An overview of the genus Coprotus (Pezizales, Ascomycota) with notes on the type species and description of C. epithecioides sp. nov.

Abstract In a mycological research performed in the Sjeverni Velebit National Park, Croatia, a new species of Coprotus was discovered, described here as C. epithecioides. Along with the microscopic examination, phylogenetic analysis of the type material, based on ITS and LSU sequences, was performed in order to evaluate its relationship with the type species, C. sexdecimsporus. The type species was sequenced in this study for the first time, providing ITS and LSU sequences from two separate collections which displayed differences in macroscopic characters and content of paraphyses. An extended description of C. sexdecimsporus based on Croatian material is also provided. A worldwide identification key to the species assigned to the genus Coprotus is presented, along with a species overview, containing a data matrix. The phylogenetic position of Coprotus in the Boubovia-Coprotus clade within Pyronemataceae s.l. is discussed. Coprotus sexdecimsporus is also reported here as new to the Croatian mycobiota.


Introduction
The name Coprotus Korf was first mentioned but not validly published by Korf (1954) as a seggregate of the genus Ascophanus Boud. (Boudier 1869) for species having iodine negative asci, hooked paraphyses and small guttulate spores. Kimbrough (1966) recognized a "Coprotus group" in Ascophanus Boud. with species that have iodine negative asci staining uniformly in Congo red and ascospores with de Bary bubbles. The genus Coprotus Korf & Kimbr. was validated by Kimbrough and Korf (1967), encompassing certain species of Ascophanus and Ryparobius Boud. (Boudier 1869), with Coprotus sexdecimsporus (P. Crouan & H. Crouan) Kimbr. & Korf chosen as the type species. Eckblad (1968) implied that Leporina Velen. (Velenovský 1947) should be the correct name instead of Coprotus, since the type specimen of Leporina multispora Velen. was found to be identical to Ryparobius sexdecimsporus (P. Crouan & H. Crouan) Sacc. This nomenclatural problem was elaborated by Kimbrough (1970), who concluded that the name Leporina should be rejected and Coprotus retained because the type material consists of mixed collections belonging to three different genera while the protologue contains "two or more entirely discordant elements". The name Coprotus was put on a without-prejudice list of generic names of fungi for protection under the International Code of Nomenclature for algae, fungi and plants (Kirk et al. 2013).
Species of the genus Coprotus are characterised by oblate to lenticular or discoid, glabrous, translucent or whitish to yellow apothecia with coprophilous ecology. Asci are functionally operculate, non-amyloid, eight-to 256-spored, producing hyaline, smooth, eguttulate ascospores, containing gaseous inclusions referred to as de Bary bubbles when placed in anhydrous conditions. Paraphyses are filiform, mostly bent to uncinate and/or swollen at the apex, hyaline or containing pigment. The excipulum is composed primarily of globose to angular cells (Kimbrough et al. 1972).
The genus Coprotus was placed in the tribe Theleboleae (family Pezizaceae) by Kimbrough and Korf (1967). In later classifications Eckblad (1968) and Kimbrough et al. (1972) placed this genus into the family Thelebolaceae (Pezizales). Kish (1974) performed cytological and ontogenetical research on C. lacteus (Cooke & W. Phillips) Kimbr., Luck-Allen & Cain using axenic cultures, and concluded that this species shows much closer affinities with the Pyronemataceae sensu Eckblad (1968) than the Thelebolaceae. Study of the apical apparatus in C. winteri (Marchal & É.J. Marchal) Kimbr. and C. lacteus by Samuelson (1978) supported this view. Using transmission electron microscopy, Van Brummelen (1998) determined that the fine ascal structure of the wall and operculum in C. lacteus is characteristic of members of the Pyronemataceae s.l. Contrary to the mentioned micromorphological and cytological evidences, all members of the Thelebolaceae, including Coprotus, were placed in the class Leotiomycetes (Kirk et al. 2008, Lumbsch andHuhndorf 2010).
