Highly divergent patterns of genetic diversity and evolution in proviral quasispecies from HIV controllers

Abstract

Background: Ongoing intra-host HIV-1 evolution has been shown in individuals that naturally suppress the viremia to low levels (HIV controllers) by the analysis of the RNA in plasma compartment. Detection of evolution at the DNA proviral compartment in HIV controllers, however, has been more challenging and the precise correlation between the systemic viral suppression level and rate of reservoir's reseeding in those individuals is not fully understood. In this sense, we examined the proviral DNA quasispecies by single genome amplification of the env gene in a cohort of 23 HIV controllers from Brazil, divided in three groups, according to the level of systemic viral suppression: (1) elite controllers with persistent undetectable viral load (PEC, n = 6); (2) elite controllers with occasional episodes of transient (51-400 copies/mL) viremia (EEC, n = 7); and (3) viremic controllers with persistent low-level (80-2000 copies/mL) viremia (VC, n = 10).

Results: The HIV-1 diversity of the PBMC-associated proviral quasispecies in EC was significantly (P < 0.01) lower than in VC, but not significantly different between PEC and EEC groups. We detected a considerable variation in the average pairwise nucleotide distance and proportion of unique sequences in the HIV-1 proviral quasispecies of PEC and EEC. Some PEC and EEC displayed highly homogenous proviral populations with large clusters of identical sequences, while others exhibited relatively diverse proviral populations with a high proportion of unique sequences comparable to VC subjects. The long-term (10-15 years) follow-up of the HIV-1 proviral populations revealed a complete evolutionary stasis in one PEC and measurable divergence rates in one EEC [3.1 (1.2-5.6) × 10^-3 substitutions/site/year and one VC [2.9 (0.7-5.1) × 10^-3 substitutions/site/year].

Conclusions: There is no simple relationship between systemic viral suppression and intra-host proviral diversity or rate of reservoir's reseeding in chronically infected HIV controllers. Our results demonstrate that very divergent patterns of intra-host viral diversity and divergence could be detected in the setting of natural suppression of HIV-1 replication and that ongoing evolution and reseeding of the PBMC proviral reservoir occurs in some elite controllers.

Keywords: Diversity; Elite controllers; Evolution; HIV-1; Reseeding; Reservoir; Viremic controllers.

Fig. 1

Clinical follow-up of the 23 HIV-1 controllers. Plasma RNA viral load (copies/mL, circles) and CD4⁺ T cell counts (cells/µL, squares) values over time (years) are shown on the left and right Y axis respectively. RNA viral loads below or above the detection limit are colored black and red, respectively. The limit of detection of RNA viral load varied over time according to the methodology used. Shaded areas indicate the time points selected in this study (gray) and previously (green) [17] for the DNA quasispecies analysis. Patient identification is shown in the upper left corner of each graph

Fig. 2

ML phylogenetic tree of env sequences from HIV-1 controllers and HIV-1 subtype reference sequences. Branches were colored according to the subtype assignment as shown in the legend at bottom left. The individual’s identification is displayed on the right side of the clusters. Sequence clusters from dual infected individuals (EEC09, VC06 and VC32) are indicated by shaded boxes. Bootstrap support for each individual cluster is shown. Black circles point to the reference sequences and asterisks highlight the sequences with APOBEC3G-mediated G to A hypermutations. Horizontal branch lengths are proportional to the bar at the bottom indicating nucleotide substitutions per site

Fig. 3

Mean nucleotide diversity (π, a) and normalized Shannon entropy (H _SN, b) of proviral env quasispecies from EC and VC. The colors of the circles represent the different levels of systemic viral suppression in HIV-1 controllers as indicated in the legend. Dotted and continuous gray lines represent the median and interquartile ranges, respectively. P values were calculated using the Mann–Whitney test

Fig. 4

Identification of divergent patterns of intra-host viral diversity among EC subjects. The normalized Shannon entropy (H _SN , y axis) of each subject’s proviral env quasispecies was plotted against the corresponding mean nucleotide diversity (π, x axis). The colors of the circles represent the different levels of systemic viral suppression in HIV-1 controllers as indicated in the legend. Dashed circles identify the two main patterns of intra-host viral diversity: proviral quasispecies of individuals from group 1 (G1, dashed orange line) displayed low diversity (π < 1%) and a high proportion of identical sequences (H _SN < 0.6); proviral quasispecies of individuals from group 2 (G2, dashed green line) exhibited larger diversity (π ≥ 2%) and a higher frequency of unique sequences (H _SN > 0.7)

Fig. 5

Longitudinal analysis of HIV-1 proviral env sequences obtained from subjects PEC52 (a, d), EEC42 (b, e) and VC05 (c, f) between 2000–2013, 2000–2012 and 2000–2011, respectively. a–c ML phylogenetic trees for each individual are shown, in which horizontal branch lengths are drawn to scale with the bar at the bottom indicating nucleotide substitutions per site. d–f Plots of the root-to-tip distance against sequence sampling time are shown below each subject tree. The slope, coefficient of regression and X intercept of linear regression analysis is indicated. The colors of the circles in phylogenetic trees and plots represent the sampling year at which sequences were obtained