Moniliellomycetes and Malasseziomycetes, two new classes in Ustilaginomycotina.

Ustilaginomycotina (Basidiomycota, Fungi) has been reclassified recently based on multiple gene sequence analyses. However, the phylogenetic placement of two yeast-like genera Malassezia and Moniliella in the subphylum remains unclear. Phylogenetic analyses using different algorithms based on the sequences of six genes, including the small subunit (18S) ribosomal DNA (rDNA), the large subunit (26S) rDNA D1/D2 domains, the internal transcribed spacer regions (ITS 1 and 2) including 5.8S rDNA, the two subunits of RNA polymerase II (RPB1 and RPB2) and the translation elongation factor 1-α (EF1-α), were performed to address their phylogenetic positions. Our analyses indicated that Malassezia and Moniliella represented two deeply rooted lineages within Ustilaginomycotina and have a sister relationship to both Ustilaginomycetes and Exobasidiomycetes. Those clades are described here as new classes, namely Moniliellomycetes with order Moniliellales, family Moniliellaceae, and genus Moniliella; and Malasseziomycetes with order Malasseziales, family Malasseziaceae, and genus Malassezia. Phenotypic differences support this classification suggesting widely different life styles among the mainly plant pathogenic Ustilaginomycotina.

INTRODUCTION

Basidiomycota (Dikarya, Fungi) contains three main phylogenetic domains, namely the subphyla Agaricomycotina, Pucciniomycotina and Ustilaginomycotina(Hibbett et al. 2007). Within Ustilaginomycotina three classes have been proposed, namely Ustilaginomycetes, Exobasidiomycetes and Entorrhizomycetes (Begerow et al. 1997, 2006, Bauer et al. 2001, 2006). Other researchers, however, questioned the status of Entorrhizomycetes and considered them incertae sedis among Basidiomycota (Matheny et al. 2006, Hibbett et al. 2007).

Ustilaginomycotina comprises plant pathogenic fungi (smuts), which are mostly dimorphic and present a yeast stage during the live cycle, and asexual fungi which are known only as yeasts or yeast-like species, e.g. Acaromyces spp., Pseudozyma spp., Meira spp., Tilletiopsis spp. and Malassezia spp. (Boekhout 1991, 1995, Boekhout et al. 1995, 2003, 2011, Begerow et al. 2000, 2006, Fell et al. 2000). Pseudozyma spp. belong to Ustilaginomycetes and the genera Acaromyces, Meira and Tilletiopsis to Exobasidiomycetes (Begerow et al. 2006, Hibbett et al. 2007, Boekhout et al. 2011). The taxonomic position of Malassezia spp., that is an important inhabitant of the human and animal skin microbiota (Guého-Kellermann et al. 2010, Sugita et al. 2010, Gaitanis et al. 2012, Findley et al. 2013), is not settled. It has been proposed that the genus belongs to Ustilaginomycotina (Begerow et al. 2000, Xu et al. 2007), but its final affiliation within this group remained problematic. The genus was treated to represent a distinct order Malasseziales in the Exobasidiomycetes based on molecular phylogenetic analyses of the nuclear ribosomal RNA genes alone or in combination with protein genes (Begerow et al. 2000, 2006, Bauer et al. 2001, Weiß et al. 2004). However, Matheny et al. (2006) suggested that the Malasseziales might be affiliated with Ustilaginomycetes. Therefore, Hibbett et al. (2007) excluded Malasseziales from Exobasidiomycetes and treated it as incertae sedis in Ustilaginomycotina.

