Rewiring Amino Acid Flux for Efficient Transglutaminase Production in Streptomyces mobaraensis

Abstract

Transglutaminase (TGase) from Streptomyces mobaraensis is widely used in food processing to improve the texture properties of food products, but its broader application is constrained by low titers. Herein, through a quantitative analysis of the intracellular amino acid pool, we identified an insufficient supply of several amino acids as the primary bottleneck for TGase overproduction. To address this limitation, we overexpressed the biosynthetic pathways of seven critically insufficient amino acids, which increased the TGase titer by 46.3% (from 41.0 to 60.0 U/mL). Additionally, repressing the expression of nonessential high-abundance proteins reduced the metabolic competition for amino acids, boosting the TGase titer by 57.5% (from 40.0 to 63.0 U/mL). Combining both strategies yielded a maximum titer of 66.9 U/mL, which is a 62.4% improvement over the starting strain. This study provides a foundation for efficient TGase production and highlights precursor flux manipulation as a general strategy for enzyme overproduction.

1. Introduction

Industrial enzymes, such as cellulase, lipase, and pectinase, have found widespread applications in diverse chemical processes to enhance the product performance because of their high catalytic efficiency and specificity. − These enzymes are widely distributed in animals, plants, and microorganisms, with those from microorganisms being particularly advantageous owing to their broad substrate specificity, stability, ease of production, and scalability. −

Microbial transglutaminase (EC 2.3.2.13, TGase), one of the most important industrial enzymes, catalyzes protein cross-linking through acyl-transfer reaction between the γ-amide group of glutamine residues and ε-amino group of lysine residues. The unique catalytic properties of microbial TGase have enabled its extensive use in modifying the functional properties of meat, soy, dairy, and wheat products over the past four decades. − In recent years, TGase has also exhibited potential applications in the pharmaceutical, textile, and leather industries. , Commercialized TGase is produced on a large scale by Streptomyces mobaraensis (formerly named Streptoverticillium mobaraensis), the only TGase-producing species approved by the U.S. Food and Drug Administration (FDA). − However, the industrial application of TGase is hampered by its low titer and high cost. Therefore, engineering microbial cell factories is essential to achieve high-yielding and cost-effective production of TGase.

Conventional strategies for enhancing enzyme production include random mutagenesis via UV irradiation (Kumar, 2015), chemical mutagenesis, and atmospheric and room-temperature plasma (ARTP) (Jiang et al., 2017; Zhang et al., 2022). In combination with conventional or high-throughput screening techniques, these strategies have resulted in enzyme-producing strains with high titers. , Another prevalent strategy involves the optimization of the fermentation process. Solid-state fermentation of filamentous fungi and submerged fermentation of bacteria and yeasts are two widely used approaches for enzyme production. , Several key fermentation conditions have been demonstrated to significantly affect enzyme production, such as various medium compositions, , inducers, , temperature, pH, agitation rates, , and non-nutritional stress. , These parameters were optimized to enhance enzyme titers.

In general, metabolic engineering strategies tend to be more rational and efficient compared to the aforementioned conventional methods. Advances in genetic engineering and omics technologies have enabled the genetic modification of microbial hosts for enzyme production through multiple strategies: (i) increasing the transcription and translation via codon usage optimization, strong promoter replacement, multicopy gene integration, translation factor overexpression, and optimization of enzyme activation processes; − (ii) enhancing secretion efficiency through signal peptide modification and overexpression of Sec/Tat pathways; , (iii) modulating the growth rate and mycelial morphology by ssgA/adpA overexpression and matAB deletion; , (iv) regulating gene expression by manipulating regulatory genes or networks; − and (v) heterologous expression in alternative hosts. , Although these strategies have resulted in remarkable improvements in enzyme titers, such production levels often remain insufficient for industrial-scale applications.

Amino acids serve as the building blocks for protein synthesis, making their adequate supply crucial for efficient enzyme production. Previous studies have focused on the supplementation of exogenous amino acids in a fermentation medium to enhance protein production. − However, this approach has contributed less to precursor supply and enzyme production due to its high cost and risk of inducing ketoacidosis. Consequently, manipulating the internal amino acid flux through systematic metabolic engineering represents a more promising alternative. Nevertheless, this strategy has remained relatively underexplored.

Our study developed a metabolic engineering approach to enhance TGase titer by optimizing the amino acid flux. Through quantitative analysis of the intracellular amino acid pool in S. mobaraensis IPIO during fermentation, we initially identified the limiting amino acids for TGase biosynthesis. To relieve this precursor supply bottleneck, we implemented systematic metabolic engineering strategies to enhance the amino acid flux toward TGase biosynthesis. Specifically, we overexpressed the biosynthetic pathways of limiting amino acids and attenuated the biosynthesis of high-abundance, nonessential proteins. This coordinated enhancement of the amino acid flux significantly boosted the TGase titer.

