Coryneumheveanum sp. nov. (Coryneaceae, Diaporthales) on twigs of Para rubber in Thailand

Abstract During studies of microfungi on para rubber in Thailand, we collected a new Coryneum species on twigs which we introduce herein as C.heveanum with support from phylogenetic analyses of LSU, ITS and TEF1 sequence data and morphological characters. Coryneumheveanum is distinct from other known taxa by its conidial measurements, number of pseudosepta and lack of a hyaline tip to the apical cell.

). The total para rubber tree plantation area in South East Asia exceeds more than 5 million hectares (Vongkhamheng et al. 2016). In Thailand, para rubber tree plantation area covers more than 3 million hectares and this number has increased every year (Kromkratoke andSuwanmaneepong 2017, Romyen et al. 2018). This important perennial crop is currently often affected by plant pathogenic fungi which can substantially decrease the quality and quantity of rubber yield ). Many taxa have proven to be serious pathogens worldwide, causing severe leaf spot formation, defoliation, shoot die-back and stem cankers (Jayasinghe 2000, Nyaka Ngobisa et al. 2015, Trakunyingcharoen et al. 2015, Liyanage et al. 2016. Nonetheless, information about the diversity of phytopathogenic taxa on para rubber from Thailand is generally lacking and currently there are only thirteen reports (Farr and Rossman 2018). Thus, the main objective of our project is to survey and study the diversity of microfungi associated with para rubber trees in Thailand. During the survey, we found a Coryneum species associated with canker disease on twigs of para rubber. This work is based on a combination of morphology and molecular data for identification this taxon.
Many Coryneum species have been reported as phytopathogens causing tree canker (Strouts 1972, Gadgil and Dick 2007, Horst 2013. This genus was introduced by Nees von Esenbeck (1817) to accommodate C. umbonatum as the type species. Historically, Coryneum species have relied on morphological studies and only a few species are supported by sequence data in GenBank. Many species causing tree canker, previously known as Coryneum were transferred to other genera e.g. Seiridium, Seimatosporium and Wilsonomyces (Sutton 1980, Raddi and Panconesi 1981, Marin-Felix et al. 2017. Recently, research has clarified the taxonomic position of the family Coryneaceae based on morphological and molecular data (Rossman et al., 2015;Senanayake et al., 2017;Wijayawardene et al., 2018). Currently 123 Coryneum species are listed in Index Fungorum (2018). Molecular analyses, using sequence data of LSU, ITS and TEF1 regions, has supplemented traditional taxonomic methods, enabling a more precise and rapid identification of species in the genus Coryneum . The correct identification of pathogenic fungi is necessary to implement appropriate quarantine decisions, suitable control strategies and to promote an understanding of the evolution of new pathogens and the movement of fungi between continents.

Collections, morphological studies and isolation
Fresh materials were collected from Chiang Rai, Thailand in 2016. Specimens were taken to the laboratory in zip lock bags and observed with a Motic SMZ 168 series stereomicroscope and photographed with an Axio camera on a Zeiss Discover V8 stereomicroscope. Sections of the conidiomata were mounted in double-distilled water (ddH 2 O) for morphological structures and photography. Images were taken us-ing a Canon 600D camera on a Nikon ECLIPSE 80i microscope. All measurements were calculated using Tarosoft® Image Framework programme v.0.9.0.7. Photoplates were made using Adobe Photoshop CS6 version 13.0 (Adobe Systems U.S.A.). The specimens were deposited in the Mae Fah Luang University Herbarium, Chiang Rai, Thailand (MFLU). Living cultures were deposited in Mae Fah Luang University Culture Collection (MFLUCC) in Thailand and duplicated at the Kunming Culture Collection (KUMCC). Faces of Fungi and Index Fungorum numbers are registered as described in Jayasiri et al. (2015) and Index Fungorum (2018).

