Transition state modeling and catalyst design for hydrogen bond-stabilized enolate formation

Article abstract of DOI:10.1021/jo0513818

A catalyst for enolate formation was designed that incorporates an amine base along with a thiourea to bind to the oxygen atom of the substrate and enolate through hydrogen bonding. A computational model of the transition state was developed in which the thiourea (modeled initially as a urea) and amine were separate molecules. This model and models incorporating one or two methanol molecules in place of the urea showed an out-of-plane hydrogen bond, apparently to the carbonyl π-bond, in addition to an in-plane hydrogen bond to an unshared electron pair. In contrast, optimized complexes of the ketone and the fully formed enolate showed only in-plane hydrogen bonding. The transition state model with the urea and amine was used to define a database search with the computer program CAVEAT to identify structures suitable for linking the amine and urea/thiourea moieties in the transition state. On the basis of a group of structures identified from this search, a flexible but conformationally biased linker was designed to connect the two catalytic moieties. The molecule having the amine and thiourea moieties connected by this linker was synthesized and was shown to catalyze proton exchange between methanol and deuterated acetone. The catalyst was about 5-fold more efficient than the amine and thiourea as separate molecules and relative to a similar but less conformationally biased catalyst.

Full text of DOI:10.1021/jo0513818

Transition State Modeling and Catalyst Design for Hydrogen
Bond-Stabilized Enolate Formation
Yimin Zhu and Dale G. Drueckhammer*
Department of Chemistry, Stony Brook University, Stony Brook, New York 11794
dale.drueckhammer@sunysb.edu
Received July 4, 2005
A catalyst for enolate formation was designed that incorporates an amine base along with a thiourea
to bind to the oxygen atom of the substrate and enolate through hydrogen bonding. A computational
model of the transition state was developed in which the thiourea (modeled initially as a urea) and
amine were separate molecules. This model and models incorporating one or two methanol molecules
in place of the urea showed an out-of-plane hydrogen bond, apparently to the carbonyl π-bond, in
addition to an in-plane hydrogen bond to an unshared electron pair. In contrast, optimized complexes
of the ketone and the fully formed enolate showed only in-plane hydrogen bonding. The transition
state model with the urea and amine was used to define a database search with the computer
program CAVEAT to identify structures suitable for linking the amine and urea/thiourea moieties
in the transition state. On the basis of a group of structures identified from this search, a flexible
but conformationally biased linker was designed to connect the two catalytic moieties. The molecule
having the amine and thiourea moieties connected by this linker was synthesized and was shown
to catalyze proton exchange between methanol and deuterated acetone. The catalyst was about
5-fold more efficient than the amine and thiourea as separate molecules and relative to a similar
but less conformationally biased catalyst.
Introduction
synthase catalyze similar reactions via enolate intermed-
iates.10-12 Simple catalysts for enolate formation could
potentially be applied to esters and other carboxylic acid
derivatives, for which enamine formation is not viable,
and could have practical use in carbon-carbon bond
forming and racemization reactions.13,14 Several catalysts
for formation of enolates or enols have been described,
including cyclodextrins having appended basic groups,15-19
The design of simple catalysts that in a general way
mimic the action of enzymes is a longstanding and
continuing goal of organic chemistry.1-4 A recent very
successful example is the use of proline and analogues
as simple catalysts for aldol condensation and related
reactions.5-8 Catalysis is achieved via an enamine mech-
anism related to the mechanism of the class I aldolases.9
Other enzymes such as the class II aldolases and citrate
(10) Hall, D. R.; Leonard, G. A.; Reed, C. D.; Watt, C. I.; Berry, A.;
Hunter, W. N. J. Mol. Biol. 1999, 287, 383-394.
(11) Kurz, L. C.; Shah, S.; Crane, B. R.; Donald, L. J.; Duckworth,
H. W.; Drysdale, G. R. Biochemistry 1992, 31, 7899-7907.
(12) Kluger, R. Chem. Rev. 1990, 90, 1151-1169.
(13) Um, P.-J.; Drueckhammer, D. G. J. Am. Chem. Soc. 1998, 120,
5605-5610.
(1) Gro¨ger, H.; Wilken, J. Angew. Chem., Int. Ed. 2001, 40, 529-
532.
(2) Horstmann, T. E.; Guerin, D. J.; Miller, S. J. Angew. Chem., Int.
Ed. 2000, 39, 3635-3638.
(3) French, R. R.; Holzer, P.; Leuenberger, M. G.; Woggon, W.-D.
Angew. Chem., Int. Ed. 2000, 39, 1267-1269.
