Integrative Taxonomy of Novel Diaporthe Species Associated with Medicinal Plants in Thailand

During our investigations of the microfungi on medicinal plants in Thailand, five isolates of Diaporthe were obtained. These isolates were identified and described using a multiproxy approach, viz. morphology, cultural characteristics, host association, the multiloci phylogeny of ITS, tef1-α, tub2, cal, and his3, and DNA comparisons. Five new species, Diaporthe afzeliae, D. bombacis, D. careyae, D. globoostiolata, and D. samaneae, are introduced as saprobes from the plant hosts, viz. Afzelia xylocarpa, Bombax ceiba, Careya sphaerica, a member of Fagaceae, and Samanea saman. Interestingly, this is the first report of Diaporthe species on these plants, except on the Fagaceae member. The morphological comparison, updated molecular phylogeny, and pairwise homoplasy index (PHI) analysis strongly support the establishment of novel species. Our phylogeny also revealed the close relationship between D. zhaoqingensis and D. chiangmaiensis; however, the evidence from the PHI test and DNA comparison indicated that they are distinct species. These findings improve the existing knowledge of taxonomy and host diversity of Diaporthe species as well as highlight the untapped potential of these medicinal plants for searching for new fungi.


Introduction
Medicinal plants are essential for sustaining human health and livelihoods according to their ethnobotanical uses and therapeutic purposes [1,2]. They have also contributed to maintaining biodiversity in forest ecosystems and supporting natural recreation in urban ecosystems [1,2]. Fungi are usually encountered in medicinal plants, where they can affect their hosts in both beneficial and harmful manners [2][3][4]. As pathogens, they impair plant health and productivity [4]; whereas, as endophytes, they promote plant growth and produce a diverse array of secondary metabolites, which have been exploited for the development of new drugs and pharmaceutical products [2,3]. Thus, studies of fungi associated with medicinal plants represent a significant repository for the estimation of fungal diversity, the discovery of novel fungi and fungal-plant interactions, as well as the bioprospecting of new bioactive compounds and their biotechnological applications [5][6][7][8][9][10][11][12].
Taxonomic studies of Diaporthe revealed a variety of medicinal plants as their hosts [38]. However, most of these studies have been conducted in temperate zones (i.e., [15][16][17]21,24,26,28]). Knowledge of Diaporthe associated with medicinal plants in the tropics is still limited [31,32]. Therefore, this study aims to identify and describe isolates of Diaporthe associated with several medicinal plants in Thailand using both morphological and molecular analyses. To better illustrate the placements of the five new species, their morphological descriptions, micrographs, and updated phylogenetic trees are presented and discussed.

Sample Collection and Morphological Examination
Fresh fungal specimens were collected from the dead leaves and woody twigs of various medicinal plants in urban parks and forest areas in the Chiang Mai and Tak provinces of Thailand in 2019 and 2022. Collected samples were investigated for macroand micro-morphological structures using a Nikon SMZ800N stereo microscope (Nikon Instruments Inc., Melville, NY, USA) and photomicrographed with a Nikon Eclipse Ni compound microscope attached to a Nikon DS-Ri2 camera system (Nikon Instruments Inc., Melville, NY, USA). The measurement of each structure (i.e., conidiomata, conidiomatal walls, conidiophores, conidiogenous cells, and conidia) was taken using the Tarosoft (R) Image Frame Work program. All figures were modified using Adobe Photoshop CS6 Extended version 10.0 software (Adobe Systems, San Jose, CA, USA).

