To a Model Pathogen, Spring Viremia of Carp Virus

Susceptibility of Zebrafish (Danio rerio ) to a Model Pathogen, Spring Viremia of Carp Virus

George E. Sanders, DVM,1, 2,* William N. Batts, and James R. Winton, PhD1

To improve our understanding of the genetic basis of fish disease, we developed a pathogen model, using zebrafish (Danio rerio ) and spring virema of carp virus (SVCV). Replicate groups of 10 fish were acclimated to 20 or 24°C, then were exposed to SVCV concentrations of 103 to 105 plaque-forming units per milliliter (PFU/ml) of water and observed daily. In a second trial, fish were acclimated to 15°C, and replicate groups of 10 fish were exposed to SVCV at a concentration of 105 PFU/ml; however, the temperature was raised 1°C/wk. Moribund fish were collected for histologic examination, and dead fish were assayed for virus by use of cell culture and reverse transcriptase-polymerase chain reaction (RT-PCR) analysis. Mortality exceeded 50% in fish exposed to 105 PFU of SVCV/ml at the lower temperatures. Clinical signs of disease became evident seven days after viral exposure and were observed most consistently in fish of the 105 PFU/ml groups. Affected zebrafish were anorectic and listless, with epidermal petechial hemorrhages followed by death. Use of plaque assays and RT-PCR analysis confirmed presence of SVCV at titers > 104 PFU/g of tissue. Histologic lesions included multifocal brachial necrosis and melanomacrophage proliferation in gills, liver, and kidneys. These results indicate that zebrafish are susceptible to infection by SVCV under conditions that mimic a natural route of exposure.

Zebrafish (Danio rerio) are small (3 to 4 cm), freshwater, tropical cyprinids that grow optimally between 25 and 31°C (Fig. 1A) (40). Females exhibit exceptional fecundity and can produce over 200 eggs/wk; resulting embryos develop ex utero, and are transparent, which facilitates visualization and ma- nipulation (4, 8-11, 37). Zebrafish have a short generation time (3 to 4 months), compared with that of other, larger teleosts used in fish disease research (e.g., rainbow trout, Oncorhynchus mykiss), but only a few common diseases affect them naturally (8, 40). These diseases, none of which currently includes a viral etiol- ogy, are the external parasites Piscinoodinium pillulare (velvet disease), Ichthyophthirius multifiliis (ichthyophthiriosis), and Gyrodactylus and Trichodina spp. and the internal freshwater pro- tozoan, Pseudoloma neurophilia (microsporidiosis), gram-negative bacteria (Aeromonas spp. and Flavobacterium columnare), acid- fast bacteria (typically, Mycobacterium marinum, M. fortuitum, or M. chelonae [mycobacteriosis]), and various monogenetic nema- todes (intestinal capillariasis) (3, 8, 20, 24, 34, 40). Due to its use in developmental and genetic research, the zebrafish has become a powerful model organism for study of vertebrate biology, and a large number of characterized strains are maintained at various research facilities and zebrafish stock centers (4, 9-11). Some of the mutant strains of zebrafish have pathologic conditions similar to human diseases, including hematopoietic, cardiovascular, renal, endocrine, and neurologic disorders (4, 9-11). The full genome sequence of the zebrafish is being completed, facilitating development of a full range of ge- Figure 1. Normal female zebrafish (A) and female zebrafish infected with 105 plaque-forming units (PFU) of spring viremia of carp virus Received: 1/27/03. Revision requested: 4/22/03. Accepted: 6/04/03. (SVCV)/ml (B). Notice presence of extensive epidermal hemorrhages. 1Western Fisheries Research Center, 6505 NE 65th Street, Seattle, Washington 98115, and 2Department of Comparative Medicine, University of Washington nomic tools and novel gene expression assays (11, 28, 39). Thus, School of Medicine, T-160 Health Sciences Center Box 357190, Seattle, Wash- ington 98195-7190. a functional disease model for this species would be a great as- *Corresponding author. set in advancing our understanding of the genetic basis of fish 514 Spring viremia of carp virus in zebrafish

