Abstract Auricularia cornea is a widely cultivated edible fungus with substantial nutritive value. This study aimed to enrich the multifunctional bionutrient element selenium in A. cornea to improve its quality and explore the accumulation of selenium in the fungus using high-throughput RNA-Seq technology. In general, the treatment group with a 100 µg/g supply of selenium outperformed the other treatment groups in terms of high yield, rich crude polysaccharides and a high total selenium concentration. Additional evidences demonstrated the budding and mature phases were two typical growth stages of A. cornea and were important for the accumulation of selenium. Therefore, the budding and mature phase tissues of A. cornea in the treatment group with a 100 µg/g supply of selenium were used for transcriptome analysis and compared to those of a control group that lacked additional selenium. A total of 2.56 × 105 unigenes from A. cornea transcriptome were assembled and annotated to five frequently used databases including NR, GO, KEGG, eggNOG and SwissProt. GO and KEGG pathway analysis revealed that genes involved in metabolic process and translation were up-expressed at the budding stage in response to selenium supplementation, including amino acid metabolism, lipid metabolism, ribosome. In addition, the differential gene expression patterns of A. cornea suggested that the up-expressed genes were more likely to be detected at the budding stage than at the mature stage. These results provide insights into the transcriptional response of A. cornea to selenium accumulation. Introduction Auricularia cornea (synonym: A. polytricha) belongs to the Basidiomycota and is an edible fungus with high nutritive value because it is rich in amino acids, polysaccharides, vitamins and mineral substances in its fruiting bodies^[44]1–[45]3. Due to its significant medicinal value and high economic returns, the cultivation area of A. cornea is constantly increasing^[46]4. The yield of A. cornea is stable with sophisticated cultivation skills. Currently, substantial efforts are underway to improve its quality and growth efficiency to meet the increased market demands. As a multifunctional bionutrient element, selenium was identified as one of the necessary trace elements for the human body by World Health Organization (WHO) Committee in 1973^[47]5 and was listed as an essential element for daily diets by the Chinese Nutrition Society in 1988^[48]6. In addition, selenium is an important component of glutathione peroxidases and selenoproteins, which are important to maintain human health because of their antioxidative abilities, antitumour effects, ability to strengthen immunity and aid in the control of various diseases, including cancer and heart disease^[49]7–[50]9. Since selenium is found in low concentrations in the soil, plants and edible mushrooms, such as Flammulina velutipes, Pleurotus ostreatus and Ganoderma lucidum, organisms that can accumulate selenium are an ideal selenium-rich food^[51]10–[52]14. Since mushrooms can biotransform selenium from inorganic to organic forms, selenium enrichment studies in mushrooms are usually performed by adding inorganic selenium to the culture medium. For instance, the addition of sodium selenite to the substrate increased the uptake and integration of selenium into proteins, polysaccharides and other components in G. lucidum^[53]15. The quality and nutritive value of edible mushrooms are generally improved by adding essential elements, such as selenium and calcium, to the substrate via irrigation water^[54]16. The addition of selenium to the substrate has a substantial impact on the growth and development of the fruiting bodies of A. cornea in direct and indirect ways, such as changing its crude polysaccharide and trace element concentrations and affecting the enzyme activities and microbial diversity in the substrate^[55]17–[56]20. As a transformation carrier of mineral elements^[57]21, A. cornea can be cultivated in substrate supplied with inorganic selenium to harvest the fruiting bodies that may have a nutritive value beyond their original state. However, A. cornea can withstand a high range of selenium concentrations in cultivation, so it is critical to identify the optimal concentration of selenium^[58]22,[59]23. Currently, next-generation sequencing (NGS) is widely utilized to investigate complex genomic questions with unprecedented handling capacity, scalability and speed^[60]24. Transcriptome sequencing explores all the mRNAs, gene composition and functions in a given sample. Comparative transcriptome analysis provides insights into the differential expression patterns of genes and related biological pathways in cells, tissues or individuals during different physiological conditions^[61]25,[62]26. For example, Yu et al.