Emergence of a Novel Mutation in the FLLA Region of Hepatitis B Virus during Lamivudine Therapy

Abstract

The emergence of resistance to lamivudine has been one of the major stumbling blocks to successful treatment and control of hepatitis B virus (HBV) infections. The major mechanism of resistance has been attributed to the alteration in the YMDD motif of the HBV polymerase due to an amino acid change of rtM204 to V/I and an accompanying rtL180M conversion. A novel mutation pattern in a patient having clinical breakthrough under lamivudine therapy was discovered. The mutant had a rtL180C/M204I genotype and was detected after 2 years of therapy with lamivudine. To characterize this novel variant, site-directed mutagenesis was performed using a vector construct containing the HBV genome. Transient transfection studies in human hepatoma cells with HBV carrying the new mutant demonstrated that the rtL180C/M204I mutant was resistant to lamivudine up to 10 μM. The resistance profile was comparable to that of the previously reported rtL180 M/M204I-containing virus. These observations were further confirmed by generation of stable cultures transfected with the mutant virus.

Hepatitis B virus (HBV) is a pathogen that afflicts over 350 million people with liver disease worldwide (27). The persistence of the infection can have devastating consequences for the infected person, as the natural progression of the disease culminates with cirrhosis of the liver and hepatocellular carcinoma. Until recently, only two treatment options, alpha interferon and lamivudine, were available for the management of HBV. Recently, adefovir dipivoxil and entecavir were approved by the U.S. Food and Drug Administration for treatment of HBV (23, 24).

While the therapeutic utility of lamivudine is clearly documented in the literature (4, 5, 12, 16, 17), the emergence of resistance to lamivudine has been the major hurdle to viral clearance since long-term lamivudine monotherapy results in clinical resistance to the drug (18). The mutations responsible for the clinical nonresponse have been characterized (1, 3, 4, 18) and can persist resulting in a major hurdle to viral clearance. While a number of mutations have been reported, the cardinal change that confers resistance has been the conversion of the methionine residue in the YMDD motif to valine or isoleucine (rtM204V/I) followed by the rtL180 M change (for review of mutations see references 19 and 23).

There are some similarities in the resistance pattern to lamivudine between human immunodeficiency virus (HIV) and HBV (10). For example, a number of HIV cases show initial replacement of methionine with valine or isoleucine at residue 184 of the polymerase (6, 25), which corresponds to the rtM204V/I in HBV polymerase (26).

Recently, mutations other than rtM204I/V have been reported for HBV (7, 19, 20, 23). These changes, albeit observed at a lesser frequency, underscore the multiple strategies employed by the virus to circumvent inhibition of the viral replication machinery. The present studies are a continuation of our efforts to identify mechanisms underlying clinical resistance to lamivudine. A novel genotype with the rtL180C change accompanied by rtM204I in the YMDD motif of HBV polymerase was identified during the course of lamivudine treatment. To the best of our knowledge, this is the first report of such an HBV variant in the clinic. We have characterized a cloned virus containing this YIDD change and related mutation (rtL180C/M204I) in the polymerase region focusing on the in vitro susceptibility to lamivudine.

MATERIALS AND METHODS

Patient history.

A 68-year-old Caucasian male individual presented initially with Kaposi's sarcoma for which chemotherapy was administered. Subsequently the patient had to undergo surgery for broken leg. After two years, the patient was also diagnosed with diabetes mellitus. The blood tests on following visits were positive for HBsAg. Liver biopsy revealed chronic active hepatitis with a histological activity index of 6, stage 1. The patient was initiated on alpha interferon therapy (6 mU three times a week) for 4 months and subsequently on lamivudine at a dose of 150 mg/day for about 3 years and 4 months. The study was conducted in accordance with the ethical guidelines of the 1995 Declaration of Helsinki. Informed consent from the patient was obtained for use of samples for the study. Protocols used in the study were approved by the Ethics Committee of the Ankara Medical School, Ankara, Turkey.

Characterization of serum HBV from the patient.

