Accurately measuring recombination between closely related HIV-1 genomes

Abstract

Retroviral recombination is thought to play an important role in the generation of immune escape and multiple drug resistance by shuffling pre-existing mutations in the viral population. Current estimates of HIV-1 recombination rates are derived from measurements within reporter gene sequences or genetically divergent HIV sequences. These measurements do not mimic the recombination occurring in vivo, between closely related genomes. Additionally, the methods used to measure recombination make a variety of assumptions about the underlying process, and often fail to account adequately for issues such as co-infection of cells or the possibility of multiple template switches between recombination sites. We have developed a HIV-1 marker system by making a small number of codon modifications in gag which allow recombination to be measured over various lengths between closely related viral genomes. We have developed statistical tools to measure recombination rates that can compensate for the possibility of multiple template switches. Our results show that when multiple template switches are ignored the error is substantial, particularly when recombination rates are high, or the genomic distance is large. We demonstrate that this system is applicable to other studies to accurately measure the recombination rate and show that recombination does not occur randomly within the HIV genome.

Figure 1. Multiple template switches change the observable recombination rate.

(A) Recombination occurs during reverse transcription when RT switches from one co-packaged RNA template to another. Marker sites in one RNA template allow the detection of recombination. However, the exact number of template switches cannot be known. That is, recombination is only observed with any odd number of template switches and recombination is not observed with zero, or any even number of template switches. (B) The probability of observing a recombination crossover changes with recombination rate (measured as “recombination events per nucleotide per round of infection [REPN]) and the length over which recombination is observed. For each length (length in nucleotides of RNA shown below each line), each recombination rate produces a unique probability of observing a recombination. (C) Profiles a snapshot of the probability of observing a recombination over different lengths with a constant recombination rate 0.001 REPN. (D and E) Using a crude formula for calculating the recombination rate (r = c/nl, where c is the number of template switches detected, n is the number of sequences, and l is the distance between marker sites) that does not take into account multiple template switches underestimates the actual rate. This error increases with genomic distance and recombination rate. The probability of observing recombination is calculated with the Poisson approximation derived in Materials and Methods (Equation A).

Figure 2. Schematic representation of the marker recombination system.

(A) Marker points were introduced by genetic changes that do not alter the amino acid sequence. (B) The distances between marker points within gag (C) Transfection-induced recombination was measured by direct sequencing of plasmid DNA extracted from co-transfected 293T cells (D) PCR-induced recombination was measured by separately infecting T-cells with homozygous virions derived from single transfections of WT and MK plasmid. Lysates were mixed before PCR. (E) Inter-virion recombination and PCR-induced recombination was measured by infecting T-cells with homozygous virions derived from single transfections of WT and MK plasmid. (F) The recombination rate in T-cells was measured by infection of T-cells with virus produced by co-transfection of 293T cells with WT and MK plasmid. (D,E,F) Recombination rates were measured by PCR of cellular lysates, cloning and sequencing.

Figure 3. HIV recombination varies across genome.

An optimal recombination rate is calculated by minimizing the chi-square value between the observed and expected frequencies of detectable template switches in each of the regions. A chi-square goodness of fit test indicates whether recombination rates are likely to vary over the entire length. (A) Variation between expected and observed frequencies (B) A dual recombination rate model was fitted and marker point 4 was the optimal location for a recombination rate switch. (C, D) In a second independent experiment we also observe that recombination is higher towards marker region 1 and lower towards marker region 6. However an F-test comparing a single recombination rate model (C) and a dual recombination rate model (D) with the switch location known from the original experiment, still did not achieve significance (p = 0.068, F-test).

Figure 4. Frequencies of multiple detected template switches follows a modified Poisson distribution.

The experimental assumption of an equal recombination rate amongst all sequences predicts that the frequency of multiple observed recombination events should be distributed as calculated by rate of recombination in each region, number of sequences and lengths over which recombination is measured. Large variation from the expected distribution indicates that this assumption may not be correct. Data shown is from the two heterozygous infection datasets combined (310 total sequences) corrected for control recombination (PCR).