintended as an applied model of ISArsquos spread

The value of m can be increased by rapidreporting and culling of infected fish (Reno 1998)Speedy removal of clinically infected fish is one of the

56


most important tools for controlling ISA (ScottishExecutive Fisheries Research Services 2000)

The appropriate units of S I and Q dependupon the mode of infection transmission (Anderson1981 Murray et al 2001b) which may be densitydependent (farms km-2) or independent (proportion offarms so N = 1) As the names imply density-dependent transmission will increase with farmdensity whereas density-independent transmissiondoes not do so

Whether transmission is density dependent orindependent it is theoretically possible to reduce Sby vaccination However current vaccines may beineffective and leave fish capable of carrying thedisease (Kibenge et al this volume) thuswidespread use is not compatible with the current EUpolicy of eradication

Different processes of pathogen exchange leadto different patterns of disease transmission (Follettet al 2002) These processes can be passiveexchange through the environment epidemic spreadwithin wild fish populations that infects farms as aside effect or vectors that actively transport infectionbetween farms

Passive exchange through the environment islimited by tidal excursion around farms (ScottishExecutive Fisheries Research Services 2000) Onlynearby adjacent sites are likely to becomecontaminated and the epidemic can be broken byphysical separation of farms Survival of pathogensin blood or other organic material could greatlyincrease the effectiveness of this form oftransmissionmdashespecially from slaughterhousesadjacent to farms (Nylund et al 1994) The expectedpattern of spread would be between neighboringfarms and would be most rapid at high density offarms or where tidal excursion is large

A second form of disease spread is among wildfish as a classic epidemic Such epidemics canspread as waves of constant velocity as hasoccurred with epidemics among North Sea seals(Swinton et al 1998) or Australian pilchards (Murrayet al 2001a) Farms could be infected as the

epidemic reaches wild populations in the area Theexpected pattern of spread would be as a waveindependent of farm density

Third natural or anthropogenic vectors mayspread disease between farm sites Natural vectorsinclude birds or fish that travel between farms Seatrout (Salmo trutta) which can act as virus-producingcarriers (Nylund and Jakobsen 1995) are more likelyto be effective vectors than fish such as saithe(Pollachius virens) which do not harbor ISA (Snow etal 2002) Animals may cover substantial distancesScottish sea trout were found carrying virulent formsof the ISAv some 84 km from the nearest infectedfarm (Raynard et al 2001) There are also manyanthropogenic exchanges between farmsmdashparticularly via shippingmdashthat may occur overhundreds of kilometers or more Thus even thoughmost movements (particularly of fish or birds) arelikely to be local infection may occur at sites distantfrom previous outbreaks Vector-transmitteddiseases may be only weakly dependent on farmdensity that is where fish farms are far apart servicevessels make long voyages (natural vectors mayshow more density dependence) Diseases likemalaria that are transmitted by insect vectors displaysuch weak density dependence (Anderson 1981)

Viral transmission depends on exchangebetween different farms and therefore varies betweenany pair of farms If a core subset of farms interactseffectively enough such that R0ij gt 1 an epidemic canoccur even if the overall R0 lt 1 (Jacquez et al 1995)As will be seen a slaughterhouse appears to haveplayed such a key role in the ISA epidemic owing tothe large number of its interactions with farms

Observed Large-Scale Spread

The observed spread of ISA in Scotland was stronglydiscontinuous (fig 1) Outbreaks were rapidlyreported over a relatively wide area and by the endof the summer in 1998 ISA had spread about500 km north to Shetland and 150 km south tomid-Argyll This area delineates almost the entireScottish salmon-farming region Only later was ISA

57


suspected at intermediate locations such as Orkneyand many areas close to outbreaks did not becomeinfected (Stagg et al 2001) Out of 343 active marinesalmon-farming sites (Stagg and Allen 2001)36 were suspected of having ISA but only 11 wereconfirmed (Stagg et al 2001) This pattern of arelatively small number of outbreaks scatteredthroughout the salmon-farming areas implies thatthere was a highly effective mechanism for long-distance dispersal that was not much more effectiveover short distances ie vectors

This observed pattern of spread indicates thatvectors transported infection over large distancesThe very large distances separating ISA outbreaksstrongly suggested anthropogenic exchanges criticaldata were available on movements of wellboatsservicing the salmon farms Records were availablefor the 12 months preceding the epidemic for the1558 visits to fish farms made by the 4 wellboats thatserviced the Loch Creran processing plantIndividual farms are not always identifiable fromvessel records which were generally grouped by

Figure 1mdashSpread of control zones around sites suspected orconfirmed with infectious salmon anemia (ISA) in Scotland duringthe epidemic of 1998ndash99 Figures show the extent of the initialoutbreak spread by the end of summer 1998 and the final extentof the area covered by ISA controls Locations named in the textare indicated by arrows in the figure where ISA first reached thearea

area Areas not visited by these wellboats were notincluded in the analysis The processing plant atLoch Creran is also excluded because it wasqualitatively different from other sites (Murray et al2002)

If ISA was confirmed in at least one site withinan area it was given a disease status of 2 If therewas suspected but no confirmed ISA in an area itwas assigned a disease status of 1 If there was nosuspected or confirmed ISA within the area it wasclassified as 0 Definitions of confirmed and suspectsites are discussed elsewhere (Stagg et al 2001Murray et al 2002)

Outbreaks of ISA showed a very clear link tothe number of visits by wellboats (p = 00015 bylinear regression) but some areas developed ISAafter relatively few wellboat visits (fig 2) whichproduced some scatter in the relationship (r 2 = 023)Particularly outstanding are Skye and Shetlandhowever these areas received shipments of live fishdirectly or indirectly from the original site of infectionat Loch Nevis (Stagg et al 2001) This is a highly

May 1998 August 1998 November 1999

LochNevis

LochBroom

Orkney

LochCreran

Skye

Argyll

Shetland

58


Skye

2Confirmed

Suspect

Clear

1

01 100010010

Number of visits

Infe

ctio

n st

atus

Shetland Loch Creran

Loch Broom

effective means of spreading diseases (Stewart1990) Loch Broom is also unusual because infectionwas only suspected in June 1999 thus it was not partof the initial spread Exclusion of these areas fromthe analysis produced a greatly improved fit(p = 0000004 r2 = 066)

Because different routes of transmission mayhave different efficiencies the vessel visits have beensubdivided into three categories movements of livefish harvest collection and other visits (eg goodsdelivery) Multiple regression has been carried out(using the data analysis package in Microsoft Excel)for infection status against these three categories ofvessel visits and distance from the processing plantThe regression was performed on three different datasets (table 1) Data sets were defined as followsI All data except Loch Creran II Exclusion of otherareas known to have directly received infected fish(Skye and Shetland) and Loch Broom III Only sites

Figure 2mdashISA disease status of areas versus the number ofwellboat visits Infection status is 0 if there was no infection1 if infection was suspected and 2 if infection was confirmedThe processing plant at Loch Creran is shown as an X Sky andShetland (large hollow squares) were infected by fish transferredfrom Loch Nevis Loch Broom is shown as a large filled squareRegression line shown (small squares only) is 0012 x visitsr 2 = 066 and p = 0000004

within 50 km of Loch Creran which excludes SkyeShetland and Loch Broom have been included

Large-scale spread of infection is accounted forby wellboat visits delivering live fish or thoseconcerned with harvest collection All the analysesshowed that there was no relationship betweengeneral-purpose visits or distance from theprocessing plant with infection status Analysis of alldata (I) excluding only Loch Creran gives a goodexplanation of the infection status (r 2 = 043) withsignificant specific relationship to movements of livefish and harvest collection When the areas that hadbecome contaminated via shipments of infected fishfrom Loch Nevis and Loch Broom (II) are excludedthe remaining infections are unrelated to live fishtransport The relationship with harvest collectionhowever is now very significant (p = 0001) andexplanation of the disease status of remaining areasis very good (r 2 = 069) A similar pattern occurs for

59


areas that were within 50 km of the processing plant(III) but is far less significant (p = 005) because therewere relatively few such data cases to analyze

Local Spread

Because analysis of wellboat movements does notpinpoint individual farms it was not useful foranalysis of transmission at distances of a fewkilometers or less The presence of infection atfarms or especially at slaughterhouses within 5 kmwas shown to be a significant risk factor for contagionof ISA into other sites (Jarp and Karlsen 1997)particularly if slaughterhouses were inadequatelydisinfected Local spread of virus with blood whenslaughter occurs at the farm may be important(Munro et al 2003) Sea lice may also transmit virusover short distances (Jarp and Karlsen 1997)Additionally local scavenging fish or bird populationsmay act as vectors that are effective only at scales ofa few kilometers or less

Eradication of ISA from the United Kingdominvolved culling of entire farm sites whereas incountries in which ISA has not been eradicated onlycages housing infected fish were culled The linkbetween cagendashonly culling and unsuccessfuleradication efforts may indicate that ISA can beeffectively transmitted between cages There wasalso probable transmission between sites at alocation such as those in Loch Nevis (Stagg et al

Table 1mdashMultivariate analysis results1

r 2 Harvest Fish movement General Distance

I 043 003 009 065 059

II 069 0001 043 028 029

III 062 005 041 082 044

I Excluding Loch Creran onlyII Excluding Loch Creran Loch Broom Skye and ShetlandIII Within 50 km of Loch Creran (excluding Loch Creran)

1Values of r2 and p of significance of relationships of infectonstatus obtained for different visit types and distance fromprocessing plant (statistically significant p values are in bold)

2001) Intersite spread of disease gave evidence ofeffective transmission at scales of tens to thousandsof meters

Discussion

The spread of ISA in Scotland provides evidence forthe anthropogenic spread of ISAv A clearstatistically significant picture of the emergence ofISA is available shipments of live fish from LochNevis spread ISA around the country and thenmovements from the harvest processing plant atLoch Creran spread the disease along the westHighland coast Similar processes probably wereinvolved in Shetland where ISA spread across thearchipelago At a local scale ISA may have spreadthrough the water particularly in the presence ofblood and perhaps parasites

The spread of ISA in Scotland has strongparallels to the spread of foot-and-mouth disease(FMD) in the United Kingdom in 2001 The viruscausing that disease was also spread over largedistances by the movement of live animals andspread between neighboring farms by a variety ofpathways (Follett et al 2002)

An important tool that minimizes ISAemergence is the control of fish transfers betweenseawater sites and this is accepted in the industrycode of practice for control of ISA Control of localspread is also achieved by separating farms intomanagement areas which are based on theparameters of tidal excursion distances (ScottishExecutive Fisheries Research Services 2000)

Evidence suggests that ISA was spread byharvesting operations Clearly harvesting is anessential part of salmon farming however harvestingpractices should be designed to reduce the risks(Munro et al 2003) Local slaughter at the farm islikely to present risks to neighboring farms whereasslaughter at a central slaughterhouse may riskwidespread transmission If fish are delivered directlyinto the processing plant for immediate slaughter andeffluent is thoroughly disinfected the risk oftransmission can be minimized

60


The virus causing ISA in salmon can be carriedby sea trout (Nylund and Jakobsen 1995) and thevirulent virus has been isolated from wild Scottishsea trout far removed from any farms (Raynard et al2001) Should virulent ISAv become established inwild populations eradication may becomeimpossible

Although ISAv has been found to bewidespread but at low prevalence in Scottish wildsalmonids in most cases it does not appear to causedisease or be associated with ISA on farms (Raynardet al 2001) Genetic evidence supports independentorigins for ISA in Canada and Norway (Blake et al1999) These facts suggest a reservoir of virus fromwhich virulent strains occasionally emerge and iftrue ISAv may resemble the closely related influenzavirus (Webster 1998 Cunningham and Snow thisvolume) Increased time or larger numbers of wild-type viruses will increase the opportunity for virulentvirus to evolve Stressed fish maintained at the samesite for generations without fallowing will provide anenvironment where disease emergence is likely TheLoch Nevis site at which ISA emerged had a chronic(non-ISA) disease history and overlappinggenerations and thus suitable conditions for ISAemergence existed (Stagg et al 2001)

Scottish ISA may have originatedindependently or it may have been imported fromNorway (Cunningham and Snow 2000) Although notidentical to any known strain Scottish ISAv was verysimilar to Norwegian ISAv isolates (Cunningham andSnow this volume) suggesting importation Scotlanddoes not allow imports of live salmonids or their eggsfrom Norway but nevertheless there are strong linksbetween the two countriesrsquo industries Much of theScottish industry is Norwegian owned and several ofthe wellboats used were built in Norway EvisceratedNorwegian and Faroe Island salmon are processed inScotland and on one occasion this involved fish froman ISA-suspect Faroese farm (Pauline Munropersonal communication) A formal risk assessmentis being conducted by Fisheries Research Servicesinspectors enabling identified risky imports to becontrolled under World Trade Organization rules

(Follett et al 2002) The risk of ISArsquos returning toScotland by either importation or viral mutationremains serious

A range of environmental and managementfactors will result in differences in ISA epidemiologybetween countries It appears that ISA wassuccessfully eradicated through a policy of culling inScotland In Norway however ISA is wellestablished and eradication has not been achievedIn Norway virulent ISAv may be established in a wildreservoir In salmon-farming areas of Atlantic NorthAmerica strong tidal currents and short distancesbetween farms may make local transmissionmechanisms more important for ISArsquos spread InPacific North America a range of different salmonidspecies may result in a complex epidemiology shouldan outbreak occur there

References Cited

Anderson R M 1981 Population ecology ofinfectious disease agents In May R M edTheoretical ecology principles and applicationsOxford Blackwell 318ndash355


Cunningham C O Snow M 2000 Genetic analysisof infectious salmon anaemia virus (ISAV) fromScotland Diseases of Aquatic Organisms 41(1) 1ndash8

Follett B Allen P Bateson P Black D Brown FEddy R Leather S Levin S Linklater KLongfield J McConnell I McLean A McMichaelA Mumford J Weiss R Westergaard J 2002Infectious disease in livestock scientific questionsrelating to the transmission prevention and control ofepidemic outbreaks of infectious disease in livestockin Great Britain London The Royal Society 160 p

61


Jacquez J Simon C Koopman J 1995 Coregroups and the R0s for subgroups in heterogeneousSIS and SI models In Mollison D ed Epidemicmodels their structure and relationship to dataCambridge UK Cambridge University Press279ndash301

Jarp J 1999 Epidemiological aspects of viraldiseases in the Norwegian farmed Atlantic salmon(Salmo salar L) Bulletin of the European Associationof Fish Pathologists 19(6) 240ndash244

Jarp J Karlsen E 1997 Infectious salmon anaemia(ISA) risk factors in sea-cultured Atlantic salmonSalmo salar Diseases of Aquatic Organisms 28(2)79ndash86

Munro P D Murray A G Fraser D I Peeler E J2003 An evaluation of the risks of infectious salmonanaemia transmission associated with differentsalmon harvesting methods in Scotland Ocean andCoastal Management 46(1ndash2) 157ndash174

Murray A G OrsquoCallaghan M Jones B 2001aSimple models of massive epidemics of herpesvirusin Australian (and New Zealand) pilchardsEnvironment International 27(2-3) 243ndash248

Murray A G OrsquoCallaghan M Jones B 2001b Amodel of transmission of a viral epidemic amongschools within a shoal of pilchards EcologicalModelling 144(2ndash3) 245ndash259

Murray A G Smith R J Stagg R M 2002Shipping and the spread of infectious salmon anemiain Scottish aquaculture Emerging InfectiousDiseases 8(1) 1ndash5

Nylund A Jakobsen P 1995 Sea trout as a carrierfor infectious salmon anaemia Journal of FishBiology 47(1) 174ndash176

Nylund A Hovland T Hodneland K Nilsen FLoslashvik P 1994 Mechanisms for transmission ofinfectious salmon anaemia (ISA) Diseases of AquaticOrganisms 19(2) 95ndash100

Raynard R S Murray A G Gregory A 2001Infectious salmon anaemia in wild fish from ScotlandDiseases of Aquatic Organisms 46(2) 93ndash100

Reno P W 1998 Factors involved in thedissemination of disease in fish populations Journalof Aquatic Animal Health 10(2) 160ndash171

Rodger H D Turnbull T Muir F Millar SRichards RH 1998 Infectious salmon anaemia(ISA) in the United Kingdom Bulletin of the EuropeanAssociation of Fish Pathologists 18(4) 115ndash116

Scottish Executive Fisheries Research Services2000 Final report of the joint governmentindustryworking group on infectious salmon anaemia (ISA)Aberdeen UK Scottish Executive FisheriesResearch Service 136 p

Snow M Raynard R Bruno D W van NieuwstadtA P Olesen N J Loslashvold T Wallace C 2002Investigation into the susceptibility of saithePollachius virens to infectious salmon anaemia virus(ISAV) and their potential role as a vector for viraltransmission Diseases of Aquatic Organisms 50(1)13ndash18

Stagg R M Allen CET 2001 Scottish fish farmsannual production survey 2000 Aberdeen UKScottish Executive Fisheries Research Services32 p

Stagg R M Bruno D W Cunnigham C ORaynard R S Munro P D Murray A G AllanCET Smail D A McVicar A H Hatings T S2001 Epizootiological investigations into an outbreakof infectious salmon anaemia (ISA) in ScotlandMarine Lab Rep 1301 Aberdeen UK FisheriesResearch Services 60 p

62


Stewart J E 1990 Introductions as factors indiseases of fish and aquatic invertebrates CanadianJournal of Fisheries and Aquatic Sciences 48(supplement) 110ndash117

Swinton J Harwood J Grenfell B T Gilligan C A1998 Persistence thresholds for phocine distempervirus in harbour seal Phoca vitulina metapopulationsJournal of Animal Ecology 67(1) 54ndash68

Webster R G 1998 Influenza an emergingmicrobial pathogen In Krause R M ed Emerginginfections San Diego CA Academic Press 275ndash300

Directory of PersonalCommunications

Pauline MunroChief InspectorFisheries Research Services Marine LaboratoryAberdeen UK

63

Improved Diagnosis of Infectious Salmon Anemia Virus by Use of a New Cell LineDerived From Atlantic Salmon Kidney Tissue

Improved Diagnosis of Infectious SalmonAnemia Virus by Use of a New Cell LineDerived From Atlantic Salmon Kidney Tissue

Jill B Rolland Deborah A Bouchard and James R Winton1

1James Winton (the author to whom correspondenceshould be addressed) is with the US Geological Survey inSeattle WA Jill Rolland worked with him at the time thispaper was prepared and is now with USDAndashAPHISrsquoVeterinary Services in Riverdale MD Deborah Bouchardis with Micro Technologies Inc of Richmond ME

Introduction

Infectious salmon anemia (ISA) is a viral diseaseresponsible for severe economic losses in the Atlanticsalmon aquaculture industry Epizootics have beenreported in Norway since 1984 (Thorud and Djupvik1988) and recently the disease has been diagnosedin Scotland (Rodger et al 1998 Stagg et al 1999Turnbull 1999) Canada (Mullins et al 1998Bouchard et al 1999 Lovely et al 1999) the UnitedStates (Bouchard et al 2001) Chile (Kibenge et al2001) and the Faroe Islands (Anonymous 2000)

Prior to 1995 routine in vitro isolation andpropagation of the ISAv was not possible becauseexisting continuous fish cell lines did not appear tosupport replication of the virus This hamperedcharacterization of the virus and consequently thedevelopment of diagnostic methods Dannevig et al(1995) reported that ISAv could be isolated usingSHKndash1 cells a cell line derived from the Atlanticsalmon (Salmo salar L) pronephros The SHKndash1 cellline is now used routinely to culture ISAv in diag-nostic and research laboratories This ability topropagate the virus has proved immensely importantas evidenced by the subsequent development of amonoclonal antibody to ISAv (Falk and Dannevig1995) and RTndashPCR diagnostic techniques (Mjaalandet al 1997) that can aid in the detection of ISAvinfections More recently it has been determined that

Abstract Infectious salmon anemia (ISA) is an importantviral disease of cultured Atlantic salmon (Salmo salar) inNorway Scotland Canada Denmark Chile and morerecently the United States Current diagnostic methodsinclude isolation of the ISA virus (ISAv) in cell culture anddetection of specific transcriptional products by reverse-transcriptasendashpolymerase chain reaction (RTndashPCR) andmicroscopic observation of viral proteins in affected cells bystaining with fluorescently labeled specific antibodies We

compared the SHKndash1 cell line to another cell line alsodeveloped from Atlantic salmon pronephros the Atlanticsalmon kidney (ASK) cell line Although ISAv can becultured in both cell lines we found ASK cells to be moredesirable for routine laboratory use Not only are ASK cellseasier to maintain they also display a distinct viral-inducedcytopathic effect (CPE) that provides the researcher anddiagnostician with a new tool to detect and quantify ISAvlevels in fish tissues

the long-established CHSEndash214 cell line (Lannan etal 1984) from chinook salmon (Oncorhynchustsawytscha) embryo (Bouchard et al 1999 Lovely etal 1999) and the AS cell line (Nicholson and Byrne1973) derived from the visceral organs of Atlanticsalmon (Sommer and Mennen 1997) also supportedthe replication of ISAv However viral production andcytopathic effect (CPE) in these cell lines is toovariable for routine use

Interestingly Falk et al (1998) reported many ofthe same problems when using the SHKndash1 cell lineincluding low yields of virus and poorly defined andslow developing CPE Kibenge et al (2000) alsoreported that not all strains of ISAv could replicate inCHSEndash214 cells and among those which didreplicate ISAv viral titers were lower than thoseproduced in SHKndash1 cells The North American ISAvisolate used in this study is known to cause CPE inthe CHSEndash214 cell line however neither the NorthAmerican nor European strains of ISAv producedCPE in a pilot study using the CHSEndash214 cell line atthe Western Fisheries Research Center

Wergeland and Jakobsen (2001) reported thedevelopment of a new cell line established fromAtlantic salmon head kidney leukocytes designatedthe TO cell line In developing the TO cell line theauthors hoped to provide researchers anddiagnosticians with a robust highly ISAv-sensitiveand stable cell line The cell line was passed morethan 150 times without changes in morphologygrowth characteristics or viral yields Howeveravailability of the TO cell line is currently very limitedbecause of patent considerations

Devold et al (2000) also reported the isolationof an Atlantic salmon head kidney cell line (ASK) that

64


is susceptible to ISAv and following infection with thevirus displays a distinct CPE (cell rounding anddetachment from the substrate) in just 7 to 8 daysBy September 2002 the cell culture had beensuccessfully passed 30 times

The SHKndash1 cell line remains the mostcommonly used cell line for the clinical diagnosis ofISA despite the fact that ISAv production can be verylow and the weak CPE has limited diagnostic valueTherefore it was of interest to find a cell line thatcould easily be adopted into a laboratoryrsquos existingcell-culture routine and that also produced high ISAvtiters as well as a distinct and complete CPE TheASK line was chosen for comparison with the SHKndash1cell line and results suggested that the ASK cell lineis superior for use when ISAv research studies orroutine diagnosis requires viral isolation

Materials and Methods

Cell Lines

Low-passage number SHKndash1 cells wereobtained from the Central Veterinary Laboratory(Oslo Norway) and high-passage number SHKndash1cells were obtained from Micro Technologies Inc(Richmond ME) The SHKndash1 cells were routinelycultured in Leibowitzrsquos Lndash15 medium supplementedwith fetal bovine serum (FBS 10-percent volume tovolume [vv]) 4 milli-molar (mM) glutamine 100 unitspenicillin 100 microg streptomycin 025 microg amphotericinB and 2ndashmercaptoethanol (01 mM) Low-passageASK cells were obtained from the University ofBergen Department of Fisheries and Marine BiologySection for Fish Health and were propagated usingthe same media formulation as for the SHKndash1 cells(Lndash15 with 10-percent FBS) with the exception of2ndashmercaptoethanol which was excluded Both celllines were incubated at 15 degC and were subculturedevery 10ndash16 days at a split ratio of 12

North American and European VirusStrains

The ISAv strain CCBB was isolated by MicroTechnologies Inc from an ISA outbreak in Atlanticsalmon in Back Bay New Brunswick (Canada) TheBremnes strain of ISAv was isolated from anoutbreak in Bremnes (Norway) by the University ofBergen Department of Fisheries and Marine BiologySection for Fish Health

Virus Production

Routine production of both strains of ISAv wasdone using the SHKndash1 cell line Following mediaremoval 75-cm2 flasks of cell monolayers wereinoculated with either 100 microL ISAv CCBB or ISAvBremnes The virus was allowed to adsorb to theSHKndash1 cells for 30 min after which Leibowitzrsquos Lndash15medium supplemented with 5-percent FBS vv 4 mMglutamine 100 units penicillin 100 microg streptomycin025 microg amphotericin B and 2ndashmercaptoethanol(01 mM) was added The SHKndash1 monolayers wereheld at 15 degC until maximum CPE was observedThe pooled culture fluids for each strain were storedin 10- or 100-mL aliquots at ndash80 degC

Virus Titrations

The sensitivity of the two cell lines to ISAv wasdetermined by titering virus supernatants in the ASKcell line using the 50-percent tissue culture infectiousdose (TCID50) viral assay with 8 replicate wells perdilution in 96-well tissue-culture microplates Cellcultures were drained and each well received 100 microLof a log10 viral dilution prepared in Leibowitzrsquos Lndash15medium supplemented as before and containing noFBS (Lndash15 0) Following virus inoculation cultureswere incubated at 15 degC in plastic containerssupplemented with a blood gas mixture untilmaximum CPE was observed (8 to 28 days) Viraltiters were calculated using the standard endpointdilution method described by Reed and Muench(1938)

65


To compare the SHKndash1 and ASK cell lines75-cm2 cell-culture flasks of confluent SHK or ASKcells were inoculated with 100 microl of ISAv eitherCCBB or Bremnes as before Cell cultures weremonitored regularly for the presence of CPE whichincluded a daily photographic record of each cultureAfter 8 to 28 days or when maximum CPE wasobserved the cell-culture supernatant was removedand serial log10 dilutions were prepared in Lndash15(0) todetermine the virus titer Whereas virus titers foreach experimental culture were measured on bothcell lines only results from titrations on the ASK cellline are reported

Results and Discussion

The virus titer of the pooled culture fluids wasdetermined to be 108 TCID50mL for both ISAv stocksFinal endpoint titers of ISAv were comparable inSHKndash1 and ASK cells on the basis of titrations usingthe ASK cell line (table 1) Accurate measurement ofvirus titers from SHKndash1 cells was impossible becauseof a lack of distinct CPE Falk et al (1997 1998)developed an immunofluorescence assay that can beadapted to quantify viral levels within the SHKndash1 cellline in the absence of CPE but we did not believe a

serological assay was appropriate for thiscomparison ISAv induces a rapid detectable CPE inthe ASK cell line (fig 1) consisting of complete lysisand detachment from the substrate within 14 days Incontrast the CPE caused by ISAv in SHKndash1 cells isnot definitive In the first 14 days after ISAvinoculation some SHK cells detach from thesubstrate These possibly represent a subpopulationof the cell line permissive to viral replication althoughthe monolayer of fibroblastlike cells appears to bemaintained (fig 2) Up to 25 days after ISAv

Table 1mdashComparison of virus titers in the NorthAmerican and European strains of ISAv that weregrown in either the SHKndash1 or ASK cell lines

Virus titer

ISAv strain Cell line TCID50

Bremnes1 SHKndash1 1064

CCBB2 1057

Bremnes ASK 106

CCBB 1061

Note Confluent monolayers of either ASK or SHK cells were inoculated witheither the CCBB or Bremnes strain of ISAv After 8 to 28 days or whenmaximum cytopathic effect was observed cell-culture supernatant wasremoved and titered on ASK cells

1European

2North American

Figure 1mdashCytopathic effect (CPE) of ISAv in ASK cells consistingof lysis and detachment of the cells from the substrate Photographtaken at 14 days after infection at 100x magnification (This and allother photographs were taken by coauthor James R Winton)

Figure 2mdashSHKndash1 cells infected with ISAv CPE consists of somecell rounding and detachment from the substrate Photographtaken at 100x magnification 14 days after infection

66


inoculation we observed morphological changes inthe SHK cells such as vacuolization that could beattributed to factors other than viral replication (fig 3)

The ASK cell line resembles epithelial cells andat the Western Fisheries Research Center hasretained that morphology with each subculture(fig 4) The SHKndash1 cell line has been partiallycharacterized and the cells were found to resemblemacrophages (Dannevig et al 1995) Howevermorphologically the cell line appears to be a mixedpopulation consisting primarily of fibroblastlike cellsbut also containing a small population of cellsresembling leukocytes (fig 5) The role of each celltype in ISAv replication is poorly understoodHowever the number of leukocytelike cells in aculture declines with each passage To avoidcomplete loss of this subpopulation fresh cells fromfrozen stocks at a low passage must be placed intoculture at regular intervals

Upon receipt of the SHKndash1 cell line it wasnecessary to acclimatize the cells using aconditioned medium Once cells were transferred togrowth medium Lndash15(10) they exhibited poor growthGrowth was regained only after the cells were placedin a medium containing Biowhittaker Australian FBSAt no time did we feel it was necessary to use aconditioned medium when culturing the ASK cellsAlthough the ASK cells appear to grow faster in amedium containing Biowhittaker Australian FBS thecell line is not dependent on this serum supplementfor growth

Use of the SHKndash1 cell line requires carefulattention to culture conditions For example theSHKndash1 cells appear to be sensitive to cell density assubculturing either before or after the cells reach100-percent confluence can result in loss of culturesIn contrast the ASK cultures appear to be moretolerant to subculture once the cells have passed100-percent confluence although like the SHKndash1ssubculture at low cell densities can also result in lossof a culture

The aim of this study was not to characterizethe ASK cell line but to compare it with the SHKndash1cell line for isolation of ISAv Results suggested that

Figure 3mdashSHKndash1 cells infected with ISAv 25 days after infectionThe cell monolayer is maintained despite infection Photographtaken at 100x magnification

Figure 5mdashA healthy culture of SHKndash1 cells consisting primarily ofcells that morphologically resemble fibroblasts and a smallproportion of cells resembling leukocytes Photograph taken at100x

Figure 4mdashA culture of ASK cells consisting of cells resemblingepithelia Photograph taken at 100x

67


the ASK cell line may be superior to the SHKndash1 cellline in terms of relative ease of use in the laboratoryMore importantly the presence of clear and distinctCPE by ISAv in ASK cells provides the fish healthspecialist with an economical method to quantify virallevels We believe that these characteristics of theASK cell line also make it appropriate for studying theepizootiology and pathogenesis of ISAv

Acknowledgments

We thank Dr Bjoslashrn Krossoslashy for the ASK cell lineDr Birgit Dannevig for the SHKndash1 cell line DebbieBouchard and Micro Technologies for ISAv strainCCBB and low-passage SHKndash1 cells Dr Are Nylundfor ISAv strain Bremnes and Ron Pascho for criticalreview of this manuscript

References Cited



Bouchard D A Brockway K Giray C Keleher WMerrill P L 2001 First report of infectious salmonanemia (ISA) in the United States Bulletin of theEuropean Association of Fish Pathologists 21 86ndash88

Dannevig B H Falk K Press C M 1995Propagation of infectious salmon anaemia (ISA) virusin cell culture Veterinary Research 26 438ndash442

Devold M Krossoslashy B Aspehaug V Nylund A2000 Use of RT-PCR for diagnosis of infectioussalmon anaemia virus (ISAV) in carrier sea troutSalmo trutta after experimental infection Diseases ofAquatic Organisms 40 9ndash18

Falk K Dannevig B H 1995 Demonstration ofinfectious salmon anemia (ISA) viral antigens in cellcultures and tissue sections Veterinary Research 26499ndash504

Falk K Namork E Rimstad E Mjaaland SDannevig B H 1997 Characterization of infectioussalmon anemia virus an Orthomyxo-like virusisolated from Atlantic salmon (Salmo salar L)Journal of Virology 71 9016ndash9023


Kibenge F S B Lyaku J R Rainnie D HammellK L 2000 Growth of infectious salmon anaemiavirus in CHSE-214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81 143ndash150

Kibenge F S B Garate O N Johnson GArriagada R Kibenge M J T Wadowska D 2001Isolation and identification of infectious salmonanaemia virus (ISAV) from coho salmon in ChileDiseases of Aquatic Organisms 45 9ndash18

Lannan C N Winton J R Fryer J L 1984 Fishcell lines establishment and characterization of ninecell lines from salmonids In Vitro 20 671ndash676




68


Nicholson B L Byrne C 1973 An established cellline from the Atlantic salmon (Salmo salar) Journal ofthe Fisheries Research Board of Canada 30913ndash916

Reed L J Muench H 1938 Simple method ofdetermining fifty percent endpoints American Journalof Hygiene 27 494ndash497

Rodger H D Turnbull T Muir F Millar F SRichards R H 1998 Infectious salmon anaemia(ISA) in the United Kingdom Bulletin of the EuropeanAssociation of Fish Pathologists 18 115ndash120


Stagg R Bruno D Cunningham C Hastings TBricknell I 1999 Focus on infectious salmonanaemia epizootiology and pathology StateVeterinary Journal 9 1ndash5

Thorud Kristin E Djupvik H O 1988 Infectiousanaemia in Atlantic salmon (Salmo salar L) Bulletinof the European Association of Fish Pathologists 8109ndash111

Turnbull T 1999 Infectious salmon anaemia in theUnited Kingdom Fish Veterinary Journal 3 64ndash67

Wergeland H I Jakobsen R A 2001 A salmonidcell line (TO) for production of infectious salmonanaemia virus (ISAV) Diseases of AquaticOrgansims 44 183ndash190

69

Evaluation of Infectious Salmon Anemia Diagnostic Tests

Evaluation of Infectious Salmon AnemiaDiagnostic Tests

Carol A McClure K Larry Hammell Ian R DohooHenrik Stryhn and Leighanne J Hawkins1

Introduction

Infectious salmon anemia (ISA) virus (ISAv) hascaused disease in farmed Atlantic salmon in NewBrunswick since 1996 (OrsquoHalloran et al 1999) Thissevere disease which is characterized by lethargyanorexia anemia death and internal organ damage(Byrne et al 1998 Thorud and Djupvik 1998) hasoccurred sporadically throughout the New Brunswickfish farms in the Bay of Fundy In 1998 about 22 ofthe 83 salmon farms were completely depopulatedfor control purposes (OrsquoHalloran et al 1999) Costlycontrol methods used on New Brunswick Atlanticsalmon farms include a surveillance program earlyharvest of fish from test-positive cages andindemnity programs

Current industry control programs require ISAtesting on dead fish at least every 6 weeks for everyfarm Such surveillance results in early slaughter of acage if there have been two positive tests on at leasttwo fish and fish in the cage have clinical signs ofISA There are several commercial diagnostic testsincluding virus isolation (VI) the indirect fluorescent

Abstract Infectious salmon anemia (ISA) is a viraldisease characterized by lethargy anorexia anemiainternal organ damage and death Costly control methodsused on the east coast of Canada include a surveillanceprogram early harvest of fish in test-positive cages andindemnity programs Test methods used for regulatorydecisions include the indirect fluorescent antibody test(IFAT) reverse-transcriptasendashpolymerase chain reaction(RTndashPCR) assay and virology Although the diagnostictests have not been validated their results are used tomake sizable monetary decisions The objective of thisstudy was to evaluate the sensitivity and specificity of ISAdiagnostic tests using data collected by the New BrunswickDepartment of Agriculture Fisheries and AquacultureBecause a ldquogold standardrdquo reference test for ISA is notavailable we used cage status as our distinguishingcriterion A pool of negative fish from farms that had neverhad the disease and a pool of positive fish from cages that

1Drs McClure Hammell Dohoo and Stryhn all work for theDepartment of Health Management Atlantic VeterinaryCollege University of Prince Edward Island inCharlottetown PE Dr Hawkins is with Maritime VeterinaryServices in St George NB

were experiencing an outbreak defined by greater than005 percent mortalities per day were obtained andassumed to be negative and positive respectively Weused results from a total of 1071 (807 negative 264positive) fish for this study On the basis of the testrsquos cutoffvalue the sensitivity and specificity for histology rangedfrom 30 percent to 73 percent and 73 percent to 99percent respectively The IFAT had sensitivities andspecificities in the range of 64 percent to 83 percent and 96percent to 100 percent respectively For the RTndashPCRassay sensitivity and specificity were 93 percent and 98percent respectively In test performance evaluation wefactored in the possible clustering of test results by farmthat might be attributed to site differences in diseaseseverity or environmental factors Slight changes insensitivities and specificities were coupled with widening ofthe estimated confidence intervals for most cases

antibody technique (IFAT) reverse-transcriptasendashpolymerase chain reaction (RTndashPCR) assay andhistology on fish tissues (Bouchard et al 1999Dannevig et al 1995ab Evensen et al 1991 Falket al 1998 Mjaaland et al 1997 Simko et al 2000Speilberg et al 1995) Performance characteristics ofthese tests are unknown and test results from thesame fish are often inconsistent Although the ISAdiagnostic tests have not been evaluated theirresults are used to make sizable monetary decisions

Because performance reliabilities for each ofthe diagnostic tests were unknown many tests wereperformed on tissue from the same fish from 1998 to2000 by the Provincial government as part of theearly surveillance program Those results were madeavailable to us for evaluation of the diagnostic testsThe objective of this study was to determine thesensitivity and specificity of as many ISA diagnostictests as possible


A total of 30255 test results were available from8167 fish Much of the data was unusable becausethe disease status of each fishrsquos cage was availableonly from April 1999 to January 2000 All fish thathad diseases other than ISA were removed from thedata set For the purpose of calculating sensitivity

70


and specificity our gold standard for disease statuswas based on the following criteria ISA-negative fishcame from farms that had no outbreak of ISA duringthe period and ISA-positive fish came from cagesthat were experiencing clinical disease defined bymortalities gt005 percent per day at the time ofsampling

After we reduced the usable data set some ofthe laboratories and tests were further dropped fromthe analysis because the numbers of samples weretoo small for statistical analysis The laboratoriesincluded in the study were the Atlantic VeterinaryCollege Diagnostic Lab and Aquatic DiagnosticServices (AVC) in Charlottetown PE the NewBrunswick Department of Agriculture Fisheries andAquaculture laboratory (DAFA) in Blacks HarbourNB and the Research and Productivity Councillaboratory (RPC) in Fredricton NB

All test results were dichotomous or ordinalHistology was reported on a scale of negativesuspect and positive For the sensitivity andspecificity histology data were analyzed in twodifferent ways first with the suspect casesconsidered positive and second with the suspectcases considered negative The IFAT results werereported as negative 1+ 2+ 3+ or 4+ based onfluorescence intensity The IFAT results wereanalyzed using two different cutoff values first using1+ 2+ 3+ or 4+ as a positive result (IFAT 1) andsecond using 1+ as a negative result and 2+ 3+ and4+ as a positive result (IFAT 2) The RTndashPCR assayand virology test have dichotomous results reportedas positive or negative Given the expense of thevirology test pools of up to five fish were tested asone sample in which all fish in the pool would have apositive result even if only one fish in the pool werepositive The data set was reduced further byidentifying the fish that were tested for virologyindividually (not in a pool) The resulting data wasanalyzed for sensitivity and specificity

Sensitivities specificities and 95-percentconfidence intervals were calculated in two differentways Initially test sensitivity specificity and 95-percent confidence intervals (exact based on thebinomial distribution [Newcombe 1998]) were

calculated from a 2 times 2 table of all fish using the goldstandards described above Secondly potential testvariation between the farms for positive and negativepopulations was taken into account by using arandom effects logistic regression model with thefarm as the random effect Sensitivity was calculatedas ey(1+ey) in which y was equal to the constantfrom the random effects logistic regression model forthe ISA-positive population divided by the square rootof (1 + 0346sigma2) in order to obtain a population-averaged estimate (Zeger et al 1988) where sigmawas the estimated dispersion of farm random effectsSpecificity was calculated using 1 ndash (ey(1+ey)) with yas above for the ISA-negative populationConfidence intervals for sensitivity and specificitywere calculated with the same formulas whensubstituting the constant by the limits of itsconfidence interval The estimated intraclasscorrelation coefficient (ICC) between samples at thesame farm was calculated as sigma2(sigma2 + 329)(Snijders and Bosker 1999) Finally 90-percentprediction intervals giving the range of farmsensitivities and specificities were computed bysimilar formulas involving sigma and the standarderror of the constant coefficient For the virologytests sensitivities specificities and 95-percentconfidence intervals were calculated only from the2 times 2 table using the results from fish testedindividually

Results

The final data set contained 3721 test results from1071 fish (807 negative and 264 positive) Thesefish came from 238 different cages and from 23different farms

Sensitivities and specificities with theirassociated confidence intervals for each testanalyzed without (combined estimate) and with(population estimate) the random effect of the farmare shown in table 1 In general the sensitivity forhistology ranged from 30 percent to 73 percent and73 percent to 99 percent respectively on the basis ofthe cutoff value The IFAT had sensitivities andspecificities in the range of 64 percent to 83 percent

71


and 96 percent to 100 pecent respectively ForRTndashPCR assay sensitivity and specificity were 93percent and 98 percent respectively Whenbetween-farm variation was taken into account theestimates changed very slightly

Discussion

The farmed Atlantic salmon industry in NewBrunswick is currently dealing with a diagnostictesting dilemma The surveillance program testsmany dead fish from all of the farms in NewBrunswick If a cage is falsely diagnosed as negative

for ISA viral loads may increase and potentiallyspread to other cages or to neighboring farms If acage is falsely diagnosed as positive with ISA thefish are harvested early resulting in tons ofnonmarket-size fish and a costly compensationpackage to the farmer Because both of thesescenarios are unacceptable the identification of adiagnostic test with high sensitivity and specificity isimperative

The results of our study found the highestsensitivities and specificities in RTndashPCR testsperformed by the RPC lab The RTndashPCR test resultsare usually returned within a few days Unfortunately

Table 1mdashThe estimated sensitivities (Se) andspecificities (Sp) for four ISA diagnostic tests in theNew Brunswick Atlantic salmon farms

Random effect

Number Combined estimate Population estimate estimated P 90 PredictedTest tested Parameter (CI)1 (CI)1 ICC1 value interval

Histology 674 Se 730 (653ndash797) 730 (655ndash793) 000 1000 668ndash783(positive)2 Sp 725 (682ndash764) 721 (646ndash794) 007 0000 528ndash866

Histology 674 Se 302 (232ndash380) 299 (219ndash393) 034 0000 47ndash753(negative)3 Sp 994 (982ndash999) 994 (981ndash998) 000 1000 984ndash998

IFAT 1 (DAFA4) 871 Se 791 (732ndash842) 794 (693ndash869) 014 0011 545ndash940Sp 955 (936ndash970) 957 (924ndash976) 011 0027 893ndash989


IFAT 1 (RPC5) 473 Se 827 (697ndash918) 827 (700ndash907) 000 1000 723ndash898Sp 983 (966ndash993) 980 (912ndash996) 031 0070 903ndash999


RTndashPCR 948 Se 926 (882ndash957) 932 (862ndash967) 010 0103 828ndash981Sp 981 (968ndash990) 967 (910ndash988) 048 0000 845ndash1000

Virology (AVC6) 21 Se No samples NA NA NA NA NASp 100 (839ndash1000) NA NA NA NA NA

Virology (RPC6) 72 Se 667 (94ndash992) NA NA NA NA NASp 986 (922ndash1000) NA NA NA NA NA

1 CI = confidence interval ICC = intraclass correlation coefficient2 Suspects were considered positive3 Suspects were considered negative4 New Brunswick Department of Agriculture Fisheries and Aquaculture5 Research and Productivity Council6 Atlantic Veterinary College

72


this testrsquos expensive price ($55 Canadian per fish)may limit its practical usefulness in the industry Thequickest and cheapest test by far is the IFATUnfortunately this testrsquos sensitivity is at best83 percent Therefore 17 percent of the truly positivefish appear as false negatives Histology did notperform very well as an ISA diagnostic test but thistest does have two advantages it is inexpensive andthere is the potential of diagnosing a concurrentdisease Performance evaluation of virology wasdifficult because most samples were pooled for thistest The final evaluation was made only on fish thatwere tested individually The small number of fishtested made it impossible to evaluate the randomeffects of the site Although the specificity of thevirology test is excellent the sensitivity was poor forRPCrsquos virology test and was not evaluated for AVCrsquosvirology test due to an insufficient number ofsamples An advantage of virology is that a positiveresult indicates there is live virus in the sampleHowever poor sensitivity high expense and longincubation periods restrict the use of this test(Dannevig et al 1995b)

Although we have estimated the sensitivitiesand specificities of these diagnostic tests a criticalreview of the methods should be discussed Definingdisease status on samples from perfectly healthysites and highly diseased cages introduces bias thatwill cause tests to appear to perform better than theywould if applied to all fish (Brenner and Gefeller1997) Fish that have just been infected and are notshowing any signs of disease may not test positive onthe available tests resulting in a loss of sensitivityFish with other types of disease may cross-react withthe tests causing false positives and a subsequentreduced specificity Because the data were trimmeddown significantly to identify obviously diseased anddisease-free fish test performance will appear betterthan it would have been had the test been applied tothe whole population

Conversely the sensitivity of the virology testmay have been falsely lowered The DAFA labpooled tissue samples from one to five fish Fish in apool usually came from the same cage If there were

five fish in a pool the cage probably had highmortalities and advanced disease If there was onlyone fish in a pool there was probably only that onedead fish in the cage Therefore it is very unlikelythat the fish in that cage had advanced clinical illnessThese fish might have been infected but might nothave had sufficiently abundant live virus to create acytopathic effect on the cell culture easily which isthe endpoint of the virology test

The random effects model was used to accountfor fish from one farm being more alike than fish fromdifferent farms This model takes into account theextra variation between farms In addition the modelprovides prediction intervals for the sensitivity (orspecificity) of the test used on fish from a new farmfrom New Brunswick When extra farm variation ispresent these intervals are wider than theconfidence intervals because they incorporate farm-to-farm variation Possible reasons for extra variationbetween farms include genetics geography age andmanagement (feed handling sea lice hygiene etc)A hypothetical scenario might be a strain of Atlanticsalmon with improved resistance to ISAv that mightnot replicate enough virus to yield a positive result onthe IFAT test This would result in an increase infalse-negative tests for fish from farms with similargenetics A geographic hypothetical example mightbe dead fish that come from more remote farms andare not processed as quickly as dead fish from localfarms As dead fish decompose the integrity of theviral RNA may be jeopardized Fish from these farmsare more likely to have false negatives on the RTndashPCR test as a result of the increased time toprocessing

