Correction for ‘Artificial intelligence-driven dynamic regulation for high-efficiency gentamicin C1a production’ by Feng Xu et al. , Green Chem. , 2025, 27 , 13436–13454, https://doi.org/10.1039/D5GC02507A.

Biosynthetic gene clusters (BGCs) encode the entire biosynthetic machinery for diverse natural products, which are critical sources of drugs and pesticides for decades. To date, a wealth of BGC engineering approaches have been developed to optimize the titers of target natural products. As one of the most extensively characterized and well-studied BGCs, the erythromycin BGC directs the biosynthesis of erythromycin, a clinically important macrolide antibiotic produced by Saccharopolyspora erythraea. In this study, strain E3-BGC1 harboring partially duplicated erythromycin BGC (SACE_0713-SACE_0726, 44 kb) that contained the complete polyketide synthase region was constructed by bacterial artificial chromosome (BAC) library construction. To alleviate cell autolysis and excessive accumulation of impurity erythromycin C in E3-BGC1, we first realized complete duplication of the erythromycin BGC (54 kb). Furthermore, the expression of eryG was precisely regulated via constitutive and inducible promoters within the complete duplicated BGC. Among all complete BGC-duplicated strains, E3-BGC7 with inducible promoter-regulated eryG exhibited the highest erythromycin A (Er-A) titer and displayed no significant metabolic burden. Otherwise, with ammonium sulfate feeding strategy adopted in complex fermentation medium, the fermentation duration was extended to 192 h, and the final Er-A titer reached 9,519.9 mg/L, representing a 70.9% increase relative to the parent strain. This constituted the highest Er-A titer ever reported. In short, the BGC precise regulation strategy established herein offers a valuable paradigm for enhancing natural product titers in alternative industrial microbial hosts.

The clinical importance of gentamicin C1a as a broad-spectrum aminoglycoside antibiotic underscores the need for efficient biomanufacturing strategies. In this study, we developed a systematic engineering framework to enhance gentamicin C1a production. First, a genome-scale metabolic model (iFX1172) was reconstructed to pinpoint critical bottlenecks in both regulatory and biosynthetic pathways. Guided by model predictions and experimental validation, we identified genC, metK, and BldD as synergistic targets. Coordinated overexpression of these genes increased gentamicin C1a titers to 198.1 mg/L, representing a 34.3% improvement over the parental strain, and also enhanced the titers of other aminoglycoside antibiotics by up to 1.6-fold, demonstrating the universality of the strategy. Metabolic flux analysis and targeted metabolomics revealed that redox homeostasis and ATP availability are pivotal for biosynthesis. Finally, process optimization in a fed-batch bioreactor using a Bayesian framework, coupled with in situ resin adsorption, yielded 964.1 mg/L gentamicin C1a with a yield of 24.1 mg/g glucose and a productivity of 6.7 mg/L/h.

Cephalosporin C (CPC)-derived antibiotics have played a vital role in improving human health and extending life expectancy. Acremonium chrysogenum remains the only microorganism capable of industrial-scale CPC production to date. However, the lack of efficient multiplex genome-editing tools has limited studies on its gene function, high-yield mechanisms as well as metabolic engineering. To overcome this limitation, a rapid and efficient CRISPR/Cas9-based multiplex genome-editing system was developed, driven by endogenous tRNA promoters, enabling one-step multilocus knockout, large-fragment DNA deletion, and gene overexpression in A. chrysogenum. Given that many strains lack visible phenotypes associated with specific genes, we introduced a visually distinguishable red phenotype by expressing the heterologous protein mCherry under a strong promoter. In the wild-type strain, 20 endogenous tRNA promoters were evaluated and compared to the heterologous Aspergillus nidulans PgpdA and Aspergillus fumigatus U6 promoters. The endogenous tRNAVal promoter showed the highest knockout efficiency (95.5%). The tRNA-gRNA array-based CRISPR/Cas9 system enabled double- and triple-site knockouts without donor DNA in industrial strain, with efficiencies of 50.0-83.3% and 14.3%, respectively. This is the first demonstration of simultaneous triple-site knockout in A. chrysogenum, especially in industrial strain. Using this system, we successfully deleted a 50.7-kb DNA fragment containing the sorbicillinoids biosynthetic gene cluster with nearly 100% efficiency and achieved overexpression of the key gene pcbAB involved in CPC biosynthesis in high-yield strain, increasing CPC titer from 5.59 g/L to 6.92 g/L with an improvement of 23.8%. Overall, this tRNA-gRNA array-based CRISPR/Cas9 multiplex gene-editing system provides an efficient and versatile platform for functional genomics and industrial strain engineering in A. chrysogenum.

