Three new species and a new combination of Triblidium

Abstract Triblidiaceae (Rhytismatales) currently consists of two genera: Triblidium and Huangshania. Triblidium is the type genus and is characterised by melanized apothecia that occur scattered or in small clusters on the substratum, cleistohymenial (opening in the mesohymenial phase), inamyloid thin-walled asci and hyaline muriform ascospores. Before this study, only the type species, Triblidium caliciiforme, had DNA sequences in the NCBI GenBank. In this study, six specimens of Triblidium were collected from China and France and new ITS, mtSSU, LSU and RPB2 sequences were generated. Our molecular phylogenetic analysis and morphological study demonstrated three new species of Triblidium, which are formally described here: T. hubeiense, T. rostriforme and T. yunnanense. Additionally, our results indicated that Huangshania that was considered to be distinct from Triblidium because of its elongated, transversely-septate ascospores, is congeneric with Triblidium. Therefore, we have placed Huangshania in synonymy under Triblidium, rendering Triblidiaceae a monotypic family.


Introduction
Triblidium Rebent.: Fr. is the type genus of Triblidiaceae Rehm (Rehm 1888(Rehm −1896(Rehm , 1912, which includes presumed saprobes on the bark of Pinaceae, Ericaceae and Fagaceae (Magnes 1997). In his monograph of the family, Magnes (1997) speculated that some species may exist in an endophytic state. Species of Triblidium are well documented in Europe, but they are poorly understood in Asia and America (Magnes 1997). Magnes (1997) revised Triblidium and accepted amongst the many included species only four species and one subspecies.
A history of Triblidiaceae is given in Karakehian et al. (2019). In brief, Magnes (1997) placed Triblidiaceae in Rhytismatales and treated Triblidiales as a synonym of Rhytismatales. Recent five-locus (Prieto et al. 2019) and 15-locus (Johnston et al. 2019) phylogeny analyses found high support for Pseudographis (Triblidiaceae) within Rhytismatales. The results of a three-gene phylogenetic analysis with expanded sampling by Karakehian et al. (2019) supported Magnes classification and the authors emended Triblidiaceae to include Triblidium and Huangshania.
We conducted a morphological analysis of a specimen of T. caliciiforme Rebent.: Fr., the type species of Triblidium and additional collections of Triblidiaceae. Phylogenetic relationships were inferred based on internal transcribed spacer (ITS), nuclear large subunit ribosomal DNA (LSU), mitochondrial small subunit ribosomal DNA (mtSSU) and the second largest subunit of RNA polymerase II (RPB2) gene.

Morphological studies and isolation
A specimen of Triblidium caliciiforme was collected in France in June 2012 on Quercus sp. Other specimens were collected in China between 2006 and 2018. Mature dried ascomata were selected for morphological observation. All observations were made from dead herbarium material. Gross morphology was observed and photographed with a dissecting microscope (Nikon SMZ-1000). Standardised colour values matching the colour of the hymenium were taken from https://www.colorhexa.com/. Microscopic preparations were observed in distilled water, Lugol's solution (IKI), 5% potassium hydroxide (KOH) and lactophenol solution. Methods for morphological analysis follow Hou et al. (2009). Measurements of asci and ascospores were made in distilled water in 2019. For each structure, at least 25 measurements were recorded. Microphotographs were obtained using an Olympus BX51 compound microscope. Specimens are deposited in the Herbarium of the College of Life Science, Capital Normal University, Beijing, China (BJTC). Fresh specimens were used to obtain cultures directly from single ascoma, after washing and surface sterilisation, as follows: 75% ethanol for 10 s, 10% sodium hypochlorite for 3 min, washing in sterile water three times. The single ascoma was dried in sterilised tissue paper, placed on potato dextrose agar (PDA) with 50 mg/l chloramphenicol and incubated at room temperature (25 °C ± 3 °C). We were unable to obtain cultures from ascomata after a month.

