Application of Truncated Immunodominant Polypeptide from Hepatitis E Virus (HEV) ORF2 in an Assay To Exclude Nonspecific Binding in Detecting Anti-HEV Immunoglobulin M

Serum samples.In total, 313 serum samples were included in the present study and were divided into three groups. One group comprised the sera from 159 healthy persons who were negative for anti-HEV IgM, as detected with a Genelabs kit (Genelabs Diagnostics Pte. Ltd., Singapore Science Park, Singapore), and negative for IgM antibodies against hepatitis A virus and hepatitis B virus core antigen and for IgG antibody against hepatitis C virus. The second group was composed of samples from 37 acute hepatitis E patients who were diagnosed by the detection of HEV RNA (7). The third group included 117 sera which were positive for anti-HEV IgM, repeatedly detected with a Genelabs kit. All sera were stored at −70°C. This study was approved by the institutional review boards of the Nanjing Drum Tower Hospital.

Cloning and expression of HEV ORF2 fragment.Total RNA was extracted from 200 μl serum from a hepatitis E patient with positive anti-HEV IgM, using Trizol (Invitrogen, CA). The ORF2 fragment was cloned by RT-PCR. The cloning primers were U1stF and U1stR as the outer pair and U2ndF and U2ndR as the inner pair (Table 1). The purified PCR products were inserted into T-tailed pGEM-T vectors (Promega, WI). Genotypes were determined by DNA sequencing.

Three fragments of different sizes, covering aa 459 to 607, 472 to 607, and 459 to 594 (residue positions refer to the prototype HEV sequence [GenBank accession number AF444003]) in ORF2, were amplified by PCR with primers containing a PstI or EcoRI site (Table 1). The enzyme-digested PCR products were ligated to the expression vector pRSET-C (Invitrogen), which has a sequence coding for six histidine residues (His₆) at the N terminus of the recombinant polypeptide for purification. The inserts in the plasmids isolated from single colonies were sequenced. Escherichia coli BL21(DE3) (Invitrogen) was transformed with each of the ORF2 recombinant plasmids, carrying the above three ORF2 fragments, cultured in LB overnight, and then subcultured in super optimal broth (SOB) at 37°C for the expression of ORF2. Bacterial pellets were subjected to SDS-PAGE to examine the polypeptides.

The recombinant polypeptides were purified by nickel-chelation affinity chromatography via the His₆ tag. Briefly, the bacterial pellets were lysed by sonication in guanidinium lysis buffer (6 M guanidine-HCl, 100 mM NaH₂PO₄, 10 mM Tris-Cl, pH 8.0). The polypeptides bound to the nickel were eluted with denaturing elution buffer (8 M urea, 100 mM NaH₂PO₄, 10 mM Tris-Cl, pH 3.5). Each polypeptide was checked by SDS-PAGE, and the concentration was determined with a bicinchoninic acid (BCA) assay (Pierce, IL). Prior to use in ELISA, the polypeptides were diluted with carbonate-bicarbonate buffer (pH 9.6) for renaturation.

Western blot analysis of anti-HEV in patient sera.The recombinant ORF2 polypeptides containing aa 459 to 607, 472 to 607, and 459 to 594 were separated by 12% SDS-PAGE and transferred to a polyvinylidene difluoride (PVDF) membrane. The membranes were blocked with 5% skim milk in 0.05% Tween 20 in phosphate-buffered saline (PBS-T) at room temperature for 1 h and then sliced into three samples. The first was incubated with peroxidase-labeled anti-His₆ (1:1,000 dilution) for 1 h. The others were incubated for 3 h with serum samples diluted 100-fold in 5% skim milk. After four washes, horseradish peroxidase (HRP)-conjugated anti-human IgG or IgM (1:2,000) was incubated with the membranes for 1 h to detect the bound antibodies. The immunoreactivity was detected using a DAB enhanced liquid substrate system (Sigma, MO).

