Polymers recognizing biomolecules based on a combination of molecular imprinting and proximity scintillation: A new sensor concept

Full text of DOI:10.1021/ja005896m

J. Am. Chem. Soc. 2001, 123, 2901-2902
2901
“reporter” into the MIP’s specific “cavity”.6 The fluorescence
intensity changes when a target analyte binds to the cavity. In
principle this sensor design should deliver superior specificity,
since only specific binding generates a sensor signal. However,
for different target analytes, new reporter monomers need to be
synthesized.
Polymers Recognizing Biomolecules Based on a
Combination of Molecular Imprinting and Proximity
Scintillation: A New Sensor Concept
Lei Ye and Klaus Mosbach*
In this communication we describe a molecularly imprinted
polymer incorporating a “universal” reporter (Figure 1). Genera-
tion of the binding signal is based on the principle of proximity
energy transfer, e.g. proximity scintillation. Due to the small size
of the MIP particles, the signal-producing element (scintillation
fluor) is located in close proximity to the MIP’s binding “cavity”.
The scintillation fluor virtually does not affect template re-binding,
therefore it can be used for both covalent and noncovalent
imprinting. In this study we use the general functional monomer,
methacrylic acid (MAA), for noncovalent imprinting of an
antagonist, (S)-propranolol (2). A scintillation monomer (1) is
covalently incorporated into our MIP microparticles during the
imprinting reaction. Rather than being confined within the binding
cavity, the fluor is randomly distributed throughout the polymer
matrix. The scintillation monomer, 4-hydroxymethyl-2,5-diphen-
yloxazole acrylate (1b), is synthesized by coupling 1a7 with
acryloyl chloride. In toluene its scintillation efficiency is ap-
proximately 62% of the commercial fluor, 2,5-diphenyloxazole
(PPO). MIPs incorporating different amounts of 1 are synthesized
by using a previously reported precipitation method (Table 1),8
in which trimethylolproprane trimethacrylate (TRIM) is used as
the cross-linking monomer. Spectrofluorometric and elemental
analyses of the obtained polymers confirm that the imprinted
polymer (IP) and nonimprinted polymer (NP) contain ap-
proximately equal amount of 1.
Despite the difference in toluene fraction used for polymer
synthesis, all the polymer microparticles obtained have an average
diameter of 0.6-1 µm. These are nonporous, and have a surface
area of approximate 7 m2 g-1. A relatively stable suspension of
polymer microparticles is readily obtained in toluene due to their
small size. The polymers bearing scintillation fluor 1 displayed
excitation and emission spectra very similar to that of the
scintillation monomer 1b (virtually identical maximum excitation
and emission wavelength). This confirmed that incorporation into
a polymer matrix does not affect the fluor’s scintillation efficacy.
To evaluate the MIPs’ response to the specific binding of the
tritium-labeled template, increasing amounts of the imprinted and
nonimprinted polymer are incubated in toluene (500 µL, contain-
ing 0.5% acetic acid (v/v)) with a fixed amount of [H3](S)-
propranolol. Samples are counted without removing the unbound
[H3](S)-propranolol (proximity scintillation counting).9 In both
cases the imprinted polymers deliver much higher counts than
the nonimprinted polymers, which is attributed to the specific
binding of the labeled template with the imprinted polymerssa
fact confirmed by quantifying the free fraction with liquid
scintillation counting.10 When 0.2 mg of polymer is used, the two
imprinted polymers generate approximately half of the maximum
Pure and Applied Biochemistry, Chemical Center
Lund UniVersity, Box 124, 221 00 Lund, Sweden
ReceiVed December 18, 2000
Modern drug discovery and demand for quick diagnostics
require development of highly sensitive and efficient analytical
methods. Sensitive techniques such as radiometric and fluorescent
measurements have been widely applied to detect trace amounts
of analytes. In many cases, for example in clinical samples and
combinatorial libraries, a target analyte coexists with a large
number of interfering compounds. To detect/quantify the target,
a separation step is often required in which a selective binding
material is employed to fish out the target analyte during the
analyses. Biological antibodies, receptors, enzymes, and single-
strand DNA segments have been utilized as the binding materials
since they are able to recognize corresponding antigens, agonists/
antagonists, substrates/inhibitors, and complementary nucleotide
sequences with high specificity. It is possible to eliminate the
separation step, thereby greatly increasing the sample throughput
by integrating the biological binding materials with appropriate
signal transduction systems, for example as in Scintillation
Proximity Assays (SPA).1
To substitute biological binding materials in general, synthetic
host molecules have been studied very intensively in supramo-
lecular chemistry,2 and more recently, in molecular imprinting
of cross-linked polymers.3 As a synthetic approach based on
template-assisted assembly of a polymeric host, molecular
imprinting generates binding “cavities” that are complementary
to the original template in both shape and functionality. Molecu-
larly imprinted polymers (MIPs) have high binding affinity and
specificity, and they are stable and relatively easy to prepare. In
addition to separation and more routine analytical applications
involving affinity adsorbents,4 they have been used as recognition
elements in various sensors.5 In most cases, an imprinted polymer
is put in physical contact with a transducer. The physicochemical
response (change in mass, resistance, capacitance, refractive index,
etc.) from binding a target analyte is translated into a sensor signal.
