Genome-wide identification of Reverse Transcriptase domains of recently inserted endogenous plant pararetrovirus (Caulimoviridae)

Abstract

Endogenous viral elements (EVEs) are viral sequences that have been integrated into the nuclear chromosomes. Endogenous pararetrovirus (EPRV) are a class of EVEs derived from DNA viruses of the family Caulimoviridae. Previous works based on a limited number of genome assemblies demonstrated that EPRVs are abundant in plants and are present in several species. The availability of genome sequences has been immensely increased in the recent years and we took advantage of these resources to have a more extensive view of the presence of EPRVs in plant genomes. We analyzed 278 genome assemblies corresponding to 267 species (254 from Viridiplantae) using tBLASTn against a collection of conserved domains of the Reverse Transcriptases (RT) of Caulimoviridae. We concentrated our search on complete and well-conserved RT domains with an uninterrupted ORF comprising the genetic information for at least 300 amino acids. We obtained 11.527 sequences from the genomes of 202 species spanning the whole Tracheophyta clade. These elements were grouped in 57 clusters and classified in 13 genera, including a newly proposed genus we called Wendovirus. Wendoviruses are characterized by the presence of four open reading frames and two of them encode for aspartic proteinases. Comparing plant genomes, we observed important differences between the plant families and genera in the number and type of EPRVs found. In general, florendoviruses are the most abundant and widely distributed EPRVs. The presence of multiple identical RT domain sequences in some of the genomes suggests their recent amplification.

Keywords: Caulimoviridae; Reverse Transcriptase (RT); endogenous; pararetrovirus; virus.

Figure 1

Phylogenetic relationships within the episomal and endogenous Caulimoviridae. Phylogram obtained from a maximum likelihood analysis with protein sequence data from RT conserved domains using 500 bootstrap replications. The size of the point indicated the bootstrap support of the tree branch. Known episomal and endogenous pararetrovirus are shown in grey and small letters. New endogenous Clusters60 are shown in bold letters. The color of the branch indicates the genus of Caulimoviridae; Bad, Badnavirus; Dio, Dioscovirus; Yen, yendovirus; Tun, tungrovirus; Zen, zendovirus; Vac, vaccinivirus; Ros, rosadnavirus; Flo, florendovirus; Gym1 and Gym2, gymnendovirus1 and 2; Pet, petuvirus; Fer, fernendovirus; Cav, cavemovirus; Sol, solendovirus; Cau, caulimovirus; Ruf, ruflodivirus; Soy, soymovirus; Xen, xendovirus; and Wen, wendovirus.

Figure 2

Phylogenetic relationships of representative sequences of the Cluster100. Representative sequences of the RT-EPRV Cluster100 (in red) were aligned with RT sequences of pararetroviral elements (in black), and a phylogenetic tree was constructed using the NJ method and 1000 bootstrap replications.

Figure 3

Schematic representation of wendovirus endogenous pararetrovirus. A scaled linear view of the genome organization of Wendovirus. The name of the sequences is the same as in Supplementary Data 4 . Grey arrows mark open reading frames and colored regions within ORFs are conserved protein domains: blue, zinc finger typically present in the coat proteins; green, Movement Protein; yellow, Aspartic Proteinase; red, Reverse Transcriptase; pink, RNaseH.