A bifunctional Molecularly Imprinted Polymer (MIP): Analysis of binding and catalysis by a thermistor

Article abstract of DOI:10.1002/anie.200601796

(Figure Presented) Binding or catalysis? Both can be distinguished with a molecularly imprinted polymer (MIP) by the different patterns of heat generation. The catalytically active sites, like in the corresponding enzyme, generate a steady-state temperature increase. Thus, enzyme-like catalysis and antibody-analogue binding are analyzed simultaneously in a bifunctional MIP for the first time (see scheme).

Full text of DOI:10.1002/anie.200601796

Communications
DOI: 10.1002/anie.200601796
Molecular Recognition
A Bifunctional Molecularly Imprinted Polymer (MIP):
Analysis of Binding and Catalysis by a Thermistor**
Kristian Lettau, Axel Warsinke, Martin Katterle, Bengt Danielsson, and
Frieder W. Scheller*
Angewandte
Chemie
6
986
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6986 –6990

Angewandte
Chemie
diameter of approximately 40 mm and catalyzed the solvolysis
N
ature has created recognition elements based either on
of non-activated phenolic esters (Scheme 1). The specific
activity of this polymer towards 4-nitrophenylacetate (NPA)
was 90 mUmg polymer and the apparent Michaelis–
proteins or on nucleic acids, showing high chemical selectiv-
ity and catalytic power. As an alternative to the evolutionary
generation of optimal binders and catalysts based amino
acids and nucleotides, totally synthetic molecularly imprinted
polymers (MIPs) have been proposed by the groups of Wulff,
À1
[
1–3]
Shea, and Mosbach.
Such bio-inspired molecularly imprinted polymers have
the potential to substitute biological recognition elements in
chemoselective affinity chromatography, biosensors, and
biochips as a result of their high chemical and thermal
stability and low cost. However, except for a few sensor
[
4,5]
configurations, widespread application has yet to come.
A
barrier for routine applications is the signal generation when
the analyte binds to the MIP. Although all marker types of
immunoassays may be applied to the MIP–analyte interac-
tion, direct (label free) transducers are particularly promis-
Scheme 1. Proposed mechanism for the MIP-catalyzed solvolysis of
the non-activated substrate (phenylacetate) to phenol and acetic acid.
[
6–10]
ing.
In this respect MIP-covered quartz crystal micro-
balances as well as surface plasmon resonance sensors have
been successfully applied for the recognition of not only low-
molecular substances, but also of proteins, viruses, and living
Menten constant (K ) 2.2 mm. The rate constants (Table 1)
M
of nitrophenylacetate solvolysis are of the same order of
[
22–26]
magnitude as those reported for other esterolytic MIPs.
[
11–13]
cells.
Another possibility for label-free detection is the use of
calorimeters. In enzyme technology the enzyme thermistor is
widely used for this purpose since most enzymatic reactions
Table 1: Kinetic parameters for the MIP-catalyzed solvolysis of phenyl-
acetate (PA) and nitrophenylacetate (NPA).
[
a]
À1
Reagent Cat. kobs [min
]
kuncat.
kobs
kuncat.
/
(kobsÀkuncat.)/
are accompanied by heat evolution in the range of 5–
À1
[min
]
kuncat.
À1 [14–16]
1
00 kJmol .
The first isothermal titration calorimetric
[
b]
À6
À6
À3
À3
À6
À3
PA
MIP 15.83”10
5.92”10
0.13”10
2.67
1.18
28.69
14.2
1.67
0.18
27.69
13.12
studies with MIPs have been performed on the chiral
recognition of phenyl-a-d /l -mannopyranoside, and Haupt
et al. have already proposed that thermistor devices can
measure adsorption and desorption processes with imprinted
NIP
MIP
NIP
7.01”10
3.83”10
1.89”10
[
c]
NPA
À1
[
17–19]
[a] 100 mgmL polymer catalyst. [b] 500 mm phenylacetate. [c] 5 mm
polymers.
nitrophenylacetate.
Herein we demonstrate for the first time the combination
of a calorimetric transducer—the thermistor—with analyte
recognition by a catalytically active MIP. Both effects, the
binding/desorption and the catalytic conversion of the
substrate, were reflected by concentration-dependant heat
signals.
In general the rate constants for phenylacetate (PA) are lower
[
23]
than reported for NPA. This result can be explained by the
fact that our polymer was originally designed for the
solvolysis of NPA, and also by the fact that NPA is more
[
23]
The catalytic polymer was prepared, as described else-
where, with a phosphonic ester monoamide (template) as the
transition-state analogue of the ester hydrolysis and 4-vinyl-
prone to hydrolysis than PA. Phenylacetate was used as the
substrate for the measurements with the thermistor. This
substrate has a lower self-hydrolysis rate than the corre-
sponding nitrophenyl ester; it is soluble in a buffer with 10%
methanol in millimolar concentrations and shows, as reported
for other substrates, no product inhibition in the expected
[
20,21]
imidazole.
NIP) was prepared from the same monomer mixture but
without template. The polymer particles had an average
The corresponding non-imprinted polymer
(
[
27]
product concentration range.
For comparison of the esterolytic polymer with an
enzyme, the catalytic cleavage of phenylacetate with esterase
immobilized on amino-functionalized porous glass was inves-
[
*] Dipl.-Biochem. K. Lettau, Dr. habil. A. Warsinke, Dr. M. Katterle,
Prof. Dr. F. W. Scheller
Institut für Biochemie und Biologie
Universität Potsdam
Karl-Liebknecht Strasse 24–25, 14476 Golm (Germany)
Fax: (+49)331-977-5050
[
28]
tigated.
The measurements with the thermistor were
performed in a flow-through reactor containing the immobi-
lized enzyme. The signal/time response shows the typical form
reported for enzyme reactions (Figure 1) and is a result of the
E-mail: fschell@uni-potsdam.de
Dr. B. Danielsson
Department of Pure and Applied Biochemistry
Lund University, P.O. Box 124, 22100 Lund (Sweden)
[14]
steady-state heat generation by the catalytic reaction. No
peaks are found, either after substrate or buffer injection,
which indicates that no measurable heating takes place on
adsorption or desorption. This result is due to the hydrophilic
surface of the particles which arises from functionalization
with the esterase. A linear dependency of the steady-state
[
**] The authors would like to thank Dr. E. Yilmazfor his ideas regarding
the thermistor measurements and the IMPRS on Biomimetic
Systems as well as the BMBF (BioHyTec InnoRegio 03i1314) for
financial support.
Angew. Chem. Int. Ed. 2006, 45, 6986 –6990
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6987

