Electrostatic acceleration of enolization in cationic ketones

Article abstract of DOI:10.1021/ja950573p

Rate constants for the water, acetate, and hydroxide ion-catalyzed enolizations of the cationic ketones 2-acetyl-1-methylpyridinium ion (3) and 1-methyl-8-oxo-5,6,7,8-tetrahydroquinolinium ion (4) have been measured at 25°C and compared with those reporte

Full text of DOI:10.1021/ja950573p

VOLUME 118, NUMBER 49
DECEMBER 11, 1996
©
Copyright 1996 by the
American Chemical Society
Electrostatic Acceleration of Enolization in Cationic Ketones
John B. Tobin and Perry A. Frey*
Contribution from the Institute for Enzyme Research, The Graduate School, and Department of
Biochemistry, College of Agricultural and Life Sciences, UniVersity of WisconsinsMadison,
Madison, Wisconsin 53705
X
ReceiVed February 17, 1996. ReVised Manuscript ReceiVed June 10, 1996
Abstract: Rate constants for the water, acetate, and hydroxide ion-catalyzed enolizations of the cationic ketones
-acetyl-1-methylpyridinium ion (3) and 1-methyl-8-oxo-5,6,7,8-tetrahydroquinolinium ion (4) have been measured
at 25 °C and compared with those reported for the enolization of 2-acetyl-3,4-dimethylthiazolium ion (2) (Halkides,
2
-
1
C. J.; Frey, P. A.; Tobin, J. B. J. Am. Chem. Soc. 1993 115, 3332). For 3, kH O, kOAc, and kOH are 1.32 ( 0.21 s ,
2
-
2
-1 -1
-1 -1
(
2.82 ( 0.95) × 10
M
s , and 187 ( 50 M s , respectively. The corresponding values for 4 are 9.17 (
-2
-1
-4
-1 -1
-1 -1
0
.24) × 10 s , (4.32 ( 0.18) × 10
M
s
and 10.9 ( 0.2 M s , respectively. The values of pKa for 3 and
4
are 11.13 and 11.90, respectively, at 25 °C. The hydration equilibrium constant Kh for 3 is 0.084 ( 0.004 at 25
°
C. The correlation of hydration constants for ketones with σ* (Greenzaid, P.; Luz, Z.; Samuel, D. J. Am. Chem.
Soc. 1967 89, 749) allow the hydration constants for 2 and 3 to be used in the estimation of σ* values for the
substituents. The σ* values are 1.68 and 1.02 for 2-(N-methylthiazolium) and 2-(N-methylpyridinium), respectively.
Use of the σ* values to estimate the inductive effects of these substituents on the pKas of 2 and 3 allows the inductive
effects to be separated from the through-space electrostatic effects. The inductive effects are estimated to contribute
4
.2- and 1.9-log units to lowering the pKas of 2 and 3, respectively. From this and the measured pKa of 11.1 for 3,
-
1
the through-space electrostatic contribution to lowering the pKa is 6.3-log units, or 8.6 kcal mol in free energy of
enolate stabilization. Assuming the same through-space effect for 2, its pKa is estimated to be 8.8. Comparisons of
these enolization rates and pKas with those for ordinary methyl and benzyl ketones indicate that the 2-(N-
methylpyridinium) substituent in 3 stabilizes the enolate by 11.2 kcal/mol in free energy, and about 8.6 kcal/mol of
this can be attributed to through-space electrostatic stabilization. The through-space electrostatic component accounts
for a 330-fold enhancement in kOH for 3 compared with a typical methyl ketone (pKa ) 19.3 for acetone). The value
6
of kOH for 2 is 1.1 × 10 times that for a typical methyl ketone (pKa ) 19.3) and 70 times that for a neutral methyl
ketone exhibiting the pKa of 8.8 estimated for 2. The through-space electrostatic effects on the enolate of 2 and the
4
transition state for its formation account for a 2.3 × 10 -fold enhancement in the enolization rate. The remaining
47-fold rate enhancement for 2 is attributed to inductive effects. The through-space electrostatic effects have important
implications for enzymatic catalysis of enolization.
I. Introduction
fully understood. In this context, electrostatic rate enhancement
has recently been emphasized, its significance being that enolate
stabilization provided by electrostatic interactions may be
The R-deprotonation of carbonyl compounds to produce
enolate ions has been the topic of many investigations.
Correlations of base-catalyzed enolization rate constants (kB )
1
2
,3
E
important for enzymatic catalysis of enolization.
with pKa values are nearly linear, while displaying a slight
curvature owing to a large Marcus intrinsic barrier, ∆Go .
However, rate constants that deviate from this curve are not
We have reported the rapid rates of enolization of the methyl
q
ketones 2-acetylthiamine pyrophosphate (1) and 2-acetyl-3,4-
3
dimethylthiazolium ion (2). The rate constant for hydroxide
X
6
Abstract published in AdVance ACS Abstracts, November 15, 1996.
ion-catalyzed enolization of 1 was found to be >10 times that
S0002-7863(95)00573-7 CCC: $12.00 © 1996 American Chemical Society

12254 J. Am. Chem. Soc., Vol. 118, No. 49, 1996
Tobin and Frey
8
.10 (t, 1H), 8.38 (d, 1H), 8.65 (t, 1H),8.77 (d, 1H). Anal. Calcd for
C
5
8
H
10INO C, 36.53; H, 3.83; N, 5.32. Found C, 36.45; H, 3.86; N,
.17.
-Methyl-8-oxo-5,6,7,8-tetrahydroquinolinium Iodide (4). The
1
7
procedure of Thummel et al. was followed with slight variations.
Reaction of 2,3-cyclohexenopyridine (13.1 g, 98 mmol) with benzal-
dehyde (14.1 g, 108 mmol) in the presence of acetic anhydride (17.7
g) at 170 °C gave 8-benzylidene-5,6,7,8-tetrahydroquinoline, which was
purified by distillation (bp 119-228 °C at 10 Torr) to give 13.2 g of
a viscous liquid. This became a yellow solid upon standing (66% yield)
7
mp 59-62 °C (lit. 62-64 °C). Ozonolysis of this compound (4.50 g,
2
0.5 mmol) was performed in 100 mL of methanol using a T-23
Welsbach Ozonator. Purification by column chromatography (silica
gel; CH Cl ) gave 0.98 g of a yellow solid (32% yield) the identity of
which was verified by its H NMR spectrum.
for acetone. The major factor in these enhanced enolization
rates has been suggested to be electrophilic stabilization of the
enolates by the positively charged thiazolium ring. Electrophilic
effects on enolization in cationic ketones have not been separated
into through-space and through-bond, inductive components.
