- Open Access
New endophytic Toxicocladosporium species from cacti in Brazil, and description of Neocladosporium gen. nov.
IMA Fungus volume 8, pages77–97(2017)
Brazil harbours a unique ecosystem, the Caatinga, which belongs to the tropical dry forest biome. This region has an important diversity of organisms, and recently several new fungal species have been described from different hosts and substrates within it. During a survey of fungal endophyte diversity from cacti in this forest, we isolated cladosporium-like fungi that were subjected to morphological and multigene phylogenetic analyses including actA, ITS, LSU, rpb2 and tub2 gene sequences. Based on these analyses we identified two new species belonging to the genus Toxicocladosporium, described here as T. cacti and T. immaculatum spp. nov., isolated from Pilosocereus gounellei subsp. gounellei and Melocactus zehntneri, respectively. To improve the species recognition and assess species diversity in Toxicocladosporium we studied all ex-type strains of the genus, for which actA, rpb2 and tub2 barcodes were also generated. After phylogenetic reconstruction using five loci, we differentiated 13 species in the genus. Toxicocladosporium velox and T. chlamydosporum are synonymized based on their phylogenetic position and limited number of unique nucleotide differences. Six strains previously assigned to T. leucadendri, including the ex-type strain (CBS 131317) of that species, were found to belong to an undescribed genus here named as Neocladosporium gen. nov, with N. leucadendri comb. nov. as type species. Furthermore, this study proposes the actA, ITS, rpb2 and tub2 as main phylogenetic loci to recognise Toxicocladosporium species.
The genus Toxicocladosporium (Cladosporiaceae, Capnodiales) was described by Crous et al. (2007) to accommodate cladosporium-like fungi having distinct “dark, thick-walled conidial and conidiophore septa, and lacking the typical coronate Cladosporium scar type”. The type species of this genus, T. irritans, was isolated from mouldy paint in Suriname and named “irritans” because of the production of several volatile metabolites in culture, causing skin irritation when there is exposure to the fungus (Crous et al. 2007). After its introduction, several new species were described in the genus, which currently comprises 13 species reported from different host plants in studies from America (Suriname and USA), Africa (Madagascar and South Africa), Asia (China), and Oceania (Australia).
Similar to Cladosporium, Toxicocladosporium exhibits a widespread distribution and the capacity to colonise distinct substrates and plant families. Almost all species in this genus were described from plant species belonging to the families Asteraceae, Cyperaceae, Myrtaceae, Pinaceae, Proteaceae, Rubiaceae, and Strelitziaceae (Crous et al. 2009a, 2010a, b, 2011a, 2012a, b, 2013, 2014, Crous & Groenewald 2011), the exception being T. irritans and T. hominis described from mouldy paint (Crous et al. 2007) and a human bronchoalveolar lavage fluid specimen (Crous et al. 2016), respectively. In addition to species descriptions in this genus, few reports of isolation of Toxicocladosporium species, mainly T. irritans, have been published in different countries. For example, studying ancient laid-paper documents of the 17th century in Portugal, Mesquita et al. (2009) reported the isolation of T. irritans. Similar results were obtained in Italy by Piñar et al. (2015) who used culture-independent molecular methods and scanning electron microscopy (SEM) to verify the fungi colonizing parchment manuscripts, and by Bonadonna et al. (2014), who reported T. irritans colonising tattoo inks. These reports may show similarities because the first isolation of T. irritans was associated with mouldy paint in Suriname (Crous et al. 2007). Toxicocladosporium irritans was also reported associated with patients having atopic dermatitis in Japan (Zhang et al. 2011), and it was isolated from human blood and a fingernail by Sandoval-Denis et al. (2015) in the USA. This species was also reported by Cruywagen et al. (2015) on baobab trees in southern Africa, and on equipment used in the International Space Station or Space Shuttle in Japan (Satoh et al. 2016).
There are also reports from other unusual substrates or hosts, including coffee scale insects in Vietnam (Nhạ et al. 2011), the vector of visceral leishmaniasis (Lutzomyia longipalpis) in Brazil (McCarthy et al. 2011), an unidentified sponge from Korea (Cho et al. 2016), patients with seborrheic dermatitis in Japan (Tanaka et al. 2014), outdoor dust samples in the USA (Barberán et al. 2015), and as a plant pathogen on African olive (Olea europaea subsp. cuspidata, Oleaceae) in Australia (Australian Government Department of Agriculture 2015). Toxicocladosporium and Cladosporium were also suggested as candidates for fungal structures found in the fossilized extinct aquatic angiosperm Eorhiza arnoldii in Canada (Klymiuk et al. 2013). These reports show that Toxicocladosporium host associations are not specific and may differ from Cladosporium, in which species tend to have confined host ranges, but with some exceptions (Bensch et al. 2012). Toxicocladosporium chlamydosporum and T. rubrigenum were, however, described from a single leaf spot of Eucalyptus camaldulensis (Myrtaceae) growing in Madagascar (Crous et al. 2009a). This example demonstrates that specimens from a single host and location can be colonized by genotypes representing different species (Bensch et al. 2012). Toxicocladosporium species may be recovered from inconspicuous substrates and extreme habitats, showing a lack of environmental preference and an ability to be associated with unusual materials and ecological conditions (McCarthy et al. 2011, Nhạ et al. 2011, Cho et al. 2016, Satoh et al. 2016).
Dematiaceous fungi isolated from different plant species in extreme environments generally live as endophytes (Redman et al. 2002, Suryanarayanan et al. 2011, Loro et al. 2012, Sun et al. 2012, Knapp et al. 2015). Although Cladosporium species are widely reported as endophytes (Bensch et al. 2012), the closely related genus Toxicocladosporium has not previously been reported as endophytic. All presently known associations of Toxicocladosporium species with plant material were as an epiphyte, saprobe, or phytopathogen, or with unusual substrates or hosts.
Plants living in dry environments are an important host for fungi with widespread distributions, and have always shown a great diversity of species (Fisher et al. 1994, Suryanarayanan et al. 2005, Khidir et al. 2010, Silva-Hughes et al. 2015, Fonseca-García et al. 2016). The Caatinga, one of the most important tropical dry forests in Brazil, harbours several cacti that prove to have a great diversity of endophytic fungi (Bezerra et al. 2012, 2013, Freire et al. 2015). Recently,
Bezerra et al. (2017) described a new order in the class Dothideomycetes for endophytes isolated from the cactus Tacinga inamoena collected in the Caatinga.
We studied all ex-type strains of Toxicocladosporium species isolated from different substrates and hosts in order to report on the isolation and to describe those we recovered as endophytes from the cacti Melocactus zehntneri and Pilosocereus gounellei subsp. gounellei growing in the Caatinga. Using morphological characters and multigene phylogenetic analyses (actA, ITS, LSU, rpb2 and tub2), the genus Toxicocladosporium and its respective species were reevaluated. We aimed to determine the phylogenetic relationship of endophytes from cacti with species of Toxicocladosporium, provide an overview of hosts and substrates amongst Toxicocladosporium species, and propose new loci to assist with species differentiation in the genus.
Materials and Methods
Endophytic fungi from cacti
Endophytic fungi were isolated as described by Bezerra et al. (2013) from the cacti Melocactus zehntneri and Pilosocereus gounellei subsp. gounellei growing in the Brazilian tropical dry forest (Caatinga), Catimbau National Park, Buique municipality, Pernambuco state, Brazil (8°36′35″S, 37°14′40″ W), and sustainable family farming plots, Itaíba municipality, Pernambuco state, Brazil (9° 08.895′ S, 37° 12.069′ W). The collections were authorized by the Ministério do Meio Ambiente (MMA)/Instituto Chico Mendes de Conservação da Biodiversidade (ICMBio); permission number: 40331-1/ authentication code 87451826 issued on 4 November, 2013. In addition, 32 isolates selected on the basis of genetic and morphological relatedness with cacti endophytes, were obtained from the collection of the Westerdijk Fungal Biodiversity Institute (formerly CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands) and the CPC collection (collection of P.W. Crous, held at CBS) and included in the analyses (Table 1).
Endophytes previously identified as belonging to Toxicocladosporium were cultured on malt extract agar (MEA), oatmeal agar (OA), potato dextrose agar (PDA), and synthetic nutrient deficient agar (SNA) (Crous et al. 2009c), and incubated at 22 °C under a natural day-night cycle. Macro- and micro-morphological features, and reproductive structures were visualized after 3 wk on MEA, OA, PDA, and/ or SNA culture media. Culture colours were evaluated using the charts of Rayner (1970). Slide preparations were mounted as described by Bensch et al. (2012) in clear lactic acid and/ or in Shear’s solution. Endophytic strains are deposited in the culture collections of Micoteca URM Prof. Maria Auxiliadora Cavalcanti (Federal University of Pernambuco, Recife, Brazil — www.ufpe.br/micoteca, WCDM 604) and the CBS collection at Westerdijk Fungal Biodiversity Institute (under Material Transfer Agreement — MTA N° 05/2015/Micoteca URM, issued on 14 April, 2015). Nomenclatural and taxonomic information were deposited in MycoBank (www.mycobank.org) (Crous et al. 2004).