disease as well as providing a possible model for diseases of buffer) was seeded into each well of a 24-well cell culture plate. other vertebrate hosts. Cells were incubated for 24 to 48 h at 25°C to become confluent The viral pathogen that we selected for this model was spring monolayers. Ten-fold, serial dilutions (10-1 to 10-5) of viral viremia of carp virus (SVCV). Currently listed as a species of the sample were created in sterile 1.5-ml Eppendorf tubes with genus Vesiculovirus within the Rhabdoviridae (2, 13, 38), SVCV is MEM-5-T. A 0.1-ml volume of each serial dilution was added to an enveloped, negative-sense, single-stranded RNA virus having a each corresponding set of three wells of the 24-well plate, and bullet-shaped virion approximately 60 to 90 nm wide and 90 to 180 the plates were incubated for 15 to 20 min to allow viral adsorp- nm long. The virus is ether, heat, and acid labile, stable at pH 7 to tion to cells. Negative-control wells received 0.1 ml of MEM-5-T, 10, and replicates optimally between 20 and 22°C (13, 14, 35, 38, and positive-control wells received 0.1 ml of a serial dilution of 41). This virus is the causative agent of acute infectious dropsy of frozen stock SVCV. After incubation, wells were subsequently the common carp (Cyprinus carpio) or spring viremia of carp (14, covered by one milliliter of methylcellulose overlay (MEM-5-T 17, 38, 41). In its normal hosts, SVCV is transmitted horizontally containing 0.75% [wt./vol.] methyl cellulose, and100 IU of peni- (including via mechanical and biological vectors), the incubation cillin, 100 mg of streptomycin [Gibco BRL], 2.5 mg of amphotericin period is temperature dependent (average 10 to 17 days), and the B [Fungizone, Gibco BRL, Grand Island, N.Y.], and 100 mg of gen- portal of entry is the gills, with subsequent viremia and dissemi- tamicin sulfate [Gibco BRL]/ml, buffered to pH 7.8 with 1M Tris nation to the kidneys, liver, spleen, heart, and gastrointestinal and sodium bicarbonate) and incubated at 20°C for 72 to 96 h. tract (1, 2, 14, 38, 41). We selected SVCV for the viral component Viral titer was expressed as plaque-forming units per milliliter. of this model because this pathogen replicates well near the nor- Animal selection, housing, and husbandry. The AB mal temperature of our host (25 to 31°C) and is extremely patho- strain of zebrafish was selected because it is a wild-type strain genic to cyprinid species indigenous to Europe (2, 25, 41). that is commonly used in research. Fish were bred in-house Because SVCV causes a disease that is notifiable to the Office In- from breeders maintained in a closed colony for multiple gen- ternational des Epizooties (O.I.E.), the in vivo component of this erations. This strain is also similar genetically to the Tübingen research was carried out within the confines of an aquatic ani- strain from Germany (12, 40). Zebrafish were reared under the mal biohazard level-three laboratory (ABL-3) located at the guidelines provided by the Guide for the Care and Use of Labo- Western Fisheries Research Center (WFRC) to prevent possible ratory Animals (27) and the United States Public Health Ser- introduction to the local environment (14, 25, 31, 32). Experimen- vice Policy on the Humane Care and use of Laboratory Animals. tal work with this pathogen should be limited to laboratories ad- The protocol for experimental use of these animals was ap- equately equipped to handle this agent safely to prevent its proved by the Institutional Animal Care and Use Committee of introduction into naïve aquatic environments. the University of Washington (Seattle, Wash.). Zebrafish were spawned using standard conditions (40), and Materials and Methods approximately 200 juveniles were transferred at two months of Cell culture. Mycoplasma-free cultures of epithelioma age to the wetlab of the WFRC from a zebrafish colony at the papulosum cyprini (EPC) cells were obtained from Dr. G. University of Washington (Seattle, Wash.). Fish were maintained Kurath (WFRC, Seattle, Wash.). Cells were grown without anti- in 5-L opaque containers, which received sand-filtered, ultravio- biotics in minimal essential medium (MEM) supplemented with let-irradiated, fresh water in flow-through manner (2.4 L/min) 0.3% tryptose phosphate broth (TPB; Difco Laboratories De- with one air-stone per container. Fish were maintained at 26°C troit, Mich.), 10% fetal bovine serum (FBS; Hyclone, Logan, and a 14/10-h cycle of light and dark. Utah), and 2 mM L-glutamine (Gibco BRL, Grand Island, N.Y.), Fish were fed No. 2 salmon starter diet (Bioproducts, Warrenton, then buffered to a pH of 7.8 by addition of sodium bicarbonate Oreg.) and were monitored twice daily. Food was withheld 24 h be- (MEM-10-SB). Cells were removed from the flasks after addi- fore and after movement. On reaching an average size of three to tion of a trypsin-versene solution and were split at a ratio of 1:3, four cm at eight months of age, fish were transferred to the ABL-3 using standard aseptic techniques (21, 30, 42). Cell cultures facility and were maintained as described. were incubated at 25°C for initial growth and were maintained Prior to the start of and during this project, zebrafish from at 20°C when the monolayers became confluent. the re-circulating system where the fish for this experiment Virus propagation. The reference strain of SVCV (ATCC were spawned, and other systems within this colony were VR-1390, Manassas, Va.) was used for in vitro and in vivo work. evaluated for previous exposure to SVCV. Results of viral isola- To prepare virus stocks, a frozen vial of SVCV containing 5.7 ´ tion and reverse transcriptase-polymerase chain reaction (RT- 106 plaque-forming units (PFU)/ml was thawed and one millili- PCR) assays of homogenized juvenile and adult zebrafish were ter of stock was used to inoculate each of four, 150-cm2 flasks negative. Histologic evaluation of juvenile and adult zebrafish containing EPC cell monolayers. Following a viral adsorption did not reveal the typical lesions associated with the presence of period of 10 to 15 min, 25 ml of MEM-10-SB was added to each previously mentioned pathogens, with the exception of low lev- flask. Cells were incubated at 25°C for 48 h until the cells be- els of microsporidiosis and mycobacteriosis. came detached, eventually lysed, and the monolayer sloughed Virus exposure of fish. Stocks of SVCV were thawed and from the bottom of the flask. The contents of the flasks were diluted in MEM-10-SB to prepare a set of virus suspensions of pooled and dispensed as aliquots, then frozen at –70°C. known titer. For experimental exposure of zebrafish to the virus, Determination of virus titer. For all viral assays, we used the water level in each container was decreased from 5 L to 1 L, the plaque assay protocol of Burke and Mulcahy (7) with EPC and five milliliters of a virus suspension prepared in MEM-10- cells. One milliliter of an EPC cell suspension in MEM-5-T SB was added to provide final concentrations in the aquaria of 3 4 5 (MEM supplemented with 0.3% TPB, 5% FBS, and 2 mM L- 10 , 10 , or 10 PFU/ml. Control groups received five milliliters glutamine and buffered to pH of 7.8 by addition of 1M Tris of MEM-10-SB without virus. Fish were allowed to remain in 515 Vol 53, No 5 Comparative Medicine October 2003 contact with the virus suspensions for 30 min. After the allotted of penicillin, 100 mg of streptomycin, 2.5 mg of amphotericin B, contact time, the water was rapidly replaced by the re-establish- and 100 mg of gentamicin sulfate/ml), then were placed in a ment of water flow to the tanks, the water level was returned to Stomacher bag (Steward Medical UAC House, London, UK) and 5 L, and aeration of the water was resumed. homogenized in a Stomacher lab-blender (Steward Medical First challenge. In the ABL-3 laboratory, 10 zebrafish per UAC House, London, UK) for 15 to 30 sec. Sample homogenates container were randomly distributed to four sets (1–4) in two were transferred from the bags to sterile centrifuge tubes and replicate tanks in each of two rooms. The water temperature in spun at 1,000 ´ g for 10 min at 4°C. A 0.1-ml volume of the ho- each room was maintained at 20 or 24°C, respectively. Experi- mogenate supernatant was used to create the series of ten-fold mental groups received five milliliters of a suspension of SVCV dilutions for the viral plaque assays, as described previously, prepared to provide challenge doses of 103, 104, or 105 PFU/ml of and the remaining volume was stored at 4°C. Titers were calcu- water. Control groups received the same volume of MEM-10-SB. lated as plaque-forming units per gram of tissue (PFU/g). A daily log, including mortality count, morbidity evaluation, Virus confirmation. Selected isolates were evaluated by use tank number, and fish behavior (feeding and activity), was kept of two methods. Primers for RT-PCR analysis were synthesized in each room during the experiment. Dead fish were frozen indi- from sequences of the SVCV glycoprotein gene (6, 18). The sense vidually in labeled whirl-pak bags (Nasco, Lincolnshire, Ill.) at – primer, 5’TCATTTAGAGCCATACG3’, encompassed the distance 70°C for future viral isolation, titer determination, and RT-PCR between nucleotides 466 and 482, and the antisense primer, analysis. Moribund fish were euthanized by use of tricaine 5’GAAAGCATGTTGAAGCT3’, encompassed the distance be- methanesulfonate (MS-222, Finquel, Argent, Redmond, Wash.) tween nucleotides1252 and 1268. When used in the RT-PCR and the bodies were fixed by perfusion of the coelomic cavity analysis, these primers directed the synthesis of an amplification with either Davidson’s or Bouin’s fixative for five to seven days product of 803 base pairs (bp). The master mixture created for before processing for routine histologic evaluation. After four each round of RT-PCR analysis contained molecular-grade dis- weeks, all remaining experimental fish were euthanized and tilled water (mdH20), 1 X PCR buffer (Promega, Madison, Wis.), the bodies were fixed for histologic evaluation. Due to the lim- 2.5 mM MgCl2 (Promega), 200 mM dNTP (200 mM each dATP, ited number of zebrafish available, remaining control fish were dCTP, dGTP, dTTP; Promega), 50 pmol of sense and antisense removed from tanks and maintained for later use in other ex- primers, 5 U of Taq DNA polymerase (Promega), 9 U of AMV re- periments. The mean day-to-death (MDD) of fish dying in each verse transcriptase (Promega), and 40 U of RNase inhibitor replicate was calculated. (RNasin, Promega). After the master mixture was created, it Second challenge. The temperature of the water provided to was aliquoted into labeled 0.2-ml thin-walled tubes and stored the containers of stock fish was decreased from 20°C to 15°C over at 4°C for a final reaction volume of 50 ml. Sample homogenate a period of five days. Fish were allowed to acclimate to the lower supernatants from frozen fish, as described previously, were di- temperature for a period of two weeks. As before, replicate groups luted 1:20 (5 ml of sample plus 95 ml of mdH20) in 1.5-ml vault- of 10 fish were challenged by immersion; however, only one con- lock tubes. Positive-control virus, from EPC-infected cells, was centration of virus, 105 PFU/ml, was used. The control replicates similarly diluted 1:20 in 1.5-ml vault-lock tubes. After appropri- were exposed to MEM-10-SB. One-week after challenge, the wa- ate mixing, unknown samples and the positive control were ter temperature was increased by 1°C/wk until a final tempera- placed in a heat block for two minutes at 95°C for release of vi- ture of 20°C was reached. As a result, the duration of this ral RNA, then were placed on ice for two minutes. experiment was six weeks. As before, a daily log was kept, dead Unknown samples and the positive control were added to the fish were frozen, and moribund fish were euthanized and fixed. master mixture and were spun briefly to accumulate fluid at bot- The MDD of fish dying in each replicate was calculated. tom of tubes. The negative-control sample contained master mix-