(4) Yan, J.; Breslow, R. Tetrahedron Lett. 2000, 41, 2059-2062.
(5) List, B.; Lerner, R. A.; Barbas, C. F., III. J. Am. Chem. Soc. 2000,
122, 2395-2396.
(14) Dinh, P. M.; Williams, J. M. J.; Harris, W. Tetrahedron Lett.
1999, 40, 749-752.
(15) Siegel, B.; Pinter, A.; Breslow, R. J. Am. Chem. Soc. 1977, 99,
2309-2312.
(6) List, B. Acc. Chem. Res. 2004, 37, 548-557.
(7) Saito, S.; Yamamoto, H. Acc. Chem. Res. 2004, 37, 570-579.
(8) Notz, W.; Tanaka, F.; Barbas, C. F., III. Acc. Chem. Res. 2004,
37, 580-591.
(16) Breslow, R.; Graf, A. J. Am. Chem. Soc. 1993, 115, 10988-
10989.
(17) Desper, J. M.; Breslow, R. J. Am. Chem. Soc. 1994, 116, 12081-
12082.
(9) Morris, A. J.; Tolan, D. R. Biochemistry 1994, 33, 12291-12297.
10.1021/jo0513818 CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/23/2005
J. Org. Chem. 2005, 70, 7755-7760
7755

Zhu and Drueckhammer
SCHEME 1
SCHEME 2
catalysis, linked to a basic group (B) positioned to
deprotonate the R-carbon. The linker should hold the
electrophilic and basic groups in proper relative position
for optimal transition state stabilization in conversion
of the substrate complex 3 to the enolate complex 4.
Described here are the computational modeling of the
transition state for this reaction, the design of an
appropriate linker, and synthesis and testing of the
designed catalyst.
aza-crown ethers,20 and a modified Kemp’s triacid.21 Such
catalysts are typically designed to have distinct func-
tional groups for deprotonation and for stabilization or
protonation of enolate oxygen. A recognized challenge in
the design of such catalysts is making the backbone
structure sufficiently rigid or constrained to prevent
collapse into a conformation in which the catalytic groups
simply bind to each other.21,22 Several receptors have been
reported that selectively bind and stabilize enolates.22-27
These receptors usually bind the enolate via hydrogen-
bonding interactions to the enolate oxygen, though acidity
enhancement by intramolecular coordination to Zn(II) ion
has also been demonstrated.25
The goal of this project was to use computer-guided
design methods based on the computer program CAVEAT
to design a simple enolate-forming catalyst. CAVEAT is
a unique program developed by Bartlett and co-workers
that searches a virtual molecular database to identify
structures having bonds that match a set of defined
vectors.28,29 Each vector defines a starting point and a
direction, with compounds identified using CAVEAT
having sets of atoms and associated bonds corresponding
to each vector. This program has been used in the design
of enzyme inhibitors and conformationally constrained
peptides28,29 and more recently in the design of chiral
ligands for asymmetric catalysis30 and a boronic acid-
based receptor and sensor for glucose.31
Results
For this initial study, an amine was chosen as the basic
group and a urea or thiourea as the electrophilic group.
The development of a computer model for the transition
state began with location of the transition state 5 for
proton transfer from the protonated form of acetone to
trimethylamine (Scheme 2). The transition state for this
reaction was located and optimized at the HF/6-31+G-
(d) level using the QST2 method in Gaussian03.32 The
resulting structure was modified by deletion of the proton
on the oxygen and introduction of N,N′-dimethylurea in
position to form two hydrogen bonds to this oxygen. One
of the amine methyl groups was replaced with a hydrogen
atom to simplify the structure.
The general approach of this project is illustrated in
Scheme 1. The catalyst 1 is envisioned having an elec-
trophilic group (E) that will coordinate to the carbonyl
oxygen of the substrate 2 and stabilize the enolate during
(18) Breslow, R.; Desper, J. M.; Huang, Y. Tetrahedron Lett. 1996,
37, 2541-2544.
(19) Binder, W. H.; Menger, F. M. Tetrahedron Lett. 1996, 37, 8963-
8966.
FIGURE 1. Complexes used in transition state models.
(20) Fenniri, H.; Lehn, J.-M.; Marquis-Rigault, A. J. Am. Chem. Soc.
1996, 35, 337-339.
The atoms of the amine, the enolate, and the transfer-
ring proton were frozen, and the position and structure
of the urea moiety was optimized at the B3LYP/6-31+G-
(d) level to give the approximate transition state struc-
ture 6 shown in Figures 1 and 2. In complexes with
acetone or its enolate, the urea aligned in approximately
the same plane as the non-hydrogen atoms of the ketone
(21) Wolfe, J.; Muehldorf, A.; Rebek, J. J. Am. Chem. Soc. 1991,
113, 1453-1454.
