The Effect of Aliphatic Alcohols and Organic Solvents on Reactions Catalyzed by 5-HydroxyJV-

SHydroxy-N-methylpyroglutamate synthetase (HMPG synthetase) catalyzes hydrolysis and acyl transfer reactions with d-substituted a-ketoglutarates, presumably through the formation of an cY-ketoglutaryl enzyme intermediate. With cr-ketoglutaramate as the substrate, maximal velocities of l&26, and 36 are observed for acyl transfer to water, methylamine, and ethanolamine, respectively. The corresponding maximal velocities with ethyl cr-ketoglutarate as the acyl donor are 19, 38, and 68. These results suggest that both acylation and deacylation are partially rate limiting for reactions involving ar-ketoglutaramate and ethyl oc-ketoglutarate. Aliphatic alcohols increase the rate of hydrolysis of cr-ketoglutaramate and ethyl or-ketoglutarate. These rate accelerations increase with increasing chain length of the alcohol until a maximal value is reached and are interpreted as a specific effect on the rate of deacylation of the acyl enzyme. Based on this assumption, the rate of transfer of the acyl group from the enzyme to water, to methylamine, and to ethanolamine has been calculated and found to be independent on the nature of the acyl donor. In contrast to the effects observed on hydrolysis, aliphatic alcohols act as competitive inhibitors with respect to amines in acyl transfer reactions and thus provide evidence for an amine-binding site on the enzyme. Organic solvents increase both hydrolysis and transfer reactions involving amines; however, these effects can be distinguished from those of alcohols by several kinetic criteria.

Aliphatic alcohols increase the rate of hydrolysis of cr-ketoglutaramate and ethyl or-ketoglutarate. These rate accelerations increase with increasing chain length of the alcohol until a maximal value is reached and are interpreted as a specific effect on the rate of deacylation of the acyl enzyme.
Based on this assumption, the rate of transfer of the acyl group from the enzyme to water, to methylamine, and to ethanolamine has been calculated and found to be independent on the nature of the acyl donor. In contrast to the effects observed on hydrolysis, aliphatic alcohols act as competitive inhibitors with respect to amines in acyl transfer reactions and thus provide evidence for an amine-binding site on the enzyme.
Organic solvents increase both hydrolysis and transfer reactions involving amines; however, these effects can be distinguished from those of alcohols by several kinetic criteria.
The enzyme, 5-hydroxy-N-methylpyroglutamate synthetase, catalyzes hydrolysis and acyl transfer reactions with 6-substi- Previous studies on the mechanism of the reaction (2,3) led to the conclusion that the reaction proceeds through the formation of an Lu-ketoglutaryl enzyme intermediate. During the course of investigating the substrate specificity of the enzyme, we noted significant increases in the rate of hydrolysis of ethyl cr-ketoglutarate in the presence of alcohols and organic solvents.
The present paper describes a detailed investigation into the nature of these rate accelerations.

MATERIALS AND METHODS
Mute&&--cu-Ketoglutaramate, &ethyl a-ketoglutarate, and s-methyl a-ketoglutarate were prepared by oxidation of the corresponding glutamic acid derivative with L-amino acid oxidase (2,4). Amines, alcohols, and organic solvents were recrystallized or redistilled before use. Bovine liver glutamate dehydrogenase (ammonium free) was obtained from Sigma Chemical Co. [i4C]Ethanolamine was obtained from Amersham-Searle Corp. 5-Hydroxy-N-methylpyroglutamate synthetase was prepared as previously described (2).
The hydrolysis of the &amide and esters of a-ketoglutarate were routinely measured spectrophotometrically by coupling the reaction to the glutamate dehydrogenase reaction (2) and following the oxidation of DPNH at 340 nm. Except when noted, all kinetic runs were performed at 30". The pH of reaction mixtures was measured at the end of each kinetic run and was found not to deviate more than ~0.05 pH units from the desired pH. The extinction coefficient of DPNH does not significantly change in the presence of alcohols or organic solvents.
