GB Virus C RNA in Serum, Liver, and Peripheral Blood Mononuclear Cells From Patients With Chronic Hepatitis B, C, and D

Transcription

1 GASTROENTEROLOGY 1997;113: GB Virus C RNA in Serum, Liver, and Peripheral Blood Mononuclear Cells From Patients With Chronic Hepatitis B, C, and D ANTONIO MADEJÓ N, MARTA FOGEDA, JAVIER BARTOLOMÉ, MARGARITA PARDO, CECILIA GONZÁLEZ, TERESA COTONAT, and VICENTE CARREÑO Department of Hepatology, Fundación Jiménez DıBaz, and Fundación para el Estudio de las Hepatitis Virales, Madrid, Spain S ince the identification of the hepatitis C virus (HCV) and hepatitis E virus (HEV) 1,2 and the development of reliable assays for their identification, most non-a, non-b hepatitis cases have been diagnosed. 3,4 However, a significant percentage of hepatitis, ranging from 10% to 20%, is still of unknown origin. 5 Simons et al. 6 re- cently isolated a novel flavivirus-like RNA genome, termed GB virus C (GBV-C), from sera of West African patients. An independent group has simultaneously isolated a flavivirus closely related to GBV-C, called the hepatitis G virus (HGV), from the plasma of a patient with chronic hepatitis. 7 GBV-C RNA sequences have been detected in sera of patients with fulminant hepatitis without hepatitis A to E virus markers 8 and in HCV chronic carriers. 9 All of Background & Aims: No conclusive data about GB virus C (GBV-C) tropism are available. We have studied the presence of genomic and antigenomic GBV-C RNA in serum, liver, and peripheral blood cells of 56 patients with chronic hepatitis B, C, or D virus infection. Methods: Genomic and antigenomic GBV-C RNA were detected by reverse-transcription nested polymerase chain reaction. Specificity was confirmed by sequenc- ing, by chemical modification of the RNA, and by using tagged primers. Results: Genomic GBV-C RNA was found in 10 of 56 (18%) of the sera. In contrast, antigenomic strand was not detected. The sequence of the amplified GBV-C RNA from 3 patients showed a 96% homology among them and from 83% to 88% with pre- viously described isolates. Genomic GBV-C RNA was found in 7 of 7 liver samples of the patients with serum GBV-C RNA. In 6 of these 7 patients (85%), antigenomic strand was found. Genomic RNA was found in 7 of 7 of the peripheral blood cell samples of the same 7 patients. Antigenomic GBV-C RNA was not found in these cells. Conclusions: These results suggest that GBV-C is a hepatotropic virus that replicates in the human liver. The data do not support a role for GBV-C in chronic liver disease. these findings suggest that the GBV-C agent may cause acute and chronic hepatitis in humans. However, the site of infection and replication of the GBV-C, whether in liver, peripheral blood mononuclear cells (PBMCs), or other tissues, has not been convincingly shown. We have studied the presence of GBV-C sequences in the serum, liver, and PBMCs of patients with chronic hepatitis caused by hepatitis B virus (HBV), HCV, and hepatitis D virus (HDV). Patients and Methods The presence of GBV-C RNA was retrospectively ana- lyzed in serum samples (stored at 020 C) from 56 patients with chronic viral hepatitis histologically proven according to international criteria (Table 1). 10 None of the patients had undergone antiviral or immunosuppressive therapy. Finally, the availability of liver and/or PBMC samples taken at the time of the analysis of GBV-C RNA was the criteria for patient selection. In addition, sera of 29 patients with chronic viral hepatitis already known to be repeatedly negative for GBV-C RNA in serum were used as controls (Table 1). Clinical features of GBV-C positive and negative (including the 29 controls) patients are summarized in Table 1. PBMCs were isolated from heparinized blood by centrifugation in a Ficoll Hypaque gradient (Biochrom KG, Berlin, Germany) and immediately stored in liquid nitrogen. Percutaneous