﻿Additional four species of Tatraea (Leotiomycetes, Helotiales) in Yunnan Province, China

﻿Abstract During the investigations of discomycetes in Yunnan, China, five species of Tatraea were discovered on decayed, decorticated oak trees or unidentified wood. All species have typical disc-like, large fruiting bodies with grey, brown or greyish-green colors. The ITS sequence analysis showed that they belong to Tatraea (Helotiaceae, Helotiales) and the LSU and ITS combination revealed a different topology within the genus. Four species, T.clepsydriformis, T.griseoturcoisina, T.yunnanensis and T.yuxiensis were established as new species, and T.aseptata was collected and described on oak woods. The pairwise homoplasy index (PHI) test results indicated that there is no significant genetic recombination (Φw = 1.0) between all related species pairs. All the species described here are supported by descriptions, illustrations and multi-gene analyses.

To date, members of Tatraea have been only found as saprobes on the rotting and permanently moist, decorticated trunks of beech wood (Fagus sylvatica), rarely occurring on Fraxinus excelsior or Betula and have been reported in Austria, China, Croatia, Denmark, France, Germany, Great Britain, Italy, Slovakia, Spain, Sweden and Switzerland (Svrček 1993;Baral et al. 1999Baral et al. , 2013aBaral et al. , 2013b;;Holec et al. 2015).The significance of fungi in forest ecosystems was reviewed by Niego et al. (2023a), highlighting their diverse functional contributions.This emphasizes the critical need to integrate fungal contributions into ecological conservation policies.Additionally, fungi hold significant value, encompassing not only the economic worth of wild and cultivated mushrooms but also the augmented value derived from fungal products and their involvement in various production processes.Furthermore, fungal involvement in ecosystem processes like carbon sequestration and recreational foraging also increases their traded value.In their study, Niego et al. (2023b) provided estimates to support more effective ecological conservation policies for fungal resources, highlighting the importance of studying and conserving these organisms.Decay fungi are able to produce enzymes that degrade components of wood, such as lignin, cellulose and xylans (Bucher et al. 2004) and are known as lignicolous fungi.Different lignicolous fungi prefer dead wood at different stages of decaying, for example, Tatraea mainly grows and decomposes the intermediate and late stages of wood decay (DS4) (Svrček 1993;Baral et al. 1999Baral et al. , 2013a;;Heilmann-Clausen 2001;Holec et al. 2015;Dvořák 2017;Ujházy et al. 2018;Kunca et al. 2022).Due to the high density of managed forests, their low understory vegetation diversity compared to that of primary forests as well as the lack of late-stage decayed wood, members of Tatraea were rarely discovered in managed forests.Thus T. dumbirensis was considered an indicator of the primary forest and forest continuity, but also rarely collected in beech-dominated managed forests (Dvořák 2017; Ujházy et al. 2018).In China, only one species (Tatraea aseptata H.L. Su & Q.Zhao) was discovered in the protected primary forests, and other species were mainly found in the Center of Europe (Baral et al. 1999).Tatraea species are mostly found in old, natural primary montane forests (Baral et al. 1999).These species may have specialized adaptations to the undisturbed, virgin primary forests, contributing to overall biodiversity.Hence, they might not thrive or even survive in disturbed or secondary forests.These findings stress the importance of accurate management of primary forests to conserve their fungal diversity as well as the fungal gene pool (Baral et al. 1999).Furthermore, the rarity of these Tatraea species also highlights the importance of conducting studies on rare Leotiomycetes fungi to conserve them before they become extinct.
We have been conducting comprehensive studies on discomycetes, encompassing investigations into taxonomy, species diversity and evolutionary research, among other aspects (Ekanayaka et al. 2017(Ekanayaka et al. , 2018(Ekanayaka et al. , 2019;;Lestari et al. 2022Lestari et al. , 2023;;Phutthacharoen et al. 2022).In this study, the authors aim to investigate the diversity of discomycetes in Yunnan Province, China.During our exploration, we discovered and collected the rare Tatraea species.In this study, we identified four new species of Tatraea on decayed and decorticated wood with detailed morphological descriptions and illustrations as recommended by Chethana et al. (2021).In previous studies, the classification of Tatraea mainly relied on morphological evidence, and only ITS sequences were available for phylogenetic analysis.Here, we provide additional gene regions and complete morphologies for future taxonomic and evolutionary research.

