A Novel Human Pegivirus, HPgV-2 (HHpgV-1), Is Tightly Associated With Hepatitis C Virus (HCV) Infection and HCV/Human Immunodeficiency Virus Type 1 Coinfection

Abstract

Background

Human pegivirus type 2 (HPgV-2) is a novel blood-borne human pegivirus that mainly infects hepatitis C virus (HCV)–infected subjects. We have investigated the prevalence of HPgV-2 in China, its association with HCV and human immunodeficiency virus type 1 (HIV-1), and the impact on HCV viral load and liver damage.

Methods

A cross-sectional study was conducted with both blood donors and HCV- and HIV-1–infected patients in Guangzhou, China. All subjects were screened for anti-HPgV-2 and HPgV-2 RNA. Demographic and clinical information were obtained from electronic medical records.

Results

We tested 8198 serum or plasma samples. Only 0.15% (6/4017) of healthy blood donors were positive for anti-HPgV-2 and negative for HPgV-2 RNA. No HPgV-2 viremia was detected in hepatitis B virus– or HIV-1–monoinfected individuals. The relatively high frequency of HPgV-2 infection was observed in 1.23% (30/2440) and 0.29% (7/2440) of HCV-infected persons by serological assay and reverse-transcription polymerase chain reaction, respectively. Furthermore, anti-HPgV-2 and HPgV-2 RNA were detected in 8.91% (18/202) and 3.47% (7/202), respectively, of HCV/HIV-1–coinfected subjects. HPgV-2 persistent infection was documented in about 30% of anti-HPgV-2–positive individuals. In addition, HPgV-2 infection may not affect HCV-related liver injury and HCV viral load.

Conclusions

Our results indicate the rarity of HPgV-2 infection in the general population and tight association with HCV, in particular with HCV/HIV-1 coinfection. HPgV-2 appears not to worsen HCV-related liver damage. Our study provides new findings about the association of HPgV-2 and HCV/HIV-1 and the impact of HPgV-2 infection on HCV replication and pathogenesis.

The Flaviviridae constitute an important and continuously expanding family. In the past several years, a growing number of new viruses that are genetically related to the Hepacivirus and Pegivirus genera within the Flaviviridae family have been identified [1, 2]. Thus, Smith et al proposed an update of the taxonomy of the genera Hepacivirus and Pegivirus to include 14 Hepacivirus species and 11 Pegivirus species [3]. Among them, a novel human pegivirus was independently identified in 2015 in the United States and provisionally assigned the name human hepegivirus 1 (HHpgV-1) by Kapoor et al [4] or human pegivirus 2 (HPgV-2) by Berg et al [5]. HHpgV-1 or HPgV-2 was also identified in people who inject drugs (PWID) in 2016 in the United Kingdom by Bonsall et al [6] and in 2017 in the United States by Kandathil et al [7]. These virus sequences share 94%–96% identity, indicating that they are actually the same virus. Although HHpgV-1/HPgV-2 shares some common features with hepatitis C virus (HCV) and other hepaciviruses, such as a typical type IV internal ribosome entry site, a core-like protein, and a highly glycosylated E2 protein [4], HHpgV-1 or HPgV-2 phylogenetically belongs to the genus of Pegivirus rather than Hepacivirus [4–6]. Therefore, in this study we prefer the name “HPgV-2” for the second human pegivirus.

The first human pegivirus (HPgV), formerly called GB virus C or hepatitis G virus [8, 9], infected approximately 1.5–2.5 billion people [10], including about 750 million people with current or active HPgV infection worldwide [11]. HPgV-2 is transmitted by transfusion, but only infects a very limited number of persons and is tightly associated with HCV infection [4–6, 12].

The pathogenicity of HPgV-2 has yet to be determined. In general, pegiviruses may lack pathogenicity [1]. However, a new pegivirus, Theiler disease–associated virus, has been found to cause Theiler disease, an acute hepatitis outbreak in horses [13]. In addition, HPgV may not cause any human diseases [10], and it displays beneficial effects on inhibition of human immunodeficiency virus type 1 (HIV-1) replication and prolonging survival of HIV-1–infected and Ebola virus–infected patients [14–16]. Therefore, it is critical to investigate the pathogenic potential of the new HPgV-2 and its impact on HCV infection and disease progression.