liver biopsy specimens were obtained using Tru-Cut needles (Baxter Healthcare Corp., Deerfield, IL). A part of the liver sample was immediately frozen in liquid nitrogen, and the remaining tissue was processed for the histologic analysis. Every reverse-transcription polymerase chain reaction (RT- PCR) run included as negative controls a no-rna reagent control and serum from 5 healthy subjects with normal alanine aminotransferase levels and negative for markers of HBV, HCV, and HCD. Abbreviations used in this paper: AMV-RT, avian myeloblastosis virus reverse transcriptase; GBV-C, GB virus C; HGV, hepatitis G virus; PBMC, peripheral blood mononuclear cell; RT-PCR, reverse- transcription polymerase chain reaction; SSC, standard saline citrate by the American Gastroenterological Association /97/$3.00

2 574 MADEJÓN ET AL. GASTROENTEROLOGY Vol. 113, No. 2 Table 1. Clinical Features of the Patients With Chronic Hepatitis Included in the Study Patients Controls GBV-C positive GBV-C negative (n Å 56) (n Å 29) (n Å 10) (n Å 75) Sex (M/F ) 50/6 26/3 9/1 67/8 Age (yr) 39.2 { { { { 13.5 Alanine aminotransferase (IU/L) { { { { Epidemiology IVDA 27 (48) 15 (51) 6 (60) 36 (48) Postransfusional 27 (48) 12 (41) 4 (40) 35 (47) Sexual 1 (2) 1 (3) 2 (3) Unknown 1 (2) 1 (3) 2 (3) Serology HBV 10 (18) 5 (17) 2 (20) 13 (17) HBV-DNA / 5/10 (50) 3/5 (60) 0/2 (0) 8/13 (62) HCV 37 (66) 19 (65) 6 (60) 50 (67) HCV-RNA / 26/37 (70) 13/19 (68) 5/6 (83) 34/50 (68) HDV 9 (16) 5 (17) 2 (20) 12 (16) HDV-RNA / 9/9 (100) 5/5 (100) 2/2 (100) 12/12 (100) Histology Mild chronic hepatitis Mild fibrosi 3 (5) 2 (7) 1 (10) 4 (5) Moderate fibrosi 34 (61) 17 (58) 6 (60) 45 (60) Moderate chronic hepatitis Mild fibrosi 2 (4) 2 (7) 1 (10) 3 (4) Moderate fibrosi 8 (14) 4 (14) 1 (10) 11 (14) Severe fibrosi 6 (11) 3 (10) 1 (10) 8 (10) Severe chronic hepatitis Cirrhosis 3 (5) 1 (3) 4 (5) NOTE. Results are expressed as mean { SD values and number (percent). No statistical differences were observed between GBV-C positive or negative patients in any of the parameters studied, as assessed by Student s t test and Fisher s Exact Test. IVDA, intravenous drug abuser. RNA Isolation Total RNA from serum, liver, and PBMC samples was isolated as previously described. 11 Liver and PBMCs were washed three times with ice-cold 11 phosphate-buffered saline (PBS) (101 PBS is 1.37 mol/l NaCl, 27 mmol/l KCl, 4.3 mmol/l Na 2 PO 4, and 1.4 mmol/l KH 2 PO 4 ) before RNA ex- traction. The last wash mixture was saved to be used as a control for the detection of GBV-C RNA in the liver and PBMCs. To test the integrity of the RNA extracted from the liver and PBMCs, b-actin messenger RNA (mrna) was amplified, using the sense (5 AGCGGGAAATCGTGCGTG3 ) and anti- sense (5 CAGGGTACATGGTGGTGCC3 ) primers, 12 which amplify a 311-nucleotide fragment. GBV-C RNA Amplification The total RNA obtained from sera, the last cell wash, and 1 mg of tissue RNA were amplified by RT-PCR using specific primers of the putative GBV-C helicase/ns3 gene and of the 5 noncoding region, designed according to the sequence reported by Leary et al. 13 (Table 2). GBV-C Genomic RNA Strand Amplification For the specific GBV-C genomic RNA strand amplification of the 5 noncoding or the NS3 regions, outer antisense primer (50A or NOA, respectively) was used to prime the complementary DNA (cdna) synthesis. After an RNA denaturation step (5 minutes at 95 C), avian myeloblastosis virus reverse transcriptase (AMV; Promega Corp., Madison, WI) was added, and reverse transcription was conducted at 42 C for 30 minutes. Finally, AMV-RT was inactivated by heating at 95 C for 1 hour. After the AMV-RT inactivation, the outer sense primer (5OS for 5 noncoding region or NOS for NS3 region amplification) was added, and PCR was performed at 94 C for 1 minute, 50 C for 1 minute, and 72 C for 1 minute for 35 cycles, with a final extension step