Enzyme-linked immunosorbent assay.An indirect ELISA was developed to detect anti-HEV IgG and IgM in sera of patients. ELISA plate wells were separately coated with 100 μl purified ORF2 polypeptides 459-607 and 472-607, diluted to 1 μg/ml in carbonate-bicarbonate buffer (pH 9.6), overnight at 4°C. The wells were washed twice with PBS-T and then blocked with 300 μl 5% skim milk in PBS-T at 37°C for 1 h. After three washes, 100 μl serum diluted 100-fold in 5% skim milk was added to the plates, followed by incubation at 37°C for 1 h. After another five washes, 100 μl HRP-conjugated anti-human IgG or IgM (Sigma), diluted 2,000-fold in PBS-T, was added to detect the bound antibodies. Following incubation at 37°C for 0.5 h, the plates were washed as described above, and 100 μl of tetramethylbenzidine solution (Sigma) was added to the wells. After incubation at room temperature for 15 min, the color development was stopped by adding 50 μl of 2 M H₂SO₄. The optical density at 450 nm (OD₄₅₀) was determined on a microplate reader (Bio-Rad, CA). The OD₄₅₀ in the well coated with the immunodominant polypeptide (aa 459 to 607) of ORF2 was denoted OD_IP, and that in the well coated with the truncated immunodominant polypeptide (aa 472 to 607) was designated OD_TP.

Anti-HEV IgM detected by commercially available kits.Commercially available ELISA kits for anti-HEV IgM were obtained from Genelabs (Genelabs Diagnostics Pte. Ltd., Singapore Science Park, Singapore) and Wantai (Wantai Biological Pharmacy Enterprise Co., Ltd., Beijing, China). The Genelabs kit is an indirect ELISA based on recombinant and synthetic peptides derived from ORF2 and ORF3 of genotypes 1 and 2 (6, 26). The Wantai kit is a capture ELISA, in which anti-human IgM (μ chain specific) is used to coat the plate and HRP-conjugated ORF2 polypeptide (genotype 1) is applied to detect the bound anti-HEV IgM; the polypeptide covers aa 394 to 606 in ORF2 (4, 28). The cutoff values were calculated per the manufacturers' instructions.

Production of recombinant ORF2 polypeptides.To obtain the HEV ORF2 gene, we amplified the fragment of interest, using RT-PCR, from four sera collected from hepatitis E patients positive for anti-HEV IgM (detected with a Genelabs kit). One of the sera was positive, and the resultant products were inserted into the pGEM-T vector and sequenced. BLAST searching in GenBank showed that the amplified sequence had the highest similarity (95.4% identity) to the corresponding region of an HEV (genotype 4) previously isolated in China (GenBank accession number EF077630). Thus, we cloned a genotype 4 ORF2 fragment.

The ORF2 protein of HEV genotype 1 has been characterized extensively through use of the recombinant polypeptide (10, 11). It may form homodimers and even higher-order oligomers (15, 23, 27), and the C-terminal portion is critical for dimerization (15, 27). The C-terminal portion covering aa 459 to 607 constitutes the neutralization and immunodominant epitopes (29, 30). However, the ORF2 polypeptide of genotype 4 is less well studied. To characterize the ORF2 polypeptide of genotype 4, we produced the corresponding ORF2 polypeptide and N- and C-terminally truncated polypeptides in E. coli, as previously described (29). Although the ORF2 polypeptide of HEV genotype 4 is putatively considered to have an additional 11 to 14 residues at the N terminus (22, 24, 25), we designated the aa positions according to the sequence of genotype 1 so that it is convenient to make comparisons between the different genotypes.

Figure 1A shows that three polypeptides, covering aa 459 to 607, 472 to 607, and 459 to 594, were all produced in E. coli. After purification with a nickel-affinity column, each polypeptide showed a single band in SDS-PAGE when the sample was heated at 95°C for 5 min in Laemmli buffer (Fig. 1B). Approximately 3 mg of each purified polypeptide was obtained from 100 ml of bacterial culture.

• Open in new tab
• Download powerpoint

FIG. 1.

Expression and purification of histidine-tagged HEV ORF2 polypeptides in E. coli. (A) SDS-PAGE of polypeptides spanning amino acids 459 to 607, 472 to 607, and 459 to 594. 0 h and 4 h, 0 and 4 h after adding IPTG (isopropyl-β-d-thiogalactopyranoside). (B) Standard SDS-PAGE of purified polypeptides. (C) SDS-PAGE of purified polypeptides, in Laemmli buffer, which were not heated.

