Characterization of a thermophilic l-arabinose isomerase from Thermoanaerobacterium saccharolyticum NTOU1

l-Arabinose isomerase (EC 5.3.1.4, l-AI) mainly catalyzes the reversible aldose–ketose isomerization between l-arabinose and l-ribulose. l-AIs can also catalyze other reactions, such as the conversion of d-galactose to d-tagatose. In this study, the araA gene encoding l-AI was PCR-cloned from Thermoanaerobacterium saccharolyticum NTOU1 and then expressed in Escherichia coli. The recombinant l-AI was purified from the cell-free extract using nickel nitrilotriacetic acid metal-affinity chromatography. The purified enzyme showed an optimal activity at 70 °C and pH 7–7.5. The enzyme was stable at pHs ranging from 6.5 to 9.5 and the activity was fully retained after 2 h incubation at 55–65 °C. The low concentrations of divalent metal ions, either 0.1 mM Mn²⁺ or 0.05 mM Co²⁺, could improve both catalytic activity and thermostability at higher temperatures. The recombinant T. saccharolyticum NTOU1 l-AI has the lowest demand for metal ions among all characterized thermophilic l-AIs. This thermophilic l-AI shows a potential to be used in industry to produce d-tagatose from d-galactose.     

Structural insights into conserved l-arabinose metabolic enzymes reveal the substrate binding site of a thermophilic l-arabinose isomerase

Structural genomics demonstrates that despite low levels of structural similarity of proteins comprising a metabolic pathway, their substrate binding regions are likely to be conserved. Herein based on the 3D-structures of the α/β-fold proteins involved in the ara operon, we attempted to predict the substrate binding residues of thermophilic Geobacillus stearothermophilusl-arabinose isomerase (GSAI) with no 3D-structure available. Comparison of the structures of l-arabinose catabolic enzymes revealed a conserved feature to form the substrate-binding modules, which can be extended to predict the substrate binding site of GSAI (i.e., D195, E261 and E333). Moreover, these data implicated that proteins in the l-arabinose metabolic pathway might retain their substrate binding niches as the modular structure through conserved molecular evolution even with totally different structural scaffolds.     

Selective reduction of xylose to xylitol from a mixture of hemicellulosic sugars

The biocatalytic reduction of d-xylose to xylitol requires separation of the substrate from l-arabinose, another major component of hemicellulosic hydrolysate. This step is necessitated by the innate promiscuity of xylose reductases, which can efficiently reduce l-arabinose to l-arabinitol, an unwanted byproduct. Unfortunately, due to the epimeric nature of d-xylose and l-arabinose, separation can be difficult, leading to high production costs. To overcome this issue, we engineered an E. coli strain to efficiently produce xylitol from d-xylose with minimal production of l-arabinitol byproduct. By combining this strain with a previously engineered xylose reductase mutant, we were able to eliminate l-arabinitol formation and produce xylitol to near 100% purity from an equiweight mixture of d-xylose, l-arabinose, and d-glucose.     

Immobilization of Bacillus licheniformis L-Arabinose Isomerase for Semi-Continuous L-Ribulose Production

Bacillus licheniformis L-arabinose isomerase (BLAI) with a broad pH range, high substrate specificity, and high catalytic efficiency for L-arabinose was immobilized on various supports. Eupergit C, activated-carboxymethylcellulose, CNBr-activated agarose, chitosan, and alginate were tested as supports, and Eupergit C was selected as the most effective. After determination of the optimum enzyme concentration, the effects of pH and temperature were investigated using a response surface methodology. The immobilized BLAI enzyme retained 86.4% of the activity of the free enzyme. The optimal pH for the immobilized BLAI was 8.0, and immobilization improved the optimal temperature from 50 °C (free enzyme) to a range between 55 and 65 °C. The half life improved from 2 at 50 °C to 212 h at 55 °C following immobilization. The immobilized BLAI was used for semi-continuous production of L-ribulose. After 8 batch cycles, 95.1% of the BLAI activity was retained. This simple immobilization procedure and the high stability of the final immobilized BLAI on Eupergit C provide a promising solution for large-scale production of L-ribulose from an inexpensive L-arabinose precursor.     