Sample collection and morphological studies
Specimens were collected from decayed wood in Yunnan Province, China, during field investigations conducted from June, 2021 to October, 2022.All samples were obtained from highly humid natural broadleaf forests and protected areas rarely accessed by humans.During the collection period, the temperature of the collection site was basically in the range of 17 °C to 27 °C, and the temperature of Jingdong County, Puer City was 14 °C to 16 °C due to the influence of high altitude.The fruiting bodies were found on the surface of extremely wet decaying wood following rainfall events.The specimens with their substrates were gently wrapped in a single layer of tissue, rotated and pinched ends tightly with a hollow center to prevent squeezing the specimen.The specimens dried naturally in air, re-wrapped in a hard-paper boxes containing a small amount of silica gel and rehydrated before being observed in the laboratory.
Fresh apothecia were photographed in the field by a Canon EOS M100 camera (Canon Co. LTD, Japan).The dried and partially fresh apothecia were captured using a Canon EOS 70D(W) digital camera attached to a C-PSN stereomicroscope (Nikon Corporation Tokyo, Japan).The dried apothecia were sectioned by hand using razor blades and photographed by a charge-coupled device SC 2000C attached to a Nikon ECLIPSE Ni-U compound microscope (Nikon Corporation Tokyo, Japan).Vertical sections were used to observe the excipulum and hymenium.Asci, ascospores and paraphyses were observed by mounting squashed mature apothecia in water.Melzer's reagent checked the blue iodine reaction of the ascus apex.
All measurements were carried out using Tarosoft (R) Image Framework program (IFW) and modified in Adobe Photoshop 2020 (Adobe system, USA).Q value indicates the length to width ratio of ascospores, n indicates the number of measured structures, and Q m indicates the average of Q value.The size of apothecia was defined as large (greater than 3.5 mm wide), moderate (greater than 2.5 mm and less than 3.5 mm wide) and small (less than 2.5 mm wide) based on mean and extreme values.The length of stipes was defined as long (longer than 1.1 mm), moderate (greater than 0.4 mm and less than 1.1 mm) and short (shorter than 0.4 mm).The colors of apothecia were determined following Kornerup and Wanscher (1967).The dried specimens were deposited at the Herbarium of Cryptogams, Kunming Institute of Botany Academia Sinica (KUN-HKAS).Facesoffungi numbers were obtained as in Jayasiri et al. (2015), and Index Fungorum numbers were obtained as in Index Fungorum (2023).Furthermore, details of all the species described in this study were uploaded to the Discomycetes website (https://discomycetes.org/,Lestari et al. 2023).

DNA extraction, PCR amplification, and sequencing
Two to three mature fruit bodies were carefully selected and thoroughly cleaned using sterilized water and 75% alcoholic solution.Subsequently, several layers of epidermal cells were meticulously removed using sterilized surgical blades.Following this step, approximately 1 mm 3 of tissue was meticulously collected from both the receptacles and stipes using new sterile surgical blades.The collected tissue was then transferred into a sterile 1.5 mL centrifuge tube.Total genomic DNA was extracted using the TriliefTM Plant Genomic DNA Kit (Tsingke Biological Technology Co., LTD, Beijing, China).The total reaction volume for the Polymerase Chain Reaction (PCR) was 25 μl, containing 12.5 μl of 2 × Power Taq PCR MasterMix, 7.5 μl of sterile deionized water, 1 μl of each primer (100 μM stock), and 3 μl of DNA template.The amplifications were performed in a TC-type gene amplifier (LifeE-CO) (Hangzhou Bori Technology Co., LTD, Hangzhou City, Zhejiang Province, China).The primers used in this study are shown in Table 1.The conditions of PCR for each gene are as follows: for the ITS, LSU, mtSSU and RPB1, initial denaturation at 94 °C for 3 min, 35 cycles of denaturation at 94 °C for 30 s, 40 s of annealing at 53 °C, 1 min elongation at 72 °C, followed by a final extension for 10 min at 72 °C; for the RPB2 initial denaturation at 94 °C for 5 min, 35 cycles of denaturation at 94 °C for 1 min, 1 min of annealing at 56 °C for RPB2, 1 min elongation at 72 °C, followed by a final extension for 10 min at 72 °C.The PCR products were verified by 1% agarose gel electrophoresis followed by staining with TS-GelRed Ver.2 10,000 × in Water (Tsingke Biological Technology Co., LTD, Beijing, China).Products were sequenced at Tsingke biological technology Co., LTD, Beijing, China.

