Solution structure of 5-keto-D-fructose: relevance to the specificity of hexose kinases

5-Keto-D-Fructose, a Natural Diketone and Potential Sugar Substitute, Significantly Reduces the Viability of Prokaryotic and Eukaryotic Cells

5-Keto-D-fructose (5-KF) is a natural diketone occurring in micromolar concentrations in honey, white wine, and vinegar. The oxidation of D-fructose to 5-KF is catalyzed by the membrane-bound fructose dehydrogenase complex found in several acetic acid bacteria. Since 5-KF has a sweetening power comparable to fructose and is presumably calorie-free, there is great interest in making the diketone commercially available as a new sugar substitute. Based on a genetically modified variant of the acetic acid bacterium Gluconobacter oxydans 621H, an efficient process for the microbial production of 5-KF was recently developed. However, data on the toxicology of the compound are completely lacking to date. Therefore, this study aimed to investigate the effect of 5-KF on the viability of prokaryotic and eukaryotic cells. It was found that the compound significantly inhibited the growth of the gram-positive and gram-negative model organisms Bacillus subtilis and Escherichia coli in a concentration-dependent manner. Furthermore, cell viability assays confirmed severe cytotoxicity of 5-KF toward the colon cancer cell line HT-29. Since these effects already occurred at concentrations of 5 mM, the use of 5-KF in the food sector should be avoided. The studies performed revealed that in the presence of amines, 5-KF promoted a strong Maillard reaction. The inherent reactivity of 5-KF as well as the Maillard products formed could be the trigger for the observed inhibition of prokaryotic and eukaryotic cells.

The 5-Ketofructose Reductase of Gluconobacter sp. Strain CHM43 Is a Novel Class in the Shikimate Dehydrogenase Family

Gluconobacter sp. strain CHM43 oxidizes mannitol to fructose and then oxidizes fructose to 5-keto-d-fructose (5KF) in the periplasmic space. Since NADPH-dependent 5KF reductase was found in the soluble fraction of Gluconobacter spp., 5KF might be transported into the cytoplasm and metabolized. Here, we identified the GLF_2050 gene as the kfr gene encoding 5KF reductase (KFR). A mutant strain devoid of the kfr gene showed lower KFR activity and no 5KF consumption. The crystal structure revealed that KFR is similar to NADP⁺-dependent shikimate dehydrogenase (SDH), which catalyzes the reversible NADP⁺-dependent oxidation of shikimate to 3-dehydroshikimate. We found that several amino acid residues in the putative substrate-binding site of KFR were different from those of SDH. Phylogenetic analyses revealed that only a subclass in the SDH family containing KFR conserved such a unique substrate-binding site. We constructed KFR derivatives with amino acid substitutions, including replacement of Asn21 in the substrate-binding site with Ser that is found in SDH. The KFR-N21S derivative showed a strong increase in the K_m value for 5KF but a higher shikimate oxidation activity than wild-type KFR, suggesting that Asn21 is important for 5KF binding. In addition, the conserved catalytic dyad Lys72 and Asp108 were individually substituted for Asn. The K72N and D108N derivatives showed only negligible activities without a dramatic change in the K_m value for 5KF, suggesting a catalytic mechanism similar to that of SDH. With these data taken together, we suggest that KFR is a new member of the SDH family. IMPORTANCE A limited number of species of acetic acid bacteria, such as Gluconobacter sp. strain CHM43, produce 5-ketofructose, a potential low-calorie sweetener, at a high yield. Here, we show that an NADPH-dependent 5-ketofructose reductase (KFR) is involved in 5-ketofructose degradation, and we characterize this enzyme with respect to its structure, phylogeny, and function. The crystal structure of KFR was similar to that of shikimate dehydrogenase, which is functionally crucial in the shikimate pathway in bacteria and plants. Phylogenetic analysis suggested that KFR is positioned in a small subgroup of the shikimate dehydrogenase family. Catalytically important amino acid residues were also conserved, and their relevance was experimentally validated. Thus, we propose KFR as a new member of shikimate dehydrogenase family.

Production of the potential sweetener 5-ketofructose from fructose in fed-batch cultivation with Gluconobacter oxydans

Sweeteners improve the dietary properties of many foods. A candidate for a new natural sweetener is 5-ketofructose. In this study a fed-batch process for the production of 5-ketofructose was developed. A Gluconobacter oxydans strain overexpressing a fructose dehydrogenase from G. japonicus was used and the sensory properties of 5-ketofructose were analyzed. The compound showed an identical sweet taste quality as fructose and a similar intrinsic sweet threshold concentration of 16.4 mmol/L. The production of 5-ketofructose was characterized online by monitoring of the respiration activity in shake flasks. Pulsed and continuous fructose feeding was realized in 2 L stirred tank reactors and maximum fructose consumption rates were determined. 5-Ketofructose concentrations of up to 489 g/L, product yields up to 0.98 g_5-KF/g_fructose and space time yields up to 8.2 g/L/h were reached highlighting the potential of the presented process.