DNA extraction, PCR and DNA sequencing
Genomic DNA was extracted from mycelium using Biospin Fungus Genomic DNA Extraction Kit (BioFlux®, Hangzhou, P.R. China) following the manufacturer's protocol. The DNA product was kept at 4 °C for the DNA amplification and maintained at -20 °C for long term storage. The DNA amplification was carried out by polymerase chain reaction (PCR) using three genes, the 28S large subunit (LSU), internal transcribed spacer (ITS) and translation elongation factor 1 alpha gene (TEF1). The LSU gene was amplified by using the primers LROR and LR5 (Vilgalys and Hester 1990), the ITS gene was amplified by using the primers ITS5 and ITS4 (White et al. 1990) and the TEF1 gene was amplified using the primers EF1-728F (Carbone and Kohn 1999) and EF2 (O'Donnell 1998). The amplification reactions were performed in 25 μl final volumes contained of 8.5 μl of sterilized ddH 2 O, 12.5 μl of 2 × Easy Taq PCR Super Mix (mixture of Easy Taq TM DNA Polymerase, dNTPs and optimised buffer (Beijing Trans Gen Biotech Co., Chaoyang District, Beijing, PR China), 1 μl of each forward and reverse primers (10 pM) and 2 μl of DNA template. The PCR thermal cycle programme for LSU and ITS gene amplification was provided as initially 94 °C for 3 mins, followed by 35 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 50 secs, elongation at 72 °C for 1 min and final extension at 72 °C for 10 mins. The PCR thermal cycle programme for TEF1 gene amplification was provided as initially 94 °C for 5 mins, followed by 40 cycles of denaturation at 94 °C for 45 secs, annealing at 52 °C for 30 secs, elongation at 72 °C for 1.30 mins and final extension at 72 °C for 6 mins. PCR products were sequenced by Sangon Biotech Co., Shanghai, China. Nucleotide sequences were deposited in GenBank (Table 1).

Phylogenetic analysis
Phylogenetic analyses were conducted based on a combined gene of LSU, ITS and TEF1 sequence data. Sequence data of Coryneaceae from previous studies and representative strains of major classes in Diaporthales were downloaded from GenBank to supplement the dataset (Table 1). The combined dataset consisted of 45 sequences including our newly generated sequences. Phaeoacremonium aleophilum (CBS 63194) and P. vibratile (CBS 117115) were selected as the outgroup taxa. The combined LSU, ITS and TEF1 gene dataset were initially aligned by using MAFFT version 7 (Katoh et al. 2017; http://mafft.cbrc.jp/alignment/server/) and improved manually, where necessary, in BioEdit v.7.0.9.1 (Hall 1999) and MEGA7 (Kumar et al. 2015). The final alignment of the combined LSU, ITS and TEF1 sequence datasets was analysed and inferred the phylogenetic tree based on maximum likelihood (ML), maximum parsimony (MP) and Bayesian inference analyses (BI). The estimated evolutionary model of Bayesian inference and maximum likelihood were performed independently for each locus using MrModeltest v. 2.3 (Nylander 2004) implemented in PAUP v. 4.0b10 (Swofford 2002). The best-fit model resulted as GTR+I+G model for each locus under the Akaike Information Criterion (AIC).
Maximum parsimony analysis was performed using PAUP v 4.0b10 (Swofford 2002). Trees were inferred using the heuristic search function with 1,000 random stepwise addition replicates and tree bisection-reconnection (TBR) as the branchswapping algorithm. All informative characters were unordered and of equal weight. The consistency index (CI), retention index (RI), rescaled consistency index (RC) and homoplasy index (HI) were calculated. Statistical supports for branches of the most parsimonious tree were estimated using maximum parsimony bootstrap (BS) analysis with 1,000 bootstrap replicates.
Bayesian inference was performed in MrBayes v. 3.2.2 (Ronquist and Huelsenbeck 2003) with the best-fit model of sequences evolution under the Akaike Information Criterion (AIC). Bayesian posterior probabilities (BY) (Rannala and Yang 1996, Zhaxybayeva andGogarten 2002) were determined by Markov Chain Monte Carlo Sampling (BM-CMC). Six simultaneous Markov chains were run from random trees for one million generations and trees were sampled every 100 th generation. The first 20% of generated trees representing the burn-in phase of the analysis were discarded and the remaining trees were used for calculating posterior probabilities (BY) in the majority rule consensus tree. The