In this study, we expand the screening of HPgV-2 infection to determine its prevalence and distribution, the association with HCV and HIV-1 infection, and the impact of HPgV-2 infection on HCV infection and liver injury in China.

MATERIALS AND METHODS

Serum or Plasma Samples

A total of 8198 de-linked human serum or plasma samples were collected from different populations (Table 1). Written informed consent was obtained from individuals enrolled in this study, which has been approved by Ethics Committees of Southern Medical University and was performed according to the approved research protocols. All samples were stored at –80°C before analyses.

Measurement of Liver Enzymes, Fibrosis, and Screening for Hepatitis C Virus, Hepatitis B Virus, and Human Immunodeficiency Virus Type 1

Alanine aminotransferase (ALT) and aspartate aminotransferase (AST) were detected using reagents from Kehua Bio-engineering Co, Ltd (Shanghai, China). The reference range of ALT and AST is 0–50 U/L and 15–40 U/L, respectively. The AST-to-platelet ratio index (APRI) and Fibrosis-4 Index for Liver Fibrosis score (FIB-4) were measured according to the published methods [17, 18].

Screening for anti-HCV, antibody against hepatitis B core antigen, and hepatitis B surface antigen was performed by using enzyme-linked immunosorbent assay (ELISA) kits from Kehua. The immunoassay for detection of anti-HIV was an in-house assay [19]. The reverse-transcription polymerase chain reaction (RT-PCR) assays for detection of HIV-1 RNA, HPgV RNA, and HCV RNA have been reported previously [20–22]. The quantification of HCV viral load was determined by COBAS AmpliPrep/COBAS TaqMan test (Roche Applied Science, Mannheim, Germany).

Detection of Anti-Human Pegivirus Type 2 (HPgV-2) Antibody and HPgV-2 RNA

ELISA for detection of the anti-HPgV-2 antibody was developed in-house (Wang H, et al, unpublished data) by using the 3 HPgV-2–specific peptides (p4, p9, and p16) [5].

The HPgV-2 5ʹ untranslated region (5ʹUTR) was amplified using the primer pairs designed in our laboratory. The primers and RT-PCR conditions for HPgV-2 NS3 gene amplification have been described previously [6]. A quantitative PCR (qPCR) assay was adapted to determine HPgV-2 viral loads using primers and probes directed against the HPgV-2 5ʹ UTR and NS3 genes [5].

Genome Sequencing by Reverse-Transcription Polymerase Chain Reaction and Phylogenetic Analysis

Six fragments covering the near full-length genome (NFLG) of HPgV-2 were amplified by RT-PCR using specific primers (Supplementary Table 1) from HPgV-2 isolate AK-790 (GenBank accession number KT439329) [4]. After purification, PCR fragments were sequenced using the ABI PRISM 3730XL DNA Analyzer (Applied Biosystems).

Multiple sequences alignment was performed by using Clustal W of Mega version 6.0. The phylogenetic trees for NS3 gene and complete sequences of Pegivirus were constructed using the neighbor-joining tree method with 1000 bootstrap replicates to assess the robustness of individual nodes of each tree.

Nucleotide Sequence Accession Number

The complete genome of HPgV-2 from hepatitis C patient HCV121 from May 2015 [HCV121(1)] and March 2017 [HCV121(2)] has been submitted to GenBank under accession numbers KX528230 and KY971607, respectively. The accession numbers for another 2 HPgV-2 Chinese strains (C346China and IDU-31China) are KY971606 and KX528231, respectively.

Statistical Analysis

Data were analyzed using SPSS 20.0 software (IBM Corporation, Armonk, New York) for calculation of odds ratios, 95% confidence intervals, 2-sample t test, Fisher exact tests, Cochran and Mantel-Haenszel statistics, and binary logistic regression analysis, as appropriate. A P value <.05 was considered statistically significant.

RESULTS

Detection of Human Pegivirus Type 2 (HPgV-2) RNA and Anti-HPgV-2 Antibody

HPgV-2 RNA was detected by RT-PCR with a lower detection limit of 1–10 copies per reaction or 50–500 copies/mL for amplification of the HPgV-2 5ʹUTR and NS3 genes, respectively (data not shown). The sensitivity of the anti-HPgV-2 immunoassay for detecting HPgV-2 RNA–positive samples was 92.9% (13/14) in our study (Table 1) and was similar to that obtained by using recombinant HPgV-2 E2 protein in the Abbott Architect System (92.9% [13/14]) or slot blot format (91.7% [11/12]) [12].

