TAURA SYNDROME VIRUS (TSV) OF PENAEID SHRIMP: INFECTION OF Penaeus monodon, RESISTANCE OF Litopenaeus vannamei AND

Transcription

1 TAURA SYNDROME VIRUS (TSV) OF PENAEID SHRIMP: INFECTION OF Penaeus monodon, RESISTANCE OF Litopenaeus vannamei AND ULTRASTRUCTURE OF THE REPLICATION SITE IN INFECTED CELLS by Thinnarat Srisuvan A Dissertation Submitted to the Faculty of the DEPARTMENT OF VETERINARY SCIENCE AND MICROBIOLOGY In Partial Fulfillment of the Requirements For the Degree of DOCTOR OF PHILOSOPHY WITH A MAJOR IN PATHOBIOLOGY In the Graduate College THE UNIVERSITY OF ARIZONA 2006

2 2 THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Thinnarat Srisuvan entitled Taura syndrome virus (TSV) of Penaeid Shrimp: Infection of Penaeus monodon, Resistance of Litopenaeus vannamei and Ultrastructure of the Replication Site in Infected Cells and recommend that it be accepted as fulfilling the dissertation requirement for the Degree of Doctor of Philosophy Date: October 23, 2006 Donald V. Lightner, Ph.D. Date: October 23, 2006 Carlos Reggiardo, D.V.M., Ph.D. Date: October 23, 2006 David G. Besselsen, D.V.M., Ph.D. Date: October 23, 2006 Kathy F. J. Tang-Nelson, Ph.D. Date: October 23, 2006 Michael W. Riggs, D.V.M., Ph.D. Final approval and acceptance of this dissertation is contingent upon the candidate s submission of the final copies of the dissertation to the Graduate College. I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement. Date: October 23, 2006 Dissertation Director: Donald V. Lightner

3 3 STATEMENT BY AUTHOR This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the library. Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interest of scholarship. In all other instances, however, permission must be obtained from the author. Thinnarat Srisuvan

4 4 DEDICATION To my parents: Ms. Charoensri Srisuvan (นางเจร ญศร ศร ส วรรณ ) and Mr. Thaveesit Srisuvan (นายทว ส ทธ ศร ส วรรณ ) กราบเท าพ อแม ล กขออ ท ศความส าเร จในการศ กษาน ให ก บพ อแม กราบขอบพระค ณท ท านได เฝ าถนอมเล ยงด ให การอบรมส งสอนให ล กเป นคนด และเป นก าล งใจในยามท ล กท อแท ล กร กท านท ง สองและร ส าน กในพระค ณท านเสมอมา ขออ านาจค ณพระศร ร ตนตร ยและส งศ กด ส ทธ ในสากล จง โปรดดลบ นดาลให ท านท งสองม ส ขภาพแข งแรง ม อาย ย นยาว อย เป นร มโพธ ร มไทรให ล กท งสองคน ตลอดไป

5 5 TABLE OF CONTENTS Page LIST OF FIGURES 6 LIST OF TABLES 8 ABSTRACT 9 CHAPTER 1. INTRODUCTION 11 CHAPTER 2. EXPERIMENTAL INFECTION OF Penaeus monodon WITH TAURA SYNDROME VIRUS (TSV) 18 Abstract 18 Introduction 19 Materials and Methods 20 Results and Discussion 23 CHAPTER 3. COMPARISON OF FOUR TAURA SYNDROME VIRUS (TSV) ISOLATES IN ORAL CHALLENGE STUDIES WITH Litopenaeus vannamei UNSELECTED OR SELECTED FOR RESISTANCE TO TSV 36 Abstract 36 Introduction 37 Materials and Methods 39 Results and Discussion 44 CHAPTER 4. ULTRASTRUCTURE OF THE REPLICATION SITE IN TAURA SYNDROME VIRUS (TSV)-INFECTED CELLS 60 Abstract 60 Introduction 61 Materials and Methods 63 Results 70 Discussion 75 REFERENCES 95

6 6 LIST OF FIGURES Figure 1.1 Geographic distribution of Taura syndrome virus (TSV) 1.2 Transmission electron micrograph of TSV 1.3 Schematic diagram of the genome organization of TSV 2.1 Photomicrographs of consecutive Penaeus monodon tissue sections Page tested by hematoxylin and eosin (H&E) stained and in situ hybridization (ISH) tissue sections illustrating TSV lesions 2.2 Phylogenetic neighbor-joining (NJ) tree of capsid protein 2 (CP2, amino acids) from 24 TSV isolates 2.3 Comparison of deduced amino acid sequences of CP2 from TSV isolates with reference to UsHi94 (GenBank no. AF277675) 3.1 Survival curves of TSV-resistant (TSR) and TSV-susceptible 52 (Kona) Litopenaeus vannamei after challenge by feeding with 4 TSV isolates: Bz01, Th04, UsHi94, and Ve Photomicrographs of consecutive Litopenaeus vannamei tissue 53 sections tested for TSV lesions by H&E staining and ISH 3.3 SYBR-Green real-time reverse transcription polymerase chain 55 reaction (RT-PCR)

7 7 LIST OF FIGURES - Continued Figure 3.4 Means ± standard errors of TSV copy numbers µl 1 RNA within Page 57 pleopods of TSR and Kona Litopenaeus vannamei after challenge by feeding with 4 virus isolates: Bz01, Th04, UsHi94, and Ve Ultrastructural changes in cells at early stages of infection with 80 TSV 4.2 Ultrastructural changes in cells at the mid-stages of an acute phase 81 infection with TSV 4.3 Ultrastructural changes in cells at late stages of infection with TSV 4.4 Transmission electron micrographs of TSV in an infected cell 4.5 Detection of TSV with ISH in paraffin-embedded tissues by light microscopy 4.6 Detection of TSV with ISH in resin-embedded tissues by light 85 microscopy 4.7 Ultrastrutural features of cells at late stages of TSV infection 86 examined by ISH using TSV-specific cdna probes Ultrastructural features of membrane rearrangement in TSVinfected cells from gills analyzed by ISH using TSV-specific cdna probes 4.9 Ultrastructural visualization of TSV in an infected cell tested by 93 ISH using TSV-specific cdna probes

8 8 LIST OF TABLES Table 2.1 Taura syndrome virus (TSV) isolates used for experimental Page 27 bioassays and sequence analysis 2.2 Initial appearance of dead Penaeus monodon and Litopenaeus 28 vannamei and cumulative mortality after TSV infection 3.1 TSV isolates used for the oral challenge studies 3.2 Reproducibility of the SYBR-Green reverse transcription polymerase chain reaction (RT-PCR) assay 3.3 Comparison of a conventional RT-PCR with the SYBR-Green 58 real-time RT-PCR for TSV within pleopods of TSV-resistant (TSR) and TSV-susceptible (Kona) Litopenaeus vannamei after challenge by feeding with 4 virus isolates

9 9 ABSTRACT Clinical signs and lesions of Taura syndrome virus (TSV) infection in Penaeus monodon were investigated by histological and in situ hybridization (ISH) analyses. Mortality among P. monodon inoculated with 2 genotypic variants of TSV (Th04Pm and Th04Lv) appeared on Day 3, with 2 out of 10 shrimp dying. Severe necrosis of cuticular epithelial cells and lymphoid organ spheroids, indicative of acute and chronic phase lesions of TSV infection, respectively, were detected in the samples. Both Th04Pm and Th04Lv belonged to a phylogenetic family of Asian TSV isolates. The results demonstrate that both mortality and histological lesions are associated with TSV infection in P. monodon. Infection with 4 genotypic variants of TSV (Bz01, Th04, UsHi94, and Ve05) in TSV-resistant (TSR) and TSV-susceptible (Kona) Litopenaeus vannamei was investigated. Survival probabilities of TSR shrimp were higher than those for Kona shrimp with all 4 variants. Th04, UsHi94, and Ve05 gave no Taura syndrome lesions with TSR shrimp. In contrast, TSR shrimp challenged with Bz01 and Kona shrimp with all 4 TSV variants exhibited severe necrosis of cuticular epithelial cells and lymphoid organ spheroids. Real-time reverse transcription polymerase chain reaction (RT-PCR) revealed that mean TSV copy numbers in TSR shrimp infected with Bz01, Th04, and UsHi94 were significantly (p < ) lower than those in Kona shrimp. In contrast, mean TSV copy numbers in TSR and Kona shrimp infected with Ve05 were not

10 10 significantly different (p > 0.4). The results show that TSR L. vannamei are susceptible to infection but give high survival rates following challenge by all 4 variants of TSV. To identify the viral replication site within shrimp infected cells, the viral RNA was located in association with virus-induced membrane rearrangement by electron microscopic ISH. Ultrastructure in the infected cells, analyzed by transmission electron microscopy, included the induction and proliferation of intracellular vesicle-like membranes, while the intracytoplasmic inclusion bodies and pyknotic nuclei were frequently seen. TSV RNA and TSV particles were found to be associated with the membranous structures. The results suggest that the proliferating membranes carry the RNA replication complex and that they are the site of nascent viral RNA synthesis.

11 11 CHAPTER 1 INTRODUCTION Taura syndrome (TS) is an economically important disease in cultured penaeid shrimp that is listed by World Organization for Animal Health (OIE 2006). It was first recognized in affected farms near the mouth of the Taura River, Ecuador, in June 1992 (Jimenez 1992, Lightner et al. 1995, 1996). Following its recognition, TS has been reported in many countries of the Americas, Asia, and Africa, and it has caused considerable economic loss to the shrimp farming industry in the last decade (Fig. 1.1) (Do et al. 2006, Hasson et al. 1995, 1999a, Lightner et al. 1995, Nielsen et al. 2005, Srisuvan et al. 2005, Tang & Lightner 2005, Tu et al. 1999, Yu & Song 2002). TS is caused by Taura syndrome virus (TSV), a member of Dicistroviridae (Mayo 2005). TSV is a non-enveloped icosahedral virus with a diameter of 32 nm (Fig. 1.2) (Bonami et al. 1997) and a single-stranded, positive-sense RNA genome of 10,205 nucleotides comprising 2 open reading frames (ORF1 and ORF2; Fig. 1.3) (Mari et al. 2002). ORF1 may code for non-structural proteins, including helicase, protease, and RNA-dependent RNA polymerase, whereas ORF2 codes for structural proteins such as the 3 major capsid proteins: CP1, CP2, and CP3.

12 12 TSV is associated with high mortalities in most life stages of Pacific white shrimp Litopenaeus vannamei (also called Penaeus vannamei) (Farfante & Kensley 1997, Lightner et al. 1996). Other species such as L. stylirostris and L. setiferus can also be infected with TSV and exhibit similar clinical signs (Overstreet et al. 1997, Erickson et al. 2002). Recently, TSV was detected in black tiger shrimp P. monodon by reverse transcription polymerase chain reaction (RT-PCR); however, no clinical signs or lesions characteristics of infection were found (Chang et al. 2004, Nielsen et al. 2005). Diagnosis of TSV infection in P. monodon is important because this species is a predominant species in shrimp aquaculture (FAO 2004). In Chapter 2, using histology and in situ hybridization (ISH), we demonstrated that both mortality and histological lesions are associated with TSV infection in P. monodon (Srisuvan et al. 2005). TSV isolated at different times and/or from different locations has been reported to display phenotypic variations. For instance, Erickson et al. (2005) demonstrated that L. vannamei experimentally infected with a Belize isolate (Bz02) showed higher mortality than they did when infected with a Hawaiian isolate (UsHi94) (referred to as BLZ02TSV and Hi94TSV, respectively, in the cited paper). Phylogenetic analysis revealed that Bz02 was distinct among 29 isolates of TSV, despite the fact that TSV displays low genetic variation from 0 to 5.6% in nucleotide sequence and from 0 to 7% in deduced amino acid sequence (Tang & Lightner 2005). There were 3 distinct phylogenetic lineages, with 2 in the Americas (UsHi94 and Bz02) and 1 in Asia. Nielsen et al. (2005) also reported that TSV isolates from Asia and the Americas were distinct.

