Introduction

Plants live in association with various microorganisms, including fungal endophytes, both in natural and agricultural environments. Fungal endophytes colonize above- and below-ground plant tissues without causing disease symptoms to their hosts, at least during some period of their life cycle (Petrini 1991; Saikkonen et al. 1998; Vandenkoornhuyse et al. 2002; Mandyam and Jumpponen 2005; Schultz and Boyle 2005; Rodriguez et al. 2009). A form-group of root colonizing fungi accommodates dark septate endophytes (DSEs, Jumpponen and Trappe 1998), classified also as class 4 endophytes (Rodriguez et al. 2009), that form melanized septate hyphae. DSEs are common in root-associated fungal communities in several biomes and climatic regions, yet several lineages represent undescribed taxa, and the ecological functions of DSEs remain elusive (Mandyam and Jumpponen 2005; Porras-Alfaro and Bayman 2011; Sieber and Grünig 2013; Knapp et al. 2018, 2019).

The orders Helotiales (Leotiomycetes) and Pleosporales (Dothideomycetes) are the two orders accommodating most DSEs (Andrade-Linares and Franken 2013; Knapp et al. 2022, 2015; Grünig et al. 2011; Newsham 2011; Sieber and Grünig 2013; Jumpponen et al. 2017). However, other DSEs can also be found in the Hypocreales (Sordariomycetes) and Chaetothyriales (Eurotiomycetes) (Sieber and Grünig 2013; Jumpponen et al. 2017; Maciá-Vicente et al. 2018). Although narrow host specificity in DSE-plant interactions is not yet proven, different DSE communities are seen in trees and forest ecosystems and grasses and grasslands. Species belonging to the order Pleosporales are prevalent DSEs of the latter habitats, comprising a plethora of grass-root endophytes (Zhang et al. 2012; Jumpponen et al. 2017) that are present in natural (Rudgers et al. 2022) and agricultural ecosystems (Gdanetz and Trail 2017). The number of formally described DSE species is continuously increasing, and the vast majority of DSEs were isolated from natural ecosystems. For example, during the last decade, 14 new DSE species have been described from natural semiarid grassland ecosystems in Hungary (Knapp et al. 2015, 2022; Ashrafi et al. 2018; Crous et al. 2019, 2021). On the other hand, we should bear in mind that non-pathogenic fungi and endophytes in agricultural environments are understudied.

As more than half of the area of Hungary represents agronomic areas and more than 80% of those areas are croplands (Hungarian Central Statistical Office), it is important to gain more information about the non-pathogenic fungi of farmland habitats. We have studied a unique, long-term agricultural experimental area near Martonvásár, Hungary that is based on wheat (Triticum aestivum) and maize (Zea mays) monocultures and was similarly managed for more than 60 years (for detailed site description, see Mayer et al. 2019; Ujvári et al. 2020; Megyes et al. 2021). During the characterization of fungal root endophytes of different parcels in these experiments, we gained hundreds of different isolates. The initial DNA barcoding using ITS (internal transcribed spacer) of nrDNA sequences revealed that several isolates from wheat represent two novel pleosporalean lineages distinct from the known species in the genus Poaceascoma (Lentitheciaceae) and the known genera in the Didymosphaeriaceae. In this study, we aimed at (i) determining the phylogenetic position of these lineages and clarify the taxonomy by providing formal descriptions of novel taxa and (ii) conducting in vitro resynthesis experiments to gain insights into their interaction with host plants.

Materials and methods

Sampling and isolation of fungal strains

Root samples of wheat were collected in experimental fields located near Martonvásár, Hungary (N47˚ 16′ 36.23′′; E18˚ 47′ 39.65′′). Root samples were collected on 08.07.2019. Healthy wheat roots were surface sterilized and processed as described in Knapp et al. (2012). We collected 32 isolates representing two distinct lineages (11 and 21 isolates). A selection of ten isolates were chosen for further analyses (Table 1), based on macroscopical characteristics on different culture media and sporulation inducing experiments. Four isolates representing a lineage in the genus Poaceascoma (Lentitheciaceae) and six Didymosphaeriaceae were chosen for molecular phylogenetic analysis (Tables 2 and 3). Three isolates of the former six of the latter were chosen for the resynthesis experiments and further sporulation generation experiments on oatmeal agar. Holotypes consisted of lyophilized cultures and were deposited as metabolically inactive samples in the herbarium of the Hungarian Natural History Museum, Budapest (BP) under the accession numbers BP112746 and BP112747. Ex-type and other cultures investigated in this study were deposited in the culture collection of the Westerdijk Fungal Biodiversity Institute (CBS 151097 and CBS 151043); nomenclatural novelties and descriptions were deposited in MycoBank (www.MycoBank.org, Crous et al. 2004).

