Sequence of reactions which follows enzymatic oxidation of propargylglycine.

The nonenzymatic reactions which follow enzymatic oxidation of the gamma-delta acetylenic amino acid propargylglycine (2-amino-4-pentynoate) have been studied. The product which accumulates in solution has been identified as 2-amino-4-hydroxy-2,4-pentadienoate gamma-lactone, formed by intramolecular attack of the carboxylate anion on the electrophilic fourth carbon of 2-iminium-3,4-pentadienoate. This previously unknown substance was characterized by its reactions in acid and base and by its nuclear magnetic resonance spectrum. The lactone is preceded in the pathway by 2-amino-2-penten-4-ynoate, a transient electron-rich species which binds tightly to D-amino-acid oxidase and induces a charge-transfer complex with the electron-deficient bound flavin coenzyme. The aminediene lactone is converted by base treatment to 2-amino-4-keto-2-pentenoate, which is also a strong inhibitor of D-amino-acid oxidase and induces a charge-transfer complex.     

Vinylglycine and proparglyglycine: complementary suicide substrates for L-amino acid oxidase and D-amino acid oxidase.

Proparglyglycine (2-amino-4-pentynoate) and vinylglycine (2-amino-3-butenoate) have been examined as substrates and possible inactivators of two flavo enzymes, D-amino acid oxidase from pig kidney and L-amino acid oxidase from Crotalus adamanteus venom. Vinylglycine is rapidly oxidized by both enzymes but only L-amino acid oxidase is inactivated under assay conditions. The loss of activity probably involves covalent modification of an active site residue rather than the flavin adenine dinucleotide coenzyme and occurs once every 20000 turnovers. We have confirmed the recent observation (Horiike, K, Hishina, Y., Miyake, Y., and Yamano, T. (1975) J, Biochem. (Tokyo), 78, 57) that D-proparglglycine is oxidized with a time-dependent loss of activity by D-amino acid oxidase and have examined some mechanistic aspects of this inactivation, The extent of residual oxidase activity, insensitive to further inactivation, is about 2%, at which point 1.7 labels/subunit have been introduced with propargly[2-14C]glycine as substrate. L-Proparglyclycine is a substrate but not an inactivator of L-amino acid oxidase and the product ahat accumulats in the nonnucleophilic N-2-hydroxyethylpiperazine-N'-2-ethanesulfonic acid buffer is acetopyruvate. In the presence of butylamine HCl, a species with lambdaman 317 nm (epsilon = 15 000) accumulates that may be a conjugated eneamine adduct. The same species accumulates from D-amino acid oxidase oxidation of D-propargylglycine prior to inactivation; the inactivated apo D-amino acid oxidase has a new peak at 317 nm that is probably a similar eneamine. A likely inactivating species is 2-keto-3,4-pentadienoate arising from facile rearrangement of the expected initial product 2-keto 4 pentynoate. Vinylglycine and proparglyglycine show inactivation specificity, then, for L-and D-amino acid oxidase, respectively.     

Expression and stereochemical and isotope effect studies of active 4-oxalocrotonate decarboxylase