2. Materials and Methods

2.1. Bacterial Strains, Plasmids, and Culture Conditions

All strains and plasmids used in this study are listed in Table S1. The oligonucleotide primers used in this study are listed in Table S2. The starting strain S. mobaraensis IPIO and its derivative strains were grown on Gauze’s Medium #1 agar plates for sporulation and ISP4 agar plates supplemented with 20 mM MgCl₂ for conjugation, both incubated at 30 °C. For genomic DNA extraction, S. mobaraensis IPIO was cultivated in the TSBY medium (3% tryptic soy broth, 10.3% sucrose, and 0.5% yeast extract) for 24–36 h at 30 °C. Escherichia coli strains were cultured in the Luria–Bertani (LB) liquid medium or on LB agar plates at 37 °C.

2.2. Determination of the Concentrations of Amino Acids in S. mobaraensis IPIO

S. mobaraensis IPIO was first cultured in 30 mL of seed medium (2% glycerol, 0.6% yeast (Angel Yeast Co., Ltd., China), 2.5% peptone from fish, 0.2% MgSO₄·7H₂O, 0.2% K₂HPO₄·3H₂O, pH 7.4) at 30 °C for 24 h. Subsequently, 3 mL of the seed culture was transferred to 30 mL of fermentation medium and incubated at 30 °C. One milliliter of the culture broths at 16, 22, and 28 h was centrifuged at 13,800g for 5 min. The supernatants were mixed with an equal volume of HCl/TCA solution (0.1 M hydrochloric acid/10% trichloroacetic acid, 1:1, v/v) to precipitate proteins and extract extracellular amino acids. Meanwhile, the harvested mycelia were washed twice with distilled water and resuspended in 1 mL of HCl/TCA solution. The cell suspensions were disrupted by sonication on ice for 10 min and then centrifuged at 13,800g for 20 min. The supernatants were injected into the automatic amino acid analyzer (Hitachi L-8900–1, Japan) for quantitative detection of amino acids.

For the derivative strains, the concentrations of the intracellular amino acids were determined by collecting 1 mL of the culture broth at 22 h during fermentation and processing as described above.

2.3. Construction of Strains Overexpressing Amino Acid Biosynthetic Genes

Taking the construction of mutant WD15 as an example, we amplified the proline biosynthetic genes (SMDS_4329–4331) and their native promoter P ₄₃₂₉ from the genomic DNA of S. mobaraensis IPIO using primers pro-F and pro-R. After digestion with XbaI/EcoRI, DNA fragments were inserted into the integrative vector pSET152 between XbaI and EcoRI and verified by sequencing. The constructed plasmid was first transformed into E. coli ET12567/pUZ8002 and then introduced into S. mobaraensis IPIO via intergeneric conjugation. Apramycin-resistant exconjugants were selected and verified by PCR amplification with primers AA-YZ-F and Pro-YZ-R. Other overexpression strains were constructed in a similar manner.

2.4. Construction of the Gene Repression Strains

The gene repression strains were constructed using a CRISPRi toolkit. The CRISPRi plasmids, which contained one or more sgRNA cassettes, were constructed according to the previously reported protocol. The resulting plasmid was then transformed into E. coli ET12567/pUZ8002 and subsequently introduced into S. mobaraensis IPIO via intergeneric conjugation. Apramycin-resistant exconjugants were selected and confirmed by PCR using primers dCas9-YZ-F and dCas9-YZ-R.

2.5. Fermentation and Analysis of TGase Production

IPIO and its derivative strains were cultured on Gauze’s Medium #1 agar plates at 30 °C for 7–8 days for sporulation. Spores harvested from one plate (approximately 10⁸ CFU) were inoculated into 30 mL of seed medium and incubated for 24 h on a rotary shaker (30 °C and 220 r.p.m.). Then, 3 mL of the seed culture was transferred into 30 mL of fermentation medium and cultivated for 28–32 h on a rotary shaker (30 °C and 220 r.p.m.), until a decrease in TGase activity was observed.

