Search

Vitamin K functions in animal cells as the cofactor of the enzyme vitamin K-dependent carboxylase which catalyzes the post-translational formation of gamma-carboxyglutamyl (Gla) residues in specific vitamin K-dependent proteins. These proteins include four blood coagulation factors (prothrombin and Factors VII, IX and X), other plasma proteins (protein C, protein S and protein Z), two proteins from bone (osteocalcin or bone Gla-protein and matrix Gla-protein), and other proteins from lung, kidney, spleen, testis, placenta and other tissues. In the proteins involved in blood coagulation the Gla residues are mandatory for the activation of the inactive proenzymes; this process occurs on phospholipid surfaces to which the proenzymes are bound via Gla residues and calcium ions. The energy needed in the carboxylation reaction is obtained from the oxidation of vitamin K hydroquinone to 2,3-epoxide of the vitamin. Specific enzymes, vitamin K epoxide reductase and vitamin K quinone reductases, catalyze consecutive reactions in which the vitamin K hydroquinone is regenerated, thus allowing continued use of the vitamin K molecule for the carboxylations. The oral anticoagulants, derivatives of 4-hydroxycoumarin and indan-1,3-dione, used as therapeutic agents in thromboembolic disease, are antagonists to vitamin K preventing the catalytic use of vitamin K in the carboxylations by irreversibly inhibiting vitamin K epoxide reductase.

Background Pre-existing chronic renal failure is a significant risk factor for acute renal failure (ARF) after cardiac surgery. N-acetylcysteine (NAC) has been shown to prevent contrast media-induced ARF. Our objective was to evaluate whether i.v. NAC has renoprotective effects in patients with mild renal failure undergoing cardiac surgery. Methods In this prospective, randomized, double-blind study, 80 patients with mild to moderate renal failure undergoing elective heart surgery with cardiopulmonary bypass were recruited. All received either i.v. NAC (n=38) or placebo (n=39) at induction of anaesthesia and then up to 20 h. Urine N-acetyl-β-d-glucosaminidase (NAG) and urine creatinine ratio, plasma creatinine, and serum cystatin C levels indicated renal function. Results Levels of urinary NAG/creatinine ratio, plasma creatinine and serum cystatin C did not significantly differ between NAC and placebo groups during five postoperative days. Urine NAG/creatinine ratio increased over 30% in 100% of patients in the NAC group vs 92.3% in the placebo group (P=0.081). Plasma creatinine increased by 25% from baseline or over 44 μmol litre−1 in 42.1% in NAC group vs 48.7% in placebo group (P=0.560). Serum cystatin C exceeded 1.4 mg litre−1 in 78.9% in NAC group vs 61.5% in placebo group (P=0.096). Conclusions Prophylactic treatment with i.v. N-acetylcysteine had no renoprotective effect in patients with pre-existing renal failure undergoing cardiac surgery.

Creatine kinase and myoglobin are markers of muscular damage in rhabdomyolysis. Whereas myoglobin is considered to be the principal compound causing tubular damage, serum creatine kinase level is presently guiding therapeutic interventions in clinical practice to prevent acute renal failure. Because differences in elimination kinetics of these two compounds may influence therapeutic decisions, we studied elimination kinetics of myoglobin and creatine kinase in patients with rhabdomyolysis.Open, noncomparative study.Intensive and intermediary care units in a university hospital.A total of 13 consecutive patients with rhabdomyolysis whose baseline serum creatine kinase exceeded 5000 IU/L. Ten of 13 patients were treated with forced alkaline diuresis, and none were dialyzed.Myoglobin had faster elimination kinetics than creatine kinase (p.01), and the average times to reach the 50% level of initial values were 12 hrs for myoglobin and 42 hrs for creatine kinase. Elimination of myoglobin was not affected by glomerular filtration rate. Compared with creatinine clearance (mean, 102 mL/min), myoglobin clearance was low (mean, 3 mL/min), both in patients with preserved renal function (n = 11) and in those with acute renal failure (n = 2).Serum myoglobin has faster elimination kinetics than creatine kinase in patients treated with forced alkaline diuresis for rhabdomyolysis. Considering the etiologic role of myoglobin, our data suggest that serum myoglobin level, rather than that of creatine kinase, should be used to guide therapy in patients with rhabdomyolysis.