DNA extraction and PCR amplification
Genomic DNA was extracted from ascomata using NuClean Plant Genomic DNA Kit (CWBIO, China), following the manufacturer's instructions and stored at -20 °C. Se-quences of ITS, LSU, mtSSU and RPB2 were obtained. PCR amplifications were undertaken using primers ITS1F/ITS4 for ITS, mrSSU1/mrSSU3R for mtSSU, LR0R/ LR5 for LSU and 5F/7CR for RPB2 (Vigalys and Hester 1990, White et al. 1990, Gardes and Bruns 1993, Rehner and Samuels 1994, Liu et al. 1999, Zoller et al. 1999). ITS, mtSSU and LSU PCR procedures in 25 µl reactions were carried out as outlined by Hou et al. (2009). PCR amplification of the RPB2 region was undertaken with an initial denaturation at 95 °C for 5 min, followed by 35 cycles of denaturation at 95 °C for 60 s, annealing at 55 °C for 60 s and elongation at 72 °C for 2 min and a final elongation at 72 °C for 10 min (Liu et al. 1999). The PCR products were purified, sequenced and edited by ZhongKe Xilin Biotechnology Co., Ltd. (Beijing, China). The new sequences were submitted to the NCBI GenBank database. Their accession numbers, as well as those for other ITS, LSU, mtSSU and RPB2 sequences downloaded from GenBank, are given in Table 1. Jacks. (Helotiales, Dermateaceae) were selected as outgroups. Maximum parsimony (MP) and Bayesian Inference (BI) analyses were performed on the concatenated ITS-LSU-mtSSU-RPB2 dataset. Each dataset was first aligned with Clustal X and then manually adjusted to allow maximum sequence similarity in Se-Al v.2.03a (Thompson et al. 1997;Rambaut 2000). Ambiguously aligned regions were excluded from the analysis by hand. Alignments were submitted to TreeBASE under accession number S25247. A partition homogeneity test was performed to determine the congruence of ITS, LSU, mtSSU and RPB2 (Farris et al. 1995;Huelsenbeck et al. 1996). After a positive outcome, the datasets were analysed together. The datasets were prepared and analysed with the maximum parsimony (MP) method using PAUP* 4.0b10 (Swofford 1998). The phylogenetic analysis was conducted using heuristic searches with 1000 replicates of random-addition sequence, tree bisection reconnection (TBR) branch swapping and no maxtree limit. All characters were equally weighted and unordered. Gaps were treated as missing data to minimise homology assumptions. A bootstrap analysis was performed with 1000 replicates, each with 100 random taxon addition sequences. Maxtrees were set to 1000 and TBR branch swapping was employed. For the Bayesian analysis, MrModeltest 2.3 with the Akaike Information Criterion (AIC) was used to choose the best-fit substitution models for the concatenated dataset: GTR+I+G for both ITS and LSU, HKY+I+G for mtSSU and SYM+G for RPB2. The Bayesian analysis was performed with MrBayes 3.1.2 (Huelsenbeck et al. 2011, Ronquist andHuelsenbeck 2003) with two sets of four chains (one cold and three heated) and the Stoprule option in effect, halting the analyses at an average standard deviation of split frequencies of 0.01. The sample frequency was set to 100 and the first 25% of trees were removed as burn-in and the remaining trees were kept and combined into one 50% majority-rule consensus tree. Bayesian Posterior Probabilities (PP) were obtained from the 50% majority consensus of the remaining trees. Clades receiving both bootstrap values of maximum parsimony (BP) ≥ 70% and PP ≥ 0.95 were considered to be significantly supported.
Conidiomata and zone lines not seen. Etymology. From Latin, rostriforme, referring to the beak-like protrusions observed at the ascospore poles.
Comments. Triblidium rostriforme is similar to T. carestiae (De Not.) Rehm but T. carestiae commonly has asci with 3-8 ascospores, ascospores with usually 7-14 transverse septa and ramose, multi-guttulate paraphyses. Diagnosis. Different from T. hafellneri by its ascospores with 6-8 transverse septa, narrow asci and geographical range. Different from its phylogenetically closest relatives (T. hubeiense and T. rostriforme) by the size and the shape of ascomata and ascospores.
Conidiomata and zone lines not seen. Etymology. Referring to the Yunnan Province where the holotype specimens were collected.

Comments.
Triblidium yunnanense is similar to T. hafellneri Magnes, but the latter has asci 20-25 µm wide, ascospores with 7 transverse septa and occurs on Vaccinum ovatum, Calluna vuglaris, Salix spp., and Nothofagus antarctica in Europe and the Americas. Triblidium yunnanense has a close relationship to the two other new species in this study, but T. rostriforme has larger ascomata, ascospores with special beak-like structures and T. hubeiense has larger ascomata, unbranched paraphyses, a moderately developed excipulum, a thicker covering stroma, basal layer and subhymenium. Notes. The placement of this species in Triblidium is demonstrated by the phylogeny presented in Fig. 1. Eriksson (1992) discussed the similarities between Huangshania and Triblidium in macro-morphology and in the morphology of hamathecial tissues and asci. The two genera differed only in ascospore morphology (elongate-phragmosporous vs. ellipsoidal-muriform). Karakehian et al. (2019) reviewed that ascospore morphology is a poor predictor of relatedness amongst these fungi. Huangshania verrucosa is the type species of the genus, therefore, Huangshania, is synonymized here under Triblidium.

Discussion
The morphological characteristics of the species described here are typical of Triblidium (Magnes 1997): ascomata on twigs of Rhododendron spp., muriform, inamyloid ascospores, and highly melanized ascomata with roughened outer surfaces. Based on our molecular phylogenetic analyses (Fig. 1), the three newly described species form a highly supported clade, sister to T. verrucosum. Triblidium yunnanense and T. hubeiense form a well-supported subclade sister to T. rostriforme. The similarity of ITS amongst these three new species is 90-95%. The sequences generated from the specimens of T. caliciiforme collected from France on bark of Quercus sp., clustered well with other sequences accepted as T. caliciiforme by Karakehian et al. (2019).
The strongly supported phylogenetic relationship justifying the synonymy of Huangshania with Triblidium was not detected by Karakehian et al. (2019)  that ascospore morphology appears to be a poor predictor of phylogenetic relationships amongst these fungi. It is worth noting that the rostrum of the ascospores in T. rostriforme and T. carestiae bear some similarity to the plug-like appendages of H. verrucosa. Furthermore, we did not transfer H. novae-fundlandiae (Rehm) Magnes, another species in Huangshania, to Triblidium since sequences were lacking.
In conclusion, three new Triblidium species from China were described in detail by both morphological and phylogenetic analyses. The new species, discovered in China, illustrate that these fungi are more widespread than previously known. Sequences from these new collections have expanded the representation of this genus in NCBI GenBank and helped our understanding of the family Triblidiaceae. Huangshania is placed in synonymy with Triblidium in order to maintain its monophyly, further demonstrating that ascospore morphology alone may be a poor predictor of evolutionary relationships.