4-Oxalocrotonate decarboxylase (4-OD) and vinylpyruvate hydratase (VPH) from Pseudomonas putida mt-2 form a complex that converts 2-oxo-3-hexenedioate to 2-oxo-4-hydroxypentanoate in the catechol meta fission pathway. To facilitate mechanistic and structural studies of the complex, the two enzymes have been coexpressed and the complex has been purified to homogeneity. In addition, Glu-106, a potential catalytic residue in VPH, has been changed to glutamine, and the resulting E106QVPH mutant has been coexpressed with 4-OD and purified to homogeneity. The 4-OD/E106QVPH complex retains full decarboxylase activity, with comparable kinetic parameters to those observed for 4-OD in the wild-type complex, but is devoid of any detectable hydratase activity. Decarboxylation of (5S)-2-oxo-3-[5-D]hexenedioate by either the 4-OD/VPH complex or the mutant complex generates 2-hydroxy-2,4E-[5-D]pentadienoate in D(2)O. Ketonization of 2-hydroxy-2,4-pentadienoate by the wild-type complex is highly stereoselective and results in the formation of 2-oxo-(3S)-[3-D]-4-pentenoate, while the mutant complex generates a racemic mixture. These results indicate that 2-hydroxy-2, 4-pentadienoate is the product of 4-OD and that 2-oxo-4-pentenoate results from a VPH-catalyzed process. On this basis, the previously proposed hypothesis for the conversion of 2-oxo-3-hexenedioate to 2-oxo-4-hydroxypentanoate has been revised [Lian, H., and Whitman, C. P. (1994) J. Am. Chem. Soc. 116, 10403-10411]. Finally, the observed (13)C kinetic isotope effect on the decarboxylation of 2-oxo-3-hexenedioate by the 4-OD/VPH complex suggests that the decarboxylation step is nearly rate-limiting. Because the value is not sensitive to either magnesium or manganese, it is likely that the transition state for carbon-carbon bond cleavage is late and that the metal positions the substrate and polarizes the carbonyl group, analogous to its role in oxalacetate decarboxylase.     

Growth of Synthrophomonas wolfei on unsaturated short chain fatty acids

The anaerobic fatty acid-degrading syntrophic bacterium, Syntrophomonas wolfei, was grown in pure culture with either trans-2-pentenoate, trans-2-hexenoate, trans-3-hexenoate, and trans, trans-2,4-hexadienoate as the substrate. Trans-2-pentenoate was fermented to acetate, propionate, butyrate, and valerate. Acetate, butyrate, and hexanoate were produced from the six-carbon mono- and di-unsaturated acids. Propionate was also product from the trans,trans-2,4-hexadienoate which suggested this compound was degraded by another pathway in addition to -oxidation. The transient production of trans-2-hexenoate from trans-3-hexenoate suggested that the position of the double bound shifted from carbon-3 to carbon-2 prior to -oxidation. The specific growth rate decreased with increasing carbon length and degree of unsaturation. Molar growth yields ranged from 8.4 to 17.5 mg (dry wt.) per mmol and suggested that energy was conserved not only from substrate-level phosphorylation, but also from the reduction of unsaturated substrate.     

Kinetic and stereochemical analysis of YwhB, a 4-oxalocrotonate tautomerase homologue in Bacillus subtilis: mechanistic implications for the YwhB- and 4-oxalocrotonate tautomerase-catalyzed reactions

YwhB, a 4-oxalocrotonate tautomerase (4-OT) homologue in Bacillus subtilis, has no known biological role, and the gene has no apparent genomic context. The kinetic and stereochemical properties of YwhB have been examined using available enol and dienol compounds. The kinetic analysis shows that YwhB has a relatively nonspecific 1,3- and 1,5-keto-enol tautomerase activity, with the former activity prevailing. Replacement of Pro-1 or Arg-11 with an alanine significantly reduces or abolishes these activities, implicating both residues as critical ones for the activities. In D2O, ketonization of two monoacid substrates (2-hydroxy-2,4-pentadienoate and phenylenolpyruvate) produces a mixture of stereoisomers {2-keto-3-[2H]-4-pentenoate and 3-[2H]-phenylpyruvate}, where the (3R)-isomers predominate. Ketonization of 2-hydroxy-2,4-hexadienedioate, a diacid, in D2O affords mostly the opposite enantiomer, (3S)-2-oxo-[3-2H]-4-hexenedioate. The mono- and diacids apparently bind in different orientations in the active site of YwhB, but the highly stereoselective nature of the YwhB reaction using a diacid suggests that the biological substrate for YwhB may be a diacid. Moreover, of the three dienols examined, 1,3- and 1,5-keto-enol tautomerization reactions are only observed for 2-hydroxy-2,4-hexadienedioate, indicating that the C-3 and C-5 positions are accessible for protonation in this compound. Incubation of 4-OT with 2-hydroxy-2,4-hexadienedioate in D2O results in a racemic mixture of 2-oxo-[3-2H]-4-hexenedioate, suggesting that 4-OT may not catalyze a 1,3-keto-enol tautomerization reaction using this dienol. It has previously been shown that 4-OT catalyzes the near stereospecific conversion of 2-hydroxy-2,4-hexadienedioate to (5S)-[5-2H]-2-oxo-3-hexenedioate in D2O. Taken together, these observations suggest that 4-OT might function as a 1,5-keto-enol tautomerase using 2-hydroxy-2,4-hexadienedioate.     

