New Insights into Lichenization in Agaricomycetes Based on an Unusual New Basidiolichen Species of Omphalina s. str.

The genus Omphalina is an ideal genus for studying the evolutionary mechanism of lichenization. Based on molecular phylogeny using ITS and nuLSU sequences by means of Bayesian and maximum likelihood analyses and morphological examination, combining the existence of green algae in basidiomata stipe and a Botrydina-type vegetative thallus, we described a bryophilous new basidiolichen species, Omphalina licheniformis, from a residential area of Jiangxi Province, China. This finding of unusual new basidiolichen species updated our understanding of the delimitation of Omphalina, indicating that both non-lichen-forming and lichen-forming fungal species are included simultaneously. The presence of algal cells in the basidiomata should receive more attention, as this would be helpful to distinguish more potential basidiolichens and explore the cryptic species diversity. This work provides new insights and evidence for understanding the significance of lichenization during the evolution of Agaricomycetes.

A cluster of bryophilous Omphalina basidiomata was found in a residential area of Jiangxi Province, China. The phylogenetic analyses of nrDNA ITS and nuLSU sequences also confirmed it to be an unknown Omphalina species. However, interestingly, it was found in the stipe existence of green algae; moreover, far fewer vegetative thalli consisting of green tiny globules (Botrydina-type) were also seen near to the hairs at the base of stipe. Therefore, an unusual new basidiolichen species of Omphalina is described and reported here. This finding indicates that Omphalina s. str. still consists of both non-lichenized and lichenized species with very a close phylogenetic relationship, and further provides new insights into the lichenization in Agaricomycetes.

Taxon Sampling and Morphological Examination
Five basidiomata specimens were collected from a residential area of Wan'an County, Jiangxi Province of China (Figure 1), and are preserved in the Herbarium Mycologicum Academiae Sinicae, Beijing, China (HMAS). Morphology and anatomy were examined using a MOTIC SMZ-168 stereomicroscope and a LEICA M125 dissecting microscope equipped with a Leica DFC450 camera.
A cluster of bryophilous Omphalina basidiomata was found in a residential area o Jiangxi Province, China. The phylogenetic analyses of nrDNA ITS and nuLSU sequences also confirmed it to be an unknown Omphalina species. However, interestingly, it wa found in the stipe existence of green algae; moreover, far fewer vegetative thalli consisting of green tiny globules (Botrydina-type) were also seen near to the hairs at the base of stipe Therefore, an unusual new basidiolichen species of Omphalina is described and reported here. This finding indicates that Omphalina s. str. still consists of both non-lichenized and lichenized species with very a close phylogenetic relationship, and further provides new insights into the lichenization in Agaricomycetes.

Taxon Sampling and Morphological Examination
Five basidiomata specimens were collected from a residential area of Wan'an County, Jiangxi Province of China (Figure 1), and are preserved in the Herbarium Myco logicum Academiae Sinicae, Beijing, China (HMAS). Morphology and anatomy were ex amined using a MOTIC SMZ-168 stereomicroscope and a LEICA M125 dissecting micro scope equipped with a Leica DFC450 camera.  Photographs of fresh specimens were taken immediately in the residential area, and the basidiomes were gathered. The morphological characteristics, including cap, stipe, pileus, and odor, were recorded. Specimens were dried in an electrical food drier at 55 • C to ensure that no moisture was left, and then were sealed in plastic bags.
To observe anatomic characteristics, parts of dried specimens were cut and mounted in 5% KOH and stained with 1% Congo Red. Anatomic characteristics including basidiospores, basidia, cystidia and pileipellis were observed under an Olympus CX31RTSF microscope (Made in Philippines, Tokyo, Japan) with at least 20 records. Data were analyzed and recorded as X = the mean of length by width ± SD, Q = the quotient of basidiospore length to width, and Qm = the mean of Q values ± SD. All of the protocols of the morphological study followed Largent's methodology [37].

DNA Extraction, PCR, and Sequencing
DNA was extracted from two fresh basidiomata (Table S1) by means of the modified CTAB method [38]. PCR was performed to amplify two gene loci: nuclear ribosomal DNA internal transcribed spacer (ITS) and large subunit (nuLSU), using primers ITS4 and ITS5 [39], and LR0R and LR5 [40], respectively. The PCR procedure followed Yang et al. [41].

Sequence Alignment and Phylogenetic Analysis
A total of 136 DNA sequences including four new sequences were used in this study (Table S1). Representative species of the lichenized genera of Hygrophoraceae (Agaricales), non-lichenized genera of Agaricales, and other related orders in the Agaricomycetes were chosen in the phylogenetic analyses. Multiclavula spp. (Clavulinaceae, Cantharellales) were taken as the outgroup.
Raw sequences were firstly assembled and edited with SeqMan [42], and then aligned using MAFFT v.7 [43]. We used the program Gblocks v.0.19b [44,45] to remove ambiguously aligned sites. The congruence of the two loci (ITS and nuLSU) was tested as described previously [46,47]. All maximum likelihood (ML) and Bayesian inference (BI) analyses were performed using the GTR + I + G model selected by jModelTest 2 [48]. The ML analysis involved 1000 pseudoreplicates with RAxML v.8.2.6 [49]. The BI analysis was performed using MrBayes v. 3.2.7 [50,51] with two parallel Markov chain Monte Carlo (MCMC), each using 5 million generations and sampling every 1000th generation. We used TRACER v.1.7.2 [52] to examine the standard deviation of split frequencies less than 0.01, reflecting the fact that the two trees differed very little, and the parameters converged. The 50% majority rule consensus tree was generated after discarding the first 25% as burn-in.

