THE ROLE of HERPESVIRUSES in MARINE TURTLE DISEASES By

THE ROLE OF HERPESVIRUSES IN MARINE TURTLE DISEASES

SADIE SHEA COBERLEY

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2002

Sadie Shea Coberley

For the turtles, and Carter and my family for encouraging me to pursue what I love.

ACKNOWLEDGEMENTS

I would like to thank my mentor, Dr. Paul Klein, for sharing his knowledge and for all of his encouragement and patience throughout my graduate education. He has been a true mentor in every sense of the word, and has done everything possible to prepare me for not only my scientific future, but phases of life outside of the laboratory as well. I would also like to thank my co-mentor, Dr. Rich Condit, first for seeing graduate student potential, and then for taking me in and helping to provide the necessary tools and expertise to cultivate it. In addition, I am indebted to Dr. Larry Herbst, who was not only my predecessor but a pioneer in FP research. His insight into studying such a complex problem has been invaluable. I am grateful for the critical analysis and raised eyebrow of

Dr. Daniel Brown and for his assistance with trouble-shooting experiments, evaluating data, and preparing manuscripts. I am also appreciative of the assistance of Dr. Elliott

Jacobson for including me in many discussions, necropsies, and analyses of marine turtles with interesting clinical signs of disease, and for sharing his vast knowledge of reptile diseases. I would like to thank Dr. Maureen Goodenow for her valuable input at committee meetings and for her assistance specifically with phylogenetic analyses. I would like to extend my gratitude to Dr. David Bloom for agreeing to join my committee so late in the project and for contributing his fresh enthusiasm and expertise in herpesviral research. I would also like to thank Dr. Wayne McCormack for his input and assistance during the final examination. It has been a blessing to have such a functional

committee and to have such an enjoyable group of researchers to help me through my scientific struggles.

I would like to extend a heartfelt thank you to Diane Duke for her constant moral support, friendship, and laboratory rescues. I would also like to extend a special thank you to all of the members of the Condit laboratory for their invaluable technical guidance, for letting me constantly interrupt them with questions, and for listening to turtle research. This has included Drs. Cari Lackner, Cindy Prins, Susan D’Costa, and

Don Latner, fellow IDP graduate student Steve Cresawn, and Jackie Fried. I would also like to say thank you to Dr. Nissin Moussatche for all of his help and for being so much fun to harass, and to Dr. Ed Niles for his timely visit to the Condit lab and for his assistance with bacterial expression that led to the expression of the LETV proteins. I would also like to thank Francesco Origgi and Joyce Merritt for their encouragement and support in the laboratory and their friendship.

I would like to thank the members of the University of Florida Interdisciplinary Center for Biotechnology Research (ICBR) Hybridoma core laboratory (Gainesville, FL, USA),

Linda Green, Shadi Bootorabi, Jamie Kelso, and Scherwin Henry, for all of their valuable help through the years. I would like to extend my gratitude to Marjorie Chow of the

ICBR Molecular Biomarker core laboratory for teaching me the technique of two- dimensional gel electrophoresis and for her encouragement and friendship. I am grateful for the assistance of Drs. Regina Shaw and Bill Farmerie of the ICBR Genomic core laboratory for construction of the LETV genomic library. Finally, I would like to thank

Susanna Lamers of Gene Genie Co. (Thibdaux, LA, USA) for her assistance with phylogenetic analysis and the construction of the phylogenetic trees.

I am indebted to all of our collaborators who have been our connection to the sea turtles. I am grateful for all of the hard work and dedication of Richie Moretti, Sue

Schaf, and Corinne Rose at the Turtle Hospital (Marathon, FL, USA). I would like to thank all of the members of the University of Central Florida Marine Turtle Research

Program headed by Dr. Llewellyn Ehrhart. I would especially like to thank Dean Bagley for all the plasma samples she has collected that we appreciate and rely on daily. I would like to thank Dr. Jeanette Wyneken for the crash course in sea turtle biology and having faith in a lab rat. It is only through these broad collaborations that it has been possible to address such complex problems. I would like to extend my gratitude to all of the members of the turtle community that I have had the privilege to work with and for their constant support and encouragement. Their dedication to the survival of marine turtles is inspiring.

Finally, I am grateful for the continued support of the U.S. Fish and Wildlife Service and the assistance of Sandy MacPherson. This research was funded by research grants

RWO 180, 194, and 213 from the U.S. Fish and Wildlife Service, Department of the

Interior, administered by the Cooperative Fish and Wildlife Unit, University of Florida,

Gainesville, FL, USA.

TABLE OF CONTENTS page ACKNOWLEDGEMENTS ...... iv

LIST OF TABLES ...... xii

LIST OF FIGURES ...... xiii

ABSTRACT...... xvi

CHAPTER

1 INTRODUCTION ...... 1

Microorganisms Associated with Diseases of Marine Turtles ...... 1 Herpesvirus Infections of Marine Turtles ...... 2 GPD-associated herpesvirus ...... 3 LETD-associated herpesvirus ...... 4 Fibropapillomatosis-associated herpesvirus ...... 5 The Herpesvirus Family...... 6 Lytic Infection...... 8 Human herpesviruses ...... 9 Chelonian herpesviruses ...... 10 Oncogenic Herpesviruses...... 11 Human simplex virus ...... 13 Epstein-Barr virus ...... 13 Human herpesvirus 6 ...... 15 Kaposi Sarcoma-associated herpesvirus ...... 16 Chelonian herpesviruses and tumors in reptiles...... 19 Methods for Linking Viruses with Disease Etiology...... 20 Henle-Koch Postulates ...... 20 Rivers’ Review...... 21 Hill’s Criteria for Causation...... 22 Evans Unified Concept...... 22 Strategies To Demonstrate Association Between Herpesviral Infection and Onset of Disease by Fulfilling Hill’s Criteria...... 22 Strength of association...... 23 Consistency of findings...... 24 Biological gradient ...... 25 Specificity of association...... 27

vii

Temporal association...... 28 Coherence to the biology of the disease...... 29 Experimental evidence...... 30 Biological plausibility and analogy ...... 32 Strategies to Investigate the Role of Herpesviruses in Marine Turtle Diseases..... 34

2 STUDIES ON THE REPLICATION OF MARINE TURTLE HERPESVIRUSES.....36

Introduction...... 36 Materials and Methods...... 37 FP Filtrate Preparation...... 37 Established Tumor Cells ...... 38 Cells and Tissue Culture ...... 39 Cultivated Chelonian Herpesviruses ...... 39 Coculture Experiments...... 39 Embryonated Eggs ...... 40 Collaborative Study with National Wildlife Health Center...... 41 Transfection Experiments ...... 42 Sample Preparation and PCR Analysis ...... 42 Results...... 43 Cultivation of the FP-associated Herpesvirus In Vitro...... 43 Established Green Turtle Tumor Cells...... 43 Embryonated Eggs Inoculated with Marine Turtle Herpesviruses ...... 44 Studies in Collaboration with the National Wildlife Health Center...... 44 Transfection of Tissue Culture Cells with Herpesviral DNA...... 45 Discussion...... 46

3 EFFORTS TO IDENTIFY SURROGATE HERPESVIRAL ANTIGENS THAT ARE CROSS-REACTIVE WITH THE FIBROPAPILLOMATOSIS-ASSOCIATED HERPESVIRUS ...... 56

Introduction...... 56 Materials and Methods...... 57 Green Turtle Plasma Samples...... 57 Human Simplex Virus ELISA Assay...... 58 Statistical Analyses ...... 60 Immunohistochemistry with LETV-infected Cells...... 60 LETV ELISA Assay...... 62 PCR Development ...... 63 Results...... 65 Antigenicity of Human Simplex Virus Glycoproteins...... 65 Cross-reactivity with LETV-infected Cells...... 67 PCR Development for Herpesvirus Glycoproteins ...... 68 Discussion...... 69

viii

4 PLAQUE-PURIFICATION, GROWTH CHARACTERISTICS, AND ENVIRONMENTAL STABILITY OF THE LUNG-EYE-TRACHEA DISEASE- ASSOCIATED HERPESVIRUS ...... 84

Introduction...... 84 Materials and Methods...... 85 Cultivation and Host Range of LETV...... 85 Plaque Purification of LETV...... 86 Growth Characteristics of LETV ...... 87 Exposure of LETV to Seawater ...... 88 Titration of viral infectivity on TH-1 cells...... 89 Electron microscopy...... 89 PCR analysis of LETV after dialysis ...... 89 Results...... 90 Cultivation and Host Range of LETV...... 90 LETV Plaque Purification Scheme and Lineage of Clones ...... 91 Growth Characteristics of LETV ...... 91 Infectivity of LETV after Exposure to Artificial and Natural Seawater...... 92 Negative stain electron microscopy...... 92 PCR analysis of LETV exposed to artificial seawater or HBSS ...... 93 Discussion...... 93

5 DEVELOPMENT AND APPLICATION OF SEROLOGICAL ASSAYS FOR DETECTING EXPOSURE OF FLORIDA MARINE TURTLES TO A DISEASE- ASSOCIATED HERPESVIRUS ...... 108

Introduction...... 108 Materials and Methods...... 109 Cultured Cells...... 109 LETV Antigen Preparation...... 109 Development of Control LETV Antisera...... 110 Development of Anti-LETV Monoclonals ...... 110 Green Turtle Plasma Samples...... 111 Development of LETV ELISA...... 113 Statistical Analyses ...... 115 Immunoblotting Protocol...... 115 Immunohistochemistry...... 116 Results...... 117 Detection of Seroconversion by ELISA...... 117 Development of LETV Monoclonals...... 118 Survey of Captive-reared Green Turtles for Antibodies to LETV...... 118 Survey of Previously Evaluated Wild Green Turtles for Antibodies to LETV ... 119 Confirmation of Anti-LETV Antibodies in Green Turtle Plasma...... 119 Survey of Green Turtle Populations at Three Florida Study Sites...... 120 Survey of Green Turtles Scored for FP Tumor Severity...... 121 Survey of Green Turtles Recaptured over Time ...... 121

Summary of ELISA Tests on Green Turtle Plasma Samples...... 122 Survey of Adult Nesting Female Turtles...... 122 Discussion...... 123

6 IDENTIFICATION AND EXPRESSION OF IMMUNOGENIC PROTEINS FOR DETECTION OF ANTIBODIES TO A DISEASE-ASSOCIATED MARINE TURTLE HERPESVIRUS ...... 143

Introduction...... 143 Materials and Methods...... 144 LETV Library Construction...... 144 Two-Dimensional Gel Electrophoresis ...... 146 Cloning of LETV UL26 and Glycoprotein B Genes...... 148 Expression and Detection of LETV Proteins ...... 149 Immunogenicity of Expressed LETV Proteins ...... 150 Phylogenetic Analyses ...... 150 Results...... 151 Partial LETV Genomic Library...... 151 Identification of the 38 kDa LETV Protein...... 151 PCR Amplification and Cloning of LETV Genes...... 152 Expression of Marine Turtle Herpesviral Proteins...... 152 Antigenicity of Expressed Herpesviral Proteins ...... 153 Phylogenetic Analysis of LETV Genes ...... 154 Discussion...... 154

7 DISCUSSION...... 171

Overview...... 171 The Etiological Role of FPHV in FP ...... 171 Limitations on Determining the Etiology of FP ...... 172 Evaluation of Antigenic Cross-reactivity among Herpesviruses ...... 173 Evaluation of Expressed Herpesviral Antigens...... 174 Immune Status of the Marine Turtle and FP...... 176 Overview...... 176 Immune function and FP...... 177 Environmental cofactors and FP...... 177 The Etiological Role of LETV in LETD ...... 179 Case Study of LETD in a Wild Loggerhead Turtle ...... 181 Molecular Virology of Marine Turtle Herpesvirus...... 184 Assignment of LETV and FPHV to the Alpha-herpesvirus Subfamily...... 187 The Role of Herpesvirus Infections in Marine Turtle Health and Conservation...... 187

REFERENCES ...... 195

BIOGRAPHICAL SKETCH ...... 208

LIST OF TABLES

Table page

3-1 Presence of cross-reactive antibodies to HSV antigens in plasma from green turtles with and without experimentally induced FP...... 82

3-2 Presence of cross-reactive anti-herpesviral antibodies in green turtle plasma...... 83

4-1 Net yield of LETV after three serial passages...... 107

5-1 Summary of monoclonal antibodies to LETV...... 140

5-2 Antibodies to LETV in wild turtles captured more than once...... 141

6-1 LETV genomic library containing homologs to human simplex virus genes...... 167

6-2 Percentage of amino acid sequence identity of UL26...... 169

6-3 Percentage of amino acid sequence identity of glycoprotein B...... 170

7-1 Anthropogenic threats to marine turtles...... 194

xii

LIST OF FIGURES

Figure page

2-1 PCR analysis of passaged primary cells established from a lung fibroma...... 50

2-2 PCR analysis of passaged tumor filtrate SSL398 and LETV in embryonated eggs during pilot study...... 51

2-3 PCR analysis of passaged tumor filtrates and LETV in embryonated eggs using degenerate primers...... 52

2-4 PCR analysis of passaged tumor filtrates and LETV in embryonated eggs using degenerate and specific primers in combination...... 53

2-5 PCR analysis of fifth passage of fat head minnow cells inoculated with tumor filtrate and LETV...... 54

2-6 PCR analysis of sixth passage of fat head minnow cells inoculated with tumor filtrate and LETV...... 55

3-1 Measure of antibodies to human simplex virus antigens in plasma from captive- reared and wild green turtles by ELISA...... 75

3-2 PCR analysis of HSV glycoprotein G using specific primer designed based on HSV1 sequences ...... 76

3-3 PCR analysis of glycoprotein G using specific primer designed based on HSV1 sequences to amplify FPHV gene homologs...... 77

3-4 Phylogenetic analysis of herpesviral glycoprotein B...... 78

3-5 PCR assay for herpesviral glycoprotein B using degenerate primers designed to group I...... 79

3-6 PCR assay for herpesviral glycoprotein B using nested degenerate primers designed to group I...... 80

3-7 PCR assay for herpesviral glycoprotein B using degenerate primers designed to group II...... 81

4-1 Schematic of plaque purification of LETV...... 99

xiii

4-2 Titer of LETV over time compared to percent cytopathic effect...... 100

4-3 Persistent infectivity of LETV after exposure to artificial seawater...... 101

4-4 Persistent infectivity of LETV after exposed to artificial seawater for 120 hr at 23 C...... 102

4-5 Persistent infectivity of LETV after exposure to artificial seawater at three temperatures...... 103

4-6 Persistent infectivity of LETV after exposure to natural seawater...... 104

4-7 Negative stain electron microscopy images of LETV after 24-hour exposure to artificial seawater ...... 105

4-8 Persistence of the LETV DNA polymerase gene after exposure to artificial seawater...... 106

5-1 Detection of seroconversion of LETV immunized green turtles ...... 131

5-2 Measure of antibodies to LETV in plasma from captive-reared and wild green turtles by ELISA ...... 132

5-3 Confirmation of the presence of anti-LETV antibodies by Western blot...... 133

5-4 Confirmation of the presence of anti-LETV antibodies by immunohistochemistry..134

5-5 Measurement of antibodies to LETV by ELISA in plasma samples from juvenile wild green turtles at three Florida study sites ...... 135

5-6 Measurement of antibodies to LETV by ELISA in plasma from wild green turtles scored for fibropapilloma severity...... 136

5-7 Summary of ELISA measurements for antibodies to LETV in plasma from 329 Florida wild green turtles ...... 137

5-8 Detection of antibodies to LETV by ELISA in plasma from nesting green and loggerhead turtles...... 138

5-9 Western blot confirms the presence of antibodies to LETV in selected plasma samples...... 139

6-1 LETV genome resolved and isolated by pulse-field gel electrophoresis...... 160

6-2 Identification of herpesviral 38 kDa protein by two-dimensional gel electrophoresis and Western blot analysis ...... 161

6-3 Amino acid sequences obtained from fractions of the P1 and P2 proteins...... 162

xiv

6-4 Schematic illustrating the process of mapping sequences of P1 and P2 to the overlapping UL26 and UL26.5 portion of the LETV genome ...... 163

6-5 Diagram representing a portion of a typical alpha-herpesvirus genome...... 164

6-6 Western blot analysis of expressed LETV proteins, UL26 and glycoprotein B...... 165

6-7 Phylogenetic analysis of LETV based on three herpesviral genes ...... 166

7-1 Amino acid sequence alignment of a portion of the herpesviral DNA polymerase gene of a LETV-like virus detected in loggerhead turtle (Cc0102) and other herpesviruses...... 191

7-2 Amino acid sequence alignment of the putative LETV semaphorin gene homolog and viral and cellular semaphorin genes...... 193

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

THE ROLE OF HERPESVIRUSES IN MARINE TURTLE DISEASES

Sadie Shea Coberley

August 2002

Chair: Dr. Paul A. Klein Cochair: Dr. Richard C. Condit Major Department: Immunology and Microbiology

Herpesviruses are associated with several diseases of marine turtles including lung- eye-trachea disease (LETD) and fibropapillomatosis (FP). Studies were designed to investigate the etiological role of herpesviruses in these diseases. Fibropapillomatosis

(FP) is a debilitating disease characterized by the development of external and internal tumors, and is increasingly prevalent among wild populations of marine turtles around the world. While a herpesvirus is the leading candidate for the etiology of FP, the FP- associated herpesvirus (FPHV) has not been isolated into pure culture, limiting definitive studies to establish causation. The clinical signs of LETD have been observed in free- ranging and captive-reared turtles worldwide, but the impact of this disease on marine turtle populations has not been evaluated. The LETD-associated herpesvirus (LETV) was successfully cultured and cloned from the index case of LETD and provided a unique resource to study the largely unexplored marine turtle herpesviruses. The ability to study

xvi

LETV concurrently with FPHV provided important research benefits. Numerous isolation attempts to cultivate FPHV in various host cell systems, though unsuccessful, were facilitated by having LETV serve as an effective virus growth control. LETV enabled studies demonstrating that this group of viruses can persist in an infectious state in a marine environment for extended periods of time. A specific serological assay for the detection of LETV antibodies was developed and exposure of wild turtles to this virus was demonstrated for the first time. The level of conservation among antigens of LETV,

FPHV, and HSV was evaluated to determine if any of these herpesviral antigens could be used in the development of specific serological assays for FPHV exposure. These studies revealed levels of immunological complexity that need to be resolved before useful seroepidemiology can be developed. Two immunogenic herpesviral proteins, LETV glycoprotein B and UL 26, were identified, cloned, and expressed. These are the first proteins from a reptilian herpesvirus to be cloned, expressed and identified as immunogenic in their host species. Serodiagnostic assays using these recombinant herpesviral antigens will be developed to evaluate exposure of marine turtle populations to LETV and FPHV and better define the role of these herpesviruses in their respective diseases.

xvii

CHAPTER 1 INTRODUCTION

All species of marine turtles have suffered serious population declines from over- harvesting for their eggs, meat, and shells, entrapment by fishing lines and nets, collisions with boats, dredging operations, and from destruction of nesting beaches and foraging habitat, and are currently either threatened or endangered (National Research Council

1990). Permission to study or conduct experiments with these animals is restricted because of their endangered or threatened status. In addition, access to all life stages of turtles is further limited by their complex life cycle that takes an individual turtle over thousands of miles of pelagic ocean. Marine turtles traverse numerous marine habitats during their life history, resulting in fragmented knowledge about marine turtle behavior and ecology at these various sites. Limited access has made it difficult to determine the impact of infectious diseases on marine turtle populations, even though it is well- established that infections with pathogens are capable of causing significant mortality in marine life (Kennedy 1998;Kennedy et al. 2000;Nettles 1992).

Microorganisms Associated with Diseases of Marine Turtles

Various microorganisms have been associated with diseases of marine turtles. Many gram positive and gram negative bacteria and fungi have been linked with disease such as ulcerative stomatitis, obstructive rhinitis, and pneumonia (Glazebrook et al. 1993;Raidal et al. 1998). Severe enteritis or encephalitis associated with infection of Caryospora cheloniae, a protozoan parasite, has been observed in free-ranging green turtles (Chelonia

1 2 mydas). This coccidial pathogen had previously only been recorded in captive-reared hatchlings (Leibovitz et al. 1978), but was suspected to be the cause of mortality in 70 sub-adult green turtles (Gordon et al. 1993). In addition, trematodes, specifically spirorchids (Learedius learedi), have been documented in free-ranging and stranded sea turtles (Gordon et al. 1998;Graczyk et al. 1995;Herbst et al. 1998b), and their role in a neurological disease of loggerhead turtles (Caretta caretta) is currently under investigation (Jacobson et al. in press). Leeches have been observed as a secondary infection of other marine turtle diseases and have been discussed as a possible vector for transmission of infectious diseases (Aguirre et al. 1994;Lackovich et al. 1999). A number of viruses have been documented in association with diseases of marine turtles, many of which have been suggested as candidate agents of fibropapillomatosis (FP). A small, naked virus-like particle was found associated with FP and based on sequence homology was hypothesized to be a papillomavirus (Lu et al. 2000a). However, efforts to reproduce this finding were unsuccessful such that a papillomavirus could not be detected by other investigators (Brown et al. 1999). A virus belonging to the family Retroviridae was identified in FP tumors as well and has since been dismissed as an endogenous marine turtle retrovirus (Casey et al. 1998). Several herpesviruses are associated with marine turtles and include the leading candidate agent of fibropapillomatosis (Jacobson et al.

1986;Jacobson et al. 1991;Rebell et al. 1975). The marine turtle herpesviruses will be reviewed at length since the focus of these studies has been on the role of these herpesviruses in marine turtle diseases.

Herpesvirus Infections of Marine Turtles

There have been three documented diseases associated with herpesvirus infections of marine turtles. A herpesvirus has been identified in green turtles with experimentally

3 induced fibropapillomatosis (FP), and has been associated with naturally occurring FP in green, loggerhead, olive ridley (Lepidochelys olivacea), and leatherback (Dermochelys coriacea) turtles (Herbst et al. 1995;Huerta et al. in press;Jacobson et al. 1991;Lackovich et al. 1999;Quackenbush et al. 1998). In addition, herpesviruses have been associated with two diseases of mariculture-reared green turtles, gray patch disease (GPD), a necrotizing dermatitis of post-hatching green turtles (Rebell et al. 1975) and lung-eye-trachea disease

(LETD), characterized by conjunctivitis, pharyngitis, tracheitis and pneumonia (Jacobson et al. 1986).

Knowledge about the prevalence of these herpesvirus infections in marine turtles and their role as disease agents is limited; little is known about the etiology and the epidemiology of the marine turtle diseases LETD, GPD, and FP. The herpesviruses associated with LETD and GPD have been documented only in mariculture turtles, and the prevalence of these diseases in wild populations has not been evaluated. In contrast, since its was first described in 1938 (Lucke 1938;Smith and Coates 1938), FP has become an increasing threat to wild marine turtle populations and the FP associated– herpesvirus (FPHV) has consistently been found associated with FP tumor tissue.

GPD-associated herpesvirus

Of the marine turtle diseases where a herpesvirus has been implicated as the etiological agent, the least is known about GPD. It has been documented twice in the literature. The first publication appeared in 1975 and described the disease in captive- reared hatchling green turtles (Rebell et al. 1975). Two types of lesions, non-spreading papules and spreading gray patches, were observed. Viral inclusion bodies were visualized by electron microscopy in tissue biopsies, and viral particles morphologically similar to a herpesvirus were documented. The herpesvirus-like particles possessed a

4 dense core and were approximately 160-220 nm in size, depending on the preparation of the sample. The overall morphology, characteristics of the lesions, and the potential mode of transmission were all typical of a herpesvirus infection. Transmission of GPD was successfully performed by scratch inoculation of naïve sea turtles with bacteria-free preparations derived from the originally observed GPD lesions (Rebell et al. 1975). In a follow up paper by the same researchers, experiments demonstrated that the severity of the lesions on turtles with GPD was influenced by the temperature of the water in which hatchling green turtles were housed. The results from these experiments suggested that manipulation of the water temperature in mariculture systems might be a method of controlling disease (Haines and Kleese 1977). Initial attempts to culture an infectious agent from turtles with GPD were successful; however, it was not possible to maintain this virus in culture and the isolate was lost (Elliott Jacobson, personal communication).

Recent attempts to isolate the GPD-associated herpesvirus (GPHV) from lesions have also been unsuccessful (Lawrence Herbst, personal communication). Since new cases of

GPD have not been reported, no additional opportunities have arisen for further attempts to isolate GPHV from infected turtles. Because of the inability to maintain the GPD associated-herpesvirus in culture, the lack of opportunities to obtain samples from turtles with GPD, and questionable relevance to sea turtle health and conservation, GPD was not included in this research project. Instead, research efforts were focused on the herpesviruses associated with LETD and FP.

LETD-associated herpesvirus

The LETD-associated herpesvirus (LETV) was originally isolated from the index case of LETD in captive-reared green turtles by Dr. Jack Gaskin at the University of Florida

(Jacobson et al. 1986). Examination of the diseased turtles revealed caseous material in

5 the trachea and lungs, and an exudate covering the eyes, all of which were consistent with a lytic viral infection. Particles morphologically similar to a herpesvirus were visualized by electron microscopy. Since its original isolation, no studies have been conducted to characterize LETV. LETV was retrieved from the original cryopreserved stock, and a large portion of this dissertation is dedicated to the characterization of this virus. As the only marine turtle herpesvirus to be successfully maintained in culture, LETV provided a unique opportunity to begin to understand herpesvirus infections of marine turtles. LETV has served as a model marine turtle herpesvirus for gathering information about the general characteristics of this largely unexplored group of viruses. In addition, this research has provided the means to assess the prevalence of LETV infections in wild populations and the threat such infections may present for marine turtle health and conservation. The clinical signs of LETD have been documented in populations of free- ranging marine turtles worldwide (Glazebrook et al. 1993), and this disease becomes of increasing concern as sea turtle hatcheries and head start programs around the world continue to release turtles annually despite enzootic and epizootic infections.

Fibropapillomatosis-associated herpesvirus

Most of the research on herpesvirus infections of marine turtles has focused on fibropapillomatosis because of its marked increase in prevalence and severity in wild populations around the world. All attempts to isolate the FP associated-herpesvirus

(FPHV) have been unsuccessful, limiting studies to test the role of this herpesvirus as the etiological agent of FP. However, FPHV is regularly found associated with FP tumors and is a leading candidate for the etiology of FP (Lackovich et al. 1999). FP of marine turtles is a debilitating disease characterized by neoplasia and hyperplasia involving both the dermal and epidermal layers resulting in tumors that give the disease its name. If the

6 initial disease is left unchecked, fibromas frequently develop internally primarily in the lungs and kidneys. Currently, there is no cure for this disease and the only treatment or therapy has been the surgical removal of the external tumors. FP was first documented in

1938 in a marine turtle housed in a NY aquarium (Lucke 1938;Smith and Coates 1938). It was not until the 1980’s that field biologists really took note of this disease in free ranging populations. By 1994, the prevalence of FP was 50% in the Indian River Lagoon, which is one of our main study sites along the east coast of Florida (Herbst 1994). The estimated prevalence is now 72.5% for this population of juvenile green turtles

(Llewellyn Ehrhart, personal communication) and as high as 92% in certain populations of Hawaiian turtles (Balazs and Pooley 1991). It is not known why the prevalence of this disease has increased so drastically over the years, but it is of great concern to conservation efforts as all species of marine turtles are endangered or threatened. Since

FP is the infectious disease of greatest threat to almost every species of marine turtle

(green, loggerhead, olive Ridley, and leatherback) (Huerta et al. 2000;Jacobson et al.

1991;Lackovich et al. 1999;Quackenbush et al. 1998), a large portion of our research efforts have focused on epidemiological studies of FP and specifically the investigation of the role of the FPHV in FP.

The Herpesvirus Family

Herpesviruses are double stranded DNA viruses with an average diameter of 150 to

200 nm. Virions consist of an icosahedral nucleocapsid surrounded by an envelope acquired from the host cell membrane. Like some other DNA viruses, herpesviruses establish latency in the neurons that innervate the site of acute infection. In the neurons herpesviruses do not replicate and instead are maintained as episomal DNA. Stress and other stimuli are thought to activate herpesvirus replication and result in reoccurring

7 lesions at the previous site of acute infection. Herpesviruses have been documented in virtually every animal where their pathogenic potential ranges from subclinical manifestations to encephalitis and cancer (Cotran et al. 1994;Fields et al. 1996).

Herpesviruses have been divided into the subfamilies of alpha, beta, and gamma. The primary purpose for the division was to enable laboratory workers and clinicians to more easily identify a new isolate and to predict the properties of an isolate based on shared properties of the group. Classification was based upon biological properties rather than evolutionary relatedness and included host species, host cell range, replication efficiency, clinical manifestations, and latently infected cell type (Karlin et al. 1994). The alpha- herpesviruses exhibit a variable host range, efficient destruction of infected cells, and latent infection of sensory ganglia. In contrast, beta-herpesviruses display a restricted host range, long reproductive cycle and slow spread in culture. Infected cells become enlarged and latency is established in lymphoreticular cells and possibly in secretory glands. Gamma-herpesviruses are limited to the family or order to which the natural host belongs, replicate in lymphoblastoid cells, and in some cases cause lytic infection of epithelioid and fibroblastic cells. Gamma-herpesviruses tend to be specific for T and B lymphocytes and infection is normally pre-lytic or lytic but often times without the production of infectious virus (Roizmann et al. 1992).

The division of herpesviruses into subfamilies by biological properties generally matches the families as they would be divided based on DNA sequence homology and gene arrangement. A key exception is Marek’s disease herpesvirus (gallid herpesvirus 1).

Originally this virus was classified as a gamma-herpesvirus based upon its development of lymphoid type tumors and tumor-like growth properties in chickens. DNA sequence

8 information and gene arrangements instead place Marek’s disease herpesvirus in the alpha-herpesvirus subfamily. Interestingly, Marek’s disease herpesvirus also possesses properties of alpha-herpesviruses including growth in fibroblasts and rapid spread in culture. This demonstrates the difficulty in classification of herpesviruses, or any virus, based upon biological properties alone since very different results can be obtained with each assay used. Prior to Marek’s disease herpesvirus, none of the known alpha- herpesviruses commonly caused tumors in their natural host and remained associated with the tumor cells. HHV-6 is another example of a herpesvirus assigned to gamma- herpesvirus subfamily based on lymphocyte tropism. Analysis of sequence homology and gene organization resulted in the reclassification of HHV-6 as a member of the beta- herpesvirus subfamilies (Roizmann et al. 1992).

Currently herpesvirus classification is being reconstructed to assign herpesviruses subfamilies based upon their evolutionary relatedness. These relationships are based upon nucleic acid and amino acid similarities of conserved genes, such as glycoprotein B (gB) and glycoprotein H (gH), and are used to separate viruses into subfamilies. It is important to note basing classification solely on DNA homology is not without its pitfalls. Recent skepticism has been raised about the real relevance of such a scheme, especially in the case of bacterial systematics (Pennisi 1999). The particular gene selected for phylogenetic analysis will profoundly influence placement of a given organism in a phylogenetic tree. It is likely that the final determination will be made based on a combination of both biological properties and genetic homologies.

Lytic Infection

Most cells infected with herpesviruses do not survive. Typically, the herpesvirus replication cycle results in structural and biochemical changes that eventually lead to cell

9 death. Infected cells balloon, round up, and fuse to form giant cells. Necrosis or lysis of these herpesvirus infected cells and the collection of edema fluid result in the formation of intraepithelial vesicles. Viral replication continues in the cells at the base of the vesicle and infectious virions are contained within the serous vesicular fluid. The fluid filled vesicles often erupt and result in the release of the infectious virions and the formation of an ulcer (Fields et al. 1996;Kumar et al. 1997).

The clinical pathology is the combined result of cell lysis induced by viral infection and immune response. Control and resolution of herpesvirus infection are the results of the joined efforts of the innate, humoral, and cellular defenses. Natural killer cells and activated macrophages are part of the innate immune system that serves as an early defense that limits the progression of the herpesvirus primary infection. In addition, interferon a and b inhibit viral replication by triggering a signal transduction cascade that results in degradation of viral RNA and inhibits a eukaryotic protein synthesis initiation factor needed for viral replication. Antibodies directed against herpesvirus glycoproteins neutralize free virus and limit the spread to additional cells. However, since viruses can escape this humoral response by spread through cell-to-cell contact, the cell-mediated immune response is essential for controlling and resolving herpesvirus infections.

Without the cell-mediated response, infection disseminates to vital organs including the brain. It is the cell-mediated and inflammatory responses that are the major cause of the symptoms of viral infection (Janeway and Travers 1996;Murray et al. 1994).

Human herpesviruses

Human simplex virus-1 (HSV-1) replicates in the skin and mucous membranes at the sites of entrance. Infections with HSV-1 range from cold-sores and gingivostomatitis to

10 life threatening disseminated visceral infections and encephalitis. Necrotizing bronchopneumonia can also be induced by infection with HSV-1. Conjunctivitis is the result of HSV-1 infection of conjunctivae. Both HSV-1 and cytomegalovirus cause ulcers in the esophagus. Histologically, HSV-1 nuclear inclusions are observed in a narrow rim of degenerating epithelial cells. Similarly, cytomegalovirus inclusions are found in capillary endothelium and stromal cells at the base of the ulcer. In HSV-1 vesicular lesions individual epidermal cells at the edge of vesicles balloon and fuse to form giant cells typical of herpesvirus infections. The combination of intraepithelial edema and degeneration or lysis of epidermal cells results in vesicles that frequently burst and crust over and may result in superficial ulcerations. Mononuclear cells are observed at the site of these lesions (Cotran et al. 1994;Knipe and Howley 2001;Kumar et al. 1997).

Chelonian herpesviruses

Infection with chelonian herpesviruses results in lesions similar to those induced by human herpesviruses. Green turtles with GPD develop lesions described as nonspreading papules and spreading gray patches with superficial necrosis (Rebell et al. 1975). LETD included periglottal necrosis, tracheitis, conjunctivitis, and pneumonia in mariculture green turtles, and severe bronchopneumonia developed in several green turtles. In addition, giant cells were observed throughout the tracheal mucosa and infiltration of mixed inflammatory was observed in the submucosa (Jacobson et al. 1986). Similarly, herpesviruses are thought to be responsible for conjunctivitis, stomatitis, tracheitis, and pneumonia in several species of tortoises, Testudo hermanni, Testudo graeca, Testudo horsfieldii, Gopherus agassizii, and Malacochersus tornieri (Biermann and Blair

1994;Kabisch and Frost 1994;Marschang et al. 1997;Pettan-Brewer et al. 1996;Une et al.

1999).

Oncogenic Herpesviruses

Members of every herpesvirus subfamily have been implicated as oncogenic, i.e., capable of inducing tumors. The most thoroughly investigated are herpesviruses that infect humans. Four human herpesviruses, human simplex virus (HSV), Epstein-Barr virus (EBV or HHV4), human herpesvirus 6 (HHV6), and Kaposi’s sarcoma-associated herpesvirus (KSHV or HHV8), have been associated with human malignancies. While none of these herpesviruses have been proved to be the sole etiological agents of human malignancies, they are consistently linked with their respective diseases and are the leading candidates. It is difficult to identify conclusively the etiology of many herpesvirus-associated malignancies due to long latency periods, low number of infected individuals who eventually develop cancer, and in some instances the requirement of additional chemical or physical factors before cancer develops (Flaitz and Hicks 1998;zur

Hausen 1991).

In general, viruses induce transformation as a result of viral replication. However, it is only when replication is aborted that transformed cells remain in the host and have the potential to lead to cancer (Fields et al. 1996). Normally viral replication kills the cell by mechanisms such as inhibiting nucleic acid or protein synthesis, disrupting the integrity of the cell membrane during entry, direct cell lysis, or displaying viral proteins on cell surface recognized by the immune system (Cotran et al. 1994). In the host, the lytic or acute phase is typically the result of host cell surface markers triggering the immune system to mount a non-specific immune response that leads to cell lysis (Flaitz and Hicks

1998).

The most common underlying cause of cancer is loss of growth control in transformed cells. The accumulation of mutations in genes that control cell growth leads to cancer.

The majority of the cells in the body are in a state of cell arrest and only a certain portion are advancing through the cell cycle. Normally cell cycle checkpoints evaluate cells with damaged or mutated DNA before progressing into replication. This allows damaged cells, which have a high potential to develop malignancies, to be either repaired or eliminated.

For example, tumor suppressor genes, such as p53 (53kD protein) and Rb (retinoblastoma tumor suppressor gene product), act within the cell cycle to allow for the repair of damaged DNA (Flaitz and Hicks 1998). Mutations introduced by viruses or other insults can lead to the inactivation of these tumor suppressors. In addition, certain genes called oncogenes that are normally inactive in the host genome can be activated by viruses or other insults. The net result is the inability to pause or arrest the cell cycle (inactivated tumor suppressors) in combination with a stimulatory signal to proliferate (activated oncogenes). The gradual accumulations of these kinds of mutations lead to the progressive loss of growth control and are mirrored in the stepwise process of tumor development. This is true not only for virally induced tumors, but also for cancers of noninfectious origins (Eckhart 1998).

Most cancers are the result of a genetic predisposition, exposure to chemical or ionizing radiation, or oxidative damage caused by normal cellular metabolism (Eckhart

1998). Virus-linked human cancers are responsible for 10-15% of the total cancer incidence (zur Hausen 1991). In latent infection, the virus genome can become a permanent part of the host’s nuclei by either integrating into the host genome or by producing certain factors to remain associated with host DNA. This association frequently results in immortalization of the cell and initiates the potential for malignant transformation. Cell associated viruses influence the cell cycle by inhibiting tumor

13 suppressors. As a result damaged cells escape repair mechanisms or targeting toward apoptosis and are stimulated to progress through the cell cycle. These cells may then accumulate further mutations by other DNA damaging events such as chemical toxins, radiation, or even other viral infections. Furthermore, certain viruses possess oncogenes, such as c-myc, TGFa (tumor growth factor a), and bcl-2 (B-cell leukemia/lymphoma 2), and once introduced into a normal cell have the potential for transformation into malignant cells (Flaitz and Hicks 1998). Unlike RNA tumor viruses that acquire cellular oncogenes by accident during the integration step of replication, DNA tumor viruses encode viral oncogenes that are an important part of overall fitness of the virus (Fields et al. 1996). Oncogenes encode growth factors that stimulate a cell to divide, and the result is the cell’s independence from growth factors normally provided by its environment. In addition, viruses may alter receptor functions that generate growth promoting signals in absence of appropriate growth factors (Eckhart 1998).

Human simplex virus

No definitive oncogenic potential for HSV has been documented. However, animal and in vitro studies suggest that HSV may participate in malignant transformation in the presence of other carcinogens or ultraviolet light and may act to enhance activation or amplification of pre-existing proto-oncogenes.

Epstein-Barr virus

Epstein-Barr virus (EBV) is described as a ubiquitous viral infection where up to

100% of the world’s adult population is infected. Most are infected in late adolescence by virus shed in saliva, which can result in the onset of infectious mononucleosis. Epstein-

Barr virus has been linked to four different human malignancies. Of these, the role of

EBV has been best established in Burkitt’s lymphoma and nasopharyngeal cancer. Some

14 evidence suggests the involvement of EBV in B cell lymphoma and Hodgkin’s lymphoma. Seroepidemiological studies revealing high antibody titers in Burkitt’s lymphoma patients and isolation of EBV from Burkitt’s lymphoma cells in tissue culture provide the best evidence for EBV as causative agent. Detection of EBV in tumors varies geographically. More than 90% of Burkitt’s lymphomas contain EBV DNA in endemic areas of Africa as compared to 15-20% in other parts of the world. EBV DNA has also been detected in “virus-free” Burkitt’s lymphoma cells lines and nasopharyngeal cancer cells (zur Hausen 1991). However, it has been shown that persistence of EBV in the tumor cells is not required (Eckhart 1998). Most Burkitt’s derived cells form tumors in nude mice and tumors in semisolid soft agar, while a subset of EBV-immortalized lymphoblastoid cells regress in nude mice and fail to grow in soft agar. EBV has been shown to immortalize B cell lymphocytes and induce “lymphoma-like” disease in New

World primates (zur Hausen 1991). EBV has also been shown to induce fusion of infected cells to form polykaryons (Cotran et al. 1994).

Nasopharyngeal carcinoma (NPC) is malignancy that occurs at a high incidence in southeast Asia and in southern China, where it is the major cause of cancer death. EBV can be detected in the majority of all NPC cells regardless of geographic origin. Latently infected cells yield an immortalized cell line in which only a small subset of genes are expressed. In a healthy host, the virus-immortalized B cells are normally eliminated by cytotoxic T cells that recognize viral antigens. Deficiencies in the T cell response play a role in the emergence of lymphomas and carcinomas in patients infected with EBV since the virally infected cells can no longer be detected by normal immune surveillance

(Eckhart 1998).

EBV latency genes, EBV-induced nuclear antigen-1 (EBNA1) and -2 (EBNA2), and latent membrane protein-1 (LMP-1) are thought to be responsible for transformation and therefore oncogenesis (Mendelsohn et al. 1995). LMP-1 has been identified as the major effector of transformation because of its ability to induce a variety of cell phenotypic changes (Eckhart 1998). EBV encodes several additional oncogenic gene products capable of inducing malignant tumor formation, including bcl-2 and jun/fos. Another nuclear gene, EBNA-LP, has been shown to interfere with p53 and Rb function and is likely to be partially responsible for cell cycle dysregulation (Flaitz and Hicks 1998).

EBV induces cellular IL-6, cyclin D, and IL-8 receptor expression. In addition, many of the malignancies associated with EBV possess a translocation event of the cellular proto- oncogene c-myc. Normally c-myc interacts with other proteins to modulate transcriptional activation. When translocated, the function of c-myc is no longer regulated. Burkitt’s lymphoma is an example of an EBV associated malignancy where c- myc gene is activated by a translocation to loci encoding immunoglobulins (Eckhart

1998). In addition, it has been shown that NPC cells are not tumorigenic in immunodeficient mice until an activated c-myc gene is introduced (Eckhart 1998).

