Department of Production Animal Medicine Faculty of Veterinary Medicine

University of Helsinki Helsinki, Finland

Staphylococcus aureus in bovine intramammary infection: molecular, clinical and epidemiological characteristics

ACADEMIC DISSERTATION

By Maarit Haveri

To be presented with the permission of the Faculty of Veterinary Medicine, University of Helsinki, for public examination in Walter Hall, Agnes Sjöbergin katu 2, Helsinki,

on November 21st 2008, at 12 noon.

Supervised by

Professor Satu Pyörälä Department of Production Animal Medicine,

Faculty of Veterinary Medicine, University of Helsinki, Finland

and

Docent Jaana Vuopio-Varkila, MD, PhD

Department of Bacterial and Inflammatory Diseases, Hospital Bacteria Laboratory,

National Public Health Institute, Helsinki, Finland

Reviewed by

Professor Julie Fitzpatrick

Moredun Research Institute, Scotland, UK

and

DVM, PhD Theo Lam

UGCN, Animal Health Service, the Netherlands

Opponent

Tore Tollersrud, DVM, PhD Department of Animal Health, Section for Immunoprofylaxis, National Veterinary Institute,

Oslo, Norway

ISBN 978-952-92-4614-4 (Paperback) ISBN 978-952-10-5034-3 (PDF)

Helsinki University Print Helsinki 2008

To my family

TABLE OF CONTENTS

ABSTRACT ............................................................................................................................. 7

LIST OF ORIGINAL ARTICLES ........................................................................................ 9

ABBREVIATIONS ............................................................................................................... 10

1. INTRODUCTION............................................................................................................. 11

2. REVIEW OF THE LITERATURE................................................................................. 12

2.1 Epidemiology and characteristics of bovine Staphylococcus aureus intramammary infection ............................................................................................... 12

2.1.1 Prevalence and significance ................................................................................... 12 2.1.2 Reservoirs and transmission................................................................................... 13 2.1.3 Clinical manifestation and outcome....................................................................... 14 2.1.4 Antimicrobial resistance of bovine Staphylococcus aureus................................... 15 2.1.5 Control.................................................................................................................... 15

2.2 Pathogenesis of bovine Staphylococcus aureus intramammary infection .................... 16 2.2.1 Entry and attachment of Staphylococcus aureus into the mammary gland ........... 16 2.2.2 Interaction between Staphylococcus aureus and the bovine immune system........ 17

2.3 The role of bacterial virulence factors in Staphylococcus aureus intramammary infection........................................................................................................................ 17

2.3.1 Control and expression of virulence factors........................................................... 18 2.3.2 Virulence factors involved in establishing an infection ......................................... 19 2.3.3 Virulence factors involved in maintenance of infection ........................................ 20 2.3.4 Immunity to Staphylococcus aureus ...................................................................... 21

2.4 Typing of bovine Staphylococcus aureus ..................................................................... 21 2.4.1 Phenotypic methods ............................................................................................... 21 2.4.2 DNA-based methods .............................................................................................. 22

3. AIMS OF THE STUDY.................................................................................................... 24

4. MATERIALS..................................................................................................................... 25

4.1 Origin of Staphylococcus aureus isolates ..................................................................... 25 4.2 Study animals and herds................................................................................................ 25

5. METHODS ........................................................................................................................ 26

5.1 Clinical examination of the cows and mastitis follow-up (II-IV) ................................. 26 5.2 Sample collection .......................................................................................................... 26

5.2.1 Milk samples .......................................................................................................... 26 5.2.2 Samples from extramammary sites (IV) ................................................................ 26

5.3 Determination of indicators of inflammation in the milk ............................................. 27

5.3.1 Milk somatic cell count .......................................................................................... 27 5.3.2 Milk N-acetyl-β-D-glucosaminidase (II) ............................................................... 27

5.4 Bacterial isolation and conventional identification....................................................... 27 5.4.1 In milk samples ...................................................................................................... 27 5.4.2 In other samples (IV).............................................................................................. 28

5.5 Antimicrobial susceptibility testing .............................................................................. 28 5.5.1 Testing for β-lactamase production........................................................................ 28 5.5.2 Determination of MIC using the microdilution method (I-II)................................ 28 5.5.3 Detection of genes encoding β-lactamase and methicillin resistance (III-IV) ....... 29

5.6. Characterisation of Staphylococcus aureus isolates using DNA-based methods (II-IV) ............................................................................................................ 29

5.6.1 DNA isolation (II-IV)............................................................................................. 29 5.6.2 Molecular typing using PFGE (II-IV).................................................................... 29 5.6.3 Detection of genes encoding virulence and adhesion factors using polymerase chain reaction (III-IV) ........................................................................ 29

5.7 Classification of clinical severity and persistence of IMI ............................................. 33 5.8 Bacteriological definitions ............................................................................................ 33 5.9 Statistical analyses......................................................................................................... 33

6. RESULTS........................................................................................................................... 35

6.1 Prevalence and occurrence of Staphylococcus aureus (I, IV)....................................... 35 6.1.1 In the herds included in the Finnish national mastitis survey (I) ........................... 35 6.1.2 In IMI and extramammary sites (IV) ..................................................................... 36

6.2 Diversity of Staphylococcus aureus strain-types originating from mastitis infections (II-III) .......................................................................................................... 37

6.2.1 Diversity and distribution of pulsotypes and gene profiles.................................... 37 6.2.2 Diversity within and between farms....................................................................... 40 6.2.3 Diversity within cows............................................................................................. 40

6.3 Antimicrobial resistance of Staphylococcus aureus (I-III) ........................................... 40 6.3.1 Resistance to penicillin (I-III) ................................................................................ 40 6.3.2 Resistance to other antimicrobials (I)..................................................................... 41

6.4 Relationship of Staphylococcus aureus strain-types to the inflammatory response of the affected quarter, clinical characteristics and persistence of mastitis (II-III) ......................................................................................................... 41

6.4.1 Staphylococcus aureus pulsotypes (II)................................................................... 41 6.4.2 Staphylococcus aureus pulsotypes and virulence gene profiles (III)..................... 42 6.4.3 The role of Staphylococcus aureus pulsotypes, virulence gene profiles, and penicillin resistance in persistent mastitis (III) ............................................................... 43

6.5 Staphylococcus aureus strain types in IMI and extramammary sites (IV) ................... 44

7. DISCUSSION .................................................................................................................... 47

7.1 Prevalence and occurrence of Staphylococcus aureus .................................................. 47 7.2. Diversity and distribution of Staphylococcus aureus strain-types ............................... 48 7.3. Resistance of S. aureus to penicillin and other antimicrobials .................................... 50 7.4. The role of strain characteristics in manifestation and persistence

of Staphylococcus aureus IMI...................................................................................... 53 7.5 Reservoirs and transmission of Staphylococcus aureus................................................ 57 7.6 Future considerations for control of S. aureus mastitis................................................. 58

8. CONCLUSIONS................................................................................................................ 61

9. ACKNOWLEDGEMENTS.............................................................................................. 63

10. REFERENCES................................................................................................................ 65

7

ABSTRACT

Staphylococcus aureus is one of the most common causes of bovine intramammary infections and mastitis in modern dairies worldwide and the most common mastitis pathogen isolated from raw milk. S. aureus is easily transmissible and infections caused by S. aureus respond poorly to treatment. The economic losses due to S. aureus mastitis are considerable, and include decrease in milk production, reduced milk quality through contamination by bacteria and increased milk SCC, veterinary and treatment costs, premature culling of cows and loss of genetic potential. Mastitis caused by S. aureus also adversely affects welfare of dairy cows. This thesis focuses on determination of the prevalence of S. aureus IMIs in Finland (I), investigating the antimicrobial resistance of S. aureus isolated from IMIs (I), in particular resistance against penicillin and methicillin (II-IV), and on studying the relationship between S. aureus strain-types and distribution (II-III), clinical severity (II-III), persistence (II-III), and transmission and reservoirs of intramammary infections (IV). Study I represents the results of a third nationwide mastitis survey conducted in 2001 in randomly selected Finnish dairy herds. S. aureus was isolated from 430 (3.4%) of the 12 661 quarters of 334 (10.2%) of the 3282 cows sampled, and from 140 (64.8%) of the 216 herds included in the survey. The results indicate that S. aureus remains a significant mastitis pathogen in Finland as both the prevalence of quarters and the prevalence of cows infected by S. aureus has remained nearly stable. The proportion of S. aureus representing a mastitis-causing agent decreased from the previous survey of 1995, but the total number of quarters that were positive for bacterial growth significantly increased. Compared with the previous surveys, the proportion of penicillin resistant, β-lactamase producing S. aureus remained almost stabile but was still high, approximately 50%. In almost half of the survey herds (study I), penicillin resistant S. aureus were isolated and were distributed throughout dairy herds and persisted for long periods, as shown in study IV. In studies II-III, clinical data and bacterial isolates from field cases of S. aureus mastitis were collected in the practice area of the Ambulatory Clinic, Faculty of Veterinary Medicine. The isolates were typed using PFGE and a selection of different virulence determinants and those encoding penicillin and methicillin resistance were identified with PCR. Among the diverse number of pulsotypes identified with PFGE typing, strains of five pulsotypes dominated as the cause of S. aureus mastitis. One pulsotype was associated with more severe signs and a more acute local response, as determined by milk NAGase activity, but was almost always eliminated from the mammary gland. There were numerous different gene profiles. The genes encoding haemolysins and LukED leukocidin were ubiquitous, whereas those for SAgs were more or less pulsotype-linked. At least one SAg-encoding gene was present in the majority of the predominant pulsotypes. Multiple genes for SAgs existed in combinations, which suggests that they are located in genetic clusters or pathogenicity islands. The mecA gene for oxacillin (methicillin) resistance was not detected. S. aureus strains that had the enterotoxin-encoding genes sed and sej were overrepresented in persistent mastitis. sed and sej permanently co-existed with the penicillin-resistance gene blaZ, suggesting the presence of a penicillinase plasmid in these strains. Within strains of pulsotype A, which typically harboured sed, sej, and blaZ, those positive for sed, sej, and blaZ persisted significantly more compared with the strains containing blaZ only. This suggests

8

that among penicillin-resistant blaZ positive strains, a subgroup of strains containing sed and sej may be more likely linked with chronic mastitis than the other subgroups. S. aureus isolates in study IV were derived from cases of mastitis during one year and those collected from extramammary sites (teat skin, teat canals and skin lesions, milking liners, hands and nostrils of the milking personnel) at one farm visit from two dairy herds with different management regimes. Selected isolates were subjected to PFGE typing and PCR analysis for genes coding for virulence and adhesion factors. Nine PFGE types were found, the two herds each having their own predominant type and sharing one type. The predominant types were present in all sites studied in both herds, suggesting that the extramammary sites can be important reservoirs for intramammary infections and mastitis, or that the strains infecting the mammary gland can spread to body sites and in the environment. In both herds, strains of the predominant pulsotypes contained adhesion genes fnbA, and fnbB, whereas only fnbA was detected in strains of the other pulsotypes. This could indicate that strains with fnbA, and fnbB exhibit higher adhesion capacity and consequently have an advantage over the other strains in the herds.

9

LIST OF ORIGINAL ARTICLES

This thesis is based on the following original articles, referred to in the text by their Roman numerals I-IV: I Pitkälä, A., M. Haveri, V. Myllys, S. Pyörälä, and T. Honkanen-Buzalski. 2004. Bovine mastitis in Finland 2001--prevalence, distribution of bacteria, and antimicrobial resistance. Journal of Dairy Science 87:2433-2341. II Haveri, M., S. Taponen, J. Vuopio-Varkila, S. Salmenlinna, and S. Pyörälä.

2005. Bacterial genotype affects the manifestation and persistence of bovine Staphylococcus aureus intramammary infection. Journal of Clinical Microbiology 43:959-61.

III Haveri, M. A. Roslöf, L. Rantala, and S. Pyörälä. 2007. Virulence genes of

bovine Staphylococcus aureus from persistent and non-persistent intramammary infections with different clinical characteristics. Journal of Applied Microbiology 103:993-1000.

IV Haveri, M., M. Hovinen, A. Roslöf, and S. Pyörälä. 2008. Molecular types and

genetic profiles of Staphylococcus aureus isolated from bovine intramammary infections and extramammary sites. Journal of Clinical Microbiology. Accepted 2008.

The original articles are reprinted with the kind permission of their copyright holders: American Dairy Society (I), American Society of Microbiology (II, IV), and Blackwell Publishing (III).

10

ABBREVIATIONS

ATCC American Type Culture Collection BMSCC bulk milk somatic cell count CCUG Culture Collection, University of Göteborg CFU colony forming unit CMT California mastitis test CNS coagulase-negative staphylococci Evira Finnish Food Safety Authority FnBP fibronectin-binding protein IMI intramammary infection MHC major histocompatibility complex MIC minimum inhibitory concentration MLST multilocus sequence typing MRSA methicillin-resistant Staphylococcus aureus MSCRAMM microbial surface components recognizing adhesive matrix

molecules NAGase N-acetyl-β-D-glucosaminidase PCR polymerase chain reaction PFGE pulsed-field gel electrophoresis PMN polymorphonuclear cell RAPD random amplified polymorphic DNA REFP restriction enzyme fragmentation pattern SAg superantigen SCC somatic cell count UH University of Helsinki

11

1. INTRODUCTION

Staphylococcus aureus is commonly resident on skin and mucous membranes of cattle. Despite its generally benign nature, in changing circumstances, as response to damage and exposure of structures below the epithelial or mucosal surface, S. aureus can behave as a pathogen. The induction of intramammary infection requires production of a variety of virulence factors enabling adherence, colonisation, invasion of the mammary cells of the bovine host by the S. aureus cells, evasion of the immune defence mechanism and survival in the host environment. The possession of virulence factors, which seems to be always increasing, varies among different S. aureus strains. Some S. aureus strains are better adapted to infect the bovine mammary gland, whereas others are less important as a cause of mastitis. Many factors, including the use of antimicrobials, management of the herd, and general and individual characteristics of the host, create continuous selective pressure on S. aureus strains. Beyond this pressure, genetic material is exchanged between the strains, and those strains better adapted to survive are selected. During this evolutionary process S. aureus strains that are more virulent or more resistant to antimicrobials are likely to develop. Identification of these strains is possible using the tools of molecular microbiology. Abundant evidence is available that virulence of S. aureus strains varies. Despite the fundamental and widely used measures to control and prevent S. aureus mastitis, this microorganism can cause herd-wide outbreaks and persist for long periods. It seems likely that the currently employed control measures are limited regarding prevention and control of the spread of particular S. aureus strains. The focus of this thesis was on characterisation of S. aureus mastitis in Finland. Prevalence, antibiotic resistance and epidemiology of intramammary infections caused by S. aureus were studied. Isolates from bovine IMI were investigated to determine whether differences in clinical significance exist between strains and to establish if certain strain types or determinants for virulence can be specifically linked with mastitis. The information presented in this thesis will be useful for planning more efficient strategies to control and prevent S. aureus mastitis.

12

2. REVIEW OF THE LITERATURE

Infections caused by Staphylococcus aureus (S. aureus) were characterised first by Sir Alexander Ogston, a Scottish surgeon, more than one hundred years ago. In 1890 S. aureus was reported to cause mastitis in cattle. S. aureus is a gram-positive, catalase-positive, usually oxidase-negative, facultative anaerobic coccus, which belongs to the family of Micrococcaceae and the group of staphylococci. S. aureus can be distinguished from other staphylococcal species on the basis of gold colony pigmentation, production of coagulase, fermentation of mannitol and trehalose, and production of heat stable thermonuclease. Most S. aureus isolates are enclosed in a polysaccharide capsule, which can be categorised into eleven different serotypes. Beneath the capsule there is a cell wall with a thick and highly cross-linked peptidoglycan layer and teichoic acid, which is typical of gram-positive bacteria (van Wely et al., 2001). The surface proteins of S. aureus share a common overall structure and are involved in binding to fibrinogen, fibronectin, collagen and IgG. Penicillin-binding proteins, PBP1-4, produced by S. aureus are involved in the cell wall peptidoglycan assembly. These proteins are able to bind to beta-lactam antibiotics. The universal features of S. aureus genome are circular 2800 kilo base pair chromosome, one or more plasmids, prophages, transposons, insertion sequences and other incompletely characterised accessory genetic elements (Projan and Novick, 1997). To date, more than 10 complete S. aureus genomes have been sequenced. The core genome of S. aureus, which includes the housekeeping genes essential for cell growth and division, is relatively conserved. Up to 20% of S. aureus chromosomes consist of mobile genetic elements, which are defined DNA segments that can replicate on their own. The major types of these elements include plasmids, S. aureus pathogenicity islands, bacteriophages, insertion sequences, and staphylococcal cassette chromosomes. Since many of these elements encode virulence or antimicrobial resistance genes, transfer of these segments into and out of S. aureus strains can affect the strain pathogenicity. The distributions of mobile genetic elements vary markedly between S. aureus strains and even within same strain lineages (Holden and Lindsay, 2008).

2.1 Epidemiology and characteristics of bovine Staphylococcus aureus intramammary infection

2.1.1 Prevalence and significance

S. aureus is one of the most frequently isolated pathogens in bovine IMI worldwide (Aarestrup et al., 1995a; Myllys et al., 1994a; Myllys et al., 1998; Acuna et al., 2001; Chaves et al., 2001; Gianneechini et al., 2002; Barrett et al., 2005; Tenhagen et al., 2006) and the most common contagious mastitis pathogen isolated from raw milk (Piccinini et al., 2003b; Olde Riekerink et al., 2006). Studies from about 25 years ago estimated that 100% of the dairy herds in California (Gonzalez et al., 1988), 48% in Vermont (Goldberg et al., 1991) and 93% in Great Britain (Wilson and Richards, 1980) included cows infected with S. aureus. The proportion of cows with S. aureus IMI has been approximately 21% in Great Britain (Wilson and Richards, 1980), 11% in Norway (Bakken, 1981), 10% in Denmark, (Aarestrup et al., 1995b), 10% in

13

Finland (Myllys et al., 1998), 9% in California (Gonzalez et al., 1988) and 5% in Vermont, USA (Goldberg, 1991; Hogan, 1997). Long-term surveys suggest that the significance of S. aureus in the dairy industry has remained unchanged (Sol, 2002; Swinkels et al., 2005). S. aureus IMI causes economic losses in several ways: decrease in milk production, reduced milk quality brought about by bacterial contamination and increased milk SCC, veterinary and treatment costs, requirement for premature culling and loss of genetic potential (Østerås et al., 2005; Hogeveen and Østerås, 2005). Cows with substantial S. aureus growth in milk samples are at high risk of being culled (Reksen et al., 2006). A study by Piccinini et al. (2003a) reported substantial milk losses in herds with S. aureus mastitis compared with herds in which the pathogen was not detected. The relative losses were greatest in herds with fewer than 150 cows. The losses listed above were of direct economic importance, but mastitis is also an animal welfare issue. Economic losses due to S. aureus mastitis may be higher than for an average case of mastitis, especially in primiparous cows (Gröhn et al., 2004). Multiparous cows are generally more often infected with S. aureus as compared with heifers (McDougall et al., 2007). However, high prevalences of S. aureus IMI have occasionally been reported for heifers. Fox et al. (1995) reported that 9% of refreshening heifers in Louisiana had S. aureus mastitis. Waage et al. (1999) recorded S. aureus in approximately 44% of the bacteriologically-positive quarters of 1040 heifers with clinical mastitis. Myllys et al. (1995a) isolated S. aureus in 20% of the quarter secretion samples taken from 265 heifers in 160 Finnish dairy farms. Under certain environmental conditions, many S. aureus strains are able to produce a variety of enterotoxins, which can cause food poisonings (Jørgensen et al., 2005b). These toxins are not inactivated by pasteurisation or other heat treatment of milk. S. aureus can also be a zoonotic pathogen, even though it is generally regarded as rather host-specific. However, direct transmission of MRSA between cows and humans was recently reported (Juhász-Kaszanyitzky et al., 2007).

2.1.2 Reservoirs and transmission

Anterior nares are the natural niche of S. aureus in humans and the microbe is frequently present on the skin, skin glands and mucous membranes of many individuals. Approximately 20% of healthy people are persistently colonised by S. aureus in the anterior nares, and up to 70% carry it at least intermittently (Kluytmans et al., 1997). Presence of S. aureus is regarded as a risk factor for staphylococcal infection (Saginur et al., 1996). Dairy cows, among other domestic artiodactyls, are considered to be temporary hosts of S. aureus, in which the microbe is frequently present as a contaminant that can multiply and persist for short periods (Kloos, 1997). S. aureus has been isolated from practically all external surfaces of healthy cows (Davidson et al., 1961; Matos et al., 1991; Roberson et al., 1994; Roberson et al., 1998), udder skin being the preferred site (Schalm et al., 1971; Jørgensen et al., 2005a). Long-term colonisation by S. aureus on teat skin and several other body sites, primarily mucosal external orifices, have been observed in heifers (Roberson et al., 1994) suggesting persistent colonisation. It its likely that a calf has its first contact with S. aureus already at birth via the genital area of the mother. Even though S. aureus has been found to survive in the barn environment, such as in bedding material, on the floor, in dust, and in feed (Davidson et al., 1961; Davidson et al., 1963; Matos et al., 1991), it cannot be considered an environmental bacterium (Kloos, 1997).

14

Mastitis caused by S. aureus is considered to be contagious, as the diversity of mastitis-causing strains is low, suggesting a common source of infection (Larsen et al., 2000a; Buzzola et al., 2001) and control programs planned for contagious mastitis have reduced occurrence of S. aureus (Hoblet and Miller, 1991; Wilson et al., 1995). Infected milk is generally considered to be the primary source of the microbe, and milking liners are the main vector of transmission of mastitis-causing S. aureus, since they have been frequently shown to be contaminated with similar S. aureus strains to those in infected milk (Fox et al., 1991; Larsen et al., 2000a; Zadoks et al., 2002; Jørgensen et al., 2005a). Traumatised sites such as abrasions on teats, legs, bends and navel, typically infected by S. aureus, are regarded as secondary sources of S. aureus causing IMI. S. aureus is regarded as a clonal organism as the populations consist of groups of genetically related strains with a common ancestor. S. aureus diversifies more often by point mutation than by horizontal gene transfer (Feil et al., 2003). S. aureus strains isolated from mastitic milk and different body sites in dairy cows have been diverse (Joo et al., 2001; Middleton et al., 2002a; Zadoks et al., 2002), but in many areas (Annemüller et al., 1999; Buzzola et al., 2001; Zschöck et al., 2004) and even worldwide (Smith et al., 2005b), a small subset of strains is considered to be involved in the majority of mastitis cases. Most studies have demonstrated that a given herd usually harbours a limited number of S. aureus strains, often with one or a few strains predominating (Kapur et al., 1995; Fitzgerald et al., 1997; Larsen et al., 2000a; Buzzola et al., 2001; Sommerhäuser et al., 2003; Mørk et al., 2005; Rabello et al., 2007). The strains of this subset have been shown to be genetically related and have probably differentiated from a common parental strain (Smith et al., 2005b). By contrast, most strains detected at extramammary sites, such as the udder and teat skin, differ genetically from the mastitis strains (Zadoks et al., 2002), which suggests that strains have evolved that specialise in infecting bovine intramammary tissue. Identical S. aureus strains have occasionally been isolated from dairy cows and hands of milking personnel (Jørgensen et al., 2005a), but strains originating from bovine mastitis in general represent a genetically different cluster than the human strains, suggesting host-specificity (van Leeuwen et al., 2005). Part of this specificity or host-adaptation may be due to acquisition or loss of mobile accessory genetic elements, which are present in different strains to various degrees (Fitzgerald and Musser, 2001).

2.1.3 Clinical manifestation and outcome

Clinical manifestation of bovine mastitis caused by S. aureus following IMI can vary from subclinical to a peracute, gangrenous form (Barkema et al., 2006). Subclinical mastitis is the most common and likely the most problematic. The only detectable response is an increased milk SCC but the pathological changes; reduced secretory cell and luminal areas, lead to lowered quality and quantity of milk production. Unfortunately, milk SCC may remain under threshold limits and infection is often chronic when detected. S. aureus mastitis generally responds poorly to treatment (Barkema et al. 2006), relapses are common, and the infection is easily transmitted to unaffected quarters of the same cow. Virulence of the infecting strain and host factors, including the efficacy of the immune response of the cow, which may be related to age and lactation stage, are assumed to affect the severity of the clinical signs and persistence of infection. Older cows are more often infected with S. aureus compared with primiparous cows and develop more severe inflammation (Pyörälä and Pyörälä, 1997; McDougall et al., 2007). High SCC in the affected quarter and multiple infected quarters in the same cow impair the prognosis (Sol et al., 2000; Zadoks et al., 2001).

15

2.1.4 Antimicrobial resistance of bovine Staphylococcus aureus

S. aureus exhibits resistance to a wide range of antimicrobial agents including disinfectants (Bjørland et al., 2001). In the Nordic countries, antimicrobial resistance of mastitis-causing bovine S. aureus is generally less common than in many other countries. In Norway and Sweden, the proportion of penicillin resistant isolates has remained below 10% (SVARM, 2002; NORM-VET, 2005). In the rest of the Europe, the proportion of penicillin resistant isolates has ranged from 23% (DANMAP, 2003) up to 69% (Nunes et al., 2007), in the United States from 38 to 61% (Erskine et al., 2002) and was reported to be 40% in Argentina (Gentilini et al., 2000). Penicillin resistance is due to expression of inducible β-lactamase (penicillinase until 1960) encoded by the blaZ gene, which causes hydrolysis of the β-lactam ring of penicillin. The first reports on the ability of S. aureus to break down penicillin were published in 1940, a year before the antimicrobial was introduced for therapeutic use (Abraham and Chain, 1940). Impaired treatment response has been associated with penicillin resistance of the infectious S. aureus strain (Sol et al., 2000; Taponen et al., 2003). However, the connection is not straightforward, which may indicate that some other bacterial factors could be involved in the phenomenon (Barkema et al., 2006). Introduction of antimicrobials such as tetracycline, aminoglycosides and macrolides into mastitis therapy has been accompanied by the emergence of corresponding resistance in bovine S. aureus strains (Myllys et al., 1998). Acquisition of modified penicillin-binding proteins (PBP) has made some S. aureus strains resistant to all β-lactams. This property, called methicillin resistance, has been uncommon among bovine S. aureus isolates to date (Monecke et al., 2007; Moon et al., 2007). The recent emergence of vancomycin resistant S. aureus strains in human medicine has made control of S. aureus infections increasingly difficult (Jones, 2008).

2.1.5 Control

The “five point plan” plan developed by mastitis experts decades ago (Neave et al., 1969) for control of contagious mastitis has been applied worldwide. The main goal of the programme was to eradicate Streptococcus agalactiae and S. aureus from cattle herds. The elements were post-milking teat disinfection, total dry cow therapy, appropriate therapy of clinical cases during lactation, proper maintenance of the milking machinery and culling of chronically infected cows. The five point plan, or some of its components, have considerably reduced Str. agalactiae mastitis, but for S. aureus mastitis the effect has been less satisfactory. Segregation of the infected cows alone has not been sufficient (Fox et al., 1991), cure rates for dry-cow therapy have been low, ranging from 40 to 70% (Leslie and Dingwell, 2003) and there is no scientific evidence to suggest that that culling alone is economic. Epidemiological studies of S. aureus in the environment of dairy cows have increased knowledge on the dynamics of S. aureus IMIs. Current strategies for control and prevention of S. aureus mastitis have been expanded to include isolation or elimination of the reservoir by segregation, therapy, and/or culling, isolation or removal of the fomites by applying improved milking hygiene, evaluation of teat skin condition, teat disinfection and backflush. In some countries, host resistance has been enhanced by improving management of the cows and vaccinating against mastitis (Talbot and Lacasse, 2005). In spite of the introduction of large-scale mastitis control programmes, S. aureus remains a major mastitis pathogen. It causes mastitis epidemics even in well-managed dairy herds (Smith et al., 1998) and can persist for long periods in the mammary glands (Raimundo et al.,

16

1999; Larsen et al., 2000a; Anderson and Lyman, 2006). The current control practices may fail to prevent the spread of particularly virulent strains. Kapur et al. (1995) suggested that control of S. aureus mastitis should be focused on specific strains causing infections in the herds.