Interaction between 1,4-thiazine derivatives and D-amino-acid oxidase

Aminoethylcysteine-ketimine (2H-1,4-thiazine-5,6-dihydro-3-carboxylic acid) strongly inhibits D-amino-acid oxidase (D-amino-acid:oxygen oxidoreductase (deaminating), EC 1.4.3.3). The inhibition is purely competitive (Ki = 3.3 X 10(-7) M). Aminoethylcysteine-ketimine modifies the visible spectrum of the enzyme: the absorption maxima of bound FAD shift from 375-455 nm to 385-445 nm with a definite shoulder at 465 nm; the appearance of a large absorption band centered at 750 nm may be due to a charge-transfer complex formation. The dissociation constant for the aminoethylcysteine-ketimine-enzyme complex, calculated by a photometric procedure (4 X 10(-7) M), is in good agreement with kinetic data. The dicarboxylic analogue of this inhibitor (lanthionine-ketimine) is ineffective in D-amino-acid oxidase inhibition and does not produce any spectral modification of the enzyme. These results confirm structural requirements for D-amino-acid oxidase inhibitor reported by other researchers. Ketimine reduced forms (thiomorpholine-2-carboxylic acid and thiomorpholine-2,6-dicarboxylic acid) are chemically synthesized and checked as D-amino-acid oxidase substrates: only thiomorpholine-2-carboxylic acid is oxidized to aminoethylcysteine-ketimine (Km = 2 X 10(-4) M).     

Properties of 5-hydroxyvalerate CoA-transferase from Clostridium aminovalericum

5-Hydroxyvalerate CoA-transferase from Clostridium aminovalericum, strain T2-7, was purified approximately 100-fold to homogeneity. The molecular mass of the native enzyme was determined by three different methods to be 178 +/- 11 kDa; that of the subunit was 47 kDa, indicating a homotetrameric structure. The following CoA esters acted as substrates (decreasing specificity, V/Km): 5-hydroxyvaleryl-CoA greater than propionyl-CoA greater than acetyl-CoA greater than (Z)-5-hydroxy-2-pentenoyl-CoA greater than butyryl-CoA greater than valeryl-CoA. 4-Pentenoate and 3-pentenoate were also activated by acetyl-CoA to the corresponding CoA esters, whereas crotonate, (E)-5-hydroxy-2-pentenoate, (E)-2-pentenoate and 2,4-pentadienoate were not attacked. 5-Hydroxyvalerate CoA-transferase showed ping-pong kinetics and was inactivated by sodium boranate only in the presence of a CoA substrate. This indicated the formation of a thiolester between a specific carboxyl group of the enzyme and CoASH during the course of the reaction. The CoA-transferase was inhibited by ATP and CTP, slightly by ADP, GTP and UTP, but not by AMP. The inhibition by ATP was competitive towards CoA esters and noncompetitive towards acetate.     

