Mechanistic imperatives for catalysis of aldol addition reactions: Partitioning of the enolate intermediate between reaction with bronsted acids and the carbonyl group

Article abstract of DOI:10.1021/ja9900297

The lyoxide ion catalyzed intramolecular aldol addition reaction of 2- (2-oxopropyl)benzaldehyde (1) to give the aldol adduct 3 proceeds via essentially irreversible formation of the acetone-like enolate intermediate 2, because reprotonation of 2 by a solvent of H₂O or D₂O (k(HOH) or k(DOD)) is much slower than intramolecular addition of the enolate to the carbonyl group (k(c)). The aldol addition reaction of 1 catalyzed by high concentrations of 3-substituted quinuclidine buffers proceeds via reversible deprotonation of 1 to give the enolate 2, and rate-determining addition of the enolate to the carbonyl group. A rate constant ratio of k(c)/k(HOH) = 35 was determined for partitioning of the enolate 2 between intramolecular addition to the carbonyl group and protonation by solvent water. The corresponding ratios k(BH)/k(c) (M^-1) for the protonation of 2 by Bronsted buffer acids and intramolecular aldol addition increase from 7 to 450 as the acidity of the buffer acid is increased from pK(BH) = 11.5 to 7.5. The data show that the electrophilic reactivity of the benzaldehyde carbonyl group toward intramolecular addition of the enolate 2 is the same as that of a hypothetical tertiary ammonium cation of pK(BH) = 13.3. The Marcus intrinsic barrier for addition of the enolate 2 to the carbonyl group is unexpectedly small, which suggests that the transition state for this reaction is stabilized by interactions between the soft-soft acid-base pair. The relevance of this work to chemical and enzymatic catalysis of aldol condensation reactions is discussed.

Full text of DOI:10.1021/ja9900297

J. Am. Chem. Soc. 1999, 121, 4763-4770
4763
Mechanistic Imperatives for Catalysis of Aldol Addition Reactions:
Partitioning of the Enolate Intermediate between Reaction with
Brønsted Acids and the Carbonyl Group
,1
†
John P. Richard* and R. W. Nagorski
Contribution from the Department of Chemistry, UniVersity at Buffalo, SUNY,
Buffalo, New York 14260-3000
ReceiVed January 4, 1999
Abstract: The lyoxide ion catalyzed intramolecular aldol addition reaction of 2-(2-oxopropyl)benzaldehyde
(1) to give the aldol adduct 3 proceeds via essentially irreversible formation of the acetone-like enolate
intermediate 2, because reprotonation of 2 by a solvent of H2O or D2O (kHOH or kDOD) is much slower than
intramolecular addition of the enolate to the carbonyl group (kc). The aldol addition reaction of 1 catalyzed by
high concentrations of 3-substituted quinuclidine buffers proceeds via reversible deprotonation of 1 to give
the enolate 2, and rate-determining addition of the enolate to the carbonyl group. A rate constant ratio of
kc/kHOH ) 35 was determined for partitioning of the enolate 2 between intramolecular addition to the carbonyl
-
1
group and protonation by solvent water. The corresponding ratios kBH/kc (M ) for the protonation of 2 by
Brønsted buffer acids and intramolecular aldol addition increase from 7 to 450 as the acidity of the buffer acid
is increased from pKBH ) 11.5 to 7.5. The data show that the electrophilic reactivity of the benzaldehyde
carbonyl group toward intramolecular addition of the enolate 2 is the same as that of a hypothetical tertiary
ammonium cation of pKBH ) 13.3. The Marcus intrinsic barrier for addition of the enolate 2 to the carbonyl
group is unexpectedly small, which suggests that the transition state for this reaction is stabilized by interactions
between the soft-soft acid-base pair. The relevance of this work to chemical and enzymatic catalysis of
aldol condensation reactions is discussed.
The direct determination of rate constants for the protonation
Scheme 1
of simple enolates by solvent water (kHOH) and buffer catalysts
(
kBH) have provided a detailed description of substituent effects
2
on the reactivity of these carbanions. Similarly, the addition
of enolates to carbonyl electrophiles in water, a reaction that
occurs broadly in chemical3 and biochemical syntheses of
carbon-carbon bonds, has been examined in Guthrie’s labora-
tory and estimates of rate constants, kc, for these reactions
a
3b
(kBH/kc) toward a simple enolate, by determining directly the
relative rates for partitioning of an enolate between reaction with
these electrophiles. Such kinetic studies have a broad signifi-
cance in chemistry and biology.
4
-6
reported.
While rate constants kBH and kc (Scheme 1) have been
determined or estimated in separate experimental studies of the
reactions of a variety of enolates with a variety of Brønsted
acids and carbonyl groups (Scheme 1), we are not aware of
any attempts to quantify these rate constants by directly
monitoring the partitioning of a simple enolate between proto-
nation and addition to the carbonyl group. The present experi-
ments were initiated to obtain a precise determination of the
relative electrophilicity of Brønsted acids and the carbonyl group
1. The magnitude of kBH/kc determines the yield of the aldol
adduct from partitioning of an enolate between protonation and
addition to the carbonyl group in water, so that good yields of
the aldol adduct will be obtained from the reaction of an enolate
7
,8
in water when this rate constant ratio is small.
2
. The rate-limiting step for aldol addition depends on the
relative barriers to partitioning of the enolate intermediate
between protonation (kBH) and addition to the carbonyl group
(kc). When kc . kBH, deprotonation of the substrate to give the
enolate, kB, is effectively irreversible and therefore rate-
determining for the overall aldol addition reaction. However,
when kBH . kc, the observed rate constant for aldol addition
will depend on kc, and the reaction will be accelerated by
reagents that lower the barrier to this rate-limiting step.
Therefore, a knowledge of the rate-determining step for aldol
addition is important for the design of simple chemical catalysts
†
Current address: Department of Chemistry, Illinois State University,
Normal, IL 61790-4160.
(
1) Tel: (716) 645 6800 ext. 2194. Fax: (716) 645 6963. E-mail:
jrichard@chem.buffalo.edu.
2) Keeffe, J. R.; Kresge, A. J. In The Chemistry of Enols; Rappoport,
Z., Ed.; Wiley: Chichester, UK, 1990; pp 399-480.
3) (a) Heathcock, C. H. In Asymmetric Synthesis; Morrison, J. D., Ed.;
(
(
Academic Press: New York, 1984; Vol. 3; pp 111-212. (b) Takayama,
S.; McGarvey, G. J.; Wong, C.-H. Annu. ReV. Microbiol. 1997, 51, 285-
3
10.
(
(
(
4) Guthrie, J. P. J. Am. Chem. Soc. 1991, 113, 7249-7255.
5) Guthrie, J. P.; Guo, J. J. Am. Chem. Soc. 1996, 118, 11472-11487.
6) Guthrie, J. P.; Barker, J. A. J. Am. Chem. Soc. 1998, 120, 6698-
(7) Bartlett, P. A.; Satake, K. J. Am. Chem. Soc. 1988, 110, 1628-1630.