The through-space component should be especially relevant to
enzymatic catalysis of enolization because of the possibility that
enolate ions at enzymatic active sites can be stabilized by the
side chains of basic amino acids.
2
2
1
Methylation of 8-oxo-5,6,7,8-tetrahydroquinolone (0.23 g, 1.6 mmol)
was carried out by stirring with trimethyloxonium tetrafluoroborate (0.21
g, 1.6 mmol) in nitromethane. Following the removal of nitromethane
under vacuum, the product was dissolved in water and passed through
an anion exchange column (Bio-Rad AG-1 × 8) in the iodide form.
Recrystallization from ethanol/diethyl ether gave 0.16 g of dark orange
crystals (34% yield), mp 168-171 °C: 1H NMR (CD
δ 2.21 (p, 2H), 2.89 (t, 2H), 3.21 (t, 2H), 4.46 (s, 3H), 8.03 (t, 1H)
3
CN, 500 MHz)
Electrophilic contributions to the enolization of cationic
ketones can be further understood through knowledge of the
thermodynamic acidity constants for their ionizations. However,
it is not feasible to measure pKa values for enolizations of
acylthiazolium ions such as 1 and 2 because they rapidly
undergo deacylation in basic solutions.4 The 2-acylpyridinium
ions 3 and 4 also undergo hydroxide ion-catalyzed deacylations,
but at slower rates than 2-acylthiazolium ions. Therefore, pKas
and enolization rates for these compounds can be measured.
Intramolecular electrostatic effects may be important in the
8
2
.51 (d, 1H), 8.69 (d, 1H); 13C NMR (DMSO-d
8.5, 49.3,129.1, 140.9, 145.9, 147.5. Anal. Calcd for C10
6
, 127 MHz) δ 20.6,
12INO:
H
C, 41.54; H, 4.18; N, 4.85. Found: C, 41.19; H, 4.24; N, 4.79.
Hydration Equilibrium. The ratios of ketone/hydrate for 3 and 4
1
in D
2
O were measured by H NMR spectroscopy (500 MHz). Solutions
,5
of the ketones were allowed to equilibrate for 30 min in 0.010 M DClO
4
+ 0.090 M NaClO at 25 °C before recording their NMR spectra.
4
Kinetics of Hydroxide Catalyzed Ionization. The rates of ioniza-
tion of 3 and 4 in aqueous sodium hydroxide solutions were determined
spectrophotometrically by monitoring the increases in absorbance at
3
50 and 400 nm. In all the kinetic solutions, the ionic strength was
maintained at 0.10 by the presence of the appropriate concentration of
NaClO . Measurements were made at 25.0 ( 0.1 °C using a Hewlett
enolization of 2-acylpyridinium ions. In fact, very fast base-
catalyzed enolization of 3 has been reported.6
4
In the present paper we have extended our study of cationic
ketones to 2-acetyl-1-methylpyridinium ion (3) and 1-methyl-
Packard 9153C Diode Array UV-vis spectrophotometer for 4 and a
stopped-flow spectrophotometer (Update Instruments System 7-RS) for
the faster ionization reactions of 3.
8
-oxo-5,6,7,8-tetrahydroquinolinium ion (4). The rates of
enolate ion formation have been obtained by direct measure-
ments in aqueous hydroxide solutions and by iodine scavenging
in acetate buffers. These results along with pKa determinations
in aqueous base have been used to evaluate the influence of
electrostatic stabilization on the ionizations of these ketones,
and the results have been compared with those for 2-acetyl-
Kinetic runs for the ionization of 3 in NaOH solution were conducted
by first allowing the reactants to equilibrate to 25 °C in the mixing
syringes of the stopped-flow spectrophotometer. Ionizations were
started by the 1:1 mixing of an aqueous solution of 2-acetyl-1-
methylpyridinium ion with an aqueous sodium hydroxide solution so
that the concentration of substrate following mixing was 4 × 10 M.
The ionization followed first-order kinetics accurately for more than
five half-lives. Rate constants were calculated by fitting the data to
the exponential form of the first-order rate equation.
-
4
3
,4-dimethylthiazolium ion 2. The electrophilic accelerations
of enolization by the positively charged nitrogen atoms in 2
and 3 have been separated into inductive and through-space
electrostatic components.
Products from the reaction of 2-acetyl-1-methylpyridinium ion in
1
NaOH solution were examined in two time course experiments by H
NMR spectroscopy. The reactions were started by rapidly dissolving
II. Experimental Section
5 mg of 3 in 9.0 mL of 0.1 M NaOH. One reaction was stopped after
5
s and the other after 4 min by adding 9.0 mL of 0.1 M HCl and
Materials. Reagents used in this study were purchased from the
Aldrich Chemical Co. Piperidine, 1-methylpiperidine, and ethanolamine
were distilled before being used to prepare buffers.
quickly mixing. Following the removal of water by lyophilization,
samples were dissolved in DMSO-d , filtered, and their NMR spectra
6
recorded.
For the ionization of 4 in NaOH solution monitored by UV
spectrophotometry, reactions were started by the addition of an aliquot
of a stock solution of ketone in acetonitrile to 3.00 mL of NaOH solution
2
-Acetyl-1-methylpyridinium Iodide (3). 2-Acetylpyridine (2.00
g, 16.5 mmol) was refluxed in excess methyl iodide (2.43 g, 17.1 mmol)
6
for 24 h. Removal of methyl iodide under vacuum gave light green
crystals, which were recrystallized three times from ethanol to give
([ketone] ) 4.4 × 10-4
M and acetonitrile was <1% by volume). In
6
0
1
.94 g of yellow crystals (22% yield), mp 160-161 °C (lit. mp 160-
these kinetic runs, however, the exponential increase in absorbance at
400 nm was accompanied by a process that brings about a decay in
absorbance. In some cases, the rates had to be corrected for the
disappearance of the enolate ion. First-order rate constants were
1
62 °C): H NMR (CD
3
CN) 500 MHz) δ 2.76 (s, 3H), 4.36 (s, 3H)
(1) For reviews see: (a) Rappoport, Z., Ed. The Chemistry of Enols;
Wiley: New York, 1990. (b) Bell, R. P. The Proton in Chemistry, 2nd ed.;
Cornell University Press: Ithaca, New York, 1993. (c) Jones, J. R. Prog.
Phys. Org. Chem. 1971, 9, 241.
calculated either by using a fit to a simple exponential function or by
8
using the equation A ) A
f
k/(l - k)[exp(-kt) - exp(-lt)], where A
f
is
(
(
2) Guthrie, J. P.; Kluger, R. J. Am. Chem. Soc. 1993, 115, 11569.
