Potential conflict of interest: Nothing to report.
Eight genotypes (A-H) of hepatitis B virus (HBV) have been identified. However, the impact of different genotypes on the clinical course of hepatitis B infection remains controversial. We investigated the frequency and clinical outcome of HBV genotypes and genotype mixtures in HBV-infected patients from Vietnam, Europe, and Africa. In addition, we analyzed the effects of genotype mixtures on alterations in in vitro viral replication. In Asian patients, seven genotypes (A-G) were detected, with A, C, and D predominating. In European and African patients, only genotypes A, C, D, and G were identified. Genotype mixtures were more frequently encountered in African than in Asian (P = .01) and European patients (P = .06). In Asian patients, the predominant genotype mixtures included A/C and C/D, compared to C/D in European and A/D in African patients. Genotype A was more frequent in asymptomatic compared with symptomatic patients (P < .0001). Genotype C was more frequent in patients with hepatocellular carcinoma (HCC; P = .02). Genotype mixtures were more frequently encountered in patients with chronic hepatitis in comparison to patients with acute hepatitis B (P = .015), liver cirrhosis (P = .013), and HCC (P = .002). Viral loads in patients infected with genotype mixtures were significantly higher in comparison to patients with a single genotype (P = .019). Genotype mixtures were also associated with increased in vitro HBV replication. In conclusion, infection with mixtures of HBV genotypes is frequent in Asia, Africa, and Europe. Differences in the replication-phenotype of single genotypes compared to genotype-mixtures suggest that co-infection with different HBV-genotypes is associated with altered pathogenesis and clinical outcome. (HEPATOLOGY 2006;43:1375–1384.)
The clinical course and long-term outcome of hepatitis B virus (HBV) infection is affected by several factors including viral genotype. For instance, it has been reported that the HBV genotype C induces more severe liver disease than HBV genotype B in Asia,1, 2 whereas HBV genotype A has been more frequently associated with the development of chronic infection than HBV genotype D in Europe.3
Eight genotypes (A to H) of HBV have been identified based on divergence over the entire genomic sequence of more than 8%.4–7 HBV genotypes have specific geographic distributions,8, 9 with genotype A found predominantly in Northwest Europe, North America, central and sub-Saharan Africa; genotypes B and C in Southeast Asia, China, and Japan; genotype D in the Mediterranean, the Middle East, and India; genotype E in Africa; genotype F in American natives, Polynesia, and Central and South America; genotype G in the United States and France,6, 10 and genotype H in Central America.7
HBV infection is endemic in Vietnam, where up to 26% of healthy individuals are seropositive for hepatitis B surface antigen (HBsAg).11 Scant information is available about the prevalence and clinical significance of the different genotypes of HBV in Vietnam. One previous study based on patients who had emigrated from Vietnam suggested that genotypes B and C may be the predominant viral genotypes in Vietnam.12 In addition, although HBV has been detected in 90% of patients diagnosed with hepatocellular carcinoma (HCC) in Vietnam,11 in contrast to Taiwan,13 China,14 and Japan,2 an association between HCC, severity of liver disease, and specific HBV genotypes has not been described in Vietnamese patients.
Little is also known about coinfection with different HBV genotypes in this population. Infection with mixtures of viral genotypes has previously been reported, although the presence of minor viral quasi-species may not be detectable using standard genotyping assays or direct sequencing.15 To overcome this limitation and to investigate the frequency of coinfection with mixtures of HBV-genotypes, Hannoun and colleagues16 developed an HBV core domain restriction fragment polymorphism (RFLP) assay to accurately determine the relative mixtures of viral genotypes in patient sera. These investigators found that genotypic coinfection is frequent in patients who are chronically infected with HBV, although in general only one viral genotype predominated.16 However, little is known about potential viral–viral interactions that occur between mixtures of HBV genotypes and of the clinical course in coinfected patients.
We have investigated the frequency of infection with different HBV genotypes, serum HBV viral loads, and clinical course in patients infected with different genotypes and mixtures of genotypes of HBV in Vietnamese (Asian) patients and compared these to European and African patients. In addition, by using an in vitro transfection assay, we also investigated the replication phenotype of mixtures of HBV genotypes compared to infection with single HBV genotypes.