The K, for ethyl a-ketoglutarate (3.8 x 10m6 M) (3) is too low to permit a detailed kinetic study by the spectrophotomet,ric assay. Alternatively, we used the glutamate dehydrogenase coupled assay but followed DPNH oxidation fluorometrically on an Eppendorf fluorometer, as previously described (3). ru'o difference was observed for DPNH standard curves prepared in buffered reaction mixtures containing alcohols or organic solvent,s as compared to standard curves prepared in buffer alone.
Transfer reactions between 14C-amines and cr-ketoglutaramate or ethyl cr-ketoglutarate were measured by the Dowex 50-H+ chromatographic assay previously described (2). The transfer reaction between higher molecular weight amines and ethyl a-ketoglutarate was assayed by measuring the disappearance of the ester. Reaction mixtures containing 50 mM potassium Triciner buffer, pH 8.6, 2 mM ethyl oc-ketoglutarate, amine and enzyme, in a final volume of 1.5 ml were incubated 1  Reaction mixtures contained 50 mM potassium tricine buffer, pH 8.0, 10 mM ammonium chloride, 1.5 mM dithiothreitol, 0.128 PM DPNH, 100 pg of glutamate dehydrogenase, 0.45 pg of enzyme, and alcohols or organic solvents as indicated in a final volume of 1.0 ml. The reaction was initiated by the addition of enzyme and followed on a Gilford recording spectrophotometer at 30". The maximal velocity for the hydrolytic reactions in the presence of organic solvents or alcohols and the concentration of alcohol or solvent required to give half-maximal rate acceleration (KA) was obtained from double reciprogal plots of l/ (v~ -u,J versus l/(alcohol or solvent), where v, is the velocity observed at a given concentration of alcohol or solvent and o. is the velocity observed in the absence of alcohol or solvent. Vmax was obtained by addition of the velocity in the absence of solvent or alcohol to the maximal velocity obtained from these double reciprocal plots. The negative reciprocal of the l/(alcohol or solvent) intercept yielded KA. In all cases, double reciprocal plots were linear.
(Since ethyl ar-ketoglutarate was used at a concentration greater than 250 times its K,, maximal velocities were assumed to prevail.) Methanol.
. a ~~~~~~~~~ or alcohon is the velocity observed in presence of alcohol or solvent and ~0 is the velocity in absence of solvent or alcohol.
Activation parameters were determined by measuring the effect of temperature on the rate of hydrolysis of ethyl a-ketoglutarate in the presence or absence of alcohols or organic solvents. Over the temperature range of 15-33", p1ot.s of log v versus l/T were linear.
Exchange between [14C]ethanol and ethyl Lu-ketoglutarate was measured in the following manner.
A 0.25.ml solution containing 1.0 mM ethyl a-ketoglutarate, 0.25 to 1.0 Y [i4C]ethanol (specific activity, 1 X lo* cpm per pmole), 50 mM potassium Tricine buffer, pH 8, 1.5 mM dithiothreitol, and 1.8 pg of enzyme was incubated for 30 min at 30". The reaction was terminated by immersing the reaction mixture in a boiling water bath for 2 min. The reaction mixture was transferred to a 20-ml scintillation vial and taken to dryness by gently blowing air over the solution.
Absolute ethanol (1 ml) was added, and the solution was again taken to dryness. The above procedure was repeated five times, after which 10 ml of scintillation fluid was added and the 14C content was determined.
Control experiments showed that no nonenzymatic exchange reaction occurred, nor was the ethyl ester hydrolyzed by this procedure.
of alcohols and solvents on the hydrolysis of B-ethyl cr-ketoglutarate catalyzed by a different enzyme, rat liver w-amidase (5). In this control experiment, one notes a slight decrease in the hydrolytic reaction, thus illustrating that the rate accelerations observed in the HMPG synthetase reactions are specific for this enzyme and are not derived from artifacts in the assay system.
In order to determine whether acyl transfer from ethyl aketoglutarate or cr-ketoglutaramate was occurring in the presence of alcohols, we measured the rate of exchange of [i4C]etha-no1 with ethyl cY-ketoglutarate.
In the presence of 1 X 10hs M ethyl oc-ketoglutarate and [i*C]ethanol (0.25 to 1.0 M), no exchange reaction was observed.
It was estimated that an exchange rate one-tenth that of the hydrolytic reaction could have been detected.