TGase activity was measured using the Grossowicz colorimetric method. TGase catalyzes the reaction between the substrates N-carbobenzoxy-glutaminyl-glycine (N-α-CBZ-Gln-Gly) and hydroxylamine to form l-glutamic acid γ-monohydroxamate. This product forms a red complex with ferric chloride in TCA (FeCl₃·TCA), the absorbance of which can be measured at 525 nm. The reaction was initiated by mixing 200 μL of TGase solution (prepared by diluting 10 μL culture supernatant with 190 μL water) at the indicated time points with 2 mL of solution A (30 mM N-α-CBZ-Gln-Gly, 100 mM hydroxylamine hydrochloride, 10 mM reduced glutathione, and 200 mM Tris–HCl, pH 6.0). After incubation at 37 °C for 10 min, the reaction was terminated by adding 2 mL of solution B (3 M HCl, 12% (w/v) TCA, and 3% (w/v) FeCl₃, all components mixed in equal volume). The reaction products were quantified based on the absorbance of the reaction mixture at 525 nm. One unit of TGase activity was defined as the amount of enzyme required to produce 1 μmol of γ-monohydroxamate per minute at 37 °C.

TGase content was analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) on a 12% separating gel. The fermentation broth from strains at 32 h was collected and centrifuged, and equal volumes of the supernatant were subjected to SDS-PAGE. The band corresponding to TGase (37.0 kDa) was visualized after staining with Coomassie Brilliant Blue R250, allowing for a comparison of TGase content between the starting and derivative strains.

2.6. Biomass Determination

The biomass of S. mobaraensis strains was determined by measuring the dry cell weight (DCW) at specified time points. One milliliter of culture broth was centrifuged at 13,800g for 2 min and washed twice with distilled water to remove residual medium solutes. The washed cell pellets were dried at 70 °C until a constant weight was achieved.

2.7. RNA Extraction and Quantitative Real-Time PCR (qRT-PCR)

S. mobaraensis mycelia were collected at 22 h during fermentation for RNA extraction using a total RNA extraction kit (SBS Genetech, China). The RNA concentration and purity were measured using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, USA). The first-strand cDNA was synthesized from 1 μg of total RNA using the RevertAid H Minus first-strand cDNA synthesis kits (Thermo Fisher Scientific, USA). The resulting cDNA served as the template for qRT-PCR with primers (listed in Table S2) using TB Green Premix Ex Taq (TaKaRa, Japan). The housekeeping gene hrdB was used as an internal reference for normalization, and all assays were performed in triplicate.

3. Results

3.1. Insufficient Supplies of Amino Acids Hampered the Biosynthesis of TGase

Since the amino acid compositions of S. mobaraensis IPIO and TGase are consistent, the availability of the intracellular amino acid pool has an impact on TGase biosynthesis (Figure S1a). To assess the balance between amino acid supply and demand for TGase biosynthesis, the concentrations of the intracellular amino acid pool in IPIO at 16 h (the early phase of TGase production), 22 h (the exponential phase of TGase production), and 28 h (the stationary phase of TGase production) during fermentation (Figures S1b and S8) were measured and compared with the theoretical amino acid composition of TGase. The measured mass percentages of intracellular amino acids were generally lower than their corresponding compositions in the TGase protein, indicating a shortage in the supply of amino acids, including cysteine, tryptophan, methionine, histidine, tyrosine, phenylalanine, proline, aspartate, arginine, and serine (Figure ). Specifically, among these insufficient amino acids, arginine, aspartate, proline, serine, tyrosine, and phenylalanine accounted for a relatively high proportion (greater than 5%) in the amino acid composition of TGase (Figure a). Further analysis revealed that the deficiency ratios of arginine, aspartate, proline, threonine, tyrosine, histidine, methionine, tryptophan, and cysteine were lower than −50% (Figure b). Combining their contribution to TGase synthesis and deficiency ratio, arginine, aspartate, proline, tyrosine, phenylalanine, histidine, methionine, threonine, and tryptophan were identified as the critical limiting amino acids. These results strongly suggested that an insufficient supply of these amino acids constitutes a critical metabolic bottleneck in the overproduction of TGase in S. mobaraensis. Therefore, enhancing the endogenous supply of these limiting amino acids becomes a potential tactic for achieving high-level production of TGase.

1.

Discovery and identification of limiting amino acids in S. mobaraensis IPIO through quantitative analysis of the intracellular amino acid pool. (a) Comparison of the relative intracellular concentrations of amino acids at 22 h fermentation (A, red bars) with the theoretical amino acid composition of TGase (T, purple bars). A was the ratio of each amino acid mass content (mg/g DCW) and the total mass content of all detected amino acids, expressed as a mass percentage. T was calculated based on the mass percentage of each amino acid in the TGase protein, as derived from its amino acid sequence. Amino acids contributing >5% to the TGase composition (black dashed line) are highlighted, as these are considered major contributors to TGase biosynthesis. (b) Amino acid deficiency ratio, which is calculated as (A–T)/T. The deficiency ratio <0 indicates that the intracellular concentration is less than the theoretical demand and is defined as deficient.