Statistical analysis. Mortality data from challenges 1 and 2 ture plus 5 µl of mdH20. Samples were placed in a thermal cycler were used to create survival curves estimated by use of Kaplan- programmed to run at 50°C for 15 min, 95°C for two minutes, 30 Meier and compared by use of log-rank tests, with significance cycles (95°C for 30 sec, 50°C for 30 sec, 72°C for 60 sec), 72°C for set at P = 0.05. The MDD data results were compared, using 7 min, and a 4°C hold indefinitely. Following standard protocols, analysis of variance (ANOVA), with significance set at P = 0.05. an electrophoretic gel was made using 1.5% DNA-grade agarose Statistics were accomplished using SYSTAT 8.0 computer soft- and 1 X Tris boric acid (ethylenediaminetetra acetic acid) solu- ware (SPSS Inc. Chicago, Ill.). tion. All PCR samples were mixed with 6 X loading dye and Histologic techniques. Davidson’s and Bouin’s fixatives were placed in wells of the gel. In the peripheral wells, 100-bp DNA prepared per standard protocols (23). The entire fish, excluding the ladders (Gibco BRL) were loaded for reference. The gel was run caudal tail fin, was placed into a standard tissue cassette for rou- for 45 min at 0.1 ampere, stained with ethidium bromide solu- tine histologic processing overnight. After embedding left side tion, and documented via photography. down in paraffin blocks, 5-mm-thick sagittal sections were cut on a A viral neutralization test, as described in the O.I.E. diagnos- rotary microtome (American Optical, Buffalo, N.Y.) and trans- tic manual for aquatic animal diseases (31), was also used to ferred to charged glass slides (Probe-On, Fisher Scientific, Pitts- confirm that the viral isolate used to expose the fish and the iso- burgh, Pa.). One set of slides was stained in routine manner with lates subsequently recovered from fish were SVCV. Neutraliz- hematoxylin and eosin (H&E) and glass coverslipped while the ing anti-SVCV polyclonal rabbit antiserum was obtained from replicates were left unstained for potential special staining. Slides the Centre for Environmental, Fisheries, & Aquaculture Science were evaluated by use of brightfield microscopy. Laboratory (Weymouth, Dorset, UK), the O.I.E. Reference Labo- Viral isolation from frozen fish. Zebrafish frozen at –70°C ratory for SVCV (31). To accurately quantify the level of virus were thawed at room temperature. Fish were weighed and di- neutralization by the serum, a quantitative plaque assay tech- luted 1:40 with MEM-0 A&F (MEM with 0.3% TPB, and 100 IU nique was used. 516 Spring viremia of carp virus in zebrafish

