Medium-chain acyl coenzyme A dehydrogenase from pig kidney has intrinsic enoyl coenzyme A hydratase activity

The flavoprotein medium-chain acyl coenzyme A (acyl-CoA) dehydrogenase from pig kidney exhibits an intrinsic hydratase activity toward crotonyl-CoA yielding L-3-hydroxybutyryl-CoA. The maximal turnover number of about 0.5 min-1 is 500-1000-fold slower than the dehydrogenation of butyryl-CoA using electron-transferring flavoprotein as terminal acceptor. trans-2-Octenoyl- and trans-2-hexadecenoyl-CoA are not hydrated significantly. Hydration is not due to contamination with the short-chain enoyl-CoA hydratase crotonase. Several lines of evidence suggest that hydration and dehydrogenation reactions probably utilize the same active site. These two activities are coordinately inhibited by 2-octynoyl-CoA and (methylenecyclopropyl)acetyl-CoA [whose targets are the protein and flavin adenine dinucleotide (FAD) moieties of the dehydrogenase, respectively]. The hydration of crotonyl-CoA is severely inhibited by octanoyl-CoA, a good substrate of the dehydrogenase. The apoenzyme is inactive as a hydratase but recovers activity on the addition of FAD. Compared with the hydratase activity of the native enzyme, the 8-fluoro-FAD enzyme exhibits a roughly 2-fold increased activity, whereas the 5-deaza-FAD dehydrogenase is only 20% as active. A mechanism for this unanticipated secondary activity of the acyl-CoA dehydrogenase is suggested.     

Kinetic characterization of the [3'-32P]coenzyme A/acetyl coenzyme A exchange catalyzed by a three-subunit form of the carbon monoxide dehydrogenase/acetyl-CoA synthase from Clostridium thermoaceticum

The ability of acetyl coenzyme A synthesizing carbon monoxide dehydrogenase isolated from Clostridium thermoaceticum to catalyze the exchange of [3'-32P]coenzyme A with acetyl coenzyme A is studied. This exchange is found to have a rate exceeding that of the acetyl coenzyme A carbonyl exchange also catalyzed by CO dehydrogenase ([1-14C]acetyl coenzyme A + CO in equilibrium acetyl coenzyme A + 14CO). These two exchanges are diagnostic of the ability of CO dehydrogenase to synthesize acetyl coenzyme A from a methyl group, coenzyme A, and carbon monoxide. The kinetic parameters for the coenzyme A exchange have been determined: Km(acetyl coenzyme A) = 1500 microM, Km(coenzyme A) = 50 microM, and Vmax = 2.5 mumol min-1 mg-1. Propionyl coenzyme A is shown to be a substrate (Km approximately 5 mM) for the coenzyme A exchange, with a rate 1/15 that of acetyl coenzyme A, but is not a substrate for the carbonyl exchange. CO dehydrogenase capable of catalyzing both these two exchanges, and the oxidation of CO to CO2, is isolated as a complex of molecular weight 410,000 consisting of three proteins in an alpha 2 beta 2 gamma 2 stoichiometry. The proposed gamma subunit, not previously reported as part of CO dehydrogenase, copurifies with the enzyme and has the same molecular weight on sodium dodecyl sulfate-polyacrylamide gel electrophoresis as the disulfide reductase previously separated from CO dehydrogenase in a final chromatographic step.     

Enhanced proteolysis of glucose-6-phosphate dehydrogenase in the presence of palmitoyl coenzyme A

Palmitoyl coenzyme A at concentrations below its critical micelle concentration increases the rate of proteolysis of baker's yeast glucose-6-phosphate dehydrogenase by proteinase A in the pH range 4-5. both glucose-6-phosphate and NADP protect glucose-6-phosphate dehydrogenase against proteolysis, but these protective effects are diminished in the presence of palmitoyl coenzyme A. Since palmitoyl coenzyme A is known to dissociate glucose-6-phosphate dehydrogenase into dimers, the results imply that the in vivo half life of glucose-6-phosphate dehydrogenase may be controlled by a process based on the regulation of the oligomeric structure of the enzyme by the collective actions of various molecules, including palmitoyl coenzyme A.     

Acetaldehyde coenzyme A dehydrogenase of Escherichia coli.

