Long-read Transcriptomics of Caviid Gammaherpesvirus 1: Compiling a Comprehensive RNA Atlas

Abstract

Caviid gammaherpesvirus 1 (CaGHV-1), formerly known as the guinea pig herpes-like virus, is an oncogenic gammaherpesvirus with a sequenced genome but an as-yet uncharacterized transcriptome. Using nanopore long-read RNA sequencing, we annotated the CaGHV-1 genome and constructed a detailed transcriptomic atlas. Our findings reveal diverse viral mRNAs and non-coding RNAs, along with mapped promoter elements for each viral gene. We demonstrated that the CaGHV-1 RTA lytic cycle transcription factor activates its own promoter, similar to KSHV, and that the CaGHV-1 ORF50 promoter responds to RTA proteins from other gammaherpesviruses, highlighting the evolutionary conservation of RTA-mediated transcriptional mechanisms. Additionally, our analysis uncovered extensive transcriptional overlap within the viral genome, suggesting a role in regulating global gene expression. Given its tumorigenic properties, broad host range, and non-human pathogenicity, this work establishes CaGHV-1 as a promising small animal model for investigating human gammaherpesvirus pathogenesis.

Keywords: Caviid gammaherpesvirus 1 (CaGHV-1); direct RNA sequencing; gammaherpesvirus; guinea pig herpes-like virus; long-read sequencing; luciferase reporter assay; nanopore sequencing; transcriptome.

Figure 1.. Coverage and average read lengths generated by the ONT-PromethION LRS platform.

(A) dcDNA-Seq sequencing illustrates the proportion of host/virus reads, while (B) presents the proportion of host /virus reads for dRNA-Seq sequencing. (C) The histogram depicts the distribution of read lengths obtained from dcDNA-Seq and dRNA-Seq in host cell and virus samples. (D) Transcript types.

Figure 2.. Characterization and distribution of the 5’ -ends of CaGHV-1 transcripts.

(A) The graph illustrates the probability distribution of nucleotides within the −50 to +10 base pair interval surrounding the transcription start sites (TSSs). For most TSSs, G/A nucleotides are preferred at the +1 and +2 positions, while C/T nucleotides are favored at the −1 position. (B) Showing the distribution of canonical eukaryotic TATA boxes in the TSSs detected by LoRTIA. (C) The gray lines depict the distribution of TSSs along the viral genome as detected by LoRTIA, with the line sizes representing the TSS counts on a logarithmic scale (Log₁₀). Red vertical arrows represent the annotated TATA boxes, black horizontal arrows indicate the ORFs of genes, and green horizontal arrows represent replication origin-associated RNA and antisense RNA molecules.

Figure 3.. Characterization and distribution of the 3’-ends of CaGHV-1 transcripts.

(A) The nucleotide probability distribution within the −50 to +10 base pair region surrounding the transcription end sites (TESs). TESs are characterized by the presence of the eukaryotic A/C cleavage site and G/U-rich sequence motifs downstream. (B) The distribution of canonical eukaryotic TATA boxes identified in TSSs by the LoRTIA program. (C) The gray lines show the distribution of TSSs across the viral genome as identified by LoRTIA, with line sizes corresponding to TSS counts on a logarithmic scale (Log₁₀). Black vertical arrows mark the annotated TATA boxes, black horizontal arrows indicate gene ORFs, and green horizontal arrows represent replication origin-associated RNAs (raRNAs) as well as antisense (as) RNA molecules.

Figure 4.. CaGHV-1 transcripts.

This figure presents the canonical mRNAs and ncRNAs of CaGHV-1 along the reference genome. Transcripts with different splicing patterns are regarded as distinct canonical transcripts. Pink arrows indicate (+)-oriented RNAs, while blue arrows represent (−)-oriented RNAs. Additionally, antisense and replication-associated RNAs are shown in green.

Figure 5.. Replication origin-associated RNAs.

(A) OriLyt-L: K3-PAN-ORF17 regions. (B) OriLyt-R: ORF69-ORF72 regions. Red arrows indicate non-coding RNAs, green arrows represent replication origin-associated RNAs (which can be either coding or non-coding), and blue arrows denote both monocistronic and polycistronic transcripts.

Figure 6.. Overlaps of raw dRNA Reads.

(A) The upper coverage plot shows the transcriptional activity of the viral genome, indicating that both DNA strands are transcriptionally active across the entire genome. Coverage values are plotted on a Log₁₀ scale, with red representing the positive strand and blue representing the negative strand. (B) This panel highlights an extremely complex meshwork of transcriptional overlaps formed by genes arranged in head-to-head (divergent) and tail-to-tail (convergent) orientations. We hypothesize strong interference between the transcriptional machineries at the overlapping regions, which may represent a novel layer of gene regulation.

Figure 7.. Evaluating the transcriptional activity of CaGHV-1 RTA.

(A) Gene structure of ORF50 encoding gpRTA in the CaGHV-1 genome. The ORF50 gene has 4 exons. The genomic coordinates are based on OQ679822.1 (GenBank). (B) Protein expression of N-terminally 3xFLAG tagged CaGHV-1 RTA in transfected 293T cells. (C) Testing the inducibility of the 3 kb promoter region of CaGHV-1 ORF50 by CaGHV-1 RTA in 293T and 104C1 cell lines using luciferase reporter assays. (D) Measuring the inducibility of the ORF50 promoter with differing lengths by ORF50 promoter in luciferase reporter assays. (E) Western blot showing the protein expression of N-terminally 3xFLAG tagged RTAs derived from the indicated gammaherpesviruses. (F) Analyzing the transcriptional activity of RTAs derived from the indicated gammaherpesviruses on the 3 kb promoter region of CaGHV-1 ORF50 in 293T and 104C1 cells.