Molecular imprinted polymers binding low functionality templates

Article abstract of DOI:10.1016/j.tetlet.2010.08.108

A series of highly specific molecular imprinted polymers (MIPs) for small, low functionality bicyclic Diels-Alder products was prepared as bulk polymers.

Full text of DOI:10.1016/j.tetlet.2010.08.108

Tetrahedron Letters 51 (2010) 5883–5885
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Molecular imprinted polymers binding low functionality templates
Yvonne Luk a,b, Christopher J. Allender a, , Thomas Wirth b,
⇑
⇑
a Welsh School of Pharmacy, Cardiff University, Redwood Building, King Edward VII Avenue, Cardiff CF10 3NB, UK
b School of Chemistry, Cardiff University, Main Building, Cardiff CF10 3AT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 25 June 2010
Revised 30 July 2010
Accepted 31 August 2010
Available online 6 September 2010
A series of highly specific molecular imprinted polymers (MIPs) for small, low functionality bicyclic
Diels–Alder products was prepared as bulk polymers.
Ó 2010 Elsevier Ltd. All rights reserved.
The technique of molecular imprinting, which is the creation of
unique 3D cavities possessing ‘lock and key’ type fit for specific tar-
get molecules, has generated a great deal of interest,1,2 due to the
high degree of molecular specificity that is achievable, thermal and
chemical stability and the ease with which such materials can be
prepared.3 This has resulted in MIPs being used in a wide range
of applications including liquid chromatography,4,5 solid phase
extraction,6 biosensors7 and chemical synthesis.8 In the area of or-
ganic chemistry, MIPs have been used as drug or lead-compound
screening aids,9 non-covalent protecting groups10 and as cata-
lysts.11 Since the specific binding of template to imprinted site
relies on non-covalent intermolecular forces, using templates that
are small and/or have limited functionality remains a challenge.12
Attempts to imprint small mono-functional molecules have been
reported, and functional MIPs for pyridine12 and propofol13 have
been prepared using conventional non-covalent and semi-covalent
(sacrificial-spacer) approaches. For both of these simple template
molecules, selectivities in general, although slightly better for MIPs
prepared using the semi-covalent approach, were poor. The effect
of template shape and functionality on the effectiveness of the
MIP has previously been studied by Matsui et al.14 who reported
that template structures with complex 3D shapes tended to result
in MIPs with higher selectivity. However, in this work the template
possesses only one functional group containing two oxygen atoms
(ester). We aimed to prepare MIPs for a mono-functional bicyclic
template prepared as the endo-product of a Diels–Alder reaction
in order to cast some light on the role of molecular shape in the
imprinting process. It is notoriously difficult to characterise MIP
selectivity on the basis of comparative template affinities for the
imprinted and non-imprinted control polymer. To use such an ap-
proach requires that non-specific binding contributions are equiv-
alent for the MIP and the control polymer and this is rarely, if ever,
the case. Furthermore, to assess specificity effectively in this way
typically involves MIP–control polymer comparative binding stud-
ies for a range of cross-reacting molecules. An alternative approach
is to use a template as a single stereoisomer so that binding selec-
tivity could be evaluated on the basis of chiral discrimination.15 In
this study selectivity was evaluated on the basis of endo/exo dis-
crimination. The endo and exo products investigated herein are dia-
stereomers containing three stereogenic centres where the
configuration at only one carbon differs in the endo and in the
exo product. For the demonstration of endo and exo selectivity,
the MIP must discriminate on the basis of a change in configuration
at a single stereocentre. Herein, we describe the preparation and
evaluation of a MIP imprinted with the endo-product of a Diels–Al-
der reaction.
A Diels–Alder reaction between cyclopentadiene 1 and benzyl
acrylate (2) at 70 °C for 17 h yielded the reaction products 3 (endo)
and 4 (exo) with a combined yield of 80% in an endo/exo ratio of 3:1
as shown in Scheme 1.16 Products 3 and 4 were then separated by
medium-pressure liquid chromatography. The major isomer 3 was
chosen as template.
In our preliminary studies using a related template (the Diels–
Alder adduct of cyclopentadiene and methyl vinyl ketone) and
acetamide (as mimic of the functional group monomer), it was
found in 1H NMR experiments that the interaction between those
compounds did not correlate with subsequent equilibrium binding
studies. Therefore, in order to identify a lead polymer composition,
a number of MIPs containing different functional monomers were
prepared and screened. This was carried out on a small scale using
bulk polymerisation initiated at 60 °C using AIBN (Table 1).17 Non-
imprinted polymers (NIPs) were also made using the same reaction
70 oC
+
+
H
CO2Bn
CO2Bn
17 h
CO2Bn
H
4
1
2
3
⇑
Corresponding authors.
