Enhanced Expression of Thaumatin-like Protein Gene (LeTLP1) Endows Resistance to Trichoderma atroviride in Lentinula edodes

Lentinula edodes (shiitake mushrooms) is heavily affected by the infection of Trichoderma atroviride, causing yield loss and decreases quality in shiitake mushrooms. The selection and breeding of fungal-resistant L. edodes species are an important approach to protecting L. edodes from T. atroviride infection. Herein, a highly resistant L. edodes strain (Y3334) and a susceptible strain (Y55) were obtained by using a resistance evaluation test. Transcriptome analyses and qRT-PCR detection showed that the expression level of LeTLP1 (LE01Gene05009) was strongly induced in response to T. atroviride infection in the resistant Y3334. Then, LeTLP1-silenced and LeTLP1-overexpression transformants were obtained. Overexpression of LeTLP1 resulted in resistance to T. atroviride. Compared with the parent strain Y3334, LeTLP1-silenced transformants had reduced resistance relative to T. atroviride. Additionally, the LeTLP1 protein (Y3334) exhibited significant antifungal activity against T. atroviride. These findings suggest that overexpression of LeTLP1 is a major mechanism for the resistance of L. edodes to T. atroviride. The molecular basis provides a theoretical basis for the breeding of resistant L. edodes strains and can eventually contribute to the mushroom cultivation industry and human health.


Introduction
Lentinula edodes (also known as Xiang Gu or shiitake), a widely cultivated edible mushroom, is famous for its nutritional properties and pharmacological effects, such as its hypocholesterolemic, anti-inflammatory, antitumor, and neuroprotective effects [1][2][3][4][5][6][7][8]. The cultivation, production, and consumption of shiitake are important nowadays not only in Asian countries but also in western countries. The white-rot basidioycete L. edodes grows on the wood logs, and it is now cultivated on the sterilized sawdust-based substrates [1,9,10]. In practice, making use of sawdust-based cultivation as a replacement for natural logs has contributed to the expansion of the production and consumption of L. edodes [11]. However, the cultivation of L. edodes is severely affected by the infection of Trichoderma spp., which overgrows mushroom mycelia and kills them, resulting in a reduction in the mushroom yield [12,13]. As a result, the green mold disease induced by Trichoderma spp. on L. edodes severely limits the sustainable development of the mushroom industry.
The main Trichoderma species affecting L. edodes in China are T. harzianum, T. atroviride, T. viride, T. longibrachiatum, T. polysporum, T. pleuroticola, and T. oblongisporum [12][13][14][15]. Among them, T. harzianum is the most general and widespread pathogen in mushrooms [14,[16][17][18]. The mycelia of T. atroviride possess a stronger capacity of overgrowth on L. edodes, and they attack the mycelium of L. edodes [12]. The other five species-namely, T. viride, T. Collection of four mycelial samples (s1, s2, w1, and w2) of peripheral hyphal zones from L. edodes at the AC (after contact for 24 h) stage. The highly resistant (Y3334) and susceptible (Y55) L. edodes strains were grown on PDA plates that were covered with cellophane at 25 °C in a dark environment. The resistant and susceptible samples of the mycelia of L. edodes (s2, w2) were harvested after the T. atrovide strain 92-1 had made contact with the L. edodes strains for 24 h. As controls, the Y3334 strain and Y55 strain of L. edodes were confronted with themselves. The mycelial samples of resistant and susceptible L. edodes (s1 and w1) were harvested after contact for 24 h; s1 represents Y3334 vs. Y3334; s2 represents Y3334 vs. T. atrovide strain 92-1; w1 represents Y55 vs. Y55; w2 represents Y55 vs. T. atrovide strain 92-1. The red square represents the sampling location.

