A class II aldolase mimic

Article abstract of DOI:10.1021/jo060608b

A class II aldolase-mimicking synthetic polymer was prepared by the molecular imprinting of a complex of cobalt (II) ion and either (1S,3S,4S)-3-benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (4a) or (1R,3R,4R)-3-benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (4b) in a 4-vinylpyridine-styrene-divinylbenzene copolymer. Evidence for the formation of interactions between the functional monomer and the template was obtained from NMR and VIS titration studies. The polymers imprinted with the template demonstrated enantioselective recognition of the corresponding template structure, and induced a 55-fold enhancement of the rate of reaction of camphor (1) with benzaldehyde (2), relative to the solution reactions, and were also compared to reactions with a series of reference polymers. Substrate chirality was observed to influence reaction rate, and the reaction could be competitively inhibited by dibenzoylmethane (6). Collectively, the results presented provide the first example of the use of enantioselective molecularly imprinted polymers for the catalysis of carbon-carbon bond formation.

Full text of DOI:10.1021/jo060608b

A Class II Aldolase Mimic
Jimmy Hedin-Dahlstro¨m, Jenny P. Rosengren-Holmberg, Sacha Legrand,
Susanne Wikman, and Ian A. Nicholls*
Bioorganic and Biophysical Chemistry Laboratory, Department of Chemistry and Biomedical Sciences,
UniVersity of Kalmar, SE-391 82 Kalmar, Sweden
ian.nicholls@hik.se
ReceiVed March 20, 2006
A class II aldolase-mimicking synthetic polymer was prepared by the molecular imprinting of a complex
of cobalt (II) ion and either (1S,3S,4S)-3-benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (4a) or
(1R,3R,4R)-3-benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (4b) in a 4-vinylpyridine-styrene-
divinylbenzene copolymer. Evidence for the formation of interactions between the functional monomer
and the template was obtained from NMR and VIS titration studies. The polymers imprinted with the
template demonstrated enantioselective recognition of the corresponding template structure, and induced
a 55-fold enhancement of the rate of reaction of camphor (1) with benzaldehyde (2), relative to the
solution reactions, and were also compared to reactions with a series of reference polymers. Substrate
chirality was observed to influence reaction rate, and the reaction could be competitively inhibited by
dibenzoylmethane (6). Collectively, the results presented provide the first example of the use of
enantioselective molecularly imprinted polymers for the catalysis of carbon-carbon bond formation.
Introduction
The molecular imprinting technique8 provides a means for
the synthesis of functionally and stereochemically defined
environments in which to perform selective reactions.7,9 The
inherent stability of these highly cross-linked polymers makes
them of particular interest for applications where extremes of
The development of new methodologies for the catalysis of
carbon-carbon bond formation remains one of the greatest
challenges for organic chemistry.1 The desire to produce systems
mimicking those demonstrated by biological macromolecular
catalysts, i.e., enzymes2 and ribozymes,3 requires not just a
capacity to enhance the rate of a given reaction, but also that
the system can provide some control over substrate selectivities
and display turnover. These additional goals exacerbate the
complexity of the task. Nonetheless, a number of quite diverse
strategies have been utilized in order to produce biomimetic
systems capable of catalyzing C-C bond formation,4 including
the use of chiral Lewis acids,5 catalytic antibody technology,6
and molecular imprinting.7
(5) Mastrorilli, P.; Nobile, C. F. Coord. Chem. ReV. 2004, 248, 377-
395.
(6) Schultz, P. G.; Yin, J.; Lerner, R. A. Angew. Chem., Int. Ed. 2002,
41, 4427-4437.
(7) Alexander, C.; Davidson, L.; Hayes, W. Tetrahedron 2003, 59, 2025-
2057.
(8) (a) Sellergren, B., Ed. Molecularly Imprinted Polymers. Man-made
Mimics of Antibodies and Their Applications in Analytical Chemistry;
Elsevier: Amsterdam, The Netherlands, 2001. (b) Yan, M.; Ramstro¨m, O.,
Eds. Molecularly Imprinted Materials. Science and Technology; Marcel
Dekker: New York, 2005. (c) Alexander, C.; Andersson, H. S.; Andersson,
L. I.; Ansell, R. J.; Kirsch, N.; Nicholls, I. A.; O’Mahony, J.; Whitcombe,
M. J. J. Mol. Recognit. 2006, 19, 106-180.
(9) (a) Whitcombe, M. J.; Alexander, C.; Vulfson, E. N. Synlett 2000,
6, 911-923. (b) Motherwell, W. B.; Bingham, M. J.; Six, Y. Tetrahedron
2001, 57, 4663-4686. (c) Bruggemann, O. Anal. Chim. Acta 2001, 435,
197-207. (d) Tada, M.; Iwasawa, Y. J. Mol. Catal. A 2003, 199, 115-
137. (e) Toorisaka, E.; Uezu, K.; Goto, M.; Furusaki, S. Biochem., Eng. J.
2003, 14, 85-91. (f) Striegler, S. J. Chromatogr. B 2004, 804, 183-195.
(g) Cheng, Z. Y.; Zhang, L. W.; Li, Y. Z. Chem. Eur. J. 2004, 10, 3555-
3561.
* Address correspondence to this author. Phone: +46-480 446258. Fax: +46-
480 446244.
(1) (a) Danishefsky, S. Science 1993, 259, 469-470. (b) Sukumaran, J.;
Hanefeld, U. Chem. Soc. ReV. 2005, 34, 530-42.
(2) (a) Bornscheuer, U. T.; Kazlauskas, R. J. Angew. Chem., Int. Ed.
2004, 43, 6032-6040. (b) Jaeger, K.-E.; Eggert, T. Curr. Opin. Biotechnol.
2004, 15, 305-313.
(3) (a) Lilley, D. M. J. Curr. Opin. Struct. Biol. 2005, 15, 313-323. (b)
Sigel, R. K. O. Eur. J. Inorg. Chem. 2005, 12, 2281-2292.
(4) Severin, K. Curr. Opin. Chem. Biol. 2000, 4, 710-714.
10.1021/jo060608b CCC: $33.50 © 2006 American Chemical Society
Published on Web 06/02/2006
J. Org. Chem. 2006, 71, 4845-4853
4845

Hedin-Dahlstro¨m et al.
SCHEME 1. Aldol Condensation between (S)-Camphor (1a)
and Benzaldehyde (2) Results in Formation of
(S)-3-Benzylidenecamphor (3a) and the Same Reaction with
(R)-Camphor (1b) Results in Formation of
(R)-3-Benzylidenecamphor (3b)
FIGURE 1. (a) Proposed transition state (TS) for the first step of the
aldol condensation reaction. (b) General structure of the putative (S)-
TSAs: exo (4a) and endo (5a). The corresponding (R)-TSAs are defined
as 4b (exo) and 5b (endo).
should also serve as suitable ligands for coordination to the metal
ion. Molecularly imprinted polymers synthesized by using
complexes of 4 with Co2+ in a 4-vinylpyridine (4-VP)-styrene-
divinylbenzene (DVB) copolymer demonstrated significant
enhancement of the rate of reaction relative to reference
polymers and the solution reaction. Moreover, polymers syn-
thesized with either enantiomer of 4 displayed selectivity for
substrate structures of the corresponding optical antipode.
