Improving xylitol production by semi-rational engineering of xylitol dehydrogenase and optimizing cofactor regeneration in Gluconobacter oxydans

Improving xylitol production by semi-rational engineering of xylitol dehydrogenase and optimizing cofactor regeneration in Gluconobacter oxydans | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Improving xylitol production by semi-rational engineering of xylitol dehydrogenase and optimizing cofactor regeneration in Gluconobacter oxydans Junhao Su, Jincheng Guo, Lu Liu, Dong Liu, Hailing Zhang This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8820590/v1 This work is licensed under a CC BY 4.0 License Status: Under Revision Version 1 posted 12 You are reading this latest preprint version Abstract Background Xylitol, a widely used sugar substitute in food and medicine, can be produced through microbial bioconversion using glucose as the primary substrate. In this process, a critical factor limiting xylitol production is the relatively low activity of xylitol dehydrogenase (XDH) during the biotransformation of D-arabitol to xylitol by resting cells of Gluconobacter oxydans . Results To improve the catalytic performance, Go XDH was engineered by site-saturation mutagenesis combined with a high-throughput screening method, and a variant Go XDH M3 (S77C/S106N/A110N) with high activity was obtained, showing a 2.49-fold increase in catalytic activity. The structural analysis revealed that the ‌S77C/S16N/A110N‌ mutations enhanced proton and electron transfer rates while stabilizing the hydrophilic substrate-binding pocket and the tetrameric structure. Additionally, by optimizing the coenzyme regeneration system and enhancing the oxygen transfer efficiency, we developed an efficient biotransformation of D-arabitol to xylitol in G. oxydans . Using resting cells of G. oxydans /XDH M3 -GDH-VHb, a xylitol titer of 29.02 g/L were achieved from 40 g/L D-arabitol within 30 h. Conclusion The findings suggest that boosting XDH activity through semi-rational engineering markedly improves xylitol productivity in G. oxydans . Xylitol Gluconobacter oxydans Xylitol dehydrogenase Site-saturation mutagenesis Coenzyme regeneration system Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Xylitol, a sugar alcohol found in nature, serves extensively as a sugar substitute in food and pharmaceutical applications ‌[1]. It is considered suitable for diabetics due to its comparable sweetness to sucrose and its insulin-independent metabolism [2]. In recent years, xylitol has been in demand globally at 161.5 million tons per annum, growing at 6% per annum, with promising applications in the food and pharmaceutical sectors [3]. There are two primary routes for xylitol production: chemical high‑pressure hydrogenation of xylose using metal catalysts, and microbial biotransformation from xylose [ 4 ]. Both methods depend on xylose derived from hemicellulose-xylan hydrolysates, but they face significant challenges like large acid/base requirements, serious environmental impact, and complicated processes [ 3 ]. In contrast, a ‌glucose-based bioconversion process‌ offers a cleaner and more cost-effective alternative for xylitol production. The process comprises two biotransformation steps: first, osmophilic yeasts ferment glucose to D-arabitol; second, xylitol is catalyzed from D-arabitol using resting cells of Gluconobacter oxydans [5]. G. oxydans is a gram‑negative obligate aerobe noted for rapid, incomplete oxidation of many sugars and alcohols via membrane‑bound dehydrogenases [6]. In this process, D-arabitol is converted to D-xylulose by the membrane-bound PQQ-dependent D-arabitol dehydrogenase (mAraDH), after which cytoplasmic NADH-dependent xylitol dehydrogenase (XDH) reduces D-xylulose to xylitol. The membrane-bound mAraDH can fully convert D‑arabitol to D‑xylulose, whereas only about 25% of D‑xylulose is subsequently reduced to xylitol. [7]. The poor catalytic efficiency and coenzyme dependence of XDH make it the key rate-limiting enzyme in xylitol biosynthesis [8]. Recently, several attempts have been made to promote xylitol production. Exogenous NADH supplementation is an alternative strategy to overcome coenzyme imbalance in XDH catalysis and achieve high xylitol yields. However, this approach remains economically unviable for industrial-scale applications because of the high cost of NADH [ 9 ]. Consequently, developing an efficient coenzyme regeneration system in recombinant cell factories has emerged as a critical approach to achieve cofactor self-sufficiency and improve xylitol production from D-arabitol in whole-cell biocatalysis. This process can be accomplished by introducing key genes encoding glucose dehydrogenase (GDH), alcohol dehydrogenase (ADH), or glucose-6-phosphate dehydrogenase (G6PDH) into Escherichia coli or G. oxydans , with ethanol or glucose serving as co-substrates to establish a robust NAD + /NADH redox cycling system [ 2 , 8 – 10 ]. However, the relatively low activity of the XDH enzyme remains a critical factor limiting xylitol synthesis. Therefore, applying directed evolution techniques to improve XDH catalytic efficiency in G. oxydans represents a promising strategy for high-yield xylitol production. Xylitol dehydrogenase (XDH; EC 1.1.1.9) is an oxidoreductase that catalyzes the reversible interconversion of xylitol and D-xylulose with NADH or NAD + as the respective cofactors [ 6 ]. The enzyme exhibits optimal reducing activity at 30°C and pH 6.0, whereas its oxidation activity peaks at 35°C and pH 9.0 [ 5 ]. Current research on XDH modification remains limited, with most studies concentrating on the regulation of its oxidative activity. To improve D-xylulose yield, XDH has been engineered by site-directed mutagenesis to modify its thermostability and coenzyme specificity. For example, the double mutant (D38S/M39R) and the triple mutant (D206A/I207R/F208S) altered the cofactor preference of XDH from NAD⁺ to NADP + [ 11 , 12 ] ‌ . In the literature of Watanabe [ 13 ], substituting S96/S99/Y102 with cysteine residues in Ps XDH introduced a structural zinc atom, thereby significantly enhancing its catalytic efficiency and thermostability (Tm increased by 4.5°C)‌. To date, no studies have employed protein engineering strategies to rationally modify XDH to enhance its reducing activity. A critical limitation lies in the fact that traditional approaches, such as site-directed mutagenesis‌ and ‌random mutagenesis‌, are labor-intensive and inefficient for rapidly screening high-activity XDH mutants. In this study, a ‌semi-rational modification‌ combining ‌structural analysis‌ with ‌computational hotspot prediction‌ (Hotspot Wizard 3.1) was employed to enhance XDH catalytic efficiency. And a high‑throughput screening method was established to accelerate and streamline the screening process. Finally, the highly active XDH was introduced into G. oxydans to construct an efficient one-step biotransformation system for converting D-arabitol to xylitol. 2. Methods and materials 2.1 Strains, plasmids, media, and culture conditions Strains and plasmids used in this study are listed in Table S1 . The recombinant pET-28b- xdh , pET-28b- gdh , and pET-28b- vhb were artificially synthesized from Sangon Biotech (Shanghai) Co., Ltd [ 14 , 15 ]. The 786 bp DNA sequence of xdh was optimized by adjusting the average GC content from 61.7% to 56.5% and increasing the CAI amount from 0.67 to 0.99, and a C-terminal His-tag was introduced to facilitate recombinant enzyme purification. These recombinant plasmids were then transformed into E. coli BL21 (DE3). For triparental mating, E. coli DH5α served as the plasmid construction and maintenance host, E. coli HB101 carrying helper plasmid pRK2013 was used for biotransformation, and the broad‑host‑range vector pBBR1MCS‑5 was employed to express native and heterologous genes in G. oxydans [ 16 ]. The strain G. oxydans 621H was purchased from BeNa Culture Collection Co., Ltd (China) and used for xylitol production [ 6 ]. E. coli was cultured in Luria-Bertani (LB) medium containing 5.0 g/L yeast extract, 10.0 g/L tryptone, and 10.0 g/L NaCl. After autoclaving at 121°C for 20 min, E. coli strains were cultivated at 37°C and 150 rpm in LB medium containing 50 mg/L kanamycin or 50 mg/L gentamicin when plasmid selection was required. G. oxydans seed medium consisted of 55.0 g/L D-sorbitol, 20.0 g/L yeast extract, 5.0 g/L KH 2 PO 4 , and 5.0 g/L K 2 HPO 4 (pH 6.5), while the fermentation medium contained 80.0 g/L D-sorbitol, 20.0 g/L yeast extract, 5.0 g/L KH 2 PO 4 , and 5.0 g/L K 2 HPO 4 (pH 6.5). Cells from a 48-h agar slant were transferred into 250 mL flasks containing 50 mL seed medium and incubated on a rotary shaker at 150 rpm and 30°C for 36 h. A 0.5 mL aliquot of this seed culture was inoculated into a 500 mL Erlenmeyer flask containing 100 mL fermentation medium, and flasks were incubated on an orbital shaker at 150 rpm and 30°C for 24 h. 2.2 Construction of site-saturation and site-directed mutagenesis libraries The primers for introducing NDT codons in site-saturation mutagenesis (SSM) are listed in Table S2. pET-28b- xdh served as the template for PCR. The SSM PCR mix contained 25 µL 2×Phanta Max Buffer, 1 µL dNTP mix, 1 µL DNA polymerase, 1 µL plasmid template (40 ng/µL), 1 µL each of forward and reverse primers (10 µM), and 20 µL deionized water. The cycling program was: 95°C for 3 min (pre-denaturation), 35 cycles of 95°C for 15 s, 56°C for 15 s, 72°C for 3 min, and a final extension at 72°C for 10 min. PCR products were digested with Dpn I at 37°C for 2 h and then transformed into E. coli BL21(DE3). Single colonies from agar plate were inoculated into a 96-deep well plate for high-throughput screening as described in section 2.4 . Positive clones were verified by DNA sequencing and stored at -80°C in sterile 30% (v/v) glycerol. Primers for site‑directed mutagenesis (SM) are listed in Table S3. Using pET‑28b‑ xdh as template, the PCR mix and cycling conditions matched those used for SSM. PCR products were digested with Dpn I at 37°C for 2 h, transformed into competent cells of E. coli , and plated for 12–16 h. Single colonies were grown in LB, and XDH activity was assayed as described in section 2.3 . 2.3 Activity assaya Cells from 40 mL broth were harvested by centrifugation (9000×g, 10 min) and resuspended in 0.8% saline for XDH activity measurement. Separately, cells were resuspended in 10 mL PBS (100 mM, pH 6.0) and disrupted by ultrasonication (40 W, 1 s/1 s) for 20 min. After centrifugation to remove debris, the supernatant was used to assay XDH activity. Screening for high-vigor mutants was performed by detecting NADH [ 17 ]. The reaction system (200 µL) consisting of PBS (0.1 mol/L, pH 6.0), 2 mM D-xylulose, and a certain amount of XDH enzyme was added. After adding 1 mmol/L NADH and mixing, absorbance changes at 340 nm were measured over 5 min. The enzyme activity unit (U) was defined as the amount of enzyme consumed to oxidize 1 µmol NADH per minute at 37°C. The enzyme activity of XDH (U/mL) = D×V×(△A/△t)/Ɛ×L×v. D: dilution multiple; V: total reaction volume; △A/△t: change of absorbance in unit time; Ɛ=6.22×10 3 L·mol − 1 ·cm − 1 ; L: optical path; v: volume of the enzyme solution. 2.4 Establishment of a high-throughput screening method for mutants 2.4.1 Preparation of cysteine-carbazole reagent The cysteine-carbazole reagent consists of 15 g/L cysteine hydrochloride solution (0.375 g cysteine hydrochloride dissolved in 25 mL of distilled water), 1.2 g/L carbazole solution (30.0 mg carbazole dissolved in 25 mL alcohol, stored away from light in a brown bottle), and 70% H 2 SO 4 . The above solutions were prepared respectively and mixed according to the ratio of 70% H 2 SO 4 : cysteine hydrochloride solution: carbazole solution = 30:1:1 when used. 2.4.2 High-throughput screening for mutants Figure S1 illustrates the high‑throughput screening procedure. Single colonies from agar plates were inoculated into individual wells of 96‑deep‑well plate X (1.0 mL LB with 50 µg/mL kanamycin per well) and incubated at 37°C for 6 h. A 100 µL aliquot from each well of plate X was transferred to the corresponding well of plate Y containing 100 µL sterile 30% (w/v) glycerol and stored at -80°C for preservation. To the remaining 900 µL culture, 100 µL fresh LB containing 50 µg/mL kanamycin and 1 mM IPTG was added, and cultured at 28°C for 16 h. Cells were harvested by centrifugation (3000×g, 30 min) and resuspended in 0.8% saline. The cells were resuspended with 200 µL reaction mixture (100 mM PBS, 100 mM D-xylulose, pH 7.0). The plates were stirred at 30 o C, 180 rpm for 1 h and then centrifugated at 3000×g for 30 min. A 40 µL aliquot of each diluted supernatant was added to a 96‑well microtiter plate containing 160 µL cysteine-carbazole reagent. The 96-well microtiter plate was incubated at 60°C for 10 min, and then colorimetric measurements were performed on a microplate reader at 540 nm. The content of D‑xylulose was calculated using the standard curve, and then the activity of XDH was calculated. One unit of the catalytic activity of XDH was defined as the amount of enzyme that consumed 1 µmoL D-xylulose per minute at 30 o C. The standard curve between the D-xylulose (mg/L) and OD 540 was y = 0.0063x + 0.1542, R 2 > 0.99. 2.4.3 Secondary screening The corresponding duplicate from plate Y was propagated in shake flasks for re-screening. This was followed by a rescreening process using the screening method for detecting NADH, as described in section 2.3 . 2.5 Protein purification and determination of kinetic parameters of wild-type and mutant XDH After spinning at 9000 × g for 10 min to pellet debris, the supernatant was passed through a Ni 2+ ‑NTA column; purified His‑tagged proteins were evaluated by SDS‑PAGE. The nickel ion column was first equilibrated using 20 mL of Binding Buffer (containing 50 mM NaCl, 20 mM PBS at pH 8.0). Then, 10 mL of crude protein was sampled, and a 20 mM imidazole solution (containing 50 mM NaCl, 20 mM imidazole, 20 mM PBS at pH 8.0) was used to elute the miscellaneous protein. The protein elution volume was 20 mL. Finally, the target protein was eluted using 5 mL of 300 mM imidazole solution (containing 50 mM NaCl, 300 mM imidazole, 20 mM PBS with pH 8.0). At the end of elution, the nickel ion column was washed with Binding Buffer and preserved with a 20% ethanol solution. The collected pure proteins were subjected to ultrafiltration membrane filtration utilizing 0.1 M PBS three times to remove excess imidazole and stored in 10% glycerol at -20°C. After centrifuging at 9000×g for 10 min at 4℃, the supernatant was collected as pure protein. The precipitate was resuspended with 10 mL of 50 mM PBS (pH 8.0). A specific amount of enzyme solution was added to the loading buffer and boiled for 15 min before sampling. Activity dependence on temperature was measured at pH 6.0 over a 20-55 o C range. The determination of the optimal pH of XDH was carried out between 5.0 and 10.0 at 30 o C, in which the buffer solution was replaced with citric acid-sodium citrate buffer solutions, pH 5.0–6.0; PBS, pH 6.0–9.0, and Gly-NaOH buffer solutions, pH 9.0–10.0, respectively. Kinetic parameters were measured with D‑xylulose concentrations of 1–10 g/L in PBS (50 mM, pH 7.5) at 35°C. Samples were centrifuged (9000×g, 10 min) and the supernatants assayed as described above. K M and V max were obtained by nonlinear regression fitting to the Michaelis-Menten equation using OriginPro 8.0. The kinetic parameters were determined with varying concentrations of D-xylulose (1–10 g/L) in PBS (50 mM, pH 7.5) at 35 o C. Samples were centrifuged at 9000×g for 10 min and the supernatants assayed as described above. The values of Km and Vmax were determined by nonlinear regression fitting to the Michaelis-Menten equation using OriginPro 8.0. 