Table 1 An overview experiments, isolates and loci of Agrorhizomyces patris and Poaceascoma zborayi described in the present study. Isolates marked with a in subscript were included in the phylogenetical analyses. Extype cultures are bolded
Full size table
Table 2 GenBank accession numbers of taxa used in the phylogenetic analysis of the Lentitheciaceae including Poaceascoma with the here described P. zborayi. Extype cultures are bolded
Full size table
Table 3 GenBank accession numbers of taxa used in the phylogenetic analysis of the Didymosphaeriaceae including the here described Agrorhizomyces patris. Extype cultures bolded
Full size table

DNA extraction and amplification

Genomic DNA was extracted from fungal mycelia using the NucleoSpin Plant II kit (using PL1, Macherey–Nagel, Germany) following the manufacturer’s instructions. Four loci were amplified and sequenced: internal transcribed spacer (ITS), partial 18S small subunit (nc SSU rDNA), and partial 28S large subunit (nc LSU rDNA) of the nrDNA, and partial translation elongation factor 1-alpha gene (TEF1–α). The following primers were used for amplification and sequencing: for ITS, ITS1F/ITS4 (White et al. 1990; Gardes and Bruns 1993); for nc SSU rDNA, NS1/NS4 (White et al. 1990); for nc LSU rDNA, LR0R/LR5 (Rehner and Samuels 1994; Vilgalys and Hester 1990); for TEF1–α, EF1-983/EF1-2218R (Rehner and Buckley 2005). The sequences were compiled from electropherograms using the Pregap4 and Gap4 software packages (Staden et al. 2000) and Sequencher 5.4 (GeneCodes Corporation, Ann Arbor, Michigan, USA), and deposited in GenBank (PP264924–PP265020 and PP273237–PP273267; Tables 1, 2 and 3). The obtained sequences were compared with the accessions in the National Center for Biotechnology Information database (NCBI, http://www.ncbi.nlm.nih.gov/Blast.cgi) using the BLASTn search (http://blast.ncbi.nlm.nih.gov/Blast.cgi) (Altschul et al. 1990). We also compared the ITS sequences with two of our unpublished datasets: one is a pilot fungal NGS metabarcoding dataset from the maize monocultures of the sampling site (see details on bacterial communities in Megyes et al. 2021), and the other is the fungal metabarcoding data from a long-term ecological experiment form the semiarid sandy grasslands of Fülöpháza, Hungary (Vajna et al. 2021).

Phylogenetic analyses

We combined and aligned the sequences of different loci with those from representative taxa in GenBank using the online version of MAFFT 7 (Katoh and Standley 2013) and the E-INS-i method. The alignments were examined and edited using MEGA 7 (Kumar et al. 2016). Two multi-locus family-level datasets for Lentitheciaceae (90 taxa) (Table 2) and Didymosphaeriaceae (110 taxa) (Table 3) were used for molecular phylogenetic analyses. Sequences highly similar to data acquired from the comparison with the accessions NCBI database using the BLASTn search (Altschul et al. 1990) and strains of representative species of the families Lentitheciaceae according to Calabon et al. (2021), Liu et al. (2022), Rajeshkumar et al. (2023), Hyde et al. (2021), and Didymosphaeriaceae according to Samarakoon et al. (2020b), Yuan et al. (2020), Crous  et al. (2022), Verkley et al. (2014) and Valenzuela-Lopez et al. (2018) were involved into the phylogenetic analyses. For both phylogenetic analyses, Corynespora cassiicola (CBS 100822) and Corynespora smithii (CABI5649b) served as outgroups. Bayesian inference (BI) analyses were performed with MRBAYES 3.1.2 (Ronquist and Huelsenbeck 2003) with a GTR + G substitution model for all nucleotide partitions. Four Markov chains were run for 10,000,000 generations, sampling every 1,000 generations with a burn-in value set at 4,000 sampled trees. Maximum likelihood (ML) phylogenetic analysis was carried out with RAXMLGUI 1.3 (Silvestro and Michalak 2012; Stamatakis 2014). The GTR + G nucleotide substitution model was also used for nucleotide partitions with ML estimation of base frequencies. ML bootstrap (BS) analysis with 1,000 replicates was used to test the support of the branches. Another dataset was created with only ITS sequences for the genus Poaceascoma, including five sequences most similar to Poaceascoma zborayi based on a BLASTn search in NCBI database. The sequences were aligned and analysed as described above using Setoseptoria arundelensis (MFLUCC 17–0759) and Setoseptoria englandensis (MFLUCC 17–0778) as outgroups. Phylogenetic trees were visualized and edited in MEGA 7 (Kumar et al. 2016) and deposited in Figshare (doi: 10.6084/m9.figshare.25164665).

