HEPATITIS E VIRUS ORF2 CAPSID POLYPEPTIDES AND USES THEREOF

Abstract

Hepatitis E virus (HEV) is responsible for over 50% of acute viral hepatitis cases worldwide. The inventors have now identified the precise sequence of infectious particle-associated ORF2 capsid protein. Strikingly, their analyses revealed that in infected patients, HEV produces three forms of the ORF2 capsid protein: ORF2i, ORF2g and ORF2c. The ORF2i protein is associated with infectious particles whereas ORF2g and ORF2c proteins are massively produced glycoproteins that are not associated with infectious particles and are the major antigens present in HEV-infected patient sera. Accordingly, the ORF2i and ORF2g proteins are thus the subject matter of the present invention as well as antibodies specific for the proteins and diagnostic assays (e.g. ELISA) for the diagnosis of Hepatitis E virus infection.

Claims

1. An isolated hepatitis E virus ORF2 capsid polypeptide comprising an ORF2i polypeptide with an amino acid sequence having at least 90% identity with the amino acid sequence as set forth in SEQ ID NO:1; or an ORF2g polypeptide with an amino acid sequence having at least 90% identity with the amino acid sequence as set forth in SEQ ID NO:2.

2. The ORF2i polypeptide of claim 1, which is not glycosylated.

3. (canceled)

4. The ORF2g polypeptide of claim 1, which is glycosylated.

5. A nucleic acid molecule encoding for the ORF2i polypeptide or the ORF2g polypeptide of claim 1.

6. An antibody specific for ORF2i polypeptide or the ORF2g polypeptide of claim 1.

7. The antibody of claim 6 which is a monoclonal antibody or a polyclonal antibody.

8. A method for detecting the presence of infectious particles of hepatitis E virus in a sample comprising contacting the sample with the antibody of claim 6 having specificity for the ORF2i polypeptide and/or the antibody of claim 6 having specificity for the ORF2g polypeptide under conditions that allow an immunocomplex of the antibody and the infectious particles to form, wherein detection of the immunocomplex indicates the presence of the infectious particles in the sample.

9. The method of claim 8 wherein the sample is selected from the group consisting of blood ascites; urine; saliva; sweat; milk; synovial fluid; peritoneal fluid; amniotic fluid; percerebrospinal fluid; lymph fluid; lung embolism; cerebrospinal fluid; pericardial fluid and faeces.

10. (canceled)

11. A method for diagnosing hepatitis E virus (HEV) infection in a subject in need thereof, comprising contacting the sample with the antibody of claim 6 having specificity for the ORF2i polypeptide under conditions that allow an immunocomplex of the antibody and infectious HEV particles to form, wherein detection of the immunocomplex indicates that the subject has an HEV infection.

12. The method of claim 11 wherein the HEV infection is an acute HEV infection, a recent HEV infection, a chronic HEV infection or a weak active HEV infection.

13. A method for diagnosing a cleared HEV infection in a subject in need thereof, comprising contacting a sample from the subject with the antibody of claim 6 having specificity for the ORF2g polypeptide under conditions that allow an immunocomplex of the antibody and the ORF2g polypeptide to form, wherein detection of the immunocomplex indicates that the subject has cleared an HEV infection.

14-18. (canceled)

Description

FIGURES

[0035] FIG. 1: Characterization of the different ORF2 products. (A) Detection by WB of ORF2 proteins in the top and pellet of the PLC3/HEV-p6 cell supernatant purified onto iodixanol cushion. (B) Comparison of ORF2 products expressed in the cell lysates and supernatant of PLC3/HEV-p6 cells, in infectious fraction 6 (F6p6) and non-infectious fraction 3 (F3p6) from iodixanol gradient, and in the pellet of PLC3/HEV-p6 cell supernatant purified on the iodixanol cushion. (C) PLC3/HEV-p6 cell lysate, supernatant and purified HEV particles were denatured and digested with indicated glycosidases (+) or not (). The dashed line shows the migration shift of ORF2 proteins following glycosidase treatment. (D) ORF2 protein detection after incubation of the supernatant of PLC3/HEV-p6 or PLC3 cells with GNA- or ProteinA-conjugated agarose beads. (E) PLC3/HEV-p6 cells were treated (+) or not () with Brefeldin A (BFA, 16 h, 1 g/ml) and the ORF2 protein expression in cells and supernatant was analyzed. (F) Sequence of ORF2 proteins analyzed by nanoLC-MS/MS. The dashed line corresponds to the signal peptide. Frames indicate potential N-glycosylation sites. The first aa of ORF2i and ORF2g identified by TMPP-Ac-OSu labeling are in bold.

[0036] FIG. 2: Characterization of the different ORF2 products. Viral particles purified on iodixanol cushion and ORF2g/ORF2c proteins immunoprecipitated with an anti-ORF2 antibody (4B2) were resolved by SDS-PAGE. Colloidal blue (CB) staining and WB are shown. The arrow indicates ORF2i (A), ORF2g (B) and ORF2c (C). The asterisk indicates human albumin. H and L indicate the heavy and light chains of immunoglobulins used in IP. ORF2 forms were digested in-gel with trypsin or AspN and analyzed by nanoLC-MS/MS. Peptide covering is highlighted in grey on the sequence of each ORF2 product. Dashed lines correspond to semi-trypsic and semi-AspN peptides. Lett.sup.14 and Ser.sup.34 in bold correspond to the first aa of ORF2i (A) and ORF2g (B), respectively, that were identified by semi-specific cleavage and TMPP-Ac-OSu labeling. No evidence was found for the N-terminal part of ORF2c (C).

