The 168-year taxonomy of Claviceps in the light of variations: From three morphological species to four sections based on multigene phylogenies

Abstract

Ergot fungi produce toxic sclerotia that have significant impacts on the agricultural, food and pharmaceutical industries. These fungi were classified in various genera (such as Sclerotium clavus, or Spermoedia clavus) before Tulasne, in 1853, erected Claviceps and described three species based on variations in morphology and host ranges, viz. C. purpurea, C. microcephala, and C. nigricans. Since then, knowledge regarding the biological, clinical, and pharmaceutical perspectives of ergot has accumulated rapidly. However, a serious taxonomic examination was lacking until Langdon’s revision accepted 25 species in 1952. That was followed by intensive regional studies by Loveless in the 1960s for Rhodesia (now Zimbabwe, Africa), and Tanda in Japan (1970s–1990s). More species names were reported and currently over 90 named taxa (species, varieties) are recorded in fungal name repositories (Mycobank and Index Fungorum). Most species were described based on morphological characteristics (sclerotia, ascomata and conidia) and host ranges. Some were characterized by their alkaloid profiles. Recently, DNA multi-locus sequence analyses (MLSA) were applied to resolve species complexes. For instance, the C. purpurea complex was separated into four species, and additional new species were recognized from South Africa and Canada. Infra- and supra-specific level genetic variations were identified via multi-locus and genomic studies. Based on five-locus phylogenies, Píchová and colleagues separated Claviceps into four sections: C. sect. Claviceps, C. sect. Citrinae, C. sect. Paspalorum and C. sect. Pusillae, for 60 species. Among these sections, several doubtful species names require clarification. Careful research on type specimens combined with molecular analyses is essential for clarifying these names.

Résumé

Les champignons de l’ergot produisent des sclérotes toxiques qui ont de sérieuses répercussions sur les industries agricole, alimentaire et pharmaceutique. Ces champignons avaient été classifiés dans différents genres (comme Sclerotium clavus ou Spermoedia clavus) avant que Tuslane (1853) élève Claviceps au rang d’espèce et en décrive trois en se basant sur les variations morphologiques et les gammes d’hôtes, à savoir C. purpurea, C. microcephala et C. nigricans. Depuis, les connaissances relatives aux perspectives biologiques, cliniques et pharmaceutiques de l’ergot se sont accumulées rapidement. Toutefois, un examen taxinomique approfondi faisait défaut jusqu’à ce que la révision de Langdon (1952) reconnaisse 25 espèces. Cela a été suivi par d’intenses études régionales menées par Loveless dans les années 1960 en Rhodésie (maintenant le Zimbabwe, en Afrique) et à Tanda au Japon (1970-1990). Plus de noms d’espèces ont été rapportés et, actuellement, plus de 90 taxons nommés (espèces, variétés) sont enregistrés dans des bases de données mycologiques (Mycobank et Index Fungorum). La plupart des espèces ont été décrites en fonction de leurs caractéristiques morphologiques (sclérotes, ascomes et conidies) et de leurs gammes d’hôtes. Certaines ont été caractérisées selon les profils de leurs alcaloïdes. Récemment, des analyses de séquences multilocus de l’ADN (MLSA) ont été utilisées pour examiner soigneusement les complexes d’espèces. Par exemple, le complexe C. purpurea a été divisé en quatre espèces et de nouvelles espèces supplémentaires sud-africaines et canadiennes ont été reconnues. Des variations génétiques aux niveaux infra et supra-spécifiques ont été constatées dans le cadre d’études multilocus et génomiques. En se basant sur des phylogénies à cinq locus, Píchová et coll. ont séparé Claviceps en quatre sections: C. sect. Claviceps, C. sect. Citrinae, C. sect. Paspalorum et C. sect. Pusillae, et ce, pour 60 espèces. Parmi ces sections, plusieurs noms controversés d’espèces requièrent d’être débrouillés. Une recherche minutieuse sur les spécimens types, combinée aux analyses moléculaires, est essentielle pour débrouiller ces noms.

