Transcriptomic Analysis Reveals Opposing Roles of CEL1B in Sophorose- and Lactose-Induced Cellulase Expression in Trichoderma reesei Rut C30

Abstract

The β-glucosidase CEL1B has been linked to regulating cellulase expression in Trichoderma reesei, yet its inducer-specific functions and broader regulatory roles remain poorly characterized. In this study, CRISPR-Cas9-mediated gene knockout was applied in the industrial high-producing T. reesei Rut C30 to investigate CEL1B function without the confounding effects of KU70 deletion. Unlike previous studies focused solely on cellulose or lactose induction, transcriptomic analysis of the CEL1B knockout strain revealed its regulatory roles under both lactose- and sophorose-rich conditions, with sophorose representing the most potent natural inducer of cellulase expression. Under lactose induction, CEL1B deletion resulted in a 52.4% increase in cellulase activity (p < 0.05), accompanied by transcriptome-wide upregulation of β-glucosidase genes (CEL3A: 729%, CEL3D: 666.8%, CEL3C: 110.9%), cellulose-sensing receptors (CRT1: 203.0%, CRT2: 105.8%), and key transcription factors (XYR1: 2.7-fold, ACE3: 2.8-fold, VIB1: 2.1-fold). Expression of ER proteostasis genes was significantly upregulated (BIP1: 3.3-fold, HSP70: 6.2-fold), contributing to enhanced enzyme secretion. Conversely, under sophorose induction, CEL1B deletion reduced cellulase activity by 25.7% (p < 0.05), which was associated with transcriptome profiling showing significant downregulation of β-glucosidase CEL3H (66.6%) and cellodextrin transporters (TrireC30_91594: 79.3%, TrireC30_127980: 76.3%), leading to reduced cellobiohydrolase expression (CEL7A: 57.8%, CEL6A: 67.8%). This first transcriptomic characterization of the CEL1B knockout strain reveals its dual opposing roles in modulating cellulase expression in response to lactose versus sophorose, providing new strategies for optimizing inducer-specific enzyme production in T. reesei.

1. Introduction

Amid the depletion of fossil fuels and escalating environmental challenges, the biorefinery industry—which utilizes the abundant lignocellulosic biomass of the Earth to produce renewable energy and bio-based chemicals—has attracted considerable attention. This emerging sector plays a crucial role in promoting global sustainable development and accelerating the transition to low-carbon energy systems [1,2,3]. Lignocellulose, primarily composed of cellulose, hemicellulose, and lignin, is inherently resistant to degradation due to its complex and heterogeneous structure [4,5]. Consequently, efficient cellulases are essential for the effective breakdown of lignocellulosic biomass, but the high cost of these enzymes remains a major bottleneck in industrial bioconversion processes [6]. Trichoderma reesei Rut C30, one of the most widely used industrial cellulase-producing strains, secretes a cellulase system dominated by cellobiohydrolase, endoglucanase, β-glucosidase, and various auxiliary proteins [7,8]. These enzymes act synergistically to convert lignocellulosic materials, such as corn stover, into fermentable glucose [9].

Cellobiohydrolase hydrolyzes cellulose chain ends to release cellobiose; endoglucanase cleaves glycosidic bonds within cellulose’s amorphous regions, thereby reducing its degree of polymerization, and β-glucosidase further converts cellobiose into glucose [9,10,11]. Auxiliary proteins—such as expansins and lytic polysaccharide monooxygenases—enhance enzymatic accessibility by disrupting cellulose crystallinity, loosening fibrillar networks, oxidizing or cleaving cellulose chains to create additional hydrolysis sites, and promoting substrate binding for core enzymes. These mechanisms collectively improve the catalytic efficiency of cellulolytic systems without directly hydrolyzing glycosidic bonds [12,13,14]. However, T. reesei produces these cellulases exclusively under inducible conditions [15,16,17].

Lactose is presently the most cost-effective soluble inducer for cellulase production in submerged liquid fermentation (SLF) [18,19,20]. However, lactose requires hydrolysis and transglycosylation by β-galactosidase or β-glucosidase to generate sophorose, the bioactive compound responsible for cellulase induction [21]. Consequently, lactose is not a direct inducer, which reduces the efficiency of cellulase biosynthesis in SLF. Sophorose (β-1,2-linked disaccharide), identified as the most potent inducer of cellulase synthesis in T. reesei, exhibits an induction capacity over 200-fold higher than lactose. Unfortunately, sophorose’s prohibitive cost precludes its industrial-scale application. Previous studies utilized a biocatalytically prepared glucose–sophorose mixture (MGD) as an inducer for T. reesei Rut C30, achieving cellulase titers of 90.3 FPU/mL [15]. Unlike lactose, direct sophorose application eliminates reliance on β-glucosidase-mediated transglycosylation. Furthermore, β-glucosidase silencing may inhibit sophorose degradation, thereby prolonging cellulase synthesis induction in T. reesei [22].

Previous studies have demonstrated that overexpression of β-glucosidase CEL1B disrupts sugar transport and endoplasmic reticulum homeostasis, thereby inhibiting cellulase synthesis during induction by cellulose or lactose. Conversely, deletion of CEL1B in T. reesei was found to enhance growth rates under lactose induction, suggesting that CEL1B may play a regulatory role in cellulase production [23]. However, its specific function under sophorose induction—the most potent inducer of cellulase synthesis—has not been reported. In our study, knockout of CEL1B resulted in a 25.7% decrease in cellulase activity under glucose–sophorose mixture (MGD) induction, but a 52.4% increase under lactose induction, highlighting its inducer-dependent dual effects on cellulase biosynthesis.

