Eleven New Species of the Genus Tarzetta (Tarzettaceae, Pezizales) from Mexico

The genus Tarzetta is distributed mainly in temperate forests and establishes ectomycorrhizal associations with angiosperms and gymnosperms. Studies on this genus are scarce in México. A visual, morphological, and molecular (ITS-LSU) description of T. americupularis, T. cupressicola, T. davidii, T. durangensis, T. mesophila, T. mexicana, T. miquihuanensis, T. poblana, T. pseudobronca, T. texcocana, and T. victoriana was carried out in this work, associated with Abies, Quercus, and Pinus. The results of SEM showed an ornamented ascospores formation by Mexican Taxa; furthermore, the results showed that T. catinus and T. cupularis are only distributed in Europe and are not associated with any American host.


Introduction
Recently, the family Tarzettaceae (Pezizales, Pezizomycetes) was erected by Ekanayaka et al. [1] based on multigene phylogenetic analysis (ITS, LSU, SSU, and tef1-α, rpb2) and was segregated from the family Pyronemataceae according to Perry et al. [2].Previously, the family Pyronemataceae was considered polyphyletic by Hansen et al. [3] with the Geopyxis lineage and by Kumar et al. [4] with the Tarzetta-Geopyxis lineage.Currently, this family is represented by Tarzetta (Cooke) Lambotte as the type genus Geopyxis (Pers.)Sacc, Hydnocystis Tul.& C. Tull., Hypotarzetta Donadini, Paurocotylis Berk., and Stephensia Tull.& C. Tul.[1].However, the Index Fungorum (https://www.indexfungorum.org/,accessed on 15 January 2024) still considers it a monogeneric family.The genus Tarzetta has a restricted distribution mostly in temperate forests, which forms ectomycorrhizal associations, generally with trees and shrubs of the genera Abies Mill., Alnus Mill., Quercus L., Pinus L., and Pseudotsuga Carrière [3,[5][6][7].Tarzetta species are characterized as small to medium apothecia (2-30 mm), sessile to stipitate, deeply cupulate, and grey to beige but sometimes ochraceous or yellowish and rarely orange.Most of the species present a hymenium whitish or concolorous to the external zone of the apothecia, with a margin that The specimens were collected in temperate forests from 2019 to 2023.The collected specimens were deposited in the José Castillo Tovar mycological herbarium of the Insti-tuto Tecnológico de Ciudad Victoria (ITCV), the herbarium of the Escuela Nacional de Ciencias Biológicas of the Instituto Politécnico Nacional (ENCB), the herbarium of the Facultad de Estudios Superiores Zaragoza (FEZA) of the Universidad Nacional Autónoma de México (UNAM), and the Mycological Collection of the Universidad Estatal de Sonora (UES).Further, herbarium specimens were analyzed in ENCB, ITCV, and FEZA Herbaria.
The macroscopic morphological characteristics of specimens such as size, shape, and color were described [7].The Illustrated Dictionary of Mycology was used for morphological terminology [20].Apothecia colors are described according to Kornerup and Wanscher [21].Longitudinal cuts of the apothecia were made and rehydrated with 70% alcohol, 5% KOH, water, and cotton blue to observe a possible ornamentation of the ascospores.The microscopic characters such as excipulum, paraphyses, asci, and ascospores were characterized for identification using an optical microscope (OM) (Axiostar plus, Zeiss, Jena, Germany; VE-B1, Velab, Ciudad de México, Mexico).The photographs were taken with a Rebel T-1i camera, a 100 mm macro lens (Canon, Tokyo, Japan), and a DCS-W630 camera (SONY, Tokyo, Japan).Scanning Electron Microscopy (SEM; SU1510, Hitachi High Technologies, Tokyo, Japan) was used to observe the ornamentation of the ascospores.