(22) Snowden, T. S.; Bisson, A. P.; Anslyn, E. V. Bioorg. Med. Chem.
2001, 9, 2467-2478.
(23) Chin, T.; Gao, Z.; Lelouche, I.; Shin, Y. K.; Purandare, A.;
Knapp, S.; Isied, S. S. J. Am. Chem. Soc. 1997, 119, 12849-12858.
(24) Kimura, E.; Kitamura, H.; Koike, T.; Shiro, M. J. Am. Chem.
Soc. 1997, 119, 10909-10919.
(25) Kimura, E.; Gotoh, T.; Koike, T.; Shiro, M. J. Am. Chem. Soc.
1999, 121, 1267-1274.
(26) Kelly-Rowley, A. M.; Lynch, V. M.; Anslyn, E. V. J. Am. Chem.
Soc. 1995, 117, 3438-3447.
(32) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson,
G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai,
H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken,
V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev,
O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P.
Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.;
Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas,
O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.;
Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov,
B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.;
Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.;
Gonzalez, C.; Pople, J. A. Gaussian 03, revision A.1; Gaussian, Inc.:
Wallingford, CT, 2004.
(27) Snowden, T. S.; Bisson, A. P.; Anslyn, E. V. J. Am. Chem. Soc.
1999, 121, 6324-6325.
(28) Lauri, G.; Bartlett, P. A. J. Comput.-Aided Mol. Des. 1994, 8,
51-66.
(29) (a) Bartlett, P. A.; Shea, G. T.; Telfer, S. J.; Waterman, S. In
Molecular Recognition: Chemical and Biological Problems; Roberts,
S. M., Ed.; Royal Society of Chemistry: London, 1989; pp 182-196.
(b) Bartlett, P. A.; Etzkorn, F. A.; Guo, T.; Lauri, G.; Liu, K.; Lipton,
M.; Morgan, B. P.; Shea, G. T. In Chemistry at the Frontiers of
Medicine, Proceedings of the Robert A. Welch Foundation Conference
on Chemical Research XXXV, Houston, TX, Oct 28-29, 1991; Robert
A. Welch Foundation: Houston, TX, 1991; pp 45-68.
(30) Kozlowski, M. C.; Panda, M. J. Mol. Graphics Modell. 2002,
20, 399-409.
(31) Yang, W.; He, H.; Drueckhammer, D. G. Angew. Chem., Int.
Ed. 2002, 40, 1714-1718.
7756 J. Org. Chem., Vol. 70, No. 19, 2005

Modeling and Catalyst Design for Enolate Formation
SCHEME 3
SCHEME 4
FIGURE 2. Transition state model 6 for the deprotonation
of the acetone-dimethylurea complex by dimethylamine.
anti conformation. The structure was further simplified
by deleting the ring and introducing a tert-butyl group
at the carbon connecting the two gauche carbon-carbon
bonds. The tert-butyl group is expected to induce the
gauche conformation of the main chain at both C-C
bonds extending from the tert-butyl substituted carbon,
as both methylene groups will prefer to be anti to the
tert-butyl group. Also, the methyl urea moiety was
replaced with the phenylthiourea to form 10.
A conformational search of 10 by AM1 identified a
structure in which the amine group is positioned to form
a somewhat distorted hydrogen bond with the more
internal N-H of the thiourea. In calculations not con-
sidering solvation (B3LYP/6-311+G(d,p)//B3LYP/6-31G-
(d)), this internally hydrogen-bonded structure was more
stable than the proposed active structure of the catalyst.
However, when a solvation model (PCM) was included
in the calculations, initially with chloroform as the
solvent, the proposed active conformation was calculated
to be more stable than the somewhat strained hydrogen-
bonded structure by about 0.3 kcal/mol.
FIGURE 3. Transition state model 7 for the deprotonation
of the acetone-bis-methanol complex by methylamine.
or enolate. However, in this and other transition state-
like structures, the urea is always significantly tilted out
of the plane of the developing enolate toward the side on
which deprotonation is occurring. To further investigate
this issue, structures with one or two methanol molecules
hydrogen-bonded to the frozen transition state were also
calculated at the B3LYP/6-31+G(d) level. The optimized
structure with a single methanol molecule had the
hydrogen bond from methanol directed perpendicular to
the plane of the developing enolate on the face nearest
the base. When a second methanol molecule was intro-
duced, the first methanol molecule remained in the same
position while the second formed a hydrogen bond in the
plane of the developing enolate (7, Figures 1 and 3).