Fungal morphology and sporulation

Potato dextrose and malt extract agar (PDA, MEA, VWR International, Belgium) plates were inoculated with colony plugs (5 mm diameter) in Petri dishes (5 cm diameter). Growth rate and colony characteristics were recorded in cultures grown for three weeks at 22 °C. To induce sporulation, isolates were cultured on autoclaved pine needles and stinging nettle stems laid on water agar (WA) in Petri dishes (9 cm diameter). The isolates were also cultured on WA media supplemented with minced vegetables (carrot, turnip, celery, and kohlrabi) at a pH of 3.5 (pH meter: Consort C830). Cultures were also grown on oatmeal agar (OA) in Petri dishes (5 cm diameter). The oatmeal agar contained 500 ml water, 15 g thin-rolled oats and 7.5 g agar. The ingredients were mixed and autoclaved following Gooding and Lucas (1959). The isolates were growing on these media at 22 °C for three months. Fungal structures and pieces of colonies were positioned and frozen onto a Peltier-cell cooled stage in distilled water mixed with glycerol. Sections of 30–50 μm thickness were cut with a Reichert microtome (Reichert, Austria) with steel microtome knife, and the sections were collected and mounted in distilled water mixed with glycerol. Morphological characteristics of the fungal structures were examined using a light microscope with differential interference contrast (DIC) optics and a Nikon Eclipse 80i microscope equipped with a Spot 7.4 Slider camera (Diagnostic Instruments, Inc.). Measurements and photographs were made using structures mounted in either water or in polyvinyl-lactoglycerol (PVLG).

Resynthesis experiment

An artificial in vitro resynthesis pot experiment was conducted for testing Poaceascoma isolates: BDT06, BDT15 and BDT33 and Didymosphaeriaceae isolates: BAT15, BAT20, BAT41, BBT01, BBT16, BBT24. In the experiment, 1.5 dl pots were filled with a twice autoclaved mixture of original soil from Martonvásár, sand, and zeolite (1:1:1). Grains of wheat were surface sterilized, allowed to germinate and each seedling was inoculated with five 5-mm fungal plugs, that were placed into the soil near the roots of seedlings. The plants were grown for eight weeks at 22 °C under a 14 h light:10 h dark cycle. Five replicates were used for each fungal isolate. The control pots were “inoculated” with five 5-mm plugs of the sterile medium. Altogether 50 plants were assessed. At the end of the inoculation experiments, the plants were harvested, and the substratum was carefully removed from the surface of the root system. Shoots were separated from the roots, and both were dried at 50 °C till constant weight, and then the shoot and root dry biomass was measured. Before drying, an equal amount of root segments was collected from each plant for microscopy analysis. To visualize the hyphae within the roots we used two methods. A wheat germ agglutinin (WGA) conjugate, WGA-Alexafluor®488 (Thermo Fisher Scientific, Lithuania) was used that stains chitin and is widely used as a fluorescent dye for fungal cell walls (Németh et al. 2022). The other method applied aniline blue as a stain for in planta visualization of fungal endophytes (Andrade-Linares et al. 2011; Knapp et al. 2019). The slides were examined on the Nikon microscope and setup described above and with excitation and emission filters for visualization of Alexa Fluor 488 probe. To test the effect of inoculation, one-way analysis of variance (ANOVA) was applied with a Tukey’s test for post-hoc analysis to identify the differences in dry biomass among plants inoculated by different isolates. For statistical analyses the GraphPad Prism 9.5.0 software was used.

Results

Molecular phylogeny

The four-loci phylogenetic analysis of Lentitheciaceae placed strains BDT06, BDT15, BDT32 and BDT33 into a clade with six previously described Poaceascoma species with high support (ML-BS = 93, B-PP = 0.99) (Fig. 4). Based on the phylogenetic analysis, the position of Poaceascoma within Lentitheciaceae is ambiguous, however, it grouped together with Stagonospora and Setoseptoria species with a moderate support. Poaceascoma taiwanense represents a distinct branch, and strains of the type species of Poaceascoma, P. helicoides, clustered with an ambiguously characterized strain P. lochii BRIP 71546 (ML-BS = 100, B-PP = 1.00). Three species, P. filiforme, P. halophilum, and our isolates grouped together (ML-BS = 71, B-PP = 0.99), and the ex-type strain, P. halophilum (MFLUCC 15–0949), formed a well-supported clade with those from wheat roots (ML-BS = 81, B-PP = 0.99). The four isolates form a strongly supported clade (ML-BS = 95, B-PP = 0.99). Relatedness of isolates BAT15, BAT20, BAT41, BBT01, BBT16 and BBT24 with Didymosphaeriaceae is highly supported (ML-BS = 100, B-PP = 1) (Fig. 5). Three clades, consisting of genera Spegazzinia and Dictyoarthrinium, and a distinct lineage represented by our isolates from Hungary, formed a strongly supported (ML-BS = 86, B-PP = 1) basal group in Didymosphaeriaceae. Each of these three clades has full support (ML-BS = 100, B-PP = 1), including the one comprising six isolates with identical sequences from wheat.

The 32 isolates, of which ten isolates were studied in detail in this study, formed two distinct lineages – these clades are considered as novel species in the genus Poaceascoma (Lentitheciaceae) and a novel species belonging to a new monotypic genus in Didymosphaeriaceae, and those taxa are formally described here.

Taxonomy

Poaceascoma zborayi Imrefi, D.G. Knapp & Kovács, sp. nov. — MycoBank MB852022; Figs. 1 and 4.