temperature, solvent regime, or pH prohibit the use of catalysts
of biological origin.10 The technique has been used with success
for preparing polymers capable of enhancing the reaction rate
of a number of types of reactions including various hydrolytic
reactions,11 transamination,12 and â-elimination.13 Previous
efforts to develop systems for the catalysis of C-C bond
forming reactions have been reported by us, aldol condensa-
tion,14 and others, Diels-Alder cyclization15 and the Suzuki
reaction.16
Results and Discussion
The enantioselective synthetic aldolase-mimicking polymers
presented in this study were designed and synthesized by using
a metal ion mediated molecular imprinting strategy. The reaction
chosen for investigation involves the condensation of camphor
(1) and benzaldehyde (2) in the presence of a mild base,
pyridine, to yield 3-benzylidenecamphor (3) (Scheme 1). The
choice was in part due to the inherent chirality present in
camphor, and in part due to the presence of a single hydrogen-
bearing R-carbon, which provides a natural limit to the number
of possible reaction products. Camphor’s chirality has previously
been utilized for steering the stereochemical outcome of aldol
reactions employing titanium enolates of camphorselenoacetone
and methyl camphorselenoacetate,25 and for a range of other
asymmetric syntheses26 involving diols and aminodiols27 and
lithium enolates of R-hydroxy ketones.28
Design and Synthesis of Transition State Analogues. The
choice of the putative TSAs (4 and 5) proposed for use in this
study was based upon our previous experience with a related
aldolase mimicking polymer selective for the production of
chalcone (8),14 whereby complexes of the TSA with Co2+ would
provide a mimic for the transition state of the aldol reaction
(Scheme 1, Figure 1). This bidentate ligand was expected to
The aldol condensation is a reaction of central importance to
both biology17,18 and synthetic organic chemistry.19,20 Accord-
ingly, significant effort has been directed to the development
of catalysts for this class of reaction, and to the establishment
of means for controlling the stereochemistry of the reaction
outcome, e.g., Evans’ oxazolidinones,21 catalytic antibodies,22
chiral Lewis acids,23 and molecular imprinting.14
In the present study we report the design, synthesis, and
evaluation of enantioselective molecularly imprinted polymers
with activity mimicking that of a class II aldolase, a metallo-
enzyme found in lower organisms such as bacteria and yeast.24
The reaction of (S)- or (R)-camphor (1) and benzaldehyde (2)
to yield the corresponding (E)-3-benzylidene-1,7,7-trimethylbi-
cyclo[2.2.1]heptan-2-one (3) was chosen for use in this study
(Scheme 1). The diketones 4 and 5 (Figure 1) were envisaged
as putative analogues for the transition state for the first step of
this Co2+ catalyzed aldol condensation reaction. Furthermore,
the diketone functionalities, or keto-enol tautomers thereof,
(10) Svenson, J.; Nicholls, I. A. Anal. Chim. Acta 2001, 435, 19-24.
(11) (a) Strikovsky, A. G.; Kasper, D.; Gru¨n, M.; Green, B. S.; Hradil,
J.; Wulff, G. J. Am. Chem. Soc. 2000, 122, 6295-6296. (b) Wulff, G.;
Gross, T.; Scho¨nfeld, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 1961-
1964. (c) Sellergren, B.; Shea, K. J. Tetrahedron: Asymmetry 1994, 5, 1403-
1406. (d) Sellergren, B.; Karmalkar, R. N.; Shea, K. J. J. Org. Chem. 2000,
65, 4009-4027.
(21) (a) Evans, D. A.; Kim, A. S. In Handbook of Reagents for Organic
Synthesis: Reagents, Auxiliaries and Catalysts for C-C Bonds; Coates, R.
M., Denmark, S. E., Eds.; John Wiley & Sons: New York, 1999; pp 91-
101. (b) Evans, D. A.; Bartroli, J.; Shih, T. L. J. Am. Chem. Soc. 1981,
103, 2127-2129. (c) Evans, D. A.; Nelson, J. V.; Taber, T. R. Top.
Stereochem. 1982, 13, 1-115.
(22) (a) Hoffmann, T.; Zhong, G.; List, B.; Shabat, D.; Anderson, J.;
Gramatikova, S.; Lerner, R. A.; Barbas, C. F., III J. Am. Chem. Soc. 1998,
120, 2768-2779. (b) Zhong, G.; Lerner, R. A.; Barbas, C. F., III Angew.
Chem., Int. Ed. 1999, 38, 3738-3741.
(23) (a) Calter, M. A.; Song, W.; Zhou, J. G. J. Org. Chem. 2004, 69,
1270-1275. (b) Saito, S.; Yamamoto, H. Acc. Chem. Res. 2004, 37, 570-
579.
(12) Svenson, J.; Zheng, N.; Nicholls, I. A. J. Am. Chem. Soc. 2004,
126, 8554-8560.
(13) (a) Mu¨ller, R.; Andersson, L. I.; Mosbach, K. Makromol. Chem.-
Rapid Commun. 1993, 14, 637-641. (b) Beach, J. V.; Shea, K. J. J. Am.
Chem. Soc. 1994, 116, 379-380.
(14) Matsui, J.; Nicholls, I. A.; Karube, I.; Mosbach, K. J. Org. Chem.
1996, 61, 5414-5417.
(15) (a) Liu, X.-C.; Mosbach, K. Macromol. Rapid Commun. 1997, 18,
609-615. (b) Visnjevski, A.; Schoma¨cker, R.; Yilmaz, E.; Bru¨ggemann,
O. Catal. Commun. 2005, 6, 601-606. (c) Busi, E.; Basosi, R.; Ponticelli,
F.; Olivucci, M. J. Mol. Catal. A: Chem. 2004, 217, 31-36.
(16) Cammidge, A. N.; Baines, N. J.; Bellingham, R. K. Chem. Commun.
2001, 2588-2589.
(24) Heron, E. J.; Caprioli, R. M. Biochim. Biophys. Acta 1975, 403,
563-572.
(17) Mikami, K.; Yajima, T.; Takasaki, T.; Matsukawa, S.; Terada, M.;
Uchimaru, T.; Maruta, M. Tetrahedron 1996, 52, 85-98.
(18) Mahrwald, R., Ed. Modern Aldol Reactions, Vol. 1: Enolates,
Organocatalysis, Biocatalysis and Natural Product Synthesis; Wiley-VCH
Verlag GmbH & Co., KGaA: Weinheim, Germany, 2004.
(19) Machajewski, T. D.; Wong, C.-H. Angew. Chem., Int. Ed. 2000,
39, 1352-1374.
(25) Tiecco, M.; Testaferri, L.; Marini, F.; Sternativo, S.; Santi. C.;
Bagnoli, L.; Temperini, A. Tetrahedron: Asymmetry 2004, 15, 783-791.
(26) Knollmuller, M.; Ferencic, M.; Ga¨rtner, P.; Mereiter, K.; Noe, C.
R. Tetrahedron: Asymmetry 1998, 9, 4009-4020.
(27) Palomo, C.; Oiarbide, M.; Aizpurua, J. M.; Gonza´lez, A.; Garc´ıa,
J. M.; Landa, C.; Odriozola, I.; Linden, A. J. Org. Chem. 1999, 64, 8193-
8200.
(20) Mahrwald, R., Ed. Modern Aldol Reactions, Vol. 2: Metal Catalysis;
Wiley-VCH Verlag GmbH & Co., KGaA: Weinheim, Germany, 2004.
(28) Palomo, C.; Oiarbide, M.; Mielgo, A.; Gonza´lez, A.; Garc´ıa, J. M.;
Landa, C.; Lecumberri, A.; Linden, A. Org. Lett. 2001, 3, 3249-3252.
4846 J. Org. Chem., Vol. 71, No. 13, 2006

A Class II Aldolase Mimic
SCHEME 2. Synthesis of Diketone 4a from (S)-Camphor
(1a)a
a Diketone 4b was obtained from (R)-camphor (1b), using the same
reaction conditions.
