Four new species of Parvoplaca (Teloschistales, Ascomycota)

. The genus Parvoplaca is extended with four new species: P. candanii from Antarctica, P. lamprocarpa from Alaska, and P. macroborealis and P. tenebrosa from Oregon. Collections identified as P. tiroliensis are shown to form two independent clades when DNA was analyzed, but it is currently not clear which clade represents the species in a strict sense. Parvoplaca athallina has been considered an Antarctic species, but according to our analyses, the species is shown to be bipolar occurring also in northern Europe, Greenland, Alaska and in California.


Introduction
The lichen genus Parvoplaca is a rather small genus in the family Teloschistaceae subfamily Xanthorioideae and was split from the large genus Caloplaca s.lat.Th.Fr. by Arup et al. (2013).It is characterized by a crustose, but usually poorly developed grey thallus and lecanorine, zeorine or sometimes biatorine yellow to orange apothecia that in several species lack anthraquinones completely or produce low amounts of the yellow-orange pigments.The anthraquinones present belong to the chemosyndrome A (Søchting 1997).The spores are polardiblastic with medium to long septum.All species are either corticolous, terricolous or grow on mosses or plant debris on the ground.No saxicolous species are currently known to us.The main distribution seems to be in the Northern Hemisphere, where all but one species occur, including the ones described in this paper, and two species occur in Antarctica.Parvoplaca differs from most other genera in the combination of being corticolous or growing on the ground, a poorly developed grey thallus that often is lacking, yellow to orange apothecium discs and usually a rather thin or suppressed thalline margin.Several species also show a tendency to darken or they have lost their orange pigments entirely, similar to the genus Pyrenode smia A. Massal.that, however, only includes saxicolous members.The most similar genus is Athallia Arup, Frödén & Søchting that includes some species with a similar combination of characteristics, i.e., A. saxifragarum (Poelt) Arup, Frödén & Søchting.Other genera with a morphology similar to Parvoplaca have either a different secondary chemistry, different spores or a different ecology.

Materials and methods
Fresh material of P. candanii was collected by MGH during a trip to Antarctica and material of P. macrobo realis and P. tenebrosa was collected by B. McCune.The material of P. alaskana was collected by B. McCune and T. Tønsberg during biodiversity assessments in Lake Clark National Park (McCune et al. 2018).Material of P. tiroliensis that was collected in Italy from the region where the species was described was borrowed from J. Nascimbene's personal herbarium with the aim to establish its identity.Collections of the described species are stored in BG, ERCH, LD and OSC.Herbarium locations are indicated after all cited collections.
The specimens were examined by interference contrast and light microscopy.Anatomical features were measured on hand-cut sections or squash preparations mounted in water.Morphological characters were measured on dry material using a dissecting microscope (40×).Spore dimensions are average values from ten measurements per specimen.Data on spore dimensions are presented in the following way: (min.extremes)xx-mean-xx(max. extremes).The measurements of anatomical and morphological characters mainly follow the guidelines of Vondrák et al. (2013).

Secondary chemistry
Secondary metabolites were identified and quantified using HPLC according to Søchting (1997).The relative composition of the secondary compounds was calculated based on absorbance at 270 nm according to Søchting (1997).All yellow, orange or reddish-pigmented parts are K+ purple.