Table 1. In vivo mortality at 20°C in experiment 1 Viral concentration Average % mortality Average mean day to death (± SEM) (± SEM)

105 55 (± 3.9) 10.3 (± 0.51) 104 15.5 (± 1.6) 7.8 (± 0.44) 103 0 0 Control 0 0

Table 2. In vivo mortality at 24°C in experiment 1 Viral concentration Average % mortality Average mean day to death (± SEM) (± SEM)

105 30 8.4 (± 0.58) 104 0 0 103 10 6 Control 0 0

Figure 3. Results from the first challenge trial of zebrafish with SVCV. Mortality shown is the average of replicate groups of 10 fish exposed to a waterborne dose of 105 (´ ), 104 (▲), 103 (❒) or 0 PFU of SVCV/ml. Fish were kept at 24°C for 30 days. Since mortality was not observed in the negative-control groups, these data were omitted for clarity.

Table 3. In vivo mortality in experiment 2 Viral concentration Average % mortality Average mean day to death (± SEM) (± SEM)

105 60 (± 3.5) 10.9 (± 0.35) Control 0 0

Figure 2. Results from the first challenge trial of zebrafish with SVCV. Mortality shown is the average of replicate groups of 10 fish exposed to a waterborne dose of 105 (´ ), 104 (▲), 103 (❒) or 0 PFU of SVCV/ml. Fish were kept at 20°C for 30 days. Since mortality was not observed in the negative-control group, these data were omitted for clarity. Error bars represent SEM at each data point.