The later steps in the catalyst design are illustrated
in Scheme 3. The C-N bonds of the amine and urea
shown in bold of the transition state complex 6 were used
to define a vector pair for a CAVEAT search. Structure
8 was found from a recently developed virtual library of
trisubstituted cyclic hydrocarbon structures, with the
C-H bonds of 8 shown in bold matching the vector pair
defined by 6. Replacement of these hydrogens by the urea
and amine groups leads to the potential catalyst structure
9. While this structure has a good deal of flexibility, it
was observed that, in the conformation that matches the
vectors defined by 6, the two C-C bonds extending from
the ring to the urea are both in gauche conformations
relative to the other carbons of the chain connecting the
urea and amine groups. In contrast, the two C-C bonds
extending to the amine group are both in the preferred
The synthesis of the designed catalyst was completed
as shown in Scheme 4. The nitroolefin 11 was prepared
by a base-catalyzed Henry reaction of trimethylacetal-
dehyde and nitromethane followed by dehydration of the
resulting nitro alcohol as previously described.33 Addition
of the Grignard reagent derived from commercially
available 3-(N,N-dimethylamino)-1-chloropropane and
reduction of the nitro group to the amine was followed
by phenylthiourea formation to make the proposed
catalyst 10. For comparison purposes, the equivalent
structure 13 lacking the tert-butyl group was prepared
by reduction of the nitrile 1234 and thiourea formation
as shown in Scheme 5.
1
The H NMR spectrum of 10 was compared to that of
N-butyl-N′-phenylthiourea to evaluate possible intramo-
lecular hydrogen-bonding interactions.35 The proton on
(33) Lucet, D.; Sabelle, S.; Kostelitz, O.; Le Gall, T.; Mioskowski, C.
Eur. J. Org. Chem. 1999, 2583-2591.
(34) Park, S.; Hayes, B. L.; Marankan, F.; Mulhearn, D. C.; Wanna,
L.; Mesecar, A. D.; Santarsiero, B. D.; Johnson, M. E.; Venton, D. L.
J. Med. Chem. 2003, 46, 936-953.
J. Org. Chem, Vol. 70, No. 19, 2005 7757

Zhu and Drueckhammer
SCHEME 5
Discussion
This work was directed at the development of a catalyst
for enolate formation designed to function in non-
hydrogen-bonding organic solvent, such that binding of
the catalyst to the carbonyl substrate could be driven by
hydrogen-bonding interactions. Urea, thiourea, and re-
lated guanidinium functionality have been used exten-
sively in catalysts for nucleophilic addition to carbonyl
groups, conjugate additions, and related reactions and
thus were chosen as simple and easily prepared enolate-
stabilizing groups for this work.36-40 A urea was used in
initial modeling, but in the actual catalyst was replaced
by the phenylthiourea, which is easily synthesized, is
expected to be more acidic and thus a better hydrogen
bond donor than the urea, and should be less prone to
self-association due to the poorer hydrogen-bonding abil-
ity of the sulfur atom. In initial modeling, it was expected
that the urea would hydrogen bond to the lone pairs on
oxygen in the substrate complex and in the transition
state and, thus, that the urea would lie in the same plane
as the carbon and oxygen atoms of the ketone and
developing enolate. While this was indeed observed in
the acetone and enolate complexes, the approximate
transition state structure gave a very nonplanar struc-
ture. Anslyn and co-workers have proposed and demon-
strated evidence for enolate stabilization by hydrogen
bond donors directed toward the π-bond rather than the
lone electron pairs.22,27 They also pointed out that hy-
drogen bonds in the enzyme p-chlorobenzoyl-CoA deha-
logenase are directed perpendicular to the plane of the
enolate, consistent with enzymatic stabilization of the
intermediate Meisenheimer complex by hydrogen bonds
directed toward the π-bond.41 Similar out-of-plane hy-
drogen bonding is observed in crotonase42 and likely in
the other members of the crotonase superfamily,22,43 and
in the acyl-CoA dehydrogenases.44,45 In contrast, the
hydrogen bonds that stabilize the enolate intermediate
in citrate synthase are in the plane of the enolate directed
toward the lone electron pairs on oxygen.46 Our calcula-
tions on the urea transition state complex are consistent
with the enzymes that exhibit out-of-plane hydrogen
bonding. In calculations in which the urea was replaced
by a pair of methanol molecules, it was found that one
methanol formed a hydrogen bond directed perpendicular
TABLE 1. Rate Constants for the Catalyzed Proton
Exchange Reactions
a
catalyst
rate constant (min-1
)
BuNMe2
0.009
0.009
0.046
0.011
BuNMe2 + BuNHCSNHPh
10
13
a Conditions: 0.1 mmol catalyst, 0.05 mL of methanol, in 0.5
mL of acetone-d6 at room temperature.
