Viral Diseases in Zebrafish: What Is Known and Unknown

Abstract

Naturally occurring viral infections have the potential to introduce confounding variability that leads to invalid and misinterpreted data. Whereas the viral diseases of research rodents are well characterized and closely monitored, no naturally occurring viral infections have been characterized for the laboratory zebrafish (Danio rerio), an increasingly important biomedical research model. Despite the ignorance about naturally occurring zebrafish viruses, zebrafish models are rapidly expanding in areas of biomedical research where the confounding effects of unknown infectious agents present a serious concern. In addition, many zebrafish research colonies remain linked to the ornamental (pet) zebrafish trade, which can contribute to the introduction of new pathogens into research colonies, whereas mice used for research are purpose bred, with no introduction of new mice from the pet industry. Identification, characterization, and monitoring of naturally occurring viruses in zebrafish are crucial to the improvement of zebrafish health, the reduction of unwanted variability, and the continued development of the zebrafish as a model organism. This article addresses the importance of identifying and characterizing the viral diseases of zebrafish as the scope of zebrafish models expands into new research areas and also briefly addresses zebrafish susceptibility to experimental viral infection and the utility of the zebrafish as an infection and immunology model.

Introduction

Animal models for biomedical research “evolve” overtime as species initially obtained from a nonlaboratory source, such as agriculture, wildlife, or the pet trade, are adapted to the laboratory, experimental tools and methodologies are developed, improvements are made in husbandry and biosecurity, and, finally, laboratory breeding colonies are developed to maintain pathogen-free, purpose-bred research animals. For example, the best-developed and most widely used species in animal-based laboratory research is the mouse, Mus musculus (Wade and Daly 2005). The development of the mouse as a laboratory model began in the 1910s and 1920s as American mouse fanciers provided a ready source of multiple lineages to researchers searching for a genetic basis for cancer (Wade and Daly 2005; Wade et al. 2002). As mouse-based research expanded into other health-related fields, outbreaks of infectious disease confounded some research studies. To counter the deleterious effects of disease outbreaks, research investigations were initiated to identify mouse pathogens. Since that time, a large number of naturally occurring viruses and other pathogens of mice have been described, and discoveries about their transmission and pathogenesis have contributed to the understanding of infectious disease. Understanding the transmission and pathogenesis of murine diseases facilitated the elimination of many pathogens from research mice, directly improving animal health as well as the utility of the mouse model. Husbandry, biosecurity, and health monitoring have since continued to improve in the rodent world, reducing variability and the number of animals necessary for individual experiments.

It is well documented and widely accepted by the biomedical research community that naturally occurring viral infections can introduce significant confounding variability, which ultimately results in invalid and misinterpreted data as well as increased animal numbers. Laboratory mice are housed in vivariums with conditions that are carefully controlled and limit pathogen transmission. Mice used for research are also purpose bred, with no introduction of new mice from the pet industry. In contrast, the development of the zebrafish (Danio rerio) model for many types of biomedical research is in its infancy. Standard controls used to prevent pathogen introduction and transmission among rodents—such as approved vendor lists, pathogen exclusion lists, and health monitoring programs—are not widely practiced in the management of zebrafish colonies (Lawrence et al. 2012). The viral infections of laboratory rodents have been extensively studied, yet virtually nothing is known about naturally occurring viruses of laboratory zebrafish. The growing importance and expansion of zebrafish models in biomedical research necessitate improvements in the standards of husbandry and biosecurity in laboratory zebrafish colonies. Improved biosecurity of zebrafish colonies requires a better understanding of natural occurring zebrafish pathogens, including viruses, their diagnosis, and how they are transmitted.

Information about the confounding effects of naturally occurring pathogens in zebrafish used for biomedical research lags far behind the information available for both aquaculture fishes and mammalian laboratory species. In fact, few naturally occurring zebrafish pathogens of any kind have been well characterized. Most of the pathogens that are known represent bacterial, fungal, or parasitic agents that were previously identified in commercially important fish species and later recognized in zebrafish. The lack of information about naturally occurring viral infections in zebrafish reflects a lack of investigation in this area rather than an inability of viral pathogens to infect zebrafish (Kent et al. 2009), as evidenced by studies documenting the experimental infection of zebrafish with viruses isolated from other fish species (LaPatra et al. 2000; Lopez-Munoz et al. 2010; Lu et al. 2008; Ludwig et al. 2011; Novoa et al. 2006; Phelan et al. 2005b; Sanders et al. 2003; Seeley et al. 1977; Xu et al. 2008) and the presence of multiple endogenous retroviruses, retrotransposons, and other retroid agents in the zebrafish genome (Basta et al. 2007; Shen and Steiner 2004).