2.2 Pathogenesis of bovine Staphylococcus aureus intramammary infection

An infection like IMI can be considered a condition where adverse colonisation by a microbe has occurred in the host animal. Inflammation, the host’s response to the microbe, is seen as swelling, redness, pain, heat and interference in the normal function of the affected organ. The general term for inflammation of the mammary gland is mastitis, which can occur with or without infection. Development of S. aureus infection in the bovine mammary gland can be divided into the following stages: entry and attachment of S. aureus bacteria in the mammary gland, interaction between the bovine immune system, evasion of immune defence, survival and tissue invasion. Entry of S. aureus into the teat canal can lead to IMI, but this depends on certain conditions as for any infection: the initial number of bacteria, access of the microbe to the target tissue, virulence of the strain and immunity of the host (Projan and Novick, 1997).

2.2.1 Entry and attachment of Staphylococcus aureus into the mammary gland

Entry of a sufficient number of S. aureus bacteria via the teat canal into the mammary gland is required for the development of natural IMI. The infective dose of S. aureus in bovine IMI is not known precisely, but quantities as small as 10 cfu have caused infection in an experimental model where bacteria were infused directly into the teat duct (Reiter et al., 1970; Postle et al., 1978). However, infection did not develop in all cows in experimental infection models (Schukken et al., 1999), which suggests that S. aureus does not always multiply sufficiently in the mammary gland, but other predisposing factors need to be present. First line defence mechanisms against invasion of S. aureus are the physical barriers of the teat canal. A tightly constricted teat canal rapidly closes the end of the teat apex between milkings. Keratin material covering intact teat canal epithelium inhibits bacterial growth, although heifers have sometimes harboured S. aureus in this tissue (Trinidad et al., 1990). Attachment of S. aureus to host cells or extracellular matrix molecules is a critical step for colonisation and intramammary infection. Adherence to these substances is thought to occur by non-specific physicochemical mechanisms and by specific bacterial host-cell binding (Kluytmans et al., 1997). The bacteria must then resist flushing of milk until adherence to the epithelial cells lining the ductules and alveoli of the mammary gland is accomplished (Frost et al., 1977). S. aureus adheres to bovine mammary epithelial cells better than most other bacterial species (Frost et al., 1977) and the presence of milk enhances adherence (Mamo and Fröman, 1994). Attachment is best to fully mature cells, such as keratinised epithelial cells, compared with the cells of the deeper epidermis (Aly and Levit, 1987). Adherence to teat canal cells depends on the origin of the cell type (Sutra and Poutrel, 1994). S. aureus adheres particularly well to the cells of the upper part of the mammary gland (Frost et al., 1977). S. aureus is also able to adhere to fat globules, which allows dissemination to the upper part of the gland by floating (Sandholm et al., 1989).

17

Teat traumas increase the risk of colonisation of the teats by S. aureus. Epithelial damage reveals the underlying subepithelial components, e.g. fibrinogen and collagen, which allows staphylococcal surface proteins to mediate adherence to the host cell matrix (Patti et al., 1994). Exotoxins secreted by S. aureus can also be involved in the epithelial damage (Mamo et al., 1988). Even minor traumas, such as those from milking, can facilitate entry of S. aureus into the teat. Callusing of the teat canal prevents tight closing and extremely callused teats were shown to be more susceptible to S. aureus mastitis (Zadoks et al., 2001). A study by Myllys et al. (1994b) demonstrated that teat canal colonisation, teat canal infection and IMI were far more common in quarters where the teat orifice epithelia were experimentally abraded compared with control quarters. Animal experiments have shown that very large numbers of viable staphylococci are needed to cause cutaneous or subcutaneous infection, unless a foreign body is present (Noble, 1965; Bunce et al., 1992). On the other hand, minor traumas such as depilation or removal of the outer layers of epithelium due to removal of adhesive surgical tapes, have led to skin infection in humans if S. aureus is applied and the area occluded (Singh et al., 1971).

2.2.2 Interaction between Staphylococcus aureus and the bovine immune system

The second line of defence comprises the innate immune system that includes defence cells (neutrophils, macrophages, natural killer cells, and dendritic cells) and humoral factors (lactoferrin, lysozyme, lactoperoxidase, protein A, and complement) present in the teat duct and in the milk (Rainard, 2003; Rainard and Riollet, 2003). Bacterial colonisation leads to influx of somatic cells, primarily polymophonuclear neutrophils (PMN), as response to cytokine production triggered by the bacteria. Phagocytosis and killing by leukocytes play the main roles in host defence against S. aureus infection. Numerous exoproteins, like SAgs secreted by S. aureus, can interfere with phagocytosis (Verhoef, 1997). Phagocytosing cells express ingested staphylococci or their products in association with MHC II molecules on their surface in order to present them to T cells. Accumulation of neutrophils and capsule formation often surrounds gland areas infected by S. aureus and the bacteria remain sheltered within furuncles or abscesses. S. aureus may avoid intracellular death and survive within the host cells (Mullarky et al., 2001), where it is protected from the host defence mechanisms and effects of antimicrobials. S. aureus is not regarded as an intracellular microorganims but replication within bovine mammary epithelial cells has been demonstrated (Almeida et al., 1996). In the intracellular environment, S. aureus can fall into nutritional stress but still survive. These so-called small-colony variants grow slowly and have typical colony morphology, due to altered metabolism, but they are infective and are highly resistant to antimicrobials (Proctor et al., 1998; Brouillette et al., 2004). Secreted cytolytic toxins allow S. aureus to be released from the intracellular environment.

2.3 The role of bacterial virulence factors in Staphylococcus aureus intramammary infection

According to the narrowest definition, a virulence factor is a substance that when purified to homogeneity and introduced into a test animal, produces a pathogenic effect. By using this definition factors involved in attachment cannot be considred to be virulence factors. In the broadest sense, any factor produced out of the bacterial cytoplasm allowing survival within or on a host organism in a non-symbiotic manner can be regarded as a virulence factor (Projan

18

and Novick, 1997). S. aureus can produce numerous putative virulence factors that allow the microbe to adhere to eukaryotic membranes, resist phagocytosis, lyse eukaryotic cells and trigger the production of a cascade of host immunomodulating molecules. Among bovine S. aureus isolated from mastitis, production of enterotoxins (Aarestrup et al., 1999; Larsen et al., 2000b; Tollersrud et al., 2000b) or genes encoding these toxins have been the most studied (Salasia et al., 2004; Zschöck et al., 2004; Fueyo et al., 2005; Srinivasan et al., 2006). Many suggest that virulence of bovine S. aureus vary among strains (Smith et al., 1998; Raimundo et al., 1999; Larsen et al., 2000a; Zadoks et al., 2000; Sommerhäuser et al., 2003), but the possible role of specific virulence factors in this phenomenon is poorly understood. In general, S. aureus virulence factors can be divided into those involved in attachment and into those enabling evasion of the host immune system and tissue invasion (maintenance of infection).

2.3.1 Control and expression of virulence factors

Virulence factors of S. aureus are not essential for bacterial cell division and growth, and many of these are needed only in certain circumstances. Continuous secretion of the virulence factors would be uneconomic for the microorganism. Therefore, expression of S. aureus surface proteins and exoenzymes is a coordinated process, which depends on the phase of bacterial growth and environmental stimuli (Horsburgh, 2008). The coordination is made by the aid of well-characterised global regulatory elements, those of sar locus (staphylococcal accessory regulator) and two-component regulatory systems like agr locus (accessory gene regulator), and sae (staphylococcal accessory element) (Fournier, 2008). By culturing S. aureus in idealised medium it has been shown that surface proteins involved in attachment are produced when the bacterial cell growth and division is exponential (Figure 1). sar upregulates production of many of the surface proteins during this phase. sar is able to up-modulate the agr locus. The agr is a quorum-sensing system, which turns off the expression of surface proteins and upregulates transcription of several virulence factors in response to increased cell density during the postexponential phase of growth. Other less well characterised regulators and proteins such as sigma factor σ and rot (Fournier, 2008) are also involved in the regulation of virulence factor production.

19

Figure 1. The expression and control of S. aureus of putative virulence factors during the bacterial growth cycle in idealized medium.↑ = upreglation, ↓ = downregulation. PXP = postexponential, STA = stationary. Adapted from Projan and Novick (1997) and from Fournier (2008).

2.3.2 Virulence factors involved in establishing an infection

S. aureus can produce several adhesins or MSCRAMMs, which have been shown to mediate attachment to various cell surface proteins such as collagens, elastin, fibrinogen, fibronectin, bone sialoprotein, laminin and thrombospondin (Foster and Höök, 1998). Ligations to these proteins enable S. aureus to colonise tissues and initiate infection (Patti et al., 1994, Brouillette et al., 2003). FnBPs have been demonstrated to be involved in cell invasion (Dziewanowska et al., 1999) and to mediate adhesion to platelets via fibronectin and fibrinogen (Heilmann et al., 2004). S. aureus is capable of invading eukaryotic cells and phagocytosing cells, and antibiotics are unable to reach the intracellular bacteria (Balwit et al., 1994). In this invasion S. aureus uses its surface-expressed FnBPs (Lammers et al., 1999) and α1β5-integrin on the host cell surface (Sinha et al., 1999). S. aureus can adhere to the polymer surface of plastic material, to form microcolonies, and to produce extracellular slime or glycocalyx, which forms a biofilm covering the bacteria (Donlan and Costerton, 2002). FnBPs are evidently required for adhesion to fibronectin-coated surfaces, e.g. those in medical devices, as double mutants have been shown to be unable to adhere (Greene et al., 1995). Furthermore, FnBPs have recently been demonstrated to mediate biofilm formation (O’Neill et al., 2008).

20

2.3.3 Virulence factors involved in maintenance of infection

The factors involved in maintenance of infection include haemolysins, leukocidins, SAgs, nucleases, proteases and lipases. Haemolysins are cytotoxic agents that damage and/or destroy cells of the defence system. Alpha-haemolysin (α-toxin), frequently produced by bovine S. aureus IMI isolates (Aarestrup et al., 1999; Akineden et al., 2001), is toxic to bovine mammary cells (Bramley et al., 1989), causes erythrolysis, disturbs the host cell ion balance, is dermonecrotic and neurotoxic (Dinges et al., 2000), is needed in biofilm formation (Caiazza et al., 2003) and plays a putative role in peracute, toxic bovine S. aureus mastitis (Jones and Wieneke, 1986; Matsunaga et al., 1993). Beta-haemolysin (β-toxin), probably produced by most of the bovine strains (Aarestrup et al., 1999), functions as sphingomyelinase of the erythrocytes. Susceptibility of erythrocytes to β-haemolysin varies depending of the sphingomyelin content of the target cell, e.g. sheep erythrocytes are easily lysed whereas rabbit erythrocytes are not (Doery et al, 1965). Based on in vitro studies, β-haemolysin has been thought to increase the damaging effects of α-haemolysin and enhance the attachment of S. aureus on bovine mammary epithelial cells (Cifrian et al., 1995; Cifrian et al., 1996). Delta-haemolysin has surfactant or channel-forming properties and it intensifies haemolytic effects of β-haemolysin (Dinges et al., 2000). Haemolysin γ is toxic to monocytes and macrophages and lyses red blood cells and phagocytes (Projan and Novick, 1997). Concomitant production of α and γ-toxin promoted S. aureus virulence in murine septic arthritis (Nilsson et al., 1999). The family of leukocidins comprises toxins that damage membranes of phagocytosing host defence cells by inducing Ca2+ influx and subsequent pore formation. Leukocidins are composed of two subunits of related proteins, class S (slow eluting) and F (fast-eluted) proteins, the toxic effect depending on the synergistic action of both of these proteins (Kamio et al., 1993). The family includes Panton-Valentine leukocidin PVL (LukS-PV + LukF-PV), γ-haemolysin and leukocidin (Hlg and Luk; HlgA + LukF and LukS + LukF), LukM/FPV (LukM + LukF-PV), LukE/D (Kamio et al., 1993; König et al. 1995; Prevost et al., 1995), and LukR-PV, detected in strain P83 originating from bovine mastitis (Supersac et al., 1993). Leukocidins are more leukotoxic that γ-haemolysin (Prevost et al., 1995). The family of staphylococcal SAgs includes enterotoxins and enterotoxin-like proteins (SEA-SEE, SEG-SER, and SEU), toxic shock syndrome toxin (TSST), and exfoliatins ETA and ETB. SAgs have at least three biological properties: pyrogenicity, superantigenicity (Dinges et al., 2000) and the ability to enhance the lethality of rabbits to endotoxin shock up to 100 000 fold in experimental models, where TSST-1 and endotoxin have been administered to test animals (Schlievert, 1982). SAgs bind to the outer surface of the MHC class II proteins and outside the site in T-cell receptors, to which antigens normally bind. The normal antigen presentation mechanism is bypassed and consequently as much as 30% of the T-cells are activated instead of the normally of 0.1‰ activated T-cells (Fleischer and Schrezenmeier, 1988). This results in massive and uncontrolled release of cytokines, and in capillary leak, hypovolemic shock and multiorgan failure as seen in the toxic shock syndrome (Dinges et al., 2000; McCormick et al., 2001; Alouf and Müller-Alouf, 2003). Peracute bovine S. aureus mastitis has been suggested to be such a reaction. The condition is rare and from the viewpoint of the bacterium it is unfavourable because the host will normally die. SAgs can stimulate macrophages and monocytes to produce several inflammatory mediators, including tumour necrosis factor-α, nitric oxide, interleukin-6 and IL-10. Furthermore, in the presence of interferon-γ, the macrophages become cytolytic (Fleming et al., 1991). SAgs are

21

considered to facilitate S. aureus to evade host defence systems and are thus suggested to contribute both to acute and chronic bovine mastitis. Ferens et al. (1998) reported that SEC affected the bovine immune system by increasing the proliferation of CD8+ T-cells, resulting in suppression of CD4+ T-cells. SAgs may help S. aureus persist in mucosal membranes (Ferens and Bohach, 2000). Most S. aureus strains produce a capsule and slime that can enhance adhesion to endothelial cells, inhibit phagocytosis, and hide the bacteria from antimicrobials and antibodies when subsequent opsonisation and phagocytosis fail (Lee et al., 1994; Verhoef, 1997). Protein A, a cell wall-associated protein, interacts with several host factors including immunoglobulins G, A, E, tumour necrosis factors, and platelets attenuating opsonisation and phagocytosis (Peterson et al., 1977; Gomez et al., 2004). Binding of IgM associated with B cells by protein A induces apoptosis, a mechanism leading to death of host cells such as PMN and macrophages (Menzies and Kourteva, 1998; Goodyear and Silverman, 2004). Coagulase, a prothrombin activator that converts fibrinogen to fibrin and causes clotting of cells, is produced by the majority of S. aureus strains. It possibly has a role as a virulence factor (Phonimdaeng et al., 1990; Lowy et al., 2000). Lipases, esterases, fatty acid modifying enzyme (FAME) and phospholipases function as surfactants against a variety of fatty acids and other lipid molecules that host cells produce in order to destroy the bacterial membrane (Chamberlain and Imanoel, 1996; Arvidson, 2000).

2.3.4 Immunity to Staphylococcus aureus

Infection by S. aureus can induce a strong antibody response to many bacterial antigens (Bonventre et al., 1984), but a long-term protective immunological memory does not seem to develop (Talbot and Lacasse, 2005). During S. aureus IMI, peripheral anti-staphylococcal antibody titers are correlated with milk antibodies, the levels being lower in milk. Bovine IgG1, IgG2 and IgM opsonise S. aureus, and IgA functions as a neutralising antibody. The levels of IgG1 are highest in mammary glands (Schalm et al., 1971). S. aureus possesses numerous mechanisms that interfere with the innate immune response of the host, including inhibition of complement activation and neutralisation of antimicrobial peptides (Chavakis et al., 2007). The effect of genetic polymorphism of innate immune factors between human individuals on nasal carriage by S. aureus was recently studied by van Belkum et al. (2007).

2.4 Typing of bovine Staphylococcus aureus

Typing of S. aureus is important for epidemiological investigations, to determine the sources of infection, routes of transmission in disease outbreaks and presence of strains of different virulence. Staphylococcal typing can be categorised as phenotypic or genotypic. During recent decades, several genotypic methods have superseded many of the traditional phenotypic methods. Genotyping is often combined with suitable phenotypic data, such as profiles for antimicrobial resistance. This is recommended, for example, in the epidemiological evaluation of clonal relatedness of isolates (Tenover et al., 1994).

2.4.1 Phenotypic methods

Antibiogram typing, i.e. determining susceptibility of the isolates to a selection of antimicrobials, has been a widely used phenotypic method in the characterisation of S. aureus originating from bovine mastitis (Myllys et al., 1998; Østerås et al, 2006). This method can be

22

useful for initial screening of multiresistant strains or strains resistant to a given, important antibiotic such as penicillin. For epidemiological purposes this method has relatively poor discriminatory power. Furthermore, many of the tested susceptibility characteristics can be very unstable due to presence of extrachromosomal plasmids for instance. In biotyping, isolates of S. aureus are differentiated by means of cultural and biochemical properties. The discriminatory power of biotyping is relatively low, and non-typable isolates are frequent. Differentiation of S. aureus capsular types using serotyping has been applied to bovine mastitis isolates (Guidry et al., 1998; Sordelli et al., 2000; Tollersrud et al., 2000a; Tollersrud et al., 2000b), aiming to develop vaccines against staphylococcal capsular polysaccharides (Sordelli et al., 2000). Non-typable isolates are common (Tollersrud et al., 2000a; Tollersrud et al., 2000b). In phage-typing, the sensitivity of staphylococcal isolates to a standard collection of bacteriophages is tested. Thorne and Wallmark (1960) were probably the first to use phage-typing in epidemiological studies on S. aureus mastitis. Fox et al. (1991) used phage typing to study the importance of milkers' hands and milking unit liners as fomites, and teat skin as a reservoir of S. aureus IMI. Phage-typing is no longer used due to lack of discrimination, large numbers of non-typable strains, changing phage patterns of a given strain over time and complex technical requirements. Multi-locus enzyme electrophoresis (MLEE) compares electrophoretic mobility variation in a set of housekeeping enzymes (Selander et al., 1986). Kapur et al. (1995) and Tollersrud et al. (2000b) applied MLEE to studying population genetics and epidemiology of mastitis-causing bovine S. aureus.

2.4.2 DNA-based methods

Ribotyping compares the numbers and sizes of DNA fragments separated by conventional electrophoresis and obtained by cleaving bacterial DNA with a frequent cutter enzyme. The method is regarded as unsuitable for routine laboratory tests as it is laborious and technically demanding (Deplano and Struelens, 1998). Strain differentiation by binary typing is made by hybridisation with RAPD-generated DNA probes to target sequences in the S. aureus genome. Absence or presence of a given sequence is binary coded, resulting in a numerical type (van Leeuwen et al., 1999). The discriminatory power of binary typing is higher than even that of PFGE (van Leeuwen et al., 1999), but the method does not provide information on genetic relatedness of the strains. MLST is based on sequence variation of seven chromosomally encoded housekeeping genes. The method provides sequence data that can be used for prediction of evolutionary development. MLST has been used in studies on host specificity (Herron-Olson et al., 2007) and global epidemiology of bovine S. aureus clones (Smith et al., 2005b; Rabello et al., 2007). In PFGE the bacterial genome is cleaved with a rare-cutting enzyme (mostly Sma1) to 10-30 large fragments ranging in size from 10 to 800 kb. Sufficient fragment separation is obtained by electrophoresis, where the current changes polarity in pulses. PFGE, currently regarded as the gold standard method, is highly reproducible and discriminatory, and coordinated inter-laboratory surveillance has allowed development of standardised protocols (van Belkum et

23

al., 1998; Murchan et al., 2003). PFGE has been widely applied in typing bovine S. aureus (Zadoks et al., 2000; Buzzola et al., 2001; Joo et al., 2001; Zadoks et al., 2002; Middleton et al., 2002a; Middleton et al., 2002b; Jørgensen et al., 2005a; Hata et al., 2006; Aires-de-Sousa et al., 2007; Peles et al., 2007; Rabello et al., 2007). The method has been specifically applied to study the effect of S. aureus genotype on manifestation of mastitis (Zadoks et al., 2000; Middleton et al., 2002a), to determine the origin of mastitis-causing strains (Zadoks et al., 2002), to evaluate the effect of genotype on long-term persistence of S. aureus in dairy herds (Anderson and Lyman, 2006) to assess recovery from mastitis caused by different S. aureus genotypes (Dingwell et al., 2006), and to study genetic relationship of strains expressing or carrying specific virulence determinants causing particular disease like food poisoning (Jørgensen et al., 2005b). PFGE is the method of choice for comparison of epidemiologically related isolates.

24

3. AIMS OF THE STUDY

1. To determine the prevalence of intramammary infections caused by S. aureus in Finnish dairy herds, based on data collected during a national mastitis survey, and to study the occurrence of S. aureus in intramammary infections and extramammary sites in two Finnish dairy herds managed under different regimes. 2. To assess the diversity of S. aureus isolates by genotyping and by identifying particular genes in S. aureus isolated from intramammary infections. 3. To investigate antimicrobial resistance of S. aureus isolated from bovine mastitis using phenotypic and genotypic methods, with special reference to penicillin and methicillin resistance. 4. To study the relationship between S. aureus genotypes, presence of specific genes, and severity and persistence of intramammary infections. 5. To study the potential reservoirs and transmission of S. aureus causing bovine mastitis in two dairy herds maintained under different management regimes.

25

4. MATERIALS

4.1 Origin of Staphylococcus aureus isolates

The first part of the thesis (I) represents results of the third nationwide survey conducted in Finland in 2001 to estimate prevalence and distribution of bovine mastitis, and in vitro antimicrobial susceptibility of the isolated mastitis pathogens. Dairy advisors collected herd data and milk samples from 3282 dairy cows on the 216 participating farms. This represented a random sample of 270 farms from a national database that included all dairy farms in Finland. Among the 12 661 milk samples collected, 430 harboured S. aureus. In studies II-III, the S. aureus isolates (n = 217), including clinical data, were collected from field cases of S. aureus mastitis in 116 cows with 141 infected quarters. The samples were collected between September 1993 and January 1997 in the practice area of the Ambulatory Clinic, Faculty of Veterinary Medicine, University of Helsinki. The 217 S. aureus isolates comprised 134 isolates collected at the onset of mastitis and 83 collected during the follow-up period. Seven of the 83 isolates were from new quarters infected during the follow-up period. In study IV, S. aureus isolates originated from two Finnish dairy herds in southern Finland. The isolates were collected from all IMIs diagnosed during a 10-month period from October 2002 to August 2003, and from extramammary sites (dairy cow teat skin, teat canals and skin lesions; milking liners; hands and nostrils of milking personnel) sampled during one farm visit to both herds. Prior to the study, S. aureus mastitis in the herds was estimated to affect 10-20% of the cows.

4.2 Study animals and herds

The cattle breeds in study I were Finnish Ayrshire (42.1%), Holstein-Friesian (14.0%), other breeds (1.0%), and mixed breeds (42.9%). The cows were from herds with a mean size of 17.3 cows, average parity of 2.5, mean annual milk production of 7505 kg and average geometric mean for SCC of 116 000 cells/ml of milk. The majority of the farms had stanchion-barns (89.8%) and the rest had loose-house systems (6.9% warm and 3.2% cold systems). Most cows in studies II-III were of the Finnish Ayrshire breed, some were Holstein Friesians or of mixed breed and very few cows were of the Finnish Landrace breed. The cows were from 70 dairy herds and represented approximately 60% of the individual cows treated for S. aureus mastitis by the Ambulatory Clinic during this period. The herds were included in a health recording system, where all veterinary treatments were recorded and BMSCC was regularly monitored. The mean milk production of these herds was not recorded. All herds were housed in stanchion barns. In study IV, Herd I comprised an average of 30 lactating Holstein-Friesian cows with mean annual milk production of 9600 kg. No replacement heifers or cows were integrated from other herds (closed herd). The cows were kept in a loose-house system during the indoor season and were milked twice a day in a milking parlor. Herd II comprised an average of 54 Ayrshire cows with mean annual milk production of 9185 kg. Young replacement heifers from all over in Finland were purchased (open herd), and the cows were housed and milked in a stanchion-barn.

26

5. METHODS

5.1 Clinical examination of the cows and mastitis follow-up (II-IV)

Clinical signs were recorded and the outcome of mastitis was followed by the veterinarians of the Ambulatory Clinic in studies II-III. The herdsmen controlled milk appearance by stripping before every milking and estimated quarter milk SCC using a CMT test. If the cow showed clinical signs of mastitis or elevated milk SCC as estimated by the CMT test (Klastrup and Schmidt Madsen, 1974), the farmers were advised to call the attending veterinarian of the Ambulatory Clinic. The veterinarian examined the cow, recorded clinical signs, scored milk SCC using CMT from each lactating quarter, collected the milk samples and started the treatment based on herd-level history of mastitis-causing bacteria and their antimicrobial susceptibility. Within approximately 20 hours of the veterinarian’s visit, preliminary results based on routine identification of bacteria and testing of β-lactamase production of staphylococci were available at the laboratory of the Ambulatory Clinic. The treatment was continued according to the laboratory results. Cows infected with β-lactamase negative isolates were treated with penicillin G i.m. for 5 days. If the isolate was β-lactamase positive, cows with clinical mastitis were treated with amoxycillin clavulanate i.m. and i.m.m. and those with subclinical mastitis with cloxacillin i.m.m. for 5 days. The veterinarian revisited the cows approximately 2 and 4 weeks post-treatment for follow-up sampling and clinical examination. In study IV herds (Herds I and II) the herdsmen monitored the cows as in studies II-III, but they contacted the herd veterinarian. In Herd II, cows with subclinical S. aureus IMI were usually not treated, but the affected quarter was prematurely dried-off, especially if the isolate was penicillin-resistant (produced β-lactamase).

5.2 Sample collection

5.2.1 Milk samples

Quarter foremilk samples were collected aseptically for bacteriological culturing as described by Honkanen-Buzalski (1995). Before sampling, the first streams of milk were discarded and teat ends were disinfected with cotton swabs soaked in 70% alcohol and allowed to dry. The samples were kept on ice until in the laboratory. The milk samples for SCC analysis by fluoro-optic counting were preserved with sodium azide (I). In study I, dairy advisors collected quarter milk samples from all milk-producing cows at the farms and submitted the samples. In study IV, both herd staff and veterinarians collected the milk samples and submitted them to the laboratories of the Finnish Food Safety Authority and Valio Ltd.

5.2.2 Samples from extramammary sites (IV)

Samples from the teat walls, from the teat ends around the teat orifice, teat canals and all clearly visible skin lesions (mainly abrasions on the hind legs and a few lesions on teat and udder skin), and hands and nostrils of the milking personnel were collected using sterile, single-packaged, cotton-tipped swabs (Technical Service Consultants Ltd., Heywood, UK). Swabs were moistened with trypticase soy broth (TSB; Oxoid Ltd., Basingstoke, UK) before

27

swabbing. Teat canal samples were taken by rotating ultra-fine, sterile, cotton-tipped swabs (Deltalab S. A., Barcelona, Spain) 360˚ in the canal. The teat was cleaned, disinfected with chlorhexidin solution and dried with a paper towel before inserting the swab into the canal. Samples were taken from the hands and nostrils of the staff by rotating the swab 360˚ between the bases of the fingers of both hands or in both nostrils. Four random swabs from milking liners in Herd I, and swabs from all milking liners in Herd II, were taken prior to milking to control the cleanliness of the liners. Milking liners were also sampled after milking each cow. The swabs were rotated 360˚ at the point where the liner joined the short milking tube. Disposable latex gloves were used during the sampling. After use all swabs were immediately placed in sterile plastic containers filled with 5 ml of TSB. The containers were cooled and transported to the laboratory within a few hours.