FIGURE 2. Proposed metal (Co2+) ion coordinated complex formation
between the enolate of the TSA (4 or 5), Co2+, and 4-vinylpyridine,
showing four of six possible coordinating functionalities.
fill two of the coordination sites of the Co2+ using its two
oxygens, while 4-vinylpyridine (and/or solvent) should fill the
remaining sites of the Co2+ complex (Figure 2). It was envisaged
that the keto-enol tautomerism available to the â-diketones
would allow for a planar geometry between the two oxygens,
as would also be the case in the corresponding enolate and its
various tautomers (see the Supporting Information).
The use of metal ions in molecular imprinting protocols can
provide a number of advantages in the preparation of synthetic
receptors29 and enzyme mimics.14,30 The general strengths of
transition metal ion-ligand coordination interactions can permit
complex formation in polar solvents not normally conducive
for use in noncovalent molecular imprinting strategies. Further-
more, the possibility for forming multiple interactions to a single
ion allows for the simultaneous coordination of multiple ligands,
e.g., reaction substrates.
The synthesis of each of the enantiomers of the diketones 4
and 5 was undertaken in order to obtain material for use in the
polymer syntheses and for polymer-ligand recognition studies.
The exo-products, 4a and 4b, were obtained in moderate yield,
as the exclusive products from the treatment of the correspond-
ing enantiomer of camphor (1a or 1b) with NaH and ethyl
benzoate (Scheme 2), using an adaption of a procedure
previously described by Togni.31
FIGURE 3. The dashed lines represent selected observed NOESY
correlations of 4a. The same correlations were observed for 4b.
FIGURE 4. Structure of the endo diketones 5a and 5b.
SCHEME 3. Synthesis of the Diketones 5a and 4a from 7a
Attempts were made to obtain the corresponding endo-
isomers, the diketones 5a and 5b (Figure 4), using a procedure
described by Wei et al.32 in order to provide alternative analytes
The benzoyl substituent of 4a was found to be in the exo-
configuration on the basis of the observed NOESY correlations
arising from the HR positioned between the two carbonyls and
the two CHendo protons (Figure 3). The 1H NMR spectra of 4a
(or 4b) revealed an equilibrium between the diketo form and
the two keto-enol forms, with a ratio of 3 to 7 in favor of the
1
for use in polymer-ligand recognition studies. The H NMR
spectra of the crude products arising from the treatment of
bromocamphor, 7a, with SmI2 in the presence of benzoyl
chloride under samarium-Barbier conditions33 demonstrated a
peak characteristic of the endo HR (doublet at 2.85-2.83 ppm)
of 4a. Furthermore, a doublet of doublets was observed at 4.25
ppm corresponding to the signal of the exo HR to the carbonyl
moieties present in 5a. This was interpreted as being indicative
of the presence of a mixture of the exo-diketone 4a and the
endo-diketone 5a (Scheme 3), in a 2:1 ratio in favor of 4a.
Attempts to separate 5a from 4a by flash chromatography on
silica failed to achieve separation, irrespective of matrix (neutral,
acidic, or basic), as shown by the disappearance of the doublet
of doublets at 4.25 ppm in the 1H NMR spectrum. This implied
that the exo-product is the thermodynamically more stable of
the two, and that the initial mixture reflects the presence of both
kinetic (endo) and thermodynamic (exo) products, for which
keto-enol tautomerism provides a mechanism for interchange
between the two (Figure 5, see also the Supporting Information
1
diketone. Partial assignment of the H and 13C NMR spectra
was accomplished by the application of a combination of
conventional 1D and 2D NMR experiments.
(29) (a) Gupta, S. N.; Neckers, D. C. J. Polym. Sci., Polym. Chem. Ed.
1982, 20, 1609-1622. (b) Vidyasankar, S.; Ru, M.; Arnold, F. H. J.
Chromatogr. A 1997, 775, 51-63. (c) Hart, B. R.; Shea, K. J. J. Am. Chem.
Soc. 2001, 123, 2072-2073. (d) Takeuchi, T.; Mukawa, T.; Matsui, J.;
Higashi, M.; Shimizu, K. D. Anal. Chem. 2001, 73, 3869-3874. (e)
Efendiev, A. A. Macromol. Symp. 1994, 80, 289-313. (f) Matsui, J.;
Nicholls, I. A.; Takeuchi, T.; Mosbach, K.; Karube, I. Anal. Chim. Acta
1996, 335, 71-77. (g) Fujii, Y.; Matsutani, K.; Kikuchi, K. J. Chem. Soc.,
Chem. Commun. 1985, 415-417. (h) Dhal, P. K.; Arnold, F. H. J. Am.
Chem. Soc. 1991, 113, 7417-7418. (i) Striegler, S.; Tewes, E. Eur. J. Inorg.
Chem. 2002, 487-495. (j) Striegler, S. Anal. Chim. Acta 2005, 539, 91-
95. (k) Striegler, S.; Dittel M. Anal. Chim. Acta 2003, 484, 53-62.
(30) (a) Brunkan, N. M.; Gagne´, M. R. J. Am. Chem. Soc. 2000, 122,
6217-6225. (b) Santora, B. P.; Larsen, A. O.; Gagne´, M. R. Organome-
tallics 1998, 17, 3138-3140. (c) Koh, J. H.; Larsen, A. O.; White, P. S.;
Gagne´, M. R. Organometallics 2002, 21, 7-9. (d) Wulff, G.; Vietmeier, J.
Makromol. Chem. - Macromol. Chem. Phys. 1989, 190, 1727-1735. (e)
Leonhardt, A.; Mosbach, K. React. Polym. 1987, 6, 285-290.
(32) Wei, H.-X.; Wang, Z.-M.; Shi, M. Chem. Pharm. Bull. 1999, 47,
909-910.
(33) Namy, J. L.; Girard, P.; Kagan, H. B. NouV. J. Chim. 1977, 1, 5-7.
(31) Togni, A. Organometallics 1990, 9, 3106-3113.
J. Org. Chem, Vol. 71, No. 13, 2006 4847

Hedin-Dahlstro¨m et al.
FIGURE 6. The dashed lines represent selected NOESY correlations
of the unsaturated ketone 3a.
the presence and strength of complexes between Co2+, TSA
(4a or 4b), and pyridine (here used as an analogue for 4-VP).
The monitoring of titrations of Co2+ with pyridine or TSA at
520 nm (294 ( 1 K) revealed complexes with apparent
dissociation constants (app Kdiss) of 228.9 ( 18.3 and 4.0 (
1.6 mM, respectively. By using conditions and concentrations
comparable to those used (see later) in the polymerization
reaction, namely using 2 equiv of pyridine per Co2+, an app
FIGURE 5. Isomerization of 5a to 4a.
SCHEME 4. Synthesis of Ketone 3a from (S)-Camphor (1a)a
Kdiss of 25.6 ( 3.8 mM was determined, indicating that the TSA
can compete for coordination of the metal ion. These data were
supported by 1H NMR studies, from which an app Kdiss of 2.50
( 0.39 mM was determined by following the downfield shift
of the HR. Complementary VIS studies with Job’s method of
continuous variation35,36 demonstrated a 1:1 stoichiometry for
the solution complexes of Co2+ and 4. On the basis of molecular
model studies, we propose that the steric bulk of 4 excludes
the formation of complexes of greater than 1:1 complex
stoichiometry under the conditions studied. On account of the
complex stabilities described above, we interpret the favorable
formation of 1:1:2 complexes of Co2+/TSA/pyridine, relative
to 1:2 complexes of Co2+/TSA on account of the relative bulk
of the TSA. Importantly, these results collectively demonstrate
that complexes of Co2+ by pyridine and TSA are formed at the
concentrations utilized in subsequent polymerization reactions.
