RICI~ARD WOLFENDEN AND WILLIAMP. JENCKS
4390
[CONTRIBCTIOS50.123 FROX THE GRADUATE DEPARTXENT O F BIOCHEXISTRY, BRASDEISUSIVERSITY,
Vol. 83 \\’ALTHXX
5.&,?ilASS.]
Acetyl Transfer Reactions of 1-Acetyl-3-methylimidazoliumChloride’ BY RICHARD WOLFENDEN AND WILLIAM P. JENCKS RECEIVEDMAY22, 1961 The reactions of 1-acetyl-3-methylimidazolium chloride ( AcMeIm +) with water, acetate monoanion, succinate and phosphate dianions and ammonia proceed with rate constants which, over a range of 106,do not differ significantly from those calculated for the corresponding reactions of S-acetylimidazoliurn i0n.z.3 This demonstrates t h a t AcMeIm + is a satisfactory non-dissociating model for AcImH + and provides further evidence t h a t many reactions of acetylimidazole with reagents of the form HY are, in fact, acid-catalyzed reactions of Y-. The rapid neutral hydrolysis of AcMeIm+ is subject to classical general base catalysis by N-methylimidazole; this reaction and the “water” hydrolysis are 2-4-fold decreased in DzO. X mechanism for the neutral hydrolysis of acetylimidazole is discussed.
The mechanisms of acyl transfer and hydrolysis reactions which follow the rate law rate
= kEY
[XcS][HY]
(1)
are often ambiguous, since such reactions may involve either a direct attack of H Y (transition state I) or a kinetically indistinguishable acidcatalyzed attack of X7- (11).
R
I
R
I1
(If a metastable tetrahedral addition intermediate4 is formed, the Y-C bond may be fully formed in the transition state, but the same considerations apply.) A similar ambiguity exists for classical general acid-base catalyzed reactions. General acid catalysis, for example, can involve catalysis either by the undissociated acid or by a proton and the conjugate base of the acid. Evidence has been presented previously*, that a number of reactions of acetylimidazole which follow rate law 1 actually involve an acid-catalyzed reaction of 1’- (mechanism 11). Since such reactions involve the acetylimidazolium cation (AcImH+, IIIa), we have studied the reactions of 1acetyl-3-methylimidazolium cation (AcMeIm+, IIIb), which would be expected to react, unambiguously, according to mechanism 11, except for the
IIIa, R = H b, R-CH,
replacement of -CH3 for -H. This compound can also be regarded as a model for acid-catalyzed amide hydrolysis, which corresponds to the Nrather than the O-protonated form. Experimental l-A4cetyl-3-~nethylirnidazoliumchloride (XcMeIm+) was prepared by dropwise addition of 0.20 g. of redistilled acetyl chloride in 10 ml. of anhydrous ether t o 0.21 g. of redistilled h7-methylimidazole5 in 15 ml. of anhydrous ether over a period of 15 minutes at room temperature with vigorous stirring. The white precipitate was filtered and washed with 200 ml. of anhydrous ether, with minimal exposure to (1) Supported by grants from the lVational Science Foundation and from t h e National Cancer Institute of the h-ational Institutes o f Health (C-3975). (2) W. P. Jencks and J. Carriuolo, J . B i d Chetn., 2 3 4 , 1272 (1959). (3) W. P. Jencks and J . Carriuolo, ibid., 234, 1280 (1959). (4) M. L. Bender, J . A m Chem. Soc., 73, 1826 (1951). ( 6 ) hl. 7I:iring. r J e h C,,tni A C , ~4 ,2 , 184.5 (ionn).