Sequence assembly and alignment
Sequences were assembled in ContigExpress (Invitrogen, USA), and then checked and edited in BioEdit 7.2.5.0 (Hall 1999).The homologous sequences were selected based on the results of the BLASTn search performed against the GenBank database available at NCBI.All new and related sequences used in this study were derived from GenBank and used for phylogenetic analyses.Two species in Chlorociboria (Chlorociboriaceae, Helotiales) were selected as the outgroup taxa.The datasets were aligned in MAFFT v. 7 (Katoh et al. 2019) with G-INS-i as the iterative refinement and default parameters were applied except for the gap penalty, which was changed to 1.00, and improved manually in BioEdit v. 7.2.5.0.Then, datasets were trimmed in TrimAl v.1.3using the gappyout option (Capella-Gutiérrez et al. 2009).The multiple loci association matrixes were concatenated to a combined dataset in SequenceMatrix 1.7.8.(Vaidya et al. 2011).Due to the lack of sequence data for protein genes, the phylogenetic tree was constructed using the ITS gene and the combined LSU and ITS gene regions (Figs 1, 2).The combined ITS, LSU, mtSSU, RPB1 and RPB2 dataset was used to analyze the recombination level within phylogenetically and closely related species (Fig. 3).The ALTER (Alignment Transformation EnviRonment) online tool was used to convert from ".fasta" to ".nexus" format.The newly generated sequences in this study were deposited in GenBank (Table 2), and the combined alignment was deposited at the TreeBASE (submission ID: 30884).

Phylogenetic analyses
Maximum likelihood (ML) analysis was performed in RAxML-HPC2 on XSEDE (8.2.12) on the CIPRES Science Gateway platform (http://www.phylo.org/por-tal2)using the GTR model with 1,000 bootstrap replications.Bayesian inference (BI) analysis was performed using MrBayes v. 3.1.2.The Markov Chain Monte Carlo sampling (MCMC) was used to evaluate the posterior probabilities (PP).The general time-reversible model with a discrete gamma distribution coupled with a proportion of an invariant (GTR+I+G) was selected for nLSU and ITS as the best model using MrModeltest v.2.3 (Nylander et al. 2004).Four simultaneous Markov Chains were run for 2,000,000 generations, with trees sampling at every 100 th generation.The 25% of the trees representing the burnin phase were discarded, and the remaining trees were used to calculate the posterior probability.The finalized phylogenetic tree was visualized in Figtree v.1.4.0 (Rambaut 2012) and edited in Adobe Illustrator 2020 and Adobe Photoshop 2020 (https://www.adobe.com/).Splitstree4 4.17.1 was used to determine the recombination level between phylogenetically and closely related but ambiguous species based on the PHI (pairwise homoplasy index) value (Taylor et al. 2000;Silvestro and Michalak 2012).The relationships between the two species were shown in splits graphs with the Log-Det transformation option.The relationship between T. macrospora and T. yunnanensis was visualized by constructing a split graph (Fig. 3A) from ITS.The relationship between the other two pairs (T.yuxiensis and T. aseptate, and T. griseoturcoisina and T. clepsydriformis) were visualized by constructing splits graphs, Fig. 3B and Fig. 3C, respectively, from 5-locus concatenated dataset.A pairwise homoplasy index below a 0.05 threshold (Φw < 0.05) indicates the presence of significant recombination between the two species (Chethana et al. 2017).