This simple method, however, often leads to MIP sensors showing
relatively low sensitivity and specificity. In a more sophisticated
manner, a fluorescent functional monomer is incorporated as a
* To whom correspondence should be addressed. Phone: +46-46-2229560.
Fax: +46-46-2224611. E-mail: Klaus.Mosbach@tbiokem.lth.se.
(1) (a) Bosworth, N.; Towers, P. Nature 1989, 341, 167-168. (b)
Udenfriend, S.; Gerber, L.; Nelson, N. Anal. Biochem. 1987, 161, 494-500.
(2) Lehn, J.-M. Angew. Chem., Int. Ed. Engl. 1988, 27, 89-112.
(3) (a) Mosbach, K.; Ramstro¨m, O. Bio/Technology 1996, 14, 163-170.
(b) Wulff, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812-1832. (c) Shea,
K. J. Trends Polym. Sci. 1994, 2, 166-173. (d) Wulff, G.; Gross, T.;
Scho¨nfeldt, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 1962-1964. (e) Lu¨bke,
C.; Lu¨bke, M.; Whitcombe, M. J.; Vulfson, E. N. Macromolecules 2000, 33,
5098-5105. (f) Alexander, C.; Smith, C. R.; Whitcombe, M. J.; Vulfson, E.
N. J. Am. Chem. Soc. 1999, 121, 6640-6651.
(4) (a) Kempe, M.; Mosbach, K. J. Chromatogr. A 1994, 664, 276-279.
(b) Sellergren, B. Anal. Chem. 1994, 66, 1578-1582. (c) Andersson, L. I.
Anal. Chem. 1996, 68, 111-117. (d) Ye, L.; Ramstro¨m, O.; Mosbach, K.
Anal. Chem. 1998, 70, 2789-2795.
(5) (a) Haupt, K.; Mosbach, K. Chem. ReV. 2000, 100, 2495-2504. (b)
Kro¨ger, S.; Turner, A. P. F.; Mosbach, K.; Haupt, K. Anal. Chem. 1999, 71,
3698-3702. (c) Malitesta, C.; Losito, I.; Zambonin, P. G. Anal. Chem. 1999,
71, 1366-1370. (d) Lin, J.-M.; Yamada M. Anal. Chem. 2000, 72, 1148-
1155.
(6) (a) Turkewitsch, P.; Wandelt, B.; Darling, G. D.; Powell, W. S. Anal.
Chem. 1998, 70, 2025-2030. (b) Wang, W.; Gao, S.; Wang, B. Org. Lett.
1999, 1, 1209-1212. (c) Matsui, J.; Higashi, M.; Takeuchi, T. J. Am. Chem.
Soc. 2000, 122, 5218-5219.
(7) (a) Clapham, B.; Richards, A. J.; Wood, M. L.; Sutherland, A. J.
Tetrahedron Lett. 1997, 38, 9061-9064. (b) Hamerton, I.; Hay, J. N.; Jones,
J. R.; Lu, S.-Y. Chem. Mater. 2000, 12, 568-572.
(8) (a) Ye, L.; Cormack, P. A. G.; Mosbach, K. Anal. Commun. 1999, 36,
35-38. (b) Ye, L.; Weiss, R.; Mosbach, K. Macromolecules 2000, 33, 8239-
8245.
(9) Samples are preincubated overnight to ensure equilibrium binding,
although a short incubation time (2 h) is sufficient. A standard liquid
scintillation counter (â-radiation counter) is used.
(10) Samples are centrifuged following the proximity scintillation counting.
