Two new Micarea species (Pilocarpaceae) from Western Europe

Two new Micarea species, M. minuta and M. pseudotsugae, are described from the Netherlands based on morphological, chemical and molecular data. Micarea minuta belongs to the M. denigrata group and can be distinguished from other similar species of this group by its small apothecia, Sedifolia-grey pigment in the epihymenium, mostly 1-septate and small ascospores, 40–80 μm wide mesopycnidia, and the lack of secondary lichen metabolites detectable by thin-layer chromatography. Micarea pseudotsugae is a member of the M. prasina group and is characterized by an areolate, granular to subsquamulose thallus, the presence of methoxymicareic acid, the lack of Sedifolia-grey pigment, the presence of crystals (visible in polarized light) in the apothecia only, and one type of conidia (mesoconidia). The phylogenetic position of the newly described species was studied based on mtSSU rDNA marker analysis; M. minuta was found to be closely related to M. denigrata, M. nitschkeana and M. subnigrata, while M. pseudotsugae was found to be closely related to M. byssacea and M. laeta. Notes on secondary chemistry, ecology, distribution and phylogeny are given.


Introduction
The lichen genus Micarea has recently been studied intensively, and several new species have been recognized worldwide (Aptroot & Cáceres 2014;Brand et al. 2014;van den Boom & Ertz 2014;Guzow-Krzemińska et al. 2016van den Boom et al. 2017a, b;2018;Kantvilas 2018;Launis & Myllys 2019;Launis et al. 2019a, b). Two more new Micarea species from Western Europe (Belgium and the Netherlands) are described below. Micarea minuta belongs to the Micarea denigrata group, and M. pseudotsugae belongs to the M. prasina group. The former is a corticolous species with very small apothecia, so far known from a few localities in the Netherlands but abundantly present at the type locality. The latter is a species from soft bark of Pseudotsuga trees and soft wood of a Pseudotsuga stump, so far known only from a few localities in the Netherlands and one locality in Belgium.

Morphological and chemical analyses of specimens
Material of the two new species was collected by the first author in the Netherlands except for one collection made in Belgium, all deposited in LG, UGDA (holotypes) and the private herbarium of the first author. The specimens used for microscopic examination were apothecial sections and squashed thallus preparations in tap water, with or without the addition of C (commercial bleach), K (water solution of potassium hydroxide), N (nitric acid) and Lugol's reagent (I) with (K/I) or without (I) pre-treatment with K. Most microscopic measurements were made in water, except those of paraphyses, which were studied in water with the addition of K (Orange et al. 2001). In each collection, ~10 ascospores, conidia and paraphyses were measured to 0.5 μm accuracy. Chemical compounds were studied by thin-layer chromatography (TLC) using standard methods (Orange et al. 2001). The crystalline granules were observed with a compound microscope under polarized light (Launis et al. 2019a, b). DNA extraction, PCR amplification and DNA sequencing DNA was extracted directly from pieces of thalli by a modified CTAB method (Guzow-Krzemińska & Węgrzyn 2000). DNA extracts were used for PCR amplification; 25 μl PCR mix contained 1U DreamTaq polymerase (Thermo Scientific) and 1X concentrated buffer, 0.2 mM of each of the four dNTP's, 0.5 μM of each mrSSU1 and mrSSU3R primer (Zoller et al. 1999) and 10-50 ng of genomic DNA. Amplifications were performed in a Mastercycler (Eppendorf) under conditions described in Guzow-Krzemińska et al. (2019b).
Subsequently, PCR products were treated with a mixture of 20 units of Exonuclease I and 2 units of FastAP TM Thermosensitive Alkaline Phosphatase enzymes (Thermo Scientific) at 37°C for 15 min, followed by incubation at 85°C for 15 min to completely inactivate both enzymes. Sequencing of each PCR product was performed in Macrogen (www.macrogen.com) using the PCR primers.

Sequence alignment and phylogenetic analysis
The newly generated mtSSU sequences were compared with the sequences available in the GenBank database (http://www.ncbi.nlm.nih.gov/BLAST/) using BLASTn search (Altschul et al. 1990) to confirm their identity. The sequences of each marker were aligned with sequences of representatives of the genus Micarea obtained from GenBank and with Byssolecania variabilis and Byssoloma leucoblepharum as outgroup (specimens and GenBank accession numbers are listed in Table 1); 193 sequences in total were analyzed. Alignment was performed using the Guidance 2 server employing the MAFFT algorithm (Landan & Graur 2008;Penn et al. 2010;Sela et al. 2015). Terminal ends and portions of the alignment with ambiguous positions that might not have been homologous were excluded using Gblocks 0.91b (Castresana 2000;Dereeper et al. 2008) with less stringent settings (i.e. allowing smaller final blocks, gap positions within the final blocks, and less strict flanking positions). The final alignment consisted of 193 sequences of 770 sites.
The data were analyzed using a Bayesian approach (MCMC) in MrBayes 3.2 (Huelsenbeck & Ronquist 2001;Ronquist & Huelsenbeck 2003) and the GTR model. Two parallel runs were performed, each using four independent chains and 10 million generations, sampling trees every 1,000th generation. Posterior probabilities (PP) were determined by calculating a majority-rule consensus tree generated from the 15,002 post-burn-in trees of the 20,002 trees sampled by the two MCMC runs using the sumt option of MrBayes.
Maximum likelihood analysis was performed using RaxML HPC v. 8 on XSEDE (Stamatakis 2014) under the GTRGAMMAI model on the CIPRES Science Gateway (Miller et al. 2010). Rapid bootstrap analyses were performed with 1,000 bootstrap replicates (BS). The RAxML tree did not contradict the Bayesian tree topology for the strongly supported branches. Therefore only the maximum likelihood tree is shown, with the posterior probabilities (PP) of the Bayesian analysis and the bootstrap support values added near the internal branches. BS ≥ 70 and PP ≥ 0.95 were considered significant. Phylogenetic trees were visualized using FigTree v. 1.4.2 (Rambaut 2012) and the clades for previously described taxa are collapsed to reduce the size of the tree.