Among the 8198 serum or plasma samples tested, 14 and 56 samples were HPgV-2 RNA and anti-HPgV-2 antibody positive, respectively, whereas 23.2% (13/56) of anti-HPgV-2 positive samples were HPgV-2 RNA positive as 1 sample was HPgV-2 RNA positive and anti-HPgV-2 negative (Table 1). The average viral load of HPgV-2 was 4.90 ± 1.35 log RNA copies/mL, and is similar to those (4.96 ± 1.30) reported by Coller et al [12]. The proportion of HPgV-2 RNA positive samples over anti-HPgV-2–positive samples observed in our study was similar to the 22.2% (8/36) value reported by Berg et al [5], and lower than the 38.7% (12/31) value in the anti-HPgV-2 E2 assay reported by Coller et al [12].

Prevalence of Human Pegivirus Type 2 Infection in Different Populations

Screening of 4017 healthy blood donors who were negative for HIV-1, HCV, and HBV yielded 6 anti-HPgV-2–weakly positive, but HPgV-2 RNA–negative samples, indicating a seroprevalence of 0.15% (Table 1). The low frequency of anti-HPgV-2 detection was also observed in HBV-monoinfected blood donors (0.2%) whereas no HPgV-2 infection was detected in 539 HIV-1–monoinfected samples. A relatively high frequency of HPgV-2 infection was observed in HCV-positive subjects with a serological prevalence of 1.23% and a frequency of HPgV-2 RNA of 0.29% in our study. HPgV-2 RNA prevalence is relatively low, though 1.35% and 1.7% of HCV-infected patients were positive in the United States [5] and in the United Kingdom [6], respectively, and the current study finds that <0.3% are positive. These results support the conclusion that HPgV-2 circulates less frequently in humans but is relatively frequent in the HCV-infected population [4–6, 12].

Interestingly, a high frequency of HPgV-2 detection was found in subjects coinfected with HCV and HIV-1 and 8.91% and 3.47% were positive for anti-HPgV-2 and HPgV-2 RNA, respectively (Tables 1 and 2). HPgV-2 RNA was also detected in 10.9% of PWIDs who were dually infected by HCV and HIV-1 in the United States [7], indicating that HCV/HIV-1–coinfected individuals represent a high-risk population for HPgV-2 infection.

Association Between Human Pegivirus Type 2 and Hepatitis C Virus (HCV), as Well as With HCV/Human Immunodeficiency Virus Type 1 Coinfection

Compared with healthy blood donors, we found that HCV-infected subjects showed an 8.33-fold higher risk of infecting HPgV-2, based on serological analysis, and an 11.49-fold higher risk of active or chronic HPgV-2 infection (Table 2). HCV/HIV-1 coinfection dramatically increased the risk of HPgV-2 infection by 71.43- and 142.86-fold based on serological and viral RNA detection, respectively (Table 2). The difference in prevalence of HPgV-2 infection and risk association was statistically significant between healthy blood donors and HCV-monoinfected or HCV/HIV-1–coinfected subjects (Table 2; P < .001), but not between HBV- or HIV-1–monoinfected individuals (data not shown).

Of note, we observed an anti-HPgV-2 prevalence of 1.88% and 0.48% in HCV-infected male and female subjects, respectively (Supplementary Table 2). The sex difference of HPgV-2 infection was 5.18-fold and statistically significant (P = .003). However, when the data were stratified according to the sources of sampling, the sex difference did not reach statistical significance (Supplementary Table 2). Importantly, these results were supported by multivariate regression analysis and indicated that only the status of HCV/HIV-1 coinfection was significantly associated with HPgV-2 infection (Supplementary Table 3; P = .021).

Persistence of Human Pegivirus Type 2 Infection

Stored plasma samples from 3 of the 14 HPgV-2 RNA–positive individuals identified in our study were tested at additional time points spanning from 68 to 1843 days. HPgV-2 RNA was persistently detected in 2 participants up to about 2 years apart (Supplementary Table 5), and no significant difference in HPgV-2 viral load was observed between the different bleeds. For patient A293, both anti-HPgV-2 and HPgV-2 RNA switched from positive to negative during a 5-year follow-up. Another 5 anti-HPgV-2–positive individuals were tested at additional time points with a median duration of 425 days. Anti-HPgV-2 was detected in all of them, but HPgV-2 RNA remained negative during the follow-up period (Supplementary Table 5). The NFLG sequences of HPgV-2 obtained from plasma samples of patient HCV121 about 2 years apart showed 97% identity at the nucleotide level and 99% amino acid sequence. The HPgV-2 strains identified from HCV121 at 2 time points were closely clustered and fell into a separate branch (Figure 1), indicating persistent infection and limited mutation of HPgV-2 [4–6].