the internal alkyl nitrogen of 10 was shifted to about 6.6
ppm from 6.0 ppm for the equivalent proton of the model
thiourea. The proton on the phenyl-substituted nitrogen
at 8.2 ppm was not shifted significantly. Compound 13
exhibited a smaller shift of the internal alkyl nitrogen
and a larger downfield shift of the phenyl-substituted
nitrogen relative to 10. Addition of 1 equiv of butyl
dimethylamine to the model thiourea (both 0.2 M)
resulted in no change in chemical shift of either of the
thiourea protons. Attempts were made to evaluate bind-
ing of acetone to 10 from the change in chemical shift of
the thiourea protons. A downfield shift of these protons
was observed in acetone-d6 compared to CDCl3, the
N-phenyl proton moving from 8.2 to 8.9 ppm and the
N-alkyl proton moving from 6.6 ppm to the region
overlapping with the phenyl protons at 7.2-7.4 ppm. As
the percentage of acetone in an acetone/CDCl3 mixture
was increased, a gradual shift of the thiourea protons
was observed but a point of saturation was not observed.
The catalysis of enolate formation by 10 was deter-
mined by deuterium exchange experiments in deuterated
acetone using methanol as a source of exchangeable
proton. The reaction was monitored by 1H NMR following
the disappearance of the methanol OH peak and appear-
ance of the acetone peak over time. For comparison
catalysis by N,N-dimethylbutylamine, a model for the
amine component of catalyst 10 was studied. Also studied
was catalysis by the same model amine in the presence
of N-butyl-N′-phenylthiourea and by the catalyst struc-
ture 13 lacking the tert-butyl group. The results are
summarized in Table 1. While no detectable exchange
was observed in the absence of any catalyst, significant
reaction was observed in the presence of butyldimethyl-
amine. No detectable enhancement of this rate was
observed when a molar equivalent of N-butyl-N′-phen-
ylthiourea relative to the amine was included. The rate
of exchange catalyzed by 10 was 5-fold greater than that
of the simple amine. In contrast, the catalyst 13 lacking
the tert-butyl group exhibited only slight rate enhance-
ment relative to the simple amine. An experiment was
also done to measure the exchange reaction between
deuterated methanol and nondeuterated acetone. The
rate was faster than that with deuterated acetone by a
factor of 1.4.
(35) Novak, L.; Hanania, M.; Kovacs, P.; Kovacs, C. E.; Kolonits,
P.; Szantay, C. Synth. Commun. 1999, 29, 1757-1766.
(36) Pihko, P. M. Angew. Chem., Int. Ed. 2004, 43, 2062-2064.
(37) Ma, D.; Cheng, K. Tetrahedron: Asymmetry 1999, 713-719.
(38) Sohtome, Y.; Tanatani, A.; Hashimoto, Y.; Nagasawa, K.
Tetrahedron Lett. 2004, 45, 5589-5592.
(39) Joly, G. D.; Jacobsen, E. N. J. Am. Chem. Soc. 2004, 126, 4102-
4103.
(40) Wenzel, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2002, 124,
12964-12965.
(41) Benning, M. M.; Taylor, K. L.; Liu, R.; Yang, G.; Xiang, H.;
Wesenberg, G.; Dunaway-Mariano, D.; Holden, H. M. Biochemistry
1996, 35, 8103-8109.
(42) Bahnson, B. J.; Anderson, V. E.; Petsko, G. A. Biochemistry
2002, 41, 2621-2629.
(43) (a) Babbit, P. C.; Gerlt, J. A. J. Biol. Chem. 1997, 272, 30591-
30594. (b) Xiang, H.; Luo, L.; Taylor, K. L.; Dunaway-Mariano, D.
Biochemistry 1999, 38, 7638-7652.
(44) Kim, J. J.; Wang, M.; Paschke, R. Proc. Natl. Acad. Sci. U.S.A.
1993, 90, 7523-7527.
(45) Djordjevic, S.; Pace, C. P.; Stankovich, M. T.; Kim, J. J.
Biochemistry 1995, 34, 2163-2171.
(46) (a) Usher, K. C.; Remington, S. J.; Martin, D. P.; Drueckham-
mer, D. G. Biochemistry 1994, 33, 7753-7759. (b) Yang, W.; Drueck-
hammer, D. G. J. Phys. Chem. B 2003, 107, 5986-5994.