Phylogenetic analysis
The dataset consisted of 45 taxa including the new taxa (Figure 1). The combined LSU, ITS and TEF1 sequence data including 2040 total characters, were analysed based on Bayesian inference, maximum likelihood and maximum parsimony analysis. RAxML analysis of the combined dataset had 996 distinct alignment patterns and 39.23% of undetermined characters or gaps. Maximum parsimony had 1191 constant characters, 151 variable parsimony-uninformative characters and 690 parsimony-informative characters. The most parsimonious tree is shown where TL = 2336, CI = 0.607, RI = 0.716, RC = 0.435, HI = 0.393. Bayesian posterior probabilities (BY) from MCMC were evaluated with the final average standard deviation of split frequencies = 0.005452. Phylogenetic analysis from ML, MP and BI gave trees with similar overall topologies of the generic placement and in agreement with previous studies ). The final RAxML tree of the combined dataset is shown in Figure 1, with a final ML optimisation likelihood value of -13004.6966291. The phylogeny shows that Coryneum heveanum forms a distinct lineage in Coryneum with strong support (94% ML, 95%MP and 1.00 BY) and in a sister clade to C. umbonatum, C. depressum, C. modonium, C. perniciosum and C. castaneicola.
Cultural characteristics. Conidia germinated on MEA within 24 h with germ tubes produced from one or both end cells, mostly from basal cell of conidia. Colonies on MEA reaching 20-25 mm diam. after 4 weeks at 25-30 °C, colonies circular, medium dense, cottony, margin wavy, superficial, slightly effuse, radially striated; colony from above, white, edges with more aerial mycelium than centre in the beginning and later become white grey, smooth with edge entire; from below: white to cream at the margin, yellowish-green in the centre in the beginning and later become dark green; not producing pigmentation in agar. Colonies on PDA reaching 10-15 mm diam. after 4 weeks at 25-30 °C, colonies circular, medium dense, cottony, slightly effuse, dark green with brown aerial mycelium on surface; not producing pigmentation in agar. Conidial masses were observed in PDA culture after 6 months at 25-30 °C. Mass of conidia dark brown to black, extruding on colony or tip of mycelium (Figure 2 q,  r). Mycelium superficial and immersed, dark brown, hyphae branched, septate, constricted at septa, thick, smooth-walled (Figure 2  Notes. Phylogenetically, Coryneum heveanum clustered in the same clade with C. umbonatum, C. depressum, C. modonium, C. perniciosum and C. castaneicola with high statistical support. Based on morphological characters, the conidia of C. castaneicola, C. depressum, C. elevatum, C. modonium and C. umbonatum have slightly curved conidia with an apical cell with a hyaline tip, while C. heveanum, C. castaneicola and C. perniciosum lack a hyaline tip ( Table 2) (Briosi and Farneti 1908, Sutton 1980, Gadgil and Dick 2007. Coryneum heveanum is similar to C. betulinum, C. perniciosum, C. psidi and C. pyricola in having broadly fusiform or clavate conidia but differs in size of conidia and number of pseudosepta (Table 2).