Although previous studies have explored the function of CEL1B in T. reesei KU70 or QM9414 [23,24], several limitations remain unresolved. Notably, functional analyses of CEL1B have been conducted using KU70-deficient strains, despite reports that deletion of KU70 may affect fungal growth [25], metabolism, and gene expression, potentially confounding the interpretation of targeted gene functions. Furthermore, other previous investigations were based on the T. reesei QM9414, which differs significantly from the industrial T. reesei Rut C30 in cellulase production capacity and regulatory background [23]; T. reesei Rut C30 harbors a mutated cre1 gene leading to partial release from carbon catabolite repression and substantially enhanced enzyme secretion [26]. In addition, prior studies have mainly focused on the effects of CEL1B under cellulose or lactose induction [24], while its role under sophorose-rich conditions—the most potent inducer of cellulase production—has not been systematically studied. Moreover, the impact of CEL1B deletion on global gene expression profiles has not yet been explored through transcriptomic analysis, limiting understanding of the underlying regulatory networks. To address these gaps, we employed a CRISPR-Cas9-based strategy to generate a CEL1B gene knockout strain directly in the Rut C30 background without KU70 disruption, and performed comprehensive transcriptomic profiling under different induction conditions. These efforts aim to elucidate the inducer-specific regulatory mechanisms of CEL1B and to provide new insights for optimizing cellulase production in T. reesei.

2. Materials and Methods

2.1. Strains

Trichoderma reesei Rut C30 (NRRL 11460) was obtained from the Agricultural Research Service (ARS) Culture Collection (NRRL; Peoria, IL, USA). The uracil-deficient strain T. reesei Δura5, utilized for CEL1B gene knockout in this study, was derived from T. reesei Rut C30, generated, and maintained in-house [9]. All T. reesei spores were stored in 50% glycerol at −80 °C.

2.2. Construction of CEL1B Gene Knockout T. reesei

The crRNA targeting the CEL1B gene of T. reesei (gene ID: TrireC30_77989) was designed using the CHOPCHOP online tool (http://chopchop.cbu.uib.no/ accessed on 3 June 2025), and the selected target sequence was 5′-CATTAACAAGCACTCCACCG-3′. The sgRNA targeting CEL1B was amplified by PCR using primers CEL1B-F (5′-CATTAACAAGCACTCCACCGGTTTTAGAGCTAGAAATAGC-3′) and CEL1B-R (5′-CACCGAGTCGGTGCTTTTTT-3′) and plasmid pTrgRNA (containing the tracrRNA sequence) as a template. The PCR products were then used as templates for a secondary amplification with primers T7-CEL1B-F (5′-TAATACGACTCACTATAGGGCATTAACAAGCACTCCACCG-3′) and T7-CEL1B-R (5′-CACCGAGTCGGTGCTTTTTT-3′), which included the T7 promoter sequence (5′-TAATACGACTCACTATAGGG-3′) [9]. The resulting DNA was transcribed into RNA using T7 RNA polymerase (Shanghai Biyuntain Biotechnology Co., Ltd., Shanghai, China), purified, and used to knock out CEL1B in T. reesei [27].

Protoplasts were prepared from T. reesei mycelia using the method described by Ogawa et al. [28], with minor modifications. A 500 μL spore suspension of T. reesei Δura5 was added to 50 mL of liquid potato dextrose medium containing 100 μL of uridine. The culture was incubated at 28 °C with agitation at 150 rpm for 12 h to facilitate germination of the spores. Subsequently, mix 3 mg/mL of Lywallzyme (Shanghai Macklin Biochemical Co., Ltd., Shanghai, China) with 2 mg/mL of Yatalase (Baori Doctor Biotechnology Co., Ltd., Beijing, China), and dissolve in 10 mL of OM solution (MgSO₄·7H₂O 73.9 g/L, Na₂HPO₄ 4.5 g/L, NaH₂PO₄ 8.17 g/L, pH = 5.8), and the germinated spores were further incubated for 12 h to digest cell walls. The resulting protoplasts were collected into 1.5 mL centrifuge tubes. Subsequently, 0.6 μg of the Poura5 expression cassette and 0.6 μg of sgRNA were introduced into the protoplast suspension, followed by incubation on ice for 30 min. After that, 600 μL of 60% (w/v) PEG4000 was added to the mixture, which was then left at room temperature for another 30 min. The transformation mixture was evenly plated onto minimal medium (MM) composed of 4 g/L Na₂HPO₄, 1.5 g/L KH₂PO₄, 1 g/L (NH₄)₂SO₄, 0.2 g/L MgSO₄·7H₂O, 0.02 g/L CaCl₂, 1 g/L NaNO₃, 5 mg/L FeSO₄·7H₂O, 1.7 mg/L MnSO₄·H₂O, 1.4 mg/L ZnSO₄·7H₂O, 2 mg/L CoCl₂, 20 g/L agar, and 10 g/L glucose. Plates were incubated at 28 °C for one week to allow colony formation [9]. Genomic DNA was extracted from transformants, and CEL1B knockout in T. reesei was confirmed by PCR and sequencing, following the method described by Ran et al. [21].

2.3. Cellulase Production

The T. reesei in which the CEL1B gene was successfully disrupted was designated as T. reesei ΔCEL1B. Spores from both the parental strain T. reesei Rut C30 and the ΔCEL1B mutant were cultivated on solid potato dextrose agar (PDA) plates at 28 °C for 7 d to promote sporulation. The harvested spores were then transferred to a seed culture medium containing 4 g/L glucose and 10 g/L corn steep liquor, and incubated at 28 °C with shaking at 150 rpm for 24 h. The resulting mycelial biomass was used to inoculate (8% v/v) a fermentation medium with either 10 g/L MGD or lactose as the sole carbon source.

The fermentation medium additionally included 2.8 g/L (NH₄)₂SO₄, 4 g/L KH₂PO₄, 0.6 g/L MgSO₄·7H₂O, 0.8 g/L CaCl₂, 10 mg/L FeSO₄·7H₂O, 1 g/L peptone, 0.3 g/L urea, 3.4 mg/L MnSO₄·H₂O, 2.8 mg/L ZnSO₄·7H₂O, 4 mg/L CoCl₂, and 0.1 M Na₂HPO₄–citrate buffer (pH 5.0). Cultures were maintained at 28 °C under agitation (150 rpm), and samples were collected at specified time points. The supernatant and pellet were separated by centrifugation and subsequently used for cellulase activity measurement and biomass analysis, respectively [24].