Estimates for sensitivities and specificities forISA diagnostic tests are helpful in choosing whichtest will most likely return a true result Howevereach test measures something different about thedisease Virus isolation measures live virus RTndashPCR measures viral RNA IFAT measures viralantigen and histology assesses lesions (Bouchard etal 1999 Dannevig et al 1995ab Evensen et al1991 Falk et al 1998 Mjaaland et al 1997 Simko etal 2000 Speilberg et al 1995) If RTndashPCR is

73


positive ISA viral RNA is most likely in the fish butthis does not necessarily indicate that the fish isclinically ill or actively shedding virus Until we arecapable of predicting the future outcome of the fishcage using diagnostic tests test results should beinterpreted cautiously

The method of choosing the gold standards forthis study was not ideal however it does give anestimation of how the tests are performing Theseresults will be used as the basis for future studiesdesigned to better estimate the sensitivities andspecificities These studies will include analyses thatare not based on a gold standard test (Hui andWalter 1980)

References Cited

Bouchard D Keleher W Opitz H M Blake SEdwards K C Nicholson B L 1999 Isolation ofinfectious salmon anemia virus (ISAV) from Atlanticsalmon in New Brunswick Canada Diseases ofAquatic Organisms 35(2) 131ndash137

Brenner H Gefeller O 1997 Variation of sensitivityspecificity likelihood ratios and predictive values withdisease prevalence Statistical Medicine 16(6) 981ndash991

Byrne P J MacPhee D D Ostland V E JohnsonG Ferguson H W 1998 Haemorrhagic kidneysyndrome of Atlantic salmon Salmo salar L Journalof Fish Diseases 21(2) 81ndash91

Dannevig B H Falk K Namork E 1995a Isolationof the causal virus of infectious salmon anaemia(ISA) in a long-term cell line from Atlantic salmonhead kidney Journal of General Virology 76(6)1353ndash1359

Dannevig B H Falk K Press C M 1995bPropagation of infectious salmon anaemia (ISA) virusin cell culture Veterinary Research 26 438ndash442

Evensen O Thorud K E Olsen Y A 1991 Amorphological study of the gross and lightmicroscopic lesions of infectious anaemia in Atlanticsalmon (Salmo salar) Research in VeterinaryScience 51(2) 215ndash222

Falk K Namork E Dannevig B H 1998Characterization and applications of a monoclonalantibody against infectious salmon anaemia virusDiseases of Aquatic Organisms 34(2) 77ndash85

Hui S L Walter S D 1980 Estimating the errorrates of diagnostic tests Biometrics 36 167ndash171

Mjaaland S Rimstad E Falk K Dannevig B H1997 Genomic characterization of the virus causinginfectious salmon anemia in Atlantic salmon (Salmosalar L) an Orthomyxo-like virus in a teleost Journalof Virology 71(10) 7681ndash7686

Newcombe R G 1998 Two-sided confidence for thesingle proportion comparison of seven methodsStatistical Medicine 17 857ndash872

OrsquoHalloran JLF LrsquoAventure J P Groman D BReid A M 1999 Infectious salmon anemia inAtlantic salmon Canadian Veterinary Journal 40(5)351ndash352

Simko E Brown L L MacKinnon A M ByrneP J Ostland V E Ferguson H W 2000Experimental infection of Atlantic salmon Salmosalar L with infectious salmon anaemia virus ahistopathological study Journal of Fish Diseases23(1) 27ndash32

Snijders TAB Bosker R J 1999 Multilevelanalysis London Sage Publications 266 p

Speilberg L Evensen O Dannevig B H 1995 Asequential study of the light and electron microscopicliver lesions of infectious anemia in Atlantic salmon(Salmo salar L) Veterinary Pathology 32(5)466ndash478

Thorud K Djupvik H O 1988 Infectious anaemia inAtlantic salmon (Salmo salar L) Bulletin of theEuropean Association of Fish Pathologists 8(5)109ndash111

Zeger S L Liang K Y Albert P S 1988 Models forlongitudinal data A generalized estimating equationapproach Biometrics 44 1049ndash1060

74


75

Development of a Strain Typing Assay forInfectious Salmon Anemia Virus (ISAv)


Marcia Cook Sherry Vincent Rachael Ritchie and Steve Griffiths1

1All authors are with the Research and Productivity Councilin Fredericton NB

Introduction

Infectious salmon anemia virus (ISAv) was firstidentified in Norway in 1984 (Thorud and Djupvik1988) and later in Canada (Mullins et al 1998 Lovelyet al 1999 Blake et al 1999 Bouchard et al 1999)Scotland (Rodger et al 1998 Rowley et al 1999)Chile (Kibenge et al 2001a) and the United States(Bouchard et al 2001) Initial focus on ISAv infectionin eastern Canada included stringent surveillanceefforts and implementation of acute managementstrategies in an attempt to limit further disseminationof the disease

Because the number of clinical cases wentdown drastically more attention has been recentlydirected toward the characterization and classificationof discrete ISAv isolates The desire to understandstrain variability comes from several directionsmdashanecdotal evidence that some ISAv outbreaks claimmore fish than others and observations in our lab thatlow-level infections exist in nonsalmonid speciessuch as plaice and haddock (unpublished)Furthermore it was established that Norwegianisolates include strains that can be differentiated by11 sequence variants within a highly polymorphicregion (HPR) of the viral segment encoding for thehemagglutinin (HA) (Devold at al 2001) The abilityto rapidly differentiate between viral strains willprovide basic knowledge on virulence epidemiologyand host specificity that can facilitate managementdecisions

The genome of ISAv consists of eight segmentsof negative-stranded ribonucleic acid) (RNA)

Abstract In 1997 infectious salmon anemia virus (ISAv)was first identified within aquaculture cages of the Bay ofFundy Canada The initial focus of surveillance andfallowing of sites with ISA-infected fish has since expandedto include characterization of ISAv isolates within AtlanticCanada This initiative was adopted to identify discretevariances in virulence of strains and to track disease Thusa diagnostic assay was developed for surveillance pro-grams capable of providing quick and reliable results for

ISAv strain identification The assay combined reverse-transcriptasendashpolymerase chain reaction (RTndashPCR) withdenaturing gradient gel electrophoresis (DGGE) to identifyand type ISAv based on nucleotide variability within specificgene segments Following initial standardization withknown isolates DGGE was performed on RTndashPCRproducts to rapidly determine strain identity without needfor additional manipulations such as sequencing orrestriction enzyme digestion

(Mjaaland et al 1997) Segment 2 is believed toencode a polymerase protein (Krossoslashy et al 1999)segments 3 and 4 to encode a nucleoprotein andpolymerase respectively (Ritchie et al 2001a)segment 6 to encode a HA (Krossoslashy et al 2001Rimstad et al 2001) and segments 7 and 8 toencode putative nonstructural and matrix proteins(Ritchie et al 2002 Cunningham and Snow 2000Biering et al 2002) Identification of these segmentsrevealed the existence of nucleotide variation amongisolates especially in segment 6 (Krossoslashy et al 2001Rimstad et al 2001 Kibenge et al 2001b Devold etal 2001 Griffiths et al 2001) Such genetic variationallows for typing of ISAv isolates

The identification of variability at the nucleotidelevel can be achieved by a number of meansincluding restriction fragment length polymorphism(RFLP) analysis and direct sequencing Howeverthese approaches can be time consuming and costlyHere we present a method that identifies variation inISAv isolates utilizing RTndashPCR and denaturinggradient gel electrophoresis (DGGE) technologies(Myers et al 1987) The latter has been used todetect mutations in disease studies (Valero et al1994) separation of alleles for sequencing (Aldridgeet al 1998) and profiling of complex microbialpopulations (Muyzer et al 1993) The DGGEtechnique is rapid and inexpensive and types multiplesamples simultaneously Following the amplificationof targeted nucleotide regions in RTndashPCR DGGEseparates the double-stranded fragments based ontheir different melting temperatures as dictated bynucleotide sequence Fragments are run in a linearascending gradient of chemical denaturant and uponmelting will be retained at a specific location withinthe gel Fragments that differ in sequence therefore

76


may be identified by comparison to known standardsHere we describe the development of an assay thatdetects genetically distinct ISAv isolates endemic tothe Bay of Fundy Canada and provides for thepossibility of detecting other strains that may beencountered in the future

Material and Methods

ISAv Isolates

New Brunswick ISAv isolates used in this studywere isolated from aquaculture Atlantic salmon(Salmo salar) in the Bay of Fundy Canada usingsalmon head kidney (SHK) cells (Ritchie et al2001b) Previous sequencing of the segment 6 HPRfrom Canadian isolates indicated that there were twomain groups of ISAv in New Brunswick (Kibenge etal 2001b Griffiths et al 2001) Isolates NB280 andNB508 were included in this study as representativesfrom each group In addition an isolate from Norway(Glesvaer) Scotland (Loch Nevis) and Nova Scotiawere included as standard reference strains Thefollowing GenBank accession numbers submitted byother researchers make reference to the same NewBrunswick isolates as used in this study the isolatenames here having been shortened AF294870(NB280) AF294874 (NB508) AF294877 (NB877)AF294871 (NB028) AF294876 (NB049) AF294875(NB002) and AF294873 (NB458) Note howeverthat the sequencing data obtained for some of theseisolates in this study does not agree with thatsubmitted to GenBank by other researchers The 30sequences generated in this study can be foundunder GenBank accession numbers AY151789ndashAY151818

RNA Extraction and RTndashPCR

Total RNA was extracted from cytopathic SHKcells using TRIzolreg LS reagent (Invitrogen) accordingto the manufacturerrsquos protocol Pellets weresuspended in diethyl pyrocarbonate (DEPC)-treatedwater (20ndash50 microL) and 2ndash4 microL used as template inRTndashPCR The RTndashPCR reactions were done usingReady-to-Gotrade RTndashPCR beads (AmershamPharmacia) as described by the manufacturerBriefly RNA was reverse transcribed in a totalvolume of 40 microL using 25 microg of random hexamerprimers at 42 degC for 30 min followed by 95 degC for5 min The PCR primers were then added to a finalconcentration of 04 microM the total PCR reactionvolume being 50 microL Primers used in this study weredesigned from the sequences of segments 6 7 and8 Segment 8 primers were based on those found inDevold et al (2000) and were included as a controlbecause they target a conserved region and ensuredthat a strictly maintained protocol existed The sameprotocol has been used for the surveillance programconducted for the New Brunswick Department ofAgriculture Fisheries and Aquaculture (NBDAFA)since 1998 Segments 6 and 7 primers wereselected to flank variable regions after comparingmultiple international ISAv sequences in the NationalCenter for Biotechnology Information databank Forsegments 6 and 7 multiple primer pairs and RTndashPCRconditions (data not shown) were tested for theamplification of all five reference isolates A 40-bpGC clamp (Myers et al 1985) was added to one ofthe primers of each primer set following Winmelttrade(BioRad) analysis to ensure that a very high meltingpoint domain existed in the amplified product Thefollowing tabulation lists the primers selected

77


Final MgCl2 concentrations in PCR were either15 mM (segment 6 and 7 primers) or 2 mM(segment 8 primers) Conditions for PCRamplification were as follows 40 cycles of 94 degC for30 sec 62 degC for 45 sec and 72 degC for 1 min 30 secAfter mixing with loading buffer amplified productswere run on 11-percent acrylamide TBE gels (MiniPROTEANreg II BioRad) at 200 V for 1 h to check forPCR product yield prior to DGGE analysis

Denaturing Gradient GelElectrophoresis

DGGE was performed using the DCodeMutation Detection System (BioRad) utilizing15-mm-thick times 15-cm long 05 times TAE gels ofvarying percentages of ureandashformamide mixture(7 M ureandash40 percent formamide) 6-percentpolyacrylamide (using 40 percent acrylamidendashbissolution 291) and a nondenaturing 4-percentstacking gel During the optimization process abroad gradient of denaturant was initially usedBased on the observed migration of RTndashPCRproducts progressively narrower ranges ofdenaturant were then tested Electrophoresis wasconducted at 80 V 60 degC for varying amounts of time(12ndash17 h)

Sequence Analysis

Isolate identification was confirmed bysequencing PCR products The PCR products were

Segment 6 HAFnew (GC clamp)

5rsquocgcccgccgcgccccgcgcccgtcccgccgcccccgcccgTKGTKAAAGANTTTGACCARACA3

and 1414mod 5rsquoACAGWGCWATCCCAAAACCTG 3

Segment 7 MAF1 (GC clamp)

5rsquocgcccgccgcgccccgcgcccgtcccgccgcccccgcccgCKGAACAAGGWGGAAAAGTGGT3

and MAR1 5rsquoTAGCAAGTTCATCAAGGAAAATG 3

Segment 8 NBFA3 (GC clamp)

5rsquocgcccgccgcgccccgcgcccgtcccgccgcccccgcccgGAGGAATCAGGATGCCAGGACG3

and RA3 5 GAAGTCGATGAACTGCAGCGA 3

purified using the Qiaquickreg PCR purification kit(Qiagen) and eluted in 50 microL of EB buffer Three microLof purified PCR product were then added to 32 pmolof sequencing primer and 8 microL of Big DyetradeTerminator solution (PE Applied Biosystems) mixed11 with half BDtrade sequencing reagent (BioCanScientific) and cycled 25 times at 96 degC for 10 sec50 degC for 5 sec and 60 degC for 4 min The totalreaction volume was 20 microL Primers used forsequencing were the same as those used in PCRFollowing the cycle sequencing reaction the mixturewas applied to Performareg DTR gel filtrationcartridges (Edge BioSystems) The eluted productwas heated to 95 degC and snap-cooled on iceSamples were then electrophoresed on an ABI3100trade Genetic Analyser and further examined usingSequenchertrade (GeneCodes)

Results

The primer sets tested including HAFnew (GCclamp)1414mod for segment 6 and MAF1 (GCclamp)MAR1 for segment 7 successfully amplifiedall five of the ISAv standard isolates as did thesegment 8 primers NBFA3 (GC clamp)RA3 Due tothe presence of insertionsdeletions (indels) in theHPR of segment 6 primers HAFnew (GC clamp)1414mod produced products of variable length inRTndashPCR ranging from 217 to 265 bp Segment 7and 8 primers amplified a 454 bp and 251 bpfragment respectively for each isolate in the panelThe three amplified regions of each isolate were

78


Figure 1amdashSequence analysis of RTndashPCR products generated bysegment 6 primers Nucleotide 1ndash40 GC clamp 1ndash63 HAFnew(GC clamp) primer 257ndash277 1414mod primer The bottom linedenotes the consensus sequence Differences are marked by anasterisk ()

79


Figure 1bmdashSequence analysis of RTndashPCR products generated bysegment 7 primers Nucleotide 1ndash40 GC clamp 1ndash62 MAF1(GC clamp) primer 432ndash454 MAR1 primer The bottom linedenotes the consensus sequence Differences are marked by anasterisk ()

80


Figure 1bmdash(continued)

81


Figure 1cmdashSequence analysis of RTndashPCR products generated bysegment 8 primers Nucleotide 1ndash40 GC clamp 1ndash62 NBFA3(GC clamp) primer 231ndash251 RA3 primer The bottom linedenotes the consensus sequence Differences are marked by anasterisk ()

82


sequenced for comparison with DGGE data(fig 1andashc)

To optimize DGGE conditions appropriate forthe resolution of each RTndashPCR product initialelectrophoresis conditions involved a broad gradientof denaturant (eg 20ndash80 percent or 30ndash80 percentureandashformamide mixture) in a 6-percent acrylamidegel run at 80 V for 17 h at a constant buffertemperature of 60 degC By estimating theconcentration of denaturant that terminated themigration of discrete RTndashPCR products for eachsegment subsequent conditions incorporatedprogressively narrower ranges of denaturant toprovide optimal resolution of fragments For thesegment 6 and 8 RTndashPCR products a 45- to65-percent gradient gel was selected for separationof fragments but a 35- to 55-percent gradient gelwas chosen for segment 7 products

When RTndashPCRDGGE was run on segment 6for the five reference isolates five distinct bands wereseen (fig 2a) Different migration patterns correlatedwith nucleotide sequence differences (fig 1a) TheNB508 and Nova Scotian standards were onlymarginally resolved under these electrophoreticconditions Accordingly we made an effort toincrease the resolution between these two isolates bydecreasing the range of gradient to 10 percent (ie a50- to 60-percent gradient gel) and by varying theelectrophoretic time Still further improvement inresolution was not obtained Nevertheless theavailability of well-resolved RTndashPCR products fromsegments 7 and 8 easily differentiated these twoisolates (see figs 2b and c)

When the RTndashPCR products that had the samemigration pattern under nondenaturing conditionswere run by DGGE for segment 7 four distinctmigration patterns were observed (fig 2b) Thevariation in migration could again be explained bynucleotide sequence variation (fig 1b) Interestinglyisolates NB280 and NB508 which differed at twonucleotide positions (positions 110 and 146 fig 1b)had seemingly identical migration patterns

When the segment 8 RTndashPCR products wererun on DGGE (fig 2c) three distinct migration

patterns were observed NB280 and NB508migrated the shortest distance Scottish and NovaScotian isolates migrated farther and the Norwegianisolate displayed the farthest migration Migrationpatterns correlated with nucleotide variation detectedby sequencing (fig 1c) The NB280 and NB508isolates were identical in sequence as were theScottish and Nova Scotian isolates Interestinglyalthough the Norwegian isolate differed from theScottish and Nova Scotian isolates by only onenucleotide (position 184 fig 1c) the change wasfrom T to C and it had a noticeable effect onmigration

To further test the ability of this assay to typeNew Brunswick ISAv we selected five isolates fromdifferent sites in the Bay of Fundy and ran them in the

Figure 2amdashDGGE analysis of segment 6 RTndashPCR products forISAv reference isolates NB280 (lane 1) NB508 (lane 2) Norway(lane 3 ) Scotland (lane 4) Nova Scotia (lane 5) RTndashPCRproducts were electrophoresed in a 45- to 65-percent denaturantgel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

83


DGGE system against the five reference ISAvisolates for segments 6 7 and 8 RTndashPCR productsSegment 6 RTndashPCRDGGE analysis typed isolatesNB877 and NB028 as reference standard NB280whereas isolates NB049 NB002 and NB458 weretyped as reference standard NB508 (fig 3a)Sequencing analysis of NB877 NB028 NB049NB002 and NB458 (data not shown) confirmed theidentifications made by the RTndashPCRDGGE assayRTndashPCR and DGGE analysis of these isolates basedon segment 7 sequences (fig 3b) typed these asNew Brunswick strains because migration patternswere equivalent to those for reference standardsNB280 and NB508 as did the analysis based onsegment 8 sequences (fig 3c)

Figure 2bmdashDGGE analysis of segment 7 RTndashPCR products forISAv reference isolates NB280 (lane 1) NB508 (lane 2) Norway(lane 3) Scotland (lane 4) Nova Scotia (lane 5) RTndashPCRproducts were electrophoresed in a 35- to 55-percent denaturantgel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

Figure 2cmdashDGGE analysis of segment 8 RTndashPCR products forISAv reference isolates NB280 (lane 1) NB508 (lane 2) Norway(lane 3) Scotland (lane 4) Nova Scotia (lane 5) RTndashPCRproducts were electrophoresed in a 45- to 65-percent denaturantgel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

Discussion

We have described three different RTndashPCRDGGEsystems based on segment 6 7 and 8 sequencesDue to the low genetic variability in segment 8amongst ISAv isolates the assay using segment 8primers rapidly distinguished New Brunswick isolatesfrom Nova Scotian and European isolates The assayusing segment 7 primers rapidly distinguished NewBrunswick isolates from Nova Scotian Norwegianand Scottish isolates The inability to distinguish thetwo New Brunswick reference isolates based onsegment 7 sequences despite the fact that theydiffered at two nucleotide positions (positions 110and 146) is unusual since the substitution of

84


G (NB280) and A (NB508) would suggest that thiswas possible If the problem lies with the proximity ofthese two base pairs to the GC clamp and theirlocation in one of the higher melting domains thenmovement of this clamp to the 3rsquo primer might allowfor the visualization of these differences HoweverWinmelt analysis would suggest that movement ofthe GC clamp to the 3rsquo primer may impede thedetection of other nucleotide differences among ISAvisolates Consequently we are currently investigatingother DGGE conditions to resolve this issue

Using RTndashPCRDGGE analysis of segment 6 itwas possible to differentiate all five ISAv referenceisolates corroborating previous suggestions that thevariability in this segment may be enough to separateISAv isolates (Krossoslashy et al 2001) The contributionof more isolates for use as reference standards thatare significantly different in the HPR region ofsegment 6 would increase the potential for typing andtracking ISAv isolates obtained in future screeningprograms If differences in segment 7 in the New

Figure 3amdashTyping of New Brunswick isolates using segment 6RTndashPCRDGGE analysis Reference isolates NB280 (lane 1)NB508 (lane 2) Norway (lane 3) Scotland (lane 4) Nova Scotia(lane 5) Test isolates NB877 (lane 6) NB028 (lane 7) NB049(lane 8) NB002 (lane 9) NB458 (lane 10) RTndashPCR productswere run in a 45- to 65-percent denaturant gel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

Figure 3bmdashTyping of New Brunswick isolates using segment 7RTndashPCRDGGE analysis Reference isolates NB280 (lane 1)NB508 (lane 2) Norway (lane 3) Scotland (lane 4) Nova Scotia(lane 5) Test isolates NB877 (lane 6) NB028 (lane 7) NB049(lane 8) NB002 (lane 9) NB458 (lane 10) RTndashPCR productswere run in a 35- to 55-percent denaturant gel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

Figure 3cmdashTyping of New Brunswick isolates using segment 8RTndashPCRDGGE analysis Reference isolates NB280 (lane 1)NB508 (lane 2) Norway (lane 3) Scotland (lane 4) Nova Scotia(lane 5) Test isolates NB877 (lane 6) NB028 (lane 7) NB049(lane 8) NB002 (lane 9) NB458 (lane 10) RTndashPCR productswere run in a 45- to 65-percent denaturant gel at 80 V 17 h 60 degC

mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

85


Brunswick isolates can be differentiated thissegment may prove more appropriate to assayrelatedness of strains as the differences here likelyarise gradually by point mutations not like theobserved HPR variation which likely occurs rapidlyby deletion events More informative than RFLPanalysis and more rapid and inexpensive thansequencing the DGGE analysis of RTndashPCR productsis a useful method to continuously monitor the typesof ISAv that currently exist to monitor their spreadand to identify the emergence of new strains

Acknowledgments

The authors thank their colleagues at the NationalVeterinary Institute (Oslo Norway) and the FisheriesResearch Services (FRS) Marine Laboratory(Aberdeen UK) for providing the Norwegian andScottish isolates respectively We also thank EricJohnsen for the maintenance of ISAv This work wassupported by the Aquaculture CollaborativeResearch and Development Program of the CanadaDepartment of Fisheries and Oceans the NewBrunswick Salmon Growersrsquo Association the NewBrunswick Department of Agriculture Fisheries andAquaculture and by Aqua Health Ltd a division ofNovartis

References Cited

Aldridge B M McGuirk S M Clark R J KnappL A Watkins D I Lunn D P 1998 Denaturinggradient gel electrophoresis a rapid method ofdifferentiating BoLAndashDRB3 alleles Animal Genetics29389ndash394

Biering E Falk K Hoel E Jegatheswaran TJoerink M Nylund A Endresen C Krossoslashy B2002 Segment 8 encodes a structural protein ofinfectious salmon anemia virus (ISAV) the co-lineartranscript from segment 7 probably encodes a non-structural or minor structural protein Diseases ofAquatic Organisms 49 117ndash122

Blake S Bouchard D Keleher W Opitz H MNicholson B L 1999 Genomic relationships of theNorth American isolate of infectious salmon anemiavirus (ISAV) to the Norwegian strain of ISAVDiseases of Aquatic Organisms 35(2) 139ndash144



Cunningham C O Snow M 2000 Genetic analysisof infectious salmon anemia virus (ISAV) fromScotland Diseases of Aquatic Organisms 41 1ndash8

Devold M Krossoslashy B Aspehaug V Nylund A(2000) Use of RTndashPCR for diagnosis of infectioussalmon anemia virus (ISAV) in carrier sea troutSalmo trutta after experimental infection Diseases ofAquatic Organisms 40 9ndash18

Devold M Falk K Dale O B Krossoslashy B BieringE Aspehaug V Nilsen F Nylund A 2001 Strainvariation based on the hemagglutinin gene inNorwegian ISA virus isolates collected from 1987 to2001 indications of recombination Diseases ofAquatic Organisms 47 119ndash128

Griffiths S Cook M Mallory B Ritchie R 2001Characterization of ISAV proteins from cell cultureDiseases of Aquatic Organisms 45 19ndash24

Kibenge F S Gaacuterate O N Johnson G ArrigadaR Kibenge MJT Wadowska D 2001a Isolationand identification of infectious salmon anemia virus(ISAV) from coho salmon in Chile Diseases ofAquatic Organisms 45 9ndash18

Kibenge FSB Kibenge MJT McKenna P KStothard P Marshall R Cusack R R McGeachyS 2001b Antigenic variation among isolates ofinfectious salmon anemia virus correlates with thegenetic variation of the viral hemagglutinin geneJournal of General Virology 82 2869ndash2879

86


Krossoslashy B Hordvik I Nilsen F Nylund AEndresen C 1999 The putative polymerasesequence of infectious salmon anemia virus suggestsa new genus within the Orthomyxoviridae Journal ofVirology 73(3) 2136ndash2142

Krossoslashy B Devold M Sanders L KnappskogP M Aspehaug V Falk K Nylund AKoumans S Endresen C Biering E 2001Cloning and identification of the infectious salmonanemia virus hemagglutinin Journal of GeneralVirology 82 1757ndash1765

Lovely J E Dannevig B H Falk K Hutchin LMacKinnon A M Melville K J Rimstad EGriffiths S G 1999 First identification of infectioussalmon anemia virus in North America withhemorrhagic kidney syndrome Diseases of AquaticOrganisms 35(2) 145ndash148

Mjaaland S Rimstad E Falk K Dannevig B H1997 Genomic characterization of the virus causinginfectious salmon anemia (ISA) in Atlantic salmon(Salmo salar L) an Orthomyxo-like virus in a teleostJournal of Virology 71(10) 7681ndash7686

Mullins J E Groman D Wadowska D 1998Infectious salmon anemia in salt water Atlanticsalmon (Salmo salar L) in New Brunswick CanadaBulletin of the European Association of FishPathologists 18 110ndash114

Muyzer G De Waal E C Uitterlinden A G 1993Profiling of complex microbial populations bydenaturing gradient gel electrophoresis analysis ofpolymerase chain reaction-amplified genes coding for16S rRNA Applied and Environmental Microbiology59(3) 695ndash700

Myers R M Fischer S G Lerman L S ManiatisT 1985 Nearly all single base substitutions in DNAfragments joined to a GC-clamp can be detected bydenaturing gradient gel electrophoresis NucleicAcids Research 13(9) 3131ndash3145

Myers R M Maniatis T Lerman L S 1987Detection and localization of single base changes bydenaturing gradient gel electrophoresis Methods inEnzymology 155 501ndash527

Rimstad E Mjaaland S Snow M Mikalsen A BCunningham C O 2001 Characterization of theinfectious salmon anemia virus genomic segmentthat encodes the putative hemagglutinin Journal ofVirology 75 5352ndash5356

Ritchie R Heppell J Cook M Griffiths G 2001aIdentification and characterization of segments 3 and4 of the ISAV genome Virus Genes 22(3) 287ndash295

Ritchie R J Cook M Melville K Simard NCusack R Griffiths S 2001b Identification ofinfectious salmon anemia virus in Atlantic salmonfrom Nova Scotia (Canada) evidence for functionalstrain differences Diseases of Aquatic Organisms 44171ndash178

Ritchie R J Bardiot A Melville K Griffiths SCunningham C O Snow M 2002 Identification andcharacterization of genomic segment 7 of theinfectious salmon anemia virus genome VirusResearch 84 161ndash170

Rodger H D Turnbull T Muir F Millar SRichards R H 1998 Infectious salmon anemia (ISA)in the United Kingdom Bulletin of the EuropeanAssociation of Fish Pathologists 18 115ndash116

Rowley H M Campbell S J Curran W L TurnbullT Bryson DG 1999 Isolation of infectious salmonanemia virus (ISAV) from Scottish farmed Atlanticsalmon (Salmo salar L) Journal of Fish Diseases 22483ndash487

Thorud K Djupvik H O 1988 Infectious salmonanemia in Atlantic salmon (Salmo salar L) Bulletin ofthe European Association of 8 Fish Pathologists109ndash111

Valero M C Velasco E Moreno F Hernandezndashhico C 1994 Characterization of four mutations inthe neurofibromatosis type I gene by denaturinggradient gel electrophoresis (DGGE) HumanMolecular Genetics 3 639ndash641

87

The Genetics of Infectious Salmon Anemia Virus

The Genetics of InfectiousSalmon Anemia Virus

Carey O Cunningham and Michael Snow1

1Both authors are with the Fisheries Research ServicesrsquoMarine Laboratory in Aberdeen UK

Introduction

The discovery of the complete genome sequence ofISAv and analysis of the products encoded by thesegenes has progressed rapidly indicating both theimportance of the disease and the value of geneticinvestigations for aquaculture research anddevelopment Although ISAv is an orthomyxovirus ithas a genome organization that is strikingly differentfrom that found in influenza viruses We will reviewthe work of major groups of researchers to elucidatethe genome of the virus and analyze individualgenes Detection of ISAv and diagnosis of ISA reliesheavily on molecular tests such as the polymerasechain reaction (PCR) We will also discuss theavailable methods and the interpretation of results

Chronology of ISAv GeneDiscovery

The genomic organization of ISAv was described byMjaaland et al (1997) as a genome typical ofmembers of the Orthomyxoviridae family beingsegmented single-stranded and negative senseThe total genome of approximately 145 kb wasshown to be composed of eight segments rangingfrom 1 to 23 kb The sequence of the smallestsegment 8 was obtained and used to design PCRprimers The sequence of segment 8 indicated thepresence of two open reading frames (ORFs) whichlike most ISAv segments did not have significanthomology with any other available nucleotide oramino acid sequence Based on extrapolation from

Abstract The complete genome sequence of infectioussalmon anemia virus (ISAv) has been determined Al-though ISAv is classified as an orthomyxovirus based onphysical characteristics and genome type it has a genomeorganization that is strikingly different from that found ininfluenza viruses Thus it has not been possible to extrapo-late directly from influenza to ISAv and the genes encodingonly one polymerase and the hemagglutinin protein havebeen established to date Genetic variation can be

extremely useful in epizootiological studies but increasedknowledge of variation in genes such as the hemagglutininhas revealed that analysis of the variation can be problem-atic without detailed understanding of the mechanisms thatcreate it Detection of ISAv and diagnosis of ISA has cometo rely heavily upon molecular tests such as the poly-merase chain reaction (PCR) These are currently the mostsensitive methods available and have proven reliable inroutine use

the genome organization of the influenza viruses itwas widely assumed that this segment encoded theNS 1 and 2 proteins (Blake et al 1999 Kibenge et al2000)

Krossoslashy et al (1999) described the sequencefrom segment 2 of ISAv This is the only gene inwhich significant homology with other members ofthe Orthomyxoviridae has been identified to dateThe 23 kb segment encodes one protein with motifsthat are conserved across the Orthomyxoviridae andin RNA-dependent RNA polymerase genes ingeneral Based on phylogenetic analysis using thissequence Krossoslashyrsquos work showed that ISAv isdistantly related to the other Orthomyxoviridae andmore closely related to the influenza viruses than theThogoto viruses Hence Krossoslashy et al proposed thata new genus be created and the name Isavirus hasbeen proposed by the International Committee onTaxonomy of Viruses Index of Viruses version 3(httpwwwncbinlmnihgovICTVdbIctvindexhtm)for this genus in preference to Aquaorthomyxoviruswhich was suggested by Krossoslashy et al (1999)

Krossoslashy et al (1999) noted conserved regionsof nucleotides at the 5rsquo termini of ISAv cDNAsequences Conservation of sequence at the 3rsquotermini of ISAv segments 7 and 8 and complemen-tarity with the conserved regions at the 5rsquo terminiwere described in detail by Sandvik et al (2000)These features have since been identified in all ISAvsegments Conservation and complementarity oftermini sequences are characteristics of orthomyxo-viruses that enable formation of panhandle structuresthat are important for viral RNA stability and theterminal sequences are important to initiatetranscription The conservation of 5rsquo and 3rsquo sequencein ISAv and evidence that ISAv requires capped

88


cellular mRNA for replication support inclusion ofISAv in the Orthomyxoviridae

By 2001 groups working in Canada andScotland had published further information on thegenome of ISAv Snow and Cunningham (2001)described segment 3 and analysis of the nucleotideand predicted amino acid sequences suggested thatthat this segment encoded the nucleoprotein (NP)The predicted molecular weight of the gene product71 kDa correlated well with previous estimates ofstructural proteins (Falk et al 1997 Griffiths et al2001) Ritchie et al (2001b) also described thesequence of segment 3 and included segment 4possibly another polymerase-encoding gene Thework of the Canadian group showed that the putativeNP was immunoreactive a finding that may beparticularly significant for development of vaccinesagainst ISA

Also important for vaccine development andperhaps the most sought-after gene was thehemagglutinin (HA) The sequence of this gene anddirect demonstration that the gene product resulted inHA via transfected cells was reported by Norwegianand Scottish groups in 2001 (Krossoslashy et al 2001aRimstad et al 2001) The product of this gene wasanalyzed by Griffiths et al in the same year The HAgene product reacts with a monoclonal antibodydeveloped some years previously by Falk et al(1998) A striking feature of this gene is the highlypolymorphic region (HPR) predicted to lieimmediately outside the viral envelope

All segments of the ISAv genome werepresented in a single publication by Clouthier et al(2002) A detailed analysis of segment 7 waspresented by Biering et al (2002) and Ritchie et al(2002) Although segment 7 was initially thought toencode the matrix proteins as it was widely believedthat the NS genes were encoded by segment 8Biering et al (2002) presented evidence thatsegment 7 probably encodes the equivalent of theinfluenza nonstructural proteins while segment 8encodes structural (possibly matrix) proteins

Work continues to complete analysis of theprotein encoded by segment 5 Thus the entire

genome of ISAv and the identity of some geneproducts have been discovered An overview of thegenome of ISAv is presented in table 1

Comparison of ISAv and OtherOthomyxoviridae

During the work to determine the ISAv genome ithas been tempting to draw parallels with the otherOrthomyxoviridae The gene order of ISAv hasimportant differences from that of influenza andhindsight has shown that this extrapolation causedconfusion such as the suggestion that segment 7may encode matrix proteins (Ritchie et al 2002) Thetable comparing the genomes of the Orthomyxo-viridae from Mjaaland et al (1997) can now beextended to give a more complete picture of thesedifferences (table 2)

The general composition of the ISAv genomeresembles the other Orthomyxoviridae in that theeight segments appear to encode polymerasenucleoprotein HA esterase matrix andnonstructural proteins Whereas influenza A and Bhave HA and neuraminidase genes on differentsegments influenza C carries HA esterase andfusion genes on a single segment ISAv hasesterase not neuraminidase activity and it has beensuggested that the esterase may be encoded on thesame segment as the HA segment 6 (Knut Falkpersonal communication) The location of the geneconferring fusion activity has not been confirmed

It is easy to envisage that genetic drift duringthe evolution of the Orthomyxoviridae could lead to adifferent gene order giving rise to situations such asinfluenza A having a nucleoprotein segmentsignificantly smaller than the polymerases whereas inISAv the nucleoprotein is larger than one polymeraseHowever the evolution of different arrangement ofsegments with multiple genes indicates that geneticrearrangement has also occurred on a much largerscale Mechanisms of recombination andreassortment have been proposed for the Orthomy-xoviridae (Gibbs et al 2001 McCullers et al 1999Worobey and Holmes 1999) perhaps these have

89


Table 1mdashInfectious salmon anemia virus (ISAv) genesand proteins described as of June 2002 andcomparison of gene order with influenza C

Gene Segment Sequence Segment ORF Protein DescriptionsISAv rsquoflu C (bp) (bp) (kDa)

Polymerase 1 1 AF404304 gt1749 gt1749 Clouthier et al 2002AJ514403 2205 2169 80 Snow et al 2003

Polymerase 2 2 AJ002475 2245 2127 805 Krossoslashy et al 1999AF404346 2185 2127 Clouthier et al 2002

Nucleoprotein 3 5 AJ276858 2069 1851 Snow and Cunningham 2001(NP) 71 Aspehaug et al 2001

AF306549 2042 1851 68 Ritchie et al 2001AF404345 2046 1851 72 Clouthier et al 2002

Polymerase 4 3 AF306548 1805 1737 653 Ritchie et al 2001AF404344 1787 1737 Clouthier et al 2002

Unconfirmed 5 1560 Snow unpublished53 Aspehaug et al 2001

AF404343 1504 1335 47 Clouthier et al 2002

Hemagglutinin 6 4 AF220607 13261321 11671175 Rimstad et al 2001esterase (HE) HEF AJ276859

AF302799ndash 1053ndash1261 gt1053ndash1168 43 Krossoslashy et al 2001AF302803AF309075 Aspehaug et al 2001

3872 Griffiths et al 2001AF404342 1323 1185 42 Clouthier et al 2002

Nonstructural 7 7 AF328627 1159 903480 Ritchie et al 2002proteins (NS) AY044132 1006 903522 354175 Biering et al 2002

AF404341 966 771441 Clouthier et al 2002AF429989ndash90

Matrix proteins Y10404 930 779gt459 Mjaaland et al 1997(M) 8 6 24 Aspehaug et al 2001

Biering et al 2002AF404340 736 705552 Clouthier et al 2002

Note Where gt is shown in open reading frame (ORF) size sequence is incomplete Only polymerase (segment 2) and HA (segment 6) gene products havebeen confirmed in ISAv

been instrumental in bringing about the diversity ofgenome organisation within the genus

Homologous recombination has beensuggested as a possible mechanism for generatingthe large variation seen in the HA gene sequence ofISAv (Devold et al 2001) Homologous recombina-tion and reassortment can only occur when twodifferent forms of virus occur in the same cellBecause more than one type of ISAv has been foundin fish from the same salmon farm (Cunningham etal 2002 Mjaaland et al 2002) it is possible thatreassortment could also occur in ISAv

Genetic Diversity of ISAv

Early comparisons of sequences of ISAv segments 2and 8 showed large differences between isolatesfrom Norway and Scotland and those from Canada(Blake et al 1999 Cunningham and Snow 2000Kibenge et al 2000 Lovely et al 1999 Ritchie et al2001a) The significant sequence variation betweenisolates provides evidence that ISAv has existed for along time It has been postulated that different ISAvisolates from Europe and Canada divergedapproximately 100 years ago (Krossoslashy et al 2001b)

90


Table 2mdashComparison of genome structure within theOrthomyxoviridae showing size (in kb) and productsof genome segments

Genome Influenzasegment A B C Thogoto Dhori ISAv

1 234 PB2 238 PB1 235 P1 238 PB2 ND 220 PB2

2 234 PB1 238 PB2 235 P2 221 PB1 22 PB1 224 PB1

3 223 PA 230 PA 215 P3 192 PA ND 207 NP

4 175 HA 188 HA 207 HEF 157 G 16 G 180 P

5 156 NP 184 NP 180 NP 141 NP 15 NP 156

6 141 NA 140 NA NB 118 M gt096 M 096 M 132 HE

7 103 M 119 M 093 NS mdash ND 116 NS

8 089 NS 109 NS mdash mdash mdash 093 M

Total 136 145 129 105 10 1328

Note Table based on table 2 in Mjaaland et al (1997) with the addition of ISAv data and Thogoto virus data from European Molecular Biology Laboratory(EMBL) nucleotide database June 2002

Gene products

ND No data available

mdash Segment not present

P P1 P2 P3 PB1 PB2 PA Polymerase

HA Hemagglutinin HEF HemagglutininEsteraseFusion

HE Hemagglutininesterase Gene product unconfirmed

NP Nucleoprotein

NA NB Neuraminidase

M Matrix

NS Nonstructural

G Glycoprotein

The presence of another variant of ISAv inCanada that has much greater sequence similarity toEuropean than to the Canadian New Brunswick ISAv(Ritchie et al 2001a and 2002) indicates that thedistribution of ISAv types is not restrictedgeographically The virus may be widely distributedthroughout the Northern Hemisphere and thisdistribution may be influenced by anthropogenicfactors

Once the HA gene of ISAv had been identifiedand analyzed the HPR of the gene became the focusof much interest The function of this HPR is unclearbut it has been used to differentiate ISAv strains(Devold et al 2001 Kibenge et al 2001a) Numerousdistinct strains can be differentiated based on theamino acid sequence within the HPR and these HPR

types may provide useful epizootiological markers(Devold et al 2001) However there are isolates thathave widely different HPR sequences but very similarsequence elsewhere in the gene and converselyisolates with the same or very similar HPRsequences but large differences elsewhere in this orother segments These variabilities complicate theuse of the HPR alone in epizootiology Eighteendifferent HPR types have been identified so far(Cunningham et al 2002 Mjaaland et al 2002)

Functional differences in cytopathogenicity andinfectivity or pathogenicity have been noted betweenISAv isolates (Kibenge et al 2000 Griffiths et al2001 Ritchie et al 2001a) and antigenic variationmay be related to sequence variation in the HPR(Kibenge et al 2001a) Analysis of this HPR also

- - - - - - - - - - - - - - - - - - - - - - - - - -

91


demonstrated the diversity of Canadian NewBrunswick and many European ISAv isolates but theaddition of further sequences from Norway some ofwhich have the same HPR type as the NewBrunswick ISAv (Krossoslashy et al 2001a) has shownthat the isolate may not be unique to CanadaIntuitively it might be expected that an RNA virus thathas been present for approximately 100 years andinfects marine fish (including Atlantic salmon) mightnow exist as a pool of genetic variants that haveevolved different phenotypes Very likely othervariants will be identified in future

It was proposed that the large variety of ISAvHA sequences could be produced by homologousrecombination (Devold et al 2001) Since thatpublication the discovery of a novel HA sequencethat contains all of the sequence motifs previouslyreported in one single sequence (Cunningham et al2002) suggests that differential deletion may be usedinstead of or as well as homologous recombinationto generate antigenically or functionally diverse formsof ISAv HA

The diversity of sequence in ISAv requires carein analysis Where recombination or reassortmentoccur more straightforward methods of phylogeneticanalysis are inappropriate (Schierup and Hein 2000)Thus the HA gene or the HPR will require differentmethods of analysis than other genes

Molecular Epizoootiology

The investigation of possible links between outbreaksof ISA has benefited from molecular study of thevirus Evidence of viral transmission was provided inthe form of HA HPR sequences (Devold et al 2001)In Scotland identical sequences found from ISAv-infected fish in different farms provided evidence thatthe outbreak stemmed from a single point source(Stagg et al 2001)

While the HPR may provide a good marker ofdifferent types or strains of ISAv viral samples withthe same HPR type might not necessarily be directlylinked Differential deletion might produce the sameHPR type from viruses that have significant

divergence in other parts of the genome and even inthe same segment as the HPR Therefore the HPRalone is not necessarily a suitable marker forepizootiological studies Nevertheless the value ofmolecular data for this work is clear

Molecular Diagnosis of ISA

Molecular methods of detecting ISAv have assumeda leading role in the diagnosis and monitoring of thisdisease Mjaaland et al (1997) developed PCRprimers that have proven reliable and sensitive fordiagnostic purposes and are still in use on a largescale Other primers developed from segment 8sequences have been developed (Devold et al 2000Loslashvdal and Enger 2002) with those of Devold et alemployed extensively in Canada and Norway(Griffiths and Melville 2000 Kibenge et al 2001b)While primers for other segments of the genomehave been developed those that anneal to segment8 seem to be most sensitive for routine diagnosticpurposes (see Mjaaland et al 2002) As that segmentappears to encode the matrix proteins (Biering et al2002) this sensitivity is to be expected as transcriptsfrom the equivalent segment are very abundant ininfluenza viruses In addition to sensitivity moleculardetection offers advantages such as the ability toidentify the tissue localization of the virus by in situprobe hybridization (Gregory 2002)

Kidney has been routinely used as the startingmaterial for PCR tests No significant difference hasbeen found in PCR results from kidney heart gillliver and blood so kidney is routinely sampled forease of sampling Nonlethal methods of sampling forPCR (Griffiths and Melville 2000) are extremelyuseful for broodfish or wild stocks

Detection of ISAv by PCR is the most sensitivemethod of detection to date (Devold et al 2000Mjaaland et al 2002 Opitz et al 2000) The viruscan be detected by PCR as early as 5 dayspostinfection (Mikalsen et al 2001) within a variety oforgans (Rimstad et al 1999) The pattern of PCRpositivity is followed approximately 10 days later by aparallel mortality pattern However it is likely that

92


variability in the susceptibilities of fish husbandrypractices and environmental factors will significantlyinfluence the outcome of infection with ISAv andfinding PCR-positive fish does not necessarily predictthat mortality will follow

The application of PCR detection methods forfish disease surveillance has attracted considerabledebate regarding specificity and the likelihood ofdetecting nonviable pathogens The specificity ofISAv detection can be confirmed by probe hybridi-zation to PCR products (McBeath et al 2000) or bysequencing the product Amplification of parts ofmore than one segment of the genome is aparticularly good indication of the presence of intactvirus but may require higher levels of infection thandetection of a small portion of segment 8 alone Lowlevels of virus may be detected after concentration orantibody capture and work continues to developadditional antibodies for this purpose using phagedisplay as well as immunization with recombinantprotein synthetic peptides and DNA vaccine vectors

Conclusions

The investigation of the genetics of ISAv has yieldedan enormous amount of extremely valuableinformation for identification and analysis of ISA andhas also provided data of great scientific interestMolecular methodology is an integral part ofdiagnosis of ISAv and will continue to provide moretools and information to improve detection of thisvirus Epizootiological studies have benefited fromISAv sequence data and development of vaccinesmay proceed more rapidly through application ofrecombinant or DNA vaccines Analysis of theOrthomyxoviridae has been given a new twist withthe discovery and examination of this distant relativeof the influenza viruses and models of the evolutionof the Orthomyxoviridae and ISAv in the futureshould prove fascinating

