Lactate Dehydrogenase-Catalyzed Regeneration of NAD from NADH
Total Page:16
File Type:pdf, Size:1020Kb
tr()()R(ir.,, . ,r,,nrrsr ny 17. 400-409(1989) LactateDehydrogenase-Catalyzed Regeneration of NADfrom NADHfor Use in Enzyme-CatalyzedSynthesis H. Kr,trH CsENnulr AND Geoncs M. WSTTESTDES Department of Chemistrt', Huruard Uniuersity, Cambridge, Massachusetts 02138 ReceiuedOctober 24, 1988 Lactate dehydrogenase(LDH, rabbit muscle. EC 1.1.1.27),coupled with pyruvic and gfyoxylic acids as oxidants, regeneratesNAD in .situfor enzyme-catalyzed oxidations. The use of LDH has advantages of convenience, stability of the regenerating system, and econ- omy. Pyruvate and glyoxylate differ in their advantages and disadvantages as oxidants. Pyruvate analogs, monomethyl oxalate and S-methyl thiooxalate, were examined as possi- ble oxidants but were not substrates of lactate dehydrogenase. o rg8gAcademicpress, Inc. INTRODUCTION We herein describethe use of l-lactate dehydrogenase(LDH.r rabbit muscle, EC Ll.1.27). with pyruvic and gly'ory'licacids as oxidants. to regenerateNAD for enzyme-cataly'zed.preparative oxidations. NAD regenerationusing LDH hasad- vantagesof convenience.stability' of the regeneratingenzy'me and reagents,and econom)'.In sy'stemsnot sensitiveto deactivationby glyoxylate.the high redox potentialof glyoxylate is an advantage. Enzymes are useful as catalysts for organic synthesis (1), and methods for cofactor regenerationhave made synthetic applicationsof even cofactor-requiring enzvmes practical (2). Although severalgood to excellent systemsregenerate the reduced nicotinamide cofactors, NAD(P)H (2, 3), methods for regeneratingthe oxidized cofactors. NAD(P), are less developed and are currently being investi- gated (4-7). A frequent problem of enzyme-catalyzedoxidations is not the co- factor-regeneratingstep but the oxidation of substratein the synthetic step. Bio- chemical oxidations may suffer from unfavorable thermodynamics and noncompetitive and uncompetitive product inhibition (4, 8). LDH catalyzes the completely stereoselectivereduction of pyruvic acid (with concomitantoxidation of NADH) to r,-lacticarcid, Eq. il1, R : CHr e, l0). LDH also acceptsas substratesa variety of other a-oxo acids (l l, l2), including gly- oxylic acid,which is reducedin the presenceof NADH to glycolicacid. Eq. [], R ' Abbreviationsused: FMN, flavin mononucleotide; G6P. glucose 6-phosphate; G6PDH, glucose-6- phosphatedehydrogenase; GluDH, glutamate dehydrogenase; Gly-Gly, glycylglycine; Hepes, 4-(2- hydroxyethyl-l-piperazineethanesulfonic acid; HLADH, horse liver alcohol dehydrogenase; aKG, ammonium a-ketoglutarate; LDH, L-lactate dehydrogenase; Mops, 4-morpholinepropanesulfonic acid: TEA. triethanolamine. 0045-2068/89$3.00 Copyright O 1989by Academic Press, Inc. All rights of reproduction in any form reserved. LACTATE DEHYDROGENASE-CAT ALY7.ED REGENERATION 401 : H (1J-15).The utility of LDH in preparativereductions. its easeof immobiliza- tion and manipulation.and its stabilitywhen protectedfrom autoxidationhave been demc-rnstrated(12, I6\. NADH NAD 8\?'n\/ ul R^coo- LDH n^coo- R= cHs, H We have examined LDH as a catalyst for NAD regeneration.The two most commonly reported methods for regeneratingNAD(P) use ammonium a-ketoglu- tarate (aKG) and glutamatedehydrogenase (GluDH, EC L4.1.3) (4,8, 17) or FMN/Or. with (6) or without (4, 18,19) FMN reductase(EC 1.6.8.1) as a catalyst. LDH is lessexpensive and higher in specificactivity than both GIuDH and FMN reductuse.rUnlike FMNior. LDH pern-ritsNAI) regenerertionunder anaerobic conditions (many enzymesare deactivatedb1, dior,r-gen. superoxide. and perox- ide). Pyruvic and glyoxylic acids are less erpensive than a-ketoglutaricacid and FMN,3 and glyoxylate(Eo: -0.90 Va)is more stron-elyoxidizing than a-KG (E; - -0.l2l V for reductive amination). Potentialobstacles to the useof pyruvateas the oridant were the nonproductive binding of pyruvate to LDH (20) and the condensationsof pyruvate with itself and NAD (10). Both Mgt* and LDH catalyzethe enolizationof pyruvate and thus the nucleophilicaddition of enol pyruvate to C--1of NAD (21). Becauseglyoxylate cannotenolize. it shouldnot undergocondensations as pyruvatedoes. Also. since glyoxylateexists in aqueoussolution as thc hydrate.it shouldbe lesssoluble in organicsolvents than p1'ruvateand thus more easily'separatedfrom products. A potentialproblem with the use of glyo.rylate.hou'ever, was the NAD-linked oxidationof glyoxylateby LDH (14,22). At high pH (pH 9), LDH oxidizesthe hydrate of glyoxylate as an analog of lactate, Eq . l2l. This reaction nor only consumesNAD but also generatesoxalate, which is a noncompetitiveinhibitor of LDH with respectto pyruvate (and presumablythe oxo form of gly,orl'latet(2J). NAD NADH g -r4.* ?' !