Ribosomal and Protein Gene Phylogeny Reveals Novel Saprobic Fungal Species From Juglans regia and Urtica dioica

During an ongoing investigation of Ascomycetes from plant substrates, three saprobic species were found from plant substrates. Two new species, Leptosphaeria regiae and Neomicrosphaeropsis juglandis were isolated from dead branches of Juglans regia from Turkey. Another species is introduced herein as Subplenodomus urticae sp. nov within the family Leptosphaeriaceae found on Urtica dioica in Italy. Multigene phylogenies based on combined LSU, ITS, SSU, and β-tubulin DNA sequence data generated from maximum likelihood and MrBayes analyses indicate that Leptosphaeria regiae is related to L. slovacica and forms an independent lineage within the genus Leptosphaeria. Subplenodomus urticae is basal to S. iridicola and its establishment as a new species is strongly supported. Neomicrosphaeropsis juglandis forms a moderately supported lineage in between N. italica and N. elaeagni in the Didymellaceae. Full morphological details are provided herein and phylogenetic relationships of the three new species are also discussed.


INTRODUCTION
The bitunicate fungi commonly known as Dothideomycetes is one of the largest group of fungi with a high level of diversity (Zhang et al., 2012;Hyde et al., 2013Hyde et al., , 2018Hyde et al., , 2019Tibpromma et al., 2017;Jayasiri et al., 2019). Most of them exist as decomposers, endophytes, epiphytes, fungicolous, lichenized, or lichenicolous fungi in diverse environments (Jeewon et al., 2013Phukhamsakda et al., 2016;Doilom et al., 2017;Wanasinghe et al., 2018;Pem et al., 2019a;Phookamsak et al., 2019). They can reproduce either sexually or asexually (Doilom et al., 2014Wijayawardene et al., 2014Wijayawardene et al., , 2018Pem et al., 2019c,d). The role of fungi as decomposers is crucial as it helps in recycling of nutrients and releasing enzymes hence maintaining the nutrient compositions of the ecosystem (Hyde and Jeewon, 2003;Hyde et al., 2005;Tang et al., 2005). Decomposers also play vital role in the ecosystem such as breakdown of rock to form soils, protection against pathogens and as a food source and alteration of pollutants (Dighton, 2016;Singh et al., 2016). Leptosphaeriaceae is a family in the order Pleosporales (Dothideomycetes, Ascomycota) introduced by Barr (1987) and typified by the genus Leptosphaeria. There are 1,800 epithets of Leptosphaeriaceae recorded in Index Fungorum (2020) with the largest number of species (1,669) occurring in the genus Leptosphaeria and 130 epithets in Mycobank (2020) but most of them lack molecular data. The family Leptosphaeriaceae is characterized by immersed to superficial ascomata, cylindrical to oblong pedicellate asci and reddish brown or yellowish brown, septate ascospores (Hyde et al., 2013;Ariyawansa et al., 2015;Dayarathne et al., 2015;Wanasinghe et al., 2016). Asexual morph are coelomycetous producing phialidic or annellidic conidiogenous cells (Wijayawardene et al., 2018). Leptosphaeriaceae species differ from other closely related families in the Dothideomycetes by the presence of a scleroplectenchymatous peridium. During our ongoing survey, another species resembling those of the asexual genus Neomicrosphaeropsis was observed. The genus Neomicrosphaeropsis was introduced by Thambugala et al. (2016) to accommodate four species namely; N. italica, N. novorossica, N. rossica, and N. tamaricicola. The type species is N. italica and was isolated from dead branches of Tamarix (Tamaricaceae) in Italy. Species of Neomicrosphaeropsis are pathogens or endophytes (Wijayawardene et al., 2017) and are morphologically characterized by hyaline to light brown, aseptate, obovoid to ellipsoidal conidia (Wijayawardene et al., 2018). The genus Neomicrosphaeropsis has been reported to comprise complex species which are morphologically similar but phylogenetically different (Thambugala et al., 2016). Neomicrosphaeropsis resembles species of Microsphaeropsis which is also accommodated in Didymellaceae as well as species of Coniothyrium in having hyaline to light brown, aseptate, obovoid to ellipsoidal, smooth-walled conidia (Verkley et al., 2014;Crous et al., 2019). There are currently 10 epithets in the genus Neomicrosphaeropsis (Index Fungorum, 2020). In this study, we introduce a new asexual species in the genus Neomicrosphaeropsis isolated from stems of Juglans regia (Juglandaceae) in Turkey using multi-gene (LSU, ITS, SSU, and β-tubulin) phylogenetic data. To the best of our knowledge, there have been no fungal species of Leptosphaeriaceae and Didymellaceae associated with Juglans regia in Turkey. We also report on a new species of Leptosphaeriaceae specifically in the genus Subplenodomus found on Urtica dioica from Italy. The aim of this study is to characterize these three fungal isolates in terms of morphology and phylogeny based on multi-gene sequence data.