13 13 Because TS causes high mortalities in L. vannamei (Lightner 1996), scientists have attempted to develop TSV-resistant shrimp populations (White et al. 2002, Wyban 2000, Xu et al. 2003). This is important because L. vannamei is the predominant cultivated species worldwide (FAO 2004). In Chapter 3, we used histology, ISH, conventional RT-PCR, and SYBR-Green real-time RT-PCR to study the course of TSV infection in a population of L. vannamei selected for TSV resistance and in an unselected population (Srisuvan et al. 2006). The intracellular biogenesis of TSV remains largely uninvestigated although the virus has become an intense subject of research for almost 15 yr. Ultrastructural pathogenesis of many viruses, e.g. Human parechovirus type 1, Dengue virus, Hepatitis C virus, Foot-and-mouth disease virus, and Severe acute respiratory syndromeassociated coronavirus, has been characterized using transmission electron microscopy (TEM), ISH, and immuno-electron microscopy (Krogerus et al. 2003, Gosert et al. 2003, Goldsmith et al. 2004, Grief et al. 1997, Monaghan et al. 2004). Additionally, our laboratory has previously developed an ISH protocol using a specific cdna probe to follow the intracellular translocation of Hepatopancreatic parvovirus (HPV) in penaeid shrimp (Pantoja & Lightner 2001). Infection by all single-stranded, positive-sense RNA viruses is believed to involve the intracellular rearrangement of membranes in the cytosol (for a review, see Mackenzie 2005, Novoa et al. 2005). The host cell membranes function as the replication site for the

14 14 synthesis of the nascent viral genomes. For instance, in many viruses such as Mouse hepatitis virus, Rubella virus, and Semliki Forest virus, these consist of generation and proliferation of endoplasmic reticulum and membrane vesicles that accumulate in the perinuclear region of infected cells (Gosert et al. 2002, Kujala et al. 2001, Magliano et al. 1998). Characterization of the viral replication complexes is an important aspect in virology and cell biology. In Chapter 4, the ultrastructure of the replication site in cells infected with TSV is visualized by TEM and ISH (Srisuvan et al. in press).

15 15 Figure 1.1. Geographic distribution of Taura syndrome virus (TSV). Stars of different colors represent 4 phylogenetic families of TSV isolates based on the capsid protein 2 (CP2) (D. V. Lightner et al. unpubl. data).

16 16 Figure 1.2. Transmission electron micrograph of Taura syndrome virus (TSV). Purified TSV suspension was subjected to a negative staining with 2% phosphotungstic acid. The photograph was taken by Jean-Robert Bonami in Scale bar = 50 nm

17 17 Figure 1.3. Schematic diagram of the genome organization of Taura syndrome virus (TSV). Numbers indicate nucleotide positions. Open reading frames (ORF1 and ORF2) are shown as open boxes and untranscribed regions (UTRs) as a single line. The approximate positions of helicase (H), protease (P), and RNA-dependent RNA polymerase (RdRp) are indicated. Arrows represent the N termini of capsid proteins (CP1, CP2, and CP3). The diagram is modified from Mari et al. (2002).

18 18 CHAPTER 2 EXPERIMENTAL INFECTION OF Penaeus monodon WITH TAURA SYNDROME VIRUS (TSV) Thinnarat Srisuvan 1, 2, *, Kathy F. J. Tang 2, Donald V. Lightner 2 1 Department of Livestock Development, 69/1 Phayathai Road, Bangkok 10400, Thailand 2 Department of Veterinary Science and Microbiology, University of Arizona, 1117 E. Lowell, Tucson, Arizona 85721, USA * thinnarat@hotmail.com Abstract: Clinical signs and lesions of Taura syndrome virus (TSV) infection in Penaeus monodon have not been documented although the virus has been detected in this shrimp species by reverse transcription polymerase chain reaction (RT-PCR). This study provides the first evidence of TSV infection in P. monodon by histological and in situ hybridization (ISH) analyses. We performed experimental bioassays with groups of P. monodon using inocula of P. monodon and Litopenaeus vannamei (Th04Pm and Th04Lv, respectively), which were collected from Thailand in 2004 and found to be positive for TSV by RT-PCR. Samples of shrimp for histological and ISH analyses were collected on Days 2, 14, and 28 post-inoculation. Mortality among TSV-inoculated P. monodon appeared on Day 3, with 2 out of 10 shrimp dying. Severe necrosis of cuticular epithelial

19 19 cells and lymphoid organ spheroids, indicative of acute and chronic phase lesions of TSV infection, respectively, were detected in the samples. Sequence analyses of the capsid protein 2 (CP2) gene showed that Th04Pm and Th04Lv isolates were different; however, both belonged to a phylogenetic family of Asian TSV isolates. The results of this study demonstrated that both mortality and histological lesions are associated with TSV infection in P. monodon. Key words: Taura syndrome virus TSV Penaeus monodon TSV infection Introduction Taura syndrome (TS) is one of the most important diseases of cultured penaeid shrimp. Since the disease was first described in Ecuador in 1992, it has rapidly spread to many countries of the world (Jimenez 1992, Hasson et al. 1995, 1999a, Lightner et al. 1995, Tu et al. 1999, Yu & Song 2002). The causative agent is Taura syndrome virus (TSV), which was placed in the family Dicistroviridae (Mayo 2002, 2005). TSV is a nonenveloped, icosahedral virus of 32 nm in diameter (Bonami et al. 1997). The genome of TSV is a single-stranded positive-sense RNA of 10,205 nucleotides and consists of 2 large open reading frames (ORF1 and ORF2) (Mari et al. 2002). ORF1 may code for non-structural proteins, including helicase, protease, and RNA-dependent RNA

20 20 polymerase, whereas ORF2 codes for structural proteins such as the 3 major capsid proteins: CP1, CP2, and CP3 (Mari et al. 2002). TS is associated with high mortalities in most life stages of Pacific white shrimp Litopenaeus vannamei (Lightner 1996). Other species such as L. stylirostris and L. setiferus can also be infected with TSV and exhibit similar clinical signs (Overstreet et al. 1997, Erickson et al. 2002). Recently, TSV was detected in black tiger shrimp Penaeus monodon by reverse transcription polymerase chain reaction (RT-PCR); however, no clinical signs or lesions characteristic of infection were found (Chang et al. 2004, Nielsen et al. 2005). Diagnosis of TSV infection in P. monodon is important because this species is a predominant species in shrimp aquaculture (FAO 2004). In the present study, using histology and in situ hybridization (ISH), we demonstrated that both mortality and histological lesions are associated with TSV infection in P. monodon. Materials and Methods Experimental bioassays. During 2004, mortalities were observed in shrimp farms in the provinces of Chachoengsao and Chumporn, Thailand. Shrimp, Penaeus monodon and Litopenaeus vannamei, were collected from the farms and sent to our laboratory for diagnosis. We detected the presence of TSV in these samples by traditional RT-PCR and real-time RT-PCR (Nunan et al. 1998, Tang et al. 2004). We did not detect the presence of any other pathogens of concern to penaeid shrimp in these samples (data

21 21 not shown). To determine if these 2 TSV isolates, Th04Pm and Th04Lv (Table 1), contained infectious TSV, we performed bioassays with groups of P. monodon (obtained from a shrimp producer in Hawaii) and specific-pathogen-free (SPF) (Lotz 1997) Kona stock L. vannamei (Oceanic Institute). These shrimp were tested and determined to be free of TSV before use. The bioassays comprised 3 experimental groups; Groups 1 and 2 served as the test groups, whereas Group 3 served as the negative control. Each group had 1 aquarium for P. monodon (10 shrimp, avg. wt = 3 g) and another for L. vannamei (10 shrimp, avg. wt = 1 g). Each shrimp in Group 1 was administered a single injection (~100 µl), into their third tail segment, of a tissue homogenate prepared from frozen TSV-infected P. monodon (Th04Pm), whereas each shrimp in Group 2 was injected (~100 µl) with a tissue homogenate prepared from frozen TSV-infected L. vannamei (Th04Lv). The tissue homogenates were prepared from shrimp cephalothoraxes as described by Hasson et al. (1995), and diluted 1:10 with 2% saline prior to inoculation. Shrimp in Group 3 were not exposed to TSV. All of the shrimp were fed once a day with a commercial pelleted feed (Rangen 35%, Buhl), for 28 d. On Days 2 and 14 postinoculation (p.i.), 2 shrimp were sampled from each test group. At termination of the experiment (28 d p.i.), all of the survivors were sampled. The tails of sampled shrimp were frozen at 70 C for RT-PCR (Nunan et al. 1998, data not shown), whereas the cephalothoraxes were fixed overnight in Davidson s fixative and transferred to 70% alcohol for hematoxylin and eosin (H&E) histological examination and ISH analyses (Lightner 1996, Mari et al. 1998). We used a mixture of probes Q1 and P15 that

22 22 hybridize the TSV genome at nucleotides 3218 to 4139 and 5915 to 7140, respectively, for ISH (Mari et al. 1998). Sequence analyses. We extracted total RNA, from either pleopods or gills of the shrimp samples, using a High Pure RNA tissue kit (Roche Biochemical) according to the manufacturer s instructions. RT-PCR was performed using a SuperScript one-step RT- PCR system with Platinum Taq DNA polymerase (Invitrogen). The CP2 region (nucleotides 7901 to 9203) of the TSV genome was amplified with primers 55P1 (5'- GGC GTA GTG AGT AAT GTA GC-3') and 55P2 (5'-CTT CAG TGA CCA CGG TAT AG-3') (Erickson et al. 2002). The RT-PCR profile was 30 min at 50 C, followed by 40 cycles of 30 s at 94 C, 30 s at 55 C, and 1.5 min at 68 C. An aliquot of the amplified product was analyzed in a 1% agarose gel containing ethidium bromide. The amplified product of CP2 was cleaned with a QIAquick PCR purification kit (Qiagen) and sequenced from both strands with an automated DNA sequencer, ABI Prism 377 (Applied Biosystems) at the sequencing facility, University of Arizona. Nucleotide sequences were aligned with Sci Ed Central software (Scientific & Educational Software), and the correct nucleotide sequences were determined. Multiple alignments of the deduced CP2 amino acid sequences (383 amino acids) from 24 TSV isolates (Table 2.1) were analyzed by CLUSTAL X (Thompson et al. 1997) and GeneDoc software (Nicholas et al. 1997). Phylogenetic analysis based on the neighbor-joining (NJ) methods of these TSV isolates was performed at 1,000 bootstrap replicates using MEGA software (Kumar et al. 2001).

23 23 Results and Discussion Mortalities and lesions of TSV infection in Penaeus monodon. To study TSV infection in Penaeus monodon as well as Litopenaeus vannamei, we performed experimental bioassays using 2 Thai TSV isolates: Th04Pm and Th04Lv (Table 2.1). For both virus isolates, inoculated P. monodon showed mortalities at Day 3 p.i., with 2 out of 10 shrimp dying and all other shrimp surviving for the rest of the study (Table 2.2). For L. vannamei, mortalities first appeared at Day 2 p.i., with 8 out of 10 shrimp dying by Day 6 p.i. These results may be because (1) L. vannamei (avg. wt 1 g) received a larger (3 times more) dose of tissue homogenate by weight than did P. monodon (avg. wt 3 g), or (2) L. vannamei were less tolerant to TSV than P. monodon. No mortality was observed among P. monodon or L. vannamei in the negative control group (Group 3). P. monodon developed lesions when injected with Th04Pm. One of 2 specimens (sampled at Day 2 p.i.) showed characteristic acute phase lesions of TSV infection (Hasson et al. 1995, Lightner et al. 1995), indicated by severe necrosis in various epithelial tissues, including gills and cuticular and stomach epithelia (Fig. 2.1A). When the consecutive tissue section of this shrimp was analyzed using ISH, a positive reaction to TSV-specific gene probes was detected as blue-black precipitates (Fig. 2.1B). This specimen exhibited a normal lymphoid organ by histology (Fig. 2.1C). However, using ISH, its consecutive lymphoid organ section displayed a strong positive reaction, and the reaction was seen only at the peripheral sheath cells of lymphoid organ tubules (Fig.

24 24 2.1D). The same lesion was also seen in L. vannamei at the early transition phase of TSV infection through ISH (Hasson et al. 1999b). Moreover, a somewhat similar lesion was also recognized in P. monodon infected with Yellow head virus (YHV) when examined by immunohistochemistry based on an YHV-specific monoclonal antibody (Soowannaya et al. 2002). In chronic phase, lesions of TSV infection and lymphoid organ spheroids (LOS) (Hasson et al. 1999b) were present in 6 specimens (2 sampled at Day 14 p.i., and 4 survivors collected at Day 28 p.i.), and consecutive lymphoid organ sections of 2 specimens (1 sampled at Day 14 p.i. and the other at Day 28 p.i.) reacted to the TSVspecific gene probes by ISH (data not shown to avoid redundancy with Fig. 2.1E to H). P. monodon also developed lesions when injected with Th04Lv. LOS were detected in 3 specimens (2 sampled at Day 14 p.i., Fig. 2.1E; 1 at Day 28 p.i., Fig. 2.1G). The consecutive lymphoid organ sections of 2 P. monodon (1 sampled at Day 14 p.i., 1 at Day 28 p.i.) reacted to the TSV-specific gene probes by ISH (Fig. 2.1F,H). This positive reaction indicated that LOS formation in these P. monodon was associated with TSV infection. Acute phase lesions of TSV infection were not detected in any P. monodon injected with Th04Lv. This result may be due to (1) an incorrect early sampling time (Day 2 p.i.), (2) the genetic characteristics of this TSV isolate, or (3) a lower inoculating dosage that resulted in a low grade acute phase infection followed by a typical chronic phase infection.

