Improving the Strategy and Performance of Molecularly Imprinted Polymers Using Cross-Linking Functional Monomers

Article abstract of DOI:10.1021/jo0352225

A new strategy for monomer design has been investigated that combines interactive monomer functionality with a cross-linking format, giving as a result noncovalent molecularly imprinted polymers (MIPs) with improved performance. This strategy was explored under the premise that more functionality could be introduced without suffering performance losses due to reduced cross-linking. While this proved to be correct, equally important contributions to selectivity enhancement at the molecular level by conformation control and diastereomeric complexation were also discovered. Monomers derived from L-serine and L-aspartic acid were synthesized and used to prepare MIPs, with the best performance obtained for the MIP formulated with the serine-based cross-linker (N,O-bis-methacryloyl L-serine, 3), versus the aspartic-acid-based cross-linkers and the traditional methacrylic acid/ethylene glycol dimethacrylate (MAA/EGDMA) formulation. Quantitative structure selectivity relationship (QSSR) studies revealed that the improved performance of 3 was due to three key factors: (1) the cross-linking nature of this monomer; (2) control of conformational flexibility; (3) a strong influence of monomer chirality on enantioselectivity in MIPs.

Full text of DOI:10.1021/jo0352225

Im p r ovin g th e Str a tegy a n d P er for m a n ce of Molecu la r ly
Im p r in ted P olym er s Usin g Cr oss-Lin k in g F u n ction a l Mon om er s
Martha Sibrian-Vazquez and David A. Spivak*
Department of Chemistry, Louisiana State University, Baton Rouge, Louisiana 70803
david_spivak@chem.lsu.edu
Received August 20, 2003
A new strategy for monomer design has been investigated that combines interactive monomer
functionality with a cross-linking format, giving as a result noncovalent molecularly imprinted
polymers (MIPs) with improved performance. This strategy was explored under the premise that
more functionality could be introduced without suffering performance losses due to reduced cross-
linking. While this proved to be correct, equally important contributions to selectivity enhancement
at the molecular level by conformation control and diastereomeric complexation were also discovered.
Monomers derived from L-serine and L-aspartic acid were synthesized and used to prepare MIPs,
with the best performance obtained for the MIP formulated with the serine-based cross-linker (N,O-
bis-methacryloyl L-serine, 3), versus the aspartic-acid-based cross-linkers and the traditional
methacrylic acid/ethylene glycol dimethacrylate (MAA/EGDMA) formulation. Quantitative structure-
selectivity relationship (QSSR) studies revealed that the improved performance of 3 was due to
three key factors: (1) the cross-linking nature of this monomer; (2) control of conformational
flexibility; (3) a strong influence of monomer chirality on enantioselectivity in MIPs.
In tr od u ction
SCHEME 1
Molecular imprinting is a maturing method of produc-
ing polymeric materials with specific recognition sites
tailored for targeted compounds.1 These have found use
as chromatographic and solid-phase extraction media,
and offer tremendous potential as the recognition element
for immunoassays and sensors.4 However, advanced
applications of MIPs such as in microanalytical systems
and microsensors will depend on the development of
higher performance MIP materials. Although the practice
of molecular imprinting is well established, new strate-
gies for the rational design of new and better materials
based on the underlying mechanisms will continue to
improve the process. This paper presents an important
example of this rational approach to improving the
performance of molecularly imprinted polymers (MIPs),
and the important discoveries that will guide the design
of monomers in the future.
-5
,5
complex (PPC), is followed by copolymerization with
cross-linking monomers to form a polymer network
around the template molecule. After removal of the
template, the resulting polymer is postulated to contain
binding cavities that are complementary in shape and
functionality to the template molecule. A seminal study
by Wulff and co-workers illustrated an important fact
that increasing the amount of cross-linking matrix
increases the specific recognition by covalently imprinted
An outline of the generally accepted mechanism for
MIP binding site formation is illustrated in Scheme 1.
Pre-organization of functional monomers in the vicinity
of the template molecule, referred to as the pre-polymer
6
polymers. However, imprinted polymers formed using
*
To whom correspondence should be addressed. Phone: (225) 578
a noncovalent strategy rely on the solution concentration
of functional monomer to form the PPC. By Le Chatlier’s
principle, increasing concentration of functional monomer
should increase the concentration of PPC, which is in
turn postulated to play a large role in the formation of
specific sites in the polymer.
2
868. Fax: (225) 578 3458.
(
(
(
(
1) Wulff, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812.
2) Mosbach, K. Trends Biochem. Sci. 1994, 19, 9.
3) Shea, K. J . Trends Polym. Sci. 1994, 2, 166.
4) (a) Molecular and Ionic Recognition with Imprinted Polymers;
Bartsch, R. A., Maeda, M., Eds.; ACS Symposium Series; American
Chemical Society: Washington, DC, 1998; Vol. 703. (b) Molecularly
Imprinted PolymerssMan made Mimics of Antibodies and Their
Application in Analytical Chemistry; Sellergren, B., Ed.; Elsevier:
Amsterdam, 2001.
(6) Wulff, G.; Vietmeier, J .; Poll, H. G. Makromol. Chem. 1987, 188,
(5) Haupt, K.; Mosbach, K. Chem. Rev. 2000, 100, 2504.
731.
1
0.1021/jo0352225 CCC: $25.00 © 2003 American Chemical Society
Published on Web 11/13/2003
9
604
J . Org. Chem. 2003, 68, 9604-9611

Molecularly Imprinted Polymers
binding the template molecule. Nontemplated polymers
showed no chiral separation, even though the polymers
had a chiral backbone. An interesting cross-linker has
been synthesized, 2,6-bis(acrylamido)pyridine (BAAP),
which uses a donor-acceptor-donor motif for forming
hydrogen bonds to a template, that mimics interactions
found in DNA.9 This example showed that the cross-
linker can also serve as the complexing functionality
needed to form the pre-polymer complex, and provide
affinity for the template upon rebinding.