Regulation of the L-arabinose catabolic pathway in Aerobacter aerogenes

Three enzymes of the l-arabinose catabolic pathway in Aerobacter aerogenes, l-arabinose isomerase, l-ribulokinase, and l-ribulose-5-phosphate 4-epimerase, are specifically induced in the presence of l-arabinose. Mutants constitutive for kinase activity are also constitutive for the isomerase and 4-epimerase activities, suggesting that these three enzymes are coordinately controlled in A. aerogenes. l-Ribulokinase activity can still be induced in the presence of l-arabinose in an isomerase-deficient strain of A. aerogenes. Since l-arabinose is not converted to l-ribulose in such a strain, it appears that l-arabinose must be the inducer of l-ribulokinase, as well as the coordinately controlled isomerase and 4-epimerase. As in the metabolism of l-arabinose, growth of A. aerogenes on l-arabitol also requires a 4-epimerase for the conversion of l-ribulose-5-phosphate to d-xylulose-5-phosphate. However, loss of ability to metabolize l-arabinose, due to a deficiency in 4-epimerase synthesis in the presence of l-arabinose, does not affect growth on l-arabitol. In addition, synthesis of the 4-epimerase associated with l-arabitol metabolism is not accompanied by l-arabinose isomerase or l-ribulokinase synthesis. These results suggest either the existence of two different l-ribulose-5-phosphate 4-epimerases in A. aerogenes, or of two different regulatory mechanisms for the control of the same epimerase.     

Structure and Engineering of l-Arabinitol 4-Dehydrogenase from Neurospora crassa

l-Arabinitol 4-dehydrogenase (LAD) catalyzes the conversion of l-arabinitol into l-xylulose with concomitant NAD⁺ reduction. It is an essential enzyme in the development of recombinant organisms that convert l-arabinose into fuels and chemicals using the fungal l-arabinose catabolic pathway. Here we report the crystal structure of LAD from the filamentous fungus Neurospora crassa at 2.6 Å resolution. In addition, we created a number of site-directed variants of N. crassa LAD that are capable of utilizing NADP⁺ as cofactor, yielding the first example of LAD with an almost completely switched cofactor specificity. This work represents the first structural data on any LAD and provides a molecular basis for understanding the existing literature on the substrate specificity and cofactor specificity of this enzyme. The engineered LAD mutants with altered cofactor specificity should be useful for applications in industrial biotechnology.     

Utilization of degraded peach gum polysaccharide by Aspergillus flavus

d-Galactose, d-mannose, d-xylose, l-arabinose, and d-glucuronic acid and its γ-lactone were examined as carbon sources for the culture of Aspergillus flavus. d-Mannose was taken up the most rapidly and d-glucuronic acid and its γ-lactone the least rapidly. A partially degraded polysaccharide from peach tree gum (Prunus persica [L.] Batsch containing the above sugars together with d-glucuronic acid and its 4-O-methyl ether was used as substrate for another A. flavus culture. It was found that d-galactose was the major sugar passing into the culture medium with lower proportions of d-xylose, l-arabinose, 2-O-β-d-glucopyranuronosyl-d-mannose, and 6-O-β-d-glucopyranuronosyl-d-galactose. This indicates that the fungus produces extracellular exo- and endo-glycanohydrolases which may be useful in structural studies on polysaccharides.     