25 25 As expected, L. vannamei, injected with either Th04Pm or Th04Lv and sampled at Day 2 p.i., showed the acute phase lesions of TSV infection, whereas other L. vannamei, which were not sampled, died by Day 6 p.i. and were found to be positive for TSV by RT-PCR (data not shown). At termination of the experiment (Day 28 p.i.), lesions indicative of TSV infection, as described by Lightner et al. (1995) and Hasson et al. (1995, 1999b), were not detected in any P. monodon or L. vannamei in the negative control group (data not shown). Sequence analyses of Th04Pm and Th04Lv. We sequenced the CP2 gene of the 2 isolates Th04Pm and Th04Lv from their original tissue samples. These 2 TSV isolates were found to be different by 1.5% in the amino acid sequence. Phylogenetic analysis of 24 TSV isolates (Fig. 2.2) also revealed that Th04Pm and Th04Lv were 2 distinct isolates although both belong to a phylogenetic family of Asian TSV isolates. This result was consistent with each isolate being collected from geographically distant locations in Thailand. Comparison of amino acid similarities of these 24 TSV isolates, according to Poch et al. 1990, revealed that the CP2 amino acid variations at positions 71, 83, 230, 301, and 366 may act as genetic markers of Asian TSV isolates because they were found in many TSV isolates collected from China, Indonesia, Taiwan, and Thailand (Fig. 2.3). The phylogenetic tree also showed that Er04Pm isolate formed a cluster with Mx98 isolate (Fig. 2.2); Er04Pm isolate was collected from a farm in Eritrea (East Africa) in which both P. monodon and L. vannamei were cultured and found to be infected with TSV (D. V. Lightner unpubl. data). The results from the sequence analyses (Figs. 2.2 & 2.3) also

26 26 suggest that TSV sequences from P. monodon and L. vannamei from the same area tend to resemble each other. In other words, P. monodon appear to be susceptible to any TSV types that are prevalent in the area where they are cultivated. In conclusion, P. monodon was found to be susceptible to TSV. Using experimental bioassays, we provide the first evidence of TSV infection in P. monodon by histological and ISH analyses. TSV may also be an important pathogen in P. monodon, as observed in Thailand and Eritrea, because of the ability of the virus to produce persistent chronic (carrier) infections in the species and because of its potential to cause mortalities under some conditions. Finally, our results indicated that additional studies are needed to elucidate the prevalence and role of TSV in P. monodon aquaculture. Acknowledgements. This work was supported by Gulf Coast Research Laboratory Consortium Marine Shrimp Farming Program, CSREES, USDA, Grant no T. S. was supported by a scholarship from the Royal Government of Thailand. We thank Mr. Robins McIntosh (Charoen Pokapand, Thailand) for providing the shrimp tissues infected with TSVTh04Pm and TSVTh04Lv.

27 27 Table 2.1. Taura syndrome virus (TSV) isolates used for experimental bioassays and sequence analysis. Sequences were retrieved as non-redundant sequences from GenBank unless otherwise indicated. Taxonomy nomenclature of penaeid shrimp is according to Farfante & Kensley (1997). TSV isolate Collection location Source species Collection year GenBank no. Bz01 Belize Litopenaeus vannamei 2001 AY Cn03-1 China L. vannamei 2003 AY Cn03-2 China L. vannamei 2003 AY Cn03-3 China L. vannamei 2003 AY Cn03-4 China L. vannamei 2003 AY Cn03-5 Hainan Island, China L. vannamei 2003 DQ Er04Pm Massawa, Eritrea Penaeus monodon 2004 DQ Id03 Surabaya, Indonesia L. vannamei 2003 DQ Mm03Pm Myanmar P. monodon 2003 AY Mx98 Sinaloa, Mexico L. vannamei 1998 AF Mx99 Nayarit, Mexico L. stylirostris 1999 AF Mx2K Sonora, Mexico L. stylirostris 2000 AF Th03-1 Chachoengsao, Thailand L. vannamei 2003 AY Th03-2 Ratchaburi, Thailand L. vannamei 2003 AY Th03-3 Nakorn Pathom, Thailand L. vannamei 2003 AY Th03-4 Nakorn Pathom, Thailand L. vannamei 2003 AY Th03-5 Samut Sakorn, Thailand L. vannamei 2003 DQ Th03-6 Chachoengsao, Thailand L. vannamei 2003 DQ Th04Lv a Chumporn, Thailand L. vannamei 2004 AY Th04Pm a Chachoengsao, Thailand P. monodon 2004 DQ Tw2KMe Taiwan Metapenaeus ensis 2000 AY Tw2KPm Taiwan P. monodon 2000 AY Tw99 Taiwan L. vannamei 1999 AF UsHi94 Hawaii, USA L. vannamei 1994 AF a Sequenced in this study

28 28 Table 2.2. Penaeus monodon and Litopenaeus vannamei. Initial appearance of dead shrimp and cumulative mortality after Taura syndrome virus (TSV) infection. Cumulative mortality: numbers of dead shrimp at Day 28 post-inoculation (p.i.); 2 shrimp were removed at Days 2 and 14 p.i. for analyses; 10 shrimp per tank were stocked at Day 0 p.i. TSV infection was determined by hematoxylin and eosin (H&E) histology, in situ hybridization (ISH), and reverse transcription polymerase chain reaction (RT-PCR) analyses. Group Species Day of first death Cumulative mortality TSV infection 1 a P. monodon 3 2 Positive L. vannamei 2 8 Positive 2 b P. monodon 3 2 Positive L. vannamei 2 8 Positive 3 c P. monodon No death 0 Negative L. vannamei No death 0 Negative a Injected with Th04Pm; b injected with Th04Lv; c negative control

29 29 Figure 2.1. Penaeus monodon. Photomicrographs of consecutive hematoxylin and eosin (H&E) stained (left column) and in situ hybridization (ISH) gene probes (right column) tissue sections illustrating Taura syndrome virus (TSV) lesions. (A, B) Cuticular epithelium with acute phase lesions in a specimen collected at Day 2 post-inoculation (p.i.); note severe necrosis of infected epithelial cells by H&E and blue-black precipitates by ISH. (C, D) Lymphoid organ of the same specimen; note normal appearance by H&E but blue-black precipitates at the peripheries of lymphoid tubules by ISH. (E H) Lymphoid organ in specimens collected at Days 14 (E, F) and 28 (G, H) p.i.; note lymphoid organ spheroids (arrows) by H&E and blue-black precipitates by ISH. Scale bars = 50 µm

30 Fig. 2.1 (continued) 30

31 31 64 Id03 12 Cn Tw2KMe Cn Th03-2 Th Tw2KPm 66 Th03-3, Mm03Pm Th03-4 Cn Cn Cn03-4 Th04Lv* Th03-6 Th03-1 Th04Pm* Tw99 84 Mx2K Mx99 UsHi94 98 Mx98 Er04Pm Bz01 Distance Figure 2.2. Phylogenetic neighbor-joining (NJ) tree of capsid protein 2 (CP2, 383 amino acids) from 24 Taura syndrome virus (TSV) isolates. Numbers on branches represent bootstrap values (%) after 1,000 replicates. *: isolates used in our experimental bioassays

32 32 * * * * * * 65 Cn I-- Cn Cn Cn Cn Id Tw Tw2KMe Th Th Th Th Th Th04Lv A Th03-3,Mm03Pm Th04Pm Tw2KPm Er04Pm L UsHi94 MTKVNAYENLPGKGFTHGVGFDYGVPLSLFPNNAIDPTIAVPEGLDEMSIEYLAQRPYML Mx L Mx Mx2K Bz * * * * * * 125 Cn K------V----V Cn K------V----V Cn K------V----V Cn K------V----V Cn K------V----V Id K------V----V Tw V----V Tw2KMe -----K------V----V Th K------V----V Th K V Th K V Th K V Th K------V----V Th04Lv -----K------V----V Th03-3,Mm03Pm -----K V Th04Pm -----K------V----V Tw2KPm -----K------V----V Er04Pm G----L UsHi94 NRYTIRGGDTPDAHGTIIADIPVSPVNFSLYGKVIAKYRTLFAAPVSLAVAMANWWRGNI Mx G----L Mx E Mx2K E Bz K------M Figure 2.3. Comparison of deduced amino acid sequences of capsid protein 2 (CP2) from 24 Taura syndrome virus (TSV) isolates with reference to UsHi94 (GenBank no. AF277675). Blanks indicate amino acids identical to those in UsHi94. A grey background indicates a non-conservative amino acid difference with respect to UsHi94; a normal print indicates a conservative difference. Sequence names in bold indicate the TSV isolates used in our bioassays. Numbers in the alignment correspond to amino acid positions of CP2 (GenBank no. AF277675).

33 33 * * * * * * 185 Cn Cn Cn Cn Cn Id Tw Tw2KMe Th Th Th Th Th Th04Lv Th03-3,Mm03Pm Th04Pm Tw2KPm Er04Pm UsHi94 NLNLRFAKTQYHQCRLLVQYLPYGSGVQPIESILSQIIDISQVDDKGIDIAFPSVYPNKW Mx Mx Mx2K Bz S * * * * * * 245 Cn H Cn H Cn H Cn H Cn H Id H Tw H Tw2KMe H Th H Th H Th H Th H Th H Th04Lv H Th03-3,Mm03Pm H Th04Pm H Tw2KPm H Er04Pm UsHi94 MRVYDPAKVGYTADCAPGRIVISVLNPLISASTVSPNIVMYPWVNWSNLEVAEPGTLAKA Mx V Mx Mx2K Bz I H Fig. 2.3 (continued)

34 34 * * * * * * 305 Cn R---- Cn R---- Cn R---- Cn R-Y-- Cn R---- Id R---- Tw R---- Tw2KMe R---- Th R---P Th R-N-- Th R---- Th R---- Th R---P Th04Lv R---- Th03-3,Mm03Pm R---- Th04Pm H K---P Tw2KPm I R---- Er04Pm VN------SK---- UsHi94 AIGFNYPADVPEEPTFSVTRAPVSGTLFTLLQDTKVSLGEADGVFSLYFTNTTTGGRHRL Mx V SR---- Mx V R--K- Mx2K G R--K- Bz D N N R---- * * * * * * 365 Cn Cn N----- Cn Cn Q----- Cn A Id A Tw99 T Tw2KMe Th QS---- Th I Th I Th I Th QS---- Th04Lv QS---- Th03-3,Mm03Pm I Th04Pm QS---- Tw2KPm Er04Pm -----S UsHi94 AYAGLPGELGSCEIVKLPQGQYSIEYAATSAPTLVLDRPIFSEPIGPKYVVTKVKNGDVV Mx Mx Mx2K Bz S Fig. 2.3 (continued)

35 35 * * 388 Cn03-1 S G----- Cn03-2 S G----- Cn03-3 S I----G----- Cn03-4 S G---T- Cn G---T- Id03 S I----G----- Tw99 S G----- Tw2KMe S G----- Th03-1 S G---- Th03-2 S V--G---V- Th03-4 S V--G---V- Th03-5 S V--G---V- Th03-6 S G---- Th04Lv S G----- Th03-3,Mm03Pm S V--G---V- Th04Pm S G----- Tw2KPm Er04Pm UsHi94 GISEETLVTCGSMAAIGEATVAL Mx Mx Mx2K Bz S Fig. 2.3 (continued)

36 36 CHAPTER 3 COMPARISON OF FOUR TAURA SYNDROME VIRUS (TSV) ISOLATES IN ORAL CHALLENGE STUDIES WITH Litopenaeus vannamei UNSELECTED OR SELECTED FOR RESISTANCE TO TSV Thinnarat Srisuvan 1, 2, *, Brenda L. Noble 2, Paul J. Schofield 2, Donald V. Lightner 2 1 Department of Livestock Development, 69/1 Phayathai Road, Bangkok 10400, Thailand 2 Department of Veterinary Science and Microbiology, University of Arizona, 1117 E. Lowell, Tucson, Arizona 85721, USA * thinnarat@hotmail.com Abstract: Taura syndrome virus (TSV) infection in TSV-resistant (TSR) and TSVsusceptible (Kona) Litopenaeus vannamei (also called Penaeus vannamei) was investigated using histology, in situ hybridization (ISH), conventional reverse transcription polymerase chain reaction (RT-PCR) assays, and SYBR-Green real-time RT-PCR analysis. The shrimp were challenged by feeding with minced tissues of L. vannamei infected with 4 genotypic variants of TSV (Bz01, Th04, UsHi94, and Ve05). Survival probabilities of TSR shrimp were higher than those for Kona shrimp with all 4 variants. Th04, UsHi94, and Ve05 gave no Taura syndrome lesions with TSR shrimp. In contrast, TSR shrimp challenged with Bz01 and Kona shrimp with all 4 TSV variants