RNA Extraction, Library Preparation, and Sequencing
Mycelial samples were ground into fine powders in liquid nitrogen. Total RNAs were extracted using RNAiso Plus (TaKaRa, Dalian, China) and Fruit-mate TM for the RNA Purification Kit (TaKaRa, Dalian, China) according to the manufacturer's instructions. RNA Figure 1. Collection of four mycelial samples (s1, s2, w1, and w2) of peripheral hyphal zones from L. edodes at the AC (after contact for 24 h) stage. The highly resistant (Y3334) and susceptible (Y55) L. edodes strains were grown on PDA plates that were covered with cellophane at 25 • C in a dark environment. The resistant and susceptible samples of the mycelia of L. edodes (s2, w2) were harvested after the T. atrovide strain 92-1 had made contact with the L. edodes strains for 24 h. As controls, the Y3334 strain and Y55 strain of L. edodes were confronted with themselves. The mycelial samples of resistant and susceptible L. edodes (s1 and w1) were harvested after contact for 24 h; s1 represents Y3334 vs. Y3334; s2 represents Y3334 vs. T. atrovide strain 92-1; w1 represents Y55 vs. Y55; w2 represents Y55 vs. T. atrovide strain 92-1. The red square represents the sampling location.

RNA Extraction, Library Preparation, and Sequencing
Mycelial samples were ground into fine powders in liquid nitrogen. Total RNAs were extracted using RNAiso Plus (TaKaRa, Dalian, China) and Fruit-mate TM for the RNA Purification Kit (TaKaRa, Dalian, China) according to the manufacturer's instructions. RNA concentrations and RNA integrity numbers were determined by using agarose gel electrophoresis and an Agilent 2100 Bioanalyzer (Agilent Technologies, Carpinteria, CA, USA), respectively. Amounts of 20 µg of RNA were equally isolated from mycelia for mRNA isolation, cDNA library construction, and sequencing according to the manufacturer's instructions (Illumina, San Diego, CA, USA). The twelve libraries were sequenced through paired-end sequencing on an Illumina Hiseq 4000 sequencer (BGI). Sequencing services were provided by the Beijing Genomics Institute (BGI)-Shenzhen, Wuhan, China.

Transcriptome Data Analysis
Twelve sets of raw data of all samples were pre-treated with FastQC, filtered, and trimmed via Trimmonmatic (parameter, PE and phred33). The indices were constructed Life 2021, 11, 863 4 of 14 via bowtie2 according to the L. edodes genome files (L. edodes W1-26 genome v1.0: http: //shiitakegdb.chenlianfu.com/ accessed on 15 November 2019), followed by mapping of the RNA-seq reads to the reference genome of L. edodes with Tophat2 by using the default settings [33]. Subsequently, Samtools was used to create SAM files for the bam files [34]. Integrated and comprehensive transcript sets were obtained via Cufflinks and Cuffmerge. We ran the Cuffdiff function to find differentially expressed genes and transcripts in the samples using the default settings [33]. One-way ANOVA was applied, and a step-up FDR p-value was calculated to sort out the statistically significant differentially expressed genes (DEGs) by filtering the data at p < 0.05 (FDR) and with >2-fold change in the expression. The GO term enrichment and functional categorization for biological processes and cellular and molecular functions were carried out using Gene Ontology at the BGI databank.

qRT-PCR
The L. edodes mycelial samples for the RNA-seq analysis were also used for qRT-PCR. Gene-specific primers were designed using the Primer Premier 5.0 [35] for the target genes as well as the reference actin-1 gene (Table S2). The relative expression levels of the ten selected genes (Table S2) were calculated by using the 2 −∆∆CT method [36].

Construction of LeTLP1-Overexpression and LeTLP1-Silencing Vectors and Fungal Transformation
The LeTLP1-overexpression vector was constructed by following the published protocols described by Yan et al. (2019) [37], with minor modifications. The CaMV 35S promoter of pCAMBIA1300 was replaced with the Legpd (L. edodes glyceraldehyde-3-phosphate dehydrogenase) promoter in order to produce the pCAMBIA1300-g vector. The Legpd promoter sequence was PCR amplified from the DNA of L. edodes strain W1-26 with the O-Legpd Promotor-F and O-Legpd Promotor-R primers ( Table 1). The full length of the LeTLP1 gene was amplified from L. edodes strain Y3334 cDNA with the O-LeTLP1-F and O-LeTLP1-R primers (Table 1), which contained the homologous arms, and then ligated into the pCAMBIA1300-g vector digested with EcoRI and KpnI in order to generate the overexpression vector pCAMBIA1300-o-LeTLP. All constructs were assessed by sequencing analysis and transferred into L. edodes strain Y55 through A. tumefaciens EHA105 infection. Table 1. Sequence information of the primers used for the construction of LeTLP1-overexpression and LeTLP1-silencing vectors.

