Near full-length HIV-1 subtype B sequences from the early South African epidemic, detecting a BD unique recombinant form (URF) from a sample in 1985

Abstract

HIV-1 subtype C is the most prevalent subtype in South Africa. Although subtype B was previously detected in South Africa, there is limited sequence information available. We characterized near full-length HIV-1 subtype B sequences from samples collected at the start of the South African HIV-1 epidemic, in the 1980s. Five samples were analysed by PCR amplification, Sanger DNA sequencing and phylogenetic analyses. The viral genomes were amplified in two overlapping fragments of 5.5 kb and 3.7 kb. The sequences were subtyped using REGA version 3.0, RIP version 3.0 and jpHMM. Maximum Likelihood phylogenetic trees were inferred with MEGA version 6. Four HIV-1 patient sequences were subtyped as pure HIV-1 subtype B. One sequence was characterized as a novel HIV-1 subtype B and D recombinant. The sequences clustered phylogenetically with other HIV-1 subtype B sequences from South Africa, Europe and the USA. We report the presence of an HIV-1 subtype B and D recombinant strain detected in the beginning of the epidemic. This indicates that viral recombination events were already happening in 1985, but could have been missed as sequence analyses were often limited to small genomic regions of HIV-1.

Figure 1

HIV-1 near full-length genome (NFLG) molecular phylogenetic analysis inferred by the maximum likelihood (ML) method. The ML phylogenetic tree inferred in MEGA version 6 contains the five new HIV-1 sequences (■), HIV-1 subtype B and D reference sequences, previously characterized sequences from South Africa and the HIV-1 Group M reference subtypes. The evolutionary distances were inferred using the general time reverse (GTR) model of nucleic acid substitution with an estimated Gamma shape parameter and invariant sites. The genetic distance is displayed in the scale bar at the bottom of the figure. Bootstrap values higher than 70% are indicated.

Figure 2

Subtyping analyses of the unique recombinant form (URF) of R605, a BD recombinant sequence according to jumping profile Hidden Markov Model (jpHMM), LANL and phylogenetic analyses. Phylogenetic analyses (ML tree) according to breakpoint located at position (862–3918 bp) in the gag - pol region clusters amongst subtype B sequences. ML tree according to breakpoint position (3918–5974 bp) in the pol – vif region clusters amongst subtype D sequences. ML tree according to breakpoint position (5974–9187 bp) in the vpu – env region clusters amongst subtype B sequences. In order to validate the sequences a total of 100 bootstrap replicates was carried to infer a ML phylogenetic tree.

Figure 3

The amplification strategy. The diagram shows amplification for whole genome. Full genome amplification was achieved by amplifying DNA in two fragments. Four primers were used for each fragment utilising pre-nested and nested PCR. Fragment 1 consisted of genes starting from gag to vpu and was approximately 5.5 Kb in length. The second fragment of genes starting from vpr until the 3′LTR and was approximately 3.7 kb in length.