Soft-sensor assisted dynamic investigation of mixed feed bioprocesses

Recombinant mixed feed bioprocesses are characterized by the controlled feeding of multiple defined carbon sources aiming at increased productivities. However, mixed feed process design is challenging due to physiological constraints such as adaptation times and catabolite repression.A novel soft-sensor assisted dynamic method that allows the science-based process design with respect to co-utilization of primary and secondary substrate was developed. The method is based on the control of the specific uptake rates of primary and secondary substrate via a combination of a rate-based soft sensor and in-line infrared spectroscopy. Maximum secondary substrate specific uptake rates and adaptation times are determined by a combination of dynamic pulse and ramp experimentation.The power of the presented method was demonstrated on a recombinant Escherichia coli pBAD mixed feed process with d-glucose as primary and l-arabinose as secondary carbon source. Onset of catabolite repression was observed once a total specific substrate uptake rate of 1.0 g/gh was exceeded. Adaptation times to l-arabinose were determined as ～10 min.The presented method can be considered generically applicable for the physiological investigation of mixed feed systems. Furthermore, metabolic capabilities of the promising but yet unexplored E. coli pBAD mixed feed system were explored for the first time.     

The ‘true’ l-xylulose reductase of filamentous fungi identified in Aspergillus niger

l-Xylulose reductase is part of the eukaryotic pathway for l-arabinose catabolism. A previously identified l-xylulose reductase in Hypocrea jecorina turned out to be not the ‘true’ one since it was not upregulated during growth on l-arabinose and the deletion strain showed no reduced l-xylulose reductase activity but instead lost the d-mannitol dehydrogenase activity [17]. In this communication we identified the ‘true’ l-xylulose reductase in Aspergillus niger. The gene, lxrA (JGI177736), is upregulated on l-arabinose and the deletion results in a strain lacking the NADPH-specific l-xylulose reductase activity and having reduced growth on l-arabinose. The purified enzyme had a K_m for l-xylulose of 25 mM and a ν_max of 650 U/mg.     

Ethanol production from pentoses and sugar-cane bagasse hemicellulose hydrolysate by Mucor and Fusarium species

The fermentation of carbohydrates and hemicellulose hydrolysate by Mucor and Fusarium species has been investigated, with the following results. Both Mucor and Fusarium species are able to ferment various sugars and alditols, including d-glucose, pentoses and xylitol, to ethanol. Mucor is able to ferment sugar-cane bagasse hemicellulose hydrolysate to ethanol. Fusarium F5 is not able to ferment sugar-cane bagasse hemicellulose hydrolysate to ethanol. During fermentation of hemicellulose hydrolysates, d-glucose was utilized first, followed by d-xylose and l-arabinose. Small amounts of xylitol were produced by Mucor from d-xylose through oxidoreduction reactions, presumably mediated by the enzyme aldose reductase¹ (alditol: NADP⁺ 1-oxidoreductase, EC 1.1.1.21). For pentose fermentation, d-xylose was the preferred substrate. Only small amounts of ethanol were produced from l-arabinose and d-arabitol. No ethanol was produced from l-xylose, d-arabinose or l-arabitol.     

A novel alpha-ketoglutaric semialdehyde dehydrogenase: evolutionary insight into an alternative pathway of bacterial L-arabinose metabolism

Azospirillum brasilense possesses an alternative pathway of l-arabinose metabolism, which is different from the known bacterial and fungal pathways. In a previous paper (Watanabe, S., Kodaki, T., and Makino, K. (2006) J. Biol. Chem. 281, 2612-2623), we identified and characterized l-arabinose 1-dehydrogenase, which catalyzes the first reaction step in this pathway, and we cloned the corresponding gene. Here we focused on the fifth enzyme, alpha-ketoglutaric semialdehyde (alphaKGSA) dehydrogenase, catalyzing the conversion of alphaKGSA to alpha-ketoglutarate. alphaKGSA dehydrogenase was purified tentatively as a NAD(+)-preferring aldehyde dehydrogenase (ALDH) with high activity for glutaraldehyde. The gene encoding this enzyme was cloned and shown to be located on the genome of A. brasilense separately from a gene cluster containing the l-arabinose 1-dehydrogenase gene, in contrast with Burkholderia thailandensis in which both genes are located in the same gene cluster. Higher catalytic efficiency of ALDH was found with alphaKGSA and succinic semialdehyde among the tested aldehyde substrates. In zymogram staining analysis with the cell-free extract, a single active band was found at the same position as the purified enzyme. Furthermore, a disruptant of the gene did not grow on l-arabinose. These results indicated that this ALDH gene was the only gene of the NAD(+)-preferring alphaKGSA dehydrogenase in A. brasilense. In the phylogenetic tree of the ALDH family, alphaKGSA dehydrogenase from A. brasilense falls into the succinic semialdehyde dehydrogenase (SSALDH) subfamily. Several putative alphaKGSA dehydrogenases from other bacteria belong to a different ALDH subfamily from SSALDH, suggesting strongly that their substrate specificities for alphaKGSA are acquired independently during the evolutionary stage. This is the first evidence of unique "convergent evolution" in the ALDH family.     