The phylogenetic affinity of Coprotus was studied using molecular data by Hansen et al. (2013), who showed that the genus belongs to the Pezizomycetes and forms a strongly supported monophyletic group with Boubovia Svrček (Pyronemataceae). This was confirmed by Lindemann et al. (2015) and Lindemann and Alvarado (2017). Wijayawardene et al. (2017) placed the genus Coprotus in the family Ascodesmidaceae (Pezizales, Pezizomycetes), and included 29 species. Additionally, isozyme analysis performed by Suárez et al. (2006) and RAPD patterns analysed by Ramos et al. (2008) detected a high intra-specific homogeneity in three Coprotus species (C. lacteus, C. niveus and C. sexdecimsporus). Furthermore, the AFLP fingerprinting technique applied to the same three Coprotus species (Ramos et al. 2015) exhibited the highest level of intraspecific variability in C. sexdecimsporus.
We began our own study of the genus Coprotus with an integrated taxonomical approach aimed at the type species, relying on vital taxonomic and phylogenetic methods. Previously only C. ochraceus (P. Crouan & H. Crouan) J. Moravec was included in phylogenetic analyses , Lindemann and Alvarado 2017. Our inventory study of fungi in the Sjeverni Velebit National Park was aimed also on fimicolous fungi resulting with a collection of a Coprotus species found on a chamois dung, Rupicapra rupicapra, that appeared different from all other known species in the genus.

Ex situ monitoring
The apothecia collected with the substrate were used for microscopic studies and DNA extraction. The remaining material (together with the original substrate) was kept in closed plastic boxes in a refridgerator under low temperature (4-8 °C) and out of doors (ca. 10-25 °C) in dark and in diffuse sunlight conditions. Over a two month period these were monitored observing a turnover of two to several generations.

Microscopic studies
Observations of apothecia were made using a stereomicroscope under magnifications up to 80×. Microscopic characters based on living cells and tissues ( * ) were recorded using vital taxonomy methods (Baral 1992), while those based on dead cells and tissues ( † ) were obtained from fixed fresh material. All described microscopic elements were observed in tap water (H 2 O); cytochemical and histochemical data were obtained using the procedure described by Kušan et al. (2015). Microscopic features were observed with transmission light microscopes (bright field, phase contrast and dark field) under magnifications up to 1000×. Drawings were made free hand to scale, and microphotographs were mostly taken with a DSLR camera mounted on the microscope's trinocular tube. Characters of apothecial construction and hymenial elements were based on a minimum of five ascomata. Spore measurements were based on samples of 50 fully mature, normally developed, and randomly selected ascospores (from living material ejected from asci). Measurements were taken directly using an ocular micrometer and from microphotographs using PIXIMÈ-TRE software ver. 5.9 (Henriot and Cheype 2017) to an accuracy of 0.1 µm. Spore wall layers were named following Heim (1962), except perispore is used rather than exospore following Harmaja (1974). Length, width and length/width ratio ("Q" value) are given as: min. -stat. mode -max. where "min." = minimum (lowest measured value), "stat. mode" = statistical mode, "max." = maximum (highest measured value). Length/width ratio (without mode value) was also introduced for asci. Dried material and accompanying data for all treated collections were deposited at the Croatian National Fungarium (CNF) in Zagreb.
A dichotomous key for identification of all putative species of Coprotus is presented. It was compiled from data derived from the literature and from our own observations. The key, except in one case, contains data for both living and dead materials. In this way the key is comprehensive. Species/character overview tables, containing supplementary data not used in the key, are presented as an aid for reliable identification (Tables 2-6). Ascus and ascospore measurements, originating from published sources, are enhanced by those obtained by measuring the original microphotographs and drawings. Ascus and ascospore "Q" values, taken from published references, were calculated from the original microphotographs and drawings.

DNA extraction, PCR amplification and DNA sequencing
Total genomic DNA was extracted from samples using DNeasy Plant kit (Qiagen Inc., USA). The LSU sequences were amplified using primers LR0R and LR7 (Vilgalys and Hester 1990). The primers ITS1-F (Gardes and Bruns 1993) and ITS4 (White et al. 1990) were used for amplification of the ITS regions (ITS1-5.8S-ITS2). All PCR amplifications consisted of 25-µL reaction volumes containing 0.2 mM of each dNTP, 0.2 µM of each primer, 1 U of Taq polymerase, 1.5 mM of MgCl and ~ 50 ng DNA. The touch-down PCR cycling profile consisted of initial 5 min at 95 °C, 10 cycles of 45 s at 95 °C, 45 s at 60 °C (decreasing 1 °C/cycle), 90 s at 72 °C, 25 cycles of 45 s at 95 °C, 45 s at 52 °C, 90 s at 72 °C, with final extension of 7 min at 72 °C. PCR products were sequenced in both directions using the same primers as for PCR by Macrogen (Macrogen Inc., Seoul, Korea). All sequences were deposited in GenBank (Table 1).

