Introduction
The family Porinaceae contains nearly 430 species, of which approximately 45% are corticolous, 35% are foliicolous, and 20% are saxicolous (McCarthy Reference McCarthy2013). In recent decades most authors have accepted only two or three genera within the family, but four were accepted by Harris (Reference Harris1995) and five by Hafellner & Kalb (Reference Hafellner and Kalb1995). Hafellner & Kalb defined their genera on the basis of characters including the presence of setae on the perithecium (Trichothelium), the type of acetone-insoluble pigment in the perithecial wall, and the presence or absence of a chitinoid ring structure in the ascus apex. However, these characters do not always agree with the authors placement of species. The newly-defined pigment Porina-yellow was said to be characteristic of Porina s. str., and absent in other genera, but it occurs as a minor pigment in several species placed in Pseudosagedia. Pseudosagedia was said to have a ring structure in the ascus, but one is absent in P. guentheri (Orange et al. Reference Orange, Purvis and James2009). Baloch & Grube (Reference Baloch and Grube2006) analyzed 28 mostly foliicolous species, using mitochondrial SSU rDNA sequences. Their analysis revealed four main clades within the material studied, and Trichothelium was nested within Porina; the generic concepts used by Hafellner & Kalb as well as Harris were not supported. Nelsen et al. (Reference Nelsen, Lücking, Andrew, Aptroot, Cáceres, Mercado-Díaz, Rivas Plata and Lümbsch2014), using mitochondrial SSU rDNA and nuclear LSU rDNA, and including some corticolous species of Porina, showed that the genus Myeloconis belongs in Porinaceae, and is sister to Porina farinosa. The genus Porina is well studied, with recent world keys to foliicolous species (Lücking Reference Lücking2004) and saxicolous species (McCarthy Reference McCarthy2000).
A distinctive Porina has been known from streams in Wales for some years, but has been misidentified as Porina lectissima. It is described here as new, based on morphology and on mitochondrial SSU rDNA and nuclear ribosomal ITS sequences.
Materials and Methods
DNA was extracted from recently collected or frozen specimens, using the Qiagen DNeasy Plant Mini Kit; the manufacturer’s instructions were followed except that warm water was used for the final elution. PCR amplification was carried out using Bioneer AccuPower PCR Premix in two 20 µl tubes. The two internal transcribed spacer regions and the 5.8S region (ITS1-5.8S-ITS2) of the nuclear ribosomal genes, and part of the small subunit of the mitochondrial ribosomal DNA (mtSSU) were amplified, using the primers ITS1F, ITS2, ITS3, ITS4, and mrSSU1, mrSSU3R. The PCR thermal cycling parameters were: initial denaturation for 5 min at 94 °C, followed by 5 cycles of 30 s at 94 °C, 30 s at 55 °C, and 1 min at 72 °C, then 30 cycles of 30 s at 94 °C, 30 s at 52 °C and 1 min at 72 °C. PCR products were visualized on agarose gels stained with ethidium bromide, and purified using the Sigma GenElute PCR Clean-Up Kit. Sequencing was performed by The Sequencing Service (College of Life Sciences, University of Dundee, www.dnaseq.co.uk) using Applied Biosystems Big-Dye Ver 3.1 chemistry on an Applied Biosystems model 3730 automated capillary DNA sequencer.
Alignment of assembled sequences was carried out using BioEdit (http://www.mbio.ncsu.edu/BioEdit/bioedit.html); ClustalW was used to create an initial alignment, which was edited manually. Ambiguously aligned regions were excluded from the analysis.
Phylogenetic relationships and support values based on mtSSU data were investigated using a Bayesian approach, and additional support values were obtained using Maximum Likelihood bootstrapping, as implemented in RaxML (Stamatakis Reference Stamatakis2006; Stamatakis et al. Reference Stamatakis, Hoover and Rougemont2008); both programs were hosted on the CIPRES Science Gateway (Miller et al. Reference Miller, Pfeiffer and Schwartz2010). The model of evolution for the Bayesian analysis (HKY+I+G) was selected using the Akaike Information Criterion (AIC) in MrModeltest 2.2 (Nylander Reference Nylander2004). Gaps were treated as missing data. Using MrBayes, two analyses of two parallel runs were carried out for 1 000 000 generations, with trees sampled every 100 generations. Stationarity was considered to have been reached when the mean standard deviation of split frequencies dropped to <0·01. A burn-in sample of 2500 trees was discarded from each run, respectively. Support values of ≥95% Bayesian posterior probabilities and ≥70% Maximum Likelihood bootstrapping were regarded as significant. ITS data were investigated using a similar Bayesian approach; models of evolution for the ITS1, 5.8S and ITS2 regions were HKY, K80 and GTR + I, respectively. A partitioned analysis was run for 500 000 generations, and a burn-in sample of 1250 trees was discarded.
Mitochondrial SSU sequences of saxicolous and corticolous species available on GenBank were included in the analysis, but not the numerous sequences of foliicolous species. No ITS sequences of Porinaceae were available from GenBank. Specimens used in the analyses are shown in Table 1.
