Xiaomei Lyu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiaomei Lyu has a total of 100 co-authors. Here are the top co-authors by number of shared publications:
Ruijin Yang
Jiangnan University
14 shared publicationsChenlu Zhu
Jiangnan University
7 shared publications- Wei Zhao
Jiangnan University
7 shared publications - Ruijin Yang
State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, Jiangsu 214122, People's Republic of China; School of Food Science and Technology, Jiangnan University, Wuxi, Jiangsu 214122, People's Republic of China
6 shared publications Han Liu
Jiangnan University
5 shared publicationsQiuqian Zeng
School of Food Science and Technology, Jiangnan University, Wuxi, Jiangsu 214122, China
4 shared publications
Latest Publications
- 19 October 2026
- Published by Royal Society of Chemistry (RSC) in Food & Function
- Vol. 17 (20) , 9049-9061
Comparative genomic insights into strain-specific zinc-promoting effects of the spore-forming probiotic Heyndrickxia coagulans
Selected H. coagulans strains enhanced zinc transport in Caco-2 cells and improved zinc status in zinc-deficient mice, while comparative genomics identified candidate genomic features associated with strain-specific zinc-promoting effects.
- 19 October 2026
- Published by Royal Society of Chemistry (RSC) in Food & Function
- Vol. 17 (20) , 9049-9061
Comparative genomic insights into strain-specific zinc-promoting effects of the spore-forming probiotic Heyndrickxia coagulans
Selected H. coagulans strains enhanced zinc transport in Caco-2 cells and improved zinc status in zinc-deficient mice, while comparative genomics identified candidate genomic features associated with strain-specific zinc-promoting effects.
- Open Access
- 25 August 2026
- Published by American Chemical Society (ACS) in Chem & Bio Engineering
Substantial Improvement of Cellobiose 2-Epimerase from Caldicellulosiruptor saccharolyticus for the Efficient Synthesis of Lactulose by Fluorescence-Quenching-Based Ultrahigh-Throughput Microdroplet Screening
Cellobiose 2-epimerases (CEs) represent a promising biocatalytic route to lactulose, yet their industrial implementations remain constrained by suboptimal catalytic efficiency. A primary bottleneck lies in the limited throughput of conventional screening methods, which restricts the exploration of sequence space and impedes systematic enzyme engineering. To address this challenge, we developed an ultrahigh-throughput (UHTP) screening platform that integrates a boronic acid-based fluorescent sensor with droplet microfluidics. This system exhibits high sensitivity toward lactulose (100 μM) and enabled rapid sorting of approximately 105 mutants from partitioned regions of Caldicellulosiruptor saccharolyticus CE (CsCE). We identified a top-performing variant, M3 (L172 V/S173L/K356L/Q371E), exhibiting a specific activity of 19.1 U/mg─the highest reported to date and a 4.68-fold increase over the wild-type─alongside a record lactulose conversion of 75%. Molecular dynamics simulations indicate that the enhanced performance stems from the optimized dynamic conformation of flexible loops at the substrate tunnel entrance, which fine-tunes the catalytic microenvironment and accelerates substrate/product flux. Collectively, this work establishes a scalable UHTP platform for CEs and delivers mechanistic insights that enhance the biosynthesis of lactulose.
- 25 August 2026
- Published by American Chemical Society (ACS) in Chem & Bio Engineering
Open AccessSubstantial Improvement of Cellobiose 2-Epimerase from Caldicellulosiruptor saccharolyticus for the Efficient Synthesis of Lactulose by Fluorescence-Quenching-Based Ultrahigh-Throughput Microdroplet Screening
Cellobiose 2-epimerases (CEs) represent a promising biocatalytic route to lactulose, yet their industrial implementations remain constrained by suboptimal catalytic efficiency. A primary bottleneck lies in the limited throughput of conventional screening methods, which restricts the exploration of sequence space and impedes systematic enzyme engineering. To address this challenge, we developed an ultrahigh-throughput (UHTP) screening platform that integrates a boronic acid-based fluorescent sensor with droplet microfluidics. This system exhibits high sensitivity toward lactulose (100 μM) and enabled rapid sorting of approximately 105 mutants from partitioned regions of Caldicellulosiruptor saccharolyticus CE (CsCE). We identified a top-performing variant, M3 (L172 V/S173L/K356L/Q371E), exhibiting a specific activity of 19.1 U/mg─the highest reported to date and a 4.68-fold increase over the wild-type─alongside a record lactulose conversion of 75%. Molecular dynamics simulations indicate that the enhanced performance stems from the optimized dynamic conformation of flexible loops at the substrate tunnel entrance, which fine-tunes the catalytic microenvironment and accelerates substrate/product flux. Collectively, this work establishes a scalable UHTP platform for CEs and delivers mechanistic insights that enhance the biosynthesis of lactulose.