2.6 Molecular homology modeling and docking Using SWISS‑MODEL, a three‑dimensional model of Go XDH was generated based on XDH (PDB: 1ZEM; 99.62% sequence identity). The structures of D‑xylulose and NADH were generated in Chem3D and docked into the homology model with AutoDock Vina, and docking results were visualized and analyzed in PyMOL. 2.7 DNA manipulation and plasmid construction in G. oxydans The strong promoter gHp0169 was chosen to drive expression of the target genes. The open reading frame (ORF) of xdh , including its endogenous ribosome‑binding site (RBS) from G. oxydans , was amplified by PCR using the primers listed in Table S4. These genes were cloned into the previously constructed plasmid pBBR- gHp0169 , resulting in plasmids pBBR- gHp0169-xdh , pBBR- gHp0169-gdh , and pBBR- gHp0169-vhb [ 18 ]. And the genes of gdh and vhb were cloned from pET-28b- gdh and pET-28b- vhb and inserted into plasmid pBBR- gHp0169 , resulting in plasmids pBBR- gHp0169-gdh and pBBR- gHp0169-vhb . According to the SM method, mutations S77C/S106N/A110N were introduced into pBBR- gHp0169-xdh , resulting in pBBR- gHp0169-xdh M3 (Fig. S2). Then the PCR product of gHp0169-gdh was amplified and inserted into pBBR- gHp0169-xdh M3 , resulting in pBBR- gHp0169-xdh M3 -gHp0169 - gdh (Fig. S2). Finally, the PCR product of gHp0169-vhb was cloned into pBBR- gHp0169-xdh M3 -gHp0169 - gdh , resulting in pBBR- gHp0169-xdh M3 -gHp0169 - gdh-gHp0169-vhb (Fig. S2). Plasmids were introduced into G. oxydans by triparental mating and selected on gentamicin, yielding recombinant strains G. oxydans /XDH, G. oxydans /XDH M3 , G. oxydans /XDH M3 ‑GDH, and G. oxydans /XDH M3 ‑GDH‑VHb [ 19 ]. Recombinant strains were cultured in fermentation medium containing 50 mg/L gentamicin, as described in Section 2.1 . Cells from 200 mL broth were harvested by centrifugation (9000×g, 10 min), resuspended in 0.8% saline, and used for biotransformation of xylitol from 30 g/L D‑arabitol. For G. oxydans /XDH M3 ‑GDH, 20 g/L glucose was added at 0, 6.5, 10, and 12 h in the reaction solution to promote NADH regeneration. After the reaction, debris was removed by centrifugation and the supernatant was analyzed for xylitol, D-arabitol, and D-xylulose concentrations. 2.8 Cell cultivation in a stirred bioreactor Cell cultivation was performed in a 10-L multi-bioreactor system (BLBIO-50SJ-5, China) using the fermentation medium. The bioreactor was inoculated at 10% (v/v) and run at 30 o C for 24 h with agitation set to 400, 500, 600, or 700 rpm and 1.5 vvm aeration. Cultures were centrifuged at 9000×g for 10 min; the supernatant was used to quantify substrate consumption, and the cell pellet was used for the biotransformation of D‑arabitol to xylitol. 2.9 Biotransformation of D-arabitol to xylitol using resting cells of recombinant G. oxydans Bioconversion of D-arabitol to xylitol was carried out in 500 mL shaking flasks using 100 mL reaction mixtures containing wet cells (100 g/L) and 40 g/L D-arabitol; 30 g/L CaCO3 was added to maintain pH. Reactions were performed at 30°C and 180 rpm. And 20 g/L glucose was added at 10 h to facilitate NADH regeneration. After completion, samples were centrifuged and the supernatants analyzed by HPLC for product concentrations. 2.10 Analytical methods Biomass was estimated by measuring optical density at 600 nm (OD 600 ) and converted to dry cell weight (DCW) using the calibration DCW (g/L) = 0.4627x-0.2685, R 2 > 0.99. Concentrations of D‑sorbitol, D‑arabitol, D‑xylulose, and xylitol were measured by HPLC with a differential refractive index detector (Waters, USA) using an Aminex HPX‑87H column (300×7.8 mm; Bio‑Rad) at 55°C, a flow rate of 0.5 mL/min, and 5 mM H 2 SO 4 as the mobile phase [ 20 ]. Experiments were performed in triplicate, and mean values ± SD from three independent batches are reported. 3. Results and Discussion 3.1 Establishment and optimization of the colorimetric assay A rapid assay for XDH activity was developed using the cysteine-carbazole chromogenic reaction with D‑xylulose. The mixture of cysteine-carbazole can react with D-xylulose under acidic conditions, showing a purple color, and the color becomes darker with the increase of D-xylulose concentration. To develop an efficient assay for monitoring D‑xylulose levels during the reduction process, the reaction conditions of cysteine-carbazole and D-xylulose were optimized. Different D‑xylulose concentrations (10–50 mg/L) were loaded into columns 1–5, and 120–200 µL of cysteine-carbazole reagent was added to wells in rows A-E, respectively (Fig. 1 a). The plate was incubated at 60°C for 10 min, producing distinct color changes in the wells. In row C, when using 160 µL cysteine-carbazole reagent as the initial developing agent volume, the color change was easier to distinguish. The absorbance spectrum of the colorimetric assay showed a characteristic peak at 540 nm, while the color reaction of cysteine-carbazole reagent with other substrates such as xylitol did not show a characteristic absorption peak in 300–700 nm wavelength range (Fig. 1 b). In addition, this method is reliable and has a good linear correlation with the XDH activity when the concentration of D-xylulose is in the range of 10–100 mg/L, making it suitable for the qualitative analysis and quantification of the products. Therefore, in the following experiments, 160 µL of cysteine-carbazole reagent and 540 nm was selected as the chromogenic volume and working wavelength. Based on the above results, we established a high-throughput screening method to identify strains with superior XDH activity. Enzymatic reactions were performed in PBS (100 mM, pH 6.0) containing 10 mM D‑xylulose and incubated at 35°C for 1 h. A 40 µL aliquot of the diluted reaction mixture was added to 160 µL cysteine-carbazole reagent, and absorbance was measured at 540 nm for the colorimetric assay. This enzyme activity detection method shortened the reaction time, avoided complicated operations, and improved the screening efficiency. 3.2 Computational identification of beneficial residue positions Semi-rational design based on structural analysis is a useful way to enhance the catalytic activity of Go XDH (xylitol dehydrogenase from G. oxydans 621H). However, the large sequence libraries make the screening process laborious and time-consuming. HotSpot Wizard 3.1 is a powerful online tool that integrates structural, functional, and evolutionary information from various bioinformatics resources to identify residue positions critical for enzymatic activity [ 21 ]. By restricting mutagenesis to a limited set of “hot spot” positions, a “small but smart” comprehensive mutant library can be constructed, which can significantly accelerate directed evolution workflows [ 22 , 23 ]. Initially, the Go XDH model was constructed using the crystal structure of XDH (PDB ID: 1zem.1. A), which shares 99.62% sequence identity‌. As a member of the short-chain dehydrogenase/reductase (SDR) superfamily, there are four conserved functional domains in Go XDH: the N -terminal glycine-rich coenzyme-binding region (N90-N91-A92-G93), the catalytic tetrad (N116-S144-Y157-K161), the central domain (T13-G14-A15-G16-G17-N18-I19-G20), and the C -terminal substrate-binding region (P187-G188) [12]. The Go XDH enzyme assembles as a tetramer, each subunit containing a Rossmann fold: a central 7‑strand β‑sheet flanked by short helices αB, αC and αG on one side and longer helices αD, αE and αF on the other (Fig. S3). Additionally, two short helices (αFG1 and αFG2) extend outward above the Rossmann fold. Subsequently, the catalytic pocket hotspots identified through molecular docking analysis, along with consensus residues predicted by the HotSpot Wizard, were designated as target sites for mutagenesis. A mutant library was constructed with 13 predicted beneficial sites: S77, F87, D103, S106, A110, V117, T118, K149, P152, A156, Y158, and G162. Finally, site‑saturation mutagenesis was conducted on these target residues, and positive variants were identified with the previously established high‑throughput assay. 3.3 Screening of variants with improved activity After several rounds of mutagenesis, four variants, S77C, S106N, and A110N, exhibiting improved activity were obtained. Compared with the wild-type Go XDH, they showed 29.8%, 42.7%, and 48.6% improvement in activity toward D-xylulose (Fig. 2 ). The structural analysis revealed that the S106N and A110N mutations located near N116 in the catalytic tetrad (N116-S144-Y157-K161) and on the helix αE (a critical interaction surface for the main subunits) led to a substantial enhancement of catalytic activity. To identify additional beneficial mutations, we performed site-saturation mutagenesis on non-conserved residues within the αE helix (D107, R111, L113, T114, I115, T118, H122, S128, R129, G130, T133, and Q134). After several rounds of mutagenesis, several new variants were identified. The T114H and I115R mutants exhibited 18.6% and 32.3% increases in their activity compared to the wild-type Go XDH. In the catalytic tetrad of Go XDH, Y157 acts as a catalytic base, S144 stabilizes a negative charge on the catalytic intermediate, and K161 interacts with the nicotinamide ribose, thus lowering the pKa of the tyrosine hydroxyl (Fig. 3 ) [ 12 ]. The function of N116 is to bridge the substrate‑binding loop and the active site using conserved water molecules [ 25 ]. These water molecules create a small hydrophilic cavity linked to other conserved residues in the fold. This pocket acts as a proton-relay system, bridging K161 to the bulk solvent and stabilizing the positively charged K161 [ 25 ]. All four residues (S106, A110, T114, and I115) on the αE helix were converted to hydrophilic amino acids (N, H, or R), indicating their likely contributes to hydrophilic pocket stabilization via improved solvation and residue-residue interactions. Furthermore, the S77C mutation on the monomer surface may have improved the protein's hydrophobicity, which is crucial for structural stabilization. Combining beneficial mutations is commonly employed to enhance enzyme performance, as mutations can act synergistically or cooperatively [ 26 ]. Therefore, combinatorial mutagenesis was conducted to generate multiple base mutations. The date revealed that the activities of the double mutant S77C/S106N and triple mutant S77C/S106N/A110N were 1.92-fold and 2.49-fold higher than that of Go XDH, respectively (Table 1 ). However, further combinations with T114H or I115R did not lead to a notable increase in activity. Accordingly, further enzymatic studies concentrated on Go XDH M3 (S77C/S106N/A110N) due to its markedly increased activity. Table 1 Enzyme activities of Go XDH and its variants towards D-xylulose. Xylitol dehydrogenase Mutational sites Enzyme activity (U/mL) Relative activity (%) Go XDH - 6.30 ± 0.13 100 Go XDH M2 S77C/S106N 12.12 ± 0.21 192 Go XDH M3 S77C/S106N/A110N 15.70 ± 0.18 249 Go XDH M4 S77C/ /S106N/A110N/T114H 10.52 ± 0.15 167 Go XDH M4’ S77C/S106N/A110N/ I115R 11.63 ± 0.12 185 3.4 Enzymatic characterization of wild-type Go XDH and Go XDH M3 To biochemically characterize the enhanced activity, both parental Go XDH and the Go XDH M3 variant were purified, and their kinetic parameters toward D‑xylulose were determined. The purified results were shown in Fig. S4. SDS-PAGE indicated a monomer size of 25–35 kDa, consistent with the calculated mass of 29.48 kDa. It could be seen that the purified enzyme contained almost no miscellaneous proteins and had a high purity, which was suitable for determining enzymatic properties. The optimal temperature of Go XDH and Go XDH M3 was approximately 35 o C. And relative activity showed a steady increase from 30 o C to a peak (100% activity), then declined sharply to its lowest level at 80°C (Fig. S5). Both Go XDH and Go XDH M3 showed higher activity at neutral pH, and the optimal pH was 7.5 (Fig. S5). These results differed from the 30 o C and pH 6.0 in the recent reports [ 5 ]. Kinetic parameters for Go XDH and Go XDH M3 were determined using purified cell extracts from the recombinant strains. Michaelis-Menten curves were fitted to the experimental data for both enzymes, showing a close fit. The extracts containing Go XDH converted D-xylulose at a maximal rate ( V max ) of 7.03 U/mg total protein, and the extracts containing Go XDH M3 showed a V max of 29.25 U/mg total protein. The Go XDH exhibited a K M for D-xylulose of 0.67 mM, and Go XDH M3 exhibited a K M of 0.52 mM. Comparative analysis showed that Go XDH M3 exhibits a V max 4.2-fold higher and a K M for D‑xylulose 23.2% lower than Go XDH. 3.5 Mechanism elucidation of activity enhancement To elucidate the enhanced catalytic mechanism, D‑xylulose was docked into models of Go XDH and Go XDH M3 . Comparison of docking poses revealed that the distance between the hydrogen of Y157 and the substrate carbonyl oxygen shortened from 3.4 Å to 2.3 Å in Go XDH M3 , while the distance between the C4 hydrogen of NADH’s nicotinamide ring and the substrate carbonyl carbon slightly extended from 3.7 Å to 4.0 Å (Fig. 4 a, b). The shortened Y157-substrate distance likely facilitates more efficient proton/electron transfer and increases substrate affinity in Go XDH M3 , accounting for its lower K M and improved catalytic efficiency. Hydrogen bond analysis reveals distinct differences in substrate binding between Go XDH and Go XDH M3 . In the wild-type Go XDH, three hydrogen bonds formed between the substrate and adjacent residues (Q95: 2.5 Å; S144: 2.2 Å; A146: 2.5 Å, Fig. 4 c). In contrast, Go XDH M3 formed five hydrogen bonds with nearby residues (S144: 1.8 Å, 2.2 Å; A146: 2.5 Å; Y157: 2.3 Å; Q199: 2.5 Å, Fig. 4 d). The expanded hydrogen bond network in Go XDH M3 contributed to more rigid substrate positioning, thus increasing catalytic performance. The reduction of D-xylulose requires cooperative participation of the catalytic tetrad (N116-S144-Y157-K161) and NADH as cofactor. First, Y157 donates a proton to the substrate carbonyl, and then a hydride is transferred to C2 of D‑xylulose (Fig. 3 ) [ 13 ]. A proton relay composed of the nicotinamide ribose 2’‑hydroxyl, K161, and a water molecule associated with N116’s backbone carbonyl facilitates the reaction. Formation of a hydrogen bond between S144 and the substrate carbonyl can stabilize the catalytic intermediate [ 25 ]. In Go XDH M3 , the additional hydrogen bond at S144 contributes to maintaining the stability of the catalytic pocket. Meanwhile, a new hydrogen bond was found between the substrate and residue Q199. Analysis of the substrate-binding pocket revealed that Q199, situated adjacent to the substrate channel inlet, pulls the substrate closer to Y157 and the pocket entrance, thereby promoting catalysis efficiency. Compared to the wild-type Go XDH, the Go XDH M3 variant exhibited significant structural modifications at the dimer interface (Fig. 5 ). Go XDH M3 introduced a new hydrogen bond between N110 and H122 on the αE helix, as well as an additional hydrogen bond between N106 and T118. These interactions collectively enhanced the structural rigidity of the dimer. In contrast, only a single hydrogen bond between S106 with K125 was found at the dimer interface of the wild-type Go XDH. These observations suggest that the introduction of new hydrogen bonds at the dimer interface, particularly in the αE helix region, contributes to stabilizing Go XDH structure. 