at 72 C for 7 minutes. The second PCR round was performed under the same conditions, using 10% of the first PCR product, adding simultaneously the inner primers (5IA and 5IS to amplify the 5 -noncoding region or NIA and NIS to amplify the NS3 region). GBV-C Antigenomic RNA Amplification For the specific antigenomic strand amplification of the 5 -noncoding or the NS3 region, outer sense primer (5OS or NOS, respectively) were used to prime the cdna synthesis as described above. After inactivation of the AMV-RT, the outer antisense primer (5OA or NOA, respectively) was added and the first PCR round was performed under the same conditions used for GBV-C genomic RNA detection. Finally, the

3 August 1997 GB VIRUS C REPLICATION IN THE HUMAN LIVER 575 Table 2. Primers and Probes Used in GBV-C RNA Amplificatio Nucleotide Size Name Nucleotide sequence position (base pairs) 5 Noncoding region Outer Sense 5OS CGGCACTGGGTGCAAGCCCCA Antisense 5OA CCCGGCCCCCACTGGTCCTTG Inner Sense 5IS CGACGCCTACTGAAGTAGACG Antisense 5IA GTACGCCTATTGGTCAAGAGA Tagged Sense T1 ATGCACATTCGCCTGCAAGACGACGCCTACTGAAGTAGACG Antisense T2 ATGCACATTCGCCTGCAAGAGTACGCCTATTGGTCAAGAGA T3 ATGCACATTCGCCTGCAAGA Probe P1 TAAATCCCGGTCATCCTGGTA NS3/helicase region Outer Sense NOS GCTCGCCTATGACTCAGCATC Antisense NOA GTCACCTCAACGACCTCCTCC Inner Sense NIS GAGACAAAGCTGGACGTTGGT Antisense NIA CAACCCACAGTCGGTGACAGA Probe P2 GGCCAGTTCTCCGCGCGGGGG NOTE. Primers T1 and T2 are designed by addition of primer T3 at 5 position of primers 5IS and 5IA, respectively. Expected size of PCR products using tagged primers: 5OS/T2, 366 base pairs; 5OA/T1, 351 base pairs; 5IS/T3, 340 base pairs; 5IA/T3, 340 base pairs. nested PCR was performed by adding the inner primers (5IS Kit; Invitrogen Corp., San Diego, CA), and three clones of and 5IA, or NIS and NIA, respectively) using the conditions each patient were sequenced by the dideoxynucleotide chain described for the genomic RNA detection. termination method (Sequenase Version 2.0; USB, Cleveland, The outer sense and outer antisense primers were added OH). separately to each sample of serum, PBMCs, and liver. The specificity of the NS3 and 5 noncoding region PCR Specificity of the Detection of Antigenomic products was also assessed by dot blot hybridization. Ten microliters of each PCR product was denaturated and spotted GBV-C RNA onto nitrocellulose membranes. Filters were prehybridized at The specificity of the antigenomic GBV-C RNA am- 70 C for 2 hours in a solution containing 21 Denhardt s plification was tested by the following control assays using solution (11 Denhardt s is 0.02% bovine serum albumin, specific primers of the 5 noncoding region: (1) GBV-C anti- 0.02% polyvinylpyrrolidone, plus 0.02% Ficoll), 51 standard genomic RNA amplification without the RT step; (2) GBV- saline citrate (SSC) (11 SSC is 150 mmol/l NaCl plus 15 C antigenomic RNA amplification without the specific primer mmol/l sodium citrate, ph 7), and 80 mg/ml of denatured, during RT; (3) addition of total RNA only after heat inactivaat sonicated salmon sperm DNA. Hybridization was conducted tion for 1 hour at 95 C of the reverse transcriptase and posterior 45 C for 18 hours with the same prehybridization solution cdna synthesis and nested PCR steps; (4) RT-nested PCR, containing 10 6 cpm/ml of the 32 P5 end labeled oligonucleo- using total RNA chemically modified at its 3 end as prestrand tides P1 (for the analysis of the genomic and antigenomic viously described 14 ; and (5) amplification of antigenomic 5 amplification of the 5 noncoding region) and P2 (for noncoding region GBV-C RNA using tagged primers in the the analysis of the genomic and antigenomic strand amplifica- RT and in the PCR 15 on unmodified and chemically modified tion of the NS3 region). After hybridization, filters were total RNA. Briefly, RT was performed in independent experi- washed five