Phylogenetic Analyses of Some Melanoleuca Species (Agaricales, Tricholomataceae) in Northern China, With Descriptions of Two New Species and the Identification of Seven Species as a First Record

Two new species (Melanoleuca galerina and M. subgrammopodia) and seven new recorded species from northern China are described here using morphological and molecular methods. Melanoleuca galerina is mainly characterized by its hygrophanous pileus, decurrent lamellae, fibrous stipe and spores with round warts. Key characteristics of M. subgrammopodia include its discolored pileus, fibrous stipe and urticiform cystidia. The divergence time of Melanoleuca fungi as well as the phylogenetic relationships within this genus were analyzed using DNA sequences of the internal transcribed spacer (ITS) and the nuclear large subunit rDNA (nrLSU) gene fragments. Analyses revealed that morphological identifications and phylogenetic relationships were consistent with the results of divergence time, thereby confirming that M. galerina and M. subgrammopodia are new species.


INTRODUCTION
Individuals from Melanoleuca Pat. are widely distributed throughout the world and include some edible species (Singer, 1986). Currently, Index Fungorum 1 lists 422 validly published species within Melanoleuca. Whereas, Kirk et al. (2008) only accepted 50 species from this genus. In respect of China, a total of 19 Melanoleuca species have been reported (Bau and Li, 1999;Zhang et al., 2001;Chen, 2007;Mao, 2009;Sun et al., 2012;Wang, 2013;He et al., 2014;Yu et al., 2014;Zhao et al., 2014;Wei et al., 2015;Du et al., 2016;Tian et al., 2018). Melanoleuca is a genus with limited morphological characteristics. It is mainly characterized by a collybioid to tricholomatoid habit, a rarely bright-colored pileus, with warty and strongly amyloid spores and having either absent or present cystidia (Singer, 1986). The divisions among Melanoleuca species are vague. Different taxonomists often have their own interpretations of morphological features, resulting in different classifications of the same species based on those differing opinions (Singer, 1986;Boekhout, 1988;Hyde et al., 2013). The following division views are currently widely approved. Singer (1962Singer ( , 1986 divided Melanoleuca into four sections primarily based on the color of the pileus or stipe and the ornamentation of spores. Boekhout (1988) focused mainly on microscopic features, on the basis of presence and shape of cystidia and divided the genus into three subgenera namely: Subg. Melanoleuca Boekhout, Subg. Urocystis Boekhout and Subg. Macrocystis Boekhout. Bon (1991) introduced the spore Q value for delimiting subsections and divided the genus into three subgenera and eight sections (Vizzini et al., 2011). Before the advent of molecular phylogenetic tools, the taxonomical units of Boekhout (1988) were most frequently used for morphological identification of Melanoleuca. However, Vizzini et al. (2011) divided Melanoleuca into two subgenera (Subg. Urticocystis Vizzini and Subg. Melanoleuca Vizzini) by constructing phylogenetic relationships with the internal transcribed spacer (ITS) sequences. Then, Yu et al. (2014) and Kalmer et al. (2018) confirmed the results of Vizzini et al. (2011). Even now, the division views proposed by Vizzini et al. (2011) are widely supported by molecular systematics (Yu et al., 2014;Kalmer et al., 2018).
Some reports have established divergence times in the fungi. For example, Hennig (1966) first recommended the use of divergence time as a universal criterion for taxa ranking. Berbee and Taylor (2010), Oberwinkler (2012), and Hibbett (2014) estimated the divergence times of Basidiomycota and Ascomycota. Time-trees indicated that Basidiomycota are estimated to have diverged around 500 million years ago (Ma), and is a sister group to Ascomycota, being of similar age. Agaricomycetes diverged about 290 Ma as estimated by Floudas et al. (2012). Zhao et al. (2016) first attempted creating a taxonomic system of fungi based on divergence time which was a reconstruction of the taxonomic system used for Agaricus. Furthermore, Chen et al. (2017) utilized the taxonomic system based on divergence time for reconstruction of Agaricus subg. Minores, A. sect. Minores (Fr.) Henn. Additionally, Zhao et al. (2017) proposed that the divergence time of a lineage could be used as a universal criterion for ranking taxa and estimated the divergence time of Basidiomycota. Until now, divergence time has not been introduced into studies of Melanoleuca, moreover, its possible usage in taxonomic studies of the genus needs further verification.
Therefore, in the present study, we made use of universal sequences in fungi (ITS + nrLSU) to estimate the divergence time of Melanoleuca, and discussed the infrageneric classifications of 19 species from northern China based on morphological identifications and clarified their phylogenetic relationships in order to provide a theoretical basis for the study of Melanoleuca.

