Engineering of multiple modules to enhance lignocellulose degradation ability in Bacillus subtilis using CRISPR/Cas9 system

Zhiwei Wang; Gongwei Liu; Zhongming Meng; Jie Xu; Haoran Tang; Shendong Wang; Caixia Zou; Caiying Ma; Yuxin Yang

doi:10.1016/j.synbio.2026.01.021

. 2026 Feb 7;13:183–195. doi: 10.1016/j.synbio.2026.01.021

Engineering of multiple modules to enhance lignocellulose degradation ability in Bacillus subtilis using CRISPR/Cas9 system

Zhiwei Wang ^a,^b,¹, Gongwei Liu ^a,^c,¹, Zhongming Meng ^a, Jie Xu ^a, Haoran Tang ^a, Shendong Wang ^b, Caixia Zou ^b, Caiying Ma ^d, Yuxin Yang ^a,^⁎

PMCID: PMC12907642 PMID: 41704465

Abstract

Bacillus subtilis is widely employed for lignocellulose degradation. However, wild-type strains typically exhibit low and incomplete cellulase activities. This study aimed to engineer recombinant B. subtilis strains harboring complete, high-efficiency cellulase systems using CRISPR/Cas9-mediated genome editing. After optimization of signal peptides, transcriptional terminators, and chromosomal integration sites, two endoglucanase expression cassettes were integrated into B. subtilis 168, yielding a strain with a 16.29-fold increase in endoglucanase activity relative to the parental strain. In parallel, a bifunctional cellulase was successfully expressed and optimized, achieving exoglucanase (549.77 U/mL) and β-glucosidase (349.26 U/mL) activities. Moreover, the strain BSKI3Cel, containing three optimized expression cassettes in its genome, exhibited high endoglucanase (129.59 U/mL), exoglucanase (596.75 U/mL), and β-glucosidase (447.42 U/mL) activities. Subsequent transformation of BSKI3Cel with plasmid pJOE2006Bf, carrying genes encoding a composite cellulase system, yielded BSK3P2C, which achieved peak cellulase activities of 533.16, 2959.83, and 2829.61 U/mL on day 8. As a proof of concept, fermentation of wheat straw using BSK3P2C was found to significantly reduce hemicellulose (16.70 %), neutral detergent fiber (7.46 %) and acid detergent fiber (9.93 %) contents. Microscopic analyses confirmed extensive lignocellulose degradation. Overall, this study establishes a high-performance B. subtilis platform with complete cellulase systems for efficient cellulosic biomass conversion.

Keywords: Bacillus subtilis, CRISPR/Cas9, Cellulase, Lignocellulose biodegradation, Expression optimization

Graphical abstract

1. Introduction

Lignocellulosic biomass, with an estimated global annual production of nearly 181.5 billion tons, is primarily composed of cellulose, hemicellulose, and lignin, accounting for approximately 30–60 %, 20–40 %, and 15–24 % of the dry matter content, respectively [1]. In addition, as a low-cost and abundant renewable resource, lignocellulosic biomass is crucial for achieving clean energy goals and reducing reliance on fossil fuels [2,3]. Cellulose, the predominant polysaccharide in plants, is the most abundant biopolymer on Earth [4]. Structurally, cellulose is constituted of linear polysaccharides composed of β-d-glucopyranose units linked by β-1,4-glycosidic bonds, with 20–300 cellulose molecules bundled together via hydrogen bonds and van der Waals interactions to form crystalline structures and certain amorphous regions [5]. Cellulose degradation can be achieved through physical, chemical, and biological methods, with cellulase-mediated biological degradation being the most environmentally sustainable approach [6].

Cellulases, the enzymes responsible for cellulose breakdown, primarily encompass endoglucanases (EC 3.2.1.4), exoglucanases (EC 3.2.1.91), and β-glucosidases (EC 3.2.1.21). Endoglucanases are rate-limiting enzymes that initiate cellulose degradation by randomly hydrolyzing β-1,4-glycosidic bonds in the cellulose backbone to produce cellooligosaccharides. Exoglucanases act on the reducing and non-reducing ends of cellulose chains, releasing large amounts of cellobiose, which β-glucosidases subsequently hydrolyze to glucose, completing the cellulose hydrolytic process [7,8]. Studies have shown that a variety of organisms, including bacteria, fungi, actinomycetes, plants, insects, and higher animals, produce cellulases [9,10]. Currently, microbial cellulases are widely applied in the management of industrial and agricultural waste, as well as in papermaking, textile, food processing, and animal feed industries [10].

Bacillus subtilis, a Gram-positive bacterium generally recognized as a safe (GRAS) probiotic [11], has been widely studied for its ability to metabolize diverse carbon-containing organic compounds, including industrial and agricultural waste, papermaking, food products, and glucose [12]. However, wild-type strains typically exhibit low cellulase activity and lack a complete cellulase system, which limits their applicability in cellulose degradation [13,14]. Despite these limitations, B. subtilis has become a model organism for the industrial production of recombinant proteins, given its non-toxic nature, high secretion capacity, ease of genetic manipulation, and simple cultivation requirements [15].

In recent years, clustered regularly interspaced short palindromic repeat (CRISPR) gene editing technology has been widely applied to B. subtilis [16]. For instance, previous studies have reported the integration of the gene ganA, which encodes β-galactosidase, into different chromosomal loci of B. subtilis using the CRISPR/Cas9 system, resulting in stable β-galactosidase expression [17]. Base editor-targeted and template-free expression regulation, a CRISPR-based technology, has allowed direct chromosomal mutations without the need for DNA library construction or synthesis, thus supporting efficient production of compounds such as lycopene [18]. Despite advancements in CRISPR-based genetic tools, only few studies have addressed the integration and optimization of multi-component cellulase systems in B. subtilis using CRISPR gene editing technology.

The present study aimed to integrate and efficiently express cellulase genes within the B. subtilis genome to construct a recombinant strain that co-expresses a complete and highly active cellulase system with strong cellulosic biomass-degrading capability. To this end, a gene exhibiting both endoglucanase and xylanase activities, together with a gene exhibiting both exoglucanase and β-glucosidase activities, was successfully amplified. Corresponding expression cassettes and CRISPR/Cas9 genome-editing vectors were systematically designed and constructed for B. subtilis. Moreover, cellulase expression in B. subtilis 168 was optimized through the rational selection of signal peptides, transcriptional terminators, and chromosomal integration sites. The resulting recombinant strain, capable of secreting a high-yield, multicomponent cellulase complex, was subsequently evaluated for its efficiency in lignocellulosic biomass degradation. This engineered strain shows potential for application in lignocellulosic biomass hydrolysis and may contribute to improved efficiency in industrial and agricultural processes.

2. Materials and methods

2.1. Strains, plasmids, culture conditions, and reagents

Primers used in this study were synthesized by Sangon Biotech, Shanghai, China. Primers sequences, constructed vectors, and recombinant strains are listed in Table S1 and Table 1. Escherichia coli competent cells (AngYuBio, Shanghai, China) and B. subtilis 168 (preserved in our laboratory) were cultured in Luria broth (LB) liquid medium (Haibo Biotech, Qingdao, China) or LB agar plates (Haibo Biotech) at 37 °C and 220 rpm for 16–20 h. For cellulase production, B. subtilis was cultured in LB liquid medium containing 1 % sodium carboxymethyl cellulose (CMC-Na) under identical cultivation conditions for 24 h. The bacterial culture broth was centrifuged at 9000 rpm for 3 min, and the supernatant was collected for enzyme activity assay. The kanamycin concentration used was 50 μg/mL, and d-mannose was added at 0.2 %.

Table 1.

Strains used in this study.

Strains	characteristic	Source
E. coli JM109	endA1, recA1, gyrA96, thi-1, hsdR17 (rk-, mk+), relA1, supE44D (lac-proAB) [F′ traD36, proAB, laqIqZ ΔM15]	Purchased from AngYuBio, Shanghai, China
E. coli JM110	rpsL(StrR), thr, leu, thi-1, lacY, galK, galT, ara, tonA, tsx, dam, dcm, supE44, Δ(lac-proAB)/F'[traD36, proAB, lacIqlacZΔM15]	Purchased from AngYuBio, Shanghai, China
B. subtilis 168	trpC2	Laboratory preservation
B. subtilis RLI2019	Produce highly active cellulase	Laboratory preservation
BSKI2006	B. subtilis 168 derivate, △egls::G2006 cassette	This study
BSsprP43PhoD2006	B. subtilis 168 derivate, △sprE::P₄₃-PhoD-G2006	This study
BSsprP43PhoB2006	B. subtilis 168 derivate, △sprE::P₄₃-PhoB-G2006	This study
BSsprP₄₃Ywb2006	B. subtilis 168 derivate, △sprE::P₄₃-YwbN-G2006	This study
BSsprP₄₃Lip2006	B. subtilis 168 derivate, △sprE::P₄₃-LipA-G2006	This study
BSsprP₄₃BglC2006	B. subtilis 168 derivate, △sprE::P₄₃-BglC-G2006	This study
BSlacP₄₃BglC2006	B. subtilis 168 derivate, △lacZ::P₄₃-BglC-G2006	This study
BSthrP₄₃BglC2006	B. subtilis 168 derivate, △thrC::P₄₃-BglC-G2006	This study
BSnprP₄₃BglC2006	B. subtilis 168 derivate, △nprE::P₄₃-BglC-G2006	This study
BSsprP₄₃BglC2006TB9	B. subtilis 168 derivate, △sprE::P₄₃-BglC-G2006-TB9	This study
BSsprP₄₃BglC2006TH1	B. subtilis 168 derivate, △sprE::P₄₃-BglC-G2006-TH1	This study
BSsprP₄₃BglC2006TB5	B. subtilis 168 derivate, △sprE::P₄₃-BglC-G2006-TB5	This study
BSsprP₄₃BglC2006TB4	B. subtilis 168 derivate, △sprE::P₄₃-BglC-G2006-TB4	This study
BSsprP₁₂BglC2006TB4	B. subtilis 168 derivate, △sprE::P₁₂-BglC-G2006-TB4	This study
BSpP₁₂BglC2006TB4	B. subtilis 168 derivate, containing plasmid pJOEP₁₂BglC2006TB4	This study
BSsprEgls2006	BSsprP₁₂BglC2006TB4 derivate, △egls::G2006 cassette	This study
BSsprlacBglC2006	BSsprP₁₂BglC2006TB4 derivate, △lacZ::P₁₂-BglC-G2006	This study
BSsprlacPhoD2006	BSsprP₁₂BglC2006TB4 derivate, △lacZ::P₁₂-PhoD-G2006	This study
BSsprlacYwb2006	BSsprP₁₂BglC2006TB4 derivate, △lacZ::P₁₂-YwbN-G2006	This study
BSsprlacPhoB2006	BSsprP₁₂BglC2006TB4 derivate, △lacZ::P₁₂-PhoB-G2006	This study
BSsprlacLip2006	BSsprP₁₂BglC2006TB4 derivate, △lacZ::P₁₂-LipA-G2006	This study
BSnprP₄₃LipBfTB9	B. subtilis 168 derivate, △nprE::P₄₃-LipA-Bf1-TB9	This study
BSnprP₄₃LipBfTB5	B. subtilis 168 derivate, △nprE::P₄₃-LipA-Bf1-TB5	This study
BSnprP₄₃LipBfTH1	B. subtilis 168 derivate, △nprE::P₄₃-LipA-Bf1-TH1	This study
BSnprP₄₃LipBfTB4	B. subtilis 168 derivate, △nprE::P₄₃-LipA-Bf1-TB4	This study
BSlacP₄₃LipBfTH1	B. subtilis 168 derivate, △LacZ::P₄₃-LipA-Bf1-TH1	This study
BSthrP₄₃LipBfTH1	B. subtilis 168 derivate, △thrC::P₄₃-LipA-Bf1-TH1	This study
BSsprP₄₃LipBfTH1	B. subtilis 168 derivate, △sprE::P₄₃-LipA-Bf1-TH1	This study
BSthrP₄₃BglCBfTH1	B. subtilis 168 derivate, △thrC::P₄₃-BglC-Bf1-TH1	This study
BSthrP₄₃YwbBfTH1	B. subtilis 168 derivate, △thrC::P₄₃-YwbN-Bf1-TH1	This study
BSthrP₄₃PhoDBfTH1	B. subtilis 168 derivate, △thrC::P₄₃-PhoD-Bf1-TH1	This study
BSthrP₄₃PhoBBfTH1	B. subtilis 168 derivate, △thrC::P₄₃-PhoB-Bf1-TH1	This study
BSthrP₁₂YwbBfTH1	B. subtilis 168 derivate, △thrC::P₁₂-YwbN-Bf1-TH1	This study
BSpP₄₃YwbBfTH1	B. subtilis 168 derivate, containing plasmid pJOEP₄₃YwbBfTH1	This study
BSthrlacLipBf	BSthrP₄₃YwbBfTH1 derivate, △lacZ::P₄₃-LipA-Bf1-TH1	This study
BSthrlacPhoBBf	BSthrP₄₃YwbBfTH1 168 derivate, △lacZ::P₄₃-PhoB-Bf1-TH1	This study
BSthrlacPhoDBf	BSthrP₄₃YwbBfTH1 derivate, △lacZ::P₄₃-PhoD-Bf1-TH1	This study
BSp2006Bf	B. subtilis 168 derivate, containing plasmid pJOE2006Bf	This study
BSspr2006thrBf	BSsprP₁₂BglC2006TB4 derivate, △thrC::P₄₃-YwbN-Bf1-TH1	This study
BSKI3Cel	BSspr2006thrBf derivate, △lacZ::P₁₂-LipA-G2006	This study
BSK3P2C	BSKI3Cel derivate, containing plasmid pJOE2006Bf	This study