References Cited

Aspehaug V Falk K Devold M Dale O BKrossoslashy B Biering E Nylund A Endresen C2001 Identification characterisation and functionalaspects of the ISA-virus structural proteins InOrthomyxoviruses 2ndash4 November 2001 TexelNetherlands Rotterdam Netherlands ErasmusUniversity 23



Clouthier S C Rector T Brown N E AndersonE D 2002 Genomic organization of infectioussalmon anaemia virus Journal of General Virology83(2) 421ndash428


Cunningham C O Gregory A Black J SimpsonI Raynard R S 2002 A novel variant of theinfectious salmon anaemia virus (ISAV)haemagglutinin gene suggests mechanisms for virusdiversity Bulletin of the European Association of FishPathologists 22(6) 366ndash374

Devold M Krossoslashy B Aspehaug V Nylund A2000 Use of RT-PCR for diagnosis of infectioussalmon anaemia virus (ISAV) in carrier sea troutSalmo trutta after experimental infection Diseases ofAquatic Organisms 40(1) 9ndash18

93


Devold M Falk K Dale O B Krossoslashy B BieringE Aspehaug V Nilsen F Nylund A 2001 Strainvariation based on the hemagglutinin gene inNorwegian ISA virus isolates collected from 1987 to2001 indications of recombination Diseases ofAquatic Organisms 47(2) 119ndash128

Falk K Namork E Rimstad E Mjaaland SDannevig B H 1997 Characterization of infectioussalmon anemia virus an orthomyxo-like virusisolated from Atlantic salmon (Salmo salar L)Journal of Virology 71(12) 9016ndash9023


Gibbs M J Armstrong J S Gibbs A J 2001 Thehemagglutinin gene but not the neuraminidase geneof lsquoSpanish flursquo was a recombinant PhilosophicalTransactions of the Royal Society of London B356(1416) 1845ndash1855

Gregory A 2002 Detection of infectious salmonanaemia virus (ISAV) by in situ hybridisationDiseases of Aquatic Organisms 50(2) 105ndash110

Griffiths S Melville K 2000 Non-lethal detection ofISAV in Atlantic salmon by RT-PCR using serum andmucus samples Bulletin of the European Associationof Fish Pathologists 20(4) 157ndash162


Kibenge FSB Lyaku J R Rainnie D HammellK L 2000 Growth of infectious salmon anaemiavirus in CHSE-214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81(1) 143ndash150

Kibenge F S Kibenge M J McKenna P KStothard P Marshall R Cusack R R McGeachyS 2001a Antigenic variation among isolates ofinfectious salmon anaemia virus correlates withgenetic variation of the viral hemagglutinin gene

Journal of General Virology 82(12) 2869ndash2879

Kibenge FSB Garate O N Johnson GArriagada R Kibenge MJT Wadowska D 2001bIsolation and identification of infectious salmonanaemia virus (ISAV) from Coho salmon in ChileDiseases of Aquatic Organisms 45(1) 9ndash18


Krossoslashy B Devold M Sanders L KnappskogP M Aspehaug V Falk K Nylund A KoumansS Endresen C Biering E 2001a Cloning andidentification of the infectious salmon anaemia virushaemagglutinin Journal of General Virology 82(7)1757ndash1765

Krossoslashy B Nilsen F Falk K Endresen C NylundA 2001b Phylogenetic analysis of infectious salmonanaemia virus isolates from Norway Canada andScotland Diseases of Aquatic Organisms 44(1) 1ndash6

Loslashvdal T Enger Oslash 2002 Detection of infectioussalmon anaemia virus in sea water by nested RT-PCR Diseases of Aquatic Organisms 49(2)123ndash128

Lovely J E Dannevig B H Falk K Hutchin LMacKinnon A M Melville K J Rimstad EGriffiths S G 1999 First identification of infectioussalmon anaemia virus in North America withhaemorrhagic kidney syndrome Diseases of AquaticOrganisms 35(2) 145ndash148

McBeath AJA Burr K L-A Cunningham C O2000 Development and use of a DNA probe forconfirmation of cDNA from infectious salmonanaemia virus (ISAV) in PCR products Bulletin of theEuropean Association of Fish Pathologists 20(4)130ndash134

McCullers J A Wang G C He S Webster R G1999 Reassortment and insertion-deletion arestrategies for the evolution of influenza B viruses innature Journal of Virology 73(9) 7343ndash7348

94


Mikalsen A B Teig A Helleman A L MjaalandS Rimstad E 2001 Detection of infectious salmonanaemia virus (ISAV) by RT-PCR after cohabitantexposure in Atlantic salmon Salmo salar Diseases ofAquatic Organisms 47(3) 175ndash181


Mjaaland S Hungnes O Teig A Dannevig B HThorud K Rimstad E 2002 Polymorphism in theinfectious salmon anaemia virus (ISAV)hemagglutinin gene importance and possibleimplications for virulence evolution and ecology ofISA disease Virology 304 379ndash391

Mjaaland S Rimstad E Cunningham C O 2002Molecular diagnosis of infectious salmon anaemia InCunningham C O ed Molecular diagnosis ofsalmonid diseases Dordrecht Kluwer AcademicPublishers 1ndash22

Opitz H M Bouchard D Anderson E Blake SNicholson B Keleher W 2000 A comparison ofmethods for the detection of experimentally inducedsubclinical infectious salmon anaemia in Atlanticsalmon Bulletin of the European Association of FishPathologists 20(1) 12ndash22

Rimstad E Falk K Mikalsen A B Teig A 1999Time course tissue distribution of infectious salmonanaemia virus in experimentally infected Atlanticsalmon Salmo salar Diseases of Aquatic Organisms36(2) 107ndash112

Rimstad E Mjaaland S Snow M Mikalsen A BCunningham C O 2001 Characterisation of theinfectious salmon anemia virus genomic segmentthat encodes the putative hemagglutinin Journal ofVirology 75(11) 5352ndash5356

Ritchie R J Cook M Mellville K Simard NCusack R Griffiths S 2001a Identification ofinfectious salmon anaemia virus in Atlantic salmonfrom Nova Scotia (Canada) evidence for functionalstrain differences Diseases of Aquatic Organisms44(3) 171ndash178


Ritchie R J Bardiot A Melville K Griffiths SCunningham C O Snow M 2002 Characterisationof genomic segment 7 of infectious salmon anaemiavirus (ISAV) Virus Research 84(1-2) 161ndash170

Sandvik T Rimstad E Mjaaland S 2000 The viralRNA 3-and 5-end structure and mRNA transcriptionof infectious salmon anaemia virus resemble those ofinfluenza viruses Archives of Virology 145(8) 1659ndash1669

Schierup M H Hein J 2000 Consequences ofrecombination on traditional phylogenetic analysisGenetics 156 879ndash891

Snow M Cunningham C O 2001 Characterisationof the putative nucleoprotein gene of infectioussalmon anaemia virus (ISAV) Virus Research 74(1ndash2) 111ndash118

Snow M Ritchie R Arnaud O Villoing SAspehaug V Cunningham C O 2003 Isolation andcharacterisation of segment 1 of the infectioussalmon anaemia virus genome Virus Research92 99ndash105

Stagg R M Bruno D W Cunningham C ORaynard R S Munro P D Murray A G AllanCET Smail D A McVicar A H Hastings T S2001 Epizootiological investigations into an outbreakof infectious salmon anaemia (ISA) in Scotland FRSMarine Lab Rep 1301 Aberdeen UK FisheriesResearch Services 60 p

95


Worobey M Holmes E C 1999 Evolutionaryaspects of recombination in RNA viruses Journal ofGeneral Virology 80(10) 2535ndash2543


Knut FalkNational Veterinary InstituteFish Health SectionUllevalsveien 68Oslo 0033Norway

96


97

Infectious Salmon Anemia in Norway and the Faroe IslandsAn Industrial Approach

Infectious Salmon Anemia in Norway and theFaroe Islands An Industrial Approach

Cato Lyngoslashy1

1Cato Lyngoslashy is the Quality Assurance Director for PanFish Group in Aringlesund Norway

Disease Status

During most of the 1990s infectious salmon anemia(ISA) was classified as a spot-vice disease in Norway(Varinggsholm et al 1994) but reemerged as a severethreat to the Atlantic salmon (Salmo salar) industry in2001 (Martin Binde personal communication)Simultaneous with its reemergence in Norwegianaquaculture (fig 1) ISA also developed withincreasing severity in the Faroe Islands (PeterOslashstergaard personal communication) When ISA isdiagnosed in Norway governmental authorities

Abstract After existing as a spot-vice disease during mostof the 1990s infectious salmon anemia (ISA) emerged as asevere threat to the salmon farming industry throughoutNorway in 2001 Simultaneously the disease also devel-oped with increasing severity in the Faroe Islands Recordscomparing the number of outbreaks today with historic datawill be shown both for Norway and the Faroe IslandsAlthough ISA has consistently caused great losses at anaffected site its economic importance has dramaticallychanged recently Insurance indemnification programseffectively reduced the economic losses for the scarce

number of outbreaks that occurred during the 1990s butthe present increase of ISA outbreaks has reduced thewillingness of insurance companies to provide sustainedcoverage This situation has forced the industry andgovernmental agencies to establish effective measuresagainst the disease Figures demonstrating the economicimpact of ISA to the industry will be presented This paperwill further discuss how changes made in the salmonindustry during the last decade have facilitated the reemer-gence of a low-virulence agent like ISAv

Year

Norway totalHordaland Sogn og Fjordane

90

80

70

60

50

Num

ber

of IS

A o

utbr

acks

40

30

20

10

0

19841985

19861987

19881989

19901991

19921993

19941995

19961997

19981999

20002001

2002

Figure 1mdashNumber of sites subject to restrictions due to ISAdiagnosis in Norway totally and in the counties of Hordaland andSogn og Fjordane

subject the affected site to specific restrictions If anyfish on the farm contract this disease the entirepopulation is affected by the restrictions

The development of ISA in Norway occurredpredominantly within the counties of Hordaland andSogn og Fjordane which represent 54 percent of thetotal number of outbreaks throughout the period(Eide 1992 Martin Binde personal communication)From 1997 until 2002 these counties reported56 percent of the total number of Norwegianoutbreaks Compared with the percentage ofproduction which is 29 this area is overrepresented

98


in the incidence of ISA (Martin Binde personalcommunication) The record of ISA in the FaroeIslands is more recent and not as historicallyestablished as it is in Norway (Peter Oslashstergaardpersonal communication)

Effect on Industry

Once an outbreak of ISA occurs current manage-ment practices require immediate harvest or cullingof the affected fish Similarly other groups of salmonwithin the affected site are subject to acceleratedharvest if their mortality exceeds 05 per thousandfish per day per pen If mortality is lower than thislimit production may proceed as normal and harvestis subsequently prolonged In other areas of theaffected farm one may agree on a harvesting plan tobe instituted based on the severity of infection Thishas been and will remain the legislative practiceuntil a new national contingency plan is implementedin Norway Healthy fish within the affected farm maybe produced for human consumption without specialtrade restrictions2

When outbreaks of ISA reach the levels seenthrough 2000 and 2001 the disease not only affectsindividual sites but also causes further complicationsfor neighboring farms within the affected area Anepidemiologic study conducted in 1995 documentedthat the risk of ISA disease increased significantly ifan adjacent site became infected (Jarp and Karlsen1997)

Contagion of sites in this manner demonstratedthe importance of site segregation because the riskof disease was considerably reduced if the distancebetween sites exceeded 5 km (Jarp and Karlsen1997) This scenario was clearly evident in Gulenand Nordfjord both of which are located in countySogn og Fjordane where the County VeterinaryOffice ordered a regional fallowing policy amongaffected farms Obviously this regime alteredproduction plans for the farming companies involved

and production was lost among those companies thatdid not have access to alternative sites

ISA Insurance Programs

Although outbreaks of ISA always cause loss ofprofit the economic significance of the disease hasvaried When insurance indemnification programswere introduced in 1996 limits were established forpotential losses that made it possible for a farmer tosurvive even if the disease forced total depopulationof saleable fish In some cases the farmer couldeven recover the book value of the fish

Insurance coverage for ISA indemnification haschanged however now that the disease is moreprevalent and more sites are affected Naturallyindemnification against ISA losses has become lessattractive to insurance companies In certain regionscompanies no longer allow fish farms to apply for ISAinsurance in other areas insurance may be obtainedonly with an elevated owner risk Unlike Norway ISAinsurance has never been commonly available in theFaroe Islands Insurance with 50-percent owner riskwas offered a short period before the ISA epidemicbegan Presently [April 2003] insurance is notavailable

The following profit-to-risk example is used todemonstrate how ISA became an intolerable financialrisk for the industry In this example a site isselected where the fish weigh an average of 1 kgThis is the point in production schedules at which theeffects of an ISA diagnosis hurt the most because itis at this weight that a fish has hardly any marketvalue it is too small to attract a fair price and it isvery costly to slaughter Consequently fish less than1 kg will usually be culled and used as a protein andfat resource in feed for fur-bearing animals Thisutilization pattern produces financial loss In fact thefarmer pays to dispose of the fish All other scenariosinvolve less expense to the farmer because the fisheither have a lower book value or there may bepartial returns by selling some portion of theproduction lot in the market Thus the economicimpact of ISA as a threat to the industry is best

2The new contingency plan will come into force fromFebruary 1 2003

99


visualized in a profit-to-risk ratio This ratio expressesthe relationship between predicted return on a three-concession site (meaning triplicate productionschedules on each site per grown year class) dividedby the estimated financial loss if ISA occurs If sucha ratio is 1 the risk equals the potential profit thusone ISA outbreak is repaid by one site with normalproduction

During 2000 the predicted profit per kilogram offish harvested was very high because supply offarmed salmon in the market was lower than thecommercial demand and the industry took advantageof cost-reducing factors in production Accordinglythe declared insurance value often exceeded bookvalue and the market price also exceeded theinsurance value Therefore insurance payments insome cases covered for book value even given 30-percent ownerrsquos risk Practically a maximum loss of$140000 could result if ISA occurred [All costfigures are expressed in US dollars] Because othersites within the same company may producenormally and gained profit up to $26 million this riskwas indeed tolerable Thus the profit-to-risk ratiowas favorable to fish farming in 2000 but this trenddid not persist (Ragnar Nystoslashyl personalcommunication) Market prices affected thewillingness of a farmer to grow more fish whichultimately led to overproduction During 2001therefore market prices dropped drasticallyMeanwhile the risk of having ISA contaminate a farmwent up This change resulted in an increasedpercentage of owner risk Simultaneously costs ofproduction also increased because feed prices rose25 centskg (internal data) Feed represents abouthalf the cost of production Consequently thedifference between insurance and book valueszeroed or turned negative Owing to the higherprevalence of ISA owner risk went from 20 percentto 50 percent and in some cases even reached 100percent Predictable loss if ISA occurredconsequently increased from $128000 in 2000 tobetween $940000 and $1740000 in 2002 Within 2years the profit-to-risk ratio had changed to an

extremely nonfavorable status and turned financialrealities (table 1) upside down for many farmers(Ragnar Nystoslashyl personal communication) A profit-to-risk ratio of 002 means that 50 normal productionsare required to repay 1 ISA outbreak

Governmental agencies within the FaroeIslands treated ISA almost alike but a distinctdifference between the Faroes and Norway was thataccessible sites were limited in the FaroesTherefore the number of fish per site in the FaroeIslands was even higher than in Norway Numbersless than 1 million fish were rare and often there wereas many 2 to 3 million fish produced per siteUnfortunately production statistics were notavailable

To date production sites in the Faroe Islandshave not sustained an escalating ISA mortality andtherefore production has not been disrupted totallyLosses have been limited to the actual pen culled orharvested However with prolonged culling ofpotentially diseased fish from a site the worst-casescenario might develop in a matter of time In such asituation the loss might go as high as $38 millionTable 2 presents a scenario for a Faroe Island siteprovided that all fish had to be culled at 1 kg duringthe 2002 production cycle (Ragnar Nystoslashyl and PeterOslashstergaard personal communication)

The salmon farming industry generally dependson financial support to produce fish and financiallending institutions are obviously risk orientedBecause of the profit-to-risk ratios presentedpreviously it became imperative for the industry inNorway and the Faroe Islands to improve the ISAsituation Together with governmental agenciesworking groups were established to develop newnational contingency plans Except for local andgeographic adaptations both national plansdeveloped similar principles to control and manageISA by modernization of hygienic measures andimplementation of vaccination programs3

3As to the use of vaccines against ISA this part of thenational plans is still subject to EUEuropean Fair TradeAssociation approval

100


Table 2mdashInfectious salmon anaemia profit-to-risk ratioand potential loss for a typical site in the Faroe Islandsholding 12 million fish during 2002 no insurance(100-percent owner risk)

Per kg Total value(fish) per site1

Book value per kg 1317 3804000Insurance value per kg 0 0Deductible owner risk 100 0Payable from insurance 0Percent coverage 00Total loss per site 3804000Normal harvest size (gutted) 45Normal predicted profit per kg 10013Normal predicted profit per site 70200Profit-to-risk ratio 002

1 Figures represent US dollars at the exchange rate of $1 = 758 Norwegiankroner

Table 1mdashInfectious salmon anemia profit-to-risk ratioand potential loss for a Norwegian three-concessionsite holding 550000 fish

a Figures calculated at 20-percent owner risk for the year2000



b Figures calculated at 50-percent owner risk for the year2002

Per kg Total value(fish) per site


c Figures calculated at 100-percent owner risk (no insurancecoverage) for the year 2002




Viral Infection of AtlanticSalmon

Even though ISA virus (ISAv) expresses differentlevels of virulence in the field in vivo challenge trialsand epidemiologic studies indicate that the virus isonly moderately contagious It is found in several fishspecies including Atlantic salmon (Salmo salar)(Nylund et al 1995) rainbow trout (Oncorhynchusmykiss) (Nylund et al 1997) brown trout (Salmotrutta) (Nylund et al 1994) Arctic charr (Salvelinisalpinus) (Snow et al 2001) and herring (Clupeaharengus) (Are Nylund personal communication)Although this disease was not described before theadvent of commercial Atlantic salmon aquaculture ithas been documented that the virus was presentbefore it was first detected as the causative agent ofISA (Krossoslashy et al 2001) The virus has evidentlysurvived and propagated in wild fish Unfortunatelywe do not know much about viral replication andsurvival in wild populations Such investigationswould provide knowledge essential to combat thedisease

Evans and Brackman (1991) have presented amodel (fig 2) that depicts the presence of a pathogen

101


in a host population This model defines four differentcategories in which the host animal may find itself(a) a healthy organism that does not harbor thepathogen (b) a healthy organism that harbors alatent or carrier infection (c) a carrier organism withclinical signs of disease and (d) a dead organismthat sheds the virus

The ratio of individuals belonging to eachcategory depends on environmental parameters thevirulence of the pathogen and the susceptibility ofthe host This model visualizes some of the problemsassociated with a moderately virulent disease likeISA Strategy is therefore directed at managementof the disease rather than elimination of the virusNevertheless it is difficult to judge whether or notthese measures are effective in a long-termperspective by assessing the outcome of clinicaloutbreaks alone As long as knowledge of theunderlying carrier status is scarce epidemiologiccontrol and transmission of the pathogen remainunclear Analytical techniques applicable for suchanalyses have become available in the past 2 yearsbut methodologies still require validation

Balances and imbalances prevail betweendifferent health categories in the pyramid (fig 2) thatmay either manifest or suppress the expression ofclinical disease and mortality The balance may bealtered owing to a variety of external and internalfactors generally referred to as stress-inducingparameters Normal production cycles containseveral stressors including operations (egtransport sea-lice treatments grading andtrenching) the maturation process itself and thepresence of concomitant diseases among othersAlso host susceptibility may vary within salmonpopulations and among different strains

Detection of ISAv with the reverse-transcriptasendashpolymerase chain reaction (RTndashPCR)technique has been used since 2001 (Devold et al2000 Griffiths and Melville 2000) Although resultsfrom different laboratories may not be comparableanalyses have revealed that a wide range of carrierstatus exists among farmed salmon populationsSome farm sites may have a prevalence of latentcarriers as high as 60 percent without manifestationof clinical ISA (internal data) Historically it has beentheorized that the vast majority of challengeoriginates with the shedding of viral particles fromdiseased and dead fish Certainly diseased anddead fish represent an important reservoir ofinfection However the virus must depend on othermechanisms of infection within the feral environmentthan those that facilitate transmission of the pathogenunder conditions of intensive pen culture Within theferal environment infection may be affected bytransmission of the virus across individuals within ashoal or when proximity of fish-to-fish contactsincreases at the time of spawning Dead or diseasedfish are left behind and are not able to follow theshoal Several viruses are present in gonadal fluidsThe spawning pattern for anadromous or pelagicspecies presents three important factors forsuccessful viral reinfection (1) concentration ofhosts (2) exposure to contaminated gonadal fluidsand (3) the possibility of ingesting contaminatedfertilized eggs or infected offspring The identificationof latent ISAv carriers in farmed salmon populationsraises questions about whether the virus may have a

Dead

Diseased

Latent carriers

Healthy non-carriers

Figure 2mdashAn iceberg model of infection depicting variousinfectious stages within a population as described by Evans andBrackman (1991)

Healthy noncarriers

102


natural mechanism of spreading without producingclinical disease This is an issue that can only beaddressed by further research

Structural Changes ThatAffect Prevalence

During the 1988ndash92 epidemic of furunculosis athorough restructuring took place within the salmon-farming industry Within this same period ISAemerged and served as an additional stimulus for therestructuring process In 1991 the first efficaciousoil-adjuvant vaccine against furunculosis becamecommercially available but neither pharmaceuticaltreatments nor vaccines were available to control ISAThus control could only be effected by improvingconventional hygienic measures

Separation of generations with ldquoall inndashall outrdquoproduction

Segregation of production sites by 3 km

Daily collection of dead fish with dead-fish brills ineach pen Hygienic measures including the installa-tion of silage tanks to handle dead fish at each site

Abandonment of shallow and low-current sites witha preference for more exposed sites with good waterexchange

Fallowing of culture sites between generations

Prohibition on movements of fish from site to site inthe sea unless such movements are for the purposeof slaughter

Implementation of sanitary packing and processingprocedures by disinfection of effluent water

Sanitation of wellboats between transports

Ultraviolet disinfection of seawater intake in smoltfarms

Collectively these measures established newstandards for the industry and helped to overcome itsfirst epidemic of ISA The increased reemergence ofISA since 2000 however indicates that thesemeasures were either insufficient or that personnelmay have become somewhat lax in their attention to

detail In the Faroe Islands some of thesefundamental hygienic measures have not yet beenfully implemented (Peter Oslashstergaard personalcommunication)

In addition to the development of hygienicstandards one of the major changes in the industryduring the last decade was a relocalization into newproduction sites More exposed sites required betterconstruction and sustained greater levels ofproduction Several concessions were alsoincorporated within the same site with theintroduction of feed barges and automatic feedsupplies With 3 or 4 concessions on the same sitethe number of fish increased to 550000 to 680000(Norwegian Export Council 2002 Directorate ofFisheries 2002) This enhanced greater possibilitiesfor ISA infection as follows

The ratio between wild and farmed fish haschanged considerably since 1990 Sea troutpopulations exhibit a regional life cycle and themajority of the population does not move more than afew kilometers from their home river (AtleKambestad personal communication) In Norwaythese populations are known to carry ISAv (Rollandand Nylund 1999 Nylund and Jakobsen 1995)Given that the size of the sea trout populationremains rather stable on a regional basis the ratiobetween sea trout and farmed salmon in some of thecore areas of ISA had changed from about 115 in1990 to 150 by summer 2002 In 1980 the ratio wasapproximately 11 (Atle Kambestad personalcommunication) This affects whether or not weshould consider wild or farmed fish as the mainreservoir for the agent In terms of populationnumbers farmed fish may represent an independentreservoir

The production pattern of transferring more fishper site also called for more fish per pen In 1990 itwas common to split a one-concession transfer into aminimum of three pens due to different sizes at thetime of transfer Today all fish may be put in one90-m polar circle Because smolts transferred perconcession are now 25 times greater than they werein the past present practices do indeed challenge the

103


smolt producer It is biologically and technicallydifficult to produce 150000 fish of consistent qualitythroughout the group Practicalities often necessitatethat the same pen contain smolts originating fromdifferent tanks in the smolt farm of different sizes andof inhomogeneous quality This situation predisposessalmon in a communal pen to social turbulences andstressors that can trigger overt disease even if thenumber of latent carriers was originally low

Also the increased number of fish per site canenhance the infectious load in the adjacent area ifISA infection is established in the population Eventhough ISA has reemerged during 2001 the numberof affected sites still remains lower than peaknumbers of sites reached in the early 1990s Thismight suggest that the present situation is not as badas it was 10 years ago That inference however isslightly incorrect because it does not take intoaccount that the number of fish per site hasincreased eightfold during the decade In 1990 a siteheld an average of 60000 to 70000 fish In 2001 atypical site was stocked with 550000 to 680000 fish(Norwegian Export Council 2002 Directorate ofFishery 2002) Again reverting to the iceberg modelwe find that the relative proportions of individualsbelonging to each level are not necessarily changedbut the numbers per category are much higherThus if an ISA outbreak takes place the number offish shedding virus and consequently the totalinfectious load to the environment and adjacent sitesare much higher One may perceive that better sitesand hygienic measures implemented by the industrywould outweigh this In fact most of the industry hascarried out higher stocking densities successfullyStill the disease has been able to recur in someregions in Norway such as Gulen and Nordfjordwhich are core areas for the disease Although theepidemiologic analysis conducted in 1995 has notbeen repeated the disease pattern in 2000ndash02 doesunderline the same risk factors despite the goodimplementation of hygienic operational measures

Another problem arising from high-stocked sitesis the time required to empty the site after ISA

diagnosis has been verified In general emptying thesite as soon as practically possible is perceived asthe most essential part of the Norwegian Contin-gency Plan Keeping the fish onsite while diseaseand infectious load are allowed to build up truly has avery negative impact on the adjacent sites as well asthe environment Still harvesting capacity may be alimiting factor in many areas In Norway a fullystocked site of harvest-size fish may require as muchas 50 working days to empty even with continuousharvest In the Faroes the job may take 100ndash120working days

Historically it has been found that challengehas been brought in from affected sites or wild fish tononaffected sites in various ways This is still of vitalimportance However at a site holding 600000 fishwith a carrier status as high as 60 percent (internalresults) chances may be higher that the agentoriginates from internal carriers (or diseased fish)than from outside introduction A self-sustaininginfection cycle cannot be excluded

On the assumption that the virus is naturallypersistent in wild fish introduction of fish farmingindeed established a new imbalance by introducing arelatively high number of susceptible hosts Thepresence of serotypes in ISAv has never beendocumented or investigated Variation in clinicalmanifestation is observed in the field Whereas somesites experience very low mortality others mayobserve an aggressive development of the diseasewith escalating mortality However the variation inmortality may also be explained by the developmentalstage at which the disease is diagnosed on a farmOn the basis of evolutionary principles one shouldexpect that introduction of a high number ofsusceptible hosts would destabilize the existingnatural balance between virus and hostsConsequently it may be questioned if a highlyvirulent virus has a greater ability to survive over timeand spread across farming units (Are Nylundpersonal communication) A comparative study ofvirulence across isolates from wild fish and farmedfish would verify this hypothesis

104


Regulation of Production byFeed Quotas

In Norway one production concession is based on acalculated volume which is equivalent to 12000 m3Within a concession there is a legislated density limitof 65 kg of fish per cubic meter of water and aguideline of 25 kgm3 based on health and animalwelfare recommendations Feed quotas perconcession were instituted in 1996 within theparameters of The Salmon Agreement negotiatedwith the European Union to regulate supply ofmarketable salmon The feed quotas have sincebeen adjusted annually by the NorwegianDepartment of Fisheries In 1996 the quota was 619tons of feed and it reached 847 tons in 2002 (fig 3)This increase occurred without adjustment of theallowed concession volume ie one concessioncould produce almost 50 percent more fish todaythan it did in 1996 Calculated maximum density hastherefore increased by 50 percent Although thefarmers have moved towards bigger pens and arethus partly compensated by retrieving a bigger actualvolume the feed quota policy has not been assessed

Year

19971996 19991998 2000 2002

900

50

60 Feed quota

Cumulative percent increase in density

30

40

20

0

10

800

700

600500

Tons

Den

sity

400

300

200

100

02001

Figure 3mdashPercentage increase in density of cultured NorwegianAtlantic salmon based on increase of feed quotas from 1996 to2002

from the perspective of salmon health Theproduction system which is based on volume (cubicmeters) is presently under revision

To the extent that limitation of density isbelieved to reduce horizontal interaction between fishand limit the potential for contagion measurementsof density as kilograms of biomass per cubic meter ofwater are somewhat inappropriate Infectious agentswhatever their nature do not infect biomass persome arbitrary unit of volume pathogens invade andinfect individual hosts Total biomass at a given sitewould have some relevance as a more appropriatemeasure than kilograms per cubic meter Althoughdensity expressed as kilograms per cubic meterpeaks just before grading and harvest animal density(fishcubic meter) peaks at transfer and declinesthroughout the production cycle (fig 4) From amicrobial perspective the chances for infecting a newhost are greatest at the start of the growth period Infigure 4 animal densities are calculated on the basisof a one-pen scenario

Also relevant from a health point of view isexamination of densities on a site basis The volumedefined by the whole farm as a unit will then apply

105


A site may consist of 6 to 12 pens These areorganized in a variety of ways in accordance with thetype of farming construction used anchoringpossibilities and the exposure of the site Basicallytwo types are availablemdashsteel construction framingall pens in a rigid structure or flexible plastic unitsorganized as separated pens In the case of steelconstruction the pens and consequently the fish arepositioned next to one another only with a walkway inbetween With flexible plastic units the units arespaced 10ndash40 m apart In the latter case thenumber of animals is stocked in a manner providingfrom two to six times more volume Normally the siteis stocked according to pen volume and that has tobe taken into account Nevertheless the steelconstruction provides neither the animal density northe biomass density afforded by the plastic circlesbased on site volume calculation (internal data TimOrsquoHara personal communication)

Introduction of feed quotas also had dramaticconsequences for feed composition The amount offeed became the legislatively regulated limiting factorfor how many tons of fish could be legally produced

per license Certainly in order to enhance productionfurther it was necessary to improve diets byincreasing energy units per kilogram of feed Since1996 the fat content in salmon diets has increasedfrom 27 to 41 percent and the energy content hasalso increased accordingly To increase theproportion of one ingredient in the feed another hasto be reduced The only way of compensating is toincrease the quality and specificity of the componentsbeing reduced Simultaneously high-energy feedallowed faster growth but reduced actual feedconversion rates (FCRs) Since 1996 and through2001 the average FCR for the Norwegian salmonindustry dropped from 1216 to 1175 (RagnarNystoslashyl personal communication) As a conse-quence of this development the prevalence ofdiseases during production had a tendency toincrease Increased growth capacity also requires amore concentrated feed for all other ingredientsRegrettably science takes time to catch up withmarket requirements In the meantime the fish maybecome compromised in diet and be adverselyaffected immunologically

30

25

20

15

10

5

0

0 1 2 3

Average weight in kg

kgm3

Numberm3

4 5

Den

sity

Figure 4mdashDevelopment of biomass density compared with animaldensity during growth in a standard salmon farm

106


Fish transportation has been restructured alongwith the development of the industry Some of thesechanges have not been favorable for preventivehealth Expansion of carrying capacity by largerboats had a very positive impact on the cost oftransport per kilogram of fish However densitiesduring transport are greater than in the pen Suchtransport also involves stress and the fish are likelyto receive and release viral agents en route topacking stations If the fish originate from a site notsubject to restrictions full water exchange duringtransportation is allowed If transportation routespass other culture sites this practice may enhancethe spread of ISA (Murray et al 2002) Alltransportation of live fish should be accomplishedwithout any emission of infective agents Boatsshould be self-contained and smolt transportationshould be carried out with particular boatsperiodically dedicated to this task

Once salmon are transported to the packingfacilities they are often loaded into resting pensThese pens were established to decrease the timerequired to unload the wellboat and allow the fish torecover from transport stress In reality howeversuch a practice allows posttransport mortality tooccur in an open environment with potential releaseof pathogens from the resting pens Such facilitiestherefore should be brought onshore in closedcompartments where they will not become factorsthat can amplify pathogens and enhance diseaseprocesses in natural environments When farmingcompanies integrated and grew bigger costreductions were effected by replacing small andineffective packing stations with new and largerfacilities To fill packing capacity fish weretransported across larger distances which couldresult in pathogens such as the ISA virus beingspread across broader areas

Oftentimes the interests of cost-effectivenessare at odds with biology and this is also true with thetransportation sector Problems have been causedby trenching of fish in the pen before loading thespeed of unloading the lack of rest for transportstaffs errors in hygienic and management routines

and so forth All these factors adversely affect theanimal the economic bottom line and ultimately totalenvironmental health It is therefore a paradox thatthe industry itself has reduced transport cost to thepoint that spread of ISA has actually been enhanced(Murray et al 2002)

Transportation of mixed species is practiced bysome of the wellboat companies Farmed fish andtrimming (byproducts) from herring and mackerel arenever transported simultaneously but these productsmay be handled by the same boat on alternate tripsISAv is isolated from and verified capable ofpropagation in herring (Are Nylund personalcommunication) Byproducts consisting of intestinesstomachs and livers definitely contain numerous viralparticles Additionally this material is often chilledand stored from 1 to 3 days before transportationAlthough the wellboat is washed with strongdetergent and disinfectants before the followingtransport the fat content of this material stronglyadheres to most surfaces Thus completedisinfection of wellboat facilities after transport ofsuch materials is a technical challenge Satisfactorycleaning after transportation of byproducts andtrimmings is virtually impossible It is not hard toimagine what might happen if the next mission of acontaminated boat involved the movement ofsusceptible salmon smolts It is the opinion of thisauthor that combined transportation of live fish andtrimmings should be prohibited

One of the most significant victories for thesalmon-farming industry was a reduction in the use ofantibiotics realized through good husbandrypreventive hygienic measures and the developmentof more efficacious oil-based adjuvant vaccinesagainst certain fish pathogens Between 1987 and1999 the salmon industry had managed to reduce itsuse of antibiotics by 99 percent and during the sameperiod total salmon production increased eightfold

Besides supporting a slow-release system forthe antigens oil-based adjuvant vaccines also createa certain nonspecific immunity This immunityprevails for ISA as well and several in vivo challengeexperiments have documented this protection (Jones

107


et al 1999 Anne Ramstad and Odd Magne Roslashdsethpersonal communication) Veso Vikan Akvavetperformed experiments in 1997 with an oil-basedvaccine that compared results against nonvaccinatedcontrols which produced a 35-percent relativepercent survival against ISA Later Intervet Norbiotested their prototype combination ISA vaccineagainst nonvaccinated fish and found that all of theoil-based vaccines provided relative percent survivalsthat ranged from 31 to 35 percent (Odd MagneRoslashdseth personal communication)

During the first half of the 1990s vaccinecompanies competed to improve the efficacy ofproducts but efforts were specifically directed againstfurunculosis This effort resulted in very effectivevaccines indeed but also increased the incidence ofadverse side effects like abdominal adhesions andmelanization (Midtlyng and Lillehaug 1998 Poppeand Breck 1997) Consequently vaccine companiesrestructured their strategies and attempted tobalance between satisfactory efficacy and acceptablelevels of side effects Some vaccines went too far inthe attempt to lower side effects at the clear expenseof efficacy In 1998 and 1999 the salmon industryactually experienced breakdowns in protectionagainst furunculosis for some vaccines No studywas conducted to investigate whether a link existedbetween the use of specific vaccines and theupcoming ISA situation that occurred from 2000 until2002 Fish that were harvested in 2000 wereevidently vaccinated from August 1998 to April 1999Statistical information about vaccine sale per productis available from distributors in Norway Collectively139 million salmon (Alistair Brown personalcommunication) were given doses originating fromvaccines linked to lowered field protection againstfurunculosis

Although documented by Jones et al (1999) inthis paper it can only be speculated that thenonspecific protection against ISA provided byeffective oil-based vaccines reduced the risk ofclinical outbreaks and that the use of less-than-efficacious vaccines accordingly increased the risk

Conclusion

Recognizing that fish farming is intensive foodproduction on the same scale as that of the poultryand swine industries we should not be surprised toencounter biological bottlenecks The emergence ofdiseases such as ISA should be perceived as abiological lighthouse that can be controlled through acombination of management and pharmaceutical andimmunological techniques The disease thereforeforces legislative and industry personnel toreevaluate current practices and develop standardsthat will lead to development of biologically balancedand sustainable production in aquaculture

References Cited

Devold M Krossoy B Aspehaug V Nylund A2000 Use of RTndashPCR for diagnosis of infectioussalmon anaemia virus (ISAV) in carrier sea troutSalmo trutta after experimental infection Diseases ofAquatic Organisms 40 9ndash18

Directorate of Fisheries Bergen Norway PublicStatistics 2001 httpwwwfiskeridirnoenglishpagesstatisticsstatisticshtml

Eide G W 1992 A retrospective analysis ofoutbreaks of infectious salmon anaemia in Sogn andFjordane in 1985ndash91 [En retrospektiv analyse avILA-utbrudd i Sogn og Fjordane 1985ndash91] NorskVeterinaertidsskrift 104 915ndash919

Evans A S Brackman P S 1991 Epidemiologicalconcepts In Bacterial infections of humans 2d ed[Place of publication unknown Plenum Medical]



108


Jones SRM Mackinnon A M Salonius K 1999Vaccination of freshwater-reared Atlantic salmonreduces mortality associated with infectious salmonanaemia virus Bulletin of the European Associationof Fish Pathologists 19 98ndash101

Krossoslashy B Nilsen F Falk K Endresen C NylundA 2001 Phylogenetic analysis of infectious salmonanaemia virus isolates from Norway Canada andScotland Diseases of Aquatic Organisms 4 1ndash6

Midtlyng P J Lillehaug A 1998 Growth of Atlanticsalmon Salmo salar after intraperitonealadministration of vaccines containing adjuvantsDiseases of Aquatic Organisms 32 91ndash97


Norwegian Export Council Eksportutvalget for fiskTromsoslash Norway Public Statistic 2001httpwwwseafoodnofaktaeksport

Nylund A Alexandersen S Loslashvik P Jakobsen P1994 The response of brown trout (Salmo trutta L)to repeated challenge with infectious salmonanaemia (ISA) Bulletin of the European Associationof Fish Pathologists 14 167ndash170

Nylund A Kvenseth A M Krossoslashy B 1995Susceptibility of wild salmon (Salmo salar L) toinfectious salmon anaemia (ISA) Bulletin of theEuropean Association of Fish Pathologists 15 152ndash155

Nylund A Jakobsen P 1995 Sea trout as a carrierof infectious salmon anemia virus Journal of FishBiology 47 174ndash176


Poppe T T Breck O 1997 Pathology of Atlanticsalmon Salmo salar intraperitoneally immunized withoil-adjuvanted vaccine A case report Diseases ofAquatic Organisms 29 3



Varinggsholm I Djupvik H O Willumsen F V TveitA M Tangen K 1994 Infectious salmon anaemia(ISA) epidemiology in Norway Preventive VeterinaryMedicine 19 277ndash290


Martin Binde Deputy Regional Veterinary OfficerNorwegian Animal Health AuthorityRegional Veterinary Office of Hordaland and Sogn ogFjordaneBontelabo 9A5001 BergenNorwayPhone +47 55363840 e-mailmartinbindedyrehelsetilsynetno

Alistair Brown Managing DirectorIntervet Norbio ASHigh Technology CenterThormoslashhlensgt 555008 BergenNorwayPhone +47 55543778 e-mailalistairbrownintervetcom

109


Atle Kambestad Wild Salmonid ManagerCounty of Hordaland Department of EnvironmentalAffairsCounty Governor of HordelandPostbox 73105020 BergenNorwayPhone +47 55572212 e-mail AtleKambestadFm-Hostatno

Are Nylund Professor in Fish DiseasesDepartment of Fish and Marine BiologyUniversity of Bergen5020 BergenNorwayPhlone +47 55584444 e-mailAreNylundifmuibno

Ragnar Nystoslashyl Aquaculture AnalystKontali Analyse ASIndustriveien 186517 KristiansundNorwayPhone +47 71683300 e-mailragnarnystoylkontalino

Tim orsquoHara Fish BiologistOmega Salmon Group Ltd124-1334 Island Highway Campbell RiverBC V9W 8C9CanadaPhone +1 250 286 1599 e-mailtimomegasalmoncom

Peter Oslashstergaard Head of DepratmentVeterinary and Environmental AgencyFish Disease LaboratoryFalkavegur 6FO-100 ToacutershavnFaroe IslandsPhone +298 356479 Cell phone +298 216479e-mail psohfsfo

Anne Ramstad Trial LeaderVESO Vikan AkvaVetAlhusstrandN-7800 NamsosNorwayPhone +47 74217770 e-mailanneramstadvesono

Odd Magne Roslashdseth Head of Research andDevelopmentIntervet Norbio ASHigh Technology CenterThormoslashlensgt 555008 BergenNorwayPhone +47 55543750 e-mailoddmagneroedsethintervetcom

110


111

The Eradication of an Outbreak of Clinical InfectiousSalmon Anemia From Scotland

The Eradication of an Outbreak of ClinicalInfectious Salmon Anemia From Scotland

Ronald M Stagg1

Introduction

Infectious salmon anemia (ISA) is a contagious andsignificant viral disease of farmed Atlantic salmon(Salmo salar) that first emerged in Norway in 1984(Thorud and Djupvik 1988) Subsequently ISA wasreported in Canada (initially as hemorrhagic kidneysyndrome (HKS) in 1996 (Bouchard et al 2001Mullins et al 1998 Lovely et al 1999) The diseasewas then reported in Scotland in 1998 (Rodger et al1998 Stagg et al 2001) the Faroe Islands in May2000 (Commission of the European Communities2002) and in the United States [in Maine] in 2001(Commission of the European Communities 2002)The outbreak in Maine was most likely a conse-quence of spread from adjacent Canadian waters

The Scottish outbreak represented the firstreport of ISA within the European CommunityPreviously ISA had not been recorded during officialsurveillance and testing of farmed or wild fish in theUnited Kingdom even though the disease has beennotifiable since 1990 (Hill 1996) This paperdescribes the Scottish outbreak and measures takento eradicate the disease It also providescommentary on the origins of the infection and themanagement measures needed for adequate controlEradication of clinical ISA from Scottish salmon farmswas successful although new legislation wasintroduced to allow flexibility for rates of depopulationof farms and opportunities for vaccination should

Abstract In 1998 an epidemic of infectious salmonanemia (ISA) occurred in marine salmon farms in ScotlandEleven farms were confirmed to have infected fish and afurther 25 farms were suspected of containing infected fishThe last suspected case was in November 1999 and theepidemic was officially over on January 16 2002 when thelast farm that shared coastal waters with a suspect casehad completed the requisite fallow period In the EuropeanUnion ISA is a List I disease and a strong legislativeframework had already been established in 1993 to bringabout its eradication This framework made the diseasenotifiable and provided measures to contain and limit

spread and to eliminate the source of infection On suspectfarms and farms in the same coastal waters the legislationprovided for restrictions on the movements of live fish andfish to harvest as well as mandatory biosecurity provisionsunder the supervision of the Official Service2 On farmswhere the disease had been confirmed immediatedepopulation followed by fallowing and disinfection wasrequired The average time for withdrawal of all fish in theScottish epidemic was 21 days and varied in accordancewith local logistics and the size of the farm This togetherwith the established legal framework to limit spread wasthought to have facilitated successful eradication

1 Dr Stagg is with the Fisheries Research Servicesrsquo MarineLaboratory in Aberdeen UK

2 The administration of the fish health legislation inScotland is a devolved function split between the ScottishExecutive Rural Affairs Department (responsible for policydevelopment) and Fisheries Research Services(responsible for administration enforcement andsurveillance) There is liaison with the UK Department ofEnvironment Food and Rural Affairs which hasresponsibility for international relations with the EU and theOffice International des Epizooties (OIE)

these be needed in the future If an eradicationstrategy is pursued affected farms should bedepopulated at the earliest opportunity Furthermoregiven that there is putative evidence for the sporadicoccurrence of the virus that causes ISA (ISAv) infarmed and wild fish preventive measures to thwartemergence of clinical disease are essential

Characteristics of ISA

The infectious nature of ISA was first demonstrated in1987 (Thorud and Djupvik 1988) In early work onthe disease the ability of liver homogenate andinfected plasma to transmit the infection was reducedafter treatment with diethyl ether and chloroformsuggesting the presence of an enveloped virus(Christie et al 1993 Dannevig et al 1995)Knowledge about the etiology of ISA progressedslowly until the virus was isolated in salmon headkidney cells (Dannevig et al 1995) Electronmicroscopy and biochemical studies on a Norwegianisolate revealed a single-stranded envelopedspherical ribonucleic acid (RNA) virus (Nylund et al1995 Dannevig et al 1995 Falk et al 1997) Genetic

112


studies (see the review by Cunningham and Snow inthis volume) showed that the ISAv is amultisegmented negative sense RNA orthomyxovirusRecent classifications showed that there weremultiple forms of the virus identified on sequencecharacteristics of the hemagglutinin (HA) gene andthat these probably vary in their pathogenicity(Rimstad and Mjaaland 2002)

Because striking parallels exist with influenzaviruses (Webster 1998) particularly important in themanagement of ISA will be the emergence of newvirulent or pathogenic forms of ISAv Recombination(Devold et al 2001) and deletion from an ancestralgene (Cunningham and Snow this volume) havebeen proposed as mechanisms for the emergence ofnew strains of ISA in relation to the hypervariableregion of the HA gene The observation of multiplestrains circulating in individual fish (reviewed inCunningham and Snowrsquos paper) enhances thelikelihood that these mechanisms occur Thesecharacteristics also have implications for the relativeimportance of vaccination and eradication as controlmeasures

