Hallmarks of hepatitis C virus in equine hepacivirus

Abstract

Equine hepacivirus (EHcV) has been identified as a closely related homologue of hepatitis C virus (HCV) in the United States, the United Kingdom, and Germany, but not in Asian countries. In this study, we genetically and serologically screened 31 serum samples obtained from Japanese-born domestic horses for EHcV infection and subsequently identified 11 PCR-positive and 7 seropositive serum samples. We determined the full sequence of the EHcV genome, including the 3' untranslated region (UTR), which had previously not been completely revealed. The polyprotein of a Japanese EHcV strain showed approximately 95% homology to those of the reported strains. HCV-like cis-acting RNA elements, including the stem-loop structures of the 3' UTR and kissing-loop interaction were deduced from regions around both UTRs of the EHcV genome. A comparison of the EHcV and HCV core proteins revealed that Ile(190) and Phe(191) of the EHcV core protein could be important for cleavage of the core protein by signal peptide peptidase (SPP) and were replaced with Ala and Leu, respectively, which inhibited intramembrane cleavage of the EHcV core protein. The loss-of-function mutant of SPP abrogated intramembrane cleavage of the EHcV core protein and bound EHcV core protein, suggesting that the EHcV core protein may be cleaved by SPP to become a mature form. The wild-type EHcV core protein, but not the SPP-resistant mutant, was localized on lipid droplets and partially on the lipid raft-like membrane in a manner similar to that of the HCV core protein. These results suggest that EHcV may conserve the genetic and biological properties of HCV.

Importance: EHcV, which shows the highest amino acid or nucleotide homology to HCV among hepaciviruses, was previously reported to infect horses from Western, but not Asian, countries. We herein report EHcV infection in Japanese-born horses. In this study, HCV-like RNA secondary structures around both UTRs were predicted by determining the whole-genome sequence of EHcV. Our results also suggest that the EHcV core protein is cleaved by SPP to become a mature form and then is localized on lipid droplets and partially on lipid raft-like membranes in a manner similar to that of the HCV core protein. Hence, EHcV was identified as a closely related homologue of HCV based on its genetic structure as well as its biological properties. A clearer understanding of the epidemiology, genetic structure, and infection mechanism of EHcV will assist in elucidating the evolution of hepaciviruses as well as the development of surrogate models for the study of HCV.

FIG 1

Detection and genetic analyses of NPHV genomic RNA in sera of Japanese domestic horses. (A) Total RNAs extracted from 31 equine sera and normal rabbit serum (Rb) as a negative control were subjected to RT-PCR analysis. Hokkaido Farm A, Tokyo Farm B, and Tokyo Farm C indicate the farms where the individual horses were reared. Three sets of primers, NPHV-F1 and NPHV-R1, NPHV-F2 and NPHV-R2, and NPHV-F3 and NPHV-R3, were used to amplify NPHV-specific gene regions. The PCR products were electrophoresed and stained with ethidium bromide. (B) Total RNAs were isolated from sera, reverse-transcribed, and estimated as a copy number per ml. Normal rabbit serum was used as a negative control. The dashed line indicates the cutoff level.

FIG 2

Serological screening of Japanese-born domestic horses. Lysates of 293FT cells transfected with an empty plasmid (a negative reference, N) or the plasmid encoding EHcVc (a positive reference, P) were subjected to Western blotting using serum from each horse. The serum response “+” indicates that the protein band with the same molecular size as the EHcV core protein was specifically detected in the “P” lane, but not in the “N” lane, while the serum response “−” indicates that the protein band with the same molecular size as the EHcV core protein was detected in neither the “P” lane nor the “N” lane. Both antibodies to the FLAG tag and to the EHcV core protein were used as serum positive controls, while protein amounts were standardized with blotting using the antibody to beta-actin. “No serum” indicates the membrane was incubated without primary antibodies but with HRP-conjugated anti-horse IgG antibodies as a background of the secondary antibody.

FIG 3

Phylogenetic analysis of the EHcV gene. Neighbor-joining trees of the nucleotide sequences from the NS3 (A) and NS5B (B) regions of the NPHV, HCV, and GBV-B strains are shown (23). Trees were constructed by the maximum composite likelihood method calculated using the program MEGA5 (24). The percentage of replicate trees in which the associated taxa were clustered together in the bootstrap test (1,000 replicates) is indicated next to the branches. Analyses were carried out using 10 strains of EhcV, JPN3/JAPAN/2013, A6-066 (GenBank accession no. JQ434003), B10-022 (GenBank accession no. JQ434004), F8-068 (GenBank accession no. JQ434005), G1-073 (GenBank accession no. JQ434002), G5-077 (GenBank accession no. JQ434006), H3-011 (GenBank accession no. JQ434008), H10-094 (GenBank accession no. JQ434007), NZP1 (GenBank accession no. JQ434001), and AAK-2011 (canine hepacivirus; GenBank accession no. JF744991); 4 strains of HCV, H77 (genotype 1a; GenBank accession no. NC004102), LyHCV (genotype 1b; GenBank accession no. AB779562), HC-J6CH (genotype 2a; GenBank accession no. NC009823), and JFH1 (genotype 2a; GenBank accession no. AB047639); 3 strains of bat hepacivirus, PDB-112 (GenBank accession no. KC796077), PDB-445 (GenBank accession no. KC796091), and PDB-829 (GenBank accession no. KC796074); 3 strains of rodent hepacivirus, RMU10-3382 (GenBank accession no. KC411777), NLR-AP-70 (GenBank accession no. KC411784), and SAR-46 (GenBank accession no. KC411807); and another primate hepacivirus, GBV-B (GenBank accession no. NC001655). The Japanese strain JPN3/JAPAN/2013 (GenBank accession no. AB863589) is underlined.