Open in a new tab

For vector construction, target fragments were amplified with high-fidelity 2 × Phanta Max Master Mix (P515; Vazyme, Nanjing, China) and vectors were confirmed by amplifying target fragments using 2 × Rapid Taq Master Mix (P222; Vazyme). The Hieff Clone Plus Multi One Step cloning kit (10912 ES; Yeasen Biotech, Shanghai, China) was used to ligate one or more target genes to the vector backbone. Point mutations were introduced using the Mut Express II Fast Mutagenesis kit V2 (C214; Vazyme). Plasmid extraction and gel recovery were performed using the SanPrep Column Plasmid Mini-Preps kit (B518191) and the SanPrep Column DNA Gel extraction kit (B518131) from Sangon Biotech, respectively. Final constructed vectors were verified via Sanger sequencing by Sangon Biotech.

2.2. Construction of gene expression vectors and recombinant strains

To insert the endoglucanase-encoding gene 2006 (G2006) from B. subtilis RLI2019 into the eglS locus of B. subtilis 168, coding sequences of endoglucanase genes from B. subtilis RLI2019 (CP123621.1) and B. subtilis 168 (CP136402.1) were retrieved and compared. A suitable single nucleotide polymorphism site in the eglS gene of B. subtilis 168 was identified, and primer pair sgEglsF/sgEglsR were designed to synthesize small-guide RNA (sgRNA). The double-stranded DNA with sticky ends was annealed and ligated to the pJOE8999 vector backbone using the golden gate assembly method (#R3733; NEB, Massachusetts, USA) to construct the JOEsgeglS vector [19]. The G2006 expression cassette and the JOEsgeglS vector backbone were amplified and ligated to construct the JOEKI2006 vector (Fig. 1A). The method of integrating gene expression cassette into the B. subtilis genome was conducted based on the approach proposed by Altenbuchner et al. (2016). The recombinant strain underwent ten serial passages at 37 °C and 220 rpm, with each passage cultured for 24 h, to assess genetic stability. The same protocol was applied for editing subsequent strains.

Fig. 1 — Schematic diagram of the CRISPR/Cas9 gene editing vector JOEKI2006 (A), the strain nomenclature (B), and the gene expression cassette and integration site (C).

2.3. Expression and optimization of endoglucanase

sgRNA primers targeting the sprE, nprE, thrC, and lacZ loci in B. subtilis 168 were designed to construct vectors JOEsgspr, JOEsgnpr, JOEsgthr, and JOEsglac, respectively. The fragments, including the upstream homologous arm of the sprE locus, P₄₃ promoter, signal peptide gene bglC and endoglucanase-encoding G2006, downstream homologous arm of the sprE locus, and JOEsgspr vector backbone, contained an overlap, were amplified and sequentially ligated to construct the JOEsprP₄₃BglC2006 vector. The edited strain was designated BSsprP₄₃BglC2006. The strain nomenclature rules and the genetic component of the expression cassettes are shown in Fig. 1B and 1C, respectively. The bglC signal peptide gene of the vector JOEsprP₄₃BglC2006 was replaced with native signal peptide genes phoB, phoD, ywbN, and lipA from B. subtilis 168 to generate vectors JOEsprP₄₃PhoB2006, JOEsprP₄₃PhoD2006, JOEsprP₄₃Ywb2006, and JOEsprP₄₃Lip2006, respectively. The resulting edited strains were designated BSsprP₄₃PhoB2006, BSsprP₄₃PhoD2006, BSsprP₄₃Ywb2006, and BSsprP₄₃Lip2006.

Similar methods were employed to construct vectors targeting the lacZ, thrC, and nprE loci, resulting in JOElacP₄₃BglC2006, JOEthrP₄₃BglC2006, and JOEnprP₄₃BglC2006. These vectors produced the edited strains BSlacP₄₃BglC2006, BSthrP₄₃BglC2006, and BSnprP₄₃BglC2006, respectively. To introduce transcriptional terminator sequences (TB9, TB4, TB5, and TH1), point mutations were added to the end of the endogluconase gene in JOEsprP₄₃BglC2006, creating JOEsprP₄₃BglC2006TB9, JOEsprP₄₃BglC2006TB4, JOEsprP₄₃BglC2006TB5, and JOEsprP₄₃BglC2006TH1, respectively. The resulting genetically edited strains were named BSsprP₄₃BglC2006TB9, BSsprP₄₃BglC2006TB4, BSsprP₄₃BglC2006TB5, and BSsprP₄₃BglC2006TH1, respectively. The 5′-untranslated region (UTR) sequence on the P₄₃ promoter was mutated to UTR-12 to construct the vector JOEsprP₁₂BglC2006TB4, resulting in the genetically edited strain BSsprP₁₂BglC2006TB4 [20].

The expression cassette P₁₂-bglC-G2006-TB4 was then combined with the pJOE8999 vector backbone (excluding the gene editing element) to construct the plasmid expression vector pJOEP₁₂BglC2006TB4, which was subsequently transformed into B. subtilis 168, generating the strain BSpP₁₂BglC2006TB4.

Additionally, the expression cassette P₁₂-bglC-G2006-TB4 was linked to the JOElacP₄₃BglC2006 vector backbone to construct the vector JOElacP₁₂BglC2006TB4. The bglC gene of the vector JOElacP₁₂BglC2006TB4 was replaced with signal peptide genes phoB, phoD, ywbN, and lipA, resulting in the construction of vectors JOElacP₁₂PhoB2006TB4, JOElacP₁₂PhoD2006TB4, JOElacP12Ywb2006TB4, and JOElacP12Lip2006TB4. Each of these five constructed vectors, along with JOEKI2006, was expressed in the strain BSsprP₁₂BglC2006TB4 to produce dual-site gene-edited strains BSsprlacBglC2006, BSsprlacPhoB2006, BSsprlacPhoD2006, BSsprlacYwb2006, BSsprlacLip2006, and BSspreglS2006, respectively.

2.4. Expression and optimization of bifunctional cellulase

Based on the genome of B. subtilis 168, codons of the bifunctional cellulase Bf1 (WP_011011185.1; Table S2) were optimized by Tsingke Biotech (Beijing, China). The synthesized gene was cloned into the pET-28a (+) vector to create the vector pET-Bf. The upstream homologous arm targeting the nprE site, P₄₃ promoter, lipA signal peptide, cellulase-encoding gene Bf1, downstream homologous arm targeting the nprE site, and JOEsgnpr vector backbone were amplified and assembled to construct the vector JOEnprP₄₃LipBf, with the resulting edited strain named BSnprP₄₃LipBf.

Point mutations introduced terminators TB9, TB5, TH1, and TB4 at the end of cellulase-encoding gene Bf1 in plasmid JOEnprP₄₃LipBf, yielded vectors JOEnprP₄₃LipBfTB9, JOEnprP₄₃LipBfTB5, JOEnprP₄₃LipBfTH1, and JOEnprP₄₃LipBfTB4, and the respective strains were designated BSnprP₄₃LipBfTB9, BSnprP₄₃LipBfTB5, BSnprP₄₃LipBfTH1, and BSnprP₄₃LipBfTB4. The upstream homologous arm targeting the lacZ site, P₄₃-lipA-Bf1-TH1 expression cassette, downstream homologous arm targeting the lacZ site, and JOEsglac vector backbone were amplified and assembled to construct the vector JOElacP₄₃LipBfTH1. Following successful gene editing, the resulting strain was named BSlacP₄₃LipBfTH1. Similarly, recombinant strains BSthrP₄₃LipBfTH1 and BSsprP₄₃LipBfTH1 were constructed using similar methods.

The signal peptide gene lipA of the vector JOEthrP₄₃LipBfTH1 was substituted with signal peptide genes bglC, ywbN, phoD, and phoB, resulting in vectors JOEthrP₄₃BglCBfTH1, JOEthrP₄₃YwbBfTH1, JOEthrP₄₃PhoDBfTH1, and JOEthrP₄₃PhoBBfTH1, and the corresponding edited strains BSthrP₄₃BglCBfTH1, BSthrP₄₃YwbBfTH1, BSthrP₄₃PhoDBfTH1, and BSthrP₄₃PhoBBfTH1. The 5′-UTR sequence of the P₄₃ promoter of the vector JOEthrP₄₃YwbBfTH1 was mutated to UTR-12 to yield JOEthrP₁₂YwbBfTH1, with the edited strain termed BSthrP₁₂YwbBfTH1.

The signal peptide gene lipA of the vector JOElacP₄₃LipBfTH1 was replaced with the signal peptides genes phoB and phoD to generate vectors JOElacP₄₃PhoBfTH1 and JOElacP₄₃PhoDBfTH1. Vectors JOElacP₄₃LipBfTH1, JOElacP₄₃PhoBfTH1, and JOElacP₄₃PhoDBfTH1 were expressed individually in the strain BSthrYwbBfTH1, and the successfully edited strains were designated BSthrlacLipBf, BSthrlacPhoBBf, and BSthrlacPhoDBf, respectively.

