Rapamycin at high concentration temporarily inhibits cellulase production
The effect of rapamycin on (hemi)cellulase production of T. reesei RUT-C30 was investigated (Fig. 1). In the presence of 0.01 µM rapamycin, all the tested cellulase activities were not changed at 24 h, but was increased at 72 h and higher than those of RUT-C30 without rapamycin during the rest fermentation time. The pNPGase, pNPCase, CMCase, and FPase activities were improved by 28.7%, 24.0%, 19.2%, and 26.2% separately at 72 h. Nevertheless, when the concentration of rapamycin was increased to 1 µM, the pNPGase, pNPCase, and CMCase activities were reduced at 24 h, but increased at 120 h and beyond. The FPase activity did not reach the same level as that without rapamycin until 168 h. A higher concentration of rapamycin than 1 µM caused a more serious reduction on cellulase activities at the early fermentation process. Especially, the addition of 100 µM rapamycin led to a decline by 67.2%, 60.2%, 56.6%, and 70.3% for pNPGase, pNPCase, CMCase, and FPase activities respectively at 24 h. Interestingly, T. reesei had the ability to recover and produce the same or even a larger amount of cellulase at 168 h, regardless of the concentrations of rapamycin. Obviously, rapamycin improved the cellulase production at low concentration, but at high concentration inhibited severely the celluase production at the early stage which was restored at the late stage. The secreted protein level followed a similar trend as celluase activity. The concentration of secreted protein was halved at 100 µM rapamycin at 24 h, as compared to the untreated sample.
The pNPXase activity did not have obvious change at 0.01 µM rapamycin. At 1 µM rapamycin, the pNPXase activity was decreased at 48 h and 72 h, but was restored to the level of that without rapamycin at 120 h. At 10 µM and 100 µM rapamycin, the pNPXase activity was sharply reduced by 68.0% and 71.9% at 24 h, and remained lower than that without rapamycin at 168 h, suggesting that high concentration of rapamycin decreased hemicellulase activity which was unable to be recovered.
A similar effect of 100 µM rapamycin on the (hemi)cellulase production was found in T. reesei grown on glucose or lactose, except that the pNPCase activity was not changed in the presence of rapamycin when using glucose as the carbon source (Figure S1). Nevertheless, the temporary inhibition effect of rapamycin on the total extracellular protein concentration was not found at the early fermentation stage of T. reesei cultivated on glucose or lactose. In addition, the inhibition effect of rapamycin on the pNPXase activity can be relieved at the late stage, different from that of T. reesei cultured on cellulose. Obviously, the effect of rapamycin on the (hemi)cellulase production in T. reesei was not dependent on the carbon sources, though some subtle differences were observed among different carbon sources.
Rapamycin induces transit morphology defect of T. reesei on cellulose, but not on lactose or glucose.
The effect of rapamycin on the morphology of T. reesei RUT-C30 cultivated in TMM + 2% cellulose, 2% lactose or 2% glucose was investigated by confocal laser scanning microscopy (CLSM) (Fig. 2 and Figure S2). The morphology of T. reesei was not affected by 0.01 µM rapamycin throughout the whole cellulase production on cellulose. When the concentration of rapamycin was increased to 1 µM and beyond, the abnormal hyphal morphology was observed at 24 h. The filamentation of T. reesei was strongly inhibited by rapamycin, showing spherical form which was reminiscent of hypha-yeast transition. However, the morphological defects by 1 µM or 10 µM rapamycin were not found at 72 h and later. Even at 100 µM rapamycin, only a few mycelia exhibited aberrant morphology at 72 h and then disappeared after 72 h. All these findings implicated that rapamycin suppresses the hyphal formation in the early stage of T. reesei grown on cellulose temporarily even at high concentration, which was restored at the late stage. However, when the carbon source was lactose or glucose, there was no notable effect of rapamycin on the mycelium morphology (Figure S2).
Rapamycin hinders cell growth and sporulation of T. reesei tentatively, but not the lipid content
The effect of rapamycin on the growth of T. reesei RUT-C30 on cellulose, lactose or glucose was studied (Fig. 3). As compared to the untreated samples, the cell growth of T. reesei on cellulose, lactose or glucose at 24 h was retarded notably with the treatment of rapamycin at concentrations no less than 1 µM. This cell growth impairment was rescued completely at later stage under cellulose or glucose condition, but not under lactose condition. Meanwhile, the spore amount of T. reesei grown on cellulose, lactose or glucose was also reduced at 24 h, which was recovered at 120 h with no significant change even with 100 µM rapamycin (Fig. 3).
It has been reported that rapamycin treatment led to an increase in the number and size of lipid droplets in the fungus S. cerevisiae (21), Podospora anserine (22, 23), and Ustilago maydis (24). Therefore, to see whether the lipid content of T. reesei was altered by rapamycin, the lipid content of T. reesei grown on cellulose, lactose or glucose was stained by Nile Red and checked under CLSM (Fig. 3 and Figure S3). There was no significant difference in the fluorescent intensity and lipid form between rapamycin-treated RUT-C30 and the untreated one, which was independent on the carbon source, implying that the lipid synthesis in T. reesei RUT-C30 was not affected by rapamycin. Overall, the cell growth and sporulation of T. reesei were reduced by high-dose rapamycin at the early fermentation stage, but not at the late stage. On the contrary, its lipid content was not influenced by rapamycin. T. reesei is highly resistant to rapamycin regardless of carbon source.
Transcription pattern of T. reesei treated with high-dose rapamycin
To gain insight into how rapamycin influences T. reesei RUT-C30 at the transcriptional level, RNA-seq analysis was performed using RUT-C30 cultured in TMM medium with or without 100 µM rapamycin for 24 h. The sequences of the total reads were mapped to the reference genome of T. reesei RUT-C30 (https://www.ncbi.nlm.nih.gov/genome/323%3fgenomeassembly_id%3d49799) with coverage of 97.68–97.76%. A total of 10048 unique transcripts were detected. Genes were considered to be differentially expressed between the two conditions when the average reads of the corresponding transcripts differed with |log2Ratio| ≥ 1 and p value ≤ 0.05. By comparing rapamycin-treated RUT-C30 to the untreated one, 484 differentially expressed genes (DEGs) were obtained, of which 201 were upregulated and 283 were downregulated (Table S1).
The enriched molecular function was mainly related to “catalytic activity” (Fig. 4A), which comprised 69 DEGs. Among them, 50 DEGs show hydrolase activity, of which 30 act on glycosyl bonds. DEGs in “cellulose binding”, “cellulase activity”, “beta-glucosidase activity” and “xylanase activity” categories were all downregulated, which are related to cellulose and hemicellulose degradation. In addition, the category “ATPase activity, coupled to transmembrane movement of substances” included 8 DEGs. Among them, 6 were predicted to be ABC transporters regarding multidrug resistance that were all upregulated to different degrees (Table S2). The increased expressions of these ABC transporters in T. reesei might be a defense mechanism against rapamycin by exporting it out of the cells, which might be worth exploring in future study.
For the enriched cellular components, 37 DEGs were under “extracellular region” category, demonstrating that rapamycin greatly impacts the extracellular enzymes (Fig. 4B). Among them, two cellobiohydrolase (CBH) (CEL7A and CEL6A), seven endoglucanase (EG) (CEL7B, CEL5A, CEL12A, CEL61A, CEL45A, CLE74A, and CEL61B), three β-glucosidase (BGL) (CEL3A, CEL3C and CEL3D), and one β-xylosidase (BXL1) were downregulated. The enriched biological processes were mainly cellulose and hemicellulose catabolic processes, of which all DEGs were downregulated (Fig. 4C). These results were consistent with the remarkably decreased pNPCase, CMCase, pNPGase and pNPXase activities in RUT-C30 treated with 100 µM rapamycin for 24 h, respectively. In addition, the most enriched pathways affected by rapamycin included “Biosynthesis of secondary metabolites”, “Biosynthesis of antibiotic”, “Starch and sucrose metabolism”, “Valine, leucine and isoleucine degradation”, and “Fructose and mannose metabolism” (Fig. 4D).
DEGs involved in the cellulase production were downregulated by rapamycin
Thirty nine DEGs were related to cellulose degradation, of which thirty seven were downregulated and only two were upregulated (Fig. 5 and Table S3). Fifteen cellulases including two cellobiohydrolases (CEL7A and CEL6A), seven endoglucanases except CEL5B, and six β-glucosidases (CEL3A, CEL1A, CEL1B, CEL3C, CEL3H and CEL3D) were notably down-expressed, which agreed with the reduced pNPCase, CMCase, pNPGase and total filter paper FPase activities as found above. The downregulated mRNA levels of BXL1 and xylanases (XYN2, XYN3, XYN4, and XYN6) were in line with decreased pNPXase activity. Moreover, nonenzymatic cellulose attacking enzymes swollenin (25), Cip1 and Cip2 (26), which acted in synergy with cellulases and hemicellulases to enhance the hydrolytic efficiency of cellulose were also downregulated significantly. Unexpectedly, the hemicellulase α-galactosidase AGL1 and β-mannosidase (M419DRAFT_93487), which hydrolyses α-D-galactosides and mannans, were upregulated, though another α-galactosidase (M419DRAFT_71638) and two other β-mannosidases (M419DRAFT_67432 and M419DRAFT_122377) were downregulated noticeably.
In addition, 7 transcriptional factors involved in cellulase production were identified to be DEGs with marked downregulation (Fig. 5D). They were cellulase transcription activators Xyr1 (27), Ace3 (28) and Clr2 (29), MFS sugar transporters Crt1 and Stp1 (30, 31), xylanase promoter binding protein Xpp1 (32), and the carbon catabolite repressor Cre4 (33). Xpp1 was firstly described as a repressor of xylanases (34) and later as a repressor of secondary metabolism (32). The downregulation of Xpp1 did not lead to the increase of xylanase production (Fig. 1), but the expression of 11 DEGs involved in KEGG category “Biosynthesis of secondary metabolites” was found to be increased significantly (Fig. 4D and Table S4), in agreement with the role of Xpp1 as a repressor of secondary metabolism.
FKBP12 is required for the temporary inhibition of rapamycin on cellulase production
Rapamycin forms a gain-of-function complex with FKBP12 first (19), which then inhibits the TOR kinases. By searching T. reesei genome for FKBP12 homologs of S. cerevisiae, we identified three FKBP orthologues including trFKBP12-1 (M419DRAFT_72966), trFKBP12-2 (M419DRAFT_140396), and trFKBP12-3 (M419DRAFT_61673). Among these three FKBP orthologues, only the transcription level of trFKBP12-1 was significantly upregulated by 1.7 fold in T. reesei treated with rapamycin (Table 1), indicating that trFKBP12-1 is probably the cellular receptor of rapamycin. Protein trFKBP12-1 was referred to as trFKBP12 in this study.
Table 1
Transcriptional level of FKBP12 and TOR complexes in T. reesei
Protein IDa
|
log2FC
|
p value
|
Name
|
Protein IDb
|
Identity
|
M419DRAFT_72966
|
1.71
|
0.0035
|
FKBP12-1
|
CAA86890
|
47.32%
|
M419DRAFT_140396
|
-0.20
|
0.0077
|
FKBP12-2
|
|
46.85%
|
M419DRAFT_61673
|
0.02
|
0.7787
|
FKBP12-3
|
|
44.44%
|
M419DRAFT_24714
|
-0.33
|
5.00E − 05
|
TOR1
|
NP_012600.1
|
49.27%
|
TOR2
|
NP_012719.2
|
49.77%
|
M419DRAFT_137337
|
0.11
|
0.2403
|
Lst8
|
NP_014392.3
|
67.00%
|
M419DRAFT_82078
|
-0.24
|
0.0072
|
Kog1
|
NP_012056.1
|
45.92%
|
M419DRAFT_26158
|
0.10
|
0.2521
|
Avo1
|
NP_014563.1
|
24.23%
|
M419DRAFT_128246
|
-0.19
|
0.0194
|
Avo3
|
NP_011018.1
|
26.88%
|
a Protein ID was assigned based on the genome database (https://fungi.ensembl.org/Trichoderma_reesei_rut_c_30_gca_000513815/Info/Index) |
b Protein ID was assigned based on the genome database (https://www.ncbi.nlm.nih.gov/) |
To determine whether trFKBP12 is the cellular receptor of rapamycin in T. reesei, gene trFKBP12 was knockout using T. reesei KU70 as the parent strain (35), obtaining mutant strain ΔtrFKBP12. KU70 was chosen as a host strain for its high efficiency of gene targeting, where ku70 was deleted in RUT-C30 (36). Similar to strain RUT-C30, the inhibition effect of 100 µM rapamycin on cellulase production was observed in strain KU70 at the early fermentation stage (Fig. 6A, 6B, and 6C), which was completely relieved at the late stage (Fig. 6D). This inhibition effect was not found in strain ΔtrFKBP12 during the whole fermentation process (Fig. 6), demonstrating that trFKBP12 was indispensable for the temporary inhibition effect of rapamycin on cellulase production in T. reesei at the early cultivation stage. It is worth noting that knockout of trFKBP12 alone led to a delay in cellulose production, similar to that caused by high-dose rapamycin. No morphology change was found in strain KU70 or ΔtrFKBP12 with the treatment of 100 µM rapamycin (Figure S4).
DEGs involved in the TOR pathway
The TOR kinase is the target of FKBP-rapamycin complex and can interact with multiple proteins to form two complexes TORC1 and TORC2. Both TORC1 and TORC2 play important roles in cell growth and metabolism, but only TORC1 was rapamycin-sensitive (13, 37). In silico analysis revealed there is only one TOR (M419DRAFT_24714) in T. reesei, with 49.27% and 49.77% sequence identity to TOR1 and TOR2 from S. cerevisiae respectively, which was coined trTOR here. The expression of gene trTOR was slightly downregulated by rapamycin (Table 1). The essential components of TORC1 (Lst8 and Kog1), and TORC2 (Avo1, Avo3, and Lst8) were identified in T. reesei, with little change at transcription levels (Table 1). The other components of the TOR complexes (Tco89, Avo2, and Bit61) were not identified in T. reesei. It seems that the effect of rapamycin on the TOR complexes was not very significant in T. reesei. The effort to delete gene trTOR was failed, indicating that trTOR is an essential gene.
Moreover, based on sequence comparisons with the homologues of genes in the TOR signal pathways (13, 38, 39), we identified a series of genes involved in TOR signal pathways in T. reesei RUT-C30, including 15 genes in ribosome biogenesis, 8 genes in cell cycle/growth, 25 genes in nutrient uptake, 1 gene in stress, 5 genes in lipid metabolism, 6 genes in cell wall integrity, and 5 genes in autophagy (Table S5). All these genes were not differentially expressed. These findings matched well with the phenotype profiling results that T. reesei displayed high resistance to rapamycin regardless of carbon sources (Fig. 1, Fig. 2, and Fig. 3).
DEGs related to gene expression