Phylogenetic analyses
The ITS + LSU alignment consisted of 1590 characters including gaps, of which 763 were conserved, 777 were variable, and 230 were parsimony informative. The LSU alignment consisted of 894 characters including gaps, of which 32 were conserved, 319 were variable, and 224 were parsimony informative. The type species Coprotus sexdecimsporus was sequenced for the first time to ascertain the real phylogenetic position of the genus Coprotus. The two phylogenies (based on LSU, and concatenate analysis of LSU and ITS) firmly nested the Coprotus species in the order Pezizales, as a member of the Boubovia-Coprotus lineage inside the Pyronemataceae s.l., in a species group next to the Geopyxis-Tarzetta and Ascodesmis-Pulvinula clades (but without high support in our contracted analyses, Figs 1, 2). In both phylogenetic trees, species in the genera Boubovia and Coprotus were clustered together, with high support values. Coprotus ochraceus showed a distant relationship to the type species C. sexdecimsporus as a phylogenetically earlier diverging lineage. Our newly described species appeared closely related to the type species. The two collections of C. sexdecimsporus sequenced displayed 100% sequence identity (ITS and LSU). As presently circumscribed, the genus Coprotus is clearly characterised by the following combination of characters: obligate coprophilous ecology, eugymnohymenial apothecial development, apothecia with reduced marginal tissue without setose hairs; inamyloid asci uniformly stainable in CR, with functional operculum; prolate, smooth (under transmission light microscope), eguttulate ascospores in all developmental stages sporoplasm of which have strong affinities to form de Bary bubble in any anhydrous conditions (especially in media such Cotton Blue). Mature spores ejected from living asci possess temporary thick and gelatinous sheath. Anamorph not known.  µm, * Q = 4.1-5.6, significantly shorter and more clavate at the marginal rim, when mature * protruding above hymenium up to 26 µm, pars sporifera * 47.3-63.3 µm, 16-spored, hyaline, base attenuated, bifurcate, arising from perforated croziers, only fully mature asci with flat lentiform operculum clearly delimited prior the spore discharge, * 6.6-8 µm in diam. and * 0.6 µm thick, lateral wall 3-layered, * 0.7-0.8 µm thick, after spore discharge operculum as a rule clearly visible; in IKI inamyloid; in CR outermost wall vividly rutilered throughout the ascal length, median layer pale rutile-yellow, innermost layer greyish; in CB cyanophobic. Ascospores * 10.7-11.7-13.8 × 6.8-7.9-8.5 µm, * Q = 1.4-1.7-1.7, ellipsoid to narrowly ellipsoid and most often radially symmetrical, with rounded-obtuse poles, rarely slightly bilaterally symmetrical with one side somewhat less convex but never flattened, 1-celled, hyaline; in living asci bi-to triseriate; when freshly ejected remain in a single group for a while due to the delicate sticky sheath enveloping individual spores; surface smooth; wall 3-layered, 0.6-0.7 µm thick, perispore dull, epispore brightly refractive, endospore layer with pale greyish-isabelline refractivity; in IKI no notable differential stainings; eguttulate, uninucleate, nucleus ±centrally to unipolarly positioned, 2.7-3 µm wide, in CRB nucleus and sheath more contrasted, perispore dull deep bluish-violet/deep cyan, epispore CRB-, endospore purplish lilac/medium violet; after applying KOH spore sheath dissolves instantly, all structures discoloured, perispore not loosening, endospore layer purplish-rosaceous; in CR perispore dull, not stained as epispore, but endospore lilac reddish; in AC completely devoid of staining; in CB de Bary bubbles present only in mature spores, perispore not loosening, weakly cyanophilic. Paraphyses cylindric, apically obtuse to subclavate, always slighty bent to uncinate, densely septate, rarely simple but often richly