DNA extraction, amplification (PCR) and sequencing
Genomic DNA extraction was performed using the Wizard® Genomic DNA Purification Kit (Promega, Madison, WI) according to the manufacturer’s instructions. The primers LR0R and LR5 (Vilgalys & Hester 1990), ITS5 and ITS4 (White et al. 1990), ACT-512F and ACT-783R (Carbone & Kohn 1999), Bt2a and Bt2b or Bt10 (Glass & Donaldson 1995) and 5f2 and 7cr (O’Donnell et al. 2010) were used to amplify part of the nuclear ribosomal large subunit (LSU) of the rDNA, the ITS region (first and second ITS regions and intervening 5.8S nrDNA), the partial actin gene (actA), partial β-tubulin gene (tub2), and a fragment of the RNA polymerase second largest subunit gene (rpb2) respectively. Amplification and sequencing reactions, sequences analyses, and consensus sequences were performed as described by O’Donnell et al. (2010) and Bezerra et al. (2017). In addition, 136 DNA sequences representing 57 taxa were retrieved from GenBank and included in the phylogenetic analyses (Table 1).
Following blast searches of the NCBI’s GenBank nucleotide database for preliminary identifications, an initial backbone tree was constructed using ITS, LSU and rpb2 sequences from Cladosporiaceae (Schubert et al. 2007a, b, Zalar et al. 2007, Crous et al. 2007, 2009b, 2011a, Bensch et al. 2010, 2012, 2015) and from the other six families in Capnodiales following Quaedvlieg et al. (2014) and Videira et al. (2016). Parastagonospora nodorum (CBS 110109) was used as outgroup. Firstly, the alignments for each locus were performed using the online MAFFT interface (Katoh & Standley 2013) followed by manual adjustments using MEGA v. 7 (Kumar et al. 2015). These alignments were used to infer preliminary phylogenetic relationships for Toxicocladosporium species in Cladosporiaceae.
A second, more inclusive analysis included actA, LSU, ITS, rpb2 and tub2 sequences derived from ex-type cultures of Toxicocladosporium species and endophytes isolated from cacti (Crous et al. 2007, 2009a, 2010a, b, 2011a, 2012a, b, 2013, 2014, 2016, Crous & Groenewald 2011). Neocladosporium leucadendri (CPC 18315 = CBS 131317), previously published as Toxicocladosporium leucadendri, was used as outgroup for that analysis.
Maximum Parsimony analyses (MP) were performed with PAUP v. 4.0b10 (Swofford 2003) and involved 1000 replicates of heuristic search with random addition of sequences. The tree bisection-reconnection option was used, with the branch swapping option set to “best-trees” only. Gaps were treated as missing data and all characters were unordered and given equal weight. The tree length (TL), consistency index (CI), Retention index (RI), and rescaled consistence index (RC) were calculated. Maximum parsimony bootstrap analyses (MP-BS) were performed using 1000 replicates. Maximum likelihood analyses (ML) were performed using RAxML-HPC2 v. 8.2.8 (Stamatakis 2014) on XSEDE in the CIPRES science gateway (http://www.phylo.org/). The robustness of the trees obtained was evaluated according to the level of bootstrap support (ML-BS), with the number of replicates determined automatically by the software. Bayesian analyses (BI) were performed using MrBayes v. 3.2.6 (Huelsenbeck & Ronquist 2001). The program was executed with four Markov chains in two simultaneous runs for 5 M generations with the stopval option on and saving trees every 1000 generations. The analyses were stopped when the two runs converged and the average standard deviation of split frequencies came below 0.01. The 50% majority-rule consensus tree and the Bayesian posterior probabilities (BPP) were calculated after discarding the first 25% of saved trees as “burn-in”. The best fit evolutionary models were calculated independently for each gene data partition using MrModelTest v. 2.3 (Nylander 2004) following the Akaike information criterion and included in the analyses, in all cases selecting the GTR+I+G model. All resulting trees were plotted using FigTree v. 1.4.2 (http://tree.bio.ed.ac.uk/software/figtree/). All the analyses were first made independently for each locus and visually inspected for topological incongruences between nodes with significant statistical support before being combined into multigene datasets (Mason-Gamer & Kellogg 1996, Wiens 1998). The new sequences generated in this study were deposited in the NCBI’s GenBank nucleotide database and the European Nucleotide Archive (Table 1) and the alignments and phylogenetic trees in TreeBASE (Study ID S20701).
In order to verify the relationship of Toxicocladosporium with other genera in the family Cladosporiaceae, we used ITS, LSU and rpb2 sequences from representatives of 19 genera from seven families in Capnodiales. Parastagonospora nodorum (CBS 110109) was used as outgroup. The final combined alignment contained 68 isolates and 1750 characters (ITS: 427, LSU: 621 and rpb2: 702) of which 770 were parsimony-informative (ITS: 178, LSU: 161 and rpb2: 431), 117 were variable and parsimony-uninformative (ITS: 37, LSU: 59 and rpb2: 21), and 848 were constant (ITS: 427, LSU: 621 and rpb2: 702). Because of the high degree of sequence conservation, the LSU analysis alone was not able to resolve the generic limits in Cladosporiaceae, i.e. the Toxicocladosporium species did not form a monophyletic clade but were intermixed with species of Cladosporium (data not shown); thus, the combined ITS, LSU, and rpb2 sequences were more informative when used in a combined alignment. Fig. 1 shows a RaxML tree and node support values obtained using MP, ML and BI analyses. Parsimony analysis resulted in 68 trees (TL = 4735; CI = 0.347; RI = 0.705; RC = 0.245). These analyses show that all Toxicocladosporium species cluster together in a clade (MP-BS 100%, ML-BS 91%, BPP 0.98) closely related to Cladosporium, with the exception of T. leucadendri. The ex-type strain of the latter species (CBS 131317) and five other isolates formed a distinct linage phylogenetically, close but unrelated to, the genera Graphiopsis and Verrucocladosporium, representing a different genus we describe here as Neocladosporium, with N. leucadendri as the type species.
The second alignment included ITS and LSU sequences from all the available ex-type strains of Toxicocladosporium species with N. leucadendri as outgroup. To further improve the species resolution, actA, rpb2 and tub2 sequences were also included in this analysis. This second phylogeny included sequences from 26 isolates (including the outgroup) and 2 562 characters (actA: 247, ITS: 396, LSU: 780, rpb2: 724 and tub2: 415) of which 482 were parsimony-informative (actA: 25, ITS: 22, LSU: 28, rpb2: 230 and tub2: 125), 215 were variable and parsimony-uninformative (actA: 35, ITS: 48, LSU: 28, rpb2: 68 and tub2: 36), and 1 820 were constant (actA: 123, ITS: 311, LSU: 726, rpb2: 420 and tub2: 240). The results of this analysis are shown in Fig. 2. Parsimony analysis resulted in a single tree showing the best score (TL = 1870; CI = 0.570; RI = 0.651; RC = 0.371). The endophytic isolates grouped in two linages; 12 isolates formed a fully-supported clade close to T. banksiae (CBS 128215) (MP-BS 100%, ML-BS 100%, BPP 1) while one isolate (URM 7491 = CBS 141540) formed a moderately supported monotypic linage closely related to but distinct from T. ficiniae (CBS 136406) and T. posoqueriae (CBS 133583) (MP-BS < 70%, ML-BS 81%, BPP 0.95). These two groups are described here as the new species Toxicocladosporium cacti (ex-type culture URM 7489= CBS 141539) and Toxicocladosporium immaculatum (ex-type culture URM 7491 = CBS 141540), respectively.
In addition, the ex-type strains of T. chlamydosporum (CBS 124157) and T. velox (CBS 124159) always clustered together with high support values (MP-BS 100%, ML-BS 100%, BPP 1.00). From these phylogenetic results and based on the few nucleotide differences between the two species (actA: 1 nt and 1 gap, ITS: 5 nt, LSU: 1 nt, rpb2:0 nt and TUB: 0 nt) and given that both species show similar morphological and ecological features, we treat the name T. velox as a synonym of T. chlamydosporum.
LSU and ITS were informative loci to verify the relationship between genera and species groups. However, the actA, rpb2 and tub2 sequences were more informative to distinguish related species, especially in the case of T. cacti, which is closely related to T. banksiae.
Our phylogenetic analyses revealed that the endophytic fungi from cactus species previously identified as Toxicocladosporium represent two new species in this genus. These newly proposed species are established based on phylogenetic analyses and morphological features. In addition, we introduce a new generic name, Neocladosporium to accommodate “Toxicocladosporium” leucadendri, which is not congeneric with Toxicocladosporium. In this section a bibliographic synopsis of the genus is compiled including key morphological features for identification, known host affiliations, substrates, and geographic distribution for all the currently accepted species of Toxicocladosporium. Table 2 summarises key morphological features of the Neocladosporium and Toxicocladosporium species included here.
Neocladosporium J.D.P. Bezerra, Sandoval-Denis, C.M. Souza-Motta & Crous, gen. nov. MycoBank MB820266
Etymology: Named because of its similarity to the genus Cladosporium.
Diagnosis: Differs from Toxicocladosporium by its verruculose to warty ramoconidia, and from Cladosporium s. str. by its dark, thick-walled conidial and conidiophore septa, also lacking the typical coronate Cladosporium scar.
Type species: Neocladosporium leucadendri (Crous) J.D.P. Bezerra et al. 2017 (syn. Toxicocladosporium leucadendri Crous 2011).
