Saccharomyces cerevisiae maltases use maltose, maltulose, turanose and maltotriose as substrates, isomaltases use isomaltose, α‐methylglucoside and palatinose and both use sucrose. These enzymes are hypothesized to have evolved from a promiscuous α‐glucosidase ancMALS through duplication and mutation of the genes. We studied substrate specificity of the maltase protein MAL1 from an earlier diverged yeast, Ogataea polymorpha (Op), in the light of this hypothesis. MAL1 has extended substrate specificity and its properties are strikingly similar to those of resurrected ancMALS. Moreover, amino acids considered to determine selective substrate binding are highly conserved between Op MAL1 and ancMALS. Op MAL1 represents an α‐glucosidase in which both maltase and isomaltase activities are well optimized in a single enzyme. Substitution of Thr200 (corresponds to Val216 in S. cerevisiae isomaltase IMA1) with Val in MAL1 drastically reduced the hydrolysis of maltose‐like substrates (α‐1,4‐glucosides), confirming the requirement of Thr at the respective position for this function. Differential scanning fluorimetry (DSF) of the catalytically inactive mutant Asp199Ala of MAL1 in the presence of its substrates and selected monosaccharides suggested that the substrate‐binding pocket of MAL1 has three subsites (–1, +1 and +2) and that binding is strongest at the –1 subsite. The DSF assay results were in good accordance with affinity (K m) and inhibition (K i) data of the enzyme for tested substrates, indicating the power of the method to predict substrate binding. Deletion of either the maltase (MAL1) or α‐glucoside permease (MAL2) gene in Op abolished the growth of yeast on MAL1 substrates, confirming the requirement of both proteins for usage of these sugars. © 2016 The Authors. Yeast published by John Wiley & Sons, Ltd.

Bacteroides thetaiotaomicron, an abundant commensal of the human gut, degrades numerous complex carbohydrates. Recently, it was reported to grow on a β-2,6-linked polyfructan levan produced by Zymomonas mobilis degrading the polymer into fructooligosaccharides (FOS) with a cell surface bound endo-levanase BT1760. The FOS are consumed by B. thetaiotaomicron, but also by other gut bacteria, including health-promoting bifidobacteria and lactobacilli. Here we characterize biochemical properties of BT1760, including the activity of BT1760 on six bacterial levans synthesized by the levansucrase Lsc3 of Pseudomonas syringae pv. tomato, its mutant Asp300Asn, levansucrases of Zymomonas mobilis, Erwinia herbicola, Halomonas smyrnensis as well as on levan isolated from timothy grass. For the first time a plant levan is shown as a perfect substrate for an endo-fructanase of a human gut bacterium. BT1760 degraded levans to FOS with degree of polymerization from 2 to 13. At optimal reaction conditions up to 1 g of FOS were produced per 1 mg of BT1760 protein. Low molecular weight (<60 kDa) levans, including timothy grass levan and levan synthesized from sucrose by the Lsc3Asp300Asn, were degraded most rapidly whilst levan produced by Lsc3 from raffinose least rapidly. BT1760 catalyzed finely at human body temperature (37°C) and in moderately acidic environment (pH 5–6) that is typical for the gut lumen. According to differential scanning fluorimetry, the Tm of the endo-levanase was 51.5°C. All tested levans were sufficiently stable in acidic conditions (pH 2.0) simulating the gastric environment. Therefore, levans of both bacterial and plant origin may serve as a prebiotic fiber for B. thetaiotaomicron and contribute to short-chain fatty acids synthesis by gut microbiota. In the genome of Bacteroides xylanisolvens of human origin a putative levan degradation locus was disclosed.

Genome of an early-diverged yeast Blastobotrys (Arxula) adeninivorans (Ba) encodes 88 glycoside hydrolases (GHs) including two α-glucosidases of GH13 family. One of those, the rna_ARAD1D20130g-encoded protein (BaAG2; 581 aa) was overexpressed in Escherichia coli, purified and characterized. We showed that maltose, other maltose-like substrates (maltulose, turanose, maltotriose, melezitose, malto-oligosaccharides of DP 4‒7) and sucrose were hydrolyzed by BaAG2, whereas isomaltose and isomaltose-like substrates (palatinose, α-methylglucoside) were not, confirming that BaAG2 is a maltase. BaAG2 was competitively inhibited by a diabetes drug acarbose (Ki = 0.8 µM) and Tris (Ki = 70.5 µM). BaAG2 was competitively inhibited also by isomaltose-like sugars and a hydrolysis product—glucose. At high maltose concentrations, BaAG2 exhibited transglycosylating ability producing potentially prebiotic di- and trisaccharides. Atypically for yeast maltases, a low but clearly recordable exo-hydrolytic activity on amylose, amylopectin and glycogen was detected. Saccharomyces cerevisiae maltase MAL62, studied for comparison, had only minimal ability to hydrolyze these polymers, and its transglycosylating activity was about three times lower compared to BaAG2. Sequence identity of BaAG2 with other maltases was only moderate being the highest (51%) with the maltase MalT of Aspergillus oryzae.

Bacterial levansucrases polymerize fructose residues of sucrose to β-2,6 linked fructans-fructooligosaccharides (FOS) and levan. While β-2,1-linked FOS are widely recognized as prebiotics, the health-related effects of β-2,6 linked FOS are scarcely studied as they are not commercially available. Levansucrase Lsc3 (Lsc-3) of Pseudomonas syringae pv. tomato has very high catalytic activity and stability making it a promising biotechnological catalyst for FOS and levan synthesis. In this study we evaluate feasibility of several high-throughput methods for screening and preliminary characterization of levansucrases using 36 Lsc3 mutants as a test panel. Heterologously expressed and purified His-tagged levansucrase variants were studied for: (1) sucrose-splitting activity; (2) FOS production; (3) ability and kinetics of levan synthesis; (4) thermostability in a Thermofluor assay. Importantly, we show that sucrose-splitting activity as well as the ability to produce FOS can both be evaluated using permeabilized levansucrase-expressing E. coli transformants as catalysts. For the first time we demonstrate the key importance of Trp109, His113, Glu146 and Glu236 for the catalysis of Lsc3. Cost-effective and high-throughput methods presented here are applicable not only in the levansucrase assay, but have a potential to be adapted for high-throughput (automated) study of other enzymes.

Hansenula polymorpha uses maltase to grow on maltose and sucrose. Inspection of genomic clones of H. polymorpha showed that the maltase gene HPMAL1 is clustered with genes corresponding to Saccharomyces cerevisiae maltose permeases and MAL activator genes orthologues. We sequenced the H. polymorpha maltose permease gene HPMAL2 of the cluster. The protein (582 amino acids) deduced from the HPMAL2 gene is predicted to have eleven transmembrane domains and shows 39-57% identity with yeast maltose permeases. The identity of the protein is highest with maltose permeases of Debaryomyces hansenii and Candida albicans. Expression of the HPMAL2 in a S. cerevisiae maltose permease-negative mutant CMY1050 proved functionality of the permease protein encoded by the gene. HPMAL1 and HPMAL2 genes are divergently positioned similarly to maltase and maltose permease genes in many yeasts. A two-reporter assay of the expression from the HPMAL1-HPMAL2 intergenic region showed that expression of both genes is coordinately regulated, repressed by glucose, induced by maltose, and that basal expression is higher in the direction of the permease gene.