37 37 exhibited severe necrosis of cuticular epithelial cells and lymphoid organ spheroids, indicative of acute and chronic phases of TSV infection, respectively. TSV was not detected by RT-PCR in TSR shrimp infected with Th04, UsHi94, and Ve05, or in Kona shrimp infected with Ve05 but was detected in TSR shrimp infected with Bz01 and in Kona shrimp infected with Bz01, Th04, and UsHi94. Real-time RT-PCR revealed that mean TSV copy numbers in TSR shrimp infected with Bz01, Th04, and UsHi94 were significantly (p < ) lower than those in Kona shrimp. In contrast, mean TSV copy numbers in TSR and Kona shrimp infected with Ve05 were not significantly different (p > 0.4). The results show that TSR L. vannamei are susceptible to infection but give high survival rates following challenge by all 4 variants of TSV. Key words: Taura syndrome virus TSV TSV variants Litopenaeus vannamei Introduction Taura syndrome (TS) is an important disease of cultured penaeid shrimp. It has been reported in many countries and has caused considerable economic loss for the last decade (Hasson et al. 1995, 1999b, Lightner et al. 1995, Tu et al. 1999, Yu & Song 2002). It is caused by Taura syndrome virus (TSV), a member of the family Dicistroviridae (Mayo 2005). TSV is a non-enveloped icosahedral virus with a diameter of 32 nm (Bonami et al. 1997) and a single-stranded, positive-sense RNA genome of 10,205

38 38 nucleotides comprising 2 open reading frames (ORF1 and ORF2) (Mari et al. 2002). ORF1 may code for non-structural proteins, including helicase, protease, and RNAdependent RNA polymerase, whereas ORF2 codes for structural proteins such as the 3 major capsid proteins: CP1, CP2, and CP3. TSV isolated at different times and/or from different locations has been reported to display phenotypic variations. For instance, Erickson et al. (2005) demonstrated that Pacific white shrimp Litopenaeus vannamei (also called Penaeus vannamei) (Farfante & Kensley 1997) experimentally infected with a Belize isolate (Bz02) showed higher mortality than they did when infected with a Hawaiian isolate (UsHi94) (referred to as BLZ02TSV and Hi94TSV, respectively, in the cited paper). Phylogenetic analysis revealed that Bz02 was distinct among 29 isolates of TSV, despite the fact that TSV displays low genetic variation from 0 to 5.6% in nucleotide sequence and from 0 to 7% in deduced amino acid sequence (Tang & Lightner 2005). There were 3 distinct phylogenetic lineages in the Americas, Asia, and Belize. Nielsen et al. (2005) also reported that TSV isolates from Asia and the Americas were distinct. Because TS causes high mortalities in L. vannamei (Lightner 1996), scientists have attempted to develop TSV-resistant shrimp populations (White et al. 2002, Wyban 2000, Xu et al. 2003). This is important because L. vannamei is the predominant cultivated species worldwide (FAO 2004). We used histology, in situ hybridization (ISH), conventional reverse transcription polymerase chain reaction (RT-PCR), and SYBR-

39 39 Green real-time RT-PCR to study the course of TSV infection in a population of L. vannamei selected for TSV resistance and in an unselected population. Materials and Methods Experimental challenges. We performed experimental challenges with 2 populations of Litopenaeus vannamei to determine possible differences in susceptibility to TSV. These 2 populations were a specific-pathogen-free (SPF) TSV-resistant (TSR) stock obtained from High Health Aquaculture, Kona, Hawaii, and an SPF TSVsusceptible (Kona) stock (Moss et al. 2005) obtained from the Oceanic Institute, Oahu, Hawaii. These shrimp were derived from stocks certified SPF for a list of shrimp pathogens including TSV for the preceding 2 yr. There were 4 viral challenge groups (Groups 1 to 4) and 1 negative control group (Group 5) for each shrimp stock. Groups 1 to 4 of each stock were challenged with TSV isolates, Bz01, Th04, UsHi94, and Ve05, respectively (Table 3.1). Each challenge group was held in a separate aquarium. Each TSR group contained ~150 shrimp (avg. wt = 2 g), and each Kona group contained 20 shrimp (avg. wt = 2 g) (see Fig. 3.1). The challenged groups were fed for 3 d (Days 0, 1, and 2 post-infection [p.i.]) at 10% body weight d 1 with minced shrimp tissues prepared from frozen L. vannamei infected with an appropriate TSV isolate. Starting at Day 3 p.i., all shrimp groups were fed once a day with a commercial pelleted feed (Rangen 35%, Buhl) for 12 d. Group 5 of each stock (negative control) that consisted of 20 shrimp per

40 40 aquarium (avg. wt = 2 g) fed only the commercial feed throughout the test period of 15 d. All of the aquaria were checked daily for moribund and dead shrimp. Shrimp were sampled for histological analysis. For TSR Group 1, 1 and 2 shrimp were sampled at Days 7 and 14 p.i., respectively. For TSR Group 2, 1 and 2 shrimp were sampled at Days 8 and 14 p.i., respectively. For TSR Groups 3 and 4, 3 shrimp were sampled from each group at Day 14 p.i. For Kona Group 1, 1 shrimp was sampled at Day 3 p.i. For Kona Group 2, 1 shrimp was sampled on Days 3 and 14 p.i. For Kona Group 3, 1 and 2 shrimp were sampled on Days 8 and 14 p.i., respectively. For Kona Group 4, 1 shrimp was sampled at Days 8 and 14 p.i. These shrimp were not tested for TSV by RT- PCR or real-time RT-PCR, but the cephalothoraxes were fixed overnight in Davidson s fixative and transferred to 70% alcohol for histological analysis by hematoxylin and eosin (H&E) staining and ISH using standard methods (Lightner 1996, Mari et al. 1998). In addition to histological analysis, shrimp were also collected for RT-PCR and real-time RT-PCR analyses. These shrimp were not used for histological analysis. These specimens were either dead shrimp collected during the challenge study, Days 0 to 13 p.i., or surviving shrimp at Day 14 p.i. Numbers of shrimp collected at specific days are provided later in Table 3.3 and Fig Total RNA was extracted from the pleopods using a High Pure RNA tissue extraction kit (Roche Biochemical) and stored at 70 C.

41 41 ISH. A mixture of probes, TS624 and TS622, was used for ISH. Probes TS624 and TS622 hybridize with the UsHi94 genome at nucleotides 3218 to 3841 and 5899 to 6520, respectively. The probes were prepared from 2 cdna clones TSV and TSV , respectively, by polymerase chain reaction (PCR) labeling with digoxigenin (DIG)-11-dUTP as described by Mari et al. (1998). Primers 3218F (5 -CAC TAC GTT AGC AGG CAA TG-3 ) and 3841R (5 -CAC TTC ACT GCA CTC GAC AC-3 ) were used to label probe TS624 (624 bp), while primers 5899F (5 -TTA AGC GCG TTG GTG ACA AG-3 ) and 6520R (5 -GCA TCC TGC GCA TCG ATA TT-3 ) were used to label probe TS622 (622 bp). Following PCR, the DIG-labeled probes TS624 and TS622 were precipitated with ethanol, re-suspended in distilled water, and stored at 20 C. SYBR-Green real-time RT-PCR. Primer Express software (Applied Biosystems) was used to design forward and reverse primers, 401F (5 -GAC TGG CTC ATA TAC TAT GGC CTC TTA T-3 ) and 545R (5 -CCG TCG CAA AGT TCC AAT TAA-3 ), respectively, from ORF1 of the Th04 genome to amplify a product of 145 bp from nucleotide positions 401 to 545. Real-time assays were performed using an ABI GeneAmp 5700 sequence detection system with SYBR-Green RT-PCR reagents (Applied Biosystems). The reaction mixture contained 1 µl of an RNA sample, 12.5 µl of 2x SYBR-Green Master Mix, 200 nm each of forward and reverse primers, and unit of Multiscribe reverse transcriptase in a final volume of 25 µl. The RT-PCR profile

42 42 was 30 min at 48 C and 10 min at 95 C, followed by 40 cycles of 95 C for 15 s and 60 C for 1 min. Data analysis was performed with GeneAmp 5700 sequence detection software. To prepare an RNA standard for the real-time assay, a TSV fragment (nucleotides 20 to 1600) was amplified from original shrimp specimens of Th04. The amplification product was then cleaned with a QIAquick PCR purification kit (Qiagen) and ligated to pgem-t-easy vector (Promega). Recombinant plasmids were cloned into competent Escherichia coli JM109 cells (Promega), and a recombinant plasmid, ptsv-1, from 1 clone was verified by sequencing with an ABI Prism automated DNA sequencer (Applied Biosystems) at University of Arizona. Then, ptsv-1 isolated with a PerfectoPrep plasmid isolation kit (Eppendorf Scientific) was linearized by SalI digestion and used as the template for an in vitro transcription with T7 RNA polymerase (Fermentas). A volume of 1 µg of plasmid was used in a 50 µl reaction at 37 C for 2 h, followed by DNase I digestion at 37 C for 30 min. The synthesis of RNA (~800 nucleotides) was confirmed by electrophoresis in a 1.2% agarose gel containing ethidium bromide. The RNA standard thus prepared was cleaned using a QIAquick PCR purification kit, quantified by a spectrophotometer, and stored at 70 C. Conventional RT-PCR. Conventional RT-PCR assays were performed with a GeneAmp EZ rtth RNA PCR kit (Applied Biosystems) using 5 µl of extracted RNA as the template and primers 9195 (5 -TCA ATG AGA GCT TGG TCC-3 ) and 9992 (5 - AAG TAG ACA GCC GCG CTT-3 ) to produce an amplicon of 231 bp (Nunan et al.

43 ). The RT-PCR protocol comprised 30 min at 60 C and 2 min at 94 C followed by 40 cycles of 45 s at 94 C, 45 s at 60 C and a final extension step for 7 min at 60 C. An aliquot of amplified products was analyzed in 1.8% agarose gel containing ethidium bromide. Statistical analyses. Statistical analyses were performed on resultant data according to Milton (1999). First, the Kaplan-Meier survival curves and cumulative survival probabilities (Bland & Altman 1998) were computed by SPSS 14.0 software for Windows. Second, an analysis-of-variance (ANOVA) technique was used to determine the reproducibility of the SYBR-Green real-time RT-PCR and to test whether there was no linear regression between the coefficient of variations (CVs) and threshold cycle (C T ). Third, a 1-way classification ANOVA, completely random design with fixed effect, was used to compare mean TSV copy numbers among 8 different shrimp subgroups and to test whether the population means were equal. Forth, once a 1-way ANOVA had been run to compare population means and the hypothesis of equality had been rejected, the investigation was continued to pinpoint the differences within 4 pairs of mean TSV copy numbers using the Bonferroni t-test. Finally, assuming that variances were unequal, the Smith-Satterthwaite t-test was performed to make inferences on difference between 2 mean TSV copy numbers (1 found to be negative and 1 found to be positive by conventional RT-PCR). The statistical analyses from steps 2 to 5 were performed using Microsoft Excel 2002 software.

44 44 Results and Discussion Survival and lesions in TSV-resistant Litopenaeus vannamei after TSV infection. The selected TSR Litopenaeus vannamei showed greater survival than did the unselected Kona stock after challenge with 4 different TSV variants. Because of the experimental design (i.e. intermittent sampling and differences in test group sizes), a simple estimation of percent survivals over time could not be precisely calculated and compared. However, a mean estimated survival, also known as a cumulative survival probability, for each group could be obtained using the Kaplan-Meier analysis (Bland & Altman 1998). Survival of the selected TSR and unselected Kona populations was determined at termination of the bioassays, Day 14 p.i.; for TSR shrimp, survival probabilities were from to 1, while for Kona shrimp survival probabilities were from 0 to (Fig. 3.1). Both TSR and Kona shrimp showed the lowest survival probabilities of and 0, respectively, when infected with Bz01 (Group 1) (Fig. 3.1A). In contrast, both TSR and Kona shrimp showed the highest survival probabilities of 1 and 0.215, respectively, when infected with UsHi94 (Group 3) (Fig. 3.1C). Our results were also consistent with Bz01 being more virulent than UsHi94 (Erickson et al. 2005, Tang & Lightner 2005). The data in Fig. 3.1 also allowed comparison of survival between TSR and Kona L. vannamei that were infected with the same TSV isolate. Specifically, survival probabilities of TSR shrimp were higher than those of Kona shrimp challenged with all 4 variants. No mortality was observed among L. vannamei in the TSR and Kona L. vannamei negative control group (Group 5).

