INTRODUCTION — Viral DNA and especially RNA genomes are inherently variable due to errors introduced during replication. These errors are in the range of 1 mutation in 1000 to 100,000 nucleotides per replication cycle for RNA viruses, and approximately 1 mutation in 100,000,000 nucleotides per replication cycle for DNA viruses [1,2]. The higher mutation rate among RNA viruses is related to the lack of proofreading functions of RNA polymerases and reverse transcriptases. As a result, mutations in viral genomic sequences are generated naturally during viral replication and should be viewed as a normal biological event.
Hepatitis B virus (HBV), a member of the Hepadnaviridae family, replicates asymmetrically via reverse transcription of an RNA intermediate, making it prone to mutations [3]. The estimated mutation rate of the hepadnavirus genome is about 2 x 104 base substitutions/site/year, about 100 times higher than that of other DNA viruses but about 100 to 1000 times lower than that of other RNA viruses [4]. Although mutations can occur randomly along the HBV genome, the overlapping open reading frames limit the number and location of viable mutations. Because chronic HBV infection frequently persists for decades, many variants may exist within the same host at any given time and each variant may have multiple base changes.
Selection by the host for the fittest variant occurs after the random mutation process. Fitness may be defined at the cellular level (viruses that replicate most efficiently in a cell type will predominate) or at the extra-cellular level (viruses that avoid immune elimination will become dominant) [5]. Thus, the accumulation of viral variants depends upon the rate at which the variants are generated and the advantage they confer to the virus.
The genetically different viral species concomitantly present in a single cell or in a single individual are termed quasi-species [1]. Interpretation of the clinical significance of HBV mutations is complicated by the lack of standardized nomenclature, differences in sensitivities of assays used in their detection, and the presence of mutations in more than one region of the HBV genome even in the same species [6,7]. Despite these difficulties, in vivo analyses of naturally occurring viral variants and in vitro mutagenesis studies have identified some mutations that have a role in viral latency, pathogenesis of liver disease, immune escape, and resistance to antiviral therapy. This topic review will summarize the clinical relevance and molecular characteristics of common HBV variants.
PRECORE AND CORE PROMOTER VARIANTS — The precore/core region of the HBV genome encodes the nucleocapsid protein (HBcAg) and hepatitis B e-antigen (HBeAg) [8,9]. The core open reading frame has two transcripts with heterogeneous 5' ends and two in-phase initiation codons. HBeAg is translated from the precore mRNA, producing a precursor polypeptide comprising the precore and the entire core region. The precore polypeptide is translocated into the endoplasmic reticulum by a signal peptide. Cleavage of the amino and carboxy termini results in a secretory protein, HBeAg. HBcAg is translated from the pregenomic RNA. Both HBcAg and HBeAg contain B- and T-cell epitopes, of which some are shared. (See "Characteristics of the hepatitis B virus and pathogenesis of infection".)
Secretion of HBeAg can be modulated at both transcriptional and translational levels. Transcription of the precore and pregenomic RNAs is under the control of regulatory elements including the basal core promoter and the core upstream regulatory sequence [10]. Mutations in these domains decrease precore mRNA transcription and HBeAg synthesis [11,12]. At the translational level, mutations within the precore region can block the translation of the precore protein and HBeAg production. The most common mutation in the precore region is a point mutation (G to A switch at nucleotide 1896), which changes codon 28 from tryptophan to a stop [13-15]. Because this mutation occurs upstream of the core start codon and because HBcAg is translated from a different transcript, HBcAg and nucleocapsid production is not interrupted.
The biological role of HBeAg in the HBV replication cycle is uncertain. Expression of HBeAg is nonessential for virus replication in animal models [16] and in humans [17]. HBeAg may act as a tolerogen or a target for immune response. In utero exposure to HBeAg can induce immune tolerance in newborn mice [18]. Perinatal transmission of HBV from HBeAg-positive mothers results in chronic HBV infection in the majority of babies [19]. In addition, HBeAg appears to modulate the host's immune response [20-23]. Precore variants that do not produce HBeAg may be selected because they can evade immune clearance [24].