HBsAg, anti-HBs, HBeAg, anti-HBe, anti-HCV, and anti-HIV were determined by the microparticle enzyme immunoassay and anti-HDV by the enzyme-immunoassay method (Abbott Laboratories, Illinois). HBV-DNA levels were determined by using a commercially available liquid-hybridization assay, having a detection limit of 5 pg/ml for viral DNA (Digene), according to the instructions provided with the kit. All serum samples were stored at −70°C.

Sequence analysis of HBV during the course of study.

The HBV DNA from the serum samples was isolated by digestion with proteinase K followed by phenol/chloroform extraction and ethanol precipitation. PCR amplification and sequencing were performed using published methodology (7).

Construction of rtM204I and rtL180C/M204I HBV mutants.

The single and double mutants of HBV were generated by conversion of the amino acid residue at position 204 to isoleucine and at 180 to cysteine or methionine. The vector designated pat0303 containing the HBV expression cassette driven by tetracycline-regulated promoter was used for mutagenesis studies (13, 14). This vector has HBV of genotype ayw and contained valine in place of methionine at position 204 (YVDD motif). Site-directed mutagenesis was performed to substitute ATT in place of GTG in the DNA coding for the YVDD motif and TGT in the place of CTG in the coding domain for the FLLA region using the QuikChange kit (Stratagene, Cedar Creek, TX). PCR amplification of the DNA polymerase region of the mutant vectors derived from the site-directed mutagenesis was performed (21). The PCR products were resolved on an agarose gel and the amplified DNA was purified and sequenced to confirm the presence of the YIDD (GTG to ATT) and CTG to TGT at position 180.

Generation of mutant HBV in hepatoma cells.

Transient transfections were performed as described earlier (21). Plasmids carrying the various HBV variants generated by site-directed mutagenesis were cotransfected with a plasmid designated pUHD, carrying the neomycin resistance gene, and a second vector carrying the β-galactosidase gene (11, 15). Huh-7 cells were plated in six-well plates and transfections were performed 24 h after seeding the cells. Treatments were performed 24 h after transfection and analysis was done 5 days after the initiation of treatment. The cell lysates were first treated with DNase I to digest input plasmids prior to processing for isolation of HBV DNA from viral particles.

Analysis of the HBV DNA was conducted as described earlier (21) with some modifications, such as the use of real-time PCR methodology for viral quantification. Purified viral DNA was analyzed by real-time PCR using HBV-specific primers and Taqman probe using an Applied Biosystems 7900 sequence detection system (7). An aliquot of the cells was also processed for β-galactosidase activity. This enzyme activity was determined according to the protocols supplied by the manufacturer (Promega Life Sciences, Madison, WI). Culture supernatants were also analyzed for hepatitis B surface antigen (HbsAg). The assay was performed using the AUSZYME monoclonal diagnostic kit (Abbott Laboratories, Chicago, IL).

Further confirmation of the effect of lamivudine on rtL180C/M204I mutant virus was performed by Southern analysis of the HBV DNA obtained from the cells transfected with the rtL180C/M204I mutant HBV vector. The DNA was resolved on a 1% agarose gel, transferred to a nylon membrane, and hybridized to HBV-specific probe and the hybridization results were visualized by phosphorimaging (22).

Transfection and selection of stable cell cultures with mutations.

The plasmid containing the mutation was transfected in HepG2 cells along with the pUHD vector carrying the tetracycline responsive elements (13, 14). Transfection was performed using the Superfect reagent as per the manufacturer's protocol (Qiagen, Valencia, CA). The transfected cultures were maintained in growth medium containing G418 and colonies that develop under these conditions were isolated and were further propagated. The culture medium and growth conditions were as described previously (15).

Characterization of the stable cell cultures.

Cell cultures derived from the G418-resistant colonies were characterized for the expression of HBV. Culture supernatants grown in the absence of tetracycline were collected and DNA isolation was performed. PCR amplification of the polymerase region was conducted using HBV-specific primers.

Analysis of the effect of lamivudine on mutant viruses from stable cell cultures.