Mutants of Escherichia coli (adh) in which alcohol dehydrogenase is derepressed under aerobic conditions were also found to overproduce acetaldehyde coenzyme a dehydrogenase. However, acetaldehyde coenzyme A dehydrogenase was induced by ethanol or acetaldehyde and subject to strong catabolite repression, whereas alcohol dehydrogenase was little affected by these conditions. Mutants no longer able to use ethanol as carbon source were isolated from an adh strain. Some of these mutants were revertants at the adh locus and no longer produced either alcohol dehydrogenase or acetaldehyde coenzyme A dehydrogenase. Others, designated acd, were found to lack only acetaldehyde coenzyme A dehydrogenase. The acd mutation was located at min 62 of the E. coli genetic map, the gene order being thyA-lysA-acd-serA-fda. Isolation of Tn10 insertions cotransducible with acd greatly simplified the mapping procedure.     

Acyl coenzyme A dependent retinol esterification by acyl coenzyme A: diacylglycerol acyltransferase 1

We provide biochemical evidence that enzymes involved in the synthesis of triacylglycerol, namely acyl coenzyme A:diacylglycerol acyltransferase (DGAT) and acyl coenzyme A:monoacylglycerol acyltransferase (MGAT), are capable of carrying out the acyl coenzyme A:retinol acyltransferase (ARAT) reaction. Among them, DGAT1 appears to have the highest specific activity. The apparent K(m) values of recombinant DGAT1/ARAT for retinol and palmitoyl coenzyme A were determined to be 25.9+/-2.1 microM and 13.9+/-0.3 microM, respectively, both of which are similar to the values previously determined for ARAT in native tissues. A novel selective DGAT1 inhibitor, XP620, inhibits recombinant DGAT1/ARAT at the retinol recognition site. In the differentiated Caco-2 cell membranes, XP620 inhibits approximately 85% of the Caco-2/ARAT activity indicating that DGAT1/ARAT may be the major source of ARAT activity in these cells. Of the two most abundant fatty acyl retinyl esters present in the intact differentiated Caco-2 cells, XP620 selectively inhibits retinyl-oleate formation without influencing the retinyl-palmitate formation. Using this inhibitor, we estimate that approximately 64% of total retinyl ester formation occurs via DGAT1/ARAT. These studies suggest that DGAT1/ARAT is the major enzyme involved in retinyl ester synthesis in Caco-2 cells.     

Molecular cloning and nucleotide sequence of cDNA encoding the entire precursor of rat liver medium chain acyl coenzyme A dehydrogenase

cDNA encoding the precursor of rat liver medium chain acyl-CoA dehydrogenase (EC 1.3.99.3) was cloned and sequenced. The longest cDNA insert isolated was 1866 bases in length. This cDNA encodes the entire protein of 421-amino acids including a 25-amino acid leader peptide and a 396-amino acid mature polypeptide. The identity of the medium chain acyl-CoA dehydrogenase clone was confirmed by matching the amino acid sequence predicted from the cDNA to the NH2-terminal and nine internal tryptic peptide sequences derived from pure rat liver medium chain acyl-CoA dehydrogenase. The calculated molecular masses of the precursor medium chain acyl-CoA dehydrogenase, the mature medium chain acyl-CoA dehydrogenase, and the leader peptide are 46,600, 43,700, and 2,900 daltons, respectively. The leader peptide contains five basic amino acids and only one acidic amino acid; thus, it is positively charged, overall. Cysteine residues are unevenly distributed in the mature portion of the protein; five of six are found within the NH2-terminal half of the polypeptide. Comparison of medium chain acyl-CoA dehydrogenase sequence to other flavoproteins and enzymes which act on coenzyme A ester substrates did not lead to unambiguous identification of a possible FAD-binding site nor a coenzyme A-binding domain. The sequencing of other homologous acyl-CoA dehydrogenases will be informative in this regard.     

Transport of coenzyme A in plant mitochondria

Oxoglutarate oxidation by purified potato mitochondria which had been stored at low temperature for 48 h or longer was stimulated by added coenzyme A. Exogenous coenzyme A was accumulated by potato mitochondria, both freshly prepared and aged, in a manner sensitive to uncouplers and low temperature. Coenzyme A was concentrated approximately 10-fold in the matrix under steady-state conditions. This coenzyme A uptake followed saturation kinetics with an apparent Km of 0.2 mM and a V of 4-6.5 nmol min-1 mg-1 protein, suggesting carrier-mediated transport. This transport was insensitive to an inhibitor of NAD+ transport. It is suggested that plant mitochondria possess a specific carrier for the net accumulation of coenzyme A.     