Mon om er a n d Cr oss-Lin k er Syn th eses. Compound
1
was synthesized following the protocol reported in the
1
0
literature, while syntheses of compounds 2 and 3 were
developed for this project. The syntheses of monomers 2
and 3 were constructed from commercially available
protected amino acids as the starting materials for the
corresponding compounds. The first step, acylation of the
amine and hydroxyl functionalities for introducing the
methacrylamide and methacrylate polymerizable groups
was carried out using traditional esterification proce-
dures. Introduction of the vinyl ketone polymerizable
group in 2 was accomplished by the use of a Weinreb
amide,11 which also uniquely served as a protecting group
for the carboxylate side chain as well as a latent elec-
trophile for the nucleophilic addition of isopropenylmag-
nesium bromide in the penultimate step. Deprotection
of the amine group in the aspartic acid derivative was
F IGURE 1. Cross-linking functional monomers and non-
cross-linking controls.
One problem with increasing the functional monomer
is that the amount of cross-linker becomes too low, and
the MIP loses its recognition properties from random
7
motion of non-cross-linked polymer domains. Thus, a
balance must be struck for optimum imprinting condi-
tions which never allows the optimum concentration of
cross-linker or functional monomer. In an effort to design
the most effective monomers for molecular imprinting,
we decided to combine the interactive monomer func-
tionality with the cross-linking requirements, thus elimi-
nating the need to limit the ratio cross-linking monomer
in favor of functional group monomers and vice versa.
While this strategy proved successful, quantitative struc-
ture-selectivity relationship (QSSR) experiments also
elucidated two other important design principles that will
guide future designs of monomers for molecular imprint-
ing.
2
achieved by acidic hydrolysis with HCl/Et O, which
provided a cleaner product versus TFA treatment. Final
deprotection of the R-carboxylic group of 8 was unsuc-
cessful using chemical procedures such as nucleophilic
dealkylation12 or basic hydrolysis, which hydrolyzed
both esters rather than the desired methyl ester saponi-
13
fication. Therefore, an enzymatic approach for this step
was investigated.1
4-16
Three enzymes were screened for
selective hydrolysis of the methyl ester: porcine pancre-
atic lipase, porcine liver esterase, and carbonic anhydrase
(bovine erythrocytes). In the case of monomer 3, porcine
pancreas lipase (PPL) gave the highest yield of the
product at 62% yield, while in the case of the aspartic
acid derivative 2, porcine liver esterase gave the highest
yield (55%). The overall syntheses for compounds 2 and
Resu lts a n d Discu ssion
For this study, functional cross-linkers incorporating
carboxylic acids were developed for comparison to the
most widely used MIP formulation, which utilizes the
combination of methacrylic acid (MAA) as a functional
monomer with ethylene glycol dimethacrylate (EGDMA)
as the cross-linker. We designed monomers 1 - 3 based
on amino acids incorporating carboxylate functionality,
while providing end groups that could be transformed
into cross-linking polymerizable groups. As non-crosslink-
ing controls, monomers 4-6 were included which incor-
porate the same interactive functionality to determine
the contributions the cross-linking in functional mono-
mers might have toward improved binding (Figure 1).
Cross-linking monomers employing both natural and
nonnatural amino acids, have previously been developed
for molecular imprinting. Phenylalanine-based N,O-
bisacryloyl-L-phenylalaninol and a similar cross-linker
N,O-dimethacryloyl phenylglycinol, were synthesized and
3
are shown in Schemes 2 and 3. Last, syntheses of non-
cross-linking functional monomers 5 and 6 as shown in
Schemes 4 and 5 were performed using known proce-
dures.1
7-19
(9) (a) Tanabe, K.; Takeuchi, T.; Matsui, J .; Ikebukuro, K.; Yano,
K.; Karube, I. J . Chem. Soc., Chem. Commun. 1995, 2303. (b) Yano,
K.; Tanabe, K.; Takeuchi, T.; Matsui, J .; Ikebukuro, K.; Karube, I. Anal.
Chim. Acta 1998, 363, 111.
(10) Spivak, D. A.; Sibrian-Vazquez, M. Bioseparation 2002, 10, 331.
(
(
11) Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.
12) (a) Haslam, E. Tetrahedron 1980, 36, 2409. (b) Dean, P. D. J .
Chem. Soc. 1965, 6655. (c) Laganis, E. D.; Chenard, B. L. Tetrahedron.
Lett. 1984, 25, 5831.
(
13) Corey, E. J .; Szekely, I.; Shiner, C. S. Tetrahedron Lett. 1977,
4
0, 3529.
(14) Ohno, M.; Ito, Y.; Arita, M.; Shibata, T.; Adachi, K.; Sawai, H.
Tetrahedron 1984, 40, 145.
15) Chenevert, R.; Belrhlid, R.; Letourneau, M.; Gagnon, R.;
(
D’Astous, L. Tetrahedron: Asymmetry 1993, 4, 1137.
(16) Wild, H. J . Org. Chem. 1994, 59, 2748.
8
tested for chiral separations. Use of these cross-linking
monomers for MIPs showed modest enantioselectivity for
(
17) Heilmann, S. M.; Smith, H. K., II. J . Appl. Polym. Sci. 1979,
4, 1551.
18) Nakaya, T.; Nishio, K.; Memita, M.; Imoto, M. Makromol. Chem.
2
(
(
7) Sellergren, B. Makromol. Chem. 1989, 190, 2703.
Rapid Commun. 1993, 14, 77.
(19) Curci, M.; Mieloszynski, J . L.; Paquer, D. Org. Prep. Proced.
Int. 1993, 25 (6), 649.
(8) Hosoya, K.; Yoshizako, K.; Shirasu, Y.; Kimata, K.; Araki, T.;
Tanaka, N.; Haginaka, J . J . Chromatogr. A, 1996, 728, 139.