To further optimize, the P₄₃-ywbN-Bf1-TH1 expression cassette was amplified and cloned into the pJOE8999 backbone, resulting in pJOEP₄₃YwbBfTH1, and was transformed into B. subtilis 168 to result in the strain BSpP₄₃YwbBfTH1.

2.5. Co-expression and optimization of complex cellulase

Expression cassettes P₁₂-bglC-G2006-TB4, P₄₃-ywbN-Bf1-TH1, and the pJOE8999 vector backbone (without genetic editing elements) were individually amplified and assembled to construct the dual cellulase expression and secretion vector pJOE2006Bf, thus harboring two cellulase genes. The recombinant plasmid was transformed into B. subtilis 168, generating the strain BSp2006Bf. In addition, JOEthrP₄₃YwbBfTH1 was expressed in the strain BSsprP₁₂BglC2006TB4, integrating the P₄₃-ywbN-Bf1-TH1 expression cassette into the thrC locus, resulting in strain BSspr2006thrBf. Furthermore, the expression of the vector JOElacP₁₂Lip2006TB4 in the strain BSspr2006thrBf integrated the P₁₂-lipA-G2006-TB4 expression cassette into the lacZ locus, thus generating the strain BSKI3Cel. Finally, transforming BSKI3Cel with the plasmid pJOE2006Bf produced the strain BSK3P2C. Strains BSKI3Cel and BSK3P2C were cultivated in cellulase production medium, and the activity of the three types of cellulase investigated herein was determined at 24-h intervals.

2.6. Determination of bacterial growth curve

Single colonies of selected recombinant strains were cultured in LB medium at 37 °C with shaking at 220 rpm for 10–12 h. Cultures were grown to an OD₆₀₀ of 0.7–0.8 and used as seed cultures. The seed cultures were subsequently inoculated at a 1.0 % (v/v) ratio into 96-well plates containing 198 μL of fresh liquid LB medium. Bacterial growth curves were analyzed using an automatic microbial growth curve analyzer (MGC-200, SCIENTZ Biotech, Ningbo, China). Each sample was tested in triplicate. The detection parameters were set as follows: un-inoculated LB medium was used as a blank control for calibration. The cultivation temperature was set at 37 °C with a shaking speed of 750 rpm. The OD₆₀₀ value was measured every 15 min over a total duration of 48 h.

2.7. Determination of enzyme activity and total protein concentration

Endoglucanase activity was determined using 1 % (w/v) CMC-Na as the substrate and quantified using the 3,5-dinitrosalicylic acid method [21]. One unit of enzyme activity was defined as the amount of enzyme required to release 1 μg of reducing sugar per minute from 1 % CMC-Na at 50 °C. Exoglucanase and β-glucosidase activities were determined using 1 mg/mL 4-nitrophenyl-β-d-cellobioside (pNPC, CAS#3482-57-3, Yuanye Bio, Shanghai, China) and 5 mM 4-nitrophenyl-β-d-glucopyranoside (pNPG, CAS#2492-87-7, Yuanye Bio) as substrates, respectively. Reaction mixtures containing the substrate and crude enzyme solution were incubated at 50 °C for 30 min, after which the reactions were terminated by the addition of 1 mol/L Na₂CO₃. Absorbance was measured at 410 nm. One unit of exoglucanase or β-glucosidase activity was defined as the amount of enzyme required to produce 1 μg of p-nitrophenol per minute from pNPC or pNPG, respectively, at 50 °C.

Hemicellulase (xylanase) activity was measured using a hemicellulase activity assay kit (ADS-W-TDX059; Aidisheng Biotechnology, Yancheng, China). One unit of activity was defined as the amount of enzyme required to produce 1 nmol of xylose per minute per milliliter of supernatant during xylan hydrolysis. Total protein concentration and specific enzyme activity were determined according to a previously described method [22].

2.8. Analysis of wheat straw enzymatic digestion

Bacterial suspensions were obtained after incubation for 24 h to reach a concentration of approximately 3.0 × 10⁸ CFU/mL. Bacterial cells were inoculated at a ratio of 5 % into the substrate mixture which contained 30 g of sterile, dried and sieved wheat straw (particle size <2 mm) and 112.5 mL of infiltration culture medium [23]. The material was transferred to a 250-mL conical flask, sealed with breathable film, and incubated under static conditions at 37 °C and approximately 80 % relative humidity. Fermentation using strain BSK3P2C was terminated after 0, 4, and 8 days. Materials from each conical flask were thoroughly mixed and sampled to generate EB0, EB4, and EB8 groups, respectively. The group fermented with B. subtilis 168 for eight days served as the WT8 control group. Each group was replicated five times.

After fermentation, moisture content and pH were measured according to previous methods [24]. Neutral detergent fiber (NDF) and acid detergent fiber (ADF) contents were analyzed using an A200I fiber analyzer (ANKOM, USA). Hemicellulose content was calculated according to the Association of Official Analytical Chemists method [25]. Reducing sugar content was determined via the 3,5-dinitrosalicylic acid method [26]. The content of lignin and total sugar was determined using specific commercial kits (BC4200 and BC2715; Solarbio, Beijing, China). Total viable count was determined following a previously described method [26].

Microscopic analysis was conducted on dried, ground, and sieved (400-mesh) enzymatically digested wheat straw samples to observe changes in fiber structure. Samples were fixed on a conductive platform, coated with gold, and observed in a field emission scanning electron microscope (SEM, Nano SEM-450, Thermo, USA). Fourier-transform infrared spectroscope (VERTEX 70, Bruker, Germany) was used for infrared spectrum analysis [27].

2.9. Data analysis

All experimental data were analyzed using SPSS 26.0 (SAS Inc., Chicago, USA). Homogeneity of variance was first assessed for each measurement index. When variance homogeneity was confirmed, intergroup comparisons were performed using the least significant difference test. When variance heterogeneity was detected, the Kruskal-Wallis H test was applied. The significance level considered statistically significant was P < 0.05, while P < 0.01 was considered extremely significant. Graphs were generated using GraphPad Prism version 8.4.3 (GraphPad Software Inc., Maryland, USA).

3. Results and discussion

3.1. Expression and optimization of endoglucanase at a single locus

The strain B. subtilis RLI2019 is derived from Reticulitermes labralis and has demonstrated high endoglucanase activity as well as strong capacity to degrade wheat straw. Moreover, the endoglucanase-encoding gene G2006 from the glycoside hydrolase family 5 was found to be the key cellulase gene in B. subtilis RLI2019 [23]. It is an acidic cellulase, with a K_m and V_max of 15.98 mg/mL and 20.78 μmol/min/mg, respectively [21]. In the present study, the cellulase expression cassettes were integrated into the B. subtilis genome using the CRISPR/Cas9 gene-editing system developed by Altenbuchner [19], achieving an editing efficiency typically exceeding 95 %. Firstly, the expression cassette from B. subtilis RLI2019 containing a promoter sequence, a signal peptide sequence, endoglucanase-encoding gene G2006, and downstream elements was integrated into the endoglucanase-encoding eglS locus of B. subtilis 168. The recombinant strain BSKI2006 exhibited an endoglucanase activity of 26.28 U/mL, marking a 59.18 % increase over B. subtilis 168 (16.51 U/mL) (Fig. 2A), indicating high expression and catalytic efficiency for the endoglucanase-encoding G2006 expression cassette.

Fig. 2 — Endoglucanase activity and growth curves of recombinant *Bacillus subtilis* strains. (A, B) Strains with different signal peptides. (C, D) Strains with expression cassettes integrated at different genomic loci. (E, F) Strains employing different transcriptional terminators. (G, H) Strains with multi-copy integrations.

**Notes:** 168 WT, Wild-type *Bacillus subtilis* 168. The genotypes of all strains are shown in Table 1. In the column chart, different lowercase letters above the bars indicate statistically significant differences (P < 0.05), while the same lowercase letters or absence of letters indicate no statistically significant difference (P > 0.05).

The most common strategies for enhancing extracellular protein expression include optimization of promoters, signal peptides, and genomic integration sites [28]. Among these, selection of an appropriate signal peptide is critical for efficient protein secretion [29]. To enhance extracellular protein expression, high-efficiency secretion signal peptides from B. subtilis 168 were added, including the Sec pathway signal peptides genes phoB and bglC, and the Tat pathway signal peptides genes phoD, lipA, and ywbN [30]. Each expression cassette was site-specifically integrated into the protease-encoding sprE locus under the control of the P₄₃ promoter. It was found that the endoglucanase activity of BSsprP₄₃BglC2006 was significantly increased by 6.82-fold compared to that of B. subtilis 168 (Fig. 2A). Similarly, the specific enzyme activity assay revealed that BSsprP₄₃BglC2006 exhibited the highest endoglucanase activity (Fig. S1A). This indicates that the signal peptide bglC derived from B. subtilis RLI2019 yielded optimal secretion efficiency for endoglucanase G2006.

Integration site selection of heterologous genes in the microbial genome has been shown to significantly influence their expression levels. Common high-expression loci for heterologous proteins include lacZ, thrC, nprE, and certain protease expression sites [31,32]. To assess the effect of integration site on endoglucanase G2006 expression, the P₄₃-bglC-G2006 expression cassette was integrated into the coding sequence regions of sprE, thrC, nprE, and lacZ loci. The results showed that the integration at the sprE locus exhibited the highest expression efficiency, with the recombinant strain BSsprP₄₃BglC2006 exhibiting an endoglucanase activity of 112.56 U/mL (Fig. 2C). Moreover, the expression efficiency of target genes integrated into the nprE and lacZ loci did not significantly change. This may result from the dual-promoter system formed by the P₄₃ promoter and the endogenous sprE promoter, enhancing transcription efficiency of the downstream gene or because the sprE-encoded protease might partially degrade endoglucanase G2006 [33].

Transcriptional terminators play a crucial role in the transcription process by preventing antisense transcription products from interfering with normal gene expression, providing regulatory structures at the 3′ end, enhancing RNA polymerase utilization and stabilizing mRNA, thereby improving upstream gene expression [34]. To increase both expression efficiency and stability of the endoglucanase gene, different terminators were included in the present study at the 3′ end of the P₄₃-bglC-G2006 expression cassette and integrated each modified cassette into the sprE locus. The results showed that the endoglucanase activity of BSsprP₄₃BglC2006TB4, BSsprP₄₃BglC2006TB5, BSsprP₄₃BglC2006TB9, and BSsprP₄₃BglC2006TH1 was 208.24, 190.47, 186.89, and 164.88 U/mL, respectively (Fig. 2C), indicating that the TB4 terminator yielded the highest expression of endoglucanase G2006.

The P₄₃ promoter from B. subtilis is a well-studied and widely used constitutive promoter. A previous study optimized its 5′-UTR to enhance target protein yield, leading to the development of a new synthetic constitutive promoter (P₁₂) [20]. In the present study, the P₄₃-bglC-G2006-TB4 expression cassette was further modified by replacing the P₄₃ promoter with the engineered P₁₂ promoter. The recombinant strain with this modification achieved a notably higher endoglucanase activity of 238.03 U/mL, suggesting that the P₁₂ promoter confers superior transcriptional activity for endoglucanase G2006 compared to the wild-type P₄₃ promoter (Fig. 2E).