7758 J. Org. Chem., Vol. 70, No. 19, 2005

Modeling and Catalyst Design for Enolate Formation
to the plane of the developing enolate directed from the
face on which deprotonation is occurring, while the
second methanol molecule formed an in-plane hydrogen
bond. It appears that the urea complex of the transition
state approximates this mode of binding to the extent
that the constraints of the urea allow. One of the
hydrogen bonds is directed approximately perpendicular
to the plane and the other approximately in the plane of
the developing enolate. This also permits the hydrogen
bonds to be more linear than would be possible if both
hydrogen bonds were directed toward the oxygen lone
electron pairs. These results support Anslyn’s observa-
tions and further suggest that the π-bond directed
hydrogen bonding may be even more important in the
transition state than in the enolate complex.
Calculations on the catalyst structure predict that a
collapsed intramolecular hydrogen-bonded structure is
possible but that such a structure should not dominate
the catalyst conformation in solution. The NMR results
are consistent with some degree of intramolecular hy-
drogen bonding between the amine and the internal
thiourea proton in 10, but with no significant intermo-
lecular hydrogen bonding at this concentration. Further
NMR shifts are observed in the presence of acetone, as
expected due to hydrogen bonding between acetone and
the thiourea. This indicates that intramolecular hydrogen
bonding of the catalyst does not prevent hydrogen bond-
ing to substrate, and substrate binding likely breaks up
the intramolecular hydrogen bonding. The absence of a
point of saturation in thiourea proton shifts upon addition
of acetone suggests fairly weak binding of acetone to the
catalyst, though an actual association constant could not
be determined. The solvent-dependent shift could also be
due in part to more general solvent effects in addition to
hydrogen bonding.
The proton exchange reactions used to demonstrate
catalysis require an exchange reaction of the protonated
amine with solvent in the enolate complex. This proton
exchange between heteroatoms is expected to be very fast
relative to the proton exchange from and to carbon. While
this was not verified in the present study, such exchange
is a classic observation in enzyme catalysis where the
intermediates are expected to be very short-lived and the
basic group in the enzyme active site could be somewhat
shielded from solvent.47 Comparison of the exchange
reaction of deuterated versus nondeuterated acetone
indicates a kinetic isotope effect that is quite small for a
reaction in which a proton is being transferred in the slow
step. The protonated versus deuterated state of the
thiourea moiety of the catalyst and the secondary effect
of the other deuterium atoms on acetone may also have
some influence on these relative rates. However, these
effects are likely to be small such that the actual primary
kinetic isotope effect is still probably much less than two.
This small kinetic isotope effect likely indicates a very
late transition state, in which the N-H (N-D) bond is
almost fully formed.
indicates the importance of proper conformation of the
catalyst. The rates compare favorably with previously
reported bifunctional catalysts for enolate or enol forma-
tion, with rate constants of about 10-3 min-1 typically
observed.15,16 An exception is the Kemp’s triacid-based
catalyst of Rebek and co-workers, in which a protonated
R-amino group of the substrate facilitates binding and
transition state stabilization, and a kcat of 0.15 min-1
was observed.21 A macrocyclic polyamine was shown by
Lehn and co-workers to catalyze H/D exchange in mal-
onate ion at a rate of 13.2 min-1, though with other
catalysts the deprotonation of acetone has been shown
to be about 1000-fold slower than deprotonation of the
more acidic malonate ion.19,20
The catalyst designed in this work and previously
reported catalysts for enolate formation all fall woefully
short of the ideal of mimicking the rates of enzyme
catalysis. A well-known challenge in enolate formation
is that the charge in the transition state resides largely
on carbon rather than oxygen due to nonperfect synchro-
nization.48,49 Computational studies modeled after en-
zyme active sites support the potential for hydrogen
bonding to carbonyl oxygen in stabilizing the transition
state for enolate formation, and indeed it appears that
such hydrogen bonding is the primary means of transi-
tion state stabilization in enolate-forming enzymes.50
However, most simple/designed enolate receptors provide
only about a 1 pKa unit increase in acidity of the
conjugate acid of their enolate ligands, though Anslyn’s
receptor having hydrogen-bonding groups directed at the
π-system showed a 2.9 pKa unit enhancement.22,27 The
nonperfect synchronization concept would suggest an
even smaller effect on the transition state for enolate
formation. An intramolecular hydrogen bond donor was
shown to provide a 10-100-fold increase in the rate of
enolate formation in a model system.48 The 5-fold en-
hancement in the rate exhibited by the thiourea moiety
in the catalyst reported in this work, though generally
disappointing, is reasonably good based on these com-
parisons. The nonlinearity of the hydrogen bonds be-
tween the oxygen and the thiourea probably diminishes
their ability to stabilize the enolate and transition state
for its formation, and groups that allow more linear
hydrogen bonds could offer some improvement. While an
even more optimal and perhaps more rigid linker might
provide some further enhancement in catalysis, other
kinds of enolate-stabilizing groups such as metal com-
plexes might be more effective than hydrogen-bonding
groups and could be amenable to the same design
strategy for linking catalytic groups.
