detection of minchinia sp., in rock oysters saccostrea cuccullata (born, 1778) using dna probes

Journal of

www.elsevier.com/locate/yjipa

Journal of Invertebrate Pathology 97 (2008) 50–60

INVERTEBRATE

PATHOLOGY

Detection of Minchinia sp., in rock oysters Saccostrea cuccullata(Born, 1778) using DNA probes

Douglas Bearham a,*, Zoe Spiers a, Shane Raidal b, J. Brian Jones c, Philip K. Nicholls a

a Murdoch University, School of Veterinary and Biomedical Sciences, Murdoch Drive, Murdoch, WA 6150, Australiab Charles Sturt University, School of Agriculture and Veterinary Science, Boorooma Street, Wagga Wagga, NSW, Australia

c Fish Health Unit, Animal Health Labs, Department of Fisheries, 3 Baron-Hay Court, South Perth, WA 6151, Australia

Received 2 March 2007; accepted 9 July 2007Available online 18 July 2007

Abstract

Haplosporidian parasites infect various invertebrate hosts including some commercially important shellfish. Haplosporidium nelsoni

(along with Perkinsus marinus) has severely affected Eastern oyster production on the eastern seaboard of the United States and flat oys-ter production in Europe has been severely impacted by Bonamia ostreae. These parasites are also often present at a very low prevalenceand there are a variety of morphologically similar species that can be difficult to differentiate during cytological or histological diagnosishence the need to develop specific tests. Recently, a Minchinia sp. was described affecting rock oysters (Saccostrea cuccullata) in northWestern Australia.

In this study, two in situ hybridisation (ISH) assays and a PCR assay have been developed and optimised for use in investigating theseparasites. The first ISH assay used a 166 bp polynucleotide probe while the second used a 30 bp oligonucleotide probe. The specificity ofeach ISH assay was assessed by applying each probe to a variety of haplosporidian (5), a paramyxian (1) or ciliophora (1) parasites. Thepolynucleotide probe produced strong hybridisation signals against all of the haplosporidian parasites tested (Minchinia sp., Minchinia

teredinis, Bonamia roughleyi, H. nelsoni and Haplosporidium costale) while the oligonucleotide probe recognised only the Minchinia sp.Both probes failed to detect the paramyxian (Marteilia sp.) or the Rhynchodid-like ciliate. The PCR assay amplifies a 220 bp region anddetected Minchinia sp. DNA from 50 ng of genomic DNA extracted from the tissues of infected oysters and 10 fg of amplified Minchinia

sp. DNA. The assay did not react to oysters infected with H. nelsoni or H. costale. The ability of the PCR and oligonucleotide ISH assayto diagnose Minchinia sp. infected oysters was compared to histological examination from a sample of 56 oysters. The PCR assayrevealed 26 infections while histological examination detected 14 infections. The oligonucleotide ISH assay detected 29 infections.The oligonucleotide ISH and PCR assays were found to be significantly more sensitive than histology for detecting the parasite.Crown copyright � 2007 Published by Elsevier Inc. All rights reserved.

Keywords: Molecular probes; Haplosporidia; Saccostrea cuccullata, Minchinia; Haplosporidium; Bonamia; Rock oyster; SSU rRNA gene; Parasite;Aquaculture

1. Introduction

The effective control of diseases of aquatic animalsrequires access to diagnostic methods that are rapid, reli-able, and highly sensitive. Histopathology has traditionallybeen used as the primary method for the diagnosis of shell-fish diseases (Mialhe et al., 1995) since it can provide con-siderable information about the general health of a shellfish

0022-2011/$ - see front matter Crown copyright � 2007 Published by Elsevie

doi:10.1016/j.jip.2007.07.002

* Corresponding author.E-mail address: [email protected] (D. Bearham).

as well as the detection of a wide range of pathogens. How-ever, histopathology requires professional training andmany pathogens are difficult to detect by this method, ifthere are low numbers of parasites present within infectedtissues. It can also be difficult to definitively diagnose infec-tions based on parasite species morphology criteria.

DNA marker technology has several advantages overtraditional techniques such as histopathology. DNA mark-ers are extremely sensitive and are ideal for specifically rec-ognising target DNA sequences regardless of the lifehistory stage present (Burreson and Ford, 2004). The small

r Inc. All rights reserved.

mailto:[email protected]

D. Bearham et al. / Journal of Invertebrate Pathology 97 (2008) 50–60 51

ribosomal subunit (SSU) region of the rRNA gene hasbeen widely targeted since it contains both conserved andvariable regions interspaced within the sequence allowingthe design of universal or specific markers. While personnelinvolved in the development of molecular markers requirea considerable amount of professional training, the molec-ular assays once developed may be used by laboratory staffand technicians familiar with molecular diagnostictechniques.

Some caution needs to be applied to the use of molecu-lar assays. Often it is not known whether an assay willdetect all strains of a pathogen throughout its range or ifthe assay reacts with other pathogen species (Burresonand Ford, 2004). The assay should be sensitive enough todetect low levels of infection in positive samples. Conse-quently, qualification of the specificity and quantificationof the sensitivity of a molecular assay is required and theassay needs to be validated against other techniques, suchas histology, so that its reliability can be determined (Bur-reson and Ford, 2004; Kleeman et al., 2002). The properqualification and quantification of a molecular assay cantake a considerable amount of time and can be an expen-sive process.

