A new lineage of mazaediate fungi in the Eurotiomycetes: Cryptocaliciomycetidae subclass. nov., based on the new species Cryptocalicium blascoi and the revision of the ascoma evolution

The class Eurotiomycetes (Ascomycota, Pezizomycotina) comprises important fungi used for medical, agricultural, industrial and scientific purposes. Eurotiomycetes is a morphologically and ecologically diverse monophyletic group. Within the Eurotiomycetes, different ascoma morphologies are found including cleistothecia and perithecia but also apothecia or stromatic forms. Mazaediate representatives (with a distinct structure in which loose masses of ascospores accumulate to be passively disseminated) have evolved independently several times. Here we describe a new mazaediate species belonging to the Eurotiomycetes. The multigene phylogeny produced (7 gene regions: nuLSU, nuSSU, 5.8S nuITS, mtSSU, RPB1, RPB2 and MCM7) placed the new species in a lineage sister to Eurotiomycetidae. Based on the evolutionary relationships and morphology, a new subclass, a new order, family and genus are described to place the new species: Cryptocalicium blascoi. This calicioid species occurs on the inner side of loose bark strips of Cupressaceae (Cupressus, Juniperus). Morphologically, C. blascoi is characterized by having minute apothecioid stalked ascomata producing mazaedia, clavate bitunicate asci with hemiamyloid reaction, presence of hamathecium and an apothecial external surface with dark violet granules that becomes turquoise green in KOH. The ancestral state reconstruction analyses support a common ancestor with open ascomata for all deep nodes in Eurotiomycetes and the evolution of closed ascomata (cleistothecioid in Eurotiomycetidae and perithecioid in Chaetothyriomycetidae) from apothecioid ancestors. The appropriateness of the description of a new subclass for this fungus is also discussed.