5.3 Determination of indicators of inflammation in the milk

5.3.1 Milk somatic cell count

Quarter-milk SCC was determined using the fluoro-optic method (Fossomatic Milk Analysis, Foss Electric, Hillerød, Denmark) in study I. In studies II-IV, milk SCC was estimated from quarter milk samples immediately after collection using the CMT test and scoring system described by Klastrup and Schmidt Madsen (1974). A cow and a quarter were considered to have S. aureus IMI when ≥ 500 cfu/ml of S. aureus in the quarter milk was detected (I-III). In study IV, the treshold for IMI was 500 cfu/ml for quarter milk samples taken by the herdsmen. From the milk samples taken by the veterinarians at the farm visit, also quarters with ≥100 cfu/ml of S. aureus in the milk were considered S. aureus IMI, if the CMT score was ≥3. Milk samples with bacterial growth of more than two species were considered contaminated (I-IV). Mastitis is a general term used to describe inflammatory conditions of the mammary gland. A cow and a quarter were defined as mastitic if the quarter milk SCC exceeded 300 000/ ml (I) or if the estimated CMT score was ≥3 using the Scandinavian scoring system (Klastrup and Schmidt Madsen, 1974) (II-IV). All quarters with S. aureus in the milk had CMT scores ≥3 at the time of the visit by the veterinarian (II-IV).

5.3.2 Milk N-acetyl-β-D-glucosaminidase (II)

NAGase activity was determined from quarter milk samples after one cycle of freezing and thawing to indicate the inflammatory response of the mammary gland. A fluorogenic NAGase assay system described by Mattila and Sandholm (1986) was used with a Fluoroskan® microtitration tray reading fluorometer (Labsystems, Helsinki, Finland). The results were expressed as arbitrary units (U).

5.4 Bacterial isolation and conventional identification

5.4.1 In milk samples

For bacterial isolation from milk samples, 0.01 ml of milk was cultured on blood-esculin agar and incubated for 48 h at 37°C. The plates were examined after 24 and 48 h of incubation and

28

bacterial species were preliminary identified as described by the National Mastitis Council (NMC, 1999). The sample was considered contaminated if more than two bacterial species were isolated (IDF, 1981; NMC, 1999). S. aureus isolates were stored frozen at -70°C in brain heart infusion broth containing 15% glycerol (I) or in commercial storage medium (II-IV; Protect Bacterial Preservers, Technical Services Consultants Limited, Heywood, UK).

5.4.2 In other samples (IV)

Isolation of S. aureus in samples from extramammary sites was based on descriptions by Lancette and Tatini (1992) with the following modifications. After keeping the samples for 30 min at room temperature, 5 ml of TSB broth containing 20% NaCl was added and the samples were vortexed and incubated for 24 h at 37˚C. Subsequently, 0.1 ml aliquots of culture were transferred to duplicate plates of Baird-Parker agar and spread with a plastic triangle. The plates were incubated at 37˚C for 24 + 24 h. Black, dark grey or shiny brown colonies with clear zones of haemolysis were selected for Gram-, catalase-, and rabbit plasma coagulase tests. Other colonies suspected of being S. aureus, but without a clear zone, were transferred on to acriflavin agar and incubated at 37˚C for 24 h. Colonies with yellow growth were tested as potential S. aureus, as previously described. The preliminary identification of S. aureus included the following tests: colony morphology, colour and haemolysis on sheep blood agar, positive Gram stain, and production of catalase, coagulase, and clumping factor using Slidex Staph Kit (bioMérieux, Marcy l’Etoile, France). The isolates were stored at –70 ˚C in Protect Bacterial Preservers (Technical Services Consultants Limited, Heywood, UK).

5.5 Antimicrobial susceptibility testing

5.5.1 Testing for β-lactamase production

Production of β-lactamase was tested for using liquid nitrocefin (II-IV), as recommended by the manufacturer (Becton Dickinson Microbiology Systems, Cockeysville, MD, USA), or using nitrocefin discs (I) (AB Biodisk, Solna, Sweden). S. aureus ATCC 29213 was used as a positive control.

5.5.2 Determination of MIC using the microdilution method (I-II)

Susceptibility of the isolates to antimicrobials was determined using the VetMIC microdilution system (SVA, Uppsala, Sweden). Minimum inhibitory concentration (MIC) was recorded as the lowest concentration that inhibited bacterial growth. Calculation of the MIC distributions and classification of the results by using breakpoints is described in detail in study I. S. aureus isolates with MIC of oxacillin = 2 µg/ ml were retested using an oxacillin-salt agar screening test as described by CLSI (2002), including methicillin-resistant and methicillin-susceptible S. aureus strains ATCC 43300 and ATCC 29213 as controls. The results were classified using the CLSI breakpoints for bacteria isolated from animals (2002), when available.

29

5.5.3 Detection of genes encoding β-lactamase and methicillin resistance (III-IV)

The β-lactamase gene blaZ was detected by primers designed by Vesterholm-Nielsen et al. (1999) in PCR conditions described by Haveri et al. (2005). The primers and conditions described by Mehrotra et al. (2000) were used for detection of the mecA gene.

5.6. Characterisation of Staphylococcus aureus isolates using DNA-based methods (II-IV)

5.6.1 DNA isolation (II-IV)

For PFGE typing, DNA was prepared in agarose blocks and digested with SmaI (Boehringer Mannheim, Germany) according to Salmenlinna et al. (2000). For PCR studies, DNA was isolated using the DNeasy Tissue Kit (QIAGEN, Hilden, Germany), with the modification that 11 U of lysostaphin (Sigma) was added at the cell lysis step. DNA quantity for each isolation was measured using a GeneQuant II RNA/DNA Calculator (Pharmacia), and diluted to a concentration of 100 ng/μl with TE buffer.

5.6.2 Molecular typing using PFGE (II-IV)

The DNA restriction fragments were separated using PFGE procedures of Murchan et al. (2003). Two λ concatamer markers (NewEngland BioLabs, USA) were included in each gel and S. aureus NCTC 8325 was used as a migration control for normalisation in every IV-VI lane. PFGE fingerprints were stored and processed with Bionumerics software (Applied Maths, Kortrijk, Belgium, version 4.00) with Dice coefficients, using clustering by the unweighted pair group method with arithmetic mean. Position tolerance was set at 1.0 and cluster cut-off at the 80% similarity level. The relatedness of the fingerprints in studies II-IV was assessed both by using computer analysis (BioNumerics, Applied Maths, Kortrijk, Belgium, version 4.00) and visual comparison. In studies II-III, where the isolates were collected during several years from 70 herds, the following criteria were used for determining the genetic relatedness of the different fingerprints in visual comparison: fingerprints with one to three band shifts were considered to be closely related and of the same pulsotype, and these were assigned the same capital letter; the closely related clones of the respective pulsotype were accordingly assigned a numeric suffix. Fingerprints with more than three band shifts were interpreted as being genetically

unrelated and of a different pulsotype. In study IV, where the isolates were obtained during a shorter period in two herds only, the criteria of Tenover et al. (1995) were applied: PFGE fingerprints with more than seven band differences were considered to be unrelated and assigned different capital letters; patterns with one to six band shifts were assigned the same capital letters and were accordingly represented by a numeric suffix.

5.6.3 Detection of genes encoding virulence and adhesion factors using polymerase chain reaction (III-IV)

PCR was used to identify the genes. The primers and PCR conditions used are listed in Table 1. In study III, multiplex PCR was conducted for the following genes sea, seb, sec, sed, see

30

(set-1), tst, mecA, eta, etb (set-2), seg, seh, sei, sej (set-3), hla, hlb (set-4), hld, hlg (set-5), sen, seo (set-6), and lukED, lukM (set-7). Individual PCR reactions were finally conducted for sel, sek, sem, and seu. In study IV, the multiplex PCRs were used for detection of sea, sec, sed (set-1), tst, mecA (set-2), seg, seh, sei (set-3), hla, hlb (set-4), hld, hlg (set-5), sen, seo (set-6), and for lukED, lukM (set-7). Individual reactions were conducted for sej, sek, sel, sem, seu, (III-IV), and for fnbA and fnbB (IV). Each 50-μl PCR mixture contained 2.5 U DyNAzymeTM DNA polymerase (Finnzymes, Espoo, Finland), 200 μM of dNTP mix, 20 pmol of each primer, and 100 ng of DNA template. The reactions were carried out in a Peltier 200 Thermal Cycler (MJ Research, Waltham, MA, USA), and each isolate was tested at least twice for each gene. The amplification products were analysed using electrophoresis through 1% agarose gels. A positive control (Table 1) and a negative control (reaction mixture without DNA template) were included in each PCR run.

31

Table 1. Primer sequences, PCR conditions and control strains used in the thesis (studies III-IV) to detect gene determinants by PCR. Gene Primer sequence (5´-3´) Reference Conditions Control strain nuc GCGATTGATGGTGATACGGTI 1 1 NCTC 8325 AGCCAAGCCTTGACGAACTAAAGC 1 blaZ AAG AGA TTT GCC TAT GCT TC 2 10 ATCC 29213 GCT TGA CCA CTT TTA TCA GC 2 femA AAAAAAGCACATAACAAGCG 3 3 NCTC 8325 GATAAAGAAGAAACCAGCAG 3 mecA ACTGCTATCCACCCTCAAAC 3 3 NCTC 8325 CTGGTGAAGTTGTAATCTGG 3 sea GGTTATCAATGTGCGGGTGG 3 3 FRI 913 CGGCACTTTTTTCTCTTCGG 3 seb GTATGGTGGTGTAACTGAGC 3 3 NO-50328 CCAAATAGTGACGAGTTAGG 3 sec AGATGAAGTAGTTGATGTGTATGG 3 3 FRI 913 CACACTTTTAGAATCAACAG 3 sed CCAATAATAGGAGAAAATAAAAG 3 3 EELA-7 ATTGGTATTTTTTTTCGTTC 3 see AGGTTTTTTCACAGGTCATCC 3 3 FRI 913 CTTTTTTTTCTTCGGTCAATC 3 eta GCAGGTGTTGATTTAGCATT 3 3 DK 959 AGATGTCCCTATTTTTGCTG 3 etb ACAAGCAAAAGAATACAGCG 3 3 DK 880 GTTTTTGGCTGCTTCTCTTG 3 tst ACCCCTGTTCCCTTATCATC 3 3 FRI 913 TTTTCAGTATTTGTAACGCC 3 seg AATTATGTGAATGCTCAACCCGATC 4 41, 32 ATCC 49775 AAACTTATATGGAACAAAAGCTACTAGTTC 4 seh CAATCACATCATATGCGAAACCAG 4 41, 32 ATCC 51811 CATCTACCCAAACATTAGCACC 4 sei CTCAAGGTGATATTGGTGTAGG 4 41, 32 ATCC 49775 AAAAAACTTACAGGCAGTCCATCTC 4 sej TAACCTCAGACATATATACTTCTTTAACG 4 41, 32 DO-204 AGTATCATAAAGTTGATTGTTTTCATGCAG 4 sek TTA GGT GTC TCT AAT AGT GCC AGC 5 5 FRI 913 AGC TGT GAC TCC GCC ATA TAT GTA 5 sel ATC ACA CCG CTT AGA ATA CTC 5 5 FRI 913 GGT ATT ATT CTT GGC GAA TCT 5 sem TGC TAT TCG CAA AAT CAT ATC GCA ACC 5 5 ATCC 49775 TCA ACT TTC GTC CTT ATA AGA TAT TTC TA 5 sen GCT TAT GAG ATT GTT CTA CAT AGC TGC 5 5 ATCC 49775 CAT TAA CGC CTA TAA CTT TCT CTT CAT C 5 seo GAG AGT TTG TGT AAG AAG TCA AGT G 5 5 ATCC 49775 GAT TCT TTA TGC TCC GAA TGA GAA 5 seu TAAAATAAATGGCTCTAAAATTGATGG 6 6 EELA-7 CGTCTAATTGCCACGTTATATCAGT 6 lukED TGAAAAAGGTTCAAAGTTGATACGAG 4 41, 32 FRI 913 TGTATTCGATAGCAAAAGCAGTGCA 4 41, 32

lukM TGGATGTTACCTATGCAACCTAC 4 41, 32 FRI 913,

ATCC 31890 GTTCGTTTCCATATAATGAATCACTAC 4 41, 32

32

Table 1. Continued Gene Primer sequence (5´-3´) Reference Conditions Control strain hla CTGATTACTATCCAAGAAATTCGATTG 4 41, 32 NCTC 8325 CTTTCCAGCCTACTTTTTTATCAGT 4 41, 32 hlb GTGCACTTACTGACAATAGTGC 4 41, 32 EELA

2285195 GTTGATGAGTAGCTACCTTCAGT 4 41, 32 hlg GCCAATCCGTTATTAGAAAATGC 7 41, 32 FRI 913 CCATAGAYGTAGCAACGGAT 8 41, 32 hld AAGAATTTTTATCTTAATTAAGGAAGGAGTG 4 41, 32 NCTC 8325 TTAGTGAATTTGTTCACTGTGTCGA 4 fnbA GCGGAGATCAAAGACAA 9 11 CCUG 47326 CCATCTATAGCTGTGTGG 9 fnbB GGAGAAGGAATTAAGGCG 9 11 CCUG 47326 GCCGTCGCCTTGAGCGT 9

1 Brakstad et al., (1992) 2 Vesterholm-Nielsen et al., (1999) 3 Mehrotra et al., (2000) 4 Jarraud et al., (2002) 5 Smyth et al., (2005) 6 Letertre et al., (2003) 7 Lina et al., (1999) 8 von Eiff et al., (2004) 9 Nashev et al., (2004) 10 Haveri et al., (2005) 11 Initial denaturation step (5 min at 94ºC) followed by 30 cycles of amplification (denaturation for 30 s at 94ºC, annealing for 30 s at 50ºC and elongation for 1 min at 72ºC) terminated with a 7 min incubation step at 72ºC 41 PCR conditions described by Jarraud et al., (2002) was used in study III 32 PCR conditions described by Mehrotra et al., (2000) was used in study IV

33

5.7 Classification of clinical severity and persistence of IMI

Clinical signs of the individual cows in studies II-III were graded as defined by the IDF (1997), with the following modifications: the type of mastitis or clinical signs of the cow were graded as subclinical (CMT score <3, no clinical signs), mild clinical (visual abnormalities in milk and/or swelling, heat, redness, or tenderness in the udder), or severe clinical, if elevated rectal temperature (>39.0°C) and other general signs (anorexia, decreased rumen motility, and general attitude) were also observed. In study II, the IMI was considered persistent and to represent the same episode, if the same pulsotype, compared with the initial isolate was isolated during the follow-up period. The IMI was classified persistent when the first follow-up isolate of S. aureus was of indistinguishable pulsotype and virulence gene profile compared with the initial isolate (II-III), or if the same S. aureus pulsotype was repeatedly isolated from the same quarter during the same lactation (IV). Quarters healthy before the farm visit but in which S. aureus was detected later during the study period were defined as new infections.

5.8 Bacteriological definitions

Isolate was considered to represent a group of monoculture bacterial cells obtained from a primary colony growing on solid medium where the sample was originally inoculated. Strain was defined as a variant of bacterial isolate having phenotypic or genotypic properties that distinguished the strain from other isolates of the same species. Clonal type, genotype, and pulsotype were used as synonyms to indicate a group of strains most likely having a common ancestor, and were assigned on the basis of fingerprints obtained with PFGE typing.

5.9 Statistical analyses

The differences in bacterial diagnoses and resistance patterns between 1995 and 2001 were compared using the two-sample proportion test in Statistix (Analytical Software, Tallahassee, FL) (I). The effect of pulsotype on the severity of the clinical signs was tested using multinomial logistic regression (proportional odds model) by including time of infection (days from calving) in the model (II). The cows were divided into two groups according to the stage of lactation: days 0-30 and days > 30 post-partum. The cows were also categorised as primiparous and multiparous (>1 parities) according to the parity. The effects of pulsotype and β-lactamase production of the S. aureus isolate on persistence of infection were tested by logistic regression by including time of infection and severity of clinical signs into the model. The influence of other measured variables, such as parity and number of infected quarters, were preliminary screened out by Fischer’s Exact Test. The effect of pulsotype, including time of infection on NAGase activity was tested by two-way analysis of variance, using log-transformed values for NAGase. One quarter value of the same cow was randomly selected into the analyses and this was repeated twice. The possible effect of genetic variation between the representative clones of pulsotype A-E on severity of signs and persistence were preliminary screened by Fischer’s Exact Test. The analyses were carried out with Stata 7.0

34

(Stata Statistical Software, College Station, Tex.). P values of <0.05 were considered to be statistically significant. The effects of each gene on the severity of the clinical signs (n=116 S. aureus isolates) was tested by using ordinal logistic regression (III). Pearson’s chi-square test or Fischer’s exact test were used to study the associations of virulence genes with persistence of IMI. The effect of each gene on persistence of IMI within-pulsotype was tested by using Pearson’s chi-square test or Fischer’s exact test. The correlation of cumulative number of virulence genes and severity of signs was tested using Pearson two-tailed correlation. The analyses were conducted with SPSS Statistical Software, version 13.0 (SPSS Inc., Chicago, Illinois).

35

6. RESULTS

6.1 Prevalence and occurrence of Staphylococcus aureus (I, IV)

6.1.1 In the herds included in the Finnish national mastitis survey (I)

Staphylococcus aureus IMI was isolated in 430 (3.4%) of the 12 661 quarters of the 3270 cows sampled in the national survey. In total 334 cows (10.2%) were positive for S. aureus, which was detected in 140 (64.8%) of the 216 farms. Herds with S. aureus positive cows had on average 17.0 cows (min. 3, max. 48) with a mean milk yield of 7670 kg per annum (min. 5000, max. 11 000), and a mean BMSCC of 146 500 cells/ml (min. 3000, max. 578 000). The majority (88.6%) were housed in stanchion-barns, nine herds had warm loose-housing (6.4%) and seven herds (5.0%) had cold loose-housing. According to the health data of the herds, an average of 4.4 mastitis treatments (min. 0, max. 19) were recorded in the herds during 2000 (Anna Pitkälä, Evira, unpublished data). Herds that were not positive for S. aureus IMI in the survey comprised on average 11.7 cows (min. 2, max. 32), had a mean milk yield of 7178 kg per annum (min. 2400, max. 11 200) and a mean BMSCC of 129 700 cells/ml (min. 20 000, max. 800 000). Seventy of the herds (92.1%) were housed in stanchion-barns and the remaining six herds (7.9%) had warm loose-housing. On average 2.9 mastitis treatments (min. 0, max. 22) were recorded in these herds in 2000 (Anna Pitkälä, unpublished data). In Table 2 characteristics of the herds grouped by number of cows in the herd are shown. Table 2. Characteristics of the Finnish survey herds with S. aureus IMIs grouped according to number of cows in the herd. In the second column, the proportion of S. aureus positive herds among all herds of the given size is shown. With permission of A. Pitkälä (unpublished data).

No. of cows in

the herd

No. of herds with S. aureus/ all herds

of the size (%)

Mean % of S. aureus

positive cows of all cows in the herd (min,

max)

Mean BMSCC cells/ml(min, max)

Mean milk yield,

kg/cow/year (min, max)

Mean no. of recorded mastitis treatments during

lactation (min, max) /cow in

2000**

1-10 32/73 (43.8)

1.9 (1, 6)* 154 484 (33 000, 578 000)

7044 (5000, 9146)

0.34 (0.0, 1.7)

11-20 69/98 (70.4)

15.0% (5.0, 63.6)

142 522 (38 000, 440 000)

7540 (5000, 10 000)

0.23 (0.0, 0.75)

21-30 29/33 (87.9)

12.1% (3.5, 26.0)

156 482 (44 000, 544 000)

8236 (6000, 11 000)

0.30 (0.0, 0.86)

>30 10/12 (83.3)

9.4% (3.2, 19.4)

135 444 (55 000,190 000)

8787 (7655, 9621)

0.032 (0.030, 0.37)

Total 140/216 (64.8)

* Mean no. of cows with S. aureus (min-max) is shown instead of % for herds with 1-10 lactating cows. **Calculated on the basis of the number of treatments in the herd health records.

36

The proportion of quarters positive for S. aureus among those with bacterial growth is shown in Table 3. S. aureus was present in 22.6% of the 155 samples harbouring two bacterial species concomitantly, usually together with CNS. Cows with S. aureus IMI had on average 1.28 infected quarters. Of these cows, 75.0% had one quarter, 16.5% two quarters, and 3.9% three quarters with S. aureus IMI. Approximately half of the quarters positive for S. aureus produced milk with an SCC over the threshold of 300 000 cells/ml. S. aureus IMI was slightly less common in cows in their first and second parities than in the older cows, but the difference was not statistically significant. Table 3. Bacterial isolates found during a survey of 2001. In the first column percentages of quarter milk samples are shown. Proportions of different pathogens are given in the second column. n=12 661 n=4237 CNS 16.61 49.63 Staphylococcus aureus 3.40 10.17 Streptococcus agalactiae 0.02 0.07 Streptococcus dysgalactiae 0.05 0.14 Streptococcus uberis 0.65 1.94 Other streptococci 0.07 0.21 Enterococcus spp. 0.40 1.20 Aerococcus viridans 0.24 0.71 Lactococcus spp. 0.24 0.73 Corynebacterium bovis 11.52 34.41 Coliforms 0.14 0.42 Other bacteria 0.12 0.37 Contaminated samples 4.15 No bacteria 62.38 Total 100.0 100.0

6.1.2 In IMI and extramammary sites (IV)

S. aureus IMI was detected in 26 quarters of 17 cows in Herd I, and in 11 quarters of 10 cows in Herd II during the study period. In Herd I, eight cows (27%) and 12 quarters (10%) had S. aureus mastitis during the follow-up period before the farm visit. For Herd II, the respective figures were five quarters (2%) in five cows (9%). At the time of the farm visit, S. aureus IMI was detected in four cows (15% of 27 cows) and seven quarters (7% of 108 quarters) in Herd I. Twenty-eight percent of the samples from the other sites harboured S. aureus (Table 4); 48% of the quarters and 85% of the cows had S. aureus on the teat walls, teat orifices and/or in teat canals. Presence of S. aureus in teat end (P<0.05) and in teat canal (P<0.01) was associated with S. aureus IMI. Eighty percent of the quarters that became infected with S. aureus during the study also had the bacterium on the teat skin and/or in teat canals, whereas for those sites on the unaffected quarters, the respective figure was only 42%. Skin lesions of seven cows (26%) supported growth of S. aureus, and two of the cows had concomitant S. aureus mastitis. In Herd II, six of the 51 cows (12%) had S. aureus IMI, with seven infected quarters (3%) during the farm visit. S. aureus was detected in 44% of samples from the other sites. Ninety-six percent of the cows and 81% of the quarters had the bacterium on the teat skin or in the teat canals. Presence of S. aureus in the teat canal was associated with S. aureus in the milk (P<0.01). Twenty-one cows had S. aureus in the skin lesions, four cows having concomitant S. aureus IMI.

37

During the follow-up period after the farm visit, S. aureus IMI was detected in 11 cows (37%) and in 14 quarters (12%) in Herd I. In Herd II, S. aureus IMI was detected in three cows (6%) and in three quarters (1%). Table 4. S. aureus isolated from different sample sites in Herds I and II during the farm visit. Sample site No. (%) of samples

with S. aureus/ No. of all samples in Herd I

No. (%) of samples with S. aureus/ No. of all samples in Herd II

Total no. (%) of samples with S.

aureus/ No. of all samples

Milk 7/108 (6.5) 7/197 (3.6) 14/305 (4.6) Teat wall 27/108 (25.0) 131/204 (64.2) 158/312 (50.6) Teat orifice 27/108 (25.0) 130/192 (67.7) 157/300 (52.3) Teat canal 32/108 (29.6) 35/191 (18.3) 67/299 (22.4) Skin lesion 7/8 26/33 (78.8) 33/41 (80.5) Liner, before milking 2/4 0/32 (0.0) 2/36 (5.6) Liner, after milking 28/108 (25.9) 48/186 (25.8) 76/294 (25.6) Milkers’ hands 2/2 0/2 2/4 Milkers’ nostrils 2/2 3/4 5/6 Total 134/556 (24.1) 380/1041 (37.5) 521/1597 (32.6)

6.2 Diversity of Staphylococcus aureus strain-types originating from mastitis infections (II-III)

6.2.1 Diversity and distribution of pulsotypes and gene profiles

The 217 S. aureus isolates collected from mastitis infections between September 1993 and January 1997 (II) comprised 32 PFGE fingerprints of 22 pulsotypes (A-V) analysed visually for genetic relatedness. Pulsotypes A to E were considered to consist of 2 to 5 closely related clones, A1-A5, B1-B2, C1-C2, D1-D2, and E1-E4. Using computer-aided analysis with an 80% cluster cut-off value, S. aureus isolates were distributed among 25 different clusters. The clones of pulsotype A and B are shown in Figure 2.

Figure 2. Clones of pulsotype A and B isolated in study II. From the left: control (NCTC 8325); pulsotype A, clones from A1 to A5; control (NCTC 8325); pulsotype B, clones B1 and B2; pulsotype B, clones B1 and B2; control (NCTC 8325).

38

Pulsotypes A to E were isolated from several cases of mastitis and these were considered to be the predominant pulsotypes. Pulsotypes F-V were each detected only once and were regarded as being sporadic pulsotypes. The occurrence of the different pulsotypes from 1993 to 1997 is presented in Figure 3. Pulsotype A and E were observed continuously and pulsotype B until 1996. Pulsotype C and D appeared in 1994 but disappeared after 1995 and 1996.

9796959493

year

15

12

9

6

3

0

N o

f her

d

type Etype Dtype Ctype Btype A

type1

Figure 3. Occurrence of the predominant S. aureus pulsotypes (A-E) isolated from bovine mastitis in various herds between 1993 and 1997 in the practice area of the Ambulatory Clinic of Helsinki University, in the southern Finland (II). The 116 pre-treatment S. aureus isolates represented pulsotypes A-E, F-I, K-O, R, T, or U, whereas pulsotypes J, P, Q, S, and V were isolated only from the follow-up samples after treatment. Among the 116 pre-treatment isolates, the genes hla, lukED, hlg, hld and hlb were the most common, being identified in 97.4%, 96.6%, 88.8%, 87.9% and 76.7% of the isolates, respectively. Sixty-nine percent of the isolates contained at least one SAg determinant, the most frequent being seo (in 44.0%), sen (39.7%), sem (36.2%), seg in combination with sei (31.9%), sed together with sej (19.8%) and seu (19.8%). lukM was identified in 32 isolates (27.6%). The genes see, eta, and etb were not detected. Forty-three virulence gene profiles were recorded. The most common combination contained hla, hlb, hld, hlg and lukED (n = 23), which was detected in pulsotype A (n=19), I, L, M, and O. hla, hlb, hld, hlg, lukED, sed, and sej was the next most common profile, being identified

39

in pulsotype A only (n=13). hla, hld, hlg, lukED, seg, sei, sem, sen, and seo (n = 8) was typical of pulsotype E (n=5), and hla, hlb, hld, hlg, lukED, lukM, seg, sei, sem, sen, seo, and seu positive isolates (n = 7) were usually of pulsotype B (n=5). The genes seg, sei, sem, sen, and seo were detected concomitantly in a total of 34 isolates (29.3%), and in 14 isolates (12% of all) they were accompanied by seu. sec and tst were present in eight isolates, those of six also had sel. The genes encoding SAg accumulated in the predominant pulsotypes A-E (73.1% positive) and were significantly less common among the sporadic pulsotypes (33.3% positive) (P<0.01). Pulsotype B harboured the highest number of virulence genes, varying between 7 (9.1%) and 15 (45.5%). Only 56% of pulsotype A carried the genes for SAg, whereas pulsotype B, C, D, and E always had the genes (Figure 4).

Figure 4. Dendrogram showing genetic similarity (%) of the pulsotypes and their clonal variants derived from 116 bovine S. aureus isolates at the onset of mastitis. The frequencies of genes encoding superantigens SAg of each pulsotype and clonal variant (A1-U) are presented. egc1 indicates the complete egc cluster, consisting of the genes seg, sei, sem, sen, and seo. egc2 indicates the complete egc cluster and seu.

40

6.2.2 Diversity within and between farms

In 54 herds (77%) only a single pulsotype was detected. Of the 29 herds where several mastitis cases (2 to 6) occurred, infections were due to the same pulsotype with indistinguishable gene profiles in 17 herds. In five herds several cases of mastitis were caused by the same pulsotype, but the number and quality of SAg genes of this pulsotype varied on different occasions. In six of the 12 herds where unrelated pulsotypes were found, the unrelated S. aureus pulsotypes caused mastitis simultaneously. The clones A1, A2, B1, B2, C1, C2, D1, D2, E1, and E2 of pulsotypes A-E were found in several herds.