J . Org. Chem, Vol. 68, No. 25, 2003 9605

Sibrian-Vazquez and Spivak
SCHEME 2a
into coarse particles and the template removed by Soxhlet
extraction with methanol. Subsequently, the polymers
were ground by mortar and pestle and the fraction
between 20 and 25 µm collected and packed into stainless
steel columns for evaluation using HPLC methods. HPLC
retention values were normalized to capacity factors (k′)
for elution of both enantiomers, and the enantioselectivity
was evaluated by comparison of separation factors, R (R
a
Reaction conditions: (a) MAA/Et3N/DMAP/DCC/CH2Cl2, rt/5
)
k′
S R
/k′ ).
d; (b) PPL (EC 3.1.1.3)/pH ) 7.5, rt/72 h.
Table 1 shows the functional cross-linkers and their
SCHEME 3a
associated MIP separation factors, which are ordered by
increasing R values. The results reveal improved reten-
tion on the MIP columns (i.e., larger k′ values) made
using each of the new cross-linkers, versus the tradition-
ally formulated MIP incorporating MAA as the functional
monomer. However, improved selectivity (i.e. a larger R
value) is observed only for the MIP using the cross-
linking functional monomer 3. To test the hypothesis that
combining fuctionality and cross-linking was a factor
responsible for the improved behavior by the 3, MIPs
formulated with different ratios of monomer/cross-linking
monomer (3/EGDMA) and S-nicotine as the template
were synthesized and their performance compared to
similar MIPs formulated with MAA.
The results shown in Table 2 (and Figure 2), indicate
that the MIPs formulated with 3 (entries 3, 5, and 7)
show improved binding affinity and selectivity as com-
pared to those formulated with MAA (entries 2, 4, and
6). The effect on selectivity is most clearly seen in Figure
2 where both compounds reach their maximum selectivity
initially at 10% functional monomer, after which 3
remains constant up to 25 mol % then decreases at a
higher ratios of 3/EGDMA. In contrast, MIPs formulated
with MAA immediately lose selectivity after reaching the
maximum at 10 mol % MAA. These results show that
noncovalent imprinting is in agreement with covalent
a
Reaction conditions: (a) NMM/i-BuCO2Cl/HCl‚HN(CH3)(OCH3)/
THF, -15 °C/1 h, rt/24 h; (b) HCl/Et2O, 0 °C/6 h, rt/18 h; (c)
H2CdC(CH3)COCl/Et3N/CH2Cl2, rt/48 h; (d) C3H5MgBr/THF, -15
C/30 min, rt/30 min; (e) PLE (EC 3.1.1.1), pH ) 8, rt/72 h.
°
SCHEME 4a
6
imprinting studies published by Wulff, establishing that
the selectivity increases with increase in the degree of
cross-linking.
a
Reaction conditions: (a) H2CdC(CH3)COCl/NaOH, rt/48 h.
Str u ctu r a l Deter m in a n ts for En h a n ced Bin d in g
by F u n ction a l Cr oss-Lin k er 3. The improved perfor-
mance by the functional cross-linker 3 over the aspartic-
acid-based monomers prompted an investigation into the
structural determinants responsible for the good binding
and selective behavior. A quantitative structure-selectiv-
ity relationship (QSSR) study was carried out by syn-
thesizing analogues of 3 to determine key elements of 3
structure leading to selective MIPs. Thus, substructures
of 3, shown in Figure 3 along with 3, were synthesized
to determine important interactions with the template.
The polymers were imprinted as before using S-nicotine
as the template with DMF as the porogen, which was
needed in order to solubilize monomer 5. Binding studies
were carried out by HPLC for imprinted and nonim-
printed polymers formulated with monomers in Figure
SCHEME 5a
a
Reaction conditions: (a) H2CdC(CH3)COCl/Et3N, 50 °C/2 h;
(b) Cr2O3/H2SO4/acetone, reflux.
Im p r in tin g a n d An a lysis of P olym er s. The best
probe for MIP selectivity due to the “imprinting effect”
is chiral differentiation, since all physical properties of
enantiomers are the same except for the three-dimen-
sional positioning of atoms in space. Thus, for our
investigation we imprinted polymers using monomers 1
-
4 as the interactive functional component (10%), and
EGDMA for the remainder of the cross-linked matrix
90%). The S enantiomer of nicotine was chosen as a
3
versus MAA. The polymers were evaluated for enan-
(
tioselectivity using the mobile phase acetonitrile/HOAc
readily available template with one chiral center for
investigating enantioselectivity, two amine functional
groups that provide binding interactions with the func-
tional monomers, and a UV chromophore for HPLC
detection. Polymerization was photochemically initiated
at room temperature and allowed to proceed for 10 h.
Once synthesized, the imprinted polymers were broken
9
8/2, which gave the best results in all cases.
The results of the binding studies in Table 3 show that
the non-crosslinking monomers as a whole exhibit poor
selectivity, even if binding (k′) values were strong.
Therefore, cross-linking provides a structural design
feature that improves selective molecular recognition by
9
606 J . Org. Chem., Vol. 68, No. 25, 2003

Molecularly Imprinted Polymers
TABLE 1. Bin d in g Stu d ies for MIP s Im p r in ted w ith (S)-Nicotin e a n d F or m u la ted Usin g th e New Cr oss-Lin k in g
a
F u n ction a l Mon om er s
a
HPLC conditions: particle size, 20-25 µm; column size, 75 mm length, 2.1 mm id.; mobile phase ) MeCN/HOAc 97/3; flow rate ) 1
mL/min; analytes: 1 mM (S)-nicotine, 1 mM (R)-nicotine, and acetone (used to determine dead volume); wavelength, 262 nm; injected
volume, 5 µL.
