A New Lichenized Fungus, Psoroglaena humidosilvae, from a Forested Wetland of Korea, with a Taxonomic Key to the Species of Psoroglaena

Psoroglaena humidosilvae Lee is described as a new lichen species from a wetland forest in South Korea. The new species is distinct from P. stigonemoides (Orange) Henssen by little projections locally present on the thallus; smaller, paler, and globose perithecia; smaller asci; and smaller ascospores generally 3-septate. Molecular analyses employing internal transcribed spacer (ITS), mitochondrial small subunit (mtSSU), and nuclear large subunit ribosomal RNA (LSU) sequences strongly support P. humidosilvae as a nonidentical species in the genus Psoroglaena. A surrogate key is provided to assist in the identification of all 22 species of Psoroglaena.

This study describes a new lichenized fungus. One of the field surveys for the lichen biodiversity in the forested wetlands of South Korea were accomplished during the summer of 2021, and five specimens in Psoroglaena were collected in a small swamp of a mountain ( Figure 1). The specimens were analyzed in ecology, morphology, chemistry, and molecular

Morphological and Chemical Analyses
Hand sections were prepared manually with a razor blade under a stereomicroscope (Olympus optical SZ51; Olympus, Tokyo, Japan), scrutinized under a compound microscope (Nikon Eclipse E400; Nikon, Tokyo, Japan), and pictured using a software program (NIS-Elements D; Nikon, Tokyo, Japan) and a DS-Fi3 camera (Nikon, Tokyo, Japan) mounted on a Nikon Eclipse Ni-U microscope (Nikon, Tokyo, Japan). The ascospores were examined at 1000× magnification in water. The length and width of the ascospores were measured and the range of spore sizes was shown with average, standard deviation (SD), length-to-width ratio, and number of measured spores. Thin-layer chromatography (TLC) was performed using solvent systems A and C according to standard methods [23].

Isolation, DNA Extraction, Amplification, and Sequencing
Hand-cut sections of ten to twenty ascomata per collected specimen were prepared for DNA isolation and DNA was extracted with a NucleoSpin Plant II Kit in line with the manufacturer's instructions (Macherey-Nagel, Düren, Germany). PCR amplification for the internal transcribed spacer region (ITS1-5.8S-ITS2 rDNA), the mitochondrial small subunit, and the nuclear large subunit ribosomal RNA genes was achieved using Bioneer's AccuPower PCR Premix (Bioneer, Daejeon, Korea) in 20-μL tubes with 16 μL of

Morphological and Chemical Analyses
Hand sections were prepared manually with a razor blade under a stereomicroscope (Olympus optical SZ51; Olympus, Tokyo, Japan), scrutinized under a compound microscope (Nikon Eclipse E400; Nikon, Tokyo, Japan), and pictured using a software program (NIS-Elements D; Nikon, Tokyo, Japan) and a DS-Fi3 camera (Nikon, Tokyo, Japan) mounted on a Nikon Eclipse Ni-U microscope (Nikon, Tokyo, Japan). The ascospores were examined at 1000× magnification in water. The length and width of the ascospores were measured and the range of spore sizes was shown with average, standard deviation (SD), length-to-width ratio, and number of measured spores. Thin-layer chromatography (TLC) was performed using solvent systems A and C according to standard methods [23].

Isolation, DNA Extraction, Amplification, and Sequencing
Hand-cut sections of ten to twenty ascomata per collected specimen were prepared for DNA isolation and DNA was extracted with a NucleoSpin Plant II Kit in line with the manufacturer's instructions (Macherey-Nagel, Düren, Germany). PCR amplification for the internal transcribed spacer region (ITS1-5.8S-ITS2 rDNA), the mitochondrial small subunit, and the nuclear large subunit ribosomal RNA genes was achieved using Bioneer's Ac-cuPower PCR Premix (Bioneer, Daejeon, Korea) in 20-µL tubes with 16 µL of distilled water, 2 µL of DNA extracts, and 2 µL of primers ITS5 and ITS4 [24], mrSSU1 and mrSSU3R [25] or LR0R and LR5 [26]. The PCR thermal cycling parameters used were 95 • C (15 s), followed by 35 cycles of 95 • C (45 s), 54 • C (45 s), and 72 • C (1 min), and a final extension at 72 • C (7 min) based on Ekman [27]. The annealing temperature was occasionally altered by ±1 degree in order to obtain a better result. PCR purification and DNA sequencing were accomplished by the genomic research company Macrogen (Seoul, Korea).

