In Vitro and In Vivo Mutational Analysis of the 3'
http://www.100md.com
病菌学杂志 2005年第2期
Molecular Hepatitis
Hepatitis Viruses Sections, Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland
Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University, Ames, Iowa
Bioqual, Incorporated, Rockville, Maryland
ABSTRACT
Hepatitis E virus (HEV) replication is not well understood, mainly because the virus does not infect cultured cells efficiently. However, Huh-7 cells transfected with full-length genomes produce open reading frame 2 protein, indicative of genome replication (6). To investigate the role of 3'-terminal sequences in RNA replication, we constructed chimeric full-length genomes with divergent 3'-terminal sequences of genotypes 2 and 3 replacing that of genotype 1 and transfected them into Huh-7 cells. The production of viral proteins by these full-length chimeras was indistinguishable from that of the wild type, suggesting that replication was not impaired. In order to better quantify HEV replication in cell culture, we constructed an HEV replicon with a reporter (luciferase). Luciferase production was cap dependent and RNA-dependent RNA polymerase dependent and increased following transfection of Huh-7 cells. Replicons harboring the 3'-terminal intergenotypic chimera sequences were also assayed for luciferase production. In spite of the large sequence differences among the 3' termini of the viruses, replication of the chimeric replicons was surprisingly similar to that of the parental replicon. However, a single unique nucleotide change within a predicted stem structure at the 3' terminus substantially reduced the efficiency of replication: RNA replication was partially restored by a covariant mutation. Similar patterns of replication were obtained when full-length genomes were inoculated into rhesus macaques, suggesting that the in vitro system could be used to predict the effect of 3'-terminal mutations in vivo. Incorporation of the 3'-terminal sequences of the swine strain of HEV into the genotype 1 human strain did not enable the human strain to infect swine.
INTRODUCTION
Hepatitis E virus (HEV) is a major cause of enterically transmitted hepatitis, especially in Asia and Africa. Serological prevalence data suggest that HEV might be endemic in industrialized countries as well, although it rarely causes disease in these countries (reviewed in references 7, 19, and 29). Antibodies to HEV (anti-HEV) have also been detected in a wide variety of animals, and HEV has been found in swine (2, 10, 12, 25, 33), deer (35), and chickens (11, 14); these findings raise the possibility that HEV may be a zoonosis (22). HEV strains infecting mammals differ at the nucleotide level by as much as 27%, whereas the avian strain differs from the mammalian strains by up to 50%. The mammalian HEV strains have been classified into four genotypes based on sequence variations (31), but they appear to comprise a single serotype (29, 45).
HEV differs enough from other viruses that it was recently classified as the sole member of the new genus Hepevirus (5). HEV is nonenveloped and contains a single-stranded RNA genome of positive sense that is capped and polyadenylated. The 7.2-kb genome contains three partially overlapping open reading frames (ORFs). ORF1 encodes a nonstructural polyprotein which provides guanylyl-methyltransferase and RNA-dependent RNA polymerase (RdRp) activities and possibly other functions (17, 21). The ORF2 and ORF3 proteins are believed to be encoded by individual subgenomic RNAs generated during replication (34, 42). ORF2 encodes the viral capsid protein; it contains a signal peptide sequence and three potential glycosylation sites (43), features which are unusual for a protein of a nonenveloped virus. ORF3 encodes a very small protein of about 123 amino acids; it contains one potential phosphorylation site (44). Recombinant ORF3 protein partitions with the cytoskeleton, and it has been suggested that an interaction between ORF2 and ORF3 is associated with the assembly of virions (36, 39). ORF3 is also believed to interact with proteins involved in the host cell signaling pathway (18).
The initiation of RNA synthesis requires the interaction of RdRp and the 3' end of the RNA virus genome. Functional analysis of HEV proteins and nucleotide sequences involved in viral replication has been a difficult task, since an efficient cell culture system for HEV is not yet available. Therefore, most studies of HEV replication to date have been performed by transfection or infection of animals (3, 8, 23, 24) or by overexpression of recombinant proteins in vitro (1, 18, 38, 39, 44). Agrawal et al. reported that the 3' noncoding region (NCR) and an adjacent region of the positive strand of genotype 1 strains of HEV form putative stem-loop structures that affect the binding of recombinant viral RdRp in an in vitro assay (1). Previously described in vivo studies of strain Sar55 (genotype 1) demonstrated that a single "silent" mutation located at nucleotide (nt) 7106 in the 3'-terminal region of ORF2 abolished the infectivity of a recombinant genome for rhesus monkeys and attenuated it for chimpanzees; these findings suggested that the sequence or structure in this region was important and that even small changes in this region might be detrimental (8). On the other hand, the 3'-terminal regions of different genotypes of HEV contain significant sequence differences, indicating that there might be considerable sequence flexibility in these regions.
In order to define more clearly the sequence or structural variability that is compatible with the viability of the virus, we have constructed 3'-terminal intergenotypic chimeras of HEV and have compared them for their ability to replicate in cell cultures and to infect animals. A replicon expressing luciferase as a reporter was constructed in order to quantify viral RNA synthesis in vitro and to assess the effect of 3'-terminal sequences on RNA replication. Serological assays and real-time reverse transcription (RT)-PCR were used to assess the replication of full-length genomes in vivo.
MATERIALS AND METHODS
Cells. Huh-7 cells, a human hepatoma cell line (26), were used for transfection assays of full-length HEV and the HEV replicon. Cells were grown as monolayers in Dulbecco's modified Eagle's medium (DMEM) supplemented with gentamicin, amphotericin B (Fungizone), and 9% fetal bovine serum (FBS) at 37°C in a 5% CO2 incubator. After transfection, the cells were maintained in DMEM supplemented with 5% FBS and incubated at 34.5°C.
Construction of plasmids. The parental clone for all molecular manipulations was pSK-HEV-2 (8) (GenBank accession no. AAF444002), which contains an infectious full-length genome of genotype 1 human HEV strain Sar55 (37). Plasmid pSK-HEV-2 is referred to as pSar55 throughout this study.
Full-length HEV chimeras were constructed in which the 3'-terminal region from nt 7086 to the poly(A) tract of pSar55 was replaced by genotype 3 swine HEV sequences (strain Meng) (25) (GenBank accession no. AF082843) or genotype 2 human HEV sequences (strain Mex-14) (13) (GenBank accession no. M74506). The desired fragments were amplified by RT-PCR either from a 10% stool suspension from pigs infected with strain Meng or from a 10% stool suspension from a cynomolgus macaque inoculated with strain Mex-14 (3); the Mex-14 strain was generously provided by the Centers for Disease Control and Prevention. These fragments were extended at their 5' ends with pSar55 sequences (nt 6982 to 7085) by fusion PCR, digested with NcoI and BglII, and finally ligated into pSar55, substituting for the comparable NcoI and BglII fragments and generating the chimera pSar/3'swine or pSar/3'Mex, respectively.
The HEV replicon expressing firefly luciferase, pE/luc, was constructed by replacing nt 5148 to 5816 of infectious full-length clone pSar55 with the luciferase gene. This procedure placed the reporter gene in frame with the ORF2 methionine initiation codon and resulted in a truncated ORF3 (N-terminal 14 amino acids) and an out-of-frame partial fragment of ORF2, as shown in Fig. 1. For cloning purposes, the luciferase gene was extended at the 5' end with nt 3963 to 5147 of pSar55, generated by fusion PCR with gel-purified PCR products derived from templates pSP-Luc+ (Promega) and pSar55. The resulting fused PCR product was digested with SfiI and EcoRI and cloned into vector pSar55 that had been digested with the same restriction enzymes.
The replicon plasmids, pE/luc-GAD, pE/luc-7106, pE/luc-7106/7097, pE/luc-7106/7144, pE/luc-3'swine, and pE/luc-3'Mex, were constructed by using site-directed mutagenesis and standard recombinant DNA techniques (30). The construct pE/luc-GAD contains mutations at nt 4674 and 4675 within the codon for aspartic acid, changing the conserved GDD motif of the RdRp of HEV into GAD (Fig. 1). The construct pE/luc-7106 represents pE/luc with a single silent G-to-T change in ORF2 at nt 7106 of pSK-HEV-3 (GenBank accession no. AAF444003); this mutation attenuates the virus for rhesus macaques and chimpanzees (8). The constructs pE/luc-7106/7097 and pE/luc-7106/7144 each contain in the pE/luc-7106 background an additional mutation which was selected during infection of a chimpanzee with SK-HEV-3. pE/luc-7106/7097 contains a G-to-A change at nt 7097, and pE/luc-7106/7144 contains a C-to-A change at nt 7144.
The replicon plasmids pE/luc-3'swine and pE/luc-3'Mex contain nt 7086 through the poly(A) tract of strain Meng and strain Mex-14, respectively, in the pE/luc background. Cloning was performed by substitution of the EcoRI-BglII fragment of pE/luc with the comparable sequences of pSar/3'swine and pSar/3'Mex (nt 5816 to 7208).
The differences within the 3'-terminal sequences of the constructed plasmids are summarized in Fig. 2A. Detailed information about the oligonucleotides used to amplify fragments for the cloning of each of the described constructs are available upon request. All plasmids were sequenced to verify that unwanted mutations had not been introduced into the HEV genome during the PCR steps.
Transcription in vitro and transfection. Plasmids were linearized at a unique BglII site located immediately downstream of the poly(A) tract of the HEV sequence. Capped transcripts were synthesized by using a T7 Riboprobe in vitro transcription system (Promega) in the presence of a cap analog. Each 100-μl reaction mixture contained 10 mM dithiothreitol, 1.6 U of RNasin/μl, 0.5 mM each ATP, CTP, and UTP, 0.25 mM GTP, 1 mM m7G(5')ppp(5')G (Ambion), 0.8 U of T7 RNA polymerase/μl, and 5 μg of BglII-linearized plasmid in transcription buffer. After 90 min of incubation at 37°C, an additional 0.4 U of T7 RNA polymerase/μl was added, and incubation was continued for an additional 1 h. The integrity and yield of the transcripts were determined by gel electrophoresis on a nondenaturing agarose gel. Noncapped RNA was prepared by the same protocol, except that the cap analog was omitted and the GTP concentration was increased to 0.5 mM. A transcription mixture lacking only T7 RNA polymerase was also prepared for mock transfection of Huh-7 cells.
Huh-7 cells were transfected with DMRIE-C transfection reagent (Invitrogen) according to the manufacturer's instructions. For each experiment, Huh-7 cells were trypsinized and grown overnight in 12- or 6-well tissue culture plates to obtain 60 to 80% confluence. Cells were washed twice with OptiMEM (Invitrogen) prior to the addition per well of 120 μl (12-well plates) or 400 μl (6-well plates) of OptiMEM containing 50 ng of DMRIE-C/μl and about 7.5 ng of transcript/μl. After 4 h of incubation at 34.5°C, the transfection mixture was exchanged with 1 ml (12-well plates) or 2.5 ml (6-well plates) of DMEM-5% FBS per well. All transfections were carried out in duplicate, and each set of experiments was repeated twice.
