Asexual and sexual morphs of Moesziomyces revisited

Abstract

Yeasts of the now unused asexually typified genus Pseudozyma belong to the smut fungi (Ustilaginales) and are mostly believed to be apathogenic asexual yeasts derived from smut fungi that have lost pathogenicity on plants. However, phylogenetic studies have shown that most Pseudozyma species are phylogenetically close to smut fungi parasitic to plants, suggesting that some of the species might represent adventitious isolations of the yeast morph of otherwise plant pathogenic smut fungi. However, there are some species, such as Moesziomyces aphidis (syn. Pseudozyma aphidis) that are isolated throughout the world and sometimes are also found in clinical samples and do not have a known plant pathogenic sexual morph. In this study, it is revealed by phylogenetic investigations that isolates of the biocontrol agent Moesziomyces aphidis are interspersed with M. bullatus sexual lineages, suggesting conspecificity. This raises doubts regarding the apathogenic nature of asexual morphs previously placed in Pseudozyma, but suggests that there might also be pathogenic sexual morph counterparts for those species known only from asexual morphs. The finding that several additional species currently only known from their yeast morphs are embedded within the genus Moesziomyces, suggests that the yeast morph might play a more dominant role in this genus as compared to other genera of Ustilaginaceae. In addition, phylogenetic reconstructions demonstrated that Moesziomyces bullatus has a narrow host range and that some previously described but not widely used species names should be applied for Moesziomyces on other host genera than Echinochloa.

Introduction

Ustilaginales is the largest order within the smut fungi (Ustilaginomycetes), including species forming a blackish to brownish powdery spore mass in different organs of monocotyledonous, and exceptionally dicotyledonous plants (Vánky 2012, Begerow et al. 2006). The order includes nine families, encompassing 54 genera. The Anthracoideaceae with the occurrence on Cyperaceae and Juncaceae, and Ustilaginaceae, with few exceptions parasitic to the Poaceae, are the largest families within the order. The latter contains the three largest smut genera, Anthracocystis, Sporisorium, and Ustilago. The difference between these three closely related genera is the almost complete lack of a plant-derived columella within sori formed by Ustilago species (Vánky 2012, McTaggart et al. 2012). Within Ustilago some species are economically important pathogens, like corn smut (Ustilago maydis) or wheat smut (Ustilago nuda). Ustilago maydis is a species for which one of the first fungal genomes was sequenced (Kämper et al. 2006). Smut fungi of the Ustilaginales usually feature both an asexual yeast morph and a sexual morph infecting host plants in a biotrophic manner. In rare cases yeasts of the Ustilaginales could also be found to be affecting humans (McNeil & Palazzi 2012, Teo & Tay 2006). The earliest case of an invasive infection with an Ustilago species, possibly U. maydis, was reported in 1946 (Moore et al. 1946). But spores of the Ustilaginales potentially also cause pneumonias, allergic reactions, or asthma (Valverde et al. 1995).

There are several studies dealing with the phylogeny of Ustilaginomycotina, mostly based on the LSU or ITS locus and some of them include asexual morphs as well (e.g. Begerow et al. 2000, 2006, Stoll et al. 2005, Wang et al. 2006, 2015, Boekhout 2011). Wang et al. (2015) link many asexual yeasts to their corresponding sexual morphs, an important step within the naming of pleomorphic fungi, as dual naming of sexual and asexual morphs is now discontinued (Hawksworth et al. 2011).

Pseudozyma has been used for species of ustilagino-mycetous yeasts belonging to Ustilaginales which are mostly believed to be apathogenic (Begerow et al. 2000, 2006). The genus was described in 1985, by Bandoni (1985), and later refined by Boekhout (1995). After Sampaio (2004) and finally Wang et al. (2015) established that the type species Pseudozyma prolifica was a synonym of Ustilago maydis, the name Pseudozyma should no longer be used. However, the phylogenetic position of some species referred to Pseudozyma is still unclear. Wang et al. (2015) suggested using the name Pseudozyma now with the addition “pro tempore” for five Pseudozyma species with an unclear phylogenetic position. In the current study we give Pseudozyma names with reference to the new combinations recently made, where possible.