Description: Mycelium consisting of pale brown, smooth, branched, septate hyphae. Conidiophores solitary, erect, unbranched or branched above, subcylindrical, straight to flexuous, apical septum becoming dark brown and thickened. Conidiogenous cells integrated, polyblastic, terminal and lateral, subcylindrical, smooth, brown; scars truncate, thickened and darkened. Ramoconidia medium brown, verruculose to warty, giving rise to branched chains of conidia, subcylindrical, polyblastic, brown, verruculose to warty, 0–1-septate, frequently forking close to apex; scars darkened, thickened. Intercalary conidia subcylindrical to fusoid-ellipsoidal, brown, smooth to somewhat warty. Small terminal conidia fusoid-ellipsoidal, brown, smooth; hila thickened and darkened.
Neocladosporium leucadendri (Crous) J.D.P. Bezerra, Sandoval-Denis, C.M. Souza-Motta & Crous, comb. nov. MycoBank MB 820267 (Fig. 3)
Basionym: Toxicocladosporium leucadendri Crous, Persoonia 27: 157 (2011).
Type: South Africa: Western Cape Province: Hermanus, Fernkloof Nature Reserve, on leaves of Leucadendron sp. (Proteaceae), 4 May 2010, P.W. Crous (CBS H-20774 — holotype; CPC 18315 = CBS 131317 — culture ex-type).
Description: Crous et al. (2011a).
Substrate and distribution: On leaves of Leucadendron sp. (Proteaceae) in the Western Cape province of South Africa (Crous et al. 2011a). On leaves of Kunzea pauciflora (Myrtaceae), and Banksia media, Hakea sp., and Petrophile sp. (Proteaceae) in Western Australia.
Other material examined: Australia: Western Australia: Albany, Fitzgerald River National Park, Point Ann, on leaves of Banksia media (Proteaceae), 21 Sep. 2015, P.W. Crous (CPC 29237); Denmark, Lights Beach, on leaves of Hakea sp. (Proteaceae), 19 Sep. 2015, P.W. Crous (CPC 29166); Wellstead, Cape Riche, on leaves of Hakea marginata (Proteaceae), 21 Sep. 2015, P.W. Crous (CPC 29092); ibid., on leaves of Kunzea pauciflora (Myrtaceae), 21 Sep. 2015, P.W. Crous (CPC 29090); Williams, Williams Nature Reserve, on leaves of Petrophile sp. (Proteaceae), 18 Sep. 2015, P.W. Crous (CPC 29545).
Notes: Crous et al. (2011a) published the strain CPC 18315 = CBS 131317 as T. leucadendri based on phylogenetic analyses using LSU and ITS sequences, and morphological characters. According to these authors, based on a combination of culture characteristics, conidiophore and conidial dimensions, it differs from known taxa, many of which also occur in the fynbos vegetation (Crous et al. 2011b). A megablast search of the NCBIs GenBank nucleotide sequence database using the ITS and LSU sequences of N. leucadendri retrieved as closest hits Graphiopsis chlorocephala and Verrucocladosporium dirinae, amongst others. In our phylogenetic analyses this strain appeared in a single lineage closely related to Graphiopsis chlorocephala and Verrucocladosporium dirinae as shown before by Crous et al. (2011a), but clearly separated from members of Toxicocladosporium (Fig. 1). Morphologically, Neocladosporium leucadendri is very similar to Toxicocladosporium species, but can be distinguished from it by size and ornamentation of ramoconidia (verruculose to warty) and ramoconidia frequently forking close to the apex. Sequences of ITS and LSU rDNA or rpb2 are the best approach to separate N. leucadendri from Toxicocladosporium species and related genera. Also very similar to Cladosporium s. str., but differing in the size and ornamentation of the ramoconidia (verruculose to warty), which are frequently forking close to apex, dark, thick-walled conidial and conidiophore septa, and lacking the typical coronate Cladosporium scar type similar to Toxicocladosporium (David 1997, Crous et al. 2007); it differs from Graphiopsis which has morphological peculiarities on its conidiophores (cladosporioid and periconioid morphs), conidiogenous loci and hila (Schubert et al. 2007a, Braun et al. 2008); from Rachicladosporium which has an apical conidiophore rachis with inconspicuous to subconspicuous scars and unthickened, not darkened-refractive conidial hila (Crous et al. 2007); and from Verrucocladosporium which has mainly an unusual conidial and hyphal ornamentation (Crous et al. 2007). Because of our phylogenetic results and morphological observations, the new generic name Neocladosporium, is proposed to accommodate N. leucadendri.
Toxicocladosporium Crous & U. Braun, Stud. Mycol. 58: 39 (2007).
Type species: Toxicocladosporium irritans Crous & U. Braun 2007.
Notes: Toxicocladosporium was introduced by Crous et al. (2007) to accommodate cladosporium-like fungi having distinct dark, thick-walled conidial and conidiophore septa, and lacking the typical coronate Cladosporium scar type. After this original publication, several new species isolated from different substrates and hosts were introduced in this genus using morphological characters and phylogenetic analyses of ITS and LSU sequences (Crous et al. 2007, 2009a, 2010a, b, 2011a, 2012a, b, 2013, 2014, 2016, Crous & Groenewald 2011).
Toxicocladosporium banksiae Crous et al., Persoonia 25: 147 (2010).
Type: Australia: Queensland: Noosa National Park, 26°34′14.0″S 153°4′21.6″E, on leaves of Banksia sp., 13 July 2009, P.W. Crous et al. (CBS H-20496 — holotype; CPC 17281, CPC 17280 = CBS 128215 — cultureex-type).
Description and illustration: Crous et al. (2010).
Substrate and distribution: On leaves of Banksia sp. (Proteaceae), Australia (Crous et al. 2010b).
Notes: According to Crous et al. (2010b), the ITS and LSU sequences of T. banksiae are close to those of T. chlamydosporum and T. irritans. The ITS sequences of T. banksiae also differ from those of T. protearum. Morphologically, T. banksiae differs from these three species in the size and shape of the intercalary and terminal conidia, ramoconidia, and presence or absence of chlamydospores. In our phylogeny using five different loci, this species is closely related to the new species T. cacti which differs in microconidiophore size [10−10 × 2.5−1 µm (aseptate) in T. banksiae vs. 21.5−34.5 × 2−5 µm (0–1-septate) in T. cacti], ramoconidia (14−25 × 2.5−1 µm vs. 10−20.5 × 2−3 µm), intercalary conidia (10−20 × 2.5−3.5 µm vs. 6−9 × 2.5−3 µm), terminal conidia (7−11 × 2−3 µm vs. 5−6 × 2−2.5 µm), and culture characteristics (colonies olivaceous grey reaching up to 7 mm diam in 2 wk in T. banksiae vs. colonies pale grey to grey, growing up to 30 mm diam in 3 wk and presence of a pale brown to brown exudate in T. cacti).
Toxicocladosporium cacti J.D.P. Bezerra, C.M. Souza-Motta & Crous, sp. nov.
Etymology: Named after the nature of the host, a cactus, from which it was isolated.
Diagnosis: Differs from T. banksiae in its slightly smaller and less septate microconidiophores and conidia, and by its pale grey to grey colonies.
Type: Brazil: Pernambuco: Catimbau National Park, 8°36′35″S 37°14′40″W, as endophytic fungus from cactus Pilosocereus gounellei subsp. gounellei, Sep. 2013, J.D.P. Bezerra (URM 90068 — holotype; URM 7489 = CBS 141539 — culture ex-type).
Other material examined: Brazil: Pernambuco: Catimbau National Park, 8°36′35″S 37°14′40″W, as endophytic fungus from cactus Pilosocereus gounellei subsp. gounellei, Sep. 2013, J.D.P. Bezerra (URM 7490 = CBS 141538, 188 JB, 191 JB, 192 JB, 195-2 JB, 225 JB, 231 JB, 235 JB, 236 JB, 226 JB, 261–2 JB).
Description: Mycelium consisting of branched, septate, smooth, brown, 2–2.5 µm wide hyphae; wall and septa becoming dark brown and thickened with age. Conidiophores dimorphic. Macroconidiophores solitary, arising from superficial mycelium, erect, brown, unbranched or branched above, finely verruculose, subcylindrical, straight to flexuous, up to 130 × 2−3.5 µm, 2–6-septate. Microconidiophores reduced to conidiogenous cells, rarely with one supporting cell, pale brown, smooth, erect, subcylindrical, 21.5−34.5 × 2−5 µm, 0–1-septate. Conidiogenous cells integrated, terminal or lateral, smooth, brown, 13−16 × 2−3 µm, proliferating sympodially with 1–2 apical loci; scars truncate, thickened and darkened, 1–1.5 µm wide. Conidia catenate in branched or unbranched chains, pale brown, thick-walled, septa dark and thick or inconspicuous, finely verruculose. Primary ramoconidia brown, finely verruculose, 0–1-septate, ellipsoidal to subcylindrical, 10−14(−20.5) × 2−3 µm; secondary ramoconidia brown, finely verruculose, 0-1-septate, ellipsoidal to subcylindrical, 7−10(−14) × 2−3 µm; scars darkened, thickened, 0.5−1 µm wide. Intercalary conidia subcylindrical to fusoid-ellipsoidal, 0–1-septate, brown, finely verruculose, 6−9 × 2.5−3 µm. Small terminal conidia fusoid-ellipsoidal, aseptate, brown, finely verruculose, 5−6 × 2−2.5 µm; hila thickened and darkened, 0.5–1 µm wide.