The yeast-like genus Moniliella was described by Stolk & Dakin (1966). In the following year the genus Trichosporonoides was introduced by Haskins & Spencer (1967). Both have a basidiomycetous affinity because they can hydrolyse urea, show a positive Diazonium Blue B (DBB) staining and have multi-lamellar cell walls. The septal pores of the species studied so far (M. oedocephala, M. spathulata, M. suaveolens) show a diversity of septa. Moniliella suaveolens has typical dolipores with an arch of endoplasmic reticulum, M. spathulata has a micropore-like structure and M. oedocephala has both (Haskins 1975, Martínez 1979). Boekhout (1998) and de Hoog & Smith (1998a, b) indicated that Trichosporonoides was probably synonymous with Moniliella. Later Rosa et al. (2008) confirmed that both genera were indeed congeneric based on 26S rRNA D1/D2 domain sequence analysis and, consequently, transferred all species into the genus Moniliella. These authors were, however, not sure about the phylogenetic relationships with either Ustilaginomycotina or Agaricomycotina.

In order to clarify the phylogenetic affiliations of Malassezia and Moniliella, we performed phylogenetic analyses based on the six genes that were used to address the higher-level phylogeny of the Fungi (James et al. 2006, Hibbett et al. 2007). Our results demonstrated that Malassezia and Moniliella belong to Ustilaginomycotina where they form deep and well-supported lineages with a sister relationship to both Ustilaginomycetes and Exobasidiomycetes. We therefore propose two new classes, Malasseziomycetes and Moniliellomycetes for these lineages, with additional support from phenotypic characters.

MATERIALS AND METHODS

Taxon sampling

Almost all the recognised Malassezia and Moniliella species were employed (Table 1). Reference species representing Pucciniomycotina, Agaricomycotina and currently recognised classes of Ustilaginomycotina were selected based on sequence availability for the six genes used in the AFTOL1 project (http://aftol.org/data.php) (Table 1). Sequence data generated in this study or data retrieved from GenBank were mostly from type strains of the taxa compared.

Table 1

Taxa sampled and sequence accession numbers employed (those in bold are determined in this study).