Keywords:

• Clavicipitaceae
• mutualism
• mycotoxin
• phylogenetic species
• plant pathogen
• polyphasic

Mots clés:

• Clavicipitacées
• mutualisme
• mycotoxine
• agent phytopathogène
• espèces phylogénétiques
• polyphasique

Introduction

Ergot fungi belong to the genus Claviceps and typically infect the florets of cereal crops, grasses (Poaceae), rushes (Juncaceae) and sedges (Cyperaceae), replacing the seeds with toxic sclerotia. Some ergot species, such as C. africana Freder., Mantle & De Milliano, C. sorghi B.G.P. Kulk., Seshadri & Hegde and C. purpurea (Fr.) Tul. can reduce the grain production in agricultural populations (Menzies and Turkington 2015; Miedaner and Geiger 2015). Often, they were considered pathogens also because of their detrimental effects on humans and animals. However, in natural populations, their aposematic function proves to be beneficial to their host plants, therefore, to a certain extent, they should be considered mutualistic parasites instead of antagonistic plant pathogens (Lev-Yadun and Halpern 2007; Wäli et al. 2013). The mycotoxins contained in ergots (sclerotia) attracted a wide scope of research, viz biochemistry, biology, food safety, pharmacology and plant pathology (Kren and Cvak 1999; Menzies and Turkington 2015; Arcella et al. 2017). Identifying and naming ergot fungi can be traced back to ancient folklores (Barger 1931). The first formal binomial name was given by de Candolle (1815) as Sclerotium clavus, but Tulasne (1853) had the first comprehensive study on ergot, discovered the sexual stage, erected the genus Claviceps and recognized three species. This is often considered the starting point of the taxonomy of ergot. Throughout history, ergot species were recognized based on the morphology of sclerotia, ascostromata (if applicable), honeydew conidia, colony features of axenic cultures, host ranges, chemistry features of the sclerotium, honeydew and culture, as well as molecular markers. The phylogenetic relationships of ergot fungi and the major characteristics of some accepted species were first reviewed by Langdon (1952) and later by Pažoutová and Parberry (1999). More recently, a polyphasic approach was applied to a large panel of species and resulted in a supra-specific classification of the genus into four sections (Píchová et al. 2018). In this short review, we highlight some major milestones in the history of ergot taxonomy and summarize recent progress in recognizing the species and genetic diversity through a molecular polyphasic approach.

Pre-history period

The 168 year history of ergot taxonomic study commences with the initial naming of Claviceps purpurea based on its sexual stage by Tulasne (1853), and includes up to the year 2021, when this paper was presented at a Canadian Phytopathological Society symposium entitled ‘The increasing issue of ergot in cereal production and the biological questions being explored’. Prior to 1853, ergot fungi had various names in the folklores of various ethnic groups. A list of prescriptions in German from 1474 and its medicinal use seems to be the oldest documentation of ergots (Píchová et al. 2018), whereas the earliest reference in literature can be traced back to Adam Lonicer’s description of the ‘long black nails between rye grains’ in a German book, Kreuterbuch, in 1582, according to Barger’s (1931) historical account. The name ‘ergot’ was used by the peasants of Sologne (France), referring to the shape of sclerotia resembling cock’s-spurs (Barger 1931). At that time, the fungus was referred to mostly as abnormal grains, i.e., sun-baked grains, excessive sap formation, or mutant grains. Claude Joseph Geoffroy, a French apothecary and chemist, was the first to recognize them as separate organisms from their host plants in the 1700s (Oberstockstall 2013). About 40–50 years later, Augustin Pyramus de Candolle (1778–1841, resourced from Index Fungorum http://www.indexfungorum.org/Names/AuthorsOfFungalNames.asp) confirmed this observation, included the fungus in his iconic treatise Famille des champignons in Flore française and named it Sclerotium clavus (de Candolle 1815). Elias Magnus Fries (1794–1878) erected the genus Spermoedia for the sclerotium state, and made the new combination Sp. clavus (DC) Fr., meanwhile also included Sp. paspali (Schwein.) Fr. in the genus (Fries 1822). Joseph-Henri Léveillé (1796–1870) described the asexual stage (conidia in honeydew) of the fungus on rye and other cereal crops and named it Sphacelia segetum (Léveillé 1827).