Primer Name
Primer Sequence (5 to 3 ) For the construction of the LeTLP1-silencing vector, a 500-bp antisense fragment from LeTLP1 was amplified with the R-LeTLP1-F and R-LeTLP1-R primers (Table 1) from the cDNA of L. edodes. The Leactin promoter fragment was amplified with the R-Leactin-F and R-Leactin-R primers (Table 1) from the cDNA of L. edodes. Then, the Leactin promoter fragment, LeTLP1 antisense fragment, and the aforesaid digested pCAMBIA1300-g vector were ligated in order to generate the silencing vector pCAMBIA1300-Ri-LeTLP. All constructs were assessed through sequencing analysis and transferred into L. edodes strain Y3334 through A. tumefaciens EHA105 infection.
The ten transformants were selected randomly to analyze their over-expression and silencing efficiency with qRT-PCR by using the primers for LeTLP1 (LE01Gene05009) and by normalizing to the actin-1 gene by using the primers for the actin-1 gene (Table S2). The relative expression levels were calculated with the 2 −∆∆CT method [36]. Three replicates for each isolate were analyzed, and the experiment was independently repeated three times.

Evaluation of the Level of Resistance against T. atroviride in LeTLP1-Overexpression and LeTLP1-Silenced Transformants
In order to test the effects of LeTLP1 over-expression and silencing on the resistance of L. edodes against T. atroviride, the confrontation assays were undertaken according to the method presented in Section 2.2. The confrontations of the LeTLP1 overexpression transformants (Y55_OE_3 and Y55_OE_10), silencing transformants (TLP1-Ri-3 and TLP1-Ri-8), and their parent isolates against T. atroviride (92-1) were observed on the 15th day after inoculation with T. atroviride in order to evaluate the level of the resistance of the transformants against T. atroviride, with five replicates for each group. Infection rates (L T :L L ) were also calculated in order to evaluate the effects of T. atroviride on the mycelia of L. edodes [32].

Assays of the Antifungal Activity of LeTLP1
The antifungal activity of LeTLP1 was measured by using the paper disc method described by Imtiaj and Lee (2007) [38]. Agar discs taken from 10 day old cultures of T. atroviride strain 92-1 were placed in the center of the Petri plates. The paper discs of Tris-HCl (pH 8.0), sterilized water, 20 µg of LeTLP (Y55), 5 µg of LeTLP1, 10 µg of LeTLP1, and 20 µg of LeTLP1 were separately placed at 1 cm from the edge of the Petri plates. As a control, agar discs of T. atroviride strain 92-1 were placed in the same manner on a fresh PDA plate with the paper discs of Tris-HCl (pH 8.0) and sterilized water. The inhibition rates (the percent inhibition of mycelial growth-PIMG) were calculated as described in Imtiaj and Lee (2007) [38]. The experiments were independently repeated three times.

Statistical Analysis
Following a one-way analysis of variance (ANOVA), Duncan's multiple range test was used to evaluate significant differences in the gene expression levels, infection rates (L T :L L ), and inhibition rates (the percent inhibition of mycelial growth-PIMG).

