We conducted our fungal collection surveys from April to October in China, and found Rhus chinensis to be one of the major tree species in Jiangxi Province. Twigs, branches and stems were carefully checked, and diseased tissues were cut into small pieces and packed in paper bags. Isolates were obtained by transferring the ascospores or conidial masses from ascomata to sterile PDA plates, incubating at 25 °C until spores germinated. Single germinating spores were transferred onto new PDA plates, which were kept at 25 °C in darkness. Specimens were deposited in the Museum of the Beijing Forestry University (BJFC) and axenic cultures maintained in the China Forestry Culture Collection Centre (CFCC).

Recognition and identification of the fungal species on Rhus chinensis was based on fruiting bodies formed on the bark. Ascomata and conidiomata were sectioned by hand using a double-edged blade, and microscopic structures were observed under a dissecting microscope. At least 10 conidiomata/ascomata, 10 asci, and 50 conidia/ascospores were measured to calculate mean and standard deviation. Measurements are reported as maxima and minima in parentheses and the range representing the mean plus and minus the standard deviation of the number of measurements given in parentheses (Voglmayr et al. 2017). Microscopy photographs were captured with a Nikon Eclipse 80i compound microscope equipped with a Nikon digital sight DS-Ri2 high definition colour camera, using differential interference contrast illumination. Nomenclatural novelties and descriptions were deposited in MycoBank (Crous et al. 2004).

Genomic DNA was extracted from colonies grown on cellophane-covered PDA plates using a modified CTAB method (Doyle and Doyle 1990). PCR amplifications were performed in a DNA Engine Peltier Thermal Cycler (PTC-200; Bio-Rad Laboratories, Hercules, CA, USA). The primer sets ITS1/ITS4 (White et al. 1990) were used to amplify the ITS region. The primer pair LR0R/LR5 (Vilgalys and Hester 1990) was used to amplify the LSU region. The primer pairs EF1-688F/EF1-986R or EF1-728F/TEF1-LLErev (Carbone and Kohn 1999; Jaklitsch et al. 2005; Alves et al. 2008) were used to amplify tef1 gene. The primer pair dRPB2-5f/dRPB2-7r (Voglmayr et al. 2016) was used to amplify the rpb2 gene. The polymerase chain reaction (PCR) assay was conducted as described by Fan et al. (2018). PCR amplification products were assayed via electrophoresis in 2 % agarose gels. DNA sequencing was performed using an ABI PRISM 3730XL DNA Analyzer with a BigDye Terminater Kit v.3.1 (Invitrogen, USA) at the Shanghai Invitrogen Biological Technology Company Limited (Beijing, China).

The preliminary identities of the isolates sequenced in this study were obtained by conducting a standard nucleotide BLAST search using the sequences generated from the above primers of the different genomic regions (ITS, LSU, tef1 and rpb2). The BLAST results showed that three isolates were grouped in the family Gnomoniaceae, and five isolates in the genus Synnemasporella. The phylogenetic analyses for the three gnomoniaceous isolates were conducted based on Senanayake et al. (2018), supplemented by sequences of Tenuignomonia styracis and Neognomoniopsis quercina from Crous et al. (2019) and Minoshima et al. (2019). Melanconis marginalis (CBS 109744) in Melanconidaceae was selected as the out-group taxon. All sequences were aligned using MAFFT v. 6 (Katoh and Toh 2010) and edited manually using MEGA v. 6 (Tamura et al. 2013). Phylogenetic analyses were performed using PAUP v. 4.0b10 for maximum parsimony (MP) analysis (Swofford 2003), and PhyML v. 3.0 for Maximum Likelihood (ML) analysis (Guindon et al. 2010).

MP analysis was run using a heuristic search option of 1000 search replicates with random additions of sequences with a tree bisection and reconnection algorithm. Other calculated parsimony scores were tree length (TL), consistency index (CI), retention index (RI), and rescaled consistency (RC). ML analysis was performed using a GTR site substitution model including a gamma-distributed rate heterogeneity and a proportion of invariant sites (Guindon et al. 2010). The branch support was evaluated using a bootstrapping method of 1000 replicates (Hillis and Bull 1993). The matrix was partitioned for the different gene regions. Phylograms were shown using FigTree v. 1.4.3 (Rambaut 2016). Novel sequences generated in the current study were deposited in GenBank (Table 1) and the aligned matrices used for phylogenetic analyses in TreeBASE (accession number: S25047).

The alignment based on the combined sequence dataset (ITS, LSU, tef1, and rpb2) included 42 in-group taxa and one out-group taxon, comprising 3368 characters in the aligned matrix. Of these, 2201 characters were constant, 224 variable characters were parsimony-uninformative and 943 characters were parsimony informative (282 from the ITS-LSU, 280 from tef1, 381 from rpb2). The MP analysis resulted in nine equally most parsimonious trees with identical tree backbone. The best ML tree (lnL = −20604.0384) was compatible with the MP strict consensus tree, except for unsupported clades in Fig. 1. As the trees obtained from different analytical methods were similar, only the ML tree was present in Fig. 1. The phylogram based on the four gene sequence matrix indicated that the three strains from the present study represent a novel genus in Gnomoniaceae.

Figure 1. 

Phylogenetic tree based on an ML analysis of a combined DNA dataset of ITS, LSU, tef1 and rpb2 gene sequences for all genera with DNA data and some species of Gnomoniaceae. Bootstrap values ≥ 50 % for MP and ML analyses are presented at the branches. The scale bar represents the number of changes per site.