Hepatitis E virus (HEV) is a positive-sense, single-stranded RNA virus, and is classified in the genus Orthohepevirus, the family Hepeviridae (Smith et al., 2014). The HEV-caused hepatitis E is generally a self-limiting disease with a case fatality rate from 0.5 to 3% in young adults but up to 30% in infected pregnant women in their third trimester of gestation (Jameel, 1999; Nan, 2014). World Health Organization (WHO) estimates that there are 20 million infections with over 3 million symptomatic cases and 56,600 deaths annually across the world (WHO, 2015). HEV is primarily transmitted via fecal-oral route. HEV infection was previously thought to be a public health problem only for the developing countries. Indeed, hepatitis E is highly endemic in East and South Asia, as well as Africa according to the WHO (WHO, 2015). HEV strains infecting humans are classified into four genotypes. HEV strains from genotypes 3 and 4 are zoonotic, whereas, those from genotypes 1 and 2 have no known animal origin. Discovery of HEV from swine and other species suggests that genotypes 3 and 4 HEV has a wide host range (Christensen et al., 2008; Dalton et al., 2008; Meng, 2013; Pavio et al., 2015). Currently, hepatitis E is frequently recognized in industrialized countries, where it was not thought to be endemic (Kwo et al., 1997; Erker et al., 1999; Schlauder et al., 1999; Worm et al., 2000; Kabrane-Lazizi et al., 2001; Mizuo et al., 2002; Sadler et al., 2006). Moreover, along with isolation of HEV from the pig, chicken, mongoose, rabbit, rat, ferret, bat, fish, and deer (Meng et al., 1997; Haqshenas et al., 2001; Li et al., 2005c; Cossaboom et al., 2012; Smith et al., 2014), cross-species infection of HEV from animal reservoirs to humans is thought to be the major cause of sporadic cases of hepatitis E in the industrialized countries (Pavio et al., 2015). Although previously thought to only cause acute infections, HEV is found in chronic infections reported both in immune compromised and immunocompetent individuals (Hoofnagle et al., 2012; Grewal et al., 2014). In addition, extrahepatic manifestations, such as neurological disorders and kidney injury in HEV infected patients have been documented (Kamar et al., 2011, 2012b; van Eijk et al., 2014; Dalton et al., 2016; Geng et al., 2016). Taken together, current knowledge for HEV implies a significant underestimation of HEV infection as a public health concern. In the following sections, recent progress in HEV biology, functions of viral proteins, cell culture system, epidemiology, viral pathogenesis, treatment, and vaccine development are reviewed and perspective new insights are discussed.

Hepatitis E was initially designated as enterically transmitted non-A, non-B hepatitis (ET-NANBH) due to similar clinical presentations to hepatitis A and B in patients, but the prospective causative agent was initially unknown (Balayan et al., 1983). Early research implied that an RNA virus was the potential pathogen for the ET-NANBH. By analysis of a cDNA library from infectious bile sample, a portion of a highly conserved RNA-dependent RNA-polymerase (RdRp) motif, commonly found in RNA viruses, was identified (Reyes et al., 1990). This new virus was designated as HEV, which was responsible for the outbreak of ET-NANBH.

The complete sequence of HEV genome was published 1 year later (Tam et al., 1991). Sequence analysis indicated that HEV contains a 7.2 kb single-stranded positive-sense RNA genome, which is capped and poly-adenylated (Ahmad et al., 2011). There are three partially overlapped open reading frames (ORFs) in an order of sequences encoding non-structural proteins (NSPs) followed by structural protein (Tam et al., 1991; Tsarev et al., 1992; Figure 1). HEV ORF1 encodes a non-structural polyprotein that consists of replicase proteins needed for HEV replication. ORF2 encodes the capsid protein, which is the major structural protein of the HEV virions, which are non-enveloped particles of 32–34 nm in diameter (Mori and Matsuura, 2011). ORF3 encodes a small multifunctional protein with a molecular mass of 13 kDa (VP13). There are also short untranslated regions (UTRs) in both the 5′ and 3′-end of the genome. Recently, an ORF4 was identified from genotype 1 HEV solely (Nair et al., 2016; Figure 1). Expression of ORF4 is cap-independent and driven by a putative IRES-like element between 2701 and 2787 nt of the HEV genome (Nair et al., 2016).

