VAAM-Jahrestagung 2011 Karlsruhe, 3.–6. April 2011

hyperthermophilic D-arabitol dehydrogenase from Thermotoga maritima,which was heterologously purified from both Escherichia coli andGluconobacter oxydans. The protein was first purified by Strep-Tactinaffinity chromatography and exclusively catalyzed NAD-dependentoxidation of D-arabitol or D-xylitol and the NADH-dependent reduction ofD-ribulose and D-xylulose. Increased catalytic rates were observed uponMg 2+ or K + addition. However, the tagged protein was not thermostable andwas rapidly inactivated at 85 o C. In contrast to the tagged protein, the D-arabitol dehydrogenase was thermostable when purified by heat precipitationand ion exchange chromatography without an affinity tag. The tag-less formof D-arabitol dehydrogenase had similar kinetic parameters compared to thetagged enzyme, demonstrating that the Strep-tag was not deleterious toprotein function but decreased protein stability. A single band at 27.6 kDawas observed on SDS-PAGE for the tag-less D-arabitol dehydrogenase andnative PAGE revealed that the protein formed homohexamer andhomododecamer. The enzyme catalyzed the oxidation of D-arabitol to D-ribulose and therefore belongs to the class of D-arabitol-2-dehydrogenases,which are typically observed in yeast not bacteria. The product D-ribulose isa rare ketopentose that has numerous industrial applications. Given itsthermostability and specificity, D-arabitol 2-dehydrogenase is a desirablebiocatalyst for the production of rare sugar precursors.GWV005Cofactor regeneration: understanding the catalyticproperties of the NAD + -reducing [NiFe]-hydrogenasefrom Ralstonia eutropha by investigating its subcomplexesL. Lauterbach* 1 , Z. Idris 2 , J. Liu 2 , K.A. Vincent 2 , O. Lenz 11 Institute of Biology/Microbiology, Humboldt-University, Berlin, Germany2 Department of Chemistry, University of Oxford, Oxford, United KingdomSeveral industrially relevant enzyme-driven syntheses, such as theproduction of L-tert leucine, require the reductant NADH, which needs to becontinuously regenerated during catalysis. Formate dehydrogenase iscurrently the only enzymatic system used for NADH regeneration atindustrial scale. However, its activity is limited, and formate leads to asignificant decrease in pH [1]. The NAD + -reducing soluble hydrogenase(SH) from Ralstonia eutropha H16 represents a promising alternative as theSH reduces NAD + to NADH at the expense of H 2 in the presence of O 2 [2-3].The SH is composed of the six subunits HoxHYFUI 2 which accommodatethe H 2-cycling catalytic Ni-Fe center, two flavin mononucleotides (FMN)and an electron relay made of iron-sulfur clusters [2]. The functional roleand interplay of these cofactors in the context of O 2 tolerance is so farunknown [4], and we have therefore investigated separately the hydrogenasemodule HoxHY and the diaphorase module HoxFU.In vitro assays and direct electrochemical studies show that as-isolatedHoxHY is catalytically inactive, but, after reductive activation at lowpotentials, exhibits both H 2 oxidation and H + reduction activities. Opticalspectroscopy revealed the presence of FMN at substoichiometric levels,which is consistent with an increase of H 2-oxidizing activity in the presenceof supplemental FMN [5].Quantification of the metal and FMN content of the HoxFU module incombination with UV/Vis spectroscopy revealed one [2Fe2S] cluster, oneFMN and a series of [4Fe4S] clusters per HoxFU, which is consistent withits close relationship to Complex I. The Michaelis constants for thesubstrates NADH, NADPH and NAD + were 56 μM, 6.78 mM and 197μM,respectively. Protein film electrochemistry revealed that NADH oxidation isproduct-inhibited by NAD + with K I = 0.3±0.2 mM [6].For HoxHY and HoxFU the overpotential relative to E(2H + /H 2) andE(NAD + /NADH) is minimal consistent with the role of the SH inbidirectional catalysis [5-6]. The results are discussed in terms of possiblecontrol mechanisms for the direction of catalysis and implications foraerobic cofactor regeneration.[1] Tishkov V.I. et al (1999): Biotechnol. Bioeng. 64, p. 187-193.[2] Burgdorf , T. et al (2005): J. Mol. Microbiol. Biotechnol. 10, p. 181-196[3] Schneider K. and Schlegel H.G. (1976) Biochim. Biophys. Acta 452, p.66-80.[4] Horch M. et al (2010): Angew. Chem. Int. Ed. 49, p. 8026-8029[5] Lauterbach, L.et al: Inorg. Chem. accepted.