Two techniques utilising DNA marker technology arethe polymerase chain reaction (PCR) and in situ hybrid-isation (ISH). Polymerase chain reaction allows highthroughput of samples and enhanced sensitivity but canlead to false negatives (Burreson, 2000; Carnegie et al.,2003). Polymerase chain reaction also requires othervisual confirmation methods so as to rule out the possibil-ity of false positives from contamination. In situ hybrid-isation can determine the locality of the parasite withinthe host and allow subsequent morphologicalcharacterisation.

The haplosporidians are a phylum of obligate parasiticprotozoa infecting a variety of invertebrate hosts. Haplo-sporidian parasites can be amongst the most lethal of allbivalve pathogens particularly where naive hosts areexposed to the parasite (Anon, 2002; Burreson et al.,2000). Several haplosporidian parasites have demonstratedthe capacity to jump hosts including Haplosporidium nel-

soni and Bonamia ostreae (Burreson et al., 2000; OIE,2006). The ability to jump host is a concern for the man-agement of these parasites since many haplosporidiansare very cryptic and are likely to be missed in histology sec-tions if they are not being specifically targeted.

Mortalities among rock oysters Saccostrea cuccullata

were first recognised by energy companies on the NorthWest Gas Shelf of Western Australia in the early 1990s(Hine and Thorne, 2002). The companies subsequentlysubmitted samples for diagnosis. Consequently, a Minchi-

nia species (Haplosporidia: Haplosporidiidae) parasitisingrock oysters on the northern Western Australian coastlinewas morphologically described by Hine and Thorne (2002).A section of the organism’s rRNA gene was sequenced andan in situ hybridisation assay was developed in our labora-tory (Bearham et al., 2007).

This study suggests a PCR assay for Minchinia sp. andcommences an assessment of its sensitivity and specificity.The ISH assay developed in Bearham et al. (2007) is alsoassessed. Both molecular methods are compared to histol-ogy. In addition, a polynucleotide ISH assay is also devel-oped and assessed for specificity.

2. Methods and materials

The sources of histologically positive bivalve sectionsinfected with target parasites are indicated in Table 1. Eth-anol stored and formalin fixed paraffin embedded samplesof Minchinia sp. were obtained from the MontebelloIslands (latitude �20.4 0S longitude 115.53 0E) as describedpreviously (Bearham et al., 2007). Extracts of genomic oys-ter DNA infected with Haplosporidium costale and H. nel-

soni were obtained from the Virginia Institute of MarineScience (VIMS).

2.1. DNA extraction

Genomic DNA from infected rock oysters was extractedfrom approximately 5 mg of gill tissue using a Master-pure� DNA purification kit (Epicentre Technologies, Syd-ney) according to the manufacturer’s protocol.

2.2. Amplification by polymerase chain reaction

The Minchinia sp. SSU rDNA gene sequence (GenBankAccession No. EF 165631) was aligned with the sequencesof the haplosporidian parasites Minchinia teredinis(U20319), Minchinia sp. from Cyrenoida floridana

(AY449712), Minchinia chitonis (AY449711), Minchinia

tapetis (AY449710), B. ostreae (AF262995), Bonamia

roughleyi (AF508801), H. costale (F387122), Haplosporidi-

um pickfordi (AY452724), Haplosporidium lusitanicum

(AY449713), H. nelsoni (U19538), Urosporidium crescens

(U47852) and host 18s sequences using the CLUSTALWalgorithm within the MEGA 3.1 sequence analysis pro-gram (Kumar et al., 2004). The variable regions wereassessed for sequences that appeared to be species specific.Sequences were then assessed for use as potential PCRprimers using the software Genetool (Doubletwist). Oncesuitable priming regions were identified the potential prim-ers were sent as queries to the GenBank database (http://www.ncbi.nlm.nih.gov/blast/) to determine if the primerswould anneal to non-target genes. Two oligonucleotides,designated SSF66 (5 0-ccgcgcatgcccagccgtat-3 0) and SSR69were selected and commercially synthesised (Geneworks,Sydney). The sequence of the SSR69 primer is the sameas the Minchinia sp. ISH probe described earlier by thislaboratory (Bearham et al., 2007). Genomic parasite androck oyster DNA was subjected to polymerase chain reac-tion using the primers designated SSF66 and SSR69 (Bear-ham et al., 2007). Each PCR was performed in a totalvolume of 25 lL and contained 67 mM Tris–HCl,16.6 mM [NH4]2SO4, 0.45% Triton X-100, 0.2 mg/mL gel-

http://www.ncbi.nlm.nih.gov/blast/

http://www.ncbi.nlm.nih.gov/blast/

Tab

le1

Th

eso

urc

eso

fre

pre

sen

tati

vep

ath

oge

ns

and

ho

stsp

ecie

su

sed

inth

ein

situ

hyb

rid

isat

ion

assa

ys

Par

asit

esp

ecie

saP

hyl

um

Ho

stsp

ecie

sn

Sam

ple

ob

tain

edfr

om

Lo

cati

on

of

infe

ctio

nb

Dat

eo

fin

fect

ion

Min

chin

iasp

.H

aplo

spo

rid

iaS

acc

ost

rea

cucc

ull

ata

56A

sin

:B

earh

amet

al.