Results Viral challenges. At thawing, stocks of SVCV that were frozen at –70°C in MEM-10-SB had titer of 2.8 ´ 107 PFU/ml. Dilu- tions of this stock were made to provide accurate doses in the experimental challenges. Disease and subsequent mortality was consistently greater with exposure of fish to 105 PFU of SVCV/ml. In the first experimental challenge, fish exposed to 105 PFU of SVCV/ml and kept at 20°C suffered mean ± SEM mortality of 55 (± 3.9) %, whereas that of fish exposed to the same dose and Figure 4. Results from the second challenge trial of zebrafish with kept at 24°C was 30% (Tables 1 and 2). Most mortality occurred SVCV. Mortality shown is the average of replicate groups of 10 fish during the first 2.5 weeks of these challenges (Fig. 2 and 3). exposed to a waterborne dose of 105 (▲) or 0 (❒) PFU of SVCV/ml. Fish In the second experimental challenge, initial virus exposure were kept 15°C for seven days, then the temperature (´ ) was increased at 15°C resulted in average mortality of 60% (Table 3). The mor- one degree per week for the remaining five weeks of the trial. tality associated with infection at the lower temperature gener- ally occurred during the first 1.5 weeks after exposure to SVCV fish available for use in each replicate. during this challenge (Fig. 4). Pathologic changes. Signs and gross pathologic changes Although the results of our challenge trials appear to reveal (Fig. 1B) associated with SVCV infection of zebrafish included: effects of both dose and temperature, statistical comparisons of partial to complete anorexia, listlessness, focal to multifocal epi- the survivorship curves from the groups receiving a dose of 105 dermal petechial to ecchymotic hemorrhages, and death. Histo- PFU/ml and kept at different temperatures in both challenge logic evaluation of moribund fish that were euthanized, trials indicated no significant differences. We believe the lack of revealed mild, diffuse branchitis; moderate, diffuse branchial statistical confirmation of the differences we observed is due to necrosis (Fig. 5B); mild, multifocal hepatic and splenic necrosis the low power of the experimental design on the basis of the few (Fig. 6B); and increased numbers of melanomacrophages in the 517 Vol 53, No 5 Comparative Medicine October 2003

Figure 5. Photomicrograph taken of transverse histologic sections of Figure 6. Photomicrograph taken of transverse histologic sections of whole fish. (A) Normal branchial tissue from a control zebrafish. (B) whole fish including pancreas (p) and spleen (s). (A) Normal splenic Severe, diffuse branchitis and branchial necrosis from a zebrafish and pancreatic tissue from a control zebrafish. (B) Normal pancreatic exposed to 105 PFU of SVCV/ml. Paraffin tissues sectioned to 5-mm tissue and severe, diffuse splenic necrosis from a zebrafish exposed to thickness. H&E stain; magnification, 20´ . 105 PFU of SVCV/ml. Paraffin tissues sectioned to 5-mm thickness. H&E stain; magnification, 20´ . gills, liver, and kidneys. Histologic evaluation of experimental zebrafish that survived serum neutralization to confirm that the virus recovered during SVCV exposure, at various concentrations of virus, that were viral plaque assays was indeed the same virus used to challenge euthanized did not reveal significant pathologic changes in the the fish and to evaluate the usefulness of the RT-PCR analysis target organs (gills, liver, pancreas, kidney, intestines, spleen, for frozen tissue samples. Amplification of genomic and messen- and heart) or other organs and vessels that could be attributed ger RNA revealed the presence of 803-bp PCR products (data to SVCV infection or disease. Additionally, there were no lesions not shown). Serum viral neutralization assay revealed a reduc- consistent with diseases attributable to other bacterial, proto- tion of > 1.7 log10 PFU/ml or > 98% reduction in the amount of zoal, or fungal etiologic pathogens. No relevant lesions were ob- virus in the treated samples (zebrafish homogenates and stock served in the control zebrafish exposed only to cell culture SVCV), compared with that in controls (data not shown). Thus, medium (Fig. 1A, 5A, and 6A), and none of these zebrafish had the serum neutralization and PCR assays confirmed the iden- lesions consistent with diseases attributable to other bacterial, tity of the virus we used as SVCV and that the RT-PCR assay protozoal, or fungal etiologic pathogens. provided accurate confirmatory data. Viral plaque assay. A total of 12 of the frozen fish from both experiments were evaluated for the presence of SVCV and all Discussion were found to be virus positive (data not shown). Viral titers Because the highest mortality occurred in fish kept at the were in excess of those (104 PFU/g) associated with death attrib- lower of the two temperatures used in the first SVCV challenge utable to SVCV infection. One fish had a titer of 103 PFU/g. (20°C), an effort was made to replicate the course of SVCV infec- Virus confirmation. Homogenates from selected samples tion in zebrafish that was similar to that described for natural used for the plaque assay were tested by RT-PCR analysis and infections of cyprinids. In European cyprinids, infection with 518 Spring viremia of carp virus in zebrafish