Three hundred seventy-five HBV-infected Vietnamese individuals were included in this study. Of these, 289 were symptomatic patients, 29 were on hemodialysis, and 86 were healthy asymptomatic chronic HBV carriers. All patients were enrolled at Tran Hung Dao Hospital, Bach Mai Hospital, 103 Military Hospital, Hanoi, Vietnam, between 2000 and 2002. The patients were subdivided into five clinical profiles including acute hepatitis B (AHB), chronic hepatitis B (CHB), liver cirrhosis (LC), HCC, and asymptomatic carriers (ASYM). In addition, 38 European patients with CHB infection were enrolled from the Department of Gastroenterology, University-Hospital of Tuebingen, Germany, and 47 hepatitis B–infected Gabonese children were enrolled from the Medical Research Unit, Albert Schweitzer Hospital, Lambaréné, Gabon. All Vietnamese subjects had a positive serum HBsAg by EIA (Abbott, Chicago, IL) and were negative for antibodies to hepatitis C virus and HIV. None of the study participants from Vietnam had a history of alcohol or drug abuse, and none received antiviral for hepatitis B infection or immunosuppressive therapy before or during the course of this study. The clinical course and severity of hepatitis infections, liver biochemical tests, serological markers of HBV, including diagnostic tests for HCC in the Vietnamese patients, have been described in detail previously.17, 18 With the exception of the ASYM group, Vietnamese patients with chronic HBV were classified on the basis of histological examination after liver biopsy into those with or without evidence of either cirrhosis or carcinoma.18, 19
European patients were classified as having active CHB with or without cirrhosis based on clinical examination, biochemical analysis of serum liver function tests, including serum alanine (ALT) and aspartate aminotransferase, positive serum HBsAg for a period of greater than 6 months, and liver biopsy.
Gabonese children admitted to hospital or treated as outpatients for malaria were enrolled in this study. The diagnosis for hepatitis B infection was based on clinical criteria and positive serum HBV-DNA(+) for a period of greater than 6 months. The clinical profiles of the African patients have also been described elsewhere.20, 21
The collection samples were performed using new equipment for every individual. Sample processing (DNA-extraction, template preparation, spinning and aliquoting, master mix preparation) and polymerase chain reaction (PCR) amplication were done in separate laboratory rooms, which are all certified for molecular diagnostics and especially for PCR, using standard precautions to prevent assay contamination.
The presence of HBV genotype mixtures in patient sera was determined by a modified RFLP PCR assay that has previously been reported by Hannoun and coworkers.16 To increase the sensitivity of the RFLP PCR assay, we used nested PCR to amplify HBV-DNA. Primers used for the first PCR were: HB-pre-16 5′-CACCTCTGCCTAATCATCTCT-3′ and HBV-590as 5′-CTGCGAGGCGAGGGAGTT-3′. Reactions were initially denatured at 94°C for 4 minutes followed by 35 cycles of 94°C for 30 seconds, 55°C for 30 seconds, and 72°C for 30 seconds followed by a final extension for 10 minutes at 72°C. The cycle conditions of the second-round PCR were as described elsewhere.16 The DNA amplicons were incubated with the restriction enzyme Tsp509I, and the restriction patterns and predicted fragment sizes determined (Fig. 1). In addition, to verify the RFLP-PCR results, direct DNA sequencing on both strands using primers for nested PCR of the HBV isolates from the serum HBV-infected patients was performed as described previously.22 The sequences obtained were matched with the National Center for Biotechnology Information GenBank and compared with recently described HBV-prototypes (Accession No. for: HBV-A Z72478; HBV-B D00329; HBV-C X01587; HBV-D V01460; HBV-E X75657; HBV-F X75658; HBV-G AF160501).
Replication-Competent HBV Plasmids.
The HBV replication-competent plasmids pHBV1.28 and pHBV1.5 (genotype A), pHBV1.2 (genotype D), and HBVpd4aI (genotype C) containing a 1.2mer, 1.28mer, 1.5mer, or dimer HBV genome, respectively, have been described previously.23–25
Cell Culture and Transfection Experiments.
