Cyprinid Herpesvirus 3

TOC summary: This virus is useful for fundamental and applied research.

T he common carp (Cyprinus carpio carpio) is a freshwater fi sh and one of the most economically valuable species in aquaculture; worldwide, 2.9 million metric tons are produced each year (1). Common carp are usually cultivated for human consumption. Koi (C. carpio koi) are an often-colorful subspecies of carp, usually grown for personal pleasure and competitive exhibitions. In the late 1990s, a highly contagious and virulent disease began to cause severe economic losses in these 2 carp industries worldwide (2) (Figure 1). The rapid spread was attributed to international fi sh trade and koi shows around the world (3). The causative agent of the disease was initially called koi herpesvirus because of its morphologic resemblance to viruses of the order Herpesvirales (3). The virus was subsequently called carp interstitial nephritis and gill necrosis virus because of the associated lesions (4). Recently, on the basis of homology of its genome with previously de-scribed cyprinid herpesviruses (5), the virus was assigned to family Alloherpesviridae, genus Cyprinivirus, species Cyprinid herpesvirus 3 and renamed cyprinid herpesvirus 3 (CyHV-3). Because of the economic losses caused by this virus, CyHV-3 rapidly became a subject for applied research. However, recent studies have demonstrated that CyHV-3 is also useful for fundamental research. We therefore summarized recent advances in CyHV-3 applied and fundamental research.

Characterization of CyHV-3 Classifi cation
CyHV-3 is a member of the order Herpesvirales and newly designated family Alloherpesviridae (5,6) ( Figure 2, panel A). Alloherpesviridae viruses infect fi sh and amphibians. The common ancestor of this family is thought to have diverged from the common ancestor of the family Herpesviridae (herpesviruses that infect reptiles, birds, and mammals) (6). According to phylogenetic analysis of specifi c genes, the family Alloherpesviridae seems to be subdivided into 2 clades (6) (Figure 2, panel B). The fi rst clade comprises anguillid and cyprinid herpesviruses, which possess the largest genomes in the order Herpesvirales (245-295 kb). The second clade comprises ictalurid, salmonid, acipenserid, and ranid herpesviruses, which have smaller DNA genomes (134-235 kb).

Structure
The CyHV-3 structure is typical of viruses of the order Herpesvirales. An icosahedral capsid contains the genome, which consists of a single, linear, double-stranded DNA molecule. The capsid is covered by a proteinaceous matrix called the tegument, which is surrounded by a lipid envelope derived from host cell trans-golgi membrane (7) (Figure 3). The envelope contains viral glycoproteins (3). The diameter of the entire CyHV-3 particle is 170-200 nm (3,8).

Genome
The genome of CyHV-3 is a 295-kb, linear, doublestranded DNA molecule consisting of a large central portion fl anked by two 22-kb repeat regions, called the left and right repeats (9). The genome size is similar to that of CyHV-1 but larger than that of other members of the order Herpesvirales, which are generally 125-240 kb.
The CyHV-3 genome encodes 156 potential proteincoding open reading frames (ORFs), including 8 ORFs encoded by the repeat regions. These 8 ORFs are consequently present as 2 copies in the genome (9). Five families of related genes have been described: ORF2, tumor necrosis factor receptor, ORF22, ORF25, and RING families. The ORF25 family consists of 6 ORFs (ORF25, ORF26, ORF27, ORF65, ORF148, and ORF149) encoding related, potential membrane glycoproteins. The expression products of 4 of the sequences were detected in mature virions (ORF25, ORF65, ORF148, and ORF149) (10 encodes several genes that could be involved in immune evasion processes, such as ORF16, which codes for a potential G-protein coupled receptor; ORF134, which codes for an IL-10 homolog; and ORF12, which codes for a tumor necrosis factor receptor homolog. Within the family Alloherpesviridae, anguillid herpesvirus 1 is the closest relative of CyHV-3 that has been sequenced (11). Each of these viruses possesses 40 ORFs exhibiting similarity. Sequencing of CyHV-1 and CyHV-2 will probably identify more CyHV-3 gene homologs. The putative products of most ORFs in the CyHV-3 genome lack obvious relatives in other organisms; 110 ORFs fall into this class. Six ORFs encode proteins with closest relatives in virus families such as Poxviridae and Iridoviridae (9). For example, CyHV-3 genes such as B22R (ORF139), thymidylate kinase (ORF140), thymidine kinase (ORF55), and subunits of ribonucleotide reductase (ORF23 and ORF141) appear to have evolved from poxvirus genes (9). Neither thymidylate kinase nor B22R has been identifi ed previously in a member of the order Herpesvirales.
Three unrelated strains of CyHV-3, isolated in Israel (CyHV-3 I), Japan (CyHV-3 J), and the United States (Cy-HV-3 U), have been fully sequenced (9). Despite their distant geographic origins, these strains exhibit high sequence identity. Low diversity of sequences among strains seems to be a characteristic of the CyHV-3 species. Despite this low diversity, molecular markers enabling discrimination among 9 genotypes (7 from Europe and 2 from Asia) have been identifi ed (12).
Because CyHV-3 possesses the largest genome among members of the order Herpesvirales, it provides a model for mutagenesis of large DNA viruses. Recently, the Cy-HV-3 genome was cloned as a stable and infectious bacterial artifi cial chromosome, which could be used to produce CyHV-3 recombinants (13).
The CyHV-3 replication cycle was recently studied by use of electron microscopy (7). Its morphologic stages suggested that it replicates in a manner similar to that of members of the family Herpesviridae. Capsids leave the nucleus by budding at the inner nuclear membrane, resulting in formation of primary enveloped virions in the perinuclear space. The primary envelope then fuses with the outer leafl et of the nuclear membrane, thereby releasing nucleocapsids into the cytoplasm. Final envelopment occurs by budding into trans-golgi vesicles. Because CyHV-3 glycoproteins have little or no similarity with those of members of the family Herpesviridae, identifi cation of the CyHV-3 glycoproteins involved in entry and egress will require further study.  Because fi sh are poikilotherms and because CyHV-3 only affects fi sh when the water temperature is 18°C-28°C, the effect of temperature on CyHV-3 replication growth in vitro has been investigated. Replication in cell culture is restricted by temperature; optimal viral growth is at 15°C-25°C. Virus propagation and virus gene transcription are turned off when cells are moved to a nonpermissive temperature of 30°C (18). Despite the absence of detectable virus replication, infected cells maintained for 30 days at 30°C preserve infectious virus, as demonstrated by viral replication when the cells are returned to permissive temperatures (18) (Figure 4). These results suggest that Cy-HV-3 can persist asymptomatically for long periods in the fi sh body when the temperature prevents virus replication; bursts of new infection occur after exposure to permissive temperatures.

Disease Caused by CyHV-3 History
In 1998, the fi rst mass deaths of common and koi carp were reported in Israel and the United States (3). However, analyses of samples from archives determined that the virus had been in wild common carp since 1996 in the United Kingdom (19). Soon after the fi rst report, outbreaks of CyHV-3 were identifi ed in countries in Europe, Asia, and Africa. Currently, CyHV-3 has been identifi ed everywhere in the world except South America, Australia, and northern Africa (20). Worldwide, CyHV-3 has caused severe fi nancial and economic losses in the koi and common carp culture industries.

Host Range
Common and koi carp are the only species known to be affected by CyHV-3 infection (21). Numerous fi sh species, cyprinid and noncyprinid, were tested for their ability to carry CyHV-3 asymptomatically and to spread it to unexposed carp (21-23) ( Table 2). CyHV-3 DNA was recovered from only 2 other fi sh species: goldfi sh and crucian carp. Cohabitation experiments suggest that goldfi sh, grass carp, and tench can carry CyHV-3 asymptomatically and spread it to unexposed common carp. Hybrids (koi-goldfi sh and koi-crucian carp) die of CyHV-3 infection (24).

