Abstract
ABSTRACTViruses of the phylum Nucleocytoviricota, or nucleo-cytoplasmic large DNA viruses (NCLDVs), undergo a cytoplasmic or nucleo-cytoplasmic cycle, the latter of which involves both nuclear and cytoplasmic compartments to proceed viral replication. Medusavirus, a recently isolated NCLDV, has a nucleo-cytoplasmic replication cycle in amoebas during which the host nuclear membrane apparently remains intact, a unique feature among amoeba-infecting NCLDVs. The medusavirus genome lacks most transcription genes but encodes a full set of histone genes. To investigate its infection strategy, we performed a time course RNA sequencing (RNA-seq) experiment. All viral genes were transcribed and classified into five temporal expression clusters. The immediate early genes (cluster 1, 42 genes) were mostly (83%) of unknown functions, frequently (95%) associated with a palindromic promoter-like motif, and often (45%) encoded putative nucleus-localized proteins. These results suggest massive reshaping of the host nuclear environment by viral proteins at an early stage of infection. Genes in other expression clusters (clusters 2 to 5) were assigned to various functional categories. The virally encoded core histone genes were in cluster 3, whereas the viral linker histone H1 gene was in cluster 1, suggesting they have distinct roles during the course of the virus infection. The transcriptional profile of the host Acanthamoeba castellanii genes was greatly altered postinfection. Several encystment-related host genes showed increased representation levels at 48 h postinfection, which is consistent with the previously reported amoeba encystment upon medusavirus infection.IMPORTANCE Medusavirus is an amoeba-infecting giant virus that was isolated from a hot spring in Japan. It belongs to the proposed family “Medusaviridae” in the phylum Nucleocytoviricota. Unlike other amoeba-infecting giant viruses, medusavirus initiates its DNA replication in the host nucleus without disrupting the nuclear membrane. Our RNA sequencing (RNA-seq) analysis of its infection course uncovered ordered viral gene expression profiles. We identified temporal expression clusters of viral genes and associated putative promoter motifs. The subcellular localization prediction showed a clear spatiotemporal correlation between gene expression timing and localization of the encoded proteins. Notably, the immediate early expression cluster was enriched in genes targeting the nucleus, suggesting the priority of remodeling the host intranuclear environment during infection. The transcriptional profile of amoeba genes was greatly altered postinfection.
Highlights
Background frequencycP value aA binomial test was used to assess whether each motif was preferentially located in intergenic regions (IRs). bOnly motifs that did not overlap predicted genes were considered to be located in IRs. cBackground frequency was calculated by dividing the sum of the length of all IRs by the length of the whole genome.Motif 1 palindromic Motif 2 Poly-ADivergentb Convergentc P valueWith motif 22 0 2.76e-06Without motif 93 88With motif 40 4 8.48e-08Without motif 75 84aOnly motifs predicted to be located in intergenic regions were used to determine their preferred location
We identified five clusters of viral gene expression profiles using the k-means method (Fig. 2B) and named these clusters as follows: cluster 1 genes showed a gradual increase in expression from 0 hpi; cluster 2 genes showed a gradual increase in expression from 1 hpi; clusters 3 and 4 genes showed a gradual increase in expression from 2 hpi; and cluster 5 genes showed a gradual increase in expression from 4 hpi
We performed RNA sequencing (RNA-seq) to dissect the transcriptional program of medusavirus
Summary
Background frequencycP value aA binomial test was used to assess whether each motif was preferentially located in intergenic regions (IRs). bOnly motifs that did not overlap predicted genes were considered to be located in IRs. cBackground frequency was calculated by dividing the sum of the length of all IRs by the length of the whole genome.Motif 1 palindromic Motif 2 Poly-ADivergentb Convergentc P valueWith motif 22 0 2.76e-06Without motif 93 88With motif 40 4 8.48e-08Without motif 75 84aOnly motifs predicted to be located in intergenic regions were used to determine their preferred location. P value aA binomial test was used to assess whether each motif was preferentially located in intergenic regions (IRs). BOnly motifs that did not overlap predicted genes were considered to be located in IRs. cBackground frequency was calculated by dividing the sum of the length of all IRs by the length of the whole genome. BDivergent cases were defined as motifs being located in the upstream regions of both neighbor genes. CThe convergent cases were defined as the motifs being located in the downstream region of both neighbor genes. Based on the host gene clusters, we performed Gene Ontology GO and KEGG pathway functional enrichment analyses (Fig. 9B to E). Cluster 1 genes were enriched in cellular transportation-related GO terms, such as “localization,” “establishment of localization,” and “transport” (Fig. 9B). Cluster 2 genes were enriched in 60 GO terms that fell into two main categories (Fig. 9D). One category comprised terms related to “cellular protein metabolic process” and “proteolysis involved in cellular protein catabolic process,” and the other category comprised stress-related terms such as “DNA repair” (Fig. S3)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