Precore variants — The most common precore variant has a point mutation at nucleotide 1896, creating a stop at codon 28 (G1896A, eW28X). This mutation is accompanied by another point mutation (G to A) at nucleotide 1899 in some patients. In vitro studies have demonstrated that precore protein can inhibit HBV replication, suggesting that precore stop codon variants may enhance viral replication [25,26]. However, patients with HBeAg-negative chronic hepatitis tend to have lower HBV DNA levels compared to those with HBeAg-positive chronic hepatitis [15,27].
Other naturally occurring mutations in the precore region include point mutations leading to initiation failure as well as deletions and insertions of nucleotides inducing frameshifts. In addition to the eW28X stop codon mutation, the most frequent precore mutations that can abrogate HBeAg production are point mutations in the start codon [28,29].
In a study involving 17 centers in the United States, precore variants were found in 27 percent of patients with chronic HBV infection, being significantly more common among the HBeAg-negative than the HBeAg-positive patients (38 versus 9 percent) [30]. A more recent study of 1036 patients (808 adults and 228 children) in the Hepatitis B Research Network found that the eW28X stop codon mutation was present as the dominant sequence in 29.4 percent of patients, and the prevalence increased with age; there was a higher prevalence in HBeAg-negative than in HBeAg-positive patients [31].
Core promoter variants — The most common core promoter variant involves a dual mutation A1762T, G1764A (TA change). A number of mutations have been found to occur together with the TA change, including changes at nucleotide 1653, 1753 to 1757, and the dual mutation C1766T G1768A. The latter mutations are most often found in patients with fulminant hepatitis [32].
Deletions within the basal core promoter, which overlap with the X gene, have been found to occur in asymptomatic carriers, patients with chronic hepatitis B or hepatocellular carcinoma (HCC), and individuals with no HBV serologic markers [33-35]. Infection with these deletion variants is usually associated with low HBV DNA levels.
In the nationwide United States study, the dual core promoter variants were found in 44 percent of patients with chronic HBV infection, being significantly more common among the HBeAg-negative than the HBeAg-positive patients (51 versus 36 percent) [30].
Detection — In addition to direct sequencing, several methods can be used to detect precore and core promoter variants including restriction fragment length polymorphism (RFLP) [36,37], line probe assay [38,39], ligase chain reaction [40], and colorimetric point mutation assay [41]. Some methods such as line probe assay and RFLP are more sensitive in detecting minor variants than direct sequencing [39].
The controversies surrounding the clinical significance of precore and core promoter variants is related to the heterogeneity of the study population, sensitivity of the methods used to detect these variants, and to the format in which results are reported. Some studies, particularly those that rely on less sensitive methods such as direct sequencing, only report the presence of variants when the variant is present as the predominant sequence, while other studies report the presence of variants when the variant is detected as a minor sequence in a mixed population.
Relationship to HBeAg status — Precore variants have been detected mainly in patients who are HBeAg negative but they can also be found in some HBeAg-positive patients [14,15,30,42]. HBeAg-positive patients with precore variants are more likely to undergo spontaneous HBeAg seroconversion and do so earlier than those with wild-type precore sequence [43,44]. These findings are in accordance with the observation that precore variants that prevent HBeAg production are selected around the time of HBeAg clearance [14,15,42]. However, not all HBeAg seroconversion requires the selection of precore mutations.
Unlike precore variants, core promoter variants can be detected in similar proportions of HBeAg-negative and HBeAg-positive patients [30,45,46]. One study demonstrated that emergence of the dual core promoter (TA) change was more common among HBeAg-positive patients who subsequently seroconverted compared to those who remained HBeAg positive, indicating that the selection of core promoter variants may play a role in HBeAg clearance [47]. This study also found that core promoter variants could be detected earlier than precore variants in patients who cleared HBeAg.
In a study from the Hepatitis B Research Network, which included 1036 participants (808 adults, 228 children) in North America who were not receiving antiviral therapy, the prevalence of the precore G1896A variant as the dominant sequence was more likely in HBeAg-negative versus HBeAg-positive participants (47.5 and 7.4 percent, respectively) [31]. The prevalence of the core promoter TA variant was also more likely in HBeAg-negative compared with HBeAg-positive participants (42.1 and 24.9 percent, respectively). Detection of these variants in HBeAg-positive patients appears to be an indicator of impending HBeAg clearance as HBeAg clearance rates were significantly higher in HBeAg-positive participants with dominant precore or core promoter variants (24.4 and 15.0 per 100 person-years) compared to those with wild-type sequence (6.0 per 100 person-years).
