Diverse Mechanisms of RNA Recombination

Recombination is widespread among RNA viruses, but many molecular mechanisms of this phenomenon are still poorly understood. It was believed until recently that the only possible mechanism of RNA recombination is replicative template switching, with synthesis of a complementary strand starting on one viral RNA molecule and being completed on another. The newly synthesized RNA is a primary recombinant molecule in this case. Recent studies have revealed other mechanisms of replicative RNA recombination. In addition, recombination between the genomes of RNA viruses can be nonreplicative, resulting from a joining of preexisting parental molecules. Recombination is a potent tool providing for both the variation and conservation of the genome in RNA viruses. Replicative and nonreplicative mechanisms may contribute differently to each of these evolutionary processes. In the form of trans splicing, nonreplicative recombination of cell RNAs plays an important role in at least some organisms. It is conceivable that RNA recombination continues to contribute to the evolution of DNA genomes.__________Translated from Molekulyarnaya Biologiya, Vol. 39, No. 4, 2005, pp. 618–632.Original Russian Text Copyright © 2005 by Gmyl, Agol.     

RNA recombination in a coronavirus: recombination between viral genomic RNA and transfected RNA fragments.

Mouse hepatitis virus (MHV), a coronavirus, has been shown to undergo a high frequency of RNA recombination both in tissue culture and in animal infection. So far, RNA recombination has been demonstrated only between genomic RNAs of two coinfecting viruses. To understand the mechanism of RNA recombination and to further explore the potential of RNA recombination, we studied whether recombination could occur between a replicating MHV RNA and transfected RNA fragments. We first used RNA fragments which represented the 5' end of genomic-sense sequences of MHV RNA for transfection. By using polymerase chain reaction amplification with two specific primers, we were able to detect recombinant RNAs which incorporated the transfected fragment into the 5' end of the viral RNA in the infected cells. Surprisingly, even the anti-genomic-sense RNA fragments complementary to the 5' end of MHV genomic RNA could also recombine with the MHV genomic RNAs. This observation suggests that RNA recombination can occur during both positive- and negative-strand RNA synthesis. Furthermore, the recombinant RNAs could be detected in the virion released from the infected cells even after several passages of virus in tissue culture cells, indicating that these recombinant RNAs represented functional virion RNAs. The crossover sites of these recombinants were detected throughout the transfected RNA fragments. However, when an RNA fragment with a nine-nucleotide (CUUUAUAAA) deletion immediately downstream of a pentanucleotide (UCUAA) repeat sequence in the leader RNA was transfected into MHV-infected cells, most of the recombinants between this RNA and the MHV genome contained crossover sites near this pentanucleotide repeat sequence. In contrast, when exogenous RNAs with the intact nine-nucleotide sequence were used in similar experiments, the crossover sites of recombinants in viral genomic RNA could be detected at more-downstream sites. This study demonstrated that recombination can occur between replicating MHV RNAs and RNA fragments which do not replicate, suggesting the potential of RNA recombination for genetic engineering.     

Group II intron RNA-catalyzed recombination of RNA in vitro.

We report the first evidence for a novel reaction mediated by the self-splicing yeast mitochondrial group II intron bl1; the site-specific recombination of RNA molecules in vitro. Upon incubation of the intron lariat with two different RNAs, each harbouring a short sequence complementary to exon binding site 1 (EBS1) of the intron, novel recombined RNAs are formed. As a result of this intron-mediated shuffling of gene segments, the 5' part of RNA1 is ligated to the 3' part of RNA2 and, reciprocally, the 5' part of RNA2 to the 3' part of RNA1. Sequence analysis of the recombinant junction shows that the site of recombination is precisely located 3' to intron binding site 1 (IBS1). The hypothesized mechanism of recombination involves exchange of RNA 5' parts after the first step of a reverse splicing reaction. The possible role of this mechanism in vivo and during prebiotic evolution is discussed.     

Generation and analysis of nonhomologous RNA-RNA recombinants in brome mosaic virus: sequence complementarities at crossover sites.