Species	Strain	D1D2	ITS	SSU	RPB1	RPB2	EF1-α
Ustilaginomycetes
Cintractia limitata	HAJB 10488	DQ645506	DQ645508	DQ645507	DQ645510	DQ645509	DQ645511
Melanotaenium endogenum	AFTOL ID1918	DQ789979	DQ789981	DQ789980	–	–	–
Melanotaenium euphorbiae	voucher HUV17733	JN367314	JN367289	JN367342	–	–	JN367365
Rhodotorula acheniorum	AS 2.3198T	AF190001	AB038128	AJ496256	KF706499	KF706522	KF706474
Schizonella melanogramma	CBS 174.42	DQ832210	DQ832212	DQ832211	DQ832214	DQ832213	DQ832215
Sporisorium reilianum	CBS 131460	KF706430	KF706438	KF706441	–	KF706511	KF706472
Urocystis colchici	AFTOL-ID1647	DQ838576	DQ839596	DQ839595	–	DQ839597	DQ839598
Urocystis eranthidis	voucher hmk292	JN367324	JN367299	JN367352	JN367428	–	JN367375
Ustanciosporium gigantosporum	HRK 023	JN367325	JN367300	JN367353	JN367429	–	JN367376
Ustilago hordei	CBS 131470	KF706429	KF706437	KF706442	KF706498	KF706521	KF706473
Ustilago maydis	CBS 504.76/MS 115	AF453938	AY854090	X62396	XM401478	AY485636	AY885160
Ustilago tritici	CBS 669.70	DQ094784	DQ846894	DQ846895	DQ846897	DQ846896	DQ846898
Exobasidiomycetes
Exobasidium gracile	DSM 4460	DQ663699	DQ663700	DQ785786	DQ663702	DQ663701	DQ663703
Exobasidium rhododendri	CBS 101457	DQ667151	DQ667153	DQ667152	DQ667155	DQ667154	DQ667156
Jaminaea angkoriensis	C5b	EU587489	EU604147	EU604148	–	–	–
Microstroma juglandis	CBS 287.63	AF009867	DQ789988	DQ789987	DQ789990	DQ789989	DQ789991
Quambalaria cyanescens	CBS 876.73	DQ317616	DQ317623	KF706440	–	KF706531	KF706485
Rhamphospora nymphaeae	CBS 72.38	DQ831032	DQ831034	DQ831033	–	DQ831035	DQ831036
Tilletiaria anomala	CBS 436.72	AJ235284	DQ234558	AY803752	DQ234571	AY803750	DQ835991
	CBS 607.83T	AJ235282	AB025704	KF706451	–	KF706530	KF706483
Tilletiopsis washingtonensis	CBS 544.50T	AJ235278	DQ835994	AJ271382	–	DQ835995	DQ835996
Malasseziomycetes
Malassezia caprae	CBS10434T	AY743616	AY743656	KF706456	KF706495	KF706513	KF706467
Malassezia dermatis	CBS 9169T	AB070365	AY390284	KF706452	KF706490	KF706532	KF706461
Malassezia equina	CBS 9969T	AY743621	KF706439	KF706454	KF706492	KF706515	KF706463
Malassezia furfur	CBS 1878T	AF063214	AY743634	KF706457	KF706497	KF706516	KF706469
Malassezia globosa	CBS 7966T	AF064025	AY387132	–	KF706493	KF706518	KF706465
Malassezia japonica	CBS 9431T	EF140672	EF140669	KF706458	–	KF706514	KF706464
Malassezia nana	CBS 9558T	EF140673	EF140667	KF706453	KF706491	KF706510	KF706462
Malassezia obtusa	CBS 7876T	AB105197	AY387137	KF706455	–	KF706519	KF706470
Malassezia pachydermatis	CBS 1879T	AY743605	AB118941	DQ457640	DQ785792	DQ408140	DQ028594
Malassezia restricta	CBS 7877T	AF064026	AY743636	EU192367	KF706496	KF706520	KF706471
Malassezia slooffiae	CBS 7956T	AJ249956	AY743633	KF706459	–	–	–
Malassezia sympodialis	CBS 7222T	AF064024	AY743632	KF706460	–	–	–
Malassezia yamatoensis	CBS 9725T	AB125263	AB125261	–	KF706494	KF706512	KF706466
Moniliellomycetes
Moniliella acetoabutens	CBS 169.66T	AF335523	EU252153	KF706443	KF706500	KF706523	KF706476
Moniliella madida	CBS 240.79T	AF335522	–	KF706447	KF706502	KF706525	KF706478
Moniliella megachiliensis	CBS 190.92T	EF137916	KF706433	KF706448	KF706501	KF706524	KF706477
Moniliella mellis	CBS 350.33T	EU545185	–	KF706446	–	KF706528	KF706481
Moniliella nigrescens	CBS 269.81T	AF335527	KF706436	–	KF706504	KF706527	KF706480
Moniliella oedocephalis	CBS 649.66T	AF335521	KF706435	KF706449	–	–	KF706484
Moniliella pollinis	CBS 461.67T	AF335525	KF706434	KF706450	KF706505	KF706529	KF706482
Moniliella spathulata	CBS 241.79T	AF335526	KF706432	KF706444	KF706503	KF706526	KF706479
Moniliella suaveolens	CBS 126.42T	AF335520	KF706431	KF706445	–	–	KF706475
Pucciniomycotina
Bensingtonia ciliata	AS 2.1945T	AF189887	AF444563	D38233	KF706509	KF706536	KF706486
Chrysomyxa arctostaphyli	CFB22246	AY700192	DQ200930	AY657009	–	DQ408138	DQ435789
Endocronartium harknessii	CFB22250	AY700193	DQ206982	AY665785	–	DQ234551	DQ234567
Erythrobasidium hasegawianum	AS 2.1923T	AF189899	AF444522	D12803	KF706506	KF706534	KF706488
Naohidea sebacea	CBS 8477	DQ831020	DQ911616	–	KF706508	KF706535	KF706487
Platygloea disciformis	IFO32431	AY629314	DQ234556	DQ234563	–	DQ234554	DQ056288
Puccinia graminis	CRL75-36-700-3/ECS	AF522177	AF468044	AY125409	XM_003334476	XM_003321826	XM_003333024
Sporidiobolus salmonicolor	CBS 490T	AF070439	AY015434	AB021697	KF706507	KF706533	KF706489
Agaricomycotina
Auriculibuller fuscus	CBS 9648T	AF444763	AF444669	KF036604	KF036314	KF036727	KF036999
Bulleromyces albus	CBS 501T	AF075500	AF444368	X60179	KF036334	KF036745	KF037016
Boletellus projectellus	MB03-118	AY684158	AY789082	AY662660	AY788850	AY787218	AY879116
Dacryopinax spathularia	GEL 5052	AY701525	AY854070	AY771603	AY857981	AY786054	AY881020
Filobasidiella depauperata	CBS 7841T	AF487884	EF211248	AJ568017	KF036417	KF036885	KF037150
Wallemia clade
Wallemia ichthyophaga	EXF994/EXF1059	DQ847516	AY302523	AY741382	DQ847522	DQ847519	DQ847525
Wallemia muriae	MZKI-B952/EXF1054	DQ847517	AY302534	AY741381	DQ847523	DQ847520	DQ847526
Wallemia sebi	EXF483	DQ847518	AY328917	AY741379	DQ847524	DQ847521	DQ847527
Ascomycota
Aleuria aurantia	OSC 100018	AY544654	DQ491495	NG013139	DQ471120	DQ247785	DQ466085
Aspergillus nidulans	NRRL 2395/FGSC4	EF652445	AY373888	U77377	XM_653321	XM_677297	XM_656730
Neurospora crassa	NRRL 13141/ICMP 6360	AY681158	AY681193	AY046271	XM_959004	AF107789	AF402094
Schizosaccharomyces pombe	972H-	CU329672	CU329672	CU329672	D13337	X56564	NC_003421