HepG2 cells were grown in Dulbecco's modified Eagle medium with 10% fetal calf serum as described previously.26 Cells were transfected with 2 μg of plasmid HBV-DNA in a total of 20 μL Fugene-6 reaction mixture (Roche Applied Science, Mannheim, Germany) as recommended by the manufacturer and harvested after 5 days. All experiments were performed in triplicate.
Quantitative Real-time HBV-PCR.
HBV-DNA quantification was performed with quantitative real-time PCR as previously described .17 Amplification and detection was carried out using the GeneAmp 7300 sequence analyzer (Applied Biosystems, Perkin-Elmer, Foster City, CA). HBV-plasmid DNA (pHBV1.28) and the WHO international standard for HBV-DNA (NIBSC Code 97/746) in increasing dilutions (5–5 × 105 genomes per vial) were included to standardize the assay. All samples were analyzed in duplicates. A quantitative PCR (qPCR) of the adenosine triphosphate synthase-6 gene was also performed, as a control for the addition of equivalent amounts of human DNA in the qPCR, as described previously.27
Isolation and Quantification of Progeny HBV-DNA.
Progeny HBV-DNA was isolated from transiently transfected HepG2 cells as previously described .28 In brief, transiently transfected cells were harvested 5 days after transfection and lysed with buffer containing 1% Igepal (Sigma-Aldrich, Munich, Germany). Intracellular HBV-core particles were immunoprecipitated with anti-HBV-core antigen antibodies (Dako, Hamburg, Germany). Contaminating HBV-DNA and HBV-plasmid DNA was eliminated by subsequent digestion with deoxyribonuclease-I (Roche Biochemica, Mannheim, Germany) and DpnI. HBV-DNA was extracted from the HBV core particles by digestion with proteinase K and 1% sodium dodecyl sulphate followed by phenol-chloroform extraction and ethanol precipitation as described.23, 28 HBV-progeny DNA was quantified using qPCR and calculated according to genome equivalents per microgram of isolated nucleic acid (GE/μg). Additionally, to verify the quantification results of the progeny HBV-DNA assays, HBV-DNA was isolated from culture medium supernatants as previously described and analyzed using qPCR.28
Statistical analysis were performed using contingency table analysis with Fisher's exact test (available at www.stata.com), and the non-parametric Mann-Whitney U-test or Kruskall-Wallis test and Spearman's rank-tests using Statview, version 4.57 (available at www.statview.com).
The study was approved by the Institutional Review Board of the Tran Hung Dao Hospital, Hanoi, Vietnam, by the ethics committee of the International Foundation for the Albert Schweitzer Hospital in Lambaréné, Gabon, and by the ethics committee of the University Hospital of Tuebingen.
Distribution of HBV-Genotypes in Three Populations: Asia, Europe, and Africa.
In Vietnamese patients, 7 genotypes A to G but not genotype H could be detected. In contrast, in European and in African patients only the four genotypes A, C, D, and G were identified (Table 1). Fifteen different genotype mixtures were detected by RFLP-PCR. These included; A/B, A/C, A/D, A/E, A/F, A/G, B/C, B/F, C/D, C/E, C/F, C/G, D/G, F/G, and A/C/D. Genotype mixtures were more frequent in African (14/47; 29.8%) than Asian (59/375; 15.7%), and European (5/38; 13.2%) patients (Table 2). However, with exception of the A/C (25/375; 6.66%), A/D (6/375; 1.6%), and C/D (10/375; 2.66%) combinations in Asian patients the number of remaining genotype mixtures was too low to permit a more detailed individual analysis. Analysis of the geographical distribution of genotypes indicated that genotype A was significantly more frequent in African patients (26/47; 55.32%), than either Asian (68/375; 18.13%) or European patients (6/38; 15.79%) (P < .0001 and P = .0002, respectively). In contrast, in Asian patients, genotype C predominated and was significantly more frequent than in either African or European patients (P = .011 and P = .017; respectively). Compared to Asian and African patients, genotype D was the predominant genotype in European patients (P = .0006 and P < .0001, respectively). Genotype B was only found in Asian patients. However, genotype D was also not infrequent in Asian patients (76/375; 20.26%; Table 1).
