Ras small GTPase RSR1 represses cellulase and xylanase production
Using the BLAST online website in NCBI, 11 putative Ras subfamily genes were found in the genome of T. reesei (Table 1), including five Ras GTPases, four Ras GEFs, and two Ras GAPs. We constructed 11 putative knockout strains (Additional file 1: Table S1) by homologous recombination in the wild-type strain QM6a to explore the effect of these genes on cellulase production. Two of the Ras GTPases (tre107035 and tre66480) and one Ras GEF (tre70548) could not be knocked out. These three genes may play a decisive role in growth, leading to death by knockout. Knockout of another three genes had a significant effect on enzyme activity (Table 1). Knockout of the TrRas2 gene led to a 60 ± 4.58% decrease in cellulase activity, which is consistent with the results obtained by Zhang et al. [37]. Knockout of tre107369 slightly increased cellulase activity by 10 ± 1.04% compared to the wild-type strain. The rsr1 deletion mutant (Δrsr1 strain) displayed significantly increased (100 ± 8.12%) cellulase activity compared to the wild-type strain QM6a in the presence of Avicel.
Table 1
Effect of knockout of 11 putative Ras subfamily genes on cellulase production
Family | Classification | Gene ID | State | Change | Multiple |
Ras | Ras GTPase | TrRas1 | √ | no change | no |
TrRas2 | √ | down | 60 ± 4.58% |
rsr1 | √ | up | 100 ± 8.12% |
tre107035 | × | unknown | unknown |
tre66480 | × | unknown | unknown |
Ras GEF | tre34726 | √ | no change | no |
tre107369 | √ | up | 10 ± 1.04% |
tre67275 | √ | no change | no |
tre70548 | × | unknown | unknown |
Ras GAP | tre61408 | √ | no change | no |
tre81785 | √ | no change | no |
×, not knocked out; √, successful knockout |
As shown in Additional file 2: Figure S1, when glucose was used as the carbon source, QM6a and Δrsr1 did not produce cellulases. Cellobiohydrolase (pNPCase), endoglucanase (CMCase), filter paper enzyme (FPase), and xylanase activities, and extracellular protein concentration were significantly improved in Δrsr1 compared to the wild-type strain QM6a in the presence of 1% (w/v) Avicel as the carbon source. The pNPCase, CMCase, and FPase activities of Δrsr1 were increased by approximately 100% compared with that of wild-type strain QM6a after 3–4 d of cultivation (Fig. 1A–C). The xylanase activity of Δrsr1 increased by 25–38% compared with that of QM6a (Fig. 1D). The extracellular protein concentration of the Δrsr1 strain was 34–60% higher than that of QM6a after 2 d of cultivation (Fig. 1E). We also found that deletion of rsr1 had almost no effect on T. reesei growth (Additional file 3: Figure S2), indicating that the enhancement of cellulase production was not achieved by controlling biomass production. Additionally, we constructed a complementation strain (RC-rsr1) based on the deletion strain Δrsr1 (Additional file 4: Figure S3). The enzyme activities in RC-rsr1 decreased to the same level as that in the wild-type strain QM6a (Fig. 1A–E). To investigate the effect of rsr1 deletion on cellulase expression at the transcriptional level, we used real-time fluorescent quantitative PCR (RT-qPCR) to detect the major cellulase and xylanase genes in the Δrsr1 strain. As shown in Fig. 2A–F, knockout of rsr1 significantly increased the expression of four major cellulase-encoding genes (cbh1, cbh2, egl1, and egl2) and two major xylanase genes (xyn1 and xyn2). These results are consistent with the increased enzyme activity in the Δrsr1 strain (Fig. 1A–E). The results indicated that rsr1 acts as a repressor in cellulase, xylanase, and secreted protein production.
To study whether rsr1 deletion can enhance cellulase production in the cellulase hyper-producer T. reesei RUT-C30, an industrial mutant [21], we also constructed the rsr1 deletion strain C30-rsr1 from RUT-C30 using the same method. The pNPCase activity of C30- rsr1 increased by > 70% (Fig. 3A) and CMCase activity increased by > 50% (Fig. 3B) compared with that of the parental strain RUT-C30 after cultivation for 3–4 d. These results suggested that rsr1 also has a negative effect on cellulase production in T. reesei mutant, and deletion of rsr1 effectively improves cellulase production in T. reesei. The data indicated the potential of the rsr1 deletion strain for practical application of high-yield cellulase production.
RSR1 negatively regulates the ACY1-cAMP-PKA pathway
The RSR1 domain was predicted using the Pfam database (http://pfam.xfam.org). Only one GTPase catalytic domain was found at position 11–177 (Fig. 4A). Phylogenetic tree analysis (Fig. 4B) showed that RSR1 universally existed in some Ascomycota, including Sordariomycetes, Eurotiomycetes, Leotiomycetes, and Pezizomycetes, with amino acid similarity > 65%. The amino acid similarity of RSR1 homologs in Trichoderma exceeded 80%, and that in Fusarium and Neurospora was 73–85%. Neurospora crassa is also an important filamentous fungus that breaks down lignocellulosic biomass and produces soluble sugars. The amino acid similarity of RSR1 homologs in this species was up to 85%. Although RSR1 is widely found in fungi, none of these putative RSR1 homologs were reported to regulate cellulase gene expression, which is worth studying.
Ras small GTPase often transmits signals through the ACY1-cAMP-PKA signaling pathway [37, 40, 43–45]. Therefore, we detected intracellular cAMP levels in the Δrsr1 strain (Fig. 5A). Intracellular cAMP levels in Δrsr1 were significantly decreased by 79.8% on 2 d of cultivation in the presence of 1% (w/v) Avicel compared to that observed in the original strain QM6a. The intracellular cAMP content in the complementation strain RC-rsr1 was consistent with that of the original strain QM6a (Fig. 5A). However, the intracellular cAMP levels in the Δrsr1 strain were not changed compared with wild-type strain QM6a and complementation strain RC-rsr1 when glucose was used as the carbon source (Fig. 5B).
We also tested the transcription levels of the genes in the ACY1-cAMP-PKA signaling pathway. The transcription levels of the acy1, pkac1, pkac2, and pkar1 genes in the knockout strain Δrsr1 were remarkably reduced compared to the wild-type and complementation strains in the presence of 1% (w/v) Avicel (Fig. 5C–F). The expression levels of ACY1-cAMP-PKA genes in RC-rsr1 recovered to the same levels as those in QM6a. However, their levels stayed consistent in Δrsr1, QM6a, and RC-rsr1 when glucose was used as the carbon source (data not shown). These results indicate that the deletion of rsr1 results in the downregulated expression levels of acy1, pkac1, pkac2, and pkar1 genes and cAMP content at cellulase inducing conditions.
RSR1 represses cellulase production through the ACY1-cAMP-PKA pathway
To address if the decreased acy1 transcriptional level and intracellular cAMP content mediates cellulase enhancement in the rsr1 deletion strain, the deletion and overexpression of acy1 were constructed in QM6a, Δrsr1, named T. reesei Δacy1, Δrsr1Δacy1, QM6a-OEacy1 and Δrsr1-OEacy1, respectively (Additional file 4: Figure S3). As shown in Fig. 6A–B, the further improvement of pNPCase and CMCase activities was observed in Δrsr1Δacy1 compared to Δrsr1 and QM6a. We found that the loss of acy1 remarkably stimulated pNPCase and CMCase activities after 3 d of fermentation in the Δrsr1Δacy1 strain, with an increase of approximately 174% and 128% compared with Δrsr1 and approximately 4.3- and 3.12-fold increase compared with QM6a, respectively. The enhanced cellulase production in rsr1 deletion strain was partly attenuated by the overexpression of acy1 (Fig. 6A, B). The loss of acy1 in the wild-type strain QM6a led to a significant improvement in pNPCase and CMCase activities (Fig. 6A, B), which also implied that acy1 acted as a negative regulator of cellulase expression. Δrsr1Δacy1 showed a higher level of cellulase production capacity compared to Δacy1 (Fig. 6A, B). The results indicated that acy1 expression and intracellular cAMP are involved in RSR1 repressing cellulase production.
To gain insight into how acy1 influences the rsr1-mediated cellulase high-yield process at the transcriptional level, RT-qPCR analysis was performed to detect the expression levels of four main cellulase-encoding genes (cbh1, cbh2, egl1, and egl2) in the QM6a, Δacy1, Δrsr1, Δrsr1Δacy1, QM6a-OEacy1 and Δrsr1-OEacy1 strains. As shown in Fig. 6C–F, and consistent with previous data regarding pNPCase and CMCase activity (Fig. 6A, B), the deletion of acy1 further enhanced the expression of cellulase-encoding genes in Δrsr1Δacy1, compared with that in Δrsr1, while the overexpression of acy1 partly abrogated the overexpression of cellulase-encoding genes in Δrsr1-OEacy1, compared with the case in Δrsr1. These results indicated that the ACY1-cAMP-PKA pathway is required for rsr1 to repress cellulase production. For the first time, we proved that ACY1 acted as a negative regulator of cellulase in the RSR1 signal transduction pathway.
Transcriptional profile of the Δrsr1 mutant cultured in the presence of Avicel
To further understand the role of rsr1 in cellulase production, we compared the two transcriptomes of the deletion strain (Δrsr1) and the original strain QM6a using 1% Avicel as the carbon source. Volcano analysis revealed significantly different expression of 172 genes (Log2fold change ≥ 1 or ≤ − 1 and P adjust < 0.05; Fig. 7A), of which 133 were upregulated and 39 were downregulated (Δrsr1_vs_QM6a). Gene ontology (GO) annotation analysis revealed different expression genes on metabolic processes, binding, catalytic activity, membrane part, cellular process, and other life activities (Fig. 7B). Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis revealed the participation of different expression genes in starch and sucrose metabolism; cyanoamino acid metabolism; valine, leucine, and isoleucine degradation; and protein processing in endoplasmic reticulum pathways (Fig. 7C). This suggested that rsr1 is a widely involved upstream signal that maintains the downstream switch ‘on’ because its deletion affects many pathways. This further suggested that rsr1 influences the entire physiological process, which is very important for cellulase production.
Comparison of cellulase and hemicellulase gene expression levels in Δrsr1 and QM6a (Additional file 5: Table S2) revealed that cellulase and xylanase gene expression levels were upregulated. Of these, endoglucanase (EGL1, EGL3, and EGL5), swollenin, and xylanase xyn4 expression levels were highly upregulated (approximately 4–8 times), consistent with our qPCR data (Fig. 7D). These data indicated that rsr1 was involved in the repression of cellulase expression during induction. We then searched for the top 10 significantly upregulated and downregulated genes (Additional file 6: Table S3). The top 10 upregulated genes included endoglucanase (TRIREDRAFT_122081; egl1) and exo-1,4-beta-xylosidase bxlB (TRIREDRAFT_58450). The expression levels of the 12 transcription factors involved in cellulase expression are listed in Additional file 7: Table S4. The cellulase activator genes xyr1, ace3, and bglr were significantly upregulated by 0.7, 1.73, and 0.29 times, respectively (Table 2), indicating their positive effect on the improvement of cellulase production in Δrsr1. These results suggested that deletion of rsr1 greatly increases expression of cellulase-related genes.
Table 2
Log2 fold change (Log2fc) of characterized transcriptional factors involved in the regulation of lignocellulase genes in QM6a and Δrsr1 strain
Gene ID | Transcription factor genes | Log2fc | Up/Down | Positive/ Negative-acting | P adjust |
77513 | ACE3 | 1.447031547 | up | Positive | 2.07207E-10 |
122208 | XYR1 | 0.766736078 | up | Positive | 0.006403219 |
52368 | BglR | 0.372849514 | up | Positive | 0.003944617 |
76817 | AreA | 0.352378556 | up | Positive | 0.049799388 |
NS: Non-significant at P adjust > 0.05 |
Previous studies have shown that sugar transporters play an important role in the induction of cellulase [46, 47]. Seven of 152 transporter genes (Additional file 8: Table S5) were differentially expressed (P < 0.05) (Table 3) in Δrsr1. The sugar transporter (TRIREDRAFT_48444) and MFS lactose permease (TRIREDRAFT_56684) were significantly upregulated by 2.72 and 2.52 times, respectively.
Table 3
Log2 fold change (Log2fc) of characterized transporters in QM6a and Δrsr1 strain
Gene ID | Description | Log2fc | Up/Down | P adjust |
69957 | Maltose permease | 1.020868586 | up | 4.32084E-05 |
104072 | Xylose transporter | 1.455447348 | up | 3.42349E-05 |
48444 | Sugar transporter | 1.896322777 | up | 2.86476E-11 |
46819 | MFS hexose transporter | 1.549980797 | up | 0.000150214 |
50894 | Sugar transporter | 1.445285702 | up | 0.002895888 |
60945 | MFS permease | 1.302824429 | up | 0.039882249 |
56684 | MFS lactose permease | 1.813960178 | up | 3.81183E-05 |
Connection between RSR1 and GPCRs
GPCRs receive extracellular signals that are ultimately transmitted to the Ras small GTPases [34]. Only three of 58 GPCR genes (Additional file 9: Table S6) showed significantly increased transcriptional levels in the RSR1 deletion strain compared to the wild-type strain (Table 4). The expression of VII GPCR (TRIREDRAFT_53238) was upregulated 1.23-fold, while the expression of the two PTH11-like GPCRs (TRIREDRAFT_58767 and TRIREDRAFT_62462) was upregulated by 1.7 and 1.4 times, respectively.
Table 4
Log2 fold change (Log2fc) of G-protein coupled receptors in QM6a and Δrsr1 strain
Gene ID | GPCR class | Log2fc | P adjust | Up/Down |
53238 | VII (related to rat growth hormone releasing factor) | 1.158695926 | 0.000705892 | up |
58767 | PTH11-like GPCRs | 1.430158452 | 0.00070725 | up |
62462 | PTH11-like GPCRs | 1.26322777 | 1.60043E-10 | up |
To further investigate the role of three transcriptionally elevated GPCRs in rsr1 deletion-mediated cellulase overexpression, double knockout strains (Δrsr1Δtre62462, Δrsr1Δtre58767, Δrsr1Δtre53238) were constructed. Figure 8A and B shows that knockout of tre58767 and tre53238 can lead to an even higher level of cellulase production than that of the parental strain Δrsr1, while the loss of tre62462 has the opposite effect. We then examined the transcript levels of cellulase-encoding genes of all strains by RT-qPCR, consistent with previous enzyme activity data (Fig. 8C–F). The Δrsr1Δtre58767 strain exhibited promotion of the transcript of cbh1 and egl2, while the Δrsr1Δtre53238 strain appeared to promote the expression of cbh1, egl1 and egl2. The expression levels of main cellulase genes in Δrsr1Δtre62462 decreased by > 50% compared to Δrsr1 at 48 h. Moreover, we found that the cellulase synthesis and cellulase gene regulation was not affected by the deletion of tre62462, tre58767, tre53238, respectively (Fig. 8A–F). All results suggested that RSR1 and GPCRs are connected and that they were all involved in a complex signal transmission pathway.