[6] Lauterbach, L. et al. submitted.GWV006Design-based construction of a lysine hyper-producingstrain by Systems Metabolic EngineeringJ. Becker* 1 , H. Schröder 2 , O. Zelder 2 , S. Haefner 2 , A. Herold 2 ,C. Klopprogge 2 , C. Wittmann 11 Institute of Biochemical Engineering, University of Technology,Braunschweig, Germany2 Research Fine Chemicals & Biotechnology, Ludwigshafen, GermanySince its discovery in the 1950´s Corynebacterium glutamicum has beenintensely exploited for amino acid production [1]. The annual productionvolume of L-lysine nowadays exceeds 1 million tons thus forming a largesector in biotechnological food and feed industry. From early on, theindustrial demand for this amino acid strongly stimulated the creation ofefficient production strains, including development of progressivetechniques that allow strain optimization. Still, the production propertiesachieved today are significantly below the theoretical capacity predicted [2].With the advent of recombinant DNA technology, a targeted geneticoptimization of C. glutamicum became possible which, in combination withsystems-level omics platforms, allows a precise and targeted optimization -systems metabolic engineering.Here we describe the stepwise construction of a genetically defined lysinehyper-producing strain. Characteristic metabolic flux patterns observedexperimentally and predicted by in silico simulation, revealed thatconcerted, systems-wide pathway engineering is required towards optimalperformance. The desired flux re-direction was achieved by only 12genome-based modifications located in the central metabolism and thelysine biosynthetic route. The finally obtained production strain achieved aremarkable carbon conversion yield of 55 %, a lysine HCl titre of 120 g L -1 ,and a productivity of 4 g L -1 h -1 . With this production performance the lysinehyper-producer created in this work is the best wild type based productionstrain so far described and lies at the maximum performance of classicallyderived strains.[1] Wittmann, C. and J. Becker (2007): The L-lysine story: From metabolic pathways to industrialproduction. In: Wendisch, V. F., (Ed.), Amino acid biosynthesis - Pathways, regulation and metabolicengineering. vol. 5. Springer Berlin/ Heidelberg pp. 39-70.[2] Wittmann, C. (2010): Analysis and Engineering of Metabolic Pathway Fluxes in Corynebacteriumglutamicum. Adv Biochem Eng Biotechnol. 120, 21 - 49.GWV007The gene rosA encoding N,N-8-amino-8-demethyl-Driboflavindimethyltransferase is located within a genecluster possibly involved in biosynthesis of roseoflavin inStreptomyces davawensisF. Jankowitsch*, M. MackInstitute of Technical Microbiology, University of Applied SciencesMannheim, GermanyStreptomyces davawensis was first isolated from a Philippine soil sample ina screening program for antibiotic-producing organisms. In the stationarygrowth phase this strain produces the antibiotic roseoflavin (RoF) (8-dimethylamino-8-demethyl-D-riboflavin) one of the few known naturalriboflavin-analogs [1,2]. It was postulated that RoF is synthesized fromriboflavin via 8-amino- (AF) and 8-methylamino-8-demethyl-D-riboflavin(MAF) [3,4]. Using a cell-free extract derived from stationary-phase S.davawensis cells, an S-adenosyl methionine dependent conversion of AFinto MAF and RoF was observed. The corresponding N,N-8-amino-8-demethyl-D-riboflavin dimethyltransferase activity was enriched by columnchromatography (3 steps). The final most active fraction still contained atleast 5 different proteins. With the major protein present in the final activefraction an enzymatic digest was performed and peptides were used for denovo sequencing by MS/MS. The obtained peptide sequence was comparedto a protein database derived from the genomic sequence of S. davawensis.The sequence matched a hypothetical protein derived from a yetuncharacterized open reading frame (ORF7678) located in the middle of a(putative) gene cluster within the S. davawensis genome. Overexpression ofORF7678 in Escherichia coli revealed that the corresponding gene producthad N,N-8-amino-8-demethyl-D-riboflavin dimethyltransferase activity. TheORF was named rosA, being the first gene of the RoF biosynthetic pathwayto be identified. RosA was purified from a recombinant E. coli strain andwas kinetically characterized (apparent K m for AF 69,46 μM; V max = 0,63μmol/min mg protein; k cat = 24,38 min -1 ). The putative S. davawensis genecluster containing rosA was overexpressed in Streptomyces lividans andStreptomyces albus, which naturally do not synthesize roseoflavin. Theresulting recombinant strains were found to not produce RoF in thespektrum | Tagungsband 2011