Results and discussion
The final DNA alignment consisted of 190 mtSSU rDNA sequences obtained from GenBank and three sequences of this marker from the newly described species and a specimen of M. microsorediata. Since the topologies from the maximum likelihood and Bayesian analyses did not show any strongly supported conflict, the Bayesian tree (Harmonic mean: -13673.39) is presented in Figure 1, with added posterior probabilities and bootstrap support from maximum likelihood analysis (Final ML Optimization Likelihood: -12102.250356).
The phylogenetic reconstruction ( Fig. 1) shows that Micarea forms two main lineages: a highly supported M. prasina group and allied species (1 PP, 98 BS), and a clade consisting of several other Micarea species (1 PP). The Micarea prasina group seems to be well studied, while the other species need further attention.
Although the M. prasina group has been widely studied recently (Guzow-Krzemińska et al. 2019a;Launis et al. 2019a, b) and numerous new species have already been recognized, here we describe another new species belonging to this group. Figure 1 shows that the entire M. prasina group is highly supported and monophyletic (1 PP, 95 BS) and agrees with previous phylogenies based  Launis et al. (2019a, b). This group accommodates the newly described M. pseudotsugae, which also produces methoxymicareic acid, and this species is closely related to M. byssacea and M. laeta  Additionally, several species closely related to the M. prasina group belong to this lineage: M. adnata, M. elachista, M. globulosella, M. misella, M. pycnidiophora, M. squamulosa and M. synotheoides. The position of M. stipitata is not resolved, as two sequences labelled as obtained from this species seem to be unrelated to each other (Fig. 1). Moreover, M. eximia also belongs to this lineage but forms a monophyletic clade together with specimens of M. misella (Fig. 1); this needs further study but was beyond the scope of this paper.
Chemistry. No lichen secondary substances detected by TLC in thallus and apothecia. Thallus C, K, KC and Pd negative. Apothecia and pycnidial tops with Sedifolia-grey pigment, K+ violet, C+ violet.
The new species is distributed throughout the south-eastern part of Noord-Brabant Province in the Netherlands, but not yet collected elsewhere. The species is very inconspicuous, even hard to observe under a binocular microscope, and it may be much more widespread than the relatively few collections suggest.
Notes. Morphologically and chemically, Micarea minuta resembles M. nitschkeana and M. sambuci and looks like a very depauperate specimen of both of them, but these two species have 3-septate ascospores and larger  (Coppins 1983;Czarnota 2007;van den Boom et al. 2018).
Habitat and distribution. The new species was found so far only on bark or on a stump of Pseudotsuga, especially the bases of trunks in forests poor in lichen species. The only accompanying species in the type locality are Coenogonium pineti, Lepraria incana, L. finkii and Porina leptalea. In the two other localities only Coenogonium pineti has been found as accompanying species.
Micarea pseudotsugae is distributed throughout the central part of the Netherlands to northern Belgium, but not yet collected elsewhere.
Notes. Micarea pseudotsugae is characterized by the areolate, continuous (only at the thallus edge with some separated areoles) thallus, consisting of flattened to convex, often subsquamulose granules, adnate to semi-immersed, mostly pale greyish brown apothecia, crystals (visible in polarized light) present only in the hymenium, conspicuous white mesopycnidia and the presence of methoxymicareic acid. It belongs to the Micarea micrococca group and is most closely related to M. byssacea and M. laeta. It differs from M. byssacea mostly by the lack of Sedifolia-grey pigment (Guzow-Krzemińska et al. 2019a;Launis et al. 2019a, b). Micarea laeta has a granular-areolate thallus but the areoles are never subsquamulose, and the thallus is also rarely continuous (Launis et al. 2019b). Due to the adnate apothecia, the new species is similar to M. microareolata, but the thallus of the latter consists of goniocysts usually coalescing to form convex to subglobose small areoles, and its paraphyses are richly branched (Launis et al. 2019b). In addition, in M. laeta and M. microareolata small crystalline granules are present in the hymenium and thallus (Launis et al. 2019b), whereas in the new species they are only in the hymenium. The crystals are so sparse in some apothecia that they may seem to be absent at first glance. Other species with similar morphology and the presence of methoxymicareic acid (e.g., M. pseudomicrococca) differ in having convex, rarely adnate and relatively smaller (up to 0.4 mm diam.) apothecia and non-subsquamulose areoles. Also, some of those (e.g., M. czarnotae) produce Sedifolia-grey pigment or their apothecia are very pale (white to whitish cream) (Launis et al. 2019b).
Micarea pseudotsugae is the second species of the M. micrococca group with crystals (visible in polarized light) only in the hymenium. Micarea czarnotae is the second one, but this species produces Sedifolia-grey pigment in the apothecia (Launis et al. 2019b).