TABLE 2. Ch r om a togr a p h ic Resu lts for MIP s Im p r in ted w ith (S)-Nicotin e a n d F or m u la ted w ith Differ en t F u n ction a l
a
Mon om er /Cr oss-Lin k in g Mon om er Ra tios
functional
monomer
functional monomer/
cross-linker monomer ratio
entry
tS (avg) (min)
tR (avg) (min)
k′S
k′R
R
1
2
3
4
5
6
7
noneb
0
0.63
0.68
15.9
1.78
122.5
0.66
0.62
0.58
8.60
1.35
64.67
0.61
0.57
0.70
41.97
3.45
339.28
0.57
59.38
0.55
0.45
22.24
2.40
178.64
0.45
38.89
1.03
1.56
1.89
1.44
1.89
1.26
1.53
c
MAA
1/9
1/9
1/3
1/3
1/1
1/1
c
3
d
MAA
3
d
e
MAA
e
3
23.55
15.56
a
HPLC conditions: particle size, 20-25 µm; column size, 100 mm length, 2.1 mm i.d.; flow rate, 1 mL/min; analytes, 1 mM (S)-
b
nicotine, 1 mM (R)-nicotine, and acetone (used to determine dead volume); wavelength, 262 nm; injected volume, 5 µL. Mobile phase )
MeCN. Mobile phase )MeCN/HOAc 97/3. Mobile phase ) MeCN/HOAc 95/5. e Mobile phase ) MeCN/HOAc 85/15.
c
d
functional groups in MIPs. One reason for this is that
cross-linking tethers the functional group covalently to
the matrix in two places, greatly reducing conformational
entropy of monomer motion. Consequently, this reduces
the conformational flexibility of the pendant functional
group, which creates a more stable interaction with the
template (i.e., reduction in the entropy of association).
However, recall that cross-linking did not help monomers
based on aspartic-acid, which showed lower degrees of
selectivity versus MAA-functionalized MIPs. This would
reason that there is more than one conformational factor
responsible for the improved performance by 3. The other
factor appears to be proximity of the pendant carboxylate
group with respect to the cross-linker backbone. From a
design point of view, it appears that providing closer
proximity of the pendant functional group to the cross-
linking group improves selectivity, again due to reducing
the conformational flexibility of the pendant group. The
effects of binding group flexibility have been shown to
take an active role in the imprinting process by Wulff
and co-workers, who examined the influence of confor-
mational flexibility of functional monomers on selective
behavior by MIPs.21 In this study, it was determined that
(20) Yano, K.; Nakagiri, T.; Takeuchi, T.; Matsui, J .; Ikebukuro, K.;
Karube, I. Anal. Chim. Acta 1997, 91.
(21) Wulff, G.; Gimpel, J . Makromol. Chem. 1982, 183, 2469.
J . Org. Chem, Vol. 68, No. 25, 2003 9607

Sibrian-Vazquez and Spivak
F IGURE 2. Enantioselectivity of MIPs imprinted with (S)-
nicotine as a function of the amount of the functional monomer.
F IGURE 4.
Scop e of Selectivity En h a n cem en t a n d Ch ir a l
In flu en ce by Cr osslin k er 3. The generality of improved
selectivity found for MIPs using the functional cross-
linker 3 was explored by imprinting several other tem-
plates. Six polymers were imprinted using each enanti-
omer of 1-naphthylethylamine (18 and 19), tryptophan
methyl ester (20 and 21), and 1,2-diphenylethylenedi-
amine (22 and 23), shown in Figure 4. Each of these
polymers was compared to a traditional MIP imprinted
with the same templates, but using MAA as the func-
tional monomer. Binding (k′) and selectivity (R) data were
again evaluated and the results are shown in Table 4.
Following the order in Table 4, entries 1 and 2 show
improved enantioselectivity for both enantiomers of
F IGURE 3.
TABLE 3. Ch r om a togr a p h ic Resu lts for MIP s
F or m u la ted w ith 3 a n d F u n ction a l Mon om er An a logu es
a n d Im p r in ted w ith (S)-Nicotin e
a
1
-naphthylethylamine over the traditional EGDMA/MAA
formulated polymers. An interesting observation from
these examples is that the polymer in entry 1 shows
higher enantioselectivity than the polymer in entry 2. It
appears that diastereomeric complexes “imprint” differ-
ently than complexes formed via nonstereogenic mono-
mers.15 The origin of this effect is either that one
diastereomeric complex is preferred in the pre-polymer
complex solution (i.e., chiral matching), or one of the
diastereomeric complexes has properties that affords a
structurally more defined binding site in the MIP.
Preferential imprinting of one diastereomeric complex
could occur from a more distinguishable geometry (e.g.,
twisted versus spherical), or by a complex having physical
properties more compatible with the polymer (e.g. a more
hydrophobic complex for the hydrophobic polymer).
Different behavior originating from diastereomeric
complexes is seen again for tryptophan methyl ester
imprinted polymers, entries 3 and 4 in Table 4, where
the “L” enantiomer imprints well, while the “D” enan-
tiomer provides MIPs with lower selectivity (even lower
than the MAA functionalized MIP). This observation
indicates that diastereomeric complexation (e.g. a chiral
mismatch of functional monomer and template) effects
can dominate any improvements to selectivity by the use
of a cross-linking functional monomer. Entries 5 and 6
in Table 4 show the results for imprinting enantiomers
a
HPLC conditions: particle size, 20-25 µm; column size, 100
mm length, 2.1 mm i.d.; mobile phase, MeCN/HOAc ) 98/2; flow
rate ) 1 mL/min; analytes, 0.1 mM (S)-nicotine, 0.1 mM (R)-
nicotine, and acetone (used to determine dead volume); wave-
length, 262 nm; injected volume, 5 µL.
decreasing binding group flexibility resulted in greater
selectivity; however, if the binding group becomes too
rigid specificity was diminished.