Therefore it is likely that oncogenesis is initiated by the immortalization of cells by EBV but this requires the introduction of additional mutations, such as the activation of c-myc, before a malignancy develops.

Human herpesvirus 6

Since its identification in 1986, HHV-6 has been associated with malignancies including non-Hodgkin’s lymphoma, Hodgkin’s disease, cervical carcinoma, and oral squamous cell carcinoma. In vitro studies have shown that HHV-6-transformed human keratinocytes produce tumors when injected into the ear of a nude mouse. In tissue

16 culture, HHV-6 induces cytopathic effect (CPE) and results in ballooning and syncytia formation in infected cells. Patients with HHV-6 associated diseases have elevated titers of anti-HHV-6 antibodies, and lymphoma biopsies and lymph nodes possess HHV-6

DNA sequences. It is hypothesized that transmission occurs via aerosols and saliva.

Serological assays have demonstrated that 80-90% of all adults have been infected during the course of their lifetimes (Drago and Rebora 1999). The direct involvement with human malignancies has not been proven. However, malignant HHV-6 infected cells have been shown to over-express TNF-a, and HHV-6 ORF-1 encodes an oncogenic protein shown to bind to p53 (Flaitz and Hicks 1998).

Kaposi Sarcoma-associated herpesvirus

Kaposi sarcoma-associated herpesvirus is associated with three diseases: Kaposi’s sarcoma (KS), primary effusion lymphoma (PEL) (or body cavity-based lymphoma

(BCBL)), and a subset of multicentric Castleman’s disease (MCD) (Schulz and Moore

1999). The strongest evidence for KSHV as an etiological agent of a disease comes from its association with KS. The association of KSHV with the diseases BCBL and MCD is less clear.

Primary effusion lymphomas (or body cavity-based lymphoma) are a distinct group of

B-cell, non-Hodgkin lymphomas. In contrast to other AIDS-related lymphomas, PELs exist as multiple lymphoma effusions in the pleural, pericardial, and/or abdominal cavities, usually without being able to identify a tumor mass (Arvanitakis et al. 1996).

PEL occurs primarily in young to middle-aged HIV-positive men although HIV-negative,

KSHV-positive cases have been described (Arvanitakis et al. 1996;Ferry and Harris

1997). The median survival time of a patient with PEL is typically only a few months.

Death is due to lymphoma or in immunosuppressed patients is the result of KS or opportunistic infections (Ferry and Harris 1997).

Multicentric Castleman’s disease is a rare and poorly understood disease characterized by massive growth of lymphoid tissues. Three defining characteristics of MCD are recognizable lymph node architecture, abnormalities of germinal centers, and plasmacytosis. Clinical characteristics of MCD are lymphadenopathy, organomegaly, and hypergammaglobulinemia. The median age for development of MCD is 50 years old with a median survival time of 29 months. Characteristics of MCD are identical to other clinical diseases such as autoimmune disorders and immunodeficiencies, and it is only by ruling out all other diseases that the diagnosis of MCD can be made. Infections are a common cause of death. Other causes include renal failure, progressive MCD, and other malignancies such as lymphoma and KS (Bowne et al. 1999;Janeway and Travers

1996;Peterson and Frizzera 1993).

Kaposi’s sarcoma has been divided among subtypes of clinical and epidemiological variations: African KS, Classic KS, post-transplant or immunosuppression-related KS, and AIDS-KS. The highest prevalence rates in the world are of the African KS subtype with no indication of immune suppression, whereas classical KS occurs predominantly in elderly men of Mediterranean descent. Post-transplant or immunosuppression-related KS is distinct in that it is highly related to iatrogenically induced immunosuppression. The most clinically aggressive subtype and most recently described is AIDS-KS. Despite the differences of these subtypes, they are histologically identical. Similar to other opportunistic infections, while illness is rare in healthy populations, it is both more common and severe in immunocompromised populations (Sarid et al. 1999).

Unlike other herpesviruses, KSHV infection is uncommon in the general population.

The prevalence of KSHV varies widely geographically. Not all of the mechanisms of transmission have been identified. Most of what is known about KSHV comes from the investigation of its role in KS. KSHV has been demonstrated to be poorly transmitted through blood products by low KS rates in transfusion recipients and hemophiliacs (Sarid et al. 1999). In addition, it is known that KSHV transmission is not limited to sexual contact. Saliva has tested polymerase chain reaction (PCR) positive for the presence of

KSHV DNA and is proposed as a mechanism of viral transmission. Studies of children in

Uganda where sero-prevalence to adult level is achieved long before puberty provide support for this hypothesis. In addition to serological studies, KS lesions have been assayed by PCR for the presence of KSHV DNA. It was demonstrated that KSHV can be detected in more than 90% of tissues by PCR (Drago and Rebora 1999) and is found in more than 90% of HIV-associated and HIV-free forms of KS (Schulz and Moore 1999).

For immunocompetent patients infection is subclinical in contrast to KSHV infection of immunocompromised individuals where tumors can be fatal (Eckhart 1998). KSHV tissue tropism appears to be endothelial cells lining the vascular spaces and the spindle cells of KS lesions (Drago and Rebora 1999). Kaposi’s sarcoma differs from other tumors in that lesions contain many different cell types. The dominant cell type is the spindle cell and is thought to be of endothelial origin. It is hypothesized that primary infected cells produce factors responsible for the recruitment of other cell types and are responsible for

KS pathogenesis (Eckhart 1998). The infectious nature of KSHV has been more recently demonstrated by isolation and propagation of a filterable agent from KS lesions in coculture. Cytopathic effect in culture and herpesviral like particles in cytoplasm and nuclei

19 visualized by EM have been revealed (Drago and Rebora 1999). The KSHV genome contains many genes homologous to cellular gene products involved in growth control and differentiation (Schulz and Moore 1999). Such “molecular piracy,” where viruses encode cellular homologs, has been documented in other gamma-herpesviruses and poxviruses (Flaitz and Hicks 1998). Specifically, the KSHV genome contains cellular gene homologs responsible for cell cycle regulation, control of apoptosis, immunoregulation and cytokine signaling. KSHV encodes v-cyclin, G-protein coupled receptors (which act as constitutively activated CXC chemokine receptors), viral IRF-1

(interferon regulatory factor, which inhibits interferon signaling), several macrophage inflammatory proteins (MIP’s) and v-IL-6. KS tumor cells have a highly restricted pattern of gene expression and few gene products can be detected including LNA (the major latent antigen), v-cyc (which has the capacity to abolish G1 checkpoint by Rb phosphorylation downregulation), and v-FLIP (homolog of a cellular apoptosis inhibitor)

(Schulz and Moore 1999). The function of viral cyclins is thought to be to maintain cell cycling in place of the cellular cyclin that gets downregulated during infection.

Disregulated cytokine signaling is thought to be a major factor in the development of KS

(Eckhart 1998). Similarly, v-IL-6 is likely to be responsible for B-cell proliferation in

Castleman’s disease. vMIP-I and vMIP-II have angiogenic properties beneficial for the nutriment of tumors. G-protein coupled receptors can also induce secretion of vascular endothelial growth factor (VEGF), the major angioproliferative cytokine (Schulz and

Moore 1999).

Chelonian herpesviruses and tumors in reptiles

The etiology of the tumors in reptiles is less well defined. There are many clinical case studies describing the presence of a tumor in reptilian species (Cooper et al. 1983;Drew

20 et al. 1999;Frye 1994;Harshbarger 1984;Helmick et al. 2000;Janert 1998;Latimer and

Rich 1998;Schultze et al. 1999). These are typically clinical case studies that are rarely pursued to determine the exact cause of tumor induction due to a number of factors including lack of reagents to test for the presence of a potential pathogen and the infrequency with which the particular tumor is observed. This is similar to instances where a virus, such as a herpesvirus, is found associated with diseased tissues (Cooper et al. 1982;Cox et al. 1980;Frye et al. 1977;Hauser et al. 1983;Jacobson et al.

1982b;Jacobson et al. 1982a;Simpson et al. 1979). The majority of chelonian diseases where a herpesvirus has been implicated as the etiological agent present clinical signs that are more consistent with a lytic herpesvirus infection (Marschang et al. 2001;Origgi and Jacobson 2000). The exception is fibropapillomatosis of marine turtles, and in fact, the majority of the recent literature pertaining to a tumor disease of a reptile is dedicated to FP.

Methods for Linking Viruses with Disease Etiology

Henle-Koch Postulates

Historically proof that an infectious microorganism is the etiological agent of a disease required fulfillment of Henle-Koch postulates. Henle proposed conditions that must be met to prove an agent as the cause of an infectious disease (Henle 1938), but it was not until his student, Robert Koch restated these postulates that the guidelines were recognized (Koch 1884). To fulfill these postulates, the suspected agent must present in the diseased host and absent from healthy individuals. The suspected agent must be isolated from the diseased host, grown in pure culture, and then be put back into a susceptible host where if it is the etiological agent, it will result in the development of the

21 original disease. Lastly, the agent should be re-isolated and purified from the experimentally infected host (Fields et al. 1996).

Rivers’ Review

Rivers (1937) reviewed the limitations of the Henle-Koch postulates specifically for proving a virus as the etiological agent of a disease. One of the predominant limitations of Koch’s postulates was the requirement that the causative agent be grown and purified.

The assumption was that all organisms could be cultivated in an artificial system and this criterion has been a major obstacle for many diseases where suspected pathogens have not been able to be grown in a laboratory. An additional limitation introduced by Koch’s postulates was the assumption that all diseases are caused by a single agent. It was not recognized that certain diseases are the result of combined or synergistic actions of two or more agents. Another limitation is that the criterion of strict association cannot always be fulfilled. According to Rivers interpretation of Henle-Koch’s postulates, the suspected agent should “occur in no other disease as a fortuitous and non-pathogenic parasite.”

However, there are cases where an individual is infected with a pathogen but does not display the disease and carriers may not display disease for years, if ever. During this time a host may be infected with another pathogen that causes a disease. The first asymptomatic infection now becomes a “fortuitous and non-pathogenic” agent (Evans

1976). Furthermore, there are instances where a pathogen cannot be put into its original host, particularly in the case of human diseases, and where no experimental or surrogate host can be identified (Rivers 1937). For viruses and other pathogens, the limitations of

Henle-Koch postulates have been a major obstacle in their recognition as a causative agent of a disease.

Hill’s Criteria for Causation

Hill (1965) proposed a set of criteria for establishing causality from epidemiological studies (Hill 1965). Those criteria are a general guideline for assessing causation of both infectious and noninfectious diseases. The criteria are strength of association, consistency, specificity, biological gradient, temporality, coherence, experimental evidence, plausibility, and analogy (Foster et al. 1993). Hill’s criteria are an important set of guidelines for establishing causation in situations where the suspected agent of a disease cannot be cultured. For instance, the classical form of Kaposi’s Sarcoma was characterized over 100 years ago (Cotran et al. 1994), but the etiology of this disease could not be established because the suspected agent could not be cultured. However, by utilizing Hill’s criteria for causation KSHV has been established as the causative agent of

KS.

Evans Unified Concept

The Henle-Koch postulates were revised by Evans (1976) to yield the Henle-Koch-

Evans postulates. With increased information about diseases and disease processes, the restrictive nature of the Henle-Koch postulates was emphasized. Evans (1976) developed a “unified” set of criteria for causation based on the compiling of all of the methods identified to overcome the limitations of Henle-Koch postulates. Evans concluded that the blind adherence to Henle-Koch postulates may act as a hindrance to scientific research instead of an aid as originally intended.

Strategies To Demonstrate Association Between Herpesviral Infection and Onset of Disease by Fulfilling Hill’s Criteria

Whenever the original Henle-Koch postulates can be fulfilled directly, the steps for establishing a suspected pathogen as the etiological agent of a disease are more

23 straightforward. This would be the situation for demonstrating the role of a culturable herpesvirus, such as LETV in LETD. However, in the case of FPHV which has not been isolated in pure culture, this becomes a more difficult task.

The process by which Kaposi’s sarcoma herpesvirus (KSHV) was shown to play an etiological role in human malignancies is informative. Given the inability to cultivate

KSHV, researchers relied on molecular strategies to gather information about the association of KSHV and human diseases to support causation. These approaches serve as an example of how to employ molecular techniques to investigate an unculturable infectious agent in order to establish causation and more specifically how to fulfill Hill’s criteria. With the inability to fulfill Koch’s postulates for both KSHV and FPHV, Hill’s criteria serve as an important guideline for evaluating the evidence for these herpesviruses as the causative agents of their respective diseases. This may be especially useful in the case of FP. The following discussion compares the application of Hill’s criteria as they have been or might be applied to KSHV, FPHV, and LETV.

Strength of association

KSHV. PCR primers were designed to detect the presence of KSHV herpesviral DNA in tumors of infected hosts. Amplified sequences were shown to have homology with other known herpesvirus sequences and confirmed that KSHV was a new human herpesvirus. Strength of association was supported by the presence of KSHV DNA in over 95% of KS lesions as detected by PCR from a wide variety of geographic settings, but was rarely found in tissues of patients without KS. In addition, it was demonstrated that KSHV-positive patients by PCR were 23 times more likely to develop KS than

KSHV-negative patients. In situ, PCR identified cells infected with virus, and probes

24 were able to detect all forms of viral DNA in isolated cells, indicative of viral replication

(Sarid et al. 1999).

FPHV. A PCR assay was also used to detect the presence of a unique herpesviruses in marine turtle fibropapillomas. Degenerate primers were designed to amplify a portion of

DNA polymerase gene conserved among the majority of herpesviruses (VanDevanter et al. 1996). A survey of Florida’s green turtle and loggerheads revealed that the majority

(96%) of tumors contained herpesvirus DNA while normal tissue from both tumor- bearing and tumor-free turtles did not contain this DNA. Sequence information of a portion of the herpesvirus DNA polymerase gene revealed that the viruses infecting all of species of marine turtles are closely related and likely to be different strains of FPHV

(Herbst et al. 1998a;Lackovich et al. 1999). Similar results have been obtained from surveys of Hawaiian green turtles (Quackenbush et al. 1998).

LETV. It is difficult to assess the strength of the association between LETV infection and LETD. Clinical signs of LETD have been observed in wild and captive reared turtles worldwide, but no virus isolation attempts have been made from wild turtles to assess the presence of LETV (Glazebrook et al. 1993). Examination of a larger population of diseased marine turtles is necessary to determine whether or not marine turtles with

LETD are exclusively infected with LETV.

Consistency of findings

KSHV. The association between KSHV infection and KS has been documented by many researchers. KSHV DNA has been detected in KS lesions with high reproducibility among different populations of patients. KSHV is present predominantly in tumors and is found at a higher copy number in fully established tumors than in early lesions. Different

25 methods have been employed to validate the presence of KSHV such as in situ hybridization and transmission electron microscopy (Sarid et al. 1999).

FPHV. FPHV infection of marine turtles with FP has been validated by several researchers. Independent studies have consistently detected FPHV DNA polymerase gene associated with FP among geographically distinct populations of diseased marine turtles.

No other herpesviruses have ever been detected. In addition, various methods have been used to assay for the presence of a herpesvirus including in situ hybridization and transmission electron microscopy.

LETV. The presence of LETV in diseased marine turtles has been documented only once in a small group of marine turtles in mariculture at the Cayman Turtle Farm

(Jacobson et al. 1986). The presence of a herpesvirus was determined by transmission electron microscopy and cultivation of a herpesvirus from LETD lesions. Further investigation is needed by multiple researchers in a larger number of diseased marine turtles.

Biological gradient

KSHV. To address the biological gradient or dose-response criterion, the location of virus at the site of the tumor was investigated. Skin adjacent to the tumor in general has a low rate of detection for KSHV by PCR and even lower rates of detection in distant unaffected skin. In situ studies have confirmed the localization of KSHV nucleic acid in the tumors and its absence in surrounding tissues. However, there were cases where

KSHV was detected in normal skin of KS patients, suggesting widespread dissemination.

Peripheral blood from patients in remission was significantly less likely to be KSHV positive than those with active KS. In addition, KS is often a slow growing tumor that can be controlled through local excisional therapy. This suggests that the cause of the tumor

26 is localized to the tumor itself and can be eliminated by removing the tumor (Sarid et al.

1999).

FPHV. The presence of the FPHV DNA polymerase gene was detected with a high frequency in fibropapillomas. Greater than 97% of all fibropapillomas and fibromas were

PCR positive for the presence of FPHV DNA polymerase gene (Lackovich et al. 1999).

This DNA was not present in healthy tissues from diseased or healthy turtles. Two experiments were conducted where tissue biopsies taken at increasing distances from a

FP tumor were assayed for the presence of FPHV by PCR (Lackovich et al. 1999). The

PCR signal decreased as a function of distance away from the tumor until FPHV could no longer be detected in the most distant biopsies (greater than 2 cm away from the tumor).

Transmission studies with a tumor extract or filtrate was capable of inducing FP at the site of injection (Herbst et al. 1995). Dose-response experiments await the cultivation of the FP-associated herpesvirus. In addition, FP tumors have been used as a source of

FPHV antigen in immunohistochemistry assays, currently the only serological test available for FPHV (Herbst et al. 1998b). The presence of the FPHV DNA in infected cells has also been demonstrated by in situ hybridization (Herbst, unpublished). There is also some suggestion that surgical intervention can be of therapeutic value if fibropapillomas are removed before development of internal fibromas. This suggests that if tissues containing FPHV, the leading candidate agent of FP, are removed then future disease (FP) can be prevented.

LETV. In the index case of LETD, herpesvirus virions were associated with lesions as visualized by transmission electron microscopy (Jacobson et al. 1986). Future

27 transmission experiments are need to evaluate the dose-response of purified LETV inoculated into marine turtles.

Specificity of association

KSHV. The specificity of the association was investigated by serological studies to detect KSHV infections. PCR had identified cells where KSHV was stability maintained and formed the basis for a serological assay using latent nuclear antigens. The investigation of the specificity of association will rely on the sensitivity of any given assay to detect the presence of KSHV. Improved detection over first and second generation assays is attributed to the use of latency associated nuclear antigens and two structural proteins found in the virion capsid and envelope. Sensitivities with the use of these antigens range from 80 to 95 percent. Whole virion antigens are hypothesized to have sensitivities approaching 100% but at the expense of a higher false-positive rates

(Schulz and Moore 1999). Seroepidemiological studies have shown that infection precedes development of KS and is associated with risk (Eckhart 1998) and rates were found to be highly variable geographically. HIV-patients who tested positive for KSHV, but had a low rate of KS, triggered interest in the possibility that low-virulence strains of

KSHV might exist, that additional factors are required for pathogenesis, or that other cross-reactive herpesviruses are present in some populations (Sarid et al. 1999).

FPHV. Similar serological analyses have been used to detect FPHV infection.

Antibodies produced by turtles with FP specifically reacted with herpesvirus-like inclusion bodies in cross-sections of fibropapillomas (Herbst et al. 1998b). Conversely, turtles without tumors did not produce antibodies against these inclusion bodies. A survey of a wild population of turtles was conducted by using plasma from 40 turtles (20 visually tumor free and 20 tumor bearing). These plasma samples were blindly tested on

28 normal tissue and tumor sections. The survey demonstrated the ability to differentiate between disease and non-diseased turtles using serology. There were 3 samples collected from tumor-free turtles that had some reactivity with FPHV inclusion bodies. It is hypothesized that these turtles were newly infected and were in the early stages of disease(Herbst et al. 1998b). This was the first evidence for the potential of a serodiagnostic assay with which populations of wild marine turtles could be evaluated for exposure to an infectious agent and then monitored for the onset of FP.

LETV. It is not known whether LETV infection is ubiquitous in marine turtle populations or if it is associated specifically with disease. A serodiagnostic assay to assess exposure to LETV has been developed and has demonstrated Florida marine turtles (21.6%) have been infected with LETV, but clinical signs of LETD were not observed (Chapter 5).

Temporal association

KSHV. Temporal association is one of the most critical criteria to fulfill. KSHV infection must precede development of KS. Because transmission studies cannot be conducted with KSHV (due to lack of a surrogate model), temporal association has been evaluated retrospectively. Serum banks for 40 HIV-infected patients were used to examine rates of seroconversion to anti-KSHV positivity. The result was that 80% of patients were seropositive for KSHV before the onset of KS (Sarid et al. 1999).

FPHV. Unlike the situation with KSHV, transmission studies can be conducted with

FPHV, but only with a limited number of endangered marine turtles. Filtrates were prepared from fibropapillomas and inoculated into green turtles (Herbst et al. 1995). Prior to inoculation and during the course of the experiment, plasma samples were taken from each turtle. These plasma have also been tested for reactivity against FPHV inclusion

29 bodies. Antibodies against FPHV developed only after inoculation with filtrate and only in turtles that developed disease. Antibodies never appeared in control animals (Herbst et al. 1998b).

LETV. Temporality has not been established because LETV has only been documented in marine turtles with fully developed LETD (Jacobson et al. 1986).

Transmission studies with inoculation of purified LETV into marine turtles are needed to assess a temporal association.

Coherence to the biology of the disease

KSHV. There is coherence between the characteristics of KSHV and what characteristics the KS agent is likely to possess. At-risk populations for KS tend to have a higher KSHV infection rate. The KSHV infection is of low prevalence and rarely results in clinical disease. However, like opportunistic infections, KSHV has a high rate of disease in immunosuppressed individuals. Coupled with other data, it is likely that KSHV infection alone is not sufficient for development of KS and that other factors are likely to play a role, specifically immune suppression (Sarid et al. 1999).

FPHV. Without seroepidemiological data from wild populations of turtles it is difficult to assess the coherence between FPHV and FP. The prevalence of FPHV infection in wild populations is not known. From transmission studies it is known that filtrates are necessary and sufficient to induce tumor formation in experimental turtles, but it is not known whether other factor(s) exist in the tumor filtrates (Herbst and Klein

1995a). There has been some suggestion that the production of okadaic acid, a known tumor promoter, by algae consumed by green turtles may have an etiological role in FP

(Landsburg et al. 1999). However, more thorough studies at three sites along the east coast of Florida have demonstrated that green turtles consuming the diet with the highest

30 amount of okadaic acid were the same population that never had a case of FP (Llewellyn

Ehrhart, personal communication). The requirement or involvement of such factors for the onset of FP has not been thoroughly investigated. Furthermore, there is no initial indication for the role of immunosuppression (Herbst and Klein 1995a). Infected turtles are capable of developing an antibody response to FPHV and parasite-specific antigens which suggests a functioning immune system (Herbst et al. 1998b). Furthermore, in situ hybridization has detected the presence of FPHV DNA polymerase gene in infiltrating mononuclear cells, suggesting that at least one branch of the non-specific immune system is also functional (Lawrence Herbst, unpublished). The role of other factors or co-factors in fibropapillomatosis etiology will require further investigation with the appropriate reagents and controls.

LETV. There is coherence between the in vitro characteristics of LETV and what characteristics the LETD agent is likely to possess. In culture, LETV infection of TH-1 cells and GTE 14D cells results in giant cell formation and rapidly progressive CPE.

These characteristics of LETV viral replication are consistent with the lesions originally observed in marine turtles with LETD.

Experimental evidence

KSHV. The experimental evidence to support KSHV as the causative agent is limited by the inability to conduct transmission studies in humans and the lack of an animal model to be utilized as a surrogate host. In the absence of transmission studies, the focus has been on the evaluation of the transformed cells. Phenotypically, KSHV possesses the shared properties of other herpesviruses. In addition, nearly all KS tumor cells are infected as determined by in situ hybridization and immunohistochemistry. KSHV has been shown to be stably maintained at high copy number in body cavity based lymphoma

31 cells. However, cultured cells from other types of lesions have low copy numbers of virus and quickly lose the KSHV genome in vitro (Drago and Rebora 1999). Similarly, KSHV can be cultivated in naturally infected lymphoma cell lines but has not been sustainably cultured from KS lesions. Primary effusion lymphomas have been demonstrated to be fully immortalized and clonal tumors. In the case of KS, studies of the clonality of tumor cells have generated contradictory results, partially due to mixed neoplastic nature of the lesions. Interestingly, tumor and cultured spindle cells do not grow in soft agar nor do they induce tumor in nude mice. It has also been found that KSHV is sensitive to the same anti-viral drugs that reduce the risk of developing KS. In addition, there are also reports of tumor regression after treatment with anti-herpesvirus agents (Sarid et al.

1999).

FPHV. The bulk of the experimental data supporting FPHV as the etiological agent of

FP comes from transmission studies. Fibropapillomas developed at the site of inoculation with filtrates, but did not form at sites inoculated with PBS controls (Herbst et al. 1995).

To obtain information about the size and characteristics of the infectious agent in the filtrate preparations, filtrates were filtered through 0.2m or 0.45m filters, or were treated with chloroform. The 0.45m filtered filtrates induced tumors at the site of inoculation while 0.2m filtered filtrates did not. This result suggested that the infectious agent(s) responsible for the induction of fibropapillomatosis is smaller than 0.45m and larger than

0.2m or that the titer was reduced below a transformation threshold. Filtrates were also treated with chloroform and then inoculated into turtles. The results indicated the infectious agent(s) are chloroform sensitive since chloroform-treated filtrates were not able to induce tumor formation. Tumor morphology and histological properties of the

32 experimentally induced fibropapillomas were identical to those found on wild turtles.

Inspection of these filtrates via transmission electron microscopy revealed herpesvirus- like particles, which was consistent with both size and chloroform sensitivity profiles

(Herbst et al. 1996).

To obtain information about the properties of these tumors, experimentally induced fibropapillomas were surgically removed and disassociated into single cells. These cells were put into a tissue culture system and assayed for properties of a transformed cell.

Cells did not demonstrate focus formation (contact inhibited) or a reduced requirement for serum. When assayed for invasiveness, tumor cells did not grow into colonies in soft agarose. However, when cells were put into the ear of a SCID mouse, a fibroma of green turtle origin was formed. Normal skin fibroblast cells were put into the ear of a SCID mouse as a control and did not form a tumor (Herbst et al. 1998c). No experiments have been conducted to determine the effectiveness of anti-herpesviral drug therapies for treatment of FP. Spontaneous tumor regression has been detected in a number of wild green turtles in Hawaii and Florida (Bennett et al. 1999;Hirama and Ehrhart in press).

LETV. The in vitro characterization of LETV has been limited to its initial isolation.

Experiments need to be designed and carried out to obtain information about the properties of LETV. Transmission studies with purified LETV would provide the best evidence to support that it is the etiological agent responsible for LETD.

Biological plausibility and analogy

KSHV. The biological plausibility and analogy criterion is satisfied by KSHV’s close relationship to other oncogenic gamma herpesviruses, including EBV and Herpesvirus saimiri (HVS). Both are known to immortalize lymphocytes and are associated with malignancies. These herpesviruses encode several oncogenes that have been implicated

33 as factors for tumorigenesis and overall pathogenesis. KSHV encodes homologs suggesting a similar process for inducing tumors. When taking into consideration the evolutionary relatedness and presence of oncogenes with all the other information known about KSHV and KS, KSHV largely fulfills Hill’s criteria as a causal factor and it is probable that KSHV is the causative agent of KS.

FPHV. Similarly, the criteria for biological plausibility and analogy are fulfilled by comparison of FPHV to other oncogenic herpesviruses. Based on DNA polymerase sequence, FPHV is an alpha-herpesvirus and shares this subfamily with the Marek’s disease virus and Lucké tumor virus. The oncogenic properties of these alpha- herpesviruses have been well characterized. Unfortunately, there is very limited sequence information available from FPHV and therefore homologs of oncogenes cannot be directly identified. However, differential display analysis has revealed under-expression of a putative thrombospondin (known tumor-suppressor and inhibitor of angiogenesis) in transformed cell obtained from fibropapillomas (Herbst et al. 2001). It is possible that

FPHV interferes with cell cycle regulation by disruption of tumor suppressor gene products or expression of virally encoded oncogenes. In addition, over-expressed genes

(putative beta-hexosaminidase and eukaryotic RF1, for instance) were also identified by differential display and homologs of these genes have been identified other types of neoplasia (Herbst et al. 2001). Evaluation of tumor cells may elucidate some of these details.

LETV. Biological plausibility is supported by LETV’s lytic replication in tissue culture. Tortoise herpesviruses demonstrate the same lytic characteristics in vitro and are known to induce lesion similar to LETV in tortoises (Biermann and Blair

1994;Marschang et al. 1997). In addition, LETV has been shown to remain infectious after 5 days of exposure to seawater (Chapter 4) and serological studies have demonstrated that marine turtles are exposed to LETV or LETV-like virus (Chapter 5).

Therefore, the replication and persistence properties of LETV and the evidence that turtles are infected with LETV suggest that this herpesvirus is likely to have the ability to induce LETD in marine turtles.

Strategies to Investigate the Role of Herpesviruses in Marine Turtle Diseases

In summary, given the difficulties in demonstrating LETV and FPHV as etiological agents (establish causation), a variety of strategies were employed in this research study to investigate the role these herpesvirus in their respective diseases. This included numerous attempts to isolate FPHV to be used in transmission studies to demonstrate causation directly. When this approach was unsuccessful, efforts turned toward the development of serological assays to establish causation via seroepidemiological studies.

These studies evaluated antigenic cross reactivity between marine turtle herpesviruses and expressed herpesviral proteins to determine whether these alternative sources of antigens could serve as surrogates for FPHV in the development of serological assays.

Studies were also conducted with LETV to obtain information about the general characteristics of marine turtle herpesviruses. In addition, serological assays specific for the detection of LETV antibodies were developed to determine whether wild populations were exposed to this disease-associated herpesvirus and to assess the feasibility of establishing a serodiagnostic specific for a given marine turtle herpesvirus. Finally, studies were performed to identify, clone, and express immunogenic proteins of marine turtle herpesvirus to be used as source of herpesviral antigens in vitro. These herpesviral proteins will provide a source of genuine FPHV and LETV antigen for the development

35 of serological assays to screen turtles for the presence of antibodies to marine turtle herpesviruses and will provide the means to establish improved assays for seroepidemiology.

CHAPTER 2 STUDIES ON THE REPLICATION OF MARINE TURTLE HERPESVIRUSES

Introduction

Current evidence supports the hypothesis that the fibropapillomatosis-associated herpesvirus (FPHV) has an etiologic (or epizootiologic) role in fibropapillomatosis.

Transmission studies clearly demonstrate the presence of an infectious agent in the tumor filtrates used to induce fibropapillomatosis (FP) in healthy turtles (Herbst et al. 1995).

Particles morphologically similar to herpesviruses have been observed in these tumor preparations. In addition, herpesviral DNA has been detected consistently in > 97% fibropapillomas and fibromas evaluated and not in unaffected tissues or samples from healthy individuals (Lackovich et al. 1999). Phylogenetic analysis of a portion of the

DNA polymerase gene determined that this virus is a unique member of the

Herpesviridae family (Lackovich et al. 1999;Quackenbush et al. 1998). Furthermore, the majority of green turtles with FP also develop antibodies to viral inclusion bodies within the tumor (Herbst et al. 1998b). Healthy turtles without FP do not develop these antibodies. These findings suggest that FPHV is not likely a bystander or secondary infection, but rather plays an etiologic role in FP. The FP-associated herpesvirus is currently the leading candidate as the causative agent of fibropapillomatosis.

The inability to cultivate FPHV has made it impossible to fulfill Koch’s postulates directly. Until FPHV can be isolated and maintained in pure culture, apart from any other viruses or agents, it will be more difficult to determine whether FPHV is the etiologic (or epizootiologic) agent of fibropapillomatosis in marine turtles. The direct method to

36 37 prove that FPHV is the causative agent of FP is to demonstrate that the isolated virus can reproduce the disease experimentally in susceptible green turtles.

This chapter is devoted to describing the numerous attempts to isolate FPHV in pure culture. These attempts utilized numerous host cell culture systems and embryonated eggs maintained at several temperatures, and included evaluation of several preparations of tumor filtrates as sources of FPHV. In addition, FP cell lines were developed and evaluated for the spontaneous production of FPHV. As the only marine turtle herpesvirus to be maintained in culture, the lung-eye-trachea disease-associated herpesvirus (LETV)

(Jacobson et al. 1986) was included in most inoculation studies as a representative of this group of marine turtle herpesviruses and positive control.

Materials and Methods

FP Filtrate Preparation

Fibropapillomas were obtained from green turtles: SSL442 and SSL446. The tumors were stored on ice during transport from the Turtle Hospital (Marathon, FL, USA) to the laboratory (Gainesville, FL, USA). The filtrates were prepared as previously described

(Herbst et al. 1995). Briefly, tumors were minced with sterile scissors. A tissue teaser was used to further mince the pieces in sterile phosphate buffered saline (less than 30% w/v). The slurry was then transferred to a sterile pre-chilled glass homogenizer. The resulting preparation was referred to as a crude tumor homogenate. Some homogenates were used directly in an unfiltered form, while others were filtered through a 0.45 mm filter to create FP filtrates. In addition, some filtrates were sonicated for 30 sec. Filtrates were either used immediately or stored at - 85 C for future use. Filtrates were our primary source of FP infectious agent (believed to be FPHV) from turtles with FP. Fresh filtrates

(not frozen) were evaluated for the presence of herpesviral DNA by PCR described

38 below. Filtrates previously prepared and stored frozen included samples from green

(SSL398, Donor 5, Donor 6, and Mate) and loggerhead (Bobbie) turtles (Herbst et al.

1995). Donor 5, Donor 6, and Mate filtrates have been previously demonstrated to be able to induce FP when inoculated into green turtles (Herbst et al. 1996). SSL398 and

Bobbie were PCR positive for herpesviral DNA. A cyst fluid (result of kidney fibromas) from experimental turtle F-3 was also used as a potential source of infectious material as well. The cyst fluid (pH 8.5) was filtered through 0.45 mm filter and was determined to be

PCR positive for herpesviral DNA before use. A filtrate was also prepared from a pool of fibropapillomas from green turtles X6837 and X6838.

Established Tumor Cells

Tumors (fibropapillomas and fibromas) from wild green turtles were collected at necropsy at the University of Florida and were transported in transport medium

(DMEM/F12 plus 5% FBS plus 3X antibiotics and antimycotics) on ice to the laboratory.

The tumors were washed three times in serum free medium and then minced with sterile scissors and forceps. Tumor pieces were either treated with collagenase (260 units) alone or with the addition of 2.5% trypsin in HBSS. Collagenase treated alone tumor pieces were incubated in T-25 flask at 28 C 5% CO2 overnight. Collagenase plus tyrpsin treated pieces were incubated at 37 C on a rotating arm for several hours. In both cases, cells were collected, pelleted, washed, and then plated in T-25 flasks. The remaining tumor fragments were plated in T-25 flasks with DMEM/F12 plus serum and antibiotics. All cells were monitored for development of cytopathic effect (CPE) or foci formation and by PCR for the presence of the herpesviral DNA polymerase gene.

Cells and Tissue Culture

Target cells included a whole 14-day green turtle embryo line (GTE14D), a 33-day whole loggerhead turtle embryo line, a green turtle embryo kidney line (Moore et al.

1998), a whole 10-day gopher tortoise (Gopherus polyphemus ) embryo line, and a terrapene heart (TH-1; ATCC No. CCL 50) line. All cell types were routinely cultured in standard plastic flasks with filtered caps and were maintained in DMEM/F12 supplemented with 5% fetal bovine serum (FBS) antibiotics and antimycotics at 28 C in humidified incubator with 5% CO2. Tissue culture cells inoculated with various virus preparations were incubated at 20, 28, or 32 C. Cultures were monitored for cytopathic effects (CPE), which became evident in the LETV- and HV4295-infected controls 3 to 6 days post-infection. Supernatants from inoculated cultures which did not show CPE were blind passaged 2 to 5 times on fresh cells.

Cultivated Chelonian Herpesviruses

Positive controls were two chelonian herpesviruses which replicated in all target cells tested: lung-eye-trachea disease-associated herpesvirus (LETV) isolated from a green turtle (Jacobson et al. 1986), and HV4295 isolated from a Hermann’s tortoise (Testudo hermanii) with vesicular stomatitis (Marschang et al. 1997).

Coculture Experiments

Cells were cultured in standard T-25 plastic flasks with filter caps at 20, 24, 28, or 32

C. Dilutions of homogenates and filtrates ranging from 1:5-1:20 were added to cell cultures which were 85-100% confluent. After 1 - 2 hours, 9 ml of fresh medium was added, and then cultures were returned to the incubator where they were incubated for up to 21 days. For co-cultivation, 104-105 fibropapilloma cells were allowed to contact target cell monolayers by settling, and were not further disturbed by medium additions.

Embryonated Eggs

Embryonated chicken eggs were evaluated for their ability to support replication of marine turtle herpesviruses. In a pilot experiment, SSL398 filtrate (0.45 m filtered) was used to inoculate embryonated eggs. LETV was included as a control. Eggs were candled to check for viability and sites were marked for either an allantoic or chorioallantoic membrane (CAM) route of infection. A 25 gauge needle was used to delivered 0.2 cc of

SSL398 or LETV into at least two eggs via each route. The eggs were sealed with glue and were incubated at 30 C (instead of 37 C) in a humidified chamber. Sterility of the remaining inoculum was evaluated by streaking the material on blood agar plates and incubating the plates overnight at 37 C. The viability of the egg was checked periodically, and after 5 days, the eggs were harvested and prepared for passage. The CAM was obtained using sterile scissors and forceps and was placed in a tube on ice. Any contaminating egg shells were removed from the CAM, and it was placed into sterile cold glass homogenizer. The sample was homogenized in 2-4 ml HBSS and was kept on ice for passage. The allantoic fluid was collected with a sterile Pasteur pipet and centrifuged at 1500 g for 10 min at 4 C to remove any red blood cells. Every egg sample was divided for passaging and for PCR analysis. The next set of eggs was inoculated with 0.2 cc of material from the first passage as before. Eggs were incubated and monitored for viability. SSL398 and LETV were passaged at total of three times.

Embryonated eggs (25-35 day of development) from the common freshwater snapping turtle (Cheydra serpentina) and the red-bellied slider (Pseudymys nelsoni) were evaluated for their ability to support the growth of marine turtle herpesviruses. Injection of virus containing filtrates (0.1-0.2 cc) into the eggs of freshwater turtles could not be achieved

41 because of the high internal positive pressure in these eggs which resulted in immediate forceful leakage of the inoculum onto the outside of the egg. Efforts were made to glue rubber vacutainer stoppers onto the surface of the egg to serve as injection ports. This technique was not successful.

Primary embryo cultures from these freshwater turtles were also developed. Snapping turtle (32 and 46 day old) and red bellied slider (15 day old) embryos were removed from the egg and minced into small fragments. The fragments were seeded into a 100 mm dish and were bathed in DMEM/F12 with 50% fetal bovine serum plus antibiotics. The fragments were incubated at 28 C to allow fragments to attach. Both primary cultures developed a heterogeneous population of fibroblast like cells. The embryo lines were incubated with filtrates or LETV, and were monitored for signs of CPE.

Collaborative Study with National Wildlife Health Center

In collaboration with the United States Geological Survey (USGS) National Wildlife

Health Center (Madison, WI), embryonated avian eggs (chicken and mallard) were further evaluated. At the National Wildlife Health Center, filtrates (0.2 cc) or LETV

(1:10) were inoculated onto the CAM of 13-16 day old avian eggs. The eggs were incubated at 30 C (instead of 37) for 5 days, and then the CAM was harvested. The CAM was washed with PBS, ground in glass homogenizer with 10 ml Hanks balanced salt solution, and then frozens. The sample was thawed and clarified at 2,000 rpm for 30 min.

Then, 0.2 cc of the supernatant was used to inoculate the next round of eggs. This process was repeated for 3 passages. Remaining supernatant and pellets from the clarification step were stored frozen at – 70 C and were sent to the University of Florida for evaluation. Green turtle embryo and fat head minnow cells were also included in studies conducted at the National Wildlife Health Center. Briefly, 0.6 cc filtrate or LETV diluted

1:4 in sucrose-EDTA was added directly onto fat head minnow cells with 4-5 ml

Leibovitz L-15 medium plus 10% FBS and onto green turtle cells with 4-5 ml RPMI

1640 plus 10% FBS. Fat head minnow cells and green turtle embryo cells were routinely maintained at 25 and 28 C, respectively. Cultures were incubated for 1 week and then were harvested freezing the entire flask at – 70 C. The flask was thawed, vortexed, and then 0.5-0.75 cc of the suspension (diluted 1:4 in sucrose-EDTA) was added to a new flask of cells. Remaining cell suspensions were centrifuged, and the resulting supernatant and pellet were frozen and sent to the University of Florida for further testing.

Transfection Experiments

Experiments were performed with lipofectin in an attempt to bypass receptor-mediated entry of virus. Filtrate (SSL398) was incubated with lipofectin as per instructions

(Invitrogen, Carlsbad, CA) and then was added to confluent green turtles cells grown in 6 well plates. The cells were monitored up to day 11. An additional experiment was performed with total genomic DNA isolated by phenol:chloroform extraction of SSL398 tumors filtrate and of LETV-infected cells. The DNA was incubated with lipofectin as instructed, was added to green turtle embryo cells, and then the cells were monitored for signs of CPE or foci formation.

Sample Preparation and PCR Analysis

Total DNA was extracted from samples with a commercially available kit (SV Total

RNA Isolation kit; Promega, Madison, WI) or by phenol:chloroform extraction. The presence of a herpesvirus was determined by using sequential nested consensus-primer

PCR (VanDevanter et al. 1996) to amplify an internal region of the herpesvirus DNA polymerase gene. The first round of amplification included 100 ng of sample template.

DNA from varicella-zoster virus (VZV; Varivax, Merck Co., West Point, PA), human

43 simplex 2 (HSV2; VR-540, ATCC, Manassas, VA) or tumor filtrate was used as positive control template. Nucleic acid-free water was used as the no template negative control. A

5 µl aliquot of that amplification mixture was subsequently used as template for the second round of amplification. Products were electrophoresed through a 1.5% agarose gel in 1X TAE buffer, and then the gel was stained with ethidium bromide and photographed on an UV gel imager. Specific primers for the DNA polymerase gene were also used in

50 ml PCR reactions (Quackenbush et al. 1998).

