Jpn. J･ Infect. Dis., 54, 55-63, 2001 Review GB VirusIC/Hepatitis G Virus KenjiAbe* Department of Pathology, National Institute of Infectious Diseases, Toyama 1-2311, Shinjuku-ku, Tobo 16218640, Japan (Received January 25, 2001. Accepted April 23, 2001) CONTENTS: Summary lntroduction History Viral genome Mutation rate of the viral genome Diagnostic assay Epidemiology and clinical significance HepatotroplSm and extrahepatic replication of the virus Relation to hepatocellular carcinoma occu汀enCe Experimental infection in chimpanzees C0-infection in human immunodeficiency virus (HIV)-infected patients Existence of viral genotypes and its geographic distribution Conclusion SUMMARY: GB vims-C (GBV-C)仇epatitis G vims (HGV) is a positive, single-strand RNA virus that has been classified in the family Flaviviridae. Interestlngly, GBVIC/HGV appears to have a truncated or absent core protein at the amino teminus of the polyprotein. GBV-C/HGV is transmitted parenterally and probably sexually. Most GBV-C作iGV infections appear to be asymptomatiC, persistent, and no correlation between virus infection and liver dysfunction although the disease-inducing activityof GBV-C什IGV remains to be investigated. Furthermore, there was no evidence ofpathogenesis in the liver by experiment with chimpanzees. From these results, GBV-C/ HGV might be considered as a kind of …orphan" vims in search of a disease. EpidemiologlCal investlgation demonstrated that GBVIC/HGV infection is present in about 1-1.4% of the healthy population in developed countries and in 8- 14.6% in developlng COuntries. The genome of GBV-C/HGV exhibits a sequence variation among different isolates. On the basis of this variation, it has been proposed that GBV-C/HGV can be classified into at least four major genotypes, consisting of type 1 (West Africa), type 2 (US/Europe), type 3 (Asia), andtype 4 (Southeast Asia). ent isolates of the same vims. This article will provide a Introd uction review of the available infbmation concemlng GBVJC/HGV After discovery of the hepatitis A and B viruses (HAV, HBV) in the 1960s and 1970S, it became apparent that an History unidentified agent was responsible for hepatltlS ln a number of patients suspected or having acute or chronic viral hepatitis. GBV-C is the acronym for a virus onglnally derived from The unknown pathogen was referred to as non-A, non-B a surgeon (whose initials were G･B･) in whom hepatitis hepatitis until the hepatitis C virus (HCV) was identified in developed in the 1960S (4)･ His plasma caused hepatitis. in 1989 (1 ) and was found to a?count for hepatitis in the majprity tamarins, and the 1 lth passage in tamarin serum was the clonlng of these patients. However, ln approximately lO% ofpatlentS source of the virus. Three di飽rent GBV agents have been cloned with liver diseases, no cause can be identified. lt has long thus far; GBV-A and GBVIB appear to be tamarin viruses, been suspected that additional hepatotroplC Viruses will be and GBV-C is a human vims (2,5). GBV-C was discovered discovered to account fわr some of these cases, especially slnCe by gene amplification with prlmerS derived from shared approximately half of these patients have a history of blood sequences in GBVIA, GBVIB, and HCV (2). On the other hand, transfusion. The cloning Ofthe HCV established that it is the initial source from which HGV was cloned was a patient among the most important human pathogens. ln 1 996, uslng with community-acquired chronic hepatitis (3 ). The sequences similar methods, two new RNA viruses were independently of GBV-C and HGV are more than 95% homologous, and discovered in human serum and were named GB virus-C the two are considered to be closely related isolates of the (GBV-C) and hepatitis G virus (HGV), respectively (2,3). Same VlmS. Database screenlng indicated that these viruses had high sequence similarity, hence they are now regarded as independ- Viral genome *Corresponding author: Tel: +81-3-528511 I 1 I ext. 2624, Fax: +81 The genomic organization of GBV-C/HGV is similar to 3 -5285-1 189, E-mail: [email protected] that ofHCV (Fig. 1). The GBVIC什IGV genome is a positive- 55 YFV (10 kb) NS4 (a+h) HCV (9.6 kb) NS4 (a+b) NS5(a+b) 111 I I I 1 I I FutLCtion UTR? ErLVelope Protease Protease/ NS3 RNA-dependent VTR proteiTI Helicase Co factor RNA poJymerase Fig. 1. Comparison ofgcnomc structure among yellow fcvcr virus (YFV), HCV, and GBVIC/HGV UTR, untranslatcd region; C, core gcnc; EI and E2, Cnvclopc gcncs I and 2; NS2 through NS5b, non-structural gcncs 2 through 5b. strand RNA or about 9400 nucleotides (nt) and it contains a data also suggested that plasma samples from individuals large open reading frame that encodes a polyprotein precursor chronically infectedwith