Background: Using phylogenetic approaches, the expectation that parallel cladogenesis should occur between parasites and hosts has been validated in some studies, but most others provided evidence for frequent host shifts. Here we examine the evolutionary history of the association between Microbotryum fungi that cause anther smut disease and their Caryophyllaceous hosts. We investigated the congruence between host and parasite phylogenies, inferred cospeciation events and host shifts, and assessed whether geography or plant ecology could have facilitated the putative host shifts identified. For cophylogeny analyses on microorganisms, parasite strains isolated from different host species are generally considered to represent independent evolutionary lineages, often without checking whether some strains actually belong to the same generalist species. Such an approach may mistake intraspecific nodes for speciation events and thus bias the results of cophylogeny analyses if generalist species are found on closely related hosts. A second aim of this study was therefore to evaluate the impact of species delimitation on the inferences of cospeciation. Results: We inferred a multiple gene phylogeny of anther smut strains from 21 host plants from several geographic origins, complementing a previous study on the delimitation of fungal species and their host specificities. We also inferred a multi-gene phylogeny of their host plants, and the two phylogenies were compared. A significant level of cospeciation was found when each host species was considered to harbour a specific parasite strain, i.e. when generalist parasite species were not recognized as such. This approach overestimated the frequency of cocladogenesis because individual parasite species capable of infecting multiple host species (i.e. generalists) were found on closely related hosts. When generalist parasite species were appropriately delimited and only a single representative of each species was retained, cospeciation events were not more frequent than expected under a random distribution, and many host shifts were inferred. Current geographic distributions of host species seemed to be of little relevance for understanding the putative historical host shifts, because most fungal species had overlapping geographic ranges. We did detect some ecological similarities, including shared pollinators and habitat types, between host species that were diseased by closely related anther smut species. Overall, genetic similarity underlying the host-parasite interactions appeared to have the most important influence on specialization and host-shifts: generalist multi-host parasite species were found on closely related plant species, and related species in the Microbotryum phylogeny were associated with members of the same host clade. Conclusion: We showed here that Microbotryum species have evolved through frequent host shifts to moderately distant hosts, and we show further that accurate delimitation of parasite species is essential for interpreting cophylogeny studies.

Summary: Genomic regions determining sexual compatibility often display recombination suppression, as occurs in sex chromosomes, plant self‐incompatibility loci and fungal mating‐type loci. Regions lacking recombination can extend beyond the genes determining sexes or mating types, by several successive steps of recombination suppression. Here we review the evidence for recombination suppression around mating‐type loci in fungi, sometimes encompassing vast regions of the mating‐type chromosomes. The suppression of recombination at mating‐type loci in fungi has long been recognized and maintains the multiallelic combinations required for correct compatibility determination. We review more recent evidence for expansions of recombination suppression beyond mating‐type genes in fungi (‘evolutionary strata’), which have been little studied and may be more pervasive than commonly thought. We discuss testable hypotheses for the ultimate (evolutionary) and proximate (mechanistic) causes for such expansions of recombination suppression, including (1) antagonistic selection, (2) association of additional functions to mating‐type, such as uniparental mitochondria inheritance, (3) accumulation in the margin of nonrecombining regions of various factors, including deleterious mutations or transposable elements resulting from relaxed selection, or neutral rearrangements resulting from genetic drift. The study of recombination suppression in fungi could thus contribute to our understanding of recombination suppression expansion across a broader range of organisms.

• Understanding disease distributions is of fundamental and applied importance, yet few studies benefit from integrating broad sampling with ecological and phylogenetic data. Here, anther-smut disease, caused by the fungus Microbotryum, was assessed using herbarium specimens of Silene and allied genera of the Caryophyllaceae. • A total of 42 000 herbarium specimens were examined, and plant geographical distributions and morphological and life history characteristics were tested as correlates of disease occurrence. Phylogenetic comparative methods were used to determine the association between disease and plant life-span. • Disease was found on 391 herbarium specimens from 114 species and all continents with native Silene. Anther smut occurred exclusively on perennial plants, consistent with the pathogen requiring living hosts to overwinter. The disease was estimated to occur in 80% of perennial species of Silene and allied genera. The correlation between plant life-span and disease was highly significant while controlling for the plant phylogeny, but the disease was not correlated with differences in floral morphology. • Using resources available in natural history collections, this study illustrates how disease distribution can be determined, not by restriction to a clade of susceptible hosts or to a limited geographical region, but by association with host life-span, a trait that has undergone frequent evolutionary transitions.