The expression of heterologous genes in microbes can be achieved through plasmid-based or genome integration approaches. Genome integration provides a stable expression of foreign genetic material, whereas plasmid-based expression typically leads to unstable expression and typically requires the introduction of antibiotic resistance genes [35]. To validate plasmid expression efficiency, this study heterologously expressed the P₁₂-BglC-2006-TB4 expression cassette using the pJOE8999 vector as the backbone. The resulting strain, BSpP₁₂BglC2006TB4, exhibited a much lower endoglucanase activity of 73.32 U/mL (Fig. 2E), suggesting that integrating the endoglucanase expression cassette into the genomic sprE locus achieved more efficient expression and secretion. This observation could be attributed to the degradation of a portion of the endoglucanase G2006 by the protease encoded by sprE.

Although B. subtilis often exhibits strong environmental stress tolerance [16], genomic overexpression of specific functional genes may lead to impaired bacterial growth [36] and significant alterations in the growth curve [37]. In this study, the growth curves of B. subtilis 168 strains expressing endoglucanase from various genomic loci using different signal peptides displayed no significant abnormalities (Fig. 2B, 2D and 2F), indicating that endoglucanase gene expression at the selected four loci did not adversely affect the normal growth of B. subtilis. This may be attributed to the high sequence similarity between G2006, derived from B. subtilis RLI2019, and the eglS protein expressed by B. subtilis 168 [21], resulting in high metabolic efficiency of the host strain for cellulase expression.

3.2. Expression and optimization of endoglucanase at two loci

To investigate the expression efficiency of endoglucanase G2006 at two genomic loci, genes with different signal peptides were respectively inserted into the eglS and lacZ loci of strain BSsprP₁₂BglC2006TB4. The results showed that, compared to BSsprP₁₂BglC2006TB4, the endoglucanase activity of BSspreglS2006, BSsprlacBglC2006, BSsprlacPhoD2006, and BSsprlacPhoB2006, decreased significantly. Only BSsprlacYwb2006 and BSsprlacLip2006 displayed a significant increase in endoglucanase activity by 6.42 % and 12.96 %, respectively (Fig. 2G). Among these, BSsprlacLip2006 demonstrated a maximum endoglucanase activity of 268.89 U/mL. BSsprlacLip2006 and BSsprlacYwb2006 also showed a significant increase in endoglucanase specific activity compared to BSsprP₁₂BglC2006TB4 (Fig. S1D). This increase may stem from enhanced endoglucanase secretion through both the Sec and Tat pathways, effectively boosting secretion efficiency of the target protein and minimizing intracellular deposition [38].

Moreover, growth curve analyses (Fig. 2H) indicated that the overexpression of endoglucanase G2006 in B. subtilis 168 resulted in a significant extension of the lag phase of bacterial growth. This is possibly due to the recombinant bacteria expressing endoglucanase, which degraded certain nutrients in the substrate into smaller molecules, hence increasing nutritional availability and decelerating the decay rate.

Taken together, the endoglucanase gene from B. subtilis RLI2019 was efficiently overexpressed in B. subtilis 168. The most suitable gene expression cassettes and insertion loci were selected. When the constructed expression cassette P₁₂-bglC-G2006-TB4 and P₁₂-lipA-G2006-TB4 were integrated into both the sprE and lacZ loci, respectively, endoglucanase activity reached 268.89 U/mL.

3.3. Expression and optimization of bifunctional cellulase at a single locus

The objective of this study was to construct a recombinant B. subtilis with ternary composite cellulase activity. However, assays revealed that the strain B. subtilis 168 lacks both exoglucanase and β-glucosidase activities. Previous studies have shown that cellulase Bf1 (Gene ID PF0073) from Pyrococcus furiosus DSM 3638, noted for its thermal stability, possesses both exoglucanase and β-glucosidase activities, with K_m and V_max values of 14.19 mg/mL and 16.38 μmoL/min/mg, respectively [21]. Both Bf1 and G2006 share an optimal reaction temperature (30 °C) and demonstrate good thermal stability within the range of 20–40 °C. Furthermore, they exhibit strong synergistic effects during lignocellulose degradation [21]. Given the prediction of high folding efficiency for Bf1 and its correlation with improved Tat-dependent secretion, this study equipped Bf1 with the commonly used and high-efficiency Tat signal peptide LipA [21,28,39]. The expression cassette P₄₃-LipA-Bf1-TB9 was constructed and inserted at the nprE site. Consequently, the resulting strain BSnprP₄₃LipBfTB9 exhibited a β-glucosidase activity of 57.46 U/mL (Fig. 3A). To assess the influence of different terminators on Bf1 expression efficiency, the TB9 terminator in this cassette was successively replaced with TB5, TB4, and TH1. The TH1 terminator showed the highest efficiency, with strain BSnprP₄₃LipBfTH1 showing a β-glucosidase activity of 136.32 U/mL. Furthermore, the expression and secretion of Bf1 in the strain B. subtilis 168 did not show a significant negative impact on strain growth (Fig. 3B).

Fig. 3 — β-Glucosidase activity and growth curves of recombinant *Bacillus subtilis* strains. (A, B) Strains employing different transcriptional terminators. (C, D) Strains with expression cassettes integrated at different genomic loci. (E, F) Strains with different signal peptides. (G, H) Strains with multi-copy integrations.

**Notes:** 168 WT, Wild-type *Bacillus subtilis* 168. The genotypes of all strains are shown in Table 1. In the column chart, different lowercase letters above the bars indicate statistically significant differences (P < 0.05), while the same lowercase letters or absence of letters indicate no statistically significant difference (*P >* 0.05).

To further investigate the expression efficiency of the P₄₃-lipA-Bf1-TH1 expression cassette at different genomic loci, it was individually inserted into the lacZ, thrC, and sprE loci of B. subtilis 168. The results showed that the highest expression efficiency was observed at the thrC locus, with a β-glucosidase activity of 181.11 U/mL (Fig. 3C). Protein secretion optimization requires extensive screening due to the lack of a universally ideal signal peptide for different microorganisms. Moreover, predictive modeling is currently inadequate, and such calculations require extensive screening in order to identify the optimal signal peptide [40]. In this study, the results showed that the Tat pathway signal peptides LipA and YwbN were the most efficient for Bf1 secretion. The recombinant strain BSthrP₄₃YwbBfTH1 exhibited exoglucanase and β-glucosidase activities of 549.77 and 349.26 U/mL (Fig. 3E), respectively, which suggests that the bifunctional cellulase Bf1 efficiently folded within the cell, thus favoring secretion via the Tat pathway [39]. Furthermore, expression of the bifunctional cellulase Bf1 using the artificial promoter P₁₂ did not result in significant change in β-glucosidase activity in the strain BSthrP₁₂YwbBfTH1. Additionally, the expression of the P₄₃-ywbN-Bf1-TH1 cassette was evaluated using the expression vector pJOE8999. The resulting strain, BSpP₄₃YwbBfTH1, exhibited a β-glucosidase activity of 203.62 U/mL. This indicates that the genomic integration of this gene expression cassette resulted in higher expression and secretion efficiency of Bf1.

3.4. Expression and optimization of bifunctional cellulase at two loci

Although multicopy integration can synergistically enhance heterologous protein yield in filamentous fungi [41], co-expression of Bf1 at two genomic loci using different signal peptides did not result in a significant increase in β-glucosidase activity (Fig. 3G). Analysis of specific activity further indicated that dual-site expression led to a significant reduction in β-glucosidase activity compared with the control strain BSthrP₄₃YwbBfTH1 (Fig. S2D). This reduction may be attributed to inhibition of transcriptional and translational processes by the heterologously expressed bifunctional cellulase Bf1 and/or the accumulation of its substrate degradation products [42]. Alternatively, the observed decrease may result from metabolic imbalance caused by the cellular biosynthetic capacity being exceeded by the burden of heterologous protein synthesis [35]. Future improvements in gene expression could be achieved by optimizing the composition of the fermentation medium and refining culture conditions to enhance cell density and cellulase expression [43]. Additionally, following dual-site expression of Bf1, the decline rate of the recombinant strain was significantly reduced (Fig. 3H). This effect may be attributed to partial alleviation of metabolic burden when the recombinant strain was cultured in medium containing complex carbon sources, potentially resulting from the knockout of the lacZ and thrC genes.

Overall, this study is the first to successfully achieve the expression and secretion of the bifunctional cellulase Bf1 in B. subtilis 168. Integration of the expression cassette P₄₃-YwbN-Bf1-TH1 into the thrC locus enabled efficient expression of Bf1. In particular, the recombinant strain BSthrP₄₃YwbBfTH1 exhibited exoglucanase and β-glucosidase activities of 549.77 and 349.26 U/mL, respectively.

3.5. Expression and optimization of complex cellulase

Various cellulolytic bacteria have been described in the literature, including Bacillus, Cellulomanas, Clostridium, Talaromyces, and Streptomyces [44,45]. However, these cellulolytic bacteria often produce either a single highly active cellulase or multiple low-activity cellulases, which significantly limits their industrial application. Herein, a recombinant B. subtilis strain expressing three cellulase activities simultaneously was constructed by inserting the expression cassettes P₁₂-bglC-G2006-TB4 and P₄₃-ywbN-Bf1-TH1 onto the plasmid pJOE2006Bf in B. subtilis 168. The resulting strain, i.e., BSp2006Bf, exhibited endoglucanase, exoglucanase, and β-glucosidase activities, i.e., 59.51, 247.99, and 183.08 U/mL, respectively (Fig. 4A–C). Subsequently, following integration of these two expression cassettes at the sprE and thrC loci using genome editing in B. subtilis 168, the recombinant strain BSspr2006thrBf exhibited higher endoglucanase, exoglucanase, and β-glucosidase activities, i.e., 106.85, 453.70, and 359.25 U/mL, respectively. This suggests that genomic integration may achieve more efficient expression and secretion of target proteins. Further integration of the P₁₂-lipA-G2006-TB4 cassette into the lacZ locus of BSspr2006thrBf yielded the strain BSKI3Cel, which exhibited the highest endoglucanase, exoglucanase, and β-glucosidase activities, i.e., 129.59, 596.75, and 447.42 U/mL, respectively. Thus, these enhanced exoglucanase and β-glucosidase activities likely resulted from the increased substrate availability of the degradation products of endoglucanase G2006 by bifunctional cellulase Bf1 [46].

Fig. 4 — Cellulase activities and growth curve of recombinant strains producing composite cellulase activities. Endoglucanase (A), exoglucanase (B), and β-glucosidase (C) activities of the recombinant strains after 24 h of cultivation. (D) Growth curve of composite cellulase-producing strains. (E) OD₆₀₀ of each strain after 12 h of cultivation. (F) Hemicellulase activity of the recombinant strains. Endoglucanase (G), exoglucanase (H), and β-glucosidase (I) activities of the BSKI3Cel and BSK3P2C at different time points.

**Notes:** 168 WT, Wild-type *Bacillus subtilis* 168. The genotypes of all strains are shown in Table 1. In the column chart, different lowercase letters above the bars indicate statistically significant differences (P < 0.05), while the same lowercase letters or absence of letters indicate no statistically significant difference (*P >* 0.05).

Furthermore, transformation of plasmid pJOE2006Bf into the recombinant strain BSKI3Cel, which generated the strain BSK3P2C, exhibited endoglucanase, exoglucanase, and β-glucosidase activities of 267.82, 1413.84, and 943.66 U/mL, respectively, after 24 h of cultivation. These findings indicate that the combined approach of plasmid-based expression and genome integration effectively enhanced the expression and secretion of heterologous proteins. Moreover, growth curve analysis revealed that the simultaneous overexpression of cellulases G2006 and Bf1 in B. subtilis 168 effectively delayed the onset of the stationary phase (Fig. 4D and 4E), a finding consistent with the effects of individually overexpressing G2006 in the strain, as described in Section 3.2.

