Evaluation and transcriptomic and metabolomic analysis of the ability of Auricularia heimuer to utilize crop straw

Sample preparation for metabolomics

An appropriate amount of the sample was mixed with pre-chilled methanol/acetonitrile/water solution (2:2:1, v/v). The mixture was vortexed, then subjected to cold ultrasonication for 30 min. It was then incubated at −20 °C for 10 min, followed by centrifugation at 14,000 g for 20 min at 4 °C. The supernatant was collected and dried under vacuum. For mass spectrometry analysis, the residue was resuspended in 100 μL of acetonitrile/water solution (1:1, v/v), vortexed, and centrifuged at 14,000 g for 15 min at 4 °C. The supernatant was collected for instrumental analysis (Doppler et al., 2016).

LC-MS/MS metabolomics experimental procedure

For the analysis of non-polar metabolites, an LC-MS/MS method was employed utilizing an UHPLC system (Vanquish; Thermo Fisher Scientific, Waltham, MA, USA) equipped with a Phenomenex Kinetex C18 column (2.1 mm × 50 mm, 2.6 μm) interfaced with an Orbitrap Exploris 120 mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA). The mobile phase consisted of two components: A (0.01% acetic acid in water) and B (isopropanol: acetonitrile, 1:1, v/v). The auto-sampler was maintained at 4 °C, with an injection volume of 2 μL. The Orbitrap Exploris 120 mass spectrometer was operated in information-dependent acquisition (IDA) mode, controlled by the Xcalibur acquisition software (Thermo Fisher Scientific, Waltham, MA, USA). In this mode, the software continuously assessed the full scan MS spectrum. The ESI source parameters were set as follows: sheath gas flow rate at 50 Arb, auxiliary gas flow rate at 15 Arb, capillary temperature at 320 °C, full MS resolution at 60,000, MS/MS resolution at 15,000, collision energy using stepped normalized collision energy (SNCE) at 20/30/40, and spray voltage at 3.8 kV (positive mode) or −3.4 kV (negative mode). The raw data were converted into the mzXML format using ProteoWizard and subsequently processed with a custom-developed program written in R, based on the XCMS package, for peak detection, extraction, alignment, and integration. The R package was applied in metabolite identification (Han et al., 2024). The identification of metabolites was based on the Human Metabolome Database (HMDB) (http://www.hmdb.ca), Metlin (http://metlin.scripps.edu), massbank (http://www.massbank.jp/), and mzclound (https://www.mzcloud.org) (Cai et al., 2015; Zhou et al., 2022).

Cellulase activity

In 1976, Mandels, Andreotti & Roche (1976) discovered cellulase. Cellulase is a multicomponent enzyme that consists of C1 enzymes (cut inside glucanase endo-1,4-β-D-glucanohydrolase, EC3.2.1.4), Cx enzymes (circumscribed glucanase), and β-glucosidase (β-D-glucosidase, EC3.2.1.21), and is usually represented by FPA. The function of C1 enzyme is to hydrolyze natural cellulose into amorphous cellulose, Cx enzyme further hydrolyzes amorphous cellulose into fiber oligosaccharides, and β-Glucosidase hydrolyzes fiber oligosaccharides into glucose. Through significant analysis of cellulase activity, the results showed that A. heimuer hyphae had different effects on cellulase activity of different crop substrates. FPA activity results are shown in Fig. 3. Figure 3A shows that rice stem (D1), corn stem (D3), and soybean stem (D5) had higher FPA enzyme activity, with D1 being the highest at 1.53 ± 0.04. The FPA enzyme activity of corn cob (D4) and rice husk (D2) was relatively low;, and the FPA enzyme activity of sawdust (D6) was the lowest at 0.06 ± 0.03. Figure 4B shows that D1 had the highest C1 enzyme activity, 6.63 ± 0.06. Figure 4C shows that D1, D3, and D6 had higher Cx enzyme activities: 1.17 ± 0.06, 1.07 ± 0.03, and 1.17 ± 0.07, respectively. Figure 4D shows that D1 and D3 had high β-G enzyme activities of 3.32 ± 0.12 and 3.17 ± 0.07, respectively. There was a positive correlation with cellulose content, mainly due to the higher cellulose content in rice husk and corn stalks. A. heimuer mycelium activity in relation to cellulase determines the decomposition ability of mycelium to substrates to a certain extent, so it is indicated that A. heimuer has the best utilization rate of D1 and D3.