PCR and DNA sequencing

Genomic DNA was extracted from the living cultures grown on the Yeast Extract Peptone Dextrose (YPD) plate using the method as described by Bolano et al. (2001). A set of six genes were selected, including three protein-coding genes, namely RPB1 (the largest subunit of RNA polymerase II), RPB2 (the second largest subunit of RNA polymerase II) and EF1-α (translation elongation factor 1-α), and three rRNA genes namely small subunit (SSU or 18S) ribosomal DNA (rDNA) D1/D2 domains of the large subunit (LSU or 26S) rDNA, and the ITS 1+2 regions (including 5.8S rDNA) of the rDNA. PCR and sequencing of the rRNA genes were performed using the methods described previously (Fell et al. 2000, Scorzetti et al. 2002, Wang et al. 2003). PCR and sequencing primers for RPB1, RPB2 and EF1-α are listed in Table 2. PCR parameters for amplifying RPB1 and RPB2 were as follows: an initial denaturation step at 94 °C for 4 min; 36 cycles of denaturation at 94 °C for 1 min, annealing at 50–52 °C for 1 min and extension at 72 °C for 1 min; and a final extension step of 8 min at 72 °C. Amplification of the EF1-α gene used a touchdown PCR: an initial denaturation at 94 °C for 4 min; an annealing temperature of 62 °C in the first cycle, successively reducing the Tm by 1 °C per cycle over the next nine cycles to reach a final Tm of 52 °C, which is used in the remaining 30 cycles and extension at 72 °C for 1 min; and a final extension step of 8 min at 72 °C. Cycle sequencing was performed using the ABI BigDye cycle sequencing kit (QIAGEN, Valencia, California). Electrophoresis was done on an ABI PRISM 3710 or 3730 DNA sequencer.

Table 2

PCR and sequence primers used.

