Review Molecular genesis of drug-resistant and vaccineescape

Transcription

1 Antiviral Therapy : (doi: /IM1499) Review Molecular genesis of drug-resistant and vaccineescape HBV mutants tephen A Locarnini 1 *, Lilly Yuen 1 1 Victorian Infectious Diseases Reference Laboratory, North Melbourne, Australia *Corresponding author tephen.locarnini@mh.org.au A high rate of viral turnover, combined with an errorprone polymerase, results in a very high frequency of mutational events during HBV replication. Not surprisingly, particular selection pressures, both endogenous (host immune clearance) and exogenous (vaccines and antivirals), readily select out new escape mutants. The introduction of nucleoside/nucleotide analogue (NA) therapy for chronic hepatitis B has witnessed the emergence of antiviral drug resistance as the major factor limiting drug efficacy. Furthermore, because of the overlap of the viral polymerase and envelope reading frames in the HBV DNA genome, NA resistance-associated mutations selected in the catalytic domains of the polymerase frequently result in important changes to the neutralizing antibody- binding domains of the hepatitis B surface antigen, including the emergence of antiviral drug-associated potential vaccine escape mutants (ADAVEMs). The public health significance of ADAVEMs is considerable in terms of the global programme for control of hepatitis B via universal infant immunization. Thus, prevention of resistance requires the adoption of strategies that not only effectively control HBV replication, but also prevent the emergence of ADAVEMs. Molecular virology of HBV HBV is a small circular DNA virus and is a member of the Hepadnaviridae family [1]. The intact virion comprises the DNA genome and an endogenous viral polymerase surrounded by structural nucleocapsid and surface proteins. The transcriptional transactivator and immune regulator proteins, as well as the structural and polymerase proteins, are encoded for in four overlapping open reading frames (ORFs) in the compact 3.2-kb partially double-stranded DNA genome. These four are the core/precore, polymerase, envelope and X ORFs. The core/precore ORF encodes both the core protein (hepatitis B core antigen), which forms the nucleocapsid, and the precore protein, also known as the hepatitis B e antigen (HBeAg), which is involved in immune evasion. The polymerase ORF encodes the viral polymerase protein, which contains a reverse transcriptase (rt) that is essential for replication. The envelope ORF encodes the pre-1, pre-2 and proteins (collectively, hepatitis B surface antigen [HBsAg]), which form the mature virion envelope. These proteins are also known as large (L), middle (M) and small () HBsAg, respectively. Finally, the X ORF encodes the X protein (hepatitis B X antigen), a transcriptional transactivator [1]. Viral entry into hepatocytes is thought to be receptormediated, although the receptors involved are yet to be identified. Following virion entry the surface proteins are shed, releasing the nucleocapsid, which is then translocated to the nucleus, with the genome entering the nucleus through the nuclear pores. The partially double- stranded DNA genome is repaired in the nucleus to a highly stable covalently closed circular DNA molecule, which is the major transcriptional template and exists as a viral minichromosome from which pregenomic RNA, precore RNA, HBx and HBsAg messenger RNAs (mrnas) are made. The pregenomic mrna is the template for both the polymerase and hepatitis B core antigen. Following translation, the HBV polymerase mediates reverse transcription of the pregenomic mrna transcript. Essentially, the HBV polymerase interacts with epsilon, a cis acting signal containing a stem loop structure on the RNA transcript prior to its encapsidation and subsequent reverse transcription within the nucleocapsid [1]. The surface proteins are synthesized and processed in the endoplasmic reticulum. Following full genome replication, the nucleocapsid translocates out of the cytosol and is packaged into mature virions that then exit the host cell via the exocytosis pathway International Medical ress (print) (online) 451

2 A Locarnini & L Yuen Figure 1. rimary resistance mutations associated with NA therapy and their location in the pol domains Terminal protein pacer Reverse transcriptase RNase H (rt1) 692 (rt344) 845 aa F V LLAQ YMDD I(G) II(F) A B C D E 3TC resistance rta181t/v rtm204v/i LdT resistance rta181t/v rtm204i ADV/TDF resistance rta181t/v rtn236t ETV resistance rti169t rtl180m rtt184/a/i/l/g/c/m rt202c/g/i rtm204v rtm250i/v The primary nucleoside/nucleotide analogue (NA) resistance-associated mutations for the l-nucleosides (lamivudine [3TC] and telbivudine [LdT]), the acyclic phosphonates (adefovir [ADV] and tenofovir disoproxil fumarate [TDF]) and the d-cyclopentane group (entecavir [ETV]). Length of the viral polymerase (845 amino acids [aa]) is based on HBV genotype A. Antiviral drug resistance and chronic hepatitis B In many parts of the world, five HBV-specific nucleoside/nucleotide analogues (NAs) targeting the viral polymerase are approved for the treatment of chronic hepatitis B (CHB): lamivudine (3TC; a cytidine l- nucleoside analogue), adefovir (ADV; an alkyl phosphonate), entecavir (ETV; d-cyclopentane), telbivudine (LdT; a thymidine l-nucleoside analogue) and tenofovir disoproxil fumarate (TDF; an alkyl phosphonate very similar to ADV). Frequently used as monotherapy, most of these NAs are associated with the evolution of antiviral drug resistance mutations in the rt domain of the HBV polymerase gene (pol). For 3TC, the prevalence of resistance rises rapidly, reaching 14% after 