Evaluation of Lentinula edodes Strains That Were Resistant to Trichoderma atrovide
The resistance levels of the 56 strains (Table S1) of L. edodes against T. atroviride were evaluated by using the resistance evaluation test ( Table 2). The results revealed that the 56 L. edodes strains (21 cultivated strains and 35 wild strains) could be divided into a highly resistant group (Y3334, Y29, Y1515, Y121, and c67), medium-resistant group (++ labeled strains in Table 2), and susceptible group (Y55). Among them, the highly resistant strain Y3334 and the susceptible strain Y55 were chosen to be representative for the subsequent research on the resistance mechanism of L. edodes against T. atroviride.
AC means after contact; − represents susceptible; + represents the low resistance level; ++ represents the medium resistance level; +++ represents the high resistance level.

Significant Differences in the Responses of the Highly Resistant and Susceptible L. edodes strains to T. atroviride Infection
On the PDA plate, significant differences were shown in the interactions of the highly resistant strain Y3334 and susceptible strain Y55 with T. atroviride at 24, 48, 72, 96, and 120 h post-infection (hpi) and 30 days post-infection (dpi), respectively ( Figure 2A). Notably, the infection rates (L T :L L ) were significantly lower in the confrontation of the highly resistant strain Y3334 with T. atroviride than for the susceptible strain Y55 at 24, 48, 72, 96, and 120 h, as well as 30 d ( Figure 2A). Furthermore, we observed that the mycelia of T. atroviride became ruptured and rough, and conidia were not generated in the zone of interaction between the hyphae of L. edodes (Y3334) and T. atroviride. By contrast, the mycelium of T. atroviride was smooth and straight, and many conidia that adhered to the surface of the susceptible L. edodes strain Y55 were generated ( Figure 2B).

Transcriptional Responses of the Highly Resistant and Susceptible L. edodes strains to T. atroviride
In order to screen the candidate genes for conferring resistance to T. atroviride in L. edodes, we synthesized twelve cDNA libraries from the mycelia of L. edodes, which were collected from the peripheral hyphal zones of Y3334 vs. T. atroviride (s2), Y55 vs. T. atroviride (w2), Y3334 vs. Y3334 (s1), and Y55 vs. Y55 (w1) (Figure 1). This was the first time that transcriptome RNA-sequencing data were generated for the Y3334 and Y55 strains of L. edodes.
The DEGs were obtained by comparing the gene expressions among the s2, w2, s1, and w1 groups. As shown in the Venn diagram, a total of 4880 DEGs were obtained. In both the Y3334 and Y55 strains, 239 up-regulated genes were identified ( Figure 3A). Among them, ten genes (Table S2) were selected to validate our differentially expressed gene results using qRT-PCR. The qRT-PCR results were correlated with the RNA-seq data (R 2 = 0.8575, Figure S1), which confirmed the high reliability of the RNA-seq data in this study. Interestingly, the expression level of the thaumatin-like protein gene LeTLP1 (LE01Gene05009) was strikingly increased in the resistant Y3334 when responding to infection with T. atro-Life 2021, 11, 863 7 of 14 viride, and the LeTLP1 expression level was extremely significantly higher than that in the susceptible Y55 in response to T. atroviride infection ( Figure 3B).

strains to T. atroviride Infection
On the PDA plate, significant differences were shown in the interactions of the highly resistant strain Y3334 and susceptible strain Y55 with T. atroviride at 24, 48, 72, 96, and 120 h post-infection (hpi) and 30 days post-infection (dpi), respectively ( Figure 2A). Notably, the infection rates (LT:LL) were significantly lower in the confrontation of the highly resistant strain Y3334 with T. atroviride than for the susceptible strain Y55 at 24, 48, 72, 96, and 120 h, as well as 30 d (Figure 2A). Furthermore, we observed that the mycelia of T. atroviride became ruptured and rough, and conidia were not generated in the zone of interaction between the hyphae of L. edodes (Y3334) and T. atroviride. By contrast, the mycelium of T. atroviride was smooth and straight, and many conidia that adhered to the surface of the susceptible L. edodes strain Y55 were generated ( Figure 2B).