Stable cells carrying rtM204I and rtL180C/M204I were seeded at a density of 10⁵ cells per 25-cm² flask in culture medium containing tetracycline (15). After 4 days, the cultures were treated with various concentrations of lamivudine in medium lacking tetracycline. On the fifth day, fresh medium with appropriate concentrations of lamivudine was added. The experiment was terminated on the eighth day. Cells were harvested and processed for HBV DNA from viral particles as described above for transient transfections. An aliquot of the purified DNA was analyzed by real-time PCR for HBV specific DNA (7). HBV DNA standards with predetermined copy numbers (genome equivalents) ranging from 8 × 10² to 8 × 10⁷ were incorporated in the experiment to monitor the copy numbers of the unknown samples as well as to determine the efficiency of PCR amplification.

RESULTS

Characteristics of HBV in patient serum.

At the initial presentation, the patient had ≥2,000 pg/ml of HBV DNA as determined by the Digene test. Blood tests revealed positivity for HBsAg and HBeAg but were negative for anti-HBe, anti-HBs, and anti-Delta. No indications of presence of HIV and HCV were noted as determined by tests specific for these viruses. Lamivudine treatment over a period of 2 years did lower the viral load to <5 pg/ml. However, 22 months after this observation the viral load increased to 1,000 pg/ml (Table 1).

TABLE 1.

Clinical and virological characteristics of patient with L180C/M204I HBV mutation^a

Diagnosis and treatment	Clinical visit date (mo.day.yr)	HBV mutation		Liver enzymesa		HBV DNA (pg/ml)	Biopsy
		L180C CTG	M204V ATG	ALT	AST
Diagnosed with Kaposi's sarcoma	1985
Received chomotherapy for 1 yr	1987
Diagnosed with diabetes mellitus	1991
Received 7 units of blood during surgery for broken leg	1992
HBsAg found to be positive	1997b
Liver biopsy done	02.21.1998					>2,000
	02.26.1998						HAI-6 stage 1
Interferon (6 mU)	03.15.1998
	07.13.1998					>2,000
Epivir (150 mg)	07.13.1998					1,137
	09.15.1998					>2,000
	06.28.1999					>2,000
	01.17.2000					<5
	10.26.2000					284
	11.05.2000	TTG	ATG
	03.13.2001	TGT	ATT	42	31	44
	05.17.2001	TGT	ATT	39	29	160
	07.17.2001	TGT	ATT	37	30	218
	09.20.2001			38	28	593
	11.19.2001b	TGT	ATT	45	32	1,074
	03.07.2002			55	34	930
	04.15.2002			47	30	791
	04.30.2002			53	36

^aALT, alanine aminotransferase; AST, aspartate aminotransferase.

^bClinical tests were positive for HBsAg, HBeAg, and anti-HBC; negative for anti-HBsAg, anti-HBeAg, anti-HCV, anti-delta, and anti-HIV.

Emergence of the mutant HBV.

The initial observation of change in the HBV polymerase sequence was noted in the FLLA region. The change detected was a conversion of CTG to TTG, which is a polymorphism. Subsequently, a two-nucleotide mutation was noticed, transforming the CTG codon (leucine) to TGT (cysteine). Simultaneously, the YMDD motif was altered from ATG to ATT, resulting in the rtL180C/M204I genotype (Fig. 1).

FIG. 1.

Evolution of the novel HBV variant from the wild-type virus. Virological assays and determinations were performed as described in Materials and Methods.

Confirmation of the mutant HBV generated by site-directed mutagenesis.

The mutant HBV vectors generated by site-directed mutagenesis showed the presence of the resistance mutations. Plasmid vectors rtL180C, rtM204I, and rtL180C/M204I were successfully generated. The sequence analysis using the forward primer showed the presence of rtM204I (nucleotide change to ATT) as well as rtL180C (conversion of CTG to TGT) (Fig. 2). There were no additional mutations in the polymerase region other than the ones introduced by site-directed mutagenesis. The sequence was also confirmed by reverse sequencing.

FIG. 2.

DNA sequence alignment of pol region of mutant HBV generated by site-directed mutagenesis. Generation of the plasmids and PCR was performed as described in Materials and Methods. The sequence of the pol region flanking the FLLA region and the YMDD motif is depicted.

Susceptibility of rtL180C/M204I HBV in transfected cells to lamivudine.