(200

7)M

on

teb

ello

Isla

nd

s20

05M

inch

inia

tere

din

isH

aplo

spo

rid

iaS

hip

wo

rmT

ered

osp

.1

Vir

gin

iaIn

stit

ute

of

Mar

ine

Sci

ence

Wac

hap

reag

ue

Vir

gin

iaU

SA

1985

Ha

plo

spo

rid

ium

cost

ale

Hap

losp

ori

dia

Cra

ssost

rea

virg

inic

a1

Vir

gin

iaIn

stit

ute

of

Mar

ine

Sci

ence

Wac

hap

reag

ue

Vir

gin

iaU

SA

2006

Ha

plo

spo

rid

ium

nel

son

iH

aplo

spo

rid

iaC

rass

ost

rea

virg

inic

a1

Vir

gin

iaIn

stit

ute

of

Mar

ine

Sci

ence

Mac

hip

on

goR

iver

,V

irgi

nia

US

A20

05B

on

am

iaro

ug

hle

yi

Hap

losp

ori

dia

Sa

cco

stre

ag

lom

era

ta4

WA

Dep

artm

ent

of

Fis

her

ies

Oys

ter

Cre

ek,

Car

nar

von

1995

Mart

eili

asp

.C

erco

zoa

Sa

cco

stre

acu

ccu

lla

ta4

WA

Dep

artm

ent

of

Fis

her

ies

No

rth

Wes

tS

hel

f19

94R

hyn

cho

did

-lik

eci

liat

esC

ilio

ph

ora

Pin

cta

da

ma

xim

a1

WA

Dep

artm

ent

of

Fis

her

ies

Mo

nte

bel

loIs

lan

ds

2001

aR

epre

sen

tati

vesa

mp

les

wer

eh

isto

logi

call

yp

osi

tive

.b

All

sam

ple

sw

ere

ob

tain

edfr

om

Wes

tern

Au

stra

lia

un

less

oth

erw

ise

stat

ed.

52 D. Bearham et al. / Journal of Invertebrate Pathology 97 (2008) 50–60

atin and 0.2 mM dNTP’s, 1.5 mM of MgCl2,40 pmol ofeach primer, and 0.55 U of Taq polymerase and templateDNA. Each of the reaction mixtures were subjected toan initial denaturation phase of 5 min at 94 �C, then 35amplification cycles, each cycle consisting of 30 s of dena-turation at 94 �C, 30 s of annealing at 62 �C, 2 min ofextension at 72 �C and then a final 7 min extension at72 �C.

A 10 lL aliquot of each of the PCR products was thenloaded on a 2% agarose gel and subjected to electrophore-sis for 20 min at 95 V. Detection of the PCR products wasperformed using ethidium bromide staining. PhiX/HaeIII(promega USA) markers were used to determine the sizeand concentration of the PCR products.

Primer specificity was tested in PCRs using genomicDNA extracted from oysters which were both positiveand negative for Minchinia sp. by histopathology. H. nel-

soni and H. costale DNA obtained from the Virginia Insti-tute of Marine Science was also used to test PCRspecificity. The infected oyster PCR contained 50 ng geno-mic DNA isolated from the gill tissue of a histologicallypositive rock oyster collected from the Montebello Islands,Western Australia. To ensure the PCR assay did not cross-react with host DNA a 500 ng aliquot of genomic oysterDNA isolated from the gill tissue of an uninfected oysterwas also tested.

2.3. Generation of polynucleotide probe for ISH

The first ISH assay utilised a 166 bp polynucleotideprobe which was generated by PCR. Genomic parasiteand rock oyster DNA was subjected to polymerase chainreaction using the primers designated Minch F1B andMinch R2B (Burreson et al., 2000; Table 1). Each PCRwas performed using reaction mixtures that had a total vol-ume of 50 lL and contained 67 mM Tris–HCl, 16.6 mM[NH4]2SO4, 0.45% Triton X-100, 0.2 mg/mL gelatin and0.2 mM dNTP’s, 2 mM of MgCl2,40 pmol of each primer,and 0.55 U of Taq polymerase and template DNA. Each ofthe reaction mixtures were subjected to an initial denatur-ation phase of 5 min at 94 �C, then 35 amplification cycles,each cycle consisting of 30 s of denaturation at 94 �C, 30 sof annealing at 48 �C, 2 min of extension at 68 �C and thena final 7 min extension at 68 �C. The successful amplifica-tion of target DNA by each PCR was determined usingthe method outlined above.

2.4. In situ hybridisation

A 166 bp polynucleotide probe obtained by PCR usingthe primers Minch F1B (Bearham et al., 2007) and MinchR2B (Bearham et al., 2007) was purified with a QiAquickPCR purification kit (Qiagen) and labelled using a nick-translation mix (Boehringer-Mannheim), both accordingto manufacturer’s instructions. The second ISH assay useda 30 bp synthetic oligonucleotide (SSR69*; Bearham et al.,

Fig. 1. Agarose gel electrophoresis of the SSF66/SSR69 PCR productsdemonstrating specificity for Minchinia sp. Samples were electrophoresedin a 2% agarose gel for 45 min at 90 V. Lane 1: PhiX/HaeIII molecularweight markers. The size of each marker in base pairs is indicated to theleft. Lane 2: No DNA negative control. Lane 3: Positive Minchinia sp.infected rock oyster sample. Lane 4: Uninfected rock oyster. Lane 5:Haplosporidium nelsoni sample. Lane 6: Haplosporidium costale sample.