The work described here demonstrates that a simple
linker designed using CAVEAT can provide an appropri-
ate relative positioning of a pair of functional groups for
bifunctional catalysis. This approach may be useful for
designing more effective catalysts for reactions that are
more easily catalyzed by a pair of properly positioned
functional groups and may thus provide a basis for future
developments in bifunctional catalysis.
The exchange rates indicate that the proper arrange-
ment of electrophilic and basic groups can provide
enhanced catalysis relative to the two groups as separate
entities. The comparison of results with 10 versus 13 also
(48) Zhong, Z.; Snowden, T. S.; Best, M. D.; Anslyn, E. V. J. Am.
Chem. Soc. 2004, 126, 3488-3495.
(49) Bernasconi, C. F. Acc. Chem. Res. 1992, 25, 9-16.
(50) Bach, R. D.; Thorpe, C.; Dmitrenko, O. J. Phys. Chem. B 2002,
106, 4325-4335.
(47) Dinovo, E. C.; Boyer, P. D. J. Biol. Chem. 1971, 246, 4586-
4593.
J. Org. Chem, Vol. 70, No. 19, 2005 7759

Zhu and Drueckhammer
2H), 2.22 (s, 6H), 1.52 (m, 3H), 1.25 (m, 1H), 1.16 (m, 1H),
0.86 (s, 9H). 13C NMR in CDCl3: 180.1, 136.6, 129.7, 126.7,
125.0, 59.5, 47.7, 46.6, 45.1, 33.2, 27.7, 26.8, 26.3. HRMS calcd
for C18H32N3S 322.2317, found 322.2322.
Experimental Section
3-Dimethylaminopropylmagnesium Chloride. To an
aqueous solution of sodium hydroxide (105 mL, 1 M) and
diethyl ether (50 mL) at 0 °C was slowly added 1-chloro-3-
(dimethylamino)propane hydrochloride (15.8 g, 100 mmol). The
layers were separated, and the aqueous phase was extracted
with diethyl ether (50 mL × 2). The combined organic layers
were washed with brine (30 mL) and carefully concentrated
under vacuum to a volume of around 20 mL. The solution was
then diluted with 50 mL of THF. To a three-necked flask
equipped with a condenser and an addition funnel under
nitrogen was added freshly ground magnesium turnings (1.9
g, 78 mmol), a crystal of iodine, THF (5 mL), a drop of ethylene
bromide, and a few drops of the 1-chloro-3-(dimethylamino)-
propane solution. The mixture was gently heated to reflux
until the color of the solution disappeared. The solution of
1-chloro-3-(dimethylamino)propane in THF was then added at
a rate sufficient to maintain reflux. The solution was heated
at reflux for an additional hour before cooling to room
temperature. The clear solution was transferred to a flask
containing a few crystals of 1,10-phenanthroline. The solution
became red after storing in the refrigerator overnight. The red
solution was titrated with 2-butanol to give a concentration
of 0.81 M.
N-5-(Dimethylamino)-2-tert-butyl-pentyl N′-Phenyl
Thiourea 10. To a solution of 1133 (0.646 g, 5 mmol) in THF
(10 mL) at -30 °C was added dropwise a solution of 3-di-
methylaminopropylmagnesium chloride in THF (7 mL, 0.81
M, 5.7 mmol). The solution was slowly warmed to -20 °C and
stirred for 1 h. The reaction was quenched with saturated
aqueous ammonium chloride (20 mL). The aqueous phase was
extracted with diethyl ether (3 × 20 mL). The combined
organic layers were dried over magnesium sulfate and con-
centrated to give N,N,5,5-tetramethyl-4-nitromethyl-1-hexa-
neamine (0.778 g, 3.6 mmol), which was used in the following
step without further purification.
