Evidence for a Mechanism of Recombination during Reverse Transcription Dependent on the Structure of the Acceptor RNA

Abstract

Genetic recombination is a major force driving retroviral evolution. In retroviruses, recombination proceeds mostly through copy choice during reverse transcription. Using a reconstituted in vitro system, we have studied the mechanism of strand transfer on a major recombination hot spot we previously identified within the genome of HIV-1. We show that on this model sequence the frequency of copy choice is strongly influenced by the folding of the RNA template, namely by the presence of a stable hairpin. This structure must be specifically present on the acceptor template. We previously proposed that strand transfer follows a two-step process: docking of the nascent DNA onto the acceptor RNA and strand invasion. The frequency of recombination under copy choice conditions was not dependent on the concentration of the acceptor RNA, in contrast with strand transfer occurring at strong arrests of reverse transcription. During copy choice strand transfer, the docking step is not rate limiting. We propose that the hairpin present on the acceptor RNA could mediate strand transfer following a mechanism reminiscent of branch migration during DNA recombination.

The Structure of HIV-1 Genomic RNA in the gp120 Gene Determines a Recombination Hot Spot in Vivo

“…We first used the same region of homology of C2 RNA. The presence of distinctive point mutations on the donor and acceptor RNAs in this region of homology allows a more precise mapping of the hot region than the one we previously performed on C2 in vitro (10). As shown in Fig.…”

“…The amplified product was purified on agarose gel, digested with SacII and BamHI, ligated into an appropriate plasmid vector, and used for transformation of E. coli. Plating on IPTG/X-Gal-containing dishes allowed blue/white screening of recombinant and parental colonies, respectively (10,21). To determine the regions where strand transfer occurred, 48 recombinant clones were analyzed by restriction mapping for each assay, as described under "Results.…”

“…We previously showed that strand transfer in vitro was decreased by a 4-fold factor in the hairpin when regions R1 and R5 were replaced by two sequences derived from the gag gene of the LAI isolate (10). This was attributed to the lack of complementarity between the sequences that constituted the base of the C2 hairpin that, as determined by in vitro probing experiments, abolished the formation of this hairpin (10). Here, we followed the same strategy generating Delhp genomic RNA (for "deleted hairpin," Fig.…”

“…2A. This sequence spans nt 6,639 -7,039 numbered according to HXB2 proviral DNA and includes the portion coding for the C2 region of the gp120 we previously identified as a recombination hot spot in a cell-free system (10). In three independent experiments the average frequency of recombination was 13.4% (standard deviation: Ϯ 0.4) in "heterozygous" samples, compared with 0.5% (Ϯ 0.2) observed in "homozygous" samples (Table I).…”

Cellular factors influencing Semliki Forest Virus vector biology

“…It is also possible that the recombination events arise through template switching promoted by the replicase, as this is the case for retroviruses. 66,67 The use of split genomes, together with the introduction of conditional mutations into structural genes, has strongly reduced the likelihood of these recombination events. 23,68 However, a risk still persists.…”