Communications
Figure 1. Thermistor response for MIP-catalyzed reactions with differ-
ent phenylacetate concentrations or 5 mm phenol, and for immobilized
esterase with 5 mm phenylacetate. Substrate flow at 0 min and start of
washing at 20 min. 20 mm sodium phosphate buffer (pH 7.5)/10%
methanol.
Figure 2. Heat signals of the reactions of the bifunctional MIP and
NIP with 5 mm phenylacetate. Start of substrate flow at 0 min and
start of washing with buffer at 45 min. The red curve represents the
MIP and the black curve the NIP. Inset shows two different inter-
actions between the polymer and the substrate inside the MIP at
times between 25 and 45 min.
signal in the range of 0.5 to 5.0 mm phenylacetate was
observed. Analysis of the phenol concentration of the effluent
with a tyrosinase electrode indicated a conversion of almost
1
00%.
To test the polymers, a 500-mL reactor containing 150 mg
tration and in the flow system used is constantly washed out
the inhibition effect should be negligible. For 5.0 mm acetic
acid there was a spike for about 1 min which decreased very
quickly back to the baseline. In contrast, with phenylacetate
and phenol, the NIP shows less than 5% of the concentration-
dependant heat signals of the MIP (Figure 2).
of MIP or NIP was used. For low phenylacetate concentra-
tions, in analogy to the esterase reactor, the imprinted
polymer (Figure 1) shows a concentration-dependent plateau
over more than 20 min, which indicates a steady state of
catalytic reaction reaching saturation at 500 mm. The con-
version rate at this substrate concentration was about 20%.
After switching back to neat buffer, the signal decreases to the
baseline. With higher concentrations of up to 5 mm of
phenylacetate a peak can be observed in the first 10 min,
followed by a decay of the signal to a saturation value. The
subsequent buffer flow generates a negative peak which
slowly comes back to baseline. The signal was recorded over a
longer time for a concentration of 5.0 mm phenylacetate
From this behavior we can conclude that as both
polymers—the MIP and the NIP—are made from the same
bulk monomer composition the signals from the MIP are due
to an imprinting effect in the MIP. From the concentration
dependence monitored with this method it seems that there
are two different kinds of binding sites in the MIP. The steady
state behavior at low substrate concentrations reflects the
presence of a group of binding sites with higher affinity which
are capable of hydrolyzing phenylacetate to phenol and acetic
acid (Figure 2, inset). This catalysis gives rise to a concen-
tration-dependant heat signal over the time when the
substrate is pumped through the reactor. The peak formed
at high concentrations is caused by a second group of binding
sites which have a weaker affinity and are capable of
reversibly binding both phenylacetate and phenol (Figure 2,
inset). The signals arising from these binding sites are a result
of the adsorption and desorption of these aromatic species.
After a certain time of substrate supply, equilibrium of
binding is reached and the signal drops to the catalytically
generated heat level until the purging phase begins. A
negative signal is then observed which can be attributed to
the endothermic desorption process of the substrate.
(
Figure 2). After the initial peak, the signal remains at a
steady state up to the switch to buffer. This switch again
results in a negative peak.
To determine the cause of the initial peak at higher
concentrations, measurements with the products of the
reaction—phenol and acetic acid—were performed. For
5
.0 mm phenol a much sharper initial peak is observed than
for 5.0 mm phenylacetate, and the subsequent decrease to the
baseline is fast (Figure 1). Solvolysis products are often
described to interfere with the binding efficiency, therefore
the influence of phenol on the PA solvolysis was investi-
gated. Experiments with running buffer containing 5 mm
phenol and injections of different phenylacetate concentra-
tions resulted in catalysis that was still detectable but
decreased by 50%. Also the initial peak is still visible but
less intense and broadened. This decrease is due to a prior
occupation of binding sites with the high excess of product.
However, since the product is only formed in a low concen-
[
27]
Phenylacetate is converted by the catalytic sites with
saturation above 500 mm. With higher concentrations, a signal
caused by adsorption to the noncatalytic binding sites is
superimposed, which corresponds to the desorption during
washing and is also visible at concentrations higher than
6
988 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6986 –6990

