Implications of molecular characters for the phylogeny of the Microbotryaceae (Basidiomycota: Urediniomycetes)

Abstract

Background: Anther smuts of the basidiomycetous genus Microbotryum on Caryophyllaceae are important model organisms for many biological disciplines. Members of Microbotryum are most commonly found parasitizing the anthers of host plants in the family Caryophyllaceae, however they can also be found on the anthers of members of the Dipsacaceae, Lamiaceae, Lentibulariaceae, and Portulacaceae. Additionally, some members of Microbotryum can be found infecting other organs of mainly Polygonaceae hosts. Based on ITS nrDNA sequences of members of almost all genera in Microbotryaceae, this study aims to resolve the phylogeny of the anther smuts and their relationship to the other members of the family of plant parasites. A multiple analysis strategy was used to correct for the effects of different equally possible ITS sequence alignments on the phylogenetic outcome, which appears to have been neglected in previous studies.

Results: The genera of Microbotryaceae were not clearly resolved, but alignment-independent moderate bootstrap support was achieved for a clade containing the majority of the Microbotryum species. The anther parasites appeared in two different well-supported lineages whose interrelationship remained unresolved. Whereas bootstrap support values for some clades were highly vulnerable to alignment conditions, other clades were more robustly supported. The differences in support between the different alignments were much larger than between the phylogenetic optimality criteria applied (maximum parsimony and maximum likelihood).

Conclusion: The study confirmed, based on a larger dataset than previous work, that the anther smuts on Caryophyllaceae are monophyletic and that there exists a native North American group that diverged from the European clade before the radiation of the European species. Also a second group of anther smuts was revealed, containing parasites on Dipsacaceae, Lamiaceae, and Lentibulariaceae. At least the majority of the parasites of Asteraceae appeared as a monophylum, but delimitations of some species in this group should be reconsidered. Parasitism on Polygonaceae is likely to be the ancestral state for the Microbotryaceae on Eudicot hosts.

Figure 1

Strict consensus of 1780 most parsimonious trees inferred from the dataset consisting of three concatenated, complete ITS alignments. The topology was rooted with Ustilentyloma brefeldii and U. fluitans. Symbols on branches indicate the magnitude of parsimony bootstrap values from analyses of the dataset after exclusion of alignment-ambiguous sites (upper left) and of the three different, complete alignments made with MAFFT (upper right), PCMA (lower left), and POA (lower right). The right side of the picture reports morphological features of the specimens included in our data set. The symbols and abbreviations used are as follows. Form of spore ornamentation: filled-in circle, smooth spores; hollow square, verrucose spores; filled-in square, reticulate spores. Height of spore ornamentation: filled-in circle, smooth spores; hollow square, flat ornamentation; filled-in square, high ornamentation. Disjunctors, columella: hollow square, absent; filled-in square, present. Forming galls with an apical lunular bed of spore masses: hollow square, absent; filled-in square, present. Sorus location: Leav, leaves; St, Stems; Inf, inflorescence axis; Ped, pedicels; Flow, swollen (and often deformed) whole flowers; Ovar, ovaries only; Seed, seeds only; Fil, filaments only; Anth, anthers only. Host family: Poac, Poaceae; Junc, Juncaceae; Poly, Polygonaceae; Lent, Lentibulariaceae; Lami, Lamiaceae; Dips, Dipsacaceae; Aste, Asteraceae; Cary, Caryophyllaceae; M., Microbotryum; M. tragopo.-pratensis, Microbotryum tragopogonis-pratensis; M. v.-irregulare, Microbotryum violaceo-irregulare; M. v.-verrucosum, Microbotryum violaceo-verrucosum; S., Sphacelotheca.

Figure 2

Maximum likelihood analyses of phylogenetic relationships of sampled Microbotryum specimens. The tree shown was inferred from the PCMA alignment with PhyML under a TrN+I+G model of site substitution. Branch lengths are scaled in terms of expected numbers of nucleotide substitutions per site. For explanation of the three Microbotryum groups on the right-hand side see discussion. Abbreviations are as in Fig. 1. The dashed line indicates that Microbotryum group 2 is not a monophyletic group.

Figure 3

Maximum likelihood majority-rule consensus inferred from the datasets of the three different alignments. The right side of the picture indicates the host plant for each parasite. Symbols on branches indicate the magnitude of likelihood bootstrap values from analyses of the dataset after exclusion of alignment-ambiguous sites (upper left) and of the three different, complete alignments made with MAFFT (upper right), PCMA (lower left), and POA (lower right). Abbreviations are as in Fig. 1.