To date, about 20 Pseudozyma species names were validly published (www.indexfungroum.org). Fifteen are now linked to a corresponding sexually typified genus (Wang et al. 2015). Of these, Pseudozyma antarctica, P. aphidis, P. parantarctica, and P. rugulosa were transferred to Moesziomyces (Wang et al. 2015).

Ustilaginalean yeasts are isolated from diverse habitats, for Ustilaginales, but mostly from grasses (Boekhout 1995, Avis et al. 2001). Some were isolated from flowers, leaves or fruits of other plants, but it is also possible to isolate them from soil or even human blood or secretory fluid (Sugita et al. 2003, Arendrup et al. 2014). Apart from the few clinical cases of Ustilago maydis infestation, it was not known until 2003 that species referred to Pseudozyma could also infect humans (Sugita et al. 2003). However, such species can cause invasive infections, especially in immunosuppressed individuals (Arendrup et al. 2014). The infection risk, according to Prakash et al. (2014), is the same as being colonised by non-albicans Candida infections. Furthermore, Avis et al. (2001) and Gafni et al. (2015) reported antifungal properties of some pseudozyma-like yeast species, including Moesziomyces aphidis, and some strains have been reported to be natural antagonists of powdery mildews (Erysiphales). Colonization of leaf surfaces by these yeasts provides a natural source of protection against some plant pathogenic fungi (Gafni et al. 2015).

Of the ustilaginalean yeasts, especially Moesziomyces antarcticus and M. aphidis, formerly treated as Pseudozyma species, have been frequently isolated from various substrates. Moesziomyces antarcticus was isolated from plants and soil, but also from blood of humans (Boekhout 2011). Moesziomyces aphidis was described in 1995 and first isolated from the secretion of an aphid, but it has later been isolated from water (Boekhout 2011) and various other sources, including soil and human blood. Wang et al. (2015) showed that these pseudozyma-like species, together with M. parantarcticus and M. rugulosus, belong to the genus Moesziomyces, which before had been generally regarded as monotypic (Vánky 2005).

Moesziomyces mainly differs from other smut fungi in having remnants of ruptured sterile cells (Vánky 1977). Vánky (1977) initially included four species in the genus Moesziomyces. Three of them, occurring on different genera of grasses (Leersia, Paspalum, Pennisetum) were later considered to be conspecific and united under the oldest name, M. bullatus (Vánky 1986, 2005). The other species, M. eriocauli (Vánky 1986), was transferred to a new genus Eriomoeszia, because of a thin cortex of sterile cells, which surrounds the spore balls (Vánky 2005). After the transfer of four pseudozyma-like species (M. antarcticus, M. aphidis, M. parantarcticus and M. rugulosus) to this genus, it now contains five species (Wang et al. 2015).

Given the high host specificity observed for most Ustilaginales species (McTaggart et al. 2012, Escudero 2015, Li et al. 2017), it seemed doubtful that Moesziomyces bullatus was parasitic to seven not closely related genera, suggesting that more species might be present in the genus, some of which might be conspecific with smuts in the past named in Pseudozyma. It was the aim of the current study to clarify the relationships of asexual and sexual morphs in the genus Moesziomyces.

Material and Methods

Fungal material

The fungal material used in this study is listed in Tables 1 and 2. The nomenclature of the hosts is derived from the latest version of The International Plant Names Index (www.ipni.org), the nomenclature of the fungi follows Vánky (2012) and MycoBank (www.mycobank.org).

Table 1 List of Moesziomyces specimens used in the present study.