All three single-stranded RNAs of the brome mosaic virus (BMV) genome contain a highly conserved, 193-base 3' noncoding region. To study the recombination between individual BMV RNA components, barley plants were infected with a mixture of in vitro-transcribed wild-type BMV RNAs 1 and 2 and an RNA3 mutant that carried a deletion near the 3' end. This generated a population of both homologous and nonhomologous 3' recombinant BMV RNA3 variants. Sequencing revealed that these recombinants were derived by either single or double crossovers with BMV RNA1 or RNA2. The primary sequences at recombinant junctions did not show any similarity. However, they could be aligned to form double-stranded heteroduplexes. This suggested that local hybridizations among BMV RNAs may support intermolecular exchanges.     

RNA-RNA recombination in Sindbis virus: roles of the 3' conserved motif, poly(A) tail, and nonviral sequences of template RNAs in polymerase recognition and template switching.

Sindbis virus (SIN), a mosquito-transmitted animal RNA virus, carries a 11.7-kb positive-sense RNA genome which is capped and polyadenylated. We recently reported that the SIN RNA-dependent RNA polymerase (RdRp) could initiate negative-strand RNA synthesis from a 0.3-kb 3'-coterminal SIN RNA fragment and undergo template switching in vivo (M. Hajjou, K. R. Hill, S. V. Subramaniam, J. Y. Hu, and R. Raju, J. Virol. 70:5153-5164, 1996). To identify and characterize the viral and nonviral sequences which regulate SIN RNA synthesis and recombination, a series of SIN RNAs carrying altered 3' ends were tested for the ability to produce infectious virus or to support recombination in BHK cells. The major findings of this report are as follows: (i) the 3'-terminal 20-nucleotides (nt) sequence along with the abutting poly(A) tail of the SIN genome fully supports negative-strand synthesis, genome replication, and template switching; (ii) a full-length SIN RNA carrying the 3'-terminal 24 nt but lacking the poly(A) tail is noninfectious; (iii) SIN RNAs which carry 3' 64 nt or more without the poly(A) tail are infectious and regain their poly(A) tail in vivo; (iv) donor templates lacking the poly(A) tail do not support template switching; (v) full-length SIN RNAs lacking the poly(A) tail but carrying 3' nonviral extensions, although debilitated to begin with, evolve into rapidly growing poly(A)-carrying mutants; (vi) poly(A) or poly(U) motifs positioned internally within the acceptor templates, in the absence of other promoter elements within the vicinity, do not induce the jumping polymerase to reinitiate at these sites; and (vii) the junction site selection on donor templates occurs independently of the sequences around the acceptor sites. In addition to furthering our understanding of RNA recombination, these studies give interesting clues as to how the alphavirus polymerase interacts with its 3' promoter elements of genomic RNA and nonreplicative RNAs. This is the first report that an in vitro-synthesized alphavirus RNA lacking a poly(A) tail can initiate infection and produce 3' polyadenylated viral genome in vivo.     

Suppression of viral RNA recombination by a host exoribonuclease

RNA viruses of humans, animals, and plants evolve rapidly due to mutations and RNA recombination. A previous genome-wide screen in Saccharomyces cerevisiae, a model host, identified five host genes, including XRN1, encoding a 5'-3' exoribonuclease, whose absence led to an approximately 10- to 50-fold enhancement of RNA recombination in Tomato bushy stunt virus (E. Serviene, N. Shapka, C. P. Cheng, T. Panavas, B. Phuangrat, J. Baker, and P. D. Nagy, Proc. Natl. Acad. Sci. USA 102:10545-10550, 2005). In this study, we found abundant 5'-truncated viral RNAs in xrn1delta mutant strains but not in the parental yeast strains, suggesting that these RNAs might serve as recombination substrates promoting RNA recombination in xrn1delta mutant yeast. This model is supported by data showing that an enhanced level of viral recombinant accumulation occurred when two different 5'-truncated viral RNAs were expressed in the parental and xrn1delta mutant yeast strains or electroporated into plant protoplasts. Moreover, we demonstrate that purified Xrn1p can degrade the 5'-truncated viral RNAs in vitro. Based on these findings, we propose that Xrn1p can suppress viral RNA recombination by rapidly removing the 5'-truncated RNAs, the substrates of recombination, and thus reducing the chance for recombination to occur in the parental yeast strain. In addition, we show that the 5'-truncated viral RNAs are generated by host endoribonucleases. Accordingly, overexpression of the Ngl2p endoribonuclease led to an increased accumulation of cleaved viral RNAs in vivo and in vitro. Altogether, this paper establishes that host ribonucleases and host-mediated viral RNA turnover play major roles in RNA virus recombination and evolution.     