Dimeric characterization of ORF2 polypeptide.Previous studies have shown that the genotype 1 ORF2 polypeptide forms homodimers and that the C-terminal portion is responsible for dimerization (15, 27). To observe whether the dimeric property is common in the genotype 4 ORF2 polypeptide, we analyzed the three polypeptides, 459-607, 472-607, and 459-594, by SDS-PAGE under different conditions. When polypeptide 459-607 was heated in Laemmli buffer, it showed a single band at the expected size of 21 kDa (Fig. 1B); however, when the polypeptide was not heated, it presented a band of the expected size (21 kDa) as well as another band corresponding to a high-molecular-mass polypeptide (42 kDa) (Fig. 1C), indicating that this band was the homodimer. On the other hand, polypeptides 459-594 and 472-607 showed single bands at the positions of the expected sizes, regardless of whether they were heated or not heated (Fig. 1B and C), suggesting that the truncated polypeptides did not form dimers. Thus, the dimerization of genotype 4 ORF2 requires aa 459 to 607 in the ORF2 protein, which is generally in accordance with the region of the genotype 1 polypeptide reported previously (15).

Characterization of antigenicity of ORF2 polypeptide.Since polypeptide 459-607 presented both monomers and dimers when it was not heated (Fig. 1C), we attempted to see whether anti-HEV from hepatitis E patients reacted with the dimeric or monomeric form or with both. After the polypeptides were transferred to the membrane, anti-HEV serum was added and then the blot was probed with enzyme-conjugated anti-human IgG and IgM. As controls, Fig. 2A presents the reactivities of anti-His₆ to the recombinant polypeptides, since each of them had a His₆ tag at the N terminus. Figures 2B and C show that both anti-HEV IgM and IgG naturally produced in the patients reacted only with the dimer, not with the monomer, of polypeptide 459-607. Furthermore, anti-HEV did not react with polypeptides 472-607 and 459-594, which did not form dimers, as demonstrated in Fig. 1. We tested 13 sera from acute hepatitis E patients, and all showed the same results. Thus, dimerization of the polypeptide spanning aa 459 to 607 in ORF2 is essential for formation of the epitopes recognized by natural infection-induced anti-HEV directed against this region.

• Open in new tab
• Download powerpoint

FIG. 2.

Antigenic characterization of ORF2 polypeptides by Western blot analysis. His₆-tagged polypeptides in Laemmli buffer were not heated before being run in the gel. Lanes 1 to 3 indicate the polypeptides covering amino acids 459 to 607, 472 to 607, and 459 to 594 in ORF2, respectively. The color was developed by the DAB enhanced liquid substrate system. (A) The membrane was probed with peroxidase-labeled anti-His₆. (B and C) The membranes were incubated with the serum from a hepatitis E patient and then probed with peroxidase-labeled anti-human IgG (B) and IgM (C).

We further analyzed the antigenicity of these polypeptides under natural conditions by using ELISA. We developed an indirect ELISA based on polypeptides 459-607 and 472-607 to observe the immunoreactivities of anti-HEV IgG and IgM, respectively. All 37 sera positive for HEV RNA reacted with polypeptide 459-607 but not with polypeptide 472-607 in the test for both anti-HEV IgG and IgM. Table 2 shows the representative OD₄₅₀ values for 10 samples. The data also demonstrated that anti-HEV IgG and IgM reacted only with polypeptide 459-607, not with polypeptide 472-607, under natural conditions, in agreement with the results of Western blot analyses.

Exclusion of false-positive anti-HEV IgM results by use of polypeptide 472-607.Based on the poor reactivity of anti-HEV to polypeptide 472-607, we assumed that the truncated polypeptide may serve as a “negative polypeptide” to rule out nonspecific binding in the detection of anti-HEV IgM. We used the indirect ELISA to measure anti-HEV IgG and IgM in the 159 sera that were negative for anti-HEV IgM by a Genelabs kit. None of them reacted with either polypeptide in the detection of IgM; however, in the detection of IgG, 19 (11.9%) sera reacted with polypeptide 459-607 but not with polypeptide 472-609. Representative results are shown in Table 2. To set the cutoff value for anti-HEV IgM in the ELISA based on polypeptide 459-607, we chose twice the average plus 2 standard deviations of the OD₄₅₀ values of the negative-control samples. Statistically, the mean OD₄₅₀ value for the 159 negative samples was 0.172 ± 0.027, and thus, a sample with an OD₄₅₀ of >0.400 was preliminarily judged to be immunoreactive. Since all sera positive for HEV RNA reacted with the immunodominant polypeptide but not with the truncated polypeptide (Fig. 2 and Table 2), we applied the truncated polypeptide as an “irrelevant polypeptide” in the ELISA. As a complementary criterion, a serum with a ratio of OD_IP to OD_TP of >2 was considered to be specifically immunoreactive, as described elsewhere (12); when the ratio was <2, a value for OD_IP minus OD_TP of >0.5 was also considered to indicate a specifically immunoreactive serum.

