Coenzyme Phosphate Research Articles

Probing the Role of Tyr 64 of Treponema denticola Cystalysin by Site-Directed Mutagenesis and Kinetic Studies

Tyr 64, hydrogen-bonded to coenzyme phosphate in Treponema denticola cystalysin, was changed to alanine by site-directed mutagenesis. Spectroscopic and kinetic properties of the Tyr 64 mutant were investigated in an effort to explore the differences in coenzyme structure and kinetic mechanism relative to those of the wild-type enzyme. The wild type displays coenzyme absorbance bands at 418 and 320 nm, previously attributed to ketoenamine and substituted aldamine, respectively. The Tyr 64 mutant exhibits absorption maxima at 412 and 325 nm. However, the fluorescence characteristics of the latter band are consistent with its assignment to the enolimine form of the Schiff base. pK(spec) values of approximately 8.3 and approximately 6.5 were observed in a pH titration of the wild-type and mutant coenzyme absorbances, respectively. Thus, Tyr 64 is probably the residue involved in the nucleophilic attack on C4' of pyridoxal 5'-phosphate (PLP) in the internal aldimine. Although the Tyr 64 mutant exhibits a lower affinity for PLP and lower turnover numbers for alpha,beta-elimination and racemization than the wild type, the pH profiles for their Kd(PLP) and kinetic parameters are very similar. Rapid scanning stopped-flow and chemical quench experiments suggest that, in contrast to the wild type, for which the rate-determining step of alpha,beta-elimination of beta-chloro-L-alanine is the release of pyruvate, the rate-determining step for the mutant in the same reaction is the formation of alpha-aminoacrylate. Altogether, these results provide new insights into the catalytic mechanism of cystalysin and highlight the functional role of Tyr 64.

Affinity assembled multilayers for new dehydrogenase biosensors

We propose a novel approach, which allows the control of the spatial arrangement of redox mediator, coenzyme and enzyme on the electrode at a molecular level, using essentially electrostatic interactions. The first step consists of adsorbing a monolayer of molecules out of a new family of redox mediators, substituted nitrofluorenones. In a second step, a monolayer of calcium cations is immobilized at the interface. It serves as a bridge between the redox mediator and the subsequently adsorbed coenzyme. The weak interaction between a carboxyl group of the redox mediator and the coenzyme's phosphate groups, revealed by QCM measurements, allows the coenzyme to keep its natural activity in the adsorbed state. In the last step, we use the intrinsic affinity of this monolayer of NAD + for dehydrogenases to build up a supramolecular sandwich composed of mediator/Ca 2+/NAD +/dehydrogenase. This simple modification procedure, which might constitute a versatile approach for the low cost assembly of well-defined biosensors surfaces, has been successfully applied to the enzymatic detection of glucose, glutamate and alcohol.

Electrodes modified with nitrofluorenone derivatives as a basis for new biosensors

We report about the electrocatalytic properties of electrodes modified by adsorption of nitro-fluorenone derivatives. The stable, adherent monolayer of these catalyst precursors can be transformed electrochemically into the corresponding hydroxylamine compounds (R-NO 2+4e+4H +⇒R-NHOH+H 2O). The completely reversible two electron oxidation of the hydroxylamine leads to the nitroso compounds (R-NHOH⇒R-NO+2e+2H +) that exhibit high catalytic activity in the electrooxidation of NADH at low overpotentials (−30 mV vs. Ag/AgCl) and therefore constitute a new family of efficient redox mediators for biosensor applications. A significant increase in catalytic activity (up to 500%) is observed after addition of calcium ions to the electrolyte. This is explained by a specific and bridging complexation between the coenzyme's phosphate groups and a carboxyl group present in the catalyst molecule. The interaction favours the contact between NADH and the surface confined catalyst, leading to a higher electron transfer efficiency. This interaction can be used in an approach of molecular level design for controlled monolayer deposition of catalyst, Ca 2+, NAD + and enzyme. A very simple and inexpensive modification scheme, essentially based on electrostatic attraction, leads to electrodes that can be employed as reagentless biosensors for the electrochemical detection of common and commercially interesting analytes like glucose.

