Taxonomy and Phylogeny of the Favolaschia calocera Complex (Mycenaceae) with Descriptions of Four New Species

: Favolaschia calocera was originally described from Madagascar, and reported to have a worldwide distribution. In the current study, samples of the Favolaschia calocera from Central America, Australia, China, Kenya, Italy, New Zealand, and Thailand were analyzed by using both morphological and molecular methods. Phylogenetic analyses were based on the internal transcribed spacer (ITS) dataset, and the combined ﬁve-locus dataset of ITS, large subunit nuclear ribosomal RNA gene (nLSU), the small subunit mitochondrial rRNA gene (mt-SSU), the small subunit of nuclear ribosomal RNA gene (nu-SSU), and the translation elongation factor 1 α (TEF1). Our study proves that Favolaschia calocera is a species complex, and six species are recognized in the complex including four new species. Three new species F. brevibasidiata , F. brevistipitata , and F. longistipitata from China; and one new species F. minutissima from Asia. In addition, Favolaschia claudopus (Singer) Q.Y. Zhang & C. Dai, earlier treated as a variety of Favolaschia calocera R. Heim, were raised to species rank. Illustrated descriptions of these ﬁve new taxa are given. An identiﬁcation key and a comparison of the characteristics of species in the Favolaschia calocera complex are provided.

Favolaschia calocera R. Heim was first described from Madagascar [11]. It differs from other species in the genus by its bright orange or yellow basidiocarps with a distinct laterally stipe and numerous gloeocystidia and acanthocysts. In addition, Favolaschia calocera is a conspicuous fungus, and it is known to occur in different environments, such as forests or shrub-lands, non-forested rural areas, home or public gardens etc., and occurs on over 50 different plant species, such as decaying dicotyledonous tree trunks, logs, and branches [12,13].
Favolaschia calocera has always been considered to be an invasive fungus by several countries [14,15]. Outside its type locality in Madagascar, it was found in New Zealand and Italy in 1969 and 1999 [13,16]. Johntson et al. [12,13] mentioned that Favolaschia calocera spread to Australia and New Zealand presumably by shipped timber. Vizzini and Zotti [16] believed that Favolaschia calocera in Norfolk Island was introduced from New Zealand. Over the past couple of decades, it has spread into Europe. In the process of the migration, the morphology and molecules of Favolaschia calocera gradually diverged. As early as in 1974, Singer [8] reported the morphological differences of Favolaschia calocera collections from Madagascar and New Zealand. Based on molecular phylogeographic analysis, Vizzini et al. [14] and Ainsworth et al. [17] indicated Favolaschia calocera is divided into two main groups: the European and Oceania samples clustered in one group, and the American and Asian materials formed another group. Until recently, there has been no systematic study on the distribution of this species in China. Our initial investigation found that Favolaschia calocera is common in tropical China, and displays a higher genetic variability.
The aim of this study was to investigate the diversity and phylogeny within the Favolaschia calocera species complex, and five new taxa were described. The key characteristics for distinguishing species are provided.

Morphological Studies
Studied specimens collected by authors or new collections from China and Australia are deposited at the herbarium of the Institute of Microbiology, Beijing Forestry University (BJFC). Sections of basidiocarps were studied microscopically according to Cui et al. [18] using a Nikon Eclipse 80i microscope with phase contrast illumination. Color terms are cited from Petersen [19]. In presenting spore size variation, 5% of measurements were excluded from each end of the range and this value is given in parentheses. The following abbreviations are used in the description: KOH = 5% potassium hydroxide; IKI = Melzer's reagent; IKI+ = amyloid; CB = Cotton Blue; CB− = acyanophilous; L = arithmetic average of all spores; W = arithmetic average of all spores; Q = the L/W ratio; n = number of spores measured from the given number of specimens.

DNA Extraction and Sequencing
A CTAB rapid plant genome extraction kit (Aidlab Biotechnologies, Co., Ltd., Beijing, China) was used to obtain PCR products from dried specimens, following the manufacturer's instructions with some modifications [20]. To generate PCR amplicons, the following primer pairs were used: ITS5 and ITS4 for ITS, and LR0R and LR7 for nLSU, MS1 and MS2 for mt-SSU, NS1 and NS4 for nu-SSU, and 983F and 1567R for TEF1 [21,22].
The PCR cycling schedules for different DNA sequences used in this study followed those used in Zhou et al. and Liu et al. [23,24] with some modifications. The PCR procedure for ITS, mt-SSU, and TEF1 was as follows: initial denaturation at 95 • C for 3 min, followed by 35 cycles at 94 • C for 40 s, 54 • C for 45 s, and 72 • C for 1 min, and a final extension of 72 • C for 10 min. The PCR procedure for nLSU was as follows: initial denaturation at 94 • C for 1 min, followed by 35 cycles at 94 • C for 30 s, 50 • C for 1 min and 72 • C for 1.5 min, and a final extension of 72 • C for 10 min. The PCR procedure for the nu-SSU was as follows: initial denaturation at 94 • C for 1 min, followed by 34 cycles at 94 • C for 30 s, 53 • C for 1 min, and 72 • C for 1.5 min, and a final extension of 72 • C for 10 min. The PCR products were purified and sequenced at the Beijing Genomics Institute (BGI), China with the same primers and the sequences were deposited in GenBank and are listed in Table 1.