The supernatants of the cultures transiently transfected with the various vectors showed the production of HBsAg, indicating the formation of viral particles in the transfected cells. Real-time PCR analysis of the DNA from control and lamivudine-treated transfected cells showed that the levels of HBV DNA were not significantly inhibited (Fig. 3). The rtM204M wild type showed sensitivity to lamivudine at 10 μM, whereas the rtL180C/M204I mutant was resistant. The pattern of the resistance, under these conditions, was similar to that of the rtM204I variant. The resistance pattern remained the same when normalized either to β-galactosidase activity or levels of HBsAg. Transfection studies of HBV with alteration only in the FLLA region (rtL180C/M204M and rtL180M/M204M) in Huh-7 cells showed that L180C and L180M are resistant to lamivudine whereas the wild-type HBV (rtL180L/M204M) was sensitive under the conditions of the assay (Fig. 4).

FIG. 3.

HBV copy numbers from wild-type (rtM204M) and mutant (rtM204I and rtL180C/M204I) viruses from transiently transfected Huh-7 cells. HBV abundance was determined using real-time PCR technology. Cells transfected with each of the vector constructs were treated with 10 μM of lamivudine for 5 days. Untreated cultures served as controls. The DNA copy numbers in each group were normalized to their HBsAg values.

FIG. 4.

HBV copy numbers from wild-type (rtL180L/M204M) and FLLA mutant (rtL180C/M204M and rtL180M/M204M) viruses from transiently transfected Huh-7 cells. HBV DNA quantification was done using real-time PCR technology. Cells transfected with the vector and a plasmid carrying the β-galactosidase gene construct were treated with 10 μM of lamivudine (3TC) for 5 days. Untreated cultures served as controls. The DNA copy numbers in each group were normalized to β-galactosidase activity.

The resistance phenotype of the virus containing rtL180C/M204I was further confirmed by visualization of the replicative intermediates on Southern blots hybridized to ³²P-labeled HBV-specific probes. β-Galactosidase activity was also detected in the cell extracts of the transfected cells, confirming the uptake and expression of the transfected vectors in the Huh-7 cells. The data, normalized to intracellular β-galactosidase activity, confirmed the lack of inhibition of antiviral activity on the novel HBV variant by lamivudine (Fig. 5A and B).

FIG. 5.

A. Replicative intermediates of rtL180C/M204I HBV from transiently transfected Huh-7 cells. Southern analysis of HBV DNA from rtL180C/M204I HBV transfections was performed as described in Materials and Methods. OC: open circular HBV DNA; DS: double-stranded HBV DNA; SS: single-stranded HBV DNA; NT: nontransfected. The positions of the size markers of HindIII-digested lambda markers (4 kb and 2 kb) are indicated. B. β-Galactosidase activity from rtL180C/M204I HBV transient transfection. β-Galactosidase activity from the cells cotransfected with a vector carrying the β-gal gene was performed as described in the Materials and Methods section.

Stable cell cultures containing the rtM204I and rtL180C/M204I mutations that were selected with G418 were positive for HBsAg in the culture supernatants (data not shown). DNA purified from the supernatants when amplified with primers specific for the DNA polymerase region, showed products consistent with that expected for HBV (Fig. 6A). Further confirmation of the resistance pattern of the novel mutant HBV genotype is depicted in the HBV response to lamivudine in the stable transfectants. Repeated treatments of lamivudine to cultures expressing the mutant virus showed no inhibitory activity on the DNA replication up to 10 μM. The data from the quantification by real-time PCR of HBV DNA from untreated cells and cells treated with lamivudine demonstrated lack of inhibitory effect (Fig. 6B).

FIG. 6.

A. rtM204M, rtM204I, and rtL180C/M204I HBVs from stably transfected HepG2 cells. The arrow indicates PCR-amplified HBV product from the supernatants of cell cultures. Lanes: 1, product from rtM204M; 2, rtM204I HBV; and 3 and 4, product from rtL180C/M204I. Lambda DNA digested with HindIII was used as the size marker. B. Effect of lamivudine on HBV DNA production from stably transfected cell cultures. Virus quantification was performed by real-time PCR methodology as previously described (7).

