Morphological and phylogenetic appraisal of Ophioceras (Ophioceraceae, Magnaporthales)

Ophioceras is accommodated in the monotypic family Ophioceraceae (Magnaporthales, Sordariomycetes), and the genus is delimited based on molecular data. During an ongoing survey of bambusicolous fungi in southwest China, we collected a submerged decaying branch of bamboo from Sichuan Province, China and an Ophioceras species occurring on this substrate was observed and isolated. An Ophioceras taxon was delimited based on morphological characteristics and combined LSU, RPB1 and ITS sequence analyses and is described as Ophioceras sichuanense sp. nov. The species formed a well-supported clade basal to Ophioceras (100% ML, 1.00 PP). Based on the updated phylogenetic tree of Magnaporthales, Ceratosphaerella castillensis (generic type) and C. rhizomorpha formed a clade within Ophioceras and morphologically resemble Ophioceras. Therefore, Ceratosphaerella is synonymized under Ophioceras. The phylogenetic relationships of Ophioceras are discussed in relation to morphological similarities of genera in Magnaporthales. The generic circumscription of Ophioceras is emended.

Single spore isolation based on the spore suspension technique [24] was carried out to obtain a pure fungal culture. Germinated ascospores were transferred to the new potato dextrose agar plates (PDA; Qingdao Daily Water Biotechnology co. LTD. Shandong, P.R. China) under aseptic conditions and grown under normal day/nightlight at room temperature. Culture characteristics (e.g., growth, shape, colour, margin, elevation, consistency) were checked and recorded after one week and four weeks.
The holotypic specimen is conserved in the herbarium of Cryptogams Kunming Institute of Botany Academia Sinica (KUN-HKAS), Yunnan, P.R. China. The isotype is stored in the herbarium of Mae Fah Luang University, Chiang Rai, Thailand (MFLU). Ex-type living cultures are preserved in the Kunming Institute of Botany Culture Collection (KUMCC) and Mae Fah Luang University Culture Collection (MFLUCC). Facesoffungi and Index Fungorum numbers were registered for the new taxon [11,25].

DNA extraction, amplification and sequencing
Fungal genomic DNA was extracted from fresh mycelia using the Biospin Fungus Genomic DNA Extraction Kit (BioFlux 1 , P.R. China) following manufacturer's instructions (Hangzhou, P.R. China) and also extracted from fruiting bodies (= pseudostromata) directly using the Forensic DNA Kit (Omega 1 , USA) for a duplicated strain. DNA amplification was performed by polymerase chain reaction (PCR). Two gene regions including the internal transcribed spacer (ITS) and 28S large subunit rDNA (LSU), were used to amplify PCR fragments using forward and reward primer pairs: ITS5/ITS4 [26] and LR0R/LR5 [27], respectively. PCR reactions were conducted in a 25 μL total volume, consisted of 2 μl of DNA template, 1 μl of each forward and reverse primer, 12.5 μl of 2× Power Taq PCR Master Mix (Beijing BioTeke Corporation, P.R. China) and 8.5 μl double-distilled water (ddH 2 O). The PCR thermal cycle program for ITS and LSU was set up following Jiang et al. [28]. PCR fragments were purified and sequenced at TsingKe Biological Technology (Beijing) Co., Ltd, P.R. China.

Molecular phylogeny
The newly generated sequences (ITS and LSU) of fungal strains were initially subjected to the basic local alignment search tool (BLASTn) via the National Center for Biotechnology Information web portal (NCBI; https://blast.ncbi.nlm.nih.gov) for discovering closely related fungal taxa. In order to clarify the phylogenetic placement of the new isolate, the representative taxa in Magnaporthales were incorporated with the new taxon to generate the sequence data matrix for further analysis. These representative taxa of Magnaporthales were downloaded from the GenBank database (Table 1) based on recent publications [2,29].
Preliminary single-gene data matrixes were aligned via MAFFT v. 7.452 [30] and improved manually in BioEdit v. 5.0.6 [31]. The single-gene alignments of LSU and ITS data matrixes were prior analyzed by maximum-likelihood (ML) criterion using RAxML v. 7.0.3 [32,33] for checking if there are any conflicts between the tree topologies. The concatenated LSU-ITS and LSU-RPB1-ITS sequence datasets were further analyzed based on maximum-likelihood (ML)  [34]. The ML + thorough bootstrap parameters were set at default values but modified as 1000 replications of bootstraps (-N 1000) and using the GTRGAMMAI model.
The best-fit evolutionary models of nucleotide substitution for LSU, RPB1 and ITS loci were evaluated by MrModeltest 2.3 [35], and the GTR+I+G substitution model under the Akaike Information Criterion (AIC) was the best-fit evolutionary model for each locus. Bayesian inference (BI) analysis was performed by MrBayes v. 3.1.2 [36]. The Markov Chain Monte Carlo sampling (MCMC) sampling method was used to determine posterior probabilities (PP) [37,38]. One million generations of six simultaneous Markov chains were run and sampled every 100th generation. MCMC heated chain was set up with a "temperature" value at 0.15. The burn-in was set to 20% of all sampled trees, meaning that sampled trees beneath the , USA) and converted to jpeg file using Adobe Photoshop CS6 (Adobe Systems Inc., USA). New sequences generated from the present study were registered for GenBank accession numbers ( Table 1). The final alignment and phylogram are submitted in TreeBASE submission ID: 28293 (http://purl.org/phylo/treebase/phylows/study/TB2: S28293?x-access-code=66338d666c9ae6b7c0a0aa779b50078d&format=html).