times (10 minutes each) at room temperature with ments, using the T1 (sense RNA strand) or T2 (antisense RNA 21 SSC, 0.1% sodium dodecyl sulfate, and five times (10 strand) primers (Table 2). The first PCR round was performed minutes each) at 70 C with 0.11 SSC and 0.1% sodium dode- by addition of the corresponding outer antisense (5OA) or cyl sulfate. After autoradiography, intensity of each spot was sense (5OS) primers. Nested PCR was performed using 10% assessed by densitometry (Personal densitometer; Molecular of the first PCR product, by adding the T3 primer and the Dynamics, Sunnyvale, CA). In all cases, genomic and antigeno- inner antisense (5IA) or sense (5IS) primers. The expected size mic GBV-C RNA amplification products from each patient of these PCR products was 340 base pairs. were analyzed in the same filter. Cloning and Sequencing Results The NS3 PCR product from three serum GBV-C After amplification of the genomic GBV-C RNA, positive patients was cloned in the PCR II vector (TA Cloning using specific primers from the NS3/helicase region, sera

4 576 MADEJÓN ET AL. GASTROENTEROLOGY Vol. 113, No. 2 Figure 1. Amino acid sequence of the amplifie GBV-C NS3 fragment, residues according to Leary et al., 13 obtained from the patients included in this study (lanes 1 3), African isolates 6 (lanes 4 6), North American isolates 6 (lanes 7 and 8), and Japanese isolates 8 (lane 9). of the 5 healthy subjects and of the 29 patients with chronic viral hepatitis previously negative to GBV-C RNA were always negative as well as for the no-rna reagent control. In contrast, the expected 272 base pair PCR product was detected in 10 of 56 (18%) of the serum samples from patients with chronic viral hepatitis. This positivity was confirmed on hybridization with the specific probe of the NS3 region. The antigenomic GBV- C strand was not found in any of the serum samples tested. The same results were obtained when the amplification was performed using primers from the 5 -noncoding region and after hybridization of the PCR products with the 5 noncoding region probe. Sequence analysis of the nine clones from 3 patients of the NS3/helicase region showed 96% homology among them at the nucleotide level. In contrast, comparison with the previously reported sequences showed a higher degree of variability, ranging from a 12% to 17% difference with respect to the sequences isolated from the U.S., West African, and Japanese patients. 6,8 However, because most of the nucleotide changes occurred at the third codon position, only 1 4 of the 66 (1.5% 6%) predicted amino acids changed with respect to the pre- viously reported isolates (Figure 1). 6,8 The presence of GBV-C RNA of genomic and anti- genomic polarity was tested in liver biopsy specimens from 7 patients with serum GBV-C RNA and from an additional 7 patients negative for the viral RNA in serum. Genomic GBV-C RNA was detected in 7 of 7 of the liver samples from patients with serum GBV-C RNA, whereas antigenomic GBV-C RNA was found in 6 of 7 (85%) of them. No GBV-C RNA of genomic or anti- genomic polarity was detected in any of the liver samples from the patients without serum GBV-C RNA. With respect to the specificity of the antigenomic GBV-C RNA detection, no amplification was obtained when the RT step was omitted or when total RNA was added to the RT reaction after reverse transcriptase inactivation. Furthermore, no antigenomic GBV-C RNA was detected when RT was performed without the corre- sponding primers. When RT was performed using chemically modified RNA, the antigenomic GBV-C RNA was detected in the same 6 liver samples in which antigenomic RNA had been detected using unmodified total RNA. RT-PCR was also performed using tagged primers with modified and unmodified total RNA. Using this methodology, the antigenomic GBV-C RNA positivity was confirmed in the six above-mentioned