Relationship to HBeAg-negative chronic hepatitis B — Seroconversion from HBeAg to anti-HBe is usually accompanied by decreased levels of HBV replication and remission of liver disease. However, some patients who become HBeAg negative continue to have high HBV DNA levels and active liver disease, so-called HBeAg-negative chronic hepatitis B (HBeAg-CHB). Many of these patients harbor HBV variants in the precore or core promoter region [13-15,42,48,49], although some continue to have low levels of wild-type HBV. (See 'Precore variants' above and 'Core promoter variants' above.)
Patients have HBeAg-CHB if they meet the following criteria [50]:
●HBsAg positive and HBeAg negative
●Serum HBV DNA >104 international units/mL
●Persistent or intermittent elevations in aminotransferase levels that cannot be attributed to other causes
There are clinical differences between patients with HBeAg-negative CHB and those who are HBeAg positive [51]. Patients with HBeAg-CHB tend to have lower serum HBV DNA levels and are more likely to have fluctuating serum aminotransferase levels [15,27,51]. Patients with HBeAg-CHB also tend to be older and have more advanced liver disease because they are in a later phase of chronic HBV infection.
Patients with HBeAg-negative CHB have been reported in all parts of the world but are more common in Mediterranean countries and Asia [46,52-55]. Such patients are estimated to account for 7 to 30 percent of those with chronic hepatitis B worldwide, based upon studies from the 1990s [56]. However, the exact prevalence of HBeAg-negative CHB is not known since large population-based studies have not been performed. The ratio of HBeAg-negative CHB to HBeAg-positive CHB has increased worldwide because of the decline in new cases of HBV infection secondary to implementation of HBV vaccination and aging of previously infected persons. In many countries including Asian countries, HBeAg-negative CHB is now more prevalent than HBeAg-positive CHB.
The geographical differences in the relative frequency of HBeAg-negative CHB and HBeAg-positive CHB are partly related to differences in the distribution of HBV genotypes [57-59]. (See "Clinical significance of hepatitis B virus genotypes".)
As examples:
●The most common HBV variant associated with HBeAg-negative CHB is the precore stop codon variant G1896A, which is usually found in association with the HBV genotype D [52]. This genotype is prevalent in the Mediterranean basin. In some countries (eg, Italy and Greece) where HBV genotype D is dominant, >90 percent of patients with chronic HBV infection are HBeAg negative.
●HBeAg-negative chronic hepatitis B (HBeAg-CHB) is rarely detected in association with genotype A (which is prevalent in the United States and North-West Europe).
●Studies in Hong Kong found that only 40 percent of patients with HBeAg-negative CHB had the precore variant. HBV genotypes B and C prevail in these areas [46].
Clinical outcomes — Initial studies suggested that precore and core promoter variants were associated with more severe liver disease and HCC [13,15,60-63]. However, these variants have also been detected in inactive carriers and patients with minimal liver disease [41,45,53,57]. Furthermore, precore and core promoter variants had no impact on pre- and post-liver transplant outcomes in a study from the United States [64]. It is possible that host immune response and mutations in other regions of the HBV genome may be involved in the pathogenesis of liver disease [65].
There are increasing data to support an association between core promoter variants and more severe liver disease. A study from Hong Kong found patients with core promoter mutation tended to have higher hepatic necroinflammation scores [66]. The same group reported the prevalence of core promoter mutations was higher among chronic hepatitis B patients who developed complications of cirrhosis and HCC [67,68].
Another study from Taiwan involving 250 chronic hepatitis B-infected patients found that the prevalence of core promoter variants increased from 3 percent among inactive carriers to 64 percent among patients with HCC [69]. When controlling for other potential confounders such as sex, age, genotype, and presence of cirrhosis, the core promoter variants (A1762T, G1764A) were still significantly associated with the development of HCC (odds ratio [OR] 10.6). A longitudinal study from Taiwan demonstrated that persistence of core promoter mutations accumulated over time before the diagnosis of HCC [70]. Another study from the United States reported that patients with core promoter variants were more likely to have hepatic decompensation [71].
