ABSTRACT
Stereum is a globally widespread genus of basidiomycete fungi with conspicuous shelf-like fruiting bodies. Several species have been extensively studied as pathogens and for medically relevant secondary metabolites, but prior to this publication the molecular phylogeny of Stereum had not been analyzed, despite pervasive morphological crypsis in the genus. Here, we provide a preliminary investigation into species boundaries among some North American Stereum. The nominal species Stereum ostrea has been referenced in field guides, textbooks, and scientific papers as a common fungus with a wide geographic range and even wider morphological variability. We use nuc rDNA ITS1-5.8S-ITS2 (ITS barcode) sequence data of specimens from midwestern and eastern North America, alongside morphological and ecological characters, to show that Stereum ostrea is a complex of at least three reproductively isolated species. Preliminary morphological analyses show that these three species correspond to three historical taxa that were previously synonymized with or confused for S. ostrea: Stereum fasciatum, Stereum lobatum, and Stereum subtomentosum. Stereum hirsutum ITS sequences taken from GenBank suggest that other Stereum species may also be species complexes. Future work should apply a multilocus approach and global sampling strategy to better resolve the taxonomy and evolutionary history of this important fungal genus.
INTRODUCTION
The most recent estimate of fungal diversity predicts that there are 2.2 to 3.8 million species of fungi, but only ~120 000 have been formally described, and fewer than 23 000 have ITS sequences available on NCBI GenBank (Hawksworth and Lücking 2017; Vu et al. 2014). New species of fungi are often discovered as cryptic species lumped under single, well-established names, such as with Amanita muscaria and Cantharellus cibarius (Geml et al. 2006; Buyck and Hofstetter 2011). While the members of some species complexes may initially be difficult to separate morphologically, distinguishing features can become clear after phylogenetic analysis and trait-mapping, as was shown with Fomes fomentarius and its neglected sister species F. inzengae (Peintner et al. 2019). Critically, differences among cryptic or apparently cryptic species can be economically relevant, such as with the fungal plant pathogen Magnaporthe grisea and the morphologically indistinguishable M. oryzae, which infect crabgrass and rice, respectively (Couch and Kohn, 2002). Fungi are not only incredibly speciose, but are also morphologically and ecologically diverse, thus presenting fantastic systems to explore how divergent selection along a variety of ecological axes might influence the origin of new biodiversity. However, in the absence of robust taxonomic resources, it is difficult to even begin studying which ecological factors might drive divergence in fungi.
One understudied fungal group that may be a good candidate for future ecological speciation research is Stereum, a diverse genus of shelf-like wood-decay fungi common in wooded biomes throughout the world. Though Stereum has been the focus of extensive bioprospecting (Doljak et al. 2006; Hybelhauerová et al. 2008; Tian et al. 2020), and some economically important species are relatively well-researched (Stenlid and Vasiliauskas 1998; Čermák et al. 2004), the below-genus level taxonomy of Stereum has not been subjected to phylogenetic analysis. This disconnect between the slow progress toward molecular taxonomy in Stereum and interest in the potentially useful properties of Stereum species can be problematic, especially where species descriptions include uninformative or deceptive morphological characters that may lead to misidentification of study specimens.
Stereum ostrea, like many other Stereum treated carelessly by systematists, is an exemplar of a species with a fraught taxonomic history (Lloyd 1913). While the name S. ostrea has been applied to collections around the world, it is unlikely that these varied collections are from a single phylogenetic species given pre-Anthropocene obstacles to dispersal. First used to describe a collection from the island of Java in Indonesia, the name S. ostrea has been consistently applied to specimens from North America since a publication by Lentz (1955) placed it in synonymy with two other putative species, S. fasciatum and S. lobatum, with which Welden (1971) and Chamuris (1985, 1988) concurred. Even before this, S. ostrea was considered either a synonym of S. fasciatum by some authors (Burt 1920; Pilát 1930; Banerjee 1935; Hendrickx 1948) or of S. lobatum by other authors (Massee 1890; Cooke 1892; Hohnel and Litschauer 1907; Reinking 1920; Boedijn 1940). Critically, Demoulin (1985) argued in favor of S. fasciatum and S. lobatum as distinct species, outlining their differences, but the literature remained divided (Eicker and Louw 1998). Currently, S. fasciatum is considered a synonym of S. ostrea (according to Index Fungorum in October 2020, http://indexfungorum.org/), and despite being an accepted name S. lobatum is rarely used. Another similar species, S. subtomentosum, has both been confused with S. fasciatum and proposed to be in a complex with S. hirsutum (Pouzar 1964; Welden 1971; Chamuris 1988; Ginns and Lefebvre 1993).