Culture characteristics (in a day-night cycle, 22 °C after 3 wk): Colonies on MEA are slightly folded and sulcate, velvety, pale grey to grey with a pale grey rim, reverse dark grey, reaching 30 mm diam; on OA flat to semi erumpent, spreading, with sparse to moderate aerial mycelium, smooth, surface and reverse pale grey to grey, to 29 mm; and on PDA surface and reverse olivaceous grey, to 25 mm. Exudate pale brown to brown observed on cultures growing on MEA and PDA.
Substrate and distribution: An endophytic fungus isolated from the cactus Pilosocereus gounellei subsp. gounellei (Cactaceae), Brazil.
Notes: Toxicocladosporium cacti is phylogenetically related to T. banksiae but differs morphologically from it in microconidiophore size and septation [21.5−34.5 × 2−5 µm (0–1-septate) vs. 10−10 × 2.5−4 µm], smaller ramoconidia (10−20.5 × 2−3 µm vs. 14−25 × 2.5−4 µm), intercalary conidia (6−9 × 2.5−3 µm vs. 10−20 × 2.5−3.5 µm), and small terminal conidia (5−6 × 2−2.5 µm vs. 7−11 × 2−3 µm). Furthermore, the culture characteristics are different from those of T. banksiae, colonies pale grey to grey, growing to 30 mm diam in 3 wk with exudate pale brown to brown in T. cacti vs. colonies olivaceous grey reaching up to 7 mm diam after 2 wk in T. banksiae.
Toxicocladosporium chlamydosporum Crous & M.J. Wingf., Persoonia 22: 90 (2009).
Synonym: Toxicocladosporium velox Crous & M.J. Wingf., Persoonia 22: 92 (2009); as ‘veloxum’.
Types: Madagascar: Morondavo, on leaf of Eucalyptus camaldulensis, Aug. 2007, M.J. Wingfield (CBS H-20193 — holotype of T. chlamydosporum; CPC 15709 = CBS 124157 — culture ex-type); ibid. (CBS H-20196 — holotype of T. velox; CPC 15736 = CBS 124159 — culture ex-type).
Description: Mycelium consisting of branched, septate, smooth, brown, 2–3 µm wide hyphae, containing swollen, globose, dark brown chlamydospore-like cells to 12 µm diam. Conidiophores dimorphic. Macroconidiophores solitary, erect, arising from superficial mycelium, penicillate, subcylindrical, straight to once geniculate-sinuous, medium to dark brown, smooth to finely verruculose, 20–60 µm long, 3–5 µm wide at base, 1–4-septate, not swollen, and lacking rhizoids. Microconidiophores erect, subcylindrical, to 15 µm tall and 5 µm wide, 0–1-septate, medium brown. Conidiogenous cells terminal, integrated, subcylindrical, straight, medium brown, 10−25 × 3−4 µm, smooth to finely verruculose; loci terminal and lateral, flat tipped, thickened, darkened, at times subdenticulate, (0.5−)1−2 µm wide. Conidia in branched chains, brown, smooth to finely verruculose, ellipsoid to cylindrical-oblong. Primary ramoconidia rarely observed, 0–1-septate, fusoid-ellipsoidal to subcylindrical, (15−)16−17(−18) × (2.5−)3−4 µm. Secondary ramoconidia 0–1-septate, fusoid-ellipsoidal, (9−)10−14(−16) × (2.5−)3−4 µm. Intercalary conidia 0–1-septate, fusoid-ellipsoidal, (8−)9− 11 (−12) × 2.5−3(−3.5) µm. Small terminal conidia aseptate, fusoid-ellipsoidal, 6−10 × 2−2.5(−3) µm (conidia dark brown and verruculose on MEA) (based on Crous et al. 2009a).
Culture characteristics (in the dark, at 25 °C after 1 mo): Colonies on MEA erumpent, spreading, with sparse aerial mycelium; surface folded, irregular and sectored, with feathery margin, centre pale olivaceous grey to fuscous-black, outer region olivaceous grey to greyish sepia; reverse iron-grey to dark grey; reaching up to 25 mm diam. Black sclerotial bodies on MEA, consisting of an agglomeration of chlamydospore-like cells; they remain sterile, and eventually resemble hollow fruiting bodies, although they lack an ostiole or defined wall. On OA spreading, flat, with sparse aerial mycelium, and even catenulate margin; surface iron-grey with patches of pale olivaceous grey to smoke-grey; colonies reaching up to 30 mm diam (Crous et al. 2009a).
Substrate and distribution: On leaves of Eucalyptus camaldulensis (Myrtaceae), Madagascar (Crous et al. 2009a).
Notes: Crous et al. (2009a) described this species using ITS and LSU sequences, and morphological characters to differentiate it from T. irritans. Toxicocladosporium chlamydosporum differs from other species in the genus in the presence of larger ramoconidia, and longer, narrower intercalary conidia, and in that it forms chlamydospores and sclerotial bodies in culture. Toxicocladosporium velox was isolated from the same leaf spot (Crous et al. 2009a). Based on the limited nucleotide differences and their morphological similarity, we consider T. velox a synonym of T. chlamydosporum. A revised description is provided to enable T. chlamydosporum in its expanded circumscription to be distinguished from other species in the genus. This species is closely related to T. protearum which differs from it mainly in the size and degree of septation of its conidiophores [20−60 µm × 3−5 µm (1–4-septate) in T. chlamydosporum vs. 30−80 µm × 3−4 µm (1–8-septate) in T. protearum], ramoconidia (15−18 × 2.5−4 µm vs. 15−20 × 2.5−3.5 µm), and intercalary and terminal conidia (8−11 × 3−3.5 µm vs. 9−16 × 2−3 µm).
Toxicocladosporium ficiniae Crous & A.R. Wood, Persoonia 31:191 (2013).
Type: South Africa: Western Cape Province: Brackenfell, Cape Town, Bracken Nature Reserve, on leaves of Ficinia indica (Cyperaceae), 18 Aug. 2012, A.R. Wood (CBS H-21413 — holotype; CPC 21283, CPC 21282 = CBS 136406 — culture ex-type).
Description and illustration: Crous et al. (2013).
Substrate and distribution: On leaves of Ficinia indica (Cyperaceae), South Africa (Crous et al. 2013).
Notes: Toxicocladosporium ficiniae is phylogenetically related to T. posoqueriae which differs in conidiophore size and septation [10−40 × 3−5 µm (1–15-septate) vs. 50−200 × 4−7 µm (1–3-septate) in T. posoqueriae], and sizes of the conidiogenous cells (5−15 × 2.5−4 µm vs. 10−20 × 4−7 µm), primary ramoconidia (15−35 × 3−4 µm vs. 5−15 × 4−5 µm), and terminal conidia (7−9 × 2.5−3 µm vs. 4−7 × 3−4 µm).
Toxicocladosporium hominis Sandoval-Denis et al., Persoonia 36: 421 (2016).
Type: USA: Florida: Daytona Beach, from human bronchoalveolar lavage fluid, D.A. Sutton (FMR H-13297 — holotype; CBS H-22331 — isotype; FMR 13297 = UTHSCSA DI-13−172 = CBS 140694 — cultures ex-type).
Description and illustration: Crous et al. (2016).
Substrate and distribution: From human bronchoalveolar lavage fluid, USA (Crous et al. 2016).
Notes: Toxicocladosporium hominis is phylogenetically related and morphologically similar to T. strelitziae (Crous et al. 2012b), but differs from T. strelitziae in the production of larger conidiogenous cells (13−30 × 3−4 µm vs. 10−15 × 2.5−3.5 µm) and intercalary conidia (9−16 × 3−4 µm vs. 10−12 × 2−2.5 µm). In addition, the latter species has smooth to verruculose ramoconidia, secondary ramoconidia and intercalary conidia, without constrictions in the medial portion or at the septum (Crous et al. 2016). Phylogenetically, this species is a distinct taxon closely related to T. strelitziae (Fig. 2).
Toxicocladosporium immaculatum J.D.P. Bezerra, C.M. Souza-Motta & Crous, sp. nov. MycoBank MB820265 (Fig. 5)
Etymology: Named after its pristine, well-developed, penicillate conidiophores.
Diagnosis: Differs from most Toxicocladosporium species by its red to dark red pigmented colonies when grown on OA. Different from T. ficiniae mainly by the larger and less septate conidiophores with shorter primary and secondary ramoconidia. Distinguished from T. posoqueriae by the slightly reduced conidiophores and conidia, and from T. rubrigenum by its less septate macroconidiophores, shorter microconidiophores and somewhat larger conidia.
Type: Brazil: Pernambuco: Itaíba, Curral Velho Farm, 9° 08.895 S 37° 12.069 W, as endophyte from cactus Tacinga inamoena, Sep. 2013, J.D.P. Bezerra (URM 90069 — holotype; URM 7491 = CBS 141540 — culture ex-type).