[0037] FIG. 3: ORF2g and ORF2c proteins are the major HEV antigens in infected patients. (A) GNA pull-down on patient sera followed by ORF2 probing. Pull-down of PLC3/HEV-p6 supernatant was used as a positive control. (B) Fractionation on iodixanol gradient (7.5-40%) of the P6 patient serum and analysis of the ORF2 content of each fraction by WB and GNA pull-down. (C) HEV RNA levels in each fraction measured by RT-qPCR. (D) Detection of HEV Ag in each gradient fraction using the Wantai HEV-Ag ELISA.sup.Plus kit. Results are expressed as signal to cut-off ratios (S/CO).

[0038] FIG. 4: Western blotting experiments showing antibodies specificity.

EXAMPLE 1

[0039] Materials and Methods

[0040] Chemicals and Cell Cultures.

[0041] PLC/PRF/5 (CRL-8024), PLC1, PLC3 and A549 (CCL-185) cells were grown in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% inactivated fetal calf serum (DMEM/FCS) at 37 C. Transfected cells were maintained at 32 C. in a medium containing DMEM/M199 (1v:1v), 1 mg/ml of lipid-rich albumin (Albumax I) and 40 nM Na.sub.2SeO.sub.3.

[0042] Plasmids and Transfection.

[0043] Plasmids expressing the cell culture adapted gt3 Kernow C-1 strain (HEV-p6, GenBank accession number JQ679013) or the replicon expressing the Gaussia luciferase gene (HEV-p6GLuc) were provided by S.U Emerson.sup.6. The replication-deficient replicon HEV-p6GLucGAD was generated by mutating the GDD motif into the RNA-dependent RNA polymerase gene.sup.7. Capped RNA transcripts were generated with the mMESSAGE mMACHINE kit (Ambion). Capped RNAs were delivered to cells by electroporation using a Gene Pulser Xcell apparatus (Bio-Rad).

[0044] Kinetics Experiments and Virus Production.

[0045] PLC1 and PLC3 cells were electroporated with capped HEV-p6 RNA (20 g/410.sup.6 cells). For kinetics experiments, every two days, supernatants were harvested and cells were lysed in buffer containing 50 mM TrisHCl (pH 7.5), 150 mM NaCl, 5 mM EDTA, 0.5% (v/v) NP40, 0.05% sodium deoxycholate, 1 mM PMSF and protease inhibitor cocktail (Complete; Roche). For virus production, cells were split every 10 days. Supernatants were collected and stored at 80 C.

[0046] Density Gradients.

[0047] PLC3/HEV-p6 supernatant was concentrated with a Vivaspin ultrafiltration spin column (Sartorius). Concentrated supernatant or patient serum (500 l) was layered on a 7.5-40% iodixanol gradient, which was centrifuged at 160,000 g for 16 h at 4 C. Twelve fractions of 1 ml were collected and their density was measured by refractometry. The HEV RNA titer was determined by RT-qPCR. Each fraction was used to infect A549 cells. Infectivity was determined by indirect immunofluorescence, immunoblotting, and RNA and viral titration. For patient samples, fractions were used for ORF2 probing, GNA pull-down, and RNA quantification.

[0048] Purification of Infectious Particles on Iodixanol Cushion.

[0049] Supernatant of PLC3/HEV-p6 was concentrated by centrifugation through a 20% iodixanol cushion at 160,000 g for 4 h at 4 C. The cushion was resuspended in PBS, ultracentrifuged a second time, and the pellet was resuspended in PBS.

[0050] Infection of Humanized Mice.

[0051] Primary human hepatocytes were transplanted into homozygous uPA.sup.+/+-SCID mice, as previously described.sup.8. Humanized mice were inoculated via intrasplenic route with HEV-p6 purified on iodixanol gradient (fraction6, F6p6, 8.710.sup.7 IU/mouse) or with the same volume of fraction6 from a control gradient prepared with concentrated supernatant of non-transfected PLC3 cells (F6 control). Non-transplanted mice inoculated with these preparations served as negative controls. A chimeric mouse inoculated with a gt1 (Sar55, 2.810.sup.5 IU/mouse) stool suspension was used as a positive control.sup.9. Stool and plasma samples were collected on a weekly basis. Viral RNA was detected and quantified in mouse plasma and stool using RT-qPCR, as described.sup.9. Sequencing of the S17 region was performed as described.sup.9.

[0052] Transmission Electron Microscopy (TEM).

[0053] The formvar-carbon TEM grids (S162, Oxford Instruments) were either incubated with 0.01% poly-L-lysine for 30 min at room temperature (RT) or with poly-L-lysine and then with anti-ORF3 (Bioss antibodies), anti-ORF2 (1E6), or isotype-matched antibodies (20 m/ml) for 1 h at RT. Grids were washed in PBS and incubated for 2 h at RT with viral samples. TEM grids were washed in PBS and incubated for 20 min in 4% paraformaldehyde and 1% glutaraldehyde in 0.1M phosphate buffer, pH7.2. Particles trapped on grids were stained with 0.5% uranyl acetate before examination under a JEOL-1230 TEM. For delipidation, particles were either treated for 1 h at 37 C. with 1% sodium deoxycholic acid, 0.1% trypsin, then processed for TEM, or treated in the same conditions then layered on an iodixanol gradient. After determining density and RNA levels, fraction 11 was processed for TEM.

[0054] Glycosidase Digestions.

[0055] Protein samples were denaturated for 10 min at 95 C. in glycoprotein denaturing buffer (New England Biolabs). Digestions with glycosidases were carried out for 4 h at 37 C. in the presence of 1% NP40 and the buffer provided by the manufacturer (NEB). Samples prepared in the same conditions but without glycosidase were used as controls.

[0056] GNA Pull-Down.

[0057] Agarose-conjugated GNA beads were incubated for 2 h at RT with 50-500 l of patient serum or 100 l of supernatant of PLC3/HEV-p6. Beads were washed 8 times with PBS 0.5% NP40. Proteins were eluted in Laemmli Buffer and analyzed by SDS-PAGE. Unconjugated agarose beads were used as a negative control.