3.6 Construction of the NADH coenzyme regeneration system in the production of xylitol by recombinant G. oxydans The beneficial mutations S77C/S106N/A110N were cloned into pBBR‑ gHp0169 ‑ xdh and introduced into G. oxydans to generate G. oxydans /XDH M3 . Flask biotransformation showed G. oxydans /XDH M3 produced 8.95 g/L xylitol after 36 h, a 36.9% increase over G. oxydans /XDH (6.54 g/L) (Table 2 ), demonstrating superior catalytic efficiency and potential application in whole-cell production of xylitol. Because Go XDH consumes large amounts of NADH to reduce D‑xylulose to xylitol, effective cofactor regeneration is required to achieve xylitol production, in which glucose dehydrogenase (GDH) is commonly used for this purpose. Therefore, GDH was co‑expressed with Go XDH M3 in G. oxydans , and biotransformation was carried out using resting recombinant cells with 20 g/L glucose as co‑substrate. As shown in Table 2 , the recombinant G. oxydans /XDH M3 -GDH achieved a significantly higher xylitol yield (11.92 g/L) compared to G. oxydans /XDH M3 . Table 2 Effects of mutation S77C/S106N/A110N and GDH on the yield of xylitol in G. oxydans . Resting cells Conversion time (h) Initial substrate concentration (g/L) Xylitol titer (g/L) Xylitol yield (%) G. oxydans /XDH 36 30 6.54 ± 0.34 21.80 G. oxydans /XDH M3 36 30 8.92 ± 0.36 29.73 G. oxydans /XDH M3 -GDH 36 30 11.92 ± 0.25 39.73 The reaction solution of 10 mL was composed of wet cells (collected from 200 mL of the fermentation broth) and 30 g/L D-arabitol, in which 30 g/L CaCO 3 was added to maintain pH. The biotransformation was performed at 30 o C and 180 rpm. For the bioconversion process of G. oxydans /XDH M3 -GDH, 20 g/L of glucose was added to the reaction solution to enable NADH regeneration. Although GDH was identified as an effective coenzyme regeneration system for xylitol production, the optimal timing for glucose addition remained unclear. For this purpose, the effects of adding glucose at different times (0, 6.5, 10, and 12 h) on the xylitol production were studied. As depicted in Fig. 6 , the yield of xylitol was significantly higher when glucose was added at 6.5 h (13.14 g/L), 10 h (15.48 g/L), or 12 h (15.36 g/L) than when added at 0 h (11.92 g/L). Results showed that peak xylitol production was achieved when glucose supplementation began at 10 h. At this time point, approximately 25 g/L of the initial D-arabitol had been consumed, accompanied by the accumulation of 23.71 g/L D-xylulose. The results indicated that when D-xylulose reached a critical concentration, the demand for NADH increased; thus, timely glucose supplementation could enhance the utilization efficiency of NADH. In contrast, early-stage glucose supplementation might enable competitive membrane-bound glucose dehydrogenase (mGDH) in G. oxydans , thereby diminishing overall glucose consumption [ 27 ]. By strategically supplementing glucose at 10 h, G. oxydans /XDH M3 -GDH achieved more efficient coenzyme regeneration, which enhanced xylitol production from D-xylulose. 3.7 Cell cultivation in a stirred bioreactor and efficient conversion of D-arabinitol to xylitol using resting cells of recombinant G. oxydans /XDH M3 -GDH-VHb As an obligate aerobic bacterium, G. oxydans requires a high oxygen supply during cell proliferation to support efficient respiratory metabolism. Vitreoscilla hemoglobin (VHb) is an oxygen-binding protein that facilitates bacterial respiration by directly channeling oxygen to terminal oxidases [ 28 ]. To enhance intracellular oxygen delivery and improve xylitol biosynthesis, the gene encoding VHb was introduced into the plasmid pBBR- gHp0169-xdh-gHp0169-gdh , resulting in recombinant G. oxydans /XDH M3 -GDH-VHb. To determine the optimal growth conditions, the recombinant G. oxydans /XDH M3 -GDH-VHb strain was cultured under varying stirring speeds (400–700 rpm). As depicted in Fig. S6, the D-sorbitol consumption rate exhibited a marked increase with rising stirring speed, and the maximal cell density at 400, 500, 600, and 700 rpm reached 2.92, 3.55, 3.39, and 3.09 gDCW/L, respectively, at 20 h. To evaluate the catalytic performance of these strains, the biotransformation of D-arabitol to xylitol was performed by resting cells. As shown in Table 3 , in the D4 experiment, the G. oxydans /XDH M3 ‑GDH‑VHb strain reached a maximum xylitol concentration of 29.02 g/L at 30 h. These results indicate a high oxygen demand during G. oxydans cell proliferation. Although higher biomass accumulation was observed at 500 and 600 rpm, enzyme production was higher at 700 rpm, resulting in enhanced catalytic activity. Table 3 Comparison of xylitol production catalyzed by different resting cells of G. oxydans /XDH M3 -GDH-VHb. Experiments Conversion time (h) Xylitol titer (g/L) Xylitol yield (%) D1 30 24.18 ± 0.64 60.45 D2 30 27.08 ± 0.54 67.70 D3 30 26.87 ± 0.57 67.18 D4 30 29.02 ± 0.48 72.55 Experiments D1, D2, D3, and D4 represent the resting cells of G. oxydans /XDH M3 -GDH-VHb cultured at stirring speeds of 400, 500, 600, and 700 rpm, respectively. The reaction solution of 100 mL was composed of wet cells (100 g/L) and 40 g/L D-arabitol, in which 30 g/L CaCO 3 was added to maintain pH. The biotransformation was performed at 30 o C and 180 rpm. Furthermore, at 10 h, 20 g/L of glucose was added to the reaction solution to facilitate NADH regeneration during the bioconversion process. Recent advancements in microbial xylitol production have increasingly emphasized genetic engineering strategies. Among these approaches, the introduction of XDH into E. coli has demonstrated significant potential for achieving high-titer xylitol production. The mixed culture of resting G. thailandicus cells with BL21- xdh and BL21- adh produced 34.34 g/L xylitol after 48 h of bioconversion, the highest xylitol titer reported to date [ 11 ]. However, this process is accomplished with the participation of two or even three strains. Therefore, constructing high-efficiency engineered G. oxydans capable of one-step conversion of D-arabitol to xylitol may be more suitable for production. Zhou [ 2 ] reported an engineered strain, G. oxydans PXPG, co-expressing XDH and GDH, which produced 12.23 g/L xylitol from 30 g/L D-arabitol in 50 h. In this study, we engineered G. oxydans 621H by co-expressing high-activity Go XDH M3 , GDH, and VHb, achieving 29.02 g/L xylitol in shake flasks within 30 h, showing superior production efficiency. In conclusion, this study provides an attractive and competitive candidate for the one-step production of xylitol in G. oxydans . 4. Conclusion In this study, a semi-rational engineering approach was developed to enhance xylitol production by modifying Go XDH through structural analysis and computational hotspot prediction (Hotspot Wizard 3.1). A cysteine-carbazole based high-throughput assay for D-xylulose was established to enable rapid screening of XDH variants with enhanced activity. This strategy yielded a variant Go XDH M3 (S77C/S106N/A110N), which exhibited a 2.49-fold increase in catalytic activity. Structural analysis revealed mutation S77C/S106N/A110N significantly shortened the gap between the hydrogen on Y157 and the carbonyl oxygen atom of D‑xylulose, accelerating the transfer rates of protons and electrons. Furthermore, new hydrogen bonds formed between the substrate and proximal residues, as well as at the dimer interface of the αE helix, contributed to the stabilization of both substrate binding and the tetramer structure. Finally, resting cells of engineered G. oxydans /XDH M3 ‑GDH‑VHb produced approximately 29.02 ± 0.48 g/L xylitol from 40 g/L D‑arabitol after 30 h of biotransformation. Abbreviations XDH, xylitol dehydrogenase G. oxydans, Gluconobacter oxydans Go XDH, xylitol dehydrogenase from G. oxydans 621H NADH, nicotinamide adenine dinucleotide GDH, glucose dehydrogenase VHb, Vitreoscilla hemoglobin mAraDH, D-arabitol dehydrogenase ADH, alcohol dehydrogenase G6PDH, glucose-6-phosphate dehydrogenase E. coli , Escherichia coli LB, Luria-Bertani IPTG, Isopropyl-β-D-1-thiogalactopyranoside SSM, site‑saturation mutagenesis SM, site‑directed mutagenesis SDS‑PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis PBS, phosphate buffer saline ORFs, open reading frames RBS, ribosomal binding site DCW, dry cell weight HPLC, high performance liquid chromatography SDR, short-chain dehydrogenase/reductase Declarations Ethics approval and consent to participate Not applicable. Consent for publication Not applicable. Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Competing Interests The authors declare no competing interests. Funding This work was financially supported by the Natural Science Foundation of Shandong Province (Grant number: ZR2022QB111). Authors' contributions JS performed the conceptualization, methodology, Data curation, Writing-original draft and Writing-review & editing. JG and LL helped in formal analysis, methodology, and writing review and editing. DL performed the conceptualization, supervision, project administration, and writing review and editing. HZ performed the conceptualization, supervision, project administration, and funding acquisition. Acknowledgements Not applicable. References Venkateswar Rao L, Goli JK, Gentela J, Koti S. Bioconversion of lignocellulosic biomass to xylitol: An overview. Bioresour Technol. 2016;213:299–310. Zhang J, Li S, Xu H, Zhou P, Zhang L, Ouyang P. Purification of xylitol dehydrogenase and improved production of xylitol by increasing XDH activity and NADH supply in Gluconobacter oxydans . J Agric Food Chem. 2013;61(11):2861–7. Dasgupta D, Bandhu S, Adhikari DK, Ghosh D. Challenges and prospects of xylitol production with whole cell bio-catalysis: A review. Microbiol Res. 2017;197:9–21. Ripoll M, Lerma-Escalera JA, Morones-Ramirez JR, Rios-Solis L, Betancor L. New perspectives into Gluconobacter -catalysed biotransformations. Biotechnol Adv. 2023;65:108127. Qi X, Zhang H, Magocha TA, An Y, Yun J, Yang M, Xue Y, Liang S, Sun W, Cao Z. Improved xylitol production by expressing a novel D-arabitol dehydrogenase from isolated Gluconobacter sp. JX-05 and co-biotransformation of whole cells. Bioresour Technol. 2017;235:50–8. Prust C, Hoffmeister M, Liesegang H, Wiezer A, Fricke WF, Ehrenreich A, Gottschalk G, Deppenmeier U. Complete genome sequence of the acetic acid bacterium Gluconobacter oxydans . Nat Biotechnol. 2005;23(2):195–200. Suzuki S, Sugiyama M, Mihara Y, Hashiguchi K, Yokozeki K. Novel enzymatic method for the production of xylitol from D-arabitol by Gluconobacter oxydans . Biosci Biotechnol Biochem. 2002;66(12):2614–20. Qi XH, Zhu JF, Yun JH, Lin J, Qi YL, Guo Q, Xu H. Enhanced xylitol production: Expression of xylitol dehydrogenase from Gluconobacter oxydans and mixed culture of resting cell. J Bioscience Bioengerring. 2016;122(3):257–62. Zhou P, Li S, Xu H, Feng X, Ouyang P. Construction and co-expression of plasmid encoding xylitol dehydrogenase and a cofactor regeneration enzyme for the production of xylitol from D-arabitol. Enzym Microb Technol. 2012;51(2):119–24. Zhang H, Yun J, Zabed H, Yang M, Zhang G, Qi Y, Guo Q, Qi X. Production of xylitol by expressing xylitol dehydrogenase and alcohol dehydrogenase from Gluconobacter thailandicus and co-biotransformation of whole cells. Bioresour Technol. 2018;257:223–8. Hong S, Wu J, Zhao H. Cloning, overexpression, purification, and site-directed mutagenesis of xylitol-2-dehydrogenase from Candida albicans . J Mol Catal B: Enzymatic. 2010;62(1):40–5. Ehrensberger AH, Elling RA, Wilson DK. Structure-guided engineering of xylitol dehydrogenase cosubstrate specificity. Structure. 2006;14(3):567–75. Watanabe S, Kodaki T, Makino K. Complete reversal of coenzyme specificity of xylitol dehydrogenase and increase of thermostability by the introduction of structural zinc. J Biol Chem. 2005;280(11):10340–9. Qiu S, Xu SY, Li SF, Meng KM, Cheng F, Wang YJ, Zheng YG. Fluorescence-based screening for engineered aldo‐keto reductase KmAKR with improved catalytic performance and extended substrate scope. Biotechnol J. 2021;16(9):2100130. Ouyang P, Wang H, Hajnal I, Wu Q, Guo Y, Chen GQ. Increasing oxygen availability for improving poly (3-hydroxybutyrate) production by Halomonas . Metab Eng. 2018;45:20–31. Liu D, Ke X, Hu ZC, Zheng YG. Combinational expression of D-sorbitol dehydrogenase and pyrroloquinoline quinone increases 6-( N -hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose production by Gluconobacter oxydans through cofactor manipulation. Enzym Microb Technol. 2020;141:109670. Jia DX, Liu D, Zheng YG. Heterologous expression, purification and enzymatic characterization of succinate-semialdehyde dehydrogenase GabD4. Food Ferment Industries. 2016;42(5):97–101. Ke X, Yu PH, Hu ZC, Chen L, Sun XQ, Zheng YG. Synergistic improvement of PQQ-dependent D-sorbitol dehydrogenase activity from Gluconobacter oxydans for the biosynthesis of miglitol precursor 6-(. J Biotechnol. 2019;300:55–62. -hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose . Liu D, Ke X, Hu ZC, Zheng YG. Improvement of pyrroloquinoline quinone-dependent d-sorbitol dehydrogenase activity from Gluconobacter oxydans via expression of Vitreoscilla hemoglobin and regulation of dissolved oxygen tension for the biosynthesis of 6-(. J Biosci Bioeng. 2021;131(5):518–24. -hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose . Liu D, Hu ZC, Ke X, Zheng YG. Breeding of Gluconobacter oxydans with high PQQ-dependent D-sorbitol dehydrogenase for improvement of 6-( N -hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose production. Biochemical Engineering Journal 2020, 161. Cai X, Shi X, Liu SQ, Qiang Y, Shen JD, Zhang B, Liu ZQ, Zheng YG. Hot spot-based engineering of ketopantoate hydroxymethyltransferase for the improvement of D-pantothenic acid production in Escherichia coli . J Biotechnol. 2023;364:40–9. Liu HT, Weng CY, Xu S-, Li SF, Wang YJ, Zheng YG. Directed evolution of a carbonyl reductase LsCR for the enantioselective synthesis of (1S)-2-chloro-1-(3, 4-difluorophenyl) ethanol. Bioorg Chem. 2022;127:105991. Bendl J, Stourac J, Sebestova E, Vavra O, Musil M, Brezovsky J, Damborsky J. HotSpot Wizard 2.0: automated design of site-specific mutations and smart libraries in protein engineering. Nucleic Acids Res. 2016;44(W1):W479–87. Cheng F, Zhai QY, Gao XF, Liu HT, Qiu S, Wang YJ, Zheng YG. Tuning enzymatic properties by protein engineering toward catalytic tetrad of carbonyl reductase. Biotechnol Bioeng. 