Results

Cultivation of the FP-associated Herpesvirus In Vitro

All attempts to cultivate the FP-associated herpesvirus from FP filtrates were unsuccessful regardless of target cell, filtrate preparation, or temperature evaluated. No signs of CPE or foci formation were ever documented and all cultures were negative by

PCR after the first passage.

Established Green Turtle Tumor Cells

Several cell strains were successfully established from tumors obtained from green turtles with fibropapillomatosis. Most cultures were negative for the presence of herpesviral DNA after the first passage in tissue culture. The one exception was cells obtained from a lung fibroma of a wild green turtle (GT1). This fibroma was PCR positive for the presence of the DNA polymerase gene of the second passage (Figure 2-1; lane 10). Cells were obtained from the fibroma by two different methods: collagenase plus trypsin treatment (Figure 2-1; lanes 4-7) and collagenase alone (Figure 2-1; lanes 8-

10). The cells released from the lung fibroma attached to tissue culture plates and grew into a monolayer of two distinct morphologies of fibroblasts. The fibroma cell strains were passaged and samples were taken to be tested by PCR for the presence of the DNA

44 polymerase gene. Cells generated by both treatments were PCR positive for the first passage (Figure 2-1; lanes 5 and 9). The second passage of cells generated by collagenase alone treatment were also PCR positive (Figure 2-1, lane 10), but not for collagenase plus trypsin treatment (Figure 2-1; lane 6).

Embryonated Eggs Inoculated with Marine Turtle Herpesviruses

Embryonated chicken eggs were inoculated with FP filtrate or LETV to evaluate the ability of avian species to support the growth of marine turtle herpesviruses. The only

PCR positive samples were from the first LETV passage (Figure 2-2). One allantoic fluid sample (Figure 2-2a; lane 7) and one CAM (Figure 2-2b; lane 6) were positive for LETV at first passage, but all subsequent passages were negative. All FPHV inoculated eggs were negative at every passage. The same is true for eggs inoculated via the chorioallantoic membrane route. Experiments with embryonated freshwater turtle eggs were further challenged by the high osmotic pressure of the egg itself. Even with the application of a rubber stopper to serve as an injection port, none of the turtle eggs were successfully inoculated.

Studies in Collaboration with the National Wildlife Health Center

Tumor filtrates and LETV were inoculated onto the CAM of chicken and mallard eggs at the National Wildlife Health Center. All of the CAM samples from chicken and mallard eggs were PCR negative for herpesviral DNA polymerase gene regardless of which primers were used (Figure 2-3; Figure 2-4). Fat head minnow cells (FHM) were also evaluated for their ability to support the growth of marine turtle herpesviruses

(Figure 2-3a; Figure 2-4a). All of the FHM samples were PCR negative when the degenerate primers were used for both rounds of amplification. All samples from FHM cells inoculated with tumor filtrate were PCR positive for the first passage when the

45 degenerate primers were used for the first round and specific primers were used for the nested amplification (Figure 2-4a; lanes 4-6). All samples tested were PCR negative for the third passage (Figure 2-4a; lanes 7 and 8). Sucrose-EDTA was added to preparations for all subsequent passages of FHM cells (Figure 2-5). Fifth passage FHM cells inoculated with LETV or tumor filtrates were all PCR positive for herpesviral DNA

(Figure 2-5; lanes 4-7). However, the sixth passage on FHM cells was PCR negative for both LETV and tumor filtrate samples (Figure 2-6). In addition, when then the fifth passage material of LETV or tumor filtrate was transferred to green turtle embryo kidney line (passage 6) no signs of CPE were observed and the PCR signal was also lost (Figure

2-6; lane 9). There were no signs of CPE or foci formation in FHM cells with tumor filtrates or LETV at any of the passages. Primary inoculation of LETV on green turtle embryo cells showed signs of CPE within 2-3 days and served as the positive control. In addition, while slow to develop, cytopathic effect was noted in LETV-infected green turtle embryo cells cultivated at 30 C.

Transfection of Tissue Culture Cells with Herpesviral DNA

Experiments were conducted with lipofectin to determine whether tumor filtrates or extracted genomic herpesviral DNA were capable of yielding virus in a tissue culture system by transfection. Total tumor DNA (positive for herpesviral DNA polymerase gene by PCR) was used to infect GTE 14D cells. Tumor DNA was incubated with lipofectin to form micelles for delivery of the DNA into the cell. LETV DNA was also evaluated.

However, no signs of foci formation or CPE were observed in cells incubated with either tumor or LETV DNA.

Discussion

More than 300 attempts to propagate the FP-associated herpesvirus were carried out under the various conditions described. While typical cytopathic effect could be induced consistently by chelonian herpesviruses LETV and HV4295, CPE was never observed in cultures inoculated with tumor filtrates or in tumor cells containing the FP-associated herpesvirus. Furthermore, FPHV DNA polymerase gene could not be detected by PCR for more than one to two passages in inoculated target cultures or in newly established FP cell lines. Newly established FP lines never developed CPE or foci formation.

Nevertheless, FP cell lines that were positive for virus may be a source of FPHVDNA and/or FPHV antigens. In addition, earlier passages may be useful targets for induction studies with various compounds such as phorbol esters or n-butyrate to promote lytic replication. This approach has been used to induce replication in several other herpesviruses (Cannon et al. 2000;Parcells et al. 2001;Westphal et al. 2000).

All embryonated eggs evaluated at the University of Florida and at the National

Wildlife Health Center could not support the growth of chelonian herpesviruses (Figure

2-1 through 2-4) even when incubated at 30 C (the upper limit for LETV). These studies have provided information about the host range of marine turtle herpesviruses and will help to narrow down the identification of suitable targets for marine turtle herpesvirus growth. Interestingly, fat head minnow cells inoculated with tumor filtrates or LETV were positive for herpesviral DNA at the fifth passage (Figure 2-5). This signal was lost after sixth passage (Figure 2-6). However, no signs of lytic replication were observed in either the tumor filtrate- or LETV-inoculated FHM cells. The persistence of herpesviral

DNA at the fifth passage is encouraging. The FHM cell culture is an epithelial cell culture system which can be incubated at temperatures that are compatible (25-30 C) with

47 reptilian body temperatures. It is possible that these cells support a low level of virus production that is below what would result in clear signs of CPE. In addition, it is worth pointing out that the signal was only detected after treating the passaged cell preparation with sucrose-EDTA, which is suggested to improve titers of Marek’s disease-associated herpesvirus (MDV) (Tina Jaquish, personal communication). LETV and MDV share the property of being tightly cell-associated and it was hypothesized that the addition of sucrose assists in liberating free virus.

Freshwater turtle eggs were evaluated as a host for marine turtle herpesviruses. The rationale for this was two-fold. (1) The first was that all species of marine turtles are either endangered or protected. Ideally marine turtle eggs would be evaluated for propagating marine turtle herpesviruses. However, access to marine turtle eggs is restricted. In addition, marine turtles grow to a large size such that there are logistical constraints to the number of marine turtles that can be maintained in captivity. Freshwater turtle eggs are readily available and the freshwater turtles can be easily maintained in aquaria. Freshwater turtles are an attractive alternative if they could serve as a surrogate animal model for epidemiological studies. (2) The second reason for conducting these experiments in freshwater turtle eggs is to provide the opportunity to troubleshoot the difficulties of inoculating an embryonated marine turtle egg due to excessive osmotic pressure. Unlike avian eggs, reptilian eggs do not have a large air space and are instead completely filled with the allantoic fluid, embryo, and associated membranes. Freshwater turtle eggs are readily available for future studies in developing a method to overcome excessive pressure when inoculating reptilian eggs.

Experiments were conducted with lipofectin to determine whether tumor filtrates or genomic herpesvirus DNA could be delivered to cells and was capable of yielding virus in tissue culture. The hypothesis was that using lipofectin as a delivery system would eliminate the receptor-mediated attachment and entry step required for herpesviruses to enter a cell. In this way, herpesvirus particles or DNA would be delivered into cells where it could be targeted into the nucleus for viral replication. Cultures were monitored for foci formation and CPE for several weeks and were discarded after neither was observed.

The inability to propagate the FP-associated herpesvirus in cultured cells has prevented the fulfillment of Koch’s postulates for evaluating the causal relationship between FPHV and FP. However, epidemiologic data may provide some alternative evidence for causation. There are several parallels between FP and its associated herpesvirus and Kaposi’s sarcoma (KS) in humans and its associated herpesvirus (KSHV, also called HHV-8). In both FP and KS, the herpesvirus is not an incidental finding in affected individuals. These herpesviruses are rare in unaffected individuals, but rarely absent from tumor tissues. In both diseases, it is unclear whether tumor growth results entirely from the proliferation of virally transformed cells or results in part from the stimulation of cell growth by signals from nearby virus-infected cells (Gallo 1998;Herbst

1994;Herbst et al. 1999). Gallo (1998) has proposed that HHV-8 promotes hyperplasia by paracrine action of infected cells. Many tumor cells are never infected with virus and true neoplastic cells are a minority in the tumors. Cell lines derived from both KS and FP develop into tumors in immunodeficient mice suggesting that some transformed cells exist in the original tumors (Gallo 1998;Herbst et al. 1998c).

FP may have a complex pathogenesis involving interactions among several viral and non-viral agents including environmental co-factors (Herbst et al. 1999;Herbst and Klein

1995a). Evaluation of the strength of this marine turtle herpesvirus as a sole risk factor for FP awaits propagation of the FP-associated virus, experimental inoculations of green turtles with the pure virus, and subsequent purification of the virus from induced tumors.

In this and other studies to date, FPHV has resisted propagation, although additional efforts ongoing in different cell culture systems may yet prove successful.

50 1 2 3 4 5 6 7 8 9 10 11

2,000 1,000 500 300 200

Figure 2-1. PCR analysis of passaged primary cells established from a lung fibroma. Degenerate primers were used to detect the herpesviral DNA polymerase gene in a lung fibroma of a wild green turtle. PCR products were resolved in a 1.5% TAE agarose gel. Lane 1: molecular weight (MW) marker. Lanes 2 and 3: VZV positive control. Lanes 4-7 include PCR products from cells obtained by collagenase plus trypsin treatment of the lung fibroma. Lane 4: primary cells. Lane 5: first passage. Lane 6: cells from second passage. Lane 7: cells from third passage. Lanes 8-10 include PCR products from cells obtained by collagenase alone treatment of the lung fibroma. Lane 8: primary cells. Lane 9: cells from first passage. Lane 10: cells from second passage. Lane 11: original fibroma tissue.

1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 500 300 200

1 2 3 4 5 6 7 2,000

1,000

500 300 200

Figure 2-2. PCR analysis of passaged tumor filtrate SSL398 and LETV in embryonated eggs during pilot study. Degenerate primers were used to assayed for the presence of the herpesviral DNA polymerase gene. PCR products were resolved in a 1.5% TAE agarose gel. (A) Samples passaged in chicken allantoic fluid. Lane 1: molecular weight marker (MW). Lane 2: HSV2 positive control. Lane 3: no template negative control. First passage of SSL398 in egg 1 (lane 4), egg 2 (lane 5), and egg 3 (lane 6). First passage of LETV in egg 1 (lane 7), egg 2 (lane 8), and egg 3 (lane 9). Second passage of SSL 398 in egg 1 (lane 10), egg 2 (lane 11), and egg 3 (lane 12). Second passage of LETV egg 1 (lane 13). (B) Samples passaged on chicken chorioallantoic membranes. Lane 1: molecular weight marker (MW). Lane 2: HSV2 positive control. Lane 3: no template negative control. Lane 4: First passage of SSL398. Lane 5: Second passage of SSL398. Lane 6: First passage of LETV. Lane 7: Second passage of LETV.

a. 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 500 300 200

b. 1 2 3 4 5 6 7 8 9 10 11 12

2,000 1,000 500 300 200

Figure 2-3. PCR analysis of passaged tumor filtrates and LETV in embryonated eggs and FHM cells using degenerate primers. Degenerate primers were used for both rounds of amplification to assay for the presence of the herpesviral DNA polymerase gene. PCR products were resolved in a 1.5% TAE agarose gel. Filtrates used for inoculation included green turtles Donor 5 plus Mate filtrate (I+II), green turtle SSL398 filtrate (III), and loggerhead turtle Bobbie filtrate (IV). (A) Samples passaged in fat head minnow cells and embryonated eggs. Lane 1: molecular weight marker (MW). Lane 2: SSL398 positive control. Lane 3: no template negative control. First passage of fat head minnow cells inoculated with filtrate I+II (lane 4), filtrate III (lane 5), and filtrate IV (lane 6).Third passage of fat head minnow cells inoculated with filtrate I+II (lane 7) and filtrate III (lane 8). Lane 9: third passage of mallard chorioallantoic membrane inoculated with LETV. Lane 10: third passage of chicken chorioallantoic membrane inoculated with LETV. First passage of mallard chorioallantoic membrane inoculated with filtrate I+II (lane 11), filtrate III (lane 12), and filtrate IV (lane 13). An arrowhead indicates the location of the expected PCR product. (B) Continued screen of samples passaged in embryonated eggs. Lane 1: molecular weight marker (MW). Fifth passage of mallard chorioallantoic membrane inoculated with filtrate I+II (lane 2), filtrate III (lane 3), and filtrate IV (lane 4). First passage of chicken chorioallantoic membrane inoculated with filtrate I+II (lane 5), filtrate III (lane 6), and filtrate IV (lane 7). Fifth passage of chicken chorioallantoic membrane inoculated with filtrate I+II (lane 8), filtrate III (lane 9), and filtrate IV (lane 10). Unfiltered (lane 11) and 0.2 micron filtered (lane 12) X6837-8 filtrate.

a. 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 500 300 200

b. 1 2 3 4 5 6 7 8 9 10 11 12

2,000 1,000 500 300 200

Figure 2-4. PCR analysis of passaged tumor filtrates and LETV in embryonated eggs and FHM cells using degenerate and specific primers in combination.Degenerate primers were used for the initial amplification and then specific primers for nested amplification to assay for the presence of the herpesviral DNA polymerase gene. PCR products were resolved in a 1.5% TAE agarose gel. Filtrates used for inoculation included green turtles Donor 5 plus Mate filtrate (I+II), green turtle SSL398 filtrate (III), and loggerhead turtle Bobbie filtrate (IV). (A) Samples passaged in fat head minnow cells and embryonated eggs. Lane 1: molecular weight marker (MW). Lane 2: SSL398 positive control. Lane 3: no template negative control. First passage of fat head minnow cells inoculated with filtrate I+II (lane 4), filtrate III (lane 5), and filtrate IV (lane 6). Third passage of fat head minnow cells inoculated with filtrate I+II (lane 7) and filtrate III (lane 8). Lane 9: third passage of mallard chorioallantoic membrane inoculated with LETV. Lane 10: third passage of chicken chorioallantoic membrane inoculated with LETV. First passage of mallard chorioallantoic membrane inoculated with filtrate I+II (lane 11), filtrate III (lane 12), and filtrate IV (lane 13). (B) Continued screen of samples passaged in embryonated eggs. Lane 1: molecular weight marker (MW). Fifth passage of mallard chorioallantoic membrane inoculated with filtrate I+II (lane 2), filtrate III (lane 3), and filtrate IV (lane 4). First passage of chicken chorioallantoic membrane inoculated with filtrate I+II (lane 5), filtrate III (lane 6), and filtrate IV (lane 7). Fifth passage of chicken chorioallantoic membrane inoculated with filtrate I+II (lane 8), filtrate III (lane 9), and filtrate IV (lane 10). Unfiltered (lane 11) and 0.2 micron filtered (lane 12) X6837-8 filtrate.

1 2 3 4 5 6 7 8 9 10 11 12 13 14

2,000 1,000 500 300 200

Figure 2-5. PCR analysis of fifth passage of fat head minnow cells inoculated with tumor filtrate and LETV. Degenerate primers were used to assay for the presence of the herpesviral DNA polymerase gene. PCR products were resolved in a 1.5% TAE agarose gel. Lane 1: molecular weight marker (MW). Lane 2: VZV positive control. Lane 3: no template negative control. Samples were run in duplicate for all remaining lanes. Lanes 4 and 5: fifth passage of tumor filtrate on fat head minnow cells. Lanes 6 and 7: fifth passage of LETV on fat head minnow cells. Lanes 8 and 9: third passage of LETV on fat head minnow cells at 30 C. Lanes 10 and 11: third passage of LETV on fat head minnow cells at 25 C. Lanes 12 and 13: third passage of LETV on green turtle embryo kidney line. Lane14: VZV plus LETV positive control.

1 2 3 4 5 6 7 8 9

2,000 300 200

Figure 2-6. PCR analysis of sixth passage of fat head minnow cells inoculated with tumor filtrate and LETV. Degenerate primers were used to assay for the presence of the herpesviral DNA polymerase gene. PCR products were resolved in a 1.5% TAE agarose gel. Lane 1: molecular weight marker (MW). Lane 2: VZV positive control. Lane 3: no template negative control. Lanes 4: fat head minnow cells alone. Lane 5: green turtle embryo kidney line alone. Lane 6: sixth passage of tumor filtrate on fat head minnow cells. Lane 7: fifth passage of tumor filtrate on fat head minnow cells after passage on green turtle embryo kidney line (six passages total). Lane 8: sixth passage of LETV on fat head minnow cells. Lane 9: fifth passage of LETV on fat head minnow cells after passage on green turtle embryo kidney line (six passages total).

CHAPTER 3 EFFORTS TO IDENTIFY SURROGATE HERPESVIRAL ANTIGENS THAT ARE CROSS-REACTIVE WITH THE FIBROPAPILLOMATOSIS-ASSOCIATED HERPESVIRUS

Introduction

Epidemiological studies of fibropapillomatosis (FP) have been limited by the inability to cultivate the fibropapillomatosis-associated herpesvirus (FPHV). As a result, it has not been possible to conduct traditional transmission studies with purified FPHV to fulfill

Henle- Koch’s postulates directly. Seroepidemiological studies provide an alternative method to establish causation. The development of serodiagnostic assays could provide information to either support or disprove the hypotheses that FPHV and LETV are etiological agents of their respective diseases. The purpose of these studies was to explore antibody detection methods and antigenic cross-reactivity among selected herpesviruses in order to develop serological tests to assess exposure of marine turtles to disease- associated herpesviruses.

Development of serological assays for FPHV has been constrained not only by the inability to cultivate this virus, but also by the fact that FP tumors do not contain adequate amounts of FPHV antigen for use in serological assays. Previous serological assays for FPHV have relied on an immunohistochemical format using viral inclusion bodies in FP tumor sections as a source of FPHV antigen. Only 8 % of FP tumor sections screened are suitable for staining (Herbst et al. 1998b) making the application of this assay impractical for screening large populations of marine turtles for antibodies to

FPHV. Two strategies were used to generate or provide a source of antigen. The first

56 57 approach was to identify surrogate herpesviruses that cross-react with anti-FPHV antibodies and that could take the place of FPHV in serodiagnostic assays. Herpesviruses, in general, are known to possess homologous epitopes that may cross-react with antibodies produced against a closely related herpesvirus, e.g. HSV1 and HSV2 (Knipe and Howley 2001). Other marine turtle herpesviruses, such as the lung-eye-trachea disease-associated herpesvirus (LETV), may display epitopes with enough homology to

FPHV that anti-FPHV antibodies recognize and bind to preparations of LETV. To explore this hypothesis, antigenic cross-reactivity between FPHV and LETV was evaluated to determine whether LETV could serve as a surrogate antigen for FPHV.

Several human herpesviral peptides were also evaluated as a source of antigen to assess the conservation of specific herpesviral antigens.

The second approach was to identify a source of antigen that could be produced in vitro for serological assays. This included PCR amplification of glycoproteins known to be immunogenic in other herpesviruses. Focus was placed on glycoprotein B and glycoprotein G since these peptides were being evaluated by ELISA. PCR assays were developed in an attempt to amplify homologs of these glycoproteins from FPHV and

LETV with the eventual goal of cloning and expressing these genes to provide a source of herpesviral antigens.

Materials and Methods

Green Turtle Plasma Samples

Tumors from wild green turtles with FP have served as a source of infectious material to be used as FP filtrates for transmission studies (Herbst et al. 1995). Plasma samples from these turtles (donors) were evaluated. Plasma from juvenile captive-reared green

58 turtles (healthy turtles with no known exposure to herpesviruses) and plasma collected from juvenile captive-reared green turtles after experimentally induced fibropapillomatosis (recipients) were evaluated by the serological assays described below. Plasma samples from some of these green turtles with long-term recurring experimentally induced FP were also tested. Plasma samples collected from juvenile wild

FP positive turtles and FP negative green turtles netted in the Indian River Lagoon

(Indian River County FL USA; 27º49N, 80º26’W), and along the Wabassco Beach

(Indian River County FL USA; 27º47N, 80º24’W) were also included. All of these plasma samples had been previously evaluated by immunohistochemistry for antibodies to FPHV (Herbst et al. 1998b). Plasma from all FP positive green turtles contained anti-

FPHV antibodies and all FP negative green turtles did not (Herbst et al. 1998b). A purified immunoglobulin fraction from a polyclonal rabbit human simplex virus 1 antiserum was used as positive control when evaluating expressed HSV antigens (DAKO

Corporation, Carpinteria, CA).

Human Simplex Virus ELISA Assay

An ELISA assay was developed to detect the presence of antibodies to marine turtle herpesviruses that cross react with yeast expressed human simplex virus (HSV-1 and

HSV-2) antigens and included: HSV-1 bovine serum albumin-conjugated glycoprotein B peptide (gB1-BSA), HSV-1 glycoprotein G (gG1m), and HSV2 glycoprotein G (gG2c).

Clones for glycoprotein G contained coding sequence for the first 190 amino acids without the transmembrane domain (238 amino acids full length). The clone for HSV-1 glycoprotein B contained an undisclosed portion of the full length gene. Samples of these purified expressed proteins were generously provided by Chiron Corporation

(Emeryville, CA) and were used as antigen to coat ELISA plates with 2 ug/ml of each

59 protein in PBS/A. Each 96 flat bottom-well microtiter plate (Nunc-Immuno MaxiSorp,

Nunc, Kamstrup, Denmark) was filled with 50 ml of a mixture of all three human simplex virus (HSV) antigens or bovine serum albumin (BSA; negative control antigen). The plates were covered with sealing tape and incubated overnight at 4 C. The wells were washed in an automatic plate washer (ELX50, Bio-Tek Instruments, Inc., Winooski, VT,

USA) 4 times with 250 ul phosphate buffered saline 0.5% Tween 20 and 0.02% sodium azide (PBS/T) and then blocked overnight with 5% w/v non-fat dry milk in PBS/T

(PBS/TM) at 4 C. After washing, 50 µl of green turtle plasma samples diluted 1:25 in

PBS/TM or HSV-1 antiserum (positive control) diluted 1:200 in PBS/A was added to the appropriate well and was incubated for 1 hour at room temperature on a Nutator (Becton

Dickinson, Sparks, MD, USA). The microplate was washed, 50 ml of biotinylated monoclonal HL858 (Herbst and Klein 1995b) and HL673 (Schumacher et al. 1993) diluted in PBS/A (1:5000 and 1:3000, respectively) was added to each well incubated with green turtle plasma, and alkaline phosphatase-conjugated goat anti-rabbit IgG

(Sigma, St. Louis, MO) diluted in 1:1000 in PBS/A was added to each well incubated with rabbit anti-HSV-1 immunoglobulin. The plate was incubated 1 hour at room temperature on a Nutator. After washing, 50 µl of alkaline phosphatase-conjugated streptavidin (Zymed Laboratories Inc., San Francisco, CA, USA) diluted 1:5000 in

PBS/A was added to each well incubated with green turtle plasma. The microplate was incubated and washed as described above, and then 100 µl of p-nitrophenyl phosphate disodium (1 mg/ml in 0.01 M sodium bicarbonate pH 9.6, 2 mM MgCl2) was added to each well. The microplate was incubated in the dark without agitation at room temperature. The A405 of each well was measured after a 1 hour incubation in an ELISA

60 plate reader (Spectra II, SLT-Labinstruments, Salzburg, Austria with DeltaSoft version

3.3 software, Biometallics, Princeton, NJ, USA). Optical density values were adjusted as follows. For each plasma sample, the optical density of the well coated with BSA was subtracted from the optical density of the well coated with expressed antigen to account for non-specific absorbance, and optical densities equal to or less than zero were adjusted to 0.001.

Statistical Analyses

Statistical analyses were conducted using SigmaStat 2.03 software (SPSS

Incorporated, Chicago IL). The Kolmgorov-Smirnov normality test was applied to optical density value data measured by ELISA from plasma of captive-reared FP negative, captive-reared FP positive, wild FP positive, and wild FP negative. Because all of the groups were non-normally distributed the following statistical analyses were used. The

Mann-Whitney rank sum test was used to compare captive-reared FP positive turtles to captive-reared FP negative turtles, and to compare wild FP positive turtles to wild FP negative turtles. The Kruskal-Wallis one-way ANOVA on ranks was used to compare all of the groups. To identify the group or groups that differ from the others, all pairwise multiple comparison procedures were performed using Dunn’s method. For estimating a positive reaction, a cutoff value was based on the optical density values obtained from the

16 healthy captive-reared green turtles. The cutoff value was calculated as the highest optical density value obtained from these healthy turtles plus three times the standard deviation (cutoff value=0.323).

Immunohistochemistry with LETV-infected Cells

Plaque purified LETV clone 292 (Chapter 4) was propagated in a terrapene heart (TH-

1; CCL-50, ATCC , Manassa, VA) line grown to confluency in DMEM/F12

61 supplemented with 5% v/v fetal bovine serum and antibiotics in 2 well glass chamber slides (Nalgene Nunc, Naperville, IL) in a 5% CO2 humidified incubator at 28 C. One well was infected with LETV 292 and after 40-50% CPE was observed, cells in the infected and the uninfected wells were fixed with 2% w/v paraformaldehyde in phosphate buffer (0.12M NaH2PO4H20, 0.28M Na2HPO4, pH 7.2), permeabilized with 0.5% Triton

X100, washed with PBS, and then air dried. On the day of the assay, slides were rehydrated in PBS and endogenous peroxidases were inactivated by adding 3% hydrogen peroxide for 30 min. Slides were blocked with PBS plus 2 % v/v horse serum (PBS/H) for 20 minutes at room temperature in a moist chamber. Then, 100 ml of green turtle plasma diluted 1:25 in PBS/TM, or LETV control plasma (Chapter 5) diluted 1:100 in

PBS/F was added to each well. The slides were incubated overnight at 4 C in a moist chamber. Each slide was washed individually in 3 changes of PBS for 10 min each. For the following incubations, 100 ml of each reagent was added to each well and the slide was incubated for 45 min at room temperature in a moist chamber. Unbiotinylated monoclonal antibodies HL858 anti-green turtle 7S IgG (1 mg/ml) (Herbst and Klein

1995b) and HL673 anti-tortoise Ig light chain (Schumacher et al. 1993) at 1 mg/ml in

PBS/H were added to each well. Slides were washed as before and then 2% v/v universal anti-IgG-conjugated biotin in PBS (Vectastain ABC kit, Vector Laboratories Inc,

Burlingame, CA) was added. The slides were washed and then incubated with horseradish peroxidases-conjugated streptavidin. After washing, slides were incubated in

Coplin jars containing chilled chromogen (0.3 mg/ml 3,3-diaminobenzidine tetrachloride in PBS) and monitored microscopically for color development. Using standard procedure, slides were counterstained with hematoxylin, dehydrated, and mounted with coverslips

(Coligan et al. 2001). Stained slides were viewed with an Olympus BX50F microscope and photographed with an Olympus digital camera.

LETV ELISA Assay

An ELISA assay specific for the detection of anti-LETV antibodies was developed

(Chapter 5). Briefly, each well of a 96 flat bottom-well microtiter plate (Nunc-Immuno

MaxiSorp, Nunc, Kamstrup, Denmark) was filled with 50 ml of LETV or uninfected cell lysate control at a concentration of 5 mg/ml, and then covered with sealing tape and incubated overnight at 4 C. The wells were washed as described above and then blocked overnight with PBS/TM at 4 C. After washing, 50 µl of green turtle plasma samples diluted 1:25 in PBS/TM or LETV negative or positive control turtle plasma (Chapter 5) diluted 1:100 in PBS plus 2% v/v FBS (PBS/F) was added to the appropriate wells. Each plasma sample was incubated with LETV and uninfected cell lysate control for 1 hour at room temperature on a Nutator. The microplate was washed and 50 ml of biotinylated monoclonal HL858 (Herbst and Klein 1995b) diluted to a concentration of 1 mg/ ml was incubated 1 hour at room temperature on a Nutator. After washing, 50 µl of alkaline phosphatase-conjugated streptavidin was added to each well. The microplate was incubated and washed as described above, and then 100 µl of substrate was added to each well. The microplate was incubated in the dark without agitation at room temperature.

The A405 of each well was measured after a 2 hour incubation and the optical density values were adjusted as follows. For each plasma sample, the optical density of the well coated with uninfected cell lysate was subtracted from the optical density of the well coated with virus to account for non-specific absorbance, and optical densities equal to or less than zero were adjusted to 0.001. For estimating seroprevalences, a cutoff value was

63 based on the optical density values obtained from the 42 healthy captive-reared green turtles. The cutoff value was calculated as the highest optical density value obtained from these healthy turtles plus three times the standard deviation. The cutoff value for the

ELISA was 0.310. The development of the ELISA specific for the detection of anti-

LETV antibodies is described in detail elsewhere (Chapter 5).

PCR Development

Specific primers were designed based on HSV-1 glycoprotein G sequence information and were labeled PAK 40, 41, and 42. A PCR assay was developed to PCR amplify glycoprotein G (gG) from HSV-1 and HSV-2 and was then evaluated for the ability to

PCR amplify gG homologs from FPHV and LETV. Briefly, 200-1000 ng of human simplex virus-1 (HSV-1;VR-260, ATCC, Manassas, VA), human simplex virus-2 (HSV-

2; VR-540, ATCC, Manassas, VA), or Varicella-Zoster virus (VZV; Varivax, Merck Co.,

West Point, PA) DNA was added to a 50 ul reaction containing 0.2 mM dNTP, 0.5 uM each primer, 2.5 units of Taq polymerase plus buffer A (Promega, Madison, WI). A range of 1.0-4.0 mM MgCl2 was evaluated, and nucleic acid-free water was the negative control. One microgram of tumor DNA extracted from green (SSL 398 and SSL406) and loggerhead (Bobbie) was tested at as a source of FPHV DNA, and 1 ug of total DNA from LETV infected TH-1 cells was tested at as a source of LETV DNA. The PCR cycling was performed in a Perkin Elmer Gene Amp 2400 thermal cycler with the following cycling conditions: denaturing step at 94 C for 1 min, annealing step at 40 C for 1 min, and extension step at 72 C for 1 min for a total of 40 cycles. PCR products were electrophoresed through a 1.5% agarose gel in 1X TAE buffer and stained with ethidium bromide to visualize and photograph on an UV gel imager.

Nucleotide sequences for herpesviral glycoprotein B were obtained from Genbank

(National Center for Biotechnology Information). Phylogenetic trees for glycoprotein B were constructed using DNAMAN software version 4.02 (Lynnon Biosoft, Vaudreuil,

Quebec, Canada) and degenerate primers were designed based on two clusters. VZV and human cytomegalovirus (HCMV; VR-538, ATCC, Manassa, VA) were selected as representatives for each of the two clusters. Degenerate primer sets to the VZV cluster were designated 5-2 , 5-7 and nested primers 5-3, 5-9, and degenerate primer sets to the

CMV cluster were designated 3-1, 3-6 and nested primers 3-3, 3-7. Briefly, 20-100 ng of

VZV or CMV DNA was added to a 50 ul reaction containing 0.4 mM dNTP, 0.5 uM each primer, 2.5 units of Taq polymerase plus buffer A (Promega, Madison, WI). A range of

1.0-4.0 mM MgCl2 was evaluated, and nucleic acid-free water was the negative control.

One microgram of tumor DNA extracted from green turtles (SSL 398 and SSL406) and a loggerhead turtle (Bobbie) was tested as a source of FPHV DNA, and 1 ug of total DNA from LETV infected TH-1 cells was tested at as a source of LETV DNA. The PCR cycling was performed in a gradient cycler (Brinkmann, Westbury, NY) with the following cycling conditions: denaturing step at 94 C for 1 min, annealing step at 37-57 C for 1 min, and extension step at 72 C for 1 min, for a total of 40 cycles. A 5 µl aliquot of the initial amplification mixture was subsequently used as template for the nested amplification. PCR products were electrophoresed through a 1.5% agarose gel in 1X

TAE buffer and stained with ethidium bromide to visualize and photograph on an UV gel imager.

Results

Antigenicity of Human Simplex Virus Glycoproteins

The antigenicity of yeast expressed glycoproteins, HSV1 gB1-BSA, HSV1 gG1m, and

HSV2 gG2c was evaluated as a mixture and are referred to as HSV antigens. Plasma samples from 16 captive-reared green turtles without experimentally induced FP were tested in the ELISA format (Figure 3-1.). Optical density values for these plasma samples were low (mean A405 = 0.034; SE=0.016). To evaluate potential cross-reactivity between anti-FPHV antibodies and HSV antigens, plasma from 16 green turtles with experimentally induced FP and anti-FPHV antibodies previously detected by immunohistochemistry (Herbst et al. 1998b) were tested by ELISA (Figure 3-1.). The optical density values for these plasma samples were also low (mean A405 =0.133;

SE=0.057). Both groups were non-normally distributed (P<0.001). There was no difference in ELISA values between plasma from the turtles with or without experimentally induced FP (Mann-Whitney test; T=251.0; P value =0.637). Based on the cutoff value of 0.322, the green turtles with experimentally induced FP were negative for antibodies that cross-react with HSV antigens (Figure 3-1). The ELISA assay was then applied to plasma previously tested for FPHV antibodies (Herbst et al. 1998b) from free- ranging FP positive (n=22) and FP negative (n=17) green turtles to assay for the presence of cross-reactive antibodies. In contrast to the captive-reared turtles, ELISA values were higher from plasma collected from both FP positive (mean A405 = 0.851, SE=0.175) and

FP negative (mean A405 =0.601; SE=0.159) wild green turtles (Figure 3-1). Both of the wild groups were non-normally distributed (P<0.001). There was no difference between

ELISA values of plasma from FP positive and FP negative turtles (Mann-Whitney rank;

T=313; P=0.453) demonstrating the lack of a correlation between FP status of wild green

66 turtles and ELISA values measured to HSV antigens. Both of the wild groups were different from the captive-reared turtles (Dunn’s method; P<0.05). Based on the cutoff

(Figure 3-1), 52.9% of the FP negative and 59% of the FP positive green turtles possessed antibodies that cross-reacted with HSV antigens in this ELISA format.

Of these same wild green turtles, 20 FP positive and 17 FP negative, all of the FP positive turtles were negative for antibodies to LETV in the ELISA (Chapter 5) and all but two of the FP negative turtles were also negative for antibodies to LETV in the

ELISA. Of these 35 LETV antibody negative plasma samples from wild green turtles,

54.1% of the samples possessed antibodies that cross-reacted with HSV antigens demonstrating a lack of correlation between LETV antibody status and reactivity with these HSV antigens. Of the four plasma samples evaluated, two LETV antibody positive/FPHV antibody positive and two LETV antibody positive/FPHV antibody negative, all possessed antibodies that cross-reacted with HSV antigens. Statistical analysis could not be performed because only a limited number of LETV positive plasma samples were tested (n=4).

To further evaluate the relationship between FP status and reactivity with HSV antigens, plasma samples collected during the course of an FP transmission study (Herbst et al. 1995) were evaluated. Plasma samples were evaluated from green turtles (donors) from which FP tumors had been removed to generate FP filtrates for the transmission study. Plasma samples collected from green turtles before and after inoculation with FP filtrates (Figure 3-1) were also included. The turtles inoculated with FP filtrates are referred to as recipients. For two green turtles with experimentally induced FP, a plasma sample was collected several years after FP induction just prior to necropsy. The results

67 of the analysis are summarized in Table 3-1. All four of the donor green turtles possessed antibodies that cross-reacted with HSV antigens (Table 3-2). Again all of the plasma samples collected from captive-reared green turtles prior to inoculation with FP filtrates did not react with HSV antigens, but two of the turtles with experimentally induced FP did cross-react with HSV antigens (plus symbol; Table 3-1). The two samples tested from recipient turtles at necropsy possessed cross-reactive antibodies to HSV antigens (Table

3-1, superscript c).

Cross-reactivity with LETV-infected Cells

Several plasma samples were tested by immunohistochemistry (IHC) using LETV- infected tissue culture cells or HSV antigens as a source of antigen (Table 3-2, superscript d). All green turtles positive for antibodies to LETV in the ELISA (Chapter 5) reacted with LETV infected cells in IHC (Donor 3, Donor 4, and wild turtle N3162;

Table 3-2). There was no clear relationship between the presence of antibodies to FPHV and a positive reaction with LETV-infected cells in IHC. While some green turtles with anti-FPHV antibodies did react with LETV (Donor 1, Donor 2, and wild turtle BP3124;

Table 3-2), there were other turtles that did not (A2 and D2 post experimentation; Table

3-2). In addition, there were green turtles that were both FPHV negative in the IHC using tumor sections as antigen and LETV antibody negative in the ELISA, but were positive for antibodies that reacted with LETV in the immunohistochemical format (A2 prior to experimentation and wild turtle N3166; Table 3-2). The HSV antigens were also included for these plasma samples to further demonstrate the complexity of these antibody-antigen reactions. Many plasma samples possessed antibodies that cross-reacted with HSV antigens (Table 3-2; superscript e), but there was no clear relationship between FPHV or

LETV antibody status and reactivity with the HSV antigens. Furthermore, a wild green

68 turtle (BP3205; Table 3-2) negative for antibodies to FPHV and LETV possessed antibodies that cross-reacted with HSV antigens.

PCR Development for Herpesvirus Glycoproteins

While the complexity of the cross-reactions observed between antibodies against various marine turtle herpesviruses, LETV-infected cells, and HSV antigens remain difficult to interpret, there was the possibility of homology between herpes simplex virus and marine turtle herpesvirus glycoproteins. Primers were designed based on the cloned regions of HSV1 gG1 (Chiron Corporation, Emeryville, CA) and a PCR assay was developed to amplify gG from HSV1 or HSV2. PCR products of the expected sizes were observed with HSV1 DNA (Figure 3-2a) and not HSV2 (Figure 3-2b). The PCR signal was strongest at 1.5 to 2.5 mM MgCl2 for HSV1 (Figure 3-2a; lanes 3-5). Once the assay was established, FP tumor DNA positive for the presence of the herpesviral DNA polymerase gene was used as template for the reactions. The results demonstrated that primers designed against HSV-1 gG were not able to amplify FPHV gG from FP tumor

DNA obtained from green (SSL) or loggerhead (Bobbie) turtles (Figure 3-3).

Glycoprotein B (gB) is the most conserved glycoprotein among all herpesviruses

(Fields et al. 1996)and nucleotide sequences encoding gB from 21 herpesviruses were aligned. All sequences together did not possess enough homology to design a degenerate primer capable of amplifying gB DNA sequence from any of the representative herpesviruses. Sets of degenerate primers were instead designed to two groups (I and II;

Figure 3-4). A representative of each of the group was selected to be serve as a positive control to establish conditions for the PCR assay, VZV for group I and HCMV for group

II (Figure 3-4). A PCR product corresponding to the expected size for VZV gB was observed at annealing temperatures ranging from 37 to 57 C at 2.5 mM MgCl2 and

between 40.5 and 51.4 C at 3.0 mM MgCl2 (Figure 3-5). In addition, PCR amplification was observed with nested primers at range of magnesium concentrations and annealing temperatures (Figure 3-6), but the strongest signal was obtained from samples with 3.0 mM MgCl2 (Figure 3-6b; bottom row). No PCR product (expected 1775 bp) for HCMV gB was observed at 1.5 and 2.5 mM MgCl2 when primers 3-1 and 3-6 were used (data not shown). However, a PCR product of the expected size (731 bp) was observed for HCMV gB after the nested reaction at 2.5 mM MgCl2 with primers 3-3 and 3-7 (Figure 3-7; bottom row). The strongest signal was obtained at annealing temperatures between 37.8

C and 48.6 C. Once conditions were established for the nested PCR assay, pooled tumor

DNA positive for the presence of the herpesvirus DNA polymerase gene was used as template. LETV DNA extracted from infected cells was also included. While amplification products could be detected from VZV and HCMV control DNA, the gB homolog was never amplified from FP tumor DNA or LETV DNA (data not shown).

Discussion

Several immunological and molecular approaches were used to explore potential antigenic cross-reactivity between marine turtle and mammalian herpesviruses in an attempt to identify surrogate antigens for FPHV. This included the development of an

ELISA and immunohistochemical assays for the detection of antibodies to marine turtle herpesviruses using LETV as antigen. In addition, an ELISA assay was developed using

HSV antigens. These serological assays demonstrated the presence of antibodies in marine turtle plasma that cross-reacted with LETV and human simplex virus proteins presumably via conserved epitopes.

The ELISA assay identified wild green turtles with antibodies that cross–react with three expressed HSV proteins (Figure 3-1). However, there was no correlation between

FP status and reactivity to these HSV antigens (Figure 3-1). In addition, among the 35

LETV antibody negative green turtles tested, approximately half reacted positively with the expressed HSV antigens demonstrating the lack of a clear relationship between LETV antibody status and cross-reactivity with HSV antigens. The presence of cross-reactive antibodies in plasma samples from some marine turtles suggests either the response to these particular herpesviral proteins (gB and gG) varies from turtle to turtle, or that some individuals are co-infected with another serologically related virus(es) that is responsible for the positive reaction with HSV antigens by ELISA. However, when focus was placed on the green turtles involved in the transmission studies (Herbst et al. 1995), two of the recipient turtles (A2 and C2; Table 3-1) possessed antibodies that cross-reacted with the

HSV antigens. The experimental turtles were hatched and raised in captivity with no known exposure to pathogens prior to experimentation suggesting any cross-reaction was the result of cross-reactivity between anti-FPHV antibodies and HSV antigens. The majority of the experimental or recipient green turtles did not recognize these HSV antigens although all were positive for antibodies to FPHV. This suggests that there is a level of conservation between human simplex virus and FPHV glycoprotein B or G, but the response to these herpesviral proteins is variable from turtle to turtle.