Louis Rene Tulasne (1815–1885) and the early twentieth century

Tulasne (1853) discovered the sexual stage during his detailed studies on l’ergot des glumacées and considered it conspecific to Sphaeria purpurea, a name coined by Fries (1823) in sanctioning Sphaeria entomorrhiza Schumach. (Schumacher 1803) based on a figure of stroma in Dickson, Fasc. pl. crypt. brit., 1, 1785 (resourced from Index Fungorum, accessed in December 2021). The genus Claviceps was erected to include both sclerotial and the sexual stages, the combination C. purpurea was made, and Sclerotium clavus DC was noted as the sclerotial state of the fungus. Besides C. purpurea, Tulasne also recognized two other species based on different host ranges and morphological and biological differences. Claviceps purpurea was on cereal crops (rye, wheat, oats) and various grasses, C. microcephala (Wallr.) Tul. on Phragmites communis Trin. and Molinia caerulea (L.) Moench, and C. nigricans Tul. on Scirporum (Scirpus, Cyperaceae). Morphologically, the ascomata of C. microcephala were characterized by producing long flexuous stipes, and very small soft heads (capitula). Those of C. nigricans had the perithecia located more distantly than the other two species. Biologically, C. microcephala differed from the other two species by the immediate germination of sclerotia on plants (Tulasne 1853). Between 1853 and the 1940s, over30 ergot species were described as Claviceps spp. by mycologists here and there (Table 1), some of which were reclassified to other genera by contemporary studies (more discussion in the section on the molecular era). A few species were described as Spermoedia spp. such as Sp. zizaniae Fyles, Sp. stevensii Seaver, Sp. rolfsii (F. Steven & J. G. Hall) Seaver, Sp. tripsaci (F. Steven & J. G. Hall) Seaver.

Table 1. Claviceps spp. described between 1853 and 1940s.

Author year	Species name
Tulasne 1853	C. microcephal, C. nigricans, C. purpurea
Cesati 1855	C. pusilla
Fuckel 1860	C. euphorbiae*
Bail 1862	C. typhinum (Pers.)*
Saccardo 1883	C. setulosa
Cooke 1884	C. wilsonii*
Hennings 1887–1990	C. pallida*, C. pallida var. orthocladae, C. patouillardiana, C. uleana
Rehm 1889	C. philippii*
Griffiths 1901–1902	C. cinerea, C. caricina*
Möller 1901	C. balansioides, C. lutea, C. ranunculoides
Stäger 1906	C. sesleriae
Adams 1907	C. junci
Stevens 1910	C. paspali, C. rolfsii, C. tripsaci
Hauman-Merck 1921	C. deliquescens
Petch 1933	C. flavella
Nakata 1934	C. virens*
Krebs 1936	C. purpurea f. secalis
Togashi 1936	C. yanagawaensis
Hansford 1941	C. digitariae
Langdon 1942	C. annulata, C. glabra, C. hirtell, C. platytricha
J. W. Groves 1943	C. grohii
Padwick & Azmatullah 1943	C. viridis
Sawada 1944	C. miscanthi, C. syntherismae
Kawatani 1946	C. litoralis

Species marked with * were reclassified to other genera in current taxonomy.

R.F.N. Langdon (1916 – 2014) – the first comprehensive revision

The first comprehensive revision was conducted by Langdon during the 1940s and the 1950s. He studied Australian ergot species and also did a systematic account on various aspects of twenty-five non-Australian Claviceps spp. including the informative characteristics for differentiating species, host ranges, distributions, and synonyms (Langdon 1952). His dissertation is still a relevant reference for current ergot taxonomy. Based on his observation, the characteristics that were useful for differentiating species include the shape and colour of sclerotia, the colour of ascostromata, the placement of perithecia (imbedded or elevated) etc. He considered other characteristics, such as the length of stipes and the dimension and shape of perithecia, as insignificant for differentiating species. The shapes and sizes of conidia varied between species, but are also variable within species (Langdon 1952), thus this might be useful to a certain extent.

