Morphology and Multi-Gene Phylogeny Reveal Pestalotiopsis pinicola sp. nov. and a New Host Record of Cladosporium anthropophilum from Edible Pine (Pinus armandii) Seeds in Yunnan Province, China

This study contributes new knowledge on the diversity of conidial fungi in edible pine (Pinus armandii) seeds found in Yunnan Province, China and emphasizes the importance of edible seed products to ensure food safety standards. We isolated two fungal species, one on the pine seed coat and the other on the endosperm of the pine seed. The two fungal species were identified as Pestalotiopsis pinicola sp. nov. and a new host record Cladosporium anthropophilum. Characteristic morphological features of Pestalotiopsis pinicola were used alongside results from multi-gene phylogenetic analysis to distinguish it from currently known species within the genus. Cladosporium anthropophilum was identified as a new host record based on morphological features and phylogenetic analysis. In addition, detailed descriptions, scanned electron microscopy morphology, illustrations, and phylogenetic trees are provided to show the placement of these species.


Introduction
Chinese white pine (Pinus armandii), one of the endemic conifer species of East Asia, is known throughout China, and particularly Yunnan Province, for its substantial ecological and economic value [1,2]. Pinus armandii seeds are suitable for use as a culinary ingredient after roasting, because the fatty acid profile of the seeds has a higher level of taxoleic acid and lower levels of octadecenoic acids compared to other species in Pinus [3,4].
Seeds are colonized by various types of fungi including fungal pathogens [5]. Several fungal species exist in seeds in the forms of spores and mycelium and can subsist for long periods of time on the seed coat and in the inner areas [6]. In general, fungi that are present within seeds are more harmful than those that merely contaminate the outer seed coat [6]. Common fungi genera that have been reported as associated with various seeds are Aspergillus, Mucor, Penicillium, Pestalotiopsis, Rhizopus,

Sample Collection and Specimen Examination
Fresh fungal structures (mycelia and spore masses) were directly isolated in potato dextrose agar (PDA) from seed coats and endosperms of Pinus armandii seeds using aseptic techniques, and the PDA plates were incubated at room temperature. Pine seeds were obtained outside Kunming, Yunnan Province, China (Figure 1). The seeds were then carefully analyzed. Morphological structures of the fungi were examined under a stereo microscope. Scanning electron microscopy (SEM) micrographs were obtained under a ZEISS GeminiSEM and ZEISS Sigma 300 apparatus, following the methods described by Figueras and Guarro [48]. To observe the fungal structures, sporulated cultures were mounted on water. Microscopic fungal structures were observed under a compound microscope and photographs were captured with a digital camera fitted on to the microscope. All microscopic structures of fungi were measured by the Tarosoft Image Framework program v.0.9.0.7., and Adobe Photoshop CS3 Extended version 10.0 (Adobe Systems, USA) was used to process and edit the images used in the figures.

Isolation
The PDA medium was used for culturing the isolated fungi. Spore masses from the seed coat and mycelia from the endosperm were aseptically transferred to PDA plates (two isolates of each species). The pure culture plates were incubated at room temperature (20-25 • C) for 14-21 days, and the fungal colonies were carefully observed and described. The herbarium specimens of the fungi were dehydrated using silica gel and deposited in the Mae Fah Luang University Herbarium. The pure cultures were deposited in the Kunming Institute of Botany Culture Collection (KMUCC). Index Fungorum (IF) and Facesoffungi (FoF) numbers were obtained as described by Index Fungorum [49] and Jayasiri et al. [50].

Isolation
The PDA medium was used for culturing the isolated fungi. Spore masses from the seed coat and mycelia from the endosperm were aseptically transferred to PDA plates (two isolates of each species). The pure culture plates were incubated at room temperature (20-25 °C) for 14-21 days, and the fungal colonies were carefully observed and described. The herbarium specimens of the fungi were dehydrated using silica gel and deposited in the Mae Fah Luang University Herbarium. The pure cultures were deposited in the Kunming Institute of Botany Culture Collection (KMUCC). Index Fungorum (IF) and Facesoffungi (FoF) numbers were obtained as described by Index Fungorum [49] and Jayasiri et al. [50].

DNA Extraction, PCR Amplification, and DNA Sequencing
The mycelia of the cultures grown on PDA at room temperature for 4 weeks were used for DNA extraction. The fungal mycelia were scraped off with a sterile scalpel and transferred to 1.5 ml microcentrifuge tubes under aseptic conditions and kept at −20 ℃ to avoid contaminations until use. The Biospin Fungal Genomic DNA Extraction Kit (BioFlux, China) was used to perform DNA extraction

DNA Extraction, PCR Amplification, and DNA Sequencing
The mycelia of the cultures grown on PDA at room temperature for 4 weeks were used for DNA extraction. The fungal mycelia were scraped off with a sterile scalpel and transferred to 1.5 mL micro-centrifuge tubes under aseptic conditions and kept at −20 • C to avoid contaminations until use. The Biospin Fungal Genomic DNA Extraction Kit (BioFlux, China) was used to perform DNA extraction from the fungal cultures, following the manufacturer's protocols. To amplify partial gene regions of the 5.8S rRNA gene in the internal transcribed spacer (ITS), translation elongation factor 1-alpha gene (TEF1), actin gene (ACT), and beta-tubulin gene (TUB2), polymerase chain reaction (PCR) was used. The PCR conditions and primers were set under standard conditions as shown in Table 1. The total volume of PCR mixtures for amplifications was set as described in Tibpromma et al. [18]. Purification and sequencing of PCR products were done by Sangon Biotech Co., Shanghai, China.