To a solution of N,N,5,5-tetramethyl-4-nitromethyl-1-hexa-
neamine (0.778 g, 3.6 mmol) in THF (15 mL) was added
lithium aluminum hydride (0.27 g, 7.1 mmol). The mixture
was heated at reflux for 3 h and carefully quenched at 0 °C
with aqueous sodium hydroxide solution (20 mL, 4 M). The
solution was extracted with diethyl ether (3 × 15 mL). The
combined organic layers were washed with brine (20 mL),
dried over magnesium sulfate, and concentrated to give
N,N,5,5-tetramethyl-4-aminomethyl-1-hexaneamine (0.649 g,
3.48 mmol), which was used in the following step without
further purification.
To a solution of phenyl isothiocyanate (0.49 g, 3.62 mmol)
in methylene chloride (3 mL) was added a solution of N,N,5,5-
tetramethyl-4-aminomethyl-1-hexaneamine (0.649 g, 3.48 mmol)
in methylene chloride (3 mL). The solution was stirred at room
temperature for 3 h. The solvent was removed under vacuum,
and the product was purified by chromatography on silica gel
(2:1 ethyl acetate/methanol) to give 10 (0.838 g, 2.6 mmol) in
53% yield over three steps. Rf 0.25 in 2:1 ethyl acetate/
methanol. 1H NMR in CDCl3: 8.20 (br, 1H), 7.37 (m, 2H), 7.25
(m, 3H), 6.58 (br, 1H), 3.77 (br, 1H), 3.49 (br, 1H), 2.33 (m,
N-5-Dimethylaminopentyl N′-Phenyl Thiourea 13. To
a solution of 5-dimethylaminovaleronitrile 1234 (0.63 g, 5
mmol) in THF (20 mL) was slowly added lithium aluminum
hydride (0.38 g, 10 mmol). The mixture was heated at reflux
overnight. The remaining lithium aluminum hydride was
carefully quenched with aqueous sodium hydroxide solution
(10 mL, 4 M). The solution was extracted with diethyl ether
(3 × 20 mL). The combined organic layers were washed with
brine (20 mL), dried over magnesium sulfate, and concentrated
to give the crude 5-dimethylaminopentylamine (0.43 g, 3.3
mmol). The crude amine was dissolved in methylene chloride
(3 mL) and then added to a solution of phenyl isothiocyanate
(0.47 g, 3.5 mmol) in methylene chloride (3 mL). The solution
was stirred at room temperature for 5 h. The solvent was
removed under vacuum, and the product was purified by
chromatography on silica gel (1:1 ethyl acetate/methanol) to
give 13 (0.62 g, 2.34 mmol, 47% yield over two steps). Rf 0.12
in 1:1 ethyl acetate/methanol. 1H NMR in CDCl3: 9.16 (br,
1H), 7.37 (m, 2H), 7.23 (m, 3H), 6.41 (br, 1H), 3.57 (q, 2H),
2.25 (t, 2H), 2.19 (s, 6H), 1.56 (m, 2H), 1.45 (m, 2H), 1.32 (q,
2H). 13C NMR in CDCl3: 179.7, 136.5, 129.3, 126.1, 124.4, 58.9,
44.8, 44.5, 28.3, 26.5, 24.0. HRMS calcd for C14H24N3S 266.1691,
found 266.1684.
Typical Procedure for Deuterium Exchange Experi-
ments. The catalyst 10 (32.1 mg, 0.1 mmol) was dissolved in
acetone-d6 (0.5 mL), and methanol (50 µL, 1.23 mmol) was
added. The exchange reaction was monitored by 1H NMR with
the disappearing OH peak of methanol and the emerging
proton peak of acetone over time. The methyl protons of
methanol were used as the internal standard.
Deuterium Exchange of Nondeuterated Acetone with
Methanol-d4. The catalyst 10 (32.1 mg, 0.1 mmol) was
dissolved in 0.5 mL of acetone, and methanol-d4 (50 µL) was
added. NMR conditions were optimized with a solution of 0.5
mL of acetone-d6 and 50 µL of methanol. The exchange
reaction was monitored by 1H NMR in the unlock mode
following the emerging OH peak of methanol over time. Both
the tert-butyl protons and the proton at δ 3.70 in the thiourea
were used as the internal standard.
Acknowledgment. Financial support from the Pe-
troleum Research Fund of the American Chemical
Society (38440-AC4) and the National Science Founda-
tion (CHE0213457) is gratefully acknowledged. NMR
facilities were supported by a grant from the National
Science Foundation (CHE0131146).
Supporting Information Available: General experimen-
tal methods, complete descriptions of the calculated geometries
1
of 5-7, and H NMR spectra for 10 and 13. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO0513818
7760 J. Org. Chem., Vol. 70, No. 19, 2005