The glucose–sophorose mixture (MGD) was synthesized via β-glucosidase-catalyzed transglycosylation. β-Glucosidase was added at 30 CBU per gram of glucose to a substrate containing 600 g/L glucose, and the reaction was conducted at 65 °C (pH 4.8) for 72 h. The mixture was heated to 100 °C for 5 min to inactivate β-glucosidase, yielding MGD. The final composition comprised 410.20 g/L glucose, 60.56 g/L laminaribiose, 9.34 g/L cellobiose, and 13.66 g/L sophorose [15].

2.4. RNA Isolation and High-Throughput RNA-Seq

Total RNA was extracted from T. reesei using a Plant Total RNA Purification Kit (Sangon Biotech, Shanghai, China), with three biological replicates prepared for RNA sequencing. RNA quantification and integrity analysis were performed using an Agilent 2100 Bioanalyzer (Agilent Technologies Co., Ltd., Beijing, China). The Agilent RNA 6000 Nano Kit (Agilent Technologies (Beijing, China) Co., Ltd.) was employed for quality control (QC), assessing RNA concentration, 28S/18S rRNA ratio, RNA Integrity Number (RIN), and fragment length distribution. RNA purity was evaluated using a NanoDrop^TM spectrophotometer (Thermo Fisher Scientific Co., Ltd., Shanghai, China).

RNA sequencing was carried out using the Illumina HiSeq 4000 platform (Shanghai Qingke Technology Co., Ltd., Shanghai, China). Following quality control, the high-quality reads were mapped to the T. reesei RUT C30 genome (https://mycocosm.jgi.doe.gov/cgi-bin/dispTranscript?db=TrireRUTC30_1&id=127115&useCoords=1 accessed on 3 June 2025) using the HISAT2 aligner (https://daehwankimlab.github.io/hisat2/ accessed on 3 June 2025). Subsequent alignment of clean reads to the reference genome was performed with Bowtie2 (http://ccb.jhu.edu/software/stringtie/ accessed on 3 June 2025), and gene expression levels were quantified using RSEM (http://ccb.jhu.edu/software/stringtie/gffcompare.shtml accessed on 3 June 2025). Differential gene expression analysis was conducted with DESeq2 (https://bio.tools/deseq accessed on 3 June 2025), applying a significance threshold of an absolute log₂ fold change ≥1 and an adjusted p-value ≤ 0.05. Within DESeq2, the mean normalized gene count served as the filtering criterion, while statistical significance was assessed via the Wald test. Functional annotation and classification of differentially expressed genes were performed based on Gene Ontology (GO) and the Eukaryotic Orthologous Groups (KOG) system, using data from the T. reesei RUT C30 genome version 1.0.

All RNA-seq data from this study have been submitted to the National Center for Biotechnology Information (NCBI) under BioProject accession number PRJNA1260437.

2.5. Determination of Copy Numbers by qPCR Analysis

Genomic DNA was extracted from T. reesei transformants using a commercial fungal DNA isolation kit (Sangon Biotech, Shanghai, China) and used as the template for quantitative PCR (qPCR). The copy number of the Poura5 expression cassette was determined following the procedure outlined by Ran et al. [21], using the single-copy Poura5 gene as an internal reference.

To assess gene expression in T. reesei Rut C30 and ΔCEL1B, qPCR was performed using the protocol previously described. Primer sequences are provided in Table S1, and the sar1 gene was used as the housekeeping reference for normalization.

2.6. Analytical Methods

The concentration of reducing sugars was measured using the 3,5-dinitrosalicylic acid (DNS) method. In brief, 1.5 mL of the appropriately diluted sample was combined with 2 mL of DNS reagent and heated in boiling water for 5 min. After cooling to room temperature, absorbance was recorded at 540 nm, and sugar concentrations were determined using a glucose standard curve as reference [29].

To evaluate cellulase activity, Whatman No. 1 filter paper was used as the substrate. A 500 μL portion of the enzyme-containing supernatant was mixed with 1 mL of 0.2 mol/L sodium acetate–acetic acid buffer (pH 4.8) and a 1 × 6 cm strip of filter paper. The mixture was incubated at 50 °C for 1 h. The amount of reducing sugars released was quantified using the DNS method. One unit (U) of cellulase activity was defined as the amount of enzyme that releases 1 μmol of reducing sugar per minute per milliliter under the assay conditions [29].

Cellobiohydrolase activity was assessed using p-nitrophenyl-β-D-cellobioside (pNPC) as the substrate, following a previously reported protocol [30]. The assay mixture included 100 μL of diluted enzyme solution and 50 μL of 1 g/L pNPC, incubated at 50 °C for 30 min. The reaction was terminated by adding 150 μL of 10% (w/v) sodium carbonate. The release of p-nitrophenol (pNP) was quantified by measuring absorbance at 415 nm. One unit (U) of cellobiohydrolase activity was defined as the amount of enzyme required to produce 1 μg of pNP per minute under assay conditions.

Endoglucanase activity was determined using carboxymethyl cellulose (CMC) as the substrate. The reaction consisted of 500 μL of appropriately diluted enzyme and 1 mL of 2% (w/v) CMC solution, incubated at 50 °C for 30 min. The amount of reducing sugars released was measured using the DNS method. One unit (U) of endoglucanase activity was defined as the amount of enzyme required to release 1 μmol of reducing sugar per minute under assay conditions [29].

β-Glucosidase activity was measured using cellobiose as the substrate. A total of 200 μL of diluted enzyme was mixed with 200 μL of 15 mmol/L cellobiose solution and incubated at 50 °C for 30 min. The reaction was terminated by boiling, and glucose production was quantified using a commercial glucose biosensor (Sieman Technology Co., Ltd., Shenzhen, China). One unit (U) of β-glucosidase activity was defined as the amount of enzyme that hydrolyzes cellobiose to generate 1 μmol of glucose per minute per milliliter under the assay conditions [29].

