Hepatitis B virus (HBV) is a small, non-cytopathic virus with a circular partially double-strand DNA genome of approximately 3.2 Kb 1. The genome has 4 overlapping genes: PreS/S, PreC/C, X, and P. The Pre S/S gene encodes for the three envelope proteins, large, middle, and small or HBsAg. The PreC/C gene encodes for the nucleocapsid protein and HBeAg. The X gene encodes for the transactivating protein X and the P gene encodes for polymerase with reverse transcriptase (RT) and RNase H activity 1. The HBsAg contains the major epitopes for induction of a protective humoral immune response 23456. These epitopes are localized in the region known as the "a" determinant, between amino acid residues 99 and 169 78910, and are involved in the binding of antibodies against HBsAg. Amino acids changes in this region render mutant strains able to escape immune responses induced by vaccines 561112. Immune-escape mutants occur naturally 131415, and in strains from patients with a weak or negative HBsAg reactivity in detection assays 1015.

HBV strains are classified into 8 main genomic groups or genotypes, designated A through H 161718, arbitrarily defined by an intergroup divergence of more than 8% based on the complete genomes 16192021. The relationship between subtypes and genotypes has been reported, and several studies have described the geographic distribution for the different HBV subtypes/genotypes 16172022232425. The study of HBV genotype has recently been focused on the clinical outcome of the infection, since studies suggest a possible association with the natural history and severity of the infection 26.

In Mexico, frequency of HBV chronic carriers is low, and the prevalence of HBsAg in blood donors ranges from 0.15 to 1.4% 2728; the frequency of hepatocellular carcinoma is also low 29. Scarce studies have addressed genetic diversity and amino acids changes in HBsAg of HBV strains from Mexico 30. The aim of the present work was to study genetic diversity in the S gene of HBV strains from asymptomatic and symptomatic carriers in Mexico. Genotype and mutations in the "a" determinant of the S gene leading to amino acids changes were compared between these two groups.

We studied 21 asymptomatic HBsAg carriers, 3 women (mean age 30 years old) and 18 men (mean age 42 years old), with average viral load of 0.7965 × 10⁴ copies/ml; and 12 symptomatic patients, 5 women (mean age 50 years old) and 7 men (mean age 39 years old), with average viral load of 2.7 × 10⁶ copies/mL (Table 1). Functional hepatic tests confirmed hepatic damage in symptomatic but not in asymptomatic cases (Table 1). Genotypes were determined by reverse hybridization and sequencing. Genotypes determined by reverse hybridization were H, C, F, C/H, C/F/H and indeterminate (ID) in asymptomatic cases, whereas in symptomatic patients only H and C/H genotypes were found (Table 2). On the other hand, the genotypes determined with the sequencing technique were only H, and C; genotype H was the predominant type (Table 2). In four asymptomatic cases the genotype could not be determined (ID) with the reverse hybridization technique, but was identified as genotype H with the sequencing method. Sequences of the "a" determinant were aligned with sequences of genotypes A-H (Z72478, X98077, AB048704, AB048701, AB032431, AB036907, AF160501, U91827, NC_003977) and analyzed to identify mutations. Mutations identified in HBV strains from the asymptomatic group were Y100N, T126I, Q129H and N146K; whereas in strains from the symptomatic group identified mutations were F93I and A128V. Subtypes found in the asymptomatic group were adw4q+, adw1q+, and adrq+; whereas in symptomatic patients subtypes detected where only adw4q+, and adrq+. Altogether, the genotype/subtype most frequent in our population was H/adw4q+ (Table 2).

In this work we analyzed genetic diversity in HBV isolates from asymptomatic (blood donors) and symptomatic chronic carriers in Mexico. Genotypes were determined with a commercially available reverse-hybridization technique and compared with the sequencing technique. Reverse hybridization allows the detection of infection with mixed genotypes, suggesting that this technique is more reliable than sequencing to detect quasispecies present in lower proportion in the sample; although in some cases it failed to establish the genotype. On the other hand, besides genotyping, nucleic acid sequencing provides information on nucleotide and amino acid sequences of the region investigated, while other molecular methods detect only the point mutations specified by the probes or primers used.