When some of these same plasma samples from wild and experimental green turtles were then tested on LETV-infected cells and the results tabulated with the other data collected, the complexity of these reactions became evident (Table 3-2). All of the LETV

ELISA-positive turtles reacted positively with the LETV-infected cells in IHC (Donor 3,

Donor 4, and N3162; Table 3-2) confirming the presence of anti-LETV antibodies. In addition, many green turtles that were LETV antibody negative (by ELISA) and FPHV

71 antibody positive (by FP tumor section IHC) were positive in the LETV IHC regardless of whether they had naturally or experimentally induced FP (Table 3-2). This demonstrated the presence of cross-reactivity either between LETV and FPHV epitopes in the IHC format or alternatively cross-reactivity between LETV epitopes and another serologically related virus(es). The antibody responses of experimental turtles A-1, D-2 and G-3 at time of necropsy provide the clearest example of cross-reactivity between

FPHV antibodies and LETV in IHC format (Table 3-2). These experimental turtles were hatched and raised in captivity with no known exposure to herpesviruses prior to experimentation. All of these turtles possessed antibodies to FPHV specifically (LETV negative by ELISA) and these antibodies were shown to cross-react with LETV in the immunohistochemical format. In addition, plasma from experimental turtle G-3 was shown to possess antibodies that cross-reacted with LETV-infected cells in a Western blot format as well.

The antibody responses of two wild green turtles (N3166 and BP3205, Table 3-2) provide the clearest example of potential exposure of marine turtles to other serologically related virus(es). Both wild green turtles were negative for antibodies to both FPHV (in

FP tumor section IHC) and LETV (in the LETV ELISA), but possessed antibodies that cross-reacted with LETV-infected cells in IHC. This suggests that green turtles may potentially be exposed to viruses other than FPHV and LETV that also possess conserved immunogenic proteins further complicating the serological analysis.

Overall, there was no correlation between the presence of FPHV antibodies and reactivity with LETV-infected cells. This again suggests that while herpesviral proteins may be conserved, the response to these proteins may vary from turtle to turtle depending

72 on a number of factors (time post-exposure to virus, severity of infection, etc.).

Therefore, while immunoperoxidase testing using FP tumor sections has been capable of distinguishing FP-positive and negative turtles (Herbst et al. 1998b), it can be concluded that the IHC format with LETV-infected cells as a source of herpesviral antigen is not specific for the detection of antibodies to FPHV or LETV, is consistently sensitive for antibodies to LETV, and may in fact detect antibodies to viruses other than LETV and

FPHV.

It is also important to recognize differences in antigen presentation and overall format of the various serological assays evaluated. The LETV ELISA assay utilizes an infected- cell lysate containing LETV virions in various stages of maturation, and these LETV proteins are presented in their native state. In contrast, the LETV-infected cells utilized in the IHC have been fixed with paraformaldehyde and permeabilized with detergent altering their native state. The FP tumor sections used for immunoperoxidase staining have been prepared similarly. While all of these assays used preparations of whole virus, the HSV ELISA utilized only three expressed proteins. These HSV glycoproteins were produced in yeast and lack host post-translational modifications, which likely influence their reactivity as antigens. In addition, secondary antibodies used to detect binding of turtle antibodies to each antigen also have the potential to influence the result of a particular assay. The ELISA for utilizing HSV antigens and the immunohistochemistry assay utilized a mixture of monoclonal anti-light chain (HL673) and anti-7S IgG (HL858) green turtle antibodies, while the LETV ELISA utilized the anti 7S IgG antibody only.

Monoclonal antibody HL673 detects all classes of green turtle antibodies including IgM

(Herbst and Klein 1995b). Due to the high anti-cellular level observed in the LETV

ELISA, HL673 was not used for routine screening. However, secondary antibodies to several classes of green turtle antibodies (light chain, 7S IgG, and 5.7S) were tested in the

LETV ELISA when screening plasma samples from the experimental turtles. There was no difference in the LETV ELISA result for each plasma sample tested regardless of secondary antibody used. Plasma samples from the experimental turtles were LETV seronegative for all turtle antibody classes evaluated (data not shown). This suggests that the secondary antibody utilized does not account for the differences observed between the various antigens and formats tested. Finally, there are differences in the sensitivities between the various assays: an ELISA assay is more sensitive and can detect a lower level of antibody in contrast to the immunohistochemical assay (Tyler and Cullor 1989).

Taken together all of these variations critically influence the ability to detect anti- herpesviral antibodies in marine turtles.

Several PCR assays were developed in an attempt to amplify herpesvirus gene homologs from LETV and FPHV. Specific primers designed to HSV1 glycoprotein G were only able to amplify gG from HSV1 (Figure 3-2a) and not HSV2 (Figure 3-2b). In fact, glycoprotein G was selected by Chiron Corporation (Emeryville, CA) as a potential herpesviral protein capable of distinguishing HSV1 and HSV2 infections in humans. Not surprisingly, the primers designed to HSV1 gG were not successful at amplifying the more distantly related FP-associated herpesviruses (Figure 3-3). PCR assays were developed that were capable of amplifying glycoprotein B of human herpesviruses

(Figures 3-5, 3-6 and 3-7), but still these assays were unable to amplify the homologous glycoproteins from marine turtle herpesviral DNA (data not shown). This demonstrated that marine turtle herpesviruses do not share enough homology with human herpesviruses

74 for even degenerate primers to be successful at amplifying the marine turtle herpesvirus homologs.

These results have revealed new levels of complexity that must be addressed before reliable serodiagnostic assays for specific herpesvirus infections of marine turtles can be developed. While there appears to be a level of broadly conserved antigenic domains among herpesviral proteins, these proteins are not conserved well enough to serve as specific surrogate antigens for studies of FPHV exposure in marine turtles. It is evident that the identification, cloning, and expression of specific herpesviral antigens are needed to provide a defined source of antigens for studies of these infections (Chapter 6) and for the development of herpesvirus-specific serological assays for marine turtles.

3.00

2.50

2.00

1.50

1.00

0.50

0.00

-0.50 Captive Captive Wild FP Wild FP FP negative FP positive Negative Positive n=16a n=16a n=17b n=20b

Figure 3-1. Measure of antibodies to human simplex virus antigens in plasma from captive-reared and wild green turtles by ELISA. Plasma from 16 captive-reared turtles with no known exposure to herpesviruses and from 16 captive-reared turtles with experimentally induced FP had low optical density values. Plasma from 20 wild FP positive and 17 wild FP negative turtles had higher ELISA values, regardless of FP status. There was no statistical difference between the two captive-reared groups indicated by superscript a or between the two wild groups indicated by superscript b. However, the captive-reared groups (a) were significantly different from the wild groups (b). The cutoff of 0.322 (dashed lined) was used to calculate number of turtles with cross- reactive antibodies. Plasma samples were tested at a 1:25 dilution.

76 a 1 2 3 4 5 6 7 8 9 10

1,353 872 603 310 281

b 1 2 3 4 5 6 7 8 9 10

1,353 872 603 310 281

Figure 3-2. PCR analysis of HSV glycoprotein G using specific primer designed based on HSV1 sequences. PCR products were resolved in a 1.5% TAE agarose gel. (A) The glycoprotein G PCR amplification products when HSV1 DNA was used as template. Lane 1: molecular weight marker. Lane 2: 1.0 mM MgCl2. Lane 3: 1.5 mM MgCl2. Lane 4: 2.0 mM MgCl2. Lane 5: 2.5 mM MgCl2. Lane 6: 3.0 mM MgCl2. Lane 7: 3.5 mM MgCl2. Lane 8: 4.0 mM MgCl2. The remaining lanes were control reactions at 2.5 mM MgCl2 and include lane 9: VZV DNA and lane 10: no template control. (B) The glycoprotein G PCR amplification products when HSV2 DNA was used as template. Lane 1: molecular weight marker. Lane 2: 1.0 mM MgCl2. Lane 3: 1.5 mM MgCl2. Lane 4: 2.0 mM MgCl2. Lane 5: 2.5 mM MgCl2. Lane 6: 3.0 mM MgCl2. Lane 7: 3.5 mM MgCl2. Lane 8: 4.0 mM MgCl2. The remaining lanes were control reactions at 2.5 mM MgCl2 and included lane 9: VZV DNA and lane 10: no template control.

1 2 3 4 5 6 7 8 9 10 11 12

1,353 872 603 310 281

Figure 3-3. PCR analysis of glycoprotein G using specific primer designed based on HSV1 sequences to amplify FPHV gene homologs. PCR products were resolved in a 1.5% TAE agarose gel. Lane 1 and 12: molecular weight marker. Lane 2: HSV1 DNA. Lane 3: VZV DNA. Lane 4: water (no template control). Lane 5: SSL398 tumor 3. Lane 6: SSL398 normal skin. Lane 7: SSL406 tumor 1. Lane 8: SSL406 tumor 2. Lane 9: SSL406 tumor 3. Lane 10: Bobbie tumor 6. Lane 11: Bobbie tumor 10.

0.05

EHV4 60 FHV 30 100 PHV I CHV7 100 VZV * HCMV * 100 HHV7 II 100 100 HHV6 EBV 100 PSV 100 100 HSV1 100 HSV2 100 EHV2 EHV5 100 100 KSHV BHV 100 PRVP 100 100 PRVB SMV THV 100 MDV Figure 3-4. Phylogenetic analysis of herpesviral glycoprotein B. Two groups were selected and are indicated by I and II. One member of group I or II was selected as a representative of the group and is indicated by an asterisk. Various herpesviruses from all three subfamilies were evaluated and identified with full name and Genbank accession number below. EHV4=Equine herpesvirus 4 (AF030027); FHV =Feline herpesvirus 1(S49775); PHV=Phocine herpesvirus 1 (U92270); CHV7=Cercopithecine herpesvirus 7 (U12388); VZV= Varicella-Zoster virus (E01150); HCMV=Human cytomegalovirus (X17403); HHV6=Human herpesvirus-6 (X83413); HHV7=Human herpesvirus 7 (AF037218); EBV=Epstein-Barr virus (V01555); PSV=Pseudorabies virus clone (D00464); HSV1 = Herpes simplex virus 1 (X14112); HSV2= Herpes simplex virus 2 (Z86099); EHV2=Equine herpesvirus 2 (U20824); Equine herpesvirus 5 (AF050671 gB); KSHV= Kaposi's sarcoma-associated herpesvirus (U93872); BHV=Bovine herpesvirus 1 (M21474); PRVP=Pseudorabies virus, Phylaxia strain (A25912); PRVB=Pseudorabies virus, Becker strain (M17321); Saimiri alpha herpesviruse (M95786); THV= Meleagrid (turkey) herpesvirus 1 (U01887 ); MDV=Marek's disease virus (D13713).

37.8 57.3 C 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 2.5 mM MgCl2 1,000 500

3.0 mM MgCl2 2,000 1,000

Figure 3-5. PCR assay for herpesviral glycoprotein B using degenerate primers designed to group I. This assay included degenerate primers 5-2 and 5-7 and VZV DNA as template (Figure 3-4). The top row of lane in the agarose gel were PCR products from reactions run in the presence of 2.5 mM MgCl2 and the bottom row of lanes in the agarose gel were reactions run in the presence of 3.0 mM MgCl2. Under both conditions, lanes 2 to lanes 13 represent PCR reactions run at annealing temperatures ranging from 37.8 C to 57.3 C. The exact temperature for each lane is listed below. Lane 1: molecular weight marker. Lane 2: 37.8 C; Lane 3: 37.7 C; Lane 4: 38.7 C; Lane 5: 40.5 C; Lane 6: 42.9 C; Lane 7: 45.7 C; Lane 8: 48.6 C; Lane 9: 51.4 C; Lane 10: 54.0 C; Lane 11: 55.9 C; Lane 12: 57.1 C; Lane 13: 57.3 C.

37.8 57.3 C a 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 1.5 mM MgCl 500 2

2.0 mM MgCl 2,000 2 1,000 500

37.8 57.3 C b 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 2.5 mM MgCl2 500

3.0 mM MgCl2 2,000 1,000 500

Figure 3-6. PCR assay for herpesviral glycoprotein B using nested degenerate primers designed to group I. This assay included nested degenerate primers 5-3 and 5-9 after initial amplification with primers 5-2 and 5-7 and VZV DNA as template (Figure 3-4). (A) The top row of lane in the agarose gel were PCR products from reactions run in the presence of 1.5 mM MgCl2 and the bottom row of lanes in the agarose gel were reactions run in the presence of 2.0 mM MgCl2. Under both conditions, lanes 2 to lanes 13 represent PCR reactions run at annealing temperatures ranging from 37.8 C to 57.3 C. The exact temperature for each lane is listed below. (B) The top row of lane in the agarose gel were PCR products from reactions run in the presence of 2.5 mM MgCl2 and the bottom row of lanes in the agarose gel were reactions run in the presence of 3.0 mM MgCl2. Under both conditions, lanes 2 to lanes 13 represent PCR reactions run at annealing temperatures ranging from 37.8 C to 57.3 C. The exact temperature for each lane is listed below. Lane 1: molecular weight marker. Lane 2: 37.8 C; Lane 3: 37.7 C; Lane 4: 38.7 C; Lane 5: 40.5 C; Lane 6: 42.9 C; Lane 7: 45.7 C; Lane 8: 48.6 C; Lane 9: 51.4 C; Lane 10: 54.0 C; Lane 11: 55.9 C; Lane 12: 57.1 C; Lane 13: 57.3 C.

37.8 57.3 C 1 2 3 4 5 6 7 8 9 10 11 12 13

2,000 1,000 1.5 mM MgCl2 700 300

2,000 2.5 mM MgCl 1,000 2 700 300

Figure 3-7. PCR assay for herpesviral glycoprotein B using degenerate primers designed to group II. This assay included degenerate nested primers 3-3 and 3-7 after initial amplification with degenerate primers 3-1 and 3-6 (Figure 3-4). HCMV DNA was used as template. The top row of lane in the agarose gel were PCR products from reactions run in the presence of 1.5 mM MgCl2 and the bottom row of lanes in the agarose gel were reactions run in the presence of 2.5 mM MgCl2. Under both conditions, lanes 2 to lanes 13 represent PCR reactions run at annealing temperatures ranging from 37.8 C to 57.3 C. The exact temperature for each lane is listed below. The exact temperature for each lane is listed below. Lane 1: molecular weight marker. Lane 2: 37.8 C; Lane 3: 37.7 C; Lane 4: 38.7 C; Lane 5: 40.5 C; Lane 6: 42.9 C; Lane 7: 45.7 C; Lane 8: 48.6 C; Lane 9: 51.4 C; Lane 10: 54.0 C; Lane 11: 55.9 C; Lane 12: 57.1 C; Lane 13: 57.3 C.

Table 3-1. Presence of cross-reactive antibodies to HSV antigens in plasma from green turtles with and without experimentally induced FP.

Recipientsa Donorb A B C D

1 -/- c -/- -/- -/-

2 -/+ -/- -/+ -/-c

3 -/- -/- -/- -/-

4 -/- -/- -/- -/-

HSV antibody status prior to FP induction / antibody status post-experimental induction of FP. + = positive for antibodies to HSV antigens (see materials and methods). - = negative for antibodies to HSV antigens (see materials and methods). aRecipient= plasma from transmission turtle with or without induced FP. bDonor= turtle that provided source of tumor used to prepare filtrates to inoculate recipients. c This recipient turtle was positive for antibodies to HSV antigens at time of necropsy.

Table 3-2. Presence of cross-reactive anti-herpesviral antibodies in green turtle plasma.

Antibody status Antibody Reactivity Turtle IDa FPHVb LETVc IHCd HSV Age

Donor 1 + - + +

Donor 2 + - + +

Donor 3 + + + +

Donor 4 + + + +

A1 necropsyf + - + +

D2 necropsy + - + +

F3 necropsy + - - +

G3 necropsy + - + -

A1 postg + - unclear -

D2 post + - - -

A2 preh - - + -

A2 post + - - +

BP3125 + - + +

BP3124 + - + -

BP3205 - - unclear +

N3162 - + + +

N3166 - - + - aTurtle ID= name or tag number of turtle; turtle ID starting with BP or N are wild turtles. bFPHV= antibody status determined by IHC using FP tumor sections as antigen (Herbst et al., 1998). cLETV= antibody status determined by LETV specific ELISA (see materials and methods). dIHC= immunohistochemistry using LETV infected cells as antigen. eHSV Ag= human simplex virus antigens used as antigen. fnecropsy= plasma sample collected from turtle with experimentally induced FP just prior to necropsy. gpost= plasma sample collected 1 year after post-experimentally induced of FP. hpre= plasma sample collected prior to experimentally induced FP.

CHAPTER 4 PLAQUE-PURIFICATION, GROWTH CHARACTERISTICS, AND ENVIRONMENTAL STABILITY OF THE LUNG-EYE-TRACHEA DISEASE- ASSOCIATED HERPESVIRUS

Introduction

Lung-eye-trachea disease-associated herpesvirus is the only marine turtle herpesvirus to be maintained in cell culture. While the relevance of LETV to marine turtle populations is not clear, the ability to cultivate this virus provides the unique opportunity to gain insight into this group of herpesviruses as a whole. This includes studies designed to obtain information about general growth characteristics, host range, and ability to persist in the environment.

Prior to these studies little was known about the general virological characteristics of marine turtle herpesviruses. The growth characteristics have not been previously described nor has the host range for LETV been evaluated. It was important to obtain some basic information about LETV to maintain this virus in the laboratory and to learn how best to cultivate this virus for various experiments. In addition, determining the host range of LETV can provide valuable information for identifying candidate cell lines to be evaluated for their ability to support the growth of other marine turtle herpesviruses, such as FPHV.

Little is known about the mechanism of transmission of chelonian herpesvirus infections in terrestrial or marine environments. Potentially, herpesviruses may be transmitted to uninfected individuals by direct contact between infected turtles or from contact with substrates harboring virus, such as sediments, contaminated surfaces, or

84 85 seawater. It has been previously demonstrated that GPD and FP can be transmitted experimentally by scratch inoculation with preparations derived from their respective lesions (Herbst et al. 1995;Rebell et al. 1975). The ability of herpesviruses to be transmitted among sea turtles in a marine environment is likely to be influenced by the ability of these viruses to maintain infectivity for extended periods in seawater. While persistence of enteric viruses in the oceans has been well documented (Patti et al. 1996) and has recently become a source of increased concern (Ezzell 1999), the ability of herpesviruses to survive in a seawater environment has not been previously investigated.

Despite numerous attempts to cultivate GPD-associated herpesvirus and FP-associated herpesvirus, the LETD-associated herpesvirus (LETV) is the only marine turtle herpesvirus to be successfully isolated and maintained in culture. As such LETV was used in these studies to obtain information about the general characteristics of this virus and to test the ability of this virus to maintain infectivity after exposure to seawater.

Materials and Methods

Cultivation and Host Range of LETV

An aliquot of the LETV isolated from the original clinical case of LETD was obtained courtesy of Dr. Jack Gaskin (University of Florida, Gainesville FL USA)

(Jacobson et al. 1986). LETV was established in cell culture from an initial infection of a green turtle (Jacobson et al. 1986). Chelonian and mammalian cells were evaluated for their ability to support the growth of LETV as determine by presence of cytopathic effect

(CPE). Virus infected cell suspensions were diluted 1:10 into Dulbecco’s modified

Eagle’s medium (DMEM/F12, Life Technologies, Gaithersburg, Maryland, USA) supplemented with 5% fetal bovine serum and antibiotics, and then incubated on 80 to

90% confluent terrapene heart cells (TH-1; ATCC No. CCL 50) green turtle embryo cells

(GTE14D;Herbst et al. 1995), and green turtle D-1 cells (Herbst et al. 1998c) grown as monolayers in vented flasks in 5% CO2 at 28 C. Rabbit skin cells (RSC) and BSC40 cells

(African green monkey kidney cells) maintained in Modified Eagle’s Medium (MEM) plus 5% calf serum (CS) grown at 37 C 5% CO2 were also evaluated. Several flasks were shifted to 28 or 37 C 1 hour prior to infection. LETV infected cell lysate was added to each flask, diluted 1:5 in DMEM/12 and incubated 1 hour at 28 or 37 C with periodic rocking. TH-1 cells were also infected and served as the positive control and were maintained at 28 or 37 C. Uninfected TH-1, RSC, and BSC40 cells were also included at each temperature to serve as the negative controls. Cultures were monitored continuously for CPE and blind passaged on day 5 or 7 post infection. A primary Balb/c mouse ear cell strain was established by incubating the tip of an ear with 1 ml 2.5% trypsin in Hank’s balanced salt solution (HBBS) plus 1 ml collagenase (300 units) at 37 C with rocking.

Cells were collected at 3 and 6 hour post incubation, pelleted, and resuspended in DMEM plus 10% FBS plus 2x penicillin, streptomycin, gentamicin, and fungizone. Cells were divided among T-25 flasks and were incubated at 28 or 37 C. Some T-25 flasks were shifted from 37 C to 28 C overnight. LETV diluted 1:10 in medium was added to cultures and incubated at 28 or 37 C. Cultures were continuously monitored for CPE and were blind passaged on day 5.

Plaque Purification of LETV

The original aliquot of LETV was expanded in two passages on TH-1 cells and plaque assays were performed to isolate individual plaques following a standard protocol

(Mahy and Kangro 1996) with the following exceptions. The virus inoculum was incubated on 100 mm dishes of TH-1 cells at 28 C for 1 hr with rocking. Dishes were overlayed with agarose (SeaPlaque, FMC BioProducts, Rockland MA) instead of agar

87 since agar has been reported as toxic to reptilian cells and some animal viruses (Clark and

Karzon 1972) (De Maeyer and Schonne 1964). The agarose, incubated at 42 C instead of

55 C due to the temperature sensitivity of LETV (Curry et al. 2000), was mixed with 2X

DMEM/F12 supplemented with FBS and antibiotics, added to the dishes, and then the dishes were incubated at 28 C. Day 13 post-infection, dishes were overlayed with agarose plus neutral red, and scored on day 14. Plaques were picked with sterile glass Pasteur pipets and transferred to PBS plus 1% v/v bovine serum albumin (PBS/B). The virus released in the PBS/B was replated on TH-1 cells in 60 mm dishes, overlayed with agarose plus neutral red on day 9, and plaques were picked on day 10 as described above.

Each twice plaque-purified clone was expanded on a 100 mm dish of TH-1 cells. Two of the LETV clones, 221 and 292, were used in this study.

Growth Characteristics of LETV

The growth characteristics of LETV 221 was evaluated to determine the net yield of virus after passaging and to determine at what time post-infection yielded the greatest amount of virus. Medium was removed from a confluent 100 mm dish of TH-1 cells (5 X

106 ), LETV supernatant (3 X 103 pfu) was added to the monolayer, and the plates were incubated for 1 hour 28 C with rocking. DMEM/F12 plus 5% FBS and antibiotics (8.5 ml) was added to the culture and returned to 28 C. Cytopathic effect was monitored daily and plates were harvested at 99-100% CPE by scraping any remaining cells in medium.

The entire cell suspension was titered by plaque assay described above without freezing.

Next, 1.5 ml of the infected cell suspension was added to a confluent TH-1 100 mm plate prepared and incubated as before. This process was repeated for a total of three passages.

A growth study was performed to determine the relationship between CPE and virus titer. LETV 221 cell suspension (0.35 ml) was added to confluent TH-1 cells in 35 mm

88 plates and incubated 1 hour at 28 C. The inoculum was recovered from the time zero plate and the monolayer was washed with 1.5ml DMEM/F12. The unabsorbed virus and the DMEM/F12 wash were titered for residual virus. For all other plates, the inoculum was aspirated, plates were washed, DMEM/F12 plus 5% FBS (1.5 ml) was added, and then the plates were returned to the incubator. On day 1, 2, 3, 4,7,10, and 16 the estimated percent CPE was noted and the amount of virus was measured by titering the culture medium and the remaining monolayer.

Exposure of LETV to Seawater

TH-1 cells were infected with LETV or HV2245 isolated from a Hermann’s tortoise

(Biermann and Blair 1994). Infected cells showed 75 to 80% cytopathic effect (CPE) in approximately 4 days. Virus was harvested by removing the majority of the medium, freezing and thawing the flask, and scraping the infected cell monolayer culture into suspension. Virus preparations were dialyzed against artificial seawater (Aquarium

Systems, Mentor, Ohio, USA), natural seawater, or Hank’s Balanced Salt Solution

(HBSS, Life Technologies, Gaithersburg, Maryland, USA). Artificial seawater salinity was adjusted to 35 parts per thousand and specific gravity to 1.026. Natural seawater was collected from the University of Florida’s Whitney Marine Laboratory (Marineland,

Florida, USA) and from Sebastian Inlet, Indian River Lagoon (Melbourne, Florida,

USA). Osmolality (mmol/kg) of HBSS, natural seawater, and artificial seawater was measured with a Vapor Pressure Osmometer (Wescor, Logan, Utah, USA). Virus preparations (0.5-1.0 ml) were transferred to 10 kD molecular weight cut off dialysis cassettes (Pierce, Rockford, Illinois, USA) and dialyzed against 1L aerated artificial seawater, natural seawater, or HBSS maintained in incubators at 15, 23, or 30 C with continuous stirring. At 0, 24, 48, 72, 96, and 120 hr, virus preparations were removed

89 from dialysis cassettes and titered for infectivity and tested by PCR for the presence of virus.

Titration of viral infectivity on TH-1 cells

Dialyzed virus preparations were serially diluted 10-fold and inoculated on TH-1 cells grown in 96-well flat-bottom plates. Each LETV preparation was titered at time zero. An aliquot of each serial dilution was plated into five wells (150 ml per well) and each well was monitored for CPE. On day 10 post-infection, cells were scored for CPE and tissue culture infectious dose 50 (log10 TCID50/ml) were calculated to determine virus titers.

Electron microscopy

LETV virions incubated for 24 hr at 23 C in artificial seawater or HBSS were visualized by negative stain electron microscopy in the Electron Microscopy Core

Laboratory (University of Florida’s Interdisciplinary Center for Biotechnology Research,

Gainesville, Florida, USA). Briefly, 10 ml samples were floated on a 400 mesh formvar coated nickel grids for 1 min, blotted, and then stained with 2% aqueous phosphotungstic acid (PTA) pH 7.0 for 30 sec. Grids were blotted, air dried, and viewed with a Hitachi H-

7000 transmission electron microscope (Hitachi Instruments, Inc. Danbury, Connecticut,

USA). Digital images were obtained with a Gatan Bio-Scan camera and Digital

Micrograph 2.5 (Gatan, Inc. Pleasanton, California, USA).

PCR analysis of LETV after dialysis

The persistence of LETV DNA following exposure of LETV to seawater was measured by detecting the herpesvirus DNA polymerase gene amplified by a nested polymerase chain reaction (PCR) (VanDevanter et al. 1996). A 500 ml sample of each dialyzed virus preparation was stored at –85 C for PCR analysis. Thawed samples were

90 desalted and concentrated by using Microcon 30 Microconcentrators (Amicon, Bedford,

Massachusetts, USA). A 10 ml aliquot of each sample was boiled for 10 min and added to a 90 ml PCR reaction mixture. The PCR cycling was performed in a Perkin Elmer Gene

Amp 2400 thermal cycler. A 10 ml aliquot of each first-stage PCR reaction was used as template for the nested PCR amplification. Products were electrophoresed through a

1.5% agarose gel in 1X TAE buffer and stained with ethidium bromide to visualize and photograph on an UV gel imager.

Results

Cultivation and Host Range of LETV

LETV was successfully cultivated in all chelonian cells evaluated. However, no signs of cytopathic effect were detected in any of the mammalian cells tested at 28 or 37 C.

Cells were blind passaged at least once and were maintained for at least 20 days after passage. Overall, mammalian cells did poorly at 28 C even when initially grown at 37 C.

Rabbit skin and BSC40 cell lines lifted from culture flasks at 28 C regardless of the presence or absence of LETV. The Balb/c mouse ear strain tolerated 28 C but visually did not look as healthy as cells maintained at 37 C. The uninfected TH-1 cells remained intact (negative control) while infected cells (positive control) demonstrated CPE after 5 days at 28 C. Infected TH-1 cells did not show signs of CPE at 37 C. In general, TH-1 cells incubated at 37 C peeled from the flasks, but did appear to adjust to this temperature after blind passage.

LETV Plaque Purification Scheme and Lineage of Clones

A plaque assay was successfully developed for the purification and titration of LETV.

Two clones of LETV, 221 and 292, were generated during the purification scheme

(Figure 4-1) and were used for all experiments.

Growth Characteristics of LETV

LETV was serial passaged three times to obtain information about the net amount of virus obtained (Table 4-1). The multiplicity of infection (MOI) varied for each passage and was as low as 0.0001 (or 1 infectious virion per 1000 cells). The average net yield of

5 virus from each passage was 9.7 X 10 pfu/ml.

A time course study provide information about the general growth characteristics of

LETV (Figure 4-2). A one hour incubation with the inoculum is sufficient for approximately 90% of the total virus to bind to cells (data not shown). Less than 10% of the virus was present in the post-absorbed inoculum (data not shown). The temporal growth study of LETV revealed that the maximal amount of virus (pfu) did not correlate with the maximal amount of CPE. Growth curves demonstrated a clear eclipse phase

(Figure 4-2; Day 1) and then a gradual increase in titer until reaching a high titer of virus on day 9. LETV titer at day 16 was less than that of day 9 indicating that LETV is not stably maintained in culture at 28 C for prolonged periods of time. There was also an expected shift from the majority of virus being present in attached cells early during infection to the majority of virus being released into the medium (includes cells released from the monolayer and free virus in the supernatant) (data not shown). It is also important to note that even on day 16 a third of the total virus present in the medium was still tightly cell-associated (data not shown).

Infectivity of LETV after Exposure to Artificial and Natural Seawater

In initial experiments, the infectivity of LETV and HV2245 virus preparations was determined after a 48 hr exposure to artificial seawater (1000 mmol/kg) or HBSS (290 mmol/kg) at 23 C. As shown in Figure 4-3, the infectivity of both LETV and HV2245 viruses persisted after a 48 hr exposure to artificial seawater with remaining titers of 4.0 and 4.8 log10TCID50/ml, respectively. Furthermore, LETV exposed to artificial seawater for up to 120 hr at 23 C remained infectious (Figure 4-4).

LETV exposed to HBSS or artificial seawater showed similar levels of infectivity at each temperature. There was persistence of LETV infectivity at 15 C and 23 C but there was an approximate 10-fold and 100-fold reduction, respectively, regardless of exposure to HBSS or artificial seawater (Figure 4-5). LETV was less stable at 30 C and showed total loss of infectivity after 48 hr.

To evaluate potential differences in infectivity between virus exposed to artificial and natural seawater, LETV preparations were dialyzed against natural seawater obtained from two locations in Florida. LETV preparations dialyzed against seawater obtained from the Sebastian Inlet, Indian River Lagoon (550 mmol/kg) maintained infectivity after

120-hr exposure (Figure 4-6). Similar results were obtained when LETV preparations were dialyzed against seawater from the Whitney Marine Laboratory (1000 mmol/kg).

Negative stain electron microscopy

The stability of LET virions was assessed after exposure to artificial seawater or

HBSS for 24 hr at 23 C. Particles with characteristic morphology of herpesviruses are evident in Figure 4-7 demonstrating the persistence of intact virus particles after artificial seawater exposure.

PCR analysis of LETV exposed to artificial seawater or HBSS

LETV DNA from virus exposed to seawater or HBSS was detected by nested PCR reactions. All samples taken throughout the experiment were PCR positive for herpesvirus DNA (Figure 4-8).

Discussion

LETV was successfully resurrected from the archived stock of the original index case of LETD. Several chelonian cell strains were identified as suitable for cultivation of

LETV and may be targets for the isolation of future marine turtle herpesvirus. Based on studies with LETV, mammalian cell lines routinely used to cultivate other herpesviruses do not support the growth of marine turtle herpesviruses. While mammalian cells cannot be completely disregarded, it appears that cells derived from lower vertebrates are more likely to be useful for the isolation and maintenance of marine turtle herpesviruses. It is difficult to determine if the inability to cultivate LETV in mammalian cells is due to temperature sensitivities of LETV and the cells themselves, or to other factors such as lack of receptors for entry. Mammalian cells thrive at 37 C which is incompatible with

LETV growth. In addition, susceptible chelonian cells do not demonstrate CPE at 37 C when infected with LETV.

Growth studies have revealed that LETV has somewhat altered growth characteristics compared to other herpesviruses. In contrast to many human herpesviruses that quickly grow to high titers, LETV is a slow growing and low titer herpesvirus. In addition, a time course study illustrated the instability of LETV as cultures reached 100% CPE (Figure 4-

2). These studies also demonstrated that 10% of the infectious virus is potentially lost by removing the initial inoculum (data not shown). The inoculum is typically removed to ensure a synchronized infection. However, because LETV is such a low titer virus, the

94 multiplicity of infection (MOI) is also very low (Table 4-1). It is not possible to achieve high enough MOI for a synchronized infection anyway so for all other experiments the inoculum was not removed. In addition, it has been shown that even after cultures reach

100%, a third of the infectious virus is still tightly cell associated as is the case with many gallid herpesviruses (Wilson and Coussens 1991). Overall these studies have indicated that it is best to harvest infected cultures when 70-80% CPE is observed and that the entire culture (cells and medium) should be collected by scraping any remaining attached cells into the culture medium. It has been noted that sonicating this preparation increases virus titer by one log while freezing this preparation reduces the titer by one log (data not shown).

Experiments utilized both artificial and natural seawater and virus infected cell lysates that simulated virally infected cells shed or sloughed from a diseased marine turtle into marine habitats. Neither the source nor the osmolality of seawater influenced the stability of LETV (Figure 4-6). However, the temperature of seawater appeared to have a profound effect on persistence of this virus because the titer declined with increasing temperatures (Figure 4-5). LETV exposed to seawater at 15 or 23 C showed similar levels of persistence while, in contrast, LETV exposed to seawater at 30 C showed a sharp decline in titer after 48 hr (Figure 4-5). Consistent results were obtained in five independent dialysis-exposure experiments at 23 C (data not shown), supporting the conclusion that the persistence of LETV infectivity after exposure to seawater is highly reproducible. Additionally, the control tortoise herpesvirus HV2245 showed similar stability following exposure to seawater (Figure 4-3), suggesting that certain chelonian

95 herpesviruses may share attributes necessary for survival under harsh environmental conditions regardless of ecological origin.

The ability of LETV to retain infectivity in a marine environment is not surprising since herpesviruses are known to be quite stable. Human herpesvirus 1 can persist for eight weeks at ambient humidity and temperature (Mahl and Sadler 1975). The swine herpesvirus, pseudorabies, remained infectious for at least 7 days in the presence of fomites such as whole corn, polypropylene, loam soil, and vinyl rubber (Schoenbaum et al. 1991). These observations suggest that marine herpesviruses, like their terrestrial counterpart, may also be able to survive out of water for extended periods of time on surfaces of equipment, instruments, boats, or in facilities that contact sea turtles infected with herpesviruses.

The stability of other viruses in marine habitats has been previously assessed

(Fujioka et al. 1980;Lo et al. 1976). Most of the research has focused on the potential contamination of marine and estuarine habitats with human enteric viruses as a result of wastewater release. Enteric viruses are of special concern as increasing amounts of sewage waste are dumped into the ocean (Ezzell 1999). If enteroviruses can survive in marine habitats, coastal waters used for recreation and harvest of shellfish may serve as a reservoir and thus become a public health hazard (Fujioka et al. 1980). Human enteric viruses including hepatitis A virus, poliovirus, echovirus, and rotavirus not only persist after exposure to seawater but have also been shown to be taken up by certain aquatic organisms (Goyal et al. 1979;Hejkal and Gerba 1981;LaBelle et al. 1980).

While these experiments have shown that the LETV maintains its infectivity after exposure to seawater in a controlled laboratory setting, it is likely that the extent of

96 herpesvirus persistence in a natural marine environment will vary depending upon the composition of seawater and numerous other environmental factors. The composition of seawater is in a constant state of flux and virus stability will likely be influenced by changes in temperature, currents, depth, sunlight, osmolality, pH, and degree of human activity. Organic material, especially protein, or cellular debris may stabilize viruses and increase survival (Lipson and Stotzky 1984). Suspended sediments, such as clay, may also have a similar effect of stabilizing infectious virus (Clark et al. 1998). In addition to the physical and chemical components, the biological composition of a specific site can negatively or positively influence virus survival. Microbes producing antiviral substances have been documented in marine habitats and have been shown to decrease the amount of time infectious virus persists in the environment (Girones et al. 1989). The presence of higher level of such microbes at one location verses the absence or lower level at another location may account for differences in virus survivability.

Because the temperature of seawater in marine turtle habitats fluctuates throughout the year, virus preparations were dialyzed against artificial seawater at temperatures ranging from 15 C (winter low) to 30 C (summer high) (Herbst et al. 1995). LETV was shown to lose infectivity within 48 hr after exposure to artificial seawater at 30 C suggesting that at higher temperatures herpesviruses shed into seawater may be inactivated more rapidly. It is not known how other chelonian herpesviruses would respond after exposure to seawater of different temperatures. In addition, temperature also affects herpesvirus-induced lesions differently than free viruses in seawater. Lesions in green turtles infected with GPD-associated herpesvirus showed faster onset, development, and increased severity at 30 C (Haines and Kleese 1977). In contrast,

97 green turtles inoculated with filtrates containing FP-associated virus began to develop tumors during the winter months (low 15 C) and continued to develop as the water temperature increased (high 30 C) (Herbst et al. 1995). It is difficult to assess the true effect of water temperature on virus survival and transmissibility since sea turtles traverse numerous marine habitats of varying water temperatures during their life history.

Transmission of mammalian herpesviruses is usually by contact of infected cells in saliva, urogenital excretions, or free virus in aerosols (Fields et al. 1996). For instance, viruses that initiate infection via the skin, eye, oral mucosa, and genital tract include herpes simplex viruses 1 and 2, cytomegalovirus, and Epstein-Barr virus. In addition, varicella-zoster virus, Epstein-Barr virus, cytomegalovirus, Marek’s disease virus, and herpes simplex viruses 1 and 2 may be transmitted via aerosols that infect the respiratory tract (Fields et al. 1990;Beasley et al. 1970).

Many factors contribute to the spread of infectious diseases in animal populations.

Stability of a pathogen outside the host is a critical factor in the epidemiology of a disease. The ability of the pathogen to remain infectious in the environment of its host may facilitate disease transmission. At this time little is known about the mechanism of transmission of herpesvirus infections in marine environments. Herpesviruses from infected turtles may potentially be transmitted to uninfected individuals by direct contact between turtles, by delivery through the egg, by vectors, by contact with virus in sediments or suspended in seawater. These experiments provide the first evidence that disease-associated chelonian herpesviruses retain their infectivity for extended periods of time outside the host in a marine environment. As a result, marine turtle herpesviruses may be transmitted by means other than direct contact between turtles. Shed herpesvirus

98 may be stably maintained in sediments or on food supplies in a marine environment. As a result, common foraging grounds which potentially attract a high density of susceptible hosts may serve as a reservoir for infectious viruses and may facilitate transmission

(Herbst and Klein 1995a).

The results of this study are significant for both researchers studying diseases of marine turtles and various aspects of marine turtle conservation. The data urge caution in the handling of marine turtles. Appropriate disinfection of instruments, tools, hands, and work surfaces that have been in contact with turtles that could harbor herpesviruses is recommended. Turtles should be handled with care to avoid spread of the infectious agent from affected to naïve turtles.

Earliest passage of LETV

Day 0: inoculated 100 mm dishes

10-2: 166 plaques

Day 14: picked 10 plaques

100 10-1 10-2 2-1 2-2 2-3 2-4 2-5 2-6 2-7 2-8 2-9 2-10

Day 10: picked 8 plaques

2-2-1 2-3-1 2-9-2 2-9-1* 2-3-2 2-9-4 2-9-3* 2-9-5*

Day 10: harvest Po

5 clones: 10 ml supernatant & 1 ml cell pellet in 1.0 PBSA Figure 4-1. Schematic of plaque purification of LETV. The earliest passage of LETV was twice plaque purified. Ten clones were picked from the first round of purification and then were plated for a second round of purification. A total of 7 plaques were picked off of 3 plates: 1 plaque from the 2-2 plate, 2 plaques from the 2-3 plate, and 5 plaques from the 2-9 plate. Each of these clones were expanded on 100 mm plate. Of the 8 clones, 5 were successfully expanded and were labeled P0. Three clones were lost to fungal contamination and are indicated with an asterisk.

100

1.0E+06 120

100

1.0E+05 80

60 Percent CPE

Virus Titer (pfu) 1.0E+04 40

1.0E+03 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Days

Figure 4-2. Titer of LETV over time compared to percent cytopathic effect. Virus was harvested and titered (solid squares), and the percent cytopathic effect (open circles) was monitored over the course of 16 days.

101

10.0 9.0 8.0 7.0 /ml)

50 6.0 5.0 TCID 4.0 10 Virus titer 3.0 (log 2.0 1.0 0.0 0 24 48 Time (hours)

Figure 4-3. Persistent infectivity of LETV after exposure to artificial seawater. Titer of LETV and tortoise herpesvirus (HV2245) after exposure to artificial seawater or Hank’s balance salt solution (HBSS) at 23 C. Titer of LETV exposed to artificial seawater (blue bars) and HBSS (black bars). Titer of HV2245 after exposure to artificial seawater (white bars) and after exposure to HBSS (gray bars).