Samples Collection, Morphological Examination, and Isolation
Specimens were collected from dead stems and branches of Juglans regia in the Corum and Kirikkale province of Turkey and on Urtica dioica in the province of Forlì-Cesena (FC) Italy (Figure 1). Samples were stored in Zip-lock bags and returned to the laboratories for examination and description of morphological characters. The specimens were observed under a Motic SMZ 168 series dissecting stereo-microscope. Free hand sections of fungal structures were taken and mounted in water for microscopic study. Photomicrography was carried out using a Canon 750D digital camera fitted to the microscope. Measurements were made with the Tarosoft (R) Image Frame Work software. The images were processed with Adobe Photoshop CS5 v. 12.0 software (Adobe Systems, United States) to illustrate fungal characters using a photoplate. Pure cultures were established from single ascospores/conidia on 2% malt extract agar (MEA; 62 g/L Criterion in distilled water) as described in Vijaykrishna et al. (2004) and Pem et al. (2019d). Cultures were incubated at 25 • C for up to 5 weeks and cultural characters were observed and measured after a week and again after 4 weeks following Liu et al. (2014). Holotype specimens are deposited in the herbarium located at Mae Fah Luang University (MFLU) and isotype specimens are deposited at the Kunming Institute of Botany, Academia Sinica Herbarium (HKAS), China. Ex-type living cultures are deposited at the Mae Fah Luang Culture Collection (MFLUCC) and duplicates at the Leibniz Institute DSMZ-German Collection of Microorganisms and Cell cultures (DSMZ). Faces of fungi numbers (Jayasiri et al., 2015) and Index Fungorum number 1 are provided.

DNA Extraction, Amplification, and Sequencing
Isolates were grown on MEA at 16 ± 2 • C for 8 weeks. DNA was extracted from fresh fungal mycelium using the DNA extraction kit (E.Z.N.A Fungal DNA Mini Kit, D3390-02, Omega Bio-Tek) following the manufacturer's protocol. Polymerase Chain Reaction (PCR) was performed to amplify specific gene regions. Primers LR0R and LR5 were used to amplify part of the nuclear ribosomal large subunit 28S rRNA gene (LSU) (Vilgalys and Hester, 1990;Rehner and Samuels, 1994). The Internal transcribed spacer rDNA region (ITS1, 5.8S rDNA, and ITS2) was determined using the primer ITS5/ITS4 and the 18S small subunit ribosomal RNA (SSU) was amplified using NS1/NS4 (White et al., 1990). For the beta-tubulin (TUB) gene, partial cds region, the primers Bt2a/Bt2b was used (Woudenberg et al., 2009). Primer sequences are available at the WASABI (Web Accessible Sequence Analysis for Biological Inference) database at the AFTOL website (aftol.org). Amplification reactions profiles for LSU, ITS, SSU, and β-tubulin gene followed Tang et al. (2007), Wang et al. (2007), and Luo et al. (2017). The analysis of PCR amplification products (amplimers) were performed by the commercial sequencing provider (BGI, Ltd Shenzhen, PR China). Same primer pairs used for amplification process were used for sequencing. The nucleotide sequence data acquired is deposited in GenBank ( Table 1). The final alignment and tree are deposited in the TreeBASE repository site 2 . The recommendations outlined by Jeewon and Hyde (2016) were used to establish the new taxa.