Description: Mycelium on SNA consisting of branched, septate, smooth to verruculose, pale brown, 2–3 µm wide hyphae. Conidiophores dimorphic, arising from superficial mycelium, erect to sinuous, brown, unbranched, finally verruculose, subcylindrical, straight to flexuous. Macroconidiophores up to 100 × 2−3.5 µm, 2–5-septate. Microconidiophores sometimes reduced to conidiogenous cells on hyphae, pale brown, smooth to finally verruculose, flexuous, subcylindrical, 12−25 × 2.5−3.5 µm, 0–1-septate. Conidiogenous cells integrated, polyblastic, terminal and lateral, smooth, becoming verruculose, brown, 10−14 × 2.5−3.5 µm; scars truncate, thickened and darkened, 1.5–2 µm wide. Primary ramoconidia medium brown, finely verruculose, 0–1-septate, subcylindrical, 14.5−22.5 × 2−4 µm. Secondary ramoconidia giving rise to branched chains of conidia, subcylindrical, polyblastic, brown, finely verruculose, 0–1-septate, (7−)8−14(−18.5) × 2−3 µm; scars darkened, thickened, 0.5–1 µm wide. Intercalary conidia subcylindrical to fusoid-ellipsoidal, brown, finely verruculose to verruculose, 11.5−13 × 2.5−3 µm. Small terminal conidia fusoid-ellipsoidal, brown, finely verruculose, 8−10(−11) × 2−3 µm; hila thickened and darkened, 0.5–1 µm wide.
Culture characteristics (in a day-night cycle, at 22 °C after 3 wk): Colonies on MEA are folded and sulcate, velvety, pale grey to olive-yellowish with a very light grey rim, reverse dark brown, reaching 33 mm diam; on OA flat, spreading, with sparse to moderate aerial mycelium, smooth, surface olive, with a light grey rim, reverse dark brown, red to dark red pigmentation produced, growing up to 33 mm diam; and on PDA surface olivaceous to olivaceous yellowish, reverse dark green, with sparse to moderate aerial mycelium, reaching up to 33 mm diam. Exudate pale brown to brown on MEA and PDA.
Substrate and distribution: As an endophyte isolated from the cactus Tacinga inamoena (Cactaceae), Brazil.
Notes: Toxicocladosporium immaculatum is phylogenetically closely related to T. ficiniae, T. posoqueriae and T. rubrigenum (Fig. 2). It differs morphologically from T. ficiniae in conidiophore size and septation [up to 100 × 2−3.5 µm (2–5-septate) vs. 10−10 × 3−5 µm (1–15-septate) in T ficiniae], conidiogenous cells (10−14 × 2.5−3.5 µm vs. 5−15 × 2.5−4 µm in T ficiniae), ramoconidia size (primary 14.5−22.5 × 2−4 µm and secondary 7−18.5 × 2−3 µm vs. primary 15−35 × 3−4 µm and secondary 12−20 × 2.5−3 µm in T ficiniae) and intercalary conidia (11.5−13 × 2.5−3 µm vs. 9−11 × 2.5−3 µm in T. ficiniae). It differs from T. posoqueriae in the size of the conidiophores [to 100 × 2−3.5 µm (2−5-septate) vs. 50−200 × 4−7 µm (1–3-septate) in T. posoqueriae], conidiogenous cells (10−14 × 2.5−3.5 µm vs. 10−20 × 4−7 µm in T. posoqueriae), ramoconidia (primary 14.5−22.5 × 2−4 µm and secondary 7−18.5 × 2−3 µm vs. 5−15 × 4−5 µm in T. posoqueriae) and terminal conidia (8−11 × 2−3 µm vs. 4−7 × 3−4 µm in T. posoqueriae). Toxicocladosporium rubrigenum differs in the size of the conidiophores [macroconidiophores to 100 × 2−3.5 µm (2–5-septate) vs. to 100 µm × 2−4 µm (1–8-septate) in T. rubrigenum and microconidiophores 12−25 × 2.5−3.5 µm vs. to 30 × 2−3 µm in T. rubrigenum], conidiogenous cells (10−14 × 2.5−3.5 µm vs. 15−20 × 2.5−3 µm in T. rubrigenum), ramoconidia (primary 14.5−22.5 × 2−4 µm and secondary 7−18.5 × 2−3 µm vs. primary 13−16 × 2.5−3.5 µm and secondary 9−14 × 2.5−3.5 µm in T. rubrigenum), intercalary conidia (11.5−13 × 2.5−3 µm vs. 7−9 × 2−2.5 µm in T. rubrigenum), and terminal conidia (8−11 × 2−3 µm vs. 4−7 × 2−2.5 µm in T. rubrigenum). Furthermore, T. immaculatum also differs from these species in colony colour, the presence of a red to dark red pigmentation in OA medium, and slower growth rates.
Toxicocladosporium irritans Crous & U. Braun, Stud. Mycol. 58: 39 (2007).
Type: Suriname: Paramaribo: isolated from mouldy paint, Feb. 1958, M.B. Schol-Schwarz (CBS H-19892 — holotype; CBS 185.58 — culture ex-type).
Description and illustration: Crous et al. (2007).
Substrate and distribution: Isolated from mouldy paint, Suriname (Crous et al. 2007); ancient laid-paper documents, Portugal (Mesquita et al. 2009); associated with patients with atopic dermatitis, Japan (Zhang et al. 2011); colonizing tattoo inks, Italy (Bonadonna et al. 2014); on parchment manuscripts, Italy (Piñar et al. 2015); from human blood and finger nail, USA (Sandoval-Denis et al. 2015); on Adansonia digitata, South Africa (Cruywagen et al. 2015); and on equipment used in the International Space Station or Space Shuttle, Japan (Satoh et al. 2016).
Notes: Crous et al. (2007) described Toxicocladosporium irritans as producing volatile metabolites, which cause a skin rash within minutes of opening an inoculated dish for microscopic examination. Morphologically and phylogenetically it is very similar to Cladosporium s. str., and produces dimorphic conidiophores, which is also a feature commonly observed in that genus. It is distinct in having dark, thick-walled conidial and conidiophore septa, and lacking the typical coronate Cladosporium scar type (David 1997, Crous et al. 2007). In our phylogenetic analyses (Fig. 2), T. irritans forms a lineage related to T. rubrigenum and T. hominis. It differs from T. rubrigenum in the size and septation of the conidiophores [30−60 × 4−6 µm (2–7-septate) vs. to 100 µm × 2−4 µm (1–8-septate)], conidiogenous cells (7−12 × 3−4 µm vs. 15−20 × 2.5−3 µm), ramoconidia (7−15 × 3−5 µm vs. primary 13−16 × 2.5−3.5 µm and secondary 9−14 × 2.5−3.5 µm) and terminal conidia (5−10 × 3−5 µm vs. 4−7 × 2−2.5 µm). It differs from T. hominis in conidiophore size (70−113 × 3−3.5 µm), conidiogenous cells (13−30 × 3−4 µm), ramoconidia (primary 15−32 × 2−4 µm and secondary 11−15 × 2.5−4 µm), and intercalary conidia (9−16 × 3−4 µm).
Toxicocladosporium pini Crous & Y. Zhang ter, Persoonia 32: 269 (2014).
Type: China: Beijing, Badaling, 40°20′45.1″N 116°00′48.3″E, on needles of Pinus sp. (Pinaceae), 1 Sept. 2013, P.W. Crous & Y. Zhang (CBS H-21719 — holotype; CPC 23639 = CBS 138005 — culture ex-type).
Description and illustration: Crous et al. (2014).
Substrate and distribution: On needles of Pinus sp. (Pinaceae), China (Crous et al. 2014).
Notes: According to Crous et al. (2014), Toxicocladosporium pini is morphologically similar to T. pseudovelox (ramoconidia 0–1-septate, broadly ellipsoid to subcylindrical, 8−15 × 2.5−4 µm; intercalary and terminal conidia ellipsoid, 6−11 × 2−3 µm) and T. protearum (ramoconidia 0–1-septate, subcylindrical, 15−20 × 2.5−3.5 µm; intercalary and terminal conidia subcylindrical to narrowly fusoid-ellipsoidal, 9−16 × 2−3 µm). Based on conidial dimensions, T. pini can be distinguished from T. protearum, but because of its morphological similarity to T. pseudovelox it can only be distinguished from that species by DNA data. Phylogenetically, this species is positioned as a distinct lineage between T. protearum and T. strelitziae (Fig. 2).
Toxicocladosporium posoqueriae Crous & R.G. Shivas, Persoonia 29: 181 (2012).
Type: Australia: Northern Territory: Darwin, on leaves of Posoqueria latifolia (Rubiaceae), 12 Apr. 2011, R.G. Shivas (CBS H-21086 — holotype; CPC 19305 = CBS 133583 — culture ex-type).
Description and illustration: Crous et al. (2012b).
Substrate and distribution: On leaves of Posoqueria latifolia (Rubiaceae), Australia (Crous et al. 2012b).
Notes: According to Crous et al. (2012b), Toxicocladosporium posoqueriae differs from other members of the genus in that it has whorls of conidiogenous cells, resembling those of Parapericoniella asterinae (Heuchert et al. 2005, Bensch et al. 2012). This species is closely related to T. ficiniae which differs in conidiophore size and septation [50−200 × 4−7 µm (1–3-septate) vs. 10−40 × 3−5 µm (1–15-septate)], conidiogenous cells (10−20 × 4−7 µm vs. 5−15 × 2.5−4 µm), ramoconidia (5−15 × 4−5 µm vs. primary 15−35 × 3−4 µm and secondary 12−20 × 2.5−3 µm), intercalary conidia (9−11 × 2.5−3 µm) and terminal conidia (4−7 × 3−4 µm vs. 7−9 × 2.5−3 µm). It is also similar to the newly described T. immaculatum which differs from in the conidiophores [macroconidiophores to 100 × 2−3.5 µm (2–5-septate) and microconidiophores 12−25 × 2.5−3.5 µm], conidiogenous cells (10−14 × 2.5−3.5 µm), ramoconidia (primary 14.5−22.5 × 2−4 µm and secondary 7−18.5 × 2−3 µm), intercalary conidia (11.5−13 × 2.5−3 µm), and terminal conidia (8−11 × 2−3 µm).
