Engineering a heme-dependent tryptophan hydroxylase pathway in E. coli for enhanced melatonin production.

Engineering a heme-dependent tryptophan hydroxylase pathway in E. coli for enhanced melatonin production - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Synth Syst Biotechnol . 2026 Apr 11;14:64–76. doi: 10.1016/j.synbio.2026.03.020 Search in PMC Search in PubMed View in NLM Catalog Add to search Engineering a heme-dependent tryptophan hydroxylase pathway in E. coli for enhanced melatonin production Luyao Zhang Luyao Zhang a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Luyao Zhang a, b, c, d , Guobin Yin Guobin Yin a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Guobin Yin a, b, c, d , Senyu Pan Senyu Pan a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Senyu Pan a, b, c, d , Xinai Zheng Xinai Zheng a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Xinai Zheng a, b, c, d , Siyan Zhou Siyan Zhou a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Siyan Zhou a, b, c, d , Guocheng Du Guocheng Du a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Guocheng Du a, b, c, d , Jianghua Li Jianghua Li a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Jianghua Li a, b, c, d , Jian Chen Jian Chen a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Jian Chen a, b, c, d , Ruirui Xu Ruirui Xu a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Ruirui Xu a, b, c, d, ⁎ , Zhen Kang Zhen Kang a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China Find articles by Zhen Kang a, b, c, d, ⁎⁎ Author information Article notes Copyright and License information a The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China b The Science Center for Future Foods, Jiangnan University, Wuxi, China c Jiangsu Province Basic Research Center for Synthetic Biology, Jiangnan University, Wuxi, China d State Key Laboratory of Food Science and Resources, Jiangnan University, Wuxi, China ⁎ Corresponding author. The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China. [email protected] ⁎⁎ Corresponding author. The Key Laboratory of Carbohydrate Chemistry and Biotechnology, Ministry of Education, School of Biotechnology, Jiangnan University, Wuxi, China. [email protected] Received 2026 Feb 2; Revised 2026 Mar 17; Accepted 2026 Mar 30; Collection date 2026 Dec. © 2026 The Authors This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). PMC Copyright notice PMCID: PMC13091554  PMID: 42006854 Abstract Melatonin is a high-value bioactive indoleamine broadly applied in the pharmaceutical, food and nutraceutical industries, yet its microbial production remains constrained by pathway complexity and low enzymatic efficiency. Here, a streamlined melatonin biosynthetic pathway was established in Escherichia coli by introducing the heme-dependent tryptophan hydroxylase Luz15. We further enhanced the intracellular heme availability and increased melatonin production by 111.45%. Fusion-tag engineering improved the solubility of all heterologous enzymes and boosted production by 50.64%. Structure-guided rational design of Luz15 subsequently yielded a beneficial D299S/W376H mutant, enhancing hydroxylation activity and contributing a 33.81% increase. To divert metabolic flux toward melatonin, a high-producing chassis was constructed via gene-editing and systematic sRNA library screening. The final strain produced 753.78 mg/L melatonin in a 5-L fed-batch bioreactor from glucose, a 38.78-fold improvement over the initial. This study establishes an efficient and scalable microbial platform for the sustainable biosynthesis of melatonin and other hydroxylated tryptophan-derived compounds. Keywords: Melatonin, Metabolic engineering, Tryptophan hydroxylase, Protein engineering, sRNA 1. Introduction Melatonin ( N -acetyl-5-methoxytryptamine) is an indoleamine hormone widely distributed in animals, plants and microorganisms. It regulates circadian rhythms and sleep–wake cycles in animals, while in plants it functions as an endogenous antioxidant and signaling molecule that supports growth and enhances tolerance to biotic and abiotic stresses. Across biological systems, melatonin exhibits pronounced antioxidant, anti-inflammatory, and immunomodulatory activities, thereby contributing to redox homeostasis and protection against oxidative damage [ 1 ]. Owing to its demonstrated efficacy in alleviating insomnia and anxiety, melatonin is widely marketed as an over-the-counter medication and dietary supplement [ 2 , 3 ]. These multifunctional properties underpin the broad application potential of melatonin in both medical and agricultural fields. Initially, melatonin was primarily extracted from the pineal glands of animals and from plants. However, animal-derived materials carry inherent risks of viral contamination, while plant-based sources are highly dependent on seasonal and geographic factors, making extraction methods insufficient to meet market demand [ 4 , 5 ]. Over the past decades, several chemical synthesis routes have been developed, which can be broadly classified into five categories according to the key starting materials and reaction schemes, including routes based on 5-methoxyindole, indole with chloroethylamine, the Fischer indole synthesis, 5-methoxyindole-3-carboxaldehyde and free-radical indole method [ [5] , [6] , [7] ]. However, chemical synthesis relies on harsh reactions and toxic reagents, causing high cost, environmental hazards and serious safety concerns. Consequently, the development of sustainable and environmentally friendly production strategies is imperative. In recent years, microbial cell factory–based melatonin production has emerged as an increasingly active area of research [ [8] , [9] , [10] , [11] ]. Most genes employed to establish melatonin biosynthesis in genetically engineered bacteria originate from animals and plants. In animals, melatonin is synthesized from tryptophan via the intermediates 5-hydroxytryptophan (5-HTP), 5-hydroxytryptamine (5-HT, serotonin), and N -acetylserotonin (NAS). Accordingly, the catalytic process involves four enzymes, including tryptophan-5-hydroxylase (TPH), tryptophan decarboxylase (TDC), serotonin N -acetyltransferase (SNAT), and N -acetylserotonin methyltransferase (ASMT) [ 9 , 12 ]. In plants, melatonin biosynthesis also begins with tryptophan and possesses an addition alternative pathway, involving four enzymatic steps. The first step is the decarboxylation of tryptophan, followed by the synthesis of serotonin catalyzed by tryptamine 5-hydroxylase (T5H), and the final conversion to melatonin is mediated by caffeic acid O -methyltransferase (COMT) [ 13 , 14 ]. Building on the current understanding of melatonin biosynthesis across diverse organisms, researchers constructed various melatonin production strains, including Saccharomyces cerevisiae and E . coli ( Table S1 ). These previous studies have consistently claimed that tryptophan hydroxylation represents the rate-limiting step in melatonin biosynthesis, which requires sufficient supply of the cofactor tetrahydrobiopterin (BH 4 ) or tetrahydromonapterin (MH 4 ) to support multiturn-over catalysis. Nonetheless, the synthesis and regeneration of BH 4 or MH 4 are arduous processes and necessitate introduction of multiple exogenous genes in E. coli [ 15 , 16 ]. BH 4 generated from guanosine triphosphate (GTP) goes through three steps, depending on enzymes GTP cyclohydrolase I (GCHI), 6-pyruvoyl-tetrahydropterin synthase (PTPS) and sepiapterin reductase (SPR), as its regeneration from pterin-4α-carbinolamine (BH 3 OH) relies on two enzymes, pterin-4α-carbinolamine dehydratase (PCD) and dihydrofolate reductase (DHPR) [ 17 ]. Therefore, construction of an efficient tryptophan hydroxylation pathway is a critical determinant for achieving high-level melatonin production. Shi et al. identified a novel bacterial tryptophan hydroxylase (bTPH) that utilizes heme as a cofactor, catalyzing tryptophan hydroxylation without requiring a biopterin synthesis–regeneration system and producing 238 mg/L melatonin in Streptomyces albus [ 18 ]. Because heme is endogenously available in E . coli , expression of Luz15 alone is sufficient to enable 5-HTP biosynthesis, thereby reducing physiological and metabolic burden and providing a new strategy for efficient hydroxylation modification of tryptophan. In this study, we established a novel melatonin biosynthesis pathway in E. coli BL21(DE3) by importing four heterologous genes luz15 , ddc , aanat and asmt . Notably, the bacterial enzyme Luz15 converts tryptophan to 5-HTP using heme, the cofactor endogenously available in E. coli , thereby eliminating the need for exogenous genes required by BH4-dependent systems, which exhibits a distinct convenience and scalability for metabolic engineering. To improve intracellular heme generation, we knocked out genes involved in heme degradation and competing pathways that consume heme precursors. Meanwhile, we also overexpressed several crucial heme synthesis genes to enhance the melatonin production, and found that constitutive expression of Rhodopseudomonas palustris hemA proved superior to all alternatives for enhancing heme yield. To promote the soluble expression of all pathway enzymes, we constructed a solubility-enhancing fusion tag library and fused them to the N-termini of relative enzymes respectively to identified the optimal tag-enzyme pairings. Considering that hydroxylation is the committed step, we focused on the interaction among enzyme Luz15, cofactor heme and substrate tryptophan by molecular docking. Structure-guided engineering of Luz15 substrate-binding pocket yielded the most favorable variant, Luz15 D299S/W376H. To redirect carbon flux towards melatonin, we not only eliminated the precursor tryptophan degradation pathway and counteracting feedback inhibition, but also constructed an sRNA library to screen potential target genes that enhance melatonin production. Consequently, the final strain MT57 expressing Luz15 mutant D299S/W376H and sRNA anti- tyrA produced 753.78 ± 32.37 mg/L melatonin from glucose in a 5-L fermenter. Overall, this study reconstructed melatonin synthesis pathway in E. coli by various efficacious strategies, showing the prospect of microbial fermentation in tryptophan derivative production. 