Previous studies have shown that B. subtilis RLI2019 exhibits efficient hemicellulose degradation [23]. In the present study, heterologous expression of its key cellulase gene G2006 resulted in a hemicellulase (xylanase) activity of 201.54 U/mL in the recombinant strain BSsprlacLip2006, whereas the strain BSK3P2C expressing composite cellulases exhibited a higher hemicellulase activity of 247.41 U/mL (Fig. 4F). These results indicate that the endoglucanase G2006 also possesses substantial xylanase activity.

Moreover, time-point analysis of cellulase activities in strains BSKI3Cel and BSK3P2C showed that both strains exhibited relatively high enzymatic activity levels from day 7–9 (Fig. 4G–I). On day 7, BSKI3Cel displayed the highest endoglucanase and exoglucanase activities (407.89 and 2291.45 U/mL, respectively), while its peak β-glucosidase activity (2168.45 U/mL) occurred on day 9. In contrast, BSK3P2C exhibited peak endoglucanase and β-glucosidase activities (536.78 and 3510.75 U/mL, respectively) on day 9, and reached its maximum exoglucanase activity of 3078.46 U/mL on day 7.

Taken together, this novel recombinant B. subtilis strain, BSKI3Cel, constructed without the use of antibiotic resistance markers, could express three complete cellulase systems by integrating three cellulase-encoding cassettes into its genome. By day 7 of cultivation, BSKI3Cel displayed endoglucanase, exoglucanase, and β-glucosidase activities of 407.89, 2291.45, and 2127.29 U/mL, respectively. Furthermore, through the integration of five cellulase expression cassettes into the genome and using a plasmid-based expression system, the resulting recombinant strain BSK3P2C exhibited endoglucanase, exoglucanase, and β-glucosidase activities of 533.16, 2959.83, and 2829.61 U/mL, respectively, on day 8 of cultivation.

3.6. Effects of recombinant strain fermentation on physicochemical properties of wheat straw

B. subtilis is a food-grade microorganism widely used for lignocellulosic biomass degradation, animal feed production, and food processing [13,15], due to its strong sporulation ability, heat and acid tolerance, and ability to express and secrete highly active proteases and cellulases [12,47]. In the present study, the moisture and temperature conditions of the ruminant rumen were simulated, and fermentation of wheat straw was performed using recombinant strains producing composite cellulases. The results showed that moisture content in the EB8 group significantly increased, while dry matter content significantly decreased (Fig. 5A and 5B), indicating that B. subtilis effectively convert organic matter in wheat straw into volatile substances, such as water and organic acids [27].

Fig. 5 — Changes in the contents of various components in the fermentation process of recombinant strain. (A) Moisture. (B) Dry matter. (C) Lignin. (D) Hemicellulose. (F) NDF. (G) ADF. (H) Total sugar. (I) Reducing sugar. (J) Viable cont. (K) pH.

**Notes:** NDF, neutral detergent fiber. ADF, acid detergent fiber. The samples of wheat straw fermented with recombinant strain BSK3P2C for 0, 4, and 8 days were designated as the EB0, EB4, and EB8 groups, respectively. The sample fermented with *B. subtilis* 168 for 8 days was designated as the WT8 group. In the column chart, different lowercase letters above the bars indicate statistically significant differences (P < 0.05), while the same lowercase letters or absence of letters indicate no statistically significant difference (P > 0.05).

After 8 days of fermentation with BSK3P2C, the contents of hemicellulose, NDF, and ADF were significantly reduced by 16.70 %, 7.46 %, and 9.93 %, respectively, in the EB8 group compared to the EB0 group (Fig. 5D, 5F and 5G). Further verification revealed that the hemicellulase activity of the EB4 and EB8 groups was significantly higher compared to that in EB0 and WT8 groups (Fig. 5E). In comparison with the WT8 group, the EB8 group showed significantly lower contents of hemicellulose, NDF, and ADF (by 15.93 %, 6.71 %, and 8.92 %, respectively). This indicates that the recombinant strain BSK3P2C could more effectively reduce the contents of hemicellulose, NDF, and ADF in wheat straw.

Endoglucanases and other enzymes are commonly used in the saccharification of agricultural residues to degrade lignocellulose into reducing sugars [48]. In this study, after 8 days of fermentation with BSK3P2C, total sugar content in wheat straw was significantly reduced compared to the EB0 group, while reducing sugar content was increased (Fig. 5H and 5I). This increase may be attributed to partial degradation of lignocellulose in wheat straw by cellulases produced by the recombinant strain, which act on glycosidic bonds to generate sugars and low-molecular-weight compounds. These metabolites can subsequently be utilized by B. subtilis for biomass synthesis, leading to a significant increase in viable cell counts in the EB8 group (Fig. 5J). This points to the promising value that the BSK3P2C can bring to lignocellulose saccharification applications.

In ruminant feeding, the addition of probiotic strains such as B. subtilis is common practice to improve ruminal degradation efficiency and maintain rumen pH, thus preventing acidosis [49,50]. The findings of the present study demonstrated that fermentation with BSK3P2C significantly increased the pH of the wheat straw by 37.81 %, compared to a 2.93 % increase with B. subtilis 168 (Fig. 5K). This indicates that BSK3P2C was more effective in improving the pH of the substrate, which is more beneficial for livestock production.

3.7. Effect of recombinant strain fermentation on the microscopic structure of wheat straw

SEM observations of the surface structure of wheat straw (Fig. 6A–D) revealed that unfermented wheat straw had a relatively smooth and flat surface. On day 4 of fermentation with BSK3P2C, wheat straw surface exhibited a rugged structure and certain areas of wave-like texture, which increased significantly by day 8, with the surface appearing loose and porous. This is likely due to enzymatic hydrolysis and fermentation that degraded cellulose, hemicellulose, lignin, and pectin on the surface layer, thus enlarging voids and pore sizes. This improved porosity likely enhances lignocellulosic biomass adsorption, thus playing an important role in modifying its physical and chemical characteristics [26]. In contrast, only slight structural changes were observed on wheat straw surface in the WT8 group, indicating that the BSK3P2C was more effective in degrading the lignocellulosic structure on wheat straw surface.

FTIR analysis further showed that the characteristic peaks of cellulose at approximately 3340 cm⁻¹ (O–H stretching), 2910 cm⁻¹ (C–H stretching), and 1020 cm⁻¹ (C–O stretching) were significantly reduced in samples treated with recombinant strain BSK3P2C on days 4 and 8 compared to unfermented wheat straw and fermented with B. subtilis 168 on day 8 (Fig. 6E) [51]. This reduction indicates that BSK3P2C could significantly change the chemical bond composition of cellulose in wheat straw, thus reducing NDF and ADF contents. Moreover, the absorption peak at approximately 1730 cm⁻¹, which indicates the C O stretching vibration of the Cacetyl group in hemicellulose, was significantly reduced in both EB4 and EB8 groups, suggesting that BSK3P2C could effectively degrade hemicellulose in wheat straw [52], consistent with the notably increased hemicellulase activity and reduced hemicellulose content in the EB8 group.

In industrial production, filamentous fungi are commonly employed for cellulase production. Currently, the most widely used strain is the engineered strain Trichoderma reesei RUT-C30, which can achieve a maximum endoglucanase activity of 3456 U/mL in fed-batch fermentation, with enzyme activities normalized according to the methodology used in the present study [3]. However, due to multiple gene mutations and deletions (over 100 kb), this strain exhibits a reduced capacity to utilize many carbon sources (such as polysaccharides and oligosaccharides). Its cellulase production process typically relies on specialized liquid fermentation techniques and the addition of specific inducers, resulting in increased process complexity and higher production costs [53]. These constraints hinder the direct applicability of the strains in lignocellulosic biomass degradation and animal feeding practices.

Accordingly, our group previously selected B. subtilis RLI2019, a strain with high endoglucanase activity, as a chassis strain and, for the first time, integrated the coding sequences of multiple cellulase genes (i.e., eglS, cel48S, and bglS) into its genome [22]. Using the same experimental procedure as in the present study, the recombinant strain B. subtilis AEA3 achieved maximum cellulase activities of 537.69, 462.54, and 793.86 U/mL, respectively. However, the genetic characteristics, metabolic mechanisms, and safety profile of this strain remain largely uncharacterized.

Therefore, the present study employed the model strain B. subtilis 168 as the chassis. Following optimization of genetic elements, the constructed recombinant strain BSK3P2C achieved maximum cellulase activities corresponding to 75.86 %, 4.95-fold, and 2.68-fold of those observed in B. subtilis AEA3, respectively. This improvement is primarily attributed to the higher catalytic efficiency of the bifunctional cellulase Bf1 compared with Cel48S and BglS [21]. Consequently, strain BSK3P2C exhibited significantly elevated exoglucanase and β-glucosidase activities. Furthermore, the strain exhibited more effective degradation of NDF in wheat straw, likely due to the synergistic interactions between the two cellulases [21,22], highlighting its potential advantages for lignocellulose degradation.

4. Conclusion

In the present study, a recombinant B. subtilis strain with high cellulase production, is optimized for a ternary cellulase system. Signal peptides, transcriptional terminators, and integration sites were carefully screened and optimized for both the endoglucanase gene G2006 and the bifunctional cellulase gene Bf1. Integration of the expression cassettes P₁₂-bglC-G2006-TB4, P₁₂-lipA-G2006-TB4, and P₁₂-lipA-G2006-TB4 into the sprE, lacZ, and thrC loci using a CRISPR-based gene-editing system, respectively, yielded a strain BSKI3Cel producing a high yield ternary complex cellulase. Following plasmid expression in BSKI3Cel, the resulting recombinant strain BSK3P2C exhibited higher endoglucanase, exoglucanase, and β-glucosidase activities, i.e., 533.16, 2959.83, and 2829.61 U/mL, respectively, after 8 days of cultivation. When the recombinant strain BSK3P2C was applied in fermentation of wheat straw for 8 days, hemicellulose, NDF, and ADF contents were significantly reduced by 16.70 %, 7.46 %, and 9.93 %, respectively, and pH was significantly increased. These findings indicate that the recombinant strain developed herein could effectively degrade lignocellulose in wheat straw, which underscores the potential industrial and agricultural application of this novel recombinant B. subtilis strain for a more efficient degradation of lignocellulosic biomass.

CRediT authorship contribution statement

Zhiwei Wang: Writing – original draft, Methodology, Funding acquisition, Conceptualization. Gongwei Liu: Data curation, Conceptualization. Zhongming Meng: Methodology, Data curation. Jie Xu: Methodology, Data curation. Haoran Tang: Methodology, Investigation. Shendong Wang: Resources, Formal analysis. Caixia Zou: Writing – review & editing. Caiying Ma: Writing – review & editing. Yuxin Yang: Writing – review & editing, Supervision, Funding acquisition.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the National Key R&D Program of China (2022YFD1300201), the High-Level Talent Research Initiation Project of Guangxi University (ZX01080033425016), and the China Agriculture Research System (CARS-39-12).

Footnotes

Peer review under the responsibility of Editorial Board of Synthetic and Systems Biotechnology.