45 45 We did not detect characteristic acute or chronic lesions of TSV infection as described by Lightner et al. (1995) and Hasson et al. (1995, 1999b), in the TSR L. vannamei that were infected with Th04, UsHi94, or Ve05 (Fig. 3.2A,B). The TSR L. vannamei, however, developed TSV lesions when infected with Bz01. One specimen (sampled at Day 7 p.i.) displayed the characteristic acute phase lesions of TSV infection, indicated by severe necrosis in various epithelial tissues, including gills, cuticular, and stomach epithelia (Fig. 3.2C). When the consecutive tissue section of this shrimp was analyzed by ISH, a positive reaction to TSV-specific gene probes was detected as blueblack precipitates (Fig. 3.2D). This TSR specimen exhibited a normal lymphoid organ by histology (Fig. 3.2E). However, by ISH, its lymphoid organ section showed a strong positive reaction, and the reaction was only seen at the peripheral cells in the lymphoid organ tubules (Fig. 3.2F). A similar distribution of TSV positive cells in the lymphoid organ was seen by ISH in L. vannamei (Hasson et al. 1999b) and Penaeus monodon (Srisuvan et al. 2005) in the transition phase of TSV infection and by immunohistochemistry in P. monodon infected with Yellow head virus (YHV) (Soowannayan et al. 2002). Two specimens infected with Bz01 and sampled at Day 14 p.i. exhibited lymphoid organ spheroids (LOS) characteristic of chronic phase lesions of TSV infection (Fig. 3.2G), and their consecutive sections reacted to the TSV-specific gene probes by ISH (Fig. 3.2H), confirming that the LOS formation was associated with TSV infection. Since only Bz01 produced the characteristic TSV lesions in the TSR L. vannamei, this population was clearly shown to be partially resistant to TSV infection using the 4 TSV variants tested in the present study.

46 46 As expected, the Kona L. vannamei, injected with all 4 TSV isolates, displayed acute and chronic phase lesions of TSV infection at Days 3, 8, 14 p.i., as illustrated in the photomicrographs for Bz01 infections in the TSR stock (Fig. 3.2C to H). In contrast, no lesions indicative of TSV infection were detected in any shrimp from the negative control group (Group 5) of either TSR or Kona stock at termination (Fig. 3.2A,B). Development and validation of a SYBR-Green real-time RT-PCR. The SYBR-Green real-time RT-PCR analysis was capable of detecting Bz01, Th05, UsHi94, and Ve05, despite the fact that there is 1 nucleotide mismatch at the primer 401F binding region for UsHi94. The dissociation curve showed a single peak for the 145 bp amplification product at a melting temperature of 79 C. The assay gave negative results for RNA extracted from YHV-infected shrimp, Infectious myonecrosis virus (IMNV)- infected shrimp, and SPF shrimp (C T = 40). Testing of 10-fold serial dilutions of the TSV RNA standard from 10 9 copies down to 1 copy revealed that 1 and 10 copies could be detected in 1 and 2 out of 5 assays, respectively, while 100 copies were detected in all 5 assays. Thus, the lower detection limit was considered to be 100 copies. A strong linear correlation (r < 0.99) was obtained between C T and RNA quantities over an 8-log range from 10 2 to 10 9 copies per reaction, indicating that this real-time assay has a large dynamic range (Fig. 3.3).

47 47 To determine the reproducibility of the SYBR-Green real-time RT-PCR assay, we compared 5 standard curves from 100 to 10 9 copies of the TSV RNA standard. The CVs within each run were between 0.11 and 0.87% for 10 9 copies and increased to 0.5 to 4.12% for 10 2 copies (Table 3.2). For these 5 independent runs, there was a significant linear relation between CVs and C T (p < 0.01, F 1, 38 = 8.62 [ANOVA, where df = 1 and 38], r 2 = , n = 40). This was in accordance with previous studies (Tang et al. 2004) showing that lower copies of RNA templates translated to higher C T and higher CVs. The CVs were 1.66% for 10 9 copies and increased to 6.28% for 10 2 copies, indicating that this real-time assay has little variation between runs. Mean viral loads after TSV challenge. By real-time RT-PCR, the TSR and Kona groups showed significant differences in mean TSV copy numbers after challenge with Bz01, Th04, or UsHi94 (p < , t 76 = 7.66, 4.82, and 10.71, respectively [the Bonferroni t-test, where df = 76]) (Fig. 3.4). In contrast, TSV copy numbers within the TSR and Kona L. vannamei infected with Ve05 showed no difference in mean effect (p > 0.4, t 76 = ). We reasoned that high mortalities in the Kona L. vannamei, infected with Bz01, may be associated with its high virus copy numbers. The real-time RT-PCR analysis showed that the TSR L. vannamei are susceptible to infection although they give high survival rates following challenge by all 4 variants of TSV. Fig. 3.4 also revealed that, for both TSR and Kona shrimp, the quantities of tissue-loaded viruses were relatively high at Days 3 to 6 p.i. and decreased at termination, Day 14 p.i. In addition, the results from the real-time RT-PCR analysis and those by histological analysis (Fig.

48 48 3.2) further support that the acute, transition, and chronic phases of TSV infection occurred following challenge as previously described by Lightner et al. (1995) and Hasson et al. (1995, 1999b). The quantities of tissue-loaded TSV may be responsible for the pathogenesis and the survivability of the affected shrimp. In contrast to real-time RT-PCR, conventional RT-PCR did not detect TSV in all TSR L. vannamei challenged with Th04, UsHi94, and Ve05, in 6 out of 10 TSR shrimp challenged with Bz01, or in all Kona shrimp challenged with Ve05 (Table 3.3). However, it did detect TSV in 4 out of 10 TSR L. vannamei infected with Bz01 and in all Kona shrimp infected with Bz01, Th04, and UsHi94. Our results in Table 3.3 were not consistent with those of D. V. Lightner et al. (unpubl. data) who found that Ve05 could be detected by the conventional RT-PCR. To investigate this apparent inconsistency, we hypothesized that only samples containing relatively high copy numbers of TSV were positive by the conventional RT-PCR. Using the SYBR-Green real-time RT-PCR, it was determined that TSV was detected in the same shrimp specimens that had given negative TSV results by the conventional RT-PCR (Table 3.3). We found that mean TSV copy numbers among the TSR shrimp infected with Th04, UsHi94, Ve05 (Shrimp 1 to 34), and 6 TSR shrimp infected with Bz01 (Shrimp 39 to 44), which were tested as negative by the conventional RT-PCR, was 1.50 x 10 3 (standard deviation [SD] = 1.57 x 10 3 ), whereas mean TSV copy in 4 TSR shrimp infected with Bz01 (Shrimp 35 to 38), which were tested as positive by the conventional RT-PCR, was 4.79 x 10 4 (SD = 3.01 x 10 4 ). Interestingly, we found that there was a significant difference in TSV copy numbers

49 49 between the TSR shrimp that TSV was not detected and those that were positive by conventional RT-PCR (p < , t = 4.87 [the Smith-Satterthwaite t-test where df = infinity]). It was also determined that mean TSV copy number in the Kona shrimp infected with Ve05 (Shrimp 45 to 54), which were detected as negative by the conventional RT-PCR, was 1.29 x 10 3 (SD = 3.43 x 10 3 ), whereas mean TSV copy in the Kona shrimp infected with Bz01, Th04, and UsHi94 (Shrimp 55 to 84), which were detected as positive by the conventional RT-PCR, was 2.65 x 10 6 (SD = 2.39 x 10 6 ). Again, we found that there was a significant difference in TSV copy numbers between the Kona shrimp that TSV was not detected and those that were positive by conventional RT-PCR (p < , t = 6.08). Thus, it is statistically necessary to conclude that the SYBR-Green real-time RT-PCR has a greater sensitivity compared to the conventional RT-PCR and that the conventional RT-PCR was only capable of detecting TSV at a relatively high virus copy number. In conclusion, the selected line of TSR L. vannamei, while susceptible to TSV infection, is resistant to development of severe TS. Using experimental challenge studies, we demonstrated the resistance to severe TSV infection (as indicated by lower TSV copy numbers in infected individuals) in the TSR L. vannamei by histological, ISH, RT-PCR, and real-time RT-PCR analyses. Resistance to TSV, as observed in the selected TSR line, may not be prevalent among wild and cultured populations of L. vannamei. Finally, our results indicate that additional studies are needed to elucidate the genetic trait and markers associated to TSV resistance in L. vannamei.

50 50 Acknowledgements. This work was supported by Gulf Coast Research Laboratory Consortium Marine Shrimp Farming Program, USDA, CSREES, Grant no T. S. was supported by a scholarship from the Royal Government of Thailand. We thank Dr. James Wyban (High Health Aquaculture) for supplying TSV-resistant (TSR) Litopenaeus vannamei for the experimental challenges and Dr. Kathy F. J. Tang-Nelson for technical assistance.

51 51 Table 3.1. Taura syndrome virus (TSV) isolates used for the oral challenge studies TSV isolate Collection location Source species Collection year GenBank no. Bz01 Belize Litopenaeus vannamei 2001 AY Th04 Chumporn, Thailand L. vannamei 2004 AY UsHi94 Hawaii, USA L. vannamei 1994 AF Ve05 Venezuela L. vannamei 2005 DQ212790

52 52 Figure 3.1. Litopenaeus vannamei. Survival curves of Taura syndrome virus (TSV)- resistant (TSR, solid lines) and TSV-susceptible (Kona, dashed lines) shrimp after challenge by feeding with 4 TSV isolates: (A) Bz01, (B) Th04, (C) UsHi94, and (D) Ve05. Numbers in bold: Kaplan-Meier survival probability values. Numbers in parentheses: numbers of shrimp sampled for histological analysis at specific days (+). n1: group total numbers of shrimp stocked at Day 0 post-infection (p.i.). n2: numbers of survivors at Day 14 p.i. A 1.0 B 1.0 n1 = 150, n2 = 123 Cumulative survival (1) n1 = 142, n2 = (1) (2) n1 = 20, n2 = 0 Cumulative survival (1) (1) (2) n1 = 20, n2 = (1) Days Days C n1 = n2 = 150 (3) D n1 = 148, n2 = 141 (3) Cumulative survival (1) n1 = 20, n2 = (2) Cumulative survival (1) n1 = 20, n2 = (1) Days Days 12 14

53 53 Figure 3.2. Litopenaeus vannamei. Photomicrographs of consecutive shrimp tissue sections tested for Taura syndrome virus (TSV) lesions by hematoxylin and eosin (H&E) staining (left column) and in situ hybridization (ISH) (right column). (A, B) Example of a normal appearing lymphoid organ (LO) from TSV-resistant (TSR) shrimp challenged by feeding with Th04, UsHi94, Ve05, and from both TSR and TSV-susceptible (Kona) shrimp in the negative control group collected at Day 14 post-infection (p.i.). (C H): Examples of the characteristic acute (C, D) and chronic phase (E H) lesions seen in TSR shrimp challenged with Bz01 and in Kona shrimp challenged with all 4 TSV isolates used in this study. (C, D) Cuticular epithelium of the stomach illustrating an acute phase lesion in a TSR shrimp challenged with Bz01 and collected at Day 7 p.i. Note the severe necrosis of infected cuticular epithelial cells as shown by H&E and blue-black precipitates by ISH with TSV-specific probes. (E, F) LO from the same TSR specimen as shown in (C, D) illustrating the early to transition phases of TSV infection. Note the normal tissue appearance by H&E but blue-black precipitates at the outermost parenchymal cells of the LO tubules by ISH. (G, H) LO from a TSR shrimp challenged with Bz01 and collected at Day 14 p.i. Note the presence of lymphoid organ spheroids (arrows in G) exemplifying the chronic phase lesion by H&E staining and a positive ISH reaction. Scale bars = 50 µm

54 Fig. 3.2 (continued) 54

55 55 Figure 3.3. SYBR-Green real-time reverse transcription polymerase chain reaction (RT- PCR). (A) Amplification plot and (B) standard curve from 10-fold serial dilutions of a Taura syndrome virus (TSV) RNA standard in duplicate measurements (squares). Rn: normalized fluorescent intensity. C T : threshold cycle. Log C: logarithm values of TSV copy numbers 10 8 A Rn Cycle numbers B Slope = Intercept = Correlation = CT Log C 7 8 9