In addition, we found that 27.8% (13/48) of anti-HPgV-2–positive samples were HPgV-2 RNA positive (Table 1), indicating that <30% of them were active and chronic cases of HPgV-2 infection. Our results are consistent with those reported previously [5, 12], and showed that about 30% of the HPgV-2 infections were persistent or, alternatively, that 70% of HPgV-2–infected subjects may have cleared the infection.

Impact of Human Pegivirus Type 2 Infection on Liver Damage and Hepatitis C Virus Replication

Because HCV can cause hepatitis, we examined possible liver damage related to HPgV-2 during coinfection of HCV and HPgV-2. We measured ALT levels and found that they were significantly higher in HCV-infected blood donors (86.1 ± 107.1 IU/mL) than in healthy blood donors (11.55 ± 5.36 IU/mL), indicating liver inflammation and signs of hepatocellular damage in HCV-infected blood donors (P < .05; Table 3). In addition, coinfection with HPgV-2 did not increase ALT levels, but instead caused a slight decrease. However, the difference was not statistically significant (P = .298; Table 3), and the standard deviations between ALT values exhibited considerable overlap. Furthermore, ALT levels were not significantly different between HPgV-2 RNA–positive and –negative subjects (50.4 ± 39.29 vs 58.5 ± 31.52 IU/mL, P = .766) or subjects with anti-HPgV-2 optical density (OD) values ≥1.0 or <1.0 (50.17 ± 43.18 vs 59.36 ± 27.56 IU/mL, P = .623) (Table 3). These results indicate that both HPgV-2 viremia and antibody titers may not affect ALT levels in HCV-infected blood donors.

To further probe the possible impact of coinfection of HPgV-2 on liver damage and HCV replication, we carried out a case-control study in which each anti-HPgV-2–positive blood donor (hereafter, “cases”) was matched with 4 anti-HPgV-2–negative blood donors (hereafter, “controls”) by age, sex, and HCV infection status. All cases and controls were male, and HCV treatment-naive. We observed slightly different HCV viral load (46.76 ± 75.99 vs 34.35 ± 51.35 × 10⁵ IU/mL) between the cases and controls (P = .849; Table 4). The ALT level was slightly lower in the case group (56.12 ± 35.24 IU/mL) than in the control group (87.75 ± 92.24 IU/mL), and the difference was statistically significant (P = .028). We repeated the case-control study by examining HPgV-2 RNA positivity and obtained similar results (Supplementary Table 4), but the difference was not statistically significant.

Because ALT reflects inflammation that may be transient, we further measured APRI ratios or FIB-4 values to represent fibrosis or cumulative liver damage in hepatitis C patients. We did not find significant differences in APRI or FIB-4 values between anti-HPgV-2–positive and –negative patients (Table 5). Kandathil et al also reported no difference of median liver stiffness between HPgV-2–positive and –negative PWID [7]. Taken together, our preliminary results indicate that coinfection with HPgV-2, regardless of the status of HPgV-2 infection (active/chronic vs resolved infection) or anti-HPgV-2 titers, may not worsen liver damage and may not affect HCV replication.

DISCUSSION

In this study, we set out to investigate the prevalence of HPgV-2 in China, its association with HCV and HIV-1, and the impact on HCV viral load and liver damage. We demonstrate very low frequency of HPgV-2 seropositivity and no HPgV-2 viremia in healthy blood donors in China. Our results are consistent with those reported in the United States and United Kingdom [5, 6, 12], indicating the rarity of HPgV-2 infection in the general population.

