The specimens examined in this study were gathered in Northwest China, and their macroscopic features were recorded in fresh as the basis for the macromorphological descriptions, including host, ecological behaviors, geographical coordinates, location, altitude, collector, and date. Photographs of fresh basidiocarps and habitat information were taken in the field. All samples studied in this study were dehydrated and stored in the herbaria at Gansu Agricultural University (MHGAU, China). Micromorphological characters were observed from dried materials and examined with a light microscope (Nikon Eclipse E 80i microscope, Nikon, Tokyo, Japan) following the general techniques used in past research (Hu et al., 2022). In the descriptions, the abbreviations mean: IKI = Melzer’s reagent, IKI– = negative reaction in Melzer’s reagent, KOH = 5% potassium hydroxide, CB = Cotton Blue, CB+ = cyanophilous, CB− = acyanophilous, L = mean spore length, W = mean spore width, Q = variation in the L/W ratio between the specimens studied, and n = number of spores measured from the number of samples. The microscopic features were examined and described in 1% Congo red aqueous solution when necessary (Wei et al., 2024).

A CTAB plant genome rapid extraction kit-DN14 (Aidlab Biotechnologies Co., Ltd.) was utilized to extract DNA from dried specimens. With slight modifications, the manufacturer’s instructions for the PCR were carried out when employing the extracted DNA (Hu et al., 2022). The ITS gene region was amplified using the primer pairs ITS5/ITS4 and LR0R/LR7 for the nLSU gene region and 983F/1567R for the TEF-1α gene region (Gardes and Bruns, 1993; Cubeta et al., 1991). The PCR procedure for ITS and TEF-1α was as follows: initial denaturation at 95°C for 3 min, followed by 35 cycles at 94°C for 40 s, 56°C (ITS) or 55°C (TEF-1α) for 45 s and 72°C for 1 min, and a final extension of 72°C for 10 min (Li et al., 2022). The PCR procedure for nrLSU was as follows: initial denaturation at 94°C for 1 min, followed by 35 cycles at 94°C for 30 s, 50°C for 1 min, 72°C for 1.5 min, and a final extension of 72°C for 10 min (Li et al., 2022). The PCR products were purified and sequenced at the Xian Genomics Institute in China using the same pair of primers. Every newly produced sequence was uploaded to GenBank (Tables 1, 2).

A combined ITS + nLSU dataset was used to examine the phylogenetic relationship of Gymnopus, and for this, Mycetinis scorodonius (Fr.) A.W. Wilson & Desjardin and Mycetinis alliaceus (Jacq.) Earle ex A.W. Wilson & Desjardin were selected as the out-groups following the study by Li et al. (2022). The combined ITS + TEF-1α dataset was used to confirm the phylogenetic relationship of Mycena, and following the study by Na et al. (2022), M. meliigena (Berk. & Cooke) Sacc. and M. arcangeliana Bres were selected as the out-group. MAFFT 7 was used to align the datasets (Ryoo et al., 2016), and manual adjustments were made using BioEdit (Katoh and Standley, 2013). The partition homogeneity test (PHT) (Hall, 1999) of the two-gene dataset was tested by PAUP v. 4.0b10, respectively (Farris et al., 1994) under 1,000 homogeneity replicates. Alignments were spliced using Mesquite v.3.2. The best-fit model of nucleotide evolution for the datasets was selected, respectively, with Akaike’s information criterion (AIC) using MrModeltest 2.3 (Swofford, 2002; Guindon and Gascuel, 2003; Darriba et al., 2012). Phylogenetic analyses were conducted according to the previous study (Hu et al., 2022).