Transcriptional Responses of the Highly Resistant and Susceptible L. edodes strains to T. atroviride
In order to screen the candidate genes for conferring resistance to T. atroviride in L. edodes, we synthesized twelve cDNA libraries from the mycelia of L. edodes, which were collected from the peripheral hyphal zones of Y3334 vs. T. atroviride (s2), Y55 vs. T. atroviride (w2), Y3334 vs. Y3334 (s1), and Y55 vs. Y55 (w1) (Figure 1). This was the first time that transcriptome RNA-sequencing data were generated for the Y3334 and Y55 strains of L. edodes.
The DEGs were obtained by comparing the gene expressions among the s2, w2, s1, and w1 groups. As shown in the Venn diagram, a total of 4880 DEGs were obtained. In both the Y3334 and Y55 strains, 239 up-regulated genes were identified ( Figure 3A). Among them, ten genes (Table S2) were selected to validate our differentially expressed gene results using qRT-PCR. The qRT-PCR results were correlated with the RNA-seq data (R 2 = 0.8575, Figure S1), which confirmed the high reliability of the RNA-seq data in this study. Interestingly, the expression level of the thaumatin-like protein gene LeTLP1 (LE01Gene05009) was strikingly increased in the resistant Y3334 when responding to infection with T. atroviride, and the LeTLP1 expression level was extremely significantly higher than that in the susceptible Y55 in response to T. atroviride infection ( Figure 3B). The Gene Ontology enrichment analysis demonstrated that the most dominant groups among the biological process components were metabolic processes and cellular processes in both the Y3334 and Y55 strains ( Figure 3C). The growth and death GO terms were only found in the highly resistant strain Y3334, while they were not found in the susceptible strain Y55. The genes in the growth and death GO terms might be resistancerelated candidate genes. Meanwhile, the catalytic and binding activities were the most abundant functional groups among the molecular functions, implying that enzyme encoding genes (such as LeTLP1) may play key roles in the regulation of the resistance of L. The Gene Ontology enrichment analysis demonstrated that the most dominant groups among the biological process components were metabolic processes and cellular processes in both the Y3334 and Y55 strains ( Figure 3C). The growth and death GO terms were only found in the highly resistant strain Y3334, while they were not found in the susceptible strain Y55. The genes in the growth and death GO terms might be resistance-related candidate genes. Meanwhile, the catalytic and binding activities were the most abundant functional groups among the molecular functions, implying that enzyme encoding genes (such as LeTLP1) may play key roles in the regulation of the resistance of L. edodes to T. atroviride.

Molecular Characterization of LeTLP1
The full length of LeTLP1 was 1993 bp. The DNA and cDNA sequence analyses revealed that the 768-bp ORF of LeTLP1 was interrupted by nine introns ( Figure 4A). The LeTLP1 gene was predicted to encode a protein of 255 amino acids that contained a signal peptide and a conserved thaumatin domain (Accession pfam00314, E-Value 1.14894e-96), which belonged to the GH64-TLP-SF superfamily ( Figure 4B). The phylogenetic analysis of the TLP indicated that LeTLP1 was clustered with the LeTLP2, while it was separate from LeTLP3, LeTLP4, and the TLP genes of other fungi ( Figure 4C). The amino acid sequences of LeTLP1 from W1-26 and Y55 were exactly the same, while the amino acid sequence of LeTLP1 from the resistant strain Y3334 exhibited two mutation sites (I-152-V and Q-245-K) compared to that of the sensitive strain Y55 ( Figure 4D).

Effects of LeTLP1 Overexpression on the Resistance of L. edodes against T. atroviride
Compared with the parent isolate Y55, the gene expression levels of LeTLP1 in the three LeTLP1-overexpression transformants Y55_OE_3, Y55_OE_5, and Y55_OE_10 increased (p < 0.01) 4.5-fold, 4.0-fold, and 6.5-fold, respectively ( Figure 5A). For two LeTLP1overexpression strains (Y55_OE_3 and Y55_OE_10), significant differences in the responses of the LeTLP1-overexpression transformants and the susceptible (Y55) L. edodes strain to the infection with T. atroviride were observed ( Figure 5B

