Redesigning transcription factor Cre1 for alleviating carbon catabolite repression in Trichoderma reesei

Carbon catabolite repression (CCR), which is mainly mediated by Cre1 and triggered by glucose, leads to a decrease in cellulase production in Trichoderma reesei. Many studies have focused on modifying Cre1 for alleviating CCR. Based on the homologous alignment of CreA from wild-type Penicillium oxalicum 114–2 (Po-0) and cellulase hyperproducer JUA10-1(Po-1), we constructed a C-terminus substitution strain—Po-2—with decreased transcriptional levels of cellulase and enhanced CCR. Results revealed that the C-terminal domain of CreAPo−1 plays an important role in alleviating CCR. Furthermore, we replaced the C-terminus of Cre1 with that of CreAPo−1 in T. reesei (Tr-0) and generated Tr-1. As a control, the C-terminus of Cre1 was truncated and Tr-2 was generated. The transcriptional profiles of these transformants revealed that the C-terminal chimera greatly improves cellulase transcription in the presence of glucose and thus upregulates cellulase in the presence of glucose and weakens CCR, consistent with truncating the C-terminus of Cre1 in Tr-0. Therefore, we propose constructing a C-terminal chimera as a new strategy to improve cellulase production and alleviate CCR in the presence of glucose.


Introduction
The abundant natural material lignocellulose, which is converted into fermentable sugars or other chemicals through enzymatic hydrolysis, is considerably under-utilized [1,2]. The biotransformation of lignocellulose is significantly hindered by the high cost and the low yield of cellulase [3]. Filamentous fungi, especially Trichoderma reesei, have been widely used to produce cellulolytic and xylanolytic enzymes [4,5,6].
Many studies have focused on the modification of Cre1, such as cre1 knockout or disruption, for alleviating CCR and improving cellulase production [12][13][14][15]. Recently, it was shown that replacement of the natural transcription factor Cre1 with an artificial minimal transcriptional activator such as Cre1-96 or other Cre1 mutants leads to attenuation of CCR, and thus to improvement in cellulase production in T. reesei [14][15][16]. Nakari-Setälä et al. [11] reported that deletion of cre1 increased the quantity of cellulases produced by the wild-type T. reesei QM6a strain. Utilization of fusion transcription factors in T. reesei has been described in recent reports. A fusion protein of Cre1 and Xyr1 resulted in enhanced cellulase production in the presence of glucose and alleviated CCR in Rut-C30 [17]. Therefore, the substantial role of Cre1 in the regulation of cellulolytic enzyme production and carbon metabolism has been reported [18,19]. However, modification of Cre1 to improve cellulase production has not yet achieved favorable results, and further research is warranted.
In the present study, a C-terminal chimera of Cre1 in T. reesei was rationally designed and the transformants were generated. The transcriptional profiles of the transformants were analyzed. The result suggested that the Cre1/CreA chimera alleviates CCR and improves the expression level of cel7a in the presence of glucose. Furthermore, multi site-directed mutagenesis at the C-terminus of Cre1 from T. reesei was performed for mimicking the dephosphorylated state and then it was proved that subcellular localization of mutant with modified Cre1 differed greatly from that with origin Cre1. Thus, we speculated that subcellular localization of Cre1 has an important effect on CCR and we proposed the rational design of transcription factor for mimicking the dephosphorylation is a novel strategy to improve cellulase production and alleviate CCR in the presence of glucose.

Strains and reagents
The T. reesei M2015804 strain was deposited at the China Center for Type Culture Collection (CCTCC). The origin strain P. oxalicum 114-2 was isolated from soil 30 years ago [20], and the cellulase hyperproducer strain JUA10-1, obtained after many rounds of mutagenesis, has been utilized in industrial processes for years [21]. The other strains used in this study are listed in Table 1.
All polymerase chain reaction (PCR) amplifications and fusion PCR were performed using DNA polymerase (Vazyme Biotech, Nanjing, China). Fungal RNA extraction and quantitative reverse transcription (qRT)-PCR experiments were performed using the PrimeScript® RT reagent Kit with gDNA Eraser (Perfect Real Time) (Takara Bio Inc., Shiga, Japan) and FastStart Essential DNA Green Master (Roche, Basel, Switzerland). All other chemicals and materials were purchased from Sinopharm Chemical Reagent Co., Ltd. (Shanghai, China). The KOD-Plus-Mutagenesis Kit was purchased from Toyobo (Osaka, Japan). Primers were synthesized by Personal Bio Biotech Co., Ltd. (Shanghai, China).