(8) Bartlett, P. A.; McLaren, K. L.; Marx, M. A. J. Org. Chem. 1994,
59, 2082-2085.
6
703.
1
0.1021/ja9900297 CCC: $18.00 © 1999 American Chemical Society
Published on Web 05/07/1999

4764 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Richard and Nagorski
9
of this reaction, and to an understanding of the imperatives
roquinuclidine hydrochloride, methanol/water; quinuclidine hydrochlo-
ride, ethanol. All other chemicals were reagent grade and were used
without further purification. The water used for kinetic and HPLC
studies was distilled and then passed through a Milli-Q water purifica-
tion system. The procedures for the preparation of 2-(2-oxopropyl)-
benzaldehyde (1) from 2-indanone and for the preparation of solutions
for kinetic studies are described in the Supporting Information.
Deuterium Exchange Reactions of 1. The deuterium exchange
10
for the catalysis of aldol addition by enzymes and enzyme
mimics.
. The determination of kBH/kc will provide an accurate value
for the relatiVe Marcus intrinsic barriers for partitioning of an
enolate between reaction with Brønsted acids and the carbonyl
1
1
3
4,5
group, provided the relative thermodynamic driving force for
reaction with the two electrophiles is also known.
reactions of 1 ([S] ) 3.5-5 mM) in D O at 25 °C and I ) 1.0 (KCl)
2
1
14,15
were monitored by H NMR spectroscopy,
using the procedures
described in the Supporting Information.
1
H NMR spectra were recorded in CDCl
3
at 25 °C on a Varian VXR-
4
00S spectrometer. Chemical shifts were referenced to CHCl
ppm. Relaxation times of the R-CH and R-CH protons of 1 were
determined to be in the range T ) 4-6 s. Spectra (32-64 transients,
0 s relaxation delay) were obtained using a sweep width of 5300 Hz,
3
at 7.27
2
3
1
6
a 90° pulse angle, and an acquisition time of 6 s.
The exchange for deuterium of the first benzylic proton of 1 in D O
2
was followed by monitoring the disappearance of the singlet at 4.146
ppm due to the R-CH group and the appearance of the triplet due to
2
the R-CHD group which is shifted 0.028 ppm upfield from the singlet.
Reaction progress, RCH2, was calculated using eq 1,16 where ACH2 and
A
CHD are the integrated areas of the singlet and triplet for the R-CH
2
It is difficult to obtain and interpret rate data for bimolecular
aldol addition because these reactions do not proceed cleanly
toward formation of a single stable product, but rather give a
and R-CHD groups, respectively. The exchange for deuterium of the
first proton of the R-CH group of 1 in D O was followed by monitoring
the disappearance of the singlet at 2.327 ppm due to the R-CH
and the appearance of the triplet due to the R-CH D group which is
shifted 0.016 ppm upfield from the singlet. Reaction progress, RCH3
was calculated using eq 2,16 where ACH3 and ACH D are the integrated
3
2
3
group
12
complex mixture of unstable products. By contrast, intramo-
lecular aldol addition reactions can be designed to proceed
cleanly toward formation of a single stable product.5 We have
shown that the intramolecular aldol condensation reaction of 1
at low concentrations (0.10-1.0 mM) proceeds through the
2
,
,13
2
areas of the singlet and triplet for the R-CH
respectively.
3
2
and R-CH D groups,
Semilogarithmic plots (not shown) of RCH2 or RCH3 against time
1
3
enolate intermediate 2 to give a good yield of 2-naphthol (4).
according to eqs 3 and 4 were linear during exchange for deuterium of
Analysis of the kinetic data for aldol addition and for the initial
deprotonation of 1 allows for the determination of rate constant
ratios kBH/kc for partitioning of the enolate 2 between protonation
by Brønsted acids and intramolecular addition to the carbonyl
group. These results provide a quantitative description of the
relative electrophilicities of Brønsted acids and the carbonyl
group toward a simple enolate, and insight into the imperatives
for catalysis of aldol addition by enzymes and enzyme mimics.
up to 35% of the first proton of the R-CH
2
3
or the R-CH group of 1.
The negative slopes of the former plots are equal to the statistically
corrected rate constant kobsd/2 for reaction of a single proton of the
R-CH
first proton of the R-CH
latter plots are equal to the statistically corrected rate constant kobsd/3
for reaction of a single proton of the R-CH group of 1, where kobsd is
2
group of 1, where kobsd is the rate constant for exchange of the
group (eq 3).16 The negative slopes of the
2
3
the rate constant for exchange of the first proton of the R-CH
3
group
(
eq 4).16 The values of kobsd were reproducible to (10%.
Experimental Section
A
CH2
Materials. 2-Indanone, methylmagnesium iodide, 3-quinuclidinone
hydrochloride, 3-quinuclidinol, 3-chloroquinuclidine hydrochloride,
quinuclidine hydrochloride, and potassium deuterioxide (40 wt %,
RCH2
)
)
(1)
A
+ ACHD
CH2
9
8+% D) were from Aldrich. Deuterium oxide (99.9% D), deuterium
A
CH3
chloride (35% w/w, 99.5% D), and CDCl (99.8% D) were from
3
RCH3
(2)
A
Cambridge Isotope Laboratories. The 3-substituted quinuclidines were
purified by recrystallization from the following solvents: 3-quinucli-
dinone hydrochloride, ethanol/water; 3-quinuclidinol, acetone; 3-chlo-
CH D
2
A
+
CH3
2
(
9) Chaperon, A. R.; Engeloch, T. M.; Neier, R. Angew. Chem., Int. Ed.
ln RCH2 ) -kobsdt/2
ln RCH3 ) -kobsdt/3
(3)
(4)
1
998, 37, 358-360. Chen, C.-T.; Chao, S.-D.; Yen, K.-C.; Chen, C.-H.;
Chou, I.-C.; Hon, S.-W. J. Am. Chem. Soc. 1997, 119, 11341-11342. Loh,
T.-P.; Pei, J.; Koh, K. S.-V.; Cao, G.-Q.; Li, X.-R. Tetrahedron Lett. 1997,
3
8, 3465-3468. Yanagisawa, A.; Matsumoto, Y.; Nakashima, H.; Asakawa,
K.; Yamamoto, H. J. Am. Chem. Soc. 1997, 119, 9319-9320. Carreira, E.
M.; Lee, W.; Singer, R. A. J. Am. Chem. Soc. 1995, 117, 3649-3650.
Yamada, Y. M. A.; Yoshikawa, N.; Sasai, H.; Shibasaki, M. Angew. Chem.,
Int. Ed. Engl. 1997, 36, 1871-1873.