9
608 J . Org. Chem., Vol. 68, No. 25, 2003

Molecularly Imprinted Polymers
TABLE 4. Ch r om a togr a p h ic Eva lu a tion of MIP s Usin g
3
a s th e F u n ction a l Mon om er a n d Im p r in ted w ith
a
Differ en t Tem p la tes
F IGURE 5. Elution profile for imprinted polymers with
(
1S,2S)-(-)-1,2-diphenylethylendiamine: (a) Imprinted poly-
mer with MAA as the functional monomer, (b) Imprinted
polymer with 3 as the functional monomer. Peak 1: (1R,2R)-
(
-)1,2-diphenylethylendiamine. Peak 2: (1S,2S)-(-)-1,2-diphen-
ylethylendiamine. Condition: mobile phase, acetonitrile/HOAc
0/10, flow rate, 1 mL/min, wavelength detection, 260 nm,
9
injected volume, 5 µL.
improved mass-transfer kinetics as a result of favorable
morphology and properties afforded by acrylamide poly-
mers. The three examples presented show that the chiral
nature of the monomer 3 often improves the performance
of imprinted polymers; although there is no general
trend. Investigation into the underlying mechanism(s) for
diastereoselective effects in MIPs will be reported in due
course.
a
HPLC conditions: particle size, 20-25 µm; column size, 100
mm length, 2.1 mm id.; flow rate, 1 mL/min.; injected volume, 5
µL; wavelength detection, 260 nm for naphthylethylamine, 250
nm for tryptophan methyl ester, and 260 nm for diphenylethyl-
amine.
of 1,2-diphenylethylenediamine; this is a more complex
system that involves two stereocenters on the template
complexed with the chiral functional cross-linker 3. The
functionalized monomer 3 imprinted (1R,2R)-(+)-1,2-
diphenylethylenediamine as the template, showed the
highest enhancement in selectivity of roughly 2.5-fold
over the traditional MAA MIP, while the opposite enan-
tiomer gave results roughly equal in terms of R to the
MAA functionalized MIP for separately injected enan-
tiomers. However, an important observation here is that
when the racemic mixture was analyzed, nearly complete
resolution was still obtained with the polymer imprinted
with the S,S derivative using 3 (Figure 5b), while the
MAA functionalized MIP (Figure 5a) did not afford any
resolution at all; even though the two different MIPs have
comparable R values. In general, polymers incorporating
the methacrylamide moiety provided greater resolution
efficiency with sharper and more well-defined peaks in
the HPLC chromatograms, whereas the methacrylate
derivatives gave broad peaks. This may be due to
Con clu sion s
A new strategy for monomer design for noncovalently
imprinted polymers has been shown to be successful for
improving molecularly imprinted materials. Quantitative
structure-selectivity relationship studies have verified a
key improvement to monomer design in these studies was
to provide the template-interactive functional group in
a cross-linking monomer format. The origins of the
improved selectivity were attributed to two factors of the
cross-linking functional monomers. First, the degree of
cross-linking is maximized without imposing restrictions
on functional group concentrations. Second, covalently
tethering of the functional group to the binding site
matrix reduces conformational entropy that would oth-
erwise interfere with specific binding. Last, a strong
influence of diastereomeric complexes on MIP selectivity
was discovered. However, more studies will be needed
to determine whether the origins of this effect are to be
J . Org. Chem, Vol. 68, No. 25, 2003 9609

Sibrian-Vazquez and Spivak
found in the solution-phase pre-polymer complex or the
final polymeric binding site.
62.5 MHz): δ 171.1, 155.4, 80.2, 62.0, 55.3, 48.1, 37.5, 32.8,
2
1
2
8.6. IR: 3327.24, 2977.50, 1740.88, 1713.49, 1663.40, 1517.88,
+
167.82, 1049.61, 989.21. HRMS (FAB) (M + H ): calcd
91.1478, found 291.1556
Exp er im en ta l Section
L-Asp a r tic Acid â-Meth yl Ester N-Meth oxy-N-m eth yl-
a m id e Hyd r och lor id e (11). Weinreb amide (10, 1.16 g, 4
mmol) was treated with 40 mL of 2 M ethereal HCl. The
temperature was maintained at 0 °C for 6 h before being
allowed to increase to room temperature and stirred for 18 h.
The excess of ethereal HCl was evaporated under a stream of
N2 and then under vacuum. The white solid residue was
filtered, washed with ethyl ether (3 × 20 mL), and dried at
N,O-Bism eth a cr yloyl, L-Ser in e r-Meth yl Ester (8). L-
Serine R-methyl ester hydrochloride (0.467 g, 3 mmol) was
dissolved in DCM (15 mL) and cooled to 0 °C, followed by
3
dropwise addition of Et N (0.607 g, 6 mmol). In another flask
methacrylic acid (0.517 g, 6 mmol) and DMAP (0.0733 g, 0.6
mmol) were dissolved in DCM (30 mL), and the resulting
solution was cooled at 0 °C. To this flask was added the
hydrochloride solution in one portion. After 5 min, DCC (1.238
g, 6 mmol) was added to the cooled solution at 0 °C and stirred
additional 30 min. After this period, the temperature was
allowed to rise to room temperature and the reaction mixture
was stirred 5 days. The DCU was filtered, the organic phase
1
room temperature to give an 89% yield of product. H NMR
[(CD ) SO, 250 MHz]: δ 8.59 (3H, s), 4.38-4.43 (1H, m), 3.72
3
2
(3H, s), 3.61 (3H, s), 3.13 (3H, s), 2.77-2.94 (2H, d, J ) 6.9
Hz). 13C NMR [(CD ) SO, 62.5 MHz]: δ 170.2, 168.3, 62.4, 52.9,
3
2
49.4, 47.4, 35.4. IR: 3416.55, 2951.36, 1730.34, 1666.27,
+
1494.94, 1212.51, 988.08, 616.22. HRMS (FAB) (M + H ):
was extracted with 0.5 M NaHCO
3
(2 × 15 mL), 0.5 M sodium
calcd 191.0954, found 191.1154.