The role of Co2+ in the complex is 2-fold, in the first instance
to provide coordination of the template during the molecular
imprinting process, and second to facilitate binding of reaction
substrates in the subsequent polymer.
Polymer Synthesis and Characterization. A series of
4-vinylpyridine-styrene-divinylbenzene copolymers were syn-
thesized by thermally induced radical polymerization with use
of azobis(cyclohexanecarbonitrile) (ABCC) as initiator (Table
1). Two polymers, one prepared in the absence of both template
(TSA) and Co2+ (P0) and another prepared in the presence of
Co2+ but without TSA (P1), were synthesized to act as
references for polymers prepared by using complexes of the
(S)- and (R)-TSA with Co2+, (P2) and (P3), respectively. The
two reference polymers were anticipated to provide insight
regarding the influence of the polymer material itself on ligand
recognition (P0) and the role of sites selective for cobalt ions
(P1). In the case of P3, its physical and chemical characteristics
were effectively identically with those of P2, though with
selectivity for the (R)-TSA (4b) and (R)-product (3b). Moreover,
no evidence of residual template was evident based upon
examination of the carbonyl region of FT-IR spectra.
a Ketone 3b was synthesized from 1b in the same manner.
for a proposed mechanism). Identical behavior was observed
in the synthesis of 5b from 7b, which resulted in isomerization
to the more favored 4b. Collectively, the results provide an
important insight into the behavior of the diketones, namely
that the keto-enol tautomerism demonstrated by the diketones
provides evidence that the TSAs can adopt a planar geometry,
as proposed for a suitable TSA.
Synthesis of Aldol Condensation Products. The products
from the aldol condensation, the R,â-unsaturated ketones 3a and
3b, were synthesized for use in the establishment of assays and
as standards for polymer-ligand recognition studies, using an
adaption of the procedure described by Chuiko et al.34 En-
antiomerically pure camphor, 1a or 1b, was reacted with
benzaldehyde (2) in the presence of n-BuLi in DMSO to furnish
the corresponding ketone, 3a or 3b, though in low yield (Scheme
4).
NOESY experiments with 3a (or 3b) showed a strong
correlation between the H4 methine and the H3d aromatic protons
(Figure 6), suggesting an (E)-configuration. The lack of any
observed correlation between H4 and H3b supported this conclu-
sion.
Template-Monomer Complexation Studies. Initial studies
on the solubility of Co(OAc)2 suggested the use of methanol
as a suitable solvent for the polymerization reactions. This protic
solvent is not normally suitable for use in noncovalent molecular
imprinting protocols; however, in this case the significant
strength of metal ion coordination surmounts the competition
from bulk solvent. A series of VIS and NMR titration studies
(see the Supporting Information) were performed to establish
(35) (a) Kim, H.; Spivak, D. A. J. Am. Chem. Soc. 2003, 125, 11269-
11275. (b) Takeuchi, T.; Dobashi, A.; Kimura, K. Anal. Chem. 2000, 72,
2418-2422. (c) Striegler, S. Bioseparation 2001, 10, 307-314. (d) Wulff,
G.; Knorr, K. Bioseparation 2001, 10, 257-276.
(36) Karlsson, J. G.; Andersson, L. I.; Nicholls, I. A. Anal. Chim. Acta
2001, 435, 57-64.
(34) Chuiko, V. A.; Vinarskaya, Z. V.; Izotova, L. V.; Tychinskaya, L.
Y. Russ. J. Org. Chem. 2002, 38, 196-199.
4848 J. Org. Chem., Vol. 71, No. 13, 2006

A Class II Aldolase Mimic
TABLE 1. Polymerization Reaction Mixture Compositions and
Polymer Physical Characterization
TABLE 2. Binding of Co2+ to Polymers after Incubation in MeOH
polymer
bounda (mM)
n (µmol/g polymer)
P0a
P1b
P2c
P3d
P0
P1
P2
P3
0.155 ( 0.149
1.352 ( 0.150
0.542 ( 0.001
0.543 ( 0.100
0.776 ( 0.747
6.762 ( 0.747
2.710 ( 0.004
2.713 ( 0.498
(S)-TSA 4a (mmol)
(R)-TSA 4b (mmol)
Co(OAc)2 (mmol)
4-VP (mmol)
styrene (mmol)
DVB (mmol)
2.0
2.0
2.0
4.0
40.0
40.0
1.2
2.0
4.0
40.0
40.0
1.2
2.0
4.0
40.0
40.0
1.2
4.0
40.0
40.0
1.2
a Incubation with Co2+ solution (8 mM) in MeOH (293 K), experiments
performed in duplicate with duplicate analyses.
ABCC (mmol)
MeOH (ml)
14.98
91.8
8.1
0.8
1.9
14.98
90.0
7.9
0.8
3.4
14.98
89.5
7.7
0.8
3.5
14.98
89.8
7.7
0.8
3.9
Polymer P3, prepared with (R)-TSA (4b), behaved similarly,
though as expected with a reversal in enantioselectivity. Under
the conditions studied, the enantioselective binding correlates
to 0.11 µmol enantioselective sites per gram of polymer, i.e.,
the difference between the binding of 4a and 4b to P2, or P3.
The ratio of enantiomer binding correlates to a difference in
free energy of binding between the two enantiomers (∆∆G) of
% C found (theoretical: 91.5)
% H found (theoretical: 7.7)
% N Found (theoretical: 0.7)
BET surface area (m2 g-1
)
micropore volume (cm3 g-1
average pore diameter (Å)
)
0.005
111.0
0.009
104.6
0.009
104.4
0.010
103.4
a Reference polymer. b Co2+ reference polymer. c (S)-TSA imprinted
polymer. d (R)-TSA imprinted polymer.
1.6 kJ mol-1 12,37
Although a significant difference, the results
.
illustrate the site heterogeneity present inherent to P2 and P3.
Binding studies were also performed in DMF at 293 ( 2 K
(see the Supporting Information), which was the solvent of
choice for use in studies on the influence of the polymers on
reaction kinetics. In DMF the polymers demonstrated greater
affinities for the template structure, in particular P0, though with
no enantioselectivity. Interestingly, and in contrast to the results
obtained in MeOH, no significant product binding was observed,
though benzaldehyde (2) displayed a markedly greater affinity
for the polymers, especially in the case of P1.
As the metal ion plays a fundamental role in the catalysis of
the aldol reaction used in this study,14,38 it was crucial to
determine the quantity of Co2+ that bound to the polymers.
Batch binding studies (Table 2) showed that the polymer
synthesized with Co2+ as template, P1, had a significantly
greater capacity for rebinding the divalent cation than that with
P2, or P3. This was interpreted as reflecting the presence of
sites selective for Co2+ rather than for Co2+-TSA complexes
where in principle coordinating moieties, the two ketones, are
lacking in the resultant polymer. Interestingly, P0 showed an
even lower capacity than the other polymers. This is attributed
to the lack of a template, which renders the polymer without
ensembles of pyridinyl functionalities in suitable spatial ar-
rangements for simultaneous interaction with the metal ion.
