Variations in the functional domain of basal core promoter of hepatitis B virus among Eastern Indian patients with prevalence of genotypes A, C, and D among the same ethnic population†
Avik Biswas, Arup Banerjee, and Partha Kumar Chandra contributed equally to this work.
Abstract
Mutations in the basal core promoter (BCP) and precore (PC) regions are associated with persistent and intermittently high hepatitis B virus (HBV) replication in several patients. The variability in the functional domains of BCP and PC region of HBV and their association with disease progression and clinical outcome were assessed in Eastern India, an unique region where three HBV genotypes, A, D, and C are prevalent among the same ethnic group. PCR amplification and direct sequencing of BCP and PC region was done on sera obtained from 130 HBsAg positive subjects with different clinical presentations. Associations of the apparent risk factors with clinical advancement were evaluated by statistical methods including multiple logistic regression analyses (MLR). HBV genotype A was present in 33.08%, C in 25.38%, and D in 41.54% cases. Genotypes A and C were associated with higher rate of T1762/A1764 mutations than the most predominant genotype D. HBeAg negative state was associated with considerably higher rate of C1753 mutation. T1762/A1764 along with C1753 was common among cirrhosis and T1762/A1764 without C1753 was frequent among chronic liver disease cases. No significant association was found between A1896 point mutation and clinical status. Multivariate analysis revealed that T1762/A1764 double mutation, HBV/A, age ≥25 years, C1753 and A1899 were critical factors for clinical advancement while age ≥25 years and C1753 as significant predictor for cirrhosis in comparison with chronic liver disease. In conclusion, the analysis of the BCP variability may help in monitoring the progression towards advanced liver disease in Eastern Indian patients. J. Med. Virol. 83:253–260, 2011. © 2010 Wiley-Liss, Inc.
INTRODUCTION
Hepatitis B virus (HBV) infection represents a major health problem worldwide, with estimated 400 million chronically infected patients at risk of developing cirrhosis or hepatocellular carcinoma. The course of HBV infection and related liver injury depend on several host factors and genetic variability of the virus, which influences the viral replication as well as expression of viral antigens [Liaw, 2006]. During the course of chronic hepatitis B, a large proportion of the carriers achieve significant reduction in virus replication with loss of hepatitis B “e” antigen (HBeAg) and seroconversion to its antibody, antiHBe [Bonino and Brunetto, 2003]. Several HBeAg-negative subjects have persistent or intermittently high HBV replication, associated mostly with mutations in the basal core promoter (BCP) and precore (PC) regions that result in reduction or prevention of HBeAg synthesis without affecting the replicative ability of the virus. The most well known is a G to A mutation at nt 1896 in the PC region, which prevents HBeAg production by converting codon 28 of this region into a stop codon with premature termination of HBeAg translation [Carman et al., 1989]. The role of A1896 to the HBV infection course is still controversial, although it has been thought to aggravate liver disease severity, associated especially with fulminant hepatitis [Hunt et al., 2000; Ozasa et al., 2006].
The BCP region, nucleotide positions 1742–1849, plays a central role in HBV replication and morphogenesis. Three TA rich sequences: TA1 (nts 1750–1755); TA2 (nts 1758–1762); TA3 (nts 1771–1775) are located 20–35 bp 5′ to the PC mRNA start sites. All three regions are required for optimal transcription of PC mRNA. TA4 (nts 1788–1795), functions as both a protein binding site for initiation of transcription 25–30 bp downstream at the start site of pgRNA and as the initiator sequence for some of the PC mRNAs. The TA rich regions are the hot spot of mutations and possibly mutations in this region have clinical implications [Okamoto et al., 1994; Kramvis and Kew, 1999]. Recently several lines of evidences have indicated that T1762/A1764 double mutations are linked to the severity of liver diseases and reduced HBeAg expression with enhanced genome replication capacity, which have been contradicted by other studies. Moreover, site-directed mutagenesis of a wild-type clone revealed gradual enhancement of genome replication and reduction of HBeAg expression when T1762/A1764 double mutation, C1753/T1762/A1764 triple mutation, and C1753/T1762/A1764/T1766 quadruple mutations were gradually introduced [Parekh et al., 2003].