6.2.3 Diversity within cows

In cows with several quarters with concomitant S. aureus mastitis, PFGE fingerprints of the isolates were indistinguishable and were considered to represent the same clone. In seven cows a new quarter was infected with S. aureus during the follow-up period, and their PFGE fingerprints (A, D, E) were indistinguishable from those of the initial S. aureus isolate. In 45 cows S. aureus mastitis was still detected after treatment in the same quarter. In 40 cows the initial S. aureus strain was found in the same quarter during the follow-up period after treatment. These infections were caused by pulsotype A, B, C, D, E, F, G, or R with identical virulence gene profiles as the initial isolate.

6.3 Antimicrobial resistance of Staphylococcus aureus (I-III)

6.3.1 Resistance to penicillin (I-III)

In the mastitis survey (I), penicillin resistant (β-lactamase positive) S. aureus strains were isolated from 69 herds, which comprised 49.3% of the herds with S. aureus IMIs and 32.1% of all herds surveyed. The herds with penicillin resistant S. aureus included 18.2 cows on average (median 16; min 3, max 48). The general herd characteristics did not differ from the mean figures for all S. aureus infected herds. On average 3.2 quarters per farm (median 2.0; min 1, max 15) had IMI due to penicillin resistant S. aureus. Thirty-four of the herds (49.3%) had both penicillin resistant and penicillin susceptible strains. In those herds, on average 56.2% (median 50.0%; min 25.0%, max 92.3%) of the quarters with S. aureus IMI harboured penicillin resistant strains. More than half (52.1%) of the 196 randomly selected S. aureus isolates collected in the survey produced β-lactamase (I). In studies II-III, 51 of the 116 isolates (44.0%), derived from cases of mastitis prior to treatment, produced β-lactamase and were positive for blaZ. Such isolates were retrieved from 34 (48.6%) of the 70 herds. The share of penicillin resistant strains was 42% for pulsotype A, 4.5% for type B, and 100% for type E. Four of the six pulsotype D isolates were penicillin resistant, but none of the pulsotype C. Pulsotype A (P<0.05), D (P<0.01), and E (P<0.001) were significantly more often penicillin resistant compared with pulsotype B. Penicillin resistance was not linked with particular clonal variants of pulsotype A and D, thus, PFGE was unable to differentiate between the penicillin resistant and penicillin susceptible strains within a particular pulsotype. Among the 17 sporadic types, pulsotype F, I, K, N, R, T, and U were penicillin resistant.

41

6.3.2 Resistance to other antimicrobials (I)

Resistance of S. aureus to different antimicrobials isolated in the survey is presented in Table 5. Multiresistance (resistance to more than two antimicrobials) was found in 2.0% of the 196 isolates tested. All of the tested isolates were susceptible to cephalotin, enrofloxacin, gentamycin, neomycin and spiramycin. Table 5. Antimicrobial resistance among the 196 Staphylococcus aureus isolates originating rom the Finnish survey herds. Antimicrobial

Breakpoint for resistance μg/ml

% resistant MIC50 (μg/ml)1

MIC90 (μg/ml)1

Range (μg/m)

Clindamycin >2 0.5 1 1 1 to 4 Erythromycin >4 1.5 0.5 0.5 0.5 to 8 Oxacillin >2 4.12 1 2 0.5 to 8 Oxytetracycline >8 5.1 1 1 0.5 to 128 Penicillin3 52.1 Streptomycin >32 4.1 4 16 1 to 256 Trimethoprim/ Sulfametoxazole4

>2 1.5 0.25 0.25 0.25 to 16

1MIC50 and MIC90 = Minimum inhibitory concentrations at which 50 and 90% of isolates are at or below 2Primary result, oxacillin susceptible according to confirmatory test 3β-Lactamase production 4Concentration of trimethoprim is given, tested at concentration ratio 1:19

6.4 Relationship of Staphylococcus aureus strain-types to the inflammatory response of the affected quarter, clinical characteristics and persistence of mastitis (II-III)

6.4.1 Staphylococcus aureus pulsotypes (II)

In the 116 cows treated for S. aureus mastitis from 1993 to 1997, S. aureus mastitis occurred between 1 and 321 days in milk, with an average of 81 days and a median of 51 days post partum. Thirty-six percent of the cases were diagnosed within the first two weeks after calving and 43% during the first month post partum. The clinical signs of the cows were more severe (P < 0.05) and milk NAGase activity values were higher (P < 0.05) within the first month after calving than during the later lactation. The severity of the clinical signs (Table 6) recorded correlated with udder inflammatory response determined by milk NAGase activity. Table 6. Characteristics of 116 cases of bovine S. aureus mastitis (II). Severity of signs

No. of herds

No. of cows

(quarters)

Mean rectal temperature ˚C (SE)

Mean NAGase

U/ml (SE)

Pulsotype of the infecting strain (No. of cows)

Subclinical 25 34 (39) 38.3 (0.11) 146 (26.7)

A (15), B (2), C (5), E (6), H (1), K (1), O (1), R (1), U (1)

Mild clinical

41 53 (62) 38.8 (0.067) 350 (29.6)

A (23), B (8), C (3), D (5), E (11), F (1), L (1), M (1), N (1)

Severe clinical

27 29 (33) 40.1 (0.11) 460 (36.6)

A (12), B (12), D (1), E (1), G (1), I (1), T (1)

All 70 116 (134) 39.4 (0.10) 323 (21.5)

A (50), B (22), C (8), D (6), E (18), F-U (1 each)

42

Association between pulsotype, severity of clinical signs and milk NAGase activity was tested for strains of pulsotypes A, B, C, D and E (n=104 strains). Infections caused by pulsotype B were related with more severe signs. Compared with this genotype, mastitis caused by pulsotype A (P<0.05), pulsotype C (P<0.01), or pulsotype E strains (P<0.05) were significantly less likely to be associated with increased severity of mastitis. Milk NAGase activity levels of quarters with mastitis caused by pulsotype B strains were significantly

higher (mean, 427 to 441 U/ml; SE 45.7-48.0) compared with those from mastitis caused by pulsotype C (mean, 124 to 127 U/ml; SE 27.0-28.2) (P < 0.05), and they tended to be higher than those associated with pulsotypes A (mean, 307 to 317 U/ml; SE 32.5-33.3) (P 0.05) and E (mean, 282 to 307 U/ml; SE 46.0-54.6). The NAGase activity values associated with pulsotype D (mean, 420 to 433 U/ml; SE 60.5-62.8) were close to those for pulsotype B strains. For pulsotypes A, B, C, D and E (No. of strains=97), persistent mastitis caused by pulsotype A (P<0.01; OR 23.1; 95% CI for OR 2.8-190.9), pulsotype C (P<0.05; OR 22.9; 95% CI for OR 1.8-288.7), pulsotype D (P<0.01; OR 41.9; 95% CI for OR 2.9-598.6) and pulsotype E (P<0.01; OR 2.4; 95% CI for OR 2.4-227.7) was significantly more common compared with pulsotype B, which were eliminated from the quarters almost always. Twenty cows (40%) infected with pulsotype A, 13 (59%) with pulsotype B, 2 out of the six cows infected with pulsotype C, and 8 cows (44%) infected with pulsotype E were all diseased within one month from calving. Mastitis due to pulsotype C was detected only during later lactation.

6.4.2 Staphylococcus aureus pulsotypes and virulence gene profiles (III)

The follow-up data were complete and persistence of mastitis could be assessed in 106 of the 116 cows. However, in five cows, the same pulsotype was isolated from the quarter but the gene profiles differed from those of the original strains. These cases were excluded from the analyses. Finally, persistent mastitis developed in 40 cows and was eliminated from 61 cows. Severity of clinical signs and persistence of mastitis caused by different S. aureus pulsotypes is shown in Table 7. Table 7. Severity of clinical signs and persistence of mastitis caused by different Staphylococcus aureus pulsotypes (PT). Clinical signs were determined for 116 cows and persistence status for 101 cows. Signs of mastitis Persistence status PT Herds Cows

(quarters) Subclinical Mild Severe Recovered Persistent ND

A 34 50 (57) 15 23 12 22 24 4 B 19 22 (25) 2 8 12 20 1 1 C 3 8 (10) 5 3 0 4 3 1 D 4 6 (8) 0 5 1 2 3 1 E 12 18 (20) 6 11 1 7 6 5 F-I, K-O, R, T, U*

12 12 (14) 5 4 3 6 3 3**

Total 116 (134) 33 54 29 61 40 15 *ND, not determined ** Three quarters of one cow were infected with pulsotype H **Persistence status for strains of pulsotypes I, T and U not determined.

43

The genes for SAg were present in 80.0% of isolates from persistent mastitis, as opposed to 63.9% of those from non-persistent mastitis. The difference was significant (31/39 vs 19/41; P<0.01) after exclusion of pulsotype B strains, which were nearly always eliminated. Pulsotype A carried most of the genes sed and sej (21 of 23), which were associated with persistent mastitis (Table 8). The link between non-persistent IMI and the genes seg, sei, sem, seo, seu, and lukM ceased to exist if pulsotype B, which usually carried these genes, were excluded. Thereafter lukM positive strains tended to dominate in moderate and severe clinical mastitis (in 9 of 10 cases), but IMI seemed to persist more often than in cases caused by lukM negative isolates. The positive correlation recorded between the cumulative number of virulence genes and the severity of clinical signs (0.215; P<0.05) was not statistically significant in the absence of pulsotype B. Table 8. Virulence genes in Staphylococcus aureus isolated from 116 cows with different clinical signs of mastitis prior to treatment. Possible persistence of mastitis was determined for 101 cows.

Severity of signs (116 cows)

Persistence of IMI (101 cows)a

Gene Subclinical (n=33)

Mild (n=54)

Severe (n=29)

Recovered (n=61)

Persistent (n=40)

sea 3 (9.1%) 4 (7.4%) 0 2 (3.3%) 4 (10.0%) seb 1 (3.0%) 0 0 0 1 (25.0%) sec 2 (6.1%) 4 (7.4%) 3 (10.3%) 9 (14.8%) 0* sed, sej 6 (18.2%) 11 (20.4%) 6 (20.7%) 5 (8.2%) 17 (42.5%) ** seg, sei 6 (18.2%) 21 (38.9%) 10 (34.5%) 23 (37.7%) 8 (20.0%)* seh 5 (15.1%) 2 (3.7%) 0 4 (6.6%) 3 (7.5%) sek 4 (12.1%) 2 (3.7%) 0 2 (3.3%) 4 (10.0%) sel 0 3 (5.6%) 3 (10.3%) 6 (9.8%) 0 sem 7 (21.2%) 24 (44.4%) 11 (37.9%) 25 (41.0%) 9 (22.5%)* sen 7 (21.2%) 26 (48.1%) 13 (44.8%) 28 (45.9%) 10 (10.0%)* seo 8 (24.2%) 27 (50.0%) 16 (55.2%) 30 (49.2%) 13 (32.5%)* seu 2 (6.1%) 8 (14.8%) 13 (44.8%) 19 (31.1%) 2 (5.0%)** tst 1 (3.0%) 4 (7.4%) 3 (10.3%) 8 (13.1%) 0* lukE-D 32 (96.7%) 51 (94.4%) 29 (100.0%) 61 (100.0%) 36 (90.0%) lukM 3 (9.1%) 13 (24.1%) 15 (51.7%) 23 (37.7%) 5 (15.0%)** α 32 (97.0%) 53 (98.1%) 28 (96.7%) 59 (96.7%) 39 (97.5%) β 25 (75.6%) 39 (72.2%) 25 (86.2%) 48 (78.7%) 29 (72.5%) δ 29 (87.9%) 48 (88.9%) 24 (82.8%) 54 (88.5%) 34 (85.0%) γ 27 (81.8%) 47 (87.0%) 27 (93.1%) 55 (90.2%) 35 (87.5%) a Differences in occurrence of genes between non-persistent and persistent mastitis. * P<0.05 ** P<0.01

6.4.3 The role of Staphylococcus aureus pulsotypes, virulence gene profiles, and penicillin resistance in persistent mastitis (III)

The blaZ gene was detected in 51 isolates (44.0%). The gene was present as often in S. aureus strains causing subclinical mastitis (in 15 of 33; 45.5%) as in those strains causing clinical mastitis (in 36 of 83; 43.3%). blaZ was detected in 57.4% of the isolates with genes for SAg, but in only 13.9% of the isolates negative for those genes (P < 0.01). Twenty-three of the blaZ positive strains (45.1%) contained sed and sej; seo was detected in 26 (51.0%), and sem and sen in 23 strains (45.1%). All isolates with sed and sej also had blaZ. Among the 65 blaZ negative strains (56.0%), seo was detected in 25 (38.5%), sen in 23 (35.4%), seu in 22

44

(33.8%), and sem in 19 strains (29.2%). The blaZ gene was not detected together with sec, sel, tst, and seu, the genes characteristic of pulsotype B (95.0% blaZ negative) or with seh and sek, which were only found from pulsotype C (always blaZ negative). Persistent mastitis was significantly more commonly caused by blaZ positive (61.0% remained persistent) than blaZ negative (25.0%) S. aureus strains (P<0.01). Among the blaZ positive strains, persistence was significantly more common (P<0.05) if the infecting isolate had sed, sej, and blaZ (17/22; 76.2% persistent) than blaZ only (9/20; 45.0% persistent). When the isolates positive for sed, sej, and blaZ were excluded, elimination rates from mastitis caused by blaZ positive strains were still lower (42.1% remained infected) than those caused by blaZ negative strains (25.0% remained infected), but the difference was not significant (P=0.15). Within pulsotype A, the most common pulsotype, 15 of the 24 strains from persistent mastitis carried sed, sej, and blaZ, whereas only five out of 22 strains from non-persistent mastitis had them (P<0.01). Treatment response to amoxycillin clavulanate and cloxacillin was lower for blaZ positive strains of pulsotype A (10/13 and 7/8 remained infected) compared with blaZ positive pulsotype E strains (2/4 and 5/9 remained infected).

6.5 Staphylococcus aureus strain types in IMI and extramammary sites (IV)

PFGE typing produced nine fingerprints, which were classified as pulsotypes A1-G. The occurrence of the pulsotypes in different sample sites in Herd I and II is shown in Table 9 and 10. Table 9. Staphylococcus aureus isolates and their PFGE types according to the origin of isolates in Herd I. Origin No. of samples with S.

aureus of total No. of samples (%)

No. of PFGE-typed S. aureus

isolates (%)

PFGE type (n)

Milk 37a (ND*)

38 (100) A1 (27), B (10), A1+B (1)

Teat wall 27/108 (25) 9 (33) A1 (7), B (1), A1+B (1) Teat orifice 27/108 (25) 13 (48) A1 (10), B (3) Teat canal 32/108 (30) 14 (44) A1 (11), B (3) Skin 7/8 (88) 7 A1 (1), B (5), A1+B (1) Liner before milking 2/4 2 A1 (2) Liner after milking 28/108 (26) 12 (43) A1 (11), B (1) Milkers’ hands 2/2 2 A1 (2) Milkers’ nostrils 2/2 2 A1 (1), C (1) Total 165 99 (ND) A1 (72), B (23), A1+B (3),

C (1) aIncludes follow-up samples taken after treatments. *ND = Not determined

45

Table 10. Staphylococcus aureus isolates and their PFGE types according to the origin of isolate in Herd II. Origin No. of samples with S.

aureus of total no. of samples (%)

No. of PFGE-typed isolates (%)

PFGE type (n)

Milk 21a (ND*) 21 (100) D (17), F (2), A2 (1), E2 (1) Teat wall 131/204 (64) 43 (34) D (38), E1 (2), A2 (1), B (1),

F (1) Teat orifice 130/192 (68) 37 (28) D (33), F (2), A2 (1), B (1) Teat canal 35/191 (18) 20 (57) D (16), B (2), A2 (1), F (1) Skin 26/33 (79) 26 (100) D (24), E1 (1), F (1) Liner before milking 0/32 0 0 Liner after milking 48/186 (26) 40 (83) D (38), B (1), E1 (1), Milkers’ hands 0/2 0 0 Milkers’ nostrils 3/4 3 D (2), G (1) Total 390 190 (ND) D (168), F (7), B (5), A2 (4),

E1 (4), E2 (1), G (1) aIncludes follow-up samples taken after treatments. *ND=Not determined Pulsotype A1 caused mastitis in 13 cows (19 quarters). In six of those cows (8 quarters), the same pulsotype was detected several times in the same quarter, indicating persistent IMI. Pulsotype A1 predominated in samples from the extramammary sites (Table 9) and was detected in 90% of S. aureus positive cows. Pulsotype A1 was present in the teat skin and/or teat canal in six cows without S. aureus IMI. Pulsotype B caused mastitis in eight cows (8 quarters). At the time of the farm visit, pulsotype B was isolated in the extramammary sites in three healthy cows. Pulsotype B was common in the skin lesions. Indistinguishable pulsotypes were isolated from quarter milk and teat skin and/or canal in the same cows. The same pulsotype was found from skin lesions and from milk of the same cows. In two cows, pulsotypes A1 and B were isolated in milk of different quarters at the same time. One cow harboured pulsotype A1 and B in the milk of the same quarter. Pulsotype D was continuously isolated from milk in Herd II. It caused mastitis in seven cows (eight quarters) and was isolated several times in the milk of three cows. At the farm visit, pulsotype D was present in the extramammary sites in 41 cows. Two preparturient heifers sampled had pulsotype D in the colostrum and on the teat skin and all three heifers had it on the skin lesions. Pulsotype F was found together with pulsotype D in the extramammary sites in four cows and pulsotype E1 in three. In one cow, pulsotype F was repeatedly isolated from the same mastitis quarter. Pulsotypes A2, B, and E2 were sporadic; each of them were isolated in single cows. Six cows had S. aureus of unrelated pulsotype in the milk compared with the teat skin or the teat canal of the same quarter, but the pulsotype found in milk was also present on the skin of another quarter. The same pulsotype D was found from skin lesions and from milk in the same cows. Two cows had pulsotype D in the teat canal and teat skin, and these quarters developed mastitis caused by that pulsotype after the sampling day. Approximately one-fifth of the milking liners were contaminated by S. aureus after milking in both herds. In Herd II, among the 40 PFGE-typed S. aureus isolated from the liners, only one S. aureus strain appeared to originate from infected quarter milk; the other strains probably originated from teat skin and teat canals of the quarters of healthy cows. In Herd I, two liners were already contaminated with S. aureus prior to milking, when the origin of S. aureus in these liners remained undetermined.

46

In 25 samples, variation in colony morphology, colour, or haemolysis was visually detected. To examine whether the colony variants represented different pulsotypes, two colonies from each of these samples were PFGE-typed. Except for three samples in Herd I (milk, teat skin, skin lesion; Table 3), each sample had only one pulsotype. Variations in the gene profiles were recorded among the most common pulsotypes, A1, B, and D. The individual genes and the different gene profiles associated with multiple S. aureus isolates were not linked with the origin of the isolate. The pulsotypes and the gene profiles are shown in Figure 5. In Herd I, pulsotype A1 from the cows and the milking liners (n=72) had hla-hlg, lukED, fnbA, and fnbB and with one exception, blaZ, sed, and sej, but seldom lukM (n=8), and only once sem and seo. Pulsotype A1 from milkers’ hands (n=2) and nostrils (n=1) were positive for hla-hlg, lukED, fnbA, fnbB, blaZ, sed, and sej, as pulsotype A1 from the cows. In Herd II, pulsotype D strains harboured hla, hld, hlg, fnbA, and fnbB and 103 had lukED, but only a few were positive for lukM (n=4), blaZ (n=3), seg and sei (n=2), sed (n=1), sem (n=1), or seo (n=1). Pulsotype D found in the nostrils of two milkers was positive for hla, hld, hlg, lukED, lukM, fnbA, and fnbB.

Figure 5. Dendrogram showing the numbers and the genetic relatedness of pulsotype A1-G detected in Staphylococcus aureus isolates originating from Herds I and II between October 2002 and August 2003. The frequencies of the different genes for each pulsotype are presented.

47

7. DISCUSSION

7.1 Prevalence and occurrence of Staphylococcus aureus

Considerable efforts have been made to control bovine mastitis in Finland since the late 1980s. The measures used have largely been successful. Regular nationwide monitoring of bovine mastitis demonstrates that mastitis incidence declined from 48% during the first survey, carried out in 1995 to 31% in 2001 (I). The control measures, including intensified culling of S. aureus-infected cows, have also contributed to reduction in the prevalence of S. aureus in Finland, which decreased from 5.1% (1988) to 3.5% (1995). In 1980, the National Veterinary and Food Research Institute reported S. aureus in about 50% of the milk samples from subclinical and clinical mastitis cases sent to the laboratory, but in 1989 this fell to 25% (Myllys, 1995). Measures implemented to control mastitis have almost eradicated Str. agalaciae, another contagious mastitis pathogen, from Finland (I). However, S. aureus has remained a significant mastitis pathogen. The same phenomenon was recently reported in a Norwegian national mastitis survey, where S. aureus was frequent but Str. agalaciae was not isolated at all (Østerås et al., 2006). Although the proportion of S. aureus among bacterial isolations declined from 16.7% to 10.2% during the past decade, both quarter (3.5-3.4%) and cow prevalence (10.1-11.1%) of S. aureus IMI in Finland has remained nearly stable (Myllys et al., 1998; study I). Bacterial growth was detected in 33.5% of milk samples, which is significantly more than in 1995 (21.0%; P<0.01). S. aureus is still the most common pathogen isolated from milk samples from clinically mastitic cows in Finland (in 18.3%) and the second commonest microorganism in subclinical mastitis samples (17.7%) (Koivula et al., 2007). This implies that measures applied to prevent and control S. aureus IMI in Finnish dairy herds have not been sufficient to eradicate S. aureus. Multiple reservoirs and routes of transmission of S. aureus exist. Nationwide studies of mastitis prevalence, including prevalence of S. aureus IMIs, are few in number and differences are apparent between studies. For example, those regarding selection of samples, culture techniques and microbiological criteria. Based on the results available, S. aureus is one of the most common mastitis pathogens in many countries. In the Norwegian national mastitis survey, which included a random sample from all dairy herds, S. aureus was found in 8.2% of the quarters with bacterial growth (Østerås et al., 2006). The detection limit for S. aureus (100 cfu/ml of milk) was lower than in our survey (500 cfu/ml of milk), and the authors estimated that by restricting the positive findings to samples with > 1500 cfu/ml of S. aureus, the prevalence would have been 4.3% (Østerås et al., 2006). In the Netherlands S. aureus was the most common pathogen, being isolated from 16% of the subclinically infected quarters with milk SCC>250 000 cells/ml (Poelarends et al., 2001). During 1987-1996, S. aureus was reported in 41.2-39.8% of subclinically and in 33.1-26.1% of clinically infected quarters in Swiss cows (Schällibaum, 1999). In Brandenburg, Germany, S. aureus was one of the three most common pathogens after coagulase-negative staphylococci (9.1%) and Corynebacterium bovis (7.3%), being isolated from 5.7% of quarters with subclinical IMI (Tenhagen et al., 2006). In Flanders, Belgium, S. aureus was isolated from 25% of culture-positive milk samples, other staphylococci (non-aureus staphylococci) being predominant (41.1%) (Piepers et al., 2007). Being a transmissible mastitis bacterium, S. aureus represents a risk for mastitis outbreaks occurring in a herd (Smith et al., 1998; Smith et al., 2005a), and it is important to monitor

48

herds where S. aureus frequently causes IMI. S. aureus has been isolated from the majority of dairy herds in many locations. S. aureus was isolated from two-thirds of herds surveyed in Finland (I). In Ontario, Canada, S. aureus-infected cows were detected in 92% of the herds (Kelton et al., 1999). On Prince Edward Island, Canada, S. aureus was recorded in 70% of the herds studied by Keefe et al. (1998). In the same area, at least 74% of the herds probably had at least one cow with S. aureus IMI, as estimated by analysis of three bulk milk cultures taken at weekly intervals (Olde Riekerink et al., 2006). In Lombardia, Italy, S. aureus was less prevalent, being found in bulk tank milk in about 30% of 1431 dairy herds investigated (Piccinini et al., 2003b). The increase of the incidence of S. aureus mastitis decades ago in Finland was suggested to be connected with the introduction of machine milking. The pathogen was typical of small herds housed in stanchion-barns (Myllys, 1995a). In the present survey (I), S. aureus was most frequent in the smallest herds (1-10 cows), but it was also common in larger herds. It was found in 10 of the 12 herds with >30 cows, indicating that it has the potential to cause problems in herds of all sizes. The predisposing factors for S. aureus IMIs can vary according to herd size. In the smallest herds, even a cow with chronic S. aureus mastitis is more likely to stay in the herd; farmers of the small herds may be reluctant to cull cows and tend to use treatments more often than those with larger herds (I). In a Swedish study, changing the housing system from traditional tie-stalls to loose-housing decreased the incidence of clinical mastitis and teat injuries (Hultgren, 2002). Teat injuries have been shown to predispose quarters to IMI with S. aureus (Myllys et al., 1994b). Olde Riekerink et al. (2006) associated S. aureus more with stanchion-barn herds than with loose-housed herds. Based on this literature it is suggested that the problem with S. aureus mastitis may be more typically associated with stanchion-barn housing. In accordance with many studies (Davidson et al., 1961; Matos et al., 1991; Roberson et al., 1994, Roberson et al., 1998; Jørgensen et al., 2005a) S. aureus was common in the environment and body sites of dairy cows, including non-mastitic cows (IV). The majority of the cows in the herds of study IV had the pathogen, but in some herds S. aureus has been unusual in samples other than milk (Smith et al., 2005a). The prevalence of S. aureus in dairy cow body sites can differ markedly even between the herds with the same prevalence of S. aureus IMI (Capurro et al., 2008), and may depend on several herd- and study-associated factors. S. aureus was a frequent finding from the teat skin and colostrum in heifers before calving (IV), in agreement with Roberson et al. (1994). These authors demonstrated that in herds with a high prevalence of S. aureus IMI (>10%) a significantly higher proportion of heifers carried the microorganism in epidermal and mucosal sites, compared with in the low prevalence (<3%) herds. Hock wounds and teat traumas were mostly contaminated by S. aureus (IV), which has been also shown by several authors (Larsen et al., 2000a; Jørgensen et al., 2005a; Capurro et al., 2008). In these sites, the pathogen can be present in high numbers (Myllys et al., 1994b).

7.2. Diversity and distribution of Staphylococcus aureus strain-types

In this thesis S. aureus isolates are differentiated on the basis of three characteristics: penicillin susceptibility (I-IV), PFGE type (II-IV), and genetic profile determined with PCR (III-IV). Although different tests were used for determining penicillin susceptibility, the results can be considered comparable as nitrocefin test results are highly correlated with the presence of the blaZ gene (Haveri et al., 2005; Pitkälä et al. 2007).