- 4 August 2026
- Published by American Chemical Society (ACS) in ACS Catalysis
- Vol. 16 (16) , 16076-16091
A Multi-Modal Pre-training Framework-Driven Active Learning System for Enhanced Protein Evolution of Cs CE
Protein language models have shown strong potential in modeling fitness landscapes for directed evolution; however, their predictive accuracy and generalization to unexplored sequence space remain limited under few-shot learning conditions. Here, we present MmALS (Multi-modal Active Learning System), a few-shot active learning framework that incorporates cold-start region scanning and multi-objective optimization to enable unbiased mutation-site selection. Its learning module employs a multimodal fusion architecture that incorporates sequence, structural, and multiple sequence alignment information, enabling accurate variant fitness prediction under low-sample conditions. The robustness of the MmALS logical framework was validated through in silico benchmarking across 12 independent datasets, in which it consistently accelerated convergence and demonstrated transferability across diverse enzyme fitness landscapes. Using Caldicellulosiruptor saccharolyticus Cellobiose 2-epimerase (CsCE) as the experimental validation, MmALS achieved a 3.88-fold enhancement of isomerization activity after only three iterative rounds (approximately 50 mutants per round), reaching a record-high activity of 20.72 U/mg. Overall, MmALS represents a data-efficient active learning framework for protein evolution, advancing computational enzyme design by enabling accurate optimization under limited experimental sampling.
- 4 August 2026
- Published by American Chemical Society (ACS) in ACS Catalysis
- Vol. 16 (16) , 16076-16091
A Multi-Modal Pre-training Framework-Driven Active Learning System for Enhanced Protein Evolution of Cs CE
Protein language models have shown strong potential in modeling fitness landscapes for directed evolution; however, their predictive accuracy and generalization to unexplored sequence space remain limited under few-shot learning conditions. Here, we present MmALS (Multi-modal Active Learning System), a few-shot active learning framework that incorporates cold-start region scanning and multi-objective optimization to enable unbiased mutation-site selection. Its learning module employs a multimodal fusion architecture that incorporates sequence, structural, and multiple sequence alignment information, enabling accurate variant fitness prediction under low-sample conditions. The robustness of the MmALS logical framework was validated through in silico benchmarking across 12 independent datasets, in which it consistently accelerated convergence and demonstrated transferability across diverse enzyme fitness landscapes. Using Caldicellulosiruptor saccharolyticus Cellobiose 2-epimerase (CsCE) as the experimental validation, MmALS achieved a 3.88-fold enhancement of isomerization activity after only three iterative rounds (approximately 50 mutants per round), reaching a record-high activity of 20.72 U/mg. Overall, MmALS represents a data-efficient active learning framework for protein evolution, advancing computational enzyme design by enabling accurate optimization under limited experimental sampling.
