Infectious hematopoietic necrosis virus: advances in diagnosis and vaccine development

Clinical diagnosis

Typically, fishes infected by IHNV will become lethargic. The infected fishes will also show abnormal swimming patterns such as sporadic whirling, spiral swimming, and flashing. Other symptoms that can be observed through the physical appearance include darkening of the skin color, exophthalmia, pale gills and mucoid, distended abdomens, opaque feces casts, and petechial hemorrhages (Fig. 2) (Woo & Cipriano, 2017).

Several reliable clinical methods for the detection and identification of IHNV are based on the gross and microscopic pathology, chemical pathology, tissue imprints, and electron microscopic analysis. Gross pathological signs of infected fishes include pale internal organs such as the liver, kidney, and spleen; distended abdomen with gelatinous substance; exophthalmia; petechial hemorrhages in the muscles and tissues surrounding the organs of the body cavity; and spinal deformities in surviving fishes. Whereas the microscopic pathological signs include necrosis of eosinophilic granular cells in the intestinal wall, and the degenerative necrosis in hematopoietic tissues, digestive tract, kidney, liver, spleen, and pancreas (Schipp, 2012). As IHNV can cause renal damage to infected fishes, it can lead to significant changes in the cellular and chemical blood constituents. By comparing with uninfected fishes, the ill fishes are anemic and leukopenic, with degenerating thrombocytes and leucocytes. A large amounts of cellular debris (necrobiotic bodies) can therefore be observed in the blood (Woo, Leatherland & Bruno, 2011).

In IHNV infected fishes, splenic and renal hematopoietic tissues are the first and most severely affected areas. Therefore, the cytopathic effect (CPE) of the virus can best be observed using tissue imprints prepared from the kidney and spleen. These imprints often show foamy macrophages and necrobiotic debris, indicating IHNV infection (Kibenge & Godoy, 2016). IHNV infection can also be identified through direct observation of virus particles using an electron microscope. The virions can be detected on the cell surface, within cell vacuoles, as well as in the intracellular spaces of the virus-infected cells (OIE, 2018).

Recently, Burbank, Fehringer & Chiaramonte (2017) reported a non-lethal sampling technique through fin clipping in adult steelhead trout, followed by the detection of IHNV with cell culture techniques. This method has been demonstrated to be more efficient than the standard lethal sampling methods, such as spleen and anterior kidney sampling (Burbank, Fehringer & Chiaramonte, 2017). The confirmation test or “gold standard” for IHNV diagnosis is by detecting the virus in cell cultures, followed by diagnosis using immunological and molecular techniques (Barlič-Maganja et al., 2002; Burbank, Fehringer & Chiaramonte, 2017; Crane & Williams, 2008; Winton, 1991; Woo, Leatherland & Bruno, 2011). The presence of IHNV is routinely assessed by observing the development of viral CPE in cell lines such as epithelioma papulosum cyprinid and fathead minnow under a phase-contrast microscope (Dixon et al., 2016). When virus-like structures are observed in cell cultures with the viral CPE using an electron microscope, a further confirmation test with either the serological method, molecular method, or a combination of both methods is required (OIE, 2018).

Serological diagnosis

Serological diagnosis often requires the use of polyclonal or monoclonal antibodies which bind specifically to the pathogen. The classic VNT is time consuming as it takes 2–8 weeks to complete. Nevertheless, VNT is still being used to detect IHNV infection without sacrificing the fish (Jenčič et al., 2014). More rapid tests based on viral antigen recognition, such as the direct and indirect fluorescent antibody tests (FAT/IFAT) (Arnzen et al., 1991; LaPatra et al., 1989; Woo, Leatherland & Bruno, 2011), ELISA (Adams & Thompson, 2011; Kim et al., 2008), peroxidase immunohistochemical and alkaline phosphatase immunocytochemical (APIC) staining (Drolet, Rohovec & Leong, 1993; Yamamoto et al., 1990), and western blotting (Ristow et al., 1993) have been successfully developed. FAT/IFAT and APIC staining are often used to detect the presence of IHNV in infected fishes through immunostaining of tissue imprints or fixed tissue sections. Drolet, Rohovec & Leong (1993) demonstrated that the APIC assay can detect IHNV in fixed tissue samples over a year old (Drolet, Rohovec & Leong, 1993).