49

Using PFGE typing, the isolates were distributed among predominant and sporadic types (II, IV). The diversity among the PFGE types was high (II), in agreement with Joo et al., (2001), Middleton et al., (2002a), and Lim et al., (2004). However, as reported by numerous authors (Akineden et al., 2001; Stephan et al., 2001; Cabral et al., 2004; Mørk et al., 2005; Anderson et al., 2006; Hata et al., 2006; Aires-de-Sousa et al., 2007; Rabello et al., 2007), a few genotypes were responsible for a large proportion of the mastitis cases (II-IV). These predominant mastitis-causing genotypes can be distributed nationwide (Buzzola et al., 2001; Mørk et al., 2005). The differences in sampling methods and variable criteria used for typing likely affect the genotype diversity between studies (Joo et al., 2001; Zadoks et al., 2000; Middleton et al., 2002a; Zadoks et al., 2002). In studies II and III, the clones of some pulsotypes were recovered during several years from numerous herds, being probably epidemiologically unrelated. In that case, MLST would have been useful complementary method for epidemiological analysis. The diversity and characteristics of S. aureus strain types in a herd can be influenced by the herd size, introduction of new strains e.g. by purchased cows (Middleton et al., 2002b; study IV), milking personnel (Larsen et al., 2000a; Zadoks et al., 2002; study IV), and the selective pressure created by management (Sommerhäuser et al., 2003; study IV) and treatment practices (IV). Among S. aureus strains in a farm, those becoming predominant can be better adapted for survival in the hostile circumstances compared with the sporadic strains. In a study on human MRSA, two epidemic strains were demonstrated to survive three months longer and in up to 1000 times higher numbers than sporadic strains (Papakyriacou et al., 2000). In herds where there is effective control of contagious mastitis, several atypical S. aureus strains started to cause mastitis (Sommerhäuser et al., 2003). Those strains were shown to have a low capacity for cow to cow spread and were difficult to eradicate using routine measures to control contagious mastitis in the herds (Sommerhäuser et al., 2003). The possible reasons for predominance of a particular strain in one herd and only sporadic appearance of the same strain in another herd (Peles et al., 2007; study IV) need to be studied further. In individual cows, concomitantly infected quarters and subsequent IMI in the same or new quarters, seem mostly to be due to the same S. aureus strains (II, IV). Repeated isolation of the same S. aureus strain from the same quarter most likely indicated persistence of the original strain because the affected quarter was continuously inflamed as indicated by elevated concentrations of inflammatory indicators in the milk. Myllys et al. (1997) demonstrated persistence of S. aureus in the mammary gland by RAPD-typing isolates obtained from pre-and post-treatment samples. The possibility of reinfection by a strain from teat skin or teat canals cannot be excluded as the strains isolated from the milk of infected quarters were, in most cases, indistinguishable from those present on the teat walls, teat ends and in teat canals (IV). With three exception, the majority of the samples harboured only one S. aureus genotype (II-IV). Young et al. (2001) demonstrated experimentally that individual quarters were colonised by more than one S. aureus strain at the same time. Recently, Smith et al. (2005a) observed multiple strains from single quarters in a natural infection. Human nasal S. aureus carriers have been shown to favour colonisation by one S. aureus strain only. In an experiment, where the carriers were artificially inoculated with a mixture of S. aureus strains including the original strain, the majority of the carriers became positive for their original resident strain (Nouwen et al., 2004). The selection of specific strains or clones involved in antimicrobial resistance like MRSA has been shown in human studies (Larsen et al., 2007). Vintov et al. (2003) demonstrated reduced

50

phage diversity and reduced number of phage groups in the countries with high occurrence of penicillin resistant strains. Associations between PFGE type and antimicrobial susceptibility profile has been recently documented (Cabral et al., 2004; Anderson et al. 2006). These reports are consistent with our findings, where significant differences were observed between pulsotypes and penicillin resistance (II-III). However, the penicillin resistant and penicillin susceptible strains of particular pulsotype were not differentiable by PFGE typing (II-IV). Forty-three virulence gene profiles were determined in S. aureus isolates originating from mastitis infections (III). The genes for haemolysins and lukED were identified in the majority of the isolates, but the other genes were more or less pulsotype-linked (Figure 4). The genes for SAg were more frequent (69% positive) than in studies of Fueyo et al. (2005) (48%) and Smyth et al. (2005) (45.5%) but less prevalent than reported by Srinivasan et al. (2006) (93.6%). Nearly 70% of the 116 isolates harboured at least one gene encoding SAg (III). The genes of SAgs are known to reside in mobile clusters or pathogenicity islands (Zhang et al., 1998; Fitzgerald et al., 2001; Jarraud et al., 2001). In agreement with previous studies (Fueyo et al., 2005; Smyth et al., 2005), the seg, sei, sem, sen, and seo genes, characteristic of the egc cluster (Jarraud et al., 2001), were the most common SAg genes. As the genes of the egc cluster seem to exhibit the widest distribution among different pulsotypes (Fueyo et al., 2005; studies III-IV), horizontal transfer of these genes appears to be common. The variation in the egc genes was higher in study III than reported earlier (Fueyo et al., 2005; Smyth et al., 2005). In agreement with the results of Becker et al. (2004), these genes do not always co-exist and may appear in various combinations in different S. aureus genotypes (III-IV). Isolates carrying sec, sel, and tst of SaPIbov (Fitzgerald et al., 2001) often also have egc and seu (Fueyo et al., 2005; studies III-IV). It is possible that unique pulsotypes have a selective system that accepts integration of only particular genes or gene clusters.

7.3. Resistance of S. aureus to penicillin and other antimicrobials

Antimicrobial susceptibility of mastitis pathogens, including S. aureus, has been studied extensively (Owens et al., 1997; Salmon et al., 1998; Gentilini et al., 2000; Werkentin et al., 2001; Erskine et al., 2002; Gianneechini et al., 2002; Rossitto et al., 2002; Guérin-Faublée et al., 2003; Makovec and Ruegg, 2003; Tenhagen et al., 2006; Østerås et al., 2006). Many countries regularly monitor resistance of animal pathogens, including mastitis-derived S. aureus (DANMAP, FINRES, MARAN, NORMVET, SVARM). There seem to be regional differences in the resistance patterns of S. aureus (Myllys et al., 1998; Salmon et al., 1998; de Oliveira et al., 2000; Orchoa-Zarzosa et al., 2008), but in most countries penicillin resistance is the commonest form of antimicrobial resistance. Based on the results of current survey (I) and previous studies from Argentina, Belgium, Brazil, Finland, Germany, Great Britain (England), Ireland, Portugal, Switzerland, the United States and Uruguay (Devriese et al., 1997; Gentilini et al., 2000; de Oliveira et al., 2000; Erskine et al., 2002; Gianneechini et al., 2002; Makovec and Ruegg, 2003; Tenhagen et al., 2006; Nunes et al., 2007; Rabello et al., 2007), approximately one-third to two-thirds of bovine S. aureus isolates are resistant to penicillin. Markedly lower figures were reported from Denmark, the Netherlands, Norway and Sweden, where only 7-23% of S. aureus isolates were penicillin resistant (SVARM, 2001; SVARM, 2002; DANMAP, 2003; MARAN, 2003; Bengtsson et al., 2005; NORM-VET, 2005; Østerås et al., 2006). Resistance to oxytetracycline is the next commonest form of resistance, the proportion of resistant isolates usually ranging from 5% (I) to 23% (Makovec and Ruegg, 2003), but reaching 58% in Italy (Moroni et al., 2006). In Norway, less than 1% of S. aureus samples isolated from mastitis were resistant to oxytetracycline (NORM-VET,

51

2005), thus the resistance figures from different countries can considerably vary, from very low to very high, probably reflecting the use of antimicrobials in those countries. Antimicrobials available for treating S. aureus mastitis in different countries are likely to affect the panel of the tested substances, which tends to vary among studies. Estimation of resistance to penicillin, for example, is not reliable if only MIC for aminopenicillins and cephalosporins are tested (Tenhagen et al., 2006), without data on β-lactamase production. The methods may not be comparable, as for example correlation between the disk diffusion method and MIC determination by the dilution method was shown to be weak for many microorganims (Kibsey et al., 1994; Kelly et al., 1999). The breakpoints used to categorise veterinary pathogens, including bovine S. aureus, into susceptible or resistant classes should be specific for that use and not be adopted from human microbiology, but specific breakpoints are not available for all substances. The proportion of penicillin resistant, β-lactamase producing S. aureus increased markedly in Finland from 1988 (32.1%) to 1995 (50.7%) and remained almost stable after that (52.1% in 2001; study I). Studies from the US reported a declining proportion of penicillin resistant S. aureus isolates during the last decade, using a disk diffusion method (Erskine et al., 2002; Makovec and Ruegg, 2003). Because the method has its limitations and β-lactamase production of the isolates was not tested, the results are not comparable with ours. In other Nordic countries the proportion of penicillin-resistant S. aureus has been found to be much lower than in Finland, 6.7% for S. aureus originating from clinical mastitis in Sweden (Bengtsson et al., 2005), 11% for isolates from subclinical mastitis in Norway (Østerås et al., 2006) and 23% for mastitis isolates from Denmark (DANMAP, 2003). Explanations for the high prevalence of penicillin resistant S. aureus in Finland are open to discussion. The rapid increase of penicillin resistance of S. aureus in the 1980s in Finland was suggested to be due to the routine use of broad-spectrum antimicrobials in mastitis therapy (Myllys et al., 1998). In Norway and Sweden with lowest figures for penicillin resistance, mainly penicillin G has been used in mastitis therapy (NORM-VET 2005). However, since the 1990s the use of antibiotics in mastitis treatment has decreased markedly in Finland (FINRES-Vet, 2005-2006). Penicillin resistance of mastitis-causing CNS, which may also function as a reservoir for resistance determinants (Couto et al., 1996), has not increased in relation to that of S. aureus (I). No within-farm correlation between penicillin resistance of S. aureus and CNS isolates was found in the survey of 1995 (Myllys et al., 1998). Guidelines for prudent use of antimicrobials in the treatment of animal infections, including bovine mastitis, to control antimicrobial resistance, have been published in Finland (Anon., 2003). In these guidelines it is recommended that treatment of mastitis should be based on testing of the antimicrobial susceptibility of the causal agent. In routine diagnostics this is normally limited to testing of β-lactamase production of staphylococci by a nitrocephin test (Pitkälä et al., 2007). As a possible consequence of changes in the treatment strategies, multiresistance (resistance to more than two antimicrobials) of S. aureus has become less common in Finland (2.0% in the survey in 2001, study I) than reported in the earlier survey of 1995 (3.2%). Resistance to oxytetracycline among S. aureus isolates decreased significantly from 1995, and resistance to first generation cephalosporins was not recorded at all in 2001 (I), as compared with 2% in 1995. In general, resistance of S. aureus to other antimicrobials than penicillin G is at a lower level in Finland as in many other countries (de Oliveira et al., 2000; MARAN, 2003; Moroni et al., 2006)

52

Penicillin resistance in Sweden was higher in S. aureus isolates from subclinical or chronic mastitis (18%) compared with isolates from clinical mastitis (6.7%) (Bengtsson et al., 2005). Consistent findings have been reported in Finland, where only 25% of isolates from clinical mastitis in the national resistance monitoring data (FINRES-Vet, 2005-2006) were resistant to penicillin. This is only half the figure reported for isolates from subclinical mastitis in the survey (I). In mastitis surveys with random sampling from all dairy herds, S. aureus isolates represent subclinical and chronic mastitis and resistance figures can be assumed to be higher than in clinical data. Specific S. aureus strains tend to be more often involved in subclinical or chronic mastitis than in clinical mastitis (II-III) and the resistance figures may vary between strains. It is possible that the use of broad-spectrum antibiotics confers penicillin resistance more than the use of penicillin G only. A link between tetracycline resistance and β-lactamase production was reported (Waage et al., 2002). Exposure of S. aureus to one antibiotic may lead to acquisition of multiple resistance determinants by the strain since the determinants often co-exist in mobile genetic elements, which can transfer as entire elements from one strain to another in a single event. In the long run, multiresistance is considered to be a burden for microbes, including S. aureus, and continuous production of resistance factors confers no additional fitness in an environment where selection pressure from antimicrobials is low (Ala’Aldeen, 2002). Susceptibility can therefore be expected to return when the microbes are no longer exposed to the given antimicrobial. The entire normal microbiota of a dairy cow can be exposed to an antimicrobial during treatment. Frequent or long-term treatments cause selection pressure for resistant skin bacteria to develop (Gustafsson et al., 2003). The sites frequently supporting S. aureus can act as resistance reservoirs (IV). Susceptible strains may acquire resistance determinants (Yazdankhah et al., 2000) from this reservoir. The determinants such as blaZ can spread by mobile genetic elements including the penicillinase plasmid, where this gene often co-exists with sed and sej genes (Zhang et al., 1998). In most Finnish bovine S. aureus strains from mastitis the blaZ gene resides in a plasmid, whereas in Swedish strains the gene is mostly chromosomal (Bagcigil et al., unpublished data). In our studies (III-IV), blaZ was often present together with sed and sej, suggesting presence of the penicillinase plasmid, but as these genes were clustered in specific genetic lineages, horizontal spread of the putative plasmid in the study population was considered to have been uncommon (III-IV). This observation is in agreement with results from a recent Danish study, where horizontal spread of penicillin resistance was indicated to be a rare event (Olsen et al., 2006). This suggests that factors other than just penicillin resistance are involved in the selection of particular strains. Penicillin resistance was associated with particular herds in Norway (Østerås et al., 2006), but the number of herds or their proportion of all herds studied was not reported. High treatment incidence in those herds was suggested to be one reason for resistance, but dominance of specific, penicillin-resistant S. aureus strains caused by events other than selection pressure was also suggested by the authors. In the Finnish survey material, approximately half of the study herds harboured penicillin resistant S. aureus strains (I). Penicillin resistant strains can be distributed throughout the farm and can persist for long periods in the herd (IV). In many countries penicillin resistance limits the treatment options for S. aureus mastitis, and it can be considered to be a therapeutic problem. Several authors reported lower cure rates for mastitis caused by penicillin resistant S. aureus isolates compared with mastitis caused by penicillin susceptible isolates (Ziv and Storper, 1985; Sol et al., 2000; Taponen et al., 2003).

53

However, results from true comparison trials, where the same treatment protocol is used for both groups in a randomized, prospective study, are not available. Two retrospective studies (Sol et al., 1997; Sol et al., 2000) showed that therapy response was lower for mastitis due to β-lactamase producing S. aureus isolates than for mastitis caused by β-lactamase negative S. aureus, although both were treated with antibiotics from groups other than beta-lactams. In the survey material (I), MIC values for oxacillin in 4% of S. aureus isolates indicated resistance, but mecA was not detected (Pitkälä, A. et al., unpublished results). Similar results were reported from Norway and Sweden (NORM-VET; 2005 SVARM, 2005). No mecA positive isolates were found among any S. aureus isolates studied in this thesis. Oxacillin (methicillin) resistant strains may occur de novo by introduction of mecDNA into a susceptible strain (Cookson, 2000), or can be transferred from humans, in which MRSA is more common (Ferry and Etienne, 2007). Spread of such strains in milk of infected dairy cows represents a risk to public health (Juhász-Kaszanyitzky et al., 2007). The emphasis in Finland as in other countries should be in the prevention and control of mastitis caused by penicillin-resistant S. aureus, instead of broadening the selection of antimicrobials used for treatment.

7.4. The role of strain characteristics in manifestation and persistence of Staphylococcus aureus IMI

It is generally accepted that virulence of S. aureus originating from bovine mastitis can differ among strains. The limited diversity of strains causing most mastitis suggests that these strains are better adapted to infect the bovine mammary gland than are the sporadic strains (II). Some strains can rapidly become predominant in a particular herd and cause herd-wide mastitis outbreaks (Smith et al., 1998) or can persist in the same herds (Anderson and Lyman, 2006) or in the same area for years (II). The variable therapy outcome within and between studies raises many questions, for which strain-associated virulence could be one answer (Sol et al., 2000; Taponen et al., 2005; Barkema et al., 2006). Some S. aureus strains, such as Newbould, have been shown to be completely unable to cause IMI or to cause only a very mild clinical response in experimental mastitis models (Schukken et al., 1999). Host response to a particular strain varies between individual cows and stage of lactation, making it complex to assess the severity of infection caused by particular S. aureus strains (II-III). Significant differences were still observed between pulsotypes in respect to severity of clinical signs and milk NAGase activity values which indicate inflammation in the quarter. Pulsotype C typically caused mild signs and the average milk NAGase activity values remained the lowest, whereas response of infection due to pulsotype B strains was significantly stronger, based on the severity of clinical signs and milk NAGase activity values (II). The number of other studies comparing host response in natural IMIs caused by different S. aureus strains is limited, but Zadoks et al. (2000) reported a similar association between clinical severity of mastitis and the PFGE type of the S. aureus strain causing the mastitis. Middleton et al. (2002a) used quarter milk SCC and NAGase activity values to study associations between inflammatory responses and PFGE strain-types derived from 191 cows with S. aureus IMI in eight herds. Using stringent strain definition criteria, the authors established no significant differences between strains for the inflammatory parameters and they concluded that response depends more on the individual host than on the infectious S. aureus strain-type. Supportive findings came from a study where the authors experimentally infected cows with different S. aureus strains. In that study, some cows developed clinical mastitis whereas others did not develop IMI despite receiving similar doses of S. aureus

54

(Middleton, J. R., unpublished data in Middleton et al., 2002a). Recently Fournier et al. (2008) analysed virulence gene patterns and genotypes of S. aureus using PCR amplification of the 16S-23S rRNA intergenic spacer for S. aureus strains derived from field cases of mastitis. The authors found one genotype to be related to high contagiousness and increased pathogenicity whereas the other types caused only sporadic infections. The different staphylococcal virulence factors have been proposed to play a role in different stages of infection, although overlap exists (Projan and Novick, 1997). Factors such as MSCRAMMs may be more important during the colonisation phase compared with factors such as superantigens, which tend to be produced by S. aureus once the infection has established. Some studies indicate the absolute importance of factors such as FnBPs for the establishment of S. aureus infection, as double mutants defective for both of the genes (fnbB and fnbA) encoding FnBPs have been shown to be incapable of adherence in vitro (Greene et al., 1995). To the best of our knowledge, such S. aureus strains have not been detected in natural bovine IMIs. In vivo studies using a murine mastitis model suggested relative virulence for particular virulence factors like HLA, because HLA-production has been correlated with lethality, and non-productive strains have induced mastitis but not death (Bramley et al., 1989). However, the absolute and relative importance of different virulence factors for inducing S. aureus infection has not been demonstrated. Determining that would be a herculean task for a pathogen with tens of putative virulence factors and inordinate numbers of combinations. Unlike the rabies virus, for example, that must infect in order to propagate, S. aureus is not an inherently pathogenic microorganism. The roles of individual virulence factors have therefore been assumed to increase the fitness and survival of the microbe in different environments. Optimal global evaluation of the differences in virulence among S. aureus strains would require epidemiological collection of the strains, standard methods for strain-typing and essential clinical information. The majority of the studies have included only some types of IMI, mostly subclinical (Fueyo et al., 2005), or clinical data or information such as milk SCC are not available. Collection of relevant clinical data from naturally occurring S. aureus mastitis is demanding and early detection of infection is required for correct characterization of the disease severity. Subclinical mastitis easily becomes chronic before it is detected. A particular strain causing mostly subclinical IMI may be easily categorised as a cause of persistent mastitis. This may have occurred for pulsotype C strains in our study (II). One relevant approach to explain virulence differences among mastitis-causing S. aureus isolates would be comparison of virulence determinants between dominant and sporadic genetic lineages. Although the gene profiles of individual strains can vary greatly among lineages, they have been shown to be relatively stable within lineages (Lindsay et al., 2006). The most frequent S. aureus genotypes can be assumed to possess such determinants that enhance their transmissibility and survivality. The genes hla, hld, hlg, lukED (III-IV) and fnbA (IV) were carried by the majority of the isolates. Although these genes may play a central role in pathogenesis of bovine IMI, their presence may not predict strain-specific virulence. By contrast, the occurrence of the fnbB, lukM, sec, sed, sei-seo, seu and tst genes varied markedly among genotypes, some of them being overrepresented in the predominant genotypes, or in genotypes causing a particular type of mastitis. The enhanced colonisation ability of some strains can increase transmission and consequently result in predominance of a specific strain. For both herds studied for this thesis the fnbA and fnbB genes were frequent and usually co-existed in the predominant pulsotypes, but only fnbA

55

was present in the sporadic types (IV). In Herd II in particular, pulsotype D may have benefited from its possibly high potential for adherence and may have therefore become predominant (IV). Increased colonisation capacity in prevalent strains could occur at the expense of virulence (Lipsitch and Moxon, 1997), compared with less common or sporadic strains. We suggest this to be the case for the predominant pulsotype D in Herd II, which carried both fnbA and fnbB but which lacked certain common virulence genes such as hlb, and generally had fewer virulence genes than several other types present in that herd (IV). The hlb gene was not identified as often as the other haemolysin genes in S. aureus strains derived from cases of mastitis (III). It typically co-existed with hla and other haemolysins in pulsotype B strains that were associated with more severe mastitis (III). Concomitant expression of HLA and HLB is considered to promote mutual effects (Dinges, 2000), when the response can be assumed to be stronger. hlb was less common in isolates from subclinical mastitis and was absent from pulsotype C strains in particular, which typically caused subclinical infection (III). In a previous study (Fueyo et al., 2005), hlb was slightly less frequent than other haemolysins and absent from specific S. aureus PFGE types. In contrast, Aarestrup et al. (1999) reported presence of hlb and HLB production to be more common than hla presence and HLA production in Danish bovine mastitis isolates, and hlb and its expression were more frequent among bovine than among human isolates. The data for the mastitis isolates, whether more clinical than subclinical, were not presented. One can also speculate that the hlb negative strains evolved from strains of human origin. From studies of human strains it is known that integration of certain bacteriophages into β-hemolysin inactivates the toxin (Coleman et al., 1991), when hlb could also remain undetected by PCR. These easily transmissible phages, named β-hemolysin (hlb)-converting bacteriophages, possess a group of genes termed the immune evasion cluster (IEC), where sea can co-exist with the genes of the newly discovered immune modulators, chemotaxis inhibitory protein of S. aureus (CHIPS), staphylococcal complement inhibitor (SCIN) (Rooijakkers et al., 2005), and staphylokinase (SAK) gene (van Wamel et al., 2006). In our study, strains from the hlb negative pulsotype C carried sea (III). Virulence determinants such as sec, tst and lukM that have been thought to cause a strong inflammatory response, were unique to or overrepresented in pulsotype B (III). TSST-1 is one of the few staphylococcal toxins where direct causality between toxin production and specific disease, toxic shock syndrome, has been demonstrated (Dinges et al., 2000). tst has been documented to be more prevalent in bovine than in human S. aureus strains (van Leeuwen et al., 2005). Production of SEC and TSST-1 was connected with severe clinical mastitis unresponsive to therapy (Jones and Wieneke, 1986), or with peracute mastitis in particular (Matsunaga et al., 1993). In Norway, 45% of S. aureus isolates from acute, chronic and subclinical bovine mastitis expressed SEC and TSST-1 and the isolates from acute clinical mastitis tended to more often produce enterotoxins and TSST-1 compared with isolates from subclinical and chronic mastitis (Tollersrud et al., 2000b). SEC was suggested to play a role in persistent bovine IMI, as stimulation by SEC caused inversion of the CD4+ T cell/ CD8+ T cell ratio (Ferens et al., 1998). The CD8+ T cell subpopulation down-regulated proliferation of CD4+ T cells, which can impair CD4+ T cell response to the pathogen (Park et al., 1993). Kuroishi et al. (2003) suggested SEC to be more important than TSST-1 in the pathogenesis of bovine IMI because intramammary inoculation by TSST-1, unlike SEC, was not associated with increase in milk SCC and severity of mastitis. Surprisingly, in the herds of study IV, tst and sec were detected in two sporadic genotypes that were not isolated from mastitis.

56

lukM was identified in approximately a quarter of the isolates from mastitis, but was clustered in pulsotype B strains, which typically caused clinical signs (III). LukM-production has been reported to be more common in clinical than in subclinical mastitis isolates (Schuberth et al., 2001). lukM seems to co-exist more often in isolates carrying SAgs than in those lacking SAgs (Fueyo et al., 2005; studies III-IV). In three in vitro assays where biological activities of purified leukotoxins on bovine PMN cells were compared by measuring spreading on solid agar, Ca2+ influx and pore formation ability, the activities for lukM/lukF’-PV were highest (Barrio et al., 2006). Rainard et al. (2003) reported that supernatants from S. aureus isolates positive for lukM were more toxic to bovine PMN cells than supernatants from lukM negative isolates. Recently Rainard (2007) suggested that the toxin interferes in vivo with the neutrophil defence mechanism in ruminant IMI. Based on these reports, LukM can be supposed to be linked more with clinical than subclinical mastitis. As LukM seems to interfere with host defence and was linked to chronic mastitis among strains other than pulsotype B (III), it may be involved in persistent infection. Secreted SAgs are able to hamper host immune response and are therefore presumed to contribute to persistence of bovine IMI (Ferens and Bohach, 2000). Use of antimicrobials in the herds may promote persistence by stimulating toxin production of S. aureus. The presence of β-lactams and fluoroquinolones has been shown in vitro to stimulate HLA production (Ohlsen et al., 1998) and TSST-1 production (Dickgiesser and Wallach, 1987); the same can be expected to occur for other SAgs. The most predictive virulence genes for the persistence of IMI were sed and sej, which were overrepresented in isolates from persistent infections (III). The majority of the predominant, herd-wide disseminated S. aureus strains typically causing recurrent or chronic mastitis in Herd I had these genes (IV). The potential role of sed and sej in promoting development of persistent infection remains open to discussion. Only one study reported in vivo production and effects of SED during mastitis (Tollersrud et al., 2006). In that study, all 17 cows became infected and developed clinical mastitis after experimental intramammary infusion of SED-producing S. aureus strain Mu60. SED was detected in mastitic milk, SED induced proliferation of bovine leukocytes in vitro, and anti-SED antibodies were elevated in serum. The sej gene, encoding SEJ, was identified in 1998 by Zhang et al. (Zhang et al., 1998). Superantigenic or emetic activity of the toxin remains to be demonstrated. SEJ is secreted during the exponential growth phase, whereas maximum production of SED is during transition from the exponential to stationary phase of bacterial growth (Zhang et al., 1998). If SEJ were superantigenic, SED and SEJ might have complementary effects on suppression of the host response. The benefit for mastitis-causing S. aureus to frequently carry egc is unclear. Although most of the egc-encoded toxins exhibit mitotic activity and can act as SAgs in conditions such as food poisoning, their importance has not yet been determined (Becker et al., 2004). Toxins coded for by egc have been associated with staphylococcal scarlet fever and toxic shock syndrome in humans (Jarraud et al., 1999), but have also been negatively correlated with the severity of infection (Ferry et al., 2005). The egc genes were demonstrated to accumulate more in carriage (nasal) strains than in invasive strains isolated from humans (van Belkum et al., 2006). Antibodies are rarely produced against egc-encoded toxins (Holtfreter et al., 2004). It has been suggested that human nostrils tolerate the toxins encoded by egc and that this locus may increase the carriage potential of toxin-positive S. aureus (van Belkum et al., 2006).

57

The sea, seb, see, seh, sek, eta, and etb genes are absent or rare in bovine mastitis isolates (Larsen et al., 2000b; Akineden et al., 2001, Larsen et al., 2002; study III-IV). It can be concluded that these genes play a minor role in the dissemination and pathogenesis of bovine IMI. In contrast, sea has been associated with invasive S. aureus isolates in humans (Peacock et al., 2002).