Protein concentrations were determined using a bicinchoninic acid (BCA) assay kit (Cat. No.: BL521A, BioSharp, Beijing, China). In a 96-well microplate, 200 μL of working solution (prepared by mixing Reagents A and B in a 50:1 volume ratio) and 20 μL of the diluted enzyme solution were combined. The mixtures were incubated at 60 °C for 30 min, and absorbance was measured at 595 nm using a microplate reader (Beijing Kaiao Technology Co., Ltd., Beijing, China). Bovine serum albumin (BSA) was used to generate the standard calibration curve.

SDS-PAGE was conducted using 10% polyacrylamide gels and culture supernatants collected after 60 h of fermentation, which corresponded to the peak of cellulase production. Samples were prepared by mixing with 5× loading buffer at a 4:1 (v/v) ratio and boiling for 10 min. Electrophoresis was performed at a constant voltage until the tracking dye reached the gel bottom. Gels were stained with Coomassie Brilliant Blue R-250 and destained until protein bands were clearly visible.

All results are presented as means ± standard deviation (SD) from three independent biological replicates. Statistical analysis between groups was conducted using Student’s t-test, with significance considered at p < 0.05 (indicated by *), unless stated otherwise.

3. Results

3.1. Gene Knockout and Phenotypic Analysis of CEL1B in Trichoderma reesei RUT C30

Trichoderma reesei RUT C30 is widely used for industrial cellulase production due to its exceptional cellulase-producing capacity. The T. reesei genome contains 11 β-glucosidase genes, whose transglycosylation activity is crucial for cellulase synthesis induced by lactose or cellulose. Additionally, these genes regulate gene expression in a non-enzymatic manner. Among them, CEL1B encodes a key intracellular β-glucosidase in T. reesei [31]. Overexpression of the CEL1B gene under lactose induction reduces sugar transport capacity, triggers endoplasmic reticulum stress, and decreases cellulase activity by over 97.0% [24,31]. Consequently, this study aimed to knock out the CEL1B gene in RUT C30 to improve cellulase production. Meanwhile, a cost-effective glucose–sophorose soluble inducer was previously developed that efficiently induces cellulase synthesis in T. reesei, yet the role of CEL1B under sophorose induction remains unexplored. In previous study, we established a CRISPR-Cas9 genome editing system in T. reesei RUT C30 and derived an uracil auxotrophic strain, T. reesei ΔTrura5. Using the Poura5 expression cassette from Penicillium oxysporum as a selection marker [32], the cassette was integrated into the TrCEL1B gene, thereby inactivating the target gene while restoring the ura5 gene.

The knockout process of the CEL1B gene in T. reesei RUT C30 is illustrated (Figure 1). The gRNA was strategically positioned 656 bp downstream of the gene’s start codon, guiding the insertion of the Poura5 expression cassette at the target site for gene disruption as shown (Figure 1A). Six potential positive transformants were obtained and designated as T. reesei ΔCEL1B1, ΔCEL1B2, ΔCEL1B3, ΔCEL1B4, ΔCEL1B5, and ΔCEL1B6. PCR validation of these transformants showed that, except for ΔCEL1B3, T. reesei ΔCEL1B1, ΔCEL1B2, ΔCEL1B4, ΔCEL1B5, and ΔCEL1B6 all exhibited successful insertion of Poura5 into the CEL1B gene (Figure 1B). Two well-growing transformants, ΔCEL1B1 and ΔCEL1B4, were selected for further study. Figure 1C show the DNA sequencing results of T. reesei ΔCEL1B1 and ΔCEL1B4, indicating that the Poura5 gene expression cassette was inserted 822 bp downstream of the start codon in ΔCEL1B1, while in ΔCEL1B4, the exogenous DNA was inserted 320 bp downstream of the start codon. As shown in Figure 1D, the growth of T. reesei RUT C30, ΔURA5, ΔCEL1B1, and ΔCEL1B4 was examined on MM medium plates without uridine and PDA medium plates containing 5-fluoroorotic acid (5-FOA) and uridine. The results indicated that T. reesei RUT C30, ΔTrCEL1B1, and ΔTrCEL1B4 exhibited identical phenotypes, growing normally on MM medium without uridine but failing to grow on PDA medium containing 5-FOA. In contrast, T. reesei ΔURA5 was able to grow on PDA plates supplemented with 5-FOA and uridine, confirming the successful restoration of the ura5 gene in ΔCEL1B1 and ΔCEL1B4. Further analysis of the URA5 expression cassette copy number (Figure 1E) revealed that T. reesei RUT C30 had a copy number of 0, whereas both ΔTrCEL1B1 and ΔTrCEL1B4 had a copy number of 1, indicating the absence of off-target effects during the gene knockout process.

3.2. Impact of CEL1B Gene Knockout on Cellulase Production

The synthesis of cellulase in T. reesei requires induction, with lactose being the most commonly used inducer [33]. Studies have shown that lactose must be hydrolyzed and transglycosylated by β-galactosidase or β-glucosidase to generate sophorose, which subsequently induces cellulase production [34]. Sophorose has been identified as the most efficient inducer of cellulase synthesis in T. reesei, exhibiting an induction capacity over 200 times higher than that of lactose. Previous studies demonstrated that a cost-effective glucose–sophorose mixture (MGD), prepared via β-glucosidase-catalyzed conversion of high-concentration glucose, can efficiently induce cellulase synthesis without relying on the transglycosylation activity of β-glucosidase [15]. However, the gene function of β-glucosidase CEL1B under lactose induction, particularly with MGD, remains uncharacterized.