3) Halkides, C. J.; Frey, P. A.; Tobin, J. B. J. Am. Chem. Soc. 1993,
the total absorbance change due to enolate ion and k and l are the rate
constants for the formation and decay of enolate ion, respectively.
1
15, 3332.
(
4) (a) Gruys, K. J.; Halkides, C. J.; Frey, P. A. Biochemistry 1987, 26,
(7) Thummel, R. P.; Lefoulon, F.; Cant, D.; Mahadevan, R. J. Org. Chem.
1984, 44, 2208.
7
575. (b) Gruys, K. J.; Datta, A.; Frey, P. A. Biochemistry 1989, 27, 9071.
(
(
5) Lienhard, G. E. J. Am. Chem. Soc. 1966, 88, 5642.
6) Cox, B. G. J. Am. Chem. Soc. 1974, 96, 6823.
(8) Espenson, J. H. Chemical Kinetics and Reaction Mechanism;
McGraw-Hill Book Co.: New York, 1981, p 66.

Enolization in Cationic Ketones
J. Am. Chem. Soc., Vol. 118, No. 49, 1996 12255
9
Kinetics of Iodination. Rates of iodination were measured in
of K by semicarbazide scavenging was not possible because
h
aqueous acetate buffer solutions by monitoring the disappearance of
the ketones reacted much too slowly, so we measured Kh by
H NMR spectroscopy. The hydrate of 4 was not observed in
-
I
3
at 354 nm. Iodine was generated by the acid-catalyzed reaction of
with KI. The iodination reactions were done under zeroth-order
1
KIO
3
the NMR spectrum in D2O, so that Kh < 0.001. Signals
assigned to the hydrate and ketone of 3 appeared in an
equilibrated solution of the ketone in D2O, and the integrals of
these signals were used to calculate (Kh)D2O ) 0.098. The
equilibrium constant in D2O was corrected to Kh in H2O by
conditions for 3 and 4.
The first-order reaction of iodine with 2-acetyl-1-methylpyridinium
iodide was carried out as follows: To 1.00 mL of aqueous buffer
-
5
solution containing 9 × 10 M I
2
was added 3.3 µL of 0.027 M
substrate stock solution in acetonitrile. The disappearance of iodine
in buffer solution in the absence of substrate corresponded to only a
small portion of the rate measured in the presence of the substrate.
Corrected absorbance decays were obtained by subtracting the back-
ground decays from the total decay curve. The resulting decay curves
fitted accurately to an exponential function to give the observed first-
order rate constants.
2
9,10
use of the equation (Kh)D O/(Kh)H O ) Φ ) 1.08.
The
2
2
corrected value of (Kh)H O was 0.084 ( 0.004.
2
Enolate Ion Formation. Reaction of the cationic ketones 3
and 4 with aqueous NaOH leads to the appearance of chro-
mophores at 350 and 400 nm, respectively. The absorption
maxima appear at longer wavelengths than those characterizing
enolates of alkylsubstituted monocarbonyl compounds (220-
260 nm). However, the λmax values for the enolates of 3 and
4 are similar to those of the enolates of the related aromatic
compounds 3-(phenylacetyl)pyridinium ion (360 nm) and
Iodinations of 3 and 4 were performed in acetate buffer solutions
under zeroth-order conditions. To 3.00 mL of acetate buffer was added
2.0 µL of substrate solution in acetonitrile. The mixture was allowed
to equilibrate to 25 °C ([ketone] ) 2-0.8 mM). The concentration of
-acetyl-1-methylpyridinium ion was measured by recording the
absorbance of the substrate solution at 310 nm, where the effective
molar extinction coefficient is 106.6 ([ketone] ) A310/ꢀ310). The kinetic
1
a
1
2
11
4
-(phenylacetyl)pyridinium ion (446 nm), respectively.
Consideration has been given to whether the chromophores
that develop quickly upon mixing 3 or 4 with hydroxide are
attributable to pseudobase addition compounds rather than to
enolate ions. These chromophores are most consistent with
enolates for the following reasons: (a) The spectral changes
are reversed upon neutralization, consistent with an acid/base
equilibration. (b) The positions of the absorption maxima are
runs were initiated by addition of 8.0 or 10.0 µL of 1.0010 mM I
.100 M NaI to the ketone solution, and the absorbances at 354 nm
were recorded as a function of time. Traces of decreasing absorption
versus time for the first 30% of I consumption were linear. The
2
/
0
2
calculated gradients, υ, were converted into observed first-order rate
constants using the equation kobs ) υ/ꢀ354[ketone], and the effective
3
-1
11
extinction coefficient, ꢀ354 ) 9.14 × 10 M (for [I
and [I ] ) 9.7 × 10 M in acetate buffer solution at pH 4.2 to 5.0 at
5 °C).
pK
determined at 25 °C in NaOH solutions and buffer solutions (piperidine,
-methylpiperidine, and ethanolamine) at an ionic strength of 0.10 by
2
] ) 9 × 10-5 M
similar to those for other, structurally related aromatic enolates.
-
-4
(c) An NMR experiment with 4 in D2O confirmed exchange of
the methylene protons adjacent to the carbonyl group occurs at
a rate comparable with development of the chromophore.
Ketone 4 was dissolved in 0.1 M NaOD, stirred for 2 min at
room temperature, and quenched with 1 equiv of DCl. Solvent
2
a
Determinations. The carbon acidities of 3 and 4 were
1
use of the Hewlett-Packard 9153C spectrophotometer. The pHs of
buffer solutions were measured by use of a Radiometer PHM 26 pH
meter.
1
was removed, and 4 was dissolved in CD3CN for H NMR
analysis, which showed >99% loss of the protons in the R-CH2
group (normally a 2H triplet at 2.89 ppm). A lower limit for
the exchange rate was estimated by considering the two-stage
exchange reaction: -CH2- f -CHD- f -CD2- governed
by rate constants k1 and k2, respectively. Setting k1 ) k and k2
f
The final absorbance values A at 350 nm (3) and 400 nm (4)
following the formation of enolate ions were used to construct titration
curves. Final absorbance values for 4 were determined as described
in the kinetics section. After dissolving 3 in aqueous solution, the
absorbance values at 350 nm were recorded one minute from the mixing
time (low pH) or by extrapolation of the exponential absorbance decay
to the time of mixing (high pH).