Metabolism of 4-pentenoic acid and inhibition of thiolase by metabolites of 4-pentenoic acid

The metabolism of 4-pentenoic acid, a hypoglycemic agent and inhibitor of fatty acid oxidation, has been studied in rat heart mitochondria. Confirmed was the conversion of 4-pentenoic acid to 2,4-pentadienoyl coenzyme A (CoA), which either is directly degraded via beta-oxidation or is first reduced in a NADPH-dependent reaction before it is further degraded by beta-oxidation. At pH 6.9, the NADPH-dependent reduction of 2,4-pentadienoyl-CoA proceeds 10 times faster than its degradation by beta-oxidation. At pH 7.8, this ratio is only 2 to 1. The direct beta-oxidation of 2,4-pentadienoyl-CoA leads to the formation of 3-keto-4-pentenoyl-CoA, which is highly reactive and spontaneously converts to another 3-ketoacyl-CoA derivative (compound X). 3-Keto-4-pentenoyl-CoA is a poor substrate of 3-ketoacyl-CoA thiolase (EC 2.3..1.16) whereas compound X is not measurably acted upon by this enzyme. The effects of several metabolites of 4-pentenoic acid on the activity of 3-ketoacyl-CoA thiolase were studied. 3,4-Pentadienoyl-CoA is a weak inhibitor of this enzyme that is protected against the inhibition by acetoacetyl-CoA. The most effective inhibitor of 3-ketoacyl-CoA thiolase was found to be 3-keto-4-pentenoyl-CoA, which inhibits the enzyme in both a reversible and irreversible manner. The reversible inhibition is possibly a consequence of the inhibitor being a poor substrate of 3-ketoacyl-CoA thiolase. It is concluded that 4-pentenoic acid is metabolized in mitochondria by two pathways. The minor yields 3-keto-4-pentenoyl-CoA, which acts both as a reversible and as a irreversible inhibitor of 3-ketoacyl-CoA thiolase and consequently of fatty acid oxidation.     

REGIONAL CHANGES IN CEREBRAL GABA CONCENTRATION AND CONVULSIONS PRODUCED BY d AND BY l-ALLYLGLYCINE

Abstract— The convulsant action of allylglycine (2-amino-4-pentenoic acid) is due to the metabolic conversion of allylglycine to 2-keto-4-pentenoic acid, a more potent glutamic acid decarboxylase inhibitor and more potent convulsant than the parent compound. We report regional changes in cerebral GABA concentration in rats after administration of d - and l -allylglycine. d -Allylglycine (3.75 mmol/kg) induced convulsions in 95–115 min, characterised by repeated clonic limb movements and rapid rotation around the head to tail axis. GABA concentrations were only reduced in cerebellum and ponsmedulla during the pre and post-convulsive periods. The localised reduction of GABA concentration is consistent with the enzymic conversion of d -allylglycine to 2-keto-4-pentenoic acid catalysed by cerebral d -amino acid oxidase, an enzyme known to be localised to the hind brain and spinal cord. l -allylglycine (1.2mmol/kg i.p.) induced convulsions in 65 -90 min, characterised by violent running followed by tonic flexion and extension. During the pre-convulsive period, GABA concentrations were reduced in all brain areas studied except the globus pallidus and ventral midbrain. The widespread decreases in GABA concentration suggest that the enzyme(s) which catalyse the conversion of l -allylglycine to 2-keto-4-pentenoic acid are widely distributed within the brain.     

A simple and rapid method to screen for mutant mice lacking D-amino-acid oxidase activity

A simple and rapid method to screen for mutant mice (Mus musculus) lacking D-amino-acid oxidase activity has been devised. Mice were given water containing small amounts (0.02%) of either D-methionine or D-phenylalanine. Urinary levels of the D-amino acid were examined using thin-layer chromatography. Some mice excreted substantial amounts of the D-amino acid through the urine. None of them had detectable D-amino-acid oxidase activity.     

Regional differences for the D-amino acid oxidase-catalysed oxidation of D-methionine in chicken small intestine.

