Epidemiological Investigations Inform Ostreid Herpesvirus 1 Disease Control In

Epidemiological investigations inform Ostreid herpesvirus 1 disease control in

Pacific oysters (Crassostrea gigas)

Maximilian Clarence de Kantzow

A thesis submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

Farm Animal Health, Sydney School of Veterinary Science

Faculty of Science

THE UNIVERSITY OF SYDNEY

2020

Declaration of Authorship

Apart from the assistance stated in the acknowledgements section, this thesis represents the original work of the author. The results of this study have not been presented for any other degree or diploma at any other university.

Maximilian Clarence de Kantzow

Bachelor of Animal and Veterinary Bioscience (Honours class 1)

February 2020

Acknowledgements

First and foremost, I would like to thank my supervisors, Dr Paul M. Hick and Professor

Richard J. Whittington for their support and guidance throughout my candidature. I am grateful for the research environment they have provided and for opportunity to contribute to aquatic animal health research at the University of Sydney. This research was funded by the

Fisheries Research and Development Corporation and by the University of Sydney through grants awarded to Professor Whittington and Dr Hick.

I would also like to thank the technical staff of the Farm Animal Health Virology Laboratory and Infectious Disease Laboratory for their support and guidance throughout my research

(Anna Waldron, Alison Tweedie, Slavicka Patten, Ann-Michele Whittington, Natalie Schiller,

Nicole Carter, Rebecca Maurer). Their advice, assistance and patience in the laboratory are very much appreciated.

I would like to thank Technical Officers Stuart Glover, Craig Kristo and Nobel Toribio for their assistance and advise over many experiments in the aquatic animal facility. Thank you to the Research fellows Dr Om Dhungyel, Dr Karren Plain, Dr Auriol Purdie, Dr Kumi de Silvia,

Dr Hannah Pooley, Dr Marine Fuhrmann, Dr Andrew McPherson and Dr Olivia Evans, and

PhD students Erandi Pathirana, Cahya Fusianto, Karen Smith, Anna Ly, Kathryn Wright and

Ed Annand of the Farm Animal Health group for their support and camaraderie. I am grateful to Dr Joy Becker and Dr Navneet Dhand for sharing their expertise and advice and for their patience. I am also very grateful to Marion Saddington for her generous assistance throughout my candidature.

Thank you to Shellfish Culture, Tasmania for supplying Crassostrea gigas spat, Bruce Alford of Broken Bay Oysters, Robert Hill and Keith Duggen of Endeavour Oysters, and to Leon and

Angela Riepsamen of Goodnight Oysters for supplying and growing the oysters what were used for the experiments in this thesis. Thank you to the oyster farmers of upper and lower

Pittwater and Blackman bay for their help and allowing me to conduct field work on their farms.

I would like to thank my parents Megan and Colin and my two sisters Rose and Lillian for their support throughout my time at university. I would like to thank the PD for its assistance while writing. Last, but certainly not least, I would like to thank my partner Clara for her love and support.

iii

Summary

Microvariant genotypes of Ostreid herpesvirus-1 (OsHV-1) first emerged in France in 2008 and have spread around Europe causing seasonal outbreaks with up to 100% mortality in

Pacific oysters (Crassostrea gigas). In 2008, the initial reference genotype of OsHV-1 was shortly replaced by the microvariants strains which spread across Europe and have been identified in Asia, Australia, and New Zealand. Since the first detection of OsHV-1 in the

Georges River, near Sydney, in 2010, this emerging pathogen has been a threat to the

Australian Pacific oyster industry in the Georges and Hawkesbury Rivers in NSW and in the waterways around Hobart in TAS. In Australia mass mortality events caused by OsHV-1 have been called Pacific oysters mortality syndrome (POMS).

Epidemiological studies of POMS indicate that the disease is multifactorial and that environmental conditions are an important determinant of cumulative mortality. The water temperature and growing conditions, including the type of farming infrastructure, all have a strong influence on mortality as described in the literature review (Chapter 1). Low stocking density production systems such as rope culture were associated with lower mortality than basket systems as was reduced feed availability and increasing age and size. Disease control methods including the use of growing structures with an increased height in the intertidal environment can reduce exposure and mortality due to OsHV-1. The use of genetically selected oysters for improved resistance to disease caused by OsHV-1, treating water before use in hatcheries, and manipulating the age and size of oysters prior to the OsHV-1 disease risk period have all been pursued as potential control strategies. An understanding of the interactions between the Pacific oyster, OsHV-1 and environment can be used to inform the continued development and refinement of disease control mechanisms.

The laboratory infection model for OsHV-1 used in Chapters 4 – 7 is detailed in Chapter 2.

This was adapted from a previously described procedure and performed in a physical containment level 2 aquatic animal facility at the University of Sydney. A cryopreserved

OsHV-1 isolate which originated from a POMS outbreak in the Georges River in 2011 was used in the infection model to maintain consistency across trials. The use of a laboratory infection model allowed the environmental conditions during each challenge to be tightly controlled and monitored to reduce the impact of confounding variables present in the field.

Additionally, the time of exposure and dose of OsHV-1 was known and could be controlled.

The outcome of the trials was determined in terms of mortality following the OsHV-1 challenge compared to authentic negative control inoculum, made from tissue taken from apparently healthy oysters. Further, OsHV-1 DNA was quantified using a qPCR assay according to a standard method in a laboratory accredited by the National Association of

Testing Authorities (NATA) according to ISO 17025:2017 (General requirements for the competence of testing and calibration laboratories).

Risk factors affecting mortality during the first occurrence POMS in Tasmania, Australia, in

January 2016 were assessed in Chapter 3. It was important to capture the effect of risk factors present during this outbreak because it was free from the confounding factor of prior infection with OsHV-1. A multivariable logistic regression model was used to analyse each risk factor while accounting for the effect of the other risk factors. There was a wide range in the mortality of farmed Pacific oysters: between 37% and 92% across the 6 farms assessed, even though the production methods and stock were relatively homogenous. The mortality also ranged from 33% at the smallest (up to 20 mm) size class to 96% at the largest (61 – 115 mm).

There was also a greater risk of mortality for oysters which were handled for grading in the

v week prior to the POMS outbreak. The variation in mortality suggested that there was an environmental effect between baskets and farms which needed further investigation. This was an initial outbreak in naïve populations, with different expectations about the incidence and severity of disease when OsHV-1 became endemic. The recommendations from the study were to reduce the handling frequency and the number of small, young oysters present on- farm during periods of high risk for POMS.

The effect of changing water temperature between multiple exposures to OsHV-1 was examined in Chapter 4. A laboratory infection model was used to evaluate how the water temperature during an oyster’s first exposure to OsHV-1 affected survival following a second exposure to OsHV-1 at 22°C which was permissive to high mortality disease. An injection challenge was used because this standardized the dose and timing of OsHV-1 exposure. The mortality in oysters across OsHV-1 exposures at 18°C was 0 – 9% whereas mortality in oysters initially exposed at 22°C was 59%. A prior OsHV-1 challenge at a water temperature of 18°C reduced the level of mortality during a second challenge at a water temperature of

22°C in six-month-old oyster spat. This suggested that a non-lethal exposure to OsHV-1 at a water temperature where low mortality occurs can have a protective effect against mortality on subsequent exposure to the virus. Further investigation of the duration and specificity of the protective effect and how this may affect the seasonal mortality patterns in endemic areas is required. This will determine if the protective effect of controlled exposure can be adapted as a disease control measure.

In Chapter 5, the laboratory infection model was used to define an in vivo growth curve for

OsHV-1 at 18°C and 22°C. The growth curve was required to further understand how the water temperature at the time oysters were exposed to OsHV-1 affected viral replication. High

vi temporal resolution was obtained by quantifying the concentration of OsHV-1 DNA in oysters sampled at 10 time points between 2 and 72 hours after exposure by injection. The rate of viral replication was the same at each water temperature, but OsHV-1 replication continued for several more hours at 22°C compared to 18°C. This resulted in a peak viral DNA concentration at 22°C that was 100 times higher than at 18°C. Mortality occurred 72 hours after the peak viral concentration at 18°C and 20 hours after the peak viral concentration at

22°C. It was also important to determine if oysters which survived the initial infection at 18°C would then experience more severe disease or mortality when the water temperature was raised to 22°C. This was examined in a separate experiment and it was determined that increasing the water temperature to 22°C 14 days after the initial OsHV-1 exposure at 18°C did not induce further mortality. These experiments demonstrated the importance of the water temperature during the initial exposure to OsHV-1 on the outcome of the infection and suggest that water temperature accounts for a large amount of the variation in mortality.

Previous experiments identified a water temperature of 18°C as the transition point above which OsHV-1 infection leads to mortality and 22°C as a water temperature where OsHV-1 consistently caused mortality in excess of 50%. The expression of important Pacific oyster immune genes in response to OsHV-1 infection were examined at each of these water temperatures in Chapter 6. Gill biopsy samples were taken at key timepoints identified based on the growth curve to further effort to understand the difference in mortality and peak viral titre between oysters exposed to OsHV-1 at a water temperature of 18°C or 22°C. A novel gill biopsy method was developed as a non-sacrificial method to sample the oysters so that the outcome of the infection could be linked with host immune gene expression data during the acute stages of infection. The mortality was 3.5 times greater in oysters exposed to OsHV-1 at

vii

22°C than at 18°C. The expression of three immune genes which had been identified as potentially important in response to OsHV-1 were measured by RT-qPCR at immediately before and at 2, 6, 12 and 24 hours post OsHV-1 exposure. The pattern of transcription of

Viperin, Myeloid differentiation factor 88 (MyD88), the apoptosis inhibitor, Baculoviral IAP repeat-containing protein 2 (IAP) and in oysters at 18°C and 22°C. These genes were upregulated at 2 hours post exposure, and down regulated 24 hours post exposure at 18°C.

However, in oysters at 22°C this pattern was reversed for transcription of Viperin and IAP. In oysters at 22°C, MyD88 was most strongly upregulated at 2 hours post exposure and most strongly down regulated at 6 hours post exposure. Down regulation of IAP was associated with mortality due to OsHV-1. Different patterns of immune gene expression in response to

OsHV-1 infection at these two water temperatures may explain, in part, the influence of water temperature on mortality.

To investigate if the protective effect provided by prior exposure to OsHV-1 required active viral replication, the effect of prior exposure to heat inactivated OsHV-1 was examined in

Chapter 7. Oysters were injected with different agents including viable OsHV-1, heat inactivated OsHV-1 and the synthetic dsRNA poly I:C at 18°C and then, 21 days later they were challenged with OsHV-1 at 22°C. These agents were all protective against mortality due to OsHV-1. This suggested that active OsHV-1 replication was not required to protect oysters from future mortality due to OsHV-1 and indicated that oysters could be conditioned to survive a POMS outbreak with heat inactivated OsHV-1 or controlled exposure to OsHV-1 at

18°C. The pattern of POMS mortality may also change in endemic waterways where oysters are being exposed to inactive OsHV-1.

viii

In conclusion, many factors in the environment of Pacific oysters, but particularly water temperature, substantially impacted the severity of disease caused by OsHV-1. Replication of

OsHV-1 and the expression of immune genes by Pacific oysters was different at 18°C and

22°C. The low mortality subsequent to OsHV-1 infection at 18°C created a pool of exposed oysters that was resistant to further mortality. The protective effect of OsHV-1 replication at a water temperature of 18°C against subsequent mortality was also induced by the non-specific immune stimuli of inactivated OsHV-1 and poly I:C. Further work is required to develop this concept of conditioning oysters by exposure to OsHV-1 under controlled conditions or use of heat inactivated OsHV-1 as a disease control strategy for POMS. A full understanding of the mechanisms that confers protection through conditioning oysters for lower mortality requires further investigation. This may lead to the development of practical and feasible method of controlling OsHV-1 exposure to minimize the impact of disease.

List of Publications and Conference Presentations

Publications de Kantzow, M.C., Hick, P.M., Dhand, N.K., Whittington, R.J., 2017. Risk factors for mortality during the first occurrence of Pacific Oyster Mortality Syndrome due to Ostreid herpesvirus 1 in Tasmania, 2016. Aquaculture 468, Part 1, 328-336. de Kantzow, M.C., Whittington, R.J., Hick, P.M., 2019. Prior exposure to Ostreid herpesvirus

1 (OsHV-1) at 18°C is associated with improved survival of juvenile Pacific oysters

(Crassostrea gigas) following challenge at 22°C, 2019. Aquaculture 507, 443-450. de Kantzow, M.C., Whittington, R.J., Hick, P.M., 2019b. Different in vivo growth of Ostreid herpesvirus 1 at 18°C and 22°C alters mortality of Pacific oysters (Crassostrea gigas). Arch.

Virol 164, Issue 12, 3035-3043.

Presentations de Kantzow, M.C., Hick, P.M., Dhand, N.K., Whittington, R.J., Factors affecting POMS mortality in Tasmania January – February 2016, 4th FRDC Australasian Aquatic Animal

Health & Biosecurity Scientific Conference, July 2017, Cairns, Australia. de Kantzow, M.C., Hick, P.M., Dhand, N.K., Whittington, R.J., Factors affecting POMS mortality in Tasmania Jan – Feb 2016, Shellfish Futures - Industry meeting of the Australian

Oyster Growers Association, June 2016, Hobart, Australia. de Kantzow, M.C., Hick, P.M., Whittington, R.J., Water temperature modulation of Pacific oysters (Crassostrea gigas) spat following controlled pre-exposure to Ostreid herpesvirus – 1

(OsHV-1), Sydney School of Veterinary Science Postgraduate Conference, November 2017,

Sydney, Australia. de Kantzow, M.C., Whittington, R.J., Hick, P.M., Conditioning Pacific oyster (Crassostrea gigas) spat for improved survival of Ostreid herpesvirus – 1 (OsHV-1) by controlled infection, 8th International Symposium on Aquatic Animal Health, September 2018,

Charlottetown, Canada. de Kantzow, M.C., Whittington, R.J., Hick, P.M., Ostreid herpesvirus - 1 (OsHV-1) in vivo growth curve and pathogenesis at a semi-permissive water temperature, 8th International

Symposium on Aquatic Animal Health, September 2018, Charlottetown, Canada.

Authorship Attribution

Chapter 3 of this thesis is published as: de Kantzow, M.C., Hick, P.M., Dhand, N.K., Whittington, R.J., 2017. Risk factors for mortality during the first occurrence of Pacific Oyster Mortality Syndrome due to Ostreid herpesvirus 1 in Tasmania, 2016. Aquaculture 468, Part 1, 328-336.

I co-designed the study, conducted the field work, analysed the data and wrote the drafts of the MS.

Chapter 4 of this thesis is published as: de Kantzow, M.C., Whittington, R.J., Hick, P.M., 2019. Prior exposure to Ostreid herpesvirus

1 (OsHV-1) at 18°C is associated with improved survival of juvenile Pacific oysters

(Crassostrea gigas) following challenge at 22°C. Aquaculture 507, 443–450.

I co-designed the study, conducted the experiment, analysed the data and wrote the drafts of the MS.

xii

Chapter 5 of this thesis is published as: de Kantzow, M.C., Whittington, R.J., Hick, P.M., 2019b. Different in vivo growth of Ostreid herpesvirus 1 at 18°C and 22°C alters mortality of Pacific oysters (Crassostrea gigas). Arch.

Virol. 164, Issue 12, 3035-3043.

I co-designed the study, conducted the experiment, analysed the data and wrote the drafts of the MS.

xiii

Attesting Authorship Attribution

In addition to the statements above, in cases where I am not the corresponding author of a published item, permission to include the published material has been granted by the corresponding author.

Maximilian C. de Kantzow

February 2020

As supervisor for the candidature upon which this thesis is based, I can confirm that the authorship attribution statements above are correct.

Senior Lecturer Paul M. Hick

February 2020

As supervisor for the candidature upon which this thesis is based, I can confirm that the authorship attribution statements above are correct.

Professor Richard J. Whittington

February 2020

xiv

Table of Contents

Declaration of Authorship ...... i

Acknowledgements ...... ii

Summary ...... iv

List of Publications and Conference Presentations ...... x

Publications ...... x

Presentations ...... x

Authorship Attribution ...... xii

Attesting Authorship Attribution ...... xiv

List of Tables ...... xxiv

List of Figures ...... xxviii

Abbreviations ...... xxxiv

1. Chapter 1 – A review of the influence of farming practices on the environment experienced by Pacific oysters (Crassostrea gigas) and implications for the control of Ostreid herpesvirus 1 (OsHV-1) ...... 1

1.1 Introduction ...... 1

1.1.1 Pacific oyster aquaculture in Australia ...... 1

1.1.2 Farming infrastructure ...... 2

1.1.3 Multifactorial disease in oyster aquaculture ...... 3

1.2 OsHV-1 ...... 5

1.2.1 Emergence ...... 5

1.2.2 Virology ...... 6

1.2.3 OsHV-1 genetics ...... 9

1.3 OsHV-1 Epidemiology ...... 10

1.4 OsHV-1 Transmission ...... 20

1.5 Pacific Oyster Immunology ...... 20

1.5.1 Antiviral immunity ...... 21

1.5.2 Immune memory and immune priming ...... 23

1.6 Management of Disease Caused by OsHV-1...... 24

1.6.1 Preventing OsHV-1 transmission ...... 25

1.6.2 Modifying farming infrastructure ...... 25

1.6.3 Manipulating age and size ...... 26

1.6.4 Antiviral immune priming ...... 26

1.6.5 Selective breeding for genetic resistance to OsHV-1 ...... 27

1.7 Conclusion ...... 28

1.8 Experimental Approach ...... 29

2. General Materials and Methods ...... 30

2.1 Reagents ...... 30

2.1.1 MilliQ water ...... 30

2.1.2 Artificial sea water...... 30

xvi

2.1.3 Fresh aquarium water ...... 31

2.1.4 Magnesium chloride solution (50 gL-1) ...... 31

2.1.5 Lysis buffer ...... 31

2.1.6 Trigene (1:100) ...... 32

2.1.7 Virkon solution (1%) ...... 32

2.1.8 Chlorine solution (1000 ppm) ...... 32

2.2 Aquatic Animal Housing Facility ...... 33

2.2.1 Biosecurity ...... 33

2.2.2 Oysters ...... 34

2.2.3 Aquaria ...... 34

2.2.4 Random allocation ...... 36

2.2.5 Water temperature monitoring...... 37

2.2.6 Water quality monitoring...... 37

2.2.7 Feed ...... 37

2.3 OsHV-1 Inoculum ...... 38

2.3.1 Cryopreserved inoculum...... 38

2.3.2 Fresh inoculum ...... 39

2.3.3 OsHV-1 challenge ...... 39

2.3.4 Oyster mortality ...... 40

2.3.5 Waste management ...... 41

xvii

2.4 Quantification of OsHV-1 DNA ...... 41

2.4.1 Tissue sampling ...... 41

2.4.2 Tissue handling for RNA extraction ...... 42

2.4.3 Tissue homogenisation ...... 42

2.4.4 Purification of nucleic acids ...... 42

2.4.5 OsHV-1 quantification by qPCR ...... 43

2.4.6 Tissue processing and PCR quality control ...... 45

2.5 Statistical Methods ...... 45

2.5.1 Data management ...... 46

2.5.2 Summary statistics ...... 46

2.5.3 Confidence intervals ...... 46

2.5.4 Generalised linear mixed models ...... 47

2.5.5 Cox proportional hazards model...... 48

3. Chapter 3 - Risk factors for mortality during the first occurrence of Pacific Oyster

Mortality Syndrome due to Ostreid herpesvirus – 1 in Tasmania, 2016 ...... 49

3.1 Abstract ...... 49

3.2 Introduction ...... 50

3.3 Methods ...... 55

3.3.1 Farms ...... 55

3.3.2 Study design ...... 56

xviii

3.3.3 Sampling ...... 57

3.3.4 Size measurement ...... 59

3.3.5 Quantification of mortality ...... 59

3.3.6 Risk factors ...... 59

3.3.7 Statistical analysis ...... 60

3.3.7.1 Descriptive analysis ...... 61

3.3.7.2 Generalized linear mixed model ...... 61

3.3.7.3 Length variation within basket...... 64

3.4 Results ...... 65

3.5 Discussion ...... 73

3.6 Conclusion ...... 77

4. Chapter 4 - Prior exposure to Ostreid herpesvirus 1 (OsHV-1) at 18°C is associated with improved survival of juvenile Pacific oysters (Crassostrea gigas) following challenge at 22°C

4.1 Abstract ...... 82

4.2 Introduction ...... 83

4.3 Methods ...... 85

4.3.1 Oysters ...... 85

4.3.2 Aquarium management...... 85

4.3.3 OsHV-1 challenge ...... 85

xix

4.3.4 Quantification of OsHV-1 DNA ...... 85

4.3.5 Experimental design ...... 86

4.3.6 Statistical analysis ...... 88

4.4 Results ...... 90

4.4.1 Mortality and prevalence of OsHV-1 ...... 90

4.4.2 Quantity of OsHV-1 DNA in oyster tissues ...... 95

4.5 Discussion ...... 100

5. Chapter 5 - Different in vivo growth of Ostreid herpesvirus 1 at 18°C and 22°C alters mortality of Pacific oysters (Crassostrea gigas) ...... 105

5.1 Abstract ...... 105

5.2 Introduction ...... 105

5.3 Methods ...... 108

5.3.1 Experiment design ...... 108

5.3.1.1 Experiment 1, OsHV-1 growth curve ...... 108

5.3.1.2 Experiment 2, temperature increase ...... 109

5.3.2 Aquarium management...... 112

5.3.3 Injection with OsHV-1 ...... 112

5.3.4 Quantification of OsHV-1 DNA ...... 112

5.3.5 Statistical analysis ...... 112

5.3.5.1 Experiment 1 ...... 112

5.3.5.2 Experiment 2 ...... 114

5.4 Results ...... 114

5.4.1 Experiment 1 – growth curve ...... 114

5.4.2 Experiment 2 - increasing temperature ...... 117

5.5 Discussion ...... 120

6. Chapter 6 - Ostreid herpesvirus 1 (OsHV-1) replication and Pacific oyster (Crassostrea gigas) immune gene expression during the first 24 hours of infection at 18°C and 22°C ..... 125

6.1 Abstract ...... 125

6.2 Introduction ...... 126

6.3 Method ...... 129

6.3.1 Aquarium management...... 129

6.3.2 OsHV-1 challenge ...... 130

6.3.3 Sampling procedure ...... 130

6.3.4 Experiment design ...... 133

6.3.5 Tissue homogenisation ...... 133

6.3.6 Total nucleic acid purification ...... 134

6.3.7 DNase treatment of nucleic acids ...... 134

6.3.8 cDNA synthesis ...... 135

6.3.9 Gene expression qPCR ...... 135

6.3.10 Quantification of OsHV-1 DNA ...... 138

xxi

6.3.11 Statistical analysis ...... 138

6.4 Results ...... 139

6.5 Discussion ...... 149

7. Chapter 7 - Immune conditioning of Pacific oysters (Crassostrea gigas): comparison of infection and immune stimulation treatments for protection against Ostreid herpesvirus 1

(OsHV-1) ...... 155

7.1 Abstract ...... 155

7.2 Introduction ...... 156

7.3 Method ...... 160

7.3.1 Experimental design ...... 160

7.3.2 Oysters ...... 162

7.3.3 Aquarium management...... 162

7.3.4 Immune conditioning and priming ...... 162

7.3.5 OsHV-1 challenge ...... 163

7.3.6 Quantification of OsHV-1 DNA ...... 163

7.3.7 Statistical analysis ...... 163

7.4 Results ...... 165

7.4.1 Immune conditioning ...... 165

7.4.2 OsHV-1 challenge ...... 165

7.5 Discussion ...... 173

xxii

8. General discussion...... 179

8.1 Introduction ...... 179

8.2 Farm Management Alters Mortality Associated with OsHV-1 ...... 180

8.3 Multiple Exposures to OsHV-1 ...... 181

8.4 Water Temperature Affects Pacific Oyster Immune Response ...... 183

8.5 Prior Exposure to OsHV-1 as a Method of Protection from Disease ...... 185

8.6 Conclusion ...... 188

9. References ...... 190

xxiii

List of Tables

Table 1.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) resulting from infection with Ostreid herpesvirus 1 (OsHV-1) at different water temperatures. Data are from studies using different challenge techniques, oyster batches, and in different locations...... 13

Table 3.1: Description of risk factors and outcome variables ...... 60

Table 3.2: Number of levels and baskets counted for the factors included as random effects in the generalised linear mixed model (GLMM)...... 63

Table 3.3: Total cumulative mortality (%) from counts of dead and live Pacific oysters

(Crassostrea gigas) among 732 baskets across 3 bays and 6 farms surveyed. Odds ratio estimates from univariable logistic regression are presented for each level of categorical variables measured following a Pacific oyster mortality syndrome (POMS) outbreak in

Tasmania, January 2016. Reference categories are indicated with an odds ratio of 1. Variables significant at p < 0.25 were considered for multivariable analysis...... 65

Table 3.4: Parameter estimates and odds ratio associated with variables included in the final multivariable model for the number of dead Pacific oysters (Crassostrea gigas) compared to the total number of oysters in a basket...... 71

Table 4.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) after the initial and second challenge with Ostreid herpesvirus 1 (OsHV-1) at a water temperature of either

18°C or 22°C. Data are total cumulative mortality and OsHV-1 prevalence with 95% confidence intervals*...... 91

Table 4.2: Cox proportional hazard model stratified by initial challenge temperature. Only groups of Pacific oysters (Crassostrea gigas) which were subjected to a second Ostreid herpesvirus 1 (OsHV-1) challenge at 22°C were included in the model...... 93

xxiv

Table 4.3: Odds ratios for the detection of Ostreid herpesvirus 1 (OsHV-1) in Pacific oysters

(Crassostrea gigas) to estimate the effect of water temperature, and OsHV-1 challenge and an

OsHV-1 re-challenge on the prevalence of OsHV-1 in oysters which survived a second challenge to OsHV-1 or a negative control inoculum. Data are from a generalized linear mixed model with tank included as a random effect...... 95

Table 4.4: Linear mixed model describing the concentration of Ostreid herpesvirus 1 (OsHV-

1) DNA in Pacific oysters (Crassostrea gigas) at the time of mortality or in apparently healthy survivors 10 days post-challenge. Only individuals which were positive for OsHV-1 when sampled were included in the analysis...... 97

Table 5.1: Parameters for a logistic growth model for the concentration of Ostreid herpesvirus 1 (OsHV-1) DNA per milligram of Pacific oyster (Crassostrea gigas) gill and mantle tissue at 18°C and 22°C during an in vivo growth curve. The model was fitted to log10 transformed OsHV-1 concentration data and the parameters were estimated and presented in the log scale...... 116

Table 5.2: Mortality in Pacific oysters (Crassostrea gigas) from all causes and that due to

Ostreid herpesvirus 1 (OsHV-1) for oysters in Experiment 2. Data for the Days 0-13 after exposure are prior to the change in water temperature while data for Days 14-28 are after the water temperature increase from 18°C to 22°C for Groups 1 and 2. The data for Days 14-28 are the proportion of oysters which survived to Day 14. Data are pooled across 6 replicate tanks in Groups 1 and 2, 2 replicate tanks in Group 3 and 8 replicate tanks in Group 4...... 119

Table 5.3: Generalised linear mixed model (GLMM) analysis of Pacific oyster (Crassostrea gigas) mortality between Days 14 and 28 for the temperature increase experiment. The odds ratio represents the difference in mortality between the treatment groups after the water

xxv temperature was increased from 18°C to 22°C in two of the groups. Tank was included as a random effect in the model...... 120

Table 6.1: Primer sequences used to quantify Pacific oyster (Crassostrea gigas) host immune gene expression by RT-qPCR...... 137

Table 6.2: Number of Pacific oysters (Crassostrea gigas) in each exposure and water temperature group biopsied at each timepoint in the first 24 hours of infection with Ostreid herpesvirus 1 (OsHV-1) or a negative control tissue homogenate. The number of Pacific oysters that died in the 14 days following the exposure is also shown. The hazard ratio reflecting the overall difference in mortality at each temperature was calculated using a Cox proportional hazards model ...... 143

Table 6.3: Linear model assessing the replication of Ostreid herpesvirus 1 (OsHV-1) during the first 24 hours after exposure to OsHV-1. Log10 transformed OsHV-1 DNA concentration in gill biopsies was used as the outcome and time and water temperature as predictors. The interaction between time and water temperature indicates that the rate of increase in OsHV-1

DNA was greater at 22°C compared to 18°C. The adjusted R2 for the model was 0.78 and the

F test on the full compared to the empty model was F = 137.7, df = 3, 116 (P < 0.001). The

OsHV-1 concentrations predicted by the model are shown in Figure 6.4...... 145

Table 7.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) 14 days after challenge with Ostreid herpesvirus 1 (OsHV-1) or a negative control. The challenge was administered 3 weeks after an immune conditioning treatment. The prevalence of OsHV-1 detected by qPCR in survivors 10 days after the OsHV-1 challenge is also shown; there was significantly lower prevalence in the oysters treated with poly I:C compared to oysters treated with ASW (P < 0.05)...... 167

xxvi

Table 7.2: Cox proportional hazard models describing the mortality in the 14 days after a challenge with Ostreid herpesvirus 1 (OsHV-1) for Pacific oysters (Crassostrea gigas) that had received immune conditioning treatments 3 weeks prior to challenge. The group conditioned with sterile artificial sea water (ASW) was used as the reference group. The negative control group was not included in the model...... 169

Table 7.3: Odds ratios for the presence of Ostreid herpesvirus 1 (OsHV-1) DNA in Pacific oysters (Crassostrea gigas) that survived a challenge with OsHV-1 by injection 3 weeks after immune conditioning treatments. A generalized linear model with OsHV-1 DNA detection as the outcome was generates with the immune conditioning treatment as the sole predictor and tank was included as a random effect to account for clustering...... 171

Table 7.4: The concentration of Ostreid herpesvirus 1 (OsHV-1) DNA in each Pacific oyster

(Crassostrea gigas) either at the time of death or at 10 days post challenge in Pacific oysters that survived a challenge with OsHV-1 by injection 3 weeks after the immune conditioning treatments. Data were analysed using two different generalized linear mixed models in R

3.3.5, one for the survivors and one for the mortalities. The outcome of the model was log10 transformed OsHV-1 concentration and the immune conditioning treatments were predictors with tank included as a random effect. The estimate for the intercept variable for each model is the mean log10 transformed viral concentration for the reference group (conditioned with sterile ASW) and the estimate for the other groups are the difference between the mean of that group and the reference group...... 172

xxvii

List of Figures

Figure 1.1: State production of edible oysters in tonnes and value in $AUD between 2010 and

2017, mostly comprising of Pacific oyster (Crassostrea gigas) and Sydney rock oyster

(Saccostrea glomerata) production. Data taken from the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) (ABARES, 2018)...... 1

Figure 1.2: Ostreid herpesvirus 1 (OsHV-1) genome structure modified from Davidson et al.

(2005). Genes that have been examined in gene expression studies are highlighted in green. .. 8

Figure 1.3: Causal diagram identifying the links between risk factors for Ostreid herpesvirus 1 (OsHV-1) mortality in Pacific oysters (Crassostrea gigas), + indicates that the presence or an increase in the risk factor is associated with higher mortality, ± indicates that the risk factor has been reported to both increase and decrease OsHV-1 mortality and - indicates that the presence or an increase in the risk factor is associated with lower OsHV-1 mortality...... 19

Figure 2.1: Diagram of the tanks and recirculation system used to maintain the Pacific oysters

(Crassostrea gigas) during laboratory experiments. B = biofilter unit and air stone...... 36

Figure 2.2: Pacific oysters (Crassostrea gigas) were injected into the adductor muscle with an

Ostreid herpesvirus 1 (OsHV-1) inoculum after relaxation in a magnesium chloride bath..... 40

Figure 3.1: Sampling locations (blocks) corresponding to leases in Pacific oyster

(Crassostrea gigas) growing areas in Upper and Lower Pitt Water (A), Blackman Bay (B).

Also shown are growing areas in which mortality occurred in January-February 2016

(Biosecurity Tasmania, 2016) in the first Pacific oyster mortality syndrome (POMS) outbreak in Tasmania, and the sampling locations for this study which were representative of these. .. 52

xxviii

Figure 3.2: Plastic baskets containing oysters on height adjustable long-lines were the most common Pacific oyster (Crassostrea gigas) growing system used on the farms surveyed in

Tasmania at the time of the initial Pacific oyster mortality syndrome (POMS) outbreak...... 54

Figure 3.3: Sampling locations in each block within each farm; grey scale indicates mortality

% in each basket sampled...... 58

Figure 3.4: Distribution of mortality (%) for baskets of Pacific oysters (Crassostrea gigas) present on farm for different times, reflecting the age of the oysters. Baskets were sampled across 6 farms following an outbreak of Pacific oyster mortality syndrome (POMS), the size category of each basket also shown. Points have been jittered around each time...... 68

Figure 3.5: Distribution of mortality attributed to Pacific oyster mortality syndrome (POMS) across size categories based on the average total shell length of Pacific oysters (Crassostrea gigas) within each basket. Whether or not oysters handled in the 7 days prior to the outbreak is also shown. Points have been jittered around each size category...... 69

Figure 3.6: Predicted mean mortality from the multivariable logistic regression model for

Pacific oysters (Crassostrea gigas) which had been on a grow-out farm for 3 – 24 months at each of 3 growing densities following the outbreak of Pacific oyster mortality syndrome

POMS in Tasmania...... 73

Figure 3.7: (Supplementary) Number of baskets counted in each bay (A) and farm (B) as well as the distribution of the time on farm (C) and average length by basket (D)...... 78

Figure 3.8: (Supplementary) Number of Pacific oysters (Crassostrea gigas) per basket sampled in each size category...... 79

Figure 3.9: (Supplementary) (A) Mortality at basket level (number of dead oysters / total number of oysters) and (B) basket level mortality on each of Pacific oyster (Crassostrea

xxix gigas) 6 farms surveyed following a Pacific oyster mortality syndrome POMS outbreak in

Tasmania, January – February 2016...... 80

Figure 3.10: (Supplementary) Distribution of age within each size category. Points have been jittered around each size category...... 81

Figure 4.1: Water temperature profile for each treatment group, showing the acclimation period prior to the first challenge with Ostreid herpesvirus 1 (OsHV-1) or the negative control inoculum, and follow up periods before and after the second challenge. The total number of

Pacific oysters (Crassostrea gigas) in each group (N) were evenly distributed between the number of replicate tanks used in the group...... 87

Figure 4.2: Kaplan-Meier survival curves for Pacific oysters (Crassostrea gigas) challenged with Ostreid herpesvirus 1 (OsHV-1) at a water temperature of 22°C after surviving a prior challenge with OsHV-1 or a negative control inoculum 35 days previously (p< 0.001, log rank test)...... 92

Figure 4.3: Ostreid herpesvirus 1 (OsHV-1) DNA copies per mg of Pacific oyster

(Crassostrea gigas) gill and mantle in individuals where OsHV-1 was detected by qPCR at the time of sampling, either upon mortality or at 10 days after a second challenge. The second challenge was either OsHV-1 inoculum (A) or an OsHV-1 free inoculum (B). The boxes show the 1st and 3rd quartiles and the horizontal line indicates the median. The whiskers show the range where less than the 1.5 times the inter quartile range (IQR). Any sample further than 1.5 times the IQR is shown as an individual point...... 98

Figure 4.4: Estimated mean Ostreid herpesvirus 1 (OsHV-1) DNA copy number per mg tissue in surviving Pacific oysters (Crassostrea gigas) 10 days after a second challenge with

OsHV-1 at 22°C showing the interaction between water temperature during the initial

xxx challenge and the initial challenge inoculum in a linear mixed model (point estimate ± standard error)...... 99

Figure 5.1: Water temperature profile and inoculum used for each treatment group in the temperature change experiment (Experiment 2)...... 111

Figure 5.2: Growth curve for Ostreid herpesvirus 1 (OsHV-1) in Pacific oyster (Crassostrea gigas) gill and mantle over 3 days post injection at 18°C and 22°C. At each time point four oysters at each temperature were sampled and the concentration of OsHV-1 was measured by qPCR. Data are the OsHV-1 concentration of OsHV-1 was measured by qPCR. Data are the

OsHV-1 concentration and the lines represent the non-linear growth curve model...... 117

Figure 6.1: The location of the gill and mantle tissue biopsy taken from sampled Pacific oysters (Crassostrea gigas) for Ostreid herpesvirus 1 (OsHV-1) quantification...... 131

Figure 6.2: Experiment design indicating the sampling time points and sample size for each group...... 132

Figure 6.3: Boxplot of the Ct values for each Pacific oyster (Crassostrea gigas) gene examined, including the housekeeper gene (EF1- α) across each exposure, water temperature, mortality outcome and sampling timepoint. The * indicates where a Kruskal-Wallis test indicated a significant difference (P < 0.05) in Ct values for a gene across factors: OsHV-1 exposure; mortality outcome at the completion of the trial; water temperature; or sampling timepoint...... 142

Figure 6.4: Kaplan-Meier survival curve showing the mortality in Pacific oysters

(Crassostrea gigas) challenged with Ostreid herpesvirus 1 (OsHV-1) or a negative control at each water temperature. The lines for the negative control group at 18°C and 22°C overlap as there was no mortality in either group...... 144

xxxi

Figure 6.5: Concentration of Ostreid herpesvirus 1 (OsHV-1) DNA per mg of Pacific oyster

(Crassostrea gigas) gill tissue in the first 24 hours of infection with OsHV-1. The data points indicate the value for each sampled oyster and the lines represent the predicted concentration at each timepoint for each water from a linear model of the log10 transformed data. The gradient of each line (see Table 6.3) represents the replication rate of OsHV-1 at each water temperature. The difference in replication rate between the two water temperatures was significant (P < 0.001)...... 146

Figure 6.6: Expression of 3 Pacific oyster (Crassostrea gigas) immune genes following challenge with Ostreid herpesvirus 1 (OsHV-1). Gene expression in gill biopsies was normalised to EF- α as a housekeeper and the fold-change was calculated compared to oysters injected with a negative control and sampled at matched time points between 0 and 24 hours after exposure. Data are split based on mortality from the OsHV-1 exposure in the 14-day follow-up period. The number of oysters out of the 12 oysters that were sampled at each timepoint that died following the challenge are indicated in Table 6.2. The experiment was conducted at two different water temperatures, 18°C (left) and 22°C (right). Grey squares indicate that no oysters were tested from the group due to either 100% mortality or 100% survival...... 148

(Crassostrea gigas) and the number of replicate tanks is shown next to each treatment group.

xxxii

Samples were taken to measure the OsHV-1 prevalence and viral load immediately prior to the OsHV-1 challenge and at the end of the trial all surviving oysters were sampled...... 161

Figure 7.2: Kaplan-Meier survival curve for an Ostreid herpesvirus 1 (OsHV-1) challenge in

Pacific oysters (Crassostrea gigas), 3 weeks after administration of immune conditioning treatments, including controls for immune conditioning...... 168

Figure 7.3: Ostreid herpesvirus 1 (OsHV-1) DNA concentration in Pacific oysters

(Crassostrea gigas) when challenged 3 weeks after receiving immune conditioning or control treatments (a) at the time of death; (b) in surviving oysters at 14 days post challenge (Day 35).

The oysters conditioned with poly I:C are not displayed here as they all survived to the end of the trial period and OsHV-1 DNA was not detected in any oyster in that group. The box is bounded with the 1st and 3rd quartile of the data and the black line indicates the mean. The whiskers extend to the maximum and minimum of the data except where data greater than 1.5 times the interquartile range from the mean are indicated by a point...... 170

xxxiii

Abbreviations

°C Degrees centigrade

µg Micrograms

µL Microlitre

ANCOVA Analysis of covariance

AS/NZS Australian Standard / New Zealand Standard

ASI Australian Seafood Industries

ASW Artificial Sea water

AUD Australian dollars

AVNV Acute Viral Necrosis Virus

BLOQ Below the limit of quantification bp base pairs cDNA Complementary deoxyribonucleic acid

CFU Colony Forming Units

CI Confidence Interval

Ct Cycle threshold

CVO Chief Veterinary Officer

CyHV 3 Cyprinid herpesvirus 3

xxxiv

DNA Deoxyribonucleic acid

DPIPWE Department of Primary Industries, Parks, Water and Environment dsDNA Double stranded deoxyribonucleic acid dsRNA Double stranded ribonucleic acid

EDTA Ethylenediaminetetraacetic acid

EFSA European Food Safety Authority

EU Endotoxin units g g-force g grams

G Needle gauge

GLM Generalized linear model

GLMM Generalized linear mixed model

GMO Genetically modified organism

HPE Host, pathogen and environment

HR Hazard ratio

HSV Human alphaherpesvirus 1

IQR Inter quartile range

Kg Kilograms

xxxv kPa Kilopascals

L litres

LAT Latency associated transcript

M Molar

Mg Milligrams mL Millilitres mm Millimetres mRNA Messenger ribonucleic acid mS Millisieverts

NATA National Association of Testing Authorities ng Nanogram nM Nanomolar

NSW New South Wales

OIE World Organisation for Animal Health

OR Odds ratio

ORF Open reading frame

OsHV-1 Ostreid herpesvirus-1 microvariants

PAMPs Pathogen associated molecular patterns

xxxvi

PC2 Physical containment level 2

PCD Programmed cell death

PCR Polymerase chain reaction poly I:C Polyinosinic:polycytidylic acid

POMS Pacific Oyster Mortality Syndrome ppm Parts per million ppt Parts per thousand

PRRs Pathogen recognition receptors qPCR Quantitative PCR

RDS Research Data Store

REML Restricted maximum likelihood

RISC Ribonucleic acid induced silencing complex

RNA Ribonucleic acid

RNAi Ribonucleic acid interference

RT Reverse transcriptase s Seconds

SA South Australia

SE Standard error

xxxvii ssRNA Single stranded ribonucleic acid

TAN Total ammonia nitrogen

TAS Tasmania

TLRs Toll-like receptors

Tm Melting temperature v Volume w Weight

WSSV White spot syndrome virus

xxxviii

1. Chapter 1 – A review of the influence of farming practices on the

environment experienced by Pacific oysters (Crassostrea gigas) and

implications for the control of Ostreid herpesvirus 1 (OsHV-1)

1.1 Introduction

1.1.1 Pacific oyster aquaculture in Australia

The aquaculture of Pacific oysters (Crassostrea gigas) in Australia began in the 1940s following their introduction to Australia and despite being an introduced species they were farmed because of their fast growth and because of the difficulty of farming Sydney Rock oysters (Saccostrea glomerata) in QX disease (infection with Marteilia sydneyi) endemic areas (Nell, 2002; Pollard and Hutchings, 1990). Edible oyster aquaculture is a major industry in the coastal estuaries and bays of New South Wales (NSW), South Australia (SA) and

Tasmania (TAS) with a $AUD 112 million farm gate price in 2016-2017 (Figure 1.1)

(ABARES, 2018).

13750 112500

12500 100000

11250 Area 87500 Area 10000 AUS 75000 AUS 8750 NSW NSW 62500 7500 SA SA 50000 6250 TAS TAS 5000 37500

3750 Edible oyster production ($ '000) ($ production oyster Edible

Edible oyster production (tonnes) production oyster Edible 25000 2500 2010- 2011- 2012- 2013- 2014- 2015- 2016- 2010- 2011- 2012- 2013- 2014- 2015- 2016- 2011 2012 2013 2014 2015 2016 2017 2011 2012 2013 2014 2015 2016 2017 Year Year

Figure 1.1: State production of edible oysters in tonnes and value in $AUD between 2010 and

2017, mostly comprising of Pacific oyster (Crassostrea gigas) and Sydney rock oyster

(Saccostrea glomerata) production. Data taken from the Australian Bureau of Agricultural and Resource Economics and Sciences (ABARES) (ABARES, 2018).