Reaction Kinetics Studies. The influence of the various
polymers on the rate of condensation of benzaldehyde (2) and
(S)- or (R)-camphor (1a, 1b) was studied by using reactions
performed in sealed tubes with DMF as solvent and elevated
temperature (100 °C). Polymers were charged with methanolic
Co2+ solutions prior to use (Table 2). A solvent reaction
containing pyridine and Co(OAc)2 was employed to allow
assessments of the influence of the polymers themselves. Since
the binding of cobalt to P0 was quite low, studies with this
polymer employed Co2+ concentrations identical with those of
the solvent reaction. An HPLC-based assay was used to monitor
the formation of reaction products 3a or 3b (Figure 8). To
provide a clear picture of the role of the polymer on the reaction
studied, product yields are presented as yield per mole sites,
where the number of sites was determined by the Co2+
concentration in the bound polymer. As stated earlier, the
presence of the metal ion is essential for the reaction to proceed
within the time frames studied.
FIGURE 7. Binding of the 0.015 mM ligand:cobalt complex (1:1) in
MeOH. Each experiment was performed in duplicate with duplicate
HPLC analyses of each sample. Error bars reflect the SD. (Figures
and uncertainties underlying the data presented in this graph are
presented in the Supporting Information, along with results of binding
in DMF.)
Evaluation of Polymer-Ligand Recognition. An assay for
TSA binding to the polymers was developed based upon a series
of polymer titration studies performed with established proce-
dures (data not shown).36 A polymer concentration of 20 mg
mL-1 was chosen for use in the investigation of polymer-
template rebinding in batch binding experiments performed in
MeOH (Figure 7).
In the case studies performed in MeOH, using P0, a polymer
devoid of the influence of both TSA and Co2+ on the polymer’s
recognition characteristics, some preference for binding of the
TSAs was observed relative to the single carbonyl containing
products (3a and 3b), though not surprisingly without any
enantioselectivity. The structurally smaller substrate, benzalde-
hyde (2), demonstrated effectively no recognition of the
polymer. In the case of the polymer synthesized in the presence
of Co2+, P1, the presence of sites selective for the cation
significantly enhanced recognition of the TSA relative to that
observed in P0, though no significant effect was seen on the
binding of 2 or 3 (a or b). This is interpreted as resulting from
the presence of sites selective for Co2+, in which the bound
ions in turn facilitate coordination of the diketone 4.
The (S)-TSA imprinted polymer P2 showed similar affinities
to 3 (a or b) as seen in the case of P1, though a substantial
increase in affinity for benzaldehyde (2). Importantly, an
increased preference for the (S)-TSA 4a, relative to 4b, was
observed, which provides strong evidence for the presence of
sites with selectivity for the (S)-enantiomer of the TSA (4a).
(37) Adbo, K.; Nicholls, I. A. Anal. Chim. Acta 2001, 435, 115-120.
(38) Watanabe, K.; Imazawa, A. Bull. Chem. Soc. Jpn. 1982, 55, 3208-
3211.
J. Org. Chem, Vol. 71, No. 13, 2006 4849

Hedin-Dahlstro¨m et al.
FIGURE 9. Structure of the inhibitor dibenzoylmethane (DBM) (6)
and chalcone (8).
FIGURE 8. Formation of (S)-product (3a) per mol site (Co2+), using
(S)- (P2) and (R)-MIPs (P3), Co2+ (P1), and nonimprinted (P0)
polymers, and solvent reaction (comparable to P0) and the correspond-
ing solvent reaction. Data were obtained from duplicate experiments
with each analysis performed in duplicate. Error bars (not discernible)
reflect SEM < 0.01 µmol/µmol sites.
TABLE 3. Turnover per Co2+ for Production of 3aa
FIGURE 10. Lineweaver-Burk plot of the formation of (S)-product
(3a) with and without the presence of an inhibitor (6), and the reaction
in solvent.
polymer
turnover (h-1
)
P0
P1
P2
P3
0.38 × 10-3
4.63 × 10-3
21.04 × 10-3
20.30 × 10-3
Information) which demonstrated no significant difference in
the swelling characteristics of the polymers used in this study.
Interestingly, the enantioselectivities observed in the binding
studies were not apparent in the studies on the influence of the
polymers on the outcome of the reaction of 1 and 2. While the
binding studies are performed under equilibrium conditions, i.e.,
thermodynamic control, the studies of the kinetics of the reaction
are never under true equilibrium conditions as the number and
type of potential ligands vying for the sites varies over time.
The results from the kinetics studies indicate that the sites
influencing enantioselectivity, perhaps those of highest affinity,
are not as effectively utilized during the reaction as in binding
studies. Comparable results have been obtained from other
systems,4,30c which may reflect either higher levels of inhibition
of these sites or that the higher affinity sites are less accessible
and that mass transfer becomes a limiting factor. It is arguable
that both factors could contribute to the observed results.
Moreover, and specifically in terms of stereochemical control,
the results may also reflect a lack of steric definition in the
vicinity of the reaction center.
Studies on the influence of the enantiomers of the TSA itself
(4a and 4b) on the reaction were performed to examine the role
of the imprinting sites on the reaction kinetics. However, the
TSA was found to rapidly degrade under the conditions
employed in the reaction assay. It is noteworthy that studies of
TSA under polymerization conditions demonstrated it to be
stable. Furthermore, the presence of the reaction products (3a
or 3b) demonstrated no significant influence on reaction rate.
An alternative strategy was to use dibenzoylmethane (6, DBM)
(Figure 9), which we have previously used as a TSA in related
studies for catalyzing production of chalcone (8).14 Reactions
performed with methanol as solvent yield no product under the
conditions employed; other solvent configurations shall be
utilized in future studies. Although 6 has a benzyl group instead
of the chiral camphor moiety, simple molecular model studies
suggested that it could fit to the volume of 4a and 4b, and
therefore should be able to access sites selective for the original
TSAs. A concentration-dependent competitive inhibition of the
reaction (Vmax) 10.11 ( 0.03 nmol/h; Km ) 126.26 ( 0.96
a (S)-Product (3a) formation based on time course experiments of 160
h.
The time course studies show that the presence of P0 has
effectively no influence on the rate of reaction, as compared to
the solution reaction performed with the same amount of Co2+
present (Figure 8 and Table 3). This implies that the polymer
matrix itself does not induce rate enhancement. However, in
the case of P1, which possesses sites selective for Co2+, a 12-
fold increase in reaction rate was obtained. This is attributed to
the presence of sites capable of binding complexes of Co2+and
substrate, i.e., sites with incomplete coordination of the metal
ion by the pyridinyl residues of the polymer allowing for access
by the substrates. This line of reasoning is supported by the
results obtained with P2, which increases reaction rate by a
factor of 55 relative to the solution and P0 reactions. Assays
run with P3, with sites selective for the (R)-enantiomer of
camphor (1b), were slightly slower suggesting either that the
sites were not as well suited for accommodating the (S)-
substrate, or that a small population of the sites are inaccessible
to 1b because of their high fidelity recognition of the (S)-
configuration of the template. Importantly, reactions performed
with P3 and 1b as substrate demonstrated the same reaction
rate enhancements as observed for reactions with P2 and 1a.
The poor stereochemical control of this reaction reflects the
outcomes of studies by Gagne´30a-c and Severin,4 where lack of
significant steric control around the reaction center may underlie
the stereochemical outcome of the reactions studied. On account
of the enantioselectivities observed in the binding studies,
experiments were also performed (at reflux and room temper-
ature) in MeOH, though no product was observed, most likely
due to competition for sites at the metal ion, and/or quenching
of any carbanion formed. Furthermore, differences between the
gas accessible surface areas of these polymers are minimal,
which allows us to exclude nonspecific surface effects as a basis
for the observed rate enhancements. This is further supported
by swelling studies performed in DMF (see the Supporting
4850 J. Org. Chem., Vol. 71, No. 13, 2006

A Class II Aldolase Mimic
organic phases were dried (MgSO4) and evaporated in vacuo. The
crude yellow oil was recrystallized from EtOH to afford white
mM; Ki ) 30.4 ( 0.35 mM) by DBM (6) was demonstrated
(Figure 10). In the presence of 20 mM of 6, the reaction rate is
reduced to that of the solution reaction. The inhibition is
indicative of the presence of sites selective for DBM which are
necessary for the catalysis of the reaction.