Disease outbreaks appear to be associated withhighly infectious forms of the virus transmitted via livefish urine feces skin mucus and dead fish Thevirus adheres to red blood cells through attachmentvia the HA protein In virulence experiments (Totlundet al 1996) blood homogenate is the most infectiousinoculum and passive transmission is assisted byvirusndashcontaminated waste such as blood inslaughterhouse effluent This finding emphasizes theneed for satisfactory disinfection practice in relationto fish farms and slaughterhouse waste and effluentto reduce the risk of ISA transmission Sea lice havebeen suggested to be a vector for transmission(Rolland and Nylund 1998) because they feed on theblood and tissue of their host The disease appearsto be restricted to marine salmon farms Severalfactors are important in determining whether or notfish at a seawater salmon site develop clinicaldisease (Jarp and Karlsen 1997)

Proximity to another ISA-affected farm

Proximity to untreated processing effluent

Sites that exchange fish between processing andgrowout areas

Multigeneration sites

Poor farm hygiene (especially bleeding at sea) and

Sites with multiple sources of smolts

Management of the Epidemic

ISA is a List I disease in the European Union and ismanaged under the auspices of two EuropeanCommission (EC) directives governing trade inaquaculture products (Commission of the EuropeanCommunities 1991) and control of fish diseases(Commission of the European Communities 1993)These require that the management of ISA inScotland be based on three elements

Passive surveillance to identify the first occurrenceof the disease at the earliest opportunity

Application of biosecurity measures to preventspread and

Elimination of the source of the infection

Within the legislation action is required if ISA issuspected to be present among fish on a farm and ifinvestigation of suspicion results in confirmation ofthe presence of the disease In Scotland farms wereconfirmed if fish in the farm showed clinical diseasecharacteristic of ISA (ie clinical signshistopathological lesion(s) pathogonomic of ISAanemia) and there was evidence of infection withISAv by culture reverse-transcriptasendashpolymerasechain reaction (RTndashPCR) assay or the indirectfluorescence antibody test (IFAT) Suspect farms hadfish with evidence of infection but tests to prove thepresence of clinical disease were inconclusiveSubsequent changes to the legislation now requirethat a site be confirmed if ISAv is isolated in twosamples from one or more fish at the farm tested onseparate occasions

113


Notification and Surveillance

Since 1991 ISA has been notifiable to the OfficialService in the United Kingdom This measure issupported by official inspection and sampling (once ayear for marine sites) to guarantee continuedfreedom from viral hemorrhagic septicemia (VHS)and infectious hematopoietic necrosis (IHN)Experience shows that although surveillanceconfirms the absence of an infectious agent it hasnot been particularly useful at detecting newoutbreaks For example initial outbreaks of VHS(Munro 1996) ISA (Rodger et al 1998) and sleepingdisease (Branson 2002) were first identified throughnotification of the disease or investigations ofabnormal mortality rather than detection by officialsurveillance It follows from this that an importantaspect of risk reduction is that notification bemandatory and that there be good relations betweenthe Official Service and industry to promote reportingFollowing the initial outbreak the Official Servicethen carried out an enhanced surveillance program inthe first instance over a wide area (40 km from theaffected sites) and subsequently in focusedsurveillance zones around suspect and confirmedsites This surveillance aided by contact tracing diddetect new secondary outbreaks and was animportant aspect of bringing the disease undercontrol

Measures To Prevent Spread

UK authorities already had strong enablinglegislation to implement the EC directive for thecontrol of fish diseases (European Commission1993) This situation allowed a range of measures tobe put in place immediately once the disease wasreported to the Official Service From the outset theobjective of the control regime was to contain anderadicate the disease On suspect farms and farmssharing coastal waters with confirmed and suspectfarms these measures included

Preventing the movements of live fish other than inexceptional circumstances when the risk of allowing amovement was less than the risk of fish staying insitu

Placing conditions on the movements of dead fishpeople and materials liable to transmit the disease tominimize the risk of transmission and

Fallowing and disinfecting the farm to break thecycle of infection

In addition complete and immediate depopulation ofall fish from confirmed farms was required Such fishcould be harvested for sale If the fish provedunsuitable for harvest or showed clinical signs ofdisease they were destroyed

Despite this legislative framework furtherdevelopment of the decisionmaking process wasrequired to manage the outbreak A rational basis fordefining the area of the coastal zone affected by thedisease was required as well as a regime todetermine the fallowing period necessary within thiscoastal zone

Determination of the Area ofthe Coastal Zone Affected

Farms near confirmed or suspect farms were at riskof contagion through local spread (Murray thisvolume) but the zone concept as it related to coastalmarine waters was poorly developed when the ISAepidemic occurred There was little information aboutthe mechanisms of ISA transport by passivedispersion Hydrographic data on individual siteswere not easily available nor was there sufficienttime to survey sites or model dispersal patternsaround individual farms It was rapidly realized thatthese constraints and a complex coastline meant thata simple and robust method would be needed toidentify adjacent farms at risk using existing sitelocations and available generic hydrographicinformation

During the epidemic some basic informationwas available on the mechanism underlyinghorizontal transmission of ISA Some researchers

114


had drawn the conclusion (Nylund et al 1994) thatpassive transmission through seawater would benegligible because of the potential dilution Thisargument was not supported by information onshedding rates dilution and the viral dose requiredto infect susceptible fish It is now understood thatthe infectious dose for ISA is low (Raynard et al2001b) and that local diffusive processes explain thespread within some sea lochs in Scotland (Murraythis volume)

To establish control and surveillance zones itwas necessary to determine the distance over whichinfection from a suspect or confirmed farm posed arisk to adjacent farms (Turrell et al 1998) Thehydrographic information used for such purposes wasthe maximum spring tide current obtained from a tidalatlas (Lee 1981) This information was translated intotidal excursions (Xt in meters) around each fish farmfrom the formula UTπ where U = tidal currentamplitude (ms-1) T = tidal period (1242 h or 44712sec) and π is a constant (31416) Most salmonfarms are located in relatively sheltered locationsand therefore the maximum tidal amplitudes wererestricted to 051 ms-1 (1 knot) in mainland Scotlandand 0255 ms-1 (05 knot) in Shetland giving tidalexcursions of 73 and 36 km respectively

The tidal excursion defines the region around afarm that is potentially infected on the basis of tidaltransport alone When tidal excursions from adjacentfarms overlap there is a risk that infection may betransmitted from one farm to another with contiguoustidal excursions Transmission from one farm toanother under this model will be broken only whenthere is a gap large enough to ensure that adjacenttidal excursions do not overlap (ie when farms areseparated by a distance greater than 2 tidalexcursions) The area enclosed by contiguous tidalexcursions formed the surveillance zone aroundfarms Extra surveillance by the Official Service andfallowing requirements were necessary in this areabecause of the perceived risk of infection becomingestablished All farms bounded by the tidal excursionof a confirmed or suspect farm were also required tocoordinate their fallowing to minimize the likelihood of

recurrence An example of the tidal excursion modelapplied to define zones around suspect farms isgiven for a typical Scottish sea loch in figure 1

The tidal excursion model accounts forwaterborne tidal infection only and does not considerresidual flows Waterborne infection includes viralparticles in the water column or bound to particulatematerial (including blood) or sea lice but cannotexplain transmission by free-swimming fish Thevirus has been detected in wild sea trout (Salmotrutta) (Raynard et al 2001a) saithe (Pollachiusvirens) from infected farms (Snow et al 2002) andescaped Atlantic salmon (Raynard et al 2001a)Although some of these species migrate over largedistances transmission by wild or escaped fish

Figure 1mdashAn example of the type of surveillance and controlzones established during the course of the Scottish ISA epidemic

West Loch Roag

Fish FarmsControl ZonesSurveillance Zone

0 5 10-km

Crown copy 2000

115


vectors was not considered to contribute significantlycompared with anthropogenic spread in the Scottishepidemic (Murray this volume) The output of thetidal excursion model also compares favorably withthe results of a Norwegian case control study (Jarpand Karlsen 1997) This study established aninflection in regressions of probability of infectionagainst distance at 5 km from an infected farmAnalysis of all RTndashPCR data reporting ISAv infectionin salmon farms following the Scottish outbreak(Stagg et al 2001) also showed that the majority ofRTndashPCR signals came from within the surveillancezones of the tidal excursion model established tocontain the disease It is possible that the tidalexcursion model may not apply to other coastalareas If so individual assessment of its usefulnesswithin different tidal regimes would be required Themodel as applied to managing the outbreak of ISA inScotland however contributed greatly to thesuccessful control of the epidemic It also formed acentral part of the risk management to reduce thelikelihood of recurrence of ISA in Scotland byproviding a rational basis for the subdivision of thecoastal zone in which area management agreementsand codes of practice (Joint GovernmentIndustryWorking Group on ISA 2000a) can be implemented

Rationale and RiskAssessment for RestockingFollowing the Outbreak

Key decisions in the management of an infectiousdisease in a commercial animal-rearing system arethe measures taken to prevent residual infection thatcould trigger a further epidemic following restockingISA affects salmon in the marine phase or inhatcheries using seawater There is some evidencethat ISAv occurs in freshwater but it is sporadic andat a low level (Raynard et al 2001a Stagg et al2001) Thus salmon smolts introduced to a new sea-cage site will be susceptible and any residualinfection with ISA may trigger a new outbreak

Management of residual infection in sea cagesrequires the cleaning and disinfection of all structuresthat may have been in contact with fish orcontaminated equipment in addition to fallowing thesite Cleaning and disinfection requires dismantlingthe cages and transporting them to shore for cleaningand disinfection Nets usually need to be transportedto a net-washing station in a skip or other containerto prevent leakage

Following the Scottish epidemic concretebarges and other large structures were cleaned tothe water line disinfected and fallowed onsiteFarmers were required to clean and disinfect by(1) scraping and brushing to remove all gross foulingand organic matter (2) cleaning and hot washing toremove remaining particulate matter fats and oilsthat were likely to bind the virus and (3) a final washwith a proprietary disinfectant Major considerationsin the selection of disinfectants were efficacy andimpact on the environment

Fallowing of a site minimizes the recurrence ofdisease from environmental reservoirs (sedimentwater fish and shellfish) Specific benefits offallowing have been shown for control of sea liceinfestations (Bron et al 1993 Grant and Treasurer1993) but results were somewhat equivocal formicroparasites (Jarp et al 1995 Wheatley et al1995) Nevertheless there is fair agreement amongindustry and regulatory experts on the benefits offallowing and year-class separation (Stewart 1998)The practice of fallowing marine cage sites was amajor factor in reducing the incidence of ISA inNorway from its peak of 101 cases in 1990 tosomewhere between 2 and 15 cases per yearbetween 1994 and 1998 (Haringstein 1997 Haringstein et al1999) The Norwegian experience showed that a 6-month fallow period was sufficient to preventrecurrence of ISA following an outbreak In theScottish epidemic however fallowing strategies alsoneeded to be developed for farms containing fishsuspected of being infected and farms within thecontrol and surveillance zones A precautionaryapproach was initially adopted based on the

116


Norwegian experience (table 1) Some suspectfarms proceeded to harvest with little additionalevidence of infection despite regular monthlysurveillance by the Official Service Fish on farmswhere this occurred typically gave positive results forIFAT and RTndashPCR assays but did not show clinicalsigns of ISA A risk-based approach to fallowingsuspect sites was therefore developed based upontwo assessments the risk of a farmrsquos carryinginfection at the time of harvest and the risk oftransmission to an adjacent farm

The infection risk was assessed by a review ofthe surveillance results for a particular farm in the 4months immediately before harvest Those farmswith high risk were those that in the last 4 monthseither had evidence of infection sufficient todesignate the farm suspect or were within the tidalexcursion of a farm that had been declared to besuspect Farms containing fish with a lower thresholdof infection with ISA were categorized as being atmedium risk and those without any subsequentevidence of infection were designated low risk Amonthly surveillance program that tested moribundfish was used to assess the infection status and wasconsidered more reliable than a single randomsampling at harvest

The transmission risk was assessed as lowmedium or high according to the likelihood of

Table 1mdashFallowing regimes used to prevent recurrenceof infectious salmon anemia following the 1998epidemic in Scotland

Fallow periodInitial fallow after riskperiod May 1998ndash assessment after

Category of farm December 1999 December 1999

Confirmed 6 months 6 monthsSuspect 6 months 3ndash6 months based

on risk assessment

Uninfected farm in 3 months 6 weekscontrol zone synchronously with synchronously with

confirmed or confirmed orsuspect farms suspect farms

Uninfected farm in 6 weeks 6 weekssurveillance zone

transmission from the suspect farm to others in thevicinity The level of risk was determined byconsideration of

Physical factors such as the local hydrography andtopography influencing the separation of sites

Biological factors such as evidence of thetransmission of infection to wild fish and

Industry factors such as the density of farms theuse of common facilities such as shore bases andother evidence of site-to-site interaction

Infection and transmission factors werecombined to provide a fallowing period that varied ona sliding scale from 3 to 6 months (fig 2)

Removal of the InfectionSource

In Scotland legislation required immediatedepopulation of fish from farms in which the diseasehad been confirmed In reality this wasaccomplished within the shortest practical time afterconsideration of the logistics required and thetonnage of fish on a site Farmers were allowed toharvest fish from sites if they did not show clinicalsigns of disease Fish that were either unsuitable forharvesting or had clinical signs of ISA were culledand the virus was inactivated by ensiling in formicacid to a pH less than 40

During the epidemic flexible schemes werediscussed based on the Norwegian experience ofdepopulating only individual cages on infected farmsif the mortality attributed to ISA approached 005percent per day (Hastein 1997) This approach hadbeen successful in reducing the prevalence of ISA inNorway but failed to eliminate new outbreaks TheControl Directive (European Commission 1993) wastherefore modified (European Commission 2000) toallow phased depopulation and the use of vaccinationprograms to help eradicate the disease

The average time for depopulation of infectedfarms in the Scottish epidemic was 21 days Thisrapid rate of depopulation is likely to have been an

117


important factor in the successful eradication of ISAfrom Scottish farms compared with other countrieswhere cages not suffering mortality or clinical diseasehave remained until harvest and epidemics havepersisted

The Scottish Epidemic

In Scotland the first case of ISA was reported inLoch Nevis in May 1998 (Rodger et al 1998) andanother 10 cases were subsequently confirmed(Stagg et al 2001) extending over the whole ofScotland including the Shetland Isles (fig 3)Twenty-five farms contained fish suspected of beinginfected but the disease did not progress sufficientlyfor it to be confirmed The incidence of new cases(farms) peaked in April 1998 (fig 4) and the lastcase was confirmed in May 1999 1 year after theinitial outbreak Suspect cases were investigateduntil November 1999 and scientific experts haveindependently drawn the conclusion that clinical

cases of ISA have been eradicated from Scottishsalmon farms (Royal Society of Edinburgh 2002)

Investigations revealed that the epidemic arosefrom a single primary site located in Loch Nevis onthe Scottish west coast (Stagg et al 2001) This wasbased on several lines of evidence

(a) The pattern of spread of the disease wasdiscontinuous and intervening unaffected areassubsequently became infected (Murray thisvolume)

(b) Virus isolates obtained from farms with clinicaldiseases (5 out of 11 confirmed farms in theepidemic) were identical in terms of the observedsequence for RNA segments 2 8 and 6(Cunningham and Snow this volume)

(c) All confirmed sites and most of the suspectsites had significant epidemiologic connections tothe primary site at Loch Nevis (Stagg et al 2001)Two of the subsequently affected areas (LochCreran and Loch Snizort on Skye) had received livefish from the primary site Another affected site in

High IR

High TR

Medium IR

Medium TR

Suspect farm

Infection Risk Assessment (IR)

Transmission Risk Assessment (TR)

Low IR

Low TR

3 months6 months 6 months 5 months 4 months

Figure 2mdashDecision tree for fallowing a salmon farm containing fishsuspected to be infected with ISA

118


Figure 3mdashLocation of ISA suspect and confirmed farms inScotland 1998ndash99

119


10

91998 1999

Confirmed

Incidence of ISA cases (farms) in Scotland January 1998 - May 2000

Suspect

2000

8

7

6

5

4

3

2

1

0

Jan Mar May Jul Sep Nov Mar MarMay Jul Sep NovJan Jan

Date (month)

Figure 4mdashIncidence of ISA cases among salmon farms inScotland from January 1998 until March 2000

Shetland had been stocked with smolts from aremote site on the Western Isles but these smoltshad been delivered by transit through the site inLoch Snizort

(d) Site-to-site movements of wellboats played asignificant role in the spread of ISA that was mainlyassociated with trips to harvest (Murray et al 2002)The spread from the primary site was alsoassociated with equipment transfer for grading orharvest of fish sharing divers and their equipmentmixing clean with dirty nets at net-washing stationsand transport of dead fish (mortalities) for disposal(Stagg et al 2001) A most significantepidemiologic factor associated with the broadcastof ISA away from the primary site was the linkage toharvesting via wellboats through a harvest stationHarvest stations are used for the temporary holdingof fish before slaughter These stations are sea

pens anchored next to a slaughterhouse andprocessing plant The creation of such harvestingstations in combination with the trafficking ofwellboats between rearing sites and the harveststation itself were identified as the highest riskfactors associated with the transmission of ISA overdistant geographic regions (Munro et al 2003)

(e) Other marine sites that had been stocked withfish from the same freshwater populations used tostock the Loch Nevis sites did not simultaneouslyhave outbreaks of disease Although some (not all)did subsequently show evidence of the disease themost likely cause was established contact with theprimary occurrence in Loch Nevis In additionsiblings and contacts still in freshwater that weretested did not show any evidence of infection

These findings suggest that the epidemic aroseat the primary site and was disseminated by the

120


actions of fish farmers However it was also evidentthat an epidemiologic link to a confirmed farm was noguarantee that the fish in a recipient farm woulddevelop clinical ISA When this occurred thereproductive ratio (R0) must have been greater thanunity (Anderson and May 1991) The size of R0 willbe dependent on the dose of ISAv required whetherconditions on a farm promote transmission of thedisease and the numbers of infected and susceptiblefish on the farm The incubation of the disease inrelation to the time from the introduction of infectionto the time to harvest was also critical in whetherdisease manifested On two farms where theintroduction of infection was actually determined theincubation time was about 3 months (Stagg et al2001) This figure compares with an incubation timeof around 21 days when Atlantic salmon wereexperimentally challenged with ISAv (Raynard et al2001) In the Scottish epidemic 19 of 25 suspectcases had clear evidence of infection and significantepidemiologic connections to the primary site (Stagget al 2001) but their fish did not develop clinicaldisease before harvest Fish from a farm in Shetlanddid not develop clinical disease despite isolation ofthe virus that was genotypically identical (segments2 6 and 8) to that isolated at the primary site

An alternative explanation for the pattern of ISAoutbreaks in Scotland is that the disease arosesporadically from a background wild source Surveyscarried out in Scotland showed that ISAv can bedetected sporadically in wild fish by RTndashPCR(Raynard et al 2001a)mdashoften in areas remote fromthe infected farms The virus was also detected byRTndashPCR in fish from some farms outside thesurveillance zones established to combat the disease(Stagg et al 2001) and five of the suspect farms didnot have any evident connection to the primaryoutbreak However it is clear that these observationsare insufficient to explain the strong linkageestablished between infected sites adequately

Origins of the ISA Epidemic

It has not been possible to identify the source of ISAvthat triggered the epidemic in Scotland definitivelyAssessment of the current evidence indicates that thevirus was either imported from infected areas inNorway by traffic of contaminated wellboats oremerged from a benign infection among wild fish(Stagg et al 2001) Since 1999 ISAv has beendetected by RTndashPCR in both farmed and wild fish Infarms the monthly mean prevalence of infection hasdeclined dramatically over the course of theepidemic Although ISAv should have disappearedaccording to the trend in these data it was still foundon isolated occasions (Stagg et al 2001) ISAv wasalso occasionally detected among fishes by RTndashPCRbeyond ISA control zones during the epidemic andamong wild salmonids throughout Scotland (Raynardet al 2001a) In none of these cases was ISAvassociated with evidence of disease This factsuggests that ISAv might exist within a widelydistributed wild host having the potential to transmitthe virus to more susceptible hosts if conditions aresuitable This supposition implies a continual albeitrare risk of reemergence The primary site of thefirst occurrence of ISA in Scotland was characterizedby high stocking densities multiple seawater-to-seawater movements of live fish continuousstocking and multiple generations (fig 5) Theseconditions give maximum opportunity for theemergence of a virulent pathogen

121


Figure 5mdashStocking regime at farms in Loch Nevis before theoutbreak of ISA Site 9801 was the primary source of theoutbreak

Implications for FutureManagement

The experience of the ISA epidemic in Scotlandhad some unforeseen consequences Industry andgovernment established a working group to addressthe risks posed by ISA (Joint GovernmentIndustryWorking Group 2000b) The practical consequenceof this work was the development of an industry codeof practice (Joint GovernmentIndustry WorkingGroup 2000a) to prevent emergence of newoutbreaks and their spread to other farms The codeof practice covers fish and equipment movements

area management by the tidal excursion modelharvest and processing practices and the disposal ofcarcasses and waste Although principally designedto prevent the spread of disease the code of practiceprovides guidance on the adoption of single-year-class strategies and fallowing Applied tomanagement zones these measures also precludeemergence of ISA

The development of ISAv vaccines (Brown et al2001) may provide another method for the controland prevention of ISA In many ways vaccination ispreferred to the current practice of depopulation offarms because of the evident economic benefits

J F M A M J J A S O N D J F M A M J J A1995 1996 1997

SITE ISA 9803

SITE ISA 9801

SITE ISA 9808

1998

1995 1996 1997 1998

Smolt unit ESmolt unit FSmolt unit G

BroodfishHatchery A

1998

S O N D J F M A M J J A S O N D J F M A M J J A S O N D

J F M A M J J A S O N DS O N D J F M A M J J A S O N D

N D

J F M A M J J A S O N D

CullSmoltsISA 9802

Cull

Harvest

Harvest

Harvest

SA 9805 SA 9805

Cull

J F M A M J J A S DO N J F M A M J J A S O N D

FALLOW

FALLOW

FALLOW

Harvest

J F M A M J J A DS O N

FALLOW

Smolt unit ASmolt unit BSmolt unit C

Sea water site BSmolt unit ASmolt unit BSmolt unit DSmolt unit E

BroodfishSea water site A



Smolt unit ASmolt unit B

GrowersISA 9816

GrowersISA 9801

ISA 9806

GrowersISA 9803

GrowersISA 9801

BroodfishSea water

site A

GrowersISA 98031994

ISA 9805

1997

MortalitiesLandfillOban

122


However a vaccination strategy should be pursuedcautiously Killed-cell vaccines may interfere withcurrent nonculture-based diagnostic methods andcomplicate differentiation between vaccinated andinfected populations Vaccine research needs toestablish that vaccines prevent carrier status as wellas clinical disease and can confer protection againstthe wide range of strains of ISA identified thus far(Cunningham and Snow this volume) Regulatorsand farmers need to be confident that thedevelopment of vaccines against one strain of ISAwill not merely increase the evolutionary pressure forthe emergence of new strains resistant to thevaccine

Acknowledgments

The management of the ISA outbreak in Scotlandcould not have been achieved without the enormoussupport of all of the staff at Fisheries ResearchServices who worked tirelessly to deal with thedisease and its consequences Special thanks go tothe Fish Health Inspectorate and the staff in thediagnostic laboratories

References Cited

Anderson R M May R M 1991 Infectiousdiseases of humans dynamics and control OxfordUK Oxford University Press 757 p


Branson E 2002 Sleeping disease in troutVeterinary Record 150 759ndash768

Bron J E Sommerville C Wootten R Rae G H1993 Fallowing of marine Atlantic salmon Salmosalar L farms as a method for control of sea liceLepeophtheirus salmonis Journal of Fish Diseases16 487ndash493

Brown L L Sperker S A Clouthier S ThorntonJ C 2001 Development of a vaccine againstinfectious salmon anemia virus (ISAV) Bulletin of theAquaculture Association of Canada 100 4ndash7

Christie K E Hjeltnes B Uglenes I Winton J R1993 Determination of buoyant density andsensitivity to chloroform and freon for the etiologicalagent of infectious salmon anemia Diseases ofAquatic Organisms 15 225ndash228

Commission of the European Communities 1991Council Directive 9167EC concerning the animalhealth conditions governing the placing on the marketof aquaculture animals and products Official Journalof the European Communities 46 1ndash18

Commission of the European Communities 1993Council Directive 9353EEC introducing theminimum Community measures for the control ofcertain fish diseases Official Journal of the EuropeanCommunities 157 23ndash31

Commission of the European Communities 2002aInformation received from the Faeroe Islandsconcerning outbreaks of infectious salmon anemia(ISA) Brussels Commission of the EuropeanCommunities [pagination unknown]

Commission of the European Communities 2002bInformation received from the United Statesconcerning outbreaks of infectious salmon anemia(ISA) Brussels Commission of the EuropeanCommunities [pagination unknown]

Dannevig B H Falk K Namork E 1995 Isolationof the causal virus of infectious salmon anemia (ISA)in a long-term cell line from Atlantic salmon headkidney Journal of General Virology 76 1353ndash1359

Devold M Falk K Dale B Krossoslashy B Biering EAspehaug V Nilsen F Nylund A 2001 Strainvariation based on the hemagglutinin gene inNorwegian ISA virus isolates collected from 1987 to2001 indications of recombination Diseases ofAquatic Organisms 47 119ndash128

123


Falk K Namork E Rimstad E Mjaaland SDannevig B H 1997 Characterization of infectioussalmon anemia virus an orthomyxo-like virusisolated from Atlantic salmon (Salmo salar L)Journal of Virology 71 9016ndash9023

Grant A N Treasurer T I 1993 The effects offallowing on caligid infestations on farmed Atlanticsalmon (Salmo salar L) in Scotland In Boxshall GA Defaye D eds Pathogens of wild and farmedfish sealice New York Ellis Horwood Ltd 255ndash260

Haringstein T 1997 Infectious salmon anemia (ISA) Ahistorical and epidemiological review of thedevelopment and spread of the disease in Norwegianfish farms In Proceedings of a workshop oninfectious salmon anemia 26 November 1997 StAndrews NB [Place of publication and publisherunknown] 5ndash14

Haringstein T Hill B J Winton J R 1999 Successfulaquatic animal disease emergency programmesReview of Scientific Technology 18 214ndash227

Hill B J 1996 National legislation in Great Britain forthe control of fish diseases Review of ScientificTechnology 15 633ndash645

Jarp J Gjevre A G Olsen A B Bruheim T 1995Risk factors for furunculosis infectious pancreaticnecrosis and mortality in post-smolt of Atlanticsalmon Salmo salar L Journal of Fish Diseases 1867ndash68

Jarp J Karlsen E 1997 Infectious salmon anemia(ISA) risk factors in sea-cultured Atlantic salmonSalmo salar Diseases of Aquatic Organisms 2879ndash86

Joint GovernmentIndustry Working Group on ISA2000a A code of practice to avoid and minimise theimpact of infectious salmon anemia (ISA) AberdeenUK Fisheries Research Services 16 p

Joint GovernmentIndustry Working Group on ISA2000b Final report of the joint governmentindustryworking group on infectious salmon anemia (ISA)Edinburgh UK The Scottish Executive 136 p

Lee A J 1981 Atlas of the seas around the BritishIsles Lowestoft UK Ministry of Agriculture Fisheriesand Food (MAFF) 92 p

Lovely J E Dannevig B H Falk K Hutchin LMacKinnon A M Melville K J Rimstad EGriffiths S G 1999 First identification of infectioussalmon anemia virus in North America withhaemorrhagic kidney syndrome Diseases of AquaticOrganisms 35 145-148


Munro ALS 1996 Report on the first recordedoutbreak of viral haemorrhagic septicaemia (VHS) inGB and subsequent actions to contain eradicate andinvestigate the origins of the infection ScottishAquacul Res Rep 3 [Place of publication andpublisher unknown] 12 p

Munro P D Murray A G Fraser D I Peeler E J2003 An evaluation of the relative risks of infectioussalmon anemia transmission associated with differentsalmon harvesting methods in Scotland Oceans andCoastal Management 46 157ndash174


Nylund A Hovland T Hodneland K Nilsen FLovik P 1994 Mechanisms for transmission ofinfectious salmon anemia (ISA) Diseases of AquaticOrganisms 19 95ndash100

Nylund A Hovland T Watanabe K Endersen C1995 Presence of infectious salmon anemia virus(ISAV) in tissues of Atlantic salmon Salmo salar Lcollected during three separate outbreaks of thedisease Journal of Fish Diseases 18 135ndash145

Raynard R S Murray A G Gregory A 2001Infectious salmon anemia virus in wild fish fromScotland Diseases of Aquatic Organisms46 93ndash100

124


Raynard R S Snow M Bruno D W 2001Experimental infection models and susceptibility ofAtlantic salmon Salmo salar to a Scottish isolate ofinfectious salmon anemia virus Diseases of AquaticOrganisms 47 169ndash174

Rimstad E Mjaaland S 2002 Infectious salmonanemia virus An orthomyxovirus causing anemerging infection in Atlantic salmon ActaPathologica Microbiologica et ImmunologicaScandinavica 110 273ndash282



Royal Society of Edinburgh 2002 The scientificissues surrounding the control of infectious salmonanaemia (ISA) in Scotland A report of the RoyalSociety of Edinburgh Working Party on InfectiousSalmon Anemia Edinburgh The Royal Society ofEdinburgh 13 p

Snow M Raynard R Bruno D W van NieuwstadtA P Olesen N J Levold T Wallace C 2002Investigation into the susceptibility of saithe(Pollachius virens) to infectious salmon anemia virus(ISAV) Potential role as a vector for viraltransmission Diseases of Aquatic Organisms 5013ndash18

Stagg R M Bruno D W Cunningham C ORaynard R S Munro P D Murray A G AllanCET Smail D A McVicar A H Hastings T S2001 Epizootiological investigations into an outbreakof infectious salmon anemia (ISA) in Scotland FRSMarine Lab Rep 1301 Aberdeen UK FisheriesResearch Services 60 p

Stewart J E 1998 Sharing the waters an evaluationof site fallowing year class separation and distancesbetween sites for health purposes on Atlantic salmonfarms Canadian Technical Report of Fisheries andAquatic Sciences 2218 1ndash56

Thorud K Djupvik H O 1988 Infectious anaemia inAtlantic salmon (Salmo salar L) Bulletin of theEuropean Association of Fish Pathologists 8109ndash111

Totland G E Hjeltnes B Flood P R 1996Transmission of infectious salmon anemia (ISA)through natural secretions and excretions frominfected smolts of Atlantic salmon Salmo salar duringtheir pre-symptomatic phase Diseases of AquaticOrganisms 26 25ndash31

Turrell W R Gillibrand P A Brown G 1998Second interim report of the Joint IndustryGovernment working group on infectious salmonanaemia ISA hydrographically defined managementareas for the Scottish aquaculture industryAberdeen UK Fisheries Research Services 32 p

Webster R G 1998 Influenza an emergingmicrobial pathogen In Krause RM ed Emerginginfections London Academic Press 275ndash300

Wheatley S B McLoughlin M F Menzies F DGoodall E A 1995 Site management factorsinfluencing mortality rates in Atlantic salmon (Salmosalar L) during marine production Aquaculture 136195ndash207

125

Practical Grower Experience in the Proactive Prevention andControl of Infectious Salmon Anemia

Practical Grower Experience in theProactive Prevention and Control ofInfectious Salmon Anemia

Sebastian M Belle1

1Mr Belle is the executive director of the Maine AquacultureAssociation

Introduction

Infectious salmon anemia (ISA) was first detected inNorway in 1984 (Thorud and Djubvik 1988 Hovlandet al 1994 Hastein 1997) Subsequently thedisease has been detected in the Canadianprovinces of New Brunswick (1986) and Nova Scotia(1998) (Bouchard et al 1999 Lovely et al 1999)Scotland (Rodger et al 1998 Rowley et al 1999)Chile (Kibenge et al 2001) and the Faroe Islands in1999 (Anonymous 2000) Some evidence suggeststhat ISA virus (ISAv) has been present in NorthAmerica for some time predating actual ISAoutbreaks (Krossoslashy et al 2001)

Clinical ISA was first reported in Maine in March2001 The Maine Aquaculture Association (MAA)and its member growers have worked closely withState and Federal authorities and fish healthprofessionals to ensure that grower practices andresponses are consistent with the best availablescientific guidance and in compliance with State andFederal environmental and fish health regulationsBeginning in 1997 Maine salmon growers developedand implemented a series of proactive and reactiveresponses to the threat of ISA These actions werebased on a philosophy of avoidance of disease

Abstract Clinical infectious salmon anemia (ISA) was firstreported in Maine waters in 2001 Maine salmon growershave developed and implemented both proactive andreactive responses to it Annual third-party biosecurityaudits were initiated in 1997 These audits proved to be avaluable learning tool for farm managers and greatlyassisted in the identification of higher risk farm practicesIn 1998 an ISA action plan was developed and imple-mented by the growers with the help of a number of fishhealth professionals This plan was revised annually basedon the evolving state of knowledge about ISA epidemiologyimproving ISAv detection methodologies and experience inISA management from other salmon-producing areasfacing ISA outbreaks In 2001 the Maine AquacultureAssociation (MAA) developed and implemented a coopera-tive bay management program designed to increase growercoordination and reduce risks associated with fish health

and environmental impact The MAA Bay ManagementAgreement was based on the most recent MAA ISA ActionPlan and systematic review of cooperative bay manage-ment plans and biosecurity protocols from around theworld The Agreement establishes minimum standardsprocedures and protocols designed to minimize diseaserisk and improve grower communications and cooperationMAA has worked closely with State and Federal authoritiesto ensure that grower practices and responses are consis-tent with the best available scientific guidance in aquaticdisease prevention and response The US Department ofAgriculturersquos (USDA) ISA Program has been critical to therapid and aggressive response to the emergence of clinicalISA in US waters Continued close cooperation betweenFederal and State authorities and the growers will signifi-cantly increase the effectiveness of any ISAv control anderadication efforts

through sound fish health and husbandry practicesand on the guiding principles for management ofinfectious diseasesmdashrisk reduction detectioncontainment and control through selectiveeradication

Grower initiatives have taken a number offorms (1) systematic reviews and revisions ofhusbandry practices and farm operations to reducefish stress (2) increased fish health surveillance andtesting (3) ISA vaccine trials (4) third-partybiosecurity audits (5) fish health action plans(6) cooperative bay management agreements and(7) assistance in the development of the ISA programand program standards by the US Department ofAgriculture (USDA) This paper will discuss onlyinitiatives 4ndash7

Biosecurity Audits

An industrywide program of annual third-partybiosecurity audits was initiated in 1997 Biosecurityaudits are proactive tools that use disease risk-analysis techniques to systematically identifycharacterize and measure sources and levels of riskThe audit characterizes by qualitative andquantitative examinations the reality of how acompany and its workers implement biosecurityprotocols not theoretical plans or company policiesAuditors specifically attempt to identify weaknesses

126


in site-specific biosecurity practices The audit formatand methods have undergone a number of revisionssince their inception Audits are conducted bylicensed and accredited veterinarians who areindependent subcontractors working for the MaineDepartment of Marine Resources or USDA Auditorsconduct unannounced visits to marine farmshatcheries processing plants and transportationboats Auditors are given uninhibited access to makedirect observations and interview employees todetermine the level of relative risk associated withoperational practices facility design and layoutEach specific area examined (up to 121 differentareas depending on the type of facility or vessel) isclassified as nolow low moderate or high riskAudit areas that are seen as particularly affectingdisease risk (35ndash45 factors) are given additionalweight during the audit analysis Facility-specificaudit reports are sent to the individual farmer and anannual comprehensive report which characterizesindustrywide risks and allows comparisons betweenfarms is sent to both the Department of MarineResources and the growersrsquo association

Biosecurity audits have proved to be a valuablelearning tool because they directly identify specificfarm practices that increase disease risk and includerecommendations on how to improve operations andreduce risks The power of the biosecurity audit as adisease management tool stems from its systematicapproach utilization of external auditorsrepeatability and direct feedback to farm managersSubsequent audits allow both regulators and growersto determine if operational biosecurity is improvingstaying the same or decreasing In generalvariances in performance between farms havedecreased over time and audit results have improvedsignificantly since the programrsquos inception (Allen andOpitz 1999 Merrill 2000 and 2001) Whilebiosecurity protocols and audits can be very effectivepreventative tools management intensity and focusare critical to creating and maintaining a climate ofconstant vigilance and prevention Effective riskcontrol mandates that biosecurity becomes aningrained set of operational habits that are constantlyreassessed and improved

Action Plans

In 1998 with the help of a number of fish healthprofessionals an ISA action plan (ISAAP) wasdeveloped and implemented by the growers (MaineAquaculture Association 1998) The ISAAP is adynamic document designed to be periodicallyrevised to incorporate the evolving knowledge aboutISA epidemiology improved ISAv detectionmethodologies and increased level of experience inISA management Formulation of the ISAAP servedas a catalyst for increasing grower cooperation andgenerating support for USDA involvement in aquaticanimal health management In order to come tocritical consensus growers learned to thinkcooperatively and develop mechanisms for actingcooperatively as they were forced to review their ownand each otherrsquos operational practices and examinehow they affected disease risks This newcooperative approach was instrumental in thedevelopment of grower initiatives for comprehensivebay management planning Thus the ISAAP formedmuch of the basis for the MAA Finfish BayManagement Agreement (FBMA) that was signed inJanuary 2002 (Maine Aquaculture Association 2002)

The ISAAP outlined a series of strategicrecommendations from the Maine salmon farmingindustry for dealing with containment and control ofISA and ISAv and documented the industryrsquosapproach to rational management of ISA throughhusbandry and biosecurity practices Some of thespecific biosecurity protocols and operationalstandards recommended in the ISAAP included

Developing veterinaryndashclient relationships for allfarms and facilities to achieve regular onsite fishhealth evaluations whose frequency would increaseshould unusual or sustained mortality occur

Initiating specific sampling protocols with increasedsampling in an ldquoISA Increased Risk Zonerdquo

Initiating standardized third-party biosecurity auditsat all farms and transport and processing facilities

Imposing severe restrictions on fish movement fromsites with confirmed ISA-positive status

127


Developing single year-class stocking andproduction strategies

Initiating sea-lice treatment programs in accord withexisting State of Maine programs

Implementing guidelines on the containmentdisinfection and disposal of all forms of wasteassociated with farm mortality and harvesting andprocessing operations

Implementing strict site equipment and personneldisinfection guidelines and specific recommendationson intersite boat equipment and personnel trafficand

Promoting specific recommendations for priorityresearch to answer critical questions about ISAmanagement

Although the ISAAP contained a number ofspecific recommendations for operational standardsand biosecurity measures it was also a policydocument In particular the ISAAP articulated theindustry position on what constituted a positive testresult and how those results should be used in theclassification of the disease status of an individualfish cagetank andor farm site The plan furtherclarified the growersrsquo perspective on these definitionsand how they related to reporting requirements Theplan highlighted the importance of indemnificationand its linkage to effective disease managementFinally the ISAAP proved to be a very effective tool inlobbying for USDA involvement in addressing the ISAissue

Cooperative Bay ManagementAgreement

To date ISA has not been eliminated in any of thecountries that have experienced clinical outbreaksGiven the many factors involved in diseasetransmission elimination of ISA may not be anattainable goal Instead loss reduction and riskminimization through effective managementtechniques such as stringent surveillance andaggressive biosecurity protocols may be morerealistic

In 2001 the MAA developed a Cooperative BayManagement Program designed to increase growercoordination and reduce risks to fish health and theenvironment associated with intensive fishproduction The MAA Finfish Bay ManagementAgreement was based on the most recent MAAISAAP and a systematic review of cooperative baymanagement plans and biosecurity protocols fromaround the world Implementation of the agreementbegan in January 2002

The agreement is an overarching legallybinding document signed by all private salmonfarmers in the State of Maine The agreementestablishes minimum standards procedures andprotocols that are designed to lessen disease riskand improve grower communications andcooperation Signatories to the agreement areobligated to establish Local Bay Management Groupsfor all eight local bay management areas in MaineThese groups of farmers are in turn obligated todevelop a local bay area management plan which isbased on the minimum standards and guidelinesestablished in the overarching agreement and whichis then reviewed and approved by the agreementsignatories Local plans may establish more stringentstandards and protocols than those contained in theagreement

The agreement includes a series of technicalappendices and a set of specific definitions for termsto be used in both the agreement and local plansThe agreement also contains a series of legal termsand conditions that refer to meeting frequencyamending procedures arbitration governing law andthe relationship between the signatories The term ofthe agreement is linked to production cycles insaltwater farms and is automatically renewedSignatories to the Agreement commit to comply withall relevant State and Federal regulations and to allcollective industry standards The agreement is notintended to coordinate production or businessstrategies for the purpose of price or marketinfluence and great care was taken to ensure theagreement did not violate antitrust laws

128


The Agreement contains a generic baymanagement template for use when developing localplans This template establishes the minimumstandards and protocols that local plans must includeand begins with a common mission statement

As marine farmers we have a strongvested interest in healthy marineecosystems and a clean marineenvironment Our mission is toachieve long-term viability andcompetitiveness in the MaineSalmon Farming Industry with acontinued commitment toenvironmental sustainability andstewardship We seek to promoteresponsible development andmanagement of the Maine SalmonFarming Industry in order to assurethe production of high quality foodwhile respecting environmentalconsiderations and consumerdemands (Maine AquacultureAssociation 2002)

The local plan template goes on to identify thespecific geographic coordinates and individual farmoccupants of each local bay management area Along-term stocking and production target for the localarea is established and any constraints inhibiting itsattainment are identified Specific actions required toovercome these constraints are identified and atimetable for achieving the long-term target isestablished Although the template includes 14 focusareas the 5 principal sections of the local plansaddress communication waste management pestmanagement fish health management andbiosecurity and disinfection

Communication

Local bay management groups must develop a BayManagement Area Communication Plan (CP) thatclearly outlines individual and collectiveresponsibilities and methods to facilitate rapid andclear communication within and between bay

management groups The CP must include methodsdesigned to reduce the risk of misinformation beingdistributed The CP establishes risk thresholds thattrigger communication among producers in eachlocal bay management area These thresholds mustrecognize that different risks engender differentpotential impacts Lower thresholds are establishedfor risks that have higher potential impacts The CPestablishes the baseline information that must becommunicated about disease risks on any individualfarm eg history character prevalence andpotential for dissemination of the identified riskspecifics of fish species affected and actions taken tocontrol and contain the risk

Waste Management

A Bay Management Area Waste ManagementPlan (WMP) must be developed This plan mustrequire all farm occupants in a bay management areato develop site-specific WMPs that are reviewed andapproved by the Bay Management Group The WMPmust clearly identify all wastes generated andclassify them with respect to any risks associatedwith their collection and appropriate disposalWhenever possible the WMP should encouragereduction reuse and recycling of waste WMPs mustaddress at a minimum human waste feed bagsscrap rope and netting fish mortalities packagingmaterials and any chemical or fuel spills The WMPis applied in concert with the specifics of the FishHealth and Biosecurity Plan (below)

Integrated Pest Management

All local Bay Management groups must developa Bay Management Area Integrated PestManagement Plan (IPMP) This plan requires alloperations in a local bay management area tocoordinate their efforts at pest monitoring and controland to use best management practices to reduce theneed for use of chemicals or medications At aminimum the IPMP includes coordinated monitoringand treatment single year-class stocking fallowingbetween year-classes pest population thresholds for

129


treatment decisions and treatment withdrawalguidelines

Fish Health and Biosecurity

All local Bay Management groups must developa Fish Health Management and Biosecurity PlanThis plan provides guidelines and protocols intendedto reduce the risk of the introduction and spread ofinfectious agents (eg ISAv) due to human activitiesThe plan requires all farms transport vessels andprocessing facilities in Maine to have regular third-party biosecurity audits Facilities that are eitherconfirmed positive for fish with ISA process fish fromsites confirmed positive for ISA or have consistentlybad audits shall be subject to biosecurity audits morefrequently Husbandry and handling protocols arealso established for all life-history stages Incoordination with the WMP the Health Managementand Biosecurity Plan establishes strict guidelines forthe collection storage and disposal of farmmortalities and blood water The plan establishesspecific protocols for appropriate use movementand disinfection of divers grading net changing andharvesting equipment Personnel and boatmovement protocols and disinfection standards arealso included Vessels are required to declare aldquohomerdquo bay management area with movementsbetween bay management areas highly restrictedand dependent on thorough disinfection Vesselsmoving between sites within a local bay managementarea must adhere to strict traffic pattern anddisinfection guidelines Finally the Fish Health andBiosecurity Plan establishes specific fallowingprocedures and plans Length of fallowing anddegree of site disinfection are determined by aspecific sitersquos health status Sites that have had anyclinical disease outbreaks or any cages depopulateddue to confirmed positive ISAv test results arerequired to fallow for longer and disinfect at higherlevels

Disinfection Protocols

Each operation in a bay management area isrequired to develop a site-specific Disinfection Plan(DP) which has to be reviewed and approved by theLocal Bay Management Group The DP specifiesmaterials methods procedures and documentationrequired in all disinfection activities of personneloperations and product addresses risk levels of allphases of cleaning disinfection and isolation andidentifies specific operational circumstances wheredisinfection is mandatory Wherever possible site-specific DPs must address all current knowntransmission and infection risks and includeprocedures that assure that subcontractorsunderstand and follow disinfection guidelines

Underlying AssumptionsThe agreement and local plans are predicated

on the assumption that disease risks and operationalpractices at one facility may affect other facilities inthe same general area Thus a collective interestcoupled with the legal obligation to work together tominimize disease risk gives the Bay ManagementAgreement its power The speed and degree ofdevelopment of local bay management plans will beone measure of the efficacy of the agreement as anadministrative and policy entity More importantly theresults of systematic statewide biosecurity audits willmeasure whether the Agreement and local plans areaffecting operational practices that increase diseaserisks Finally in the event of clinical diseaseoutbreaks the speed and degree of collectiveresponse will also be a measure of the agreementssuccess in developing improved cooperation andcoordination

USDA ISA Program

The industry is committed to responsible ISAmanagement and seeks supportive andcomplementary ISA policy development from bothFederal and State regulators in order to foster optimaldisease management Other authors at this

130


symposium will give a more complete summary ofthe USDA ISA Program My presentation will focuson the intent and implementation of the program fromthe industry perspective

In 2000 based on the landed value of farmedfinfish and shellfish Maine was the largest marineaquaculture producer in the Nation Farm gate salesfor Atlantic salmon alone exceeded $100 million TheMaine salmon farming industry accounts for morethan 1000 full-time jobs located in some of thepoorest counties in the United States ISA lossesand indirect costs to the Maine industry exceeded$20 million in 2001 and 2002

