Beta-tubulin and Actin gene phylogeny supports Phaeoacremoniumovale as a new species from freshwater habitats in China

Abstract A new species of Phaeoacremonium, P.ovale (Togniniaceae), was isolated during a diversity study of freshwater fungi from Yunnan Province in China. Morphological and cultural studies of the fungus were carried out and its sexual and asexual morphs (holomorph) are introduced herein. This species is characterised by peculiar long-necked, semi-immersed ascomata with oval to ellipsoid ascospores and ellipsoid to ovoid conidia. Phylogenetic analyses of a combined TUB and ACT gene dataset revealed that strains of P.ovale constitute a strongly supported independent lineage and are related to P.griseo-olivaceum and P.africanum. The number of nucleotide differences, across the genes analysed, also supports establishment of P.ovale as a new species.


Introduction
Lignicolous freshwater fungi are important in nutrient recycling . A number of taxonomic studies have focused on the diversity of such fungi in the South East Asian region and these investigations have reported a number of novel species (e.g. Jeewon et al. 2003;Cabanela et al. 2007;Zhang et al. 2008;Luo et al. 2018). In this study, we report a new species of Phaeoacremonium isolated from decaying wood from a stream in Yunnan Province, China.
Phaeoacremonium (= Togninia), introduced by Crous et al. (1996), is typified by P. parasiticum and it belongs to Togniniaceae (Gramaje et al. 2015). Phaeoacremonium was reported to be the asexual morph of Togninia (Mostert et al. 2003(Mostert et al. , 2006aPascoe et al. 2004). Gramaje et al. (2015) proposed Phaeoacremonium over Togninia as the correct name based on common usage and this has been listed in Réblová et al. (2016) and followed in Wijayawardene et al. (2018). The species are basically characterised by black ascomata with a long neck and clavate to cylindrical asci with oval to ellipsoid, hyaline ascospores and straight or flexuous mononematous conidiophores with oval to reniform phialo-conidia (Marin-Felix et al. 2018;Spies et al. 2018).
Most species of Phaeoacremonium are plant or/and human pathogens and some have been recorded on arthropods or in soil (Groenewald et al. 2001;Guarro et al. 2003;Hemashettar et al. 2006;Mostert et al. 2006a;Damm et al. 2008;Gramaje et al. 2015) while others are causal agents of Petri disease and esca of grapevines (Pascoe et al. 2004;Rooney-Latham et al. 2005a;Mostert et al. 2006b). Phaeoacremonium species can also infect a wide range of woody hosts, such as cherry, apricot, olive and peach trees (Rumbos 1986;Di Marco et al. 2004;Kubátová et al. 2004). Recent studies have reported the importance of Phaeoacremonium species in causing brown wood streaking of Olea spp. and Prunus spp. (Mostert et al. 2006b;Damm et al. 2008;Gramaje et al. 2012;Nigro et al. 2013;Olmo et al. 2014;Carlucci et al. 2015). Rooney-Latham et al. (2004, 2005a reported that, in the presence of water, spores in some Phaeoacremonium species are forcibly discharged from perithecia through the long neck and exit the ostiole to be dispersed by wind, rain or insects in order to colonise other substrates. Recently Hu et al. (2012) introduced a freshwater inhabiting species, Phaeoacremonium aquaticum (= Togninia aquatica).
Species of Togniniaceae have been reported to colonise substrates in different types of habitats and recent taxonomic studies have revealed additional new species (Gramaje et al. 2015). We have been studying fungi along a north-south gradient in the Asian region ) and, in this study, we report on two collections of Phaeoacremonium from China. The aim here is to characterise these two strains as one novel species based on morphology as well as to investigate their phylogenetic affinities with previously known Togniniaceae species based on partial TUB and ACT genes.

Sample collection, morphological studies and isolation
Submerged dead wood was collected from Baoshan, Yunnan Province in China in October 2016, brought to the laboratory in zip lock plastic bags and treated in the laboratory following procedures detailed in Luo et al. (2018). Fruiting bodies were found growing on decaying wood in a sterile plastic box after two weeks of incubation and the fungus was subsequently isolated based on the method of Chomnunti et al. (2014). Specimens were examined by a Motic SMZ 168 stereomicroscope. Micromorphological characters were examined using a Nikon ECLIPSE 80i compound microscope and images were captured with a Canon EOS 600D digital camera. Identification of colours was based on Ridgway (1912). The Taro soft Image Framework programme version 0.9.0.7 was used for measurements. Single spores were isolated and grown on water agar (WA) and potato dextrose agar (PDA) media. Ascospores germinated on PDA within 1 week. The colonies were transferred to WA and PDA to promote sporulation (sporulation occurred after 30 days in PDA). The cultures were checked 2 to 3 times per week and all procedures were performed in a sterile environment and at room temperature. The morphological characters of the asexual morph were examined after sporulation. Specimens are deposited in the Kunming Institute of Botany, Academia Sinica (KUN) and duplicated in Mae Fah Luang University (MFLU) Herbarium, Chiang Rai, Thailand. Facesoffungi numbers (FoF) (http://www.facesoffungi. org/) were obtained as stated in Jayasiri et al. (2015) and Index Fungorum numbers (IF) (http://www.indexfungorum.org/names/IndexFungorumRegisterName.asp).

DNA extraction, PCR amplification and sequencing
Total genomic DNA was extracted from mycelium using a Trelief Plant Genomic DNA Kit following the instructions of the manufacturer. The genomic DNA was amplified by using polymerase chain reaction (PCR) in a 25 μl reaction mixture. Partial regions of the beta-tubulin (TUB) and Actin (ACT) gene were amplified using the primer pairs T1 (O'Donnell and Cigelnik 1997) and Bt2b (Glass and Donaldson 1995), ACT-513F and ACT-783R (Carbone and Kohn 1999), respectively. The internal transcribed spacers (ITS) regions of the rDNA (ITS1-5.8S-ITS2) were also amplified using primer pairs ITS5 and ITS4 (White et al. 1990) but no further analyses were done on these due to lack of sequence data. The PCR conditions for these regions were as follows: an initial denaturation at 94 °C for 3 min, followed by 35 cycles of denaturation at 94 °C for 30 sec, annealing at 51 °C (TUB) or 60 °C (ACT) or 55 °C (ITS) for 50 sec and extension at 72 °C for 1 min, with a final extension at 72 °C for 10 min. PCR products were then sequenced with the primers mentioned above by a commercial sequencing provider (Tsingke Company, Beijing, P.R. China).

Phylogenetic analysis
The quality of the amplified nucleotide sequences was checked and combined by Seq-Man version 7.1.0 (44.1) and Finch TV version 1.4.0 (www.geospiza.com). Sequences used by Marin-Felix et al. (2018), Spies et al. (2018) and the closest matches for our strains were retrieved from the National Center for Biotechnology Information (NCBI) by nucleotide BLAST. Sequences were aligned in MAFFT v. 7.310 (http:// mafft.cbrc.jp/alignment/server/index.html) (Katoh and Standley 2016) and manually corrected in Bioedit 7.0.9.0 (Hall 1999).
The phylogenetic analyses of combined gene regions (TUB and ACT) were performed using maximum-likelihood (ML) and Bayesian Inference (BI) methods. The best-fit model (GTR+G+I) was obtained using jModelTest 2.1.10 under the Akaike Information Criterion (AIC) calculations (Darriba et al. 2012). The ML analysis was enforced with RAxML-HPC v.8 on XSEDE (Stamatakis 2014;Miller et al. 2015) with 1000 rapid bootstrap replicates. Bayesian inference was implemented by MrBayes v. 3.0b4 (Ronquist and Huelsenbeck 2003). Four simultaneous Markov chains were run for 5,000,000 generations sampling one tree every 1000 th generations and other criteria as outlined by Hongsanan et al. (2017). The temperature value was lowered to 0.15, burn-in was set to 0.25. Gaps were treated as missing data with no differential weighting of transitions against transversions and the partition homogeneity test was performed to assess whether datasets from different genes were congruent. Phylogenetic trees were viewed with FigTree v1.4.0 (http:// tree.bio.ed.ac.uk/software/figtree/) and processed by Adobe Illustrator CS5. Alignment and trees were deposited in Tree-BASE (submission ID: 22810). The nucleotide sequence data of the new taxon have been deposited in GenBank (Table 1).
Culture characteristics. Ascospore germinating on PDA within 1 week at 23°C, germ tubes produced from ends. Colonies growing on PDA, reaching 2 cm diam. and sporulating after 30 days. Colonies semi-immersed to superficial, irregular in shape, flat, slightly raised, with undulate edge, slightly rough on surface, cottony to fairly fluffy, colony from above, greyish-brown (5F3-5, Ridgway 1912) at the margin, initially write to cream (5A1-3) in the centre, becoming dark brown (5F7-8) at the margin, orange-white (5B1-3) at the centre; from below, initially, greyish-brown at the margin, white at the centre, becoming dark brown at the margin, orange-white at the centre, producing brown pigmentation in agar.
In this study, we introduce a novel taxon of Phaeoacremonium from dead wood collected in a stream in the Yunnan Province, China and describe its sexual and asexual morph. Examination of morphological characters reveal that our species is sufficiently distinct from extant species to establish it as a new species. Analyses of the combined DNA sequence dataset from partial TUB and ACT genes also support that this taxon is a Phaeoacremonium species and phylogenetically distinct from other species (Fig. 1). The two strains of P. ovale constitute a strongly supported independent lineage close to other species as depicted in Clade I. Phylogeny also reveals a close relationship to P. griseo-olivaceum, but with low support. To further support P. ovale as a new species, we compared nucleotide differences with other related species as recommended by Jeewon and Hyde (2016). Comparison of the 533 nucleotides across the TUB region reveals 43 bp (10%) differences, 256 bp of the ACT region reveals 22 bp (8.5%) differences and 517 bp of the ITS region reveals 4 bp (1%) differences compared to P. griseo-olivaceum (CBS 120857). Examination of the TUB region reveals 59 bp (11%) difference compared to P. africanum (CBS 120863) while the ACT region reveals 19 bp (7%) and ITS region reveals 17 bp (3%) differences, but the latter clusters in a different subclade in our phylogeny and is therefore considered distinct. There are also some morphological similarities between P. ovale and P. africanum in terms of black ascomata with a long neck, clavate asci and small, oval to ellipsoid ascospores in sexual morph and ellipsoid to ovoid, aseptate conidia in asexual morph (Damm et al. 2008). Despite a morphological resemblance to P. africanum and close relationship to P. griseoolivaceum, there are other differences across these species. Phaeoacremonium ovale was collected from an aquatic habitat and from dead wood in China whereas the former two species were collected from Prunus spp. in South Africa (Damm et al. 2008). In addition, conidial size of P. africanum and P. griseo-olivaceum are 5-12 × 1.5-2 μm and 5-8 × 1.5-2 μm, whereas conidia of P. ovale measure 2.5-6 × 1-2.5 μm (Damm et al. 2008;Fig. 3). No sequence data of the TUB and ACT gene are available for P. aquaticum and P. leptorrhynchum and therefore we provide ITS sequences of our strains and compare them with those two species. Comparison of ITS regions reveals 61 bp (12%) differences with P. aquaticum (IFRDCC 3035) and 11 bp (2%) differences with P. leptorrhynchum (UAMH9590). In addition, our new species is also morphologically different from them. Phaeoacremonium ovale is morphologically different as ascospores of P. aquaticum and P. leptorrhynchum are reniform (ascospores of P. ovale are oval/ellipsoid) and measure 5-6 × 1-1.5 μm and 7-10 × 1-1.5 μm, respectively. Phaeoacremonium inconspicuum as described by Gramaje et al. (2015) also appears morphologically similar to P. ovale in terms of clavate asci and hyaline, aseptate ascospores (Eriksson and Yue 1990), but could not be included in our analyses as DNA sequences are unavailable. However, the ascospore shape and size of P. inconspicuum is different (allantoid, measuring 7-10 × 1.5-2 μm) (Eriksson and Yue 1990;Réblová 2011).