56 56 Table 3.2. Reproducibility of the SYBR-Green reverse transcription polymerase chain reaction (RT-PCR) assay. TSV: Taura syndrome virus; CV: coefficient of variation; C T : threshold cycle TSV CV (%) within assay in 5 runs (mean C T from duplicate measurements) copies µl CV (%) between assays (mean C T ) (8.79) 0.71 (8.51) 0.11 (8.44) 0.12 (8.50) 0.87 (8.66) 1.66 (8.58) (12.13) 0.25 (11.91) 1.01 (11.84) 0.81 (11.86) 0.08 (12.18) 1.34 (12.11) (15.25) 0.82 (15.94) 0.46 (15.26) 0.93 (15.56) 0.76 (15.84) 2.05 (15.57) (20.09) 0.55 (19.82) 1.11 (19.74) 0.70 (20.13) 0.35 (20.20) 1.02 (20.00) (22.81) 0.61 (22.92) 1.25 (23.13) 1.36 (23.54) 1.21 (23.99) 2.09 (23.27) (26.26) 1.72 (26.69) 0.84 (26.23) 0.04 (26.16) 0.18 (26.75) 1.06 (26.42) (27.43) 0.67 (29.81) 1.98 (27.42) 0.44 (27.20) 0.18 (27.80) 3.84 (27.93) (29.02) 2.02 (33.43) 4.12 (30.63) 0.70 (28.59) 0.50 (30.03) 6.28 (30.34)

57 57 Figure 3.4. Litopenaeus vannamei. Means ± standard errors of Taura syndrome virus (TSV) copy numbers µl 1 RNA within pleopods of TSV-resistant (TSR) and TSVsusceptible (Kona) shrimp after challenge by feeding with 4 virus isolates: (A) Bz01, (B) Th04, (C) UsHi94, and (D) Ve05. Numbers near squares: numbers of shrimp collected on specific days. N/A: not applicable A TSV copies µl 1 RNA B TSV copies µl 1 RNA Day 3 Day 5 Day 9 Day 14 TSR 6.03E Kona 1.21E E+06 N/A N/A 1 Day 4 Day 5 Day 7 Day 14 TSR Kona 5.79E E E+05 N/A C TSV copies µl 1 RNA D TSV copies µl 1 RNA Day 5 Day 8 Day 10 Day 14 TSR N/A N/A N/A Kona 1.36E E E E+05 1 Day 6 Day 8 Day 10 Day 14 TSR N/A N/A N/A Kona

58 58 Table 3.3. Litopenaeus vannamei. Comparison of a conventional reverse transcription polymerase chain reaction (RT-PCR) with the SYBR-Green real-time RT-PCR for Taura syndrome virus (TSV) within pleopods of TSV-resistant (TSR) and TSV-susceptible (Kona) shrimp after challenge by feeding with 4 virus isolates. (+) presence or ( ) absence of the specific 231 bp amplicon Population TSV isolate TSR Ve05 UsHi94 Th04 Bz01 Shrimp no. a Means of duplicate copy numbers Collection day post-infection Conventional RT-PCR Real-time RT-PCR a (TSV copy numbers µl 1 RNA)

59 59 Table 3.3 (continued) Population TSV isolate Kona Ve05 UsHi94 Th04 Bz01 Shrimp no. a Means of duplicate copy numbers Collection day post-infection Conventional RT-PCR Real-time RT-PCR a (TSV copy numbers µl 1 RNA) x x x x x

60 60 CHAPTER 4 ULTRASTRUCTURE OF THE REPLICATION SITE IN TAURA SYNDROME VIRUS (TSV)-INFECTED CELLS Thinnarat Srisuvan 1, 2, *, Carlos R. Pantoja 2, Rita M. Redman 2, Donald V. Lightner 2 1 Department of Livestock Development, 69/1 Phayathai Road, Bangkok 10400, Thailand 2 Department of Veterinary Science and Microbiology, University of Arizona, 1117 E. Lowell, Tucson, Arizona 85721, USA * thinnarat@hotmail.com Abstract: Taura syndrome virus (TSV) is a member of the family Dicistroviridae that infects Pacific white shrimp Litopenaeus vannamei (also called Penaeus vannamei), and its replication strategy is largely unknown. In order to identify the viral replication site within shrimp infected cells, the viral RNA was located in correlation to virus-induced membrane rearrangement. Ultrastructural changes in the infected cells, analyzed by transmission electron microscopy (TEM), included the induction and proliferation of intracellular vesicle-like membranes, while the intracytoplasmic inclusion bodies and pyknotic nuclei indicative of TSV infection were frequently seen. TSV plus-strand RNA, localized by electron microscopic in situ hybridization (EM-ISH) using TSV-specific cdna probes, was found to be associated with the membranous structures. Moreover,

61 61 TSV particles were observed in infected cells by TEM, and following EM-ISH, they were also seen in close association to the proliferating membranes. Taken together, our results suggest that the membranous vesicle-like structures carry the TSV RNA replication complex and that they are the site of nascent viral RNA synthesis. Further investigations on cellular origins and biochemical compositions of these membranous structures will elucidate the morphogenesis and propagation strategy of TSV. Key words: Taura syndrome virus TSV in situ hybridization Transmission electron microscopy Litopenaeus vannamei Introduction Taura syndrome (TS) is an economically important disease listed by World Organization for Animal Health (OIE 2006). Since its first recognition in Ecuador in 1992, TS has rapidly spread to cultured penaeid shrimp-farming regions in many countries of the Americas, Asia, and Africa and continued to devastate the shrimp industry for the last decade (Lightner et al. 1995, Nielsen et al. 2005, Srisuvan et al. 2005, Tang & Lightner 2005). The causative agent of TS is Taura syndrome virus (TSV), a member of the family Dicistroviridae, which is a non-enveloped icosahedral virus with a diameter of 32 nm and a single-stranded, positive-sense RNA genome of 10,205 nucleotides (Bonami et al. 1997, Mari et al. 2002, Mayo 2005). The principal host of

62 62 TSV is Pacific white shrimp Litopenaeus vannamei (also called Penaeus vannamei) (Fanfante & Kensley 1997, Lightner et al. 1995). The intracellular biogenesis of TSV remains largely uninvestigated although the virus has become an intense subject of research for almost 15 yr. Ultrastructural pathogenesis of many viruses, e.g. Human parechovirus type 1, Dengue virus, Hepatitis C virus, Foot-and-mouth disease virus (FMDV), and Severe acute respiratory syndromeassociated coronavirus (SARS-CoV), has been characterized using transmission electron microscopy (TEM), in situ hybridization (ISH), and immunoelectron microscopy (IEM) (Krogerus et al. 2003, Gosert et al. 2003, Goldsmith et al. 2004, Grief et al. 1997, Monaghan et al. 2004). Additionally, our laboratory has previously developed an ISH protocol using a specific cdna probe to follow the intracellular translocation of Hepatopancreatic parvovirus (HPV) in penaeid shrimp (Pantoja & Lightner 2001). Infection by all single-stranded, positive-sense RNA viruses is believed to involve the intracellular rearrangement of membranes in the cytosol (for a review, see Mackenzie 2005, Novoa et al. 2005). The host cell membranes function as the replication site for the synthesis of the nascent viral genomes. For instance, in many viruses such as Mouse hepatitis virus, Rubella virus, and Semliki Forest virus, these consist of generation and proliferation of endoplasmic reticulum (ER) and membrane vesicles that accumulate in the perinuclear region of infected cells (Gosert et al. 2002, Kujala et al. 2001, Magliano et al. 1998). Characterization of the viral replication complexes is an important aspect in

63 63 virology and cell biology. In the present paper, the ultrastructure of the replication site in cells infected with TSV is visualized by TEM and ISH. Materials and Methods Shrimp specimens. Litopenaeus vannamei were collected from affected farms in Ecuador, Peru, and Columbia, fixed in 6% glutaraldehyde in phosphate buffer, and processed for TEM as previously described by Bonami et al. (1992). Histological examinations revealed that they exhibited characteristic lesions of TSV infection as previously reported by Lightner et al. (1995) (not shown). These shrimp were not tested for TSV by ISH. Specimens used for ISH were embedded with paraffin and the hydrophilic Unicryl resin (British Bio-Cell International). For paraffin embedding, the specimen was an L. vannamei (wt = 1 g) derived from a specific-pathogen-free (SPF) Kona stock (Moss et al. 2005), obtained from the Oceanic Institute, Oahu, Hawaii, USA. This shrimp was inoculated with a tissue homogenate prepared from frozen TSV-infected L. vannamei, colleted from Thailand in 2004 (Th04, GenBank no. AY997025) (for a detailed inoculation method, see Srisuvan et al. 2005). The cephalothorax was fixed with Davidson s fixative and embedded in paraffin for histological analysis using standard methods (Ligthner 1996). The paraffin-embedded tissue sections were used for light microscopic (LM)-ISH.

64 64 In addition, specimens for Unicryl resin embedding were generated in experimental challenge studies using 20 SPF Kona L. vannamei (avg. wt = 1 g). Each shrimp was administered a single injection (~100 µl), into their third tail segment, of a tissue homogenate prepared from frozen TSV-infected L. vannamei, collected from Belize in 2001 (Bz01, GenBank no. AY590471). The tissue homogenate was prepared from shrimp cephalothoraxes as described by Hasson et al. (1995), and diluted 1:150 with 2% saline prior to inoculation. All shrimp were fed once a day with a commercial pelleted feed (Rangen 35%, Buhl), starting at Day 0 post-inoculation (p.i.). The aquarium was observed twice a day for moribund and dead shrimp. The gills of 11 moribund shrimp, sampled at Days 3 to 5 p.i., were processed for both LM-ISH and electron microscopic (EM)-ISH as subsequently described. Fixation, embedding, and sectioning for ISH. The fixative was 6% glutaraldehyde prepared with 0.15 M Millonig s phosphate buffer (ph 7.0) supplemented with 1% sodium chloride and 0.5% sucrose. The gills of each TSV-infected shrimp were collected and cut into small pieces (~1 mm 3 ) in ice-cold phosphate buffer. Tissue specimens from each shrimp were transferred to the ice-cold fixative, ~10 times of the volume of the shrimp tissues (~1 ml), and fixed for 6 h under refrigeration (4 C). After fixation, specimens were dehydrated at room temperature (RT, ~25 C) in a graded series of ethanol (15 min each in 30, 50, 70, 80, and 95%, and twice in absolute ethanol). The dehydrated specimens were infiltrated at 4 C with increasing

65 65 concentrations of Unicryl resin as follows: 24 h in resin: absolute ethanol (1:2), 24 h in resin: absolute alcohol (2:1), and 24 h in pure resin. Resin-infiltrated specimens were transferred into Beem capsules containing fresh resin and polymerized at 10 C for 5 d by exposure to ultra-violet (UV) light provided by 2 x 15 W Phillips UV lamps, 360 nm wavelength, set at approximately 15 cm under the Beem capsules. Semi-thin sections (1 µm thickness) were placed on a drop of HPLC (high performance liquid chromatography) water on a microscope glass slide, heat-dried at 60 C for ~2 min, stained with 0.5% toluidine blue in 1% sodium borate at 60 C for 1 min, and then observed with a light microscope for the presence of characteristic lesions of TSV infection as described by Lightner et al. (1995). Consecutive semi-thin sections were placed on drops of HPLC water on a Superfrost/Plus positively charged microscope slides (Fisher Scientific), heat-dried, and stored at RT until the time of analysis. Five slides of semi-thin sections were prepared from each block. Five to 7 consecutive ultrathin sections (gold interference color) from the same blocks were also placed on carbon/formvar-coated 100-mesh nickel grids and stored, unstained, at RT. Preparation of TSV-specific cdna probes. A mixture of probes, TS624 and TS622, was used for ISH (Srisuvan et al. 2006). Probes TS624 and TS622 hybridize with the TSV genome (GenBank no. AF277675) at nucleotides 3218 to 3841 and 5899 to 6520, respectively. They were prepared from 2 cdna clones TSV and TSV , respectively, by polymerase chain reaction (PCR) labeling with

66 66 digoxigenin (DIG)-11-dUTP as described by Mari et al. (1998). Primers 3218F (5 -CAC TAC GTT AGC AGG CAA TG-3 ) and 3841R (5 -CAC TTC ACT GCA CTC GAC AC-3 ) were used to label probe TS624 (624 bp), while primers 5899F (5 -TTA AGC GCG TTG GTG ACA AG-3 ) and 6520R (5 -GCA TCC TGC GCA TCG ATA TT-3 ) were used to label probe TS622 (622 bp). The reaction mixture contained 5 µl of an appropriate cdna clone, 10 µl of 10x PCR buffer (Applied Biosystems), 100 µm of dttp, 100 µm of a mixture of datp, dctp, and dgtp, 10 µl of 10x DIG-DNA labeling mix (Roche), 2 mm of MgCl 2, 1 mm each of forward and reverse primers, and 0.05 unit of AmpliTaq Gold DNA Polymerase in a final volume of 100 µl. The PCR profile was 5 min at 94 C, followed by 40 cycles of 94 C for 1 min, 55 C for 1 min, 72 C for 2 min, and a final extension step at 72 C for 7 min. An aliquot of each DIG-labeled probe, TS624 and TS622, was analyzed in a 1% agarose gel containing ethidium bromide. Following PCR, each probe (~99 µl) was precipitated with 360 µl of absolute ethanol, containing 1 µl of 20 mg ml 1 glycogen, 10 µl of 200 mm EDTA (ph 8.0), and 11 µl of 4 M LiCl. The probe suspension was mixed well, placed at 20 C overnight (~16 h), and centrifuged at 4 C and 13,000 g for 30 min. The supernatant was carefully decanted, and the pellet was washed with 0.5 ml of cold 70% ethanol, followed by a centrifugation for 10 min at 4 C and 13,000 g. The supernatant was decanted, and the pellet was air-dried for 20 min. Finally, each probe was re-suspended in 100 µl of HPLC water, placed at 37 C for 10 min, and stored at 20 C.