Although the Maine salmon-farming sector issignificant in the United States and Maine itrepresents less than 2 percent of world production offarmed salmon Maine salmon farmers work in ahighly competitive world market with other producingareas that have significant government programs thatsupport their salmon producers This has beenrecognized by the International Trade Commissionwhich has twice awarded Maine salmon growersjudgments finding that foreign producers enjoysignificant competitive advantages due togovernment support

The ability of the salmon-farming industry inMaine to effectively manage any disease isdependent on grower initiatives effective detectionand treatment methods sound fish health regulatorypolicy and government assistance Governmentassistance should take three formsmdashthe same formsthat are typical of US Government assistance forterrestrial farmers

Disease surveillance and monitoring includingpathogen surveys biosecurity audits and qualitycontrol of testing laboratories and detection methods

Epidemiologic analysis and research designed todevelop farming methods that reduce disease risksand

The development and implementation of anadequately funded indemnification plan tocompensate private farmers for the value of seizedanimals they are forced to depopulate under disease

control programs Indemnification should also fullycover the costs of depopulation cleaning anddisinfection

Although the USDA program contains elementsof these three important areas the Maine salmon-farming industry is concerned about the continuedcommitment of USDA to domestic salmon growersIndemnity is a key concept to encourage active earlyand sustained participation in an effective ISAmanagement program Commercial insurancepolicies of the type popular in Norway typically havenot found favor with Maine salmon farmers Wherepartial losses may be covered large deductiblesapply on a per-incident basis Importantly lossesthat are the result of the depopulation of cages eitherelectively or in compliance with regulatory agencymandates are not covered

The industry believes that an adequately fundedand professionally administered indemnity program iscritical in order to facilitate the implementation of ISABest Management Plans being implemented byindividual companies and the Bay ManagementPlans These initiatives are detailed and aggressiveattempts to improve operations and reduce the risk ofdisease

The industry has spent and continues to spendsignificant capital to implement these initiativesCurrent indemnification funding levels aresubstantially below actual industry costs Futurefunding levels appear likely to be even lower thancurrent levels

Cooperation is needed between the privatepublic and academic sectors to research developand fund better indemnity strategies Options suchas State Federal andor industry-sponsoredindemnification should all be explored fully in order toincrease the total funding available forindemnification In particular Maine State policyprioritizes the responsible development of marineaquaculture as a means of diversifying coastaleconomies (Maine Aquaculture DevelopmentCommittee 1990 Maine Department of MarineResources and Maine Coastal Program 1997)Farmed salmon is currently the second most valuable

131


seafood product landed in Maine As of September2002 the State has made no direct financialcontribution to indemnification

Where Do We Go From Here

Four critical areas deserve more attention in order toimprove the efficacy of ISA control and managementin Maine

Significantly increased biosecurity auditing

Research on the epidemiology of ISA

Increased indemnification funding and

Increased cooperation between Canadian andUnited States growers and regulators

Frequent and comprehensive third-partybiosecurity audits provide direct feedback to farmersand quickly address sources of risk and improveoperational practices The results of biosecurityaudits should be openly discussed in local baymanagement group meetings in order tocooperatively solve issues and use peer pressureand threats of peer litigation to improve operations onsubstandard farms

Systematic epidemiologic research should bedone to improve our understanding of diseasevectors and risk factors In order to be most effectivethis research must be conducted in a manner thatdoes not compromise veterinaryndashclient confidentialityResearch results should be used to review andimprove current bay management biosecurity andfish health protocols

Current indemnification levels are not adequateThe industryrsquos position remains that mandatorydepopulation orders constitute a seizure of privateproperty and that under Federal and State law thegovernment is required to compensate the owner ofsaid property The industry is committed however toexamining the possibility and potential form of a self-insurance component in a cooperativeindemnification program

In New Brunswick Canada the fish-farmingindustry sought changes to existing insurance

regulations to form a self-managed fund Currentlythis fund receives annual contributions from individualfarms paid in at the rate of 3 cents per pound ofoverall production This level is matched bygovernment funding The industry component aloneis not enough to support the current level of claimsagainst the fund In Maine a similar industry matchwould generate only $780000 Given the level ofMaine farmersrsquo losses to ISA in 2001 (approximately$20 million) and the proposed funding andcompensation levels (40 to 60 percent of industryproduction costs) for indemnification in 2003 thereappears to be a significant gap between availablefunds and industry costs The MAA and its membergrowers stand ready to work cooperatively on thisserious challenge

Maine salmon farms in Cobscook Bay shareessentially the same water resources as Canadianfarms In several instances farms from the twocountries are located within 2000 feet of each otherAlthough governance structures are differentCanadian and US fish farms and their respectivepractices directly affect each other Coordination ofISA management programs in Canada and Maine isessential in order to reduce risks in both jurisdictionsThe MAA has opened discussions with its sisterproducer organizations in New Brunswick and NovaScotia to improve communication and harmonize baymanagement practices Similar discussions areoccurring between regulatory authorities in bothjurisdictions These discussions must beaccelerated and legal structures must be developedthat facilitate this cross-border initiative

USDArsquos ISA program has been critical to therapid and aggressive response to the emergence ofclinical ISA in US waters Regional and fieldpersonnel from USDArsquos Animal and Plant HealthInspection Service have provided invaluableassistance to Maine salmon farmers in a timely andprofessional manner Headquarters staff and theSecretary of Agriculture have provided vitalleadership and support for a new programContinued close cooperation between Federal andState authorities and the growers will significantly

132


increase the effectiveness of any ISAv control anddiscuss eradication efforts

Acknowledgments

The author thanks the members of the MaineAquaculture Association Fish Health TechnicalCommittee for their hard work addressing thechallenges ISA has presented over the last severalyears The committee has been assisted by anumber of external fish health professionals Twopeople in particular deserve recognition for theirextensive work and patience Dr Michael Opitz wasinstrumental in developing the first biosecurity auditsand forcing us as growers to face our own mistakesand weaknesses Dr Peter Merrill has workedtirelessly to help write the industryrsquos ISA Action Planand has continued the development andimplementation of Dr Opitzrsquos biosecurity audits Ithank them both for their efforts time and dedicationto the continuing the viability of Maine aquaculture

References Cited

Allen E Opitz H M 1999 Biosecurity audit ofMarine salmon production sites and salmonprocessing plants in Maine Richmond MEMicrotechnologies 56 p

Anonymous 2000 ISA hits the Faeroes FishFarming International 27 47

Aquaculture Development Committee 1990 Anaquaculture development strategy for the State ofMaine Augusta ME Maine State Planning Office105 p

Bouchard D Kelleher W Opitz H M Blake SEdwards K Nicholson B Taylor S 1997 A casestudy of HKS in New Brunswick histopathology virusisolation electron microscopy and molecularcharacterization In HKS workshop 5 November1997 St Andrews NB [Place of publicationunknown] HKS Industry Committee 14ndash18

Bouchard D Kelleher W Opitz H M Blake SEdwards K Nicholson B 1999 Isolation ofinfectious salmon anemia virus (ISAv) from Atlanticsalmon in New Brunswick Canada Diseases ofAquatic Organisms 35 131ndash137

Haringstein T 1997 Infectious salmon anemia (ISA) ahistorical and epidemiological review of thedevelopment of the disease in Norwegian fish farmsIn Workshop on infectious salmon anemia 26November 1997 St Andrews NB Fredericton NBNew Brunswick Department of Fisheries andAquaculture 5ndash12

Hovland T Nylund A Watanabe K Endresen C1994 Observation of infectious salmon anemia virusin Atlantic salmon Salmo salar L Journal of FishDiseases 17 291ndash296

Kibenge FSB Gaacuterate O N Johnson GArriagada R Kibenge MJT Wadowska D 2001Isolation and identification of infectious salmonanemia virus (ISAV) from Coho salmon in ChileDiseases of Aquatic Organisms 45 9ndash18

Krossoslashy B Nilsen F Falk K Endresen C andNylund A 2001 Phylogenetic analysis of infectiousDiseases of Aquatic Organisms 44 1ndash6

Lovely J E Dannewig B H Falk K Hutchin LMacKinnon A M Melvill K J Rimstad E GriffithsS G 1999 First identification of infectious salmonanemia virus in North America with hemorrhagickidney syndrome Disease of Aquatic Organisms 25145ndash148

Maine Aquaculture Association 1998 Infectioussalmon anemia action plan Brewer ME MaineAquaculture Association Fish Health Committee16 p

Maine Aquaculture Association 2002 MaineAquaculture Association finfish bay managementagreement Hallowell ME Maine Aquaculture FinfishSteering Committee 39 p

133


Maine Department of Marine Resources and MaineCoastal Program 1997 Mainersquos aquaculture strategyAugusta ME Maine State Planning Office [Totalpages unknown]

Merrill P L 2001 2000 Biosecurity audit reportfarmed salmon marine sites amp processing plantsRichmond ME Microtechnologies 78 p

Merrill P L 2000 1999 Biosecurity audit reportRichmond ME Microtechnologies 62 p


Rowley H M Campbell S J Curran W LTurnbull T Bryson D G 1999 Isolation of infectioussalmon anemia virus (ISAv) from Scottish farmedAtlantic salmon Salmo salar L Journal of FishDiseases 22 483ndash488

Thorud K Djubvik H O 1988 Infectious anemia inAtlantic salmon (Salmo salar) Bulletin of theEuropean Association of Fish Pathologists 8109ndash111

134


135

Survey of Nonsalmonid Marine Fishes for Detection of InfectiousSalmon Anemia Virus and Other Salmonid Pathogens

Survey of Nonsalmonid Marine Fishes forDetection of Infectious Salmon Anemia Virusand Other Salmonid Pathogens

Sharon A MacLean Deborah A Bouchard and Stephen K Ellis1

Introduction

Infectious salmon anemia (ISA)mdasha viral diseasecaused by an orthomyxovirus possibly a member ofa new proposed genus Aquaorthomyxovirus(Krossoslashy et al 1999)mdashhas resulted in seriousimpacts to the Atlantic salmon (Salmo salar) farmingindustry in several countries The disease was firstreported from Norway in 1984 (Thorud and Djupvik1988) In more recent years cases of the diseasehave been reported from eastern Canada (Mullins etal 1998) Scotland (Rodger et al 1998) the FaroeIslands (Office International des Epizooties [OIE]2000) Chile (Kibenge et al 2001) and theNortheastern United States (Bouchard et al 2001)

Outbreaks of ISA in the United States havebeen confined to Cobscook Bay Maine a fishery thataccounted for more than 50 percent of the 36 millionpounds of Atlantic salmon culture produced in theState in 2000 Within 10 months of the first officiallyreported case of ISA in February 2001 fish in morethan 90 percent of the culture sites in Cobscook Bay

Abstract In an effort to identify potential reservoirs ofsalmonid pathogens nearly 3000 fish including alewife(Alosa pseudoharengus) American eel (Anguilla rostrata)Atlantic herring (Clupea harengus harengus) Atlanticmackerel (Scomber scombrus) Atlantic cod (Gadusmorhua) haddock (Melanogrammus aeglefinus) Atlantichalibut (Hippoglossus hippoglossus) pollock (Pollachiusvirens) American shad (Alosa sapidissima) and winterflounder (Pseudopleuronectes americanus) were sampledfrom the natural environment Pollock cod and lumpfish(Cyclopterus lumpus) also were sampled from within cagesholding infectious salmon anemia (ISA)-diseased salmonAssays included cell culture for listed salmonid viruses thedirect fluorescent antibody test for Renibacteriumsalmoninarum and reverse-transcriptasendashpolymerasechain reaction (RTndashPCR) for ISA virus (ISAv) All of thefish collected from the natural environment tested negativeby any assay method Two of 12 pollock taken from insidea cage with ISA-diseased salmon showed weak RTndashPCR

positive results and were cell-culture negative Ninetypollock collected outside a cage with diseased salmontested negative for viruses and R salmoninarum One of24 pools (5 fish per pool) of tissues from cod taken from awellboat holding salmon from a cage with clinically dis-eased fish produced cytopathic effects (CPE) characteristicof ISAv on salmon head kidney cells This finding wasconfirmed by RTndashPCR of cell culture supernatant Viralpathogens and R salmoninarum were not detected in 26lumpfish collected from inside diseased cages These datasuggest a need for attention to biosecurity practicesconcerning nonsalmonids retained in and harvested fromsalmon cages These results indicate that pollock and codcan harbor ISAv however it was not determined if the viruscan replicate within these hosts The significance of suchpotential carriers to the epizootiology of ISA needs furtherinvestigation as a source of the virus in the wild and toexamine potential impacts on nonsalmonid populations

1 Ms MacLean is with the National Oceanic andAtmospheric Administrationrsquos National Marine FisheriesService in Narragansett RI Ms Bouchard is with MicroTechnologies Inc in Richmond ME Dr Ellis is withUSDA APHIS Veterinary Services in Eastport ME

were diseased This situation caused the State ofMaine Department of Marine Resources (DMR) andthe US Department of Agriculture Animal and PlantHealth Inspection Service (USDAndashAPHIS) to ordereradication of approximately 15 million ISAv-infectedor -exposed fish Government-mandateddepopulation followed earlier removal by the industryof more than 1 million ISA-exposed fish in efforts tocontrol the disease

Understanding the epizootiology of the diseaseis important in its control therefore identification ofpotential reservoirs of the pathogen in the naturalenvironment becomes key to development of diseasecontrol measures In laboratory studies brown trout(Salmo trutta) and rainbow trout (Oncorhynchusmykiss) have been shown to be asymptomaticcarriers of ISAv that can transmit the virus to salmonby cohabitation (Nylund and Jakobsen 1995 Nylundet al 1995 and 1997) Results of recent studiesconducted in Scotland and Canada indicate that ISAvexists at a low prevalence level in wild salmonids andthat ISAv has been found in Atlantic salmonaquaculture escapees (Olivier 2002 Raynard et al2001)

The National Oceanic and AtmosphericAdministrationrsquos (NOAA) National Marine FisheriesService Northeast Fisheries Science Center

136


(NMFSndashNEFSC) has initiated a broad survey toexamine possible reservoirs of ISAv and othersalmonid pathogens in wild marine fishes BecauseCobscook Bay of Maine has relatively low salmonidpopulations NMFS scientists hypothesized thatnonsalmonids potentially could be reservoirs andpossibly play a role in transmission of the virusbetween culture facilities Species that are commonto coastal Maine and Atlantic salmon culture siteswere targeted in this study These fishes werecollected adjacent to and distant from culture sites inCobscook Bay

Furthermore in November 2000 Atlanticsalmon in eight Maine rivers and streams (theDennys East Machias Machias PleasantNarraguagus Ducktrap and Sheepscot rivers andCove Brook) were listed as endangered under theEndangered Species Act Due to the close proximityof salmon cages to some of these rivers there wasconcern about the potential threat of disease to theseendangered salmon In addition transmission ofISAv from asymptomatic infected trout to Atlanticsalmon by cohabitation has been demonstrated(Nylund and Jakobsen 1995) This fact causedconsideration of potentially infected nonsalmonids toaffect migratory Atlantic salmon as well Thereforesampling included nonsalmonids collected fromselected Atlantic salmon rivers in Maine


Fishes Sampled

Data presented here are for nonsalmonidmarine fishes collected during 2000ndash02 fromlocations adjacent to and distant from Atlantic salmonculture operations in Cobscook Bay With theexception of fishes collected from within or nearAtlantic salmon culture cages all fish collectionswere made ancillary to field programs having otherobjectives Fishes were collected from the naturalenvironment and from Atlantic salmon culture cagesFishes from the natural environment were collected

(1) From the Dennys Narragaugus PleasantSheepscot and Penobscot rivers in Maine duringspringtime smolt trapping operations

(2) From coastal Maine during DMR trawl surveysincluding areas from Penobscot Bay Maine to theborder with Canada

(3) From offshore areas in the Northwest Atlanticduring NMFS bottom trawl surveys from New Jerseyto the Gulf of Maine and

(4) By angler catches within 10 m of Atlantic salmonculture cages in Cobscook Bay Maine

Fishes from the caged environment were taken frominside cages or boat wells during harvest operationsof ISA-diseased Atlantic salmon in Cobscook Bay

Initially fishes were frozen for ease of handlingin the field Subsequently fishes were chilledtransported to the laboratory of Micro TechnologiesInc (Richmond ME) or to field stations fordissection or were dissected fresh onboard surveyships Tissues were appropriately preserved for thevarious assays (Thoesen 1994)

We aimed to sample each species at least atthe 5-percent prevalence detection level (ie N = 60)and oftentimes were able to sample at the 2-percentprevalence detection level (N = 150) Speciessampled from the natural environment includedalewife (Alosa pseudoharengus) (N = 1059)American eel (Anguilla rostrata) (N = 297) Atlanticherring (Clupea harengus harengus) (N = 684)Atlantic mackerel (Scomber scombrus) (N = 211)Atlantic cod (Gadus morhua) (N = 115) haddock(Melanogrammus aeglefinus) (N = 55) Atlantichalibut (Hippoglossus hippoglossus) (N = 2) pollock(Pollachius virens) (N = 123) American shad (Alosasapidissima) (N = 3) and winter flounder(Pseudopleuronectes americanus) (N = 259)Nonsalmonid species collected from inside Atlanticsalmon culture cages included Atlantic cod (N = 120)pollock (N = 16) and lumpfish (Cyclopterus lumpus)(N = 26) Table 1 summarizes the species andnumbers of fishes tested and their general locationsof capture Figure 1 illustrates the general locationsof fishes sampled from the entire survey area and

137


figure 2 highlights sampling locations in Mainewaters

Assay Procedures

The tissue sampling and assay protocols wereperformed according to ldquoSuggested Procedures forthe Detection of Certain Finfish and ShellfishPathogensrdquo (Thoesen 1994) Fishes in the studywere sampled for the following assays viralisolations ISAv-specific RTndashPCR and directfluorescent antibody test (DFAT) for Renibacteriumsalmoninarum causative agent of bacterial kidneydisease Not all assays were conducted on all fishAssay selection was dependent on the field situation(for example NMFSndashNEFSC bottom trawl surveysrun 12 consecutive days at sea in an environment notconducive to cell culture)

Viral IsolationsmdashKidney spleen and gill tissuesfrom a maximum of five fish per individual specieswere pooled placed into phosphate-buffered saline(PBS) and kept chilled until further processed within24 hours Tissues were homogenized and diluted110 in PBS and were further diluted 110 in MinimumEssential Medium (MEM) with Hankrsquos salts Lndashglutamine and 2-percent fetal bovine serum Tissuehomogenates were inoculated onto each of three celllines salmon head kidney (SHK) chinook salmonembryo (CHSEndash214) and Epithelioma papillosumcyprini (EPC) All cultures were monitored for anycytopathic effect (CPE)-producing agents for 28 daysat 16 degC Suspected ISAv CPE was confirmed byISAvRTndashPCR on cell culture supernatants

ISAvRTndashPCRmdashIndividual middle kidney tissue orblood were sampled from fishes and placed inRNALaterTM (Ambion Inc Austin TX USA)Samples were kept chilled at 4 degC and assayed withina few days or were kept frozen at ndash20 degC untilassayed within a few weeks RNA was extractedfrom kidney tissue or blood using the RNeasyreg MiniKit (Qiagen) and subsequently tested by an RTndashPCRprotocol using the ISAv 1D2 primer set Based onpositive controls and appropriate sized marker bandsevaluators considered a resulting amplified bandpositive when it occurred at the 493 base pairmigration position RTndashPCR products were purifiedand submitted for gene sequence analysis

BKDndashDFATmdashSmears of kidney tissue from individualfishes were prepared fixed in methanol andrefrigerated until processed Fixed smears werestained with a fluorescent-labeled polyclonalRenibacterium salmoninarum-specific antiserumaccording to standard DFAT protocols Stained slideswere observed for bacterial fluorescence indicative ofR salmoninarum

Results

A total of 2970 fishes representing 11 species wereassayed for salmonid viral pathogens and Rsalmoninarum the causative agent of bacterialkidney disease (see table 1)

Table 1mdashGeneral locations fishes and numbersampled for disease assays

Location Common name N

Natural environmentRivers in Maine Alewife 944

American eel 297

Coastal Maine Alewife 22Atlantic herring 230Atlantic cod 90Pollock 7Winter flounder 189

Ocean Alewife 93Atlantic herring 454Atlantic mackerel 211Atlantic cod 25Haddock 55Atlantic halibut 2Pollock 26American shad 3Winter flounder 10

Near cages with ISA- Pollock 90diseased salmon Winter flounder 60

Within cages holding Atlantic cod 120diseased salmon Lumpfish 26

Pollock 16

138


Figure 1mdashLocations of fish collections New Jersey to Maine 2000ndash02

139


Cobscook BayDennys

Sheepscot

Penobscot Pleasant

Narraguagus

Figure 2mdashLocations of fish collections in Maine 2000ndash02(Stars = coastalocean sites triangles = river sites)

140


Fishes From the Natural Environment

Excluding Cobscook Bay Maine variousspecies totaling 2658 fish were collected from riversand coastal areas in Maine and from ocean areas inMaine to as far south as New Jersey WithinCobscook Bay 150 fish were collected near salmoncages 90 pollock were taken within 10 m of cageswith ISA-diseased fish and 60 winter flounder werecollected from beneath cages holding uninfectedsalmon

All fishes collected from the natural environmenttested negative for R salmoninarum by BKDndashDFATand for ISAv by RTndashPCR No CPE characteristic ofISAv or other salmonid viruses was observed in cellcultures of fishes from the wild

Fishes From Cages With ISA-DiseasedSalmon

Fishes collected from cages of diseasedAtlantic salmon came from two different samplingevents One event provided 12 pollock the secondevent provided 120 cod 26 lumpfish and 4 pollockwhile cages were being harvested

Two of 12 pollock taken from a cage with ISA-diseased salmon showed a weak positive band byRTndashPCR for ISAv CPE was not observed incorresponding cell-culture samples Conclusive DNAsequencing results were not obtained from the PCRproducts due to the weak amplifications in these twopollock samples

One of 24 pools (5 fishpool) of tissues fromcod taken from a wellboat containing salmonexhibiting clinical ISA yielded CPE characteristic ofISAv on SHK cells RTndashPCR assay of the cell culturesupernatant confirmed the agent as ISAv RTndashPCRassays of the individual fishes that comprised thepositive pool of tissues were negative for ISAv Genesequencing of the RTndashPCR product confirmed thatthe CPE was due to ISAv and demonstrated 99-percent homology with the North American ISAvisolate Assays on 26 lumpfish and 4 pollockcollected along with the cod did not indicate thepresence of viral pathogens

No BKDndashDFAT-positive results were detectedamong the several species of fishes taken fromculture cages or the harvest boat

Discussion

The results of this study indicate that Rsalmoninarum ISAv and other cultured fish virusesare not present at a significant level in the species offishes tested from the natural environment ISAvhowever was detected and isolated fromnonsalmonid fishes that as age-0 juveniles becameentrained in Atlantic salmon culture cages

The isolation of ISAv from tissues of cod takenfrom the harvest boat well presents an interestingand perhaps significant finding Viral culture canstand alone as a definitive test for ISAv as it indicatesthe presence of viable viral pathogen Yet it is notedthat the corresponding direct-tissue RTndashPCRsamples from the individual fish comprising thepositive pool tested negative repeatedly Theseconfounding results have possible explanationsBecause the cod were taken from a wellboatcontaining ISAv-infected fish from a site wheresalmon displayed clinical ISA and gill lamellae wereincluded in the tissue sample exogenous viralparticles could have adhered to the gill tissue as aresult of a high viral load in the boat well and therebybe carried into the cell culture assay That scenariowould account for the negative RTndashPCR of individualkidney tissues of the five fishes that made up thepool However if the viral load in the harvest wellwas substantially high it would seem probable thatmore than 1 of the 30 pools of fishes (cod lumpfishand pollock) collected at that time would have testedpositive

It is also possible that the cod were infectedwith ISAv but with a low viral load as might beexpected in a carrier state Under thesecircumstances it is possible for cell culture to detectvirus that is not detected by RTndashPCR depending onthe tissue distribution of low numbers of virus Inother laboratory studies that we have conducted tocompare various tissue sources and ISAv detection

141


methods CPE has been observed occasionally incell culture without positive RTndashPCR bands for thesame tissue taken from known infected fish(unpublished data) Similar results were obtained ina wild fish survey in which ISAv was isolated from fiveseparate sea trout each of which tested negative byRTndashPCR (Raynard et al 2001)

Two of 12 pollock collected at 1 time from withina cage holding ISA-diseased salmon showed weakbands by ISAvRTndashPCR assay Ninety pollockcollected adjacent to a cage holding ISA-diseasedsalmon tested negative for viruses by viral cultureand ISAvRTndashPCR The RTndashPCR reactions from thetwo pollock did not produce substantial quantities ofpurified nucleic acid consequently the PCR productsdid not result in good gene sequencing reactionsAlthough it is uncertain if those two fish actually wereinfected with ISAv it is possible that the bandsrepresent amplification of degraded ISAv genome ora very low number of ISAv particles It is worthnoting that ISAv was not detected by RTndashPCR insaithe (Pollachius virens) at a minimum of 7 daysafter intraperitoneal injection with the Norwegianstrain of the virus nor in saithe cohabiting withinfected Atlantic salmon (Snow et al 2002) Becausethe corresponding cell cultures for the two RTndashPCR-positive pollock were negative the gel bands wereweak and the gene sequencing reaction was poorthe significance of the positive RTndashPCR resultsremains unclear

These results indicate that cod and pollock canharbor ISAv however it is unknown if the virus canreplicate within these hosts The cod which werecollected in January ranged from 14 to 26 cm (mean186 cm) total length Based on growth rates for theGulf of Maine these cod were about a year old(Pentilla et al 1989) It is unknown just how long thecod and pollock were within the cages althoughbased on published growth rates mesh size of thenets on salmon cages and assumed more rapidgrowth for entrapped fish we may roughly estimatethat these cod were in the cages for 8 months Alsounknown are the extent and duration of theirexposure to ISAv Regardless the positive test

results from cod and pollock in cages containing ISA-diseased salmon is useful information for industryand should bring attention to biosecurity practicesconcerning the handling and disposition ofnonsalmonids retained in or harvested from salmoncages The significance of these potential carriers tothe epizootiology of ISA remains to be investigated

In autumn 2000 and spring 2001 fishescollected for assay were frozen for ease of handlingin the field Although ISAv is still infective afterfreezing at ndash20 oC (Thorud and Torgersen 1994 asreported by Nylund et al 1995) and can be isolatedfrom frozen tissue we have found the rate ofrecovery from frozen tissue is slightly less than 50percent (unpublished data) It is possible that theviral assays of the frozen fishes in this study mayhave been compromised If ISAv indeed is present inthese fishes at a 2- to 5-percent prevalence levels atwhich we sampled we would expect to detect it inassays on fresh tissues from subsequent years butwe did not The BKDndashDFAT results on frozen tissueare valid and in none of the samples (fresh frozennatural or retained) was there a positive result

Studies in Scotland and Canada indicate thatISAv is present in wild salmonid populations but ISAvwas not detected in the wild nonsalmonids examined(Raynard et al 2001 Olivier 2002) We did not assaywild salmonids from Maine for two reasons wildAtlantic salmon in Maine rivers are an endangeredspecies so the few salmon returning to rivers shouldnot be put at risk and populations of other nativeanadromous salmonids are small and scattered inMaine (Joan Trial personal communication) We didtest Atlantic salmon from the commercial WestGreenland fishery and found 1 of 19 positive by RTndashPCR (MacLean and Brown 2002) The genesequencing of this PCR product confirmed ISAv andindicated the viral RNA was of the North Americanstrain Microsatellite DNA analysis to determinecontinent of origin of this fish indicated that itoriginated from North America (King et al 2002)

Recent work by the US Fish and WildlifeService (USFWS) suggests that ISAv may be presentin the wild salmonid population in Maine When first

142


brought to the hatchery 1 of 68 brood Atlanticsalmon collected from the Penobscot River in Maineduring 2001 tested positive for ISAv in blood usingRTndashPCR Gene sequencing of this product indicatedthat the viral RNA was most similar to the Norwegianand Scottish strain of ISAv Two subsequent testsover a 4ndashweek period gave negative RTndashPCR resultsand the fish remained asymptomatic (PatriciaBarbash personal communication) Further assay ofincoming brood fish showed no positive RTndashPCR orviral assay results of 126 Atlantic salmon testedthrough August 2002 These fish will be tested againbefore spawning (Patricia Barbash personalcommunication)

As the testing of wild salmon in Maine has beenlimited it remains to be determined if a reservoir ofthe pathogen resides in salmonids andornonsalmonid species in the waters of MaineTherefore speculation on the source of the ISA virusin cultured Atlantic salmon in Maine cannot be madenow

Acknowledgments

We thank the many individuals involved in fieldcollections of fishes for analysis including NOAAndashNMFSndashNEFSC and Maine Atlantic SalmonCommission field crews Sally Sherman Maine DMRChris Bartlett Maine Sea Grant Walt MalkowskiNOAAndashNMFS and the Maine salmon aquacultureindustry We also thank the University of MainersquosDNA Sequencing Facility for gene sequenceanalyses and Timothy E Sheehan NOAAndashNMFS forgraphics

References Cited

Bouchard Deborah Brockway Keith Giray CemKeleher William Merrill Peter L 2001 First report ofinfectious salmon anemia in the United StatesBulletin of the European Association of FishPathologists 21 86ndash88

Kibenge Frederick S B Garate Oscar N JohnsonGerald Arriagada Roxana Kibenge Molly J TWadowska Dorota 2001 Isolation and identificationof infectious salmon anemia virus (ISAV) from cohosalmon in Chile Diseases of Aquatic Organisms 459ndash18

King T L Redding D G Brown R W KanneworffP 2002 Continent of origin of Atlantic salmoncollected at West Greenland 2001 Work Pap 200234 Copenhagen International Council for theExploration of the Sea North Atlantic SalmonWorking Group 13 p

Krossoslashy Bjorn Hordvik Ivar Nilsen Frank NylundAre Endresen Curt 1999 The putative polymerasesequence of infectious salmon anemia virus suggestsa new genus within the Orthomyxoviridae Journal ofVirology 73 2136ndash2142

MacLean Sharon A Brown Russell W 2002 ISAoutbreaks and disease testing of Atlantic salmon inthe United States and the West Greenland fisheryWork Pap 200220 Copenhagen InternationalCouncil for the Exploration of the Sea North AtlanticSalmon Working Group 3 p



Nylund A Alexandersen S Rolland J BJakobsen P 1995 Infectious salmon anemia virus(ISAV) in brown trout Journal of Aquatic AnimalHealth 7 236ndash240

Nylund A Kvenseth A M Krossoslashy B HodnelandK 1997 Replication of the infectious salmon anemiavirus (ISAV) in rainbow trout Oncorhynchus mykiss

143


(Walbaum) Journal of Fish Diseases 20 275ndash279

Office International des Epizooties 2000 Infectioussalmon anemia in the Faroe Islands OIE DiseaseInformation 13(14) 53

Olivier Gilles 2002 Disease interactions betweenwild and cultured fishmdashperspectives from theAmerican Northeast (Atlantic Provinces) Bulletin ofthe European Association of Fish Pathologists 22103ndash109

Pentilla Judith A Nelson Gary A Burnett John MIII 1989 Guidelines for estimating lengths at age for18 northwest Atlantic finfish and shellfish speciesNOAA Tech Memo NMFSndashFNECndash66 NationalOceanic and Atmospheric Administration NationalMarine Fisheries Service 39 p

Raynard R S Murray A G Gregory A 2001Infectious salmon anemia virus in wild fish inScotland Diseases of Aquatic Organisms 46 93ndash100

Rodger H D Turnbull T Muir F Millar SRichards R H 1998 Infectious salmon anaemia(ISA) in the United Kingdom Bulletin of theEuropean Association of Fish Pathologists 18 115ndash116

Snow M Raynard R Bruno D W Van NieuwstadtA P Olosen N J Lovold T Wallace C 2002Investigation into the susceptibility of saithePollachius virens to infectious salmon anaemia virus(ISAV) and their potential role as a vector for viraltransmission Diseases of Aquatic Organisms 50 13ndash18

Thoesen John C ed 1994 Suggested proceduresfor the detection and identification of certain finfishand shellfish pathogens 4th ed version 1 RockvilleMD American Fisheries Society Fish Health Section


Thorud K Torgersen Y 1994 ILAmdashKan sloslashyd fiskfra smittede anlegg vaeligre smittefarlig NorskFiskeoppdrett 9 44ndash45


Patricia BarbashUS Department of the InteriorUS Fish and Wildlife ServiceFish Health Unit400 Washington AvenueLamar PA 16848

Joan TrialMaine Atlantic Salmon Commission650 State StreetBangor ME 04412ndash5654

144


145


Infectious Salmon Anemia in New BrunswickAn Historical Perspective and Updateon Control and Management Practices(1997ndash2002)

Sandi M McGeachy and Mark J Moore1

Introduction and History ofISA

The New Brunswick Atlantic salmon (Salmo salar)farming industry has grown from one farm in 1979 to96 farms in 2002 and current production is estimatedat 40000 metric tons annually All of these farms arelocated in southwestern New Brunswick with close to50 percent of the farms being within 30 km of eachother The industry has established 22 baymanagement areas (BMAs) within the southwesternpart of the Bay of Fundy in an effort to facilitate singleyear-class farming and agreements on codes ofpractice

In the summer of 1996 two sites in two BMAs(Lime Kiln BaymdashBMA 10 and Bliss HarbourmdashBMA 9)experienced unexplained elevated levels of mortalityamong premarket size Atlantic salmon of the 1995-year class Seventeen farms were located in thesetwo bay areas within 5 km of each other (fig 1)Disease investigations concentrated on bacterialpathogens before viral toxicological and histologicalassays were conducted In the fall of 1996 additionalfarms in these bays also sustained elevated fishmortality Various fish health specialists and servicesbecame involved in attempts to identify possibledisease or toxicological agents associated with theselosses It was not until late in 1996 when commonhistological lesions were identified in the kidneys andlivers of affected fish that some form of a diagnostictest was used to describe the distribution of similar

Abstract Although infectious salmon anemia (ISA) hasbeen present in New Brunswick since 1996 the causativeagent was not identified until 1997 As a result ISA spreadfrom just a few farms to several farms in three bay manage-ment areas over a 2-year period (1996ndash97) A number ofmanagement and control procedures have helped reduceISA viral loads and infection rates these include single

1Dr McGeachy is with the Department of AgricultureFisheries and Aquaculture in Fredericton NB Dr Mooreworked there during preparation of this paper but is nowwith Maritime Veterinary Services Ltd in St George NB

year-class farming early detection and removal of infectedfish fallowing and containment of bloodwater and process-ing waste However ISA continues to be a serious diseaseof concern in New Brunswick This manuscript discussesthese management and control strategies as well as furthermeasures that may be implemented to control ISA in NewBrunswick

mortality events At this time the condition wastermed hemorrhagic kidney syndrome (HKS)because the histopathology lesions were notconsistent with those produced by any other knowndisease agents (Bryne et al 1998)

Some researchers and companies alsosuspected that HKS was a new clinical presentationof bacterial kidney disease (BKD) due to the fact thatfish at a number of farms with HKS problems hadBKD Managers of farms in these areas attemptedtreatments with various antibiotics but did not curtailHKS

As part of disease testing and surveillancethroughout late 1996 a togavirus was isolated in fishassociated with HKS (Kibenge et al 2000a)However further transmission work underexperimental conditions showed that this virus did notcause mortality in salmon and the etiology of HKSremained unknown

With consistent histological lesions beingassociated with the high levels of mortality the NewBrunswick Department of Fisheries and Aquaculture(NBDFA) completed an extensive histologicalsampling survey of fish from all of the farms in theindustry between February and September 1997The early focus of the screening was placed on theoriginal affected BMAs (9 10 and 20) andneighboring BMAs (6 7 8 and 19) in an attempt toascertain the spread of HKS epidemiologically Atotal of 4723 kidney samples from 767 cages on 69sites were tested by histological examination Therewere 373 positive fish from 109 cages at 20 siteslocated in 5 BMAs From the fish tested the percentbreakdown of positive samples by year-class was asfollows 73 percent among the spring 1995-year

146


class in 11 cages 78 percent among the fall 1995-year class in 8 cages 623 percent among the spring1996-year class in 60 cages 189 percent among thefall 1996 in 16 cages abd 32 percent among spring1997-year class in 5 cages Prevalence of HKS was05 percent among broodstock and an additional 540fish were labeled as ldquosuspectrdquo The survey also notedthat 951 fish (18 percent) were found to havehistological evidence of BKD infection but only 2 ofthese fish were positive for HKS and 8 wereconsidered suspect

Because the cause of the mortality remainedunknown it was difficult to prevent the stocking ofsites in affected areas (eg BMA 9 10 and 20)Although a few farms did not place smolt on theirsites in the spring of 1997 most farms did Thehistological survey in 1997 had shown that HKS wascontained within Lime Kiln Bay Bliss Harbour and

Figure 1mdashMap of aquaculture bay management areas (BMAs)and sites for Lime Kiln Bay (BMA 10) Bliss Harbour (BMA 9) andBack Bay (BMA 8) in New Brunswick Canada

Seal Cove (BMA 20) located on Grand Manan IslandHowever some survey work indicated that a limitednumber of fish tested positive for HKS in Back Bay(BMA 8) and Deer Island East (BMA 6) indicatingpotential spread into neighboring areas by thesummer of 1997

Further research and testing for a causativeagent resulted in the isolation and identification of theinfectious salmon anemia virus (ISAv) by theResearch and Productivity Council (RPC) inSeptember 1997 This virus was conclusivelydemonstrated to be the etiologic agent for HKS(Lovely et al 1999) and further confirmation wasprovided by identification of ISAv samples sent to theNational Veterinary Institute in Norway

This was the first time ISA had been reportedoutside of Norway This confirmation became aturning point that enabled the New Brunswick

147


government and industry to implement appropriatecontrol and management strategies A number of riskfactors concurrent with ISA were already identified inNorway (Varinggsholm et al 1994 Jarp and Karlsen1997) Risks included

Timing of detection and slaughter of infected fishand cages

Number of generations on a site

Proper handling and disposal of dead fish

A 5-km zone of influence for slaughterhouses andeffluent discharges from processing plants near fishfarms

Harvest procedures and bloodwater containment

Transportation corridors

Zoning of infected farms in high-risk ldquocombatrdquozones

Fallowing of sites and

Disinfection protocols

As a result of reviewing these risk factors andinformation contained in the action plan known asldquoStop ISAmdashNorway 1994rdquo (Anonymous 1994) newcontrol and management strategies wereimplemented in New Brunswick (NB) The NBIntegrated Management Plan for ISA was based ondetection containment control and prevention TheNB Department of Fisheries and Aquaculture(NBDFA) established an extensive ISAv SurveillanceProgram An Eradication Program based onMinisterial-ordered depopulations of clinically infectedfish was also established

The Government of Canada supported initialfunding for the compensation program under theauspices of the National Disaster Relief ProgramIndustry and NBDFA also recognized the need for acomprehensive ISA Management and ControlProgram involving the adoption of stringenthusbandry harvesting processing and transportpractices including the disinfection of equipment andthe replacement of wooden cages wooden harvestbarges and wooden feed boats A Market ProtectionProgram was also developed by the industry and apublic relations firm was hired to prepare a

comprehensive plan that addressed issues related tohuman health and consumer perceptions

Depopulation of infected stocks began withwhole site removals in Lime Kiln Bay Bliss Harbourand Seal Cove These areas were completelyemptied of salmon from cages in June 1998 and thenfallowed for 10 to 12 months prior to smolt placementin the spring of 1999 Work on disinfectionprocedures and harvesting guidelines were alsobeing developed (Torgersen and Haringstein 1995Washburn and Gillis Associates Ltd 1998 Neill andGunter Ltd 1999) Zones or areas of concern wereimplemented to some degree with mandatoryharvesting of market-size fish and restriction on smoltplacements in some areas (eg Back BaymdashBMA 8)

During the spring of 1998 the New Brunswickgovernment and industry established a Fish HealthTechnical Committee comprised of industryveterinarians and government representatives (bothFederal and Provincial) to look at fish health issuesfor farmers in the Bay of Fundy and to provide adviceto the New Brunswick Minister of Fisheries andAquaculture One such recommendation of theCommittee was that smolt should not be placed intoBack Bay (BMA 8) due to its proximity to Lime KilnBay and Bliss Harbour (within 2 km of affectedfarms) As it turned out fish from three out of eightfarms in Back Bay became infected within months ofthe smolt restriction order

In 1998 a total of six additional farms with1997-year class fish became infected with ISA as thedisease spread to Back Bay (BMA 8) LrsquoEtete (BMA7) and Beaver Harbour (BMA 12) By this time all ofthe farms in the three BMAs originally affected werefallowed and it was apparent that total containmentof ISA was not going to be achieved

Total site depopulation was not an option at thattime but industry and government agreed that cage-by-cage removal of infected stocks was the bestmethod of reducing the spread of ISA within andbetween sites Further recommendations andMinisterial orders came for the early harvest ofmarket fish (1997-year class) in Back Bay LrsquoEteteand Beaver Harbour by December 31 1998 By the

148


end of 1998 a total of 24 sites that had 1997-yearclass smolts were affected by ISA This resulted inthe depopulation of nearly 17 million salmon(table 1)

Table 1mdashAtlantic salmon farms affected by infectioussalmon anemia (ISA) in New Brunswick Canada

Mean no ofsalmon

No of ISA- No of salmon depopulatedYear class affected farms depopulated per farm

1995 4 NA NA1996 17 NA NA1997 24 1667870 694951998 16 1190511 744061999 24 1636518 681882000 9 221700 246332001 15 985000 65667

Data for 1995 and 1996 were estimated due to unknown cause of mortalityprior to the point at which a confirmatory diagnosis of hemorrhagic kidneysyndrome and ISA was made

19950

5

10

15

20

25

1996 1997 1998 1999 2000 2001

Year-class production cycle

No

of I

SA

-affe

cted

farm

s

Data for 1995 and 1996 were estimated due to unknown cause of mortality prior to the point at which a confirmatory diagnosis of hemorrhagic kidney syndrom and ISA was made

Figure 2mdashnumber of Atlantic salmon farms affected by ISA in NewBrunswick Canada by year-class production cycle

Newly placed smolts of the 1998-year classbecame infected in Beaver Harbour in August (within4 months of transfer to saltwater) while newly placedsmolts in LrsquoEtete became infected in the fall of 1998(6ndash8 months after saltwater transfer) Sixteen siteswith 1998-year class salmon became affected by ISAduring their production cycle (fig 2)

A 4-month fallow period was implemented formany of the BMAs affected by ISA Fallowing was adifficult decision to make because scientific data didnot exist on the minimum acceptable fallow periodOn the west side of Deer Island 9 out of 12 sitesbecame affected (1998-year class) of which 6 siteswere multiyear class (1998-year class and 1999-yearclass) As a result newly placed smolts of the 1999-year class became infected within 4 months oftransfer to saltwater Once again this infectionscenario necessitated whole-site depopulations andharvesting of market fish to contain the spread ofISA

By the end of the 1999-year class productioncycle ISA had affected close to 50 of NewBrunswickrsquos 86 farms and more than 4000000salmon had been depopulated ISA had now spreadbeyond the inshore salmon sites in 1999 and wasprogressively moving down the west side of DeerIsland In total 24 farms became affected by ISA inthe 1999-year class (fig 2) Over the next 2 yearsISA continued to spread slowly Although the extentof the infection and losses were high the losses werebelow the levels experienced with the 1997 to 1999-year classes (table 1)

Control and Management ofISA

Some of the the main components of the NBIntegrated Management Plan were modeled oninformation from Norway and based on advice fromthe Fish Health Technical Committee One of themajor components was the ISAv SurveillanceProgram which enabled the early detection and earlyremoval of affected cages Other components werebased on restructuring the industry to single year-

149


class farming compensation for ordered depopulatedstocks disinfection controlling vectors such as sealice research and containment of bloodwater andprocessing waste All of the salmon processingplants completed infrastructure upgrades to allow forproper containment and treatment of processingwaste

ISAv Surveillance Program

Once ISAv was identified as the causative agent forHKS (Lovely et al 1999) industry and governmentestablished an ISAv Surveillance Program based onan extensive surveillance of the entire industryUnder guidance from the Fish Health TechnicalCommittee the ISAv Surveillance Program wasbased on early detection and removal of ISA-infectedstocks on a cage-by-cage basis Either governmentbiologistsveterinarians or industry veterinariansvisited sites showing no evidence of infection every6ndash8 weeks but suspect or positive sites were visitedevery 2ndash4 weeks Weak or moribund fish werecollected and necropsied at the NBDFA laboratorywhere a full range of ISAv tests were completed(NBDFA 1998) The main objectives of the ISAvSurveillance Program were to detect the presence ofsubclinical ISA within cages and on sites at theearliest opportunity and to detect the earlyemergence of clinical outbreaks of ISA at the cageand site level

Before the early 1990s ISA diagnosis inNorway was based on evaluation of clinical signsnecropsy findings histological changes andhematological findings (Anonymous 1994 Haringstein1997) It was not until the mid-1990s that Norwegiandiagnosticians used virus isolation (Dannevig et al1995) and experimented with indirectimmunofluorescent antibody (IFAT) tests (Falk 1997Falk et al 1998) and reverse-transcriptasendashpolymerase chain reaction (RTndashPCR) tests (DeVoldet al 2000) The IFAT and RTndashPCR assays were notrecognized by the Office International des Epizooties(OIE) as official or approved tests for ISA (OIE 1995and 1997) In fact the OIE (1997) clearly stated that

the diagnostic procedures for ISA were to be basedon clinical pathological histopathological andhematological changes The IFAT test could be usedto confirm the presence of viral antigens and toresolve cases that were otherwise inconclusive

When the New Brunswick ISAv SurveillanceProgram was developed it was decided to test eachsampled fish for ISA by histological examinationvirus isolation on SHK cell lines (Dannevig et al1995) IFAT and RTndashPCR By 2000 histologicaltests were dropped as the reliability and repeatabilityof the IFAT and RTndashPCR appeared to be sufficient forsurveillance purposes Currently veterinary clinicalimpressions and IFAT tests on moribund fish samplesare used for preliminary screening on all sites BothRTndashPCR and viral tissue culture are then used toconfirm positive IFAT results The ISAv SurveillanceProgram is in its sixth year and an average of about8000 samples are analyzed annually