Oysters, including Pacific oysters also play important roles in estuarine ecosystems and protecting the coastline by forming oyster reefs (Dumbauld et al., 2009; Sarà and Mazzola,

1997). Oyster populations, including farmed oysters, help to maintain estuarine water quality by filtering large volumes of water in the estuary and providing a habitat for fish (Dumbauld et al., 2009; Forrest et al., 2009). However, the structures used in oyster farming have a negative impact by shading sea grass (Dumbauld et al., 2009) and can also increase sedimentation below the racks or baskets (Forrest et al., 2009). Pacific oysters are not native to Australia and are classed as a feral pest species for which control programs are in place to limit their spread (Fisheries Management Act 1994). This impacts the Pacific oyster aquaculture industry because farmers are restricted to using triploid oysters, which have three sets of chromosomes, in some locations. Triploid oysters are considered to be reproductively sterile and not able to contribute to populating estuaries (Nell, 2002, Fisheries Management

Act 1994).

1.1.2 Farming infrastructure

Pacific oysters naturally settle on fixed structures, often oyster reefs, during their development and then grow onto the adult phase. In Pacific oyster aquaculture, hatchery larvae are produced so that they can be handled individually instead of forming reef structures. In hatcheries the oysters are grown in flow-through or recirculating aquaculture systems and are fed intensively produced algae. Algae production is done on-site at the hatchery with a number of species that can be mixed to form the diet for the oysters. Pacific oysters are sold from the hatchery at various sizes ranging from of 2 mm to 20 mm. They are then moved onto 2 a variety of different types of farming infrastructure in open estuary systems to grow up to a marketable size in open estuary systems. In Australia, rack, basket, and long line infrastructure are most commonly used, with many farms using a variety of different growing infrastructure depending on the farm location and the age and size of the oysters. Rack systems are a vertical post and horizontal rail structure with oysters held in semi-ridged baskets held between the rails. Long line systems use posts to hold up a flexible line, baskets containing oysters are then suspended from the line. Many farms use a variety of farming infrastructure across the farm.

1.1.3 Multifactorial disease in oyster aquaculture

Disease in aquatic animals is often multifactorial with the right combination of host, pathogen and environmental factors required for a disease to emerge. Mass mortality events, likely caused by emerging diseases, have historically impacted the oyster industry with the shifts in production species in France from Crassostrea angulata to Ostrea edulis to Crassostrea gigas each occurring through the 20th century (Goulletquer and Heral, 1997). Emerging diseases also impact the natural environment and wild fish populations making disease control important for maintaining the sustainability of aquaculture which impacts production efficiency and food security (Murray and Peeler, 2005). Understanding the interactions between the host, pathogen and environment allows all three areas to be used to inform integrated, multifactorial control strategies (Pernet et al., 2016).

The pathogens affecting oysters vary between geographic locations and oyster species.

Commensal or environmental bacteria such as Vibrio spp. and Rickettsia-like-organisms are common opportunistic pathogens in multiple oyster species and Nocardia crassostreae has been reported in Pacific oysters (Elston, 1993; Friedman and Hedrick, 1991; Travers et al.,

2015). In Pacific oysters, Vibrio aestuarianus has been associated with mortality in Europe

3 since 2001 (Barbosa Solomieu et al., 2015; Travers et al., 2015) and has also been isolated in the absence of mortality, suggesting that the environmental conditions contribute to mortality

(Barbosa Solomieu et al., 2015; King et al., 2019). Vibrio spp. were also suggested as a cause for some summer mortality events (Schikorski et al., 2011b; Segarra et al., 2010). The presence of Vibrio spp. in association with oyster mortality events is inconclusive as many species were commensals or opportunistic pathogens which caused mortality in a previously compromised host, suggesting multifactorial disease. Further epidemiological studies are required to show causality and identify the specific conditions under which Vibrio spp. cause disease. In these diseases a combination of factors including the host species, pathogens and an environment conducive to disease mortality was required together to result in mortality.

Changes to farming infrastructure have historically been effective in controlling pathogens in oysters. The introduction of shell borers or spionid polychaete mud worms (Polydora spp. and

Boccardia spp.) to Australia probably occurred around 1880 as a result of importing oysters from New Zealand. This resulted in mass mortality events in Sydney rock oysters (Saccostrea glomerata) between 1880 and 1900 was caused by ‘worm disease’ in estuaries where imported oysters were located (Ogburn et al., 2007). Mud worm disease has been subsequently controlled by raising the height of growing infrastructure so that the oysters to at least 0.5m above the mud substrate. Increasing the growing height reduced exposure to the parasites and reduced the fitness of the mud worms on the outside of the shell due to drying

(Handley and Bergquist, 1997).

The aim of this literature review is to identify how farming practices can influence the environment experienced by Pacific oysters. This has important implications for the nature of exposure to Ostreid herpesvirus 1 OsHV-1 and the response of the oysters to infection. This

4 review will focus on the environmental factors that affect the pathogenesis of disease caused by OsHV-1 and identify how these factors can be manipulated in farming systems to influence interactions between Pacific oysters and OsHV-1.

1.2 OsHV-1

1.2.1 Emergence

Ostreid herpesvirus 1 is an emerging pathogen which threatens the Pacific oyster aquaculture industry (Pernet et al., 2016). A combination of the widespread distribution of Pacific oysters, changes in the OsHV-1 genome and conducive farming practices have created conditions which have allowed OsHV-1 to emerge as a multifactorial disease on a global scale. Herpes- like viruses were first observed in Eastern oysters (Crassostrea virginica) in 1970 by electron microscopy of individuals which died during an experiment (Farley et al., 1972).

Herpesviruses associated with mortality of Pacific oysters were later identified and isolated from outbreaks of mortality in Pacific oyster spat during summer in France in 1992 and 1993

(Le Deuff and Renault, 1999; Renault et al., 1994). Herpesvirus-like particles have been identified alongside Vibrio spp. following summer mortality events, suggesting that summer mortality probably consisted of mortality events caused by several different pathogens

(Schikorski et al., 2011b; Segarra et al., 2010). The microvariant genotype of Ostreid herpesvirus 1 (OsHV-1) was detected in France in 2008 and identified as the cause of severe mortality events during the summer which caused mass mortality of up to 100% in Pacific oysters (Crassostrea gigas) (EFSA, 2010; Schikorski et al., 2011a; Segarra et al., 2010). A similar emerging disease associated with mass mortalities in the Chinese scallop (Chamys farreri) in China was caused by a virus called acute viral necrosis virus (AVNV) (Ren et al.,

2013). AVNV was found to be 97% similar to OsHV-1 based on the full genome sequence

(Ren et al., 2013).

Microvariant genotypes of OsHV-1 are an emerging pathogen that appears to have spread and caused disease in both farmed and wild Pacific oysters. Since 2008, OsHV-1 has been detected in association with mass mortality in Pacific oysters in Ireland (Clegg et al., 2014;

EFSA, 2010; Peeler et al., 2012), Spain (EFSA, 2010; Roque et al., 2012), Netherlands

(Gittenberger et al., 2016), New Zealand (Keeling et al., 2014) and in Australia (Jenkins et al., 2013; Paul-Pont et al., 2014). In Australia, mass mortality associated with OsHV-1 in farmed Pacific oysters first occurred in the Georges River, NSW, in November (spring) 2010

(Jenkins et al., 2013) followed by the Hawkesbury River in January 2013 (Paul-Pont et al.,

2014). The disease was termed Pacific Oyster Mortality Syndrome (POMS) and caused mortalities of up to 100% in oyster spat and up to 60% in larger oysters (50 – 60 mm).

Recurrent seasonal disease has occurred in the Georges River and the Hawkesbury River

(Paul-Pont et al., 2014, Whittington et al., 2019). The economic losses associated with recurrent mortality events which occurred in the warmer seasons of each year led to cessation of Pacific oyster production in both of these river systems as farming wasn’t economically viable (Paul-Pont et al., 2014). OsHV-1 was detected in Tasmania (TAS) in 2016 and caused mass mortality in farmed Pacific oysters in the waterways near Hobart including Upper and

Lower Pittwater, Blackman Bay and Pipeclay Lagoon (de Kantzow et al., 2017; Ugalde et al.,

2018). In 2018, OsHV-1 was detected during surveillance of oysters on a barge and in feral oysters in Port Adelaide, South Australia (SA) (Animal Health Australia, 2019).

1.2.2 Virology

Ostreid herpesvirus 1 (OsHV-1) is the type species of the genus Ostreavirus in the family

Malacoherpesviridae, Order Herpesvirales (Davison et al., 2011, 2005; Siddell et al., 2011).

The OsHV-1 capsid has a triangulation number of T = 16 and is approximately 116nm in diameter, within an outer envelope (Davison et al., 2005). The genome (Figure 1.2) is composed of double stranded DNA which was 207.4 kilobase pairs (kbp) in length for the

OsHV-1 reference genome (GenBank Accession: AY509253). This was predicted to encode

124 genes, 12 of which were duplicated in inverted repeats at each end of the genome

(Davison et al., 2005).

Figure 1.2: Ostreid herpesvirus 1 (OsHV-1) genome structure modified from Davidson et al.

(2005). Genes that have been examined in gene expression studies are highlighted in green.

1.2.3 OsHV-1 genetics

Genotypic variants of OsHV-1 called OsHV-1 microvariants were identified from samples taken in 2008 during outbreaks of high mortality disease (Segarra et al., 2010). The OsHV-1 microvariant (µVar) was initially characterised using a 709 base pair (bp) section of the C region of the genome (ORF4) which included a 12 base pair microsatellite deletion and two non-synonymous substitutions in the ORF4 coding region (Segarra et al., 2010). OsHV-1

µVar sequence also contained one deletion and one synonymous substitution in a 607 bp section of the IA region (ORF43) compared to the OsHV-1 reference isolate (Segarra et al.,

2010). Related microvariant genotypes were identified in France between 2008 and 2010 with

15bp or 9bp deletions instead of a 12bp deletion in the C region and a different range of polymorphisms have been identified through sequencing of the C region of the genome

(Martenot et al., 2015b, 2013, 2012, 2011). The OIE defined OsHV-1 microvariants as any

OsHV-1 genotype with sequence variations in the microsatellite locus upstream of the open reading frame (ORF) 4 and in ORF4 and ORF42/43 when compared with the OsHV-1 reference sequence (OIE, 2016). Between 2008 and 2010 the microvariants genotype became the predominant genotype of OsHV-1 in France (Martenot et al., 2015b, 2011). Distinct genotypes of OsHV-1 have been identified in different geographic locations using both full and partial genome sequencing (Arzul et al., 2001a, 2001b; Bai et al., 2019; Batista et al.,

2015, 2007; Burioli et al., 2017; Ren et al., 2013). There was 93% nucleotide sequence similarity across the full genome between European isolates and isolates from China, including OsHV-1-SB which was isolated from a different species of bivalve (Scapharca broughtonii) and Acute Viral Necrosis Virus (AVNV) isolated from Chlamys farreri (Bai et al., 2019; Xia et al., 2015). The isolate sequenced following the initial outbreak in Australia in

2012 was the microvariants genotype (Jenkins et al., 2013). The significance of polymorphisms in the C and IA regions of the genome in determining the virulence of OsHV-

1 or infectivity for a particular bivalve species have not been determined. Additionally, the majority of bioinformatically predicted genes have not been further characterised (Bai et al.,

2019; Burioli et al., 2016). Further characterisation of predicted OsHV-1 genes would allow a better understanding of how polymorphisms are likely to affect virulence and host specificity.

Despite the large variety of genotypes discovered, there have been no direct measurements of the mutation rate (Bai et al., 2019). However, long term surveillance of OsHV-1 outbreaks in

NSW, Australia, suggest that OsHV-1 may have undergone attenuation, based on a reduction in mortality (Whittington et al., 2019).

1.3 OsHV-1 Epidemiology

Describing the interactions between the host, pathogen and environment provides a way of understanding the risk factors for a disease and how they influence the incidence and severity of disease caused by a pathogen (Thrusfield, 2007). Modifying risk factors for disease severity can reduce mortality during an outbreak and together with identification of conditions conducive to transmission of OsHV-1, provide the basis for developing a disease control strategy. Many of the risk factors for mortality caused by OsHV-1 have been identified as environmental, such as water temperature, or are intrinsic to the oyster, such as age and size.

Outbreaks of mortality associated with OsHV-1 infection occur in the seasons with warmer water; in Australia outbreaks have occurred from October to May (Whittington et al., 2019). A well defined risk factor is an increase in average daily water temperature to above 18°C to

20°C in Australia or 16°C in Europe (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a; Pernet et al., 2012; Petton et al., 2015a; Renault et al., 2014; Whittington et al., 2019). Time periods

10 where the water temperature is above these thresholds can occur multiple times over summer in endemic estuaries (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a; Pernet et al., 2012;

Petton et al., 2015a; Renault et al., 2014; Whittington et al., 2019). Experimental infection studies have shown that both the incubation period and the total cumulative mortality depend on the exposure dose and the water temperature (de Kantzow et al., 2016; Paul-Pont et al.,

2015). The incubation following an injection challenge period before mortality ranges between 24 hours at a constant water temperature of 26°C to 96 hours at 18°C (de Kantzow et al., 2016; Petton et al., 2013). Following a cohabitation challenge the incubation period ranges from a minimum of 48 hours at 22°C to 144 hours at 14°C (Petton et al., 2013; Schikorski et al., 2011a). Under field conditions, the water temperature can be variable over the incubation period resulting in greater variability in the incubation period (Paul-Pont et al., 2014; Petton et al., 2015a).

Under controlled laboratory conditions, oysters have shown no signs of disease when exposed to OsHV-1 at 14°C, low mortality (< 25%) at 18°C and mortality exceeding 90% at 26°C

(Table 1.1) (de Kantzow et al., 2016). Using a laboratory to control the conditions removes the impact of variability and confounders that are present in field conditions, such as growth rate and feeding. At a water temperature of 22°C the incubation period was increased from 96 hours when oysters were fed twice daily to 144 hours when the oysters were not fed (Evans et al., 2015). Mortality is dose dependent when challenged by injection and cohabitation with

OsHV-1 infected oysters (de Kantzow et al., 2016; Evans et al., 2015; Paul-Pont et al., 2015,

2014). The concentration of OsHV-1 DNA was higher at increased water temperature for oysters which died following OsHV-1 exposure as well as those which survived (de Kantzow et al., 2016; Petton et al., 2013). The mortality caused by OsHV-1 was reduced at a water

11 temperature of 28°C compared to 26°C (Delisle et al., 2018; Petton et al., 2013). Low water temperatures of 10°C to 13°C arrest mortality in oysters exposed at 22°C, suggesting that water temperature changes following the initial exposure are important in determining the pattern of mortality (Pernet et al., 2015; Petton et al., 2013).

Table 1.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) resulting from infection with Ostreid herpesvirus 1

(OsHV-1) at different water temperatures. Data are from studies using different challenge techniques, oyster batches, and in different locations.

Challenge Water Total cumulative Comment Reference temperature (°C) mortality (%) Laboratory, 14 0 (de Kantzow et al., 2016) Injection 18 23 (de Kantzow et al., 2016) 20 63-88 Range between trials (Hick et al., 2016) 22 77 (de Kantzow et al., 2016) 70, 65, 90 3 different homogenates were tested (Schikorski et al., 2011b) 87-100 Control, range between replicates (Martenot et al., 2015a) 40-93 Inoculum held at 16°C for 48-54 hours (Martenot et al., 2015a) 7-53 Inoculum held at 16°C for 30 hours (Martenot et al., 2015a) 73 Inoculum held at 25°C <33 hours (Martenot et al., 2015a) 13 ±6 Mean +- SE (Green et al., 2015b) 90 (Schikorski et al., 2011a) 26 84 (de Kantzow et al., 2016) Laboratory, 20 50-71 Range between trials (Hick et al., 2016) Immersion Laboratory, 13 2 (Petton et al., 2013) cohabitation 14 18 (Petton et al., 2013) 15 18 (Petton et al., 2013) 16 30 (Petton et al., 2013) 18 30 (Petton et al., 2013)

20 21-50 Range between replicates (Petton et al., 2013) 22 72-100 Range between replicates (Evans et al., 2015) 50 (Schikorski et al., 2011a) 35 (Petton et al., 2013) 27 20 Range between replicates (Petton et al., 2013) 29 17 Range between replicates (Petton et al., 2013) Field exposure 17 50-68 Range between replicates (Petton et al., 2013) 18 62 (Petton et al., 2013) 20 75-90 Range between replicates (Petton et al., 2013) 21 87 Field challenge at 18°C, increased to 21°C (Pernet et al., 2015)

OsHV-1 has been detected by qPCR in Pacific oysters where the water temperature is above

20°C without an associated mortality event, suggesting subclinical infection can occur

(Whittington et al., 2019). Comorbidity from infection with OsHV-1 and Vibrio spp. has been suggested to explain why OsHV-1 does not always cause mortality (Azéma et al., 2016; de

Lorgeril et al., 2018; Pernet et al., 2012; Petton et al., 2015b, 2013). The presence of high concentrations of Vibrio spp. immediately prior to mortality due to OsHV-1 suggest that infection with OsHV-1 may cause immune suppression and the resulting bacteraemia may be the direct cause of mortality (de Lorgeril et al., 2018; Petton et al., 2015b). Vibrio spp. are present in the surrounding environment and make up a small part of the oyster’s normal microbiome as commensals (Pathirana et al., 2019). The population of commensal Vibrio spp. changes depending on the location and environment where the oysters are grown (Barbosa

Solomieu et al., 2015; Pathirana et al., 2019; Petton et al., 2015b). Further research is required to understand if increasing concentrations of Vibrio spp. following infection with OsHV-1 are the cause of mortality or reflect dysbiosis in a disease pathogenesis that would be fatal regardless. Water temperature also affects the growth of Vibrio spp. suggesting their role in oyster disease may change with water temperature (Motes et al., 1998; Petton et al., 2015b).

The first sign of an outbreak is unusually high mortality with affected oysters also gaping or leaking water when brought to the surface (Clegg et al., 2014; EFSA, 2010; Gittenberger et al., 2016; Jenkins et al., 2013; Peeler et al., 2012; Roque et al., 2012). OsHV-1 causes higher mortality in spat and juvenile oysters compared to adult oysters and greater mortality in smaller oysters compared to larger oysters (Azéma et al., 2017; Dégremont, 2013; Hick et al.,

2018; Paul-Pont et al., 2014). There was 20% higher average mortality in small (0.43 g) oysters compared to large (4.73 g) oysters following field exposure to OsHV-1 in an endemic

15 estuary (Azéma et al., 2017). However, age and size are often confounded in observational studies and are connected by the growth rate of the oyster. A protective effect of increasing age against mortality is dependent on the size of the oyster (Azéma et al., 2017; Dégremont,

2013; Paul-Pont et al., 2014). However, deliberately manipulating the size did not change the level of mortality due to OsHV-1 in a field challenge (Hick et al., 2018). Spat under 12 months of age are still likely to experience very high mortality even with a protective size effect (Hick et al., 2018; Petton et al., 2015a). The annual production calendar as part of an integrated program to reduce the overall impact of OsHV-1 (Carrasco et al., 2017). The efficacy of combining changes to the production calendar with manipulating the size of oysters relative to their age under commercial production conditions requires further investigation.

The farm management decisions that affect the risk of OsHV-1 mortality includes the growing structure and location, the stocking density and the handling procedures. Increasing the height of intertidal growing structures reduced the mortality of adult oysters but not spat, possibly due to reduced exposure to OsHV-1 (Paul-Pont et al., 2013a; Whittington et al., 2015a). Low density growing structures such as rope culture may have provided some protection from

OsHV-1 mortality in France (Pernet et al., 2012). A high density of spat in a single location was associated with higher mortality than areas with a mix of adults and spat (Pernet et al.,

2014a, 2012). Machine handling during grading is associated with reduced mortality compared to manual grading, possibly because manual handling is rougher (Peeler et al.,

2012).

The oyster’s genotype, following selective breeding of family lines, had a direct effect on susceptibility to OsHV-1 and on the mortality following an outbreak and also affected the

16 growth rate of the oysters (Pernet et al., 2012). Oysters breed from pedigrees selected for an

OsHV-1 survival phenotype had increased resistance to OsHV-1 following a controlled laboratory challenge; however, specific genes or genotypes which are associated with resistance have not been identified (Degremont et al., 2015). Although genotype has an effect on mortality, the ploidy of oysters did not influence the level of mortality due to OsHV-1

(Dégremont et al., 2016; Peeler et al., 2012).

Energy reserves, measured as carbohydrate and triglyceride content in oyster tissue samples, are correlated with greater feed availability, which in turn, may affect OsHV-1 transmission

(Evans et al., 2016, 2015; Pernet et al., 2019, 2014b). Oysters with higher energy reserves had a lower risk of OsHV-1 mortality in a laboratory cohabitation trial (Pernet et al., 2019). This suggests that higher mortality due to OsHV-1 and more energy reserves in dead oysters compared to survivors may both be due to greater feed availability resulting in more exposure to OsHV-1. Oysters which were exposed to OsHV-1 had a greater energy demand compared to unexposed oysters, possibly due to the energy required to mount an immune response

(Tamayo et al., 2014).

Water temperature, age, size, genetics and the growing conditions are all factors relating to the host, pathogen and environment which affect the level of mortality due to OsHV-1 (Figure

1.3). The combination of these factors which are required for OsHV-1 to cause high mortality in Pacific oysters makes mortality due to OsHV-1 a multifactorial disease. Currently, understanding how the host, pathogen and environmental risk factors contribute to OsHV-1 disease is limited by confounding and interactions between these factors in field studies. To overcome this limitation, laboratory experiments in controlled conditions are required to assess the effect of individual risk factors such as environmental parameters on mortality

17 caused by OsHV-1 in Pacific oysters. This approach has been used to examine the effect of pH , salinity and food availability (Evans et al., 2015; Fuhrmann et al., 2018).

+ High feed OsHV-1 availability Exposure +

Previous Water + OsHV-1 - Temperature + Exposure

Stocking density ± - ± Growing structure - height -

± Production ± calendar Size OsHV-1 - Mortality ±

Age + ± - Production Genotype

Resistance - Genotype

1.4 OsHV-1 Transmission

The pattern of mortality caused by OsHV-1 is very patchy at the tray or basket level, indicating that the epidemic does not propagate through the population within these growing units (Paul-Pont et al., 2014; Whittington et al., 2018, 2015a). Outbreaks of mortality due to

OsHV-1 have followed a point source epidemic pattern, consistent with all the farmed oysters being exposed at once from an environmental reservoir as the main source of transmission initiating an outbreak (Whittington et al., 2018). However, infection by cohabitation under laboratory conditions indicates that horizontal transmission can occur between oysters in close proximity (Evans et al., 2015; Paul-Pont et al., 2015, 2014; Schikorski et al., 2011b;

Whittington et al., 2018). The unique local conditions including feed availability and water temperature affect the efficiency of horizontal transmission in laboratory trials (Evans et al.,

2016, 2015). There was a low prevalence (7.3%) of OsHV-1 in a population of surviving oysters from a previous field outbreak which may have served as a reservoir that initiated subsequent disease outbreaks (Evans et al., 2017). OsHV-1 does not persist in sea water for longer than 2 days, or 7 days in dead oyster tissue (Hick et al., 2016), which is unlikely to be long enough to travel long distances between countries, especially on ocean currents. Bivalve biofouling on shipping and debris floating in the ocean and has been suggested as a possible mechanism for transmission of OsHV-1 between growing areas in different waterways as farmed oysters in endemic areas are often the subject of movement restrictions (Deveney et al., 2017; Gittenberger et al., 2016; Whittington et al., 2018).

1.5 Pacific Oyster Immunology

Pacific oysters live in a unique intertidal estuarine environment where obligate and opportunistic pathogens are constantly present. To thrive in this environment requires an

20 effective immune system to resist infectious disease. The immune system of Pacific oysters utilizes the shell and mucus layers to physically protect the outer surfaces and hemocytes and immune molecules in the hemolymph internally as the functional effectors of the immune system (Allam and Raftos, 2015; Wang et al., 2018). Oysters have an open circulatory system and so the hemolymph and hemocytes are present within the tissues and will concentrate at sites of infection and inflammation through chemotaxis (Allam and Raftos, 2015; Green et al.,

2015a; Green and Speck, 2018). Unlike mammals, invertebrates do not have T-lymphocytes or produce antibodies to recognize and bind to antigens associated with specific pathogens

(Allam and Raftos, 2015). The immune system recognizes pathogens through pathogen recognition receptors (PRRs) such as Toll-like receptors (TLRs) (Allam and Raftos, 2015;

Green et al., 2014a). There is evidence that non-specific immune priming occurs in oysters and so they are able to produce a more effective immune response upon a second encounter with a pathogen (Green and Montagnani, 2013; Lafont et al., 2017).

1.5.1 Antiviral immunity

The antiviral immune system of the Pacific oyster uses PRRs to recognize a viral infection which can trigger a variety of responses including programmed cell death, RNA interference

(RNAi) which can selectively target and degrade viral mRNA and antiviral effectors such as viperin to fight the infection (Allam and Raftos, 2015; Green et al., 2015c, 2015a). PRRs recognize pathogen associated molecular patterns (PAMPs) which are motifs conserved between pathogens, these include dsRNA and viral envelope proteins (Allam and Raftos,

2015). A wide array of receptor molecules which recognize pathogen associated molecular patterns (PAMPs) give the mollusc an antiviral response with broad range of effective pathogen recognition (Green et al., 2015a; Martenot et al., 2017; Moreau et al., 2015; Pauletto

21 et al., 2017; Schulenburg et al., 2007). A PAMP such as dsRNA binding to a PRR such as

TLR3 activates an immune signalling pathway which can initiate an antiviral response and release cytokines such as interferon (Green et al., 2015a; Green and Montagnani, 2013).

Interferon and other cytokines can bind to receptors in the cellular membrane of nearby cells and cause Janus kinase (JAK) to phosphorylate which in turn causes signal transducer and activator of transcription protein (STAT) phosphorylation in a homologue of the JAK/STAT pathway which then leads to the transcription of antiviral response genes (Green et al., 2015a,

2014b). The functional outcomes of the immune response to OsHV-1 that have been identified are apoptosis, autophagy, migration of hemocytes to the infected tissue and an increase in the transcription of genes encoding viperin and antimicrobial peptides in infected tissue (Green et al., 2015a, 2014b; Moreau et al., 2015; Picot et al., 2019). Other effectors of the invertebrate immune response including RNA interference (Owens and Malham, 2015; Rosani et al.,

2016), humoral antimicrobial peptides and increased phagocytosis and autophagy may also be activated (Green and Speck, 2018; Moreau et al., 2015; Rowley and Pope, 2012).

The anti-viral immune response of Pacific oysters has mostly been investigated by measuring gene expression in response to OsHV-1 and inferred from exposure to poly I:C using reverse transcriptase quantitative polymerase chain reaction (RT-qPCR) (Green et al., 2016b; Green and Montagnani, 2013; Lafont et al., 2017). Poly I:C (polyinosinic:polycytidylic acid) is a synthetic polymer analogue of dsRNA which binds to PRRs which detect intracellular dsRNA, it is used to artificially induce an antiviral response (Green and Montagnani, 2013;

He et al., 2015; Ma and He, 2014). Many of the target genes measured have been putative immune genes where the function was predicted in-silico based on homogeneity with model species such as mice and humans (Green et al., 2016b; Green and Montagnani, 2013; Lafont

22 et al., 2017). This method has the limitation that it may not identify novel immune mechanisms and using a single virus model to investigate the immune response means the breadth of possible mechanisms may not be identified. Experiments which have used poly I:C have the limitation that it only binds to dsRNA receptors and not the array of PRRs that the

PAMPs associated with a virus or viral infection may bind. Full transcriptome sequencing has been used to capture the full range of genes which were transcribed in OsHV-1 infected oysters, including unknown genes which have not been studied using RT-qPCR (Rosani et al.,

2015). This approach is advantageous because it may be able to identify novel genes which do not have known homologues. Proteomic analysis has been performed after exposure to OsHV-

1 and poly I:C in order to identify which transcribed genes are translated to proteins

(Corporeau et al., 2014; Green et al., 2016a). These studies identified unknown genes and proteins which were being expressed in response to OsHV-1 or poly I:C but did not identify the function (Corporeau et al., 2014; Green et al., 2016a; Rosani et al., 2015). The function of these differentially expressed genes needs to be confirmed in order to further understand the antiviral immune response of the oyster.

1.5.2 Immune memory and immune priming

Immune priming is a process whereby exposure to a pathogen or immunostimulant induces a more effective immune response, resulting in a greater chance of survival if the animal is exposed to a pathogen within the period the priming effect lasts (Green and Montagnani,

2013; Kurtz, 2004; Lafont et al., 2017; Milutinović and Kurtz, 2016; Schulenburg et al.,

2007). Immune memory is more specific than priming and requires the immune system to retain information gained about a pathogen and use the information upon a second exposure to the pathogen (Milutinović and Kurtz, 2016). Invertebrate immune memory was first identified

23 in the copepod Macrocyclops albidus in response to the tapeworm Schistocephalus solidus, this suggests that oysters may have the mechanisms required for immune memory (Kurtz,

2004; Kurtz and Franz, 2003a). Oysters that were immune primed by injection with poly I:C had a greater chance of survival following subsequent exposure to OsHV-1 (Green and

Montagnani, 2013; Lafont et al., 2017). Immune priming with poly I:C also resulted in a lower concentration of OsHV-1 in oysters which survive an OsHV-1 exposure compared to oysters which have not been immune primed (Green and Montagnani, 2013; Lafont et al., 2017).

Oyster spat which were exposed to OsHV-1 in the field and were subsequently maintained at

10°C or 13°C for 64 days experienced similar mortality to naïve oysters when they were exposed to OsHV-1 in the field for a second time, suggesting they were not immune primed

(Pernet et al., 2015). Antiviral immune priming using poly I:C has only been investigated as a method of protecting against mortality due to OsHV-1 and immune memory has not been directly demonstrated in oysters. Both immune priming and immune memory require further investigation to understand their potential for disease control in Pacific oysters.

1.6 Management of Disease Caused by OsHV-1

Multifactorial disease occurs when a combination of risk factors in the host, pathogen and environment combine to provide the conditions which permit disease to occur. Control strategies should take a multifactorial approach and consider a combination of host, pathogen and environmental risk factors and identify those which can be manipulated to reduce the incidence and severity of disease effectively. Additional control strategies have been suggested or evaluated including immune priming brood stock with poly I:C, breeding for resistance to mortality caused by OsHV-1, changing the time of year spat are brought onto the farms, buying larger spat onto the farm and modifying growing structures to raise the growing

24 height. Pacific oysters are farmed extensively but, in high numbers, with approximately 80 million oysters on Tasmanian farms in 2016 (Ugalde et al., 2018), mostly in intertidal racks.

For a disease control strategy to be successful it must be able to be efficiently applied to oyster growing systems in order to have significant uptake among farmers.

1.6.1 Preventing OsHV-1 transmission

Following the detection of OsHV-1 in NSW in 2010, movement restrictions were implemented to prevent the spread of OsHV-1 to other growing areas. The movement restrictions applied to both Pacific oysters and oyster farming equipment (Animal Health

Australia, 2019; Department of Agriculture, 2015; NSW DPI, 2014). Despite these restrictions, there were additional OsHV-1 outbreaks in the Hawkesbury River in 2013, in

Tasmania in 2016, and the detection of OsHV-1 in Port Adelaide in 2018 (Animal Health

Australia, 2019; Jenkins et al., 2013; Paul-Pont et al., 2014; Whittington et al., 2018).

However, the movement restrictions may have slowed the spread of OsHV-1 as it did not spread as quickly as in Europe and the spread of OsHV-1 across the globe suggests that moment restrictions alone are not able to stop the spread as other mechanisms of transmission exist (Whittington et al., 2018).

1.6.2 Modifying farming infrastructure

Modifying the growing structures has allowed farmers to change the oyster’s growing environment in a way that reduces the incidence and severity of disease. Raising the height of oyster racks or baskets by 300mm reduced the mortality due to OsHV-1 but also reduced the growth rate (Paul-Pont et al., 2013a; Whittington et al., 2015a). Modifying basket and longline systems to increase the growing height may be a practical solution in areas of Australia where

25 variable and fixed baskets on long lines are the predominant growing structure where height of the lines may be increased on the existing posts (Ugalde et al., 2018).This approach is limited by the requirement to produce a profitable number of oysters on the available oyster leases which limits the minimum practical stocking density. Oyster spat in upweller facilities can be protected from OsHV-1 by filtering or holding inlet water for 48 hours before it enters the upweller (Whittington et al., 2015b). Part of the effect of raising growing structures and filtering water may be due to reduced feed and OsHV-1 exposure (Evans et al., 2019, 2015).

1.6.3 Manipulating age and size

Reducing mortality by manipulating age and size requires changing the production calendar so that oysters are largest and oldest when they are at the highest risk of exposure to OsHV-1

(Azéma et al., 2017; Carrasco et al., 2017; Hick et al., 2018). The major limitation of this strategy is that outbreaks occur during a window each year when the water temperature rises above 18°C, this doesn’t leave enough time to grow oysters to over 12 months of age when they are afforded a significant level of protection against mortality (Hick et al., 2018). As part of an overall disease control strategy changing the management calendar to maximize the age and size of oysters during the period of increased risk OsHV-1 is likely to effectively lower overall OsHV-1 mortality (Carrasco et al., 2017; Hick et al., 2018). However, this approach may not produce enough oysters of a saleable size during the period of peak demand.

1.6.4 Antiviral immune priming

Antiviral immune priming may be an effective way to reduce mortality, but it requires significant modification to make the technique practical. Priming oysters directly with poly

I:C reduces mortality in juvenile oysters (> 30 mm) but it is not practical to inject hundreds of

26 thousands of oysters used in aquaculture production (Green and Montagnani, 2013; Lafont et al., 2017). Further, the use of poly I:C is toxic to Rhesus monkey, mice and dogs and is unlikely to be acceptable from a food safety standpoint and may have undesirable impacts on estuarine ecosystems (Hartmann et al., 1987; Homan et al., 1972). Brood stock primed with poly I:C have produced large quantities of larvae which were resistant to an OsHV-1 challenge under laboratory conditions (Green et al., 2016b). However, spat from brood stock primed with poly I:C have not been evaluated under field conditions and the length of time the larvae/spat retain relative protection from OsHV-1 is unknown. Limitations to this approach are a lack of understanding of the mechanism of actions, whether side effects such as hindered growth or reproduction or increase susceptibility to bacterial disease occur (Green et al.,

2015b; Green and Montagnani, 2013; Green and Speck, 2018). The combined effect of immune printing and environmental conditions including water temperature and other immune interactions such as with opportunistic bacterial pathogens are not known.

1.6.5 Selective breeding for genetic resistance to OsHV-1

Improving genetic resistance of Pacific oysters to OsHV-1 associated disease has been investigated as a method of reducing the mortality (Dégremont, 2013; Degremont et al., 2015;

Dégremont et al., 2010; Normand et al., 2014b; Sauvage et al., 2009). Selection of successive generations of oysters for resistance to OsHV-1 has been attempted in France and produced family lines which had 66% and 73% survival against a field challenge with OsHV-1 compared to less than 8% survival in unselected control families (Dégremont et al., 2015b).

Increased growth was also observed in the resistant lines compared to the susceptible family lines in the same trial. Laboratory trials have also demonstrated 90% difference in survival between susceptible and resistant family lines following an OsHV-1 challenge (Dégremont et

27 al., 2015a; Segarra et al., 2014c). However, a genome wide association study found a heritability of 0.078 – 0.25 for survival, much lower than the 0.48 – 0.71 reported based on pedigree (Gutierrez et al., 2018). Although the transcriptional response of some known immune genes to OsHV-1 exposure has been examined in resistant and susceptible oysters, the genes which are under selective pressure have not been identified or characterized

(Gutierrez et al., 2018; Segarra et al., 2014c). Additionally, there are physical traits that impact resistance such as size and it is important to assess if selecting for these traits will confer resistance. Commercial oyster stock which have been bred for OsHV-1 resistance have been widely discussed and progressively introduced into commercial production through hatcheries in Australia but the efficacy of the breeding programs have not been independently assessed (Ugalde et al., 2018).

1.7 Conclusion

Novel pathogens can emerge when multiple host, pathogen and environmental factors combine to allow disease to occur. Microvariant genotypes of OsHV-1 emerged as a pathogen of Pacific oysters when multiple environmental factors allowed the interactions between the host and pathogen to result in oyster mortality due to OsHV-1. In Pacific oyster aquaculture, the farming practices and infrastructure influence the environment experienced by the oysters play an important role in determining interactions between the host, pathogen, and environment that lead to this multifactorial disease. These interactions need to be understood in order to develop strategies to reduce the impact of disease. Multifactorial diseases require a multifactorial approach to disease control which takes all aspects of the host, pathogen and environment into account. To build a picture of how risk factors interact to determine mortality, each risk factor needs to be assessed while considering the effect each other risk

28 factor has on mortality. Field investigations have been important for identifying factors which impact mortality due to OsHV-1 in the context of all the host, pathogen and environment interactions and laboratory trials have been used to isolate individual factors from confounders to gain a better understanding of their impact. Water temperature is an environmental factor which is dynamic over time and affects the way the host interacts with the pathogen. This makes further investigation of the effect of water temperature on the effectiveness of control strategies and other host, pathogens and environment factors as well as mortality an important step in understanding how to reduce the mortality caused by OsHV-1. Understanding how the water temperature affects the Pacific oyster and the way it interacts with OsHV-1 may present the basis for developing control strategies for mortality caused by OsHV-1.

1.8 Experimental Approach

The two aims of this thesis are i) to examine the effect of farm management and other risk factors on mortality caused by OsHV-1 during a natural outbreak and ii) to use a laboratory infection model for OsHV-1 that provides control of the environment to examine the specific risk factors and there interactions on mortality due to OsHV-1. By isolating and controlling or statistically accounting for different interactions between the host, pathogen and environment it may be possible to elucidate the effect of each factor on mortality.

2. General Materials and Methods

2.1 Reagents

2.1.1 MilliQ water

MiliQ water was produced by a Milli-Q® Biocel Ultrapure water purification system

(Millipore).

Specifications:

Resistivity at 25°C 18.2 mS

Total organic carbons 5 – 10 µgL-1

Pyrogens < 0.001 EUmL-1

Bacteria < 1 CFUmL-1

Stored at room temperature.

2.1.2 Artificial sea water

Artificial sea water (ASW) was prepared in a 3000 L polyethylene storage tank using:

Salt (Red Sea 15 kg

Municipal tap water 500 L

Mixing and water circulation was maintained by an external electric pump. The salinity was

30 ppt, as measured using a refractometer at 22°C (Red Sea). The refractometer was recalibrated at 0 ppt using municipal tap water prior to each use. ASW was prepared at least

48 hours prior to use to allow residual chlorine in the tap water to dissipate into the

30 atmosphere and equilibration to the room temperature which was maintained using a reverse cycle air conditioner at the intended water temperature for use in experiments.

For use as a laboratory reagent, small volumes of ASW (< 250 mL) were prepared at the same ratio using MilliQ water instead of municipal tap water and filtered to 0.22 µm using a syringe

(Thermo) and syringe filter (Millipore).

2.1.3 Fresh aquarium water

Tap water was dechlorinated by aeration for 48 hours to produce fresh aquarium water or was dechlorinated with sodium thiosulfate to produce fresh aquarium water as follows:

Municipal tap water 76 L

Tap water conditioner (API Fish Care, 30.2% w/w sodium thiosulfate, 9.8% w/w

EDTA tetrasodium salt) 1 mL

The water was stored in 500 L polyethylene tanks and circulated using a pump in an external canister filter (Fluval 206) which contained activated charcoal. The fresh aquarium water was stored at the same temperature as intended use.

2.1.4 Magnesium chloride solution (50 gL-1)

Magnesium chloride hexahydrate (Sigma) 250 g

Fresh aquarium water 5 L

Stored at the same temperature as intended use.

2.1.5 Lysis buffer

RLT lysis buffer was mixed with the antifoaming agent DX for use in mechanical tissue homogenisation.

RLT lysis buffer (Qiagen) 300 mL

DX (Qiagen) 2 mL

β – Mercaptoethanol 98% molecular biology grade (Sigma) 3 mL

Stored at room temperature

2.1.6 Trigene (1:100)

Trigene Advance (MediChem International) 10 mL

Municipal tap water 1 L

Stored at room temperature.

2.1.7 Virkon solution (1%)

The Virkon solution (1%) was prepared immediately before use as a surface disinfectant. For use as a footbath the Virkon solution (1%) was used until it turned colourless or for 7 days, whichever was less.

Virkon® powder (Antec International) 20 g

Municipal tap water 2 L

Stored at room temperature.

2.1.8 Chlorine solution (1000 ppm)

A fresh stock of chlorine solution was made each day.

Pool chlorine, 12.5% available chlorine (Hy-Clor) 40 mL

Municipal tap water 960 mL

Stored at room temperature.

2.2 Aquatic Animal Housing Facility

2.2.1 Biosecurity

All animal and molecular biology experiments were undertaken in a physical containment level 2 (PC2) certified laboratory or aquatic animal facility. These facilities were certified to comply with the guidelines for certification of a PC2 laboratory (version 3.2) or PC2 aquatic animal facility (version 3.1) as set out by the Office of the Gene Technology Regulator. These facilities also comply with Australian standard AS/NZS 2243.3 Safety in laboratories Part 3:

Microbiological safety and containment as required by the University of Sydney institutional biosafety committee for handling risk group 2 organisms (The University of Sydney, 2019).

Genetically modified organisms (GMOs) were handled in accordance with the requirements of the Office of the Gene Technology Regulator and the University of Sydney institutional

Biosafety committee as required by the Gene Technology Act 2000 and the Gene Technology

Regulations Act 2001.

Pacific oysters (Crassostrea gigas) were transported by air from Tasmania (TAS) to New

South Wales (NSW) with a certificate of freedom for movement. Upon arrival to the laboratory the oysters were held in ASW overnight to remove contaminants and inspected for general health. A random sample of 30 oysters was measured with vernier callipers to determine size and tested to determine that the batch was free from OsHV-1. The criteria for a

33 healthy batch of oysters was all closed, no damage to the shell, free from OsHV-1 and good condition in the sampled oysters.

2.2.2 Oysters

Triploid Pacific oyster spat (2240 size, 3 months old) were procured from batches of stock produced commercially for supply of farms by Shellfish Culture, Tasmania and transported to

Patonga Creek, NSW. The oysters were grown to 50 – 60 mm in length under commercial growing conditions in an area which remained free of OsHV-1 during an ongoing active surveillance program. These oysters were then transported to the laboratory for use in experiments. The brood stock for these oysters had been selectively bred for resistance to

OsHV-1 and were expected to have survival 20% better than stock which had not been selectively bred. There is published data on the success of the Australian Seafood Industries

(ASI) Pacific oyster breeding program but their estimated breeding values for OsHV-1 resistance have not been externally verified. However, the use of ASI brood stock is prevalent in Australian Pacific oyster hatcheries and they represent the most prominent genetic background of the hatchery reared Pacific oysters in Australia.

Animal ethics approval was not required for working with Pacific oysters as they do not come under the definition of an Animal for the purposes of the Australian Code for the Care and

Use of Animals for Scientific Purposes which governs the use of animals in scientific research

(National Health and Medical Research Council, 2013). This code defines animals as: any live non-human vertebrate (that is, fish, amphibians, reptiles, birds and mammals, encompassing domestic animals, purpose-bred animals, livestock, wildlife) and cephalopods.

2.2.3 Aquaria

In each laboratory experiment, groups of oysters were maintained in separate tanks within a recirculating system which was used as a water bath to maintain the water temperature. The water in the recirculation system housing each tank and the tanks housing the oysters was not allowed to mix. Each oyster tank contained 12 L of ASW an air stone to aerate the water to maintain optimal dissolved oxygen saturation and a biofilter which removed particulates and ammonia from the water. Up to 30 oysters were housed per tank, the oysters were placed on perforated polycarbonate racks to allow water flow around each oyster.

Four identical recirculation systems, each containing 6 oyster tanks, were maintained within the PC2 Aquatic animal facility at the University of Sydney. Water within the recirculation system was pumped (Uno HMC-75LX, 65 L min-1) out of the sump and into a chiller unit

(DBA-110, Daeil) and then onto the tanks through an inlet manifold above the tanks. The water level within the outer tanks was maintained with a standpipe and the water drained back into the sump through an outlet manifold (Figure 2.1). A thermostatically controlled submersible aquarium heater (3619 Aquarium Heater, Eheim) placed in the sump was used in conjunction with the chiller unit to maintain the water temperature in the recirculating water which also maintained the water temperature of the oyster tanks. The air temperature in the room was maintained by a reverse cycle air conditioner (Mitsubishi Heavy Industries) and was set to 22°C.