1 year [2] and 80% after 4 years [3] of monotherapy. Resistance develops less rapidly with ADV (5.9% over 144 weeks) [4] and ETV (<1% after 3 years but 5.8% after 1 year in 3TC-treated patients), and there is evidence that previous 3TC therapy accelerates ETV resistance [5,6]. Antiviral drug resistance to these NAs is associated with the selection of primary resistance mutations in the pol/rt-coding domains, which facilitate ongoing viral DNA synthesis in the presence of the drug. Continued therapy in the setting of drug failure leads to the accumulation of compensatory (or secondary) mutations that restore viral fitness. With 3TC, the primary amino acid substitutions, rtm204v/i, arise in the YMDD motif of pol, and the compensatory changes include rtl180m [7,8] and rtv173l [9]. ADV resistance is characterized by rtn236t and/or rta181v/t [10,11]. ETV resistance tends to emerge in a stepwise manner with the rt202i change occurring sequentially in viruses already bearing 3TC resistance (rtl180m plus rtm204v) mutations [6,12], although primary ETV resistance has been reported where these mutations have appeared simultaneously [13]. For LdT, the year 2 resistance rates were 29% in HBeAgpositive and 11% in HBeAg-negative patients with the primary resistance mutation rtm204i. econdary mutations rtl80i/v and rtl80i/v plus rtl180m can accompany this signature mutation in approximately 2 5% of cases [14,15]. No definitive TDF-associated mutations have been described in treatment-naive patients taking TDF for up to week 72 [16,17], but the main ADV- associated resistance mutations, rta181t/v and rtn236t, do confer reduced sensitivity both clinically [18] and virologically [10]. The full list of primary resistance mutations associated with NA therapy and their location in the pol/rt domains are listed in Figure 1. To date, there are eight codons associated with primary antiviral drug resistance in CHB, which map to four of the functional domains of pol [19]: B- domain at codons rti169, rtl180, rta181 and rtt184; C- domain at codons rt202 and rtm204; D-domain at codon rtn236; and E-domain at codon rtm250. Recent bioinformatic studies [20] have shown that these eight codons are associated with four major International Medical ress

3 Molecular genesis of HBV mutants pathways of antiviral resistance: the l- nucleoside pathway, which is identified by the NA resistanceassociated mutation, rtm204v/i, resulting in resistance to 3TC and LdT; the acyclic/alkyl phosphonate sugar pathway, which is identified by the rtn236t substitution, resulting in resistance to ADV and reduced sensitivity to TDF; the shared pathway, which is identified by the mutation in pol of rta181t/v and results in resistance to both the l-nucleosides (3TC and LdT) as well as ADV, and reduced sensitivity to TDF; and the treatment-naive ETV resistance pathway, which is identified by detection of the mutations rtl180m plus rtm204v/i plus at least one substitution in one of the rtt184, rt202 and rtm250 amino acid positions. Finally, various combinations and permutations of these four major pathways have been described in patients with multidrug-resistant HBV, such as rta181t plus rtn236t plus rtm250l [21]. Causes of antiviral drug resistance Antiviral drug resistance reflects the reduced susceptibility of a virus to the inhibitory effect of a drug [22] and results from a process of adaptive mutations under NA therapy. ix important factors combine that account for the timing and pattern of resistance: HBV viral replication rates, low fidelity of pol, selective pressure of the drug, the genetic barrier of the drug, role of replication space (hepatocyte turnover) and fitness of the resistance mutant. In a chronically infected individual, the extent of HBV replication is considerable, reaching >10 12 virions per day. As pol is an rt that lacks proof-reading capacity, HBV replication is also associated with a high mutational rate of 10-5 substitutions/base/cycle. Thus, all possible single-base changes in the 3,200-nucleotide genome are generated every day, thereby accounting for the observation that single and double mutations associated with NA resistance existing in patients prior to therapy, whilst triple or quadruple mutations require replication in the presence of NA-imposed selection pressure and so rarely pre-exist. Thus, it is not surprising that NAs that have low potency and a low genetic barrier (one or two primary resistance mutations) to resistance (for example, 3TC and LdT), when used as monotherapy will fail quickly, whereas NAs that are potent (for example, TDF and ETV) and have a high genetic barrier (for example, ETV with at least three mutations) to resistance are associated with extremely low resistance rates after 3 5 years of monotherapy [13]. Cross-resistance Cross-resistance is defined as resistance to drug(s) that a virus has never been exposed to. The drug resistanceassociated mutations selected by particular groups of NA (for example, l-nucleosides, acyclic phosphonates or cyclopentanes) could diminish the antiviral activity of other drugs [23]. This should be considered before any antiviral drug is prescribed and the physician should plan for eventual treatment failure. The initial selection and subsequent rescue therapies should be based on knowledge of cross-resistance [23], so that the second agent used will not induce cross-resistance with the failing agent [24]. referably by using the add-on/combination approach [24]. The advantage of using combinations of NA with complementary cross- resistance profiles has recently been highlighted [23] and a summary of crossresistance profiles based on the viral resistance