E-mail address: wirth@cf.ac.uk (T. Wirth).
Scheme 1. Synthesis of the template
cyclopentadiene (1) and benzyl acrylate (2).
3
via a Diels–Alder reaction between
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.08.108

5884
Y. Luk et al. / Tetrahedron Letters 51 (2010) 5883–5885
Table 1
nificant apparent specific binding was observed. A predictable de-
crease in apparent specific binding was observed as initial
template concentration increased.
Compositions of MIPs P1–P5a
P1
P2
P3
P4
P5b
A number of interesting observations arose when the percent-
age bound of P2 and P3 NIP and MIP were plotted against the ini-
O
O
O
O
N
OH
—
NH2
N
N
tial template concentrations of 10, 25 and 50
MIP binds a very similar percentage (ꢁ50%) of template over the
range 10–50 M, whilst for the P2 control polymer there is a grad-
lM (Fig. 1). Firstly, P2
H
H
O
O
l
ual increase in the percentage bound (11–42%) over the same tem-
plate concentration range. For P3 MIP, over the same concentration
range, the percentage bound falls from 50% to ꢁ40%, whilst for the
control polymer the percentage bound remains relatively consis-
tent. It can be misleading to over interpret such binding data since
MIP binding site affinity is extremely polyclonal and estimations of
available binding site concentration is fraught with difficulties.
However, it was surprising, yet reassuring, that given the differ-
ence in monomer composition, P2 and P3 MIPs and control poly-
mers behaved in a similar manner.
4 equiv.
4 equiv.
4 equiv.
4 equiv.
O
O
CHCl3c
(0.5 mL)
CHCl3
(0.515 mL)
CHCl3
CHCl3
(0.434 mL)
CHCl3
(0.5 mL)
c
c
c
c
(0.434 mL)
NIPs were prepared in the same manner, but in the absence of the template 3.
a
All polymers also contain 0.01 equiv of azobisisobutyronitrile (AIBN). MIPs also
contain 0.25 equiv of template 3. The reactants were placed in polymerisation vials
and purged in nitrogen for 5 min and then heated at 60 °C for 17 h. Polymers were
ground and washed prior to use.
In an attempt to further reduce non-specific interaction
between template and polymer, an alternative cross-linking mono-
mer, ethylene glycol dimethacrylate (EGMA), was evaluated. Previ-
ously it had been reported that EGMA can result in reduced
non-specific binding (in apolar solvents) and can provide the addi-
tional benefit of improving polymer flexibility and accessibility.20
However, when the binding of 3 to a non-imprinted EGMA poly-
mer P5 was evaluated, non-specific binding was found to be great-
er than that of an equivalent divinylbenzene (DVB) cross-linked
polymer when evaluated under the same conditions. As a result,
DVB was favoured as the cross-linker. A further polymer modifica-
tion was evaluated where the acrylamide in P3 was replaced with
an equimolar amount of N,N0-methylene bisacrylamide (MBA) in
P4. The reason for this modification was to create conformational
dependence between adjacent amide groups and increased rigidity
in and around the imprinted site. In previous studies this had been
shown to give rise to an improved MIP performance. However, by
maintaining equimolar amounts of acrylamide and MBA the num-
ber of amide groups in P4 was doubled compared with P3. There-
fore, in order to make valid comparisons between acrylamide and
MBA-containing polymers a further polymer (P3–8) was prepared
containing 8 equiv of acrylamide. For completeness a polymer con-
taining 8 equiv of 4-vinylpyridine was also synthesised (P2–8).