[0058] Patient Samples.

[0059] Patient samples were collected in France between 2012 and 2017. This was a non-interventional study. Samples were obtained only via standard viral diagnostics following a physician's order (no supplemental or modified sampling). Data were analyzed anonymously. The French Public Health Law (CSP Art L 1121-1.1) does not require written informed consent from patients for such a protocol.

[0060] Detection of HEV-Ag.

[0061] Gradient fractions were diluted in PBS and Ag levels were measured with the Wanta HEV-Ag ELISA.sup.Plus kit (Wanta Biological Pharmacy Enterprise) according to the manufacturer's instructions.

[0062] Results:

[0063] Establishment of an Efficient HEV Cell Culture System

[0064] In an attempt to establish a cell culture system for HEV, we first analyzed HEV replication in two subclones of the PLC/PRF/5 cell line, namely PLC1 and PLC3 cells. We used the Gaussia luciferase (GLuc)-expressing HEV gt3 replicon named HEV-p6GLuc.sup.6, in which the ORF2 coding sequence was replaced with the secreted GLuc sequence. Thus, the amount of secreted GLuc is proportional to viral RNA synthesis and consequently to HEV replication. PLC1, PLC3, and parental PLC/PRF/5 cells were electroporated with HEV-p6GLuc RNA or with a non-replicative HEV-p6GLuc genome (HEV-p6GLucGAD). The replication levels of HEV-p6GLuc steadily increased over time in the three cell lines (data not shown), but PLC1 and PLC3 cells showed higher HEV replication fold increase, as compared to PLC/PRF/5 cells. Experiments were thereafter conducted using PLC1 and PLC3 cells.

[0065] We assessed the ability of PLC1 and PLC3 cells to express viral proteins and produce infectious particles in kinetics experiments. PLC1 and PLC3 cells were electroporated with the HEV-p6 RNA strain.sup.6. Immunofluorescence with anti-ORF2 and anti-ORF3 antibodies showed that over 80% of cells expressed viral proteins (data not shown), indicating that PLC1 and PLC3 cells are highly transfectable, and that robust replication and expression of viral genome likely occur in these cells. It must be noted that ORF2 expression was inhibited in the presence of Sofosbuvir (SFV) (data not shown), confirming the specificity of the signal.

[0066] We next analyzed the expression of ORF2 and ORF3 proteins in the cell lysates and supernatants of HEV-p6 electroporated PLC1 and PLC3 cells by western blotting (WB) at different time points. Importantly, the ORF2 protein was detected in the cell lysates and supernatant of PLC1 cells as early as 2 and 4 days post-electroporation (p.e.), respectively (data not shown). In HEV-p6 electroporated PLC3 cells, expression of the ORF2 protein was slightly delayed. It must be noted that PLC3 cells grow more slowly than PLC1 cells, which could explain the observed differences. In addition to the major product of ORF2, ORF2-related proteins with lower molecular weights were also detected in cells and supernatants, indicating that the ORF2 capsid protein likely undergoes post-translational modifications. The ORF3 protein was also early detected at 2 and 4 days p.e. in PLC1 and PLC3 cell lysates, respectively (data not shown). In contrast, the ORF3 protein was faintly detected in the supernatants of electroporated cells, indicating that this protein is mainly expressed intracellularly.

[0067] The major ORF2 product in the supernatant (data not shown) displayed a higher apparent molecular weight than the intracellular form (data not shown), suggesting that the highly secreted ORF2 protein likely undergoes post-translational modifications. Comparison of intracellular and extracellular ORF3 proteins showed that the secreted ORF3 protein migrated slightly faster than the intracellular protein (data not shown) likely reflecting differences in protein phosphorylation.sup.10 or undescribed modifications of the ORF3 protein.

[0068] Viral RNA and infectious particles were also monitored in the supernatants of transfected cells collected on different days p.e. HEV RNA levels were assessed by RT-qPCR. High RNA levels were detected as early as 2 days p.e. (data not shown). RNA titers increased progressively and reached 1.110.sup.8 and 3.310.sup.7 copies/ml at day 10 for PLC1 and PLC3 cells, respectively (data not shown). In parallel, infectious viral titers were determined and reached 710.sup.3 and 1.510.sup.3 ffu/ml for PLC1 and PLC3 cells, respectively. These results indicate that assembly of infectious viral particles occurs very early and equally in HEV-p6 electroporated PLC1 and PLC3 cells. However, HEV infectivity represents 1.510.sup.4 RNA copies/ffu for PLC1 cells and 2.210.sup.4 RNA copies/ffu for PLC3 cells, indicating that the assembly of infectious HEV particles is likely an inefficient process. However, we cannot exclude the possibility that a viral or cellular inhibitor blocks HEV infection and biases the infectious titers. Further experiments were carried out using PLC3 cells.

[0069] ORF2 Capsid Protein is Massively Produced but Only a Small Fraction is Assembled into Infectious Particles.

[0070] To produce large amounts of infectious supernatants, we cultured HEV-p6 transfected PLC3 cells during 47 days. Supernatants were pooled, concentrated, and fractionated on an iodixanol gradient. The distribution of ORF2 protein, ORF3 protein, RNA and infectious viral particles was analyzed from each fraction (data not shown). ORF2 protein was detected from fractions 2 to 7 but more abundantly in fractions 3 to 5 (data not shown). As observed in the supernatants of transfected cells (FIG. 1C), ORF2 protein in fractions 2 to 5 was detected as two products, one major product of approximately 90 kDa (ORF2g) and a smaller product of around 75 kDa (ORF2c). ORF2-related proteins with lower molecular weights were also detected at the top of the gradient in fractions 2 to 5, which might correspond to additional processed forms of the proteins. In contrast, the ORF2 protein was mainly detected as an 80 kDa product (ORF2i) in fractions 6 and 7, which corresponds to the size of intracellular ORF2 (FIG. 1C). The ORF3 protein was exclusively detected in fractions 5 to 7 but most abundantly in fraction 6 (data not shown). Interestingly, only one major peak of RNA was detected in fraction 6, with a density of 1.10 g/ml (data not shown). Therefore, a large amount of capsid protein (fractions 2 to 4) was not associated with the viral RNA while the ORF3 protein was.