Fig. 2. Agarose gel electrophoresis of the SSF66/SSR69 PCR products withvarying amounts of amplified Minchinia sp. DNA. Samples were electro-phoresed in a 2% agarose gel for 45 min at 90 V. Lane 1: PhiX/HaeIIImolecular weight markers. The size of each marker in base pairs is indicatedto the left. Lane 2: No DNA negative control. Lane 3: 1 ng of template DNA.Lane 4: 100 pg. Lane 5: 10 pg. Lane 6: 1 pg. Lane 7: 10 fg. Lane 8: 1 fg.


2007) and was labelled using a DIG Oligonucleotide Tail-ing Kit (Roche diagnostics).

The presence and location of each parasite species wasconfirmed using hematoxylin-eosin (H&E) stained serialsections. The in situ hybridisation procedure was per-formed as in Bearham et al., 2007. The assay was optimisedby varying the Proteinase K incubation time and the con-centration of the DIG-labelled probe. The optimal Protein-ase K incubation time was found to be 15 min (at 50 lg/mL) while the optimal concentration of DIG-labelledprobe was found to be 2 ng/mL of hybridisation mix. Neg-ative controls included histologically uninfected sectionsand incubation with an identical hybridisation mix butcontaining an irrelevant DNA probe. Parasite identityand location were confirmed in adjacent sections stainedwith H&E.

2.5. Minchinia sp. diagnosis

In order to compare the diagnostic capabilities of theMinchinia sp. primers and the oligonucleotide ISH assayagainst the established techniques of histological examina-tion, 56 oysters were tested using all three techniques todetermine the presence of the Minchinia sp. in each individ-ual. Histological examination using H&E stained sectionswere first examined by the first author for approximately5 min. The PCR assay was performed next followed byISH. The oligonucleotide ISH assay was performed usingserial sections cut adjacent to the sections stained forH&E examination. Since the rate of detection in histologydepends in part on the ability of the operator, a retrospec-tive examination of the H&E stained serial sections wasperformed excepting sections that were negative by allthree tests. The use of serial sections between the ISHand H&E stained sections meant the retrospective histolog-ical examination could be performed in less than 5 minsince the location of the parasites could be identified withinthe positive ISH section. Each oyster was examined onlyonce by the PCR and ISH assays and only a single H&Esection was taken from each oyster.

The results of the comparison were evaluated usingstandard epidemiological methods (OIE, 2006) for sensi-tivity (proportion of oysters with Minchinia sp. that testpositive—proportion of true positives), specificity (propor-tion of oysters without Minchinia sp., that test negative—proportion of true negatives), positive predictive value(PPV; the probability that an oyster returning a positivetest is actually positive) and negative predictive value(NPV; the probability that an oyster that returns a nega-tive test actually does not have Minchinia sp.). The resultsfrom the molecular methods were assessed against each ofthe histological examinations. A chi-squared test was per-formed using SPSS for Windows (version 13.0) and com-pared the number of positives generated by eachdiagnostic test. The results for each diagnostic test weretreated as either positive or negative regardless of the levelof infection.

3. Results

3.1. Detection by polymerase chain reaction

The PCR primer pair SSF66/SSR69 amplified a 220 bpregion of the Minchinia sp. small subunit rRNA gene(Fig. 1). The assay did not amplify the SSU sequences fromH. nelsoni or H. costale infected oyster genomic DNA(Fig. 1). The primers did not produce a product from500 ng of genomic DNA from an oyster not infected withMinchinia DNA (Fig. 1). The product was easily detectableafter PCR amplification using 10 fg of Minchinia sp. ampli-fied DNA (Fig. 2). The primers amplified the Minchinia sp.


SSU rDNA from 50 ng of genomic DNA from an oysterhistologically positive for the Minchinia sp. (Fig. 2).

It was evident that increasing amounts of host DNAhindered the PCR when concentrations greater than250 ng were added to the reaction (Fig. 3). The PCR failedwhen 300 ng was added to the reaction.

3.2. In situ hybridisation: DNA probe specificity

In order to assess the specificity of the in situ hybridisa-tion assays using two different SSU rDNA probes, paraffinembedded samples of oysters infected with a variety of par-asites from the phyla Haplosporidia, and Ciliophora aswell as a paramyxian were subjected to in situ hybridisa-tion. Parasite identity and location were confirmed in adja-cent sections after staining with hematoxylin-eosin (H&E;Fig. 4).

Both the ISH probes were found to hybridise to all pres-porulating and sporulating stages of the Minchinia sp. inrock oyster sections (S. cuccullata; Fig. 4). The polynucleo-tide probe also detected intracellular B. roughleyi infectingthe digestive gland of a Sydney rock oyster (Fig. 5) and pro-duced positive results from the plasmodial stages of all ofthe other haplosporidian parasites tested (Figs. 5 and 6)including H. nelsoni (Fig. 5) and H. costale infecting thedigestive gland of an Eastern Oyster (Crassostrea virginica;Fig. 6) and M. teredinis from the gills of the shipworm Ter-

edo sp. (Fig. 6). Both probes produced little backgroundstaining and did not reproduce a signal in tissues processedfrom uninfected oysters or in negative control material(Figs. 4–6). Both probes failed to detect the paramyxianparasite (Marteilia sp.) or the ciliate (data not shown).