Table 1 Specimens used in the phylogenetic analysis. Accession numbers refer to mtSSU sequences unless otherwise indicated. New sequences are in bold
Acetone-insoluble pigments in the perithecium were observed in water and in 10% KOH.
Results
Fifteen mtSSU sequences were prepared for the new species and for several additional taxa. The ITS region was unexpectedly difficult to amplify, and a number of sequences obtained proved to be contaminants. Twelve ITS sequences were prepared.
The aligned mtSSU region comprised 1150 sites, of which 625 were removed as ambiguous. The 50% majority-rule consensus tree from the Bayesian analysis of mtSSU sequences is shown in Fig. 1. The lower nodes of the trees are poorly supported. Well-supported clades include: 1) Porina lectissima together with the single sequence of P. austroatlantica; 2) P. aenea and P. chlorotica; 3) the two sequences of P. rivalis together with one unidentified sequence; and 4) five sequences of P. byssophila. One sequence named as P. byssophila from GenBank clustered with Porina aenea/chlorotica, but is evidently based on a misidentified specimen.
Fig. 1 Phylogenetic relationships of Porina species, based on a Bayesian analysis of the mitochondrial SSU. The tree was rooted using Belonia russula and Gyalecta ulmi. The two support branches associated with each branch are posterior probabilities (PP) and maximum likelihood bootstrap (MLb) values, respectively. Branches in bold indicate a support of PP ≥95% and MLb ≥70%.
The aligned ITS1-5.8S-ITS2 region comprised 583 sites, of which 230 were deleted as ambiguous. The unrooted 50% majority-rule consensus tree from the Bayesian analysis of the ITS region is shown in Figure 2, and supports the distinctness of the new species from P. lectissima and others.
Fig. 2 Phylogenetic relationships of Porina, based on a Bayesian analysis of the ITS region. The tree is unrooted.
The Species
Porina rivalis Orange sp. nov.
MycoBank No.: MB814184
Thallus thin; perithecia prominent, involucrellum with acetone-insoluble pigments Sagedia-red and Porina-yellow; ascospores 3-septate, 13·0–18·5×4·0–5·5 µm.
Type: Great Britain, Wales, Breconshire, near Llanwrtyd Wells, Nant Walch, 22/8551.4971, alt. 230 m, on stones submerged in stream, shaded; water pH 6·6, conductivity 43 µS cm−1, 16 September 2011, Alan Orange 20628 (NMW − C.2015.005.8—holotype; GenBank accession nos: KR108900 (ITS), KR108913 (mtSSU)).
Fig. 3 Porina rivalis (holotype). Scale=1 mm.
Fig. 4 Porina rivalis, sections of perithecia. A, Orange 20439 in water; B, Orange 20439 in K; C, holotype in water; D, holotype in K; E, Orange 20644 in K; F, Orange 20443 in K. Scale=100 µm.
Prothallus brown, very thin (seen once). Thallus light orange-brown to grey-brown or dark grey (orange tints disappearing on storage), thin, 20–70 µm thick, continuous or with scattered cracks. Photobiont trentepohlioid. Perithecia prominent, 160–400 µm diam., dark brown or black, sometimes orange-brown or brown at extreme base. Involucrellum of isodiametric thick-walled cells, enclosing numerous photobiont cells, without crystals; inner part yellow to orange, K+ orange-red (Porina-yellow), near upper surface dark grey to purplish red, K+ dark grey or bluish grey (Sagedia-red at least in part), a small area adjacent to ostiole often dark dull violet. Centrum 185–295 µm diam.; exciple colourless or yellow (Porina-yellow). Ascus ± cylindrical, thin-walled, I−, with truncate apex, with a ring structure. Ascospores 3-septate, narrowly ellipsoid, 13·0–15·3–17·5(–18·5)×4·0–4·8–5·5 µm, 2·3–3·2–4·1(–4·3) times as long as wide (based on 61 spores from 10 specimens).
Conidiomata up to 80 µm diam., occurring near junctions of conspecific thalli; conidia rod-shaped, aseptate, 4·0–4·5×1·2 µm.
Ecology and distribution. On frequently inundated siliceous rocks beside streams that are neither strongly acidified nor nutrient-enriched. The water chemistry at two sites in Mid-Wales was recorded on eight (calcium) to ten (others) occasions during 2011–2013 (mean values in bold):
Irfon: pH 5·8–6·5–7·2, conductivity 35–42–48 µS cm–1, calcium 1·7–2·1–2·7 mg l–1;
Nant Walch: pH 5·1–6·1–7·0, conductivity 32–44–56 µS cm–1, calcium 1·7–2·4–4·6 mg l–1.
Associated species include Dermatocarpon luridum, Ephebe lanata, Ionaspis lacustris, Rhizocarpon lavatum, Verrucaria consociata, V. rosula, V. sublobulata, and the bryophytes Heterocladium heteropterum, Hygrohypnum ochraceum, Hyocomium armoricum, Marsupella emarginata, Racomitrium aciculare and Scapania undulata. Porina rivalis has been recorded from at least 10 streams in Mid and North Wales, and one stream each in South-west England and North England.