2. Materials and methods 2.1. Strains and plasmids E. coli JM109 was used for plasmids construction, amplification and preservation. E. coli BL21(DE3), designated as Trp0, and its derivatives were used to express protein and produce melatonin. The genes luz15 from Actinomadura luzonensis , ddc from Candidatus Koribacter versatilis Ellin 345, aanat from Streptomyces griseofuscus and asmt from Homo sapiens were synthesized by GENEWIZ (Suzhou, China), then the gene fragments upon were fused into vector pET-28a(+) and pCDFDuet-1 to construct origin plasmids pET28a(+)-P lamdalac - ddc - aanat -P T7 - asmt and pCDF-P T7 - luz15 , using primers luz1 5-fg-F/R, pCDF-vector-F/R, ddc -fg-F/R, aanat -fg-F/R, asmt -fg-F/R and pET-vector-F/R. Gene hemA from R. palustris obtained from lab preservation, while genes hemB , hemE , hemF , hemG , hemH amplified from E. coli BL21(DE3) with primers hemA -fg-F/R, hemB -fg-F/R, hemE -fg-F/R, hemF -fg-F/R, hemG -fg-F/R and hemH -fg-F/R, were aimed to construct plasmids to enhance heme synthesis. Additionally, solubility-enhancing fusion tag genes sumo , mbp , gst , trxA , nusA , gb1 , and dsbC , which were stored in the laboratory, were fused at the N-terminal of Luz15, Ddc, Aanat and ASMT respectively with designated primers. The sRNA expression plasmids were constructed via PCR with corresponding primer pairs and self-cyclization using endogenous DNA recombinase. All strains, plasmids and primers are listed in Tables S2–S4 . 2.2. Medium and cultivation E. coli JM109 was cultivated at 37 °C in LB medium (10 g/L tryptone, 10 g/L NaCl, 5 g/L yeast extract). The fermentation medium for melatonin production in shake-flask level contains 10 g/L glucose, 2 g/L yeast extract, 16 g/L (NH 4 ) 2 SO 4 , 16 g/L Na 2 HPO 4 ·12H 2 O, 3 g/L KH 2 PO 4 , 1 g/L MgSO 4 ·7H 2 O, 0.01 g/L MnSO 4 ·H 2 O, 0.1 g/L ferric citrate, 1 g/L glycine, 1 g/L tryptophan and 1 g/L methionine. Correct colonies on agar were transferred into LB medium with corresponding antibiotic and incubated overnight at 37 °C. Then inoculation with an initial OD 600 of 0.6, the seed culture was introduced into a 250 mL shake flask containing 25 mL fermentation medium and cultivated at 30 °C. After 2 h, inducer IPTG with a final concentration of 0.1 mM was added into the culture to induce melatonin production. When necessary, medium contains kanamycin (50 μg/mL), streptomycin (50 μg/mL) or chloramphenicol (15 μg/mL). 2.3. Protein purification The cells expressing Luz15 wild type, Luz15 wild type co-expressed with HemA, or Luz15 variant D299S/W376H were harvested and resuspended in lysis buffer (300 mM NaCl, 50 mM NaH 2 PO 4 , 10 mM imidazole, pH 8.0). Cells were disrupted by high-pressure homogenization and the lysates were centrifuged at 12,000 rpm for 10 min to collect supernatant. His-tagged Luz15 and its variant were purified using Ni-TED sepharose affinity column. The columns were initially washed with washing buffer (300 mM NaCl, 50 mM NaH 2 PO 4 , 25 mM imidazole, pH 8.0) and subsequently proteins were eluted with elution buffer (300 mM NaCl, 50 mM NaH 2 PO 4 , 300 mM imidazole, pH 8.0). Protein concentration was determined using Bradford protein assay. The heme occupancy of enzyme Luz15, with or without HemA constitutively co-expressed, was evaluated by recording UV-visible absorption spectra from 350 nm to 600 nm using a microplate reader (TECAN). Protein samples were adjusted to a final concentration of 20 μM and heme-bound Luz15 exhibited a characteristic absorption peak at approximately 410 nm, indicating holo-Luz15 formation. 2.4. In vitro biochemical assays For the in vitro kinetic analysis of Luz15 and its variant D299S/W376H with O 2 as oxidant, reactions were performed in a volume of 100 μL, containing 10 μM purified enzyme, 25 mM l -ascorbate and various concentrations of substrate l -tryptophan in 50 mM HEPES buffer (pH 8.0). Reactions were incubated for 10 min at 30 °C and subsequently quenched by the addition of 300 μL methanol. Initial reaction rates (V 0 ) were determined from product formation and substrate consumption, which were detected by HPLC. Kinetic parameters K m and k cat were obtained by fitting V 0 as a function of substrate concentration to the Michaelis-Menten equation using software GraphPad Prism 10. 2.5. Bioinformatics tools Structure of wild-type Luz15 and D299S/W376H mutant proteins was predicted by AlphaFold3 and the relative positions of the substrate tryptophan and the cofactor heme were referred to the crystal structure of XcTDO (PDB 2NW8 ). Based on the above understanding, molecular docking was executed with pyRosetta. Multiple sequence alignment was performed by MEGA11 [ 19 ] and visualized by ESPript 3.0 [ 20 ]. Virtual saturation mutagenesis was carried out and scored by pyRosetta. Molecular dynamics (MD) simulation was conducted using GROMACS version 2022 with CHARMM36 all-atom force field at the designated target temperature. Simulations were performed for 100 ns using a 2-fs integration time step. Root-mean-square deviation (RMSD) trajectories were calculated based on backbone Cα atomic coordinates for Luz15 WT and D299S/W376H mutant systems at 300 K. Root-mean-square (RMSF) values were also computed for Cα atoms of all residues in Luz15 WT and mutant D299S/W376H systems at 300 K. 2.6. High-throughput screening of the concentration of 5-HTP 5-HTP has a color reaction with Gibbs reagent, 0.5% 2,6-dibromoquinone chlorimide ethanol solution, at pH 10–11 and the product has an absorption peak at 606 nm. This reaction enables the development of a high-throughput method for screening 5-HTP high-yield strains [ 21 ]. Single colonies were cultivated in LB medium overnight at 37 °C and then transferred into 1.5 mL fermentation medium with an initial OD of 0.2 in 24-well plates. After incubating 1.5 h, culture was added 0.1 mM IPTG and went on fermenting 36 h at 30 °C. Subsequently the culture was centrifuged at 12,000 rpm for 5 min, and the supernatant was diluted to the appropriate concentration. What's more, different concentrations (0, 2, 4, 6, 8, 10 mg/L) of 5-HTP standard solution was prepared to make a standard curve to measure 5-HTP concentration of culture. Then the samples and 5-HTP standard solution were adjusted to the pH value of 10–11 and mixed with Gibbs reagent in a 25:1 vol ratio. After reacting for 30 min, the absorbance at 606 nm was measured to quantify the titer of 5-HTP according to the standard curve. 2.7. Construction of sRNA library We used a designed artificially sRNAs to inhibit the gene expression in post-transcription level. The standardized sRNAs consist of a 24 nt base-pairing region to target genes, a single stem-loop for transcription termination and a polyU tail [ 22 , 23 ]. According to the metabolic network in E . coli BL21(DE3), we selected 88 genes as potential targets for further screening. Subsequently, we utilized the web server sslRNAD ( http://www.kangzlab.cn/ ) to design sRNAs sequence and corresponding primers. Then we mixed primers together and took plasmid pCDF-P J23119 - hemA -P T7 - luz15 as template to perform one-pot PCR. The resulting PCR product was transformed into E. coli JM109 for self-cyclization and amplification. Through the above procedures, the desired sRNA library was constructed successfully. 2.8. Fed-batch fermentation For the fed-batch fermentation of MT57, a 5-L fermenter containing 2 L fed-batch fermentation medium was applied. The primary seed culture was cultivated in LB medium at 37 °C overnight and then was transferred into secondary seed medium (1.5 g/L yeast exact, 0.58 g/L NaCl, 5.29 g/L (NH 4 ) 2 SO 4 , 17.41 g/L K 2 HPO 4 , 13.61 g/L KH 2 PO 4 , 0.25 g/L MgSO 4 ·7H 2 O, 1 μg/L Thiamin·HCl, 3 mg/L vitamin B3, 1 mg/L vitamin B2, 5.51 mg/L CaCl 2 , 16.66 mg/L FeCl 3 ·6H 2 O, 1 mg/L MnCl 2 ·4H 2 O, 1.7 mg/L ZnCl 2 , 340 μg/L CuCl 2 , 470 μg/L CoCl 2 , 600 μg/L Na 2 MoO 4 ·2H 2 O) to propagate at 37 °C for 8-10 h. Fed-batch fermentation medium consists of the following components: 10 g/L glucose, 13.3 g/L KH 2 PO 4 , 4 g/L (NH 4 ) 2 HPO 4 , 1.2 g/L MgSO 4 ·7H 2 O, 1.7 g/L citric acid, 8.4 mg/L EDTA, 2.5 mg/L CoCl 2 ·6H 2 O, 15 mg/L CuCl 2 ·2H 2 O, 3 mg/L HBO 3 , 2.5 mg/L Na 2 MoO 4 ·2H 2 O, 13 mg/L Zn(CH 3 COO) 2 ·2H 2 O, 0.1 g/L Fe(III)citrate, 4.5 mg/L pyridoxal phosphate (PLP), and 4.5 mg/L thiamine·HCl. If requireed, 1 g/L methionine was added into fed-batch fermentation. 