Using the heuristic search, maximum parsimony (MP) analysis was carried out using PAUP*version 4.0b10. Gaps were considered as missing data, and all characters were given the same weight. This study chose the heuristic search option to determine the trees, which included TBR branch switching and 1,000 random sequence additions. All parsimonious trees were preserved, zero-length branches were collapsed, and the maximum number of trees was set to 5,000. A bootstrap analysis with 1,000 replicates was used to test the clade’s robustness (Posada and Crandall, 1998). For each maximum parsimonious tree (MPT), descriptive tree statistics including tree length (TL), consistency index (CI), retention index (RI), rescaled consistency index (RC), and homoplasy index (HI) were computed. RAxmL v.7.2.8 was used to carry out maximum likelihood (ML) analysis with a GTR + G + I model (Felsenstein, 1985). The computer calculated all model parameters, but only the best maximum likelihood tree from each search was retained. The best-fit evolution model for Bayesian inference (BI) was used utilizing MrModeltest 2.3 (Stamatakis, 2006; Posada and Crandall, 1998). MrBayes 3.2.6 was used for BI, conducting two distinct runs, each starting from random trees with four independent chains running at the same time, doing 2 million repeats, and sampling a tree every 100 generations through the online platform CIPRES Science Gateway (www.phylo.org, accessed on 23 April 2024) (Nylander, 2004). The sequence alignment was deposited at TreeBase (Gymnopus submission ID: 31485; Mycena submission ID: 31513). A majority rule consensus tree of all the remaining trees was determined, and the first 25% of the sampled trees were eliminated as burn-in. Branches were considered significantly supported when the bootstrap supports for MP and ML were greater than or equal to 50%, and the Bayesian inference (BI) was greater than or equal to 0.95. FigTree v1.4.2 was used for the visualization of phylogenetic trees.

The sequences involved in the ITS + nLSU dataset of Gymnopus were obtained from 47 specimens representing 28 species. The dataset contains 47 ITS and 26 nLSU sequences; among them, four ITS and two nLSU sequences are newly generated. The dataset contained 1,217 characters, including gaps (418 characters for ITS, 799 characters for nLSU), with 930 characters constant, 55 variable and parsimonious uninformative, and 232 parsimonious informative. The maximum parsimony analysis of Gymnopus produced 4,953 equally parsimonious trees (TL = 578, CI = 0.633, RI = 0.822, RC = 0.521, HI = 0.367). The GTR + I + G models were selected as the most effective models for every region of the combined ITS + nLSU sequence dataset, and they were used in the Bayesian analysis. The topology obtained from MP and Bayesian analysis was similar to that of ML analysis. The Bayesian analysis yielded a topology that was concordant, with an average split frequency standard deviation of 0.005966. Only the ML tree is shown in Figure 2; bootstrap support values for MP and ML ≥50% and BI ≥0.95 are noted at the nodes, respectively.

The BLAST results show that the specimen FLF541 (Gymnopus gansuensis, holotype) is similar to Gymnopus sp. (taxon: 3341947), G. barbipes (taxon: 1460870), and G. impudicus (taxon: 206322), with a sequence similarity of 91.87%–95.73% from NCBI based on ITS with the top 10. Gymnopus subsepiiconicus (FLF393, Holotype) has a sequence similarity with G. erythropus (taxon: 230774), G. longisterigmaticus (taxon: 2935534), and G. longus (taxon: 2906493) ranging from 95.66% to 96.06% based on ITS with top 10.

The ITS + nLSU phylogenetic tree (Figure 2) confirmed the affinities of Gymnopus gansuensis and Gymnopus subsepiiconicus within Gymnopus, and both of them formed distinct, well-supported lineages (Figure 2).

The sequences involved in the ITS + TEF-1α dataset of Mycena were obtained from 66 specimens representing 37 taxa. The datasets contained 65 ITS and 51 TEF-1α sequences; among them, two ITS and two nLSU sequences are newly generated. The dataset contained 1,088 characters, including gaps (710 characters for ITS, 378 characters for TEF-1α), with 628 characters constant, 45 variable and parsimonious uninformative, and 415 parsimonious informative. The maximum parsimony analysis of Mycena produced 201 equally parsimonious trees (TL = 1,054, CI = 0.583, RI = 0.806, RC = 0.470, HI = 0.417). The GTR + I + G models were selected as the most effective models for every region of the combined ITS + TEF-1α sequence dataset, and it was used in the Bayesian analysis. The topology obtained from MP and Bayesian analysis was similar to that of ML analysis. The Bayesian analysis yielded a topology that was concordant, with an average split frequency standard deviation of 0.009123. Only the MP tree is shown in Figure 1; bootstrap support values for MP and ML ≥50% and BI ≥0.95 are noted at the nodes, respectively.