3.3. Sensitivity

Both probes recognised all presporulating Minchinia sp.stages in repeated assays including uni nucleate and bi

Fig. 3. Agarose gel electrophoresis of the SSF66/SSR69 PCR productswith of 1 pg of amplified Minchinia sp. DNA and varying amounts of hostDNA. Samples were electrophoresed in a 2% agarose gel for 45 min at90 V. Lane 1: PhiX/HaeIII molecular weight markers. The size of eachmarker in base pairs is indicated to the left. Lane 2: No DNA negativecontrol. Lane 3: Positive control. Lane 4: 200 ng of host DNA. Lane 5:250 ng. Lane 6: 300 ng. Lane 7: 400 ng. Lane 8: 500 ng.

nucleate stages (Fig. 4). Plasmodia also produced strongpositive results along with immature sporonts. The assaydid not detect Minchinia sp. spores.

3.4. Diagnosis comparison

Infection by Minchinia sp. was detected by histologicalexamination (an initial examination and a retrospectiveexamination) PCR and in situ hybridisation with the oligo-nucleotide probe. The parasites were generally present atlow levels of infection making them difficult to detect byhistological examination alone. Parasites were detected inthe reproductive follicles, digestive gland and gills of thehost. The initial histological examination detected 14infected oysters among the 56 oysters examined whilePCR amplification detected 26 infected oysters (Table 2).In situ hybridisation detected a total of 29 infected oysters(Table 2).

Initial histological examination diagnosed thirteen oys-ters as uninfected that were described as infected by PCR(Oysters 7, 8, 11, 14, 17, 18, 20, 35, 37, 43, 47, 53, and56; Table 2). These oysters were diagnosed as a light infec-tion by in situ hybridisation. Once PCR and ISH assayswere competed, a retrospective histological examinationwas performed. The retrospective histological examinationdetected parasites in eight of these oysters (oysters: 14, 17,18, 20, 33, 37, and 47) previously described as uninfected.The intensity of infection determined by histological exam-ination was often lower than that determined by in situ

hybridisation. PCR diagnosed three oysters as negativethat were described as a light infection by histologicalexamination and in situ hybridisation (oyster 10, 17 and43; Table 2).

The traditional diagnostic method provided an appar-ent Minchinia sp. prevalence of 25% (Table 3). In compar-ison, ISH provided the highest apparent prevalence of52% with the highest sensitivity of 100% and an NPVof 100%. However, the specificity and PPV of ISH wasrelatively low (64.2% and 48.2%; Table 3) compared tohistology because ISH allowed visualisation of low inten-sity Minchinia infections in 15 oysters which were negativeby histology. A chi-squared test detected a significantlydifferent result between the ISH assay and histology buta significantly different result could not be obtained com-pared to PCR. The retrospective histological examination,performed with the benefit of a serial ISH section, diag-nosed 23 positive oysters at 41% prevalence. A chi-squared test did not detect a significantly different resultbetween the retrospective examination and the PCR andISH assays.

Polymerase chain reaction had a sensitivity of 92.8%and this method recorded an apparent prevalence of46% (Table 3). The NPV of PCR was relatively low at75% since this method failed to detect three low intensityMinchinia sp. infections in three oysters, which werepositive in ISH and retrospective histological examina-tion. Specificity and PPV were again relatively low

Fig. 4. (A) Hematoxylin-eosin stained section containing Minchinia sp. in rock oyster gill tissue. (B) Serial section used in an in situ hybridisation with thepolynucleotide probe. Parasites are identified by a darker colouration. (C) Serial section used in an in situ hybridisation with the oligonucleotide probe.Sections are counterstained in Brazilian hematoxylin. (D) Negative control in situ hybridisation with an irrelevant probe. Scale bar is equal to 15 lm andapplies to each picture. Example parasites are indicated by the arrow.


(69% and 50%) compared to histology because PCRobtained positive results from 12 oysters which were neg-ative by histology (but positive by ISH). A chi-squaredtest detected a significantly different result between thePCR assay compared to the initial histologicalexamination.

When histology was compared to the pooled PCR andISH results, the molecular methods provided an apparentMinchinia sp. prevalence of 52% (Table 4). The sensitivityof histology compared to the pooled molecular methodswas relatively low at 48% with an NPV of 64%. This wasbecause of 15 false negative results when compared to themolecular methods. Specificity and PPV were both 100%for histology (Table 4).

4. Discussion

These results indicate that both the PCR assay devel-oped in this study and the in situ hybridisation assay devel-oped in Bearham et al. (2007) were capable of detectingMinchinia sp. in rock oysters. The comparison of diagnos-tic techniques suggested the ISH assay was the most sensi-tive followed by PCR (Tables 3 and 4).

4.1. Specificity and sensitivity of the PCR assay

The SSF66/SSR69 PCR assay targeted variable sectionswithin the SSU region of the parasite’s rRNA gene. Theassay recognised only the Minchinia sp. infected rock oys-

Fig. 5. (A) Hematoxylin-eosin (H&E) stained section containing Bonamia roughleyi in a Sydney rock oyster digestive diverticula. Scale bar = 5 lm. (B)Polynucleotide in situ hybridisation from the same region. Parasites are identified by a darker colouration. Scale bar = 15 lm. (C) Oligonucleotide in situ

hybridisation also from the same region. Scale bar = 5 lm. Sections are counterstained in a Brazilian hematoxylin. (D) H&E of Haplosporidium nelsoni

plasmodia in an Eastern Oyster (Crassostrea virginica) digestive gland. Scale bar = 5 lm. (E) Serial section used in an in situ hybridisation with thepolynucleotide probe. Parasites are identified by a darker colouration. Scale bar = 15 lm. (F) Serial section used in an in situ hybridisation with theoligonucleotide probe. Scale bar = 15 lm. Sections are counterstained in a Brazilian hematoxylin.


ter and did not detect either H. nelsoni or H. costale DNA(Fig. 1). Further testing is required using closely relatedMinchinia species such as M. tapetis and M. teredinis inorder to confirm the PCR assay does not cross-react withthese species. All of the negative control material includinguninfected oysters and no DNA controls were negative.