Nonhomologous RNA-RNA recombination events at the 3' nontranslated region of the Sindbis virus genome: hot spots and utilization of nonviral sequences.

The mechanism of RNA-RNA recombination at the 3' nontranslated region (3'NTR) of the Sindbis virus (SIN) genome was studied by using nonreplicative RNA precursors. The 11.7-kb SIN genome was transcribed in vitro as two nonoverlapping RNA fragments. RNA-1 contained the entire 11.4-kb protein coding sequence of SIN and also carried an additional 1.8-kb nonviral sequence at its 3' end. RNA-2 carried the remaining 0.26 or 0.3 kb of the SIN genome containing the 3'NTR. Transfection of these two fragments into BHK cells resulted in vivo RNA-RNA recombination and release of infectious SIN recombinants. Eighteen plaque-purified recombinant viruses were sequenced to precisely map the RNA-RNA crossover sites at the 3'NTR. Sixteen of the 18 recombinants were found to be genetically heterogeneous at the 3'NTR. Two major clustered sites within the 3'NTR of RNA-2 were found to be fused to multiple locations on the nonviral sequence of RNA-1, resulting in insertions of 10 to 1,085 nucleotides at the 3'NTR. Sequence analysis of crossover sites suggested only limited homology and heteroduplex-forming capability between substrate RNAs. Analysis of additional 23 recombinant viruses generated by mutagenized donor and acceptor templates supports the occurrence of recombination hot spots on donor templates. Introduction of a 17-nucleotide rudimentary replicase recognition signal in the acceptor template alone did not induce the polymerase to reinitiate at the 17-nucleotide signal. Interestingly, deletion of a 24-nucleotide hot spot locus on the donor template abolished crossover events at one of the two sites and allowed the polymerase to reinitiate at the 17-nucleotide replicase recognition signal inserted at the acceptor template. The possible roles of RNA-protein and RNA-RNA interactions in the differential regulation of apparent pausing, template selection, and reinitiation are discussed.     

Lsm proteins bind and stabilize RNAs containing 5' poly(A) tracts

Many orthopoxvirus messenger RNAs have an unusual nontemplated poly(A) tract of 5 to 40 residues at the 5' end. The precise function of this feature is unknown. Here we show that 5' poly(A) tracts are able to repress RNA decay by inhibiting 3'-to-5' exonucleases as well as decapping of RNA substrates. UV cross-linking analysis demonstrated that the Lsm complex associates with the 5' poly(A) tract. Furthermore, recombinant Lsm1-7 complex specifically binds 5' poly(A) tracts 10 to 21 nucleotides in length, consistent with the length of 5' poly(A) required for stabilization. Knockdown of Lsm1 abrogates RNA stabilization by the 5' poly(A) tract. We propose that the Lsm complex simultaneously binds the 3' and 5' ends of these unusual messenger RNAs and thereby prevents 3'-to-5' decay. The implications of this phenomenon for cellular mRNA decay are discussed.     

At least 104 nucleotides are transposed from the 5' terminus of the avian sarcoma virus genome to the 5' termini of smaller viral mRNAs.