citrate (2 × 15 mL), dried over MgSO
4
, and the solvent was
L-Asp a r tic Acid â-Meth yl Ester N-Meth a cr yloyl-N-
m eth oxy-N-m eth yla m id e (12). N-Methoxy-N-methylamide
hydrochloride (11, 0.910 g, 4 mmol) was suspended in 5 mL of
evaporated under vacuum giving an orange oil. The product
was isolated as a yellow oil by flash chromatography using
1
EtOAc/hexanes 50/50 in 71% yield. H NMR (CDCl
3
, 250
3
DCM, and the mixture was neutralized using Et N until pH
MHz): δ 6.67-6.70 (1H, d, J ) 7.58n Hz), 5.99 (1H, d, J )
7
-8 (a white precipitate was formed in this step). The mixture
0
5
.95 Hz), 5.68 (1H, d, J ) 0.95 Hz), 5.50 (1H, d, J ) 1.58 Hz),
.30 (1H, d, J ) 0.95 Hz), 4.80-4.87 (1H, m), 4.41-4.43 (2H,
3
was cooled to 0 °C and another aliquot of Et N added for a
3
total of 20 mmol of Et N in the reaction mixture, followed by
dd, J ) 4.10, 1.42 Hz), 3.69 (3H, s), 1.88 (3H, s), 1.82 (3H, s).
dropwise addition of methacryloyl chloride (1.045 g, 10 mmol).
The mixture was stirred at 0 °C for 30 min, and then the
temperature was allowed to rise to room temperature and
stirring continued for 48 h. The reaction mixture was filtered
1
3
C NMR (CDCl , 62.5 MHz): δ 170.4, 168.3, 167.2, 139.6,
3
1
2
9
35.9, 126.8, 120.9, 64.3, 53.1, 52.4, 18.7, 18.5. IR: 3337.66,
956.58, 1724.14, 1663.26, 1625.90, 1522.48, 1162.74, 1018.98,
43.15. HRMS (FAB) (M ): calcd 255.1107, found 255.1107.
+
and the filtrate washed with 0.5 M NaHCO
3
(3 × 15 mL) and
N,O-Bism eth a cr yloyl, L-Ser in e (3). In a 100 mL amber
0.5 M sodium citrate (3 × 15 mL), dried over magnesium
sulfate, and evaporated under vacuum to leave a brown oil.
The compound was isolated by flash chromatography using
bottle with cap was dissolved N,O-bismethacryloyl, L-serine
R-methyl ester, 8 (0.334 g, 1.3 mmol), in acetone (5 mL)
followed by the addition of 40 mL of 0.1 M phosphate buffer
of pH 7.5. To this mixture was added porcine pancreatic lipase,
EC 3.1.1.3 (100 mg). The mixture was sonicated for 1 min and
then shaken for 72 h at room temperature. The reaction
mixture was acidified to pH 3.0 with 1.0 M HCl. The aqueous
phase was extracted with EtOAc (3 × 20 mL), and the
combined organic extracts were washed with water (2 × 20
EtOAc/hexane 60/40, followed by EtOAc 100%, to give a yellow-
1
orange oil in 39% yield. H NMR (CDCl
3
, 250 MHz): δ 6.89-
6.92 (1H, d, J ) 8.21 Hz). 5.66 (1H, s), 5.26 (1H, s), 5.15 (1H,
m), 3.69 (3H, s), 3.54 (3H, s), 3.1 (3H, s), 2.63-2.69 (2H, dd, J
) 6.02, 5.84 Hz), 1.85 (3H, s). 13C NMR (CDCl
, 62.5 MHz): δ
3
171.1, 168.0, 139.6, 120.7, 62.0, 52.2, 46.8, 36.6, 32.6, 18.7.
IR: 333.83, 2953.69, 1739.36, 1655.33, 1623.80, 1526.06,
+
mL). The organic phase was dried over MgSO
4
and the solvent
1438.21, 1175.38, 988.38. HRMS (FAB) (M + H ): calcd
259.2710, found 259.1287.
evaporated under vacuum to give a yellow oil. The product
was isolated by flash chromatography using only EtOAc to give
N,r-Bism eth a cr yloyl, L-Asp a r tic Acid â-Meth yl Ester
(13). L-Aspartic acid â-methyl ester N-methacryloyl-N-meth-
oxy-N-methylamide (13, 2.58 g, 10 mmol) was dissolved in 10
1
an isolated yield of 61.8%. H NMR (CDCl
3
, 250 MHz): δ 10.68
(
1H, s, broad), 6.87-6.90 (1H, d, J ) 7.35 Hz), 6.04-6.05 (1H,
d, J ) 0.99 Hz), 5.74-5.75 (1H, d, J ) 0.79 Hz), 5.54-5.55
1H, d, J ) 1.54 Hz), 5.37-5.38 (1H, d, J ) 0.79 Hz), 4.85-
.91 (1H, m), 4.51-4.52 (2H, dd, J ) 3.95, 1.26 Hz), 1.91 (3H,
mL of dry THF and cooled to -60 °C under N . A 30 mL
2
(
portion of 0.5 M isopropenylmagnesium bromide (15 mmol)
was added dropwise over a period of 5 min, during which time
a pale yellow precipitate formed. The mixture was stirred for
30 min at -60 °C after complete addition of the Grignard
reagent, and then the mixture was brought to 0 °C and stirred
for another 30 min. After this period, 7 mL of saturated NH4-
Cl solution cooled at 0 °C was added to quench the reaction.
THF was removed under vacuum followed by addition of ethyl
ether (30 mL), after which the phases were separated. The
4
1
3
s), 1.85 (3H, s). C NMR (CDCl
3
, 62.5 MHz): δ 172.5, 169.3,
1
1
1
2
67.7, 139.0, 135.9, 127.2, 122.1, 76.1, 64.3, 52.9, 30.7, 18.7,
8.6. IR: 3366.15, 2958.72, 1718.42, 1658.39, 1619.96, 1542.53,
165.28, 1020.30, 943.78. HRMS (FAB) (M + H ): calcd
+
42.1095, found 242.1036.