Yeast cultivation

Fresh material of Moesziomyces bullatus collected in 2015 (GLM-F105817) was used for yeast cultivation. A suspension of spores in 2 mL water was prepared. Three tubes with 200 µL spore suspension were exposed to three different conditions: (1) heating of the suspension on a thermomixer at 45 °C for 5 min (Shetty & Safeeulla 1979); (2) chilling the suspension overnight on ice; and (3) incubation for 5 min at room temperature (ca. 20 °C). From each tube 20 µL suspension was each plated on two plates of SAM (Thines lab Standard Agar Medium, consisting of 20 g agar, 20 g PDB, 10 g yeast extract, 10 g malt extract, 40 mL clarified vegetable juice, 960 mL water) with the addition of 75 mg Rifampicin/L. One set of plates was incubated at 30 °C, the other set at room temperature. After 3 d on every plate abundant yeast growth was recognized. Pure cultures were produced by picking and transferring individual single colonies to the SAM medium (isolate A1–A10). To isolate pseudozyma-morphs associated with Albugo laibachii on Arabidopsis thaliana, A. laibachii spores were suspended in water and treated with antibiotics to remove bacteria. Subsequently, the suspension was plated on PDA at 20 °C and colonies were singled out after 7 d.

DNA extraction, PCR and sequencing

In total 5–20 mg of infected plant tissue from herbarium specimens and yeast colonies were disrupted in a mixer mill (MM2, Retsch), using two iron beads of 3 mm and 5–8 iron beads of 1 mm diam per sample and shaking at 25 Hz for 5–10 min. Genomic DNA was extracted using the BioSprint 96 DNA Plant Kit (Qiagen, Hilden) on a KingFisher Flex robot (Thermo Scientific, Dreieich). PCR amplification of the complete ITS nrDNA was performed using the M-ITS1 forward primer (Stoll et al. 2003) and the ITS4 (White et al. 1990) or smITS-R1 reverse primer (Kruse et al. 2017).

The reaction was performed in a thermocycler (Eppendorf Mastercycler 96 vapo protect, Eppendorf, Hamburg) with an initial denaturation at 95 °C for 4 min, 36 PCR cycles of denaturation at 95 °C for 40 s, annealing at 56 °C for 40 s and elongation at 72° C for 60 s, followed by a final elongation at 72° C for 4 min. The resulting amplicons were sequenced at the Biodiversity and Climate Research Centre (BiK-F) laboratory using the PCR primers. Sequences were deposited in GenBank (NCBI, Table 1).

Phylogenetic reconstructions

The dataset included sequences of Moesziomyces species sexual and asexual morphs, both newly sequenced (Table 1) and downloaded from GenBank (Table 2). First all available sequences were extracted from NCBI on the basis of sequence similarity. Subsequently sequences were removed that were: (1) highly redundant with already-included ITS genotypes; or (2) of doubtful sequence quality, i.e. with mutations at positions highly conserved or with nucleotide changes only towards one end of the sequences.

Table 2 List of additional sequences used in the phylogenetic tree, downloaded from GenBank.

Macalpinomyces eriachnes was selected as an outgroup, based on the phylogenetic tree in Shivas et al. (2013). Alignments were made using mafft v. 7 (Katoh & Standley 2013) employing the Q-INS-I algorithm and removing leading and trailing gaps. The resulting total alignment length was 576 bp. For phylogenetic analyses, Minimum Evolution (ME) analysis was done with Mega v. 6.06 (Tamura et al. 2013), using the Tamura-Nei substitution model, assuming partial deletion at a cut-off of 80% and 1000 bootstrap replicates. Maximum Likelihood (ML) analysis was done using RAxML on the webserver TrEase (www.thines-lab.senckenberg.de/trease) with all parameters were set to default values. For Bayesian analysis also the webserver TrEase was used for calculating 10 M tree generations on four incrementally heated MC chains. The first 30% of the trees obtained this way were discarded to ensure sampling of the stationary phase. All other parameters were set to default.

Morphological examination

For light microscopy, the herbarium specimens GLM-F105814 and GLM-F105812 were transferred to distilled water on a slide. Morphological examination was carried out using a Zeiss Imager M2 AX10 microscope (Carl Zeiss, Göttingen). Measurements of the spore balls and spores were performed at ×400. Measurements are reported as maxima and minima in parentheses, and the mean plus and minus the standard deviation of a number of measurements given in parenthesis; the means are given in italics (Table 3).