The DeLey-Doudoroff Pathway of Galactose Metabolism in Azotobacter vinelandii

Azotobacter vinelandii cell extracts reduced NAD and oxidized d-galactose to galactonate that subsequently was converted to 2-keto-3-deoxy-galactonate. Further metabolism of 2-keto-3-deoxy-galactonate required the presence of ATP and resulted in the formation of pyruvate and glyceraldehyde 3-P. Radiorespirometry indicated a preferential release of CO(2) at the first carbon position of the d-galactose molecule. This suggested that Azotobacter vinelandii metabolizes d-galactose via the DeLey-Doudoroff pathway. The first enzyme of this pathway, d-galactose dehydrogenase, was partially characterized. It has a molecular weight of about 74,000 Da and an isoelectric point of 6.15. The pH optimum of the galactose dehydrogenase was about 9. The apparent K(m)s for NAD and d-galactose were 0.125 and 0.56 mM, respectively. Besides d-galactose, the active fraction of this galactose dehydrogenase also oxidized l-arabinose effectively. The electron acceptor for d-galactose or l-arabinose oxidation, NAD, could not be replaced by NADP. These substrate specificities were different from those reported in Pseudomonas saccharophila, Pseudomonas fluorescens, and Rhizobium meliloti.     

Engineering nonphosphorylative metabolism to synthesize mesaconate from lignocellulosic sugars in Escherichia coli

Dicarboxylic acids are attractive biosynthetic targets due to their broad applications and their challenging manufacturing process from fossil fuel feedstock. Mesaconate is a branched, unsaturated dicarboxylic acid that can be used as a co-monomer to produce hydrogels and fire-retardant materials. In this study, we engineered nonphosphorylative metabolism to produce mesaconate from d-xylose and l-arabinose. This nonphosphorylative metabolism is orthogonal to the intrinsic pentose metabolism in Escherichia coli and has fewer enzymatic steps and a higher theoretical yield to TCA cycle intermediates than the pentose phosphate pathway. Here mesaconate production was enabled from the d-xylose pathway and the l-arabinose pathway. To enhance the transportation of d-xylose and l-arabinose, pentose transporters were examined. We identified the pentose/proton symporter, AraE, as the most effective transporter for both d-xylose and l-arabinose in mesaconate production process. Further production optimization was achieved by operon screening and metabolic engineering. These efforts led to the engineered strains that produced 12.5 g/l and 13.2 g/l mesaconate after 48 h from 20 g/l of d-xylose and l-arabinose, respectively. Finally, the engineered strain overexpressing both l-arabinose and d-xylose operons produced 14.7 g/l mesaconate from a 1:1 d-xylose and l-arabinose mixture with a yield of 85% of the theoretical maximum. (0.87 g/g). This work demonstrates an effective system that converts pentoses into a value-added chemical, mesaconate, with promising titer, rate, and yield.     