Cultivation substrate plays a crucial role in the yield, mycelial growth rate, biological efficiency, and nutritional parameters of edible and medicinal mushrooms (An, Wu & Dai, 2019). The ability of heimuer to utilize corn cobs as a substrate and produce cellulose-degrading enzymes has been confirmed (Yu et al., 2022). Enzyme activity assays indicated that the filter paper enzyme activity of heimuer significantly increased when cultivated on a corn cob substrate. This is consistent with the high cellulase and filter paper enzyme activities of D1 and D3 in this study. Research has found that the activity of cellulase is closely related to the growth and yield of heimuer. Optimized extraction conditions can significantly enhance the activity of cellulase, thereby promoting the growth of heimuer. The potential to effectively degrade cellulose and hemicellulose in corn cobs supports the use of corn cobs as a sustainable substrate for heimuer cultivation (Geng et al., 2025). We speculate that rice straw and corn stalks are also excellent agricultural straw substrates for heimuer cultivation. Therefore, we chose this substrate for subsequent experiments.

Hemicellulase and liginase activities

Hemicellulose is a large class of carbohydrates second only to starch and cellulose. For example, the content of hemicellulose in straw accounts for 25% to 50% of its dry weight (Li et al., 2008). The degradation of hemicellulose by edible bacteria is mainly caused by β-1, 4-endoxylanase, β-xylosidase, α-arabinofuranosidase, acetylxylanesterase, and α-glucuronidase. Endo-1, 4-β-xylanase (EC3.2.1.8) is abbreviated as xylanase. Xylanase, a key enzyme in the cellulase series, is mainly extracellular and acts synergically with xylosidase intracellular enzyme. Currently, xylanase has been the most extensively studied. As a physical barrier, lignin is embedded in cellulose and hemicellulose to form a dense matrix, which hinders the biodegradation and utilization of straw (Mei et al., 2020). Currently, there are four enzyme activities that degrade lignin in plant cell walls (de Gonzalo et al., 2016; Falade et al., 2017): Lips (EC 1.11.1.14), MnPs (EC 1.11.1.13), and LACC/Lac (EC 1.10.3.2). These enzymes, as biocatalysts for lignin biodegradation, can increase the degradation capacity and speed. However, because lignin-decomposing enzymes are too large to penetrate undecayed plant cell walls, reactive oxygen species may be the factor that causes local lignin decay (Pollegioni, Tonin & Rosini, 2015). To sum up, it is necessary to investigate the correlation between the activities of hemicellulase and lignin enzyme and the growth of A. heimuer mycelia.

In this study, hemicellulase activity was highest in D1 and D5, reaching as high as 75.78 ± 1 and 76.92 ± 0.38 U/mL, respectively, with no significant difference between them. This was followed by D4 and D6, which were 31.96 ± 0.65 and 33.85 ± 0.76 U/mL, respectively, and then finally by D2 and D3, which had activity of 22.89 ± 0.65 U/mL and 23.27 ± 0.38 U/mL, respectively (Fig. 4A). D1, D3, and D6 showed significant differences in their ability to utilize hemicellulose. As shown in Figs. 4B, 4C, and 4D, agaric hyphae could produce Lac, MnP, and Lip simultaneously. Lip activity showed significant differences, with the highest being 729.66 ± 1a and the lowest being 693.14 ± 2.04c. Differential effects of fungi on single lignin-modifying enzyme (LME) production at different lignin growth substrates and nitrogen sources have been identified (Rusitashvili, Kobakhidze & Elisashvili, 2024). The activities of MnP and Lac in D6 were the highest, at 666.35 ± 2.03 and 63.89 ± 1.46, respectively, followed by D3, at 641.71 ± 3.22 and 56.71 ± 1.23, respectively. We speculate that the similarity in the utilization of LACC is related to the decomposition of dense lignin in corn stalk. Currently, only certain strains of WRF can simultaneously co-produce LACC, Lip, and MnP, making them highly valuable for research (Wang, Yao & Su, 2018).

Correlation analysis between different substrates, extracellular enzyme activity, and mycelial growth rate

The C content of substrates mainly comes from lignocellulose. The content of cellulose, hemicellulose, and lignin varies in different substrates, which affects the ability of edible fungi to decompose substrates. The above experiment shows that different substrates, including carbon and nitrogen content, wood fiber content, and extracellular enzyme activity, have significant effects on the growth of A. heimuer hyphae. Consequently, we performed correlation analysis (Fig. 5) to analyze the potential relationship between carbon and nitrogen content, wood fiber composition, key extracellular enzyme activity in substrates, and the growth of A. heimuer hyphae. The results showed that there was a significant positive correlation between carbon content and lignin content in substrates. Furthermore, we also found that cellulose, hemicellulose, and lignin did not show a significant positive correlation with the growth rate of W-ZD22 mycelium (0.05 ≥ p > 0), and the difference was not significant.

In addition, LACC also showed its multiple effects in this study. It not only has a significant positive correlation with lignin content, but also significantly affects the activity of exoglucanase, further highlighting its complex role and importance in lignocellulosic degradation.