Phylogenetic Analysis
The aligned matrix contained 2011 unambiguous nucleotide (1129 ITS and 882 nuLSU) position characteristics for the full dataset of 93 members. BI and ML phylogenetic trees were constructed, and they had similar topological structures. The RAxML tree is shown in Figure 2 with both bootstrap support (BS) and posterior probability (PP) values of BI analysis. In the tree, all of the Omphalina species clustered into a well-supported monophyletic clade (BS 100/PP 1.00), obviously separated from other groups, in which the two samples (Coll. Nos. JX001 and ZRL20220005) are included and co-formed into a separate branch (BS 100/PP 0.99), indicating that this is a new species also supported by the morphological characteristics (see below). The BI phylogenetic tree and two single-gene-locus RAxML trees are shown in Figures S1-S3.

Taxonomy
The genus Omphalina is known to be a non-lichenized basidiolichen genus, because the original lichenized species contained in this genus have been separated out and formed totally different other genera such as Lichenomphalia [2]. However, an Omphalina new species was found to be lichenized in this study, and is described below. Therefore, the definition of the genus Omphalina also needs to be redefined as the genus with lichenized species in some cases.
Omphalina the morphological characteristics (see below). The BI phylogenetic tree and two singlegene-locus RAxML trees are shown in Figures S1-S3.

Taxonomy
The genus Omphalina is known to be a non-lichenized basidiolichen genus, because the original lichenized species contained in this genus have been separated out and formed totally different other genera such as Lichenomphalia [2]. However, an Omphalina new species was found to be lichenized in this study, and is described below. Therefore,  [56], and the algal cells a are also observed ( Figure 3J). However, the new species is distant from Lich spp., but clusters within the genus Omphalina, close to O. pyxidata (Bull.) Qu chionophila Lamoure in phylogeny (Figure 2), which also have brown caps, sm and decurrent lamellae [57], but are not known to have a lichenized form.

Discussion
Phylogenetic and morphological analyses support this new species as a member of Omphalina s. stricto. In the phylogenetic tree (Figure 2), O. chionophile is sister to O. licheniformis. However, in morphology, they can be distinguished by the size of the basidospores-that of O. chionophile is 8-10 × 5-6 μm [25,54,58], and that of O. licheniformis is 4.9-5.5 × 3.8-4.6 μm. Furthermore O. chionophile lacks cheilocystidia. The type species O. pyxidata is also phylogenetically close to O. licheniformis, sharing similar basidiome features which remain difficult to distinguish in situ. However, they can be distinctly separated under a microscope according to the basidiospore and cheilocystidia, as the basidiospores of O. pyxidata (7-8 × 5-6 μm) are larger than those of O. licheniformis (4.9-5.5 × 3.8-4.6 μm), Diagnosis: Omphalina licheniformis is distinguished from other species of this genus by having smaller basidiospores (4.9-5.5 × 3.8-4.6 µm) and presenting distinctive cheilocystidia. It is also characterized by the presence of far fewer vegetative thalli consisting of green tiny globules (Botrydina-type) near to the hairs at the base of the stipe, and green algae in the stipe.
Basidiomes small, omphalinoid. Pileus 7-15 mm in diam., subhemispherical when young and subfunnel when mature, distinctly depressed, hygrophanous, red brown to yellowish brown, edge paler, becoming alutaceous when drying, margin involute when young, wavy and striate when mature. Context thin, up to 1 mm, concolorous with pileus surface. Lamellae decurrent, distant, thick, forked and anastomosing, concolorous with pileus, but paler. Stipe 15-20 × 3-5 mm, cylindrical, hollow, smooth to white fibrillose, concolorous with the pileus, darker at the middle and lower parts with white mycelium-like cilia. Smell and taste indistinct. Notes: The vegetative thalli of this species are so tiny and few in number that they can very easily be ignored. The coexistence of algal cells in the base of the stipe near to the hairs is a very new finding, because previously, algal cells were only reported in the vegetative thallus of basidiolichens and known as green algae Coccomyxa [2,34,35]. The algal cells found in the new species are also unicellular and green, and unfortunately, this algal species has not been identified. However, the possibility that they are moss chloroplasts can be excluded, although moss chloroplasts are also unicellular and green, because chloroplasts are organelles within the cells of moss and need to live in the cytoplasm [55], and so it seems unlikely that the moss chloroplasts would escape from the moss cells and exist separately, trapped in the fungal hyphae. Furthermore, the moss chloroplasts are oval and 2.5-3 × 5-6 µm in size ( Figure 3K), different from the algal cells (6-10 µm in diam., Figure 3H). Inconspicuous or absent thalli are common in the basidiolichens of Agaricales, for example in the bryophilous or phycophilous basidiolichen genus Lichenomphalia, and the thalli are not obvious in some species such as L. umbellifera (L.) Redhead, Lutzoni, Moncalvo & Vilgalys and L. velutina (Quél.) Redhead, Lutzoni, Moncalvo & Vilgalys [2,34]. The Botrydina-type globular thallus of the new species ( Figure 3I) is also similar to L. meridionalis (Contu & La Rocca) P.A. Moreau & Courtec. [56], and the algal cells and hyphae are also observed ( Figure 3J). However, the new species is distant from Lichenomphalia spp., but clusters within the genus Omphalina, close to O. pyxidata (Bull.) Quél. and O. chionophila Lamoure in phylogeny (Figure 2), which also have brown caps, small agarics, and decurrent lamellae [57], but are not known to have a lichenized form.