2021;118(12):4643–54. Filling C, Berndt KD, Benach J, Knapp S, Prozorovski T, Nordling E, Ladenstein R, Jörnvall H, Oppermann U. Critical residues for structure and catalysis in short-chain dehydrogenases/reductases. J Biol Chem. 2002;277(28):25677–84. Zhang Q, Lu XF, Zhang Y, Tang XL, Zheng RC, Zheng YG. Development of a robust nitrilase by fragment swapping and semi-rational design for efficient biosynthesis of pregabalin precursor. Biotechnol Bioeng. 2020;117(2):318–29. Krajewski V, Simić P, Mouncey NJ, Bringer S, Sahm H, Bott M. Metabolic Engineering of Gluconobacter oxydans for improved growth rate and growth yield on glucose by elimination of gluconate formation. Appl Environ Microbiol. 2010;76(13):4369. Zhang L, Li Y, Wang Z, Xia Y, Chen W, Tang K. Recent developments and future prospects of Vitreoscilla hemoglobin application in metabolic engineering. Biotechnol Adv. 2007;25(2):123–36. Additional Declarations No competing interests reported. Supplementary Files Supplymentmetariel.docx Cite Share Download PDF Status: Under Revision Version 1 posted Editorial decision: Revision requested 11 May, 2026 Reviews received at journal 04 May, 2026 Reviews received at journal 25 Apr, 2026 Reviewers agreed at journal 25 Apr, 2026 Reviewers agreed at journal 13 Apr, 2026 Reviews received at journal 13 Mar, 2026 Reviewers agreed at journal 11 Mar, 2026 Reviewers agreed at journal 09 Mar, 2026 Reviewers invited by journal 09 Mar, 2026 Editor assigned by journal 15 Feb, 2026 Submission checks completed at journal 15 Feb, 2026 First submitted to journal 15 Feb, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8820590","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":604736346,"identity":"2bae8f2a-5f15-48e6-beac-6725ed3ca4f0","order_by":0,"name":"Junhao Su","email":"","orcid":"","institution":"Yantai University","correspondingAuthor":false,"prefix":"","firstName":"Junhao","middleName":"","lastName":"Su","suffix":""},{"id":604736347,"identity":"b5fdbe99-53db-4f0e-88e2-3ea195d746c2","order_by":1,"name":"Jincheng Guo","email":"","orcid":"","institution":"Yantai University","correspondingAuthor":false,"prefix":"","firstName":"Jincheng","middleName":"","lastName":"Guo","suffix":""},{"id":604736348,"identity":"fa1a349e-27d6-405c-b08f-3aff213ddcc6","order_by":2,"name":"Lu Liu","email":"","orcid":"","institution":"Yantai University","correspondingAuthor":false,"prefix":"","firstName":"Lu","middleName":"","lastName":"Liu","suffix":""},{"id":604736349,"identity":"eb84a594-332b-468a-854e-7a0c22e00f60","order_by":3,"name":"Dong Liu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYBACPmYGBgkgLcfG3tj4ACKWgF8LG1SLMT/P4WYDiGpCWhggWhJnzkhvkyBOCzuP4Y2PO2oZNxxIbKvm/XGYgZ89x4Dh5w58DuMxtpx55jizwYGDbbd5Eg4zSPa8MWDsPYNXi5k0b9sxNoODjRAtBjdyDJgZ2whr4TE4zNhWDNJiT6SWGgnJNsY2ZrAtEgS1sBVbzmw7YMDPw9gsOSctnUfizLOCg714tPDzH95442NbXX2b/POHH97YWMvxtydvfPATjxYoOAwmmXgYGHhAjAMENTAw1IFJxh9EKB0Fo2AUjIKRBwBR4ktNx7sTQwAAAABJRU5ErkJggg==","orcid":"","institution":"Yantai University","correspondingAuthor":true,"prefix":"","firstName":"Dong","middleName":"","lastName":"Liu","suffix":""},{"id":604736350,"identity":"3ea0c866-93c8-427a-8533-ec8bc7f2b237","order_by":4,"name":"Hailing Zhang","email":"","orcid":"","institution":"Yantai University","correspondingAuthor":false,"prefix":"","firstName":"Hailing","middleName":"","lastName":"Zhang","suffix":""}],"badges":[],"createdAt":"2026-02-08 09:38:42","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8820590/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8820590/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104781314,"identity":"52dbcb33-8c9a-4655-8f14-9466e2003de6","added_by":"auto","created_at":"2026-03-17 07:55:23","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":111933,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Reaction of different forms of D-xylulose with cysteine-carbazole reagent; (b) Absorbance spectra in the full-wavelength scan (300-700 nm): D-xylulose, xylitol, cysteine-carbazole reagent, cysteine-carbazole reagent and xylitol, and cysteine-carbazole reagent and D-xylulose.\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/1e35a2e32d28c75cb44c6a56.jpg"},{"id":104489399,"identity":"1e81dfc9-0959-4f16-9445-03366e6a5b9d","added_by":"auto","created_at":"2026-03-12 11:11:51","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":81982,"visible":true,"origin":"","legend":"\u003cp\u003eSingle point mutation of \u003cem\u003eGo\u003c/em\u003eXDH.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/a84cccfe1d453161d498bea1.jpg"},{"id":104489400,"identity":"11e9c92d-e7cd-4650-ad78-f7f2ff4f4da1","added_by":"auto","created_at":"2026-03-12 11:11:51","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":37298,"visible":true,"origin":"","legend":"\u003cp\u003eThe proposed catalytic mechanism of \u003cem\u003eGo\u003c/em\u003eXDH.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/97f9e6f734df4001d3c59690.jpg"},{"id":104489398,"identity":"573e538e-5d9b-46fb-a770-02e5adb3532d","added_by":"auto","created_at":"2026-03-12 11:11:51","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":172526,"visible":true,"origin":"","legend":"\u003cp\u003eMolecular docking analysis of NADH and D-xylulose into wild-type \u003cem\u003eGo\u003c/em\u003eXDH (a) and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (b); Representation of hydrogen bond networks between key residues in wild-type \u003cem\u003eGo\u003c/em\u003eXDH (c) and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (d).\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/2755b4323e39a14fc05b7488.jpg"},{"id":104783554,"identity":"4c6cfab1-4b8a-474e-983c-88d161604935","added_by":"auto","created_at":"2026-03-17 08:01:32","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":117785,"visible":true,"origin":"","legend":"\u003cp\u003eAnalysis of hydrogen bond interactions at the dimer interface of wild-type \u003cem\u003eGo\u003c/em\u003eXDH (a) and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (b).\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/ec28205720540a5cf6d486e4.jpg"},{"id":104780891,"identity":"df2f3837-8940-4288-8a99-19ea97affed1","added_by":"auto","created_at":"2026-03-17 07:54:12","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":74602,"visible":true,"origin":"","legend":"\u003cp\u003eEffects of different glucose addition times on the production of xylitol in recombinant \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH (a) 0 h; (b) 6.5 h; (c) 10 h; (a) 12 h.\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/4dc2ee09a41593fd1ac58cd7.jpg"},{"id":104785964,"identity":"9b971626-16bc-45a9-a832-3d8e65d65440","added_by":"auto","created_at":"2026-03-17 08:14:01","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1966964,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/dad88fba-d79e-4cb5-aa38-a18362287ae5.pdf"},{"id":104489401,"identity":"171fbd71-6730-44de-9ce5-88b7766099c2","added_by":"auto","created_at":"2026-03-12 11:11:51","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":1139267,"visible":true,"origin":"","legend":"","description":"","filename":"Supplymentmetariel.docx","url":"https://assets-eu.researchsquare.com/files/rs-8820590/v1/9b1ffeca3d0c0c55293a037f.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eImproving xylitol production by semi-rational engineering of xylitol dehydrogenase and optimizing cofactor regeneration in Gluconobacter oxydans\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eXylitol, a sugar alcohol found in nature, serves extensively as a sugar substitute in food and pharmaceutical applications \u0026zwnj;[1]. It is considered suitable for diabetics due to its comparable sweetness to sucrose and its insulin-independent metabolism [2]. In recent years, xylitol has been in demand globally at 161.5\u0026nbsp;million tons per annum, growing at 6% per annum, with promising applications in the food and pharmaceutical sectors [3].\u003c/p\u003e \u003cp\u003eThere are two primary routes for xylitol production: chemical high‑pressure hydrogenation of xylose using metal catalysts, and microbial biotransformation from xylose [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Both methods depend on xylose derived from hemicellulose-xylan hydrolysates, but they face significant challenges like large acid/base requirements, serious environmental impact, and complicated processes [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. In contrast, a \u0026zwnj;glucose-based bioconversion process\u0026zwnj; offers a cleaner and more cost-effective alternative for xylitol production. The process comprises two biotransformation steps: first, osmophilic yeasts ferment glucose to D-arabitol; second, xylitol is catalyzed from D-arabitol using resting cells of \u003cem\u003eGluconobacter oxydans\u003c/em\u003e [5]. \u003cem\u003eG. oxydans\u003c/em\u003e is a gram‑negative obligate aerobe noted for rapid, incomplete oxidation of many sugars and alcohols via membrane‑bound dehydrogenases [6]. In this process, D-arabitol is converted to D-xylulose by the membrane-bound PQQ-dependent D-arabitol dehydrogenase (mAraDH), after which cytoplasmic NADH-dependent xylitol dehydrogenase (XDH) reduces D-xylulose to xylitol. The membrane-bound mAraDH can fully convert D‑arabitol to D‑xylulose, whereas only about 25% of D‑xylulose is subsequently reduced to xylitol. [7]. The poor catalytic efficiency and coenzyme dependence of XDH make it the key rate-limiting enzyme in xylitol biosynthesis [8].\u003c/p\u003e \u003cp\u003eRecently, several attempts have been made to promote xylitol production. Exogenous NADH supplementation is an alternative strategy to overcome coenzyme imbalance in XDH catalysis and achieve high xylitol yields. However, this approach remains economically unviable for industrial-scale applications because of the high cost of NADH [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Consequently, developing an efficient coenzyme regeneration system in recombinant cell factories has emerged as a critical approach to achieve cofactor self-sufficiency and improve xylitol production from D-arabitol in whole-cell biocatalysis. This process can be accomplished by introducing key genes encoding glucose dehydrogenase (GDH), alcohol dehydrogenase (ADH), or glucose-6-phosphate dehydrogenase (G6PDH) into \u003cem\u003eEscherichia coli\u003c/em\u003e or \u003cem\u003eG. oxydans\u003c/em\u003e, with ethanol or glucose serving as co-substrates to establish a robust NAD\u003csup\u003e+\u003c/sup\u003e/NADH redox cycling system [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e, \u003cspan additionalcitationids=\"CR9\" citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. However, the relatively low activity of the XDH enzyme remains a critical factor limiting xylitol synthesis. Therefore, applying directed evolution techniques to improve XDH catalytic efficiency in \u003cem\u003eG. oxydans\u003c/em\u003e represents a promising strategy for high-yield xylitol production.\u003c/p\u003e \u003cp\u003eXylitol dehydrogenase (XDH; EC 1.1.1.9) is an oxidoreductase that catalyzes the reversible interconversion of xylitol and D-xylulose with NADH or NAD\u003csup\u003e+\u003c/sup\u003e as the respective cofactors [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. The enzyme exhibits optimal reducing activity at 30\u0026deg;C and pH 6.0, whereas its oxidation activity peaks at 35\u0026deg;C and pH 9.0 [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Current research on XDH modification remains limited, with most studies concentrating on the regulation of its oxidative activity. To improve D-xylulose yield, XDH has been engineered by site-directed mutagenesis to modify its thermostability and coenzyme specificity. For example, the double mutant (D38S/M39R) and the triple mutant (D206A/I207R/F208S) altered the cofactor preference of XDH from NAD⁺ to NADP\u003csup\u003e+\u003c/sup\u003e [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]\u003csup\u003e\u0026zwnj;\u003c/sup\u003e. In the literature of Watanabe [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e], substituting S96/S99/Y102 with cysteine residues in \u003cem\u003ePs\u003c/em\u003eXDH introduced a structural zinc atom, thereby significantly enhancing its catalytic efficiency and thermostability (Tm increased by 4.5\u0026deg;C)\u0026zwnj;. To date, no studies have employed protein engineering strategies to rationally modify XDH to enhance its reducing activity. A critical limitation lies in the fact that traditional approaches, such as site-directed mutagenesis\u0026zwnj; and \u0026zwnj;random mutagenesis\u0026zwnj;, are labor-intensive and inefficient for rapidly screening high-activity XDH mutants.\u003c/p\u003e \u003cp\u003eIn this study, a \u0026zwnj;semi-rational modification\u0026zwnj; combining \u0026zwnj;structural analysis\u0026zwnj; with \u0026zwnj;computational hotspot prediction\u0026zwnj; (Hotspot Wizard 3.1) was employed to enhance XDH catalytic efficiency. And a high‑throughput screening method was established to accelerate and streamline the screening process. Finally, the highly active XDH was introduced into \u003cem\u003eG. oxydans\u003c/em\u003e to construct an efficient one-step biotransformation system for converting D-arabitol to xylitol.\u003c/p\u003e"},{"header":"2. Methods and materials","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Strains, plasmids, media, and culture conditions\u003c/h2\u003e \u003cp\u003eStrains and plasmids used in this study are listed in Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e. The recombinant pET-28b-\u003cem\u003exdh\u003c/em\u003e, pET-28b-\u003cem\u003egdh\u003c/em\u003e, and pET-28b-\u003cem\u003evhb\u003c/em\u003e were artificially synthesized from Sangon Biotech (Shanghai) Co., Ltd [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e]. The 786 bp DNA sequence of \u003cem\u003exdh\u003c/em\u003e was optimized by adjusting the average GC content from 61.7% to 56.5% and increasing the CAI amount from 0.67 to 0.99, and a C-terminal His-tag was introduced to facilitate recombinant enzyme purification. These recombinant plasmids were then transformed into \u003cem\u003eE. coli\u003c/em\u003e BL21 (DE3). For triparental mating, \u003cem\u003eE. coli\u003c/em\u003e DH5α served as the plasmid construction and maintenance host, \u003cem\u003eE. coli\u003c/em\u003e HB101 carrying helper plasmid pRK2013 was used for biotransformation, and the broad‑host‑range vector pBBR1MCS‑5 was employed to express native and heterologous genes in \u003cem\u003eG. oxydans\u003c/em\u003e [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. The strain \u003cem\u003eG. oxydans\u003c/em\u003e 621H was purchased from BeNa Culture Collection Co., Ltd (China) and used for xylitol production [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cem\u003eE. coli\u003c/em\u003e was cultured in Luria-Bertani (LB) medium containing 5.0 g/L yeast extract, 10.0 g/L tryptone, and 10.0 g/L NaCl. After autoclaving at 121\u0026deg;C for 20 min, \u003cem\u003eE. coli\u003c/em\u003e strains were cultivated at 37\u0026deg;C and 150 rpm in LB medium containing 50 mg/L kanamycin or 50 mg/L gentamicin when plasmid selection was required.