β-Phenylethanol (2-PE) is a natural aroma compound used in food and cosmetic applications, and its efficient microbial production requires the coordinated control of glucose and precursor availability. Here, Raman and near-infrared (NIR) spectroscopy were implemented as soft sensors to guide glucose/l-phenylalanine (l-Phe) cofeeding during 2-PE biosynthesis by Saccharomyces cerevisiae. Raman-Artificial Neural Network and NIR-Partial Least Squares models showed the best performance (external validation R2 up to 0.99). A soft-sensing-guided glucose-feeding strategy maintained glucose below 20 g/L and triggered refeeding after ethanol depletion. Under these conditions, the optimal l-Phe supply range was found to be 5-15 g/L, yielding 16.72 g/L 2-PE and 0.66 g/g l-Phe-to-2-PE conversion yield. Transcriptomics at low, intermediate, and high l-Phe levels indicated that intermediate l-Phe promoted SPS-module/Ehrlich-pathway responses, whereas low or excessive l-Phe impaired nitrogen- and redox-related metabolism. These results support the process and strain optimization for 2-PE biomanufacturing.

The aromatic compound β-phenylethanol (2-PE) is inherently toxic and can inhibit cell activity in Saccharomyces cerevisiae, making it highly challenging to enhance strain tolerance through rational design due to the lack of reliable connections between tolerance phenotype and genetic loci. This study employed adaptive laboratory evolution strategy to investigate the tolerance characteristics of S. cerevisiae S288C under inhibitory concentrations of 2-PE. The tolerant mutant SEC4.0 was characterized through comprehensive analysis of whole genome sequence, transcriptome, and phosphoproteome. The findings revealed that the high resistance of SEC4.0 was not primarily due to large-scale transcriptional upregulation of stress response genes, but rather through alterations in the phosphorylation levels of lipid-related pathways. PKC1 mutations that affect stress signal transduction and SPT3 mutations that affect arginine biosynthesis have been shown to significantly enhance 2-PE resistance. This study also investigated the effects of exogenous amino acid addition and synergistic effects with two key mutanted genes on 2-PE resistance. This study provides a foundation for enhancing yeast tolerance to this aromatic compound through rational design strategies.

Cephalosporin C (CPC) is a natural product that serves as the key precursor for various semisynthetic cephalosporins. Its industrial production primarily relies on Acremonium chrysogenum. However, as the exclusive microbial strain used for large-scale CPC fermentation, A. chrysogenum faces challenges in metabolic engineering owing to the absence of an efficient multigene coexpression system. This study presented, for the first time, the development and application of a 2A peptide-based multigene coexpression system in A. chrysogenum. The self-cleavage efficiencies of ten virus-derived 2A peptides were systematically evaluated, ranging from 66.5% to 88.2%, with P2A showing the best performance. The most efficient P2A peptide was then selected, enabling both high- and low-level precalibrated expression based on the transcriptome data and supporting the coexpression of three genes in A. chrysogenum, including the driver gene. This system was subsequently applied to coexpress the key CPC biosynthetic genes, cefEF and cefG, with the driver gene of ACRE_076110, resulting in a significant 3.19-fold increase in CPC titer compared to the wild-type strain. Furthermore, the strategy was successfully extended to an industrial high-yield strain, with CPC titer increasing from 6.09 g/L to 7.45 g/L, receiving a 22.2% improvement. Overall, this study provides a valuable tool for metabolic engineering efforts aimed at enhancing CPC production in A. chrysogenum.