)
k/2, assuming a negligible secondary isotope effect, the total
proton content for the two-proton exchange is given by [H] )
1
2
[
H0] exp(-kt/2). An estimate of the minimum exchange rate
constant leading to 99% exchange, which corresponds to at least
The expression for the final absorbance values is given by the sum
-
1
7
half-lives in 120 s, is k g 0.08 s , and the enolization rate
of the absorption contributions from each of the species in solution A
f
-
1
-
constant is 2k g 0.15 s for 4 in 0.1 M NaOD. The minimum
enolization rate constant as estimated by NMR is consistent with
the spectrophotometrically measured value of 0.3 s in 0.1 M
NaOH.
)
ꢀ [K] + ꢀ
K
E
[E] + ꢀ
h
[h] + ꢀh-[h ]: ꢀ
X
is the extinction coefficient
-
for a given species, X ) K (ketone), E (enolate ion), h (hydrate), h
hydrate anion). The pH-dependent absorbance in terms of total
-
1
(
substrate concentration, [S]tot is
Over a longer time interval, the chromophores attributed to
enolate ions decreased in intensity. N-Methylpyridinium ion
was identified by H NMR as the product arising from this
+
h
{
(ꢀ + ꢀ K )[H ] + (ꢀ K + ꢀ K K )}[S]
h
h
h
E
a
h-
h
a
tot
1
Af )
)
+
h
(
1 + K )[H ] + K + K K )
h a h a
absorbance decay in the reaction of 3 in NaOH. We propose
that the disappearance of the enolate ions derived from 3 and 4
results from deacylation by way of the hydrate anion species,
as shown in eq 1 for the reaction of 3. This deacylation is
analogous to the well-known deacylation of 2-acetyl-3,4-
+
+
(
A[H ] + B)[S] /(K[H ] + C)
tot
For 3 the value of K on the right side of the equation was set at K )
+ K ) 1.084. For 4 the value of K was set at 1.00 because the
hydration of this ketone is negligible. Curve fits for both substates
were done according to the above equation, and the acid dissociation
1
h
5
dimethylthiazolium ions, but it takes place on a much slower
time scale, allowing enolization to be observed spectrophoto-
metrically at high pH values. In this sense, the deacylations of
constants were obtained using the kinetically determined value of K
Z
h
h
()K
h
K
a
) and K
a
h a
) C - K K (see Results).
3
2
and 4 validates the 2-acylpyridinium ions as analogues of
-acetylthiazolium ions.
III. Results
Kinetics of Enolate Ion Formation. The time dependencies
Hydration Equilibrium. The hydration equilibrium con-
stants of 3 and 4 were determined in aqueous solutions at 25
(9) Chiang, Y.; Kresge, A. J.; Krogh, E. T. J. Am. Chem. Soc. 1988,
10, 2600.
1
(
(
10) Bone, R.; Wolfenden, R. J. Am. Chem. Soc. 1985, 107, 4772.
11) Bunting, J. W.; Stefanidis, D. J. Am. Chem. Soc. 1988, 110, 4008.
°C (ionic strength 0.10) by measuring the proportions of hydrate
and ketone at equilibrium. A spectrophotometric determination
(12) Kresge, A. J.; Tobin, J. B. J. Org. Chem. 1993, 58, 2652.

12256 J. Am. Chem. Soc., Vol. 118, No. 49, 1996
Tobin and Frey
to 50%). This behavior was observed previously for 2-acetyl-
6
1
-methylpyridinium ion by Cox, who also demonstrated that
the rates were independent of the halogen used.
The rates of enolization were found to be proportional to the
buffer concentration. Therefore, the data were fitted to eq 3 to
obtain the buffer catalytic coefficients kcat. For each data set at
a given buffer ratio, kcat was then fitted to eq 4, in which kHA
kobs ) k + k [buffer]
(3)
(4)
o
cat
kcat ) kHA + (k - kHA)fB
B
and kB are the buffer-acid and buffer-base catalytic coefficients,
respectively, and fB is the fraction of buffer in the basic form
of the increases in absorption accompanying enolate ion
formation in aqueous sodium hydroxide solutions were measured
by stopped-flow spectrophotometry in the case of 3 and
conventional spectrophotometry for 4 at 25 °C and ionic strength
-
-
(
[
fB ) [AcO ]/([HOAc] + [AcO ]). Linear plots of kcat versus
buffer] with zero intercepts showed that the reaction is subject
to general base catalysis but not general acid catalysis.
Plots of the intercepts at zero buffer concentration (ko) versus
hydroxide concentration were made in order to determine the
0
.10. The rates of enolate formation observed at 350 nm for 3
were first order in ketone concentration for more than 98% of
the reaction. Observed first-order rate constants were obtained
in NaOH solutions over the concentration range 0.004-0.10
M. The increase in absorbance at 400 nm corresponding to
the ionization of 4 was followed by its decay. In hydroxide
solutions of 0.8-9.9 mM, the decay of the band at 400 nm was
slow relative to its formation; therefore, first-order rate constants
were evaluated using a single exponential function. At base
concentrations of 20-99 mM, however, the secondary hydroly-
sis reaction interfered, and the corrected rate constants for
enolate ion formation were determined as described in the
Experimental Section.
rate constants kH O and kOH by fitting to eq 5. For both ketones
2
+ kOH[OH-]
(5)
ko ) k
H2O
the slope was zero within error, indicating that the effect of
hydroxide ion in acetate buffer solutions was small. Upper
3
2
-1
limits of kOH were found to be 6.3 × 10 and 1.7 × 10 M
for 3 and 4, respectively.
The rate constants determined in acetate buffer studies were
-
1
s
the catalytic coefficients referring to the unit concentration of
an equilibrium mixture of ketone and its hydrate. They were
converted to true catalytic coefficients based on unit ketone
concentration through multiplication by (1 + Kh), Where Kh )
The first-order rate constants for ionizations of 3 and 4
increase with the hydroxide concentration at the lower end of
the concentration range and, as shown in Figures 1 and 2, reach
limiting values at higher concentrations. This behavior is consis-
tent with an ionization reaction in which there is a kinetically
0
.084 and 0 for 3 and 4, respectively (see below). The rate
constants corrected for hydration are given in Table 1. The
-
rate constant for the enolization of 3 by acetate agrees with the
controlled conjugate base, Z , formed with an equilibrium
-
3
-1 -1 6
previously determined value of 8.3 × 10
M
s .
constant of KZ. In the case of the reaction of 3 according to eq
-
pK Determination. The ketone acidity constants for 3 and
1
, Z is the hydrate anion and KZ is the equilibrium constant
a
4
in aqueous solution were determined spectrophotometrically.
for its formation. The applicable rate law is given by eq 2,
The final absorbance values following enolate ion formation in
kinetic runs (350 nm for 3 and 400 nm for 4) were used to
construct the titration curves in Figures 3 and 4 according to
the procedure described in the Experimental section. The
calculated values of pKa for compounds 3 and 4 are given in
Table 1.