Extraction, Amplification, and Sequencing of DNA
The DNA was extracted from dried herbarium specimens.Genomic DNA was extracted using the CTAB method [22].Two molecular markers were used, and these were the Internal Transcribed Spacer region of nuclear ribosomal DNA (ITS1-5.8-ITS2nrDNA; hereafter ITS) and the large subunit nrDNA (28S).PCR amplification included 35 cycles with an annealing temperature of 54 • C. It was carried out with the ITS5 and ITS4 primers [23] for the ITS nrDNA region and the LROR and LR5 primers [24] for the 28S nrDNA region (LSU).The PCR products were verified using agarose gel electrophoresis.The gels were run for 1 h at 95 V cm −3 in 1.5% agarose and 1× TAE buffer (Tris Acetate-EDTA, Saint Lois, MO, USA).The gel was stained with GelRed (Biotium, Fremont, CA, USA) and the bands were visualized in an Infinity 3000 transilluminator (Vilber Lourmat, Baden-Wurtemberg, Germany).The amplified products were purified with the ExoSAP Purification kit (Affymetrix, Santa Clara, CA, USA), following the manufacturer's instructions.They were quantified and prepared for the sequence reaction using a BigDye Terminator v. 3.1 (Applied Biosystems, Waltham, MA, USA).These products were sequenced in both directions with an Applied Biosystems model 3730XL (Applied BioSystems, Waltham, MA, USA) at the Instituto de Biología of the Universidad Nacional Autónoma de México (UNAM).

Sequence Assembly
The sequences of both strands of each of the genes were analyzed, edited, and assembled using BioEdit version 7.0.5 [25] to generate a consensus sequence, which was compared with those deposited in the GenBank of the National Center for Biotechnology Information (NCBI), using the tool BLASTN 2.2.19 [26].

Phylogenetic Analysis
To study phylogenetic relationships, our newly produced sequences of twenty-six individuals were added to reference sequences of ITS and LSU nrDNA deposited in the NCBI database (http://www.ncbi.nlm.nih.gov/genbank/,accessed on 25 January 2024), and an alignment was performed based on the taxonomic sampling employed by Van Vooren et al. [7] and Healy et al. [27] (Table 1).Each region was aligned using the online version of MAFFT v. 7 [28,29].The alignment was revised in PhyDE v. 10.0 [30], followed by minor manual adjustments to ensure character homology between taxa.The matrix was composed for ITS by 55 taxa (692 characters) and LSU by 61 taxa (800 characters).The data were analyzed using maximum parsimony (MP), maximum likelihood (ML), and Bayesian inference (BI).Maximum parsimony analyses were carried out in PAUP* 4.0b10 [31] using the heuristic search mode, 1000 random starting replicates, and TBR branch swapping with MULTREES and Collapse on.Bootstrap values were estimated using 1000 bootstrap replicates under the heuristic search mode, each with 100 random starting replicates.Maximum likelihood analyses were carried out in RAxML v. 8.2.10 [32] with a GTR + G model of nucleotide substitution.To assess branch support, 10,000 rapid bootstrap replicates were run with the GTRGAMMA model.Bayesian inference was carried out in MrBayes v. 3.2.6 x64 [33] with four chains, and the best evolutionary model for alignment was sought using PartitionFinder v. 2 [34][35][36].The information block for the matrix includes two simultaneous runs, four Montecarlo chains, a temperature set to 0.2, and a sampling of 10 million generations (standard deviation ≤ 0.1) with trees sampled every 1000 generations.The first 25% of samples were discarded as burn-in, and convergence was evaluated by examining the standard deviation of split frequencies among runs and by plotting the log-likelihood values from each run using Tracer v. 1 [37].The remaining trees were used to calculate a 50% majority-rule consensus topology and posterior probabilities (PP).Trees were visualized and optimized in FigTree v. 1.4.4 [38].

Molecular Analysis
Phylogenetic reconstruction was based on the alignment of the nrITS + LSU dataset (56 taxa, 1520 characters, including gaps).The three phylogenetic analyses of the dataset, MP, ML, and BI, recovered similar topologies (Figure 1).No significant conflict (bootstrap value > 70%) was detected among the topologies obtained via separate phylogenetic analyses.The parsimony analysis of the alignment found 1205 trees of 291 steps (CI = 0.5022, HI = 0.1475, RI = 0.4785, RC = 0.3785).The best RA×ML tree with a final likelihood value of -44,572.924927 is presented.The matrix had 1095 distinct alignment patterns, with 5.15% undetermined characters or gaps.Estimated base frequencies were as follows: A = 0.114712, C = 0.191626, G = 0.180634, T = 0.213028; substitution rates AC = 1.007806,AG= 1.154719, AT = 1.290447,CG = 1.045887,CT = 4.696475, GT = 1.000000; and gamma distribution shape parameter α= 0.002898.In the Bayesian analysis, the standard deviation between the chains stabilized at 0.00002 after 3 million generations.No significant changes in tree topology trace or cumulative split frequencies of selected nodes were observed after approximately 0.25 million generations, which were discarded as 25% burn-in.