- 22 June 2026
- Published by Elsevier in Bioresource Technology
- Vol. 459, 135217
Production of phloretin by multi-module carbon flux rewiring in Saccharomyces cerevisiae
- 22 June 2026
- Published by Elsevier in Bioresource Technology
- Vol. 459, 135217
Production of phloretin by multi-module carbon flux rewiring in Saccharomyces cerevisiae
- 22 May 2026
- Published by Elsevier in Food Bioscience
- Vol. 81
Biosensor-driven evolution of oxidoreductive pathway for efficient tagatose production by Saccharomyces cerevisiae
- 22 May 2026
- Published by Elsevier in Food Bioscience
- Vol. 81
Biosensor-driven evolution of oxidoreductive pathway for efficient tagatose production by Saccharomyces cerevisiae
- 24 March 2026
- Published by American Chemical Society (ACS) in Journal of Agricultural and Food Chemistry
- Vol. 74 (12) , 10375-10386
Engineering a Tyrosol-Responsive Dynamic System in Saccharomyces cerevisiae for Adaptive Regulation of Hydroxytyrosol Biosynthesis
Static metabolic regulation during microbial synthesis of polyphenolic compounds leads to tyrosol accumulation, inducing cellular toxicity and growth inhibition. Dynamic regulation effectively alleviates the imbalance between cell growth and production; yet, a tyrosol-responsive dynamic system for yeast adaptive regulation remains unreported. In this study, endogenous tyrosol-responsive promoters were mined via transcriptomics, and the optimal promoter PYOR153W was identified. Subsequently, a novel tyrosol-responsive biosensor composed of the responsive promoter and CRISPR/dSpCas9-VPR was designed to dynamically regulate green fluorescent protein (GFP) expression and hydroxytyrosol biosynthesis, respectively. This strategy successfully enhanced hydroxytyrosol production by 1.53-fold with negligible tyrosol accumulation. Furthermore, this dynamic system improved strain robustness during longer yeast fermentation periods, and the hydroxytyrosol titer increased to 4.26 g/L in a 5 L bioreactor. Our findings establish the first tyrosol-responsive biosensor in yeast and provide a framework for the efficient biosynthesis of high-value tyrosol derivatives.
- 24 March 2026
- Published by American Chemical Society (ACS) in Journal of Agricultural and Food Chemistry
- Vol. 74 (12) , 10375-10386
Engineering a Tyrosol-Responsive Dynamic System in Saccharomyces cerevisiae for Adaptive Regulation of Hydroxytyrosol Biosynthesis
Static metabolic regulation during microbial synthesis of polyphenolic compounds leads to tyrosol accumulation, inducing cellular toxicity and growth inhibition. Dynamic regulation effectively alleviates the imbalance between cell growth and production; yet, a tyrosol-responsive dynamic system for yeast adaptive regulation remains unreported. In this study, endogenous tyrosol-responsive promoters were mined via transcriptomics, and the optimal promoter PYOR153W was identified. Subsequently, a novel tyrosol-responsive biosensor composed of the responsive promoter and CRISPR/dSpCas9-VPR was designed to dynamically regulate green fluorescent protein (GFP) expression and hydroxytyrosol biosynthesis, respectively. This strategy successfully enhanced hydroxytyrosol production by 1.53-fold with negligible tyrosol accumulation. Furthermore, this dynamic system improved strain robustness during longer yeast fermentation periods, and the hydroxytyrosol titer increased to 4.26 g/L in a 5 L bioreactor. Our findings establish the first tyrosol-responsive biosensor in yeast and provide a framework for the efficient biosynthesis of high-value tyrosol derivatives.
- 19 March 2026
- Published by American Chemical Society (ACS) in Journal of Agricultural and Food Chemistry
- Vol. 74 (12) , 10220-10231
Metabolomics-Guided Integration of a “Push–Pull-Restrain” Strategy for Enhanced LNnT Production in Saccharomyces cerevisiae
Lacto-N-neotetraose (LNnT) is a prominent neutral human milk oligosaccharide (HMO) with limited availability. Here, Saccharomyces cerevisiae (GRAS strain, Generally Recognized as Safe) was metabolically engineered to enhance LNnT production. Metabolomics analysis revealed that competition from chitin synthesis and insufficient metabolic driving force for heterologous glycosyltransferases in yeast constituted major metabolic bottlenecks. To overcome these limitations, a “push–pull-restraint” strategy was applied, integrating glycosyltransferase screening, protein tag fusion evaluation, targeted chitin disruption of chitin biosynthetic genes, and balanced modulation of UDP-Gal/UDP-GlcNAc supply. These interventions increased extracellular LNnT titer in shake-flask cultures from 14.65 mg/L to 1.66 g/L, with total LNnT reaching 2.01 g/L. Further scale-up in a 5 L bioreactor under fed-batch conditions resulted in a total LNnT titer of 4.70 g/L. These results demonstrate the potential for the production of LNnT by S. cerevisiae and open up an innovative framework for the effective synthesis of LNnT.
