Abstract

PIP: DNA sequences encoding the C2-V3 regions or the C2-V5 regions of the surface glycoprotein gp 120 of the human immunodeficiency virus type 1 (HIV-1) were amplified by the polymerase chain reaction (PCR) from peripheral blood mononuclear cells obtained in 1990/1992 from 20 infected Ugandans. The PCR-amplified DNA was cloned into a phagemid vector and between 1 and 12 clones from each provirus were sequenced. 32 clones derived from 6 proviruses were sequenced and their consensus C2-V5 nucleotide sequences (approximately 720 nucleotides) were analyzed by fastDNAml program. Sequence conservation in the C2-V5 regions of env was about 75%, and most of the variations were due to length differences (234 to 243 amino acids), and most of these variations occurred in the V4 region. In the analysis of the C2-V3 regions, a total of 130 clones were sequenced and found to contain between 250 and 320 nucleotides. 4 of the sequences contained stop codons. The phylogenetic tree obtained by fastDNAml analysis of consensus nucleotide sequences for the C2-V3 regions assigned the Ugandan proviruses to 4 subtypes: 5 in global subtype A, 2 in subtype B, 1 in subtype C, and 12 in subtype D. Analysis of the deduced consensus amino acid sequences (about 81 nucleotides) of the C2-V3 regions of 41 proviruses by a maximum parsimony program gave a similar phylogenetic relationship. The data indicated that phylogenetic analysis of nucleotide and/or amino acid sequences from the C2-V3 regions is a reliable method of sub-type determination. The consensus amino acid sequence of the subtype A and D proviruses were almost identical to those of the Albert et al. group B and group A proviruses, respectively. The deduced amino acid sequences of the C2-V5 regions of 6 of these proviruses showed considerable diversity both between patients and within patients. The region varied in length between 234 and 243 amino acids and included deletions and repetitions, particularly in the V4 region.