Aldol Condensation Reactions of 1. All reactions were carried out
at 25 °C and ionic strength 1.0 (KCl). Reactions were initiated by
making a 2500-fold dilution of a solution of 1 in acetonitrile into 3
mL of the appropriate reaction mixture to give a final substrate
concentration of 0.1-0.3 mM. Rate constants for the conversion of 1
to 2-naphthol (4) were determined spectrophotometrically by following
(
10) Fessner, W.-D.; Schneider, A.; Held, H.; Sinerius, G.; Walter, C.;
Hixon, M.; Schloss, J. V. Angew. Chem., Int. Ed. Engl. 1996, 35, 2219-
2
221.
(11) Desper, J. M.; Breslow, R. J. Am. Chem. Soc. 1994, 116, 12081-
1
1
3
2082. Koh, J. T.; Delaude, L.; Breslow, R. J. Am. Chem. Soc. 1994, 116,
1234-11240. Breslow, R.; Desper, J.; Huang, Y. Tetrahedron Lett. 1996,
7, 2541-4.
(14) (a) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 1992, 114,
10297-10302. (b) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 1996,
118, 3129-3141.
(15) Rios, A.; Richard, J. P. J. Am. Chem. Soc. 1997, 119, 8375-8376.
Richard, J. P.; Williams, G.; Gao, J. J. Am. Chem. Soc. 1999, 121, 715-
726.
(12) Bell, R. P. J. Chem. Soc. 1937, 1637-1640. Bell, R. P.; Lidwell,
O. M. Proc. R. Soc. London, Ser. A 1940, 176, 88-121. Bell, R. P.; Smith,
M. Y. J. Chem. Soc. 1958, 1691-1696. Bell, R. P.; McTigue, P. T. J. Chem.
Soc. 1960, 2983-2994. Guthrie, J. P. Can. J. Chem. 1974, 52, 2037-2040.
(13) Nagorski, R. W.; Mizerski, T.; Richard, J. P. J. Am. Chem. Soc.
(16) Halkides, C. J.; Frey, P. A.; Tobin, J. B. J. Am. Chem. Soc. 1993,
115, 3332-3333.
1
995, 117, 4718-4719.

Catalysis of Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4765
Scheme 2
the appearance of 4 at 340 nm (usual method) or in some cases the
-
1
disappearance of 1 at 256 nm. First-order rate constants, kobsd (s ),
were determined from the slopes of semilogarithmic plots of reaction
progress against time, which were linear for at least three reaction
halftimes. The values of kobsd were reproducible to (5%.
Rate constants for the slow reactions of 1 in the presence of low
concentrations of 3-quinuclidinone at pH 8.8 (pH maintained with 10
mM pyrophosphate buffer) were determined from the initial velocity
of formation of 4 during the first 6% of the reaction, with the assumption
that the total change in absorbance at 340 nm is the same as that
observed for the faster reaction of an equal concentration of 1 in the
presence of high concentrations of 3-quinuclidinone.
Figure 1. Dependence of kobsd (s-1) for the conversion of 1 to
2-naphthol (4) in H O on the total concentration of 3-substituted
2
quinuclidine buffer at 25 °C (I ) 1.0, KCl). The solid lines are the
nonlinear least-squares fits of the data to eq 5, calculated as described
+
in the text. (A) Catalysis by quinuclidine buffers. Key: (b), [B]/[BH ]
+
) 1.0; (9), [B]/[BH ] ) 0.25. (B) Catalysis by 3-quinuclidinol buffers.
+
+
+
Key: (1), [B]/[BH ] ) 4; (b), [B]/[BH ] ) 1.0; (9), [B]/[BH ] )
+
HPLC product analyses were carried out as described in earlier
work1 with peak detection by a Waters 996 diode array detector.
0.25. Catalysis by 3-chloroquinuclidine buffers: ([), [B]/[BH ] ) 9.0.
7,18
19
+
(C) Catalysis by 3-quinuclidinone (pKBH ) 7.5), [B]/[BH ] ) 20, at
pH 8.8 in the presence of 10 mM pyrophosphate buffer.
Results
Figure S2B of the Supporting Information shows the depen-
Figure S1 of the Supporting Information shows representative
partial H NMR spectra at 400 MHz of 1 recovered after its
incubation in D2O buffered with 3-quinuclidinone at pD 8.3
and 25 °C (I ) 1.0, KCl). At early reaction times (Figure S1A)
the disappearance of the singlet at 4.146 ppm due to the benzylic
R-CH2 group is accompanied by the appearance of a triplet at
.118 ppm (J HD ) 2 Hz) due to the R-CHD group (Scheme
A). At longer times (Figures S1B and S1C) the disappearance
of the singlet at 2.327 ppm due to the R-CH3 group is
accompanied by the appearance of a triplet at 2.311 ppm (J HD
-1
1
dence of kobsd (s ) for exchange for deuterium of the first proton
of the R-CH3 group of 1 on the concentration of deuterioxide
ion in D2O at 25 °C (I ) 1.0, KCl). The slope of this correlation
-
1
-1
is (kDO)ex ) 0.024 M s , the second-order rate constant for
exchange of the first proton of the R-CH3 group of 1 catalyzed
by deuterioxide ion. Signals for the aromatic protons of
4
2
1
2
-naphthol (4) were detected by H NMR spectroscopy during
the course of these exchange reactions.
The aldol condensation reaction of 1 to give 2-naphthol (4)
in 0.1 M potassium hydroxide in H2O at 25 °C (I ) 1.0, KCl)
was monitored spectrophotometrically between 200 and 400 nm.
The maximum decrease in absorbance due to the disappearance
of 1 occurs at 256 nm, and the maximum increase in absorbance
due to the formation of 4 occurs at 340 nm. Sharp isosbestic
points were observed at 250, 267, 299, and 307 nm. No reaction
intermediates were detected by HPLC analysis with peak
detection at 256 nm during the conversion of 1 to 4 in the
presence of 3-quinuclidinone buffer at pH 7.0.
)
2 Hz) due to the R-CH2D group (Scheme 2B). Semiloga-
rithmic plots (not shown) of reaction progress against time
according to eqs 3 and 4 were linear during exchange of 35%
of the first proton of the R-CH2 or the R-CH3 group of 1. The
negative slopes of these plots are equal to kobsd/2 (R-CH2 group,
-
1
eq 3) or kobsd/3 (R-CH3 group, eq 4), where kobsd (s ) is the
observed first-order rate constant for exchange of the first proton
1
6
of the R-CH2 group or the R-CH3 group of 1 (Scheme 2).
-
1
The slope of the linear plot of kobsd (s ) for exchange for
deuterium of the first proton of the R-CH2 group of 1 against
the concentration of the basic form of 3-quinuclidinone buffer
in D2O at pD 8.3 and 25 °C (I ) 1.0, KCl) (not shown) is kB
s , the second-order rate constant for
exchange of the first proton of the R-CH2 group of 1 catalyzed
by 3-quinuclidinone (Scheme 2A).