Depending upon >8% divergences in complete genome, HBV has been classified into eight genotypes (HBV/A to H), which are associated with different mutation rates in the BCP/PC region [Funk et al., 2002]. Previous studies on patients infected with HBV in India mainly focused on one particular point mutation like A1896 and/or T1762/A1764, which showed no association of this mutation with chronic or fulminant hepatitis in Western India [Gandhe et al., 2003a]. The present study was carried out to test the hypothesis that besides these mutations, other mutation(s) in the functional domains of BCP might be associated with disease severity. Therefore, to examine the mutations present in the functional domain of BCP region, a sequence-based study was performed on 130 patients with different genotypes (HBV/A, C, and D) from Kolkata, Eastern India and the mutations were analyzed among different clinical outcomes.
MATERIALS AND METHODS
Patients
One hundred and eighty-five HBsAg positive subjects, who were either voluntary blood donors or patients referred from different medical centers of the state of West Bengal, with or without clinical features of chronic liver disease were clinically examined, evaluated, particularly for antiviral therapy. Of these, 130 consecutive patients were chosen (111 male and 19 female) who fulfilled the inclusion criteria of this study. This study was carried out from July 2006 to January 2008.
Only those patients were included who had never received any antiviral therapy. Patients with history of alcohol intake, diagnosed with HIV and/or HCV infection markers or having history of high-risk behavior were excluded. Also patients with any other potential etiology of chronic liver disease (autoimmune liver disease, Wilson's disease) were excluded. The procedures to be employed were explained to the patients and only after obtaining informed consent, they were included in this study. The work was a part of the study approved by the Institutional Ethical Committee.
The patients were classified into three categories on the basis of clinical examination, liver biochemistry, ultrasonography, and liver biopsy. However, patients included in Group-I was not considered for liver biopsy.
Group-I (inactive HBsAg carrier state)
Forty-five asymptomatic carriers, mainly voluntary blood donors who were incidentally detected as HBsAg positive, normal ALT (<40 IU/L; evaluated 2–3 times at 3 months interval) and normal liver ultrasonography for a period of more than 6 months were included.
Group-II (chronic hepatitis B or chronic liver disease state)
Fifty symptomatic or asymptomatic patients, with persistently or intermittently elevated ALT (1.5 times upper limit of normal) for a period of more than 6 months and liver biopsy showing necro-inflammatory changes and fibrosis without nodule formation, and with or without ultrasonographic evidence of portal hypertension were included.
Group-III (cirrhosis)
Thirty-five patients, with clinical features of liver cell failure, portal hypertension and liver biopsy showing fibrosis with nodule formation, with ultrasonographic evidence of portal hypertension were included.
Patients were admitted as and when necessary and were evaluated in the Liver Clinic, Kolkata Medical College, Kolkata during the study. All patients underwent blood tests to evaluate the liver biochemistry (liver function tests, prothrombin time) and to establish the nature of chronic liver disease. Viral markers, autoimmune markers like anti-nuclear factor, ceruloplasmin and other tests were done as and when required. Routine investigations (hemogram, serum urea, creatinine, electrolytes) were done. All patients underwent ultrasonography with or without Doppler study. Upper gastrointestinal endoscopy was performed as and when required.
Blood samples were collected, serum separated, and stored in aliquots at −80°C for further screening. Liver biopsy was done in patients with persistently or intermittently elevated ALT, with or without evidence of liver cell failure or chronic liver disease, using the standard protocol.
Serologic Markers
All sera were screened for HBeAg, antiHBe (EQUIPAR, Varese, Italy) and HBsAg (Organon Teknika, Boxtel, The Nederland) by commercial (ELISA) method.
HBV-DNA Extraction, Quantification, Detection of BCP, PC Mutations, and HBV Genotypes
HBV-DNA extraction, amplification of BCP and PC regions were done as described earlier [Banerjee et al., 2005], followed by direct sequencing. Serum HBV-DNA was quantified by real-time Taqman PCR assay [Abe et al., 1999]. Variants in the BCP and PC region were identified by comparing with HBV full genome sequences belonging to different genotypes, previously reported from Eastern Indian population [Banerjee et al., 2006a]. Substitutions that were related to specific genotype were excluded from mutational analysis. Viral genotypes were established by restriction fragment length polymorphism (RFLP) method as described previously [Lindh et al., 1997], were confirmed by direct surface gene sequencing and phylogenetic analyses as described earlier [Banerjee et al., 2006b].