We used the ELISA developed in this study to retest anti-HEV IgM in 117 serum samples which were positive for anti-HEV IgM, as detected by a Genelabs kit. Surprisingly, only 34 serum samples were positive for IgM, and the other 83 samples were negative. Since the assay developed in this study was an indirect ELISA and the Genelabs kit is also an indirect ELISA, the high proportion of discrepancy led us to retest these samples with an assay based on different principles. Thus, we measured anti-HEV IgM in these 117 sera with a Wantai ELISA kit and found that 35 of them were positive and the other 82 were negative. There were 32 samples which showed positive results and 80 samples which were negative in both the ELISA in this study and the Wantai ELISA kit.

To clarify whether the sera which had discordant results in different assays were positive or negative for anti-HEV IgM, we performed Western blotting to analyze the reactivities of 43 discordant sera and 14 dual-positive sera to polypeptides 459-607 and 472-607. While all dual-positive sera reacted with the dimer of polypeptide 459-607, none of the discordant sera showed a positive reaction (data not shown). Additionally, we checked the alanine aminotransferase (ALT) levels of these patients and found no patients with abnormal results (data not shown). Together, these data strongly suggest that the positive results for discordant samples detected by the Genelabs kit were caused by nonspecific binding.

In measuring the IgM level in 117 sera with the ELISA used in this study, we also found that eight sera had higher OD values (OD₄₅₀ of >0.4) in the plate wells coated with polypeptide 459-607 and polypeptide 472-607 (Table 3); however, only three, E4, E14, and E96, had an OD_IP markedly higher than the OD_TP (OD_IP minus OD_TP > 0.5). To further determine the true status of these eight samples, we used Western blotting to analyze anti-HEV IgM; the results showed that only the three sera with values for OD_IP minus OD_TP of >0.5 reacted with the dimer of polypeptide 459-607 and that the five others did not react with either of the polypeptides (Fig. 3), highly indicating that E4, E14, and E96 were truly positive and that the others were negative for anti-HEV IgM.

• Open in new tab
• Download powerpoint

FIG. 3.

Western blot analysis of eight sera that reacted with both immunodominant and truncated ORF2 polypeptides. His₆-tagged polypeptides in Laemmli buffer were not heated before being run in the gel. Lanes 1 and 2, histidine-tagged polypeptides covering aa 459 to 607 and aa 472 to 607 in ORF2, respectively. Anti-His₆, the membranes were probed with peroxidase-labeled anti-His₆; E04 to E90, the membranes were incubated with each of the sera and then probed with peroxidase-labeled anti-human IgM. The color was developed by the DAB enhanced liquid substrate system.

Additionally, we performed other experiments to gain more evidence to support the above results. Since specific IgM and IgG immune responses to HEV occur early in the acute phase of the infection, anti-HEV IgM and IgG may be detected simultaneously in the vast majority of acute infections (21). Thus, we tested anti-HEV IgG in these eight sera and found that all three IgM-positive sera were also positive for IgG, whereas only one of the five IgM-negative sera was positive for IgG (Table 3). Measurement of ALT levels in the eight sera showed that only E4, E14, and E96 were abnormal, while five others had no elevation (Table 3). Moreover, sera E14 and E96 had detectable HEV RNA, yet none of the five other sera were positive. Together, these results are supportive of the above findings that among the eight samples with high OD values, only the three sera with OD₄₅₀ differences of >0.5 were truly positive, while the other five were negative for anti-HEV IgM, indicating that the truncated polypeptide 472-607 may be used to exclude false-positive results caused by nonspecific binding.