Similarly, ELISA, dot blotting, and western blotting are used to confirm the presence of IHNV by detecting the viral components with antibodies which bind specifically to the viral antigens. To further contribute to the serological detection of IHNV, Xu et al. (2016) performed a high throughput screening method by using the flow cytometry to select recombinant antibodies which could be used as potential universal diagnostic reagents. Another rapid detection method which is known as the staphylococcal coagglutination test can be used to diagnose IHNV within 15 min (Bootland & Leong, 1992; Kim, Winton & Leong, 1994). With the aid of a portable light microscope, this method has the potential to be used as an on-site or point-of-care diagnostic test. However, the staphylococcal coagglutination test is rarely used in the past decade, possibly due to advancements in point-of-care diagnosis using molecular methods. Apart from using specific antibodies, nucleic acid hybridization probes labeled with biotin or alkaline phosphatase can also be employed to detect the presence of IHNV genomic materials (Gonzalez et al., 1997).

Molecular diagnosis

The application of molecular diagnosis in clinical microbiology laboratories accelerates the detection and identification of IHNV. Molecular diagnostic methods are generally better than serological methods in terms of sensitivity, as the presence of IHNV genes can be easily amplified with methods such as PCR and LAMP (OIE, 2018). Since IHNV is an RNA virus, reverse transcription- (RT-) PCR is often used to detect the N and G genes of IHNV (Emmenegger et al., 2000; Knüsel et al., 2007). In addition, real-time RT-PCR (qRT-PCR) is also commonly used to detect IHNV. qRT-PCR generally has a lower risk of contamination, greater sensitivity, and exclusion of post-PCR analysis as compared to RT-PCR (Dixon et al., 2016; Woo, Leatherland & Bruno, 2011). More importantly, qRT-PCR is capable of quantifying the viral genome or transcripts, thereby could be used to determine the health status of an infected fish (Overturf, LaPatra & Powell, 2001).

Purcell et al. (2013) developed a universal qRT-PCR targeting the N gene of IHNV, and reported a sensitivity and specificity of 100%. As quantitation with qRT-PCR requires the establishment of a standard curve, the results generated from different laboratories could be different. Therefore, RT-droplet digital PCR (RT-ddPCR) has been employed for quantitative detection of IHNV as an alternative to qRT-PCR (Jia et al., 2017). In addition, Pinheiro et al. (2016) developed a multiplex RT-PCR (mRT-PCR) for simultaneous detection of major viruses that infect rainbow trout, including IHNV. In the following year, Tong et al. (2017) also developed a liquid chip technique for simultaneous detection of IHNV, spring viremia of carp virus (SVCV), and viral hemorrhagic septicemia virus (VHSV) in salmonids, through the use of fluorescence-coded microspheres for hybridization with the RT-PCR products.

Loop-mediated isothermal amplification or RT-LAMP is a powerful diagnostic tool to detect aquaculture diseases as it is rapid and highly sensitive, in which a few copies of cDNA can be amplified by 10⁹ folds in less than an hour (Biswas & Sakai, 2014; Fu et al., 2011; Suebsing et al., 2011b). Suebsing et al. (2011a) demonstrated that RT-LAMP can detect as little as 0.01 fg of RNA extracted from IHNV-infected cells. In addition, RT-LAMP is suitable to be applied as a point-of-care IHNV detection tool as the amplification of DNA does not require an expensive thermal cycle, which is a must in PCR-based methods. One of the advantages of LAMP in detecting IHNV is that it allows direct and rapid visualization of the amplified products with naked eyes due to the formation of magnesium pyrophosphate (white precipitate byproduct generated from LAMP), which indicates a successful amplification of the target genomic region (Dhama et al., 2014; Gunimaladevi et al., 2005). This feature makes it applicable in laboratory and field conditions.

