An antibody-catalyzed [2,3]-elimination reaction

Full text of DOI:10.1021/ja962257w

11686
J. Am. Chem. Soc. 1996, 118, 11686-11687
Scheme 1. Antibody-Catalyzed Cope Elimination Reaction
and Transition State Analogue
An Antibody-Catalyzed [2,3]-Elimination Reaction
Seung Soo Yoon,† Yoko Oei, Elizabeth Sweet, and
Peter G. Schultz*
Howard Hughes Medical Institute
Department of Chemistry, UniVersity of California
Berkeley, California 94720
ReceiVed July 3, 1996
Although there are no obvious limitations to the types of
chemical transformations that can be catalyzed by enzymes, only
a few examples are known of what are formally enzyme-
catalyzed, pericyclic reactions.1 Consequently, there has been
considerable interest in generating antibodies that catalyze this
class of reactions, both to increase the scope of biological
catalysis as well as to gain insight into the mechanisms by which
such catalytic functions might evolve.2 We now report an
example of an antibody-catalyzed [2,3]-sigmatropic reaction,
the Cope elimination3 of N-oxide 1 to dimethylhydroxyamine
and 4-methoxystyrene 2 (Scheme 1).
Balb/c mice. Twenty-three monoclonal antibodies specific for
hapten 3 were obtained by standard protocols,7 and purified by
chromatography on protein A-coupled Sepharose 4B.8 The
conversion of N-oxide 1 to product in aqueous 5 mM NaCl, 50
mM Na2PO4 buffer, pH 7.5, was followed by high-performance
liquid chromatography (HPLC).9 One of the twenty-three
antibodies, 21B12.1, was found to catalyze the [2,3]-elimination
reaction over the background reaction, with initial rates con-
sistent with Michaelis-Menten kinetics. The values of kcat and
Km were determined by fitting the kinetic data to the Michaelis-
Menten equation using a nonlinear regression program10 and
are 1.44 × 10-3 h-1 and 235 µM, respectively, at 37 °C. The
unimolecular reaction rate constant (kuncat) for the uncatalyzed
In order to catalyze this reaction, monoclonal antibodies were
generated against the keyhole limpet hemocyanin (KLH)
conjugate of substituted tetrahydrofuran 3, which resembles the
conformationally restricted transition state of the reaction.4
Moreover, the reduced dipole monent of hapten 3 relative to
the substrate is likely to induce a low-dielectric environment in
the corresponding antibody combining site. Such an environ-
ment might be expected to accommodate the charge dispersion
on going from substrate to transition state better than water.
This effect has previously been exploited in an antibody-
catalyzed decarboxylation reaction.5 Consequently, an antibody
generated to hapten 3 might be expected to catalyze the Cope
elimination of substrate 1 through a combination of entropy and
medium effects. The dissociative nature of the reaction should
minimize product inhibition.
reaction under the same conditions is 1.58 × 10-6 h-1
,
corresponding to a rate enhancement of roughly 103 over the
uncatalyzed reaction. Antibody 21B12.1 did not measurably
catalyze the Cope elimination reaction of the demethylated
substrate analogue, 2-(4′-hydroxyphenyl)ethyldimethylamine
oxide, or the elimination reaction of the corresponding sulfoxide,
2-(4′-methoxyphenyl)ethyl methyl sulfoxide, indicating that the
antibody-catalyzed reaction is highly selective. Hapten 3
inhibits the antibody-catalyzed reaction competitively with a
Ki of 200 nM at 37 °C.11 The ratio of Ki to Km correlates
roughly with the observed rate acceleration, consistent with the
notion that the rate enhancement is due to the preferential
binding of the transition state by antibody. The deuterated
substrate, 1-2,2-d2, was prepared12 and kinetic isotope effects
were measured. The value of kcatH/kcatD is 2.78 for the antibody-
The N-hydroxysuccinyl ester of hapten 36 was coupled to
KLH and the resulting protein conjugate was used to immunize
* Author to whom correspondence should be addressed.
† Current address: Department of Chemistry, Sung Kyun Kwan Uni-
versity, Natural Science Campus, Suwon 440-746, Korea.
(1) (a) Andrews, G. D.; Smith, G. D.; Young, I. G. Biochemistry 1973,
12, 3492. (b) Gorisch, H. Biochemistry 1978, 17, 3700. (c) Guilford, W.
J.; Copley, S. D.; Knowles, J. R. J. Am. Chem. Soc. 1987, 109, 5013. (d)
Laschat, S. Angew. Chem., Int. Ed. Engl. 1996, 35, 289. (e) Oikawa, H.;
Katayama, K.; Suzuki, Y.; Ichihara, A. J. Chem. Soc., Chem. Commun.
1995, 1321.
(2) (a) Hilvert, D.; Hill, K. W.; Nared, K. D.; Auditor, M. J. Am. Chem.
