ORF57 Overcomes the Detrimental Sequence Bias of Kaposi's Sarcoma-Associated Herpesvirus Lytic Genes

Abstract

Kaposi's sarcoma-associated herpesvirus (KSHV) encodes ORF57, which enhances the expression of intronless KSHV genes on multiple posttranscriptional levels. However, it remains elusive how ORF57 recognizes viral RNAs. Here, we demonstrate that ORF57 also increases the expression of the multiple intron-containing K15 gene. The nucleotide bias of the K15 cDNA revealed an unusual high AT content. Thus, we optimized the K15 cDNA by raising the frequency of GC nucleotides, yielding an ORF57-independent version. To further prove the importance of the sequence bias of ORF57-dependent RNAs, we grouped KSHV mRNAs according to their AT content and found a correlation between AT-richness and ORF57 dependency. More importantly, latent genes, which have to be expressed in the absence of ORF57, have a low AT content and are indeed ORF57 independent. The nucleotide composition of K15 resembles that of HIV gag, which cannot be expressed unless RNA export is facilitated by the HIV Rev protein. Interestingly, ORF57 can partially rescue HIV Gag expression. Thus, the KSHV target RNAs of ORF57 and HIV gag RNA may share certain motifs based on the nucleotide bias. A bioinformatic comparison between wild-type and sequenceoptimized K15 revealed a higher density for hnRNP-binding motifs in the former. We speculate that binding of particular hnRNPs to KSHV lytic transcripts is the prerequisite for ORF57 to enhance their expression. IMPORTANCEThe mostly intronless genes of KSHV are only expressed in the presence of the viral regulator protein ORF57, but how ORF57 recognizes viral RNAs remains elusive. We focused on the multiple intron-containing KSHV gene K15 and revealed that its expression is also increased by ORF57. Moreover, sequences in the K15 cDNA mediate this enhancement. The quest for a target sequence or a response element for ORF57 in the lytic genes was not successful. Instead, we found the nucleotide bias to be the critical determinant of ORF57 dependency. Based on the fact that ORF57 has only a weak affinity for nucleic acids, we speculate that a cellular RNA-binding protein provides the sequence preference for ORF57. This study provides evidence that herpesviral RNA regulator proteins use the sequence bias of lytic genes and the resulting composition of the viral mRNP to distinguish between viral and cellular mRNAs. K aposi's sarcoma-associated herpesvirus (KSHV) is a human gammaherpesvirus that was identified in KS tissues from AIDS patients (1). It is also associated with two lymphoproliferative disorders, the primary effusion lymphoma and the plasma cell variant of multicentric Castleman's disease. KSHVassociated malignancies are rare in immunocompetent people but more common in the context of immunosuppression, especially in HIV patients or transplant recipients. The viral life cycle is divided into a latent and a lytic phase. In contrast to other herpesviruses, tumorigenesis is also a consequence of the lytic cycle (2-5).Herpesviruses replicate in the nucleus, and viral gene expression is dependent on the host ...

The KSHV RNA regulator ORF57: target specificity and its role in the viral life cycle

“…Screening the target RNAs for the previously identified ORE was not successful . This is consistent with a similar bioinformatic analysis performed by our group which also failed to identify putative OREs in other ORF57 dependent genes . In sum, both studies extend the list of viral and cellular ORF57 targets, but the exact mode of recruitment still remains unclear.…”

“…In our experimental system we used K15 and ORF47 as reporters for ORF57 activity and observed an up to 30 fold increase in nuclear RNA levels. Export was detectable but only to a minor degree . The situation may be similar to most transient transfection experiments where also intron‐less cDNAs are expressed in the absence of splicing.…”

Identification and Characterization of the Physiological Gene Targets of the Essential Lytic Replicative Epstein-Barr Virus SM Protein

“…Differential GC content, the presence of specific RNA motifs in responsive transcripts, atypical codon usage, and differential transcriptional activation have all been proposed to explain specificity of action of SM homologs (33,36,46,(51)(52)(53). Our data demonstrate that while the basis for SM responsiveness does appear to reside in the coding sequences of the relevant genes, there is no clear relationship of the SM effect to any of the above characteristics in the SM-dependent gene set.…”

“…8A). It has been suggested that one basis for the gene specificity of the KSHV SM homolog ORF57 is the GC content of its target genes (33). While the SM-dependent gene set had a slightly lower GC content than the comparator sets of SMindependent genes, all other lytic genes, or latent genes, the average values were not overtly different (Fig.…”

Transcriptional and Post-Transcriptional Regulation of Viral Gene Expression in the Gamma-Herpesvirus Kaposi’s Sarcoma-Associated Herpesvirus

“…The mechanism by which ORF57 selectively increases their expression remains unclear, as it depends on the sequence of the coding region rather than promoter sequences or polyadenylation signals [85]. Interestingly, Vogt et al found a similar effect, reporting that ORF57 is required to sustain the levels of some viral mRNAs containing higher AT nucleotide content, which they propose is suboptimal compared to cellular nucleotide content [86]. In a separate study vFLIP was demonstrated to have suboptimal codon usage, which impaired vFLIP RNA accumulation [87].…”