GBV-C/HGV contained antibody of about 2900 amiヮ0.acid (aa) residues･ Like HCV, GBV-C/ to the GBV-C/HGV core protein peptide. These data indicate HGV has charactenstlCS Of a navivirusllike genome. The two that GBV-C/HGV has a nucleocapsid and that at least part of viruses thus represent a new genus in the family Flaviviridae the putative core reglOn Ofthe virus is expressed in vivo･ ln includingflaviviruses and pestiviruses, Since the genetic distance the literature, infectious particles lacking such a core have between GBV-C/HGV and HCV is too great to consider GBV- been generated artificially with the vesicular stomatitis virus C/HGV as a different genotype ofHCV (2). Like HCV, the glycop.rotein (8)･ The 5'UTR coヮtains an intemal ribosome genome or GBVC/HGV is preceded by a 5'-untranslated entry site that is capable of directlng CAP-independent trans- region (UTR), which is followed by a long open reading lation of the polyprotein (9, 1 0)･ Multiple sequenc.e alignments frame. The genome terminates with a 3'UTR, but it has no demonstrated that the 5'UTR and 3'UTR contalnS blocks of poly(A)-tail･ ln contrast to HCV, the sequeヮce variability of highly conserved (>95%) sequences･ This cons?rvation has envelope 2 (E2) is very low among different lSOlates collected enabled the optlmization of RT-PCR assays uslng PrlmerS worldwide and no hypervariable reglOn exists in the envelope from these conserved regions. reglOn･ The presumed polyprotein containsstructural envelope 1 (E 1 )and E2 glycoproteins at the amin0-terminal end,followed Mutation rate or the viral genome by nonstmctural (NS) proteins (NS2, NS3, NS4, NS5) at the carboxy-terminal end. The E2 glycoprotein has three potential Nakao et al. (1 1 ) reported that GBV-C/HGV was estimated glycosylation sites and 1 8 cysteine residues that might be to have a mutation rate of 3.9 × loヰ base substitutions per involved in disulfide bonds. Interestlngly, the GBV-C/HGV site per year･ We also dete-ined the mutation rate of GBV- genome lacks a clearly identifiable core gene which encodes C/HGV For this purpose, we isolated and compared two a nucleocapsid protein. ln our study, from the size of the GBV-C/HGV genomes (HGV-IM68 isolate and HGV-IM7 1 putative core reglOn at the aa level in 54 GBV-C什lGV isolates, isolate) from the same Japanese patient. The IM68 isolate they can be divided into four different groups (unpublished had been obtained from the patient 12 years before the study data). ln other words, 2 isolates with 107 aa (group 1) were took place. The IM68 sequence differed from lM71 sequence found, 4 isolates with 84 aa (group 2), 9 isolates with 47 aa at 59 (1%) of the 5867 nt･ The differe.nces between those (group 3), and 39 isolates with 16 aa (group 4). 1n addition, isolates were exclusively nucleotide polnt mutations. Based one isolate lacked the core reglOn COmpletely. 1n contrast, the on these results, the mutation rate ofHGV-lM7 I was estimated core reglOn OfHCV consists of 19l aa. The lack ofa core-like to be 0.8 × 10~3 base substitutions per site per year during a protein at the N-teminus of the viral polyprotein distinguishes 12-year period. This result suggests that the mutation rate of GBVIV/HGV from all other members of the Flaviviridae. the GBV-C什lGV genome is almost the same as that ofHCV Thus, GBVIC/HGV appears to be distinct from enveloped (12-14). However, unlike HCV; GBV-C/HGV does not have vimses in general because they tend to encode a basic protein a hypeⅣariable reglOn in the entire gene. The highest rate of that mediates the packaging of the viral nucleic acid into the mis-sense mutation was found in the E2 region, Which was virion envelope. However, Schmolke et al. (6) reported that followed by the rate of mutations in El and NS4 reglOnS. No GBV-C/HGV builds classic viral particles displaylng E2 mi§-sense mutation was seen in the core reglOn, nor in the envelope protein on their outer surfaces by uslng a Panel of NS5b reglOn. Moreover, no nucleotide mutation was obseⅣed eight monoclonal antibodies agalnSt the putative E2 protein in the 5 ′-UTR. Potential Asn-linked glycosylation (N-X-T/S) following DNA immunization. In addition, Xiang et al. (7) motifs located within the El and E2 reg10nS Were found reported that GBV-C/HGV particles included an extremely- to be conseⅣed entirely between the HGV-lM71 and -lM68 low density virion particle (1.07 to 1.09 g/ml) and a nucleo- isolates. capsid of～日8 g/ml, as shown on a sucrose gradient. Their 56 GBV-C/HGV RNA in serum reflects active viral infection, Diagnostic assay but the presence of anti-E2 antibody has been reported to RT-PCR is the only diagnostic tool to detect GBV-C作iGV correlate with recovery from viral infection, because GBVI RNA.