DNA extraction
Genomic DNA extraction was performed directly using 5-10 fresh apothecia (fruitbodies).Total DNA was extracted using the DNeasy Plant Mini Kit from Qiagen (Cat.No./ID: 69204) following the accompanying protocol or the sequences were produced using direct PCR according to Arup et al. (2015).PCR was performed to amplify two gene loci: nuclear ribosomal DNA internal transcribed spacer (nrITS) and mitochondrial small subunit (mrSSU) rDNA.The nuclear rDNA ITS gene region was amplified by using the fungal-specific primer ITS1F and the universal primer ITS4 (Gardes & Bruns 1993;White et al. 1990).The mrSSU gene region was amplified by using the primers mtSSU1F (Zhou& Stanosz 2001) and mtSSU3R (Zoller et al. 1999) or mrSSU1 (Zoller et al. 1999) and mrSSU7 (Zhou & Stanosz 2001).ITS and mrSSU were amplified using the primers ITS1F and ITS4, mtSS1F and mtSSU3R in 30 µL reactions containing 15 µL 2 × Taq PCR MasterMix, 1 µL each primer solution (10 µM), 12 µL ddH2O and 2 µL genomic DNA, and the PCR cycling conditions comprised of an initial denaturation at 94°C for 5 min; 30-34 denaturation cycles at 94°C for 30 s, annealing at 53-56°C (ITS) or 50-54°C (mrSSU) for 30-60 s, extension at 72°C for 30-60 s; and a final extension at 72°C for 3-8 min.A touchdown procedure was used for some reactions, decreasing 1°C per cycle for the first 6 cycles.The target PCR products were checked by electrophoresis on 1.5% agarose gels.Sequence analyses of the lichen samples obtained from the PCR products were performed by the BM Labosis Laboratory (Ankara, Turkey) or by Macrogen Inc., South Korea, using the same primers as for the PCR.The two resulting strands were assembled using CLC Main Workbench 4.1.2™ or Geneious 11.1.5.Subsequent alignments were done in the same programs and adjusted manually.Sequences have been submitted to GenBank as indicated in Table 1.

Sequence alignment
Two different alignments were prepared, one for a combined analysis of the genes nrITS and mrSSU and one alignment of only nrITS sequences.The combined analysis included 33 sequences of the subfamily Xanthorio ideae (Arup et al. 2013) and the ITS alignment included 51 sequences of the genus Parvoplaca and phylogenetically closely related genera.In the first alignment, Cero thallia luteoalba was used as outgroup, and in the second analysis, Xanthomendoza mendozae was used as outgroup with Pachypeltis castellana as internal outgroup.The alignment program MAFFT as implemented in Geneious 11.1.5.was chosen for the initial alignments using the option auto and thereafter adjusted manually.Unalignable ends, introns in all the aligned genes and ambiguously aligned parts were excluded from the alignment.The alignments of the two different genes were first analysed separately to check for incongruence between genes, but none were detected.A conflict between the datasets was assumed to be significant if two different relationships were both supported with posterior probabilities of 0.95 or higher.

Phylogenetic analysis
We generated 26 new nrITS sequences and 12 new mrSSU sequences for this study.Phylogenetic relationships were inferred using Maximum Likelihood (ML) as implemented in IQ-TREE (Nguyen et al. 2015) and Bayesian tree inference was carried out using Markov chain Monte Carlo (MCMC) as implemented in MrBayes ver.3.2.4(Ronquist et al. 2012).In the combined analysis, the two included genes were treated as separate partitions.A suitable model of molecular evolution was found by IQ-TREE using the option model search in that program evaluating only the 24 models available in MrBayes, and later also used in the Bayesian analyses.For the combined analysis, the SYM+G model was found to be optimal for the nrITS and HKY+I for the mrSSU data set.For the pure ITS data set, SYM+G was again found to be optimal.Three parallel runs with 100,000,000 generations starting with a random tree and employing 6 simultaneous chains were executed, of which 5 were incrementally heated with a temperature of 0.10.Analyses were diagnosed every 1,000 generations in the last 50% of the tree sample and automatically halted when convergence was reached.Convergence was defined as a standard deviation of splits (of frequency 0.1) between runs below 0.01.Every 1,000 th tree was sampled.A majority-rule consensus tree was constructed from the post-burn-in tree samples.The consensus trees were visualized using FigTree 1.4.4 and re-drawn in Adobe Illustrator.