2.9. Analytical methods Samples were prepared by adding an equal volume of 100% acetonitrile, vortexed sufficiently, and centrifuged at 12,000 rpm for 10 min. Then 10 μL supernatant was filtered through a 0.22 μm nylon filter and detected by high-performance liquid chromatograph (HPLC). Analysis was performed on an Agilent 1260 HPLC equipped with an Agilent ZORBAX Eclipse XDB-C18 analytical column (3.0 mm × 250 mm, 5 μm). The column temperature was maintained at 30 °C, and a UV detector operated at a wavelength of 280 nm. Elution was performed at 0.6 mL/min with mobile phases of acetonitrile (A) and water (B) both containing 0.1% formic acid: 0 min, 5% A; 12 min, 59% A; 13 min, 95% A; 17 min, 95% A. Concentration of 1, 10, 25, 50, 75 and 100 mg/L of tryptophan, 5-HTP or melatonin were used to create standard curves for HPLC quantification of fermentation broth. The liquid chromatography-mass spectrometry (LC–MS) analysis to detect 5-HTP or melatonin was performed under the same conditions on a Waters Quattro Premier XE mass spectrometer (Micromass, Manchester, UK), operating in positive mode. LC–MS parameters were capillary voltage 3000 V, gas temperature 350 °C, gas flow 12.0 L/min, nebulizer pressure 35 psig and full scan mode ( m / z 100–500) [ 18 ]. Data were processed by MassLynx software. For intracellular heme quantification, cells were resuspended in 1 M NaOH solution and disrupted by ultrasonication. The lysates were centrifuged at 12,000 rpm for 10 min to collect supernatant and then the concentration of heme was measured with Heme assay kit (BioAssay Systems) at 400 nm [ 24 ], where the intensity of color is proportional to the heme concentration. 3. Results and discussion 3.1. Construction of a heme-dependent tryptophan hydroxylation pathway for melatonin biosynthesis in E. coli Tryptophan hydroxylation is the committed and rate-limiting step in melatonin biosynthesis [ 9 , 12 ]. Most previously reported microbial platforms rely on animal-derived tryptophan hydroxylases (aTPHs), which strictly require BH 4 as a redox cofactor. As a result, functional expression of aTPHs in E . coli necessitates reconstruction of a complex biopterin biosynthesis and regeneration system involving multiple auxiliary enzymes [ 16 , 25 , 26 ], substantially increasing pathway complexity, metabolic burden and redox imbalance, thereby limiting scalability. In contrast, recently identified bTPHs represent a distinct class of enzymes that utilize heme as the sole cofactor and operate independently of biopterin synthesis–regeneration systems [ 18 ]. Among them, Luz15 from Aurantimonas luzonensis was selected in this study due to its robust catalytic activity toward l -tryptophan in heterologous hosts and its reliance only on endogenous heme in E. coli . This single-enzyme dependency markedly simplifies pathway architecture and minimizes metabolic interference, offering a more economical and scalable strategy for microbial melatonin production. Accordingly, Luz15 was introduced into E. coli to establish a heme-dependent tryptophan hydroxylation module ( Fig. 1 A). The empty vector pCDFDuet-1 or the expression plasmid pCDF-P T7 - luz15 was transformed into E. coli BL21(DE3), generating the control strain MT0 and the Luz15-expressing strain MT1, respectively. Following shake-flask fermentation, 5-hydroxytryptophan (5-HTP) was detected in MT1 by LC–MS analysis ( Fig. 1 B). Quantitative HPLC analysis further confirmed a 5-HTP titer of 30.23 ± 1.00 mg/L in MT1, whereas no detectable 5-HTP was observed in MT0 ( Fig. 1 C), demonstrating the effective in vivo activity of Luz15. Subsequently, to establish a complete melatonin biosynthetic pathway, an aromatic l -amino-acid decarboxylase (Ddc) from Corynebacterium koribacter , an N -acetyltransferase (Aanat) from Streptomyces griseofuscus , and an acetylserotonin O -methyltransferase (ASMT) from H . sapiens were heterologously expressed in E. coli , following a previously reported pathway design ( Fig. 1 D) [ 27 ]. This multienzyme cascade enabled the stepwise conversion of 5-HTP to melatonin. Fig. 1. Open in a new tab Construction of a new melatonin biosynthesis pathway in E. coli . (A) Different pathways for 5-hydroxytryptophan synthesis from tryptophan utilizing tryptophan hydroxylase from bacteria or animals in E. coli . (B) LC–MS identification of the synthesized 5-HTP. (C). 5-HTP production in MT1 and the blank control strain MT0. (D) A biosynthesis pathway of melatonin from 5-HTP in E. coli [ 27 ]. (E) LC–MS identification of the synthesized melatonin. (F) Time-course situations of melatonin production and cell growth in strain MT2 during 36 h shake-flask fermentation. Samples were taken every 4 h to measure melatonin titer and OD 600 . All the data are presented as mean values ± S.D. from three independent biological replicates (n = 3). Incorporation of the complete melatonin biosynthetic pathway into this chassis generated the initial production strain MT2. Following 36 h of shake-flask cultivation, SDS-PAGE analysis confirmed the successful expression of all enzymes involved in the melatonin biosynthetic pathway ( Fig. S1 ). Quantitative HPLC analysis indicated that MT2 achieved a melatonin titer of 18.96 ± 0.23 mg/L ( Fig. 1 E and F; Fig. S2 ). Collectively, these results demonstrate that bacterial heme-dependent tryptophan hydroxylases, exemplified by Luz15, provide a superior enzymatic platform for melatonin biosynthesis in E. coli by circumventing the cofactor constraints associated with animal TPHs. This strategy establishes a simplified, robust and industrially attractive route for the microbial production of melatonin and other value-added tryptophan derivatives. 3.2. Enhancing heme availability to relieve the rate-limiting tryptophan hydroxylation step Liquid chromatographic analysis revealed that in the fermentation broth pathway intermediates, containing 5-HTP, 5-HT and NAS, accumulated at relatively low levels compared to final product melatonin ( Fig. S2 ), corroborating previous reports that tryptophan hydroxylation constitutes a rate-limiting step in melatonin biosynthesis. Given that Luz15 is a heme-dependent enzyme, we sought to enhance intracellular heme availability to increase the proportion of catalytically active holo-Luz15 in E. coli . Notably, excessive heme accumulation and its synthetic intermediates adversely affected cell growth [ [28] , [29] , [30] ], indicating that an optimal balance between holoenzyme formation and cellular fitness is required ( Fig. 2 A). Fig. 2. Open in a new tab Metabolic engineering to optimize the supply of Luz15 cofactor heme. (A) The balance between melatonin production and cellular fitness through modifying heme biosynthesis and the metabolic pathway of heme in E . coli . Overexpressed genes are highlighted in pink and knockout genes are highlighted in purple. (B) Effect of additional 5-ALA on melatonin production. 5-ALA with final concentration from 0 to 0.5 g/L was added to medium and the production of melatonin in MT2 was measured after 36 h of 24-well-plate fermentation. (C) The melatonin titers and OD 600 after knocking out gene cyoE or cysG in strain MT2, constructing strain MT3 and MT4. (D) The effect of overexpressing key genes about heme biosynthesis on melatonin titers and OD 600 in MT5-MT13 without 5-ALA supplementation. (E) The change of melatonin production and OD 600 at 4-h intervals in 36 h shake-flask fermentation through optimizing hemA expression. In (B), (C), (D) and (E), all the data are presented as mean values ± S.D. from three independent biological replicates (n = 3). In E. coli , heme is synthesized from 5-aminolevulinic acid (5-ALA) through a seven-step pathway involving hemB , hemC , hemD , hemE , hemF , hemG , and hemH ( Fig. 2 A). To preliminarily assess the role of heme availability in tryptophan hydroxylation, 5-ALA was supplemented into the culture medium at final concentrations ranging from 0.1 to 0.5 g/L. Following 24-well plate fermentation, melatonin production increased by more than two-fold compared with cultures without 5-ALA supplementation ( Fig. 2 B), indicating that external addition of 5-ALA enhanced melatonin production. However, there was a production plateau at higher 5-ALA concentration, which may be associated with tight cellular regulation of heme biosynthesis restricting the conversion of 5-ALA to heme [ 31 , 32 ], or the capacity of the 5-ALA transport system limiting intracellular availability of 5-ALA [ [33] , [34] , [35] , [36] ]. Considering that direct supplementation of 5-ALA is economically impractical, we next engineered the endogenous heme metabolic network to enhance heme availability without external addition of 5-ALA. Genes cyoE and cysG are involved in heme consumption or diversion into competing pathways, as cyoE encodes protoheme IX farnesyltransferase that converts heme into heme O, while cysG encodes siroheme synthase that channels the heme precursor uroporphyrinogen III (UPG III) toward siroheme biosynthesis [ 37 ]. Accordingly, these two genes were