[0071] The infectivity of each fraction was analyzed by infecting A549 cells. Expression of ORF2 and ORF3 proteins was analyzed by WB (data not shown) and indirect immunofluorescence (data not shown) five days post-infection. ORF2i and ORF3 proteins were detected in the lysates of cells infected with fractions 5, 6 and 7 (data not shown), suggesting that infectious particles were associated with these fractions. Indeed, immunofluorescence staining of A549 cells inoculated with fractions 5, 6 and 7 were positive for ORF2 protein expression whereas cells inoculated with fractions 1 to 4 and fractions 8 to 12 were negative (data not shown). Fraction 6 showed the highest infectious titer (510.sup.6 ffu/ml). While ORF2g and ORF2c proteins were detected in cells inoculated with fractions 3 and 4 (data not shown), no specific infection was observed with these fractions by immunofluorescence (data not shown), indicating that ORF2g and ORF2c are likely very stable proteins binding to the cell surface that can be detected after several days of incubation.

[0072] It must be noted that the exosomal CD81 tetraspanin was detected in infectious fractions (data not shown), supporting the hypothesis that HEV particles likely exploit the exosomal secretory pathway for their egress.sup.11.

[0073] Together, our results indicate that during the HEV lifecycle, the ORF2 capsid protein is massively produced, but only a small fraction (ORF2i) is assembled into infectious particles that are secreted through the exosomal pathway.

[0074] Infection of Human Liver Chimeric Mice with HEVcc Particles.

[0075] Recently, human liver chimeric mice have been described as valuable models for studying in vivo chronic HEV infection and evaluating antiviral molecules.sup.9,12-14. Establishment of HEV infection in chimeric mice was achieved after inoculation of gt1 or gt3 virions. However, inoculation of mice with non-treated cell culture supernatant-derived particles did not result in robust infection. Therefore, we next evaluated whether our HEVcc particles were able to infect primary human hepatocyte (PHH)-transplanted mice.sup.9. Chimeric mice were inoculated via intrasplenic route with either fraction 6 of HEV-p6 (F6p6) purified on iodixanol gradient (FIG. 2) or fraction 6 of a control gradient prepared from non-transfected PLC3 cells supernatant (F6 control, neg 1) (data not shown). A human chimeric mouse inoculated with a gt1 stool suspension (Sar55) was used as a positive control.sup.9. A non-transplanted mouse inoculated with F6p6 was used as a second negative control (neg 2). While to a lesser extent than the gt1-infected mouse, the F6p6-inoculated chimeric mouse still showed signs of active infection (data not shown). Indeed, HEV RNA was detected in stools of F6p6-inoculated chimeric mouse from week 2 to 10 with a significant increase from week 8 to 10 (data not shown). Genomic HEV RNA was also detected in mouse liver (data not shown) and small intestine content (data not shown). RNA load in the F6p6-inoculated mouse increased more slowly as compared to the gt1-inoculated mouse, as previously described.sup.9 but for the first time, we demonstrate a robust infection of a chimeric mouse inoculated with HEV particles derived from cell culture supernatant.

[0076] Since HEV-p6 strain contains the human S17 ribosomal protein fragment that confers a growth advantage.sup.6,15, we next sequenced this region of HEV-p6 from the F6p6-infected mouse (stool week 9 p.i. and one liver piece) and compared it to the inoculum. We did not find any reversion of adaptive mutations in the S17 region (data not shown).

[0077] Ultrastructure of HEV Particles.

[0078] Recently, a new strategy based on the direct specific immunocapture (IC) of hepatitis C virus (HCV) particles on transmission electron microscopy (TEM) grids led, for the first time, to the precise description of their ultrastructure.sup.16. To define the morphology of HEVcc particles, which has never been precisely described, we used the same approach on isolated infectious particles. We first analyzed material captured on grids coated with poly-L lysine, a polycationic attachment factor. We observed a fairly homogeneous population of particles of 40-70 nm in size (data not shown). We next performed IC with an antibody specific to the ORF3 protein that had been described as exposed on lipid-associated HEV particles. We observed particles with an apparent internal icosahedral substructure likely corresponding to the capsid (data not shown). The calculation of size distribution showed that particle sizes ranged from 40 nm to 70 nm. In contrast, no particles were observed when an anti-ORF2 (1E6) (data not shown) or isotype control antibodies (data not shown) were used. However, when viral preparation was first partially delipidated by sodium deoxycholate and trypsin (DT) treatment before IC with the anti-ORF2 antibody, two populations of particles were immunocaptured: a major population consisting of particles of 30-50 nm in size and a more heterogeneous population with particles of 90-140 nm (data not shown). Thus, DT treatment unmasked the 1E6 epitope on viral particles and led to IC of small HEV particles that likely correspond to naked virions. However, DT treatment did not fully remove ORF3 and associated lipids since some particles were still captured by an anti-ORF3 antibody (data not shown). The larger size of these particles is likely due to dissociation of their lipid coats. In contrast, when DT-treated particles were purified on density gradient before IC (data not shown), a highly homogeneous population of particles of approximately 32 nm was immunocaptured with 1E6 while no particles were captured by IC anti-ORF3. These particles, with a density of 1.18 g/ml, displayed an icosahedral structure likely corresponding to naked capsids.