The ORF1 of HEV can be translated directly from the genomic RNA, whereas ORF2 and ORF3 are translated soly from the sub-genomic RNA in alternative frames (Graff et al., 2006). In an earlier report, three RNA species were detected in liver tissue of experimentally infected macaques, with sizes of 7.2, 3.7, and 2 kb (Tam et al., 1991). The 3.7 and 2 kb RNA species were thought to be sub-genomic RNAs for translation of ORF2 and ORF3, respectively. However, a later study in Huh7 cells only identified one capped 2.2 kb sub-genomic RNA, which is a bicistronic mRNA for translation of both ORF2 and ORF3 (Graff et al., 2006). Transcription of this sub-genomic RNA initiates at nucleotide position 5122 in the Sar55 strain, which is located downstream of the first two methionine codons of the initially presumed ORF3. The same conclusion was drawn from another in vitro study of genotype 3 HEV infection in PLC/PRF/5 hepatoma cells (Ichiyama et al., 2009). The HEV genome contains two cis-reactive elements (CREs) that are essential for the viral replication (Cao et al., 2010; Parvez, 2015b). The first CRE overlaps the 3′ end of ORF2 and the 3′ UTR and is essential for HEV replication. The second CRE locates in the intergenic region of the HEV genome and forms a stem–loop structure that may be the promoter for synthesis of the 2.2-kb subgenomic RNA (Cao et al., 2010).

Hepatitis E virus was initially classified as a member of the Caliciviridae family. However, sequence analysis of HEV ORF1 indicated no similarity to Caliciviruses, or other picorna-like viruses. On the other hand, there is limited but significant similarity to the alphavirus-like superfamily of RNA viruses, specifically, the rubella virus (Berke and Matson, 2000). Consequently, HEV was classified into the family Hepeviridae (Berke and Matson, 2000; Emerson and Purcell, 2003).

Although HEV strains are highly diverse and heterogenic, only one serotype of HEV exists. Classification of HEV strains is under transition due to the different criteria used (Smith et al., 2013, 2014). Recently, a new proposal for the classification of the family Hepeviridae was published (Smith et al., 2014). In this proposal, the family Hepeviridae contains two genera: Orthohepevirus (all mammalian and avian HEV isolates) and Piscihepevirus (trout HEV; Table 1). Within the genus Orthohepevirus, four different species (A, B, C, D) are designated to include isolates from different hosts (Smith et al., 2014). All four previously recognized HEV genotypes (1–4) that infect humans belong to the Orthohepevirus A virus (Smith et al., 2014).

The previously recognized HEV genotypes 1–4 classification system was based on complete genomic sequences (Lu et al., 2006). HEV genotype 1 is the most conserved among the four genotypes. There is only one full-length genotype 2 sequence available (Smith et al., 2016). Both HEV genotypes 1 and 2 are restricted to humans with no known animal reservoirs, whereas genotypes 3 and 4 are zoonotic with an expanded host range (Table 1; Meng, 2010; Ahmad et al., 2011). Therefore, genotypes 3 and 4 HEV strains are highly diverse (Lu et al., 2006; Smith et al., 2014). Since the constant discovery of new HEV or HEV-related isolates from rabbit, rat, ferret, bat, moose, farmed mink, camel, and wild boar (Lorenzo et al., 2007; Zhao et al., 2009; Johne et al., 2010; Geng et al., 2011; Takahashi et al., 2011; Drexler et al., 2012; Raj et al., 2012; Krog et al., 2013; Lin et al., 2014; Woo et al., 2014), four genotypes are no longer satisfying classification of expanding HEV isolates. In the new classification system, some HEV strains from wild boars in Japan with unique viral nucleotide sequences are designated as genotypes 5 and 6, while HEV from camel is classified as genotype 7 of Orthohepevirus A (Smith et al., 2014). The genotypes that infect humans include 1, 2, 3, 4, and 7 (Table 1; Smith et al., 2016).