Interestingly, substitution of acrylamide in P3 with an equimo-
lar amount of MBA in P4 resulted in a slight increase in MIP and
b
P5 was prepared to investigate whether the non-specific binding could be
reduced.
c
Solvent volumes were varied so as to maintain a constant monomer: solvent (v/
v) ratio.
conditions but in the absence of the template. The ability of each of
these polymers to bind the template was evaluated using simple
single point equilibrium binding assays (2 mg of polymer in 2 mL
of MeCN; template concentration of 10–500 lM, mechanical shak-
ing for 17 h) where equilibrium ‘free’ template concentrations
were determined by HPLC. [HPLC-conditions: RP Kromasil C18
5
l
m, 250 mm ꢀ 4.6 mm; 20
lL; 1 mL/ min; 215 and 235 nm; sol-
vent: MeOH/H2O (90/10)]. Preliminary studies demonstrated high
non-specific binding (giving a poor imprinting affect) when assays
were carried out in chloroform. This situation improved when ace-
tonitrile was used. Initially, three different divinylbenzene (DVB)
cross-linked MIPs were evaluated containing an acidic (metha-
crylic acid, P1), a basic (4-vinylpyridine, P2) and a neutral func-
tional monomer (acrylamide, P3) at a monomer template ratio of
(4:1) (Table 1). For P1, binding studies showed no difference be-
tween the amount of template binding to the MIP compared to
the control. For polymers P2 and P3 the MIPs bound 6.6% and
7.7%, respectively, more template than did their controls. This
was taken as an indication of an imprinting affect.
NIP binding for low template concentrations (10
lM) and a slight
Although it is speculative to hypothesise as to the type and
number of interactions responsible for binding between template
and polymer it is interesting to note that for the methacrylic acid
containing P1, there is no suggestion of an imprinting effect,
whereas for both the 4-vinylpyridine- and acrylamide-containing
polymers (P2 and P3) binding is favoured to the MIP. Therefore,
neither acid nor basic group appears to be a prerequisite for pro-
ducing an imprinting affect. The most likely point of interaction
for P2 is between the nitrogen of 4-vinylpyridine and the carbonyl
carbon of the template, whilst for P3 a number of different hydro-
gen bonds are conceivable between the amide group of acrylamide
and the template carbonyl group. Given the nature of the environ-
ment it is also possible that interfacial hydrogen bonding, between
the acrylamide amide and template benzyl group, might also make
a contribution.18,19
decrease for higher template concentrations (25 and 50
is despite the fact that P4 contained twice the number of amide
residues compared to P3. However, the difference between MIP
l
M). This
P2 NIP
IF= 2.4
P2 MIP
IF= 3.8
4.6
IF= 1.1
IF= 1.4
11.5
60
50
40
30
20
10
0
P3 NIP
P3 MIP
5.3
IF= 2.0
10.4
23.8 IF= 1.4
20.9
19.1
8.3
13.6
2.2
5.3
1.2
Following these initial observations, more extensive binding
studies were undertaken in order to construct binding isotherms
(not shown) for the binding of template 3 to P2 and P3 (MIP and
10
25
Concentration (µM)
50
control polymers). At starting concentrations >100 lM apparent
specific binding (nmol bound per mg MIP—nmol bound per mg
control polymer) was low suggesting a saturation of available
MIP binding sites whilst at template concentrations <100 lM, sig-
Figure 1. The percentage of 3 bound to P2 and P3 NIPs and MIPs. The amount of 3 per
milligram of polymer is given at the top of each bar in nmol/mg. IF (imprinting factor):
ratio between the amount of 3 bound to MIP and the amount of 3 bound to NIP.

Y. Luk et al. / Tetrahedron Letters 51 (2010) 5883–5885
5885
IF= 2.9
4.6
Figure 3 clearly shows that there was no difference between the
amount of 4 binding to the MIP as compared to the NIP. It is inter-
esting to note that, particularly at lower ligand concentrations,
non-specific binding was significantly greater for the exo com-
IF= 1.2
24.0
60
50
40
30
20
10
0
P4 NIP
P4 MIP
IF= 1.6
8.5
19.5
pound 4 as compared to the endo compound 3 [10
NIP = 1.2 nmol/mg; 4 to NIP = 3.3 nmol/mg].
lM 3 to
5.3
In conclusion, a series of MIPs were prepared against a small,
poorly functional template 3 and their specificity was evaluated
using a simple equilibrium-binding assay. For a number of polymer
systems, the amount of endo template 3 binding to the MIP was
consistently greater than that binding to the NIP. When the exo
form of the template 4 was evaluated, the amount binding MIP
and NIP was similar.
1.6
10
25
Concentration (µM)
50
Figure 2. The percentage bound for P4 NIPs and MIPs. The amount of 3 per
milligram of polymer is given at the top of each bar in nmol/mg. IF (imprinting
factor): ratio between the amount of 3 bound to MIP and the amount of 3 bound to
NIP.
Acknowledgements
We are grateful to Cardiff University for financial support and
the EPSRC National Mass Spectrometry Service Centre, Swansea,
for mass spectrometric data.