Soc. 1989, 111, 9261. (b) Braisted, A. C.; Schultz, P. G. J. Am. Chem.
Soc. 1990, 112, 7430. (c) Hilvert, D.; Carpenter, S. H.; Nared, K. D.;
Auditor, M. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 4953. (d) Jackson, D.
Y.; Jacobs, J. W.; Sugasawara, R.; Reich, S. H.; Bartlett, P. A.; Schultz, P.
G. J. Am. Chem. Soc. 1988, 110, 4841. (e) Braisted, A. C.; Schultz, P. G.
J. Am. Chem. Soc. 1994, 116, 2211. (f) Gouverneur, V. E.; Houk, K. N.;
Pascul-Teresa, B. D.; Beno, B.; Janda, K. D.; Lerner, R. A. Science 1993,
262, 204. (g) Yli-Kauhaluoma, J. T.; Ashley, J. A.; Lo, C.-H.; Tucker, L.;
Wolfe, M. M.; Janda, K. D. J. Am. Chem. Soc. 1995, 117, 7041. (h)
Suckling, C. J.; Tedford, M. C.; Bence, L. M.; Irvine, J. I.; Stimson, W. H.
J. Chem. Soc., Perkin Trans. 1 1993, 1925.
(7) Jacobs, J. W. Ph.D. Thesis, University of California, Berkeley, CA
94720.
(8) (a) Kronvall, G.; Grey, H.; Williams, R. J. Immunol. 1972, 105, 1116.
(b) Harlow, E.; Lane, D. Antibodies. A Laboratory Manual; Cold Spring
Harbor Laboratory: New York, 1988.
(9) HPLC assays were carried out with a Microsorb C18 reverse-phase
column with a gradient starting a 10% acetonitrile in water and increasing
to 100% acetonitrile over 25 min. Product formation was monitored at
270 nm and quantitated against the internal standard, 4-(N-ethylamido)-
benzoic acid methyl ester. The retention time of the product formed in the
catalyzed and uncatalyzed reaction is identical with that of commercially
available 4-methoxystyrene.
(10) Initial rates were measured at five different substrate concentrations
(100, 175, 250, 500, and 1000 µM) and the initial rate data were fitted to
the Michaelis-Menten equation V ) Vmax[1]/(Km + [1]) using the Leven-
verg-Marquart algorithm of the Kaleida Graph computer program (Abel-
beck software). Antibody concentrations were 4.3 µM in binding sites.
(11) Inhibition data were fitted to following equation:
(3) Cope, A. C.; Foster, T. T.; Towel, P. H. J. Am. Chem. Soc. 1949,
71, 3939.
(4) Kwart, H. Acc. Chem. Res. 1982, 15, 401.
(5) (a) Lewis, C.; Kramer, T.; Robinson, S.; Hilvert, D. Nature 1991,
253, 1019. (b) Lewis, C.; Paneth, P.; O’Leahy, M. H.; Hilvert, D. J. Am.
Chem. Soc. 1993, 115, 1410.
(6) Hapten 3 was synthesized from the commercially available starting
material, 2(5H)-furanone. A palladium-catalyzed coupling reaction of
starting material with 4-iodoanisole followed by hydrogenation, Grignard
reaction with methyl magnesium bromide, and acid-catalyzed cyclization
in the presence of p-toluenesulfonic acid afforded 2-dimethyl-4-(4′-
methoxyphenyl)tetrahydrofuran. Removal of the methyl protecting group
of phenol with sodium hydride and ethanethiol, followed by Mitsunobu
alkylation with ethyl 6-hydroxyhexanoate and subsequent hydrolysis of the
ethyl ester, provided hapten 3. Substrate 1 was synthesized by an ethyl-
(3-dimethylaminopropyl)dicarbodiimide coupling reaction of 4-methoxy-
phenylacetic acid with dimethylamine, followed by reduction with borane-
THF complex and oxidation with m-chloroperbenzoic acid.
Vi/Vo ) ([E]t - [I] - Ki′ + {([E]t - [I] + Ki′)2 + 4Ki′[I]}0.5)/[E]t
where [E]t ) total concentration of antibody binding sites, [I] ) concentra-
tion of inhibitor, Ki′ ) Ki(1 + [S]/Km), [S] ) substrate concentration, Vi )
initial velocity measured in the presence of inhibitor, and v0 ) initial velocity
in the absence of inhibitor: Williams, J. W.; Morrison, J. F. Methods
Enzymol. 1979, 63, 437.
(12) Reduction of p-anisoyl chloride with sodium borodeuteride followed
by mesylation, replacement with cyanide, reduction with LiAlH4, reductive
dimethylation, and oxidation with m-chloroperbenzoic acid provided the
deuterated substrate.
S0002-7863(96)02257-3 CCC: $12.00 © 1996 American Chemical Society

Communications to the Editor
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11687
conformation for the elimination reaction. However, the
antibody functions primarily by lowering the enthalpy of
activation for reaction, consistent with solvation effects on the
reaction.