Hepatitis E virus-like virus isolated from avian species is called avian HEV, which shares less than 50% nucleotide identity but common antigen epitopes in the capsid protein with mammalian HEV (Haqshenas et al., 2001, 2002; Huang et al., 2004). Currently, avian HEV is classified into the species Orthohepevirus B (Smith et al., 2014). HEV strains from rat, ferret and bat are classified into the species of Orthohepevirus C and D, respectively (Smith et al., 2014). The cutthroat trout virus (CTV) is identified as an HEV-like virus in retrospective studies. CTV shares even lower sequence identity with mammalian and avian HEV and is now classified as a member of the genus Piscihepevirus (Batts et al., 2011; Smith et al., 2014).

For the four genotypes (1–4) of HEV infecting humans, there are differences in their geographic distributions. Genotype 1 HEV mainly includes strains from Asia and Africa including the Sar55 isolate, while genotype 2 contains a Mexican strain and variants from Africa. Genotypes 3, including human and swine HEV, is mainly found in the industrialized countries (Purcell and Emerson, 2008). The genotype 4 is previously thought to be found only in China (Purcell and Emerson, 2008), however, recent reports show that genotype 4 HEV strains are also isolated in other countries, including India, Indonesia, Japan, Vietnam, Spain, France, and Italy (Okamoto, 2007; Midgley et al., 2014; Lapa et al., 2015). For details about molecular epidemiology and viral evolution of HEV, please refer to the article by Purdy and Khudyakov (2011).

Due to the lack of an effective in vitro cell culture system for HEV, the replication cycle of HEV is largely unknown. The capsid protein is believed to bind to an unidentified cellular receptor to initiate viral entry. HEV-VLPs generated from recombinant ORF2 protein attach to cells via heparin sulfate proteoglycans (HSPGs; Kalia et al., 2009). Moreover, one study based on a viral overlay protein binding assay (VOPBA) suggests that a protein with molecular weight about 55 kDa could be the candidate receptor for HEV entry; but mass spectrometry revealed that this virus binding band contained 31 different proteins (Zhang et al., 2011). Another study suggests that aa 458–607 located in the C-terminal region of the capsid protein (M domain) may be the putative receptor binding site of HEV virions (He et al., 2008). Moreover, structure and sequence analyses suggest that the putative binding motif of the capsid protein is conserved among all four major mammalian HEV genotypes (Guu et al., 2009). Heat shock cognate protein 70 (HSC70), HSPGs and Grp78 are found to be involved in either cell surface binding with HEV capsids or intra-cellular transport in different models and are potential cellular receptors or essential factors for HEV proliferation (Kalia et al., 2009; Yu et al., 2011; Cao and Meng, 2012). However, further investigation is needed to confirm if these molecules truly act as receptors for HEV. After binding with its receptor, HEV particles are internalized via a dynamin-2, clathrin, and membrane cholesterol-dependent pathway (Kapur et al., 2012; Holla et al., 2015). In addition, a recent report suggests the quasi-enveloped HEV particles enter cells via a distinct pathway that involves in degradation of the lipid membrane in the lysosome (Yin et al., 2016).

After entry into permissive cells, the HEV capsid is uncoated by unknown mechanisms. In one study utilizing VLP from the truncated capsid protein HEV239, an HSP90-specific inhibitor (geldanamycin) blocks the intracellular transport of the HEV239 VLP without affecting its entry (Zheng et al., 2010). This suggests that HSP90 may play a role in the intracellular transport of HEV particles. After uncoating, the HEV ORF1 translation is followed. HEV genomic RNA replication relies on the replicase encoded by ORF1. Along with the generation of the sub-genomic RNA, translation of ORF2 and ORF3 occurs, followed by virion packing and egress. For the release of HEV particles, multivesicular body (MVB) pathway and ESCRT machinery in the cytoplasm are used (Nagashima et al., 2014).

Since the discovery of HEV, many efforts have been made to develop a rigorous in vitro cell culture system. However, the cell culture system of HEV is still limited and relatively ineffective, especially for genotype 1 HEV. An early study tried to use primary hepatocytes from macaques with serum-free medium for HEV propagation; however, HEV replication was limited and the detection of HEV in the medium relied on PCR amplification (Tam et al., 1996). A group from Japan reports that HEV isolate 87A is able to replicate in A549 cells; however, PCR was also used to detect viral RNA in the cell culture supernatant (Huang et al., 1995), instead of immunofluorescence assay to detect viral proteins. Another group also showed that the A549 cell line could be used effectively for passaging two Chinese HEV isolates (Wei et al., 2000).

