Three new species of Phanerochaete (Polyporales, Basidiomycota)

Abstract Phanerochaetecanobrunnea, P.cystidiata and P.fusca are presented as new species, supported by morphological studies and two sets of phylogenetic analyses. The 5.8S+nuc 28S+rpb1 dataset shows the generic placement of the three species within the phlebioid clade of Polyporales. The ITS+nuc 28S dataset displays relationships for the new taxa within Phanerochaete s.s. Phanerochaetecanobrunnea grew on angiosperm branches in subtropical Taiwan and is characterised by greyish brown hymenial surface, brown generative hyphae and skeletal hyphae and absence of cystidia. Phanerochaetecystidiata grew on angiosperm branches above 1000 m in montane Taiwan and SW Yunnan Province of China and is characterised by cream to yellowish hymenial surface and more or less encrusted leptocystidia. Phanerochaetefusca grew on angiosperm branches at 1700 m in Hubei Province of China and is characterised by dark brown hymenial surface, leptocystidia, brown subicular hyphae and colourless to brownish basidiospores.

The field survey of the corticioid fungi from Taiwan and mainland China conducted in 2014, 2015 and 2017, have revealed three new species of Phanerochaete s.s. presented herein, based on morphological and phylogenetic evidence.

Morphological studies
Voucher specimens are deposited at the herbarium of National Museum of Natural Science of ROC (TNM). We used three mounting media for microscopic studies: 5% potassium hydroxide (KOH) with 1% phloxine was used for observation and measurements; Melzer's reagent (IKI) was utilised to determine amyloidity and dextrinoidity and Cotton blue (CB) was utilised to check cyanophily. A standard method of measurement for microscopic characters follows Wu (1990). Below abbreviations were used when presenting statistic measurements of basidiospores: L = mean basidiospore length, W = mean basidiospore width, Q = variation in L/W ratio, n = number of measured spores. The terminology of microscopic characters followed Wu (1990).
RNA polymerase II largest subunit (rpb1) was amplified using the primer pair RPB1-Af/RPB1-Cr (Stiller and Hall 1997;Matheny et al. 2002). Both RPB1-2.1f and RPB1-2.2f were used as alternative primers to pair with RPB1-Cr (Frøslev et al. 2005). The PCR protocols for ITS, nuc 28S and rpb1 followed Wu et al. (2018). PCR products were directly purified and sequenced by the MB Mission Biotech Company (Taipei, Taiwan). We determined the identity and accuracy of newly obtained sequences by comparing them to sequences in GenBank and assembled them using BioEdit (Hall 1999). Newly obtained sequences were then submitted to GenBank (https://www. ncbi.nlm.nih.gov/genbank/; Table 1).

Phylogenetic analyses
We included two datasets for phylogenetic analyses. The 5.8S+nuc 28S+rpb1 was compiled for inferring generic classification of target species within the phlebioid clade of Polyporales. The ITS+nuc 28S was compiled for getting better resolutions on species level within Phanerochaete s.s. The selection of strains and species consulted Wu et al. (2010), Floudas and Hibbett (2015), Volobuev et al. (2015), Liu and He (2016), Miettinen et al. (2016) and Wu et al. (2018). MAFFT v. 7 was used to align sequences of each marker with default settings (Katoh and Standley 2013). The resulting alignments were manually adjusted in MEGA 7 (Kumar et al. 2016). Hyphoderma litschaueri (Burt) J. Erikss. & Å. Strid and H. mutatum (Peck) Donk, were chosen as the outgroup in the 3-marker dataset. Phlebiopsis gigantea (Fr.) Jülich was chosen as the outgroup in the 2-marker dataset. Final datasets were deposited at TreeBASE (submission ID 23083).
For both datasets, Maximum Likelihood (ML) and Bayesian Inference (BI) analyses were performed, respectively, using RAxML BlackBox (Stamatakis et al. 2014) and MrBayes v. 3.2.6 (Ronquist et al. 2012) at the CIPRES Science Gateway (Miller et al. 2010; http://www.phylo.org/). For BI analysis, jModeltest 2.1.10 (Darriba et al. 2012) was first carried out to determine the best-fit substitution model for each marker based on Akaike Information Criterion (AIC). The GTR+I+G was used as the substitution model for the entire alignment of the 3-marker dataset, while, for the 2-marker dataset, the HKY+I+G and the GTR+I+G were used for the alignments of ITS and nuc 28S, respectively. The parameters for BI analyses were as follows: ngen = 10000000, samplefreq = 100, nchains = 4, nst = 6 for GTR, nst = 2 for HKY, rates = invgamma, burn-in = 25000. Fifty percent majority-rule consensus phylograms with posterior probability values (PP) were obtained when the average standard deviation of split frequencies was below 0.001. For ML analysis, the bestscoring tree with values of bootstrap (BS) was constructed using the GTR model with one hundred rapid bootstrap inferences. Gaps were regarded as missing data. Phylograms were visualised and edited by TreeGraph 2 (Stöver and Müller 2010) and Adobe Illustrator (Adobe Systems, Inc).