Mannose is an aldohexose component of a number of glycoproteins in cellular membranes and blood plasma. Free (unbound) mannose is a normal blood plasma constituent and its concentration is elevated in diabetes mellitus and chronic glomerulonephritis. We devised an enzymatic method for the determination of free mannose in which mannose is converted to glucose-6-phosphate and measured spectrophotometrically using glucose-6-phosphate dehydrogenase and nicotinamide adenine dinucleotide phosphate (NADP). Accumulation of reduced NADP in the assay was verified by spectral analysis and by finding rapid disappearance of absorbance at 340 nm on addition of glutathione reductase and oxidized glutathione into the reaction mixture. The method necessitates prior removal of glucose from the samples. This we accomplished using glucose-6-phosphate dehydrogenase and a surplus amount of NADP, followed by elimination of reduced NADP by acidification of the reaction mixture. The assays may be run in parallel for expediency. Concentration of free mannose in serum was 18.5 +/- 5.5 mumol/l in healthy fasting female adults. The analytical recovery was 90.2 +/- 10.2% and the between-run imprecision was 13.5% (18.5 +/- 5.5 mumol/l, mean +/- SD) and 10.4% (75.3 +/- 10.3 mumol/l). The assay showed rectilinearity up to 220 mumol/l, which covers the measuring range to which the mannose concentrations in normal and clinical samples may be expected to fall.

Formaldehyde is oxidized in animal cells into formate in two consecutive reactions catalyzed by the specific enzymes formaldehyde dehydrogenase (EC 1.2.1.1) and S-formylglutathione hydrolase (EC 3.1.2.12) (Uotila and Koivusalo, 1974a; 1974b). Formaldehyde reacts nonenzymically with glutathione (GSH) to form the adduct S-hydroxymethylglutathione. Formaldehyde dehydrogenase catalyzes the NAD-dependent oxidation of this adduct to S-formylglutathione. S-Formylglutathione hydrolase catalyzes the hydrolysis of S-formylglutathione to formate and GSH.

Formaldehyde is oxidized in mammalian tissues and in several other sources to formate in two consecutive reactions catalyzed by separate enzymes (Uotila and Koivusalo, 1974a; 1974b) (Scheme 1). Formaldehyde and glutathione react first non-enzymically to form a hemithioacetal adduct (S-hydroxymethylglutathione) (Reaction 1). The adduct is the substrate of formaldehyde dehydrogenase (EC 1.2.1.1) and is oxidized in the NAD-depend-ent reaction to S-formylglutathione (Reaction 2). Formaldehyde dehydrogenase has been shown to be identical with the class III alcohol dehydrogenase (Koivusalo et al., 1989). For a comprehensive review on formaldehyde dehydrogenase, see Uotila and Koivusalo (1989). The hydrolysis of S-formylglutathione to reduced glutathione and formate (Reaction 3) is catalyzed by a specific enzyme, S-formylglutathione hydrolase (EC 3.1.2.12). Formate can be oxidized by the catalase reaction or is transferred to tetrahydrofolate to form 10-formyl-tetrahydrofolate. Formate is thus incorporated into the reactions of the C-1 metabolism by which it can be oxidized to carbon dioxide in the 10-tetrahydrofolate dehydrogenase reaction (Kutzbach and Stokstad, 1968).

Glutathione-dependent formaldehyde dehydrogenase (EC 1.2.1.1) was first described from beef and chicken livers (Strittmatter and Ball, 1955). The enzyme was first purified to homogeneity and characterized from human liver by Uotila and Koivusalo (1974a) who showed that the enzyme catalyzes a reversible reaction in which formaldehyde, GSH and NAD are converted into S-formyl-glutathione and NADH. The actual substrate of the enzyme is S-hydroxy-methylglutathione, a hemimercaptal adduct nonenzymically formed from formaldehyde and GSH (Uotila and Mannervik, 1979). S-Formylglutathione is irreversibly hydrolyzed into formate and GSH in the reaction catalyzed by a specific S-formylglutathione hydrolase (EC 3.1.2.12) which has also been purified and characterized from human liver (Uotila and Koivusalo, 1974b). Both formaldehyde dehydrogenase and S-formylglutathione hydrolase are widely occurring enzymes and have later been isolated and characterized from a number of other sources in addition to human liver (Uotila and Koivusalo, 1989; Uotila, 1989).