individually deleted in strain MT2 to generate MT3 (MT2 Δ cyoE ) and MT4 (MT2 Δ cysG ). As a result, melatonin titers reached 16.78 ± 2.53 mg/L in MT3 and decreased sharply to 1.61 ± 0.48 mg/L in MT4 ( Fig. 2 C). These results indicate that deletion of cyoE exerts minimal impact on melatonin production, whereas cysG deletion severely impairs cell growth and productivity ( Fig. S3 ). Consequently, MT2 was remained as the chassis strain for subsequent engineering. In addition to blocking competing pathways, overexpression of heme biosynthetic genes has been shown to effectively elevate intracellular heme levels [ 32 , 38 , 39 ]. Therefore, hemB , hemE , hemF , hemG , and hemH from E. coli and hemA from Rhodopseudomonas palustris were overexpressed in various combinations to construct strains MT5-MT13. Among these engineered strains, MT13 exhibited the highest melatonin titer of 35.18 ± 5.04 mg/L without compromising cell growth, representing an 85.57% increase relative to MT2 ( Fig. 2 D and Fig. S3 ). Although strains MT8, MT9 and MT12 accumulated substantial levels of intracellular heme, their melatonin production was unexpectedly low, supposing excessive heme and its reactive intermediates exerted cytotoxicity and impaired cell viability, thereby limiting melatonin production. To further improve the temporal coordination between heme biosynthesis and Luz15 expression, the inducible promoter P trc controlling hemA was replaced with the constitutive promoter P J23119 , yielding strain MT14. This strategy enabled heme accumulation prior to HemA expression, thereby facilitating efficient holoenzyme assembly and improving cellular adaptation. Time-course analysis demonstrated that MT14 maintained normal growth and initiated melatonin production earlier, ultimately achieving a titer of 40.09 ± 0.98 mg/L ( Fig. 2 E). Compared with modulation of downstream heme biosynthetic enzymes, which possibly causes accumulation of toxic intermediates and feedback regulation, overexpression of 5-ALA synthase enhances the committed precursor 5-ALA generation and determines the overall pathway flux. Furthermore, the intracellular heme concentrations of strain MT2-MT14 were quantified with MT14 exhibiting an enhanced intracellular heme concentration of 1.51 ± 0.13 mg/L from 1.06 ± 0.16 mg/L in MT2 ( Fig. S4 and Fig. S5 ), suggesting that constitutive expression of HemA effectively optimized heme supply. Comparison between intracellular heme levels and corresponding melatonin titers reveals a positive correlation, suggesting that moderate improvement of heme availability facilitates melatonin biosynthesis without compromising cell growth. Consistently, UV-visible spectroscopic analysis revealed that while purified Luz15 from MT2 or MT14 exhibited identical absorption wavelength, the MT14-derived Luz15 displayed a 3.8-fold higher absorbance at 410 nm, directly indicating a higher proportion of active holo-Luz15 formation ( Fig. S6 and Fig. S7 ). In conclusion, the results point the importance of cofactor engineering as a complementary strategy for pathway optimization. 3.3. Improving enzyme solubility through fusion tag engineering to enhance melatonin production SDS-PAGE analysis ( Fig. S1 ) confirmed successful expression of all pathway enzymes, but their solubility remained suboptimal. To address this limitation, a comprehensive solubility-enhancing fusion tag library was constructed [ 40 ], including SUMO [ 41 ], MBP [ 42 ], GST [ 43 ], TrxA [ 44 ], NusA [ 45 ], GB1 [ 46 ], DsbC [ 47 ], and NT11 [ 48 ]. SUMO (small ubiquitin-like modifier) enhances protein stability through post-translational conjugation, whereas MBP (maltose-binding protein) is widely used to improve recombinant protein solubility in E. coli . GST (glutathione S -transferase) can function as a molecular chaperone to facilitate proper protein folding. TrxA (thioredoxin) reduces inclusion body formation via its oxidoreductase activity, while NusA stabilizes nascent polypeptides during translation. GB1 promotes compact folding and high-level expression, DsbC facilitates disulfide bond formation and rearrangement of misoxidized proteins, and NT11, derived from the N-terminal half of a duplicated carbonic anhydrase, has been reported to markedly enhance protein production. Each tag was individually fused to the N-terminus of the target enzymes to systematically identify optimal tag-enzyme combinations. Initially, fusion of these tags to the N-terminus of Luz15 did not improve melatonin production ( Fig. 3 A and E), indicating that Luz15 is predominantly soluble and that additional fusion tags may interfere with its structural integrity or catalytic performance. In contrast, fusion of the solubility-enhancing tag library to Ddc yielded variable effects, resulting in melatonin titers corresponding to 95.88%, 51.74%, 139.91%, 66.09%, 18.43%, 65.07%, 26.91%, and 107.60% of the original level, respectively ( Fig. 3 B and F). Among these, GST and NT11 fusions significantly enhanced melatonin production, likely by improving the soluble expression of Ddc. Fig. 3. Open in a new tab Establishment of solubility-enhancing fusion tag library to improve pathway enzymes expression. (A), (B), (C) and (D) Relative melatonin yields of strains expressing Luz15, Ddc, Aanat and ASMT fused with different solubility tags successively. Data are normalized to the corresponding strains. (The production of the strains expressing Luz15 or Ddc fused with tags was compared with MT14, whereas relative yields of strains expressing tagged Aanat or ASMT were calculated versus MT24 or MT30.) (E) SDS-PAGE analysis of expression of Luz15 fused with tags. (M: protein marker, WT: wild-type Luz15, SUMO: Luz15 fused with tag SUMO, MBP: Luz15 fused with tag MBP, GST: Luz15 fused with tag GST, TrxA: Luz15 fused with tag TrxA, NusA: Luz15 fused with tag NusA, GB1: Luz15 fused with tag GB1, DsbC: Luz15 fused with tag DsbC.) The upon gel picture is representative of cell lysate supernatant. The below gel picture exhibits cell lysate precipitation. (F), (G) and (H) adopt the similar representation with (E). (I) Schematic overview of the strategy of establishing fusion tag library by stepwise fusion of tags to the N-terminus of enzymes and the best configuration and the final titer in MT38. In (A), (B), (C), (D) and (I), all the data are presented as mean values ± S.D. from three independent biological replicates (n = 3). Significance (P value) was evaluated by bilateral t -test, with ∗, ∗∗, and ∗∗∗ indicating P values < 0.05, <0.01, and <0.001, respectively. Subsequently, SUMO was identified as the optimal fusion tag for Aanat, leading to a 12.87% increase in melatonin titer ( Fig. 3 C and G), while screening of ASMT fusion tags resulted in an additional 12.44% improvement ( Fig. 3 D and H). Through this stepwise optimization, the combination of GST-tagged ckDdc, SUMO-tagged sgAanat and NT11-tagged ASMT was identified as the most effective configuration. Under this optimized setup, melatonin production reached 60.39 ± 7.29 mg/L ( Fig. 3 I). SDS-PAGE analysis further confirmed that fusion tag engineering substantially enhanced enzyme solubility. These results indicate that enhancing the soluble and highly active expression of enzymes in the melatonin biosynthetic pathway is an effective strategy for improving melatonin production. 3.4. Structure-guided protein engineering of Luz15 improves hydroxylation capacity Protein engineering represents a powerful strategy for optimizing biosynthetic pathways, as both rational redesign of catalytic centers and mutations at distal sites has been shown to critically influence catalytic activity, protein expression, and thermal stability [ [49] , [50] , [51] , [52] ]. Previous studies have shown that the AlphaFold-predicted structure of Luz15 closely resembles the crystal structure of canonical tryptophan 2,3-dioxygenase (TDO), an enzyme responsible for oxidative cleavage of the indole ring of tryptophan. Structural superposition of Luz15 with TDO revealed the approximate reaction region accommodating the enzyme, substrate tryptophan, and cofactor heme, and identified a C-terminal TDO-like domain together with a conserved N-terminal YXXY motif [ 18 ]. Based on these insights, we aligned sequences from the bTPH family ( Fig. S8 ) and performed molecular docking to preliminarily partition Luz15 into a catalytic region and a distal stability region ( Fig. 4 A). To systematically explore the sequence space of the distal stability region, in silico saturation mutagenesis was conducted using pyRosetta. Each single point variant was subjected to structural relaxation and evaluated using the Rosetta energy function, with variants exhibiting lower scores than the wild type considered potentially more stable. This analysis yielded 21 target residues generating 263 variants and among them we picked 6 variants, C23S, C23T, P113L, P305A, R307Q and L319P, for further evaluation ( Fig. S9 ). In parallel, cluster analysis of Luz15 with 1586 homologous enzymes revealed a low sequence conservation of approximately 19.63% within the catalytic region ( Fig. S10 ), indicating substantial evolutionary divergence and providing guidance for identifying residues critical for catalysis. Directed