Two new species of Neopestalotiopsis from southern China

Abstract Background Pestalotiopsis-like fungi are widely distributed in many plants and include endophytes, pathogens and saprobes. Five strains of Neopestalotiopsis were isolated from diseased leaves of Rhapisexcelsa (Principes, Palmae), Rhododendronsimsii and Rho.championiae (Ericales, Ericaceae) and Erythropalumscandens (Santalales, Olacaceae) in southern China. New information Based on morphology and multi-gene (ITS, tub2, tef1) phylogeny, our five strains of Neopestalotiopsis represent two new species and one extant species. Descriptions, illustrations and notes are also provided for the new species.


Introduction
Sporocadaceae was introduced by Corda (1842) and comprised abundant endophytic, plant pathogenic or saprobic taxa (Liu et al. 2019). A great part of Sporocadaceae species were reported as important plant pathogenic fungi that mainly harm various economic crops, such as tea, blueberry and elephant apple (Fernández et al. 2015, Tsai et al. 2020. Jaklitsch et al. (2016) synonymised Bartaliniaceae, Discosiaceae, Pestalotiopsidaceae and Robillardaceae under Sporocadaceae. Liu et al. (2019) studied the taxonomy of Sporocadaceae and accommodated 30 genera in it.  and  placed Sporocadaceae in Amphisphaeriales and accepted 33 genera.
Amongst surveys of microfungi in southern China, we made five collections of Neopestalotiopsis from four host plants. Based on morphological descriptions and molecular analyses of three gene loci, our strains represent two new species and one known species.

Sample collection and fungi isolation
Diseased leaf samples with fruiting bodies were collected from major botanical gardens in Yunnan, Guangxi and Guizhou Provinces in southern China. After surface disinfection of the diseased tissues , the single-spore method was used for obtaining a pure culture (Senanayake et al. 2020). The isolates were transferred to new potato dextrose agar (PDA) plates to obtain a pure strain.

Morphology study
Cultures growing on potato dextrose agar (PDA) were incubated under moderate temperatures (28ºC) for 2−4 weeks in 12 h daylight. The diameter of cultures was measured after 1 week and the colour was determined with the colour charts of Rayner (1970). The morphological features were noted and recorded following Hu et al. (2007). Microscopic preparations were prepared in lactophenol and over 30 measurements were obtained per structure. Photographs were taken using a compound microscope (Olympus BX53, Japan). The holotype specimens were deposited in the Herbarium of Department of Plant Pathology, Agricultural College, Guizhou University (HGUP). Ex-type cultures were deposited in the Culture Collection at the Department of Plant Pathology, Agriculture College, Guizhou University, China (GUCC).

DNA extraction and PCR amplification
DNA extraction and PCR amplification follow  with some minor changes. A Fungus Genomic DNA Extraction Kit (Biomiga#GD2416, San Diego, California, USA) was used to extract fungal genome DNA. DNA amplification was performed in a 25 µl reaction mixture which contains 2.5 µl 10 × PCR buffer, 1 µl of each primer (10 µM), 1 µl template DNA and 0.25 µl Taq DNA polymerase (Promega, Madison, WI, USA). The ITS rDNA region was amplified using primer pairs ITS4 and ITS5 (White et al. 1990). The partial tub2 gene region was amplified with primer pairs T1 and Bt2b ( Glass andDonaldson 1995, O'Donnell andCigelnik 1997). The tef1 gene fragment was amplified using the primer pairs EF1-728F and EF-2 (O'Donnell et al. 1998, Carbone andKohn 1999). PCR amplification conditions were performed according to the methods described by Norphanphoun et al. (2019). The PCR products were sent to SinoGenoMax company (Beijing, China) which used the fluorescently-labelled Sanger method for sequencing. The resulting DNA sequences were submitted to GenBank (https://www.ncbi.nlm.nih.gov/ genbank/) and their accession numbers were provided in Table 1

Sequence alignment and phylogenetic analyses
The reference sequences were downloaded from GenBank for phylogenetic analyses ( Markov chains were run for 1000000 generations and trees were sampled every 100th generation (printfreq = 100) and 10000 trees were obtained. The last standard deviation of split frequencies was below 0.01. Initial trees were discarded (25% burn-in value) and the remaining trees were used to evaluate posterior probabilities (PP) in the majority rule consensus tree.
PAUP v. 4.0b10 (Swofford 2002) was used to perform Maximum Parsimony (MP) analyses. Trees were inferred by using the heuristic search option with 1,000 random sequence additions and tree bisection and reconnection (TBR) as the branch-swapping algorithm. The maxtrees were set as 5000. Descriptive tree statistics for parsimony (tree length (TL), consistency index (CI), retention index (RI), related consistency index (RC) and homoplasy index (HI)) were calculated.

Description
Disease symptom: Pathogenic causing spots on leaves tip of Rhapis excelsa. Leaf spots shape irregular, brown, slightly sunken on leaves tip. Small brown spots appeared initially and then gradually enlarged, changing to dark brown spots with a yellow border and jagged edge.

Etymology
Latin, rhapidis, refers to the host plant (Rhapis excelsa) from which the fungus was isolated.