A. R. Loveless – characteristics of honeydew conidia

Loveless put more emphasis on the morphology of conidia as useful features for distinguishing species. He studied the conidia from honeydew from 34 grass species collected around Rhodesia (a British colony in territory equivalent to present-day Zimbabwe). Using a piece of special graphic equipment, he outlined the shapes and sizes of conidia randomly selected from honeydew samples from 34 grass species, and categorized the conidial shapes and sizes into 13 groups, six of which were associated with a certain species, i.e. group 2, 5, 6, 8, 9 and 12 corresponding to C. paspali F. Stevens & J.G. Hall, C. digitariae Hansf., C. sulcata Langdon, C. maximensis T. Theis, C. pusilla (Ces.) Ces. and C. cynodontis Langdon. Consequently, he predicted seven more species to be described in Rhodesia (Loveless 1964). Later, several species described in the area indeed corresponded to the size ranges of conidial groups, such as C. africana to group 10 (Frederickson et al. 1991), C. fusiformis Loveless to group 7 (Loveless 1967), C. rhynchelytri Herd & Loveless to group 1 (Herd and Loveless 1965), Claviceps loudetiae Pažoutová & Freder. to group 13, and C. truncatispora Pažoutová & Freder. to group 11 (Pažoutová et al. 2011).

S. Tanda – ergot fungi in Japan between 1970s and 1990s

Another frequently encountered name in the ergot taxonomy literature is Seinosuke Tanda. Tanda had a series of publications on ergot fungi in Japan between the 1970s to 1990s. He described not only some species, C. imperatae Tanda & Kawat. (Tanda and Kawatani 1976), C. microspora Tanda (Tanda 1985), C. panicoidearum Tanda & Y. Harada (Tanda and Harada 1989), C. bothriochloae Tanda & Y. Muray (Tanda 1991), C. amamiensis Tanda (Tanda 1992), but also characterized a lot of varieties (Tanda 1977, 1978a, 1978b, 1978c, 1979a, 1979b, 1979c, 1980a, 1980b, 1981) based on very careful examinations, measurements, and comparisons using many aspects of features including sclerotia, sexual stage, asexual stage, culture, alkaloid tests, and sugar consumption in culture. Even though he noted the ergot on Phalaris arundinacea L. had significantly larger conidia, he was not convinced it was a new species. He called it a new variety, viz C. purpurea var. phalaridis (Tanda 1979b), which turned out to be a new species based on molecular evidence (see later discussion about C. humidiphila Pažoutová & M. Kolařík). In the same way, the real identities of those varieties he described need to be validated.

Chemistry features applied for Claviceps taxonomy

The sugar composition in honeydew and its significance for taxonomy was investigated by Mower and Hancock (1975). Through analyzing the sugar composition of nine determined and several undetermined Claviceps spp., they found significant variation in sugar profiles associated with several taxonomic sub-groups. For example, C. gigantea S.F. Fuentes, C. cinerea Griffiths and C. grohii J.W. Groves were rich in arabinitol or/and mannitol, whereas C. purpurea, C. fusiformis, and C. nigricans were rich in glucose and fructose. In contrast, there was a lot more research on the alkaloid content in sclerotia or in vitro cultures as a result of their toxic effects and pharmaceutical potentials. As three groups of ergot alkaloids (EAs) were first isolated from the sclerotia of C. purpurea (Stoll 1950), a wide spectrum of active compounds have been continuously reported from various Claviceps spp. through many analytical technics (Tanaka and Sugawa 1952; Mantle 1968; Porter et al. 1974), also reviewed by Flieger et al. (1997). Consequently, the characteristics of EA production were incorporated in species descriptions. Notable examples are the studies of ergot in Japan by Tanda in that the presence/absence of EA production was almost always featured in the descriptions of species or/and varieties (Tanda 1979b, 1991). Lately, the more advanced technologies, i.e. high performance liquid chromatography (HPLC) and ultra high performance liquid chromatography (UPLC) enabled the highly sensitive and reliable separation of compounds and profiling of EA production. Varied species can be characterized by a unique combination of EA components in different quantities (Negård et al. 2015; Liu et al. 2020). Furthermore, indole diterpenes or paspaline-like compounds were detected in several more species in addition to C. paspali and C. cynodontis (Uhlig et al. 2014, 2021; Negård et al. 2015). In addition to alkaloids, various pigments present in the sclerotia are often in equal or greater amounts (Buchta and Cvak 1999). Among them, ergochromes showing toxicity to animals can contribute to overall ergot toxicity, and their spectrum represents another distinguishing feature between species (Flieger et al. 2019).

