Clitopilus lampangensis (Agaricales, Entolomataceae), a new species from northern Thailand

Abstract A new species of agaricomycetes, Clitopilus lampangensis, is described based on collections from northern Thailand. This species was distinguished from previously described Clitopilus species by its pale yellow to grayish yellow pileus with the presence of wider caulocystidia. Molecular phylogenetic analyses, based on the data of the internal transcribed spacers (ITS) and the large subunit (LSU) of the nuclear ribosomal DNA, and the second largest subunit of RNA polymerase II (rbp2) genes, also support the finding that C. lampangensis is distinct from other species within the genus Clitopilus. A full description, color photographs, illustrations and a phylogenetic tree showing the position of C. lampangensis are provided.


Introduction
2008; Hartley et al. 2009;Crous et al. 2012;Raj and Manimohan 2018). Clitopilus is characterized by basidiocarps that are clitocyboid, omphalinoid or pleurotoid, mostly whitish or occasionally grayish or brownish in color, with pink or pinkish brown spore prints, ellipsoid basidiospores with longitudinal ridges that appear angular in a polar view, and hyphae lack clamp connections (Singer 1986;Noordeloos 1988). There are 30 species of Clitopilus worldwide (Kirk et al. 2008), although there are 201 species names recorded in the Index Fungorum (http://www.indexfungorum.org/Names/ Names.asp). The taxa list in the Index Fungorum includes synonyms and misidentifications, as well as some species that are not well documented. Formerly, the genus Clitopilus included Rhodocybe (Moncalvo et al. 2002;Co-David et al. 2009;Vizzini et al. 2011a). However, molecular phylogenetic analyses have provided powerful tools for the identification of Clitopilus, leading to the separation of Clitopilus from Rhodocybe as well as the related genera (Clitocella and Clitopilopsis) (Cooper 2014;Kluting et al. 2014;Raj and Manimohan 2018).
Only six species, Clitopilus apalus (Berk. & Br.) Petch, C. crispus Pat. C. doimaesalongensis Jatuwong, Karun. & K.D. Hyde, C. chalybescens T.J. Baroni & Desjardin, C. peri (Berk. & Br.) Petch and C. prunulus, have been reported in Thailand (Baroni et al. 2001;Chandrasrikul et al. 2011;Kluting et al. 2014;Jatuwong et al. 2017). During an investigation of macrofungi in northern Thailand, we found a population of Clitopilus which we describe here as a new species based on the morphological and molecular characteristics. To confirm its taxonomic status, the phylogenetic relationship of the new species was determined by the ITS and LSU of the rDNA, and the rbp2 genes.

Sample collection
Basidiocarps were collected in Mae Moh District, Lampang Province, northern Thailand in 2018. Basidiocarps were wrapped in aluminum foil and kept in plastic specimen boxes to be transported to the laboratory. Notes on the macromorphological features and photographs were obtained within 24 h of collection. The specimens were dried at 40-45 °C and deposited at the Herbarium of the Sustainable Development of Biological Resources Laboratory, Faculty of Science, Chiang Mai University (SDBR-CMU), and BIOTEC Bangkok Herbarium (BBH), Pathumthani, Thailand.

Morphological studies
Macromorphological data were recorded from fresh specimens. The recording of color names and codes followed Kornerup and Wanscher (1978). Micromorphological data were recorded from dry specimens rehydrated in 95% ethanol followed by distilled water, 3% KOH or Melzer's reagent. Anatomical features were based on at least 50 measurements of each structure as seen under a light microscope (Olympus CX51, Japan). For spore statistics, Q is the ratio of spore length divided by spore width and Q is the average Q of all specimens ± standard deviation.