67 67 LM-ISH. Paraffin and Unicryl resin-embedded tissue sections of L. vannamei were used for LM-ISH. For paraffin-embedded specimens, the hybridization procedure was identical to that of resin-embedded sections as subsequently described, except for the counterstaining. Specifically, after the silver enhancement step, slides were counterstained for 5 min with 0.5% Bismarck Brown (Science Lab) and dehydrated as follows: 3 x 10 dips each in 95%, absolute ethanol, and 4 x 10 dips in Clear-Rite (Richard-Allan Scientific). Then slides were mounted with Permount (Fisher Scientific) and examined under a light microscope. The hybridization using only hybridization buffer without the TSV-specific gene probes was also performed on appropriate slides (negative control). For Unicryl resin-embedded tissue sections, the hybridization protocol was modified from a protocol that was developed for HPV (Pantoja & Lightner 2001). Specifically, semi-thin sections were first re-hydrated at RT for 10 min each with HPLC water and 1x TNE (50 mm Tris-HCl, 10 mm NaCl, 1 mm EDTA, ph 7.4). Proteolytic digestion was performed in a humid incubator at 37 C for 15 min with 500 µl of 100 µg ml 1 freshly prepared Proteinase K (Sigma Chemical) diluted in 1x TNE. The digestion was inactivated for 5 min in 0.4% cold formaldehyde, and the sections were rinsed for 5 min at RT with 2x standard saline citrate (SSC) (1x = 0.15 M NaCl, M sodium citrate, ph 7.0). Pre-hybridization was performed by pouring, onto the slides, 200 µl of hybridization buffer (50% formamide, 0.02% Ficoll 400, 0.02% polyvinylpyrolydone 360, 0.02% bovine serum albumin, 5% dextran sulfate, 0.5 mg ml 1 denatured salmon sperm

68 68 DNA, 4x SSC), and the slides were incubated in a humid incubator at 37 C for 30 min. Denatured probes were prepared by adding 2 µl each of probes TS622 and TS624 to 500 µl of hybridization buffer. The mixture of probes was then boiled at 100 C for 10 min and quenched on ice for 5 min. The denatured probes (100 µl) were placed onto each slide, and slides were incubated overnight (~20 h) in a humid incubator at 37 C. The resin sections were subjected to post-hybridization washes with decreasing concentrations of SSC at 37 C (2 x 5 min each in 2x, 1x, 0.5x, and 0.1x SSC). The slides were soaked in Buffer I (0.1 M Tris-HCl, 0.15 M NaCl, ph 7.5) at 37 C for 5 min and blocked at 37 C for 15 min with 125 µl of Buffer II (blocking buffer) (0.5 ml of 10 mg ml 1 Blocking reagent [Roche] in Buffer I). Detection of the hybridized probes was performed using a sheep anti-dig antibody conjugated with 15 nm-colloidal gold particles (Electron Microscopy Sciences), diluted 1:50 in Buffer II. The slides were incubated in a humid incubator at 37 C for 2 h. Gold particles not bound to the probes were extensively washed 4 x 5 min each with Buffer I and HPLC water at RT. Amplification of the reacted gold particles was performed using a silver enhancer (Electron Microscopy Sciences), which was placed onto each slide (0.5 ml), and slides were incubated for 15 min in a dark humid chamber at RT. The silver enhancement was terminated in HPLC water for 15 min at RT. Slides were heat-dried at 60 C for 1 min, counterstained for 30 sec with 0.5% toluidine blue in 1% sodium borate, mounted with Permount, and examined under a light microscope.

69 69 EM-ISH. Ultra-thin sections of resin-embedded specimens were processed for electron microscopy. The hybridization procedures were almost identical to those developed for LM-ISH except that the sections were on grids and not glass slides. All reagents were placed as a drop (25 µl) on a piece of hydrophobic film and grids, containing tissue sections, were floated on the drops with the section side facing down, while reagents destined for incubation were placed on a hydrophobic film in a Petri dish with a moistened filter paper in the bottom. Grids were re-hydrated for 5 min each in HPLC water and 1x TNE, and incubated with Proteinase K for 5 min. The reaction was inactivated in cold formaldehyde for 5 min. The grids were washed for 5 min at RT in 2x SSC and pre-hybridized in hybridization buffer for 30 min. Hybridization was performed overnight (~20 h) at 37 C in a humid incubator on drops of a mixture of denatured probes TS622 and TS624. Post-hybridization washes were followed the above procedures, and the grids were blocked in Buffer II for 15 min at 37 C. The detection of probes was performed for 2 h at 37 C using the anti-dig gold conjugated antibody (20 µl). Grids were rinsed 4 x 5 min each in Buffer I and HPLC water. Silver enhancement was performed by floating the grids on a drop (20 µl) of the silver enhancer in a dark humid chamber at RT for 15 min. The reaction was stopped by floating the grids on a drop of HPLC water for 15 min followed by air-drying on a piece of filter paper for at least 3 h.

70 70 Counterstaining was performed using uranyl acetate (UA) and lead citrate (LC) according to Morel et al. (2001). Specifically, for UA staining a grid was floated on a drop of 5% aqueous UA (30 µl) with the tissue section facing down, and incubation was performed in the dark for 30 min. The grid was swirled for 10 sec in a beaker of HPLC water and air-dried for 30 min on a piece of filter paper in a Petri dish. For LC staining, a drop of 10% LC (30 µl) was placed on hydrophilic film in a Petri dish in which a pellet of NaOH had been placed for 3 min to capture carbon dioxide. Then, a grid was floated on the drop with the tissue section facing down for 5 min, washed, and air-dried as described for UA. EM analysis in this study was performed using a JEOL 100CXII model at University of Arizona. Results TEM. The ultrastructure of cells infected with TSV was first characterized by TEM. Structural changes of TSV-infected cells were classified into 3 stages: early, mid, and late stages of TSV infection. The morphological identities of cells at the early stages of infection were the accumulation of cytoplasmic organelles, including rough ER (RER), Golgi that had a normal stacked appearance, and mitochondria, in the perinuclear region of the cells (Fig. 4.1). Also at this stage of infection, intracellular inclusion bodies, indicative of TSV infection, were occasionally seen in the perinuclear region, while the nuclei usually exhibited the normal appearance.

71 71 At the mid-stages of an acute phase infection, the majority of cells developed a large number of RER that had occupied a significant proportion of the cytoplasm, while the nuclei remained largely normal (Fig. 4.2A). The intracellular inclusion bodies were also seen in the perinuclear region of infected cells. Various organelles, including RER, vesicles, and mitochondria, filled the rest of the cytoplasm. Some of the mitochondria were rounded and slightly distended (Fig. 4.2B). More importantly, clusters of developing small membranous vesicles (SMVs) were seen within the cytosol. The RER contained electron-dense materials and were usually adjacent to SMVs. The inner surfaces of RER were sometimes covered by small invaginations. At the late stages of infection, a significant number of intracellular membranes, i.e. RER and SMVs, and free ribosomes, occupied the large proportion of the cytosol that was virtually devoid of other cytoplasmic organelles (Fig. 4.3). The RER contained electron-dense materials, while the rest of the cytoplasm was also filled with electrondense particles. Some RER were contiguous with the outer nuclear membrane, which usually displayed a pyknotic appearance, indicative of TSV infection. More interestingly, TSV particles seen as spherical bodies were observed within the cytoplasm of cells at late stages of infection (Fig. 4.4). Again, the cells had lost all cytoplasmic organelles and largely reduced in volume. It is also worth noting that the pyknotic nuclei of TSVinfected cells were eccentric on one side of the cells; these results determined by TEM are in accordance with those by EM-ISH illustrated later in Figs. 4.7 to 4.9.

72 72 LM-ISH. Hybridization by TSV-specific cdna probes was first investigated by light microscopy using paraffin-embedded tissue sections of TSV-infected shrimp. Hybridization signals were visualized using a sheep anti-dig antibody conjugated to 15 nm-colloidal gold particles. Black precipitates indicative of the presence of the TSV RNA genomes were observed in various epithelial tissues, including gills, cuticular, and stomach epithelia (Fig. 4.5A). Weak labeling signals were also seen within antennal glands and lymphoid organ (not shown). Background labeling caused by a non-specific deposition of silver was detected, but it was not associated with TSV-infected cells. No labeling was observed in tissues that were treated only with hybridization buffer without the TSV-specific gene probes (negative control) (Fig. 4.5B). The positive reaction to the TSV-specific cdna probes was also seen on semithin sections of Unicryl resin-embedded tissues. Epithelial cells of the gills displayed strong labeling with the TSV-specific gene probes (Fig. 4.6). The hybridization signals appeared almost exclusively within the cytoplasm of infected cells. Pyknotic and/or karyorrhectic nuclei and intracytoplasmic inclusion bodies can also be seen within TSVinfected cells. Additionally, intriguing ordered membranous structures, which may be derived from degenerated distended mitochondria, were observed in the cytosol (Fig. 4.6B) (see also Figs. 4.7 & 4.8). The intensity of hybridization signal on resin-embedded tissue sections was further investigated and optimized. Additional testing revealed that the labeling intensity

73 73 can be significantly increased during 3 steps: (1) probe detection, (2) blocking, and (3) silver-enhancing steps (not shown). More specifically, hybridization yielded the best signal intensity with the anti-dig antibody at dilution of 1:50, followed by incubation for ~15 min each with blocking buffer and silver enhancer. Besides, non-specific deposition of silver was found to occur frequently when the incubation time exceeded 30 min. These optimized time and concentration were restrictedly applied for EM-ISH. EM-ISH. Hybridization signals to the TSV-specific cdna probes were also observed by electron microscopy in ultra-thin sections of resin-embedded tissues. TSVinfected cells of the gill epithelium could be identified because they specifically reacted to the hybridized probes seen as black precipitates. Their nuclei, which were relatively free of hybridization signals in the nucleoplasm, showed pyknosis that is characteristic to late stages of TSV infection (Fig. 4.7). Unidentified membranous structures, which may be derived from distended degenerated mitochondria as illustrated by light microscopy (Fig. 4.6B), were also visualized in the cytoplasm by electron microscopy (Fig. 4.7A). A positive reaction to the TSV-specific gene probes was also detected within circulating cells in the hemocoel (Fig. 4.7C); these cells may be hemocytes or necrotic cells that sloughed from the gill epithelium or stromal matrix. In addition, the nuclei of cells that had lost much of their cytoplasmic contents, found in the hemocoel, displayed a very weak positive reaction by ISH (Fig. 4.7D). It was also noted that a positive reaction by ISH was present in between the inner and outer nuclear membranes (Figs. 4.7A,C,