The water quality in each of the oyster tanks was maintained by the biofilter and through water changes to keep Total Ammonia Nitrogen (TAN) < 2 ppm and nitrate < 160ppm. The pH was adjusted to 8.2 using sodium bicarbonate (Pureblue) when it was ≤ 7.8.

B B B B B B 1 2 3 4 5 6

Air blower

Sump Heater Chiller

Pump

Figure 2.1: Diagram of the tanks and recirculation system used to maintain the Pacific oysters

(Crassostrea gigas) during laboratory experiments. B = biofilter unit and air stone.

2.2.4 Random allocation

Oysters were randomly allocated to a tank at the beginning of the acclimation period or at the time of inoculation. A list of tank numbers was generated in excel such that the tank number was repeated the same number of times as the number of oysters intended for that tank. This list was then randomly ordered using the random number generating function in excel to produce a random number for each line item, the items in the list were then ordered in descending order of the random number. The oysters were then placed in a large holding tank and each oyster was systematically removed from the holding tank and placed in the tank number that was next on the list.

To randomly allocate treatments to tanks a similar approach was used. A list of treatments was generated which included one line for each replicate, this list was then randomly ordered using random numbers generated in Microsoft Excel as described above. The water temperature profiles were assigned to the systems in the same way.

2.2.5 Water temperature monitoring

The water temperature in the recirculating tanks and the oyster tanks was recorded every 30 minutes using a commercial, factory calibrated temperature probe (Thermocron iButton,

DS1921G). The probes were sealed in a watertight container and weighted down so they sat at a depth of 10 cm and 20 cm within the water column. The probes were placed in Tanks 2 and

5 and the sump of each recirculating system.

The water temperature was also observed directly using submersible thermometers and kept within 1°C of the target water temperature throughout each experiment by adjusting the aquarium heater and the chiller units.

2.2.6 Water quality monitoring

The pH, TAN, nitrate and nitrite was measured in each tank using a commercially available saltwater master test kit (API) according to the instructions. A sample of tank water was taken using a sterile transfer pipette and placed in each test tube and the test reagents were added and mixed. The results were then visually compared to the reference card.

2.2.7 Feed

The oysters were fed a commercial live algae concentrate (Shellfish Diet 1800, Reed

Mariculture) which contains a mix of Isochrysis spp., Pavlova spp., Tetraselmis spp.,

Chaetoceros calcitrans, Thalassiosira weissflogii and Thalassiosira pseudonana at a combined concentration of 8x109 cells mL-1 (8% dry weight).

Dry weight composition (from manufacturer):

Protein 52.0%

Lipid 16.1%

Carbohydrate 22.0%

Ash 9.9%

Oysters were fed 2 mL per tank every second day. This level was calculated as a maintenance ration intended to balance feed availability and water quality during each experiment.

2.3 OsHV-1 Inoculum

2.3.1 Cryopreserved inoculum

A cryopreserved stock of clarified tissue homogenate produced from oysters that were naturally infected with OsHV-1 and sampled from the Georges River in 2011 by Paul-Pont et al. (2013b) was used as an inoculum as described by Hick et al. (2016). Briefly, the gill and mantle were excised using a scalpel and homogenized in a stomaching machine with 10% w/v

ASW. The diluted homogenate was successively filtered to 5 µm, 0.45 µm and 0.22 µm using syringe filters (Micropore) and frozen at -80°C with 10% v/v foetal bovine serum (Sigma) and

10% v/v molecular biology grade glycerol (Sigma).

For use as an inoculum, a new aliquot of cryopreserved homogenate was used for each challenge to avoid repeated freeze-thaw cycles. The cryopreserved stock was diluted 1 in 100

38 with sterile ASW. The undiluted stock contained 1.5x105 OsHV-1 DNA copies.µL-1 the dilution and 100 µL dose combine to give 1.5x106 OsHV-1 DNA copies per oyster.

The negative control was prepared using the same procedure with gill and mantle tissue from a population of apparently healthy oysters which were considered as free from infection with

OsHV-1 based on qPCR testing.

2.3.2 Fresh inoculum

A fresh inoculum was prepared by challenging oysters by injection with the cryopreserved

OsHV-1 stock. They were and maintained in tanks at 22°C for 36 hours. The oysters were collected at the time of mortality or at 36 hours and the gill and mantle were dissected. The gill and mantle tissues (0.1 g) were homogenized in a 2 mL tube with 1.0 mL of ultrapure water (Life Technologies) with 0.4 mL of silica zirconia beads (Daintree Scientific). The mechanical homogenisation was done using a TissueLyser II (Qiagen) for 240 s at 30 oscillations per second and the tube holder was inverted after 120 s. The tubes were centrifuged at 900 g for 5 minutes to help remove remaining large particles and 200 µL of the supernatant removed. This was added to 19.8 mL of sterile ASW to achieve a final dilution of

1 in 1000 w/v. The diluted supernatant was then successively filtered to 5 µm, 0.45 µm and

0.22 µm and stored on ice for no longer than 4 hours prior to use.

The negative control was prepared by dissecting 0.1 g of gill and mantle tissue from an apparently healthy oysters from the same cohort used to produce the OsHV-1 inoculum and processed in the same way to the same dilution.

2.3.3 OsHV-1 challenge

Prior to injection (Figure 2.2), the oysters were removed from the water for 12 hours and then

-1 immersed for 6 hours in 50 gl MgCl2 dissolved in fresh aquarium water to relax the adductor muscle. A dose of 100 µL of inoculum was injected into the adductor muscle using a 1 mL syringe and a 25G needle. After injection the oysters were rinsed in ASW to remove residual magnesium chloride solution and immediately placed back into their tank.

Figure 2.2: Pacific oysters (Crassostrea gigas) were injected into the adductor muscle with an

Ostreid herpesvirus 1 (OsHV-1) inoculum after relaxation in a magnesium chloride bath.

2.3.4 Oyster mortality

Oysters were defined as moribund when their valves did not close after 5 minutes of air exposure or were gaping open and did not retract the mantle when it was pricked with a 25G needle. Moribund oysters which showed any signs of tissue degradation were defined as dead.

2.3.5 Waste management

Laboratory waste and solid waste from the aquatic animal facility was disinfected using an autoclave (Cisa) which reached a temperature of 121°C and pressure of 103 kPa for a minimum of 15 minutes.

Liquid waste from the aquatic facility was disinfected by the addition of pool chlorine (12.5% available chlorine, manufacturer) to an available chlorine concentration of 500 ppm for 1 hour. The water was then held for 24 hours and discharged over 8 hours.

Surfaces and equipment were cleaned using Trigene (1:100) and paper towel. Clean surfaces were disinfected with a Virkon solution (1%) which was sprayed to cover onto exposed surfaces and given a 10 minute contact time. The surfaces and equipment were then rinsed with municipal tap water to remove residual Virkon and dried with paper towel.

2.4 Quantification of OsHV-1 DNA

2.4.1 Tissue sampling

Oysters sampled at the time of mortality, designated sampling time or at the end of an experiment were stored at -80°C until dissection. For tissue sampling each oyster was thawed, shucked and a biopsy of gill and mantle (combined mass: 0.08-0.12 g) was placed into a 2.0 mL tube containing 1.0 mL of DNA/RNA free ultrapure water (Life Technologies) and 0.4 g of 0.1 mm silica-zirconia beads (Daintree Scientific).

To reduce cross contamination between samples each oyster was opened and dissected on a separate section of an absorbent bench mat (Thermofisher Scientific) placed on a plastic chopping board. Gloves were changed and the bench and balance disinfected with chlorine

41 solution (1000 ppm) when contaminated or between oysters from different tanks. The cutting board was changed between oysters from different tanks and disinfected with chlorine solution (1000 ppm) for a contact time of 10 minutes.

2.4.2 Tissue handling for RNA extraction

Tissue samples (0.04 g) were placed in 500 µL of RNALater (Sigma) in a 1.5 mL flip top tube and stored at -20°C until tissue homogenisation. Prior to tissue homogenisation, the tissue samples were removed from RNALater (Sigma) with a sterile 10 µL pipette tip placed in a 2.0 mL tube containing 400 µL of RLT buffer with DX and 0.4 g of 0.1 mm silica-zirconia beads

(Daintree Scientific).

Dedicated aliquots of reagents were used for working with tissue samples to be used for RNA extraction and working with RNA samples. All work with these samples was done in a PC2 cabinet (Email Air Handling) that was cleaned with RNAZap (Sigma) and all equipment used for working with RNA was cleaned with RNAZap prior to use.

2.4.3 Tissue homogenisation

The tissue samples were mechanically homogenized by bead beating using a TissueLyser

(Qiagen) for 240 s at 30 oscillations per second with an inversion of the tube holding insert after 120 s. The samples were then centrifuged at 900 g for 10 mins. The supernatant was aliquoted into DNA and RNA free tubes and stored at -80°C until nucleic acid extraction.

2.4.4 Purification of nucleic acids

Total nucleic acids were purified using the MagMAX-96 Viral RNA Isolation Kit (Life

Technologies) and a magnetic particle processor (MagMAX Express-96 Applied Biosystems).

The kit was used according to the manufacturer’s directions. Briefly, 50 µL of the sample supernatant was added to the lysis buffer and magnetic bead mix, incubated and the nucleic acids attached to the beads were passed through two wash buffers and eluted into the elution buffer. Purified nucleic acids were stored in the elution buffer at -20°C prior to quantification.

A negative control well with no sample was included on each plate and was subsequently tested for OsHV-1 DNA contamination by qPCR.

2.4.5 OsHV-1 quantification by qPCR

The number of copies of OsHV-1 DNA in the samples was quantified using the real-time quantitative PCR assay described by Martenot et al. (2010) modified as previously described

(Evans et al., 2014). The assay uses TaqMan chemistry and the amplification target and probe are in the B-region of the OsHV-1 genome (ORF 4). Reactions were prepared in duplicate 25

µL volumes with Path-ID qPCR Master Mix (Applied Biosystems). Each reaction contained 5

µL of purified nucleic acid sample as template with, 12.5 µL reaction buffer, 1 µL enzyme mix, 250 nM OsHV-1B probe (5’ TGCCCCTGTCATCTTGAGGTATAGACAATC 3’), 900 nM each of the primers BF (5’ GTCGCATCTTTG- GATTTAACAA 3’) and B4 (5’

ACTGGGATCCGACTGACAAC 3’), and molecular grade water. The cycling conditions were 10 minutes at 95 °C followed by 40 cycles of 95 °C for 15 s and 60 °C for 45 s run in a

7500 FAST thermocycler (ABI) or Mx3000P (Stratagene) qPCR instrument.

A standard curve for quantification ranging from 107 copies to 100 copies of template per reaction was derived from a 10-fold dilution of plasmid DNA containing the target region

(pOSHV-Breg). The pOSHV-Breg plasmid was produced by Alison Tweedie by amplifying a

324 bp region of the OsHV-1 genome using the Expand High Fidelity PCR system (Roche) with the primers (Forward: 5’ CACATCAACTGCTGGTGCTGAT 3’, Reverse: 5’

ACTGGGATCCGACTGACAAC 3’). The amplicon was inserted into the qCR 4-TOPO TA vector and cloned using a DNA TA cloning kit (Thermofisher Scientific) in OneShot TOP10 chemically competent Escherichia coli cells. The plasmid was then purified using a Plasmid

Quikpure kit (Macherey Nagel). The quantity of purified plasmid was measured using a Qubit

(Thermofisher Scientific) with a dsDNA BR Assay kit (Thermofisher Scientific), the purified plasmid was then diluted to 5x106 copiesµL-1 and 20 µL aliquots were stored at -80°C for use in preparing the standard curves.

During the qPCR experiment carboxyrhodamine (ROX) and 6-carboxy-fluorescein (FAM) fluorescence data were collected during each cycle. The FAM data was normalised to the

ROX reference dye to account for well to well differences in florescence. The ROX normalised fluorescence data baseline corrected with the baseline set between cycles 3 and 15.

The threshold line was then set to the lowest level where it was above the baseline noise and the efficiency of the reactions based on the standard curve was between 95% and 105%.

Samples were defined as positive if the ROX normalised and baseline corrected FAM florescence increased exponentially and exceeded the threshold, the cycle where the magnitude of the florescence exceeded the threshold was defined as the cycle threshold (Ct).

Samples which did not meet this criterion were negative. The number of copies of the OsHV-

1 B-region in each sample in the PCR was extrapolated from a standard curve generated by plotting the Ct of the samples in the standard curve against they known number of copies of the B-region. The number of copies in the sample in each qPCR reaction was converted to copies per mg of tissue using the equation:

10 × (((푠푎푚푝푙푒 푚푎푠푠 (푚푔) × 100) + 1000푢푙 푤푎푡푒푟)/(푠푎푚푝푙푒 푚푎푠푠(푚푔) × 100) which 3 accounts for the dilution of the sample mass in 1000 µL in ultrapure water, the dilution of 50

µL into 75 µL during the nucleic acid extraction and a 5 µL sample being measured by qPCR.

2.4.6 Tissue processing and PCR quality control

A tissue processing batch was defined as a group of 45 samples which underwent the tissue homogenisation process described in section 2.4.3 at the same time. An empty tube was included at the beginning and end of each batch as negative control to identify contamination.

A known positive tissue sample was also included at the end of each batch as a positive control to ensure that if virus was present it was being extracted and as a measure of extraction efficiency. A known positive and negative sample was also included on each PCR plate. The negative control samples were used to identify any contamination in the reagents used in processing the tissue samples and the positive controls were used to monitor the efficiency of

OsHV-1 recovery between batches. A positive control sample of 1x102 copies of the pOSHV-

Breg plasmid was also included in each PCR plate.

2.5 Statistical Methods

Statistical analyses were performed using Microsoft R Open version 3.4.3 (R Core Team,

2017) and SAS™ 9.3 software (SAS Institute Inc.). Data were summarized, graphed and analysed using the Tidyverse, plyr, survival, survminer, lme4, lmerTest and binom packages

(Bates et al., 2015; Dorai-Raj, 2014; Kassambara and Kosinski, 2017; Kuznetsova et al., 2017;

Müller and Wickham, 2017; Therneau, 2015; Therneau and Grambsch, 2000; Wickham et al.,

2017b, 2017a, Wickham, 2017, 2011, 2009; Wickham and Henry, 2017).

General statistical methods are defined here, their application in each experiment is explained in the methods of each chapter.

2.5.1 Data management

Research data was stored on the University of Sydney Research Data Store (RDS) in accordance with the University’s policy on the storage of research data. The data were only accessible by researchers directly involved in the project. The RDS is backed up hourly and offline backups are created each week to protect the data from loss.

Transcribed data was checked for errors by double checking each entry then checking the number of entries per group against the experiment design and assessing high or low data points.

2.5.2 Summary statistics

Summary statistics were generated for categorial data by creating frequency tables of the outcome and predictor variables. For mortality data the number of dead and total number of oysters were used to calculate the percentage mortality in each group with an associated confidence interval. For continuous variables, boxplots by treatment group were used to visually assess the differences between groups, the normality of the data and the spread of the data. Numerical summaries including the maximum, minimum, mean and median were also used to assess normality and spread and differences between groups. Data that were not normally distributed were log10 transformed were required to meet the assumption of normality or where required for interpretation.

2.5.3 Confidence intervals

Confidence intervals for normally distributed data were calculated at the 95% using the mean and standard error with the calculation 95% CI = mean ± 1.96 x standard error. Where the data were log10 transformed to meet the assumption of normality the mean and associated confidence intervals were back transformed using the equation 10n, where n is the mean or

95% CI. Confidence intervals for binary data, including mortality, were calculated using

Fishers’ exact method in R using the function binom.logit in the binom package. Confidence intervals for odds ratios and hazard ratios were calculated using the confint function in R which uses the equation 95% CI = parameter estimate ± 1.96 x standard error. For odds ratio estimates the confidence intervals and point estimate were then exponentiated to obtain the odds ratio from log odds.

2.5.4 Generalised linear mixed models

Generalised linear models were fitted using the MIXED procedure in SAS 9.3 software or the glmer function in R. The parameters were estimated by maximum likelihood, where the outcome variable was binary the logit link function was used. Random effects were added to the model to account for clustering based on the study design. The fit of the model was assessed using the Akaike information criterion and Bayesian information criterion in comparison to a null model. The assumptions of the model were assessed graphically using plots of the standardised model residuals. The quantile-quantile comparison plot of the residuals was used to assess their normality, this was accepted when data generally followed a

1:1 gradient. The residuals-fitted plot was used to assess the assumption of a linear relationship of the variables to the outcome, this was accepted when no structure not associated with binary predictors could be seen in the plot and the trendline had a gradient of

0. The scale-location plot was used to assess the assumption of equal variance, this was

47 accepted when the was no visible structure not associated with binary predictors to the plot and the trendline had a gradient of 0. The residuals-leverage graph was used to assess if any datum was having a particularly strong effect on the model, this was accepted if no point was outside Cook’s distance.

2.5.5 Cox proportional hazards model

A Kaplan-Meier survival curve was generated for each treatment group using the survfit function from the survival package and ploted with the ggsurvplot function from ggsurvival package with visual elements controlled with the ggplot2 package.

A Cox proportional hazards model was then used to analyse the mortality data with the survival package in R using the coxph function. Clustering, including within replicates, was accounted for by adding a clustering variable as required by the study design. The parameters of the model were estimated by maximum likelihood. The Wald test statistic was used to assess the significance of the independent variables. The assumption of proportional hazards was assessed using Schoenfeld residuals, the residuals were assessed for each variable using the cox.zph function and visually assed using the ggcoxzph function to plot the residuals. The assumption was met if the residual plot had no discernible structure not associated with binary predictors and the trendline had a gradient of 0.

3. Chapter 3 - Risk factors for mortality during the first occurrence of

Pacific Oyster Mortality Syndrome due to Ostreid herpesvirus – 1 in

Tasmania, 2016

3.1 Abstract

The microvariant genotype of Ostreid herpes virus 1 (OsHV-1 μVar) has caused repeated seasonal outbreaks of mass mortality in Pacific oyster (Crassostrea gigas) populations in

Europe, New Zealand and in New South Wales, Australia. The first occurrence of OsHV-1 related mortality in Tasmania, Australia was in January 2016; mortality was documented and multivariable logistic regression analysis was undertaken to examine risk factors. Overall mortality was 78% but varied from 37% – 92% across 6 farms and between 96% (0 – 20 mm length) and 33% (61 – 115 mm) across size classes. The odds of mortality were 3.2 to 3.8 times greater for baskets containing oysters with an average shell length ≤ 40 mm compared to those with an average < 61 mm. Generally, the greater the time on farm the lower the mortality, consistent with the increasing age and size of the oysters. The effect of stocking density on mortality changed with increasing time on farm (age). The standard density produced lower mortality than high or low stocking density particularly for spat. Mortality was nearly twice as likely when oysters were handled for routine husbandry in the 7 days prior to the outbreak compared to not handled. Ploidy and hatchery of origin did not alter the disease outcome. The variation in mortality with husbandry requires further investigation to determine if it can be exploited to help maintain productivity. Statement of relevance: This descriptive epidemiological study documents the impact of the first occurrence of disease associated with Ostreid herpesvirus - 1 (OsHV-1) in Tasmania, Australia. OsHV-1 is an emerging and economically important pathogen. Describing the impact of OsHV-1 in new

49 environments and identifying the association between farm management and mortality is important to develop effective disease control measures.

3.2 Introduction

The microvariant genotype of Ostreid herpes virus 1 (OsHV-1 μVar) was first detected in

France in 2008 and was identified as the cause of mass mortality of Pacific oysters

(Crassostrea gigas) during summer (EFSA, 2010; Schikorski et al., 2011a; Segarra et al.,

2010). Microvariant genotypes of OsHV-1 have since been detected in association with mass mortality in Pacific oysters in several European countries (Clegg et al., 2014; Gittenberger et al., 2016; Lynch et al., 2012; Roque et al., 2012), New Zealand (Bingham et al., 2013) and in

New South Wales (NSW), Australia (Jenkins et al., 2013). In NSW, mass mortality associated with OsHV-1 μVar in farmed Pacific oysters first occurred in the Georges River in 2010

(Jenkins et al., 2013) followed by the Hawkesbury River in 2013 (Paul-Pont et al., 2014) and was termed Pacific Oyster Mortality Syndrome (POMS). The high mortality rate and seasonal recurrence of POMS has caused significant economic damage to the NSW oyster industry with cessation of production and consequent job losses in waterways where OsHV-1 has become endemic.

The Tasmanian oyster industry produced more than 4000 tons of oysters in 2012 worth $24 million with the majority being Pacific oysters (ABARES, 2013). In Tasmania, oysters are grown predominantly on the north, east and south east coast. A survey in 2011 and ongoing monitoring prior to 2016, conducted according to government biosecurity regulations, demonstrated Tasmania to be free of OsHV-1 µVar (Animal Health Australia, 2016).

However, between 28/01/2016 and 18/03/16 farmers reported mortality of up to 100% in baskets harvested from Upper and Lower Pitt Water and Blackman Bay (Biosecurity

Tasmania, 2016). OsHV-1 μVar was detected in samples taken from these oysters on

01/02/2016 by the Department of Primary Industries, Parks, Water and Environment

(DPIPWE) (Biosecurity Tasmania, 2016). Testing was consistent with OIE and Department of

Agriculture and Water Resources AQUAVETPLAN guidelines including viral DNA sequence analysis (Department of Agriculture, 2015; OIE, 2016). Initially a control area preventing the movement of oysters and oyster growing equipment was enforced throughout Tasmania by the

Tasmanian Chief Veterinary Officer (CVO), consistent with a pre-existing emergency response plan (Department of Agriculture, 2015).

The affected growing areas with high levels of mortality were east of Hobart and largely within Upper and Lower Pitt Water, Pipe Clay Lagoon and Blackman Bay in south east

Tasmania, approximately 20% of growing areas in the state (Figure 3.1). The index case occurred in Lower Pitt Water. Lower Pitt Water is connected to Frederick Henry Bay on the southern end and to Upper Pitt Water to the north. Upper Pitt Water has a generally muddy/silted bottom, is 1.2-1.8 m deep at high tide on flats where the oysters are grown and is up to 6 m deep in the channels. Lower Pitt Water has a sandy substrate and oysters are grown on sand banks where the intertidal depth is between 1m and 2.5 m channels are also wider, up to 8 m deep, and have fast flowing currents compared to Upper Pitt Water.

Figure 3.1: Sampling locations (blocks) corresponding to leases in Pacific oyster

(Crassostrea gigas) growing areas in Upper and Lower Pitt Water (A), Blackman Bay (B).

Also shown are growing areas in which mortality occurred in January-February 2016

(Biosecurity Tasmania, 2016) in the first Pacific oyster mortality syndrome (POMS) outbreak in Tasmania, and the sampling locations for this study which were representative of these.

Blackman Bay is connected to Dunalley Bay/Frederick Henry Bay on the west by the Denison

Canal and to the Pacific Ocean to the east; the water in the bay is fully exchanged with every tide and has a tidal flow of up 8 knots in the channel. Oysters are grown on intertidal leases on sand flats with an intertidal depth of 1-2 m and in sub-tidal leases 2-6 m deep. Channels in

Blackman Bay are up to 12 m deep and some sub-tidal leases are located in shallower parts of the channels.

The types of growing infrastructure used in Tasmania are common to other Australian oyster farming areas and comprise a mixture of fixed and adjustable height long-lines and fixed

52 height racks. In the long-line system the oysters are held in closed, rigid mesh plastic baskets which are suspended from horizontal lines. The lines are typically 100 m long and tensioned with plastic or wooden posts at ~5 m intervals (Figure 3.2). Adjustable long-line systems differ only in that the lines can be moved up and down the post using clips at different heights

(clip height) to adjust the position of baskets in the water column, immersion time and exposure to waves and weather. Rack systems have a vertical post and horizontal rail structure with oysters held in semi-rigid baskets held between the rails. Two forms of sub-tidal infrastructure are used: deep-water packs made from 5-6 horizontally stacked long-line baskets suspended ~0.5 m below the surface and floating, semi rigid, plastic baskets attached to a rope floating on the surface in a similar configuration to long-lines.

Figure 3.2: Plastic baskets containing oysters on height adjustable long-lines were the most common Pacific oyster (Crassostrea gigas) growing system used on the farms surveyed in

Tasmania at the time of the initial Pacific oyster mortality syndrome (POMS) outbreak.

Environmental and husbandry factors including water temperature and clip height on intertidal long-lines are associated with the occurrence and severity of POMS outbreaks (Paul-Pont et al., 2014; Peeler et al., 2012; Pernet et al., 2012). Decreasing the immersion time by raising the height of growing infrastructure by 300 mm above the industry standard reduced mortality of adult oysters due to POMS in NSW by half (Paul-Pont et al., 2013a; Whittington et al.,

2015a). The use of baskets compared to cementing oysters onto rope as growing infrastructure apparently increased mortality in France (Pernet et al., 2014a). Oyster age, size, ploidy and

54 time on farm have all been associated with differing mortality in field trials and previous outbreaks of POMS (Dégremont et al., 2016; Paul-Pont et al., 2014; Peeler et al., 2012).

Critically examining the distribution of mortality and the range of growing methods used at the time of the POMS outbreak in Tasmania presented an opportunity to identify husbandry and management factors which may exacerbate mortality or have a protective effect against

POMS. In this outbreak where there were commercial quantities of oysters in the water and none of these oysters had been previously exposed to OsHV-1. Consequently, the aim of this investigation was to estimate total cumulative mortality and identify husbandry and oyster factors that influenced mortality during the first POMS outbreak in Tasmania. A detailed study was undertaken on 6 farms across 3 bays that were representative of the outbreak.

3.3 Methods

The investigation was conducted between the 21st and the 24th of March and the 11th to the

22nd of April, 2016 corresponding to the recommencement of farming operations following the POMS outbreak. All farms included in the study had been classed by DPIPWE as POMS- affected following testing for OsHV-1. All dead shells were assumed to be oysters that had died of POMS. Prior to this outbreak, the baseline mortality reported by farmers was < 1% across all size classes.

3.3.1 Farms

Farmers affected by the outbreak were invited to participate according to the availability of detailed farm records and their willingness to participate. This required farmers to physically assist researchers in sampling and examining stock. In Upper and Lower Pitt Water there were

5 farms of which 3 participated (Figure 3.1). In Blackman Bay there were 6 farms of which 3 participated. Selection of farms was by convenience due to logistic constraints (Figure 3.1).

A farm was defined as a set of one or more blocks (oyster leases) in a single bay managed as a single commercial enterprise with a consistent management strategy. A block was defined as an area of parallel rows of infrastructure around 100m long, consisting of either long-lines in sets of four, floating lines or fixed height baskets, which were separated by a gap of 10-20m.

These blocks were surrounded by management channels > 20m wide to allow for boat movement. Long-lines were held up by sequentially numbered posts at 5m intervals; these were used to indicate the position of each basket along the row. A batch of oysters was defined as a group from the same spawning which came onto the farm as a single group. A batch may have been split into further groups based on size or distributed across discontinuous segments of growing infrastructure across the farm according to the husbandry requirements of the farmer. Where a batch was split into these different management units they were termed a sub-batch.

3.3.2 Study design

A target sample size of 7991 oysters, comprising 80 baskets with about 100 oysters each was calculated to be required at each farm. This design was to estimate farm-level mortality with a precision of 5% and a confidence of 95% and was calculated based on an estimated mortality of 50% and adjusted for clustering within baskets with an intra-class correlation coefficient of

0.2 (Dhand and Khatkar, 2014). The sampling frame for each farm was determined by examining farm records of oysters present on the farm at the time of the outbreak. Sampling on each farm was stratified with preferential selection of batches that contained sub-batches across more than one location or growing structure. Sample size was proportionate to the size

56 of the sub-batch within each farm. The sample size that was achieved at each farm depended on logistical considerations including the time that was available due to tides, the number of baskets that were present and the time required to examine the oysters in each basket, which in turn depended on their size, number per basket and the degree of biofouling.

3.3.3 Sampling

Systematic random sampling was used to obtain a representative sample of baskets within each sub-batch. A random number between 1 and n was generated at the time of sampling

(https://www.random.org), where n was the total number of baskets divided by the sample size. This determined the first basket to be sampled according to position on the row.

Subsequently each nth basket was sampled and its position was recorded by attaching a label to the basket (Figure 3.3). The basket position along the row was designated by counting the baskets on each of the 4 lines from left to right while moving along the row.

Farm 1 Farm 4

1 2 3 9 10 10 20 5 15

11 12 Row 10 Row 10 5 5 0 0 10 20 30 0 10 20 30 0 10 20 30 5 10 15 5 10 15 Post number Post number Farm 2 Farm 5

4 5 13 Mortality % 10 100 80

60 Row Row 5 40 5 20 0 0 10 20 30 0 10 20 30 5 10 15 Post number Post number Farm 3 Farm 6

6 7 8 14 15 16

15 30 20 10 10

17 18

Row Row 5 30 20 10

0 5 10 15 20 0 5 10 15 20 0 5 10 15 20 5 10 5 10 Post number Post number

Figure 3.3: Sampling locations in each block within each farm; grey scale indicates mortality

% in each basket sampled.

The size and age range of oysters that were available for the survey reflected the stage in the production cycle at that time of year. The peak sales of oysters for Christmas had occurred immediately prior to the outbreak therefore the number of plate-size stock present on each farm was low. Husbandry conditions were also affected by oyster sales prior to the outbreak; much of the stock on the top clip height had been sold before the outbreak and stock on the lower clip height had been brought up to the middle clip height to replenish these at the time

58 of the outbreak. On Farms 2 and 3, cleanup operations had commenced prior to the survey; this decreased the availability of larger stock sizes to sample as they were being preferentially targeted due to higher immediate market value.

3.3.4 Size measurement

The average size of the oysters within each basket on all farms was determined at the time of sampling according to the standard categorical sizes used for stock management and marketing. Size was assessed by the farmers according to the average top shell length corresponding to the categories in Table 3.3; sizes larger than 71mm were grouped due to small sample size. In addition, on Farm 1 the total shell length of every oyster in all baskets surveyed was determined to the nearest millimeter using a carpenter’s square. On Farm 2 the length of oysters in the two baskets with the greatest mortality and the two with the lowest mortality for each batch sampled was determined by the same method. Lengths were recorded for both live and dead oysters. These data were analyzed separately to the categorical size data.

3.3.5 Quantification of mortality

The total cumulative mortality in each basket was determined by counting the number of apparently live oysters and the number of empty shells. All mortality was attributed to POMS for the purposes of the survey. Farmer reported mortality was not included in the survey.

3.3.6 Risk factors

Farm management including location, husbandry and growing conditions and biological parameters including oyster size were recorded for each basket (Table 3.1). Stocking density was evaluated based on the number of oysters in each basket for each size category: baskets in

59 the first quartile were designated as high stocking density, baskets in 4th quartile were designated as low stocking density and those in the 2nd and 3rd quartiles were designated as standard stocking density. Time on farm was calculated from the arrival date of each batch and, in the case of on-growers (i.e. oysters acquired as juveniles or adults from another farmer rather than as spat from a hatchery), adjusted for time spent on the previous farm, as estimated by the farmer.

Table 3.1: Description of risk factors and outcome variables

Variable Description

Bay One of three bays containing the farms which were surveyed Farm Unique and anonymous number for each farm Block Uniquely identifies each block, nested within farm Batch Uniquely identifies each group of oysters as they came onto the farm Sub-batch Uniquely identifies sub-groups of a batch split across different locations Dead Count of dead oysters in a basket, outcome variable Total Total count of oysters and shells in a basket, outcome variable Size Total top shell length standardized across farms, categorized based on management size categories Time on farm Number of months spent on a farm since spat left the hatchery of origin Density Designated as high for the top quartile, standard for middle two quartiles or low if in the lowest quartile for stocking density for the size category in each basket On-grower Whether or not the batch had been bought from another farm for grow-out and sale Handled Whether or not the basket had been subjected routine management procedures in the 7 days prior to the outbreak Ploidy Genome status, diploid or triploid Hatchery The hatchery of origin, one of two local hatcheries Clip height Ordinal descriptor of relative basket height, including sub-tidal growing, which determines immersion time

Handling was defined as any routine management procedure applied to the basket in the seven days prior to the outbreak; this included grading, movement between leases and arrival of oysters onto the farm from a hatchery or another farm.

3.3.7 Statistical analysis

Statistical analysis was performed using the SAS (release 9.4, SAS Institute Inc.) GLIMMIX and MIXED procedures. Plots were created using Microsoft R open (version 3.2.5, 2016-04-

14) with the ggplot2 and ggmap packages (David Kahle, 2013; R Core Team, 2017;

Wickham, 2009) and

ArcGIS (Version: 10.2, ESRI Corporation). Map data were sourced from the Australian

Bureau of Statistics Australian Statistical Geography Standard 2016 (Australian Bureau of

Statistics, 2016).

3.3.7.1 Descriptive analysis

Data summaries, histograms and scatter plots against mortality as a percent of total basket count were used to assess normality of the data and screen for any graphical association with mortality. The data were visually checked for errors and outliers, no outliers were removed.

3.3.7.2 Generalized linear mixed model

A generalized linear mixed model (GLMM) was used to assess the effect of management practices across farms and intrinsic oyster factors on mortality. The count of dead oysters relative to the total number of oysters per basket was used as the outcome variable

(dead/total).

The random effects farm, block and sub-batch were included in the model to account for the variation arising from differences in geographical location and positioning due to husbandry practices. Management differences were accounted for at the farm level, differences due to location were accounted for at the block level, nested within farms. Over-dispersion was reduced with the addition of a random effect at the sub-batch level, nested within blocks, and

61 accounted for in the standard errors of the parameter estimates (Table 3.2). Random effects were modelled with random intercepts.

Table 3.2: Number of levels and baskets counted for the factors included as random effects in the generalised linear mixed model

(GLMM).

Intra-class Random Number Variance Number of baskets surveyed at each level Description correlation effect of levels (SE) coefficient Mean Maximum Median Minimum Farm Unique and anonymous number for each farm 6 0 0.000 122.0 229 89 42 Block Uniquely identifies each block, nested in farm 45 0.15 (0.22) 0.036 16.3 154 7 1 Sub-Batch Uniquely identifies each batch split across 131 0.80 (0.16) 0.225 6.1 38 1 1 each location

The initial explanatory variables added to the model were time on farm in months, size category, ploidy, density, hatchery, bay, clip height, on-grower status and if the basket had been handled in the 7 days prior to the outbreak (Table 3.1). These risk factors were included as fixed effects and were screened for association (p <0.25) with mortality (dead/total) in univariable analyses. The assumption of linearity was tested graphically for each continuous variable. Only variables with a significant association (p <0.25) were included in multivariable analyses. Forward stepwise multivariable logistic regression analysis was performed between each unique combination of factors starting with two variable models and progressing with the addition of further variables to significant models. The significance of each variable was assessed using a Wald test.

Two-way interactions then were assessed between all variables selected for inclusion in the final model. The model was assessed using the log likelihood ratio test (P < 0.001) and the presence of influential points was assessed graphically using Cook’s distance.

3.3.7.3 Length variation within basket

A similar GLMM as above was used for univariable analysis to assess the effect of length variation of oysters within a basket for Farms 1 and 2 where individual lengths were measured, with mortality (dead/total) as the outcome variable. The explanatory variables of length range, the range of lengths within each basket, and the standard deviation of lengths within the basket were tested separately in the model. The lengths of live and dead oysters within a basket were compared using a restricted maximum likelihood (REML) analysis with length as the outcome, death as the explanatory variable and sub-batch and block level random effects. Assumptions of the model were evaluated using residual diagnostics.

3.4 Results

A total of 6 farms were surveyed, and there were between 42 and 299 baskets sampled at each farm (Tables 3.2 and 3.3). Of the baskets surveyed, 67.5% contained oysters with a mean length between 41 mm and 60 mm (Table 3.3, Figure 3.7 Supplementary). Time on farm was between 1 and 39 months with an average of 12.2 months (Figure 3.7 Supplementary).

Growing infrastructure varied between farms with sub-tidal, long-line, floating baskets and fixed rack systems all in use. However, the majority of baskets surveyed (92%) were on a height adjustable long-line system. On-growers were present only on Farms 2 and 4 and had been on farm for between 1 and 2 months at the time of the outbreak. Density was greatest and had the largest range in the smallest size category but was relatively consistent across the larger size categories (Figure 3.8 Supplementary).

Total mortality was 78.3% with a range of 0% to 100% between baskets (Figure 3.9

Supplementary). The range of mortality was similar across all 3 bays however overall mortality was higher in Blackman Bay (87.1%) than in Upper or Lower Pitt Water (53.5% and

54.5% respectively) (Table 3.3). Farm level mortality ranged from 37.3% to 92.3%.

Table 3.3: Total cumulative mortality (%) from counts of dead and live Pacific oysters

Tasmania, January 2016. Reference categories are indicated with an odds ratio of 1. Variables significant at p < 0.25 were considered for multivariable analysis.

Variable Levels Mortality % (95% CI) Odds Ratio (95% CI) P value Number of baskets Bay Upper Pitt Water 53.51 (53.62 - 53.40) 1.00 < 0.001 86 Lower Pitt Water 54.53 (54.58 - 54.48) 0.78 (0.44 - 1.39) 391 Blackman Bay 87.12 (87.16 - 87.08) 2.55 (1.42 - 4.57) 255 Farm 1 53.51 (53.62 - 53.40) - - 86 2 50.70 (50.76 - 50.64) - 299 3 92.28 (92.32 - 92.24) - 165 4 76.79 (76.91 - 76.67) - 48 5 37.31 (37.46 - 37.16) - 42 6 59.53 (59.63 - 59.43) - 92 Size 0-20mm 95.88 (95.93 - 95.83) 3.93 (1.89 - 8.15) < 0.001 70 21-30mm 78.67 (78.77 - 78.57) 3.80 (1.83 - 7.89) 71 31-40mm 69.68 (69.82 - 69.54) 3.55 (1.75 - 7.19) 40 41-50mm 49.49 (49.55 - 49.43) 2.35 (1.11 - 4.94) 233 51-60mm 44.35 (44.41 - 44.29) 1.17 (0.59 - 2.32) 261 61-115mm 32.94 (33.06 - 32.82) 1.00 57 Clip height Sub-tidal 23.37 (23.71 - 23.03) 0.23 (0.05 - 0.97) 0.043 6 Low clip 56.83 (56.92 - 56.74) 2.24 (0.61 - 8.25) 128 Middle clip 82.28 (82.32 - 82.24) 1.52 (0.89 - 2.60) 455 High clip 48.21 (48.29 - 48.13) 1.00 143 Handling Not handled 50.20 (50.25 - 50.15) 0.49 (0.26 - 0.90) 0.021 414 Handled 85.38 (85.42 - 85.34) 1.00 318 On-grower Non on-grower 80.53 (80.56 - 80.50) 0.45 (0.15 - 1.40) 0.169 503 On-grower 54.47 (54.53 - 54.41) 1.00 229 Hatchery Unknown 34.12 (34.37 - 33.87) - < 0.001 14 A 88.82 (88.86 - 88.78) 3.38 (2.02 - 5.67) 224 B 52.37 (52.41 - 52.33) 1.00 494 Genotype Unknown 34.12 (34.37 - 33.87) - 0.007 2 Diploid 81.98 (82.01 - 81.95) 1.99 (1.21 - 3.28) 514 Triploid 54.03 (54.10 - 53.96) 1.00 216 Density Low 62.10 (62.17 - 62.03) 0.94 (1.11 - 0.80) < 0.001 176 Standard 74.59 (74.63 - 74.55) 0.81 (0.90 - 0.72) 376 High 84.56 (84.61 - 84.51) 1.00 180

Mortality was 62.9% lower in baskets containing the largest compared to the smallest size categories (Table 3.3). Mortality in spat < 20mm top shell length was 95.9% and decreased with increasing size of the oysters to 33.9% in the largest group (> 71mm top shell length)

(Figures 3.4 and 3.5, Table 3.3). Univariable analysis indicated that mower mortality was associated with both longer time on farm and larger size categories (Figures 3.4 and 3.5). The lowest density groups had the lowest mortality (62.1%) compared to the standard (74.6%) and high (84.6%) density groups (Table 3.3). Mortality in diploid oysters was 82.0% and in triploids was 54.1%. Mortality for oysters handled in the 7 days prior to the outbreak was

35.2% higher than in those which were not handled (Table 3.3). Multivariable analysis was then undertaken to separate factors that were not significant in the presence of variation in mortality that was due to other factors.

100

Size category 0-20mm 21-30mm 50 31-40mm

41-50mm Mortaity (%) Mortaity 51-60mm 61-115mm

0 10 20 30 40 Time on farm (months)

100

Not handled 50

Handled Mortality (%) Mortality

0-20mm 21-30mm 31-40mm 41-50mm 51-60mm 61-115mm Size catagory

Seven out of the nine proposed husbandry or oyster factors had a significant association with mortality at p < 0.25 (Table 3.3). The final multivariable model contained four significant (P

<0.05) explanatory variables and a single non-significant variable as part of an interaction.

The random effects structure detailed above accounted for clustering in the data. This model included time on farm, size category, stocking density, and the recently handled status (Table

3.4). Handling the oysters in the seven days prior to the outbreak was associated with almost double the odds of mortality compared to oysters which were not handled, after accounting for all other effects included in the final model. The odds of mortality in oysters in size categories of <40mm mean length was more than 3 times that of oysters >61mm in length. There was an interaction between time on farm and density indicating the effect on mortality of time on farm was different at each level of density (Figure 3.6). The probability of death was highest for oysters stocked at low density or high density when the time on farm was short (3-6 months, i.e. when oysters were relatively small and young) compared to a long period (21-24 months, i.e. when oysters were relatively large and old), but for oysters kept at standard density the probability of death was the same regardless of how long they had been present on farm (1 to 24 months) (Figure 3.6). The standard density produced the lowest mortality until

18 months on farm, after which both high and low density stocking had slightly reduced mortality.

Variable Level Estimate Std. Error Odds Ratio (95% CI) P value Intercept - 0.488 0.441 - - Time on farm - -0.028 0.013 - 0.135 Density Low -0.169 0.132 - < 0.001 Standard -0.549 0.084 - High - - - Density*Time on farm Low 0.002 0.013 * 0.001 Standard 0.028 0.008 High - - Handling Not handled -0.619 0.265 0.54 (0.32 - 0.91) 0.020 Handled - - 1 Size Category (0-20mm) 1.347 0.371 3.85 (1.85 - 7.97) < 0.001 (21-30mm) 1.165 0.396 3.21 (1.47 - 6.98) (31-40mm) 1.254 0.357 3.50 (1.73 - 7.06) (41-50mm) 0.690 0.362 1.99 (0.98 - 4.06) (51-60mm) -0.045 0.341 0.96 (0.49 - 1.87) (61-115mm) - - 1 *The interaction between density and time on farm is presented graphically in Figure 3.6.

Low density 1.0

0.8

0.6

0.4

0.2

0.0

3 6 9 12 15 18 21 24

Standard density 1.0

0.8

0.6

0.4

0.2

Predicted mean mortality mean Predicted 0.0 3 6 9 12 15 18 21 24

High density

1.0

0.8

0.6

0.4

0.2

0.0 3 6 9 12 15 18 21 24 Time on farm (months)

Figure 3.6: Predicted mean mortality from the multivariable logistic regression model for

Pacific oysters (Crassostrea gigas) which had been on a grow-out farm for 3 – 24 months at each of 3 growing densities following the outbreak of Pacific oyster mortality syndrome

POMS in Tasmania.

There were 129 baskets from Farms 1 and 2 included in an analysis of the effect of length variation within a basket on mortality. These baskets had a mean mortality of 46.5% (range

4.1 - 94.9%) the range of lengths in each basket was 27.1mm ±9.9mm. There was no association between mortality and either the range of lengths in a basket (P = 0.055) or the standard deviation of lengths in a basket (P = 0.351) in univariable analyses. Greater individual oyster length was associated with a greater number of dead oysters in the two highest and two lowest mortality baskets from each sub-batch counted on Farm 2 (P < 0.001).