Phylogenetic Analyses
The ITS and TEF1 sequence data produced in this study were used in BLAST searches in the GenBank database (www.http://blast.ncbi.nlm.nih.gov/) to determine their most probable closely related taxa. The sequence data generated in this study were analyzed with closely related taxa retrieved from GenBank based on BLAST searches and recent publications [9,16,18,55,56]. Single gene sequence datasets were aligned using the MAFFT v.7.215 website [57] and manually edited in BioEdit v.7.0 when necessary [58]. Single sequence alignment datasets were combined using BioEdit v.7.2.5 [58]. The alignment of combined datasets in FASTA format was converted to PHYLIP and NEXUS formats using the Alignment Transformation Environment (ALTER) website [59]. Phylogenetic trees were run in randomized accelerated maximum likelihood (RAxML) and Bayesian posterior probabilities (BYPPs). The maximum likelihood (ML) analysis was performed via the CIPRES Science Gateway [60] using the RAxML-HPC BlackBox (8.2.4) section [61,62] with the general time reversible model (GTR) using a discrete gamma distribution as the evolutionary model. To carry out Bayesian analysis, the model of evolution was estimated using MrModeltest 2.2 [63] with HKY+I+G (for the Pestalotiopsis dataset) and GTR+I+G (for the Cladosporium dataset) as nucleotide substitution models selected for combined datasets. Posterior probabilities (PPs) [64] were determined by Markov chain Monte Carlo sampling (MCMC) in MrBayes v.3.0b4 [65]. The parameters were set as six simultaneous Markov chains ran for 5,000,000 generations and sampling every 100th generation for a total of 50,000 trees [66]. The first trees representing the burn-in phase of the analysis (20%) were discarded and the remaining (post-burn) trees were used for calculating PPs in the majority rule consensus tree (the critical value for the topological convergence diagnostic values reached 0.01) [67,68].
The phylograms were figured in FigTree v.1.4 [69] and reorganized using Microsoft Office PowerPoint 2007 and Adobe illustrator CS3 (Adobe Systems Inc., USA). The sequences generated in this study were submitted to GenBank (Tables 2 and 3).

Phylogenetic Analysis of Combined Sequence Data
The combined dataset of genera Cladosporium and Pestalotiopsis were analyzed using maximum likelihood and Bayesian analyses (Figure 2; Figure 4). Both the ML and BYPP trees showed similar results in topology and no significant differences were seen (data not presented).
In the Cladosporium tree (Figure 2), the final alignments contained 104 strains with 1484 characters, including 594 characters for TEF1, 306 characters for ACT, and 584 characters for ITS. Cercospora beticola (CBS 116456) was used as an outgroup taxon. The tree topology of the ML analysis was similar to the BYPP. The best scoring RAxML tree with a final likelihood value of −14,457.527098 is presented. The matrix had 681 distinct alignment patterns with 30.20% undetermined characters or gaps. Estimated base frequencies were as follows: A = 0.228336, C = 0.290122, G = 0.251877, T = 0.229664; substitution rates AC = 1.724785, AG = 2.866615, AT = 1.692026, CG = 1.001444, CT = 5.300862, GT = 1.000000; gamma distribution shape parameter a = 0.312597. The phylogram of the genus Cladosporium based on a combined dataset showed that our strains grouped together with Cladosporium anthropophilum clade with relatively high bootstrap supports ( Figure 2).
In the Pestalotiopsis tree (Figure 4), the final alignments contained 81 strains with 1562 characters, including 549 characters for TEF1, 440 characters for TUB2, and 573 characters for ITS. Neopestalotiopsis formicarum (CBS 362.72) and N. clavispora (CBS 447.73) were used as outgroup taxa. The tree topology of the ML analysis was similar to the BYPP. The best scoring RAxML tree with a final likelihood value of −11413.131729 is presented. The matrix had 696 distinct alignment patterns, with 12.40% undetermined characters or gaps. Estimated base frequencies were as follows: A = 0.235816, C = 0.293897, G = 0.211788, T = 0.258500; substitution rates AC = 1.049115, AG = 3.327441, AT = 1.067008, CG = 0.861291, CT = 3.485808, GT = 1.000000; gamma distribution shape parameter a = 0.276615. The Pestalotiopsis phylogram, based on a combined dataset, showed that our new species, Pestalotiopsis pinicola, was well separated from P. rosea with relatively high bootstrap supports (100% ML/1 BYPP, Figure 4). Therefore, we propose Pestalotiopsis pinicola as a distinct new species and Cladosporium anthropophilum as a previously known species. Phylogram generated from RAxML analysis based on combined TEF1, ACT and ITS sequence data of the genus Cladosporium. Related sequences were obtained from Sandoval-Denis et al. [16] and Bensch et al. [56]. Bootstrap support values for ML equal to or greater than 60% and BYPP Figure 2. Phylogram generated from RAxML analysis based on combined TEF1, ACT and ITS sequence data of the genus Cladosporium. Related sequences were obtained from Sandoval-Denis et al. [16] and Bensch et al. [56]. Bootstrap support values for ML equal to or greater than 60% and BYPP from MCMC analyses equal to or greater than 0.95 are given above/below the nodes. The ex-type strains are indicated in bold type. Newly generated sequences are indicated in red with two asterisks.
Note that Cladosporium anthropophilum was established by Sandoval-Denis et al. [16] which belongs to the C. cladosporioides species complex. Cladosporium anthropophilum is probably known as a common saprobic fungus and also represents a clinically relevant fungus [16,70]. In this study, we found a strain of C. anthropophilum from a seed coat of Pinus armandii which was confirmed based on morphology and multi-gene analysis (Figures 2 and 3). The morphology of our strain was similar to the C. anthropophilum described by Sandoval-Denis et al. [16]. In addition, this is the first report of C. anthropophilum from P. armandii (Figure 4).
Culture characteristics: Colonies on PDA reaching 9 cm in diameter after 2 weeks at room temperature, edge undulate with curled, whitish, aerial mycelium on surface, spore masses form after 1 month, black spore masses; reverse of culture yellow-white to yellow-orange with black dots.
Note that Pestalotiopsis pinicola is introduced based on morphological and phylogenetic data. In the phylogenetic analysis, our new species cluster with P. rosea Maharachch. and K.D. Hyde [40] but are well separated with high support (100% ML/1 BYPP, Figure 4). In addition, base pair differences of our new taxa with closest taxa were checked based on the recommendations of Jeewon and Hyde [71]; our isolate differs from P. rosea (MFLUCC12-0258 and CL0441) with five ITS base pairs (2.65%), four TUB base pairs (1.64%), and ten RPB2 base pairs (4.65%). In addition, the culture of P. rosea was seen as a reddish colony [40], while our new species produces a whitish colony. In