The maximum
rates obtained for a-ketoglutaramate and &ethyl a-ketoglutarate hydrolysis, in the presence of aliphatic alcohols and solvents, is shown in Table II. All organic solvents give the same maximal rate enhancement for a particular substrate, whereas the rate accelerations with primary alcohols increase with increasing chain length until a maximum value is reached. The effect of a number of aliphatic alcohols and organic In order to establish whether alcohols and organic solvents solvents on the HMPG synthetase catalyzed hydrolysis of a-increase the hydrolytic reactions by the same mechanism, the ketoglutaramate, &ethyl a-ketoglutarate, and &methyl Luketo-effect of these compounds on the kinetics of ethyl ac-ketoglutaglutarate is shown in Table I. All of the alcohols and solvents rate hydrolysis was examined. As shown in Fig. 1, alcohols tested accelerate hydrolysis.
Also shown in Table I  The values of KA and V,,,,, were obtained as described under "Materials and Methods," utilizing the reaction mixtures described in Table I In addition to the hydrolytic reaction, HMPG synthetase catalyzes acyl transfer reaction between &substituted or-ketoglutarates and amines (2,3). The effect of alcohols on the transfer reaction, with S-ethyl a-ketoglutarate and methylamine as substrates, is shown in Fig. 2. In contrast to the results obtained for the hydrolytic reaction, butanol and pentanol are competitive inhibitors with respect to methylamine, while The samples were incubated for 10 min at 30", after which time the reaction was terminated by the addition of 0.05 ml of 20% trichloracetic acid. [WI-Amide formation was measured by the Dowex-50 chromatographic assay (see "Methods").
Control experiments showed that under the above conditions the reaction was linear for at least 40 min. E/v expressed as l/pmoles per min per mg of protein. Inhibition constants were calculated from plots of l/v versus (I) in reaction mixtures identical with those described in Table I ethanol and propanol have no effect. Also shown in Fig. 2 is the effect of dioxane on the transfer reaction. In this case, dioxane acts as a noncompetitive activator of the reaction. Using a different amine acceptor, ethanolamine, competitive inhibition between the amine and propanol, butanol, and penta-no1 was observed, while noncompetitive activation occurred in the presence of dioxane, acetone, and dimethyl formamide. These results serve to illustrate that kinetically the effect of alcohols can be distinguished from those of organic solvents.
The observed competitive inhibition between alcohols and amines suggests the presence of an amine-binding site on the enzyme. To establish more firmly the presence of such a bind- ing site and to estimate its size, we determined the KI values for a series of amines with respect to ethyl or-ketoglutarate hydrolysis.
,4s shown in Table III, the Kr values decrease with increasing chain length and reach a maximum at about 7 carbon atoms.
The nature of the active form of the amine substrate (i.e. free base or protonated amine) was investigated by studying the dependence of Kr on pH. The results of one such experiment, with heptylamine as the inhibitor, is shown in Fig. 3. It is apparent that there is a marked change in Kr with pH.
Calculation of the Kr for heptylamine for the free base form of the amine yielded values of 1.8, 1.8, and 1.9 X lo+ M at pH 7.2, 7.7, and 8.1, respectively.
In a similar type of experiment we found that the KA for heptanol activation was independent of pH over the range of 7.2 to 8.1.
formation of an acyl enzyme (2,3). In order to gain insight into the mechanism of the observed effects of alcohols and solvents, it was desirable to establish whether acylation, deacylation, or perhaps both, were rate limiting for the reaction studied. Table IV lists the maximal velocities for the hydrolysis and transfer reactions (utilizing methylamine and ethanolamine as acceptors) with ethyl cr-ketoglutarate and or-ketoglutaramate as acyl donors.
These results, along with the observation that at low amine concentrations the transfer reaction proceeds without a stoichiometric decrease in the hydrolytic reactions, suggest that both acylation and deacylat.ion are partially rate limiting for the reactions involving both ethyl a-ketoglutarate and (Yketoglutaramate.
In the presence of a series of aliphatic amines the rate of acyl transfer decreases with increasing chain length of the amine, Table V. Table VI lists the activation parameters for the hydrolysis of ethyl ac-ketoglutarate in water, acetone, dioxane, propanol, and pentanol.