\u003c/p\u003e \u003cp\u003e \u003cem\u003eG. oxydans\u003c/em\u003e seed medium consisted of 55.0 g/L D-sorbitol, 20.0 g/L yeast extract, 5.0 g/L KH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, and 5.0 g/L K\u003csub\u003e2\u003c/sub\u003eHPO\u003csub\u003e4\u003c/sub\u003e (pH 6.5), while the fermentation medium contained 80.0 g/L D-sorbitol, 20.0 g/L yeast extract, 5.0 g/L KH\u003csub\u003e2\u003c/sub\u003ePO\u003csub\u003e4\u003c/sub\u003e, and 5.0 g/L K\u003csub\u003e2\u003c/sub\u003eHPO\u003csub\u003e4\u003c/sub\u003e (pH 6.5). Cells from a 48-h agar slant were transferred into 250 mL flasks containing 50 mL seed medium and incubated on a rotary shaker at 150 rpm and 30\u0026deg;C for 36 h. A 0.5 mL aliquot of this seed culture was inoculated into a 500 mL Erlenmeyer flask containing 100 mL fermentation medium, and flasks were incubated on an orbital shaker at 150 rpm and 30\u0026deg;C for 24 h.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Construction of site-saturation and site-directed mutagenesis libraries\u003c/h2\u003e \u003cp\u003eThe primers for introducing NDT codons in site-saturation mutagenesis (SSM) are listed in Table S2. pET-28b-\u003cem\u003exdh\u003c/em\u003e served as the template for PCR. The SSM PCR mix contained 25 \u0026micro;L 2\u0026times;Phanta Max Buffer, 1 \u0026micro;L dNTP mix, 1 \u0026micro;L DNA polymerase, 1 \u0026micro;L plasmid template (40 ng/\u0026micro;L), 1 \u0026micro;L each of forward and reverse primers (10 \u0026micro;M), and 20 \u0026micro;L deionized water. The cycling program was: 95\u0026deg;C for 3 min (pre-denaturation), 35 cycles of 95\u0026deg;C for 15 s, 56\u0026deg;C for 15 s, 72\u0026deg;C for 3 min, and a final extension at 72\u0026deg;C for 10 min. PCR products were digested with \u003cem\u003eDpn\u003c/em\u003eI at 37\u0026deg;C for 2 h and then transformed into \u003cem\u003eE. coli\u003c/em\u003e BL21(DE3). Single colonies from agar plate were inoculated into a 96-deep well plate for high-throughput screening as described in section \u003cspan refid=\"Sec6\" class=\"InternalRef\"\u003e2.4\u003c/span\u003e. Positive clones were verified by DNA sequencing and stored at -80\u0026deg;C in sterile 30% (v/v) glycerol.\u003c/p\u003e \u003cp\u003ePrimers for site‑directed mutagenesis (SM) are listed in Table S3. Using pET‑28b‑\u003cem\u003exdh\u003c/em\u003e as template, the PCR mix and cycling conditions matched those used for SSM. PCR products were digested with \u003cem\u003eDpn\u003c/em\u003eI at 37\u0026deg;C for 2 h, transformed into competent cells of \u003cem\u003eE. coli\u003c/em\u003e, and plated for 12\u0026ndash;16 h. Single colonies were grown in LB, and XDH activity was assayed as described in section \u003cspan refid=\"Sec5\" class=\"InternalRef\"\u003e2.3\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Activity assaya\u003c/h2\u003e \u003cp\u003eCells from 40 mL broth were harvested by centrifugation (9000\u0026times;g, 10 min) and resuspended in 0.8% saline for XDH activity measurement. Separately, cells were resuspended in 10 mL PBS (100 mM, pH 6.0) and disrupted by ultrasonication (40 W, 1 s/1 s) for 20 min. After centrifugation to remove debris, the supernatant was used to assay XDH activity.\u003c/p\u003e \u003cp\u003eScreening for high-vigor mutants was performed by detecting NADH [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]. The reaction system (200 \u0026micro;L) consisting of PBS (0.1 mol/L, pH 6.0), 2 mM D-xylulose, and a certain amount of XDH enzyme was added. After adding 1 mmol/L NADH and mixing, absorbance changes at 340 nm were measured over 5 min. The enzyme activity unit (U) was defined as the amount of enzyme consumed to oxidize 1 \u0026micro;mol NADH per minute at 37\u0026deg;C. The enzyme activity of XDH (U/mL)\u0026thinsp;=\u0026thinsp;D\u0026times;V\u0026times;(△A/△t)/Ɛ\u0026times;L\u0026times;v. D: dilution multiple; V: total reaction volume; △A/△t: change of absorbance in unit time; Ɛ=6.22\u0026times;10\u003csup\u003e3\u003c/sup\u003e L\u0026middot;mol\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e\u0026middot;cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e; L: optical path; v: volume of the enzyme solution.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Establishment of a high-throughput screening method for mutants\u003c/h2\u003e \u003cdiv id=\"Sec7\" class=\"Section3\"\u003e \u003ch2\u003e2.4.1 Preparation of cysteine-carbazole reagent\u003c/h2\u003e \u003cp\u003eThe cysteine-carbazole reagent consists of 15 g/L cysteine hydrochloride solution (0.375 g cysteine hydrochloride dissolved in 25 mL of distilled water), 1.2 g/L carbazole solution (30.0 mg carbazole dissolved in 25 mL alcohol, stored away from light in a brown bottle), and 70% H\u003csub\u003e2\u003c/sub\u003eSO\u003csub\u003e4\u003c/sub\u003e. The above solutions were prepared respectively and mixed according to the ratio of 70% H\u003csub\u003e2\u003c/sub\u003eSO\u003csub\u003e4\u003c/sub\u003e: cysteine hydrochloride solution: carbazole solution\u0026thinsp;=\u0026thinsp;30:1:1 when used.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section3\"\u003e \u003ch2\u003e2.4.2 High-throughput screening for mutants\u003c/h2\u003e \u003cp\u003eFigure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e illustrates the high‑throughput screening procedure. Single colonies from agar plates were inoculated into individual wells of 96‑deep‑well plate X (1.0 mL LB with 50 \u0026micro;g/mL kanamycin per well) and incubated at 37\u0026deg;C for 6 h. A 100 \u0026micro;L aliquot from each well of plate X was transferred to the corresponding well of plate Y containing 100 \u0026micro;L sterile 30% (w/v) glycerol and stored at -80\u0026deg;C for preservation. To the remaining 900 \u0026micro;L culture, 100 \u0026micro;L fresh LB containing 50 \u0026micro;g/mL kanamycin and 1 mM IPTG was added, and cultured at 28\u0026deg;C for 16 h. Cells were harvested by centrifugation (3000\u0026times;g, 30 min) and resuspended in 0.8% saline. The cells were resuspended with 200 \u0026micro;L reaction mixture (100 mM PBS, 100 mM D-xylulose, pH 7.0). The plates were stirred at 30\u003csup\u003eo\u003c/sup\u003eC, 180 rpm for 1 h and then centrifugated at 3000\u0026times;g for 30 min. A 40 \u0026micro;L aliquot of each diluted supernatant was added to a 96‑well microtiter plate containing 160 \u0026micro;L cysteine-carbazole reagent. The 96-well microtiter plate was incubated at 60\u0026deg;C for 10 min, and then colorimetric measurements were performed on a microplate reader at 540 nm. The content of D‑xylulose was calculated using the standard curve, and then the activity of XDH was calculated. One unit of the catalytic activity of XDH was defined as the amount of enzyme that consumed 1 \u0026micro;moL D-xylulose per minute at 30\u003csup\u003eo\u003c/sup\u003eC. The standard curve between the D-xylulose (mg/L) and OD\u003csub\u003e540\u003c/sub\u003e was y\u0026thinsp;=\u0026thinsp;0.0063x\u0026thinsp;+\u0026thinsp;0.1542, R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.99.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section3\"\u003e \u003ch2\u003e2.4.3 Secondary screening\u003c/h2\u003e \u003cp\u003eThe corresponding duplicate from plate Y was propagated in shake flasks for re-screening. This was followed by a rescreening process using the screening method for detecting NADH, as described in section \u003cspan refid=\"Sec5\" class=\"InternalRef\"\u003e2.3\u003c/span\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Protein purification and determination of kinetic parameters of wild-type and mutant XDH\u003c/h2\u003e \u003cp\u003eAfter spinning at 9000 \u0026times; g for 10 min to pellet debris, the supernatant was passed through a Ni\u003csup\u003e2+\u003c/sup\u003e‑NTA column; purified His‑tagged proteins were evaluated by SDS‑PAGE. The nickel ion column was first equilibrated using 20 mL of Binding Buffer (containing 50 mM NaCl, 20 mM PBS at pH 8.0). Then, 10 mL of crude protein was sampled, and a 20 mM imidazole solution (containing 50 mM NaCl, 20 mM imidazole, 20 mM PBS at pH 8.0) was used to elute the miscellaneous protein. The protein elution volume was 20 mL. Finally, the target protein was eluted using 5 mL of 300 mM imidazole solution (containing 50 mM NaCl, 300 mM imidazole, 20 mM PBS with pH 8.0). At the end of elution, the nickel ion column was washed with Binding Buffer and preserved with a 20% ethanol solution. The collected pure proteins were subjected to ultrafiltration membrane filtration utilizing 0.1 M PBS three times to remove excess imidazole and stored in 10% glycerol at -20\u0026deg;C. After centrifuging at 9000\u0026times;g for 10 min at 4℃, the supernatant was collected as pure protein. The precipitate was resuspended with 10 mL of 50 mM PBS (pH 8.0). A specific amount of enzyme solution was added to the loading buffer and boiled for 15 min before sampling.\u003c/p\u003e \u003cp\u003eActivity dependence on temperature was measured at pH 6.0 over a 20-55\u003csup\u003eo\u003c/sup\u003eC range. The determination of the optimal pH of XDH was carried out between 5.0 and 10.0 at 30\u003csup\u003eo\u003c/sup\u003eC, in which the buffer solution was replaced with citric acid-sodium citrate buffer solutions, pH 5.0\u0026ndash;6.0; PBS, pH 6.0\u0026ndash;9.0, and Gly-NaOH buffer solutions, pH 9.0\u0026ndash;10.0, respectively. Kinetic parameters were measured with D‑xylulose concentrations of 1\u0026ndash;10 g/L in PBS (50 mM, pH 7.5) at 35\u0026deg;C. Samples were centrifuged (9000\u0026times;g, 10 min) and the supernatants assayed as described above. \u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003eM\u003c/em\u003e\u003c/sub\u003e and \u003cem\u003eV\u003c/em\u003e\u003csub\u003e\u003cem\u003emax\u003c/em\u003e\u003c/sub\u003e were obtained by nonlinear regression fitting to the Michaelis-Menten equation using OriginPro 8.0.\u003c/p\u003e \u003cp\u003eThe kinetic parameters were determined with varying concentrations of D-xylulose (1\u0026ndash;10 g/L) in PBS (50 mM, pH 7.5) at 35\u003csup\u003eo\u003c/sup\u003eC. Samples were centrifuged at 9000\u0026times;g for 10 min and the supernatants assayed as described above. The values of \u003cem\u003eKm\u003c/em\u003e and \u003cem\u003eVmax\u003c/em\u003e were determined by nonlinear regression fitting to the Michaelis-Menten equation using OriginPro 8.0.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003e2.6 Molecular homology modeling and docking\u003c/h2\u003e \u003cp\u003eUsing SWISS‑MODEL, a three‑dimensional model of \u003cem\u003eGo\u003c/em\u003eXDH was generated based on XDH (PDB: 1ZEM; 99.62% sequence identity). The structures of D‑xylulose and NADH were generated in Chem3D and docked into the homology model with AutoDock Vina, and docking results were visualized and analyzed in PyMOL.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003e2.7 DNA manipulation and plasmid construction in \u003cem\u003eG. oxydans\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eThe strong promoter \u003cem\u003egHp0169\u003c/em\u003e was chosen to drive expression of the target genes. The open reading frame (ORF) of \u003cem\u003exdh\u003c/em\u003e, including its endogenous ribosome‑binding site (RBS) from \u003cem\u003eG. oxydans\u003c/em\u003e, was amplified by PCR using the primers listed in Table S4. These genes were cloned into the previously constructed plasmid pBBR-\u003cem\u003egHp0169\u003c/em\u003e, resulting in plasmids pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e, pBBR-\u003cem\u003egHp0169-gdh\u003c/em\u003e, and pBBR-\u003cem\u003egHp0169-vhb\u003c/em\u003e [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. And the genes of \u003cem\u003egdh\u003c/em\u003e and \u003cem\u003evhb\u003c/em\u003e were cloned from pET-28b-\u003cem\u003egdh\u003c/em\u003e and pET-28b-\u003cem\u003evhb\u003c/em\u003e and inserted into plasmid pBBR-\u003cem\u003egHp0169\u003c/em\u003e, resulting in plasmids pBBR-\u003cem\u003egHp0169-gdh\u003c/em\u003e and pBBR-\u003cem\u003egHp0169-vhb\u003c/em\u003e. According to the SM method, mutations S77C/S106N/A110N were introduced into pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e, resulting in pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e\u003csub\u003eM3\u003c/sub\u003e (Fig. S2). Then the PCR product of \u003cem\u003egHp0169-gdh\u003c/em\u003e was amplified and inserted into pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e\u003csub\u003eM3\u003c/sub\u003e, resulting in pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e\u003csub\u003eM3\u003c/sub\u003e\u003cem\u003e-gHp0169\u003c/em\u003e-\u003cem\u003egdh\u003c/em\u003e (Fig. S2). Finally, the PCR product of \u003cem\u003egHp0169-vhb\u003c/em\u003e was cloned into pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e\u003csub\u003eM3\u003c/sub\u003e\u003cem\u003e-gHp0169\u003c/em\u003e-\u003cem\u003egdh\u003c/em\u003e, resulting in pBBR-\u003cem\u003egHp0169-xdh\u003c/em\u003e\u003csub\u003eM3\u003c/sub\u003e\u003cem\u003e-gHp0169\u003c/em\u003e-\u003cem\u003egdh-gHp0169-vhb\u003c/em\u003e (Fig. S2). Plasmids were introduced into \u003cem\u003eG. oxydans\u003c/em\u003e by triparental mating and selected on gentamicin, yielding recombinant strains \u003cem\u003eG. oxydans\u003c/em\u003e/XDH, \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e, \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e‑GDH, and \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e‑GDH‑VHb [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Recombinant strains were cultured in fermentation medium containing 50 mg/L gentamicin, as described in Section \u003cspan refid=\"Sec3\" class=\"InternalRef\"\u003e2.1\u003c/span\u003e. Cells from 200 mL broth were harvested by centrifugation (9000\u0026times;g, 10 min), resuspended in 0.8% saline, and used for biotransformation of xylitol from 30 g/L D‑arabitol. For \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e‑GDH, 20 g/L glucose was added at 0, 6.5, 10, and 12 h in the reaction solution to promote NADH regeneration. After the reaction, debris was removed by centrifugation and the supernatant was analyzed for xylitol, D-arabitol, and D-xylulose concentrations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003e2.8 Cell cultivation in a stirred bioreactor\u003c/h2\u003e \u003cp\u003eCell cultivation was performed in a 10-L multi-bioreactor system (BLBIO-50SJ-5, China) using the fermentation medium. The bioreactor was inoculated at 10% (v/v) and run at 30\u003csup\u003eo\u003c/sup\u003eC for 24 h with agitation set to 400, 500, 600, or 700 rpm and 1.5 vvm aeration. Cultures were centrifuged at 9000\u0026times;g for 10 min; the supernatant was used to quantify substrate consumption, and the cell pellet was used for the biotransformation of D‑arabitol to xylitol.