First-order rate constants for the reaction of 1 to give 4 were
determined spectrophotometrically by following the appearance
of the product at 340 nm. Figure 1A shows the dependence of
-
2
-1 -1
-1
)
4.0 × 10
M
kobsd (s ) for the conversion of 1 to 4 in H2O on the
concentration of quinuclidine buffers at 25 °C (I ) 1.0, KCl)
and Figure 1B shows the corresponding data for 3-quinuclidinol
and 3-chloroquinuclidine buffers. Figure 1C shows the depen-
Figure S2A of the Supporting Information shows the linear
-
1
-1
dence of kobsd (s ) for the conversion of 1 to 4 on the
dependence of kobsd (s ) for exchange for deuterium of the first
proton of the R-CH3 group of 1 on the concentration of the
basic form of 3-quinuclidinone buffer in D2O at pD ) 7.7 or
1
9
concentration of 3-quinuclidinone (pKBH ) 7.5) in solutions
that were buffered with 10 mM pyrophosphate at pH 8.8 and
2
5 °C (I ) 1.0, KCl).
8
×
.3 and 25 °C (I ) 1.0, KCl). The slope of this plot is kB ) 3.4
-
4
-1 -1
Figure S2B of the Supporting Information shows the depen-
10
M
s , the second-order rate constant for exchange
-
1
dence of kobsd (s ) for the conversion of 1 to 4 on the
concentration of hydroxide ion in H2O and deuterioxide ion in
D2O at 25 °C (I ) 1.0, KCl). The slopes of these plots are
of the first proton of the R-CH3 group of 1 catalyzed by
-quinuclidinone (Scheme 2B).
3
(
17) Richard, J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc.
984, 106, 1361-1372.
18) Richard, J. P. J. Am. Chem. Soc. 1989, 111, 1455-1465.
1
(19) Gresser, M. J.; Jencks, W. P. J. Am. Chem. Soc. 1977, 99, 6963-
6980.
(

4766 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Richard and Nagorski
Scheme 3
Scheme 4
Scheme 5
Mechanism of Aldol Addition. The second-order rate
constant for exchange for deuterium of the first proton of the
R-CH3 group of 1 catalyzed by deuterioxide ion in D2O, (kDO)ex
-
1
-1
-1
)
0.024 M s , is 6-fold smaller than (kDO)aldol ) 0.15 M
-
1
s
for the deuterioxide ion catalyzed aldol addition reaction
of 1 in D2O to give 3. However, this value of (kDO)aldol is very
close to the expected second-order rate constant for deproto-
nation of 1 by deuterioxide ion to give the acetone-like enolate
2
2,23
2
.
These results show that the deuterioxide ion catalyzed
incorporation of deuterium into the R-CH3 group of 1 is slower
than both the deprotonation and aldol condensation reactions
1
of 1. This is consistent with the detection of 4 by H NMR
spectroscopy during these deuterium exchange reactions (see
1
Results). By contrast, no 4 was detected by H NMR spectros-
kHO)aldol ) 0.10 M-1 s-1 and (kDO)aldol ) 0.15 M-1 s , the
second-order rate constants for the reaction of 1 to give 4
catalyzed by hydroxide in H2O and deuterioxide ion in D2O,
respectively.
-1
copy during exchange for deuterium of 35% of the first proton
of the R-CH3 group of 1 in D2O buffered with g0.10 M
(
3
-quinuclidinone at pD ≈ 8.0. Therefore, under these reaction
conditions, the partitioning of the enolate intermediate 2 strongly
favors the formation of deuterium-labeled 1 over intramolecular
aldol addition to give 3.
Discussion
These data show that the aldol addition reaction of 1 proceeds
by a two-step mechanism in which lyoxide ion catalyzed
deprotonation of 1 to give the enolate 2 is effectively irreversible
and rate-determining for the formation of 3, so that the observed
second-order rate constant for lyoxide ion catalyzed aldol
addition is essentially equal to the rate constant for deprotonation
of 1, (kLO)aldol ) kLO (Scheme 5). The observed second-order
rate constant for deuterioxide ion catalyzed exchange of
deuterium into the R-CH3 group of 1 is smaller than that for
deprotonation of 1, (kDO)ex < kDO, because the enolate inter-
mediate 2 undergoes ring closure (kc) faster than its reprotonation
by solvent D2O (kc > kDOD, Scheme 5). By contrast, the strong
buffer catalysis of proton transfer from 1 results in a change to
fast and reversible proton transfer in the presence of high
The observation of several isosbestic points when the
intramolecular reaction of 1 to give 2-naphthol (4) was followed
spectrophotometrically and the absence of the formation of any
detectable intermediates as determined by HPLC analysis shows
that there is no significant accumulation of 3 and 5 during these
reactions. This is a consequence of the large overall thermo-
dynamic driving force for conversion of 1 to 4 (ca. 23 kcal/
mol) and the nearly equal division of the overall change in the
standard Gibbs free energy between the three reaction steps, as
determined in the elegant analysis of Guthrie et al. (Scheme
5
3
).
4
A rate constant ratio of kelim/kretro ) 3.5 × 10 has been
estimated for the partitioning of 6 between elimination of water
5
and retroaldol cleavage (Scheme 4). An even larger value of
+
concentrations of 3-quinuclidinone buffer, kBD[BD ] . kc
this rate constant ratio is expected for the partitioning of 3 for
which the elimination product 5 is extensively conjugated.
Similarly, the thermodynamically favorable tautomerization of
(
Scheme 5), and a change in rate-limiting step for the overall
-
aldol addition from proton transfer to form 2, kDO[DO ] +
kB[B], to the ring closure reaction of 2, kc (Scheme 5).
5
to give 4 is expected to be much faster than the unfavorable
This change in rate-determining step for aldol addition is
addition of water to 5 to give 3. We therefore conclude that the
formation of 3 from 1 is effectively irreversible and rate-
determining for the formation of 4. Therefore, the observed rate
constants for the reaction of 1 to give 4 determined in this work
represent rate constants for the intramolecular aldol addition
reaction of 1 to give 3 (Scheme 5).
-1
responsible for the downward curvature in the plots of kobsd (s )
for the conversion of 1 to 3 in H2O against the concentration
of the buffer catalyst (Figure 1). At low buffer concentrations
where proton transfer is rate-determining for the overall aldol
+
addition (kc . kHOH + kBH[BH ]) the general base catalysis of
proton transfer results in a linear dependence of kobsd on buffer
concentration (Scheme 5). However, on proceeding to high
concentrations of the buffer catalyst the change in rate-limiting
step from proton transfer to intramolecular addition of the
The exchange for deuterium of the first proton of the R-CH2
and R-CH3 groups of 1 (Scheme 2) in D2O at 25 °C (I ) 1.0,
1
KCl) was followed by H NMR spectroscopy (Figure S1 of the
Supporting Information). The second-order rate constants for
catalysis of these exchange reactions by 3-quinuclidinone are
(
21) A pKa of 19.3 has been determined for the carbon acidity of acetone
ref 2). A statistical correction gives pKa ) 19.6 for a single methyl group
of acetone.