IF= 0.97
60
50
40
30
20
10
0
IF= 0.92
11.8
P3 NIP
P3 MIP
24.8
24.0
Supplementary data
10.9
IF= 1.1
3.3 3.6
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.tetlet.2010.08.108.
References and notes
1. Alexander, C.; Andersson, H. S.; Andersson, L. I.; Ansell, R. J.; Kirsch, N.; Nicholls,
I. A.; Mahony, J. O.; Whitecombe, M. J. J. Mol. Recognit. 2006, 19, 106–180.
2. Yan, M.; Ramstrom, O. In Molecular Imprinting Materials: Science and
Technology; CRC press: London, 2004.
3. Zhang, H.; Piacham, T.; Drew, M.; Patek, M.; Mosbach, K.; Ye, L. J. Am. Chem. Soc.
2006, 128, 4178–4179.
10
25
Concentration (µM)
50
4. Sanbe, H.; Hosoya, K.; Haginaka, J. Anal. Sci. 2003, 19, 715–719.
5. Matsui, J.; Nicholls, I.; Takeuchi, T. Anal. Chim. Acta 1998, 343, 89–93.
6. Liu, Y.; Liu, X.; Wang, J. Anal. Lett. 2003, 36, 1631–1645.
7. Ye, L.; Mosbach, K. J. Am. Chem. Soc. 2001, 123, 2901–2902.
8. Alexander, C.; Davidson, L.; Haynes, W. Tetrahedron 2003, 59, 2025–2057.
9. Ramstrom, O.; Ye, L.; Mosbach, K. Anal. Commun. 1998, 5, 9–11.
10. Alexander, C.; Smith, C. R.; Whitcombe, M. J.; Vulfson, E. N. J. Am. Chem. Soc.
1999, 121, 6640–6651.
11. Matsui, J.; Nicholls, I. A.; Karube, I.; Mosbach, K. J. Org. Chem. 1996, 61, 5414–
5417.
12. Kirsch, N.; Alexander, C.; Davies, S.; Whitcombe, M. J. Anal. Chim. Acta 2004,
504, 63–71.
13. Petcu, M.; Karlsson, J. G.; Whitcombe, M. J.; Nicholls, I. A. J. Mol. Recognit. 2009,
22, 18–25.
14. Matsui, J.; Nicholls, I. A.; Takeuchi, T. Tetrahedron: Asymmetry 1996, 7, 1357–
1361.
15. Spivak, D. A. Adv. Drug Delivery Rev. 2005, 57, 1779–1794.
16. Essers, M.; Mück-Lichtenfeld, C.; Haufe, G. J. Org. Chem. 2002, 67, 4715–4721.
17. Lanza, F.; Sellergren, B. Macromol. Rapid Commun. 2004, 25, 59–68.
18. Mitchell, J. B. O.; Nandi, C. L.; Ali, S.; McDonald, I. K.; Thornton, J. M.; Price, S. L.;
Singh, J. Nature 1993, 366, 413.
Figure 3. The percentage of 4 bound to P3 NIPs and MIPs. The amount of 4 per
milligram of polymer is given at the top of each bar in nmol/mg. IF (imprinting
factor): ratio between the amount of 3 bound to MIP and the amount of 3 bound to
NIP.
and NIP binding was similar (ꢁ14% at 25
mentary data) and P4 (see Fig. 2).
lM) for P3–8 (see Supple-
At 25 lM, the imprinting factors (IF) for P3, P4 and P3–8 were
2, 1.6 and 1.7, respectively. This indicated that the imprinting ef-
fect was not affected by the number of acrylamide groups. The
amounts of non-specific binding for P3 and P4 were the same
(5.3 nmol/mg) even though P4 had twice as many acrylamide
groups. However, P3–8 had lower non-specific binding compared
to P4 (3.2 nmol/mg for P3–8 and 5.3 nmol/mg for P4) even though
both polymers contained the same number of acrylamides.
From the different polymers, P3 gave the largest amount of spe-
cific binding at higher concentrations (Fig. 1). In order to evaluate
the endo/exo selectivity the equilibrium binding of 4 (exo form of
the product) to P3 (MIP imprinted with 3 and NIP) was evaluated.
19. Adams, H.; Carver, J.; Hunter, C. A.; Osborne, N. J. Chem. Commun. 1996, 2529–
2530.
20. Wulff, G.; Vietmeier, J.; Poll, H. G. Makromol. Chem. 1987, 188, 731–740.