Statistical Analysis
Statistical analysis has been carried out using MINITAB Release 13.31 and Microsoft EXCEL. For continuous variables, like age, HBV-DNA and ALT, normality tests have been done, and for those which were found to be normal, tests for equality of averages were carried out using t-test, and for the others non-parametric two sample median tests were used. For testing proportions, asymptotic normality was used.
RESULTS
Characteristics of Patients
The demographic, clinical, and virological data of the 130 patients with different clinical outcomes included in this study were shown in Table I. Out of these 130 patients, 23 patients were HBeAg positive and 106 patients were HBeAg negative (Table III). The mean age was significantly higher among cirrhosis patients than among inactive carriers and chronic liver disease patients (P-value = 0.016 and P-value = 0.008, respectively). The ALT level was significantly lower in inactive carriers group than chronic liver disease and cirrhosis group (P-values < 0.0001). Significantly higher ALT was observed among chronic liver disease patients in comparison with cirrhosis patients (P-value = 0.008).
| Characteristic | Group-I (n = 45) | Group-II (n = 50) | Group-III (n = 35) | Total (n = 130) |
|---|---|---|---|---|
| M/F | 38/7 | 43/7 | 30/5 | 111/19 |
| Age (year)* | 33.04 ± 11.943a | 32.40 ± 11.794b | 39.31 ± 13.387a,b | 34.48 ± 12.55 |
| ALT (IU/L)* | 32.04 ± 10.205c,d | 87.36 ± 41.993c,e | 76.28 ± 59.077d,e | 71.99 ± 48.96 |
| HBeAg (+ve/−ve) | 8/37 | 10/40 | 5/29 | 23/106 |
| HBV genotypes, A/C/D | 10/9/26 | 16/14/20 | 17/10/8 | 43/33/54 |
| HBV-DNA (log10 copies/ml)* (n) | 3.46 ± 0.782f,g (n = 23) | 5.58 ± 0.837f (n = 39) | 5.06 ± 1.513g (n = 18) | 4.85 ± 1.35 (n = 80) |
| C1753 | 7 (15.56)h,i | 15 (30.00)h,j | 20 (57.14)i,j | 42 (32.31) |
| T1762/A1764 | 13 (28.89)k,l | 36 (72.00)k,m | 31 (88.57)l,m | 80 (61.54) |
| T1766/A1768 | 0 | 2 (4.0) | 4 (11.43) | 6 (4.62) |
| C1814 | 0 | 5 (10) | 0 | 5 (3.85) |
| A1896 | 13 (28.89) | 9 (18.00) | 6 (17.14) | 28 (21.58) |
| A1899 | 2 (4.44)n | 6 (12.00) | 6 (17.14)n | 14 (10.76) |
- a P = 0.016.
- b P = 0.008.
- c P < 0.0001.
- d P < 0.0001.
- e P = 0.008.
- f P < 0.0001.
- g P = 0.0003.
- h P = 0.043.
- i P < 0.0001.
- j P = 0.005.
- k P < 0.0001.
- l P < 0.0001.
- m P = 0.023.
- n P = 0.036.
- * Mean ± SD.
The genotypes and the sequences of the entire BCP and PC regions were determined in all the samples; of them 43 (33.08%), 33 (25.38%), and 54 (41.54%) had genotypes HBV/A, HBV/C, and HBV/D, respectively. HBV/D (26/45; 57.78%) was predominant among inactive carriers and HBV/A (17/35, 48.57%) among cirrhosis cases (Table I). Notably, the presence of HBV genotype A or C increased significantly with advancing clinical stages (P-value = 0.039 for inactive carriers versus chronic liver disease, and P-value = 0.042 for chronic liver disease vs. cirrhosis) although all genotypes were found in all the three groups.
When the HBeAg positive and HBeAg negative cases of inactive carrier group was studied separately, no significant biochemical or virological differences was observed.
Frequency of Mutations in the TA Rich BCP Region
TA rich regions of BCP were examined for mutations. Mutations were mostly concentrated at nt 1753 in TA1 region; at nt 1762 in TA2 region and at nt 1764. In this region, T1762/A1764 was most frequent mutation found in 80/130 (61.54%) study samples followed by C1753 in TA1 region (42/130; 32.31%) (Table I). In addition, 8 base deletion (nt 1763–1770), in one genotype C sample isolated from one cirrhosis patient was also observed.