The BLAST results showed that the specimen FLF449 (Mycena glabera, Holotype) is similar to M. seminau (taxon: 1524328), M. rufobrunnea (taxon: 2942240), and M. sinar (taxon: 1524325), with a sequence similarity of 96.46%–97.83% from NCBI based on ITS with the top 10.

The ITS + TEF-1α phylogenetic tree (Figure 1) confirmed the affinities of Mycena glabera in Mycena sect. Calodontes, and two newly collected specimens formed distinct well-supported lineages (Figure 1).

Gymnopus gansuensis, B.Y. Wang, T. F. Ma & L.F. Fan sp. nov. (Figures 3a,b, 4).

MycoBank: 855041.

Etymology: refers to the place of origin in Gansu Province, China.

Diagnosis: Differs from other Gymnopus species by pileus honey yellow at the center, margin pinkish buff to buff, stipe pinkish buff to fuscous, and basidiospores elliptic to briolette.

Holotype: CHINA. Gansu Province: Tibetan Autonomous Prefecture of Ganan, Taohe National Nature Reserve, N34°40′70″, E103°53′26″, in coniferous forest, 29 July 2023, MHGAU FLF541.

Basidiomata small-to-medium size, solitary. Pileus usually applanate or slightly convex, 1.0–2.5 cm in diameter, striated, hygrophanus, honey yellow at the center, margin pinkish buff to buff, entire, tomentose near margin. Context thin, fleshy, white to cinnamon buff, odorless. Lamellae subfree to adnate, pinkish buff to honey yellow, crowded. Stipe center, cylindrical, smooth in the upper part, covered with white hairs up to 2/3 (from the base upwards), fistulose, fibrous, 3.2–5.0 cm × 0.2–0.4 cm, pinkish buff near the pileus, fuscous at the bottom.

Hyphal system composed of generative hyphae with clamp connections, IKI–, CB–; tissues turn to black in KOH. Generative hyphae in context light brown to brown, thin-walled, occasionally branched, regularly arranged, 3–8 μm in diameter. Generative hyphae in lamella light brown, thin-walled, occasionally branched, regularly arranged, 5–11 μm in diameter.

Basidiospores elliptic to briolette, smooth, hyaline, IKI–, thin-walled, (6–) 6.5–8 (9) × 3–4 (−4.5) μm, L = 7.5 μm, W = 3.48 μm, Q = 2.16–2.25, (n = 60/2). Basidia clavate, smooth, hyaline, thin-walled, 27–35 μm × 6–7 μm, two- or four-spored, sterigmata 1.5–3.5 μm long. Cheilocystidia abundant, clavate, with obtuse at the center, smooth, hyaline, thin-walled, 51–63 μm × 5–6 μm. Pileipellis a cutis, made up of irregularly branched or weakly coralloid hyphae, inflated, smooth, hyaline to light brown, thin-walled, 4–8 μm wide.

Additional specimen (paratype) examined: CHINA. Gansu Province: Tibetan Autonomous Prefecture of Ganan, Taohe National Nature Reserve, N34°40′70″, E103°53′26″, in coniferous forest, 29 July 2023, MHGAU WBY541 (duplicate).

Gymnopus subsepiiconicus, B.Y. Wang, T. F. Ma & L.F. Fan sp. nov. (Figures 3c,d).

MycoBank: 855412.

Etymology: refers to the species being similar to Gymnopus sepiiconicus.

Diagnosis: Differs from other Gymnopus species by pileus clay buff to grayish brown at the center, margin pinkish buff to fawn, stipe dark brown to fuscous, basidiospores elliptic.

Holotype: CHINA. Gansu Province: Lintan County Prefecture of Ganan, Yeliguan National Forest Park, N34°40′70″, E103°53′26″, in coniferous forest land, 27 July 2023, MHGAU FLF393.

Basidiomata in small-to-medium size, solitary to gregarious. Pileus usually applanate or slightly convex, 1.5 cm–2.2 cm in diameter, fluted, hygrophanus, clay buff to grayish brown at the center, margin pinkish buff to fawn, entire. Context thin, fleshy, light grayish brown, odorless. Lamellae subfree to adnate, honey yellow, crowded. Stipe central, cylindrical, 3.2–4.8 cm × 0.2–0.3 cm, grayish brown to dark brown near the pileus, fuscous at the bottom, smooth, fistulose, and fibrous.