Butenyl-spinosyn is a high-quality biological insecticide produced by Saccharopolyspora pogona that effectively targets a broad range of insect pests. However, the large-scale production of this insecticide is hindered by its low yield. Herein, based on prior comparative genomic analysis, five mutations were individually overexpressed in aG6. Subsequently, the combinatorial overexpression of sp1322 (encoding NAD-glutamate dehydrogenase) and sp6746 (encoding dTDP-glucose 4,6-dehydratase) in aG6 resulted in strain O1322-6746. The production of butenyl-spinosyn in O1322-6746 was 77.1% higher than that in aG6. Comparative targeted metabolomic analysis uncovered that O1322-6746 exhibited increased metabolic flux toward butenyl-spinosyn precursors. Furthermore, single-factor experiments, Plackett-Burman analysis and response surface methodology were performed to optimize the fermentation medium for O1322-6746. Ultimately, butenyl-spinosyn production was enhanced to 298.5 mg/L in a 5-L bioreactor, marking the highest yield ever reported. This work demonstrated that combining metabolic engineering with medium optimization is an effective strategy to improve butenyl-spinosyn production.

beta-Phenylethanol (2-PE) is the second most widely used flavoring agent after vanillin. Saccharomyces cerevisiae possesses the ability to synthesize 2-PE via two distinct pathways: the Ehrlich pathway and the de novo synthesis pathway. While the conversion efficiencies of these pathways have been improved independently through metabolic engineering strategies, the strong inhibitory effect of L-phenylalanine (L-Phe), the substrate of Ehrlich pathway, limits their synergistic potential for enhancing 2-PE production. In this study, we observed that when the specific growth rate exceeded 0.236 h- 1, the strain exhibited a notable increase in flux through the de novo synthesis pathway. Meanwhile, the Ehrlich pathway maintained a high level of activity in the conversion of LPhe to 2-PE. This metabolic state has been defined as the "Coexisting" metabolism. However, enabling the occurrence of Coexisting metabolism at or below the Critical Specific Growth Rate (0.193 h- 1 in this study) is essential to minimize byproduct formation of ethanol. Metabolic flux analysis has indicated that insufficient intracellular NADH levels, along with significant diversion of flux through the L-tyrosine (L-Tyr) biosynthesis pathway, present major challenges. To identify potential regulatory targets, we utilized transcriptional and protein profiles. Furthermore, reverse metabolic engineering has validated TYR1, the first enzyme in the L-Tyr biosynthetic pathway, as a key regulatory node. By downregulating TYR1 and overexpressing IDH1 to enhance intracellular NADH supply, an engineered strain achieved a conversion rate of 115.2 % at mu = 0.193 h- 1 in chemostat cultivation, surpassing the theoretical maximum of the Ehrlich pathway and indicating the emergence of a significant Coexisting metabolic state. In Fed-Batch bioconversion, this strain showed a 33.2 % increase in conversion rate during the rapid synthesis period, and the final titer of 2-PE increased from 11.2 g/L to 12.1 g/L, representing an 8.0 % improvement. In short, this study provided a new, efficient and green approach to 2-PE production.

The industrial biosynthesis of complex antibiotics such as gentamicin C1a requires precise, dynamic control of microbial metabolism. In this study, an artificial intelligence (AI)-driven control framework was developed that integrated data-driven decision-making with real-time sensing to optimize the green production of gentamicin C1a. The system consists of four tightly coupled modules encompassing backpropagation neural network (BPNN)-based kinetic modeling, multi-objective optimization (NSGA-II), dual-spectroscopy monitoring (near-infrared and Raman), and closed-loop feedback control. The BPNN model accurately captured nonlinear correlations between specific substrate consumption rates, specific growth rates, and specific gentamicin C1a production rates, with R2 values of 0.9631, 0.9578, and 0.9689, respectively. By resolving phase-specific trade-offs in metabolic demands, the AI-driven dynamic regulation enabled real-time coordination between carbon, nitrogen, and oxygen supplementation as well as cellular requirements. Through the implementation of an AI-driven dynamic regulation system, gentamicin C1a production was significantly enhanced, achieving a titer of 430.5 mg L-1, a 75.7% improvement over traditional fed-batch fermentation. Notably, the gentamicin C1a yield reached 10.3 mg g-1 and the specific productivity attained 0.079 mg gDCW-1 h-1, both of which represented the highest levels reported to date. During the late fermentation phase, integrated metabolomics and metabolic flux analyses revealed a dynamic reorganization of the metabolic network, characterized by increased flux through the pentose phosphate pathway, enhanced NADPH generation and consumption, and improved carbon-nitrogen allocation favoring gentamicin C1a biosynthesis. Finally, the integrated techno-economic analysis and life cycle assessment confirm the commercial feasibility and greenhouse gas mitigation potential of gentamicin C1a production via AI-enhanced fermentation. These results demonstrate the power of AI-enhanced bioprocesses for intelligent fermentation control and establish a scalable, mechanistically informed strategy for the green industrial production of secondary metabolites.