the atmosphere. An ether slurry of the precipitate was poured into vials and the ether removed by evacuation in a desiccator over P20:. The product, used without further purification, was an extremely hygroscopic white powder which melted at 110” and gave a € 1 2 7yield ~ of acetohydroxamic acid on addition to neutral aqueous hydroxylamine solution.6 The ultraviolet spectrum of the product in ethanol exhibits a shoulder at 240 mp corresponding t o the acetylimidazole peak at 245 mp,7 but differs from the acetylimidazole spectrum in t h a t the extinction rises rapidly below 230 mp. Water, deuterium oxide, ethanol and dioxane were redistilled before use; water was made lo-’ M in ethylenediaminetetraacetic acid. Kinetic Measurements.-The rate of disappearance of AcMeIm + was followed by measuring the decrease in absorbance of water solutions at 245 m p with a Zeiss model P X Q I1 spectrophototneter, equipped with a thermostated cuvette compartment. Unless otherwise specified, the temperature \vas maintained a t 25.0 =t0.1” and the ionic strength was maintained at 0.20 by appropriate addition of XaC1. Reactions were initiated by the addition of approxinmtely 0.1 xng. of the solid amide, with stirring, t o 3 nil. of reaction mixture, to give a solution approximately 2 x lo-“ Ili in amide. Reactions were followed in duplicate for at least 4 halftimes with at least ten optical density readings. Buffer concentrations were in large excess so t h a t the pH, measured with the glass electrode of a Radiometer model 4 pH meter, remained constant throughout and good first-order kinetics mere obtained. Pseudo-first-order rate constants were calculated from thc equation k = 0.693/tl/,. Sodium phosphate, sodium acetate, sodium succinate and S-methylimidazole buffers, neutralized n-ith HC1, were used. Pseudo-firstorder rate constants obtained a t each of a iiuniber of ccncentrations of buffer a t the same PH were plotted against total buffer concentration, yielding a line xvhose slopc gave the second-order rate constant for reaction with the buffer a t the fiH of measurement and which could be extrapolated t o give the rate constant for hydrolysis at zero buffer concentration at the pH of measurement. The reaction with ammonia was studied at several pH values in N-methylimidazole buffers, since the rates are too fast for study in ammonia buffers; the observed rates in the presence of 0.1 ;21atnmonium chloride were corrected for uncatalyzed and buffercatalyzed hydrolysis. I n each case second-order rate constants, k2‘, were determined at three or four PH values.
Results The rate of hydrolysis of AcMeIm+ remains constant from pH 1 to p H 6 and increases as neutrality is approached. In Fig. 1 the observed rate of hydrolysis a t 25.0’ and ionic strength 0.20, extrapolated to zero buffer concentration, is plotted as a function of pH. The dashed line represents the corresponding rate profile for acetylimidazole.2 The increase in the rate of AcPIIeIm+ hydrolysis near neutrality is proportional to (OH-) and the hydrolysis of ,4ckIeIm+ follows the rate law (G) F. Lipmann and L. C . Tuttle, J . B i d . Chem., 159, 21 (1945). (7) E. R . Stadtman in “The Mechanism of Enzyme Action,” ed. W. D. McElroy and B. Glass, Johns Hopkins Press, Baltimore, LTd., l M 4 , p, 581.
Nov. 5 , l O G l rate
=
~-L~CETYL-~-METHYLIIIDAZOLIUM CHLORIDE :
2 8 [XcMeIm+]
ACETYL TRANSFER
4391
+ 9.0 X 106[[AcMeIm'](OH-) (2)
The calculated rate from this equation is shown in Fig. 1 as the solid line. In 0.10 AI DC1 in DnO a t the same ionic strength and temperature kohs = 1.08min.-1, giving k H 2 0 / k D 2 0 = 2.6. I
I
I
YO 0 I M HCI in dioxane. Fig. 2.-Rates of hydrolysis of AcMeIm+ ( A ) and B c I m H + ( 0 )in water-dioxane mixtures a t 25".
-
I .
c ._
E
Y
v)
n
',
0;
i
0
Y
AcIm
i i 4 ; ; ; PH.
6-1
Fig. 1.-Rate of hydrolysis of h c M e I m + , extrapolated to zero buffer concentration, as a function uf pH, a t 25' and ionic strength 0.2.
The rates of solvolysis of AcMeIm+ in aqueous dioxane and ethanol are shown in Figs. 2 and 3. With increasing dioxane concentration the rates of AchieIm+ and AcImH + solvolysis decrease approximately proportionally to the wster concentration, while the reaction of AcMeIm+ with ethanol a t two temperatures, which includes ethanolysis as well as hydrolysis, does not begin to decrease until high concentrations of ethanol are reached, and even exhibits a small maximum in the rate a t intermediate ethanol concentrations. Such a rate maximum suggests that the reaction mechanism involves an interaction of the two components of this mixed solvent (cf. refs. 8, 9). The effect of increasing concentrations of NaCl on the rate of AcMeIm+ solvolysis was also found to be identical with that obtained2with AcImH+. The data obtained for the nucleophilic reactions of various bases with AcMeIm+ are given in Table I. In each case the reaction occurs mainly with a single ionic species, namely, the acetate monoanion, the succinate and phosphate dianions, and neutral ammonia. The possibility of a second-order reaction of ammonia due to general base-catalyzed ammonolysis, as previously described for the ammonolysis of acetylimidazole, is not excluded by (8) J. Koskikallio, Acta Chrm. Scand., lS, 665 (1950). (0) J. a, Hyne, J , Am. Cham. Soc,, 91, 5128 ( 1 P b O ) .
0
.,
05
10
Fraction H,O.
