RAPID PAPER

Prevalence of GBV-C/HGV, a novel ‘hepatitis’ virus,in patients with aplastic anaemia

KEVIN E. BROWN, SUSAN WONG AND NEAL S. YOUNG Hematology Branch,National Heart, Lung, and Blood Institute, Bethesda, Maryland, U.S.A.

Received 25 October 1996; accepted for publication 21 February 1997

Summary. GBV-C or hepatitis G virus (GBV-C/HGV) is a novelRNA virus with similarities to members of the Flaviviridaefamily, especially hepatitis C. Viral RNA is detected in about1·5% of American blood donors, with higher prevalence inmultiply transfused patients and in individuals with hepatitisor liver disease. Some cases of aplastic anaemia followapparent non-A, non-B, non-C viral hepatitis, and GBV-Cviraemia has been described in three case reports ofhepatitis-associated aplastic anaemia. We tested clinicalsamples from patients with aplastic anaemia with or withoutrecent hepatitis for the presence of GBV-C/HGV. Virus was

detected in a total of 15/57 (26·3%) of patients withaplastic anaemia and 12/52 (23·1%) of multiply transfusedcontrol patients. Sequencing of the 188 base pair NS3helicase PCR product in the serum of five individualsindicated the same high degree of sequence variation ashas been seen among other isolates of the virus. GBV-C/HGVdoes not appear to be implicated in the aetiology of aplasticanaemia.

Keywords: Flaviviridae, non-A, non-B hepatitis, epidemiology,prevalence, aplastic anaemia.

Despite the identification of two new hepatitis viruses withinthe last 10 years, the aetiology of approximately 10–20% ofhepatitis cases remains unknown (Purcell, 1994; Alter &Bradley, 1995), including sero-negative (non-A–E) commu-nity-acquired hepatitis (Alter et al, 1992), most fulminanthepatitis of childhood (Sallie et al, 1994; Kuwada et al,1994), and the hepatitis/aplastic anaemia syndrome (Hibbset al, 1992a). Recently, two independent laboratories inpursuit of novel hepatitis viruses described the identificationof a new RNA virus, provisionally designated either GB virus-C (GBV-C) (Simons et al, 1995a) or hepatitis G virus (HGV)(Linnen et al, 1996). GBV-C was isolated from a West Africanserum sample that contained antibodies to two newlydescribed viruses GBV-A and GBV-B (Simons et al, 1995b).Hemi-nested RT-PCR of the serum with degenerate primersderived from the NS3 region of GBV-A, GBV-B and hepatitis Cvirus (HCV) amplified a product of the expected size, butsequencing showed that the product was from a novel virus,related to but distinct from GBV-A, GBV-B and HCV. Genomeextension was performed to clone the entire sequence (Learyet al, 1996a). The second group identified part of the HGVgenome by screening an expression library made from cDNA

from plasma of a patient with chronic hepatitis; the completeHGV sequence was then obtained by anchored polymerasechain reaction (PCR) (Linnen et al, 1996).

Both GBV-C and HGV are positive-stranded RNA viruses ofapproximately 10 kilobases, with similarities at the genomiclevel to members of the Flaviviridae family, especially tohepatitis C, as well as to GBV-A and GBV-B (Leary et al,1996a). Comparison of the structure and sequence of GBV-Cand HGV confirm that they are isolates of the same virus(Zuckerman, 1996). GBV-C/HGV can be transmitted byblood transfusion (Simons et al, 1995a), and viral RNA hasbeen detected in plasma pools and immunoglobulin pre-parations (Nubling & Lower, 1996). Viraemia was found in1·5% of volunteer blood donors in the United States (Linnenet al, 1996; Dawson et al, 1996) and 5·7% of normalindividuals in Vietnam (Brown et al, 1997b).