Locus	Primers (5’-3’)
RPB1	RPB1-Af: GAR TGY CCD GGD CAY TTY GG
	RPB1-Cr: CCN GCD ATN TCR TTR TCC ATR TA
RPB2	fRPB2-5F: GAY GAY MGW GAT CAY TTY GG
	fRPB2-7cR: CCC ATR GCT TGY TTR CCC AT
	bRPB2-6F: TGG GGY ATG GTN TGY CCY GC
	gRPB2-6R: GCA GGR CAR ACC AWM CCC CA
EF1-α	EF1-983F: GCY CCY GGH CAY CGT GAY TTY AT
	EF1-2218R: AT GAC ACC RAC RGC RAC RGT YTG
	EF1-1577F: CAR GAY GN TAC AAG ATY GGT GG
	EF1-1567R: AC HGT RCC RAT ACC ACC RAT CTT

Molecular phylogenetic analyses

Sequences were aligned with the Clustal X program (Thompson et al. 1997). The alignment datasets were firstly analysed with Modeltest v. 3.04 (Posada & Crandall 1998) using the Akaike information criterion (AIC) to find the most appropriate model of DNA substitution. A general time-reversible model of DNA substitution additionally assuming a percentage of invariable sites and Γ-distributed substitution rates at the remaining sites (GTR+I+G) was selected for further analyses. Maximum likelihood (ML) analyses were conducted in MEGA v. 5 (Tamura et al. 2011). Maximum parsimony (MP) analysis was conducted in PAUP v. 4.0b10 (Swofford 2002) where the support of the branching topologies was derived from 1 000 replicates with 10 random additions. Three to four ascomycetous species were used as outgroups. Bayesian posterior probability analyses were conducted in MrBayes v. 3.2 (Ronquist et al. 2012) with parameters set to 1 000 000 generations, four runs and four chains. The chains were heated to 0.25 and a stop value of 0.01 was used. The alignment matrix was deposited in TreeBASE (www.treebase.org) with submission ID 14907.

RESULTS

A total of 108 new sequences were generated from 31 species in this study (Table 1). New sequences and those retrieved from GenBank generated from the same species were concatenated in different combinations to form three separate datasets as follows: 1) a rRNA gene dataset formed by SSU, LSU D1/D2 and ITS (including 5.8S rDNA); 2) a protein gene dataset consisted of RPB1, RPB2 and EF1-α; and 3) a six-gene dataset formed by the combination of the former two datasets as used in James et al. (2006). All the datasets were subjected to ML, MP and Bayesian analyses respectively and the topologies of the trees obtained were visually examined for phylogenetic concordance.

In the majority of the trees, four distinct monophyletic clades, namely Exobasidiomycetes, Malassezia, Moniliella and Ustilaginomycetes, were consistently resolved within Ustilaginomycotina (Table 3, Fig. 1, 2). The Malassezia, Moniliella and Ustilaginomycetes clades each received 1.0 post probability and 99–100 % bootstrap support in the trees constructed using different algorithms based on the six-gene dataset. These three clades were also clearly resolved and strongly supported (1.0 post probability and 92–100 % bootstrap values) in the trees drawn from the rRNA gene and the protein gene datasets, except for the Ustilaginomycetes clade which received weak (51–65 %) bootstrap support in the ML and MP trees drawn from the rRNA gene dataset (Table 3, Fig. 2). The Exobasidiomycetes was resolved as a monophyletic clade in the trees constructed from the rRNA gene and the six-gene datasets, but only received strong support in the Bayesian and ML trees drawn from the rRNA gene dataset. This clade was shown to be non-monophyletic in all the trees drawn from the protein gene dataset (Table 3, Fig. 1, 2). In the Bayesian and ML trees (Fig. 2d, e), Tilletiopsis fulvescens and Tilletiaria anomala formed a separate branch, while the remaining taxa of the Exobasidiomycetes formed an unsupported (0.6 post probability) group in the Bayesian tree (Fig. 2d) and two separate groups in the ML tree (Fig. 2e). In the MP tree (Fig. 2f), two Exobasidium species formed a separate branch from the other Exobasidiomycetes taxa, which in turn formed an unsupported group.