-
2
-1 -1
-4
kB ) 4.0 × 10
(R-CH3 group). The difference in these rate constants
is consistent with the known greater acidity of the benzylic
M
s
(R-CH2 group) and kB ) 3.4 × 10
(
-
1 -1
M
s
22) A value of kDO ) 0.32 M s-1 for deprotonation of acetone by
-
1
(
-1 -1
20
deuterioxide ion in D2O can be estimated from kHO ) 0.22 M
deprotonation by hydroxide ion in H2O (ref 23) and the secondary solvent
s
for
protons of benzyl methyl ketone (pKa ≈ 16) than of the protons
of a single R-methyl group of acetone (statistically corrected
isotope effect of kDO/kHO ) 1.46 (ref 24). A statistical correction for the
2
1
-1 -1
pKa ) 19.6).
presence of two methyl groups at acetone gives kDO ) 0.16 M
deprotonation of 1 by deuterioxide ion in D2O.
s
for
(
20) Keeffe, J. R.; Kresge, A. J.; Yin, Y. J. Am. Chem. Soc. 1988, 110,
(23) Chiang, Y.; Kresge, A. J.; Morimoto, H.; Williams, P. G. J. Am.
Chem. Soc. 1992, 114, 3981-3982.
8
201-8206.

Catalysis of Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4767
Table 1. Rate Constants for the Deprotonation of 1 by Brønsted Bases and the Reverse Protonation of 2 by Brønsted Acids, and Rate
Constant Ratios for Partitioning of the Enolate Carbon of 2 between Protonation by Brønsted Acids and Intramolecular Aldol Addition in
Water at 25 °C (I ) 1.0, KCl) (Scheme 5)
Brønsted base pKBHa [HO ] (M)b [B]/[BH+]
-
k
(M-1 s-1)c
k
BH/k
(M-1)c
k
BH (M-1 s-1)d
k
lim (s-1)e
k
lim/kHO[HO ] ) k
-
/kHOH
f
B
c
c
-
-
3
-2
7 g
-2
1
1
1.5
0.0
4.0 × 10
1.0
0.25
8.8 × 10
9.8 × 10
6.6
6.1
1.1 × 10
1.3 × 10
4.0 × 10
33
40
3
-2
-3
1
.0 × 10
-
-
-
4
4
5
-2
-2
-2
-3
-4
-4
6.3 × 10
4.0
1.0
0.25
2.1 × 10
2.0 × 10
2.1 × 10
40
34
34
2.1 × 10
5.9 × 10
1.5 × 10
33
37
38
7 g
1
4
.6 × 10
.0 × 10
6.5 × 10
-
-
4
6
-3
-4
8
-4
-5
9
7
.0
.5
1.3 × 10
9.0
5.7 × 10
120
450
2.2 × 10
4.3 × 10
33
33
8.0 × 10
20h
5.8 × 10
8.1 × 10
8
2.6 × 10
a
Apparent acidity of the corresponding 3-substituted quinuclidinium cation at 25 °C and I ) 1.0 (KCl). Data taken from ref 19. b The concentration
c
of hydroxide ion was calculated from the solution pH as described in the Supporting Information. Determined from the nonlinear least-squares fit
of the data from Figure 1 to eq 5 (see text). Calculated from the value of kBH/k
at high concentrations of buffer, calculated from the data in this table using eq 7 (see text). Calculated from [HO ] and kHO ) 0.10 M
the deprotonation of 1 by hydroxide ion (see text). Average of values at determined at different buffer ratios [B]/[BH ]. At pH 8.8 in the
d
-1
6
-1
e
c
(M ) and k
c
) 1.8 × 10 s (see text). Limiting rate constant
f
- -1 -1
s
for
g
+
h
presence of 10 mM pyrophosphate buffer.
+
Scheme 6
enolate to the carbonyl group (kc , kHOH + kBH[BH ]) results
in a downward break and values of kobsd that are independent
of buffer concentration (Scheme 5).
-
kHO[HO ] + k f [B]
B B
T
kobsd
)
(5)
1
+ (kBH/kc)fBH[B]T
-
1
Equation 5 gives the relationship between kobsd (s ) for the
overall aldol addition reaction of 1 to give 3 and the total
concentration of the buffer catalyst, [B]T (Figure 1). This
equation was derived for the mechanism shown in Scheme 5,
with the simplifying assumption that the barrier for reprotonation
of the enolate intermediate 2 by solvent is significantly larger
than the barrier to intramolecular addition of the enolate to the
carbonyl group (kHOH , kc, see following section). The known
parameters in eq 5 are kHO ) (kHO)aldol ) 0.10 M- s , the
observed second-order rate constant for hydroxide ion catalysis
of the aldol addition reaction of 1 with rate-determining
deprotonation (see Results section); [HO ], determined from
the observed pH of the solution (Table 1); and fB and fBH, the
1
-1
-
1
where klim (s ) is the limiting first-order rate constant for the
overall aldol addition reaction of 1 to give 3 in the presence of
high concentrations of a buffer catalyst at a specified pH. At
-
+
high concentrations of buffer, kBH[BH ] . kc, kc is fully rate-
determining, so that kobsd ) klim and ∆G
6). Equation 7 is the limiting form of eq 5 at high concentrations
fraction of the buffer present in the basic (B) and acidic forms
q
obsd
q
+
) ∆G lim (Scheme
(
BH ), respectively (Table 1). The nonlinear least-squares fits
-
1
of the data in Figure 1 to eq 5 gave the rate constants kB (M
+
-1
-1
of buffer where kobsd ) klim and fB/fBH ) [B]/[BH ].
s ) and the partitioning ratios kBH/kc (M ) given in Table 1.
The uncertainties in kB and kBH/kc, estimated from the range of
the values determined for catalysis by a single buffer at different
acid/base ratios are (10% and (20%, respectively (Table 1).
Enolate Addition to the Carbonyl Group. The reaction
coordinate profile in Scheme 6 shows that the difference in the
activation barriers for partitioning of 2 between protonation by
solvent water (kHOH) and intramolecular addition of the enolate
+
k
([B]/[BH ])
B
klim
)
(7)
(kBH/kc)
+
The values of kB, kBH/kc, and [B]/[BH ] in Table 1 were
substituted into eq 7 to give values of klim. These values of klim
were then combined with kHO ) 0.10 M
according to eq 6 to give the values of kc/kHOH in Table 1. The
values of kc/kHOH determined for several pH values and buffer
catalysts range from 33 to 40 with an average value of kc/kHOH
-
1
-1
-
s
and [HO ]
q
carbon to the carbonyl group (kc), ∆∆G , is given by the
difference in the activation barriers for the overall aldol addition
q
reaction of 1 when proton transfer (∆G HO) and enolate addition
q
-
are rate-determining (∆G lim), so that kc/kHOH ) klim/kHO[HO ]
eq 6),
)
35 ( 5.
(
The enolate of acetone was generated by laser flash photolysis
and the rate constant for its protonation by solvent water
klim
kHO[HO-] )
kc
4
-1 25
determined as kHOH ) 5.0 × 10 s . The rate constants for
(6)
kHOH
(24) Pocker, Y. Chem. Ind. 1959, 1383-1384.