Hyphal system composed of generative hyphae with clamp connections, IKI–, CB–; tissues turn to black in KOH. Generative hyphae in context light brown to brown, thin-walled, occasionally branched, regularly arranged, 3 μm –6 μm in diameter. Generative hyphae in lamella grayish brown to dark brown, thin-walled, occasionally branched, regularly arranged, 4 μm–8 μm in diameter.

Basidiospores elliptic, smooth, hyaline, inamyloid, thin-walled, 6–9 × 3–3.5 (−4) μm, L = 7.25 μm, W = 3.25 μm, Q = 2.05–2.4 (n = 60/2). Basidia clavate, smooth, hyaline, thin-walled, 19–40 μm × 5–7 μm, two- or four-spored sterigmata 2–5 μm long. Cheilocystidia abundant, clavate, smooth, hyaline, thin-walled, 52–111 μm× 5–6.5 μm. Pileipellis a cutis, made up of irregularly branched or weakly coralloid hyphae, inflated, smooth, hyaline to light brown, thin-walled, 5–15 μm wide.

Additional specimens (paratype) examined: CHINA. Gansu Province: Lintan County Prefecture of Ganan, Yeliguan National Forest Park, N34°56′58.924″, E103°35′47.026″, in coniferous forest, 12 September 2023, MHGAU FLF739, 13 September 2023, MHGAU FLF757.

Mycena glabera, B.Y. Wang, T. F. Ma & L.F. Fan sp. nov. (Figures 5, 6).

MycoBank: 855042.

Etymology: refers to the smooth and glabrous pileus.

Diagnosis: Mycena glabera is characterized by pileus convex or hemispherical, margin decurved, white to cream, slightly wavy, smooth, hygrophanus, surface glabrous, pale pink–yellow to fleshy pink at the center, basidiospores long ellipsoid to cylindrical, thin-walled, cheilocystidia fusiform, thin-walled.

Holotype: CHINA. Gansu Province: Tibetan Autonomous Prefecture of Ganan, Taohe National Nature Reserve, N34°40′70″, E103°53′26″, in coniferous forest, 28 July 2023, MHGAU FLF449.

Basidiomata thin and small, solitary, or scattered. Pileus convex or hemispherical to applanate, margin decurved, white to cream, slightly wavy, smooth, hygrophanus, with vertical stripes or grooves, surface glabrous, old lace to moccasin at the center, 10–20 mm in diameter. Context white, fleshy, very thin (< 1 mm). Lamellae subdecurrent, white to pale pink-yellow, distant. Stipe centrally attached, cylindrical, equal, smooth, fragile, hollow, 30–35 mm × 2–3 mm, white to cream near the pileus, pale wax yellow at the bottom.

Hyphal system composed of generative hyphae with clamp connections, IKI–, CB–; slightly inflated in KOH. Generative hyphae smooth, hyaline, thin-walled, regularly arranged, 3–9 μm in diameter.

Basidiospores long ellipsoid to cylindrical, smooth, hyaline, inamyloid, thin-walled, with obviously oil drop, (5.0–) 6.3–10.2 (−11.7) μm × 3.7–5.5 (−6.7) μm, L = 8.7 μm, W = 4.3 μm, Q = 2.0–2.15, (n = 60/2). Basidia clavate, smooth, hyaline, thin-walled, 29.0–45.0 μm × 6.3–12.0 μm, four-spored when mature, sterigmata up to 15.5 μm long. Cheilocystidia abundant, fusiform to subfusiform, with obtuse at the center, smooth, hyaline, thin-walled, 48.0–77.4 μm × 10.0–16.4 μm.

Additional specimen (paratype) examined: CHINA. Gansu Province: Tibetan Autonomous Prefecture of Ganan, Taohe National Nature Reserve, N34°40′70″, E103°53′26″, in coniferous forest, 28 July 2023, MHGAU WBY449 (duplicate) (Figure 7).