On the other hand, as the commonly employed method for single-stranded, positive-sense RNA virus, transfection of capped RNA from an HEV cDNA infectious clone via in vitro transcription to PLC/PRF/5 (hepatocellular carcinoma) and Huh7 cells demonstrates limited replication of HEV (Emerson et al., 2004). Although cell lysates from the RNA transfected cells is infectious in rhesus monkeys, cell to cell spread of the virus in cultured cells is not observed (Emerson et al., 2004). S10-3 cell line, a subclone of Huh7 hepatoma cell line, has improved replication efficiency of HEV for the Sar55 strain (Graff et al., 2006; Shukla et al., 2011). But this assay still relies on transfection of the cells with full length HEV RNA. A recent report suggests that replication efficiency of genotype 1 HEV in human hepatoma cell lines (Huh7, Huh7.5, and HepG2/C3A) is affected by innate immune response (Devhare et al., 2016).

A Japanese group reports that a genotype 3 isolate from acute hepatitis patient propagates in PLC/PRF/5 and A549 cells (Tanaka et al., 2007; Okamoto, 2011). After A549 cells were seeded in a six-well plate and inoculated with HEV at 1.0 × 10⁴ and 1.0 × 10⁵ RNA copies per well, HEV RNA reached the highest titer of 10⁷ copies/ml at 50 days post-inoculation. However, PLC/PRF/5 cells could only support efficient growth as A549 with a higher MOI (1.0 × 10⁵ viral RNA copies per well). Moreover, in this HEV cell culture system, HEV infected cells need to be maintained at 35.5°C and cultured with a mixed cell culture medium (50% Dulbecco’s Modified Eagle Medium and 50% Medium 199) supplemented with 2% (v/v) fetal bovine serum and 30 mM MgCl₂. The same group also reports that a genotype 4 HEV from a fulminant hepatitis patient can grow in PLC/PRF/5 and A549 cells and reach a titer of 1.3 × 10⁷ copies/ml within 10–20 days incubation period (Tanaka et al., 2009). Moreover, a human hepatoma-derived cell line HepaRG and a porcine embryonic stem cell-derived cell line PICM-19, which have morphological and functional properties similar to primary hepatocytes, were shown to support HEV replication (Rogee et al., 2013). However, the HEV replication level in these two cell lines is very low and requires 1 month incubation (Rogee et al., 2013).

In HEV Kernow-C1 p6 cell culture system, the recombinant virus with human S17 gene insertion was presumed as a minor species in the host but was selectively adapted to cells after six passages (Shukla et al., 2011). Replication of Kernow-C1 p6 is 7.5-fold higher in HepG2/C3A human hepatoma cells than in Huh7.5, PLC/PRF/5, A549, Caco-2 or rhesus kidney cells in a 7-day incubation period, suggesting that HepG2/C3A cell line is the most permissive. Moreover, this HEV isolate is also able to infect a variety of non-primate cells, including cow, mouse, chicken, cat, dog, and rabbit cells, albeit with lower efficiency. Although it is still unclear how the insertion of S17 occurred in HEV infected patient, Okamoto’s group demonstrated that two cell adapted HEV strains HEV JE03-1760F (genotype 3) and HEV JF5/15F (genotype 4) did not shown any recombination with cellular S17 gene after 53 and 33 generations of passages in PLC/PRF/5 and A549 cells, respectively (Okamoto, 2013), which suggested the recombination and insertion of S17 may occur in patients rather than in cultured cell. Besides human HEV isolates, animal HEV strains from domestic pigs, wild boars, rabbits, and rats can be propagated in human hepatoma cell lines as well (Jirintai et al., 2012, 2014; Takahashi et al., 2012). A recent report demonstrates that pluripotent stem cell derived hepatocytes support HEV replication in vitro (Helsen et al., 2016).