4768 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Richard and Nagorski
nation of 2 (eq 9) meet two requirements for an internally
log k ) -7.3 + 0.55pK
(8)
(9)
B
BH
log kBH ) 12.5 - 0.47pKBH
consistent set of data: (1) The sum of the Brønsted exponents
obtained from eqs 8 and 9 is R + â ) 1.02, which is in
agreement with the required value of 1.00 for the logarithmic
change in the overall equilibrium constant with changing basicity
of the buffer catalyst, pKBH. (2) The y-intercepts given by eqs
8
and 9 correspond to the rate constants for deprotonation of 1
and the reverse protonation of 2 by a hypothetical buffer catalyst
of pKBH ) 0, and the difference of these intercepts, log(kBH/
kB)0 ) 19.8, is in agreement with the required value of 19.6,
which is the difference between pKBH ) 0 for this hypothetical
buffer catalyst and pKa ) 19.6 for the R-CH3 group of 1.21
The acidity of a hypothetical tertiary ammonium cation that
exhibits the same electrophilic reactivity toward 2 in an
Figure 2. Brønsted correlations of rate constants for reversible proton
transfer from 1 to 3-substituted quinuclidines (pKBH ) 7.5-11.5)19 in
H O at 25 °C (I ) 1.0, KCl). (A) Brønsted correlation for deprotonation
2
6
-1 -1
intermolecular reaction (kBH ) 1.8 × 10 M s ) as the
-
1
-1
of 1 by 3-substituted quinuclidines. Values of k
B
(M
s ) were taken
benzaldehyde carbonyl group in an intramolecular reaction (kc
-
1
-1
from Table 1. The open circle corresponds to kHO ) 0.10 M
s
for
6
-1
)
1.8 × 10 s ) can be calculated from eq 9 as pKBH ) 13.3.
deprotonation of 1 by hydroxide ion (see text). (B) Brønsted correlation
for protonation of 2 by 3-substituted quinuclidinium cations. Values
By contrast, the acidity of a hypothetical tertiary ammonium
cation that exhibits the same electrophilic reactivity toward 2
in a bimolecular reaction as the benzaldehyde carbonyl group
in a bimolecular reaction (kBH ) (kc)bi ) 2.4 × 10 M s )
is pKBH ) 17.3. The higher pKa estimated for a hypothetical
Brønsted acid with the same chemical reactivity as the benzal-
dehyde carbonyl group in inter- (pKBH ) 17.3) compared to
intramolecular (pKBH ) 13.3) reactions is a direct consequence
of the effective molarity of ≈75 M for intramolecular addition
of the enolate carbon of 2 to the benzaldehyde carbonyl group
-
1 -1
of kBH (M
s ) were taken from Table 1. The open circle corresponds
-
1 -1
to (kHOH/55 M) ) 900 M
s
for protonation of 2 by water (see text).
4
-1 -1 4
hydroxide ion catalyzed deprotonation of the R-CH3 group of
-
1
-1
1
(kHO ) 0.10 M s , see above) and a single R-CH3 group
-
1 -1 23
of acetone (kHO ) 0.11 M s ) are nearly identical, so that
it is reasonable to assume that the rate constant for the reverse
protonation of the acetone-like enolate 2 by solvent water is
the same as that for the enolate of acetone. Therefore, the rate
constant ratio kc/kHOH ) 35 for the partitioning 2 can be
(see above).
Deuterioxide Ion Catalyzed Exchange. The observation that
buffer catalysts result in a limiting ca. 35-fold increase in kobsd
for the overall aldol addition reaction of 1 (Table 1 and Figure
4
-1
6
combined with kHOH ) 5.0 × 10 s to give kc ) 1.8 × 10
-
1
s
for intramolecular addition of the enolate carbon of 2 to
the carbonyl group. This can be compared with the estimated
second-order rate constant for bimolecular addition of the
1
) requires that kc for the intramolecular addition reaction of 2
be 35-fold larger than kHOH for its protonation by solvent water
eq 6). The corresponding partitioning ratio in D2O should be
4
-1 -1 4
acetone enolate to benzaldehyde, (kc)bi ) 2.4 × 10 M s ,
to give an approximate effective molarity of 75 M for the
intramolecular enolate addition reaction of 2.26 Effective mo-
larities of 0.1-50 M have been reported for intramolecular
deprotonation of R-carbonyl carbon,26,27 and the range of
effective molarities must be the same for reactions in the reverse
direction of enolate protonation. This shows that there is, at
best, only a modest advantage to intramolecular addition of an
enolate to the carbonyl group over its intramolecular protonation
by Brønsted acids.
(
even larger due to a primary isotope effect on the protonation
of 2 by solvent, so that kc/kDOD > 35. This would require that
deuterium exchange of the protons of the R-CH3 group of 1
occur less than once every 35 times that 2 is generated by
deprotonation of 1 by deuterioxide ion, so that almost all the
deprotonation events result in ring closure to give the aldol
-1 -1
adduct 3, and kDO ) (kDO)aldol ) 0.15 M s . Therefore, the
rate constant for deuterioxide ion catalyzed exchange of the first
proton of the R-CH3 group of 1 through the enolate intermediate
-1
Brønsted Relationships. The rate constant ratios kBH/kc (M )
for partitioning of the enolate 2 between protonation by buffer
acids and intramolecular aldol addition were combined with kc
-3
-1 -1
2
is expected to be (kDO)ex e kDO/35 ) 4.3 × 10
M
s .
This is significantly smaller than the obserVed value of (kDO)ex
0.024 M s . We conclude that while the deuterioxide ion
catalyzed exchange of the protons of the R-CH3 group of 1
(kDO)ex ) 0.024 M s ] is significantly slower than depro-
tonation of 1 [(kDO)aldol ) 0.15 M s ], it is also unexpectedly
fast. The mechanism for this exchange reaction is discussed
further in the Supporting Information.
-1 -1
)
6
-1
)
1.8 × 10 s (see above) to give absolute values of kBH
-
1
-1
(
M
s ) (Table 1). Figure 2 shows Brønsted correlations of
-1 -1
[
rate constants for deprotonation of 1 by 3-substituted quinu-
clidines, kB (Figure 2A), and the reverse protonation of 2 by
the corresponding tertiary ammonium cations, kBH (Figure 2B),
with the pKa of the buffer catalyst, pKBH. The solid lines show
the fits of the data to eqs 8 and 9 which give â ) 0.55 (Figure
-1 -1
Electrophilicity of Brønsted Acids and the Carbonyl
Group. A comparison of kc for the intramolecular aldol addition
reaction of 2 with the rate constants for protonation of 2 (Figure
2B) shows that Brønsted acids may be either more or less
reactive than a benzaldehyde carbonyl group. The importance
of the intrinsic reaction barrier and thermodynamic driving force
in the determination of these relative rate constants is examined
in Table 2, which summarizes rate and equilibrium constants
for reactions of the acetone-like enolate 2 with Brønsted acids
2
A) and R ) 0.47 (Figure 2B).