74 74 4.8A,C, and 4.9); however, higher magnification revealed that the region did not contain TSV-like particles. Cells infected with TSV displayed the membrane rearrangement within the cytoplasm. A positive reaction to the TSV-specific gene probes was restricted to the cytosol, confirming that it is the replication site of the new virus progenies as suggested and illustrated in Figs. 4.1 to 4.4 by TEM. Specifically, a large number of RER and SMVs were dispersed throughout the cytosol of TSV-infected cells (Fig. 4.8A,B). The RER were frequently juxtaposed to mitochondria and showed continuity to the outer nuclear membrane (Fig. 4.8A,C). There was also some evidence of small invaginations and electron-dense materials within these membranous structures; more intriguingly, the small invaginations found on the inner surfaces of the RER displayed a positive reaction to the TSV-specific gene probes (Fig. 4.8C,D). Intracellular inclusion bodies were also demonstrated by EM-ISH; they comprised SMVs and unclassified membranous structures (Fig. 4.8E). The SMVs observed within the inclusion bodies did react to the TSV-specific gene probes by ISH. In contrast, the unclassified membranous structures that may be derived from distended degenerated mitochondria, sometimes seen in the perinuclear region, exhibited a relatively weak reaction to the TSV-specific gene probes when compared to the developing replication sites (Fig. 4.8E,F). Electron-dense spherical bodies, or spherules approximately 30 nm in diameter when compared to 15-nm gold particles, were observed in the cytoplasm of TSV-infected

75 75 cells (Fig. 4.9A). More interestingly, the gold particles, indicative of the presence of the TSV RNA genomes, were seen in close association to the spherical bodies on the proliferating membranes (Fig. 4.9B). These findings strongly suggest that the spherical bodies were the TSV particles. Moreover, the infected epithelial cells usually contained a large number of replication sites within the cytoplasm, and occasionally they were undergoing a sloughing process since they were almost excluded from intercellular matrices, reduced in volume, and had relatively fewer cytoplasmic organelles when compared to uninfected cells. Discussion Replication of all single-stranded positive-sense RNA viruses is thought to be tightly linked to the rearrangement of cellular membranes that ultimately wrap around the viral replication complexes (Mackenzie 2005). In this present study, several ultrastructural features that are hallmarks of the membrane rearrangement were observed by TEM. At early to late stages of TSV infection, cellular organelles such as RER appeared to cluster within the perinuclear region, while mitochondria were largely seen at the periphery of infected cells (Figs. 4.1 to 4.3). Eccentricity of the nuclei as illustrated in Figs. 4.3, 4.4, and 4.7 to 4.9 also suggests that the viral replication takes place exclusively on one side of the cells. In cells infected by Bunyaviruses, FMDV, and SARS-CoV, ultrastructural analysis has recently shown that the cellular organelles moved to the perinuclear region on one side of the cells at the early stages of infection (Goldsmith et al.

76 , Monaghan et al. 2004, Novoa et al. 2005). Ng & Hong (1989) also reported that the proliferation of the RER is the earliest visible event after Kunjin virus infection (also reviewed by Novoa et al. 2005). Thus, the characteristic distribution and proliferation of the cellular organelles observed in this present work suggests that the TSV replication might take place in a defined region of the cytoplasm. Additionally in this present study, TEM analysis at a high magnification revealed that SMVs were generated within the cytosol of TSV-infected cells (Fig. 4.2B). The cluster of SMVs consisted of numerous vesicles and was surrounded by distended RER that contained electron-dense materials. The induction and proliferation of these unique cytoplasmic membrane structures have been previously described as convoluted membranes, paracrystalline arrays, and small vesicular structures or vesicle packets (Mackenzie et al. 2005). More importantly, replication of Flaviviruses such as Kunjin virus, West Nile virus (WNV), and Tick-borne encephalitis virus, is believed to take place in the membrane structures (Hong & Ng 1987, Ng et al. 1994, Lorenz et al. 2003). The proximity between the clustered SMVs and the RER as illustrated in Figs. 4.2B, 4.3, and 4.8D also suggests an association between these two structures. For Poliovirus, the replication complex has also been shown to occur within ER-derived vesicles (Rust et al. 2001), while the replication of Flaviviruses has been reported to take place in Golgiderived vesicles (Mackenzie et al. 1999).

77 77 Our results obtained by ISH are in accordance with those illustrated by TEM. A positive reaction to TSV-specific gene probes was seen in close association to the membranous structures (Figs. 4.7 & 4.8). These confirmed that the viral RNA is associated with the proliferating membranes. Unlike the RER and SMVs, mitochondria may not be directly involved in the biosynthesis of TSV RNA genomes because they did not react to the TSV-specific gene probes by ISH (Fig. 4.7). However, the significant accumulation of mitochondria in the cytosol at early to mid stages of TSV infection (Figs. 4.1 & 4.2) and their presence within intracellular inclusion bodies and the perinuclear region (Fig. 4.7E,F) suggest that they actively participate in the formation of viral replication complex. Miller et al. (2001) demonstrated that Flock house virus RNA replicates on outer mitochondrial membranes, and Novoa et al. (2005) suggested that mitochondria, cytoplasmic membranes, and cytoskeletons supply factors for key steps of the viral replication. Replication of all RNA viruses is believed to take place exclusively within the cytoplasm. However, recent studies have shown that a significant proportion (20%) of the total RNA-dependent RNA polymerase (RdRp) activity from cells infected with Dengue virus, Japanese encephalitis virus (JEV), and WNV is resident within the nucleus; furthermore, the major replicase proteins of JEV also localized within the nucleus by confocal microscopy and IEM (Uchil et al. 2006). Therefore, the host cell nucleus clearly functions as an additional site for the replication of Flaviviruses. In the present study, a positive reaction by ISH was evident in between the inner and outer nuclear membranes

78 78 (Figs. 4.7A,C, 4.8A,C, and 4.9). These results implicate that there is the accumulation and association of TSV RNA with the nuclear membrane although the viral particles were not seen within the nucleus by TEM or ISH. In addition, the findings indicate that additional studies will be needed to elucidate the presence of functionally active TSV RNA synthesis within the host cell nucleus or in association with the nuclear membrane. Following TEM, inclusion bodies were frequently observed, and they occupied a large area within the cytoplasm of TSV-infected cells (Figs. 4.1 to 4.3). Unfortunately, we were unable to identify immature or mature TSV particles within the inclusion bodies stained only with UA and LC; however, the results in Fig. 4.8E demonstrated that the TSV RNA genomes were present within the inclusion bodies because the viral genomes reacted to the TSV-specific gene probes by ISH. More intriguingly, TSV particles seen as spherical bodies were observed by TEM (Fig. 4.4), and following EM-ISH, they were seen with gold particles near the vesicular membranes (Fig. 4.9). Thus, these results confirmed that the cellular membranes carry TSV RNA genomes and that they are important for the morphogenesis and propagation of TSV. In conclusion, infection by TSV results in alterations of intracellular membranes within the cytosol. We demonstrated here by TEM and EM-ISH that TSV RNA genomes and TSV particles are associated with the proliferating membranes. Further characterization on cellular origin and biochemical compositions of these vesicular

79 79 membranes will shed light on biogenesis and propagation strategy of TSV in shrimp infected cells. Acknowledgements. This work was supported by Gulf Coast Research Laboratory Consortium Marine Shrimp Farming Program, CSREES, USDA, Grant no T. S. was supported by a scholarship from the Royal Government of Thailand. We thank Dr. David L. Bentley and Dr. Kathy F. J. Tang-Nelson for technical assistance.

80 80 Figure 4.1. Litopenaeus vannamei. Ultrastructural changes in cells at early stages of infection with Taura syndrome virus (TSV). The nucleus (N) exhibits a normal appearance. The cytoplasmic organelles, including Golgi (short arrows), rough endoplasmic reticulum (RER, long arrows), and mitochondria (Mi) can be seen in the perinuclear region. Also shown in the cytoplasm are large electron-dense inclusion bodies (I), endocytic vacuoles (E), and lysosomes (L). Scale bar = 0.5 µm

81 Figure 4.2. Litopenaeus vannamei. Ultrastructural changes in cells at the mid-stages of an acute phase infection with Taura syndrome virus (TSV). (A) Nucleus (N) displays a normal appearance, while the cytoplasm contains a large number of rough endoplasmic reticulum (RER) and intracellular inclusion bodies (I). (B) Higher magnification of an equivalent region shows that the cytosol contains clusters of developing small membranous vesicles (SMVs), distended mitochondria (Mi), and RER that are covered by small invaginations (arrows) studded with electron-dense particles. Scale bars = 0.5 µm 81

82 82 Figure 4.3. Litopenaeus vannamei. Ultrastructural changes in cells at late stages of infection with Taura syndrome virus (TSV). (A) The pyknotic nucleus (N) is surrounded by vesicular distribution of the nuclear membrane. The cytoplasm contains a large number of small membranous vesicles (SMVs) and rough endoplasmic reticulum (RER) that carry electron-dense materials (arrow). (B) Higher magnification of the rectangular area in (A), rotated by 90, shows SMVs and RER. Scale bars = 0.5 µm

83 83 Figure 4.4. Litopenaeus vannamei. Transmission electron micrographs of Taura syndrome virus (TSV) in an infected cell. At (A) low and (B) high magnifications, the nucleus (N in A) shows developing pyknosis as indicated by increased electron density of the nucleoplasm, while TSV particles (arrowhead in B) are present in the cytoplasm. Scale bars = 0.5 µm

84 84 Figure 4.5. Litopenaeus vannamei. Detection of Taura syndrome virus (TSV) with in situ hybridization (ISH) in paraffin-embedded tissues by light microscopy. (A) The cuticular epithelium of the stomach displays a positive hybridization reaction (arrows) to TSVspecific cdna probes using an anti-digoxigenin antibody coupled with 15 nm-colloidal gold particles. (B) Negative control. Scale bars = 50 µm

85 85 Figure 4.6. Litopenaeus vannamei. Detection of Taura syndrome virus (TSV) with in situ hybridization (ISH) in resin-embedded tissues by light microscopy. (A) Gill filaments; (B) cross section of gill central axis. TSV RNA is detected by TSV-specific cdna probes with an anti-digoxigenin antibody coupled to 15 nm-colloidal gold particles (dashed circles). Also note that TSV-infected cells display pyknotic nuclei (white arrowhead in A) and cytoplasmic inclusion bodies (asterisks in B) and contain unknown membranous structures, which may be distended degenerated mitochondria (black arrowheads in B). Scale bars = 25 µm

86 86 Figure 4.7. Litopenaeus vannamei. Ultrastrutural features of cells at late stages of Taura syndrome virus (TSV) infection examined by in situ hybridization (ISH) using TSVspecific cdna probes. Infected cells in the gills are highly vacuolated and display strong (A C) to weak (D) labeling signals within the cytoplasm, while the nuclei (N) are completely devoid of signal. Also note (A) an unknown membranous structure, which may be degenerated distended mitochondrion (Mi); (C) an epithelial cell undergoing a sloughing process found in the hemocoel (Hec); (D) a nucleus with juxtaposed vacuoles and few cytoplasmic contents from a cell that had been separated from the gill epithelium. Scale bars = 1 µm

87 Fig. 4.7 (continued) 87

88 Fig. 4.7 (continued) 88

89 89 Figure 4.8. Litopenaeus vannamei. Ultrastructural features of membrane rearrangement in Taura syndrome virus (TSV)-infected cells from gills analyzed by in situ hybridization (ISH) using TSV-specific cdna probes. (A) Infected cell illustrating a pyknotic nucleus (N) and contiguous rough endoplasmic reticulum (RER) that is presumably connected to mitochondria (Mi). (B) Higher magnification of an equivalent region within the cytosol that is highly vacuolated. (C, D): Positive reaction by ISH (arrows) within RER (C) and small membranous vesicles (SMVs in D). Also indicated are electron-dense materials (short arrow in C) and SMVs unbound to gold particles (arrowheads in D). (E, F): Unknown membranous structures (clear arrows) that may be derived from distended degenerated mitochondria and which are relatively devoid of hybridization signals. (E) Randomly organized membranes are located within inclusion bodies. (F) Higher magnification of Fig. 4.7A displays uniformly organized membranes in the perinuclear area. Scale bars = 0.5 µm

90 Fig. 4.8 (continued) 90

91 Fig. 4.8 (continued) 91

92 Fig. 4.8 (continued) 92

93 93 Figure 4.9. Litopenaeus vannamei. Ultrastructural visualization of Taura syndrome virus (TSV) in an infected cell tested by in situ hybridization (ISH) using TSV-specific cdna probes. The same cell from gill epithelium is shown at (A) low and (B) high magnifications. Gold particles (15 nm, white arrowheads in B), indicative of the TSV RNA genomes, are closely associated with TSV particles seen as electron-dense spherical bodies (approximately 30 nm in diameter, black arrowheads in B). Pyknotic nucleus (N) and mitochondria (Mi) are also indicated. Scale bars = 0.5 µm

94 Fig. 4.9 (continued) 94