L-arabinose binding protein from Escherichia coli B-r

A protein which is capable of binding l-arabinose-1-(14)C has been isolated from l-arabinose-induced cultures of Escherichia coli B/r. Analysis for this l-arabinose-binding protein (ABP) in a number of l-arabinose-negative mutants suggests that the ABP is not coded for by any of the known genetic units of the l-arabinose complex yet is under the control of the regulator gene araC. The ABP has been purified and found to bind l-arabinose, d-fucose, d-xylose, and l-ribulose with decreasing affinities. The K(m) for l-arabinose is 5.7 x 10(-6)m. The molecular weight, as determined by equilibrium centrifugation, was found to be 32,000. The protein was observed to have many features that liken it to other recently isolated binding proteins that have been implicated in the active transport of small molecules.     

Alginate immobilization of recombinant Escherichia coli whole cells harboring l-arabinose isomerase for l-ribulose production

Recombinant Escherichia coli whole cells harboring Bacillus licheniformis l-arabinose isomerase (BLAI) were immobilized with alginate. The operational conditions for immobilization were optimized with response surface methodology. Optimal alginate concentration, Ca²⁺ concentration, and cell mass loading were 1.8% (w/v), 0.1 M, and 44.5 g L⁻¹, respectively. The interactions between Ca²⁺ concentration, alginate concentration, and initial cell mass were significant. After immobilization of BLAI, cross-linking with 0.1% glutaraldehyde significantly reduced cell leakage. The half-life of immobilized whole cells was 150 days, which was 50-fold longer than that of free cells. In seven repeated batches for l-ribulose production, the productivity was as high as 56.7 g L⁻¹ h⁻¹ at 400 g L⁻¹ substrate concentration. The immobilized cells retained 89% of the initial yield after 33 days of reaction. Immobilization of whole cells harboring BLAI, therefore, makes a suitable biocatalyst for the production of l-ribulose, particularly because of its high stability and low cost.     

发酵乳杆菌来源l-阿拉伯糖异构酶理性设计及在d-塔格糖生产中的应用

L-阿拉伯糖异构酶(L-arabinose isomerase,L-AI)是D-半乳糖异构化生成D-塔格糖的关键酶。为提高L-阿拉伯糖异构酶对D-半乳糖的活性及在生物转化中的转化率,本研究对发酵乳杆菌(Lactobacillus fermentum)CGMCC2921来源的L-阿拉伯糖异构酶进行重组表达和生物转化应用,并对其底物结合口袋进行理性设计以提高酶对D-半乳糖亲和力和催化活性。结果显示,突变体F279I对D-半乳糖的转化率提高至野生型酶的1.4倍,进一步叠加获得的双突变体M185A/F279I的K_m和k_cat分别为530.8mmol/L与19.9s^-1,底物亲和力显著提高,催化效率提高至野生型酶的8.2倍。以400 g/L乳糖为底物时,突变酶M185A/F279I转化率高达22.8%。本研究在乳糖高值化生产塔格糖方面具有重要的应用价值。     

The Hypocrea jecorina (syn. Trichoderma reesei) lxr1 gene encodes a d-mannitol dehydrogenase and is not involved in l-arabinose catabolism

The Hypocrea jecorina LXR1 was described as the first fungal l-xylulose reductase responsible for NADPH dependent reduction of l-xylulose to xylitol in l-arabinose catabolism. Phylogenetic analysis now reveals that LXR1 forms a clade with fungal d-mannitol 2-dehydrogenases. Lxr1 and the orthologous Aspergillus nigermtdA are not induced by l-arabinose but expressed at low levels during growth on different carbon sources. Deletion of lxr1 does not affect growth on l-arabinose and l-xylulose reductase activity remains unaltered whereas d-mannitol 2-dehydrogenase activities are reduced. We conclude that LXR1 is a d-mannitol 2-dehydrogenase and that a true LXR1 is still awaiting discovery.     