The Brønsted correlations determined in this work for the
deprotonation of 1 (eq 8) and the microscopic reverse proto-
(
25) Chiang, Y.; Kresge, A. J.; Tang, Y. S.; Wirz, J. J. Am. Chem. Soc.
1
984, 106, 460-462.
(
(
26) Kirby, A. J. AdV. Phys. Org. Chem. 1980, 17, 183-278.
27) Richard, J. P. J. Am. Chem. Soc. 1984, 106, 4926-4936.

Catalysis of Aldol Addition Reactions
J. Am. Chem. Soc., Vol. 121, No. 20, 1999 4769
Table 2. Rate and Equilibrium Constants for Reactions of the Acetone-like Enolate 2 with Brønsted Acids and a Benzaldehyde-type Carbonyl
Group in Water at 25 °C (I ) 1.0, KCl)
Ka
k
k
Λ (kcal/mol) log k
b
c
reaction
pKBH
f
r
0
8
d
1.8 × 10 (s-1)e
6
5.5 × 10 (s-1)f
-3
3
.3 × 10
14.1
2.4
5.7 5.0 × 10 (M) 5.0 × 10 (s-1)h
5
g
4
0.10 (M-1 s-1)i
14.7
2.0
1
j
6 g
6
-1 -1
1.0 (M-1
-1)k
1
1
3.3 1.8 × 10
1.8 × 10 (M
s
)
s
12.8
12.9
3.4
3.3
l
8
7
-1
-1) 4.3 × 10 (M s-1)k
m
-2
-1
0.8 3.3 × 10
1.4 × 10 (M
s
a
. b Marcus intrinsic barrier (kcal/mol) for the reaction of 2, calculated
r
Equilibrium constant for the reaction of 2 as shown, given by K ) k
f
/k
c
from the rate and equilibrium constants using eq 10. Logarithm of the Marcus intrinsic rate constant for the reaction of 2, calculated from intrinsic
d
5
barrier Λ using eq 11. Calculated from K ) 1.2 × 10 for reaction of 2 to give the ketol 3 [ref 5], pK
a
) 19.6 for the R-CH
) k
/K. g Calculated as K ) k
) kBH ) 1.8
3
protons of 1 [ref
e
6
-1
f
h
2
1], and pK
a
) 15.9 for the ketol hydroxyl of 3 [ref 5].
k
c
) 1.8 × 10 s , see text. Calculated as k
l
r
f
f r
/k . Data
i
j
for enolate of acetone [ref 25]. kHO ) (kHO
10 M
protonation of 2 is equal to that for intramolecular addition of 2 to the carbonyl group, calculated using pK
1]. Calculated as k
)
aldol, data from this work. Acidity of hypothetical tertiary ammonium cation for which k
(see text). Calculated from pKBH using eq 9. Acidity of hypothetical tertiary ammonium cation for which K ) 3.3 × 10 for
c
6
-1 -1
k
8
×
s
a
) 19.6 for the R-CH protons of 1 [ref
3
m
2
f
r
) Kk .
Table 3. Intrinsic Rate Constants for Proton Transfer and
Nucleophilic Addition Reactions of Hard and Soft Acids and Bases
in Water
and the benzaldehyde carbonyl group. Marcus intrinsic barriers
Λ (kcal/mol, Table 2) for the hypothetical thermoneutral
reactions (∆G° ) 0) were estimated from the appropriate rate
4,28-30
and equilibrium constants using eq 10 (derived at 298 K).
The Marcus intrinsic rate constants for these thermoneutral
reactions, k0, were then estimated from these intrinsic barriers
using eq 11 (derived at 298 K).
1
.36
1.36 log K 2
log kobsd
)
17.44 - Λ 1 -
(10)
(11)
{
(
) }
1
4Λ
Λ
log k ) 12.8 -
0
1
.36
The data in Table 2 show that the enolate 2 has a greater
reactivity toward the benzaldehyde carbonyl group (kc ) 1.8 ×
6
-1
4
1
0 s , Table 2) than toward solvent water (kHOH ) 5 × 10
-
1
s ), but the similar values of ko for these reactions show that
this difference in reactivity is due mainly to the larger
thermodynamic driving force for carbonyl addition. As noted
by Guthrie in an earlier study of related intermolecular and
intramolecular aldol addition reactions,4 the intrinsic barriers
for reaction of an acetone-like enolate with a benzaldehyde
carbonyl group (Λ ) 14.1 kcal/mol, Table 2) and solvent water
a
Data from refs 36 and 37. b Reference 38. c Calculated from log
k
0
) 2.0 (Table 2). d Calculated from log k
0
) 2.4 (Table 2).
,5
2 by a tertiary ammonium cation whose pKBH of 10.8 was
chosen so that the two reactions would have the same
7
-1 -1
(
Λ ) 14.7 kcal/mol, Table 2) are very similar. The larger
thermodynamic driving force (kBH ) 1.4 × 10 M s ) shows
that there is a larger intrinsic barrier for the former reaction
(Table 2).
intrinsic barrier for protonation of 2 by solvent water (Λ ) 14.7
kcal/mol) than by a tertiary ammonium cation of pKBH )13.3
with the same reactivity toward 2 (Λ ) 12.8 kcal/mol) is
generally observed for proton transfer at R-carbonyl carbon, and
is manifested as negative deviations of rate constants for solvent-
catalyzed proton transfer at carbon from Brønsted correlations
Table 3 compares the Marcus intrinsic rate constants, k0, for
the reactions of “hard” (water and hydroxide ion) and “soft”
(carbonyl group and enolate ion) Lewis/Brønsted acids and
bases. These data follow the general trend that reactions between
soft acids and soft bases are intrinsically faster than reactions
(Figure 2, open symbols). Possible explanations for the deviation
of the point for lyoxide ion from rate-equilibrium correlations
34,35
between hard and soft pairs of acids and bases.
have been discussed elsewhere.2
7,30-33
1
. The intrinsic reactivity of the hard base hydroxide ion with
The smaller rate constant for the intramolecular aldol addition
9
-1 -1 36,37
the hard acid water, k0 ) 4 × 10 M s ,
is much larger
6
-1
reaction of 2 (kc ) 1.8 × 10 s ) compared to protonation of
-
1
than that of the “soft” enolate ion with water, k0 ) 100 s .
(
(
(
(
28) Marcus, R. A. J. Phys. Chem. 1968, 72, 891-899.
29) Marcus, R. A. J. Am. Chem. Soc. 1970, 91, 7224-7225.
30) Kresge, A. J. Chem. Soc. ReV. 1973, 2, 475-503.
(34) Hard and Soft Acids and Bases; Pearson, R. G., Ed.; Dowden,
Hutchinson and Ross: Stroudsberg, PA, 1973.
(35) Pearson, R. G. Chemical Hardness; John Wiley & Sons: New York,
1997; p 208.
(36) Meiboom, S. J. Chem. Phys. 1961, 34, 375-388.