T. reesei RUT C30, ΔTrCEL1B1, and ΔTrCEL1B4 were cultured with 10 g/L MGD or lactose as inducers for cellulase synthesis (Figure 2). Under MGD or lactose induction, cellulase production in all strains peaked at 60 h, with T. reesei ΔTrCEL1B1 and ΔTrCEL1B4 exhibiting identical production trends. When induced by MGD, cellulase and cellobiohydrolase activities in T. reesei ΔCEL1B were reduced by 25.7% and 22.7%, respectively, compared to T. reesei RUT C30 (p < 0.05), while β-glucosidase and endoglucanase activities showed no significant differences. This suggests that CEL1B gene knockout increased cellobiohydrolase production, thereby promoting cellulase synthesis. In contrast, under lactose induction, T. reesei ΔCEL1B exhibited increased activities of cellulase (52.4%), cellobiohydrolase (20.1%), β-glucosidase (34.1%), and endoglucanase (10.0%) compared to T. reesei RUT C30 (p < 0.05), indicating that the conversion of lactose to sophorose does not rely on the transglycosylation activity of CEL1B. In summary, sophorose and lactose exhibit distinct molecular mechanisms in inducing cellulase synthesis in T. reesei, and CEL1B negatively regulates cellulase synthesis in an enzyme activity-independent manner.

To further analyze extracellular protein concentrations, the results are shown in Figure 3. Under MGD induction, there was no significant difference in protein secretion between T. reesei RUT C30 and ΔTrCEL1B (p > 0.05). The extracellular protein concentration of T. reesei ΔTrCEL1B increased by 31.0% compared to T. reesei RUT C30 using lactose as inducer (p < 0.05), which was consistent with the cellulase production trend. Moreover, the results of sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) were in agreement with the extracellular protein concentration measurements.

Although a complemented strain was not constructed, two independent CEL1B knockout mutants were generated and analyzed. Both strains exhibited consistent changes in cellulase activity and gene expression, supporting that the observed effects are attributable to the loss of CEL1B.

3.3. Transcriptome Analysis

Given that T. reesei RUT C30 and ΔCEL1B exhibited opposite cellulase production patterns under lactose and MGD induction, respectively, transcriptomic analysis was performed to investigate the underlying molecular mechanisms. Sequencing was conducted on 12 samples, generating approximately 91.48 Gb of data per sample using the Illumina HiSeq platform. The raw sequencing reads contained low-quality bases, adapter contamination, and a high proportion of unknown bases (N), which were removed before downstream analysis. The quality metrics of the clean reads are presented in Table S2. Subsequently, clean reads were aligned to the reference genome using HISAT. The mapping results were consistent across samples (Table S3), confirming their comparability. Based on gene expression data, box plots (Figure S1A) and Principal Component Analysis (PCA) (Figure S1B) indicated that the normalized expression profiles of the samples were similar. All samples used for transcriptomic analysis were highly correlated (Figure S1C).

Based on gene expression analysis, differentially expressed genes (DEGs) were identified among samples or groups. Under MGD induction, 256 DEGs were identified between T. reesei RUT C30 and ΔCEL1B using DESeq2, with 57 genes upregulated and 199 genes downregulated. Similarly, under lactose induction, 949 DEGs were identified, of which 561 and 388 genes were upregulated and downregulated in RUT C30/ΔCEL1B (Figure 4A). The number of DEGs under lactose induction was 4.5 times higher than that under MGD induction, indicating that CEL1B gene knockout had a greater impact when lactose was used as an inducer. Gene Ontology (GO) enrichment analysis of the 199 and 899 CEL1B-specific DEGs revealed that, regardless of induction by MGD or lactose, the functions of DEGs were primarily related to lignocellulose metabolism (including cellulose, hemicellulose, and lignin) and cellulase activity (Figure 4B) and (Figure S2). However, under MGD induction, all lignocellulose metabolism-related genes were downregulated, explaining the decrease in cellulase production following CEL1B gene knockout. In contrast, under lactose induction, these genes were all upregulated, and CEL1B was also found to influence sugar transport processes.

3.4. GH Family

To comprehensively analyze the transcriptional differences in lignocellulose-degrading enzyme genes of T. reesei induced by MGD or lactose after knockout of the CEL1B gene, we examined the differential expression of all annotated glycoside hydrolase (GH) genes [35]. The average fragments per kilobase of transcript per million mapped reads (FPKM) were calculated for each GH family, and the sum of all FPKM averages within a GH family reflected the overall transcription level (Figure 5, Table S4). When MGD was used as the inducer, the total FPKM mean of all GH families in T. reesei ΔCEL1B decreased by 43.1% compared to T. reesei RUT C30. In contrast, lactose induction resulted in a 204% increase, consistent with cellulase production (Figure 2). Among all GH families, GH1 [36,37], GH5 [37,38,39], GH6 [40] and GH7 [41] accounted for the highest proportion, primarily encoding β-glucosidase, endoglucanase and cellobiohydrolase. Under MGD induction, the transcription levels of these four GH families in T. reesei ΔCEL1B constituted 58.0% of all GH genes, showing a 53.1% decrease compared to T. reesei RUT C30, with GH6 exhibiting the most significant reduction (65.4%), while GH1 decreased by 1.6%, GH5 by 60.9%, and GH7 by 55.9%. Conversely, under lactose induction, the transcription levels of GH1, GH5, GH6, and GH7 in T. reesei ΔCEL1B increased by 275.2% compared to T. reesei RUT C30, with GH1, GH5, GH6, and GH7 rising by 37.1%, 275.7%, 406.2%, and 366.6% respectively. These genes account for 50.0% of all GH genes.

Table 1. Log2-fold change (FC) of the main secreted cellulases expressed in MGD/lactose.