N-t-Boc-L-a sp a r tic Acid â-Meth yl Ester N-Meth oxy-N-
Meth yla m id e (10). To a solution of the protected amino acid
(
9, 0.742 g, 3 mmol) in DCM (11 mL) was added N-methyl-
morpholine (0.668 g, 6.6 mmol). The mixture was cooled to -15
C, isobutylchloroformate (0.451 g, 3.3 mmol) was added, and
organic layer was washed with H
2
O (2 × 15 mL) and brine (1
× 20 mL), dried over MgSO
4
, filtered, and concentrated under
°
reduced pressure. The residue, a pale yellow oil, was separated
then the mixture was stirred at -15 °C for 15 min, followed
by addition of N,O-dimethylhydroxylamine hydrochloride (0.325
g, 3.3 mmol). The mixture was kept at -15 °C for an additional
by flash chromatography with EtOAc/hexane 50/50 giving 44%
1
yield of a yellow oil. H NMR (CDCl
(
(
3
, 250 MHz): δ 6.98-7.007
1H, d, J ) 6.79 Hz), 6.08 (1H, s), 5.86 (1H, t, J ) 1.42), 5.70
1H, d, J ) 0.79 Hz), 5.47-5.54 (1H, m), 5.33 (1H, d, J ) 0.95
1
h, allowed to warm to room temperature, and stirred for 24
Hz), 3.62 (3H, s), 2.66-2.87 (2H, dd, J ) 5.37, 5.21 Hz), 1.91
3H, s), 1.86 (3H, s). 13C NMR (CDCl
, 62.5 MHz): δ 198.2,
h. The reaction mixture was poured into water (11 mL) and
(
3
the aqueous phase extracted with DCM (2 × 6 mL). The
1
1
71.7, 167.8, 142.1, 139.7, 126.8, 120.9, 52.3, 50.3 37.1, 18.8,
8.2. HRMS (FAB) (M ): calcd 240.2677, found 240.1247.
4
combined organic extracts were dried over MgSO and filtered,
+
and the solvent was removed under vacuum to give pale yellow
oil. The product was isolated by flash chromatography using
N,r-Bism eth a cr yloyl, L-Asp a r tic Acid (2). In a 100 mL
1
EtOAc/hexane 50/50 to give a final yield of 98.0%. H NMR
amber bottle with cap was dissolved N,R-bismethacryloyl,
L-aspartic acid â-methyl ester (13, 0.406 g, 1.7 mmol) in 5 mL
of acetone, followed by the addition of pH 8.0 (0.1 M) phosphate
buffer (40 mL). To this mixture was added porcine liver
(
4
CDCl
3
, 250 MHz): δ 5.36-5.39 (1H, d, J ) 6.95 Hz), 4.96-
.98 (1H, m), 3.75 (3H, s), 3.62 (3H, s), 3.19 (3H, s), 2.55-2.78
1
3
(2H, dd, J ) 6.79, 5.52 Hz), 1.49 (9H, s). C NMR (CDCl
3
,
9
610 J . Org. Chem., Vol. 68, No. 25, 2003

Molecularly Imprinted Polymers
esterase, EC 3.1.1.1 (119 mg). The mixture was sonicated for
phase was washed with water (1 × 30 mL), brine (3 × 25 mL),
dried over MgSO , and the solvent was evaporated under
1
min and then shaken for 72 h at room temperature. The
4
reaction mixture was acidified to pH 3.0 with 1 N HCl. The
aqueous phase was extracted with EtOAc (3 × 20 mL). The
combined organic extracts were washed with water (2 × 20
vacuum. The target compound was isolated by flash chroma-
tography on silica gel using EtOAc/hexanes 50/50 as eluent.
The target compound was obtained in a 47% yield as a pale
1
mL). The organic phase was dried over MgSO
4
and the solvent
yellow oil that crystallized upon standing. H NMR (CDCl
3
,
evaporated under vacuum giving a yellow oil. The product was
isolated by flash chromatography with 100% EtOAc to give a
250 MHz): δ 10.49 (1H, broad), 6.04-6.05 (1H, t, J ) 1.60,
0.91 Hz), 5.51-5.53 (1H, t, J ) 1.60 Hz), 4.34-4.39 (2H, t, J
) 6.17 Hz), 2.67-2.70 (2H, t, J ) 6.17 Hz), 1.87-1.91 (3H, s).
1
5
7
5
5
5% yield. H NMR (CDCl
3
, 250 MHz): δ 9.27 (1H, s, broad),
1
3
.08-7.24 (1H, d, J ) 7.58 Hz), 6.10 (1H, d, J ) 0.79 Hz),
.89 (1H, d, J ) 1.42 Hz), 5.75 (1H, d, J ) 0.79 Hz), 5.50-
.57 (1H, m), 5.37 (1H, t, J ) 0.95, 0.79 Hz), 2.70-2.91 (2H,
dd, J ) 5.21, 5.05 Hz), 1.93 (3H, s), 1.88 (3H, s). C NMR
CDCl , 62.5 MHz): δ 198.7, 175.2, 168.4, 141.9, 139.4 127.3,
21.7, 50.5, 37.7, 18.8, 18.3. IR: 3359.50, 2928.88, 1722.55,
687.87. 1614.67, 1525.64, 1215.27, 940.79. HRMS (FAB)
3
C NMR (CDCl , 62.5 MHz): δ 177.2, 167.6, 136.3, 126.5, 60.2,
34.0, 18.6. IR: 3450, 2963.3, 1718.2, 1700.1, 1634.97, 1454.15,
1404.79, 1325.07, 1300.43, 1166.68, 1023.63, 947.37, 814.56.
1
3
+
HRMS (FAB) (M + H ): calcd 159.0579, found 159.0485.