Phylogenetic analyses
The dataset for the phylogenetic analysis based on the ITS gene consists of 69 taxa, represented by 81 isolates, including two outgroup taxa, Chlorociboria aeruginosa (TNS:F13596) and Chlorociboria aeruginascens (TNS:F36241).The dataset contains 550 total characters with gaps.The combined alignment contains 239 constant characters, 54 variable and parsimony uninformative characters and 254 parsimony-informative characters.The RAxML analysis of the ITS gene dataset yielded the best-scoring tree with a final likelihood value of -8336.600892(Fig. 1).The maximum likelihood matrix comprises 366 distinct alignment patterns with 8.85% undetermined characters or gaps.Estimated base frequencies are as follows: A = 0.219317, C = 0.265186, G = 0.257196, T = 0.258302; substitution rates AC = 1.749614,AG = 2.352229, AT = 1.537478,CG = 0.647334, CT = 5.060114, GT = 1.000000; gamma distribution shape parameter α = 0.360268.The LSU and ITS concatenated dataset consists of 9 taxa, represented by 21 isolates, including two outgroup taxa, Hymenoscyphus aurantiacus (HMAS 264143) and Phaeohelotium epiphyllum (TNS:F40042).The concatenated dataset contains 1744 aligned nucleotide sites, including 1256 bp for the LSU region and 488 bp for the ITS region with gaps.The combined alignment contains 1471 constant characters, 89 variable and parsimony uninformative characters and 184 parsimony-informative characters.The RAxML analysis of the combined dataset (LSU and ITS) yielded the best-scoring tree with a final likelihood value of -4243.290550(Fig. 2).The dataset comprises 259 distinct alignment patterns with 31.66%undetermined characters or gaps.Estimated base frequencies are as follows: A = 0.227902, C = 0.244223, G = 0.295968, T = 0.231906; substitution rates AC = 0.961500, AG = 2.058057, AT = 0.608426, CG = 0.460988, CT = 7.662978, GT = 1.000000; gamma distribution shape parameter α = 0.183205.Species in Hymenoscyphus showed different topologies in ML and BI analyses, but the support values for the nodes are less.Despite the different taxon sampling, the topological structure of the phylogenetic tree shown in Fig. 1 is similar to that of Johnston et al. (2019).In the ML and Bayes analyses, Tatraea formed a monophyletic clade within Helotiaceae with 67% ML bootstrap support and 0.98 Bayesian probability in the ITS phylogeny (Fig. 1).Some nodes in Clade II and Clade III have low support values (Fig. 1).In Clade I, our collections of Tatraea aseptata clustered with the type species and formed a sub-clade sister to T. yuxiensis with 98% ML support and 1.00 Bayesian probability support.However, this clade comprising T. aseptata and T. yuxiensis separated from T. dumbirensis with 74% ML bootstrap support and 0.57 Bayesian probability.Collections of T. clepsydriformis and T. griseoturcoisina formed two individual clades in Fig. 1 with 50% ML bootstrap support and 0.80 Bayesian probability, and 52% ML bootstrap support and 0.92 Bayesian probability, respectively.However, these two species clustered as a separate sub-clade with 85%  ML support and 0.96 Bayesian probability support in the ITS and LSU combined phylogeny (Fig. 2).Tatraea yunnanensis clustered with T. macrospora with 98% ML bootstrap support and 0.99 Bayesian probability support in the LSU and ITS phylogeny (Fig. 2).A pairwise homoplasy index below 0.05 typically indicated the presence of significant recombination among the groups.In our analysis, the pairwise homoplasy index (PHI or Φw) for three pairs of species (T.macrospora vs. T. yunnanensis, T. yuxiensis vs. T. aseptata and T. griseoturcoisina vs. T. clepsydriformis) were 1.0, 1.0 and 0.4185, respectively.These results indicated no significant recombination among these pairs.Type material.Holotype.HKAS 124623.
Notes.The distinctive characteristics of T. clepsydriformis are moderate-sized apothecia (2.5 mm wide when fresh), with fresh brown receptacles and stipes, light brown to hazel brown at dry condition, stipes concolorous to receptacles, pale yellow, proliferous cells of ectal excipulum not observed, aporhynchous asci and small, ellipsoidal ascospores without septa.