In summary, the current cell culture systems for HEV have limitations. So far, only one report shows limited replication of a genotype 1 HEV strain from serum sample in cell culture without RNA transfection (Takahashi et al., 2010). On the other hand, although several groups have demonstrated that genotypes 3 or 4 HEV strains can be adapted to cultured cells and are able to re-infect new cells, a long incubation time is needed in comparison to other RNA viruses with good cell culture systems. Moreover, the cell culture adapted Kernow-C1 virus may have a different phenotype compared with its parental wild type virus as this cell culture adapted virus contains host gene sequence.

The HEV is primarily transmitted via fecal-oral route. The most common source of infection is contaminated drinking water in developing countries. For a long time, hepatitis E was thought to be a public health problem only for developing countries. However, hepatitis E is now frequently recognized in industrialized countries where it was not thought to be endemic previously (Kwo et al., 1997; Erker et al., 1999; Schlauder et al., 1999; Worm et al., 2000; Kabrane-Lazizi et al., 2001; Mizuo et al., 2002; Sadler et al., 2006).

World Health Organization estimates that there are 20 million HEV infections annually across the world. Among these cases, there are over 3 million symptomatic cases and 56,600 deaths (WHO, 2015). Hepatitis E is highly endemic in East and South Asia. Data indicates over 50% of global hepatitis E deaths occur in this region. In East Asia, large outbreaks of hepatitis E have only been described in China. Hepatitis E accounts for 20–50% of acute hepatitis cases in this region. The seroprevalence of anti-HEV antibodies in the region varies from 10 to 50%, indicating that hepatitis E is hyperendemic in this region. In South Asia, outbreaks of hepatitis E have been reported in most countries in this region, but variable in scale (WHO, 2015). HEV accounts for 20–60% of sporadic acute hepatitis and fulminant liver failure in this region. In particular, the rates of fulminant liver failure are usually higher in pregnant patients. A recent paper reported that HEV infection causes 49% acute viral hepatitis and 75% fulminant hepatic failure in pregnant women in one area in India (Kumar et al., 2014). However, the seroprevalence rates of prior exposure to HEV are relatively low, ranging from 10 to 40% in most studies.

In the developed countries, such as North America, Western Europe and Japan, no outbreaks have been reported. These areas are considered as low or non-endemic for HEV. However, sporadic cases of hepatitis E have been reported. Transmission of HEV from animal reservoirs to humans is assumed to be the major cause of those sporadic cases. A series of cases of HEV infection in people who ate undercooked deer meat 6–7 weeks before the onset of disease have been reported (Tei et al., 2003; Yazaki et al., 2003; Li et al., 2005c). HEV RNA recovered from the leftover deer meat was found to be identical in sequence to the HEV RNA recovered from the patients (Takahashi et al., 2004). Consumption of shellfish is considered a risk factor in a documented case (Koizumi et al., 2004). Thus, foodborne infection may occur from the consumption of uncooked/undercooked products from infected animals. Moreover, blood transfusion and solid organ transplant mediated HEV transmission are reported (Pas et al., 2012; Wedemeyer et al., 2012; Sue et al., 2016). IgM and IgG against HEV are detected in recipients of blood transfusions (Wedemeyer et al., 2012).

Hepatitis E virus genotype 1 is responsible for most endemic and epidemic cases of hepatitis E in Asia, and genotype 2 is prevalent in Central America and Africa (Purcell and Emerson, 2008). There is no known animal reservoir for HEV genotypes 1 and 2 (Wedemeyer et al., 2012). Genotypes 3 and 4 are zoonotic and can cause HEV infections in the developed countries. For the detailed geographical distribution of hepatitis E virus genotypes, please refer to these reviews (Dalton et al., 2008; Kamar et al., 2012a).

