Loss Of Sequence Information Research Articles

Mutational Disruption of cis-Acting Replication Element 2C in Coxsackievirus B3 Leads to 5'-Terminal Genomic Deletions.

ABSTRACTFollowing natural human or experimental murine infections and in cell culture, coxsackievirus B (CVB) RNA can persist for weeks in the absence of a cytopathic effect, yet viral RNA remains detectable. Our earlier studies demonstrated that this persistence produced viral RNA with up to 49 nucleotide deletions at the genomic 5′ terminus which partially degraded the cloverleaf (or domain I), an RNA structure required for efficient viral replication. A cis-acting replication element (CRE) in the 2C protein-coding region [CRE(2C)] templates the addition of two uridine residues to the virus genome-encoded RNA replication primer VPg prior to positive-strand synthesis. Because our previous work also demonstrated that the genomes of CVB with a 5′-terminal deletion (CVB-TD) have VPg covalently linked, even though they rarely terminate in the canonical UU donated by CRE(2C)-mediated uridylylation of VPg, we hypothesized that a functional (uridylylating) CRE(2C) would be unnecessary for CVB-TD replication. Using the same 16 mutations in the CVB3 CRE(2C) structure that were considered lethal for this virus by others, we demonstrate here both in infected cell cultures and in mice that wild-type (wt) and CVB3-TD strains carrying these mutations with a nonuridylylating CRE(2C) are viable. While the wt genome with the mutated CRE(2C) displays suppressed replication levels similar to those observed in a CVB3-TD strain, mutation of the CRE(2C) function in a CVB3-TD strain does not further decrease replication. Finally, we show that replication of the parental CVB3 strain containing the mutated CRE(2C) drives the de novo generation of genomic deletions at the 5′ terminus.IMPORTANCE In this report, we demonstrate that while CVB can replicate without a uridylylating CRE(2C), the replication rate suffers significantly. Further, deletions at the 5′ terminus of the genome are generated in this virus population, with this virus population supplanting the wild-type population. This demonstrates that VPg can prime without being specifically uridylylated and that this priming is error prone, resulting in the loss of sequence information from the 5′ terminus. These findings have significance when considering the replication of human enteroviruses, and we believe that these data are unattainable in a cell-free system due to the poor replication of these CRE-deficient viruses.

RNA-conjugated template-switching RT-PCR method for generating an Escherichia coli cDNA library for small RNAs

The cDNA conversion of RNA molecules is a prerequisite for their analysis. In the case of prokaryotic RNAs, cDNA conversion is difficult due to a lack of the long poly(A) tails that are found in eukaryotic mRNAs. The full cDNAs for eukaryotic mRNAs can be amplified by the reverse transcription polymerase chain reaction (RT-PCR) using the template-switching method together with an oligo(dT) primer. To amplify the full cDNAs for prokaryotic RNAs, we modified the template-switching RT-PCR method by adopting an RNA linker at the 3′ end of the target RNAs. Using this method, which we named as RNA-conjugated template-switching RT-PCR (RC/TS RT-PCR), we constructed a cDNA library for small RNAs from cold-shock-treated Escherichia coli cells. To confirm that the cDNAs were amplified by RC/TS RT-PCR without a loss of sequence information, clones carrying the 6S RNA sequence were analyzed from the cDNA library for small RNAs ranging from 130 to 350 nt. We found that the 6S RNA sequences were fully converted into the corresponding cDNAs, confirming that RC/TS RT-PCR is a useful method for constructing a cDNA library for small RNAs in E. coli. This method can be also used to construct a cDNA library for non-poly(A)-containing RNAs from eukaryotic cells.