Phylogenetic Analysis
SeqMan v. 7.0.0 (DNASTAR, Madison, WI, United States) was used to assemble consensus sequences. Sequences of closely related strains were recovered from BLAST searches of GenBank 3 together with sequences of representative species used by Chen et al. (2015), Thambugala et al. (2016), andPhookamsak et al. (2019) and these are listed in Table 1. Sequences were aligned with online MAFFT v. 7 (Kuraku and Katoh, 2013;Katoh et al., 2019) 4 . The alignments were checked visually and improved manually where necessary using BioEdit v. 7.0.5.2 (Hall, 1999). Ambiguous regions were excluded from the analyses and gaps were treated as missing data. All novel sequences were deposited in GenBank and the final alignment and tree deposited in TreeBASE 2 . Phylogenetic analyses were based on maximum likelihood (ML) and Bayesian inference (BI) methods. Maximum likelihood analyses (ML), for single and combined gene alignments included 1,000 bootstrap replicates and was performed using RAxML-HPC2 run on XSEDE (8.2.8) (Stamatakis, 2014) in the CIPRES Science Gateway platform (Miller et al., 2011) using GTR+I+G model of evolution. The final tree was selected among suboptimal trees from each run by comparing likelihood scores with the GTRGAMMA nucleotide 3 http: //www.ncbi.nlm.nih.gov 4 http://mafft.cbrc.jp/alignment/server/index.html substitution model. The best fitting substitution model for each single gene partition and the concatenated data set was determined in MrModeltest 2.3 (Nylander, 2004) for Bayesian inference posterior probabilities (PP). GTR+I+G model was used for each partition or each gene separately, and incorporated into the analysis. The Bayesian inference posterior probabilities (PP) distribution (Zhaxybayeva and Gogarten, 2002) was estimated by Markov Chain Monte Carlo sampling (MCMC) in MrBayes v. 3.2.2 (Ronquist et al., 2012). The MCMC analyses, with six chains were run, started from random tree topology and lasted 1,000,000 generations and sampled every 100 generations (Nylander et al., 2008). The Tracer v. 1.5.0 software program was used to calculate the distribution of log-likelihood scores in order to determine the stationary phase for each search, to check whether extra runs were required to achieve convergence, the stable likelihood plateaus and burn-in value (Drummond et al., 2012). The first 2,000 generations were excluded as burn-in and 10,000 trees were obtained. Maximum likelihood bootstrap values equal or greater than 50% and Bayesian inference posterior probabilities (PP) equal or greater than 0.90 are given in black below or above each node (Figure 1). The phylograms were viewed in FigTree v1.4 5 and edited using Microsoft PowerPoint 2016.    Saprobic on dead stem of Juglans regia. Sexual morph: Ascomata 315-377 µm high, 364-410 µm diam., solitary or gregarious, superficial or semi-immersed on host tissue, visible as black spots on host surface, brown to dark brown. Ostiole apex dark brown to black, ostiolar canal filled with periphyses, papilla not conspicuous. Peridium 41-50 µm wide, comprising two cell types, outer layer composed of large, heavily pigmented, thick-walled cells of textura angularis, inner layer composed of scleroplectenchymatous cells of textura angularis. Hamathecium comprising numerous, 1.4-2.6 µm diam., slime coated, branched, cellular pseudoparaphyses. Asci 99-130 × 9-10 µm (x = 104.8 × 9.9 µm, n = 30), 8-spored, bitunicate, numerous, cylindrical to cylindric-clavate, short pedicellate, apically rounded, with indistinct ocular chamber. Ascospores 15-18 × 6-7 µm (x = 17.1 × 7.1 µm, n = 30), uni to bi-seriate, hyaline brown when immature, becoming yellowish brown to brown at maturity, ellipsoid to broadly fusiform, with rounded to acute ends, slightly clavate, narrow toward the base, 3-septate,  Notes -The new isolate Leptosphaeria regiae was obtained from dead stem of Juglans regia. In the NCBI BLASTn search of ITS sequence L. regiae has a closest match with L. sclerotioides (Preuss ex Sacc.) de Gruyter et al. (2013) (LP7-MRL) with identities 477/492 (97%) and 1% gaps. In our multigene phylogenetic analysis, L. regiae clusters close to L. slovacica (CBS 125975) with strong bootstrap support (100% ML, 1.00 PP). A comparison of 528 base pairs across the ITS (+5.8S) regions shows 71 (11.9%) base pair differences between L. regiae and L. slovacica. Morphologically, L. regiae differs from L. slovacica in its smaller ascospores (15-18 µm vs. 18-22 µm). We therefore, introduce Leptosphaeria regiae as a new species in the genus Leptosphaeria based on differences in morphology and DNA sequence data.