[0079] We next analyzed the ultrastructure of authentic particles from three HEV-infected patient sera (HEVser). As for HEVcc, no particles were observed by IC anti-ORF2 or IC with irrelevant antibodies (data not shown). In contrast, IC anti-ORF3 resulted in the efficient isolation of pleomorphic populations. HEVser1 and HEVserST-1 particles had a similar morphology to HEVcc particles while HEVser2 particles displayed a thick outer layer likely corresponding to lipids (data not shown). The calculation of size distribution of immunocaptured particles showed that the mean and median diameters of HEVser particles were generally larger than those of HEVcc particles and differed between patients (data not shown). Particles from HEVser2 were the largest with a mean diameter of 120 nm. These results indicate that the lipid content of particles likely determine particle size, as described for HCV particles.sup.16.

[0080] Lastly, we analyzed the ultrastructure of particles from an HEV-infected patient stool, as described above. In contrast to HEVcc and HEVser, stool particles were captured by IC anti-ORF2, whereas no particles were captured by IC anti-ORF3 (data not shown). Although particles appeared to be entangled in impurities limiting calculation of size distribution, the mean diameter of HEV stool particles was 28 nm (n=34, mean=28 nm, median=28 nm, SD=4 nm), which is in accordance with previous observations.sup.17. As for delipidated HEVcc, stool particles displayed an icosahedral structure likely corresponding to naked capsids.

[0081] Characterization of the Different Forms of ORF2 Protein.

[0082] We developed a purification system using a 20% iodixanol cushion to easily separate infectious particles (ORF2i, FIG. 1A, Pellet) from ORF2g/ORF2c proteins that are not associated with infectious particles (FIG. 1A, Top).

[0083] HEV produces large amounts of secreted ORF2g/ORF2c proteins that are likely glycosylated/processed forms of the ORF2 protein. In contrast, infectious particle-associated ORF2 protein (ORF2i) purified by iodixanol gradient (F6p6) or cushion (Pellet/Cushion) displayed the same size as intracellular ORF2 protein (Cells) (FIG. 1B), indicating that ORF2i likely does not undergo post-translational modifications.

[0084] Since the ORF2 protein sequence contains three potential sites for N-linked glycosylation and multiple sites for O-linked glycosylation, we next analyzed the sensitivity of ORF2 proteins to treatment with different glycosidases (FIG. 1C). Endoglycosidase H (EndoH) cleaves within the chitobiose core of high mannose and some hybrid oligosaccharides from N-glycoproteins. Peptide-N-Glycosidase F (PNGaseF) cleaves between the innermost N-acetyl glucosamine and asparagine residues of high mannose, hybrid and complex oligosaccharides from N-glycoproteins. O-Glycosidase (O-Gly) catalyzes the removal of some O-linked disaccharides. Neuraminidase (Neura) catalyzes the hydrolysis of sialic acid residues from N- and O-glycoproteins. It should be noted that terminal sialic acids block the action of O-Gly. The ORF2i protein expressed from cell lysates or purified HEV particles was resistant to glycosidase digestions (FIG. 1C), indicating that this protein is not N- or O-glycosylated. In contrast, secreted ORF2g/ORF2c proteins were sensitive to digestion with EndoH and PNGaseF glycosidases, as attested by the mobility shift upon treatment, indicating that these proteins are N-glycoproteins. ORF2g/ORF2c proteins were resistant to O-Gly treatment but sensitive to treatment with Neura and more markedly to a combination of Neura and O-Gly. These results indicate that ORF2g/ORF2c proteins are sialylated and O-glycosylated proteins. Importantly, ORF2g/ORF2c proteins were precipitated by a Galanthus nivalis Agglutinin (GNA) pull-down, a lectin that interacts with high-mannose-type glycans confirming the oligomannosidic nature of ORF2g/ORF2c N-glycans (FIG. 1D) and allowing the easy and specific precipitation of ORF2g/ORF2c proteins.

[0085] We demonstrated that secreted ORF2g/ORF2c are glycosylated proteins, indicating that these proteins go through the secretory pathway. However, ORF2g/ORF2c proteins were not detected at the intracellular level. We therefore hypothesized that ORF2g/ORF2c proteins might move rapidly through the secretory pathway and be secreted quickly without intracellular accumulation. We thus treated HEV-p6 expressing PLC3 cells with Brefeldin A (BFA), a protein secretion inhibitor.sup.18 (FIG. 1E). Upon BFA treatment, secretion of ORF2g/ORF2c proteins was blocked. Interestingly, a diffuse band likely corresponding to ORF2g and a faint band likely corresponding to ORF2c were detected in the lysate of BFA-treated cells, indicating that upon protein secretion blocking, ORF2g/ORF2c proteins accumulate in the cell, thus validating our hypothesis.