pathways approach is shown in Table 1. Table 1. Cross-resistance analysis for the nucleoside/nucleotide analogues approved for chronic hepatitis B Resistance mutation a Lamivudine/telbivudineresistant (L180M with or Adefovir-resistant Adefovir-resistant Entecavir- Telbivudine- Mutation effect without M204V/I) (N236T) (A181T/V) resistant b resistant c Confers complete Telbivudine Lamivudine, Lamivudine resistance telbivudine Confers some degree Entecavir Tenofovir disoproxil Lamivudine, telbivudine, Entecavir of reduced sensitivity fumarate tenofovir disoproxil to listed drugs fumarate [18,23] d Drugs remaining Adefovir, tenofovir Entecavir, lamivudine, Entecavir Adefovir, tenofovir Adefovir, tenofovir fully active [23,61] disoproxil fumarate telbivudine disoproxil fumarate disoproxil fumarate everal drug resistance mutations affect sensitivity to other drugs in laboratory analyses, but their effect on treatment outcomes is unclear. a First virological breakthrough should be managed with an add-on strategy (combination), instead of switch (sequential) monotherapy. b Entecavir-resistant mutations are rtm204i/v with or without rtl180m, plus one or more substitutions at rtt184, rt202 or rtm250, with or without an rti169 substitution in the HBV polymerase. c Telbivudine-resistant mutations are rtm204i or rtl180m plus rtm204v. d The effect on sensitivity is variable as results are according to laboratory analyses and not patient studies. Tenofovir disoproxil fumarate resistance was not determined because of treatment intensification after week 72 [16,17]. Antiviral Therapy 15:3 t B 453

4 A Locarnini & L Yuen Figure 2. olymerase and envelope link of HBV: relevance to antiviral drug resistance and vaccine escape Env re-1 re- 2 a determinant Terminal protein pacer GFA B CDE RNase H ol rt domains Envelope mutants olymerase mutants a Ag Ab binding IC 50, µg/ml Wild type Wild type 1.09 HBIG escape sg145r rtw153q >55.0 sd144e/g145r rtg153e >55.0 Drug-resistant sf158y rtf166l 1.86 se164d rtv173l sw196 rtm204i 8.29 si195m rtm204v 5.26 sm198i rtv207i 12.5 se164d/i195m rtv173/rtl180m/rtm204v Adapted from [35]. a rtl180m in polymerase (ol) causes no change in envelope (Env). Ab, antibody; Ag, antigen; HBIG, hepatitis B immunoglobulin; IC 50, 50% inhibitory concentration; rt, reverse transcriptase. Consequences of antiviral drug resistance in chronic hepatitis B The consequences of emergence of NA-resistant HBV in treated patients include virological breakthrough and rebound in viral load, biochemical breakthrough including hepatic flares, histological progression of liver disease, clinical manifestations including hepatic decompensation, increased viral recurrence post liver transplantation and even possible increased tumourigenicity [25]. everal public health issues have been identified including an alteration in the HBsAg antigenicity [26], transmission of drug-resistant HBV [27,28] and a possible threat to the various immunization programmes designed to control hepatitis B [29]. This includes the emergence of antiviral drug-associated potential vaccineescape mutants (ADAVEMs) [27]. The remainder of this article will focus on these public health issues. ublic health significance of the pol/ gene overlap The envelope ORF of HBV encodes three co-carboxylterminal proteins, the shortest of which is, the principal component of the external envelope of the 42 nm virion. The protein carries the major target of neutralizing antibody, the a determinant. Although it is well known that resistance mutations result in nonsynonymous changes in the of HBsAg [30 34], the first study to demonstrate that these substitutions might affect HBsAg protein conformation, and more importantly its antigenicity, was by Torresi et al. [35]. These investigators showed that the common 3TCresistant substitutions (rtm204v/si195m, rtm204i/ sw196, rtm204i/sw196l and rtv173l/se164d plus rtl180m plus rtm204v/si195m) resulted in reductions in the reactivity of the altered HBsAg with vaccine-induced antibody against HBsAg (anti-hbs). In addition, the converse has been shown in that changes in the gene introduce changes in pol that correspond with 3TC- resistant compensatory-type mutations [36]. These studies have now been independently confirmed using different in vitro models, including mammalian cell culture transfection and epitope density mapping [37]. These latter studies used fully glycosylated HBsAg recombinant proteins, whereas the study of Torresi et al. [35] used yeast-expressed recombinant HBsAg proteins that were not glycosylated. The key findings are summarized in Figure 2. These in vitro studies have been now extended to in vivo analyses, in which the common 3TC-resistant mutation, rtv173l plus rtl180m plus rtm204v [9], which displays the se164d plus si195m change in HBsAg, successfully infected hepatitis B-immunized International Medical ress