Acyl coenzyme A inhibition of Leuconostoc mesenteroides glucose-6-phosphate dehydrogenase: a comparison of the TPN and DPN linked reactions

The inhibitory effects of ATP, coenzyme A, and acetyl, malonyl, and oleyl derivatives of coenzyme A on the TPN and DPN dependent activities of Leuconostoc glucose-6-phosphate dehydrogenase are compared. At pH 7.8, 24°, saturating levels of DPN or TPN, and inhibitor concentrations of 2–4 mM only ATP has an appreciable effect on the TPN dependent reaction, but all were potent inhibitors of the DPN dependent reaction. Oleyl coenzyme A was the most effective (K_i ～ 0.15 mM against glucose-6-phosphate) while acetyl coenzyme A was least effective (K_i ～ 1.0 mM). A possible regulatory role of this inhibition in fatty acid synthesis is suggested.     

The origin of chloroplastic acetyl coenzyme A

Pyruvic dehydrogenase activity has been examined in a number of highly purified leaf organelles. In spinach leaf cell, the major activity is in the mitochrondrion with low activity in isolated chloroplasts. The major source of CO₂ derived from pyruvic acid metabolism in the isolated chloroplast is via the acetolactic synthase reaction localized in the chloroplast. Evidence is presented that the leaf mitochondrion contains both the pyruvic acid dehydrogenase and an acetyl coenzyme A hydrolase. It is suggested that free acetic acid is generated in the mitochrondrion and then moves to the chloroplast where acetyl coenzyme synthetase converts it from the metabolically inert acid to the very metabolically active acetyl coenzyme A.     

Functional arginine residues involved in coenzyme binding by glutamate dehydrogenases.

Reaction of the NADP-dependent glutamate dehydrogenase of Neurospora with 1,2-cyclohexanedione results in a biphasic loss of enzyme activity. At the end of the rapid phase of the reaction (t1/2 = 1.5 min) the enzyme activity is diminished by approximately 60% with the simultaneous loss of 1 residue of arginine per subunit. After 60 min, the enzyme activity is completely lost with the modification of a total of 2 arginine residues per subunit. Reaction of bovine liver glutamate dehydrogenase with cyclohexanedione causes a rapid loss of approximately 45% of the enzyme activity and modification of about 1.5 residues of arginine per subunit. More prolonged treatment results in reaction of an additional 4 residues of arginine per subunit but is without further effect on the residual activity. The activity of the Neurospora enzyme is not protected by substrate, coenzyme, or a combination of both; however, the activity of the bovine enzyme is partially protected by high levels of NAD or NADP. Although the Km for alpha-ketoglutarate is unchanged by a limited modification of either enzyme with cyclohexanedione, the Km for coenzyme is increased about 2-fold for the Neurospora enzyme and about 1.5-fold for the bovine enzyme. The Ki of the Neurospora dehydrogenase for the competitive inhibitor 2'-monophosphoadenosine-5'-diphosphoribose is unchanged by the enzyme modification, but nicotinamide mononucleotide, a competitive inhibitor for the native Neurospora enzyme, does not inhibit the glutamate dehydrogenase with 1 modified arginine residue. This finding implies that the modified arginine is at or near the nicotinamide binding iste of the enzyme.     

Autotrophic acetyl coenzyme A biosynthesis in Methanococcus maripaludis.

To detect autotrophic CO2 assimilation in cell extracts of Methanococcus maripaludis, lactate dehydrogenase and NADH were added to convert pyruvate formed from autotrophically synthesized acetyl coenzyme A to lactate. The lactate produced was determined spectrophotometrically. When CO2 fixation was pulled in the direction of lactate synthesis, CO2 reduction to methane was inhibited. Bromoethanesulfonate (BES), a potent inhibitor of methanogenesis, enhanced lactate synthesis, and methyl coenzyme M inhibited it in the absence of BES. Lactate synthesis was dependent on CO2 and H2, but H2 + CO2-independent synthesis was also observed. In cell extracts, the rate of lactate synthesis was about 1.2 nmol min-1 mg of protein-1. When BES was added, the rate of lactate synthesis increased to 2.3 nmol min-1 mg of protein-1. Because acetyl coenzyme A did not stimulate lactate synthesis, pyruvate synthase may have been the limiting activity in these assays. Radiolabel from 14CO2 was incorporated into lactate. The percentages of radiolabel in the C-1, C-2, and C-3 positions of lactate were 73, 33, and 11%, respectively. Both carbon monoxide and formaldehyde stimulated lactate synthesis. 14CH2O was specifically incorporated into the C-3 of lactate, and 14CO was incorporated into the C-1 and C-2 positions. Low concentrations of cyanide also inhibited autotrophic growth, CO dehydrogenase activity, and autotrophic lactate synthesis. These observations are in agreement with the acetogenic pathway of autotrophic CO2 assimilation.     