Category	Gene ID	Description/Name	MGD Log2FC	Lac Log2FC
Cellulase	125125	Exoglucanase I/CEL7A	−1.3	2.4
	122470	Exoglucanase II/CEL6A	−1.7	2.6
	5304	Endoglucanase I/CEL7B	−1.6	3.4
	72489	Endoglucanase III/CEL5A	−1.7	2.5
	124438	Endoglucanase II/CEL12A	−1.4	3.3
	74496	Endoglucanase V/CEL45A	NS	NS
	136547	β-glucosidase/CEL3A	−0.8	3.0
	82126	β-glucosidase/CEL3H	−1.6	NS
Auxiliary protein	104220	Swollenin	−1.5	2.3
	139633	Endoglucanase IV (AA9)	NS	2.0
	122518	Endoglucanase VII (AA9)	−1.2	1.8
	121449	Cellulose binding domain/CIP1	−1.5	2.2
	125575	Cellulose binding domain/CIP2	−1.6	2.3

NS, nonsignificant at p > 0.05; Lac: Lactose.

details the major lignocellulose-degrading genes differentially expressed in T. reesei Rut C30 and ΔCEL1B under different inducers, showing expression patterns consistent with cellulase fermentation results. When MGD was used as the inducer, the expression levels of the two cellobiohydrolases (CEL7A [33,42,43] and CEL6A [44,45,46]) in T. reesei ΔCEL1B decreased by 57.8% and 67.8%, respectively. Since cellobiohydrolases play a critical role in cellulose degradation, this may be the primary reason for the reduced cellulase production in T. reesei ΔCEL1B [47,48]. Among endoglucanases, CEL5A [49] and CEL12A [50,51] were downregulated by 69.8% and 61.2%, respectively, while CEL61A [52] and CEL45A [51] showed no significant difference. Similarly, the expression of the major extracellular β-glucosidase CEL3A [9,21,31,48] was not significantly altered, consistent with the β-glucosidase fermentation results. However, CEL3H expression decreased by 66.6%; previous studies reported that deletion of CEL3H impairs sophorose transport, which may explain the reduced cellulase production upon sophorose induction following CEL1B deletion [53]. In addition, auxiliary activity proteins (AA), including AA9 (lytic polysaccharide monooxygenase), swollenin, Cip1 [33,54], and Cip2 [55], were significantly downregulated. AA9 oxidatively disrupts the crystalline structure of cellulose, creating cleavage sites [56]; swollenin physically loosens the hydrogen-bond network of cellulose, increasing substrate accessibility [53]; Cip1 and Cip2, as non-catalytic auxiliaryproteins, enhance enzyme adsorption stability and promote inter-enzyme synergy, respectively, thereby optimizing enzyme–substrate interactions [33,54,55]. Together, these four components form a cooperative “pre-treatment–oxidation–synergism” system that significantly reduces the recalcitrance of cellulose and supports efficient lignocellulose bioconversion. Therefore, when MGD is used as the inducer, the cellulase cocktail produced by T. reesei ΔCEL1B is less effective in hydrolyzing lignocellulosic biomass such as corn stover. In contrast, under lactose induction, most lignocellulose-degrading genes were significantly upregulated in T. reesei ΔCEL1B, consistent with the fermentation results of filter paper activity, pNPCase, CMCase, and β-glucosidase. Notably, CEL3H expression was not significantly affected, suggesting that CEL1B does not directly regulate CEL3H expression and that its regulation is related to the type of inducer used. Whereas the expression of β-glucosidase CEL3A was significantly upregulated by 729.0% when lactose was used as an inducer, our previous study showed that deletion of CEL3A reduced cellulase production by 19.2%, suggesting that the transglycosylation activity of CEL3A is important for lactose induction. This is also an important reason for the increased cellulase production by T. reesei ΔCEL1B during lactose induction.

3.5. Transporters

It is predicted that the T. reesei genome contains 113 sugar transporter proteins, compared to only 43 in Saccharomyces cerevisiae. This is expected, as T. reesei has a much broader range of sugar substrates than S. cerevisiae, which can grow only on a limited number of monosaccharides and disaccharides. The lifestyle of T. reesei requires the utilization of various pentoses, hexoses, and oligosaccharides, primarily derived from lignocellulose degradation [19,57]. Inducers of cellulase production in T. reesei include cellobiose, lactose, sorbose, and sophorose, while glucose inhibits this induction [9,15,33,58]. The mechanism by which T. reesei senses the presence of inducing sugars remains unclear; however, studies suggest that membrane proteins are involved in this process. As shown in Table S5 and Figure 6, 28 DEGs related to transporter proteins were identified under MGD induction, whereas 98 DEGs were identified under lactose induction.

Table 2. Log2-fold change (FC) of characterized transporters in MGD/lactose.

Gene ID	Name	MGD Up/Dwon	Lac Up/Dwon	Description
97259	Hgt1	NS	Up	Glucose transporter
109243	Crt1	NS	Up	Cellulose sensor
136988	Stp1	NS	Up	Cellobiose transporter
138519	Str1	Down	Up	Xylose transporter
33630	Xlt1	Down	Up	Xylose transporter
38765		NS	Up	Xylose transporter
91594		Down	Up	Cellobiose/Xylose transporter
127980		Down	Up	Cellodextrin transporter
55710		NS	Up	Permease of the major facilitator superfamily
137001	Crt2	NS	Up	Cellulose sensor
137795		NS	Up	Cellobiose transporter
79984		NS	Up	Cellobiose transporter
124396	Hxt1	NS	Down	Glucose transporter

NS, nonsignificant at p > 0.05; Lac: Lactose.

summarizes the expression profiles of transporter-encoding genes in T. reesei, including identified sugar transporters and cellulose sensors. Under MGD induction, four sugar transporter genes—TrireC30_91594, TrireC30_127980, TrireC30_33630 (Xlt1) [59], and TrireC30_138519 (Str1) [60,61]—were downregulated by 79.3%, 76.3%, 65.0%, and 52.0%, respectively, among 28 differentially expressed transporters. TrireC30_91594 and TrireC30_127980, predicted to encode cellodextrin transporters, exhibited reduced expression, potentially impairing sophorose transport under MGD and thereby attenuating cellulase gene induction. TrireC30_33630 (encoding the monosaccharide transporter Xlt1) and TrireC30_138519 (Str1) are established monosaccharide transporters. In T. reesei, monosaccharide transport (e.g., glucose and xylose) involves multiple genes, including the high-affinity transporter Hxt1 [62] and low-affinity transporter Stp1 [63]. Consequently, MGD induction minimally impacted monosaccharide transport in the T. reesei ΔCEL1B.