1. Enzymic oxidation of D-[1-14C]methionine (D-met) to 2-keto-4-methylthiobutanoate (KMB) has been determined using 100,000 g supernatants prepared from chicken tissue homogenates. 2. The small intestinal mucosa contains substantial oxidative activity towards D-met, which represents about one-half and one-tenth the hepatic and renal activity, respectively. 3. KMB is poorly decarboxylated and rather transaminated to L-met. 4. The specific activity for D-met oxidation in the duodenal mucosa is 1.5- and 4.0-fold than in the jejunal and ileal mucosa, respectively. 5. The intestinal D-met-oxidizing activity is dramatically altered by the D-amino acid oxidase specific inhibitor benzoate.     

Chloroperoxidase-catalyzed cyclodimerization of methyl (2E)-2,4-pentadienoate: a [4+2] cycloaddition product

The chloroperoxidase (CPO)-catalyzed oxidation of the methyl (2E)-2,4-pentadienoate gives the terminal double bond epoxide (25%) and a cyclodimerization compound (63%) as the major products.     

Identification of methionine-110 as the residue covalently modified in the electrophilic inactivation of D-amino-acid oxidase by O-(2,4-dinitrophenyl) hydroxylamine

The reaction of O-(2,4-dinitrophenyl)hydroxylamine with D-amino-acid oxidase leads to complete inactivation which can be protected against by the competitive inhibitor benzoate [D'Silva, C., Williams, C. H., Jr., & Massey, V. (1986) Biochemistry 25, 5602-5608]. The residue modified has been identified as methionine-110. Differential high-performance liquid chromatography mapping of tryptic digests of D-amino-acid oxidase modified in the absence and presence of benzoate allows the isolation of a single methionine-containing tryptic peptide corresponding to residues 100-115 and referred to as T6-T7. In unmodified enzyme, the bond involving Arg-108 is readily cleaved and T6 and T7 are isolated. Brief treatment of peptide T6-T7 with carboxypeptidase Y released residues 112-115, and the residual peptide was isolated in good yield. Further treatment of this peptide (residues 100-111) with carboxypeptidase Y released Val and an unknown amino acid that comigrated with synthetically prepared S-aminomethionine sulfonium salt. The unknown compound and S-aminomethionine break down to methionine on treatment with dithiothreitol.     

Enzymatic determination of several D-amino acids using luminol-mediated chemiluminescence

A method for the quantitative determination of several D-amino acids in the range of 0.05-1 nmol per assay (0.25-5 microM) is described. It is insensitive to the presence of excesses of the respective L-amino acids. The assay system employs D-amino-acid oxidase (hog kidney), peroxidase (horse radish) and luminol; the total photon output elicited by the oxidation of the D-amino acids is determined. The different reactivity of individual D-amino acids with D-amino-acid oxidase limits the applicability of the assay. Indications for the usefulness of immobilized enzymes in D-amino-acid analysers are also given.     

Covalent and Noncovalent Dimers of the Cyanide-Resistant Alternative Oxidase Protein in Higher Plant Mitochondria and Their Relationship to Enzyme Activity

Evidence for a mixed population of covalently and noncovalently associated dimers of the cyanide-resistant alternative oxidase protein in plant mitochondria is presented. High molecular mass (oxidized) species of the alternative oxidase protein, having masses predicted for homodimers, appeared on immunoblots when the sulfhydryl reductant, dithiothreitol (DTT), was omitted from sodium dodecyl sulfate-polyacrylamide gel sample buffer. These oxidized species were observed in mitochondria from soybean (Glycine max [L.] Merr. cv Ransom), Sauromatum guttatum Schott, and mung bean (Vigna radiata [L.] R. Wilcz). Reduced species of the alternative oxidase were also present in the same mitochondrial samples. The reduced and oxidized species in isolated soybean cotyledon mitochondria could be interconverted by incubation with the sulfhydryl reagents DTT and azodicarboxylic acid bis(dimethylamide) (diamide). Treatment with chemical cross-linkers resulted in cross-linking of the reduced species, indicating a noncovalent dimeric association among the reduced alternative oxidase molecules. Alternative pathway activity of soybean mitochondria increased following reduction of the alternative oxidase protein with DTT and decreased following oxidation with diamide, indicating that electron flow through the alternative pathway is sensitive to the sulfhydryl/disulfide redox poise. In mitochondria from S. guttatum floral appendix tissue, the proportion of the reduced species increased as development progressed through thermogenesis.     