Results
The alignment for the first analysis consisted of 33 terminals of 1,339 aligned nucleotide sites, of which 389 were parsimony informative.The nrITS partition consisted of 527 sites (170 informative) and the mrSSU partition of 814 sites (63 informative).The first Bayesian analysis halted after 1,600,000 generations and a 50% majority-rule tree with posterior probabilities is shown in Fig. 1.The second analysis of only nrITS data consisted of 51 terminals of 530 sites of which 114 were parsimony informative.This analysis halted after 346,000 generations and the 50% majority-rule tree with posterior probabilities is shown above the branches in Fig. 2. The Maximum Likelihood analysis yielded trees very similar to the Bayesian one and only these are shown.
Maximum Likelihood bootstrap values are shown under the branches in the same figure.
The analysis of the combined data set of ITS and mrSSU sequences show the new species to be positioned well within the the genus Parvoplaca (Fig. 1).The genus has full support with Xanthomendoza and Pachypeltis in a sister position, similar to the results presented by Arup et al. (2013), where the genus was first recognized.
In the second analysis with only ITS sequences, nine monophyletic, fully supported clades representing species are found in addition to two species represented by only one sequence each (Fig. 2).The whole genus split at the bottom in two clades, however, one without support, with a larger supported clade including seven species.Of these, four are very similar to the appearance of P. tiroliensis s.lat., whereas, P. candani, P. tenebrosa and P. macrobo realis differ from the other four.The unsupported clade contains P. lamprocarpa, described here, and the two darkfruited species, P. servitiana and P. suspiciosa, together with the orange-fruited P. nigroblastidiata.

Discussion
We are now recognizing 11 species in the genus Parvo placa with P. tiroliensis as type.At the time of recognition of the genus (Arup et al. 2013)    & Arup 2021) and A. soropelta (Søchting & Castello 2012) have been firmly documented to be bipolar using molecular data.In the Northern Hemisphere, P. athallina has previously been confused with P. tiroliensis from which it can be separated only by molecular methods.The occurrence of cryptic species that cannot be recognized by any other means than DNA is really not practical, in particular for amateur lichenologists.If the similar species would form a separate monophyletic clade it would have been a possible solution to unite them all under one name, but the molecular data implies that these taxa are the result of convergent evolution in independent linages mixed with other morphologically recognizable species.Only P. chelyae and P. tiroliensis seem to be sister species, but this relationship has no support with the currently available data.Thus, for the time being, we recommend to keep the use of already described species, even if cryptic.
(Fig. 3A-G) MycoBank MB 849837 Diagnosis: Differs from other terricolous species of the genus by its continuous, well developed chalky white thallus.
Etymology.Named in honor of Prof. Dr. Mehmet Candan, a lichenologist from Turkey and the best friend of the third author MGH.

MycoBank MB 8498312
Diagnosis: Similar to Parvoplaca macroborealis, but differs in the smaller apothecia, shorter spores and the presences of pycnidia.
Etymology.The name reflects the bright orange apothecia from the greek words lamprós which means bright or shining and carpus which means fruit.
Chemistry.The apothecial disc and true exciple contain parietin as a major compound, and small amounts of fallacinal, emodin, teloschistin and parietinic acid, which corresponds to chemosyndrome A of Søchting (1997).The thallus and thalline margin contain no anthraquinones, but the pigment sedifolia-grey, which reacts K+ violet and N+ brownish red.
Habitat and distribution.The new species is so far known only from Lake Clark National Park in Alaska.It was found there on two different localities, once on Populus balsamifera and once on an unknown hardwood in a Picea forest, near a lake.Rather few accompanying species were found in the collections, only fragments of Phaeophyscia, probably P. ciliata (Hoffm.)Moberg, Scutula circumspecta (Vain.)Kistenich et al., Bacidia rosellizans S. Ekman and a Biatora.
Remarks.This species is characterized by the very intense orange apothecia on a rather thin and inconspicuous thallus.It resembles P. macroborealis and may, as this species, have a rather thin grey thalline margin, but differs in the smaller spores with a longer septum/spore length ratio, in having pycnidia and in the narrower paraphysis tips.Athallia pyracea is also similar, but has a less distinct thallus that usually shows some shades of yellow or orange, the spores are in general slightly longer with a longer septum (10.0-15.0µm and 3.8-5.5 µm versus 10.5-12.5 µm and septum 3.0-4.5 µm), and pycnidia are rare or absent in A. pyracea.

