Rev-Rev Response Element Activity Selection Bias at the HIV Transmission Bottleneck

Abstract

HIV is not efficiently transmitted between hosts, and selection of viral variants occurs during the process of sexual transmission. The factors that confer selective advantage at the transmission bottleneck remain incompletely understood. We explored whether differences in the Rev-Rev Response Element (RRE) regulatory axis of HIV affect transmission fitness, since functional variation in the Rev-RRE axis in different viral isolates has been shown to affect replication kinetics and relative expression of many HIV proteins. Single genome HIV sequences were identified from nine linked subject pairs near the time of female-to-male transmission. Using a rapid flow-cytometric assay, we found that the functional Rev-RRE activity varied significantly between isolates. Moreover, it was generally lower in recipients' viruses compared to the corresponding donor viruses. In six of nine transmission events, recipient virus Rev-RRE activity clustered at the extreme low end of the range of donor virus activity. Rev-RRE pair activity was an unpredictable product of component Rev and RRE activity variation. These data indicate selection pressure on the Rev-RRE axis during female-to-male sexual transmission. Variation in the activity of the Rev-RRE axis may permit viral adaptation to different fitness landscapes and could play an important role in HIV pathogenesis.

Figure 1.

Phylogenetic trees for individual transmission pairs. A phylogenetic tree was generated using the neighbor joining method for 401 single genome HIV sequences. The sequences of four hundred primary isolates associated with eighteen individual subjects in nine linked female-to-male HIV transmission pairs were obtained from GenBank. The laboratory strain NL4–3 was included in tree generation as an outgroup but was excluded from the figure display for clarity. Portions of the tree corresponding to the individual transmission pairs, A through I, are displayed separately. Tip symbols differentiate sequences from donors and recipients, as well as genomes containing Rev-RRE pairs that were selected or not selected for inclusion in functional assays. Horizontal bars represent nucleotide substitutions per site. See also Figure S1.

Figure 2.

Proportion of viral variants represented in functional activity assays. A subset of the unique Rev-RRE cognate pairs from primary isolates were included in the functional activity assays. For each subject, the proportion of the sequenced viral variants from that individual’s plasma containing a Rev-RRE sequence included in the functional assays is shown.

Figure 3.

Rev-RRE functional activity of viral variants from donors and recipients. The relative functional activity of selected Rev-RRE pairs from primary isolates in both donors and recipients is shown. Each HIV transmission pair, A through I, is displayed, with Rev-RRE cognate pairs from donor viral sequences on the left of each box and Rev-RRE cognate pairs from recipient viral sequences on the right. Each bubble is a unique Rev-RRE pair in the indicated individual’s quasispecies. The position of the bubble on the y-axis represents the relative level of Rev-RRE functional activity for that pair. The area of each bubble is scaled to the relative prevalence of the Rev-RRE pair sequence within the individual’s sequenced quasispecies. Activity units are multiples of the Rev-RRE activity of the laboratory strain NL4–3. D – donor, R – recipient.

Figure 4.

Contributions of Rev and the RRE to cognate pair functional activity. The relative functional activity of Rev-RRE cognate pairs from primary isolates, as well as the activity of the component Rev with NL4–3 RRE and the component RRE with NL4–3 Rev is shown. Relative activity is shown as multiples of the activity of the NL4–3 Rev/NL4–3 RRE cognate pair without units. Observations from technical replicates (n = 5 or 6) are shown as dots. Bars represent the mean value of individual observations and error bars represent standard deviation. Statistical comparisons were performed using a one-way ANOVA with adjustment for multiple comparisons using Dunnett’s T3 method. Primary isolate 1 was obtained from donor I (accession KR820312), isolate 2 was obtained from recipient B (KR820385), isolate 3 was obtained from donor G (KY112428), isolate 4 was obtained from donor C (KY112275), and isolate 5 was obtained from recipient A (JX973051). Cognate pair activity values replicate values shown in Figure 3. NS not significant, * p <0.05, ** p <0.01, *** p<0.001.