Ideally, methods established to diagnose IHNV should not be limited to laboratories as they can also be applied in farms which involve a large number of samples. These methods have to be simple, user-friendly, specific, sensitive, rapid, and affordable to fish farmers.

Development of DNA vaccine

DNA vaccine is a type of genetic vaccine which involves the introduction of recombinant plasmid encoding an immunogenic antigen into host cells, whereby the antigen could be translated and primes the immune system (Van Drunen Littel-Van Den Hurk, Loehr & Babiuk, 2001). Along with the advancement in genetic engineering, numerous DNA vaccines have been invented in the past three decades and many have entered clinical trials (Ferraro et al., 2011). Although the developed DNA vaccines are more focused on targeting human diseases, DNA vaccines against IHNV were also frequently reported (Alonso et al., 2003; Ferraro et al., 2011; Tonheim, Bogwald & Dalmo, 2008). An obvious advantage of DNA vaccines over protein-based vaccines is the scalability and lower cost of production with reduced complexities. DNA vaccines could circumvent most of the problematic issues associated with protein-based vaccines including challenges in protein purification, low protein expression, low protein solubility, and protein misfolding (Leitner, Ying & Restifo, 1999). Importantly, most DNA vaccines were also demonstrated to be capable of inducing both the cellular and humoral immune responses similar to the live attenuated vaccines (Wang et al., 1998). Moreover, DNA vaccines have a better safety profile in contrary to live attenuated vaccines comprising attenuated pathogens, which may pose a risk of regaining virulence in the host (Pliaka, Kyriakopoulou & Markoulatos, 2012). In addition, plasmid DNA containing immunostimulatory sequence (CpG motifs) also increases the immunogenicity of the vaccine and reduces the reliance on toxic adjuvants which often result in adverse inflammation (Coombes & Mahony, 2001). Plasmid DNA could also be engineered to encode multiple viral antigens to generate multivalent DNA vaccines (Tonheim, Bogwald & Dalmo, 2008).

Majority of the IHNV DNA vaccines were developed based on the G protein of the IHNV M and U genotypes, which were found to induce strong humoral immune responses in immunized fishes (Nichol, Rowe & Winton, 1995; Peñaranda, LaPatra & Kurath, 2011). DNA vaccines designed based on other internal viral proteins of IHNV such as N, P, M, and NV did not induce any significant protective immunogenicity (Corbeil et al., 1999). Recently, an IHNV DNA vaccine encoding the G protein of the J genotype was found to be effective against a wide range of IHNV strains by eliciting strong neutralizing antibody responses and upregulation of Mx-1 gene, an IFN-inducible antiviral effector (Xu et al., 2017a). On other hand, DNA vaccines which consist of recombinant plasmid encoding the G protein derived from other serologically distant rhabdoviruses: SVCV or snakehead rhabdovirus (SHRV) were also shown to induce notable cross protections in the early (30 days post-vaccination) but not the late (70 days post-vaccination) lethal IHNV challenges (Kim et al., 2000). The G proteins of IHNV, SVCV, and SHRV shared only about 11% homology in amino acid sequences, therefore, protective responses observed during the early IHNV challenge could largely attributed to the non-specific innate immune responses conferred by IFN-inducible antiviral Mx-1 protein. As the non-specific immune responses faded over time, immunized fishes become more vulnerable to the late IHNV challenge, thus an increased mortality was observed. Nevertheless, fishes immunized with DNA vaccine encoding the G protein of IHNV survived in both the early and late IHNV challenges, suggesting that a long term effective protection requires specific immune responses (Kim et al., 2000). Previous studies have also indicated that co-infection and interactions between infectious pancreatic necrosis virus (IPNV) and IHNV have led to the loss of infective titer of IHNV due to the early release of interfering cytokines which inhibit the viral activities (Alonso, Rodriguez & Perez-Prieto, 1999; Saint-Jean & Perez-Prieto, 2007; Tafalla, Rodriguez Saint-Jean & Perez-Prieto, 2006). To investigate the capability of IPNV in inducing early cross protection against IHNV, De Las Heras, Perez Prieto & Rodriguez Saint-Jean (2009) created a DNA vaccine encoding the VP2 protein of IPNV, and demonstrated its protective efficacy against early heterologous IHNV challenges. Similar results were obtained when DNA vaccine against another rhabdovirus, VHSV was recruited for early IHNV challenge (LaPatra et al., 2001; Lorenzen et al., 2002b). However, the early non-specific cross protection conferred by the rhabdovirus DNA vaccines was shown to be restricted to viral but not bacterial infection as no increment in survival rate was detected when the immunized trout were challenged with bacterial pathogens (Lorenzen et al., 2002b).