Plasticity of the Tryptophan Synthase Active Site Probed by31P NMR Spectroscopy

The functional properties of tryptophan synthase alpha2beta2 complex are modulated by a variety of allosteric effectors, including pH, monovalent cations, and alpha-subunit ligands. The dynamic properties of the beta-active site were probed by 31P NMR spectroscopy of the enzyme-bound coenzyme pyridoxal 5'-phosphate. The 31P NMR signal of the cofactor phosphate of the internal aldimine exhibits a single peak at 3.73 ppm with a line width of 12 Hz. In the presence of saturating concentrations of sodium ions, the 31P signal shifts to 3.97 ppm concomitant with a change in line width to 35 Hz. The latter indicates that sodium ions decrease the conformational flexibility of the coenzyme. In the absence of ions, lowering pH leads to the appearance of a second peak at 4.11 ppm, the intensity of which decreases in the presence of cesium ions. Addition of L-serine in the presence of sodium ions leads to the formation of the external aldimine, the first metastable catalytic intermediate. The 31P signal does not change its position, but a change in line width from 35 to 5 Hz is observed, revealing that this species is characterized by a considerable degree of rotational freedom around the coenzyme C-O bond. In the presence of L-serine and either cesium ions or the allosteric effector indole-3-acetylglycine, the accumulation of the second catalytic intermediate, alpha-aminoacrylate, is observed. The 31P signal is centered at 3.73 ppm with a line width of 5 Hz, indicating that the phosphate group of the coenzyme in the external aldimine and the alpha-aminoacrylate exhibits the same flexibility but a slightly different state of ionization. Because the alpha-aminoacrylate intermediate but not the external aldimine triggers the allosteric signal to the alpha-subunit, other portions of the beta-active site modify their dynamic properties in response to the progress of the catalytic process. A narrow line width was also observed for the quinonoid species formed by nucleophilic attack of indoline to the alpha-aminoacrylate. The 31P signal moves downfield to 4.2 ppm, indicating a possible change of the ionization state of the phosphate group. Thus, the modification of either the ionization state of the coenzyme phosphate or its flexibility or both are, at least in part, responsible for the conformational events that accompany the catalytic process.

Microscopic protonation equilibria and solution conformations of coenzyme A and coenzyme A disulfides

Microscopic acid dissociation constants have been determined for the protonated N1 nitrogen of the adenine ring and the 3'-phosphate and thiol groups of coenzyme A (CoASH) and the protonated N1 nitrogen and 3'-phosphate groups of symmetrical coenzyme A disulfide (CoASSCoA) and coenzyme A-glutathione mixed disulfide (CoASSG) from chemical shift vs pD data. The microscopic constants for the N1 nitrogen and 3'-phosphate groups in the three compounds are essentially identical

Kinetics and equilibria for the reactions of coenzymes with wild type and the Y70F mutant of Escherichia coli aspartate aminotransferase

The Y70F mutant of aspartate aminotransferase has reduced affinity for coenzymes compared to the wild type. The equilibrium dissociation constants for pyridoxamine phosphate (PMP) holoenzymes, KPMPdiss, were determined from the association and dissociation rate constants to be 1.3 nM and 30 nM for the wild type and mutant, respectively. This increase in KPMPdiss for Y70F is due to a 27-fold increase in the dissociation rate constant. Pyridoxal phosphate (PLP) association kinetics are complex, with three kinetic processes detectable for wild type and two for Y70F. A directly determined, accurate value of KPLPdiss for wild type enzyme has been difficult to obtain because of the low value of this constant. The values of KPLPdiss for the holoenzymes were determined indirectly through the measured values for KPMPdiss, glutamate-alpha-ketoglutarate half-reaction equilibrium constants, and the equilibrium constant for the transamination of PLP by glutamate catalyzed by Y70F. The values of KPLPdiss obtained by this procedure are 0.4 pM for wild type and 40 pM for Y70F. The increases in KPMPdiss and KPLPdiss for Y70F correspond to delta delta G values of 1.9 and 2.7 kcal/mol, respectively, and are directly attributed to the loss of the hydrogen bond from the phenolic hydroxyl group of Tyr70 to the coenzyme phosphate. The delta G for association of PLP with wild type enzyme is 4.7 kcal/mol more favorable than that for PMP.