Basal Core Promoter Mutant and Clinical Outcome
Of 130 patients, 80 (61.54%) had BCP T1762/A1764 double mutations. The presence of T1762/A1764 double mutations significantly increased from inactive carrier to chronic liver disease patients (P < 0.0001) and cirrhosis patients (P < 0.0001) and is also significantly higher among cirrhosis patients than among chronic liver disease patients (P-value = 0.023) (Table I).
In TA rich region C1753 was associated exclusively with BCP mutations (T1762/A1764). Statistical analysis revealed that the presence of C1753 is significantly higher among chronic liver disease patients (P = 0.043) and cirrhosis patients (P < 0.0001) in comparison with inactive carrier cases, and is also significantly higher among cirrhosis patients than among chronic liver disease patients (P = 0.005) (Table I).
When the prevalence of BCP mutations were compared on the basis of HBV genotypes the proportions of T1762/A1764 double mutations in patients with HBV genotypes A, C, and D are found, respectively, 28/43 (65.12%), 30/33 (90.91%), and 22/54 (40.74%). Statistically, the proportion of T1762/A1764 double mutation in genotype A is significantly lower than that in genotype C (P = 0.002), but considerably higher than that in genotype D.
Eighty patients with T1762/A1764 mutations are again divided into two categories based on the presence of C1753 mutation (Table II). The mean age of patients with C1753 is significantly higher than those without C1753 (P = 0.049). The average HBV-DNA level is significantly higher in patients with C1753 in comparison to those without C1753 (P < 0.0001). Finally, statistically the chance of T1762/A1764 mutations with C1753 increases significantly with advancing clinical stages of liver disease (P = 0.043 for inactive carrier versus chronic liver disease, P = 0.005 for chronic liver disease vs. cirrhosis).
| Variables | Total | T1762/A1764 with C1753 (A) | T1762/A1764 without C1753 (B) | Total BCP mutation (A + B) | Wild type in BCP |
|---|---|---|---|---|---|
| N | 130 | 42 | 38 | 80 | 50 |
| Age (years)* | 34.48 ± 12.55 | 36.69 ± 12.66a | 35.08 ± 12.34a | 35.92 ± 12.46 | 32.18 ± 12.47 |
| Sex, M/F (%) | 111/19 | 34/8 (80.95%) | 32/6 (82.21%) | 66/14 (82.50%) | 45/5 (90.00%) |
| ALT levels* | 108 | 81.20 ± 62.27 | 79.44 ± 32.75 | 81.15 ± 51.17 | 55.14 ± 38.36 |
| Normal (≤40 IU/L) | 35 | 31.75 ± 6.96 | 29.75 ± 5.19 | 31.25 ± 6.45 | 30.00 ± 8.19 |
| Elevated ALT | 73 | 102.4 ± 63.8 | 86.54 ± 31.07 | 94.46 ± 50.33 | 83.24 ± 39.41 |
| HBV-DNA* (log10 copies/ml) (n) | 80 | 5.36 ± 1.048b (27) | 5.34 ± 1.353b (23) | 5.35 ± 1.185 (50) | 4.03 ± 1.221 (30) |
- a P = 0.049.
- b P < 0.0001.
- * Mean ± SD.
Mutations in the PC Region
In the PC region, the most common mutation was A1896 (28/130, 21.58%) and majority was in genotype D (18/28, 64.29%) and none in genotype A. The presence of A1896 does not differ significantly among the three categories of patients (Table I) but was found to be significantly higher in HBeAg negative than in HBeAg positive patients (P = 0.014) (Table III).
| Serology | HBeAg +ve | HBeAg −ve |
|---|---|---|
| N** | 23 | 106 |
| M/F | 20/3 | 90/16 |
| Age* | 32.83 ± 12.331 | 34.77 ± 12.664 |
| ALT* | 67.83 ± 41.598 | 71.89 ± 45.594 |
| HBV-DNA (log10 copies/ml) | 5.45 ± 1.69 | 4.74 ± 1.26 |
| Genotype (A/C/D) (%) | 8/6/9 (34.78/26.09/39.13) | 34/27/45 (32.08/25.47/42.45) |
| T1762/A1764 (A/G) (%) | 12/11 (52.17/47.83) | 68/38 (64.15/35.85) |
| C1753 (%) | 5 (21.74) | 37 (34.90) |
| A/G1896 (%) | 2/21a (8.70/91.30) | 26/80a (24.53/75.47) |
| A/G1899 | 2/21 (8.70/91.30) | 12/94 (11.32/88.68) |
- a P = 0.014.
- * Mean ± SD.