Many aquatic viruses are not host-specific, possibly reflecting evolutionary access to a wider range of potential hosts when virions are distributed in the water column. For example, viral hemorrhagic septicemia virus (VHSV¹) infects dozens of species of both marine and freshwater fish, and some aquatic viruses can naturally infect both fishes and amphibians or both fishes and aquatic invertebrates. Because many aquatic viruses can infect multiple species, zebrafish are likely to be susceptible to viral pathogens that have already been identified in other fish species. The number and the diversity of viruses that have been described in other teleost fishes are quite large (Table 1). Research efforts to identify naturally occurring viruses in zebrafish should therefore include a search for novel viruses and a search for viruses previously isolated from other teleost fishes, especially tropical cyprinid species.

Table 1.

Viruses described in teleost fishes

Viruses	References
DNA viruses
Adenoviridae	Benko et al. 2002; Harrach and Benko 2007
Circoviridae	Lorincz et al. 2011
Herpesviridae	Choi et al. 2004; Graham et al. 2004; Jeffery et al. 2007; Shlapobersky et al. 2010
Iridoviridae	Go et al. 2006; Kitamura et al. 2006; Tsai et al. 2005; Weber et al. 2009; Whittington et al. 2010
Polyomaviridae	Essbauer and Ahne 2001
Poxviridae	Nylund et al. 2008b
RNA viruses
Birnaviridae	Zhao et al. 2008
Calciviridae	Smith et al. 1980a,b
Coronaviridae	Miyazaki et al. 2000; Schutze et al. 2006
Hepeviridae	Batts et al. 2011
Nodaviridae	Bigarre et al. 2009; Gomez et al. 2008; Hegde et al. 2003; Montes et al. 2010; Moody et al. 2009; Nylund et al. 2008a
Orthomyxoviridae	Falk et al. 1997; Mjaaland et al. 1997
Paramyxoviridae	Falk et al. 2008
Picornaviridae	Berthiaume et al. 1982; Robin and Dery 1982; Robin and Lariviere-Durand 1983
Reoviridae	Goodwin et al. 2010; LaPatra et al. 1995; Mohd Jaafar et al. 2008
Retroviridae	Francis-Floyd et al. 1993
Rhabdoviridae	Gagne et al. 2007; Tao et al. 2008; Teng et al. 2007

This article addresses the importance of identifying and characterizing the naturally occurring viral infections of zebrafish as the scope of zebrafish models expands into new research areas, and it briefly discusses zebrafish susceptibility to experimental viral infection and the characteristics of zebrafish models that enable the study of host factors and viral immunity.

Importance of Identifying Naturally Occurring Viruses in Zebrafish

Historically, the zebrafish has served as a model for vertebrate development and genetics, with virtually all experimentation occurring during the first few days postfertilization; thus, adult zebrafish have been traditionally maintained exclusively as breeding stock to produce embryos for experimentation. The lack of information about viral infections in zebrafish can be partially explained by the fact that, in the absence of an epizootic event, chronic morbidity and mortality in a zebrafish colony generally have not prevented researchers from collecting a sufficient number of zebrafish embryos to conduct their studies. Consequently, many researchers are accustomed to accepting a level of morbidity and mortality. Although the zebrafish research community has not historically exhibited the level of concern for eliminating infectious diseases that is now common in the rodent research community, the advantages of zebrafish as a model organism have resulted in its recent expansion in areas of biomedical research where the confounding effects due to unknown infectious agents are a serious concern. Research areas such as aging, cancer, immunity, infection, and toxicology often require that the Zebrafish be maintained for a much greater portion of their life span and that the histopathologic changes in adult animals be assessed. Increased mortality, underlying chronic inflammation, altered cytokine levels, tissue damage, and tissue repair resulting from natural infections are likely to be important confounding variables in these types of studies.