Effects of LeTLP1 Overexpression on the Resistance of L. edodes against T. atroviride
Compared with the parent isolate Y55, the gene expression levels of LeTLP1 in the three LeTLP1-overexpression transformants Y55_OE_3, Y55_OE_5, and Y55_OE_10 increased (p < 0.01) 4.5-fold, 4.0-fold, and 6.5-fold, respectively ( Figure 5A). For two LeTLP1overexpression strains (Y55_OE_3 and Y55_OE_10), significant differences in the responses of the LeTLP1-overexpression transformants and the susceptible (Y55) L. edodes strain to the infection with T. atroviride were observed ( Figure 5B). The susceptible (Y55) L. edodes strain was completely covered by the mycelia and spores of T. atroviride, while only a few T. atroviride spores were observed on the surface of the mycelia of the LeTLP1-overexpression transformants. Notably, the infection rates (L T :L L ) of T. atroviride in the mycelia of LeTLP1overexpression transformants were significantly decreased compared to those in the parent isolate Y55 ( Figure 5C). The results demonstrated that the overexpression of LeTLP1 played a crucial role in the resistance of L. edodes to T. atroviride. to those in the parent isolate Y55 ( Figure 5C). The results demonstrated that the overexpression of LeTLP1 played a crucial role in the resistance of L. edodes to T. atroviride. Compared with the parent isolate Y3334, the transcription levels of LeTLP1 in the three silenced transformants TLP1-Ri-3, TLP1-Ri-8, and TLP1-Ri-9 declined by 35%, 43%, and 38%, respectively ( Figure 5D). For two LeTLP1 RNAi strains (TLP1-Ri-3 and TLP1-Ri-8), deep black compounds were generated in the interaction zone between the mycelia of the LeTLP1 RNAi strains and the T. atroviride strains (Figure 5E), and the infection rates (LT:LL) of T. atroviride in the mycelia of L. edodes increased dramatically compared to those in the parent isolate Y3334 ( Figure 5F). The results indicated that the resistance levels significantly declined when LeTLP1 was silenced in L. edodes.
3.6. Antifungal Activity of LeTLP1 against T. atroviride Compared with the parent isolate Y3334, the transcription levels of LeTLP1 in the three silenced transformants TLP1-Ri-3, TLP1-Ri-8, and TLP1-Ri-9 declined by 35%, 43%, and 38%, respectively ( Figure 5D). For two LeTLP1 RNAi strains (TLP1-Ri-3 and TLP1-Ri-8), deep black compounds were generated in the interaction zone between the mycelia of the LeTLP1 RNAi strains and the T. atroviride strains (Figure 5E), and the infection rates (L T :L L ) Life 2021, 11, 863 10 of 14 of T. atroviride in the mycelia of L. edodes increased dramatically compared to those in the parent isolate Y3334 ( Figure 5F). The results indicated that the resistance levels significantly declined when LeTLP1 was silenced in L. edodes.
3.6. Antifungal Activity of LeTLP1 against T. atroviride The filter paper discs that contained the purified LeTLP1 obviously inhibited the mycelial growth of T. atroviride ( Figure 6A,B). On the other hand, LeTLP1 was found to be more effective than LeTLP (Y55) against the mycelial growth of T. atroviride ( Figure 6C). The results demonstrated that LeTLP1 exhibited significant antifungal activity against T. atroviride.