7.5 Reservoirs and transmission of Staphylococcus aureus

The most important reservoir of S. aureus IMI is the infected mammary gland (Davidson et al., 1961; Roberson et al., 1998) in herds with S. aureus mastitis. The numbers of S. aureus secreted by the infected gland can be very high, exceeding millions of cfu/ml of milk (Kitchen, 1981). Prevalence of S. aureus IMI is considerably higher during lactation than during the dry period, suggesting that transmission of the pathogen occurs mostly via milk from infected quarters. In cases where heifers freshen with S. aureus (Roberson et al., 1994; study IV), multiparous cows develop S. aureus IMI during the dry period, or in herds, where S. aureus mastitis has been successfully controlled (Sommerhäuser et al., 2003), transmission mechanisms other than contact with infected milk are more likely. Extramammary sites and environment can form an important reservoir of S. aureus IMI, and the pathogen may survive in these sites despite control measures. Milking liners were demonstrated to be important vectors of mastitis-causing S. aureus strains (IV). This is in agreement with previous studies (Larsen et al., 2000a; Jørgensen et al., 2005a; Smith et al., 2005a). Transmission of S. aureus from the teat canals and skin of cows that did not have infected udder quarters to milking liners was apparent (IV). In Herd II, in 39 out of 40 PFGE typed cases the liner contamination originated eg. from the skin of.quarters without S. aureus IMI (IV). The numbers of viable S. aureus transmitting from intact skin to milking liners can be expected to be lower than in the mastitic milk, but in damaged skin, the pathogen can be present in large numbers. S. aureus isolated from dairy herds has been suggested to form two distinguishable populations; those adapted to infect the mammary gland and those representing residents of extramammary sites (Fox et al., 1991; Zadoks et al., 2002; Smith et al., 2005b; Barkema et al., 2006). Davidson et al. (1961) reported udders and teats to be the most important reservoirs for mastitis-causing S. aureus, but they also found phage types that persistently colonised these sites without causing IMI, suggesting strain-associated tissue adaptation. The authors also introduced one S. aureus strain into a group of cows and demonstrated the transmission of the strain to the cows, causing skin colonisation and IMI. Subsequently, many studies have reported mastitis-causing strains to be mostly different from those isolated from udder and teat skin (Fox et al., 1991; Zadoks et al., 2002; Smith et al., 2005b), but these findings do not apply to all herds, as shown by our results (IV) and other studies (Middleton et al., 2002b; Jørgensen et al., 2005a). These contradicotry findings may depend on factors like the size and management of the herd, and sampling criteria used in the study. Milk from an infected quarter is considered the primary source of mastitis-causing S. aureus. Sites in frequent contact with milk, such as teat canals, teat skin and milking liners can be expected to be contaminated with strains originating from the milk (Jørgensen et al., 2005a; study IV). As entry of S. aureus to the mammary gland is considered to occur via the teat canal (Anderson, 1983), preceded by teat skin and teat orifice colonisation by the bacteria (Bramley et al., 1979), mastitis-causing S. aureus strains can be expected to be found from these sites prior to IMI. This was directly demonstrated in two of the cows in Herd II, where

58

the same genotype that infected the udder quarter was also isolated from the teat skin prior to IMI (IV). Preparturient heifers can have their udders or colostrum colonised by the same S. aureus strains which cause the majority of IMIs in a herd (Middleton et al., 2002b; study IV), indicating that heifers can function as important reservoirs of S. aureus IMI. In a study by Roberson et al. (1994), almost one-third of new cases of S. aureus IMI in lactating herds were due to heifers that freshened with S. aureus IMI. Isolation of a heifer calf from S. aureus may be impossible as the microbe may transmit due to exposure to vaginal S. aureus during parturition, or possibly via uterine S. aureus infection (Davidson et al., 1961; Eduvie et al., 1984). The strains in heifers that have been raised in the same herd are likely to be similar or genetically related to strains in adult cows, as shown in our study (IV). Preparturient S. aureus colonisation may predispose heifers to S. aureus mastitis after calving (Nickerson et al., 1995). The risk for S. aureus IMI developing at calving in heifers with S. aureus colonisation of the udder area was shown to be 3.4 times as high as in S. aureus-negative heifers (Roberson et al., 1994). The prevalence of heifer mastitis caused by S. aureus varies geographically. In Norway S. aureus was the most common cause (approximately 46% of the isolated bacteria) of clinical mastitis of heifers within two weeks of calving (Waage et al., 1999). In a Danish study, prevalence of S. aureus mastitis in heifers was 1% at parturition, but increased almost up to 10% during the first month after calving (Aarestrup and Jensen, 1997), indicating that heifers became infected after introduction to milking. In this study, hands and the nostrils of farm personnel harboured S. aureus strains that caused bovine IMI. Such findings agree with results of several previous studies (Fox et al., 1995; Roberson et al., 1998; Larsen et al., 2000a; Jørgensen et al., 2005a). Mastitis-derived bovine strains of S. aureus have been shown to be genetically distant from those causing human infections (van Leeuwen et al., 2005; Hata et al., 2008), but individuals, including farm personnel in close contact with animals, can be expected to share the same strains at least temporarily. Human hands can at least act as vectors, but their significance as long-term reservoirs was not the subject of this study. Variation within and between S. aureus clonal lineages originating from different host animals seems to be high (van Leeuwen et al., 2005; Herron-Olson et al., 2007; Ben Zakour et al., 2008; Sung et al., 2008), but strains from bovine S. aureus mastitis appear to carry several unique genes, which may be involved in adaptation of S. aureus to bovine udder. Some other genes such as bap involved in biofilm formation or gene complexes like "immune evasion cluster" may be less important for IMI, as these genes, in contrast to human isolates, are rare in bovine isolates (Monecke et al., 2007; Sung et al., 2008).

7.6 Future considerations for control of S. aureus mastitis

The significance of S. aureus as a cause of mastitis differs among herds, suggesting a herd-level approach. As a whole, most mastitis-causing S. aureus comprise a limited number of clones. However, the number and diversity of infecting S. aureus strains in a particular herd is affected by numerous factors, including the number of reservoirs, routes of transmission, and features that create selection pressure, such as treatment incidences etc. Some clinically important bacterial characteristics such as penicillin resistance appear to vary among isolates within identical clonal types. Clinical consequences can be different for the clonal variants and different mastitis control measures may be required. This suggests that antimicrobial

59

resistance and perhaps certain virulence properties of S. aureus should be tested in addition to molecular typing. Molecular typing can be used to formulate strategies for control of S. aureus mastitis, by identifying the reservoirs and fomites of mastitis-causing S. aureus in a herd. If the diversity of S. aureus strains has been determined for a herd, the information can be used to predict characteristics of the future S. aureus isolates in the same herd. Grinberg et al. (2005) applied information on penicillin-resistance status and genotypes of S. aureus isolates in a herd to predict the resistance status of subsequent isolates on the same farm, supposing the IMIs to be due to genetically related strains. Early detection of S. aureus IMI is critical to limit exposure time of the other quarters of the same cow and other cows to the pathogen, and to achieve satisfactory therapy responses for the cows to be treated. For that purpose, the threshold level for detection of mastitis using CMT (> 300 000/ml), as introduced 30 years ago by Klastrup and Schmidt Madsen (1974), is too high to detect IMI. Not more than about half the quarters with S. aureus IMI in the survey material had milk SCC over this threshold (I). In a more recent study, the SCC of healthy quarters was usually below 50 000 cells/ml (Barkema et al., 1999). Detection of mastitis represents a special challenge in herds milked using automatic systems. These herds are usually large and are loose-housed, where control, eradication and prevention of S. aureus can be difficult. The ability of an automatic milking system to detect mastitis has not been satisfactory (Hovinen et al., 2006), and teat-cleaning and disinfection of milking units between milkings may be inadequate (Hovinen et al. 2005). As milking liners appear to play a central role in transmission of S. aureus, cleaning of milking clusters and liners by backflushing or dipping in disinfectant (followed by thorough rinsing of the liners) has been proposed to eliminate S. aureus from the liners (IDF, 2006). However, S. aureus strains can survive in the liners despite standard post-milking washing procedures (IV). Post-milking teat disinfection, by dipping or spraying, is regarded as the single most important and probably the most effective mastitis control procedure (IDF, 2006). The large numbers of S. aureus on the teat skin despite post-milking teat-dipping suggests that either the dipping does not kill the bacteria or the bacteria are reintroduced from an unknown source. Whist et al. (2006) demonstrated under field conditions that efficacy of teat dips may not be as good as suggested, mainly based on the results of experimental challenge studies. In the light of the present and earlier studies, the prevalence and incidence of S. aureus mastitis can be successfully controlled, but total eradication of S. aureus from a herd may not be realistic. Control programmes should include measures to identify animals at risk to S. aureus IMI and those that enhance the host response. Genetic characteristics of individual cows have been shown to contribute to susceptibility of S. aureus mastitis (Schukken et al., 1994), when host response can be improved via testing and selection of cows. Continuous monitoring of teat condition is required, as traumatised or abraded teats are likely to support S. aureus growth (Zadoks et al., 2001). The role of cows with traumatised or callused teats as S. aureus carriers should be studied. Recent studies on development of vaccines against S. aureus have focused on developing antibodies against specific virulence factors. The ideal factors would be those involved in the adhesion of S. aureus, since many of the genes encode these factors, e.g. fnbA is present in most strains. In experimental models, administration of vaccines to MSCRAMMs has

60

conferred broad, albeit not complete, protection against S. aureus IMIs (Brouillette et al., 2002; Carter and Kerr, 2003). Another approach would be to develop strategies to prevent adherence of S. aureus by specifically blocking the interaction of the pathogen with fibronectin by antagonist ligands, as suggested by Arciola et al. (2005).

61

8. CONCLUSIONS

1. During the past decade, mastitis control measures in Finland have contributed to reduction of mastitis prevalence. The relative proportion of S. aureus positive milk samples has diminished from 17 % in 1995 to 11 % in 2001, but the prevalence of cows infected by S. aureus has remained nearly stable, about 10 %, with 1.3 quarters infected by S. aureus per cow. S. aureus is present in two-thirds of the Finnish dairy herds. The significance of this pathogen differs in the herds, and a herd-level approach is needed for control of S. aureus mastitis. Thorough comparison between herds positive and negative for S. aureus would be helpful to explore herd-level factors exposing to S. aureus mastitis. In the two study herds with problems of S. aureus mastitis, in average 48% of quarters and 85% of cows harboured S. aureus on the teat skin or in the teat canal, indicating the wide spread of the bacterium in these herds.

2. S. aureus isolates collected during 1993-1997 can be divided into five predominant

and 17 sporadic genotypes. The diversity of S. aureus genotypes appears to be wide, but a limited number of types predominate as mastitis-causing agents. Isolates belonging to certain genotypes were detected during several years in the same herds or in the same area. To find out if these isolates share the same ancestor or epidemiological origin would require typing by MLST. The factors encoding haemolysins and leukocidin ED were the most common virulence factors, carried by the majority of S. aureus isolates. Horizontal transfer may be common for egc-encoded superantigens as these were found from several different PFGE types. The genes for other superantigens were PFGE type linked.

3. S. aureus penicillin resistance is very common and widely distributed in Finnish herds.

The blaZ gene coding for penicillin resistance seems to be often linked to sed and sej enterotoxin genes, which could indicate the presence of a penicillinase plasmid in those strains. These genes were present only in certain predominating PFGE types. This may indicate that horizontal spread of the putative penicillinase plasmid has not been common. Resistance to other antimicrobials was limited and no methicillin resistant isolates were found.

4. On the basis of severity of clinical signs and milk NAGase activity, virulence of S.

aureus differs between strains. Strains carrying particularly genes sed, sej, and blaZ were associated with persistent mastitis. One pulsotype causing mild to severe clinical mastitis instead of subclinical mastitis was elimiminated from the udder almost always.

5. The same S. aureus strains which caused intramammary infections were also detected

from the extramammary sites. The same strains frequently resided on healthy teat skin and in teat canals of cows free from intramammary infection. These sites can act as reservoirs of mastitis causing S. aureus, as well as wounds and skin lesions of the cows. Hands of the farm personnel can at least temporarily be contaminated with the same strains. Predominant S. aureus strains had the adhesion determinants fnbA and fnbB more often than the sporadic strains, which may have contributed to the better survival and spread of these strains in the herds. The diversity of S. aureus pulsotypes was higher in the herd which purchased cattle as compared with the closed herd.

62

6. Information on the ecology, epidemiology and characteristics of S. aureus presented in this thesis may be useful for planning control strategies in herds with problems of S. aureus mastitis. Diagnosis to strain level and determination of certain virulence determinants would be necessary to target measures against infections caused by specific strains. This would require molecular techniques feasible also for field use.

63

9. ACKNOWLEDGEMENTS

This work was carried out in the Department of Production Animal Medicine, Faculty of Veterinary Medicine, University of Helsinki. I wish to express my deepest gratitude to: My supervisor Professor Satu Pyörälä, DVM, PhD, for continuous support and guidance. Without Satu’s optimism and encouragement this work would not have been completed. Thank you, Satu, for sharing your vast knowledge with me. My second supervisor, Docent Jaana Vuopio-Varkila, MD, PhD, who provided me excellent research facilites in Hospital Bacteria Laboratory, National Public Health Institute and who guided me in the world of molecular biology. All the people at the Department of Bacteriology, Finnish Food and Safety Authority; especially my co-author Anna Pitkälä, DVM, for introducing me the field of mastitis pathogens; Mrs. Kaisa Rouhiainen for valuable advice and skilful laboratory assistance; Leila Rantala, DI, for guidance and help in PCR work; and Vesa Myllys, DVM, PhD, and Professor Tuula Honkanen-Buzalski for collaboration. The staff of the Department of Production Animal Medicine, especially Mrs. Marja-Liisa Tasanko, Taina Lehto, Merja Pöytäkangas, and Ilkka Saastamoinen for laboratory help; and Mrs. Maarit Mäki for helping me in all kinds of practical things. Saara Salmenlinna, M.Sc., PhD, and Mrs. Elina Siren, and all the personnel in Hospital Bacteria Laboratory, National Public Health Institute, for friedly and excellent guidance. My pre-examiners, Professor Julie Fitzpatrick, DVM, PhD, and Theo Lam, DVM, PhD, for their constructive criticism and valuable comments; Tore Tollersrud, DVM, PhD, for agreeing to serve as my opponent. Hannu Rita, M.Sc., PhD, Ian Dohoo, DVM, PhD, and Arto Ketola, M.Sc., for statistical advices; Dr. Jonathan Robinson for thorough linguistic revision of the manuscript; and Professor Tapani Alatossava for collaboration and reviewing of the article manuscripts. My friends and colleagues; especially Mari, for high contribution to my work as a co-author; Taina for encouraging me to go forward; Anu, for nice coffee breaks and doing enormous laboratory work; all the present and former members of our “Dynamic mastitis research group”, Suvi, Heli, Leena, Paula, Veera, Tanja, and Sami for inspiring collaboration. My veterinary colleagues in the field; thank you for your encouragement in the final step of my work. My parents, Maija-Liisa and Mikael, for taking care, supporting, and helping me and my family in so many ways; my brothers Heikki, Mikko and their families for sharing moments of fun with me and my family.

64

My sister Hanna and her family, especially for scientific interest in my work and countless refreshing and inspiring discussions. Ilkka, my love, and our three little sunshines Viola, Helka and Hilma; thank you for sharing this life with me. I owe my grateful thanks to the Faculty of Veterinary Medicine at the University of Helsinki, The Finnish Ministry of Agriculture and Forestry, Valio Ltd., Walter Ehrström Foundation, Orion-Pharmos, Mercedes Zacchariassen, and The Finnish Veterinary Society for financial support of my work.

65

10. REFERENCES

1. Aarestrup., F., C. Dangler, and L. Sordillo. 1995a. Prevalence and duration of intramammary infections in Danish heifers during the peripartum period. J. Dairy Sci. 80:301-312.

2. Aarestrup, F. M., H. C. Wegener, and V. T. Rosdahl. 1995b. A comparative study of Staphylococcus aureus strains isolated from bovine subclinical mastitis during 1952-1956 and 1992. Acta vet. scand. 36:237-243.

3. Aarestrup, F. M. and N. E. Jensen. 1997. Prevalence and duration of intramammary infection in Danish heifers during the peripartum period. J. Dairy Sci. 80:307-312.

4. Aarestrup, F. M., H. D. Larsen, N. H. Eriksen, C. S. Elsberg, and N. E. Jensen.1999. Frequency of alpha- and beta-haemolysin in Staphylococcus aureus of bovine and human origin. A comparison between pheno- and genotype and variation in phenotypic expression. APMIS 107: 425-430.

5. Abraham, E.P. and E. Chain. 1940. An enzyme from bacteria able to destroy penicillin. Nature 3713:837.

6. Acuna, C. N., R. E. Chertoff, M. B. Martinez, and J. M. Nimo. 2001. Udder pathogens prevalence in dairy cows from Argentina. 40th Ann. Meet. Natl. Mastitis Council. Reno, Nevada, pp. 177-178.

7. Aires-de-Sousa, M., C. E. S. R. Parente, O. Vieira-da-Motta, I. C. F. Bonna, D. A. Silva, and H. de Lencastre. 2007. Characterization of Staphylococcus aureus isolates from buffalo, bovine, ovine, and caprine milk samples collected in Rio de Janeiro State, Brazil. Appl. Environ. Microbiol. 73:3845-3849.

8. Akineden, O., C. Annemuller, A. A. Hassan, C. Lämmler, W. Wolter, and M. Zschöck. 2001. Toxin genes and other characteristics of Staphylococcus aureus isolates from milk of cows with mastitis. Clin. Diagn. Lab. Immunol. 8:959-964.

9. Ala’Aldeen, D. 2002. A nonmultiresistant community MRSA exposes its genome. Lancet 359:1791-1792.

10. Almeida, R. A., K. R. Matthews, E. Cifrian, A. J. Guidry, and S. P. Oliver. 1996. Staphylococcus aureus invasion of bovine mammary epithelial cells. J. Dairy Sci.79:1021-1026.

11. Alouf, J. E. and H. Müller-Alouf. 2003. Staphylococcal and streptococcal superantigens: molecular, biological and clinical aspects. Int. J. Med. Microbiol. 292:429-440.

12. Aly, S. and R. Levit. 1987. Adherence of Staphylococcus aureus to squamous epithelium: role of fibronectin and teichoic acid. Rev. Infect. Dis. 9:341-350.

13. Anderson, J.C. 1983. Veterinary aspects of staphylococci. In: Staphylococci and staphylococcal infections, vol. 1. Easmon, C. S. F. and C. Adlam (eds). London Academic Press, London, pp 193-241.

14. Anderson, K. L. and R. L. Lyman. 2006. Long-term persistence of specific genetic types of mastitis-causing Staphylococcus aureus on three dairies. J. Dairy Sci. 89:4551-4556.

15. Anderson, K. L., R. L. Lyman, S. M. Bodeis-Jones, and D. G. White. 2006. Genetic diversity and antimicrobial susceptibility profiles among mastitis-causing Staphylococcus aureus isolated from bovine milk samples. Am. J. Vet. Res. 67:1185-1191.

16. Annemüller, C., C. Lämmler, and M. Zschöck. 1999. Genotyping of Staphylococcus aureus isolated from bovine mastitis. Vet. Microbiol. 69:217-224.

17. Anonymous, 2003. Recommendations for the use of antimicrobial agents in the treatment of the most significant infectious diseases in animals. Memorandum 3. Ministry of Agriculture and Forestry, Helsinki, Finland, 58 pages.

18. Arciola, C. R., D. Campoccia, S. Gamberini, L. Baldassarri, and L. Montanaro. 2005. Prevalence of cna, fnbA and fnbB adhesin genes among Staphylococcus aureus isolates from orthopedic infections associated to different types of implant. FEMS Microbiol Lett. 246:81-86.

19. Arvidson, S. 2000. Extracellular enzymes. In: Gram-positive pathogens. Fischetti, V. A., R. P. Novick, J. J. Ferreti, D. A. Portnoy, and J. I. Rood, (eds). American Society for Microbiology, Washington, DC, pp. 379-385.

66

20. Bakken, G. 1981. Environment and bovine udder diseases in the loose housing system for dairy cows with reference to relevant data from the cowhouse system. Acta Agric. Scand.31:438-444.

21. Balwit, J. M., P. van Langevelde, J. M. Vann, and R. A. Proctor. 1994. Gentamicin-resistant menadione and hemin auxotrophic Staphylococcus aureus persist within cultured endothelial cells. J. Infect. Dis.170: 1033-1037.

22. Barkema, H. W., H. A. Deluyker, Y. H. Schukken, and T. J. G. M. Lam. 1999. Quarter-milk somatic cell count at calving and the first six milkings after calving. Prev. Vet. Med. 38:1-9.

23. Barkema, H. W., Y. H. Schukken, and R. N. Zadoks. 2006. Invited Review: The role of cow, pathogen, and treatment regimen in the therapeutic success of bovine Staphylococcus aureus mastitis. J. Dairy Sci. 89:1877-1895. Review.

24. Barrett, D. J., A. M. Healy, F. C. Leonard, and M. L. Doherty. 2005. Prevalence of pathogens causing subclinical mastitis in 15 dairy herds in the Republic of Ireland. Irish Veterinary Journal 58:333-337.

25. Barrio, M. B., P. Rainard, and G. Prévost. 2006. LukM/LukF'-PV is the most active Staphylococcus aureus leukotoxin on bovine neutrophils. Microbes Infect. 8:2068-2074.

26. Becker, K., A. W. Friedrich, G. Peters, C. and von Eiff. 2004. Systematic survey on the prevalence of genes coding for staphylococcal enterotoxins SElM, SElO, and SElN. Mol. Nutr. Food Res. 48: 488-495.

27. van Belkum, A., W. van Leeuwen, M. E. Kaufmann, B. Cookson, F. Forey, J. Etienne, R. Goering, F. Tenover, C. Steward, F. O'Brien, W. Grubb, P. Tassios, N. Legakis, A. Morvan, N. El Solh, R. de Ryck, M. Struelens, S. Salmenlinna, J. Vuopio-Varkila, M. Kooistra, A. Talens, W. Witte, and H. Verbrugh. 1998. Assessment of resolution and intercenter reproducibility of results of genotyping Staphylococcus aureus by pulsed-field gel electrophoresis of SmaI macrorestriction fragments: a multicenter study. J. Clin. Microbiol. 3:1653-1659.

28. van Belkum, A., D. C. Melles, S. Snijders, W. B. van Leeuwen, H. F. L. Wertheim, J. L. Nouwen, H. A. Verbrugh, and J. Etienne. 2006. Clonal distribution and differential occurrence of the enterotoxin gene cluster, egc, in carriage- versus bacteremia-associated isolates of Staphylococcus aureus. J. Clin. Microbiol. 44:1555-1557.

29. van Belkum, A., M. Emonts, H. Wertheim, C. de Jongh, J. Nouwen, H. Bartels, A. Cole, P. Hermans, H. Boelens, N. L. Toom, S. Snijders, H. Verbrugh, and W. van Leeuwen. 2007. The role of human innate immune factors in nasal colonization by Staphylococcus aureus. Microbes Infect. 9:1471-1477.

30. Bengtsson, B. 2005. Antibiotic susceptibility of bacteria isolated from clinical and subclinical mastitis, Sweden. In: Report from Nordic Meeting on Mastitis Diagnostics. Uppsala 1-3 October 2003. SVA, Uppsala, Sweden.

31. Bjørland, J., M. Sunde, and S. Waage. 2001. Plasmid-borne smr gene causes resistance to quaternary ammonium compounds in bovine Staphylococcus aureus. J. Clin. Microbiol. 39: 3999–4004.

32. Bonventre, P. F., C. Linneman, L. S. Weckbach, J. L. Staneck, C. R. Buncher, E. Vigdorth, H. Ritz, D. Archer, and B. Smith. 1984. Antibody responses to toxic-shock-syndrome (TSS) toxin by patients with TSS and by healthy staphylococcal carriers. J. Infect. Dis. 150:662-666.

33. Brakstad, O.G., K. Aasbakk, and J. A. Maeland. 1992. Detection of Staphylococcus aureus by polymerase chain reaction amplification of the nuc gene. J. Clin. Microbiol. 30:1654-1660.

34. Bramley, A. J., J. S. King, T. M. Higgs, and F. K. Neave. 1979. Colonization of the bovine teat duct following inoculation with Staphylococcus aureus and Escherichia coli. Br. Vet. J. 135:149.

35. Bramley, A. J., A. H. Patel, M. O’Reilly, R. Foster, and T. J. Foster.1989. Roles of alpha-toxin and beta-toxin in virulence of Staphylococcus aureus for the mouse mammary gland. Infect. Immun. 57:2489-2494.

36. Brouillette, E., P. Lacasse, L. Shkreta, J. Bélanger, G. Grondin, M. S. Diarra, S. Fournier, and B. G. Talbot. 2002. DNA immunization against the clumping factor A (ClfA) of Staphylococcus aureus.

37. Vaccine. 20:2348-2357.

67

38. Brouillette, E., B. G. Talbot, and F. Malouin. 2003. The fibronectin-binding proteins of Staphylococcus aureus may promote mammary gland colonization in a lactating mouse model of mastitis. Infect. Immun. 71:2292-2295.

39. Brouillette E., A. Martinez, B. J. Boyll, N. E. Allen, and F. Malouin. 2004. Persistence of a Staphylococcus aureus small-colony variant under antibiotic pressure in vivo. FEMS Immunol. Med. Microbiol. 41:35-41.

40. Bunce, C., L. Wheeler, and G. Reed. 1992. Murine model of cutaneous infection with gram-positive cocci. Infect. Immun. 60:2636-2640.

41. Buzzola, F. R., L. Quelle, M. I. Gomez, M. Catalano, L. Steele-Moore, D. Berg, E. Gentilini, G. Denamiel, and D. O. Sordelli. 2001. Genotypic analysis of Staphylococcus aureus from milk of dairy cows with mastitis in Argentina. Epidemiol. Infect. 126:445–452.

42. Cabral, K. G., C. Lämmler, M. Zschöck, H. Langoni, M.E. de Sá, C. Victória, and A. da Silva. 2004. Pheno and genotyping of Staphylococcus aureus, isolated from bovine milk samples from São Paulo State, Brazil. Can. J. Microbiol. 50:901-909.

43. Caiazza, N.C. and G. A. O`Toole. 2003. Alpha-toxin is required for biofilm formation by Staphylococcus aureus. J. Bacteriol. 185:3214-3217.

44. Capurro, A., K. Persson Waller, A. Aspàn, H. Ericsson Unnerstad, and Karin Artursson. 2008. Identification of sources of Staphylococcus aureus in herds with mastitis problems. Annu. Meet. Natl. Mastitis Council, Proc., pp. 222-223.

45. Carter, E. W. and D. E. Kerr. 2003. Optimization of DNA-based vaccination in cows using green fluorescent protein and protein A as a prelude to immunization against staphylococcal mastitis. J. Dairy Sci. 86:1177-1186.

46. Chamberlain, N. R., and P. Imanoel. 1996. Genetic regulation of fatty acid modifying enzyme from Staphylococcus aureus. J. Med. Microbiol. 44:125-129.

47. Chavakis, T., K. T. Preissner, and M. Herrmann. 2007. The anti-inflammatory activities of Staphylococcus aureus. Trends Immunol. 28:408-418. Review.

48. Chaves, C. J., L. Tirante, M. Pol., D. Bas, R. Vandoni, and R. Olivieri. 2001. Prevalence of intramammary infections in 74 dairy herds located in Argentina. 40th Annu. Meet. Natl. Mastitis Council, Reno, Nevada, pp. 205-206.

49. Cifrian, E., A. J. Guidry, A. J. Bramley, N. L. Norcross, F. D. Bastilda-Corcuera, and W. W. Marquardt. 1996. Effect of staphylococcal beta-toxin on the cytotoxicity, proliferation and adherence of Staphylococcus aureus to bovine mammary epithelial cells. Vet. Microbiol. 48:187-198.

50. Cifrian, E., A. J. Guidry, C. N. O'Brien, and W. W. Marquardt. 1995. Effect of alpha-toxin and capsular exopolysaccharide on the adherence of Staphylococcus aureus to cultured teat, ductal, and secretory mammary epithelial cells. Res. Vet. Sci. 58:20-25.

51. CLSI. 2002. Clinical and Laboratory Standards Institute standards (Former: National Committee for Clinical Laboratory Standards). 2002a. Performance Standards for Antimicrobial Disk and Dilution Susceptibility Tests for Bacteria Isolated from Animals; Approved Standard. 2nd ed. NCCLS document M31-A2. NCCLS, Wayne, PA.

52. Coleman, D., J. Knights, R. Russell, D. Shanley, T. H. Birkbeck, G. Dougan, and I. Charles. 1991. Insertional inactivation of the Staphylococcus aureus beta-toxin by bacteriophage phi 13 occurs by site- and orientation-specific integration of the phi 13 genome. Mol. Microbiol. 5:933-939.

53. Cookson, B. D. 2000. Methicillin resistant Staphylococcus aureus in the community: new battlefronts, or are the battles lost? Infect. Control Hosp. Epidemiol. 21:398-403.

54. Couto, I., H. de Lencastre, E. Severina, W. Kloos, J. A. Webster, R. J. Hubner, I. S. Sanches, and A. Tomasz. 1996. Ubiquitous presence of a mecA homologue in natural isolates of Staphylococcus sciuri. Microb Drug Resist 2:377-91.

55. DANMAP 2003. DANMAP - Use of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from food animals, foods and humans in Denmark. 2004. Available at: http://www.danmap.org/pdfFiles/Danmap_2003.pdf.

56. Davidson, I. 1961. The epidemiology of staphylococcal mastitis. Vet. Rec. 73:1015-1018. 57. Davidson, I. 1963. Experiments on controlling staphylococcal mastitis. Res. Vet. Sci. 4:64-

76. 58. Dinges, M. M., P. M. Orwin, and P. M. Schlievert. 2000. Exotoxins of Staphylococcus

aureus. Clin. Microbio. Rev. 13:16-34.