Notes
Neopestalotiopsis rhapidis clustered with N. cocoes (MFLUCC 15-0152) with 85% ML support, although without enough MP and BI support. Within comparison of the three gene regions, there were only three character differences in the ITS region, but 27 in the tef1 region. Neopestalotiopsis rhapidis has longer conidia and shorter apical appendages than those of N. cocoes (19-22.5 ×7.5-9.5 µm; 14.9-21 µm)

Description
Disease symptom: Associated with leaf spots of Rhododendron simsii. The leaf spots are small irregular to subcircular shape, brown, slightly sunken spots appear on surface leaves of R. simsii, which scattered on the surface leaves tip and eventually develops into a large lesion. Small off-white spots appeared initially and then gradually enlarged, changing to light brown circular ring spots with a dark brown border.

Notes
In the multi-gene analysis, strain GUCC 21504 formed a distinct clade with a sister strain GUCC 21505, but the node support values were 68/90/-(MP/ML/BI) and these two strains were close to N. protearum (CBS 114178). When comparing the polymorphic nucleotide differences of our two strains, there are 18 base pair differences, seven in ITS, two in tub2 and nine in tef1, but without obvious distinction (higher than 98.5%). Compared with N. protearum and our ex-type strain (GUCC 21504), there were six character differences with N. protearum in the ITS region, three character differences with N. protearum in the tub2 region, but 12 character differences from N. protearum in the tef1 region; thus the DNA base pair differences were mainly in the tef1 gene regions. The morphological differences between our strains and N. protearum were wider conidia (N. protearum: 24.8 ± 1.5 × 8.5 ± 0.6 µm), more apical appendages (N. protearum: 3-5) and shorter basal appendages (N.

Description
Disease symptom: Pathogenic causing spots on leaves of Erythropalum scandens. Leaf spots shape irregular, brown to reddish-brown, slightly sunken spots appear on surface leaves of E. scandens, which scattered on the leaves tip. Small brown spots appeared initially and then gradually enlarged, changing to reddish-brown spots with a yellow border.

Notes
GUCC 21506 and GUCC 21507 with the same nucleotides sequences were related to N. dendrobii  and N. saprophytica (CBS 115452). There were ten character differences with N. dendrobii and 11 character differences with N. saprophytica, but the most differences (nine character differences) between our strains and N. saprophytica were only in the tef1 region. Alternatively, collection differed to N. dendrobii in having more apical appendages ( N. dendrobii: 2-3) and much longer apical appendages (N. dendrobii: 6 ± 0.9 µm) (Ma et al. 2019). Morphological
In the phylogenetic analyses, GUCC 21501 was sister to N. cocoes (MFLUCC 15-0152 ), but only with a 85% ML bootstrap support. GUCC 21504 and GUCC 21505 formed an independent clade with MP and ML (68/90) supports and were close to N. protearum (CBS 111506 ). GUCC 21506 and GUCC 21507 clustered with moderate and high supports (65/99/1: MP/ML/BI) and kept a very close relationship with N. saprophytica (CBS 115452) by credible statistic support (100/67/1: MP/ML/BI). DNA sequence differences between our strains and related species are listed in Table 2. The phylogram generated from MP analysis, based on combined ITS, tub2 and tef1 sequence data of Neopestalotiopsis. The tree was rooted with Pestalotiopsis diversiseta (MFLUCC 12-0287) and P. trachicarpicola (OP068). Maximum Parsimony and Maximum Likelihood bootstrap values ≥ 50%, Bayesian posterior probabilities ≥ 0.90 (MPBS/MLBS/PPBY) were given at the nodes. Our strains in this study were in green. Ex-type strains were marked by T. Discussion Hu et al. (2007) believed that pestalotiopsis-like fungi had different phenotypes in conidial morphology.  summarised some stable characteristics for determining pestaloids, such as the length and width of conidia, length of the apical appendages, presence or absence of knobbed apices and the position of the apical appendage attached to the conidial body. However, as these characteristics were often similar or overlapped, sequence data are crucial for the identification of pestalotioid, and as well as for the introduction of new species (Norphanphoun et al. 2019).
In this study, we describe two new species as Neopestalotiopsis rhapidis and N. rhododendri. The species were distinct from extant Neopestalotiopsis species, based on morphological and phylogenetic analyses. However, the statistical support of main nodes for the genus were very low (Fig. 4). The reason might be that the reference sequences we used were short, including the short tef1 and tub2 sequences (Ran et al. 2017). Longer sequences with more informative data are needed to solve this problem. Furthermore, our study also found that the evolutionary relationships amongst species of Neopestalotiopsis are unstable , Jiang et al. 2018, Kumar et al. 2019, Tsai et al. 2020. Therefore, other genes are needed to distinguish the inter-species T T T Several indicators could be used in the classification of Neopestalotiopsis in this study, such as the size of conidia and the number and length of appendages (Maha rachchikumbura et al. 2014, Freitas et al. 2019, Kumar et al. 2019). The differences in the colour of three median cells and the length of other cells, however, lacked significant variation to clearly distinguish the species of Neopestalotiopsis. Therefore, as the morphological identification alone cannot accurately identify the fungi of the genus Neopestalotiopsis, it must be combined with the phylogenetic tree (Liu et al. 2019, Norphanphoun et al. 2019, Tsai et al. 2020, Jiang et al. 2021).