Re-collection of Dermeaprunus in China, with a description of D.chinensis sp. nov.

Abstract Dermea was protected against its synonym, Foveostroma, due to its well-circumscribed generic concept and more frequent use. We describe and illustrate Dermeachinensissp. nov. based on its morphological characteristics and a molecular analysis of the internal transcribed spacer (ITS) and large subunit (LSU) sequence data. Dermeachinensis is isolated from Betulaalbosinensis with sexual and asexual morphs and can be distinguished from D.molliuscula on Betula trees by its aseptate and wider ascospores. The connection between the two morphs is proved based on sequence data. Here, we describe the asexual morph of D.pruni for the first time based on morphological and molecular data from the same host and country of origin, and compare it with other species of Prunus.

Dermea species are generally considered highly host-specific (Groves 1946(Groves , 1951. The plant genus Prunus is the major host for Dermea, with D. cerasi, D. padi, D. prunastri, and D. pruni described from them (Groves 1946(Groves , 1951. However, ascospores in D. pruni are larger than those from the other three species (Groves 1951). Dermea cerasi, D. padi, and D. prunastri can be easily distinguished by the macroconidial and microconidial dimensions (Groves 1946). Among these four species, D. cerasi, D. padi, and D. prunastri were recognized based on both sexual and asexual fruiting bodies (Groves 1946), but D. pruni was proposed only with a sexual morph based on a specimen (Teng #3352, preserved in the herbarium of the University of Michigan) collected from China (Groves 1951). Hence, the re-collection of D. pruni specimens aiming for an asexual morph from the original host and country seems meaningful. Additionally, few sequence data are available for most Dermea species, and considering that the host associations may be incorrect and that many geographical areas are still insufficiently studied, the synonymies and actual numbers of Dermea species are still unclear.
Dermea species were considered pathogenic to their hosts (Groves 1951;Abeln et al. 2000). For example, D. abietinum (syn. D. balsamea) caused hemlock dieback (Dodge 1932) and D. prunastri was considered the cause of greengage plums die-back (Dowson 1913). However, members of Dermea have not been recently reported to cause serious plant diseases.
During our fungal collection surveys conducted in China, we collected several Dermea specimens from two species of tree, Betula albosinensis and Prunus cerasifera f. atropurpurea. We identified fungi species using both morphological and molecular approaches; as a result, a novel species and the asexual morph of D. pruni are described herein for the first time.

Sample collections and fungal isolates
Fresh specimens of Dermea were collected from tree barks during our fungal collection trip in China. We obtained single ascospore and conidia isolates by removing a mucoid spore mass from apothecia or conidiomata and spreading the suspension on the surface of 2% malt extract agar (MEA; 20 g malt extract, 20 g agar, 1 L water). After inoculation, agar plates were incubated at 25 °C to induce germination of spores. Single germinating spores were then transferred to clean plates under a dissecting microscope with a sterile needle. Specimens and isolates were deposited in the Museum of Beijing Forestry University (BJFC). Axenic cultures are maintained in the China Forestry Culture Collection Center (CFCC).

Morphological analysis
Species identification was based on the morphological characters of apothecia and conidiomata produced on natural substrates. Cross-sections were prepared manually using a double-edged blade under a Leica stereomicroscope (M205 FA). Photomicrographs were captured with a Nikon Eclipse 80i microscope equipped with a Nikon digital sight DS-Ri2 high-definition colour camera, using differential interference contrast (DIC) illumination and the Nikon software, NIS-Elements D Package 3.00. Measurements of ascospores and conidia are reported as the maximum and minimum in parentheses and the range representing the mean ± standard deviation of the number of measurements is given in parentheses. Cultural characteristics of isolates incubated on MEA in the dark at 25 °C were recorded.

DNA extraction, PCR amplification and sequencing
Genomic DNA was extracted from axenic living cultures on MEA with cellophane using a modified CTAB method (Doyle and Doyle 1990). The internal transcribed spacer (ITS) region was amplified with primers ITS1 and ITS4 (White et al. 1990), and the large subunit (LSU) region with the primers LR0R and LR5 (Vilgalys and Hester 1990). Amplification of ITS and LSU were accomplished by an initial step of 2 min at 95 °C, followed by 35 cycles of 30 s at 95 °C, 30 s at 51 °C, and 40 s at 72 °C, with a final extension of 10 min at 72 °C. DNA sequencing was performed on an ABI PRISM 3730XL DNA Analyzer using BigDye Terminater Kit 3.1 (Invitrogen) at the Shanghai Invitrogen Biological Technology Company Limited (Beijing, China).