Parvoplaca macroborealis
MycoBank MB 849832 Diagnosis: Similar to Athallia borealis, but differs in larger apothecia that are zeorine, not biatorine, also similar to Parvoplaca lam procarpa, but differs in longer spores with a smaller septum/spore length ratio, in lacking pycnidia and in the wider paraphysis tips.Etymology.The new species is similar to Caloplaca borealis, but generally larger.
Chemistry.The apothecial disc and true exciple contain parietin as a major compound, and small amounts of fallacinal, emodin, teloschistin and parietinic acid, which corresponds to chemosyndrome A of Søchting (1997).The thallus and thalline margin contains no anthraquinones, but probably the pigment Sedifolia-grey, which reacts K+ violet and N+ brownish red, but the spot test reaction is hardly detectable.
Distribution and ecology.So far this species is only known from Wheeler County in Oregon where it is found in Umatilla National Forest (Fig. 6) on three localities  fairly close to one another.In all three cases the new species was found on decorticated twigs of Juniperus in a Juniperus -Pinus ponderosa savanna with volcanic boulders and thin soil at elevations between 1,312 and 1,358 m.The lichen community seems rich in lichen species according to the accompanying species on the twigs.Notes.Caloplaca macroborealis and Caloplaca borealis (Vain.)Poelt are similar in their grey thin thallus and orange apothecia that are sometimes grey on the outside of their apothecial margin.The grey color in C. borealis is not due to a thalline margin being present, it has biatorine apothecia, the outher part of the proper margin just turns grey to black rather often.In P. macroborealis the grey color is caused by a thalline margin that is often present in young apothecia, but often reduced and inconspicuous in older ones.The apothecia in C. borealis are normally smaller than in P. macroborealis, 0.2-0.4(-0.6)mm, whereas those in P. macroborealis are up to 1.3 mm in diameter.Athallia pyracea is also similar, but has a less distinct thallus that usually shows some shades of yellow or orange, the spore septum is generally longer, 3.8-5.5 µm versus 3.0-4.0µm.Description.Thallus discontinuous, as a thin blackish grey film over the bryophyte, somewhat uneven; prothallus not observed.Apothecia present, aggregate, adnate, round to somewhat irregular, zeorine, 0.2-0.65 mm diam.; disc somewhat concave to flat, blackish brown; proper margin raised above to level with disc, even, 35-70 µm thick, concolorous with disc, K-, consisting of radiating hyphae with elongate cells, 5-10 × 2 µm; thalline margin blackish grey, very inconspicuous, up to 30 µm thick, but often strongly supressed, cortex of inflated cells, 15-25 µm thick, pigmented part K+ violet, N+ red, inner parts with small POL+ crystals, not dissolving in K; epihymenium grey-to olive-brown, K+ brownish purple; N-; hymenium ~75 µm thick, hyaline; hypothecium ~50 µm thick, hyaline; paraphyses simple or sometimes branched once above, 1.5-2.0µm broad with upper cells wider, up to 6.0 µm, sometimes with oil droplets; asci cylindrical, 55-65 × 14-17 µm, (4-)8-spored; spores polardiblastic, fusiform to broadly ellipsoid, 10.0-13.42-15.5 × 7.0-8.17-10.0µm, septum 3.0-3.41-4.0µm (n = 9), ratio of spore length/width 1.25-1.66-2.0,ratio of septum/spore length 0.25-0.26-0.35,spores in 4-spored asci may be longer, up to 20 µm.Pycnidia not observed.
Chemistry.Apothecia and thallus without anthraquinones, but epihymenium with a grey-brown pigment, K+ brownish purple, N-.This pigment reminds of Verrucarioides-brown in Meyer & Printzen (2000), but is probably not identical with that.
Distribution and ecology.So far this species is only known from the type locality in Lake County, Oregon, where it was found on high altitude growing in a bryophyte cushion of the genus Grimma on rock outcrops in a semi-open mixed coniferous forest.It was associated with Phaeophyscia decolor (Kashiw.)Essl.and Parvo placa tiroliensis s.lat.(McCune pers. com.)Etymology.The name tenebrosa means dark and refers to the dark apotheca of the species.
Notes.Parvoplaca tenebrosa is morphologically very similar to forms of P. tiroliensis with dark apothecia where the anthraquinone crystals are few or entirely lacking.Such forms have also been recognized as separate species, Caloplaca friesii Magn., that was based on Caloplaca ferruginea var.melanocarpa Th.Fr.Also anatomically such forms are very similar to the new species, but they differ in their set of pigments.A brownish pigment in the cortex of P. tenebrosa reacts K+ violet and N+ red and in the type of C. friesii this pigment seems to be lacking.Furthermore, the epihymenium of P. tenebrosa is N-but in C. friesii it is dull orange reddish.Caloplaca friesii is currently regarded as a synonym of Parvoplaca tirolien sis, but their relationship needs further studies.Very dark forms of P. tiroliensis have a brownish pigment in the cortex that reacts K+ reddish that C. friesii seems to lack.