5 Molecular genesis of HBV mutants chimpanzees that carried high titres of circulating anti- HBs pre-challenge [27]. This chimpanzee study also established the genetic stability of the rtv173l plus rtl180m plus rtm204v variant in a non-immunized chimpanzee, in whom no revertants to wild type (WT) were detected over time compared with infection with the sg145r vaccine-escape mutant, which quickly back-reverted to WT [27]. This latter observation reveals the important role of compensatory mutations in fixing the genetic archive, especially in the setting of transmission of NA resistance. Hepatitis B immunization programme ince its widespread introduction, the hepatitis B vaccine has become an essential part of global infant immunization programmes. The vaccine has been particularly important for countries where the incidence of HBVrelated hepatocellular carcinoma (HCC) is high. The hepatitis B vaccine is an effective means of preventing HBV infection, producing protective levels of antibodies in up to 95% of recipients [38]. The current commercially available hepatitis B vaccine used in most programmes is a yeast-derived recombinant HBsAg, containing only the small protein. As discussed above, neutralizing antibodies (anti-hbs) elicited by hepatitis B vaccines specifically target the a determinant of HBsAg. It has been recognized that the administration of hepatitis B vaccine can result in the selection of antibody escape variants such as sg145r [39]. In high-prevalence countries, such as China, Thailand and Taiwan, longterm monitoring has shown that hepatitis B immunization programmes have increased the incidence of HBV variants with a number of signature substitutions associated with vaccine failure in the HBsAg protein [39,40], although the vaccine programme reduces the overall burden of hepatitis B [41]. Mutations in and around the a determinant might lead to an alteration in the antigenicity of the HBsAg protein so that anti- HBs might fail to neutralize the virus [42 50]. Infection of immunized individuals with these vaccine-escape mutants (VEMs) [45] is therefore possible. VEMs are typically characterized by the presence of particular single amino acid changes in the protein, for example, sg145r [46,48]. ome of these variants have been found in individuals for at least a decade, suggesting they are stable and potentially transmissible variants of HBV. In Taiwan, up to 28% of children with CHB also harbour HBsAg mutants. Not surprisingly then, VEMs capable of causing infection in fully immunized individuals are common in countries with high rates of endemic HBV infection and universal hepatitis B infant immunization programmes [45,48]. To date, the emergence of VEMs has not had a known negative effect on the immunization programme of any particular country [51], but many of these programmes do not include the use of a birth-dose of hepatitis B immunoglobulin, which is known to accelerate VEM selection [44,45,47,51]. In those populations where 3TC has been widely used to treat patients continuously for periods of several years, viruses with alterations in HBsAg are likely to occur relatively frequently, and some will have the potential to be ADAVEMs [29]. The emergence of such ADAVEMs in 3TC-treated patients does not readily translate to a significant imminent threat to the global hepatitis B immunization programme [29]. For a new viral species to pose a threat in this setting, Clements et al. [29] have proposed that an ADAVEM would need to possess the following four characteristics: it must be a stable mutant; it must have undergone sufficient changes in antigenicity such that anti-hbs generated by the current vaccine no longer neutralizes it; it must be transmissible and cause infection in immunized individuals, and so have the opportunity for ongoing spread; and it must cause disease (acute or chronic) in infected individuals. Of these four characteristics, there is evidence to date that the first three have been acquired; it is not known if ADAVEMs have the same propensity to cause disease as do current circulating strains of HBV [29], although one case of primary infection with a 3TC-resistant HBV was associated with acute hepatitis [28]. Clearly, further studies are needed to fully elucidate the clinical, pathological and epidemiological significance of these emerging ADAVEMs. Evolution of HBsAg during antiviral therapy In order to study the effects of long-term antiviral therapy on the HBsAg of HBV, we have performed an analysis of the pol/rt sequences and, therefore, the gene, obtained from serum samples of 2,136 patients with either HBV monoinfection or HIV HBV coinfection that were submitted for drug resistance testing from 1991 to 2009 to the Victorian Infectious Diseases Reference Laboratory, North Melbourne, Australia. These sequences have been analysed using the eqhepb programme [52]. A function of this programme is to identify and differentiate polymorphic and reportable amino acid substitutions within pol and gene, taking into account the HBV genotypes. Based on whether the pol/rt sequences have amino acid substitutions at any of the eight rt codons (Figure 1) that are associated with drug resistance (rt169, rt180, rt181, rt184, rt202, rt204, rt236 and/or rt250), the patients were categorized as either having drug-resistant (DR; n=737) or drug-susceptible (D; n=1,399) HBV infections. The frequency of reportable amino acid substitutions within the HBsAg for these two types of HBV isolates (DR or D for the total cohort) is shown in Figure 3. As highlighted in this figure, a number of HBsAg codons in the DR isolates have significantly higher rates of amino acid substitutions compared with equivalent codons in Antiviral Therapy 15:3 t B 455