Inactivation of two-electron reduced medium chain acyl-CoA dehydrogenase by 2-octynoyl-CoA

The acetylenic thioester, 2-octynoyl-CoA, inactivates medium chain acyl-CoA dehydrogenase from pig kidney by two distinct pathways depending on the redox state of the FAD prosthetic group. Inactivation of the oxidized dehydrogenase occurs with labeling of an active site glutamate residue and elimination of CoASH. Incubation of the reduced dehydrogenase with 2-octynoyl-CoA rapidly forms a kinetically stable dihydroflavin species which is resistant to reoxidation using trans-2-octenoyl-CoA, molecular oxygen, or electron transferring flavoprotein. The reduced enzyme derivative shows extensive bleaching at 446 nm with shoulders at 320 and 380 nm. Denaturation of the reduced derivative in 80% methanol yields a mixture of products which was characterized by HPLC, by uv/vis, and by radiolabeling experiments. Approximately 20% of the flavin is recovered as oxidized FAD, about 40% is retained covalently attached to the protein, and the remainder is distributed between several species eluting after FAD on reverse-phase HPLC. The spectrum of one of these species ressembles that of a N(5)-C(4a) dihydroflavin adduct. These data suggest that a primary reduced flavin species undergoes various rearrangements during release from the protein. The possibility that the inactive modified enzyme represents a covalent adduct between 2-octynoyl-CoA and reduced flavin is discussed. Analogous experiments using enzyme substituted with 1,5-dihydro-5-deaza-FAD show rapid and quantitative reoxidation of the flavin by 0.5 eq of 2-octynoyl-CoA.     

Nicotinamide adenine dinucleotide phosphate-specific glutamate dehydrogenase of Neurospora. III. Inactivation by nitration of a tyrosine residue involved in coenzyme binding.

Neurospora glutamate dehydrogenase (NADP-specific) is rapidly inactivated upon reaction with tetranitromethane. This inactivation is completely prevented by the presence of coenzyme (NADP) or nicotinamide mononucleotide (NMN) but not by substrate. NADH, or 2'-monophosphoadenosine-5'-diphosphoribose. Amino acid analysis indicates that the primary effect of modification is nitration of a single residue of tyrosine per polypeptide chain. We have identified the reactive tyrosine by isolation of a single, uniquely labeled peptide after hydrolysis with trypsin followed by cleavage with cyanogen bromide. The modified residue proved to be tyrosine-168 in the linear sequence. This residue is not present in the part of the sequence that had been previously implicated as involved in the binding of the adenylate portion of the coenzyme. Both NMN and 2-monophosphoadenosine-5'-diphosphoribose act as competitive inhibitors of NADP in the oxidation of glutamate with Ki values of 4.65 x 10(-4) M and 4.30 x 10(-4) M, respectively. Thus, the specific protection afforded by NADP and NMN, but not by 2'-monophosphoadenosine-5'-diphosphoribose, indicates that tyrosine-168 is involved in binding the nicotinamide portion of the coenzyme.     

3-Hydroxy-3-methylglutaryl coenzyme A lyase: affinity labeling of the Pseudomonas mevalonii enzyme and assignment of cysteine-237 to the active site.