Discussion
The nutritional compounds and pharmacological properties of Lentinula edodes are beneficial to human health [1][2][3][4][5][6][7][8]. Unfortunately, L. edodes is heavily attacked by Trichoderma spp., which compete for limited resources and space for mycelial growth, and it harms the yield and quality of shiitake cultivation [12,16,17,39,40]. In order to protect L. edodes from the infection with T. atroviride, the screenings of resistant L. edodes germplasms and breeding of resistant L. edodes strains are eco-friendly and safe alternatives to the use of chemical fungicides [12,16,18]. Deciphering the molecular mechanisms of the resistance of L. edodes to T. atroviride will provide a theoretical basis for the breeding of resistant L. edodes strains.
In this study, significant differences in the responses of the highly resistant (Y3334) and susceptible (Y55) L. edodes strains to infection with T. atroviride were observed ( Figure  2). A previous study demonstrated that T. atroviride possesses a strong capacity for attacking the mycelia of the susceptible L. edodes strain [12]. The SEM observations ( Figure 2B)  . (A,B) The antifungal activity of LeTLP1 was measured by using the paper disc method described by Imtiaj and Lee (2007) [38]. Agar discs taken from 10 day old cultures of T. atroviride strain 92-1 and were placed in the center of the Petri plates. 1-6: The paper discs of Tris-HCl (pH 8.0), sterilized water, 20 µg of LeTLP (Y55), 5 µg of LeTLP1, 10 µg of LeTLP1, and 20 µg of LeTLP1. (C) The inhibition rates (the percent inhibition of mycelial growth-PIMG) were calculated as described in Imtiaj and Lee (2007) [38].

Discussion
The nutritional compounds and pharmacological properties of Lentinula edodes are beneficial to human health [1][2][3][4][5][6][7][8]. Unfortunately, L. edodes is heavily attacked by Trichoderma spp., which compete for limited resources and space for mycelial growth, and it harms the yield and quality of shiitake cultivation [12,16,17,39,40]. In order to protect L. edodes from the infection with T. atroviride, the screenings of resistant L. edodes germplasms and breeding of resistant L. edodes strains are eco-friendly and safe alternatives to the use of chemical fungicides [12,16,18]. Deciphering the molecular mechanisms of the resistance of L. edodes to T. atroviride will provide a theoretical basis for the breeding of resistant L. edodes strains.
In this study, significant differences in the responses of the highly resistant (Y3334) and susceptible (Y55) L. edodes strains to infection with T. atroviride were observed (Figure 2). A previous study demonstrated that T. atroviride possesses a strong capacity for attacking the mycelia of the susceptible L. edodes strain [12]. The SEM observations ( Figure 2B) showed that the susceptible Y55 strain was heavily infected by T. atroviride, while the highly resistant Y3334 strain grew healthily. Therefore, the highly resistant Y3334 strain and the susceptible Y55 strain were proper materials for studying the molecular basis of the resistance of L. edodes to T. atroviride.
By using a transcriptomic analysis, we identified a novel candidate resistance-related gene, LeTLP1 (LE01Gene05009), which contains a conserved thaumatin domain (Figures 3 and 4B) and belongs to the group of thaumatin-like protein (TLPs). LeTLP1 was strikingly significantly induced when the resistant L. edodes strain Y3334 responded to T. atroviride infection in comparison with the response of susceptible L. edodes, and this was verified by using qRT-PCR detection ( Figure 3B). Similarly, thaumatin-like proteins (TLPs) were shown to be induced in rice plants that were infected with Rhizoctonia solani [41]. LeTLP1 ( Figure S2) shares 16 conserved cysteines (Cys) that are required for eight disulfide bonds, as well as other TLPs, which play important roles in maintaining the structure and activity [42,43]. However, the functions of the novel LeTLP1 in L. edodes and whether it regulates the resistance to T. atroviride are yet unknown.
In plants, TLPs are known as PR-5 proteins, and some of TLPs are known to exhibit both β-1,3-glucan binding [27,44,45] and endo-β-1,3-glucanase activities [46]. β-glucan, the most abundant fungal cell wall polysaccharide in fungi, serves in microbe-associated molecular patterns (MAMPs) [47]. In fungal cell walls, the most abundant β-glucan is β-1,3glucan, which consists of dominant β-glucan content [48]. The TLPs crack the β-1,3-glucan in the cell walls of fungi and exhibit antifungal activity [27,46]. Furthermore, recent studies found that the TLPs also occurred in L. edodes [42,46], and they were highly conserved in plants exhibiting endo-β-1,3-glucanase activity. A previous study indicated that TLPs exhibit antifungal activity by lysing the β-1, 3-glucan in the cell walls of pathogenic fungi. TLPs may participate in regulating the level of resistance of L. edodes relative to invading fungi. In addition, the overexpression of the rice TLP gene in transgenic cassava resulted in enhanced tolerance to Colletotrichum gloeosporioides [49]. Thaumatin-like protein (Pe-TLP) acted as a positive factor in the enhanced resistance to spot disease in transgenic poplars [50]. Therefore, we asked whether the novel LeTLP1 regulates the resistance level in L. edodes. Interestingly, the LeTLP1-overexpression experiment ( Figure 5) demonstrated that the overexpression of LeTLP1 contributes to the enhancement of the level of resistance of L. edodes against T. atroviride. The resistance levels of LeTLP1-silenced transformants (TLP1-Ri-3 and TLP1-Ri-8) declined. Our results demonstrated that the overexpression of LeTLP1 plays a crucial role in the resistance of L. edodes to T. atroviride.
It is noteworthy that a thaumatin-like protein (TLG1) from L. edodes degrades lentinan (a β-1,3-glucan found in its own cell wall) and, thus, may play a role in cell wall degradation [42]. How does L. edodes protect itself against the activity of LeTLP1? Our results showed that the expression level of LeTLP1 was very low without T. atroviride infection ( Figure 3B). Therefore, the production of LeTLP1 is likely to be very limited in mycelia grown in a sawdust culture. This indicates that there is not enough to inflict any harm on itself. However, the LeTLP1 gene expression level was strongly increased in the resistant L. edodes in response to the infection with T. atroviride ( Figure 3B). LeTLP1 might affect the mycelia of L. edodes in response to the infection with T. atroviride. However, the SEM observations ( Figure 2B) confirmed that the highly resistant L. edodes Y3334 strain grew healthily in a dual culture with T. atroviride. Consistently, L. edodes tlg1 was not transcribed in vegetative mycelia grown in a liquid culture or sawdust culture, suggesting that it was induced under stress conditions [42]. Notably, a previous study indicated that the thaumatin-like protein (TLG1) might not act as an antifungal agent [42]. However, our results demonstrated that the LeTLP1 protein exhibited significant antifungal activity against T. atroviride ( Figure 6). Taking the results together, this study demonstrates that the overexpression of LeTLP1 in L. edodes results in resistance to T. atroviride.