Fungal Isolation and Preservation
Pure cultures were obtained from single spore isolation on 2% water agar (WA), and germinated conidia were aseptically transferred to potato dextrose agar (PDA) [39]. Fungal cultures were incubated at 25 • C for four to six weeks and then examined for colony morphology and spore production. Herbarium material and pure culture of Diaporthe globoostiolata were deposited in the herbarium of Mae

Phylogenetic Analyses
The sequences obtained in this study were submitted through a BLASTn search in GenBank (www.ncbi.nlm.nih.gov/blast/, assessed on 1 March 2023) to determine the most similar taxa. The initial phylogenetic analysis was conducted based on the ITS sequence dataset from Norphanphoun et al. [32] to identify the placement of our isolates within species complexes. The newly generated sequences and their related sequences were then selected for the concatenated ITS, tef1-α, tub2, cal, and his3 sequence dataset based on the BLASTn search results and updated literature [18,22,32,[46][47][48] (Table 1). The alignment of a single locus dataset was performed using MAFFT v.7 (http://mafft.cbrc.jp/alignment/ server/index.html, assessed on 1 March 2023) [49] and the ambiguous sites were manually adjusted using BioEdit 7.1.3.0 [50]. The phylogenetic trees of single locus and combined datasets were analyzed using maximum likelihood (ML) and Bayesian inference (BI) criteria. Tree topologies from single locus analyses were also compared and no conflicts were found.
Taxa Names Culture Accession No.
GenBank Accession No.    [52][53][54] in the CIPRES Science Platform V3.3 (https://www. phylo.org/portal2/home.action, assessed on 1 March 2023) [55]. The GTRGAMMA model of the bootstrapping phase with 1000 bootstrap iterations was set as the parameter for ML analysis [51]. The best nucleotide substitution model was determined using MrModeltest v.2.3 [56], and GTR + I + G was selected as the best-fitting model for the ITS, tef1-α, tub2, cal, and his3 datasets. For BI analysis, six simultaneous Markov chains were set to run 10,000,000 generations with a sampling frequency of 100 generations. The burn-in phase was set as 0.25, and the posterior probabilities (PP) were evaluated from the remaining trees. The phylogenetic trees resulting from the ML and BI analyses were visualized by the FigTree v1.4.0 program [57] and adjusted using Adobe Photoshop CS6 software (Adobe Systems, San Jose, CA, the USA). Novel obtained sequences were registered for GenBank accession numbers.

Genealogical Concordance Phylogenetic Species Recognition Analysis
The recombination level between new species and their most closely related taxa was examined using the Genealogical Concordance Phylogenetic Species Recognition (GCPSR) model [58,59]. A pairwise homoplasy index (PHI) test was implemented by SplitsTree4 using the LogDet transformation and split decomposition options [60,61]. A PHI test result (Φw) above 0.05 indicated no significant recombination in the dataset. In addition, split graphs were generated for visualization of the relationship between closely related species.

Genealogical Concordance Phylogenetic Species Recognition Analysis
In the PHI analysis, there was no evidence of significant recombination (Φw > 0.05) between each new species (Diaporthe afzeliae, D. bombacis, D. globoostiolata, and D. samaneae) and their closely related taxa in the combined ITS, tef1-α, tub2, cal, and his3 sequence dataset (Figure 2a-d). The results of PHI analysis also revealed no significant recombination (Φw > 0.05) between D. zhaoqingensis and D. chiangmaiensis (Figure 2e). This evidence confirms that they are distinct species.
Culture characteristics: Colonies on PDA reached 5 cm diam. after 10 days at 25 • C, effuse, fluffy, lobate at the margin, originally white, becoming yellowish to pale brown mycelium with age, yellowish to pale brown in reverse, with numerous black dots developing as the fruiting bodies (conidial production not seen).
Culture characteristics: Colonies on PDA reached 9 cm diam. after 10 days at 25 • C, effuse, sparse hyphae, filiform margin, originally white, becoming pale yellowish mycelium with age, yellowish to pale brown in reverse, with numerous black dots developing as the fruiting bodies (conidial production not seen).