SVCV typically occurs in the fall and winter among susceptible cause of death. This last point is quite important because the populations as the water temperature decreases with seasonal systemic amount of virus in the host may indicate how well the change (1, 2, 5, 14, 25, 41); mortality and morbidity, if not seen immune system is handling exposure to this pathogen. Titer > initially, usually occur as the water temperature increases with 104 PFU/g of tissue is indicative of death associated with SVCV the approach of spring. following exposure via immersion or by intraperitoneal injection The higher percentage of mortality observed in the second (1, 14, 16, 41). As a result, all of the dead fish with this titer or experiment correlated well with natural conditions and with higher most likely died due to direct effects of SVCV infection. It the known functions of the teleost immune system. Adult carp is interesting that the only sample with titer < 104 PFU/g was produce neutralizing antibodies more rapidly and uniformly obtained from a fish that died late in the first experiment. This when kept at 25 rather than 14°C, and viremia with associated fish was exposed to 103 PFU of SVCV/ml in the 24°C challenge mortality occurs in nature, mainly at temperatures below 15°C of experiment 1. In this instance, it may mean that lethal dam- (15, 16). This phenomenon is directly related to the fact that the age is possible at titer < 104 PFU/g of tissue or that the fish im- fish immune system functions optimally near the high end of a mune system had begun to clear the virus, yet the fish given species’ thermo-neutral range. When fish are exposed to succumbed due to failure of damaged organs. The immune sys- temperatures below or at the low end of their range, their hu- tem of most zebrafish that survived initial infection was most moral, cell-mediated, and non-specific immune responses are less likely able to limit the infection of SVCV and eventually clear able to prevent or limit infection (19, 33). Thus, we believe the the virus. To ascertain whether this assumption is correct, decreases in temperature to 20°C (challenge 1) or 15°C (chal- zebrafish that survive exposure to SVCV will have to be evalu- lenge 2) were immunosuppressive for the zebrafish and the re- ated for presence and amount of virus at selected time points. sult was a more pathogenic response of the virus in the host, as Amplification of an RT-PCR product of 803 bp and a serum evident by the increase in mortality. Additionally, this may also neutralization assay were used in support of the viral plaque reflect the ratio of viral replication in comparison with recruit- assay to confirm that the virus recovered from dead zebrafish ment of host’s immune defenses, especially if the rate of the was the same as that used for the experimentally induced infec- former overwhelmed the rate of response of the latter (14, 19, 26). tions. All homogenates from zebrafish not exposed to SVCV had Gross pathologic changes observed in zebrafish were some- negative viral plaque assay results (data not shown). When what similar, without the characteristic coelomic distention, as freshly thawed homogenate was passed once through EPC cells, described for SVCV infections of other cyprinids (1, 2, 14, 25, 35, amplification of appropriately sized bands by use of RT-PCR 41). However, histologic lesions, such as edema, hemorrhage, in- analysis was more consistently observed. This is most likely due flammation, and necrosis, in previously mentioned target or- to the presence of RNA inhibitors released from the homogeni- gans were not similar in presence, severity, or distribution to zation of the entire zebrafish in addition to the normal degrada- those typically associated with SVCV infection of other cyprin- tion of RNA quality during the freezing process, suggesting that ids (1, 2, 14, 29, 35, 41). Most of these lesions have been tradi- an RNA extraction method should be added to the protocol. By tionally observed either in acute experimental studies or during no means is our RT-PCR assay the most sensitive test for evalu- natural disease outbreaks of other cyprinids. As a result, it is ating frozen tissues for the presence of SVCV, and this problem not too surprising that most of the histologic lesions seen in the has been seen with attempts to use PCR-based methods to de- zebrafish were during the acute to sub-acute phase of infection. tect the presence of other rhabdoviruses from frozen fish tissue. The severity of the branchitis and branchial necrosis observed However, this technique has the potential to be at least as spe- in SVCV-infected zebrafish was most likely the cause of death cific as the serum neutralization test for confirming the identity due to the impairment and loss of the respiratory and osmo- of a virus isolate as SVCV. regulatory functions of this organ. Overall, we have developed a reproducible and controlled Zebrafish that survived the viral induced gill damage lived model for a rhabdovirus infection in zebrafish that closely ap- sufficiently long to develop subsequent hepatic and splenic ne- proximates natural infection of cyprinids with SVCV. Due to the crosis from viral dissemination and replication. The increased fact that this pathogen induces a disease in cyprinids that is numbers of melanomacrophages present in specific organs is reportable to national and international agencies, experimental attributable to the presence of SVCV and the host’s immune re- work with this pathogen should be limited to those laboratories sponse to this pathogen. However, the absence of lesions in adequately equipped to handle this agent safely to prevent its some zebrafish surviving exposure to SVCV was curious. It is introduction into naïve aquatic environments. possible that zebrafish without histologic lesions may not have Although others have infected zebrafish with the fish viruses, been individually infected by the virus, however they did come infectious pancreatic necrosis virus or infectious hematopoietic from tanks in which the virus was isolated from other fish. Viral necrosis virus, to study specific aspects of vertical transmission plaque assay was not performed on zebrafish that survived vi- (36) or hematopoiesis (22), important features of this model are rus exposure and did not have pathologic changes because, due that it involves a waterborne challenge and that the course and to its small size, the entire fish is required for either histologic severity of disease can be modulated by alteration of the viral examination or viral identification. To further characterize the dose or temperature. We have recovered and identified SVCV pathologic changes associated with SVCV infection in this model, from moribund fish at titers compatible with death via exposure zebrafish will need to be sampled at selected postexposure inter- to this pathogen and have made a primary assessment to the vals for histologic evaluation. clinical signs of disease and gross and histopathologic changes Re-isolation of SVCV from the frozen fish, determination of associated with this particular virus in this host. However, more virus titer in tissues, and confirmatory identification of the vi- work needs to be done on the pathogenesis of the disease caused rus were used to complete Koch’s postulates and confirm the by SVCV in the zebrafish host. To help in this endeavor, an im- 519 Vol 53, No 5 Comparative Medicine October 2003