Multivalent DNA vaccines

Viral hemorrhagic septicemia virus and IHNV are common pathogens endemic to rainbow trout in Europe. Co-administration of IHNV and VHSV DNA vaccines in a single injection in rainbow trout was previously reported to induce long-lasting protections against both individual and combined virus challenges (Boudinot et al., 1998; Einer-Jensen et al., 2009). Dual DNA vaccination could be a viable alternative to avoid repeated stressful vaccination procedures in rainbow trout. However, simultaneous vaccinations of several plasmid DNA encoding different antigens have also been reported to reduce the immunogenicity of the vaccines compared to those administered alone (Sedegah et al., 2004). Remarkably, a recent bivalent DNA vaccine encoding both the G protein of IHNV and VP2–VP3 of IPNV was shown to be highly effective in rainbow trout against individual and simultaneous IHNV and IPNV challenges. In all cases, the RPS was over 90% (Xu et al., 2017b). Multivalent vaccines have an added advantage over multi-DNA vaccination due to lower cost of production, as only one type of plasmid is required to produce multiple immunogenic antigens. However, the size of plasmid can affect its transformation into both the prokaryotic and eukaryotic cells (Kreiss et al., 1999; Ohse et al., 1995). Therefore, efforts should be given while designing DNA vaccines to minimize the size of recombinant plasmids, particularly those of multivalent vaccines. Studies on IHNV DNA vaccines are summarized in Table 2.

Factors affecting the efficacy of DNA vaccines

Protective immune responses induced by DNA vaccines could vary widely based on the route of immunization. Intramuscular injection is the most common DNA immunization technique employed in aquaculture, particularly fishes (Corbeil, Kurath & LaPatra, 2000; Garver, LaPatra & Kurath, 2005; Lorenzen et al., 2002a; Peñaranda, LaPatra & Kurath, 2011; Xu et al., 2017a). Corbeil, Kurath & LaPatra (2000) demonstrated that gene gun and intramuscular injection are the most efficient DNA delivery methods as measured by the protective efficacy on the immunized rainbow trout fry challenged with IHNV, whereas intraperitoneal injection induced partial protection. Nevertheless, other routes of DNA immunization including intrabuccal administration, scarification of the skin, and the immersion method were shown to be ineffective against IHNV challenge (Corbeil, Kurath & LaPatra, 2000). Although DNA vaccination via injection method is highly effective against IHNV, this technique is stressful to fishes, time consuming and laborious (Corbeil, Kurath & LaPatra, 2000).

A more cost-effective route of vaccination includes oral DNA vaccination. However, an oral vaccination requires the DNA to be protected from degradation in the digestive tract. Adomako et al. (2012) utilized a copolymer, poly(D,L-lactic-co-glycolic acid) as a nanocarrier for the delivery of oral DNA vaccine, where a slight protection towards immunized fishes was reported. Recently, Ballesteros et al. (2015) encapsulated the DNA encoding G protein of IHNV with an alginate microsphere, and orally vaccinated the rainbow trout. Their results revealed that the DNA vaccine was effectively protected in the fish gut by the alginate microsphere, resulting in a significant reduction in mortality of the immunized fishes. To date, oral vaccination is less effective compared to intramuscular injection. However, further optimization in the future could possibly enhance the protective efficacy of these vaccines. Therefore, it represents a viable alternative in aquaculture, in which it is more practical: lower cost and less laborious.