Table 3

Statistical support values for the major clades resolved in Ustilaginomycotina.

Clade	Three rRNA genes			Three protein genes			Combined six genes
	PP	MLBP	MPBP	PP	MLBP	MPBP	PP	MLBP	MPBP
Exobasidiomycetes	1.0	76	< 50	nm	nm	nm	1.0	51	< 50
Malassezia	1.0	100	100	1.0	100	100	1.0	100	100
Moniliella	1.0	100	100	1.0	100	100	1.0	100	100
Ustilaginomycetes	1.0	51	65	1.0	92	98	1.0	99	99

MLBP = bootstrap percentage from maximum likelihood analysis.

MPBP = bootstrap percentage from maximum parsimony analysis.

PP = posterior probability from Bayesian analysis.

Nm = not monophyletic.

Fig. 1

Phylogenetic tree constructed using Bayesian analysis of the combined sequences of 18S rDNA, 26S rDNA D1/D2 domains, ITS regions (including 5.8S rDNA), RPB1, RPB2 and EF1-α depicting the phylogenetic placements of genera Moniliella and Malassezia within the Ustilaginomycotina. Branch lengths are scaled in terms of expected numbers of nucleotide substitutions per site. Bayesian posterior probabilities and bootstrap percentages from 1 000 replicates of maximum likelihood and maximum parsimony analyses are shown respectively from left to right on the deep and major branches resolved.

Fig. 2

Phylogenetic trees based on datasets of rRNA genes including 18S, D1/D2 domains of 26S and ITS-5.8S (a, b and c); protein genes including RPB1, RPB2 and EF1-α (d, e and f) and the combined six genes (g, h and i) using Bayesian (a, d and g), maximum likelihood (b, e and h) and maximum parsimony (c, f and i) analyses, showing the major clades resolved within the Ustilaginomycotina. Bayesian posterior probabilities above 0.7 or bootstrap percentages over 50 % from 1 000 replicates are shown.

The phylogenetic relationships among the Exobasidiomycetes, Malassezia, Moniliella and Ustilaginomycetes clades were clearly resolved by Bayesian analysis of the six-gene dataset, showing that the Exobasidiomycetes clade was located basal to the other three clades; Malassezia and Moniliella were sister clades and Ustilaginomycetes was basal to them. Their relationships were all strongly supported by 1.0 post probability (Fig. 1, 2g). However, the relationships among the four clades within Ustilaginomycotina were largely unresolved in the other trees (Fig. 2).

DISCUSSION

Basidiomycetous species in the subphylum Ustilaginomycotina are usually dimorphic, producing a saprobic haploid yeast phase and a parasitic dikaryotic hyphal phase. A considerable number of cultivable ustilaginomycetous fungi are only known by asexual yeasts and yeast-like microbes and are classified mainly based on physiological, biochemical and molecular criteria commonly used for yeasts, forming a taxonomic system hitherto independent from that of filamentous fungi (Boekhout 1991, Boekhout et al. 2011). Molecular methods recently showed the affiliation of these yeasts with various filamentous smuts and merged the two groups into a unified taxonomic system (Bauer et al. 2001, Begerow et al. 2000, 2006, Weiß et al. 2004, Matheny et al. 2006, Boekhout et al. 2011). However, the fine phylogenetic positions of the yeast and yeast-like groups have been a matter of debate. Here we show that the Malassezia and Moniliella clades present two deeply rooted lineages within the smut fungi (Ustilaginomycotina, Basidiomycota, Fungi). They also possess unique phenotypic (morphological, ultrastructural, physiological and biochemical) characters distinct from those of Ustilaginomycetes and Exobasidiomycetes.