^{Appendix A}

Supplementary data to this article can be found online at https://doi.org/10.1016/j.synbio.2026.01.021.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Multimedia component 1

mmc1.xlsx^{(23.4KB, xlsx)}

Multimedia component 2

mmc2.docx^{(834.7KB, docx)}

References

1.Saini S., Sharma K.K. Fungal lignocellulolytic enzymes and lignocellulose: a critical review on their contribution to multiproduct biorefinery and global biofuel research. Int J Biol Macromol. 2021;193:2304–2319. doi: 10.1016/j.ijbiomac.2021.11.063. [DOI] [PubMed] [Google Scholar]
2.Dias A.H.S., Skaf M.S., Silveira R.L. Computational insights into glucose tolerance and stimulation in a family 1 β-glucosidase. J Chem Inf Model. 2025;65:7102–7112. doi: 10.1021/acs.jcim.5c00922. [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Huang X., Deng L., Zhang X., Zhao X., Liu C., Bai F. Developing a soluble inducer for robust production of cellulases by Trichoderma reesei through submerged fermentation. Green Chem. 2025;27:2733. doi: 10.1039/d4gc06001f. [DOI] [Google Scholar]
4.Li W., Huan X., Zhou Y., Ma Q., Chen Y. Simultaneous cloning and expression of two cellulase genes from Bacillus subtilis newly isolated from Golden Takin (Budorcas taxicolor Bedfordi) Biochem Biophys Res Commun. 2009;383:397–400. doi: 10.1016/j.bbrc.2009.04.027. [DOI] [PubMed] [Google Scholar]
5.John J.A., Samuel M.S., Govarthanan M., Selvarajan E. A comprehensive review on strategic study of cellulase producing marine actinobacteria for biofuel applications. Environ Res. 2022;214 doi: 10.1016/j.envres.2022.114018. [DOI] [PubMed] [Google Scholar]
6.Khan S.I., Sahinkaya M., Colak D.N., Zada N.S., Uzuner U., Belduz A.O., et al. Production and characterization of novel thermostable CotA-laccase from Bacillus altitudinis SL7 and its application for lignin degradation. Enzym Microb Technol. 2024;172 doi: 10.1016/j.enzmictec.2023.110329. [DOI] [PubMed] [Google Scholar]
7.Santos C.A., Morais M.A.B., Mandelli F., Lima E.A., Miyamoto R.Y., Higasi P.M.R., et al. A metagenomic ‘dark matter’ enzyme catalyses oxidative cellulose conversion. Nature. 2025;639:1076–1083. doi: 10.1016/j.biortech.2021.126263. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Guo H., He T., Lee D. Contemporary proteomic research on lignocellulosic enzymes and enzymolysis: a review. Bioresour Technol. 2022;344 doi: 10.1038/s41586-024-08553-z. [DOI] [PubMed] [Google Scholar]
9.Watanabe H., Tokuda G. Animal cellulases. Cell Mol Life Sci. 2001;58:1167–1178. doi: 10.1186/s43141-023-00619-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Hussain S., Yasin M.T., Ahmad K., Khan S., Ahmad R., Khan J., et al. Enhancement effect of AgO nanoparticles on fermentative cellulase activity from thermophilic Bacillus subtilis Ag-PQ. J Genet Eng Biotechnol. 2023;21:151. doi: 10.1007/PL00000931. [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Liu Y., Cao X., Wang X., Chen R., Yuan B., Li S., et al. Multiplex base editing of RBSs rewires Bacillus subtilis metabolism for lycopene overproduction. Metab Eng. 2026;93:169–183. doi: 10.1016/j.ymben.2025.10.005. [DOI] [PubMed] [Google Scholar]
12.Park S.A., Bhatia S.K., Park H.A., Kim S.Y., Sudheer P.D.V.N., Yang Y., et al. Bacillus subtilis as a robust host for biochemical production utilizing biomass. Crit Rev Biotechnol. 2021;41:827–848. doi: 10.1080/07388551.2021.1888069. [DOI] [PubMed] [Google Scholar]
13.Ma L., Aizhan R., Wang X., Yi Y., Shan Y., Liu B., et al. Cloning and characterization of low-temperature adapted GH5-CBM3 endo-cellulase from Bacillus subtilis 1AJ3 and their application in the saccharification of switchgrass and coffee grounds. AMB Express. 2020;10:42. doi: 10.3390/polym15204142. [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Li H., Zhang M., Zhang Y., Xu X., Zhao Y., Jiang X., et al. Characterization of cellulose-degrading bacteria isolated from silkworm excrement and optimization of its cellulase production. Polymers (Basel) 2023;15(20):4142. doi: 10.1186/s13568-020-00975-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Li Y., Wu Y., Liu Y., Li J., Du G., Lv X., et al. A genetic toolkit for efficient production of secretory protein in Bacillus subtilis. Bioresour Technol. 2022;363 doi: 10.1016/j.biortech.2022.127885. [DOI] [PubMed] [Google Scholar]
16.Liu Y., Cheng H., Li H., Zhang Y., Wang M. A programmable CRISPR/Cas9 toolkit improves lycopene production in Bacillus subtilis. Appl Environ Microbiol. 2023;89 doi: 10.1128/aem.00230-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Watzlawick H., Altenbuchner J. Multiple integration of the gene ganA into the Bacillus subtilis chromosome for enhanced beta-galactosidase production using the CRISPR/Cas9 system. AMB Express. 2019;9:158. doi: 10.1186/s13568-019-0884-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Wang Y., Cheng H., Liu Y., Liu Y., Wen X., Zhang K., et al. In-situ generation of large numbers of genetic combinations for metabolic reprogramming via CRISPR-guided base editing. Nat Commun. 2021;12:678. doi: 10.1038/s41467-021-21003-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Altenbuchner J. Editing of the Bacillus subtilis genome by the CRISPR-Cas9 system. Appl Environ Microbiol. 2016;82:5421–5427. doi: 10.1128/AEM.01453-16. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Xiao J., Peng B., Su Z., Liu A., Hu Y., Nomura C.T., et al. Facilitating protein expression with portable 5’-UTR secondary structures in Bacillus licheniformis. ACS Synth Biol. 2020;9:1051–1058. doi: 10.1021/acssynbio.9b00355. [DOI] [PubMed] [Google Scholar]
21.Wang Z., Tang H., Li Y., Yang B., Liang X., Gong H., et al. Characterization and synergistic activity of heterologously expressed microbial-derived endoglucanase and bifunctional cellulase on wheat straw. Int J Biol Macromol. 2024;282 doi: 10.1016/j.ijbiomac.2024.137485. [DOI] [PubMed] [Google Scholar]
22.Liu G., Gong H., Tang H., Meng Z., Wang Z., Cui W., et al. Enhanced lignocellulose degradation in Bacillus subtilis RLI2019 through CRISPR/Cas9-mediated chromosomal integration of ternary cellulase genes. Int J Biol Macromol. 2025;306 doi: 10.1016/j.biortech.2023.129457. [DOI] [PubMed] [Google Scholar]
23.Liu G., Zhang K., Gong H., Yang K., Wang X., Zhou G., et al. Whole genome sequencing and the lignocellulose degradation potential of Bacillus subtilis RLI2019 isolated from the intestine of termites. Biotechnol Biofuels Bioprod. 2023;16:130. doi: 10.1186/s13068-023-02375-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Su W., Jiang Z., Wang C., Xu B., Lu Z., Wang F., et al. Dynamics of defatted rice bran in physicochemical characteristics, microbiota and metabolic functions during two-stage co-fermentation. Int J Food Microbiol. 2022;362 doi: 10.1016/j.ijfoodmicro.2021.109489. [DOI] [PubMed] [Google Scholar]
25.Pedrazzani C., Viscusi P., Fuso A., Caligiani A. Development of a protocol for fractionating and characterising fibres from lignocellulosic food waste. Food Chem X. 2024;22 doi: 10.1016/j.fochx.2024.101501. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Wang Z., Tang H., Liu G., Gong H., Li Y., Chen Y., et al. Compound probiotics producing cellulase could replace cellulase preparations during solid-state fermentation of millet bran. Bioresour Technol. 2023;385 doi: 10.1016/j.biortech.2023.129457. [DOI] [PubMed] [Google Scholar]
27.Wang Z., Tang H., Li Y., Tian L., Ye B., Yan W., et al. Evaluating the dynamic effects of complex probiotics as cellulase replacements during fermentation of apple pomace. Int J Food Microbiol. 2024;425 doi: 10.1016/j.ijfoodmicro.2024.110896. [DOI] [PubMed] [Google Scholar]
28.Zhang M., Zhen J., Teng J., Zhao X., Fu X., Song H., et al. N-Terminal sequences of signal peptides assuming critical roles in expression of heterologous proteins in Bacillus subtilis. Microorganisms. 2024;12:1275. doi: 10.3390/microorganisms12071275. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Tian J., Long X., Tian Y., Shi B. Enhanced extracellular recombinant keratinase activity in Bacillus subtilis SCK6 through signal peptide optimization and site-directed mutagenesis. RSC Adv. 2019;9:33337–33344. doi: 10.1039/c9ra07866e. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Song Y., Nikoloff J.M., Zhang D. Improving protein production on the level of regulation of both expression and secretion pathways in Bacillus subtilis. J Microbiol Biotechnol. 2015;25:963–977. doi: 10.4014/jmb.1501.01028. [DOI] [PubMed] [Google Scholar]
31.Sauer C., Syvertsson S., Bohorquez L.C., Cruz R., Harwood C.R., van Rij T., et al. Effect of genome position on heterologous gene expression in Bacillus subtilis: an unbiased analysis. ACS Synth Biol. 2016;5:942–947. doi: 10.1021/acssynbio.6b00065. [DOI] [PubMed] [Google Scholar]
32.Jeong D., So Y., Park S., Park S., Choi S. Random knock-in expression system for high yield production of heterologous protein in Bacillus subtilis. J Biotechnol. 2018;266:50–58. doi: 10.1016/j.jbiotec.2017.12.007. [DOI] [PubMed] [Google Scholar]
33.Ozturk S., Ergun B.G., Calik P. Double promoter expression systems for recombinant protein production by industrial microorganisms. Appl Microbiol Biotechnol. 2017;101:7459–7475. doi: 10.1007/s00253-017-8487-y. [DOI] [PubMed] [Google Scholar]
34.Ray-Soni A., Bellecourt M.J., Landick R. Mechanisms of bacterial transcription termination: all good things must end. Annu Rev Biochem. 2016;85:319–347. doi: 10.1146/annurev-biochem-060815-014844. [DOI] [PubMed] [Google Scholar]
35.Yu X., Lv H., Luo H., Zhu X., Wu J., Zhang K. High level food-grade expression of maltogenic amylase in Bacillus subtilis through genomic integration and comA overexpression. Int J Biol Macromol. 2025;309 doi: 10.1016/j.ijbiomac.2025.143060. [DOI] [PubMed] [Google Scholar]
36.Liu L., Ding D., Wang H., Ren X., Lee S.Y., Zhang D. Balancing cell growth and product synthesis for efficient microbial cell factories. Adv Sci (Weinh) 2025 doi: 10.1002/advs.202510649. [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Wu Y., Li Y., Liu Y., Xiu X., Liu J., Zhang L., et al. Multiplexed in-situ mutagenesis driven by a dCas12a-based dual-function base editor. Nucleic Acids Res. 2024;52:4739–4755. doi: 10.1093/nar/gkae228. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Ejaz S., Khan H., Sarwar N., Aqeel S.M., Al-Adeeb A., Liu S. A review on recent advancement in expression strategies used in Bacillus subtilis. Protein Pept Lett. 2022;29:733–743. doi: 10.2174/0929866529666220803163335. [DOI] [PubMed] [Google Scholar]
39.Owji H., Nezafat N., Negahdaripour M., Hajiebrahimi A., Ghasemi Y. A comprehensive review of signal peptides: structure, roles, and applications. Eur J Cell Biol. 2018;97:422–441. doi: 10.1016/j.ejcb.2018.06.003. [DOI] [PubMed] [Google Scholar]
40.Wang B., Wu Y., Lv X., Liu L., Li J., Du G., et al. Synergistic regulation of chassis cell growth and screening of promoters, signal peptides and fusion protein linkers for enhanced recombinant protein expression in Bacillus subtilis. Int J Biol Macromol. 2024;280 doi: 10.1016/j.ijbiomac.2024.136037. [DOI] [PubMed] [Google Scholar]
41.Shi Z., Li J., Zhu F., Liu X., Zhou J., Li J., et al. Multidimensional strategies for efficient heterologous protein expression in Aspergillus niger. Biotechnol Adv. 2025;84 doi: 10.1016/j.biotechadv.2025.108690. [DOI] [PubMed] [Google Scholar]
42.Görke B., Stülke J. Carbon catabolite repression in bacteria: many ways to make the most out of nutrients. Nat Rev Microbiol. 2008;6:613–624. doi: 10.1038/nrmicro1932. [DOI] [PubMed] [Google Scholar]
43.Xiang M., Kang Q., Zhang D. Advances on systems metabolic engineering of Bacillus subtilis as a chassis cell. Synth Syst Biotechnol. 2020;5:245–251. doi: 10.1016/j.synbio.2020.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Chukwuma O.B., Rafatullah M., Tajarudin H.A., Ismail N. A review on bacterial contribution to lignocellulose breakdown into useful bio-products. Int J Environ Res Publ Health. 2021;18(11):6001. doi: 10.3390/ijerph18116001. [DOI] [PMC free article] [PubMed] [Google Scholar]
45.Sun Y., Sun Q., Gao T., Ren S., Hu X. Competitive binding of TeXlnR and TeAmyR regulates cellulase production in Talaromyces endophyticus. Synth Syst Biotechnol. 2026;11:356–369. doi: 10.1016/j.synbio.2025.10.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Xu C., Tong S., Sun L., Gu X. Cellulase immobilization to enhance enzymatic hydrolysis of lignocellulosic biomass: an all-inclusive review. Carbohydr Polym. 2023;321 doi: 10.1016/j.carbpol.2023.121319. [DOI] [PubMed] [Google Scholar]
47.Stulke J., Gruppen A., Bramkamp M., Pelzer S. Bacillus subtilis, a swiss army knife in science and biotechnology. J Bacteriol. 2023;205 doi: 10.1128/jb.00102-23. [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Huang Z., Ni G., Zhao X., Wang F., Qu M. Characterization of a GH8 beta-1,4-glucanase from Bacillus subtilis B111 and its saccharification potential for agricultural straws. J Microbiol Biotechnol. 2021;31:1446–1454. doi: 10.4014/jmb.2105.05026. [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Guo X., Sha Y., Lv W., Pu X., Liu X., Luo Y., et al. Sex differences in rumen fermentation and microbiota of Tibetan goat. Microb Cell Fact. 2022;21:55. doi: 10.1186/s12934-022-01783-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Kamal M., Linlin K., Gao J., Xinrui Z., Xinming C., Haibo W., et al. Effects of Saccharomyces cerevisiae and Bacillus subtilis on in vitro fermentation in the rumen of Hu sheep. J Sci Food Agric. 2025;105:498–506. doi: 10.1002/jsfa.13848. [DOI] [PubMed] [Google Scholar]
51.Song P., Zhou F., Li F., Han Z., Wang L., Xu J., et al. Superfine pulverisation pretreatment to enhance crystallinity of cellulose from Lycium barbarum L. leaves. Carbohydr Polym. 2021;253 doi: 10.1016/j.carbpol.2020.117207. [DOI] [PubMed] [Google Scholar]
52.Astruc J., Nagalakshmaiah M., Laroche G., Grandbois M., Elkoun S., Robert M. Isolation of cellulose-II nanospheres from flax stems and their physical and morphological properties. Carbohydr Polym. 2017;178:352–359. doi: 10.1016/j.carbpol.2017.08.138. [DOI] [PubMed] [Google Scholar]
53.Peterson R., Nevalainen H. Trichoderma reesei RUT-C30 – thirty years of strain improvement. Microbiology (Read) 2012;158:58–68. doi: 10.1099/mic.0.054031-0. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Multimedia component 1