Hepatitis E virus infection mainly causes acute hepatitis with a case fatality rate from 0.5 to 3% in young adults (Jameel, 1999). Remarkably, case fatality rate resulting from HEV-related fulminant liver failure can reach up to 30% in infected pregnant women in their third trimester of gestation (Jameel, 1999). Generally, HEV has an incubation period of 2–8 weeks (Purcell and Emerson, 2008). The initial symptoms of acute hepatitis E are unspecific and flu-like, such as myalgia, arthralgia, and weakness. After this short prodromal phase, a period of symptoms such as vomiting, itching, uncolored stools, darkened urine and jaundice could last for days to several weeks accompanied by increased levels of liver transaminases, bilirubin, alkaline phosphatase, and γ-glutamyltransferase (Hoofnagle et al., 2012; Wedemeyer et al., 2012). Current case reports indicate that most cases are self-limited and do not result in chronic hepatitis (Hoofnagle et al., 2012). An investigation on pregnancy outcomes in hepatitis E shows higher HEV loads in pregnant women with acute viral hepatitis and fulminant hepatic failure, and higher levels of TNF-α, IL-6, IFN-γ, and TGF-β1 than non-pregnant women, which suggests that high cytokine levels are correlated with severe liver injury in HEV infection (Kumar et al., 2014). A recent study highlights the role of TLR3 and IFN-γ in HEV pathogenesis. Patients with high levels of TLR3 and robust IFN-γ response are observed in self-limiting acute viral hepatitis cases, and are able to limit the disease and recover uneventfully (Majumdar et al., 2015). However, patients with lower expression of TLR3 and IFN-γ progress to acute liver failure (Majumdar et al., 2015).

HEV can cause chronic infection as well. Although chronic HEV infection was initially reported only in immuno compromised persons, such as organ transplant recipients, patients receiving cancer chemotherapy and HIV-infected persons (Hoofnagle et al., 2012), latest reports show that chronic HEV infection also occurs in an immunocompetent individual with systemic lupus erythematosus (SLE; Grewal et al., 2014). However, since this kind of cases are rare, data available so far are not sufficient to consider this patient as an immunocompetent individual (Kamar and Izopet, 2014). In organ transplant recipients, the chronic course leads to persistent increases in levels of alanine aminotransferase, significant histological activity and fibrosis in some cases (Wedemeyer et al., 2012). HIV-infected individuals have higher positive rate of anti-HEV antibody than individuals without HIV infection (Wedemeyer et al., 2012).

Besides hepatitis, extrahepatic manifestations have been documented. Neurological disorders, such as polyradiculopathy, Guillain–Barré syndrome, bilateral brachial neuritis, encephalitis and proximal myopathy, and neuralgic amyotrophy are reported in patients with acute and chronic HEV infections (Kamar et al., 2011, 2014; van den Berg et al., 2014; van Eijk et al., 2014; Dalton et al., 2016; Drave et al., 2016). The kidney injury caused by HEV infection is reported and also documented in monkeys infected experimentally with HEV as well (Kamar et al., 2005, 2012b; Geng et al., 2016). Furthermore, a recent report provides evidence that extrahepatic replication of HEV in the placenta of infected mothers, which may be associated with fetal mortality (Bose et al., 2014). It also raises the concern for the vertical transmission of HEV to fetus and newborn by an infected mother (Krain et al., 2014).

A report suggests the association between the outcome of HEV infection in solid-organ transplant patients and the genetic heterogeneity of HEV quasispecies in ORF1 (Lhomme et al., 2014). Analysis of the viral genetic heterogeneity indicates that both nucleotide complexity and genetic distance of the ORF1 proline-rich domain in patients whose infection became chronic are higher than the patients who cleared the virus (Lhomme et al., 2014).

Although diagnostic tests for HEV are commercially available, none of them have been formally approved in the United States by the Food and Drug Administration (FDA; Hoofnagle et al., 2012). Current tests mainly target anti-HEV antibodies, including IgG and IgM. However, several assays are based on antigens expressed by a single HEV genotype, especially genotype 3, and might be limited for the detection of all HEV genotypes. Indeed, there are variations in sensitivity, specificity and agreement in the results of these assays, which may account for the discrepancies among positive rates of anti-HEV antibodies in various populations (Mast et al., 1998; Herremans et al., 2007; Drobeniuc et al., 2010). It is also notable that a recent study demonstrates the false positive result in HEV IgM test due to cross reaction with EBV and CMV, which heavily affects the accuracy of HEV serology testing (Hyams et al., 2014). Only 13.3% of the total samples with the positive HEV IgM were PCR positive for HEV RNA. The cross reactivity of IgM against HEV, EBV, and CMV is very high. These data suggest that to confirm HEV infection in patients, clinical features, blood ALT level and PCR testing should be all included in addition to serological test alone.