-
-
kobs ) (kOH[OH ] + k )/(1 + K [OH ]/K )
(2)
H O
Z
w
2
where Kw is the ion product of water and kH O is the rate constant
for uncatalyzed enolization. The constants kOH, kH O, and KZ
were evaluated for 3 and 4 by fitting data to eq 2. Because
these ketones are in equilibrium with their hydrate forms, the
rate constants are multiplied by the correction factor (Kh + 1)
to convert to the “true” rate constants, which are given in Table
2
2
IV. Discussion
When compared with acetone, 2 and 3 undergo hydroxide-
6
3
catalyzed ionization about 10 and 1.7 × 10 times times faster,
respectively, and their rates of base-catalyzed enolization are
also orders of magnitude faster than that of acetaldehyde.
1
.
The hydrate anion 6 (eq 1) and its analogue derived from 4
are the most obvious species that can be postulated for the
structure of an anionic decomposition intermediate Z . The
formation of hydrate anions has also been postulated to explain
saturation kinetics in the deprotonation reactions of 1-methyl-
13
Cationic ketones 8 and 9 also ionize very fast, and the ketones
0 and 11 undergo hydroxide ion-catalyzed enolization faster
-
1
3
-acetylpyridinium and 4-(arylacetyl)pyridinium ions in basic
11
aqueous solutions. The equilibrium constant KZ is, therefore,
related to the acid dissociation constant for the hydrate, Ka :
KZ ) KhKa .
h
h
Rate constants for enolate ion formation in acetic acid/acetate
buffers were measured indirectly by iodine scavenging. Rate
measurements were made in a series of solutions at constant
ionic strength and buffer ratio while varying the buffer
concentration. Rates of enolization were measured in a series
than their unmethylated forms by factors of 2600 and 15,
respectively. Fast enolizations of these compounds have been
-
(13) (a) Carey, A. R. E.; Al-Quatami, S.; More O’Ferrall, R. A.; Murray,
B. A. J. Chem. Soc., Chem. Commun. 1988, 1097. (b) Personal communica-
tion to J. R. Keeffe and A. J. Kresge: see ref 1a, p 469. (c) Murray, B. A.
Ph. D. Thesis, University College, Dublin, 1988.
of experiments at four buffer ratios: [HOAc]/AcO ] ) 0.49,
0
.97, 1.38, and 1.92. The enolizations were zero order with
respect to iodine in at least the first portion of the reaction (up

Enolization in Cationic Ketones
J. Am. Chem. Soc., Vol. 118, No. 49, 1996 12257
Table 1. Summary of Rate and Equilibrium Constants for the Interconversions of 2-Acetyl-3,4-dimethylthiazolium and
a
2
-Acetyl-1-methylpyridinium Ions in Aqueous Solution
parameter/compd
23
3
4
-
1
5 b
2 b
k
k
k
pK
pK
OH (M s-1
)
(1.20 ( 0.20) × 10
(1.87 ( 0.50) × 10
10.9 ( 0.2
-
1
-2
-4
H O (s
)
1.32 ( 0.21
(9.17 ( 0.24) × 10
(4.32 ( 0.18) × 10
11.90 ( 0.03
2
OAc (M- s-1
1
)
(7.63 ( 0.95) × 10
-3 b
(2.82 ( 0.95) × 10
11.13 ( 0.05
12.53 ( 0.12
0.084 ( 0.04
-2 b
c
a
8.8
(-log K
h
K
a
h)
12.32 ( 0.04
<0.001
Z
K
h
1.10 ( 0.10
a
Ionic strength ) 0.10, 25 °C. b Corrected for hydration equilibrium. c Estimated as described in the text.
Figure 1. Dependence of ionization rate on NaOH concentration for
-acetyl-1-methylpyridinium ion 3. The first order rate constants are
plotted versus the concentration of aqueous NaOH for the ionization
of 2-acetyl-1-methylpyridinium ion 3. Each point represents the average
of 5 to 7 rate measurements.
Figure 3. Titration curve for the ionization of 2-acetyl-1-methylpy-
ridinium ion (3) in aqueous NaOH. Absorbance at 350 nm is plotted
against pH. Each point represents the average of five to six determina-
tions.
2
Figure 4. Titration curve for the ionization of 1-methyl-8-oxo-5,6,7,8-
tetrahydroquinolinium ion (4) in aqueous NaOH. Absorbance at 400
nm is plotted against pH. Each point represents the average of six
determinations.
Figure 2. Dependence of ionization rate on NaOH concentration for
1
-methyl-5,6,7,8-tetrahdroquinolinium ion (4). The first-order rate
constants are plotted versus the concentration of aqueous NaOH for
the ionization of 1-methyl-5,6,7,8-tetrahdroquinolinium ion (4). Each
point represents the average of 6 rate measurements.
easily separated and have not been for 8-11. In the following
discussion, we shall attempt to delineate the inductive and
through-space electrostatic effects of the 2-(N-methylpyridinium)
and 2-(N-methylthiazolium) substituents on the enolizations of
attributed to electrostatic stabilization of the developing negative
charge on the carbonyl oxygen by the positive charge on
1
4
nitrogen in the transition state for enolate formation. The
electrophilic effects of positively charged nitrogen in the
enolization of these molecules can be conceptualized as the
product of two factors, the inductive effects of the electropositive
substituents and the through-space electrostatic effects of the
positive charges. Each of these effects can be manifested in
both the ground state for the enolate ion and the transition state
for its formation. Inductive and through-space effects are not
2
and 3. The through-space electrostatic components are of
special interest for understanding enzymatic enolization, in
which positively charged side chains of basic amino acid
residues may facilitate the formation of enolate ions by
stabilizing them in their ground states and in the transition states
for their formation.
Kinetics of Enolate Ion Formation. Rates of base-catalyzed
enolate ion formation vary with the structure of the carbonyl
molecule. In general, the more acidic the molecule the faster
the ionization reaction. Kresge and Keeffe have constructed a
(
14) Cox, B. G.; De Maria, P; Fini, A. J. Chem. Soc., Perkin Trans. 2
1
984, 44, 3308.

12258 J. Am. Chem. Soc., Vol. 118, No. 49, 1996
Tobin and Frey
transition state and the final ground state for the enolate. The
two solvation sites are more remote from each other than in
the case of a “normal acid”, in which the negative charge is
substantially borne by the atom from which the proton has been
abstracted. In enolate ion formation, therefore, substantial
reorganization of the solvent must take place in the transition
19
state, and this slows the reaction. In a series of ketones closely
related in structure, as represented by the 17 open circles for
benzyl and methyl ketones in Figure 5, the asymmetries of the
solvation requirements are similar, so that the ionization rates
are correlated by the line. In cases in which requirements for
solvation differ from those of ordinary benzyl and methyl
ketones, deviations from the correlation line will be observed.