Phylogenetic Analyses
All ITS, mtSSU, and LSU sequences were aligned and edited manually using ClustalW in Bioedit V7.2.6.1 [28]. All ambiguous characters were removed and only unambiguous characters were analyzed. The final alignment comprised 2338 (ITS), 968 (mtSSU), and 2404 (LSU) columns. Phylogenetic trees with bootstrap values were obtained in RAxML GUI 2.0.6 [29] using the Maximum Likelihood method with a rapid bootstrap with 1000 bootstrap replications and GTR GAMMA (TrN + I + G4 for ITS, TPM3uf + G4 for mtSSU, and TIM2 + G4 for LSU) for the substitution matrix, as the best models produced by the model test in the software. The posterior probabilities were obtained in BEAST 2.6.4 [30] using the GTR 121131/TN93 (ITS), the GTR 123123 (mtSSU), and the GTR 121343 (LSU) models, as the appropriate models of nucleotide substitution produced by the Bayesian model averaging methods with bModelTest [31], empirical base frequencies, gamma for the site heterogeneity model, four categories for gamma, and a 10,000,000 Markov chain Monte Carlo chain length with a 10,000-echo state screening and 1000 log parameters. Then, a consensus tree was constructed in TreeAnnotator 2.6.4 [30] with the first 25% discard as a burn-in, no posterior probability limit, a maximum clade credibility tree for the target tree type, and median node heights. All trees were displayed in FigTree 1.4.2 [32] and edited in Microsoft Paint. The phylogenetic trees and DNA sequence alignments are deposited in TreeBASE under the study ID 29486. Most sequences employed for the analyses are based on Muggia et al. [9]. Overall analyses in the materials and methods were accomplished based on Lee and Hur [33].

Phylogenetic Analyses
Three independent phylogenetic trees for the genus Psoroglaena and related genera in the family Verrucariaceae were produced from 91 sequences (27 for ITS, 33 for mtSSU, and 31 for LSU) from GenBank, and nine new sequences (3 for each locus) from the new species (Table 1). The new species is positioned in Psoroglaena in all ITS, mtSSU, and LSU trees. The ITS tree shows that the new species is clustered with P. stigonemoides and P. sp., represented by a bootstrap value of 100 and a posterior probability of 1.00 for the branch (Figure 2). Although the new species is neighboring with other Psoroglaena species in the branch, the species is separately located without coupling with any identical species. The mtSSU tree illustrates that the new species is grouped with P. abscondita (Coppins and Vězda) Hafellner and Türk, P. biatorella (Arnold) Lücking and Sérus., and P. sp., represented by a bootstrap value of 100 and a posterior probability of 1.0 for the branch (Figure 3). Although located closer to P. abscondita and P. biatorella than P. sp. in the branch, the new species is distinctly situated away from the close species. The LSU tree explains that the new species is arranged with other Psoroglaena species, i.e., P. biatorella and P. stigonemoides, represented by a bootstrap value of 57 (not shown) and a posterior probability of 1.00 (not shown) for the branch (Figure 4). Although the new species is grouped with the Psoroglaena species those are classified in a different clade separated from the new species. Unexpectedly, Psoroglaena abscondita is located in a clade of the genus Placidium, far from all other Psoroglaena species. This species is further considered in the discussion.    Phylogenetic relationships among available species in the genus Psoroglaena based on a maximum likelihood analysis of the dataset of ITS sequences. The tree was rooted with the sequences of the genus Placidium based on Muggia et al. [9]. Maximum likelihood bootstrap values ≥70% and posterior probabilities ≥ 95% are shown above internal branches. Branches with bootstrap values ≥90% are shown as fatty lines. The new species, P. humidosilvae, is presented in bold as their DNA sequences were produced from this study. All species names are followed by the GenBank accession numbers.      Distribution and ecology: The species occurs on the bark of Salix koreensis. The species is currently known from the type collections.
Etymology: The species epithet indicates the lichen's habitat, i.e., a humid forest. Notes: The new species is most similar to P. stigonemoides in having projections on thallus and transversely septate ascospores among corticolous species. However, P. stigonemoides differs from the new species by heavy projections covered on thallus, larger, darker, and ovoid-obpyriform perithecia, larger asci, and larger ascospores up to 5-septate [34].
Distribution and ecology: The species occurs on the bark of Salix koreensis. The species is currently known from the type collections.
Etymology: The species epithet indicates the lichen's habitat, i.e., a humid forest. Notes: The new species is most similar to P. stigonemoides in having projections on thallus and transversely septate ascospores among corticolous species. However, P. stigonemoides differs from the new species by heavy projections covered on thallus, larger, darker, and ovoid-obpyriform perithecia, larger asci, and larger ascospores up to 5-septate [34].
The new species can be compared with P. abscondita and P. chirisanensis in having transversely 3-septate ascospores among corticolous species. However, P. abscondita represents leprose thallus, darker perithecia, narrower exciple, and longer but narrower ascospores [34,35]. Psoroglaena chirisanensis differs from the new species by thallus without projections and larger ascospores with distinct constriction at septa [18] (Table 2).  The morphological and ecological characteristics for the closely related species are referenced from the previous literature. The measurements with asterisk marks in the column of P. stigonemoides are referenced from Lücking [4], which describes the characteristics of foliicolous P. stigonemoides possibly. All information on the new species is produced from selected specimens (KBA-L-0002000, KBA-L-0002005, KBA-L-0002006, KBA-L-0002007) in this study.