(37) Loewenstein, A.; Sz o¨ ke, A. J. Am. Chem. Soc. 1962, 84, 1151-
1154.
31) Jencks, W. P.; Brant, S. R.; Gandler, J. R.; Fendrich, G.; Nakamura,
C. J. Am. Chem. Soc. 1982, 104, 7045-7051.
(
(
32) Pohl, E. R.; Hupe, D. J. J. Am. Chem. Soc. 1978, 100, 8130-8133.
33) Washabaugh, M. W.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111,
6
83-692.

4770 J. Am. Chem. Soc., Vol. 121, No. 20, 1999
Richard and Nagorski
for partitioning of an enolate ion between intramolecular addition
to the carbonyl group and intermolecular protonation by
Brønsted acids that have been determined in this work, ∆∆Gq
)
4.6 kcal/mol. Similar free energy profiles are expected for
termolecular aldol addition reactions, except that in this second
case partitioning of the enolate between intermolecular reactions
with Brønsted acids and the carbonyl group might show a small
increase in kc for carbonyl addition, relative to kBH, due to the
larger effective molarity of ca. 75 M for the former reaction
(see above).
The value of pKBH ) 6 was chosen for the catalytic base in
Figure 3 because enzymes usually maintain full catalytic activity
throughout the physiological pH range, which requires pKa <
7
for essential basic residues. The profile in Figure 3 demon-
strates two important roles of Class II aldolases in stabilization
of the transition state for aldol addition reactions.
1
. Interactions that stabilize the enolate relative to the keto
form of the substrate will result in a decrease in the ca. 18 kcal/
mol barrier to enolization and they will also stabilize the enolate-
ion-like transition state for the rate-determining addition of the
enolate to a carbonyl electrophile (kc, Figure 3). Such interac-
tions might include the formation of a strong, single-potential,
Figure 3. Free energy reaction coordinate profile for catalysis of the
intramolecular aldol addition reaction of 1 by 1.0 M of a Brønsted
base catalyst of pKBH ) 6. The profile was constructed as described in
the text.
40
hydrogen bond between the enzyme and the enolate ion, or
binding interactions which are expressed at the enzyme-enolate
4
1
complex, but not at the enzyme-ketone complex.
. Specific stabilization of the rate-determining transition state
2
2
. The requirement for rehybridization and electronic reor-
for addition of the enolate to the carbonyl electrophile from
electrophilic catalysis by an acidic amino acid side chain or a
metal cation. This could lower the overall barrier to the general-
base-catalyzed aldol addition, for which the enolate addition
step is rate-determining, by 4-5 kcal/mol, until the point where
substrate deprotonation again becomes rate-determining for the
overall aldol addition (Figure 3). For the reaction shown in
Figure 3, this would correspond to a net rate acceleration of ca.
ganization at the benzaldehyde carbonyl group upon addition
of nucleophiles results in a smaller intrinsic rate constant for
3
reaction of benzaldehyde with hydroxide ion (k0 ) 6 × 10
-
1 -1
M
s ) compared with proton transfer from water to hydroxide
9
-1 -1
ion (k0 ) 4 × 10
M
s ). This same requirement for
rehybridization and electronic reorganization at the benzaldehyde
carbonyl group might have been expected to result in a decrease
in the intrinsic rate constant for its reaction with 2, compared
with the intrinsic rate constant for the reaction of water with 2.
However, the intrinsic rate constant for protonation of the
acetone-like enolate 2 (a soft base) by the hard acid water, k0
2
000-fold.
Acknowledgment. This work was supported by grants GM
3
9754 and 47307 from the National Institutes of Health. We
-
1
-1
)
2
100 s , is slightly smaller than ko ) 250 s for addition of
to a benzaldehyde-type carbonyl group (a soft acid) (Table
thank Dr. Tadeusz Mizerski for the synthesis of 1 and Dr. Tina
L. Amyes for helpful discussion.
3
). We suggest that the expected decrease in the intrinsic rate
Supporting Information Available: Experimental details
of the following: (a) Procedures for the synthesis of 1. (b)
Procedures for the preparation of solutions used for kinetic
studies. (c) Procedures for monitoring the deuterium exchange
reactions of 1. A discussion of the mechanism for the deuteri-
oxide ion catalyzed exchange of the protons of the R-CH3 group
constant (increase in the intrinsic barrier Λ) for the latter reaction
is masked by a compensating stabilization of the transition state
from favorable interactions between the “soft” Lewis acid-
3
4,35
base pair.
Therefore, a simple consideration of the barriers
to geometric and electronic rearrangement may not be sufficient
39
to model the activation barrier for reactions, because these
barriers are also affected by developing covalent/ionic bonding
interactions between the reacting atoms in the transition state.
1
of 1. Figure S1: Representative H NMR spectra obtained
during exchange for deuterium of the protons of the R-CH2 and
-
1
R-CH3 groups of 1. Figure S2: Dependence of kobsd (s ) for
exchange for deuterium of the first proton of the R-CH3 group
of 1 on the concentration of the basic form of 3-quinuclidinone
Enzymatic Catalysis of Aldol Addition. Figure 3 shows a
free energy reaction coordinate profile for reaction of a general
base catalyst of pKBH ) 6 and a ketone with an aldehyde to
give the products of an intramolecular aldol addition reaction.
This profile was constructed using rate constants for a catalyst
of pKBH ) 6 calculated from eq 8 (log kB ) -4.0) and eq 9
-
1
buffer and deuterioxide ion in D2O; dependence of kobsd (s )
for the conversion of 1 to 4 on the concentration of hydroxide
ion in H2O and deuterioxide ion in D2O. This material is
available free of charge via the Internet at http://pubs.acs.org.
(
log kBH ) 9.7), log kc ) 6.3, and pKa ) 19.6 as the carbon
21
acidity of the ketone. This figure depicts the relative barriers
JA9900297
(
40) Gerlt, J. A.; Kozarich, J. W.; Kenyon, G. L.; Gassman, P. G. J.
(
38) Calculated by interpolation of the linear plot of rate (log kHO) and
Am. Chem. Soc. 1991, 113, 9667-9669. Gerlt, J. A.; Gassman, P. G. J.
Am. Chem. Soc. 1992, 114, 5928-5934. Gerlt, J. A.; Gassman, P. G. J.
Am. Chem. Soc. 1993, 115, 11552-11568. Gerlt, J. A.; Gassman, P. G.
Biochemistry 1993, 32, 2, 11943-11952.
(41) Jencks, W. P. In AdV. Enzymol. Relat. Areas Mol. Biol.; Meister,
A., Ed.; John Wiley and Sons: 1975; Vol. 43; pp 219-410.
equilibrium (log KHO) constants for addition of hydroxide ion to ring-
substituted benzaldehydes in water: McClelland, R. A.; Coe, M. J. Am.
Chem. Soc. 1983, 105, 2718-2725. The least-squares correlation line for
these data is log kHO ) 3.8 + 0.81 log KHO.
(
39) Guthrie, J. P. J. Am. Chem. Soc. 1997, 119, 1151-1152.