Notes.The distinctive characteristics of Tatraea griseoturcoisina are greyish-green apothecia, with yellowish-white to snow white discs when dry, narrow hyphae of medullary excipulum, short aporhynchous asci and slightly narrow ellipsoidal ascospores without septa.
Phylogenetically, T. griseoturcoisina grouped with T. clepsydriformis with 85% ML bootstrap support and 0.96 Bayesian probability in the combined LSU and ITS phylogeny (Fig. 2).A pairwise homoplasy index (PHI) test was conducted using a five-gene dataset (ITS, LSU, mtSSU, RPB1 and RPB2) to assess the recombination level between clades of T. griseoturcoisina and T. clepsydriformis.The results revealed that there were no significant recombination events observed between these two groups (Фw > 0.05), indicating that they are genetically isolated and thus supporting them as distinct species (Fig. 3).Tatraea griseoturcoisina is distinct from all other species based on its unique macro-characteristics of greyish-green apothecia, dried discs and receptacles.Micro-characteristics of T. griseoturcoisina resemble T. clepsydriformis by having narrow hyphae of medullary excipulum, short asci and smaller ellipsoidal ascospores, but it is distinct from T. clepsydriformis by having a thinner medullary excipulum (164-308 μm vs.335-535 μm), longer (17.1 × 5.4 μm vs. 15.2 × 5.7 μm) and curved ascospores.Tatraea griseoturcoisina can be distinguished from the other five species (T.aseptata, T. dumbirensis, T. macrospora, T. yunnanensis and T. yuxiensis) based on its short asci (109-122 μm) and ascospores (14.6-20.4μm) (see Suppl.material 1).Tatraea yunnanensis C.J.Y.Li & Q.Zhao, sp.nov.Index Fungorum: IF901180 Facesoffungi Number: Fo15192 Fig. 7 Etymology.The specific epithet refers to the locality from where the type species was collected.
Notes.The distinctive characteristics of T. yunnanensis are large (4.8 mm wide), brown apothecia with pastel green to light green discs and short stipes, thick medullary excipulum comprising 2-10 layers of inverted proliferous cells, arched or irregular shaped, pleurorhynchous asci with J+ pores, filiform, 3-septate paraphyses and elongated fusiform ascospores without septa.
Phylogenetically, our collections clustered sister to T. macrospora with 98% ML bootstrap support and 1.0 Bayesian probability in the combined LSU and ITS phylogeny (Fig. 2).It was shown that the two species do not have any genetic recombination (Φw = 1.0) based on the pairwise homoplasy index (PHI) value (Fig. 3).Tatraea yunnanensis resembles T. macrospora in having cupulate apothecia, yellowish-white to orange-white discs when fresh and large ascospores.In contrast, our species differ from T. macrospora by having longer ascospores (32.5-42.4μm vs. 22-40 μm) with 3-8-septate ascospores (Baral et al. 1999).For T. macrospora, the morphological descriptions provided in previous studies are incomplete and lack details on the apothecial size and color, stalk and spores.
Tatraea yuxiensis C.J.Y.Li & Q.Zhao, sp.nov.Index Fungorum: IF901187 Facesoffungi Number: Fo15193 Fig. 8 Etymology.The specific epithet refers to the locality from where the type species was collected.
Notes.The distinctive characteristics of T. yuxiensis are orange-grey to brownish-grey discs when fresh, dark brownish-black or dark ochraceous-brown when dry, short and regular cylindrical stipe, 3-4 layers of proliferous minimal cells, pleurorhynchous asci and elongated ellipsoidal and laterally asymmetrical ascospores.
Phylogenetically, T. yuxiensis clustered sister to T. aseptata with 100% ML bootstrap support and 1.0 Bayesian probability in the combined LSU and ITS phylogeny (Fig. 2).The pairwise homoplasy index (PHI) indicated no significant genetic recombination (Φw = 1.0) between T. yuxiensis and T. aseptata and confirmed that they are different species (Fig. 3).