- 19 March 2026
- Published by American Chemical Society (ACS) in Journal of Agricultural and Food Chemistry
- Vol. 74 (12) , 10220-10231
Metabolomics-Guided Integration of a “Push–Pull-Restrain” Strategy for Enhanced LNnT Production in Saccharomyces cerevisiae
Lacto-N-neotetraose (LNnT) is a prominent neutral human milk oligosaccharide (HMO) with limited availability. Here, Saccharomyces cerevisiae (GRAS strain, Generally Recognized as Safe) was metabolically engineered to enhance LNnT production. Metabolomics analysis revealed that competition from chitin synthesis and insufficient metabolic driving force for heterologous glycosyltransferases in yeast constituted major metabolic bottlenecks. To overcome these limitations, a “push–pull-restraint” strategy was applied, integrating glycosyltransferase screening, protein tag fusion evaluation, targeted chitin disruption of chitin biosynthetic genes, and balanced modulation of UDP-Gal/UDP-GlcNAc supply. These interventions increased extracellular LNnT titer in shake-flask cultures from 14.65 mg/L to 1.66 g/L, with total LNnT reaching 2.01 g/L. Further scale-up in a 5 L bioreactor under fed-batch conditions resulted in a total LNnT titer of 4.70 g/L. These results demonstrate the potential for the production of LNnT by S. cerevisiae and open up an innovative framework for the effective synthesis of LNnT.
- 19 March 2026
- Published by Elsevier in Food Chemistry
- Vol. 512, 148942
Development of a cold-adapted enzymatic cascade system for in situ value-added conversion of milk
- 19 March 2026
- Published by Elsevier in Food Chemistry
- Vol. 512, 148942
Development of a cold-adapted enzymatic cascade system for in situ value-added conversion of milk
- Open Access
- 1 February 2026
- Published by Tsinghua University Press in Food Science and Human Wellness
- Vol. 15 (2)
De novo production of lacto- N -neotetraose by metabolic engineering of Saccharomyces cerevisiae as a cell factory
Lacto-N-neotetraose (LNnT) is a crucial neutral core human milk oligosaccharide (HMO). In this study, we established a LNnT-producing Saccharomyces cerevisiae cell factory was established through comprehensive metabolic engineering. Specifically, the de novo biosynthetic pathway of LNnT was assembled by heterologously expressing the lactose permease (Lac12) from Kluyveromyces lactis and the glycosyltransferase from Neisseria meningitidis in Saccharomyces cerevisiae. Subsequently, carbon source regulation based on the glucose-sensitive GAL regulatory system was employed to optimize the expression time of heterologous genes, achieving a production of 15.61 mg/L of LNnT in shake-flask fermentation. In addition, the key rate-limiting steps involved in LNnT synthesis pathway were identified and the corresponding genes were overexpressed to enhance LNnT production, resulting in an eight-fold increase in LNnT titer compared to that of parental strain. To our knowledge, this is the first report on LNnT biosynthesis in Saccharomyces cerevisiae, opening up the possibility of green production of LNnT using food-safe microorganisms.
- 1 February 2026
- Published by Tsinghua University Press in Food Science and Human Wellness
- Vol. 15 (2)
Open AccessDe novo production of lacto- N -neotetraose by metabolic engineering of Saccharomyces cerevisiae as a cell factory
Lacto-N-neotetraose (LNnT) is a crucial neutral core human milk oligosaccharide (HMO). In this study, we established a LNnT-producing Saccharomyces cerevisiae cell factory was established through comprehensive metabolic engineering. Specifically, the de novo biosynthetic pathway of LNnT was assembled by heterologously expressing the lactose permease (Lac12) from Kluyveromyces lactis and the glycosyltransferase from Neisseria meningitidis in Saccharomyces cerevisiae. Subsequently, carbon source regulation based on the glucose-sensitive GAL regulatory system was employed to optimize the expression time of heterologous genes, achieving a production of 15.61 mg/L of LNnT in shake-flask fermentation. In addition, the key rate-limiting steps involved in LNnT synthesis pathway were identified and the corresponding genes were overexpressed to enhance LNnT production, resulting in an eight-fold increase in LNnT titer compared to that of parental strain. To our knowledge, this is the first report on LNnT biosynthesis in Saccharomyces cerevisiae, opening up the possibility of green production of LNnT using food-safe microorganisms.
- 20 November 2025
- Published by Elsevier in Biosensors and Bioelectronics
- Vol. 295, 118244
Genetically encoded biosensors in microbes for Tumor targeting
- 20 November 2025
- Published by Elsevier in Biosensors and Bioelectronics
- Vol. 295, 118244
Genetically encoded biosensors in microbes for Tumor targeting