To provide additional evidence for this idea, we measured
the elimination reaction of substrate 1 in dimethylformamide
(DMF) and 1,4-dioxane. The values of kuncat are 2.76 × 10-4
h-1 in 1,4-dioxane (ET ) 36)16 and 5.79 × 10-3 h-1 in DMF
(ET ) 43)16 at 37 °C. The activation parameters for the
uncatalyzed reaction in 1,4-dioxane are ∆Hq ) 26.6 kcal/mol
and ∆Sq ) -5.71 eu. The values for the uncatalyzed reaction
in DMF are ∆Hq ) 25.3 kcal/mol and ∆Sq ) -3.55 eu.16 The
change of reaction medium from water (ET ) 63.1)16 to organic
solvent was found to significantly accelerate the N-oxide
elimination reaction, largely by lowering the enthalpy of
activation. These results are consistent with the notion that the
medium effects play a significant role in this antibody-catalyzed
reaction. The lack of a quantitative correlation of kuncat with
the solvent parameter ET16 likely reflects the differing abilities
of DMF and dioxane to stabilize both substrate 1 and a transition
state in which partial proton transfer is occurring. Given the
substrate specificity of the antibody and the small fraction of
active antibodies relative to those that bind hapten 3, other
factors are clearly influencing catalysis. Additional mechanistic
and structural studies on antibody 21B12.1, the germline
precursor, and noncatalytic antibodies specific for 3 may clarify
this issue.
Figure 1. Arrhenius plot for the antibody 21B12.1 catalyzed reaction
(b) and the uncatalyzed reactions in 1,4-dioxane ()), DMF (4), and
aqueous buffer (O).
catalyzed reaction, suggesting that the rate-determining transition
state for the reaction involves some degree of C-H bond
breaking. For the uncatalyzed reaction in aqueous buffer, the
value of kuncatH/kuncatD is 2.34.14
To gain additional insight into the mechanism of catalysis
by antibody, 21B12.1, the kinetics of the catalyzed (kcat) and
uncatalyzed (kuncat) elimination reactions were examined as a
function of temperature (Figure 1). The activation parameters
for the uncatalyzed reaction are ∆Hq ) 30.6 kcal/mol and ∆Sq
) -2.84 eu and the values for the antibody-catalyzed reaction
are ∆Hq ) 27.2 kcal/mol and ∆Sq ) -0.29 eu, in aqueous 5
mM NaCl, 50 mM Na2PO4 buffer, pH 7.5.15 The difference in
∆Gq between the antibody-catalyzed reaction and the uncata-
lyzed reaction is 4.2 kcal/mol at 37 °C, consistent with the
roughly 1000-fold rate acceleration. The fact that the entropy
of activation for the antibody-catalyzed reaction is about zero
suggests that the antibody constrains the substrate in a favorable
In summary, we have successfully elicited an antibody that
significantly accelerates the rate of [2,3]-sigmatropic elimination
reaction. Mechanistic studies underscore the importance of
medium effects in this antibody-catalyzed reaction and suggest
that such effects,17 when combined with entropic effects as well
as general acid-base and nucleophilic catalysis, could lead to
large additive reductions in the ∆Gq of catalytic antibodies.
Acknowledgment. We are grateful for financial support for this
work from the National Institutes of Health (Grant No. R01AI24695).
P.G.S. is a Howard Hughes Medical Institute Investigator. S.S.Y. is
supported by a fellowship from the Helen Hay Whitney Foundation.
(13) Equimolar concentrations of 1 and 1-2,2-d2 were combined and
added to the antibody solution. Reaction products (<1% conversion) were
purified by HPLC and analyzed by mass spectrometry. The signal resulting
from 2 and the deuterated product were separately integrated and the ratio
of the integral (PH/PD) was taken as the mole ratio of product. The value
JA962257W
of kcatH kcatD is uncorrected for secondary isotope effects.
/
(14) For the uncatalyzed reaction, the values of kH/kD are 2.52 and 2.54
in dioxane and DMF, respectively.
(16) Dimroth, K.; Reichardt, C.; Siepmann, T.; Bohlmann, F. Ann. Chem.
1963, 661, 1.
(15) The reaction rates (kcat and kuncat) were measured at four different
temperatures (47, 37, 27, and 17 °C) for the antibody-catalyzed reaction in
water and the uncatalyzed reaction in 1,4-dioxane and DMF, and (67, 57,
47, and 37 °C) for the uncatalyzed reaction in aqueous buffer. The activation
parameters were determined from an Arrhenius plot.18
(17) (a) Chiao, W.-B.; Saunders, W. H., Jr. J. Am. Chem. Soc. 1978,
100, 2802. (b) Cram, D. J.; Sahyun, M. R.; Knox, G. R. J. Am. Chem.
Soc. 1962, 84, 1734.
(18) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper & Row: New York, 1987; p 209.