Discussion
In this paper, we describe a novel taxon belonging to Pestalotiopsis and a new host record of Cladosporium isolated from seeds of Pinus armandii obtained from Yunnan Province, China. Mature agar colonies sporulated in cultures with masses of conidia.
We isolated a new Pestalotiopsis species from mycelia-covered endosperms of pine seeds. Past research has yielded new species from Pestalotiopsis with similar origins; for example, several endophytic Pestalotiopsis species were isolated from the bark and needles of Pinus armandii in China [38]. Furthermore, Pestalotiopsis brassicae and P. oryzae were isolated from seeds from Oryza sativa and Brassica napus [42]. Several have often been isolated as endophytes and many pathogens or endophytes may persist as saprobes, which mean Pestalotiopsis species are able to switch life-modes [42]. The present study illustrates a novel species of Pestalotiopsis as Pestalotiopsis pinicola, taking both morphology and phylogeny into consideration (Figures 4 and 5). The phylogenetic tree construction of the DNA sequences of single and combined genes provides confirmation with high bootstrap support that P. pinicola is a characteristic new species separate from other known species of the genus (Figure 4). Moreover, this genus is known as one of the fungal groups that can produce a wide range of chemically novel secondary metabolites and mycotoxins; for example, pestaloside exhibiting significant antifungal properties was produced by P. microspora, obtained from Torreya taxifolia [42,[72][73][74]. There is, consequently, a potential health threat in the sale of these seeds as an edible foodstuff. Follow-up research investigating the potential toxins produced by P. pinicola should be conducted to clarify this issue. We conclude that fungi live inside seeds as endophytes and then switch life-modes to saprobes or weak pathogens when conditions become unfavorable. In the future, knowledge about pestalotioid fungi associated with seeds will help provide a basis for developing proper management of these pathogens.
We found another species, Cladosporium anthropophilum, growing on pine seed coats. The etymology of this species comes from Greek which refers to the sample's source which was isolated from a human clinical sample [16]. This species can be found in human clinical samples, indoor air, food and plant materials, such as seeds or leaves, and it is also a common saprobic fungus [56]. In addition, this species is known as the second-most prevalent species from clinical environments from the US after C. halotolerans, and it also has been isolated quite frequently from indoor environments [16,69]. However, we continue to lack information about the chemistry or secondary metabolites of this species along with the potential serious health effects associated with long-term exposure to a large amount of Cladosporium anthropophilum.
The present study illustrates two species of Pestalotiopsis and Cladosporium based on both morphology and phylogeny. These two species of fungi were isolated from pine seeds from Yunnan Province, China. The fungal mycelia in the seeds were observed after the seeds were broken open to eat, and these seeds can be found in many food markets around Yunnan Province. We recommend that consumers should carefully check seed products before purchase and consumption, as these fungi may cause adverse health effects in the long term. Therefore, to address this health concern, in the future we will focus our research on the secondary metabolites and mycotoxins of Cladosporium anthropophilum and Pestalotiopsis pinicola.