The major drawback of the above studies is that the conclusions were based on cross-sectional analyses. As a result, large-scale longitudinal studies with consideration of other host and viral confounders are needed to confirm the relationship between core promoter variants and advanced liver disease. Several studies have confirmed an association between basal core promoter mutations and hepatocellular carcinoma; in a few studies, there has been an association with cirrhosis as well [72-75]. One study included 2762 Taiwanese patients with chronic HBV followed between 1991 and 2004 (for a total of 33,747 person years), during which a total of 153 incident cases of HCC were detected [76]. After adjusting for baseline HBV DNA levels, sex, and age of recruitment, the basal core promoter variant A1762T/G1764A and genotype C were predictors of increased risk of HCC, while the precore variant G1896A was associated with a lower risk of HCC [73].
Fulminant hepatitis — Precore and core promoter variants have been described in association with fulminant hepatitis [7,77-80]. It has been suggested that these variants result in a fulminant course because of enhanced HBV replication or a more aggressive immune response. However, in some studies, these variants were also detected in asymptomatic source patients who led to the outbreak of fulminant hepatitis B [81]. The mechanism by which precore variants cause fulminant hepatitis in the new host but an inactive liver disease in the original host is unclear. It is possible that de novo infection with precore variants elicits a more aggressive immune response while precore variants that are selected during the course of infection may escape immune clearance and are associated with less active disease.
Serum quantitative HBV DNA levels — There is a paucity of data on the relationship between precore/core promoter variants and serum quantitative HBV DNA levels. The most comprehensive study was done in the United States and involved a total of 694 consecutive chronic HBV-infected patients [71]. HBeAg-positive patients with core promoter variant had significantly lower serum HBV DNA levels compared to those with wild-type sequence in the core promoter region. A possible explanation was the selection of core promoter variants during the process of immune clearance. By contrast, precore and/or core promoter variants were associated with higher serum HBV DNA levels than wild-type sequence in HBeAg-negative patients, which may be because precore/core promoter variants can evade the host immune response and can thus maintain higher levels of HBV replication. This observation has been confirmed by subsequent studies [31,82,83].
Interferon therapy — Several studies have found that interferon (IFN) therapy is associated with a low rate of sustained response among patients with HBeAg-negative CHB [84,85]. These studies suggested that a longer duration of treatment (>12 months) was necessary to decrease the rate of post-treatment relapse [86].
Relatively few studies have evaluated the relationship between precore and core promoter mutations and response to IFN therapy and results have been conflicting [43,44,87-90].
●HBeAg-positive patients: Some studies [43,44] found that patients who were HBeAg positive with precore variant were more likely to achieve HBeAg loss/seroconversion with IFN therapy while disparate results have been reported by others [87-89,91,92]. Whether the conflicting results could be accounted for by differences in HBV genotypes or baseline HBV DNA levels is unknown.
●HBeAg-negative patients: One study from Italy showed that patients with exclusively or predominantly precore variant had a higher rate of relapse after initial response to IFN [84]. A study from Germany found that responders to IFN therapy had significantly fewer mutations in the basic core promoter region compared with nonresponders [88].
Nucleoside analog therapy — Most patients with HBeAg-negative CHB relapse after one year of nucleoside analog treatment (see appropriate topic reviews). As with IFN, longer duration of treatment is needed to maintain viral suppression.
There is very little information on the correlation between precore and core promoter mutations and response or resistance during nucleoside analog therapy. Clinical studies of lamivudine, adefovir, entecavir, telbivudine, and tenofovir show that nucleoside analogs have similar potency in HBeAg-positive and in HBeAg-negative patients. (See appropriate topic reviews.)
Summary — Interpretation of the biologic and pathogenic significance of precore and core promoter variants should be made cautiously because many factors such as HBV genotype, viral quasispecies, mutations in other regions of the HBV genome, host immune response, and sensitivity of detection methods can influence the results. Testing precore and core promoter variants especially in patients suspected to have HBeAg-negative CHB can help further characterize the patient but these tests are not routinely indicated for clinical management because the precise role of these variants in predicting the severity of liver disease or response to antiviral therapy remain unclear.