Fig. 1
figure 1

Colony and microscopic images of Poaceascoma zborayi strain BDT15 (CBS 151097). a On potato dextrose agar (PDA) incubated at 22 °C in darkness b Globose structure formed on oatmeal agar (OA) after three weeks at room temperature c Conidia-like structures formed on oatmeal agar (OA) after three weeks at room temperature. d Conidia-like structures formed on oatmeal agar (OA) after three weeks at room temperature. Scale bars: b–c 200 μm; d100 μm

Full size image

Etymology: We name the species in honour of Géza Zboray (1941–2023), an outstanding teacher at the Institute of Biology, Eötvös Loránd University (Budapest, Hungary), who had considerable influence on generations of Hungarian biologists all around the country, including all here involved authors.

Typification: Hungary: Martonvásár, isolated from healthy roots of wheat collected from agricultural fields, N47˚ 16′ 36.23′′; E18˚ 47′ 39.65′′, July 2019, I. Imrefi, a dried biologically inert agar culture (holotype BP112746, deposited under the barcode HNHM-MYC 033463; ex-type culture BDT15 = CBS 151097). GenBank: ITS = PP264928; nc LSU rDNA = PP264982; nc SSU rDNA = PP265014; TEF1–α = PP273262.

Diagnosis: Based on the phylogenetic analyses Poaceascoma zborayi differs from the taxa phylogenetically analysed in Fig. 4 by unique fixed alleles in the TEF1-α and nc ITS rDNA loci, identified based on alignments of separate loci deposited at Figshare repository (doi: 10.6084/m9.figshare.25164665): TEF1-α positions: 629 (T), 695 (T), 941 (T); nc ITS rDNA positions: a unique nucleotide motif at 350–363 (CGTGCGTTGGACCT). Additionally to those, Poaceascoma zborayi differs from congeneric Poaceascoma species by unique fixed alleles in the TEF1-α, nc ITS rDNA and nc LSU rDNA identified based on alignments of separate loci deposited at Figshare repository (doi: 10.6084/m9.figshare.25164665): TEF1-α positions: 97 (C), 244 (T), 374 (T), 389 (T), 407 (A), 416 (T), 443 (T), 479 (G), 482 (C), 656 (T), 731 (T), 758 (C), 779 (C), 791 (C), 803 (T), 824 (T), 833 (T), 854 (C); nc ITS rDNA positions: 36 (A), 81 (C), 82 (DEL), 87 (T), 103 (G), 120 (A), a unique nucleotide motif at 331–369 (CTGGGTGTTGTCCCGCCTCGTGCGTTGGACCTCGCCCG), 439 (C), 468 (T), 489 (A), 495 (A), 511 (C); nc LSU rDNA positions: 158 (T), 375 (T).

Additional specimens examined: Hungary: Martonvásár, roots of wheat collected from agricultural fields N47˚ 16′ 36.23′′; E18˚ 47′ 39.65′′, July 2019, I. Imrefi (BDT06); ibid. (BDT32); ibid. (BDT33); ibid. (BDT05); ibid. (BDT07); ibid. (BDT14); ibid. (BDT46); ibid. (BDT47); ibid. (BDT49); ibid. (BDT50). (Table 1).

Description: Poaceascoma zborayi is a dark septate endophyte colonizing wheat roots, also in in vitro resynthesis experiments, with no negative effects on the host. The isolates of the species showed almost the same morphology and dark colour on all the media tested. Colonies moderately slow-growing at 22 °C, on PDA covered the 5 cm Petri dishes, on MEA reaching 41 mm diam in 21 d. Colonies on PDA brownish grey with dark grey marginal zone and flat, with sparse to dense aerial mycelium, radially striate as the centre is lighter brown than the younger part of the colony, with dark exudates in the media. On MEA, colonies dark brown with a pale grey marginal zone and flat, with sparse aerial mycelium and radially striate, without secreted exudate droplets (Fig. 1a). Conidia-like structures produced by BDT15 on OA medium (Fig. 1c, d): conidiophore-like structures formed by branched, segmented hyaline hyphae, often found in groups. Conidia-like structures emerging on these structures were observed, brownish-red, rough-walled, warted, having variable sizes (20–40 μm). Immature, globose structures also formed by BDT15 (CBS 151097) on OA (500–800 µm diam) (Fig. 1b). Isolates BDT15 and BDT33 also formed hyphal loops on OA. The ITS blast search (08/01/2024) found several GenBank entries with more than 99% similarity with the ITS of P. zborayi, for example: MK808094 (Rudgers et al. 2022), OM106729 (Beschoren da Costa et al. 2022), KJ188723 (Luo et al. 2014), DQ420978 (Waldrop et al. 2006), MT683270 (Fors et al. 2020). Based on the ITS search in two of our unpublished fungal metabarcoding datasets, a fungus with 99.5% similarity to P. zborayi was found in the soil samples of maize monoculture on Martonvásár, but P. zborayi was not found in the soil from natural sandy grassland near Fülöpháza.