Discussion
The phylogenetic results strongly support that the new species is unique in Psoroglaena in all ITS, mtSSU, and LSU trees. Although previously recorded DNA sequences for Psoroglaena are not plentiful enough, all the sequences of the new species are clearly positioned in the Psoroglaena group. Whilst a closely related genus Agonimia was analyzed with the new species in the molecular phylogeny, lack of the DNA sequences for Phylloblastia, another comparable genus, did not allow for the classification in the phylogeny. However, the taxonomy of the new species was diagnosed by comparing with Phylloblastia in ecology and morphology. All species in Phylloblastia are known as foliicolous, if the corticolous P. gyeongsangbukensis is disregarded, and the new corticolous species in Psoroglaena can be distinguished from Phylloblastia.
A few molecular studies have previously been accomplished on Psoroglaena at the family level [9,10]. The genus Psoroglaena was classified on its own clade without any closely related genus in both studies. Psoroglaena is solely located in a clade without having any other genera. Although Psoroglaena is close to Agonimia or Phylloblastia in morphology, Agonimia was located far from Psoroglaena in their phylogeny, and Phylloblastia was not concerned in the phylogeny due to the lack of DNA data. This study explains that Psoroglaena is classified as distinct in Verrucariaceae as well. Although the LSU tree shows Psoroglaena is positioned close to Agonimia and Flakea, the ITS and mtSSU trees show that our genus is located on its own clade and unique in the family. Interestingly, P. abscondita in Muggia et al. [9] is classified far from other Psoroglaena species as the species is weirdly positioned in the LSU tree in this study. They decided that the distinct position of P. abscondita is due to the difference in the ribosomal DNA. We also experienced such a difference and the LSU locus showed high variability from Pyrenodesmia [36] and Arthonia (a manuscript in preparation). The high variability of the LSU sequences does not allow a clear classification in molecular phylogeny.
Five species of Psoroglaena were previously reported in Korea, which includes P. chirisanensis, P. coreana, P. gangwondoensis, P. japonica, and P. sunchonensis [18][19][20][21][22]. Corticolous species are P. chirisanensis, P. gangwondoensis, and P. sunchonensis, and other two are saxicolous. Here, the corticolous species are further considered to identify the new species. Both P. chirisanensis and P. gangwondoensis differ from the new species by inhabiting bark of Robinia psuedoacasia L. and representing larger ascospores without any filamentous projections on the thallus. Psoroglaena sunchonensis inhabits a similar substrate (i.e., Salix alba L.) and shows minute projections on thallus as the new species have such filamentous projections locally. Although sharing above similarities in ecology and morphology, the former represents whitish ascomata, different from constantly pale-yellow ones of the latter, and much wider (8-9 µm wide) and muriform ascospores unlike the purely transversely septate spores of the latter. All recorded Korean species of Psoroglaena have no DNA sequence data, and were not analyzed in molecular phylogeny.
Psoroglaena species can be collected by chance in the field as they are so minute and can be easily confused with something like algae or bryophytes. Particularly, Korean Psoroglaena species were found on planted trees just nearby a stream flowing through a local city (e.g., P. sunchonensis), or on a common tree species, Robinia pseudoacacia (e.g., P. chirisanensis and P. gangwondoensis), which is one of the representative invasive species to a disturbed area by artificial change. The new species was collected in a small swamp just below a forest road of a local mountain. Further fortuitous collections in an urban or a disturbed area for Psoroglaena and following molecular results from them will clarify their taxonomy beyond morphology.