In the current and earlier work, HPgV-2 was detected more frequently in HCV-infected subjects, with a serological prevalence of 1.2% in China and 2.6%–3.3% in the United States [5, 12]. Furthermore, about 0.3% of both anti-HCV and HCV RNA positive subjects were HPgV-2 RNA positive in China. Values of 1.4% and 1.5% were reported in the United States [5, 12] and 1.7% in the United Kingdom [6]. Kandathil et al and the current study found that 10.9% [7] and 3.5% (Table 1), respectively, of individuals dually infected with HCV and HIV-1 are HPgV-2 RNA positive. The prevalence difference of HPgV-2 infection among various populations and countries remains to be verified due to the limited number of samples and regions tested and lack of well-evaluated test assays. It would be worthwhile to determine if there are high-risk populations or subpopulations as well as regions in which HPgV-2 infection is epidemic.

It was originally proposed that HPgV-2 is a blood-borne virus [4, 5], as individuals with hemophilia and PWID with multiple blood exposures and infected with HCV and/or HIV-1 are more frequently infected by HPgV-2 [6, 7]. However, HPgV-2 infection, in particular HPgV-2 viremia, was not detected in individuals singly infected with HBV or HIV-1 [5, 12], although HBV, HCV, and HIV-1 are all transmitted parenterally. These results suggest that blood exposure alone may not be sufficient to explain the tight association between HPgV-2 and HCV infection as well as HCV/HIV-1 coinfection. It is interesting to know if there are any other determinants that associate with high frequency of HPgV-2 infection in HCV-infected populations. The tight association of hepatitis D virus (HDV) and HBV has been well documented and is due to the dependence of HDV replication on the envelope of HBV [23]. It would be interesting to know if the common features shared by HPgV-2 and HCV could explain the association of these 2 viruses.

Although it is rare to detect HPgV-2 infection in subjects singly infected with HIV-1, the unique finding in our study (Tables 1 and 2) and in Kandathil et al’s study [7] showed that coinfection of HCV and HIV-1 can dramatically increase the frequency and risk of HPgV-2 infection. It is well established that HIV-1/HCV-coinfected individuals have higher HCV viral loads [24] and a higher persistence rate of HCV infection, probably due to CD4⁺ cell depletion caused by HIV-1 infection [25]. It has also been reported that HPgV-2 infection appears more likely to be associated with HCV viremia [4, 5, 12, 26]. In our study, coinfection of HIV-1 and HCV were clearly shown to increase the persistence rate of HPgV-2 infection by about 10% (Table 1). These results suggest the possibility that HIV-1 increases replication and viral load of HCV, which in turn supports the replication and persistent infection of HPgV-2 and results in high prevalence of HPgV-2 infection among individuals coinfected with HCV and HIV-1.

HPgV-2 can establish chronic and persistent infection [4, 5]. However, the percentage of chronic HPgV-2 infection is much lower than for HCV. Approximately 55%–85% of HCV-infected individuals develop persistent long-term infections [27]. In our study, about 72.2% (35/48) of anti-HPgV-2–positive subjects were HPgV-2 RNA negative. Berg et al reported that 57.9% (11/19) of HPgV-2 seropositive subjects were HPgV-2 RNA negative [5]. Coller et al found that 61.3% (19/31) of anti-HPgV-2 E2 positive subjects were HPgV-2 RNA negative [12]. These results indicate that resolved infections occur in about 60%–70% of HPgV-2–infected individuals, similar to HPgV infection, in which about 25% of infections persist and the other 75% clear viremia within 2 years of infection [28, 29].

As HPgV-2 is a novel human pegivirus, it is critical to determine its pathogenesis and its association with human diseases. Our initial results show that the influence of HPgV-2 on HCV viral load and liver damage including fibrosis in HCV-infected subjects may be limited. Recently, Kandathil et al reported similar results—that is, no evidence that HPgV-2 contributed to liver disease [7]. These findings provide us with clues about the medical consequences of HPgV-2 infection, raising the possibility that infection with HPgV-2 may result in limited or absence of pathogenicity.

In conclusion, our current study provides evidence of infrequent HPgV-2 infection in the general Chinese population, but a tight association between HPgV-2 and HCV infection, in particular HCV and HIV-1 coinfection. Further studies are necessary to clarify the interaction between HPgV-2 and HCV and/or HIV-1, and to investigate possible pathogenesis and disease association of HPgV-2 infection.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Financial support. This work was supported by the Bureau of Science and Information Technology of Guangzhou Municipality (grant numbers 201604020011 and 2014Y2-00550); and Guangdong Medical Scientific Research Foundation (grant number A2015012).

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

Author notes