FIG 4

RNA structure analysis of the 5′ and 3′ ends of EHcV. (A) Alignment of the 5′ UTRs of the EHcV strains. Asterisks and black lines indicate variable residues and miRNA-targeting regions, respectively. SL structures are enclosed in rectangles. GenBank numbers are listed in the legend to Fig. 3. (B) Alignment of the 3′ UTRs of the EHcV strains. The 3′-terminal sequences with GenBank numbers AB921150 and AB921151 were determined using serum samples 5 and 1, respectively, and were aligned with that of JPN3/JAPAN/2013 (GenBank accession no. AB863589). (C) The secondary structures of the 5′ UTR, NS5B-coding region, and 3′ UTR were predicted on the basis of minimum free energy predictions. The stem loops in the 5′ UTR (left) were designated according to the stem loops of the HCV 5′-UTR structures (30). A gray line, double line, and single line indicate a pseudoknot structure, miR-122-target region, and complementary sequence, respectively. The stem loops in the NS5B-coding region and 3′ UTR (right) were designated to correspond to the stem loops embedded in the HCV 3′-terminal region (32, 33). The NS5B-coding region, (A)-rich region, and 3′-X-tail sequence are indicated under the schematic structure. Start and stop codons are enclosed by rectangles.

FIG 5

Amino acid alignment and hydrophobicity of EHcV and HCV core proteins. (A) Alignment of the core proteins of EHcV (JPN3/JAPAN/2013) and HCV genotype1b (Con1; GenBank accession number AJ238799). Asterisks indicate identical amino acid residues. Bars indicate gaps to achieve maximum amino acid matching. The black and white arrowheads indicate the predicted cleavage site of the core protein of HCV by SPP and signal peptidase, respectively. The EHcV core protein was composed of three domains, domain 1 (a black line, residues 2 to 132), domain 2 (a broken black line, residues 133 to 187), and domain 3 (a gray line, residues 188 to 204), relative to those of the HCV core protein (42). (B) Hydrophobicity plots of the EHcV and HCV core proteins were prepared by the method of Kyte and Doolittle (26). The horizontal and vertical axes represent amino acid position and hydrophobicity, respectively.

FIG 6

Intramembrane processing of the EHcV core protein by SPP. (A) The plasmids encoding HCVc, HCVc-mt, EHcVc, and EHcVc-mt are shown as a schematic diagram. Their C-terminal regions (171 to 190, HCV core protein; 185 to 203, EHcV core protein) were aligned. The C-terminal Ala of each core protein was replaced with Arg (R) to prevent signal peptidase-dependent cleavage for the detection of the SPP-uncleaved core protein with the anti-HA antibody. Bars indicate the amino acids that were the same as those of the wild-type residues. (B) The secondary protein structures in the C-terminal transmembrane regions of the HCV and EHcV core proteins and mutants were predicted by the method of Garnier et al. (25). Arrows indicate putative SPP cleavage sites. (C) HCVc, HCVc-mt, EHcVc, and EHcVc-mt were expressed in 293FT cells and immunoblotted with the anti-FLAG and -HA antibodies. (D) HCVc or EHcVc was expressed with SPP-wt or SPP-D219A in the 293FT cell line. HCVc-mt and EHcVc-mt were expressed in the absence of SPP-wt and SPP-D219A as uncleavable controls. (E) HCVc or EHcVc was coexpressed with or without SPP-D219A. SPP-D219A was pulled down with Ni beads. Coprecipitated proteins were immunoblotted with the anti-FLAG antibody.

FIG 7

Intracellular localization of hepacivirus core proteins. HCVc, HCVc-mt, EHcVc, or EHcVc-mt was expressed in the Huh7OK1 cell line. The resulting cells were stained with Bodipy 558/568 (red) and then fixed with 4% paraformaldehyde at 24 h posttransfection, permeabilized, and subjected to indirect immunofluorescence staining. Each core protein was detected using mouse anti-FLAG antibodies and then Alexa 488-conjugated anti-mouse IgG (green). Cell nuclei were stained with DAPI after fixation (blue).

FIG 8

The EHcV core protein partially migrated to the DRM fraction after SPP-dependent processing. 293FT cells expressing either EHcVc or EHcVc-mt were homogenized with or without 1% Triton X-100 and then subjected to a flotation assay. Proteins in each fraction were concentrated with cold acetone and then subjected to Western blotting using the anti-FLAG, anti-calreticulin, or anti-caveolin-1 antibody.