PRE-S/S VARIANTS — The major B-cell neutralizing epitopes of HBsAg reside in the "a" determinant region located at amino acid positions 124 to 149 [93,94]. Antibodies to the "a" determinant confer protection against all serotypes of HBV [95]. The pre-S and S regions of the HBV genome also contain putative HLA class I-restricted cytotoxic T lymphocyte (CTL) epitopes [96]. Mutations involving T- or B-cell epitopes in the pre-S and S regions may result in immune escape mutants [97]. In addition, mutations in the pre-S region may affect the secretion of small S protein. Clinically significant pre-S/S variants will be discussed below.
Vaccine-escape mutants — HBV S gene mutants have been described in infants who were infected with hepatitis B despite an adequate anti-HBs response to hepatitis B vaccination. These mutations have been observed in many parts of the world, including Italy, Singapore, Taiwan, Japan, China, and Africa. The most common mutation involves glycine to arginine substitution at codon 145 (G145R) in the "a" determinant of HBsAg. In vitro studies confirmed that HBsAg with G145R mutation had decreased binding to hepatitis B surface antibodies [98]. Other mutations involving amino acids 126, 129, and 141 have also been reported among vaccinees but the clinical significance of these mutants is not well documented.
Most reports found that the vaccine-related HBV S mutants were not detected in the maternal carriers, suggesting that the mutations were selected by vaccine and/or hepatitis B immune globulin (HBIG). However, a few reports suggested that these mutants were transmitted from the mothers to the babies. One study reported horizontal transmission of G145R mutation to intrafamilial members [99].
A study from Taiwan [100] found that the prevalence of HBV S mutants among HBsAg-positive children increased after the introduction of universal vaccination, from 7.8 percent in 1984 to 25 percent in 1994, and about 23 percent in 1999. However, the carrier rate among Taiwanese children decreased from 9.8 to 0.7 percent in the same period. A subsequent update that included data from 2009 found no increase in breakthrough HBV infection of HBV S mutants over the 25-year period since the initiation of universal vaccination in Taiwan [101]. These data suggest the emergence of vaccine escape mutants is rare and has not diminished the efficacy of HBV vaccines. (See "Hepatitis B virus immunization in adults".)
Immunoprophylaxis failure in orthotopic liver transplant recipients — HBV reinfection in patients receiving maintenance HBIG immunoprophylaxis may be due to inadequate dosing or to the emergence of mutant virus that can escape neutralization. Several studies reported the detection of S mutants in patients who developed recurrent hepatitis B after liver transplantation despite HBIG prophylaxis. As with vaccine failure, G145R is the most common mutation. One study found significant correlation between these mutants and the duration of HBIG therapy and reversion of the mutation to wild-type sequence after withdrawal of HBIG. The use of nucleos(t)ide analogs with potent antiviral activity and a high barrier to resistance prior to and after liver transplant has greatly reduced the need for HBIG, the rate of HBV reinfection post-liver transplant, and the emergence of HBV S mutants. (See "Liver transplantation in adults: Preventing hepatitis B virus infection in liver transplant recipients".)
Viral persistence in HBsAg-negative patients — Loss of HBsAg usually indicates clearance of HBV. However, low levels of HBV DNA can be detected in serum and liver from some HBsAg-negative individuals using PCR assays [102]. In most instances of occult HBV infection (HBsAg negative but HBV DNA positive by PCR), HBV DNA is present in very low concentrations with sub-detectable amounts of HBsAg. However, in rare instances, mutations in the HBV genome down-regulate S gene expression or produce aberrant surface proteins. As a result, HBsAg is not detectable by conventional serologic assays [103-106]. Occult HBV infection may have an impact on blood transfusion, organ transplantation, diagnosis of cryptogenic liver disease, and hepatocellular carcinoma (HCC) development [107,108]. (See "Hepatitis B virus: Screening and diagnosis in adults".)