Production of 5-ketofructose from fructose or sucrose using genetically modified Gluconobacter oxydans strains

The growing consumer demand for low-calorie, sugar-free foodstuff motivated us to search for alternative non-nutritive sweeteners. A promising sweet-tasting compound is 5-keto-D-fructose (5-KF), which is formed by membrane-bound fructose dehydrogenases (Fdh) in some Gluconobacter strains. The plasmid-based expression of the fdh genes in Gluconobacter (G.) oxydans resulted in a much higher Fdh activity in comparison to the native host G. japonicus. Growth experiments with G. oxydans fdh in fructose-containing media indicated that 5-KF was rapidly formed with a conversion efficiency of 90%. 5-KF production from fructose was also observed using resting cells with a yield of about 100%. In addition, a new approach was tested for the production of the sweetener 5-KF by using sucrose as a substrate. To this end, a two-strain system composed of the fdh-expressing strain and a G. oxydans strain that produced the sucrose hydrolyzing SacC was developed. The strains were co-cultured in sucrose medium and converted 92.5% of the available fructose units into 5-KF. The glucose moiety of sucrose was converted to 2-ketogluconate and acetate. With regard to the development of a sustainable and resource-saving process for the production of 5-KF, sugar beet extract was used as substrate for the two-strain system. Fructose as product from sucrose cleavage was mainly oxidized to 5-KF which was detected in a concentration of over 200 mM at the end of the fermentation process. In summary, the two-strain system was able to convert fructose units of sugar beet extract to 5-KF with an efficiency of 82 ± 5%.

Rates of spontaneous cleavage of glucose, fructose, sucrose, and trehalose in water, and the catalytic proficiencies of invertase and trehalas

The half-lives for spontaneous hydrolysis of trehalose and sucrose at 25 °C are 6.6 × 10⁶ years and 440 years. The half-lives for decomposition of the hydrolysis products glucose and fructose are 96 years and 70 days, respectively. Whereas sucrose and trehalose differ by a factor of 15000 in their rates of uncatalyzed hydrolysis, the reactions catalyzed by invertase (EC 3.2.1.26) and trehalase (EC 3.2.1.28) proceed at similar rates. Accordingly, the attainments of invertase as a catalyst are modest, but the rate enhancement and catalytic proficiency produced by trehalase approach the high levels achieved by polysaccharide hydrolases.

Carbohydrate substrate specificity of bacterial and plant pyrophosphate-dependent phosphofructokinases

Pyrophosphate-dependent phosphofructokinase from the facultative anaerobic bacterium Propionibacterium freudenreichii and from the mung bean Phaseolus aureus has been purified to homogeneity. Potential utilization of carbohydrate substrate analogues for each enzyme was initially screened by using Fourier transform 31P NMR at pH 8 and 25 degrees C and monitoring the appearance of the phosphate resonance in the direction of D-fructose 6-phosphate phosphorylation (forward reaction direction) and, with the bisphosphate analogues, the appearance of the pyrophosphate resonance in the direction of phosphate phosphorylation (reverse reaction direction). Both enzymes are strict in their requirements for the sugar phosphate substrate, with only D-fructose 6-phosphate, D-sedoheptulose 7-phosphate, and 2,5-anhydro-D-mannitol 6-phosphate, or their respective bisphosphates in the reverse reaction direction, utilized as substrates at detectable levels. The dissociation constants for D-psicose 6-phosphate, D-tagatose 6-phosphate, and L-sorbose 6-phosphate are an order of magnitude larger than that for D-fructose 6-phosphate, indicating a stringent steric requirement for the D-threo (trans) configuration at the two nonanomeric furan ring hydroxyl groups. These results strongly suggest that the anomeric, epimeric, and tautomeric form of the sugar phosphate substrates favored by both enzymes is the beta-D-fructofuranose form. Dissociation constants for nonsubstrate analogues were used to provide information on the nature of the active site. Competitive inhibition patterns vs. fructose 1,6-bisphosphate were obtained for a series of 1,n-alkanediol bisphosphates (where n = 2-9).(ABSTRACT TRUNCATED AT 250 WORDS)

Structures of D-threo-2,5-hexodiulose 1-phosphate and D-threo-2,5-hexodiulose 1,6-bisphosphate (5-keto-D-fructose mono- and bis-phosphate) in solution by 13C-N.M.R. spectroscopy

The mono- (2) and bis-phosphate (3) derivatives of D-threo-2,5-hexodiulose (1) (5-keto-D-fructose) were synthesized enzymically and purified by anion-exchange chromatography. The proportions, sizes of ring, and anomeric configurations were determined by F.t. 31P- and 13C-n.m.r. spectroscopy. Compound 2 was found to exist preponderantly (70-78%) in the beta-pyranose form with the remainder existing in the 2R,5R-furanose form. Compound 3 assumes two different furanose forms in solution, one (77-84%) being the 2R,5R-furanose form and the other the 2S,5R-furanose form.