Discussion
Tatraea was initially collected a 100 years ago from the Nizke Tatra Mountains in Europe and was subsequently found in several countries (Velenovský 1934;Baral et al. 1999).In Britain and Croatia, T. dumbirensis is listed on the Red List as a threatened species (Ujházy et al. 2018).Tatraea macrospora appeared in some countries, but no official records were found.Therefore, the accuracy of species identification could not be confirmed.Since the study in 1999, there have been no new species reports in Tatraea from other continents.All collections in this study were collected from Yunnan, China, of which most were collected from protected natural forests and from areas comprising mainly oak trees.Some species are found in plantations that have been protected and nursed for many years, however, the host is too decayed to identify.In most cases, the decayed oak wood is still the main nutrient provider in forests.After the addition of our collections (T.clepsydriformis and T. griseoturcoisina), the description of Tatraea should be extended from long asci and large ascospores to include slightly shorter asci and smaller spores, as well as the initial greyish turquoise color of the apothecia.
In the past, the type species, T. dumbirensis was incorrectly recognized to be a member of the Leotiaceae or Sclerotiniaceae (Velenovský 1934;Baral et al. 1999).The exclusion from the Sclerotiniaceae was due to the absence of darkening and sclerotia formation in the cultures (Baral et al. 1999).In the present taxonomic study, Tatraea was included in Helotiaceae, and we also agreed on this treatment based on the ITS analysis in our study (Vasilyeva 2010).The genus-level placements of each species in Tatraea changed after adding data from other genera into the analysis.Previously, T. clepsydriformis and T. griseoturcoisina clustered into separate clades, later clustered into a single main clade as sister sub-clades after adding more taxa, and their micro-morphological characteristics were more similar.In the phylogenetic analyses, the taxonomic status of species is provisional due to the lack of genetic information for the type species.To assess the significant recombination levels of related species, we performed five gene analyses individually and for the combined dataset, both of which provided evidence for them being different species.The dilemma for conducting research is the paucity of available molecular information for the known species.More informative loci were provided in this study, including mitochondrial genes and protein genes, hence, future taxonomy, phylogeny research and evolutionary studies in Helotiaceae can be benefited from this study.Additionally, some species with similar morphological characteristics to Tatraea, such as Ciboria fusispora, are currently unable to transfer due to a lack of evidence and fresh samples.Therefore, more research with more fresh specimens is essential to facilitate the classification of these species.

Figure 1 .
Figure 1.Maximum likelihood tree based on the ITS sequence, showing the phylogenetic position of Tatraea.The ML bootstrap proportions (ML-BP) equal to or higher than 70% and Bayesian posterior proportions (BI-PP) equal to or higher than 0.90 are shown near the branches of the phylogenetic tree.Newly generated isolates of the current study are shown in blue and ex-types in bold.

Figure 2 .
Figure 2. Maximum likelihood tree based on a combined dataset of LSU and ITS sequences for the genus Tatraea.The ML bootstrap proportions (ML-BP) equal to or higher than 70% and Bayesian posterior proportions (BI-PP) equal to or higher than 0.90 are shown near the branches on the phylogenetic tree.Newly generated isolates of the current study are shown in blue and ex-types are shown in bold.

Table 1 .
Primers used for the PCR amplifications in this study.

Table 2 .
Detailed information and corresponding GenBank accession numbers of the taxa used in the phylogenetic analyses of this study.' † ' Denotes type species.Newly generated sequences are shown in bold.'-': indicates sequence data is not available.