Although no cases of naturally occurring viral infections have been reported in zebrafish research facilities, such infections probably occur, considering what is known about other fishes held in captivity. If viral infections occur, they may seriously affect many types of research. The task of identifying and characterizing naturally occurring viral infections in zebrafish is thus critically important. The potential cost of undiagnosed viral infections may be increasing as the nature of zebrafish research changes and as zebrafish facilities become more centralized. Although some viral infections in any host species may produce epizootic events, high mortality, or noticeable clinical signs, many viral infections cause only subclinical or low-grade infections. Moreover, the negative effects of subclinical infections on research are rarely reported (see Kent et al. 2012, in this issue). Even unnoticed viral infections may alter the immune system and confound research (Baker 1998). The risk of confounding effects is also related to the type of experimentation. Because zebrafish models have expanded from developmental biology and genetics to include models for toxicology, aging, cancer, infection, and immunology, many kinds of research may be seriously and adversely affected by unidentified underlying viral infections.

Because no naturally occurring viral infections have been reported in zebrafish, it is not possible to provide specific examples of how naturally occurring viral infections have confounded zebrafish research. However, examples demonstrating the significant effects of viral infections on subsequent infections in other fish species illustrate the potential confounding effects of unrecognized viral infections on the use of zebrafish as an infection model. For example, multiple instances of viral infections nonspecifically conferring either increased susceptibility or increased resistance to a subsequent infection with other pathogens have been reported. When Olive flounder (Paralichthys olivaceous) were infected with flounder birnavirus and then subsequently infected with Edwardsiella tarda, Streptococcus iniae, or VHSV, the flounder birnavirus–infected flounder were more resistant to VHSV infection than controls but less resistant to bacterial infections (Pakingking et al. 2003). Later experiments showed that flounder birnavirus–infected flounder exhibited significant protection from VHSV when they were exposed 3, 7, or 14 days following flounder birnavirus exposure, but there was no significant mortality difference from controls when the VHSV exposure occurred 21 days after flounder birnavirus infection (Pakingking 2004). Similarly, flounder birnavirus infection conferred complete protection in sevenband grouper (Epinephelus septemfasciatus) to subsequent infection with red-spotted grouper nervous necrosis virus, whereas 80% mortality was observed in sevenband grouper not exposed to flounder birnavirus (Pakingking et al. 2005).

Regardless of whether clinical signs are evident, underlying infection with an unknown viral agent may confound experimental infection studies with a different pathogen. Flounder birnavirus does not normally cause disease in infected flounder, but it profoundly influences the outcome of infection studies on other pathogens. This set of studies illustrates the type of confounding effects that might be expected when infection studies are carried out in zebrafish that have been unknowingly exposed to subclinical viral infections. Similarly, it is well documented in salmonid fishes that recent or simultaneous viral infection can dramatically alter the outcomes of other infections. If these viral infections had been unrecognized, the vastly different experimental outcomes would have been difficult to interpret and may have erroneously been attributed to other factors. For example, Atlantic salmon (Salmo salar) acutely infected with infectious pancreatic necrosis virus (IPNV¹) and infectious salmon anemia virus display significantly lower mortality than salmon infected with infectious salmon anemia virus only (Johansen and Sommer 2001), but Atlantic salmon coinfected with IPNV and either of the bacterial pathogens Vibrio salmonicida or Aeromonas salmonicida exhibited significantly higher mortality than salmon infected with V. salmonicida or A. salmonicida alone (Johansen and Sommer 2001; Johansen et al. 2009). Rainbow trout coinfected with IPNV and infectious hematopoietic necrosis virus (IHNV¹) displayed 50% less mortality than trout infected with either virus individually (Alonso et al. 2003). In a similar study, coinfected trout displayed a reduced tissue distribution of IHNV among the internal organs compared with trout infected with IHNV alone (Brudeseth et al. 2002). Notably, rainbow trout infected with IHNV 14 days after infection with IPNV exhibited only 2% mortality, whereas trout infected with IHNV alone exhibited 72% mortality (Byrne et al. 2008). Finally, Hedrick and colleagues (1994) showed that preexposure to Cutthroat trout virus protected rainbow trout from subsequent IHNV challenge for up to 4 weeks. If extrapolating from salmonids to zebrafish, unrecognized viral infections in zebrafish could be anticipated to interfere with subsequent experimental infections. It therefore seems prudent to search for naturally occurring viruses in zebrafish because the zebrafish is an important infection model for aquaculture and biomedical research and uncovering any underlying viral infections is critical to a correct interpretation of experimental infections.

The potential effects of unrecognized viral infections may, in some cases, be similar to the confounding effects documented for parasitic and bacterial infections in zebrafish and other laboratory fishes. Mycobacterium spp. are the most important bacterial pathogens of laboratory zebrafish and cause a wide array of chronic inflammatory lesions that complicate histopathologic interpretation (see Kent et al. 2012, in this issue; Sanders et al. 2012, in this issue). Similarly, viral infections may confound toxicologic studies or other studies that rely on histopathology by introducing unexplained histopathologic changes. Underlying parasitic infections have also been shown to complicate the interpretation of histopathologic changes in fish toxicologic studies. Parasitic infections make fish more susceptible to the toxic effects of zinc (Boyce and Yamada 1977), cadmium chloride (Pascoe and Cram 1977), and petroleum hydrocarbons (Moles 1980). Similarly, detrimental effects of polychlorinated biphenyls on anterior kidney leukocytes were more severe in parasitized juvenile salmon (Jacobson et al. 2003). Underlying infections may also alter immune function; for example, chronic subclinical infections contribute to stress, which can subsequently contribute to immunosuppression. When experimentally parasitized zebrafish were subjected to other stressors, they suffered increased mortality, earlier onset of infection, reduced weight, and reduced fecundity compared with unparasitized zebrafish (Ramsay et al. 2009). Subclinical infections probably also alter the type or amount of inflammation. For example, increased inflammation and altered cytokine production caused by underlying infections can affect cell proliferation rates, creating a confounding variable in tumorigenesis models. In fact, the incidence of intestinal neoplasia following treatment with dimethylbenzathracine was significantly higher in zebrafish infected by the intestinal nematode Pseudocapillaria tomentosa than in similarly treated uninfected zebrafish (Kent et al. 2002).

In recent years, researchers have developed immunocompromised zebrafish to aid in the study of hematopoiesis, tumorigenesis, infection, and immunity. As with other species, immunocompromised zebrafish are likely to be more susceptible to viral infections than wild-type fish, exhibiting higher mortality, higher morbidity, more clinical signs, and more severe histopathologic lesions. Thus, in experimental models where immunocompromised zebrafish are required, the use of virus-free fish will play a major role in reducing the variability in data because of confounding factors, such as inflammation and other host responses to infection, and will simultaneously reduce the number of fish required to achieve adequate statistical power.

Naturally occurring infections along with oncogenic viruses may also play a role in confounding experiments, including cancer studies using zebrafish (Kent et al. 2009), because viruses are associated with tumorigenesis in several other fish species (Bowser et al. 2005; Francis-Floyd et al. 1993). Spontaneously occurring neoplastic lesions are relatively common in laboratory zebrafish. Most commonly reported are spermacytic seminomas affecting the testes, spindle cell sarcomas (malignant nerve sheath tumors), ultimobranchial gland tumors, and gastrointestinal tumors, which may have a neoplastic or preneoplastic prevalence of greater than 30% in some zebrafish facilities (M.L. Kent, personal communication). These gastrointestinal tumors display no sex predilection, occur in multiple background strains, and appear to be confined to recirculating systems. To date, no dietary or waterborne carcinogens have been identified, suggesting the possibility of an infectious component. Florida strain wild-type zebrafish treated with N-nitroso-N-ethylurea in one study exhibited 100% incidence of cutaneous papillomas (Beckwith et al. 2000); however, no cutaneous papillomas were observed in similar experiments conducted at several other research institutions, suggesting the possibility of an unrecognized oncogenic virus (Kent et al. 2009). Thus, zebrafish are subject to neoplasia, which varies by population, suggesting that zebrafish may be hosts to unknown oncogenic viruses. Oncogenesis due to unrecognized viruses may therefore be a confounding variable in cancer research using zebrafish. A detailed review of neoplasia in zebrafish is included in this issue (Spitsbergen et al. 2012).

The importance of pathogen-free animals for experimentation is widely accepted for mammalian laboratory models. The only current effort to generate pathogen-free zebrafish is being conducted at the Sinnhuber Aquatic Resources Laboratory at Oregon State University. To date, this laboratory has eliminated one pathogen, the microsporidian Pseudoloma neurophilia (Kent et al. 2011). For development of pathogen-free zebrafish to be successful, however, naturally occurring viral pathogens affecting zebrafish colonies must be identified and diagnostic assays to detect viral infection must be developed to facilitate the elimination of viral pathogens.