The amount of host DNA in the reproductive tissue canbe substantial and needs to be greatly diluted to avoid inhi-

bition of the PCR. A small quantity of target DNA in avery high amount of host DNA can result in a markeddecrease in PCR sensitivity, perhaps because of decreasedpotential for primer template binding (Zimmermannet al., 1994). The assay was successful in detecting 10 fgof template DNA in 250 ng of host DNA (Fig. 3).

The assay did appear to suffer from the focal or patchydistribution of the parasite in the host. One oyster pro-

Fig. 6. (A) Hematoxylin-eosin (H&E) stained section containing Minchinia teredinis in Teredo sp. (B) Serial section used in an in situ hybridisation withthe polynucleotide probe. Scale bar = 15 lm. Parasites are identified by a darker colouration. (C) Serial section used in an in situ hybridisation with theoligonucleotide probe. Scale bar = 15 lm. (D) H&E stained section containing of Haplosporidium costale plasmodia in an Eastern Oyster (Crassostrea

virginica) digestive gland. Scale bar = 10 lm. (E) Serial section used in an in situ hybridisation with the polynucleotide probe. Parasites are identified by adarker colouration. Scale bar = 10 lm. (F) Serial section used in an in situ hybridisation with the oligonucleotide probe. Scale bar = 10 lm. Sections arecounterstained in a Brazilian hematoxylin.


duced a negative result by PCR which was subsequentlyfound to be positive by the initial histological examination(oyster 10; Table 2) while two oysters were found to be verylightly infected in the retrospective histological examina-tion (oysters 17 and 43; Table 2). All three of these oysterswere diagnosed as infected by the oligonucleotide ISHassay.

While PCR does not allow quantification of infectionintensity, it did detect low intensity infections in themajority of cases when the results of the tests were com-pared for each oyster (Table 2). Histological examinationdid not detect fifteen oysters described as infected by PCRand/or in situ hybridisation. Subsequent re-examination ofthese oysters resulted in an additional nine infections

Table 2Comparison of histology, in situ hybridisation (ISH) and PCR for detecting Minchinia sp. in 56 rock oysters

Oyster number Histology PCR ISH Retrospective histology All tests combined

1 Heavy Positive Heavy Heavy Heavy2 Medium Positive Heavy Medium Heavy3 Negative Negative Negative Negative4 Light Positive Light Light Light5 Medium Positive Medium Medium Medium6 Negative Negative Negative Negative7 Negative Positive Light Negative Light8 Negative Positive Light Negative Light9 Negative Negative Negative Negative

10 Light Negative Light Light Light11 Negative Positive Light Negative Light12 Negative Negative Negative Negative13 Negative Negative Negative Negative14 Negative Positive Light Very lightb Light15 Negative Negative Negative Negative16 Negative Negative Negative Negative17 Negative Negative Light Very lightb Light18 Negative Positive Light Very lightb Light19 Negative Negative Negative Negative20 Negative Positive Light Very lightb Light21 Light Positive Medium Light Medium22 Negative Negative Negative Negative23 Light Positive Light Light Light24 Negative Negative Negative Negative25 Light Positive Light Light Light26 Negative Positive Light Negative Light27 Negative Negative Negative Negative28 Negative Positive Light Very lightb Light29 Light Positive Light Light Light30 Negative Negative Negative Negative31 Negative Negative Negative Negative32 Negative Negative Negative Negative33 Negative Positive Light Very lightb Light34 Light Positive Light Light Light35 Negative Negative Negative Negative36 Medium Positive Medium Medium Medium37 Negative Positive Light Very lightb Light38 Negative Negative Negative Negative39 Negative Negative Negative Negative40 Negative Negative Negative Negative41 Negative Negative Negative Negative42 Light Positive Medium Light Light43 Negative Negative Light Very lightb Light44 Negative Negative Negative Negative45 Light Positive Medium Light Light46 Negative Negative Negative Negative47 Negative Positive Light Very lightb Light48 Negative Negative Negative Negative49 Negative Negative Negative Negative50 Light Positive Light Light Light51 Negative Negative Negative Negative52 Negative Negative Negative Negative53 Negative Positive Light Negative Light54 Negative Negative Negative Negative55 Negative Negative Negative Negative56 Negative Positive Very Light Negative Very lightDiagnosed positive (n) 14 26 29 23 29Apparent prevalence 25% 46% 52% 41% 52%False negativesa 26.8% 5.4% 0% 10.7% 0%Chi-square statistic 14.0 0.286 0.71 1.786df 1 1 1 1Significance 0.000 0.593 0.789 0.181

Heavy is defined as more than 20 parasites per 20· field of view, medium is between 10 and 20 parasites per 20· field of view light is between 4 and 10 parasites per20· field of view while very light refers between 1 and 3 parasites per 20· field of view.

a False negatives defined relative to the all tests combined category.b Represents a change between the initial histological examination and the retrospective histological examination. Diagnoses in bold are in disagreement with the

alternative tests.