Such deviations have been referred to as transition state
Figure 5. Correlation of kOH for enolization with pK
Relationship between the rate constants for hydroxide-catalyzed enolate
ion formation, logkOH/p, and carbon acid ionization constants, log(K
p), for carbonyl molecules at 25 °C. Symbols: (open circles) 17 methyl
a
values of ketones.
19
imbalance or imperfect synchronization.
The fact that values of log(kOH/p) for cationic ketones 2, 8,
and 9 lie well above the line in Figure 5 suggests that stabilizing
interactions appear in their enolization transition states that are
more pronounced than in the enolate ions. The positive
deviations correspond to 70-, 920-, and 30-fold enhancements
in rates for ketones 2, 8, and 9, respectively, relative to the rates
that would corresond to their pKa values. The ionization rate
for 2-acetyl-1-methylpyridinium ion 3 is well correlated to its
acidity. In the case of the cyclic ketone 1-methyl-8-oxo-5,6,7,8-
tetrahydroquinolinium ion (4), the carbonyl group appears to
be in position for an electrostatic interaction with the enolate
oxygen; however, this ketone ionizes more slowly than expected
a
/
1a
and benzyl ketones; the correlation line is given by eq 6 ; (filled
3
triangle) 2-acetyl-3,4-dimethylthiazolium ion (2); (filled circle) 2-acetyl-
1
-methylpyridinium ion (3); (filled square) 1-methyl-8-oxo-5,6,7,8-
tetrahydroquinolinium ion; (open square) R-(1-pyridinium)acetophenone
1
4
14
(
8); (open triangle) R-3-(1-methyl-pyridinium)acetophenone (9); (×)
1
6
17
isobutyrophenone; (+) cyclohexanone.
logarithmic correlation of rate constants for hydroxide-catalyzed
enolate formation with pKa values for monocarbonyl compounds
1a
in water at 25 °C. A good linear relationship has been found
for 17 benzyl and methyl ketones,15 as shown by the open circles
in Figure 5. The data were fitted to eq 6, where p is the number
from its pK . This may to be due to the fact that it is a cyclic
a
ketone that is ring-fused to the pyridinium ion, which may
impose conformational constraints on the enolization process.
Through-space electrostatic stabilization that appears to be made
possible by its structure does not overcome the internal
conformational constraints. Examples from the literature of
log(kOH/p) + (0.40 ( 0.01) log(K /p) + (6.47 ( 0.13) (6)
a
of chemically equivalent acidic protons in the carbonyl molecule.
16
1
6,17
other negative deviations include isobutyrophenone and cy-
Cationic ketones 8 and 9 and other selected ketones
are
1
7
clohexanone, which also fall below the line in Figure 5. The
second R-alkyl substituent in isobutyrophenone is thought to
destabilize the negative charge on the incipient enolate ion,
bringing about a transition state imbalance. This effect together
with a probable steric retardation on the deprotonation causes
the slower rate.
included in Figure 5 to illustrate deviations from the overall
correlation. The 2-acetyl-3,4-dimethylthiazolium ion (2) is also
included in Figure 5 as a positive deviation, based on its
3
ionization rate and its estimated pKa (see below). Positive
deviations have been attributed to intramolecular electrostatic
1
a,18
acceleration.
Estimation of pKa for 2-Acetyl-3,4-dimethylthiazolium Ion.
Electrostatic Acceleration. The central fact about enolate
ion formation is that it is very slow when compared with the
dissociation of protons from “normal acids” of comparable pKa.
The most probable reason for the slow rate is suggested by the
structural formulas in eq 7, which illustrates the fact that the
negative charge generated by abstraction of a proton from the
R-carbon resides mainly on the enolate-oxygen atom. The
attacking hydroxide ion must be solvated in the ground state,
whereas the enolate oxygen must be solvated both in the
2
-Acetyl-3,4-dimethylthiazolium ion (2) ionizes extremely
rapidly, and 2-acetylthiamine pyrophosphate (1) ionizes about
3
4
times faster. The ionization rates for these compounds can
be compared with those of other methyl ketones if the ionization
constant for enolate formation can be evaluated. The pKas of
1
and 2 cannot be measured directly owing to their rapid
hydrolysis in basic solutions. However, the acidity of 2 may
be estimated by comparing its ketonic properties with those of
3
described in this paper.
(15) (a) Chiang, Y.; Kresge, A. J.; Tang, Y. S.; Wirz, J. J. Am. Chem.
The inductive effect of the 2-(N-methylthiazolium) substituent
Soc. 1984, 106, 460. (b) Jones, J. R.; Mark, R. E.; Subba Rao, S. C. Trans.
Faraday Soc. 1967, 63, 111. (c) Guthrie, J. P.; Cossar, J.; Klym, A. Can.
J. Chem. 1987, 65, 2154. (d) Keeffe, J. R.; Kresge, A. J.; Toullec, J. Can.
J. Chem. 1986, 64, 1224. (e) Chiang, Y.; Kresge, A. J.; Krough, E. T. J.
Am. Chem. Soc. 1988, 110, 2600. (f) Chiang, Y.; Hojatti, M.; Keeffe, J. R.;
Kresge, A. J.; Schepp, N. P.; Wirz, J. J. Am. Chem. Soc. 1987, 109, 4000.
on the carbonyl group of 2 has been recognized as a factor in
5
its favorable hydration equilibrium. In a series of methyl
ketones, the hydration constants are related to the polar
2
0a
substituent parameter σ* by eq 8.
Hydration does not entail
(g) Chiang, Y.; Kresge, A. J.; Walsh, P. A.; Yin, Y J. Chem. Soc., Chem.
Commun. 1989, 869. (h) Yin, Y. Ph. D. Dissertation, Department of
Chemistry, University of Toronto, 1988. (i) Pollack, R. M.; Mack, J. P. G.;
Eldin, S. J. Am. Chem. Soc. 1989, 111, 6419.
-
log(1/K ) ) (1.70 ( 0.07)Σσ* - (2.81 ( 0.31) (8)
h
(
16) Pruszynski, P.; Chiang, Y.; Kresge, A. J.; Schepp, N. P.; Walsh, P.
an alteration in electrostatic charge, so that the effects of the
A. J. Phys. Chem. 1986, 90, 3760.
(
17) Schepp, N. P. Unpublished work.