On the other hand, although HEV RNA can also be detected in blood and stool for several weeks after acute HEV infection, in addition to a narrow detectable window of HEV viremia (Hyams et al., 2014), current HEV RNA tests are still experimental since they have not been standardized yet (Wedemeyer et al., 2012). Furthermore, diagnostic support for IgM and IgG anti-HEV detection in clinical samples using commercially available kits and PCR assay for detection of HEV RNA in serum and stool samples are also available from the Division of Viral Hepatitis in the Centers for Disease Control and Prevention (CDC, 2016).

Hepatitis E virus infection mainly causes a self-limited disease and most infected individuals are able to clear it spontaneously. Although the case fatality rate in adults is 0.5–3%, the rate can increase to 30% in pregnant women during their third trimester of gestation in South Asia (Jameel, 1999). Therefore, antiviral therapy is needed. Although no specific treatment has been approved for HEV, off-label application of ribavirin as monotherapy for HEV has demonstrated promising results in both acute and chronic hepatitis E patients (Kamar et al., 2010; Mallet et al., 2010; Gerolami et al., 2011). In vitro assay showed that ribavirin could inhibit replication of genotypes 1–3 HEV through the depletion of intracellular GTP pools in HEV infected cells (Debing et al., 2014a). For immunosuppressed patients, a reduction of immunosuppression has shown efficacy in the treatment of chronic HEV infection (Wedemeyer et al., 2012). Moreover, application of pegylated interferon in combination with ribavirin has been reported as a treatment for chronic HEV infection but only shown moderately synergistic effect (Wedemeyer et al., 2012; Debing et al., 2014a). However, due to the evidence of embryolethality and teratogenicity revealed by animal study, ribavirin has been assigned to pregnancy category X by the FDA and contraindicated in women who are pregnant and in the male partners of women who are pregnant (Sayed et al., 2015). Moreover, ribavirin-induced G1634R mutation was reported and associated with treatment failure of ribavirin monotherapy in solid-organ transplant patients (Debing et al., 2014b; Lhomme et al., 2015; Todt et al., 2016). Therefore, viral specific treatment for HEV is needed.

Our laboratory has successfully tested application of peptide-conjugated morpholino oligomers (PPMOs) as novel anti-HEV compounds (Nan et al., 2015). PPMOs are water soluble, nuclease-resistant single-stranded DNA analogs containing a backbone of morpholine rings and phosphorodiamidate linkages along with conjugation of arginine-rich cell penetrating peptide for facilitating cell delivery (Summerton, 1999; Abes et al., 2006). PPMOs bind to mRNA by Watson–Crick base pairing and interfere with translation through steric blockade of the AUG-translation initiating region. Antisense morpholino oligomers are currently tested in clinical trials for treating Duchenne muscular dystrophy in humans and has been documented as effective against numerous types of viral infections in experimental animal models (Anthony et al., 2012; Mendell et al., 2013; Moulton, 2013). Importantly, upon systemic administration, PPMOs distribute to liver, remain pharmacologically viable, and are effective at reducing viral titers (Amantana et al., 2007; Burrer et al., 2007; Paessler et al., 2008). In our study, PPMO HP1 targeting 5′ UTR of HEV genotype 1 Sar55 strain demonstrates strong inhibition of HEV replication (Nan et al., 2015). Since the 5′ UTR of HEV genome is highly conserved among different HEV genotypes infecting humans, the PPMO HP1 may be an HEV-specific inhibitor with antiviral activity across multiple HEV genotypes (Nan et al., 2015). These qualities, along with the in vitro efficacy against HEV (Nan et al., 2015), make PPMOs be appealing for consideration as a novel inhibitor of HEV infections.