Table 1. Distribution of HBV Single Genotypes in Three Populations
Asiana n = 316/375 (%)
Europeanb n = 33/38 (%)
Africanc n = 33/47 (%)
Total N = 382/460 (%)
Abbreviation: NA, not applicable.
a vs. c < .0001 (33.36) b vs. c = .0002 (13.99)
a vs. b = .017 (5.66) a vs. c = .011 (6.42)
a vs. b = .0006 (11.84) a vs. c = .0024 (9.21) b vs. c < .0001 (22.85)
a vs. b = .0012 (10.48) b vs. c = .035 (4.45)
Table 2. Distribution of HBV Genotype Mixtures in Three Populations
Asiana n = 59/375 (%)
Europeanb n = 5/38 (%)
Africanc n = 14/47 (%)
Abbreviation: NA, not applicable.
a vs. b > .05 a vs. c = 0.13 (6.16)
a vs. b = .045 (4.0) a vs. c < .001 (32.0)
Other remaining genotype mixtures
Clinical Symptoms and Liver Parameters of HBV-Infected Vietnamese.
The clinical profiles and serum biochemical analyses of the different Vietnamese patient groups are shown in Table 3. Patients with AHB were significantly younger than those in the other clinical categories (P < .001). As expected, the serum ALT and aspartate aminotransferase levels in the AHB group were significantly higher compared with the other groups (P < .001). The prothrombin time was significantly prolonged in patients with LC compared with the other clinical groups (P < .001, Table 3).
Table 3. Characteristics of Vietnamese Patients With HBV Infection
Abbreviations: ND, not done; IU, international units.
P < .001.
P < .05 for comparison with all other groups.
Patients with chronic hepatitis B were defined as having a serum AST and ALT 2-fold higher than healthy controls, positive serum HBsAg for greater than 6 months, anti-HBc-IgG(+), anti-HBc-IgM(−), no ascites, peripheral edema or splenomegaly, no clinical additional clinical evidence of cirrhosis, and liver histopathology consistent with chronic hepatitis.
Asymptomatic, no jaundice, normal AST and ALT, HBsAg(+), but HBeAg(−).
Distribution of HBV Genotypes in Vietnamese Population.
In Vietnamese patients, the genotypes most frequently detected were C (94/375; 25.06%), D (76/375; 20.26%), and A (68/375; 18.13%) (Table 4). Genotype A was significantly more frequent in asymptomatic carriers (P < .0001, chi-squared(1) = 18.26) and patients without HCC (P = .0016, chi-squared(1) = 9.99). In contrast, genotype C was significantly more frequent in patients with HCC (P = .023, chi-squared(1) = 5.15) and asymptomatic HBV (P = .035, chi-squared (1) = 4.45). In addition, by examining the distribution of genotypes in the different clinical groups HCC occurred significantly more frequently in patients infected with genotype C than those infected with genotypes B and A (P = .0005 and P = .0001; respectively). Genotype D was identified significantly more frequently in patients with LC than other clinical groups (P < .0001).
Table 4. Distribution of HBV Genotypes in Vietnamese Patients With HBV Infection
ASYM n = 86 (%)
AHB n = 43 (%)
CHB n = 70 (%)
LC n = 92 (%)
HCC n = 84 (%)
Total n = 375 (%)
Other remaining genotype mixtures
Distribution of Genotype-Mixtures in HBV-Infected Vietnamese Patients.
The distribution of genotype mixtures in the Vietnamese patients according to the different clinical categories is shown in Table 5. Genotype mixtures were significantly more frequent in patients with CHB compared with patients with AHB, LC, and HCC (P < .02) and in patients with AHB, CHB, and ASYM compared with patients with severe disease (LC and HCC, P = .028; chi-squared(1) = 4.78). No significant difference was detected across the different clinical categories in patients infected with a single genotype.