with homology analysis, we constructed 14 mutants, Y106D, F229I, I233Q, F236E, Y240K, L283E, F291L, R295D, S302G, A303F, H375L, I383G, Y397L and L398R to assess catalytic efficiency. Fig. 4. Open in a new tab Catalysis mechanism and rational design of Luz15. (A) Regional structure of Luz15 divided into two parts, a catalysis region and a distal stability region, and the key amino acid residues which interact with cofactor heme and substrate tryptophan. (B) The 5-HTP relative yields of single-point mutants compared to WT after 36 h 24-well-plate fermentation. (C) The melatonin relative yield of single-point mutants and combination mutants versus WT after 36 h shake-flask fermentation. (D) The local structure of Luz15 wild type and the best mutant D299S/W376H. (E) RMSD analysis of Luz15 and its mutant D299S/W376H at 300 K. (F) RMSF analysis of Luz15 and its mutant D299S/W376H at 300 K. In (B) and (C), all the data are presented as mean values ± S.D. from three independent biological replicates (n = 3). To further elucidate the catalytic mechanism of Luz15, molecular docking was employed to analyze interactions between the enzyme, tryptophan and heme. Thirteen residues were identified to interact with heme via hydrophobic interactions, hydrogen bonding and π-π stacking ( Fig. 4 A). Among these, five residues played pivotal roles in ligand positioning and stabilization. Specifically, Arg295 and Thr298 formed salt bridges with the amino and carboxyl groups of tryptophan, Ala303 and Tyr106 anchored the indole moiety through hydrophobic interactions, and Arg301 modulated indole ring orientation via side chain interactions, collectively enabling regioselective C5 hydroxylation of tryptophan. Therefore, we additionally created 7 mutants in catalytic region for further examination, including T298D, T298E, T298S, D299S, D299E, W376D and W376H. Guided by these structural and mechanistic insights, totally 27 Luz15 variants were constructed to enhance catalytic activity. High-throughput screening based on relative 5-HTP production identified four superior single point mutants, C23T, T298S, D299S and W376H, exhibiting activity increases of 59.71%, 66.54%, 135.53% and 109.73%, respectively ( Fig. 4 B). The C23T mutant, located in the distal region, likely improved solubility and flexibility by preventing disulfide bond formation. T298S relieved steric hindrance and strengthened electrostatic interactions with the amino group of tryptophan. Substitution of Asp299 with Serine expanded the substrate binding pocket, facilitating substrate entry, while W376H potentially promoted the formation of a stabilizing salt bridge between adjacent alpha helices ( Fig. S11 ). These beneficial mutants were subsequently introduced individually into the complete melatonin biosynthetic pathway for secondary screening. Among them, D299S exhibited the greatest enhancement in melatonin production and was therefore selected as the parent for combinatorial mutagenesis. Three double mutants and one triple mutant were constructed, of which D299S/W376H displayed the highest activity, increasing the melatonin titer by 33.81% to 80.74 ± 1.75 mg/L ( Fig. 4 C). In vitro biochemical characterization of purified enzymes revealed that the variant D299S/W376H exhibited a 2.62-fold increase in the turnover number ( k cat ) and an enhanced overall catalytic efficiency ( k cat / K m ) which increasing from 2.18 ± 0.46 h −1 mM −1 to 2.79 ± 0.55 h −1 mM −1 , confirming a substantial improvement in its intrinsic hydroxylation activity ( Fig. S12 and Table S4 ). To elucidate the molecular basis underlying this improvement, protein conformational analysis and molecular dynamics simulations were performed. Replacement of Trp376 with His enabled the imidazole side chain of His376 to form a salt bridge with the carboxylate group of heme, thereby strengthening cofactor binding ( Fig. 4 D). RMSD analysis demonstrated that the D299S/W376H mutant reaches structural equilibrium more rapidly than the wild-type enzyme ( Fig. 4 E), while root RMSF analysis revealed increased flexibility in regions spanning residues 297 to 299 and 385 to 393, which likely facilitates substrate and cofactor access to the active site ( Fig. 4 F). 3.5. sRNA-guided identification of metabolic targets to enhance tryptophan availability Owing to the high cost and limited economic feasibility of directly supplementing tryptophan in the fermentation medium, the establishment of a robust chassis strain with optimized intrinsic tryptophan biosynthetic capacity is essential. However, endogenous tryptophan biosynthesis in E. coli is governed by a highly complex and tightly regulated metabolic network ( Fig. 5 A), which is subject to multilayered control mechanisms, including transcriptional repression, feedback inhibition and transcriptional attenuation. These regulatory constraints significantly limit intracellular tryptophan accumulation and must be systematically relieved to enable efficient downstream melatonin biosynthesis. Fig. 5. Open in a new tab Large-scale screening of target genes with an sRNA library and CRISPR/Cas9 system to improve melatonin biosynthesis from glucose. (A) Metabolic pathway of substrate tryptophan and cofactor SAM and AcCoA. The pathway of tryptophan biosynthesis is outlined in orange. The overexpressed genes are marked in red while the effectively inhibited genes are marked in blue. (B) Construction of effective tryptophan-producing chassis. The concentration of tryptophan in culture was measured after 36 h shake-flask fermentation by HPLC. (C) Streamlined workflow of the sRNA library construction and target screening. One-pot PCR was performed with 88 pairs of primer mixture to generate sRNA library and then the library was transformed into strain harboring plasmid pET-P lamdalac -GST- ckddc -SUMO- sgaanat -P T7 -NT11- asmt . Transformants were picked randomly and inoculated in 24-well plates. After fermentation, HPLC was utilized to quantify the melatonin titer of each transformant. The top several transformants were selected for gene sequencing and secondary screening. (D) Primary screening of the sRNA library at the 24-well-plate level. The melatonin titers of fermentation broth in each well were measured with HPLC. (E) Secondary screening of the sRNA library at the shake-flask level. Melatonin production and OD 600 of strains inhibiting 8 different genes respectively were analyzed compared with the control and the function of efficacious genes was displayed in table. In (B) and (E), all the data are presented as mean values ± S.D. from three independent biological replicates (n = 3). To reduce intracellular tryptophan degradation, the gene tnaA , encoding tryptophanase responsible for tryptophan catabolism, was deleted as this strategy has been previously demonstrated to effectively enhance tryptophan retention [ 53 ]. In addition, in E. coli , the tryptophan repressor TrpR, encoded by trpR , forms a complex with tryptophan under sufficient intracellular concentration and binds to the trp operon to repress its transcription [ 54 ]. Accordingly, a double knockout strain, E. coli BL21(DE3) Δ tnaA Δ trpR , designated Trp1, was constructed as the initial chassis for further metabolic engineering. Beyond transcriptional regulation, feedback inhibition at the enzymatic level also poses a major bottleneck. The gene trpE encodes anthranilate synthase component I, which catalyzes the first committed step of tryptophan biosynthesis and is highly sensitive to feedback inhibition by tryptophan. To overcome this limitation, a feedback resistant variant harboring mutants TrpE S40F, Q71K, S94N, C465Y (TrpE fbr ) was engineered [ [55] , [56] , [57] ]. A P T7 - trpE fbr expression cassette was subsequently integrated into the chromosomal malEFG locus, yielding strain Trp2. Building upon this foundation, further optimization of the tryptophan biosynthetic pathway was pursued by modular engineering, which can be conceptually divided into three interconnected modules: central carbon metabolism, the shikimate (SK) pathway, and the chorismate (CHO) pathway. Phosphoenolpyruvate (PEP) and erythrose-4-phosphate (E4P), which serve as essential precursors for aromatic amino acid biosynthesis, are condensed by 3-deoxy-D-arabino heptulosonate 7-phosphate synthase (DAHPS) to form DAHP, the first committed intermediate of the SK pathway. In E. coli , DAHPS is encoded by three isoenzymes ( aroG , aroF , and aroH ), among which aroG plays the dominant role in directing carbon flux toward aromatic amino acid biosynthesis. To alleviate feedback inhibition on DAHPS and enhance precursor flux into the shikimate pathway, a feedback resistant mutant, AroG S211F (AroG fbr ) was constructed [ 58 , 59 ]. A P T7 - aroG fbr - aroC expression cassette was integrated into the chromosomal ymgF locus, resulting in strain Trp3. This modification significantly increased tryptophan production to 15.59 ± 1.10 mg per liter, showcasing over 7 times than production of the initial strain and thereby establishing a efficient tryptophan producing chassis ( Fig. 5 B). To redirect metabolic flux toward target compounds, programmable transcriptional repression tools offer a precise, non-editing strategy for