Sidney A. Bernhard

On the interaction of the active site of α-chymotrypsin with chromophores: Proflavin binding and enzyme conformation during catalysis

The acridine dye, proflavin, binds firmly to α -chymotrypsin. This binding alters the visible absorption spectrum of the dye. Evidence is presented that the dye binding is stoichiometric with the number of catalytically active sites. The evidence is as follows. (1) Spectrophotometric measurements of adsorption isotherms are consistent with a value of 1·0 adsorption sites/active site. (2) Competitive inhibition of specific substrate reaction occurs in the presence of dye. The enzyme-competitive inhibitor dissociation constant is the same as the dissociation constant determined (in the absence of substrate) from the adsorption isotherms. (3) Stoichiometric titration of the active site with an acylating reagent completely abolishes the dye binding site; removal of the acyl blocking group completely restores the binding site. Non-stereospecific acylating agents can react with the active site—dye complex, but the resultant acylated enzyme cannot bind dye. This finding is presented as a strong argument for a conformational change at the site during catalysis. The change in visible absorption spectrum between dye in water and dye in enzyme site is in accord with the supposed hydrophobic (non-polar) nature of the site. Due to this large change in light absorption at select wavelengths, the dye can be used as a quantitative indicator of the number of otherwise non-complexed catalytic sites, and hence as a tool for the measurement of equilibrium and kinetic (transient) interactions with inhibitors and substrates. Some examples are illustrated.

Subunit conformation and catalytic function in rabbit-muscle glyceraldehyde-3-phosphate dehydrogenase☆☆☆

Abstract This paper describes the functional properties of glyceraldehyde-3-phosphate dehydrogenase which has been chemically modified at the active site. Using the fact that in the tetrameric enzyme only two of the four presumably equivalent thiol groups are readily acylated by substrate, a diacylated dicarboxymethylated enzyme was prepared by reaction of acyl-enzyme with iodoacetate. The acyl groups were then catalytically completely removed. The resulting carboxymethyl-enzyme is presumably a single homogeneous species, potentially distinguishable from carboxymethyl-enzyme prepared by limited alkylation of native (non-acylated) enzyme. However, the specific enzymic activity and number of acylatable active sites were independent of the method of preparation of carboxymethyl-enzyme. Two acyl groups could be removed from diacyl-dicarboxymethyl-enzyme but only one could be re-introduced. These and other properties of the enzyme are explainable in terms of an intramolecular subunit rearrangement within the tetrameric molecule. It is concluded that those models which assume that the unmodified enzyme contains four equivalent sites are inconsistent with the acylation and alkylation results.

Half-site reactivity and the “induced-fit” hypothesis☆

Abstract Many subunit enzymes show the phenomenon of half-site reactivity, that is, the reaction with a substrate or substrate analogue shows a stoichiometry equal to one-half the number of “identical” subunits. One obvious potential explanation for this phenomenon is that reaction with one substrate molecule induces a change in an adjacent subunit, preventing a subsequent substrate molecule from reacting. Such an explanation has been proposed for the half-site reaction of the enzyme cytidine triphosphate synthetase with a substrate analogue (Levitzki et al. , 1971). Analogous studies with glyceraldehyde-3-phosphate dehydrogenase reveal that the four active-site (Cys-149) sulfhydryl groups per tetrameric molecule react equivalently with iodoacetate whereas only two of the four sites undergo facile acylation with the substrate. The simple fact that the 2:1 stoichiometry ratio for the alkylation-acylation reactions is independent of the degree of prior alkylation rules out the ligand-induced asymmetry model as an explanation of the stoichiometries. Rather, it suggests that in muscle dehydrogenase there is a pre-existent non-equivalence among the subunits. On these bases we propose a procedure for distinguishing induced from pre-existent asymmetry in quaternary structure.