102

6.0 1200

5.0 1000

/ml) 4.0 800 50 3.0 600 TCID 10 Virus Titer 2.0 400 (log

1.0 200 Osmolality (mmol/kg)

0.0 0 0 24 48 72 96 120 Time (hours)

Figure 4-4. Persistent infectivity of LETV after exposed to artificial seawater for 120 hr at 23 C. Titers of LETV exposed to artificial seawater (blue bars) and Hank’s balance salt solution (HBSS; black bars) are demonstrated. Osmolality of LETV preparation exposed to artificial seawater (open triangles) and to HBSS (open circles) are also indicated.

103

4.5 4.0 3.5

/ml) 3.0 50 2.5

TCID 2.0 10 Virus Titer 1.5

(log 1.0 0.5 0.0 0 24 48 72 96 Time (hours)

Figure 4-5. Persistent infectivity of LETV after exposure to artificial seawater at three temperatures. LETV exposed to artificial seawater (solid line) or Hank’s balance salt solution (dashed lines) at 15 C (triangles), 23 C (circles) and 30 C (squares).

104

7.0 6.0 5.0 /ml) 50 4.0

TCID 3.0 10 Virus Titer 2.0 (log 1.0 0.0 0 24 48 72 96 120 Time (hours)

Figure 4-6. Persistent infectivity of LETV after exposure to natural seawater. LETV was exposed to artificial seawater (open squares), Hank’s balance salt solution (HBSS; open triangles), and natural seawater from Sebastian Inlet, Florida (open circle) at 23 C.

105

100 nm 200 nm

200 nm 200 nm

Figure 4-7. Negative stain electron microscopy images of LETV after 24-hour exposure to artificial seawater. The top panels illustrate LETV virions after exposure to Hank’s balance salt solution at 23 C. The bottom panels illustrates LETV virions after exposure to artificial seawater at 23 C.

106

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

2,000 1,000 500 300 200

Figure 4-8. Persistence of the LETV DNA polymerase gene after exposure to artificial seawater. Lane 1 and 15: molecular weight marker. Lane 2: VZV DNA. Lane 3: no template negative control. Lane 4: original LETV preparation (t=0) used for the dialysis experiments. Lanes 5, 7, 9, 11, and 13: LETV preparations after t= 24, 48, 72, 96, and 120 hr exposure to artificial seawater at 23 C, respectively. Lanes 6, 8, 10, 12, and 14: LETV preparations after t=24, 48, 72, 96, and 120 hr exposure to Hank’s balance salt solution at 23 C, respectively.

107

Table 4-1. Net yield of LETV after three serial passages.

Passage Inputa Pre-Titerb MOIc Harvestd Post-Titere Netf

2 2 4 5 P0 6.9 X 10 4.6 X 10 0.0001 13 5.4 X 10 4.9 X 10

4 4 5 5 P1 8.1 X 10 5.4 X 10 0.02 7 1.1 X 10 9.7 X 10

5 5 4 6 P2 1.7 X 10 1.1 X 10 0.03 7 3.1 X 10 1.1 X 10

4 4 5 6 P3 4.7 X 10 3.1 X 10 0.009 10 2.5 X 10 2.2 X 10

aTotal plaque forming units (pfu) added to cells. bAmount of virus added in pfu per milliliter (ml) to cells. c Multiplicity of infection equal to pfu added to five million cells. dNumber of days post-infection harvested LETV-infected cells. eAmount of virus recovered from infected cells (pfu/ml). fNet amount of virus (pfu) equal to total pfu post-infection minus input pfu.

CHAPTER 5 DEVELOPMENT AND APPLICATION OF SEROLOGICAL ASSAYS FOR DETECTING EXPOSURE OF FLORIDA MARINE TURTLES TO A DISEASE- ASSOCIATED HERPESVIRUS

Introduction

Demonstrating an etiological role for marine turtle disease-associated herpesviruses is limited either by the lack of access to marine turtle life stages with clinical disease

(LETD), or the inability to cultivate the associated herpesvirus (FPHV). Lung-eye- trachea disease (LETD) occurs during life stages of the turtle that are both pelagic and disperse. As a result, the chance of encountering turtles with clinical signs of LETD is highly unlikely. Furthermore, although FP has reached epizootic proportions in some wild marine turtle populations, the relationship between infection with FPHV and development of FP in wild populations has not been studied because FPHV has not been isolated in culture to allow fulfillment of Koch’s postulates. As a result of these limitations, focus has been placed on developing serological assays to provide information on the prevalence of infection with these viruses in wild turtles. The hope is that the information collected through seroepidemiology might shed light on the etiological role of these marine turtle herpesviruses in their respective diseases.

This chapter describes the development of an enzyme-linked immunosorbent assay

(ELISA) for the detection of specific antibodies to the marine turtle herpesvirus, LETV.

The rationale for this effort was to examine whether LETV infections occur in wild marine turtle populations and also to develop an assay capable of distinguishing LETV from FP-associated herpesvirus infections. The LETV-specific ELISA assay was applied

108 109 to a large population of free-ranging juvenile green turtles at three study sites in Florida to assess the exposure of these turtles to LETV. In addition, the relationship between FP and LETV antibodies was investigated further by analyzing the influence of tumor severity and duration of tumor status on the presence of antibodies to LETV.

Materials and Methods

Cultured Cells

A 1.5 cm skin biopsy was taken from the ventral surface of the rear flipper of a 1 year-old captive-reared green turtle, 99A-1, and was digested in 10 ml collagenase (260 units/ml) and 2.5% w/v trypsin in Hank’s balanced salt solution (HBSS). The digestion medium was collected and cells were pelleted at 500 g for 10 min at 4 C. Cells were resuspended in Dulbecco’s Modified Eagle Medium (DMEM/F12) supplemented with

5% v/v fetal bovine serum (FBS), penicillin/streptomycin (2 mg/ml), gentamicin (60 mg/ml), and fungizone (1 mg/ml). The cell suspension was added to 25 cm2 vented flasks and grown in a 5% CO2 humidified incubator at 28 C. The 99A-1 primary cell strain was maintained and passaged in continuous culture. The D-1 green turtle fibroblastic cell strain was previously developed (Herbst et al. 1998c). Terrapene heart cell monolayer lines (TH-1; ATCC No. CCL 50) were also used.

LETV Antigen Preparation

Plaque purified LETV clones 292 and 221 (Chapter 4) were propagated in TH-1, D-1, or 99A-1 cells grown in DMEM/F12 supplemented with 5% v/v FBS and antibiotics in vented flasks in a 5% CO2 humidified incubator at 28 C. Monolayers with 75-80% cytopathic effect (CPE) were scraped into medium and stored frozen. The cell suspension was thawed, sonicated for 60 sec, and clarified by centrifugation at 2500 g for 20 min at 4

C. Virus was pelleted from the supernatant by ultracentrifugation at 40,000 g for 1 hr at 4

110

C. Virus pellets were resuspended in phosphate buffered saline (PBS) and repelleted under the same conditions. The washed virus pellet was resuspended in 2-4 ml PBS and sonicated for 60 sec. Virus titers (plaque forming units/ml) were determined by plaque assay as described in Chapter 4. Uninfected cells were scraped into medium, pelleted, resuspended in PBS, and sonicated to serve as uninfected cell lysate controls. Protein concentrations of virus and uninfected cell controls antigens were measured by Bradford assay (Pierce, Rockford, IL).

Development of Control LETV Antisera

Two juvenile captive-reared green turtles, 99C-1 and 99A-1, received subcutaneous immunizations with inactivated LETV preparations to produce anti-LETV antibodies to serve as reagent controls. Antigen was prepared as described above for LETV clones 292 and 221 and then inactivated by dialysis against 0.02% v/v formaldehyde at 30 C for 48 hrs (Laufs and Steinke 1976). Residual formaldehyde was removed by dialysis against 3 changes of PBS for 24 hrs. Inactivation was confirmed by plaque assay as described above. Turtle 99C-1 was immunized with LETV clone 292 grown in D-1 cells, and turtle

99A-1 was immunized with LETV clone 221 grown in 99A-1 cells to minimize development of antibodies to cellular antigens. Each turtle received 5 immunizations every 2 weeks with 1 X 104 plaque forming units/injection plus an equal volume of 2X

Ribi’s adjuvant (Sigma, St. Louis, MI).

Development of Anti-LETV Monoclonals

To generate monoclonal antibodies to LETV, three Balb/c mice (age 6-8 weeks) were immunized with inactivated LETV. Each mouse was inoculated subcutaneously with100 ml of inactivated LETV containing 4000 plaque forming units/injection plus an equal volume of 2X Ribi’s adjuvant (Sigma, St. Louis, MO). Additional injections with

111 inactivated LETV were given two weeks and then 1 month post-immunization with the same LETV plus adjuvant preparation. A booster immunization were repeated 2 months post-immunization with 100 ml live LETV plus an equal volume of 2X Ribi’s adjuvant.

The final booster was given 3 months post-immunization with 100 ml live LETV without adjuvant. Test bleeds were drawn throughout the course of the immunizations and the serum was assayed for the presence of antibodies to LETV by ELISA as described below.

The spleen was harvested from an immunized mouse three days after final immunization and monoclonal antibodies were prepared as previously described (Liddell and Cryer

1991;Simrell and Klein 1979). Briefly, splenocytes were fused with SP2/0 myeloma cells and were plated in 96 wells with fusion medium (DMEA(GL), 1X HAT, 25% Spo/2 conditioned medium, and 20% horse serum; Invitrogen, Carlsbad, CA). Each well was monitored microscopically for growth of hybridomas. The supernatant was monitored for the secretion of antibodies by ELISA (described below) with LETV 292 clone grown in

D-1 cells as antigen. Wells containing cells secreting antibody to LETV with low anti- cellular background were identified by ELISA and the cells were transferred to 24 well plates for expansion. The supernatant was monitored for production of antibodies to

LETV by ELISA and antibodies were isotyped (ISOStrip Kit; Roche Diagnostic

Corporation, Indianapolis, IN). Cells producing high levels of IgG anti-LETV antibodies with low cellular background were further evaluated by ELISA and Western blot analyses. Two hybridomas were further cloned, 2A4 and 4C11.

Green Turtle Plasma Samples

Plasma from 42 juvenile captive-reared green turtles (healthy turtles with no known exposure to herpesviruses) and plasma collected from 30 juvenile captive-reared green

112 turtles after experimentally induced fibropapillomatosis (Herbst et al. 1995) were evaluated by the serological assays described below (9 to 11 month-old and 21 to 23 month-old, respectively). Plasma samples collected from 27 juvenile wild FP positive turtles and 19 juvenile wild FP negative green turtles netted in the Indian River Lagoon

(Indian River County FL USA; 27º49’57’’N, 80º26’18’’W), and along the Sebastian Inlet

(Indian River County FL USA; 27º47’ 38’’N, 80º24’34’’W) were also included (Herbst et al. 1998b). All of these plasma samples had been previously evaluated by immunohistochemistry for antibodies to FPHV (Herbst et al. 1998b). Plasma from all FP positive green turtles contained anti-FPHV antibodies and all FP negative green turtles did not. Plasma from green turtle 99C-1 pre- and 2 month post-immunization with LETV served as the negative and positive LETV antibody controls, respectively.

A total of 180 plasma samples were obtained from juvenile green turtles netted in the

Indian River lagoon, along the reef of the Sebastian Inlet, or in the Trident Basin (Cape

Canaveral, Brevard County, Florida, USA). Blood was drawn from the dorsal cervical sinus and was collected in lithium heparin-treated vacutainer tubes. The blood samples were transported on ice to the laboratory where the samples were then centrifuged at maximum speed in an Adams Physicians Centrifuge (Becton Dickinson, Sparks, MD) for

10 minutes. Plasma was collected and stored at – 20 C. Of the 60 plasma samples collected from each study site, twenty were collected in each of the years 1997, 1998, and

1999. Plasma samples were selected to prevent sampling an individual turtle at more than one location or from more than one year. A random number generator was applied to the database blinded for FP status to select 20 plasma samples from each site for each year.

113

A separate survey tested 133 plasma samples collected from January 1998 to

December 1999 from green turtles netted in the Indian River lagoon. These green turtles were scored for fibropapilloma tumor severity (Hirama and Ehrhart in press). Green turtles were assigned to a tumor severity score ranging from 0 (no tumor) to 3 (most severe) as previously described (Balazs and Pooley 1991). Plasma samples from green turtles of each tumor severity score were randomly selected from those available in the plasma bank.

Plasma samples collected between 1995 and 1999 from 16 individual wild green turtles captured more than once (recaptures) were also tested by ELISA. Tag numbers allowed individuals to be identified from year-to-year. In addition, plasma collected from a small group of adult nesting female turtles were tested for the presence of antibodies to

LETV. Samples from 9 green turtles and 4 loggerhead turtles nesting on Melbourne

Beach (Indian River County, Florida, USA) were assayed by ELISA. In all assays, plasma from two captive-reared green turtles, 99C-1 and 99A-1, prior to and after immunization with inactivated LETV served as positive and negative anti-LETV antibody controls (Coberley et al. 2001).

Development of LETV ELISA

Each well of a 96 flat bottom-well microtiter plate (Nunc-Immuno MaxiSorp, Nunc,

Kamstrup, Denmark) was filled with 50 ml of 292 LETV or uninfected cell lysate control, diluted in sodium phosphate buffer (pH 7.2) containing 0.15 M NaCl and 0.02% NaN3

(PBS/A) to concentration of 5 mg/ml, and then covered with sealing tape and incubated overnight at 4 C. The wells were washed 4 times with 250 ml PBS/A containing 0.05%

Tween 20 (PBS/T) using an automatic microplate washer (EAWII, SLT-Labinstruments,

114

Salzburg, Austria), and then blocked overnight with 300 µl per well of PBS/T containing

5% w/v non-fat dry milk (PBS/TM) at 4 C. After 4 more washes, 50 µl of green turtle plasma samples diluted 1:25 in PBS/TM or negative or positive LETV antibody control plasma diluted 1:100 in PBS plus 2% v/v FBS (PBS/F) were added to the appropriate wells. For titration experiments, plasma samples were two-fold serially diluted. Each plasma sample was incubated with 292 LETV and uninfected cell lysate control for 1 hour at room temperature on a Nutator (Becton Dickinson, Sparks MD). The microplate was washed as described above and 50 ml of biotinylated monoclonal HL858 (anti-green turtle 7S IgG) (Herbst and Klein 1995b) diluted in PBS/A to a concentration of 1 mg/ ml was incubated 1 hour at room temperature on a Nutator. After washing, 50 µl of alkaline phosphatase-conjugated streptavidin (Zymed Laboratories Inc., San Francisco CA) diluted 1:5000 in PBS/A was added to each well. The microplate was incubated and washed as described above, and then 100 µl of p-nitrophenyl phosphate disodium (1 mg/ml in 0.01 M sodium bicarbonate pH 9.6, 2 mM MgCl2) was added to each well. The microplate was incubated in the dark without agitation at room temperature. The A405 of each well was measured after a 2 hour incubation in an ELISA plate reader (Spectra II,

SLT-Labinstruments, Salzburg, Austria with DeltaSoft version 3.3 software,

Biometallics, Princeton NJ). Optical density values were adjusted as follows. For each plasma sample, the optical density of the well coated with uninfected cell lysate was subtracted from the optical density of the well coated with virus to account for non- specific absorbance. Optical densities equal to or less than zero were adjusted to 0.001 for statistical analysis.

115

Statistical Analyses

Statistical analyses were conducted using SigmaStat 2.03 software (SPSS

Incorporated, Chicago IL). The Kolmgorov-Smirnov normality test was applied to optical density value data measured by ELISA from plasma of captive-reared FP negative, captive-reared FP positive, wild FP positive, and wild FP negative. Because many of the groups were non-normally distributed the following statistical analyses were used. The

Mann-Whitney rank sum test was used to compare captive-reared FP positive to captive- reared FP negative turtles, and to compare wild FP positive turtles to wild FP negative turtles. The Kruskal-Wallis one-way ANOVA on ranks was used to compare all of the groups. To identify the group or groups that differ from the others, all pairwise multiple comparison procedures were performed using Tukey test or Dunn’s method. For estimating seroprevalences, a cutoff value was based on the optical density values obtained from the 42 healthy captive-reared green turtles. The cutoff value was calculated as the highest optical density value obtained from these healthy turtles plus three times the standard deviation (cutoff value=0.310). Chi-square analysis was used to compare the relative frequencies of seroprevalence.

Immunoblotting Protocol

The specificities of the polyclonal anti-LETV antibodies from the LETV-immunized green turtles and plasma from FP positive and FP negative turtles seropositive for anti-

LETV antibodies were tested by Western immunoblotting. Clone 292 LETV and uninfected cell lysate control antigens were separated by using polyacrylamide gel electrophoresis (PAGE) under denaturing conditions with a precast 10% w/v polyacrylamide bis-tris gel and MOPS running buffer (NuPAGE, Novex, San Diego,

CA). The separated proteins were electroblotted from the gel onto a nitrocellulose

116 membrane, and then the membrane was blocked overnight with PBS/TM at 4 ºC. The blocked blot was washed in PBS/T, and then incubated with green turtle plasma at a 1:25 dilution in PBS/TM or LETV control plasma diluted 1:100 in PBS/F for 60 min at room temperature with rocking in a manifold (Pierce, Rockford IL). After washing, the blot was removed from the manifold and incubated with HL858 as described for 60 min at room temperature on a rocker. After washing the membrane, it was incubated in alkaline phosphatase-conjugated streptavidin as described for 60 min at room temperature on a rocker. After final washing, the blot was developed in substrate buffer (0.1 M Tris HCl pH 8.8, 1 mM MgCl2) containing nitroblue tetrazolium chloride and 5-bromo-4-chloro-3- indolylphosphate p-toluidine salt (Promega, Madison, WI).

Immunohistochemistry

TH-1 cells were cultured as described above in 2 well glass chamber slides (Nalgene

Nunc, Naperville, IL) and grown to confluency. One well was infected with LETV 292 and after 40-50% CPE was observed, cells in the infected and the uninfected wells were fixed with 2% w/v paraformaldehyde in phosphate buffer (0.12M NaH2PO4H20, 0.28M

Na2HPO4, pH 7.2), permeabilized with 0.5% TritonX 100, washed with PBS, and then air dried. On the day of the assay, slides were rehydrated in PBS and endogenous peroxidases were inactivated by adding 3% hydrogen peroxide for 30 min. Slides were blocked with PBS plus 2 % v/v horse serum (PBS/H) for 20 minutes at room temperature in a moist chamber. Then, 100 ml of green turtle plasma diluted 1:25 in PBS/TM, or

LETV control plasma diluted 1:100 in PBS/F was added to each well. The slides were incubated overnight at 4 C in a moist chamber. Each slide was washed individually in 3 changes of PBS for 10 min each. For the following incubations, 100 ml of each reagent

117 was added to each well and the slide was incubated for 45 min at room temperature in a moist chamber. Unbiotinylated monoclonal antibodies HL858 anti-green turtle 7S IgG

(Herbst and Klein 1995b) and HL673 anti-tortoise Ig light chain (Schumacher et al.

1993) at 1 mg/ml in PBS/H were added to each well. Slides were washed as before and then 2% v/v universal anti-IgG-conjugated biotin in PBS (Vectastain ABC kit, Vector

Laboratories Inc, Burlingame, CA) was added. The slides were washed and then incubated with horseradish peroxidases-conjugated streptavidin. After washing, slides were incubated in Coplin jars containing chilled chromogen (0.3 mg/ml 3,3- diaminobenzidine tetrachloride in PBS) and monitored microscopically for color development relative to the positive and negative LETV antibody controls. Using standard procedure, slides were counterstained with hematoxylin, dehydrated, and mounted with coverslips (Coligan et al. 2001). Stained slides were viewed with an

Olympus BX50F microscope and photographed with an Olympus digital camera.

Results

Detection of Seroconversion by ELISA

Positive control antisera to LETV were generated by immunizing two captive-reared juvenile green turtles, 99A-1 and 99C-1, with inactivated LETV clones 221 and 292, respectively. The ELISA was able to detect antibodies to LETV in the plasma of both turtles as early as 1-2 months post-immunization (Figure 5-1). Anti-LETV was still detectable at least 4.5 months after the last immunization. Plasma from 99C-1 before and

2 months after LETV immunization served as negative and positive controls, respectively, for all future experiments. To assess the interassay reproducibility of the

ELISA assay, the mean A405, standard deviation and coefficient of variation for the negative and positive reagent controls of 10 ELISAs were calculated. The positive

118

control (mean A405=1.874, SD=0.186) with a coefficient of variance value equal to 9.9% demonstrated the reproducibility of the assay. A CV value <10% is considered excellent

(Crowther 2001). The negative control (mean A405= 0.007; SD=0.015) had a CV value

(SD/ mean A405 X 100) equal to 210 since both the mean A405 and standard deviation were such small values.

Development of LETV Monoclonals

Three mice were successfully immunized to generate control monoclonal anti-LETV antibodies for various serological assays. Fusion supernatants were evaluated by ELISA, immunohistochemistry, and Western blot analyses and the results are summarized in

Table 5-1. Hybridoma 2A4 was selected for cloning and monoclonal antibody purification for use as a positive control for LETV in the ELISA format.

Survey of Captive-reared Green Turtles for Antibodies to LETV

Plasma samples from captive-reared green turtles without experimentally induced FP were tested in the ELISA format (Figure 5-2). Optical density values for these plasma samples were very low (mean A405 = 0.020; SE=0.006). To evaluate potential cross- reactivity between anti-FPHV antibodies and LETV antigens, plasma from 30 turtles with experimentally induced FP and anti-FPHV antibodies previously detected by immunohistochemistry (Herbst et al. 1998) were tested by ELISA (Figure 5-2). The optical density values for these plasma samples were also very low (mean A405=0.020;

SE=0.004). Both groups were non-normally distributed (P<0.001). There was no difference in ELISA values between plasma from the turtles with or without experimentally induced FP (Mann-Whitney test; T=1238.0; P value =0.153). Based on the cutoff value of 0.310, the green turtles with experimentally induced FP were seronegative for antibodies to LETV (Figure 5-2).

119

Survey of Previously Evaluated Wild Green Turtles for Antibodies to LETV

The ELISA assay was then applied to plasma previously tested for FPHV antibodies

(Herbst et al. 1998b) from free-ranging FP positive (n=27) and FP negative (n=19) green turtles to assay for the presence of anti-LETV antibodies. In contrast to the captive-reared turtles, ELISA values were higher from plasma collected from both FP positive (mean

A405 = 0.192, SE=0.041) and FP negative (mean A405 =0.143; SE=0.027) wild green turtles (Figure 5-2). The wild FP negative plasma was normally distributed (P>0.200), but the wild FP positive plasma was not normally distributed (P<0.001). There was no difference between ELISA values of plasma from FP positive and FP negative turtles

(Mann-Whitney rank; T=425.5; P=0.647) demonstrating the lack of a correlation between

FP status of wild green turtles and ELISA values measured to LETV. Both of the wild groups were different from the captive-reared turtles (Dunn’s method; P<0.05). Based on the cutoff (Figure 5-2), 10.5% of the FP negative and 14.8% of the FP positive green turtles were seropositive for antibodies to LETV. There was no statistically significant correlation between LETV seroprevalence and FP status (Chi-square = 0.195, P > 0.50).

The seroprevalence overall for Florida green turtles tested was 13%. These results were confirmed by Western blot and immunohistochemistry.

Confirmation of Anti-LETV Antibodies in Green Turtle Plasma

Plasma samples from FP positive and FP negative wild green turtles that were LETV seropositive by ELISA were titered and tested by two complementary serological assays,

Western blot and immunohistochemistry. The anti-LETV antibodies displayed a normal titration pattern reflecting their specific interaction with LETV antigen in the ELISA

(data not shown). Plasma samples that were LETV seropositive by ELISA also reacted with LETV infected cell preparations in a Western blot format (Figure 5-3). Antibodies in

120 plasma from LETV immunized, FP positive, and FP negative turtles seropositive by

ELISA recognized similar bands in the LETV infected cells. Although antibodies in plasma bound to a number of bands, the immunodominant bands were approximately 38 and 19 kDa (Figure 5-3, arrowheads). Faint background or non-specific binding to various proteins in the uninfected cell lysate control, containing medium and serum proteins, was also noted. When assayed by immunohistochemistry, the plasma containing

LETV antibodies, as measured by ELISA, reacted specifically with LETV infected cells in the zones of cytopathic effect and not with uninfected cells (Figure 5-4).

Survey of Green Turtle Populations at Three Florida Study Sites

Plasma samples from green turtles netted in the Indian River lagoon, along the

Sebastian reef, and in Trident Basin were tested for the presence of anti-LETV antibodies by ELISA (Figure 5-5a). Anti-LETV antibodies were detected in plasma collected from the Indian River lagoon (mean A405 =0.248; SE=0.031; n=60), Sebastian reef (mean A405

=0.167; SE=0.020; n=60) and Trident basin (mean A405 =0.208 ; SE=0.046; n=60 (Figure

5-5a). All groups were non-normally distributed (P£0.002). While anti-LETV antibodies were detected in some individuals from all three study sites, plasma samples collected from turtles in the Indian River lagoon had statistically higher ELISA values compared to the Sebastian reef and Trident basin (Tukey test; P<0.05). The estimated seroprevalence was 21.6% for the Indian River lagoon, 10% for the Sebastian reef, and 18.3% for the

Trident basin. Data from each year (1997, 1998, 1999) at each study site were combined to examine if there was a difference in the level of anti-LETV antibodies over time in

Florida (Figure 5-5b). All groups were non-normally distributed (P£0.001). There was no statistical difference in the ELISA values obtained from Florida green turtle plasma

121 samples collected in 1997, 1998 and 1999 (Kruskal-Wallis ANOVA; P=0.393; n=60 per year). The presence of antibodies to LETV in selected plasma samples with ELISA optical density values above the cutoff was confirmed by Western blot analysis (data not shown).

Survey of Green Turtles Scored for FP Tumor Severity

To explore the relationship between FP and anti-LETV antibodies, plasma from green turtles scored for FP tumor severity were tested (Figure 5-6). Antibodies to LETV were detected by ELISA in the plasma of wild green turtles from score 0 (mean A405 =0.238;

SE=0.038;n=36), score 1 (mean A405 =0.228; SE=0.043; n=35), score 2 (mean A405

=0.237; SE=0.032; n=36), and score 3 (mean A405 =0.390; SE=0.071; n=26). All groups were non-normally distributed (P£ 0.010). There was no correlation between FP tumor severity and the presence of antibodies to LETV measured by ELISA (Kruskal-Wallis

ANOVA; P=0.073). The presence of antibodies to LETV in selected plasma samples with

ELISA optical density values above the cutoff was confirmed by Western blot analysis

(data not shown).

Survey of Green Turtles Recaptured over Time

To examine changes in the level of antibodies to LETV in an individual turtle over time, plasma from green turtles captured more than once (recaptures) were tested by

ELISA (Table 5-2). Plasma samples from turtles at first capture were normally distributed (P>0.200), but plasma samples from these turtles at last capture were not normally distributed (P<0.001). There was no statistical difference between anti-LETV antibody levels measured by ELISA in plasma collected from an individual turtle the first time captured and the last time captured (Mann-Whitney rank; T=269.5; P = 0.851; n=16). Plasma samples from green turtles were divided among the following categories

122 relative to their FP status: FP positive, FP negative, regressors (tumor status change from

FP positive to FP negative), progressors (tumor status change from FP negative to FP positive), and other (changing status) (Table 5-2). There was no clear relationship between FP status and anti-LETV antibody optical density values measured in the

ELISA.

Summary of ELISA Tests on Green Turtle Plasma Samples

To estimate the seroprevalence of LETV in free-ranging juvenile green turtles on the east coast of Florida, optical density values obtained from all plasma samples (n=329) screened by ELISA in this study were combined (Figure 5-7). The distribution of ELISA values from plasma of wild green turtles was compared to the distribution of the 42 healthy, capture-reared juvenile green turtles (mean A405 =0.020; SE=0.006). A cutoff value was calculated as the highest optical density value obtained from these healthy turtles plus three times the standard deviation (cutoff value=0.310). Based on this cutoff value, 71 out of the 329 wild Florida green turtles tested were seropositive for LETV antibodies (seroprevalence = 21.6%).

Survey of Adult Nesting Female Turtles

Plasma from adult nesting females were tested for the presence of anti-LETV antibodies by ELISA. Antibodies to LETV were detected in plasma of both green (mean

A405 =2.114, SE=0.356, n=9) and loggerhead (mean A405 = 1.614, SE=0.787, n=4) turtles

(Figure 5-8). Selected green and loggerhead plasma samples with the ELISA optical density values above the cutoff were tested by Western blot (Figure 5-9). Plasma samples reacted with proteins in the LETV infected cell lysates not present in the uninfected cell lysate controls.

123

Discussion

This is the first ELISA developed for a specific herpesvirus infection of marine turtles.

Immunization of two captive-reared green turtles with inactivated LETV produced an

ELISA detectable antibody response against LETV within one to two months (Figure 5-

1). The temporal pattern of this antibody response was similar to that obtained from previous immunization experiments conducted in green turtles using dinitrophenol (DNP) conjugated bovine serum albumin (Herbst and Klein 1995b).

A monoclonal antibody (2A4) was developed to serve as an additional anti-LETV positive control for serological tests (Table 5-1). Screening of the hybridoma fusion supernatants revealed major differences in the reactivity of these monoclonal antibodies with LETV presented in the ELISA, Western blot, or immunohistochemistry format

(Table 5-1). This was consistent with earlier observations on the variable reactivity of anti-herpesvirus antibodies in turtle plasma with LETV in different assay formats

(Chapter 3). None of the evaluated monoclonal antibodies reacted with bands similar to the molecular weight of proteins presumed to be immunodominant (e.g.,19 and 38 kDa proteins).

Assay of plasma collected from captive-reared and wild green turtles illustrated the absence of a correlation between the presence of anti-FPHV antibodies and reactivity with LETV in the ELISA. The 30 captive-reared green turtles with experimentally induced fibropapillomatosis (Figure 5-2) were of particular interest since they were raised in captivity with no known exposure to herpesviruses and no clinical herpesvirus disease prior to their use in FP transmission studies. All 30 turtles with anti-FPHV antibodies

(Herbst et al. 1998b) were seronegative for LETV. Thus, in contrast to the immunohistochemical format, in which the plasma from some of these FP positive turtles

124 with anti-FPHV antibodies reacted with LETV (presumably through cross-reactive

FPHV/LETV antigens) (Chapter 3) the ELISA is specific for the detection of 7S IgG anti-LETV antibodies.

There was no correlation between FP status (anti-FPHV antibodies) and reactivity with LETV in the ELISA among the previously evaluated wild green turtles (potentially exposed to many herpesviruses) in Florida (Figure 5-2). This further supports the hypothesis that these two viral infections are independent of one another. Antibodies to

LETV detected by ELISA and confirmed by Western blot (Figure 5-3) and immunohistochemistry (Figure 5-4) in plasma of the wild green turtles (both FP positive and negative) indicated that these turtles were exposed to LETV or an antigenically related LETV-like virus. In the case of the FP positive turtles, antibodies to LETV suggested co-infection.

Antibodies to LETV were detected in the plasma of some green turtles at each study site: the Indian River lagoon, the Sebastian reef, and the Trident basin. Interestingly, the estimated seroprevalence and levels of anti-LETV antibodies were slightly higher for green turtles in the Indian River lagoon (Figure 5-5a). It is known that, compared to other nearby sites, green turtles in the Indian River lagoon have a higher prevalence of fibropapillomatosis which is associated with the herpesvirus, FPHV (Lackovich et al.

1999). For example in 1998, a survey of green turtles in the Indian River lagoon showed a FP prevalence of 72.5% (Llewellyn Ehrhart, personal communication) compared to the

Trident basin where FP has never been documented (Dean Bagley, personal communication). These data suggest that conditions in the Indian River lagoon may facilitate the transmission of infectious diseases and even increase disease severity. These

125 factors may include the presence of vectors for transmission, the presence of virus stabilizing conditions, and favorable water temperature (Curry et al. 2000;Herbst and

Klein 1995a).

There was no obvious change in LETV antibody levels in the samples tested over a three-year study period suggesting that LETV exposure, in general, was not increasing

(Figure 5-5b). This was further supported by data obtained from ELISA assays of plasma from recaptured individuals tested over time (1995-1999). There was no statistical change in optical density values to LETV from the first time a turtle was captured compared to last time it was captured (Table 5-2). Thus LETV is not likely to represent a new epizootic at these sites but may instead be part of an endemic or sporadic infection. This is similar to the pattern of endemic herpesvirus infections in other vertebrates (Fields et al. 1996).

There appears to be no relationship between FP tumor severity or status and the presence of anti-LETV antibodies as detected by ELISA. Antibody optical density values were similar in all tumor severity score categories (Figure 5-6). Although score 3 category optical density values were slightly higher on average, the two extremes, score 0

(no tumor) to score 3 (most severe FP), were not statistically different from each other.

One concern when testing for the presence of LETV antibodies and the association of

LETV exposure with FP is that while a tumor may be grossly visible, the turtle may not yet have mounted a detectable immune response to the virus. It is assumed that a turtle with a tumor severity score of 3 is in the advanced stage of the disease, has been afflicted with FP for a prolonged duration, and would have ample time to develop an anti-FPHV antibody response capable of cross-reacting with LETV or to develop an anti-LETV

126 response if LETV was associated with the disease, FP. It is known that green turtles with experimentally induced fibropapillomatosis develop detectable antibody responses to

FPHV within 1 year post inoculation (Herbst et al. 1998b) suggesting that a wild turtle with tumor for a year or more would likely have mounted an immune response to the etiologic agent. It was shown that turtles with FP for up to 3 years had very low optical density values to LETV (BP5603; Table 5-2) again supporting the hypothesis that it is unlikely that LETV has a role in FP. Instead, detection of antibodies to LETV in an FP positive turtle is likely due to co-infection with LETV or a LETV-like virus.

When few known-positive plasma samples are available, a cutoff value based on the distribution of a known-negative population can be used to classify unknown samples as potentially positive or negative (Crowther 2001). In this study, a cutoff value (optical density=0.310) based on 42 healthy captive-reared green turtles was used to calculate an estimated seroprevalence of LETV at a Florida study site. Based on this cutoff, a larger survey of plasma samples from 329 wild green turtles from three study sites in Florida revealed a seroprevalence of 21.6% (Figure 5-7). However, it is difficult to determine, using this test alone, what is the true extent of LETV exposure in the wild. Transmission studies with infectious LETV or identification of naturally infected turtles diagnosed with

LETD are needed to provide additional positive reference plasma to refine this test further, and to provide information about the relationship between LETV infection, the development of class-specific (IgM, 7S IgG, and 5.7S IgG) antibody responses, and the clinical course of the disease, LETD.

Preliminary examination of plasma collected from 13 nesting green and loggerhead turtles revealed the highest optical densities measured to LETV by ELISA (Figure 5-8).

127

The seroprevalence in this small sample was 100% for the green and 75% for the loggerhead nesting turtles and the presence of anti-LETV antibodies was confirmed by

Western blot (Figure 5-9). The implications of this data for turtle populations are unknown at this time, but suggest widespread exposure. One possibility is that as an adult turtle, nesting females have a constant high level of antibodies to LETV. This would be similar to human and other mammals in which titers to herpesviruses persist for their lifetime (Collier 2000;Fields et al. 1996;Hoang et al. 1999). For example, >95% of the adult human population have relatively high levels of antibodies to varicella-zoster virus, human herpesvirus 6 and 7, and Epstein-Barr virus as a result of chronic exposure and latent infection with these herpesviruses throughout their lifetimes (Hoang et al. 1999). If this were the case for turtles, then an estimated seroprevalence of 21.6% in juveniles would indicate that at this life stage marine turtles are at the beginning of their exposure to LETV, and as the turtle population ages seroprevalence would increase. Another possibility is that the stress or hormonal changes associated with nesting causes recrudescence of virus shedding in infected females causing a “booster” effect on the antibody titer. Alternatively, it may be that prior to nesting there may be a hormone- mediated increase in antibody levels to the entire spectrum of antigens/pathogens that turtles have encountered in their lifetime in preparation for vertical transfer of these antibodies to offspring through the egg. Because all of the adult samples are from nesting females a hormonal effect or sex differences cannot be ruled out and requires further study. However, one nesting loggerhead turtle had a low optical density reading for

LETV (Figure 5-8) suggesting that not all adult females may have been exposed, and that elevated antibody levels are specific for LETV and may not reflect a generalized

128 polyclonal activation of antibody synthesis prior to nesting. Currently, the sample size is small and it is not known if this is an accurate representation of the adult population.

Future collaborations will work to increase the plasma sample size from nesting and non- nesting females and to begin to collect plasma from adult males to better assess exposure to LETV.

Several adult nesting female loggerhead turtles had high levels of antibodies to LETV.

This is the first data to suggest that loggerheads in Florida may also be infected with

LETV or a LETV-like herpesvirus. This finding is not surprising since the closely related

FPHV virus is associated with FP in green, loggerhead, and olive ridley turtles (Herbst et al. 1998a;Jacobson et al. 1991;Lackovich et al. 1999;Quackenbush et al. 1998). The

ELISA should be applied to assess exposure of other turtle species to LETV.

Western blot analysis of green turtle plasma samples with ELISA optical density values above the cutoff were used to confirm the presence of anti-LETV antibodies in this study (Figures 5-3 and 5-9). Western blot analysis revealed several virus-specific bands of interest in the LETV infected cells. Future studies focused on the identification of several of these viral antigens, in particular the 19 and 38 kDa proteins (Chapter 6).

Some of these antigens may be useful for the development and refinement of new serological tests for other sea turtle herpesviruses.

The ability to grow LETV in culture has facilitated the development of an ELISA to assess exposure of populations of wild turtles to this herpesvirus. The serological data obtained from the application of this assay indicate that wild green turtles in Florida are exposed to LETV and identify the urgent need to investigate LETD further. Infection with LETV is likely to result in clinical disease and mortality in wild green turtle

129 populations. The transient nature of herpesviral lesions and the pelagic and dispersed life stages of post-hatchling animals, makes it difficult to document active disease among wild turtles. Serological assays such as the ELISA described here will enable researchers to monitor wild populations of turtles for exposure to LETV in the absence of overt disease. This assay also may serve as a health screening tool for head start programs and other release programs to reduce the risk of introducing animals harboring infection into wild turtle populations and introducing virus into marine habitats. The mechanism(s) of herpesvirus transmission in the marine environment is unknown, but LETV could potentially be transmitted to uninfected individuals by direct contact between infected turtles or by contact with substrates harboring virus, such as sediments, contaminated surfaces, or seawater. We have previously demonstrated that LETV remains infectious in seawater for up to 5 days (Curry et al. 2000).

In addition to LETV’s potential impact on marine turtle health and conservation,

LETV infection is a potential confounder in serological studies of the fibropapillomatosis-associated herpesvirus. The only method presently available for detecting FP-associated herpesvirus antibodies (IHC) (Herbst et al. 1998b), presents a high levels of cross-reactivity between the two viruses when LETV is used as antigen

(Chapter 3), therefore is not adequate for use in seroepidemiology of free-ranging populations. The specific ELISA for LETV antibodies described here provides one of the tools needed to distinguish these different herpesvirus exposures in wild turtles, much as

ELISA assays are being developed to distinguish herpes simplex viruses (HSV-1 and

HSV-2) exposures in humans.

130

While antibodies were detected in plasma from green turtles at the current study sites in Florida, clinical signs of disease have not been documented. In captivity, disease was documented in 1-year-old turtles (Jacobson et al. 1986). This life stage of free-ranging turtles is pelagic and is not frequently witnessed by researchers and turtle biologists. In addition, factors or stresses that influence viral pathogenesis or influence disease severity in a mariculture system may not be present in natural marine environments (Collier 2000;

Haines and Kleese 1977). It may be that infections with LETV in wild populations are subclinical and rarely exhibit the clinical severity demonstrated in mariculture unless animals are stressed or debilitated. Although these analyses will always be limited by the largely inaccessible life stages of these endangered wild species and complicated by the known significant effects of age, temperature, season, and hormones on antibody responses in these reptiles (Ambrosius 1976), the development of the LETV specific

ELISA allows further exploration of the impact of this herpesvirus on marine turtle health.

The application of the LETV-specific ELISA has allowed documentation of the presence and prevalence of antibodies to LETV in wild Florida marine turtles and has provided evidence for the occurrence of infection in these populations. Future research will explore the impact LETV may have on marine turtle populations and will investigate the relationship between LETV and LETD. Further studies like this are needed to enhance the application of seroepidemiological approaches to establish cause-effect disease relationships in marine turtles exposed to other herpesviruses, including FPHV.

131

2 99A-1 99C-1 1

Optical Density (405 nm) 0

0 1 2 3 4 5 Time (months)

Figure 5-1. Detection of seroconversion of LETV immunized green turtles. 99A-1 (open circles) and green turtle 99C-1 (closed circles). Antibody levels to LETV were detected in plasma from the immunized 99A-1 (open circles) and 99C-1 (closed circles) green turtles by the ELISA over the course of the immunization. Plasma samples were tested at a 1:100 dilution.

132

1.5

1.0

0.5 Cutoff =0.310 0.0 Optical Density (405 nm) -0.5 Captive Captive Wild FP Wild FP FP negative FP positive negative positive n=42 a n=30 a n=19 b n=27 b

Figure 5-2. Measure of antibodies to LETV in plasma from captive-reared and wild green turtles by ELISA. Plasma from 42 captive-reared turtles with no known exposure to herpesviruses and from 30 captive-reared turtles with experimentally induced FP had low optical density values (seronegative). Plasma from 27 wild FP positive and 19 wild FP negative turtles had higher ELISA values, regardless of FP status. The cutoff of 0.310 (dashed lined) was used to calculate seroprevalence.