Table 3 Measurements from 25 spore balls and 100 teliospores for collections of Moesziomyces bullatus on Echinochloa crus-galli from two different clades.

Results

The isolated yeasts from fresh Moesziomyces bullatus samples from Echinochloa crus-galli and E. muricatus were fast-growing on SAM medium. The colour of the yeasts was cream to light reddish, and the shape of the colonies was regular and roundish.

A phylogenetic hypothesis for the sampled Moesziomyces species and the cultivated yeast asexual morphs is given in Fig. 1. The results of the Minimum Evolution, Maximum Likelihood and Bayesian Analyses were congruent. The clade comprising the type of M. aphidis also includes M. bullatus s. str. from Echinochloa crus-galli and E. muricata, as well as the sequence of the type of M. rugulosa, an isolate of Moesziomyces from Albugo laibachii on Arabidopsis thaliana and many other isolates not determined to the species level from various sources. While visual inspection of the alignments revealed that there was some sequence variation within the Moesziomyces bullatus clade, the relationships of the four subgroups was not resolved apart from the clustering of Pseudozyma aphidis and the majority of M. bullatus isolates with the clade containing the type of P. rugulosa. Collections from E. crus-galli with smut symptoms were present in two different clades. The morphological investigation of a sexual morph from each clade (GLM-F105812 and GLM-F105814) revealed no morphological differences. Moesziomyces bullatus clustering within the majority of M. aphidis had the following spore characteristics: sporeballs variable in shape and size, 52.5–156.5 × 41.5–113 µm, spores ovoid, globose, often irregular, pale yellow-brown, (6.5−)7–7.8–8.5(−10) × (5.5−)6–6.8–7.5(−8) µm, a length/breadth ratio of 1.01–1.15–1.39 (n = 100) (Fig. 2). In comparison, the collection of M. bullatus clustering together with the sequence of the type species of M. rugulosus showed the following spore characteristics: sporeballs variable in shape and size, 58.5–144.5 × 44–105 µm, spores ovoid, globose, often irregular, pale yellow-brown, (6−)7–7.9–8.5(−9.5) × (5.5−)6–6.7–7.5(− 8) µm, a length/breadth ratio of 1.01–1.19–1.38 (n = 100) (Fig. 2). The sister group to M. bullatus was formed by M. antarcticus. The four lineages of M. bullatus formed an isolated clade with high to maximum support in all analyses together with samples classified as M. antarcticus.

Fig. 1

Phylogenetic tree based on Minimum Evolution analyses of nrITS sequences of Moesziomyces spp., rooted with Macalpinomyces eriachnes. Numbers on branches denote bootstrap support in Minimum Evolution, Maximum Likelihood and a posteriori probabilities from Bayesian Analyses. Values below 55% are not shown. The bar indicates expected substitutions per site. GenBank numbers precede taxon names, and are followed by the name of the host or isolation source of the fungus.

Fig. 2

Sori and spores of Moesziomyces bullatus on Echinochloa crus-galli. A, D. Sori. B, E. Teliospore balls. C, F. Teliospores. A–C (GLM-F105812), D–F (GLM-F105814).

Apart from the groups mentioned above, four additional distinct groups were revealed. Two of these corresponded to lineages formed by sexual smuts of the genus Moesziomyces isolated from plants with smut disease symptoms. One of these corresponded to M. bullatus s. lat. on Paspalum distichum, and the other to Eriomoeszia eriocauli. The remaining two lineages formed a monophyletic clade with high support in Minimum Evolution Analysis. One lineage included sequences of yeasts classified as M. parantarcticus, the other a sexual morph of a plant pathogenic fungus of the genus Moesziomyces from Pennisetum glaucum, as well asexual morphs isolated from symptom-free barley and a preterm infant.