Collectively, the rate enhancing influence of the TSA
imprinted polymers, together with the concentration inhibitory
effect of 6, demonstrates that sites selective for the transition
state analogue are responsible for the catalysis of this otherwise
extremely slow C-C bond forming reaction, with some en-
antioselectivity. Longer studies, 450 h, resulted in a proportional
increase in the amount of product formed, which highlights the
resilience of these materials to harsh environments.
crystals of 3a (0.42 g, 9%). Mp 84-87 °C; [R]20 -369 (c 1.07,
D
acetone); λmax 290.0 (c 40 µM, log ꢀ 4.38, MeOH); IR (KBr) 3024
(CH arom), 2956 (CH), 1720 (CdO), 1648 (CdC); 1H NMR (400
3
MHz, CDCl3, 25 °C) δ 7.50-7.48 (2H, d, J ) 7.3, H3d), 7.42-
7.40 (2H, t, 3J ) 7.3, H3e), 7.38-7.34 (1H, d, 3J ) 7.3, H3f), 7.25
3
(1H, s, H3b), 3.12-3.10 (1H, d, J ) 4.2, H4), 2.22-2.17 (1H, tt,
3
3
3
3J ) 4.2, J ) 11.5, H5′), 1.83-1.76 (1H, dt, J ) 11.5, J ) 2.8,
H6′), 1.64-1.50 (2H, m, H6′′ and H5′′), 1.04 (s, 3H, Me1), 1.01 (s,
3H, Me7′), 0.81 (s, 3H, Me7′′); 13C NMR (63 MHz, CDCl3, 25 °C)
δ 208.7 (CdO), 142.5 (C3a), 136.1 (C3c), 130.2 (C3dH), 129.1
(C3fH), 129.0 (C3eH), 127.9 (C3bH), 57.5 (C7), 49.6 (C4H), 47.1
(C1), 31.1 (C6H2), 26.4 (C5′H2), 21.0 (C7′′H3), 18.7 (C7′H3), 9.7
(C1H3); MS 240 (M+, 100%), 225, 212, 197, 184, 169, 157, 141,
128, 115, 103, 91, 77, 55, 41. Anal. Calcd for C17H20O: C, 84.96;
H, 8.39. Found: C, 85.27; H, 8.47.
Conclusions
The development of new methods for the catalysis of carbon-
carbon bond formation remains one of the great challenges for
synthetic organic chemistry. In this study we have demonstrated
that molecularly imprinted polymers selective for a complex
of Co2+ and a transition state analogue (4) for the aldol reaction
of camphor (1) and benzaldehyde (2) can result in polymeric
materials which increase reaction rate by a factor of over 50.
Importantly, these polymers demonstrate enantioselective rec-
ognition of substrate and turnover. This study provides the first
example of an enantioselective molecularly imprinted polymer
capable of catalysis of carbon-carbon bond formation.
(1R,4S)-(E)-3-Benzylidene-1,7,7-trimethylbicyclo[2.2.1]heptan-
2-one (3b). The same procedure as for 3a was employed, but with
1b as starting material. CH2Cl2 was used for the extraction of 3b,
which was isolated as white crystals (0.44 g, 9%). Mp 95-97 °C;
[R]20 +412 (c 1.00, acetone); λmax 289.0 (c 40 µM, log ꢀ 4.30,
D
MeOH); IR (KBr) 3026 (CH arom), 2953 (CH), 1723 (CdO), 1650
1
(CdC); H NMR (400 MHz, CDCl3, 25 °C) δ 7.50-7.48 (2H, d,
3
3J ) 7.3, H3d), 7.42-7.39 (2H, t, J ) 7.4, H3e), 7.36-7.34 (1H,
3
3
d, J ) 7.2, H3f), 7.25 (1H, s, H3b), 3.13-3.12 (1H, d, J ) 4.2,
H4), 2.24-2.16 (1H, tt, 3J ) 4.5, 3J ) 11.5, 5′), 1.83-1.76 (1H, dt,
3J ) 12.1, 3J ) 3.0, H6′), 1.64-1.50 (2H, m, H6′′ and H5′′), 1.04 (s,
3H, Me1), 1.01 (s, 3H, Me7′), 0.81 (s, 3H, Me7′′); 13C NMR (63
MHz, CDCl3, 25 °C) δ 208.7 (CdO), 142.5 (C3a), 136.1 (C3c), 130.2
(C3dH), 129.1 (C3fH), 129.0 (C3eH), 127.9 (C3bH), 57.5 (C7), 49.6
(C4H), 47.1 (C1), 31.1 (C6H2), 26.4 (C5′H2), 21.0 (C7′′H3), 18.7
(C7′H3), 9.7 (C1H3); MS 240 (M+, 100%), 225, 212, 197, 184, 169,
157, 141, 128, 115, 103, 91, 77, 55, 41. Anal. Calcd for C17H20O:
C, 84.96; H, 8.39. Found: C, 85.05; H, 8.30.
Experimental Section:
General. All reactions were performed under inert atmosphere.
Benzaldehyde was freshly distilled before use. Benzoyl chloride
was distilled from Ca and THF was dried over Na/benzophenone.
MeOH was dried over I2/Mg and freshly distilled prior to use.
Divinylbenzene (DVB) was extracted three times with a solution
of NaOH (0.1 M), dried over MgSO4, filtered, and passed through
basic Al2O3 before use. Azobis(cyclohexanecarbonitrile) (ABCC)
was recrystallized from MeOH. Anhydrous DMSO (99.9%),
anhydrous DME (99.5%), (R)-camphor (98%), (S)-camphor (99%),
ethyl benzoate (99%), sodium hydride (95%), styrene (99%),
4-vinylpyridine (95%), n-BuLi (2.5 M in toluene), and Co(OAc)2‚
4H2O were used as received.
(1S,3S,4S)-3-Benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-
one (4a). A solution of (S)-camphor 1a (2.00 g, 13.1 mmol)
dissolved in DME (12 mL) was added to a suspension of NaH
(1.13 g, 47.3 mmol) in DME (18 mL). The mixture was refluxed
for 1 h, whereupon ethyl benzoate (2.17 g, 14.6 mmol) dissolved
in 12 mL of DME was added to the reaction mixture under reflux.
After being stirred at reflux temperature overnight, the reaction was
quenched by addition of 10 mL of EtOH (95%). The mixture was
poured onto 60 mL of water and acidified with HCl until pH 1.
