The Strange Lifestyle of Multipartite Viruses

Abstract

Multipartite viruses have one of the most puzzling genetic organizations found in living organisms. These viruses have several genome segments, each containing only a part of the genetic information, and each individually encapsidated into a separate virus particle. While countless studies on molecular and cellular mechanisms of the infection cycle of multipartite viruses are available, just as for other virus types, very seldom is their lifestyle questioned at the viral system level. Moreover, the rare available "system" studies are purely theoretical, and their predictions on the putative benefit/cost balance of this peculiar genetic organization have not received experimental support. In light of ongoing progresses in general virology, we here challenge the current hypotheses explaining the evolutionary success of multipartite viruses and emphasize their shortcomings. We also discuss alternative ideas and research avenues to be explored in the future in order to solve the long-standing mystery of how viral systems composed of interdependent but physically separated information units can actually be functional.

Fig 1. Schematic representation of the genome formula of Faba bean necrotic stunt virus (FBNSV) in two different host plant species.

The genome formulae presented are in Vicia faba (A) and Medicago truncatula (B). The relative frequencies of the eight FBNSV segments have been calculated in within-host viral populations. The rounded median copy number of each segment is represented relative to the less abundant segment, here arbitrarily set to one. The core genome corresponds to the classical conception of a viral genome (rectangle). Adapted from reference [50].

Fig 2. Comparison of the genome or segment sizes in the three types of viral genome organization.

The listed families or genera are those with significant differences in genome architectures. Not all viral families are represented in this figure. We have chosen families with a large range of total genome sizes, but avoiding those monopartite virus groups with an immense genome that would have compressed too much the scale of the graphic. We also show viral families containing both mono- and multipartite member species. The size range of whole genomes and that of individual segments are illustrated by black and grey lines, respectively. All size data come from the website (http://viralzone.expasy.org/), from the ninth International Committee on Taxonomy of Viruses (ICTV) report, and from specific literature. *All genera of the family Virgaviridae except the genus Tobamovirus are composed of multipartite virus species. **The genus Begomovirus is composed of both monopartite and bipartite virus species. ***All genera of the family Closteroviridae are composed of monopartite virus species, except for the genus Crinivirus. ****All genera of the family Geminiviridae are composed of monopartite virus species, except for the genus Begomovirus. *****All genera of the family Potyviridae are composed of monopartite virus species, except for the genus Bymovirus.

Fig 3. Relationship between the phylogeny of movement proteins of the 30K superfamily and the genome organization of corresponding viruses.

This tree was constructed from all movement protein sequences available at the time of reference [85] using parsimony analysis. For more details on the construction of the tree, see reference [85]. 0, RT, N, A, I, II, and III represent the type of polymerase encoded by the viruses: none, RNA-dependent DNA polymerase, negative-strand virus, ambisens-strand virus and positive-strand virus, and supergroups I, II, III RNA-dependent RNA polymerases, respectively. The thin-lined polygon encloses those movement proteins known to form virion-bearing tubules. Genera with a red asterisk are those whose member species are multipartite viruses (N.B.: The genus Begomovirus is composed of both monopartite and bipartite viruses). Reproduced and adapted from reference [85].