Culture characteristics -Colonies growing on MEA, reaching a diameter of 25 mm after 7 days at 25 • C, circular to irregular, flat to slightly raised, mycelium medium sparse, surface initially white, becoming pale saffron to pale white, reverse dark-gray with whitish edge, smooth at surface with entire to slightly filamentous edge, thinly hairy. Notes -Our new taxon Neomicrosphaeropsis juglandis is characterized by large, aseptate conidia with a unique yellowish or greenish brown color and measures 8-11 × 6-7 µm, compared to N. italica (3.6-6.2 µm × 2.9-4.6 µm), the type species of the genus Neomicrosphaeropsis. The ITS sequence comparison of N. juglandis with N. tamaricicola, N. italica and N. rossica reveals a difference of 0.8% (5 base pairs difference). However, strong evidence to support N. juglandis as a new species comes from the comparison of the RPB2 gene of our new species, N. juglandis to that of N. italica which shows a pairwise difference of 1.6% as well as that of β-tubulin with N. italica showing a difference of 2.1%. We therefore, introduce N. juglandis as a new species in the genus Neomicrosphaeropsis (Didymellaceae) based on morphological and phylogenetic evidence derived especially from protein coding genes.
Culture characteristics -Circular, surface rough, entire edge, in the middle powdery, on the edge thinly hairy, margin welldefined and slightly radiating, white and slightly raised in the middle, greenish gray at the edge; reverse white in the middle, strongly radiating, cracking the media, greenish-gray at the edges.  µm vs. 80-100 × 10-15 µm) and shorter ascospores (19-24 × 4.4-5.6 µm vs. 21-25 × 6-7 µm). Other species of Subplenodomus are in their asexual morph and thus cannot be compared. A comparison of 528 ITS (+5.8S) nucleotides between S. urticae and S. galiicola shows 57 (9.5%) base pair difference while that of S. urticae and S. iridicola shows 69 (13.1%) base pair difference. Thus, a new taxon is introduced as S. urticae based on the recommendations provided by Jeewon and Hyde (2016).

Phylogenetic Analyses
In the multi-locus phylogeny inferred from the combined dataset of LSU, ITS, SSU, and β-tubulin, several wellsupported clades can be recognized which are used for the delimitation of the 10 genera namely, Plenodomus, Alternariaster, Sphaerellopsis, Leptosphaeria, Alloleptosphaeria, Pseudoleptosphaeria, Subplenodomus, Paraleptosphaeria, and Heterospora (Figure 5). The genus Plenodomus forms a wellsupported clade within the family Leptosphaeriaceae and comprised 26 strains as well as the type species Plenodomus lingam (Tode: Fr.) Höhn. The genus Alternariaster also forms a well-supported clade sister to the Plenodomus clade and included four strains namely, A. trigonosporus, A. centaureaediffusae, A. bidentis along with A. helianthi, the type species of the genus Alternariaster. Sphaerellopsis strains together with the type strain, Sphaerellopsis filum (Biv.) B. Sutton clustered in a distinct clade, fully supported in all analyses (89% ML, 1.00 PP). Leptosphaeria sensu stricto forms a wellsupported clade in the family Leptosphaeriaceae comprising L. doliolum strains, the type species, strains of 12 other species along with the new species Leptosphaeria regiae (MFLUCC 18-1137). Our new taxon, L. regiae is close to L. slovacica (CBS 125975). The monotypic genera Alloleptosphaeria and Pseudoleptosphaeria form a distinct clade sister to each other and close to the genus Neoleptosphaeria which comprise three strains with N. rubefaciens as type species. Our new taxon, Subplenodomus urticae constitutes an independent lineage and fits within the genus Subplenodomus. The genus Paraleptosphaeria forms a well-supported monophyletic clade (89% ML, 1.00 PP) with four strains together with the type species Paraleptosphaeria nitschkei. The new species Neomicrosphaeropsis juglandis clusters in the family Didymellaceae with moderate support (85% ML, 1.00 PP). The multigene analyses show that N. juglandis (MFLUCC 18-0795) is phylogenetically related to species in the genus Neomicrosphaeropsis in particular to N. italica (Figure 5). We therefore, describe the three taxa as new based on the recommendations outlined by Jeewon and Hyde (2016).