The genus Moniliella was originally classified in the order Moniliales of fungi imperfecti (Stolk & Dakin 1966). A sexual morph has not been observed (de Hoog 1979a) and all members of the genus have greyish to olivaceous black colonies, yeast-like growth, hyphae that disarticulate in arthroconidia (Martinez et al. 1979, de Hoog 1979b), and CoQ-9 as the major ubiquinone (de Hoog et al. 2011). All species, except M. fonsecae, can ferment glucose and some species also galactose, sucrose and/or raffinose (Martínez et al. 1979, de Hoog et al. 2011), which among Basidiomycota is a rare trait. Some species are considered xerophilic (Hocking & Pitt 1981, de Hoog et al. 2011) and most are known from industrial settings, food stuffs, fats, oils and acids or substrates with low water activity or were isolated from flowers in tropical forest ecosystems (de Hoog et al. 2011). Thus their origin and ecology also differs from the vast majority of Ustilaginomycotina that typically are plant pathogens (Begerow et al. 2006). Glucose, galactose and mannose (low) were found to be present in the cell walls of Moniliella species, whereas xylose and fucose were absent (Weijman 1979), which led this author to conclude that the genus may not be related to the Agaricomycotina. Ustilaginomycotina, as analysed thus far, have glucose as the dominant cell wall sugar, with some galactose and mannose, but without xylose (Dörfler 1990, Boekhout et al. 1992, Bauer et al. 1997, 2001, Prillinger et al. 2011, van der Klei et al. 2011), thus agreeing with the cell wall composition as known from Moniliella species.

The septal ultrastructure of the Moniliella species as investigated so far revealed a rather complex pattern with dolipores in M. suaveolens, micro-pore-like structures in M. spathulata and both types occurring in M. oedocephala. The dolipores resemble those of Agaricomycetes but with an arch of endoplasmic reticulum instead of parenthesomes (Haskins 1975, Martinez 1979, Bauer et al. 2006, van Driel et al. 2009). These differences in pore structures may be growth stage dependent, and differ between yeast-like, hyphal and arthroconidial stages, but they may also be caused by the fixation protocols used. The dolipore structure is quite different from the simple pore with membrane caps observed in the Ustilaginomycotina and the simple pore without membrane apparatus typical for the Pucciniomycotina (Boekhout et al. 1992, Bauer et al. 1997, 2001, 2006, van der Klei et al. 2011). The biochemical and ultrastructural characters made the taxonomic and phylogenetic positions of Moniliella elusive for a long time. However, our results clearly show that the Moniliella species belong to the Ustilaginomycotina with strong support (Fig. 1, 2).

Our analyses showed that the species of Malassezia formed a highly supported deep lineage in the Ustilaginomycotina that we rank at the class level. This is further sustained by the unique monopolar budding, the thick helicoidal cell wall, the lipid dependency of most Malassezia species and the lipophily of M. pachydermatis (Guého-Kellermann et al. 2011). Malassezia species occur commonly on human and animal skin, but metagenomic data also revealed the presence of the species in some unexpected habits, such as forest soil, nematodes or corals (Guého-Kellermann et al. 2010, Gaitanis et al. 2012). The genome comparison between Ustilago maydis and M. globosa revealed some major differences in the enzyme armamentarium used by human inhabiting Malassezia yeasts that differs from that of the plant pathogen U. maydis (Xu et al. 2007, Sun et al. 2013).

Our analyses of the three rRNA genes and the combined six genes indicated that Exobasidiomycetes seem to be monophyletic when the Malassezia clade is excluded, but it remains polyphyletic in the analysis of the three protein-coding genes. Begerow et al. (1997, 2000) and Bauer et al. (2001) showed that the Exobasidiomycetes were weakly supported or with 56–85 % bootstrap support values based on the LSU rDNA analysis. Begerow et al. (2006) indicated that Exobasidiomycetes were paraphyletic using a combined analysis of SSU, D1/D2, ITS, atp6 and β-tubulin sequences, but this lineage was found to be monophyletic when the SSU data were excluded, but with weak bootstrap support. Matheny et al. (2006) indicated that the taxa within Exobasidiomycetes except the Malassezia clustered together with or without statistical support based on the different gene combination datasets. Thus, for a better understanding of the phylogeny of Exobasidiomycetes further analyses using more molecular data or genomic data are needed.