Nomenclature
The electronic version of this article in Portable Document Format (PDF) in a work with an ISSN or ISBN will represent a published work according to the International Code of Nomenclature for algae, fungi, and plants, and hence the new names contained in the electronic publication of a PLOS ONE article are effectively published under that Code from the electronic edition alone, so there is no longer any need to provide printed copies.
In addition, new names contained in this work have been submitted to Index Fungorum from where they will be made available to the Global Names Index. The unique Index Fungorum number can be resolved, and the associated information viewed through any standard web browser by appending the Index Fungorum number contained in this publication to the prefix www.indexfungorum.org/. The online version of this work is archived and available from the following digital repositories: PubMed Central and LOCKSS.

Compliance with ethical standards
There is no conflict of interest (financial or non-financial) and all authors have agreed to submission of paper. The authors also declare that they have no conflict of interest and confirm that the field studies did not involve endangered or protected species.

Molecular phylogeny
Based on the results from the nucleotide BLAST search tool of LSU sequence, our new strains (KUMCC 20-0213 and KUN-HKAS 107677) are closely related to species of Ophioceras, whereas ITS sequence revealed that our new strains are similar to the unidentified fungal endophyte isolate 4583 (86.79% similarity) and other taxa in Magnaporthales. The concatenated LSU-RPB1-ITS dataset included 2594 total characters with gaps (LSU: 1-905 bp, RPB1: 906-1877 bp, ITS: 1878-2594 bp). The best scoring ML tree with the final ML optimization likelihood value of -23570.931644 (ln) was selected to represent the phylogenetic relationships of taxa in Magnaporthales (Fig 1). All free model parameters were estimated using the GTRGAMMAI model, with 1354 distinct alignment patterns and 40.89% of undetermined characters or gaps. Estimated base frequencies were as follows: A = 0.247576, C = 0.255381, G = 0.292293, T = 0.204750, with substitution rates AC = 1.508478, AG = 2.781197, AT = 1.835341, CG = 0.983246, CT = 6.427316, GT = 1.000000. The Tree-Length = 7.381586 and the gamma distribution shape parameter α = 0.605644. The evaluation of Bayesian posterior probabilities (BYPP) from MCMC was carried out with the final average standard deviation of split frequencies reached 0.009301.
The tree topology from ML analysis showed similar results with the BI analysis and comparing LSU-ITS, and LSU-RPB1-ITS phylograms also revealed similarities in overall