downregulating competing or bypass pathways [ [60] , [61] , [62] ]. Given the fragile regulation of tryptophan metabolism and the additional metabolic demands imposed by melatonin biosynthesis, we therefore employed a post-transcriptional sRNA library to identify potential gene targets for metabolic flux optimization [ 22 ]. A de novo synthetic sRNA library was designed to target 88 candidate genes involved in competing pathways or cofactor metabolism, including those related to tryptophan consumption as well as S -adenosylmethionine (SAM) and acetyl-CoA (AcCoA) biosynthesis. The impact of repressing different targets on melatonin production was subsequently evaluated ( Fig. 5 C). This high throughput sRNA-based screening platform enabled rapid identification of novel regulatory targets that constrain melatonin synthesis and provided valuable guidance for subsequent strain optimization. For primary screening, 117 sRNA transformants together with three parental strains were cultivated in 24-well plates. HPLC analysis revealed that more than 70% of the transformants exhibited improved melatonin production relative to the control strains ( Fig. 5 D). The top 15 performing strains were selected for sequencing to identify the corresponding repressed genes. Based on this analysis, eight candidate targets were subjected to secondary screening at the shake-flask level. Ultimately, six genes were identified as promising regulatory targets, including pfkA encoding ATP-dependent 6-phosphofructokinase, tyrA encoding prephenate dehydrogenase, sdhA encoding the flavoprotein subunit of succinate dehydrogenase, csrB encoding a regulatory non-coding RNA, sucD encoding succinyl-CoA ligase subunit alpha, and aceA encoding isocitrate lyase. Repression of pfkA , sdhA , sucD , and aceA likely reshapes central carbon metabolism by attenuating glycolysis and the tricarboxylic acid cycle, thereby increasing the availability of E4P and reducing power in the form of NADPH, both of which are critical for tryptophan biosynthesis [ [63] , [64] , [65] , [66] ]. In contrast, downregulation of tyrA effectively blocks carbon flux toward the competing tyrosine biosynthetic branch, redirecting chorismate toward tryptophan formation [ [67] , [68] , [69] ]. Among all evaluated strains, repression of tyrA resulted in the most pronounced and stable improvement in melatonin production. The corresponding strain, MT57 achieved a melatonin titer of 100.72 ± 1.08 mg per liter ( Fig. 5 E). These results collectively demonstrate that strengthening the intrinsic tryptophan biosynthetic capacity of the host strain is sufficient to sustain high melatonin productivity even in the absence of exogenous tryptophan supplementation, thereby providing a more economical and scalable platform for melatonin biosynthesis. 3.6. Fed-batch fermentation of engineered E. coli enables high-level melatonin production To further evaluate the production potential of the engineered Luz15 variant D299S/W376H without tryptophan supplement, a 5-L fed-batch fermentation was conducted using strain MT57. Throughout the entire fermentation process, the temperature was maintained at 30 °C, while the pH was controlled at 6.5 via automatic supplementation with 30% aqueous ammonia solution. Glucose feeding was regulated automatically based on real time glucose consumption, ensuring that the residual glucose concentration was dynamically maintained within the range of 0 to 1 g per liter. At 8 h post inoculation, IPTG was added with a final concentration of 0.1 mM to induce gene expression, together with supplementation of methionine at 1 g per liter. Cell growth and melatonin accumulation were continuously monitored during the fermentation process. After 72 h of cultivation, the cell density reached an OD 600 of 105.79, and melatonin titer peaked at 753.78 ± 32.37 mg/L with a volumetric productivity of 10.46 mg/L/h and a carbon yield from glucose of 1.88 mg/g. The titer represented a 7.5-fold increase compared with shake-flask fermentation and was 38.78-fold higher than that of the initial production strain ( Fig. 6 ). Notably, melatonin accumulation closely correlated with biomass formation during the fed-batch process, indicating that cellular growth capacity strongly influences overall production performance. These results suggest that further refinement of feeding strategies and process control parameters, such as carbon source delivery and induction timing, may provide additional improvements in melatonin titer. Moreover, beyond ensuring sufficient intracellular tryptophan availability, enhancing acetyl-CoA generation and optimizing the S -adenosylmethionine regeneration cycle are expected to be effective strategies for further boosting melatonin biosynthesis in future process development. Fig. 6. Open in a new tab Fed-batch fermentation of E. coli strain MT57 from glucose. Fed-batch fermentation was performed in a 5-L bioreactor with 2 L fermentation medium. The time course of cell growth, glucose consumption and melatonin production were investigated in detail. Data are presented as mean values ± SD from three independent biological replicates (n = 3). 4. Conclusion In this study, we successfully established a microbial melatonin biosynthetic pathway in E . coli by incorporating the newly identified bacterial tryptophan hydroxylase Luz15. Through systematic integration of metabolic pathway engineering and structure-guided protein engineering, a high-performance melatonin producing strain with clear industrial relevance was constructed. Targeted optimization of intracellular heme availability, together with systematic screening of solubility enhancing fusion tags, substantially improved pathway efficiency and enzyme functionality. Guided by comprehensive analyses of enzyme stability, sequence conservation, and enzyme–cofactor–substrate interactions, the Luz15 variant D299S/W376H was identified as the optimal mutant, resulting in a 33.81% increase in melatonin production relative to the wild-type enzyme. In parallel, in depth interrogation of the metabolic network underlying melatonin biosynthesis, combined with the application of a de novo sRNA repression library, effectively alleviated constraints imposed by precursor tryptophan availability and redirected metabolic flux toward the target pathway. Ultimately, the final engineered strain MT57 harboring the Luz15 D299S/W376H variant achieved a melatonin titer of 753.78 ± 32.37 mg per liter in a 5-L fed batch bioreactor without exogenous tryptophan supplementation, corresponding to a 38.78 folds improvement compared with the initial production strain. Collectively, the integration of metabolic engineering and enzyme engineering strategies enables the construction of a melatonin-producing strain and provides a valuable foundation for further optimization and industrial application. These results demonstrate the robustness and scalability of the newly established melatonin biosynthetic route and highlight its potential as a versatile platform for the microbial production of melatonin and other high value tryptophan derived compounds. CRediT authorship contribution statement Luyao Zhang: Writing – original draft, Visualization, Methodology, Investigation, Data curation. Guobin Yin: Investigation, Funding acquisition, Conceptualization. Senyu Pan: Investigation. Xinai Zheng: Validation. Siyan Zhou: Writing – review & editing, Investigation. Guocheng Du: Supervision. Jianghua Li: Supervision. Jian Chen: Supervision, Funding acquisition. Ruirui Xu: Writing – review & editing, Validation, Resources, Funding acquisition. Zhen Kang: Writing – review & editing, Validation, Resources, 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. Acknowledgements This work was financially supported by the National Key Research and Development Program of China (2024YFF1106300), the National Natural Science Foundation of China (U24A20368, 32370066), the Jiangsu Basic Research Center for Synthetic Biology (BK20233003), the Fundamental Research Funds for the Central Universities (JUSRP622003), the China Postdoctoral Science Foundation (2025M772527), the Postdoctoral Fellowship Program of CPSF (GZC20251638), the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX23_2493) and the Guangdong S&T Program (2024B1111160002). 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.2026.03.020 . Contributor Information Ruirui Xu, Email: [email protected]. Zhen Kang, Email: [email protected]. Appendix A. Supplementary data The following is the Supplementary data to this article: Multimedia component 1 mmc1.docx (5.9MB, docx) References 1. Arnao M.B., Hernández-Ruiz J. Melatonin and its relationship to plant hormones. Ann Bot. 2017;121:195–207. doi: 10.1093/aob/mcx114. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Sheikh B.A., Aarif A., Midhat B., Shahzada M.R., Amir B.W., Rahil R.B., Muneeb U.R. Melatonin and health: insights of melatonin action, biological functions, and associated disorders. Cell Mol Neurobiol. 2023;43:2437–2458. doi: 10.1007/s10571-023-01324-w. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 3. Russel J.R., Ramaswamy S., Maira S.C., Dun X.T., Sergio R.C., Giuseppe G., Luiz G.A.C. Brain washing and neural health: role of age, sleep, and the cerebrospinal fluid melatonin rhythm. Cell Mol Life Sci. 2023;80:88. doi: 10.1007/s00018-023-04736-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 4. Lerner A.B., Case J.D., Takahashi Y., Lee T.H., Mori W. Isolation of melatonin, the pineal gland factor that lightens melanocytes1. J Am Chem Soc. 1958;80:2587. doi: 10.1021/ja01543a060. [ DOI ] [ Google Scholar ] 5. Szmuszkovicz J., Anthony W., Heinzelman R. Synthesis of N-acetyl-5-methoxytryptamine. J Org Chem. 1960;25:857–859. doi: 10.1021/jo01075a623. [ DOI ] [ Google Scholar ] 6. He L., Li J.L., Zhang J.J., Su P., Zheng S.L. Microwave assisted synthesis of melatonin. Synth Commun. 2003;33:741–747. doi: 10.1081/SCC-120016317. [ DOI ] [ Google Scholar ] 7. Verspui G., Elbertse G., Sheldon F.A., Hacking M.A.P.J., Sheldon R.A. Selective hydroformylation of N-allylacetamide in an inverted aqueous two-phase catalytic system, enabling a short synthesis of melatonin. Chem Commun. 2000;15:1363–1364. doi: 10.1039/B003715J. [ DOI ] [ Google Scholar ] 8. Wang Q., Li C.C., Yuan B., Yu A.Q., Qu G., Sun Z.T. Engineering the activity of a newly identified arylalkylamine N-acetyltransferase in the acetylation of 5-hydroxytryptamine. Chembiochem. 2024;25(9) doi: 10.1002/cbic.202400069. [ DOI ] [ PubMed ] [ Google Scholar ] 9. Xie X., Ding D.Q., Bai D.Y., Zhu Y.R., Sun W., Sun Y.M., Zhang D.W. Melatonin biosynthesis pathways in nature and its production in engineered microorganisms. Synth Syst Biotechnol. 2022;7:544–553. doi: 10.1016/j.synbio.2021.12.011. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 10. Zhang Y.F., He Y.Z., Zhang N., Gan J.J., Zhang S., Dong Z.Y. Combining protein and metabolic engineering strategies for biosynthesis of melatonin in Escherichia coli. Microb Cell Fact. 2021;20:170. doi: 10.1186/s12934-021-01662-8. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 11. Germann S.M., Jacobsen S.A.B., Schneider K., Harrison S.J., Jensen N.B., Chen X., Stahlhut S.G., Borodina I., Luo H., Zhu J.F., Maury J., Forster J. Glucose-based microbial production of the hormone melatonin in yeast Saccharomyces cerevisiae. Biotechnol J. 2016;11:717–724. doi: 10.1002/biot.201500143. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Xiao S.J., Wang Z., Wang B.X., Hou B., Cheng J., Bai T., Zhang Y., Wang W., Yan L.X., Zhang J.M. Expanding the application of tryptophan: industrial biomanufacturing of tryptophan derivatives. Front Microbiol. 2023;14 doi: 10.3389/fmicb.2023.1099098. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Back K., Tan D.X., Reiter R.J. Melatonin biosynthesis in plants: multiple pathways catalyze tryptophan to melatonin in the cytoplasm or chloroplasts. J Pineal Res. 2016;61:426–437. doi: 10.1111/jpi.12364. [ DOI ] [ PubMed ] [ Google Scholar ] 14. Byeon Y., Back K. Melatonin production in Escherichia coli by dual expression of serotonin N-acetyltransferase and caffeic acid O-methyltransferase. Appl Microbiol Biotechnol. 2016;100:6683–6691. doi: 10.1007/s00253-016-7458-z. [ DOI ] [ PubMed ] [ Google Scholar ] 15. Wang H.J., Liu W.Q., Shi F., Huang L., Lian J.Z., Qu L., Cai J., Xu Z.N. Metabolic pathway engineering for high-level production of 5-hydroxytryptophan in Escherichia coli. Metab Eng. 2018;48:279–287. doi: 10.1016/j.ymben.2018.06.007. [ DOI ] [ PubMed ] [ Google Scholar ] 16. Liu W.W., Yu Z.C., Chen Z.C., Ma L., Liu Y.H., Zhang Z., Xu Q.Y. Multiple metabolic strategies for green and efficient 5-hydroxytryptamine biosynthesis in Escherichia coli. ACS Sustainable Chem Eng. 2025;13:3706–3717. doi: 10.1021/acssuschemeng.4c10140. [ DOI ] [ Google Scholar ] 17. Thöny B., Auerbach G., Blau N. Tetrahydrobiopterin biosynthesis, regeneration and functions. Biochem J. 2000;347:1–16. doi: 10.1016/j.phrs.2020.104827. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Shi X.J., Zhao G.Y., Li H., Zhao Z.J., Li W., Wu M.L., Du Y.L. Hydroxytryptophan biosynthesis by a family of heme-dependent enzymes in bacteria. Nat Chem Biol. 2023;19:1415–1422. doi: 10.1038/s41589-023-01416-0. [ DOI ] [ PubMed ] [ Google Scholar ] 19. Tamura K., Stecher G., Kumar S. MEGA11: molecular evolutionary genetics analysis version 11. Mol Biol Evol. 2021;38:3022–3027. doi: 10.1093/molbev/msab120. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 20. Gouet P., Courcelle E., Stuart D.I., Metoz F. ESPript: analysis of multiple sequence alignments in PostScript. Bioinformatics. 1999;15:305–308. doi: 10.1093/bioinformatics/15.4.305. [ DOI ] [ PubMed ] [ Google Scholar ] 21. Wang B.B., Xu J.Z., Liu S., Rao Z.M., Zhang W.G. Engineering of human tryptophan hydroxylase 2 for efficient synthesis of 5-hydroxytryptophan. Int J Biol Macromol. 2024;260 doi: 10.1016/j.ijbiomac.2024.129484. [ DOI ] [ PubMed ] [ Google Scholar ] 22. Wang Y., Yin G.B., Weng H.J., Zhang L.Y., Du G.C., Chen J., Kang Z. Gene knockdown by structure defined single-stem loop small non-coding RNAs with programmable regulatory activities. Synth Syst Biotechnol. 2023;8:86–96. doi: 10.1016/j.synbio.2022.11.006. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Yin G.B., Peng A.Q., Zhang L.Y., Wang Y., Du G.C., Chen J., Kang Z. Design of artificial small regulatory trans-RNA for gene knockdown in Bacillus subtilis. Synth Syst Biotechnol. 2023;8:61–68. doi: 10.1016/j.synbio.2022.11.003. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Hopp M.T., Vaidya S.M., Grimming K.M., Srtudthoff L.J., Clauser J.C., Yuan X.J., Singh S., Müller J., Oldenburg J., Hamza I., Imhof D. Quantitative analysis of heme and hemoglobin for the detection of intravascular hemolysis. Anal Chim Acta. 2024;1312 doi: 10.1016/j.aca.2024.342766. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Hara R., Kino K. Enhanced synthesis of 5-hydroxy-L-tryptophan through tetrahydropterin regeneration. AMB Express. 2013;3:70. doi: 10.1186/2191-0855-3-70. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Luo H., Yang L., Kim S.H., Wulff T., Feist A.M., Herrgård M., Palsson B.O. Directed metabolic pathway evolution eables functional pterin-dependent aromatic-amino-acid hydroxylation in Escherichia coli. ACS Synth Biol. 2020;9:494–499. doi: 10.1021/acssynbio.9b00488. [ DOI ] [ PubMed ] [ Google Scholar ] 27. Luo H., Schneider K., Christensen U., Yang L., Herrgård M., Palsson B.O. Microbial synthesis of human-hormone melatonin at gram scales. ACS Synth Biol. 2020;9:1240–1245. doi: 10.1021/acssynbio.0c00065. [ DOI ] [ PubMed ] [ Google Scholar ] 28. Chen H.H., Wang Y.H., Wang W.S., Cao T., Zhang L., Wang Z.D., Chi X.R., Shi T., Wang H.W., He X.W., Liang M.D., Yang M.X., Jiang W.Y., Lv D.Y., Yu J.M., Zhu G.L., Xie Y.T., Gao B., Wang X.Y., Liu X.T., Li Y.Y., OuYang L.M., Zhang J.Y., Liu H.M., Li Z.L., Tong Y.J., Xia X.K., Tan G.Y., Zhang L.X. High-yield porphyrin production through metabolic engineering and biocatalysis. Nat Biotechnol. 2025;43:1717–1727. doi: 10.1038/s41587-024-02267-3. [ DOI ] [ PubMed ] [ Google Scholar ] 29. Anzaldi L.L., Skaar E.P. Overcoming the heme paradox: heme toxicity and tolerance in bacterial pathogens. Infect Immun. 2010;78:12. doi: 10.1128/iai.00613-10. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Yang S.M., Guo Z.H., Sun J.Y., Wei J.X., Ma Q.Y., Gao X.Z. Recent advances in microbial synthesis of free heme. Appl Microbiol Biotechnol. 2024;108:68. doi: 10.1007/s00253-023-12968-5. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Choi K.R., Yu H.E., Lee H., Lee S.Y. Improved production of heme using metabolically engineered Escherichia coli. Biotechnol Bioeng. 2022;19:3178–3193. doi: 10.1002/bit.28194. [ DOI ] [ PubMed ] [ Google Scholar ] 32. Zhao X.R., Choi K.R., Lee S.Y. Metabolic engineering of Escherichia coli for secretory production of free haem. Nat Catal. 2018;1:720–728. doi: 10.1038/s41929-018-0126-1. [ DOI ] [ Google Scholar ] 33. Verkamp E., Backman V.M., Björnsson J.M., Söll D., Eggertsson G. The periplasmic dipeptide permease system transports 5-aminolevulinic acid in Escherichia coli. J Bacteriol. 1993;175:1452–1456. doi: 10.1128/jb.175.5.1452-1456.1993. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Delepelaire P. Bacterial ABC transporters of iron containing compounds. Res Microbiol. 2019;170:345–357. doi: 10.1016/j.resmic.2019.10.008. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Hu B.D., Zhao X.R., Zhou J.W., Li J.H., Chen J., Du G.C. Efficient hydroxylation of flavonoids by using whole-cell P450 sca-2 biocatalyst in Escherichia coli. Front Bioeng Biotechnol. 2023;11 doi: 10.3389/fbioe.2023.1138376. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 36. Kato S., Abe M., Qkahashi N., Ariyasu S., Matsuda F., Shoji O., Hayashi T. Metabolic engineering of the 5-aminolevulinate biosynthetic pathway in E. coli improves efficiency of hemoprotein-based biocatalysis. Angew Chem Int Ed. 2025;64(39) doi: 10.1002/anie.202512156. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Geng Z.X., Ge J.X., Cui W., Zhou H., Deng J.Y., Xu B.C. Efficient de novo biosynthesis of heme by membrane engineering in Escherichia coli. Int J Mol Sci. 2022;23 doi: 10.3390/ijms232415524. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Sun C.X., Hu B.D., Li Y.C., Wu Z.M., Zhou J.W., Li J.H., Chen J., Du G.C., Zhao X.R. Efficient stereoselective hydroxylation of deoxycholic acid by the robust whole-cell cytochrome P450 CYP107D1 biocatalyst. Synth Syst Biotechnol. 