Role of donor genital tract HIV-1 diversity in the transmission bottleneck

Abstract

The predominant mode of HIV-1 infection is heterosexual transmission, where a genetic bottleneck is imposed on the virus quasispecies. To probe whether limited genetic diversity in the genital tract (GT) of the transmitting partner drives this bottleneck, viral envelope sequences from the blood and genital fluids of eight transmission pairs from Rwanda and Zambia were analyzed. The chronically infected transmitting partner's virus population was heterogeneous with distinct genital subpopulations, and the virus populations within the GT of two of four women sampled longitudinally exhibited evidence of stability over time intervals on the order of weeks to months. Surprisingly, the transmitted founder variant was not derived from the predominant GT subpopulations. Rather, in each case, the transmitting variant was phylogenetically distinct from the sampled locally replicating population. Although the exact distribution of the virus population present in the GT at the time of transmission cannot be unambiguously defined in these human studies, it is unlikely, based on these data, that the transmission bottleneck is driven in every case by limited viral diversity in the donor GT or that HIV transmission is solely a stochastic event.

Fig. 1.

Phylogenetic analysis of env V1–V4 sequences from the blood and GT in eight heterosexual transmission pairs. Env V1–V4 nucleotide sequences were aligned for each linked transmission pair, and ML trees were drawn to reveal sequence heterogeneity and homogeneity. Donor blood sequences are shown in green (dark green-filled squares, PB; light green-filled squares, PL), donor GT sequences are shown in red (red +, CA; red x, CF; red-filled circles, SW), and recipient blood sequences are shown in blue (dark blue-filled squares, PB; light blue-filled squares, PL). Highlighter analysis revealed the donor GT (red arrow) and blood sequence (green arrow) most related to the consensus recipient sequence (blue open square). Horizontal branch lengths are drawn to scale, with the bar representing 10-nt changes.

Fig. 2.

Highlighter and phylogenetic analysis reveal distinct GT populations. Aligned Env V1–V4 nucleotide sequences for transmission pairs were analyzed by the Los Alamos Highlighter tool. (A–C) Output files with Highlighter plots aligned to the phylogenetic trees. Tick marks indicate nucleotide differences from the recipient consensus sequence (blue open square). Nucleotide differences are color-coded and marked according to their genetic location along the length of V1–V4. Colors are as follows: green, A; red, T; orange, G; blue, C; gray, gaps. Arrows point to those variants in the blood (green) and GT (red) most closely related to the transmitted founder virus.

Fig. 3.

Phylogenetic analysis of longitudinal samples. Longitudinal sequences (V1–V4) from chronically infected females, blood PL and GT, were aligned, and ML phylogenetic trees were drawn for chronically infected female subjects ZM1149F (A), ZM1862F (B), ZM323F (C), and ZM1165F (D). Blood-derived sequences are shown in green, and GT-derived sequences are shown in red. Closed arrowheads in A identify identical or nearly identical sequences from both the same time points [GTDY28 time point 3 (TP3)], and open arrowheads identify identical or nearly identical sequences from different time points [GTDY14 time point 2 (TP2) and GTDY28 time point 3 (TP3)]. Four Zambian subtype C reference sequences were used to root the tree. Horizontal branch lengths are drawn to the scale shown.

Fig. P1.

Heterosexual transmission across a mucosal barrier. (A) GT-derived virus population at the time of sampling contains a large subpopulation of virus (red virions), as well as numerous other variants. In this study, only one of these variants, distinct from the large variant population, successfully crosses the mucosal epidermis barrier of the genitals (B) to generate an infection capable of spreading to all lymphoid tissues of the recipient (C).