The difference in length between the live (72.7 ±7.7mm) and dead oysters (69.3 ±7.7mm) was

3.4mm.

3.5 Discussion

The impact of the first POMS outbreak in Tasmania included the loss of 78% of current stock on the farms surveyed in the current study. Overall, mortality was similar to that observed in the Hawkesbury River in 2013 (Paul-Pont et al., 2014) and in France at the time of the emergence of OsHV-1 µVar in 2008 (Segarra et al., 2010) but was higher than that seen in

Ireland (Peeler et al., 2012). The continued impact of OsHV-1 has prevented farming of

Pacific oysters in NSW, Australia, whereas in France production has continued with spat mortality increasing from 14% in the period 2004 – 2007 to 58% in the period 2008 – 2014

(EFSA, 2015).

During a massive outbreak of POMS in NSW older and larger oysters had higher survival, although age exerted a greater effect than size (Paul-Pont et al., 2014). Conversely, size was the most important factor in determining mortality, both compared to age and all other factors examined in the current outbreak, as indicated by the greatest odds ratios of all the factors included in the model (Table 3.4). In the current study, time on farm serves as a proxy for age

(Figure 3.10 Supplementary) and the estimates presented here, based on time on farm (Figure

3.6), can be interpreted as an indicator of age. The difference between time on farm and age is the time spent in the hatchery between the date of spawning and the date of sale; this can be up to 6 months. The interaction that was observed between stocking density and time on farm indicates that management is an important factor in mortality due to POMS and that either a substantial increase or decrease in density from that most commonly used for each size category could be detrimental for young stock. Further investigation is required to determine the effect and mechanisms for higher and lower stocking density on mortality. High stocking density might intuitively be associated with greater transmission of a pathogen and this was demonstrated in laboratory trials where mortality was increased with greater density (Evans et al., 2015; Petton et al., 2015a).

Currently oyster farmers can manipulate the stocking density, location and position in the water column as well as the timing of the arrival of spat onto the farm to grow oysters to satisfy peak market demands. The density of oysters in a basket at each size is modulated by the farmer through regular grading to optimize growth and conditioning of the oysters, with the optimum number of oysters per basket decreasing as size increases. Although mean oyster length and number are important at a basket level, the observed difference in size between live and dead oysters was minimal and could be accounted for by growth of 1mm per week, which

74 is expected by the farmer, and may have occurred over the period between the outbreak and when the study took place; it is unlikely to be biologically significant.

Handling oysters within 7 days prior to the outbreak, including transferring oysters onto the farm, was associated with higher mortality. In Ireland manually handling oysters was associated with increased mortality when compared to machine handling (Peeler et al., 2012).

Under laboratory conditions, physically induced stress has a suppressive effect on the immune system over a short period (8 hours) (Lacoste et al., 2002). Physical stress is also associated with greater mortality resulting from pathogenic infection with Vibrio sp. (Lacoste et al.,

2001a). The magnitude of the stress effect due to handling in field conditions is likely to be influenced by environmental conditions such as differences between air and water temperature which can also affect the immune system (Green et al., 2014b). It is also possible that increased feeding following time out of water caused greater uptake of the virus. Regardless of the mechanism, higher mortality following handling of oysters suggests that farmers should not attempt to move or handle stock during high risk periods for an outbreak as the higher mortality resulting from the movement itself is likely to negate any benefit of the husbandry activity.

In the current study clip height and the use of sub-tidal systems was not found to affect mortality. The difference in clip heights on the adjustable height long line growing systems was approximately 100–150 mm. The clip height used was determined by local growing conditions and was chosen by farmers to optimize growth and conditioning. For adult oysters, high growing height was protective against POMS in repeated, replicated, controlled experiments in NSW and an observational study in Ireland (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a; Whittington et al., 2015a). In NSW this was associated with a growing

75 height 300mm above the farmer’s standard growing height. The different growing environment and tidal conditions in Tasmania make comparisons to growing heights in NSW very difficult. However, for Tasmania where a relatively homogenous range of growing systems was used, trialling greater differences in clip heights may produce an improvement in survival. Accurate measurement of immersion times at different growing heights in both NSW and Tasmania is warranted to enable further comparisons to be made. A prospective longitudinal study of affected growing areas in future outbreaks would allow investigation of more diverse and novel growing methods and overcome the limitations of this retrospective cross sectional study, including recruitment bias in the farms surveyed.

Sub-tidal systems where oysters are grown between 0.5m and 2.0 m below the surface were in use in some affected bays and favourable survival was noted by farmers. However, this cultivation method was uncommon and the oysters involved were the subject of emergency harvest, resulting in an insufficient sample size to compare to intertidal systems in this study.

Oysters grown in deeper water may experience lower overall temperatures, more stable environmental conditions and uniform growth rates (Sarà and Mazzola, 1997; Zrnčić et al.,

2007). Those areas where sub-tidal growing occurred had stronger tidal flow than intertidal areas. Higher flow rates were associated with lower mortality where OsHV-1 infection was transmitted by cohabitation in flow-through experimental systems (Petton et al., 2015a). These observations suggest that the local environmental factors present in the locations where sub- tidal systems are used may not be conducive to the expression of disease caused by OsHV-1

µVar. A controlled study examining the effect of a less homogenous range of infrastructure is required to investigate other growing practices that may reduce the risk of mortality, such as deep sub-tidal cultivation. Such a study was conducted in France (Pernet et al., 2014a).

Differences in ploidy and the use of on-growers had no significant effect on mortality in the current outbreak after accounting for time on farm, size, density and handling, and as such were not included in the GLMM. Previously, laboratory trials and field observations have indicated that ploidy is not biologically relevant in an OsHV-1 outbreak (Dégremont, 2011;

Dégremont et al., 2016; Degremont and Benabdelmouna, 2014; Gagnaire et al., 2006; Peeler et al., 2012). However, on-growers are an important component of production for many farmers. The ability to continue to use on-growers as well as both triploid and diploid stock will reduce the disruption to current farming systems caused by POMS by allowing for a wide range of sources of stock for grow out.

3.6 Conclusion

The high total mortality associated with OsHV-1 µVar in Tasmania, Australia had a significant economic impact on local communities in which oyster farming is an important industry. Multivariable logistic regression analysis was used to reveal important factors while separating out factors which were not significant but which may have been falsely attributed importance during the outbreak due to unrepresentative observations. After allowing for time on farm, density and handling, baskets containing oysters with a greater average length had a lower mortality than baskets containing smaller oysters. Handling oysters in the week prior to the outbreak resulted in higher mortality from POMS and should be avoided during a period of high risk. As such, any husbandry changes to reduce mortality should be pre-emptive.

Changes to ploidy, clip height, density and the source of oysters (on-growers or hatchery) should not be implemented for the purpose of reducing the risk of mortality in a POMS outbreak and oyster growers are advised to maintain normal practices in regards to these factors with the aim of maximizing production. Further investigation, such as prospective

77 longitudinal studies of future outbreaks will provide further insights into the disease impact in

Tasmania. Investigation of modified or novel management practices is recommended for the potential to reduce the risk of mortality in a POMS outbreak and integrate disease management into an efficient farming system.

400 400 A B 300 300

200 200

Baskets Baskets 100 100

0 0 1 2 3 1 2 3 4 5 6 Bay Farm

400 400 C D 300 300

200 200

Baskets Baskets 100 100

0 0 0 10 20 30 40 20 40 60 80 Time on farm (months) Average oyster length (mm)

Figure 3.7: (Supplementary) Number of baskets counted in each bay (A) and farm (B) as well as the distribution of the time on farm (C) and average length by basket (D).

3000

2000

1000 Number of oysters per basket

0 0-20mm 21-30mm 31-40mm 41-50mm 51-60mm 61-115mm Size category

Figure 3.8: (Supplementary) Number of Pacific oysters (Crassostrea gigas) per basket sampled in each size category.

100

60 75

Baskets Mortality

20 25

0 0 0 25 50 75 100 1 2 3 4 5 6 Mortality (%) Farm

Figure 3.9: (Supplementary) (A) Mortality at basket level (number of dead oysters / total number of oysters) and (B) basket level mortality on each of Pacific oyster (Crassostrea gigas) 6 farms surveyed following a Pacific oyster mortality syndrome POMS outbreak in

Tasmania, January – February 2016.

Time on farm (months) 10

0 0-20mm 21-30mm 31-40mm 41-50mm 51-60mm 61-115mm Size catagory

Figure 3.10: (Supplementary) Distribution of age within each size category. Points have been jittered around each size category.

4. Chapter 4 - Prior exposure to Ostreid herpesvirus 1 (OsHV-1) at 18°C is

associated with improved survival of juvenile Pacific oysters (Crassostrea

gigas) following challenge at 22°C

4.1 Abstract

The high mortality and economic loss in farmed Pacific oysters (Crassostrea gigas) caused by

Ostreid herpesvirus 1 (OsHV-1) has spurred research into strategies to mitigate the impact of the disease. Water temperature strongly influences the outcome of exposure with 14°C not being permissive for an epidemic, infection occurring with limited mortality at 18°C if there is a high viral dose, and full disease expression occurring at 22°C. Observations in the field suggest improved survival to subsequent exposures in cohorts of oysters that have survived an

OsHV-1 outbreak, but this assessment is complicated by the increasing age of the oysters, exposure to different viral loads, different environmental conditions and the removal of naturally susceptible individuals. A laboratory infection model was used to evaluate the effect on survival of two temporally-spaced challenges with OsHV-1 under different water temperature regimes. Triploid Pacific oyster spat were recruited from commercial farm stocks at 6 months of age and exposed to OsHV-1 by intramuscular injection at 18°C or 22°C and then re-exposed at either 18°C or 22°C. Mortality did not occur in oysters exposed at 18°C or when the temperature was increased to 22°C at 14 days post challenge. Oysters challenged with OsHV-1 at 22°C were protected if pre-exposed to OsHV-1 at 18°C (Hazard ratio: 0.22,

95% CI: 0.06–0.80). The present study suggests that non-lethal exposure to OsHV-1 can reduce the mortality on subsequent exposure to the virus, but further experiments are required to investigate the duration of the protective response and its effect in different environments.

Such protection may help describe the dynamics of seasonally recurrent disease epidemics and provide a novel approach to disease management.

4.2 Introduction

High mortality in farmed Pacific oysters (Crassostrea gigas) due to disease outbreaks caused by microvariant genotypes of Ostreid herpesvirus 1 (OsHV-1) has impacted aquaculture of this species across Europe, Australia, and New Zealand (EFSA, 2010; Jenkins et al., 2013;

Keeling et al., 2014; Renault et al., 2014, 2012; Segarra et al., 2010; Ugalde et al., 2018;

Webb et al., 2007). Recurrent disease outbreaks have prevented Pacific oyster production in

New South Wales waterways where OsHV-1 is endemic and have caused loss of production and economic hardship in Tasmania (de Kantzow et al., 2017; Paul-Pont et al., 2014; Ugalde et al., 2018). Reducing the impact of this disease is important to maintain the viability of the industry. Modifying the interactions between the host, pathogen and environment can reduce the severity of disease resulting in lower mortality. The importance of host and environmental factors that can be manipulated by farmers in outbreaks of OsHV-1 is illustrated by reduced mortality in genetically different groups of oysters (Degremont et al., 2015), in low density growing structures (Pernet et al., 2012) and higher intertidal growing height of oysters in baskets and racks (Paul-Pont et al., 2013a; Whittington et al., 2015a, 2015b). The use of additional management techniques to control disease caused by OsHV-1 is a high priority.

Vaccination is commonly used as a method of controlling viral diseases in finfish aquaculture, however, this approach is not applicable to oysters which do not have an adaptive immune response.

Repeated exposure to OsHV-1 occurs naturally (Whittington et al., 2018) and survivors of disease have been observed to survive subsequent outbreaks (Evans et al., 2017) suggesting

83 that the host response to OsHV-1 may be influenced by prior exposure. Injection of the double stranded RNA analogue polyinosinic:polycytidylic acid (poly I:C) into the adductor muscle reduced OsHV-1 prevalence 48 hours after exposure from 100% to 11% (Green and

Montagnani, 2013). Collectively, these data raise the possibility that a systemic stimulus such as non-lethal viral exposure might alter the pathogenesis of infection and increase host survival.

Water temperature is a well-studied environmental factor which profoundly alters the course of OsHV-1 infection (de Kantzow et al., 2016; Hick et al., 2016; Pernet et al., 2015; Petton et al., 2015a, 2013; Renault et al., 2014). Water temperature can be monitored easily and changes are predictable with the time of year, coinciding with seasonally recurrent OsHV-1 disease outbreaks (Evans et al., 2016; Garcia et al., 2011; Paul-Pont et al., 2014). Infection of oysters in tanks at water temperatures where mortality has been observed to occur in the field

(18°C) showed an interaction between dose and water temperature as a high dose of OsHV-1 was required to induce mortality at 18°C, with a prevalence in survivors of 38% (de Kantzow et al., 2016). Oysters challenged by OsHV-1 under field conditions at 22°C apparently maintained a persistent infection over 83 days at 10°C - 13°C (Pernet et al., 2015). While persistent infection was not detected by PCR, the disease expressed when the water temperature was raised to 22°C (Pernet et al., 2015). A water temperature of 10°C - 14°C is not permissive for establishment of a new infection or disease pathogenesis (de Kantzow et al., 2016; Petton et al., 2013).

The effect of altering water temperature after exposure to OsHV-1 merits further examination to determine if the pathogenesis is altered in favour of higher survival. More specifically, controlled exposure of oysters to OsHV-1 at a water temperature close to the lower limit

84 permissive to infection may enhance the proportion of survivors of infection that have an increased probability of survival with subsequent OsHV-1 challenge. The aim of the present study was to evaluate the impact of controlled temperature regimes on the susceptibility to disease when survivors of OsHV-1 exposure at 18°C and 22°C were re-challenged two weeks later with OsHV-1 at 22°C.

4.3 Methods

4.3.1 Oysters

Triploid Pacific oysters were sourced from a commercial hatchery (Shellfish Culture

Tasmania, Batch SPL16A) and were grown in Patonga Creek, New South Wales until 6 months of age and 30-50mm in length as described in Chapter 2.2.2. A random sample of oysters tested negative for OsHV-1 by qPCR prior to the trial (n=30). The oysters were randomly allocated to tanks as described in Chapter 2.2.4 and acclimated for 7 days at water temperature maintained of either 18°C or 22°C prior to the experiment.

4.3.2 Aquarium management

The tanks were set up and maintained as described in Chapter 2.2.3. The water temperature and quality monitoring were as described in Chapter 2.2.5-6. Access to food which was as described in Chapter 2.2.7.

4.3.3 OsHV-1 challenge

A fresh OsHV-1 inoculum was prepared, and oysters were challenged with OsHV-1 as per

Chapter 2.3.

4.3.4 Quantification of OsHV-1 DNA

Samples were taken from oysters as per Chapter 2.4.1 and nucleic acid extraction and quantification of OsHV-1 was conducted as described in Chapter 2.4.3 - 4 and 2.4.7 – 9.

4.3.5 Experimental design

Six treatment groups were defined by the water temperature profile throughout the experiment which included 2 challenges with either OsHV-1 suspension or a negative control suspension.

The numbers of tanks and oysters is shown in Figure 4.1. The initial challenge included 2 groups at 22°C, one challenged with OsHV-1 and the other with a negative control suspension. The remaining four treatment groups were at 18°C for the initial challenge; two of these groups were challenged with OsHV-1, and the other two with the negative control suspension. One group with each challenge suspension was maintained at 18°C for the remainder of the trial and the water temperature in the other two groups was increased to 22°C over 6 hours at day 14 after the initial challenge. The second challenge was performed 35 days after the initial challenge in all treatment groups.

Oysters were examined every 12 hours and dead and moribund oysters were removed and stored at -80°C. At the conclusion of the trial (Day 45) all remaining live oysters were sampled.

The time elapsed to mortality and the OsHV-1 DNA concentration in gill and mantle tissues were determined for all individuals that died. The OsHV-1 DNA concentration was determined in the tissues of all individuals that survived to the end of the trial.

Water Initial Second Group (No. temperature challenge challenge oysters/No. tanks)

4 (207/8) 5 (143/6)

22°C

1 (134/6) 18 °C 2 (144/6) Initial 3 (47/2) inoculum 6 (55/2) OsHV-1 35 Days Control -7 Days 0 Days 14 Days 45 Days

Figure 4.1: Water temperature profile for each treatment group, showing the acclimation period prior to the first challenge with

Ostreid herpesvirus 1 (OsHV-1) or the negative control inoculum, and follow up periods before and after the second challenge. The total number of Pacific oysters (Crassostrea gigas) in each group (N) were evenly distributed between the number of replicate tanks used in the group.

4.3.6 Statistical analysis

The initial sample size was calculated to provide 80% power to detect a 20% difference in mortality after the second challenge in the treatment and the controls with α = 0.05 and hazard ratio (HR) = 2. The required sample size was then determined for each group based on an expected mortality from the initial OsHV-1 challenge of 30% at 18°C and 50% at 22°C (de

Kantzow et al., 2016; Petton et al., 2013). Calculations were done using the powerSurvEpi package in R. Total cumulative mortality and prevalence of OsHV-1 in oysters that had died and in survivors was calculated as the aggregate for replicate tanks for each treatment group.

Statistical analyses were performed using Microsoft R Open version 3.4.3 (R Core Team,

2017) and SAS™ 9.3 software (SAS Institute Inc.). Data were summarized, graphed and analysed using the Tidyverse, plyr, survival, survminer, lme4, lmerTest and binom packages

(Bates et al., 2015; Dorai-Raj, 2014; Kassambara and Kosinski, 2017; Kuznetsova et al., 2017;

Müller and Wickham, 2017; Therneau, 2015; Therneau and Grambsch, 2000; Wickham et al.,

2017b, 2017a; Wickham, 2017, 2011, 2009; Wickham and Henry, 2017).

A Kaplan-Meier survival curve was generated for each group given a second challenge with

OsHV-1 at 22°C using the ggsurvplot function from ggsurvival package and the ggplot and survival packages. A Cox proportional hazards model was used to analyse the mortality data with the survival package in R. The failure condition was oysters which had died and had detectable level of OsHV-1 DNA; oysters in which OsHV-1 DNA was not detected by qPCR were censored at the time of death. The model was built using a backwards stepwise process beginning with both initial challenge temperature and the initial challenge suspension (OsHV-

1 or control) and their interaction. Each iteration, the variable with the largest P value was removed until only significant variables (P<0.05) remained. The tank identification number

88 was included in all models to account for clustering between individuals housed within the same tank. The Wald test statistic was used to assess the significance of the independent variables. The assumption of proportional hazards was assessed using Schoenfeld residuals.

A liner mixed effects model was used to determine the effect of initial water temperature, initial challenge suspension, second challenge suspension and mortality on the concentration of OsHV-1 in gill and mantle tissues of individuals where OsHV-1 was detected. A separate model was made for the oysters at the time of death and survivors in order to assess if different parameters were important for each group. The model was fitted using the MIXED procedure in SAS 9.3 software and estimates of the outcome for the terms in the interaction variable were computed for each combination with the lsmeans step and graphed using ggplot2 in R 3.4.3.

A generalized linear mixed model was used to determine the effect of initial water temperature, water temperature at second challenge, initial challenge suspension type and second challenge suspension on the prevalence of OsHV-1 in surviving oysters after the second challenge. For both models a stepwise backwards elimination process was used to build the model. Interactions were considered between the remaining variables. The tank identification was included in both models as a random effect to account for clustering. Plots of the residuals for both models were inspected to assess compliance with the assumptions of each model.

4.4 Results

4.4.1 Mortality and prevalence of OsHV-1

The total cumulative mortality (TCM) attributable to OsHV-1 was higher at 22°C (range 5.1 to 45.9% between groups) than at 18°C (range 0 to 8.7%) between groups regardless of first or second challenge with OsHV-1 (Table 4.1). The first challenge of oysters at 18°C was associated with reduced mortality after the second challenge at 22°C. The TCM for oysters that were held at 18°C for the first viral challenge then at 22°C for the second viral or control challenge (Group 1 and Group 2) was 5.1% and 20.4%, respectively (Table 4.1). The hazard of death for the former was 0.22 (95% CI: 0.06 to 0.80) times that of the latter (Figure 4.2,

Table 4.2), suggesting that prior challenge with OsHV-1 at 18°C protected against a later challenge at 22°C. After a second viral challenge at 22°C, the oysters which had survived the first viral challenge at 22°C (Group 4) had a TCM associated with OsHV-1 infection of 24.4% compared with 33.3% in oysters that had been injected with the negative control suspension on Day 0 (Group 5); this difference was not significant (Table 4.2). The TCM from all causes was 3 - 10% higher than the mortality associated with OsHV-1 (Table 4.1). Oysters challenged with OsHV-1 at 22°C then held at that temperature (Group 4, control) had ongoing mortality (7.7%) associated with OsHV-1 while those challenged at 18°C then held at 22°C

(Group 1, control) did not experience any mortality after the increase in temperature.

Table 4.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) after the initial and second challenge with Ostreid

herpesvirus 1 (OsHV-1) at a water temperature of either 18°C or 22°C. Data are total cumulative mortality and OsHV-1 prevalence

with 95% confidence intervals*.

Group Initial challenge N All-cause OsHV-1 Second Second N All-cause OsHV-1 N OsHV-1 conditions mortality % Mortality % challenge challenge mortality % Mortality % survivors prevalence in (95% CI) (95% CI) temperature (°C) (95% CI) (95% CI) survivors (95% CI) 1 18°C, OsHV-1 134 0.0 (0.0 - 2.7) 0.0 (0.0 - 2.7) 22 OsHV-1 99 10.1 (5.0 - 17.8) 5.1 (1.7 - 11.4) 89 68.5 (57.8 - 78.0)

Control 19 0.0 (0.0 - 17.6) - 19 26.3 (9.1 - 51.2)

2 18°C, Control 144 0.0 (0.0 - 2.5) 0.0 (0.0 - 2.5) 22 OsHV-1 113 23.9 (16.4 - 32.8) 20.4 (13.4 - 29) 86 55.8 (44.7 - 66.5)

Control 24 0.0 (0.0 - 14.2) - 24 0.0 (0.0 - 14.2)

3 18°C, OsHV-1 47 2.1 (0.1 - 11.3) 2.1 (0.1 - 11.3) 18 OsHV-1 23 13.0 (2.8 - 33.6) 8.7 (1.1 - 28) 20 60.0 (36.1 - 80.9)

Control 23 0.0 (0.0 - 14.8) - 23 47.8 (26.8 - 69.4)

4 22°C, OsHV-1 207 59.4 (52.4 - 66.2) 45.9 (39 - 52.9) 22 OsHV-1 41 34.1 (20.1 - 50.6) 24.4 (12.4 - 40.3) 27 77.8 (57.7 - 91.4)

Control 13 23.1 (5.0 - 53.8) 33.3 (0.8 - 90.6) 10 70.0 (34.8 - 93.3)

5 22°C, Control 143 0.0 (0.0 - 2.5) 0.0 (0.0 - 2.5) 22 OsHV-1 114 40.4 (31.3 - 49.9) 33.3 (24.8 - 42.8) 68 64.7 (52.2 - 75.9)

Control 24 100 (85.8 - 100) 8.3 (1.0 - 27.0) 0 -

6 18°C, Control 55 1.8 (0.0 - 9.7) 0.0 (0.0 - 6.5) 18 OsHV-1 27 3.7 (0.1 - 19) 3.7 (0.1 - 19) 86 34.6 (17.2 - 55.7) Control 26 0.0 (0.0 - 13.2) - 26 0.0 (0.0 - 13.2) *95% Confidence intervals can be practically interpreted to mean that the value is within this range with 95% confidence.

1.00

0.75

Prior exposure challenge: OsHV-1 at 18°C 0.50 Negative control at 18°C OsHV-1 at 22°C

Negative control at 22°C Survival probability Survival 0.25

0.00 0 48 96 144 192 240 Time (hours)

Initial challenge Initial challenge Coefficient Hazard ratio temperature (°C) inoculum estimate (95% CI) 18 OsHV-1 -1.51 0.22 (0.06 - 0.80) * Negative control - 1 22 OsHV-1 -0.32 0.73 (0.28 - 1.88) Negative control - 1 *95% Confidence interval does not include 1, indicating the result is significant (P <

0.05).

After the second challenge, all of the oysters in one of the control tanks/groups died associated with poor water quality due to a failed biofilter (Group 5 control). Although these had not been deliberately exposed to OsHV-1, 8.3% (2 out of 24) of these had detectable OsHV-1 in gill and mantle, presumably due to cross contamination between tanks. There was no mortality in the other non-exposed controls (Group 2 control and Group 6 control) and OsHV-1 was not detected in any of these oysters.

The prevalence of OsHV-1 infection in oysters that survived to Day 45 ranged from 34% to

78% among groups that were injected with the virus at Day 35 (Groups 1, 2, 3, 4, 5, 6) compared to a range of 26% to 70% in those injected with OsHV-1 only on Day 0 (Groups 1 control, 3 control and 4 control). The odds of a surviving oyster testing positive for OsHV-1 at

Day 45 were 2.3 times greater if they were initially challenged with OsHV-1 at 22°C compared to 18°C (Table 4.3). Interestingly, water temperature during the second challenge at

Day 35 had no significant association with the prevalence of PCR positive status in surviving oysters (P > 0.05) and was not included in the final model (Table 4.3).

Table 4.3: Odds ratios for the detection of Ostreid herpesvirus 1 (OsHV-1) in Pacific oysters

(Crassostrea gigas) to estimate the effect of water temperature, and OsHV-1 challenge and an

Parameter Estimate Std. Error P-value Odds ratio (95% CI) * Intercept -1.9809 0.2963 < 0.001 - Initial temperature 22°C 0.8214 0.2542 0.001 2.27 (1.3 - 4.08) Second challenge with OsHV-1 1.7875 0.2832 < 0.001 5.97 (3.37 - 12.95) Initial challenge with OsHV-1 1.0805 0.2274 < 0.001 2.95 (1.82 - 5.23) *95% Confidence intervals which do not include 1 are statistically significant (P <

0.05).

4.4.2 Quantity of OsHV-1 DNA in oyster tissues

The concentration of OsHV-1 DNA was much lower in the survivors compared to the dead oysters, regardless of whether there were one or two challenges with OsHV-1 (Figure 4.3).

Following the second challenge, the OsHV-1 DNA concentration in the survivors was

1.21x104 (geometric mean; 95% CI: 1.18x104 – 1.24x104 (95% CI) copies per mg and in the mortalities was 1.23x105 (geometric mean; 95% CI: 1.10 x105 – 1.37 x105) copies per mg.

The OsHV-1 DNA concentration in the survivors at Day 45, 10 days after the second OsHV-1 challenge, depended on the water temperature for the initial challenge at Day 0, and whether the initial suspension contained OsHV-1 (Table 4.4). There was a significant interaction between the water temperature at challenge and the suspension (OsHV-1 or negative control)

95 meaning that the effect of the initial challenge suspension on the amount of OsHV-1 DNA depended on the initial water temperature. For oysters which survived both challenges, the concentration of OsHV-1 DNA was higher if the initial OsHV-1 challenge temperature was

18°C rather than 22°C. For the oysters which were first challenged with a control suspension, the concentration of OsHV-1 DNA in the survivors was higher when they had been at 22°C rather than at 18°C (Figure 4.4). For dead oysters following the second challenge, the OsHV-1

DNA concentration depended only on the initial challenge suspension and was significantly lower in oysters which had already been challenged with OsHV-1 (Table 4.4, Figure 4.3 A).

Table 4.4: Linear mixed model describing the concentration of Ostreid herpesvirus 1 (OsHV-1) DNA in Pacific oysters

(Crassostrea gigas) at the time of mortality or in apparently healthy survivors 10 days post-challenge. Only individuals which were positive for OsHV-1 when sampled were included in the analysis.

Group Parameter Estimate Std. Error P-value Survivors Intercept 4.2046 0.1261 < 0.001 Initial temperature (22°C) 0.8917 0.1913 < 0.001 Initial challenge (OsHV-1) -0.4238 0.1689 0.023 Initial temperature (22°C) * initial challenge (OsHV-1) -2.0010 0.2987 < 0.001 Mortalities Intercept 5.6256 0.2248 < 0.001 Initial challenge (OsHV-1) -1.7028 0.4673 < 0.001

8 A

6 Survivors Mortalities 4

2 18°C, OsHV-1 18°C, Control 18°C, OsHV-1 22°C, OsHV-1 22°C, Control 18°C, Control at 22°C at 22°C at 18°C at 22°C at 22°C at 18°C

6 B

5 1 genome equivalents per mg tissue mg per equivalents genome 1

- Survivors

4 Mortalities OsHV 3

10 log 18°C, OsHV-1 18°C, Control 18°C, OsHV-1 22°C, OsHV-1 22°C, Control 18C, Control at 22°C at 22°C at 18°C at 22°C at 22°C at 18°C Initial challenge conditions and second challenge temperature

Figure 4.3: Ostreid herpesvirus 1 (OsHV-1) DNA copies per mg of Pacific oyster

tissue 5

per mg mg per Initial challenge 4 Control

OsHV-1

1 DNA copy DNA 1 -

OsHV

18 19 20 21 22 Log Water temperature during initial challenge (°C)

Figure 4.4: Estimated mean Ostreid herpesvirus 1 (OsHV-1) DNA copy number per mg tissue in surviving Pacific oysters (Crassostrea gigas) 10 days after a second challenge with

OsHV-1 at 22°C showing the interaction between water temperature during the initial challenge and the initial challenge inoculum in a linear mixed model (point estimate ± standard error).

4.5 Discussion

This study has shown that mortality due to OsHV-1 infection can be reduced by prior exposure to OsHV-1 when combined with a specific water temperature regime. This is evidence that the first exposure to OsHV-1 reduces the mortality resulting from exposure to the same pathogen when challenged for a second time. Immune priming has been previously observed in molluscs, administration the dsRNA poly I:C reduced mortality from a subsequent

OsHV-1 challenge but did not affect mortality from a challenge with the pathogenic bacteria

Vibrio splendidus (Green and Montagnani, 2013; Lafont et al., 2017). Immune memory of a specific pathogen has been observed in invertebrates with repeated challenges with the same parasite in copepods (Kurtz and Franz, 2003b). Consistent with previous laboratory and field studies, the outcome of OsHV-1 infection was highly temperature dependent (de Kantzow et al., 2016; Pernet et al., 2015; Petton et al., 2013; Renault et al., 2014).

Methods of experimental infection with OsHV-1 include cohabitation within the same tank as infected oysters (Evans et al., 2015; Schikorski et al., 2011b), immersion in or injection with an OsHV-1 suspension prepared from infected tissue (Hick et al., 2018; Schikorski et al.,

2011b), and field exposure to a natural epizootic (Pernet et al., 2015; Petton et al., 2013).

Exposing oysters to OsHV-1 in an aquarium system allows direct control over many environmental conditions including the timing and dose of the OsHV-1 exposure, water temperature, and the age, size and physiological state of the oysters. The reproducibility of the laboratory infection model used in this study was demonstrated by the similar outcome for oysters exposed at 22°C in the initial challenge to that in previous experiments (de Kantzow et al., 2016; Paul-Pont et al., 2015).

100

Random allocation of oysters in this study, accounted for variability in individual OsHV-1 susceptibility and ensured that each treatment group contained equal distributions of the variation that occur within each batch. Previous studies indicate high variability of OsHV-1 resistance within commercial batches of oysters (de Kantzow et al., 2016; Pernet et al., 2015;

Petton et al., 2013; Schikorski et al., 2011b, 2011a), including some with the benefit of selective breeding for resistance (Degremont et al., 2015). If previous treatment or exposure to

OsHV-1 had no effect on expression of viral infection, the expected total cumulative mortality would reach the same proportion in all of the groups following the second challenge with

OsHV-1 at 22°C. However, mortality following the second OsHV-1 challenge at 22°C was lower in oysters which had been previously challenged with OsHV-1 at 18°C compared to oysters which had been previously exposed to a negative control suspension. Thus, prior exposure history clearly reduced mortality. The protective effect observed in this study was attributed specifically to the presence of OsHV-1 in the initial suspension at 18°C based on the response of the control groups. The OsHV-1 suspension preparation may have contained immunogenic factors which were not removed by the filtration process or diluted beyond a meaningful dose during the suspension preparation which may have affected the response to the second OsHV-1 challenge. The suspension may have also contained some antiviral components from the donor oyster which had survived cryopreservation. The immunogenic factors were accounted for by using an equally filtered and diluted tissue homogenate as a negative control suspension and a control group was used to assess the effect of a physiological response to the change in water temperature from 18°C to 22°C.

Protection against mortality and lower OsHV-1 DNA concentration in survivors following the second challenge indicated that the response to OsHV-1 infection was altered by the initial

101 challenge with OsHV-1 at 18°C. Immune response pathways which are theoretically present in oysters including the JAK/STAT pathway, apoptosis, autophagy or RNA interference and a wide array of receptor molecules which recognize pathogen associated molecular patterns

(PAMPs) (Green et al., 2015a; Martenot et al., 2017; Moreau et al., 2015; Pauletto et al.,

2017; Schulenburg et al., 2007). The pathways may act together to provide greater specificity in higher animals may increase the antiviral effectiveness of the mollusc immune response

(Green et al., 2015a; Schulenburg et al., 2007). These and other aspects of the invertebrate immune response including RNA interference (Owens and Malham, 2015), humoral antimicrobial peptides and increased phagocytosis and autophagy (Moreau et al., 2015;

Pauletto et al., 2017; Rowley and Pope, 2012) have been experimentally linked to a change in mortality resulting from a viral infection in oysters. They may also be important in determining future mortality and immune memory. Further investigation is warranted to evaluate the immune mechanism that explains the results of the present study.

The high prevalence of OsHV-1 in surviving oysters and the absence of a second wave of mortality following the increase in water temperature from 18°C to 22°C contrasted with previous observations from one experiment in France. There, oysters which were exposed to

OsHV-1 at 18°C during a natural outbreak and then placed at 10°C or 13°C in a laboratory had a mortality of 35-83% when the temperature was increased to 21°C between 6 and 64 days later (Pernet et al., 2015). However, the current observations are in agreement with another experiment in France in which there was no recurrent mortality when oysters that had been exposed to a natural outbreak at 18°C then held at 13°C for 40 days were raised to 20°C

(Petton et al., 2013). The variable exposure conditions during natural disease outbreaks may influence both mortality and the OsHV-1 infection status of surviving oysters and might

102 explain the variability observed in mortality. The experimental infection model used in the current study removes this variability but at the cost of bypassing the mechanism of natural infection.

Although the mortality was reduced in the present experiment, the prevalence of OsHV-1 in survivors increased with each OsHV-1 challenge. A cumulative increase in OsHV-1 prevalence has not been documented in oysters subject to recurrent OsHV-1 exposure in endemic waters (Evans et al., 2017). Outbreaks generally occur in waters above 18°C, or 16°C in Europe, and there is no evidence of transmission below this temperature (Clegg et al., 2014;

Paul-Pont et al., 2014, 2013a; Pernet et al., 2012; Petton et al., 2015a; Renault et al., 2014).

Understanding drivers of OsHV-1 prevalence and persistence in exposed oysters and the causes of recurrent disease outbreaks in late spring and summer are important for effective disease control (Whittington et al., 2018). Although the detection of OsHV-1 by qPCR does not indicate the presence of infective virus, using this technology to investigate whether shedding of OsHV-1 occurs during infection would also be important to disease control.

Deliberate exposure to OsHV-1 at a water temperature that minimizes mortality presents a potential disease control strategy for OsHV-1 in farmed Pacific oysters in endemic areas. As the duration of the present study was only 45 days, further evaluation of the duration of the protective effect is required in both controlled conditions and in field experiments to establish the efficacy of the management strategy in a production setting in an endemic estuary. It will also be necessary to evaluate the model using a more practical exposure, for example by immersion or cohabitation, and to evaluate younger oysters. The addition of a specific immune priming infection to induce a response to survival over the period of greatest infection risk would bring management further into line with vaccine preventable diseases in

103 finfish aquaculture (Dadar et al., 2017). The period where oysters are at risk depends on the local conditions over spring and summer (Clegg et al., 2014; Paul-Pont et al., 2014; Pernet et al., 2012). A management calendar for OsHV-1 has been suggested based on reducing the number of highly susceptible stock during the period of greatest risk (Carrasco et al., 2017).

In conclusion, water temperature was confirmed to be an extremely important factor determining mortality after infection with OsHV-1 and also impacted the prevalence of subclinical (i.e. no mortality) infection, concentration of viral DNA and the response of oysters to subsequent challenge. A protective effect from OsHV-1 exposure at 18°C, which was permissive for infection but not disease, was demonstrated for a subsequent OsHV-1 challenge at 22°C. These observations are important to direct further investigations into the pathobiology of OsHV-1 infection and to inform the management of oysters farmed in OsHV-

1 endemic waters.

104

5. Chapter 5 - Different in vivo growth of Ostreid herpesvirus 1 at 18°C and

22°C alters mortality of Pacific oysters (Crassostrea gigas)

5.1 Abstract

Seasonally recurrent outbreaks of mass mortality in Pacific oysters (Crassostrea gigas) caused by microvariant genotypes of Ostreid herpesvirus 1 (OsHV-1) occur in Europe, New Zealand and Australia. The incubation period for OsHV-1 under experimental conditions is 48-72 hours and depends on water temperature, as does the mortality. An in vivo growth curve for

OsHV-1 was determined by quantifying OsHV-1 DNA at 10 time points between 2 and 72 hours after exposure to OsHV-1. The peak replication rate was the same at 18 °C and 22 °C; however, there was a longer period of amplification leading to a higher peak concentration at

22 °C (2.34 × 107 copies/mg at 18 hours) compared to 18 °C (1.38 × 105 copies/mg at 12 hours). The peak viral concentration preceded mortality by 72 hours and 20 hours at 18 °C and 22 °C, respectively. Cumulative mortality to day 14 was 45.9% at 22 °C compared to

0.3% at 18 °C. The prevalence of OsHV-1 infection after 14 days at 18 °C was 33.3%. No mortality from OsHV-1 occurred when the water temperature in tanks of oysters challenged at

18 °C was increased to 22 °C for 14 days. The influence of water temperature prior to exposure to OsHV-1 and during the initial virus replication is an important determinant of the outcome of infection in C. gigas.

5.2 Introduction

Recurrent outbreaks of disease with high mortality in Pacific oysters (Crassostrea gigas) are caused by microvariant genotypes of the species Ostreid herpesvirus 1 (OsHV-1) (Order

Herpesvirales, Family Malacoherpesviridae, Genus Ostreavirus) in endemic waters across

105

Europe, Ireland, New Zealand and Australia (EFSA, 2010; Jenkins et al., 2013; Keeling et al.,

2014; Ugalde et al., 2018; Whittington et al., 2015a). The disease has put considerable financial pressure on farming and research has focused on reducing the impact of the disease in regions where OsHV-1 is endemic (EFSA, 2010; Pernet et al., 2016; Ugalde et al., 2018).

The onset of mortality in estuaries where OsHV-1 is endemic is associated with seasonal increases in water temperature. In New South Wales, Australia the water temperature was

19°C – 24°C during the initial outbreaks (Paul-Pont et al., 2014, 2013a). A lower water temperature threshold of 16°C for OsHV-1 disease expression has been identified in France

(Pernet et al., 2012; Renault et al., 2014).

The severity of mortality and the effectiveness of transmission of OsHV-1 are strongly influenced by water temperature. In France, a water temperature of 14°C reduced transmission of OsHV-1 between cohabitating oysters while oysters challenged above 18°C and then held at 10°C - 13°C were not impacted by the infection until the temperature was increased (Petton et al., 2013). In Australia, reducing the temperature of estuarine water by 2 - 3°C from 20 -

25°C reduced the transmission of OsHV-1 to spat held in an onshore flow-through upweller

(Whittington et al., 2015b). Pacific oysters had no signs of disease when exposed to OsHV-1 at 14°C under laboratory conditions, and had low mortality (< 25%) at 18°C when exposed to a high dose (de Kantzow et al., 2016). Oysters exposed at 22-26°C had a higher concentration of OsHV-1 in their tissues, a shorter incubation period and the mortality was higher compared to 10-18°C, (de Kantzow et al., 2016; Pernet et al., 2015). Further investigation of the long- term outcomes for oysters which survive OsHV-1 exposure is important for understanding how recurrent epidemics occur. In a laboratory setting, 38% of oysters challenged with OsHV-

1 at 18°C remained infected 14 days after exposure (de Kantzow et al., 2016), and long term

106 carrier states occur in estuaries among the survivors of outbreaks (Evans et al., 2017). Oysters which survived OsHV-1 exposure in the field have subsequently died of the disease when recruited to experimental systems and exposed to an increase in water temperature (Pernet et al., 2015; Petton et al., 2013).

The outcome for oysters which do not immediately die because of infection with OsHV-1 may be determined by the conditions prior to and during infection and by the environment in the weeks following the infection. The infection status of oysters exposed below a permissive water temperature for disease (<18°C in Australia) was unclear as OsHV-1 was either not detectable or not quantifiable by real-time qPCR in the gill and mantle (de Kantzow et al.,

2016). A threshold exceeding 104 copies of OsHV-1 DNA per milligram of oyster tissue is generally indicative of OsHV-1 induced mortality in the field compared to lower concentrations in surviving oysters (Oden et al., 2011). The replication rate and concentration of OsHV-1 during the early stages of pathogenesis may have an important impact on the persistence of OsHV-1 and also the potential for mortality.

Herpesviruses of terrestrial animals (Family Herpesviridae) and fish (Family

Alloherpesviridae) cause latent infections which recrudesce to cause clinical disease when the host immune system is compromised or under conducive environmental conditions (Stevens,

1994). Latent infections occur when the infection is not cleared and the viral genome persists in a host cell, often only transcribing low levels of latency associated genes (LATs). Cyprinid herpesvirus 3 (CyHV-3) in carp (Cyprinus carpio) provides an example of herpesvirus latency in an aquatic host (Dishon et al., 2007), where the mortality associated with infection is also temperature-dependent (Eide et al., 2011b, 2011a; Xu et al., 2013). Observations consistent with latency of OsHV-1 have been observed in field exposure trials (Pernet et al., 2015; Petton

107 et al., 2013). However, these observations might also reflect persistent or recurrent infection of oysters which have very different immune systems compared to fish and mammals (Klein,

1989). Laboratory models for OsHV-1 infection of Pacific oysters enable control of the environment to determine the impact of individual factors on viral pathogenesis (Paul-Pont et al., 2015; Schikorski et al., 2011b). Experimental challenge by injection allows the pathogenesis of OsHV-1 to be observed under standardized conditions, without the influence of factors that impact transmission, co-infection, variable dose and timing of exposure that occur with cohabitation models and natural infection (de Kantzow et al., 2016; Paul-Pont et al., 2015; Schikorski et al., 2011a). Sampling with high temporal resolution is required to determine if variation in the initial replication of OsHV-1 contributes to the different disease outcomes at different water temperatures. The replication kinetics for OsHV-1 measured during the acute phase of infection at different water temperatures may explain the role of the environment on the later variable mortality outcome. The aim of this study was to address two key aspects of the effect of water temperature on infection with OsHV-1: i) to quantify the in vivo growth of OsHV-1 and identify differences in the replication rate and peak viral concentration at 18°C and 22°C; and ii) to evaluate whether a subclinical infection established at the lower temperature progresses to clinical disease after an increase in temperature.