Materials and Morphological Observations
All samples were collected from 2011 to 2018 from northern China, and have been deposited in the Herbarium Mycology of Jilin Agricultural University (HMJAU) and Herbarium Mycology of Jinlin Agricultural Science and Technology University (HMJU). The specific details are shown in Table 1. Pictures of the habitats were taken by a Canon 80D camera. Macroscopic features were recorded using fresh collections. Color descriptions were based upon the classifications made by Kornerup and Wanscher (1978). Dried specimens were used for microscopic observations, using 5% KOH as the floating agent, Melzer's reagent was used to examine the presence of amyloid or dextrinoid reactions. Slices of lamellae and pileipellis were observed under the Olympus BX 53 microscope. Free hand drawings were made from all microscopic observations. Shooting and measurements of anatomical features were presented in the Cellsens Standard. The data was recorded by (a) b-c × d-e (f), n was the number of examined basidiospores and Q (length: breadth ratios) was calculated from 30 mature basidiospores of 3 basidiocarps. Cystidial shapes were described as per Vizzini et al. (2011). In addition, basidiosopres were observed under the scanning electron microscope, using the following procedure: gills were attached to specimen holders by carbon tape, coated with platinum-palladium using a Hitachi MC 1000 Ion Sputter Coater and examined with a FEI Quanta 200 FE-SEM operated at 5-10 kV.

DNA Extraction, PCR, Purification and Sequencing
Genomic DNA was extracted from the dried specimens following the procedure described by Zhao et al. (2011). Polymerase chain reaction amplified sequences of the ITS and the nrLSU regions. Primers ITS1 and ITS4 (White et al., 1990) were used for the ITS region while primers LROR (Rehner and Samuels, 1994) and LR7 (Vilgalys and Hester, 1990) were used for the nrLSU region. The total volume of the PCR amplification reaction system was 50 µL containing of 10 µL of 5 × PCR buffer (Dingguo, Beijing, China), 4 µL of 200 µmol/L deoxynucleoside triphosphates, 1 µL of 200 µmol/L each primer, 5 U of Taq DNA polymerase and 10 µL of template DNA. The program parameters were set as follows, for ITS: initial denaturation at 94 • C for 4 min; repeated for 30 cycles, denaturation at 94 • C for 1 min, with annealing at 55 • C for 1 min, extension at 72 • C for 1 min, left at 72 • C for 5 min and saved at 4 • C; nrLSU: initial denaturation at 94 • C for 4 min, repeated for 30 cycles, denaturation at 94 • C for 90 s, with annealing at 55 • C for 90 s, extension at 72 • C for 90 s, left at 72 • C for 5 min and saved at 4 • C.
The products of PCR amplification were purified with the EasyPure Plasmid MiniPrep Kit (TransGen Biotech Co., Ltd., Beijing, China.), and resolved on a 1.0% agarose gel and subsequently submitted for sequencing (sequencing was completed by BGI Co., Ltd., Beijing, China).

Phylogenetic Analysis
A total of 110 sequences (ITS and nrLSU) representing 29 species were incorporated in the phylogenetic analyses, of which 26 sequences were retrieved from GenBank. Clitocybe subditopoda was used as the outgroup. Detailed specimen information appears   (Lin et al., 2017). The conservative region was selected in Gblock 2 and the vacancy gap in the data were treated as missing data (Talavera and Castresana, 2007). Saturation was tested using DAMBE 5.2 (Posada and Crandall, 1998) (model = test by Xia, 2013). MrModel Test 2.3 were used to select the fragment models (Wilgenbusch and Swofford, 2003;Nylander, 2004). The best model was used (ITS-nrLSU: GC) to construct a Maximum likelihood (ML) tree with PhyML (Guindon et al., 2009). The ML tree was evaluated by bootstrap analysis with 1000 replicates (Stamatakis, 2006). Bootstrap values greater than or equal to 60% were indicated along nodes.