Conclusions
We obtained a highly resistant L. edodes strain (Y3334) and a susceptible strain (Y55) and found a novel resistance-related gene, LeTLP1 (LE01Gene05009), in the resistant L. edodes strain (Y3334). Further analysis showed that the LeTLP1 gene expression level strongly increased in the resistant L. edodes in response to the infection with T. atroviride compared with the susceptible L. edodes strain. However, the expression level of LeTLP1 was very low without the T. atroviride infection. More importantly, the experiment with LeTLP1 overexpression and silencing clarified that the expression of LeTLP1 was highly correlated with the level of resistance of L. edodes against T. atroviride. In addition, the LeTLP1 protein (Y3334) exhibited significant antifungal activity against T. atroviride. LeTLP1 was found to be more effective than LeTLP (Y55) against the mycelial growth of T. atroviride. This study demonstrates that resistance to T. atroviride is endowed by the enhanced expression of LeTLP1 in L. edodes, thus providing insights into the molecular mechanisms of the resistance of L. edodes against T. atroviride. Further characterization of the upstream activators and downstream targets of LeTLP1 will be used to decipher the resistance mechanisms in L. edodes and other basidiomycetes.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/life11080863/s1, Figure S1: qRT-PCR results of ten selected genes and the correlations between transcripotome data and real-time PCR results, Figure S2: Alignment of the amino acid sequences of LeTLP1 (Y3334), LeTLP (Y55), and L. edodes TLP (GenBank: AB244759), Table S1: Cultivated and wild strains of Lentinula edodes used in this study, Table S2: Primers used for real-time quantitative PCR.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable. This study did not involve humans.

Data Availability Statement:
The RNA-seq data generated for this study have been deposited into the Biotechnology Information (NCBI) Sequence Read Archive (SRA) under accession number: PRJNA590542.