branched in the upper part; apically producing abundant medium to strongly refractive golden-yellow to cinnamon-yellow granular exudate, in IKI copper orange, in CRB dark grey blue, after applying KOH rubis red-grey; apical cells * 6.9-16.4 × 2-3.4 µm, † 1.4-2.8 µm wide, wall thin and hyaline, cells in the upper half contain minute medium to strongly refractive hyaline globules * 0.2-1 µm wide or in pigmented apothecia with beer-yellow to beer-orange scattered dotted granules which are in IKI greyish green, in CRB deep purplish-lilac to deep violet; in CB wall cyanophobic, cytoplasm weakly cyanophilic. Subhymenium only slightly differentiated from medullary excipulum, * 12-19 µm thick, composed of hyaline textura globulosa-angularis, cells * 3.8-7.5 µm wide. Medullary excipulum hyaline, in the middle flank * 12-22 µm thick, composed of textura porrecta, cells runing parallel to the surface, * 1.4-4.8 µm wide. Margin subhyaline, fairly reduced to a thin cellular zone * 9.6-11.3 µm thick at ½ of hymenium height, composed of small celled textura angularis 1-2 cell thick, cells clavate or elongated angular, 2.4-8.8 µm wide, marginal rim composed of prismatic terminal cells which do not protrude above hymenium; in CB cell walls strongly cyanophilic. Ectal excipulum hyaline, in the middle flank * 48-56 µm thick, composed of textura globulosa, cells * 7.2-16 µm wide, walls yellowish; in IKI some cells with visible moderate accumulations of glycogene; in CB cell walls slightly cyanophilic; in AC cell walls and cytoplasm deeply lilac. Overall excipulum devoid of crystalline matter, without colouring in KOH, in IKI completely inamyloid. Anamorph not found. Distribution and ecology. The species has a cosmopolitan distribution and can be found on dung of various wild and domestic animals, mainly herbivores (especially ruminant animals and rodents). In the temperate zone it is distributed in the habitats from maritime to alpine zones. Notes. De Sloover (2002) summarises the data on the distribution of pigments in microscopic elements in the Coprotus species described up to that time. His overview suggests that paraphyses are not the only cause of the overall apothecial pigmentation. However, our detailed study on living material of C. sexdecimsporus over a period of two months clearly showed that cytoplasmic pigments in the paraphyses develop with exposure to light. These observations used apothecia on original substrate and were carried out under controlled conditions. The pigments developed under sunlight or artificial light with a sufficient amount of the ultraviolet wave-length. On the other hand, pigmentation was completely absent if apothecia were grown continually under dark or low-light conditions. There is considerable variability in ascospore dimensions given in the literature. Although it seems that ascospore length may vary regardless of any presently visible cause, the ascospore diameter seems to be smaller in material from the Southern Europe / Mediterranean region. Accordingly, material from Italy (Doveri 2004) and Tunisia (Häffner 1996), almost completely overlap with our studied material from the East Adriatic region. These are in the range of ascospore widths from 6.9-8.5 µm. Specimens from the European Atlantic (Crouan's material restudied by Le Gal, 1960), Norway (Aas, 1983) and both Americas (Kimbrough et al. 1972, Dokmetzian et al. 2005 have spores with greater spore widths, ranging from 7.5-10 µm. These differences might point to some ecologicalgeographical causes. The type material is missing according to Kimbrough et al. (1972). Etymology. The specific epithet refers to epithecium-like ascal protective formation composed of swollen apical paraphyses cells.
Distribution and ecology. The species is known so far only from Mt. Velebit, Croatia. The only collection originates from chamois dung in the alpine karstic habitat.