Discussion

Moesziomyces is a morphologically well-defined genus in the smut fungi, mainly characterised by ruptured sterile cells in the sori around the spores. The genus was believed to be monotypic by Vánky (2012), but phylogenetic investigations of the past decade have shown that several species previously assigned to the asexually typified yeast genus Pseudozyma, were closely related to Moesziomyces bullatus (Begerow et al. 2000, 2006, Wang et al. 2006, 2015). In the latest edition of the International Code of Nomenclature for algae, fungi and plants (ICN) it is ruled that the dual naming for asexual and sexual morphs of fungi has been discontinued (McNeill et al. 2012). Consequently, Wang et al. (2015) attempted to resolve the names of species placed in the genus Pseudozyma as far as possible, and combined those related to Moesziomyces bullatus s.lat. into Moesziomyces.

The yeast asexual morphs were, for example, found to live epiphytically on different hosts (Boekhout 1995), but also to occur on a variety of other substrates. Due to their asexual reproduction with pullulating and division, it is possible for them to colonize suitable habitats in a short period of time. Of these yeasts, Pseudozyma aphidis is often considered as a biocontrol agent for plant pathogenic fungi (Avis et al. 2001, Buxdorf et al. 2013). Thus it is noteworthy that one isolate of this species co-occurred in Albugo laibachii lesions on Arabidopsis thaliana, indicating only no or only limited antagonism against this specialised white blister rust species (Thines et al. 2009). It is commonly believed that most Pseudozyma species have lost pathogenicity, which is seemingly supported by recent genomic analyses (Lefebvre et al. 2013). However, it should be noted that if a different start codon is taken for translation than the one predicted, all Pseudozyma yeasts included by Lefebvre et al. (2013) have a functional copy of PEP1, a conserved effector among smut fungi of the Ustilaginales (Sharma et al. 2014, Hemetsberger et al. 2015), suggesting the possibility of a misannotation of the start codon. In-depth bioinformatic analyses and functional testing will be needed to clarify this situation.

Deducing the conspecificity of Moesziomyces bullatus with Pseudozyma aphidis and P. rugulosa was not possible for Wang et al. (2015), as they did not include sequences from the type host of M. bullatus, Echinochloa crus-galli, but only from M. verrucosus on Paspalum distichum, which they erroneously assumed to be conspecific with M. bullatus. However, the smut sexual morphs from the type host, E. crusgalli from Germany, are placed in two of the four subclusters of a larger cluster, which is interpreted here as representing M. bullatus. It is noteworthy that three of the four subclusters of M. bullatus contain environmental samples from various sources. In conjunction with the ease of cultivation observed for M. bullatus from E. crus-galli, it is concluded that unlike the vast majority of genera of Ustilaginales, the asexual yeast morph plays a major role as a proliferating life-cycle stage in Moesziomyces and that the plant-parasitic dikaryophase is probably mainly important for maintaining the possibility of sexual recombination. As the two subclades previously referred to as M. aphidis and M. rugulosus are interspersed with the morphologically identically lineages of M. bullatus from E. crus-galli, they are probably better included in M. bullatus until more sequence data become available. It seems probable that, with the inclusion of additional smut samples from Echinochloa, additional smut-causing members of the subclades will be discovered. Sampling in Africa seems to be promising in this respect, as the species diversity of Echinochloa is highest on this continent. Also, the notion that yeasts of the subclade containing the ex-type culture of Pseudozyma aphidis can withstand high temperatures, such as the human body temperature, is suggestive of a subtropical to tropical origin of this lineage.

Further, our investigations show that the older name M. eriocauli for Eriomoeszia eriocauli, should be taken up again, as this species was found embedded within Moesziomyces.

With the synonymy of the generic name Eriomoeszia and the reappraisal of the hardly used Moesziomyces names of the smut fungi of Paspalum and Pennisetum, Moesziomyces now includes six species. It is, however, likely that additional species will have to be added because smut samples from some Poaceae genera listed as host plants for the M. bullatus complex in Vénky (2012) could not be included in the current study, such as smuts from Leersia, Panicum, Polytrias, and Uranthoecium. Given the apparently high host specificity of Moesziomyces species, it seems likely that these pathogens represent species independent from M. bullatus.