Single Year-Class Farming

As already mentioned production of multiple year-class generations on a given farm is a key factor thatincreases the risk of ISA (Jarp and Karlsen 1997) Itis estimated that in 1996 more than 60 percent of thesalmon farms in New Brunswick were run asmultiyear-class operations that had as many as threeor four generations present on some sites More andmore farms have restructured to single-year sites andsingle year-class bays Today 98 percent of thefarms are single year-class with 100 percent of thefarms completing this transition in the fall of 2002However it should be noted that single year-classfarming as currently defined by policy in NewBrunswick allows for up to 20 percent of the marketfish to be held over on a site into their third year for a4-month overlap with newly placed smolts (NBDAFA2000) The holding over of market fish is contingentupon ISA status for that farm and farms in thatparticular BMA Since it takes 14 to 16 months inseawater for salmon to reach market size true singleyear-class sites would require the industry tomaintain more than two sites to allow for full crop

150


rotation and complete fallowing between yearclasses

The incidence of ISA on multiyear-class sites forthe 1999-year class of salmon which becameinfected with ISA was double the level on single-yearclass sites (table 2) Affected multiyear-class sitesremoved on average about six cages per site ascompared to three cages on affected single year-class sites (McGeachy 2001) Fish on multiyear-class sites also became infected within 6 months ofsmolt transfer whereas infection did not occur until12 to 14 months after transfer among single year-class sites

A comparison of the data involving the numberof affected farms and depopulation numbers wascompleted on four BMAs that had ISA in 1997ndash98The bay areas of Lime Kiln Bay (BMA 10) BlissHarbour (BMA 9) Back Bay (BMA 8) and Seal Cove

Table 2mdashData on ISA at New Brunswick salmon farmscomparing ISA-affected single year-class (SYC) andmultiyear-class (MYC) sites

Year-class 1998 1999 1999SYC sites SYC sites MYC sites

Mean time (months) toISA clinical disease 14 12 6

Mean no of cagesdepopulated peraffected farm 0 3 6

Table 3mdashThe history of ISA infection in four BMAs in New Brunswick Canada during three production cycles(1997 1999 and 2001 year-classes)

- - - - - - 1997 year-class - - - - - - - - - - - - 1999 year-class - - - - - - - - - - - - 2001 year-class - - - - - -Mean fish Mean fish Mean fish

ISA Fish removed ISA Fish removed ISA Fish removedBMA Farms farms removed per farm farms removed per farm farms removed per farm

Lime Kiln 10 9 583881 64875 2 25577 12788 3 95000 31667BlissHarbour 7 6 583990 97332 2 215707 107853 4 408700 102175

Back Bay 8 3 114760 38253 1 33000 33000 3 62034 20678Seal Cove 4 3 206823 68941 0 0 0 0 0 0

Totals 29 21 1489454 70926 5 274284 54865 10 565734 56573

on Grand Manan (BMA 20) were the first to fullyapply initial control and management plans for ISA byimplementing single year-class sites and baysdisease surveillance and containment of bloodwaterand processing plant wastes Comparison of diseasestatistics within these four bay areas indicated thatthe prevalence of ISA in the 1999-year class wasreduced by 60 percent to 70 percent in comparison tothe 1997-year class (table 3) In fact ISA has notrecurred in the Seal Cove BMA for two productioncycles With the exception of Bliss Harbour lossesand forced depopulations have been reduced from900000 salmon for the 1997-year class to just over150000 salmon in the 2001-year class Theprevalence in and losses of salmon to ISA in BlissHarbour have not changed dramatically whichsuggests that factors other than single year-classfarming are important for controlling ISA and must beaddressed to completely effect disease control Themean number of salmon removed per affected site inBliss Harbour has remained constant at about100000 fish per affected site Current investigationsare being conducted within this area to identifyfurther control strategies It should be noted that sixpreviously affected BMAs have successfully gonethrough complete production cycles (18ndash24 months)and taken salmon to market without furtherreinfection

151


Compensation forDepopulation

Depopulation or stamping-out procedures have beenconsidered a crucial component for the control ofnontreatable viral infections such as ISA (Anonymous1994) At the time when HKS and ISA was identifiedin New Brunswick there was and still remains nonational compensation program for the ordereddepopulation of diseased fish such as those infectedwith ISA As noted earlier the Government ofCanada jointly funded compensation with theProvince of New Brunswick for a 3-year period Notuntil 1999 were regulatory amendments completed tomake it mandatory for farms to secure financing forcompensation (self-compensation through industry)While depopulations were occurring (1998ndash2000)funds for compensation were always on a year-to-year basis which was disconcerting to the farmers

A long-term compensation program remains anissue with the industry and government Under theauspices of the National Aquatic Animal HealthProgram (NAAHP) indemnification is being proposedto compensate farmers for cases such as ISA andother exotic diseases The NAAHP not only includesa component on compensation but it will also developguidelines and practices for disease surveillancezoning research husbandry practices and qualitycontrolquality assurance in support of animal healthfor the aquaculture industry This Program is stillbeing developed jointly between the provincesterritories industry and the Federal Government ofCanada

Research

After the industry was affected by ISA and thecausative agent was identified government andindustry soon realized that a number of facts aboutISA remained unknown An HKS Science Committeeof researchers institutions and veterinarians wasorganized by government and industry to reviewresearch priorities and complete such researchStudies were completed on strain identification

(Kibenge et al 2000b Ritchie et al 2001) verticaltransmission of the virus from broodfish to offspring(Melville and Griffiths 1999) nonlethal detectionmethods (Griffiths and Mellville 2000) andepidemiology (Hammell and Dohoo 1999) Suchresearch has filled scientific voids in basic knowledgeconcerning the disease process and remains an areaof critical importance

Current and Future Initiatives

Current and future initiatives will focus on moving totrue single year-class farming to allow for all-in all-out production (complete fallow periods on all farms)improvements in biosecurity and husbandrysustaining the industry from a fish health perspectiveimproving key infrastructures (eg wharves) andepidemiologic investigations to assist industry andgovernment on improving upon current control andmanagement practices for ISA

Conclusion

More than 55 percent of the salmon farms in NewBrunswick have been adversely impacted by ISA atone time or another during the past five years (1997ndash2002) Although the disease has seriously affectedthe salmon farming industry considerable progresshas been made to control infection reduce mortalityand compensate for lost revenue With the exceptionof a few BMAs forced depopulation due to ISA hasbeen reduced by as much as tenfold Managementfactors that have been implemented to control thedisease are at least partially responsible for the factthat ISA has not recurred in one bay area for threeproduction cycles and in another six bay areas thedisease has been absent for up to two productioncycles Thus it is promising to note that ISA can beeffectively controlled and managed Furtherimprovements in management practices fosteredthrough advances in scientific research will helpindustry and government authorities alleviate thedeleterious biological and financial impacts of thisdisease

152


References Cited

Anonymous 1994 Plan for fighting infectious salmonanemia Proposed action plan Oslo RoyalNorwegian Ministry of Agriculture 61 p

Bryne P J MacPhee D D Ostland V E JohnsonG Ferguson H W 1998 Haemorrhagic kidneysyndrome of Atlantic salmon Salmo salar L Journalof Fish Diseases 21 81ndash91

Dannevig B H Falk K Namork E 1995 Isolationof the causal virus of infectious salmon anaemia(ISA) in a long-term cell line from Atlantic salmonhead kidney Journal of General Virology 76 1353ndash1359


Falk K 1997 Diagnostic methods for detection ofISA virus In Proceedings workshop on infectioussalmon anemia 26 November 1997 St Andrews NBFredericton NB New Brunswick Department ofFisheries and Aquaculture 27ndash43

Falk K Namork E Dannevig B H 1998Characterization and application of a monoclonalantibody against infectious salmon anaemia virusDiseases of Aquatic Organisms 34 77ndash85


Hammell K L Dohoo I R 1999 Epidemiologicstudy of risk factors for mortalities in Atlantic salmonin areas of New Brunswick affected withhaemorrhagic kidney syndromemdashinfectious salmonanemia Tech Rep Charlottetown PE AtlanticVeterinary College for Department of Fisheries andOceans Canada 98 p

Haringstein T 1997 Infectious salmon anaemia (ISA) Ahistorical and epidemiological review of thedevelopment and spread of the disease in Norwegianfish farms In Proceedings workshop on infectioussalmon anemia 26 November 1997 St Andrews NBFredericton NB New Brunswick Department ofFisheries and Aquaculture 6ndash12

Jarp J Karlsen E 1997 Infectious salmon anaemia(ISA) risk factors in sea-cultured Atlantic salmonSalmo salar Diseases of Aquatic Organisms 28 79ndash86

Kibenge FSB Whyte S K Hammell K LRainnie D Kibenge M T Martin C K 2000a Adual infection of infectious salmon anaemia (ISA)virus and a togavirus-like virus in ISA of Atlanticsalmon Salmo salar in New Brunswick CanadaDiseases of Aquatic Organisms 42 11ndash15

Kibenge FSB Lyaku J R Rainnie D Hammell KL 2000b Growth of infectious salmon anaemia virusin CHSE-214 cells and evidence for phenotypicdifferences between virus strains Journal of GeneralVirology 81 143ndash150

Lovely J E Dannevig B H Falk K Hutchin LMacKinnon A M Melville K J Rimstad EGriffiths S G 1999 First identification of infectionsalmon anaemia virus in North America withhaemorrhagic kidney syndrome Diseases of AquaticOrganisms 35 145ndash148

McGeachy S M 2001 ISA in New Brunswick updateand control strategies In Proceedings 9th annualNew England farmed fish health managementworkshop 5 April 2001 Machias ME Orono MEUniversity of Maine Cooperative Extension 13

Melville K Griffiths S 1999 Absence of verticaltransmission of infectious salmon anemia virus(ISAV) from individually infected Atlantic salmonSalmo salar Diseases of Aquatic Organisms 38231ndash234

153


Neill and Gunter Ltd 1999 Procedures for approvalof disinfection systems Rep 9390 for New BrunswickDepartment of Fisheries and AquacultureFredericton NB Neill and Gunter Ltd 14 p

New Brunswick Department of Agriculture Fisheriesand Aquaculture (NBDAFA) 2000 Bay of FundyMarine Aquaculture Site Allocation PolicyFredericton NB New Brunswick Department ofAgriculture Fisheries and Aquaculture 21 p

New Brunswick Department of Fisheries andAquaculture (NBDFA) 1998 ISAV SurveillanceProgram Fredericton NB New BrunswickDepartment of Fisheries and Aquaculture 6 p

Office International des Epizooties (OIE) 1995Infectious salmon anaemia In Diagnostic manual foraquatic animal diseases 1st ed Paris OfficeInternational des Epizooties 101ndash107

Office International des Epizooties (OIE) 1997Infectious salmon anaemia In Diagnostic manual foraquatic animal diseases 2d ed Paris OfficeInternational des Epizooties 119ndash126

Ritchie R J Cook M Melville K Simard NCusack R Griffiths S 2001 Identification ofinfectious salmon anemia virus in Atlantic salmonfrom Nova Scotia (Canada) evidence for functionalstrain differences Diseases of Aquatic Organisms 44171ndash178

Torgersen Y Haringstein T 1995 Disinfection inaquaculture Revue scientifique et technique OfficeInternational des Epizooties 14(2) 419ndash434

Varinggsholm I Djupvik H O Willumsen F V Tveit AM Tangen K 1994 Infectious salmon anaemia (ISA)epidemiology in Norway Preventive VeterinaryMedicine 19 277ndash290

Washburn and Gillis Associates Ltd 1998Guidelines for disinfection hygiene standards foraquaculture marine grow-out operations andassociated services specific to infectious salmonanaemia Fredericton NB New BrunswickDepartment of Fisheries and Aquaculture 60 p

154


155

Experiences With Regulatory Responses toInfectious Salmon Anemia in Norway

Experiences With Regulatory Responsesto Infectious Salmon Anemia in Norway

Kristin E Thorud and Tore Haringstein1

1 Dr Thorud is with the Norwegian Animal Health Authorityin Oslo Norway and Dr Haringstein is with the NationalVeterinary Institute also in Oslo

History of ISA in Norway

Norway was the first country to experience thechallenge of ISA The first outbreak occurred in fall1984 in a hatchery in Hordaland County in thesouthern part of Norway Disease with similar clinicaland pathological signs of infection occurred the nextspring in farms that had received smolts from theoriginally affected hatchery In northern Norway thedisease was first recognized in 1988 but one casemight have occurred in 1986 that was not diagnosed

Within salmon farms the disease spreadrelatively slowly from one cage to another and itgenerally took about 1 month for the disease to showup in fish in adjacent cages Contagion ofneighboring farms usually required about 6 to 9months before clinical disease occurred The firstoutbreak in an area usually occurred in a farm acouple of weeks after medication against anotherdisease problem such as sea lice Hitra disease ortapeworm infestations If the first outbreak within afarm or area was initiated by some stress factor itwas difficult to stop the disease from spreading evenafter the specific stress was alleviated

Although there were outbreaks without obviousorigins spread into new areas was initially seen afterpurchase of infected smolts or in farms located nearslaughterhouses and processing plants In Norwayand in Hordaland County particularly the disease

Abstract Norway was the first country to experience thechallenge of infectious salmon anemia (ISA) After the firstoutbreak in 1984 the number of cases of ISA exploded in1989 and reached a peak of 80 new infected farms in 1990The disease became notifiable in 1988 when scientistsrealized that ISA was contagious and putatively caused bya virus Once the disease caused serious losses to fishfarming regulatory responses were developed in parallelwith increased scientific knowledge To eradicate ISA frominfected farms and affected areas control was mainlyexercised through depopulation of the affected farm andeventually the surrounding area To prevent introduction of

ISA into nonaffected areas controlling transfers of live anddead salmon between areas has been effective Within thelast few years there has been a resurgence in the numberof outbreaks There may be several reasons for this noneof which has been scientifically investigated These includestructural changes in the industry concerning transport ofsalmon for rearing or slaughter and enhancement ofnonspecific yet protective immunity afforded by vaccinationagainst other diseases Although vaccination was not analternative in 1990 it is currently considered to be a tool forthe control of ISA within affected areas

exploded in 1989 (fig 1) It is interesting to noticethat this happened the year after there had beenmassive towing of sea cages in all directions acrossthe Hardangerfjord which was done as an attempt toescape from a toxic algal bloom

Strategy for Controlling ISA

ISA became a notifiable disease within Norway in1988 as a list B disease when it was demonstratedthat ISA was contagious (Thorud and Djupvik 1988)This gave the authorities the ability to introduceregulatory measures to control outbreaks Still ittook many years before the virus was isolated(Dannevig et al 1995) and preliminarily characterized(Mjaaland et al 1997 Falk et al 1997) In theinterim ISA caused serious losses in the fish farmingindustry and the authorities had to develop a strategyto control it This strategy included general measuresadopted to prevent spread of virus and manageoutbreaks at farms The measures were based onknowledge of risk factors disclosed in fieldexperiences transmission trials and epidemiologicstudies The strategy was developed in parallel withincreasing scientific knowledge about the virus andrisk factors for disease

Two epidemiologic studies that were conductedin the first half of 1990s (Varinggsholm et al 1994 Jarpand Karlsen 1997) demonstrated that the risk for ISAwas significantly related to the location of the farmsand certain management factors The risk for fish ina sea farm to contract ISA increased by 88 to ashigh as 139 when a farm was located closer than

156


5 km from a location with ISA-infected fish Riskincreased by a factor of 146 when a farm waslocated closer than 5 km from slaughterhouses andby 38 for processing plants that did not disinfect offaland wastewater Risk from slaughterhouses wasconsiderably reduced after disinfection of wastewaterbut disinfection seemed ineffective on wastewaterreleased from processing plants

Management practices also increased the riskof contracting ISA on the farm For example thepractice of sharing staffs with other farms increasedthe risk for ISA by 07 to 39 and removing dead fishless frequently than once a day increased the risk by30 Mixing smolts from more than one hatcheryslightly increased the risk for ISA by 16 to 29 butreceipt of fish that were previously kept in other

seawater locations sometimes increased the risk forISA by 562

General Preventive MeasuresTo Reduce Spread of ISA

A connection was observed between the spread ofISA among fish farms and particularly to new areaswith purchase of infected smolts and with release ofuntreated water into the sea from nearbyslaughterhouses and processing plants Regulationstherefore targeted hygienic procedures at hatcheriesslaughterhouses and processing plants

To protect smolts produced in the hatcheries itwas considered important to ensure a pathogen-freewater supply Because ISA was only seen in fish kept

Figure 1mdashVerified outbreaks of infectious salmon anemia thatoccurred within Norway from 1984 until September 2002

1984 1987 1990 1993 1996 1999 2002

First occurrence of ISA

Regulations on transport (vehicles and hygienic measures)

Regulation on disinfection of wastewater from slaughterhousesand processing plants and of the seawater supply in hatcheries

80

70

60

50

40

30

20

10

0

ISA proven to be contagiousISA made notifiable

Introduction of zones to combat outbreaks

Official guidelines for dealing with outbreaks

Obligatory health certicateObligatory health control in hatcheriesBan on use of seawater in hatcheriesBan on moving fish already put in sea

157


in seawater disinfection of the seawater supply intohatcheries was introduced This measuresupplemented other general hygienic measures inthe hatcheries which required at least 12 regularhealth assessments per year These controlsfacilitated early detection of disease and alsoallowed the confident purchase of smolts from aparticular hatchery Disinfection of eggs is also partof these precautions

To protect salmon in sea cages nearslaughterhouses and processing plants theauthorities required that offal and wastewater fromsuch facilities be properly managed and disinfectedOther measures included a ban on movement of fishafter transfer to seawater and regulations for hygienicstandards and procedures on transport vehiclesparticularly wellboats Demand for daily uptake ofdead fish in summertime and every second day inwinter was introduced in the regulations as a generalprecaution for fish disease control

Measures To CombatOutbreaks

From the very beginning of ISArsquos history in Norwaythe disease tended not to disappear from affectedfarms until each site had been emptied of salmonand rainbow trout About 1990 it became obviousthat the disease could hardly be eradicated from thearea until the authorities and the farmers agreed on aprocess to empty all farms and fallow the whole areafor a designated period of time

Based on these experiences and theinformation gained from epidemiologic studies in1996 the authorities introduced official guidelines todeal with outbreaks of ISA Within the farms all fishin sea cages had to be killed or slaughtered if theirdaily mortality exceeded 1 fish per 2000 in the cageDead fish had to be removed daily and handledproperly The farmerrsquos choice of slaughterhouse hadto be approved by the local authorities who closelyreviewed the transport route for the fish to theslaughterhouse as well as the hygienic standardpracticed at the slaughterhouse

After each affected farm was depopulated itwas cleaned disinfected and then fallowed forusually 6 months before any restocking was allowedat the site

Measures Taken Within theZones

The official guidelines for dealing with outbreaks ofISA introduced both combat zones and observationzones around an affected farm From theepidemiologic study demonstrating particular risk forsites located closer than 5 km from ISA-affectedlocations a radius of about 5 km was used toestablish the combat zone According to theguidelines managers were not allowed to put smoltsinto any site located within the combat zoneAdditionally a farm had to be fallowed for at least 1month after it was depopulated and disinfectedbefore permission could be given to restock the farmwithin the zone

Intense official surveillance was conducted in allfarms giving priority to farms in the defined combatzone Furthermore official guidelines bannedtransports of smolts by wellboats closer than 20 kmfrom infected sites and transport of fish for slaughterhad to pass more than 5 km away from ISA-affectedfarms

ISA Situation in Norwaymdash2002

The number of outbreaks in Norway increaseddramatically in 1989 and reached a peak of 80 newlyaffected farms in 1990 Since 1993 the number ofoutbreaks decreased which suggested that theofficial control strategy was successful even thoughcurrent diagnostic tools were not available (fig 1)Within the last 4 years the number of outbreaks hasincreased again Several reasons have beenhypothesized for this reemergence but none of thesetheories has been reviewed scientifically

Certain areas have been subjected to moreoutbreaks than others Still the number of outbreaks

158


within the combat zones around the outbreaks hasbeen few compared to the number of outbreaksoccurring outside these zones (table 1) This factindicated that the measures in force at affected farmshave been relatively successful in preventing spreadto neighboring farms but the procedures to preventemanation of the disease spread from affected areashave been less successful

As a consequence of the introduction ofregulatory measures connected to hygienicprocedures of inlet water in hatcheries andwastewater in slaughterhouses and processingplants the first outbreaks in new areas these daysare rarely related to purchasing infected smolts or tobeing located close to slaughterhouses Transportsof smolts and fish for slaughter though have beensuspected as a probable route of transmission forspread of ISA Structural changes in the industryleading to long transports of large amounts of salmonfor rearing and for slaughter may be one factorUnspecific immunization due to vaccinationprocedures against other diseases may to someextent have allowed spread of virus with undetectedinfected fish

To reduce the risk for transports of infected fishthrough or to unaffected areas or transports ofuninfected smolts through affected areas there is aneed for early detection of infection Early detectionrequires reliable diagnostic tools but there must alsobe a stimulus for farmers to report suspiciousfindings This is especially important because controlof ISA is partly a question of confidence andcooperation between the authorities and the farmersStill strict regulations by the authorities are

necessary The decade that has passed since ISAwas a major problem for the industry gives thefarmers today little experience in dealing with ISAand a lower level of awareness about necessaryprecautions

Present Strategy

Every year since 1984 Norway has dealt withdifferent numbers of outbreaks of ISA This scenariosuggests that a control strategy allowing vaccinationagainst the disease may be more practical in Norwaythan a total eradication strategy Still Norway haslessened the impact of ISA because regulationscombined general preventive measures to reducedisease spread with an eradication strategy in andaround affected farms Structural changes in theindustry and in management procedures have led torecognition of the need to enforce control in certainsituations Therefore farmers and authorities inNorway have recently agreed to replace theguidelines for dealing with outbreaks of ISA with acontingency plan that in some respects goes evenfurther than the regulations enacted by the EuropeanUnion

The aim of the contingency plan is to furtherreduce the risk for spread of virus from affected farmsand areas to unaffected farms and areas Strictregulations of aquaculture activities related toaffected farms and the surrounding zones has beenof utmost importance At the farm level the principalchange in the new contingency plan requiresdepopulation of all fish within 80 days Controls ontransport of fish for slaughter and regulations onslaughterhouses will be strengthened Furthermorethe contingency plan proposes a ban on all transportof live fish through combat zones and observationzones Vaccination will be considered within thezones when several outbreaks occur in the samearea

Early diagnosis of infected fish is important forsuccess in the combat strategy against ISA Toenable farmers to detect ISA and to take necessaryprecautions to protect against it the authorities will

Table 1mdashOutbreaks of ISA in areas surroundingaffected farms (= within combat zones) compared tooutbreaks in unaffected areas (= outside combatzones)

1998 1999 2000 2001

Outbreaks within combat zones 2 2 2 6

Outbreaks outside combat zones 11 6 13 14

Verified outbreaks in total 13 8 17 20

159


plan a program to increase knowledge about signs ofISA risk factors and consequences of the diseaseamong the farmers and the fish health servicesEconomic compensation to owners of ISA-affectedfarms for losses due to official restrictions after anoutbreak of ISA would further encourage the farmersto report findings at an early stage Suchindemnification programs are under consideration bythe Norwegian government

Conclusions

In summary the Norwegian experiences with theregulatory responses in force support the principlesfor the strategy in force Todayrsquos regulationsemphasize general protective measures to preventthe spread of disease in combination with controlstrategies once outbreaks occur Althoughvaccination was not a choice in 1990 it is consideredwithin the framework of the new contingency plan asa tool to control ISA within heavily affected areas Acommon vaccination strategy is not yet the choice

References Cited

Dannevig B H Falk K Namork E 1995 Isolationof the causal virus of infectious salmon anaemia(ISA) in a long-term cell line for Atlantic salmon headkidney Journal of General Virology 76 1353ndash1359

Falk K Namork E Rimstad E Mjaaland SDannevig B H 1997 Characterization of infectioussalmon anaemia virus an orthomyxo-like virusisolated from Atlantic salmon (Salmo salar L)Journal of Virology 71 9016ndash9023

Jarp J Karlsen E 1997 Infectious salmon anaemia(ISA) risk factors in sea-cultured Atlantic salmon(Salmo salar) Diseases of Aquatic Organisms 2879ndash86

Mjaaland S Rimstad E Falk K Dannevig B H1997 Genome characterization of the virus causinginfectious salmon anaemia in Atlantic salmon (Salmosalar L) an Orthomyxo-like virus in a teleost Journalof Virology 71 7681ndash7686

Thorud K E Djupvik H O 1988 Infectious salmonanaemia in Atlantic salmon (Salmo salar L) Bulletinof the European Association of Fish Pathologists 8109ndash111

Varinggsholm I Djupvik H O Willumsen F V Tveit AM Tangen K 1994 Infectious salmon anaemia (ISA)epidemiology in Norway Preventive VeterinaryResearch 19 277ndash290

160


161

Regulatory Aspects of Infectious Salmon AnemiaManagement in Scotland

Regulatory Aspects of Infectious SalmonAnemia Management in Scotland

Alasdair H McVicar1

Introduction

Similar to other major salmon farming countries(Norway Canada the Faroes Chile the UnitedStates) Scotland has experienced an outbreak ofinfectious salmon anemia (ISA) In May 1998 thedisease was first detected in one Atlantic salmonfarm Principally associated with farm managementpractices the disease subsequently spread from thatoriginal source to another 10 farms that were widelydistributed throughout the Scottish salmon farmingindustry

Based on ISA experiences in other countriesparticularly Norway the risk of such an outbreakoccurring in Scotland and more generally in theEuropean Union (EU) had been previouslyrecognized Consequently in the national regulationsof both the EU and United Kingdom (UK)contingency measures already existed to managesuch a situation This paper describes the regulatorystructures in place within Scotland to

(a) Require notification of the national OfficialService for fish disease control of the occurrence ofthe disease

(b) Enable enactment of containment measures torestrict further spread

(c) Enable the introduction of managementmeasures and

(d) Ensure that management measures wereeffective before relaxation of controls

Abstract Infectious salmon anemia (ISA) was firstdetected in a Scottish salmon farm in May 1998 BecauseISA is a List I disease in European Union regulations theoutbreak was subject to immediate control measures by theCompetent Authority in Scotland responsible for aquaticanimal health measures This included the removal of allstocks from the affected farm the tracing of all contactsthe designation of infected and surveillance zones sur-rounding all foci of infection and an epizootiologic investi-

gation into the spread and possible sources of the infectionAlthough ISA had spread to 10 additional farms the controlmeasures that were implemented successfully removed thedisease from Scottish salmon farms within a year of thefirst outbreak without any subsequent recurrence Thedecision process for the management measures that wereimplemented was firmly based on principles of risk assess-ment

1Alasdair McVicar is with the Fisheries Research ServicesrsquoMarine Laboratory in Aberdeen UK He is on secondmentto the Canada Department of Fisheries and Oceans inOttawa presently

Role of Regulations in DiseaseManagement

In comparison with other approaches regulatoryintervention has a clearly defined but relatively limitedrole as a disease management tool Mostimportantly it can be directed only at restrictingfreedom of action of human activities that aresignificantly linked to a risk of disease occurrenceThe objectives of regulatory controls are usuallydirected at a common or long-term benefit and oftenhave a negative impact on an individual or restrictedsector of the population Realistically regulatoryintervention for significant diseases should beconsidered as an option only when other diseasemanagement measures are significantly less effectiveor are unavailable Legislation is effective andcredible only when it is based on scientificallyvalidated information when it can be practicallyimplemented and when its negative impacts areproportional to the benefits Whenever there isnoncompliance with a disease managementregulation these questions must be asked

Is there either a lack of conviction of the short- orlonger term benefits in which case an educationalrequirement is apparent or

Is the regulation wrong by not meeting theobjectives disproportionate in its effects or inferior toanother alternative in which case the regulation mayneed to be changed

As has been apparent from the ISA outbreak inScotland disease management regulations musthave flexibility to accommodate new or unforeseencircumstances Thus when some aspects of the EUregulations were believed to be impractical changes

162


were implemented in the immediacy of withdrawaluse of vaccination in disease containment andprovision of an option for compensation (CouncilDirective 200027EC Council Decision 2001572EC)

Role of Risk Assessment

Countries that are members of the World TradeOrganization (WTO) have rights and obligations withrespect to trade which are outlined as internationalstandards in the Sanitary and Phytosanitary (SPS)Agreement of that organization As the agencydelegated by the WTO to deal with animal health theOffice International des Epizooties (OIE) providesdetails of these standards relevant to aquatic animalsthrough the International Aquatic Animal HealthCode The Code provides guidelines that reduce therisk of spreading disease while facilitating tradeEthical standards and transparency are integral to theSPS Agreement the OIE International AquaticAnimal Health Code and national regulations

When controls are imposed that implementregulatory disease management the decisionprocess demands scientific credibility based on riskassessment The OIE International Aquatic AnimalHealth Code has now formally outlined thecomponents of risk analysis (viz hazardidentification consequences risk of an incidentroutine management measures available andcommunication) These factors have long beenrecognized as an integral part of many fish diseasemanagement schemes that extend from farmhusbandry practices to regulatory controls Thisincludes the UK Diseases of Fish Act 1937 1983which is one of the oldest and most comprehensivefish disease-management measures Although thisAct was developed to deal with the spread offurunculosis in wild salmonids within Britain it provedhighly adaptable for other disease risks associatedwith fish farming including ISA

It was therefore no surprise when ISA wasfirst detected in Scotland in May 1988 that the suiteof actions taken by the Competent Authority closely

followed the OIE recommended pattern (McVicar2001) Because data that supported a quantitativerisk assessment approach on ISA management inScotland were limited considerable emphasis wasinitially placed on qualitative assessments of thedisease in previously affected areas Considerablecooperation was obtained from regulators anddisease practitioners in both Norway and Canada inproviding the latest information on the detectiondiagnosis and epizootiology of ISA in their respectivecountries Such information proved to be invaluablein developing details of the management response toISA in Scotland However they were treated withsome caution because it could not be assumed thatthe disease would behave in the same way andpresent the same level of hazard in different affectedareas There were major differences in the localhydrography ecosystems and farming practicesAny of these could affect the transmission and effectsof infection Because such differences could alsoaffect outbreaks in different regions within Scotlandeach new incident of ISA or of suspected ISA in aScottish salmon farm was subjected to a reevaluationof the risk assessment by an ad hoc group in theCompetent Authority taking into account all availablerelevant information This response permitted thebest use of the relevant local experience whichenabled a progressive move toward a quantitativeapproach in risk identification and management

Hazard Identification andNotification

ISA comfortably met the main criteria of a diseasehazard suitable for control by regulations

It was of proven serious economic impact to farmedsalmon

Currently avoidance cannot be assured andavailable control measures do not prevent occurrenceof serious disease and

It has a restricted distribution or occurrence withinindustries and geographically a situation that can bemaintained by restrictions

163


These features are well documented for ISA in thescientific literature

Early official awareness of the occurrence ofincidents of a disease of concern such as ISA is acritical requirement that accompanies regulatorycontrol This is recognized by the OIE InternationalAquatic Animal Health Code where a compulsorynotifiable status is required for a disease to beincluded in a zonation system ISA is included by OIEin the list of ldquoother significant diseasesrdquo In the UnitedKingdom the risks from the disease were apparentfrom the situation in Norway during the 1980s andas a result ISA was made a UK-notifiable diseasein 1990 under the Diseases of Fish Acts 1937 1983Around the same time the risks to salmon stocksthroughout the EU were recognized and ISA wasmade an EU List I Disease in 1991 under CouncilDirective 9167EEC Contingency plans to deal withsuspected and confirmed outbreaks of ISA werespecified in Council Directive 9353EEC

These EU Directives were incorporated intoUK national law by enabling legislation in the FishHealth Regulations 1992 Diseases of Fish (Control)Regulations 1994 and the Fish Health Regulations1997 This suite of regulations provided theCompetent Authority with the ability to establish andmaintain Scotland as a zone free of ISA consistentwith OIE recommended standards With the outbreakof ISA in 1998 it was then possible to establish andmaintain separation between infected zonessurveillance zones and ISA-free zones by the samestandards Similarly following clearance of thedisease steps were taken to reestablish the ISA free-zone status to Scotland

Aspects of ISA ControlAvailable Under UKRegulations

Whenever EU Directives deal with an aspect of fishdisease management by regulation and enablinglegislation is introduced into the UK as is required bythe Directives these new regulations take

precedence over previous official controls Existingnational power(s) remain where an area in existingregulations is not covered by the EU regulations

The powers available to the CompetentAuthority of Scotland that were most critical to thesuccessful management of the ISA outbreak include

requirements for notification of suspicion orconfirmation of infection in any susceptible species

introduction of disease containment measures onaffected areas (infected farms affected zones andsurrounding surveillance zones) with clearance(depopulation) of affected sites

provisions for an epizootiologic study to investigatethe origin of the disease and its risk of transfer toother areas and

provision to require fallowing of infected sites for aspecified period before those sites can be restocked

Early Warning

For management of disease by regulation andto restrict its spread it is essential that any evidenceabout the occurrence of the disease even suspicion(both defined by the regulations) be brought to theattention of the Official Authority without delayInternationally it is a requirement of disease controlthrough zonation as in the OIE International AquaticAnimal Health Code that a disease must becompulsorily notifiable This principle is alsoembedded in both the United Kingdom and EUregulations The requirement to notify the officialauthority in the UK is described as immediate andfailure to comply is treated as an offense Forprofessional diagnosticians such an offense is likelyto be considered as unethical All susceptible fishspecies are covered by the regulations an importantpoint for emerging diseases because the full range ofsusceptible host species may be unknown

Prevention of Spread

Linked to immediate notification it is self-evident that when an infectious disease is discoveredor suspected in a location the more rapid practical

164


disease containment measures are implemented thebetter the opportunity to prevent spread When ISAwas detected in Scotland it was possible toimmediately place controls on affected facilitiesregarding movements of stock equipment personsvehicles (including boats) and dead fish Biosecurityprovisions at entrances to the facility were alsorequired Similar restrictions were also placed onother farms in the designated infected (control) andsurveillance zones surrounding the affected farmBecause ISA was a List I disease as defined inCouncil Directive 9167EEC (ie considered to beexotic to the EU) there was an obligation toimplement an immediate compulsory program ofdepopulation Depopulation involved immediateremoval of all stocks from affected farms in a mannerthat managed the risk of further spread of theinfection during processing or disposal of stocksThis proved difficult to achieve in large farms thatproduced several hundred tons of fish Thus CouncilDirective 200027EC amended this requirement to amore practical consideration in which all fish werewithdrawn from the affected facility in accordancewith a scheme established by the CompetentAuthority and approved by the Commission of the EU

At the time of the outbreak there was noprovision in the EU or UK regulations for directcompensation for any of the stocks that weredestroyed The possibility of indemnification waslater introduced through Council Decision 2001572EC

After total depopulation facilities and equipmentwere disinfected according to standards specified bythe Competent Authority As a possible additionaltool for containment of infection Council Directive200027EC lifted the total ban on the use of vaccinesagainst ISA and permitted such immunization as aderogation within part of a contingency plan for anoutbreak of ISA

Delineation of the Extent of theDisease

Risk of spread of the infection from the detectedsource of the disease was evaluated through an

immediate census of stocks on site Furthermore anevaluation of official stock movement recordsmaintained by each farm provided information onstock origins and shipments into and out of theaffected facility Through such an analysis it waspossible to ldquoring fencerdquo the ISA outbreak with respectto live fish movements by introducing furthercontainment measures as necessary

Information on possible sources of the diseaseand on the most important factors involved in thespread of the infection to other farms were obtainedfrom the epizootic study detailed results of which areprovided in Alexander Murrayrsquos article in this bookThese studies provided the data on the mostimportant areas where focus was required to supporta transparent decisionmaking process on both themanagement measures needed to control theoutbreak and ways to deal with possible further risksof infection

Removal of Infection Foci

To prevent recycling of ISA through different agegroups of farm stocks the Scottish regulationsprovided the authority to require the fallowing of anaffected farm after an outbreak of the diseaseHowever it is a difficult task to validate a negativepoint biologically Deciding to permit restocking of anaffected farm without risk of recurrence of theinfection is therefore difficult Qualitative riskanalysis on experiences elsewhere can provideinformation but it could not be presumed that thoseprocedures associated with ISA outbreaks in Norwayand eastern Canada were fully relevant to theScottish situation Regulatory powers existed inScotland to determine the period of fallowing beforerestocking a facility As described in Ronald Staggrsquospaper in this book the absence of recurringinfections in restocked farms after an outbreak onthese indicated that a 6-month fallow period wassufficient to break any possible recurrence of thedisease Whether a shorter fallow period would beequally as effective is unknown Based on anassessment of the infection risk (eg test results)and of the transmission risk (eg example the

165


proximity of other farms and traffic between farms) itwas possible to allocate different minimum fallowperiods in the control zones to farms where therewas some suspicion of ISA infection (as defined bythe regulations)and where there was no evidence ofinfection

Conclusions

Strong regulations to manage an outbreak of ISAexisted in Scotland and it was likely that the earlyimplementation of these regulations had largelycontributed to the elimination of this diseaseEradication of ISA from the Scottish salmon farmingindustry was accomplished within one year of the firstoutbreak as described in Staggrsquos chapter

References Cited

Council Directive of 28 January 1991 9167EECconcerning the animal health conditions governingthe placing on the market of aquaculture animals andtheir products Official Journal of the EuropeanCommunities No L 46 190291

Council Directive of 24 June 1993 introducingminimum Community measures for the control ofcertain fish diseases 9353EEC Official Journal ofthe European Communities No L 175 19071993

Council Directive 200027EC of 2 May 2000amending Directive 9353EEC introducing minimumCommunity measures for the control of certain fishdiseases Official Journal of the EuropeanCommunities 11428 1352000

Council Decision of 23 July 2001 amending Decision90424EEC on expenditure in the veterinary field(2001527EC) Official Journal of the EuropeanCommunities No L 20316 2872001

Diseases of Fish Act 1937 1983 revised to 1stDecember 1986 Her Majestyrsquos Stationery OfficeLondon

McVicar A H 2001 Practical application of riskassessment in the control of infectious salmonanaemia in Scotland In Rogers C J ed Riskanalysis in aquatic animal health Paris WorldOrganization for Animal Health 99ndash107

The Fish Health Regulations 1992 Edinburgh UKHer Majestyrsquos Stationery Office Edinburgh Press

The Diseases of Fish (Control) Regulations 1994Edinburgh UK Her majestyrsquos Stationery OfficeEdinburgh Press

The Fish Health Regulations 1997 London TheStationery Office Limited

166


167

Design and Implementation of anInfectious Salmon Anemia Program


Otis Miller Jr1

1Dr Miller is with USDAndashAPHISndashVS in Riverdale MD

Introduction APHISrsquoInvolvement in AquacultureBefore ISA

During the late 1990s APHIS received nearly twodozen petitions to promulgate animal healthregulations and regulatory programs for the control offarm-raised fin fish as livestock These petitionscame from State farm bureaus industry associationsindividual producers State officials and businessesthat serve aquaculture industries These petitionscovered dozens of issues and raised hundreds ofquestions Although these requests were broad inscope there was one underlying message thataquatic farmers should receive the same servicesthat domestic producers of livestock receive foranimals moving in interstate and foreign commerce

APHIS is authorized to promulgate regulationsto protect the health of livestock and poultry in theUnited States and manages regulatory programscovering poultry horses swine cattle and otherlivestock Additionally APHIS programs coveranimals that could transmit diseases or pests tolivestock Programs for terrestrial livestock areintended to (1) prevent the importation of diseasesand pests (2) regulate interstate movement in auniform manner (3) provide diagnostic laboratoryservices (4) regulate vaccines and biologic reagents

Abstract Late in 2001 the Secretary of Agricultureannounced an emergency threat to the US livestockindustry and authorized the transfer of funds from theCommodity Credit Corporation to establish an infectioussalmon anemia (ISA) program Effective December 132001 about $83 million was authorized for the USDepartment of Agriculture (USDA)ndashAnimal and PlantHealth Inspection Servicersquos (APHIS) Veterinary Services(VS) to implement an ISA control and indemnity programfor farm-raised salmon in the United States In addition tothe payment of indemnity these funds were designated toassist the State of Maine with program activities such asdepopulation and disposal cleanup and disinfectionestablishment of surveillance programs epidemiologic anddiagnostic support and training for producers and veteri-narians Maine officials have taken steps to prevent the

spread of ISA However Federal assistance was deemednecessary to control this threat to animal health and theUS economy effectively APHISrsquo goal is to control andcontain the ISA virus through rapid detection and depopu-lation of salmon that have been infected with or exposedto ISA virus (ISAv) ISAv can be controlled within high-riskzones through surveillance vaccination and effectivemanagement practices ISA control requires depopulationof all pens holding infected fish Indemnification is neces-sary to provide an incentive for salmon farmers to reportdiseased fish and to continue testing

Therefore State officials asked USDA to assist withthe epidemiology surveillance and indemnificationprograms This report will explain the events that led to thisemergency declaration and the current results of the newlyinstituted ISA indemnity and control program

used in animals and (5) control or eradicate diseasesand pests already found in the United StatesMoreover under the 2002 farm bill the Animal HealthProtection Act of 2002 APHIS has been granted thespecific authority to promulgate regulations to protectthe health of farm-raised aquatic animals

In response to the petitions received APHISconsidered expanding its services and programs toinclude farm-raised finfish Several services werealready being provided to the aquaculture industryincluding laboratory diagnostic work endorsement ofcertificates of inspection for export for aquaticanimals and aquatic animal products and licensureof vaccines and biologic reagents for use in aquaticanimals The agency also controls damage by wildbirds and other animals to farmed aquatic animalsUser fees and cooperative agreements with Stateand local governments and industry pay for some ofthese programs and service but expanding theagencyrsquos services and programs would requireadditional funding

To assess the specific needs of the aquacultureindustry and to determine what role if any APHISshould play in regulating this industry the agencyissued an advance notice of proposed rulemaking inthe Federal Register (64 FR 23795ndash23796) on May4 1999 entitled ldquoAquaculture Farm-Raised FinFishrdquo In that notice APHIS requested views andrecommendations from all interested persons onseveral specific issues First should the agency

168


consider regulating only domesticated farm-raisedfinfish or should it also consider regulating otheraquatic animals Further what additional species ifany should APHIS include in a potentially broaderprogram Second if APHIS expanded the range ofits services what new or additional services shouldbe provided Additionally should APHIS consideradopting specific regulations to prevent theintroduction of diseases and pests of aquatic animalspecies and if so should those regulations be similarto those already in place for poultry and terrestriallivestock Also what form would any potentialrulemaking take with regard to industry and Statecooperative programs Finally should APHIS adoptregulations to control the interstate movement ofaquatic species to prevent the interstate spread ofdiseases and pests If so should the agencyconsider including voluntary industry-drivenprograms to help producers control and eliminatedisease

APHIS received 55 written comments inresponse to the advance notice of proposedrulemaking Additionally to facilitate publicparticipation in the process APHIS held a series ofeight public meetings throughout the country in 2001They were well attended and brought together acrosssection of industry representatives State andlocal officials and other interested parties Meetingtranscripts are available on the Internet at httpwwwaphisudagovppdradaquaculturehtml

Currently APHIS is evaluating commentsreceived in response to the advance notice ofproposed rulemaking and the comments we receivedat the public meetings Once comment review iscomplete APHIS will publish a notice in the FederalRegister Certainly however the agencyrsquos ISAcontrol efforts indicate that preventing theestablishment of foreign animal diseases in aquaticspecies is a vital priority within its mission to protectUS agriculture

APHIS and Foreign AnimalDiseases

APHIS has regulations in part 53 of title 9 of theCode of Federal Regulations that provide for thecontrol and eradication of diseases including foot-and-mouth disease rinderpest contagiouspleuropneumonia exotic Newcastle disease highlypathogenic avian influenza or other communicablediseases of livestock or poultry that in the opinion ofthe Secretary of Agriculture constitute an emergencyand threaten the livestock or poultry of the UnitedStates The regulations authorize payments for thefair market value of the animals destroyed as well aspayments for the cost of their destruction anddisposal The regulations also authorize paymentsfor materials that must be cleaned and disinfected ordestroyed because of disease exposure

Although these diseases are included in theregulations because they constitute an emergencyand threaten the livestock or poultry of the UnitedStates they also pose a significant risk to countriesthat rely on the international trade of animals andanimal products To better facilitate internationaltrade of animals and animal products the UnitedStates is a member country represented at the OfficeInternational des Epizooties (OIE a world animalhealth organization) OIE is an internationalveterinary organization comprising at press time(April 2003) 162 member countries in which APHISndashVS serves as delegate and Chief Veterinary Officerfor the United States Through this relationshipAPHIS has mandates under the Agreement onSanitary and Phytosanitary Measures (SPSAgreement) of the World Trade Organization (WTO)and among other standards disease reportingrequirements to OIE Although ISA is classified byOIE as an ldquoother significant diseaserdquo it requiresreporting within 24 hours because in the UnitedStates it is a new finding with the likelihood ofexceptional epidemiologic significance

169


The Emergence of ISAmdashTimeline of APHISrsquo Action

The first case of ISA in the United States wasconfirmed in Maine on February 15 2001 ByDecember 2001 19 cases of ISA had beenconfirmed in 16 net-pen sites in Maine

On April 24 2001 the Maine Department ofMarine Resources the Maine AquacultureAssociation and the Maine State Veterinarianrequested that USDA provide the State withassistance in the areas of indemnificationepidemiology and surveillance for ISASubsequently APHIS entered into a cooperative ISAcontrol program with Maine to help safeguard thesalmon industry from future incursions of this exoticdisease and to monitor and manage the ISA status ofsalmonid aquaculture sites in that State

Effective December 13 2001 and published inthe Federal Register on December 20 2001 (66 FR65679ndash65680 Docket No 01ndash082ndash1) the Secretarydeclared an emergency because of ISA Theemergency declaration process has several levels ofclearances The time between the submission foremergency status and the actual declaration of theemergency for a new disease in a new animalspecies can be lengthy