133 MW 1 2 3 4 MW 5 6 MW 7 8

191

64 51 39

28 19 14 N/A

Figure 5-3. Confirmation of the presence of anti-LETV antibodies by Western blot. Western blot confirms presence of antibodies to LETV in green turtle plasma samples with the highest optical density values in the ELISA. LETV-infected cell lysate(10 mg) was used as antigen in odd numbered lanes. Cell lysate (10 mg) used as control antigen in even numbered lanes. MW: molecular weight marker. Lane 1 and 2: pre-immune 99C-1 plasma. Lane 3 and 4: 2 month post-immune plasma. Lane 5 and 6: FP positive plasma. Lane 7 and 8 FP negative plasma. The immunodominant proteins appear to be the 19 and 38 kDa bands (arrowheads).

134

Post-Immune Wild FP Wild FP Pre-Immune Control Positive Negative Control

LETV Infected

Uninfected

Figure 5-4. Confirmation of the presence of anti-LETV antibodies by immunohistochemistry. Immunohistochemistry confirms presence of antibodies to LETV in green turtle plasma samples with the highest optical density values in the ELISA. Plasma samples from FP positive and FP negative turtles and from the pre-immune (negative control) and 2 month post-immunized (positive control) 99C-1 turtle were incubated with LETV infected and uninfected TH-1 monolayers. Magnification: 400X.

135

a 3.0

2.5

2.0

1.5

1.0

0.5 Optical Density (405 nm) 0.0

-0.5 Lagoon Reef Trident n=60 n=60 n=60 b 3.0

2.5

2.0

1.5

1.0

0.5 Optical Density (405 nm) 0.0

-0.5 1997 1998 1999 n=60 n=60 n=60

Figure 5-5. Measurement of antibodies to LETV by ELISA in plasma samples from juvenile wild green turtles at three Florida study sites. (A.) Measurement of antibodies to LETV in plasma samples collected from green turtles captured in the Indian River lagoon, along the Sebastian reef, or in the Trident basin. (B.) Measurement of antibodies to LETV in plasma samples collected green turtles from each of the three study sites (Indian River lagoon, Sebastian reef, and Trident basin) in 1997, 1998, and 1999.

136

2.0

1.5

1.0

0.5

Optical Density (405 nm) 0.0

-0.5 Score 0 Score 1 Score 2 Score 3 n=36 n=35 n=36 n=26

Figure 5-6. Measurement of antibodies to LETV by ELISA in plasma from wild green turtles scored for fibropapilloma severity. Plasma from individual turtles in all tumor score categories (score 0=no tumor to score 3=most severe) contained antibodies to LETV.

137

3.0

2.5

2.0

1.5 Wild Green Turtles Captive Green Turtles 1.0 0.310

0.5 Optical Density (405 nm)

0.0 0 100 200 300 400 -0.5 Number of Green Turtles

Figure 5-7. Summary of ELISA measurements for antibodies to LETV in plasma from 329 Florida wild green turtles. A cutoff value of OD=0.310 was used to identify samples positive for antibodies to LETV (see Materials and Methods). The estimated seroprevalence was 21.6%.

138

3.5

3.0

2.5

2.0

1.5

1.0

0.5 Optical Density (405 nm)

0.0

-0.5 Nesting Green Nesting Loggerhead Turtles Turtles n=9 n=4

Figure 5-8. Detection of antibodies to LETV by ELISA in plasma from nesting green and loggerhead turtles. Positive control 1.144; SE =0.094; negative control 0.014 SE=0.004; 2 replicates. An arbitrary cutoff of OD=0.310 was used to identify samples as positive for antibodies to LETV (see Materials and Methods).

139

a b MW 1 2 3 4 MW 5 6 7 8 MW 1 2 3 4

191

64 51 39 28 19 14

Figure 5-9. Western blot confirms the presence of antibodies to LETV in selected plasma samples. (A.) Detection of anti-LETV antibodies in plasma from nesting green and loggerhead turtles with ELISA values above the cutoff (OD=0.310). LETV-infected cell lysate (10 mg) was used as antigen in lanes 1-4; cell lysate (10 mg) used as control antigen in lanes 5-8. MW: molecular weight marker. Lanes1 and 5: P2840 green turtle plasma. Lane 2 and 6: N6385 green turtle plasma. Lanes 3 and 7: N9897 loggerhead turtle plasma. Lane 4 and 8: N9954 loggerhead turtle plasma. (B.) Western blot incubated with control plasma. LETV-infected cell lysate (10 mg) was used as antigen in lanes 1 and 3. Uninfected cell lysate (10 mg) was used as antigen in lanes 2 and 4. Lanes 1 and 2: negative control plasma. Lanes 3 and 4: positive control plasma.

140

Table 5-1. Summary of monoclonal antibodies to LETV.

ELISA assayb Immunohistochemistry

Hyba Screen 1c Screen 2c WBd Infectede Uninfectedf 1H6 0.944 1.208 - - - 2A4 1.984 2.401 - + - 2C3 0.577 0.540 + unclear - 2G11 0.776 0.805 + + + 2H9 1.049 1.349 - - - 3B1 0.954 1.328 - unclear - 3B7 1.670 1.743 + + - 3B11 0.643 1.128 + - - 3H8 0.907 1.178 - unclear - 4C7 1.023 0.221 + unclear - 4C11 0.900 0.489 + + - 5A7 1.884 1.965 - - - 5B3 0.767 0.974 - + - 5E4 0.851 1.100 + - - 5E11 0.675 0.777 + unclear -

aHyb= hybridoma fusion supernatants bELISA assay with LETV-infected cell lysate as antigen c OD values (A405 ) of antibodies measured to LETV-infected cell lysates dWB= Western blot analysis with LETV-infected cell lysate as antigen eImmunohistochemistry analysis with LETV-infected cells as antigen fImmunohistochemical analysis with uninfected cells as antigen

141

Table 5-2. Antibodies to LETV in wild turtles captured more than once.

Category Turtle ID Sampling Intervala ELISAb FP Statusc

FP Positive BP4527 0 mo - + 31 mo - + BP4529 0 mo - + 9 mo - + BP4567 0 mo - + 25 mo - + 33 mo - + BP5648 0 mo - + 15 mo - + BP5603 0 mo - + 37 mo - +

FP Negative BP4572 0 mo - - 31 mo + -

BP4595 0 mo + - 33 mo + -

BP5622 0 mo + - 42 mo - - 47 mo - -

BP5570 0 mo + - 22 mo - - 29 mo - -

BP5663 0 mo - - 42 mo - -

a0 month= first time sampled bPresence of antibody to LETV cFP status at time of sampling

142

Table 5-2. Continued

Category Turtle ID Sampling Intervala ELISAb FP Statusc

Regressors BP5607 0 mo - + 36 mo - -

BP5635 0 mo - + 25 mo - - 38 mo - -

BP5637 0 mo + + 19 mo - - 37 mo - -

Progressors BP6641 0 mo - - 18 mo - +

BP7397 0 mo - - 31 mo - +

Other BP3295 0 mo - - 6 mo - + 11 mo - -

a0 month= first time sampled bPresence of antibody to LETV cFP status at time of sampling

CHAPTER 6 IDENTIFICATION AND EXPRESSION OF IMMUNOGENIC PROTEINS FOR DETECTION OF ANTIBODIES TO A DISEASE-ASSOCIATED MARINE TURTLE HERPESVIRUS

Introduction

Efforts to understand the role and impact of these herpesviruses on marine turtle populations have focused on the development of serological assays for seroepidemiological studies. A LETV-specific ELISA assay has been developed using

LETV-infected cell lysates as antigen for the detection of antibodies. However, LETV is a relatively low titer and slow growing virus. Generation of antigen for the LETV ELISA assay is labor intensive and subject to batch-to-batch variation. In addition, the use of

LETV-infected cell lysates includes the potential for high anti-cellular background and requires all plasma samples to be tested on both infected and uninfected cell lysate preparations. Expressed recombinant herpesviral proteins could provide a source of antigen for improved serological assays for LETV. Expressed LETV proteins should circumvent many of the time constraints of cultivating virus, and reduce ELISA background problems.

In this study two strategies were used to select LETV proteins for expression. One protein was previously identified as a potential immunodominant LETV antigen (Chapter

5). This 38 kDa protein was recognized by many, but not all, individual wild turtles with antibodies to LETV, and was targeted for cloning and expression. A second LETV protein, glycoprotein B, was also selected for cloning and expression. Glycoprotein B

(gB) has been identified as a major target of the anti-herpesviral immune response in

143 144 many host species and includes virus-neutralizing activity (Knipe and Howley

2001;Speckner et al. 2000). Glycoprotein B also has been extensively targeted for the development of many mammalian herpesvirus vaccines and for the development of serological assays for detection of herpesvirus infections. Glycoprotein B is the most conserved glycoprotein (Pereira 1994) and provides good representation of the herpesvirus subfamily taxonomy (Knipe and Howley 2001).

LETV glycoprotein B and the 38 kDa protein were cloned and expressed, and shown to react in Western blots. In future seroepidemiological studies, these proteins will be evaluated for their potential to detect exposure of wild turtle populations to disease- associated herpesviruses.

Materials and Methods

LETV Library Construction

Terrapene heart cells (1.0X107) were infected with LETV clone 221 (Chapter 4).

After 6-7 days post infection, cells demonstrating 90-100% cytopathic effect (CPE) were collected from the supernatant by centrifugation 2500 g 10 min at 4 C. The infected cell pellets were resuspended in L buffer (0.1M EDTA pH 8.0, 0.01 M Tris HCl pH 7.6,

0.02M NaCl), mixed with equal volume 1.2% low temperature agarose, and poured in plug mold apparatus. Embedded infected cells were lysed in situ with 2 changes of L buffer plus 1% v/v Sarkosyl plus 0.1% w/v proteinase K for 48 hrs at 50 C. After 3 washes in L buffer, plugs were kept in TE pH 8.0 at 4 C. A 10 well 1.0% 0.5X TBE low temperature agarose gel was poured and the embedded cells were loaded into each of the

8 wells. Lambda Ladder PFG marker (New England Biolabs, Beverly, MA) and Low

Range PFG Marker (New England Biolabs, Beverly, MA) were loaded into the remaining wells. LETV 221 genomic DNA was separated by pulse-field gel electrophoresis (200V,

145

24 hours, 50-90 sec) at 4 C. The gel was stained in ethidium bromide bath, and bands were visualized on long wavelength UV box. Bands corresponding to the LETV genome

(approx 150 kb) were cut out of the gel and were melted at 70 C for 20 min. Melted gel slices were digested in b-agarase (5 units/ 100 mg) overnight at 45 C. The mixture was transferred to Centricon YM-100 (Millipore, Medford, MA) and concentrated by centrifugation at 500 g for 20 min at room temperature. The centricon was washed with

20% isopropanol, reconcentrated, and then washed with 2 volumes of TE pH 7.5. The centricon was inverted and the DNA sample was collected by centrifugation. Next, 20 mg of LETV 221 DNA was sonicated to generate fragments 1-2 kb in size, and then the

DNA was size fractionated on CHROMA SPIN-100 gel filtration columns (Clontech,

Palo Alto, CA) to remove DNA molecules smaller than 300 bp. The DNA was eluted, ethanol precipitated, and blunt ends were formed using T4 DNA polymerase. The DNA was ligated into SmaI linearized pUC18 cloning vector and then the ligated product was ethanol precipitated. The ligated product was electroporated into E. coli host TOP10 and clones were blue/white selected on Luria-Bertani (LB) agar plates plus 5- bromo 4- chloro-3 indoyl b-D galactopyranoside (X-gal) and isopropyl b-D-1- thiogalactopyranoside (IPTG). Clones were grown in LB plus ampicillin for 20 hours in

96 well plate format with agitation. Bacterial clones were harvested and DNA extracted with Qiagen Biorobot (Qiagen, Valencia, CA). DNA samples were quantified, and then sequenced on MegaBACE sequencer using ET terminator mix in MJ thermocycler

(Amersham Pharmacia, Biotech, Piscataway, NJ). Sequences were compared to other sequences by blastx (basic local alignment search tool in protein database) searching the

NCBI (National Center for Biotechnology Information) database. Clones matching

146 herpesvirus genes were arranged according to the gene order of the model alpha- herpesvirus, herpes simplex virus 1 (HSV1).

Two -Dimensional Gel Electrophoresis

LETV 221 was grown in TH-1 cells and after 75-80% CPE was observed, cells were scraped into medium and the cell suspension was stored at –85 C until needed. LETV 221 cell suspension was thawed in lukewarm bath, was sonicated for 30 sec on ice, and then was clarified at 2500 g for 20 min at 4 C. The supernatants were collected and pooled, and then virus was pelleted by centrifugation at 40,000 g for 1 hr 4 C in SW28 rotor. The supernatant was discarded, pellets were washed by resuspending in 30 ml cold 10 mM

Tris pH 7.4, and then virus was pelleted again as described above. The supernatant was discarded and the pelleted virus was washed again. Finally, the pelleted virus was resuspended in 1 ml 9 M urea, 4% CHAPS, 2 M thiourea, 65 mM DTT, 0.1 % SDS, and trace Orange G. The virus preparation was sonicated until homogenous (approximately

1.5 min) and total protein was measured by Bradford assay. The virus sample was prepared and resolved by 2D electrophoresis as follows: 549 ug of total protein plus 2% ampholyte pH 3-10NL (Amersham Pharmacia Biotech, Piscataway, NJ) was clarified by centrifugation 15,000 g for 20 min at room temp for each gel. The supernatant was collected with a 27 gauge needle to sheer genomic DNA and was incubated with

Immobiline strips pH 3-10 NL (Amersham Pharmacia Biotech, Piscataway, NJ) overnight at room temperature covered in mineral oil. The Immobiline strips were assembled with wicks in an isoelectric focusing apparatus (Amersham Pharmacia

Biotech, Piscataway, NJ), and the proteins were resolved in the first dimension under the following conditions: 300 V, 100 mA, 30 W, 0:01 hours; 300V, 150 mA, 30 W, 2:30 hours; 3500V 100 mA, 50W 3:00 hours, 3500 V 150 mA 100W 22 hours. The Immobline

147 strip were then incubated in equilibration buffer I (Genomic Solutions, Ann Arbor, MI) for 15 min at room temperature with rocking. Buffer I was discarded and then the strips were incubated in equilibration buffer II (Genomic Solutions, Ann Arbor, MI) for 15 min at room temperature with rocking. Finally, the strips were loaded and secured with agarose into 4-20% Tris-Glycine ZOOM gels (Invitrogen, Carlsbad, CA) to resolve the second dimension at 125 V. One gel was stained with colloidal blue stain (Pierce,

Rockford, IL) or Coomassie stain. Another gel was electroblotted onto PVDF membrane at 30 V for 1.5 hours. The membrane was blocked with 5% nonfat dried milk in phosphate buffered saline 0.5% Tween 20 and 0.02% sodium azide (PBS/TM) overnight at 4 C. After washing, the membrane was incubated with 1 month post-immunization plasma from the LETV immunized turtle 99A-1 diluted 1:100 in 2% FBS in phosphate buffered saline and 0.02% sodium azide (PBS/F) for 1 h with rocking to identify the 38 kDa protein. The membrane was washed, and then incubated with biotinylated anti-green turtle 7s IgG monoclonal antibody HL858 (Herbst and Klein 1995b)at 1 ug/ml in PBS/A for 1 hour at room temperature with rocking. After washing, the membrane was incubated with Streptavidin-conjugated alkaline phosphatase (Zymed Laboratories Inc,

San Francisco, CA) diluted 1:5000 in PBS/A for 1 hour at room temperature with rocking. After final washing, the blot was developed in substrate buffer (0.1 M Tris HCl pH 8.8, 1 mM MgCl2) containing nitroblue tetrazolium chloride and 5-bromo-4-chloro-3- indolylphosphate p-toluidine salt (Promega, Madison, WI). The reaction was stopped in water. Initially, the two proteins recognized (referred to as P1 and P2) were cut out of the

PVDF membrane with a scalpel for N-terminal protein sequencing. After determining both proteins were N-terminally blocked, Western blots were aligned with corresponding

148

Coomassie stained gel to match the proteins recognized by antibodies in the Western to the proteins stained in the gel. These proteins were cut out of the gel with a sterile scalpel. Homologous protein spots from 7 gels were pooled to obtain enough material for protein sequencing. A portion of the gel with no proteins present were used as negative or background control. The proteins were extracted, endolyse C digested, and protein fragments were purified by HPLC. The protein fractions not present in the gel alone or background control were sequenced.

Cloning of LETV UL26 and Glycoprotein B Genes

Primers were designed based on sequence information obtained from the LETV partial genomic library to PCR amplify the open reading frame of UL26 or gB. The primers were also designed to include restriction enzyme sites for cloning into several expression systems. Total DNA from LETV infected cells (100 ng) was added to a 50 ml

PCR reaction mixture containing 1.5 mM MgCl2, 0.4 mM dNTP, 0.2 mM of each primer, and 2 units of Proof start polymerase with supplied buffer (Qiagen, Valencia, CA). The

PCR cycles were performed in a Perkin Elmer Gene Amp 2400 thermal cycler under the following conditions: 2 min 94 C; denaturing step at 94 C for 10 sec, annealing step at 65

C for 30 sec, and extension step at 72 C for 4 min for a total of 35 cycles. The resulting

PCR product was gel purified (Roche Indianapolis, IN), the purified PCR product was sequenced directly, and then was T/A cloned into pTAdv plasmid (Clontech, Palo Alto,

CA). Each clone of the appropriate size (gB: 2638 bp; UL26: 1777 bp) was sequenced and compared to other herpesvirus homologs in the database. Plasmid DNA was generated and double digested in 1 unit each of Nde I and Xho I in preparation for

149 subcloning into Nde/Xho linearized pET16b expression vector (Novagen, Madison, WI).

The insert sequence of the pET16b clones was confirmed by sequencing.

Expression and Detection of LETV Proteins

Plasmid DNA was transformed into the HMS174 (DE3) strain ( Novagen, Madison,

WI) for expression. An 11 ml culture was prepared and once the optical density value

(A600) of culture medium reached 0.8, the culture was split into two tubes. One tube served as the uninduced control, and to the second tube 1 mM IPTG was added to induce expression in the culture. Both cultures were returned to a 37 C incubator for 2.5 hours.

The bacteria were collected by centrifugation at 5000 g for 5 min at 4 C. The pellets were washed twice with cold 20 mM Tris-HCl pH 8.0. The washed bacterial pellets were stored frozen overnight. The pellet was resuspended in 0.5 ml cold PBS, and then benzonase (25 units) was added to the lysed bacteria and incubated at room temp for 20 min. Samples were prepared, boiled, and equal amounts of each sample were loaded onto a NuPage 10% Bis-Tris gel (Invitrogen, Carlsbad, CA) and run at 175 V in MOPS buffer.

The gel was then incubated in colloidal blue stain or electroblotted onto nitrocellulose membrane at 30V for 1 hour. Nitrocellulose membranes were blocked with 1% Alkali- soluble casein (Novagen, Madison, WI) overnight at 4 C. The membrane was washed three times in TBST (Tris-buffered saline plus Tween 20) and was incubated with 0.2 ug/ml monoclonal anti-His antibody (Novagen, Madison, WI) for 1 hour with rocking.

The membrane was washed and incubated with rabbit anti-mouse IgG-alkaline phosphate

(Sigma, St. Louis, MO) for 1 hour at room temperature with rocking. The membrane was washed and finally incubated in substrate buffer (0.1 M Tris HCl pH 8.8, 1 mM MgCl2) containing nitroblue tetrazolium chloride and 5-bromo-4-chloro-3-indolylphosphate p- toluidine salt (Promega, Madison, WI). The color reaction was stopped with water.

150

Immunogenicity of Expressed LETV Proteins

The immunogenicity of the expressed proteins was evaluated by Western blot analysis. The induced and uninduced bacterial lysates were resolved and transferred to nitrocellulose membrane as described above. The blocked membrane was incubated with

1 month and 3 month post-LETV immunization plasma from green turtle 99A-1 (Chapter

5) diluted 1:100 in 2% FBS/A or with wild green turtle plasma diluted 1:25 in 5% nonfat dried milk in TBST for 1 hour with rocking at room temperature. The membrane was washed and incubated with biotinylated 1 ug/ml HL858 monoclonal antibody. After a 1 hour incubation, the membrane was washed and incubated with 1:5000 Streptavidin- conjugated alkaline phosphatase in PBS/A for 1 hour with rocking. The membrane was washed a last time and incubated in substrate buffer (0.1 M Tris HCl pH 8.8, 1 mM

MgCl2) containing nitroblue tetrazolium chloride and 5-bromo-4-chloro-3- indolylphosphate p-toluidine salt (Promega, Madison, WI).

Phylogenetic Analyses

Amino acid sequences of herpesvirus UL26, DNA polymerase, and glycoprotein B genes were obtained from NCBI (National Center for Biotechnology Information) database. LETV sequences were obtained from the LETV partial genomic library or from

LETV clones described above. Protein sequences were aligned using the ClustalX analysis program at the website of the European Bioinformatics Institute

(http://www2.ebi.ac.uk/clustalw/). Phylogenetic trees were generated using the programs

PROTDIST and FITCH (Fitch-Margoliash method) from the PHYLIP package

(Phylogeny Inference Package; University of Washington, Seattle, WA). The phylogenetic trees were constructed using the cladogram output option and the program

TREEVIEW version 1.6.6 (Page 1996). The percentage of amino acid sequence identity

151 was also determined for UL26 and glycoprotein B genes of LETV to other alpha- herpesviruses.

Results

Partial LETV Genomic Library

The LETV genome was estimated to be approximately 150 kb in size based on its mobility in a pulse-field electrophoresis gel (Figure 6-1). A LETV partial genomic library was successfully constructed in pUC-18 and many of the clones were matched to herpesvirus gene homologs in the NCBI database. Homologs to 29 HSV1 genes were detected (Table 6-1). The LETV partial genomic library contains about 18.7% of the

LETV genome.

Identification of the 38 kDa LETV Protein

The 38 kDa protein was detectable in virus-infected cell lysate preparations, but not in uninfected cell controls. The 38 kDa band visualized in a one-dimensional gel (Figure 6-

2a) resolved into two protein spots, P1 and P2, when visualized by two-dimensional gel electrophoresis (Figure 6-2b). When the protein spots were cut out of the PVDF membrane and were submitted for direct sequencing, both were N-terminally blocked.

The corresponding proteins spots were identified in the colloidal blue stained two- dimensional gels (Figure 6-2c) and submitted for analysis. Protein sequence was obtained from three fractions for each of the protein spots (Figure 6-3). The sequence information identified these two spots as the same protein, and the two spots are likely the result of differential post-translation modification such as phosphorylation. As demonstrated in

Figures 6-3 and 6-4d, the 38 kDa sequence fractions matched a portion of the LETV genome containing the overlapping open reading frames corresponding to HSV1 genes

UL26 (minor capsid protein or protease) and UL26.5 (assembly or scaffolding protein).

152

In other herpesviruses, UL26 encodes a polyprotein that is autoproteolytically cleaved to yield protease and another scaffolding protein (Hoog et al. 1997;Sheaffer et al.

2000;Tigue et al. 1996). The UL26 LETV homolog is diagrammed with these putative cleavage sites (Figure 6-4b) and the two hypothetical protein products, protease and scaffolding protein (Figure 6-4c). The 2D fraction sequences map to the putative LETV assembly protein (shaded boxes; Figure 6-4d) suggesting that the 38 kDa protein is likely a scaffolding protein responsible for proper capsid formation.

PCR Amplification and Cloning of LETV Genes

Primers were synthesized based on sequences from the LETV partial genomic library, and a PCR assay was successfully developed to amplify the region of the LETV genome containing the open reading frame for glycoprotein B (UL27) and UL26. The open reading frame for gB was initially cloned in two parts (Figure 6-5). The LETV genome organization matched that of an alpha-herpesvirus and was consistent with the results of phylogenetic analysis based on these genes (Figure 6-7). Once the entire sequence of gB was obtained, primers were designed to amplify the entire open reading frame for cloning and expression. Similarly, a region of the genome spanning UL26 was PCR amplified, and primers were designed for cloning and expression of UL26.

Expression of Marine Turtle Herpesviral Proteins

LETV UL26 and gB were expressed successfully in HMS174 cells. Expression of the clone LETV UL26 gene resulted in two proteins: 26 and 38 kDa in size (Figure 6-6a, lane

1). Based on homologs of UL26 in other herpesviruses, this gene encodes a polyprotein that is autoproteolytically cleaved during expression and yields the assembly protein and protease (Hoog et al. 1997;Sheaffer et al. 2000;Tigue et al. 1996). The two expressed

LETV proteins likely correspond to these protein products. As diagrammed in Figure 6-4,

153 it is hypothesized that the protease portion of the putative polyprotein maintained the His- tag while serendipitously the assembly or scaffolding protein contained an internal His rich region (HPHHHH) such that the monoclonal anti-His antibody recognized both products of the UL26 (Figure6-6a, lane 1). A band corresponding the expected molecular weight of glycoprotein B was also observed (Figure 6-6b, lane 1).

Antigenicity of Expressed Herpesviral Proteins

Both the UL26 and glycoprotein B were expressed successfully as detected by anti-

His antibody. The antigenicity of the expressed proteins was evaluated by Western blot analysis using plasma from a LETV immunized green turtle and a wild LETV antibody positive green turtle. Plasma from the LETV immunized green turtle was incubated with the UL26 expression products and recognized the 38 kDa protein specifically and not the

26 kDa protein (Figure 6-6a, lane 3). This is consistent with the 2D gel electrophoresis study where the fraction sequences were mapped to the putative assembly protein portion of the polyprotein (Figure 6-4d). A wild LETV positive green turtle plasma sample, shown previously to recognize the 38 kDa protein in LETV infected cell lysates (Chapter

5), recognized the E. coli expressed 38 kDa protein (Figure 6-6a, lane 5). Many other proteins in the induced and uninduced preparations were also recognized demonstrating a high level of background when an E. coli lysate is used as antigen (Figure 6-6a. lane 5 and 6). It was expected that green turtles would have natural antibodies to E. coli.

Plasma from a LETV immunized green turtle and a wild LETV antibody positive green turtle were also incubated with the expressed LETV glycoprotein B. The plasma from the LETV immunized green turtle reacted strongly with a protein corresponding to the appropriate weight of glycoprotein B (Figure 6-6b, lane 3). Furthermore, the antibodies in the plasma from the LETV immunized turtle did not react with the

154 uninduced E. coli lysate that lacked glycoprotein B (Figure 6-6b, lane 4). The interpretation of the reaction of the wild turtle plasma with glycoprotein B is more difficult. While a weak reaction with a protein corresponding to the appropriate molecular weight of glycoprotein B was observed, there is a very high level of background (Figure 6-6b, lanes 5 and 6).

Phylogenetic Analysis of LETV Genes

Phylogenetic trees were constructed to determine the relationship of LETV to other herpesviruses. Independent analysis was performed with the herpesvirus UL26, DNA polymerase gene, and glycoprotein B. LETV consistently clustered with alpha herpesviruses (Figure 5-7a through c). Based on this analysis LETV was determined to be most similar to other herpesviruses associated with fibropapillomatosis of several species of marine turtles (Figure 5-7b). The percentage of amino acid sequence identity of LETV

UL26 (Table 6-2) and glycoprotein B (Table 6-3) genes to other alpha-herpesviruses was also determined.

Discussion

Improved diagnostic assays are needed to detect exposure of marine turtles to disease- associated herpesviruses. Recombinant herpesviral proteins can serve as a valuable source of antigen for the development of immunodiagnostic assays in the absence of cultivated virus and as an alternative to virus-infected cell lysates. Two strategies were used in this study to identify useful antigenic targets for cloning and expression. The first focused on a 38 kDa protein recognized on Western blots by antibodies in LETV immune plasma. The second relied on available information on immunodominant herpesvirus proteins in other species and lead to the selection of glycoprotein B for cloning and expression. To our knowledge, these are not only the first proteins from a reptilian

155 herpesvirus to be cloned and expressed, but they represent the first reptilian herpesviral proteins to be identified as immunogenic in their host species.

Prior to this study, genomic sequence information for marine turtle herpesviruses was limited. The only genomic sequence information for LETV was a 181 bp portion of the herpesviral DNA polymerase gene (Herbst et al. 1998a). To obtain additional sequence information, a partial LETV genomic library was constructed using DNA obtained by pulse-field gel electrophoresis. Pulse-field gel electrophoresis is a useful tool for isolating genomic DNA from herpesviruses that are tightly cell-associated and has been utilized for other herpesviruses such as Marek’s disease-associated herpesvirus (Wilson and

Coussens 1991). LETV genomic DNA was identified by pulse-field gel electrophoresis and was determined to be approximately 150 kb in size (Figure 6-1). This is within the range of genome sizes of members of the Herpesviridae family (Knipe and Howley

2001). While this library only contained 18.7% of the genome (Table 6-1), it provided enough sequence information to PCR amplify genes of interest.

The entire coding sequence for LETV glycoprotein B was obtained and was similar to that of other herpesviruses (Figure 6-7c; Table 6-3). Glycoprotein B is the most conserved herpesviral glycoprotein and is the representative gene routinely used for phylogenetic studies (Knipe and Howley 2001). When the LETV gB was compared to glycoprotein B of other herpesviruses it occupied a distinct branch (Figure 6-7c). This demonstrates that while glycoprotein B is relatively conserved overall, marine turtle herpesviruses are unique members of the Herpesviridae family. Currently, the herpesvirus subfamily to which LETV belongs is unknown. However, the organization of the portion of the LETV genome containing glycoprotein B supports the hypothesis that LETV is an

156 alpha-herpesvirus. In addition, in phylogenetic analysis LETV glycoprotein B clustered with other alpha-herpesviruses (Figure 6-7c).

Immunogenic proteins of marine turtle herpesviruses have not been previously described. Two-dimensional gel electrophoresis allowed the separation of complex virus- infected cell lysate preparations, and in combination with Western blot analysis, identified proteins recognized by the green turtle antibody response to LETV (Figure 6-

2b). The 38 kDa protein recognized by antibodies of several marine turtles was identified as a scaffolding protein encoded by the overlapping LETV UL26 and UL26.5 genes.

Translation of the LETV UL26 message resulted in two products which are presumed to be the result of autoproteolytic cleavage of the putative polyprotein as occurs in other herpesviruses (Hoog et al. 1997;Sheaffer et al. 2000;Tigue et al. 1996). Based on what is known about other herpesviruses, the proteins represent herpesviral protease and scaffolding protein. The putative protease portion of this hypothetical polyprotein is well conserved with approximately 25 % identity to human herpesviral protease amino acid sequence, while the scaffolding protein is more poorly conserved (> 19% homology).

Phylogenetic analysis revealed similar branched patterns for trees constructed based on amino acid sequence of UL26 (Figure 6-7a). Consistent with analyses for glycoprotein B and DNA polymerase, the phylogeny trees also placed LETV with alpha-herpesviruses.

Table 1 demonstrated the overall percentage of amino acid sequence identity for the entire LETV UL26 gene compared to other alpha-herpesviruses.

The coding regions of UL26 and glycoprotein B were successfully PCR amplified, sequenced, and cloned into expression vectors. The UL26 expression protein was recognized (Figure 6-6a, lanes 3 and 4) by the same plasma sample (from a LETV

157 immunized turtle) used to identify the 38 kDa protein in LETV infected cell lysates by

2D gel electrophoresis (Figure 6-2b). Plasma from a wild green turtle, previously demonstrated to recognize a 38 kDa protein in LETV infected cells (Chapter 5), also recognized the UL26 expressed protein (Figure 6-6a, lanes 5). This indicated that the 38 kDa protein recognized by the immunized turtles is the same herpesviral protein recognized by wild turtles naturally exposed to LETV. Antibodies to histidine identified a protein of the appropriate molecular weight as expressed his-tagged glycoprotein B. In addition, plasma from the LETV immunized green turtle also reacted strongly with a protein corresponding to the appropriate molecular weight of glycoprotein B (Figure 6-

6b, lane 3). However, it is difficult to interpret the reaction of the plasma with from the wild turtle (Figure 6-6b, lanes 5 and 6). First, this wild turtle plasma sample was selected because it had high levels of antibody to many LETV antigens (Chapter 5), but its reactivity with glycoprotein B was completely unknown. In addition, there is currently no positive control antibody specific for LETV glycoprotein B. The antigenicity of glycoprotein B in herpesvirus infected wild turtles is unresolved until plasma from large numbers of turtles can be tested on purified antigen and a specific control antibody against LETV glycoprotein B is generated. Such large scale experiments are planned using these newly available expressed LETV proteins.

Recombinant expressed UL26 and glycoprotein B have been evaluated previously for development of serological assays in other species. The product of UL26 (minor capsid protein) has been previously used as antigen in immunodiagnostic assays for the detection of exposure to other disease-associated herpesviruses. Specifically, peptides derived from UL26 gene (or ORF65) was used as antigen in ELISA assays for the

158 detection of antibodies to Kaposi’s sarcoma-associated herpesvirus, KSHV (Pau et al.

1998). This suggests that expressed 38 kDa marine turtle herpesviral proteins may be useful antigens as well. While this protein alone was not diagnostic in the case of KSHV, it is hypothesized that this antigen could be of value when used in combination with other immunogenic herpesviral proteins. Glycoprotein B has also been used extensively for the development of serological assays (Caselli et al. 2000;Franti et al. 2002;Loomis-Huff et al. 2001), and like UL26, had the most diagnostic strength when used in combination with other herpesviral proteins (Corchero et al. 2001). A similar strategy of using a combination of recombinant herpesviral proteins will be needed for development of future assays for exposure to marine turtle herpesviruses.

LETV is the only marine turtle herpesvirus to be maintained in culture and as such it can serve as a model for marine turtle herpesviruses. Phylogenetic analysis has demonstrated a closer evolutionary relationship between LETV and FPHV than other herpesviruses currently in the database (Figure 6-7b). Furthermore, antibodies to FPHV cross-react with LETV, suggesting a level of conservation of viral proteins between these two viruses (Chapter 3). Based on these findings, the assumption is that many of the immunodominant proteins of marine turtle herpesviruses are likely to be conserved.

Degenerate primers are currently being developed to PCR amplify FPHV gene homologs.

This will potentially provide the first source of genuine FPHV antigens in vitro.

Identification of immunogenic herpesviral proteins such as the 38 kDa protein and glycoprotein B provide a rational approach for developing the next generation of immunodiagnostic assays for herpesvirus-associated diseases of marine turtles.

159

Seroepidemiological studies using these assays will help determine the role of these herpesviruses in their respective diseases.

160

1 2 3 4 5

339.5 291.0 242.5 194.0 194.0 145.5 145.5 97.0 97.0 48.5 48.5 23.1

9.42

6.55

Figure 6-1. LETV genome resolved and isolated by pulse-field gel electrophoresis. Lane 1: Lambda marker. Lane 2: Low range molecular weight marker. Lane 3: LETV-infected cells. Band corresponding to the LETV genomic DNA is indicated by arrowhead Lane 4: LETV genomic DNA cut out for construction of partial genomic library.

161

a b 191

64 148 P1 51 60 39 42 30 28 22 19 17 P2 14 6 N/A 4

148 P1 60 42 30 22 17 P2 6 4

Figure 6-2. Identification of herpesviral 38 kDa protein by two-dimensional gel electrophoresis and Western blot analysis. (A) The 38 kDa protein is recognized in a LETV infected cell lysate by plasma from a LETV immunized green turtle assayed by Western blot. (B) The same plasma sample recognizes two protein spots (P1 and P2 indicated by arrows) when the LETV infected cell lysate was resolved by two- dimensional gel electrophoresis and then assayed by Western blot. (C) The LETV infected cell lysate was resolved by two-dimensional gel electrophoresis and then Coomassie stained to identify proteins corresponding to the two spots recognized in the Western blot. The corresponding proteins, P1 and P2, are indicated by the arrows.

162

Assembly Protein (UL26.5) Protease (UL26)

580 1

492 476 427 409 NLSSNVDAVSPSPAPKKVKTPLR(34aa)AADGGVRKFAVGRSTIIKEYYPATTK P1: LXSNVDAVSPSPAPK P2: NVDAVSPSPAPI P1: KFAVGRSTIIK P2: KFAVXRXTMQ P1: KEYYPATAK P2: KEYYPATTK

Figure 6-3. Amino acid sequences obtained from fractions of the P1 and P2 proteins. The fraction sequences matched a portion of the LETV genome containing the overlapping open reading frames corresponding to HSV1 genes UL26 and UL26.5. The top row of amino acids corresponds to LETV sequence for this region of the genome. The fractions obtain from digestion products of P1 and P2 are shown aligned. The P1 and P2 were determined to be the same protein.

163

a Assembly Protein (UL26.5) Protease (UL26)

UL26 UL26.5

545/544 238/237

580aa 1

c HHHHPH His-tag

scaffolding protein protease (4 kDa) (34 kDa) (26 kDa)

d 492-476aa 427-409aa

Figure 6-4. Schematic illustrating the process of mapping sequences of P1 and P2 to the overlapping UL26 and UL26.5 portion of the LETV genome. (A) This diagram represents the two putative overlapping LETV open reading frames of UL26 and UL26.5 (indicated by arrows). (B) The putative polyprotein encoded by LETV UL26 is shown as a solid box with the two hypothetical cleavage sites indicated with arrowheads. (C) The two protein products, protease and scaffolding protein, expected from cleavage of the cloned LETV UL26 are shown with the anticipated sizes. The protease portion possesses a histidine tag and the scaffolding protein has a histidine rich region as shown. (D) The protein sequences from the 38 kD resolved by two-dimensional electrophoresis are aligned with the scaffolding portion of the overlapping region of UL26 and UL26.5 and are indicated by the hatched boxes.

164

ssDNA Assembly Assembly DNA polymerase binding Glycoprotein B Protease Protein Protein (UL30) protein (UL27) (UL26) (UL28) (UL26.5) (UL29)

9 6

5 7

Figure 6-5. Diagram representing a portion of a typical alpha-herpesvirus genome. Clones from the LETV partial genomic library containing the corresponding portion of the genome are indicated by hatched boxes. Primers (shown as arrows with numbers) were designed based on the sequence of these clones to PCR amplify the entire open reading frame of the LETV glycoprotein B gene.

165

a MW 1 2 MW 3 4 5 6

97 64 51 39

28 19 14

b MW 1 2 MW 3 4 MW 5 6

97 64 51

28 19 14

Figure 6-6. Western blot analysis of expressed LETV proteins, UL26 and glycoprotein B. In the odd numbered lanes, lysates of induced E. coli were used as a source of antigen, and in the even number lanes, lysates of the uninduced E. coli control were used as a source of antigen. The molecular weight marker is indicated by MW. Lanes 1 and 2 were incubated with monoclonal anti-His antibody to identify expression of cloned herpesviral proteins. (A) The LETV scaffolding protein (indicated by arrowhead) and protease were recognized by anti-histidine antibody. Lanes 3 and 4 were incubated with 1 month post- LETV immunization plasma. Lanes 5 and 6 were incubated with plasma from a wild LETV antibody-positive green turtle shown previously to recognize the 38 kDa protein in LETV infected cell lysates. (B) The expressed LETV glycoprotein B (indicated by arrowhead) was recognized by anti-histidine antibody. Lanes 3 and 4 were incubated with 3 month post-LETV immunization plasma. Lanes 5 and 6 were incubated with plasma from an adult LETV antibody-positive green turtle.

166

a b c

MEHV1 EHV4 LETV PSV EHV1 VZV BHV1 PSV CHV7 EHV4 BHV1 EHV1 BHV1 PSV GALL1 MEHV EHV4 GALL2 GALL3 EHV1 GALL3 GALL2 HSV2 GALL1 HSV2 HSV1 HSV1 VZV LETV MEHV CHV7 HSV2 GALL2 LETV HSV1 GALL3 ORTHV VZV GALL1 HGTHV CHV7 LTHV HCMV FGTHV TUHV TUHV HCMV HHV7 HCMV HHV7 HHV6 TUHV HHV7 HHV6 MHV4 HHV6 EBV EHV2 ATHV3 MHV4 KSHV ATHV3 EBV ATHV3 KSHV AHV1 EHV2 EBV KSHV MHV4 AHV1 EHV2 AHV1

Figure 6-7. Phylogenetic analysis of LETV based on three herpesviral genes. (A) Comparison of the herpesviral UL26 genes. (B) Comparison of the herpesviral DNA polymerase genes. (C) Comparison of the herpesviral glycoprotein B genes. Various herpesviruses from all three subfamilies were evaluated and identified with full name and Genbank accession number below. Alpha-herpesviruses: GALL1= Gallid herpesvirus 1 (D13713; AF168792; AB012137), GALL2= Gallid herpesvirus 2 (NC_002229), GALL3= Gallid herpesvirus 3 (NC_002577), HSV1 = Human simplex virus 1 (X14112), HSV2= Human simplex virus 2 (Z86099), HGTHV= Hawaiian green turtle herpesvirus DNA polymerase (AF035003), LTHV= Loggerhead turtle herpesvirus DNA polymerase (AF035005), CHV7= Cercopithecine herpesvirus 7 (NC_002686), ORTHV = Olive ridley turtle herpesvirus DNA polymerase (AF049904), VZV= Varicella-Zoster virus (X04370), FGTHV= Florida green turtle herpesvirus DNA polymerase (AF035004), BHV1= Bovine herpesvirus 1 (NC_001847), EHV4= Equine herpesvirus 4 (AF0370027), EHV1= Equine herpesvirus 1 (NC_001491), PSV= Pseudorabies virus (M17321; X95710; L24487), MEHV= Meleagrid herpesvirus 1 (NC_002641). Beta herpesviruses: HHV6= Human herpesvirus (X83413), HHV7= Human herpesvirus 7 (AF037218), HCMV= Human cytomegalovirus (X17403), TUHV= Tupaia herpesvirus (NC_002794). Gamma herpesviruses: EBV= Epstein-Barr virus (V01555), KSHV= Kaposi’s sarcoma- associated herpesvirus (U93872), EHV2= Equine herpesvirus 2 (U20824), AHV1= Alcelaphine herpesvirus 1 (NC_002531), ATHV3= Ateline herpesvirus 3 (NC_001987), MHV4= Murid herpesvirus 4 (NC_001826 ).

167

Table 6-1. LETV genomic library containing homologs to human simplex virus genes.