Escherichia coli maltodextrin phosphorylase: contribution of active site residues glutamate-637 and tyrosine-538 to the phosphorolytic cleavage of .alpha.-glucans

The role of Escherichia coli maltodextrin phosphorylase (EC 2.4.1.1) active site residues Glu637 and Tyr538 which line the sugar-phosphate contact region of the enzyme was investigated by site-directed mutagenesis. Substitution of Glu637 by an Asp or Gln residue reduced kcat to approximately 0.2% of wild-type activity, while the Km values were affected to a minor extent. This indicated participation of Glu637 in transition-state binding rather than in ground-state binding. 31P NMR analysis of the ionization state of enzyme-bound pyridoxal phosphate suggested that Glu637 is also involved in modulation of the protonation state of the coenzyme phosphate observed during catalysis. Despite loss of proposed hydrogen-bonded substrate contacts, the Tyr538Phe mutant enzyme retained more than 10% activity; the apparent affinity of all substrates was slightly decreased. Mutations at either site affected the error rate of the enzyme (ratio of hydrolysis/phosphorolysis). Besides a role in substrate binding, the hydrogen-bond network of Tyr538 supports the exclusion of water from the active site.

Tyrosine 70 increases the coenzyme affinity of aspartate aminotransferase. A site-directed mutagenesis study.

The crucial step in enzymatic transamination is the tautomerization of aldimine/ketimine intermediates, formed between the pyridoxyl coenzyme and the amino/keto acid substrate, which is catalyzed primarily by the active site residue Lys-258 (Malcolm, B. A., and Kirsch, J. F. (1985) Biochem. Biophys. Res. Commun. 132, 915-921; W. L. Finlayson and J. F. Kirsch, in preparation). Tyr-70 is localized in close proximity to Lys-258 and, in addition, forms a hydrogen bond with the coenzyme phosphate. Tyr-70 has been postulated to have an important role in the tautomerization (Kirsch, J. F., Eichele, G., Ford, G. C., Vincent, M. G., Jansonius, J. N., Gehring, H., and Christen, P. (1984) J. Mol. Biol. 174, 497-525). This hypothesis has now been tested by the construction and analysis of a mutant Escherichia coli aspartate aminotransferase in which Tyr-70 has been changed to Phe (Y70F). Y70F retains at least 15% of the maximal activity of the wild type enzyme (WT) (kcat = 170 +/- 15 s-1 for WT versus greater than or equal to 26 +/- 3 s-1 for Y70F and shows increased Michaelis constants for both substrates (KmAsp = 2.5 +/- 0.4 mM; Km alpha Kg = 0.59 +/- 0.08 mM for WT versus KmAsp = 3.9 +/- 0.3 mM; Km alpha Kg = 2.70 +/- 0.02 mM for Y70F (where alpha Kg is alpha-ketoglutarate) ). The spectrophotometrically determined pK a values of the internal aldimines formed between pyridoxal 5'-phosphate (PLP) and Lys-258 are identical for WT and Y70F. In assays where excess L-aspartate and excess PLP are incubated with either WT or Y70F, the mutant enzyme converts the free PLP to free pyridoxamine 5'-phosphate 80-fold faster than WT (k = (3.75 +/- 0.23) X 10(-2)s-1 for Y70F versus (4.90 +/- 0.02) X 10(-4)s-1 for WT). Y70F also converts free pyridoxamine 5'-phosphate to free PLP faster than WT. Thus, Y70F dissociates coenzyme more readily than does WT. It therefore appears that the role of Tyr-70 is mainly in preventing the dissociation of the coenzyme from the enzyme. Tyr-70 does not function in an essential chemical step.

31P NMR relaxation studies of the activation of the coenzyme phosphate of glycogen phosphorylase. The role of motion of the bound phosphate