Although GBV-C/HGV has not been definitively associatedwith any disease (Alter, 1996), an increased prevalence ofviraemia was reported for persons with a history ofparenteral exposures or undergoing haemodialysis (Masukoet al, 1996; de Lamballerie et al, 1996), and in patients withliver disease (Linnen et al, 1996; Aikawa et al, 1996; Yoshibaet al, 1995; Fiordalisi et al, 1996). In addition, GBV-C/HGVhas been reported in three cases of hepatitis associated withaplastic anaemia (hepatitis/aplastic anaemia syndrome),suggesting an aetiologic role (Simons et al, 1995a; Zaidl

British Journal of Haematology, 1997, 97, 492–496

492 q 1997 Blackwell Science Ltd

Correspondence: Dr K. E. Brown, Bldg 10/Rm 7C218, NationalInstitutes of Health, 9000 Rockville Pike, Bethesda, MD 20892-1652, U.S.A.

et al, 1996; Byrnes et al, 1996). We therefore undertook astudy to assess the prevalence of this novel virus in patientswith aplastic anaemia that is both associated with hepatitisand idiopathic.

MATERIALS AND METHODS

Sera. The study was approved by the National Heart, Lung,and Blood Institutional Review Board. Serum samples wereobtained from three different population groups. Groups 1and 2 were patients with aplastic anaemia; group 1 had ahistory of clinical hepatitis within 3 months of the onset ofaplasia and group 2 had no such history of hepatitis. As acontrol, group 3 consisted of multiply transfused haematol-ogy patients also attending Hematology Branch clinics. In allgroups, to minimize the role of transfusion-acquired infec-tion, the earliest available sample after referral to theNational Institutes of Health was tested (1981–96). Serumwas separated from whole blood within 12 h of collectionand immediately stored at ¹208C until tested.

RT-PCR for GBV-C/HGV. RNA was extracted from serumsamples (140 ml) using the Qiamp HCV kit (Qiagen Inc.),according to the manufacturer‘s protocol. Reverse transcrip-tion was performed using random hexamers and murineleukaemia virus reverse transcriptase (708C for 10 min, 428Cfor 30 min, 958C for 5 min; GeneAmp RNA PCR kit, PerkinElmer). Amplification was performed using degenerate primersGBV-s1 and GBV-a1 in a ‘touchdown’ thermocycling protocolas previously described (Leary et al, 1996b). The products(5 ml) were analysed by Southern hybridization using aninternal 125 nucleotide 32P-nick translated probe generatedfrom a GBV-C positive serum.

Sequencing of isolates. Following amplification, PCR pro-ducts that gave a clear band on ethidium bromide stainingwere ligated into a cloning vector (TA cloning, Invitrogen)and sequenced either by dideoxynucleotide chain termina-tion using Sequenase II (Amersham, U.K.) or automatedcycle sequencing with Taq polymerase (Perkin Elmer). Thesequences were aligned using DNAStar (DNAStar Inc.), andsimilarities calculated using the clustal method withstandard weighted residue weight table.

Serology for hepatitis A, B and C. Sera were tested forhepatitis A antibody (IgM and IgG), hepatitis B surfaceantigen and antibody (HbsAg and HbsAb) and hepatitis Bcore antibody (HbcAb), and hepatitis C antibody by standardcommercial assays. In addition, hepatitis C was assayed byRT-PCR. RNA was extracted as above, and amplification anddetection of positive samples was performed using nestedPCR as previously described (Hibbs et al, 1992b).

RESULTS

Seroprevalence of GBV-C/HGV in patients with hepatitis-associated aplastic anaemiaSerum samples were tested from 10 patients with the clinicalsyndrome of hepatitis-associated aplastic anaemia (Brown etal, 1997a): three had detectable GBV-C/HGV viraemia(Table I). All three patients had received multiple bloodtransfusions (16 units in Saudi Arabia, 20 units in

Venezuela, and 35 units in Canada respectively). Serumaliquots were available from 12/16 units donated to the firstof these patients, and testing of the samples for HGV-C/HGVshowed that one of the units (from a family member) alsocontained GBV-C/HGV.

One patient, born in Asia, had evidence of previousinfection with hepatitis B (HbsAg negative, HbsAb andHbcAb positive). In the other nine patients, serological testsfor hepatitis A (IgM), B (HbsAg and HbcAb) and C (HCV Aband HCV RNA) were all negative, confirming that none hadevidence of recent or current infection with any of theseviruses. In addition, there was no evidence of active CMV orEBV infection (data not shown).

