Resumo

Early region 3 (E3) of the human adenovirus (HAdV) genome consists of a cassette of genes that encodes several proteins implicated in modulating the host immune response. The size and composition of the E3 transcripts vary among the HAdV species and serotypes. The aim of this study was to describe the variability of the nucleotide and predicted amino acid sequences of the HAdV-3 and HAdV-7 E3 proteins: 12.5K, 16.1K, 19.5K and 7.7K. Nasopharyngeal samples were collected from children with acute respiratory illnesses in São Paulo, Brazil, between 1995 and 2006. During the studied period 677 HAdVs were detected and typed by PCR and partial nucleotide sequencing of the hexon and fiber genes. The entire E3 region of 139 HAdV-3 and HAdV-7 strains were amplified and sequenced and the nucleotide and the predicted amino acid sequences were aligned with published sequences. Phylogenetic trees were constructed using MEGA 5 software. ORF 12.5K was the most conserved. A unique amino acid substitution in residue 37 (S/P/F) was observed, separating the samples into 3 groups: HAd-7p/-7h, -3a1 and -3p /-3a2. In ORF 16K, 5 distinct amino acid substitutions were observed at positions 46 (L-F), 51 (T-N), 52 (F-S), 69 (A-G) and 77 (L-G) separated the viruses into 4 phylogenetically distinct clusters: HAd-7p/-7h, -3p, -3a1 and -3a2. In ORF 19K, 4 amino acid substitutions at positions 21 (L-M), 61 (S-F), 85 (E-Q) and 137 (A-V) separated the viruses into 4 groups with mixed types: HAdV3p; -7p/-3a/-7h; -7h/-7p/-3a; and -3a/-7p. ORF 7.7K was the most variable. The proteins of HAd-7p,-3p, -3a and -7h exhibited different sizes and sequences (HAd3p-94aa, -7p-66aa, -3a and -7h-45 aa). The samples typed as HAdV-3p always grouped in the same cluster in all protein analyzed. HAd-3a1, -3a2, -7p and -7h were grouped in different clusters based on ORFs 19K and 7.7K sequences. These ORFs could be considered “hot spots” for mutation. Financial support: Fapesp 03/020555