Other specimens examined. None. Notes. Coprotus epithecioides has several characters making it distinct from other species in the genus. The paraphyses are of two types, along with the usual filiformclavate ones, there are also an abundance of those with very short, swollen apical cells, that mutually form an epithecioid protective layer over immature asci, a character not recorded so far in the genus Coprotus. Additionally, in the epithecioid type, 1-2 subapical cells are often also swollen. This gives the paraphyses a moniliform appearance. When present, paraphysal pigments are most often orange to reddish-orange and crystalloid, i.e. of fibrillar shape, resembling the carotenoid pigmentation of Scutellinia species. Spores are highly bilaterally symmetric compared to C. glaucellus, C. subcylindrosporus, C. argenteus and C. sexdecimsporus (which has only inconspicuously and partly bilaterally symmetric spores) and the spores are significantly shorter than those of C. subcylindrosporus, C. argenteus and C. sexdecimsporus. Coprotus glaucellus differs by the presence of only apically uninflated to subclavate paraphyses which do not form an epithecioid protective cover over immature asci. Also it has notably elongated cells at the marginal edge. As elaborated above, paraphysal cytoplasmic pigments normally also develop in this species if the fungus is strongly exposed to sunlight or artificial light with ultraviolet wave-lengths. The pigmentation is completely absent if the apothecia is grown continually under dark or low-light conditions (see notes under C. sexdecimsporus).

Discussion
Together with the newly described species, 29 species are currently accepted in the genus Coprotus. One species is published invalidly (Häffner 1996), while four misapplied species concepts were recognized in our study and considered as separate taxonomic entities: Ascophanus aurantiacus Velen. (Velenovský 1934, Svrček 1976, which is erroneously synonymised by Kimbrough et al. (1972)  In this, our first contribution to the knowledge of the genus Coprotus, we aimed to ascertain the exact phylogenetic position of the genus, bearing in mind that the type species C. sexdecimsporus had not previously been sequenced. We also undertook to determine the variability in colour noted in this species. To do this a typical nonpigmented sample of C. sexdecimsporus and a pigmented 16-spored Coprotus collection were analysed using molecular and vital taxonomic methods. The non-pigmented C. sexdecimsporus and the pigmented form proved to be the same species with 100% bp identity, showing that the apothecia of C. sexdecimsporus may be pigmented or not. The same behaviour regarding pigmentation was also recorded in the newly described C. epithecioides by performing the same light-test procedure through prolonged monitoring of apothecia on original substrate. The apothecia of both C. sexdecimsporus and C. epithecioides, fully grown in dark first, were devoid of any notable pigmentation in the paraphyses, while new generations of apothecia started to develop pigment granules soon after exposure to sunlight or artificial light rich in UV radiation. This would indicate that future testing along these lines on other species in the genus would be fruitful and informative in further developing the identification key. All Coprotus keys published so far, that containing significant numbers of species (Kimbrough et al. 1972, Aas 1983, Prokhorov 1998, Doveri 2004, Melo et al. 2015 use paraphysal and apothecial pigmentation that we show are unstable/unreliable. Phylogenetic analyses of both forms of the type species confirmed the position of the genus Coprotus in the order Pezizales, inside a large species group of the Pyronemataceae s.l., placing the Coprotus-Boubovia lineage next to the Ascodesmis species group but without high support in our contracted analyses (cf. also Hansen et al. 2013, Lindemann and Alvarado 2017. In our study C. epithecioides clustered in the Coprotus core group (sister to the type species). Our analysis confirmed that both eight-spored and multispored (in our case 16-spored) species belong in the genus Coprotus (cf. Hansen et al. 2013).
Previously only C. ochraceus was included in phylogenetic analyses (cf. , Lindemann and Alvarado 2017. In our analyses, this species clearly falls outside both the Coprotus core group and the group containing putative members of the genus Boubovia (Figs 1, 2). The isolated position of C. ochraceus is furthermore supported by the detailed re-examination of Crouan's material by Le Gal (1960), who managed to observe several to many granules inside the sporoplasm that could not represent de Bary bubbles, a feature that is absent in all other known Coprotus species. However, paraphyletic relationship of analysed members of Boubovia should be clarified in future studies with more species and more DNA regions included. A number of Coprotus species (but not C. ochraceus) that we have studied so far in detail, including the type species C. sexdecimsporus and the new species C. epithecioides, did not possess any refractive granular / guttulate content in the sporoplasm at any developmental stage (see also Kimbrough 1966, Kimbrough andKorf 1967). All known species of Coprotus are obligatory fimicolous (cf. Doveri 2011). Those species in the closely related genus Boubovia, that were included in our phylogenetic analyses, placed next to each other (Figs 1, 2), are principally found on other types of substrate (dump soil, pebbles, litter and decayed organic material), and their ascospores possess internal guttules, at least during the early stages of development (Svrček 1977, Yao andSpooner 1996). The present study implies the necessity for further phylogenetic studies of more Coprotus collections and species (reliably identified), as well as more DNA regions. Until more research is done, we restrict the genus to strictly fimicolous species, the spores of which are smooth under the light microscope, and are devoid of any internal refractive granular content at any developmental stage. Also, freshly ejected ascospores of all the species analysed by us possessed thick and sticky temporary sheaths in the living state, a rarely reported, but important character, also detected by Le Gal (1960). An example of the importance of such a character in generic characterisation is the encapsulating, rather firm spore sheath present in the genus Paratricharina Van Vooren, U. Lindemann, M. Vega, Ribes, Illescas & Matočec (VanVooren et al. 2015) but absent from almost all pezizalean genera.