Gene Homologb Total bpc Gene Encodesd

UL5 1767 Component of DNA helicase-primase

UL6 2019 Minor capsid protein

UL8 958 Component of DNA helicase-primase

UL9 1636 Origin binding protein

UL12 1044 Deoxyribonuclease

UL15 838 Terminase

UL16 654 Involved in host range

UL19 523 Major capsid protein

UL21 883 Tegument protein

UL23 806 Thymidine kinase

UL25 1268 Capsid associated tegument protein

UL26 600 Protease

UL27 272 Glycoprotein B

UL28 524 Assembly protein

UL29 666 Single stranded DNA binding protein

aHSV= human simplex virus 1 bLETV sequences matched to homolog of HSV gene cTotal number of base pairs for given gene contained in the LETV library

168

Table 6-1. Continued

Gene Homologb Total bpc Gene Encodesd

UL30 422 DNA polymerase

UL31/32 666 Matrix protein/membrane protein

UL33 873 Assembly protein

UL36 701 Tegument

UL38 901 DNA capsid protein

UL39 1315 Ribonucleotide reductase, lg. subunit

UL40 775 Ribonucleotide reductase, sm. subunit

UL44 295 Glycoprotein C

UL45 519 Thymidyl synthase

UL49 832 Tegument protein

UL50 871 Deoxyuridine triphosphatase (dUTPase)

UL52 1303 Component of DNA helicase-primase

US3 885 Protein kinase

UL8 899 Glycoprotein E

NEd 2337 Semaphorin

aHSV= human simplex virus 1 bLETV sequences matched to homolog of HSV gene cTotal number of base pairs for given gene contained in the LETV library dNE= not equivalent to any HSV gene

169

Table 6-2. Percentage of amino acid sequence identity of UL26. LETVa GALL1b HSV1c BHV1d EHV1e PSVf

LETV ------

GALL1 25.34 ------

HSV1 25.32 32.27 ------

BHV1 27.34 33.38 37.74 ------

EHV1 27.77 32.72 41.09 40.79 ------

PSV 27.44 29.90 34.82 37.48 40.99 ------

aLung-eye-trachea disease-associated herpesvirus bGallid herpesvirus 1 cHuman simplex virus 1 dBovine herpesvirus 1 eEquine herpesvirus 1 fPseudorabies virus

170

Table 6-3. Percentage of amino acid sequence identity of glycoprotein B. LETVa GALL1b HSV1c BHV1d EHV1e PSVf

LETV ------

GALL1 34.34 ------

HSV1 31.14 44.65 ------

BHV1 29.65 45.40 48.14 ------

EHV1 31.21 44.79 44.74 50.65 ------

PSV 31.02 45.71 49.68 59.37 50.75 ------

aLung-eye-trachea disease-associated herpesvirus bGallid herpesvirus 1 cHuman simplex virus 1 dBovine herpesvirus 1 eEquine herpesvirus 1 fPseudorabies virus

CHAPTER 7 DISCUSSION

Overview

Herpesviruses have been implicated as the etiological agents of three marine turtle diseases, FP, LETD, and GPD. These studies have focused on FP and LETD and the role of FPHV and LETV in their respective diseases. The data collected support the hypothesis that these herpesviruses are credible pathogens in the marine environment and presumed etiological agents of their associated marine turtle diseases.

The Etiological Role of FPHV in FP

In the case of FP, transmission studies have demonstrated that a sub-cellular infectious agent is responsible for the induction of tumors in the absence of additional cofactors

(Herbst et al. 1995). This infectious agent, still unidentified, is successful at inducing FP in marine turtles of different genetic backgrounds (Herbst et al. 1995). Currently, all evidence points to FPHV as the etiological agent of FP. FPHV is specifically and consistently associated with FP diseased tissues (Lackovich et al. 1999). Serological studies have shown that the development of antibodies to FPHV correlates with the development of FP (Herbst et al. 1998b). FPHV DNA has also been shown to be associated with cultured FP cell lines (Figure 2-1; Lawrence Herbst, personal communication). Differential display has identified altered expression of genes with homology to genes associated with other neoplastic diseases, such as thrombospondin and beta-hexosaminidase (Chapter 1; Herbst et al. 2001).

171 172

FPHV and LETV are phylogenetically related (Figure 6-7) and likely share similar physical properties. Studies on LETV have revealed that this virus retained infectivity for at least 5 days in seawater at 23 C (Figure 4-4). Stability in the environment has been well documented for mammalian herpesviruses (Mahl and Sadler 1975;Schoenbaum et al. 1991) and for enteroviruses frequently detected in marine environments (Fujioka et al.

1980;Hejkal and Gerba 1981;LaBelle et al. 1980;Lo et al. 1976). These data suggest that marine turtle herpesviruses have the potential to be maintained on various surfaces in a marine environment (sediments, rubbing rocks, foods sources, etc.) that could serve as vectors for transmission. Interestingly, FPHV DNA was detected on the gills of wrasse cleaner fish (Thalassoma duperrey) that groom sea turtles and has been suggested as one such possible vector for transmission of FPHV (Lu et al. 2000b).

Limitations on Determining the Etiology of FP

The inability to isolate FPHV into pure culture has been the single most important obstacle to testing the etiological role of this herpesvirus in FP. Numerous culture systems under a variety of conditions traditionally used for isolating many viruses were not successful (Chapter 2). Studies on LETV showed that marine turtle herpesviruses appear to have a limited host range, suggesting that future virus isolation efforts should focus on chelonian derived cells as a potential host for FPHV.

As an alternative to demonstrating FP causation directly with purified virus, emphasis was instead placed on developing the means to conduct seroepidemiological studies.

Seroepidemiology has been used extensively in human medicine, where for obvious reasons, transmission studies cannot be performed to investigate the role of suspected pathogens in disease (Dillner 1999;Hashido 2000;Koff 1995;Richardson 1997;Siscovick et al. 2000).

173

Evaluation of Antigenic Cross-reactivity among Herpesviruses

An initial approach to address the limited availability of a source of FPHV antigen was to evaluate, the antigenic cross-reactivity between marine turtle and human herpesvirus proteins to determine if other herpesviruses or herpesvirus antigens could take the place of FPHV in serological assays (Chapter 3). Cross-reactivity was observed in several assay formats and indicated that some epitopes of herpesviruses are conserved.

For instance, it was demonstrated that antibodies to FPHV or LETV cross-reacted with

HSV antigens (Table 3-2). The HSV antigens were explored because they were an available source of antigen and they provided the opportunity to begin to assess the conservation of gB and gG among herpesviruses. Previous studies had examined the reactions of a limited number of plasma samples on commercially available test strips constructed with many different HSV antigens (Chiron Corporation) and commercially available diagnostic kits for VZV (Biowhittaker, Inc., Walkersville, MD) (Lawrence

Herbst, personal communication). These studies suggested that cross-reactivity might exist. To evaluate this initial observation further, the HSV antigens were tested with a larger number of green turtle plasma samples. Serendipitously, phylogenetic analysis demonstrated that LETV gB is more closely related to HSV1 and HSV2 than any of the other herpesviruses evaluated (Figure 6-7c). Given this, it is not surprising that some antibodies to LETV or FPHV cross-reacted with HSV antigens that included gB.

Antigenic cross-reactivity was also observed between FPHV and LETV (Table 3-2).

This too was not surprising given the close phylogenetic relationship between these two viruses (Figure 6-7b). However, it was determined that similar to the results obtained using HSV antigens, LETV could not take the place of FPHV for generating an FPHV- specific serological assay. In addition, the LETV immunized turtles (Figure 5-1) were

174 subsequently challenged with FP filtrate to determine whether immunity to LETV provided protection against infection with FPHV and development of FP (unpublished observations). Some of these filtrates were also incubated with plasma from a LETV- immunized turtle that appeared to have anti-LETV neutralizing activity. Tumors developed at all sites injected with FP filtrates regardless of whether the filtrates were incubated with LETV neutralizing plasma or not. There are many caveats to this experiment, including the lack of a LETV challenge positive control. However, while there are signs of cross-reactivity between LETV and FPHV, this does not appear to be sufficient to provide protection from infection with FPHV and development of FP.

While the inability of LETV to serve as a surrogate antigen for FPHV or protect against FPHV infection is discouraging, these experiments provided valuable insight for the development of future virus-specific serological assays. The need for specificity in serological assays is especially important when one considers that turtles are likely exposed to many herpesviruses including those that have yet to be identified. Diagnostic assays must be able to capable of distinguish among marine turtle herpesviral infections.

The level of antigenic conservation observed between herpesviruses as distantly related as human and marine turtles points to the complexity of designing such assays.

Evaluation of Expressed Herpesviral Antigens

Future serological assays will benefit from using herpesvirus proteins produced in vitro. Analysis of immunological responses to LETV proteins (38 kDa UL26 scaffolding protein; Figure 6-2) and identification of genes known to be immunodominant in other herpesviruses (glycoprotein B; Figure 6-5) have enabled specific herpesviral genes to be targeted as potentially useful sources of antigen. Both LETV UL26 and gB were successfully cloned and expressed, and were determined to be antigenic by incubation

175 with plasma from LETV immunized turtles (Figure 6-6). It is difficult to predict their diagnostic value without further study. The first consideration is whether all turtles will develop an immunological response to these particular proteins during the course of natural virus infection. The second consideration is whether these assays will be complicated by cross-reactivity given the level of conservation between LETV and

FPHV. In fact, there is already some evidence that specific anti-FPHV antibodies in plasma from a green turtle with experimentally induced FP cross-reacted with the 38 kDa protein in a Western blot where LETV-infected cells were used as a source of antigen.

However, the assay format selected to evaluate responses to these expressed herpesvirus antigens will likely have a profound influence on the predictive value of using these expressed antigens. What is known is that the presence of antibodies that react with these proteins in a Western blot was not capable of blocking infection with FPHV. The LETV immunized green turtle, 99A-1, possessed antibodies that recognized the expressed 38 kDa UL26 protein and glycoprotein B (Figure 6-6), but developed tumors and FPHV infection when challenged with FPHV. Of course antibody reactivity and viral neutralization are two very different endpoints, and while these proteins may not provide protection, the recombinant herpesviral proteins may provide a valuable source of antigen. Plans are underway to evaluate these expressed LETV proteins to determine whether they can take the place of LETV-infected cells lysates as a source of antigen in

ELISA assays. In addition, primers have been designed to develop PCR assay to amplify the FPHV gene homologs. The hope is that these expressed proteins will be useful source of antigen for screening marine turtle populations for exposure to herpesviruses and to establish seroepidemiology for these infections.

176

Immune Status of the Marine Turtle and FP

Overview

There is currently much debate about the health status of marine turtles and its role in

FP. The immune system of reptiles in general is not well characterized. Unlike fish, birds, and mammals, reptiles are of low economic importance and there is historically little interest in the health of these animals. In addition, there are many restrictions on collecting certain species of reptile for immunological studies due to their endangered or threatened status. Furthermore, no inbred strains of reptiles are currently available to eliminate noise cause by genetic variability. Studies of reptile immunology, particularly in the case of sea turtles, have been further limited by difficulties in establishing and maintaining the appropriate housing conditions for study in captivity. The conditions in which a reptile is kept profoundly influences its immune status and immune responses

(Munoz and De la Fuente 2001). Factors such as sex, age, season of the year, nutrition, reproductive status (hormones), and infection status have also been shown to influence the functioning of the immune system (Nelson and Demas 1996;Turner 1994;).

To further complicate issues, the types, numbers, and ratios of circulating immune cells vary drastically from species to species, such that normal values established for one species are unlikely to be accurate for evaluation of another species and in most cases will not be appropriate (Turner 1994). Thus, baseline values for a “normal” immune system must be established for each species studied. In most cases critical tools, such as monoclonal antibodies (Mabs) and lymphocyte markers, needed to rigorously define and quantify immune cells have not been developed. This is true of marine turtle immunology where although Mabs to immunoglobulin subclasses have been developed (Herbst and

Klein 1995b), Mabs to immune cell types and lymphocyte markers have not.

177

Immune function and FP

Several attempts have been made to draw correlations between the immune status of marine turtles and the prevalence of fibropapillomatosis. These studies have consistently shown that any changes observed in the immune status of a turtle (using a variety of assays to assess immune status or function) are a consequence of chronic or severe fibropapillomatosis (score 2 or higher) and not a prerequisite for the development of FP

(Work et al. 2001). This is best illustrated by one captive-reared green turtle that developed FP during the course of a mitogen study. The response of this turtle to mitogens was similar to that of the other healthy captive-reared turtles that were being evaluated to define the normal response to mitogen stimulation (Work et al. 2000). In addition, transmission studies (Herbst et al. 1995) have demonstrated that normal captive-reared turtles, similar to those used for mitogen studies, developed FP following inoculation with FP filtrates. This suggests that impairment of the immune system is not required for FP induction. Finally, it is important to note that all captive-reared green turtles with experimentally-induced FP developed an antibody response to FPHV indicating a functioning humoral immune system (Herbst et al. 1998b), and two captive- reared green turtles capable of mounting a rigorous antibody response to LETV were fully susceptible to FPHV. These studies all support the conclusion that any putative signs of a depressed immune system are likely the consequence of chronic FP and are not a prerequisite for development of FP. The effects of chronic disease regardless of its origin are well documented (Inui 2002;Nelson 2000).

Environmental cofactors and FP

Studies have also been conducted in attempt to implicate various chemicals or biotoxins as either directly involved with the etiology of FP or as cofactors influencing

178 disease severity. One study examined three turtles that stranded and died shortly thereafter in Hawaii (Miao et al. 2001). In all three cases, turtles had detectable levels of polychlorinated biphenyls (PCBs) regardless of FP status. These PCB levels were as great as ten-fold higher in adipose tissue than had been previously reported in turtles

(Miao et al. 2001). Given the small number of turtles assayed, the fact that all turtles had stranded (suggesting other factors involved leading to death, and differences in sex, reproductive maturation, and age), it is impossible to discern any useful information from this study with regard PCBs to FP etiology in spite of the claim that “chemical pollution may have a role in the etiology of FP.” What can be concluded is that three Hawaiian green turtles possessed PCBs in their tissues. The issue of the impact of chemical pollution on marine species, especially an endangered one, is of obvious importance and merits further investigation.

A similar situation exists concerning the potential role of biotoxins such as okadaic acid as a cofactor in FP (Landsburg et al. 1999). These studies suggested an association between the prevalence of benthic dinoflagellates (potential source of okadaic acid) on substrates presumed to be a preferred food source of marine turtles, and the prevalence of

FP. However, analysis of stranded green turtles with and without FP revealed no statistical difference between disease severity and okadaic acid levels in tissues. A more direct study has been performed by analyzing samples obtained from tracheal leavage of

Florida juvenile green turtles to assay for the presence of dinoflagellates on plant material actually consumed by turtles. Initial analysis had determined that the highest concentrations of dinoflagellates were observed in a population of juvenile green turtles where FP had never been documented (Llewellyn Ehrhart, personal communication).

179

While all of these studies remain in their infancy, there is currently no convincing data to indicate a role for immune suppression, pesticides, or a biotoxin in the etiology of FP.

It should be emphasized that there are currently no reagents available to assess accurately the immune status of a marine turtle. What is clear is that there is wide variability between individual marine turtles, and is consistent with the characteristics of the immune system of other reptiles. Furthermore, there is variation in clinical values obtained from captive-reared verses wild green turtles (Swimmer 2000). Baseline data must be established to define normal values for wild turtles. Until this is achieved, all studies claiming to evaluate the disease-related immune status of marine turtles should be assessed with extreme caution. There is currently a consortium of researchers in the State of Florida charged with developing reagents to define these parameters and develop improved methods to assess the health of marine turtles.

The Etiological Role of LETV in LETD

Since access to the life stage during which marine turtles would demonstrate clinical signs of LETD is limited, the significance of this disease for marine turtle health and the role of LETV in this disease are unknown. Immunological assays can determine whether marine turtles are exposed to LETV and this would begin to provide information about the significance of this virus infection to marine turtle populations. LETV exposure is of particular concern because head start facilities around the world continue to release thousands of turtles annually despite enzootic and epizootic infections.

To determine whether free-ranging green turtles were exposed to LETV, an ELISA assay specific for the detection of LETV was developed and applied to three study sites along the east coast of Florida (Chapter 5). It was determined that 21.6% of these juvenile green turtles are exposed to LETV (Figure 5-7), and based on a limited number of nesting

180 turtles, it appears that almost all adult marine turtles have been infected with LETV at some point during their lifetime (Figure 5-8). As a result, it is difficult to assess the impact of LETV infection on marine turtle populations. Without morbidity or mortality data that might accompany LETV infection and LETD in the wild, it is not possible to make statements about the role of LETV in marine turtle health. The results of LETV seroepidemiology suggests that there are wild marine turtles that survived infection with

LETV and mounted an antibody response to the infection. In addition, the high prevalence of antibodies to LETV in the small number of adult turtles evaluated suggests that most individuals are exposed to LETV at some point during their lifetime (Figure 5-

8). This is similar to the high seroprevalance of herpesviral infections in humans (Hoang et al. 1999).

While mortality rates were as high as 38% in captive-reared turtles in the index case of

LETD, is impossible to extrapolate this to predict the mortality rates of LETD in wild populations because the conditions of mariculture are significantly different than those in natural habitats. For instance, mariculture systems contain a high density of turtles that are susceptible to water quality and temperature issues that have been documented to influence the severity of herpesvirus-induced diseases (Haines and Kleese 1977). It is not known if LETD would be more or less severe in the absence of these factors in the wild.

However, free-ranging turtles face difficulties captive-reared turtles do not. Wild turtles must be self-sufficient and are not provided supportive care when they become ill.

Cconsiderations of LETD in wild turtles should not be abruptly dismissed. In fact, the signs of stomatitis, glottitis, tracheitis, and bronchopneumonia, characteristic of LETD, have been documented in mariculture and oceanarium-reared green turtles worldwide,

181 and mortality in post-hatchling and juvenile turtles can reach 70% (Glazebrook et al.

1993). Unfortunately, of the cases in wild turtles with clinical signs similar to LETD, none of these individuals were tested for the presence of or exposure to LETV. This is due to an number of factors including the lack of diagnostic assays available at the time

(PCR had not yet been developed to assay for the herpesviral DNA polymerase gene) and the lack of communication between researchers studying epidemiology and field biologists and clinicians who have contact with diseased marine turtles. Routinely, specimens that are collected are formalin fixed for histological examination. This treatment inactivates any viral or other infectious agents that may be present. Instead, some samples should be collected and preserved specifically for testing of infectious agents such as viruses. In addition, plasma should be added to the battery of samples collected and should either be banked for use in existing assays or for retrospective analysis as future tests become available.

Case Study of LETD in a Wild Loggerhead Turtle

A recent example of the value of collecting such samples and the importance of clear communication and appropriate sample preservation is with a loggerhead turtle that stranded in Monroe County, Florida. Two lung samples from this loggerhead turtle were sent to our laboratory for virological analysis. Unfortunately, the samples had completely thawed during transport and remained at ambient temperature for many hours. In addition, this sample was sent without information about the identification of this turtle

(where the turtle came from, and what were the circumstances that lead to its death, etc.).

The two lung samples were evaluated by PCR with the hope of sorting out the paperwork at a later time. Both samples were determined to be positive for the presence of the herpesviral DNA polymerase gene. A portion of each of the two lung pieces was

182 prepared independently and the total PCR product for each sample was sequenced using both forward and reverse primers. The amplification product of the two samples were identical and had an amino acid sequence that was four amino acids different that the corresponding portion of the LETV DNA polymerase gene (Figure 7-1). A filtrate was prepared from each of the lung pieces and was used as inoculum for virus isolation on tissue culture cells. These cells were maintained for 15 days, blind passaged, and maintained for another 19 days. No signs of CPE were detected, and the cultures suffered from extensive fungal contamination. The absence of infectious virus in the original tissues cannot be ruled out given the condition of the samples when they reached the laboratory. Subsequent requests to receive additional material that was still frozen were unsuccessful.

Meanwhile, as the sample was tracked down, it was determined that this one turtle had been given four different names or identification codes as it traveled from the beach to rehabilitation center to veterinary hospital and finally to the necropsy table. This included

“Hawkeye,” “Skippy,” “Cc0102,” and simply “the superbowl turtle.” As a result, it was difficult to obtained information because people involved in the intermediate steps were not aware of what had become of the turtle or of the new identification it had been given once it left their facility. What was eventually determined was that this turtle had originally stranded in Monroe County, Florida and was brought into the Turtle Hospital,

Marathon, FL for evaluation. Once this was determined it was easy to piece together the history of this turtle because of the excellent records this facility maintains. Copies of the description of condition of this turtle were obtained. The Turtle hospital described the turtle as “lethargic” with no other gross signs of disease. The turtle was transported to

183

Harmony Animal Hospital where a pathologist described mild proliferation in the trachea and superficial necrosis of mucosal epithelium with occasional synctial cell formation and numerous intranuclear inclusions. Their notes indicated a suspicion of the involvement of a virus, specifically a herpesvirus, which brought these samples to our laboratory.

Fortuitously, the Turtle hospital had collected a plasma sample when the turtle was first brought in and it was evaluated by the LETV-specific ELISA assay (Chapter 5). This is the first wild turtle with clinical signs similar to LETD to be shown also to possess antibodies to LETV. In addition, this is the only example of a LETV-related virus being detected in tissues of a wild marine turtle and is the only example of direct detection of a

LETV-related virus in the tissues of a loggerhead.

The existence of broad collaborations between field biologists, clinicians at rehabilitation centers, and research institutions, made it possible to put together a complete medical history and profile to provide information about factors potentially responsible for the disease observed in this turtle. Given the fact that most turtle rescue/rehabilitation facilities rely on a workforce of volunteers, the quality of documentation and availability of samples is extraordinary. Efforts are continually being made to develop and unify protocols for collecting samples and for improving communication between the different facilities so that these collaborations can continue to provide insight into disease of marine turtles. All individuals involved with turtle conservation are aware of how critical this is to the success of treating injured and diseased marine turtles.

184

Molecular Virology of Marine Turtle Herpesvirus

Useful information was obtained during the entire course of constructing the partial genomic LETV library (Chapter 6). This included devising solutions for obtaining genomic DNA from a virus that is tightly cell-associated. Repeated attempts to band

LETV in density gradients were unsuccessful (data not shown). Titering various fractions revealed that the majority of virus was still tightly associated with the cellular debris maintained at the top of the gradient. Pulse-field gel electrophoresis was capable of separating viral DNA from cellular genomic DNA providing a pure source of LETV

DNA for the genomic library. This revealed that the approximate genome size of the

LETV genome was 150 kilobases (Figure 6-1).

While this library contained only 18.7% of the genome (Table 6-1), it provided enough sequence information to target selected genes of interest. This was particularly important for the amplification of LETV gB (Figure 6-5). While herpesviruses are well conserved, the level of conservation was not great enough to design primers to PCR amplify gene homologs based on the sequences of herpesviruses. All attempts to PCR amplify gB from LETV and FPHV based on gB sequence information of other herpesviruses were unsuccessful (Figures 3-4 through 3-7). Genuine LETV sequence had to be obtained for the design of specific primers to PCR amplify the region of the genome containing LETV gB (Figure 6-5).

In addition, when the fraction sequences obtained from the 2D gel (Figure 6-3) were not contained in the LETV partial genomic library, these sequences were compared to a number of searchable public databases (NCBI BLAST, EMBL,SWISS-PROT). The fraction sequences could not be matched or identified to a particular herpesviral gene demonstrating the lack of conservation in this portion of the Herpesviridae genome.

185

Serendipitously, the sequence for the LETV assembly or scaffolding protein was included in one of the clones generated during the amplification of the region of the genome spanning glycoprotein B (Figure 6-3). If genuine LETV sequence had not been obtained, it would not have been possible to identify the 38 kDa protein as the LETV assembly protein (Figure 6-4).

The LETV partial genomic library was also able to provide information about genes that were not the focus of this research. Such a gene was the identification of the putative

LETV semaphorin gene (Table 6-1). In general, semaphorins are chemoattractant and/or repulsive factors essential for neuronal and lymphocyte development (He et al. 2002) and have also been implicated in bone skeleton and heart formation, tumor suppression, and immune function (Xu et al. 1998). Semaphorins have been identified in other viruses including vaccinia (VV) (Gardner et al. 2001), fowlpox (FV) (Afonso et al. 2000), and ectromelia virus (cause of mousepox) (Gardner et al. 2001;Smith and Alcami 2002). The alcelaphine herpesvirus 1 (AHV-1) is the only herpesvirus to date to be identified as possessing a semaphorin gene (Ensser et al. 1997;Ensser and Fleckenstein 1995).

Members of this gene family all have a conserved 500 amino acid “sema” domain

(Inagaki et al. 1995). Viral semaphorins, class V semaphorins, are secreted due to their lack of the transmembrane domains (found in most vertebrate and invertebrate homologs)

(Nakamura et al. 2000) and are hypothesized to have special functions in the immune system (Lange et al. 1998).

One of the clones in the LETV library contained partial sequence of a semaphorin gene (Table 6-1). The clone sequence from the library was aligned with nucleotide sequences from other virally-encoded semaphorin genes and from the nucleotides

186 sequences from human and mouse semaphorin messenger RNA (Figure 7-2). To further assess whether this was viral or host (TH-1 cells) semaphorin, primers were designed to the semaphorin library clone and a PCR assay was developed. Special care was taken to avoid designing primers to regions spanning introns that may potentially result in too great of a target for PCR. This was accomplished by identifying the position of introns in human semaphorin sequences (arrowheads in Figure 7-2) as guidelines for the potential sites of introns in the terrapene genomic sequence of semaphorin since the sequence for terrapene semaphorin is not available. A PCR product of the appropriate size was obtained when total DNA from LETV-infect cells was used as template, but not when total DNA from uninfected cell was used as template. The PCR product from the LETV- infected cells was not sequenced to confirm that the amplification product was semaphorin. However, it is hypothesized that the sequence information obtained from the

LETV genomic library corresponds to viral and not cellular semaphorin gene, given the complete absence of any introns in the library clone (as opposed to 9 introns in same region of the human homolog; Figure 7-2) and the absence of a PCR product from the uninfected cells.

Based on the high amino acid similarity between the secreted viral proteins and the cellular proteins it has been suggested that these genes were acquired from their host and modified during evolution (Gardner et al. 2001). This would be another example of molecular piracy (Chapter 1) that is well described for many oncogenic human herpesviruses (Eckhart 1998;Flaitz and Hicks 1998;Schulz and Moore 1999). The presence of this gene in LETV, or any other marine turtle herpesvirus, may provide insight in to the virulence and pathogenicity of this group of viruses.

187

Assignment of LETV and FPHV to the Alpha-herpesvirus Subfamily

Analysis of marine turtle herpesviruses LETV and FPHV has indicated these herpesviruses are more closely related to each other than to any of the other herpesviruses included in the evaluation (Figure 6-7b). In addition, evaluation of each of these herpesviruses has independently confirmed that these viruses are both appropriately placed in the alpha-herpesvirus subfamily. In the case of LETV, the characteristic of lytic replication in tissue culture, the clinical signs of LETD in both mariculture and wild green turtles, the genome organization specifically around the region containing DNA polymerase and gB (Figure 6-3), and the phylogenetic analysis fulfill all of the current criteria for assignment of a herpesvirus to a given subfamily (Chapter 1). While not as complete, similar evidence does exist for FPHV. The genome organization around DNA polymerase (Yu et al. 2000), the persistence of FPHV DNA in infected cells (Herbst et al. 1998c), and phylogenetic analysis (Figure 6-7b) all support the hypothesis that FPHV is also an alpha-herpesvirus.

The Role of Herpesvirus Infections in Marine Turtle Health and Conservation

It is difficult to assess the implications of FPHV and LETV herpesvirus infections on the global issue of marine turtle health and conservation. Limited access to turtles and their endangered status makes it impossible to follow an individual wild turtle from the time of initial infection with a herpesvirus to the onset of clinical disease (if any). As a result, the morbidity and mortality rates due to herpesvirus infections for a given population are unknown.

What is known is that some green turtles recover from FP while in others the disease progresses and is so severe that they die. Spontaneous regression has been documented by photographically monitoring individual Hawaiian green turtles over the course of

188 several years (Bennett et al. 1999). In addition, One Florida juvenile green turtle at the first encounter was FP negative, 6 months later was FP positive, and then 6 months after that was FP negative (Table 5-2). Conversely, there are documented cases of wild turtles that are brought in to rehabilitation centers where their FP tumors are removed, but these individuals continue to develop new tumors or re-growths while in captivity. Many develop internal fibromas for which there is no therapy.

It is important to note that the evaluation of the health status of any individual turtle does not necessarily reflect the status of that species of marine turtle throughout the world. For instance, the population of Florida juvenile green turtles evaluated in these studies (Chapter 5) represents only a small portion of green turtles worldwide in the same age class (1-5 years). Analysis of this group of juvenile turtles is further complicated by the fact that an individual turtle may only be evaluated one time, and this population is in a constant state of flux with turtles constantly leaving and entering this population. As such, it is not possible to assess the significance of a FP prevalence of 72.5% in the

Indian River lagoon in Florida on the population dynamics or health of marine turtles worldwide. Until there is greater access to all life stages for greater periods of time, it is very difficult to make predictions about the impact of any disease on marine turtle population dynamics.

There are projects underway to address the absence of information about the absolute numbers and geographic locations of all marine species. Researchers have come together to form a database called the global ocean biogeographic information system (OBIS) as a starting point toward developing a “consensus of marine life” (Malakoff 2000). However, even if the information on the location of every marine turtle were to become available,

189 to what does one compare this information? Records exist for certain populations of marine turtles over a period of several years, decades, or perhaps even a couple centuries.

It becomes a daunting task to try to place the current status of a given population into historical perspective. An excellent discussion of the limitation of establishing what is

“natural” or “baseline” for a species is provided by Jeremy Jackson (Jackson 2001) who argues that the definition for these terms are typically based on the perception of how things appear now or when they were first encountered and rarely does this reflect the true natural or pristine state of an ecosystem.

What cannot be denied is there has been drastic decline in numbers of specific populations of marine turtles over the past few centuries (Jackson 2001). In terms of assessing or quantifying the effects of human activity, Jackson best states this as

“changes caused by humans are the signal and the natural variability constitutes the noise that obscure the human footprint” (Jackson 2001). Given the natural variability in habitats (waxing and waning nesting beaches, constant changes in ecology of foraging grounds, etc.) it is often difficult to separate out the effect of one particular factor, such as a disease, on a complex marine ecosystem. However, there are more obvious examples of direct loss of turtles as a direct result of human activities (Table 7-1): hundreds of dead turtles in fishing nets, marine turtle eggs and carapaces for sale in international market places, and washed out nests due to an accelerated rate of beach erosion cause by construction of a seawall are a few examples. What is known is that the loss of marine turtles is not quickly replaced due to the late sexual maturation times of turtles and the fact that only a few hatchling turtles survive to reach reproductive maturity (Jackson

190

2001). This creates a situation where the impact of any given negative activity or conservation effort may have a lag time of decades (Jackson et al. 2001).

While these studies presented in this dissertation have provided important new information about the role of herpesviruses in specific diseases of marine turtles, the larger question of the impact of disease on the long-term survival of the world’s marine turtles is not something that can be answered at this time.

191

HSV1 QHGLLPCLHVAATVTTIGREMLLATREYVHARWA.AFEQLGLLPCL VAATVTTIGR MLLATREYVH RWA FE L 39

GALL1 SNGLLPCIDVAATVTTIGRNMLLTVRDYIHKQWG.TRDALGLLPCIDVAATVTTIGR MLL RDYIH WG DAL 39 HV1976 AMGLLPCLEVAATVTTVGRNMLLDTRDYIHKRWS.EREKFA GLLPCLEVAATVTTVGR MLL TRDYIH RWS E E F 39 ORTHV ATGFLPCLEVAATVTTVGRDMLLATRDFIHTRWGTDFEALA GFLPCLEVAATVTTVGR MLLATRDFIH RWG DFEAL 40 LTHV ATGFLPCLEVAATVTTVGRDMLLATRDFIHTRWGTDFEALA GFLPCLEVAATVTTVGR MLLATRDFIH RWG DFEAL 40 HGTHV ATGFLPCLEVAATVTTVGRDMLLATRDFIHTRWGTDFEALA GFLPCLEVAATVTTVGR MLLATRDFIH RWG DFEAL 40 FGTHV ATGFLPCLEVAATVTTVGRDMLLATRDFIHTRWGTDFEALA GFLPCLEVAATVTTVGR MLLATRDFIH RWG DFEAL 40 LETV GAGLLPCLEIAATVTTVGRAMLLSTRDYIHERWR.DRERFG GLLPCLEIAATVTTVGR MLL TRDYIH RW D E F 39 Cc0102 GAGLLPCLEIAATVTTVGRAMLLATRDYIHERWR.DRDRFG GLLPCLEIAATVTTVGR MLLATRDYIH RW D D F 39

HSV1 LADFPEAADMRAPG.PYSMRL DFPE A R PG SMR 58 GALL1 LREFPNLSNFMRPE.DYSVSL EFPNL F RPE SV 58 HV1976 LVDFPQMSQYVIREEPHSMRL DFPQM Y E SMR 59 ORTHV LADAPELAAFRRPESLFGLRL D PELA FRRPE GLR 60 LTHV LTDAPELAAFRRPESLFGLRL D PELA FRRPE GLR 60 HGTHV LVDAPELAAFRRPESLFGLRL D PELA FRRPE GLR 60 FGTHV LVDAPELAAFRRPESLFGLRL D PELA FRRPE GLR 60 LETV LLDFPEFREHVVSDAPHSLRL DFPEF D SLR 59 Cc0102 LLDFPEFREHVAADAPHSLRL DFPEF AD SLR 59

Figure 7-1. Amino acid sequence alignment of a portion of the herpesviral DNA polymerase gene of a LETV-like virus detected in loggerhead turtle (Cc0102) and other herpesviruses. Cc0102 was different from LETV by four amino acids (arrowheads). Various herpesviruses were evaluated and identified with full name and Genbank accession number below. Alpha-herpesviruses: HSV1 = Human simplex virus 1 (X14112), GALL1= Gallid herpesvirus 1 (AF168792), HV1976 = herpesvirus isolated from tortoise {Biermann & Blair 1994 148 /id}(unpublished sequence), ORTHV = Olive ridley turtle herpesvirus DNA polymerase (AF049904), LTHV= Loggerhead turtle herpesvirus DNA polymerase (AF035005), HGTHV= Hawaiian green turtle herpesvirus DNA polymerase (AF035003), FGTHV= Florida green turtle herpesvirus DNA polymerase (AF035004), and LETV= Lung-eye-trachea disease-associated herpesvirus (unpublished sequence).

192

Human RSEG...LLACGTNARHPSCWNLVNGTVVP.LGEMRGYAPEG LL CGTN PSCW L N LG RGYAP 99 Mouse RGNG...LLVCGTNARKPSCWNLVNDSVVMSLGEMKGYAPNG LLVCGTN PSCW L N LG KGYAP 169 LETV ...... GRSTG 4 AHV1 QTDG...LLACGTNSQKPSCWLINNLTTQF.LGPKLGLAPDG LL CGTN PSCW I N LG G AP 193 FV YDDK...ILVCGTNSSSPTCWYINGTIKEP.TPYGRGLSPD ILVCGTN P CW I RG P 135 VV VEDADKDTLVCGTNNGNPKCWKIDGSDDPK..HRGRGYAPDA LVCGTN P CW I RGYAP 106

Human FSPDENSLVLFEGDEVYSTIRKQEYNGKIPRFRRIRGESEFSPD NSLVLF G E YSTI K G I RFRRIRG E 139 Mouse FSPDENSLVLFEGDEVYSTIRKQEYNGKIPRFRRIRGESEFSPD NSLVLF G E YSTI K G I RFRRIRG E 209 LETV GS...... VSTYNLRDTVSYYKRFRRIRGGAPS ST S RFRRIRG 30 AHV1 FSPSSGNLVLFDQNDTYSTINLYKSLSGSHKFRRIAGQVEFSP LVLF D YSTI S KFRRI G E 233 FV ESYDMTGLVLIDGKEIYSTIKKYSHLSTG..FSRIVGKPVS D GLVLI G E YSTI K S F RI G 173 VV YQ...NSKVTIISYNECVLSDINISKEGIKRWRRFDGPCGY NS V I S N I RWRRF G G 143

Human LYTSDTVMQNPQFIKATIVHQDQAYDDKIYYFFREDNPDKLYTSDT M NPQF AT VH YDDKIY FF ED K 179 Mouse LYTSDTVMQNPQFIKATIVHQDQAYDDKIYYFFREDNPDKLYTSDT M NPQF AT VH YDDKIY FF ED K 249 LETV LTTSETSMRNP...... DYVRAATWDGKVYIFFNEDVEAGL TSET M NP WDGKVYIFF ED 64 AHV1 LYTSDTAMHRPQFVQATAVHKNESYDDKIYFFFQENSHSDLYTSDT M PQF AT VH YDDKIYFFF EN 273 FV LYTSSSTMKNPKFVHLVSLQETNSINDTIYIFFQEEGMAKLYTS M NP F LQ ND IYIFF EE K 213 VV YDLYTADNVIPKDGLRGAFVDKDGTYDKVYILFTDTIGSKP DKVYILF D K 183

Human NPEAPLNVSRVAQLCRGDQGGESSLSVSKWNTFLKAMLVCP V RVAQLCR DQGGESSLSVSKW TFLKA LVC 219 Mouse NPEAPLNVSRVAQLCRGDQGGESSLSVSKWNTFLKAMLVCP V RVAQLCR DQGGESSLSVSKW TFLKA LVC 289 LETV ..SIGKRVARVGQVCRNDDGGEGAYAQAKWTTFLKTELVCV RVGQVCR DDGGEG KW TFLK LVC 102 AHV1 FKQFPHTVPRVGQVCSSDQGGESSLSVYKWTTFLKARLACP V RVGQVC DQGGESSLSV KW TFLKA L C 313 FV .....VSR.....VCKHDQGGSGSLSGSKWSTFLKSIMICVCK DQGG GSLS SKW TFLK MIC 243 VV R...IVKIPYIAQMCLNDEGGPSSLSSHRWSTFLKVELECI IAQMC DEGG SSLS RW TFLK L C 220

Human SDAATNKNFNRLQDVFLLPDPSGQWRDTRVYGVFSNPWNYD T FNRLQDVF LP P G W DT YGVFSNPWNY 259 Mouse SDAATNRNFNRLQDVFLLPDPSGQWRDTRVYGVFSNPWNYD T FNRLQDVF LP P G W DT YGVFSNPWNY 329 LETV VDDGTGARFSRLVSVDFLPR.FGSWKNGLAYGLFSNEFGFD T F RL V LP GSW N YGLFSN F F 141 AHV1 VDYDTGRIYNELQDIFIWQAPENSWEETLIYGLFLSPWNFD T YN LQDIF W P SW ET YGLF PWNF 353 FV EDLN..VRFNYLKDVVVIKG..KSPNETIIYGLFFNEWNYD FN L DV I S ET YGLF N WNY 279 VV DIDG...... RSYRQIIHSRTIKTDNDTILYVFFDSPYSKR I DT Y FF PY 254

Human SAVCVYSLGDIDKVFRTSSLKGYHSSLPN.PRPGKCLPDQSAVCVYS GDID VF TS LKGYH LP PRPG CLP 298 Mouse SAVCVYSLGDIDRVFRTSSLKGYHMGLPN.PRPGMCLPKKSAVCVYS GDID VF TS LKGYH LP PRPG CLP 368 LETV SAVCVYHIGDVMQLFVTSPLKGYSGPLPV.PRPGTCVPSGSAVCVY GDV LF TS LKGY LP PRPG CVP 180 AHV1 SAVCVFTVKDIDHVFKTSKLKNYHHKLPT.PRPGQCMKNHSAVCVF DID VF TS LK YH LP PRPG CM 392 FV SAVCMFKFDKIQNNFNTSPLKGYSGGKVLSVRPGTCLNTSSAVCMF D IQ F TS LKGY RPG CL 319 VV SALCTYSMNTIKQSFSTSKLEGYTKQLPS.PAPGICLPAGSALC YS I F TS L GY LP P PG CLP 293

Human QPIPTETFQVADRHPEVAQRVEPMGPLKTPLFHSKYHYQKQ IPTETFQVAD PEVA V P P PLF SKY Y K 338 Mouse QPIPTETFQVADSHPEVAQRVEPMGPLKTPLFHSKYHYQKQ IPTETFQVAD PEVA V P P PLF SKY Y K 408 LETV K.VPASTFGVTLKYPELLLPMLPTVPANSALFFNEIPYVKVP TF V YPEL M P P ALF Y K 219 AHV1 QHVPTETFQVADRYPEVADPVYQKNNAMFPIIQSKYIYTKQ VPTETFQVAD YPEVA V PII SKY Y K 432 FV T..PRDTFEVIDLYPETLYGVKG.....DFIFKTKYTYTHP DTFEV D YPE V IF KY Y 352 VV KVVSHTTFEVIEKYNVLDDIIKPLSN...... V TFEV E Y L I P 319 Figure 7-2. Amino acid sequence alignment of the putative LETV semaphorin gene homolog and viral and cellular semaphorin genes. Human= H-SEMA-L mRNA (AAC34741), Mouse= Sema7a mRNA (NP_035482 ), AHV1= alcelaphine herpesvirus 1 semaphorin gene (T03102), FW= Fowlpox A39R gene (NP_039010), and VV= Vaccinia virus A39R gene (NC_001559). Arrowheads indicated position of introns in genomic sequence of human semaphorin (H-sema-L) gene.

