Open Access

Caroline Judith-Hertz

• The fungal genus Pestalotiopsis s.l. contains approximately 300 described species and is globally distributed. The monotypic genus Pestalotia is considered the closest relative of Pestalotiopsis s.l. This study aims to investigate the diversity and systematics within Pestalotiopsis s.l. and its relation to Pestalotia. Therefore, an integrative approach is used considering molecular phylogeny methods as well as examination of morphological characters. Recently, Pestalotiopsis s.l. was split into three genera with the addition of the newly erected Neopestalotiopsis and Pseudopestalotiopsis. The species of these genera are usually saprotrophic, phytoparasitic, or endophytic, and have been isolated from soil, air, and many kinds of anorganic material. The asexual fruiting bodies appear on infected plant material as black acervuli that release conidia. The conidia are important to examine for morphological taxon recognition. The number of conidial cells is the feature that distinguishes Pestalotiopsis s.l. spp. with five celled conidia, from Pestalotia pezizoides with six celled conidia. However, the significance of morphological characters is controversially discussed among mycologists. In recent years, 55 new species were described based on minor genetic distances and marginal or no morphological differences. Thus, the value of certain morphological characters and genetic markers need to be reconsidered. In this study, 102 herbarium specimens of 26 described species, with an emphasis on plant pathogenic species from North America, have been morphologically examined and documented through drawings and photographs. Morphological examination was complemented with a comprehensive molecular dataset obtained from 191 cultures representing the genera Neopestalotiopsis, Pestalotia, Pestalotiopsis, Pseudopestalotiopsis, and Truncatella. One novelty of this work is that, besides the well-established markers ITS, TEF1, and ß-tubulin, the protein-coding genes MCM7 and TSR1 were successfully sequenced and included in the analyses. Phylogenies using Maximum Likelihood and Bayesian inference methods of single loci and the combined dataset were calculated. By comparison of these phylogenies, MCM7 was identified as the most powerful one in terms of phylogenetic resolution and statistical support of nodes and is proposed as an additional barcoding marker in Pestalotiopsis s.l. In Pestalotiopsis, species delimitation was tested using the Baysian Phylogenetics and Phylogeography (BP&P) program that tests an existing species scenario against Bayesian inference methods under a multispecies coalescent model. The program supported only ten species out of the predetermined 19 species scenario. Measurements of conidia for species detected by BP&P were explored using a TukeyHSD-Test in the program R to find means that are significantly different from each other. This test revealed that combinations of morphological characters are required to distinguish between the ten species found by BP&P. Another purpose of this work was to clarify the status of Pestalotia with regard to Pestalotiopsis s.l. Therefore, fresh epitypic material of Pestalotia pezizoides, was collected, isolated, and cultivated. The molecular analysis of a combined dataset of the gene regions ITS and LSU for species of Amphisphaeriales nested P. pezizoides in the genus Seiridium. Thus, synonymy of Pestalotia with Seiridium is proposed here. This is supported by morphology of the conidia. Further, an epitype is proposed for the type species of Pestalotiopsis, P. maculans. On the other hand, the recently proposed epitype of P. adusta is rejected here as it conflicts with the taxonomic hypothesis obtained in this study and its introduction is inconsistent with the formal requirements for epitypification. A new topotypic specimen is proposed instead. Additionally, several nomenclatural changes become necessary in many species examined. These include three new combinations and six synonyms of species of Pestalotiopsis s.l. The conclusion of this work is that morphological data have potential as a valuable, inexpensive and easy way to recognize species. However, it is not the best method for species discovery and delimitation bearing in mind that in microfungi and many other organisms, individual plasticity and analogous structures are inadequately investigated. By phylogenetic analyses of molecular sequence data, it is possible to compare a great amount of equivalent characters and to delimit species that are morphologically cryptic. This is especially important since species of Pestalotiopsis s.l. mostly lack sexual structures that are helpful for morphological species delimitation in other groups of fungi. Thus, the Genealogical Concordance Species Concept (GCSC) finds its application in many fungal taxa. Conflicts in the genealogy between phylogenetic trees of different markers are interpreted as recombination of the genetic material within a linage. Accordingly, the change from conflict to congruence in a set of different phylogenetic trees can be seen as the species limit. It can be expected that increased application of the GCSC will lead to further approximation of described species numbers to the real number of species, especially in complicated groups like asexual microfungi.