3.2. Strengthening the Biosynthesis of Limiting Amino Acids to Enhance TGase Titer

To enhance the TGase titer in S. mobaraensis IPIO, we focused on increasing the intracellular pool of limiting amino acids. The candidate limiting amino acids, including arginine, the aspartate family (aspartate, threonine, and methionine), proline, histidine, and the aromatic amino acids (phenylalanine, tyrosine, and tryptophan), were selected for overexpression. The corresponding biosynthetic genes of these amino acids in the S. mobaraensis IPIO genome were overexpressed under the control of their native promoters, generating engineered strains WD22, WD16, WD17, WD15, WD18, and WD19, respectively (Figure a and Table S1). The control strain IPIO:pSET152 carrying vector pSET152 was also constructed. Compared with the control IPIO:pSET152, the TGase activities of strains WD22, WD16, WD17, WD15, WD18, and WD19 were increased by 12.7, 17.2, 13.0, 20.4, 34.0, and 39.8%, respectively (Figure b). No significant difference in cell growth was observed among these strains during fermentation (Figure S2a). Further analysis of the fermentation supernatants via SDS-PAGE confirmed the increased TGase titer, as shown by more intense TGase bands in the overexpression strains than in the control, consistent with the results of TGase activity assays (Figure S2b).

2.

Enhanced endogenous supply of limiting amino acids through overexpression of their biosynthetic pathways. (a) Schematic overview of biosynthetic pathways in S. mobaraensis IPIO that were individually overexpressed to strengthen the biosynthesis of limiting amino acids. Genes argCJBDF and argGH were overexpressed to enhance arginine biosynthesis (yielding WD22); two aspartate aminotransferase (AspC)-encoding genes, aspC, were individually overexpressed to enhance aspartate biosynthesis (yielding WD16 and WD17); proBAC were overexpressed to enhance proline biosynthesis (yielding WD15); hisDCBNHAF were overexpressed to enhance histidine biosynthesis (yielding WD18); and aroECKBQ were overexpressed to enhance phenylalanine, tyrosine, and tryptophan biosynthesis (yielding WD19). TCA cycle: tricarboxylic acid cycle; G6P: glucose 6-phosphate; PEP: phosphoenolpyruvate; α-KG: α-ketoglutarate; OAA: oxaloacetate. (b, c) TGase activity at 32 h fermentation (b) and the relative intracellular concentrations of the corresponding amino acids at 22 h fermentation (c) in strains IPIO:pSET152 (as control), WD22, WD16, WD17, WD15, WD18, and WD19. (d) Comparisons of the intracellular amino acid pools between IPIO:pSET152 (gray) and WD22 (blue), WD15 (green), WD18 (orange), and WD19 (purple). (e) TGase activity of IPIO:pSET152, WD18, WD19, and WD28 (co-overexpression of hisDCBNHAF and aroECKBQ in IPIO). Relative amino acid concentration was the ratio of each amino acid concentration and DCW, and normalized to the level of the control strain IPIO:pSET152, which was set to 1.0. Statistical significance was determined by two-tailed paired t tests. Error bars represent mean ± SD (n = 3 biological replicates). *p < 0.05; **p < 0.01; ***p < 0.001; ns: not significant.

To validate the improved amino acid supply in these engineered strains, the intracellular concentrations of these amino acids were measured at 22 h of fermentation. As observed in Figure c, the intracellular concentrations of amino acids were markedly altered in the overexpression strains compared to the control, with most amino acids showing substantial increases. As expected, the concentrations of the targeted amino acids were significantly elevated in their respective overexpression strains, indicating a direct correlation between these amino acid increases and the observed improvements in the TGase titer (Figure d).

To further improve the TGase titer, the gene cassettes for histidine and aromatic amino acid biosynthesis were co-overexpressed in IPIO, yielding a strain WD28. Compared to strain WD19 with the aromatic amino acid overexpressed, WD28 exhibited a higher titer, from 57.0 to 60.0 U/mL. Therefore, the combined overexpression of limiting amino acid biosynthetic pathways further increased the supply of these amino acids and thereby enhanced the TGase titer (Figure e).

3.3. Suppression of Competing Protein Biosynthesis to Rewire Amino Acid Flux toward TGase Production

The presence of numerous intracellular and extracellular proteins in S. mobaraensis IPIO indicated a robust translation and intense competition for amino acid flux, which might limit the metabolic flux available for TGase biosynthesis and ultimately impede TGase production (Figure S3). Reducing their utilization in the biosynthesis of these proteins to redirect amino acid flux toward TGase production could be another strategy to enhance the TGase titer.

First, quantitative proteomic analysis was conducted to identify highly expressed proteins that compete for amino acids and pinpoint potential engineering targets. The abundances of 157 extracellular proteins and 2583 intracellular proteins were determined in the total proteome (Table S3). Notably, TGase itself accounted for only 0.8% of the total detected protein abundance, indicating that a significant fraction of amino acid flux was diverted toward the biosynthesis of other high-abundance proteins, leading to a substantial loss of amino acids from TGase production.