Table 5. Distribution of Genotype Mixtures in HBV-Infected Vietnamese Patients
Analysis of serum HBV-DNA showed significantly higher viral loads in the AHB, CHB, and ASYM groups compared with either the LC or HCC groups (P < .05; Fig. 2A). The viral load in patients with HCC was significantly lower than in patients without HCC (P < .01; Fig. 2B). In the analysis of the relationship between viral replication and genotype, we found that serum HBV-DNA levels in patients singly infected with HBV genotype A and C were significantly higher than those infected with genotype D (P = .0002 and P = .0001, respectively, Fig. 3A). Interestingly, the viral load in patients infected with genotype mixtures was significant higher than those infected with single genotype overall (P = .019; Fig. 3B) and in LC group (P = .009; Fig. 3C). The HBV-DNA level in patients infected by genotype mixture C/D was significantly higher than in patients infected with single-genotype C and D (P < .0001; Fig. 5A).
Correlation of HBV Genotypes With HBV-DNA and Liver Disease.
Multivariate analysis indicated that genotypes C (R = 0.38; P < .001) and D (R = 0.22; P = .001) and age (R = 0.19; P = .007) were positively correlated to HCC development, whereas ALT (R = −0.24; P < .001) and mixed infections (R = −0.31; P < .001) were negatively correlated to HCC. Genotype D (R = 0.28; P < .001) and age (R = 0.14; P = .04) were positively correlated to LC. Moreover, mixed genotypes were positively correlated with CHB (R = 0.24; P = .001) and negatively correlated to age (R = −0.25; P < .001). Genotype A (R = 0.3; P < .001) was positively correlated to ASYM. Genotype C (R = 0.29; P = .048) was positively correlated to histological severity. Additionally, hepatitis B e antigen(+) (R = 0.41; P < .001) and ALT (R = 0.13; P = .04) were positively correlated to HBV-DNA. However, using pair-wise analysis, mixed infections (R = 0.13; P = .01) were positively correlated to HBV-DNA.
In Vitro Replication Levels of Different HBV Genotypes.
Because the serum viral loads in mixed infections were significantly higher than single-genotype infections, we performed in vitro transfection experiments in HepG2 cells to determine whether potential interactions between different viral genotypes altered HBV replication. Progeny HBV-DNA and HBV-DNA in secreted virions were extracted and used to evaluate the replication competence of different HBV genotypes by real-time PCR and reverse transcription PCR. Interestingly, consistent with the results of the in vivo viral loads, the in vitro experiments demonstrated that the replication level of HBV genotype A was significantly higher than the levels observed in cells transfected with genotypes C and D (P < .001; Fig. 4A–B). The results of the progeny HBV-DNA assays were in agreement with the qPCR of culture medium supernatants (data not shown). Moreover, the in vitro experiments revealed higher levels of HBV-replication in cells co-transfected with genotype mixtures compared to single-genotype transfections (P = .001; Fig. 4C). In accordance with the results of the serum HBV-DNA, the in vitro experiments demonstrated that the level of HBV-DNA of genotype mixture C/D was significantly higher than those transfected with single-genotype C and D (P = .0016; Fig. 5B).
Relation Between the Different HBV Genotypes In Vivo and In Vitro.
To investigate the relationship between the levels of replication of individual genotypes in patients infected with genotype mixtures, we performed semiquantitative PCR (sqPCR) analysis for genotype mixtures A/C, A/D, and C/D of both patient sera (Fig. 6A,C) and in vitro experiments (Fig. 6B,D). sqPCR analysis of genotype mixtures A/C and A/D demonstrated that HBV-genotype A dominated over genotypes C and D (Fig. 6A–D). No difference was detected between HBV genotypes C and D in patients or in in vitro experiments with genotype mixture C/D (Fig. 6A–D).