POLYMERASE GENE MUTATIONS — The HBV DNA polymerase gene consists of four distinct regions [109]:
●A primer involved in priming of reverse transcription
●A spacer with no known function
●A reverse transcriptase/DNA polymerase which is responsible for reverse transcription of the pregenomic RNA into the first (-) strand HBV DNA and for synthesis of the second (+) strand HBV DNA
●An RNAse H, which removes the RNA template
The reverse transcriptase/DNA polymerase region has five conserved domains: A, B, C, D, E. Domains A, C, and D are involved in nucleoside triphosphate binding and catalysis, while domains B and E participate in the positioning of the RNA template and the primer, relative to the catalytic site [110,111]. The putative catalytic domain is believed to reside in the tyrosine-methionine-aspartate-aspartate (YMDD) locus in domain C. This locus is conserved in all viral reverse transcriptases as well as in all isolates of hepadnaviruses [109]. Polymerase gene mutations that are associated with antiviral resistance can arise spontaneously but are rarely detected unless they have been selected by exposure to oral nucleos(t)ide analogs. (See "Hepatitis B virus: Overview of management", section on 'Nucleos(t)ide analogs'.)
X GENE VARIANTS — The X region is crucial for the replication and expression of HBV because the X protein regulates HBV replication through activating and regulating viral and cellular genes [112,113]. The X region overlaps the core promoter, enhancer II, and two direct repeats that are involved in regulating precore and pregenomic mRNA transcription. HBx protein is believed to be important in hepatic carcinogenesis. Several studies reported that mutations affecting codons 130 and 131 of the X gene are more common in patients with hepatocellular carcinoma (HCC) than in patients with chronic hepatitis B [62,63,114]. These mutations correspond to the dual core promoter mutations A1762T and G1764A.
Although many studies have shown a convincing relationship between core promoter variants and HCC [68-70], it is unclear if core promoter variant is directly involved in the process of carcinogenesis or the increased risk of HCC is related to more active and rapidly progressive liver disease. Because of the association between core promoter variants and HBV genotype C, and more active hepatic necroinflammation, careful analysis is needed to determine if core promoter mutation is independently related to HCC development. A prospective cohort study from Taiwan found that core promoter but not precore mutation was associated with an increased risk of HCC and the association was independent of HBV genotype [76]. However, more studies are needed to clarify these associations. Whether X gene mutations that are unrelated to the common dual core promoter mutation increase the risk of HCC also needs further clarifications.
SUMMARY
●Impact of variants – Because chronic hepatitis B virus (HBV) infection frequently persists for decades, many variants may exist within the same host at any given time, and each variant may have multiple base changes. Some mutations have a role in viral latency, pathogenesis of liver disease, immune escape, and resistance to antiviral therapy. (See 'Introduction' above.)
●Types of variants
•Precore/core region variants – The precore/core region of the HBV genome encodes the nucleocapsid protein (HBcAg) and hepatitis B e-antigen (HBeAg). Initial studies suggested that precore and core promoter variants were associated with more severe liver disease. However, these variants have also been detected in inactive carriers and patients with minimal liver disease. There are increasing data to support an association between core promoter variants and more severe liver disease, including hepatocellular carcinoma (HCC). (See 'Precore and core promoter variants' above.)
•The pre-S and S region variants – The pre-S and S regions of the HBV genome contain putative HLA class I-restricted cytotoxic T lymphocyte (CTL) epitope. Several studies reported the detection of S mutants in patients who developed recurrent hepatitis B after liver transplantation despite hepatitis B immune globulin (HBIG) prophylaxis, or in children born to HBsAg carrier mothers who developed HBV infection despite receiving HBIG and HBV vaccination. (See 'Immunoprophylaxis failure in orthotopic liver transplant recipients' above.)
•Polymerase gene mutations – Polymerase gene mutations can lead to development of resistance against antiviral agents. They can arise spontaneously but are rarely detected unless they have been selected by exposure to oral nucleos(t)ide analogs. (See 'Polymerase gene mutations' above.)
•X gene variants – The X region is crucial for the replication and expression of HBV because the X protein regulates HBV replication through activating and regulating viral and cellular genes. Several studies reported that mutations affecting codons 130 and 131 of the X gene are more common in patients with HCC than in patients without HCC. These X gene mutations correspond to core promoter mutations that had been reported to be associated with an increased risk of HCC. (See 'X gene variants' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Dr. Chi-Jen Chu, who contributed to an earlier version of this topic review.
Do you want to add Medilib to your home screen?