Immunofluorescence microscopy. Huh-7 cells were grown in six-well plates to 60 to 80% confluence for transfection by the protocol described above. Cells were split 1:3 at 5 days posttransfection, used to seed two-well glass chamber slides (Nalge Nunc International), grown for an additional 2 days, and then were fixed with acetone for indirect immunofluorescence microscopy as described previously (6). Following fixation, the cells were incubated for 20 min at room temperature with a 1:1 mixture of 10% bovine serum albumin and phosphate-buffered saline (PBS) containing a mixture of monoclonal anti-ORF2 (HEV#4; diluted 1:10) and polyclonal anti-ORF3 (diluted 1:1,000) antibodies as primary antibodies for the simultaneous detection of the two viral proteins. HEV#4 is an ORF2-specific Fab fragment derived from a chimpanzee (32), and the ORF3-specific antibody was produced in rabbits immunized against a synthetic peptide comprising amino acids 91 to 123 of human HEV strain Sar55 (Lofstrand). After being washed with PBS, the slides were incubated for 20 min at room temperature with a mixture of Alexa Fluor 488-conjugated goat anti-human immunoglobulin G (IgG) (Molecular Probes) and Alexa Fluor 568-conjugated goat anti-rabbit IgG (Molecular Probes) as secondary antibodies. Controls included identically treated Huh-7 cells mock transfected with the mixture lacking T7 polymerase. The samples were subjected to confocal microscopy at the Biological Image Facility at the National Institute of Allergy and Infectious Diseases.
Reporter gene assay. Transfected cells were washed twice with PBS and harvested into 200 μl of passive lysis buffer (Promega). Lysates were immediately frozen at –80°C. The samples were frozen and thawed twice and clarified by centrifugation for 2 min, and a 20-μl aliquot of each cell extract was assayed for luciferase activity by using a luciferase assay system (Promega) according to the manufacturer's instructions. A TRILUX 1450 luminescence counter (Perkin-Elmer) was used for detection.
Inoculation of rhesus macaques with HEV genomic RNA. RNA from 10 μg of BglII-linearized plasmid was transcribed in vitro in a 200-μl reaction mixture in the presence of the m7G(5')ppp(5')G cap analog by using a T7 Megascript kit (Ambion) according to the manufacturer's instructions. The integrity and yield of the RNA were verified by gel electrophoresis of 5 μl of transcription reaction mixture on a nondenaturing agarose gel. The remaining 195 μl of the transcription reaction mixture was diluted with 805 μl of PBS without calcium and magnesium and injected into multiple sites in the liver of a rhesus macaque by percutaneous intrahepatic injection guided by ultrasound as described previously for SK-HEV-2 (8). Each chimeric construct was tested in two rhesus monkeys simultaneously. Serum samples were monitored weekly for serum liver enzyme levels (alanine aminotransferase [ALT]) (Anilytics, Gaithersburg, Md.), anti-HEV, and viral genomes. ALT levels that were 2 times the mean preinoculation level were considered elevated and indicated hepatitis. The preinoculation level was calculated as the geometric mean for three weekly samples collected at week zero and before inoculation. Anti-HEV were detected by using an in-house enzyme-linked immunosorbent assay (ELISA) as described previously (9). Viral genomes were quantified by HEV-specific real-time RT-PCR of total RNA extracted from 100 μl of serum as described previously (8).
The macaques used in the study were housed at Bioqual (Rockville, Md.). The housing, maintenance, and care of the animals met or exceeded all requirements for primate husbandry, as specified in the Guide for the Care and Use of Laboratory Animals (27).
Generation of SK-HEV-2 and Sar/3'swine pool. Feces collected from rhesus macaques infected by intrahepatic transfection (8) with Sar55 and Sar/3'swine recombinant genomes were diluted to 10% with PBS, clarified by centrifugation at 3,000 rpm in an IEC Centra-8R centrifuge (International Equipment Company) for 30 min at 4°C, and stored frozen in liquid nitrogen. Genome titers determined by real-time RT-PCR as described previously (8) were 4.8 x 106 genomes of Sar55 per ml and 1.9 x 106 genomes of Sar/3'swine per ml.
Inoculation of pigs with HEV. Two each of 60-day-old specific-pathogen-free (SPF) pigs, housed at the University of Iowa, were inoculated intraveneously with Sar55 and Sar/3'swine as described previously (23). All pigs were found to be seronegative for IgG anti-HEV by an ELISA prior to inoculation. Two of the SPF pigs were inoculated intravenously with 1 ml each of a 10% fecal suspension containing 4.8 x 106 genomes of HEV strain Sar55. The other two were inoculated intravenously with 1 ml each of a 10% fecal suspension containing 1.9 x 106 genomes of HEV chimera Sar/3'swine. Serum samples collected prior to inoculation and weekly until the end of the study were monitored for anti-HEV (9).
Sequence analysis. Genome fragments amplified by HEV-specific RT-PCR from the serum of inoculated animals were purified after agarose gel electrophoresis (QiaQuick; Qiagen) and sequenced directly with an automated sequencer.
RNA structural predictions were performed with the MFOLD program of the Genetics Computer Group (GCG) suite (Wisconsin package, version 10.0) (46).
RESULTS
3'-Terminal intergenotypic chimeras of HEV replicate in cell cultures. Previously it was shown that a single silent mutation introduced into the 3'-terminal region of ORF2 during cDNA cloning abolished or diminished the viability of a recombinant genotype 1 virus in chimpanzees and rhesus macaques (8). This mutation is in a region of genotype 1 strains predicted to form stem-loop structures that are involved in binding to the viral RdRp (1). Since this mutation would have disrupted only 1 bp in a predicted stem structure, it was possible that there were rigid requirements for the functionality of this region. However, the corresponding 3'-terminal regions of strain Mex-14 (genotype 2) (13) and strain Meng (genotype 3) (25) share only 81 and 74% nucleotide identities, respectively, with this region of HEV strain Sar55 (genotype 1) (Fig. 2A), and the MFOLD program predicts structures with different lengths of stems and loops among the three viral genomes (Fig. 2B). The polymerases of the selected genotype 2 and 3 strains also are not identical and share 89.5 and 88% amino acid identities, respectively, with that of Sar55. Therefore, it seemed possible that the polymerase and the 3'-terminal region of each virus had coevolved and that replication would occur only with the homologous combination of the two.
In order to examine the degree of tolerance for structural and sequence variations in the 3'-terminal region of HEV, we constructed chimeric genomes containing the backbone of the Sar55 infectious genome substituted at the 3' end [nt 7086 to the poly(A) tract] with the equivalent region from genotype 2 strain Mex-14 or genotype 3 swine strain Meng. Capped RNA transcripts produced by transcription in vitro of each chimeric cDNA clone and the parental clone, pSar55, were transfected into Huh-7 cells, and replication was ascertained by the detection of ORF2 and ORF3 proteins by confocal microscopy. The merged images of cells doubly labeled for ORF2 protein (green) and ORF3 protein (red) are shown in Fig. 3. Tested separately, neither primary antibody cross-reacted with the opposite secondary antibody (data not shown). Surprisingly, not only did both chimeras replicate (Fig. 3A and B), but also there was no discernible difference in the number of positive cells, time of appearance, or intensity of ORF2 and ORF3 staining compared with the results from a transfection with the parental clone, pSar55 (data not shown).
Evaluation of an HEV replicon expressing luciferase. In order to compare directly the replicative abilities of the chimeras, a quantitative assay was required. For this purpose, an HEV replicon, pE/luc, in which luciferase translation began at the methionine initiation codon of ORF2 was constructed (Fig. 1). The replicon contained the complete 5' NCR, ORF1, and the 3'-terminal region of the infectious genome, but ORF3 sequences were truncated to encode only 14 N-terminal amino acids and ORF2 expression was abolished due to deletion of the N-terminal half of the gene and a resultant frameshift that affected the remainder. As a negative control, the highly conserved GDD motif found in the catalytic portion of RdRp (15, 16) was mutated to GAD (pE/luc-GAD) to inactivate the polymerase expressed by pE/luc (20, 41) (Fig. 1).
Capped or uncapped transcripts synthesized in vitro from pE/luc or pE/luc-GAD, respectively, were transfected into Huh-7 cells, and intracellular luciferase activity was monitored. The luciferase activity measured up to 8 h posttransfection was equivalent to the background emission from uninfected cells and from cells transfected with the negative control transcription mixture lacking T7 RNA polymerase (mock-transfected cells) (data not shown). Capped or uncapped E/luc or E/luc-GAD expressed very low levels of luciferase during the first 24 h posttransfection. However, whereas the levels of luciferase expressed by the uncapped genome and by the E/luc-GAD genome remained very low, the levels of luciferase expressed by the E/luc genome increased dramatically after 1 day posttransfection. Figure 4A shows a representative profile of the expression of luciferase monitored as luciferase activity posttransfection. Whereas luciferase activity in E/luc-GAD-transfected cells increased less than 2-fold between days 1 and 2 posttransfection, in E/luc-transfected cells there was a significant increase in luciferase activity of 12-fold. The difference between E/luc and E/luc-GAD continued to increase through day 4, resulting in a level of luciferase activity in E/luc 122-fold higher than that in E/luc-GAD in this set of experiments. In repeated experiments, E/luc showed, on average, 93-fold more luciferase activity than E/luc-GAD by day 4. Data obtained after 4 days posttransfection were less reliable and showed reduced luciferase activity, most likely due to a loss of cells from the monolayer. Comparable kinetics were obtained when the transcription preparation was treated twice with DNase prior to transfection (data not shown). Therefore, the GAD mutant was unable to replicate or did so very inefficiently. Omission of the cap from pE/luc transcripts prevented the differential increase in luciferase activity and resulted in luciferase levels comparable to those in capped or uncapped E/luc-GAD (Fig. 4A). In comparison, E/luc RNA, when capped, was highly replication competent in Huh-7 cells.
Despite the convincing replication of the E/luc replicon compared to the negative control, hybridization analyses failed to detect newly synthesized positive- or negative-strand RNA. The large number of input transcripts, the modest number of cells successfully transfected, and the apparently low efficiency of RNA replication in the cells made it impossible to detect newly synthesized replicon RNA by Northern blot analysis or real-time RT-PCR (data not shown). E/luc- and E/luc-GAD-transfected cells both contained steadily degrading input RNA that was easily detected as a large smear by Northern blotting over the course of 6 days. Unfortunately, newly synthesized RNA was not detectable against this background.
Influence of 3'-terminal sequences on replication ability. The 3'-terminal regions comprising the intergenotypic chimeras or the single mutation at nt 7106 which had attenuated Sar55 for nonhuman primates was introduced into the replicon in place of the wild-type Sar55 sequence, and the ability of each mutant replicon was monitored by quantifying luciferase activity after transfection of Huh-7 cells. Once again, all replicons produced little luciferase during the first 24 h posttransfection, but between 2 and 4 days posttransfection, differences became evident (Fig. 4B). Genomes containing the single mutation at nucleotide 7106 replicated, albeit inefficiently. Although the luciferase activity in E/luc-7106 was 17-fold higher than that in E/luc-GAD by day 4, it was 5-fold lower than that of the parental replicon (E/luc) at that time. In contrast, throughout the study, luciferase production by the E/luc-3'Mex or E/luc-3'swine chimeric replicon was always higher than that by E/luc-7106, and the kinetics of luciferase production by the intergenotypic chimeras were more similar to those of parental replicon E/luc than to those of E/luc-7106 (Fig. 4B). At 4 days posttransfection, luciferase expression was higher than that of E/luc-GAD by 44-fold for E/luc-3'Mex and by 50-fold for E/luc-3'swine (52- and 55-fold, respectively, in a different experiment). Thus, the multiple sequence differences and predicted structure variations typical of genotype 2 and 3 strains had significantly less effect on the ability of the Sar55 polymerase to replicate the genome than did the single mutation at nt 7106.