DNA vaccine delivery by attenuated bacteria via horizontal gene transfer was also previously suggested to be a suitable route of immunization in aquaculture due to its low labor cost. Despite successful demonstration of GFP gene transfer into salmonid fish cells via attenuated invasive E. coli, in vitro or in vivo gene transfer of IHNV G protein into fish cells has not been conducted (Simon & Leong, 2002).

The efficacy of IHNV DNA vaccines has been reported to be dose dependent (Ballesteros et al., 2015; Corbeil et al., 2000; Garver, LaPatra & Kurath, 2005; LaPatra et al., 2000). In general, a larger fish requires a higher vaccination dose for effective protection against IHNV. A 120 g-fish requires about 100 times higher dosage to achieve similar protective immunity compared to fingerlings of one to three g (LaPatra et al., 2000). LaPatra, Lorenzen & Kurath (2002) later demonstrated that 0.1 µg of IHNV DNA vaccine is sufficient to induce significant protection in rainbow trout fry. As a rule of thumb, to induce sufficient protective immune response in rainbow trout, intramuscular vaccination of at least 10 ng DNA per gram body weight is required (Lorenzen et al., 2002a).

IHNV DNA vaccines are most commonly tested on rainbow trout, often resulting in high neutralizing antibodies and survival rate in the fish (Corbeil et al., 1999, 2000; Kim et al., 2000; LaPatra, Lorenzen & Kurath, 2002; Xu et al., 2017a). Atlantic salmon was also recruited as an animal model to study the efficacy of IHNV DNA vaccine, in which they were greatly protected from IHNV immersion and cohabitation challenges (over 90% RPS). Furthermore, passive serum transfer from the immunized Atlantic salmon to rainbow trout has also increased the survival rate of the recipients (Traxler et al., 1999). On the other hand, Chinook and sockeye salmon immunized with DNA vaccines also exhibited increased survivability, although to a lesser extend compared to Atlantic salmon and rainbow trout (Garver, LaPatra & Kurath, 2005).

Apart from host differences, external parameters such as temperature also play an important role in determining the efficacy of the DNA vaccines. Lorenzen et al. (2002a) suggested that the DNA vaccine encoding G protein of rabies virus failed to elicit early unspecific protection against IHNV could be due to the low water temperature. Intriguingly, Lorenzen et al. (2009) later demonstrated that IHNV and VHSV DNA vaccines induced different defense mechanisms in rainbow trout upon VSHV challenge at different temperatures. At low temperature of 5 and 10 °C, IHNV DNA vaccine could induce a prolonged cross protection against VSHV challenge but no significant protection was observed at 15 °C. In addition, the activity of Mx protein and the level of neutralizing antibody of the immunized fish were also found to vary at different temperatures (Lorenzen et al., 2009). Therefore, the effect of vaccines at different water temperatures should be studied to achieve an optimal protection.