To prioritize potential engineering targets, we focused on the top 10 most abundant extracellular proteins and the top 20 most abundant intracellular proteins. Among these, five candidates were predicted to be nonessential or functionally redundant by the Swiss-Prot and antiSMASH databases, including the extracellular SMDS_111 (chitosanase), SMDS_5177 (TGase-activating metalloprotease inhibitor), and SMDS_6427 (Dispase autolysis-inducing protein), as well as the intracellular SMDS_758 (NRPS condensation domain-containing protein) and SMDS_5989 (type I polyketide β-ketoacyl synthase) (Table ).

1. Intracellular and Extracellular Nonessential Proteins with High Abundance Identified through Proteomic Analysis of S. mobaraensis IPIO Collected at 22 h Fermentation.

	accession	normalized abundance	annotated function
extracellular	SMDS_111	113.188	chitosanase
	SMDS_6427	108.317	dispase autolysis-inducing protein
	SMDS_5177	104.896	transglutaminase-activating metalloprotease inhibitor
intracellular	SMDS_758	269.992	NRPS condensation domain-containing protein
	SMDS_5989	248.453	type I polyketide β-ketoacyl synthase

We have established a CRISPRi system for gene repression in IPIO (Figure S4). To validate the effect of repressing these candidate nonessential genes on TGase production, two different gRNA (guide RNA) were designed for the repression of three extracellular protein-coding genes: gRNA-P targeting the promoter regions and gRNA-NT targeting the nontemplate (NT) strand within the 5′ coding region (within 100 bp from the start codon ATG). This dual-targeting approach was implemented because sgRNA binding location significantly influences the repression efficiency by affecting RNA polymerase binding and early transcriptional elongation (Figure S4). , A total of six sgRNA (synthetic guide RNA) cassettes were individually cloned into the vector pSET-dCas9/sgRNA and then introduced into IPIO. The qRT-PCR analysis first verified their successful repression in the mutants (Figures S11 and S12). Expression repression of SMDS_5177 and SMDS_6427 resulted in a significant increase in TGase titer and productivity (the ratio of titers of TGase and DCW) via targeting either the promoter or coding region (Figures a, S5a, and S11). In contrast, the repression of SMDS_111 expression did not enhance TGase titer (Figure a).

3.

Repression of high-abundance nonessential proteins using the CRISPRi system. (a, b) CRISPRi-guided repression of high-abundance proteins. (a) TGase activity at 32 h fermentation in strains IPIO:pSET-dCas9 (as control), WBH03 (repression of SMDS_5177 by targeting its promoter in IPIO), WBH04 (repression of SMDS_5177 by targeting its coding region in IPIO), WBH08 (repression of SMDS_6427 by targeting its promoter in IPIO), WBH09 (repression of SMDS_6427 by targeting its coding region in IPIO), WBH10 (repression of SMDS_111 by targeting its promoter in IPIO), and WBH11 (repression of SMDS_111 by targeting its coding region in IPIO) at 32 h fermentation. The sites of gRNA-P (orange) targeted the template (T) or nontemplate (NT) strands of the promoter, and gRNA-NT (green) targeted the nontemplate strand of the coding region of SMDS_5177, SMDS_6427, and SMDS_111. (b) TGase activity at 32 h fermentation in strains IPIO:pSET-dCas9 (as control), WBH12 (repression of SMDS_759 operon by targeting its promoter in IPIO), and WBH07 (repression of SMDS_5988 operon by targeting its promoter in IPIO). The sites of gRNA-P (orange) targeted the template or nontemplate strands of the promoter of SMDS_759 and SMDS_5988. (c) Randomly combined repression of two genes, three genes, or four target genes in IPIO. TGase activity at 32 h fermentation in strains WBH14 (repression of SMDS_5177 and SMDS_5988), WBH15 (repression of SMDS_759 and SMDS_5988), WBH16 (repression of SMDS_5177 and SMDS_759), WBH17 (repression of SMDS_6427 and SMDS_759), WBH18 (repression of SMDS_5177 and SMDS_6427), WBH19 (repression of SMDS_6427 and SMDS_5988), and WBH20 (repression of SMDS_5177, SMDS_6427, and SMDS_5988), and WBH21 (repression of SMDS_5177, SMDS_6427, SMDS_5988, and SMDS_759). Statistical significance was determined by two-tailed paired t tests. Error bars represent mean ± SD (n = 2 or 3 biological replicates). *p < 0.05; **p < 0.01; ***p < 0.001; ns: not significant. (d) Relative concentrations of the intracellular amino acids in strains IPIO:pSET-dCas9 (gray circle), WBH03 (orange rhombus), WBH07 (green rhombus), and WBH14 (blue circle). Relative amino acid concentration was the ratio of each amino acid concentration and DCW, and normalized to the level of the control strain IPIO:pSET-dCas9, which was set to 1.0.