(19) (a) Bernasconi, C. F. Pure Appl. Chem. 1982, 54, 2335. (b)
Bernasconi, C. F. Acc. Chem. Res. 1987, 20, 301.
(18) Kirby, A. AdV. Phys. Org. Chem. 1980, 17, 183.

Enolization in Cationic Ketones
J. Am. Chem. Soc., Vol. 118, No. 49, 1996 12259
Table 2. Carbonyl Stretching Frequencies of Acylpyridine and
Acylthiazole Compoundsa
ν (cm-1)
CN
∆ν (cm-1)
b
ketone
CH
3
D
2
O
3 2
CH CN D O
2
-acetylpyridine
1701.4 1691.8
1720.7 1716.9
1715.1
+
- c
N -D Cl
19.3
18.3
25.1
23.3
-4.9
+
I- (3)
N -CH
3
N-Cu2
+ d
1686.9
3
-acetylpyridine
N -D Cl
N -CH
1692.8 1685.5
1711.1 1708.8
1705.1
+
- c
23.3
23.3
+
I-
3
8
-oxy-5,6,7,8-tetrahydroquinoline 1703.4
+
-
N -CH
3
I (3)
1716.9
1689.9
1716.9
13.5
27.0
2
-acetyl-4-methylthiazole
+
-
N -CH
3
I (2)
a
Concentration ) 0.02 M. b
(N-substituted ketone) - ν(neutral ketone). c Ketone
ν
dissolved in 0.10 M DCl/D
O.
O. d Ketone dissolved in 0.20 M CuCl
/
2 2
D
2
Figure 6. Correlation of pK
nones. Literature values of pK
are plotted for the compounds p-methoxyacetophenone, p-methylac-
etophenone, acetophenone, p-fluoroacetophenone, p-chloroacetophe-
none, p-bromoacetophenone, and p-nitroacetophenone. The correlation
a
with σ* in phenyl-substituted acetophe-
and σ* for their phenyl substituents21
The pKa of 2 is remarkably low. We know of no other methyl
a
ketone that is so acidic. Other ketones that are comparably
acidic contain two or more activating functional groups on the
ionizing carbon atom, one example being acetylacetone (pKa
)
9).
line is given by pK ) (-2.13 ( 1.05)σ* + 19.79.
a
Spectroscopic Properties of Cationic Ketones. The induc-
tive acid-strengthening effect of the 2-(N-methylthiazolium)
substituent in 2 is larger than that of the 2-(N-methylpyridinium)
substituent in 3. Examination of the spectroscopic properties
of the ketonic groups can give information about whether the
acid strengthening features of the substituents may be attributed
in part to differential destabilization of the keto forms. The
carbonyl stretching frequency is sensitive to the electronic
structure and is, therefore, sensitive to substituent inductive
effects. The carbonyl stretching frequencies are given in Table
2
-(N-methylthiazolium) and 2-(N-methylpyridinium) substituents
on the hydration of 2 and 3 should be attributed to their inductive
effects. On the basis of the published value of Kh for 2 and
3
the present evaluation of Kh for 3, the values of σ* calculated
from eq 8 are 1.68 and 1.02 for 2 and 3, respectively. These
values of σ* can be used to estimate the inductive effects of
the 2-(N-methylthiazolium) and 2-(N-methylpyridinium) sub-
stituents on the pKas of 2 and 3.2
0b
-1
2
. The frequency for 3-acetylpyridine is 1680-1690 cm , and
-
1
A linear correlation between pKa and σ* values of R-groups
in a series of phenyl-substituted acetophenones is shown in
Figure 6, which is plotted using literature data.21 This correla-
tion should pertain to 2 and 3 because all of the data are for
for 2-acetylpyridine, it is 1690-1701 cm , depending on the
solvent. A slight shift of ν to lower frequency upon changing
from organic solvent to D2O may result from the increase in
polarity and/or the increase in hydrogen bonding.
2
methyl ketones bonded to aromatic rings through sp carbons.
Methylation and deuteration of the ring nitrogens lead to 10-
-
1
Therefore, the σ* values for 2 and 3 can be used with the
correlation line in Figure 6 to determine what the pKas of 2
and 3 would be based soley on the inductive effects of 2-(N-
methylthiazolium) and 2-(N-methylpyridinium) substituents.
This operation gives values of 15.1 and 17.4 as the pKas for 2
and 3, respectively. The measured pKa for 3 (11.13,Table 1) is
27 cm shifts in the CdO frequency for the ketones. It is
significant that the carbonyl stretching frequencies for the
cationic forms are relatively unaffected by solvent. The
explanation of this in the case of 3-acetylpyridine has been
suggested to be that resonance forms carrying positive charge
on ring carbons make a significant contribution to the electronic
2
3
structures of aromatic ketones. In the quaternary system,
resonance forms that tend to increase positive charge on the
positively charged ring are disfavored, so that the contributions
of these forms are less important than in the uncharged parent
ketones. The νCdO for the N-methylated forms are much more
6
.3-log units lower than that predicted by the inductive effect.
The additional acid strengthening should be attributed to
through-space electrostatic stabilization of the enolate ion by
the cationic nitrogen. On structural grounds, we believe that
the same electrostatic effect should increase the acidity of of
the 2-acetyl-3,4-dimethylthiazolium ion (2). We therefore
estimate the pKa of 2-acetyl-3,4-dimethylthiazolium ion as 15.1
-
1
like unperturbed carbonyl stretches (1700-1720 cm ). As
shown in the bottom line of Table 2, the effect of the quaternary
nitrogen on νCdO in the thiazolium system is substantially more
pronounced than in the pyridinium system. This can be
correlated with the greater inductive effect in the thiazolium
system. The greater electrophilicity of the 2-(N-methylthiazo-
-
6.3, or 8.8.22
(20) (a) Greenzaid, P.; Luz, Z.; Samuel, D. J. Am. Chem. Soc. 1967, 89,
7
49. (b) A similar analysis could be applied to 2-acetylthiamine pyrophos-
4
a
phate 1, for which values of (Kh)D O are available. However it is
(22) A much less reliable estimate of pKa for 2 could be obtained by
placing its value for log(kOH/3) on the correlation line in Figure 5. The
resulting value of 4.2 is an unreasonably low pKa for a methyl ketone, 4.6-
log units lower than the much more reasonable estimate of 8.8 found above
and shown in Table 1. This supports our conclusion that 2 enolizes much
more rapidly than can be accounted for by its pKa, and it must result from
an internal lowering of the transition state energy for enolization. The most
obvious source of this is an electrostatic interaction of the positively charged
ring with the developing negative charge in the transition state for
enolization.