Divergence-Time Analysis
Sequences incorporated in phylogenetic analyses were used to estimate the divergence time. All the sequences were aligned by MEGA v.7.0 3 . Four species of Tricholomataceae were used as the outgroup. We used divergence times of Boletales (189 ± 20Ma), Archaeomarasmius leggetti (90 Ma), Quatsinoporises cranhanaii (125 Ma) and Mycena plumbea (90 ± 30 Ma) as calibration points (Hibbett et al., 1997;Smith et al., 2004;Feng, 2012). Divergence time was estimated in BEAST v.2.5.1. The best substitution model for each partition was inferred with the program MrModeltest 2.2 (Nylander, 2004): GTR + G + I for ITS and nrLSU. The number of substitution rate categories and Gamma shape parameters were 4.0362 and 4.0343, respectively. BEAUti v.2.5.1 4 was used to construct an XML file. The Relaxed clock model was selected according to the ESS value exceeding 200. Substitution models were independently estimated for each gene partition. The Yule speciation prior set was used to estimate the divergence time and the corresponding credibility intervals were constructed using treeModel. We ran an independent Monte Carlo Markov Chains (MCMC) of 10 million generations, logging states every 10,000 generations.
The checking for convergence and mixing of Log files were completed in Tracer v1.6 5 . Tree files were summarized by the TreeAnnotator v.2.5.1., discarding 10% of states as burnin and annotating clades with ≥0.8 posterior probability, and the maximum-clade-credibility tree (MCC) was generated. The resulting files were viewed using Figtree v.1.4 6 .
Etymology: referring to the basidiospores, which are similar to Galerina.
Habitat and distribution: Scattered on grass. Known from Jilin Province in China.
Notes: The main characteristics of M. subgrammopodia are its larger basidiocarps, white lamellae, short stipe and urticiform cystidia. Melanoleuca subgrammopodia is related to members of the section Grammpodiae in the subgenus Urticocystis (Boekhout, 1988). Melanoleuca floridana Murrill is distinct from M. subgrammopodia in having a subtomentose pileus and stipe which are basal part clavate (Hesler, 2013a). Melanoleuca juliannae Rimóczi, Antonín, L. Nagy and Tomšovský is characterized by a violaceous-blue stipe and two types of cheilocystidia (the exscissa-type is most common while the brevipes-type is less frequent) (Vladimír et al., 2014). However, there is only one type of cheilocystidia in M. subgrammopodia (brevipes-type). Melanoleuca subgrammopodia is also close to M. subacris Murrill and M. subcylindrispora Murrill, M. subacris is distinct from M. subgrammopodia in having a uniform white pileus. The main characteristic of M. subcylindrispora are its subcylindrical spores, but the spores of M. subgrammopodia are elliptical to subovoid (Hesler, 2013b). Pileus 5-7 cm diameter, depressed at disc, dull olive brown, surface glabrous, dry, smooth, subviscid when moist, margin incurved, becoming plane or uplifted, paler grayish brown. Lamellae adnate, pale gray to pallid or white, moderately broad, narrowed toward the margin, 8-9 lamellae/cm in the edge of the pileus, with intercalated lamellulae, the edges even and staining brownish where bruised. Stipe 5-6 × 1-1.2 cm, surface concolorous with the pileus or paler, slightly flared at the base, longitudinally fibrous striate, glabrous, hollow when matured. Context 2-3 mm thick at pileus center, watery gray when moist, pallid when faded, odor none, taste mild. Spore print white.
Habitat and distribution: Solitary, on ground of coniferous mixed forest. Known from China and America.
Notes: M. angelesiana is placed in clade A and belonging to subgenus Urticocystis. The Chinese collection shares very similar morphological features and DNA sequence (ITS and nrLSU). Differing from holotype (Smith, 1944), the material from Heilongjiang produces a dry pileus which is not umbonate.
Habitat and distribution: Scattered on sandy soil. Known from China and Korea.
Notes: M. griseobrunnea is placed in clade A and belonging to subgenus Urticocystis. The species is related to M. porphyropoda, but M. griseobrunnea is distinguished by having urticiform cystidia (Yu et al., 2014). Differing from holotype (Antonín et al., 2017), the edges of lamellae for Chinese collections are jagged, and the stipes are slightly whitish pruinose.
Habitat and distribution: Scattered on grass. Known from China and Italy.  Notes: M. microcephala is included in clade A and belonging to subgenus Urticocystis. Differing from interpretations of Fontenla et al. (2013), the stipe of the material from Gansu is slightly longer, and the color of stipe is darker.
Notes: M. pseudopaedida is included in clade A and belonging to subgenus Urticocystis. Differing from holotype (Bon, 1990), the lamellae of the Chinese collections are whitish pruinose and the color is paler.

Phylogenetic Analysis
For the ITS-nrLSU ML analysis, 110 sequences relating to 29 species were added. The ML tree represented as Figure 12 shows detailed results with high bootstrapping values. Phylogenetic analysis produced two Clades: A and B. Clade A was formed in I node, with the bootstrapping value of 81%, while Clade B was formed in II node, having a bootstrapping value of 100%. Melanoleuca galerina and M. subgrammopodia were independently separated in III node, with a bootstrapping value of 87%. Both M. galerina and M. subgrammopodia were included in Clade A.

DISCUSSION
In this study, two new species from northeastern China were described in detail. Based on Boekhout (1988) M. galerina belongs to the subgenus Melanoleuca and M. subgrammopodia belongs to the section Grammpodiae in the subgenus Urticocystis. However, using phylogenetic analyses, the species in our investigations are divided into two Clades. The species in clade A are with urticiform cystidia or without cystidia, and in clade B have macrocystidia. This result was corroborated by Vizzini et al. (2011) whereby the authors divided the genus Melanoleuca into two subgenera: the subgenus Urticocystis mainly included the species with urticiform cystidia or without cystidia, while the subgenus Melanoleuca mainly included the species with macrocystidiate. Therefore, in this study, our phylogenetic analyses of Melanoleuca species using combined ITS and nrLSU sequences suggests that M. galerina and M. subgrammopodia are new species belonging to the subgenus Urticocystis.
In our investigations, the results of divergence time were in line with phylogenetic analysis, and supports that M. galerina and M. subgrammopodia are new species. In addition, results also indicate that divergence time of a lineage could be used as a criterion for ranking taxa (Zhao et al., 2016;Chen et al., 2017). But, the selection of proper calibration points provided by fossils is an important aspect. Moreover, reliable calibration points can provide evidence for divergence time (Feng, 2012).

DATA AVAILABILITY STATEMENT
The datasets generated can be found in NCBI, accession numbers can be found in Table 1.