Cells producing avian sarcoma virus (ASV) contain at least three virus-specific mRNAs, two of which are encoded within the 3' half of the viral genome. Each of these viral RNAs can hybridize with single-stranded DNA(cDNA5') that is complementary to a sequence of 101 nucleotides found at the 5' terminus of the ASV genome, but not within the 3' half of the genome. We proposed previously (Weiss, Varmus and Bishop, 1977) that this nucleotide sequence may be transposed to the 5' termini of viral mRNAs during the genesis of these RNAs. We now substantiate this proposal by reporting the isolation and chemical characterization of the nucleotide sequences complementary to cDNA5' in the genome and mRNAs of the Prague B strain of ASV. We isolated the three identified classes of ASVmRNA (38, 28 and 21S) by molecular hybridization; each class of RNA contained a "capped" oligonucleotide identical to that found at the 5' terminus of the ASV genome. When hybridized with cDNA5', each class of RNA gave rise to RNAase-resistant duplex hybrids that probably encompassed the full extent of cDNA5'. The molar yields of duplex conformed approximately to the number of virus-specific RNA molecules in the initial samples; hence most if not all of the molecules of virus-specific RNA could give rise to the duplexes. The duplexes prepared from the various RNAs all contained the capped oligonucleotide found at the 5' terminus of the viral genome and had identical "fingerprints" when analyzed by two-dimensional fractionation following hydrolysis with RNAase T1. In contrast, RNA representing the 3' half of the ASV genome did not form hybrids with cDNA5'. We conclude that a sequence of more than 100 nucleotides is transposed from the 5' end of the ASV genome to the 5' termini of smaller viral RNAs during the genesis of these RNAs. Transposition of nucleotide sequences during the production of mRNA has now been described for three families of animal viruses and may be a common feature of mRNA biogenesis in eucaryotic cells. The mechanism of transposition, however, and the function of the transposed sequences are not known.     

Efficient system of homologous RNA recombination in brome mosaic virus: sequence and structure requirements and accuracy of crossovers.

Brome mosaic virus (BMV), a tripartite positive-stranded RNA virus of plants engineered to support intersegment RNA recombination, was used for the determination of sequence and structural requirements of homologous crossovers. A 60-nucleotide (nt) sequence, common between wild-type RNA2 and mutant RNA3, supported efficient repair (90%) of a modified 3' noncoding region in the RNA3 segment by homologous recombination with wild-type RNA2 3' noncoding sequences. Deletions within this sequence in RNA3 demonstrated that a nucleotide identity as short as 15 nt can support efficient homologous recombination events, while shorter (5-nt) sequence identity resulted in reduced recombination frequency (5%) within this region. Three or more mismatches within a downstream portion of the common 60-nt RNA3 sequence affected both the incidence of recombination and the distribution of crossover sites, suggesting that besides the length, the extent of sequence identity between two recombining BMV RNAs is an important factor in homologous recombination. Site-directed mutagenesis of the common sequence in RNA3 did not reveal a clear correlation between the stability of predicted secondary structures and recombination activity. This indicates that homologous recombination does not require similar secondary structures between two recombining RNAs at the sites of crossovers. Nearly 20% of homologous recombinants were imprecise (aberrant), containing either nucleotide mismatches, small deletions, or small insertions within the region of crossovers. This implies that homologous RNA recombination is not as accurate as proposed previously. Our results provide experimental evidence that the requirements and thus the mechanism of homologous recombination in BMV differ from those of previously described heteroduplex-mediated nonhomologous recombination (P. D. Nagy and J. J. Bujarski, Proc. Natl. Acad. Sci. USA 90:6390-6394, 1993).     

Inhibition of HIV derived lentiviral production by TAR RNA binding domain of TAT protein

Background

Retroviruses have a diploid genome and recombine at high frequency. Recombinant proviruses can be generated when two genetically different RNA genomes are packaged into the same retroviral particle. It was shown in several studies that recombinant proviruses could be generated in each round of HIV-1 replication, whereas the recombination rates of SNV and Mo-MuLV are 5 to 10-fold lower. The reason for these differences is not clear. One possibility is that these retroviruses may differ in their ability to copackage genomic RNAs produced at different chromosomal loci.

Results

To investigate whether there is a difference in the efficiency of heterodimer formation when two proviruses have the same or different chromosomal locations, we introduced two different Mo-MuLV-based retroviral vectors into the packaging cell line using either the cotransfection or sequential transfection procedure. The comparative study has shown that the frequency of recombination increased about four-fold when the cotransfection procedure was used. This difference was not associated with possible recombination of retroviral vectors during or after cotransfection and the ratios of retroviral virion RNAs were the same for two variants of transfection.

Conclusions

The results of this study indicate that a mechanism exists to enable the preferential copackaging of Mo-MuLV genomic RNA molecules that are transcribed on the same DNA template. The properties of Mo-MuLV genomic RNAs transport, processing or dimerization might be responsible for this preference. The data presented in this report can be useful when designing methods to study different aspects of replication and recombination of a diploid retroviral genome.