The declaration of emergency was issued formany reasons The declaration considered ISArsquossevere economic threat not only to Mainersquos industrybut also to the viability and sustainability of salmonaquaculture in the entire United States Salmonproduction in Maine exceeds 362 million poundsannually with an industry value of $101 million(Maine Aquaculture Association 2001) In an attemptto control the ISA outbreak by the time of theSecretaryrsquos declaration of emergency the Statersquossalmonid industry had already voluntarilydepopulated about 900000 salmon worth nearly $11million This loss is even greater when capitalexpenditures such as labor costs and equipment areconsidered

Additionally the existence of ISA in Maine hasaffected other States well outside the threat ofimmediate disease transfer owing to its ramificationsfor international trade As a party to the WTO theUnited States must follow SPS Agreements Forexample when ISA emerged in Maine Chile and theEuropean Union prohibited the importation of troutand salmon eggs from Washington Maine Idahoand any other State that would export them Theresulting trade loss of salmonid eggs for the year2001 was estimated at $2 million

As a result of the Secretaryrsquos declaration ofemergency about $83 million was transferred fromthe Commodity Credit Corporation for theDepartmentrsquos ISA control and indemnification effortsfor fiscal year 2002 which ended September 302002 ISA Funding for fiscal year 2003 would have tobe requested through the Office of Management andBudget

The Interim RulemdashThe ControlProgram and Indemnity

After the declaration of emergency APHIS beganwork to create a successful ISA control program thatwould gain the support of salmon producers OnApril 11 2002 APHIS published an interim rule (67FR 17605ndash17611 Docket No 01ndash126ndash1) that waseffective retroactively to the Secretaryrsquos declaration ofemergency dated December 13 2001 The interimrule provided for among other things indemnificationfor fish depopulated because of ISA on or after theDecember 13 2001 declaration of emergencySince then slightly more than $61 million has beenspent on indemnity cleaning and disinfection

Additionally the interim rule amended thedefinition of ldquodiseaserdquo which is found in section 531of title 9 of the Code of Federal Regulations toinclude ISA among the other listed foreign animaldiseases Additionally APHIS added other definitionsto that section that help explain program elementsFor example a definition for the term ldquoISA ProgramVeterinarianrdquo was added to identify the veterinarianas the person assigned to be the main point of

170


contact at the field level to manage the ISA programfor APHIS in Maine

The interim rule also provided requirements foreligible producers to receive indemnity funds whichare described in more detail next

Finally the interim rule evaluated potentialeconomic consequences of the ISA control programin Maine Because the entire farmed Atlantic salmonindustry in Maine is at risk if ISA is not controlledAPHIS determined that the $100-million-dollar-per-year industry outweighed the cost of the programSpecifically APHIS anticipated potential benefits tothe control and indemnity program First indemnityprovides producers who have not been participatingin ISA control with an incentive to do so AdditionallyAPHIS recognized that a more aggressive approachearly on although the number of affected sites wasrelatively small might obviate the need for higherfuture Federal costs to contain a more widespreadoutbreak Additionally the agency expected that thecontrol and indemnity program might also reduce theimpact of trade restrictions due to ISA This reductioncould produce third-party trade benefits bydemonstrating to trading partners the intent andability of the United States to protect its animalindustries thus enhancing US ability to negotiateaccess to foreign markets

Infectious Salmon AnemiaProgram Standards

The ISA program standards establish procedures forthe prevention and containment of ISA from farm-raised Atlantic salmon These standards which werecodeveloped by the ISA Technical Board andapproved by APHISndashVS Mainersquos Department ofMarine Resources and the Maine AquacultureAssociation provide industry accreditedveterinarians approved laboratory personnel andregulators with a template to follow to meet the goalsof this Federal program Indemnity payments anintegral part of this program will be made toproducers only if the established procedures andstandards are followed However as described in the

interim rule these indemnity funds are provided onlyfor a 2-year period during which time the industryshould establish and implement a self-compensationindemnity fund based on a certain financial amountset aside per fish produced Further these standardsare amendable as science and circumstances evolve

Program Requirements

To receive payment for ISA indemnity claimsproducers must at a minimum observe the followingrequirements

1 Establish and maintain a veterinary clientndashpatientrelationship with an APHIS-accredited veterinarianand inform the ISA Program Veterinarian in writing ofthe name of their accredited veterinarian at the timethe participant enrolls in the ISA program and within15 days of any change in accredited veterinarians

The farmed salmon industry uses highlyqualified personnel experienced in all aspects of fishculture husbandry and health managementAlthough most industry members have establishedinhouse procedures for increased diseasesurveillance and a working relationship withaquaculture veterinarians and diagnostic laboratoriesto provide further technical expertise thisrequirement will ensure that all participants haveready access to an APHIS-accredited veterinarianwho will conduct the surveillance testing andreporting activities discussed in the next paragraphand will assist participants in carrying out the otheraspects of the program discussed later

2 Cooperate with and assist in periodic onsitedisease surveillance testing and reporting activitiesfor ISA which will be conducted by the producersrsquoAPHIS accredited veterinarian or a State or Federalofficial as directed by the ISA Program Veterinarian

Surveillance ensures that resources and theattention of producers will be directed at routine andregularly scheduled inspections and healthassessments of fish so that ISA will be quicklydiagnosed Testing with the best and mostscientifically sound assays at an approved laboratorywill ensure prompt and accurate diagnosis

171


Reporting procedures ensure that once infected ordiseased fish are identified control measures anddepopulation can proceed rapidly

3 Develop and implement biosecurity protocols foruse at all participant-leased finfish sites andparticipant-operated vessels engaged in aquacultureoperations throughout Maine A copy of theseprotocols shall be submitted to the ISA ProgramVeterinarian at the time the participant enrolls in theISA program and within 15 days of any change in theprotocols

The implementation of effective biosecurityprotocols will reduce the risk of introducing andspreading ISA into and between marine sites andcages by movement of farmed fish equipment andpeople

4 Develop with the involvement of the participantrsquosaccredited veterinarian and the fish site healthmanager a site-specific ISA action plan for thecontrol and management of ISA A copy of the actionplan shall be submitted to APHIS for review at thetime the participant enrolls in the ISA program andwithin 15 days of any change in the action plan

The action plan is a document developed foreach site that defines the response contingencies forISA eg activities to be undertaken upon diseasedetection notification procedures etc) should thedisease emerge at the site

5 Participate in the State of Mainersquos integrated pestmanagement (IPM) program for the control of sea liceon salmonids A copy of the management plandeveloped by the participant for the State IPMprogram shall be submitted to APHIS for review atthe time the participant enrolls in the ISA programand within 15 days of any change in the managementplan

Sea lice are copepod arthropods belonging tothe genera Lepeophtheirus and Caligus Species ofboth genera infest Atlantic salmon and live in themucus layer where they attach and suck blood orcause sores The larger Lepeophtheirus species aregenerally regarded as capable of transmitting ISASea lice of both genera can cause stress on fish

which adversely affects the immune response TheMaine IPM program for sea lice provides formonitoring treatment and management practicesdesigned to minimize the presence of sea lice in pensites and reduce the need for the use of chemicalsand medications APHIS considers control of sea liceto be a vital component of the ISA control program inMaine therefore the agency will in cooperation withthe State of Maine review and verify the adequacy ofeach participantrsquos sea lice management plan

6 Submit to the ISA Program Veterinarian at thetime the participant enrolls in the ISA programcomplete and current fish inventory information foreach participant-leased finfish site with site and cageidentifiers Fish inventory information must includethe numbers age date of saltwater transfervaccination status and previous therapeutant historyfor all fish in each participant-leased finfish site

This information will provide APHIS with thedata necessary to establish disease control actionscomplete epidemiologic assessments and increaseits ability to monitor fish populations effectively

7 Maintain and make available to the ISA ProgramVeterinarian upon request mortality data for eachparticipant-leased finfish site and pen in production

This can be accomplished using existingindustry records systems and log sheets Themortality data will be used by APHIS in conjunctionwith the fish inventory information discussedpreviously to establish disease control actionscomplete epidemiologic assessments and increasethe agencyrsquos ability to monitor fish populationseffectively

8 Cooperate with and assist APHIS in thecompletion of biosecurity audits at all participant-leased finfish sites and participant-operated vesselsinvolved in salmonid aquaculture

These audits will be performed to assess theefficacy of the biosecurity protocols established bythe participants to reduce the risk of introducing andspreading ISA into and between marine sites andcages by movement of farmed fish equipment andpeople

172


Program AdministrationISA Technical BoardmdashThe ISA Technical Boardconsists of seven individuals the USDAndashAPHISndashISAProgram Veterinarian two Maine Department ofMarine Resources (DMR) representatives threeindustry representatives appointed by the MaineAquaculture Association and a chairperson Theboard makes recommendations to the USDAndashAPHISndashArea Veterinarian-in-Charge (AVIC) ISAProject Manager or the Maine DMR Commissioneror all three to consider positive laboratory resultsepidemiologic data audit reports or other informationregarding reported disease risks or conditionsrequiring action under the terms of this program Aquorum will be four people (at least two industry andtwo regulatory agency representatives) for any givenmeeting or conference call The chairperson willserve as facilitator and vote only in the case of a tie

ISA Technical Board Committee Members

Chair Michael Opitz University of MaineStephen Ellis USDAndashAPHISndashVSPaul Waterstrat Maine DMRAndrew Fisk Maine DMRNils Steine Atlantic Salmon of MaineDan MacPhee Maritime Veterinary ServicesJulia Mullens Heritage Salmon

Program ProgressmdashThe Science of ISAControl

Surveillance Diagnostic Sampling and TestingmdashAs provided in the Program Standards participatingproducers must submit to diagnostic testingprocedures Currently surveillance sampling isconducted by APHIS on a monthly basis at high- andlow-risk zone net-pen sites throughout the Gulf ofMaine With the assistance of the National MarineFisheries Services which takes wild fish samples inthe Gulf of Maine and the US Fish and WildlifeService which conducts a national wild fish healthsurvey routine sampling among wild and culturedsalmon and nonsalmonid species is providing solid

evidence of the status of ISA and overall aquaticanimal health status in the United States

Sample Specifications

Table 1 lists sample specifications for the ISAtests to be performed the number of fish per poolthe required tissues and the collection vessel usedfor each test

Identification of ISAv

To ensure that surveillance and diagnosticprocedures are standardized throughout the industrythe following diagnostic methods to determine thepresence of ISA are acceptable histopathologyvirology RTndashPCR IFAT and gross pathology

All official diagnostic tests must be performedby a USDAndashAPHIS-approved laboratory All ISAdiagnostic test procedures will be performedaccording to the diagnostic test procedures approvedby APHIS Reasonable confirmatory evidence of ISAis based on clinical and postmortem findings inaccordance with criteria described in section 21 ofthe ISA chapter in the OIE Diagnostic Manual for

Table 1mdashSample specifications for ISA Program-approved tests1

Tissue Collecting vessel Test sample required preservative medium

Histo- Kidney spleen Tube10 formalinpathology liver gill

pyloric cecae

RTndashPCR2 025 cm3 midkidney Screw cap 2-mL tubeRNA

IFAT3 Midkidney im- Frosted end slidepression

Virology Kidney spleen Specimen cupphosphate-in cell gill pyloric cecae buffered salineculture

1 All tests are for single fish in a pool

2 RTndashPCR = reverse-transcriptasendashpolymerase chain reaction

3 IFAT = indirect fluorescent antibody test All tests were performedaccording to the OIE Diagnostic Manual for Aquatic Animal Diseases

173


Aquatic Animals and the ISA virus is detected byperforming the diagnostic tests mentioned aboveaccording to the OIE Diagnostic Manual for AquaticAnimal Diseases and other references establishedand filed with APHIS

Cleaning and Disinfection

Cleaning and disinfecting net-pen cages andassociated structures are a vital part of the ISAcontrol program APHISndashVS had all 1107 net-pencages primary nets and predator nets cleaned bypower washing scraping steaming or all threemethods (table 2) Nets were removed from sitesand cleaned at an authorized facility As provided inthe interim rule the agency paid the farmers forcleaning and disinfection costs where VS providedoversight of the process Table 2 details nets cleanedand disinfected under APHIS supervision

EradicationndashDepopulationndashRestockingUpdate of Salmon in Maine

Currently there is no disease outbreak of ISA inMaine and routine monthly surveillance sampleshave all tested negative As of this writing 17 sitesholding 1561005 fish were depopulated Of these17 8 were depopulated under the ISA Program and 9were depopulated voluntarily before the December 13

Table 2mdashCages and nets disinfected as of May 5 2002(N = 738)

Cages Nets Total Type and size Number Primary Predator

12-m2 steel 46 46 46 138

15-m2 steel 140 140 140 420

20-m2 steel 12 12 12 36

24-m2 steel 40 40 40 120

70-m polar 65 65 65 195

100-m polarcircle 66 66 66 198

2001 declaration of emergency All infected andexposed sites in the high-risk zone in the Gulf ofMaine were fallowed for up to 90 days prior torestocking Of those 17 sites 6 have been restockedwith a total of 1800320 smolts and all of thosesmolts have been vaccinated against ISA

A report of this current ISA update wassubmitted to OIE for redesignation of the UnitedStates as an ISA-free country on September 5 2002The United States is again officially ISA free

Acknowledgments

This report could not have been written had it notbeen for the efforts of many people from StateProvincial Federal and industry organizations allworking for a common cause related to the control ofISA I want to acknowledge the individuals that I amaware of who contributed to the emergencydeclarationrsquos becoming a reality and to theestablishment of the ISA indemnity and controlprogram

USDA Secretary Ann Veneman

APHIS Administrator Bobby Acord

(former) VS Deputy Administrator Alfonso Torres

Commissioner George Lapointe Maine DMR

Paul Waterstrat Maine DMR

Senator Olympia Snowe

Congressman John Elias Baldacci

AVIC William Smith

ISA Program Manager Stephen Ellis

Economist John Green

Budget Analyst Toni Harris

Regulatory WriterEditor Elizabeth Barrett

174


References Cited

Maine Aquaculture Association 2001 Infectioussalmon anemia action plan Hallowell ME MaineAquaculture Association Salmon Industry FishHealth Committee (Caroline Graham chair) 9 p

US Department of Agriculture Animal and PlantHealth Inspection Service July 2001 Infectioussalmon anemia Tech Note Washington DC USDepartment of Agriculture Animal and Plant HealthInspection Service 1 p

US Government Printing Office 2002 Code ofFederal Regulations title 9 part 53 vol 67 70 April11 Washington DC US Government Printing Office17605ndash17610

US Government Printing Office 2001 Declaration ofemergency because of ISA In Federal Register vol66 245 December 65679mdash65680

Additional References

Animal and Plant Health Inspection Service 2001Infectious salmon anemia Maine USA March 232001 Impact Worksheet June 25 2001 RiverdaleMD US Department of Agriculture Animal and PlantHealth Inspection Service 5 p

US Department of Agriculture Animal and PlantHealth Inspection Service 2002 Infectious salmonanemia program standards Washington DC USDepartment of Agriculture Animal and Plant HealthInspection Service 52 p [Final draft April 30 2002]

175

Presentations From the Open Forum

Dr David ScarfeAmerican Veterinary Medical AssociationRichmond VA

Good afternoon Irsquom David Scarfe and I will bemoderating this last session I hope that this willbring together many ideas and thoughts of manypeople with ISA experience I think what wersquoveexperienced in ISA outbreaks over the last few yearsis very unfortunate not only in one country but amultitude of different places throughout the worldBut countries have now responded stronglyaquaculture which has always had a tendency to bethe stepchild of agriculture is now getting nationalrecognition or equivalency

This afternoonrsquos session will be brief to thepoint Wersquoll try to keep a lot of the session as focusedas possible but it may overflow into numerous topicsThe intention for this afternoon is to try to focus onsome innovative ideas about responding to ISA

Obviously some of this concept will apply toresponses to other aquatic diseases But Irsquod like youto focus on three elements of the responseprevention control and eradication

In looking forward utility should be a practicaltheoretical and problematic approach for us toconsider in the future

Wersquove asked everybody wanting to offercomments for discussion to register a brief topic oncards which have been grouped in an attempt toprovide some structure and continuity Initiallyregistrants will be called on to make a briefpresentation followed by comments from the floorbefore moving onto the next topic If needed anoverhead projector will be available to illustratepoints

A court reporter will capture the salient pointspresented today to be added to the publishedproceedings after editing for clarity I invite JillRolland to initiate the first topic for discussion


Note The first time each speaker addressed the forum we are listing his or heraffiliation in short form For complete contact information please see appendix 2in the back matter

Ms Jill RollandUSDAndashAPHIS Veterinary ServicesRiverdale MD

On the west coast of the United States where wehave quite a bit of Atlantic salmon aquaculture inaddition to five native species of Pacific salmonmany people are interested to know if our Pacificsalmon stocks will be at risk should infectious salmonanemia make its way to the west coast The USGeological Surveyrsquos lab in Seattle is fortunate to havea BL-3 lab a biosafety level 3 containment lab whichallows us to work with exotic pathogens Thisenabled us to perform experiments testing thesusceptibility of Pacific salmonids to the ISA virusThe species used in our experiments were chumchinook coho and sockeye and for a positive controlwe used Atlantic salmon which were flown in fromthe east coast Two strains of ISAv were used theNorth American strain CCBB and the Norwegianstrain Bremnes

The experiment was split into two trials Thefirst trial was carried out in the fall using threedifferent viral doses of the Bremnes strain Duringthe experimental period we sampled the mortalitiesfor virus isolation and histopathology in addition torandom samples taken from all species at specifiedtime intervals

The mortality rate in the Atlantic salmon waslower than expected so we began a second trial inFebruary As only a limited number of fish remainedwe limited our trial to using only one dose of virus ahigh dose and we used both the north American andBremnes strains of virus Again we sampled bothmortalities and took random samples throughout theexperimental period

In both trials all Pacific salmon mortalitiesoccurred within 2 days after experimental challengeThese fish had no signs of disease as expected andprobably died from stress related to the challengeThe Atlantic salmon positive controls had low-levelmortality in trial 1 but mortality reached 96 to 100percent in trial 2 Therefore it would seem that

176


Pacific salmonid species are relatively resistant toISAv as compared to Atlantic salmon

Whether or not these species can function ascarriers and whether or not the virus is able toreplicate in these species was not determined norwas it the goal of this study Instead we wereattempting to answer the question of whether or notPacific salmon are susceptible to the disease ISAand based on our laboratory study it would seemPacific salmonids are resistant We were able to getsome low-level titers from some of the Pacific salmonat different time points and whether wersquore reisolatinga virus that we had already injected or wersquore actuallyseeing virus replication has not been determined

Dr David Scarfe

Thank you If no one has comments related to thattopic wersquoll go on to the next topic Ron Stagg wouldlike to make some comments on management of ISAand brood stock farms

Dr Ron StaggFisheries Research Servicesrsquo Marine LaboratoryAberdeen UK

The issue I want to talk about is how we mightmanage ISA if it occurs on a brood stock farm Thecontext of that discussion is that in Scotland at leastwe now have some different ways of managing broodstock Some people grow the brood stock in cagesand then transport eggs into fresh water sometransport fish into fresh water and we also have anumber of specialist brood stock farms in land-basedsystems What would we do if we had ISA on thosefarms The risk of spread of ISA is high if there is arisk of horizontal transfer to smolts On the otherhand these farms are the future of the industry in thesense that theyrsquore developing and refining the broodstock and performance So we have this paradoxthese are very valuable fish but if they contract ISAthey are also a high risk for spreading the disease

Have people thought about this in terms ofmanagement The Royal Society of Edinburghreport on ISA raised this issue in Scotland and Iwonder what experiences are in other countries andhow they might proceed to manage their situation

Dr David Scarfe

Are you suggesting something like specific pathogenfree (SPF) certification such as in some otheragriculture industries for example the swineindustry

Dr Ron Stagg

Yes possibly pathogen free Maybe those farmsneed to have much higher security so that theyprevent the possible ingress of ISA in the farms Butof course thatrsquos a very expensive option

Dr Paul MidtlyngVESO Veterinary ResearchOslo NO

You probably remember that there has been onecase before of ISAv in Norway where the diseasewas present in a broodfish farm At the time thedisease was not notifiable so there were noinstruments in place to prevent those eggs frombeing sold What happened was that doubledisinfection procedures were applied beforedispatching the fertilized eggs This was in 1987 isnrsquotthat correct I donrsquot think that any epidemiologicevidence has since come up to suggest that therewas a spread of infection from those positive fish

Your lab Ron has recently publisheddisinfection trials with the ISA virus I believe it waslast year [2001] So in the present case given thehighly particular value of brood fish populations Iwould let those live I would question the justificationof culling down superiormdashhighly superiormdashbroodstock for a disease like ISA There are other meansof control which to some degree could remedy theproblem

177


Dr Ronald RobertsUniversity of IdahoTwin Falls ID

Irsquom from the University of Idaho and Irsquom also directorof Landcatch Ltd which is the largest independentland-based brood stock farm I think in the world Icertainly would like to comment on this because ofconcerns which we have in Landcatch havinginvested pound5 or pound6 millionmdashthatrsquos $10 millionmdashinto thedevelopment of highly selected brood stock only tofind that the authorities insisted upon smolts beingrequired to be slaughtered irrespective of genotypefrom special lines just for environmental surveyswithout any validity to distinguish between the eliteones of which we are very much concerned and thenonelite And so I am very grateful that Ron Stagghighlighted this as a problem and also that Ron istaking it very seriously

The future of the industrymdashas with poultry andas with pigsmdashis going to depend upon the quality ofbrood stock and specifically pathogen-free broodstock And if we are to be continually at risk from acondition which in my view is endemic (and I knowRon doesnrsquot agree with that) If this virus stays inour natural environment and if we could be underextreme sanction if at any time it happened to befound in our waters or in a single fish then nobody isgoing to invest in the development of brood stock

It has been argued that we should disinfect ourcommon water We have 6 tonssecond of seawatergoing through our farm I donrsquot believe that a methodfor disinfecting those volumes of water exists even inthe US

So I feel this is a very high-priority concern andI would definitely agree with Dr Midtlyngrsquos statementthat there is no evidence that this is anunmanageable disease All our eggs are tripledisinfected before they get to the customer And thesituation with regard to the importance of developingbrood stock is such and the nature of the ISAdisease is such that we cannot make the statementthat we must put high-value brood stock at risk for adisease like ISA That should be the message

Mr Sebastian BelleMaine Aquaculture AssociationHallowell ME

My topic is somewhat related to this issue becausethe strategy that our growers have chosen to takewith respect to brood stock is no pun intended not toput all of our eggs in one basket They have decidedto move brood stock to land-based holding farms andto locate those brood stock at more than one facilityso that if a single facility is positive at least you havethe lines maintained in other facilities Howeverthatrsquos a very expensive process and we are far fromachieving that at this stage of the game Itrsquos a long-term goal on our part

I do have some questions though A number oftimes during the symposium there were references tothe detection of ISA in fresh water There were acouple of different instances where it was shown andI would like some clarification If this relates toholding brood stock in land-based facilities as wellwhat do people really know about ISA in fresh waterAre there in fact good verified instances of positiveclinical ISA in fresh water Are we talking about justpositive disease tests And if thatrsquos the case howdoes that relate to how we might manage thedisease as it were in our brood stock programsthemselves How do you use that to youradvantage

Dr Ron Stagg

In the fresh-water work wersquove done at the FisheriesResearch Servicesrsquo lab in Aberdeen wersquove neverobserved clinical disease in fresh water nor have weisolated ISAv All of our positives have beendetermined by RTndashPCR I think we have todifferentiate here between a background sporadicinfection and the risk of a potential disease outbreakThe latter will depend on the form of the pathogenand the sustained nature of the infection

In relation to brood stock farms I would like toask Mr Belle whether they separate brood-stockholding areas from where they are growing smoltsthat come from the eggs from those brood stock

178


I ask this in the context of the risk that might bepresented in the event of an ISA outbreak and thepossibility of cross-contamination in the event ofthese being shipped to ongrowing farms

Mr Sebastian Belle

Well we have different strategies in different facilitiesand I think everybody accepts the fact that you haveto try as much as you can to separate the broodstock from obviously the incubation of the smoltfacilities We have an added wrinkle in our case andI think this is also the case in Scotland to someextent we also have farms that are maintainingbrood stock in marine sites and those sites inparticular are obviously at great risk We are veryconcerned about them Thatrsquos one of the reasons wewent to the strategy of trying to divide the linesbetween different facilities

But I absolutely support Ron Robertsrsquo point thatthe volume of water that is coming into the land-based facilities is enormous and to ensure 100-percent disinfection of that incoming water is veryvery difficult

I mean the real world often does not perform tothe same performance standards that equipmentmanufacturers tell you they do I think itrsquos difficult toachieve that unless you build in multiple systems andthat causes you to have very high capitalinvestments

Dr Sandi McGeachyNew Brunswick Depratment of AgricultureFisheries and AquacultureFredericton NB

In New Brunswick I think we have a number of land-based brood stock facilities now in place where theeggs and offspring havenrsquot tasted saltwater Andwhat these facilities are doing is recirculating thewater using low water flow pathogen-free waterWhat they do with their siblings is send those out tosaltwater to test their performance at sea and theyselect their land-based siblings to obtain the betterperformance

I know you donrsquot get the same genetic gain asyou would if you actually selected the actual fishperforming in the saltwater but within the brood-stockprogram in New Brunswick thatrsquos how the plans arefor training in a biosecure source

Again this is spread across three facilities inNova Scotia one in Prince Edward Island and a fewin New Brunswick

I know Roland Cusack is here He might beable to speak about what is happening in the NovaScotia industry Those plans started a number ofyears ago so we are now at the point of becomingself-contained or self-sufficient

Dr Peter MerrillMicro Technologies IncRichmond ME

Keep in mind that there are no theoreticalimpediments to ISAv protection in fresh waterbecause in effect any experiments in fresh waterhave demonstrated that fish can become infected andcome down with clinical disease Irsquom still not awareof any published reports although I think [someinvestigators in] Norway reported some RTndashPCRproblems

In Maine with the majority of Maine companiesfollowing the ISA and the RTndashPCR certificationprotocols all brood stock are completely tested forISAv at the time of spawning The eggs are doubledisinfected and before they are released fromquarantine in a certified facility the progeny fromthose eggs are retested for ISAv and then testedagain before theyrsquore transferred to the mainland Sotherersquos a fair amount of redundant testing involvedThis doesnrsquot complete the testing but it doesseparate a particular type brood stock from blood-sampled brood stock

Dr David Scarfe

Thank you If we can move to the next topic AlasdairMcVicar would like to comment on the current ISAsituation in Ireland

179


Dr Alasdair McVicarCanada Department of Fisheries and OceansOttawa ON

Irsquom here to speak about ISAv in Ireland at the requestof the Competent Authority there because they wereunable to attend They have isolated ISA virus on theSHKndash1 cells and confirmed it by using IFAT and PCRThey used laboratories in Dublin (the Institute) theMarine Laboratory in Aberdeen and the CommunityReference Laboratory in Aarhaus A report was sentto the OIE Reference Laboratory in Oslo

The Irish particularly wanted to acknowledgethe cooperation obtained from these labs andespecially Tore Haringstein in Norway

The date of suspicion was July 5 2002 and thedate of confirmation August 1 All aquaculturefacilities in Ireland have been tested for the presenceof ISA using virus-isolation

Irsquom just looking through their report to theStanding Veterinary Committee of the EC here tosummarize the main points The source of the virusis unknown and the two affected rainbow trout farmsboth in seawater are isolated from other farms inIreland The authorities do not know when thedisease was introduced Strict biosecurity measureswere in place in both sites on July 5

At both sites control zones were establishedaccording to the Directives An effluent filtrationsystem was installed in the processing units withblood-water disinfection processing waste disposedof by rendering The number of the fish in the twosites involved is 140000 at an average weight of 25kilos in one and at the second site 220000 at 400grams and another 34000 at 14 kg One site hasfish of mixed sizes The other site with market-sizedfish has actually been harvested

Currently [September 2002] at the second sitethey are doing a trial involving testing thepathogenicity of the isolated virus in Atlantic salmonUntil those tests are completed the fate of thesmaller fish on that second site is uncertain

For this ISAv problem there are continuedhigher levels of surveillance of the still-stockedaffected salmon farm and the nearest salmon farmsThe way I see it they are not finding any evidence ofISAv or disease in any of the farms

Dr David Scarfe

One question related to that Is what you weredescribing specifically confined to salmon

Dr Alasdair McVicar

No The tests were performed on troutOncorhynchus mykiss only on trout There were nosalmon on these sites These were two seawaterrainbow trout farms

Dr David Scarfe

Did the animals show clinical signs and do you haveany idea what led someone to look for the ISA virus

Dr Alasdair McVicar

The fact that ISA was found in Scotland generatedconcern in Ireland and subsequent study at a highofficial level That started the surveillance and Irishauthorities have since been doing this on a routinebasis So this culture the presentation of virus iswhat came up from that

Dr David Scarfe

Would anybody like to speculate with respect toSebastian Bellersquos question concerning the presenceof ISA in fresh water environments and any possibleassociation between ISA in salmon and trout If notwe will move on to a countryrsquos perspective ofresponding to the threat of an outbreak of ISA IainEast has asked if he could make some commentsabout the Australian contingency plan for a potentialISA outbreak

180


Dr Iain EastAgriculture Fisheries ForestrymdashAustraliaCanberra AU

Irsquod just like to make some brief comments First off inour case itrsquos prevention and the steps wersquove taken todo that Then second Irsquoll very briefly touch on someof the issues that were raised during the last day anda half

In Australia wersquove taken a proactive approachto try and deal with these things before they happenItrsquos interesting to hear that Canada is currently tryingto develop an aquatic animal health plan We havehad one in place for the last 4 years Itrsquos a plan thatwe have deliberately used to get industryrsquosinvolvement because as you all know if you donrsquottake industry along with you plans are not worth thepiece of paper theyrsquore written on

Some of you may have picked up a copy ofAQUAPLAN at the American Veterinary MedicalAssociationrsquos table earlier this week If you didnrsquotcopies are available on our Web site or if you giveme your business card I can mail you a copy of theplan It is designed to cover all aspects of health asshown there and the reason is that if you donrsquotaddress all of those issues then there will be gaps

AQUAVETPLAN is our emergency managementresponse plan Itrsquos a series of manuals that coverAustraliarsquos predetermined response plan and how werespond to a range of diseases To some extent itrsquosstill under development but there are a number ofcomponents Again there were some CDndashROMs onthe AVMA table earlier this week and if you got oneof those congratulations I think there were only 10 Ican send those to you or you can download thecontent from our Web site

We have a plan under development for ISAwhich includes a full scientific review of the diseasethe principles of control and eradication and thepreferred control policy in Australia When this isfinished it will be endorsed by all sectors ofgovernment and by the industry

The value of this is that if we get the diseasewe can pull the manual off the shelf and we knowwhat wersquore going to do We donrsquot want to spend daysor weeks arguing with industry and governments asto what wersquore going to do

Many of you will be aware of the import riskanalysis for salmonid product and the dispute that wehad with Canada I donrsquot want to take sides butquarantine is one aspect of our policy as is thesurveillance program that we have up and runningand this predetermined response plan

Switching subjects a bit I would say that acouple of things Irsquove heard today and yesterday are abit worrying They may be for your country ahistorical development but Irsquod just like to give you a30-second overview of how the management of ourpearling industry works The major component iswhat we call the ldquo5 and 2 rulerdquo which means that notwo farms can be closer 5 km together unless there isa specific written agreement between the two farmsthat covers health management and then we willallow them no closer than 2 km Having salmonfarms 300 m apart as mentioned in a previous talkmdashhow can I say it without starting an incidentmdashitrsquosless than optimal

In the Australian pearling industry we also havepredetermined zoning Brood stock can come out ofonly one zone to a hatchery and the offspring fromthat hatchery go back to that zone There is noswitching of animals across zones unless you arewilling to put the animals at a quarantine site for 6months and then have them health-tested after that

There have been occasions where companieshave taken the risk of doing that and theyrsquove hadtheir stock slaughtered out This we believe is theonly way that we can protect our industry and thesesorts of things do occur in our salmon industrySome of you will be aware that wersquove had problemswith aquabirnavirus in one region in Tasmania Thatarea is totally zoned off You cannot take live animalsout of there for any reason whatsoever Productcoming out of there has to be heads off gills off

181


gutted before it can move and there is a verydifferent level of surveillance in that area

If we get disease we insist on slaughtering outand there is no government compensation at themoment in Australia But this is the level we believethat we need to have to protect our industry from thesort of circumstances that other countries areunfortunately facing at the moment

Dr David Scarfe

Thank you Dan MacPhee would like to comment onfreedom from ISA

Dr Dan MacPheeMaritime Veterinary ServicesSt George NB

The question I want to raise is Is Scotland free ofISA The reason I ask is because as I understandthe program instituted in Scotland if a farm wasshown to be infected the farm was required toremove all of their stock without compensationFurthermore farms within a 5-km zone were placedunder increased surveillance That would seem tome to create a huge incentive for farmers to avoidhaving ISA at all costs First because if they doreport ISA theyrsquore going to suffer a loss of millions ofdollars which is almost certainly not insurableSecond because they may bring the same fate upontheir neighbors Third because in the coastalcommunities in which aquaculture exists theseactions may cause the loss of numerous jobs andthe effects would be felt throughout the community

In one respect I admire the strict regulatoryapproach that was adopted in Scotland because Isupport regulatory medicine where appropriate and Ibelieve it is appropriate in the case of ISA

If governments are going to take such a drasticand Draconian action as eradication withoutcompensation I believe that they have aresponsibility to determine whether or not the actionthey took was effective The only means I know of todetermine if such regulatory action was effective is to

measure the outcome But in my experience it maybe very important to a farm to avoid the truth of thefact that they have ISA Therefore if there is a controlprogram involving total eradication withoutcompensation the program probably wonrsquot have 100-percent industry support in terms of diseasereporting In that scenario there needs to be a veryvery intensive surveillance program before anyconclusions can be drawn about the status of ISA inan area

The first question Irsquod like to ask is addressed tothe Scottish team What are the details of thesurveillance program instituted that has allowed themto declare that Scotland is free of ISA Also it hasbeen mentioned that ISA is a reportable disease inScotland My second question is whether or not thedisease is reportable by the farm or is it alsoreportable by others such as private veterinarians ordiagnostic laboratories My third question isaddressed to the audience in general What is anadequate testing protocol in order to determine if anarea is free of ISA assuming that there will not be fullcooperation on the part of the farms

Dr David Scarfe

Thank you I see a tremendous number of issuesbeing raised Possibly most important OIE has setsome clear guidance for response to aquatic animaldiseases perhaps Tore Haringstein would like tocomment I donrsquot believe the OIE Aquatic AnimalCode has suggested full guidelines for a validdeclaration of ISA pathogen or disease freedom Asa general principle continued active and passivesurveillance are mandatory requirements for such adeclaration Concerning the Scottish suggestion offreedom (and Irsquoll obviously defer to comments fromthe Scots) the information I read is that a statementof freedom originated from the Royal Society inEdinburgh an independent scientific body who wererequested to evaluate the information and provide anassessment of the ISA situation in Scotland

There is undoubtedly a need for continuingintentional trade and recognition of disease-free

182


status but there is going to have to be a definition ofwhat constitutes freedom from an aquatic animaldisease particularly when there are very largereservoirs in wild aquatic populations for whichdisease surveillance is very difficult

Dr Tore HaringsteinNational Veterinary InstituteOslo NO

In reference to the OIE Code and Manual OIE iscurrently writing a new schedule (introductorychapter) for testing in farming that will be published inthe next edition of the Manual The new generalchapter for the Manual is currently being evaluated byan ad hoc group on risk analysis in aquatic animalhealth that will advise the OIE Fish DiseasesCommission on amendments that need to be madeto the surveillance and sampling methods in order tomake the chapter as valid as possible We sent itover to the risk analysis people to make sure itrsquosokay

Dr Ron Stagg

Irsquoll try to be very brief and factual I think thatrsquos thebest way to deal with this problem First I think wersquoretalking about the eradication in Scotland of clinicaldisease and I think we should take the messagefrom the epidemiologic assessment wersquove made theepidemic being the point source and about theevidence of sporadic occurrence of virus in theenvironment I think the two are very differentSecond the basis of our surveillance program isinspection for clinical disease On that basis the lastclinical outbreak in Scotland was in February 1999This approach reflects what Tore Haringstein has justsaid about the methods available for screening forISAv

In relation to the question whether industrywould hide the presence of this disease I think thatrsquosvery unlikely given what happens in clinical cases ofISA and the way the mortality progresses I think wealso have to consider the very good workingrelationships with the industry and in particular the

joint response to dealing with the ISA crisis inScotland For example in one case the farm wascleared while we were in the process of confirmingThat showed industryrsquos dedication to getting rid ofthis disease So we not only have surveillance by theofficial service we actually have surveillance by theindustry on the industry and I think we shouldnrsquotforget that And last in response to the question ofreporting in Scotland it is a legal requirement for anyperson who suspects the presence of ISA to reportthis to the authorities

Dr David Scarfe

Thank you very much Ron Roberts would like toaddress the issue of control programs withoutindemnification in the United Kingdom

Dr Ronald Roberts

I want to raise the question of control methods orcontrol systems without indemnification for tworeasons I agree with Ron Stagg that if the lab isreporting ISA it would be almost impossible to hideit and certainly no one should do so Also virtuallyall fish farms in Scotland have their own veterinarysurgeon and their own regular vet inspections As aveterinary surgeon Irsquom very aware of this I have nodoubts whatsoever concerning owners of ISA-affected farms not reporting to the appropriateauthorities any real suspicion of any notifiable fishfarm disease On the other hand there may be aneed for acknowledging the fact that opportunitiesmay need to be arranged for tracing back conditionswhich we donrsquot know enough about yet that could beconfused with ISA which could be in the backgroundI donrsquot think anybody in the industry would jump toreport such a condition to the department if therewere no solid grounds for having made such a reportItrsquos human nature to remain silent if one is not surethat he has good grounds for suspicion Also theresults of reporting a disease are Draconian for thefarm even if in the end the report proves to benegative

184


both from an economic point of view for the farmerbut also for the accountability of the regulatoryagency to ensure that fair compensationindemnification is garnered before a diseaseregulation or program is put into place I think itwould be too cavalier to suggest that we can have aprogram in place that will work indefinitely or work toeradicate diseases like ISA without compensationSo Irsquom not going to carry on I think thatrsquos been dealtwith adequately by previous speakers

Dr David Scarfe

Thank you Trevor Hastings would like to offercomments on diagnostic testing

Dr Trevor HastingsFisheries Research Servicesrsquo Marine LaboratoryAberdeen UK

We had some interesting papers during the workshopdescribing new cell lines for isolation of ISA virusvalidation of methods specificity and so on Irsquom alsostruck by the fact that there are no accepted methodsfor demonstrating the absence of ISA In talking withcolleagues at diagnostic laboratories in differentcountries Irsquom struck by the differences that exist incriteria used for confirming the presence of ISA Forexample there are people who would not wish toconfirm ISA unless disease was present There arethose who would confirm ISA on the basis of a virusisolation And the consequences of these differenthuddle criteria for confirming disease can beextremely serious I just think it would be useful fordiagnosticians from different countries tocommunicate to each other and to try get organizedon the standards they use for the diagnosis of ISA

Mr Ragnar ThorarinssonALPharmaBergen NO

Well therersquos talk about vaccinations already now inEurope What Irsquove been thinking about is basically aquestion to the diagnosticians in the audience What

do you believe will be the value or lack of value withRTndashPCR when our stocks of fish have beenimmunized with oil-based vaccines (containinginactivated whole virions as the ISA antigen) to allviral infections of the ISA virus

Dr Peter Merrill

Two comments One itrsquos important to remember thatthe diagnosis of ISA disease among commerciallyraised fish is a rightmdashin the United States at leastmdashexclusively reserved for veterinarians who have thefinal say whereas successful pathogen detection canbe accomplished through any credible laboratory Ijust wanted to mention that because I agree thatthere should be standards in place to harmonizemethods by which a pathogen is detected But therealso has to be some leeway for the veterinarian toassess those findings in light of the clinical picture inwhich he needs to test the fish As far as RTndashPCRresults showing up in postvaccinated fish thatactually hasnrsquot been the case in my experience Isuppose the vaccine can affect it but I did run a verylarge-scale field trial for an ISAv vaccine and wetested a good number of fish at intervals of 1 day4 days 7 days 14 days 28 days and monthlythereafter and did not find any evidence of positiveresults of RTndashPCR in postvaccinated fish

Dr Paul Midtlyng

I would like to add that should RTndashPCR-positivesoccur in vaccinated fish oil-adjuvant vaccineformulations by themselves have what you could calla marker which is the antibody response levels tothose antigens present in the formulation I havenrsquotheard of similar antibody levels occurring in naturallyinfected fish Neither would unvaccinated fish displayany intraabdominal vaccine remains or adhesionslike we see in vaccinated individuals So I think thatfor many practical reasons fish that have receivedvaccine might be identified by a combination of testsagainst ISA virus and other constituents or antigensin those formulations I would be surprised if anymajor misclassification would occur in such cases

185


Mr John ReddingtonDiagXoticsWilton CT

Yes I think that there is an acute need forharmonization in these detection systems And oneof the issues is that there is fragmentation within theUnited States on what criteria this testing needs tomeet what approval processes need to be overcomeWersquove heard from several people that they getcontradictory results from a plethora of labs

My recommendation at least as we movethrough the process is that we donrsquot focus so muchon technologies as we start thinking aboutharmonization through a panel of characterizedsamples I just hope to give those people who areinterpreting the results some confidence of whatabilities those labs have in terms of the reproducibilityof the results that theyrsquore putting out

Dr Oslashystein EvensenNorwegian College of Veterinary MedicineOslo NO

This might be a little bit outside of the area that wersquorediscussing but I want to comment on what was saidabout the importance of the hypervariable region ofthe HA protein and possible implications of this invaccine development Therersquos really a need to test ifthe variation seen in the various isolates correlateswith the immune response and if it has anyimplication for protective immune response seen inthe fish It is not obvious that this is the fact

Dr David Scarfe

The virus wersquore dealing with is probably capable ofrapid DNA change (evolution) typical of mostinfluenza-type viruses and I wonder to what degreewe are discussing actual virus presence versusdisease And from a clinical point of view Irsquove alwaysvery clearly separated diagnosing disease (resultingin actual pathology) from virus identification Eventhough these are highly interrelated and asdiagnostic techniques for this disease become

refined the epidemiologic understanding of thisdisease needs a clear distinction between presenceof diseases and presence of pathogen TrishBarbash wanted to comment on the disease in wildpopulations

Ms Patricia BarbashUS Fish and Wildlife ServiceNortheast Fisheries CenterLamar PA

I would like to direct your attention back to wild fishfor a few moments and I want to share someinformation with you on some of the efforts myagency has made in evaluating the status of ISAv inwild Atlantic salmon In the Northeastern UnitedStates my agency has been struggling to restoreandor maintain Atlantic salmon runs to about adozen rivers All but two of these rivers are in MaineSalmon stocks have been designated under theEndangered Species Act as distinct populationsegments for eight rivers in Maine All of thosepopulations are perilously low and our managementof disease in the populations can be very tricky

The best historical run remains the PenobscotRiver in Maine which flows through Bangor It canhave a thousand salmon or more annually migratingfrom the sea The fish are trapped at the VeazieDam and approximately half of the run is transportedto a holding facility at the Craig Brook National FishHatchery and held for spawning in covered poolsEach pool will hold in excess of a hundred fishdepending on the run The Craig Brook National FishHatchery also maintains endangered stocks ofsalmon from six rivers held in separate rearing unitsfor the purposes of enhancement and restoration

There is good deal of concern that sea-run fishcarrying ISA from the ocean may put the endangeredstocks at risk Last year we implemented a samplingregime on the Penobscot in order to accomplishseveral tasks one to determine the prevalence ofISA in fish caught from the Penobscot River duringthe upstream migration two to assess the risk ofcollecting these fish and holding them at a facility

186


which also rears endangered stocks three to assessthe risk of transmission of the virus horizontally andvertically in fresh water four to track any detectedinfections as the fish remains in fresh water and fiveto determine best management strategies forminimizing the impacts of ISAv on these fish

In 2001 we attempted to sample blood from asmany incoming Penobscot sea runs as we could Wewere only able to get 68 fish for reasons Irsquom not goingto discuss here The results were from RTndashPCR andsome were other assays but one blood sample inthose produced a weak RTndashPCR result which wascorroborated by another lab and sequencing found itto be about 97 percent homologous to the ScottishNorwegian strain of ISAv The virus however wasnever isolated on cell-culture assays After 1 monthof isolation in a pool of fresh water this fish wassampled again with negative results

All the fish were sampled right before spawningblood samples and all the results and cell cultures forRTndashPCR were negative In 2002 we attempted thesame thing We collected 126 samples from morethan 380 fish brought to the facility So far all thetests that wersquove run are negative for ISAv We doplan to test again before spawning

It is impossible to draw conclusions from oneRTndashPCR finding from the 2001 sampling althoughour results have raised some eyebrows and didindeed indicate that our wild salmon are most likelyexposed to the virus at some unknown point whilemigrating and feeding in the ocean environmentUnfortunately therersquos very scant information on thestatus of these fish when theyrsquore in the ocean

Cooperation with the different resourceagencies such as the National Marine FisheriesService and the US Fish and Wildlife Service andsome of the State agencies is a good example of howwe can continue to make efforts and learn about thestatus ISAv in wild populations

Dr David Scarfe

Thank you very much Irsquod like to point out that thistype of approach to active surveillance is absolutely

critical albeit that it is fraught with all sorts ofproblems such as how to validly sample populationsLetrsquos move on to looking at some vaccines KiraSalonius has asked to make some comments

Ms Kira SaloniusNovartis Aqua HealthVictoria PE

First of all Irsquod just like to invite you to the session [ofthe overall symposium] on vaccines where I willpresent a paper on the efficacy of ISAv vaccines infresh water Irsquod like to break out a little bit of datafrom this presentation that are relevant to some of thethings that we discussed here yesterday Itrsquos mostlyindustry data about the experiences withdepopulation and how that relates to vaccinationstatus so there are a few assumptions to be madeWhen you consider the presentation of the data Irsquomgoing to show you the first assumption is that it wascollected from farms and records of vaccinationsspecifically with the assistance of the New BrunswickSalmon Growers Association What makes thiseasier for us is because there are only twocompanies involved so there are only two productstwo different products being used on the fish at thefish-farm level And wersquore basing this data on theassumption that all fish on exposed sites are all fishin one population so wersquore dealing with results thatare percentages calculated on numbers of fish notsites We are comparing the total number of fishdepopulated to the total number of fish vaccinatedVaccination status also has been assumed not to bebiased to the criteria for depopulation Most of thetime or at least in our experience the veterinariansare not using vaccination status as an influence onthe decision to depopulate