In addition to the Clustal X, we also used the MAFFT program (Katoh & Standley 2013) to align the sequences and the Gblocks program (Talavera & Castresana 2007) to remove ambiguously aligned blocks from the alignments. The alignments produced by the MAFFT and those treated by the Gblocks were all subjected to Bayesian, ML and MP analyses. The consensus was that the Malassezia, Moniliella and Ustilaginomycetes taxa were respectively resolved as strongly supported monophyletic clades in all the trees obtained, while the Exobasidiomycetes was monophyletic without statistic support in the trees drawn from the six-gene dataset but polyphyletic in the trees drawn from the rRNA gene and the protein gene datasets (data not shown). The polyphyletic nature of the Exobasidiomycetes was magnified by using the new programs, implying that this group may not represent a single class.

In conclusion, multiple gene phylogenetic analyses and phenotypic comparisons suggest that the species of Malassezia and Moniliella form two independent deep lineages representing sister groups to the recognised classes Ustilaginomycetes and Exobasidiomycetes within Ustilaginomycotina. Therefore, we propose two new classes to accommodate these fungi.

TAXONOMY

Q.M. Wang, F.Y. Bai & Boekhout, class. nov. — MycoBank MB805229

Type order. Moniliellales.

Etymology. The nomenclature of the class is derived from the generic name of Moniliella Stolk & Dakin, Antonie van Leeuwenhoek 32: 399. 1966.

Member of Ustilaginomycotina. Sexual morph unknown. Colonies are smooth or velvety, greyish to olivaceous black. Budding cells are ellipsoidal and form terminally on true hyphae that disarticulate with artroconidia. Pseudohyphae and chlamydospores may be present. Cell walls are multi-lamellar. Hyphal septa typically possess dolipores with an arch of endoplasmic reticulum, but ‘micropore’-like structures may also be present. Sugars are fermented by most species. Nitrate is assimilated. Urease and diazonium blue B (DBB) reactions are positive. Coenzyme Q-9 is present. Xylose and fucose are absent from whole-cell hydrolysates.

Q.M. Wang, F.Y. Bai & Boekhout, ord. nov. — MycoBank MB805230

Type family. Moniliellaceae.

The diagnosis of the order Moniliellales is based on the description of the class Monilielliomycetes. The nomenclature of the order is based on the genus Moniliella Stolk & Dakin, Antonie van Leeuwenhoek 32: 399. 1966.

Q.M. Wang, F.Y. Bai & Boekhout, fam. nov. — MycoBank MB805231

Type genus. Moniliella Stolk & Dakin, Antonie van Leeuwenhoek 32: 399. 1966.

The diagnosis of the family Moniliellaceae is based on the description of the order Moniliellales. The nomenclature of the family is based on the genus Moniliella Stolk & Dakin, Antonie van Leeuwenhoek 32: 399. 1966.

Boekhout, Q.M. Wang & F.Y. Bai, class. nov. — MycoBank MB805514

Type order. Malasseziales R.T. Moore with family Malasseziaceae Denchev & R.T. Moore, Mycotaxon 110: 379. 2009, and genus Malassezia Baillon (1889).

Etymology. The nomenclature of the class is derived from the order name of Malasseziales R.T. Moore, Bot. Mar. 23: 371. 1980, emend. Begerow et al., Mycol. Res. 104: 59. 2000.

Member of Ustilaginomycotina. Sexual morph unknown. Cells are globose, ovoid or cylindrical. Budding is typically monopolar on a more or less broad base, enteroblastic and percurrent. The cell wall is multi-lamellate, and the inner layer of the cell wall is corrugated with a groove spiralling from the bud site. The species are lipid dependent or lipophilic. Sugars are not fermented. Urease and diazonium blue B (DBB) reactions are positive. Coenzyme Q-9 is formed. Xylose is absent in whole-cell hydrolysates.