Are the structure and function of an enzyme the same in aqueous solution and in the wet crystal

0-[β-(3-indole)acryloyl]-Ser195-α-chymotrypsin [I] is hydrolyzed via specific multifunctional intramolecular catalysis dependent on the specific configuration of the “native” polypeptide. When the polypeptide is in its native configuration the intense ultraviolet-visible absorption spectrum of the chromophoric acylenzyme (ester) is highly distinctive (atypical of indoleacryloyl esters). Cleavage of the indoleacryloyl residue by hydrolysis results in the elimination (at select wavelengths) of the above mentioned intense absorptivity. The absorption spectrum of (I) and its rate of disappearance (as indicated by the decrease in absorptivity) can be determined both in solution and in the wet crystal. Both the distinctive ultraviolet-visible spectrum of native I and its specific rate of catalyzed hydrolysis are independent of phase (solution or crystal). Hence, molecular inferences regarding the catalytic mechanism of enzyme action, derived from crystallographic diffraction studies of acyl-chymotrypsins, are validly extendable to aqueous solution.

On the relationship between the conformation and the catalyzed reactivity of acyl-chymotrypsin

Abstract The stability of acyl-chymotrypsins, in particular O-[β-(3-indole)acryloyl]-Ser195-α-chymotrypsin, towards (aqueous) solvent denaturants has been studied by examining the composition of final products and the kinetics of their formation following incubation in particular denaturing solvents. Near neutral pH, the originally “native” acylenzyme might, a priori, be anticipated to undergo either intramolecular catalytic hydrolysis yielding carboxylate anion, or denaturation yielding the unreactive “denatured” acylenzyme. Near neutral pH, both these processes occur at relatively low velocity (half-lives of the order of minutes or hours) in solvents in which denaturation of “native” unsubstituted enzyme is a rapid process (half-lives of the order of milliseconds or seconds). The partitioning of product between carboxylate anion (the catalytic product) and denatured acylenzyme is a function of the composition of the particul r denaturing solvent. Over a restricted but catalytically significant pH range, the total rate of product formation is unaffected by the particular (solvent-dependent) partitioning ratio, suggesting a common rate-determining intermediate for catalysis and for denaturation. Changes in the catalytic velocity (vp), which can be effected either by the addition of competitive nucleophiles (such as hydroxylamine), or by the introduction of acyl derivatives with different reactivities, lead to corresponding changes in the velocity of denaturation (vp) and hence have little effect on the partitioning ratio ( v p v d ). This latter result indicates that a covalently modified acylenzyme-nucleophile complex is an obligatory intermediate in both catalysis and denaturation of the acylenzyme, and that catalysis and (variable) conformation of the polypeptide are intimately related. The nature of the structural and chemical changes occurring during catalysis are considered.

On the function of half-site reactivity: intersubunit NAD+-dependent activation of acyl-glyceraldehyde 3-phosphate dehydrogenase reduction by NADH.

Glyceraldehyde 3-phosphate dehydrogenase extracted from sturgeon muscle, exhibits half-site reactivity in its reaction with the pseudo-substrate β -(2-furyl) acryloyl phosphate ( Malhotra & Bernhard, 1968 ). The product is the difurylacryloyl thiol ester enzyme tetramer formed from the active site cysteinyl residue. The electronic spectrum of the furylacryloyl thiol ester linkage is perturbed upon binding of oxidized coenzyme (NAD + ) at the acyl site ( Malhotra & Bernhard, 1973 ). Likewise, we now report the perturbation in electronic spectrum of this furylacryloyl thiol ester upon interaction with NADH at the acyl site. The spectral perturbation accompanying NADH binding is large and in the opposite direction from that due to the binding of NAD + . The “color” of the chromophoretic acyl-enzyme linkage is a reflection of its chemical properties: furylacryloyl-enzyme-NAD complex is “red-shifter” (λ max =360 nm), and reacts uniquely with inorganic phosphate to yield the corresponding acyl phosphate. Furylacryloyl-enzyme-NADH complex is “blue-shifter” (λ max =330 nm) and is uniquely capable of reduction by NADH. Despite the presence of a ubiquitous NADH-induced color change, NADH reduction occurs only when NAD is bound to the remaining two non-acylated subunits. Hence, these results provide a direct example of allosteric regulation of enzyme function by ligand interaction at an adjacent subunit site. This site is known to be at least 17 A removed from the site of reduction ( Buehner et al. , 1974 ). Carboxymethylation of the two remaining (active) thiol groups of the diacyl-enzyme tetramer, a process which modifies the affinity for NAD at this site to only a small extent, does not affect the NAD-dependent activation of the acyl-NADH site for reduction. Qualitative transient kinetic evidence suggests that NAD binding to the non-acylated sites favors a slow conformational change leading to reducibility at acyl sites. The affinity of the various sites, acylated and non-acylated, for coenzyme ligands varies with the extent of acylation and the extent and nature of the ligand occupancies at adjacent subunits. Since the molar concentration of glyceraldehyde 3-phosphate dehydrogenase sites in muscle sarcoplasmic fluid is high (of the order of 1 pm), these results suggest that the “enzyme” functions as a regulatory warehouse for energy via acyl-enzyme-ATP interconversion and NADH/NAD redox potential regulation.