In this study we have analyzed the distribution of HBV genotypes and the associated clinical outcome of 460 HBV-infected individuals from three geographically distinct populations, Vietnam, Africa, and Europe, and the associated clinical outcome for the Vietnamese group. A sensitive PCR-RFLP developed by Hannoun et al.16 was adapted for our study, and this assay enabled us to detect 7 HBV-genotypes (A, B, C, D, E, F, and G), with a predominance of genotypes A, C, and D, in Vietnamese patients. In European and African patients, only HBV genotypes A, C, D, and G were identified. In accordance with previous studies, genotypes A and D predominated in European and African patients.8, 9 However, genotype A was most frequently detected in Vietnamese patients; this is in contrast to previous studies, which have shown a predominance of genotypes B and C in Asian patients.8, 13 However, Hannoun and colleagues recently reported the occurrence of recombinant HBV genotypes A and C strains from five Vietnamese patients,29 suggesting that genotype A may be more common in Vietnam than previously expected. Moreover, a recent study showed that recombination between HBV genotypes B and C is also frequent in Asia, including Vietnam.30
Increasingly, evidence now supports the view that different HBV genotypes have significant roles in determining the clinical outcome of liver disease and the response to antiviral therapies. This has been shown with genotypes A and D, which are prevalent in Europe and the United States, and genotypes B and C, which are prevalent in Asia.2, 3, 30–32 To identify the impact of HBV genotypes on the clinical outcome of HBV infection, we have characterized the genotype distribution and clinical outcome of subjects with well-characterized clinical profiles of liver disease and infection. In this study, we demonstrated that in Vietnam genotypes A and C were the predominant viruses in ASYM patients in comparison with symptomatic patients and patients with HCC in whom the predominant genotypes were C and D. This suggests that pathogenesis of genotype A may be different from the other viral genotypes. Multivariate analysis showed HBV genotypes C and D to be positively correlated to HCC, whereas genotype A was correlated to ASYM. This is further consolidated by our findings demonstrating an association between the development of HCC and genotype C and D but not with genotype A. This association has also been reported in studies from Taiwan,13 Japan,2 and China.14 Interestingly, in our Vietnamese population, genotype D was also prevalent in patients with LC disease.
In addition to demonstrating an association between HBV-genotypes and severity of clinical disease, by using real-time PCR, we were able to show that the serum HBV-DNA in patients infected with HBV genotypes C and D were significant lower compared to those infected with genotype A. Our in vitro cell culture findings demonstrating higher levels of HBV replication of genotype A compared to genotypes C and D were consistent with the results of the in vivo viral loads. We also found that, in accordance with the in vivo results, HBV genotype A predominated over the replication of genotypes C and D in genotype mixtures A/C and A/D. This, together with the higher level of replication of genotype A, may partly explain the observed association between genotype A and the more frequent progression to CHB than genotype D.3
The reasons for the differences observed in clinical course of disease and viral loads is not clear, although one could postulate that these may be the consequence of genotype-specific differences in viral replication, variability in the host immune response to different HBV genotypes33 possibly resulting in greater severity of liver disease and ultimately the development of LC and adverse effects on cell signaling events with the consequent development of HCC.34–36
Few studies have reported the occurrence of mixed HBV genotype infections16 or the effects of mixed infections on clinical outcome or viral replication. The study by Hannoun and coworkers16 included 30 patients who had received treatment with interferon alpha. Using a PCR-RFLP assay, these investigators demonstrated that 67% of patients were infected with more than one genotype.16 Recently, using the DHBV model, Walters and coworkers31 demonstrated that superinfection with a second duck HBV was not possible once the liver has become populated with a single species of duck HBV supercoiled DNA.37 However, this is in contrast to our finding of mixed HBV genotype infections in patients with CHB infection and argues that coinfection or superinfection with a second species of HBV is possible in human hosts.
In our study, infection with mixed HBV genotype was found in 16.9% (78/460) of patients. However, in patients with chronic hepatitis, the proportion of patients with genotype mixtures was significantly higher (28.6%; Table 5), again supporting the notion that infection of the human liver with more than a single viral population is possible. The results of the current study are consistent with those of previous reports, which suggest that genotypic coinfection is not infrequent in chronic HBV infection and that higher levels of HBV replication are associated with mixed-genotype infections.16 In summary, our report is the first showing an association between the distribution and impact of HBV genotypes and mixed genotype infections and of viral replication on clinical disease severity in HBV-endemic areas. Attributing precise roles to any given genotype or mixed genotype infection in the context of HBV pathogenesis or clinical progression is unclear. However, our findings indicate that the clinical course of CHB and the development of HCC is influenced by HBV genotype and by the occurrence of mixed genotype infection and warrants further investigation to understand these effects from a virological and pathogenesis perspective.
The authors thank Don L.V., Thang H.H., Nhuong D.X., Lan D.T., Binh D.T., Thoa L.T., Lam H.T., and Dung D.T. for their help in collecting samples and S. Illmann, H. Kaiser, and G. Janke for excellent technical assistance.