6 A Locarnini & L Yuen the D isolates (<0.05; determined by Fisher s exact and Holm s method of -value correction). The HBsAg codons were s47 (rt55), s75 (rt84), s134 (rt142), s164 (rt173), s172 (rt181), s173 (rt181), s175 (rt184), s192 (rt200), s195 (rt204), s196 (rt204), s199 (rt207), s220 (rt229) and s221 (rt229; Figure 3). Importantly, no HBsAg codons were identified by statistical analysis to have higher rates of reportable amino acid substitutions in the D isolates compared with equivalent regions in the DR isolates. Figure 3 also includes a plot of the location of the rt domains, the major B-cell epitopes of HBsAg and the CTL-T helper epitopes. The patients were then further categorized into HBV monoinfections (D n=1,229 and DR n=615) or HIV HBV coinfection (D n=181 and DR n=111). Interestingly, the same HBsAg codons as seen in the total cohort was found to have higher rates of reportable amino acid substitutions in the DR isolates of patients with HBV monoinfections, and these were statistically significant (<0.05). Only the HBsAg codons s164 (rt173), s195 (rt204) and s196 (rt204) showed similar differences between the DR and D isolates of patients coinfected with HIV and HBV. This difference in the number of HBsAg codons observed between the two patient groups might be explained by the small number of patients with HIV HBV coinfection in the study. Nonetheless, it was interesting to note that the rates of amino acid substitutions in s164 and s195 were substantially higher in the DR isolates of HIV HBV- coinfected patients (39% for s164 and 96% for s195) compared with HBV-monoinfected patients (7% for s164 and 40% for s195), whereas the converse was observed for s196 (coinfected 13% and monoinfected 46%). These observations reflect the higher rate of detection for the rtv173l (se164d) plus rtl180m (silent in ) plus rtm204v (si195m) 3TC resistance mutational profile in coinfected patients compared with the rtm204i (sw196stop//l) profile. The ADAVEM potential of rtv173l plus rtl180m plus rtm204v (se164d plus si195m) has already been discussed above and is shown in Figure 2. Figure 3. The HBsAg amino acid mutation frequency from the eqhepb database from 1991 to rt F A B C D B-cell epitopes a Determinant M 1 2 CTL & T-helper Epitopes HLA class I HLA class II HLA class I HLA class I HLA class I 50 s Resistant isolates (n=737) ensitive isolates (n=1,399) s196 roportion, % s47 s s75 s134 s173 s172 s175 s192 s199 s220 s221 0 s1 s3 s5 s7 s9 s11 s13 s15 s17 s19 s21 s23 s25 s27 s29 s31 s33 s35 s37 s39 s41 s43 s45 s47 s49 s51 s53 s55 s57 s59 s61 s63 s65 s67 s69 s71 s73 s75 s77 s79 s81 s83 s85 s87 s89 s91 s93 s95 s97 s99 s101 s103 s105 s107 s109 s111 s113 s115 s117 s119 s121 s123 s125 s127 s129 s131 s133 s135 s137 s139 s141 s143 s145 s147 s149 s151 s153 s155 s157 s159 s161 s163 s165 s167 s169 s171 s173 s175 s177 s179 s181 s183 s185 s187 s189 s191 s193 s195 s197 s199 s201 s203 s205 s207 s209 s211 s213 s215 s217 s219 s221 s223 s225 HBsAg codon positions Data from [52]. Frequency of reportable amino acid substitutions within each hepatitits B surface antigen (HBsAg) codon position (n=2,136). The length of HBsAg varied between sequences, and this has been taken into account during proportion calculation. The locations of the functional domains in the overlapping pol are shown, coplotted with the major B-cell and T-cell epitopes. CTL, cytotoxic T-lymphocyte; HLA, human leukocyte antigen; M, mini-loop; rt, reverse transcriptase; 1, loop 1; 2, loop International Medical ress

7 Molecular genesis of HBV mutants Of the reportable amino acid substitutions detected in HBsAg codons that had significantly higher substitution rates in DR isolates than in D isolates, not all were the result of an overlap with the known drug resistance-associated mutations within pol. Nonetheless, most of them are located in a specific rt catalytic domain rather than an interdomain [19]: s75 in domain A; s164, s172 and s173 in domain B; and s192, s195, s196 and s199 in domain C. It is possible that the rt of DR HBV isolates is under greater positive selective pressure than the D isolates. In the context of envelope functional domains, s134 (loop 1) and s164 (downstream of loop 2) are both located in the a determinant of HBsAg [42,48]. Importantly, the common 3TC-associated HBsAg mutational profile (rtv173l/se164d plus rtl180m plus rtm204v/ si195m) has already been shown to cause a reduction in the reactivity of HBsAg with antihbs in vitro [35] and in vivo [27]. Finally, s175 is located in an important HLA class-i-restricted epitope initially described by Chisari and Ferrari [53] and classified in their paper as ENV (assigned s175 s184 in Figure 3), implying an immunological rather than an NA-driven selection pressure. As well as non-synonymous substitutions, antiviral therapy also resulted in HBsAg truncation that occurred more frequently in HBV isolates that are DR compared with those that are D; in particular, the presence of stop codons in the C-terminal (s172, s196 and s199) end of HBsAg (Figure 4). It is noteworthy that the occurrence of sw182stop (rtv191i) was substantially higher in HBV isolated from HIV HBV-coinfected patients. Investigation into the clinical effects and significance of these C-terminal truncations is ongoing. Molecular pathogenesis and oncogenic potential everal HBV proteins are involved in the development of HCC, transcribed from either integrated HBV DNA or the HBV genome. The envelope ORF encodes three coterminal surface proteins: L, translated from pre-1 plus pre-2 plus genes; M, translated from pre-2 plus genes; and, from the gene. re-2/ genes truncated at the 3 end (C-terminal) have been isolated from integrated HBV DNA sequences in HCC [54,55]. The L and M proteins with C-terminal truncations have transcriptional transactivation potential, a function not exhibited by the full-length forms. Truncated HBV surface proteins have been implicated Figure 4. Frequency of HBsAg stop codon mutations in HBV-monoinfected and HIV HBV-coinfected patients 5 4 Monoinfected sensitive group Monoinfected resistant group Coinfected sensitive group Coinfected resistant group roportion, % s21 s53 s61 s69 s74 s76 s94 s104 s159 s172 s182 s191 s196 s199 s201 s213 s216 s220 HBsAg codon positions The hepatitis B surface antigen (HBsAg) stop codon mutation frequency in HBV-monoinfected patients (n=1,844) was compared with HIV HBV-coinfected patients (n=292) in either the nucleoside/nucleotide drug-sensitive (D) or drug-resistant (DR) group. The length of HBsAg varied between sequences, and this has been taken into account during proportion calculation. Antiviral Therapy 15:3 t B 457