193

Human VAVHRMQASHGET...FHVLYLTTDRGTIHKVVEPGEQEHVHR YLTT GTIHKVV E 375 Mouse VVVHRMQASNGET...FHVLYLTTDRGTIHKVVESGDQDHVHR YLTT GTIHKVV D 445 LETV IALHQIAVR...... ETQYLVMYLSTDVGVHRITQLMSLH YL ST V Q 251 AHV1 LLVYRVEYGGVFW...ATIFYLTTIKGTIHIYVRYEDSNSV R YLTT GTIH V D 469 FV IVINTAVINYQHKDYRVTTFYLSTSDGKIHKVVVYEDG..IN YL T G IHKVV D 390 VV ...... QPIFEG...... PSGG 328

Human SFAFNIMEIQPFRRAAAIQTMSLDAERRKLYVSSQWEVSQFNI EI PF R AAIQ M LD KLYVSSQWEVSQ 415 Mouse SFVFNIMEIQPFHRAAAIQAISLDADRRKLYVTSQWEVSQVFNI EI PF R AAIQ I LD KLYV SQWEVSQ 485 LETV GEVINALELRPP..FDAVSDFALDPLLVNAIVSSDRRAVAVIN EL P AV F LD VSSDR 289 AHV1 TTALNILEINPFQKPAPIQNILLDNTNLKLYVNSEWEVSELNI EI PF K APIQ I LD KLYV SEWEVSE 509 FV ..VINVIELTLKQYPSPVLALVSDERSEKLFVSYNDSTIEVINV EL PV L D KLFVS E 428 VV VKWFDIKEKENEHREYRIYFIKENSIYSFDTKSKQTRSSQFDI E R I I N S Q SQ 368

Human VPLDLCEVYGGGCHGCLMSRDPYCGWDQGRCISIYSSERSVPLDLC VYGG C CLMSRDPYCGW Q C S 455 Mouse VPLDMCEVYSGGCHGCLMSRDPYCGWDQDRCVSIYSSQRSVPLDMC VYSG C CLMSRDPYCGW Q C S 525 LETV YRLDACTWYPDACEDCILARDPYCGWGHRGCESAIT..AALD C Y D C CIL RDPYCGW H C S 327 AHV1 VPLDLCSVYGNDCFSCFMSRDPLCTWYNNTCSFKQRVSVEVPLDLC VYG C CFMSRDP C W C 549 FV LPLAFCHLYGGTCDSCLLSRDPHCGWTNIDCVYGGEKKLLLPL FC LYGG C CLLSRDP CGW C 468 VV VDARLFSVMVTSKPLFIADIGIGVGMPQMKKILKMV L V I G G Q 403

Human VLQSINPAEPHKECPNPKPDKAPLQKVSLAPNSRYYLSCPV Q IN A C KP P KVSL S YLSCP 495 Mouse VLQSINPAEPHRECPNPKPDEAPLQKVSLARNSRYYLTCPV Q IN A C KP P KVSL S YL CP 565 LETV VFQAIRFAKIGRVCNLTPHPAVPVRNVRVTVGADTYFRCRV Q I A C P V V YF C 367 AHV1 TGGPANRTLSEMCGDHYAP.TVVKHQVSIPLLSNSYLSCPN P VSI S YLSCP 588 FV QKDIYDVPKNICSGSLIKR.EPFSRKVYLSSSSYHVLSCPD D P K KV L S LSCP 507 VV

Human MESRHATYSWRHKENVEQSCEPGHQSPNCILFIENLTAQQES HATY W N E CE CILFI NLTA 535 Mouse MESRHATYLWRHEENVEQSCEPGHQSPSCILFIENLTARQES HATY W N E CE CILFI NLTA 605 LETV VDSYWATHVWTKDKVEVLRCEKDR..LPCVYYVANATAADDS AT W CE CV YV N TA 405 AHV1 AVSNHADYFWTKDGFTEKRCHVKTHKNDCILLIANSTTATS HA Y W E C CILLI N T 628 FV IESHQANYVWVNKHNKTIVDCGPDNNDMCYFFIYNLYDDNES QA Y W N C FFI NL 547 VV

Human YGHYFCEAQEGSYFREAQHWQLLPEDGIMAEHLLGHACA.G Y C EG R L D 574 Mouse YGHYRCEAQEGSYLREAQHWELLPEDRALAEQLMGHARA.G Y C EG R L D 644 LETV AGTYACRESERDVERVKTIVRLTVARPKLNQAGRQITTRRG Y C E R L 445 AHV1 NGTHVCNMKE...... D...... G C E D 639 FV FGKYTCTSEEGWNKETVMIEELSKVNKDYSKRQLQEHAVNG Y C EG L N 587 VV

Human ...... LAASLWLGVLPTLTLGLLVS L V TL V 593 Mouse ...... LAASFWLGVLPTLILGLLVS L V TL V 663 LETV PPVTTERPRPRTQTRTERPIT 466 AHV1 ...... SVTVKLLEVNVTLML V TL 653 FV KKGNTLITCSKFLIIFVYLFVFIIVS L I L V 612 VV

Figure 7-2 Continued

194

Table 7-1. Anthropogenic threats to marine turtles.a

Activity Brief explanation

Habitat Loss Encroachment of buildings; harvesting sea grass beds Harvest of meat, shell, Activities still active around the world & eggs Beach driving Direct interference with turtles; compaction of sand & existing nests Sea walls Block assess to nesting beach; promote erosion result in complete loss

Beach renourishment Sand type & consistency incompatible with netting; loss of original site

Beach clean-up Modification to beach characteristics (compaction); disruption of nests Sand mining On- or off-shore results in loss or modification to habitat; see “Dredging” Humans on beach Disruption of nesting turtle & obstruction to nesting sites (false-crawl); disruption of nests Dredging Uptake by dredge resulting in injury or death; destruction of foraging grounds Nuclear power plant intake Uptake into channels & potential death to turtle Fisheries by-catch Drowning or injury in nets or ropes or fishing lines Entanglement Drowning or injury in discarded trash or fishing gear Chemical pollution Death or illness caused by petroleum products, pesticides, etc. Boat strikes Drowning or injury Artificial lighting Interference with nesting turtle & hatchling disorientation Increased predation Attracted to urban settings & habitat loss of predator aFor complete review see Lutz and Musick, 1996.

REFERENCES

Afonso, C. L., Tulman, E. R., Lu, Z., Zsak, L., Kutish, G. F., and Rock, D. L. (2000). The genome of fowlpox virus. J.Virol. 74, 3815-3831.

Aguirre, A. A., Balazs, G. H., Zimmerman, B., and Spraker, T. R. (1994). Evaluation of Hawaiian green turtles (Chelonia mydas) for potential pathogens associated with fibropapillomas. J.Wildl.Dis. 30, 8-15.

Ambrosius, H. (1976). Immunoglobulins and antibody production in reptiles. pp. 298- 334. (Blackwell Scientific Publications: Oxford.)

Arvanitakis, L., Mesri, E. A., Nador, R. G., Said, J. W., Asch, A. S., Knowles, D. M., and Cesarman, E. (1996). Establishment and characterization of a primary effusion (body cavity-based) lymphoma cell line (BC-3) harboring Kaposi's sarcoma- associated herpesvirus (KSHV/HHV-8) in the absence of Epstein-Barr virus. Blood 88, 2648-2654.

Balazs, G. H. and Pooley, G. S. (1991). Current status of fibropapillomas in the Hawiian green turtle, Chelonia mydas. US Dept Commerce. NOAA Tech Memo NMFS- SEFSC-156, 47-51.

Beasley, J. N., Patterson, L. T., and McWade, D. H. (1970). Transmission of Marek's disease by poultry house dust and chicken dander. Am.J.Vet.Res. 31, 339-344.

Bennett, P. A., Keuper-Bennett, U, and Balazs, G. H. (1999). Photographic evidence for the regression of fibropapillomas afflicting green turtles at Honokowai, Maui, in the Hawaiian Islands In Proceedings of the Twentieth Annual Symposium on Sea Turtle Biology and Conservation. US Dept Commerce. NOAA Tech Memo NMFS-SEFSC-443.

Biermann, B. and Blair, S. (1994). First isolation of a herpesvirus from tortoises with diptheroid-necrotizing stomatitis. Second World Congress of Herpetology. Abstracts 6, 27.

Bowne, W. B., Lewis, J. J., Filippa, D. A., Niesvizky, R., Brooks, A. D., Burt, M. E., and Brennan, M. F. (1999). The management of unicentric and multicentric Castleman's disease: a report of 16 cases and a review of the literature. Cancer 85, 706-717.

195 196

Brown, D. R.,Lackovich, J. K.,and Klein, P. A. (1999) Further evidence for absence of papillomaviruses from sea turtle fibropapillomas. Veterinary Record 145, 616- 617.

Brown, D. R., Schumacher, I. M., Nogueira, M. F., Richey, L. J., Zacher, L. A., Schoeb, T. R., Vliet, K. A., Bennett, R. A., Jacobson, E. R., and Brown, M. B. (2001). Detection of antibodies to a pathogenic mycoplasma in american alligators (Alligator mississippiensis), broad-nosed caimans (Caiman latirostris), and siamese crocodiles (Crocodylus siamensis) J.Clin.Microbiol. 39, 285-292.

Cannon, J. S., Ciufo, D., Hawkins, A. L., Griffin, C. A., Borowitz, M. J., Hayward, G. S., and Ambinder, R. F. (2000). A new primary effusion lymphoma-derived cell line yields a highly infectious Kaposi's sarcoma herpesvirus-containing supernatant. J.Virol. 74, 10187-10193.

Caselli, E., Balboni, P. G., Incorvaia, C., Argnani, R., Parmeggiani, F., Cassai, E., and Manservigi, R. (2000). Local and systemic inoculation of DNA or protein gB1s- based vaccines induce a protective immunity against rabbit ocular HSV-1 infection. Vaccine 19, 1225-1231.

Casey, R. N., Quackenbush, S. L., Work, T. M., Balazs, G. H., Bowser, P. R., and Casey, J. W. (1998). Evidence for retrovirus infections in green turtles, Chelonia mydas, from the Hawaiian islands. Dis.Aquat.Organ. 31, 1-7.

Clark, H. F. and Karzon, D. T. (1972). Iguana virus, a herpes-like virus isolated from cultured cells of a lizard, Iguana iguana. Infect.Immun. 5, 559-569.

Clark, K. J., Grant, P. G., Sarr, A. B., Belakere, J. R., Swaggerty, C. L., Phillips, T. D., and Woode, G. N. (1998). An in vitro study of the aflavins extracted from black tea to neutralize bovine rotavirus and bovine coronavirus infections. Vet.Microbiol. 63, 147-157.

Coligan, J. E., Kruisbeek, A. M., Margulies, D. H., Shevach, E. M., and Strober, W. (2001). Current Protocols in Immunology. (John Wiley & Sons, Inc.: New York.)

Collier, L. (2000). Human Virology. (Oxford University Press: New York.)

Cooper, J. E., Gschmeissner, S., and Holt, P. E. (1982). Viral particles in a papilloma from a green lizard (Lacerta viridis). Lab.Anim. 16, 12-13.

Cooper, J. E., Jackson, O. F., and Harshbarger, J. C. (1983). A neurilemmal sarcoma in a tortoise (Testudo hermanni). J.Comp.Pathol. 93, 541-545.

Corchero, J. L., Mar, E. C., Spira, T. J., Pellett, P. E., and Inoue, N. (2001). Comparison of serologic assays for detection of antibodies against human herpesvirus 8. Clin.Diagn.Lab Immunol. 8, 913-921.

197

Cotran, R. S., Kumar, V., and Robbins, S. L. (1994). Pathologic Basis of Disease. (W.B. Sanders Company: Philadelphia.)

Cox, W. R., Rapley, W. A., and Barker, I. K. (1980). Herpesvirus-like infection in a painted turtle (Chrysemys picta). J.Wildl.Dis. 16, 445-449.

Crowther, J. R. (2001). Methods in Molecular Biology. The ELISA Guidebook. (Humana Press: Totowa.)

Curry, S. S., Brown, D. R., Gaskin, J. M., Jacobson, E. R., Ehrhart, L. M., Blahak, S., Herbst, L. H., and Klein, P. A. (2000). Persistent infectivity of a disease- associated herpesvirus in green turtles after exposure to seawater. J.Wildl.Dis. 36, 792-797.

De Maeyer, E. and Schonne, E. (1964). Starch gel as an overlay for the plaque assays of animal viruses. Virology 24, 13-18.

Dillner, J. (1999). The serological response to papillomaviruses. Semin.Cancer Biol. 9, 423-430.

Drago, F. and Rebora, A. (1999). The new herpesviruses: emerging pathogens of dermatological interest. Arch.Dermatol. 135, 71-75.

Drew, M. L., Phalen, D. N., Berridge, B. R., Johnson, T. L., Bouley, D., Weeks, B. R., Miller, H. A., and Walker, M. A. (1999). Partial tracheal obstruction due to chondromas in ball pythons (Python regius). J.Zoo.Wildl.Med. 30, 151-157.

Eckhart, W. (1998). Viruses and human cancer. Sci.Prog. 81 ( Pt 4), 315-328.

Ensser, A. and Fleckenstein, B. (1995). Alcelaphine herpesvirus type 1 has a semaphorin- like gene. J.Gen.Virol. 76 ( Pt 4), 1063-1067.

Ensser, A., Pflanz, R., and Fleckenstein, B. (1997). Primary structure of the alcelaphine herpesvirus 1 genome. J.Virol. 71, 6517-6525.

Evans, A. S. (1976). Causation and disease: the Henle-Koch postulates revisited. Yale J.Biol.Med. 49, 175-195.

Ezzell, C. (1999). It came from the deep. Sci.Am. 280, 22, 24.

Fields, B.N., Knipe, D.M., Chanock, M.S., Hirsch, M.S., Melnick, J.L., Monath,T.P., and Roizmann, B (1990). Fields Virology. Second Edition. (Raven Press: New York.)

Fields, B.N., Knipe, D.M., and Howley, P.M. (1996). Fundamental Virology. Third Edition. (Lippincott-Raven Publishers: Philadelphia.)

Ferry, J. A. and Harris, N. L. (1997). Atlas of lymphoid hyperplasia and lymphoma. (W.B. Saunders Company: Philadelphia.)

198

Flaitz, C. M. and Hicks, M. J. (1998). Molecular piracy: the viral link to carcinogenesis. Oral Oncol. 34, 448-453.

Foster, K. R., Bernstein, D. E., and Huber, P. W. (1993). Phantom Risk, Scientific Inference and the Law. (The MIT Press: London.)

Franti, M., Aubin, J. T., Saint-Maur, G., Kosuge, H., Yamanishi, K., Gautheret-Dejean, A., Garbarg-Chenon, A., Huraux, J. M., and Agut, H. (2002). Immune reactivity of human sera to the glycoprotein B of human herpesvirus 7. J.Clin.Microbiol. 40, 44-51.

Frye, F. L. (1994). Diagnosis and surgical treatment of reptilian neoplasms with a compilation of cases 1966-1993. In Vivo 8, 885-892.

Frye, F. L., Oshiro, L. S., Dutra, F. R., and Carney, J. D. (1977). Herpesvirus-like infection in two Pacific pond turtles. J.Am.Vet.Med.Assoc. 171, 882-884.

Fujioka, R. S., Loh, P. C., and Lau, L. S. (1980). Survival of human enteroviruses in the Hawaiian ocean environment: evidence for virus-inactivating microorganisms. Appl.Environ.Microbiol. 39, 1105-1110.

Gallo, R. C. (1998). The enigmas of Kaposi's sarcoma. Science 282, 1837-1839.

Gardner, J. D., Tscharke, D. C., Reading, P. C., and Smith, G. L. (2001). Vaccinia virus semaphorin A39R is a 50-55 kDa secreted glycoprotein that affects the outcome of infection in a murine intradermal model. J.Gen.Virol. 82, 2083-2093.

Girones, R., Jofre, J. T., and Bosch, A. (1989). Isolation of marine bacteria with antiviral properties. Can.J.Microbiol. 35, 1015-1021.

Glazebrook, J. S., Campbell, R. S., and Thomas, A. T. (1993). Studies on an ulcerative stomatitis-obstructive rhinitis-pneumonia disease complex in hatchling and juvenile sea turtles, Chelonia mydas and Caretta caretta. Dis.Aquat.Organ. 16, 133-147.

Gordon, A. N., Kelly, W. R., and Cribb, T. H. (1998). Lesions caused by cardiovascular flukes (Digenea: Spirorchidae) in stranded green turtles (Chelonia mydas). Vet.Pathol. 35, 21-30.

Gordon, A. N., Kelly, W. R., and Lester, R. J. (1993). Epizootic mortality of free-living green turtles, Chelonia mydas, due to coccidiosis. J.Wildl.Dis. 29, 490-494.

Goyal, S. M., Gerba, C. P., and Melnick, J. L. (1979). Human enteroviruses in oysters and their overlying waters. Appl.Environ.Microbiol. 37, 572-581.

199

Graczyk, T. K., Aguirre, A. A., and Balazs, G. H. (1995). Detection by ELISA of circulating anti-blood fluke (Carettacola, Hapalotrema, and Learedius) immunoglobulins in Hawaiian green turtles (Chelonia mydas). J.Parasitol. 81, 416-421.

Haines, H. and Kleese, W. C. (1977). Effect of water temperature on a herpesvirus infection of sea turtles. Infect.Immun. 15, 756-759.

Harshbarger, J. C. (1984). Pseudoneoplasms in ectothermic animals. Natl.Cancer Inst.Monogr. 65, 251-273.

Hashido, M. (2000). [Seroepidemiology of alphaherpesviruses]. Nippon.Rinsho. 58, 828- 837.

Hauser, B., Mettler, F., and Rubel, A. (1983). Herpesvirus-like infection in two young boas. J.Comp.Pathol. 93, 515-519.

He, Z., Wang, K. C., Koprivica, V., Ming, G., and Song, H. J. (2002). Knowing how to navigate: mechanisms of semaphorin signaling in the nervous system. Sci.STKE. 2002, RE1.

Hejkal, T. W. and Gerba, C. P. (1981). Uptake and survival of enteric viruses in the blue crab, Callinectes sapidus. Appl.Environ.Microbiol. 41, 207-211.

Helmick, K. E., Bennett, R. A., Ginn, P., DiMarco, N., Beaver, D. P., and Dennis, P. M. (2000). Intestinal volvulus and stricture associated with a leiomyoma in a green turtle (Chelonia mydas). J.Zoo.Wildl.Med. 31, 221-227.

Henle, J. (1938). On Miasmata and Contagie. (Johns Hopkins Press:Baltimore.)

Herbst, L. H. (1994). Fibropapillomatosis of marine turtles. Ann.Rev.Fish Dis. 4, 389- 425.

Herbst, L. H., Chakrabarti, R., Klein, P. A., and Achary, M. (2001). Differential gene expression associated with tumorigenicity of cultured green turtle fibropapilloma- derived fibroblasts. Cancer Genet.Cytogenet. 129, 35-39.

Herbst, L. H., Garber, R. L., Lopez, C., and Klein, P. A. (1998a). Molecular biological evidence for the involvement of a unique herpevirus in the etiology of green turtle fibropapillomatosis. In Proceedings of the Sixteenth Annual Symposium on Sea Turtle Biology and Conservation.US Dept Commerce. NOAA Tech Memo NMFS-SEFSC-412, 96.

Herbst, L. H., Greiner, E. C., Ehrhart, L. M., Bagley, D. A., and Klein, P. A. (1998b). Serological association between spirorchidiasis, herpesvirus infection, and fibropapillomatosis in green turtles from Florida. J.Wildl.Dis. 34, 496-507.

200

Herbst, L. H., Jacobson, E. R., Klein, P. A., Balazs, G. H., Moretti, R., Brown, T., and Sundberg, J. P. (1999). Comparative pathology and pathogenesis of spontaneous and experimentally induced fibropapillomas of green turtles (Chelonia mydas). Vet.Pathol. 36, 551-564.

Herbst, L. H., Jacobson, E. R., Moretti, R., Brown, T., Sundberg, J. P., and Klein, P. A. (1995). Experimental transmission of green turtle fibropapillomatosis using cell- free tumor extracts. Dis.Aquat.Organ. 22, 1-12.

Herbst, L. H. and Klein, P. A. (1995a). Green turtle fibropapillomatosis: challenges to assessing the role of environmental cofactors. Environ.Health Perspect. 103 Suppl 4, 27-30.

Herbst, L. H. and Klein, P. A. (1995b). Monoclonal antibodies for the measurement of class-specific antibody responses in the green turtle, Chelonia mydas. Vet.Immunol.Immunopathol. 46, 317-335.

Herbst, L. H., Moretti, R. H., Brown, T., and Klein, P. A. (1996). Sensitivity of the transmissible green turtle fibropapillomatosis agent to chloroform and ultracentrifugation conditions. Dis.Aquat.Organ. 25, 225-228.

Herbst, L. H., Sundberg, J. P., Shultz, L. D., Gray, B. A., and Klein, P. A. (1998c). Tumorigenicity of green turtle fibropapilloma-derived fibroblast lines in immunodeficient mice. Lab.Anim.Sci. 48, 162-167.

Hill, A. G. (1965). Environment and disease: Association or causation? Proc.R.Soc.Med. 58, 295-300.

Hirama, S. and Ehrhart, L. M. Epizootiology of green turtle fibropapillomatosis on the Florida Atlantic coast (USA). In Proceedings of the Twentieth Annual Symposium on Sea Turtle Biology and Conservation. US Dept Commerce. NOAA Tech Memo NMFS-SEFSC, in press.

Hoang, M. P., Rogers, B. B., Dawson, D. B., and Scheuermann, R. H. (1999). Quantitation of 8 human herpesviruses in peripheral blood of human immunodeficiency virus-infected patients and healthy blood donors by polymerase chain reaction. Am.J Clin.Pathol. 111, 655-659.

Hoog, S. S., Smith, W. W., Qiu, X., Janson, C. A., Hellmig, B., McQueney, M. S., O'Donnell, K., O'Shannessy, D., DiLella, A. G., Debouck, C., and Abdel-Meguid, S. S. (1997). Active site cavity of herpesvirus proteases revealed by the crystal structure of herpes simplex virus protease/inhibitor complex. Biochemistry 36, 14023-14029.

201

Huerta, P., Pineda, H., Aguirre, A. A., Spraker, T. R., Sarti, L., and Barragan, A. First confirmed case of fibropapilloma in a leatherback turtle (Dermochelys coriacea). In Proceedings of the Twentieth Annual Symposium on Sea Turtle Biology and Conservation. U.S. Dep. Commerce, NOAA Tech. Memo. NMFS-SEFSC, in press.

Inagaki, S., Furuyama, T., and Iwahashi, Y. (1995). Identification of a member of mouse semaphorin family. FEBS Lett. 370, 269-272.

Inui, A. (2002). Cancer anorexia-cachexia syndrome: current issues in research and management. CA Cancer J.Clin. 52, 72-91.

Jackson, J. B. (2001). What was natural in the coastal oceans? Proc.Natl.Acad.Sci. U.S.A 98, 5411-5418.

Jackson, J. B., Kirby, M. X., Berger, W. H., Bjorndal, K. A., Botsford, L. W., Bourque, B. J., Bradbury, R. H., Cooke, R., Erlandson, J., Estes, J. A., Hughes, T. P., Kidwell, S., Lange, C. B., Lenihan, H. S., Pandolfi, J. M., Peterson, C. H., Steneck, R. S., Tegner, M. J., and Warner, R. R. (2001). Historical overfishing and the recent collapse of coastal ecosystems. Science 293, 629-637.

Jacobson, E., Chrisman, C., Homer, B. L., Detrisac, C., Dunmore, D., Lopez, A., Greiner, E., Szabo, N., Coberley, S., Klein, P., Ewing, R., Mader, D, Schaf, S., Moretti, R., Mette, N., Landsburg, J. H., and Shelton, D. Pathologic findings in loggerhead, Caretta caretta, with polyneuropathy in coastal waters off south Florida. In Proceedings of the Reptile and Amphibian Pathology Colloquium, in press.

Jacobson, E. R., Gaskin, J. M., Clubb, S., and Calderwood, M. B. (1982a). Papilloma-like virus infection in Bolivian side-neck turtles. J.Am.Vet.Med.Assoc. 181, 1325- 1328.

Jacobson, E. R., Gaskin, J. M., Roelke, M., Greiner, E. C., and Allen, J. (1986). Conjunctivitis, tracheitis, and pneumonia associated with herpesvirus infection in green sea turtles. J.Am.Vet.Med.Assoc. 189, 1020-1023.

Jacobson, E. R., Gaskin, J. M., and Wahlquist, H. (1982b). Herpesvirus-like infection in map turtles. J Am.Vet.Med.Assoc. 181, 1322-1324.

Jacobson, E. R., Harris, R. K., Buregelt, C., and Williams, B. (1991). Herpesvirus in cutaneous fibropapillomas of the green turtle, Chelonia mydas. Dis.Aquat.Organ. 12, 1-16.

Janert, B. (1998). A fibrosarcoma in a Siamese crocodile (Crocodylus siamensis). J.Zoo.Wildl.Med. 29, 72-77.

Janeway, C. A. and Travers, P. (1996). Immunobiology: the immune system in health and disease. (Current Biology Ltd./Garland Publishing Inc.: New York.)

202

Kabisch, D. and Frost, J. W. (1994). Isolation of herpesvirus from Testudo hermanni and Agrionemys horsfieldii. Verh ber Erkrg Zootiere 36, 241-245.

Karlin, S., Mocarski, E. S., and Schachtel, G. A. (1994). Molecular evolution of herpesviruses: genomic and protein sequence comparisons. J.Virol. 68, 1886- 1902.

Kennedy, S. (1998). Morbillivirus infections in aquatic mammals. J.Comp.Pathol. 119, 201-225.

Kennedy, S., Kuiken, T., Jepson, P. D., Deaville, R., Forsyth, M., Barrett, T., van de Bildt, M. W., Osterhaus, A. D., Eybatov, T., Duck, C., Kydyrmanov, A., Mitrofanov, I., and Wilson, S. (2000). Mass die-off of Caspian seals caused by canine distemper virus. Emerg.Infect.Dis. 6, 637-639.

Knipe, D.M., and Howley P.M. (2001). Fields Virology. (Lippincott Williams and Wilkins: Philadelphia.)

Koch, R. (1884). Die Aetiologie der Tuberculose. Mitt.Kaiser.Gesundh. 2, 1.

Koff, R. S. (1995). Seroepidemiology of hepatitis A in the United States. J.Infect.Dis. 171, S19-S23.

Kumar, V., Cotran, R. S., and Robbins, S. L. (1997). Basic Pathology.(W.B. Sanders Company: Philadelphia.)

LaBelle, R. L., Gerba, C. P., Goyal, S. M., Melnick, J. L., Cech, I., and Bogdan, G. F. (1980). Relationships between environmental factors, bacterial indicators, and the occurrence of enteric viruses in estuarine sediments. Appl.Environ.Microbiol. 39, 588-596.

Lackovich, J. K., Brown, D. R., Homer, B. L., Garber, R. L., Mader, D. R., Moretti, R. H., Patterson, A. D., Herbst, L. H., Oros, J., Jacobson, E. R., Curry, S. S., and Klein, P. A. (1999). Association of herpesvirus with fibropapillomatosis of the green turtle Chelonia mydas and the loggerhead turtle Caretta caretta in Florida. Dis.Aquat.Organ. 37, 89-97.

Landsburg, J. H., Balazs, G. H., Steidinger, K. A., Baden, D. G., Work, T. M., and Russell, D. J. (1999). The potential role of natural tumor promoters in marine turtle fibropapillomatosis. J.Aquat.Anim.Health 11, 199-210.

Lange, C., Liehr, T., Goen, M., Gebhart, E., Fleckenstein, B., and Ensser, A. (1998). New eukaryotic semaphorins with close homology to semaphorins of DNA viruses. Genomics 51, 340-350.

Latimer, K. S. and Rich, G. A. (1998). Colonic adenocarcinoma in a corn snake (Elaphe guttata guttata). J.Zoo.Wildl.Med. 29, 344-346.

203

Laufs, R. and Steinke, H. (1976). Vaccination of nonhuman primates with killed oncogenic herpesviruses. Cancer Res. 36, 704-706.

Leibovitz, L., Rebell, G., and Boucher, G. C. (1978). Caryospora cheloniae sp. n.: a coccidial pathogen of mariculture-reared green sea turtles (Chelonia mydas mydas). J.Wildl.Dis. 14, 269-275.

Liddell, J. E. and Cryer, A. (1991). A Practical Guide to Monoclonal Antibodies. (John Wiley: New York.)

Lipson, S. M. and Stotzky, G. (1984). Effect of proteins on reovirus adsorption to clay minerals. Appl.Environ.Microbiol. 48, 525-530.

Lo, S., Gilbert, J., and Hetrick, F. (1976). Stability of human enteroviruses in estuarine and marine waters. Appl.Environ.Microbiol. 32, 245-249.

Loomis-Huff, J. E., Eberle, R., Lockridge, K. M., Rhodes, G., and Barry, P. A. (2001). Immunogenicity of a DNA vaccine against herpes B virus in mice and rhesus macaques. Vaccine 19, 4865-4873.

Lu, Y., Aguirre, A. A., Work, T. M., Balazs, G. H., Nerurkar, V. R., and Yanagihara, R. (2000a). Identification of a small, naked virus in tumor-like aggregates in cell lines derived from a green turtle, Chelonia mydas, with fibropapillomas. J.Virol.Methods 86, 25-33.

Lu, Y., Qigui, Y., Zamzow, J. P., Wang, Y., Losey, G. S., Balazs, G. H., Nerurkar, V. R., and Yanagihara, R. (2000b). Detection of green turtle herpesviral sequence in saddleback wrasse Thalassoma duperrey: A possible mode of transmission of green turtle fibropapilloma. J.Aquat.Anim.Health 12, 58-63.

Lucke, B. (1938). Studies of tumors in cold-blooded vertebrates. Annual Report of the Tortugas Laboratory of the Carnegie Institute 38, 92-94.

Lutz, P. L. and Musick, J. A. (1997). The Biology of Sea Turtles. (CRC Press: Boca Raton.)

Mahl, M. C. and Sadler, C. (1975). Virus survival on inanimate surfaces. Can.J.Microbiol. 21, 819-823.

Mahy, B. W. J. and Kangro, H. O. (1996). Virology Methods Manual. (Academic Press: San Diego.)

Malakoff, D. (2000). Marine census. Grants kick off ambitious count of all ocean life. Science 288, 1575-1576.

204

Marschang, R. E., Frost, J. W., Gravendyck, M., and Kaleta, E. F. (2001). Comparison of 16 chelonid herpesviruses by virus neutralization tests and restriction endonuclease digestion of viral DNA. J.Vet.Med.B Infect.Dis.Vet.Public Health 48, 393-399.

Marschang, R. E., Gravendyck, M., and Kaleta, E. F. (1997). Herpesviruses in tortoises: investigations into virus isolation and the treatment of viral stomatitis in Testudo hermanni and T. graeca. Zentralbl.Veterinarmed.[B] 44, 385-394.

Mendelsohn, J., Howley, P. M., Israel, M. A., and Liotta, L. A. (1995). The Molecular Basis of Cancer. (W.B. Sanders Company: Philadelphia.)

Miao, X. S., Balazs, G. H., Murakawa, S. K., and Li, Q. X. (2001). Congener-specific profile and toxicity assessment of PCBs in green turtles (Chelonia mydas) from the Hawaiian Islands. Sci. Total Environ. 281, 247-253.

Moore, M. K., Work, T. M., Balazs, G. H., and Docherty, D. E. (1998). Preparation, cryopreservation, and growth of cells prepared from the green turtle (Chelonia mydas). Meth.Cell Sci. 19, 168.

Munoz, F. J. and De la Fuente, M. (2001). The effect of the seasonal cycle on the splenic leukocyte functions in the turtle Mauremys caspica. Physiol.Biochem.Zool. 74, 660-667.

Murray, P. R., Kobayashi, G. S., Pfaller, M. A., and Rosenthal, K. S. (1994). Medical Microbiology. (Mosby-Year Book, Inc.: St. Louis.)

Nakamura, F., Kalb, R. G., and Strittmatter, S. M. (2000). Molecular basis of semaphorin-mediated axon guidance. J.Neurobiol. 44, 219-229.

National Research Council (1990). Decline of the sea turtles: causes and prevention. (National Academy Press: Washington, D.C.)

Nelson, K. A. (2000). The cancer anorexia-cachexia syndrome. Semin.Oncol. 27, 64-68.

Nelson, R. J. and Demas, G. E. (1996). Seasonal changes in immune function. Q.Rev.Biol. 71, 511-548.

Nettles, V. F. Jr. (1992). Wildlife diseases and population medicine. J.Am.Vet.Med.Assoc. 200, 648-652.

Origgi, F. C. and Jacobson, E. R. (2000). Diseases of the respiratory tract of chelonians. Veterinary Clin.North Am.Exot.Anim.Pract. 3, 537-49, viii.

Page, R. D. (1996). TreeView: an application to display phylogenetic trees on personal computers. Comput.Appl.Biosci. 12, 357-358.

205

Parcells, M. S., Lin, S. F., Dienglewicz, R. L., Majerciak, V., Robinson, D. R., Chen, H. C., Wu, Z., Dubyak, G. R., Brunovskis, P., Hunt, H. D., Lee, L. F., and Kung, H. J. (2001). Marek's disease virus (MDV) encodes an interleukin-8 homolog (vIL- 8): characterization of the vIL-8 protein and a vIL-8 deletion mutant MDV. J.Virol. 75, 5159-5173.

Patti, A. M., Bagnod, S., Marinaro, Manduca A., Santi, A. L., Mastroeni, I., and Fara, G. M. (1996). Survival of enteric viruses in the marine environment. Ann.Ig 8, 341- 348.

Pau, C. P., Lam, L. L., Spira, T. J., Black, J. B., Stewart, J. A., Pellett, P. E., and Respess, R. A. (1998). Mapping and serodiagnostic application of a dominant epitope within the human herpesvirus 8 ORF 65-encoded protein. J.Clin.Microbiol. 36, 1574-1577.

Pennisi, E. (1999). Is it time to uproot the tree of life? Science 284, 1305-1307.

Pereira, L. (1994). Function of glycoprotein B homologues of the family herpesviridae. Infect.Agents Dis. 3, 9-28.

Peterson, B. A. and Frizzera, G. (1993). Multicentric Castleman's disease. Semin.Oncol. 20, 636-647.

Pettan-Brewer, K. C., Drew, M. L., Ramsay, E., Mohr, F. C., and Lowenstine, L. J. (1996). Herpesvirus particles associated with oral and respiratory lesions in a California desert tortoise (Gopherus agassizii). J.Wildl.Dis. 32, 521-526.

Quackenbush, S. L., Work, T. M., Balazs, G. H., Casey, R. N., Rovnak, J., Chaves, A., duToit, L., Baines, J. D., Parrish, C. R., Bowser, P. R., and Casey, J. W. (1998). Three closely related herpesviruses are associated with fibropapillomatosis in marine turtles. Virology 246, 392-399.

Raidal, S. R., Ohara, M., Hobbs, R. P., and Prince, R. I. (1998). Gram-negative bacterial infections and cardiovascular parasitism in green sea turtles (Chelonia mydas). Aust.Vet.J. 76, 415-417.

Rebell, G., Rywlin, A., and Haines, H. (1975). A herpesvirus-type agent associated with skin lesions of green sea turtles in aquaculture. Am.J.Vet.Res. 36, 1221-1224.

Richardson, A. (1997). Is breast cancer caused by late exposure to a common virus? Med.Hypotheses 48, 491-497.

Rivers, T. M. (1937). Viruses and Koch's postulates. J.Bacteriol. 33, 1-12.

Roizmann, B., Desrosiers, R. C., Fleckenstein, B., Lopez, C., Minson, A. C., and Studdert, M. J. (1992). The family Herpesviridae: an update. The Herpesvirus Study Group of the International Committee on Taxonomy of Viruses. Arch.Virol. 123, 425-449.

206

Sarid, R., Olsen, S. J., and Moore, P. S. (1999). Kaposi's sarcoma-associated herpesvirus: epidemiology, virology, and molecular biology. Adv.Virus Res. 52, 139-232.

Schoenbaum, M. A., Freund, J. D., and Beran, G. W. (1991). Survival of pseudorabies virus in the presence of selected diluents and fomites. J.Am.Vet.Med.Assoc. 198, 1393-1397.

Schultze, A. E., Mason, G. L., and Clyde, V. L. (1999). Lymphosarcoma with leukemic blood profile in a Savannah monitor lizard (Varanus exanthematicus). J.Zoo.Wildl.Med. 30, 158-164.

Schulz, T. F. and Moore, P. S. (1999). Kaposi's sarcoma-associated herpesvirus: a new human tumor virus, but how? Trends Microbiol. 7, 196-200.

Schumacher, I. M., Brown, M. B., Jacobson, E. R., Collins, B. R., and Klein, P. A. (1993). Detection of antibodies to a pathogenic mycoplasma in desert tortoises (Gopherus agassizii) with upper respiratory tract disease. J.Clin.Microbiol. 31, 1454-1460.

Sheaffer, A. K., Newcomb, W. W., Brown, J. C., Gao, M., Weller, S. K., and Tenney, D. J. (2000). Evidence for controlled incorporation of herpes simplex virus type 1 UL26 protease into capsids. J.Virol. 74, 6838-6848.

Simpson, C. F., Jacobson, E. R., and Gaskin, J. M. (1979). Herpesvirus-like infection of the venom gland of Siamese cobras. J.Am.Vet.Med.Assoc. 175, 941-943.

Simrell, C. R. and Klein, P. A. (1979). Antibody responses of tumor-bearing mice to their own tumors captured and perpetuated as hybridomas. J.Immunol. 123, 2386-2394.

Siscovick, D. S., Schwartz, S. M., Caps, M., Wang, S. P., and Grayston, J. T. (2000). Chlamydia pneumoniae and atherosclerotic risk in populations: the role of seroepidemiology. J.Infect.Dis. 181 Suppl 3, S417-S420.

Smith, G. M. and Coates, C. W. (1938). Fibro-epithelial growths of the skin in large marine turtles Chelonia mydas. Zoologica 23, 93-98.

Smith, V. P. and Alcami, A. (2002). Inhibition of interferons by ectromelia virus. J.Virol. 76, 1124-1134.

Speckner, A., Kropff, B., Knor, S., and Mach, M. (2000). The antigenic domain 1 of human cytomegalovirus glycoprotein B contains an intramolecular disulphide bond. J.Gen.Virol. 81, 2659-2663.

Swimmer, J. Y. (2000). Biochemical responses to fibropapilloma and captivity in the green turtle. J.Wildl.Dis. 36, 102-110.

207

Tigue, N. J., Matharu, P. J., Roberts, N. A., Mills, J. S., Kay, J., and Jupp, R. (1996). Cloning, expression and characterization of the proteinase from human herpesvirus 6. J.Virol. 70, 4136-4141.

Turner, R.J. (1994). Immunology: A Comparative Approach. (John Wiley & Sons: West Sussex.)

Tyler, J. W. and Cullor, J. S. (1989). Titers, tests, and truisms: rational interpretation of diagnostic serologic testing. J.Am.Vet.Med.Assoc. 194, 1550-1558.

Une, Y., Uemura, K., Nakano, Y., Kamiie, J., Ishibashi, T., and Nomura, Y. (1999). Herpesvirus infection in tortoises (Malacochersus tornieri and Testudo horsfieldii). Vet.Pathol. 36, 624-627.

VanDevanter, D. R., Warrener, P., Bennett, L., Schultz, E. R., Coulter, S., Garber, R. L., and Rose, T. M. (1996). Detection and analysis of diverse herpesviral species by consensus primer PCR. J.Clin.Microbiol. 34, 1666-1671.

Westphal, E. M., Blackstock, W., Feng, W., Israel, B., and Kenney, S. C. (2000). Activation of lytic Epstein-Barr virus (EBV) infection by radiation and sodium butyrate in vitro and in vivo: a potential method for treating EBV-positive malignancies. Cancer Res. 60, 5781-5788.

Wilson, M. R. and Coussens, P. M. (1991). Purification and Characterization of infectious Marek's disease virus genomes using pulse field electrophoresis. Virology 185, 673-680.

Work, T. M., Balazs, G. H., Rameyer, R. A., Chang, S. P., and Berestecky, J. (2000). Assessing humoral and cell-mediated immune response in Hawaiian green turtles, Chelonia mydas. Vet.Immunol.Immunopathol. 74, 179-194.

Work, T. M., Rameyer, R. A., Balazs, G. H., Cray, C., and Chang, S. P. (2001). Immune status of free-ranging green turtles with fibropapillomatosis from Hawaii. J.Wildl.Dis. 37, 574-581.

Xu, X., Ng, S., Wu, Z. L., Nguyen, D., Homburger, S., Seidel-Dugan, C., Ebens, A., and Luo, Y. (1998). Human semaphorin K1 is glycosylphosphatidylinositol-linked and defines a new subfamily of viral-related semaphorins. J.Biol.Chem. 273, 22428-22434.

Yu, Q., Lu, Y., Nerurkar, V. R., and Yanagihara, R. (2000). Amplification and analysis of DNA flanking known sequences of a novel herpesvirus from green turtles with fibropapilloma Brief report. Arch.Virol. 145, 2669-2676. zur Hausen, H. (1991). Viruses in human cancers. Science 254, 1167-1173.

BIOGRAPHICAL SKETCH

Sadie Shea Coberley was born in Gold Beach, Oregon, September 26, 1974, and grew up in the small coastal community of Brookings, Oregon. After graduating from high school she attended Oregon State University and received her Bachelor of Science degree with honors in June of 1996. In the spring of 1995, she attended OSU Hatfield Marine

Science Center, where she received advanced training in marine biology. In 1997 she began her studies toward a Doctor of Philosophy degree in the Interdisciplinary Program in Biomedical Sciences. During this period she received several awards for best student poster at International Sea Turtle meetings and at University of Florida’s Graduate

Student Council forum. In 2000, she was awarded a fellowship for Outstanding Graduate

Research.

Sadie was married November 20, 1999, to Carter Coberley, formerly of Portland,

Oregon, and now they continue to purse their research interests at the U.S. Army Medical

Research Institute of Infectious Diseases at Fort Detrick, Frederick, Maryland, and Gene

Logic Corporation, Gaithersburg, Maryland, respectively.

208