- ** For one sample HBeAg status cannot be determined.
A1899 mutation was also found in 14/130 (10.77%) patients most of them (8/14; 57.14%) were associated with A1896. The presence of A1899 is significantly higher among cirrhosis patients than among inactive carrier cases (P = 0.036). However, the HBeAg initiation codon (C1814) mutation was detected only among chronic liver disease patients (5/50, 10.0%) infected with HBV/A (3/5, 60%) and HBV/D (2/5, 40%), who were HBeAg negative (data not shown). In the other two groups no such mutation was found.
HBV Viral Load
HBV-DNA viral load could be quantified in 80 samples (Table I). The average HBV-DNA is significantly higher in chronic liver disease (P < 0.0001) and cirrhosis patients (P = 0.0003) in comparison to inactive carrier cases, but the difference between chronic liver disease and cirrhosis patients is not statistically significant (P = 0.0862) (Table I). Difference in viral load was considerably higher among HBeAg positive group, but the difference was not statistically significant (P = 0.0694) (Table III).
Multivariate Logistic Regression Analysis
- (i)
Stepwise logistic regression
In order to study the factors associated with the different clinical stages of hepatitis B infection, stepwise binary logistic regression was performed (Table IV).
The significant risk factors for cirrhosis were the presence of the T1762/A1764 double mutation, age ≥25 and HBV genotype A. However, when only HBeAg-negative patients were considered, the risk factors for cirrhosis were found to be T1762/A1764 double mutation and C1753 mutation.
The presence of wild-type BCP sequences showed significant association with inactive carrier group, irrespective of HBeAg status.
In chronic liver disease patients the presence of T1762/A1764 double mutation was only marginally significant (P = 0.05; OR = 2.10; 95% CI = 0.99–4.99) (data not shown in the table), with the confidence interval (CI) indicating that in chronic liver disease patients both A1762/G1764 and T1762/A1764 can be found, though the chance of getting T1762/A1764 seems comparatively higher.
- (ii)
Ordinal logistic regression
An ordinal logistic regression was performed taking the factors as mentioned above. As, it was possible to order the clinical stages of hepatitis B infection according to the increasing level of disease severity; such as level 1 (inactive carrier), level 2 (chronic liver disease), and level 3 (cirrhosis). The reference level was taken to be level 1.
Ordinal logistic regression analysis revealed that T1762/A1764 double mutation (P < 0.0005; odds ratio [OR] = 7.51; 95% CI = 3.07–18.41), A1899 mutation (P = 0.029; OR = 3.83; 95% CI = 1.15–12.78), and HBV genotype A (P = 0.044; OR = 2.26; 95% CI = 1.02–5.00) to be associated with higher levels of infection. While for HBeAg negative patients it is only T1762/A1764 double mutation.
- (iii)
Logistic regression for class of patients with either chronic liver disease or cirrhosis
To study the factors associated with cirrhosis when compared with chronic liver disease, stepwise binary logistic regression was performed for the 85 patients in these two classes, by including the above-mentioned factors. The significant predictive factors for cirrhosis were age ≥25 years (P-value = 0.023; OR = 4.80; 95% CI = 1.24–18.60) and presence of C1753 (P-value = 0.022; OR = 2.99; 95% CI = 1.17–7.63).