^[63]27 studied the changes in gene expression and biological pathways from the mycelia to the initial primordial stages of G. lucidum using comparative transcriptome analysis. In addition, RT-PCR technology has been widely used in the quantitative detection of genes and viruses due to its high speed, sensitivity and accuracy^[64]28–[65]30. For instance, real-time quantitative PCR was used to study the stability of housekeeping genes in the symbiotic fungus Tuber melanosporum during its development and discovered various genes that were stably expressed at different stages^[66]31. Although selenium enrichment in edible fungi has been widely reported, few studies have explored the mechanisms of selenium accumulation in A. cornea using NGS technology. In this study, the agronomic traits and nutrient concentrations in A. cornea treated with different amounts of selenium were investigated. The transcriptome of A. cornea at two typical stages (budding and mature) was explored to reveal the stage-specific mechanisms of selenium enrichment using NGS technology. Four differentially expressed genes that may affect the quality and agronomic traits of A. cornea were selected for RT-PCR validation, with the goal of providing a theoretical basis for future studies on selenium enrichment in edible fungi. Results Effect of selenium on the growth of A. cornea Selenium addition affected the agronomic traits and nutrient contents of A. cornea (Table [67]1 and Fig. [68]1). The mean cluster weight increased first, then decreased following the increased concentration of selenium supply. It peaked (211.29 g) with the selenium concentration of 100 µg/g in the substrate, and was significantly higher than that of the control group. The mean size of a single random auricular patch was the biggest when supplied with 20 µg/g and 50 µg/g of selenium in the substrate. What was more, the mean fresh yield exhibited a first increase and afterward decrease trend, and it peaked with 100 µg/g of selenium augmented in the substrate, 47.05% higher than that of control group (473.99 g). However, the mycelial growth rate and thickness of A. cornea were significantly lower in the selenium-treated groups than those of control group. The crude polysaccharide content in A. cornea reached the highest (23.69%) with 150 µg/g of selenium supplied in the substrate. The total selenium concentration in the fruiting body increased with the increasing concentration of selenium, and it peaked (50.88 µg/g) with 200 µg/g of selenium added in the substrate, equivalent to 106 folds of control group. Table 1. A. cornea’s agronomic traits and nutrient content of fruiting bodies in different treatments NO. MGR (mm/d) Size (cm^2) Thickness (mm) SW(g) CW (g) AC Yield (g) CP (%) TSC (µg/g) ACK 4.89 ± 0.52b 87.86 ± 24.47a 1.64 ± 0.19b 42.60 ± 11.60a 172.40 ± 18.06a 7.70 ± 1.64a 473.99 ± 20.71a 7.31 ± 0.34a 0.48 ± 0.02a A20 4.19 ± 0.50a 127.24 ± 33.04bc 1.34 ± 0.23a 43.83 ± 15.02a 167.39 ± 38.45a 6.50 ± 1.27a 563.71 ± 23.19ab 12.26 ± 1.45b 0.59 ± 0.02a A50 4.53 ± 0.59a 136.25 ± 47.66c 1.32 ± 0.15a 52.42 ± 21.65a 194.52 ± 50.66ab 6.50 ± 1.78a 599.61 ± 9.47bc 21.44 ± 0.01c 8.5 ± 0.23b A100 4.23 ± 0.37a 121.32 ± 41.29abc 1.35 ± 0.13a 50.65 ± 15.17a 211.29 ± 50.82b 7.20 ± 0.92a 697.00 ± 49.64c 22.45 ± 0.22de 32.24 ± 1.22c A150 4.46 ± 0.67a 113.59 ± 43.98abc 1.26 ± 0.13a 46.41 ± 13.23a 198.64 ± 28.33ab 7.80 ± 2.04a 605.88 ± 42.88bc 23.69 ± 0.24e 43.24 ± 1.01d A200 4.32 ± 0.63a 89.82 ± 34.50ab 1.26 ± 0.16a 38.81 ± 11.50a 163.42 ± 27.31a 6.90 ± 1.10a 633.41 ± 28.82bc 18.66 ± 0.05d 50.88 ± 0.55e [69]Open in a new tab Abbreviations: NO. cultivating formula; MGR mycelial growth rate; Size the size of a single random auricular patch; Thickness the thickness of a single random auricular patch; SW the weight of a single random auricular patch; CW the weight of a random cluster; AC the amount of a random cluster; Yield the fresh yield per bag; CP the content of the crude polysaccharide in the mature fruiting bodies; TSC the total selenium concentration in the mature fruiting bodies; ACK the control group without the addition of selenium; A20 the treatment group with 20 µg/g sodium selenite addition; A50 the treatment group with 50 µg/g sodium selenite addition; A100 the treatment group with 100 µg/g sodium selenite addition; A150 the treatment group with 150 µg/g sodium selenite addition; A200 the treatment group with 200 µg/g sodium selenite addition. Data with different lower-case letters display significant differences (p-value < 0.05) by the LSD method using a one-way ANOVA. MGR, Size, Thickness, SW, CW, AC and Yield were replicated 8–10 times, while CP and TSC were replicated 3 times. Figure 1. [70]Figure 1 [71]Open in a new tab Fruiting bodies of A. cornea at the mature stage in different treatments. Abbreviations: (a) ACK the control group without selenium addition; (b) A20 the treatment group with 20 µg/g sodium selenite addition; (c) A50 the treatment group with 50 µg/g sodium selenite addition; (d) A100 the treatment group with 100 µg/g sodium selenite addition; (e) A150 the treatment group with 150 µg/g sodium selenite addition; (f) A200 the treatment group with 200 µg/g sodium selenite addition. Generally, A. cornea cultured in the substrate with 100 µg/g of selenium outperformed other treatments in terms of improved agronomic traits and rich nutrients. Besides, the total selenium contents were significantly different in four growing periods between A100 and CK (P < 0.05) (Table [72]S1). The highest total selenium content was tested at budding stage (136.46 µg/g), and subsequently dropped to 32.24 µg/g at mature stage. Thus, the treatment was subsequently selected for transcriptome analysis to explore the effect of selenium accumulation on the gene expression in A. cornea at the budding and mature stages. Sequencing of A. cornea Transcriptome and De Novo Assembly High-throughput RNA sequencing was used to overview the gene expression profiles of A. cornea between the selenium-treated samples at the