Figure 1 .
Figure 1.Majority-rule consensus tree based on a Bayesian MCMC analysis of a combined data set of the nuITS and mtSSU genes showing the phylogentic relationship of Parvoplaca to the nearest genera.Posterior probabilities ≥ 0.95 are shown above the branches and Maximum Likelihood bootstrap values ≥ 75 are shown below the branches.Species marked with an * are corticolous, all other Parvoplaca are terricolous, muscicolous or occurs on plant debris.

Figure 2 .
Figure 2. Majority-rule consensus tree based on a Bayesian MCMC analysis of ITS data of Parvoplaca.Branches with posterior probabilities higher or equal to 0.95 are shown in bold.Posterior probabilities ≥ 0.95 are shown above the branches and bootstrap values ≥ 75 are shown below the branches.Species marked with an * are corticolous, all other Parvoplaca are terricolous, muscicolous or occurs on plant debris.

Figure 6 .
Figure6.Pinus ponderosa savanna with thin soil where P. macroborealis was found on twigs of Juniperus.

Table 1 .
Location, collector and GenBank accession numbers of sequences of species used in the analyses.Sequences in bold are newly produced.
, there was no problem selecting this species as type.Based on the data presented (Darbishire 1912 however, two candidate clades for the application of the name P. tiroliensis as there are no molecular data from the more than 120 year old type collection.Clade A includes a mixture of northern European sequences and sequences from Austria and Italy, whereas, clade B includes two sequences from Italy together with a Chinese sequence.Since the type locality for P. tiroliensis is situated in South Tyrol, northeastern Italy, both clades are possible representatives.None of the sequenced specimens are from the type locality, only as close as possible to the same region.Specimens of the two clades are more or less identical and cannot be If one of the two candidate clades is considered to represent P. tiroliensis in a strict sense, then the other clade needs a name and there are at least two names to explore, viz.Caloplaca arctica H. Magn.and C. jungermanniae subsp.subolivaceum(Callopismasubolivaceum(Th.Fr) Räsänen).However, in our opinion the genetic profile of a specimen from the type locality needs to be established before any decision can be made, unless a differentiating character can be found.Parvoplaca athallina was described in 1912 from Graham Island in Antarctica(Darbishire 1912) and has been regarded as an Antarctic species ever since.However, our results clearly show that it also occurs in Norway and Sweden in Europe, in Greenland, in Alaska and in Yosemite National Park in California.It can thus be considered bipolar and belonging to a very small group of Teloschistaceae species with this distribution pattern.Within the family, only Xanthomendoza borealis (Lindblom & Søchting 2008), Austroplaca sibirica(Søchting separated by morphology, anatomy, chemistry, ecology or altitude; in other words, they are cryptic species.Also P. athallina and P. chelyae are very similar to P. tiroliensis and difficult to determine, but the name P. athallina can be safely applied to the clade including Antarctic specimens and P. chelyae to the clade including specimens from the Canary Islands, where their respective type localities are situated. Melanohalea elegantula (Zahlbr.)O. Blanco et al. and M. exasperatula (Nyl.)O. Blanco et al. are among the dominant species, but several crustose species also accur, such as Lecanora laxa (Śliwa & Wetmore) Printzen, Rino dina pyrina (Ach.)Arnold, Amandinea punctata (Hoffm.)Coppins.& Scheid., Candelaria pacifica Westb.& Arup, Letharia vulpina (L.) Hue s.lat.and unidentified species of Buellia, Lecidella and Bryoria.The composition indicates a somewhat enriched environment, perhaps by dust from the vulcanic boulders.