| Clinical stage | Factor | P-value | Odds ratio | 95% CI | |
|---|---|---|---|---|---|
| Lower | Upper | ||||
| Cirrhosis (n = 130) | T1762/A1764 | 0.001 | 7.04 | 2.23 | 22.23 |
| HBV genotype A | 0.035 | 2.59 | 1.07 | 6.27 | |
| Age ≥25 years | 0.033 | 4.20 | 1.12 | 15.78 | |
| Cirrhosis with HBeAg negative (n = 106) | T1762/A1764 | 0.028 | 6.26 | 1.22 | 32.13 |
| C1753 | 0.035 | 3.03 | 1.08 | 8.51 | |
| Inactive carrier (n = 130) | A1762/G1764 | <0.0005 | 9.48 | 4.06 | 22.13 |
| Inactive carrier with HBeAg negative (n = 106) | A1762/G1764 | <0.0005 | 8.97 | 3.59 | 22.42 |
DISCUSSION
Identification of host and viral factors associated with HBV-induced liver disease is necessary for developing effective management strategies. Accumulating evidences suggest association of HBV genotypes with disease progression [Sumi et al., 2003]. Because of distinct geographical distribution of HBV genotypes, comparison of HBV genotype and associated mutations was done in population where two genotypes are prevalent [Tanaka and Mizokami, 2007]. However, the results vary in different geographical areas due to differences in host and environmental factors [Datta et al., 2008a], India is a country with more than 43 million HBV carriers [Sarin, 1996]. HBV genotype distribution has been reported to vary in different parts of India [Thakur et al., 2002; Gandhe et al., 2003b; Vivekanandan et al., 2004; Kumar et al., 2005; Banerjee et al., 2006b; Chattopadhyay et al., 2006; Chauhan et al., 2006]. Limited data are available on HBV mutants from other parts of India and most of the study was performed on patients infected with HBV/A and HBV/D [Gandhe et al., 2003a; Chauhan et al., 2006]. In Eastern India, the distribution of HBV genotypes is unique, where in addition to HBV/A and HBV/D, considerable proportion of HBV/C is present in the same ethnic group [Vivekanandan et al., 2004; Banerjee et al., 2006a,b]. Thus to our knowledge, this is the first report that assessed the mutations in the functional domain of BCP region, among three HBV genotypes and their clinical significance among the same ethnic group.
In India, HBeAg negative chronic hepatitis B infections are common [Gandhe et al., 2003a; Banerjee et al., 2005; Chauhan et al., 2006]. A recent report from Northern India showed that BCP mutations at T1762/A1764 were more common in HBeAg negative, compared to HBeAg positive patients, which is similar to our findings [Chauhan et al., 2006]. Whereas another study from Western India [Gandhe et al., 2003a,b] reported that clinical manifestations were independent of BCP mutations (T1762/A1764). In the present study, the T1762/A1764 mutation occurred frequently in patients with chronic liver disease and cirrhosis than inactive carrier cases (Table I). Also the prevalence of T1762/A1764 mutation was significantly higher among the cirrhosis group than the chronic liver disease group. This finding suggests that the T1762/A1764 mutations occur in early immune clearance phase regardless of clinical outcomes and these mutants are selected persistently in the natural course of chronic HBV infection which supports previous observations [Gandhe et al., 2003a; Chauhan et al., 2006].
Several studies [Kidd-Ljunggren et al., 1997; Kramvis and Kew, 1999; Parekh et al., 2003; Jardi et al., 2004; Tong et al., 2005] have highlighted the importance of mutations in the TA rich BCP region. Besides common T1762/A1764 mutations, other mutations in TA1 and TA3 may play an important role in relation to viral replication and disease manifestation. Previous studies [Banerjee et al., 2005; Datta et al., 2008a] from this laboratory was focused on point mutation T1762/A1764, detected by RFLP method and thus other mutations in TA rich BCP region were not considered at that time. However, in this study, we amplified successfully and sequenced the BCP region of all 130 patients of different genotypes and clinical status, which gave us an apt opportunity to scan for other mutations in the functional part of the BCP region. In this study, among the functional region of BCP, TA1, TA2 regions were found to be variable, while very few mutations were detected in TA3 and TA4 regions. Among the mutated sequences we found 32.31% mutations at TA1 and 61.54% mutations at TA2 region. One of the mutations, C1753, was predominant in TA1 region and this mutation was associated exclusively with the occurrence of T1762/A1764 double mutation. Notably, from Table II it is clear that the triple mutation results significantly higher ALT and HBV-DNA load. Also the chance of triple mutation increases significantly with advancing clinical stages of liver disorder. T1766/A1768 double mutation (4.62%) and C1814 (3.85%) mutations were not considered for statistical analysis because of their very low prevalence in the study population, but it is important to note that 11.43% of the cirrhosis cases were found to have T1766/A1768 mutation when compared to chronic liver disease (4.0%) and inactive carrier (0%).
Several studies have indicated that genotype C is associated more frequently with BCP mutations as well as liver damage [Nakashima et al., 2004; Tanaka et al., 2006]. The genotype distribution documented in the present study is concordant with previous studies from Eastern India [Banerjee et al., 2005; Datta et al., 2008a]. In the present study, T1762/A1764 double mutation and C1753 mutation could be detected among subjects infected with all the three genotypes, which seem to indicate that this mutation may develop irrespective of the genotypes (A, C, and D) but the occurrences of these mutations were significantly higher among subjects infected with genotype A and genotype C. Hence considering the association of these mutations in BCP region with advanced liver diseases on one hand and with genotypes (A and C) on the other, the present results clearly support the previous observations from Eastern India, that genotypes A and C are associated more frequently with progressive liver disease in Eastern Indian population [Datta et al., 2008a]. In contrast, majority of the study samples from Western India belonged to HBV/D where C1753 mutation was rare [Gandhe et al., 2003a].