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003e2.9 Biotransformation of D-arabitol to xylitol using resting cells of recombinant \u003cem\u003eG. oxydans\u003c/em\u003e\u003c/h2\u003e \u003cp\u003eBioconversion of D-arabitol to xylitol was carried out in 500 mL shaking flasks using 100 mL reaction mixtures containing wet cells (100 g/L) and 40 g/L D-arabitol; 30 g/L CaCO3 was added to maintain pH. Reactions were performed at 30\u0026deg;C and 180 rpm. And 20 g/L glucose was added at 10 h to facilitate NADH regeneration. After completion, samples were centrifuged and the supernatants analyzed by HPLC for product concentrations.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec15\" class=\"Section2\"\u003e \u003ch2\u003e2.10 Analytical methods\u003c/h2\u003e \u003cp\u003eBiomass was estimated by measuring optical density at 600 nm (OD\u003csub\u003e600\u003c/sub\u003e) and converted to dry cell weight (DCW) using the calibration DCW (g/L)\u0026thinsp;=\u0026thinsp;0.4627x-0.2685, R\u003csup\u003e2\u003c/sup\u003e\u0026thinsp;\u0026gt;\u0026thinsp;0.99. Concentrations of D‑sorbitol, D‑arabitol, D‑xylulose, and xylitol were measured by HPLC with a differential refractive index detector (Waters, USA) using an Aminex HPX‑87H column (300\u0026times;7.8 mm; Bio‑Rad) at 55\u0026deg;C, a flow rate of 0.5 mL/min, and 5 mM H\u003csub\u003e2\u003c/sub\u003eSO\u003csub\u003e4\u003c/sub\u003e as the mobile phase [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. Experiments were performed in triplicate, and mean values\u0026thinsp;\u0026plusmn;\u0026thinsp;SD from three independent batches are reported.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results and Discussion","content":"\u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003e3.1 Establishment and optimization of the colorimetric assay\u003c/h2\u003e \u003cp\u003eA rapid assay for XDH activity was developed using the cysteine-carbazole chromogenic reaction with D‑xylulose. The mixture of cysteine-carbazole can react with D-xylulose under acidic conditions, showing a purple color, and the color becomes darker with the increase of D-xylulose concentration. To develop an efficient assay for monitoring D‑xylulose levels during the reduction process, the reaction conditions of cysteine-carbazole and D-xylulose were optimized. Different D‑xylulose concentrations (10\u0026ndash;50 mg/L) were loaded into columns 1\u0026ndash;5, and 120\u0026ndash;200 \u0026micro;L of cysteine-carbazole reagent was added to wells in rows A-E, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). The plate was incubated at 60\u0026deg;C for 10 min, producing distinct color changes in the wells. In row C, when using 160 \u0026micro;L cysteine-carbazole reagent as the initial developing agent volume, the color change was easier to distinguish. The absorbance spectrum of the colorimetric assay showed a characteristic peak at 540 nm, while the color reaction of cysteine-carbazole reagent with other substrates such as xylitol did not show a characteristic absorption peak in 300\u0026ndash;700 nm wavelength range (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). In addition, this method is reliable and has a good linear correlation with the XDH activity when the concentration of D-xylulose is in the range of 10\u0026ndash;100 mg/L, making it suitable for the qualitative analysis and quantification of the products. Therefore, in the following experiments, 160 \u0026micro;L of cysteine-carbazole reagent and 540 nm was selected as the chromogenic volume and working wavelength.\u003c/p\u003e \u003cp\u003eBased on the above results, we established a high-throughput screening method to identify strains with superior XDH activity. Enzymatic reactions were performed in PBS (100 mM, pH 6.0) containing 10 mM D‑xylulose and incubated at 35\u0026deg;C for 1 h. A 40 \u0026micro;L aliquot of the diluted reaction mixture was added to 160 \u0026micro;L cysteine-carbazole reagent, and absorbance was measured at 540 nm for the colorimetric assay. This enzyme activity detection method shortened the reaction time, avoided complicated operations, and improved the screening efficiency.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003e3.2 Computational identification of beneficial residue positions\u003c/h2\u003e \u003cp\u003eSemi-rational design based on structural analysis is a useful way to enhance the catalytic activity of \u003cem\u003eGo\u003c/em\u003eXDH (xylitol dehydrogenase from \u003cem\u003eG. oxydans\u003c/em\u003e 621H). However, the large sequence libraries make the screening process laborious and time-consuming. HotSpot Wizard 3.1 is a powerful online tool that integrates structural, functional, and evolutionary information from various bioinformatics resources to identify residue positions critical for enzymatic activity [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. By restricting mutagenesis to a limited set of \u0026ldquo;hot spot\u0026rdquo; positions, a \u0026ldquo;small but smart\u0026rdquo; comprehensive mutant library can be constructed, which can significantly accelerate directed evolution workflows [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eInitially, the \u003cem\u003eGo\u003c/em\u003eXDH model was constructed using the crystal structure of XDH (PDB ID: 1zem.1. A), which shares 99.62% sequence identity\u0026zwnj;. As a member of the short-chain dehydrogenase/reductase (SDR) superfamily, there are four conserved functional domains in \u003cem\u003eGo\u003c/em\u003eXDH: the \u003cem\u003eN\u003c/em\u003e-terminal glycine-rich coenzyme-binding region (N90-N91-A92-G93), the catalytic tetrad (N116-S144-Y157-K161), the central domain (T13-G14-A15-G16-G17-N18-I19-G20), and the \u003cem\u003eC\u003c/em\u003e-terminal substrate-binding region (P187-G188) [12]. The \u003cem\u003eGo\u003c/em\u003eXDH enzyme assembles as a tetramer, each subunit containing a Rossmann fold: a central 7‑strand β‑sheet flanked by short helices αB, αC and αG on one side and longer helices αD, αE and αF on the other (Fig. S3). Additionally, two short helices (αFG1 and αFG2) extend outward above the Rossmann fold. Subsequently, the catalytic pocket hotspots identified through molecular docking analysis, along with consensus residues predicted by the HotSpot Wizard, were designated as target sites for mutagenesis. A mutant library was constructed with 13 predicted beneficial sites: S77, F87, D103, S106, A110, V117, T118, K149, P152, A156, Y158, and G162. Finally, site‑saturation mutagenesis was conducted on these target residues, and positive variants were identified with the previously established high‑throughput assay.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003e3.3 Screening of variants with improved activity\u003c/h2\u003e \u003cp\u003eAfter several rounds of mutagenesis, four variants, S77C, S106N, and A110N, exhibiting improved activity were obtained. Compared with the wild-type \u003cem\u003eGo\u003c/em\u003eXDH, they showed 29.8%, 42.7%, and 48.6% improvement in activity toward D-xylulose (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003e). The structural analysis revealed that the S106N and A110N mutations located near N116 in the catalytic tetrad (N116-S144-Y157-K161) and on the helix αE (a critical interaction surface for the main subunits) led to a substantial enhancement of catalytic activity. To identify additional beneficial mutations, we performed site-saturation mutagenesis on non-conserved residues within the αE helix (D107, R111, L113, T114, I115, T118, H122, S128, R129, G130, T133, and Q134). After several rounds of mutagenesis, several new variants were identified. The T114H and I115R mutants exhibited 18.6% and 32.3% increases in their activity compared to the wild-type \u003cem\u003eGo\u003c/em\u003eXDH. In the catalytic tetrad of \u003cem\u003eGo\u003c/em\u003eXDH, Y157 acts as a catalytic base, S144 stabilizes a negative charge on the catalytic intermediate, and K161 interacts with the nicotinamide ribose, thus lowering the \u003cem\u003epKa\u003c/em\u003e of the tyrosine hydroxyl (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003e) [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. The function of N116 is to bridge the substrate‑binding loop and the active site using conserved water molecules [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. These water molecules create a small hydrophilic cavity linked to other conserved residues in the fold. This pocket acts as a proton-relay system, bridging K161 to the bulk solvent and stabilizing the positively charged K161 [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. All four residues (S106, A110, T114, and I115) on the αE helix were converted to hydrophilic amino acids (N, H, or R), indicating their likely contributes to hydrophilic pocket stabilization via improved solvation and residue-residue interactions. Furthermore, the S77C mutation on the monomer surface may have improved the protein's hydrophobicity, which is crucial for structural stabilization.\u003c/p\u003e \u003cp\u003eCombining beneficial mutations is commonly employed to enhance enzyme performance, as mutations can act synergistically or cooperatively [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. Therefore, combinatorial mutagenesis was conducted to generate multiple base mutations. The date revealed that the activities of the double mutant S77C/S106N and triple mutant S77C/S106N/A110N were 1.92-fold and 2.49-fold higher than that of \u003cem\u003eGo\u003c/em\u003eXDH, respectively (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). However, further combinations with T114H or I115R did not lead to a notable increase in activity. Accordingly, further enzymatic studies concentrated on \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (S77C/S106N/A110N) due to its markedly increased activity.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEnzyme activities of \u003cem\u003eGo\u003c/em\u003eXDH and its variants towards D-xylulose.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eXylitol dehydrogenase\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMutational sites\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eEnzyme activity (U/mL)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRelative activity (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e6.30\u0026thinsp;\u0026plusmn;\u0026thinsp;0.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e100\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM2\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS77C/S106N\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e12.12\u0026thinsp;\u0026plusmn;\u0026thinsp;0.21\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e192\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS77C/S106N/A110N\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e15.70\u0026thinsp;\u0026plusmn;\u0026thinsp;0.18\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e249\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM4\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS77C/ /S106N/A110N/T114H\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e10.52\u0026thinsp;\u0026plusmn;\u0026thinsp;0.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e167\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM4\u0026rsquo;\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eS77C/S106N/A110N/ I115R\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e11.63\u0026thinsp;\u0026plusmn;\u0026thinsp;0.12\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e185\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003e3.4 Enzymatic characterization of wild-type \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e\u003c/h2\u003e \u003cp\u003eTo biochemically characterize the enhanced activity, both parental \u003cem\u003eGo\u003c/em\u003eXDH and the \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e variant were purified, and their kinetic parameters toward D‑xylulose were determined. The purified results were shown in Fig. S4. SDS-PAGE indicated a monomer size of 25\u0026ndash;35 kDa, consistent with the calculated mass of 29.48 kDa. It could be seen that the purified enzyme contained almost no miscellaneous proteins and had a high purity, which was suitable for determining enzymatic properties.\u003c/p\u003e \u003cp\u003eThe optimal temperature of \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e was approximately 35\u003csup\u003eo\u003c/sup\u003eC. And relative activity showed a steady increase from 30\u003csup\u003eo\u003c/sup\u003eC to a peak (100% activity), then declined sharply to its lowest level at 80\u0026deg;C (Fig. S5). Both \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e showed higher activity at neutral pH, and the optimal pH was 7.5 (Fig. S5). These results differed from the 30\u003csup\u003eo\u003c/sup\u003eC and pH 6.0 in the recent reports [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Kinetic parameters for \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e were determined using purified cell extracts from the recombinant strains. Michaelis-Menten curves were fitted to the experimental data for both enzymes, showing a close fit. The extracts containing \u003cem\u003eGo\u003c/em\u003eXDH converted D-xylulose at a maximal rate (\u003cem\u003eV\u003c/em\u003e\u003csub\u003e\u003cem\u003emax\u003c/em\u003e\u003c/sub\u003e) of 7.03 U/mg total protein, and the extracts containing \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e showed a \u003cem\u003eV\u003c/em\u003e\u003csub\u003e\u003cem\u003emax\u003c/em\u003e\u003c/sub\u003e of 29.25 U/mg total protein. The \u003cem\u003eGo\u003c/em\u003eXDH exhibited a \u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003eM\u003c/em\u003e\u003c/sub\u003e for D-xylulose of 0.67 mM, and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e exhibited a \u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003eM\u003c/em\u003e\u003c/sub\u003e of 0.52 mM. Comparative analysis showed that \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e exhibits a \u003cem\u003eV\u003c/em\u003e\u003csub\u003e\u003cem\u003emax\u003c/em\u003e\u003c/sub\u003e 4.2-fold higher and a \u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003eM\u003c/em\u003e\u003c/sub\u003e for D‑xylulose 23.2% lower than \u003cem\u003eGo\u003c/em\u003eXDH.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003e3.5 Mechanism elucidation of activity enhancement\u003c/h2\u003e \u003cp\u003eTo elucidate the enhanced catalytic mechanism, D‑xylulose was docked into models of \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e. Comparison of docking poses revealed that the distance between the hydrogen of Y157 and the substrate carbonyl oxygen shortened from 3.4 \u0026Aring; to 2.3 \u0026Aring; in \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e, while the distance between the C4 hydrogen of NADH\u0026rsquo;s nicotinamide ring and the substrate carbonyl carbon slightly extended from 3.7 \u0026Aring; to 4.0 \u0026Aring; (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, b). The shortened Y157-substrate distance likely facilitates more efficient proton/electron transfer and increases substrate affinity in \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e, accounting for its lower \u003cem\u003eK\u003c/em\u003e\u003csub\u003e\u003cem\u003eM\u003c/em\u003e\u003c/sub\u003e and improved catalytic efficiency.