Taxonomy

Based on the phylogenetic data presented here, the following nomenclature and taxonomic changes are made.

Moesziomyces antarcticus (Goto et al.) Q.M. Wang et al., Stud. Mycol. 81: 81 (2015).

Basionym: Sporobolomyces antarcticus Goto et al., Mycologia 61: 759 (1969).

Synonyms: Candida antarctica (Goto et al.) Kurtzman et al., Yeasts: 86 (1983).

Vanrija antarctica (Goto et al.) R.T. Moore, Bibltheca Mycol. 108: 167 (1987).

Pseudozyma antarctica (Goto et al.) Boekhout, J. Gen. Appl. Microbiol. 41: 364 (1995).

Moesziomyces bullatus (J. Schröt.) Vánky, Bot. Notiser 130: 133 (1977).

Basionym: Sorosporium bullatum J. Schröt., Abh. Schles. Ges. Vaterl. Cult., Abth. Naturwiss. 72: 6 (1869).

Synonyms: Tolyposporium bullatum J. Schröt, in Cohn., Krypt. Fl. Schles. 3(1): 276 (1887).

Sterigmatomyces aphidis Henninger & Windisch, Arch. Microbiol. 105: 50 (1975).

Tolypoderma bullata (J. Schröt.) Thirum. & M.J. O’Brien, Friesia 11: 190 (1978) [“1977”].

Sporothrix rugulosa Traquair et al., Canad. J. Bot. 66: 929 (1988).

Stephanoascus rugulosus Traquair et al., Canad. J. Bot. 66: 929 (1988).

Pseudozyma aphidis (Henninger & Windisch) Boekhout, J. Gen. Appl. Microbiol. 41: 364 (1995).

Pseudozyma rugulosa (Traquair, et al.) Boekhout & Traquair, J. Gen. Appl. Microbiol. 41: 364 (1995).

Moesziomyces aphidis (Henninger & Windisch) Q.M. Wang et al., Stud. Mycol. 81: 81 (2015).

Moesziomyces rugulosus (Traquair, et al.) Q.M. Wang et al., Stud. Mycol. 81: 81 (2015).

Moesziomyces eriocauli (G.P. Clinton) Vánky, Nordic J. Bot. 6: 71 (1986).

Basionym: Tolyposporium eriocauli G.P. Clinton, Rhodora 2: 82 (1901).

Synonyms: Dermatosorus eriocauli (G.P. Clinton) M.D. Whitehead & Thirum., Mycologia 64: 128 (1972).

Tolypoderma eriocauli (G.P. Clinton) Thirum., Friesia 11: 191 (1978).

Eriomoeszia eriocauli (G.P. Clinton) Vánky, Mycol. Balcanica 2: 106 (2005).

Moesziomyces parantarcticus (Sugita et al.) Q.M. Wang et al., Stud. Mycol. 81: 81 (2015).

Basionym: Pseudozyma parantarctica Sugita et al., Microbiol. Immun. 47: 186 (2003).

Moesziomyces verrucosus (J. Schröt.) J. Kruse & Thines, comb. nov.

MycoBank MB819410

Basionym: Ustilago verrucosa J. Schröt., Hedwigia 35: 214 (1896).

Synonyms: Tolyposporium evernium Syd., Ann. Mycol. 37: 443 (1939).

Tolyposporium paspali Langdon, Univ. Queensland Dept. Biol. Pap 2(9): 4 (1948).

Moesziomyces evernius (Syd.) Vánky, Bot. Notiser 130: 135 (1977).

Tolyposporidium evernium (Syd.) Thirum. & Neerg., Friesia 11: 180 (1978) [“1977”].

Moesziomyces penicillariae (Bref.) Vánky, Bot. Notiser 130: 135 (1977).

Basionym: Tolyposporium penicillariae Bref., Unters. Gesammtgeb. Mykol. 12: 154 (1895).

Synonym: Tolyposporidium penicillariae (Bref.) Thirum. & Neerg., Friesia 11: 181 (1978) [“1977”].