Further analysis of the 3'-terminal regions was performed with replicons harboring additional mutations in these regions. Inoculation of chimpanzees with capped transcripts synthesized in vitro from cDNA clone pSK-HEV-3 (containing the replication-inhibitory mutation at nt 7106) resulted in the selection of new 3'-terminal region mutations that were postulated to increase replication efficiency in vivo (8). The virus isolated from one animal had at nt 7097 a new mutation (G to A) which did not restore the wild-type structure (Sar55) but did cause slight changes in the stem downstream of the mutation at nt 7106 in the predicted secondary structure (Fig. 2B). The virus isolated from the second animal had at a different site (nt 7144) a new mutation (C to A) which was a covariant of the mutation at nt 7106 in the predicted secondary structure of the 3'-terminal region and restored the wild-type structure (Fig. 2B). In order to determine whether these mutations did indeed affect viral replication, replicons pE/luc-7106/7097 and pE/luc-7106/7144 were constructed to contain the additional mutations at nt 7097 and nt 7144, respectively, in the pE/luc-7106 background. RNA replication was again evaluated by monitoring luciferase activity in transfected Huh-7 cells over a period of 4 days and was compared to RNA replication in parental replicon E/luc, E/luc-7106, E/luc-GAD, and mock-transfected cells (Fig. 4C). Luciferase production by E/luc-7106 was higher than that by the null polymerase mutant E/luc-GAD but still was sixfold lower than that by wild-type E/luc on day 4. The addition of the mutation at nt 7097 to E/luc-7106 increased luciferase production 2.4-fold over that by E/luc-7106 by day 4, and the addition of the mutation at nt 7144 to E/luc-7106 increased luciferase activity even more (4-fold over that in E/luc-7106). These results confirmed the hypothesis that these mutations had been selected because they increased viral replication and suggested that mutations which affect the secondary structure in this region can influence RNA replication.
3'-terminal HEV chimeric genomes are infectious in vivo. Since Huh-7 cells grown in cell cultures are not the same as hepatocytes within a liver, it was of interest to determine whether the 3'-terminal chimeras also could replicate in vivo. Capped transcripts synthesized in vitro from full-length cDNA clones pSar/3'swine and pSar/3'Mex were tested for their infectivity by intrahepatic inoculation into the livers of two rhesus monkeys each. Seroconversion to anti-HEV, serum liver enzyme levels, and viremia were monitored weekly for 12 weeks and compared with the results obtained with previously tested transfected Sar55 (wild type) (8) (Fig. 5). All six animals developed anti-HEV between 5 and 8 weeks posttransfection, concurrent with or shortly after the time at which peak amounts of viral RNA were detected in the serum. Overall, there were no differences among the animals with regard to any of the parameters measured that could not be explained by normal biological variations; e.g., the variations between two animals receiving the same inoculum were comparable to those between two animals receiving different inocula. The 3'-terminal sequences of HEV were amplified from each of the six rhesus monkeys by RT-PCR and were found to be identical to those of the input genome for each monkey. Therefore, in vivo, the 3'-terminal chimeras replicated with an efficiency indistinguishable from that of the wild-type parent.
3'-Terminal sequences do not specify host range. It was previously shown that Sar55 virus was infectious for nonhuman primates but was not able to infect swine even when a well-characterized infectious stock was inoculated intravenously (23). In contrast, swine HEV strain Meng was able to infect rhesus monkeys and chimpanzees as well as swine, indicating that it had a broader host range that allowed it to cross species barriers (24).
The 3' NCR of viruses has been reported to interact with host proteins (4, 28, 40), and since there was a 26% difference in nucleotide sequence in this region between strains Sar55 and Meng, it seemed possible that this region might be responsible for the ability of the swine virus to infect different species. To test this hypothesis, we made one virus pool containing HEV from a rhesus monkey that had been transfected with transcripts synthesized in vitro from pSar55 and another virus pool from a monkey that had been transfected with transcripts synthesized in vitro from pSar/3'swine; we inoculated approximately 106 genomes of Sar55 or of Sar/3'swine into swine. SPF pigs were monitored weekly for 15.5 weeks. Seroconversion was not detected in any pig (data not shown), confirming that genotype 1 human strain Sar55 could not infect pigs and indicating that the 3'-terminal region of the swine HEV genome was not sufficient to confer to Sar55 the ability to infect swine productively.
DISCUSSION
The production of luciferase encoded by a replicon of HEV provided a convenient method of demonstrating and analyzing HEV genomic replication. The results obtained with the replicon containing wild-type Sar55 sequences indicated that HEV RNA replication in cell cultures occurred independently of functional ORF2 and ORF3 proteins, confirming results obtained previously with a green fluorescent protein (GFP)-producing replicon (6), for which different parameters were measured. In the GFP-producing system, fluorescence-activated cell sorting analysis was used to quantify the numbers of transfected cells (about 10% maximum), thus providing an estimate of individual genome viability. That study showed that although uncapped genomes could initiate an infection in vitro, it was a rare event compared to the number of infections initiated with capped genomes. Additionally, detectable GFP was not produced when cells were transfected with genomes containing the GAD mutant polymerase. In the replicon system described here, luciferase production was averaged over the entire population of cells, and the amount of luciferase produced by the uncapped transcripts was not larger than that produced by the GAD mutant. Thus, data from both systems confirmed that capping of the HEV genome is very important for viability, but the luciferase system was not as sensitive as the GFP system for detecting rare events. However, the luciferase system was quite useful for making other quantitative comparisons.
The luciferase-expressing replicons all produced very low and nearly identical levels of luciferase during the first 12 h posttransfection. This result most likely was due to translation of the transfecting RNA by an unknown mechanism. This early expression of the luciferase gene did not seem to require a cap, because the uncapped and capped genomes produced equivalent luciferase activities during this time. It also did not reflect replication, because the activities of the GAD and GDD constructs consistently differed by less than twofold at 12 h posttransfection. The amount of luciferase produced should reflect the synthesis and replication of the subgenomic mRNA believed to encode ORF2 protein normally and luciferase in this case (6, 34, 42). Therefore, the increase in luciferase activity observed between 1 and 2 days with wild-type replicon E/luc compared to polymerase-defective replicon E/luc-GAD suggested that the replication complexes or viral enzymes required for genome replication did not accumulate to significant levels until that time, that the replication of HEV is a relatively inefficient process, and that the replication cycle is prolonged, at least in cell cultures.
The inefficiency of HEV replication made it necessary to rely on a reporter such as luciferase to detect differences in RNA replication between the wild-type genome and one encoding mutant RdRp or containing 3'-terminal mutations. The failure to detect newly synthesized replicon RNA in our system reflects the large amount of transcribed RNA necessary for transfection, which would contain a mixture of capped and uncapped RNAs. The large amount of input RNA required to transfect a significant number of cells caused such a high background that it and its degradation products prevented detection by Northern blotting of any newly synthesized replicon RNA or subgenomic RNAs. Similarly, the input RNA produced such a high signal in RT-PCR and newly synthesized RNA represented such a small fraction that it could not be reliably detected.
The luciferase system was ideal for studying the effect on RNA replication of mutations in the 3'-terminal region. Of particular interest was the mutation at nt 7106, which was previously shown to attenuate HEV for nonhuman primates (8). This attenuation could reflect the inefficient replication of viral RNA or the inability to package the genome into infectious virions. The results of the replicon assays clearly demonstrated that luciferase activity in E/luc-7106 was diminished compared to that in the parent replicon and thus established the fact that this single mutation was in a region critical for RNA replication.
The decreased level of replication induced by this single mutation was in contrast to the more robust level of replication observed after transfection with E/luc-3'Mex or E/luc-3'swine. In the 109-nt tract exchanged to generate the chimeras, strain Mex-14 differed from Sar55 at 21 nt, including nine insertions; thus, there was about a 20% difference in the sequences. Swine strain Meng differed even more from Sar55 in this region, with 29 nt differences, including six insertions (for a nucleotide sequence difference of 26%). Additionally, the predicted secondary structures of the genomes were all different from each other in this region. However, in spite of these large differences, both chimeric replicons replicated more efficiently in cell cultures than did the virus with a mutation at nt 7106 and, as full-length chimeras, both produced infections that were indistinguishable from those produced by wild-type Sar55.
Agrawal et al. reported that 3'-terminal sequences of genotype 1 HEV genomes form two putative stem-loop structures, which seem to be important in concert for binding to RdRp in vitro (1). In light of the results obtained with the chimeras, there seems to be a considerable degree of flexibility, with regard to RNA replication, in the precise sequence and structure: the formation of the two stem structures by itself is not sufficient, since the seemingly minor change caused by the mutation at nt 7106, which would abolish only one putative base pair within stem-loop 2, had a significant inhibitory effect on RNA replication. Restoration of this particular part of stem-loop 2 by a covariant mutation significantly increased the efficiency of RNA replication, although not to wild-type levels; these results suggested that the general structure generated by the 3'-terminal sequences, including the ORF2 sequences, is important but that the sequences themselves may also have an effect. It is interesting that the region around nt 7106 is completely base paired in strain Mex-14, whereas in both Sar55 and the swine strain, this region contains a small loop. The stem mutation at nt 7097, selected in the second chimpanzee, had a larger effect on the predicted structure and increased viral replication in vitro to a lesser extent, probably because the structure, although more favorable, still deviated from that of the wild type. However, the hypothesis that the mutations at nt 7097 and nt 7144 were selected in the chimpanzees because they promoted genome replication was supported by the results of the luciferase assays.
The accumulating data suggesting that HEV may be a zoonosis make it interesting to determine why strain Meng isolated from swine was able to cross species borders, whereas strains Sar55 and Mex-14 isolated from humans were not. The 3' NCRs of some viruses have been shown to interact with host proteins (4, 28, 40). Therefore, given the extent of differences between strains Sar55 and Meng in these regions, it appeared reasonable to determine whether the Sar/3'swine chimera could infect pigs. The fact that a large dose of chimeric virus did not infect either of the two inoculated pigs indicated that the 3'-terminal region is not responsible for host range determination or that, if it is involved, other regions of the swine genome are also required for infection of pigs. Once an infectious clone of swine HEV is developed, it should be possible to answer the question of how host range is determined.
The robust viability of the full-length chimeras and the appreciable replication competence of the corresponding replicons are in contrast to the diminished replication of the virus with a mutation at nt 7106 and its replicon and underscore the difficulty in trying to determine which sequences or structures in the 3'-terminal region are critical for replication. The current study validates the use of a luciferase-producing replicon to study these parameters in vitro to reliably predict their involvement in vivo. It should now be possible to undertake saturation-mutagenesis studies of the replicon to determine which mutants can replicate in vitro and then transfect those into animals to determine which regions are critical for replication in vivo.
ACKNOWLEDGMENTS
This work was supported in part by National Institute of Allergy and Infectious Diseases contract no. 1-A0-02733.
We thank Ron Engle and Claro Yu for performing the ELISA, Owen Schwartz for guidance and help with confocal microscopy, and Natalia Teterina for valuable advice on RNA detection.
REFERENCES
Agrawal, S., D. Gupta, and S. K. Panda. 2001. The 3' end of hepatitis E virus (HEV) genome binds specifically to the viral RNA-dependent RNA polymerase (RdRp). Virology 282:87-101.