Molecular phylogenetic studies
Genomic DNA of dry specimens (1-10 mg) was extracted using a Genomic DNA Extraction Mini-Kit (FAVORGEN, Taiwan). The ITS region of DNA was amplified by polymerase chain reactions (PCR) using ITS4 and ITS5 primers (White et al. 1990), the LSU of rDNA gene were amplified with LROR and LRO5 primers (Vilgalys and Hester 1990), and rbp2 gene was amplified with the bRBP2-6F and bRBP2-7.1R primers (Matheny 2005). The amplification program for these three domains was performed in separated PCR reaction and consisted of an initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 30 s, annealing at 52 °C for 30 s (ITS), 52 °C for 45 s (LSU), and 54 °C for 1 min (rpb2), and extension at 72 °C for 1 min on a peqSTAR thermal cycler (PEQLAB Ltd., UK). PCR products were checked on 1 % agarose gels stained with ethidium bromide under UV light. PCR products were purified using a PCR clean up Gel Extraction NucleoSpin Gel and PCR Clean-up Kit (Macherey-Nagel, Germany) following the manufacturer's protocol. The purified PCR products were directly sequenced. Sequencing reactions were performed and the sequences were automatically determined in the genetic analyzer at 1 st Base company (Kembangan, Malaysia) using the PCR primers mentioned above. Sequences were used to query GenBank via BLAST (http://blast.ddbj.nig. ac.jp/top-e.html).
For phylogenetic analyses, the sequences from this study, previous studies and the GenBank database were used and provided in Table 1. The multiple sequence alignment was carried out using MUSCLE (Edgar 2004), and the combined ITS and LSU alignment, and rpb2 alignment were deposited in TreeBASE under the study ID 24373 and 24374, respectively. Phylogenetic trees were constructed using maximum likelihood (ML) and Bayesian inference (BI) algorithms, implemented by RAxML v7.0.3 (Stamatakis 2006) and MrBayes v3.2.6 (Ronquist et al. 2012), respectively. Rhodocybe griseoaurantia and R. pallidogrisea were used as outgroup. The best-fit substitution model for BI and ML analyses were estimated by jModeltest 2.1.10 (Darriba et al. 2012) using Akaike information criterion (AIC). For ML analysis, the bootstrap (BS) replicates were set as 1000 and used to test phylogeny (Felsenstein 1985). Clades with bootstrap values (BS) of ≥ 70% were considered significantly supported (Hillis and Bull 1993). For the BI analysis, the Markov chains were run for one million generations, with six chains and random starting trees. The chains were sampled every 100 generations. Among these, the first 2,000 trees were discarded as burn-in, while the postburn-in trees were used to construct the 50% majority-rule consensus phylogram with calculated Bayesian posterior probabilities. Bayesian posterior probabilities (PP) ≥ 0.95 were considered significant support (Alfaro et al. 2003).

Phylogenetic analyses
The topology of each single-gene of ITS and LSU, and the combined ITS and LSU phylograms were found to be similar. However, differences were observed in the topology of the rbp2 gene. Therefore, we present only the combined ITS and LSU gene phylogram (Fig. 1), and the single rbp2 gene phylogram (Fig. 2). The combined ITS and LSU sequence dataset consisted of 34 taxa and were comprised of 1774 characters including gaps (ITS: 1-779, LSU: 780-1774). The sequence dataset of rbp2 consisted of 27 taxa and the aligned dataset was comprised of 620 characters that included gaps. The GTR model with gamma rate heterogeneity and invariant sites (GTR+G+I) was the best-fit model used for both ML and BI analyses that were selected by AIC. The average standard deviation of the split frequencies fell to 0.011364 and 0.009837 in the BI analysis of the combined ITS and LSU, and rbp2 sequences, respectively after one million generations. This was observed after the 50% majority consensus phylogram was constructed. The ML analysis of the combined ITS and LSU sequences was based on the parameters estimated under the GTR+I+G model, and the proportion of the invariable sites and the gamma shape parameters were 0.0250 and 0.9320, respectively. Additionally, the tree with log likelihood (-8211.7515) was built after 1000 bootstrapping replications. In the ML analysis of the rbp2 sequence that was based on the GTR+I+G model, the proportion of the invariable sites and the gamma shape parameters were 0.5400 and 1.7960, respectively, while the tree with log likelihood (-3640.1616) was built after 1000 bootstrapping replications. Both the combined ITS and LSU, and the rbp2 phylograms indicated that the sequences were of a new species, C. lampangensis, that had formed a monophyletic clade with high BS (100 %) and PP (1.0) support (Figs 1, 2). A combined ITS and LSU phylogram revealed that the new species was a sister taxon to C. chalybescens. In addition, the rbp2 phylogram indicated that the new species was a sister taxon to C. chalybescens and C. peri. Diagnosis. Distinguished from other Clitopilus species by its pale yellow to grayish yellow pileus with the presence of caulocystidia, and from C. chalybescens by its wider caulocystidia, longer basidiospores, and lack of grayish blue color change on the pileus and stipe when bruised.
The phylogenetic analyses of the combined ITS and LSU, and rpb2 sequences confirmed that C. lampangensis formed a monophyletic clade which clearly separated it from the other Clitopilus species. Clitopilus lampangensis forms a sister taxon to C. chalybescens and C. peri. Clitopilus peri differs from C. lampangensis by its smaller white basidiocarps (8-22 mm in diameter) and the absence of caulocystidia (Pegler 1986).
Additionally, the different morphological characteristics that exist between C. lampangensis and C. chalybescens have been mentioned above.
Therefore, a combination of the morphological characteristics and the molecular analyses strongly support recognition of a new fungus species. This discovery is considered important in terms of stimulating a deeper investigation of macrofungi in Thailand, and will help researchers to better understand the distribution and ecology of Clitopilus.
Key to Clitopilus species known from Thailand