2023;8:741–748. doi: 10.1016/j.synbio.2023.11.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Wang W.Q., Xiang Y.L., Yin G.B., Hu S., Cheng J., Chen J., Du G.C., Kang Z., Wang Y. Construction of 5-aminolevulinic acid microbial cell factories through identification of novel synthases and metabolic pathway screens and transporters. J Agric Food Chem. 2024;72:8006–8017. doi: 10.1021/acs.jafc.4c00903. [ DOI ] [ PubMed ] [ Google Scholar ] 40. Yang F.L., Sun X.Y., Lu J., Zhang L.P., Du G.C., Li J.H., Xu R.R., Kang Z. Immobilized high-performance heparin lyase III for efficient preparation of low molecular weight heparin. Int J Biol Macromol. 2024;280 doi: 10.1016/j.ijbiomac.2024.135833. [ DOI ] [ PubMed ] [ Google Scholar ] 41. Kim K.I., Baek S.H., Chung C.H. Versatile protein tag, SUMO: its enzymology and biological function. J Cell Physiol. 2002;191:257–268. doi: 10.1002/jcp.10100. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Kapust R.B., Waugh D.S. Escherichia coli maltose-binding protein is uncommonly effective at promoting the solubility of polypeptides to which it is fused. Protein Sci. 1999;8:1668–1674. doi: 10.1110/ps.8.8.1668. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Allocati N., Federici L., Masulli M., Carmine D.I. Glutathione transferases in bacteria. FEBS J. 2009;276:58–75. doi: 10.1111/j.1742-4658.2008.06743.x. [ DOI ] [ PubMed ] [ Google Scholar ] 44. Costa S., Almeida A., Castro A., Domingues L. Fusion tags for protein solubility, purification and immunogenicity in Escherichia coli: the novel Fh8 system. Front Microbiol. 2014;5:63. doi: 10.3389/fmicb.2014.00063. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Marco A.D. Two-step metal affinity purification of double-tagged (NusA-His6) fusion proteins. Nat Protoc. 2006;1:1538–1543. doi: 10.1038/nprot.2006.289. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Song S.J., Diao H.P., Moon B., Yun A., Hwang I. The B1 domain of streptococcal protein G serves as a multi-functional tag for recombinant protein production in plants. Front Plant Sci. 2022;13 doi: 10.3389/fpls.2022.878677. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 47. Haebel P.W., Goldstone D., Katzen F., Beckwith J., Metcalf P. The disulfide bond isomerase DsbC is activated by an immunoglobulin-fold thiol oxidoreductase: crystal structure of the DsbC-DsbDalpha complex. EMBO J. 2002;21:4774–4784. doi: 10.1093/emboj/cdf489. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 48. Nguyen T.K.M., Ki M.R., Son R.G., Pack S.P. The NT11, a novel fusion tag for enhancing protein expression in Escherichia coli. Appl Microbiol Biotechnol. 2019;103:2205–2216. doi: 10.1007/s00253-018-09595-w. [ DOI ] [ PubMed ] [ Google Scholar ] 49. Jiang P., Jin H.B., Zhang G.T., Zhang W.J., Liu W., Zhu Y.G., Zhang C.S., Zhang L.P. A mchanistic uderstanding of the dstinct rgio- and cemoselectivity of mltifunctional P450s by sructural cmparison of IkaD and CftA cmplexed with cmmon sbstrates. Angew Chem Int Ed. 2023;2 doi: 10.1002/anie.202310728. [ DOI ] [ PubMed ] [ Google Scholar ] 50. Grobe S., Badenhorst C., Bayer T., Hamnevik E., Wu S., Grathwol C., Link A., Koban S., Brundiek H., Großjohann B., Bornscheuer U. Engineering rgioselectivity of a P450 mnooxygenase eables the synthesis of ursodeoxycholic acid via 7β-hydroxylation of lithocholic acid. Angew Chem Int Ed. 2021;60:753–757. doi: 10.1002/anie.202012675. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 51. Yang F.L., Xu R.R., Huang H., Lu J., Feng J.M., Sun X.Y., Li J.H., Du G.C., Kang Z. Controllable polysaccharide depolymerization by heparinases for efficient low molecular weight heparin production. Carbohydr Polym Tech. 2026;13 doi: 10.1016/j.carpta.2026.101087. [ DOI ] [ Google Scholar ] 52. Xu R.R., Zhang W.J., Xi X.T., Chen J.M., Wang Y., Du G.C., Li J.H., Chen J., Kang Z. Engineering sulfonate group donor regeneration systems to boost biosynthesis of sulfated compounds. Nat Commun. 2023;14:7297. doi: 10.1038/s41467-023-43195-1. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 53. Li G., Young K.D. Indole production by the tryptophanase TnaA in Escherichia coli is determined by the amount of exogenous tryptophan. Microbiology. 2013;159:402–410. doi: 10.1099/mic.0.064139-0. [ DOI ] [ PubMed ] [ Google Scholar ] 54. YangJ Gunasekera A., Lavoie T.A., Jin L.H., Lewis D.E., Carey J. In vivo and in vitro studies of TrpR-DNA interactions. J Mol Biol. 1996;258:37–52. doi: 10.1006/jmbi.1996.0232. [ DOI ] [ PubMed ] [ Google Scholar ] 55. Li H.B., Zhang J.M., Gao J.X., Luo Y.Z. Advances in the microbial biosynthesis of L-tryptophan and its derivatives. BioDesign Res. 2025;7 doi: 10.1016/j.bidere.2025.100046. [ DOI ] [ Google Scholar ] 56. Caligiuri M.G., Bauerle R. Subunit communication in the anthranilate synthase complex from Salmonella typhimurium. Science. 1991;252:1845–1848. doi: 10.1126/science.2063197. [ DOI ] [ PubMed ] [ Google Scholar ] 57. Zhang Z., Yu Z.C., Wang J.D., Yu Y.F., Li L.X., Sun P.J., Fan X.G., Xu Q.Y. Metabolic engineering of Escherichia coli for efficient production of L-5-hydroxytryptophan from glucose. Microb Cell Fact. 2022;21:198. doi: 10.1186/s12934-022-01920-3. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 58. Du L.H., Zhang Z., Xu Q.Y., Chen N. Central metabolic pathway modification to improve L-tryptophan production in Escherichia coli. Bioengineered. 2019;10:59–70. doi: 10.1080/21655979.2019.1592417. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 59. Liu S., Wang B.B., Xu J.Z., Zhang W.G. Engineering of shikimate pathway and terminal branch for efficient production of L-tryptophan in Escherichia coli. Int J Mol Sci. 2023;24 doi: 10.3390/ijms241411866. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 60. Elhadi D., Lv L., Jiang X.R., Wu H., Chen G.Q. CRISPRi engineering E. coli for morphology diversification. Metab Eng. 2016;38:358–369. doi: 10.1016/j.ymben.2016.09.001. [ DOI ] [ PubMed ] [ Google Scholar ] 61. Zheng Y., Mo Y.X., Yuan Y.B., Su T.Y., Qi Q. A rapid and efficient strategy for combinatorial repression of multiple genes in Escherichia coli. Microb Cell Fact. 2025;24:74. doi: 10.1186/s12934-025-02697-x. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 62. Yin G.B., Zhang L.Y., Zhou D., Zhou S.Y., Xu R.R., Du G.C., Chen J., Kang Z. Design of orthogonal and portable RNA devices for post-transcriptional regulation by reverse engineering type I toxin-antitoxin systems. Nucleic Acids Res. 2025;53 doi: 10.1093/nar/gkaf1395. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 63. Zheng L.J., Xu J.J., Huang S.H., Wen X.Y., Shi L., Zeng G.J., Zheng H.D., Fan L.H., Guo Q. Improvement of D-allulose synthesis from sucrose through dual PTS-dependent transport in engineered Escherichia coli. J Agric Food Chem. 2025;73:18889–18898. doi: 10.1021/acs.jafc.5c04449. [ DOI ] [ PubMed ] [ Google Scholar ] 64. Qiao J., Tan X., Ren H.Y., Wu Z., Hu X.Q., Wang X.Y. Construction of an Escherichia coli strain lacking fimbriae by deleting 64 genes and its application for efficient production of poly(3-hydroxybutyrate) and L-Threonine. Appl Environ Microbiol. 2021;87 doi: 10.1128/AEM.00381-21. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 65. Liu P., Liu J.S., Zhu W.Y., Zhang B., Liu Z.Q., Zheng Y.G. O-Succinyl-L-homoserine overproduction with enhancement of the precursor succinyl-CoA supply by engineered Escherichia coli. J Biotechnol. 2021;325:164–172. doi: 10.1016/j.jbiotec.2020.11.002. [ DOI ] [ PubMed ] [ Google Scholar ] 66. Guo L.L., Cheng Y., Tan X.X., Zhang H.X., Yu H.R., Tian M.Y., Liu M.Y., Sun W.K., Gao K.Y., Jiang T.Y., Kang Q.J., Meng W.S., Liu Y.D., Kang Z.Q., Lü C.J., Gao C., Xu P., Ma C.Q. Efficient trans-aconitic acid production using systematically metabolic engineered Escherichia coli. Synth Syst Biotechnol. 2026;11:110–116. doi: 10.1016/j.synbio.2025.09.009. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 67. Chen L., Chen M.L., Ma C.W., Zeng A.P. Discovery of feed-forward regulation in L-tryptophan biosynthesis and its use in metabolic engineering of E. coli for efficient tryptophan bioproduction. Metab Eng. 2018;47:434–444. doi: 10.1016/j.ymben.2018.05.001. [ DOI ] [ PubMed ] [ Google Scholar ] 68. Gong X.Y., Teng Y.X., Zhang J.L., Gan Q., Song M., Alaraj A., Kner P., Yan Y.J. Architecting a transcriptional repressor-based genetic inverter for tryptophan derived pathway regulation in Escherichia coli. Metab Eng. 2024;86:66–77. doi: 10.1016/j.ymben.2024.09.008. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 69. Schmid J.W., Mauch K., Reuss M., Gilles E.D., Kremling A. Metabolic design based on a coupled gene expression—metabolic network model of tryptophan production in Escherichia coli. Metab Eng. 2004;6:364–377. doi: 10.1016/j.ymben.2004.06.003. [ 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.docx (5.9MB, docx) Articles from Synthetic and Systems Biotechnology are provided here courtesy of KeAi Publishing ACTIONS View on publisher site PDF (9.8 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top

Related documents

• Relapse prediction and individualized treatment-effect modeling in relapsing multiple sclerosis: a systematic review.#1016102