Pathogenesis
Several researchers have postulated that the gills might be the portal of entry for CyHV-3 (17,(26)(27)(28); however, this hypothesis was recently refuted (29). Bioluminescent imaging and an original system for performing percutaneous infection restricted to the posterior part of the fi sh showed that the skin covering the fi n and body mediated entry of CyHV-3 into carp (29) ( Figure 5). This study, together with an earlier study of the portal of entry of a rhabdovirus (infectious hematopoietic necrosis virus) in salmonids (30), suggests that the skin of teleost fi sh represents an effi cient portal of entry for certain viruses. The skin of teleost fi sh is a stratifi ed squamous epithelium that, unlike its mammalian counterpart, is living and capable of mitotic division at all levels, even the outermost squamous layer. The scales are dermal structures. More extensive studies are needed to demonstrate that the skin is the only portal of entry of CyHV-3 into carp.
After initial replication in the epidermis (29), the virus is postulated to spread rapidly in infected fi sh, as indicated by detection of CyHV-3 DNA in fi sh tissues (27). As early as 24 hours postinfection, CyHV-3 DNA was recovered from almost all internal tissues (including liver, kidney, gut, spleen, and brain) (27), where viral replication occurs at later stages of infection and causes lesions. One hypothesis regarding the rapid and systemic dissemination indicated by PCR is that CyHV-3 secondarily infects blood cells. Virus replication in organs such as the gills, skin, and gut at the later stages of infection represents sources of viral excretion into the environment. After natural infection under permissive temperatures (18°C-28°C), the highest mortality rates occur 8-12 days postinfection (dpi) (21). Gilad et al. suggest that death is due to loss of the osmoregulatory functions of the gills, kidneys, and gut (27).
All members of the family Herpesviridae exhibit 2 distinct life-cycle phases: lytic replication and latency. Laten-cy is characterized by maintenance of the viral genome as a nonintegrated episome and expression of a limited number of viral genes and microRNAs. At the time of reactivation, latency is replaced by lytic replication. Latency has not been demonstrated conclusively in members of the family Alloherpesviridae. However, some evidence supports existence of a latent phase. CyHV-3 DNA has been detected by real-time PCR at 65 dpi in clinically healthy fi sh (27). Furthermore, the virus persisted in a wild population of common carp for at least 2 years after the initial outbreak (31). Finally, St-Hilaire et al. demonstrated the possibility of a temperature-dependent reactivation of CyHV-3 lytic infection several months after initial exposure to the virus (32). This fi nding suggests that the temperature of the water could control the outcome of the infection (replicative/ nonreplicative). Whether the observations described above refl ect latent infection, as described for the family Herpesviridae, or some type of chronic infection, remains to be determined. Similarly, the carp organs that support this latent or chronic infection still need to be identifi ed.

Transmission
Horizontal transmission of CyHV-3 in feces (26) and secretion of viral particles into water (21) have been demonstrated. The skin of carp acts as the portal of entry of CyHV-3 and the site of early replication (29). The early replication of the virus at the portal of entry could contribute not only to the spread of the virus within infected fi sh but also to the spread of the virus throughout the fi sh population. As early as 2-3 dpi, infected fi sh rubbed against other fi sh or against objects. This behavior could contribute to a skin-to-skin mode of transmission. Later during infection, this mode of transmission could also occur when uninfected fi sh pick at the macroscopic herpetic skin lesions on infected fi sh. To date, no evidence of vertical transmission of CyHV-3 has been found. The fi rst signs appear at 2-3 dpi. The fi sh exhibit appetite loss and lethargy and lie at the bottom of the tank with the dorsal fi n folded. Depending on the stage of the infection, the skin exhibits different clinical signs, such as hyperemia, particularly at the base of the fi ns and on the abdomen; mucus hypersecretion; and herpetic lesions ( Figure  6). The gills frequently become necrotic and hypersecrete mucus, which suffocates the fi sh. Bilateral enophthalmia is observed in the later stages of infection. Some fi sh show neurologic signs in the fi nal stage of the disease, when they become disoriented and lose equilibrium (3,19,21).

Histopathologic Findings
In CyHV-3 infected fi sh, prominent pathologic changes occur in the gill, skin, kidney, liver, spleen, gastrointestinal system, and brain (3,17,21,28). Histopathologic changes appear in the gills as early as 2 dpi and involve the epitheli-al cells of the gill fi laments. These cells exhibit hyperplasia, hypertrophy, and/or nuclear degeneration (3,17,21,28). Severe infl ammation leads to the fusion of respiratory epithelial cells with cells of the neighboring lamellae, resulting in lamellar fusion (17,28). In the kidney, a weak peritubular infl ammatory infi ltrate is evident as early as 2 dpi and, along with blood vessel congestion and degeneration of the tubular epithelium in many nephrons, increases with time (17). In the spleen and liver, splenocytes and hepatocytes, respectively, are the most obviously infected cells (28). In brain of fi sh that showed neurologic signs, congestion of capillaries and small veins are apparent in the valvula cerebelli and medulla oblongata, associated with edematous dissociation of nerve fi bers (28).

Diagnosis
Diagnosis of CyHV-3 infection is described elsewhere (20). Suspicion of CyHV-3 infection is based on clinical signs and histopathologic fi ndings. Since initial isolation of CyHV-3 in 1999, complementary diagnostic methods have been developed. Virus isolation from infected fi sh tissues in cell culture (C. carpio carp brain and koi fi n cells) was the fi rst method to be developed (3). This time-consuming approach is still the most effective method for detecting infectious particles during an outbreak of CyHV-3 infection. A complete set of techniques for detecting viral genesincluding PCR (20), nested PCR (33), TaqMan PCR (27), and loop-mediated isothermal amplifi cation (34)-has been developed. Real-time TaqMan PCR has been used to detect CyHV-3 in freshwater environments after concentration of viral particles (2). Finally, ELISAs have been developed to detect specifi c anti-CyHV-3 antibodies in the blood of carp (35) and to detect CyHV-3 antigens in samples (17,26).