Alcohol dehydrogenases (EC 1.1.1.1) from human (Vallee and Bazzone, 1983), rat (Julia et al., 1987b), mouse (Algar et al., 1983) and other mammalian species (Julia et al., 1987a) are divided into three classes, I, H and III, which are characterized by their specific, differing electrophoretic mobilities, reactivities with different alcohols and sensitivities to pyrazole and its derivatives. Class III alcohol dehydrogenases have anodic electrophoretic mobility at neutral to alkaline pH, and they are further distinguished by their insensitivity to pyrazole and poor use of ethanol and other short-chain alcohols as substrate. Long-chain alcohols like 1-octanol are, in contrast, used well as substrates by these enzymes (Vallee and Bazzone, 1983). The class III alcohol dehydrogenases occur in mammalian tissues much more widely than the enzymes from the other classes (Julia et al., 1987b). The class III enzymes have been purified from human liver (Wagner et al., 1984) and brain (Beisswenger et al., 1985), and from the livers of horse (Kaiser et al., 1989), rat (Julia et al., 1987b) and mouse (Algar et al., 1983) but the physiological significance of these enzymes has remained obscure although their involvement in the oxidation of long-chain fatty alcohols or ω-hydroxy fatty acids has been postulated (Giri et al., 1989).

The glutathione-dependent formaldehyde dehydrogenase (EC 1.2.1.1) is a ubiquitous cytoplasmic enzyme, which was first described in 1955 Strittmatter and Ball from beef and chicken livers. It was purified to homogeneity and characterized from human liver by Uotila and Koivusalo in 1974 (1974a) and has later also been purified from some other sources (Uotila and Koivusalo 1989, Pourmotabbed et al. 1989). Formaldehyde ehydrogenase catalyzes the following reversible reaction: $$Formaldehyde + GSH + NA{D^ + } \leftrightarrow S - formylglutathione + NADH + {H^ + }$$

The steady-state kinetic mechanism of vitamin K-dependent carboxylase from calf liver has been investigated by initial-velocity measurements with varying concentrations of two carboxylase substrates and constant, nonsaturating concentrations of the other two substrates. With all combinations of the varied substrates tested linear kinetics were obtained with lines intersecting on the left side of the 1 v axis in double-reciprocal plots. Thus the carboxylase has a sequential reaction mechanism which includes the quinternary complex of the enzyme with its four substrates. A mechanism with the ordered steady-state addition of all substrates to the enzyme accords well with the results. A totally random mechanism was excluded but the alternative possibility remained that part of the substrates are added in a rapid-equilibrium random reaction. Experiments with saturating constant concentrations of sodium bicarbonate and varying concentrations of the other substrates suggest that bicarbonate (CO2) is either the first or, more probably, the last substrate bound to the enzyme.

A new specific and sensitive assay method for reduced glutathione has been developed. It is based on the reaction HCHO + NAD + + H 2 O → HCOOH + NADH + H + , catalyzed by formaldehyde dehydrogenase (formaldehyde: NAD oxidoreductase, EC.1.2.1.1) in the presence of reduced glutathione. Oxidized glutathione and other thiols do not interfere in this reaction. A purification procedure for formaldehyde dehydrogenase from beef liver is presented. The influence of cysteine and some other thiols, leucine, ascorbate, lactate, pyruvate and four acids generally used for deproteinization is reported. The results obtained by this method from human blood and rat tissues are compared with those obtained by Ellman's method.

Glyoxalase II (S-(2-hydroxyacyl)glutathione hydrolase, EC 3.1.2.6) was purified to homogeneity and separated into two forms (alpha, pI = 8.0; beta, pI = 7.4) from both liver and brain of wistar rats by column isoelectric focusing. These forms were also found to have different electrophoretic mobilities. No significant differences were found between the alpha and beta forms from either source in the relative molecular mass (about 24,000) or in Km values using three substrates. The temperature-inactivation profiles were also similar, the two forms being stable up to 50 degrees C. Chemical modification studies with phenylglyoxal suggest that these enzyme forms probably contain arginine residues near the active site. Inactivation of alpha and beta forms by diethylpyrocarbonate and by photooxidation with methylene blue, and protection by S-D-mandeloylglutathione, a slowly reacting substrate, suggest the presence of histidine at the active site. The alpha and beta forms show different half-life values in inactivation by histidine reagents, which may be due to a difference in the active-site structures of these enzymes. The results probably indicate distinct structures (sequences) for alpha and beta forms.