concentration of 100 µg/g (A100) and the control (ACK) at the budding and mature stages based on the Illumina NextSeq500 platform. After quality examination and filtering, an average of 3.09 × 10^7 clean reads per sample were obtained (Table [73]S2). The clean reads were assembled to produce 7.77 × 10^5 contigs, 3.06 × 10^5 transcripts and 2.56 × 10^5 unigenes (Table [74]S3). The average sizes of them were 251.34 bp, 520.54 bp and 452.75 bp, respectively. The length of the unigenes ranged from 200 bp to over 5,000 bp. Those ranging from 200 to 700 bp accounted for about 87% of the total number of the unigenes according to the sequence length distribution analysis (Table [75]S4). In general, the RNA-Seq data were of high quality for subsequent transcriptome analysis. Functional Annotation of A. cornea Transcriptome All the assembled unigenes from clean reads were aligned against the sequences from five frequently used databases for functional annotation, including NR, GO, KEGG, eggNOG and SwissProt (Fig. [76]S1). The proportions of the unigenes matched to the databases were 43.34% (NR), 14.90% (GO), 2.06% (KEGG), 34.33% (eggNOG) and 36.34% (SwissProt), respectively, with a common percentage of 0.73% in all the five databases. A total of 110783 unigenes from A. cornea transcriptome were matched to the reference sequences of 782 other species in the NR database. In particular, 42% of the total unigenes were found to match to those of A. delicata TFB-10046 SS5 (Fig. [77]S2). In addition, the unigenes with more than 1% of significant hits included Batrachochytrium dendrobatidis JAM81, Rhizopus delemar RA 99–880, Arthrobotrys oligospora ATCC 24927, etc. (Table [78]S5). The unigenes with homologs found from other species accounted for 37.49%. A total of 213521 unigenes in the A. cornea transcriptome were annotated to GO database and grouped into 50 functional groups (Fig. [79]2). There were 19 functional groups involved in biological process, 16 in cellular component and 15 in molecular function, accounting for 38.54%, 41.08%, and 20.39% of the total annotated unigenes, respectively. Among all the GO classifications, “cellular process”, “metabolic process”, “cell”, “catalytic activity”, “cell part”, “binding”, “single-organism process”, “organelle”, “membrane” and “macromolecular complex” were high represented (Table [80]S6). However, few genes related to “biological adhesion”, “locomotion”, “extracellular matrix”, “extracellular matrix component” and “nutrient reservoir activity” were found. Figure 2. [81]Figure 2 [82]Open in a new tab GO classification of the A. cornea transcriptome. X-axis represents the number of unigenes involved in the distinct GO terms. Y-axis displays the descriptions of GO terms. To obtain more information on the functions of each protein, eggNOG analysis was performed. Finally, a total of 26 eggNOG categories was detected (Fig. [83]3). The categories of “function unknown” and “general function prediction only” were dominant, accounting for 16.28% and 14.18%, respectively. Figure 3. [84]Figure 3 [85]Open in a new tab Classification of eggNOG annotations of the A. cornea transcriptome. The unigenes were annotated using eggNOG. The capital letters on the X-axis represent different eggNOG categories. Y-axis shows the number of unigenes in each eggNOG category. The unigenes were mapped to the reference pathways in the KEGG database to identify the biological pathways in A. cornea. A total of 5721 unigenes were annotated and mapped to 35 KEGG pathways (Fig. [86]4 and Table [87]S7). The pathways with the most unigenes in A. cornea were involved in “translation” (13.13%), “carbohydrate metabolism” (7.32%), “folding, sorting and degradation” (6.68%) and “signal transduction” (6.55%). None of the unigenes were assigned to the pathways of “enzyme families”, “RNA family” and “environmental adaptation”. The type of “metabolism” maintained the most unigenes, accounting for 41.57% in A. cornea. Figure 4. [88]Figure 4 [89]Open in a new tab Classification of the KEGG pathways based on the A. cornea transcriptome. The grey blocks on the right represent the corresponding KEGG classes of the KEGG pathways. The annotations to these databases provided a basic resource in A. cornea, and it aided further understanding of the effect of selenium accumulation on the biological activities in this fungus. Analysis of differentially expressed genes The differential gene expression patterns of A. cornea between the selenium-treated group (A100) and the control group (ACK) at the budding (b) and mature stages (m) were shown in Fig. [90]5. The RPKM method was applied to calculate the expression level of each unigene. The growth stage separated the A. cornea transcriptomes into two clusters. The one at the budding stage was further separated into two sub-clusters by two treatments. Interestingly, the up-expressed genes at the budding stage in both treatments became down-expressed at the mature stage. As illustrated in Fig. [91]S3 and Table [92]S8, selenium-related gene expression in A. cornea was more active at the budding stage than that at the mature stage. Among the 3907 differentially expressed unigenes (DEGs) between the selenium-treated and control groups at the budding stage, 1956 were up-expressed and 1951 were down-expressed. Besides, there were 891 selenium-responsive DEGs detected at the mature stage, including 575 up-expressed and 316 down-expressed