Based on the above findings, we selected representative substrates with decomposition ability (D1, D3, D6) for transcriptome and metabolomics studies. These studies will further analyze the differences in transcription and metabolism of A. heimuer in the process of using crop stalk and other substrates, particularly in the expression and regulation of key enzymes such as endoglucanase, exoglucanase, MnP, and LACC.

DEG analysis

We used RNA-seq to detect DEGs on D1, D3, and D6 substrates during the growth of A. heimuer hyphae. 5,149, 2,740, and 2,933 DEGs were detected in the D1-vs-D3, D1-vs-D6, and D3-vs-D6 groups, respectively. For the comparison group of D1-vs-D3, 2,689 DEGs were upregulated and 2,460 DEGs were downregulated. For the comparison group of D1-vs-D6, 1,296 DEGs were upregulated and 1,444 DEGs were downregulated. Finally, in the comparison group of D3-vs-D6, 1,599 upregulated DEGs and 1,374 downregulated DEGs were obtained (Fig. 6A). As shown in Fig. 6B, 526 consensus DEGs were prevalent in the three comparison groups, which indicates that the expression of these genes are consistent in A. heimuer hyphae in response to different substrates. Across the three comparison groups, there were 507 (D1-vs-D6), 1,499 (D1-vs-D3), and 470 (D6-vs-D3) specific DEGs. The D6-vs-D3 comparison group had the least specific DEGs, which we will focus on later.

To understand the expression patterns of DEGs on the three substrates, bidirectional clustering analysis was performed on the union and samples of the differential genes of all comparison groups (Fig. 6C). The heatmap showed that the biological duplication of the samples was good. The three groups of DEGs were mainly divided into highly-expressed genes (red) and low-expressed genes (green). These 526 consensus DEGs showed the expression levels of the same gene in different samples and expression patterns of different genes in the same sample, with the expression patterns of D1 and D6 being most similar.

Functional annotation and enrichment analysis of DEGs

In order to study the biological processes involved in these DEGs in three comparative groups, GO annotation was performed on the DEGs present in each comparative group. GO classification was performed based on molecular function (MF), biological process (BP), and cellular component (CC). The GO annotations were mapped to GO terms, and the number of annotated genes on the second level classification are shown in Fig. 7A. These DEGs are significantly enriched in the metabolic processes of BP, the cell and organelle parts of CC, and binding and catalytic activity of MF.

KO annotation was completed by KOBAS annotation system and statistical results of the KEGG pathway were obtained (Fig. 7B). All 35 second level branches under five levels (metabolism, genetic information processing, environmental information processing, cellular processes, organic systems) have mappings. The dominant pathways are carbohydrate metabolism, signal transduction, transport, and catabolism. Lignocellulase genes mainly exist in these metabolic pathways, which are also the key pathways to pay attention to.

To further analyze DEG-related pathways, these transcripts were KEGG annotated and enriched to 111, 106, and 102 pathways in the three comparison groups, respectively. Based on the previous experiments, we focused on the analysis of the D6-vs-D3 group. All single-feature genes were functionally predicted and classified in KEGG database. They were divided into five main functional categories and 18 sub-categories (Fig. 7B). The largest category was the metabolic cluster, which accounted for about 69 items, and of which carbohydrate metabolism (14 items, 13.7%) was in the subcategory. Figure 7C shows the bubble diagram of 14 enriched carbohydrate metabolism pathways, and 15.66% DEGs were enriched in the starch and sucrose metabolism pathways (ko00500).

Screening and analysis of differential metabolites

In order to better understand the metabolic mechanisms of different substrate media during the growth process of A. heimuer hyphae, non-targeted metabolomics analysis was conducted on A. heimuer hyphae grown on different substrate formulation media. Under different carbon-induced conditions, the degradation ability of A. heimuer strains to lignin, cellulose, and hemicellulose varies. LC-MS analysis was performed on the samples treated with fermentation culture. As shown in Fig. 8A, PCA results showed good sample repeatability, indicating that metabolome data could be used for subsequent analysis. For the comparison of inter-group differences between different groups, the supervised orthogonal partial least squares discriminant analysis (OPLS-DA) model was used, and permutation tests were used to identify that supervised learning methods did not obtain classification by chance (Fig. 8B). Using this approach, it is possible to better distinguish between group differences and improve their effectiveness and parsing ability (Meng et al., 2022).