8 A Locarnini & L Yuen Figure 5. HBsAg mutations selected during NA therapy Mini-loop 120T C K/R T C 121-s-s-124 T/M G T/I/A T/V Loop 1 of a determinant /C A/V T Q 161 L T A F/L A/G K/R W G F/Y /T N/T L 99 /T D G W 164 D M Y E F/Y I Q I/L W G L A 194 F M I L 107 C 137-s-s-149 C V C V C T 138 C 139-s-s-147 R C F 173 F /T N * K G 145R W Loop 2 of a determinant /T D 144E L L -s-s- *Glycosylation site 172 top 176 V 3TC or LdT ADV/TDF ETV /L 196 L//top I A/V W 195 M M M/I 226 Adapted from [42] and [48]. The hepatitis B surface antigen (HBsAg) mutations selected during nucleoside/nucleotide analogue (NA) treatment with l-nucleosides (such as lamivudine [3TC] or telbivudine [LdT]), acyclic phosphonates (such as adefovir [ADV] or tenofovir disoproxil fumarate [TDF]) or d-cyclopentane (such as entecavir [ETV]). The mini-loop (codons ), and loop 1 and loop 2 regions of the a determinant are shown. in the progression to HCC as they possess transactivational activity, revealed by increased nuclear factor κb or activator protein 1 promoter activity [56]. Of clinical significance is the recent observation that NA therapy selects for HBV mutants that encode truncated surface proteins and therefore could theoretically accelerate the progression to HCC [25,57]. Treatment of CHB with all the NAs can result in the selection of HBV variants with point mutations in pol that not only confer NA resistance but also result in changes to HBsAg (Figure 5) [26,58]. In particular, the point mutation that causes the rta181t change in pol also encodes a stop codon (sw172*) in the overlapping surface proteins (Figure 5), resulting in truncation of the last 55 amino acids of the C-terminal hydrophilic region of the HBsAg. In vitro analysis of rta181t/sw172* HBV has shown that it is defective in secretion of viral particles resulting in intracellular retention of surface proteins and has a dominant negative effect on WT virion secretion resulting in lower viral loads extracellularly [58]. Two recent reports [25,57] have now provided evidence for involvement of HBV encoding the rta181t/ sw172* mutation in the pathogenesis of and progression to HCC. Analyses of HBV DNA from patients who developed HCC despite 3TC therapy revealed stop codon mutations in the envelope gene in seven of eight patients compared with the control group, in which no patients developed HCC. Using expression constructs encoding the HBV surface proteins, these investigators demonstrated that surface proteins truncated at amino acids sl21, sw156 or sw172 (the last of which corresponds to the surface proteins expressed from rta181t/ International Medical ress

9 Molecular genesis of HBV mutants sw172*) transactivated the c-myc and V40 promoters. NIH-3T3 cells transfected with these constructs were also tumourigenic when injected into nude mice, whereas the WT full-length surface proteins were not [25,58]. Nonetheless, further studies are needed to confirm the clinical significance of these findings. Another common 3TC resistance mutation is rtm204i/sw196* stop (Figure 5), which is observed in up to 10% of 3TCresistant patients [52,58]; however, to date, there is no data available concerning its effects on viral replication or hepatocyte biology. Although NA therapies significantly decrease viral load and improve patient survival in the short term [59], it appears that they might unexpectedly select for HBV variants that are potentially oncogenic, negating the overall efficacy of NAs in preventing hepatocarcinogenesis, the main long-term goal of antiviral therapy in CHB. Future challenges in the treatment of CHB involve the development of antiviral therapies that do not select for potentially oncogenic DR HBV, as well as the development of treatment strategies that effectively inhibit HBV replication eliminating the risk of drug resistance and therefore the emergence of these C-terminally truncated HBsAg HBV variants. Conclusions The current patterns of antiviral drug resistance in CHB are complex. However, four major pathways can be defined in most cases, based on rtm204v/i, rtn236t, rta181t/v and ETV-associated mutations (rtl180m plus rtm204v plus one of rtt184, rt202 or rtm250), with the emergence of multidrug resistance a clear cause for concern in the longer term. Furthermore, broad clusters of compensatory mutations during 3TC therapy will compromise future rescue therapy options with the newer more potent drugs, such as ETV. The best costeffective strategy is to prevent or avoid the emergence of antiviral drug resistance in the first place [22]. This is especially relevant from a public health perspective because most antiviral-resistant HBVs have an HBsAg and have the potential to behave as vaccine escape mutants, be transmitted and infect immunized individuals. Furthermore, there is preliminary evidence that these viruses can be associated with increased oncogenicity. The global programme for control of hepatitis B is built on the foundation of universal infant immunization and will continue to reduce new incident infections of hepatitis B [60]. The challenge of drug resistance to this highly successful campaign requires timely and effective involvement of public health groups cooperatively working with treating physicians to ensure that successful and appropriate therapy guidelines for hepatitis B are achieved and implemented, and that this strategy minimizes or