68

59. Devriese, L. A., F. Haesebrouck, J. Hommez, and R. Vandermeersch. 1997. A 25-year survey of antibiotic susceptibility testing in Staphylococcus aureus from bovine mastitis in Belgium, with special reference to penicillinase. Vlaams Diergeneeskundig Tijdschrift. 66:170–173.

60. Dickgiesser, N. and U. Wallach.1987. Toxic shock syndrome toxin-1 (TSST-1): influence of its production by subinhibitory antibiotic concentrations. Infection. 15:351-353.

61. Dinges, M. M., P. M. Orwin, and P. M. Schlievert. 2000. Exotoxins of Staphylococcus aureus. Clin. Microbiol. Rev.13:16-34. Review.

62. Dingwell,R. T., K. E. Leslie, P. Sabour, D. Lepp, and J. Pacan. 2006. Influence of the genotype of Staphylococcus aureus, determined by pulsed-field gel electrophoresis, on dry-period elimination of subclinical mastitis in Canadian dairy herds. Can. J. Vet. Res. 70:115-120.

63. Doery, H. M., B. J. Magnusson, J. Gulasekharam, and J. E. Pearson.1965. The properties of phospholipase enzymes in staphylococcal toxins. J. Gen. Microbiol. 40:283-296.

64. Dziewanowska, K., J. M. Patti, C. F. Deobald, K. W. Bayles, W. R. Trumble, and G. A. Bohach. 1999. Fibronectin binding protein and host cell tyrosine kinase are required for internalization of Staphylococcus aureus by epithelial cells. Infect. Immun. 67:4673-4678.

65. von Eiff, C., A. W. Friedrich, G. Peters, and K. Becker. 2004. Prevalence of genes encoding for members of the staphylococcal leukotoxin family among clinical isolates of Staphylococcus aureus. Diagn. Microb. Inf. Dis. 49:157-162.

66. Eduvie, L.O., D. I. K. Osori, P. B. Addo, and C. O. Njoku. 1984. Bacteriological investigation of the postpartum uterus: Relationship to involution and histopathological findings. Therio. 21:733.

67. Erskine, R. J., R. D. Walker, C. A. Bolin, P. C. Bartlett, and D. G. White. 2002. Trends in antibacterial susceptibility of mastitis pathogens during a seven-year period. J. Dairy Sci. 85:1111–1118.

68. Feil, E. J., J. E. Cooper, H. Grundmann, D. A. Robinson, M. C. Enright, T. Berendt, S. J. Peacock, J. M. Smith, M. Murphy, B. G. Spratt, C. E. Moore, and N. P. Day. 2003. How clonal is Staphylococcus aureus? J. Bacteriol. 185:3307-3316.

69. Ferens, W.A., W. C. Davis, M.J. Hamilton, Y. H. Park YH, C. F. Deobald, L. Fox, and G. Bohach. 1998. Activation of bovine lymphocyte subpopulations by staphylococcal enterotoxin C. Infect Immun. 66:573-80.

70. Ferens, W.A. and G. A. Bohach. 2000. Persistence of Staphylococcus aureus on mucosal membranes: superantigens and internalization by host cells. Review. J. Lab. Clin. Med. 135:225-230.

71. Ferry, T., Thomas, D., Genestier, A. L., Bes, M., Lina, G, Vandenesch, F., and Etienne, J. 2005. Comparative prevalence of superantigen genes in Staphylococcus aureus isolates causing sepsis with or without septic shock. Clin. Infect. Dis. 41:771-777.

72. Ferry, T. and J. Etienne. 2007. Community acquired MRSA in Europe. BMJ. 335:947-948. 73. FINRES-Vet 2005-2006. Finnish veterinary antimicrobial resistance monitoring and

consumption of antimicrobial agents. Available at: http://www.evira.fi/uploads/WebShopFiles/1198141211941.pdf.

74. Fitzgerald, J. R., W. J. Meaney, P. J. Hartigan, C. J. Smyth, and V. Kapur. 1997. Fine-structure molecular epidemiological analysis of Staphylococcus aureus recovered from cows. Epidemiol. Infect. 119:261-269.

75. Fitzgerald, J. R., S. R. Monday, T. J. Foster, G. A. Bohach, P. J. Hartigan, W. J. Meaney, and C. J. Smyth. 2001. Characterization of a putative pathogenicity island from bovine Staphylococcus aureus encoding multiple superantigens. J. Bacteriol. 183:63-70. Erratum in: J. Bacteriol. 166:following 4259.

76. Fitzgerald, J. R. and J. M. Musser. 2001. Evolutionary genomics of pathogenic bacteria. Trends Microbiol. 9:547-553.

77. Fleischer, B. and H. Schrezenmeier. 1988. T cell stimulation by staphylococcal enterotoxins. Clonally variable response and requirement for major histocompatibility complex class II molecules on accessory or target cells.

78. Foster, T. J. and M. Höök. 1998. Surface protein adhesins of Staphylococcus aureus. Trends Microbiol. 6:484-8. Review.

69

79. Fournier, J. M. 1990.Capsular polysaccharides of Staphylococcus aureus. p. 533. In Wädström T., Eliasson I., Holder I, and A Ljungh (eds): Pathogenesis of wound and biomaterial-associated infections. Springer-Verlag, London, UK.

80. Fournier, B. 2008. Global regulators of Staphylococcus aureus virulence genes. In: Staphylococcus molecular genetics. Lindsay, J. A. (ed), Caister Academic Press, Norfolk, pp.131-183.

81. Fournier, C., P. Kuhnert, J. Frey, R. Miserez, M. Kirchhofer, T. Kaufmann, A. Steiner, and H. U. Graber. 2008. Bovine Staphylococcus aureus: Association of virulence genes, genotypes and clinical outcome. Res. Vet. Sci. 2008 Mar 19. Epub ahead of print.

82. Fox, L. K., M. Gershman, D. D. Hancock, and C. T. Hutton. 1991. Fomites and reservoirs of Staphylococcus aureus causing intramammary infections as determined by phage typing – the effect of milking time hygiene practices. Cornell vet. 81:183-193.

83. Fox, L.K., S. T. Chester, S. C. Nickerson, J. W. Pankey, and L. D. Weaver. 1995. Survey of intramammary infections in dairy heifers at breeding age and first parturition. J. Dairy Sci. 78:1619-1628.

84. Frost, A. J., D. D. Wanasinghe, and J. B. Woolcock. 1977. Some factors affecting selective adherence of microorgansims in the bovine mammary gland. Infect. Immun. 15:245-253.

85. Fueyo, J. M., M. C. Mendoza, M. R. Rodicio, J. Muniz, M. A. Alvarez, and M. C. Martin. 2005. Cytotoxin and pyrogenic toxin superantigen profiles of Staphylococcus aureus associated with subclinical mastitis in dairy cows and relationships with macrorestriction genomic profiles. J. Clin. Microbiol. 43:1278-1284.

86. Gentilini, E., G. Denamiel, P. Llorente, S. Godaly, M. Rebuelto, and O. DeGregorio. 2000. Antimicrobial susceptibility of Staphylococcus aureus isolated from bovine mastitis in Argentina. J. Dairy Sci. 83:1224–1227.

87. Gianneechini, R. E., C. Concha, and A. Franklin. 2002. Antimicrobial susceptibility of udder pathogens isolated from dairy herds in the West Littoral region of Uruguay. Acta vet. Scand. 43:31–41.

88. Goldberg, J. J., J. Pankey, P. A. Drechsler, P. A. Murdough, and D. B. Howard. 1991. An update survey of bulk milk quality in Vermont. J. Food Protect. 54:549-553.

89. Gómez, M. I., A. Lee, B. Reddy, A. Muir, G. Soong, A. Pitt, A. Cheung, and A. Prince. 2004. Staphylococcus aureus protein A induces airway epithelial inflammatory responses by activating TNFR1. Nat Med. 10:842-848.

90. Gonzalez, R.N., D.E. Jasper, T.B. Farver, R.B. Bushnell, and C.E. Franti. 1988. Prevalence of udder infections and mastitis in 50 California dairy herds. JAVMA 193: 323-328.

91. Goodyear, C. S. and G. J. Silverman. 2004. Staphylococcal toxin induced preferential and prolonged in vivo deletion of innate-like B lymphocytes. Proc. Natl. Acad. Sci. USA. 101:11392-11397.

92. Greene, C., D. McDevitt, and P. Francois. 1995. Adhesion properties of mutants of Staphylococcus aureus defective in fibronectin-binding proteins and studies on the expression of fnb genes.

93. Mol. Microbiol.17:1143-1152. 94. Grinberg, A., A. Hittman, M. Leyland, L. Rogers, and B. le Quensne. 2005. Epidemiological

and molecular evidence of a monophyletic infection with Staphylococcus aureus causing a purulent dermatitis in a dairy farmer and multiple cases of mastitis in his cows. Epidemiol. Inf.132:507-513.

95. Gröhn, Y., D. J. Wilson, R. N. Gonzalez, J. A. Hertl, H. Schulte, G. Bennett, and Y. H. Schukken. 2004. Effect of pathogen-specific clinical mastitis on milk yield in dairy cows. J. Dairy Sci. 87:3358-3374.

96. Gustafsson, I., M. Sjölund, E. Torell, M. Johannesson, L. Engstrand, O. Cars, and D. I. Andersson. 2003. Bacteria with increased mutation frequency and antibiotic resistance are enriched in the commensal flora of patients with high antibiotic usage. J. Antimicrob. Chemother. 52:645-650.

97. Guidry, A., A. Fattom, A. Patel, C. O’Brien, S. Shepherd, and J. Lohuis. 1998. Serotyping scheme for Staphylococcus aureus isolated from cows with mastitis. Am. J. Vet. Res. 59:1537-1539.

70

98. Hata, E., K. Katsuda, H. Kobayashi, T. Ogawa, T. Endo, and M. Eguchi. 2006. Characteristics and epidemiologic genotyping of Staphylococcus aureus isolates from bovine mastitic milk in Hokkaido, Japan. J. Vet. Med. Sci. 68:165.

99. Hata, E., K. Katsuda, H. Kobayashi, K. Nishimori, I. Uchida, M. Higashide, E. Ishikawa, T. Sasaki, and M. Eguchi. 2008. Bacteriological characteristics of Staphylococcus aureus isolates from humans and bulk milk. J. Dairy Sci. 91:564-569.

100. Haveri, M., S. Suominen, L. Rantala, T. Honkanen-Buzalski, and S. Pyörälä. 2005. Comparison of phenotypic and genotypic detection of penicillin G resistance of Staphylococcus aureus isolated from bovine intramammary infection. Vet. Microbiol.106:97-102.

101. Heilmann, C., S. Niemann, B. Sinha, M. Herrmann, B. E. Kehrel, and G. Peters. 2004. Staphylococcus aureus fibronectin-binding protein (FnBP)-mediated adherence to platelets, and aggregation of platelets induced by FnBPA but not by FnBPB. J. Infect. Dis. 190:321-329.

102. Herron-Olson, L., J. R. Fitzgerald, J. M. Musser, and V. Kapur. 2007. Molecular correlates of host specialization in Staphylococcus aureus. PLoS ONE. 2:e1120.

103. Hoblet, K. H. and G. Y. Miller. 1991. Use of partial budgeting to determine the economis outcome of Staphylococcus aureus intramammary infection reduction strategies in three Ohio dairy herds. JAVMA 199:714-720.

104. Hogan, J. 1997. Occurrence of clinical and subclinical environmental mastitis. Udder health management for environmental streptococci. Guelph, ON 22/06/97:36-41.

105. Hogeveen, H. and O. Østerås. 2005. Mastitis management in an economic framework. In Hogeveen, H. (ed), Proc. 4th Int. Mastitis Conf., Maastrich, the Netherlands, pp. 41-52.

106. Holden, M. T. G. and J. A. Lindsay, J. 2008. Whole genomes: sequence, microarray and systems biology. In: Staphylococcus molecular genetics. Lindsay, J. A. (ed), Caister Academic Press, Norfolk, pp. 1-28.

107. Holtfreter, S., K. Bauer, D. Thomas, C. Feig, V. Lorenz, K. Roschanck, E. Friebe, K. Selleng, S. Lovenich, T. Greve, A. Greinacher, B. Panzig, S. Engelmann, G. Lina, and B. M. Broker. 2004. egc-encoded superantigens from Staphylococcus aureus are neutralized by human sera much less efficiently than are classical staphylococcal enterotoxins or toxic shock syndrome toxin. Infect. Immun. 72:4061-4071.

108. Honkanen-Buzalski, T., 1995. Sampling technique, transportation and history. In: The Bovine Udder and Mastitis. Sandholm, M., T. Honkanen-Buzalski, L. Kaartinen, and S. Pyörälä (eds.), Gummerus Press, Jyväskylä, Finland, pp. 111-114.

109. Horsburgh, M. J. 2008. The response of Staphylococcus aureus to environmental stimuli. Staphylococcus molecular genetics. Lindsay, J. A. (ed), Caister Academic Press, Norfolk, pp.185-206.

110. Hovinen, M., A.-M. Aisla, and S. Pyörälä. 2005.Visual detection of technical success and effectiveness of teat cleaning in two automatic milking systems J. Dairy Sci. 88:3354-3362.

111. Hovinen, M. A.-M. Aisla, and S. Pyörälä. 2006. Accuracy and reliability of mastitis detection with electrical conductivity and milk colour measurement in automatic milking. Acta Agric. Scand., Section A-Animal Sciences, 56:121-127.

112. Hultgren, J. 2002. Foot/leg and udder health in relation to housing changes in Swedish dairy herds. Prev. Vet. Med. 53:167–189.

113. IDF (International Dairy Federation). 1981. Laboratory methods for use in mastitis work. IDF Bulletin 132/1981. IDF, Geneve, Switzerland.

114. IDF (International Dairy Federation). 2006. Staphylococcus aureus intramammary infections. Zecconi, A., L. Calvinho, and L. K. Fox (eds). IDF bulletin 408/2006, pp.21-22.

115. IDF (International Dairy Federation). 1997. Recommendations for presentation of mastitis-related data. Smith, L. K. (ed). IDF Bulletin 321/1997. IDF, Geneve, Switzerland.

116. Jarraud, S., M. A. Peyrat, A. Lim, A. Tristan, M. Bes, J. Etienne, and G. Lina. 1999. Involvement of enterotoxins G and I in staphylococcal toxic shock syndrome and staphylococcal scarlet fever. J. Clin. Microbiol. 37:2446-2449.

117. Jarraud, S., M. A. Peyrat, A. Lim, A. Tristan, M. Bes, C. Mougel, J. Etienne, F. Vandenesch, M. Bonneville, and G. Lina. 2001. egc, a highly prevalent operon of enterotoxin gene, forms a putative nursery of superantigens in Staphylococcus aureus. J. Immunol. 166:669-677.

71

118. Jarraud, S., C. Mougel, J. Thioulouse, G. Lina, H. Meugnier, F. Forey, X. Nesme, J. Etienne, and F. Vandenesch. 2002. Relationships between Staphylococcus aureus genetic background, virulence factors, agr groups (alleles), and human disease. Infect. Immun. 70:631-641.

119. Jones, R. N. 2008. Key considerations in the treatment of complicated staphylococcal infections. Clin. Microbiol. Infect.14 (Suppl 2):3-9. Review.

120. Jones, T. O. and A. A. Wieneke. 1986. Staphylococcal toxic shock syndrome. Vet. Rec. 119:435-436.

121. Joo, Y. S., L. K. Fox, W. C. Davis, G. A. Bohach, and Y. H. Park. 2001. Staphylococcus aureus associated with mammary glands of cows: genotyping to distinguish different strains among herds.Vet. Microbiol. 80:131-138.

122. Jørgensen, H. J., T. Mørk, T., and L. M. Rørvik. 2005a. The occurrence of Staphylococcus aureus on a farm with small-scale production of raw milk cheese. J. Dairy Sci. 8:3810-3817.

123. Jørgensen, H. J., T. Mørk, D. A. Caugant, A. Kearns, and L. M. Rørvik. 2005b.Genetic variation among Staphylococcus aureus strains from Norwegian bulk milk. Appl. Environ. Microbiol. 71:8352-8361.

124. Juhász-Kaszanyitzky, É., S. Jánosi, P. Somogyi, Á. Dán, L. van der Graaf-van Bloois, E. van Duijkeren, and J. A. Wagenaar. 2007. MRSA Transmission between cows and humans. Emerg. Infect. Dis. 13:630-632.

125. Kamio, Y., A. Rahman, H. Nariva, T. Ozawa, and K. Izaki. 1993. The two staphylococcal bi-component toxins, leukocidin and gamma-hemolysin, share one component in common. FEBS Lett. 321:15-18.

126. Kapur, V., W. M. Sischo, R. S. Greer, T. S. Whittam, and J. M. Musser. 1995. Molecular population genetic analysis of Staphylococcus aureus recovered from cows. J. Clin. Microbiol. 33:376-380.

127. Keefe, G. K., J. van Leeuwen, and J. Wichtel. 1998. Evaluation of extended therapy with pirlimycin hydrochloride (Pirsue ®) for the elimination of existing Staphylococcus aureus infection during lactation (interim results). Proc. Atlantic. Vet. Coll. Contin. Educ. Fall. Conf.

128. Kelly, L. M., R. M. Jacobs, and P. C. Appelbaum. 1999. Comparison of agar dilution, microdilution, E-test, and disk diffusion methods for testing activity of cefditoren against Streptococcus pneumoniae. J. Clin. Microbiol. 37:3296–3299.

129. Kelton, D., D. Alves, N. Smart, A. Godkin, and P. Darden. 1999. Bulk tank culture for major contagious mastitis pathogens — lessons from the Sentinel herds. Proc. 38th Annu. Meet. Natl. Mastitis Counc., pp. 144–145.

130. Kibsey, P. C., R. P. Rennie, and J. E. Rushton. 1994. Disk diffusion versus broth microdilution susceptibility testing of Haemophilus species and Moraxella catarrhalis using seven oral antimicrobial agents: Application of updated susceptibility guidelines of the National Committee for Clinical Laboratory Standards. J. Clin. Microbiol. 32:2786–2790.

131. Kitchen, B. J. 1981. Review of the progress of dairy science: bovine mastitis: milk compositional changes and related diagnostic tests. J. Dairy Res. 48:167-88. Review.

132. Klastrup, O., and P. Schmidt Madsen. 1974. Nordiske rekommendationer vedrøende mastitundesøgelser af kiertelprøver. Nord. Vet. Med. 26:197–204.

133. Kloos, W. 1997. Taxonomy and systematics of staphylococci indigenous to humans. The molecular basis of pathogenesis. In: The staphylococci in human disease. Crossley, K. B. and G. L. Archer (eds.). Churchill Livingstone, N. Y., pp. 113-137.

134. Kluytmans, J., A. van Belkum, and H. Verbrugh. 1997. Nasal carriage of Staphylococcus aureus: epidemiology, underlying mechanisms, and associated risks. Clin. Microbiol. Rev., Jul 1997, 505-520, Vol 10, No. 3.

135. Koivula, M., A. Pitkälä, S. Pyörälä, and E. Mäntysaari. 2007. Distribution of bacteria and seasonal and regional effects in a new database for mastitis pathogens in Finland. Acta Agric. Scand. A. In press.

136. Kuroishi, T., K. Komine, K. Kai, M. Itagaki, J. Kobayashi, M. Ohta, S. Kamata, and K. Kumagai. 2003. Concentrations and specific antibodies to staphylococcal enterotoxin-C and toxic shock syndrome toxin-1 in bovine mammary gland secretions, and inflammatory response to the intramammary inoculation of these toxins. J. Vet. Med. Sci. 65:899-906.

72

137. König, B., G. Prévost, Y. Piémont, and W. König. 1995. Effects of Staphylococcus aureus leukocidins on inflammatory mediator release from human granulocytes. J. Infect. Dis. 171:607-613.

138. Lammers, A., P. J. Nuijten, and H. E. Smith.1999. The fibronectin binding proteins of Staphylococcus aureus are required for adhesion to and invasion of bovine mammary gland cells. FEMS Microbiol. Lett. 180:103-109.

139. Lancette, G. A. and S. R. Tatini. 1992. Staphylococcus aureus. In: Compendium of methods for the microbiological examination of foods. Vanderzant, C. and D. F. Splittstoesser (eds.). American Public Health Association, Washington, DC, pp. 533-550.

140. Larsen, H. D., K. H. Sloth, C. Elsberg, C. Enevoldsen, L. H. Pedersen, N. H. Eriksen, F. M. Aarestrup, and N. E. Jensen. 2000a. The dynamics of Staphylococcus aureus intramammary infection in nine Danish dairy herds. Vet. Microbiol. 71:89-101.

141. Larsen, H. D., A. Huda, N. H. R. Eriksen, and N. E. Jensen. 2000b. Differences between Danish bovine and human Staphylococcus aureus isolates in possession of superantigens. Vet. Microbiol. 76:153-162.

142. Larsen, H. D., F. M. Aarestrup, and N. E. Jensen. 2002. Geographical variation in the presence of genes encoding superantigenic exotoxins and beta-hemolysin among Staphylococcus aureus isolated from bovine mastitis in Europe and USA.Vet. Microbiol. 85:61-67.

143. Larsen, A., M. Stegger, R. Goering, M. Sørum, and R. Skov. 2007. Emergence and dissemination of the methicillin resistant Staphylococcus aureus USA300 clone in Denmark (2000-2005). Euro Surveill. 20;12.Epub ahead of print.

144. Lee, J. C., S. Xu, A. Albus, and P. J. Livolsi. 1994. Genetic analysis of type 5 capsular polysaccharide expression by Staphylococcus aureus. J. Bacteriol. 176:4883-4889.

145. van Leeuwen, W., H. Verbrugh, J. van der Velden, N. van Leeuwen, M. Heck, and A. van Belkum. 1999. Validation of binary typing for Staphylococcus aureus strains. J. Clin. Microbiol. 37:664-674.

146. van Leeuwen, W., Melles, D. C., Alaidan, A., Al-Ahdal, M., Boelens, H. A. M., Snijders, S. V., Wertheim, H., van Duijkeren, E., Peeters, J. K., van der Spek, P. J., Gorkink, R., Simons, G., Verbrugh, H. A., and van Belkum, A. 2005. Host- and tissue-specific pathogenic traits of Staphylococcus aureus. J. Bacteriol. 187:4584-4591.

147. Leslie, K. E., and R. T. Dingwell. 2003. Background to dry cow therapy: What, where, why—Is it still relevant? Pages 5–17 in 42nd National Mastitis Council Annual Meeting, Fort Worth, Texas. Natl. Mastitis Counc., Madison, WI.

148. Letertre, C., S. Perelle, F. Dilasser, and P. Fach. 2003. Identification of a new putative enterotoxin SEU encoded by the egc cluster of Staphylococcus aureus. J. Appl. Microbiol. 95:38-43.

149. Lim, S., Y. Joo, J. Moon, A. Lee, H. Nam, S. Wee, and H. Koh. 2004. Molecular typing of enterotoxigenic Staphylococcus aureus isolated from bovine mastitis in Korea. J. Vet. Med. Sci. 66:581-584.

150. Lindsay, J. A., C. E. Moore, N. P. Day, S. J. Peacock, A. A. Witney, R. A. Stabler, S. E. Husain, P. D. Butcher, and J. Hinds. 2006. Microarrays reveal that each of the ten dominant lineages of Staphylococcus aureus has a unique combination of surface-associated and regulatory genes. J. Bacteriol.188: 669–676.

151. Lina, G., Y. Piemont, F. Godail-Gamot, M. Bes, M. O. Peter, V. Gauduchon, F. Vandenesch, and J. Etienne. 1999. Involvement of Panton-Valentine leukocidin producing Staphylococcus aureus in primary skin infections and pneumonia. Clin. Infect. Dis. 29:1128-1132.

152. Lipsitch, M. and E. R. Moxon. 1997. Virulence and transmissibility of pathogens: what is the relationship? Trends Microbiol 5:31-37.

153. Lowy, F. D. 2000. Is Staphylococcus aureus an intracellular pathogen? Trends Microbiol. 8:341-343. Review.

154. Makovec, J. A., and P. L. Ruegg. 2003. Antimicrobial resistance of bacteria isolated from dairy cow milk samples submitted for bacterial culture: 8905 samples (1994–2001). J. Am. Vet. Med. Assoc. 222:1582–1589.

155. Mamo, W. and G. Fröman. 1994. Adhesion of Staphylococcus aureus to bovine mammary epithelial cells induced by growth in milk whey. Microbiol. Immunol. 38: 305-308.

73

156. Mamo, W., G. Fröman, and T. Wadström. 1988. Interaction of subepithelial connective tissue components with Staphylococcus aureus and coagulase negative staphylococci from bovine mastitis. Vet. Microbiol. 18:163-176.

157. MARAN 2003. Monitoring of antimicrobial resistance and antibiotic usage in animals in the Netherlands. Available at: http://www.cidc-lelystad.wur.nl/UK/publications/

158. Matos, J. S., D. G. White, R. J. Harmon, and B. E. Langlois. 1991. Isolation of Staphylococcus aureus from sites other than the lactating mammary gland. J. Dairy Sci. 74:1544-1549.

159. Mattila, T. and M. Sandholm. 1986. Milk plasmin, N-acetyl-beta-D-glucosaminidase, and antitrypsin as determinants of bacterial replication rates in whey. J. Dairy Sci. 69:670-675.

160. Matsunaga, T., S. Kamata, N. Kakiichi, and K. Uchida. 1993. Characteristics of Staphylococcus aureus isolated from peracute, acute and chronic bovine mastitis. J. Vet. Med. Sci. 55:297-300.

161. McDougall, S., K. E. Agnew, R. Cursons, X. X. Hou, and C. R. W. Compton. 2007. Parenteral treatment of clinical mastitis with tylosin base or penethamate hydriodide in dairy cattle.

162. J. Dairy Sci. 90:779-789. 163. Mehrotra, M., G. Wang, and W. Johnson. 2000. Multiplex PCR for detection of genes for

Staphylococcus aureus enterotoxins, exfoliative toxins, toxic shock syndrome toxin 1, and methicillin resistance. J. Clin. Microbiol. 38:1032-1035.

164. Menzies, B. E. and I. Kourteva. 1998. Internalization of Staphylococcus aureus by endothelial cells induces apoptosis. Infect. Immun. 66:5994-5998.

165. Middleton, J., L. Fox, J. Gay, J. Tyler, and T. Besser. 2002a. Use of pulsed-field gel electrophoresis for detecting differences in Staphylococcus aureus strain populations between dairy herds with different cattle importation practices. Epidemiol. Infect. 129:387-395.

166. Middleton, J. R., L. K. Fox, J. M. Gay, J. W. Tyler, and T. E. Besser. 2002b. Influence of Staphylococcus aureus strain-type on mammary quarter milk somatic cell count and milk N-acetyl-beta-D-glucosaminidase activity in cattle from eight dairies. J. Dairy Sci. 85:1133-1140.

167. Monecke, S., P. Kuhnert, H. Hotzel, P. Slickers, and R. Ehricht. 2007. Microarray based study on virulence-associated genes and resistance determinants of Staphylococcus aureus isolates from cattle. Vet. Microbiol. 125:128-140.

168. Moon, J. S., A. R. Lee, H. M. Kang, E. S. Lee, M. N. Kim, Y. H. Paik, Y. H. Park, Y. S. Joo, and H. C. Koo. 2007. Phenotypic and genetic antibiogram of methicillin resistant staphylococci isolated from bovine mastitis in Korea. J. Dairy Sci. 90:1176-1185.

169. Moroni, P., G. Pisoni, M. Antonini, R. Villa, B. Boettcher, and S. Carli. 2006. Short communication: Antimicrobial drug susceptibility of Staphylococcus aureus from subclinical mastitis in Italy. J. Dairy Sci. 89:2973-2976.

170. Mullarky, I. K., C. Su, N. Frieze, Y. H. Park, and L. M. Sordillo. 2001. Staphylococcus aureus agr genotypes with enterotoxin production capabilities can resist neutrophil bactericidal activity. Infect. Immun. 69:45-51.