The primary effect of these activators is to increase the entropy of activation, with little or no effect on the enthalpy of activation. DISCUSSION We have previously presented evidence that the hydrolysis The HMPG synthetase-catalyzed hydrolysis of S-substituted and acyl transfer reactions proceed through the int'ermediate ac-ketoglutarates is accelerated up to 7-fold by organic solvents and aliphatic alcohols. Similar solvent-and alcohol-mediated rate accelerations have been observed for a number of proteolytic enzymes (6)(7)(8)(9)(10)(11)(12)(13)(14)(15).
The increase in hydrolytic activity cannot be a consequence of acyl transfer to the alcohol or solvent, since the assay system employed specifically measures the product of the hydrolytic reaction, a-ketoglutarate.
Any such transfer reaction would result in no effect or apparent inhibition of the hydrolytic reaction (16).
The data presented in this paper provide quantitative evidence for the previously proposed two-step mechanism involving an acyl enzyme intermediate (2,3) as shown in Scheme 2 where Kx represents the equilibrium constant for binding of the substrate X to the enzyme, and kl, kz, etc. represent the rate constants for the covalent bond-making and bond-breaking steps.
If a common intermediate (i.e. acyl enzyme) is formed in the reaction, then the rate of reaction of this intermediate with water or with a nucleophile (such as an amine) should be independent of the acyl donor (see Scheme 3). Although this criterion can easily be tested when formation of the acyl enzyme is fast relative to its decomposition (3,17), it is more difficult to demonstrate when acylation is rate limiting or partially rate limiting.
That acylation and deacylation are both partially rate limiting for the hydrolysis and acyl transfer reactions with a-ketoglutaramate as the acyl donor and for acyl transfer reactions with ethyl a-ketoglutarate as the substrate is demonstrated by the observation t,hat both substrates are hydrolyzed at nearly identical rates but show rate differences of a-fold for acyl transfer to ethanolamine (Table IV). If deacylation were entirely rate limiting, both cY-ketoglutaramate and ethyl a-ketoglutarat,e would exhibit identical maximal velocities for acyl transfer to ethanolamine.
On the other hand, if acylation were entirely rate limiting, the same maximal velocity for hydrolysis and acyl transfer should occur for each substrate.
The rate of hydrolysis of ethyl a-ketoglutarate (and to a much smaller extent, cu-ketoglutaramate) increases with increasing chain length of the alcohol until a maximal value is reached. This observation is consistent with an increase by alcohols on the rate of acylation, such that deacylation becomes rate limiting, or a,n effect on deacylation such that acylation becomes rate + RCOOH ENZ t RCONHR  Table IV. 6 Taken from the data in Table II. c Calculated from the equation: Voba = V aoy~ation X vdeacylation V acrk.tion + Vdeaeylation ' imiting.
If the effect were on acylation, a-ketoglutaramate and ethyl oc-ketoglutarate would show identical maximal velocities. The fact that these rates differ by a factor of 2 suggests that the effect of alcohols is to increase the deacylation step, and that the maximum rate that can be obtained in the presence of an alcohol reflects the rate of the acylation step.
The rate of acyl transfer from the enzyme to water or to amines can be calculated by the equation where Vobs corresponds to the observed maximal velocity, Vacyl corresponds to the rate of acylation of the enzyme by either ethyl ac-ketoglutarate or oc-ketoglutaramate, and VdheacyI corresponds to the rate of transfer of the acyl group from the enzyme to an acceptor molecule.
Assuming that V,,l is equal to the highest maximal velocity obtained in the presence of alcohols, we have calculated v&oyl for transfer to water, transfer to methylamine, and transfer to ethanolamine. As seen in Table VII, identical values are obtained for both t,he ester, ethyl cr-ketoglutarate, and the amide, a-ketoglutaramate, and thus provide strong evidence for both the acyl enzyme mechanism and the postulate that alcohols specifically effect the rate of hydrolysis of the acyl enzyme.
The inclusion of an amine-binding site in the proposed mechanism is required to account for the observed competitive Louis B. Hersh -methylpyroglutamate Synthetase N 5-Hydroxy-