We hypothesize that S. ostrea in North America is not a single species but a complex of several species. To test this, and to begin developing a more integrated approach to Stereum taxonomy, we collected over 50 Stereum specimens from eastern North America and used nuc rDNA ITS1-5.8S-ITS2 (ITS barcode) sequence data to create a phylogenetic tree. Because sequence information from a single gene is not always enough to reliably delineate species, we coupled our phylogenetic approach with morphological and ecological data. Following our hypothesis that S. ostrea is a species complex, we predicted that differences in ITS sequences would correlate with differences in morphology and ecology (e.g., substrate), and as such would reveal evidence of more than one reproductively isolated Stereum species.
MATERIALS AND METHODS
Collection and Identification
We collected Stereum basidiocarps from locations in midwestern and eastern North America, with the most intensive sampling in Iowa and some supplemental collections made in Alabama and Florida (SUPPLEMENTARY TABLE 1). For identification we used the dichotomous key and morphological descriptions from Chamuris (1988), the most comprehensive and recent publication on Stereum. Based on morphology, we identified our collections as S. ostrea, S. hirsutum, S. complicatum, S. gausapatum, S. sanguinolentum, and S. striatum. We used iNaturalist to record photographs, dates of collection, and approximate GPS coordinates of collection location. We also recorded whether the substrate from which samples were collected was hardwood or conifer, and if the hymenium changed color when bruised. We air dried all collections and preserved samples in polyethylene plastic bags. In total, we collected and analyzed 49 samples, 36 of which are included in this paper (SUPPLEMENTARY TABLE 1).
All Stereum species in this study are found primarily on angiospermous wood, except for S. sanguinolentum which is primarily found on conifers. Stereum are occasionally reported on rare hosts and have been sequenced from atypical host species (Ginns and Lefebvre 1993; Jusino 2014). SD50 S. complicatum was found growing on a dead branch of loblolly pine still attached to the tree.
DNA Extraction, PCR and sequencing
We used a CTAB and liquid nitrogen method adapted from Chen et al. (2010) to extract DNA from 3 × 3 mm pieces of basidiocarp from each collection, taking care to exclude as much dirt and debris as possible. We used DNA diluted 1:20 with molecular grade water for PCR amplification using ITS1-F (CTTGGTCATTTAGAGGAAGTAA) and ITS4 (TCCTCCGCTTATTGATATGC) primers (Gardes and Bruns 1993; White et al. 1990), with the following thermocycler program: 3 min @ 94 C, 20-30 cycles (30 s @ 94 C, 30 s @ 55 C, 30 s @ 72 C), 2 min @ 72 C. We cleaned PCR products with Exo-SAP following manufacturer protocols. We sequenced PCR products in forward and reverse directions on an Applied Biosystems ABI 3730 DNA Analyzer (Thermo Fisher Scientific, Massachusetts) housed in the Roy J. Carver Center for Genomics in the University of Iowa Biology Department.
Phylogenetic Trees
We used Geneious 8.1.7 (http://www.geneious.com/) for alignment of forward and reverse sequences for each collection, which we then manually checked and trimmed. We chose Xylobolus subpileatus, another member of the Stereaceae, as an outgroup. This specimen was collected in Florida by S.G.D. and sequenced by the Smith Lab at the University of Florida. We also obtained North American and Eurasian S. hirsutum sequences from BOLD and GenBank. We used MAFFT (Katoh et al. 2002) via the CIPRES server (Miller et al. 2010) to align chosen sequences, then generated a maximum likelihood tree using RAxML (Stamatakis 2014) via CIPRES. For computation of the Bayesian tree, we used MrBayes 3.2.7 (Ronquist et al. 2012) with a GTR+I+Γ substitution model for 200 000 generations, with sampling every 100 generations.
RESULTS
Maximum likelihood (FIG. 1) and Bayesian trees (SUPPLEMENTARY FIG. 1) were largely in agreement, except for the placement of the single Stereum striatum sequence. Both trees show Stereum sorting out into several clades. Specimens initially identified as S. ostrea were monophyletic but split into three well-differentiated clades (FIG. 1). Sequences within the same clade differed from one another by 1–3%, while sequences among the three clades differed from one another by 7–14%.
Maximum likelihood phylogeny generated from ITS sequence data. Bootstrap values are above branches. Scale bar represents the number of nucleotide changes per site. The first three clades show strong support, suggesting S. ostrea in midwestern and eastern North America consists of three distinct species, which we identify as S. fasciatum, S. lobatum, and S. subtomentosum.
The “S. ostrea” specimens in the first clade (FIG. 1) all share features of the previously synonymized species S. fasciatum, as described by many authors including Burt (1920) and Demoulin (1985). These specimens feature a cap clothed in coarse hair that is resistant to wearing off in bands, and these hairs gather in individual clumps that are best observed with a hand lens or dissecting microscope (FIG. 2A–B). If bruised (when basidiocarps are fresh) or wetted (when dried) the hymenium (undersurface) does not stain color (FIG. 2A).