Table 3Evaluation of PCR and ISH relative to histology

PCR (%) ISH (%) Histology (%)

Sensitivity 92.8 100 100Specificity 69.0 64.2 100PPVa 50 48.2 100NPVa 75 100 100Apparent prevalence 46 52 25True prevalence 25 25 25

a PPV is the positive predictive value, NPV is the negative predictivevalue.

Table 4Evaluation of histology compared to ISH and PCR

Histology (%) PCR/ISH (%)

Sensitivity 48 100Specificity 100 100PPVa 100 100NPVa 64 100Apparent prevalence 25 52True prevalence 52 52

a PPV is the positive predictive value, NPV is the negative predictivevalue.


being detected (Table 2). The retrospective examinationindicated the parasite was present in the H&E sectionsbut was not easily detectable after 5 min of searching. Itis possible that the efficiency of detection for histologicalexamination may be different for alternative personneldepending on their skill, level of experience and the lengthof time they spend examining the sections. The retrospec-tive histological examination did not detect infection insix oysters described as infected by PCR and in situ

hybridisation. Overall, the PCR assay was found to bemore sensitive than the initial histological examinationfor detecting the Minchinia sp. infecting rock oysters(Tables 2–4). No statistically significant difference wasdetected between the PCR and ISH assays and thereforePCR assay may provide an inexpensive method fordetecting the parasite.

4.2. The 30 bp oligonucleotide ISH assay

The 30 bp oligonucleotide probe targeted a variable sec-tion of the SSU region of the parasite’s rRNA gene in orderto confer a higher level of specificity than the polynucleo-tide ISH probe. The oligonucleotide probe only recognisedthe Minchinia sp. infecting rock oysters. The probe did notrecognise closely related haplosporidian parasites such asM. teredinis, B. roughleyi, H. costale or H. nelsoni orcross-react with any of the host species tested (Figs. 4–6).Consequently, the ISH assay utilising the 30 bp oligonu-cleotide probe is likely to be species specific. It is impossibleto ensure the assay does not react to uncharacterised haplo-sporidian parasites. The assay was successful in localisingindividual parasites thereby detecting low levels ofinfection.

The comparison of diagnostic techniques summarised inTable 2 suggests the oligonucleotide ISH assay is more sen-sitive than the initial histological examination (Tables 2–4).These data highlights the difficulty detecting low numbersof relatively cryptic parasites in H&E sections that maybe present in a variety of different tissue types. The oligonu-cleotide ISH assay detected the highest number of infectedoysters in the comparison (29 from 56 oysters examined;Table 2). The sensitivity of the assay means the pathogenis detected earlier in the infection cycle and lower levelsof infection can be detected. In situ hybridisation costs con-siderably more that PCR and therefore it may be most sui-ted to detecting the pathogen in low numbers of oysterswhere high sensitivity is required.

Further testing of rock oysters infected with Minchinia

sp. is required to ensure the assay detects all strains ofthe parasite throughout its range. The oligonucleotideassay will be a useful confirmatory diagnostic test to ensureno contamination of PCR has occurred since it allows visu-alisation of the parasite within host tissues.

4.3. The polynucleotide ISH assay

A polynucleotide ISH probe was used by Cochennecet al. (2000) to detect B. ostreae in Ostrea edulis and wasfound to also detect Bonamia sp. in infected Tiostrea chilen-

sis and H. nelsoni in infected C. virginica. The 166 bp poly-nucleotide probe obtained from the Minchinia sp. producedstrong hybridisation signals with all of the haplosporidianparasites tested (Minchinia sp., M. teredinis, H. nelsoni, H.

costale and B. roughleyi) and did not cross-react with anyhost tissues in the five bivalve species tested (S. cuccullata,Saccostrea glomerata, C. virginica, Pinctada maxima andTeredo sp.; Figs. 5 and 6). Further testing is required inorder to assess whether the assay recognises basal or puta-tive haplosporidian parasites such as the parasite detectedin New Zealand paua, Haliotis iris (Diggles et al., 2002)or Urosporidium species. However, since the polynucleotideprobe recognised three of the four Haplosporidia genera(Minchinia, Haplosporidium and Bonamia) the ISH assaymay be a phylum specific test. The polynucleotide ISHprobe failed to recognise Marteilia sp. from rock oysters(S. cuccullata) or the Rhynchodid-like ciliate (Ciliophora).In addition, incidental parasites that did not belong to theHaplosporidia, found in rock oysters (copepods, grega-rines) did not cross-react (unpublished observations).

Disease has not been a problem for the mollusc indus-tries of Western Australia. However, it is certain that manymore pathogenic organisms remain to be discovered, espe-cially as molluscs become subject to aquaculture or aresubjected to environmental stresses associated with eco-nomic activity (Jones and Creeper, 2006). Because of theage of the Australian continent and its relative isolation,many of these pathogens may prove to be unique to Wes-tern Australia (Jones and Creeper, 2006). The polynucleo-tide ISH assay may be a useful tool for disease surveys ofwild bivalve populations for haplosporidian parasites at


low levels of infection. Many haplosporidians are known tocause considerable losses in naı̈ve hosts following translo-cation (Burreson and Ford, 2004; Burreson et al., 2000).Eradication of these parasites, once they have becomeestablished is usually not feasible. The polynucleotideISH assay is able to detect a variety of haplosporidian par-asites in a variety of bivalve species. Thus, it may providetarget tissues for further analysis in the detection of unde-scribed or difficult to detect haplosporidian parasites at lowlevels of infection. Once a positive result has been obtained,the polynucleotide ISH assay must be used in conjunctionwith alternative assays (such as PCR, histological examina-tion or electron microscopy) that will confirm the haplo-sporidian infection and allow speciation. It is possiblethat the polynucleotide assay may cross-react with otheruntested parasite groups and consequently needs to be usedwith caution just as it is also possible for untested haplo-sporidian parasite species to cross-react with the oligonu-cleotide probe.