unigenes. Figure 5. Figure 5 [93]Open in a new tab  A heatmap showing the log[2](FoldChange) values of the selenium-responsive DEGs (n = 3) in each sample. The DEGs and samples were subject to bidirectional clustering analysis using the R package Pheatmap based on the Euclidean distance and complete linkage clustering. The up-expressed DEGs are coloured in red and the down-regulated DEGs in green, respectively. KEGG pathway enrichment analysis was used to illustrate the overrepresented metabolic and signalling pathways involved in the growth of A. cornea in response to the selenium supply at different growth stages (Table [94]2 and Fig. [95]6). At the budding stage, the majority of selenium-responsive DEGs involved in metabolism and genetic information processing pathways were up-expressed, representing 86.57% (116) and 98.13% (105), respectively. However, only a few selenium-responsive DEGs were related to metabolism (5) and genetic information processing (9) at the mature stage, of which, 4 pathways related to metabolism were up-expressed and 7 genetic information processing pathways were up-expressed. In addition, all the DEGs involved in the organismal systems (endocrine system) were up-expressed in the presence of selenium at the budding and mature stages (e.g., immune systems and nervous system). Table 2. Overrepresented KEGG pathways between the selenium-treated and control groups of A. cornea at the budding and mature stages, respectively. Pathway ID KEGG name Class Level 1 Class Level 2 Down-DEGs Up-DEGs Total DEGs Total unigenes p-value At the budding stage ko04152 AMPK signaling pathway Environmental Information Processing Signal transduction 0 6 6 39 0.042 ko03010 Ribosome Genetic Information Processing Translation 2 105 107 437 0.000 ko00250 Alanine, aspartate and glutamate metabolism Metabolism Amino acid metabolism 0 8 8 38 0.003 ko00220 Arginine biosynthesis Metabolism Amino acid metabolism 2 5 7 29 0.002 ko00400 Phenylalanine, tyrosine and tryptophan biosynthesis Metabolism Amino acid metabolism 0 5 5 25 0.022 ko00350 Tyrosine metabolism Metabolism Amino acid metabolism 1 4 5 30 0.045 ko00290 Valine, leucine and isoleucine biosynthesis Metabolism Amino acid metabolism 1 8 9 27 0.000 ko00254 Aflatoxin biosynthesis Metabolism Biosynthesis of other secondary metabolites 0 3 3 4 0.001 ko00620 Pyruvate metabolism Metabolism Carbohydrate metabolism 0 10 10 62 0.007 ko00500 Starch and sucrose metabolism Metabolism Carbohydrate metabolism 6 4 10 75 0.025 ko00920 Sulfur metabolism Metabolism Energy metabolism 0 4 4 21 0.047 ko00061 Fatty acid biosynthesis Metabolism Lipid metabolism 0 5 5 16 0.003 ko00100 Steroid biosynthesis Metabolism Lipid metabolism 0 8 8 20 0.000 ko00770 Pantothenate and CoA biosynthesis Metabolism Metabolism of cofactors and vitamins 2 7 9 37 0.001 ko01210 2-Oxocarboxylic acid metabolism Metabolism Overview 3 15 18 57 0.000 ko01230 Biosynthesis of amino acids Metabolism Overview 3 23 26 187 0.000 ko01212 Fatty acid metabolism Metabolism Overview 0 7 7 51 0.050 ko04922 Glucagon signaling pathway Organismal Systems Endocrine system 0 6 6 40 0.046 At the mature stage ko04010 MAPK signaling pathway Environmental Information Processing Signal transduction 2 0 2 34 0.027 ko04015 Rap1 signaling pathway Environmental Information Processing Signal transduction 2 0 2 46 0.047 ko03010 Ribosome Genetic Information Processing Translation 7 2 9 437 0.005 ko04062 Chemokine signaling pathway Organismal Systems Immune system 2 0 2 35 0.029 ko04720 Long-term potentiation Organismal Systems Nervous system 2 0 2 36 0.030 ko00232 Caffeine metabolism Metabolism Biosynthesis of other secondary metabolites 1 0 1 3 0.023 ko00140 Steroid hormone biosynthesis Metabolism Lipid metabolism 1 0 1 5 0.037 ko00410 beta-Alanine metabolism Metabolism Metabolism of other amino acids 1 1 2 24 0.014 [96]Open in a new tab Abbreviations: Down-DEGs number of the down-expressed genes in response to selenium addition; Up-DEGs number of the up-expressed genes in response to selenium addition; Total DEGs total number of differentially expressed genes in response to selenium addition; Total unigenes total number of unigenes. Figure 6. [97]Figure 6 [98]Open in a new tab KEGG Pathway Enrichment of ACKb vs A100b (a) and ACKm vs A100m (b) FDR, false discovery rate adjusted p-value in multiple tests. Genes involved in metabolism The metabolism-related pathways were overrepresented at the budding stage, and a large number of up-expressed selenium-related genes participated in these pathways. Among them, five were involved in amino acid metabolism. These associations included unigenes encoding for alanine, aspartate and glutamate metabolism (ko00250), arginine biosynthesis (ko00220), phenylalanine, tyrosine and tryptophan biosynthesis (ko00400), tyrosine metabolism (ko00350) and valine, leucine and isoleucine biosynthesis (ko00290). Besides, two pathways subordinated to lipid metabolism, including fatty acid biosynthesis (ko00061) and steroid biosynthesis (ko00100). Other metabolism-related pathways were also revealed up-expressed, like 2-Oxocarboxylic acid metabolism (ko01210), biosynthesis of amino acids (ko01230), pyruvate metabolism (ko00620), etc. However, the pathway of starch and sucrose metabolism (ko00500) showed down-expressed in response to the selenium supply. At the mature stage, three metabolism-related pathways displayed lower-expressed states, and