A fraction of supernatant is mixed with a commercial scintillation liquid and
counted with a â-radiation counter.
10.1021/ja005896m CCC: $20.00 © 2001 American Chemical Society
Published on Web 03/06/2001

2902 J. Am. Chem. Soc., Vol. 123, No. 12, 2001
Communications to the Editor
Figure 2. Competitive assay of (S)-propranolol with polymer IP 1 (solid
square) and NP 1 (open square). [H3](S)-propranolol (2 pmol) and polymer
microparticles (0.2 mg) are incubated in toluene (500 µL, containing 0.5%
acetic acid (v/v)) with an increasing amount of (S)-propranolol prior to
proximity scintillation counting. ∆PSC is the change in proximity
scintillation counts caused by the unlabeled analyte competing for the
binding sites. Data are mean values of triplicate measurements.
Figure 1. Schematic representation of chemical sensing with an imprinted
polymer. Scintillation fluor (1) incorporated by its copolymerization into
the polymer converts â-radiation from the bound tritium-labeled template
into a fluorescent signal. The unbound labeled compound is too far away
from 1 for effective energy transfer, therefore no fluorescent light is
generated. The MIP is useful in two aspects: (a) in direct quantification
of the template that coexists with other compounds in a radioisotope-
labeled sample and (b) in competitive assay of the template where the
labeled template is used as a tracer, whose binding is inhibited by the
unlabeled template. In the present study, (S)-propranolol (2) is used as a
model template for demonstration.
Figure 3. Competitive assay of (S)-propranolol (solid square) and (R)-
propranolol (solid triangle) with IP 2. [H3](S)-propranolol (2 pmol) and
the polymer (0.1 mg) are incubated in toluene (500 µL, containing 0.5%
acetic acid (v/v)) with an increasing amount of analytes prior to proximity
scintillation counting.
Table 1. Syntheses of Imprinted and Nonimprinted Polymersa
monomer
concentration, e.g. this chemo sensor displays a lower detection
limit (Figure 3). The sensor is chiral selective, since the antipode
of the template, (R)-propranolol, causes a much smaller PSC
change. This chiral discrimination can be further improved if
required, although more important in this communication is the
chiral specificity as such.
The present example has demonstrated a new sensor concept
using molecular imprinting and proximity scintillation. We have
used a photomultiplier tube (PMT) for signal quantification. When
combined with arrays of PMTs or imaging systems, for example
with a CCD camera, imprinted scintillation polymers can be
deposited in microtiter plate wells and used for high throughput
screening purposes. MIP sensors based on other proximity energy
transfer systems are also envisioned.
polymer 2 (mmol) MAA (mmol) 1b (mmol) TRIM (mmol)
IP 1
NP 1
IP 2
0.7
1.39
1.39
0.78
0.78
0.69
0.69
0.16
0.16
1.39
1.39
0.78
0.78
0.6
NP 2
a The reaction components and the initiator, azobisisobutyronitrile
(2 wt % of monomer), are dissolved in anhydrous toluene (40 mL).
The solution is saturated with nitrogen and heated at 60 °C for 16 h.
Discrete polymer microparticles are obtained after a brief treatment in
an ultrasonic water bath followed by centrifugation. The template is
removed by repetitively washing the polymers with methanol-acetic
acid (90:10, v/v) and acetone. The polymers are dried in a vacuum.
fluorescence counts, which are two times higher than the
nonimprinted polymers. The MIPs bearing scintillation fluor
specifically bind their templates. More importantly, the specific
binding event is translated into a scintillation signal.
In a competitive mode, binding of tritium-labeled (S)-propra-
nolol to the imprinted polymer is inhibited by increasing the
amount of the unlabeled template, which results in decreased
proximity scintillation counts (PSC) (Figure 2). With the non-
imprinted polymer, however, the nonlabeled template does not
change the PSC significantly.
Acknowledgment. This work is in part supported by the Swedish
Research Council for Engineering Sciences (TFR). The authors thank
Dr. Kenneth Wa¨rnmark, Organic Chemistry I, Lund University for
assistance in NMR measurement.
Supporting Information Available: Synthesis and 1H NMR data
for 1b and fluorescent spectra for 1b and polymer IP 1 (PDF). This
material is available free of charge via the Internet at http://pubs.acs.org.
When less scintillation monomer is incorporated, the obtained
polymer IP 2 generates PSC change at even lower analyte
JA005896M