liver samples (Figure 2). No GBV-C RNA of genomic or antigenomic polarity was found in the last wash mixture of the liver biopsy specimens. The presence of GBV-C RNA was also tested in un- modified and chemically modified RNA from PBMCs of the 7 patients with viral RNA in serum and liver and from the 7 patients without GBV-C RNA in serum and liver. The genomic RNA strand was detected in the 7 patients with GBV-C RNA in serum and liver. In con- trast, GBV-C RNA of antigenomic polarity was not found in any of the 7 PBMC samples. No GBV-C RNA of either genomic or antigenomic polarity was detected in the 7 PBMC samples from the patients without viral RNA in serum and liver. As specificity controls, the final PBS wash of the PBMCs was also tested for the presence of GBV-C RNA, with negative results in all cases. Finally, b-actin mrna was positive by single PCR in all the liver biopsy and PBMC samples tested, showing that the lack of GBV-C RNA detection was not a result of RNA degradation (data not shown). The specificity of the amplification reactions was also proved by hybridization of all PCR products with the corresponding 5 noncoding or NS3 region oligonucleo- tide probes. Densitometric analysis of the autoradiographies showed that the genomic/antigenomic GBV-C RNA ratio in the liver cells ranged from 5/1 to 25/1 (data not shown). Discussion A novel hepatitis-associated agent, termed GBV- C or HGV, has been identified recently. 6,7 GBV-C RNA has been detected by PCR in sera of patients with fulmi-

5 August 1997 GB VIRUS C REPLICATION IN THE HUMAN LIVER 577 Figure 2. GBV-C RNA amplificatio using primers from the 5 noncoding region, from (A) serum, (B) PBMCs, and (C ) liver of one of the patients included in the study. (A) G, genomic strand; AG, antigenomic strand; 1, serum from a healthy subject; 2, no RNA negative control. (B) G, genomic strand; W1, PBS wash; AG, antigenomic strand; W2, PBS wash; 1, PBMCs from a patient without GBV-C RNA in serum; 2, no RNA negative control. (C ) G, genomic strand; W1, PBS wash; AG, antigenomic strand; W2, PBS wash; 1, amplificatio of AG without RT; 2, amplificatio of AG without primers during RT; 3, amplificatio of AG adding total RNA after heat inactivation of reverse transcriptase; 4, amplificatio of AG using chemically modifie total RNA; 5, amplificatio of AG using tagged primers; 6, no RNA negative control. M, 100 base pair DNA ladder (GIBCO BRL, Paisley, Scotland). nant hepatitis 8 and chronic hepatitis 9 as well as in voluntary blood donors without evidence of liver disease. 7 However, the possible infection or replication of GBV- C in liver, or in other tissues such as PBMCs, has not been convincingly shown. In this study, we tested for the presence of GBV-C RNA in the serum, liver, and PBMC samples of patients with chronic hepatitis B, C, and D. Genetic amplification using specific primers of the NS3 and the 5 noncoding regions showed serum genomic GBV-C RNA positivity in 10 of 56 patients (18%) included in this study. Sequence analysis of PCR products showed a high percentage of homology among clones of the same patient and between clones of different patients. However, comparison with previously published sequences 6,8 showed a lower homology, with differences at the nucleotide level ranging from 12% to 17%. However, because most of the changes observed occurred at the third codon position, the differences were only between 1.5% and 6% at the amino acid level with respect to the clones isolated in the United States, West Africa, and Japan. This high degree of amino acid conservation suggests an important role of the NS3 product in the virus biology. With respect to the detection of GBV-C in the liver biopsy specimens, viral RNA of genomic polarity was detected in 7 of 7 of the liver samples from patients with GBV-C RNA in serum. The absence of genomic strand in the last liver biopsy wash mixture supports the conclusion that the positive signals obtained were not caused by blood contamination of the liver samples. It could be argued that the detection of the genomic GBV-C RNA in liver is caused by viral particles bound to the cell membrane. However, because viral RNA of antigenomic polarity was also found in liver samples and this strand was not detected in serum, this explanation seems unlikely. On the other hand, the presence of antigenomic RNA strand is considered to be evidence of viral replication in flaviviruses. 16 In this study, GBV-C RNA of antigenomic polarity was detected in 6 of 7 of the liver samples, with genomic GBV-C RNA in the serum and liver. Recent studies 14,15 have suggested the existence of false-positive results in the detection of antigenomic HCV-RNA when a specific antigenomic primer is used for the RT-nested PCR. This is probably caused by nonspecific priming of the genomic strand by cellular RNA fragments produced during the extraction steps or by self-priming of the genomic RNA. Therefore, we have shown the specificity of the detection of the antigenomic GBV-C RNA strand in several experiments. RT-nested PCR was performed without addition of primers during the reverse tran- scriptase step, and the lack of detection of the antigeno- mic RNA showed that, in the experimental conditions assayed, no nonspecific priming by cellular RNA or self- priming occurred. To further assure the specificity of the results, detection of the antigenomic strand was performed on total liver RNA chemically modified at its 3 end to impede cellular RNA priming. This experimental approach confirmed antigenomic GBV-C RNA positivity in the same above-mentioned 6 liver samples, showing the specificity of the antigenomic GBV-C RNA detec- tion. Furthermore, the use of tagged primers during the RT and PCR for the amplification of unmodified or chemically modified RNA confirmed the specificity of the detection of the antigenomic RNA because the same

6 578 MADEJÓN ET AL. GASTROENTEROLOGY Vol. 113, No Yoshiba M, Okamoto H, Mishiro S. Detection of the GBV-C hepati- tis virus genome in serum from patients with fulminant hepatitis 6 samples were positive. All of these results, as well as the fact that antigenomic strand was never found in the last wash supernatant of the 7 liver samples or in serum, show that GBV-C replicates in the human liver. This conclusion is also supported by the semiquantitative analysis of the dot blot hybridization of genomic and antigenomic PCR products. This analysis showed that the amount of genomic GBV-C RNA in liver samples is 5 25 times higher than the antigenomic viral RNA. These results are similar to those found for HCV. 17 Finally, we also investigated the presence of GBV-C 1996; 271: in PBMCs of the same 7 patients in whom genomic and antigenomic GBV-C RNA were studied in the liver. of unknown aetiology. Lancet 1995; 346: GBV-C RNA of genomic polarity was found in the PBMCs of the 7 patients, whereas antigenomic GBV-C 196. RNA was not found in any of the PBMC samples tested. These data suggest that GBV-C can infect, but does Classificatio ing. Hepatology 1994; 19: not replicate, in human PBMCs. However, it cannot be discarded that the genomic GBV-C RNA detected in the PBMCs came from viral particles tightly attached to Biochem 1987; 162: the cell membrane, rather than from virus infecting the third generation tests and polymerase chain reaction. J Hepatol 1993; 19: Simons JN, Leary TP, Dawson GJ, Pilot-Matias TJ, Muerhoff AS, Schlauder GG, Desai SM, Mushahwar IK. Isolation of novel viruslike sequences associated with human hepatitis. Nature Med 1995; 1: Linnen J, Wages J Jr, Zhang-Keck Z-Y, Fry KE, Krawczynski KZ, Alter H, Koonin E, Gallagher M, Alter M, Hadziyannis S, Karayiannis P, Fung K, Nakatsuji Y, Shih JW-K, Young L, Piatak M Jr, Hoover