Importance of Biosecurity for Laboratory Zebrafish

Because naturally occurring viral infections can confound research and therefore pose risks to zebrafish facilities, it is important to prevent the introduction of new viruses into naive fish colonies and to identify and manage existing pathogens within a colony. Biosecurity of laboratory zebrafish is affected not only by the current lack of knowledge regarding naturally occurring zebrafish viruses but also by the introduction of zebrafish raised by the pet and aquarium trade into laboratory zebrafish facilities. In contrast to purpose-bred research mice, ornamental zebrafish are often raised together with other species that harbor numerous pathogens. A recent example of pathogen transmission from the pet trade to laboratory zebrafish is neon tetra disease, caused by the microsporidian Pleistophora hyphessobryconis, which was detected in three unrelated laboratory zebrafish facilities (Sanders et al. 2010). These parasitic outbreaks were directly linked to zebrafish purchased from the pet trade (Sanders et al. 2010). In addition, the diagnostic pathology service at the Zebrafish International Resource Center occasionally detects common aquarium fish helminth parasites, indicating that the submitted zebrafish were raised in outdoor ponds (M.L. Kent, personal communication). The often unrestricted relationship between the pet trade and research colonies undoubtedly facilitates the introduction of viral pathogens, as well as parasites and bacteria, into laboratory zebrafish colonies. Although knowledge of the viruses that infect the aquarium fish species that are raised with zebrafish in the pet trade is limited, the viruses that have been described include herpesviruses (Gilad et al. 2002; Jeffery et al. 2007), iridoviruses (Go et al. 2006; Hossain et al. 2008; Jeong et al. 2008a,b), a nodavirus (Hegde et al. 2003), and a reovirus (Seng et al. 2004). This research not only demonstrates that laboratory zebrafish may still harbor a variety of pathogens from their origin in the pet trade but also provides evidence that zebrafish raised in outdoor ponds continue to directly enter laboratory colonies, providing a constant source of new pathogens from other pet and aquarium species. It is, therefore, also likely that at least some laboratory zebrafish are currently infected with viruses prevalent in the ornamental fish trade.

Moreover, many newer zebrafish facilities are centralized; therefore, an epizootic event caused by breaches in biosecurity has the potential to be devastating by exposing the zebrafish colonies of multiple investigators to pathogens during an outbreak of disease (Kent et al. 2009). The risk of viral spread between zebrafish populations within a centralized facility is increased by the close proximity of aquatic systems, increased traffic, and overlap in personnel and other resources.

Zebrafish as a Viral Infection and Host Defense Model

The zebrafish is a useful and attractive model for infectious disease and immunity research and is considered a refinement over the use of mammalian infection models (Burgos et al. 2008). The zebrafish model boasts the capacity to allow investigation of specific immune system components at various stages of immunologic development, and extensive molecular, genetic, and imaging tools are available for this species. The utility of the zebrafish model for infection and immunity experiments has been extensively reviewed (Meeker and Trede 2008; Phelps and Neely 2005; Sullivan and Kim 2008; Trede et al. 2004; van der Sar et al. 2004; Yoder et al. 2002) and hinges on the functional similarity of the zebrafish immune system and the mammalian immune system.

Briefly, zebrafish exhibit both the innate and adaptive arms of the immune system, including leukocyte populations, inflammatory mediators, and signaling molecules that are similar to those of the mammalian immune system. Zebrafish have B and T lymphocytes, monocytes, and phagocytic tissue macrophages that are similar to their mammalian counterparts, as well as at least two granulocyte lineages (Phelps and Neely 2005). Of these granulocyte lineages, zebrafish neutrophils exhibit segmented nuclei and cytoplasmic granules that are myeloperoxidase positive, whereas the second granulocyte lineage displays characteristics of both eosinophils and basophils (Bennett et al. 2001). Zebrafish B and T lymphocytes have many similarities to their mammalian counterparts; both rag1 and rag2 have been identified in zebrafish (Willett et al. 1997a,b), and both T-cell receptor genes and B-cell immunoglobulin genes exhibit V(D)J recombination (Haire et al. 2000). Whereas mammalian adaptive immune systems include five immunoglobulin (Ig) classes—IgA, IgD, IgE, IgG, and IgM—the zebrafish exhibits three known immunoglobulin classes—IgD, IgM, and IgZ—as well as a second IgZ-like immunoglobulin, IgZ-2 (Danilova et al. 2005; Hu et al. 2010b).