Agrorhizomyces Imrefi, D.G. Knapp & Kovács, gen. nov. — MycoBank MB852052; Figs. 2 and 5.

Fig. 2
figure 2

Colonies with different morphology and microscopic images of Agrorhizomyces patris a–c Colonies incubated on potato dextrose agar at 22 °C in darkness. a Isolate BBT01 (CBS 151043) b Isolate BAT20 c: Isolate BAT15 d Sterile, globose structure formed by isolate BBT01 after three weeks at room temperature on oatmeal agar. Scale bar: 200 μm

Full size image

Etymology: Agro (from the Greek word agros, meaning field, a reference to the sampling site of the isolates, also to the Agricultural Institute of Centre for Agricultural Research) + rhizo (from the Greek word rhiza, meaning root, referring to the fact that the isolates were collected from surface sterilized roots).

Type species: Agrorhizomyces patris Imrefi, D.G. Knapp & Kovács.

Notes: The genus Agrorhizomyces contains one root endophytic species. Isolates can be collected from surface sterilized roots and can be cultured and maintained on general media.

Phylogenetically, the monotypic Agrorhizomyces groups within the Didymosphaeriaceae with full support.

Agrorhizomyces patris Imrefi, D.G. Knapp & Kovács, sp. nov. — MycoBank MB852054; Figs. 2 and 5.

Etymology: The species was named “patris” in honor and memory of Ildikó Imrefi’s father.

Typification: Hungary: Martonvásár, isolated from healthy roots of wheat collected from agricultural fields N47˚ 16′ 36.23′′; E18˚ 47′ 39.65′′, July 2019, I. Imrefi, a dried biologically inert agar culture (holotype BP112747, deposited under the barcode HNHM-MYC 033464; ex-type culture BBT01 = CBS 151043). GenBank: ITS = PP264945; nc LSU rDNA = PP264967; nc SSU rDNA = PP264999; TEF1–α = PP273247.

Diagnosis:

Based on the phylogenetic analyses Agrorhizomyces patris differs from the taxa phylogenetically analysed in Fig. 5 by unique fixed alleles in the TEF1-α, nc ITS rDNA and nc LSU rDNA loci identified based on the alignments of separate loci deposited at Figshare repository (doi: 10.6084/m9.figshare.25164665): TEF1-α position: 218 (T); nc ITS rDNA positions: a unique nucleotid motif at 226–232 (GTATACC), 246 (C), 482–484 (TAA), a unique nucleotid motif at 561–595 (AGAGTAGGCGGTTGCTCGAGGCTTT); nc LSU rDNA positions: 290 (C), 297 (G), 369 (G), 375–376 (AA), 426–427 (TC), 564 (C).

Additional specimens examined: Hungary: Martonvásár, roots of wheat collected from agricultural fields N47˚ 16′ 36.23′′; E18˚ 47′ 39.65′′, July 2019, I. Imrefi (BAT15); ibid. (BAT20); ibid. (BAT41); ibid. (BBT16); ibid. (BBT24); ibid. (BAT14); ibid. (BAT17); ibid. (BAT19); ibid. (BAT21); ibid. (BAT25); ibid. (BAT39); ibid. (BAT40); ibid. (BBT01); ibid. (BBT02); ibid. (BBT04); ibid. (BBT13); ibid. (BBT15); ibid. (BBT24); ibid. (BBT41); ibid. (BBT42); ibid. (BBT43); ibid. (BBT55).

Description: Agrorhizomyces patris is a dark septate endophyte colonizing wheat roots, also in in vitro resynthesis experiments, with no or slightly negative effects to the host. The isolates show variant colony morphologies and colours and growing characteristics on different media (Fig. 2a–c).Colonies moderately fast-growing on PDA, growing approximately 0.7 cm a week, on other examined culture media (MEA) at 22 °C covered 5 cm Petri dishes in 21 d. Colonies on PDA pale brown grey with creamy marginal zone, colony surface slightly elevated at the centre, with dense aerial mycelium, radially striate, with sparse secreted exudate droplets on the surface of the mycelia. On MEA, colonies dark brown-grey with a greyish inner zone, colony surface slightly elevated at the centre, with dense aerial mycelium, radially striate, with sparse secreted exudate droplets on the surface of the mycelia (Fig. 2a–c). Sexual morph unknown. Globose structures formed by BBT01 (CBS 151043) on OA medium (Fig. 2d), consisting of wall, approximately 200 μm thick surrounding a cavity. These structures are strongly pigmented, resembling immature sporocarps with variable sizes (1.5–2 mm diam). Sporulation was not observed. The ITS BLASTn search (08/01/2024) in GenBank resulted only three unpublished environmental sequences with ~ 93% similarity with the ITS sequence of the A. patris isolate BBT01. These three almost identical sequences were: MF569188, KX192630, ON696029. Based on the ITS search in two of our unpublished fungal metabarcoding datasets a fungus with 100% similarity to A. patris was found in the soils of maize monoculture at Martonvásár, but A. patris was not found in the soil from a natural grassland near Fülöpháza.