Neither ISH assay recognised mature haplosporidianspores. The inability of the assay to detect mature sporesis consistent with other haplosporidian parasites. Stokesand Burreson (1995) and Stokes et al. (1995) reported sim-ilar results in two haplosporidian species (H. nelsoni andM. teredinis). These findings were attributed to the inabilityof either the probe and/or the anti-dioxigenin antibody topenetrate the spore wall. The lack of spore recognitionmay hamper the use of the assay in life cycle studies, suchas the identification of an intermediate host. These resultscontrast with Carnegie et al. (2006) whose ISH assay didrecognise the spores of Bonamia perspora.

The PCR and in situ hybridisation tests developed andoptimised in this study should prove a valuable researchtool. Highly sensitive and reliable detection methods forthe Minchinia sp. infecting rock oysters will facilitate andspeed up further research into this parasite and the interac-tions between the parasite and its host.

Acknowledgments

We thank Nancy Stokes at the Virginia Institute of Mar-ine Science for providing H. nelsoni, H. costale andM. teredinis sections as well as H. nelsoni and H. costale

DNA. Our gratitude also goes to Michael Slaven and Ger-ard Spoelstra (Murdoch University) who undertook thehistological preparations for the study. The authors alsothank Dr. Patrick Shearer for his editorial comments andwould like to acknowledge the contribution of the review-ers to improving this manuscript. This work is supported

by the Australian Government’s Fisheries Research andDevelopment Corporation Project No. 2006/064, MurdochUniversity and the Pearl Producers Association.

References

Anon, 2002. Import risk analysis (IRA) of non-viable bivalve molluscs,Technical issues paper, Australian Government. Agriculture, Fisheriesand Forestry.

Bearham, D. et al., 2007. Molecular characterisation of a Haplosporidianparasite infecting rock oysters Saccostrea cuccullata in north WesternAustralia. J. Invertebr. Pathol. 95, 33–40.

Burreson, E.M., 2000. Disease diagnosis by PCR: foolproof or foolhardy?J. Shellfish Res. 19, 642.

Burreson, E.M., Ford, S.E., 2004. A review of recent information on theHaplosporidia, with special reference to Haplosporidium nelsoni (MSXdisease). Aquat. Living Resour. 17, 499–517.

Burreson, E.M. et al., 2000. Increased virulence in an introducedpathogen: Haplosporidium nelsoni (MSX) in the eastern oyster Cras-

sostrea virginica. J. Aquat. Anim. Health, 1–8.Carnegie, R.B. et al., 2006. Bonamia perspora n.sp (Haplosporidia), a

parasite of the oyster Ostreola equestris is the first Bonamiaspecies known to produce spores. J. Eukaryot. Microbiol. 53,232–245.

Carnegie, R.B. et al., 2003. Molecular detection of the oyster parasiteMikrocytos mackini, and a preliminary phylogenetic analysis. Dis.Aquat. Organ. 54, 219–227.

Cochennec, N. et al., 2000. Detection of Bonamia ostreae based on smallsubunit ribosomal probe. J. Invertebr. Pathol. 76, 26–32.

Diggles, B.K. et al., 2002. Pathology of the cultured paua Haliotis iris

infected with a novel haplosporidian, with some observations on thecourse of disease. Dis. Aquat. Organ. 50, 219–231.

Hine, P.M., Thorne, T., 2002. Haplosporidium sp. (Alveolata: Haplospo-ridia) associated with mortalities among rock oysters Saccostrea

cuccullata in north Western Australia. Dis. Aquat. Organ. 51, 123–133.Jones, J.B., Creeper, J.H., 2006. Diseases of pearl oysters and other

molluscs: a Western Australian perspective. J. Shellfish Res. 25, 233–238.

Kleeman, S.N. et al., 2002. Specificity of PCR and in situ hybridizationassays designed for detection of Marteilia sydneyi and M. refringens.Parasitology 125, 131–141.

Kumar, S. et al., 2004. MEGA3: integrated software for molecularevolutionary genetic analysis and sequence alignment. Brief. Bioin-form., 5.

Mialhe, E. et al., 1995. Future of biotechnology-based control of diseasein marine invertebrates. Mol. Mar. Biol. Biotechnol. 4, 275–283.

OIE, Manual of diagnostic tests for aquatic animals, 2006.Stokes, N.A., Burreson, E.M., 1995. A sensitive and specific DNA probe

for the oyster pathogen Haplosporidium nelsoni. J. Eukaryot. Micro-biol. 42, 350–357.

Stokes, N.A. et al., 1995. Small subunit ribosomal RNA gene sequence ofMinchinia teredinis (Haplosporidia: Haplosporidiidae) and a specificDNA probe and PCR primers for its detection. J. Invertebr. Pathol.65, 300–308.

Zimmermann, K. et al., 1994. Nested primer PCR detection limits ofHIV-1 in the background of increasing numbers of lysed cells.Biotechniques 17, 18–20.

detection of minchinia sp., in rock oysters saccostrea cuccullata (born, 1778) using dna probes

Documents