Angewandte
Chemie
5
00 mm. Kirchner et al. mentioned a similar effect, where the
on two different types of recognition sets in the MIP, which
most perfect cavities were preferably filled at low analyte
concentrations, and less perfect cavities at higher concen-
trations, resulting in two different binding enthalpies.
With this knowledge, we can now determine apparent
kinetic data for both functional sites. For the catalytic site the
heat signal at 20 min prior to washing and for the binding site
the signal at 4 min is plotted against the substrate concen-
tration used (Figure 3). The calculated apparent kinetic data
are shown in Table 2.
can be distinguished by the differences in heat generation
using a flow-through thermistor. Thus, the MIP thermistor
allows measurements of two events: substrate conversion in
the catalytic sites and its adsorption at the binding sites. With
this method it will be possible to construct label-free multi-
analyte detectors and gain a closer insight into the interac-
tions between the polymer catalyst and the substrate which is
necessary to optimize the polymer synthesis for an efficient
tailor-made catalyst.
[
19]
In conclusion, enzyme-like catalysis and antibody-like
binding of a bifunctional MIP can be simultaneously resolved
by this method for the first time. The bifunctionality is based Experimental Section
[
20]
Polymers were prepared as described in Lettau et al. Samples were
dissolved in 20 mm sodium phosphate buffer (pH 7.5)/10% methanol
immediately before use. Flow and physical conditions for the
À1
thermistor measurements: 1.0 mLmin 20 mm sodium phosphate
buffer (pH 7.5)/10% methanol and a constant T of 303.15 K. Rate
constants were determined in 20 mm sodium phosphate buffer
(pH 7.5)/10% methanol.
Received: May 8, 2006
Revised: June 15, 2006
Keywords: biosensors · calorimetry · enzyme models ·
.
molecular recognition · solvolysis
[
[
[
1] G. Wulff, W. Vesper, R. Grobe-Einsler, A. Sarhan, Macromol.
Chem. Phys. 1977, 178, 2799 – 2816.
2] K. J. Shea, E. A. Thompson, S. D. Pandey, P. S. Beauchamp, J.
Am. Chem. Soc. 1980, 102, 3149 – 3155.
3] R. Arshady, K. Mosbach, Macromol. Chem. Phys. 1981, 182,
687 – 692.
[
[
4] S. A. Piletsky, A. P. F. Turner, Electroanalysis 2002, 14, 317 – 323.
5] F. L. Dickert, O. Hayden, P. Lieberzeit, C. Haderspoeck, R.
Bindeus, C. Palfinger, B. Wirl, Synth. Met. 2003, 138, 65 – 69.
6] I. Surugiu, L. Ye, E. Yilmaz, A. Dzgoev, B. Danielsson, K.
Mosbach, K. Haupt, Analyst 2000, 125, 13 – 16.
[
[
[
7] B. Sellergren, L. I. Andersson, Methods 2000, 22, 92 – 106.
8] K. Haupt, A. G. Mayes, K. Mosbach, Anal. Chem. 1998, 70,
3936 – 3939.
[
9] R. J. Ansell, K. Mosbach, Analyst 1998, 123, 1611 – 1616.
[
[
[
10] E. P. C. Lai, A. Fafara, V. A. VanderNoot, M. Kono, B. Polsky,
Can. J. Chem. 1998, 76, 265 – 273.
11] T. Kobayashi, Y. Murawaki, P. S. Reddy, M. Abe, N. Fujii, Anal.
Chim. Acta 2001, 435, 141 – 149.
12] F. L. Dickert, P. Lieberzeit, O. Hayden, Anal. Bioanal. Chem.
2003, 377, 540 – 549.
Figure 3. Michaelis–Menten plots of signals taken at a) 4 min and
b) 20 min. Squares and circles are the mean values of three measure-
ments for the MIP and the NIP, respectively, and error bars indicate
the standard deviations.
[13] A. Kugimiya, T. Takeuchi, Biosens. Bioelectron. 2001, 16, 1059 –
1062.
[14] K. Mosbach, B. Danielsson, Anal. Chem. 1981, 53, A83.
[15] B. Danielsson, L. Buelow, C. R. Lowe, I. Satoh, K. Mosbach,
Anal. Biochem. 1981, 117, 84 – 93.
[
[
[
[
[
16] I. Satoh in Chemical Sensor Technology, Vol. 2 (Ed.: T. Seiyama),
Elsevier, Amsterdam, 1989.
17] K. Haupt in Ultrathin Electrochemical Chemo- and Biosensors,
Vol. 2 (Ed.: V. M. Mirsky), Springer, Berlin, 2004, pp. 23 – 37.
18] A. Weber, M. Dettling, H. Brunner, G. E. M. Tovar, Macromol.
Rapid Commun. 2002, 23, 824 – 828.
19] R. Kirchner, J. Seidel, G. Wolf, G. Wulff, J. Inclusion Phenom.
Macrocyclic Chem. 2002, 43, 279 – 283.
20] K. Lettau, A. Warsinke, A. Laschewsky, K. Mosbach, E. Yilmaz,
F. W. Scheller, Chem. Mater. 2004, 16, 2745 – 2749.
Table 2: Apparent kinetic data in the flow.
[
a]
Signal
Cat.
DTmax
K
Steady-state
MIP
NIP
MIP
NIP
4.16Æ0.25 mK
322.17Æ57.88 mm
[
b]
[b]
n.d.
n.d.
Peak
15.72Æ0.94 mK
1836.28Æ187.76 mm
[
b]
[b]
n.d.
n.d.
[a] KM for steady-state signal; KD for peak signal. [b] Not determinable.
Angew. Chem. Int. Ed. 2006, 45, 6986 –6990
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 6989