(
1
1
3
P olym er P r ep a r a tion . Typical formulation: in a 13 × 100
mm test tube, 0.0698 g (0.43 mmol) of (S)-nicotine was
3
dissolved in 2 mL of CHCl . To this solution was added 1.530
+
(
M ): calcd 226.1001, found 226.1090.
g (7.73 mmol) of EGDMA, 0.206 g (0.86 mmol) of 3, and 0.028
g (0.172 mmol) of AIBN. The solution was purged by bubbling
nitrogen gas into the mixture for 5 min, capped, and sealed
with Teflon tape and Parafilm. The samples were inserted into
a photochemical turntable reactor, which was immersed in a
constant-temperature bath. A standard laboratory UV light
source (medium pressure 450 W mercury arc lamp) jacketed
in a borosilicate double-walled immersion well was placed at
the center of the turntable. The polymerization was initiated
photochemically at 20 °C and the temperature maintained by
both the cooling jacket surrounding the lamp and the constant-
temperature bath holding the entire apparatus. The poly-
merization was allowed to proceed for 10 h and then used for
chromatographic experiments.
Ch r om a togr a p h ic Exp er im en ts. Removal of the template
was achieved by Soxhlet extraction with methanol for 48 h.
Then the polymers were ground using a mortar and pestle,
the particles were sized using U.S.A. Standard Testing Sieves,
and the fraction between 20 and 25 µm was collected. The
particles were slurry packed, using a solvent delivery module,
into stainless steel columns (length, 75 or 100 mm; i.d., 2.1
mm) to full volume for chromatographic experiments. The
polymers were then washed on line overnight using MeCN/
HOAc 90/10, at a flow rate of 0.1 mL/min to remove any
residual template. HPLC analyses were performed isocratically
at room temperature (21 °C). The void volume was determined
using acetone as an inert substrate. The separation factors
N-Meth a cr yloylglycin e (5). Compound 5 was synthesized
as a white powder in 42% yield, following the protocol reported
1
7 1
by Heilmann. H NMR (MeOD, 250 MHz): δ 5.60-5.61 (1H,
t, J ) 0.95 Hz), 5.24-5.25 (1H, t, J ) 1.42 Hz), 4.77 (2H, s),
1
3
1
1
1
1
.99 (3H, s), C NMR (MeOD, 62.5 MHz): δ 172.1, 170.4,
39.8, 120.1, 41.0, 17.6. IR: 3359.85, 2983.28, 175.03, 1587.79,
205.06, 1027.07, 960.75, 895.30. HRMS (FAB) (M ): calcd
44.0582, found 144.0665.
+
Hyd r oxyp r op yl Meth a cr yla te (16). A solution of 1,3-
propanediol (15, 15.0 g, 197 mmol) and Et
3
N (33.1 g, 45.6 mL,
3
27 mmol) in 200 mL of dry THF was cooled to 0 °C.
Methacryloyl chloride (20.6 g, 197 mmol) was added dropwise
over a period of 1 h with stirring, keeping the temperature
between 0 and 3 °C. The reaction mixture was allowed to warm
to 50 °C and kept at this temperature for 2 h. After this period,
the reaction mixture was cooled and the precipitate filtered.
The solvent was evaporated under vacuum and the residue
redissolved in EtOAc (150 mL). The organic phase was washed
with 0.5 M NaHCO
3
(2 × 100 mL) and 0.5 M sodium citrate
(2 × 100 mL), dried over MgSO
4
, and filtered and the solvent
evaporated under vacuum. The residue was purified by flash
chromatography on silica gel sequentially using the eluents
EtOAc/hexanes 20/80, EtOAc/hexanes 50/50 to give a colorless
1
liquid (39% yield). H NMR (CDCl
(
3
, 250 MHz): δ 5.98-5.99
1H, t, J ) 1.60, 0.91 Hz), 5.44-5.46 (1H, t, J ) 1.60 Hz), 4.14-
4
(
.19 (2H, t, J ) 6.17 Hz), 3.51-3.61 (2H, t, J ) 6.17 Hz), 3.03
13
1H, broad), 1.77-1.85 (5H, m). C NMR (CDCl
3
, 62.5 MHz):
L D
(R) were measured as the ratio of capacity factors k′ / k′ . The
δ 168.1, 136.5, 126.0, 62.0, 59.2, 32.0, 18.4. IR: 3419.3,
capacity factors were determined by the relation k′ ) (R
v
v
-
2
1
961.11, 2929.92, 2890.13, 1716.47, 1636.78, 1454.14, 1322.62,
D
D
)/D
v v
, where R is the retention volume of the substrate and
298.62, 1173.11, 1053, 944.66, 816.36.. HRMS (FAB) (M +
H ): calcd 145.0786, found 145.0871.
v
is the void volume.
+
2
-Meth yla cr ylic Acid 2-Ca r boxyeth yl Ester (6). Hy-
Ack n ow led gm en t. This research was supported in
droxypropyl methacrylate (16, 1.15 g, 8 mmol) and 150 mg of
BHT were dissolved in acetone (10 mL), and the solution was
cooled to 5-10 °C. To the cooled solution was added J ones
reagent (6.5 mL) dropwise over a period of 30 min, keeping
the temperature between 5 and 10 °C. Afterward, the solution
was allowed to warm to room temperature and stirred for
additional 30 min. Then the reaction mixture was cooled at
0 °C, and excess J ones reagent was quenched by dropwise
addition of 2 mL of 2-propanol. A 25 mL portion of H O was
added to dissolve completely the solid formed, and the reaction
part by Research Corp.n, through Cottrell Scholar
Award CS0801, the Petroleum Research Fund (ACS-
PRF# 34485-G7), and an NSF CAREER Program award
CHE-0134290 (D.A.S.).
Su p p or tin g In for m a tion Ava ila ble: General experimen-
1
13
tal and copies of H NMR and C NMR spectra of all
synthesized compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
1
2
mixture was extracted with EtOAc (3 × 25 mL). The organic
J O0352225
J . Org. Chem, Vol. 68, No. 25, 2003 9611