Vaccine efficacy can be affected by the route of vaccine delivery as different vaccination approaches influence vaccine localization and priming of the immune cells, and consequently affect the systemic immune responses (Zhang, Wang & Wang, 2015). Due to the complexity of different vaccines, hosts, vaccine dosages, types of adjuvant involved, injection volumes and intervals between injections, thus relative immunogenicity of the vaccines administered by different routes could vary considerably (Zhang, Wang & Wang, 2015). The underlying mechanism of different routes of vaccination in affecting DNA vaccine’s efficacy in fishes remains elusive. Nevertheless, intramuscular injection is the most widely used method for DNA vaccination in fishes due to its ability to induce potent immune responses (Tonheim, Bogwald & Dalmo, 2008). Studies in mice demonstrated the distribution of plasmid DNA between the muscle body and epimysium following a DNA vaccination, subsequently myocytes and mononuclear cells were shown to rapidly uptake plasmid DNA shortly after intramuscular injection (Hølvold, Myhr & Dalmo, 2014). DNA immunization by gene gun, on the other hand, introduced the DNA plasmid directly into the cytoplasm, presumably resulting in the DNA being processed by antigen presenting cells, and subsequently activating the adaptive immunity (Wang et al., 2008). DNA vaccines delivered via oral route are relatively less laborious but they were shown to be less effective. Oral DNA vaccination required special protection for the plasmid DNA against hostile fish digestive system to prevent DNA degradation before cellular uptake. Even the DNA plasmid was protected from degradation via certain approaches, transfection efficiency of the plasmid DNA in the fish digestive system poses another challenge (Corbeil, Kurath & LaPatra, 2000). Immersion route is simple and suitable for mass vaccination of fishes. However, transfection efficiency of the plasmid DNA delivered via immersion route is heavily affected by many factors such as the length of immersion time, size of the fish, stress, pH, osmolarity of the vaccine buffer, the water temperature, and the physical properties (particulate or soluble) of the antigen. Each parameter has to be optimized to improve transfection efficiency and immunogenicity of the DNA vaccine (Nakanishi & Ototake, 1997).

Controversial in DNA vaccination

Despite the tremendous amount of promising results yielded by DNA vaccines against fish pathogens, the introduction of foreign DNA into human foods has always been controversial throughout the past decades (Alonso & Leong, 2013). There is a possibility that the plasmid DNA could integrate into the host genome, leading to insertion mutations. Nevertheless, plasmid DNA delivered via intramuscular injection into muscle cells exists as an extra-chromosomal DNA, and its integration into the host genome was reported to be negligible (Kanellos et al., 1999; Ledwith et al., 2000; Nichols et al., 1995). To further mitigate this issue, Alonso et al. (2003) developed a DNA vaccine based on the G gene of IHNV, controlled by the interferon regulatory factor 1A promoter originated from rainbow trout to prevent its expression in human. In addition, a study by Salonius et al. (2007) also indicated that the potential risk of spontaneous mutations in Atlantic salmon was about 43-folds higher than that caused by DNA vaccination. Furthermore, several studies have suggested that IHNV DNA vaccination of rainbow trout only caused transient histopathological changes in multiple tissues, and no long-term histopathological damage was observed (Garver et al., 2005; Kurath et al., 2006). However, certain regulations such as the Norwegian Gene Technology Act which categorized DNA vaccinated animals as genetically modified organisms presents a stringent policy, eventually leading to low public acceptance (Alonso, Chiou & Leong, 2011). To eliminate this concern, a self-destructive IHNV DNA vaccine was designed (Alonso, Chiou & Leong, 2011). The plasmid DNA contains an inducible fish cell promoter which regulates the expression of G glycoprotein for protective immune responses, and a ZnCl₂ inducible promoter which controls the expression of IHNV M protein inducing apoptosis of the transfected cells. Upon successful vaccination, fishes were significantly protected from lethal IHNV challenge, and exposure to ZnCl₂ induced apoptosis in fish cells containing the DNA vaccine without causing serious toxicity to the fishes (Alonso, Chiou & Leong, 2011). This approach could pave a way to the development of safer DNA vaccines with higher public acceptance.

Although many research groups have patented their inventions, including Kurath et al. (1985) (Patent No.: 5354555), Salonius et al. (2007) (Patent No.: EP1553979A1, CA2498896C, CN100339131C, JP4578973B2, ES2288627T3, DK1553979T3, DE60315858T2, PT1553979E, AT370746T, AU2003277863B2, WO2004026338A1, NO20051840L, HK1082666A1, CY1107784T1), Alonso et al. (2003) and Alonso, Chiou & Leong (2011) (Patent No.: WO/2002/069840), and Xu et al. (2017a) (Patent No.: CN105816871A, CN105861450A), to date, the Apex-IHN^® manufactured by Aqua Health Ltd (an affiliate of Novartis) is the only licensed IHNV DNA vaccine in Canada and the USA (Grunwald & Ulbert, 2015; USDA, 2014).