\u003c/p\u003e \u003cp\u003eHydrogen bond analysis reveals distinct differences in substrate binding between \u003cem\u003eGo\u003c/em\u003eXDH and \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e. In the wild-type \u003cem\u003eGo\u003c/em\u003eXDH, three hydrogen bonds formed between the substrate and adjacent residues (Q95: 2.5 \u0026Aring;; S144: 2.2 \u0026Aring;; A146: 2.5 \u0026Aring;, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003ec). In contrast, \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e formed five hydrogen bonds with nearby residues (S144: 1.8 \u0026Aring;, 2.2 \u0026Aring;; A146: 2.5 \u0026Aring;; Y157: 2.3 \u0026Aring;; Q199: 2.5 \u0026Aring;, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003ed). The expanded hydrogen bond network in \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e contributed to more rigid substrate positioning, thus increasing catalytic performance. The reduction of D-xylulose requires cooperative participation of the catalytic tetrad (N116-S144-Y157-K161) and NADH as cofactor. First, Y157 donates a proton to the substrate carbonyl, and then a hydride is transferred to C2 of D‑xylulose (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003e) [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. A proton relay composed of the nicotinamide ribose 2\u0026rsquo;‑hydroxyl, K161, and a water molecule associated with N116\u0026rsquo;s backbone carbonyl facilitates the reaction. Formation of a hydrogen bond between S144 and the substrate carbonyl can stabilize the catalytic intermediate [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. In \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e, the additional hydrogen bond at S144 contributes to maintaining the stability of the catalytic pocket. Meanwhile, a new hydrogen bond was found between the substrate and residue Q199. Analysis of the substrate-binding pocket revealed that Q199, situated adjacent to the substrate channel inlet, pulls the substrate closer to Y157 and the pocket entrance, thereby promoting catalysis efficiency.\u003c/p\u003e \u003cp\u003eCompared to the wild-type \u003cem\u003eGo\u003c/em\u003eXDH, the \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e variant exhibited significant structural modifications at the dimer interface (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e5\u003c/span\u003e). \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e introduced a new hydrogen bond between N110 and H122 on the αE helix, as well as an additional hydrogen bond between N106 and T118. These interactions collectively enhanced the structural rigidity of the dimer. In contrast, only a single hydrogen bond between S106 with K125 was found at the dimer interface of the wild-type \u003cem\u003eGo\u003c/em\u003eXDH. These observations suggest that the introduction of new hydrogen bonds at the dimer interface, particularly in the αE helix region, contributes to stabilizing \u003cem\u003eGo\u003c/em\u003eXDH structure.\u003c/p\u003e \u003cp\u003e \u003cb\u003e3.6 Construction of the NADH coenzyme regeneration system in the production of xylitol by recombinant\u003c/b\u003e \u003cb\u003eG. oxydans\u003c/b\u003e\u003c/p\u003e \u003cp\u003eThe beneficial mutations S77C/S106N/A110N were cloned into pBBR‑\u003cem\u003egHp0169\u003c/em\u003e‑\u003cem\u003exdh\u003c/em\u003e and introduced into \u003cem\u003eG. oxydans\u003c/em\u003e to generate \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e. Flask biotransformation showed \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e produced 8.95 g/L xylitol after 36 h, a 36.9% increase over \u003cem\u003eG. oxydans\u003c/em\u003e/XDH (6.54 g/L) (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), demonstrating superior catalytic efficiency and potential application in whole-cell production of xylitol. Because \u003cem\u003eGo\u003c/em\u003eXDH consumes large amounts of NADH to reduce D‑xylulose to xylitol, effective cofactor regeneration is required to achieve xylitol production, in which glucose dehydrogenase (GDH) is commonly used for this purpose. Therefore, GDH was co‑expressed with \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e in \u003cem\u003eG. oxydans\u003c/em\u003e, and biotransformation was carried out using resting recombinant cells with 20 g/L glucose as co‑substrate. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the recombinant \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH achieved a significantly higher xylitol yield (11.92 g/L) compared to \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eEffects of mutation S77C/S106N/A110N and GDH on the yield of xylitol in \u003cem\u003eG. oxydans\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResting cells\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConversion time (h)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eInitial substrate concentration (g/L)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eXylitol titer (g/L)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eXylitol yield (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e6.54\u0026thinsp;\u0026plusmn;\u0026thinsp;0.34\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e21.80\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e8.92\u0026thinsp;\u0026plusmn;\u0026thinsp;0.36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e29.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e36\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c4\"\u003e \u003cp\u003e11.92\u0026thinsp;\u0026plusmn;\u0026thinsp;0.25\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e39.73\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003eThe reaction solution of 10 mL was composed of wet cells (collected from 200 mL of the fermentation broth) and 30 g/L D-arabitol, in which 30 g/L CaCO\u003csub\u003e3\u003c/sub\u003e was added to maintain pH. The biotransformation was performed at 30\u003csup\u003eo\u003c/sup\u003eC and 180 rpm. For the bioconversion process of \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH, 20 g/L of glucose was added to the reaction solution to enable NADH regeneration.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eAlthough GDH was identified as an effective coenzyme regeneration system for xylitol production, the optimal timing for glucose addition remained unclear. For this purpose, the effects of adding glucose at different times (0, 6.5, 10, and 12 h) on the xylitol production were studied. As depicted in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e6\u003c/span\u003e, the yield of xylitol was significantly higher when glucose was added at 6.5 h (13.14 g/L), 10 h (15.48 g/L), or 12 h (15.36 g/L) than when added at 0 h (11.92 g/L). Results showed that peak xylitol production was achieved when glucose supplementation began at 10 h. At this time point, approximately 25 g/L of the initial D-arabitol had been consumed, accompanied by the accumulation of 23.71 g/L D-xylulose. The results indicated that when D-xylulose reached a critical concentration, the demand for NADH increased; thus, timely glucose supplementation could enhance the utilization efficiency of NADH. In contrast, early-stage glucose supplementation might enable competitive membrane-bound glucose dehydrogenase (mGDH) in \u003cem\u003eG. oxydans\u003c/em\u003e, thereby diminishing overall glucose consumption [\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. By strategically supplementing glucose at 10 h, \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH achieved more efficient coenzyme regeneration, which enhanced xylitol production from D-xylulose.\u003c/p\u003e \u003cp\u003e \u003cb\u003e3.7 Cell cultivation in a stirred bioreactor and efficient conversion of D-arabinitol to xylitol using resting cells of recombinant\u003c/b\u003e \u003cb\u003eG. oxydans\u003c/b\u003e\u003cb\u003e/XDH\u003c/b\u003e\u003csub\u003e\u003cb\u003eM3\u003c/b\u003e\u003c/sub\u003e\u003cb\u003e-GDH-VHb\u003c/b\u003e\u003c/p\u003e \u003cp\u003eAs an obligate aerobic bacterium, \u003cem\u003eG. oxydans\u003c/em\u003e requires a high oxygen supply during cell proliferation to support efficient respiratory metabolism. \u003cem\u003eVitreoscilla\u003c/em\u003e hemoglobin (VHb) is an oxygen-binding protein that facilitates bacterial respiration by directly channeling oxygen to terminal oxidases [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. To enhance intracellular oxygen delivery and improve xylitol biosynthesis, the gene encoding VHb was introduced into the plasmid pBBR-\u003cem\u003egHp0169-xdh-gHp0169-gdh\u003c/em\u003e, resulting in recombinant \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH-VHb. To determine the optimal growth conditions, the recombinant \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH-VHb strain was cultured under varying stirring speeds (400\u0026ndash;700 rpm). As depicted in Fig. S6, the D-sorbitol consumption rate exhibited a marked increase with rising stirring speed, and the maximal cell density at 400, 500, 600, and 700 rpm reached 2.92, 3.55, 3.39, and 3.09 gDCW/L, respectively, at 20 h. To evaluate the catalytic performance of these strains, the biotransformation of D-arabitol to xylitol was performed by resting cells. As shown in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e, in the D4 experiment, the \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e‑GDH‑VHb strain reached a maximum xylitol concentration of 29.02 g/L at 30 h. These results indicate a high oxygen demand during \u003cem\u003eG. oxydans\u003c/em\u003e cell proliferation. Although higher biomass accumulation was observed at 500 and 600 rpm, enzyme production was higher at 700 rpm, resulting in enhanced catalytic activity.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eComparison of xylitol production catalyzed by different resting cells of \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH-VHb.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\"\u0026plusmn;\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eExperiments\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eConversion time (h)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eXylitol titer (g/L)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eXylitol yield (%)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e24.18\u0026thinsp;\u0026plusmn;\u0026thinsp;0.64\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e60.45\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e27.08\u0026thinsp;\u0026plusmn;\u0026thinsp;0.54\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e67.70\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e26.87\u0026thinsp;\u0026plusmn;\u0026thinsp;0.57\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e67.18\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eD4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c2\"\u003e \u003cp\u003e30\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\"\u0026plusmn;\" colname=\"c3\"\u003e \u003cp\u003e29.02\u0026thinsp;\u0026plusmn;\u0026thinsp;0.48\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c4\"\u003e \u003cp\u003e72.55\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"4\"\u003eExperiments D1, D2, D3, and D4 represent the resting cells of \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH-VHb cultured at stirring speeds of 400, 500, 600, and 700 rpm, respectively. The reaction solution of 100 mL was composed of wet cells (100 g/L) and 40 g/L D-arabitol, in which 30 g/L CaCO\u003csub\u003e3\u003c/sub\u003e was added to maintain pH. The biotransformation was performed at 30\u003csup\u003eo\u003c/sup\u003eC and 180 rpm. Furthermore, at 10 h, 20 g/L of glucose was added to the reaction solution to facilitate NADH regeneration during the bioconversion process.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eRecent advancements in microbial xylitol production have increasingly emphasized genetic engineering strategies. Among these approaches, the introduction of XDH into \u003cem\u003eE. coli\u003c/em\u003e has demonstrated significant potential for achieving high-titer xylitol production. The mixed culture of resting \u003cem\u003eG. thailandicus\u003c/em\u003e cells with BL21-\u003cem\u003exdh\u003c/em\u003e and BL21-\u003cem\u003eadh\u003c/em\u003e produced 34.34 g/L xylitol after 48 h of bioconversion, the highest xylitol titer reported to date [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. However, this process is accomplished with the participation of two or even three strains. Therefore, constructing high-efficiency engineered \u003cem\u003eG. oxydans\u003c/em\u003e capable of one-step conversion of D-arabitol to xylitol may be more suitable for production. Zhou [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] reported an engineered strain, \u003cem\u003eG. oxydans\u003c/em\u003e PXPG, co-expressing XDH and GDH, which produced 12.23 g/L xylitol from 30 g/L D-arabitol in 50 h. In this study, we engineered \u003cem\u003eG. oxydans\u003c/em\u003e 621H by co-expressing high-activity \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e, GDH, and VHb, achieving 29.02 g/L xylitol in shake flasks within 30 h, showing superior production efficiency. In conclusion, this study provides an attractive and competitive candidate for the one-step production of xylitol in \u003cem\u003eG. oxydans\u003c/em\u003e.\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Conclusion","content":"\u003cp\u003eIn this study, a semi-rational engineering approach was developed to enhance xylitol production by modifying \u003cem\u003eGo\u003c/em\u003eXDH through structural analysis and computational hotspot prediction (Hotspot Wizard 3.1). A cysteine-carbazole based high-throughput assay for D-xylulose was established to enable rapid screening of XDH variants with enhanced activity. This strategy yielded a variant \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (S77C/S106N/A110N), which exhibited a 2.49-fold increase in catalytic activity. Structural analysis revealed mutation S77C/S106N/A110N significantly shortened the gap between the hydrogen on Y157 and the carbonyl oxygen atom of D‑xylulose, accelerating the transfer rates of protons and electrons. Furthermore, new hydrogen bonds formed between the substrate and proximal residues, as well as at the dimer interface of the αE helix, contributed to the stabilization of both substrate binding and the tetramer structure. Finally, resting cells of engineered \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e‑GDH‑VHb produced approximately 29.02\u0026thinsp;\u0026plusmn;\u0026thinsp;0.48 g/L xylitol from 40 g/L D‑arabitol after 30 h of biotransformation.