The aqueous phase was extracted with pentane (3 × 75 mL). The
combined organic phases were washed with an aqueous solution
of NaHCO3 (5%, 75 mL) and brine (75 mL). After drying of the
organic phase over MgSO4 and evaporation of the solvents, the
yellow crude crystals were recrystallized from pentane to give 4a
as pale yellow crystals (1.38 g, 42%). Mp 65-67 °C; [R]20D -268
(c 0.99, CH2Cl2); λmax 309.4 (c 80 µM, log ꢀ 4.38); IR (KBr) 3200-
2600 (br OH), 3051 (CH arom), 2968 (CH), 1663 (CdC), 1617
(CdO, â-diketone/enol); 1H NMR (250 MHz, CDCl3, 25 °C) (both
diketo and keto-enol forms) δ 8.63 (0.3H, br s, OH-enol), 7.68-
7.64 (2H, m, H arom), 7.43-7.42 (3H, m, H arom), 2.85-2.83
(0.7H, d, 3J ) 3.8, OCCHCO), 2.22-2.11 (1H, m, CH, CHC(CH3)2),
1.83-1.74 (1H, m, CH), 1.67-1.48 (3H, m, CH2 and CH), 1.02
(3H, s, CH3), 0.94 (3H, s, CH3), 0.82 (3H, s, CH3); 13C NMR (66
MHz, CDCl3, 25 °C) (diketo and keto-enol forms) δ 213.2, 212.8,
210.6, 197.2, 193.3 (all CdO and CdC(OH)keto-enol 2), 161.8
1H and 13C NMR spectra were recorded at 500, 400, 270, or
250 MHz and 125, 100, 68, or 63 MHz, respectively. CDCl3 and
C6D6 were used as solvents, and the signals of the solvents served
as internal standards. Signals of methyl, methylene, and quaternary
carbon atoms were distinguished by DEPT experiments. Homo-
nuclear 1H connectivities were determined by using COSY experi-
1
ments. Heteronuclear H-13C connectivities were determined by
using HSQC and HMBC experiments. Absolute configurations were
resolved by NOESY experiments. Chemical shifts (δ) are reported
in ppm and J values are presented in hertz. Mass spectra of positive
ions obtained by electron impact (EI, 70 eV) were measured with
an Agilent 6890 GC-system with a Agilent 5973 MS detector. FT-
IR spectra were recorded with samples dispersed in KBr on a
Nicolette Avatar FT-IR spectrophotometer by diffuse reflectance
IR spectroscopy. VIS studies were performed on a Hitachi U2000
spectrophotometer. The data analyses were conducted with the
software package Prism (version 3.03, GraphPad Software, USA).
(1S,4R)-(E)-3-Benzylidene-1,7,7-trimethylbicyclo[2.2.1]heptan-
2-one (3a). To a cold (ice bath) solution of n-BuLi (2.5 M in
toluene, 11 mL, 27.58 mmol) dissolved in DMSO (10 mL) was
added dropwise a solution of (S)-camphor 1a (3.00 g, 19.70 mmol)
and benzaldehyde 2 (2.20 mL, 21.67 mmol) in DMSO (15 mL).
The reaction mixture was stirred at room temperature overnight,
then poured into ice water (250 mL) containing 10 mL of HOAc.
The resulting yellow oil was extracted with Et2O. The combined
(CdC(OH)
1), 136.4, 134.1 (both Cq arom), 133.4, 133.1,
keto-enol
130.3, 129.9, 128.7, 128.3, 128.1, 127.8 (all CH arom), 115.4
(CdC(OH) ), 63.8, 58.8 (both CH), 57.7, 57.6, 50.0 (both
keto-enol 2
Cq), 48.6, 48.4 (both CH), 46.4, 46.3 (both Cq), 45.2 (CH), 30.6,
30.2, 28.9, 27.9, 27.1, 22.1 (all CH2), 21.6, 20.3, 19.7, 19.6, 18.9,
18.8, 9.6, 8.8 (all CH3); MS 256 (M+), 241, 228, 213, 196, 185,
171, 147, 135, 123, 105 (100%), 91, 77, 55, 41. Anal. Calcd for
C17H20O2: C, 79.65; H, 7.86. Found: C, 80.10; H, 7.96.
J. Org. Chem, Vol. 71, No. 13, 2006 4851

Hedin-Dahlstro¨m et al.
(1R,3R,4R)-3-Benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-
one (4b). The same procedure as for 4a was employed, with the
(R)-camphor 1b as starting material. The product, 4b, was isolated
as pale yellow crystals (2.91 g 58%). Mp 84-86 °C; [R]20D +277
(c 1.00, CHCl3); λmax 306.0 (c 80 µM, log ꢀ 4.08, MeOH); IR (KBr)
3200-2600 (br s, OH), 3057 (CH arom), 2959 (CH), 1669 (Cd
C), 1607 (CdO, â-diketone/enol); 1H NMR (250 MHz, CDCl3, 25
°C) (both diketo and keto-enol forms) δ 8.64 (0.15H, br s, OH-
enol), 7.69-7.65 (2H, m, H arom), 7.45-7.43 (3H, m, H arom),
change in absorbance at 520 nm followed by fitting the data to a
one-site binding model. The goodness of fit (R2) was 0.9880 or
better in all cases.
Polymer Synthesis. 4-Vinylpyridine (430 µL, 4.0 mmol), styrene
(4580 µL, 40.0 mmol), and divinylbenzene (5690 µL, 40.0 mmol)
were mixed with 4a or 4b (512.7 mg, 2.0 mmol), azobis-
(cyclohexanecarbonitrile) (ABCC) (293.2 mg, 1.2 mmol), and Co-
(OAc)4‚4H2O (498.2 mg, 2.0 mmol) in MeOH (14.98 mL), and
briefly sonicated. The mixture was degassed by repeated freeze-
thaw cycles (three times) and after the last cycle left under vacuum.
Polymerization was carried out at 55 °C (36 h) to obtain polymers
P2 (4a) and P3 (4b). The bulk polymers were ground and sieved
through a 63 µm sieve and then wet sieved (acetone) through a 25
µm sieve. Particles in the range of 63-25 µm were collected. The
fine particles were removed by repeated sedimentation from acetone
3
2.85-2.84 (0.87H, d, J ) 3.8, OCCHCO), 2.22-2.11 (1H, m,
CH, CHC(CH3)2), 1.83-1.74 (1H, m, CH), 1.67-1.49 (3H, m, CH2
and CH), 1.03 (3H, s, CH3), 0.94 (3H, s, CH3), 0.83 (3H, s, CH3);
13C NMR (66 MHz, CDCl3, 25 °C) (diketo and keto-enol forms)
δ 213.2, 212.9, 210.7, 197.2, 193.3 (all CdO and CdC(OH)
keto-enol
2), 161.8 (CdC(OH)keto-enol 1), 136.5, 134.1 (both Cq arom), 133.4,
133.1, 130.3, 129.9, 128.7, 128.3, 128.1, 127.8 (all CH arom), 115.4
(CdC(OH)keto-enol 1), 63.8 (Cq), 58.8 (CH), 57.7, 50.1 (both Cq),
48.6, 48.4 (both CH), 46.4, 46.3 (both Cq), 45.2 (CH), 30.6, 30.2,
28.9, 27.9, 27.1, 22.1 (all CH2), 21.6, 20.3, 19.7, 19.6, 19.0, 18.8,
9.7, 8.9 (all CH3); MS 256 (M+), 241, 228, 213, 196, 185, 171,
147, 135, 123, 105 (100%), 91, 77, 55, 41. Anal. Calcd for
C17H20O2: C, 79.65; H, 7.86. Found: C, 79.45; H, 8.00.
(6 × 400 mL). The print molecule complex (4a-Co2+ and 4b-Co2+
,
respectively) was removed by packing the polymer (4 g) in an
HPLC column and washing with acetic acid/MeOH 7:3 (400 mL),
MeOH (100 mL), 45 mM Na2-EDTA in MeOH/water (400 mL),
MeOH (50 mL), and acetone (100 mL). Two reference polymers
were also synthesized as described above, P0 (absence of 4 and
Co2+) and P1 (absence of 4).