Resynthesis experiment

In the resynthesis experiment, all nine isolates were able to colonize the wheat roots, forming characteristic extra- and intraradical hyphal structures and microsclerotia (Online Resource 1 b, c). Intraradical hyphae of both species were septate (Online Resource 1 a, b, d). Intraradical hyphae of Agrorhizomyces patris were predominantly hyaline and could be stained with the WGA-Alexafluor®488 fluorescent dye (Online Resource 1c). Those of Poaceascoma zboray were melanized and could not be visualized by WGA-Alexafluor®488 fluorescent dye (Online Resource 1a, b). The presence of the isolates effected the biomass of wheat in the pot culture experiment based on one-way ANOVA (p < 0.01). However, based on the results of Tukey’s test, only one of the isolates effected the biomass significantly (p < 0.05) compared to the control plant: BBT16 caused a decrease of shoot and root dry weight (Fig. 3).

Fig. 3
figure 3

Box-plot diagram of a: shoot and b: root dry weight of wheat inoculated with three Poaceascoma zborayi and six Agrorhizomyces patris isolates. The dots represent the outliers. Dry weight was measured after 8-wk long experiment

Full size image

Discussion

Here we introduced and formally described two novel pleosporalean DSE species, Poaceascoma zborayi and Agrorhizomyces patris isolated from healthy wheat roots from an agricultural field. There are more than 40 well-identified and investigated pleosporalean species considered as DSEs in genera Acrocalymma, Alternaria, Alfoldia, Aquilomyces, Curvularia, Darksidea, Delitchia, Drechslera, Flavomyces, Fuscosphaeria, Kiskunsagia, Laburnicola, Murispora, Periconia, Polydomus, Posidoniomyces, Setophoma and numerous further undescribed lineages belonging to families mainly of suborders Pleosporineae and Massarineae (Knapp et al. 2012, 2015, 2018, 2022; Zhang et al. 2012; Andrade-Linares and Franken 2013; Sieber and Grünig 2013; Knapp and Kovács 2016; Jumpponen et al. 2017; Vohník et al. 2019; Crous et al. 2019; Pereira et al. 2019; Crous et al. 2021; Pintye and Knapp 2021; Romero-Jiménez et al. 2022; Ashrafi et al. 2023). The majority of the known pleosporalean DSE species originated from natural ecosystems. However, one of the best-known and characterized DSE species, Periconia macrospinosa (Periconiaceae, Pleosporales) (Mandyam et al. 2010; Knapp et al. 2018) was originally isolated and described from roots of Sorghum vulgare (Lefebvre et al. 1949). Later it was regularly isolated from winter wheat roots (Hall 1986). The DSE species Murispora kazachstanica (Crous et al. 2021) and Laburnicola radiciphila (Knapp et al. 2022), were isolated from gramineous crops.

The pleosporalean family Lentitheciaceae was introduced by Zhang et al. (2009) based on multi-gene phylogeny. Lentitheciaceae currently accommodates 18 genera (see Calabon et al. 2021; Liu et al. 2022; Yang et al. 2022). Most of these genera comprise species characterized by lenticular to globose ascomata. Two genera (Phragmocamarosporium and Towyspora) accommodate only anamorphic species. Both are characterized by coelomycetous asexual morphs (Liu et al. 2022) (Figs. 4 and 5).

Fig. 4
figure 4

a Maximum Likelihood tree of concatenated ITS, nc LSU rDNA, nc SSU rDNA and TEF1-α sequences of representative species and genera of the Lentitheciaceae tree including Poaceascoma species. ML bootstrap support values (≥ 70) are shown before slashes or above branches, Bayesian posterior probabilities (≥ 0.90) are shown after slashes or below branches. Highlighted margin sections indicate names of taxa, the here described Poaceascoma zborayi is in bold. Corynespora cassiicola (CBS 100822) and Corynespora smithii (CABI5649b) served as outgroups. The scale bar indicates expected changes per site per branch. b Maximum Likelihood (RAxML) tree of ITS sequences of Poaceascoma species and P. zborayi isolates with bootstrap values and scale bar as described above. Setoseptoria arundelensis (MFLUCC 17–0759) and Setoseptoria englandensis (MFLUCC 17–0778) served as outgroups

Full size image
Fig. 5
figure 5

Maximum Likelihood tree of concatenated ITS, nc LSU rDNA, nc SSU rDNA and TEF1-α sequences of representative species and genera of the Didymosphaeriaceae tree including Agrorhizomyces patris. ML bootstrap support values (≥ 70) are shown before slashes or above branches, Bayesian posterior probabilities (≥ 0.90) are shown after slashes or below branches. Highlighted margin sections indicate the names of taxa, the here described A. patris is in bold. Corynespora cassiicola (CBS 100822) and Corynespora smithii (CABI5649b) served as outgroups. The scale bar indicates expected changes per site per branch