Arankalle, V. A., L. P. Chobe, A. M. Walimbe, P. N. Yergolkar, and G. P. Jacob. 2003. Swine HEV infection in south India and phylogenetic analysis (1985-1999). J. Med. Virol. 69:391-396.
Bradley, D. W., K. Krawczynski, E. H. Cook, Jr., K. A. McCaustland, C. D. Humphrey, J. E. Spelbring, H. Myint, and J. E. Maynard. 1987. Enterically transmitted non-A, non-B hepatitis: serial passage of disease in cynomolgus macaques and tamarins and recovery of disease-associated 27- to 34-nm viruslike particles. Proc. Natl. Acad. Sci. USA 84:6277-6281.
Dollenmaier, G., and M. Weitz. 2003. Interaction of glyceraldehyde-3-phosphate dehydrogenase with secondary and tertiary RNA structural elements of the hepatitis A virus 3' translated and non-translated regions. J. Gen. Virol. 84:403-414.
Emerson, S. U., D. Anderson, A. Arankalle, X.-J. Meng, M. Purdy, G. G. Schlauder, and S. A. Tsarev. 2004. Hepevirus, p. 851-855. In C. M. Fauquet, M. A. Mayo, J. Maniloff, U. Desselberger, and L. A. Ball (ed.), Virus taxonomy. VIIIth report of the ICTV. Elsevier/Academic Press, London, United Kingdom.
Emerson, S. U., H. Nguyen, J. Graff, D. A. Stephany, A. Brockington, and R. H. Purcell. 2004. In vitro replication of hepatitis E virus (HEV) genomes and of an HEV replicon expressing green fluorescent protein. J. Virol. 78:4838-4846.
Emerson, S. U., and R. H. Purcell. 2003. Hepatitis E virus. Rev. Med. Virol. 13:145-154.
Emerson, S. U., M. Zhang, X.-J. Meng, H. Nguyen, M. St. Claire, S. Govindarajan, Y. K. Huang, and R. H. Purcell. 2001. Recombinant hepatitis E virus genomes infectious for primates: importance of capping and discovery of a cis-reactive element. Proc. Natl. Acad. Sci. USA 98:15270-15275.
Engle, R. E., C. Yu, S. U. Emerson, X. J. Meng, and R. H. Purcell. 2002. Hepatitis E virus (HEV) capsid antigens derived from viruses of human and swine origin are equally efficient for detecting anti-HEV by enzyme immunoassay. J. Clin. Microbiol. 40:4576-4580.
Garkavenko, O., A. Obriadina, J. Meng, D. A. Anderson, H. J. Benard, B. A. Schroeder, Y. E. Khudyakov, H. A. Fields, and M. C. Croxson. 2001. Detection and characterisation of swine hepatitis E virus in New Zealand. J. Med. Virol. 65:525-529.
Haqshenas, G., H. L. Shivaprasad, P. R. Woolcock, D. H. Read, and X. J. Meng. 2001. Genetic identification and characterization of a novel virus related to human hepatitis E virus from chickens with hepatitis-splenomegaly syndrome in the United States. J. Gen. Virol. 82:2449-2462.
Hsieh, S. Y., X. J. Meng, Y. H. Wu, S. T. Liu, A. W. Tam, D. Y. Lin, and Y. F. Liaw. 1999. Identity of a novel swine hepatitis E virus in Taiwan forming a monophyletic group with Taiwan isolates of human hepatitis E virus. J. Clin. Microbiol. 37:3828-3834.
Huang, C. C., D. Nguyen, J. Fernandez, K. Y. Yun, K. E. Fry, D. W. Bradley, A. W. Tam, and G. R. Reyes. 1992. Molecular cloning and sequencing of the Mexico isolate of hepatitis E virus (HEV). Virology 191:550-558.
Huang, F. F., G. Haqshenas, H. L. Shivaprasad, D. K. Guenette, P. R. Woolcock, C. T. Larsen, F. W. Pierson, F. Elvinger, T. E. Toth, and X. J. Meng. 2002. Heterogeneity and seroprevalence of a newly identified avian hepatitis E virus from chickens in the United States. J. Clin. Microbiol. 40:4197-4202.
Kamer, G., and P. Argos. 1984. Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12:7269-7282.
Koonin, E. V. 1991. The phylogeny of RNA-dependent RNA polymerases of positive-strand RNA viruses. J. Gen. Virol. 72:2197-2206.
Koonin, E. V., A. E. Gorbalenya, M. A. Purdy, M. N. Rozanov, G. R. Reyes, and D. W. Bradley. 1992. Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses. Proc. Natl. Acad. Sci. USA 89:8259-8263.
Korkaya, H., S. Jameel, D. Gupta, S. Tyagi, R. Kumar, M. Zafrullah, M. Mazumdar, S. K. Lal, L. Xiaofang, D. Sehgal, S. R. Das, and D. Sahal. 2001. The ORF3 protein of hepatitis E virus binds to Src homology 3 domains and activates MAPK. J. Biol. Chem. 276:42389-42400.
Krawczynski, K., S. Kamili, and R. Aggarwal. 2001. Global epidemiology and medical aspects of hepatitis E. Forum (Genova) 11:166-179.
Lohmann, V., F. Korner, U. Herian, and R. Bartenschlager. 1997. Biochemical properties of hepatitis C virus NS5B RNA-dependent RNA polymerase and identification of amino acid sequence motifs essential for enzymatic activity. J. Virol. 71:8416-8428.
Magden, J., N. Takeda, T. Li, P. Auvinen, T. Ahola, T. Miyamura, A. Merits, and L. Kaariainen. 2001. Virus-specific mRNA capping enzyme encoded by hepatitis E virus. J. Virol. 75:6249-6255.
Meng, X. J. 2000. Zoonotic and xenozoonotic risks of hepatitis E virus. Infect. Dis. Rev. 2:35-41.
Meng, X. J., P. G. Halbur, J. S. Haynes, T. S. Tsareva, J. D. Bruna, R. L. Royer, R. H. Purcell, and S. U. Emerson. 1998. Experimental infection of pigs with the newly identified swine hepatitis E virus (swine HEV), but not with human strains of HEV. Arch. Virol. 143:1405-1415.
Meng, X. J., P. G. Halbur, M. S. Shapiro, S. Govindarajan, J. D. Bruna, I. K. Mushahwar, R. H. Purcell, and S. U. Emerson. 1998. Genetic and experimental evidence for cross-species infection by swine hepatitis E virus. J. Virol. 72:9714-9721.
Meng, X. J., R. H. Purcell, P. G. Halbur, J. R. Lehman, D. M. Webb, T. S. Tsareva, J. S. Haynes, B. J. Thacker, and S. U. Emerson. 1997. A novel virus in swine is closely related to the human hepatitis E virus. Proc. Natl. Acad. Sci. USA 94:9860-9865.
Nakabayashi, H., K. Taketa, K. Miyano, T. Yamane, and J. Sato. 1982. Growth of human hepatoma cell lines with differentiated functions in chemically defined medium. Cancer Res. 42:3858-3863.
National Research Council. 1996. Guide for the care and use of laboratory animals. National Academy Press, Washington, D.C.
Petrik, J., H. Parker, and G. J. Alexander. 1999. Human hepatic glyceraldehyde-3-phosphate dehydrogenase binds to the poly(U) tract of the 3' non-coding region of hepatitis C virus genomic RNA. J. Gen. Virol. 80:3109-3113.
Purcell, R. H., and S. U. Emerson. 2001. Hepatitis E virus, p. 3051-3061. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schlauder, G. G., and I. K. Mushahwar. 2001. Genetic heterogeneity of hepatitis E virus. J. Med. Virol. 65:282-292.
Schofield, D. J., J. Glamann, S. U. Emerson, and R. H. Purcell. 2000. Identification by phage display and characterization of two neutralizing chimpanzee monoclonal antibodies to the hepatitis E virus capsid protein. J. Virol. 74:5548-5555.
Takahashi, M., T. Nishizawa, H. Miyajima, Y. Gotanda, T. Iita, F. Tsuda, and H. Okamoto. 2003. Swine hepatitis E virus strains in Japan form four phylogenetic clusters comparable with those of Japanese isolates of human hepatitis E virus. J. Gen. Virol. 84:851-862.
Tam, A. W., M. M. Smith, M. E. Guerra, C. C. Huang, D. W. Bradley, K. E. Fry, and G. R. Reyes. 1991. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 185:120-131.
Tei, S., N. Kitajima, K. Takahashi, and S. Mishiro. 2003. Zoonotic transmission of hepatitis E virus from deer to human beings. Lancet 362:371-373.
Torresi, J., F. Li, S. A. Locarnini, and D. A. Anderson. 1999. Only the non-glycosylated fraction of hepatitis E virus capsid (open reading frame 2) protein is stable in mammalian cells. J. Gen. Virol. 80:1185-1188.
Tsarev, S. A., S. U. Emerson, G. R. Reyes, T. S. Tsareva, L. J. Legters, I. A. Malik, M. Iqbal, and R. H. Purcell. 1992. Characterization of a prototype strain of hepatitis E virus. Proc. Natl. Acad. Sci. USA 89:559-563.
Tyagi, S., S. Jameel, and S. K. Lal. 2001. Self-association and mapping of the interaction domain of hepatitis E virus ORF3 protein. J. Virol. 75:2493-2498.
Tyagi, S., H. Korkaya, M. Zafrullah, S. Jameel, and S. K. Lal. 2002. The phosphorylated form of the ORF3 protein of hepatitis E virus interacts with its non-glycosylated form of the major capsid protein, ORF2. J. Biol. Chem. 277:22759-22767.
Waggoner, S., and P. Sarnow. 1998. Viral ribonucleoprotein complex formation and nucleolar-cytoplasmic relocalization of nucleolin in poliovirus-infected cells. J. Virol. 72:6699-6709.
Wang, X., and S. Gillam. 2001. Mutations in the GDD motif of rubella virus putative RNA-dependent RNA polymerase affect virus replication. Virology 285:322-331.
Xia, X., R. Huang, and D. Li. 2000. Studies on the subgenomic RNAs of hepatitis E virus. Wei Sheng Wu Xue Bao 41:622-627.
Zafrullah, M., M. H. Ozdener, R. Kumar, S. K. Panda, and S. Jameel. 1999. Mutational analysis of glycosylation, membrane translocation, and cell surface expression of the hepatitis E virus ORF2 protein. J. Virol. 73:4074-4082.
Zafrullah, M., M. H. Ozdener, S. K. Panda, and S. Jameel. 1997. The ORF3 protein of hepatitis E virus is a phosphoprotein that associates with the cytoskeleton. J. Virol. 71:9045-9053.
Zhou, Y.-H., R. H. Purcell, and S. U. Emerson. 2004. An ELISA for putative neutralizing antibidies to hepatitis E virus detects antibodies to genotype 1, 2, 3 and 4. Vaccine 22:2578-2585.