When lactose served as the inducer, 98 differentially expressed genes (DEGs) were linked to transporter-encoding genes, comprising 73 up- and 25 downregulated genes. Among these, 17 were sugar transporter genes (14 upregulated, 3 downregulated), representing 17.3% of transporter-related DEGs. The cellulose sensor Crt1 (TrireC30_109243), a predicted β-disaccharide transporter critical for cellulase induction, exhibited a 203% upregulation in T. reesei ΔCEL1B [64,65,66,67]. Additionally, Crt2 (TrireC30_137001) expression increased by 105.8% [64]. Although Crt2 is not directly involved in β-disaccharide transport, it regulates cellulase gene expression by activating the transcription factor Xyr1 and subsequently Crt1. These elevated expression levels of Crt1 and Crt2 likely contribute significantly to enhanced cellulase production. With the exception of Hxt1 [62], all other mono- and disaccharide transporters were significantly upregulated, indicating that CEL1B deletion enhances cellulase biosynthesis under lactose induction. The divergent expression profiles of sugar transporters between MGD and lactose conditions suggest inducer-specific regulation. This supports a model where CEL1B deletion promotes lactose-to-sophorose conversion, amplifying cellulase induction.

3.6. Transcription Factors

The T. reesei genome contains 758 predicted transcription factor (TF)-encoding genes. Under MGD induction, six TFs exhibited differential expression (three upregulated, three downregulated), while lactose induction altered 46 TFs (26 upregulated, 20 downregulated) (Table S6). The expression profiles of TF-encoding genes regulating lignocellulase expression are summarized in

Table 3. Log2-fold change (FC) of characterized transcription factors involved in the regulation of cellulase genes.

Gene ID	Name	Positive/Negative-Acting	MGD Up/Down	MGD Log2FC	Lac Up/Down	Lac Log2FC
98788	Xyr1	Positive	NS	−0.1	Up	1.7
98455	Ace3	Positive	NS	−0.7	Up	1.8
125610	Vib1	Positive	NS	−0.2	Up	1.3
122363	Ace1	Negative	NS	0.2	NS	−0.6
32395	Ace2	Positive	NS	−0.1	NS	0.5
93466	Hap2	Positive	NS	0.1	NS	−0.1
24298	Hap3	Positive	NS	−0.1	NS	0.2
95791	Pacc	Negative	NS	0.6	NS	−0.2
140814	Area	Positive	NS	0.3	NS	0.1
91236	Bglr	Positive	NS	0.3	NS	0.1
68701	Clr-1	Positive	NS	0.3	NS	0.3
76250	Clr-2	Positive	NS	−0.2	NS	0.7
93861	Ypr1	Positive	NS	−0.5	NS	0.8
112571	Tmac1	Positive	NS	0.2	NS	−0.3
23706	Cre1	Negative	NS	−0.6	NS	0.7
6520	Rce1	Negative	NS	−0.2	NS	0.7

NS, nonsignificant at p > 0.05; Lac: lactose.

. Under MGD induction, no significant changes (p > 0.05) occurred in known cellulase gene regulators. In contrast, lactose induction in T. reesei ΔCEL1B upregulated cellulase-activating TFs Xyr1 (2.7-fold), Ace3 (2.8-fold), and Vib1 (2.1-fold) compared to T. reesei Rut C30, consistent with cellulase production trends. Xyr1 and Ace3 are indispensable for cellulase gene expression, as their deletion severely impairs production and transcription [55,68,69,70,71,72,73], Ace3 [74,75,76,77]. Vib1 further enhances cellulase expression; its overexpression in T. reesei Rut C30 increased cellulase yields and secretion by 200.0% and 219.0%, respectively [76,78,79]. The elevated cellulase production in T. reesei ΔCEL1B under lactose likely stems from TF upregulation. Since lactose requires β-glucosidase-mediated transglycosylation to generate sophorose—the true cellulase inducer—CEL1B is not central to this conversion. Direct use of sophorose-containing MGD reduced only cellobiohydrolase in T. reesei ΔCEL1B, minimally affecting overall cellulase production. Conversely, lactose induction significantly boosted cellulase yields in T. reesei ΔCEL1B, likely due to enhanced sophorose formation and induction efficiency.

As shown in Figure 7, the two most differentially expressed transcription factors (TFs) in T. reesei ΔCEL1B—TrireC30_106452 and TrireC30_133861—showed divergent expression patterns depending on the inducer. Under MGD induction, their mRNA levels were suppressed by 70.7% and 52.0%, respectively, whereas lactose induction markedly upregulated them by 185.6% and 257.1%. Both TFs belong to the Gcn5-related N-acetyltransferase (GNAT) family [80,81], which regulates cellulase gene transcription via histone acetylation. GNAT acetyltransferases, such as Gcn5 [82,83], catalyze lysine acetylation on histone tails (e.g., H3K9/K14) at promoter regions, loosening chromatin structure to activate transcription. The pronounced upregulation of these GNAT members under lactose correlates with enhanced cellulase production, while their repression under MGD aligns with reduced enzyme yields, suggesting GNAT-mediated chromatin remodeling directly modulates cellulase expression. Notably, no other differentially expressed TFs exhibited this regulatory association, highlighting the specificity of GNAT-driven epigenetic control. These findings establish GNAT acetyltransferases as critical regulators of cellulase biosynthesis, with their inducer-dependent expression shifts reflecting metabolic adaptations caused by CEL1B gene deletion, favoring lactose-to-sophorose conversion over direct transcriptional regulation.