mmc1.xlsx^{(23.4KB, xlsx)}

Multimedia component 2

mmc2.docx^{(834.7KB, docx)}

[bib1] 1.Saini S., Sharma K.K. Fungal lignocellulolytic enzymes and lignocellulose: a critical review on their contribution to multiproduct biorefinery and global biofuel research. Int J Biol Macromol. 2021;193:2304–2319. doi: 10.1016/j.ijbiomac.2021.11.063. [DOI] [PubMed] [Google Scholar]

[bib2] 2.Dias A.H.S., Skaf M.S., Silveira R.L. Computational insights into glucose tolerance and stimulation in a family 1 β-glucosidase. J Chem Inf Model. 2025;65:7102–7112. doi: 10.1021/acs.jcim.5c00922. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib3] 3.Huang X., Deng L., Zhang X., Zhao X., Liu C., Bai F. Developing a soluble inducer for robust production of cellulases by Trichoderma reesei through submerged fermentation. Green Chem. 2025;27:2733. doi: 10.1039/d4gc06001f. [DOI] [Google Scholar]

[bib4] 4.Li W., Huan X., Zhou Y., Ma Q., Chen Y. Simultaneous cloning and expression of two cellulase genes from Bacillus subtilis newly isolated from Golden Takin (Budorcas taxicolor Bedfordi) Biochem Biophys Res Commun. 2009;383:397–400. doi: 10.1016/j.bbrc.2009.04.027. [DOI] [PubMed] [Google Scholar]

[bib5] 5.John J.A., Samuel M.S., Govarthanan M., Selvarajan E. A comprehensive review on strategic study of cellulase producing marine actinobacteria for biofuel applications. Environ Res. 2022;214 doi: 10.1016/j.envres.2022.114018. [DOI] [PubMed] [Google Scholar]

[bib6] 6.Khan S.I., Sahinkaya M., Colak D.N., Zada N.S., Uzuner U., Belduz A.O., et al. Production and characterization of novel thermostable CotA-laccase from Bacillus altitudinis SL7 and its application for lignin degradation. Enzym Microb Technol. 2024;172 doi: 10.1016/j.enzmictec.2023.110329. [DOI] [PubMed] [Google Scholar]

[bib7] 7.Santos C.A., Morais M.A.B., Mandelli F., Lima E.A., Miyamoto R.Y., Higasi P.M.R., et al. A metagenomic ‘dark matter’ enzyme catalyses oxidative cellulose conversion. Nature. 2025;639:1076–1083. doi: 10.1016/j.biortech.2021.126263. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] 8.Guo H., He T., Lee D. Contemporary proteomic research on lignocellulosic enzymes and enzymolysis: a review. Bioresour Technol. 2022;344 doi: 10.1038/s41586-024-08553-z. [DOI] [PubMed] [Google Scholar]

[bib9] 9.Watanabe H., Tokuda G. Animal cellulases. Cell Mol Life Sci. 2001;58:1167–1178. doi: 10.1186/s43141-023-00619-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.Hussain S., Yasin M.T., Ahmad K., Khan S., Ahmad R., Khan J., et al. Enhancement effect of AgO nanoparticles on fermentative cellulase activity from thermophilic Bacillus subtilis Ag-PQ. J Genet Eng Biotechnol. 2023;21:151. doi: 10.1007/PL00000931. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib11] 11.Liu Y., Cao X., Wang X., Chen R., Yuan B., Li S., et al. Multiplex base editing of RBSs rewires Bacillus subtilis metabolism for lycopene overproduction. Metab Eng. 2026;93:169–183. doi: 10.1016/j.ymben.2025.10.005. [DOI] [PubMed] [Google Scholar]

[bib12] 12.Park S.A., Bhatia S.K., Park H.A., Kim S.Y., Sudheer P.D.V.N., Yang Y., et al. Bacillus subtilis as a robust host for biochemical production utilizing biomass. Crit Rev Biotechnol. 2021;41:827–848. doi: 10.1080/07388551.2021.1888069. [DOI] [PubMed] [Google Scholar]

[bib13] 13.Ma L., Aizhan R., Wang X., Yi Y., Shan Y., Liu B., et al. Cloning and characterization of low-temperature adapted GH5-CBM3 endo-cellulase from Bacillus subtilis 1AJ3 and their application in the saccharification of switchgrass and coffee grounds. AMB Express. 2020;10:42. doi: 10.3390/polym15204142. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib14] 14.Li H., Zhang M., Zhang Y., Xu X., Zhao Y., Jiang X., et al. Characterization of cellulose-degrading bacteria isolated from silkworm excrement and optimization of its cellulase production. Polymers (Basel) 2023;15(20):4142. doi: 10.1186/s13568-020-00975-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib15] 15.Li Y., Wu Y., Liu Y., Li J., Du G., Lv X., et al. A genetic toolkit for efficient production of secretory protein in Bacillus subtilis. Bioresour Technol. 2022;363 doi: 10.1016/j.biortech.2022.127885. [DOI] [PubMed] [Google Scholar]

[bib16] 16.Liu Y., Cheng H., Li H., Zhang Y., Wang M. A programmable CRISPR/Cas9 toolkit improves lycopene production in Bacillus subtilis. Appl Environ Microbiol. 2023;89 doi: 10.1128/aem.00230-23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17.Watzlawick H., Altenbuchner J. Multiple integration of the gene ganA into the Bacillus subtilis chromosome for enhanced beta-galactosidase production using the CRISPR/Cas9 system. AMB Express. 2019;9:158. doi: 10.1186/s13568-019-0884-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib18] 18.Wang Y., Cheng H., Liu Y., Liu Y., Wen X., Zhang K., et al. In-situ generation of large numbers of genetic combinations for metabolic reprogramming via CRISPR-guided base editing. Nat Commun. 2021;12:678. doi: 10.1038/s41467-021-21003-y. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib19] 19.Altenbuchner J. Editing of the Bacillus subtilis genome by the CRISPR-Cas9 system. Appl Environ Microbiol. 2016;82:5421–5427. doi: 10.1128/AEM.01453-16. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib20] 20.Xiao J., Peng B., Su Z., Liu A., Hu Y., Nomura C.T., et al. Facilitating protein expression with portable 5’-UTR secondary structures in Bacillus licheniformis. ACS Synth Biol. 2020;9:1051–1058. doi: 10.1021/acssynbio.9b00355. [DOI] [PubMed] [Google Scholar]

[bib21] 21.Wang Z., Tang H., Li Y., Yang B., Liang X., Gong H., et al. Characterization and synergistic activity of heterologously expressed microbial-derived endoglucanase and bifunctional cellulase on wheat straw. Int J Biol Macromol. 2024;282 doi: 10.1016/j.ijbiomac.2024.137485. [DOI] [PubMed] [Google Scholar]

[bib22] 22.Liu G., Gong H., Tang H., Meng Z., Wang Z., Cui W., et al. Enhanced lignocellulose degradation in Bacillus subtilis RLI2019 through CRISPR/Cas9-mediated chromosomal integration of ternary cellulase genes. Int J Biol Macromol. 2025;306 doi: 10.1016/j.biortech.2023.129457. [DOI] [PubMed] [Google Scholar]