Ergot taxonomy in the molecular era – a polyphasic approach

The Pažoutová group in the Czech Republic should be considered the pioneer in the application of molecular markers for differentiating Claviceps species with their publication of characterizing C. citrina Pažoutová, Fučík., Leyva-Mir & Flieger, a species on Distichlis spicata (L.) Greene from Mexico. They demonstrated the different fingerprints of the random amplified polymorphism DNA (RAPD) of C. citrina from eight other species and variations in the rDNA internal transcribed spacer (ITS) (Pažoutová et al. 1998). Combined with its specific host range and other unique morphological features, especially the lemon-yellow colour of capitula, C. citrina was so named. During their study, it was found that on a high-sugar content media (T2), most Claviceps species grow faster, with less aerial hyphae showing more compact colonies that differed in texture and colour (Pažoutová et al. 1998, 2008a). Conidia formation on T2 medium appeared more abundant, which allowed more opportunity to observe the features of conidia and conidiogenesis. Building on previous observations of the phialidic conidiogenesis in Claviceps (Luttrell 1980; Rykard et al. 1984) and the occasional formation of the secondary conidia, Pažoutová and colleagues observed pleomorphic conidiation, i.e. primary enteroblastic (Sphacelia-like), holoblastic formation of secondary conidia (such as in C. africana) and sympodial holoblastic in C. citrina and C. zizaniae producing conidia with truncate ends (Pažoutová et al. 2004). The presence/absence of micro-conidia and the secondary conidia proved to be species-specific characteristics.

A phylogenetic study of C. purpurea and close relatives using RAPD fingerprinting and ITS rDNA sequences showed the presence of three genetic groups, G1 – G3, distinguishable by conidia morphology and alkaloid spectrum (Pažoutová et al. 2008b). This was later verified by multi-locus sequence analyses, and the three genetic groups were proven to be associated with different ecological niches, suggesting ecological cryptic speciation (Douhan et al. 2008). Until 2015, a formal taxonomic treatment was applied, and species names were designated to the three groups, i.e. G1 = C. purpurea s. str, G2 = C. humidiphila and G3 = C. spartinae (R.A. Duncan & J.F. White) Pažoutová & M. Kolařík. In addition, the new lineage G2a was named C. arundinis (Pažoutová et al. 2015). In terms of the host range and morphology, C. arundinis corresponds well with Tulasne’s concept of C. microcephala (Wallr.) Tul. However, the type specimen connected to C. microcephala (basionym: Kentrosporium microcephalum Wallr. 1846) was on an insect larva, which conflicts with the host of C. arundinis, therefore a new name was proposed with a holotype on Phragmites australis (Cav.) Trin. Ex Steud. (PRM922710). Claviceps humidiphila was characterized by often producing larger conidia than C. purpurea, and commonly found on Calamagrostis, Deschampsia cespitosa (L.) P. Beauv. and Phalaris arundinacea, corresponding well to C. purpurea var. phalaridis Tanda, therefore Pažoutová et al. (2015) considered these two conspecific and designated a specimen collected by Tanda from Japan as holotype and a specimen from Germany as an epitype. A follow-up study proved the loss of the holotype specimen, and another specimen collected from the type location indicated C. purpurea var. phalaridis Tanda belonged to a different species from the epitype of C. humidiphila from Germany. A taxonomic rectification was conducted in that the specimen from Japan was designated as a neotype of C. humidiphila, whereas the German species was given a new name C. bavariensis (Liu et al. 2022). The delimitation of these species was confirmed by a study of Norweigian samples and also supported by the different indole alkaloid profiles produced by these species (Negård et al. 2015).

Host specificity of species and/or varieties was a long-standing topic of ergot study in that some species have a restricted host range, while others are more generalists (Campbell 1957; Alderman et al. 2004). The four phylogenetic species mentioned earlier showed a complex pattern of host preference. Some plant species (universal host) can host more ergot species (e.g. Poa, Dactylis) than others, whereas in particular areas (Central Europe vs Scandinavia) host specificity is more evident at a certain level, such that cultivated cereals in Europe are exclusively infected by C. purpurea s.str. (Negård et al. 2015; Pažoutová et al. 2015). Pažoutová et al. (2008b) observed that C. purpurea s.str. prefers a dryer and open localities, whereas C. humidiphila is more often in wet and shady places. This pattern was challenged by later studies (Negård et al. 2015; Liu et al. 2020). Further evaluation with proper sampling on universal hosts could bring more insights. The DNA sequence-based polyphasic approach was also applied to genus-wide phylogenetic study and classification. Píchová et al. (2018) conducted comprehensive phylogenetic analyses based on multiple protein-coding genes, resulting in the supra-specific classification of the genus into four sections, i.e. Claviceps sect. Claviceps, C. sect. Citrinae, C. sect. Paspalorum, and C. sect. Pusillae, associated with host and geographic ranges, and ergochrome and ergot alkaloid profiles. This comprehensive phylogenetic tree was further supplemented with a few more species from Canada (Fig. 1). In a supplementary Table, Píchová et al. (2018) also listed a tentative assessment for ninety published names of Claviceps spp. building a foundation for a taxonomic revision.

Fig. 1 (Colour online) Phylogenetic relationships of ergot species with available DNA sequences. The classification into four sections (Citrinae, Claviceps, Paspalorum, Pusillae) and their characterization (toxicity, host range, geography and others), were based on.Píchová et al. (2018) and Liu et al. (2020). Ergot species in bold are of agricultural importance: C. africana on sorghum, C. cynodontis on Bermuda grass, C. fusiformis on pearl millet, C. gigantea on maize, C. paspali on paspalum and C. purpurea on cereals and forage grasses are important food and feed contaminants causing toxicity to human and livestock. Claviceps bavariensis infecting common pasture grasses may impact livestock. The maximum likelihood tree generated in PhyML 3.0. was constructed using ITS-LSU rDNA, TUB2, TEF1-α and RPB2 sequences from Píchová et al. (2018) and Liu et al. (2020).

Genetic diversity in Canada and the western USA

In the Canadian National Mycological Herbarium (DAOM), 90% of over 700 specimens of ergot fungi were identified as C. purpurea partly because ergot has never been systematically studied by taxonomists in Canada. In addition to C. purpurea, five other species were in the collection, including C. grohii, C. junci J. Adams, C. microcephala (possibly C. arundinis, see more discussion earlier), C. nigricans and C. zizaniae (Fyles) Pantidou ex Redhead, M.E. Corlett & M.N.L. Lefebvre.

Multi-gene DNA sequence analyses of the samples collected in Canadian agricultural and surrounding areas recovered seven lineages within the pre-molecular concept of C. purpurea (Shoukouhi et al. 2019). Three of them corresponded to C. humidiphila (now C. bavariensis), C. spartinae, and C. purpurea s.str., whereas the other four new lineages represented four new species (Liu et al. 2020), i.e. C. ripicola M. Liu, J.G. Menzies & S.A. Wyka on Ammophila, Phalaris and Calamagrostis from semi-wet habitat, river/lake banks, and sand dunes (British Columbia, Manitoba, Ontario, Quebec; USA: Colorado), C. perihumidiphila M. Liu on Agrostis capillaris L. and Elymus albicans (Scribn. & J. G. Sm.) Á. Löve from marshlands or terrestrial (Alberta, Newfoundland), C. quebecensis M. Liu & J. Cayouetteon Ammophila breviligulata Fernald (Quebec) distinguishable by ovoid shaped conidia, C. occidentalis M.Liu as axenic culture from Canadian Collection of Fungal Culture, and originally on Bromus inermis Leyss., and Phleum pratense L. from Alberta. The genetic diversity and distribution of C. purpurea s.str. in Canada and the Western USA were investigated through phylogenetic network analyses and population demographic parameters, resulting in the discovery of three subdivided genetic clusters, however the subdivision is not correlated with geographic regions or host groups. It was speculated that the intrinsic mechanisms that led to the cessation of gene flow among these three generic clusters might be mating and/or vegetative incompatibility, genomic adaptation etc. (Liu et al. 2021).

