INVESTIGADORES
TORRES Carolina
congresos y reuniones científicas
Título:
Site-specific variation in Hepatitis B virus
Autor/es:
TORRES, C.; CAMPOS, R.H.; MBAYED, V.A.
Lugar:
Rotterdam
Reunión:
Workshop; 15TH INTERNATIONAL BIOINFORMATICS WORKSHOP ON VIRUS EVOLUTION AND MOLECULAR EPIDEMIOLOGY; 2009
Institución organizadora:
The Erasmus Postgraduate School Molecular Medicine (Rotterdam, Holanda) y la Katholieke Universiteit Leuven Bélgica.
Resumen:
Hepatitis B virus (HBV) has a DNA genome, but it replicates via a RNA intermediate. The substitution rate of HBV is expected to be determined in part by the low fidelity of its reverse transcriptase, but also, by the partial overlapping of the four open reading frames (S, Core, X, Pol), which jointly with secondary structures of the HBV might constrain the number and nature of substitutions occurring in the HBV genome. The aim of our work was: a) to analyze the site-specific variation for the 1st, 2nd and 3rd codon position of all the HBV genes in both, overlapping (phase +1, +2) and nonoverlapping regions; and b) to evaluate the extent of substitution rate variation for each codon position in different regions of the polymerase gene. Complete genomes belonging to all genotypes (A – H), including 33 isolates sequenced in our laboratory, were obtained from Genbank. Sequences with insertion, deletion or any molecular marker of HBeAg (-) stage were eliminated, with exception of all genotype G sequences, which possess two constitutive translational stop codons at positions 2 and 28 of preCore/Core. The data set resulting covered 652 sequences, for genotype A (n=126), B (n=125), C (n=144), D (n=109), E (n=46), F (n=69), G (n=15) and H (n=18). Analyses were performed on genotypes and subgenotypes separately. Site specific variation was evaluated using the Entropy method implemented in Bioedit v7.0.9.0 software and relative substitution rates obtained by Maximum Likelihood approaches implemented in DNArates v1.0 software and in the module Baseml of PAML v4.0 package. Genome regions analyzed were preS1, preS2, S, ORF S, Core, X and Pol, in both overlapping and nonoverlapping regions. Variations in codon position were compared by a Kruskall-Wallis test (nonparametric ANOVA). The module Baseml of PAML v4.0 package was used for obtaining estimators of α value (parameter of gamma distribution) of each partition corresponding to each codon position. In general, all HBV genotypes and subgenotypes showed that the most variable codon position in the ORF S was the 3rd position that represents the 1st position of ORF P (3S/1P). In contrast, for other regions (overlapping C/P, X/P and nonoverlapping Pol) the third codon position of Pol was the most variable, especially in nonoverlapping Pol (3P). Besides, for genotypes A to H, 3P displayed α~1 while most of other codon position in Pol showed α5 for all codon position. In conclusion, different regions of Pol have dissimilar pattern in relative variation of codon position. In the overlapping S/P region, the only one in phase +1, synonyms changes in S prevail over non synonyms changes in Pol, whereas synonyms changes in Pol are apparently preferred in the other regions, which are in phase +2 or nonoverlapping. Besides, variation observed in the 1st and 2nd codon position of Pol is restricted to few sites with very high rates (α≤1). However, this fact is less marked in the 3rd codon position of nonoverlapping Pol region (α~1), which would display more sites with high substitution rates.