Acremonium chrysogenum is the major industrial producer of cephalosporin C (CPC), which is used as raw material for the production of significant cephalosporin antibiotics. Due to the lack of diverse promoter elements, the development of metabolic engineering transformation is relatively slow, resulting in a limited improvement on CPC production. In this study, based on the analysis of the transcriptome profile, 27 candidate promoters were selected to drive the expression of the reporter genes. The promoter activities of this library ranged from 0.0075 to 101 times of the control promoter PAngpdA . Simultaneously, a rapid screening method for potential bidirectional promoters was developed and 4 strong bidirectional promoters from 27 candidate options were identified and validated. Finally, the Golden Gate method was employed to combine promoter modules from the library with various target genes. Through a mixed transformation and screening process, high-yielding strains AG-6, AG-18, and AG-41 were identified, exhibiting an increase in CPC production of 30%, 35%, and 29%, respectively, compared to the control strain Ac-∆axl2:: eGFP. Therefore, the utilization of this promoter library offers a broader range of synthetic biology toolkits for the genetic engineering transformation of A. chrysogenum, thus establishing a solid foundation for the precise regulation of gene expression.

β-Phenylethanol (2-PE), as an important flavor component in wine, is widely used in the fields of flavor chemistry and food health. 2-PE can be sustainably produced through Saccharomyces cerevisiae. Although significant progress has been made in obtaining high-yield strains, as well as improving the synthesis pathways of 2-PE, there still lies a gap between these two fields to unpin. In this study, the macroscopic metabolic characteristics of high-yield and low-yield 2-PE strains were systematically compared and analyzed. The results indicated that the production potential of the high-yield strain might be contributed to the enhancement of respiratory metabolism and the high tolerance to 2-PE. Furthermore, this hypothesis was confirmed through comparative genomics. Meanwhile, transcriptome analysis at key specific growth rates revealed that the collective upregulation of mitochondrial functional gene clusters plays a more prominent role in the production process of 2-PE. Finally, findings from untargeted metabolomics suggested that by enhancing respiratory metabolism and reducing the Crabtree effect, the accumulation of metabolites resisting high 2-PE stress was observed, such as intracellular amino acids and purines. Hence, this strategy provided a richer supply of precursors and cofactors, effectively promoting the synthesis of 2-PE. In short, this study provides a bridge for studying the metabolic mechanism of high-yield 2-PE strains with the subsequent targeted strengthening of relevant synthetic pathways. It also provides insights for the synthesis of nonalcoholic products in S. cerevisiae.

Porcine circovirus Type 2 (PCV2) is a primary etiological pathogen of post-weaning multi-systemic wasting syndrome (PMWS). The capsid protein of PCV2 is the crucial immunogenic protein which can induce antibody generation and immune responses. However, there is still a lack of efficient PCV2 vaccines with high immunogenicity. In the current study, we developed a novel engineered PCV2 capsid (∆1-41aa)-pFc fusion protein (PCFP), which comprised a truncated capsid protein of PCV2 and a porcine IgG Fc fragment, fused to the capsid protein of PCV2 at the C-terminus. We found that this novel fusion protein could auto-assemble into virus-like nanoparticles with an estimated mean diameter of 22.6 nm, characterized by transmission electron microscopy. Immunization of BALB/c mice with this fusion protein significantly increased the production levels of anti-PCV2-capsid protein antibody in serum. Besides, the virus-like nanoparticles, PCFP was demonstrated to induce efficient cellular immune responses in mice, as evident by the high specific T cell reactivity to the PCFP fusion protein and the high production of the immune cytokines IFN-γ and IL-10 in an ex vivo re-stimulation system. Collectively, these findings demonstrate that the PCV2 truncated capsid subunit Fc-fusion protein can induce both cellular and humoral immune responses, and it displays great application potential.