Construction of mutants
All CreA/Cre1 mutation replacement cassettes were constructed based on PCR and fusion PCR, which was mediated using the KOD-Plus-Mutagenesis Kit and DNA polymerase. The primers used for the construction of the cassettes are listed in Supplementary Table S1. All the plasmids and linearized expression cassettes in this study were listed in Supplementary Table S2. The transformation of the plasmids and cassettes into P. oxalicum or T. reesei was performed using protoplast transformation based on a protocol published previously [22]. The transformants were selected on plates containing minimal medium (MM) supplemented with 2% glucose and 200 μg/mL hygromycin or 0.3 μg/ml pyrithiamine. As shown in Supplementary Figs. S1-S4, the transformants were verified using PCR and DNA sequencing (not shown) based on our previous methods [22].

Analysis of growth phenotype and transcriptional profile
The method used for culturing fungi for growth phenotype, transcript analysis, and RT-qPCR assays was already described previously [22]. The primers used for qRT-PCR are listed in Supplementary Table S1.

Microscopic observation of hyphal morphology
Microscopic observation of hyphal morphology was performed based on a previously described method [22]. Approximately, 1 × 10 3 spores were inoculated on slides with solidified medium containing PDA, or MM with 2% glucose at 30°C for 48 h. Microscopic images of hyphae were observed using an optical microscope (Nikon Eclipse E100, Japan) at 400x magnification.
Furthermore, the conidia were inoculated to MM salts supplemented with 2% (w/v) glucose and cultivated for 12 h at 26°C. concentration of 10 μg/mL for 20 min, washed and resuspended in 2% (w/v) glucose medium. The preparations were imaged on an inverted fluorescence microscope (Nikon Ti-E, Nikon Corporation, Melville NY).

Investigation of the C-terminus of CreA from P. oxalicum
The amino acid (aa) sequences of CreA between Po-0 and Po-1 were compared using ClustalX2 and then the sequences at the C-terminus of CreA Po−1 greatly differed from those of the original CreA Po−0 (Fig. 1a). It was obvious that a frameshift mutation at site 1205 of nucleotide sequence of CreA Po−0 resulted in the amino acid sequences at the Cterminus. To investigate the effect of this frameshift mutation on the regulation of cellulase expression, the C-terminal sequence SCPQIIP-WRAMIWPIAFKKKSLRLSRLLARGLSGSTLVHTLMSFAFHARYETI of CreA Po−1 extending from aa 401 to aa 453 were substituted with the Cterminal sequence SSTNNSVAGNDLADRF of CreA Po−0 extending from aa 401 to aa 417, thereby generating a new chimera transcription factor named CreA Po−2 . CreA Po−2 was recombined into Po-1, generating the transformant Po-2. Cel7a is a major cellulose-encoding gene, and xlnR is the gene encoding transactivator in P. oxalicum. Subsequently, the expression levels of cel7a and xlnR in the parent and transformant strains were determined using glucose or avicel as the sole carbon source (Fig. 2). No significant changes in the transcriptional levels of xlnR were observed between the transformants and the parent strain grown in either glucose or avicel. In contrast, the transcriptional level of cel7a in Po-2 was sharply decreased using glucose as the sole carbon source, but it was unchanged using avicel as the sole carbon source. It was therefore suggested that the C-terminal frameshift mutation of CreA in Po-1 alleviated CCR, resulting in cellulase hyperproduction.

Rational design of Cre1 in T. reesei
Given the importance of the C-terminus of CreA in alleviating CCR in P. oxalicum, we redesigned the transcription factor Cre1 in T. reesei.
First, the aa sequences of CreA/Cre1 from Po-0 and Po-1 as well as Tr-0 were compared using ClustalX2 (Fig. 1a). We found that the transactivation domain (TAD), NLS, NES, or repression domain in CreA/Cre1 have been reported previously (Fig. 1b). However, to date, no report exists on the function of the C-terminus of CreA/Cre1 from Po-0 or Tr-0 (Fig. 1b). A motif (SSTTGS) from aa 387 to aa 392 at the Cterminus of Cre1 in T. reesei was found to be similar to a motif (SSSTNNS) of CreA in wild-type Po-0 (Fig. 1a). In addition, the motif at the C-terminus of Cre1 Tr−0 (RSSTTGSLAGGDLMDRM) from aa 386 to aa 402 was replaced with that (SCPQIIPWRAMIWPIAFKKKSLRLSRLL-ARGLSGSTLVHTLMSFAFHARYETI) of CreA P°−1 , thus generating a new transcription factor-Cre1 Tr−1 . Subsequently, the C-terminal motif of Cre1 Tr−0 (RSSTTGSLAGGDLMDRM) from aa 386 to aa 402 was deleted as a control, resulting in Cre1 Tr−2 . Cre1 Tr−1 and Cre1 Tr−2 were then introduced into Tr-0 to replace Cre1 Tr−0 , generating the transformants Tr-1 and Tr-2. We found that the transcriptional level of cel7a in Tr-1 was significantly increased by 20-fold than that in Tr-0, which was consistent with that observed in Tr-2; however, the transcriptional level of xyr1 remained unchanged compared with those of the parent strain grown with glucose as the sole carbon source (Fig. 3a). The transcriptional levels of cel7a and xyr1 in Tr-1 presented no significant changes compared with those in Tr-0 grown using avicel as the sole carbon source (Fig. 3b). These results provide strong evidence that modification of the C-terminus by constructing the chimera successfully alleviated CCR and then improved the transcriptional levels of cellulosedegrading enzymes.
Subsequently, the transcription levels of xyr1 and cel7a in the parent and transformant strains were compared using glucose or avicel as a sole carbon source. The xyr1 and cel7a encoded Xyr1 that is a positive transcription factor and CBHI that is a major cellulolytic enzyme, respectively. The transcription level of cel7a in Tr_ Cre1 5M was great higher than that of parent strain with glucose as a sole carbon source. On the contrary, no significant changes in xyr1 and cel7a expression were observed at the transcriptional levels among Tr_ Cre1 5M and the parent strain when avicel was used as a sole carbon source (Fig. 4c). Furthermore, we found that FPase and pNPCase activities, as well as soluble protein from Tr_ Cre1 5M were significantly increased by 1.69-, 1.5-, and 1.62-fold, respectively, compared with the parent strain grown in the media using glucose as a sole carbon source after 6 days culture (Fig. 4d). When avicel was used as a sole carbon source, the only little changes on FPase and pNPCase activities were observed. Our findings proved that five mimicking dephosphorylation sites at the Cterminus of Cre1 result in an improvement of cellulase production in the presence of glucose.
Plasmid pUG-Cre1 5M -GFP and pUG-Cre1-GFP were introduced into the parent strain, generating the transformants named Tr_Cre1-GFP and Tr_Cre1 5M -GFP. Then, these strains were cultured in the medium containing glucose as carbon source and their mycelium were observed with inverted fluorescence microscope for investigating the influence of phosphorylation at the C-terminus of Cre1 on subcellular localization. We found that the local of fluorescence signal of Tr_Cre1 5M -GFP greatly differed from that of Tr_Cre1-GFP (Fig. 5). The fluorescence signals were distributed in the whole mycelium in Tr_Cre1 5M -GFP, on the contrary, fluorescence signal was focused on the center of nuclei of Tr_Cre1-GFP. Thus, this result suggested that phosphorylations at the Cterminus of Cre1 have an important effect on subcellular localization of Cre1 in T. reesei.

Discussion
The CCR effect facilitates the preferential assimilation of energyefficient and readily available carbon sources such as glucose or xylose by inhibiting the expression of enzymes involved in the catabolism of other carbon sources [26,27]. Therefore, many attempts have been made to suppress CCR and/or its effect to improve the expression of cellulase. Nakari-Setälä et al. [11] reported that the deletion of cre1 increased the quantity of cellulases produced by the wild-type T. reesei Fig. 2. Comparison of the transcriptional levels among transformant Po-2 and parent strains Po-1 in P. oxalicum. Relative gene expression of cbh1 and xlnR between Po-1 (blank) and its complemented strain Po-2 (dark gray) grown on culture using (a) glucose or (b) avicel as the sole carbon source. Gene expression levels were normalized (logarithm-2 -ΔΔCT analysis) to that of the reference sample Po-1 using the reference gene βactin. Mean values are presented; error bars indicate standard deviation from three independently grown cultures. Fig. 3. Transcript levels of cel7a and xyr1 in the transformant strains Tr-1 and Tr-2. The transcriptional levels of cel7a and xyr1 between Tr-0 (blank) and its mutated strains Tr-1 (gray), Tr-2 (dark gray) grown in culture using (a) glucose or (b) avicel as the sole carbon source. Gene expression levels were normalized (2 -ΔΔCT analysis) to that of β-actin gene. Mean values are presented; error bars indicate standard deviation from three independently grown cultures.
QM6a strain. However, it was also reported that deletion of cre1 did not result in inceased total cellulase production in T. reesei because of the remarkably slower growth rate and biomass reduction [28]. These results indicate that Cre1 is necessary during cellulase production.
Heinzelman et al. [29] applied SCHEMA (a structure-guided protein recombination route) on three fungal cellobiohydrolase II (CBH-II) enzymes and engineered a series of highly thermostable CBH-II variants. Herein, based on the homologous alignment of CreA from wildtype Po-0 and cellulase hyperproducer Po-1, we substituted the C-terminus of Po-1 with that of Po-0. The transcription level of cel7a growing in glucose showed that the C-terminus replacement efficiently alleviated CCR (Fig. 2a). Our results revealed that the C-terminus of CreA in the hyperproducer Po-1 plays an important role in alleviating CCR. Furthermore, according to structure analysis (Fig. 1b), the C-terminus of CreA Po−1 did not exist in the functional area of CreA. Therefore, we speculate that the regulation of CreA is relatively complex and may be affected by non-functional domains.
Herein, CreA from Po-0 and Po-1 and Cre1 from Tr-0 were compared based on homologous alignment. The C-terminus of Cre1 in Tr-0 was replaced by the C-terminus of CreA from Po-1, generating Tr-1. As a control, the C-terminus of Cre1 was removed, generating Tr-2. The CCR effect of Tr-1 was successfully alleviated, as detected by the transcription level of cel7a growing on glucose, similar to that of Tr-2 (Fig. 3a). These data revealed that the C-terminal chimera Cre1 Tr−1 greatly improves cellulase transcription in the presence of glucose and alleviates CCR in a similar manner, similar to that observed by truncating the Cterminus of Cre1. Thus, we thought that the C-terminus of Cre1/A plays an important role on CCR and cellulase transcription in the presence of glucose.
Nguyen et al. [25] reported that the phosphorylation of this peptide (SSTTGSLAGGDLMDRM) of Cre1 was identified using liquid chromatography-mass spectrometry [25]. Therefore, five potential phosphorylation target sites including S387, S388, T389, T390, and S392 were  simultaneously mutated to valine for eliminating the phosphorylation state of these site. Moreover, it was proved that dephosphorylations at the C-terminus of Cre1 changed subcellular localization of Cre1 in T. reesei and then improved the transcription levels of cellulolytic enzyme and cellulase activities in the presence of glucose (Fig. 5). This result is consistent with previous study that phosphorylation of Cre1 affects its nuclear import and export [30] and plays an important role in cellulase production in T. reesei [31]. Thus, we suggested that phosphorylation of the C-terminal of Cre1 plays an important role on its subcellular localization and has a direct relationship with CCR.

Conclusions
Based on the findings of the present study, we suggest that phosphorylation of the C-terminus of CreA/Cre1 is one of the reasons driving CCR. We demonstrated that constructing a C-terminal chimera of Cre1 in T. reesei improved the transcription levels of cellulase and alleviated CCR in the presence of glucose. Notably, we proved that phosphorylation of the C-terminus of Cre1 plays an important role on its subcellular localization and has a direct relationship with CCR. Taken together, these results demonstrate a new perspectives or strategies for designing Cre1 for allevating CCR and providing a basis for Cre1 phosphorylation-based research involved in CCR.

Declaration of competing interest
Author Shaoli Hou was employed by the company Shandong Henglu Biological Technology Co. Ltd. The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.