Toxicocladosporium protearum Crous & Roets, Persoonia 25: 135 (2010).
Type: South Africa: Western Cape Province: Stellenbosch, J.S. Marais Garden, on leaves of Protea sp., 22 Apr. 2008, F. Roets (CBS H-20490 — holotype; CPC 15254 = CBS 126499 — culture ex-type).
Description and illustration: Crous et al. (2010a).
Substrate and distribution: On leaves of Protea sp. (Proteaceae), South Africa (Crous et al. 2010a).
Notes: Blast analyses of the LSU and ITS sequences of Toxicocladosporium protearum showed that it is closely related to T. chlamydosporum and T. irritans (Crous et al. 2010a). Morphologically it differs from T. chlamydosporum which has smaller intercalary (8−11 × 3−3.5 µm) and terminal (6−10 × 2−3) conidia. Our phylogenetic analyses place T. protearum as a distinct lineage between T. chlamydosporum and T. pini which has larger macroconidiophores (30−90 × 3−4 µm), intercalary conidia (12−14 × 3 µm, 0–1-septate), and smaller terminal conidia (8−11 × 2.5−3 µm, 0–1-septate) (Crous et al. 2014).
Toxicocladosporium pseudovelox Crous, Persoonia 26: 81 (2011); as ‘pseudoveloxum’.
Type: South Africa: Western Cape Province: Hermanus, Fernkloof Nature Reserve, 34°23′38″S 19°16′9.7″E, on leaf bracts of Phaenocoma prolifera, 2 May 2010, K.L. Crous & P.W. Crous (CBS H-20535 — holotype; CPC 18257 = CBS 128775 — culture ex-type).
Description and illustration: Crous & Groenewald (2011).
Substrate and distribution: On leaf bracts of Phaenocoma prolifera (Asteraceae), South Africa (Crous & Groenewald 2011).
Notes: Crous & Groenewald (2011) showed that Toxicocladosporium pseudovelox was similar to T. chlamydosporum and other Toxicocladosporium species, but has shorter ramoconidia (8−15 × 2.5−4 µm) than T. chlamydosporum (15−18 × 2.5−4 µm). Toxicocladosporium pseudovelox is closely related to T. pini, which has larger conidiophores [macroconidiophores 30−90 × 3−4 µm (2–8-septate) and microconidiophores 10−17 × 3−4 µm], conidiogenous cells (5−20 × 3−3.5 µm), and intercalary conidia (12−14 × 3 µm, 0–1-septate). Toxicocladosporium pseudovelox was placed in a basal position at a highly supported node, which clustered it with T. pini, T. protearum, and T. chlamydosporum (Fig. 2).
Toxicocladosporium rubrigenum Crous & M.J. Wingf., Persoonia 22: 91 (2009).
Type: Madagascar: Morondavo, on leaf of Eucalyptus camaldulensis, Aug. 2007, M.J. Wingfield (CBS H-20195 — holotype; CPC 15735 = CBS 124158 — culture ex-type).
Description and illustration: Crous et al. (2009a).
Substrate and distribution: On leaf of Eucalyptus camaldulensis (Myrtaceae), Madagascar (Crous et al. 2009a).
Notes: This species differs from other Toxicocladosporium species in the production of densely branched penicillate conidiophores, and colonies that form a prominent red pigment on OA (Crous et al. 2009a). Toxicocladosporium rubrigenum is phylogenetically related to T. irritans and the new species T. immaculatum (Fig. 2). It differs from T. irritans in having longer and narrower conidiophores and conidiogenous cells (to 100 µm × 2−4 µm and 15−20 × 2.5−3 µm), as well as narrower ramoconidia (13−16 × 2.5−3.5 µm); and from T. immaculatum in the size of the conidiophores [macroconidiophores to 100 × 2−3.5 µm (2–5-septate) and microconidiophores 12−25 × 2.5−3.5 µm], conidiogenous cells (10−14 × 2.5−3.5 µm), ramoconidia (primary 14.5−22.5 × 2−4 µm and secondary 7−18.5 × 2−3 µm), intercalary conidia (11.5−13 × 2.5−3 µm), terminal conidia (8−11 × 2−3 µm), and culture characteristics (culture colour, pigmentation in the culture medium, production of exudates, and growth rates).
Toxicocladosporium strelitziae Crous, Persoonia 28: 179 (2012).
Type: South Africa: Mpumalanga Province: Kruger Game Reserve, Satara Rest Camp, on leaves of Strelitzia reginae (Strelitziaceae), 11 July 2011, P.W. Crous (CBS H-20970 — holotype; CPC 19763, CPC 19762 = CBS 132535 — culture ex-type).
Description and illustration: Crous et al. (2012b).
Substrates and distribution: On leaves of Strelitzia reginae (Strelitziaceae), South Africa (Crous et al. 2012b).
Notes: In a previous phylogenetic analysis, Toxicocladosporium strelitziae was placed in close proximity to T. pseudovelox (Crous et al. 2012b), but in the present analysis is placed in a lineage distant from that species with T. hominis as the closest relative (Fig. 2). Toxicocladosporium strelitziae is distinct from T. pseudovelox in having longer, narrower conidiophores (40−70 × 2−3.5 µm vs. 20−50 × 3−4 µm in T. pseudovelox), and larger, aseptate ramoconidia (12−20 × 2−3.5 µm vs. 8−15 × 2.5−4 µm, 0–1-septate in T. pseudovelox), and from T. hominis which has larger conidiophores (40−70 × 2−3.5 µm in T. strelitziae vs. 70−113 × 3−3.5 µm in T. hominis), conidiogenous cells (10−15 × 2.5−3.5 µm in T. strelitziae vs. 13−30 × 3−4 µm), ramoconidia [primary 12−20 × 2−3.5 µm (aseptate) and secondary 10−17 × 2−3.5 µm (aseptate) in T. strelitziae vs. primary 15−32 × 2−4 µm (0–2-septate) and secondary 11−15 × 2.5−4 µm (0−1-septate) in T. hominis], and intercalary conidia [10−12 × 2−2.5 µm in T. strelitziae vs. 9−16 × 3−4 µm (0–1-septate) in T. hominis].
The generic name Toxicocladosporium was introduced by Crous et al. (2007) to accommodate fungi similar to Cladosporium species but with different conidiophore and conidium morphology and phylogeny. Following this description, several new species were reported mainly as epiphytic, saprobic or phytopathogenic fungi from all continents (Crous et al. 2009a, 2010a, b, 2011a, 2012a, b, 2013, 2014, 2016, Crous & Groenewald 2011). However, in contrast to Cladosporium, Toxicocladosporium species had not previously been reported as endophytic fungi (Bensch et al. 2012, Bezerra et al. 2012, 2013). The isolation of novel Toxicocladosporium species as endophytic fungi from cacti in a tropical dry forest (Caatinga) in Brazil is reported here for the first time, and illustrates the diversity of fungi present as endophytes in different hosts and ecosystems.
In this study we revisited all currently published species of Toxicocladosporium using morphology and phylogenetic analyses (including three new loci). Based on these data we proposed two new species and one new closely related genus. Using a multigene phylogeny to recognise taxa in Dothideomycetes, Schoch et al. (2006) showed that Cladosporium belongs to the family Cladosporiaceae (an older name for the previously published Davidiellaceae). Later, during the investigation of cladosporium-like taxa, Crous et al. (2007) studied several isolates and proposed different genera based on their morphology and phylogeny, using sequences of part of the LSU nrDNA. In their phylogenetic reconstruction, six new genera were proposed, including Rachicladosporium, Toxicocladosporium, and Verrucocladosporium as incertae sedis. Bensch et al. (2012) monographed the genus Cladosporium and showed that it belongs to the family Cladosporiaceae (Capnodiales, Dothideomycetes) along with other four genera, Graphiopsis, Rachicladosporium, Toxicocladosporium, and Verrucocladosporium. Using ITS, LSU and rpb2 sequences from these genera, from all Toxicocladosporium species and from the other six families in Capnodiales we reconstructed the phylogenetic relationships of Cladosporiaceae and determined the phylogenetic position of each genus, including the newly described genus Neocladosporium (Fig. 1). Our results are similar to those of Bensch et al. (2012), who used LSU sequences to verify the relationship among these genera of the Cladosporiaceae.
Sandoval-Denis et al. (2015) studied clinical samples from the USA and reported the isolation of Cladosporium and Toxicocladosporium mainly obtained from respiratory specimens. These authors used phylogenetic analyses from all the available ITS and LSU sequences of Toxicocladosporium species except T. leucadendri, as well as morphological characters to identify two isolates as T. irritans, while a third isolate was unidentified, but phylogenetically positioned in a lineage between T. rubrigenum and T. strelitziae. In a subsequent paper the unidentified isolate was published as a new species, T. hominis (Crous et al. 2016). Sandoval-Denis et al. (2015) may not have included sequences from T. leucadendri in their analyses because this species appeared as a different genus, not belonging to Toxicocladosporium s. str. Toxicocladosporium leucadendri (CPC 18315 = CBS 131317) was published by Crous et al. (2011a) based on megablast searches in combination with culture characteristics, and conidiophore and conidial dimensions. Also, the phylogenetic analyses of the ITS and LSU sequences showed this strain in a single clade between Graphiopsis chlorocephala and Verrucocladosporium dirinae. The same result was observed in our phylogenetic analyses using the same loci, and also using actA, rpb2 and tub2 sequences. Based on these results, we introduced the new genus, Neocladosporium,with N. leucadendri as type species. Furthermore, based on the phylogenetic position and the small nucleotide differences between T. chlamydosporum and T. velox, we treat them as conspecific. These similarities can be also observed in the phylogenetic reconstruction published by Crous & Groenewald (2011), where T. velox and T. chlamydosporum are placed in the same clade with a high bootstrap support value. In addition, these authors used few morphological characters, such as the colour and size of conidia (darker brown and somewhat larger), absence and/or presence of chlamydospores and growth in culture to separate these species. These features are now combined in the revised circumscription of T. chlamydosporum presented here.