Of interest, a higher diversity in the infecting genotypes was found among asymptomatic carriers (H, C, F, C/H, C/F/H), than in symptomatic patients, where genotype H was found in all cases and in only one case a C/H infection was detected. These results would suggest that in our population, genotype H strains are more prevalent than C and F genotypes in symptomatic cases. This observation is relevant since studies indicate certain association of the genotype with the clinical outcome of the infection; thus, in hepatocellular carcinoma, genotype C was more prevalent in patients >50 years and genotype B more prevalent in patients <50 years of age, than in age-matched asymptomatic carriers 37. Of interest, in our study, genotype C was identified in four of the 21 asymptomatic cases. Alternatively, LMV treatment in the symptomatic cases might be selecting for the H genotype. Diversity of HBV genotypes may also affect the accuracy of diagnostic tests and therapeutic decisions 32.

HBV has been classified in eight genotypes, A through H 18, which show a geographic distribution. In this study, genotype H was found as the predominant genotype in Mexican strains, followed by genotypes C and F. Genotype H has a close phylogenetic relationship with genotype F 1718. The genotype C is reported as prevalent in Asian populations 2033, and patients observed in our population might correspond to imported cases, due to the increased mobility among Asian and Latin-American countries; in fact, recent reports have documented the presence of HBV genotype H strains among Japanese blood donors 34, as an evidence of the global mobility. The genotype prevalent in a population may determine the type of mutations prevalent in the infecting strains 35; for example, the immune-escape mutant G145R is closely associated with genotype D 36. HBsAg has also been classified in subtypes based on the sequence of the S gene and in identification of the amino acids encoded at specific positions 33; in our population, subtypes, like genotypes, were more diverse among strains from asymptomatic carriers than in symptomatic cases.

The hepatic function tests and HBV viral load showed that in our symptomatic patients there was hepatic damage and high viral replication, which suggests that the antiviral treatment with LMV was not working, even thought the drug was administered for at least one year in all cases. It is likely that after a long failed therapy, resistance to LMV has been developed. The main mutations associated with LMV resistance are located at the RT gene in the 204 position (rtM204I/V), which is in the catalytic YMDD motif. The P gene overlaps the S gene and thus mutations selected during antiviral treatment may cause concomitant changes to the overlapping reading frame; in particular altering the C-terminal region of HBsAg. Thus for mutations associated with LMV resistance, the rtM204V change is associated with a I195M change in the S gene (sI195M), while the rtM204I change is associated with three possible changes in the S gene, sW196S, sW196L, or a stop codon. In addition, mutation rtA181T is associated with the stop mutation sW172stop in the S gene 37. However, in our study, none of these mutations were found in the S gene of the strains from symptomatic cases.

There has been several reports on HBV S gene mutants affecting amino acid position 120, 123, 124 126, 129, 131, 141, 144 and 145 of the "a" determinant and preS region. The most relevant mutations seem to be the substitutions of amino acid G145R, K141E and T131I and insertion of 3 amino acids between residues 123 and 124, since they markedly affect the antigenic structure of HBsAg 38. Mexican strains of the present study showed mutations at the first and second loop of the "a" determinant; although it should be noted that these mutations differed between asymptomatic and symptomatic strains. Thus, in the asymptomatic cases mutations were found at positions 100, 126, 129, and 146, whereas in symptomatic patients mutation were in positions 93 and128; none of the above mutations are reported to affect significantly the antigenic structure of the protein. It should be emphasized that all cases included in the study were selected because they were HBsAg positive and so those patients with mutations rendering the S protein undetectable with the antibodies tested, were excluded.

In conclusion, we report the genotypes of HVB prevalent in symptomatic and asymptomatic patients in Mexico and show that there is more genetic diversity and mutations in the "a" determinant of HBV strains infecting asymptomatic carriers than in symptomatic patients. In HBV strains from our population genotype H and subtype adw4q+ were predominant. The role of viral genotype and subtype, as well as the ethnicity of the host in the determination of the clinical outcome of HBV infection is still unclear, and studies on genotypes infecting populations should help to clarify this issue.

The authors declare that they have no competing interests.

MERT: Designed and supervised most of the molecular studies; analyzed sequences and interpreted results. Participated in the preparation of the manuscript.

HAVS: carried out molecular techniques and participated in the analysis of results.

VJB: carried out immunoassays and interpreted the results.

MDV: responsible for the clinical aspects of the study, including the recruitment of patients; participated in the writing of the manuscript.

JT: Participated in the designed of the study and in the analyses and interpretation of the results. Critical review and edition of the manuscript.

OMH: Participated in the clinical designed of the study and in obtaining financial support for the study.

MTAM: responsible for the design of the study and for acquiring financial support. Supervision of the experimental work. Participated in the drafting of the manuscript.