Presence of D-amino-acid oxidase protein in mutant mice lacking D-amino-acid oxidase activity.

1. Mutant mice lacking D-amino-acid oxidase activity were examined as to whether they possessed the enzyme protein. 2. Immunoblotting using an antibody against hog kidney D-amino-acid oxidase showed that kidney homogenates of the mutant mice as well as that of the normal mice had proteins reactive to the antibody. 3. Peroxisomal proteins of the kidney cells of the mutant mice were not different from those of the normal mice. 4. The peroxisomes of the mutant mice possessed a protein reactive to the antibody in the immunoblotting whose size was the same as the D-amino-acid oxidase protein present in the peroxisomes of the normal mice. 5. These results suggest that the mutant mice synthesize the D-amino-acid oxidase protein and integrate it into peroxisomes, though it is a nonfunctional enzyme.     

Dehalogenation and Deamination of l-2-Amino-4-chloro-4-pentenoic Acid by Proteus mirabilis

Eighty-one strains of bacteria were tested for their ability to catalyze the release of chloride ion from Dl-2-amino-4-chloro-4-pentenoic acid. A dehalogenating enzyme was obtained from the cells of Proteus mirabilis IFO 3849, which can use the l-isomer. The enzyme was constitutively produced. The conversion of l-2-amino-4-chloro-4-pentenoic acid to 2-keto-4-pentenoic acid, ammonia, and chloride ion was demonstrated. The reaction product, 2-keto-4-pentenoic acid, was isolated as its 2,4-dinitrophenylhydrazone and identified by catalytic hydrogenolysis of the hydrazone to the corresponding amino acid, norvaline.     

Stereospecificity in meta-fission catabolic pathways

The stereospecificities of 2-keto-4-pentenoate hydratase and 4-hydroxy-2-ketovalerate aldolase were studied in Escherichia coli, Pseudomonas putida, and Acinetobacter sp. Hydration was stereospecific in all three; however, only P. putida and Acinetobacter sp. showed stereospecificity in their aldolase reactions.     

Biotransformation of benzothiophene by isopropylbenzene-degrading bacteria.