The mechanism of succinate or fumarate transfer in the tricarboxylic acid cycle allows molecular rotation of the intermediate

Mitochondria were incubated with L[5-13C]glutamic acid and the distribution of the label between the two carboxyl carbon atoms of the L-aspartic acid formed was determined by 13C NMR. The reaction sequence leading from L-glutamic acid to L-aspartic acid spans the tricarboxylic acid cycle reactions involving the two symmetrical intermediates succinate and fumarate. The C2 symmetry of these intermediates in principle permits a discrimination of the mechanism of their transfer between their enzyme sites of production and utilization. A direct transfer of metabolite from site to site by translation alone predicts an unequal distribution of 13C between the C1 and C4 of aspartate, whereas molecular rotation during transfer allows for a scrambling of the original C5 label. Under several conditions of different glutamate concentrations and solvent osmotic pressures, equal labeling in the C1 and C4 carbons of aspartate is observed. This observation is inconsistent with a transfer mechanism restricting molecular rotation for both intermediates but is compatible with both a random diffusion and a direct transfer mechanism provided the latter allows molecular rotation.

Automatic determination of amino acid sequences

A fundamental problem for biochemistry is the determination of the linear sequence of amino acids in proteins. This paper describes a computer-oriented logic for obtaining such determination. The logic applies successively stronger decision rules to extract the required information on the protein sequence.

The intracellular equilibrium thermodynamic and steady-state concentrations of metabolites.

A new model for the organization and flow of metabolites through a metabolic pathway is presented. The model is based on four major findings. (1) The intracellular concentrations of enzyme sites exceed the concentrations of intermediary metabolites that bind specifically to these sites. (2) The concentration of the excessive enzyme sites in the cell is sufficiently high so that nearly all the cellular intermediary metabolites are enzyme-bound. (3) Enzyme conformations are perturbed by the interactions with substrates and products; the conformations of enzyme-substrate and enzyme-product complexes are different. (4) Two enzymes, catalyzing reactions that are sequential in a metabolic pathway, transfer the common metabolite back and forth via an enzyme-enzyme complex without the intervention of the solvent environment. The model proposes that the enzyme-enzyme recognition is ligand-induced. Conversion of E2S and E2P results in the loss of recognition of E2 by E1 and the concomitant recognition of E2 by E3. This model substantially alters existent views of the bioenergetics and the kinetics of intracellular metabolism. The rates of direct transfer of metabolite from enzyme to enzyme are comparable to the rates of interconversion between substrate and product within an individual enzyme. Consequently, intermediary metabolites are nearly equipartitioned among their high-affinity enzyme sites within a metabolic pathway. Metabolic flux involves the direct transfer of metabolite from enzyme to enzyme via a set of low and nearly equal energy barriers.

NUCLEOPHILIC DISPLACEMENT REACTIONS AT ESTER AND THIOLESTER BONDS

This paper deals with the influence of the environment on the rates and equilibria of nucleophilic displacement reactions at a carbonyl carbon atom. We consider three kinds of environmental perturbations: (1) “solvent” environment, which influences the polarity of the carbonyl group and hence its interaction with a polar or charged nucleophile; (2) covalently adjacent functional groups, whose electronic properties affect the reactivity at the carbonyl carbon; (3) protein conformation surrounding a carbonyl group, which can influence reactivity either via environmental perturbations as in (1) or (2), or by introducing strain, and consequent reactivity, at a carbonyl bond. Of particular interest to us are enzymic reactions with substrates involving the formation of metastable acyl enzymes, since some of these reactions lead to metastable thiolesters. Complementary studies with small model peptides sometimes involve the transient formation of a conventionally “high energy” acyl bond as a metastable intermediate in the pathway of reaction, as in the formation of acyl phosphates and thiolesters from acyl esters (vide infra). The transient and equilibrium thermodynamic properties of transient acyl derivatives are discussed in the following section.