eliminates antiviral drug resistance [23]. Acknowledgements The authors wish to thank Tania Candy for editorial support and the NIH for its funding from the RO1 grant AIO Disclosure statement The authors declare no competing interests. References 1. Harrison TJ. Hepatitis B virus: molecular virology and common mutants. emin Liver Dis 2006; 26: Lai CL, Chien RN, Leung NW, et al. A one-year trial of lamivudine for chronic hepatitis B. Asia Hepatitis Lamivudine tudy Group. N Engl J Med 1998; 339: Wright TL. Clinical trial results and treatment resistance with lamivudine in hepatitis B. emin Liver Dis 2004; 24 uppl 1: Hadziyannis J, Tassopoulos NC, Heathcote EJ, et al. Longterm therapy with adefovir dipivoxil for HBeAg-negative chronic hepatitis B. N Engl J Med 2005; 352: herman M, Yurdaydin C, ollano J, et al. Entecavir for treatment of lamivudine-refractory, HBeAg-positive chronic hepatitis B. Gastroenterology 2006; 130: Tenney DJ, Levine M, Rose RE, et al. Clinical emergence of entecavir-resistant hepatitis B virus requires additional substitutions in virus already resistant to lamivudine. Antimicrob Agents Chemother 2004; 48: Chang TT, Gish RG, Hadziyannis J, et al. A doseranging study of the efficacy and tolerability of entecavir in lamivudine-refractory chronic hepatitis B patients. Gastroenterology 2005; 129: Ling R, Harrison TJ. Functional analysis of mutations conferring lamivudine resistance on hepatitis B virus. J Gen Virol 1999; 80: Delaney WE, IV, Yang H, Westland CE, et al. The hepatitis B virus polymerase mutation rtv173l is selected during lamivudine therapy and enhances viral replication in vitro. J Virol 2003; 77: Angus, Vaughan R, Xiong, et al. Resistance to adefovir dipivoxil therapy associated with the selection of a novel mutation in the HBV polymerase. Gastroenterology 2003; 125: Osiowy C, Villeneuve J, Heathcote EJ, Giles E, Borlang J. Detection of rtn236t and rta181v/t mutations associated with resistance to adefovir dipivoxil in samples from patients with chronic hepatitis B virus infection by the INNO-LiA HBV DR line probe assay (version 2). J Clin Microbiol 2006; 44: Villet, Ollivet A, ichoud C, et al. tepwise process for the development of entecavir resistance in a chronic hepatitis B virus infected patient. J Hepatol 2007; 46: Colonno RJ, Rose R, Baldick CJ, et al. Entecavir resistance is rare in nucleoside naive patients with hepatitis B. Hepatology 2006; 44: Lai CL, Leung N, Teo EK, et al. A 1-year trial of telbivudine, lamivudine, and the combination in patients with hepatitis B e antigen-positive chronic hepatitis B. Gastroenterology 2005; 129: Liaw YF, Gane E, Leung N, et al. 2-Year GLOBE trial results: telbivudine is superior to lamivudine in patients with chronic hepatitis B. Gastroenterology 2009; 136: Heathcote EJ, Gane E, de Man RA, et al. Two year tenofovir disoproxil fumarate (TDF) treatment and adefovir dipivoxil (ADV) switch data in HBeAg-positive patients with chronic hepatitis B (study 103), preliminary analysis. Hepatology 2008; 48 uppl 1:376A. Antiviral Therapy 15:3 t B 459

10 A Locarnini & L Yuen 17. Marcellin, Muti M, Krastev Z, et al. Two year tenofovir disoproxil fumarate (TDF) treatment and adefovir dipivoxil (ADV) switch data in HBeAg-negative patients with chronic hepatitis B (study 102), preliminary analysis. Hepatology 2008; 48 uppl 1:370A. 18. van Bommel F, Feucht H, chlosser B, et al. Detection of mutations associated with resistance to nucleos(t)ide analogs in patients with HBV infection during treatment with tenofovir. J Hepatol 2008; 48 uppl 2: Bartholomeusz A, Tehan BG, Chalmers DK. Comparisons of the HBV and HIV polymerase, and antiviral resistance mutations. Antivir Ther 2004; 9: Yuen LK, Bartholomeusz A, Ayres A, Littlejohn M, Locarnini. Multidrug resistance and cross-resistance pathways in HBV as a consequence of treatment failure. Hepatology 2007; 46 uppl 1:659A. 21. haw T, ozzi T, Locarnini. Activity of nucleos(t)ide analog reverse transcriptase inhibitor (NRTI) combinations against drug resistant HBV: Analysis and optimisation using doseresponse surfaces. Hepatology 2008; 48 uppl 1:739A. 22. Richman DD. The impact of drug resistance on the effectiveness of chemotherapy for chronic hepatitis B. Hepatology 2000; 32: European Association for the tudy of the Liver. EAL Clinical ractice Guidelines: management of chronic hepatitis B. J Hepatol 2009; 50: Lampertico, Vigano M, Manenti E, Iavarone M, Lunghi G, Colombo M. Adefovir rapidly suppresses hepatitis B in HBeAg-negative patients developing genotypic resistance to lamivudine. Hepatology 2005; 42: Lai MW, Huang F, Hsu CW, Chang MH, Liaw YF, Yeh CT. Identification of nonsense mutations in hepatitis B virus gene in patients with hepatocellular carcinoma developed after lamivudine therapy. Antivir Ther 2009; 14: Locarnini A. Hepatitis B virus surface antigen and polymerase gene variants: potential virological and clinical significance. Hepatology 1998; 27: Kamili, ozzi V, Thompson G, et al. Efficacy of hepatitis B vaccine against antiviral drug-resistant hepatitis B virus mutants in the chimpanzee model. Hepatology 2009; 49: Thibault V, Aubron-Olivier C, Agut H, Katlama C. rimary infection with a lamivudine-resistant hepatitis B virus. AID 2002; 16: Clements CJ, Coghlan B, Creati M, Locarnini, Tedder R, Torresi J. Global control of hepatitis B virus: does treatment-induced antigenic change affect immunization? Bull World Health Organ 2010; 88: Bartholomew MM, Jansen RW, Jeffers LJ, et al. Hepatitis-Bvirus resistance to lamivudine given for recurrent infection after orthotopic liver transplantation. Lancet 1997; 349: Ogura Y, Kurosaki M, Asahina Y, Enomoto N, Marumo F, ato