5.3 Methods

5.3.1 Experiment design

5.3.1.1 Experiment 1, OsHV-1 growth curve

To create an in-vivo growth curve, the concentration of OsHV-1 DNA in gill and mantle tissue was measured over 3 days post exposure in oysters held at a water temperature of either 18°C

108 or 22°C. Prior to the experiment, oysters were randomly assigned to any of three replicate treatment tanks (n=16/tank) or two control tanks (one negative, one positive) (n=5/tank) at each water temperature, 18°C or 22°C (58 oysters total per temperature). This was done using a random number generator (www.random.org) to randomly select a tank number for each oyster until each tank reached quota. During the experiment, four oysters were randomly sampled from among the treatment tanks at 2, 4, 6, 8, 10, 12, 18, 24, 48 and 72 hours post exposure, with selection to ensure at least one was selected from each tank. Cumulative mortality was not assessed in the treatment tanks as most of the oysters were sampled before mortality commenced. The oysters in the positive control tank were used to determine the time of onset for mortality at each temperature and the concentration of OsHV-1 at that time. All oysters in the positive control tank were sampled when at the onset of mortality; the negative control oysters were monitored for 7 days post injection and sampled then, at the conclusion of the trial.

5.3.1.2 Experiment 2, temperature increase

Oysters challenged with OsHV-1 at 18°C were maintained at this temperature for 14 days then the water temperature was increased to 22°C over a period of 6 hours and maintained at 22°C for a further 14 days. Four treatment groups were used in the experiment (Figure 5.1). Three groups were challenged at 18°C: two with OsHV-1 and one with a negative control inoculum.

The fourth group was challenged with OsHV-1 at 22°C. Positive and negative control groups were maintained at 18°C for 28 days. All oysters were examined every 12 hours for 28 days following exposure and dead and moribund oysters were removed and placed at -80°C.

Oysters with shell valves that did not close after 5 minutes of air exposure were defined as dead.

109

The sample size required for each group at the time of the water temperature increase, after any mortality in the first 14 days after exposure was calculated to provide 80% power to detect a mortality difference of 10% in oysters inoculated with OsHV-1 at 18°C and 30% in oysters inoculated with OsHV-1 at 22°C, with α = 0.05 and RR = 2. Calculations were performed with the powerSurvEpi package in R 3.5.1 (R Core Team, 2017; Weiliang et al.,

2018). The sample size was then increased to account for an expected mortality of 30% at

18°C and 50% at 22°C after the initial exposure. The final sample size for Experiment 2 was:

134 (Group 1); 144 (Group 2); 47 (Group 3); and 207 (Group 4) (Figure 5.1, Table 5.2).

110

Water Group Exposure to Temperature End point temperature (No. Oysters / OsHV-1 increase No. tanks) 4 (207 / 8)

22°C

18 °C 1 (134 / 6) 2 (144 / 6) Inoculum 3 (47 / 2) OsHV-1 Control -7 days 0 days 14 Days 28 Days

Figure 5.1: Water temperature profile and inoculum used for each treatment group in the temperature change experiment

(Experiment 2).

111

5.3.2 Aquarium management

For each experiment Pacific oysters were procured and housed in tanks as described in sections 2.2.2 – 7.

5.3.3 Injection with OsHV-1

Oysters were exposed by injection with an inoculum prepared as per section 2.3. A fresh inoculum was produced for Experiment 1 as per Chapter 2.3.2 to avoid a freeze thaw cycle which may impact the initial virus replication, resulting in one additional in vivo passage compared to the cryopreserved stock. In Experiment 2, a different single use aliquot of the original cryopreserved stock was used.

5.3.4 Quantification of OsHV-1 DNA

OsHV-1 was quantified as per sections 2.4.3 - 4 and 2.4.7 – 9 in samples were taken from oysters as per section 2.4.1.

5.3.5 Statistical analysis

Statistical analysis was performed using SAS version 9.3 (SAS institute, Carey, NC, USA) and Microsoft R Open (R Core Team, 2017). Data were stored, aggregated and summarized using Microsoft SQL Server 2016 (Microsoft Corporation, Redmond, WA, USA). Exploratory statistical analysis was performed graphically using ggplot2 in R and numerically using SQL

Server 2016 and R 3.5.1.

5.3.5.1 Experiment 1

The viral DNA concentration determined by qPCR was log10 transformed for modeling and graphical presentation. Graphs of the OsHV-1 concentration at each temperature at each time

112 point were produced using the ggplot2 package in R (Wickham, 2009). The approach to the statistical analysis of the growth curve was similar to the method used to model lactation curves of cattle (Raadsma et al., 2009). Each parameter for the growth curve model was estimated for the 18°C data along with a corresponding parameter that is an adjustment to fit the 22°C data. This approach allows the adjustment parameters which are the difference between the growth curve at each temperature to be assessed for statistical significance. The growth curves were based on the basic growth curve form:

푁 푁 = 푚푎푥 푡 푁 − 푁 1 + ( 푚푎푥 0)푒−푟푡 푁0

where: Nt is the viral titer at time = t, Nmax is the maximum viral titer, N0 is the viral titer at t =

0, t is the time in hours and r is the growth rate. To determine the difference between the growth curves at each temperature, an ANCOVA approach was used in a non-linear framework to fit the model

푁 + 푁 푋 푁 = 푚푎푥 푚푎푥1 1 푡 (푁 + 푁 푋 ) − (푁 + 푁 푋 ) 1 + ( 푚푎푥 푚푎푥1 1 0 01 1 )푒−(푟+푟1푋1)푡 푁0 + 푁01푋1 to the transformed OsHV-1 concentration data using the nls function in base R. In this equation X1 is a dummy variable used to allow the inclusion of the data at a second water temperature with the parameters of the second water temperature denoted with a subscript 1

(Table 5.1). Graphs of the raw data with model predictions and the predicted verses residual values were used to assess the fit of the model.

113

5.3.5.2 Experiment 2

Mortality curves for the trial period were graphed in R using ggplot2. The difference in total cumulative mortality due to OsHV-1 following the water temperature increase (Days 14 – 28) was analyzed using a generalized linear mixed model (GLMM) in R using the glmer function in the lme4 and lmerTest packages (Bates et al., 2015; Kuznetsova et al., 2017). Mortality was used as the outcome variable, treatment group as the factor and the replicate tanks as a random effect in the model. Mortality was defined as an oyster which died between days 14 and 28 inclusive and had a level of OsHV-1 detectable by qPCR. Oysters that died without detectable

OsHV-1 were censored at the time of death. For analysis, a single mortality was added to each group with no mortality to allow the model to converge. Residuals were graphed and inspected to assess if the assumptions of the model had been met.

5.4 Results

5.4.1 Experiment 1 – growth curve

At 22°C, mortality was first observed 48 hours post exposure and all oysters in the positive control tank were dead or moribund at this time. At 18°C, the first mortality occurred at 84 hours post injection and all the positive control oysters were sampled at this time without determining the final cumulative mortality. The concentration of OsHV-1 DNA in the positive control oysters at 18°C when moribund or dead was 2.32 x106 copies.mg-1 (95% CI: 1.06x106

– 5.09x106) and at 22°C was 1.96x108 copies.mg-1 (95% CI: 1.68x108 – 2.28x108). Mortality did not occur in the negative control oysters at either temperature and OsHV-1 DNA was not detected in the negative control group at 18°C. OsHV-1 DNA was detected in 1 of 5 negative control oysters at 22°C at a concentration of 1.68x103 OsHV-1 copies.mg-1; this is thought to

114 be due to contamination after the live animal experiment during the processing of tissue samples.

In oysters in the 18°C treatment, OsHV-1 DNA was initially detected 6 hours post exposure, but was below the limit of quantification of the assay (<100 copies.mg-1). The concentration increased during the exponential growth phase to peak at 1.38x105 copies.mg-1 of gill and mantle tissue (95% CI: 3.45x104 - 5.51x105) at 12 hours post inoculation (Figure 5.2). At

22°C, OsHV-1 DNA was initially detected 4 hours post exposure (below the limit of quantification) and peaked 18 hours post exposure at 2.34x107 copies.mg-1 (95% CI: 1.16x107

7 - 4.72x10 ). The initial OsHV-1 concentration predicted by the model (N0, N0 1) was close to zero at both temperatures and was not significantly different between them, which agreed with the observed values in the gill and mantle tissue at 0 and 2 hours post exposure. Plateaus in the modelled viral DNA concentration occurred after 12 hours at 18°C and after 18 hours at 22°C, similar to the observed data. The maximum predicted concentration of OsHV-1 DNA was approximately 25 times higher at 22°C (9.08x106 copies.mg-1, 95% CI: 5.12x106 – 1.61x107) compared to 18°C (3.56x105 copies.mg-1, 95% CI: 8.35x105 – 1.52x105) (Table 5.1). These predicted values were also similar to the observed values up to 72 hours for each temperature.

However, there was a widespread in the OsHV-1 DNA concentration at each point in time.

The viral loads at both temperatures plateaued 20 hours before any oysters succumbed to mortality at 22°C and 72 hours before they succumbed at 18°C. There was no significant difference in the peak rate of OsHV-1 replication at 18°C or 22°C (P = 0.092), with the maximum replication rate approximately doubling the concentration every hour.

115

Description Parameter Estimate Std. Error P value Plateau at 18°C K 5.551 0.189 < 0.001 Change in 22°C plateau from 18°C K1 1.407 0.277 < 0.001

Starting concentration at 18°C N0 0.002 0.002 0.697

Change in 22°C starting concentration N0 1 0.129 0.09 0.154 from 18°C Growth rate at 18°C r 1.226 0.343 0.001 Change in 22°C growth rate from 18°C r1 -0.613 0.36 0.092

116

OsHV-1 concentration and the lines represent the non-linear growth curve model.

5.4.2 Experiment 2 - increasing temperature

During the first 14 days, there was one OsHV-1 related mortality out of 325 oysters (0.3%,

95% CI: 0.01% - 1.70%) in the groups exposed at 18°C. The OsHV-1 concentration was

8.99x105 OsHV-1 genome copies.mg-1 in this oyster at the time of death. In the 14 days following the increase in water temperature to 22°C, there was 3% (95% CI: 0.8% - 7.5%) and

4.9% (95% CI: 2% - 9.8%) mortality in the groups challenged at 18°C, but OsHV-1 was not detected at the time of death in any of these oysters (Table 5.2). Immediately prior to the water

117 temperature increase the prevalence of OsHV-1 was 33% (95% CI: 9.9% - 65.1%) in a random sample of oysters that were exposed to OsHV-1 at 18°C. The OsHV-1 concentration in these positive individuals was 1.90x103 copies.mg-1 (95% CI: 1.14x102 – 3.17x104).

In the group exposed at 22°C, the mortality due to OsHV-1 was 45.9% (95% CI: 39.0% -

52.9%) up to Day 13, and a further 23.8% (95% CI: 15.2% - 34.3%) died between Day 14 and

28 (Table 5.2). Up to Day 13 the concentration of OsHV-1 in the oysters at the time of death was 6.72x106 copies.mg-1 (95% CI: 4.46x106 – 1.01x107) and between Days 14 and 28 it was

9.28x104 copies.mg-1 (95% CI: 1.88x104 – 4.58x105).

118

Table 5.2: Mortality in Pacific oysters (Crassostrea gigas) from all causes and that due to Ostreid herpesvirus 1 (OsHV-1) for

oysters in Experiment 2. Data for the Days 0-13 after exposure are prior to the change in water temperature while data for Days 14-

28 are after the water temperature increase from 18°C to 22°C for Groups 1 and 2. The data for Days 14-28 are the proportion of

oysters which survived to Day 14. Data are pooled across 6 replicate tanks in Groups 1 and 2, 2 replicate tanks in Group 3 and 8

replicate tanks in Group 4.

Group Initial Final Exposure N Days 0-13 Days 14-28 temperature temperature (°C) (°C) All mortality % OsHV-1 mortality All mortality % OsHV-1 mortality (95% CI) % (95% CI) (95% CI) % (95% CI) 1 18 22 OsHV-1 134 0 (0 - 2.7) 0 (0 - 2.7) 3 (0.8 - 7.5) 0 (0 - 2.7) 2 18 22 Negative 144 0 (0 - 2.5) 0 (0 - 2.5) 4.9 (2 - 9.8) 0 (0 - 2.5) 3 18 18 OsHV-1 47 2.1 (0.1 - 11.3) 2.1 (0.1 - 11.3) 0 (0 - 7.7) 0 (0 - 7.7) 4 22 22 OsHV-1 207 59.4 (52.4 - 66.2) 45.9 (39 - 52.9) 35.7 (25.6 - 46.9) 23.8 (15.2 - 34.3)

119

Table 5.3: Generalised linear mixed model (GLMM) analysis of Pacific oyster (Crassostrea

gigas) mortality between Days 14 and 28 for the temperature increase experiment. The odds

ratio represents the difference in mortality between the treatment groups after the water

temperature was increased from 18°C to 22°C in two of the groups. Tank was included as a

random effect in the model.

Parameter (Group) Estimate Std. Error P-value Odds Ratio (95% CI) Intercept -4.89 1 < 0.001 - OsHV-1 at 18°C, increase to 22°C (Reference) - - - 1 Control at 18°C, increase to 22°C -0.07 1.42 0.96 0.93 (0.06 - 15.08) OsHV-1 at 18°C 1.08 1.42 0.45 2.94 (0.18 - 47.62) OsHV-1 at 22°C 3.73 1.04 < 0.001 41.68 (5.43 - 320.03)

5.5 Discussion

The results of this study indicated that the replication rate of OsHV-1 in C. gigas was the

same at both 18°C and 22°C, but that viral replication occurred for a longer period at 22°C

compared to at 18°C. The plateau phase of replication was reached 6 hours later at 22°C than

at 18°C and at a higher concentration of OsHV-1. This is consistent with previous studies

where greater viral loads in oyster tissues were observed at the time of mortality at 22°C

compared to 18°C (de Kantzow et al., 2016; Pernet et al., 2015; Petton et al., 2013). Despite

33% of oysters still containing OsHV-1 at 14 days following exposure at 18°C to OsHV-1 by

injection, an increase in water temperature to 22°C did not result in mortality of these oysters

over the following 14 days. Combined, the results from the two experiments suggest that

OsHV-1 can infect oysters at 18°C and 22°C, but replication, pathogenesis and host response

are different depending on the water temperature. Consequently, estuarine water temperature

120 before and during OsHV-1 exposure may affect the outcome of OsHV-1 epidemics and the infection status of the survivors.

The cumulative mortality due to OsHV-1 in this study was variable between experiments and compared to previous studies. In Experiment 2, mortality due to OsHV-1 at 22°C was lower

(46%) than the 60-70% previously reported from both laboratory and field experiments (de

Kantzow et al., 2016; Pernet et al., 2015; Petton et al., 2013; Schikorski et al., 2011a, 2011b).

The age, size and genetic background of the oysters used may have influenced the level of mortality (de Kantzow et al., 2016; Dégremont, 2011; Hick et al., 2018). Two different batches of oysters were used and although the source hatchery and growing conditions were similar, there are likely to be differences in susceptibility between oysters of different genetic origin (Segarra et al., 2014c). These factors may have also influenced the mortality in the positive control for the growth curve where mortality was comparatively high (100%) at 22°C.

The use of a fresh OsHV-1 inoculum may also have impacted the observed mortality, even though the dose was standardized after quantification by qPCR. Storage and dose of OsHV-1 inoculum have both been shown to influence mortality (de Kantzow et al., 2016; Paul-Pont et al., 2015).

The lower peak viral DNA concentration at 18°C compared to 22°C suggests that viral replication ceases earlier at 18°C compared to at 22°C. Following injection with poly I:C, anti-viral gene expression in oysters at 12°C was delayed compared to oysters at 22°C (Green et al., 2014b). Altered immune gene expression at different water temperature might explain the difference in OsHV-1 replication kinetics. Expression of antiviral genes including apoptosis inhibitors at 12 and 24 hours after infection with OsHV-1 at 22°C was demonstrated by Segarra et al. (2014c). Additionally, the biochemistry governing viral replication may also

121 have limited viral replication earlier, due to lower availability of substrates for enzymes or lower enzymatic activity at 18°C compared to 22°C. A shorter exponential growth phase for

OsHV-1 may contribute to the lower mortality observed at 18°C and indicates that this initial stage of replication is important in determining the outcome of the infection. The present study focused on quantification of viral DNA as an indicator of OsHV-1 replication, which was likely to influence the pathogenesis of infection. It has been demonstrated that the quantity of viral DNA is closely correlated with infectivity titer for another temperature dependent aquatic herpesvirus, CyHV 3 (Mletzko et al., 2017). Previously, field and experimental studies have identified a threshold concentration of 104 OsHV-1 genome copies.mg-1 associated in oysters that succumb to mortality (de Kantzow et al., 2016; Oden et al., 2011; Schikorski et al., 2011b). The growth curve determined in Experiment 1 indicated this threshold could be reached as early as 6 hours after challenge by injection at 22°C or after

8 hours at 18°C. Interactions between the host and pathogen in the initial phase of infection that are under the influence of environmental conditions determine the peak viral titer before mortality occurs. Apparently healthy oysters at 18°C that were sampled on 14 days after

OsHV-1 challenge had a moderate proportion (2/12) with OsHV-1 DNA concentration above

104 genome copies.mg-1. As there was no mortality caused by OsHV-1 after Day 14, this suggests that some oysters can tolerate a viral concentration that exceeds this threshold at some stage of the infection and go on to survive.

The initial environmental conditions may be important for determining the long-term outcome for survivors of OsHV-1 exposure. The absence of a peak in mortality following the increase in water temperature from 18°C to 22°C in oysters which had been exposed to OsHV-1 at

18°C contrasted with a study by Pernet et al. (2015). In the later study, high mortality

122 followed a more extreme thermal profile with a shift from 13°C to 21°C following exposure at

18°C. The magnitude of temperature change or other uncharacterized stress factors might influence the thermal challenge test described for identification of subclinical OsHV-1 infection (Normand et al., 2014a). Further investigation is required to establish if latent infection under the control of latency associate transcripts occurs for OsHV-1 and what factors would influence recrudescence (Segarra et al., 2016). The growth curve indicates that OsHV-1 replication peaks before mortality and suggests that the temperature in the initial 12 to 24 hours of infection is important for determining both initial mortality and the long-term outcome. The diurnal changes in water temperature in an estuary may be enough to alter mortality given that the maximum and minimum temperatures may be 5°C above or 2 -3°C below the daily average in Australian estuaries (Whittington et al., 2015b). The current study highlights a possible difference in temperature response for OsHV-1 in oysters between

Australia and Europe, as recrudescence does not appear to occur in Australia at a similar temperature to that which caused apparent recrudescence in European conditions (Pernet et al., 2015). The effect of changing water temperatures during the period prior to and immediately after exposure to OsHV-1 requires further investigation as a potential disease control method and for understanding mortality patterns during outbreaks.

The results of the present experiments establish that a higher concentration of OsHV-1 in oysters infected at 22°C compared to 18°C arose due to a longer period of replication during the initial stages of infection. The non-linear ANCOVA model provides an appropriate statistical approach to determine the influence of different factors that can affect parameters of the viral growth curve. Experiment 2 demonstrated that 6 month old Pacific oyster spat which were exposed to OsHV-1 at 18°C experienced low mortality (< 5%), and whilst a subset

123 remained PCR positive, there was no disease expression when water temperature was raised to

22°C. This observation in a controlled laboratory environment overcomes the influence of confounding factors that impact field surveillance to examine the effect of a change in water temperature on mortality for oysters previously infected with OsHV-1. These data provide an insight into the important impact of water temperature on the early replication of OsHV-1 and the subsequent pathogenesis of disease. Control strategies for OsHV-1 will be devised with knowledge of the influence of water temperature on the disease outcome, both at the time of infection and subsequently during a persistent or latent infection (de Kantzow et al., 2019a). In conclusion, the growth rate of OsHV-1 is not dependent on temperature in the range 18-22°C.

However, replication continued for a longer period before reaching a plateau at a higher concentration at 22°C compared to 18°C. Further studies of the immune responses of oysters to OsHV-1 at different water temperatures will generate an improved understanding of the pathogenesis of disease caused by this virus.

124

6. Chapter 6 - Ostreid herpesvirus 1 (OsHV-1) replication and Pacific oyster

(Crassostrea gigas) immune gene expression during the first 24 hours of

infection at 18°C and 22°C

6.1 Abstract

Water temperature is a strong determinant of mortality of Pacific oysters (Crassostrea gigas) infected with Ostreid herpesvirus 1 (OsHV-1), however, the influence of water temperature on expression of host immune genes following OsHV-1 exposure is unknown. Additionally, the current sampling methods used to measure gene expression in target tissues are lethal, so the outcome of OsHV-1 infection cannot be assessed in the same oysters during the acute stage of infection. The aim of this experiment was to compare oyster immune gene expression in response to OsHV-1 at two different water temperatures and to reconcile this with the outcome of infection in the same individuals. To achieve this, oysters maintained at 18°C or

22°C were exposed to either OsHV-1 or a negative control. A novel biopsy method was used to take gill samples at 0, 2, 6 12- and 24-hours post-exposure for RNA purification.

Expression of selected host immune genes was quantified using RT-qPCR and fold change in gene expression was calculated using the delta-delta Ct method. Mortality in oysters exposed to OsHV-1 at 18°C was 18.3% (95% CI: 10.5 - 30.2%) and at 22°C was 63.3% (95% CI: 50.5

- 74.5%) at 14 days post-exposure. This corresponded to a hazard ratio for mortality of 5.19

(95% CI: 2.61 - 10.32%) at 22°C compared to 18°C. Following OsHV-1 exposure at 18°C there was variation in the expression of Viperin (Χ2 = 17.90, df = 3, P < 0.001), IAP repeat- containing protein 2 (IAP) (Χ2 =35.51, df = 3, P < 0.001), and Myeloid differentiation factor

88 (MyD88) (Χ2 = 17.38, df = 3, P < 0.001) between timepoints with upregulation at 2 hours post exposure and down regulation at 24 hours post exposure. At 22°C, there was lower

125 expression of Viperin (Χ2 = 12.60, df = 3, P = 0.005) and IAP (Χ2 = 26.01, df = 3, P < 0.001) at 2 hours post exposure than 24 hours post exposure. While MyD88 expression was also highest at 2 hours post exposure at a water temperature of 22°C it was lowest at 6 hours post exposure (Χ2 = 25.61, df = 3, P < 0.001). IAP expression was lower in oysters which went on to die due to OsHV-1 in the 14 days post exposure (Χ2 = 6.01, df = 1, P = 0.014). This suggests that the immune responses at different water temperatures explains, in part, the influence of water temperature on mortality with OsHV-1 infection.

6.2 Introduction

Disease caused by microvariant genotypes of Ostreid herpesvirus 1 (OsHV-1) have caused recurrent mortality of up to 100% in farmed Pacific oysters (Crassostrea gigas) in Europe, including France, Ireland, Spain (Clegg et al., 2014; EFSA, 2010), Australia (de Kantzow et al., 2017; Jenkins et al., 2013; Paul-Pont et al., 2014) and New Zealand (Keeling et al., 2014;

Webb et al., 2007). The water temperature has a strong, direct effect on the level of mortality caused by OsHV-1 (de Kantzow et al., 2016; Martenot et al., 2015a; Pernet et al., 2015; Petton et al., 2013). This effect is an important factor in the epidemiology of OsHV-1, with outbreaks occurring seasonally when the water temperature rises above 18°C to 22°C in Australia or

16°C in Europe (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a; Pernet et al., 2012; Petton et al., 2015a; Renault et al., 2014; Whittington et al., 2019). The seasonal nature of the outbreaks has led to changes to the production calendar to move to spat placed on the farm in autumn so that they are as old and large as possible during the periods with high water temperatures in summer (Carrasco et al., 2017). Oysters are ectothermic and their physiology, including their immune system is impacted by the water temperature (Green et al., 2014b). Exploring the interactions between the host and pathogen at different water temperatures will provide an

126 understanding of different outcomes with higher mortality at 22°C compared to 18°C and no evidence of infectivity below 14°C (de Kantzow et al., 2016; Petton et al., 2013).

The outcome of OsHV-1 infection is influenced by viral replication kinetics and the interaction the virus has with the host immune response (Green et al., 2015b; Schikorski et al.,

2011a). There are three possible outcomes: mortality of the oyster; survival with persistent

OsHV-1 detectable by qPCR; or survival with apparent freedom from OsHV-1 (Evans et al.,

2017; Green et al., 2015a; Green and Speck, 2018; Petton et al., 2015a). The antiviral immune response of Pacific oysters utilises apoptosis, and autophagy, RNA interference (RNAi) and antiviral effectors or peptides (Green et al., 2015a; Green and Speck, 2018). Apoptosis autophagy aims to kill an infected cell to limit viral replication and the potential for the virus to spread to other cells. These processes are triggered within cells, including haemocytes, by the presence or replication of OsHV-1 (Allam and Raftos, 2015; Green et al., 2015a; Moreau et al., 2015; Picot et al., 2019; Wang et al., 2018). Apoptosis inhibitors are present in the

OsHV-1 genome and are expressed early during infection to overcome the apoptosis aspect of the immune response and favour continued viral replication (Green et al., 2015b; Segarra et al., 2014c). Baculoviral IAP repeat-containing protein 2 (IAP) is an apoptosis inhibitor present in Pacific oysters which is upregulated during OsHV-1 infection and in oysters which have survived an OsHV-1 challenge (Segarra et al., 2014b). Antiviral effectors such as viperin are proteins encoded by the host genome which produce direct antiviral activity. Viperin is an extensively studied antiviral effector in response to OsHV-1 (Green et al., 2015c; Green et al.,

2016; Green and Speck, 2018; Rosani et al., 2015). The presence of viperin inhibits OsHV-1 replication but the mechanism is unknown (Dang et al., 2015; Minton, 2018). The synthetic double stranded RNA (dsRNA) analogue Poly I:C induced greater upregulation of antiviral

127 immune genes at 22°C compared to 12°C suggesting that the antiviral response is water temperature dependent (Green et al., 2014b).

Immune signalling pathways within and between host cells have been hypothesised based on gene expression in response to the immune stimulus provided by experiments using OsHV-1 infection and injection of poly I:C (Green et al., 2015a; Green and Montagnani, 2013; He et al., 2015; Jouaux et al., 2013; Segarra et al., 2014a). Immune signalling is initiated by detection of a pathogen associated molecular pattern (PAMP) by a pathogen recognition receptor (PRR). For example, Toll like receptor 3 (TLR3) recognises the PAMP dsRNA which may be produced during viral replication and is upregulated in response to poly I:C

(Green and Montagnani, 2013; Jacobs and Langland, 1996; Segarra et al., 2014b; Weber et al.,

2006). TLRs then require a signalling protein to generate a downstream immune response within the cell, such as Myeloid differentiation factor 88 (MyD88), which has been demonstrated to be upregulated during infection with OsHV-1 or exposure to poly I:C (Du et al., 2013a; Green and Montagnani, 2013; Segarra et al., 2014c, 2014b; Tang et al., 2017). The presence of viral infection is signalled to surrounding cells via the interferon pathway which also induces expression of antiviral proteins such as viperin in the surrounding cells (Green and Speck, 2018). Immune regulation genes including interferon regulatory gene 2 (IRF2) are also upregulated during an OsHV-1 infection (Green and Montagnani, 2013; Renault et al.,

2011; Rosani and Gerdol, 2017). Interferon induced immune activity in surrounding cells is suggested to occur through the JAK/STAT immune pathway which recognises Interferon

Induced Protein 44 (IFI44) as an intracellular signalling protein (Green and Montagnani,

2013; Green and Speck, 2018; He et al., 2015). The immune pathways and interactions between the oyster and OsHV-1 require further examination under controlled conditions to

128 understand how they are altered by water temperature. The length of time OsHV-1 replicates and the peak viral DNA concentration are temperature dependent (de Kantzow et al., 2019b).

The level of host gene expression during the period of replication may determine mortality and similarly be influenced by water temperature (Green et al., 2014b).

Current sampling methods for gene expression in Pacific oysters have prevented longitudinal sampling of individuals because they are either lethal or require relaxation in a magnesium chloride solution which may affect the course of disease. The current experiment used a non- lethal sampling method which allowed gill biopsies to be taken during the acute stages of infection and for the individual to be followed until the endpoint of the infection. A hole drilled in the shell prior to the experiment allowed the gill tissue to be accessed for a biopsy to be taken with less impact on the oyster and more accurate timing than relaxing the adductor muscle with a magnesium chloride bath. This enabled the concentration of OsHV-1 DNA and gene expression to be determined during the acute stages of infection together with determining the subsequent survival of the individual at the completion of the disease course.

Longitudinal gene expression data compared against the final survival outcome are important for understanding factors which influence the outcome of infection with OsHV-1. The aim of the present experiment was: (i) compare the expression of key immune genes and OsHV-1 concentration during the acute stages of a controlled laboratory infection with the outcome of infection in the same individual oysters, and (ii) determine if differential expression of key oyster immune genes at different water temperature corresponds with different mortality outcomes from OsHV-1 infection.

6.3 Method

6.3.1 Aquarium management

129

Triploid Pacific oysters (50 – 60 mm) (section 2.2.2) were housed in tanks managed as described in section 2.2 for the duration of the acclimation period and experiment. The oysters were acclimated to either 22°C or 18°C for 7 days (Figure 6.1).

6.3.2 OsHV-1 challenge

A cryopreserved stock of OsHV-1 was produced as described in section 2.3.1 and used to challenge the oysters as described in section 2.3.3-4.

6.3.3 Sampling procedure

An access portal was created in the side inferior valve of each oyster 7 days prior to the

OsHV-1 challenge to allow for the injection challenge and non-lethal tissue sampling. This was done using a 10 mm diamond core drill bit (Sutton Tools) in a drill press, the drill bit had a flat cutting surface which caused minimal tissue damage and no oysters died as a result of the procedure (Figure 6.1). At the time of sampling, the portal was used to access the gill with

Tischler biopsy forceps (3x7 mm, Armo) which were used to remove 60 mg of gill tissue. The tissue sample was placed in 500 µL of RNALater (Sigma) in a 1.5 mL screw cap tube. The sample was fixed in RNALater for 48 hours at 4 °C and then stored at -20 °C until nucleic acid extraction. The tissue sample handled as per section 2.4.2.

130

Sampling portal Gill

Inferior valve Viscera Adductor muscle

Figure 6.1: The location of the gill and mantle tissue biopsy taken from sampled Pacific oysters (Crassostrea gigas) for Ostreid herpesvirus 1 (OsHV-1) quantification.

131

Sampling OsHV-1 Water Group (N/tanks) Challenge Mortalities temperature Mortalities N = 60/3 1 Survivors N = 60/3 Mortalities 22°C 2 Survivors Mortalities N = 60/3 18 °C 3 Survivors Inoculum 4 N = 60/3 Survivors OsHV-1 Control 0 2 6 12 24 Drill 336 186 Hours access Hours Hours Hours Hours Hours acclimation Hours portal

Figure 6.2: Experiment design indicating the sampling time points and sample size for each group.

132

6.3.4 Experiment design

There were four treatment groups defined by the water temperatures of 22°C and 18°C and

OsHV-1 exposure, each containing 60 oysters divided equally across 3 replicate tanks per treatment. At each water temperature the oysters in 3 tanks were exposed to OsHV-1 and the oysters in the other 3 tanks were exposed to the negative control by intramuscular injection.

Immediately prior to exposure and at 2, 6, 12 and 24 hours post-exposure a biopsy was taken from 12 oysters in each group (Figure 6.2). The oysters were monitored for 14 days post- exposure with dead and moribund oysters removed every 12 hours. A second gill biopsy was taken at the time of death, or at the completion of the trial in surviving oysters. The target genes Viperin, IAP and MyD88 were chosen as they represent diverse aspects of the Pacific oyster immune response and have all been studied in association with OsHV-1 infection.

However, the effect of water temperature on the expression of these genes has not been previously investigated.

6.3.5 Tissue homogenisation

Prior to tissue homogenisation, the tissue samples were removed from the ethanol by manipulating with a sterile 20 µL pipette tip and were placed in a 2.0 mL tube containing 600

µL of RLT buffer with 0.5% v/v DX reagent (Qiagen), 1% v/v β – mercaptoethanol (Qiagen) and 0.4 g of 0.1 mm silica-zirconia beads (Daintree Scientific). The tissue samples were mechanically homogenized by bead beating using a TissueLyser (Qiagen) for 240 s at 30 oscillations per second with an inversion of the tube holding insert after 120 s. The samples were then centrifuged at 900 g for 10 mins. A 50 µL aliquot of supernatant was taken for immediate nucleic acid extraction and the remainder stored at -80°C.

133

6.3.6 Total nucleic acid purification

Total nucleic acids were purified from 50 µL of the sample supernatant using the MagMAX-

96 Viral RNA Isolation Kit (Life Technologies) and a magnetic particle processor (MagMAX

Express-96 Applied Biosystems) into 75 µL of elution buffer. Purified nucleic acids were stored at -80°C prior to preparation for RT-qPCR for gene expression and OsHV-1 DNA quantification. A negative control well with no sample was included on each plate and was subsequently tested for OsHV-1 DNA by qPCR. The concentration of dsDNA in each nucleic acid extract was measured using a Qubit fluorometer (Thermofisher Scientific).

6.3.7 DNase treatment of nucleic acids

Aliquots of the total nucleic acid extracts were treated with DNase 1 (Sigma) to remove genomic DNA prior to reverse transcription. An 8 µL aliquot of total nucleic acid extract

(containing less than 2 mg DNA) was treated with 2 µL of DNase 1 (1 unit/µL) in a total volume of 20 µL with 2 µL of 10x DNase reaction buffer (Sigma) and 8 µL of RNase free water (Life Technologies). The reaction was incubated at 37°C for 20 minutes before 2 µL of stop solution (50 mM EDTA, Sigma) was added and the DNase was deactivated by incubation at 70°C for 10 minutes. RNA in the samples was then purified by ethanol precipitation to remove the DNase and reaction buffers. For the precipitation, 2.2 µL of 3M sodium acetate and 55 µL absolute ethanol was added to the 22 µL of reaction products and incubated at -

20°C overnight. The samples were then centrifuged at 16,000 g for 10 mins at 4°C and the supernatant was discarded. The pellet was washed twice with 50 µL of 70% ethanol at -20°C with centrifugation at 16,000 g for 3 mins at 4°C between washes. The pellet was then dissolved in 20 µL of ultrapure DNase and RNase free water. The concentration of ssRNA in each sample was measured using a Qubit fluorometer with a ssRNA kit (Thermofisher

134

Scientific) according to the manufacturer’s directions. The purity of the RNA was assessed by spectrophotometry using a NanoDrop1000 (Thermofisher Scientific). Extracts with an

A260/A280 ratio greater than 2 and an A260/A230 ratio between 2 and 2.2 were considered free of contamination. The absence of DNA in each sample was confirmed negative results when the samples were tested for the Pacific oyster housekeeper gene (EF1- α) according to the method described in section 6.2.9, without reverse transcription.

6.3.8 cDNA synthesis

Reverse transcription was performed using the SensiFAST cDNA synthesis kit (Bioline) according to the manufacturer’s instructions. Briefly, each reaction contained 5 µL of sample

RNA (500 ng), 4 µL of 5x TransAmp buffer, 1 µL Reverse transcriptase (200 u/µL) and 10

µL of ultrapure DNase and RNase free water. The reactions were mixed by pipetting and incubated in a thermocycler (Corbet) at 25°C for 10 minutes then, 42°C for 15 minutes and

48°C for 15 minutes followed by 85°C for 5 minutes; then held on ice. The cDNA was diluted

1 in 5 with ultrapure water and stored at -20°C until quantification by qPCR.

6.3.9 Gene expression qPCR

Relative expression levels of selected Pacific oyster and OsHV-1 genes were assessed by qPCR (Table 1) using modifications of previously described assays (Segarra et al., 2014). The host immune genes under evaluation were selected because they had were reported to be differentially expressed in oysters in response to OsHV-1 exposure or in oysters exposed to poly I:C (Green and Montagnani, 2013; Segarra et al., 2014c). The Pacific oyster Elongation factor 1-α (EF1-α) gene was measured as a housekeeper according to a previously described method (Du et al., 2013b). Reactions were prepared in a 10 µL volume containing 5 µL of 2x

135

Quantitech low ROX SYBR ([2-[N-(3-dimethylaminopropyl)-N-propylamino]-4-[2,3-dihydro-

3-methyl-(benzo-1,3-thiazol-2-yl)-methylidene]-1-phenyl-quinolinium]) reaction buffer

(Qiagen), 200nM forward and reverse primers, 2.96 µL DNase/RNase free water and 2 µL template cDNA. The reactions were run in a 7500 FAST (ABI), Mx3000P (Stratagene) or

AriaMx (Stratagene) thermocycler. Each experimental sample was run in two technical replicates on the same plate for each target. A standard positive sample from an oyster infected with OsHV-1 and no template control was included in each qPCR run. The reaction conditions were 95°C for 10 minutes followed by 40 cycles of denaturation at 94°C for 15s, annealing at 60°C for 30s and extension at 72°C for 30s. A dissociation curve was then run from 60°C to 95°C and the melting temperature (Tm) was defined as the temperature where there was a reduction in fluorescence. Samples which had an exponential increase in ROX normalised, baseline corrected SYBR fluorescence and a Tm within ± 0.5°C as the positive control sample used for the standard curve for that target were considered to have target specific amplification for the gene of interest. A fluorescence threshold that passed through the exponential phase of each positive sample was calculated for each plate such that a reference sample included on each plate had the same Ct for the same target gene across each plate. The cycle threshold (Ct) for each other sample was defined as the cycle where the fluorescence for that sample exceeded the fluorescence threshold. The efficiency of the reaction for each target was calculated using a standard curve created with a 5-step 5-fold dilution series of cDNA from a sample with a high concentration of target cDNA.

136

Table 6.1: Primer sequences used to quantify Pacific oyster (Crassostrea gigas) host immune gene expression by RT-qPCR.

Gene Forward Reverse Amplicon (bp) Protein and expression stage Reference EF-1α AGTCACCAAGGCTGCACAGAAAG TCCGACGTATTTCTTTGCGATGT 200 Translation (Segarra et al., 2014c) (housekeeper) (Green and Montagnani, MyD88 AGGTACCGGCTGTGATACGA TTCAAACGCCACCAAGACTG 217 Myeloid differentiation factor 88 2013) Viperin GCTTTGACCCGGAAACCAAC TGACACCAATCCCGAACTCG 97 Anti-viral effector (Green et al., 2014b) Baculoviral IAP repeat- IAP CCCGAAAACGTAACCTCAGA TTTCGTTTGCTGCTCATTTG 288 (Segarra et al., 2014c) containing protein 2

137

6.3.10 Quantification of OsHV-1 DNA

The number of copies of OsHV-1 DNA in the samples was quantified as described in section

2.4.7.

6.3.11 Statistical analysis

Statistical analysis was performed using R version 3.5.1 (R Core Team, 2017). Exploratory data analysis and data cleaning was done using the tidy verse packages (Müller and Wickham,

2017; Wickham, 2011, 2009; Wickham et al., 2017b). A Kaplan-Meier survival curve was calculated using the survminer package (Kassambara and Kosinski, 2017). A Cox proportional hazards model was used to assess the effect of water temperature and OsHV-1 DNA concentration at the time of biopsy on mortality. The failure condition for the model was mortality during the follow-up period. Only oysters exposed to OsHV-1 were included in this model. Both variables were included in the initial model and OsHV-1 DNA concentration at the time of biopsy was then removed from the final model as it was not a significant predictor of mortality (P < 0.05) was removed in a backward stepwise approach. The Cox proportional hazard model was run using the coxph function in the survival package for R (Therneau,

2015) and the Schoenfeld residuals were plotted to assess the assumption of proportional hazards.

To evaluate the assumption that expression of the Pacific oyster housekeeper gene, EF1-α, was consistent across the experimental conditions, a Kruskal-Wallis rank sum test was used to compare the EF1-α Ct values between the two water temperatures, OsHV-1 exposure status and biopsy time point. The coefficient of variation (CV) of the Ct values for EF1-α for the same groupings was calculated by dividing the standard deviation by the mean. Where EF1-α

138 or the gene under investigation could not be amplified the sample was omitted from fold change calculation.

The relative expression of each Pacific oyster immune gene under evaluation was calculated separately relative to the housekeeper, EF1-α, to compare between the OsHV-1 exposed and negative control groups using the delta-delta Ct (∆∆Ct) method (Livak and Schmittgen, 2001).

The relative expression was calculated separately for oysters at each water temperature and sampling time point after exposure. Differences in the housekeeper normalised target gene expression for treatment groups were calculated and raised to the power of 2 to calculate fold change according to the equation:

퐹퐶 = 2−[(푇푒푥푝표푠푒푑−퐻푒푥푝표푠푒푑)−푎푣푔(푇푐표푛푡푟표푙−퐻푐표푛푡푟표푙)]

where: FC = fold change, T1 = Ct of the target gene in Group 1, H1 = Ct of the housekeeper gene in Group 1, T2 = Ct of the target gene in Group 2, H2 = Ct of the target gene in Group 2.

The delta Ct values (target – housekeeper) were averaged across each negative control group and used in the calculation with the delta Ct values from individual OsHV-1 exposed oysters.

This was done because there was no natural pairing individual OsHV-1 exposed and negative control oysters which could be used as the basis for pairing in the calculation. The fold change in the expression for each gene at different water temperature and timepoint was plotted using the ggplot2 package in R. The fold change equation using the ∆∆Ct value was calculated in R.

The Kruskal-Wallis rank sum test was then used to assess the differences in fold change of each gene between sampling timepoints, across water temperatures and between mortalities and survivors.

6.4 Results

139

Mortality first occurred 60 hours after the oysters were exposed to OsHV-1 at 22°C, whilst for oysters at 18°C mortality did not occur until 120 hours after exposure (Figure 6.3). The total cumulative mortality after OsHV-1 exposure was much lower for the oysters at 18°C (18.3%;

95% CI: 10.5 - 30.2) compared to 22°C (63.3%; 95% CI: 50.5 - 74.5) (Table 6.2). Mortality did not occur in the oysters which were injected with the negative control tissue homogenate at either 18°C or 22°C, indicating that this and the biopsy procedure did not cause mortality.

Water temperature was a significant predictor of oyster mortality: the hazard of death following the OsHV-1 exposure was 5.19 (95% CI: 2.61 - 10.32) times greater at 22°C compared to at 18°C.

OsHV-1 DNA was first detected in gill tissue biopsies 2 hours after OsHV-1 exposure for 4 out of 12 oysters sampled at 18°C and in 5 out of 12 oysters at 22°C (Figure 6.4). By 6 hours after exposure and at all subsequent biopsy timepoints, OsHV-1 DNA was detected in all oysters at each water temperature (n = 12). The concentration of OsHV-1 DNA in the gill tissue was much lower at the time of the final biopsy, 24 hours post exposure, in oysters at

18°C (5.62x102 copies per mg of tissue; 95% CI: 4.17x102 – 7.59x102) compared to those at

22°C (1.35x105 copies per mg of tissue; 95% CI: 7.59x104 – 2.40x105) (Figure 6.4). The concentration of OsHV-1 DNA detected in biopsies at any of the timepoints did not predict mortality of individual oysters, when assessed with the Cox proportional-hazards model to account for the effect of water temperature.

The average Ct for the housekeeper gene (EF1- α) was lower than for target genes, however, there was a wide distribution in the Ct values of all 4 genes (Figure 6.5). The coefficient of variation (CV) for EF1- α Ct values across all the oysters that were sampled was 16.1% despite standardisation of tissue mass. The CV of EF1- α at 18°C was 15.7% and the CV of

140

EF1- α at 22°C was 15.4%. In additional to the individual variation, the Ct values indicated lower expression of EF1- α in oysters at 18°C compared to 22°C (Kruskal-Wallis: Χ2 = 33.01,

P < 0.001, df = 1). Because the expression of EF1- α was different at each water temperature, the relative gene expression of the immune genes from the oysters at each water temperature was considered separately. Further, EF1- α expression also varied with the biopsy time between 0 to 24 hours post OsHV-1 exposure (Kruskal-Wallis: Χ2 = 26.83, P < 0.001, df = 4).

The level of expression was at 2 hours post exposure and the highest was at 12 hours post exposure. Therefore, gene expression fold change relative to this housekeeper was not compared across time points. There was not a significant difference in EF1- α expression (Ct values) between oysters exposed to OsHV-1 or the negative homogenate (Kruskal-Wallis: Χ2

= 2.24, P = 0.135, df = 1). As EF1- α expression (Ct values) was stable across exposure, gene expression fold change was compared between oysters exposed to OsHV-1 or a negative homogenate.