The number and the substrate specificities of glutathione thiol esterases of human red blood cells have been investigated by gel electrophoresis and isoelectric focusing and staining methods devised for the location of these enzymes on gels. Several glutathione thiol esterase forms, both unspecific (with respect to the S-acyl group of the substrate) and specific were found. Electrophoresis on both polyacrylamide and agarose gels resolved three enzyme components with apparently similar substrate specificity. Isoelectric focusing in liquid column separated two unspecific thiol esterase components with S -lactoylglutathione (p I =8.4) and S -propionylglutathione (p I =8.1) as the best substrates, respectively, and two specific enzymes, S -formylglutathione hydrolase (p I =5.2) and S -succinylglutathione hydrolase (p I =9.0). Isoelectric focusing on polyacrylamide gel resolved nine unspecific glutathione thiol esterase bands (between pH values 7.0 and 8.4). Partially purified glyoxalase II ( S -2-hydroxyacylglutathione hydrolase, EC 3.1.2.6) from erythrocytes or liver still gave three components on electrophoresis and several activity bands on gel electrofocusing. These results indicate that human red cells contain at least four separate glutathione thiol esterases. Glyoxalase II, one of these enzymes, apparently occurs in multiple forms. These were neither influenced by pretreatment of the samples with neuraminidase or thiols nor were interconvertible during the fractionations.

We have purified formaldehyde dehydrogenase (EC 1.2.1.1) 1390-fold from human liver. The final preparation, which has a specific activity of 3.20 i.u. per mg of protein (25°) is homogeneous according to electrophoretic criteria. S-Formylglutathione rather than formate is formed from formaldehyde and reduced glutathione in the reaction catalyzed by purified formaldehyde dehydrogenase. The enzyme is not strictly NAD-specific; NADP can also be used although NADP has a much higher Km value than NAD, especially at high pH values. S-Formylglutathione is reduced by the enzyme to formaldehyde with either NADH or NADPH as cofactors. Methylglyoxal and some other ketoaldehydes also can be used as the substrates of formaldehyde dehydrogenase. The product obtained from methylglyoxal is probably S-pyruvylglutathione. S-Formylglutathione is hydrolyzed very actively in crude human liver preparations. All of this activity is removed during the purification of formaldehyde dehydrogenase. Formaldehyde dehydrogenase has an apparent molecular weight of 81,400 according to gel filtration and a subunit molecular weight of 39,500 according to dodecyl sulfate gel electrophoresis. Isoelectric focusing experiments gave an isoelectric point, pI of 6.35. The enzyme is very sensitive to mercaptide-forming reagents. NAD and NADH protect the enzyme. NAD and especially NADH stabilize the enzyme during storage and against denaturation at high temperatures.

Two separate pools of glyoxalase II were demonstrated in rat liver mitochondria, one in the intermembrane space and the other in the matrix. The enzyme was purified from both sources by affinity chromatography on S-(carbobenzoxy)glutathione-Affi-Gel 40. From both crude and purified preparations polyacrylamide gel-electrophoresis resolved multiple forms of glyoxalase II, two from the intermembrane space and five from the matrix. Among the thioesters of glutathione tested as substrates, S-D-lactoylglutathione was hydrolyzed most efficiently by the enzymes from both sources. Significant differences were observed in the specificities between the intermembrane space and matrix enzymes with S-acetoacetylglutathione, S-acetylglutathione, S-propionylglutathione and S-succinylglutathione as substrates. Pure glyoxalase II from rat liver cytosol was chemically polymerized and used as antigen. Antibodies were raised in rabbits and the antiserum was used for comparison of the two purified mitochondrial enzymes with cytosolic glyoxalase II by immunoblotting. The enzyme purified from the intermembrane space cross-reacted with the antiserum, but the matrix glyoxalase II did not. The results give evidence for the presence in rat liver mitochondria of two species of glyoxalase II with differing characteristics. Only the enzyme from the intermembrane space appears to resemble the cytosolic glyoxalase II forms.