Full size image

The initially monoptypic genus Poaceascoma was introduced with the species P. helicoides occurring on dead stems and roots of Digitaria sanguinalis (Poaceae) collected in a natural terrestrial habitat (Phookamsak et al. 2015). Later, five additional species were described. All six species were described based on their sexual morphs. Typically, ascomata are known from natural substrates (Poaceascoma helicoides, P. aquaticum, P. halophila, and P. taiwanense), while those of P. filiforme were described from oatmeal agar (Crous et al. 2020). These ascomata differed in several characteristics but were semi-immersed to erumpent with filiform, generally septate and spirally twisted ascospores. In phylogenetic analyses of different combinations of DNA loci, generally ITS, nc LSU rDNA, nc SSU rDNA, TEF1-α and RPB2, Poaceascoma species grouped together in well supported clades (Phookamsak et al. 2015; Luo et al. 2016; Hyde et al. 2017, 2018; Crous et al. 2020).

Further five Poaceascoma species (P. aquaticum, P. halophilum, P. taiwanense, P. filiforme, P. lochii) were described and reported as saprobic fungi from different gramineous plant materials (Luo et al. 2016; Hyde et al. 2017, 2018; Crous et al. 2020; Boonmee et al. 2021; Tan et al. 2021).

Here we introduced the seventh species in the genus Poaceascoma, P. zborayi, also collected from a gramineous plant species (Triticum aestivum) supporting the association of the genus to Poaceae plants. Asexual morph of none of the five already described Poaceascoma species was determined. Here we could observe the production of two different structures: P. zborayi (isolate BTD15) formed globose immature sporocarps that we could neither identify as ascomata nor pycnidia (Fig. 1b). They were larger (500–800 µm diam) than most of the ascomata of known Poaceascoma species (Hyde et al. 2018). Brownish-red, rough-walled, warted, segmented conidium-like structures were repeatedly observed, emerging from conidiophore structures with variable sizes (20–40 μm) (Fig. 1c, d). Such asexual propagules were neither described for other Poaceascoma species nor in other members of the family before.

Although the majority of Poaceascoma species and materials were collected from dead plant shoot tissues, Phookamsak et al. (2015) noted that ascomata can also be found on roots of Digitaria sanguinalis and observed P. helicoides as saprobic on grass culms and roots. Since P. zborayi is characterized as a root endophyte, not causing symptoms and with no negative effect in in vitro experiments, we might hypothesize that Poaceascoma species could have an endophytic lifestyle beside being collected as saprobes on plant debris. The ITS blast search (08/01/2024) showed several GenBank entries with more than 99% similarity with the ITS of P. zborayi illustrating its wide distribution: Rudgers et al. (2022) collected two isolates (e.g. MK808094) from healthy roots of the foundation grass Schizachyrium scoparium in North American plains. Numerous sequences (e.g., OM106729) were originated from the USA from roots of switchgrass (Panicum virgatum) in long-term bioenergy research sites in Michigan and Wisconsin (Beschoren da Costa et al. 2022). Luo et al. (2014) collected an isolate (KJ188723) from the roots of a dominant grass species in temperate pine barrens in New Jersey, USA. Waldrop et al. (2006) gained 100% similar uncultured clone sequences (e.g. DQ420978) from grass-dominated experimental plots in Minnesota, USA. The blast of ITS2 resulted additional 100% similar hits: sequence (ON696510) originating from the soil of an experimental site at Richmond, NSW, Australia (Hassan et al. 2022) and sequence (MF568887) originating from soil samples Bloomington-Normal, IL USA (Beck 2017). Based on similar sequences, other Poaceascoma isolates were found in various geographic areas also associated with agricultural plants. For instance, Raza et al. (2019) collected P. helicoides isolates from the gramineous crop sugarcane (Saccharum officinarum) in China. Several Poaceascoma isolates (e.g., A122, MT683270) were collected also from healthy roots of sugarcane in Brazil (Fors et al. 2020). These isolations of different Poaceascoma species from healthy roots of agricultural plants further strengthen the hypothesis that members of this genus could have an endophytic lifestyle.

The non-pathogenic association of the genus to the Poaceae was hypothesized by Wu et al. (2022). During their experiments on a chemical fertilizer reduction system using substitution by organic material inputs, a member of the genus Poaceascoma was a key player in significant changes of the potential fungal functions in co‑occurrence network patterns of bacterium‑fungus‑nematode communities. Poaceascoma zborayi colonized the host plant in our resynthesis experiments, causing no symptoms and without negative effect on the biomass of shoot and root, either. Therefore, we consider this species as a root endophyte and not a pathogen. Still, information on the functional role of Poaceascoma species remains limited.