To improve the discrimination of species in the genus Toxicocladosporium, we generated actA, rpb2 and tub2 sequences from all the available ex-type strains as well as endophytic isolates generated in this study (Fig. 2). In our analyses using a combined matrix of ITS, LSU, actA, rpb2 and tub2 sequences, we recognise 13 species in this genus, including the two new species, T. cacti and T. immaculatum. As previously demonstrated in Cladosporium by different authors (Braun et al. 2003, Schubert et al. 2007b, Zalar et al. 2007, Bensch et al. 2010, 2012, 2015, Sandoval-Denis et al. 2016), ITS, and LSU sequences are less informative than actA, rpb2 and tub2 sequences to separate species in Toxicocladosporium. In our analyses, rDNA sequences were very similar among some species, but are useful to separate genera (LSU) and species groups (ITS). After inclusion of actA sequences, the third most informative region after rpb2 and tub2, respectively, the separation of species was improved. Sequences of rpb2, followed by tub2 were the best loci to recognise species in our analyses. We therefore recommend these markers as barcoding targets for species recognition as well as for the description of new taxa in addition to ITS and actA sequences in this genus. The inclusion of actA, rpb2 and tub2 sequences in our analyses was crucial to facilitate the separation of T. cacti from T. banksiae, since rDNA sequences from endophytic isolates were closely related to T. banksiae, but could not unambiguously resolve both species. In contrast, the actA, rpb2 or tub2 loci consistently separate these two taxa with high statistical confidence (data not shown). A similar situation was observed in Cladosporium for which a combined phylogenetic analysis including ITS, translation elongation factor 1-alpha (tef1) and actA loci has been adopted in order to separate species within that genus, with ITS being the least informative locus (Bensch et al. 2012, 2015, Sandoval-Denis et al. 2016).
Our study shows that Toxicocladosporium species, as those of Cladosporium (Bensch et al. 2012, Bezerra et al. 2012, 2013), may be isolated as endophytic fungi from plants growing in tropical dry regions. This report also expands our knowledge about endophytes associated with cacti and highlights the mostly underestimated fungal diversity associated with this little-studied group of host plants, and as well as the importance of protecting them in their natural habitats.
Australian Government Department of Agriculture (2015) Norfolk Island Quarantine Survey 2012–2014. https://doi.org/www.norfolkonlinenews.com/Literatu reRetrieve.aspx?ID=207325.
Barberan A, Ladaub J, Leffa JW, Pollardb KS, Menningerd HL, et al. (2015) Continental-scale distributions of dust-associated bacteria and fungi. Proceedings of the National Academy of Sciences, USA 112: 5756–5761.
Bensch K, Groenewald JZ, Dijksterhuis J, Starink-Willemse M, Andersen B, et al. (2010) Species and ecological diversity within the Cladosporium cladosporioides complex (Davidiellaceae, Capnodiales). Studies in Mycology 67: 1–94.
Bensch K, Braun U, Groenewald JZ, Crous PW (2012) The genus Cladosporium. Studies in Mycology 72: 1–401.
Bensch K, Groenewald JZ, Braun U, Dijksterhuis J, de Jesus Yanez-Morales M, et al. (2015) Common but different: The expanding realm of Cladosporium. Studies in Mycology 82: 23–74.
Bezerra JDP, Santos MGS, Svedese VM, Lima DMM, Fernandes MJS, et al. (2012) Richness of endophytic fungi isolated from Opuntia ficus-indica Mill. (Cactaceae) and preliminary screening for enzyme production. World Journal of Microbiology and Biotechnology 28: 1989–1995.
Bezerra JDP, Santos MGS, Barbosa RN, Svedese VM, Lima DMM, et al. (2013) Fungal endophytes from cactus Cereus jamacaru in Brazilian tropical dry forest: a first study. Symbiosis 60: 53–63.
Bezerra JDP, Oliveira RJV, Paiva LM, Silva GA, Groenewald JZ, et al. (2017) Bezerromycetales and Wiesneriomycetales ord. nov. (class Dothideomycetes), with two novel genera to accommodate endophytic fungi from Brazilian cactus. Mycological Progress 16: 297–309.
Bonadonna L, Briancesco R, Coccia AN, Fonda A, Rosa AL, et al. (2014) Valutazione delle caratteristiche microbiologiche di inchiostri per tatuaggi in confezioni integre e dopo I’apertura. Microbiologia Medica 29: 4807.
Braun U, Crous PW, Dugan FM, Groenewald JZ, de Hoog GS (2003) Phylogeny and taxonomy of cladosporium-like hyphomycetes, including Davidiella gen. nov, the teleomorph of Cladosporium s. str. Mycological Progress 2: 3–18.
Braun U, Crous PW, Schubert K (2008) Taxonomic revision of the genus Cladosporium s. lat. 8. Reintroduction of Graphiopsis (=Dichocladosporium) with further reassessments of clado-sporioid hyphomycetes. Mycotaxon 103: 207–216.
Carbone I, Kohn LM (1999) A method for designing primer sets for speciation studies in filamentous ascomycetes. Mycologia 91: 553–556.
Cho K-H, Kim D-C, Yoon C-S, Ko WM, Lee SJ, et al. (2016) Anti-neuroinflammatory effects of citreohybridonol involving TLR4-MyD88-mediated inhibition of NF-KB and MAPK signaling pathways in lipopolysaccharide-stimulated BV2 cells. Neuro-chemistry International 95: 55–62.
Crous PW, Gams W, Stalpers JA, Robert V, Stegehuis G (2004) MycoBank: an online initiative to launch mycology into the 21st century. Studies in Mycology 50: 19–22.
Crous PW, Braun U, Schubert K, Groenewald JZ (2007) Delimiting Cladosporium from morphologically similar genera. Studies in Mycology 58: 33–56.
Crous PW, Wingfield MJ, Groenewald JZ (2009a) Niche sharing reflects a poorly understood biodiversity phenomenon. Persoonia 22: 83–94.
Crous PW, Schoch CL, Hyde KD, Wood AR, Gueidan C, et al. (2009b) Phylogenetic lineages in the Capnodiales. Studies in Mycology 64: 17–47.
Crous PW, Verkley GJM, Groenewald JZ, Samson RA (2009c) Fungal Biodiversity. [CBS Laboratory Manual Series No. 1.] Utrecht: Centraalbureau voor Schimmelcultures.
Crous PW, Groenewald JZ, Roets F (2010a) Fungal Planet 57. Toxicocladosporium protearum. Persoonia 25: 134–135.
Crous PW, Groenewald JZ, Shivas RG, McTaggart AR (2010b) Fungal Planet 63. Toxicocladosporium banksiae. Persoonia 25: 146–147.
Crous PW, Groenewald, JZ (2011) Why everlastings don’t last. Persoonia 26: 70–84.
Crous PW, Summerell BA, Shivas RG, Romberg M, Mel’nik VA, et al. (2011a) Fungal Planet description sheets: 92–106. Persoonia 27: 130–162.
Crous PW, Summerell BA, Swart L, Denman S, Taylor JE, et al. (2011b) Fungal pathogens of Proteaceae. Persoonia 27: 20–45.
Crous PW, Summerell BA, Shivas RG, Burgess TI, Decock CA, et al. (2012a) Fungal Planet description sheets: 107–127. Persoonia 28: 138–182.
Crous PW, Shivas RG, Wingfield MJ, Summerell BA, Rossman AY, et al. (2012b) Fungal Planet description sheets: 128–153. Persoonia 29: 146–201.
Crous PW, Wingfield MJ, Guarro J, Cheewangkoon R, van der Bank M, et al. (2013) Fungal Planet description sheets: 154–213. Persoonia 31: 188–296.
Crous PW, Shivas RG, Quaedvlieg W, van der Bank M, Zhang Y, et al. (2014) Fungal Planet description sheets: 214–280. Persoonia 32: 184–306.
Crous PW, Wingfield MJ, Richardson DM, Leroux JJ, Strasberg D, et al. (2016) Fungal Planet description sheets: 400–468. Persoonia 36: 316–458.
Cruywagen EM, Crous PW, Roux J, Slippers B, Wingfield MJ (2015) Fungi associated with black mould on baobab trees in southern Africa. Antonie van Leeuwenhoek 108: 85–95.
David JC (1997) A contribution to the systematics of Cladosporium: revision of the fungi previously referred to Heterosporium. Mycological Papers 172: 1–157.
Fisher PJ, Sutton BC, Petrini LE, Petrini O (1994) Fungal endophytes from Opuntia stricta: a first report. Nova Hedwigia 59: 195–200.