they were caffeine metabolism (ko00232), steroid hormone biosynthesis (ko00140) and beta-alanine metabolism (ko00410). Genes involved in ribosome and other pathways We found the greatest number of up-expressed unigenes in ribosome (ko03010) in response to the selenium supply at the budding stage. But the pathway of ribosome got lower-expressed at the mature stage. Other detected genes that significantly changed between treatments were mainly involved in enzyme activity and protein synthesis. At the budding stage, an average of 57.24% genes were related to protein synthesis (e.g. ribosomal protein, other predicted protein), and an average of 21.15% participated in enzyme activities (e.g. peroxidase, lipase, glycoside hydrolase), and a smaller portion of genes regulated other cellular life activities, such as transport, ATP, translation. It was noteworthy that a total of 7 genes that related to biological metabolism of glutathione exhibited up-expressed in the selenium-treated group. Validation of Transcriptome data by RT-qPCR To validate the gene expression profiles in transcriptome of A. cornea, four DEGs were selected for real-time reverse transcription PCR (RT-qPCR). The genes c108476_g4 related to “biosynthesis of amino acids” and c102698_g1 related to “metabolism of xenobiotics by cytochrome P450”were up-expressed in control group. While the gene c96467_g1 related to “bile secretion” was down-expressed in control group. Furthermore, c135295_g1, annotated neither in GO nor KEGG database, upregulated in control group at both stages, in which, it was with a higher expression level at the budding stage. All of the four selected genes were successfully amplified, and the expression patterns were highly consistent with the DEG results obtained from the transcriptome analysis, confirming the reliability of the DEG analysis from RNA-Seq data (Table [99]3). Table 3. Unigenes used for validation of gene expression profiles in A. cornea transcriptome. Group Gene ID DESeq analysis based on RNA-seq Validation of the DEGs by qRT-PCR analysis Log[2]Fold Change p-value ACK (2-ΔΔCt) A100 (2-ΔΔCt) p-value A100b vs ACKb c135295_g1 −1.65 1.33E-05 1.00 ± 0.11 0.72 ± 0.08 9.14E-03 c108476_g4 −2.45 9.44E-10 1.00 ± 0.11 0.30 ± 0.05 5.89E-07 c102698_g1 −3.07 8.07E-14 1.00 ± 0.02 0.14 ± 0.04 3.13E-08 c96467_g1 2.25 1.78E-08 1.01 ± 0.13 9.76 ± 0.27 1.80E-27 A100m vs ACKm c135295_g1 −6.12 8.45E-08 1.01 ± 0.13 0.01 ± 0.00 2.41E-09 [100]Open in a new tab 2-ΔΔCt displays relative gene expression level using RPL4 as a reference gene in RT-qPCR analysis. Data are presented as the means ± standard deviation of three replicates. Abbreviations: ACK control group without selenium addition in the substrate; A100 treatment group with 100 µg/g of selenium addition in the substrate; m at the mature stage; b at the budding stage. Discussion Selenium supply affects A. cornea physiological growth In the present study, A. cornea maintained its normal growth with selenium supplementation in the substrate ranging from 20 µg/g to 200 µg/g. However, the mycelial growth of A. cornea slowed down in response to selenium supplementation. Malinowska et al. found that sodium selenite with low concentration of 25 mg/L did not significantly affect mycelial growth of Hericium erinaceum, whereas larger supplementation of Na[2]SeO[3] resulted in a mycelial yield decrease^[101]32, which was consistent with our study. Furthermore, increased selenium supplementation caused thinning and extensive branching of mycelia of G. lucidum^[102]33. Besides, a dynamic change of selenium enrichment was observed, and selenium was finally accumulated efficiently in A. cornea. According to Goyal et al., the uptake of selenium in mycelia of G. lucidum showed an increasing trend as selenium supplementation increased^[103]33. The selenium absorption of mycelia might consequently affect selenium accumulation in fruiting bodies, and its accumulation differs from edible fungi. G. lucidum was reported to contain up to 72 μg Se/g in its fruiting body^[104]34, while the maximum selenium content in the studied A. cornea was approximately 51 μg Se/g. Moreover, the selenium concentration in Lentinula edodes first ascended and then descended over time^[105]35, which was consistent with the present study. The present study revealed richer polysaccharides in fruiting bodies of A. cornea grown in the selenium-treated substrate. As a significant trace element, selenium is usually supplemented into the mushroom media, and the inorganic selenium is then bio-absorbed and subsequently transformed to organically bound one^[106]36. Selenium is demonstrated to bind with proteins and polysaccharides^[107]37, or form a conjugated complex with polysaccharides of edible fungi^[108]38. Chen et al. reported that, sodium selenite supplementation in the media of Spirulina platensis contributed an increase to extracellular polysaccharides in dose-dependent manner, and the extracellular polysaccharide was probably produced by mycelia in response to oxidative damage^[109]39. Additionally, the synthesis of proteins and polysaccharides in G. lucidum was enhanced with the selenium concentration below 100 ug/g in the substrate^[110]15. Consequently, a proper range of selenium supplementation is suggested for A. cornea cultivation to improve the agronomic properties (e.g. fresh yield, polysaccharide content). De Novo assembly of transcriptome reveals