Phylogenetic analyses
Sequences from this study and reference sequences obtained from GenBank (Table 1) were aligned and edited manually using MEGA6 (Tamura et al. 2013). The alignments were concatenated for phylogenetic analyses. Maximum parsimony (MP) analyses were conducted with PAUP 4.0b10 (Swofford 2003), using 1000 heuristic search replicates with random-additions of sequences along with the tree bisection and reconnection (TBR) branch swapping algorithm (MULTREES option in effect, steepest descent option not in effect). All molecular characters were unordered and given equal weight; analyses were performed with gaps treated as missing data; the COLLAPSE command was set to minbrlen, maxtrees were set to 5000. All equally parsimonious trees found were saved in the MP analyses. Other calculated parsimony scores were tree length (TL), consistency index (CI), retention index (RI), and rescaled consistency (RC). MP bootstrap analyses with 1000 replicates were performed in the same manner, with 10 rounds of heuristic search replicates with random addition of sequences and subsequent TBR branch swapping during each bootstrap replicate. ML analyses were conducted using RAxML (Stamatakis 2006) as implemented in raxmlGUI 1.3 (Silvestro and Michalak 2012), using the ML + rapid bootstrap setting and the GTRGAMMA substitution model with 1000 bootstrap replicates. Taxonomic novelties were deposited in MycoBank.

Phylogenetic analyses
The alignment based on the combined sequence dataset (ITS and LSU) contained 1431 characters. Of these, 1136 characters were constant, 103 variable characters were parsi- mony-uninformative, and 192 parsimony informative. The MP analyses resulted in five equally most parsimonious trees, with the first tree (TL = 601, CI = 0.647, RI = 0.807, RC = 0.522), which is shown in Figure 1. Tree topologies of the best tree revealed by the ML analyses was identical to those of the MP tree (not shown). The two species from this study appeared in two distinct clades, and three strains of Dermea chinensis from the Betula albosinensis cluster in a well-supported clade (MP/ML = 100/100) (Fig. 1). Taxonomy
Culture characters. On MEA at 25 °C colonies grow slowly, reaching 50 mm diameter within 50 d, at first pale yellow, gradually becoming dark brown with scanty aerial mycelium.
Habitat and host range. On dying stems and branches of Prunus cerasifera f. atropurpurea.
Specimens Notes. Dermea pruni was proposed based on a specimen collected from Prunus branches in Sichuan province, China. However, no living culture or DNA data were available (Groves 1951). In addition, the asexual morph was not included in the original description (Groves 1951). During our fungal collection trip in China, two Dermea specimens were accidentally discovered on a common road tree, Prunus cerasifera f. atropurpurea in Shaanxi province, which borders Sichuan province, the original collection province of the holotype. Asexual fruiting bodies were observed on the whole trees, from stems to branches. However, no sexual morph was found, even though we investigated all Prunus trees along the road. Conidial size was compared among our collections, D. cerasi, D. padi, and D. prunastri, which can distinguish them ( Table 2). Considering that our collections and the type specimen (Teng #3352, preserved in the herbarium of the University of Michigan) of D. pruni were collected from the same hosts and from nearby regions (Groves 1951), our specimens were identified and treated here as D. pruni. However, more detailed taxonomic studies are needed, including DNA extraction from the holotype of D. pruni to compare ITS sequences of our collections and the holotype.   (Groves 1946(Groves , 1951. These four species can be obviously distinguished by both morphological and molecular approaches. We update the asexual morph and molecular data of D. pruni. The genus Pezicula is a phylogenetically close to Dermea species and has recently been confirmed based on an ITS-28S-16S rDNA analysis (Mehrabi et al. 2018). However, Pezicula is characterized by typically bright-coloured, yellowish to ochraceous, more fleshy-waxy apothecia, broader and more clavate asci, and more broadly ellipsoid to oblong-ellipsoid or ovoid ascospores (Grove 1946). Our phylogenenetic analysis of Dermea and related genera based on the combined ITS and LSU sequence data (Fig. 1) showed that Pezicula is well-supported as a separate clade with high values (MP/ML = 96/98). Dermea was thought to be a monophyletic group (Abeln et al. 2000), but Dermea was not well-supported, as D. persica was included in the analysis (Mehrabi et al. 2018). We added additional DNA sequence data in our study (Fig. 1), which indicates that Dermea is not monophyletic.
Species of Dermea are well-circumscribed by morphological characteristics. However, only 10 species (Table 1) are currently characterized by molecular data, and most species remain unconfirmed by phylogenetic examination. Hence, DNA data from type or ex-strains and newly obtained collections are essential in subsequent taxonomic work.