Taxonomy
Eleven species of the genus Tarzetta are described as a new species, and they are based on morphological, ecological, and molecular characteristics.Furthermore, a map of Mexico shows the distribution of the type species (Figure 2); a comparative table of Mexican species and some American and European species with morphological and ecological characteristics (vegetation type and ectomycorrizal host) (Table 2) and the taxonomic key of the Mexican species of Tarzetta are included.

Taxonomy
Eleven species of the genus Tarzetta are described as a new species, and they are based on morphological, ecological, and molecular characteristics.Furthermore, a map of Mexico shows the distribution of the type species (Figure 2); a comparative table of Mexican species and some American and European species with morphological and ecological characteristics (vegetation type and ectomycorrizal host) (Table 2) and the taxonomic key of the Mexican species of Tarzetta are included.GenBank: ITS: PP825384, LSU: PP825427.

Discussion
In this study, 11 species of Tarzetta are described from Mexico, based on ecological, morphological, and molecular data.Combined analyses of two datasets (ITS and LSU) showed two strongly supported clades described as follows: Clade I, with most of the species growing on the Sierra Madre Oriental, is mainly associated with Quercus species, except T. davidii, which is putatively associated with Abies religiosa at least within the Transversal Neovolcanic Axis [42].Species of this clade show paraphyses generally slightly branched, filiform, and septate.Clade II, with most of the species growing on the Sierra Madre Occidental, is associated with conifers; T. texcocana and T. pseudobronca seem to be the exception, where the former may be associated with Quercus species, while the second associated with Pinus cembroides.Most of the species of this clade have paraphyses bifurcate with few septa; nevertheless, T. pseudobronca can show branched paraphysis.
According to Van Vooren et al. [7], the main diagnostic characteristics for the description of Tarzetta species are the size of the ascoma and ascospores, as well as the host.In this work, two clades are observed, separated mainly by the paraphyses structure.Van Vooren et al. [7] considered this character of little taxonomic value because the apical area of the paraphyses seems to be highly variable in the same species, depending on the development stage of the area where the observations are made.However, there is a tendency for Mexican species to group, according to this character, although it is not consistent.According to the host, a marked separation of the clades is observed, with the species of Clades I and II mostly associated with conifers and Quercus species, respectively.In the case of T. cupressicola, this species grows under Cupressus lusitanica; however, there is no evidence that it is directly associated with mycorrhiza.
Although most species of this genus have been described and cited with smooth spores, except J. jafneospora with verrucose ascospores [7,9,39], all Mexican species seen in OM are smooth but under SEM they show finely rugose ornamentation (Figures 14 and 15).

Figure 1 .
Figure 1.Maximum likelihood phylogeny based on the nrITS + LSU sequence data.Maximum parsimony and Bayesian analyses recovered identical topologies concerning the relationships among the main clades of the Tarzetta.For each node, the following values are provided: maximum parsimony bootstrap (%)/maximum likelihood bootstrap (%)/ and posterior confidence (p-value).The scale bar represents the expected number of nucleotide substitutions per site.The new species of Tarzetta are shown in bold.

Figure 1 .
Figure 1.Maximum likelihood phylogeny based on the nrITS + LSU sequence data.Maximum parsimony and Bayesian analyses recovered identical topologies concerning the relationships among the main clades of the Tarzetta.For each node, the following values are provided: maximum parsimony bootstrap (%)/maximum likelihood bootstrap (%)/ and posterior confidence (p-value).The scale bar represents the expected number of nucleotide substitutions per site.The new species of Tarzetta are shown in bold.

Table 1 .
Taxa information and GenBank accession numbers of the sequences used in this study.