Fig. 3.-Rates of solvolysis of AcMeIm" in ethanolwater mixtures a t 25': a t 0.6" and 25.0"; 0, in ethanolwater mixtures in dioxane with a total concentration of hydroxylic solvent of 16.7 M.
these data, since such a reaction would have become manifest only a t rates too fast to measure by the present methods. TABLE I RATE CONSTANTS FOR REACTIONOF AcMeIm+ WITH NUCLEOPHILIC REAGENTS AT 25.0" A N D IONIC STRENGTH 0.2 Reactant
A- (pK)
C0ncn.a range
Acetate(4.76)
0.05-0.20 .05- .20 .05- . 2 0 .os- . 2 0 Succinate='d .02- . 2 0 (5.35) .02- . 2 0 Phosphate' ,005-0.02 (6.86) ,025- . l 0 .04 .01 Ammonia .10 (9.21) .10 .IO
-
pH
A-/total A
1. mole-' ka',b min.-1
5.56 5.04 4.64 4.24 6.50 5.77 6.09 5.78 5.63 6.72 6.27 5.82
0.86 .65 .43 .ll .9$ ,725 ,145 ,077 ,0555 3 . 2 4 X lo-' 1.15 X 10-0 0.41 X 10-9
14.5 11.0 7.2 3.0 39.5 30.0 376 168 136 198 71.2 27.2
I. mole-' ki,c min.-l 16.8 16.8 16.7 14.4 42.2 41.4 2590 2180 2450 61,100 61,900 64,100
=Average of six determinations a t each pH. b k z ' 3 - khydrola)/[tOtal A ] . k2 = (hob. * k h v d r o l n ) / [ A - ] ~ Ionic strength 0.60for cucsinate.
(kabs
TABLE I1 SECOND-ORDER RATE CONSTANTS FOR XUCLEOPIIILIC ACTIONS Vi1111 ,4CETYLIMTDAZOLE*~3AND AchlcIm + AT
RE25"
kw-
[AcIml-
[HY],
Reactant,
2 f
a
10 M [ MeIm] f r e e base.
Fig 4.-General base catalysis of AchlcIni + hydrolysis hy N-methylimidazole a t 25" and ionic strcngtli 0 2. Rntc constants determined in tlic PH range 6.36 and 7 A7 a11d corrcctcd for uncntalpzccl hydrolysis, after extrapolation t o zero buffer concentration.
At constant pH and ionic strength the rate of hydrolysis of AcMeIm+ increases with the concentration of N-methylimidazole buffer. This increase is greater a t more alkaline pH, and a plot of the increase in rate against the concentration of N-methylimidazole present as the free base a t each of several pH values is linear (Fig. 4). This requires the addition of a term for general base catalysis to the rate law for AclIeIm+ solvolysis =
kl[AcMeIm+]
1 mole - 1 ntin - 1
ky-[AcImH + I [Y-1, !. mole 1 mtn - 1
kr-[AcMeIrn+][Y-1, ! myle-1 min.
Water 0 051" 0.051" CH3COO1.3 19 17 Succinate' 14 78 42 HPOI" 1 0 2100 2230 PiHs 0.20 81000 62000 a Second-order constant, calculated for a water concentration of 55 111.
05
rate
Y-
+ k2[AcMeIm+](OH-) +
[ilcMeIm+] Z&B,[B,] (3) with k B = 18 1. mole-' rnin.-l for N-methylimidazole catalysis. The rate of the catalyzed reaction in D20 solution is decreased to a value too small for accurate measurement (Fig. 4); an estimated value of 4.5 for the catalytic constant in D20 gives a value of k l 1 2 0 / k D 2 0 = 4 for the catalyzed reaction.
Discussion The rate of hydrolysis of AcMeIm+ below pH 6 is identical with that of acetylimidazole in dilute acid solution, i.e., to that of acetylimidazolium cation (AcImH+).Z The effect of dioxane and NaC1 on the hydrolysis of AchleIm+ in 0.1 M HCI and the 2.6-fold rate reduction in D?O correspond closely with the solvent and deuterium isotope effects for the hydrolysis of protonated acetylimidazole under the same conditions. This constitutes strong evidence that protonation of acetylimidazole occurs on the free ring nitrogen and that .i\cMeIm+ is a satisfactory model for reactions of the conjugate acid of acetylimidazole (AcImH +). The observed rate constants, expressed according to rate law 1, for the reactions of acetylimidazole with a series of reagents, HY, are summarized in column 2 of Table 11. In column 3 are given the corresponding rate constants for the acid-catalyzed reaction of ET- according to mechanism I1 and rate law 4 rate = ky-[AcXH +I [Y-1 (41 calculated from the relation ky- = kHYKAoxn+/KnY
(5)
in which KA~XH+ and I