Zuker, M., D. H. Mathews, and D. H. Turner. 1999. Algorithms and thermodynamics for RNA secondary structure prediction: a practical guide, p. 11-43. In J. Barciszewski and B. F. C. Clark (ed.), RNA biochemistry and biotechnology. NATO ASI series. Kluwer Academic Press, Boston, Mass.(Judith Graff, Hanh Nguyen)
Hepatitis Viruses Sections, Laboratory of Infectious Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland
Department of Veterinary Diagnostic and Production Animal Medicine, College of Veterinary Medicine, Iowa State University, Ames, Iowa
Bioqual, Incorporated, Rockville, Maryland
ABSTRACT
Hepatitis E virus (HEV) replication is not well understood, mainly because the virus does not infect cultured cells efficiently. However, Huh-7 cells transfected with full-length genomes produce open reading frame 2 protein, indicative of genome replication (6). To investigate the role of 3'-terminal sequences in RNA replication, we constructed chimeric full-length genomes with divergent 3'-terminal sequences of genotypes 2 and 3 replacing that of genotype 1 and transfected them into Huh-7 cells. The production of viral proteins by these full-length chimeras was indistinguishable from that of the wild type, suggesting that replication was not impaired. In order to better quantify HEV replication in cell culture, we constructed an HEV replicon with a reporter (luciferase). Luciferase production was cap dependent and RNA-dependent RNA polymerase dependent and increased following transfection of Huh-7 cells. Replicons harboring the 3'-terminal intergenotypic chimera sequences were also assayed for luciferase production. In spite of the large sequence differences among the 3' termini of the viruses, replication of the chimeric replicons was surprisingly similar to that of the parental replicon. However, a single unique nucleotide change within a predicted stem structure at the 3' terminus substantially reduced the efficiency of replication: RNA replication was partially restored by a covariant mutation. Similar patterns of replication were obtained when full-length genomes were inoculated into rhesus macaques, suggesting that the in vitro system could be used to predict the effect of 3'-terminal mutations in vivo. Incorporation of the 3'-terminal sequences of the swine strain of HEV into the genotype 1 human strain did not enable the human strain to infect swine.
INTRODUCTION
Hepatitis E virus (HEV) is a major cause of enterically transmitted hepatitis, especially in Asia and Africa. Serological prevalence data suggest that HEV might be endemic in industrialized countries as well, although it rarely causes disease in these countries (reviewed in references 7, 19, and 29). Antibodies to HEV (anti-HEV) have also been detected in a wide variety of animals, and HEV has been found in swine (2, 10, 12, 25, 33), deer (35), and chickens (11, 14); these findings raise the possibility that HEV may be a zoonosis (22). HEV strains infecting mammals differ at the nucleotide level by as much as 27%, whereas the avian strain differs from the mammalian strains by up to 50%. The mammalian HEV strains have been classified into four genotypes based on sequence variations (31), but they appear to comprise a single serotype (29, 45).
HEV differs enough from other viruses that it was recently classified as the sole member of the new genus Hepevirus (5). HEV is nonenveloped and contains a single-stranded RNA genome of positive sense that is capped and polyadenylated. The 7.2-kb genome contains three partially overlapping open reading frames (ORFs). ORF1 encodes a nonstructural polyprotein which provides guanylyl-methyltransferase and RNA-dependent RNA polymerase (RdRp) activities and possibly other functions (17, 21). The ORF2 and ORF3 proteins are believed to be encoded by individual subgenomic RNAs generated during replication (34, 42). ORF2 encodes the viral capsid protein; it contains a signal peptide sequence and three potential glycosylation sites (43), features which are unusual for a protein of a nonenveloped virus. ORF3 encodes a very small protein of about 123 amino acids; it contains one potential phosphorylation site (44). Recombinant ORF3 protein partitions with the cytoskeleton, and it has been suggested that an interaction between ORF2 and ORF3 is associated with the assembly of virions (36, 39). ORF3 is also believed to interact with proteins involved in the host cell signaling pathway (18).
The initiation of RNA synthesis requires the interaction of RdRp and the 3' end of the RNA virus genome. Functional analysis of HEV proteins and nucleotide sequences involved in viral replication has been a difficult task, since an efficient cell culture system for HEV is not yet available. Therefore, most studies of HEV replication to date have been performed by transfection or infection of animals (3, 8, 23, 24) or by overexpression of recombinant proteins in vitro (1, 18, 38, 39, 44). Agrawal et al. reported that the 3' noncoding region (NCR) and an adjacent region of the positive strand of genotype 1 strains of HEV form putative stem-loop structures that affect the binding of recombinant viral RdRp in an in vitro assay (1). Previously described in vivo studies of strain Sar55 (genotype 1) demonstrated that a single "silent" mutation located at nucleotide (nt) 7106 in the 3'-terminal region of ORF2 abolished the infectivity of a recombinant genome for rhesus monkeys and attenuated it for chimpanzees; these findings suggested that the sequence or structure in this region was important and that even small changes in this region might be detrimental (8). On the other hand, the 3'-terminal regions of different genotypes of HEV contain significant sequence differences, indicating that there might be considerable sequence flexibility in these regions.
In order to define more clearly the sequence or structural variability that is compatible with the viability of the virus, we have constructed 3'-terminal intergenotypic chimeras of HEV and have compared them for their ability to replicate in cell cultures and to infect animals. A replicon expressing luciferase as a reporter was constructed in order to quantify viral RNA synthesis in vitro and to assess the effect of 3'-terminal sequences on RNA replication. Serological assays and real-time reverse transcription (RT)-PCR were used to assess the replication of full-length genomes in vivo.
MATERIALS AND METHODS
Cells. Huh-7 cells, a human hepatoma cell line (26), were used for transfection assays of full-length HEV and the HEV replicon. Cells were grown as monolayers in Dulbecco's modified Eagle's medium (DMEM) supplemented with gentamicin, amphotericin B (Fungizone), and 9% fetal bovine serum (FBS) at 37°C in a 5% CO2 incubator. After transfection, the cells were maintained in DMEM supplemented with 5% FBS and incubated at 34.5°C.
Construction of plasmids. The parental clone for all molecular manipulations was pSK-HEV-2 (8) (GenBank accession no. AAF444002), which contains an infectious full-length genome of genotype 1 human HEV strain Sar55 (37). Plasmid pSK-HEV-2 is referred to as pSar55 throughout this study.
Full-length HEV chimeras were constructed in which the 3'-terminal region from nt 7086 to the poly(A) tract of pSar55 was replaced by genotype 3 swine HEV sequences (strain Meng) (25) (GenBank accession no. AF082843) or genotype 2 human HEV sequences (strain Mex-14) (13) (GenBank accession no. M74506). The desired fragments were amplified by RT-PCR either from a 10% stool suspension from pigs infected with strain Meng or from a 10% stool suspension from a cynomolgus macaque inoculated with strain Mex-14 (3); the Mex-14 strain was generously provided by the Centers for Disease Control and Prevention. These fragments were extended at their 5' ends with pSar55 sequences (nt 6982 to 7085) by fusion PCR, digested with NcoI and BglII, and finally ligated into pSar55, substituting for the comparable NcoI and BglII fragments and generating the chimera pSar/3'swine or pSar/3'Mex, respectively.
The HEV replicon expressing firefly luciferase, pE/luc, was constructed by replacing nt 5148 to 5816 of infectious full-length clone pSar55 with the luciferase gene. This procedure placed the reporter gene in frame with the ORF2 methionine initiation codon and resulted in a truncated ORF3 (N-terminal 14 amino acids) and an out-of-frame partial fragment of ORF2, as shown in Fig. 1. For cloning purposes, the luciferase gene was extended at the 5' end with nt 3963 to 5147 of pSar55, generated by fusion PCR with gel-purified PCR products derived from templates pSP-Luc+ (Promega) and pSar55. The resulting fused PCR product was digested with SfiI and EcoRI and cloned into vector pSar55 that had been digested with the same restriction enzymes.
The replicon plasmids, pE/luc-GAD, pE/luc-7106, pE/luc-7106/7097, pE/luc-7106/7144, pE/luc-3'swine, and pE/luc-3'Mex, were constructed by using site-directed mutagenesis and standard recombinant DNA techniques (30). The construct pE/luc-GAD contains mutations at nt 4674 and 4675 within the codon for aspartic acid, changing the conserved GDD motif of the RdRp of HEV into GAD (Fig. 1). The construct pE/luc-7106 represents pE/luc with a single silent G-to-T change in ORF2 at nt 7106 of pSK-HEV-3 (GenBank accession no. AAF444003); this mutation attenuates the virus for rhesus macaques and chimpanzees (8). The constructs pE/luc-7106/7097 and pE/luc-7106/7144 each contain in the pE/luc-7106 background an additional mutation which was selected during infection of a chimpanzee with SK-HEV-3. pE/luc-7106/7097 contains a G-to-A change at nt 7097, and pE/luc-7106/7144 contains a C-to-A change at nt 7144.
The replicon plasmids pE/luc-3'swine and pE/luc-3'Mex contain nt 7086 through the poly(A) tract of strain Meng and strain Mex-14, respectively, in the pE/luc background. Cloning was performed by substitution of the EcoRI-BglII fragment of pE/luc with the comparable sequences of pSar/3'swine and pSar/3'Mex (nt 5816 to 7208).
The differences within the 3'-terminal sequences of the constructed plasmids are summarized in Fig. 2A. Detailed information about the oligonucleotides used to amplify fragments for the cloning of each of the described constructs are available upon request. All plasmids were sequenced to verify that unwanted mutations had not been introduced into the HEV genome during the PCR steps.
Transcription in vitro and transfection. Plasmids were linearized at a unique BglII site located immediately downstream of the poly(A) tract of the HEV sequence. Capped transcripts were synthesized by using a T7 Riboprobe in vitro transcription system (Promega) in the presence of a cap analog. Each 100-μl reaction mixture contained 10 mM dithiothreitol, 1.6 U of RNasin/μl, 0.5 mM each ATP, CTP, and UTP, 0.25 mM GTP, 1 mM m7G(5')ppp(5')G (Ambion), 0.8 U of T7 RNA polymerase/μl, and 5 μg of BglII-linearized plasmid in transcription buffer. After 90 min of incubation at 37°C, an additional 0.4 U of T7 RNA polymerase/μl was added, and incubation was continued for an additional 1 h. The integrity and yield of the transcripts were determined by gel electrophoresis on a nondenaturing agarose gel. Noncapped RNA was prepared by the same protocol, except that the cap analog was omitted and the GTP concentration was increased to 0.5 mM. A transcription mixture lacking only T7 RNA polymerase was also prepared for mock transfection of Huh-7 cells.
Huh-7 cells were transfected with DMRIE-C transfection reagent (Invitrogen) according to the manufacturer's instructions. For each experiment, Huh-7 cells were trypsinized and grown overnight in 12- or 6-well tissue culture plates to obtain 60 to 80% confluence. Cells were washed twice with OptiMEM (Invitrogen) prior to the addition per well of 120 μl (12-well plates) or 400 μl (6-well plates) of OptiMEM containing 50 ng of DMRIE-C/μl and about 7.5 ng of transcript/μl. After 4 h of incubation at 34.5°C, the transfection mixture was exchanged with 1 ml (12-well plates) or 2.5 ml (6-well plates) of DMEM-5% FBS per well. All transfections were carried out in duplicate, and each set of experiments was repeated twice.