The Didymosphaeriaceae represents a well-supported clade within the Massarineae (Tanaka et al. 2015; Yuan et al. 2020). Ariyawansa et al. (2014) synonymized Montagnulaceae under Didymosphaeriaceae, accommodating 39 genera including our new genus Agrorhizomycota (Wijayawardene et al. 2014, 2022; Liu et al. 2022 Tanaka et al. 2015; Wanasinghe et al. 2016; Yuan et al. 2020, Ren et al. 2022). More than two third of the genera in the family were introduced based on their sexual morphs. Their ascomata are generally globose to sub-globose, have a central ostiole, and peridia with several layers of lightly pigmented to dark brown or black cells of angular texture. The asci are 2–4 or 8-spored, comprising 1–2-seriate, overlapping, ellipsoid or oblong, 1–3-septate or muriform ascospores (Ren et al. 2022). Other genera were described by asexual morphs, such as Alloconiothyrium, Dictyoarthrinium, Paraconiothyrium, and Spegazzinia were introduced based on only their asexual morphs characters (Ren et al. 2022).

In the present study, the multi-locus analyses showed that Agrorhizomyces grouped with Spegazzinia and Dictyoarthrinium, forming a basal clade in Didymosphaeriaceae. Spegazzinia was established in 1880 by Saccardo based on S. ornate. Micromorphology supports its classification, in the Apiosporaceae (Sordariomycetes) (Samarakoon et al. 2020a). Based on a multi-locus phylogeny using sequences of S. deightonii and S. tessarthra, Tanaka et al. (2015) placed Spegazzinia in Didymosphaeriaceae (Dothideomycetes). The further newly described Spegazzinia species supported the placement of the genus as a basal clade in Didymosphaeriaceae (Thambugala et al. 2017; Samarakoon et al. 2020a). However, only the asexual morph of Spegazzinia species is determined, their typical conidia originate terminally at the apex of basauxic conidiophores (Mena-Portales et al. 2017), separating them from the further species in the Didymosphaeriaceae with only coelomycetous asexual morphs (Thambugala et al. 2017). The other genus with which Agrorhizomyces was in sister position, Dictyoarthrinium, was introduced in 1952 by Hughes based on D. quadratum (Samarakoon et al. 2020b). Until Vu et al. (2019) sequenced D. sacchari (CBS 529.73), all Dictyoarthrinium species were introduced based on their morphological data. Although Hyde et al. (2020a, b) still accommodated the genus in the Apiosporaceae, based on their basauxic conidiogenous cells that are like those formed by Spegazzinia species, Samarakoon et al. (2020b) placed Dictyoarthrinium in the Didymosphaeriaceae, as well. Agrorhizomyces grouped with two asexual hyphomycete genera in the family that consist of mostly ascomata-forming species and anamorphic taxa with coelomycetous asexual morphs. We found coelomycetous structures, that were globose and sterile without conidia and conidiophores and resembled those formed by Spegazzinia and Dictyoarthrinium. Although each morph was sterile or immature, comprising no spores and other characteristic features, the presence of these globose structures, strengthened the affiliation of A. patris to the Didymosphaeriaceae, although they were never observed in the two neighboring genera.

The ITS BLASTn search (08/01/2024) in GenBank resulted only three unpublished, almost identical (> 99% similarity) environmental uncultured sequences with ~ 93% similarity with our ITS sequence from A. patris. One (MF569188) was obtained from soil from the above-mentioned location (Bloomington-Normal, IL USA; Beck 2017), while the other (KX192630) was derived from soil from Illinois, Champaign County, USA. The third (ON696029) originated from the above listed Australian site (Hassan et al. 2022). Albeit the sampling numbers are small, it seems, the novel DSE taxa (and their close relative) co-occurred at the same sites at their original location and also at North American and Australian locations (Beck 2017; Hassan et al. 2022).

Spegazzinia and Dictyoarthrinium species, are found on a wide range of dead plant material in various habitats, the former also on various host species. Although we have limited information on their lifestyle and function, we are not aware that they are plant pathogenic. Some Spegazzinia species, such as S. tessarthra, have been identified as saprobes and endophytes (Samarakoon et al. 2020a). Most of the Dictyoarthrinium species are considered saprobes showing affinity to monocotyledonous plants. The genus is widely distributed across the tropics, mainly in terrestrial environments (Samarakoon et al. 2020b). We are not aware of that A. patris was isolated from soil, living plant tissues or debris. We found no similar ITS sequences of A. patris in the GenBank and neither in our metabarcode database of natural grassland ecosystem. The dominant presence of A. patris in the roots of wheat in our sampling site was strengthened by the ITS2-based metabarcoding study of the soil fungal community in the same sites and parcels. Each three genera are plant-associated, and endophytic lifestyle cannot be ruled out in the case of Spegazzinia and Dictyoarthrinium species either. Although one of the six A. patris isolates had a slight but significant negative effect on the host biomass, the here reported isolates might be beneficial also in ways other than plant growth promotion (Ruotsalainen et al. 2022).

Our results illustrate also that environments where the taxon diversity of nonpathogenic fungi are less studied, like the soils of agronomic croplands, might be an unexplored reservoir of hidden fungal diversity. Revealing this diversity could help us understanding the spread and functional role of root endophytic fungi and might lead to beneficial agricultural applications.