Phylogenetic Analyses
Besides the newly generated sequences for this study, other related sequences from GenBank were also included in the phylogenetic analysis. Panellus stipticus (Bull.) P. Karst. is used as the outgroup in Figure 1 [26], and Favolaschia varariotecta Singer is used as the outgroup in Figure 2 [14]. The sequences used ClustalX [27] and were manually adjusted in BioEdit [28]. Trees were shown in TreeView 1.6.6 [29].
New sequences are shown in bold. Maximum parsimony (MP) analysis was used for the ITS and the ITS + nLSU + mt-SSU + nu-SSU + TEF1 datasets in PAUP* 4.0b10 [30]. Trees were generated using 100 replicates of random stepwise addition of sequence and tree-bisection reconnection (TBR) branch-swapping algorithm, with all characters given equal weight. Branch supports for all parsimony analyses were estimated by performing 1000 bootstrap (BT) replicates [31]. Descriptive tree statistics, tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RC), and homoplasy index (HI) were calculated for each maximum parsimonious tree generated. RAxML 7.2.8 (GitHub, San Francisco, CA, USA) was used for maximum likelihood (ML) analysis. The default settings of the GTR+I+G model were used for all parameters in the ML analysis [32]. The branch support values were obtained using nonparametric bootstrapping with 1000 replicates [33].
Bayesian interference (BI) analyses were calculated with MrBayes 3.1.2 [34]. Four Markov chains were run for 5,000,000 generations until the split deviation frequency value was less than 0.01 and trees were sampled every 100 generations. Branches that received bootstrap support for BS (bootstrap support for MP and ML) values and BPPs (Bayesian posterior probabilities for BI) simultaneously not less than 50% and 0.70, respectively, were considered as significantly supported.

Molecular Phylogeny
The ITS base contained 57 sequences representing 23 taxa. The ITS dataset had an aligned length of 816 characters, of which 522 characters are constant, 122 characters are variable but parsimony uninformative, and 172 characters are parsimony informative. MP analysis yielded two equally parsimonious trees (TL = 542, CI = 0.675, RI = 0.847, RC = 0.572, HI = 0.325). BI and ML analyses produced consensus trees similar to MP analysis, and only the MP tree is shown. BI showed an average standard deviation of split frequencies = 0.006573.
The combined five-gene (ITS + nLSU + mt-SSU + nu-SSU + TEF1) dataset of 25 samples represented 7 taxa. It had an aligned length of 3470 characters, of which 3358 characters are constant, 65 characters are variable but parsimony uninformative, and 87 characters are parsimony informative. MP analysis yielded one tree (TL = 126, CI = 0.913, RI = 0.941, RC = 0.859, HI = 0.087). BI and ML analyses produced consensus trees similar to MP analysis, and only the MP tree is shown. BI showed an average standard deviation of split frequencies = 0.009683.
The phylogeny (Figure 1), based on ITS, shows that the five new taxa and Favolaschia calocera from Madagascar form six distinct lineages with robust support and cluster in the F. calocera complex clade. The phylogeny (Figure 2) based on ITS + nLSU + mt-SSU + nu-SSU + TEF1 results in a similar topology to the phylogeny based on ITS sequences, and is nested within the Favolaschia calocera complex, forming six distinct lineages. Etymology. Brevibasidiata (Lat.): referring to the species having short basidia. Basidiocarps annual, gregarious, gelatinous. Pileus 2-7 × 1.5-4 mm, orbicular, apricot-orange when fresh, cream buff when dry; pileal surface slightly undulated in a reticulate pattern matching the pores below, sometimes faintly pruinose when dry. Hymenophore concolorous with pileal surface, poroid, 50-120 pores per basidiocarp; pores 0.4-1 mm in diam, pentagonal to somewhat irregular or polygonal in shape, larger near the base and smaller near the edge, the marginal pores often incomplete, pore edges pruinose when dry. Stipe obvious, laterally attached, concolorous with pileus, cylindrical or tapered to a slightly wider base, sometimes curved, very finely velutinate under a lens, 2-8 mm long.
According to the molecular analysis performed by Johnston et al. [13] and Vizzini et al. [14], these collections of Favolaschia calocera from New Zealand, Kenya, and Italy clustered in one group represent very recent, probably human-vectored introductions; on the other hand, samples of Madagascar and Asia (southern China and Thailand) display a higher genetic variability and could represent the natural distribution of this fungus. Our study is consistent with the speculation that the Australian samples cluster with the specimens from New Zealand, Kenya, and Italy and are described as Favolaschia claudopus.
The natural distribution of Favolaschia calocera complex is a controversial question, which is being explored by taxonomists. Two hypotheses have emerged. One hypothesis suggested that Favolaschia calocera as well as other Favolaschia species, during the break-up of Gondwana (about 150 million years ago), drifted away from Madagascar, where they likely evolved, with the India-Seychelles landmass, and subsequently dispersed through Asia. A second hypothesis, contrary to the first, suggested that Favolaschia calocera was accidentally introduced into Madagascar from Asia, by an impressive one-way human migratory flow from Southeast Asia (Borneo) to Madagascar in the centuries between 200 and 500 Anno Domini [14,[35][36][37][38]. As in our analysis, these Asian samples of Favolaschia calocera complex show higher variability than Madagascar. Our investigation revealed that Favolaschia calocera complex is common in tropical China and four new species have been discovered. However, to date, only Favolaschia calocera has been recorded in Madagascar in F. calocera complex, as mycologists have been active in Central America since the late1800s and it is unlikely they would have overlooked such a brightly colored and conspicuous fungus. Thus, our data tend to support the second hypothesis, but to confirm these hypotheses, a molecular clock analysis on all the Favolaschia species is needed.
Key to species of the Favolaschia calocera complex.