For the intracellular targets, SMDS_758 and SMDS_5989 are predicted to be involved in secondary metabolite biosynthesis and are coexpressed with their respective gene cluster proteins SMDS_759 and SMDS_5988 (Figure S6). Therefore, gRNA-Ps targeting the promoter region were designed for the repression of the SMDS_759 and SMDS_5988 operons, respectively. The SMDS_759 operon-repressing strain WBH12 and the SMDS_5988 operon-repressing strain WBH07 showed an increase in TGase titer by 47.2 and 38.5%, respectively, and an increase in productivity by 41.4 and 36.6%, respectively, compared to the control strain IPIO:pSET-dCas9 (Figures b, S5b, and S11). No significant difference in cell growth was observed among these strains during fermentation (Figure S9).

Since a significant improvement in TGase titer has been achieved by repressing the expression of SMDS_5177, SMDS_6427, SMDS_758, and SMDS_5989, combined repression was designed to further minimize the amino acid competition. CRISPRi plasmids targeting two, three, or four targets were constructed, in which each sgRNA cassette targeted the promoter region, and these plasmids were individually introduced into IPIO. In total, eight strains of combined gene repression were generated. Notably, dual repression of the SMDS_5988 operon and SMDS_5177 (strain WBH14) and dual repression of the SMDS_5988 operon and SMDS_6427 (strain WBH19) resulted in a synergistic enhancement of TGase titer: 16.1% higher than that of the SMDS_5988 operon-repressing strain WBH07 and 60.8% higher than that of the control strain IPIO:pSET-dCas9 (Figure c). The productivity was also consistently increased, as cell growth was generally unaffected (Figures S9 and S10). However, more complex combinations involving simultaneous repression of three or four targets did not yield a further increase in TGase titer and sometimes even reduced it below the level of the best dual-gene repression (Figure C). Intracellular amino acid pool analysis of selected strains WBH03, WBH14, and IPIO:pSET-dCas9 confirmed that the expression repression of these proteins led to the accumulation of most amino acids (Figure d), thereby enriching the precursor pool available for TGase biosynthesis.

3.4. Improving TGase Titer by Integrating Both Effective Strategies

To maximize the supply of amino acids for TGase production, the most effective cassettes from these two strategies were integrated to engineer higher-yield strains. The initial plasmids pSET152 and pSET-dCas9/sgRNA are incompatible in IPIO because they both utilize the same φC31-attB/attP integration site and apramycin resistance selection marker. To enable their co-integration, the dCas9 and sgRNA cassettes from pSET-dCas9/sgRNA were inserted into plasmid pLQ2120, which contains a φBT1-attP site, a φBT1 integrase gene, and the thiostrepton resistance gene tsr, generating a new CRISPRi vector pRT-dCas9/sgRNA (Figure S7).

Subsequently, four single-repression plasmids were constructed based on pRT-dCas9/sgRNA and then individually introduced into WD28, which overproduced aromatic amino acids and histidine. As shown in Figures and S10, four resulting strains WBH21-WBH24 exhibited further improvement in TGase titer and productivity. Additionally, two combinatorial repression plasmids, pRT-dCas9/sgRNA _{SMDS_6427} -sgRNA _{SMDS_5988} and pRT-dCas9/sgRNA _{SMDS_5177} -sgRNA _{SMDS_5988} , were also introduced into the strain WD28, generating the final engineered strains WBH25 and WBH26, respectively. The final engineered strain WBH26 achieved the highest TGase titer of 66.9 U/mL, representing a 62.6% increase compared to the original control strain IPIO:pSET152 and a 16.0% increase compared to strain WD28:pRT-dCas9 (Figure ).

4.

Integration of two effective engineering strategies for enhanced TGase titer. TGase activity in IPIO:pSET152 (control), WD28:pRT-dCas9 (control), and WBH21 (repression of SMDS_5177 in WD28), WBH22 (repression of SMDS_6427 in WD28), WBH23 (repression of SMDS_759 in WD28), WBH24 (repression of SMDS_5988 in WD28), WBH25 (repression of SMDS_6427 and SMDS_5988 in WD28), and WBH26 (repression of SMDS_5177 and SMDS_5988 in WD28) at 32 h fermentation. Statistical significance was determined using two-tailed paired t tests. Error bars represent mean ± SD (n = 2 or 3 biological replicates). *p < 0.05; **p < 0.01; ***p < 0.001.