Since the need for the standardisation of defining taxonomic characters (especially spore shapes) is already elaborated in Kušan et al. (2014), we tested the shape of the asci as a useful taxonomic character too. The asci of the genus Coprotus vary considerably in both shape (from broad clavate to narrow cylindric) and size (38-210 × 6-55 µm) (Table 4). However, individual species in this genus mostly possess asci with comparatively little variation in size and shape. This prompted us to introduce a standardisation of ascus shape types and length/width ratio ("Q" value) for describing asci, in order to enhance differentiation between Coprotus species. Ascus shape types were grouped in the current study into three series, defined by the position of its broadest point and "Q" value: clavate, cylindric and fusiform (see explanation under the Table 4). Baral (1992) observed that considerable alterations in quantitative taxonomic characters between dead and living cells exist in Ascomycota, due to the turgor loss causing cell shrinkage (especially in hymenial elements). This phenomenon, resulting in significantly lower measurements in dead cells, was recorded during the current study in ascal length and width (frequently with altered length/width ratio), and paraphysal width in all Coprotus collections studied in the living state. Therefore, great care should be taken when measuring the asci and paraphyses in order not to mix up the measurements of living and dead cells. On the other hand, ascospores in Coprotus showed little quantitative alteration. This can be explained by rigid spore walls and the capability of the sporoplasm to reversibly reduce its volume (caused by loss of cytoplasmic water) by forming gaseous de Bary bubble without significant cell shrinkage. This behaviour is not only characteristic to the genus Coprotus, but also to other phylogenetically closely related genera such as Boubovia (cf. Kristiansen and Schumacher 1993) and Lasiobolus Sacc. (cf. Kimbrough and Korf 1967). The ascospores of a number of more distantly related fungi usually possess pliant and thin walls, that easily irreversibly collapse unilaterally, together with the sporoplasm (e.g. Peziza, Iodophanus or Morchella), or both the wall and the sporoplasm irreversibly shrink, decreasing the ascospore's size ±evenly in all parts (numerous species of Helotiales), as shown diagrammatically in Baral (1992).
We recommend that future studies of newly collected material of Coprotus include careful observations of microscopic characters in the living state, especially in cases of rare and potentially new species, for the following reasons: (1) Living mature asci, besides representing a valuable standard for measurement and shape definition, may with proper orientation display useful characteristics related to the dehiscence apparatus as it appears immediately before spore ejection. This is also the case if living material is directly fixed with CB ( Fig. 5a) or CR; (2) Freshly ejected ascospores are normally at a uniform ontogenetic, mature stage, structurally complete and presumably viable, thus in this condition they represent a valuable standard for measurement, vital staining and description of structural features. Spores shape is unaltered because they are fully hydrated. This allows the differentiation of bilateral symmetry from those spores that may appear to have bilateral symmetry because of collapse due to the loss of turgor. We repeatedly recorded this alteration not only in this genus but throughout different pezizalean taxa; (3) A spontaneous (natural) spore discharge from living mature asci enables the monitoring of the presence and properties of the ascospore sheath. This structural detail can be of great help in taxonomical studies of every single species putatively assigned to the genus Coprotus, as well as to related taxa. It is already known that the presence or absence of such structures represents important taxonomic information in a number of ascomycetous taxa; (4) Both the paraphysal internal pigmentation and the exudate may disappear in older dried material. Observation of shrunken paraphysis tips on dead material minimises the difference among a number of species. All the abovementioned characters, are only visible in the living state. However, they can be easily recorded (e.g. microphotography) for future use from every fresh and viable collection.