Enhanced activity and stability of <Emphasis Type="SmallCaps">l</Emphasis>-arabinose isomerase by immobilization on aminopropyl glass

Immobilization of Bacillus licheniformis l-arabinose isomerase (BLAI) on aminopropyl glass modified with glutaraldehyde (4 mg protein g support⁻¹) was found to enhance the enzyme activity. The immobilization yield of BLAI was proportional to the quantity of amino groups on the surface of support. Reducing particle size increased the adsorption capacity (q _m) and affinity (k _a). The pH and temperature for immobilization were optimized to be pH 7.1 and 33°C using response surface methodology (RSM). The immobilized enzyme was characterized and compared to the free enzyme. There is no change in optimal pH and temperature before and after immobilization. However, the immobilized BLAI enzyme achieved 145% of the activity of the free enzyme. Correspondingly, the catalytic efficiency (k _cat/K _m) was improved 1.47-fold after immobilization compared to the free enzyme. The thermal stability was improved 138-fold (t _1/2 increased from 2 to 275 h) at 50°C following immobilization.     

Hydrazinolysis of monosaccharides: a two-step synthesis of chiral pentane-1,2,3-triols from pentoses and observations on the hydrazinolysis of glycoproteins and glycopeptides

1,2-Dideoxyalditols, the corresponding 1-alkenes, and 1-deoxyalditols are formed in various proportions from d-glucose, d-mannose, l-arabinose, and d-xylose by the action of refluxing hydrazine. Sequential hydrazinolysis, catalytic hydrogenation, and chromatography afford a route to 1,2-dideoxyalditols. For example, 1,2-dideoxy-l-erythro-pentitol is formed from l-arabinose in 42% yield, and d-xylose is a source of 1,2-dideoxy-d-threo-pentitol (50%). Under the conditions (anhydrous hydrazine at 100° for 30 h in the absence of air) used by Montreuil for the hydrazinolysis of glycoproteins and glycopeptides, no 1,2-dideoxyalditol was formed; degradation was incomplete, there being some aldose hydrazone present. Under Kochetkov's hydrazinolysis conditions (105° for 10 h with hydrazinium sulphate), less degradation occurred and the product from d-galactose was identified as 1-deoxy-d-tagatose hydrazone.     

Codon-optimized bacterial genes improve L-Arabinose fermentation in recombinant Saccharomyces cerevisiae

Bioethanol produced by microbial fermentations of plant biomass hydrolysates consisting of hexose and pentose mixtures is an excellent alternative to fossil transportation fuels. However, the yeast Saccharomyces cerevisiae, commonly used in bioethanol production, can utilize pentose sugars like l-arabinose or d-xylose only after heterologous expression of corresponding metabolic pathways from other organisms. Here we report the improvement of a bacterial l-arabinose utilization pathway consisting of l-arabinose isomerase from Bacillus subtilis and l-ribulokinase and l-ribulose-5-P 4-epimerase from Escherichia coli after expression of the corresponding genes in S. cerevisiae. l-Arabinose isomerase from B. subtilis turned out to be the limiting step for growth on l-arabinose as the sole carbon source. The corresponding enzyme could be effectively replaced by the enzyme from Bacillus licheniformis, leading to a considerably decreased lag phase. Subsequently, the codon usage of all the genes involved in the l-arabinose pathway was adapted to that of the highly expressed genes encoding glycolytic enzymes in S. cerevisiae. Yeast transformants expressing the codon-optimized genes showed strongly improved l-arabinose conversion rates. With this rational approach, the ethanol production rate from l-arabinose could be increased more than 2.5-fold from 0.014 g ethanol h(-1) (g dry weight)(-1) to 0.036 g ethanol h(-1) (g dry weight)(-1) and the ethanol yield could be increased from 0.24 g ethanol (g consumed l-arabinose)(-1) to 0.39 g ethanol (g consumed l-arabinose)(-1). These improvements make up a new starting point for the construction of more-efficient industrial l-arabinose-fermenting yeast strains by evolutionary engineering.