The main indicators for industrial production of high-quality lactic acid at elevated temperatures are high titer, productivity, yield, and optical purity. However, no such strains have been reported to meet all these requirements simultaneously. In this study, a high optical purity L-lactic acid producing strain is developed through the CRISPR-Cas9 gene editing platform. Further, adaptive evolution was used to breed and select a highperformance strain (NCBIO01-M2-ldhL1-HT) that could efficiently produce L-lactic acid at a high temperature of 45celcius. This strain produced 221.0 g/L of L-lactic acid in open fermentation with high initial glucose concentration. Also, L-lactic acid productivity and yield was above 7.5 g/L/h and 0.96 g/g respectively, as well as the optical purity of L-lactic acid in the fermentation broth exceeded 99.1%. In short, this breeding strain possess high potential to be considered for the commercial production of polymer-grade L-lactic acid.

Embryonic Sertoli cells (eSCs) possess multiple supporting functions and research value in gonadal development and sex determination. However, the limitation of acquiring quality eSCs had hindered the further application. Herein, we successfully derived non-genetically modified (non-GM)-induced embryonic Sertoli-like cells (eSLCs) from mouse embryonic stem cells (ESCs) with a TM4 cell-derived conditioned medium containing recombinant endogenous protein factors Sry, Sox9, Sf1, Wt1, Gata4, and Dmrt1. These eSLCs were determined through morphology; transcriptional expression levels of stage-specific, epithelial, and mesenchymal marker genes; flow cytometry, immunofluorescence; and immunocytochemistry and functionally determined by coculture with spermatogonia stem cells. Results indicated that these eSLCs performed similarly to eSCs in specific biomarkers and expression of marker genes and supported the maturation of spermatogonia. The study induced eSLCs from mouse ESCs by defined protein factors. However, the inducing efficiency of the non-GM method was still lower than that of the lentiviral transduction method. Thus, this work established a foundation for future production of non-GM eSLCs for clinical applications and fundamental theory research.

Propanol had been widely used as a precursor for erythromycin synthesis in industrial production. However, the knowledge on the exact metabolic fate of propanol was still unclear. In the present study, the metabolic fate of propanol in industrial erythromycin-producing strain Saccharopolyspora erythraea E3 was explored via 13C labeling experiments. An unexpected pathway in which propanol was channeled into tricarboxylic acid cycle was uncovered, resulting in uneconomic catabolism of propanol. By deleting the sucC gene, which encodes succinyl-CoA synthetase that catalyse a reaction in the unexpected propanol utilization pathway, a novel strain E3-ΔsucC was constructed. The strain E3-ΔsucC showed a significant enhancement in erythromycin production in the chemically defined medium compared to E3 (786.61 vs 392.94 mg/L). Isotopically nonstationary 13C metabolic flux analysis were employed to characterize the metabolic differences between Saccharopolyspora erythraea E3 and E3-ΔsucC. The results showed that compared with the starting strain E3, the fluxes of pentose phosphate pathway in E3-△sucC increased by almost 200%. The flux of the metabolic reaction catalyzed by succinyl-CoA synthetase in E3-ΔsucC was almost zero, while the glyoxylate bypass flux significantly increased. These new insights into the precursor utilization of antibiotic biosynthesis by rational metabolic engineering in Saccharopolyspora erythraea provided the new vision in increasing industrial production of secondary metabolites.