Challenges and future perspectives

The internal transcribed spacer (ITS) region was widely accepted as a marker in species identifications and phylogenetic analyses for many fungal groups and was proposed by the Consortium for the Barcode of Life as the primary universal DNA barcoding locus for fungi (Schoch et al. 2012). However, for some fungal groups, the ITS appeared to be a poor marker as a result of the presence of non-orthologous copies (O’Donnell and Cigelnik 1997), or low resolution at the species level (Geiser et al. 2007). In Claviceps, non-orthologous ITS copies were found in C. purpurea s.str. (G1) and C. occidentalis (G4) through examination of the ITS tree generated for Canadian samples (Shoukouhi et al. 2019). The low resolution at the species level is another shortcoming from which the ITS region in Claviceps spp. suffers; for instance, it could neither separate C. ripicola (G6, 7) from C. spartinae (G3), nor C. arundinis (G2a), C. humidiphila (G2) and C. perihumidiphila (G2b) (Negård et al. 2015; Pažoutová et al. 2015; Shoukouhi et al. 2019). In contrast, the protein-coding genes provide better resolutions, although, for certain species, a fragment of the second largest subunit of RNA polymerase II (RPB2) and the translation elongation factor 1-α (EF1-α) displayed a varied pattern of affinities among species. For example, RPB2 inferred C. humidiphila (G2) and C. arundinis (G2a) are more closely related than each of them to C. perihumidiphila (G2b), whereas TEF1 indicated C. humidiphila is more closely related with C. perihumidiphila (Shoukouhi et al. 2019). A follow-up study using the neutrality test indicated that these two genes are likely linked with genes under positive selection, and have undergone ‘genetic hitchhiking’ or ‘selective sweeping’ (Liu et al. 2021). Given these limitations of individual genes, it is recommended to apply a holistic MLSA approach for species recognition in Claviceps.

The drastically increased number of whole genome sequences of Claviceps confirmed the species recognition based on MLSA, and also allowed the genome level understanding of the speciation mechanism associated with genome landscape and host expansion (Wyka et al. 2021). This could also provide opportunities for developing more gene markers for species recognitions and species-level phylogenetic studies. The question then arise is what would be the best strategy to apply genome data in fungal taxonomy. Similar to the selection of standard DNA barcoding loci, there is a need to select a standard panel of genes for species recognition in various fungal groups. Target enrichment approaches increased the efficiency of obtaining pre-selected genome regions that are suitable for specific phylogenetic levels (Albert et al. 2007; Gnirke et al. 2009; Jones and Good 2016). The technology has been used for certain fungi and fungus-like groups for exploring phylogenetic informative markers (Grewe et al. 2020; Nguyen et al. 2021). In the course of selecting the target genome regions, the paralogous gene copies caused by ancient lineage sorting, hybridization, recombination, horizontal gene transfer etc. were still among the biggest concerns (Andermann et al. 2020; Dettman and Eggertson 2021). Some strategies were proposed for distinguishing them, including identifying the unusually high levels of variations or checking the presence of extra haplotypes. However, none of the tactics can ensure the elimination of paralogous genes (Andermann et al. 2020). A thorough understanding of specific genome structures through careful genome wide examination is deemed to be an essential step. The concept and recognition of species have been a centre of debates since the birth of biological species concepts (Mayr 1942; Dobzhansky 1970) that criticized the shortcomings of the traditional typological species concepts or morphological species concepts (Mayr 1996). The debates became more heated with the emergence of diverse species concepts coming along with the molecular markers applied for classification, for example, the genealogical species concept (Baum and Shaw 1995), Hennigian species concept (Meier and Willmann 2000), monophyletic species concept (Donoghue 1985), phylogenetic species concepts (Nixon and Wheeler 1990) etc, see review by Luckow (1995). In fungal molecular systematic, the Genealogical Concordance Phylogenetic Species Recognition (GCPSR, Taylor et al. (2000)) was widely applied in fungal species recognition although many researches demonstrated that paraphyletic groups could be considered as species as well (Hörandl and Stuessy 2010). It is foreseeable that the debates over how to recognize fungal species will continue in the genomic era (Matute and Sepúlveda 2019; Xu 2020). We argue that despite the rapid advances in DNA-based technology may reveal more genotypic diversity, the recognition of species should still be based on the combination of genotypic and phenotypic features (polyphasic approach) because it is the phenotypes that are tangible features directly impacting humans and eco-systems (Lücking et al. 2020).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This research was funded by Agriculture and Agri-Food Canada’s Growing Forward 2 for a research network on Emerging Mycotoxins [EmTox, project # J-000048], Agriculture and Agri-Food Canada, Science and Technology Branch fungal and bacterial biosystematics [J-002272].