Previously in several reports, combined prevalence of both BCP (T1762/A1764) and PC (A1896) mutations showed association with the degree of fibrosis as well as with severity of liver disease [Sato et al., 1995; Lindh et al., 1999; Jardi et al., 2004; Song et al., 2006] as reflected by raised ALT levels and histological changes. However, in this study, the rate of PC mutation was low (28/130; 21.58%) in spite of high frequency of T1858 in samples of genotypes C and D, which supports the previous observations from this region [Banerjee et al., 2005; Chandra et al., 2007]. No significant association was found among clinical status and A1896 point mutation in the PC region. On the other hand, the present study highlights the importance of triple mutation (C1753/T1762/A1764) as an indicator of severe liver disease in our population. This severity may be due to the fact that mutations in TA1 were stem-disturbing mutations according to the model for the folding of pgRNA proposed by Kidd and Kidd-Ljunggren [1996] and reduce the amount of base pairing in the basal stem thereby facilitating reverse transcription.
Therefore, these findings suggest that these mutations in the late course of chronic HBV infection are associated with severity of liver disease. However, India is in low HCC prevalence zone [Datta et al., 2008b], therefore the role of these mutations in HCC cases could not be evaluated due to the presence of only a few HCC cases in the study population. Recently the importance of mutation at nt 1753 among HCC cases in comparison to non-HCC cases has been documented [Tanaka et al., 2006; Elkady et al., 2008]. The mutation at 1753 had earlier been associated with HCC among individuals infected either with HBV/C among Chinese/Japanese/Vietnamese [Tanaka et al., 2006; Truong et al., 2007; Yuan et al., 2007], or HBV/D [Elkady et al., 2008] among Mongolian population and among cirrhosis cases among Indonesian population [Utama et al., 2009]. Interestingly, the present study underscores the association of C1753 mutation with cirrhosis cases among three different genotypes, in the same ethnic population. The results of the present study are also supported by a recent in vitro study using site-directed mutagenesis of wild-type clone of genotype A, showing a gradual enhancement of viral replication by the double T1762/A1764, triple C1753/T1762/A1764, and quadruple C1753/T1762/A1764/T1766 mutations, along with decrease in HBeAg expression [Tong et al., 2005]. However, we found increased triple mutation with increasing severity of liver disease. Notably, we could detect very few C1753/T1762/A1764/T1766 quadruple mutations.
Multiple logistic regression (MLR) is an accepted statistical method for assessing association between a risk factor and a quanta outcome (probability of presence of a disease), after statistically adjusting for potential confounding effects of other risk factors. Stepwise MLR is used to identify the statistically significant risk factors. If the disease can be categorized into more than two categories according to its severity, ordinal logistic regression analysis is used. As laboratory data are usually categorical in nature, and can be explored in the form of OR, MLR turned out to be an effective tool for analysis in the present study. By applying different MLR approach age ≥25 years, HBV genotype A, T1762/A1764 double mutation, C1753 mutation and A1899 were found to be the critical factors influencing clinical advancement. Interestingly, in HBeAg negative patients mutations in TA1 and TA2 regions of BCP are risk factors for cirrhosis. While, age above 25 years and C1753 mutations are significant predictive factors for cirrhosis in comparison to chronic liver disease cases.
In conclusion, this study showed the importance of mutation in the functional domains of BCP region in relation to disease manifestation in Eastern Indian population. The significantly higher occurrence of these mutations and occurrence of advanced liver diseases among HBV genotypes A and C infected patients were also documented. In addition, using different MLR approach age ≥25 years, HBV genotype A, T1762/A1764 double mutation, C1753 mutation and A1899 were identified as significant risk factors for disease severity. While, age above 25 years and C1753 mutations are significant predictors for cirrhosis in comparison with chronic liver disease cases. Taken together, these data highlight the importance of BCP mutations and the clinical relevance of HBV genetic variability in identifying useful biomarkers for advanced liver disease.
Acknowledgements
We thank Sri Tapan Chakraborty and Sri Srikanta Deb for excellent technical assistance.