Attempted Synthesis of (1S,3R,4S)-3-Benzoyl-1,7,7-trimethylbi-
cyclo[2.2.1]heptan-2-one (5a). A solution of SmI2 in THF (0.1
M) was prepared by adding THF (100 mL) to Sm (1.80 g, 12 mmol)
and I2 (2.54 g, 10 mmol) and stirring the reaction mixture vigorously
at 22 °C overnight. The color of the reaction changed from brown
to green and then to Prussian blue. Then, (S)-bromocamphor 7a
(1.15 g, 5 mmol) and benzoyl chloride (0.70 g, 5 mmol) were
dissolved in THF (10 mL) and the solution was added slowly at 0
°C to the solution of SmI2 in THF. The resulting brownish mixture
was stirred at room temperature overnight. The solvent was
evaporated and the residue was hydrolyzed with HCl (10 mL, 15%).
The aqueous phase was extracted 3 times with Et2O. The combined
organic phases were dried over MgSO4 and evaporated to give a
Polymer Titrations. To duplicate samples of blank polymer (P0)
and (R)-MIP (P3) (1 to 20 ( 0.05 mg) solutions of 0.1 or 0.015
mM of 4b:Co2+ (1:1) in MeOH were added and the samples were
incubated at room temperature for 19 h. The samples were filtered
trough 13 mm syringe filters with 0.2 µm PTFE membranes and
analyzed on a Kromasil C18 column (5 µm 150 mm × 4.6 mm) at
295 nm on a HP 1050 HPLC with the mobile phase MeOH/water
(9:1) and the flow 1.0 mL/min.
Batch Binding Studies. On the basis of the polymer titration
results, batch binding studies were performed in MeOH or DMF,
using 20 mg of polymer (P0, P1, P2, and P3) and various ligands
(2, 3a, 3b, 4a, and 4b), 0.015 mM. All samples were incubated for
19 h at room temperature. Determinations of bound ligand were
performed as described above. All studies were performed in at
least duplicate, with duplicate analysis of all points.
Reaction Assays. Polymer assays were performed according to
Matsui et al.14 with minor modifications. Polymer samples (P0,
P1, P2, and P3) were incubated at room temperature for 19 h with
Co(OAc)4‚4H2O (1 mg/100 mg of polymer) in MeOH (0.5 mL).
The samples were filtered and the concentration of bound Co2+
was established by analysis of the residual Co2+ present in the
filtrate by quantitative spectrophotometric analysis (520 nm). The
polymers were then dried under vacuum overnight at room
temperature. Cobalt treated polymer samples (200 mg) were
incubated with 1a or 1b (200 µmol) and 2 (200 µmol) in dry DMF
(1.0 mL). Solution reactions were carried out as above with pyridine
(10 µL) and Co(OAc)4‚4H2O (8 µmol). The reactions were
performed in sealed tubes at 100 °C in a thermostated oil bath.
Samples (10 µL) were taken directly from the reaction mixtures
and diluted 100-fold before filtration and analysis by HPLC, using
a Kromasil C18 5 µm 150 mm × 4.6 mm column at 295 nm. HPLC
analyses were run isocratically with MeOH/water (9:1) as the mobile
phase at 1.0 mL/min. Standard curves of concentration versus peak
area were prepared in triplicate over the concentration ranges used
in the assay for calculation of the product yield.
1
brown oil containing 4a and 5a in a ratio of 2:1. Partial H NMR
spectrum of 5a (250 MHz, CDCl3, 25 °C) δ 4.25-4.22 (1H, dd, 3J
3
) 1.3, J ) 4.3, OCCHCO). Purification of the crude product by
flash chromatography on silica gel (eluent: Et2O/cyclohexane 1:6,
triethylamine 1%) gave exclusively 4a.
Attempted Synthesis of (1R,4S,4R)-3-Benzoyl-1,7,7-trimethyl-
bicyclo[2.2.1]heptan-2-one (5b). The same procedure as for 5a
was employed but with the (R)-bromocamphor 7b as starting
material. The crude product was also isolated as a brown oil
containing 4b and 5b. Partial 1H NMR spectrum of 5b (250 MHz,
CDCl3, 25 °C) δ 4.27-4.25 (1H, dd, 3J ) 1.2, 3J ) 4.8, OCCHCO).
The crude product was purified by flash chromatography on neutral
alumina (eluent: Et2O/cyclohexane 1:6) to exclusively give 4b.
NMR Titrations. A solution of 4a (10 mM) and pyridine-d5
(20 mM) in CD3OD was titrated by consecutive additions of a
solution containing Co(OAc)4‚4H2O (40 mM), 4a (10 mM), and
1
pyridine-d5 (20 mM) in CD3OD. H NMR spectra were recorded
at 250 MHz at 298 K. CD3OD (99.8%), pyridine-d5 (99%), and
CDCl3 (99.9%) were used as solvents. Apparent dissociation
constants were calculated with nonlinear line fitting to a one-site
model where each regression was based on no less than 8 data
points and results are presented with the standard error. The
goodness of fit (R2) was 0.9898 or better in all cases.
Inhibition Studies. Samples were prepared in triplicate with Co-
(OAc)4‚4H2O treated polymer (P2) (200 mg) incubated with 1a
(100 µmol) and 2 (50 to 400 µmol) in dry DMF (0.5 mL). As
controls, solution reactions were carried out as described above.
Inhibition was performed by addition of 0, 5, and 20 mM of 6.
The reactions were performed in sealed tubes at 100 °C in a
thermostated oil bath. Samples (10 µL) were taken directly from
the reaction mixtures and diluted 100-fold before filtration and
analysis by HPLC, using a Kromasil C18 5 µm 150 mm × 4.6
mm column at 295 nm. HPLC analyses were run isocratically with
MeOH/water (9:1) as the mobile phase at 1.0 mL/min.
VIS Titrations. Formation of prepolymerization complexes was
studied by titrating a solution of Co(OAc)2‚4H2O (20 mM) in
MeOH containing 40 mM pyridine with a solution of 4b (80 mM)
in MeOH containing 40 mM pyridine. The effect of the different
components on complexation strength was elucidated by titrating
a solution of Co(OAc)4‚4H2O (10 or 5 mM) in MeOH with a
solution of 4b (40 mM) or pyridine (5000 mM) in MeOH. Job’s
method of continuous variation was employed for determining the
stoichiometric relationship between Co2+ and 4b in MeOH. The
change in absorbance was recorded at 400-700 nm and apparent
dissociation constants (app Kdiss) were calculated by plotting the
4852 J. Org. Chem., Vol. 71, No. 13, 2006

A Class II Aldolase Mimic
Acknowledgment. We thank Dr. Jesper G. Karlsson (Uni-
versity of Kalmar, Sweden), Hannu Luukinen (University of
Oulu, Finland), and Dr. Mats Malmberg (Synthelec AB,
Sweden) for assistance with NMR measurements. We also thank
Dr. Håkan S. Andersson (University of Kalmar, Sweden) and
Dr. Michael J. Whitcombe (Cranfield University, UK) for
fruitful discussions. The financial support of the Swedish
Research Council (VR), National Research School in Pharma-
ceutical Sciences (Fla¨k), Swedish Knowledge Foundation
(KKS), and the University of Kalmar is most gratefully
acknowledged.
Supporting Information Available: Spectroscopic data (NMR)
for synthesis products, proposed mechanism for keto-enol tau-
tomerism, additional spectroscopic titration data, additional binding
study data, and polymer swelling studies. This material is available
free of charge via the Internet at http://pubs.acs.org.
JO060608B
J. Org. Chem, Vol. 71, No. 13, 2006 4853