Fonseca-Garc’ia C, Coleman-Derr D, Garrido E, Visel A, Tringe SG, et al. (2016) The cacti microbiome: interplay between habitat-filtering and host-specificity. Frontiers in Microbiology 7: 150.
Freire KTLS, Ara’ujo GR, Bezerra JDP, Barbosa RN, Silva DCV, et al. (2015) Fungos endofiticos de Opuntia ficus-indica (L.) Mill. (Cactaceae) sadia e infestada por Dactylopius opuntiae (Cockerell, 1896) (Hemiptera: Dactylopiidae). Gaia Scientia 9: 104–110.
Glass NL, Donaldson GC (1995) Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes. Applied and Environmental Microbiology 61: 1323–1330.
Heuchert B, Braun U, Schubert K (2005) Morphotaxonomic revision of fungicolous Cladosporium species (hyphomycetes). Schlechtendalia 13: 1–78.
Huelsenbeck JP, Ronquist F (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17: 754–755.
Katoh K, Standley DM (2013) MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Molecular Biology and Evolution 30: 772–780.
Khidir HH, Eudy DM, Porras-Alfaro A, Herrera J, Natvig DO, et al. (2010) A general suite of fungal endophytes dominate the roots of two dominant grasses in a semiarid grassland. Journal of Arid Environments 74: 35–42.
Klymiuk AA, Taylor TN, Taylor EL, Krings M (2013) Paleomycology of the Princeton Chert 11. Dark-septate fungi in the aquatic angiosperm Eorhiza arnoldii indicate a diverse assemblage of root-colonizing fungi during the Eocene. Mycologia 105: 1100–1109.
Knapp DG, Kov’acs GM, Zajta E, Groenewald JZ, Crous PW (2015) Dark septate endophytic pleosporalean genera from semiarid areas. Persoonia 35: 87–100.
Kumar S, Stecher G, Tamura K (2015) MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Molecular Biology and Evolution 33: 1870–1874.
Loro M, Valero-Jim’enez CA, Nozawac S, M’arquez LM (2012) Diversity and composition of fungal endophytes in semiarid Northwest Venezuela. Journal of Arid Environments 85: 46–55.
Mason-Gamer R, Kellogg E (1996) Testing for phylogenetic conflict among molecular data sets in the tribe Triticeae (Gramineae). Systematic Biology 45: 524–545.
McCarthy CB, Diambra LA, Rivera Pomar RV (2011) Metagenomic analysis of taxa associated with Lutzomyia longipalpis, vector of visceral leishmaniasis using an unbiased high-throughput approach. PLoS Neglected Tropical Diseases 5: e1304.
Mesquita N, Portugal A, Videira S, Rodr’iguez-Echeverr’ia S, Bandeira AML, et al. (2009) Fungal diversity in ancient documentsL a case study on the Archive of the University of Coimbra. International Biodeterioration and Biodegradation 63: 626–629.
Nhạ PV, Giang HTT, Vượng PT, Thanh DT, Tuyết TT, et al. (2011) Giám dịnh một số chủng nấm ký sinh rệp sáp hai cà phê bằng phương pháp DNA. [Identification of some fungal strains parasitic on coffee scale insect by DNA.] Tạp chí Khoa học và Phát triển 5: 713–718.
Nylander JA (2004) MrModeltest. Version 2. Uppsala: Evolutionary Biology Centre, Uppsala University.
O’Donnell K, Sutton DA, Rinaldi MG, Sarver BA, Balajee SA, et al. (2010) Internet-accessible DNA sequence database for identifying fusaria from human and animal infections. Journal of Clinical Microbiology 48: 3708–3718.
Piñar G, Sterflinger K, Pinzari F (2015) Unmasking the measleslike parchment discoloration: molecular and microanalytical approach. Environmental Microbiology 17: 427–443.
Quaedvlieg W, Binder M, Groenewald JZ, Summerell BA, Carnegie AJ, et al. (2014) Introducing the Consolidated Species Concept to resolve species in the Teratosphaeriaceae. Persoonia 33: 1–40.
Rayner RW. (1970) A Mycological Colour Chart. Kew: Commonwealth Mycological Institute.
Redman RS, Sheehan KB, Stout RG, Rodriguez RJ, Henson JN (2002) Thermotolerance generated by plant/fungal symbiosis. Science 298: 1581
Sandoval-Denis M, Sutton DA, Martin-Vicente A, Cano-Lira JF, Wiederhold N, et al. (2015) Cladosporium species recovered from clinical samples in the United States. Journal of Clinical Microbiology 53: 2990–3000.
Sandoval-Denis M, Gen’e J, Sutton DA, Wiederhold NP, Cano-Lira JF, et al. (2016) New species of Cladosporium associated with human and animal infections. Persoonia 36: 281–298.
Satoh K, Yamazaki T, Nakayama T, Umeda Y, Alshahni MM, et al. (2016) Characterization of fungi isolated from the equipment used in the International Space Station or Space Shuttle. Microbiology and Immunology 60: 295–302.
Schoch C, Shoemaker RA, Seifert KA, Hambleton S, Spatafora JW, et al. (2006) A multigene phylogeny of the Dothideomycetes using four nuclear loci. Mycologia 98: 1041–1052.
Schubert K, Braun U, Groenewald JZ, Crous PW (2007a) Cladosporium leaf-blotch and stem rot of Paeonia spp. caused by Dichocladosporium chlorocephalum gen. nov. Studies in Mycology 58: 95–104.
Schubert K, Groenewald JZ, Braun U, Dijksterhuis J, Starink MS, et al. (2007b) Biodiversity in the Cladosporium herbarum complex (Davidiellaceae, Capnodiales), with standardisation of methods for Cladosporium taxonomy and diagnostics. Studies in Mycology 58: 105–156.
Silva-Hughes AF, Wedge DE, Cantrell CL, Carvalho CR, Pan Z, et al. (2015) Diversity and antifungal activity of the endophytic fungi associated with the native medicinal cactus Opuntia humifusa (Cactaceae) from the United States. Microbiological Research 175: 67–77.
Stamatakis A (2014) RAxML Version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics 30: 1312–1313.
Sun Y, Wang Q, Lu X, Okane I, Kakishima M (2012) Endophytic fungal community in stems and leaves of plants from desert areas in China. Mycological Progress 11: 781–790.
Suryanarayanan TS, Wittlinger SK, Faeth SH (2005) Endophytic fungi associated with cacti in Arizona. Mycological Research 109: 635–639.
Suryanarayanan TS, Murali TS, Thirunavukkarasu N, Rajulu MBG, Venkatesan G, et al. (2011) Endophytic fungal communities in woody perennials of three tropical forest types of the Western Ghats, southern India. Biodiversity and Conservation 20: 913–928.
Swofford DL (2003) PAUP*: phylogenetic analysis using parsimony (* and other methods). Version 4.0 beta. Sunderland, MA: Sinauer Associates.
Tanaka A, Cho O, Saito M, Tsuboi R, Kurakado S, et al. (2014) Molecular characterization of the skin fungal microbiota in patients with seborrheic dermatitis. Journal of Clinical & Experimental Dermatology 5: 239.
Videira SI, Groenewald JZ, Braun U, Shin HD, Crous PW (2016) All that glitters is not Ramularia. Studies in Mycology 83: 49–163.
Vilgalys R, Hester M (1990) Rapid identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. Journal of Bacteriology 172: 4238–4246.
White TJ, Bruns T, Lee J, Taylor J (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: PCR Protocols: a guide to methods and applications (Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds): 315–322. San Diego: Academic Press.
Wiens JJ (1998) Testing phylogenetic methods with tree congruence: phylogenetic analysis of polymorphic morphological characters in phrynosomatid lizards. Systematic Biology 47: 427–444.
Zalar P, de Hoog GS, Schroers H-J, Crous PW, Groenewald JZ, et al. (2007) Phylogeny and ecology of the ubiquitous saprobe Cladosporium sphaerospermum, with descriptions of seven new species from hypersaline environments. Studies in Mycology 58: 157–183.
Zhang E, Tanaka T, Tajima M, Tsuboi R, Nishikawa A, et al. (2011) Characterization of the skin fungal microbiota in patients with atopic dermatitis and in healthy subjects. Microbiology and Immunology 55: 625–632.
We thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) (Process 203132/2014-9), the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) and the Fundação de Amparo à Ciência e Tecnologia de Pernambuco (FACEPE) of Brazil for financial support and scholarships. We also thank Konstanze Bensch and David L. Hawksworth for the valuable comments and suggestions to improve the manuscript. We extend our thanks to the Universidade Federal de Pernambuco and to the technical staff, Eliane Silva-Nogueira and Luan Amim from the URM Culture Collection, and to Marjan Vermaas, Arien van Iperen and Mieke Starink-Willemse from the Westerdijk Fungal Biodiversity Institute. We also thank Tamara Caldas, Greicilene Albuquerque, Gianne Rizzuto, Karla Freire and other students of the Laboratório de Micologia Ambiental/UFPE for their technical help and processing of samples.
About this article
Cite this article
Bezerra, J.D.P., Sandoval-Denis, M., Paiva, L.M. et al. New endophytic Toxicocladosporium species from cacti in Brazil, and description of Neocladosporium gen. nov.. IMA Fungus 8, 77–97 (2017). https://doi.org/10.5598/imafungus.2017.08.01.06
- Endophytic fungi
- Multigene phylogeny