Phylogenetic studies have contributed to unravel the evolutionary relationships within the class (e.g. Geiser et al. 2006;Wood et al. 2016;Réblová et al. 2017) and have been used to re-classify taxa accordingly. Recently, a new order and two subclasses have been described (Chen et al. 2015;Gueidan et al. 2014;Wood et al. 2016;Réblová et al. 2017) in the Section Editor: Gerhard Rambold * M. Prieto maria.prieto@urjc.es Eurotiomycetes. Current problems in the classification of families and orders within this class are the high amount of unknown or undescribed species (Chen et al. 2015), the high presence of species only known from their anamorph (Gibas et al. 2002a;Stchigel et al. 2013), un-assigned paraphyletic taxa (Gueidan et al. 2014) or limited availability of sequence data (Quan et al. 2020). Currently, Eurotiomycetes is generally accepted to comprise 5 subclasses: Chaetothyriomycetidae, C o r y n e l i o m y c e t i d a e , E u r o t i o m y c e t i d a e , Mycocaliciomycetidae and Sclerococcomycetidae (Dowell 2001;Geiser et al. 2006;Hibbett et al. 2007;Wood et al. 2016;Réblová et al. 2017). Eurotiomycetidae encompasses three morphologically diverse orders: Arachnomycetales, Eurotiales and Onygenales. Eriksson (1999) classified these orders in the Eurotiomycetes, but they were instead treated as Plectomycetes by Geiser and LoBuglio (2001). Members of Eurotiomycetidae produce different ascoma types including cleistothecia, ascostromata or gymnothecia (Currah 1985, Gibas et al. 2002a, Hibbett et al. 2007, Jaklitsch et al. 2016. In Eurotiomycetidae, asci are usually evanescent, sometimes bitunicate, scattered throughout the ascoma without hamathecium elements and with eight ascospores (Hibbett et al. 2007). They include mostly saprophytic species, and some produce toxic and economically important secondary metabolites, food fermentation enzymes, or are used as genetic models, as Aspergillus nidulans (Jaklitsch et al. 2016). Animal-associated fungi are also known (pathogens of vertebrates in Onygenales, Jaklitsch et al. 2016). Chaetothyriomycetidae includes the common black yeast fungi, some of which are pathogens of humans and animals, but also contains lichenized groups and endophytes (Gueidan et al. 2014;Chen et al. 2015). Within this subclass, members of Verrucariales, Chaetothyriales and Pyrenulales are characterized by having perithecial ascomata and bitunicate or secondarily prototunicate asci with a dehiscence ranging from fissitunicate to evanescent and presence of hamathecium structure, producing pseudoparaphyses or periphysoids (Barr 1983). Phaeomoniellales, sister to the clade including Verrucariales and Chaetothyriales, comprises mainly endophytes and plant pathogens with perithecial ascomata when sexual states are present (Chen et al. 2015). Mycocaliciomycetidae includes non-lichenized members of the families Mycocaliciaceae and Sphinctrinaceae (Mycocaliciales, Tibell and Wedin 2000). They produce stalked or sessile apothecial ascomata, and spore dispersal is active or passive when the ascomata develop a mazaedium. Asci are unitunicate, cylindrical, mostly with a distinctly thickened apex and with 8 spores. Members of Mycocaliciomycetidae are parasites or commensals on lichenized or saprotrophic fungi. The new subclass Coryneliomycetidae was introduced by Wood et al. (2016) for the Coryneliaceae. Most members of this group form pseudothecial mazaedial ascomata, with initially bitunicate asci in which the outer wall disintegrates (prototunicate) and the ascospores finally accumulate as a dry mass at the ascoma beak (Johnston and Minter 1989;Geiser et al. 2006); but not all are mazaediate (Garrido-Benavent and Pérez-Ortega 2015). They are mostly biotrophic on leaves and stems, especially on Podocarpaceae in the southern hemisphere with some northern temperate species that occur on conifers (Garrido-Benavent and Pérez-Ortega 2015). Although the Sclerococcum-Dactylospora lineage was recovered as a distinct lineage in the broad multigene phylogenies by Schoch et al. (2009) andChen et al. (2015), Sclerococcomycetidae was first formally recognized by Réblová et al. (2017) including in it the single order Sclerococcales. Later, Dactylospora was merged with Sclerococcum based on the close phylogenetic relationship of their type species (Diederich et al. 2018;Olariaga et al. 2019). Members of this subclass are characterized by apothecial ascomata with unitunicate, non-amyloid asci, covered with an amyloid or hemiamyloid gelatinous c a p . T h e h a m a t h e c i u m c o n s i s t s o f p e r s i s t e n t pseudoparaphyses. They are non-lichenized, terrestrial, marine, bryophytic, corticolous, lignicolous, lichenicolous or are associated with beetles as a part of their intestinal microbiota. Olariaga et al. (2019) also pointed out the existence of several dematiaceous phialidic fungi within this clade, several strains isolated from digestive tracts of beetles and a new aquatic hyphomycete. Thus, more hyphomycetes are likely to belong to the Sclerococcomycetidae.
Most higher-level Ascomycota systematics hypotheses are based on ascoma type (Nannfeldt 1932;Luttrell 1955;Henssen and Jahns 1974) and ascus morphology and dehiscence (Luttrell 1951;Eriksson 1981;Hafellner 1984). Ancestral character state reconstruction analyses can provide valuable insights to understand the evolution (Schmitt et al. 2009a;Joy et al. 2016) of groups that are prone to shifts in ecological and morphological character states as the Eurotiomycetes (Geiser et al. 2006;Schoch et al. 2009). Regarding ascomata, ancestral state reconstruction analyses support the ancestor of Ascomycota as producing apothecia and show multiple transitions from apothecioid to perithecioid and cleistothecioid ascomata (Schoch et al. 2009), as Nannfeldt (1932) previously hypothesized. In the Eurotiomycetes, the ancestor has been reconstructed as most probably having apothecia with independent origins of perithecioid ascomata in Chaetothyriomycetidae and Eurotiomycetidae (Eurotiomycetes) and cleistothecioid ascomata in Eurotiales and Onygenales (Schoch et al. 2009). Regarding mazaediate ascomata, ancestral state reconstruction analyses showed a high degree of parallel evolution with multiple independent gains of the mazaedium in the Ascomycota (Prieto et al. 2012). Concerning asci, the ancestor of the Eurotiomycetidae and Chaeothyriomycetidae is reconstructed as producing fissitunicate asci; however, the ancestral state of the most basal node of the Eurotiomycetes is not resolved (Schoch et al. 2009). This suggests the origin of eurotialean prototunicate, deliquescent asci from a fissitunicate ancestor (Geiser et al. 2006). However, new discoveries of novel lineages or the inclusion of new taxa in phylogenies is likely to change our understanding of the evolutionary history of this labile group of fungi concerning ascoma morphology.
The study of a new mazaediate calicioid species suggested a phylogenetic placement within the Eurotiomycetes. Thus, the main aims of this work were to describe this new taxon, to explore its phylogenetic relationships and to analyse the evolution of the ascoma type in the Eurotiomycetes.

Taxon sampling
Preliminar Blast searches (https://blast.ncbi.nlm.nih.gov/ Blast.cgi) suggested a placement of the new species within the Eurotiomycetes. Thus, in order to place the new species in the Eurotiomycetes, a larger sample of representatives of all subclasses and orders downloaded from GenBank was assembled (Table 1).

Morphological study
The macro-and microscopic description was made from fresh material and was completed by observing material soaked in water or in KOH 5%. Structures were measured from fresh material or from rehydrated material in H 2 O. Only mature spores discharged from asci were measured. Spore statistics were calculated for each collection based on measures of 25 spores. Abbreviations of statistics referring to ascospores are L m = mean length, W m = mean width and Q m = L m /W m . Mounting reagents used were water, Congo red in sodium dodecyl sulphate, Lugol's solution (IKI) and KOH 5%. Material is deposited in ARAN-Fungi, MAF herbaria (Thiers 2014), as well as in J. Etayo's private herbarium.

Culture observations
A dikaryotic culture of Cryptocalicium blascoi was obtained by depositing ascospores of a mazaedium on 2% malt extract agar (MEA; 2% malt extract, 2% agar-agar). The plate was sealed with laboratory film and incubated at room temperature. The culture was deposited at the Spanish Type Culture Collection (Spain, CECT). The identity of the culture was confirmed by comparing its nuITS region and sequences obtained from ascomata (Table 1).

Alignments and phylogenetic analyses
Sequences were aligned manually using AliView v.1.26 (Larsson 2014) and translated to amino acids in protein coding loci. Ambiguous regions (sensu Lutzoni et al. 2000) and introns were delimited manually and excluded from phylogenetic analyses. We also used MAFFT v. 7 (Katoh and Standley 2013) to align automatically and Gblocks 0.91b (Castresana 2000) to identify ambiguous regions. Since the maximum likelihood results were very similar between MAFFT-Gblocks and manually constructed matrices, the last ones were used for analyses. The combined alignment is available at TreeBASE (S28198). Individual gene regions were analysed using maximum likelihood-based inference (ML) in RAxML ver. 8.2.12 (Stamatakis 2014) with a GTRGAMMA model for tree inference and rapid bootstrapping with a GTRCAT model. Gene-tree incongruence was checked by comparing maximum likelihood bootstrap values (ML-BS) among individual gene trees. Clades were considered in conflict when a supported clade (bootstrap support >70%) for one marker was contradicted with

Evolution/ancestral state reconstruction for ascoma morphology
Although a high diversity of ascoma types is found in the Eurotiomycetes (Alexopoulous et al. 1996;Jaklitsch et al. 2016), states were coded as open (apothecioid), operculate (perithecioid) or closed (cleistothecioid) ascomata. A fourth character state was coded as absent when asci are not produced in a sporocarp or sexual morphs are unknown. We inferred ancestral states and traced the evolution of ascoma morphology, using different methodologies and the last 5000 trees that resulted from each run from the Bayesian analysis of the concatenated data set (20000 trees in total). Maximum likelihood and parsimony ancestral state reconstructions were performed in Mesquite 2.75 (Maddison and Maddison 2019) with the ML model Mk1 and equal   Gibas et al. 2004, 3 Gibas et al. 2002b, 4 Geiser et al. 2006, 5 Brasch and Graser 2005, 6 Untereiner et al. 2002, 7 Wilmotte et al. 1993, 8 Anderson et al. 1998, 9 Klinger et al. 2013, 10 Hinrikson et al. 2005, 11 Nikkuni et al. 1998, 12 Nierman et al. 2005, 13 Berbee and Taylor 1992b, 14 LoBuglio et al. 1993, 15 Liu et al. 1999, 16 Spatafora et al. 200617 Untereiner and Naveau 1999, 18 Bhattacharya et al. 2000, 19 Berbee 1996, 20 Lutzoni et al. 2004, 21 Lindemuth et al. 2001, 22 Reblova et al. 2013, 23 Prieto et al. 2013  probability for any particular character change. To account for topological uncertainty, we used the "trace character over trees" option that summarizes the ASR over a series of trees. We reconstructed the ancestral states using the Bayesian approach described by Pagel et al. (2004) and Pagel and Meade (2006), implemented in BayesTraits v. 3.02 (www.evolution. rdg.ac.uk). For this purpose, we used the same posterior tree sample as for the maximum likelihood and parsimony ancestral state reconstructions, with an exponential prior with the mean drawn from a uniform hyperprior on the 0-10 interval. The MCMC was ran for 100M generations, sampling every 1000 generations. The first 10M generations were discarded as burn-in.

Results
Based  Etymology: named after Javier Blasco Zumeta, outstanding Aragonese naturalist, who showed to us the first locality where this species was found and who provided us with further material of it.

Comments
Cryptocalicium blascoi is probably the tiniest known calicioid fungus. The unusual ecology on the underside of bark strips of Cupressaceae, the presence of a mazaedium, the clavate hemiamyloid asci, hamathecial filaments and the dark violet pigment granules that turn blue-green in KOH make C. blascoi unique. Our attempts to morphologically identify C. blascoi with the existing literature on calicioid fungi (e.g. Tibell 1999) failed, and molecular data confirmed that C. blascoi does not group in any lineage of known calicioid fungi. The clavate asci, with a pedicel, bitunicate and evanescent, are consistent with the phylogenetic position of C. blascoi between the Eurotiomycetidae and the Coryneliomycetidae, both having these characteristics as well.
The hemiamyloid reaction of asci and the presence of hamathecial filaments, however, have not been cited for any species in those groups. So far, C. blascoi has been found in areas with large ancient Juniperus oxycedrus and J. thurifera trees in continental Mediterranean areas. In central Spain (Ávila, Madrid, Soria, Toledo), C. blascoi has been detected in almost every large J. oxycedrus tree sampled in search for it. Although C. blascoi can be very hardly observed in nature even using a hand lens of 10-20 ×, it can be collected by gathering loose bark strips and by scrutinizing those under the dissecting scope. Cryptocalicium blascoi grows directly on the inner part of the bark and occurs often on blackened zones. We have observed several times that a few ascomata grew on solid resin drops. Our efforts to find C. blascoi in areas with higher precipitation rates or where large ancient Cupressaceae do not exist (Huesca, Navarre) failed so far. Thus, it is possible that Fig. 1 Best maximum likelihood tree obtained from RAxML based on a 7-locus data set (5.8S nuITS,nuLSU,nuSSU,mtSSU,MCM7,RPB1 and RPB2). Nodes show bootstrap support (ML-BS) from maximum likelihood, and posterior probabilities (PP) obtained in the Bayesian analysis, ordered as ML-BS/PP. ML-BS below 60% and PP below 0.85 are not shown. Results from the ancestral state reconstruction with BayesTraits are shown in the studied nodes. Ascoma types are also indicated except for those species devoid of sporocarps or with unknown sexual morphs Fig. 2 Cryptocalicium blascoi (holotype). A Type locality. B Juniperus oxycedrus tree with loose bark strips. White arrows indicate suitable places for C. blascoi. C Gregarious ascomata. D Young ascoma with unexposed hymenium. E Older ascoma with exposed hymenium. F-G Ascomata with mazaedia. H Old ascoma without mazaedium (Etayo 31798). Scale bars = 100 μm. Photographs I. Olariaga C. blascoi is restricted to well-preserved stands of ancient Juniperus and Cupressus. Due to the minute size of C. blascoi and to the fact that one of the collections was made on Cupressus sempervirens, it is likely to have been overlooked and to have a considerably wider distribution.

Discussion
The phylogenetic position of the new species constitutes a further example of the high ecological and morphological diversity of the Eurotiomycetes. Other members of Eurotiomycetes with mazaediate and stipitate open ascomata belong in the Mycocaliciomycetidae (Tibell and Wedin 2000), which mainly differ from C. blascoi in having cylindrical asci without iodine reactions. Within Coryneliomycetidae some species also have mazaediate ascomata, but in these, the spores accumulate in perithecial beaks that arise from a stromatic tissue (pseudothecia) (Garrido-Benavent and Pérez-Ortega 2015; Wood et al. 2016). Within both Eurotiomycetidae and Chaetothyriomycetidae, several genera such as Onygena, Pseudotulostoma, Pyrgillus or Trichocoma also include mazaediate-perithecioid or cleistothecioidmembers. Ancestral state reconstruction analyses have shown that several independent gains of the mazaedium have occurred in Ascomycota and that this character is highly homoplastic (Prieto et al. 2012). Regarding ecology, many Eurotiomycetes colonize dead plant tissues and living animals (Jaklitsch et al. 2016;Quan et al. 2020). Eurotiomycetidae includes mostly saprophytic species occurring in various substrates like wood, compost, dung, decaying plant material or fo od st uff s ( Ja kli tsc h et al. 2 01 6) . M em be rs of Coryneliomycetidae are mostly biotrophic on leaves and stems of Podocarpaceae (Garrido-Benavent and Pérez-Ortega 2015; Wood et al. 2016). Mycocaliciomycetidae includes saprophytic species on bark, wood or lichenicolous (Jaklitsch et al. 2016), whereas Sclerococcomycetidae comprises also corticolous and lignicolous species (Réblová et al. 2017). Thus, the ecology of C. blascoi is not surprising as some species of related lineages share similar trophic modes (i.e. saprophytic).
The sister relationship of Cryptocaliciomycetidae with the rest of Eurotiomycetidae and Coryneliomycetidae opens the possibility of including both Coryneliomycetidae and Cryptocaliciomycetidae within the Eurotiomycetidae or of describing a new subclass (Cryptocaliciomycetidae) following the same criteria as in Wood et al. (2016). These authors  introduced Coryneliomycetidae to encompass the Coryneliaceae and argued its unique position in the Eurotiomycetes and its morphology. They underlined the presence of pseudothecial mazaedial ascomata containing initially double-walled asci, with the outer layer deliquescing as opposed to Eurotiomycetidae (usually with cleistothecia/ gymnothecia and prototunicate asci). The newly described species C. blascoi differs also from members of Eurotiomycetidae in having apothecia and presence of hamathecium and hemiamyloid asci and furthermore represents a genetically distinct lineage.
Ascoma types including open (apothecioid) or closed (perithecioid, cleistothecioid) forms have traditionally been used as key paradigms for ascomycete classification (Schmitt et al. 2009a). Molecular phylogenies show that ascoma evolution is complex, with multiple phylogenetic origins, and that ascoma type is an inappropriate character to circumscribe classes (Schmitt et al. 2009a;Schmitt 2011). Stchigel and Guarro (2007) underlined the high diversity of ascoma types in the Eurotiales: true cleistothecia (e.g. Chaetosartorya, Dichotomomyces, Eurotium, Hemisartorya, Neosartorya and Sclerocleista); asci borne in hyphal masses or tufts (e.g. Byssochlamys, Dendrosphaera, Sagenoma, Talaromyces and Trichocoma); asci sitting on a stroma (e.g. Eupenicillium, Hamigera, Hemicarpenteles, Neocarpenteles, Penicillliopsis, Thermoascus and Warcupiella); cleistothecia produced in a stroma or surrounded by a mass of Hülle cells (e.g. Cristaspora, Dichlaena, Emericella, Fennellia and Petromyces) or naked asci without an ascoma (e.g. Edyuillia, syn. Eurotium). Moreover, in the case of Chaetothyriomycetidae, both ascolocular and ascohymenial developmental types of ascomata exist in Pyrenulales, Verrucariales and Chaetothyriales (Schmitt 2011). However, it is unknown how these ascoma types have evolved, and regardless of their ontogeny, these types may be considered as closed ascomata (perithecioid and cleistothecioid). Our results show that all these ascoma types have evolved from an open ascoma type and support the use of the ascoma gross morphology to define (together with additional characters) subclasses within Eurotiomycetes. Stchigel and Guarro (2007) also stated that taxonomic schemas based on a few morphological characters have proved to be unstable and suggested that an appropriate approach should reflect a natural classification following the evolutionary relationships between the considered organisms. Our ancestral state reconstruction analyses of the ascoma type and the differences with the rest of members of Eurotiomycetidae support the decision of describing new subclasses for both Coryneliomycetidae (Wood et al. 2016) and Cryptocaliciomycetidae, thus reflecting the natural evolution within Eurotiomycetes.
Concerning ascoma evolution, our results, as those by Schoch et al. (2009), support that the ancestor of Eurotiomycetes produced an open ascoma. It is shown here that this character is more stable in Chaetothyriomycetidae, Sclerococcomycetidae and Mycocaliciomycetidae than in the clade including Eurotiomycetidae, Coryneliomycetidae and Cryptocaliciomycetidae. Molecular data have shown that cleistothecia originated several times within the Sordariomycetes-a group producing predominantly perithecioid ascomata-through the loss of the ostiolar canal (Berbee and Taylor 1992a;Rehner and Samuels 1995;Suh and Blackwell 1999). Within the Eurotiomycetes, Schoch et al. (2009) also suggest that cleistothecia have evolved from perithecia. Our results disagree in this respect as it is here suggested that both perithecioid and cleistothecioid forms h a v e a r i s e n f r o m a p o t h e c i o i d a n c e s t o r s i n t h e Eurotiomycetes. Within Lecanoromycetes (Schmitt et al. 2009a) and Leotiomycetes (Schoch et al. 2009), perithecia and cleistothecia have evolved independently several times from ancestors producing apothecia. All in all, it seems that the evolution of different types of ascomata shows different patterns in different groups of Ascomycota.
Code availability Not applicable.
Author contribution MP, IO and JE designed the work; MP, IO and JE analysed the data; and MP, IO and JE contributed in writing the manuscript.
Funding Open access funding by Swedish Museum of Natural History. The first author acknowledges grant dha 2016-23 4.3 from the Swedish Taxonomy Initiative (Svenska artprojektet) administered by the Swedish Species Information Center (ArtDatabanken).

Conflict of interest The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.