Pseudomonas mevalonii 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) lyase is irreversibly inactivated by the reactive substrate analog 2-butynoyl-CoA. Enzyme inactivation, which follows pseudo-first-order kinetics, is saturable with a KI = 65 microM and a limiting k(inact) of 0.073 min-1 at 23 degrees C, pH 7.2. Protection against inactivation is afforded by the competitive inhibitor 3-hydroxyglutaryl-CoA. Labeling of the bacterial enzyme with [1-14C]-2-butynoyl-CoA demonstrates that inactivation coincides with covalent incorporation of inhibitor, with an observed stoichiometry of modification of 0.65 per site. Avian HMG-CoA lyase is also irreversibly inactivated by 2-butynoyl-CoA with a stoichiometry of modification of 0.9 per site. Incubation of 2-butynoyl-CoA with mercaptans such as dithiothreitol results in the formation of a UV absorbance peak at 310 nm. Enzyme inactivation is also accompanied by the development of a UV absorbance peak at 310 nm indicating that 2-butynoyl-CoA modifies a cysteine residue in HMG-CoA lyase. Tryptic digestion and reverse-phase HPLC of the affinity-labeled protein reveal a single radiolabeled peptide. Isolation and sequence analysis of this peptide and a smaller chymotryptic peptide indicate that the radiolabeled residue is contained within the sequence GGXPY. Mapping of this peptide within the cDNA-deduced sequence of P. mevalonii HMG-CoA lyase [Anderson, D. H., & Rodwell, V. W. (1989) J. Bacteriol. 171, 6468-6472] confirms that a cysteine at position 237 is the site of modification. These data represent the first identification of an active-site residue in HMG-CoA lyase.     

Covalent binding of an NAD analogue to liver alcohol dehydrogenase resulting in an enzyme-coenzyme complex not requiring exogenous coenzyme for activity.

1. The NAD analogue, N6-[N-(6-aminohexyl)carbamoylmethyl]-NAD, was covalently bound to horse liver alcohol dehydrogenase in a carbodiimide-mediated reaction and in such a way that it was active with the very same enzyme molecule to which it was coupled. 2. The degree of substitution, i.e. the number of NAD analogues per enzyme subunit, could be varied (0.3-1.6). In one preparation 1.6 coenzyme molecules were bound per subunit; the alcohol dehydrogenase activity of this preparation was 40% of the activity obtained after addition of free NAD in excess. 3. It was calculated that every fourth active site of this preparation was provided with a covalently bound functioning coenzyme analogue, and that this analogue had a cycling rate of about 40 000 cycles/h in a coupled substrate assay. 4. The presence of the covalently bound coenzyme made the active sites difficult to inhibit with a competitive inhibitor. For example, 10 mM AMP inhibited the activity of the preparation by 50% whereas a reference system containing native alcohol dehydrogenase was inhibited by 80% in spite of the fact that the reference system contained about 20 000 times as high a concentration of coenzyme.     

Induction of acyl coenzyme A synthetase and hydroxyacyl coenzyme A dehydrogenase during fatty acid degradation in Neurospora crassa

Neurospora crassa is able to use long-chain fatty acids as the sole carbon and energy source. After growth on oleate there was nearly a 10-fold induction of the acyl coenzyme A (CoA) synthetase and a fivefold increase in the activity of the 3-hydroxyacyl-CoA dehydrogenase. There was a slight induction of the enoyl-CoA hydratase and 3-ketoacyl-CoA thiolase, but no apparent induction of the flavin-linked acyl-CoA dehydrogenase. These noncoordinate changes in the fatty acid degradation enzymes suggest that they are not organized into a multienzyme complex as is found in bacteria.     

Acyl coenzyme A dependent retinol esterification by acyl coenzyme A:diacylglycerol acyltransferase 1

We provide biochemical evidence that enzymes involved in the synthesis of triacylglycerol, namely acyl coenzyme A:diacylglycerol acyltransferase (DGAT) and acyl coenzyme A:monoacylglycerol acyltransferase (MGAT), are capable of carrying out the acyl coenzyme A:retinol acyltransferase (ARAT) reaction. Among them, DGAT1 appears to have the highest specific activity. The apparent K_m values of recombinant DGAT1/ARAT for retinol and palmitoyl coenzyme A were determined to be 25.9 ± 2.1 μM and 13.9 ± 0.3 μM, respectively, both of which are similar to the values previously determined for ARAT in native tissues. A novel selective DGAT1 inhibitor, XP620, inhibits recombinant DGAT1/ARAT at the retinol recognition site. In the differentiated Caco-2 cell membranes, XP620 inhibits ～85% of the Caco-2/ARAT activity indicating that DGAT1/ARAT may be the major source of ARAT activity in these cells. Of the two most abundant fatty acyl retinyl esters present in the intact differentiated Caco-2 cells, XP620 selectively inhibits retinyl–oleate formation without influencing the retinyl–palmitate formation. Using this inhibitor, we estimate that ～64% of total retinyl ester formation occurs via DGAT1/ARAT. These studies suggest that DGAT1/ARAT is the major enzyme involved in retinyl ester synthesis in Caco-2 cells.     