141

40 * 40 * * 35 Exposure 35 Mortality

30 Negative 30 Ct Ct Surviors control 25 25 OsHV - 1 Mortalities 20 20

EF1 IAP MyD88 Viperin EF1 IAP MyD88 Viperin Gene Gene Sampling 40 40 timepoint (hours) 35 * * * Water 35 * * temperature 0

30 30 Ct Ct 18°C 2 25 25 22°C 6 20 20 12 EF1 IAP MyD88 Viperin EF1 IAP MyD88 Viperin 24 Gene Gene

Figure 6.3: Boxplot of the Ct values for each Pacific oyster (Crassostrea gigas) gene examined, including the housekeeper gene

(EF1- α) across each exposure, water temperature, mortality outcome and sampling timepoint. The * indicates where a Kruskal-Wallis test indicated a significant difference (P < 0.05) in Ct values for a gene across factors: OsHV-1 exposure; mortality outcome at the completion of the trial; water temperature; or sampling timepoint.

142

Table 6.2: Number of Pacific oysters (Crassostrea gigas) in each exposure and water temperature group biopsied at each timepoint

in the first 24 hours of infection with Ostreid herpesvirus 1 (OsHV-1) or a negative control tissue homogenate. The number of

Pacific oysters that died in the 14 days following the exposure is also shown. The hazard ratio reflecting the overall difference in

mortality at each temperature was calculated using a Cox proportional hazards model

Water N per Mortalities per biopsy time point (hours post Total Mortality% (95% temperature Exposure N total time exposure) Cox proportional hazards model mortality CI) (°C) point 0 2 6 12 24 Estimate Hazard Ratio (95% CI) 18 Control 60 12 0 0 0 0 0 0 0 (0 - 0.06) - -

OsHV-1 60 12 4 1 2 2 2 11 18.3 (10.5 - 30.2) Reference 1 22 Control 60 12 0 0 0 0 0 0 0 (0 - 0.06) - - OsHV-1 60 12 11 7 9 6 5 38 63.3 (50.5 - 74.5) 1.647 5.19 (2.61 - 10.32)

143

Figure 6.4: Kaplan-Meier survival curve showing the mortality in Pacific oysters

144

DNA was greater at 22°C compared to 18°C. The adjusted R2 for the model was 0.78 and the

F test on the full compared to the empty model was F = 137.7, df = 3, 116 (P < 0.001). The

OsHV-1 concentrations predicted by the model are shown in Figure 6.4.

Variable Estimate Std. Error P value Intercept -0.4118 0.16625 0.0147 Intercept (OsHV-1 DNA conc. at 0.11419 0.01348 < 0.001 t = 0, water temperature = 18°C) Water temperature (22°C) 0.03289 0.23511 0.889 Time * Water temperature (22°C) 0.11878 0.01907 < 0.001

145

Figure 6.5: Concentration of Ostreid herpesvirus 1 (OsHV-1) DNA per mg of Pacific oyster

146

The expression of the host immune genes Viperin, MyD88 and IAP in oysters challenged with

OsHV-1 relative to oysters challenged with a negative control inoculum between 2 hours and

24 hours post exposure was different at a water temperature of 18°C and 22°C (Figure 6.6). At

18°C relative expression of all three host immune genes was highest at 2 hours post exposure and lowest at 24 hours post exposure. This difference in fold change between sampling time points was significant in Viperin (Χ2 = 17.90, df = 3, P < 0.001), IAP (Χ2 =35.51, df = 3, P <

0.001) and MyD88 (Χ2 = 17.38, df = 3, P < 0.001). Conversely, at 22°C, expression of the

Viperin (Χ2 = 12.60, df = 3, P = 0.005) and IAP (Χ2 = 26.01, df = 3, P < 0.001) was lower at 2 hours than at 24 hours post exposure. At 22°C, relative expression of MyD88 was highest at 2 hours post exposure and lowest at 6 hours most exposure (Χ2 = 25.61, df = 3, P < 0.001).

The relative expression of all three host immune genes was lower in oysters which died following the OsHV-1 challenge than those which survived. However, this difference was significant for relative expression of IAP (Χ2 = 6.01, df = 1, P = 0.014) and not Viperin (Χ2 =

0.43, df = 1, P = 0.511) or MyD88 (Χ2 = 2.22, df = 1, P = 0.136).

147

Water temperature of 18°C Water temperature of 22°C

Viperin Survivors

MyD88 Log2 fold change 10 IAP 5

Gene 0 Viperin Mortalities -5 MyD88

IAP

2 6 12 24 2 6 12 24 Sampling time point (hours)

148

6.5 Discussion

In this study, expression pattern of the immune genes viperin, IAP and MyD88 in Pacific oyster gill tissue biopsies during the first 24 hours after OsHV-1 exposure was different when the water temperature was 18°C or 22°C. The expression of IAP in the 24 hours after OsHV-1 exposure was also associated with the risk of mortality during the next 14 days. This is the first study to report oyster immune gene expression during the acute stages of OsHV-1 infection in conjunction the mortality outcome of the same individuals. Although EF1- α has been previously validated and used as a housekeeper in Pacific oyster gene expression experiments (Du et al., 2013b; Segarra et al., 2014c), in the current study the expression was variable across water temperature and the sampling time.

The current study confirms the effect of water temperature on mortality due to OsHV-1. The total cumulative mortality was similar to previous laboratory experiments at the same water temperatures (de Kantzow et al., 2016, 2019a, 2019b; Pernet et al., 2015; Petton et al., 2013;

Renault et al., 2014). The initial detection of OsHV-1 DNA in gill tissue occurred at a similar time and the concentration was similar to previous observations (de Kantzow et al., 2019b;

Segarra et al., 2014a). OsHV-1 DNA was measured to quantify viral growth because it gave an indication of the number of viral genomes being produced and was also comparable to the previous growth curve and OsHV-1 quantification data. OsHV-1 replicated faster at 22°C compared to at 18°C in the present study, whereas in Chapter 5 the OsHV-1 replication rate was similar at 18°C and 22°C. This discrepancy may be because there were fewer timepoint measured in the present study. This meant the growth model developed in Chapter 5 was not suited to these data as did not include the plateau phase which is required for fitting the model

149 used in Chapter 5. The two different modelling approaches may have contributed to the difference in the effect of water temperature on growth rate estimated in the two studies.

The housekeeper gene, EF1- α, has previously been used in gene expression experiments looking at OsHV-1 and Pacific oyster gene expression (Du et al., 2013b; Segarra et al.,

2014c). However, in this study EF1- α was not a stable housekeeper between the two water temperatures or across the 24 hour period subsequent to the injection challenge. The exposure, either to OsHV-1 or a negative control homogenate was not associated with a difference in

EF1- α expression. This makes interpreting gene expression data more difficult, even if a gene with stable expression is found using techniques such as transcriptome sequencing there may not be a biological justification for its use as a housekeeper. This limitation notwithstanding,

EF1- α was used as a housekeeper for calculating differences in gene expression between oysters infected with OsHV-1 and control oysters.

Viperin was expressed at higher levels in Pacific oysters at a water temperature of 22°C compared to 12°C in response to poly I:C exposure (Green et al., 2014b). Whereas, in the current experiment time series data revealed that the expression in of Viperin increased over the 24 hours following exposure to OsHV-1 at 22°C, but decreased over the same period following exposure to OsHV-1 at 18°C. Contrasting immune expression patterns at 18°C and

22°C were similar across the three genes examined in the current study. Given the large differential in mortality between 18°C and 22°C (de Kantzow et al., 2016, 2019a; Pernet et al.,

2015; Petton et al., 2015a, 2013), understanding the temporal patterns in gene expression responses to OsHV-1 at each temperature during the acute stages of infection is important to understand the reasons for the difference in mortality.

150

MyD88 was also initially upregulated in Pacific oysters after exposure to OsHV-1 at 18°C, but was downregulated at 24 hours after exposure to OsHV-1 at 18°C. MyD88 did not follow the opposite pattern of down regulation to upregulation at a water temperature of 22°C, unlike

IAP and viperin. Expression of MyD88 has been previously reported as upregulated in response to OsHV-1 and poly I:C at a water temperature of 22°C (Green et al., 2015b; Green and Montagnani, 2013; Renault et al., 2011; Rosani et al., 2015). MyD88 has a proposed role as an intermediate step in multiple immune pathways in Pacific oysters (Segarra et al., 2014b,

2014c). The four genes which were differentially expressed in the 24 hours after OsHV-1 exposure and the two genes which predicted mortality were a PRR and signalling genes which detect a pathogen and then initiate and regulate an immune response (Segarra et al., 2014b,

2014c). OsHV-1 infection was suggested to supress the pathways that activate and regulate the antiviral immune response of the oyster during the early stages of infection at 22°C but not at 18°C (Green et al., 2015b; Rosani et al., 2015; Segarra et al., 2014a). This suggests that suppression of immune response at a water temperature of 22°C that allows the virus to replicate and cause mortality at this temperature rather than an overactive immune response triggered by OsHV-1, as previously suggested (Green et al., 2014b).

IAP and viperin are expressed in response to detecting a pathogen in order to overcome the infection and neither were differentially expressed in response to exposure to OsHV-1 compared to a negative control homogenate. Only changes in expression of IAP was associated with a difference in mortality in oysters which were exposed to OsHV-1 in this study. However, up regulation of Viperin has previously been observed in response to both

OsHV-1 and poly I:C exposure (Green et al., 2015c, 2015b; Green and Montagnani, 2013;

Renault et al., 2011; Rosani et al., 2015). IAP has been previously reported as upregulated in

151 response to OsHV-1 and poly I:C (Green et al., 2015b; Green and Montagnani, 2013; Renault et al., 2011; Rosani et al., 2015). However, viral homologs of Pacific oyster IAP expressed by

OsHV-1 may affect endogenous IAP expression (Green et al., 2015b), and IAP may be differentially expressed later than 24 hours after exposure (Segarra et al., 2014c, 2014a). IAP and MyD88 have also been upregulated in response to OsHV-1 exposure in oysters which were genetically selected for resistance to OsHV-1 mortality (Segarra et al., 2014c). The difference in mortality between the resistant and susceptible lines was 88% and represented the effect of genetic resistance to OsHV-1 (Segarra et al., 2014c), whereas the difference in mortality between oysters at different temperatures in the present study was independent of genetic resistance to OsHV-1. This indicates that while the expression of the selected immune receptor and signalling genes is dependent on OsHV-1 exposure and can affect the level of mortality, Pacific oyster immune genes which are key to overcoming OsHV-1 infection are yet to be identified. Key elements of the Pacific oyster immune response that provide resistance to OsHV-1 may not have homologues in vertebrate animals, making them difficult to identify by studying vertebrate model species (Allam and Raftos, 2015; Green et al.,

2015a).

The present study used a drilling technique to provide an opening in the shell which provided access for injection into the adductor muscle and gill biopsy. The advantage was collection of samples ante-mortem during the acute stages of infection while retaining the ability to observe the still know the outcome for the individual. This procedure may have resulted in less disturbance to the disease process compared to relaxing the oysters to obtain the biopsy. The use of this technique allowed the present study to demonstrate that the altered expression of putative Pacific oyster immune genes was associated with mortality in the same individual.

152

The present study focused on a small number of genes previously associated with an immune response to OsHV-1 (Green et al., 2015b, 2014b; Rosani et al., 2015; Segarra et al., 2014b).

However, this approach could be shifted to make use of unbiased approaches such as RNA sequencing to assess the full oyster and OsHV-1 transcriptome (Rosani et al., 2019). The current study was limited by only having one biopsy sample from each oyster in the 24 hours after the OsHV-1 exposure, this was done to avoid excessive tissue damage or mortality associated with taking multiple gill biopsy samples. The use of multiple longitudinal biopsy samples from the same oyster, all linked to survival data may produce a more powerful dataset for understanding immune response to OsHV-1. At present, studies have considered gene expression at different water temperatures (Green et al., 2014b) or on OsHV-1 exposure on gene expression at a single temperature (Green et al., 2015c; Green and Montagnani, 2013;

Segarra et al., 2014b, 2014c). The different association between gene expression and OsHV-1 mortality in the current study and previous gene expression studies (Green et al., 2014b) may be due to a difference in the age of the oysters as this affects mortality due to OsHV-1 (de

Kantzow et al., 2016; Paul-Pont et al., 2015; Whittington et al., 2018). Examination of the changes in the proteome of gill tissue during the early stages of OsHV-1 infection, as previously done on haemolymph following poly I:C exposure (Green et al., 2016a), may provide a more direct insight into the changes in cellular function during pathogenesis.

However, the production of mRNA for each gene may not directly correlate with a change or increase in the level of function of the transcribed protein.

At a water temperature of 18°C, the expression of the putative Pacific oyster immune genes

MyD88, IAP and Viperin was initially upregulated in oysters exposed to OsHV-1 compared to oysters not exposed to an OsHV-1 free negative control homogenate. Whereas, at 22°C, two

153 of the same genes were initially down regulated and then expression increased over the 24 hours following exposure to OsHV-1. Mortality due to OsHV-1 was significantly higher at a water temperature of 22°C compared to 18°C. When including oysters at both temperatures, down regulation of the putative Pacific oyster immune gene IAP was associated with oysters that died in 14 days following OsHV-1 exposure. This suggests that the immune responses to

OsHV-1 depends on the water temperatures and this may explain, in part, the influence of water temperature on mortality with OsHV-1 infection. The biopsy technique introduced in the present study is suited to further studies considering a broad wide range of immune genes is required to further investigate the mechanisms that drive mortality caused by OsHV-1 infection.

154

7. Chapter 7 - Immune conditioning of Pacific oysters (Crassostrea gigas):

comparison of infection and immune stimulation treatments for protection

against Ostreid herpesvirus 1 (OsHV-1)

7.1 Abstract

Microvariant genotypes of Ostreid herpesvirus 1 (OsHV-1) are the cause of a potentially high mortality disease of Pacific oysters (Crassostrea gigas) in aquaculture. Oysters that survive natural exposure to OsHV-1 in the field have higher survival during subsequent disease outbreaks. This has been replicated under laboratory conditions, indicating that previous exposure specifically provides a protective effect against OsHV-1. The aim of the present study was to compare the effect of prior to infection with OsHV-1 with prior exposure to heat inactivated OsHV-1 or poly I:C which has previously been shown to elicit a protective response on mortality following exposure to OsHV-1. The treatments were administered by injection and the oysters were maintained in laboratory conditions at a water temperature of

18°C for 14 days before the temperature was increased to 22°C. The total cumulative mortality (TCM) in the 21 days after immune conditioning with OsHV-1 was 12.5% (95% CI:

8.21 – 18.58), with no mortality in the other groups. Three weeks after the treatments, the oysters were then challenged with OsHV-1 and mortality was monitored for 14 days. The

TCM in oysters treated with artificial sea water (ASW) and OsHV-1 free homogenate was

53% and 55% respectively. Compared to oysters conditioned with ASW, significant protection against mortality was provided by heat inactivated OsHV-1 (HR: 0.41, 95% CI:

0.25 - 0.67) and infectious OsHV-1 (HR: 0.18, 95% CI: 0.11 - 0.32) and poly I:C (HR: 0.02,

95% CI: 0.00 - 0.11). Conditioning for increased survival by prior exposure to inactivated or infectious OsHV-1 in environmental conditions that favour survival provides an alternative to

155 non-specific immune priming agents such as poly I:C for protection against disease caused by

OsHV-1. Further understanding of the immune modulatory mechanisms and conditions under which this can be elicited is recommended to develop practical disease control applications.

7.2 Introduction

Pacific oysters (Crassostrea gigas) are affected by seasonal outbreaks of disease caused by microvariant genotypes of Ostreid herpesvirus 1 (OsHV-1) in Europe (EFSA, 2010), Australia

(Jenkins et al., 2013) and New Zealand (Keeling et al., 2014). Reducing the impact of OsHV-

1 on Pacific oyster aquaculture requires an integrated approach to managing host and environmental factors to favour a balance with the pathogen which is conducive to health and productivity (Pernet et al., 2016). Currently the focus of disease control includes genetic selection for oysters with disease resistance (Dégremont et al., 2015b) and changing growing conditions to favourably modify the environment (Paul-Pont et al., 2013b; Pernet et al., 2012;

Whittington et al., 2015a, 2015b). Immune priming with poly I:C has also been explored as a method of reducing mortality with a subsequent OsHV-1 challenge (Green et al., 2016b;

Green and Montagnani, 2013; Lafont et al., 2017). Manipulating the interaction between the host and pathogen to give the host an advantage, for example by raising the growing height, is an effective technique for maintaining healthy and productive livestock (Whittington et al.,

2015a). Vaccines are frequently used in finfish aquaculture to protect the host from infection by developing specific antibodies against a pathogen (Dadar et al., 2017; Dalmo, 2018). This approach is not available for oysters as they do not produce antibodies which are a key part of the adaptive immune mechanism that produces a highly specific and effective immune response following vaccination of vertebrates (Wang et al., 2018). There is evidence that invertebrates may have a form of immune memory arising from prior exposure which reduces

156 mortality on subsequent exposures (Kurtz and Franz, 2003b; Milutinović and Kurtz, 2016).

Administration of inactivated White spot syndrome virus (WSSV) with the immunostimulant glucan or inactivated Vibrio penaeicida achieved a reduction in mortality Penaeus japonicus following a WSSV challenge (Namikoshi et al., 2004). The suggested mechanism that allowed increased survival of the shrimp was increased phagocytic activity initiated by pattern recognition receptor (PRR) proteins which bind to pathogen associated molecular patterns

(PAMPs) on the surface of the pathogen (Rowley and Pope, 2012).

Immune priming of invertebrates has been defined as an exposure to an immune stimulant or a pathogen that results in greater survival on an encounter with the same pathogen by virtue of a change induced by the first exposure (Kurtz, 2004; Milutinović and Kurtz, 2016; Schulenburg et al., 2007). A specific immune response in invertebrates was first observed in the copepod

Macrocyclops albidus with an initial exposure to a natural parasite, the tapeworm

Schistocephalus solidus (Kurtz and Franz, 2003b). This resulted in a lower level reinfection and infection intensity with sibling tapeworm larvae 3 days later compared to unrelated tapeworm larvae (Kurtz and Franz, 2003b). This example suggested the presence of immune memory in invertebrates. Immune memory is the ability to store and reuse information about a previously encountered pathogen, such as in the form of long lived RNA-induced silencing complexes (RISCs) incorporating viral gene specific miRNA which selectively silence viral gene expression (Milutinović and Kurtz, 2016; Olejniczak et al., 2013). Administration of dsRNA of the VP26 or VP28 WSSV structural protein genes amplified by PCR or recombinant VP26 or VP28 protein protected against a lethal WSSV challenge in Litopenaeus vannamei and Penaeus japonicus, respectively (Escobedo-Bonilla et al., 2015; Namikoshi et al., 2004). Immune conditioning is reinforcing a response through exposure. Exposure to

157

OsHV-1 at 18°C reduced mortality following a second OsHV-1 exposure compared to oysters with no history of OsHV-1 exposure (de Kantzow et al., 2019a). The immune response to

OsHV-1 was reinforced through exposure to OsHV-1, as measured by lower mortality in conditioned oysters. Injecting Pacific oysters with polyinosinic:polycytidylic acid (poly I:C), a synthetic double stranded homopolymer used to mimic dsRNA in cells, primed oysters and gave protection against OsHV-1 (Green et al., 2016a; Lafont et al., 2017; Pauletto et al.,

2017). However, these experiments did not show if the protective effect against mortality is specific to infection with OsHV-1.

Poly I:C is a potent immune stimulant which binds to conserved PAMP receptors such as toll- like receptors, as it is an analogue of viral dsRNA. Administering poly I:C to oysters prevented mortality from an OsHV-1 challenge up to 18 weeks later (Green et al., 2016b;

Green and Montagnani, 2013; Lafont et al., 2017; Pauletto et al., 2017). A transgenerational effect has also been reported: at 96 hours after exposure to OsHV-1, the mortality in larvae of brood stock treated with poly I:C was halved compared to larvae of brood stock treated with seawater (Green et al., 2016b). Administration of poly I:C changes gene expression patterns in immune pathways presumed to act as antiviral defences (Green and Montagnani, 2013; Green and Speck, 2018). Because poly I:C is not comprised of the bases adenine, cytosine, guanine or uracil, it has nothing uniquely in common with OsHV-1 over other viruses. Administration of two consecutive doses of poly I:C did not induce changes in transcription of immune associated genes on the second administration compared to oysters which did not receive a second dose of poly I:C (Green et al., 2014a). Poly I:C did not provide protection against infection with the pathogen Vibrio tasmaniensis, suggesting that the protective effect of the immune stimulation may be limited to viruses (Green and Montagnani, 2013; Lafont et al.,

158

2017). This is consistent with poly I:C stimulating TLR3 which recognizes dsRNA which is a signal for viral replication.

Invoking a specific response which allows the immune system to differentiate between pathogens may have a lower physiological cost than a general response, but would require using an immune stimulant with specific characteristics of the pathogen to produce an immune memory (Kurtz, 2004). Field observations suggest that oysters which have previously been exposed to OsHV-1 during an outbreak are likely to survive subsequent outbreaks (Evans et al., 2017; Whittington et al., 2018). A proportion of the increased survival is due to increasing age and size which reduces the susceptibility of the oysters to OsHV-1 between the outbreaks

(Hick et al., 2018). The role of prior exposure to OsHV-1 in reducing mortality was demonstrated under laboratory conditions (de Kantzow et al., 2019a). Challenge with OsHV-1 with the water temperature at 18°C resulted in infection without high mortality and this exposure was protective against a subsequent OsHV-1 challenge at 22°C, a water temperature at which high mortality occurred in control oysters without the prior exposure (de Kantzow et al., 2019a).

The aim of this study was to determine if active infection with OsHV-1 was required to provide a protective effect against mortality caused by subsequent exposure to OsHV-1 or if a similar response could be induced by exposure to inactivated OsHV-1. Conditioning oysters with heat inactivated OsHV-1 may allow a response to the capsid, viral nucleic acid and other structural components without viral replication, reducing the risk of mortality and the risk of translocating virus in treated oysters. Controlled exposure to OsHV-1 using a specific water temperature regime that had previously enabled protection by initiating a low mortality infection was used as a control to lead to protection after exposure to OsHV-1 (de Kantzow et

159 al., 2019a). Poly I:C was also included to assess the relative level of protection it provided against mortality (Green et al., 2016b; Green and Montagnani, 2013; Lafont et al., 2017).

7.3 Method

7.3.1 Experimental design

Different immune conditioning treatments were tested for efficacy in protecting C. gigas from mortality when they were challenged with OsHV-1 infection 3 weeks later. The immune conditioning treatments were: infectious OsHV-1 (3x106 genome copies per oyster); heat inactivated OsHV-1 (3x106 genome copies per oyster); and poly I:C (250 µg/oyster). The control treatments were an OsHV-1 free oyster tissue filtrate and sterile artificial sea water

(ASW). Each treatment was administered by injection and the oysters were maintained in laboratory conditions with the water temperature at 18°C for 14 days before the temperature was increased to 22°C over a period of 6 hours (Figure 7.1). Three weeks after the immune conditioning, 3 out of the 4 replicate tanks for each treatment group were randomly selected and the oysters in these were challenged with OsHV-1. Mortality was recorded for 14 days.

Oysters in the remaining replicate tank were negative controls that were challenged with an

OsHV-1 free oyster tissue filtrate and monitored for 14 days.

160

Tank water Acclimation Immune conditioning OsHV-1 Challenge End point

temperature

22°C

18 °C Sterile ASW (80 / 4) Poly I:C (80 / 4) OsHV-1 (160 / 8) OsHV-1, heat inactivated (80 / 4) Oyster tissue homogenate (80 / 4) -7 days 0 days 14 Days 21 Days 35 Days

Figure 7.1: Experiment timeline showing the water temperature profile with the acclimation period and the timing of the immune stimulation and Ostreid herpesvirus 1 (OsHV-1) challenges. Each treatment group is shown with a separate line. The water temperature was increased from 18°C to 22°C on day 14 in each group. The initial number of Pacific oysters (Crassostrea gigas) and the number of replicate tanks is shown next to each treatment group. Samples were taken to measure the OsHV-1 prevalence and viral load immediately prior to the OsHV-1 challenge and at the end of the trial all surviving oysters were sampled.

161

7.3.2 Oysters

Oysters for the trial were as described in section 2.2.2.

7.3.3 Aquarium management

Oysters were housed at a density of 20 oysters per tank in aquaria managed as described in sections 2.2.3 – 7.

7.3.4 Immune conditioning and priming

Each tank was randomly allocated to one of the five treatments using the RAND function in

Microsoft Office Excel 365 (Microsoft Corporation, Seattle). The oysters were removed from the water 12 hours prior to treatment and placed in a solution of 50 gL-1 magnesium chloride

(MgCl2, Sigma) in fresh aquarium water for 4 hours to relax the adductor muscle. All the oysters in a tank were injected with a 50 µL volume of the allotted priming or conditioning agent into the adductor muscle using a 1 mL syringe and a 25G needle.

The ASW used as a negative control and for dilution of reagents was prepared from sea salt

(Red Sea Salt) at 30 ppt in MilliQ water and was filtered to 0.22 µm (Micropore). The OsHV-

1 free oyster tissue filtrate was prepared from apparently healthy oysters of the same batch according to the method for tissue homogenisation described in Section 2.6. To achieve a final dilution of 1 in 1000 w/v, 200 µL of the supernatant was added to 19.8 mL of ASW and filtered with a 0.22 µm syringe filter. The infectious OsHV-1 inoculum was prepared from a cryopreserved oyster tissue homogenate which was diluted to 1 in 1000 w/v final dilution with

ASW. The cryopreserved virus stock was produced from naturally infected oysters sampled from the Georges River in 2011 (Paul-Pont et al., 2013b) and prepared as described by Hick et al. (2016). Briefly, the gill and mantle of 15 oysters were excised and pooled, homogenized by

162 stomaching and diluted to 10% w/v in sterile ASW. The diluted homogenate was filtered to

0.22 µm and frozen at -80°C with 10% v/v foetal bovine serum (Sigma) and 10% v/v glycerol

(Sigma). The heat inactivated OsHV-1 was prepared by incubating a 10 mL aliquot of the

OsHV-1 inoculum at 50°C in a hybridization oven for 20 minutes. The poly I:C inoculum was prepared by diluting lyophilised poly I:C (Sigma) to 5 mgmL-1 in ultrapure water (Life

Technologies). All treatments were prepared immediately prior to use.

7.3.5 OsHV-1 challenge

The oysters were challenged with a cryopreserved inoculum prepared as per section 2.3.1 and administered as per sections 2.3.3 – 4.

7.3.6 Quantification of OsHV-1 DNA

Gill and mantle samples were taken according to section 2.4.1 and processed as described in sections 2.4.3 – 4. OsHV-1 was quantified according to sections 2.4.7 – 9.

7.3.7 Statistical analysis

Statistical analysis was performed in Microsoft R Open version 3.4.5 (R Core Team, 2017) using the tibble (Müller and Wickham, 2017), readr (Wickham et al., 2017b), tidyr (Wickham and Henry, 2017), plyr (Wickham, 2011), forcats (Wickham, 2017) , survival (Therneau,

2015; Therneau and Grambsch, 2000), ggplot2 (Wickham, 2009) and survminer (Kassambara and Kosinski, 2017) packages.

The sample size was calculated for an expected hazard ratio between treatment and control groups of 4 with 80% power and α = 0.05 using the powerSurvEpi package in R 3.4.5

(Weiliang et al., 2018). The calculation required an assumption that mortality would be 10%

163 in oysters pre-exposed to OsHV-1 at 18°C, 50% in oysters without immune conditioning (pre- exposed to ASW) and 20% in oysters conditioned with the negative inoculum. The sample size for each group was then rounded up to a multiple of 20 to allow for 20 oysters per replicate tank.

Summary tables of the total cumulative mortality and Kaplan-Meier survival curves were produced (Kaplan and Meier, 1958). The mortality data were analysed using a Cox proportional hazards model (Breslow, 1992; Cox, 1972; Therneau and Grambsch, 2000). The failure condition was mortality within the follow up period for the challenge (14 days) and a positive OsHV-1 PCR result at the time of mortality. The model included treatment group as the only independent variable and tank identification number was included as a random effect to account for clustering within replicate tanks. The Wald test statistic was used to assess the significance of each treatment and the Schoenfeld residuals were plotted to assess the assumption of proportional hazards.

OsHV-1 DNA concentration was log10 transformed for graphical presentation and to satisfy the assumption of normality for further analysis. Boxplots were created from the transformed data for each challenge using ggplot2. Confidence intervals for the mean OsHV-1 concentration from each group were calculated with 95% intervals and back transformed for presentation. The OsHV-1 DNA concentration in the survivors at the end of the trial and at the time of death in the oysters that died were analysed separately. Both data sets were analysed using a linear mixed model with the glm function in R with log10 transformed OsHV-1 concentration as the outcome and treatment group as the explanatory variable. Tank was included as a random effect to account for clustering within replicate tanks. Confidence intervals for the model parameters were calculated at the 95% level.

164

The prevalence of OsHV-1 3 weeks after the immune conditioning treatments was determined using a random sample of 20 surviving oysters. The prevalence was calculated as the number of oysters that tested positive for OsHV-1 DNA divided by the total number of oysters sampled from that treatment group. The 95% confidence intervals were calculated using the binom.logit function from the binom package (Dorai-Raj, 2014). The difference in OsHV-1 prevalence in the survivors of the OsHV-1 challenge between groups was determined using a logistic regression model with the glm function. The parameter estimate for each treatment group was exponentiated to calculate the odds ratio for the detection of OsHV-1 between the reference group (Sterile ASW) and each treatment group. The odds ratio for the reference groups was set at 1. The outcome for the model was the detection of OsHV-1 in the surviving oysters from each treatment group and the sole explanatory variable was the treatment group.

The tank number of each oyster was included as a random effect.

7.4 Results

7.4.1 Immune conditioning

In the 21 days after the immune conditioning there was 12.5% (95% CI: 8.21 – 18.58) mortality in oysters injected with OsHV-1 at 18°C (Table 7.1). The average OsHV-1 DNA concentration at the time of death was 4.58 x103 (95% CI: 1.83x103 – 1.14x104) copies.mg-1.

There was no mortality in the first 21 days following the immune conditioning treatment in any of the other treatment groups.

7.4.2 OsHV-1 challenge

High mortality occurred in the control groups: 53% in those treated with ASW and 55% in those treated with OsHV-1 free oyster tissue homogenate (Figure 7.2, Table 7.1). The total

165 cumulative mortality, after both treatment and challenge, was 26.7% for oysters that had been conditioned with heat inactivated OsHV-1 and 12.8% for the oysters that had received infectious OsHV-1 at 18°C (Table 7.1). This corresponded to significant protection against mortality with hazard ratios of 0.41 (95% CI: 0.25 – 0.67) for heat inactivated OsHV-1 and

0.18 (95% CI: 0.11 – 0.32) for virulent OsHV-1 compared to the ASW control (Figure 7.2,

Table 7.2). There was no mortality after the OsHV-1 challenge at 22°C in the group conditioned with poly I:C indicating a strong protective effect (HR: 0.02, 95% CI: 0.00 –

0.11) compared to the ASW control. The OsHV-1 free oyster tissue homogenate did not have any immune conditioning properties compared to injection with ASW (Hazard ratio, HR:

1.03, 95% CI: 0.66 – 1.59) (Table 7.2). There was no mortality in oysters injected with the negative control for any treatment groups.

166

Table 7.1: Total cumulative mortality of Pacific oysters (Crassostrea gigas) 14 days after challenge with Ostreid herpesvirus 1

(OsHV-1) or a negative control. The challenge was administered 3 weeks after an immune conditioning treatment. The prevalence of OsHV-1 detected by qPCR in survivors 10 days after the OsHV-1 challenge is also shown; there was significantly lower prevalence in the oysters treated with poly I:C compared to oysters treated with ASW (P < 0.05).

Challenge Treatment group N Mortality % (95% CI) Prevalence % (95% CI) Negative Heat inactivated OsHV-1 20 0 (0 - 16.8) - control Negative homogenate 20 0 (0 - 16.8) - OsHV-1 homogenate 30 0 (0 - 11.6) - Poly I:C 20 0 (0 - 16.8) - Sterile ASW 20 0 (0 - 16.8) - OsHV-1 Heat inactivated OsHV-1 60 26.7 (17.0 - 39.2) 55.6 (33.0 – 76.0) Negative homogenate 60 55.0 (42.4 - 67.0) 55.0 (33.6 – 74.7) OsHV-1 homogenate 94 12.8 (7.4 - 21.2) 55.6 (33.0 – 76.0) Poly I:C 60 0.0 (0 – 6.0) 0.0 (0.0 – 16.8) Sterile ASW 60 53.3 (40.8 - 65.5) 70.0 (47.3 – 85.9)

167

Figure 7.2: Kaplan-Meier survival curve for an Ostreid herpesvirus 1 (OsHV-1) challenge in

Pacific oysters (Crassostrea gigas), 3 weeks after administration of immune conditioning treatments, including controls for immune conditioning.

168

Treatment group Coefficient Hazard Ratio (95% CI) Sterile ASW - 1 Heat inactivated OsHV-1 -0.895 0.41 (0.25 - 0.67) Negative homogenate 0.025 1.03 (0.66 - 1.59) Poly I:C -3.802 0.02 (0.00 - 0.11) OsHV-1 -1.699 0.18 (0.11 - 0.32) *95% Confidence intervals which do not include 1 are statistically significant (P < 0.05).

The concentration of OsHV-1 DNA at the time of death for oysters that died following the challenge with OsHV-1 at 22°C was lowest in the group that was conditioned with infectious

OsHV-1 at 18°C (1.30 x104; 95% CI: 5.28x103 – 3.22x104 copies.mg-1) (Figure 7.3), but this was not significantly different from the other groups (P > 0.05) (Table 7.3). Amongst the survivors at the end of the trial, the only group which had a significantly lower concentration of OsHV-1 DNA compared to oysters conditioned with sterile ASW were the oysters conditioned with poly I:C (P ≤ 0.001) (Table 7.3).

At the conclusion of the trial, the prevalence of OsHV-1 in the survivors of the OsHV-1 challenge at 22°C was highest in the group conditioned with sterile ASW at 70% (95% CI:

47.28 – 85.86) and lowest in the group conditioned with poly I:C at 0% (95% CI: 0.00 –

16.84) (Table 7.1). The odds of an oyster in the group conditioned with poly I:C which survived the challenge with OsHV-1 at 22°C testing positive for OsHV-1 at the conclusion of the trial was 0.02 (95% CI: 0 - 0.15) that of the control group given ASW (Table 7.4).

169

6 A

t g

m 5

o n

e 4

V H

s 3

o L 2 Heat inactivated OsHV-1Negative homogenate OsHV-1 homogenate Sterile ASW

6 B

i t

g 5

p s

e 4

e g

1 3

V H

s 2

o L

Heat inactivated OsHV-1Negative homogenate OsHV-1 homogenate Sterile ASW

Figure 7.3: Ostreid herpesvirus 1 (OsHV-1) DNA concentration in Pacific oysters

(Crassostrea gigas) when challenged 3 weeks after receiving immune conditioning or control treatments (a) at the time of death; (b) in surviving oysters at 14 days post challenge (Day 35).

The oysters conditioned with poly I:C are not displayed here as they all survived to the end of the trial period and OsHV-1 DNA was not detected in any oyster in that group. The box is

170 bounded with the 1st and 3rd quartile of the data and the black line indicates the mean. The whiskers extend to the maximum and minimum of the data except where data greater than 1.5 times the interquartile range from the mean are indicated by a point.

Treatment group (Parameter) Estimate Std. Error P value Odds ratio (95% CI) Sterile ASW (Reference) 0.8473 0.488 0.082 1 Heat inactivated OsHV-1 -0.6242 0.6805 0.359 0.54 (0.14 - 2.02) Negative homogenate -0.6466 0.6634 0.33 0.52 (0.14 - 1.9) OsHV-1 homogenate -0.6242 0.6805 0.359 0.54 (0.14 - 2.02) Poly I:C -3.7917 1.1361 < 0.001 0.02 (0 - 0.15)

171

Table 7.4: The concentration of Ostreid herpesvirus 1 (OsHV-1) DNA in each Pacific oyster (Crassostrea gigas) either at the time of death or at 10 days post challenge in Pacific oysters that survived a challenge with OsHV-1 by injection 3 weeks after the immune conditioning treatments. Data were analysed using two different generalized linear mixed models in R 3.3.5, one for the survivors and one for the mortalities. The outcome of the model was log10 transformed OsHV-1 concentration and the immune conditioning treatments were predictors with tank included as a random effect. The estimate for the intercept variable for each model is the mean log10 transformed viral concentration for the reference group (conditioned with sterile ASW) and the estimate for the other groups are the difference between the mean of that group and the reference group.

OsHV-1 challenge Treatment group Estimated OsHV-1 Estimate Std. Error P value 95% CI outcome (Parameter) genomes.mg-1 Survivors Sterile ASW (Intercept) 1.253 0.253 < 0.001 0.76 - 1.75 17.91 Heat inactivated OsHV-1 0.133 0.379 0.726 -0.61 - 0.88 24.32 Negative homogenate -0.048 0.357 0.892 -0.75 - 0.65 16.03 OsHV-1 homogenate -0.113 0.367 0.759 -0.83 - 0.61 13.8 Poly I:C -1.253 0.357 < 0.001 -1.95 - -0.55 1

OsHV-1 challenge Treatment group Estimated OsHV-1 Estimate Std. Error P value 95% CI outcome (Parameter) genomes.mg-1 Mortality Sterile ASW (Intercept) 4.651 0.162 < 0.001 4.33 - 4.97 4.48x104 Heat inactivated OsHV-1 -0.19 0.277 0.495 -0.73 - 0.35 2.89x104 Negative homogenate -0.102 0.221 0.645 -0.54 - 0.33 3.54x104 OsHV-1 homogenate -0.536 0.301 0.079 -1.12 - 0.05 1.30x104

172

7.5 Discussion

Conditioning Pacific oysters with heat inactivated OsHV-1 produced a strong protective effect against subsequent mortality due to OsHV-1, suggesting that conditioning does not require viral replication. The current study confirmed the observation that exposure to OsHV-1 at a temperature that is not permissive for mortality reduces mortality during a second challenge with OsHV-1 at a permissive temperature. A protective effect was also observed with the administration of heat inactivated OsHV-1 at 18°C and poly I:C. Protection against mortality due to OsHV-1 has been previously observed following immune conditioning with an OsHV-

1 exposure (de Kantzow et al., 2019a) The oysters administered sterile ASW at the time of conditioning had similar mortality when challenged with OsHV-1 at 22°C to that described in previous laboratory experiments at this water temperature using oysters that had not been conditioned for protection in any way (de Kantzow et al., 2016; Petton et al., 2015a, 2013;

Renault et al., 2014).

Poly I:C was used as a control in this experiment to compare the relative protective effect of this non-specific immune stimulant (Green and Montagnani, 2013; Lafont et al., 2017) with that provided by OsHV-1 (de Kantzow et al., 2019a). Poly I:C is recognized by highly conserved intracellular Toll-like receptors as a pathogen associated molecular pattern (PAMP) and may lead to non-specific antiviral immune conditioning pathway (Green et al., 2015a). It is assumed that the mode of action is through detection by the TLR 3 receptor which binds to dsRNA and induces an antiviral immune response through the JAK/STAT pathway (Green and Montagnani, 2013). The level of protection against mortality due to OsHV-1 afforded by immune conditioning with poly I:C is dependent on the dose with 19 µg per gram resulting in

173

100% survival and 1.9 µg per gram resulting in 50% survival (Lafont et al., 2017). Previously protection has been achieved with doses of 250 µg per oyster (Green and Montagnani, 2013),

500 µg per oyster (Green et al., 2016b). In the current study, the dose of poly I:C for each oyster was 250 µg. The differences in protection afforded by injection of poly I:C, infectious

OsHV-1 and heat inactivated OsHV-1 that were observed in the current study may be due, in part, to the different quantities of nucleic acid given or produced by each treatment. The amount of viral dsDNA in the heat inactivated preparation was likely to be less than 1 ng per oyster, based on the OsHV-1 genome length and a dose of 106 OsHV-1 genomes. Although most viruses produce dsRNA it is a by-product in dsDNA viruses and only present in a very low concentration during replication, if at all (Jacobs and Langland, 1996). A low level or absence of viral dsRNA in oysters conditioned with infective OsHV-1 compared to oysters conditioned with poly I:C may have led to a lower level of protection against mortality through less excitation of the TLR3 receptor pathways suggested to induce resistance (Green and Montagnani, 2013). Expression of the receptor TLR 3 was not differentially expressed between oysters which had and had not been genetically selected for OsHV-1 (Segarra et al.,

2014c). Further research is required to identify other PRRs which detect PAMPs associated with OsHV-1. The relative protective potential of dsRNA, viral nucleic acid and virion components needs further investigation to understand the overall level of immune conditioning or priming resulting from each treatment. The water temperature may have also altered the response to immune stimulation (Green et al., 2014b).

The mechanisms which underly priming or conditioning in oysters require further investigation to understand the advantages and limitations of the techniques. If OsHV-1 is similar to other large dsDNA viruses, infection with OsHV-1 could produce dsRNA during

174 replication and so may stimulate similar pathways to poly I:C (Jacobs and Langland, 1996;

Weber et al., 2006). However, OsHV-1 also produces mRNA which is targeted by RNAi and has capsid and envelope proteins which are also PAMPs which may stimulate different receptors to dsRNA (Grundhoff and Sullivan, 2011; Pauletto et al., 2017). Heating OsHV-1 at

50°C for 5 mins was an effective method of disinfection when infectivity was evaluated with a bioassay (Hick et al., 2016). Viral proteins remain recognizable by host immune receptors after heating for 30 minutes at 56°C, allowing the heat inactivated Human alphaherpesvirus 1

(HSV) to adsorb to host cells and initiate a response (Lancz, 1980). The protective effect from heat inactivated OsHV-1 indicated a relatively small dose of viral dsDNA and/or capsid and envelope proteins from the virus provided sufficient immune stimulation to result in a significant protective effect when the exposure was by intramuscular injection. Toll like receptor 3 (TLR3), which recognizes dsRNA, is up regulated in oysters injected with poly I:C but not in oysters infected with OsHV-1, suggesting there is less viral dsRNA present than poly I:C or that different receptors or immune pathways are involved (Green and Montagnani,

2013; Pauletto et al., 2017; Segarra et al., 2014c, 2014a). Changes in expression of TLR3 may not be directly linked to the amount of dsRNA the receptor is binding to as TLR3 may have additional roles within the cell. TLR3 expression affects glucose homeostasis, suggesting that the role of TLR3 is more complex than PAMP recognition alone (Strodthoff et al., 2015).

Further investigation of the immune pathways activated in oysters by exogenous nucleic acids and other PAMPs are required to differentiate the mechanisms of the protective effect afforded by the different treatments.

To produce a protective response upon encountering a pathogen for the second time, molluscs require mechanisms for immune memory (Milutinović and Kurtz, 2016). Viral replication

175 may be required for some proposed mechanisms to function, such as for RNAi which requires pathogen mRNA to be cleaved and used as a guide RNA as part of an RNA induced silencing complex (RISC), which allows targeted cleavage of viral mRNA (Grundhoff and Sullivan,

2011; Kloosterman and Plasterk, 2006; Milutinović and Kurtz, 2016; Olejniczak et al., 2013).

The increased protection afforded by priming with live OsHV-1 compared to heat inactivated

OsHV-1 suggests that a replicating infection may be important for some aspects of immune memory. Poly I:C treatment influences immune gene expression and may prevent apoptosis or an immunopathological response killing the oyster through inflammation and tissue damage

(Green et al., 2016b; Green and Montagnani, 2013; Lafont et al., 2017; Pauletto et al., 2017).

Poly I:C may also have a direct effect on OsHV-1 such as altering OsHV-1 gene transcription.