There are six morphological species in the genus Subplenodomus (Species Fungorum, 2020) and all six species are described based on DNA sequence data (Tibpromma et al., 2017;Crous et al., 2018). The new species Subplenodomus urticae is morphologically similar to S. iridicola in sharing superficial or semi-immersed black ascomata and cylindrical asci with club-shaped pedicel but distinct in that the latter was described from Iris sp. (Iridaceae) from UK and has larger ascospores (21-25 × 5-7 µm vs. 19-24 × 4.4-5.6 µm) (Crous et al., 2018). A synopsis of Subplenodomus species is provided in Table 3. Phylogenetically, S. urticae clusters in the genus Subplenodomus basal to S. iridicola. Subplenodomus violicola is the type species of Subplenodomus and was established by de Gruyter et al. (2013). Since then, five additional species have been described in the genus. Subplenodomus urticae differs from S. iridicola by 13.1% nucleotide differences in the ITS regions. In our multi-gene analysis, the affinities of Subplenodomus corroborates those reported by previous studies (Schoch et al., 2009;Zhang et al., 2012;Crous et al., 2018;Phookamsak et al., 2019). Subplenodomus apiicola, S. drobnjacensis, S. valerianae, and S. violicola all produce pycnidia with an elongated neck. The pycnidial wall is pseudoparenchymatous. The new species S. urticae is unique and well-distinct among all the species reported in the genus Subplenodomus. Subplenodomus urticae is the first Subplenodomus species reported from Urtica dioica (Urticaceae) and is unique in having shorter cylindrical asci as well as ascospores compared to other species of Subplenodomus.
Our new species Neomicrosphaeropsis juglandis is an independent lineage close to N. italica. Among the several genes regions analyzed, it was noted that β-tubulin DNA sequence data generated relatively well-resolved topologies to support intergeneric relationships within the Didymellaceae and particularly in connection with Neomicrosphaeropsis (data not shown). Our new taxon is an addition to Neomicrosphaeropsis and is also the first record of the genus on Juglans regia in Turkey. A synopsis of the asexual morph of existing species of Neomicrosphaeropsis is provided in Table 4. Our new taxon is unique in that it produces larger aseptate conidia compared to   Neomicrosphaeropsis rossica 3-4.6 × 2-4 4.4-5.7 × 2.9-3.9 Tamarix ramosissima Ledeb.
Italy Thambugala et al., 2016 New species described in this study are indicated in bold.
other Neomicrosphaeropsis species and has been reported from a different host. Among the phenotypically diverse species, the genus Neomicrosphaeropsis as well as Didymellocamarosporium in Didymellaceae produce pigmented, muriform spores Thambugala et al., 2016). One interesting finding is also that Neomicrosphaeropsis cytisi, N. cytisicola and N. cytisina are morphologically similar with conidial measurement ranging between 4-7.9 × 2.5-3.5 µm and all of them were isolated from Cytisus sp. (Fabaceae). The authors differentiate the two species based on size and form of conidiomata.
However, it is highly probable that N. cytisi, N. Cytisicola, and N. cytisina are all same species as the ITS sequences of the three species are same (no base pair difference). RPB2 comparison of N. cytisi and N. cytisicola shows only three base pair difference across 1089 nucleotides examined. No TEF and BTUB sequences are available for the three species for comparison. Likewise, N. tamaricicola, N. rossica, N. Novorossica, and N. italica have been isolated from Tamarix species and are morphologically similar with conidial measurements ranging from 3.5-6.6 × 2.5-4.6 µm. Comparison of RPB2 and BTUB gene sequences between N. italica and N. tamaricola shows no base pair difference. However, comparison of TEF gene sequences between N. italica to N. rossica and N. tamaricicola shows 7 or 1.0% and 9 or 1.3% base pair differences, respectively, while there was no base pair differences between N. rossica and N. tamaricicola. It can be possible that some of these species are the same and need to be synonymised in future studies. In our phylogenetic analyses, N. minima is closely related to N. cytisi with strong bootstrap support. Comparison of available ITS and TEF sequences of N. minima and N. cytisi reveals zero base pair difference, however, there are some slight differences in conidial sizes (N. minima: 2.8-5.4 × 9.2-3.6 µm vs. N. cytisi: 4.5-7.9 × 3-5 µm). Neomicrosphaeropsis minima also differs from N. cytisi in conidiomatal size (N. minima: 60-80 µm diam., 60-95 µm high vs. N. cytisi: 75-155 µm diam., 75-130 µm high) and were isolated from different hosts. Whether these species are distinct and merit a specific taxonomic rank warrant further investigations given that minor conidial and conidiomatal size differences could vary under cultural and host conditions. Recollecting and sequencing of more coniothyrium-like and more fungi similar to Neomicrosphaeropsis from different geographical regions are also essential to clarify the placement of species and to infer relationships in Neomicrosphaeropsis and Neomicrosphaeropsis-like genera within the Didymellaceae (Pleosporales).

AUTHOR CONTRIBUTIONS
DP, RJ, and KH designed the study. FS did the sample collections. DP and RJ were involved in the phylogenetic analyses. SL contributed to the research funds. All authors contributed to the article and approved the submitted version.

FUNDING
This research work was partially supported by the Chiang Mai University.