the close relationship between A. cornea and A. delicate The GC content of A. cornea was approximately 57.93%, which was lower than what was reported by Zhou et al. (61.3%)^[111]40. No reference genome was found, so an approximation of transcriptome coverage for assembly was made. Based on NR database, most hit genes were obtained in A. delicata TFB-10046 SS5. The two species were highly similar in genes, which was consistent with Zhou et al. study^[112]40. Furthermore, none of the four tested restriction enzymes including Hae ill, Taq i, Hinfl and Msp 1.ms could separate A. cornea and A. delicate, suggesting their exact relationship of affinis species^[113]41. Besides, the unigenes with more than 1% of significant hits included Batrachochytrium dendrobatidis JAM81, a fungus that affects the occurrence of Amphibians; Rhizopus delemar RA 99–880, a fungus that belongs to the Zygomycota; Arthrobotrys oligospora ATCC 24927; Dichomitus squalens LYAD-421 SS1 and three Trichoderma species. Functional annotation reveals the importance of protein synthesis throughout the growth of A. cornea The three databases GO, GOSlim and KEGG complemented for functional annotation of genes. GO analysis presented catalytic activity as one important functional category during the growth of A. cornea. Various enzymes, including cellulose, lignin peroxidase, were involved in catalysis during mushroom growth, and catalytic activity dominated such biological processes like metabolism, nutrition and energy conversion^[114]42–[115]44. The present study revealed the largest number of unigenes involved in translation, agreeing with the previous study^[116]40. Translation is considered as an indispensable process in the genetic central dogma and plays a key role in biological growth. Notably, two KEGG pathways “ribosome” and “biosynthesis of amino acids” were detected at both budding and mature stages, suggesting the significance of protein synthesis during these two periods. Expression of the genes is growth stage dependent Interestingly, we found a higher number of DEGs at the budding stage than at the mature stage. And the up-expressed genes at the budding stage had lower expression level at the mature stage. Different gene expression patterns indicated their significance at the respective developmental stages of mushrooms. Some genes function differently at different growth stages^[117]45,[118]46. Based on the transcriptional analysis of G. lucidum, numerous metabolic pathways involved in biosynthesis upregulated during the mycelium stage, whereas, unigenes related to degradation activity were over-expressed during the fruiting body stage^[119]27. Morin et al. compared gene expression in Agaricus bisporus of fruiting bodies and undifferentiated mycelium, and found genes coding for hydrophobins, lectins were over-expressed at fruiting body stage^[120]47. Besides, the transcriptomes of Hypsizygus marmoreus at four growth stages revealed that, stress signals including MAPK, cAMP might be the most important pathways during the fruiting body stage^[121]26. What’s more, in analysis of DEGs in Agrocybe aegerita transcriptome, Wang et al. uncovered genes related to polysaccharide and steroid biosynthesis were differentially expressed at mycelium and fruiting body stages^[122]48. In the present study on A. cornea, unigenes involved in biosynthesis of amino acids were high expressed at the budding stage, and the ribosome pathway relating to translation was in high expression at the budding and mature stages. Physiological metabolisms vary in different growth periods, and all kinds of genes are responsible to regulate and control diverse life activities^[123]49,[124]50. At the budding stage of A. cornea, reproductive growth dominates with primordium differentiation. Instead, nutrients are enriched in fruiting bodies at the mature stage. As one of the most complex and rapid developmental stage of mushroom growth, the fruiting process is reported to be regulated by cellular processes^[125]51. At this growth stage, dramatic morphological changes are caused by complicated regulations on transcription and translation^[126]52. And fruiting body development is often motivated by environmental changes^[127]53. It’s inferred that, most of physiological activities in A. cornea tend to slow down or even cease when its tissues get mature at the mature stage. As a result, massive unigenes are down-expressed or shutdown. Selenium supply significantly affects the metabolism of A. cornea The transcriptome analysis was performed to show the effect of selenium addition on A.cornea at different growth stages. Selenium stimulated the synthesis of various proteins, and it would improve the medical use of A. cornea. Selenium was absorbed and converted into various substances, including selenocysteine, selenomethionine and other compounds^[128]35. It was demonstrated that the accumulated selenium in G. lucidum protein showed antioxidant activities and selenium protected DNA from being oxidized strongly^[129]54. In the present study, 7 genes related to biological metabolism of glutathione were up-expressed. As was reported, selenium contributes to be part of selenoproteins, especailly glutathione peroxidase^[130]55. And glutathione peroxidase contributes to protect exposed organisms from damage caused by reactive oxygen species^[131]56. In addition, selenium upregulated genes of glucanase synthesis in