The increased physiological demand due to increased antiviral immune activity may impact on the oyster’s growth rate and condition. This would be an important consideration in understanding immune regulation and on the potential for using immune priming with poly

I:C in oyster aquaculture. Oysters which are in good condition and use a larger proportion of their energy reserves during the period of exposure to OsHV-1 have greater survival, possibly due to the ability to mount a more effective immune response than oysters in a poor condition.

Targeted investigation of immune responses identified in transcriptional studies are needed to understand the differences in immune response and potential mechanisms for immune memory. A further limitation is that OsHV-1 is the only virus of Pacific oysters currently available for use in an infection model, which makes it difficult to assess the specificity of immune conditioning or priming.

The application of conditioning or priming of Pacific oysters as a disease mitigation measure in aquaculture has several practical limitations. The conditioning method presented here

176 requires injecting each oyster individually, a problem given the level of commercial production at 11, 345 tons in 2015-16 (ABARES, 2018). Therefore, alternative delivery mechanisms to injection such as immersion are needed and should be investigated.

Transgenerational immune priming with Poly I:C has been suggested but has only been demonstrated in larvae not in spat or adult oysters which would be required for the technique to be effective for use in aquaculture (Green et al., 2016b). The use of poly I:C in commercial oysters used for food is unlikely to be acceptable for food safety reasons. The toxic effects observed in mammals included fever, hypotension, depressed total blood count and anaemia, weight loss, focal haemorrhages and congestion (Hartmann et al., 1987, 1986; Homan et al.,

1972; Krown et al., 1985). Mortality followed a single 30 mg/kg dose of poly I:C in rhesus monkeys and 5 mg/kg in beagles (Hartmann et al., 1987; Homan et al., 1972). The use of poly

I:C would require research on poly I:C residues in oyster tissues, and on potential withholding periods as these are required to inform the use of chemicals in food producing animals

(Agricultural and Veterinary Chemicals Code Act 1994). For this reason, conditioning with live or heat inactivated OsHV-1 is potentially more acceptable to regulatory authorities than priming with poly I:C. The use of heat inactivated OsHV-1 rather than live OsHV-1 has the advantage of reducing the biosecurity risk and the risk of initial mortality due to viable OsHV-

1, even at low water temperature (de Kantzow et al., 2019a). The use of OsHV-1 for an immune conditioning treatment may introduce a biosecurity risk due to the potential to move infected oysters into an OsHV-1 free area and in interpreting the results of disease surveys as oysters have remained OsHV-1 positive 9 weeks after a laboratory OsHV-1 challenge (de

Kantzow et al., 2019a). The persistence of detectable levels of OsHV-1 in exposed oysters would make mapping the spread of an outbreak more difficult. Further evaluation of treatment

177 with inactivated OsHV-1 under field conditions is required to assess the duration of the effect, the ages of the oysters when it can be applied and the consistency of the protective effect between batches of oysters.

In conclusion, oysters can be conditioned to increase their survival when challenged with

OsHV-1 with a controlled prior exposure to OsHV-1 and manipulation of the water temperature. Priming oysters with the non-specific immune priming agent poly I:C also reduced mortality due to OsHV-1. There was no mortality associated with immune conditioning by injection of heat inactivated OsHV-1. Additional development of more efficient administration techniques immune conditioning if heat inactivated OsHV-1 were to be used for reducing mortality caused by OsHV-1. This technique also has the drawback that diagnostic testing in the event of an outbreak may not be able to distinguish between oysters conditioned with heat inactivated OsHV-1 and oysters naturally infected with OsHV-1.

Further work in the field with larger numbers of oysters may be required to understand how long OsHV-1 may be detected in a cohort following conditioning when using different sampling strategies. The specificity of the protective effect from immune conditioning requires further evaluation using other potential pathogens. This would also provide a pathway to investigate and understand the mollusc immune system and the physiological cost of immune conditioning on the oysters. The potential for cross-protection against other pathogens requires further investigation. In addition, the duration of the protective effect needs to be determined along with interactions with water temperature, which plays an important role in the outcome of OsHV-1 infection and conditioning with OsHV-1.

178

8. General discussion

8.1 Introduction

Microvariant genotypes of Ostreid herpesvirus-1 (OsHV-1) emerged in France in 2008 causing mass mortalities in Pacific oysters (Crassostrea gigas) over summer (Martenot et al.,

2011; Segarra et al., 2010). Since its detection in the Georges River in 2010, OsHV-1 has had a detrimental impact on the Pacific oyster industry in NSW and Tasmania and been a threat to the rest of the Australian Pacific oyster industry (Chapter 3) (Jenkins et al., 2013; Paul-Pont et al., 2014). The immediate growing environment of the oyster, which is influenced by stocking density, height in the water column, growing location, hydrodynamics and the water temperature, is important in determining the level of mortality caused by OsHV-1 during an outbreak (de Kantzow et al., 2016, 2017; Paul-Pont et al., 2013a; Pernet et al., 2014a, 2012;

Petton et al., 2015a; Whittington et al., 2019, 2015a). Water temperature is a driving factor, with outbreaks beginning to occur each season as the water temperature increases to between

16°C and 18°C, depending on the region (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a;

Pernet et al., 2012; Petton et al., 2015a; Renault et al., 2014; Whittington et al., 2018).

However, OsHV-1 has been detected prior to outbreaks at this water temperature without concurrent mortality (Clegg et al., 2014; Paul-Pont et al., 2014, 2013a; Pernet et al., 2012;

Petton et al., 2015a; Renault et al., 2014; Whittington et al., 2018).

The research described in this thesis examined the effect of oyster farm management on the severity of disease caused by OsHV-1. Then, the use of controlled laboratory experiments enabled examination of the interactions between Pacific oysters and OsHV-1 under different water temperature conditions. The laboratory model reduced the impacts of confounding

179 variables which are associated with different water temperatures in the field, and identified the specific effects of the thermal profile on mortality.

8.2 Farm Management Alters Mortality Associated with OsHV-1

The growing environment provided to farmed Pacific oysters strongly influenced their risk of mortality during an OsHV-1 outbreak. In Chapter 3, there was an interaction between the length of time oysters had been on the farm, which reflected their age, and the stocking density within baskets. Young spat which had been on the farm for 3 to 6 months had a lower mortality when kept at the median stocking density for their age, whereas oysters which had been on the farm for 18 to 24 months had lower mortality when housed at either a high or low stocking density compared to the median stocking density for that age group. Handling oysters

(for example for size grading) within 7 days of the outbreak was also associated with increased mortality. This may have been due to stress reducing the ability of the oysters to mount an effective immune response (Duchemin et al., 2007; Lacoste et al., 2001b). A lack of handling over the entire summer season has been associated with high mortality during

OsHV-1 outbreaks (Ugalde et al., 2018), however the increase in mortality may have also been associated with the baskets becoming over-crowded as the oysters grew without size grading over the season. Low stocking density production systems such as rope culture have previously been associated with lower mortality than basket systems across all age groups

(Pernet et al., 2014a; Petton et al., 2015a). However, in the Tasmanian outbreak the highest and lowest stocking densities had the greatest mortality in younger stock (Chapter 3). The time of year when new oyster spat are purchased and placed into the estuary influences their age and size during the OsHV-1 risk period (Carrasco et al., 2017; Hick et al., 2018). Oysters which are larger and older at the time of an OsHV-1 outbreak have a lower risk of mortality

180

(Carrasco et al., 2017; Hick et al., 2018). To reduce the risk of mortality, the production calendar can be adjusted in order to maximise the age and size of oysters during summer. This could be done by purchasing larger spat and placing them out in autumn after the OsHV-1 risk period so they have the maximum time to grow before summer (Carrasco et al., 2017; Hick et al., 2018). Increasing the height of rack systems or baskets in the intertidal environment reduced how long the oysters were underwater during each tidal cycle and reduced mortality due to OsHV-1 during natural outbreaks. A reduction in the amount of time oysters were underwater may have reduced their exposure to OsHV-1 and directly altered their susceptibility (Evans et al., 2015; 2017; Whittington et al., 2015a). Oysters which are grown in an endemic location have the potential be exposed to low levels of OsHV-1 early in the season when the average water temperature is still at or below 16°C to18°C: this may reduce their risk of mortality later during the season. In endemic waterways where mortality is recurrent, the level mortality due to OsHV-1 is altered by both the local environment of the oyster and pre-exposure to OsHV-1.

8.3 Multiple Exposures to OsHV-1

In endemic estuaries, the pattern of mortality due to OsHV-1 may be affected by the water temperature during the period oysters are first exposed and infected with OsHV-1. Water temperature in estuaries is variable around the reported average over the course of a day in the intertidal zone where oysters are grown than in areas of the estuary with deeper water where average water temperature is measured. The variation around the average daily water temperature of 16°C to 20°C when POMS begins to occur may impact the occurrence of mortality (Clegg et al., 2014; Paul-Pont et al., 2014, 2013b; Petton et al., 2015a; Renault et al.,

2014; Whittington et al., 2019). The infection model used in this thesis focuses on the effect of

181 the average water temperature by maintaining a stable temperature with controlled changes at specific timepoints, which removes the effect of daily water temperature variation. Removing the variable exposure dose and temperature and the change in age and size which are present in the field allowed the effect of prior exposure on future mortality caused by OsHV-1 to be examined directly for the first time. Exposure to OsHV-1 when the water temperature was

18°C reduced the risk of mortality in oysters when they were exposed to OsHV-1 a second time at the higher water temperature of 22°C (Chapter 4). This indicated that the host response during the second exposure was changed by the prior exposure to OsHV-1 at 18°C. In waterways where OsHV-1 is endemic, oysters which had survived one OsHV-1 outbreak appeared to survive further outbreaks, suggesting that exposure to OsHV-1 was protective against further mortality (Evans et al., 2017; Whittington et al., 2018). The water temperature at which an oyster is first exposed to OsHV-1 changes not only the level of mortality but also the characteristics of viral replication during the initial infection (de Kantzow et al., 2016,

Chapter 5). This suggested that mortality due to OsHV-1 may be reduced in farmed oysters if they are naturally exposed to OsHV-1 at 18°C, before summer. Farmed oysters could only be exposed if effective transmission of OsHV-1 was possible at a water temperature of 18°C, but transmission was not able to be assessed while using the OsHV-1 injection exposure method used in this thesis. These effects of water temperature on OsHV-1 transmission need to be evaluated using a cohabitation exposure model where naïve oysters are housed with infected individuals (Evans et al., 2015).

The increase in water temperature over summer may cause oysters which were previously exposed to OsHV-1 to die from OsHV-1 if they are still infected or they may survive if they overcame the infection at the lower water temperature. Oysters that were exposed during an

182 outbreak at 18.1°C and then chilled to 10°C or 13°C experienced mortality once the water temperature was raised to 25°C (Pernet et al., 2015). However, it was demonstrated that following exposure to OsHV-1 at 18°C, the increase in water temperature to 22°C itself did not produce mortality in the previously exposed oysters (Chapter 5). OsHV-1 DNA was detected in 33% of oysters immediately prior to the water temperature increase, suggesting that a proportion may have been infected (Chapter 5). The oysters that were exposed and that did not succumb to OsHV-1 at 18°C may have been able to overcome the infection at either

18°C or after the increase in water temperature to 22°C. It is probable that oysters exposed to

OsHV-1 in the field during a natural outbreak were exposed to OsHV-1 above 18°C (Pernet et al., 2015). Additionally, variation in the water temperature in an estuary during transmission and the first hours of infection may have changed the way an oyster responds to an OsHV-1 infection following an increase in water temperature compared to exposure at a constant water temperature (Pernet et al., 2015) (Chapter 5). Understanding differences in the Pacific oyster immune response when exposed to OsHV-1 at different water temperatures may provide some insight into the mechanism that reduces mortality upon the second exposure. Further work is also required to understand how the magnitude of dynamic variation of the water temperature around the daily mean might affect the mortality caused by OsHV-1.

8.4 Water Temperature Affects Pacific Oyster Immune Response

The strong effect of water temperature on the level of mortality caused by OsHV-1 and the concentration of OsHV-1 DNA in oyster tissues following exposure suggests that water temperature may have an impact on the immune response of the Pacific oyster. The immediate transcriptional response of some Pacific oyster immune genes to stimulation with the synthetic dsRNA poly I:C is different at a water temperature of 12°C and 22°C (Green et al., 2014b). In

183

Chapter 6, the pattern of relative expression of the three selected Pacific oyster immune genes, viperin, Myeloid differentiation factor 88 (MyD88), and IAP repeat-containing protein 2

(IAP) in oysters exposed to OsHV-1 compared to control oysters that were not exposed to

OsHV-1 was different at a water temperature of 22°C and 18°C. At 18°C the three immune genes were up regulated in oysters sampled 2 hours after they were exposed to OsHV-1 and down regulated 24 hours after exposure to OsHV-1. However, at a water temperature of 22°C viperin and IAP expression followed a different pattern and were down regulated at 2 hours and upregulated at 24 hours post exposure. This contrasted with upregulated response of some immune genes, including IRF2 and Viperin, that were measured in response to stimulation with poly I:C at a water temperature of 22°C compared to those at 12°C (Green et al., 2014b).

The immune response to infection with OsHV-1 and stimulation with poly I:C are likely to be different, even though some genes expressed in response to stimulation with poly I:C are homologous to vertebrate antiviral genes (Green and Montagnani, 2013). Lower expression of

IAP in the 24 hours after OsHV-1 exposure was associated with mortality in the same individuals in the 14 days following OsHV-1 exposure (Chapter 6). This indicated that water temperature affected the regulation of some known immune genes and identified that this may help explain the effect of water temperature on the level of mortality caused by OsHV-1 infection. The transcriptome of OsHV-1 infected Pacific oysters revealed a number of unknown genes which were expressed at relatively high levels compared to known immune genes (Rosani et al., 2015). The use of less targeted techniques such as RNA sequencing to assess the transcriptome or proteomic analysis may be better suited to identifying candidate genes (Rosani et al., 2019).

184

The biopsy method used in Chapter 6 was the first time this method was used to allow direct linking of the gene expression in the 24 hours after exposure to the mortality data for the same individual over the next 14 days. Combining this sampling method with an analytical technique such as complete RNA transcriptome sequencing to measure changes in the transcriptome would allow for a more complete evaluation of genes that are affected by water temperature and which predict mortality. An approach using proteomics that has been applied to measure the response to stimulation with poly I:C and infection with OsHV-1 (Green et al.,

2016a; Tamayo et al., 2014) could also be applied at different water temperatures and to identify proteins associated with changes in mortality. An -omics approach using a study design that allows the results to be linked to the infection outcome is a promising area for future research in Pacific oyster immunology. Genes which are associated with reduced mortality due to OsHV-1 may also provide promising targets for genetic selection of OsHV-1 resistant oysters.

8.5 Prior Exposure to OsHV-1 as a Method of Protection from Disease

Development of effective and acceptable disease control techniques is important to reduce the impact of OsHV-1 on Pacific oyster aquaculture. In the present studies, Pacific oysters were protected from mortality caused by OsHV-1 infection by prior exposure to either infection- competent or heat-inactivated OsHV-1 at a water temperature of 18°C (Chapter 5, Chapter 7).

Other herpesviruses of aquatic animals can cause latent infections and it is possible that this may occur in the surviving oysters at a water temperature of either 18°C or 22°C. Further experiments over a longer period of time such as up to 24 months post OsHV-1 exposure are required to understand if a latent infection could occur and if this would pose a risk to other oysters over the period the OsHV-1 exposed oysters are likely to be on a farm. Non-OsHV-1

185 specific immune stimulation with poly I:C (a synthetic polymer) before oysters were exposed to OsHV-1 at 22°C also prevented mortality due to OsHV-1. This indicated that OsHV-1 replication was not required to produce a protective effect against mortality from OsHV-1

(Chapter 7). Further investigation is required to ascertain which aspects of the OsHV-1 exposure apart from water temperature are key to producing protection against mortality and to examine if the protective effect is specific to the genotype of OsHV-1 used in the initial exposure.

Previously, poly I:C has been shown to elicit an immune response homologous to a vertebrate antiviral immune response in oysters, and to provide protection against mortality caused by

OsHV-1 (Green and Montagnani, 2013; Lafont et al., 2017). However, of the immune genes that were differentially expressed in response to stimulation by poly I:C not all were differentially expressed following an OsHV-1 challenge in oysters which were selected for resistance to mortality and had not been exposed to poly I:C (Segarra et al., 2014, Chapter 6).

Additionally, expression levels of these genes did not predict mortality due to OsHV-1

(Segarra et al., 2014, Chapter 6). This indicates that a different set of immune genes are involved in the response to an OsHV-1 infection compared to those which are stimulated by exposure to poly I:C, although there is overlap. The mismatch in oyster genes expression following exposure to OsHV-1 and poly I:C limit the use of poly I:C exposure studies in understanding the immune response of the Pacific oyster to OsHV-1. The age, size and genotype of the oysters may have also impacted the immune response, as these factors affect the level of susceptibility to OsHV-1.

There are several challenges to the use of prior exposure with OsHV-1 as a disease control strategy, relating to issues of practicality and biosecurity. In order to be effective, a disease

186 control strategy must be able to be applied to the population at risk, in this instance commercially farmed Pacific oysters. The current method of prior exposure with OsHV-1 requires injection of individuals which is time consuming, technically difficult and is not suitable to be applied to thousands of oysters efficiently (section 2.3.3). Oysters have previously been successfully exposed to OsHV-1 by immersion in an OsHV-1 suspension, although this requires relaxation in an MgCl2 solution, and also by cohabitation in the same tank as infected individuals (Evans et al., 2015; Hick et al., 2018). Brood stock exposed to poly I:C have produced OsHV-1 resistant larvae (Green et al., 2016b). Further investigation is required to understand if brood stock can be exposed to either infection competent or heat inactivated OsHV-1 and then be used to produce OsHV-1 resistant larvae. Alternative techniques to injection exposure such as exposure by immersion in an OsHV-1 suspension

(Hick et al., 2016) may be more practical for large numbers of oysters and also warrant further development.

Acceptance of prior exposure as a disease control strategy will also depend on the biosecurity implications for growing areas (Department of Agriculture, 2015; OIE, 2016). While the use of prior exposure to OsHV-1 to reduce oyster mortality does not risk the introduction of

OsHV-1 in waterways where OsHV-1 is endemic, the use of stock which have been exposed to OsHV-1 as a control measure in areas which are at risk of OsHV-1 introduction would not be acceptable under the current movement restrictions for oysters (Department of Agriculture,

2015; OIE, 2016). The possibility of OsHV-1 causing a latent infection that could recrudesce and transmit OsHV-1 to naive oysters would have to be investigated to fully understand the biosecurity risk. However, the use of prior exposure to heat inactivated OsHV-1 or an attenuated OsHV-1 strain may be an acceptable preventative measure in areas which are

187 deemed at imminent risk of OsHV-1 introduction (Chapter 7). However, this does pose the risk that oysters that have been exposed will test positive for OsHV-1 by qPCR in surveillance programs. OsHV-1 DNA was detected in 33% of live, randomly sampled oysters 14 days after injection with OsHV-1 at 18°C (Chapter 7). OsHV-1 DNA was also detected in live oysters

45 days after injection at 18°C or 22°C (Chapter 7). This indicates that surveillance programs using qPCR to detect OsHV-1 DNA would need to take prior exposure into account, combine the qPCR data with mortality data and possibly use unexposed sentinels. It also indicates that

OsHV-1 DNA detected in the absence of mortality as part of a surveillance program may be the result of oysters being exposed to OsHV-1 at water temperatures at or around 18°C up to

45 days prior to sampling. Further work is required to test the efficacy of prior exposure to

OsHV-1 as a control measure on Pacific oyster farms in endemic waterways and to determine if it can be implemented at a scale suitable for commercial aquaculture.

8.6 Conclusion

Disease control recommendations for OsHV-1 can be developed with a greater understanding of how the interactions between Pacific oysters and OsHV-1 are affected by the environment and how this impacts disease incidence and severity. The environment, especially water temperature during the time of the initial exposure to OsHV-1, is important in determining the extent of mortality. Combined with the protective effect of prior exposure, the strong effect that the water temperature has on OsHV-1 pathogenesis in Pacific oysters could be leveraged to develop ways to reduce mortality due to OsHV-1 in commercial oyster aquaculture.

Deliberately exposing oysters to OsHV-1 at a low water temperature (18°C) was an effective way to reduce mortality when oysters were exposed to OsHV-1 at a higher water temperature which is normally associated with mass mortality in Australia (22°C). The combination of

188 controlled changes in water temperature and prior exposure to OsHV-1 also provided an insight into Pacific oyster immune function. The different pattern of expression of some immune genes at 18°C compared to at 22°C suggested that the response to OsHV-1 is influenced by different water temperatures and may be more effective against OsHV-1 at

18°C than at 22°C. Prior exposure to OsHV-1 should be evaluated for further development as a control method to reduce the impact of mortality caused by OsHV-1. Future work is required to understand what aspects of infection or exposure are required for a protective effect against mortality caused by OsHV-1. The application of a variety of technologies are required to analyse the effect of water temperature on the oyster immune response to OsHV-1 in greater detail in order to identify which aspects have the greatest influence on mortality.

189

9. References

ABARES, 2018. Australian fisheries and aquaculture statistics 2017. Canberra, ACT.

ABARES, 2013. Australian fisheries and aquaculture statistics 2012. Canberra, ACT.

Agricultural and Veterinary Chemicals Code Act 1994 (Commonwealth). Retrieved from

https://www.legislation.gov.au/Details/C2016C00999.

Allam, B., Raftos, D., 2015. Immune responses to infectious diseases in bivalves. J. Invertebr.

Pathol. 131, 121–136. doi:10.1016/j.jip.2015.05.005

Animal Health Australia, 2019. Animal health in Australia 2018, Tropical Animal Health and

Production. Canberra. doi:10.1007/bf02359722

Animal Health Australia, 2016. Animal Health Surveillance Quarterly Report - Aquatic

Animal Health. Animal Health Surveillance Quarterly Report. Canberra, ACT.

Arzul, I., Nicolas, J., Davison, A., Renault, T., 2001a. French Scallops: A New Host for

Ostreid Herpesvirus-1. Virology 290, 342–349. doi:10.1006/viro.2001.1186

Arzul, I., Renault, T., Lipart, C., Davison, A.J., 2001b. Evidence for interspecies transmission

of oyster herpesvirus in marine bivalves. J. Gen. Virol. 82, 865–870. doi:10.1099/0022-

1317-82-4-865

Australian Bureau of Statistics, 2016. Australian Statistical Geography Standard (ASGS):

Volume 1 - Main Structure and Greater Capital City Statistical Areas, July 2016.

Canberra, ACT.

Azéma, P., Maurouard, E., Lamy, J.-B., Dégremont, L., 2017. The use of size and growing

height to improve Crassostrea gigas farming and breeding techniques against OsHV-1.

190

Aquaculture 471, 121–129. doi:10.1016/j.aquaculture.2017.01.011

Azéma, P., Travers, M.-A., Benabdelmouna, A., Dégremont, L., 2016. Single or dual

experimental infections with Vibrio aestuarianus and OsHV-1 in diploid and triploid

Crassostrea gigas at the spat, juvenile and adult stages. J. Invert. Pathol. 139, 92-101.

doi:10.1016/j.jip.2016.08.002

Bai, C.-M., Morga, B., Rosani, U., Shi, J., Li, C., Xin, L.-S., Wang, C.-M., 2019. Long-range

PCR and high-throughput sequencing of Ostreid herpesvirus 1 indicate high genetic

diversity and complex evolution process. Virology 526, 81–90.

doi:10.1016/j.virol.2018.09.026

Barbosa Solomieu, V.V., Renault, T., Travers, M.-A.A., 2015. Mass mortality in bivalves and

the intricate case of the Pacific oyster, Crassostrea gigas. J. Invertebr. Pathol. 131, 2–10.

doi:10.1016/j.jip.2015.07.011

Bates, D., Maechler, M., Bolker, B., Walker, S., 2015. Fitting Linear Mixed-Effects Models

Using {lme4}. J. Stat. Softw. 67, 1–48. doi:10.18637/jss.v067.i01

Batista, F.M., Arzul, I., Pepin, J.F., Ruano, F., Friedman, C.S., Boudry, P., Renault, T., 2007.

Detection of Ostreid herpesvirus 1 DNA by PCR in bivalve molluscs: A critical review.

J. Virol. Methods 139, 1–11. doi:10.1016/j.jviromet.2006.09.005

Batista, F.M., López-Sanmartín, M., Grade, A., Morgado, I., Valente, M., Navas, J.I., Power,

D.M., Ruano, F., 2015. Sequence variation in Ostreid herpesvirus 1 microvar isolates

detected in dying and asymptomatic Crassostrea angulata adults in the Iberian

Peninsula: Insights into viral origin and spread. Aquaculture 435, 43–51.

doi:https://doi.org/10.1016/j.aquaculture.2014.09.016

191

Bingham, P., Brangenberg, N., Williams, R., Andel, M. V., 2013. Marine and freshwater

investigation into the first diagnosis of Ostreid herpesvirus 1 in Pacific oysters. Surveill.

40, 20–24.

Biosecurity Tasmania, 2016. Pacific Oyster Mortality Syndrome (POMS). Hobart, TAS.

Bly, J.E., Clem, L.W., 1992. Temperature and teleost immune functions. Fish Shellfish

Immunol. 2, 159–171. doi:10.1016/S1050-4648(05)80056-7

Boyd, C., 2012. Water quality, in: Lucas, J., Southgate, P. (Eds.), Aquaculture: Farming

Aquatic Animals and Plants. Wiley-Blackwell, Oxford, pp. 52–82.

Breslow, N.E., 1992. Introduction to Kaplan and Meier (1958) Nonparametric Estimation

from Incomplete Observations. Break. Stat. 311–318. doi:10.1007/978-1-4612-4380-

9_24

Burioli, E.A. V, Prearo, M., Houssin, M., 2017. Complete genome sequence of Ostreid

herpesvirus type 1 µVar isolated during mortality events in the Pacific oyster Crassostrea

gigas in France and Ireland. Virol. 509, 239-251. doi:10.1016/j.virol.2017.06.027

Burioli, E.A. V, Prearo, M., Riina, M. V., Bona, M.C., Fioravanti, M.L., Arcangeli, G.,

Houssin, M., 2016. Ostreid herpesvirus type 1 genomic diversity in wild populations of

Pacific oyster Crassostrea gigas from Italian coasts. J. Invertebr. Pathol. 137, 71–83.

doi:10.1016/j.jip.2016.05.004

Butt, D., Raftos, D., 2007. Immunosuppression in Sydney rock oysters (Saccostrea

glomerata) and QX disease in the Hawkesbury River, Sydney. Mar. Freshw. Res. 58,

213–221.

192

Callaway, R., Shinn, A.P., Grenfell, S.E., Bron, J.E., Burnell, G., Cook, E.J., Crumlish, M.,

Culloty, S., Davidson, K., Ellis, R.P., Flynn, K.J., Fox, C., Green, D.M., Hays, G.C.,

Hughes, A.D., Johnston, E., Lowe, C.D., Lupatsch, I., Malham, S., Mendzil, A.F.,

Nickell, T., Pickerell, T., Rowley, A.F., Stanley, M.S., Tocher, D.R., Turnbull, J.F.,

Webb, G., Wootton, E., Shields, R.J., 2012. Review of climate change impacts on marine

aquaculture in the UK and Ireland. Aquat. Conserv. Mar. Freshw. Ecosyst. 22, 389–421.

doi:10.1002/aqc.2247

Carrasco, N., Gairin, I., Pérez, J., Andree, K.B., Roque, A., Fernández-Tejedor, M., Rodgers,

C.J., Aguilera, C., Furones, M.D., 2017. A production calendar based on water

temperature, spat size, and husbandry practices reduce OsHV-1 μvar impact on cultured

Pacific oyster Crassostrea gigas in the Ebro Delta (Catalonia), Mediterranean coast of

Spain. Front. Physiol. 8. doi:10.3389/fphys.2017.00125

Carrasco, N., Green, T., Itoh, N., 2015. Marteilia spp. parasites in bivalves: A revision of

recent studies. J. Invertebr. Pathol. 131, 43–57. doi:10.1016/j.jip.2015.07.016

Child, A.R., Laing, I., 1998. Comparative low temperature tolerance of small juvenile

European, Ostrea edulis L., and Pacific oysters, Crassostrea gigas Thunberg. Aquac.

Res. 29, 103–113. doi:10.1046/j.1365-2109.1998.00934.x

Clarke, A., Fraser, K.P.P., 2004. Why does metabolism scale with temperature? Funct. Ecol.

18, 243–251. doi:10.1111/j.0269-8463.2004.00841.x

Clarke, A., Johnston, N.M., 1999. Scaling of metabolic rate with body mass and temperature

in teleost fish. J. Anim. Ecol. 68, 893–905. doi:10.1046/j.1365-2656.1999.00337.x

Clegg, T.A., Morrissey, T., Geoghegan, F., Martin, S.W., Lyons, K., Ashe, S., More, S.J.,

193

2014. Risk factors associated with increased mortality of farmed Pacific oysters in

Ireland during 2011. Prev. Vet. Med. 113, 257–267.

doi:10.1016/j.prevetmed.2013.10.023

Corporeau, C., Tamayo, D., Pernet, F., Quere, C., Madec, S., 2014. Proteomic signatures of

the oyster metabolic response to herpesvirus OsHV-1 μVar infection. J. Proteomics 109,

176–187. doi:10.1016/j.jprot.2014.06.030

Cox, D., 1972. Regression Models and Life-Tables. J. R. Stat. Soc. 34, 187–220.

Dadar, M., Dhama, K., Vakharia, V.N., Hoseinifar, S.H., Karthik, K., Tiwari, R., Khandia, R.,

Munjal, A., Salgado-Miranda, C., Joshi, S.K., 2017. Advances in aquaculture vaccines

against fish pathogens: global status and current trends. Rev. Fish. Sci. Aquac. 25, 184–

217. doi:10.1080/23308249.2016.1261277

Dalmo, R.A., 2018. DNA vaccines for fish: Review and perspectives on correlates of

protection. J. Fish Dis. 41, 1–9. doi:10.1111/jfd.12727

Dang, V.T., Benkendorff, K., Green, T., Speck, P., 2015. Marine snails and slugs: a great

place to look for antiviral drugs: TABLE 1. J. Virol. 89, 8114–8118.

doi:10.1128/JVI.00287-15

David Kahle, H.W., 2013. ggmap: spatial visualization with ggplot2. R J. 5, 144–161.

Davison, A.J., Eberle, R., Ehlers, B., Hayward, G.S., McGeoch, D.J., Minson, A.C., Pellett,

P.E., Roizman, B., Studdert, M.J., Thiry, E., 2011. The order Herpesvirales. Arch. Virol.

46, 564–574. doi:10.1016/j.cortex.2009.08.003.Predictive

Davison, A.J., Trus, B.L., Cheng, N., Steven, A., Watson, M.S., Cunningham, C., Le Deuff,

194

R.M., Renault, T., 2005. A novel class of herpesvirus with bivalve hosts. J. Gen. Virol.

86, 41–53. doi:10.1099/vir.0.80382-0 de Kantzow, M., Hick, P., Becker, J., Whittington, R., 2016. Effect of water temperature on

mortality of Pacific oysters Crassostrea gigas associated with microvariant Ostreid

herpesvirus 1 (OsHV-1 µVar). Aquac. Environ. Interact. 8, 419–428.

doi:10.3354/aei00186 de Kantzow, M.C., Hick, P.M., Dhand, N.K., Whittington, R.J., 2017. Risk factors for

mortality during the first occurrence of Pacific Oyster Mortality Syndrome due to Ostreid

herpesvirus – 1 in Tasmania, 2016. Aquaculture 468, 328–336.

doi:10.1016/j.aquaculture.2016.10.025 de Kantzow, M.C., Whittington, R.J., Hick, P., 2019a. Prior exposure to Ostreid herpesvirus 1

(OsHV-1) at 18 °C is associated with improved survival of juvenile Pacific oysters

(Crassostrea gigas) following challenge at 22 °C. Aquaculture 507, 443–450.

doi:https://doi.org/10.1016/j.aquaculture.2019.04.035 de Kantzow, M.C., Whittington, R.J., Hick, P.M., 2019b. Different in vivo growth of Ostreid

herpesvirus 1 at 18°C and 22°C alters mortality of Pacific oysters (Crassostrea gigas).

Arch. Virol. doi:10.1007/s00705-019-04427-2 de Lorgeril, J., Lucasson, A., Petton, B., Toulza, E., Montagnani, C., Clerissi, C., Vidal-

Dupiol, J., Chaparro, C., Galinier, R., Escoubas, J.M., Haffner, P., Dégremont, L.,

Charrière, G.M., Lafont, M., Delort, A., Vergnes, A., Chiarello, M., Faury, N., Rubio, T.,

Leroy, M.A., Pérignon, A., Régler, D., Morga, B., Alunno-Bruscia, M., Boudry, P., Le

Roux, F., Destoumieux-Garzόn, D., Gueguen, Y., Mitta, G., 2018. Immune-suppression

195

by OsHV-1 viral infection causes fatal bacteraemia in Pacific oysters. Nat. Commun. 9.

doi:10.1038/s41467-018-06659-3

Deane, E.E., Woo, N.Y.S., 2009. Modulation of fish growth hormone levels by salinity,

temperature, pollutants and aquaculture related stress: A review. Rev. Fish Biol. Fish. 19,

97–120. doi:10.1007/s11160-008-9091-0

Dégremont, L., 2013. Size and genotype affect resistance to mortality caused by OsHV-1 in

Crassostrea gigas. Aquaculture 416–417, 129–134.

doi:10.1016/j.aquaculture.2013.09.011

Dégremont, L., 2011. Evidence of herpesvirus (OsHV-1) resistance in juvenile Crassostrea

gigas selected for high resistance to the summer mortality phenomenon. Aquaculture

317, 94–98. doi:10.1016/j.aquaculture.2011.04.029

Dégremont, L., Bédier, E., Boudry, P., 2010. Summer mortality of hatchery-produced Pacific

oyster spat (Crassostrea gigas). II. Response to selection for survival and its influence on

growth and yield. Aquaculture 299, 21–29. doi:10.1016/j.aquaculture.2009.11.017

Degremont, L., Benabdelmouna, A., 2014. Mortality associated with OsHV-1 in spat

Crassostrea gigas: role of wild-caught spat in the horizontal transmission of the disease.

Aquac. Int. 22, 1767–1781. doi:10.1007/s10499-014-9781-7

Degremont, L., Garcia, C., Allen, S.K., 2015. Genetic improvement for disease resistance in

oysters: A review. J. Invertebr. Pathol. 131, 226–241. doi:10.1016/j.jip.2015.05.010

Dégremont, L., Lamy, J.B., Pépin, J.F., Travers, M.A., Renault, T., 2015a. New insight for the

genetic evaluation of resistance to Ostreid herpesvirus infection, a worldwide disease, in

196

Crassostrea gigas. PLoS One 10, 1–12. doi:10.1371/journal.pone.0127917

Dégremont, L., Ledu, C., Maurouard, E., Nourry, M., Benabdelmouna, A., 2016. Effect of

ploidy on the mortality of Crassostrea gigas spat caused by OsHV-1 in France using

unselected and selected OsHV-1 resistant oysters. Aquac. Res. 47, 777–786.

doi:10.1111/are.12536

Dégremont, L., Nourry, M., Maurouard, E., 2015b. Mass selection for survival and resistance

to OsHV-1 infection in Crassostrea gigas spat in field conditions: Response to selection

after four generations. Aquaculture 446, 111–121. doi:10.1016/j.aquaculture.2015.04.029

Delisle, L., Petton, B., Burguin, J.F., Morga, B., Corporeau, C., Pernet, F., 2018. Temperature

modulate disease susceptibility of the Pacific oyster Crassostrea gigas and virulence of

the Ostreid herpesvirus type 1. Fish Shellfish Immunol. 80, 71–79.

doi:https://doi.org/10.1016/j.fsi.2018.05.056

Department of Agriculture, 2015. AQUAVETPLAN Disease Strategy Infection with Ostreid

herpesvirus-1 microvariant. Canberra, ACT. URL

https://www.agriculture.gov.au/animal/aquatic/aquavetplan/ostreid-herpesvirus

Deveney, M., Roberts, S., Moody, N., Crane, M., Ellard, K., 2017. Biofouling as a long

distance vector for pathogens, in: Proceedings of the 4th FRDC Australasian Aquatic

Animal Health & Biosecurity Scientific Conference. Fisheries Research and Develoment

Corporation, Canberra, p. 35.

Dhand, N.K., Khatkar, M.S., 2014. Statulator: An online statistical calculator. Sample Size

Calculator for Estimating a Single Proportion. URL: http://statulator.com.

197

Dios, S., Romero, A., Chamorro, R., Figueras, A., Novoa, B., 2010. Effect of the temperature

during antiviral immune response ontogeny in teleosts. Fish Shellfish Immunol. 29,

1019–1027. doi:10.1016/j.fsi.2010.08.006

Dishon, A., Davidovich, M., Ilouze, M., Kotler, M., 2007. Persistence of Cyprinid herpesvirus

3 in infected cultured carp cells. J. Virol. 81, 4828 LP – 4836. doi:10.1128/JVI.02188-06

Dorai-Raj, S., 2014. Package ‘binom’: Binomial Confidence Intervals For Several

Parameterizations. URL https://cran.r-project.org/web/packages/binom/binom.pdf

Du, Y., Zhang, L., Huang, B., Guan, X., Li, L., Zhang, G., 2013a. Molecular cloning,

characterization, and expression of two myeloid differentiation factor 88 (Myd88) in

Pacific oyster, Crassostrea gigas. J. World Aquac. Soc. 44, 759–774.

doi:10.1111/jwas.12077

Du, Y., Zhang, L., Xu, F., Huang, B., Zhang, G., Li, L., 2013b. Validation of housekeeping

genes as internal controls for studying gene expression during Pacific oyster (Crassostrea

gigas) development by quantitative real-time PCR. Fish Shellfish Immunol. 34, 939–945.

doi:10.1016/j.fsi.2012.12.007

Duchemin, M.B., Fournier, M., Auffret, M., 2007. Seasonal variations of immune parameters

in diploid and triploid Pacific oysters, Crassostrea gigas (Thunberg). Aquaculture 264,

73–81. doi:10.1016/j.aquaculture.2006.12.030

Dumbauld, B.R., Ruesink, J.L., Rumrill, S.S., 2009. The ecological role of bivalve shellfish

aquaculture in the estuarine environment: A review with application to oyster and clam

culture in West Coast (USA) estuaries. Aquaculture 290, 196–223.

doi:10.1016/j.aquaculture.2009.02.033

198

EFSA, 2015. European Food Safety Authority Panel on Animal HealthWelfare. Oyster

mortality. EFSA J. 13.

EFSA, 2010. Scientific opinion on the increased mortality events in Pacific oysters. Eur. Food

Saf. Auth. J. 8, 1–60. doi:10.2903/j.efsa.2010.1894.

Eide, K., Miller-Morgan, T., Heidel, J., Bildfell, R., Jin, L., 2011a. Results of total DNA

measurement in koi tissue by Koi herpesvirus real-time PCR. J. Virol. Methods 172, 81–

84. doi:https://doi.org/10.1016/j.jviromet.2010.12.012

Eide, K., Miller-Morgan, T., Heidel, J.R., Kent, M.L., Bildfell, R.J., Lapatra, S., Watson, G.,

Jin, L., 2011b. Investigation of Koi herpesvirus latency in koi. J. Virol. 85, 4954–4962.

doi:10.1128/JVI.01384-10

Elston, R.A., 1993. Infectious diseases of the Pacific oyster, Crassostrea gigas. Annu. Rev.

Fish Dis. 3, 259–276. doi:10.1016/0959-8030(93)90038-D

Escobedo-Bonilla, C.M., Vega-Peña, S., Mejía-Ruiz, C.H., 2015. Efficacy of double-stranded

RNA against white spot syndrome virus (WSSV) non-structural (orf89, wsv191) and

structural (vp28, vp26) genes in the Pacific white shrimp Litopenaeus vannamei. J. King

Saud Univ. - Sci. 27, 182–188. doi:10.1016/j.jksus.2014.11.004

Evans, O., Hick, P., Dhand, N., Whittington, R., 2015. Transmission of Ostreid herpesvirus-1

in Crassostrea gigas by cohabitation: effects of food and number of infected donor

oysters. Aquac. Environ. Interact. 7, 281–295. doi:10.3354/aei00160

Evans, O., Hick, P., Whittington, R.J., 2017. Detection of Ostreid herpesvirus-1 microvariants

in healthy Crassostrea gigas following disease events and their possible role as

199

reservoirs of infection. J. Invertebr. Pathol. 148, 20–33. doi:10.1016/j.jip.2017.05.004

Evans, O., Hick, P., Whittington, R.J., 2016. Distribution of Ostreid herpesvirus-1 (OsHV-1)

microvariant in seawater in a recirculating aquaculture system. Aquaculture 458, 21–28.

doi:10.1016/j.aquaculture.2016.02.027

Evans, O., Kan, J.Z.F., Pathirana, E., Whittington, R.J., Dhand, N., Hick, P., 2019. Effect of

emersion on the mortality of Pacific oysters (Crassostrea gigas) infected with Ostreid

herpesvirus-1 (OsHV-1). Aquaculture 505, 157–166.

doi:https://doi.org/10.1016/j.aquaculture.2019.02.041

Evans, O., Paul-Pont, I., Hick, P., Whittington, R.J.J., 2014. A simple centrifugation method

for improving the detection of Ostreid herpesvirus-1 (OsHV-1) in natural seawater

samples with an assessment of the potential for particulate attachment. J. Virol. Methods

210, 59–66. doi:10.1016/j.jviromet.2014.09.023

Farley, C.A., Banfield, W.G., Kasnic, G., Foster, W.S., 1972. Oyster herpes-type virus.

Science (80-. ). 178, 759–760.

Fisheries Management Act 1994 (NSW). Retrieved from

https://www.legislation.nsw.gov.au/#/view/act/1994/38

Forrest, B.M., Keeley, N.B., Hopkins, G.A., Webb, S.C., Clement, D.M., 2009. Bivalve

aquaculture in estuaries: Review and synthesis of oyster cultivation effects. Aquaculture

298, 1–15. doi:10.1016/j.aquaculture.2009.09.032

Friedman, C.S., Hedrick, R.P., 1991. Pacific oyster nocardiosis: Isolation of the bacterium and

induction of laboratory infections. J. Invertebr. Pathol. 57, 109–120. doi:10.1016/0022-

200

2011(91)90047-T

Fuhrmann, M., Delisle, L., Petton, B., Corporeau, C., Pernet, F., 2018. Metabolism of the

Pacific oyster, Crassostrea gigas, is influenced by salinity and modulates survival to the

Ostreid herpesvirus OsHV-1 . Biol. Open 7, bio028134. doi:10.1242/bio.028134

Gagnaire, B., Frouin, H., Moreau, K., Thomas-Guyon, H., Renault, T., 2006. Effects of

temperature and salinity on haemocyte activities of the Pacific oyster, Crassostrea gigas

(Thunberg). Fish Shellfish Immunol. 20, 536–547. doi:10.1016/j.fsi.2005.07.003

Garcia, C., Thébault, A., Dégremont, L., Arzul, I., Miossec, L., Robert, M., Chollet, B.,

François, C., Joly, J.-P.P., Ferrand, S., Kerdudou, N., Renault, T., 2011. Ostreid

herpesvirus 1 detection and relationship with Crassostrea gigas spat mortality in France

between 1998 and 2006. Vet. Res. 42, 73. doi:10.1186/1297-9716-42-73

Gene Technology Act 2000 (Commonwealth). Retrieved from

https://www.legislation.gov.au/Details/C2016C00792

Gene Technology Regulations Act 2001 (Commonwealth). Retrieved from

https://www.legislation.gov.au/Details/F2020C00651

Gittenberger, A., Voorbergen-Laarman, M.A., Engelsma, M.Y., 2016. Ostreid herpesvirus

OsHV-1 μVar in Pacific oysters Crassostrea gigas (Thunberg 1793) of the Wadden Sea,

a UNESCO world heritage site. J. Fish Dis. 39, 105–109. doi:10.1111/jfd.12332

Goulletquer, P., Heral, M., 1997. Marine Molluscan Production Trends in France: From

Fisheries to Aquaculture. C.L. MacKenzie, V.G. Burrell, A. Rosenfield, H. W. (Ed.),

Hist. Present Cond. Futur. Molluscan Fish. North Am. Eur. NOAA Tech. Rep. NMFS

201

129, Dep. Commer. Seattle, Washington, 1997, 137-164. NO 137–164.