In contradiction to the presentation yesterdayonly 40 percent of the fish in the 2001 year-class arevaccinated not 100 percent as was indicated In the2002 year-class the majority has been vaccinatedagainst ISAv In 2001 approximately 60 percent ofthe population was not vaccinated specifically againstISAv

187


Forte V1 is a Novartis Aqua Health product andit represents only 23 percent of the population in the2001 year-class Our competitor Bayer is the othercompany in the Bay of Fundy marketing vaccines forfish and we donrsquot have any transparency with themso I canrsquot tell you what proportion of those fish areISA-specific vaccinates This [slide] is the ISAvstatus for the 2001 year-class from New Brunswickbased on those data from industry So the totalnumber of fish in 2001 year-class on the positivesites deemed positive by the criteria for ISAv statusis 227 percent The total number of fish eradicatedas of August 1 was 61 percent ldquodepopulatedrdquo Ithink is a better term The vaccination status versusdepopulation status in the 2001 year-class for ForteV1 which is a vaccine specific for ISAv Forte andTriple (multivalent bacterins) and a competitorvaccine are shown [in the slide] As I said we canrsquotcomment on the proportion in this population that arespecifically vaccinated with ISAv There is a fourfolddecrease in the number of fish depopulated that werevaccinated with Forte V1 versus Forte-vaccinatedfish

We can state that vaccination reduces the riskof depopulation This will be assessed further in mypresentation [in the main symposium] tomorrow AndI would say that the situation for disease in the Bay ofFundy isnrsquot particularly better this year than it hasbeen in other years but itrsquos important to separatewhen you speak of that what the vaccination statusof the population is Thank you to Dr Miller for theopportunity to present this in brief

Dr David Scarfe

Thank you very much Trevor wanted to make somemention of field trials

Dr Trevor Hastings

It was interesting to hear that last report It is inrelation to that sort of thing that I want to speak Irsquomstruck by the absence of good field data on theefficacy of vaccines and it seems to me that vaccineuse is pretty unregulated Virtually anybody seems to

be able to use vaccines or not use them according totheir wish Those who regulate the use of ISAvaccines donrsquot appear to have specified that properfield trials be carried out and Irsquod like to be corrected ifIrsquom wrong Thatrsquos really what Irsquom talking aboutwhoever does allow these other ISA vaccines shouldplace a requirement that they be properly evaluatedrather than just used without advance testing

Dr David Scarfe

From the perspective of the American VeterinaryMedical Association and the veterinary profession inthe United States we do have concern related to theunderstanding of efficacy of vaccines specifically thedegree of efficacy required of vaccines in theregulatory (licensing) process and the publicperception of how effective vaccines are In myopinion the false public perception of ldquooncevaccinated always free of pathogens or diseasesrdquoprobably gleaned from the success against humanpolio and smallpox is damaging to disease responseThe AVMA is working with US Federal agenciesright now to try get appropriate labeling on vaccinesand biologics to help correct these misperceptions

Dr Paul Midtlyng

I would like to comment on Trevorrsquos statements FirstIrsquom not surprised about the quality of field data Iwould rather say this problem is normal for field trialsunless the organizers put in a huge effort andmanpower to follow up and manage their data toenable a high-quality efficiency evaluation I think Irsquovebeen perhaps one of the last persons conducting ahuge field trial (on furunculosis) in a situation like thisbut that was only made possible through full-timeemployment in followup After 3 to 4 years of work Iwas able to extract information out of whichreasonably reliable results were obtained But as faras I know the ISA vaccinations have so far beenoperational rather than for vaccine documentationpurposes

I think that for fish vaccines their protectivecapacity must be demonstrated through controlled

188


experimental trials The results may then bevalidated through field trials which cannot as Imentioned be as well controlled as experimentalwork I donrsquot think that data coming out of the ISAfield trials are worse than those coming out of mostother field vaccine trials but they are too uncontrolledto document the efficacy of the vaccines on theirown

Mr Pawan AgrawalCanadian Food Inspection AgencyOttawa ON

I would like to clarify some of the vaccine-relatedissues Our office is responsible for licensingveterinary biologics in Canada In early 1998 weapproved the use of autogenous ISA vaccines Wemade certain exceptions to meet an urgent need forthe vaccine at that time But later the ISA vaccineswere licensed They went through the laboratoryapproval process which is similar to that formammalian vaccines There are no exceptionsmade

Efficacy of the ISAv vaccine is based on thevaccination-challenge model in the lab situationRight now therersquos no requirement for a field efficacyfor ISAv vaccine I agree with the previous speakersin that it would be very difficult to generate the fieldefficacy data There are several complications withthat

The first one is that the natural challenge is notvery predictable Another reason is itrsquos veryexpensive to run large-scale field efficacy studiesAlso the time required to generate the field efficacydata is very long To generate valid data we need tomaintain vaccinated and unvaccinated fish in thesame cage and I donrsquot think the fish farmers wouldlike to do that

During our licensing process we consult withother agencies and departments such as theDepartment of Fisheries and Oceans Canada andalso with the stakeholders as necessary And ourrole is to approve the vaccine Itrsquos other relevantFederal and Provincial authoritiesrsquo responsibility to

decide whether to use the vaccine in certainsituations or not That is not our responsibility

There were some issues raised about thepotential for antigenic changes in ISAv And early onwhen we reviewed the ISAv vaccine submissions werealized the need for addressing this issueTherefore we required that manufacturers reevaluatethe efficacy of ISAv vaccine at an appropriate intervalor whenever therersquos an indication that the vaccine isnot protecting against the current ISAv strains Themanufacturers have to reevaluate the efficacy of thevaccine using the most recent field isolate

For potency of current ISAv vaccine each batchof the vaccine is tested by the manufacturer by avaccination challenge method and our office alsoevaluates it before we allow the distribution of ISAvvaccine for market We welcome feedback fromusers

Dr Peter Merrill

I can give a little insight from the US perspectivehaving some experience with innocuous ISAvvaccines I want to point out that homogenousvaccines do not imply nor do they bear anyrequirement to [be effective in order to be] licensedvaccines under USDA regulations in the CFR Fielddata may be offered in support of the efficacy but itrsquosnot a requirement And the only other comment Irsquodlike to offer is that when field trials were begun therewas a very fine line set to both protect the proprietaryinterest of the vaccine manufacturer and still provideenough supportive information that can be sharedand disseminated to encourage the use of aparticular vaccine There are so many mechanicalfactors involved in running a field trial particularly inthe Passamaquoddy BayndashCobscook Bay area wherethere are differing approaches from government togovernment as far as protection and management ofISAv and ISA respectively that itrsquos almost impossibleto draw solid conclusions from field trials even underthe best of circumstances

189


Dr David StarlingUSDAndashAPHIS Veterinary ServicesCenter for Veterinary BiologicsAmes IA

Irsquod like to address some of topics just brought upOne is the efficacy of vaccines A lot of information isborne on the label In most cases licensed biologicshere in the United States carry on the label thephrase ldquoas a significant aid in the prevention ofrdquoVeterinary Servicesrsquo memorandum 800200mdashitrsquosposted on the Web sitemdashcan be read By definitionif a particular vaccine is effective under controlledconditions therersquos a significant difference between avaccinated group of animals and an unvaccinatedgroup of animals when both groups are challengedby virulent organisms The reason that wording isthere is because the VirusndashSerumndashToxin Act of 1913specifically prohibits dangerous contaminatedworthless products from being put into the marketThe label endorsement saying a vaccine is ldquoasignificant aid in prevention ofrdquo a given disease is aclear indication to the consumer that this is not aworthless product Products can be licensed andbear that label even though theyrsquove demonstrated agreater amount of efficacy Most of this area is left tothe manufacturer to choose and they have reasonswhy they choose to go to a lower level of labelingAlso the comment was made of the titersrsquo not havingefficacy testing and so forth That is true Butaccording to section 113113 of the Code of FederalRegulations there shall be an expectation of efficacy

Dr Hugh Mitchell

Irsquod certainly like to dispel any notion that therersquos nowillingness from the manufacturers fromveterinarians or farmers to get field data I mean italways seems to be the Holy Grail The question isDoes this product work in the real world Well one

of the problems in the real world is the inherent[variability] and itrsquos one of the most fascinating andaggravating challenges to set up and run a field trialin the real world for the various vaccines Manytimes if you manage to accomplish that the diseaseschange or the formulation of that particular vaccinechanges So I donrsquot think it means we give up ondoing field trials but I do think it means we have torealize the limitations of such trials Veterinariansmanufacturers and farmers have to cooperate inunprecedented fashion to generate that kind of datawhere the willingness to get that data is there

Dr Paul Midtlyng

Irsquod like to start some discussion on thenonvaccination policy or the prohibition of ISAvaccination To put it bluntly I believe there are noreally good arguments to prohibit ISA vaccination asa part of ISA control programs For the moment weare talking about inactivated vaccines We are havingan uncontrollable source of potential infection in thewild We all hope that this will trigger only occasionalnew outbreaks but the risk is obviously there andcannot be completely controlled The idea of keepinga sentinel population in such a situation where theISA diagnosis will first be confirmed after the animalshave started to shed virus is a terrible one fordisease control And this is unfortunately the currentstatus of ISA carrier diagnosismdashdespite PCR anddespite serology

I would hypothesize that the current policyimplemented by the EU is a kind of carryover fromterrestrial animals where farmed populations can beeffectively isolated from the reservoir fauna Wereally need to look into the details and therequirements necessary for nonvaccinated policy tobe successful I would very much appreciatediscussing this controversial issue in greater depth inorder to optimize the ISA control strategies

Dr David Scarfe

I thank everyone for their thoughts and comment

190


191

Sebastian M Belle MScExecutive DirectorMaine Aquaculture AssociationBox 148Hallowell ME 04347Phone (207) 622ndash0136Fax (207) 622ndash0576futureseasaolcom

Deborah A BouchardMicroTechnologies Inc41 Main StreetRichmond ME 04357Phone (207) 737ndash2637Fax (207) 737ndash4504dbouchardmicrotechnologiesbiz

Rocco Cipriano PhDUS Geological SurveyNational Fish Health Research

Laboratory11700 Leetown RoadKearneysville WV 25430Phone (304) 724ndash4432Fax (304) 724ndash4435rocco_ciprianousgsgov

Marcia Cook MScResearch and Productivity Council921 College Hill RoadFredericton NB CANADA E3B 6Z9Phone (506) 452ndash1212Fax (506) 452ndash1395mcookrpcunbca

Carey O Cunningham PhDFRS Marine LaboratoryPO Box 101375 Victoria RoadAberdeen UK AB11 9DBPhone +44 1224 295 634Fax +44 1224 295 667ccunninghammarlabacuk

Ian R DohooDepartment of Health ManagementAtlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 566ndash0640Fax (902) 566ndash0823dohooupeica

Appendix 1Authorsrsquo Affiliations

Stephen K Ellis DVMUSDA APHIS Veterinary ServicesMarine Trades CenterndashWCTC16 Deep Cove RoadEastport ME 04631Phone (207) 853ndash2581Fax (207) 853ndash2783stephenkellisaphisusdagov

Steve GriffithsResearch and Productivity Council921 College Hill RoadFredericton NB CANADA E3B6Z9Phone (506) 452ndash0581Fax (506) 452ndash1395sgriffitrpcunbca

Lori Gustafson DVM PhDUSDA APHIS Veterinary Services16 Deep Cove RoadEastport ME 04631Phone (207) 853ndash2581Fax (207) 853ndash2783lorilgustafsonaphisusdagov

Larry Hammell DVM MScAtlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 566ndash0728Fax (902) 566ndash0823lhammellupeica

Tore Haringstein PhDNational Veterinary InstitutePO Box 8156 Dep0033 OsloNorwayPhone +47 23 21 61 50Fax +47 23 21 60 01torehasteinvetinstno

Leighanne J Hawkins DVMMaritime Veterinary Services246 Main StreetPO Box 1105St George NB E5C 3S9 CanadaPhone (506) 566ndash8908Fax (506) 566ndash8909

Tomy Joseph DVM MScDepartment of Pathology and

MicrobiologyAtlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 566ndash0779Fax (902) 566ndash0851tjosephupeica

Frederick SB Kibenge BVM PhDDipl AC

Department of Pathology andMicrobiology

Atlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 566-0967Fax (902) 566-0851kibengeupeica

Molly JT Kibenge BVM MScPhD

Pacific Biological StationDepartment of Fisheries and OceansNanaimo BC V9T 6N7 CanadaPhone (250) 729ndash8380Fax (250) 756ndash7053KibengeMpacdfo-mpogcca

Cato LyngoslashyQuality Assurance DirectorPan Fish GroupGrimmergt 56002 AringlesundNorwayPhone +47 70116100Fax +47 70116161Cellular +47 930 11330catolyngoypanfishno

Sharon A MacLeanUS Department of CommerceNational Oceanic and Atmospheric

AdministrationNational Marine Fisheries Service28 Tarzwell DriveNarragansett RI 02882ndash1199Phone (401) 782ndash3258Fax (401) 782ndash3201smacleanmolananmfsgov

192


Carol A McClure DVM MSDip ACT

Department of Health ManagementAtlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 566ndash0609Fax (902) 566ndash0823cmcclureupeica

John McDougallAlpharma ASHarbitzalleacuteen 3PO Box 158 SkoslashyenNndash0212 Oslo NorwayPhone 47 22 52 90 81Fax 47 22 52 90 90johnmcdougallalpharmano

Sandi M McGeachy PhDPO Box 6000Department of Agriculture Fisheries

and AquacultureFredericton NB Canada E3B 5H1Phone (506) 453ndash8515Fax (506) 453ndash5210sandimcgeachygnbca

Alasdair H McVicarDepartment of Fisheries and Oceans200 Kent Street 12ndashWndash118Ottawa ON K1A 0E6 CanadaPhone (613) 990ndash0273Fax (613) 993ndash7665mcvicaradfo-mpogcca

Peter L Merrill DVMMicro Technologies Inc41 Main StreetRichmond ME 04357Phone (207) 737ndash2637Fax (207) 737ndash4504pmerrillmicrotechnologiesbiz

Otis Miller Jr DVM MSUSDAndashAPHISndashVeterinary Services4700 River Road Unit 46Riverdale MD 20737Phone (301) 734ndash6188Fax (301) 734ndash4982otismilleraphisusdagov

Mark J Moore DVMPO Box 129Department of Agriculture Fisheries

and AquaculturePO Box 1037St George NB E5C 3S9 CanadaPhone (506) 755ndash5067 or 755ndash4000Fax (506) 755ndash8909 or 755ndash4001markjmooregnbca

Alexander G Murray PhDFisheries Research Services Marine

LaboratoryPO Box 101375 Victoria RoadAberdeen AB11 9DB UKPhone +44 (0)1224 295616Fax +44 (0)1224 295667murraysmarlabacuk

Rachael RitchieResearch and Productivity Council921 College Hill RoadFredericton NB E3B 6Z9 CanadaPhone (506) 460ndash5670Fax (506) 452ndash1395rritchierpcunbca

Jill B RollandUSDAndashAPHISndashVeterinary ServicesUnit 464700 River RoadRiverdale MD 20737ndash1231Phone (301) 734ndash7727Fax (301) 734ndash4982jillbrollandaphisusdagov

Mike Snow PhDFisheries Research Services Marine

LaboratoryPO Box 101375 Victoria RoadAberdeen AB11 9DB UKPhone +44 1224 295 611Fax +44 1224 295 667msnowmarlabacuk

Ronald M StaggFisheries Research Services Marine

LaboratoryPO Box 101375 Victoria RoadAberdeen AB11 9DB UKPhone +44 1224 295 321Fax +44 1224 295 511staggrmarlabacuk

Henrik StryhnDepartment of Health ManagementAtlantic Veterinary CollegeUniversity of Prince Edward IslandCharlottetown PE C1A 4P3 CanadaPhone (902) 894ndash2847Fax (902) 566ndash0823hstryhnupeica

Kristin E Thorud DVM PhDNorwegian Animal Health AuthorityPO Box 8147 DepN-0033 OsloNorwayPhone + 47 23 21 65 61Fax + 47 23 21 65 01ketdyrehelsetilsynetno

Sherry VincentResearch and Productivity Council921 College Hill RoadFredericton NB E3B 6Z9 CanadaPhone (506) 460ndash5667Fax (506) 452ndash1395svincentrpcunbca

James R WintonWestern Fisheries Research CenterBiological Resources DisciplineUS Geological Survey6505 NE 65th StreetSeattle WA 98115Phone (206) 526-6282Fax (206) 526-6654Jim_Wintonusgsgov

193

Pawan AgrawalVeterinary Biologics SectionCanadian Food Inspection Agency59 Camelot DriveOttawa Ontario K2J 4C3CanadaPhone (613) 225ndash2342Fax (613) 228ndash6612pagrawalinspectiongcca

Patricia BarbashUS Fish and Wildlife ServiceNortheast Fisheries CenterFish Health SectionPO Box 155Lamar PA 16848Phone (570) 726ndash6611Fax (570) 726ndash7379patricia_barbashfwsgov


Iain EastAgriculture Fisheries Forestry-

AustraliaOffice of the Chief Veterinary OfficerGPO Box 858Canberra 2611 AustraliaPhone 61ndash2ndash6272ndash3106Fax 61ndash2ndash6272ndash3150iaineastaffagovau

Oslashystein EvensenNorwegian School of Veterinary

SciencePO Box 8146 DepNndash0033 Oslo NorwayPhone +47 22 96 45 00Fax +47 22 56 57 04oysteinevensenvethsno

Appendix 2Contributors to the Open Forum

Tore Haringstein PhDNational Veterinary InstitutePO Box 8156 Dep0033 Oslo NorwayPhone +47 23 21 61 50Fax +47 23 21 60 01torehasteinvetinstno

Trevor HastingsFisheries Research Services Marine

LaboratoryPO Box 101375 Victoria RoadAberdeen AB13 0DR UKPhone +44 1224 876544Fax +44 1224 295620hastingsmarlabacuk

Daniel MacPhee DVM DVScMaritime Veterinary Services LtdPO Box 1105 246 Main StSt George NB E5B 1L6 CanadaPhone (506) 755ndash8908Fax (506) 755ndash8909ddm-mvsnb-aibncom


and AquacultureFredericton NB E3B 5H1 CanadaPhone (506) 453ndash8515Fax (506) 461ndash5210sandimcgeachygnbca



Paul Midtlyng DVM PhDVESOVet ResearchPO Box 8109 DepNOndash0032 Oslo NorwayPhone +47 2296 1120Fax +47 2296 1101paulmidtlyngvesono

Hugh MitchellNoavartis Aqua HealthProfessional Services9500 NE 141 PlaceBothell WA 98011Phone (425) 821ndash6821Fax (425) 814ndash6946hughmitchellahnovartiscom

John ReddingtonDiagXotics Inc27 Cannon RoadWilton CT 06897Phone (203) 762ndash0279Fax (203) 762ndash2378marketingdiagxoticscom

Ronald RobertsUniversity of IdahoHagerman Lab9 Alexander Dr Bridge of AllanStirling FK9 4QB UKPhone +44 17 86 83 30 78Fax +44 17 86 83 30 87heronpiscescompuservecom


Kira SaloniusAqua Health LtdNovartis Animal Vaccines797 Victoria RoadVictoria PE C0A 2G0 CanadaPhone (902) 658ndash2260Fax (902) 658ndash2261kirasaloniusahnovartiscom

194


A David Scarfe PhD DVMMRSSA

American Veterinary MedicalAssociation

1931 N Meacham Rd Ste 100Schaumberg IL 60173ndash4360Phone (847) 925ndash8070 ext 6634 or (800) 248ndash2862Fax (847) 285ndash6634DScarfeavmaorg



David E Starling DVMUSDAndashAPHISndashVeterinary ServicesCenter for Veterinary Biologics510 S 17th St Ste 104Ames IA 50010Phone (515) 232ndash5785 ext 145Fax (515) 232ndash7120davidestarlingaphisusdagov

Ragnar ThorarinssonAlpharma Animal Health Division Aquatic ProductsC Sundtsgate 62N-5004 Bergen NorwayPhone +47 55 96 20 60Fax +47 55 96 01 05ragnarthorarinssonalpharmano

The US Departments of Agriculture (USDA) the Interior (USDI)and Commerce prohibit discrimination in all their programs andactivities on the basis of race color national origin sex religionage disability political beliefs sexual orientation or marital orfamily status (Not all prohibited bases apply to all programs)Persons with disabilities who require alternative means forcommunication of program information (Braille large printaudiotape etc) should contact USDArsquos TARGET Center at (202)720ndash2600 (voice and TDD)

To file a complaint of discrimination write USDA Director Office ofCivil Rights Room 326ndashW Whitten Building 1400 IndependenceAvenue SW Washington DC 20250ndash9410 or call (202) 720ndash5964(voice and TDD) USDA USDI and Commerce are equalopportunity providers and employers

The opinions expressed by individuals in this report do notnecessarily represent the policies of USDA USDI or Commerce

Mention of companies or commercial products does not implyrecommendation or endorsement by USDA USDI or Commerceover others not mentioned The Federal Government neitherguarantees nor warrants the standard of any product mentionedProduct names are mentioned solely to report factually on availabledata and to provide specific information

Photo credits The background illustration on the front cover wassupplied as a photo micrograph by Michael Opitz of the Universityof Maine and is reproduced by permission The line art of salmoncame from the 1973 book ldquoThe Salmon Their Fight for Survivalrdquo byAnthony Netboy Houghton Mifflin Company is the copyright holderon the line art which is reproduced by permission Images insidethe proceedings were supplied by the senior authors For reproduc-tion rights please consult each senior author using contactinformation in appendix 1 ldquoAuthorsrsquo Affiliationsrdquo

Issued April 2003




US GeologicalSurvey








i



ii








iii






Foreword

iv




v





Acknowledgments

vi







vii




viii




Tore Haringstein



Peter L Merrill






Alexander G Murray




Contents

ix









Cato Lyngoslashy


Ronald M Stagg


Sebastian M Belle





x





Alasdair H McVicar


Otis Miller Jr




xi

1









Etiology


2







3



Pathogenicity







Transmission


4










5








Management




6


Vaccination


References Cited









7




















8


















9



















10



















11












12

13





Tore Haringstein1

Introduction


















14

















15




Definitions





Contingency plans

















16







Contingency Plans







17



















18















Urgent Notification

Listed diseases





Zoonotic potential

Nonlisted diseases



19











Shetland Islands

















ot ot

20










Confirmation of ISA












21
























22



Conclusions





References Cited

















23









24


25



Introduction





Peter L Merrill1







26


Lab Tests



Sampling



27










28



Preserving Samples










ISAV Cell Culture


29






ISAv RTndashPCR





30






ISAvndashIFAT





31




Histology






32











33



Comparing Tests






















34










Neg control 2 0 0 0 0 0 00 00 Na 00



Totals 62 37 59 60 60 57 00 3025 Na 1313




1045 8 882 111 45 15

35








36



Acknowledgments


References Cited













37




















38


39




Introduction







40







25

20

15

10

5

01995 1996


Num

ber

of IS

A-a

ffect

ed fa

rms


41





Outcome Indicators















42






Envelope








43


European HA subtype


Can-14

Can-8

Can-1

Can-10

Can-11

Can-9

Can-15

Can-3

Can-4

Can-6

Can-12

Can-5

Can-7

Can-1

Can-2

Can-13

Nor-5

Nor-6

Nor-1

Nor-4

Nor-3

Scot-2

Scot-1

Nor-2

01





44












45


Virus

VirusVirus Antibody

Virus receptor

Fc receptor

Cell

Cell













46







Conclusion







F03 + 03 plusmn0021 (+) 0025 plusmn 0004

F04 + 214 plusmn 03 (+) 00


EF06 + 031 plusmn 0016 (+) 0061 plusmn 0038




47



References Cited














48


















49













50


51





Introduction


The Problem







52



Potential Solutions







53





Conclusion


Acknowledgments



References Cited








54


55



Alexander G Murray1










dQdt = mI (3)


R0 = βSm (4)



56













57









LochNevis

LochBroom

Orkney

LochCreran

Skye

Argyll

Shetland

58


Skye

2Confirmed

Suspect

Clear

1

01 100010010

Number of visits

Infe

ctio

n st

atus


Loch Broom






59



Local Spread





I 043 003 009 065 059

II 069 0001 043 028 029

III 062 005 041 082 044




Discussion





60







References Cited





61


















62







63





Introduction









64





Cell Lines




Virus Production


Virus Titrations


65







Virus titer



CCBB2 1057

Bremnes ASK 106

CCBB 1061


1European

2North American



66










67



Acknowledgments


References Cited















68










69




Introduction










70







Results



71



Discussion





Random effect












72








73




References Cited

















74


75





Introduction








76




ISAv Isolates




77





Sequence Analysis












Results


78



79



80



81



82










mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

83




mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

Discussion


84





mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

85



Acknowledgments


References Cited












86


















87





Introduction









88












89























90




1 234 PB2 238 PB1 235 P1 238 PB2 ND 220 PB2

2 234 PB1 238 PB2 235 P2 221 PB1 22 PB1 224 PB1

3 223 PA 230 PA 215 P3 192 PA ND 207 NP

4 175 HA 188 HA 207 HEF 157 G 16 G 180 P

5 156 NP 184 NP 180 NP 141 NP 15 NP 156

6 141 NA 140 NA NB 118 M gt096 M 096 M 132 HE

7 103 M 119 M 093 NS mdash ND 116 NS


Total 136 145 129 105 10 1328


Gene products






NP Nucleoprotein

NA NB Neuraminidase

M Matrix

NS Nonstructural

G Glycoprotein





- - - - - - - - - - - - - - - - - - - - - - - - - -

91













92




Conclusions


References Cited








93




















94

















95





96


97



Cato Lyngoslashy1


Disease Status




Year


90

80

70

60

50

Num

ber

of IS

A o

utbr

acks

40

30

20

10

0

19841985

19861987

19881989

19901991

19921993

19941995

19961997

19981999

20002001

2002




98



Effect on Industry










99









100




















101






Dead

Diseased

Latent carriers



Healthy noncarriers

102



















103








104




Year

19971996 19991998 2000 2002

900

50

60 Feed quota


30

40

20

0

10

800

700

600500

Tons

Den

sity

400

300

200

100

02001





105





30

25

20

15

10

5

0

0 1 2 3


kgm3

Numberm3

4 5

Den

sity


106









107





Conclusion


References Cited







108


















109









110


111



Ronald M Stagg1

Introduction










112

















113














114








West Loch Roag


0 5 10-km

Crown copy 2000

115








116









on risk assessment













117










High IR

High TR

Medium IR

Medium TR

Suspect farm



Low IR

Low TR



118



119


10

91998 1999

Confirmed


Suspect

2000

8

7

6

5

4

3

2

1

0


Date (month)







120






121








SITE ISA 9803

SITE ISA 9801

SITE ISA 9808

1998

1995 1996 1997 1998


BroodfishHatchery A

1998



N D


CullSmoltsISA 9802

Cull

Harvest

Harvest

Harvest

SA 9805 SA 9805

Cull


FALLOW

FALLOW

FALLOW

Harvest


FALLOW







GrowersISA 9816

GrowersISA 9801

ISA 9806

GrowersISA 9803

GrowersISA 9801

BroodfishSea water

site A

GrowersISA 98031994

ISA 9805

1997


122



Acknowledgments


References Cited













123




















124















125



Sebastian M Belle1


Introduction







Biosecurity Audits


126




Action Plans







127













128





Communication



Waste Management




129









USDA ISA Program


130














131
















132



Acknowledgments


References Cited













133








134


135




Introduction









136





Fishes Sampled









137



Assay Procedures





Results





American eel 297





Pollock 16

138



139


Cobscook BayDennys

Sheepscot

Penobscot Pleasant

Narraguagus


140










Discussion




141









142




Acknowledgments


References Cited










143















144


145














146








147



















148




Mean no ofsalmon


1995 4 NA NA1996 17 NA NA1997 24 1667870 694951998 16 1190511 744061999 24 1636518 681882000 9 221700 246332001 15 985000 65667


19950

5

10

15

20

25

1996 1997 1998 1999 2000 2001


No

of I

SA

-affe

cted

farm

s








149










150














Totals 29 21 1489454 70926 5 274284 54865 10 565734 56573


151





Research





Conclusion


152


References Cited
















153











154


155















156









1984 1987 1990 1993 1996 1999 2002




80

70

60

50

40

30

20

10

0





157














158






Present Strategy





1998 1999 2000 2001




159



Conclusions


References Cited







160


161



Alasdair H McVicar1

Introduction















162













163













Early Warning




164











165



Conclusions


References Cited










166


167



Otis Miller Jr1











168








169












170














171
















172
















pyloric cecae







173









12-m2 steel 46 46 46 138

15-m2 steel 140 140 140 420

20-m2 steel 12 12 12 36

24-m2 steel 40 40 40 120

70-m polar 65 65 65 195




Acknowledgments









AVIC William Smith





174


References Cited








175
















176




Dr David Scarfe





Dr David Scarfe


Dr Ron Stagg





177










Dr Ron Stagg



178



Mr Sebastian Belle












Dr David Scarfe


179










Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


180













181




Dr David Scarfe








Dr David Scarfe



182





Dr Ron Stagg




Dr David Scarfe


Dr Ronald Roberts


184



Dr David Scarfe







Dr Peter Merrill


Dr Paul Midtlyng


185







Dr David Scarfe







186







Dr David Scarfe






187




Dr David Scarfe


Dr Trevor Hastings



Dr David Scarfe


Dr Paul Midtlyng



188











Dr Peter Merrill


189




Dr Hugh Mitchell



Dr Paul Midtlyng



Dr David Scarfe


190


191

























192
























193























194












US GeologicalSurvey








i



ii








iii






Foreword

iv




v





Acknowledgments

vi







vii




viii




Tore Haringstein



Peter L Merrill






Alexander G Murray




Contents

ix









Cato Lyngoslashy


Ronald M Stagg


Sebastian M Belle





x





Alasdair H McVicar


Otis Miller Jr




xi

1









Etiology


2







3



Pathogenicity







Transmission


4










5








Management




6


Vaccination


References Cited









7




















8


















9



















10



















11












12

13





Tore Haringstein1

Introduction


















14

















15




Definitions





Contingency plans

















16







Contingency Plans







17



















18















Urgent Notification

Listed diseases





Zoonotic potential

Nonlisted diseases



19











Shetland Islands

















ot ot

20










Confirmation of ISA












21
























22



Conclusions





References Cited

















23









24


25



Introduction





Peter L Merrill1







26


Lab Tests



Sampling



27










28



Preserving Samples










ISAV Cell Culture


29






ISAv RTndashPCR





30






ISAvndashIFAT





31




Histology






32











33



Comparing Tests






















34










Neg control 2 0 0 0 0 0 00 00 Na 00



Totals 62 37 59 60 60 57 00 3025 Na 1313




1045 8 882 111 45 15

35








36



Acknowledgments


References Cited













37




















38


39




Introduction







40







25

20

15

10

5

01995 1996


Num

ber

of IS

A-a

ffect

ed fa

rms


41





Outcome Indicators















42






Envelope








43


European HA subtype


Can-14

Can-8

Can-1

Can-10

Can-11

Can-9

Can-15

Can-3

Can-4

Can-6

Can-12

Can-5

Can-7

Can-1

Can-2

Can-13

Nor-5

Nor-6

Nor-1

Nor-4

Nor-3

Scot-2

Scot-1

Nor-2

01





44












45


Virus

VirusVirus Antibody

Virus receptor

Fc receptor

Cell

Cell













46







Conclusion







F03 + 03 plusmn0021 (+) 0025 plusmn 0004

F04 + 214 plusmn 03 (+) 00


EF06 + 031 plusmn 0016 (+) 0061 plusmn 0038




47



References Cited














48


















49













50


51





Introduction


The Problem







52



Potential Solutions







53





Conclusion


Acknowledgments



References Cited








54


55



Alexander G Murray1










dQdt = mI (3)


R0 = βSm (4)



56













57









LochNevis

LochBroom

Orkney

LochCreran

Skye

Argyll

Shetland

58


Skye

2Confirmed

Suspect

Clear

1

01 100010010

Number of visits

Infe

ctio

n st

atus


Loch Broom






59



Local Spread





I 043 003 009 065 059

II 069 0001 043 028 029

III 062 005 041 082 044




Discussion





60







References Cited





61


















62







63





Introduction









64





Cell Lines




Virus Production


Virus Titrations


65







Virus titer



CCBB2 1057

Bremnes ASK 106

CCBB 1061


1European

2North American



66










67



Acknowledgments


References Cited















68










69




Introduction










70







Results



71



Discussion





Random effect












72








73




References Cited

















74


75





Introduction








76




ISAv Isolates




77





Sequence Analysis












Results


78



79



80



81



82










mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

83




mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

Discussion


84





mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

85



Acknowledgments


References Cited












86


















87





Introduction









88












89























90




1 234 PB2 238 PB1 235 P1 238 PB2 ND 220 PB2

2 234 PB1 238 PB2 235 P2 221 PB1 22 PB1 224 PB1

3 223 PA 230 PA 215 P3 192 PA ND 207 NP

4 175 HA 188 HA 207 HEF 157 G 16 G 180 P

5 156 NP 184 NP 180 NP 141 NP 15 NP 156

6 141 NA 140 NA NB 118 M gt096 M 096 M 132 HE

7 103 M 119 M 093 NS mdash ND 116 NS


Total 136 145 129 105 10 1328


Gene products






NP Nucleoprotein

NA NB Neuraminidase

M Matrix

NS Nonstructural

G Glycoprotein





- - - - - - - - - - - - - - - - - - - - - - - - - -

91













92




Conclusions


References Cited








93




















94

















95





96


97



Cato Lyngoslashy1


Disease Status




Year


90

80

70

60

50

Num

ber

of IS

A o

utbr

acks

40

30

20

10

0

19841985

19861987

19881989

19901991

19921993

19941995

19961997

19981999

20002001

2002




98



Effect on Industry










99









100




















101






Dead

Diseased

Latent carriers



Healthy noncarriers

102



















103








104




Year

19971996 19991998 2000 2002

900

50

60 Feed quota


30

40

20

0

10

800

700

600500

Tons

Den

sity

400

300

200

100

02001





105





30

25

20

15

10

5

0

0 1 2 3


kgm3

Numberm3

4 5

Den

sity


106









107





Conclusion


References Cited







108


















109









110


111



Ronald M Stagg1

Introduction










112

















113














114








West Loch Roag


0 5 10-km

Crown copy 2000

115








116









on risk assessment













117










High IR

High TR

Medium IR

Medium TR

Suspect farm



Low IR

Low TR



118



119


10

91998 1999

Confirmed


Suspect

2000

8

7

6

5

4

3

2

1

0


Date (month)







120






121








SITE ISA 9803

SITE ISA 9801

SITE ISA 9808

1998

1995 1996 1997 1998


BroodfishHatchery A

1998



N D


CullSmoltsISA 9802

Cull

Harvest

Harvest

Harvest

SA 9805 SA 9805

Cull


FALLOW

FALLOW

FALLOW

Harvest


FALLOW







GrowersISA 9816

GrowersISA 9801

ISA 9806

GrowersISA 9803

GrowersISA 9801

BroodfishSea water

site A

GrowersISA 98031994

ISA 9805

1997


122



Acknowledgments


References Cited













123




















124















125



Sebastian M Belle1


Introduction







Biosecurity Audits


126




Action Plans







127













128





Communication



Waste Management




129









USDA ISA Program


130














131
















132



Acknowledgments


References Cited













133








134


135




Introduction









136





Fishes Sampled









137



Assay Procedures





Results





American eel 297





Pollock 16

138



139


Cobscook BayDennys

Sheepscot

Penobscot Pleasant

Narraguagus


140










Discussion




141









142




Acknowledgments


References Cited










143















144


145














146








147



















148




Mean no ofsalmon


1995 4 NA NA1996 17 NA NA1997 24 1667870 694951998 16 1190511 744061999 24 1636518 681882000 9 221700 246332001 15 985000 65667


19950

5

10

15

20

25

1996 1997 1998 1999 2000 2001


No

of I

SA

-affe

cted

farm

s








149










150














Totals 29 21 1489454 70926 5 274284 54865 10 565734 56573


151





Research





Conclusion


152


References Cited
















153











154


155















156









1984 1987 1990 1993 1996 1999 2002




80

70

60

50

40

30

20

10

0





157














158






Present Strategy





1998 1999 2000 2001




159



Conclusions


References Cited







160


161



Alasdair H McVicar1

Introduction















162













163













Early Warning




164











165



Conclusions


References Cited










166


167



Otis Miller Jr1











168








169












170














171
















172
















pyloric cecae







173









12-m2 steel 46 46 46 138

15-m2 steel 140 140 140 420

20-m2 steel 12 12 12 36

24-m2 steel 40 40 40 120

70-m polar 65 65 65 195




Acknowledgments









AVIC William Smith





174


References Cited








175
















176




Dr David Scarfe





Dr David Scarfe


Dr Ron Stagg





177










Dr Ron Stagg



178



Mr Sebastian Belle












Dr David Scarfe


179










Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


180













181




Dr David Scarfe








Dr David Scarfe



182





Dr Ron Stagg




Dr David Scarfe


Dr Ronald Roberts


184



Dr David Scarfe







Dr Peter Merrill


Dr Paul Midtlyng


185







Dr David Scarfe







186







Dr David Scarfe






187




Dr David Scarfe


Dr Trevor Hastings



Dr David Scarfe


Dr Paul Midtlyng



188











Dr Peter Merrill


189




Dr Hugh Mitchell



Dr Paul Midtlyng



Dr David Scarfe


190


191

























192
























193























194











ii








iii






Foreword

iv




v





Acknowledgments

vi







vii




viii




Tore Haringstein



Peter L Merrill






Alexander G Murray




Contents

ix









Cato Lyngoslashy


Ronald M Stagg


Sebastian M Belle





x





Alasdair H McVicar


Otis Miller Jr




xi

1









Etiology


2







3



Pathogenicity







Transmission


4










5








Management




6


Vaccination


References Cited









7




















8


















9



















10



















11












12

13





Tore Haringstein1

Introduction


















14

















15




Definitions





Contingency plans

















16







Contingency Plans







17



















18















Urgent Notification

Listed diseases





Zoonotic potential

Nonlisted diseases



19











Shetland Islands

















ot ot

20










Confirmation of ISA












21
























22



Conclusions





References Cited

















23









24


25



Introduction





Peter L Merrill1







26


Lab Tests



Sampling



27










28



Preserving Samples










ISAV Cell Culture


29






ISAv RTndashPCR





30






ISAvndashIFAT





31




Histology






32











33



Comparing Tests






















34










Neg control 2 0 0 0 0 0 00 00 Na 00



Totals 62 37 59 60 60 57 00 3025 Na 1313




1045 8 882 111 45 15

35








36



Acknowledgments


References Cited













37




















38


39




Introduction







40







25

20

15

10

5

01995 1996


Num

ber

of IS

A-a

ffect

ed fa

rms


41





Outcome Indicators















42






Envelope








43


European HA subtype


Can-14

Can-8

Can-1

Can-10

Can-11

Can-9

Can-15

Can-3

Can-4

Can-6

Can-12

Can-5

Can-7

Can-1

Can-2

Can-13

Nor-5

Nor-6

Nor-1

Nor-4

Nor-3

Scot-2

Scot-1

Nor-2

01





44












45


Virus

VirusVirus Antibody

Virus receptor

Fc receptor

Cell

Cell













46







Conclusion







F03 + 03 plusmn0021 (+) 0025 plusmn 0004

F04 + 214 plusmn 03 (+) 00


EF06 + 031 plusmn 0016 (+) 0061 plusmn 0038




47



References Cited














48


















49













50


51





Introduction


The Problem







52



Potential Solutions







53





Conclusion


Acknowledgments



References Cited








54


55



Alexander G Murray1










dQdt = mI (3)


R0 = βSm (4)



56













57









LochNevis

LochBroom

Orkney

LochCreran

Skye

Argyll

Shetland

58


Skye

2Confirmed

Suspect

Clear

1

01 100010010

Number of visits

Infe

ctio

n st

atus


Loch Broom






59



Local Spread





I 043 003 009 065 059

II 069 0001 043 028 029

III 062 005 041 082 044




Discussion





60







References Cited





61


















62







63





Introduction









64





Cell Lines




Virus Production


Virus Titrations


65







Virus titer



CCBB2 1057

Bremnes ASK 106

CCBB 1061


1European

2North American



66










67



Acknowledgments


References Cited















68










69




Introduction










70







Results



71



Discussion





Random effect












72








73




References Cited

















74


75





Introduction








76




ISAv Isolates




77





Sequence Analysis












Results


78



79



80



81



82










mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

83




mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

Discussion


84





mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10


mdash L

ane

1

mdash L

ane

2

mdash L

ane

3

mdash L

ane

4

mdash L

ane

5

mdash L

ane

6

mdash L

ane

7

mdash L

ane

8

mdash L

ane

9

mdash L

ane

10

85



Acknowledgments


References Cited












86


















87





Introduction









88












89























90




1 234 PB2 238 PB1 235 P1 238 PB2 ND 220 PB2

2 234 PB1 238 PB2 235 P2 221 PB1 22 PB1 224 PB1

3 223 PA 230 PA 215 P3 192 PA ND 207 NP

4 175 HA 188 HA 207 HEF 157 G 16 G 180 P

5 156 NP 184 NP 180 NP 141 NP 15 NP 156

6 141 NA 140 NA NB 118 M gt096 M 096 M 132 HE

7 103 M 119 M 093 NS mdash ND 116 NS


Total 136 145 129 105 10 1328


Gene products






NP Nucleoprotein

NA NB Neuraminidase

M Matrix

NS Nonstructural

G Glycoprotein





- - - - - - - - - - - - - - - - - - - - - - - - - -

91













92




Conclusions


References Cited








93




















94

















95





96


97



Cato Lyngoslashy1


Disease Status




Year


90

80

70

60

50

Num

ber

of IS

A o

utbr

acks

40

30

20

10

0

19841985

19861987

19881989

19901991

19921993

19941995

19961997

19981999

20002001

2002




98



Effect on Industry










99









100




















101






Dead

Diseased

Latent carriers



Healthy noncarriers

102



















103








104




Year

19971996 19991998 2000 2002

900

50

60 Feed quota


30

40

20

0

10

800

700

600500

Tons

Den

sity

400

300

200

100

02001





105





30

25

20

15

10

5

0

0 1 2 3


kgm3

Numberm3

4 5

Den

sity


106









107





Conclusion


References Cited







108


















109









110


111



Ronald M Stagg1

Introduction










112

















113














114








West Loch Roag


0 5 10-km

Crown copy 2000

115








116









on risk assessment













117










High IR

High TR

Medium IR

Medium TR

Suspect farm



Low IR

Low TR



118



119


10

91998 1999

Confirmed


Suspect

2000

8

7

6

5

4

3

2

1

0


Date (month)







120






121








SITE ISA 9803

SITE ISA 9801

SITE ISA 9808

1998

1995 1996 1997 1998


BroodfishHatchery A

1998



N D


CullSmoltsISA 9802

Cull

Harvest

Harvest

Harvest

SA 9805 SA 9805

Cull


FALLOW

FALLOW

FALLOW

Harvest


FALLOW







GrowersISA 9816

GrowersISA 9801

ISA 9806

GrowersISA 9803

GrowersISA 9801

BroodfishSea water

site A

GrowersISA 98031994

ISA 9805

1997


122



Acknowledgments


References Cited













123




















124















125



Sebastian M Belle1


Introduction







Biosecurity Audits


126




Action Plans







127













128





Communication



Waste Management




129









USDA ISA Program


130














131
















132



Acknowledgments


References Cited













133








134


135




Introduction









136





Fishes Sampled









137



Assay Procedures





Results





American eel 297





Pollock 16

138



139


Cobscook BayDennys

Sheepscot

Penobscot Pleasant

Narraguagus


140










Discussion




141









142




Acknowledgments


References Cited










143















144


145














146








147



















148




Mean no ofsalmon


1995 4 NA NA1996 17 NA NA1997 24 1667870 694951998 16 1190511 744061999 24 1636518 681882000 9 221700 246332001 15 985000 65667


19950

5

10

15

20

25

1996 1997 1998 1999 2000 2001


No

of I

SA

-affe

cted

farm

s








149










150














Totals 29 21 1489454 70926 5 274284 54865 10 565734 56573


151





Research





Conclusion


152


References Cited
















153











154


155















156









1984 1987 1990 1993 1996 1999 2002




80

70

60

50

40

30

20

10

0





157














158






Present Strategy





1998 1999 2000 2001




159



Conclusions


References Cited







160


161



Alasdair H McVicar1

Introduction















162













163













Early Warning




164











165



Conclusions


References Cited










166


167



Otis Miller Jr1











168








169












170














171
















172
















pyloric cecae







173









12-m2 steel 46 46 46 138

15-m2 steel 140 140 140 420

20-m2 steel 12 12 12 36

24-m2 steel 40 40 40 120

70-m polar 65 65 65 195




Acknowledgments









AVIC William Smith





174


References Cited








175
















176




Dr David Scarfe





Dr David Scarfe


Dr Ron Stagg





177










Dr Ron Stagg



178



Mr Sebastian Belle












Dr David Scarfe


179










Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


Dr Alasdair McVicar


Dr David Scarfe


180













181




Dr David Scarfe








Dr David Scarfe



182





Dr Ron Stagg




Dr David Scarfe


Dr Ronald Roberts


184



Dr David Scarfe







Dr Peter Merrill


Dr Paul Midtlyng


185







Dr David Scarfe







186







Dr David Scarfe






187




Dr David Scarfe


Dr Trevor Hastings



Dr David Scarfe


Dr Paul Midtlyng



188











Dr Peter Merrill


189




Dr Hugh Mitchell



Dr Paul Midtlyng



Dr David Scarfe


190


191

























192
























193























194









international response to infectious salmon anemia: prevention

Documents