Genetic identity, biological phenotype, and evolutionary pathways of transmitted/founder viruses in acute and early HIV-1 infection

Abstract

Identification of full-length transmitted HIV-1 genomes could be instrumental in HIV-1 pathogenesis, microbicide, and vaccine research by enabling the direct analysis of those viruses actually responsible for productive clinical infection. We show in 12 acutely infected subjects (9 clade B and 3 clade C) that complete HIV-1 genomes of transmitted/founder viruses can be inferred by single genome amplification and sequencing of plasma virion RNA. This allowed for the molecular cloning and biological analysis of transmitted/founder viruses and a comprehensive genome-wide assessment of the genetic imprint left on the evolving virus quasispecies by a composite of host selection pressures. Transmitted viruses encoded intact canonical genes (gag-pol-vif-vpr-tat-rev-vpu-env-nef) and replicated efficiently in primary human CD4(+) T lymphocytes but much less so in monocyte-derived macrophages. Transmitted viruses were CD4 and CCR5 tropic and demonstrated concealment of coreceptor binding surfaces of the envelope bridging sheet and variable loop 3. 2 mo after infection, transmitted/founder viruses in three subjects were nearly completely replaced by viruses differing at two to five highly selected genomic loci; by 12-20 mo, viruses exhibited concentrated mutations at 17-34 discrete locations. These findings reveal viral properties associated with mucosal HIV-1 transmission and a limited set of rapidly evolving adaptive mutations driven primarily, but not exclusively, by early cytotoxic T cell responses.

Figure 1.

Maximum likelihood phylogenetic tree of full-length HIV-1 genome sequences from 12 subjects. Individual sequences are shown as filled ovals. HIV-1 clade B and C reference sequences from the LANL database (http://www.hiv.lanl.gov/content/index) are shown in gray. Numerals at nodes indicate phylogenetic support for sequences included at that node; maximum likelihood bootstrap support ≥ 70% is shown in italics. Bayesian posterior probability values ≥ 0.9 are shown in bold. The scale bar indicates 5% genetic distance (5 nt differences per 100 sites).

Figure 2.

Phylogenetic and Highlighter analyses of HIV-1 sequences from subject WITO4160. WITO_fl.CON is the inferred transmitted/founder sequence. In the phylogram (left), asterisks indicate sequences that had International Union of Pure and Applied Chemistry ambiguous state assignments caused by mixed bases. Mixed bases were attributed to Taq polymerase errors and, thus, are not reflected in branch lengths of phylograms in this or other figures. The scale bar indicates a genetic distance of 0.0001 (0.01%, or 1 nt difference per genome). In the Highlighter plots (middle and right), nucleotide differences from the transmitted/founder consensus sequence are indicated by tic marks. Brackets denote insertions of single bases. Gray tics and boxes indicate a deleted sequence. The consensus (CON) is the same whether or not mixed bases are included. The middle Highlighter plot shows sequences without ambiguous (mixed) base assignments. The Highlighter plot on the right includes sequences with one or more ambiguous bases (shown as dark blue tic marks). Sequences H3 and C3 contained APOBEC-related G-to-A hypermutation. The horizontal axis indicates nucleotide positions in the HIV-1 genome. Sequences begin at nucleotide position 582 in the 5′ LTR (U5) and extend to position 9,606 in the 3′ LTR (R) based on HXB2 reference sequence numbering (http://www.hiv.lanl.gov/content/sequence/HIV/REVIEWS/HXB2.html).

Figure 3.

Phylogenetic and Highlighter analyses of HIV-1 sequences from subject ZM247F. Subject ZM247F had sequences comprising two distinct lineages that were 2.4% different from each other, as shown by the phylogram (left) and Highlighter plot (right). ZM247F_fl.CON1 is the consensus of lineage 1. The second lineage (sequences B8, E10, E11, and G11) differs from ZM247F_fl.CON1 by 213–217 nt differences, as indicated by tic marks.

Figure 4.

Phylogenetic and Highlighter analyses of HIV-1 sequences. (A and B) Analyses are from subjects CH40 (A) and CH77 (B). In Highlighter plots, the boxed tic marks indicate nucleotide polymorphisms shared between two or more sequences. Such shared mutations are also reflected by branches emanating from common nodes in the phylograms. In B, sequences are rooted on sequence A2, not on a consensus of the sequences. This is because sequence A2 was most similar to the transmitted/founder sequence, differing from it by only a single nucleotide at position 6,021 (see Results for explanation).

Figure 5.

Highlighter plots of sequences from longitudinal samples. (A–C) Samples were from subjects CH40 (A), CH58 (B), and CH77 (C). The dark line represents the transmitted/founder sequence (T/F) for each subject. Sequences from different sample dates are indicated by the grouped horizontal lines, with tic marks indicating differences from the T/F sequence. The screening sample (S) was in each case obtained from a preseroconversion Fiebig stage II sample near peak viremia. The sampling times in days after screening and corresponding viral loads are indicated (left). Light blue tic marks indicate differences in noncoding regions of the genomes. Gray tic marks indicate deletions. Green tic marks indicate differences that are synonymous in the reading frames indicated at the bottom of the figure. Red tic marks indicate differences that are nonsynonymous in a coding gene. Short horizontal lines (labeled A, B, and C) indicate short (∼1-kb) sequences from a screening sample used to verify the presence or absence of a shared polymorphism (Figs. S3–S6). The horizontal axis indicates nucleotide positions in the alignment beginning at position 582 in the 5′ LTR (U5) and extending to position 9,606 in the 3′ LTR (R) based on HXB2 reference sequence numbering (http://www.hiv.lanl.gov/content/sequence/HIV/REVIEWS/HXB2.html).

Figure 6.

Cloning strategy and functional analysis of three subtype C full-length molecular clones from subjects ZM246F and ZM247F. (A) The full-length transmitted/founder proviral sequence for ZM246F was chemically synthesized in three overlapping fragments. The two transmitted/founder proviruses in subject ZM247F were PCR amplified from PBMC DNA in two overlapping halves. Proviral fragments were cloned into the indicated plasmid vectors. (B) Plasmid DNAs were transfected into 293T cells, and progeny viruses were examined for their ability to replicate in activated primary human CD4⁺ lymphocytes (top) and monocyte-derived macrophages (bottom) from the same donor. Culture supernatants were analyzed at days 1–16 for p24 production. The results shown were replicated four times in lymphocytes and macrophages from four different normal blood donors.