Binding of coenzyme and substrate and coenzyme analogues to 6-phosphogluconate dehydrogenase from sheep liver. An X-ray study at 0.6 nm resolution.

The analogues of the coenzyme NADP+, nicotinamide--8-bromo-adenine dinucleotide phosphate (Nbr8ADP+) and 3-iodopyridine--adenine dinucleotide phosphate (io3PdADP+), were prepared. Nbr8ADP+ was found to be active in the hydrogen transfer adn io3PdADP+ is a coenzyme competitive inhibitor for 6-phosphogluconate dehydrogenase. The binding of NADP+, NADPH and NADPH together with 6-phosphogluconate as well as that of both analogues to crystals of the enzyme 6-phosphogluconate dehydrogenase has been investigated at 0.6-nm resolution using difference electron density maps. The molecules bind in a similar position in a cleft in the enzyme subunit distant from the dimer interface. The orientation of the coenzyme in the site has been determined from the io3PdADP+ -NADP+ difference density. The ternary complex difference density extends beyond that of the nicotinamide moiety of the coenzyme and tentatively indicates substrate binding. No clear identification of the bromine atom of Nbr8ADP+ can be made. However, the analogue is bound more deeply in the cleft than is NADP+. The NADPH density is the most clearly defined and has thus been used to fit a molecular model using an interactive graphics system, checking for preferred geometry. A possible conformation is presented which is significantly different from that of NAD+ in the lactate dehydrogenase ternary complex.     

Regulation of carbon and electron flow in Propionispira arboris: relationship of catabolic enzyme levels to carbon substrates fermented during propionate formation via the methylmalonyl coenzyme A pathway.

A detailed study of the glucose fermentation pathway and the modulation of catabolic oxidoreductase activities by energy sources (i.e., glucose versus lactate or fumarate) in Propionispira arboris was performed. 14C radiotracer data show the CO2 produced from pyruvate oxidation comes exclusively from the C-3 and C-4 positions of glucose. Significant specific activities of glyceraldehyde-3-phosphate dehydrogenase and fructose-1,6-bisphosphate aldolase were detected, which substantiates the utilization of the Embden-Meyerhoff-Parnas path for glucose metabolism. The methylmalonyl coenzyme A pathway for pyruvate reduction to propionate was established by detection of significant activities (greater than 16 nmol/min per mg of protein) of methylmalonyl coenzyme A transcarboxylase, malate dehydrogenase, and fumarate reductase in cell-free extracts and by 13C nuclear magnetic resonance spectroscopic demonstration of randomization of label from [2-13C]pyruvate into positions 2 and 3 of propionate. The specific activity of pyruvate-ferredoxin oxidoreductase, malate dehydrogenase, fumarate reductase, and transcarboxylase varied significantly in cells grown on different energy sources. D-Lactate dehydrogenase (non-NADH linked) was present in cells of P. arboris grown on lactate but not in cells grown on glucose or fumarate. These results indicate that growth substrates regulate synthesis of enzymes specific for the methylmalonyl coenzyme A path and initial substrate transformation.     

GroEL-assisted dehydrogenase folding mediated by coenzyme is ATP-independent.

It has been commonly accepted that GroEL functions as a chaperone by modulation of its affinity for folding intermediates through binding and hydrolysis of ATP. However, we have found that NAD, as a coenzyme of d-glyceraldehyde-3-phosphate dehydrogenase (GAPDH), also stimulates the discharge of GAPDH folding intermediate from its stable complex with GroEL formed in the absence of ATP and assists refolding with the same yield as ATP/Mg(2+) does. The reactivation further increases when ATP is also present, but addition of Mg(2+) has no more effect. NADP, a coenzyme of glucose-6-phosphate dehydrogenase, also releases its folding intermediates from GroEL and increases reactivation. Different from ATP, NAD triggers the release of GAPDH intermediates bound by GroEL via binding with GAPDH itself but not with GroEL, and the released intermediates all folded to native molecules without the formation of aggregation. The collaborative effects of coenzyme and GroEL mediate GroEL-assisted dehydrogenase folding in an ATP-independent way.