DISCUSSION

Currently, alpha interferon, lamivudine, adefovir dipivoxil, and entecavir are the only four treatment modalities used for the management of HBV infection, although other anti-HBV agents are now in Phase 3 clinical trials (4, 23). Long-term use of lamivudine in the clinic has generated a plethora of information with respect to the clinical response, as well as the emergence of resistant phenotypes of HBV (18). While there are some similarities with respect to the mutations in the active site of both the HBV and HIV polymerases selected by lamivudine, similar resistant variant of HBV exhibit some unique mutations outside the catalytic region. It is postulated that these changes are paramount for the virus empowering it with the advantage required for its survival in the presence of the drug. The most common change observed in the FLLA region is the conversion of amino acid leucine (L) to methionine (M) at position 180.

The present study stems from the observation in the clinic of a novel mutation in an individual undergoing lamivudine therapy for HBV infection. The virus exhibited a gradual progression from wild type to a rtM204I variant in the pol gene. Also accompanying the change in the YMDD motif was the conversion of L (leucine) to C (cysteine) at the amino acid position 180, leading to FLCA instead of FLLA. This amino acid change did not create a stop codon in the overlapping surface antigen sequence open reading frame.

We have for the first time documented the emergence and persistence of a novel rtL180C/M204I variant of HBV during lamivudine therapy. The mutation rtM204I has been described earlier to appear in response to lamivudine monotherapy. However, some studies have shown the preponderance of rtM204I in the second year of treatment in comparison to rtM204V during the first year (8, 9). The conversion of methionine (M) to valine/isoleucine (V/I) at position 204 in the YMDD motif renders HBV resistant to lamivudine (1). The virus with rtM204V can also acquire rtL180M change and these are the most common genotypes seen in the clinic in response to long term lamivudine therapy. Our studies show that in addition to the above changes that confer resistance, novel amino acid changes such as leucine to cysteine could culminate in the same end point, which is drug-resistant HBV.

We have reported recently an rtM204S mutation that resulted in virus resistant to lamivudine by generating stable transfectants that express HBV with the YSDD motif (7). A similar observation was also made by other researchers in the clinic (20). Also, unique mutations, rtA222T and rtL336V, accompanied with mutations in the basal core promoter, core, surface antigen, and X protein regions have been reported (3).

The fluctuations in the viral load in the patient seen during the course of treatment conform to characteristic pattern observed in the clinic during antiviral therapy. There is a steady and profound initial response to the drug due to the sensitivity of the wild-type virus followed by the slow but marginal increase in the viral load. The latter phase is contributed by the emerging resistant variant(s). Even though this variant population does not reach the initial wild-type levels, they present a formidable barrier to successful treatment due to their recalcitrance to the antiviral used. The low replicative capacity of some of these mutants may render them inept to achieve high viral loads (7, 23).

Our studies also underscore the challenges faced in tackling HBV resistance. Even though variants such as rtM204S as well as the one we report in this manuscript are not as frequent as rtL180M/M204V, they could exist in subjects as minor populations and contribute to the clinical resistance. Based on the proposed model for HBV and the interaction of lamivudine with the polymerase, it is suggested that the compensatory nature of rtL180M mutation is due to the interaction of the methionine to its neighboring amino acids (10) and steric hindrance (28). Based on the minimal energy fit of this model, leucine does not cause this interference in stacking. Since both methionine and cysteine are sulfur-containing amino acids, with similar side chains (S-CH₃ for methionine and S-H for cysteine) similar interactions can be envisioned.

Our studies with rtL180C change in the FLLA region leading to resistance to lamivudine are similar to rtL180M-mediated resistance (21). Mutations in the HBV polymerase generated by other nucleosides such as adefovir dipivoxil emphasize the multiple strategies the virus can adopt to elude inhibition (2). Understanding the interaction of the drugs with the active site of the viral enzyme is critical for the development of therapies to combat HBV. Mechanisms by which the virus circumvents the inhibitory activity would shed light on the strategies that need to be implemented for the successful therapeutic intervention. Furthermore, understanding the biology of the resistant virus is critical for the determination of the mechanism of resistance as well as the management of individuals with HBV.