the present study. Accordingly, selenium promoted adenosine accumulation in Cordyceps militaris during its growth^[132]57. Meanwhile, one of the selected genes (c102698_g1) for verification was related to “metabolism of xenobiotics by cytochrome P450”. Cytochrome P450 (CYPs) is considered as an important role for biological metabolism. Morin et al. detected genes of cytochrome P450 in Agaricus bisporus grown on compost, with a relatively low number but in high expression^[133]58. The significance of the present study Edible mushrooms are good for human health and longevity due to their natural antioxidant compounds^[134]59,[135]60. A. cornea is a particularly popular edible mushroom, and has attracted substantial interest^[136]21,[137]61. Most of the edible fungi have a strong ability to enrich mineral elements^[138]62,[139]63. The enriched elements have been demonstrated to increase the nutritive value of edible fungi, among which selenium is an outstanding one^[140]64,[141]65. As an essential trace element for humans, selenium is feasible to be added into the substrates for producing selenium-biofortified mushrooms with great agronomic value^[142]66. Mushrooms including Agaricus bisporus, Lentinula edodes, are capable of accumulating selenium through enzymatic digestion and transforming it into organic components that are beneficial to human^[143]67–[144]70. The bioavailability of organically bound selenium from dietary supplements is better with lower toxicity for the mushrooms than inorganic selenium compounds^[145]5. The effect of selenium on A. cornea in the present study strongly supported the above-mentioned views. The quality of A. cornea in the selenium-treated cultivation substrate is strongly enhanced by the accumulation of this trace element and indirectly by the selenium-enhanced biological activities. The present study provides a solid support for inorganic selenium addition into the substrate during the commercial cultivation of A. cornea in agricultural practice. Methods Cultivation of A. cornea The A. cornea cultivar named Shanghai No. 1 was provided by the Soil and Fertilizer Research Institute, Sichuan Academy of Agricultural Science. The substrate comprised cottonseed hull (10%), corn cob (30%), sawdust (33%), rice bran (20%), corn flour (2%), gypsum (1%) and lime (4%), and all the materials were fresh, dry and unspoiled. A sodium selenite (Na[2]SeO[3]) solution was added to the substrate as the selenium source. After thorough mixing, the final concentrations of sodium selenite in the substrate were determined to be 20 µg/g, 50 µg/g, 100 µg/g, 150 µg/g, and 200 µg/g and were labelled A20, A50, A100, A150, A200, respectively. The control group without the addition of selenium to the substrate was labelled ACK. The substrate was placed in polyethylene cultivation bags (size: 42 cm × 22 cm × 0.005 cm) and atmospheric sterilized at 100 °C for 14 hours. After sterilization, the bags were cooled to room temperature and placed in a laminar flow hood for the inoculation of A. cornea. After inoculation, they were planted in the cultivation base of Jiandi, Shifang, China (N 31°12′20.33″, E 104°01′42.35″). The cultivation area was ventilated. The cultivation area had been cleaned and disinfected before the experiment was conducted. The mycelial growth rate of A. cornea was measured at the hyphal stage. Other agronomic traits including the size, thickness and weight of a single random auricular patch, fresh yield, crude polysaccharide and total selenium concentration in the fruiting bodies in the different treatments were investigated at the mature stage. There were three replicates per treatment. We used 15 cultivation bags in each replicate and a total of 270 cultivation bags were used for the study. Excel and SPSS13.0 were used for statistical analysis. Data with different lower-case letters display significant differences (p-value < 0.05) by the LSD method of a one-way ANOVA. Sample collection for transcriptome analysis The tissues of A. cornea cultured in ACK and A100 were sampled with three replicates at the budding and mature phases and denominated ACKb, ACKm, A100b and A100m. Disposable disinfected gloves, sterilized tweezers and knives were prepared and used to take samples. Fresh tissues of the same treatment were collected and pooled, weighing over 500 mg per sample. The samples were filled in 2 mL Eppendorf tubes (Eppendorf, Germany), stored in liquid nitrogen and immediately sent to Personalbio (Shanghai, China) for RNA extraction and transcriptome sequencing. RNA extraction, library preparation and sequencing Total RNA from the tissues of A. cornea was extracted using a Qiagen RNeasy mini kit (Qiagen, Germany) including DNAase treatment following the manufacturer’s instructions. The quality of the extracted RNA was examined using agarose gel electrophoresis. The ribosomal RNA that had contaminated the total RNA was eliminated using a Ribosomal RNA Removal Kit (Illumina, San Diego, CA, USA). The remaining RNA was cleaved into small fragments with lengths of 200–300 bp. The RNA was reverse transcribed to cDNA using a Fast Quant RT Kit (with gDNAase) (TIANGEN, Beijing) according to the manufacturer’s instructions. The first strand of cDNA was synthesized with random hexamers and reverse transcriptase using RNA as the template. The second strand was synthesized based on the first strand with the base T replaced by U. To