[bib23] 23.Liu G., Zhang K., Gong H., Yang K., Wang X., Zhou G., et al. Whole genome sequencing and the lignocellulose degradation potential of Bacillus subtilis RLI2019 isolated from the intestine of termites. Biotechnol Biofuels Bioprod. 2023;16:130. doi: 10.1186/s13068-023-02375-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24.Su W., Jiang Z., Wang C., Xu B., Lu Z., Wang F., et al. Dynamics of defatted rice bran in physicochemical characteristics, microbiota and metabolic functions during two-stage co-fermentation. Int J Food Microbiol. 2022;362 doi: 10.1016/j.ijfoodmicro.2021.109489. [DOI] [PubMed] [Google Scholar]

[bib25] 25.Pedrazzani C., Viscusi P., Fuso A., Caligiani A. Development of a protocol for fractionating and characterising fibres from lignocellulosic food waste. Food Chem X. 2024;22 doi: 10.1016/j.fochx.2024.101501. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26.Wang Z., Tang H., Liu G., Gong H., Li Y., Chen Y., et al. Compound probiotics producing cellulase could replace cellulase preparations during solid-state fermentation of millet bran. Bioresour Technol. 2023;385 doi: 10.1016/j.biortech.2023.129457. [DOI] [PubMed] [Google Scholar]

[bib27] 27.Wang Z., Tang H., Li Y., Tian L., Ye B., Yan W., et al. Evaluating the dynamic effects of complex probiotics as cellulase replacements during fermentation of apple pomace. Int J Food Microbiol. 2024;425 doi: 10.1016/j.ijfoodmicro.2024.110896. [DOI] [PubMed] [Google Scholar]

[bib28] 28.Zhang M., Zhen J., Teng J., Zhao X., Fu X., Song H., et al. N-Terminal sequences of signal peptides assuming critical roles in expression of heterologous proteins in Bacillus subtilis. Microorganisms. 2024;12:1275. doi: 10.3390/microorganisms12071275. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib29] 29.Tian J., Long X., Tian Y., Shi B. Enhanced extracellular recombinant keratinase activity in Bacillus subtilis SCK6 through signal peptide optimization and site-directed mutagenesis. RSC Adv. 2019;9:33337–33344. doi: 10.1039/c9ra07866e. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib30] 30.Song Y., Nikoloff J.M., Zhang D. Improving protein production on the level of regulation of both expression and secretion pathways in Bacillus subtilis. J Microbiol Biotechnol. 2015;25:963–977. doi: 10.4014/jmb.1501.01028. [DOI] [PubMed] [Google Scholar]

[bib31] 31.Sauer C., Syvertsson S., Bohorquez L.C., Cruz R., Harwood C.R., van Rij T., et al. Effect of genome position on heterologous gene expression in Bacillus subtilis: an unbiased analysis. ACS Synth Biol. 2016;5:942–947. doi: 10.1021/acssynbio.6b00065. [DOI] [PubMed] [Google Scholar]

[bib32] 32.Jeong D., So Y., Park S., Park S., Choi S. Random knock-in expression system for high yield production of heterologous protein in Bacillus subtilis. J Biotechnol. 2018;266:50–58. doi: 10.1016/j.jbiotec.2017.12.007. [DOI] [PubMed] [Google Scholar]

[bib33] 33.Ozturk S., Ergun B.G., Calik P. Double promoter expression systems for recombinant protein production by industrial microorganisms. Appl Microbiol Biotechnol. 2017;101:7459–7475. doi: 10.1007/s00253-017-8487-y. [DOI] [PubMed] [Google Scholar]

[bib34] 34.Ray-Soni A., Bellecourt M.J., Landick R. Mechanisms of bacterial transcription termination: all good things must end. Annu Rev Biochem. 2016;85:319–347. doi: 10.1146/annurev-biochem-060815-014844. [DOI] [PubMed] [Google Scholar]

[bib35] 35.Yu X., Lv H., Luo H., Zhu X., Wu J., Zhang K. High level food-grade expression of maltogenic amylase in Bacillus subtilis through genomic integration and comA overexpression. Int J Biol Macromol. 2025;309 doi: 10.1016/j.ijbiomac.2025.143060. [DOI] [PubMed] [Google Scholar]

[bib36] 36.Liu L., Ding D., Wang H., Ren X., Lee S.Y., Zhang D. Balancing cell growth and product synthesis for efficient microbial cell factories. Adv Sci (Weinh) 2025 doi: 10.1002/advs.202510649. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib37] 37.Wu Y., Li Y., Liu Y., Xiu X., Liu J., Zhang L., et al. Multiplexed in-situ mutagenesis driven by a dCas12a-based dual-function base editor. Nucleic Acids Res. 2024;52:4739–4755. doi: 10.1093/nar/gkae228. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] 38.Ejaz S., Khan H., Sarwar N., Aqeel S.M., Al-Adeeb A., Liu S. A review on recent advancement in expression strategies used in Bacillus subtilis. Protein Pept Lett. 2022;29:733–743. doi: 10.2174/0929866529666220803163335. [DOI] [PubMed] [Google Scholar]

[bib39] 39.Owji H., Nezafat N., Negahdaripour M., Hajiebrahimi A., Ghasemi Y. A comprehensive review of signal peptides: structure, roles, and applications. Eur J Cell Biol. 2018;97:422–441. doi: 10.1016/j.ejcb.2018.06.003. [DOI] [PubMed] [Google Scholar]

[bib40] 40.Wang B., Wu Y., Lv X., Liu L., Li J., Du G., et al. Synergistic regulation of chassis cell growth and screening of promoters, signal peptides and fusion protein linkers for enhanced recombinant protein expression in Bacillus subtilis. Int J Biol Macromol. 2024;280 doi: 10.1016/j.ijbiomac.2024.136037. [DOI] [PubMed] [Google Scholar]

[bib41] 41.Shi Z., Li J., Zhu F., Liu X., Zhou J., Li J., et al. Multidimensional strategies for efficient heterologous protein expression in Aspergillus niger. Biotechnol Adv. 2025;84 doi: 10.1016/j.biotechadv.2025.108690. [DOI] [PubMed] [Google Scholar]

[bib42] 42.Görke B., Stülke J. Carbon catabolite repression in bacteria: many ways to make the most out of nutrients. Nat Rev Microbiol. 2008;6:613–624. doi: 10.1038/nrmicro1932. [DOI] [PubMed] [Google Scholar]

[bib43] 43.Xiang M., Kang Q., Zhang D. Advances on systems metabolic engineering of Bacillus subtilis as a chassis cell. Synth Syst Biotechnol. 2020;5:245–251. doi: 10.1016/j.synbio.2020.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib44] 44.Chukwuma O.B., Rafatullah M., Tajarudin H.A., Ismail N. A review on bacterial contribution to lignocellulose breakdown into useful bio-products. Int J Environ Res Publ Health. 2021;18(11):6001. doi: 10.3390/ijerph18116001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib45] 45.Sun Y., Sun Q., Gao T., Ren S., Hu X. Competitive binding of TeXlnR and TeAmyR regulates cellulase production in Talaromyces endophyticus. Synth Syst Biotechnol. 2026;11:356–369. doi: 10.1016/j.synbio.2025.10.010. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib46] 46.Xu C., Tong S., Sun L., Gu X. Cellulase immobilization to enhance enzymatic hydrolysis of lignocellulosic biomass: an all-inclusive review. Carbohydr Polym. 2023;321 doi: 10.1016/j.carbpol.2023.121319. [DOI] [PubMed] [Google Scholar]

[bib47] 47.Stulke J., Gruppen A., Bramkamp M., Pelzer S. Bacillus subtilis, a swiss army knife in science and biotechnology. J Bacteriol. 2023;205 doi: 10.1128/jb.00102-23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib48] 48.Huang Z., Ni G., Zhao X., Wang F., Qu M. Characterization of a GH8 beta-1,4-glucanase from Bacillus subtilis B111 and its saccharification potential for agricultural straws. J Microbiol Biotechnol. 2021;31:1446–1454. doi: 10.4014/jmb.2105.05026. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib49] 49.Guo X., Sha Y., Lv W., Pu X., Liu X., Luo Y., et al. Sex differences in rumen fermentation and microbiota of Tibetan goat. Microb Cell Fact. 2022;21:55. doi: 10.1186/s12934-022-01783-8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib50] 50.Kamal M., Linlin K., Gao J., Xinrui Z., Xinming C., Haibo W., et al. Effects of Saccharomyces cerevisiae and Bacillus subtilis on in vitro fermentation in the rumen of Hu sheep. J Sci Food Agric. 2025;105:498–506. doi: 10.1002/jsfa.13848. [DOI] [PubMed] [Google Scholar]

[bib51] 51.Song P., Zhou F., Li F., Han Z., Wang L., Xu J., et al. Superfine pulverisation pretreatment to enhance crystallinity of cellulose from Lycium barbarum L. leaves. Carbohydr Polym. 2021;253 doi: 10.1016/j.carbpol.2020.117207. [DOI] [PubMed] [Google Scholar]

[bib52] 52.Astruc J., Nagalakshmaiah M., Laroche G., Grandbois M., Elkoun S., Robert M. Isolation of cellulose-II nanospheres from flax stems and their physical and morphological properties. Carbohydr Polym. 2017;178:352–359. doi: 10.1016/j.carbpol.2017.08.138. [DOI] [PubMed] [Google Scholar]

[bib53] 53.Peterson R., Nevalainen H. Trichoderma reesei RUT-C30 – thirty years of strain improvement. Microbiology (Read) 2012;158:58–68. doi: 10.1099/mic.0.054031-0. [DOI] [PubMed] [Google Scholar]

PERMALINK

Engineering of multiple modules to enhance lignocellulose degradation ability in Bacillus subtilis using CRISPR/Cas9 system

Zhiwei Wang

Gongwei Liu

Zhongming Meng

Jie Xu

Haoran Tang

Shendong Wang

Caixia Zou

Caiying Ma

Yuxin Yang

Abstract

Graphical abstract

1. Introduction

2. Materials and methods

2.1. Strains, plasmids, culture conditions, and reagents

Table 1.

2.2. Construction of gene expression vectors and recombinant strains

Fig. 1.

2.3. Expression and optimization of endoglucanase

2.4. Expression and optimization of bifunctional cellulase

2.5. Co-expression and optimization of complex cellulase

2.6. Determination of bacterial growth curve

2.7. Determination of enzyme activity and total protein concentration

2.8. Analysis of wheat straw enzymatic digestion

2.9. Data analysis

3. Results and discussion

3.1. Expression and optimization of endoglucanase at a single locus

Fig. 2.

3.2. Expression and optimization of endoglucanase at two loci

3.3. Expression and optimization of bifunctional cellulase at a single locus

Fig. 3.

3.4. Expression and optimization of bifunctional cellulase at two loci

3.5. Expression and optimization of complex cellulase

Fig. 4.

3.6. Effects of recombinant strain fermentation on physicochemical properties of wheat straw

Fig. 5.

3.7. Effect of recombinant strain fermentation on the microscopic structure of wheat straw

Fig. 6.

4. Conclusion

CRediT authorship contribution statement

Declaration of competing interest

Acknowledgments

Footnotes

Appendix A. Supplementary data

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases