Gammaherpesvirus Readthrough Transcription Generates a Long Non-Coding RNA That Is Regulated by Antisense miRNAs and Correlates with Enhanced Lytic Replication In Vivo

Abstract

Gammaherpesviruses, including the human pathogens Epstein–Barr virus (EBV) and Kaposi’s sarcoma-associated herpesvirus (KSHV) are oncogenic viruses that establish lifelong infections in hosts and are associated with the development of lymphoproliferative diseases and lymphomas. Recent studies have shown that the majority of the mammalian genome is transcribed and gives rise to numerous long non-coding RNAs (lncRNAs). Likewise, the large double-stranded DNA virus genomes of herpesviruses undergo pervasive transcription, including the expression of many as yet uncharacterized lncRNAs. Murine gammaperherpesvirus 68 (MHV68, MuHV-4, HV68) is a natural pathogen of rodents, and is genetically and pathogenically related to EBV and KSHV, providing a highly tractable model for studies of gammaherpesvirus biology and pathogenesis. Through the integrated use of parallel data sets from multiple sequencing platforms, we previously resolved transcripts throughout the MHV68 genome, including at least 144 novel transcript isoforms. Here, we sought to molecularly validate novel transcripts identified within the M3/M2 locus, which harbors genes that code for the chemokine binding protein M3, the latency B cell signaling protein M2, and 10 microRNAs (miRNAs). Using strand-specific northern blots, we validated the presence of M3-04, a 3.91 kb polyadenylated transcript that initiates at the M3 transcription start site and reads through the M3 open reading frame (ORF), the M3 poly(a) signal sequence, and the M2 ORF. This unexpected transcript was solely localized to the nucleus, strongly suggesting that it is not translated and instead may function as a lncRNA. Use of an MHV68 mutant lacking two M3-04-antisense pre-miRNA stem loops resulted in highly increased expression of M3-04 and increased virus replication in the lungs of infected mice, demonstrating a key role for these RNAs in regulation of lytic infection. Together these findings suggest the possibility of a tripartite regulatory relationship between the lncRNA M3-04, antisense miRNAs, and the latency gene M2.

Multiple Long-Read Sequencing Survey of Herpes Simplex Virus Dynamic Transcriptome

“…This study revealed an immense complexity of transcriptional overlaps ( Figure 6 and Table 6 ). These overlaps are produced by either transcriptional read-through events between transcripts oriented in parallel [as described in Kara et al (2019)], or in a convergent manner (thereby generating rtRNAs), or through the use of long TSS isoforms pertaining to one or of both partners of divergently-oriented genes. Transcriptional overlaps can also be produced by antisense transcripts controlled by their own promoters, as seen in LAT transcripts.…”

Comprehensive Annotations of Human Herpesvirus 6A and 6B Genomes Reveal Novel and Conserved Genomic Features

“…Laborious follow-up molecular work revealed the transcriptional architecture of individual genomic loci, but for most HHV-6 genes annotations are still based on these initial in-silico ORF predictions. In recent years, major advances in high-throughput sequencing approaches have revealed that the transcriptome and translatome of herpesviruses are extremely complex, encompassing large numbers of overlapping transcripts, extensive splicing and many non-canonical translation products (Arias et al, 2014;Balázs et al, 2017;Bencun et al, 2018;Depledge et al, 2019;Gatherer et al, 2011;Kara et al, 2019;O'Grady et al, 2019O'Grady et al, , 2016Tombácz et al, 2017;Whisnant et al, 2019). Our own work in which we employed ribosome profiling and systematic transcript analysis to decipher HCMV genome complexity revealed a rich collection of coding sequences that include many viral short ORFs (sORFs), uORFs and alternative translation products that generate extensions or truncations of canonical proteins (Stern-Ginossar et al, 2012).…”

“…The annotation of HHV-6 coding capacity has traditionally relied on open reading frame (ORF)-based analyses using canonical translational start and stop sequences and arbitrary size restriction to demarcate putative protein-coding genes, resulting in a list of around 100 ORFs for each virus (Dominguez et al, 1999;Gompels et al, 1995;Gravel et al, 2013). In recent years, genome wide analysis of herpesviruses using short RNA sequencing (RNA-seq) reads, and recently also direct and long-read RNA-seq revealed very complex transcriptomes Depledge et al, 2019;Gatherer et al, 2011;Kara et al, 2019;O'Grady et al, 2019O'Grady et al, , 2016Tombácz et al, 2017), and combined with genome-wide mapping of translation, revealed hundreds of new viral ORFs (Arias et al, 2014;Bencun et al, 2018;Stern-Ginossar et al, 2012;Whisnant et al, 2019).…”

Long noncoding RNAs: Novel regulators of virus‐host interactions

“…Gammaherpesviruses transcribe their own structurally and functionally intricate lncRNAs, which modulate cellular and viral gene expression to maintain viral latency or to induce lytic reactivation . In addition, a 3.91 kb lncRNA M3‐04 generated by gammaherpesviruses regulates viral replication in mice by interacting with antisense miRNAs and the latency gene M2 . Other herpesviruses also encode multiple functional lncRNAs, such as the Epstein‐Barr virus encoding ncRNA, BamH I‐A rightward transcripts, Kaposi's sarcoma‐associated herpesvirus (KSHV) encoding UCA1 , and herpesvirus saimiri encoding U‐rich RNAs .…”