The innate arm of the zebrafish immune system likewise bears considerable resemblance to the mammalian system. For example, 24 putative toll-like receptor (TLR) variants have been identified in zebrafish, including at least one homologue for each of the following mammalian TLRs: TLR1, TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, and TLR9, in addition to fish-specific TLRs (Jault et al. 2004; Meijer et al. 2004). Gene homologues encoding adaptor proteins important for signal transduction also have been identified in zebrafish, including toll-interleukin 1 receptor domain–containing adaptor protein (TIRAP), myeloid differentiation primary response gene 88 (MyD88), sterile alpha and Armadillo motif–containing protein (SARM), and toll-interleukin 1 receptor–containing adapter molecule (TICAM) (Jault et al. 2004; Meijer et al. 2004). Homologues have also been identified for IRAK-4 and TRAF6, key proteins in the TLR signaling pathway (Phelan et al. 2005a; Sullivan and Kim 2008). When ZFL cells were infected with snakehead rhabdovirus (SHRV¹), the zebrafish homologues for TLR3 and TRAF6 were upregulated, demonstrating conserved TLR signaling in zebrafish (Phelan et al. 2005a).

Many components of the complement system have been identified in zebrafish, including B (Gongora et al. 1998), C1q (Hu et al. 2010a), C1s (Nakao et al. 2011), C3 (Samonte et al. 2002; Vo et al. 2009a,b), C4 (Samonte et al. 2002), C6, C7, C8a, C8b, C8y, and C9 (Nakao et al. 2011), H (Sun et al. 2010), I, mannose-binding lectin–associated serine protease (MASP)-1, MASP-2, MASP-3, and mannose-binding lectin (Nakao et al. 2011). The classical pathway, alternative pathway, and lectin pathways all activate complements in teleost fishes (Holland and Lambris 2002; Nakao et al. 2011; Sunyer and Lambris 1998). Similarly, homologues to many mammalian cytokines have been identified in zebrafish, including interleukin 1β (Pressley et al. 2005), type I (Altmann et al. 2003) and type II (Igawa et al. 2006) interferons, tumor necrosis factor α (Praveen et al. 2006; Pressley et al. 2005), and several interleukins (Sullivan and Kim 2008).

Moreover, several innate immune system components with antiviral properties have been identified in zebrafish. For example, as in the mammalian antiviral response, zebrafish type I interferon directly induces myxovirus resistance (Mx) expression (Haller et al. 2007). Mx proteins are a group of high molecular weight dynamin-like proteins whose antiviral properties were first recognized when inbred mouse strains with mutations at the Mx locus were found to be susceptible to influenza A virus, an orthomyxovirus (Haller et al. 1979; Lindenmann 1962, 1964; Workenhe et al. 2010). Since then, Mx proteins have been shown to have antiviral effects against several viral families, including Bunyaviridae, Orthomyxoviridae, Paramyxoviridae, Rhabdoviridae, and Togaviridae. Because the Mx gene is not normally expressed in the absence of viral infection, Mx expression is commonly used as a marker for type I interferon activity (Haller et al. 2007) and can be used as evidence of viral infection in the zebrafish, as demonstrated by the upregulation of interferon and Mx RNA in zebrafish experimentally infected with SHRV (Phelan et al. 2005b).

Thus, the similarity between the zebrafish immune system and the mammalian immune system, the genetic and molecular tools available for zebrafish, and the ability to image an entire infected embryo or fish over the course of infection make the zebrafish an excellent model to investigate viral pathogenesis and host defenses. Investigating natural infection with zebrafish viruses in addition to experimental infections will improve the usefulness of the zebrafish as a model organism, not only elucidating routes of infection, virulence factors, host defenses, and viral countermeasures to host defenses but also providing information about potential confounding effects, modes of transmission, and necessary biosecurity measures for improved operation of bioaquatics facilities.

Experimental Viral Infections in Zebrafish

Infection studies designed to develop the zebrafish as a model for viral infection in commercially important fish species have demonstrated the experimental susceptibility of zebrafish to infection by several families of viruses, including Birnaviridae (IPNV) (LaPatra et al. 2000; Seeley et al. 1977); Rhabdoviridae (IHNV) (LaPatra et al. 2000; Ludwig et al. 2011), Spring viremia of carp virus (Lopez-Munoz et al. 2010; Sanders et al. 2003), VHSV (Novoa et al. 2006) and SHRV (Phelan et al. 2005b); Nodaviridae (Malabar grouper nervous necrosis virus) (Lu et al. 2008); and Iridoviridae (infectious spleen and kidney necrosis virus) (Xu et al. 2008). The susceptibility of zebrafish to experimental infections with this broad range of viruses suggests not only that naturally occurring viruses occur in zebrafish but also that many viral families may be represented among the as-yet unidentified zebrafish viruses.

