Distoseptispora bambusae sp. nov. (Distoseptisporaceae) on bamboo from China and Thailand

Abstract Background Bamboo is a widespread plant with medicinal value. During our taxonomic study on medicinal plants, three collections of Distoseptispora were made from China and Thailand. Phylogenetic analyses of combined LSU, ITS and RPB2 sequence data showed that two collections represented a new species, phylogenetically distinct from other described species in Distoseptispora. New information This new species has macronematous, mononematous conidiophores, polyblastic or monoblastic conidiogenous cells and acrogenous, solitary, straight, obclavate, multi-septate, thick-walled conidia. Distoseptispora bambusae sp. nov. is introduced with illustrations and a comprehensive description. The third collection on dead wood from Thailand was identified as D. tectona with newly-generated molecular data for this taxon.

of dead wood was collected from the Botanical Garden, Mae Fah Luang University, Chiang Rai, Thailand (November 2019). The samples were processed and examined following the method described by Dai et al. (2017). Samples were brought to the laboratory in an envelope after recording the collection details including hosts, places and dates. Morphological observations were made using a stereomicroscope (SteREO Discovery. V12,Carl Zeiss Microscopy GmBH,Germany). Fruiting bodies were transferred with a needle and placed in a drop of distilled water on a glass slide, then covered with the cover slip for microscopic studies and photomicrography. The morphological figures were captured using a Nikon ECLIPSE Ni compound microscope (Nikon, Japan) fitted with a NikonDS-Ri2 digital camera (Nikon, Japan). Measurements were made using the Tarosoft (R) Image Frame Work software. Photo-plates were made with Adobe Photoshop CS6 software (Adobe Systems, USA).
Single-spore isolations were done following the method described in (Chomnunti et al. 2014). Germinated spores were transferred to potato dextrose agar (PDA: 39 g/l sterile distilled water, Difco potato dextrose) plates and incubated at room temperature for 4 weeks. Herbarium materials were deposited in the Fungarium of Mae Fah Luang University (MFLU), Chiang Rai, Thailand. Pure cultures were deposited in the Mae Fah Luang University Culture Collection (MFLUCC) and International Collection of Microorganisms from Plants (ICMP). Facesoffungi (FoF) and Index Fungorum numbers were acquired as described in  and Index Fungorum (http://www.indexfungorum.org).

DNA extraction, PCR amplification and sequencing
Fresh fungal mycelia were scraped with sterilised scalpels. Genomic DNA was extracted using Genomic DNA Extraction Kit (GD2416) following the manufacture's protocol. PCR amplifications were performed in a 20 μl reaction volume, with 10 μl of 10 × PCR Master Mix, 1 μl of each primer, 1 μl template DNA and 7 μl ddH O. Primers used and PCR thermal cycle programmers are listed in Table 1

Phylogenetic analyses
Sequences (Table 2) generated during this study were complemented with sequences from previous studies , Su et al. 2016, Luo et al. 2018   The Maximum Likelihood (ML) analysis was performed using IQ-tree (Nguyen et al. 2015, Chernomor et al. 2016. Nucleotide substitution models were selected under the Akaike Information Criterion (AIC) by jModelTest2 (Darriba et al. 2012) on XSEDE in the CIPRES web portal (Miller et al. 2010). For ITS dataset, the GTR+I+G model was selected (-lnL= 3364.5406), for LSU, the TIM2+I+G model (-lnL = 959.3999), and for RPB2, the GTR+I+G (-lnL= 5111.0788). ML was inferred under partitioned models. Non-parametric bootstrap analysis was implemented with 1000 replicates.
Maximum Parsimony (MP) analysis was carried out with the heuristic search in PAUP v. 4.0b10 (Swofford 2002). All characters were unordered and of equal weight, and gaps were treated as missing data. Maxtrees were unlimited, branches of zero length were collapsed and all multiple, equally-parsimonious trees were saved. Clade stability was assessed using a bootstrap (BT) analysis with 1,000 replicates, each with 10 replicates of random stepwise addition of taxa (Hillis and Bull 1993).
Bayesian Inference (BI) analysis was performed by the Markov Chain Monte Carlo sampling (MCMC) coalescent approach implemented in BEAST v1.8.4 (Drummond et al. 2012), with an uncorrelated lognormal relaxed clock. The Birth-Death Incomplete Sampling speciation model (Stadler 2009) was selected as tree prior. The nucleotide substitution models were the same as above. Markov chains were run for 1,000,000 generations and trees were sampled every 1000th generation. The XML file generated by BEAUti (Drummond et al. 2012)   Culture characteristics: Conidia germinated on PDA within 12 hours and germ tubes were produced from both ends. Colony reached 30 mm in 4 weeks at 26℃ on PDA media, circular, flat, surface rough, grey from above, brown from below, edge entire.

Notes:
The morphological characteristics of Distoseptispora bambusae match well with the generic concept of Distoseptispora (Su et al. 2016). Multi-gene analyses showed that D. bambusae is a phylogenetically-distinct species, most closely related to D. suoluoensis, a species isolated from submerged wood in a freshwater habitat . Distoseptispora bambusae has shorter conidiophores (40-96 vs. 80-250 μm) and shorter conidia (45-74 vs. (65-) 80-125(-145) μm) than those of D. suoluoensis . Our two specimens of D. bambusae were similar in morphology, but polyblastic conidiogenous cells were observed from the Chinese specimen, while the Thai specimen has only monoblastic conidiogenous cells. These may be due to geographical differences and the different observation period. Although the two strains clustered together with short branches in the phylogenetic tree, comparisons of ITS sequences showed that there are 3 bp (base pair) differences without gaps between two strains and we identified them as the same species following the guidelines for species delineation proposed by Jeewon and .

Etymology
Bambusae, referring to the host.

Analysis
Partial nucleotide sequences of the LSU, ITS and RPB2 were used to determine the phylogenetic position of the taxa isolated. Sequences of 47 strains retrieved from GenBank, representing species of Distoseptispora and two outgroups A. fusiformis (MFLU 18-1601) and A. lignicola (MFLUCC 15-0377), were analysed. Single gene analyses were done to compare the topologies and clade stabilities, respectively. Nucleotide substitution models were selected by jModelTest2 on XEDE (Drummond et al. 2012). For the ITS and RPB2 dataset, the GTR+I+G model was selected, for LSU, the TIM2+I+G. The manuallyadjusted LSU, ITS and RPB2 alignment comprised a total of 2,246 characters (768 for LSU; 436 for ITS; 1,042 for RPB2), including coded alignment gaps. Amongst them, 1,471 characters were constant, 195 variable characters were parsimony-uninformative and number of parsimony-informative characters was 580. One thousand equally most parsimonious trees (Tree length = 1799, CI = 0.640, RI = 0.733, RC = 0.469, HI = 0.360) were yielded from the heuristic search. MP, ML and Bayesian analyses of the combined dataset inferred similar topologies, respectively. The "most likelihood" tree is presented (Fig. 4).

Discussion
In this study, two collections from China and Thailand, representing a new Distoseptispora species, is introduced, based on morphology and phylogenetic analysis. The two samples were both found on bamboo from terrestrial habitats. It is the fourth species found from medicinal plants. The other three are D. palmarum, D. thailandica and D. xishuangbannaensis (Tibpromma et al. 2018.
Distoseptispora species does not seem to have specific habitat preferences. Most of them are reported from submerged wood in freshwater habitats, while some species have been introduced from terrestrial habitats (Luo et al. 2018, Tibpromma et al. 2018. So far, Distoseptispora were only found in China and Thailand. They may exist in other countries, waiting to be discovered on the basis of their diverse habitats.
The asexual morph of Distoseptispora is similar to Sporidesmium in producing holoblastic, euseptate or distoseptate conidia and blastic, terminal conidiogenous cells (Shenoy et al. 2006, Luo et al. 2018. Sexual morphs of Distoseptispora have not been reported. Acrodictys martini was transferred to Distoseptispora as D. martini by Xia et al. (2017), based on their phylogenetic analysis. However, this species morphologically resembles Acrodictys rather than Distoseptispora. Therefore, the molecular data of Distoseptispora martini may need further verification (Luo et al. 2018).
It is interesting to note that, in most species of Distoseptispora, the conidia are longer than their conidiophores, while in some, they are shorter than their conidiophores. However, this characteristic does not reflect their phylogenetic position. For example, D. obpyriformis Z.L.
Luo & H.Y. Su, a species that has long conidia and short conidiophores and D. rostrata Z.L. Luo, K.D. Hyde & H.Y. Su that has longer conidiophores, but shorter conidia, form a sister clade in the phylogenetic tree.