Mycologia, 105(1), 2013, pp. 112–124. DOI: 10.3852/12-136 # 2013 by The Mycological Society of America, Lawrence, KS 66044-8897 Phylogenetic analyses of Coprinopsis sections Lanatuli and Atramentarii identify multiple species within morphologically defined taxa La´szlo´ G. Nagy1 ically indistinguishable lineages within C. lagopus, Department of Microbiology, Faculty of Science and which included C. lagopus var. vacillans, an ephem- Informatics, University of Szeged, Ko¨ze´p fasor 52, eral, developmental variant. Morphological traits H-6726 Szeged, Hungary supporting the inferred clade structure are discussed. Dennis E. Desjardin Three new taxa (C. fusispora, C. babosiae, C. villosa), Department of Biology, San Francisco State University, and one new combination (C. mitraespora)are 1600 Holloway Avenue, San Francisco, California proposed. 94132 Key words: Coprinopsis cinerea, C. lagopus, Copri- Csaba Va´gvo¨lgyi nus sensu lato, phylogeny, taxonomy Department of Microbiology, Faculty of Science and Informatics, University of Szeged, Ko¨ze´p fasor 52, INTRODUCTION H-6726 Szeged, Hungary Taxonomic research on species of Coprinus s.l. has Roger Kemp played a pivotal role in forming our views on species 4 Grove End, Lasswade, Midlothian, EH18 ILJ, Scotland, UK concepts, ontogeny, as well as other aspects of the biology of fungi (Lange 1952; Kemp 1974, 1975, 1980, Tama´s Papp 1985a, b; Ku¨es 2000). Having been subjected to tests Department of Microbiology, Faculty of Science and of the biological species concept, section Lanatuli Informatics, University of Szeged, Ko¨ze´p fasor 52, became the primary source of information about the H-6726 Szeged, Hungary mating system of Agaricomycetes and processes of speciation and reproductive isolation. As a forerunner of recent phylogenetic evidence for cryptic speciation Abstract: Sections Lanatuli and Atramentarii of the and reproductive isolation without discernable mor- genus Coprinopsis contain some of the best known phological divergence in mushrooms (Kauserud et al. and most important agaric species, including C. 2006, 2007; Matute et al. 2006, Le Gac et al. 2007, Sato cinerea and C. lagopus, yet a critical, phylogeny-based et al. 2007, Crespo and Lumbsch 2011), mating tests assessment of the species limits has not been carried out. Taxa have been characterized chiefly on the basis recovered significant conflict between morphologi- of morphological characters, which however show cally defined species and incompatibility groups, or little discriminatory power and/or considerable over- biological species (Kemp 1975, Flynn 1990, Vilgalys lap between several species pairs. We used ITS and 1991, Rowe 2011, Matute et al. 2006, Le Gac et al. LSU sequence data of 29 described taxa in Coprinopsis 2007, Crespo and Lumbsch 2011). Recent phyloge- sections Lanatuli and Atramentarii to infer species netic studies have suggested the widespread occur- limits and the correspondence between morpholog- rence of cryptic species in fungi (Bidochka et al. 2005, ical characters and species lineages, as well as to Alamouti et al. 2011, Chen et al. 2011), which in part examine the phylogenetic affinities of sections could explain the discrepancy observed between the Lanatuli and Atramentarii. Our analyses recovered number of morphologically defined species and three large clades, implying a paraphyly for section estimates of species diversity based on next genera- Lanatuli. Based on morphology and clade structure, tion sequencing technology (Hibbett et al. 2009, Nagy we estimate ca. 38 species in the two sections, et al. 2011). including several potentially new taxa, three of which The taxonomy of section Lanatuli has been are described herein. Coprinopsis pachyderma, C. determined predominantly by morphological and lagopus var. vacillans, C. acuminata, C. spelaiophila, ecological characters (Orton and Watling 1979, Ulje´ Coprinus citrinovelatus and Cop. brunneistrangulatus and Noordeloos 1999, Ulje´ et al. 2000, Ulje´ 2005). were found to be synonymous with other, earlier Macroscopic characters are largely uninformative for described species. Congruent with previous mating species identification, with the exception of the size of studies, our analyses recovered multiple, morpholog- basidiomes, which can be used to delimit certain species. Of the micromorphological features, spore Submitted 19 Apr 2012; accepted for publication 26 Jun 2012. shape and size have the greatest diversity in section 1 Corresponding author. E-mail: [email protected] Lanatuli. Additionally, the number of sterigmata and 112 NAGY ET AL.: COPRINOPSIS PHYLOGENY 113 characteristics of the veil have been used to define phylogenetic analyses. Furthermore, the LSU sequence species. Based on these traits, approx. three species of