Morphological comparison of a-b. SD01 Stereum fasciatum, c-d. SD03 S. lobatum, e-f. SD12 S. subtomentosum. Arrows point to areas where hairs have worn away in concentric bands, exposing the chestnut-brown context; a. few very thin bands, c. wide bands, starting from cap edge, and e. irregular, uneven banding. The second column shows differences between species in cap hair texture; b. coarse and clumped in tufts, d. short, matted, and felted, and f. long, wooly and tufted.
Similarly, the “S. ostrea” specimens in the second clade (FIG. 1) showed features consistent with the previously synonymized S. lobatum (Burt 1920; Demoulin 1985). The cap is clothed in matted, felted hairs that quickly begin wearing off, from the edge of the cap inwards, in concentric bands exposing the chestnut-brown context beneath (FIG. 2C–D). When bruised or wetted the hymenium stains a bright yellow color (FIG. 2B).
“S. ostrea” specimens in the third clade shared some features with both S. fasciatum and S. lobatum, however the basidiocarps were usually thicker, more irregular, and a richer brown color both cap and hymenium. The cap is covered clumping hairs that are typically longer and woolier than specimens of the first clade (S. fasciatum), and wear off in bands more readily but not to the extent of specimens in the second clade (S. lobatum) (FIG. 2E–F). When bruised or wetted the hymenium stains a bright yellow color like S. lobatum (FIG. 2B). Overall, the morphology agrees with descriptions of S. subtomentosum (Pouzar 1964; Chamuris 1988).
Specimens identified before sequencing as belonging to other species of Stereum formed clades with ITS sequences differing within clades by 15%. These included S. complicatum, S. gausapatum, and S. sanguinolentum. The single sequence of S. striatum was placed differently in the two trees but differed from all other sequences by more than 7%. S. hirsutum sequences acquired from GenBank and BOLD databases differed from one another by 1–6%, but only as much as 4% within the two geographically isolated (North America vs. Eurasia) clades.
DISCUSSION
Together, our results support the hypothesis that the North American fungi that have been lumped under the name “S. ostrea” are actually at least three distinct species, reproductively isolated and differing in morphology. The inference of reproductive isolation is based on the observation that while all three species can be found growing together on the same substrate, there is strong concordance between ITS sequence identity and morphology, such that they appear not to be hybridizing despite often living in sympatry. Demoulin (1985) argued that S. fasciatum and S. lobatum were distinct species based on differing morphology and lack of fusion while growing on the same substrate, and our results support this claim with new genetic evidence.
As a consequence of these new data, the Stereum in the three “S. ostrea” clades should be recognized as distinct species, and those in synonymy with S. ostrea should revert to their original names. Our FIGS. 2 and 3 along with images for each specimen on iNaturalist (SUPPLEMENTARY TABLE 1) are provided to assist future naturalists and researchers in distinguishing among these taxa. Most importantly, these three species in midwestern and eastern North America can be differentiated by 1) texture of cap hairs, 2) presence and extent of banding, and 3) color staining of the hymenium when bruised or wetted (FIGS. 2 and 3).
Comparison of color staining between a. S. fasciatum and b. S. lobatum, which show slight darkening, and bright yellow staining, respectively. S. subtomentosum (not shown) exhibits the same bright yellow staining as S. lobatum. Note that both specimens were dry, and re-wetted to investigate color staining.
A more general conclusion of this research is that an integrative approach to Stereum taxonomy that includes DNA sequence data will be helpful in resolving species boundaries. In many cases, original descriptions are only a few sentences long or overly vague. Further, Stereum taxonomy has a long history of synonymization efforts such that descriptions of Stereum species from different authors are often conflicting. Historically, Stereum has been used as a genus name to describe many corticioid fungi, many of which have been transferred to other genera both in the Stereaceae and to other genetically distant families (Larsson and Larsson 2003). Many Stereum are accepted as synonyms of better-known Stereum species, but if our study is of any indication, some of those names may also need to be resurrected.
Our work here demonstrates a need to explore the rest of the Stereum genus, so that we may better understand their ecological roles, how they might be useful to us, and how they evolved. Molecular phylogenetic approaches will be necessary as a supplemental tool to delineate Stereum species, as morphology alone has proved inadequate. By using a multilocus approach and worldwide sampling strategy, we can achieve greater resolution within this genus and other members of the Stereaceae.
ACKNOWLEDGEMENTS
Financial support for this project was provided by awards to S.G.D. in the form of University of Iowa’s Maureen Medberry Snell CLAS Award and two Iowa Center for Research by Undergraduates (ICRU) fellowships. We thank Dr. Andrew Forbes for his mentorship and contributions to the preparation of this manuscript, and Dr. Rosanne Healy for providing the outgroup sequence and her valuable advice. We are grateful to Dean Abel for his encouragement and donation of valuable literature on Stereum.
Footnotes
Small edits and formatted for journal submission. Added Figure 3, Table 1 is now Supplementary Table 1, updated with GenBank Accession numbers and ISC herbarium voucher numbers.
LITERATURE CITED
