Evolution of the Old World Arenaviridae and their rodent hosts: generalized host-transfer or association by descent?

Ten scenarios optimizing the number of cospeciation events between the phylogenies of the Old World Arenaviridae (OWA) and their murine hosts are tested while attempting to answer the following questions. Does the coevolutionary model explain their respective distribution? What kind of evolutionary events could have most frequently contributed to the horizontal and/or vertical transmission of the OWA? How to define secondary hosts and to interpret their existence in the evolutionary process? Where are the geographical origins of the OWA? All scenarios support the "diffuse coevolution" hypothesis previously proposed for the OWA, in which parallel phylogeny and/or host switches on closely related hosts can be considered as the most common mechanisms of transmission. The scenarios allow defining more precisely the concepts of principal and secondary hosts. Such scenarios also suggest that the diversity of the viruses and their rodent hosts could be higher than currently expected and that cophylogeny could have been underestimated. The "diffuse coevolution" hypothesis permits to interpret the transfer of the viruses to distant hosts as a result of a disturbance in their regular mode of dispersion, which could match with the periods of emergence as human parasites. The comparison of the viral phylogeny with the host cladogram also suggests that the viruses parasitized the Murinae before several lineages became distinct and spread in Africa. This supposes that the origin of the arenaviruses has to be found out of Africa. (C) 2001 Elsevier Science B. V. All rights reserved.