Communications
[
[
[
[
[
[
[
[
21] E. Yilmaz, K. Haupt, K. Mosbach, Angew. Chem. 2000, 112,
178 – 2181; Angew. Chem. Int. Ed. 2000, 39, 2115 – 2118.
22] C. G. Overberger, J. C. Salamone, Acc. Chem. Res. 1969, 2, 217 –
24.
23] K. Ohkubo, Y. Urata, S. Hirota, Y. Honda, Y. Fujishita, T.
Sagawa, J. Mol. Catal. 1994, 93, 189 – 193.
24] K. Ohkubo, Y. Urata, S. Hirota, Y. Honda, T. Sagawa, J. Mol.
Catal. 1994, 87, L21 – L24.
25] J. M. Kim, K. D. Ahn, G. Wulff, Macromol. Chem. Phys. 2001,
2
2
202, 1105 – 1108.
26] J. Q. Liu, G. Wulff, Angew. Chem. 2004, 116, 1307 – 1311; Angew.
Chem. Int. Ed. 2004, 43, 1287 – 1290.
27] G. Wulff, T. Gross, R. Schꢀnfeld, Angew. Chem. 1997, 109, 2050 –
2052; Angew. Chem. Int. Ed. Engl. 1997, 36, 1962 – 1964.
28] G. T. Hermanson, Bioconjugate Techniques, Academic Press,
San Diego, 1996.
6
990 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 6986 –6990