Immunofluorescence microscopy. Huh-7 cells were grown in six-well plates to 60 to 80% confluence for transfection by the protocol described above. Cells were split 1:3 at 5 days posttransfection, used to seed two-well glass chamber slides (Nalge Nunc International), grown for an additional 2 days, and then were fixed with acetone for indirect immunofluorescence microscopy as described previously (6). Following fixation, the cells were incubated for 20 min at room temperature with a 1:1 mixture of 10% bovine serum albumin and phosphate-buffered saline (PBS) containing a mixture of monoclonal anti-ORF2 (HEV#4; diluted 1:10) and polyclonal anti-ORF3 (diluted 1:1,000) antibodies as primary antibodies for the simultaneous detection of the two viral proteins. HEV#4 is an ORF2-specific Fab fragment derived from a chimpanzee (32), and the ORF3-specific antibody was produced in rabbits immunized against a synthetic peptide comprising amino acids 91 to 123 of human HEV strain Sar55 (Lofstrand). After being washed with PBS, the slides were incubated for 20 min at room temperature with a mixture of Alexa Fluor 488-conjugated goat anti-human immunoglobulin G (IgG) (Molecular Probes) and Alexa Fluor 568-conjugated goat anti-rabbit IgG (Molecular Probes) as secondary antibodies. Controls included identically treated Huh-7 cells mock transfected with the mixture lacking T7 polymerase. The samples were subjected to confocal microscopy at the Biological Image Facility at the National Institute of Allergy and Infectious Diseases.
Reporter gene assay. Transfected cells were washed twice with PBS and harvested into 200 μl of passive lysis buffer (Promega). Lysates were immediately frozen at –80°C. The samples were frozen and thawed twice and clarified by centrifugation for 2 min, and a 20-μl aliquot of each cell extract was assayed for luciferase activity by using a luciferase assay system (Promega) according to the manufacturer's instructions. A TRILUX 1450 luminescence counter (Perkin-Elmer) was used for detection.
Inoculation of rhesus macaques with HEV genomic RNA. RNA from 10 μg of BglII-linearized plasmid was transcribed in vitro in a 200-μl reaction mixture in the presence of the m7G(5')ppp(5')G cap analog by using a T7 Megascript kit (Ambion) according to the manufacturer's instructions. The integrity and yield of the RNA were verified by gel electrophoresis of 5 μl of transcription reaction mixture on a nondenaturing agarose gel. The remaining 195 μl of the transcription reaction mixture was diluted with 805 μl of PBS without calcium and magnesium and injected into multiple sites in the liver of a rhesus macaque by percutaneous intrahepatic injection guided by ultrasound as described previously for SK-HEV-2 (8). Each chimeric construct was tested in two rhesus monkeys simultaneously. Serum samples were monitored weekly for serum liver enzyme levels (alanine aminotransferase [ALT]) (Anilytics, Gaithersburg, Md.), anti-HEV, and viral genomes. ALT levels that were 2 times the mean preinoculation level were considered elevated and indicated hepatitis. The preinoculation level was calculated as the geometric mean for three weekly samples collected at week zero and before inoculation. Anti-HEV were detected by using an in-house enzyme-linked immunosorbent assay (ELISA) as described previously (9). Viral genomes were quantified by HEV-specific real-time RT-PCR of total RNA extracted from 100 μl of serum as described previously (8).
The macaques used in the study were housed at Bioqual (Rockville, Md.). The housing, maintenance, and care of the animals met or exceeded all requirements for primate husbandry, as specified in the Guide for the Care and Use of Laboratory Animals (27).
Generation of SK-HEV-2 and Sar/3'swine pool. Feces collected from rhesus macaques infected by intrahepatic transfection (8) with Sar55 and Sar/3'swine recombinant genomes were diluted to 10% with PBS, clarified by centrifugation at 3,000 rpm in an IEC Centra-8R centrifuge (International Equipment Company) for 30 min at 4°C, and stored frozen in liquid nitrogen. Genome titers determined by real-time RT-PCR as described previously (8) were 4.8 x 106 genomes of Sar55 per ml and 1.9 x 106 genomes of Sar/3'swine per ml.
Inoculation of pigs with HEV. Two each of 60-day-old specific-pathogen-free (SPF) pigs, housed at the University of Iowa, were inoculated intraveneously with Sar55 and Sar/3'swine as described previously (23). All pigs were found to be seronegative for IgG anti-HEV by an ELISA prior to inoculation. Two of the SPF pigs were inoculated intravenously with 1 ml each of a 10% fecal suspension containing 4.8 x 106 genomes of HEV strain Sar55. The other two were inoculated intravenously with 1 ml each of a 10% fecal suspension containing 1.9 x 106 genomes of HEV chimera Sar/3'swine. Serum samples collected prior to inoculation and weekly until the end of the study were monitored for anti-HEV (9).
Sequence analysis. Genome fragments amplified by HEV-specific RT-PCR from the serum of inoculated animals were purified after agarose gel electrophoresis (QiaQuick; Qiagen) and sequenced directly with an automated sequencer.
RNA structural predictions were performed with the MFOLD program of the Genetics Computer Group (GCG) suite (Wisconsin package, version 10.0) (46).
RESULTS
3'-Terminal intergenotypic chimeras of HEV replicate in cell cultures. Previously it was shown that a single silent mutation introduced into the 3'-terminal region of ORF2 during cDNA cloning abolished or diminished the viability of a recombinant genotype 1 virus in chimpanzees and rhesus macaques (8). This mutation is in a region of genotype 1 strains predicted to form stem-loop structures that are involved in binding to the viral RdRp (1). Since this mutation would have disrupted only 1 bp in a predicted stem structure, it was possible that there were rigid requirements for the functionality of this region. However, the corresponding 3'-terminal regions of strain Mex-14 (genotype 2) (13) and strain Meng (genotype 3) (25) share only 81 and 74% nucleotide identities, respectively, with this region of HEV strain Sar55 (genotype 1) (Fig. 2A), and the MFOLD program predicts structures with different lengths of stems and loops among the three viral genomes (Fig. 2B). The polymerases of the selected genotype 2 and 3 strains also are not identical and share 89.5 and 88% amino acid identities, respectively, with that of Sar55. Therefore, it seemed possible that the polymerase and the 3'-terminal region of each virus had coevolved and that replication would occur only with the homologous combination of the two.
In order to examine the degree of tolerance for structural and sequence variations in the 3'-terminal region of HEV, we constructed chimeric genomes containing the backbone of the Sar55 infectious genome substituted at the 3' end [nt 7086 to the poly(A) tract] with the equivalent region from genotype 2 strain Mex-14 or genotype 3 swine strain Meng. Capped RNA transcripts produced by transcription in vitro of each chimeric cDNA clone and the parental clone, pSar55, were transfected into Huh-7 cells, and replication was ascertained by the detection of ORF2 and ORF3 proteins by confocal microscopy. The merged images of cells doubly labeled for ORF2 protein (green) and ORF3 protein (red) are shown in Fig. 3. Tested separately, neither primary antibody cross-reacted with the opposite secondary antibody (data not shown). Surprisingly, not only did both chimeras replicate (Fig. 3A and B), but also there was no discernible difference in the number of positive cells, time of appearance, or intensity of ORF2 and ORF3 staining compared with the results from a transfection with the parental clone, pSar55 (data not shown).
Evaluation of an HEV replicon expressing luciferase. In order to compare directly the replicative abilities of the chimeras, a quantitative assay was required. For this purpose, an HEV replicon, pE/luc, in which luciferase translation began at the methionine initiation codon of ORF2 was constructed (Fig. 1). The replicon contained the complete 5' NCR, ORF1, and the 3'-terminal region of the infectious genome, but ORF3 sequences were truncated to encode only 14 N-terminal amino acids and ORF2 expression was abolished due to deletion of the N-terminal half of the gene and a resultant frameshift that affected the remainder. As a negative control, the highly conserved GDD motif found in the catalytic portion of RdRp (15, 16) was mutated to GAD (pE/luc-GAD) to inactivate the polymerase expressed by pE/luc (20, 41) (Fig. 1).
Capped or uncapped transcripts synthesized in vitro from pE/luc or pE/luc-GAD, respectively, were transfected into Huh-7 cells, and intracellular luciferase activity was monitored. The luciferase activity measured up to 8 h posttransfection was equivalent to the background emission from uninfected cells and from cells transfected with the negative control transcription mixture lacking T7 RNA polymerase (mock-transfected cells) (data not shown). Capped or uncapped E/luc or E/luc-GAD expressed very low levels of luciferase during the first 24 h posttransfection. However, whereas the levels of luciferase expressed by the uncapped genome and by the E/luc-GAD genome remained very low, the levels of luciferase expressed by the E/luc genome increased dramatically after 1 day posttransfection. Figure 4A shows a representative profile of the expression of luciferase monitored as luciferase activity posttransfection. Whereas luciferase activity in E/luc-GAD-transfected cells increased less than 2-fold between days 1 and 2 posttransfection, in E/luc-transfected cells there was a significant increase in luciferase activity of 12-fold. The difference between E/luc and E/luc-GAD continued to increase through day 4, resulting in a level of luciferase activity in E/luc 122-fold higher than that in E/luc-GAD in this set of experiments. In repeated experiments, E/luc showed, on average, 93-fold more luciferase activity than E/luc-GAD by day 4. Data obtained after 4 days posttransfection were less reliable and showed reduced luciferase activity, most likely due to a loss of cells from the monolayer. Comparable kinetics were obtained when the transcription preparation was treated twice with DNase prior to transfection (data not shown). Therefore, the GAD mutant was unable to replicate or did so very inefficiently. Omission of the cap from pE/luc transcripts prevented the differential increase in luciferase activity and resulted in luciferase levels comparable to those in capped or uncapped E/luc-GAD (Fig. 4A). In comparison, E/luc RNA, when capped, was highly replication competent in Huh-7 cells.
Despite the convincing replication of the E/luc replicon compared to the negative control, hybridization analyses failed to detect newly synthesized positive- or negative-strand RNA. The large number of input transcripts, the modest number of cells successfully transfected, and the apparently low efficiency of RNA replication in the cells made it impossible to detect newly synthesized replicon RNA by Northern blot analysis or real-time RT-PCR (data not shown). E/luc- and E/luc-GAD-transfected cells both contained steadily degrading input RNA that was easily detected as a large smear by Northern blotting over the course of 6 days. Unfortunately, newly synthesized RNA was not detectable against this background.
Influence of 3'-terminal sequences on replication ability. The 3'-terminal regions comprising the intergenotypic chimeras or the single mutation at nt 7106 which had attenuated Sar55 for nonhuman primates was introduced into the replicon in place of the wild-type Sar55 sequence, and the ability of each mutant replicon was monitored by quantifying luciferase activity after transfection of Huh-7 cells. Once again, all replicons produced little luciferase during the first 24 h posttransfection, but between 2 and 4 days posttransfection, differences became evident (Fig. 4B). Genomes containing the single mutation at nucleotide 7106 replicated, albeit inefficiently. Although the luciferase activity in E/luc-7106 was 17-fold higher than that in E/luc-GAD by day 4, it was 5-fold lower than that of the parental replicon (E/luc) at that time. In contrast, throughout the study, luciferase production by the E/luc-3'Mex or E/luc-3'swine chimeric replicon was always higher than that by E/luc-7106, and the kinetics of luciferase production by the intergenotypic chimeras were more similar to those of parental replicon E/luc than to those of E/luc-7106 (Fig. 4B). At 4 days posttransfection, luciferase expression was higher than that of E/luc-GAD by 44-fold for E/luc-3'Mex and by 50-fold for E/luc-3'swine (52- and 55-fold, respectively, in a different experiment). Thus, the multiple sequence differences and predicted structure variations typical of genotype 2 and 3 strains had significantly less effect on the ability of the Sar55 polymerase to replicate the genome than did the single mutation at nt 7106.