2
questionable whether the hydration of 1 can be correlated with that of 2 or
other ordinary ketones because of the steric interference with hydration
that can be expected on the part of the N-pyrimidinyl group on the thiazolium
ring of 1. The pyrimidinyl group lies near the ketonic function, and for this
reason the major form of 1 in aqueous solution is the internally cyclized
carbinolamine form, in which the pyrimidinyl amino group forms an adduct
with the acetyl carbonyl group.
21) The σ* values are from: Perrin, D. D.; Dempsey, B.; Sergeant, E.
P. pKa Prediction for Organic Acids and Bases; Chapman and Hall: New
York, 1981. The pKa values are from: Keeffe, J. R.; Kresge, A. J. In The
Chemistry of Enols; Rappoport, Z., Ed.; Wiley: New York, 1990.
4
a
(
(23) Patrick, D. E., II; Wilson, J. E.; Leroi, G. E. Biochemistry 1974,
13, 2813.

12260 J. Am. Chem. Soc., Vol. 118, No. 49, 1996
Tobin and Frey
lium) group may be due to its lesser aromaticity relative to the
ketones, thio esters, and carboxylic acids at very fast rates (ms)
at pH 7. Solvent reorganization should not retard enzymatic
enolization, because solvation of the enolate can be expected
2-(N-methylpyridinium) group, which should allow the positive
charge to be more highly dispersed in the pyridinium ring. A
lesser dispersal of positive charge in the thiazolium ring would
tend to make carbon-2 more electropositive than it is in the
pyridinium ring.
19b
to be provided by the enzymatic binding site.
However, high
pKa values remain as significant barriers. Two major classes
of interactions between carbonyl compounds and enzymatic sites
have been proposed to contribute to overcoming these barriers,
electrostatic acceleration by metal ion cofactors or basic amino
Through-Space Electrostatic Acceleration of Enolization.
The through-space electrostatic effects of quaternary nitrogen
on the ionization rates of 2 and 3 can be conceptually separated
into two categories: (1) electrostatic stabilization of the enolate
ion in its ground state, which manifests itself as increased carbon
acidity (decreased pKa), and (2) additional electrostatic stabiliza-
tion of the incipient enolate oxyanion in the transition state,
which is manifested in a positive deviation from the correlation
line in Figure 5. The through-space electrostatic effect on the
carbon acidities of 2 and 3 is here estimated as corresponding
to a 6.3-log unit downward perturbation of pKa, and it
corresponds to a 330-fold acceleration of the enolization rate
for both compounds. The additional through-space electrostatic
stabilization of the transition state in the case of 2 leads to an
additional 70-fold rate enhancement. The latter effect may be
attributed to a decreased requirement for solvent reorganization
in the transition state owing to electrostatic stabilization. The
overall through-space electrostatic acceleration for 2 is, there-
2
acid groups in the active site, and low-barrier hydrogen bonds
formed between substrate carbonyl groups and electrophilic
2
5
groups in the active site. We have here described the fast
enolization of two cationic ketones in which hydrogen bonding
and metal coordination can play no role.
The present results show that purely electrostatic, through-
space stabilization of the enolate ions and the transition states
for enolate formation for 2 and 3 in aqueous solutions are
significant. Analogous electrostatic effects in enzymatic active
sites can be expected to be at least comparable, and even larger
when the effective dielectric constants are smaller than that of
water. Cationic groups of enzymes that may participate in
enolization include the basic groups of amino acid side chains
such as lysine, arginine, and histidine. All of these groups can
contribute hydrogen bonds as well as positive charges toward
stabilizing enolates and transition states, and the hydrogen bonds
will further enhance enolization rates. Metal ions can be as
effective or more effective than basic groups. Low-barrier
hydrogen bonds may be brought into play, especially in sites
lacking metal ions or cationic amino acids, such as the
enolization site of triose phosphate isomerase.
4
24
fore, (330)(70) or 2.3 × 10 .
The inductive effects of the 2-(N-methylthiazolium) and 2-(N-
methylpyridinium) substituents on enolization rates can also be
estimated. Using acetone (pKa ) 19.3) as a standard methyl
ketone, and ignoring for the moment the through-space elec-
trostatic contributions to the acidities, the electron withdrawing
properties of the 2-(N-methylthiazolium) and 2-(N-methylpy-
ridinium) substituents would have the effect of lowering the
pKa values to 15.1 and 17.4 for 2 and 3, respectively. On the
basis of eq 6, the values of kOH for methyl ketones exhibiting
these pKas will be 46- and 5.7-fold higher, respectively, than
that for acetone. The inductive effect of the 2-(N-methylthi-
azolium) group on the rate is larger than for the 2-(N-
methylpyridinium) group, and both are smaller than the through-
space acceleration.
Acknowledgment. We thank Ray Hansen of the Update
Instrument Co. for the use of stopped flow equipment. We
thank Professor Charles P. Casey of the Department of
Chemistry for allowing us to use his ozonolysis apparatus. This
research was supported by Grant No. DK 28607 from the
National Institute of Diabetes and Digestive and Kidney
Diseases, USPHS. This study made use of the National
Magnetic Resonance Facility at Madison, which is supported
by NIH Grant RR02301 from the Biomedical Research Tech-
nology Program, National Center for Research Resources.
Equipment in the facility was purchased with funds from the
University of Wisconsin, the NSF Biological Instrumentation
Program (Grant DMB-8415048), NIH Biomedical Research
Technology Program (Grant RR02301), NIH Shared Instru-
mentation Program (Grant RR02781), and the U.S. Department
of Agriculture.
Implications of Electrostatic Acceleration for Enzymatic
Enolizations. Enzymes catalyze enolizations of aldehydes,
(24) A reviewer has suggested that a part of the inductive effect may
arise from a field or through-space effect. We have here taken the inductive
effect to be that exerted by the electropositivity of the substituent carbon,
carbon-2 of 2 or 3, on the methyl ketone groups, whatever the origins o2f
electropositivity. In 2 and 3, the sources of electropositivity are the sp
character of carbon-2 and electron withdrawal by the quaternary nitrogen
to which it is bonded. In these cases, the most direct and simple medium
for electropositivity to be transmitted from the quaternary nitrogen to
carbon-2 is through the covalent linkage between them. It is difficult to
imagine a nonbonded field between nitrogen-1 and carbon-2 that would be
a better conduit for electron withdrawal than the bond linking them.
JA950573P
(25) (a) Gerlt, J. A.; Gassman, P. G. J. Am. Chem. Soc. 1993, 115, 1152.
(b) Gerlt, J. A.; Gassman, P. G. Biochemistry 1993, 32, 11943.