PLOS ONE
Ophioceras (Ophioceraceae, Magnaporthales) topologies (Fig 1 and S2 Fig). Thus, we will use the LSU-RPB1-ITS topology for further discussion. Five families of Magnaporthales were included in the presented phylogenetic analyses viz. Ceratosphaeriaceae, Magnaporthaceae, Ophioceraceae, Pseudohalonectriaceae and Pyriculariaceae. These five families formed well-resolved monophyletic clades within Magnaporthales with significant support (greater than 70% ML and 0.95 PP) in our combined gene analyses (Fig 1 and S2 Fig). Ophioceraceae has a close relationship with Ceratosphaeriaceae and Pseudohalonectriaceae. However, the phylogenetic relationships of these three families are not well resolved and pending further clarification.
Phylogenetic analyses of the LSU-RPB1-ITS sequence matrix revealed that the investigated specimen (KUN-HKAS 107677) and its pure culture (KUMCC 20-0213) are grouped together and form an independent lineage basal to Ophioceras in Ophioceraceae with high statistical support (100% ML, 1.00 PP; Fig 1). Considering the phylogenetic results and morphology, we propose a novel species, Ophioceras sichuanense, occurring on submerged bamboo in Sichuan Province, China.
Culture characteristics: Ascospores germinated on PDA within 24 hours at room temperature under normal condition. Mycelium superficial to immersed in agar medium, branched,  [17]. Ophioceras sichuanense resembles O. leptosporum and O. tenuisporum due to the size ranges of asci and ascospores and extremely long ostiolar necks. However, the species differs from the latter two species in their ascospore septation ( Detailed description and illustration: see Huhndorf et al. [8].
Known hosts/ habitat and distribution: Saprobic on bark or wood in terrestrial habitat. To date, Ophioceras castillensis is only reported from Costa Rica, Nicaragua, and Puerto Rico [8].
Notes-In this study, Ceratosphaerella castillensis is transferred to Ophioceras as O. castillensis based on a concatenated LSU-RPB1-ITS analyses coupled with morphological similarity of the ascomata and ascomatal wall which is typical Ophioceras. Ophioceras castillensis can be separated from other Ophioceras species in forming ascomata on large clusters, superficial on sparse, subicular hyphae and having clavate asci and hyaline to pale brown, fusiform, 3-septate ascospores [8]. Detailed morphological comparison and taxa habitats in Ophioceras are described in Table 2.

PLOS ONE
available in GenBank. In preliminary phylogenetic analysis of SSU sequence matrix, the species formed a stable clade within Magnaporthaceae rather than Ophioceraceae. Ophioceras arcuatisporum needs to be re-visited and re-illustrated, incorporating details from molecular data. Luo et al. [2] performed combined LSU and TEF1-α phylogenetic analyses to investigate the relationships of taxa in Magnaporthales. In their phylogeny, Ceratosphaeria grouped with Pseudohalonectria (Pseudohalonectriaceae) and separated from Ophioceras (Ophioceraceae). Based on the fact that Ceratosphaeria differs from Pseudohalonectria in having narrow cylindric-fusiform to filiform and longer ascospores, Ceratosphaeriaceae was thus introduced as a new family within Magnaporthales to accommodate Ceratosphaeria [2]. In the present study, we performed an updated phylogenetic tree based on a concatenated LSU-RPB1-ITS sequences and showed that Ceratosphaeria (Ceratosphaeriaceae) clustered with Ophioceras (Ophioceraceae) with low statistical support, suggesting that gene selection in the data matrix affects the tree topology at the familial levels in Magnaporthaceae. Ceratosphaeria is morphologically similar to Ceratosphaerella [8]. Although Ceratosphaeria clustered with Ophioceras with low statistical support, Ceratosphaeria possibly belongs to Ophioceraceae. However, the phylogenetic status of Ceratosphaeria needs to be clarified with more evidence in the future studies.
In the present study, we synonymize Ceratosphaerella under Ophioceras based on molecular phylogeny coupled with similar ascomatal morphology. Phylogenetic analyses based on the LSU sequence dataset (S1 Fig) and the concatenated LSU-ITS (S2 Fig) and the LSU-RPB1-ITS (Fig 1)  Hence, more reliable gene loci (e.g., SSU, ITS, RPB1 and TEF1-α) from the ex-type strain of O. rhizomorpha should be obtained and the epitype of O. castillensis should be designated and incorporated with morpho-molecular based taxonomic treatment. Furthermore, the new collections and sequence data of taxa in Ophioceras are required to provide a better taxonomic resolution for robust species delineations in this genus.
Many genera in Magnaporthales have similar morphological characteristics with Ophioceras (Table 3). However, these genera are considered distinct genera based on phylogenetic investigations [1,2,8,[59][60][61]. Pseudohalonectria (Pseudohalonectriaceae) is also similar to Ophioceras in its ascomata and asci. However, Pseudohalonectria varied in shape of ascospores, such as ellipsoidal, fusiform or filiform [62]. In the present study, Pseudohalonectria (Pseudohalonectriaceae) formed an independent lineage separate from other families in Magnaporthales; however, Pseudohalonectria could not be resolved at the species level such as in Perera et al. [62]. It may be because molecular data of most taxa in this genus are unavailable in GenBank database. Moreover, some sequences of P. lignicola deposited in GenBank are likely to be misidentified [62]. Therefore, sequences of Pseudohalonectria species used for phylogenetic analyses are limited.