Further analysis of the 3'-terminal regions was performed with replicons harboring additional mutations in these regions. Inoculation of chimpanzees with capped transcripts synthesized in vitro from cDNA clone pSK-HEV-3 (containing the replication-inhibitory mutation at nt 7106) resulted in the selection of new 3'-terminal region mutations that were postulated to increase replication efficiency in vivo (8). The virus isolated from one animal had at nt 7097 a new mutation (G to A) which did not restore the wild-type structure (Sar55) but did cause slight changes in the stem downstream of the mutation at nt 7106 in the predicted secondary structure (Fig. 2B). The virus isolated from the second animal had at a different site (nt 7144) a new mutation (C to A) which was a covariant of the mutation at nt 7106 in the predicted secondary structure of the 3'-terminal region and restored the wild-type structure (Fig. 2B). In order to determine whether these mutations did indeed affect viral replication, replicons pE/luc-7106/7097 and pE/luc-7106/7144 were constructed to contain the additional mutations at nt 7097 and nt 7144, respectively, in the pE/luc-7106 background. RNA replication was again evaluated by monitoring luciferase activity in transfected Huh-7 cells over a period of 4 days and was compared to RNA replication in parental replicon E/luc, E/luc-7106, E/luc-GAD, and mock-transfected cells (Fig. 4C). Luciferase production by E/luc-7106 was higher than that by the null polymerase mutant E/luc-GAD but still was sixfold lower than that by wild-type E/luc on day 4. The addition of the mutation at nt 7097 to E/luc-7106 increased luciferase production 2.4-fold over that by E/luc-7106 by day 4, and the addition of the mutation at nt 7144 to E/luc-7106 increased luciferase activity even more (4-fold over that in E/luc-7106). These results confirmed the hypothesis that these mutations had been selected because they increased viral replication and suggested that mutations which affect the secondary structure in this region can influence RNA replication.
3'-terminal HEV chimeric genomes are infectious in vivo. Since Huh-7 cells grown in cell cultures are not the same as hepatocytes within a liver, it was of interest to determine whether the 3'-terminal chimeras also could replicate in vivo. Capped transcripts synthesized in vitro from full-length cDNA clones pSar/3'swine and pSar/3'Mex were tested for their infectivity by intrahepatic inoculation into the livers of two rhesus monkeys each. Seroconversion to anti-HEV, serum liver enzyme levels, and viremia were monitored weekly for 12 weeks and compared with the results obtained with previously tested transfected Sar55 (wild type) (8) (Fig. 5). All six animals developed anti-HEV between 5 and 8 weeks posttransfection, concurrent with or shortly after the time at which peak amounts of viral RNA were detected in the serum. Overall, there were no differences among the animals with regard to any of the parameters measured that could not be explained by normal biological variations; e.g., the variations between two animals receiving the same inoculum were comparable to those between two animals receiving different inocula. The 3'-terminal sequences of HEV were amplified from each of the six rhesus monkeys by RT-PCR and were found to be identical to those of the input genome for each monkey. Therefore, in vivo, the 3'-terminal chimeras replicated with an efficiency indistinguishable from that of the wild-type parent.
3'-Terminal sequences do not specify host range. It was previously shown that Sar55 virus was infectious for nonhuman primates but was not able to infect swine even when a well-characterized infectious stock was inoculated intravenously (23). In contrast, swine HEV strain Meng was able to infect rhesus monkeys and chimpanzees as well as swine, indicating that it had a broader host range that allowed it to cross species barriers (24).
The 3' NCR of viruses has been reported to interact with host proteins (4, 28, 40), and since there was a 26% difference in nucleotide sequence in this region between strains Sar55 and Meng, it seemed possible that this region might be responsible for the ability of the swine virus to infect different species. To test this hypothesis, we made one virus pool containing HEV from a rhesus monkey that had been transfected with transcripts synthesized in vitro from pSar55 and another virus pool from a monkey that had been transfected with transcripts synthesized in vitro from pSar/3'swine; we inoculated approximately 106 genomes of Sar55 or of Sar/3'swine into swine. SPF pigs were monitored weekly for 15.5 weeks. Seroconversion was not detected in any pig (data not shown), confirming that genotype 1 human strain Sar55 could not infect pigs and indicating that the 3'-terminal region of the swine HEV genome was not sufficient to confer to Sar55 the ability to infect swine productively.
DISCUSSION
The production of luciferase encoded by a replicon of HEV provided a convenient method of demonstrating and analyzing HEV genomic replication. The results obtained with the replicon containing wild-type Sar55 sequences indicated that HEV RNA replication in cell cultures occurred independently of functional ORF2 and ORF3 proteins, confirming results obtained previously with a green fluorescent protein (GFP)-producing replicon (6), for which different parameters were measured. In the GFP-producing system, fluorescence-activated cell sorting analysis was used to quantify the numbers of transfected cells (about 10% maximum), thus providing an estimate of individual genome viability. That study showed that although uncapped genomes could initiate an infection in vitro, it was a rare event compared to the number of infections initiated with capped genomes. Additionally, detectable GFP was not produced when cells were transfected with genomes containing the GAD mutant polymerase. In the replicon system described here, luciferase production was averaged over the entire population of cells, and the amount of luciferase produced by the uncapped transcripts was not larger than that produced by the GAD mutant. Thus, data from both systems confirmed that capping of the HEV genome is very important for viability, but the luciferase system was not as sensitive as the GFP system for detecting rare events. However, the luciferase system was quite useful for making other quantitative comparisons.
The luciferase-expressing replicons all produced very low and nearly identical levels of luciferase during the first 12 h posttransfection. This result most likely was due to translation of the transfecting RNA by an unknown mechanism. This early expression of the luciferase gene did not seem to require a cap, because the uncapped and capped genomes produced equivalent luciferase activities during this time. It also did not reflect replication, because the activities of the GAD and GDD constructs consistently differed by less than twofold at 12 h posttransfection. The amount of luciferase produced should reflect the synthesis and replication of the subgenomic mRNA believed to encode ORF2 protein normally and luciferase in this case (6, 34, 42). Therefore, the increase in luciferase activity observed between 1 and 2 days with wild-type replicon E/luc compared to polymerase-defective replicon E/luc-GAD suggested that the replication complexes or viral enzymes required for genome replication did not accumulate to significant levels until that time, that the replication of HEV is a relatively inefficient process, and that the replication cycle is prolonged, at least in cell cultures.
The inefficiency of HEV replication made it necessary to rely on a reporter such as luciferase to detect differences in RNA replication between the wild-type genome and one encoding mutant RdRp or containing 3'-terminal mutations. The failure to detect newly synthesized replicon RNA in our system reflects the large amount of transcribed RNA necessary for transfection, which would contain a mixture of capped and uncapped RNAs. The large amount of input RNA required to transfect a significant number of cells caused such a high background that it and its degradation products prevented detection by Northern blotting of any newly synthesized replicon RNA or subgenomic RNAs. Similarly, the input RNA produced such a high signal in RT-PCR and newly synthesized RNA represented such a small fraction that it could not be reliably detected.
The luciferase system was ideal for studying the effect on RNA replication of mutations in the 3'-terminal region. Of particular interest was the mutation at nt 7106, which was previously shown to attenuate HEV for nonhuman primates (8). This attenuation could reflect the inefficient replication of viral RNA or the inability to package the genome into infectious virions. The results of the replicon assays clearly demonstrated that luciferase activity in E/luc-7106 was diminished compared to that in the parent replicon and thus established the fact that this single mutation was in a region critical for RNA replication.
The decreased level of replication induced by this single mutation was in contrast to the more robust level of replication observed after transfection with E/luc-3'Mex or E/luc-3'swine. In the 109-nt tract exchanged to generate the chimeras, strain Mex-14 differed from Sar55 at 21 nt, including nine insertions; thus, there was about a 20% difference in the sequences. Swine strain Meng differed even more from Sar55 in this region, with 29 nt differences, including six insertions (for a nucleotide sequence difference of 26%). Additionally, the predicted secondary structures of the genomes were all different from each other in this region. However, in spite of these large differences, both chimeric replicons replicated more efficiently in cell cultures than did the virus with a mutation at nt 7106 and, as full-length chimeras, both produced infections that were indistinguishable from those produced by wild-type Sar55.
Agrawal et al. reported that 3'-terminal sequences of genotype 1 HEV genomes form two putative stem-loop structures, which seem to be important in concert for binding to RdRp in vitro (1). In light of the results obtained with the chimeras, there seems to be a considerable degree of flexibility, with regard to RNA replication, in the precise sequence and structure: the formation of the two stem structures by itself is not sufficient, since the seemingly minor change caused by the mutation at nt 7106, which would abolish only one putative base pair within stem-loop 2, had a significant inhibitory effect on RNA replication. Restoration of this particular part of stem-loop 2 by a covariant mutation significantly increased the efficiency of RNA replication, although not to wild-type levels; these results suggested that the general structure generated by the 3'-terminal sequences, including the ORF2 sequences, is important but that the sequences themselves may also have an effect. It is interesting that the region around nt 7106 is completely base paired in strain Mex-14, whereas in both Sar55 and the swine strain, this region contains a small loop. The stem mutation at nt 7097, selected in the second chimpanzee, had a larger effect on the predicted structure and increased viral replication in vitro to a lesser extent, probably because the structure, although more favorable, still deviated from that of the wild type. However, the hypothesis that the mutations at nt 7097 and nt 7144 were selected in the chimpanzees because they promoted genome replication was supported by the results of the luciferase assays.
The accumulating data suggesting that HEV may be a zoonosis make it interesting to determine why strain Meng isolated from swine was able to cross species borders, whereas strains Sar55 and Mex-14 isolated from humans were not. The 3' NCRs of some viruses have been shown to interact with host proteins (4, 28, 40). Therefore, given the extent of differences between strains Sar55 and Meng in these regions, it appeared reasonable to determine whether the Sar/3'swine chimera could infect pigs. The fact that a large dose of chimeric virus did not infect either of the two inoculated pigs indicated that the 3'-terminal region is not responsible for host range determination or that, if it is involved, other regions of the swine genome are also required for infection of pigs. Once an infectious clone of swine HEV is developed, it should be possible to answer the question of how host range is determined.
The robust viability of the full-length chimeras and the appreciable replication competence of the corresponding replicons are in contrast to the diminished replication of the virus with a mutation at nt 7106 and its replicon and underscore the difficulty in trying to determine which sequences or structures in the 3'-terminal region are critical for replication. The current study validates the use of a luciferase-producing replicon to study these parameters in vitro to reliably predict their involvement in vivo. It should now be possible to undertake saturation-mutagenesis studies of the replicon to determine which mutants can replicate in vitro and then transfect those into animals to determine which regions are critical for replication in vivo.
ACKNOWLEDGMENTS
This work was supported in part by National Institute of Allergy and Infectious Diseases contract no. 1-A0-02733.
We thank Ron Engle and Claro Yu for performing the ELISA, Owen Schwartz for guidance and help with confocal microscopy, and Natalia Teterina for valuable advice on RNA detection.
REFERENCES
Agrawal, S., D. Gupta, and S. K. Panda. 2001. The 3' end of hepatitis E virus (HEV) genome binds specifically to the viral RNA-dependent RNA polymerase (RdRp). Virology 282:87-101.
Arankalle, V. A., L. P. Chobe, A. M. Walimbe, P. N. Yergolkar, and G. P. Jacob. 2003. Swine HEV infection in south India and phylogenetic analysis (1985-1999). J. Med. Virol. 69:391-396.