from the recently published genome of C. cinerea (Stajich Atramentarii and 19 of Lanatuli (including one et al. 2010) has been extracted and included in the variety) were recognized in a recent monographic alignments, because most developmental studies utilize the sequenced strain Okoyama 7_130. An ITS sequence of treatment of European species (Ulje´ 2005). A number C. lagopus (HM126487) was obtained from GenBank; of additional species have been described from the additional GenBank sequences from this group were tropics, which fit either section Lanatuli or Atramen- disregarded because of the unavailability of voucher tarii, including C. brunneofibrillosus (Dennis 1961), C. specimens and the possibility of misidentifications. In total, macropus, C. africanus and C. fibrillosus (Berkeley and 121 specimens were collected and analyzed. Coprinus Broome 1871, Pegler 1966). Coprinus neolagopus was cortinatus was used as outgroup. The abbreviations C. and described from Japan, while van de Bogart (1975, Cop. are used for Coprinopsis and Coprinus respectively 1979) published several provisionally named species throughout the manuscript. from North America. Coprinopsis nevillei and C. Morphological examinations.—Notes on macromorphology annulopora were described from dead herbaceous were made from freshly collected specimens. Dried herbar- stems, from France and Germany respectively (En- ium specimens were revived in 5% KOH for light microscopy. derle 2004, Garcia and Vellinga 2010). Measurements were made with a calibrated ocular microme- In addition to serving as a model for studies of a ter. Spores were measured at 10003, whereas other details biological species concept, various taxa of section were measured at 4003 magnification. At least 20 measure- Lanatuli are model organisms in developmental ments were made on each cell type on each specimen. The biology, fungal cell biology or enzyme production numbers after the word ‘‘Basidiospores’’ refers to the (Ku¨es 2000) and have been repeatedly isolated from number of spores measured and the number of basidiomes and collections they are originating from respectively. human fungal mycoses (as the anamorph, Hormogra- Terminology of shapes follows Vellinga (1988). phiella aspergillata, e.g. Lagrou et al. 2005). A complete genome sequence was published for C. Laboratory protocols.—Genomic DNA was extracted from cinerea (Stajich et al. 2010). In light of the popularity 1–10 mg dried basidiome, ground to a fine powder with of Lanatuli species as experimental model organisms micropestles, with the DNEasy Plant Mini Kit (QIAGEN) and the cumulative knowledge of their biology, a following the manufacturer’s instructions. We targeted the modern, phylogeny-based revision of the species in ITS1-5.8S-ITS2 locus and an approximately 1.5 kb portion of the large ribosomal subunit RNA gene (nrLSU) of the this group is highly desirable. nuclear ribosomal gene cluster. PCR amplification and In this study we set out to test the monophyly of sequencing followed standard protocols and was performed sections Lanatuli and Atramentarii and examine the with the primer combinations ITS1-ITS4 and LROR-LR7 species relationships within them. We infer trees from (White et al. 1990). Sequencing was performed commer- a combined alignment of sequences of two loci (ITS cially by LGC Genomics (Berlin). Individual readings were and LSU) and compare the phylogenetic patterns to assembled to contigs with the Pregap and Gap4 programs of established morphology-based classification to help the Staden package (Staden et al. 2000). delimit species in these groups. Alignments and phylogenetic analyses.—LSU sequences were aligned by Clustal W (Thompson et al. 2002), while the MATERIALS AND METHODS large number of indels in the ITS region necessitated a more sophisticated alignment strategy. First, we ran PRANK Taxon sampling.—The following strategy was used to design (Lo¨ytynoja et al. 2008), a probabilistic algorithm with the taxon sampling for this study: First, we attempted to collect ability to distinguish insertions and deletions with default specimens for all described species within sections Atra- parameters and two replicates to build an initial alignment. mentarii and Lanatuli (FIG. 1), which were supplemented This alignment, plus the LSU alignment was used to with materials selected by morphological examinations of compute a maximum likelihood tree (see below), which specimens preserved in two of the largest collections of was used as a guide tree in the second round of PRANK coprinoid fungi, L and SZMC (Szeged Microbiological alignment, with the +F option specified. This resulted in Collections, each totaling . 1500 and . 2000 specimens many fewer incorrectly aligned residues (based on visual respectively).