munohistochemical assay using a monoclonal antibody against 12. Driever, W. and Z. Rangini. 1993. Characterization of a cell SVCV would be of assistance in determining the route of infec- line derived from zebrafish (Brachydanio rerio) embryos. In Vitro tion and the organs affected in fish sampled at selected times Cell. Dev. Biol. 29:749-754. 13. Emerson, S. U. 1986. Rhabdoviruses, p. 477-490. In B. N. Fields after infection. and D. M. Knipe (ed.), Fundamental virology. Raven Press, New The increasing use of zebrafish in biological research has re- York. sulted in the creation of a large number of mutant strains hav- 14. Fijan, N. 1999. Spring viraemia of carp and other viral diseases ing specific gene deletions, and determination of the full genome and agents or warm-water fish, p. 177-244. In P. T. K. Woo and D. W. Bruno (ed.), Fish diseases and disorders: vol. 3 viral, bacte- sequence will lead to development of a broad range of gene ex- rial, and fungal infections. CABI Publishing, Wallingford, United pression assays. These tools offer unique opportunities to further Kingdom. our understanding of vertebrate biology. The future use of this 15. Fijan, N. 1988. Vaccination against spring viraemia of carp, p. virus disease model with various stocks of mutant zebrafish and 204-215. In A. E. Ellis (ed.). Fish vaccination. Academic Press, novel gene expression assays will be of assistance in studying the London. 16. Fijan, N., Z. Pertinec, Z. Stancl, N. Kezic, and E. Teskeredzic. genetic basis of viral pathogenesis and host response in fish. 1977. Vaccination of carp against spring viraemia: comparison of intaperitoneal and per oral application of live virus to fish kept in ponds. Bull. Off. Int. Epizoot. 87:441-442. 17. Fijan, N., Z. Pertinec, D. Sulimanovic, and L. O. Zwillenberg. Acknowledgments 1971. Isolation of the viral causative agent from the acute form We would like to thank Carla Conway at the Western Fisheries of infectious dropsy of carp. Veteterinarski Archiv., Zagreb. 41:125- Research Center and Charlene Karma of the Department of Com- 138. parative Medicine at the University of Washington for technical as- 18. Goodwin, A. E. and J. R. Winton. In press. V. Spring Viremia sistance in the histology portion of this study. In addition, Denny of Carp. In J.C. Thoesen (ed), Blue book—suggested procedures Liggitt, Gerald Van Hoosier, and Brian Iritani of the Department of for the detection and identification of certain finfish and shell- Comparative Medicine provided invaluable support and assistance fish pathogens, 5th ed. Fish Health Section, American Fisheries during the project. We especially thank Maureen Purcell for her as- Society, Bethesda, Md. sistance with the statistical evaluation of data and Barry Hill for pro- 19. Kennedy-Stoskopf, S. 1993. Immunology, p. 149-159. In M. K. vision of the neutralizing anti-SVCV serum. Finally, we thank David Stoskopf (ed.), Fish medicine. W. B. Saunders Co., Philadelphia. Raible and staff for providing the fish used for this research and the 20. Kent, M. L., J. K. Bishop-Stewart, J. L. Matthews, and J. M. opportunity to learn more about the husbandry of this species. This Spitsbergen. 2002. Pseudocapillaria tomentosa, a nematode research was supported, in part, by grant T32 RR07019 from the Na- pathogen, and associated neoplasms of zebrafish (Danio rerio) tional Institutes of Health in partial fulfillment of the Master’s de- kept in research colonies. Comp. Med. 52:345-358. gree completed by the first author at the University of Washington. 21. Lannan, C. N. 1994. Fish cell culture: a protocol for quality control. J. Tissue Cult. Methods 16:95-98. 22. LaPatra, S. E., L. Barone, G. R. Jones, and L. I. Zon. 2000. References Effects of infectious hematopoietic necrosis virus and infectious 1. Ahne, W. 1978. Uptake and multiplication of spring viraemia of pancreatic necrosis virus infection on hematopoietic precursors carp virus in carp, Cyprinus carpio L. J. Fish Dis. 1:265-268. of the zebrafish. Blood Cells Molecules Dis. 26:445-452. 2. Ahne, W., H. V. Bjorklund, S. Essbauer, N. Fijan, G. Kurath, 23. Luna, L. G. 1968. Preparation of tissues, p. 4-5. In L. G. Luna and J. R. Winton. 2002. Spring viremia of carp (SVC). Dis. Aquat. (ed.), Manual of histologic staining methods of the Armed Forces Org. 52:261-272. Institute of Pathology, 3rd ed. McGraw-Hill, Inc., New York. 3. Astrofsky, K. M., M. D. Schrenzel, R. A. Bullis, R. M. 24. Matthews, J. L., A. M. Brown, J. K. Bishop-Stewart, P. Smolowitz, and J. G. Fox. 2000. Diagnosis and management of Rogers, and M. L. Kent. 2001. Pseudoloma neurophilia n.g., atypical Mycobacterium spp. infections in established laboratory n.sp., a new microsporidium from the central nervous system of zebrafish (Brachydanio rerio) facilities. Comp. Med. 50:666-672. the zebrafish (Danio rerio). J. Eukaryot. Microbiol. 48:227-233. 4. Barut, B. A. and L. I. Zon. 2000. Realizing the potential of 25. McAllister, P. E. 1993. Goldfish, koi, and carp viruses, p. 478- zebrafish as a model for human disease. Physiol. Genomics 2:49- 486. In M. K. Stoskopf (ed.), Fish medicine. W. B. Saunders Co., 51. Philadelphia. 5. Baudouy, A. M., M. Danton, and G. Merle. 1980. Experimen- 26. Mims, C. A. 1964. Aspects of the pathogenesis of virus diseases. tal infection of susceptible carp fingerlings with spring viremia Bact. Rev. 28:30-71. of carp virus, under wintering environmental conditions, p. 23- 27. National Research Council. 1996. Guide for the care and use 27. In W. Ahne (ed.), Fish diseases, Third COPRAQ-Session. of laboratory animals. National Academy Press, Washington, D.C. Springer-Verlag, Berlin. 28. Nature. 2002. News in brief. Nature 418:472. 6. Björklund, H. V., K. H. Higman, and G. Kurath. 1996. The 29. Negele, R. D. 1977. Histopathological changes in some organs glycoprotein genes and gene junctions of the fish rhabdoviruses of experimentally infected carp fingerling with Rhabdovirus spring viremia of carp virus and hirame rhabdovirus: analysis of carpio. Bull. Off. Int. Epiz. 87:449-450. relationships with other rhabdoviruses. Virus Res. 42:65-80. 30. Nicholson, B. L. 1988. Fish cell culture: an overview, p. 191- 7. Burke, J. A. and D. Mulcahy. 1980. Plaque procedure for infec- 194. In Y. Kuroda, E. Kurstak, and K. Maramorosch (ed.), Inver- tious hematopoietic necrosis virus. Appl. Environ. Microbiol. tebrate and fish tissue culture. Japanese Scientific Press, Tokyo. 39:872-876. 31. Office International Des Epizooties. 2000. Spring viraemia 8. Casebolt, D. B., D. J. Speare, and B. S. Horney. 1998. Care of carp, p. 46-52. In OIE Fish Disease Commission (ed.). Diag- and use of fish as laboratory animals: current state of knowl- nostic manual for aquatic azimal diseases, Paris. edge. Lab. Anim. Sci. 48:124-136. 32. Office International Des Epizooties. 2000. Spring viraemia 9. Detrich, H. W., M. Westerfield, and L. I. Zon. 1999. Overview of carp, p. 62-64. In OIE Fish Disease Commission (ed.). Interna- of the zebrafish system. Methods Cell. Biol. 59:3-10. tional Aquatic Animal Health Code, Paris. 10. Dodd, A., P. M. Curtis, L. C. Williams, and D. R. Love. 2000. 33. Press, C. McL. 1998. Immunology of fishes, p. 3-62. In P-P. Zebrafish: bridging the gap between development and disease. Pastoret, P Griebel, H. Bazin, A. Govaerts (ed.), Handbook of ver- Hum. Mol. Genet. 9:2443-2449. tebrate immunology. Academic Press, San Diego, Calif. 11. Dooley, K. and L. I. Zon. 2000. Zebrafish: a model system for 34. Pullium, J. K., D. L. Dillehay, and S. Webb. 1999. High mor- the study of human disease. Curr. Opin. Genet. Dev. 10:252-256. tality in zebrafish (Danio rerio). Contemp. Top. Lab. Anim. Sci. 38:80-83.