Discussion
This study describes five novel species of Diaporthe in Thailand. Aside from the phenotypic traits, phylogenetic and PHI analyses based on the combined sequence datasets of ITS, tef1-α, tub2, cal, and his3 were successfully applied to justify the novel species. In particular, tub2, cal, and his3 have a high discrimination power for distinguishing species in Diaporthe, and this is consistent with the results from other studies [15,18,22,[35][36][37].
Our study also gains better insight into the phylogenetic relationships within Diaporthe, especially in the D. arecae species complex. Diaporthe zhaoqingensis and D. chiangmaiensis were clustered together in the same clade (98% ML, 1.00 PP) and not so well separated ( Figure 1). Therefore, we compared the base pair differences between the type strains of D. zhaoqingensis ZHKUCC 22-0056 and D. chiangmaiensis MFLUCC 18-0544. There are 1.38% base pair differences in ITS (7/508 bp) between the ex-type of both strains. In the tef1-α gene region, there are 0.33% base pair differences (1/300 bp) between the type strains of D. chiangmaiensis MFLUCC 18-0544 and D. zhaoqingensis ZHKUCC 22-0057. There are 4.94% base pair differences (19/385 bp) in the tub2 gene region, between D. chiangmaiensis MFLUCC 21-0212 and the type strain of D. zhaoqingensis ZHKUCC 22-0056. However, some genes from the type strains were not available to compare. The PHI test result also showed that D. zhaoqingensis and D. chiangmaiensis were not conspecific, indicating that they are different species (Figure 2e). Diaporthe zhaoqingensis was isolated as an endophyte on Morinda officinalis [18], and D. chiangmaiensis was isolated from Magnolia lilifera as an endophyte and saprobe [47]. However, the morphological characteristics of these two species could not be compared as only gamma conidia were observed in D. zhaoqingensis while alpha conidia were observed in D. chiangmaiensis [18,47]. Therefore, more sequence data such as the tub2, cal, and his3 of the type strain of D. chiangmaiensis are needed to resolve their taxonomic placements and confirm whether they are distinct species.
Furthermore, the new species, D. careyae, was shown to be distinct from other Diaporthe species based on its morphology and phylogeny. The conidia of D. careyae were 0-1(-2) septate, whereas aseptate conidia were a typical characteristic of Diaporthe. The septation of conidia has been reported in some Diaporthe species (e.g., D. foeniculina and D. saccarata) [17,68], however, their phylogenetic placements were not closely related to D. careyae. It is noteworthy that there are some singleton species that were not grouped into any species complex, and their taxonomic positions remain unclear [32]. In addition, most species of Diaporthe lack sequence data and have incomplete morphological descriptions [31,32]; therefore, further extensive sampling is needed in order to unravel the taxonomic circumscription of this genus.
The newly introduced species of Diaporthe were associated with different medicinal plants, comprising D. afzeliae on Afzelia xylocarpa, D. bombacis on Bombax ceiba, D. careyae on Careya sphaerica, and D. samaneae on Samanea saman. These plant species have been used as traditional medicines in tropical countries, including Thailand, and have been reported on concerning their various phytochemicals and pharmacological activities [69][70][71][72][73][74][75]. To the best of our knowledge, none of the Diaporthe species have been isolated from these host genera, making this the first report of such a host association [38]. Moreover, a new species, D. globoostiolata, was found on a member of Fagaceae. Some plant genera in Fagaceae, such as Castanopsis, Quercus, and Lithocarpus, have also been reported on regarding their medicinal usage and pharmacological properties [76][77][78][79]. Furthermore, more than 30 Diaporthe species have been recorded from the host family Fagaceae [38]. This study reflects the high genetic diversity and phenotypic variation within Diaporthe and expands our understanding of the diversity and host relationships of the Diaporthe species associated with medicinal plants in tropical regions. However, future studies are necessary to investigate the disease symptoms and evaluate the pathogenicity of these Diaporthe isolates as they are important for tree health assessments and management.  Data Availability Statement: All sequences generated in this study were submitted to GenBank (https://www.ncbi.nlm.nih.gov, accessed on 1 April 2023).