171. Murchan, S., M. E. Kauffmann, A. Deplano, R. de Ryck, M. Struelens, C. E. Zinn, V. Fussing, S. Salmenlinna, J. Vuopio-Varkila, N. El Soth, C. Cuny, W. Witte, P. T. Tassios, N. Legakis, W. van Leeuwen, A. van Belkum, A. Vindel, I. Laconcha, J. Garaizar, S. Häggman, B. Olsson-Liljequist, U. Ramsjö, G. Coombes, and B. Cookson. 2003. Harmonization of pulsed-field gel electrophoresis protocols for epidemiological typing of Staphylococcus aureus: a single approach developed by consensus in 10 European laboratories and its application for tracing the spread of related strains. J. Clin. Microbiol. 41: 1574-1585.

172. Myllys, V., T. Honkanen-Buzalski, P. Huovinen, M. Sandholm, and E. Nurmi. 1994a. Association of changes in the bacterial ecology of bovine mastitis with changes in the use of milking machines and antibacterial drugs. Acta vet. scand. 35:363-369.

173. Myllys, V., T. Honkanen-Buzalski, H. Virtanen, S. Pyörälä, and H.-P. Müller. 1994b. Effect of abrasion of teat orifice epithelium on development of bovine staphylococcal mastitis. J. Dairy Sci. 77:446-452.

74

174. Myllys, V. 1995a. Staphylococcal mastitis in heifers and dairy cows. Ph. D. Thesis. National Veterinary Food Reearch Institute and Department of Clinical Veterinary Medicine. Helsinki. ISBN 952-90-6288-5.

175. Myllys, V. 1995b. Staphylococcal mastitis in heifers before and after parturition. J. Dairy Res. 62:51-60.

176. Myllys, V., J. Ridell, J. Björkroth, I. Biese, and S Pyörälä. 1997. Persistence in bovine mastitis of Staphylococcus aureus clones as assessed by random amplified polymorphic DNA analysis, ribotyping and biotyping. Vet. Microbiol. 57:245-251.

177. Myllys, V., K. Asplund, E. Brofeldt, V. Hirvela¨-Koski, T. Honkanen-Buzalski, J. Junttila, L. Kulkas, O. Myllykangas, M. Niskanen, H. Saloniemi, M. Sandholm, and T. Saranpää. 1998. Bovine mastitis in Finland in 1988 and 1995-Changes in prevalence and antimicrobial resistance. Acta Vet. Scand. 39:119–126.

178. Mørk, T., T. Tollersrud, B. Kvitle, H. J. Jørgensen, and S. Waage. 2005. Comparison of Staphylococcus aureus genotypes recovered from cases of bovine, ovine, and caprine mastitis. J. Clin. Microbiol. 43:3979–3984.

179. Nashev, D., K. Toshkova, S. I. Salasia, A. A. Hassan, C. Lämmler, and M. Zschöck. 2004. Distribution of virulence genes of Staphylococcus aureus isolated from stable nasal carriers. FEMS Microbiol. Lett. 233:45-52.

180. NMC (National Mastitis Council). 1999. Laboratory Handbook on Bovine Mastitis. NMC Inc., Madison, WI.

181. Neave, F. K., F. H. Dodd, R. G. Kingwill, and D. R. Westgarth. 1969. Control of mastitis in the dairy herd by hygiene and management. J. Dairy Sci. 52:696-707.

182. Nilsson, I. M., O. Hartford, T. Foster, and A. Tarkowski. 1999. Alpha-toxin and gamma-toxin jointly promote Staphylococcus aureus virulence in murine septic arthritis. Infect. Immun. 67:1045-1049.

183. Noble, W. C. 1965. The production of subcutaneous staphylococcal skin lesions in mice. Br. J. Exp. Pathol. 46:254-262.

184. NORM-VET 2005. Usage of antimicrobial agents and occurrence of antimicrobial resistance in Norway. 2006. Available at: http://www.vetinst.no/eng/Research/Publications2/Norm-Norm-Vet-Report http://www.vetinst.no/Arkiv/Zoonosesenteret/NORM_NORM-VET_2005.pdf.

185. Nouwen, J., H. Boelens, A. van Belkum, and H. Verbrugh. 2004. Human factor in Staphylococcus aureus nasal carriage. Infect.Immun. 72:6685-6688.

186. Nunes, S. F., R. Bexiga, L. M. Cavaco, and C. L. Vilela. 2007. Technical Note: Antimicrobial susceptibility of Portuguese isolates of Staphylococcus aureus and Staphylococcus epidermidis in subclinical bovine mastitis. J. Dairy Sci. 90:3242-3246.

187. Olde Riekerink, R. G. M., H. W. Barkema, S. Veenstra, D. E. Poole, R. T. Dingwell, and G. P. Keefe. 2006. Prevalence of contagious mastitis pathogens in bulk tank milk in Prince Edward Island. Can. Vet. J. 47: 567–572.

188. de Oliveira, A. P., J. L. Watts, S. A. Salmon, and F. M. Aarestrup. 2000. Antimicrobial susceptibility of Staphylococcus aureus isolated from bovine mastitis in Europe and the United States. J. Dairy Sci. 83:855-862.

189. Ohlsen, K., W. Ziebuhr, K. P. Koller, W. Hell, T. A. Wichelhaus, and J. Hacker. 1998. Effects of subinhibitory concentrations of antibiotics on alpha-toxin (hla) gene expression of methicillin-sensitive and methicillin-resistant Staphylococcus aureus isolates. Antimicrob. Agents Chemother. 42:2817-2823.

190. Olsen, J. E., H. Christensen, and F. M. Aarestrup. 2006. Diversity and evolution of blaZ from Staphylococcus aureus and coagulase-negative staphylococci. Antimicrob. Chemother. 57:450-460.

191. O'Neill, E., C. Pozzi, P. Houston, H. Humphreys, D. A. Robinson, A. Loughman, T. J. Foster, and J. P. O'Gara. 2008. A novel Staphylococcus aureus biofilm phenotype mediated by the fibronectin-binding proteins, FnBPA and FnBPB. J. Bacteriol. 190:3835-3850.

192. Orchoa-Zarzosa, A., P. D. Loeza-Lara, F. Torres-Rodríguez, H. Loeza-Ángeles, N. Mascot-Chiquito, S. Sánchez-Baca, and L. E. López-Meza. 2008. Antimicrobial susceptibility and invasive ability of Staphylococcus aureus isolates from mastitis from dairy backyard systems. Antonie Van Leeuwenhoek. Epub. ahead of print.

75

193. Papakyriacou, H., D. Vaz, A. Simor, M. Louie, and M. J. McGavin. 2000. Molecular analysis of the accessory gene regulator (agr) locus and balance of virulence factor expression in epidemic methicillin-resistant Staphylococcus aureus. J. Infect. Dis 181:990-1000.

194. Park, Y. H., L. K. Fox, M. J. Hamilton, and W. C. Davis. 1993. Suppression of proliferative response of BoCD4+ T lymphocytes by activated BoCD8+ T lymphocytes in the mammary gland of cows with Staphylococcus aureus mastitis. Vet. Immunol. Immunopathol. 36:137-151.

195. Patti, J. M., B. L. Allen, M. McGavin, and M. Höök. 1994. MSCRAMM-mediated adherence of microorganisms to host tissues. Annu. Rev. Microbiol. 48:585.

196. Peacock, S. E., C. E. Moore, A. Justice, M. Kantzadou, L. Story, K. Mackie, G. O’Neill, and N. P. Day. 2002. Virulent combinations of adhesin and toxin genes in natural populations of Staphylococcus aureus. Infect. Immun. 70:4987-4996.

197. Peles, F., M. Wagner, L. Varga, I. Hein I, P. Rieck, K. Gutser K, P. Keresztúri, G. Kardos, I. Turcsányi, B. Béri, and A. Szabó. 2007. Characterization of Staphylococcus aureus strains isolated from bovine milk in Hungary. Int. J. Food Microbiol. 118:186-193.

198. Peterson, P., J. Verhoef, L. D. Sabath, and P. G. Quie. 1977. Effect of protein A on staphylococcal opsonization. Infect. Immun. 15:760-764.

199. Phonimdaeng, P., M. O'Reilly, P. Nowlan, A. J. Bramley, and T. J. Foster. 1990. The coagulase of Staphylococcus aureus 8325-4. Sequence analysis and virulence of site-specific coagulase-deficient mutants. Mol. Microbiol.4:393-404.

200. Piccinini, R., E. Binda, L. Zanini, and A. Zecconi. 2003a. Contagious pathogens reduce milk production. Informatore Agrario 59:55-57.

201. Piccinini, R., E. Binda, A. Zecconi, and L. Zanini. 2003b. Prevalence study on bulk tank milk cultures in 1000 dairy herds in Lombardia (Italy). 42nd Ann. Meet. Natl. Mastitis Council, Fort Worth, pp. 396-397.

202. Piepers, S., L. de Meulemeester, A. de Kruif, G. Opsomer, H. W. Barkema, and S. de Vliegher. 2007. Prevalence and distribution of mastitis pathogens in subclinically infected dairy cows in Flanders, Belgium. J. Dairy Res. 74:478-483.

203. Pitkälä, A., L. Salmikivi, P. Bredbacka, A.-L. Myllyniemi, and M. T. Koskinen. 2007. Comparison of tests for detection of beta-lactamase-producing staphylococci. J. Clin. Microbiol. 45:2031-2033.

204. Poelarends, J.J., Hogeveen, H., Sampimon, O.C., Sol, J. 2001. Monitoring subclinical mastitis in Dutch dairy herds. Proc. 2nd International Symposium on Mastitis and Milk Quality, Vancouver, BC, Canada, pp.145-149.

205. Postle, D. S., M. Roguinsky, and B. Poutrel. 1978. Incuced staphylococcal infections in the bovine mammary gland. Am. J. Vet. Res. 39:29-35

206. Prevost, G., B. Cribier, P. Couppie, P. Petiau, G. Supersac, V. Finck-Barbancon, H. Monteil, and Y. Piemont. 1995. Panton-Valentine leucocidin and gamma-hemolysin from Staphylococcus aureus ATCC 49775 are encoded by distinct genetic loci and have different biological activities. Infect. Immun. 63:4121-4129.

207. Proctor, R. A., B. Kahl, C. von Eiff, P. E. Vandaux, D. P. Lew, and G. Peters. 1998. Staphylococcal small colony variants have novel mechanisms for antibiotic resistance. Clin. Infect. Dis. Suppl. 1:S68-74. Review.

208. Projan, S. J. and R. P. Novick.1997. The molecular basis of pathogenesis. In: The staphylococci in human disease. Crossley, K. B. and G. L. Archer (Eds.). Churchill Livingstone, N. Y., pp. 55-81.

209. Pyörälä, S. and E. Pyörälä. 1997. Accuracy of methods using somatic cell count and N-acetyl-β-D-glucosaminidase activity in milk to assess the bacteriological cure of bovine clinical mastitis. J. Dairy Sci. 80:2820-2825.

210. Rabello, R. F., B. M. Moreira, R. M. M. Lopes, L. M. Teixeira, L. W. Riley, and A. C. D. Castro. 2007. Multilocus sequence typing of Staphylococcus aureus isolates recovered from cows with mastitis in Brazilian dairy herds. J. Med. Microbiol. 56:1505-1511.

211. Raimundo, O., M. Deighton, J. Capstick, and N. Gerraty. 1999. Molecular typing of Staphylococcus aureus of bovine origin by polymorphisms of the coagulase gene. Vet. Microbiol. 66:275–284.

212. Rainard, P. 2003. The complement in milk and defense of the bovine mammary gland against infections.Vet. Res. 34:647-70. Review.

76

213. Rainard, P. and C. Riollet. 2003. Mobilization of neutrophils and defense of the bovine mammary gland. Reprod. Nutr. Dev. 43:439-57. Review.

214. Rainard, P., J. C. Corrales, M. B. Barrio, T. Cochard T, and B. Poutrel. 2003. Leucotoxic activities of Staphylococcus aureus strains isolated from cows, ewes, and goats with mastitis: importance of LukM/LukF'-PV leukotoxin. Clin. Diagn. Lab. Immunol.10:272-277.

215. Rainard, P. 2007. Staphylococcus aureus leucotoxin LukM/F' is secreted and stimulates neutralising antibody response in the course of intramammary infection.Vet. Res. 38:685-96.

216. Reiter, B., M.E. Sharpe, and T.M. Higgs. 1970. Experimental infection of the non-lactating bovine udder with Staphylococcus aureus and Streptococcus uberis. Res. Vet. Sci.11:18-27.

217. Reksen, O., L. Sølverød, A. J. Branscum, and O. Østerås. 2006. Relationships between milk culture results and treatment for clinical mastitis or culling in Norwegian dairy cattle. J. Dairy Sci. 89:2928-2937.

218. Roberson, J. R., K. L. Fox, D. D. Hancock, J. M. Gay, and T. E. Besser. 1994. Ecology of Staphylococcus aureus isolated from various sites of dairy farms. J. Dairy Sci. 77:3354-3364.

219. Roberson, J. R., K. L. Fox, D. D. Hancock, J. M. Gay, and T. E. Besser.1998. Sources of intramammary infections from Staphylococcus aureus in dairy heifers at first parturition. J. Dairy Sci. 8:687-693.

220. Rooijakkers, S. H. M., M. Ruyken, A. Roos, H. R. Daha, J. S. Presanis, R. B. Sim, W. J. B. Van Wamel, K. P. M. Van Kessel, and J. A. G. van Strijp. 2005. Immune evasion by a staphylococcal complement inhibitor that acts on C3 convertases. Nat. Immunol. 6:920-927.

221. Salasia, S. I., Z. Kushnan, C. Lämmler, and M. Zschöck. 2004. Comparative studies on pheno- and genotypic properties of Staphylococcus aureus isolated from bovine subclinical mastitis in central Java in Indonesia and Hesse in Germany. J. Vet. Sci. 5:103-109.

222. Salmenlinna S, O. Lyytikäinen, P. Kotilainen, R. Scotford, E. Siren, and J. Vuopio-Varkila. 2000. Molecular epidemiology of methicillin-resistant Staphylococcus aureus in Finland. Eur. J. Clin. Microbiol. Infect. Dis.19:101-107.

223. Saginur, R, K. Sleigh, H. Marks, M. G. Bergeron, A.W. Chow, and the ESPRIT Study Group. 1996. S. aureus nasal carriage and post-operative wound infection. Abstract, 308. In 8th International Symposium of staphylococci and staphylococcal infections.

224. Salmon, S. A., J. L. Watts, J. W. Aarestrup, and R. J. Yancey Jr. 1998. Minimum Inhibitory concentrations for selected antimicrobial agents against organisms isolated from the mammary glands of dairy heifers in New Zealand and Denmark. J. Dairy Sci. 81:570–578.

225. Sandholm M., L. Kaartinen, P. Hyvönen, K. Veijalainen, and P. L. Kuosa. 1989. Flotation of mastitis pathogens with cream from subclinically infected quarters. Prospects for developing a cream-rising test for detecting mastitis caused by major mastitis pathogens. Zentralbl. Veterinarmed. [B]. 36:27-34.

226. Schlievert, P. M. 1982. Enhancement of host susceptibility to lethal endotoxin shock by staphylococcal pyrogenic exotoxin type C. Infect. Immun. 36:123-128.

227. Schuberth, H.-J., C. Krueger, H. Zerbe, E. Bleckmann, and W. Leibold. 2001. Characterization of leukocytotoxic and superantigen-like factors produced by Staphylococcus aureus isolates from milk of cows with mastitis. Vet. Microbiol. 82:187-199.

228. Schukken, Y. H., B. A. Mallard, J. C. Dekkers, K. E. Leslie, and M. J. Stear. 1994. Genetic impact on the risk of intramammary infections following Staphylococcus aureus challenge. J. Dairy Sci. 77:639–647.

229. Schukken, Y. H., K. E. Leslie, D. A. Barnum, B. A. Mallard, J. H. Lumsden, P. C. Dick, G. H. Vessie, and M. E. Kehrli. 1999. Experimental Staphylococcus aureus intramammary challenge in late lactation dairy cows: quarter and cow effects determining the probability of infection. J. Dairy Sci. 82:2393-2401.

230. Schällibaum, M. 1999. Mastitis-pathogens isolated in Switzerland: 1987-1996. IDF Mastitis Newsletter 24:38.

231. Selander, R. K., D. A. Caugant, H. Ochman, J. M. Musser, M. N. Gilmour, and T. S. Whittam. 1986. Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl. Environ. Microbiol. 51:873-884.

232. Singh, G., R. R. Marples, and A. M.Klingman. 1971. Experimental Staphylococcus aureus infections in humans. J. Invest. Dermatol. 57:149-162.

77

233. Sinha, B., P. P. Francois, O. Nusse, M. Foti, O. M. Hartford, P. Vaudaux, T. J. Foster, D. P. Lew M. Herrmann, and K. H. Krause. 1999. Fibronectin-binding protein acts as Staphylococcus aureus invasin via fibronectin bridging to integrin alpha5beta1. Cell Microbiol. 1:101-117.

234. Smith, E. M., L. E. Green, G. F. Medley, H. E. Bird, and C. G. Dowson. 2005a. Multilocus sequence typing of Staphylococcus aureus isolated from high-somatic-cell-count cows and the environment of an organic dairy farm in the United Kingdom. J. Clin. Microbiol. 43: 4731–4736.

235. Smith, E. M., Green, L. E., Medley, G. F., Bird, H. E., Fox, L. K., Schukken, Y. H., Kruze, J. V., Bradley, A. J., Zadoks, R. N., and Dowson, C. G. 2005b. Multilocus sequence typing of intercontinental bovine Staphylococcus aureus isolates. J. Clin. Microbiol. 43:4737-4743.

236. Smith, T. H., L. K. Fox, and J. R. Middleton. 1998. Outbreak of mastitis caused by one strain of Staphylococcus aureus in a closed dairy herd. J. Am. Vet. Med. Assoc. 212:553-556.

237. Smyth, D. S., P. J. Hartigan, W. J. Meaney, J. R. Fitzgerald, C. F. Deobald, G. A. Bohach, and C. J. Smyth. 2005. Superantigen genes encoded by the egc cluster and SaPIbov are predominant among Staphylococcus aureus isolates from cows, goats, sheep, rabbits and poultry. J. Med. Microbiol. 54:401-411.

238. Sol, J., O. C. Sampimon, J. J. Snoep, and Y. H. Schukken. 1997. Factors associated with bacteriological cure during lactation after therapy for subclinical mastitis caused by Staphylococcus aureus. J. Dairy Sci. 80:2803–2808.

239. Sol, J., Sampimon, O. C. , Barkema, H. W., Schukken, Y. H., 2000. Factors associated with cure after therapy of clinical mastitis caused by Staphylococcus aureus. J. Dairy Sci. 83, 278-84.

240. Sol, J. 2002. Cure of Staphylococcus aureus mastitis in Dutch dairy cows. Ph.D. Thesis. University of Utrecht. Utrecht.

241. Sommerhäuser, J., B. Kloppert, W. Wolter, M. Zschöck, A. Sobiraj, and K. Failing. 2003. The epidemiology of Staphylococcus aureus infections from subclinical mastitis in dairy cows during a control programme. Vet. Microbiol. 96:91-102.

242. Sordelli, D. O., F. R. Buzzola, M. I. Gomez, L. Steele-Moore, D. Berg, E. Gentilini, M. Catalano, A. J. Reitz, T. Tollersrud, G. Denamiel, P. Jeric, and J. C. Lee. 2000. Capsule expression by bovine isolates of Staphylococcus aureus from Argentina: genetic and epidemiologic analyses. J. Clin. Microbiol. 38:846-850.

243. Srinivasan, V., B. E. Gillespie, M. J. Lewis, L. T. Nguyen, S. I. Headrick, Y. H. Schukken, and S. P. Oliver. 2006. Prevalence of enterotoxin and toxic shock syndrome toxin genes in Staphylococcus aureus isolated from milk of cows with mastitis. Foodborne Pathog. Dis. 3:274-283.

244. Stephan, R., C. Annemüller, A. A. Hassan, and C. Lämmler. 2001. Characterization of enterotoxigenic Staphylococcus aureus strains isolated from bovine mastitis in north-east Switzerland.Vet. Microbiol. 78:373-382.

245. Sung, J. M., D. H. Lloyd, and J. A. Lindsay. 2008. Staphylococcus aureus host specificity: comparative genomics of human versus animal isolates by multi-strain microarray. Microbiology. 154:1949-1959.

246. Sutra, L. and B. Poutrel. 1994. Virulence factors involved in the pathogenesis of bovine intramammary infections due to Staphylococcus aureus. J. Med. Microbiol. 40:79-89. Review.

247. SVARM 2001. Swedish Veterinary Antimicrobial Resistance Monitoring. 2002. Available at. http://www.sva.se/upload/pdf/Tjänster%20och%20produkter/Trycksaker/svarm2001.pdf

248. SVARM 2002. Swedish Veterinary Antimicrobial Resistance Monitoring. 2003. Available at: http://www.sva.se/upload/pdf/Tjänster%20och%20produkter/Trycksaker/svarm2002.pdf

249. Swinkels, J. M., H. Hogeveen, and R. N. Zadoks. 2005. A partial budget model to estimate economic benefits of lactational treatment of subclinical Staphylococcus aureus mastitis. J. Dairy Sci. 88:4273-4287.

250. Talbot, B. G. and P. Lacasse. 2005. Progress in the development of mastitis vaccines. Livestock production science 98: 101-113.

251. Taponen, S., A. Jantunen, E. Pyörälä, and S. Pyörälä. 2003. Efficacy of targeted 5-day combined parenteral and intramammary treatment of clinical mastitis caused by penicillin-susceptible or penicillin-resistant Staphylococcus aureus. Acta vet. Scand. 44:53-62.

78

252. Tenhagen, B.-A., G. Köster, J. Wallmann, and W. Heuwieser. 2006. Prevalence of mastitis pathogens and their resistance against antimicrobial agents in dairy cows in Brandenburg, Germany. J. Dairy Sci. 89:2542-2551.

253. Tenover, F. C., R. Arbeit, G. Archer, J. Biddle, S. Byrne, R. Goering, G. Hancock, G. A. Hébert, B. Hill, R. Hollis, et al. 1994. Comparison of traditional and molecular methods of typing isolates of Staphylococcus aureus. J. Clin. Microbiol. 32: 407–415.

254. Tenover, F. C., R. D. Arbeit, R. V. Goering, G. Hancock, P. A. Mickelsen, B. E. Murray, D. H. Persing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33:2233-2239.

255. Thorne, H. and G. Wallmark. 1960. Studies on staphylococcal mastitis in an individual herd with special reference to phage-types of isolated staphylococci. Acta. vet. Scand. 1:114.

256. Tollersrud, T., K. Kenny, A. J. Reitz Jr., and J. C. Lee. 2000a. Genetic and serologic evaluation of capsule production by bovine mammary isolates of Staphylococcus aureus and other Staphylococcus spp. from Europe and the United States. J. Clin. Microbiol. 38:2998-3003.

257. Tollersrud, T., Kenny, K., Caugant, D.A., and Lund, A. 2000b. Characterization of isolates of Staphylococcus aureus from acute, chronic and subclinical mastitis in cows in Norway. APMIS 108:565-572.

258. Tollersrud, T., A. H. Kampen, and K. Kenny. 2006. Staphylococcus aureus enterotoxin D is secreted in milk and stimulates specific antibody responses in cows in the course of experimental intramammary infection. Infect. Immun. 74:3507-3512.

259. Trinidad, P., S. C. Nickerson, and T. K. Alley. 1990. Prevalence of intramammary infection and teat canal colonization in unbred and primigravid dairy heifers. J. Dairy Sci. 73:107-114.

260. Verhoef, J. 1997. In: The staphylococci in human disease. Crossley, K. B. and G. L. Archer (Eds.). Churchill Livingstone, N. Y., pp. 213-232.

261. Vesterholm-Nielsen, M., M. Ø. Larsen, and J. E. Olsen. 1999. Occurrence of blaZ gene in penicillin resistant Staphylococcus aureus isolated from bovine mastitis in Denmark. Acta vet. Scand. 40:279-286.

262. Waage, S., T. Mørk, A. Røros, D. Aasland, A. Hunshamar, and A. Ødegaard. 1999. Bacteria associated with clinical mastitis in dairy heifers. J. Dairy Sci. 82:712-719.

263. Waage, S., J. Bjørland, D. A. Caugant, H. Oppegaard, T. Tollersrud, T. Mørk, and F. M. Aarestrup. 2002. Spread of Staphylococcus aureus resistant to penicillin and tetracycline within and between dairy herds. Epidemiol. Infect. 129:193-202.

264. van Wamel, W. J. B., S. H. M. Rooijakkers, M. Ruyken, K. P. M. van Kessel, and J. A. G. van Strijp. 2006. The innate immune modulators staphylococcal complement inhibitor and chemotaxis inhibitory protein of Staphylococcus aureus are located on β-hemolysin-converting bacteriophages. J. Bacteriol. 188:1310–1315.

265. van Wely, K. H., J. Swaving, R. Freudl, and A. J. Driessen. 2001. Translocation of proteins across the cell envelope of Gram-positive bacteria. FEMS Microbiol. Rev.25:437-54.

266. Whist, A. C., O. Österås, and L. Solverod. 2006. Clinical mastitis in Norwegian herds after a combined selective dry-cow therapy and teat-dipping trial. J. Dairy Sci. 89:4649-4659.

267. Wilson C. D. and M. S. Richards. 1980. A survey of mastitis in the British dairy herd. Vet Rec. 106:431–435.

268. Wilson, D. J., R. N. Gonzales, and P. M. Sears. 1995. Segregation or use of separate milking units for cows infected with Staphylococcus aureus: effects on prevalence of infection and bulk tank somatic cell count. J. Dairy Sci. 78:2083-5.

269. Young, B., D. Platt, D. Logue, H. Ternent, and J. Fitzpatrick. 2001. Bovine Staphylococcus aureus mastitis: strain recognition and dynamics of infection. J. Dairy Res. 68:377-88.

270. Zadoks, R., W. van Leeuwen, H. Barkema, O. Sampimon, H. Verbrugh, Y. H. Schukken, and A. van Belkum. 2000. Application of pulsed-field gel electrophoresis and binary typing as tools in veterinary clinical microbiology and molecular epidemiologic analysis of bovine and human Staphylococcus aureus isolates. J. Clin. Microbiol. 38:1931-1939.

271. Zadoks, R. N., H. G. Allore, H. W. Barkema, O. C. Sampimon, G. J. Wellenberg, Y. T. Gröhn, and Y. H. Schukken. 2001. Cow-and quarter-level risk factors for Streptococcus uberis and Staphylococcus aureus mastitis. J. Dairy Sci. 84:2649–2663.

79

272. Zadoks, R. N., W. B. van Leeuwen, D. Kreft, L. K. Fox, H. W. Barkema, Y. H. Schukken, and A. van Belkum. 2002. Comparison of Staphylococcus aureus isolates from bovine and human skin, milking-equipment, and bovine milk by phage typing, pulsed-field gel electrophoresis, and binary typing. J. Clin. Microbiol. 40:3894-3902.

273. Zhang, S., J. J. Iandolo, and G. C. Stewart. 1998. The enterotoxin D plasmid of Staphylococcus aureus encodes a second enterotoxin determinant (sej). FEMS Microbiol. Lett. 168:227-233.

274. Ziv, G. and M. Storper. 1985. Intramuscular treatment of subclinical staphylococcal mastitis in lactating cows with penicillin G, methicillin and their esters. J. Vet. Pharmacol. Ther. 8:276–283.

275. Zschöck, M., K. Risse, and J. Sommerhäuser. 2004. Occurrence and clonal relatedness of sec/tst-gene positive Staphylococcus aureus isolates of quartermilk samples of cows suffering from mastitis. Lett. Appl. Microbiol. 38:493-498.

276. Yazdankhah, S. P., H. Sørum, and H. Oppegaard. 2000. Comparison of genes involved in penicillin resistance in staphylococci of bovine origin. Microb. Drug Resist. 6:29-36.

277. Østerås, O., H. Hogeveen, D. K. Singh, and K. E. Leslie. 2005. Economic consequences of mastitis. IDF Bulletin 394/2005. International Dairy Federation, Bruxelles, Belgium.

278. Østerås, O., L. Sølverød, and O. Reksen. 2006. Milk culture results in a large Norwegian survey- effects of season, parity, days in milk, resistance, and clustering. J. Dairy Sci. 89:1010-1023.