Non-targeted metabolomics analysis showed that 453 differential metabolites of D1-vs-D6 were up-regulated and 422 were down-regulated, 282 differential metabolites of D6-vs-D3 were up-regulated and 233 down-regulated, and 362 differential metabolites of D1-vs-D3 were up-regulated and 343 down-regulated (Fig. 9A). The differential metabolites were clustered (Fig. 9B). The top 20 metabolites included (Fig. 9C) phloroglucinaldehyde, (4-(β-D-Glucopyranosyloxy) phenylacetic acid, (R)-2-hydroxystearate, luminol, gentisic acid, 3,4-dihydro-6-hydroxy-alpha,2,5,7,8-pentamethyl-2H-1-benzopyran-2-pentanoic acid, 3alpha-hydroxyoreadone, fumaritine N-oxide, vicenin2, 1-(3,5-dihydroxyphenyl)-12-hydroxytridecan-2-one, 4-[2-[(1R,4aS,5R,6R,8aS)-6-hydroxy-5-(hydroxymethyl)-5,8a-dimethyl-2-methylidene-3,4,4a,6,7,8-hexahydro-1H-naphthalen-1-yl]acetyl]-2H-furan-5-one, 1’H-spiro[cyclopentane-1,2’-quinazolin]-4′(3′H)-one, pyridine, 2-hydrazinyl-5-nitro-3,4-dihydroxybenzoic acid, 11-deoxyprostaglandin E2, isoschaftoside, schaftoside, 2,3-dihydroxybenzoic acid, 4-oxododecanedioic acid, and armillyl_everninate.

KEGG enrichment of differential metabolites

According to mycelial growth characteristics and enzyme activity, the utilization of corn stalk and sawdust was more similar. Therefore, D3 and D6 were selected for enrichment of metabolic pathways. The top 20 enrichment pathways were ABC transporters, metabolic pathways, biosynthesis of amino acids, arginine and proline metabolism, D-amino acid metabolism, C5-branched dibasic acid metabolism, pyrimidine metabolism, glyoxylate and dicarboxylate metabolism, 2-oxocarboxylic acid metabolism, caffeine metabolism, arginine biosynthesis, valine, leucine and isoleucine biosynthesis, arachidonic acid metabolism, β-alanine metabolism, biosynthesis of cofactors, starch and sucrose metabolism, glutathione metabolism, carbon metabolism, tyrosine metabolism, and phenylalanine metabolism (Fig. 10).

ABC transporters play important roles in various cellular processes (Bonifer & Glaubitz, 2021). In naphthylacetic acid-promoted cordycepin studies, significant enrichment of the ABC transporter pathway was found. ABC transporters are known to transport multiple amino acids, such as L-glutamate, and are involved in amino acid metabolism that affects cordyceps synthesis (Wang et al., 2023). Compared with classical amino acids, the distribution, metabolism and function of natural non-classical amino acids are still relatively obscure. Natural non-classical amino acids are mainly found in plants as secondary metabolites and play a variety of physiological functions. It is necessary to study and clarify the metabolic pathways and key enzymes of non-classical amino acids (Zou et al., 2018). An extracellular effector depends on the production and release of extracellular D-amino acids that regulate various cellular processes, such as cell wall biogenesis, biofilm integrity, and spore germination. Non-classical D-amino acids are mainly produced by broad-spectrum racemases (Bsr). The promiscuous nature of Bsr allows it to produce high concentrations of D-amino acids in environments with different L-amino acid compositions. However, it has not been clear until recently whether these molecules exhibit different functions (Aliashkevich, Alvarez & Cava, 2018). Studies have shown that in Candida albicans, the TCA cycle not only occupies the central position of cell metabolism, but also regulates other biological processes such as CO2 sensing and mycelial development through integration with the Ras1-cAMP signaling pathway (Tao et al., 2017). Over the past few decades, carbon catabolic inhibition (CCR) has fascinated scientists and researchers worldwide. This important mechanism allows priority to be given to energy efficient and readily available carbon sources over those that are less readily available. This mechanism helps the microbes get the maximum amount of glucose to keep up with their metabolism. Microbes absorb glucose and highly favorable sugars before switching to less desirable carbon sources, such as organic acids and alcohols. In the CCR of filamentous fungi, CreA, as a transcription factor, is regulated to a certain extent by ubiquitination (Adnan et al., 2017).

Metabolic analysis showed that specific metabolites, including various amino acids, pyrimidines, acids, lipids, and carbohydrates, were produced when different substrates were decomposed. The synthesis and metabolism of amino acids have significant influences on the growth and metabolic product synthesis of fungi (Erdmann et al., 2024; Verwaal et al., 2007). Pyrimidine metabolism plays an important regulatory role in the growth and development of fungi (Sha et al., 2025), and acidic substances have significant effects on the flavor and nutritional components of A. heimuer. Carbohydrates are the main energy source for A. heimuer, providing energy for growth and metabolism (Yao et al., 2019). The enrichment analysis showed that these substances had strong biological activity to promote catabolism and biosynthesis, and were involved in amino acid and sugar metabolism. Therefore, these metabolic pathways may be an important metabolic pathway in the production of A. heimuer from corn stalk. Studying the transcription factors that correspond to these metabolites facilitates targeted breeding. In particular, we focus on starch and sucrose metabolism, carbon metabolism, and redox reactions.