4. Discussion

Since the first reports on microbial TGase sources in the 1980s, enzymes produced by Streptoverticillium species have been identified. Among these, S. mobaraensis was approved as the sole commercial production strain on an industrial scale. ,, Owing to its unique properties and wide applications, the demand for microbial TGase is rising in the food industry. Engineering S. mobaraensis for TGase overproduction is of great significance for industrial development.

The increased precursor supply strategy is commonly employed in microbial strain engineering to boost the production of secondary metabolites. − A classic example is the overexpression of the accA2 gene (encoding a subunit of acetyl-CoA carboxylase) to engineer the malonyl-CoA supply, leading to a significant increase in polyketide antibiotic actinorhodin in Streptomyces coelicolor. In light of this, overexpressing genes involved in key precursor biosynthesis and diverting their use from other biosynthetic pathways represent effective strategies to enhance the production of targeted metabolites. However, this approach has rarely been applied to optimize the amino acid precursor flux for enhanced enzyme biosynthesis.

Our study establishes a comprehensive metabolic engineering framework for enhancing microbial enzyme production by systematically optimizing the amino acid flux. To compensate for the supply of limiting amino acids, we overexpressed their biosynthetic pathways, such as hisDCBNHAF, to increase the availability of corresponding amino acids, and most amino acids in the pool also showed a general increase. These global metabolic changes likely contributed to the significant improvement in the TGase titer observed in these strains (Figure b). To redirect amino acid flux toward TGase biosynthesis, we reduced the expression of the highly abundant nonessential proteins. The resultant mutants accumulated higher levels of most of the amino acids. These results demonstrated that, as expected, TGase titers were enhanced by overexpressing the biosynthetic cassettes of amino acids and blocking the biosynthesis of competing nonessential proteins, as amino acids are the direct building blocks of TGase.

However, some other combinatorial overexpression or repression failed to yield the anticipated increase in TGase titer, suggesting that imbalanced metabolic flux imposes a substantial metabolic burden, ultimately impairing cell growth and triggering regulatory dysfunction. , These findings highlight the necessity of employing more sophisticated synthetic biology strategies, such as fine-tuned control of gene expression or dynamic regulatory circuits, to optimize metabolic flux distribution while maintaining cellular fitness.

TGase is initially synthesized as an inactive prepro-TGase and secreted via the Tat pathway. , After secretion, the pro-TGase zymogen is activated through a two-step maturation process. , First, the N-terminal zymogen peptide (44 amino acids in S. mobaraensis) is cleaved by a transglutaminase-activating metalloprotease (TAMEP). Subsequently, the remaining tetrapeptide is removed by a tripeptidyl aminopeptidase that belongs to the serine protease family, yielding catalytically active mature TGase. Notably, SMDS_5177 is a transglutaminase-activating metalloprotease inhibitor. The TAMEP inhibitory protein that belongs to the large Streptomyces subtilisin inhibitor (SSI) family has been shown to inhibit metalloproteases. Therefore, we hypothesized that the repression of SMDS_5177 expression might relieve the inhibition of TAMEP-mediated mature TGase activation, thereby facilitating TGase maturation and production. SDS-PAGE analysis revealed that the extracellular pro-TGase and TGase level ratio in the two repression mutants was higher than that in the control strain (Figure S13a,b), indicating that the repression of SMDS_5177 expression did not significantly promote TGase activation. Moreover, WBH03 exhibited a 30% increase in the total amino acid pool concentration compared to the control (Figures S13c and d), confirming that the enhanced TGase titer results from improved amino acid availability rather than enhanced zymogen activation. These results also demonstrate that TGase secretion and activation are not limiting factors for TGase overproduction in S. mobaraensis IPIO.

To investigate whether the observed increase in TGase production results from global metabolic perturbations or changes in energy balance, we performed targeted metabolomic analysis of key pathways, including glycolysis, the tricarboxylic acid (TCA) cycle, and the pentose phosphate pathway. The results showed that most intermediates remained largely unchanged in the engineered strains (e.g., WD28, WBH14) compared to the control. Likewise, the NAD⁺ concentration was not significantly altered (Figure S14a). These findings indicate that overexpression of the amino acid biosynthetic pathways and repression of high-abundance protein expression did not induce global reprogramming of central carbon metabolism and substantial perturbation of cellular redox balance. Consequently, the enhanced production of TGase is likely to result from the increased amino acid pool (Figures d and d), rather than alterations in central carbon flux and energy metabolism. Thus, it can be more reasonably attributed to the specific engineering of the targeted biosynthetic pathway, supporting the validity and effectiveness of our strategy. Notably, however, GTP and ATP levels were significantly elevated (Figure S14b), which can be explained by reduced energy demand due to the repression of high-abundance protein expression, as well as relaxed (p)­ppGpp-mediated stringent response induced by enhanced amino acid biosynthesis. The resulting decrease in (p)­ppGpp levels alleviates the inhibition of purine nucleotide biosynthesis and promotes the transcription of genes encoding ribosomal proteins and amino acid biosynthesis enzymes, collectively contributing to the observed increase in GTP and ATP pools. ,