Seroprevalence of GBV-C/HGV in patients with aplastic anaemiaSerum samples were tested from 47 cases of aplastic anaemianot preceded by hepatitis. Most patients had receivedmultiple blood transfusions prior to analysis, and thereforemultiply transfused anaemic patients were used as a controlpopulation (Table I). Of the patients with aplastic anaemia,12 (25·5%) were viraemic, compared to 12/52 (23·1%) ofcontrols. This difference was not statistically significant(x2 = 0·08; P = 0·78; relative risk = 0·9; 95% confidencelimits 0·45–1·81).

A detailed transfusion history was available for 42 ofthe aplastic anaemia patients (Table II) and 30 of thethalassaemia/sickle cell anaemia patients. Most of thethalassaemia or sickle cell anaemia patients (24/30) hadreceived > 100 units of blood, with a mean of 350 units. Incontrast, only five aplastic anaemia patients had received> 100 units; there was a significant difference betweenviraemia in those who had received blood products beforeserum sampling (12/34) compared to those who had notreceived transfusions (0/8; x2 = 3·95; P = 0·05; lower 95%confidence limit 0·76).

Variation in genome sequenceThe 188 base pair product from reverse gene amplification ofthe virus was sequenced from five (two from hepatitis–aplasia patients, and three from non-hepatitis-associatedaplasia) of the positive samples. All of the sequences were

493Rapid Paper

q 1997 Blackwell Science Ltd, British Journal of Haematology 97: 492–496

Table I. Seroprevalence of GBV-C/HGV in individuals with aplasticanaemia and in multiply transfused patients with other haemato-logical diseases.

No. positive Percent

Aplastic anaemia patients 15/57 26·3Hepatitis/aplastic anaemia 3/10 30·0Aplastic anaemia 12/47 25·5

Control patients 12/52 18·4Sickle cell anaemia 4/22Thalassaemia 4/32Others* 4/10

* Others: sideroblastic anaemia (1), pancytopaenia (1), myelodys-plastic syndrome (4), ITP (2), Fanconi’s anaemia (2).

different (Fig 1), indicating that they did not representlaboratory cross contamination. The sequences hadbetween 75% and 80% identity to each other or to otherpublished sequences of GBV-C/HGV, demonstrating a highdegree of sequence variation even within this putativelywell-conserved region of the viral genome. Isolates fromthe patients with either hepatitis/aplastic anaemia oraplastic anaemia of other causes did not appear to beany more closely related to each other than to otherisolates.

DISCUSSION

Although the aetiology of aplastic anaemia remains unknown,several lines of evidence have suggested an immune-mediatedprocess with a viral infection as the precipitating event in somecases: exposure to known viruses (Linet et al, 1989); geographicclusters (Linet et al, 1985); a higher incidence in East Asia,where dense human and animal populations live in closeproximity (Young et al, 1986); and immunologic abnormalitiessimilar to the response evoked by a viral infection (Zoumbos etal, 1985). The most common apparently infectious antecedentof aplastic anaemia is hepatitis, and the hepatitis/aplasticanaemia syndrome is a recognized variant of aplastic anaemia.The presumed cause of the hepatitis is unknown, but mostcases are seronegative (non-A, non-B, non-C) (Hibbs et al,1992a; Pol et al, 1990).

GBV-C/HGV viraemia has recently been descibed in threecases of hepatitis/aplastic anaemia syndrome (Simons et al,1995a; Zaidl et al, 1996; Byrnes et al, 1996), suggesting thatit might be implicated as the aetiological agent of thepreceding hepatitis and a trigger for the subsequent marrowfailure. However, in the first case, both sera taken at the timeof hepatitis and prior to the administration of blood productswere negative (Simons et al, 1995a); in the other twopatients, GBV-C/HGV was reportedly detected in serum priorto treatment with blood products. In our analysis of NIHpatients with hepatitis-associated aplastic anaemia, GBV-C/

494 Rapid Paper

q 1997 Blackwell Science Ltd, British Journal of Haematology 97: 492–496

Table II. Correlation of number of blood transfusions and GBV-C/HGV viraemia in (a) aplastic anaemia patients and (b) sickle cell/thalassaemic patients.

No. of blood transfusions received

0 1–9 10–99 >100

(a) Negative 8 6 14 2Positive 0 2 7 3

Total 8 8 21 5

(b) Negative 0 1 4 19Positive 0 0 1 5

Total 0 1 5 24

Fig 1. Sequence alignments of the PCR products from isolates obtained from individuals with aplastic anaemia (1 and 2, hepatitis-associatedaplastic anaemia; 3, non-hepatitis aplastic anaemia) or multiply transfused patients (4 and 5). Sequences are compared to published sequences(GBV-C [GenBank U36380] (Leary et al, 1996a); HGV [U44402] and HGV-2 [U45966] (Linnen et al, 1996)).

HGV viraemia was detected in only three of 10 individuals, allthree having received multiple blood transfusions prior totesting (Brown et al, 1997a); viraemia was not detected inanother three individuals who had not received any bloodproducts. In addition, in the current study the prevalence ofGBV-C/HGV viraemia in non-hepatitis-associated aplasticanaemia patients was not significantly greater than that inthe anaemic patients who had received multiple transfusionsfor other indications. Ideally, studies should be performed onmany more patients and controls, and, in addition, aplasticanaemia patients prior to transfusion should be tested.However, such patients in large numbers are not generallyavailable for study. We were unable to detect HGV/GBV-Cviraemia in any of our patients who had not received bloodtransfusions. Nevertheless, given the presence of GBV-C/HGVin some apparently untransfused aplastic anaemia patients,the high degree of genetic heterogeneity of GBV-C/HGV, andthe similarity of the human immune response to flaviviruses(such as dengue) and abnormalities of the immune system inacquired aplastic anaemia, a member of this family, perhapsdistantly related to GBV-C/HGV, remains a possible candidateagent in the hepatitis/aplasia syndrome and in aplasticanaemia in general. Alternatively, and more likely in ourview, is the existence of a novel viral agent that is aetiologic inhepatitis/aplasia and other unexplained hepatitis syndromes.

Very little is known about the epidemiology or pathogeni-city of GBV-C/HGV. The prevalence of infection in the generalpopulation can only be roughly estimated, as there arecurrently no methods for the detection of antibody to GBV-C/HGV. In addition, the duration of GBV-C/HGV viraemiafollowing infection is unknown; in one study, in only one ofeight patients followed for 7–16 years did virus becomeundetectable, and this patient had been viraemic for over 10years (Masuko et al, 1996). Thus, after acquisition the virusmay persist and circulate for life. Also, it is not known if thevirus remains latent in the body following infection and ifreactivation may occur with immunosuppression, informa-tion important in assessing the role of a virus in the setting ofaplastic anaemia; it is possible that the increased detection ofGBV-C/HGV viraemia in some individuals with aplasticanaemia may indicate reactivation of latent virus, ratherthan its role as an aetiologic agent.

In conclusion, GBV-C/HGV viraemia is frequent in patientswith aplastic anaemia. However, it is known that the virus istransmitted by blood transfusion, and the presence ofviraemia in patients with aplastic anaemia correlates withtreatment with blood products, rather than as the aetiolo-gical agent of aplasia with or without prior hepatitis. In thisrespect GBV-C/HGV is similar to hepatitis C, in that thefrequent detection of hepatitis C viral genome in serum frompatients with hepatitis/aplastic anaemia syndrome is alsoassociated with blood transfusion rather than the aetiologi-cal agent of bone marrow failure (Hibbs et al, 1992a).

REFERENCES

Aikawa, T., Sugai, Y. & Okamoto, H. (1996) Hepatitis G infection indrug abusers with chronic hepatitis C. (Letter). New EnglandJournal of Medicine, 334, 195–196.

Alter, H.J. (1996) The cloning and clinical implications of HGV andHGBV-C. (Editorial). New England Journal of Medicine, 334, 1536–1537.

Alter, H.J. & Bradley, D.W. (1995) Non-A, non-B hepatitis unrelatedto the hepatitis C virus (non-ABC). Seminars in Liver Disease, 15,110–120.

Alter, M.J., Margolis, H.S., Krawczynski, K., Judson, F.N., Mares, A.,Alexander, W.J., Hu, P.Y., Miller, J.K., Gerber, M.A., Sampliner,R.E., Meeks, E.L. & Beach, M.J. (1992) The natural history ofcommunity acquired hepatitis C in the United States. New EnglandJournal of Medicine, 327, 1899–1905.

Brown, K.E., Tisdale, J., Barrett, A.J., Dunbar, C.E. & Young, N.S.(1997a) Hepatitis associated aplastic anemia. New England Journalof Medicine, 336, 1059–1064.

Brown, K.E., Wong, S., Mat, B., Binh, T.V., Be, T.V. & Young, N.S.(1997b) High prevalence of GB virus C/hepatitis G virus inhealthy persons in Ho Chi Minh City, Vietnam. Journal of InfectiousDiseases, 175, 450–453.

Byrnes, J.J., Banks, A.T., Piatack, M., Jr & Kim, J.P. (1996) HepatitisG-associated aplastic anaemia. Lancet, 348, 472.

Dawson, G.J., Schlauder, G.G., Pilot-Matias, T.J., Theile, D., Leary,T.P., Murphy, P., Rosenblatt, J.E., Simons, J.N., Martinson, F.E.A.,Gutierrez, R.A., Lentino, J.R., Pachucki, C., Muerhoff, A.S., Widell,A., Tegtmeier, G., Desai, S. & Mushahwar, I.K. (1996) Prevalencestudies of GB virus-C infection using reverse transcriptase-polymerase chain reaction. Journal of Medical Virology, 50, 97–103.

de Lamballerie, X., Charrel, R.N. & Dussol, B. (1996) Hepatitis GBvirus C in patients on hemodialysis. (Letter). New England Journalof Medicine, 334, 1549.

Fiordalisi, G., Zanella, I., Mantero, G., Bettinardi, A., Stellini, R.,Paraninfo, G., Cadeo, G. & Primi, D. (1996) High prevalence of GBvirus C infection in a group of Italian patients with hepatitis ofunknown etiology. Journal of Infectious Diseases, 174, 181–183.

Hibbs, J.R., Frickhofen, N., Rosenfeld, S.J., Feinstone, S.M., Kojima,S., Bacigalupo, A., Locasciulli, A., Tzakis, A.G., Alter, H.J. &Young, N.S. (1992a) Aplastic anemia and viral hepatitis: non-A,non-B, non-C? Journal of the American Medical Association, 267,2051–2054.

Hibbs, J.R., Issaragrisil, S. & Young, N.S. (1992b) High prevalence ofhepatitis C viremia among aplastic anemia patients and controlsfrom Thailand. American Journal of Tropical Medicine and Hygiene,46, 564–570.

Kuwada, S.K., Patel, V.M., Hollinger, F.B., Lin, H.J., Yarbough, P.O.,Wiesner, R.H., Kaese, D. & Rakela, J. (1994) Non-A, non-Bfulminant hepatitis is also non-E and non-C. American Journal ofGastroenterology, 89, 57–61.

Leary, T.P., Muerhoff, A.S., Simons, J.N., Pilot-Matias, T.J., Erker, J.C.,Chalmers, M.L., Schlauder, G.G., Dawson, G.J., Desai, S.M. &Mushahwar, I.K. (1996a) Sequence and genomic organization ofGBV-C: a novel member of the Flaviviridae associated with humannon-A–E hepatitis. Journal of Medical Virology, 48, 60–67.

Leary, T.P., Muerhoff, A.S., Simons, J.N., Pilot-Matias, T.J., Erker, J.C.,Chalmers, M.L., Schlauder, G.G., Dawson, G.J., Desai, S.M. &Mushahwar, I.K. (1996b) Consensus oligonucleotide primers forthe efficient detection of GB virus C. Journal of Virological Methods,56, 119–121.

Linet, M.S., Markowitz, J.A., Sensenbrenner, L.L., Warm, S.G., Weida,S., Van Natta, M.L. & Szklo, M. (1989) A case–control study ofaplastic anemia. Leukemia Research, 13, 3–11.

Linet, M.S., Tielsch, J.M., Markowitz, J.A., Sensenbrenner, L.L.,McCaffrey, L.D., Warm, S.G., Vanderslice, S.F., Morgan, W.F.,Bearden, J.D. & Szklo, M. (1985) An apparent cluster of aplasticanemia in a small population of teenagers. Archives of InternalMedicine, 145, 635–640.

495Rapid Paper

q 1997 Blackwell Science Ltd, British Journal of Haematology 97: 492–496

Linnen, J., Wages, J., Zhang-Keck, Z., Fry, K.E., Krawczynski, K.Z.,Alter, H., Koonin, E., Gallagher, M., Alter, M., Hadziyannis, S.,Karayiannis, P., Fung, K., Nakatsuji, Y., Shih, J.W., Young, L.,Piatak, M., Hoover, C., Fernandez, J., Chen, S., Zou, J., Morris, T.,Hyams, K.C., Ismay, S., Lifson, J.D., Hess, G., Foung, S.K., Thomas,H., Bradley, D., Margolis, H. & Kim, J.P. (1996) Molecular cloningand disease association of hepatitis G virus: a transfusion-transmissible agent. Science, 271, 505–508.

Masuko, K., Mitsui, T., Iwano, K., Yamazaki, C., Okuda, K., Meguro,T., Murayama, N., Inoue, T., Tsuda, F., Okamoto, H., Miyakawa, Y.& Mayumi, M. (1996) Infection with hepatitis GB virus C inpatients on maintenance hemodialysis. New England Journal ofMedicine, 334, 1485–1490.

Nubling, C.M. & Lower, J. (1996) GB-C genomes in a high risk group,in plasma pools, and in intravenous immunoglobulin. (Letter).Lancet, 347, 68.

Pol, S., Driss, F., Devergie, A., Brechot, C., Berthelot, P. &Gluckman, E. (1990) Is hepatitis C virus involved in hepatitis-associated aplastic anemia? Annals of Internal Medicine, 113, 435–437.

Purcell, R.H. (1994) Hepatitis viruses: changing patterns of humandisease. Proceedings of the National Academy of Sciences of the UnitedStates of America, 91, 2401–2406.

Sallie, R., Silva, A.E., Purdy, M., Smith, H., McCaustland, K., Tibbs,C., Portman, B., Eddleston, A., Bradley, D. & Williams, R. (1994)

Hepatitis C and E in non-A non-B fulminant hepatic failure: apolymerase chain reaction and serological study. Journal ofHepatology, 20, 580–588.

Simons, J.N., Leary, T.P., Dawson, G.J., Pilot-Matias, T.J., Muerhoff,A.S., Schlauder, G.G., Desai, S.M. & Mushahwar, I.K. (1995a)Isolation of novel virus-like sequences associated with humanhepatitis. Nature Medicine, 1, 564–569.

Simons, J.N., Pilot-Matias, T.J., Leary, T.P., Dawson, G.J., Desai, S.M.,Schlauder, G.G., Muerhoff, A.S., Erker, J.C., Buijk, S.L., Chalmers,M.L., VanSant, C.L. & Mushahwar, I.K. (1995b) Identification oftwo flavivirus-like genomes in the GB hepatitis agent. Proceedingsof the National Academy of Sciences of the United States of America,92, 3401–3405.

Yoshiba, M., Okamoto, H. & Mishiro, S. (1995) Detection of the GBV-C hepatitis virus genome in serum from patients with fulminanthepatitis of unknown aetiology. Lancet, 346, 1131–1132.

Young, N.S., Issaragrisil, S., Chieh, C.W. & Takaku, F. (1986) Aplasticanaemia in the Orient. British Journal of Haematology, 62, 1–6.

Zaidl, Y., Chapman, C.S. & Myint, S. (1996) Aplastic anaemia afterHGV infection. Lancet, 348, 471–472.

Zoumbos, N.C., Gascon, P., Djeu, J.Y., Trost, S.R. & Young, N.S.(1985) Circulating activated suppressor T lymphocytes in aplasticanemia. New England Journal of Medicine, 312, 257–265.

Zuckerman, A.J. (1996) Alphabet of hepatitis viruses. Lancet, 347,558–559.

496 Rapid Paper

q 1997 Blackwell Science Ltd, British Journal of Haematology 97: 492–496