Green, T., Chataway, T., Melwani, A.R., Raftos, D.A., 2016a. Proteomic analysis of

hemolymph from poly(I:C)-stimulated Crassostrea gigas. Fish Shellfish Immunol. 48,

39–42. doi:10.1016/j.fsi.2015.11.018

Green, T., Helbig, K., Speck, P., Raftos, D., 2016b. Primed for success: Oyster parents treated

with poly(I:C) produce offspring with enhanced protection against Ostreid herpesvirus

type I infection. Mol. Immunol. 78. doi:10.1016/j.molimm.2016.09.002

Green, T.J., Benkendorff, K., Robinson, N., Raftos, D., Speck, P., 2014a. Anti-viral gene

induction is absent upon secondary challenge with double-stranded RNA in the Pacific

oyster, Crassostrea gigas. Fish Shellfish Immunol. 39, 492–497.

doi:10.1016/j.fsi.2014.06.010

Green, T.J., Montagnani, C., 2013. Poly I: C induces a protective antiviral immune response

in the Pacific oyster (Crassostrea gigas) against subsequent challenge with Ostreid

herpesvirus (OsHV-1 μvar). Fish Shellfish Immunol. 35, 382–388.

doi:10.1016/j.fsi.2013.04.051

Green, T.J., Montagnani, C., Benkendorff, K., Robinson, N., Speck, P., 2014b. Ontogeny and

water temperature influences the antiviral response of the Pacific oyster, Crassostrea

gigas. Fish Shellfish Immunol. 36, 151–157. doi:10.1016/j.fsi.2013.10.026

Green, T.J., Raftos, D., O’Connor, W., Adlard, R.D., Barnes, A.C., 2011. Disease prevention

strategies for QX disease (Marteilia sydneyi) of Sydney rock oysters (Saccostrea

glomerata) . J. Shellfish Res. 30, 47–53. doi:10.2983/035.030.0108

202

Green, T.J., Raftos, D., Speck, P., Montagnani, C., 2015a. Antiviral immunity in marine

molluscs. J. Gen. Virol. 96, 2471–2482. doi:10.1099/jgv.0.000244

Green, T.J., Rolland, J.-L.L., Vergnes, A., Raftos, D., Montagnani, C., 2015b. OsHV-1

countermeasures to the Pacific oyster’s anti-viral response. Fish Shellfish Immunol. 47,

435–443. doi:10.1016/j.fsi.2015.09.025

Green, T.J., Speck, P., 2018. Antiviral Defense and Innate Immune Memory in the Oyster.

Viruses 10. doi:10.3390/v10030133

Green, T.J., Speck, P., Geng, L., Raftos, D., Beard, M.R., Helbig, K.J., 2015c. Oyster viperin

retains direct antiviral activity and its transcription occurs via a signalling pathway

involving a heat-stable haemolymph protein. J. Gen. Virol. 96, 3587–3597.

doi:10.1099/jgv.0.000300

Green, T.J., Vergnes, A., Montagnani, C., de Lorgeril, J., 2016. Distinct immune responses of

juvenile and adult oysters (Crassostrea gigas) to viral and bacterial infections. Vet. Res.

47, 47:72. doi:10.1186/s13567-016-0356-7

Grundhoff, A., Sullivan, C.S., 2011. Virus-encoded microRNAs. Virology 411, 325–343.

doi:10.1016/j.virol.2011.01.002

Gutierrez, A.P., Bean, T.P., Hooper, C., Stenton, C.A., Sanders, M.B., Paley, R.K., Rastas, P.,

Bryrom, M., Matika, O., Houston, R.D., 2018. A genome-wide association study for host

resistance to Ostreid herpesvirus in Pacific oysters (Crassostrea gigas). G3 (Bethesda).

8, 1273–1280. doi:10.1534/g3.118.200113

Handley, S.J., 1995. Spionid polychaetes in Pacific oysters, Crassostrea gigas (thunberg)

203

from Admiralty Bay, Marlborough Sounds, New Zealand. New Zeal. J. Mar. Freshw.

Res. 29, 305–309. doi:10.1080/00288330.1995.9516665

Handley, S.J., Bergquist, P.R., 1997. Spionid polychaete infestations of intertidal Pacific

oysters Crassostrea gigas (Thunberg), Mahurangi Harbour, northern New Zealand.

Aquaculture 153, 191–205. doi:10.1016/S0044-8486(97)00032-X

Hartmann, D., Adams, J.S., Meeker, A.K., Schneider, M.A., Lenz, B.F., Talmadge, J.E., 1986.

Dissociation of therapeutic and toxic effects of polyinosinic-polycytidylic acid admixed

with poly-lysine and solubilized with carboxymethyl cellulose in tumor-bearing mice.

Cancer Res. 46, 1331 LP – 1338.

Hartmann, D., Schneider, M., Lenz, B., Talmadge, J., 1987. Toxicity of polyinosinic-

polycytidylic acid admixed with poly-L-lysine and solubilized with

carboxymethylcellulose in mice. Pathol Immunopathol Res. 6, 37–50.

He, Y., Jouaux, A., Ford, S.E., Lelong, C., Sourdaine, P., Mathieu, M., Guo, X., 2015.

Transcriptome analysis reveals strong and complex antiviral response in a mollusc. Fish

Shellfish Immunol. 46, 131–144. doi:10.1016/j.fsi.2015.05.023

Hick, P., Evans, O., Looi, R., English, C., Whittington, R.J., 2016. Stability of Ostreid

herpesvirus-1 (OsHV-1) and assessment of disinfection of seawater and oyster tissues

using a bioassay. Aquaculture 450, 412–421.

doi:https://doi.org/10.1016/j.aquaculture.2015.08.025

Hick, P.M., Evans, O., Rubio, A., Dhand, N.K., Whittington, R.J., 2018. Both age and size

influence susceptibility of Pacific oysters (Crassostrea gigas) to disease caused by

Ostreid herpesvirus-1 (OsHV-1) in replicated field and laboratory experiments.

204

Aquaculture 489, 110–120. doi:https://doi.org/10.1016/j.aquaculture.2018.02.013

Homan, E.R., Zendzian, R.P., Schott, L.D., Levy, H.B., Adamson, R.H., 1972. Studies on

Poly I:C toxicity in experimental animals. Toxicol. Appl. Pharmacol. 23, 579–588.

doi:https://doi.org/10.1016/0041-008X(72)90098-1

In, V., O’Connor, W., Dove, M., Knibb, W., 2016. Can genetic diversity be maintained across

multiple mass selection lines of Sydney rock oyster, Saccostrea glomerata despite loss

within each. Aquaculture 454, 210–216.

doi:https://doi.org/10.1016/j.aquaculture.2015.12.030

Jacobs, B.L., Langland, J.O., 1996. When two strands are better than one: the mediators and

modulators of the cellular responses to double-stranded RNA. Virology 219, 339–349.

doi:10.1006/viro.1996.0259

Jenkins, C., Hick, P., Gabor, M., Spiers, Z., Fell, S.A., Gu, X., Read, A., Go, J., Dove, M.,

O’Connor, W., Kirkland, P.D., Frances, J., 2013. Identification and characterisation of an

Ostreid herpesvirus-1 microvariant (OsHV-1 μVar) in Crassostrea gigas (Pacific

oysters) in Australia. Dis. Aquat. Organ. 105, 109–126. doi:10.3354/dao02623

Jouaux, A., Lafont, M., Blin, J.-L., Houssin, M., Mathieu, M., Lelong, C., 2013. Physiological

change under OsHV-1 contamination in Pacific oyster Crassostrea gigas through

massive mortality events on fields. BMC Genomics 14, 1–14. doi:10.1186/1471-2164-

14-590

Kaplan, E.L., Meier, P., 1958. Nonparametric Estimation from Incomplete Observations. J.

Am. Stat. Assoc. 53, 457–481. doi:10.2307/2281868

205

Karvonen, A., Rintamäki, P., Jokela, J., Valtonen, E.T., 2010. Increasing water temperature

and disease risks in aquatic systems: Climate change increases the risk of some, but not

all, diseases. Int. J. Parasitol. 40, 1483–1488. doi:10.1016/j.ijpara.2010.04.015

Kassambara, A., Kosinski, M., 2017. survminer: Drawing Survival Curves using “ggplot2.”

Keeling, S., Brosnahan, C., Williams, R., Gias, E., Hannah, M., Bueno, R., McDonald, W.,

Johnston, C., 2014. New Zealand juvenile oyster mortality associated with Ostreid

herpesvirus 1 - an opportunistic longitudinal study. Dis. Aquat. Organ. 109, 231–239.

King, W.L., Jenkins, C., Seymour, J.R., Labbate, M., 2019. Oyster disease in a changing

environment: Decrypting the link between pathogen, microbiome and environment. Mar.

Environ. Res. 143, 124–140. doi:10.1016/j.marenvres.2018.11.007

Klein, J., 1989. Are Invertebrates Capable of Anticipatory Immune Responses? Scand. J.

Immunol. 29, 499–505. doi:10.1111/j.1365-3083.1989.tb01152.x

Klinger, D.H., Levin, S.A., Watson, J.R., 2017. The growth of finfish in global open-ocean

aquaculture under climate change. Proc. R. Soc. B Biol. Sci. 284.

doi:10.1098/rspb.2017.0834

Kloosterman, W.P., Plasterk, R.H.A., 2006. The diverse functions of micrornas in animal

development and disease. Dev. Cell. 11, 441–450. doi:10.1016/j.devcel.2006.09.009

Krown, S., Kerr, D., Stewart, W., Field, A., Oettgen, H., 1985. Phase I trials of poly(I,C)

complexes in advanced cancer. J. Biol. Response Mod. 4, 640–649.

Kurtz, J., 2004. Memory in the innate and adaptive immune systems. Microbes Infect. 6,

1410–1417. doi:10.1016/j.micinf.2004.10.002

206

Kurtz, J., Franz, K., 2003a. Evidence for memory in invertebrate immunity. Nature 425, 37–

38. doi:10.1038/425037a

Kurtz, J., Franz, K., 2003b. Innate defence: evidence for memory in invertebrate immunity.

Nature 425, 37–38. doi:10.1038/425037a

Kuznetsova, A., Brockhoff, P.B., Christensen, R.H.B., 2017. lmerTest Package: Tests in linear

mixed effects models. J. Stat. Softw. 82, 1–26. doi:10.18637/jss.v082.i13

Lacoste, A., Jalabert, F., Malham, S.K., Cueff, A., Poulet, S.A., 2001a. Stress and stress-

induced neuroendocrine changes increase the susceptibility of juvenile oysters

(Crassostrea gigas) to Vibrio splendidus. Appl. Environ. Microbiol. 67, 2304–2309.

doi:10.1128/AEM.67.5.2304-2309.2001

Lacoste, A, Malham, S.K., Cueff, A, Poulet, S. A, 2001b. Stress-induced catecholamine

changes in the hemolymph of the oyster Crassostrea gigas. Gen. Comp. Endocrinol. 122,

181–188. doi:10.1006/gcen.2001.7629

Lacoste, A., Malham, S.K., Gélébart, F., Cueff, A., Poulet, S.A., 2002. Stress-induced

immune changes in the oyster Crassostrea gigas. Dev. Comp. Immunol. 26, 1–9.

Lafferty, K.D., Porter, J.W., Ford, S.E., 2004. Are diseases increasing in the ocean? Annu.

Rev. Ecol. Evol. Syst. 35, 31–54. doi:10.1146/annurev.ecolsys.35.021103.105704

Lafont, M., Petton, B., Vergnes, A., Pauletto, M., Segarra, A., Gourbal, B., Montagnani, C.,

2017. Long-lasting antiviral innate immune priming in the Lophotrochozoan Pacific

oyster, Crassostrea gigas. Sci. Rep. 7, 13143. doi:10.1038/s41598-017-13564-0

Lancz, G.J., 1980. Physical integrity of herpes simplex virus following thermal inactivation.

207

Arch. Virol. 64, 375–381. doi:10.1007/BF01320622

Le Deuff, R.M., Renault, T., 1999. Purification and partial genome characterization of a

herpes-like virus infecting the Japanese oyster, Crassostrea gigas. J. Gen. Virol. 80,

1317–1322.

Le Morvan, C., Troutaud, D., Deschaux, P., 1998. Differential effects of temperature on

specific and nonspecific immune defences in fish. J. Exp. Biol. 201, 165–168.

Ling, S.-W., Mumaw, L., 1977. Aquaculture in Southeast Asia : a historical overview.

University of Washington Press, Seattle.

Lipp, E.K., Huq, A., Colwell, R.R., 2002. Effects of global climate on infectious disease: the

cholera model. Clin. Microbiol. Rev. 15, 757 LP – 770. doi:10.1128/CMR.15.4.757-

770.2002

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-time

quantitative PCR and the 2-ΔΔCT method. Methods 25, 402–408.

doi:10.1006/meth.2001.1262

Lynch, S.A., Carlsson, J., Reilly, A.O., Cotter, E., Culloty, S.C., 2012. A previously

undescribed Ostreid herpesvirus 1 (OsHV-1) genotype detected in the Pacific oyster,

Crassostrea gigas, in Ireland. Parasitology 139, 1526–1532. doi:DOI:

10.1017/S0031182012000881

Makrinos, D.L., Bowden, T.J., 2016. Natural environmental impacts on teleost immune

function. Fish Shellfish Immunol. 53, 50–57. doi:10.1016/j.fsi.2016.03.008

Marcogliese, D., 2001. Implications of climate change for parasitism of animals in the aquatic

208

environment. Can. J. Zool. 79, 1331–1352. doi:10.1139/cjz-79-8-1331

Martenot, C., Denechère, L., Hubert, P., Metayer, L., Oden, E., Trancart, S., Travaillé, E.,

Houssin, M., 2015a. Virulence of Ostreid herpesvirus 1 μVar in sea water at 16°C and

25°C. Aquaculture 439, 1–6. doi:10.1016/j.aquaculture.2015.01.012

Martenot, C., Fourour, S., Oden, E., Jouaux, A., Travaillé, E., Malas, J.P., Houssin, M., 2012.

Detection of the OsHV-1 μVar in the Pacific oyster Crassostrea gigas before 2008 in

France and description of two new microvariants of the Ostreid Herpesvirus 1 (OsHV-1).

Aquaculture. doi:10.1016/j.aquaculture.2011.12.030

Martenot, C., Gervais, O., Chollet, B., Houssin, M., Renault, T., 2017. Haemocytes collected

from experimentally infected Pacific oysters, Crassostrea gigas: Detection of Ostreid

herpesvirus 1 DNA, RNA, and proteins in relation with inhibition of apoptosis. PLoS

One 12, e0177448.

Martenot, C., Lethuillier, O., Fourour, S., Oden, E., Trancart, S., Travaillé, E., Houssin, M.,

2015b. Detection of undescribed Ostreid herpesvirus 1 (OsHV-1) specimens from Pacific

oyster, Crassostrea gigas. J. Invertebr. Pathol. 132, 182–189.

doi:10.1016/j.jip.2015.10.005

Martenot, C., Oden, E., Travaillé, E., Malas, J.-P.P., Houssin, M., Travaille, E., Malas, J.-P.P.,

Houssin, M., Travaillé, E., Malas, J.-P.P., Houssin, M., 2011. Detection of different

variants of Ostreid herpesvirus 1 in the Pacific oyster, Crassostrea gigas between 2008

and 2010. Virus Res. 160, 25–31. doi:10.1016/j.virusres.2011.04.012

Martenot, C., Oden, E., Travaillé, E., Malas, J.P., Houssin, M., 2010. Comparison of two real-

time PCR methods for detection of Ostreid herpesvirus 1 in the Pacific oyster

209

Crassostrea gigas. J. Virol. Methods 170, 86–89. doi:10.1016/j.jviromet.2010.09.003

Martenot, C., Travaillé, E., Lethuillier, O., Lelong, C., Houssin, M., 2013. Genome

exploration of six variants of the Ostreid herpesvirus 1 and characterization of large

deletion in OsHV-1μVar specimens. Virus Res. 178, 462–470.

doi:10.1016/j.virusres.2013.08.006

Milutinović, B., Kurtz, J., 2016. Immune memory in invertebrates. Semin. Immunol. 28, 328–

342. doi:https://doi.org/10.1016/j.smim.2016.05.004

Minton, K., 2018. Viperin breaks viral chains. Nat. Rev. Immunol. 18, 480–481.

doi:10.1038/s41577-018-0035-1

Mletzko, A., Amtmann, A., Bergmann, S., Lee, P., Christian, J., Buchholz, R., Becker, A.,

2017. Inoculation of Cyprinid herpesvirus 3 (CyHV-3) on common carp brain cells—

influence of process parameters on virus yield. Vitr. Cell. Dev. Biol. Anim. 53, 579–585.

Moreau, P., Moreau, K., Segarra, A., Tourbiez, D., Travers, M.A., Rubinsztein, D.C., Renault,

T., 2015. Autophagy plays an important role in protecting Pacific oysters from OsHV-1

and Vibrio aestuarianus infections. Autophagy 11, 516–526.

doi:10.1080/15548627.2015.1017188

Motes, M.L., DePaola, A., Cook, D.W., Veazey, J.E., Hunsucker, J.C., Garthright, W.E.,

Blodgett, R.J., Chirtel, S.J., 1998. Influence of water temperature and salinity on Vibrio

vulnificus in Northern Gulf and Atlantic Coast oysters (Crassostrea virginica). Appl.

Environ. Microbiol. 64, 1459–1465.

Müller, K., Wickham, H., 2017. tibble: Simple Data Frames.

210

Murray, A.G., Peeler, E.J., 2005. A framework for understanding the potential for emerging

diseases in aquaculture. Prev. Vet. Med. 67, 223–235.

doi:https://doi.org/10.1016/j.prevetmed.2004.10.012

Namikoshi, A., Wu, J.L., Yamashita, T., Nishizawa, T., Nishioka, T., Arimoto, M., Muroga,

K., 2004. Vaccination trials with Penaeus japonicus to induce resistance to white spot

syndrome virus. Aquaculture 229, 25–35. doi:10.1016/S0044-8486(03)00363-6

National Health and Medical Research Council, 2013. Australian Code of practice for the care

and use of animals for scientific purposes, 8th ed. URL https://www.nhmrc.gov.au/about-

us/publications/australian-code-care-and-use-animals-scientific-purposes

Nell, J.A., 2002. Farming triploid oysters. Aquaculture 210, 69–88. doi:10.1016/S0044-

8486(01)00861-4

Normand, J., Blin, J.L., Jouaux, A., 2014a. Rearing practices identified as risk factors for

Ostreid herpesvirus 1 (OsHV-1) infection in Pacific oyster Crassostrea gigas spat. Dis.

Aquat. Organ. 110, 201–211. doi:10.3354/dao02756

Normand, J., Li, R., Quillien, V., Nicolas, J.L., Boudry, P., Pernet, F., Huvet, A., 2014b.

Contrasted survival under field or controlled conditions displays associations between

mRNA levels of candidate genes and response to OsHV-1 infection in the Pacific oyster

Crassostrea gigas. Mar. Genomics 15, 95–102. doi:10.1016/j.margen.2014.02.002

NSW DPI, 2014. Procedure Biosecurity - Pacific Oyster Mortality Syndrome incursion

response. URL https://www.dpi.nsw.gov.au/__data/assets/pdf_file/0003/637680/poms-

incursion-response-policy.pdf

211

Oden, E., Martenot, C., Berthaux, M., Travaillé, E., Malas, J.P., Houssin, M., 2011.

Quantification of Ostreid herpesvirus 1 (OsHV-1) in Crassostrea gigas by real-time

PCR: Determination of a viral load threshold to prevent summer mortalities. Aquaculture

317, 27–31. doi:10.1016/j.aquaculture.2011.04.001

Ogburn, D., White, I., Mcphee, D., 2007. The disappearance of oyster reefs from Eastern

Australian Estuaries - Impact of colonial settlement or mudworm invasion? Coast.

Manag. 35, 271–287. doi:10.1080/08920750601169618

Oh, S.Y., Oh, M.J., Nishizawa, T., 2014. Potential for a live red seabream iridovirus (RSIV)

vaccine in rock bream Oplegnathus fasciatus at a low rearing temperature. Vaccine 32,

363–368. doi:10.1016/j.vaccine.2013.11.030

OIE, 2019. Aquatic Animal Health Code, 22nd ed. URL https://www.oie.int/standard-

setting/aquatic-code/access-online/

OIE, 2016. Infection with Ostreid herpesvirus 1 microvariant. Man. Diagnostic Tests Aquat.

Anim. 1–14.

Olejniczak, S.H., La Rocca, G., Gruber, J.J., Thompson, C.B., 2013. Long-lived microRNA-

Argonaute complexes in quiescent cells can be activated to regulate mitogenic responses.

Proc. Natl. Acad. Sci. U. S. A. 110, 157–62. doi:10.1073/pnas.1219958110

Owens, L., Malham, S., 2015. Review of the RNA interference pathway in molluscs including

some possibilities for use in bivalves in aquaculture. J. Mar. Sci. Eng. 3, 87–99.

doi:10.3390/jmse3010087

Pankhurst, N.W., King, H.R., 2010. Temperature and salmonid reproduction: Implications for

212

aquaculture. J. Fish Biol. 76, 69–85. doi:10.1111/j.1095-8649.2009.02484.x

Pathirana, E., Fuhrmann, M., Whittington, R., Hick, P., 2019. Influence of environment on the

pathogenesis of Ostreid herpesvirus-1 (OsHV-1) infections in Pacific oysters

(Crassostrea gigas) through differential microbiome responses. Heliyon 5, e02101.

doi:10.1016/j.heliyon.2019.e02101

Paul-Pont, I., Dhand, N.K., Whittington, R.J., 2013a. Influence of husbandry practices on

OsHV-1 associated mortality of Pacific oysters Crassostrea gigas. Aquaculture 412–413,

202–214. doi:10.1016/j.aquaculture.2013.07.038

Paul-Pont, I., Dhand, N.K., Whittington, R.J., 2013b. Spatial distribution of mortality in

Pacific oysters Crassostrea gigas: Reflection on mechanisms of OsHV-1 transmission.

Dis. Aquat. Organ. 105, 127–138. doi:10.3354/dao02615

Paul-Pont, I., Evans, O., Dhand, N.K., Rubio, A., Coad, P., Whittington, R.J., 2014.

Descriptive epidemiology of mass mortality due to Ostreid herpesvirus-1 (OsHV-1) in

commercially farmed Pacific oysters (Crassostrea gigas) in the Hawkesbury River

estuary, Australia. Aquaculture 422–423, 146–159.

doi:10.1016/j.aquaculture.2013.12.009

Paul-Pont, I., Evans, O., Dhand, N.K., Whittington, R.J., 2015. Experimental infections of

Pacific oyster Crassostrea gigas using the Australian Ostreid herpesvirus-1 (OsHV-1)

μVar strain. Dis. Aquat. Organ. 113, 137–147. doi:10.3354/dao02826

Pauletto, M., Segarra, A., Montagnani, C., Quillien, V., Faury, N., Le Grand, J., Miner, P.,

Petton, B., Labreuche, Y., Fleury, E., Fabioux, C., Bargelloni, L., Renault, T., Huvet, A.,

2017. Long dsRNAs promote an anti-viral response in Pacific oyster hampering Ostreid

213

herpesvirus 1 replication. J. Exp. Biol. 220, 3671 LP – 3685. doi:10.1242/jeb.156299

Peeler, E.J., Allan Reese, R., Cheslett, D.L., Geoghegan, F., Power, A., Thrush, M.A., 2012.

Investigation of mortality in Pacific oysters associated with Ostreid herpesvirus-1 μVar

in the Republic of Ireland in 2009. Prev. Vet. Med. 105, 136–143.

doi:10.1016/j.prevetmed.2012.02.001

Perkins, F.O., Wolf, P.H., 1976. Fine structure of Marteilia sydneyi sp. n.-haplosporidan

pathogen of Australian oysters. J. Parasitol. 62, 528–538.

Pernet, F., Barret, J., Le Gall, P., Corporeau, C., Dégremont, L., Lagarde, F., Pépin, J.F.,

Keck, N., 2012. Mass mortalities of Pacific oysters Crassostrea gigas reflect infectious

diseases and vary with farming practices in the Mediterranean Thau lagoon, France.

Aquac. Environ. Interact. 2, 215–237. doi:10.3354/aei00041

Pernet, F., Lagarde, F., Gall, P. Le, D’Orbcastel, E.R., 2014a. Associations between farming

practices and disease mortality of Pacific oyster Crassostrea gigas in a Mediterranean

lagoon. Aquac. Environ. Interact. 5, 99–106. doi:10.3354/aei00096

Pernet, F., Lagarde, F., Jeanné, N., Daigle, G., Barret, J., Le Gall, P., Quere, C., D’orbcastel,

E.R., 2014b. Spatial and temporal dynamics of mass mortalities in oysters is influenced

by energetic reserves and food quality. PLoS One 9. doi:10.1371/journal.pone.0088469

Pernet, F., Lupo, C., Bacher, C., Whittington, R.J., 2016. Infectious diseases in oyster

aquaculture require a new integrated approach 371. Phil. Trans. R. Soc. B 371:

20150213. doi:10.1098/rstb.2015.0213

Pernet, F., Tamayo, D., Fuhrmann, M., Petton, B., 2019. Deciphering the effect of food

214

availability, growth and host condition on disease susceptibility in a marine invertebrate

1. J. Experimental Biol. 210534. doi:10.1242/jeb.210534

Pernet, F., Tamayo, D., Petton, B., 2015. Influence of low temperatures on the survival of the

Pacific oyster (Crassostrea gigas) infected with Ostreid herpesvirus type 1. Aquaculture

445, 57–62. doi:10.1016/j.aquaculture.2015.04.010

Petton, B., Boudry, P., Alunno-Bruscia, M., Pernet, F., 2015a. Factors influencing disease-

induced mortality of Pacific oysters Crassostrea gigas. Aquac. Environ. Interact. 6, 205–

222. doi:10.3354/aei00125

Petton, B., Bruto, M., James, A., Labreuche, Y., Alunno-Bruscia, M., Le Roux, F., 2015b.

Crassostrea gigas mortality in France: The usual suspect, a herpesvirus, may not be the

killer in this polymicrobial opportunistic disease. Front. Microbiol. 6.

doi:10.3389/fmicb.2015.00686

Petton, B., Pernet, F., Robert, R., Boudry, P., 2013. Temperature influence on pathogen

transmission and subsequent mortalities in juvenile Pacific oysters Crassostrea gigas.

Aquac. Environ. Interact. 3, 257–273. doi:10.3354/aei00070

Picot, S., Morga, B., Faury, N., Chollet, B., Dégremont, L., Travers, M.A., Renault, T., Arzul,

I., 2019. A study of autophagy in hemocytes of the Pacific oyster, Crassostrea gigas.

Autophagy 15, 1801–1809. doi:10.1080/15548627.2019.1596490

Pollard, D.., Hutchings, P.A., 1990. A review of exotic marine organisms introduced to the

Australian region. 2. Invertebrates and Algae. Asian Fish. Sci. 3, 223–250.

R Core Team, 2017. R: A Language and Environment for Statistical Computing.

215

Raadsma, H.W., Jonas, E., McGill, D., Hobbs, M., Lam, M.K., Thomson, P.C., 2009.

Mapping quantitative trait loci (QTL) in sheep. II. Meta-assembly and identification of

novel QTL for milk production traits in sheep. Genet. Sel. Evol. 41, 45.

doi:10.1186/1297-9686-41-45

Raftos, D.A., Kuchel, R., Aladaileh, S., Butt, D., 2014. Infectious microbial diseases and host

defense responses in Sydney rock oysters. Front. Microbiol. 5, 1–12.

doi:10.3389/fmicb.2014.00135

Rajagopal, S., Van Der Velde, G., Jansen, J., Van Der Gaag, M., Atsma, G., Janssen-

Mommen, J.P.M., Polman, H., Jenner, H.A., 2005. Thermal tolerance of the invasive

oyster Crassostrea gigas: Feasibility of heat treatment as an antifouling option. Water

Res. 39, 4335–4342. doi:10.1016/j.watres.2005.08.021

Read, G.B., 2010. Comparison and history of Polydora websteri and P. haswelli (Polychaeta:

Spionidae) as mud-blister worms in New Zealand shellfish. New Zeal. J. Mar. Freshw.

Res. 44, 83–100. doi:10.1080/00288330.2010.482969

Ren, W., Chen, H., Renault, T., Cai, Y., Bai, C., Wang, C., Huang, J., 2013. Complete

genome sequence of acute viral necrosis virus associated with massive mortality

outbreaks in the Chinese scallop, Chlamys farreri. Virol. J. 10, 110. doi:10.1186/1743-

422X-10-110

Renault, T., Bouquet, A.L., Maurice, J.-T.T., Lupo, C., Blachier, P., 2014. Ostreid herpesvirus

1 infection among Pacific oyster (Crassostrea gigas) Spat: Relevance of water

temperature to virus replication and circulation prior to the onset of mortality. Appl.

Environ. Microbiol. 80, 5419–5426. doi:10.1128/AEM.00484-14

216

Renault, T., Faury, N., Barbosa-Solomieu, V., Moreau, K., 2011. Suppression substractive

hybridisation (SSH) and real time PCR reveal differential gene expression in the Pacific

cupped oyster, Crassostrea gigas, challenged with Ostreid herpesvirus 1. Dev. Comp.

Immunol. 35, 725–735. doi:10.1016/j.dci.2011.02.004

Renault, T., Le Deuff, R.-M., Cochennec, N., Maffart, P., 1994. Herpesviruses associated with

mortalities among Pacific oyster, Crassostrea gigas, in France-Comparative study. Rev.

Med. Vet. (Toulouse).

Renault, T., Moreau, P., Faury, N., Pepin, J.-F., Segarra, A., Webb, S., 2012. Analysis of

clinical Ostreid herpesvirus 1 (Malacoherpesviridae) specimens by sequencing amplified

fragments from three virus genome areas. J. Virol. 86, 5942–5947.

doi:10.1128/JVI.06534-11

Roque, A., Carrasco, N., Andree, K.B., Lacuesta, B., Elandaloussi, L., Gairin, I., Rodgers,

C.J., Furones, M.D., 2012. First report of OsHV-1 microvar in Pacific oyster

(Crassostrea gigas) cultured in Spain. Aquaculture 324–325, 303–306.

doi:10.1016/j.aquaculture.2011.10.018

Rosa, R., Marques, A., Nunes, M.L., 2012. Impact of climate change in Mediterranean

aquaculture. Rev. Aquac. 4, 163–177. doi:10.1111/j.1753-5131.2012.01071.x

Rosani, U., Gerdol, M., 2017. A bioinformatics approach reveals seven nearly-complete

RNA-virus genomes in bivalve RNA-seq data. Virus Res. 239, 33–42.

doi:10.1016/j.virusres.2016.10.009

Rosani, U., Pallavicini, A., Venier, P., 2016. The miRNA biogenesis in marine bivalves. PeerJ

4, e1763. doi:10.7717/peerj.1763

217

Rosani, U., Varotto, L., Domeneghetti, S., Arcangeli, G., Pallavicini, A., Venier, P., 2015.

Dual analysis of host and pathogen transcriptomes in Ostreid herpesvirus 1-positive

Crassostrea gigas. Environ. Microbiol. 17, 4200–4212. doi:10.1111/1462-2920.12706

Rosani, U., Young, T., Bai, C.M., Alfaro, A.C., Venier, P., 2019. Dual analysis of virus-host

interactions: The case of Ostreid herpesvirus 1 and the cupped oyster Crassostrea gigas.

Evol. Bioinforma. 15. doi:10.1177/1176934319831305

Rowley, A.F., Pope, E.C., 2012. Vaccines and crustacean aquaculture-A mechanistic

exploration. Aquaculture 334–337, 1–11. doi:10.1016/j.aquaculture.2011.12.011

Sarà, G., Mazzola, A., 1997. Effects of trophic and environmental conditions on the growth of

Crassostrea gigas in culture. Aquaculture 153, 81–91. doi:10.1016/S0044-

8486(97)00009-4

Sato-Okoshi, W., Abe, H., 2013. Morphology and molecular analysis of the 18S rRNA gene

of oyster shell borers, Polydora species (Polychaeta: Spionidae), from Japan and

Australia. J. Mar. Biol. Assoc. United Kingdom 93, 1279–1286.

doi:10.1017/S002531541200152X

Sauvage, C., Pépin, J.F., Lapègue, S., Boudry, P., Renault, T., 2009. Ostreid herpesvirus 1

infection in families of the Pacific oyster, Crassostrea gigas, during a summer mortality

outbreak: Differences in viral DNA detection and quantification using real-time PCR.

Virus Res. 142, 181–187. doi:10.1016/j.virusres.2009.02.013

Schikorski, D., Faury, N., Pepin, J.F., Saulnier, D., Tourbiez, D., Renault, T., 2011a.

Experimental Ostreid herpesvirus 1 infection of the Pacific oyster Crassostrea gigas:

Kinetics of virus DNA detection by q-PCR in seawater and in oyster samples. Virus Res.

218

155, 28–34. doi:10.1016/j.virusres.2010.07.031

Schikorski, D., Renault, T., Saulnier, D., Faury, N., Moreau, P., Pépin, J., 2011b.

Experimental infection of Pacific oyster Crassostrea gigas spat by Ostreid herpesvirus 1:

Demonstration of oyster spat susceptibility. Vet. Res. 42, 27. doi:10.1186/1297-9716-42-

Schulenburg, H., Boehnisch, C., Michiels, N.K., 2007. How do invertebrates generate a highly

specific innate immune response? Mol. Immunol. 44, 3338–3344.

doi:10.1016/j.molimm.2007.02.019

Segarra, A., Baillon, L., Faury, N., Tourbiez, D., Renault, T., 2016. Detection and distribution

of Ostreid herpesvirus 1 in experimentally infected Pacific oyster spat. J. Invertebr.

Pathol. 133, 59–65.

Segarra, A., Baillon, L., Tourbiez, D., Benabdelmouna, A., Faury, N., Bourgougnon, N.,

Renault, T., 2014a. Ostreid herpesvirus type 1 replication and host response in adult

Pacific oysters, Crassostrea gigas. Vet. Res. 45, 30–32. doi:10.1186/s13567-014-0103-x

Segarra, A., Faury, N., Pépin, J.F., Renault, T., 2014b. Transcriptomic study of 39 Ostreid

herpesvirus 1 genes during an experimental infection. J. Invertebr. Pathol. 119, 5–11.

doi:10.1016/j.jip.2014.03.002

Segarra, A., Mauduit, F., Faury, N., Trancart, S., Dégremont, L., Tourbiez, D., Haffner, P.,

Barbosa-Solomieu, V., Pépin, J.-F., Travers, M.-A., Renault, T., 2014c. Dual

transcriptomics of virus-host interactions: comparing two Pacific oyster families

presenting contrasted susceptibility to Ostreid herpesvirus 1. BMC Genomics 15, 580.

doi:10.1186/1471-2164-15-580

219

Segarra, A., Pépin, J.F., Arzul, I., Morga, B., Faury, N., Renault, T., 2010. Detection and

description of a particular Ostreid herpesvirus 1 genotype associated with massive

mortality outbreaks of Pacific oysters, Crassostrea gigas, in France in 2008. Virus Res.

153, 92–99. doi:10.1016/j.virusres.2010.07.011

Siddell, S.G., Davison, A.J., Lefkowitz, E.J., Sabanadzovic, S., Simmonds, P., Smith, D.B.,

Orton, R.J., Harrach, B., Knowles, N.J., Kropinski, A.M., Kuhn, J.H., Nibert, M.L.,

Zerbini, F.M., 2011. ICTV Report on Virus Taxonomy Classification and Nomenclature

of Viruses. URL https://talk.ictvonline.org/ictv-reports/.

Stevens, J.G., 1994. Overview of herpesvirus latency. Semin. Virol. 5, 191–196.

doi:https://doi.org/10.1006/smvy.1994.1020

Strodthoff, D., Ma, Z., Wirström, T., Strawbridge, R.J., Ketelhuth, D.F.J., Engel, D., Clarke,

R., Falkmer, S., Hamsten, A., Hansson, G.K., Björklund, A., Lundberg, A.M., 2015.

Toll-like receptor 3 influences glucose homeostasis and β-cell insulin secretion. Diabetes

64, 3425–3438. doi:10.2337/db14-0838

Tamayo, D., Corporeau, C., Petton, B., Quere, C., Pernet, F., 2014. Physiological changes in

Pacific oyster Crassostrea gigas exposed to the herpesvirus OsHV-1μVar. Aquaculture

432, 304–310. doi:10.1016/j.aquaculture.2014.05.023

Tang, X., Huang, B., Zhang, L., Li, L., Zhang, G., 2017. Molecular characterization of Pacific

oyster (Crassostrea gigas) IRAK4 gene and its role in MyD88-dependent pathway. Dev.

Comp. Immunol. 72, 21–29. doi:10.1016/j.dci.2017.02.004

The University of Sydney, 2019. Biological safety [WWW Document]. Biol. Saf. URL

https://intranet.sydney.edu.au/employment/safety-wellbeing/standards-

220

guidelines/biological-safety.html (accessed 8.20.19).

Therneau, T.M., 2015. A Package for Survival Analysis in S.

Therneau, T.M., Grambsch, P.M., 2000. Modeling Survival Data: Extending the Cox Model.

Springer, New York.

Thrusfield, M. V., 2007. Veterinary epidemiology, 3rd ed. Blackwell, Oxford.

Travers, M.A., Boettcher Miller, K., Roque, A., Friedman, C.S., 2015. Bacterial diseases in

marine bivalves. J. Invertebr. Pathol. 131, 11–31. doi:10.1016/j.jip.2015.07.010

Ugalde, S.C., Preston, J., Ogier, E., Crawford, C., 2018. Analysis of farm management

strategies following herpesvirus (OsHV-1) disease outbreaks in Pacific oysters in

Tasmania, Australia. Aquaculture 495, 179–186. doi:10.1016/j.aquaculture.2018.05.019

Vezzulli, L., Colwell, R.R., Pruzzo, C., 2013. Ocean warming and spread of pathogenic

Vibrios in the aquatic environment. Microb. Ecol. 65, 817–825. doi:10.1007/s00248-012-

0163-2

Vezzulli, L., Pezzati, E., Brettar, I., Höfle, M., Pruzzo, C., 2015. Effects of global warming on

Vibrio ecology. Microbiol. Spectr. 3.

Wang, L., Song, X., Song, L., 2018. The oyster immunity. Dev. Comp. Immunol. 80, 99–118.

doi:10.1016/j.dci.2017.05.025

Wang, L.F., Crameri, G., 2014. Emerging zoonotic viral diseases. OIE Rev. Sci. Tech. 33,

569–581. doi:10.20506/rst.33.2.2311

Ward, J.R., Lafferty, K.D., 2004. The elusive baseline of marine disease: Are diseases in

ocean ecosystems increasing? PLoS Biol. 2, 542–547. doi:10.1371/journal.pbio.0020120

221

Webb, S.C., Fidler, A., Renault, T., 2007. Primers for PCR-based detection of Ostreid

herpesvirus-1 (OsHV-1): Application in a survey of New Zealand molluscs. Aquaculture

272, 126–139. doi:10.1016/j.aquaculture.2007.07.224

Weber, F., Wagner, V., Rasmussen, S.B., Hartmann, R., Paludan, S.R., 2006. Double-stranded

RNA is produced by positive-strand RNA viruses and DNA viruses but not in detectable

amountsby negative-strand RNA viruses. J. of Virol. 80, 5059–5064.

doi:10.1128/JVI.80.10.5059

Weiliang, Q., Chavarro, J., Lazarus, R., Rosner, B., Ma, J., 2018. powerSurvEpi: Power and

Sample Size Calculation for Survival Analysis of Epidemiological Studies. R package

version 0.1.0.

Whittington, R., Reddacliff, G., 1995. Influence of environmental temperature on

experimental infection of redfin perch (Perca fluviatilis) and rainbow trout

(Oncorhynchus mykiss) with epizootic haematopoietic necrosis virus, an Australian

iridovirus. Aust. Vet. J. 72, 421–424. doi:10.1111/j.1751-0813.1995.tb06193.x

Whittington, R.J., Dhand, N.K., Evans, O., Paul-Pont, I., 2015a. Further observations on the

influence of husbandry practices on OsHV-1 μVar mortality in Pacific oysters

Crassostrea gigas: Age, cultivation structures and growing height. Aquaculture 438, 82–

97. doi:10.1016/j.aquaculture.2014.12.040

Whittington, R.J., Hick, P.M., Evans, O., Rubio, A., Alford, B., Dhand, N., Paul-Pont, I.,

2015b. Protection of Pacific oyster (Crassostrea gigas) spat from mortality due to

Ostreid herpesvirus 1 (OsHV-1 μVar) using simple treatments of incoming seawater in

land-based upwellers. Aquaculture 437, 10–20. doi:10.1016/j.aquaculture.2014.11.016

222

Whittington, R.J., Liu, O., Hick, P.M., Dhand, N., Rubio, A., 2019. Long-term temporal and

spatial patterns of Ostreid herpesvirus 1 (OsHV-1) infection and mortality in sentinel

Pacific oyster spat (Crassostrea gigas) inform farm management. Aquaculture 513,

734395. doi:https://doi.org/10.1016/j.aquaculture.2019.734395

Whittington, R.J., Paul-Pont, I., Evans, O., Hick, P., Dhand, N.K., 2018. Counting the dead to

determine the source and transmission of the marine herpesvirus OsHV-1 in Crassostrea

gigas. Vet. Res. 49, 34. doi:10.1186/s13567-018-0529-7

Wickham, H., 2017. forcats: Tools for working with categorical variables (Factors).

Wickham, H., 2011. The split-apply-combine strategy for data analysis. J. Stat. Softw. 40, 1–

29.

Wickham, H., 2009. ggplot2: Elegant graphics for data analysis. Springer-Verlag New York.

Wickham, H., Francois, R., Henry, L., Müller, K., 2017a. dplyr: A Grammar of Data

Manipulation.

Wickham, H., Henry, L., 2017. tidyr: Easily Tidy Data with “spread()” and “gather()”

Functions.

Wickham, H., Hester, J., Francois, R., 2017b. readr: Read rectangular text data.

Wolf, P., 1972. Occurrence of a haplosporidian in Sydney rock oysters (Crossostrea

commercialis) from Moreton Bay, Queensland, Australia. J. Invertebr. Pathol. 19, 416–

417.

World Bank, 2014. Reducing Disease Risk in Aquaculture. The World Bank, Washington.

Xia, J., Bai, C., Wang, C., Song, X., Huang, J., 2015. Complete genome sequence of Ostreid

223

herpesvirus-1 associated with mortalities of Scapharca broughtonii broodstocks. Virol. J.

12, 110. doi:10.1186/s12985-015-0334-0

Xu, J.-R., Bently, J., Beck, L., Reed, A., Miller-Morgan, T., Heidel, J.R., Kent, M.L., Rockey,

D.D., Jin, L., 2013. Analysis of koi herpesvirus latency in wild common carp and

ornamental koi in Oregon, USA. J. Virol. Methods 187, 372–379.

doi:https://doi.org/10.1016/j.jviromet.2012.11.015

Yijie Ma, He, B., 2014. Recognition of herpes simplex viruses: toll-like receptors and beyond.

J Mol Biol. 426, 1133–1147. doi: 10.1016/j.jmb.2013.11.012

Yuan, J.S., Reed, A., Chen, F., Stewart, C.N., 2006. Statistical analysis of real-time PCR data.

BMC Bioinformatics 7, 1–12. doi:10.1186/1471-2105-7-85

Zrnčić, S., Oraić, D., Mihaljević, Ž., Zanella, D., 2007. Impact of varying cultivation depths

on growth rate and survival of the European flat oyster Ostrea Edulis, L. Aquac. Res. 38,

1305–1310. doi:10.1111/j.1365-2109.2007.01804.x

224