Isopropylbenzene-degrading bacteria, including Pseudomonas putida RE204, transform benzothiophene to a mixture of compounds. Induced strain RE204 and a number of its Tn5 mutant derivatives were used to accumulate these compounds and their precursors from benzothiophene. These metabolites were subsequently identified by 1H and 13C nuclear magnetic resonance spectroscopy and gas chromatography-mass spectrometry. When strain RE204 was incubated with benzothiophene, it produced a bright yellow compound, identified as trans-4-[3-hydroxy-2-thienyl]-2-oxobut-3-enoate, formed by the rearrangement of cis-4-(3-keto-2,3-dihydrothienyl)-2-hydroxybuta-2,4-dieno ate, the product of 3-isopropylcatechol-2,3-dioxygenase-catalyzed ring cleavage of 4,5-dihydroxybenzothiophene, as well as 2-mercaptophenylglyoxalate and 2'-mercaptomandelaldehyde. A dihydrodiol dehydrogenase-deficient mutant, strain RE213, converted benzothiophene to cis-4,5-dihydroxy-4,5-dihydrobenzothiophene and 2'-mercaptomandelaldehyde; neither trans-4-[3-hydroxy-2-thienyl]-2-oxobut-3-enoate nor 2-mercaptophenylglyoxalate was detected. Cell extracts of strain RE204 catalyzed the conversion of cis-4,5-dihydroxy-4,5-dihydrobenzothiophene to trans-4-[3-hydroxy-2-thienyl]-2-oxobut-3-enoate in the presence of NAD+. Under the same conditions, extracts of the 3-isopropylcatechol-2,3-dioxygenase-deficient mutant RE215 acted on cis-4,5-dihydroxy-4,5-dihydrobenzothiophene, forming 4,5-dihydroxybenzothiophene. These data indicate that oxidation of benzothiophene by strain RE204 is initiated at either ring. Transformation initiated at the 4,5 position on the benzene ring proceeds by three enzyme-catalyzed reactions through ring cleavage. The sequence of events that occurs following attack at the 2,3 position of the thiophene ring is less clear, but it is proposed that 2,3 dioxygenation yields a product that is both a cis-dihydrodiol and a thiohemiacetal, which as a result of this structure undergoes two competing reactions: either spontaneous opening of the ring, yielding 2'-mercaptomandelaldehyde, or oxidation by the dihydrodiol dehydrogenase to another thiohemiacetal, 2-hydroxy-3-oxo-2,3-dihydrobenzothiophene, which is not a substrate for the ring cleavage dioxygenase but which spontaneously opens to form 2-mercaptophenylglyoxaldehyde and subsequently 2-mercaptophenylglyoxalate. The yellow product, trans-4-[3-hydroxy-2-thienyl]-2-oxobut-3-enoate, is a structural analog of trans-o-hydroxybenzylidenepyruvate, an intermediate of the naphthalene catabolic pathway; extracts of recombinant bacteria containing trans-o-hydroxybenzylidenepyruvate hydratase-aldolase catalyzed the conversion of trans-4-[3-hydroxy-2-thienyl]-2-oxobut-3-enoate to 3-hydroxythiophene-2-carboxaldehyde, which could then be further acted on, in the presence of NAD+, by extracts of recombinant bacteria containing the subsequent enzyme of the naphthalene pathway, salicylaldehyde dehydrogenase.     

L-fucose metabolism in mammals. Kinetic studies on pork liver 2-keto-3-deoxy-L-fuconate:NAD+ oxidoreductase.

Pork liver 2-keto-3-deoxy-L-fuconate:NAD+ oxidoreductase has been shown to convert 2-keto-3-deoxy-L-fuconate to a 6-carbon acid tentatively identified as 2,4(or 5)-diketo-5(or 4)-monohydroxyhexanoate. The enzyme has a pH optimum of 10. 5 or higher. It is stabilized by dithiothereitol and inhibited by p-hydroxymercuribenzoate and heavy metals (Ag+, Hg2+, Co2+, Cd2+, Pb2+, Zn2+, and Cu2+), suggesting the presence of a functionally essential sulfhydryl group; pre-treatment of enzyme with NAD+ prevents inhibition by p-hydrocymercuribenzoate and heavy metals indicating that this sulfhydryl group may be near the NAD+ binding site. The enzyme has an absolute requirement for NAD+; NADP+ is an ineffective coenzyme. Several lines of evidence indicate that the same enzyme acts on both 2-keto-3-deocy-L-fuconate and 2-keto-3-deoxy-D-arabonate; thus, the pure enzyme acts on both substrates, the two substrates have very similar kinetic parameters (Km values are: 2-keto-3-deocy-L-fuconate, 0.20 mM; 2-keto-3-deoxy-D-arabonate, 0.25 mM; NAD+ for either substrate, 0.22 to 0.25 mM), the two substrates show identical pH and temperature profiles and the two substrates compete for common enzyme active sites. A large number of other sugars and sugar acids, including several 2-keto-3-deoxyaldonates, were ineffective as substrates. The dehydrogenase was also found in calf, beef, lamb, mouse, and rat liver. These studies when considered together with previous studies on the metabolism of L-fucose in pork liver indicate the presence of a soluble enzyme pathway capable of converting L-fucose to 2,4(or 5)-diketo-5(or 4)-monohydroxyhexanoate; this pathway can also convert D-arabinose, and probably L-galactose, to the analogous derivatives (diketomonohydroxypentanoate and diketodihydroxyhexanoate, respectively.