Spin-lattice and spin-spin relaxation rates (1/T1 and 1/T2) have been determined for the catalytically essential coenzyme phosphate at the active site of glycogen phosphorylase in both activated (R state) and inactive (T state) conformations of the enzyme. Dipolar contributions to 31P relaxation due to exchangeable protons on the phosphate group have been determined by measurement of relaxation rates at different concentrations of H2O and D2O, and field dependence studies have been performed to estimate the contribution of chemical shift anisotropy to the remaining 31P relaxation in D2O. At 109 MHz, dipolar relaxation from exchangeable protons was found to account for 50% of the spin-lattice relaxation for activated phosphorylase in 75% H2O, the remainder being due to chemical shift anisotropy. The spin-lattice relaxation rates in D2O for R-state glycogen phosphorylase are very similar to those measured for other proteins of very different size such as actin (Brauer, M., and B. D. Sykes, 1981, Biochemistry. 20:6767-6775), alkaline phosphatase (Coleman, J. E., I. D. Armitage, J. F. Chlebowski, J. D. Otvos, and A. J. M. S. Uiterkamp, 1979), and phosphoglucomutase (Rhyu, G. I., W. J. Ray, Jr., and J. L. Markley, 1984, Biochemistry. 23:252-260). In inactive (T state) phosphorylase the spin-lattice relaxation rates were almost an order of magnitude slower, while the spin-spin relaxation rates were essentially identical. These results have been analyzed by calculating the theoretically expected 31P relaxation rates in the presence of internal motions that are included in the relaxation calculation using the model-free approach of Lipari and Szabo (1982, J. Am. Chem. Soc. 104:4564-4559). The analysis suggests the coenzyme phosphate is relatively immobilized in the activated enzymic conformation, but in the inactive (Tstate) conformation it is considerably more mobile with a rotational correlation time one to two orders of magnitude smaller. Since the spin-lattice relaxation rate for the active R-state (immobilized) phosphate is similar to that observed in other phosphoenzymes of different size it is suggested that a librational motion on the nanosecond time scale may constitute a common spin-lattice relaxation pathway for phosphates in macromolecules. The consequences of phosphate motion in terms of recent suggestions concerning the environment and the catalytic role of the coenzyme phosphate are discussed.

Ligand binding and stabilization of malate- and lactate dehydrogenase.

Binding of coenzymes, coenzyme fragments and phenolate ligands to malate- and lactate dehydrogenase was studied. From linear competition in titration experiments, the coenzyme binding site was concluded to bind all the ligands employed. The analogy between the phenolate ligands and tetraiodofluorescein which is known to bind at the adenosine binding site suggests binding of phenolates at this site. Coenzymes and coenzyme fragments retard the irreversible thermal inactivation of the enzymes. The retardation effect decreases in the order NADH greater than NAD greater than ADPR greater than or equal to AMP for both enzymes. Phenolate ligands binding to the adenosine pocket do not stabilize the enzymes. The stabilization is concluded to originate from the interaction of coenzyme phosphate and nicotinamide with the enzymes. The interactions with the adenosine moiety and with the second ribose seem to be ineffective in retardation of thermal denaturation.

Pyridoxal(5')diphospho(1)-alpha-D-glucose. A potent R-state inhibitor of glycogen phosphorylase.

Pyridoxal(5')diphospho(1)-alpha-D-glucose has been tested as an inhibitor of native glycogen phosphorylases a and b. Its inhibition patterns with respect to substrate, glucose 1-phosphate, and activator, adenosine monophosphate, show it to be a potent (Ki = 40 microM) R-state inhibitor of phosphorylase b, mimicking the binding of glucose-1-phosphate, and, as predicted for an R-state inhibitor, its binding to AMP-activated phosphorylase a is even tighter (Ki = 10 microM). Moreover, it is demonstrated that its binding does not involve covalent imine formation from the pyridoxal aldehyde to an active-site lysine residue. It thus represents the tightest binding R-state inhibitor reported to date, and a 31P NMR study of the effects of binding of this inhibitor upon 31P resonances for the coenzyme phosphate and that of the nucleotide activator is presented. Results obtained are essentially identical to those obtained previously using glucose cyclic 1,2-phosphate, corroborating the previous conclusions. A rationale for the tightness of the binding is presented, as are other possible uses of this compound in studies on glycogen phosphorylase and other similar enzymes.

Pyridoxal phosphate is not the acid catalyst in the glycogen phosphorylase catalytic mechanism

Pyridoxal-reconstituted phosphorylase has been shown previously (Parrish, R.F., Uhing, R.J. and Graves, D.J. (1977), Biochemistry 16 , 4824) to be activated to a very similar extent by phosphite (pK a=6.6) and fluorophosphate (pK a=4.8). This paper shows that the pH dependence of the enzyme activity is also similar for these two activators. This is interpreted to mean that the coenzyme phosphate is not involved as an essential acid catalyst in the actual reaction mechanism and is probably not involved in any form of proton transfer process. The pH/activity profile for native phosphorylase b in the direction of glycogen synthesis is also presented and a comparison of the two sets of profiles is discussed.

Catalytic mechanism of glycogen phosphorylase: pyridoxal(5')diphospho(1)-alpha-D-glucose as a transition-state analogue.