Another nucleic-acid based strategy, siRNA, has also been reported to be effective in inhibiting HEV replication. An siRNA targeting HEV RdRp was reported to inhibit HEV replication in A549 cells and in piglets (Huang et al., 2009). In another report, siRNA targeting a 3′ cis-acting element and viral nucleotide sequences coding for helicase and RdRp are effective against HEV in HepG2 cells (Kumar et al., 2010). However, it is generally acknowledged that siRNA needs considerable improvements in their delivery to relevant targets in vivo before they can be considered for clinical applications involving systemic delivery against virus infections.

Current prevention for HEV relies on sanitary measures, such as providing clean water, and appropriately cooked food to avoid transmission from undercooked food (Kamar et al., 2012a). Since in vitro culturing of HEV is limited and ineffective, HEV vaccine development mainly focuses on the expression of the capsid protein as a subunit vaccine. The capsid protein shares over 85% identity among the four major HEV genotypes in mammalian hosts (Mori and Matsuura, 2011). The capsid protein from genotype 1 HEV expressed by baculovirus or bacterial vectors has been tested in clinical trials. The first candidate was a 56 kDa protein expressed in insect cells. In a phase 2 trial in Nepal, the vaccine is well-tolerated and highly immunogenic, with 95% efficacy for protection against hepatitis E (Shrestha et al., 2007). The second vaccine, HEV239, encompasses aa 368–606 of ORF2 product, is a 26 kDa truncated protein expressed in E. coli (Li et al., 2005b). This vaccine is well-tolerated with an efficacy of 100% protection after three doses in a population tested in China, which included both men and women aged 16–65 years (Zhu et al., 2010). The HEV239 vaccine was approved and marketed in China in 2012. Whether it will be endorsed in other countries or how effective it is against all other genotypes of HEV infecting humans remains unknown. Moreover, a study shows that HEV239 vaccine could protect rabbits against homologous and heterologous HEV challenge (Liu et al., 2014; Zhang et al., 2015). A recent report demonstrates that a hybrid protein fusing protruding (P) domains from capsid proteins of both Norovirus (NoV) and HEV induces a higher antibody titer than either P domain alone (Wang et al., 2014a). Subunit vaccine candidates containing antigens of HEV, rotavirus, and astrovirus are reported as well (Xia et al., 2016).

More than 20 years have passed since the discovery and complete genome sequencing of HEV. Our understanding of HEV is still limited, though ongoing research continues to reveal more and more information about this virus. Currently, we know that HEV is not only a public health concern in developing countries as previously thought, but also a concern with a more complicated scenario in the developed countries. More and more animal reservoirs are revealed and we now understand that genotypes 3 and 4 HEV are zoonotic and foodborne pathogens. However, the cross-species transmission and host tropism of different HEV genotypes are still elusive. Current data imply certain viral proteins such as ORF1 product plays a role in the host tropism of HEV. Further investigation is needed to elucidate the basic biology of HEV.

On the one hand, although approved in China, the HEV239 vaccine is still unavailable to most of the world, despite the fact that serum surveillance indicates a high prevalence rate of HEV throughout the world. Moreover, recent discoveries about the antigenicity variation between HEV genotypes and quasi-enveloped viral particles hidden from neutralizing antibody suggest new challenges and questions about the efficacy of the approved vaccine. Further investigation about the vaccine efficacy against multiple HEV genotypes or seeking for an improved vaccine is needed. In addition, virus specific treatment for HEV infection is not available yet. Although, the off-label using of pegylated IFNs and antiviral drugs for general purposes have demonstrated efficacy against HEV, safety is still a concern as no validation has yet been conducted for these treatments. Therefore, a HEV-specific treatment such as PPMOs is needed.

Due to the absence of a suitable animal model and a simple cell culture system, many details about this virus and its infection, such as its biology, pathogenesis, strain variances, genotype differences, molecular mechanisms and vaccine efficacy for cross protection are still incomplete. However, current information also indicates that it is possible to establish a useful cell culture system using certain HEV strains such as the cell culture-adapted Kernow-C1 strain. These recent advances will facilitate further studies, which hopefully will reveal more insights about the basic biology of HEV, such as proteolytic processing of ORF1 product, functions of the viral proteins, HEV pathogenesis, effective therapeutics and a better vaccine.

All authors listed, have made substantial, direct and intellectual contribution to the work, and approved it for publication.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.