Immune Response
Immunity in ectothermic vertebrates differs in several ways from that of their mammalian counterparts. Environmental temperature has drastic effects on the fi sh immune system. In carp, for example, at <14°C, adaptive immunity is inhibited, but the innate immune response remains functional (36). As mentioned above, host temperature also has an effect on CyHV-3 replication, which can occur only at 18°C-28°C. In carp that are infected and maintained at 24°C, antibody titers begin to rise at ≈10 dpi and plateau at 20-40 dpi (37). In the absence of antigenic reexposure, the specifi c antibodies gradually decrease over 6 months to a level slightly above or comparable to that of unexposed fi sh. Although protection against CyHV-3 is proportional to the titer of specifi c antibodies during primary infection, immunized fi sh, even those in which antibodies are no longer detectable, are resistant to a lethal challenge, possibly because of the subsequent rapid response of B and T memory cells to antigen restimulation (37).

Prophylaxis and Control
For CyHV-3 control, 3 approaches are being developed. They are 1) management and commercial measures to enhance the international market of certifi ed CyHV-3 -free carp and to favor eradication of CyHV-3, 2) selection of CyHV-3-resistant carp, and 3) development of safe and effi cacious vaccines.

Selection of CyHV-3-Resistant Carp
Carp resistance to CyHV-3 might be affected by host genetic factors. Shapira et al. demonstrated differential resistance to CyHV-3 (survival rates 8%-60%) by crossbreeding sensitive domesticate strains and a resistant wild strain of carp (38). Further supporting the role of host genetic factors in CyHV-3 resistance, major histocompatibility class II genes were recently shown to affect carp resistance (39).

Vaccination of Carp
Soon after the characterization of CyHV-3, a protocol to induce a protective adaptive immune response in carp was developed. This approach is based on the fact that CyHV-3 induces fatal infections only when the water temperature is 18°C-28°C.
According to this protocol, healthy, uninfected fi sh are exposed to CyHV-3 infected fi sh for 3-5 days at permissive temperature (22°C-23°C) and then transferred for 30 days to ponds at a nonpermissive temperature (≈30°C). After this procedure, 60% of fi sh become resistant to further challenge with CyHV-3 (4). Despite its ingenuity, this method has several disadvantages: 1) increasing the water temperature to 30°C makes the fi sh more susceptible to secondary infection by other pathogens and requires a large amount of energy in places where the water is naturally cool; 2) the protection is observed in only 60% of fi sh; 3) carp that are "vaccinated" by using this protocol have been exposed to wild-type virulent CyHV-3 and could therefore represent a potential source of CyHV-3 outbreaks if they later come into contact with an unexposed carp.
Attenuated live vaccine appears to be the most appropriate for mass vaccination of carp. Attenuated vaccine candidates have been produced by successive passages in cell culture (4). The vaccine strain candidate was further attenuated by UV irradiation to increase the mutation rate of the viral genome (4,37). A vaccine strain obtained by this process has been produced by KoVax Ltd. (Jerusalem, Israel) and has been shown to confer protection against a virulent challenge. However, this vaccine is available in only Israel and has 2 main disadvantages: 1) the molecular basis for the reduced virulence is unknown, and consequently, reversions to a pathogenic phenotype cannot be excluded; and 2) under certain conditions, the produced attenuated strain could retain residual virulence that could be lethal for a portion of the vaccinated fi sh (37).
An inactivated vaccine candidate was described by Yasumoto et al. (40). It consists of formalin-inactivated CyHV-3 trapped within a liposomal compartment. This vaccine can be used for oral immunization in fi sh food. Protection effi cacy for carp is 70% (40).

Conclusions
Because CyHV-3 causes severe fi nancial losses in the common carp and koi culture industries worldwide, it is Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 16 a useful subject for applied science. Safe and effi cacious vaccines adapted to mass vaccination of carp and effi cient diagnostic methods need to be developed. Several aspects of CyHV-3 make it also useful for fundamental science. These aspects are its large genome, the relationship between CyHV-3 infectivity and temperature, and the low similarity between CyHV-3 genes and the genes of other members of the order Herpesvirales that have been studied. Further studies are needed to identify the roles of CyHV-3 genes in viral entry, egress, and disease pathogenesis.
This work was supported by a grant from the University of Liège (Crédit d'Impulsion) and a Fonds de la Recherche Fondamentale Collective grant of the Fonds National de la Recherche Scientifi que (2.4622.10).
Dr Michel is a biologist at the University of Liège. His research interests include cyprinid herpesvirus 3.