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eXDH, xylitol dehydrogenase\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eG. oxydans, Gluconobacter oxydans\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eGo\u003c/em\u003eXDH, xylitol dehydrogenase from \u003cem\u003eG. oxydans\u003c/em\u003e 621H\u003c/p\u003e\n\u003cp\u003eNADH, nicotinamide adenine dinucleotide\u003c/p\u003e\n\u003cp\u003eGDH, glucose dehydrogenase\u003c/p\u003e\n\u003cp\u003eVHb, \u003cem\u003eVitreoscilla\u003c/em\u003e hemoglobin\u003c/p\u003e\n\u003cp\u003emAraDH,\u0026nbsp;D-arabitol dehydrogenase\u003c/p\u003e\n\u003cp\u003eADH, alcohol dehydrogenase\u003c/p\u003e\n\u003cp\u003eG6PDH, glucose-6-phosphate dehydrogenase\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eE. coli\u003c/em\u003e, \u003cem\u003eEscherichia coli\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eLB, Luria-Bertani\u003c/p\u003e\n\u003cp\u003eIPTG, Isopropyl-\u0026beta;-D-1-thiogalactopyranoside\u003c/p\u003e\n\u003cp\u003eSSM, site‑saturation mutagenesis\u003c/p\u003e\n\u003cp\u003eSM, site‑directed mutagenesis\u003c/p\u003e\n\u003cp\u003eSDS‑PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis\u003c/p\u003e\n\u003cp\u003ePBS, phosphate buffer saline\u003c/p\u003e\n\u003cp\u003eORFs, open reading frames\u003c/p\u003e\n\u003cp\u003eRBS, ribosomal binding site\u003c/p\u003e\n\u003cp\u003eDCW, dry cell weight\u003c/p\u003e\n\u003cp\u003eHPLC, high performance liquid chromatography\u003c/p\u003e\n\u003cp\u003eSDR, short-chain dehydrogenase/reductase\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;Availability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting Interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was financially supported by the Natural Science Foundation of Shandong Province (Grant number: ZR2022QB111).\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;Authors\u0026apos; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJS performed the conceptualization, methodology, Data curation, Writing-original draft and Writing-review \u0026amp; editing. JG and LL helped in formal analysis, methodology, and writing review and editing. DL performed the conceptualization, supervision, project administration, and writing review and editing. HZ performed the conceptualization, supervision, project administration, and funding acquisition.\u003cstrong\u003e\u003cbr\u003e\u0026nbsp;Acknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eVenkateswar Rao L, Goli JK, Gentela J, Koti S. Bioconversion of lignocellulosic biomass to xylitol: An overview. Bioresour Technol. 2016;213:299\u0026ndash;310.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Li S, Xu H, Zhou P, Zhang L, Ouyang P. Purification of xylitol dehydrogenase and improved production of xylitol by increasing XDH activity and NADH supply in \u003cem\u003eGluconobacter oxydans\u003c/em\u003e. J Agric Food Chem. 2013;61(11):2861\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDasgupta D, Bandhu S, Adhikari DK, Ghosh D. Challenges and prospects of xylitol production with whole cell bio-catalysis: A review. Microbiol Res. 2017;197:9\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRipoll M, Lerma-Escalera JA, Morones-Ramirez JR, Rios-Solis L, Betancor L. New perspectives into \u003cem\u003eGluconobacter\u003c/em\u003e-catalysed biotransformations. Biotechnol Adv. 2023;65:108127.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi X, Zhang H, Magocha TA, An Y, Yun J, Yang M, Xue Y, Liang S, Sun W, Cao Z. Improved xylitol production by expressing a novel D-arabitol dehydrogenase from isolated \u003cem\u003eGluconobacter\u003c/em\u003e sp. JX-05 and co-biotransformation of whole cells. Bioresour Technol. 2017;235:50\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePrust C, Hoffmeister M, Liesegang H, Wiezer A, Fricke WF, Ehrenreich A, Gottschalk G, Deppenmeier U. Complete genome sequence of the acetic acid bacterium \u003cem\u003eGluconobacter oxydans\u003c/em\u003e. Nat Biotechnol. 2005;23(2):195\u0026ndash;200.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSuzuki S, Sugiyama M, Mihara Y, Hashiguchi K, Yokozeki K. Novel enzymatic method for the production of xylitol from D-arabitol by \u003cem\u003eGluconobacter oxydans\u003c/em\u003e. Biosci Biotechnol Biochem. 2002;66(12):2614\u0026ndash;20.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi XH, Zhu JF, Yun JH, Lin J, Qi YL, Guo Q, Xu H. Enhanced xylitol production: Expression of xylitol dehydrogenase from \u003cem\u003eGluconobacter oxydans\u003c/em\u003e and mixed culture of resting cell. J Bioscience Bioengerring. 2016;122(3):257\u0026ndash;62.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou P, Li S, Xu H, Feng X, Ouyang P. Construction and co-expression of plasmid encoding xylitol dehydrogenase and a cofactor regeneration enzyme for the production of xylitol from D-arabitol. Enzym Microb Technol. 2012;51(2):119\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang H, Yun J, Zabed H, Yang M, Zhang G, Qi Y, Guo Q, Qi X. Production of xylitol by expressing xylitol dehydrogenase and alcohol dehydrogenase from Gluconobacter thailandicus and co-biotransformation of whole cells. Bioresour Technol. 2018;257:223\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHong S, Wu J, Zhao H. Cloning, overexpression, purification, and site-directed mutagenesis of xylitol-2-dehydrogenase from \u003cem\u003eCandida albicans\u003c/em\u003e. J Mol Catal B: Enzymatic. 2010;62(1):40\u0026ndash;5.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEhrensberger AH, Elling RA, Wilson DK. Structure-guided engineering of xylitol dehydrogenase cosubstrate specificity. Structure. 2006;14(3):567\u0026ndash;75.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWatanabe S, Kodaki T, Makino K. Complete reversal of coenzyme specificity of xylitol dehydrogenase and increase of thermostability by the introduction of structural zinc. J Biol Chem. 2005;280(11):10340\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQiu S, Xu SY, Li SF, Meng KM, Cheng F, Wang YJ, Zheng YG. Fluorescence-based screening for engineered aldo‐keto reductase KmAKR with improved catalytic performance and extended substrate scope. Biotechnol J. 2021;16(9):2100130.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOuyang P, Wang H, Hajnal I, Wu Q, Guo Y, Chen GQ. Increasing oxygen availability for improving poly (3-hydroxybutyrate) production by \u003cem\u003eHalomonas\u003c/em\u003e. Metab Eng. 2018;45:20\u0026ndash;31.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu D, Ke X, Hu ZC, Zheng YG. Combinational expression of D-sorbitol dehydrogenase and pyrroloquinoline quinone increases 6-(\u003cem\u003eN\u003c/em\u003e-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose production by Gluconobacter oxydans through cofactor manipulation. Enzym Microb Technol. 2020;141:109670.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJia DX, Liu D, Zheng YG. Heterologous expression, purification and enzymatic characterization of succinate-semialdehyde dehydrogenase GabD4. Food Ferment Industries. 2016;42(5):97\u0026ndash;101.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKe X, Yu PH, Hu ZC, Chen L, Sun XQ, Zheng YG. Synergistic improvement of PQQ-dependent D-sorbitol dehydrogenase activity from \u003cem\u003eGluconobacter oxydans\u003c/em\u003e for the biosynthesis of miglitol precursor 6-(. J Biotechnol. 2019;300:55\u0026ndash;62. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\u003c/span\u003e\u003cspan address=\"http://-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu D, Ke X, Hu ZC, Zheng YG. Improvement of pyrroloquinoline quinone-dependent d-sorbitol dehydrogenase activity from \u003cem\u003eGluconobacter oxydans\u003c/em\u003e via expression of Vitreoscilla hemoglobin and regulation of dissolved oxygen tension for the biosynthesis of 6-(. J Biosci Bioeng. 2021;131(5):518\u0026ndash;24. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\u003c/span\u003e\u003cspan address=\"http://-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu D, Hu ZC, Ke X, Zheng YG. Breeding of \u003cem\u003eGluconobacter oxydans\u003c/em\u003e with high PQQ-dependent D-sorbitol dehydrogenase for improvement of 6-(\u003cem\u003eN\u003c/em\u003e\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\u003c/span\u003e\u003cspan address=\"http://-hydroxyethyl)-amino-6-deoxy-α-L-sorbofuranose\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e production. \u003cem\u003eBiochemical Engineering Journal\u003c/em\u003e 2020, 161.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai X, Shi X, Liu SQ, Qiang Y, Shen JD, Zhang B, Liu ZQ, Zheng YG. Hot spot-based engineering of ketopantoate hydroxymethyltransferase for the improvement of D-pantothenic acid production in \u003cem\u003eEscherichia coli\u003c/em\u003e. J Biotechnol. 2023;364:40\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu HT, Weng CY, Xu S-, Li SF, Wang YJ, Zheng YG. Directed evolution of a carbonyl reductase LsCR for the enantioselective synthesis of (1S)-2-chloro-1-(3, 4-difluorophenyl) ethanol. Bioorg Chem. 2022;127:105991.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBendl J, Stourac J, Sebestova E, Vavra O, Musil M, Brezovsky J, Damborsky J. HotSpot Wizard 2.0: automated design of site-specific mutations and smart libraries in protein engineering. Nucleic Acids Res. 2016;44(W1):W479\u0026ndash;87.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCheng F, Zhai QY, Gao XF, Liu HT, Qiu S, Wang YJ, Zheng YG. Tuning enzymatic properties by protein engineering toward catalytic tetrad of carbonyl reductase. Biotechnol Bioeng. 2021;118(12):4643\u0026ndash;54.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFilling C, Berndt KD, Benach J, Knapp S, Prozorovski T, Nordling E, Ladenstein R, J\u0026ouml;rnvall H, Oppermann U. Critical residues for structure and catalysis in short-chain dehydrogenases/reductases. J Biol Chem. 2002;277(28):25677\u0026ndash;84.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Q, Lu XF, Zhang Y, Tang XL, Zheng RC, Zheng YG. Development of a robust nitrilase by fragment swapping and semi-rational design for efficient biosynthesis of pregabalin precursor. Biotechnol Bioeng. 2020;117(2):318\u0026ndash;29.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrajewski V, Simić P, Mouncey NJ, Bringer S, Sahm H, Bott M. Metabolic Engineering of \u003cem\u003eGluconobacter oxydans\u003c/em\u003e for improved growth rate and growth yield on glucose by elimination of gluconate formation. Appl Environ Microbiol. 2010;76(13):4369.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang L, Li Y, Wang Z, Xia Y, Chen W, Tang K. Recent developments and future prospects of \u003cem\u003eVitreoscilla\u003c/em\u003e hemoglobin application in metabolic engineering. Biotechnol Adv. 2007;25(2):123\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-biotechnology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bbit","sideBox":"Learn more about [BMC Biotechnology](http://bmcbiotechnol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bbit/default.aspx","title":"BMC Biotechnology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Xylitol, Gluconobacter oxydans, Xylitol dehydrogenase, Site-saturation mutagenesis, Coenzyme regeneration system","lastPublishedDoi":"10.21203/rs.3.rs-8820590/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8820590/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eXylitol, a widely used sugar substitute in food and medicine, can be produced through microbial bioconversion using glucose as the primary substrate. In this process, a critical factor limiting xylitol production is the relatively low activity of xylitol dehydrogenase (XDH) during the biotransformation of D-arabitol to xylitol by resting cells of \u003cem\u003eGluconobacter oxydans\u003c/em\u003e.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eTo improve the catalytic performance, \u003cem\u003eGo\u003c/em\u003eXDH was engineered by site-saturation mutagenesis combined with a high-throughput screening method, and a variant \u003cem\u003eGo\u003c/em\u003eXDH\u003csub\u003eM3\u003c/sub\u003e (S77C/S106N/A110N) with high activity was obtained, showing a 2.49-fold increase in catalytic activity. The structural analysis revealed that the \u0026zwnj;S77C/S16N/A110N\u0026zwnj; mutations enhanced proton and electron transfer rates while stabilizing the hydrophilic substrate-binding pocket and the tetrameric structure. Additionally, by optimizing the coenzyme regeneration system and enhancing the oxygen transfer efficiency, we developed an efficient biotransformation of D-arabitol to xylitol in \u003cem\u003eG. oxydans\u003c/em\u003e. Using resting cells of \u003cem\u003eG. oxydans\u003c/em\u003e/XDH\u003csub\u003eM3\u003c/sub\u003e-GDH-VHb, a xylitol titer of 29.02 g/L were achieved from 40 g/L D-arabitol within 30 h.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e \u003cp\u003eThe findings suggest that boosting XDH activity through semi-rational engineering markedly improves xylitol productivity in \u003cem\u003eG. oxydans\u003c/em\u003e.\u003c/p\u003e","manuscriptTitle":"Improving xylitol production by semi-rational engineering of xylitol dehydrogenase and optimizing cofactor regeneration in Gluconobacter oxydans","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-12 11:11:46","doi":"10.21203/rs.3.rs-8820590/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2026-05-11T04:54:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T18:04:35+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-04-26T02:23:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"78861955135337271214569348462797458716","date":"2026-04-26T01:35:50+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"21265595779068801640258567280473834143","date":"2026-04-13T14:58:32+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-03-13T09:01:29+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"178670957358461459399826660747857736703","date":"2026-03-12T01:19:16+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"267367263605623976431543231963980842613","date":"2026-03-09T13:35:47+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-03-09T13:31:14+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-02-16T03:24:54+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-02-15T15:22:52+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Biotechnology","date":"2026-02-15T15:18:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-biotechnology","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"bbit","sideBox":"Learn more about [BMC Biotechnology](http://bmcbiotechnol.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/bbit/default.aspx","title":"BMC Biotechnology","twitterHandle":"BMC_series","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"09f69b02-31b2-4648-81c3-5ae32e5c389e","owner":[],"postedDate":"March 12th, 2026","published":true,"recentEditorialEvents":[{"type":"decision","content":"Revision requested","date":"2026-05-11T04:54:58+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2026-05-04T18:04:35+00:00","index":78,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"in-revision","subjectAreas":[],"tags":[],"updatedAt":"2026-05-11T05:10:02+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-12 11:11:46","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8820590","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8820590","identity":"rs-8820590","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}