Formaldehyde dehydrogenase (EC 1.2.1.1) is a widely occurring enzyme which catalyzes the oxidation of S-hydroxymethylglutathione, formed from formaldehyde and glutathione, into S-formylglutathione in the presence of NAD. We determined the amino acid sequences for 5 tryptic peptides (containing altogether 57 amino acids) from electrophoretically homogeneous rat liver formaldehyde dehydrogenase and found that they all were exactly homologous to the sequence of rat liver class III alcohol dehydrogenase (ADH-2). Formaldehyde dehydrogenase was found to be able at high pH values to catalyze the NAD-dependent oxidation of long-chain aliphatic alcohols like n-octanol and 12-hydroxydodecanoate but ethanol was used only at very high substrate concentrations and pyrazole was not inhibitory. The amino acid sequence homology and identical structural and kinetic properties indicate that formaldehyde dehydrogenase and the mammalian class III alcohol dehydrogenases are identical enzymes.Formaldehyde dehydrogenase; Alcohol dehydrogenase, class III; Sequence homology; Amino acid sequence

S-Formylglutathione hydrolase, a new glutathione thiol esterase from human liver, has been purified into homogeneity according to disc electrophoretic and ultracentrifugal criteria. The final preparation catalyzes the hydrolysis of 4100 µmoles of S-formylglutathione per min per mg of protein at 25° and represents 2,350-fold purification over 23,000 x g supernatant of liver homogenate. In addition to S-formylglutathione, the enzyme can catalyze the hydrolysis of S-acetylglutathione at a 200-fold slower velocity but has no activity with S-lactylglutathione. The elution volume of the enzyme in a gel filtration column gives an apparent molecular weight of 52,500 and a diffusion coefficient of 6.57 x 10-7 cm2 s-1. The enzyme has a sedimentation coefficient of 4.24 S as judged from sedimentation velocity in analytical ultracentrifuge. From a sedimentation equilibrium experiment, the molecular weight of the enzyme is 55,500. Dodecyl sulfate gel electrophoresis indicates a subunit molecular weight of 30,000. Thus, the enzyme probably consists of 2 subunits. The isoelectric point of the enzyme obtained from electrofocusing experiments is 5.41. The enzyme apparently contains a reactive —SH group and purification is possible only in the presence of thiols. Activity is rapidly lost when the thiols are removed, but dithiothreitol can regenerate part of the activity. Several types of —SH reagents are inhibitory. Dithiothreitol can reverse completely the effect of mercaptide-forming and oxidizing agents. Amino group reagents also rapidly inactivate the enzyme. In addition, ascorbate and folate are inhibitory. Chelating agents and organophosphates do not inhibit the enzyme.

Publisher Summary This chapter presents a procedure for the preparation of formaldehyde dehydrogenase. Formaldehyde dehydrogenase can be assayed by recording the initial velocity spectrophotometrically at 340 nm and 25°. Either the formation of nicotinamide adenine dinucleotide–hydrogen (NADH) in the forward reaction, or the disappearance of NADH in the reverse reaction can be measured. For crude tissue preparations, only the forward reaction assay may be used because of the presence of a large excess of activity catalyzing the hydrolysis of S -formylglutathione. The enzyme can be purified to homogeneity from the human liver, involving extraction; salt fractionation; washing on diethylaminoethyl (DEAE)-cellulose 1and DEAE-cellulose 2; isoelectric focusing; and finally washing on Sephadex G-100, hdroxylapatite, and QAE-Sephadex column. The final preparation is homogeneous in polyacrylamide electrophoresis under denaturing and nondenaturing conditions and represents about 1400-fold purification over the crude soluble fraction of the human liver.

The steady-state kinetic mechanism of human liver formaldehyde dehydrogenase (formaldehyde:NAD+ oxidoreductase (glutathione-formylating), EC 1.2.1.1) was investigated by product inhibition of the forward and the reverse reactions catalyzed by the enzyme. The results are compatible with a mechanism which contains the random addition to the enzyme of NAD+ and S-hydroxymethylglutathione (the adduct of glutathione and formaldehyde), or NADH and S-formylglutathione, and free glutathione as the allosteric activator of the enzyme (Uotila, L. and Mannervik, B. (1979) Biochem. J. 177, 869–878).