Proteomic analysis also revealed that SMDS_1309 and SMDS_6382, both associated with secondary metabolism, are highly expressed in S. mobaraensis IPIO (Table S3). Inactivating these genes or their respective operons may further enhance the TGase titer. Genome sequencing showed that S. mobaraensis IPIO harbors 33 putative secondary metabolite biosynthetic gene clusters (Table S4). These functional clusters not only compete for amino acids but also utilize other essential metabolic resources (e.g., organic acids, energy, and redox equivalents) during metabolite biosynthesis. , Consequently, further studies will prioritize tracking the availability of critical precursors, eliminating competing gene clusters, and overexpressing rate-limiting biosynthetic genes to achieve high production of TGase in S. mobaraensis IPIO.

To the best of our knowledge, the final enzyme activity of 66.9 U/mL represents the highest TGase yield ever achieved in shake-flask fermentation of S. mobaraensis, surpassing the enzyme activity levels obtained through both optimized fermentation and other genetic engineering approaches. Numerous studies have focused on optimizing the cultivation conditions or culture media to enhance TGase yields. − ,, For example, Kolotylo et al. optimized the fermentation medium composition, as well as inoculum incubation time and dosage, achieving the highest TGase activity at 6.59 U/mL. Zhang et al. reported a maximal TGase activity of 4.3 U/mL under MgCl₂ stress conditions, representing a 2-fold increase over the control (2.1 U/mL). Further scale-up fermentation and fermentation optimization remain our focus for further improvements in TGase production.

In summary, our study develops a coordinated “push-and-block” strategy to enhance TGase titer through rational metabolic engineering. This approach demonstrates potential applicability to various industrial enzyme-producing strains and provides a framework for more precise strain improvement. Future integration with a genome-scale metabolic model and AI-assisted analysis may further optimize precursor flux control and identify new engineering targets, potentially advancing microbial enzyme production technologies.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jafc.6c02313.

• Strains and plasmids used in this study (Table S1); primers used in this study (Table S2); information on extracellular and intracellular proteins with high abundance (Table S3); putative biosynthetic gene clusters of IPIO genome (Table S4); heatmap for amino acid compositions of TGase and IPIO genome (Figure S1a); time courses of cell growth and TGase production of S. mobaraensis IPIO (Figure S1b); strengthening the biosynthesis of limiting amino acids (Figure S2); SDS-PAGE analysis of the extracellular and intracellular proteins during fermentation (Figure S3); validation of the availability of CRISPRi system in S. mobaraensis IPIO (Figure S4); repressing the candidate nonessential proteins with high abundance (Figure S5); biosynthetic gene clusters containing the high-abundance proteins SMDS_758 and SMDS_5989, as well as the abundance of their associated cluster proteins (Figure S6); construction scheme of the CRISPRi plasmid pRT-dCas9/sgRNA (Figure S7); comparison of the extracellular or intracellular amino acid pools of IPIO with the theoretical ones of TGase (Figure S8); the biomass of the repression strains (Figure S9); calculation of the productivity of the repression strains (Figure S10); transcription analysis of target genes in IPIO:pSET-dCas9 and derived repression mutants (Figure S11); transcription analysis of SMDS_5177 and SMDS_5989 in WD28:pRT-dCas9 and WBH26 (Figure S12); TGase expression and amino acid pool in SMDS_5177-repression mutants (Figure S13); targeted metabolomic analysis of key metabolites in central carbon metabolism (Figure S14) (PDF)

†.

D.W. and B.W. contributed equally to this work. D.W.: conceptualization, investigation, methodology, data collection and analysis, writing of the original draft, and review and editing. B.W.: investigation, methodology, data collection and analysis, writing of the original draft. X.L.: methodology and review and editing. Y.Y.: methodology and data curation. Q.K.: methodology, data curation, and review and editing. L.B.: conceptualization, project administration, funding acquisition, and review and editing.

This work was financially supported by the Shanghai Municipal Science and Technology Major Project, Shanghai Committee of Science and Technology (grant No. 24HC2810200), and Dongsheng Biotech (Taixing) Co., Ltd. to L.B.

The authors have filed two patents (CN114181878A and CN115029290A) to protect the strategy invention and TGase production.

The authors declare no competing financial interest.