Discussion
Fungi on para rubber (Hevea brasiliensis) can be pathogens, saprobes or endophytes (Rocha et al. 2011, Seephueak et al. 2011, Ghazali 2013, Nyaka Ngobisa et al. 2015, Senwanna 2017. Fungal endophytes on para rubber have been comparatively well-studied (Gasiz and Chaverri 2010, Rocha et al. 2011, Déon et al. 2012, while few studies have investigated saprobic fungi or fungi associated with para rubber (Cai et al. 2013, Trakunyingcharoen et al. 2015. However, previous studies reporting saprobic taxa based on morphology, are available (Seephueak et al. 2011, Seephueak 2012. In this study, we introduced a new species, Coryneum heveanum, found on twigs of para rubber, based on morphological characters and phylogenetic analyses. Coryneum species are phytopathogenic fungi associated with twig blight, canker and dieback disease with some species reported as saprobes (Carter 1914, Strouts 1972, Gadgil and Dick 2007, Senanayake et al. 2018. Host-specificity of Coryneum has not yet been clarified and species have been recorded from various plant families worldwide (i.e. Betulaceae, Clusiaceae, Cupressaceae, Fagaceae, Hippocastanoideae, Malvaceae, Myrtaceae, Rosaceae, Ulmaceae) (Wehmeyer 1926, Strouts 1972, Sutton 1980 (Sutton 1980) --31-36 × 14-17; 4-5 Betula rubrum (Betulaceae) C. castaneicola (Sutton 1980) --57-80 × 10-13; apical cell with a hyaline tip; 6-7 Castanea dentata (Fagaceae) C. depressum (Sutton 1980) --44-53 × 19-23; apical cell with a hyaline tip; 4-5(-6) Quercus spp. (Fagaceae) C. elevatum (Sutton 1980) Sutton (1980) and only a few species are supported by molecular data with nine sequences from six species available in GenBank. However, we do not include Coryneum foliicola (CBS 153.32) sequence data in our analyses as its phylogenetic affinities are distant from Coryneaceae (data not shown). Therefore, we use reliable sequences from GenBank to determine the taxonomic placement of our new species. Based on morphological characters, there are some similarities between Coryneum heveanum and related Coryneum species, such as acervular conidiomata, fusiform or clavate conidia with pseudosepta (Sutton 1980, Maharachchimbura et al. 2016. However, C. heveanum is distinct from other known taxa including Coryneum umbonatum (type species) by conidial measurements, number of pseudosepta and lack of a hyaline tip to the apical cell (Table 2) (Briosi and Farneti 1908, Sutton 1980, Gadgil and Dick 2007. Current phylogenetic analyses of combined LSU, ITS and TEF1 alignment are used to clarify the species relationships in Coryneum (Figure 1), following Senanayake et al. (2017) and Fan et al. (2018). The phylogenetic tree shows that our species clearly groups with Coryneum. In addition, pairwise dissimilarities of DNA sequences of ITS regions between C. heveanum and other Coryneum species also provide further evidence to justify C. heveanum as a new species . Comparison of 599 nucleotides of the ITS nucleotides between C. heveanum and C. umbonatum (MFLUCC 13-0658 and MFLUCC 15-1110) reveals 90 base pair differences. Comparison of 536 nucleotides of the ITS nucleotides between C. heveanum and C. castaneicola (43_2) reveals 90 base pair differences. Comparison of 620 nucleotides of the ITS nucleotides between C. heveanum and C. umbonatum (CBS 199.68) reveals 91 base pair differences. Comparison of 598 nucleotides of the ITS nucleotides between C. heveanum and C. perniciosum (CBS 130.25) reveals 77 base pair differences. Coryneum umbonatum strains (AR3541, MFLUCC 13-0658 and MFLUCC 15-1110) form a distinct lineage, which is in agreement with the results of Fan et al. (2018). However, Coryneum umbonatum (CBS 199.68) forms a separate clade with C. umbonatum strain AR 3541, MFLUCC 13-0658 and MFLUCC 15-1110 and we cannot verify this taxon based on morphological characters. Previous studies have described the morphological features of Coryneum umbonatum but conidial dimensions and number of pseudosepta reported varies significantly from each other (Sutton 1980, Gadgil and Dick 2007 (Table 2). In addition, some of the Coryneum sequences deposited in GenBank (i.e. C. castaneicola, C. depressum, C. foliicola, C. monodia and C. perniciosum, C. umbonatum) lack morphological characteristics and their identities cannot be confirmed. Therefore, these taxa need to be recollected, described and sequenced to determine their taxonomic placement in this family.