1 Glyoxalase I has been obtained in electrophoretically pure form from sheep liver by a procedure which includes ammonium sulfate and poly(ethylene glycol) fractionations and column chromatographies on hydroxyapatite, Cibacron Blue–Sephadex G-100 and DEAE-cellulose. The specific activity of the homogeneous preparations is about 4000 units/mg of protein (25° C). Three separate peaks of activity were obtained in the last column on DEAE-cellulose (DE-32). No signs of heterogeneity were seen in the previous steps. Purified but not crude preparations gave two activity peaks on disc gel electrophoresis. 2 The isoelectric point of metal-free apoglyoxalase I is 5.0 by isoelectric focusing. The apparent molecular weight of glyoxalase I is 45 900 from gel chromatography. From dodecylsulfate gel electrophoresis the enzyme has a subunit molecular weight of 21000. 3 In addition to methylglyoxal, phenylglyoxal and kethoxal are good substrates of sheep liver glyoxalase I. Hydroxypyruvaldehyde and glyoxal react more slowly. 4 Catalytically inactive apoenzyme of sheep liver glyoxalase I has been prepared by dialysis against EDTA and Chelex-100. All of the activity is restored by Mg2+ ions; Zn2+, Mn2+, Co2+, Ni2+ and Ca2+ ions, in decreasing order of maximum velocity, give partial reactivation. Of these metals, Mg2+ has the highest (2.8 mM) and Zn2+ the lowest (0.007 mM) apparent half-saturation concentration when assayed in 80 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonate (Hepes) buffer pH 6.8. 5 Several chelating agents are inhibitors of sheep liver glyoxalase I. In all cases Mg2+ reverses the inhibition after a short incubation time.

Glyoxalase II (S-(2-hydroxyacyl)glutathione hydrolase, EC 3.1.2.6), which has been regarded as a cytosolic enzyme, was also found in rat liver mitochondria. The mitochondrial fraction contained about 10–15% of the total glyoxalase II activity in liver. The actual existence of the specific mitochondrial glyoxalase II was verified by showing that all of the activity of the crude mitochondrial pellet was still present in purified mitochondria prepared in a Ficoll gradient. Subfractionation of the mitochondria by digitonin treatment showed that 56% of the activity resided in the mitochondrial matrix and 19% in the intermembrane space. Partial purification of the enzyme (420-fold) was also achieved. Statistically significant differences were found in the substrate specificities of the mitochondrial and the cytosolic glyoxalase II. Electrophoresis and isoelectric focusing of either the crude mitochondrial extract or of the purified mitochondrial glyoxalase II resolved the enzyme activity into five forms with the respective pI values of 8.1, 7.5, 7.0, 6.85 and 6.6. Three of these forms (pI values 7.0-6.6) were exclusively mitochondrial, with no counterpart in the cytosol. The relative molecular mass of the partially purified enzyme, as estimated by Superose 12 gel chromatography, was 21 000. These results give evidence for the presence of mitochondrial glyoxalase II which is different from the cytosolic enzymes in several characteristics.

Clinical use of the P3-lactam cephalosporins has resulted in a number of reports of vitamin K-responsive bleeding abnormalities. The effects of these antibiotics on the activities of vitamin K-dependent enzymes in vitro were therefore assessed. The activity of vitamin K-dependent carboxylase was inhibited by 50% by 6-10 mM cefamandole or structurally analogous cephalosporins. The inhibition was noncompetitive with respect to the pentapeptide substrate, vitamin KH2, and CO2. Vitamin K epoxide reductase was less sensitive to and only partially inhibited by cefamandole and the analogs. The possible physiologic significance of the inhibition of these enzymes in clinical hemorrhagic problems was considered. Vitamin K is a required cofactor for the posttrans

Formaldehyde dehydrogenase (EC 1.2.1.1) and formate dehydrogenase (EC 1.2.1.2) have been isolated in pure form from pea seeds by a rapid procedure which employs column chromatographies on 5′-AMP-Sepharose, Sephacryl S-200, and DE32 cellulose. The apparent molecular weights of formaldehyde and formate dehydrogenases are, respectively, 82,300 and 80,300 by gel chromatography, and they both consist of two similar subunits. The isoelectric point of formaldehyde dehydrogenase is 5.8 and that of formate dehydrogenase is 6.2. The purified formate dehydrogenase gave three corresponding protein and activity bands in electrophoresis and isoelectric focusing on polyacrylamide gel whereas formaldehyde dehydrogenase gave only one band. Formaldehyde dehydrogenase catalyzes the formation of S-formylglutathione from formaldehyde, and glutathione. Formate dehydrogenase can, besides formate, also use S-formylglutathione and two other formate esters as substrates. S-Formylglutathione has a lower Km value (0.45 m m ) than formate (2.1 m m ) but the maximum velocity of S-formylglutathione is only 5.5% of that of formate. Pea extracts also contain a highly active S-formylglutathione hydrolase which has been separated from glyoxalase II (EC 3.1.2.6) and partially purified. S-Formylglutathione hydrolase is apparently needed between formaldehyde and formate dehydrogenases in the metabolism of formaldehyde in pea seeds, in contrast to what was recently reported for Hansenula polymorpha, a yeast grown on methanol.