obtain more accurate results for the subsequent analysis in the gene functional annotation and expression, a TruSeq Stranded mRNA Library Prep Kit (Illumina, San Diego, CA, USA) that can determine the direction of justice chains and antisense chains was adopted to construct the libraries^[146]71 The cDNA fragments were enriched using PCR amplification after the generation of the library. The libraries with fragment lengths between 300 and 400 bp were examined using an Agilent 2100 Bioanalyzer (Agilent Technologies, CA, USA). The total and effective concentrations of the libraries were determined. The libraries with varied index sequences were pooled in proportion based on the effective concentrations of the libraries and the amount of data required by the libraries. The pooled libraries were uniformly diluted to 2 nM and formed single stranded libraries using alkaline denaturation. Finally, the libraries were paired-end sequenced based using an Illumina NextSeq500 (Illumina, San Diego, CA, USA). All the raw sequences were deposited in the NCBI Sequence Read Archive (SRA) database with the accession NO. SAMN07184775-SAMN07184786. De novo transcriptome assembly and gene functional annotation Raw reads in the FASTQ files were checked and filtered by quality control using the FastQC program^[147]72 The adapters were removed from the sequences. The sequences with a length less than 50 bp and with an average sequence quality score lower than Q20 were removed. Trinity^[148]73, a short reads assembling program, was used to perform de novo assembly of the clean reads as described by Zhou^[149]40. High-quality sequences were alternatively spliced from the beginning to obtain transcript sequences using Inchworm, Chrysalis and Butterfly. The longest sequence of every cluster was selected as one unigene, which was finally annotated against the databases of GO (Gene Ontology)^[150]74, KEGG (Kyoto Encyclopedia of Genes and Genome)^[151]75, eggNOG (evolutionary genealogy of genes: Non-supervised Orthologous Groups)^[152]76 and SwissProt (Swiss-Prot protein)^[153]77. The alignment result (the read count of each unigene) was obtained by aligning the clean sequences with the corresponding unigene for the following differential gene expression and enrichment analysis. Differentially expressed genes The expression quantification of RNA-Seq data was performed using RSEM, software for accurate transcript quantification^[154]78. The de novo assembled transcriptome was used as a reference. The alignment result was investigated. The FPKM value of each unigene was calculated as an estimation of the expression level. Within RNA-Seq technology, FPKM represents the expected number of fragments per kilobase of transcript sequence per million reads sequenced^[155]79. The differences in unigene expression were analysed using DESeq (Version 1.18.0)^[156]78. The significantly differentially expressed genes (DEGs) were screened with the threshold: |log[2](FoldChange)| > 1 and p-value < 0.05. Two-way hierarchical clustering (complete linkage) was conducted to show the expression pattern of each DEG across all the samples using the R package Pheatmap^[157]80 based on the Euclidean distance. Volcano plots were drawn to show the patterns of unigene expression between the selenium-treated (A100) and control groups (ACK) at the budding and mature stages using the R package ggplot2^[158]81. Functional enrichment analysis GO term enrichment analysis was conducted using the GOseq R packages based on Wallenius non-central hyper-geometric distribution^[159]82. All the DEGs were first assigned to a set of predefined Terms of Gene Ontology (TOG) to profile their biological functions. The TOGs overrepresented with DEGs were examined compared to a randomly sampled set of genes using hypergeometric testing^[160]83. The KEGG pathways enrichment analysis was performed similarly based on the hypergeometric distribution principle using KOBAS software^[161]84. A p-value < 0.05 was used as the threshold to determine the significance of the enrichment pattern of each TOG or KEGG pathway between the selenium treated (A100) and control groups (ACK) at the budding and mature stages, respectively. Validation of differentially expressed genes by RT-qPCR A total of four unigenes identified by RNA-Seq were selected for expression validation using the RT-qPCR analysis. Approximately 3 mg of RNA of each sample was reversely transcribed using a Super RT Kit (TaKaRa, Osaka, Japan), following the manufacturer’s instructions. The sequences of the primers used to amply the selected genes are shown in Table [162]S9. RPL4 (Ribosomal Protein L4) was amplified as a reference gene to normalize gene expression for each RT-PCR using primers described by Xu J et al.^[163]85. At least three independent RT-PCR reactions were performed to confirm the expression of each gene in one sample. Supplementary information [164]Supplementary Figure 1^ (134.7KB, pdf) [165]Supplementary Figure 2^ (112.5KB, pdf) [166]Supplementary Figure 3^ (92.1KB, pdf) [167]Supplementary Table 1^ (103.3KB, pdf) [168]Supplementary Table 2^ (88.1KB, pdf) [169]Supplementary Table 3^ (92.1KB, pdf) [170]Supplementary Table 4^ (74.9KB, pdf) [171]Supplementary Table 5^ (89.5KB, pdf) [172]Supplementary Table 6^ (73.3KB, pdf) [173]Supplementary Table 7^ (78KB, pdf) [174]Supplementary Table 8^ (98KB, pdf) [175]Supplementary Table 9^ (72.3KB, pdf) Acknowledgements