Bradley, D. W., K. Krawczynski, E. H. Cook, Jr., K. A. McCaustland, C. D. Humphrey, J. E. Spelbring, H. Myint, and J. E. Maynard. 1987. Enterically transmitted non-A, non-B hepatitis: serial passage of disease in cynomolgus macaques and tamarins and recovery of disease-associated 27- to 34-nm viruslike particles. Proc. Natl. Acad. Sci. USA 84:6277-6281.
Dollenmaier, G., and M. Weitz. 2003. Interaction of glyceraldehyde-3-phosphate dehydrogenase with secondary and tertiary RNA structural elements of the hepatitis A virus 3' translated and non-translated regions. J. Gen. Virol. 84:403-414.
Emerson, S. U., D. Anderson, A. Arankalle, X.-J. Meng, M. Purdy, G. G. Schlauder, and S. A. Tsarev. 2004. Hepevirus, p. 851-855. In C. M. Fauquet, M. A. Mayo, J. Maniloff, U. Desselberger, and L. A. Ball (ed.), Virus taxonomy. VIIIth report of the ICTV. Elsevier/Academic Press, London, United Kingdom.
Emerson, S. U., H. Nguyen, J. Graff, D. A. Stephany, A. Brockington, and R. H. Purcell. 2004. In vitro replication of hepatitis E virus (HEV) genomes and of an HEV replicon expressing green fluorescent protein. J. Virol. 78:4838-4846.
Emerson, S. U., and R. H. Purcell. 2003. Hepatitis E virus. Rev. Med. Virol. 13:145-154.
Emerson, S. U., M. Zhang, X.-J. Meng, H. Nguyen, M. St. Claire, S. Govindarajan, Y. K. Huang, and R. H. Purcell. 2001. Recombinant hepatitis E virus genomes infectious for primates: importance of capping and discovery of a cis-reactive element. Proc. Natl. Acad. Sci. USA 98:15270-15275.
Engle, R. E., C. Yu, S. U. Emerson, X. J. Meng, and R. H. Purcell. 2002. Hepatitis E virus (HEV) capsid antigens derived from viruses of human and swine origin are equally efficient for detecting anti-HEV by enzyme immunoassay. J. Clin. Microbiol. 40:4576-4580.
Garkavenko, O., A. Obriadina, J. Meng, D. A. Anderson, H. J. Benard, B. A. Schroeder, Y. E. Khudyakov, H. A. Fields, and M. C. Croxson. 2001. Detection and characterisation of swine hepatitis E virus in New Zealand. J. Med. Virol. 65:525-529.
Haqshenas, G., H. L. Shivaprasad, P. R. Woolcock, D. H. Read, and X. J. Meng. 2001. Genetic identification and characterization of a novel virus related to human hepatitis E virus from chickens with hepatitis-splenomegaly syndrome in the United States. J. Gen. Virol. 82:2449-2462.
Hsieh, S. Y., X. J. Meng, Y. H. Wu, S. T. Liu, A. W. Tam, D. Y. Lin, and Y. F. Liaw. 1999. Identity of a novel swine hepatitis E virus in Taiwan forming a monophyletic group with Taiwan isolates of human hepatitis E virus. J. Clin. Microbiol. 37:3828-3834.
Huang, C. C., D. Nguyen, J. Fernandez, K. Y. Yun, K. E. Fry, D. W. Bradley, A. W. Tam, and G. R. Reyes. 1992. Molecular cloning and sequencing of the Mexico isolate of hepatitis E virus (HEV). Virology 191:550-558.
Huang, F. F., G. Haqshenas, H. L. Shivaprasad, D. K. Guenette, P. R. Woolcock, C. T. Larsen, F. W. Pierson, F. Elvinger, T. E. Toth, and X. J. Meng. 2002. Heterogeneity and seroprevalence of a newly identified avian hepatitis E virus from chickens in the United States. J. Clin. Microbiol. 40:4197-4202.
Kamer, G., and P. Argos. 1984. Primary structural comparison of RNA-dependent polymerases from plant, animal and bacterial viruses. Nucleic Acids Res. 12:7269-7282.
Koonin, E. V. 1991. The phylogeny of RNA-dependent RNA polymerases of positive-strand RNA viruses. J. Gen. Virol. 72:2197-2206.
Koonin, E. V., A. E. Gorbalenya, M. A. Purdy, M. N. Rozanov, G. R. Reyes, and D. W. Bradley. 1992. Computer-assisted assignment of functional domains in the nonstructural polyprotein of hepatitis E virus: delineation of an additional group of positive-strand RNA plant and animal viruses. Proc. Natl. Acad. Sci. USA 89:8259-8263.
Korkaya, H., S. Jameel, D. Gupta, S. Tyagi, R. Kumar, M. Zafrullah, M. Mazumdar, S. K. Lal, L. Xiaofang, D. Sehgal, S. R. Das, and D. Sahal. 2001. The ORF3 protein of hepatitis E virus binds to Src homology 3 domains and activates MAPK. J. Biol. Chem. 276:42389-42400.
Krawczynski, K., S. Kamili, and R. Aggarwal. 2001. Global epidemiology and medical aspects of hepatitis E. Forum (Genova) 11:166-179.
Lohmann, V., F. Korner, U. Herian, and R. Bartenschlager. 1997. Biochemical properties of hepatitis C virus NS5B RNA-dependent RNA polymerase and identification of amino acid sequence motifs essential for enzymatic activity. J. Virol. 71:8416-8428.
Magden, J., N. Takeda, T. Li, P. Auvinen, T. Ahola, T. Miyamura, A. Merits, and L. Kaariainen. 2001. Virus-specific mRNA capping enzyme encoded by hepatitis E virus. J. Virol. 75:6249-6255.
Meng, X. J. 2000. Zoonotic and xenozoonotic risks of hepatitis E virus. Infect. Dis. Rev. 2:35-41.
Meng, X. J., P. G. Halbur, J. S. Haynes, T. S. Tsareva, J. D. Bruna, R. L. Royer, R. H. Purcell, and S. U. Emerson. 1998. Experimental infection of pigs with the newly identified swine hepatitis E virus (swine HEV), but not with human strains of HEV. Arch. Virol. 143:1405-1415.
Meng, X. J., P. G. Halbur, M. S. Shapiro, S. Govindarajan, J. D. Bruna, I. K. Mushahwar, R. H. Purcell, and S. U. Emerson. 1998. Genetic and experimental evidence for cross-species infection by swine hepatitis E virus. J. Virol. 72:9714-9721.
Meng, X. J., R. H. Purcell, P. G. Halbur, J. R. Lehman, D. M. Webb, T. S. Tsareva, J. S. Haynes, B. J. Thacker, and S. U. Emerson. 1997. A novel virus in swine is closely related to the human hepatitis E virus. Proc. Natl. Acad. Sci. USA 94:9860-9865.
Nakabayashi, H., K. Taketa, K. Miyano, T. Yamane, and J. Sato. 1982. Growth of human hepatoma cell lines with differentiated functions in chemically defined medium. Cancer Res. 42:3858-3863.
National Research Council. 1996. Guide for the care and use of laboratory animals. National Academy Press, Washington, D.C.
Petrik, J., H. Parker, and G. J. Alexander. 1999. Human hepatic glyceraldehyde-3-phosphate dehydrogenase binds to the poly(U) tract of the 3' non-coding region of hepatitis C virus genomic RNA. J. Gen. Virol. 80:3109-3113.
Purcell, R. H., and S. U. Emerson. 2001. Hepatitis E virus, p. 3051-3061. In D. M. Knipe, P. M. Howley, D. E. Griffin, R. A. Lamb, M. A. Martin, B. Roizman, and S. E. Straus (ed.), Fields virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Sambrook, J., and D. W. Russell. 2001. Molecular cloning: a laboratory manual, 3rd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y.
Schlauder, G. G., and I. K. Mushahwar. 2001. Genetic heterogeneity of hepatitis E virus. J. Med. Virol. 65:282-292.
Schofield, D. J., J. Glamann, S. U. Emerson, and R. H. Purcell. 2000. Identification by phage display and characterization of two neutralizing chimpanzee monoclonal antibodies to the hepatitis E virus capsid protein. J. Virol. 74:5548-5555.
Takahashi, M., T. Nishizawa, H. Miyajima, Y. Gotanda, T. Iita, F. Tsuda, and H. Okamoto. 2003. Swine hepatitis E virus strains in Japan form four phylogenetic clusters comparable with those of Japanese isolates of human hepatitis E virus. J. Gen. Virol. 84:851-862.
Tam, A. W., M. M. Smith, M. E. Guerra, C. C. Huang, D. W. Bradley, K. E. Fry, and G. R. Reyes. 1991. Hepatitis E virus (HEV): molecular cloning and sequencing of the full-length viral genome. Virology 185:120-131.
Tei, S., N. Kitajima, K. Takahashi, and S. Mishiro. 2003. Zoonotic transmission of hepatitis E virus from deer to human beings. Lancet 362:371-373.
Torresi, J., F. Li, S. A. Locarnini, and D. A. Anderson. 1999. Only the non-glycosylated fraction of hepatitis E virus capsid (open reading frame 2) protein is stable in mammalian cells. J. Gen. Virol. 80:1185-1188.
Tsarev, S. A., S. U. Emerson, G. R. Reyes, T. S. Tsareva, L. J. Legters, I. A. Malik, M. Iqbal, and R. H. Purcell. 1992. Characterization of a prototype strain of hepatitis E virus. Proc. Natl. Acad. Sci. USA 89:559-563.
Tyagi, S., S. Jameel, and S. K. Lal. 2001. Self-association and mapping of the interaction domain of hepatitis E virus ORF3 protein. J. Virol. 75:2493-2498.
Tyagi, S., H. Korkaya, M. Zafrullah, S. Jameel, and S. K. Lal. 2002. The phosphorylated form of the ORF3 protein of hepatitis E virus interacts with its non-glycosylated form of the major capsid protein, ORF2. J. Biol. Chem. 277:22759-22767.
Waggoner, S., and P. Sarnow. 1998. Viral ribonucleoprotein complex formation and nucleolar-cytoplasmic relocalization of nucleolin in poliovirus-infected cells. J. Virol. 72:6699-6709.
Wang, X., and S. Gillam. 2001. Mutations in the GDD motif of rubella virus putative RNA-dependent RNA polymerase affect virus replication. Virology 285:322-331.
Xia, X., R. Huang, and D. Li. 2000. Studies on the subgenomic RNAs of hepatitis E virus. Wei Sheng Wu Xue Bao 41:622-627.
Zafrullah, M., M. H. Ozdener, R. Kumar, S. K. Panda, and S. Jameel. 1999. Mutational analysis of glycosylation, membrane translocation, and cell surface expression of the hepatitis E virus ORF2 protein. J. Virol. 73:4074-4082.
Zafrullah, M., M. H. Ozdener, S. K. Panda, and S. Jameel. 1997. The ORF3 protein of hepatitis E virus is a phosphoprotein that associates with the cytoskeleton. J. Virol. 71:9045-9053.
Zhou, Y.-H., R. H. Purcell, and S. U. Emerson. 2004. An ELISA for putative neutralizing antibidies to hepatitis E virus detects antibodies to genotype 1, 2, 3 and 4. Vaccine 22:2578-2585.
Zuker, M., D. H. Mathews, and D. H. Turner. 1999. Algorithms and thermodynamics for RNA secondary structure prediction: a practical guide, p. 11-43. In J. Barciszewski and B. F. C. Clark (ed.), RNA biochemistry and biotechnology. NATO ASI series. Kluwer Academic Press, Boston, Mass.(Judith Graff, Hanh Nguyen)