Discussion
Phylogenetic and morphological analyses support this new species as a member of Omphalina s. stricto. In the phylogenetic tree (Figure 2), O. chionophile is sister to O. licheniformis. However, in morphology, they can be distinguished by the size of the basidospores-that of O. chionophile is 8-10 × 5-6 µm [25,54,58], and that of O. licheniformis is 4.9-5.5 × 3.8-4.6 µm. Furthermore O. chionophile lacks cheilocystidia. The type species O. pyxidata is also phylogenetically close to O. licheniformis, sharing similar basidiome features which remain difficult to distinguish in situ. However, they can be distinctly separated under a microscope according to the basidiospore and cheilocystidia, as the basidiospores of O. pyxidata (7-8 × 5-6 µm) are larger than those of O. licheniformis (4.9-5.5 × 3.8-4.6 µm), and the cheilocystidia of O. pyxidata are often branched, while the cheilocystidia of O. licheniformis are cylindrical and flexuose, narrowly clavate or lageniform, and not branched [53,54].
Lichen-forming fungi are an important component of the kingdom Fungi, making up nearly 20% of the known fungal species, among which over 99% of species belong to Ascomycota [1]. Previous studies showed that more losses than gains of lichenization have occurred during the evolution of Ascomycota, resulting in lichen-forming fungi becoming the ancestors of major lineages of non-lichen-forming fungi in Ascomycota [62]. Compared with ascolichens, basidiolichens are much rarer, and comprise less than 1% of species of the known lichen-forming fungi [1]; however, lichenization in the evolution of Basidiomycota is also very important in related research, and is even treated as one of necessary models to study the evolution of lichens [63].
The genus Omphalina has been taken as a model system to study lichenization since the late 1990s due to its variable nutritional modes [4]; however, after a series of species such as Omphalina umbellifera, etc., were transferred to other genera, no lichen-forming species were reported in Omphalina s.str., until the finding of Omphalina licheniformis in this study. Our study indicates that within the Agaricales, the lichenization process also occurred in the Omphalinaceae of the suborder Tricholomatineae. In the previous reports on basidiolichens, algal cells have never been found in the fruiting body structures [2,34,35,63], but indeed existed in the stipe of Omphalina licheniformis (Figure 3). This finding of unusual new species updated our understanding of the delimitation of Omphalina, indicating that both nonlichen-forming and lichen-forming fungal species are included simultaneously. Moreover, these results provide new insights and evidence for understanding the significance of lichenization during the evolution of Basidiomycota. Through this study, it should be noted that we need to pay more attention to the Basidiomycota fungi, especially to whether the algal cells are present in the fruiting bodies, which would be very helpful to distinguish more potential basidiolichens and explore the cryptic species diversity through these algal cell examinations.
The presence of algal cells as lichen photobionts is well-known to provide a carbon source for the mycobiont [64], which is relatively easy to understand in ascolichens, because fruiting bodies such as apothecia and pycnidia are closely connected parts of the lichen thallus, and the photobiont can be found both in the thallus and fruiting bodies, except the lecideine-type apothecia and pycnidia without hymenial algae. The algal cells in basidiolichens are assumed to be similar in function [5,65], but there is still an absence of strong evidence, especially due to the fruiting bodies of basidiolichens, which often look separable from the thallus in most cases, and algal cells have only been reported in the thallus previously [2,34,35,63]; moreover, sometimes the thallus is not obvious [2,34].
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/jof8101033/s1, Table S1: Specimens used for DNA extraction and GenBank accession number of all samples used in this study, Figure S1: The Bayesian tree based on the concatenated ITS + nuLSU (two genes) dataset, Figure S2: The maximum likelihood tree based on the ITS dataset, Figure S3: The maximum likelihood tree based on the nuLSU dataset.
Author Contributions: X.W. and R.Z. conceived and designed the study. Z.C. and W.G. discovered and collected specimens. T.Z., X.Z. and B.L. generated the DNA sequence data, X.Z., S.Q., X.W. and R.Z. performed the phenotypic analysis, T.Z., X.Z., A.V., X.W. and R.Z. analyzed the DNA data and checked issues related to nomenclatural articles. X.W. wrote the manuscript draft. A.V. and R.Z. All authors have read and agreed to the published version of the manuscript.