Notes. Porina rivalis is best detected in the field by its distinctly semi-aquatic habit. In some specimens the perithecium is black above but concolorous with the thallus below, giving a distinctive appearance (Fig. 3), but in others it is entirely black. In section, the involucrellum has a cap of dark pigment over an inner layer containing yellow to brownish orange pigment. The amount of dark pigment is variable in extent, and it can be sparse in shaded habitats. The contrasting pigments are more easily seen in K, when the outer layer is grey, and the inner layer is often a rich orange-red (Fig. 4). Porina rivalis is likely to be confused with other saxicolous species with 3-septate spores. Porina lectissima has a thin to rather thick thallus, 40–200 µm thick, which is often cracked; the perithecia are somewhat larger, 320–500 µm diam. The perithecia of P. lectissima are often orange or orange-brown, at least when young or wet, but in a few specimens from exposed habitats they appear almost black when dry. In these cases, the pigment is still almost exclusively Porina-yellow. The ascospores of P. lectissima are considerably larger, (21·5–)22·0–26·5–31·0×4·5–5·5–6·5 µm. It is frequent on moist or flushed siliceous rocks, and while it can occur in places where it is submerged by high river flows, it is not a member of the distinctly semi-aquatic community of lichens and bryophytes. Porina leptalea has small perithecia, 240–320 µm diam., containing only Porina-yellow, and the ascospores are 14·5–23·0×3·5–5·0 µm. It is often found on bark, but can also occur on moist rocks. Porina chlorotica has a purplish brown pigment in the involucrellum, and Porina-yellow is absent or very localized, while P. byssophila has an involucrellum that is dominated by dark pigments, with only small amounts of Porina-yellow, and grows on non-aquatic rocks and bark. Porina pacifica Brodo (Brodo Reference Brodo2004) from western North America has the inner layer of the involucrellum containing Porina-yellow (best seen in K), but the thicker outer layer is dull purple-brown (colour only visible in very thin section). An unidentified collection from British Columbia (Orange 22778) on shaded stones has an involucrellum dominated by Porina-yellow, with a trace of darker pigment at the ostiole; the ascospores are slightly larger than in P. rivalis, 18–21(–23)×4·5 µm. It is basal to P. rivalis in the mtSSU tree.
Additional selected specimens examined. Great Britain: Wales: V.C. 48, Merioneth: near Ganllwyd, Afon Gain near confluence with Afon Mawddach, 23/735.264, 1996, A. Orange 11137 (NMW – C97.35.184); above Llyn Celyn, Afon Tryweryn, 23/843.399, 1998, A. Orange 12098 (NMW – C.1999.011.74); near Llanfrothen, Dolfriog, near Corlwyni, 23/6146.4656, 2002, A. Orange 14153 (NMW – C.2003.002.141); Ganllwyd, Rhaeadr Mawddach, 23/7360.2757, 2011, A. Orange 20644 (NMW – C.2015.005.9); same locality and date, A. Orange 20649 (NMW – C.2013.001.126). V.C. 49, Caernarvonshire: Llanberis, Afon Hwch, 23/579.590, 1994, A. Orange 10058 (NMW – C95.38.121); east of Beddgelert, stream below Llyn Llagi, 23/6368.4890, 2011, A. Orange 20463 (NMW – C.2015.005.7); east of Beddgelert, Gelli Iago, 23/6334.4823, 2011. A. Orange 20443 (NMW – C.2015.005.6); same locality and date, A. Orange 20439 (NMW – C.2015.005.5). England: V.C. 4, North Devon: Belstone, River Okement, iv 1959, A. E. Wade (NMW 61.48.312). V.C. 57, Derbyshire: Goyts Clough, River Goyt, SK 01.73, xi 2010, S. Price (NMW–C.2015.005.10).
Discussion
The analysis of mtSSU sequences confirms that Porina rivalis is not closely related to P. aenea, P. austroatlantica, P. byssophila, P. chlorotica, P. lectissima or P. pacifica, some of the other saxicolous species with 3-septate ascospores, although it forms a well-supported clade with a single sequence from an unidentified collection from British Columbia. The analysis of the few available ITS sequences confirms that P. rivalis is not closely related to P. byssophila, P. guentheri var. lucens, P. interjungens, P. lectissima or P. pacifica. However, the phylogeny of the genus will only be resolved using a much broader sampling of taxa, and additional gene regions.
The acetone-insoluble pigments of the involucrellum are important in defining the new species. Such pigments have not been analyzed chemically, but a standard system of naming them, defined on the basis of their colour and colour changes in acids and alkalis, was proposed by Meyer & Printzen (Reference Meyer and Printzen2000). Four named pigments, two of them newly defined, were reported from Porina s. lat. by Hafellner & Kalb (Reference Hafellner and Kalb1995), but the writer has found these difficult to apply, with the exception of Porina-yellow. It is also difficult to be certain whether a colour seen in a section is a ‘different’ pigment to adjacent areas, or the effect of a mixture of pigments. The pigments found in each species of Porina need to be described carefully, and the formal names used only with caution.
I am grateful to Dr Ingrid Jüttner for allowing the use of water chemistry data, obtained as part of the ISAAC Liming Project.