[0086] We next analyzed the sequence of ORF2 proteins by mass spectrometry. Viral particles purified on iodixanol cushion and ORF2g/ORF2c proteins immunoprecipitated with an anti-ORF2 antibody (4B2) were resolved by SDS-PAGE. The Colloidal blue stained bands corresponding to ORF2i, ORF2g and ORF2c in WB (FIG. 2) were digested in-gel with trypsin or AspN and then analyzed by nanoLC-MS/MS. Identified peptides and sequence covering are shown in FIG. 2. For the three proteins, the C-terminal end was fully covered, demonstrating the absence of processing at their C-termini. For ORF2i protein, a semi-tryptic peptide covering half of the signal peptide (SP) was unexpectedly identified (FIG. 2A, dashed line), suggesting that the SP of ORF2i might not be cleaved. For ORF2g, tryptic and semi-AspN peptides starting with Ser.sup.34 were identified, suggesting that the first aa of the ORF2g protein might correspond to Ser.sup.34 (FIG. 2B). For ORF2c, tryptic peptides starting with Iso.sup.234 were identified, suggesting that the first aa of the ORF2c protein might correspond to one of the aa in proximity to Iso.sup.234 (FIG. 2C). Since semi-tryptic and semi-AspN peptides correspond to either natural protein processing or non-specific proteolytic events during proteomics analyses, we further performed labeling with N-succinimidyloxycarbonylmethyl tris (2,4,6-trimethoxyphenyl) phosphonium bromide (TMPPAc-OSu), which binds specifically to the N-terminus of intact proteins.sup.19. Peptides identified with TMPP-Ac-OSu modification confirmed that the first aa of ORF2i and ORF2g corresponds to Leu.sup.14 and Ser.sup.34, respectively (FIGS. 1F, 2). In contrast, as TMPP-Ac-OSu labeling did not identify the first aa of ORF2c, further studies are necessary to clarify this observation. Together, these data indicate that the ORF2g protein loses its SP and is likely processed by secretory pathway proteases. In contrast, the ORF2i protein is not processed by a signal peptidase and therefore likely not translocated into the ER lumen.

[0087] ORF2g/ORF2c Proteins are the Major Antigens in HEV-Infected Patient Sera

[0088] Since we demonstrated that in cell culture, HEV mainly produces glycosylated forms of ORF2 proteins that are not associated with infectious particles, we next sought to determine whether the same occurs in infected patients. Sera from 17 infected HEV patients and from five HEV-negative patients were precipitated by GNA pull-down and probed for the ORF2 protein (FIG. 3A), as described above. GNA-pulldown of PLC3/HEV-p6 supernatant was used as a positive control. Strikingly, 13 and 7 out of 17 HEV-positive sera displayed large amounts of the ORF2g and ORF2c proteins, respectively. ORF2 proteins were not detected in HEV-negative sera. In addition, the detection of ORF2g/ORF2c proteins was dependent neither on the patient's HEV strain nor on the serum viral load (Table 1). Importantly, fractionation of a patient serum (P6 patient) on an iodixanol gradient followed by GNA pull-down of each fraction (FIG. 3B) demonstrated that, as in cell culture, very large amounts of ORF2g/ORF2c proteins were isolated in light fractions (fractions 4 and 5) and pull-down by GNA. However, these fractions were not associated with infectious material, since they were not associated with the viral genome (FIG. 3C). It should be noted that the ORF2 proteins present in fraction 4 were likely not precipitated by GNA beads due to the presence of high amounts of human albumin. Our results indicate that ORF2g/ORF2c proteins are likely the major antigens in HEV-infected patient sera. Therefore, we quantified ORF2 proteins in each fraction of the gradient with the Wanta HEV-antigen ELISA.sup.Plus assay (FIG. 3D). Interestingly, the highest amount of antigen was detected in fraction 4, containing the ORF2g/ORF2c proteins, whereas lower amounts of antigens were detected in the other fractions and notably in fraction 7, the infectious fraction. Together, our results indicate that in infected patients, HEV produces high amounts of glycosylated antigen protein that are not associated with infectious particles and might lead to a bias in diagnosis of active HEV infection.

TABLE-US-00003 TABLE 1 Features of HEV patient sera used in GNA pull-down Amount used Pa- Geno- Serum in GNA Ag ORF2 tients type titer .sup.a pull-down .sup.b levels .sup.c proteins .sup.d P1 3c 1.3 10.sup.8 6.5 10.sup.6 17.4 ORF2g P2 3c 3.7 10.sup.7 1.8 10.sup.6 19.4 ORF2g P3 3c 2.1 10.sup.7 2.1 10.sup.6 17.4 ORF2g P4 ND 2.0 10.sup.5 1.0 10.sup.5 20.2 ORF2g/ORF2c P5 3f 7.0 10.sup.5 3.5 10.sup.5 ND P6 3c 1.6 10.sup.8 1.6 10.sup.7 19.4 ORF2g/ORF2c P7 3c 1.2 10.sup.7 1.2 10.sup.6 18.6 ORF2g/ORF2c P8 3f 1.2 10.sup.5 0.6 10.sup.5 5 P9 3f 1.5 10.sup.5 0.7 10.sup.5 19.7 ORF2g/ORF2c P10 3f 1.2 10.sup.6 0.6 10.sup.6 18.9 ORF2g/ORF2c P11 ND ND ND 18.3 ORF2g/ORF2c P12 3f 2.0 10.sup.6 1.0 10.sup.6 ND ORF2g P13 3f 2.1 10.sup.6 1.0 10.sup.6 ND ORF2g P14 3e 1.0 10.sup.6 2.0 10.sup.5 ND P15 3 1.1 10.sup.6 2.0 10.sup.5 ND P16 3c ND ND ND ORF2g P17 3c ND ND ND ORF2g/ORF2c .sup.a in HEV RNA copies/ml .sup.b in HEV RNA copies .sup.c in S/CO values. Samples were diluted 100 times in PBS and quantified with the Wanta HEV-Ag-ELISA kit. .sup.d detected in GNA pull-down + WB anti-ORF2

[0089] Discussion:

[0090] By combining the highly replicative and cell culture-adapted p6 strain.sup.6 and highly transfectable subclones of PLC/PRF/5 cells, we developed a new cell culture system for which viral replication and protein expression were detected very early post-transfection. Time-course experiments showed that the ORF2 protein was early and massively secreted into the supernatant of transfected cells, as recently reported.sup.20,21. Interestingly, ORF2 and ORF3 proteins migrated differently when detected either in cell lysate or supernatant, indicating that these proteins likely undergo post-translational modifications during their secretion. Further experiments would be necessary to identify such modifications in the ORF3 protein. The ORF2 protein contains three potential sites for N-linked glycosylation and multiple sites for O-linked glycosylation. Using expression vectors, it has been previously shown that the ORF2 protein is glycosylated and expressed on the cell surface.sup.22-24, but it is not clear whether the glycosylated ORF2 protein is the natural form of the virion. Recently, it has been suggested that the HEV virion capsid is likely glycosylated.sup.20, although Graff et al. showed the opposite.sup.25. Our study revealed that HEV produces large amounts of ORF2 proteins, named ORF2g and ORF2c, that are secreted, sialylated, N- and O-glycosylated but are not associated with infectious virions. A large portion of ORF2 proteins is likely translocated into the ER lumen where they are N-glycosylated and likely processed by proteases to generate ORF2g and ORF2c proteins. These two proteins move rapidly through the secretory pathway where they are O-glycosylated and sialylated, then quickly secreted. Interestingly, the presence of RGRR residues upstream of the ORF2g N-terminus suggests that a furin-like protease might be involved in its maturation. Further experiments would be necessary to characterize glycosylation and processing sites in the ORF2 sequence, notably the mechanism leading to the production of the ORF2c protein. In contrast, the ORF2i protein seen on an intracellular level and in virions is likely not translocated into the ER lumen and stays in the cytosolic compartment. Our results suggest the existence of two production pathways for the HEV capsid protein: (i) a major non-productive pathway in which ORF2 proteins are delivered to the secretion route where they are processed and quickly secreted. (ii) a productive pathway in which cytosolic ORF2 proteins are delivered to the virion assembly sites. Further studies are needed to thoroughly investigate these pathways.

[0091] In contrast to previous studies.sup.9,13, we succeeded in infecting chimeric mice with gt3 HEVcc particles. The separation of infectious particles from ORF2g/ORF2c proteins by density gradient might explain our success, as these proteins interfere with the capacity of virions to infect target cells, using a mechanism which needs to be elucidated.

[0092] We found that during its lifecycle, HEV highly secretes glycosylated forms of the ORF2 protein, which circulate in infected patients and are the major antigens in patient sera. It would be interesting to define which form of ORF2 protein is recognized by antibodies from patients that have resolved their infections. HEV may produce ORF2g/ORF2c proteins as immunological bait. Interestingly, using TEM analyses, we found that ORF2g/ORF2c proteins do not form particulate material (data not shown) in contrast to other viruses such as hepatitis B virus, where the surface antigen forms subviral particles.

[0093] Our TEM analyses showed that HEVcc are particles of 40-70 nm in size displaying internal structures and the ORF3 protein at their surface. DT treatment and ultracentrifugation uncloaked small icosahedral capsids similar to those found in patient stools (.sup.17 and our study), indicating that HEVcc are associated with lipids in which ORF3 is embedded. Our analyses showed that, although much larger, HEVser particles have a morphology similar to HEVcc particles. HEVser virions display the ORF3 protein at their surface and are likely highly associated with lipids. As HEVser diameters differed between patients, it would be interesting to determine if the observed size variations are related to lipid content, as recently demonstrated for HCV particles.sup.16.

[0094] Recently, the Wanta HEV-Ag-ELISA assay was marketed for diagnosing HEV infection. Comparative studies of this assay showed high levels of specificity and sensitivity 26-29. However, some discrepancies between antigen detection and RNA quantification were found in some samples in which the HEV antigen was detected in the absence of detectable RNA.sup.27-29. In our hands, the antigen assay was also very sensitive. However, we found that ORF2g/ORF2c proteins, which are not associated with infectious virions, were the main antigens recognized by the kit. Together, these data indicate that the results obtained with the HEV-Ag-ELISA assay must be cautiously interpreted. Indeed, this assay might not necessarily detect an active infection, since ORF2g/ORF2c proteins seem to be very stable in cell culture but likely also in patient sera.sup.28, suggesting that these proteins might remain in patient sera even when HEV is no longer replicating in the liver.

EXAMPLE 2

[0095] Generation of Specific Antibodies Directed Against the ORF2i Protein.

[0096] A peptide (P1) that derives from the polypeptide of the present invention (ORF2i) was synthetised and coupled to KLH. Five mice were immunized three times at three weeks intervals with the P1 peptide. Freund's complete and incomplete adjuvants were used during immunisation. The animals were immunized by subcutaneous and intraperitoneal routes. Ten days after the third immunisation, mice have been bleeded and their sera tested for immunoreactivity. Sera were first assayed by indirect ELISA on plates coated with the P1 peptide (data not shown). Their specificity was next analyzed in western blotting experiments with a mixture of ORF2 proteins (ORF2i and ORF2g/ORF2c) as antigens (FIG. 4). Two (P1S2 and P1S5) out of five sera showed a highly specific recognition of the ORF2i protein, with no cross-reaction with the ORF2g/c proteins, as compared to the 1E6 antibody that recognizes the three forms (FIG. 4). Cloning and production of monoclonal antibodies are ongoing.

[0097] These results indicate that antibodies specifically directed against the ORF2i polypeptide can be produced. Such antibodies will be very suitable for determining presence of infectious particles of hepatitis E virus in a sample. More particularly, detection of the ORF2i or ORF2g polypeptide of the present invention is suitable for diagnosing hepatitis E virus infection in a subject.