Zebrafish have also been used as models of mammalian viral infection, although there are some limitations to their use. For example, zebrafish are often maintained at approximately 28°C, whereas some mammalian viruses adapted to replicate at 37°C may not be pathogenic at 28°C. However, advantages to zebrafish models for mammalian viral infections include the capacity for live imaging, whole-organism histopathology and immunohistochemistry, and temperature-shift experiments. Because zebrafish larvae are small and transparent, every tissue in the zebrafish can be observed while still functioning in the live animal (Ludwig et al. 2011). To accomplish a similar objective in mice would require mice to be sacrificed at multiple time points with multiple harvests of selected tissues in an attempt to piece together the overall picture. Furthermore, as poikilotherms, zebrafish can survive over a range of temperatures, allowing some viral infections to be “halted” by shifting the temperature so that viral replication does not continue, as was recently accomplished with IHNV (Ludwig et al. 2011). By shifting the temperature of live infected embryos from 24°C to 28°C, the infectious process was halted at various points, facilitating the characterization of the course of infection (Ludwig et al. 2011).

The first zebrafish infection studies using a mammalian virus demonstrated dose-dependent infection of the adult zebrafish nervous system with herpes simplex virus type 1 (Herpesviridae) (Burgos et al. 2008; Hubbard et al. 2010). Zebrafish treated with the antiviral acyclovir exhibited significantly reduced viral load in all examined regions (head, dorsal midbody, and ventral midbody), whereas zebrafish treated with cyclophosphamide exhibited significantly higher mortality and increased viral loads (Burgos et al. 2008).

In summary, zebrafish are experimentally susceptible to a variety of viruses described in other fish species. The viruses include the following: IPNV, IHNV, Malabar grouper nervous necrosis virus, Spring viremia of carp virus, SHRV, VHSV, infectious spleen and kidney necrosis virus, and herpes simplex virus type 1. These experimental infections document zebrafish susceptibility to five viral families: Birnaviridae, Rhabdoviridae, Nodaviridae, Iridoviridae, and Herpesviridae. These experiments provide evidence that laboratory zebrafish are infected with naturally occurring viruses.

Conclusion

The similarity between the zebrafish and mammalian immune systems, the capacity of the zebrafish model to allow investigation of specific immune system components at different stages of immunologic development, and the molecular, genetic, and imaging tools available for this species make the zebrafish particularly useful and attractive as a model for infectious disease and immunity research. However, zebrafish studies of infection and immunity, as well as other types of zebrafish research, are at risk of being confounded by unrecognized naturally occurring viral infections. To date, what is known about viral disease in zebrafish is entirely the result of experimental infections using viruses isolated from other species; no naturally occurring viral infections have been reported in the zebrafish. However, the lack of information regarding naturally occurring viral infections in zebrafish does not reflect an inability of viral pathogens to infect zebrafish because zebrafish are easily experimentally infected with a variety of viruses isolated from other fishes. It is imperative that naturally occurring viruses of zebrafish be identified and characterized so that sensitive diagnostic tests can be designed and adequate health monitoring can be implemented. Furthermore, continued research to elucidate the specific pathogenesis and transmission of each virus will be necessary to determine which pathogens are of concern for various areas of research, and such research will aid in the design of biosecurity protocols. This process is crucial to improvement of laboratory zebrafish health, reduction of unwanted variability, and continued development of the zebrafish as a model organism. Moreover, the connection between laboratory zebrafish and the ornamental fish trade must be severed so that viruses and other pathogens are not easily introduced into research colonies. Zebrafish facilities should exclusively use independent sources of purpose-bred laboratory zebrafish, such as the Zebrafish International Resource Center and the Sinnhuber Aquatic Resources Laboratory. The molecular, genetic, and imaging tools available for the zebrafish have developed much more quickly than background knowledge of viral diseases. The best time to address the question of underlying viral disease in zebrafish is now, to ensure the maximum return on the efforts and funds invested in this worthwhile animal model.