520 Spring viremia of carp virus in zebrafish

35. Smail, D. M. and L. S. Munro. 1989. The virology of teleosts, p. 39. Vogel, G. 2000. Sanger will sequence zebrafish genome. Science 173-241. In R. J. Roberts (ed.), Fish pathology. Bailliere Tindall, 290:1671. London. 40. Westerfield, M. 1989. In The zebrafish book: a guide for the labo- 36. Seeley, R. J., A. Perlmutter, and V. A. Seeley. 1977. Inherit- ratory use of zebrafish (Brachydanio rerio). University of Oregon ance and longevity of infectious pancreatic necrosis virus in ze- Press, Eugene, Oreg. bra fish, Brachydanio rerio (Hamilton-Buchanan). Appl. Environ. 41. Wolf, K. 1988. Spring viremia of carp, p. 191-216. In Fish viruses Microbiol. 34:50-55. and fish viral diseases. Cornell University Press, Ithaca, N.Y. 37. Streisinger, G., C. Walker, N. Dower, D. Knauber, and F. 42. Wolf, K. and M. C. Quimby. 1976. Procedures for subculturing Singer. 1981. Production of clones of homozygous diploid zebra fish cell and propagating fish cell lines, p. 471-474. In V. J. Evans, fish (Brachydanio rerio). Nature 291:293-296. V. P. Perry, and M. M. Vincent (ed.), TCA manual, vol. 2. Tissue 38. Van Regenmortel, M. H. V., C. M. Fauquet, and D. H. L. Bishop. Culture Association, Rockville, Md. 2000. The viruses, p. 563-569. In Virus taxonomy classification and nomenclature of viruses. Academic Press, San Diego, Calif.

521