REFERENCES

[0098] Throughout this application, various references describe the state of the art to which this invention pertains. The disclosures of these references are hereby incorporated by reference into the present disclosure. [0099] 1. Debing Y, Moradpour D, Neyts J, et al. Update on hepatitis E virology: Implications for clinical practice. Journal of Hepatology 2016; 65:200-212. [0100] 2. Pischke S, Hartl J, Pas S D, et al. Hepatitis E virus: Infection beyond the liver? Journal of Hepatology 2017; 66:1082-1095. [0101] 3. Khuroo M S, Khuroo M S. Hepatitis E: an emerging global diseasefrom discovery towards control and cure. Journal of Viral Hepatitis 2016; 23:68-79. [0102] 4. Feng Z, Hensley L, McKnight K L, et al. A pathogenic picornavirus acquires an envelope by hijacking cellular membranes. Nature 2013; 496:367-371. [0103] 5. Okamoto H. Culture systems for hepatitis E virus. J Gastroenterol 2012; 48:147-158. [0104] 6. Shukla P, Nguyen H T, Faulk K, et al. Adaptation of a genotype 3 hepatitis E virus to efficient growth in cell culture depends on an inserted human gene segment acquired by recombination. J Virol 2012; 86:5697-5707. [0105] 7. Emerson S U, Nguyen H, Graff J, et al. In vitro replication of hepatitis E virus (HEV) genomes and of an HEV replicon expressing green fluorescent protein. J Virol 2004; 78:4838-4846. [0106] 8. Meuleman P, Libbrecht L, De Vos R, et al. Morphological and biochemical characterization of a human liver in a uPA-SCID mouse chimera. Hepatology 2005; 41:847-856. [0107] 9. Sayed I M, Verhoye L, Cocquerel L, et al. Study of hepatitis E virus infection of genotype 1 and 3 in mice with humanised liver. Gut 2017; 66:920-929. [0108] 10. Zafrullah M, Ozdener M H, Panda S K, et al. The ORF3 protein of hepatitis E virus is a phosphoprotein that associates with the cytoskeleton. J Virol 1997; 71:9045-9053. [0109] 11. Nagashima S, Jirintai S, Takahashi M, et al. Hepatitis E virus egress depends on the exosomal pathway, with secretory exosomes derived from multivesicular bodies. Journal of General Virology 2014; 95:2166-2175. [0110] 12. Allweiss L, Gass S, Giersch K, et al. Human liver chimeric mice as a new model of chronic hepatitis E virus infection and preclinical drug evaluation. Journal of Hepatology 2016; 64:1033-1040. [0111] 13. van de Garde MDB, Pas S D, van der Net G, et al. Hepatitis E Virus (HEV) Genotype 3 Infection of Human Liver Chimeric Mice as a Model for Chronic HEV Infection. J Virol 2016; 90:4394-4401. [0112] 14. Sayed I M, Foquet L, Verhoye L, et al. Antiviral Research. Antiviral Research 2017; 141:150-154. [0113] 15. Shukla P, Nguyen H T, Torian U, et al. Cross-species infections of cultured cells by hepatitis E virus and discovery of an infectious virus-host recombinant. Proc. Natl. Acad. Sci. U.S.A. 2011; 108:2438-2443. [0114] 16. Piver E, Boyer A, Gaillard J, et al. Ultrastructural organisation of HCV from the bloodstream of infected patients revealed by electron microscopy after specific immunocapture. Gut 2016. [0115] 17. Balayan M S, Andjaparidze A G, Savinskaya S S, et al. Evidence for a virus in non-A, non-B hepatitis transmitted via the fecal-oral route. Intervirology 1983; 20:23-31. [0116] 18. Klausner R D, Donaldson J G, Lippincott-Schwartz J. Brefeldin A: insights into the control of membrane traffic and organelle structure. J Cell Biol 1992; 116:1071-1080. [0117] 19. Gallien S, Perrodou E, Carapito C, et al. Ortho-proteogenomics: Multiple proteomes investigation through orthology and a new M S-based protocol. Genome Research 2008; 19:128-135. [0118] 20. Qi Y, Zhang F, Zhang L, et al. Hepatitis E Virus Produced from Cell Culture Has a Lipid Envelope. PLoS ONE 2015; 10:e0132503. [0119] 21. Shiota T, Li T C, Yoshizaki S, et al. Establishment of hepatitis E virus infection-permissive and -non-permissive human hepatoma PLC/PRF/5 subclones. Microbiol. Immunol. 2015; 59:89-94. [0120] 22. Jameel S, Zafrullah M, Ozdener M H, et al. Expression in animal cells and characterization of the hepatitis E virus structural proteins. J Virol 1996; 70:207-216. [0121] 23. Torresi J, Meanger J, Lambert P, et al. High level expression of the capsid protein of hepatitis E virus in diverse eukaryotic cells using the Semliki Forest virus replicon. J. Virol. Methods 1997; 69:81-91. [0122] 24. Zafrullah M, Ozdener M H, Kumar R, et al. Mutational analysis of glycosylation, membrane translocation, and cell surface expression of the hepatitis E virus ORF2 protein. J Virol 1999; 73:4074-4082. [0123] 25. Graff J, Zhou Y-H, Torian U, et al. Mutations within potential glycosylation sites in the capsid protein of hepatitis E virus prevent the formation of infectious virus particles. J Virol 2008; 82:1185-1194. [0124] 26. Trmeaux P, Lhomme S, Chapuy-Regaud S, et al. Performance of an antigen assay for diagnosing acute hepatitis E virus genotype 3 infection. J. Clin. Virol. 2016; 79:1-5. [0125] 27. Zhao C, Geng Y, Harrison T J, et al. Evaluation of an antigen-capture EIA for the diagnosis of hepatitis E virus infection. Journal of Viral Hepatitis 2015; 22:957-963. [0126] 28. Behrendt P, Bremer B, Todt D, et al. Hepatitis E Virus (HEV) ORF2 Antigen Levels Differentiate Between Acute and Chronic HEV Infection. Journal of Infectious Diseases 2016; 214:361-368. [0127] 29. Geng Y, Zhao C, Huang W, et al. Detection and assessment of infectivity of hepatitis E virus in urine. Journal of Hepatology 2016; 64:37-43.