Recombinant viruses initiated the early HIV-1 epidemic in Burkina Faso

Abstract

We analyzed genetic diversity and phylogenetic relationships among 124 HIV-1 and 19 HIV-2 strains in sera collected in 1986 from patients of the state hospital in Ouagadougou, Burkina Faso. Phylogenetic analysis of the HIV-1 env gp41 region of 65 sequences characterized 37 (56.9%) as CRF06_cpx strains, 25 (38.5%) as CRF02_AG, 2 (3.1%) as CRF09_cpx, and 1 (1.5%) as subtype A. Similarly, phylogenetic analysis of the protease (PR) gene region of 73 sequences identified 52 (71.2%) as CRF06_cpx, 15 (20.5%) as CRF02_AG, 5 (6.8%) as subtype A, and 1 (1.4%) was a unique strain that clustered along the B/D lineage but basal to the node connecting the two lineages. HIV-2 PR or integrase (INT) groups A (n = 17 [89.5%]) and B (n = 2 [10.5%]) were found in both monotypic (n = 11) and heterotypic HIV-1/HIV-2 (n = 8) infections, with few HIV-2 group B infections. Based on limited available sampling, evidence suggests two recombinant viruses, CRF06_cpx and CRF02_AG, appear to have driven the beginning of the mid-1980s HIV-1 epidemic in Burkina Faso.

Figure 1. Neighbor-joining tree of HIV-1 env gp41 (n = 65) and protease (73) sequences collected in 1986.

The Burkino Faso sequences are identified by the prefix, BF. Sequences from mixed HIV-1/HIV-2 infections are highlighted in gray. The reference HIV-1 group M and CRF sequences, respectively, have prefixes A, B, C, D, F1, F2, G, H, J, K, and potential L to denote the subtype or sub-subtype, or number indicating which CRF they represent, i.e., 1 for CRF01, 2 for CRF02, etc. The position of the outgroup (SIV-cpz sequence) is not shown. The numbers on nodal branches represent the bootstrap values (out of 100 replicates); only values 68% or greater are shown. The scale bar indicates an evolutionary distance of 0.1 nucleotides per site. Vertical distances are for clarity only.

Figure 2. Neighbor-joining tree of 18 HIV-2protease sequences; the prefix BF represents sequences recovered from Burkina Faso patients in 1986.

For comparison, 7 1980s sequences from Ivory Coast were included in the tree and contain the prefixes IC. Highlighted sequences in gray were from mixed HIV-1/HIV-2 infections; the remainders were from monotypic HIV-2 infections; groups #1 and #2 within subtype A represent distinct subclusters of closely related sequences. Viruses from sooty mangabeys are shown (SIVsm). The position of the outgroup (SIV-cpz sequence) is not shown. The number on the node indicates the bootstrap value; only values 68% or greater are shown. The scale bar indicates an evolutionary distance of 0.1 nucleotides per site. Vertical distances are for clarity only.