C, Fernandez J, Chen S, Zou J-C, Morris T, Hyams KC, Ismay S, Lifson JD, Hess G, Foung SKH, Thomas H, Bradley D, Margolis H, Kim JP. Molecular cloning and disease association of hepatitis G virus: a transfusion-transmissible agent. Science 9. Aikawa T, Sugai Y, Okamoto H. Hepatitis G infection in drug abusers with chronic hepatitis C. N Engl J Med 1996; 334: Desmet VJ, Gerber M, Hoofnagle JM, Manns M, Scheuer PJ. of chronic hepatitis: diagnosis, grading and stag- 11. Chomzynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal 12. Reiling N, Ulmer AJ, Duchrow M, Ernst M, Flad H-D, Hanschidt S. Nitric oxide synthase: mrna expression of different isoforms PBMCs. in human monocytes/macrophages. Eur J Immunol 1994; 24: In conclusion, our results suggest that GBV-C is a hepatotropic virus that replicates in the human liver. 13. Leary TP, Muerhoff AS, Simons JN, Pilot-Matias TJ, Erker JC, Chalmers ML, Schlauder GG, Dawson GJ, Desai SM, Mushahwar However, the results obtained in this work, in agreement IK. Sequence and genomic organization of GBV-C: a novel memwith previous reports, 18,19 do not support a role for GBV- ber of the flavivirida associated with human non A-E hepatitis. C in chronic liver disease, because clinical features of J Med Virol 1996; 48: Gunjy T, Kato N, Hijikata M, Hayashi K, Saitoh S, Shimotohno GBV-C positive or negative patients were similar. K. Specifi detection of positive and negative stranded hepatitis Whether different GBV-C subtypes are associated with C viral RNA using chemical RNA modification Arch Virol 1994; different pathogenicity should be further studied. 134: Lanford RE, Sureau C, Jacob JR, White R, Fuerst TR. Demonstration of in vitro infection of chimpanzee hepatocytes with hepatitis References C virus using strand-specifi RT/PCR. Virology 1994; 202: Choo QL, Kuo G, Weiner AJ, Overby LR, Bradley DW, Houghton 614. M. Isolation of a cdna clone derived from a bloodborne non-a, 16. Westaway EG. Flavivirus replication strategy. Adv Virus Res 1987; non-b viral hepatitis genome. Science 1989; 244: : Gregory RR, Purdy MA, Kim JP, Luk K-C, Young LM, Fry KE, Bradley 17. Lanford RE, Chavez D, Chisari FV, Sureau C. Lack of detection DW. Isolation of a cdna from the virus responsible for enterically of negative-strand hepatitis C virus RNA in peripheral blood monotransmitted non-a, non-b hepatitis. Science 1990; 247:1335 nuclear cells and other extrahepatic tissues by the highly strand specifi rtth reverse transcriptase PCR. J Virol 1995; 69: Kuo G, Choo Q-L, Alter HJ, Gitnick GL, Redeker AG, Purcell RH, Miyamura T, Dienstag JL, Alter MJ, Stevens CE, Tegtmeier GE, 18. Alter MJ, Gallagher M, Morris TT, Moyer LA, Meeks EL, Krawczyn- Bonino F, Colombo M, Lee W-S, Kuo C, Berger K, Shuster JR, ski K, Kim JP, Margolis HS. Acute non A-E hepatitis in the United Overby LR, Bradley DW, Houghton M. An assay for circulating States and the role of hepatitis G virus infection. New Engl J Med antibodies to a major etiologic virus of human non-a, non-b hepa- 1997; 336: titis. Science 1989; 244: Alter HJ, Nakatsuji Y, Melporder J, Wages J, Wesley R, Shih 4. Dawson GJ, Chau KH, Cabal CM, Yarbough PO, Reyes GR, Mus- JW-K, Kim JP. The incidence of transfusion associated hepatitis hahwar IK. Solid-phase enzyme-linked immunosorbent assay for G virus infection and its relation to liver damage. New Engl J Med hepatitis E virus IgG and IgM antibodies utilizing recombinant 1997; 336: antigens and synthetic peptides. J Virol Methods 1992; 38: Received August 2, Accepted April 1, Marcellin P, Martinot-Peignoux M, Gabriel F, Branger M, Degott Address requests for reprints to: Vicente Carreño, M.D., Depart- C, Elias A, Xu LZ, Larzul D, Erlinger S, Benhamou JP. Chronic ment of Hepatology, Fundación Jiménez DıB az, Avda. Reyes Católicos, non-b, non-c hepatitis among blood donors assessed with HCV 2, Madrid, Spain. Fax: (34)