C. revalence and significance of naturally occurring mutations in the surface and polymerase genes of hepatitis B virus. J Infect Dis 1999; 180: hields L, Owsianka A, Carman WF, et al. election of hepatitis B surface escape mutants during passive immune prophylaxis following liver transplantation: potential impact of genetic changes on polymerase protein function. Gut 1999; 45: Terrault NA, Zhou, McCory RW, et al. Incidence and clinical consequences of surface and polymerase gene mutations in liver transplant recipients on hepatitis B immunoglobulin. Hepatology 1998; 28: Tipples GA, Ma MM, Fischer K, Bain VG, Kneteman NM, Tyrrell DL. Mutation in HBV RNA-dependent DNA polymerase confers resistance to lamivudine in vivo. Hepatology 1996; 24: Torresi J, Earnest-ilveira L, Deliyannis G, et al. Reduced antigenicity of the hepatitis B virus HBsAg protein arising as a consequence of sequence changes in the overlapping polymerase gene that are selected by lamivudine therapy. Virology 2002; 293: Torresi J, Earnest-ilveira L, Civitico G, et al. Restoration of replication phenotype of lamivudine-resistant hepatitis B virus mutants by compensatory changes in the fingers subdomain of the viral polymerase selected as a consequence of mutations in the overlapping gene. Virology 2002; 299: loan RD, Ijaz, Moore L, Harrison TJ, Teo CG, Tedder R. Antiviral resistance mutations potentiate hepatitis B virus immune evasion through disruption of its surface antigen a determinant. Antivir Ther 2008; 13: Hessel l, West DJ. Antibody responses to recombinant hepatitis B vaccines. Vaccine 2002; 20: He C, Nomura F, Itoga, Isobe K, Nakai T. revalence of vaccine-induced escape mutants of hepatitis B virus in the adult population in China: a prospective study in 176 restaurant employees. J Gastroenterol Hepatol 2001; 16: Theamboonlers A, Chongsrisawat V, Jantaradsamee, oovorawan Y. Variants within the a determinant of HBs gene in children and adolescents with and without hepatitis B vaccination as part of Thailand s Expanded rogram on Immunization (EI). Tohoku J Exp Med 2001; 193: u FH, Chen JD, Cheng H, Lin CH, Liu YH, Chu FY. eroprevalence of Hepatitis-B infection amongst Taiwanese university students 18 years following the commencement of a national hepatitis B vaccination program. J Med Virol 2007; 79: Carman WF. The clinical significance of surface antigen variants of hepatitis B virus. J Viral Hepat 1997; 4 uppl 1: Carman WF, Korula J, Wallace L, Machee R, Mimms L, Decker R. Fulminant reactivation of hepatitis B due to envelope protein mutant that escaped detection by monoclonal HBsAg ELIA. Lancet 1995; 345: Carman WF, Trautwein C, van Deursen FJ, et al. Hepatitis B virus envelope variation after transplantation with and without hepatitis B immune globulin prophylaxis. Hepatology 1996; 24: Hsu HY, Chang MH, Liaw H, Ni YH, Chen HL. Changes of hepatitis B surface antigen variants in carrier children before and after universal vaccination in Taiwan. Hepatology 1999; 30: Karthigesu VD, Allison LM, Fortuin M, Mendy M, Whittle HC, Howard CR. A novel hepatitis B virus variant in the sera of immunized children. J Gen Virol 1994; 75: Oon CJ, Lim GK, Ye Z, et al. Molecular epidemiology of hepatitis B virus vaccine variants in ingapore. Vaccine 1995; 13: Wallace W, Carman W. urface variation of HBV: scientific and medical relevance. Viral Hepatitis Reviews 1997; 3: Wands JR, Fujita YK, Isselbacher KJ, et al. Identification and transmission of hepatitis B virus-related variants. roc Natl Acad ci U A 1986; 83: Waters JA, Kennedy M, Voet, et al. Loss of the common a determinant of hepatitis B surface antigen by a vaccineinduced escape mutant. J Clin Invest 1992; 90: Basuni AA, Butterworth L, Cooksley G, Locarnini, Carman WF. revalence of HBsAg mutants and impact of hepatitis B infant immunisation in four acific island countries. Vaccine 2004; 22: Yuen LK, Ayres A, Littlejohn M, et al. EQHEB: a sequence analysis program and relational database system for chronic hepatitis B. Antiviral Res 2007; 75: Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Annu Rev Immunol 1995; 13: Caselmann WH, Meyer M, Kekule A, Lauer U, Hofschneider H, Koshy R. A trans-activator function is generated by integration of hepatitis B virus pre/ sequences in human hepatocellular carcinoma DNA. roc Natl Acad ci U A 1990; 87: Kekule A, Lauer U, Meyer M, Caselmann WH, Hofschneider H, Koshy R. The pre2/ region of integrated hepatitis B virus DNA encodes a transcriptional transactivator. Nature 1990; 343: International Medical ress

11 Molecular genesis of HBV mutants 56. chluter V, Meyer M, Hofschneider H, Koshy R, Caselmann WH. Integrated hepatitis B virus X and 3 truncated pre/ sequences derived from human hepatomas encode functionally active transactivators. Oncogene 1994; 9: Lai MW, Yeh CT. The oncogenic potential of hepatitis B virus rta181t/ surface truncation mutant. Antivir Ther 2008; 13: Warner N, Locarnini. The antiviral drug selected hepatitis B virus rta181t/sw172* mutant has a dominant negative secretion defect and alters the typical profile of viral rebound. Hepatology 2008; 48: Liaw YF, ung JJ, Chow WC, et al. Lamivudine for patients with chronic hepatitis B and advanced liver disease. N Engl J Med 2004; 351: Chang MH, Chen CJ, Lai M, et al. Universal hepatitis B vaccination in Taiwan and the incidence of hepatocellular carcinoma in children. Taiwan Childhood Hepatoma tudy Group. N Engl J Med 1997; 336: Zoulim F. Mechanism of viral persistence and resistance to nucleoside and nucleotide analogs in chronic hepatitis B virus infection. Antiviral Res 2004; 64:1 15. Accepted for publication 14 October 2009 Antiviral Therapy 15:3 t B 461