Pyridoxal(5')diphospho(1)-alpha-D-glucose was used to reconstitute glycogen phosphorylase beta (1,4-alpha-D-glucan:orthophosphate alpha-D-glucosyltransferase, EC 2.4.1.1) from rabbit muscle, replacing the natural pyridoxal 5'-phosphate coenzyme. Incubation of the reconstituted enzyme alone resulted in the gradual cleavage of the synthetic cofactor to pyridoxal 5'-phosphate, which caused slow reactivation of the enzyme. The addition of maltopentaose or glycogen altered the mode of cleavage; the cofactor was rapidly decomposed to pyridoxal 5'-diphosphate. The radioactive glucose moiety released from pyridoxal(5')diphospho(1)-alpha-D-[14C]glucose was incorporated into the outer chain of glycogen, forming an alpha-1,4-glucosidic linkage. These results show that the glucosyl transfer reaction discovered mimics the normal catalysis of this enzyme, and they strongly support the catalytic mechanism in which the coenzyme phosphate acts as a catalyst by direct interaction with the phosphate of the substrate, forming the pyrophosphate-like transition intermediate.

Covalently bound non-coenzyme phosphorus residues in flavoproteins: 31 P nuclear magnetic resonance studies of Azotobacter flavodoxin

In addition to the 5'-phosphate ester on its flavin mononucleotide (FMN) moiety, flavodoxin from Azotobacter vinelandii contains 2 moles of tightly bound phosphate. One non-coenzyme phosphate group is covalently bound to the protein, as it remains with the protein on acid precipitation, whereas the other phosphate is released. The invariance of the (31)P nuclear magnetic resonance chemical shift of the covalently bound phosphate (-0.8 ppm relative to 85% phosphoric acid) with pH, even in the presence of protein denaturants, implies it is in a diester linkage to the protein. Because no evidence could be found for the presence of covalently bound sugars, nucleotides, or phospholipids, it is suggested that the phosphate residue forms a diester linkage with two hydroxyl amino acids in the protein. The only other suggestion of a phosphodiester linkage in proteins is from previous studies on pepsin and pepsinogen [Perlmann, G. E. (1955) Adv. Prot. Chem. 10, 1-30]. The observed changes in (31)P chemical shift with pH show that the covalent phosphorus in pepsinogen has ionization properties of a monoester rather than a diester. The (31)P resonance of the FMN phosphate occurs at -5.6 ppm in native Azotobacter flavodoxin. No ionization of the protein-bound FMN phosphate is observed since the chemical shift does not change appreciably in the pH range of 5.5-9.5. The chemical shift data suggest, but do not prove, that the coenzyme phosphate in its protein-bound form is dianionic. Chemical analysis of several other flavoenzymes from a variety of sources shows the presence of covalently bound phosphorus in quantities stoichiometric with the flavin content in most of the enzymes tested. Thus, the presence of covalent phosphorus in flavoenzymes may be a general phenomenon with currently unknown catalytic significance.

Alpha-Isopropylmalate synthase from Alcaligenes eutrophus H 16. II. Substrate specificity and kinetics.

The purified isopropylmalate synthase of Alcaligenes eutrophus H 16 reacted with the following alpha-keto acids and acyl-coenzyme A derivatives (in the sequence of decreasing affinities): alpha-ketoisovalerate, alpha-keto-n-valerate, alpha-ketobutyrate and pyruvate; acetyl-CoA, propionyl-CoA, butyryl-CoA, malonyl-CoA, valeryl-CoA, and crotonyl0CoA. alpha-Ketoisocaproate, however, is a strong inhibitor of the enzyme. All reactions catalyzed by isopropylmalate synthase were inhibited to the same extent by the endproduct L-leucine. The substrate saturation curves of alpha-ketoisovalerate or other alpha-keto acids and of acetyl-coenzyme A or other acyl-CoA derivatives had intermediary plateau regions; the Hill coefficient alternated between nH-values higher and lower than 1.0, indicating changes from positive to negative and from negative to positive cooperativity for the substrates. The products, isopropylmalate and free coenzyme A, showed competitive inhibition patterns against both substrates (alpha-ketoisovalerate and acetyl-CoA). Free coenzyme A (1 micronM) inactivated the enzyme irreversibly. The 3'-phosphate of coenzyme A and the free carboxyl group of alpha-ketoisovalerate were involved in optimal binding of these substrates, but 3'-dephospho-acetyl-coenzyme A and the methylester of alpha-keto-isovalerate were also converted by this enzyme. A CH3--CH2-grouping of the alpha-keto acids seemed to be necessary for binding this substrate.