_-/ o L2l lrAcoo- *o^coo- LDH ,-,oAaoo- RESULTSAND DISCUSSION Our work began with a searchfor substratesof LDH other than pyruvate which would be good oxidizing agentsfor NAD regeneration.We measuredthe kinetic r Cost and specificactivity of LDH vs GIuDH are $0.71l1000U and 1000U/mg vs $4.7/1000U and 40 U/mg. respectively.FMN reductase(50 U/mg) requiresisolation (Ref. (9)). Prices(Sigma. 1988)are for enzymes as researchbiochemicals and thereforerepresent upper limits. One thousandunits (l U : I pmol/min) is approximately the amount of enzymatic activity required to generate1 mol of product/ day. 3 Prices of pyruvic (957o),glyoxylic (50% wlw aqueous solution), and a-ketoglutaric g5%) acids are 7.6, 3.5, and 15 $/mol, respectively (Aldrich and ICN Biochemicals, 1988). a Reduction potentialsare relative to the half-reaction2H* + 2 e --->H". for which E^: _0.42Y at pH7. 402 CHENAULT AND WHITESIDES .g E 200 E 100 0 06121824 V/[glyoxylate](mL min -1) Ftc. L Eadie-Hofstee plot used to determineK^(app) of LDH for glyoxylate in the absence(O) and presence (C) of I mn sodium vanadate. constants of LDH with glyoxylate (pH 7.0, 25'C) and found the apparent Michaelis constant,5K,,(app), to be 25 mr'a(Fig. 1). This value was in reasonable agreementwith previousdeterminations (13-15) and was significantlyhigher than thatfor pyruvate(K,,:0.15-0.33 mM) ( 13,15,25). The maximalvelocity (y,u*) for the reductionof glyoxylate was approximatelythe sameas that with pyruvate and was at least l0a times faster than the oxidation of gly'oxirlatc(ne detectedno oxidation with 50 mv glyoxylate and 2.8 mr'aNAD). At concentrationsabove -50 mu, glyoxylate inhibited its own reduction by LDH. Since glyoxylate exists in aqueous solution as the hydrate and since vanadate has been shown to catalyzethe hydration/dehydration of glyoxylate (25), we also measuredthe kinetic constants of LDH with glyoxylate in the presenceof I mrra vanadate(Fig. l). The K,,,(app)was reducedto 8 mv. but Vn,,,,also dropped by nearly half. In the presenceof l0 mHrvanadate. LDH lost all activity. We did not pursue the mechanism of this inhibition nor the further use of vanadate as a catalyst in NAD-regeneratingsystems. Monomethyl oxalate (1) and S-methyl thiooxalate (2) were tested as pyruvate analogs,but neither was a substrateof LDH (V < 0.005%of that for pyruvate). o o tl tl ^,, ^A^^^- \/n3\-./ UUU cH.s^coo- 1 2 ScsEuE I To measure the stability of the pyruvate (glyoxylate)/LDH system, we incu- bated 0.1 mrraNAD and 50 mM pyruvate (glyoxylate) with and without LDH. The activities of the speciesin solution were determinedperiodically by the enzymatic assay.In all tests, the organic acids and enzyme remainedfully active after 120h. 5 The K- measured is an apparent value because it is based on the total concentration of glyoxylate, regardless of the form(s) it exists in and the species bound by the enzyme. The actual K- is probably much smaller since glyoxylate exists in solution almost completely as the hydrate and yet LDH binds the free aldehydic species as the substrate for reduction. LACTATE DEHYDROGENAS E-C AT ALY ZED REGENERAT I ON 403 In the presenceof pyruvate, active NAD decreasedsomewhat: After 120h, in the presence and absence of LDH, 45 and 60Vo,respectively, of the original active NAD remained. With glyoxylate, in the presence and absence of LDH, 82 and 8970,respectively, of the original active NAD remained after 120 h. LDH, coupled with pyruvic and glyoxylic acids as oxidants, successfullyregen- erated NAD in oxidations of 4 mmol of glucose 6-phosphate (G6P) to 6-phos- phogluconatecatalyzedby glucose-6-phosphatedehydrogenase (G6PDH) Eq. t3l. The G6P/G6PDH system was chosenfor initial demonstrationsbecause the oxida- tion of G6P is thermodynamically favorable (spontaneoushydrolysis of 6-phos- phogluconolactonerenders it irreversible)and not subject to product inhibition by 6-phosphogluconate.6-Phosphogluconate was isolatedas its barium salt (26) in 70-80% yield. Total turnover numbers (TTN)6 (J) for NAD(H) were 930-1070.At the end of the reactions (4-6 h), both enzymes remained fully active, and the nicotinamide cofactor retainedl6-99% of its original activity. In separatestability studies, soluble G6PDH incubated in the presenceof 50 mM pyruvic or glyoxylic acid (25'C, ambient atmosphere)remained 86 and707o active, respectively, after 95 h. When 50 mr',rG6P was included in the incubation mixture, G6PDH lost no activity after 95 h. aoPot2- to,*\ .-: Qr- NAD n^coo- bu'oH 1 \/,/ tou /\I lrRDH/ \ 3 l3l n ^nn- n vvv \ \*,"Nft:"*,o To compare the present method of NAD regenerationto methods using aKG/ GIuDH and FMN/Or. we used pyruvate/LDH to regenerate NAD for the HLADH-catalyzed oxidation of c'is-1,2-bis(hydroxymethyl)cyclohexane(3) to (+)-(lR,65)-cis-8-oxabicyclo[4.3.0]nonan-7-one(4),Eq. [4] (Tablel) (4,27). Both PAN-immobilized (28) and membrane-enclosed (29) LDH regeneratedNAD