The stability and high productivity of heterogeneous terpenoid production in Escherichia coli expression system is one of the most key issues for its large scale industrialization. In the current study on taking lycopene biosynthesis as an example, an integrated Escherichia coli system has been generated successfully, which resulted into stable and high lycopene production. In this process, two modules of mevalonate (MVA) pathway and one module of lycopene expression pathway were completely integrated in the chromosome. Firstly, the copy number and integrated position of three modules of heterologous pathways were rationally optimized. Later, a strain DH416 equipped with heterogeneous expression pathways through chromosomal integration was efficiently derived from parental strain DH411. The evolving DH416 strain efficiently produced the lycopene level of 1.22 g/L (49.9 mg/g DCW) in a 5 L fermenter with mean productivity of 61.0 mg/L/h. Additionally, the integrated strain showed more genetic stability than the plasmid systems after successive 21st passage.

In this study we have established a rational and high performance high-throughput screening system to select for a sophorolipids (SLs) high-producing strain of Candida bombicola. Introduction of mutagen combination, relaxation culture and multi-stress significantly improved both mutation and positive mutation rates. A high-performing strain, Ncbio 5, was selected out of 6212 mutants. Final SLs titer, productivity and yield of this strain in 5 L bioreactors were 26.9%, 27.0% and 35.0% higher than in the original strain, respectively. The improved fermentation performance in Ncbio 5 is contributed by the longer production period, higher SLs productivity and more efficient oil utilization. The strategy adopted herein to optimize the high-throughput screening system should be readily extendable to other similar systems.

麦芽糖和葡萄糖对粪产碱杆菌发酵合成凝胶多糖有着显著的影响,为了详细分析两种底物对凝胶多糖合成的影响机制,利用恒化培养实验及稳态碳平衡代谢分析,研究发现在稀释速率为0.1h-1时,利用麦芽糖和葡萄糖为碳源底物的条件下粪产碱杆菌的微观代谢途径通量有较大的差异.以麦芽糖为底物时凝胶多糖的摩尔得率为53.8％,比葡萄糖为碳源时的摩尔得率(36.9％)高出了45.8％以上.同时以麦芽糖为碳源时HMP途径的绝对代谢通量比葡萄糖时的通量提升了40％以上.这条途径通量的增加,提升了NADPH还原力供给速率,促进了依赖于还原力NADPH的凝胶多糖合成途径通量,提升了碳源底物向产物的摩尔转化速率.而且代谢流分析结果显示ED途径通量和能量提供也是影响粪产碱杆菌凝胶多糖合成效率的关键因素.麦芽糖作为碳源底物过程中维持的较低的残留葡萄糖浓度解除了高葡萄糖浓度条件下对凝胶多糖合成的抑制,能够实现更高通量的ATP能量提供效率,更加促进了凝胶多糖合成通量.

Background Embryonic Sertoli cells (eSCs) have been known for playing important roles in male reproductive development system. In current studies, eSCs were mainly generated from induced intermediate mesoderm. The deriving mechanism of eSCs has been unclear so far. Therefore, this work was aimed to reveal the molecular pathways during derivation of eSCs. Methods In this scenario, a differentiation model from mouse embryonic stem cells (mESCs) to eSCs was established through spatiotemporal control of 5 key factors, Wilms tumor 1 homolog (Wt1), GATA binding protein 4 (Gata4), nuclear receptor subfamily 5, group A, member 1 (Nr5a1, i.e., Sf1), SRY (sex determining region Y)-box 9 (Sox9), doublesex, and mab-3 related transcription factor 1 (Dmrt1). To investigate the molecular mechanism, these key factors were respectively manipulated through a light-switchable (light-on) system, tetracycline-switchable (Tet-on) system, and CRISPR/Cas9 knock out (KO) system. Results Via the established approach, some embryonic Sertoli-like cells (eSLCs) were induced from mESCs and formed ring-like or tubular-like structures. The key factors were respectively manipulated and revealed their roles in the derivation of these eSLCs. Based on these results, some molecular pathways were mapped during the development of coelomic epithelial somatic cells to eSCs. Conclusions This differentiation model provided a high controllability of some key factors and brought a novel insight into the deriving mechanism of Sertoli cells.