Carboxymethylation of horse-liver alcohol dehydrogenase in the crystalline state. The active-site zinc region and general anion-binding site of the enzyme correlated in primary and teritiary structures.

Horse liver alcohol dehydrogenase (isozyme EE) in the crystalline state was alkylated with iodoacetate under conditions resulting in the single substitution of Cys-46, which is a ligand to the active-site zinc atom. Alkylation was facilitated by the prior formation of a complex with imidazole bound to the zinc atom. Extent and specificity of the reaction were determined by use of 14C-labelled iodoacetate and by analyses of radioactive peptides after cleavage with trypsin. Ternary complexes of the enzyme with coenzymes and inhibitors effectively protected the protein against alkylation. ADP-ribose, Pt(CN)2-/4 , 1,10-phenanthroline, Au(CN)-/2 and AMP also prevented alkylation with decreasing effectiveness. Crystallographic studies of the alkylated enzyme show that the carboyxmethylated sulfur atom of Cys-46 is still liganded to the active-site zinc atom and that the iodide ion liberated during alkylation is bound as the fourth ligand to zinc, displacing imidazole. Crystallographic analyses were also performed of the binding of AMP and Pt(CN2-/4 to the enzyme. It was found that Arg-47 interacts with the phosphate moiety of the nucleotide. Lys-228 and Arg-47 interact in the platinate complex with the bulky anion, the center of which coincides with the position of the nucleotide phosphate. Some of the cyano-ligands to platinum occupy a crevice between the coenzyme phosphate binding site and the active-site zinc atom. The results of the combined studies on primary and tertiary structures confirm previous suggestions that iodoacetate enters the active site via reversible binding to an anion-binding site. This site interacts with the negatively charged groups of the coenzyme as well as with ADP-ribose, Pt(CN2-/4 and to a lesser extent Au(CN)-/2 and AMP, which therefore prevent the reversible binding of iodoacetate. 1,10-Phenanthroline does not block the binding site but interferes with alkylation presumably by changing the coordination of zinc. Identificationof this labelled residue in both chemical and crystallographic studies correlates the primary and tertiary structures. Characterizations of the active-site zinc region and the general anion-binding site are also presented.

The Binding of Coenzymes and Analogues of the Substrate-Coenzyme Complex to Tyrosine Aminotransferase

The interaction of tyrosine apo-aminotransferase with coenzymes and coenzyme derivatives has been studied. The derivatives include analogues of substrate-coenzyme complexes formed in the course of the transamination (pyridoxyl derivatives) and coenzyme analogues (pyridoxine 5′-phosphate, 4′-deoxypyridoxine 5′-phosphate, 1-methyl-pyridoxal 5′-phosphate. 1-methyl-pyridoxamine 5′-phosphate). From a comparison of the behaviour of the different compounds, the ΔGo values for the binding of the coenzyme phosphate group and of the substrate (tyrosine) carboxyl and phenyl groups have been determined as equal to -7.1, -3.0 and -2.7 kcal/mol (-29.7, -12.5 and -11.3 kJ/mol) respectively. In the binding of the substrate to the enzyme a significant fraction of the intrinsic ΔGo appears to be used up for some associated endoergonic process. A comparison of the interaction between the enzyme and pyridoxamine 5′-phosphate and some of its derivatives has shown that a positive charge and a large substituent at position 4′ of the coenzyme have an adverse effect on the binding. Methylation of position 1 of pyridoxal 5′-phosphate and pyridoxamine 5′-phosphate makes their ΔGo of binding more positive by 4.0 and 1.8 kcal/mol (16.7 and 7.5 kJ/mol) respectively. ΔHo and ΔSo of binding for the different pyridoxyl-derivatives showed a much greater variability than ΔGo; ΔHo and ΔSo appear to be very sensitive indicators of the correct alignment of the substrate-coenzyme complex at the active site. On the basis of these results a hypothesis on the mode binding of the pyridoxyl-amino acids is presented, which is compatible with the scheme of transamination proposed earlier for aspartate transaminase by Braunstein and by Ivanov and Karpeisky.