3.7. Endoplasmic Reticulum (ER) Protein Processing Pathway

In filamentous fungi, nascent polypeptides translocated into the endoplasmic reticulum undergo coordinated processing events, including signal peptide cleavage, glycosylation, and chaperone-mediated structural maturation [84]. Notably, a comparative analysis demonstrated that T. reesei ΔCEL1B exhibited significantly enhanced secretory capacity under lactose induction compared to the parental T. reesei Rut C30 strain (p < 0.05), whereas both strains showed comparable extracellular protein yields when cultured with MGD as the inducer. These findings underscore the critical challenge of maintaining the fidelity of the secretory pathway, wherein aberrant polypeptides are systematically recognized and targeted for cytoplasmic degradation via the ubiquitin–proteasome system (UPS) [53].

As illustrated in

Table 4. Differential transcription of genes involved in protein processing in the endoplasmic reticulum (ER).

Gene ID	Name	MGD Up/Dwon	Lac Up/Dwon	Description
124246	Fes1	NS	Up	Hsp70 nucleotide exchange factor
121926		NS	Up	Heat shock protein 70 kDa
102206		NS	Up	Heat shock protein 90
122594		NS	Up	DnaJ domain-containing protein
25648	Bip1	Down	Up	Heat shock protein 70 kDa
83445	ERManI	NS	Up	1, 2-α-mannosidase
137683	NEF	NS	Up	Armadillo/beta-catenin-like repeat-containing protein
101105	ERManI	NS	Up	1, 2-α-mannosidase
116265	SPCS1	Down	NS	Microsomal signal peptidase 12 kDa subunit

NS, nonsignificant at p > 0.05; Lac: Lactose.

, transcriptional profiling revealed differential expression patterns of genes involved in protein folding and post-translational processing. Under lactose induction, eight DEGs were identified, all of which were highly expressed. The 6.2-fold upregulation of the Hsp70 nucleotide exchange factor (TrireC30_124246) suggests an enhanced protein quality control capacity, augmenting Hsp70 chaperone activity to manage cellular stress and protein misfolding [85,86]. The observed upregulation of molecular chaperones—including Hsp90 (TrireC30_102206, 2.0-fold) [86], the DnaJ-domain co-chaperone (TrireC30_122594, 1.1-fold), and BiP1 (TrireC30_25648, 3.3-fold) [84,85,87]—suggests a coordinated enhancement of protein quality control mechanisms, likely improving folding efficiency (via Hsp70–Hsp40 collaboration), stabilizing stress-responsive clients (through Hsp90 activity), and reinforcing endoplasmic reticulum homeostasis (via the BiP1-driven unfolded protein response).

The marked upregulation of ERManI (TrireC30_83445: 2.8-fold; TrireC30_101105: 2.1-fold) [87], which encodes 1,2-α-mannosidase critical for N-glycan trimming and glycoprotein quality control in the ER, suggests enhanced processing of misfolded glycoproteins, potentially accelerating ER-associated degradation (ERAD) or maturation pathways. Concurrently, the slight increase in NEF (TrireC30_137683: 1.0-fold), an Armadillo/β-catenin-like repeat-containing protein, may synergize with molecular chaperones to regulate protein folding or substrate release, collectively fine-tuning ER proteostasis under stress conditions [24,33]. The significant downregulation of BiP1 (53.8% reduction) and SPCS1 (TrireC30_116265, 54.7% reduction) [24] in T. reesei ΔCEL1B under MGD induction indicates disruptions in endoplasmic reticulum (ER) proteostasis and secretory pathway efficiency. In summary, the use of lactose as an inducer avoids excessive ER stress, which is advantageous for both T. reesei survival and cellulase production. Notably, the contrasting responses of the ER protein processing pathway under lactose versus MGD induction in the T. reesei ΔCEL1B strain suggest that CEL1B deletion primarily facilitates lactose-to-sophorose conversion rather than directly modulating ER proteostasis.

3.8. RT-qPCR

The transcriptomic data were validated by RT-qPCR analysis of mRNA expression levels for 22 genes in T. reesei Rut C30 and ΔCEL1B (Figure 8). The Log₂ fold change (Log₂FC) values obtained from both transcriptomic analysis and RT-qPCR (averaged from three replicates) showed a strong Pearson correlation (R² = 0.9101), indicating the reliability of the transcriptome data.

4. Conclusions

The knockout of CEL1B in Trichoderma reesei exerts inducer-specific regulatory effects on cellulase biosynthesis. Under lactose induction, CEL1B gene deletion elevated cellulase production by 52.4% (p < 0.05), driven by the upregulation of β-glucosidase genes (Cel3A: 729.0%, Cel3D, and Cel3C), which enhanced lactose-to-sophorose conversion. This activated cellulose sensors Crt1 (203% upregulation) and Crt2 (105.8% upregulation), stimulating transcription factors Xyr1 (2.7-fold), Ace3 (2.8-fold), and Vib1 (2.1-fold) to amplify cellulase gene expression. Concurrently, the upregulation of ER proteostasis genes (BiP1: 3.3-fold, Hsp70: 6.2-fold, Hsp90: 2.0-fold) improved enzyme folding and secretion efficiency. In contrast, under MGD induction, CEL1B knockout reduced cellulase activity by 25.7% (p < 0.05) through the downregulation of β-glucosidase Cel3H (66.6%) and cellodextrin transporters (TrireC30_91594: 79.3%, TrireC30_127980: 76.3%), which impaired sophorose transport and suppressed cellobiohydrolases (CEL7A: 57.8%, CEL6A: 67.8%). These results demonstrate that CEL1B acts as a metabolic switch, repressing cellulase synthesis under sophorose-rich conditions while enhancing lactose-driven induction via transcriptional and secretory pathway optimization. This dual regulatory mechanism provides a foundation for engineering T. reesei with tailored inducer responsiveness for industrial lignocellulose bioconversion.