Soluble single-molecule nanogels of controlled structure as a matrix for efficient artificial enzymes

Article abstract of DOI:10.1002/anie.200503926

(Figure Presented) Copy cats! Soluble single-molecule nanogels (see STEM micrograph; scale-bar = 20 nm) with a densely packed structure are prepared by a new method, and for the first time efficient molecular imprinting is possible. These nanogels with defined molecular weights of, for example, 40 000 Da, low polydispersities, and on average one active site per particle proved to be catalytically active enzyme mimics.

Full text of DOI:10.1002/anie.200503926

Angewandte
Chemie
Nanogels
to avoid macrogelation during the polymerization and to
produce separated, intramolecularly cross-linked macromo-
lecules. Such polymers can be obtained, for example, by
emulsion polymerization.[9] An elegant approach avoiding
aqueous solutions during preparation has been described by
Graham et al.[10] The polymerization is conducted in homo-
geneous solution at higher dilution in solvents with a
solubility parameter d similar to that of the corresponding
polymer. Earlier work in our group[11] has shown that it is
indeed possible to obtain molecularly imprinted, soluble,
highly cross-linked nanogels by this method; however, the
selectivity was rather low. Similar results were obtained by
Resmini et al.;[12] the catalytic activity of their nanogels was
also low with only 1% of the prepared sites being catalytically
active. They also did not control the size and the molecular
weight of the nanoparticles. In general, nanogels of this type
seem to be unsuitable for molecular imprinting because they
possess a noncompact “fractal”[13] structure.
DOI: 10.1002/anie.200503926
Soluble Single-Molecule Nanogels of Controlled
Structure as a Matrix for Efficient Artificial
Enzymes**
Günter Wulff,* Byong-Oh Chong, and Ute Kolb
Dedicated to Professor David C.Sherrington
on the occasion of his 60th birthday
For years, the construction of efficient artificial mimics of
natural enzymes has been a challenging topic for scientists
from a variety of disciplines.[1] Macrocyclic compounds,
polymers, micelles, and others have been used as matrices
for the construction of these mimics.[2] More recently, inspired
by the concept of transition-state stabilization in enzyme
catalysis[3] researchers have raised catalytically quite active
antibodies against stable transition-state analogues of the
reaction of interest.[4] Similarly, molecularly imprinted poly-
mers[5] offer excellent possibilities for mimicking the active
sites of enzymes since not only the shape of the transition
state can be mimicked by imprinting but also suitable catalytic
groups and binding sites can be introduced in a predeter-
mined orientation.[6] However, only recently have compounds
with strong catalytic activity clearly surpassing that of the
corresponding catalytic antibodies been prepared.[7] The
disadvantage of these molecularly imprinted model systems
is their insolubility and the heterogeneity of the active sites.
Here we report on a new procedure to obtain molecularly
imprinted catalysts in the form of soluble, single-molecule
nanogels of defined structure. As they contain just one active
site per particle, they are very similar to natural enzymes.
More homogeneous imprinting has already been achieved by
Zimmerman et al.,[8] who performed molecular imprinting
inside dendrimers. But this method requires elaborate
syntheses, and the selectivity in molecular recognition was
modest. A more general method should be the synthesis of
highly cross-linked, soluble nanogels. In this case it is crucial
During the development of more efficient nanogels we
used a rather simple catalytic system that was already
employed with insoluble, molecularly imprinted polymers.[14]
Imprinted nanogels were thus prepared by polymerization
and cross-linking of complex 1, which consists of a diphenyl
phosphate template as the stable transition-state analogue of
the carbonate hydrolysis and N,N’-diethyl-4-vinylbenzami-
dine as the functional monomer. Removal of the template
leads to active sites with a shape corresponding to the shape of
the transition state and with one amidine group each. The
ratio of the monomers and the content of cross-linker were
the same as that in the preparation of insoluble macroporous
polymers[14] (see Table 1). Radical polymerization was per-
formed in dilute solution in cyclopentanone (0.1–1.5 wt%).
(Cyclopentanone proved to be the best solvent for this
purpose.) The resultant nanogels could be isolated from the
completely transparent solution by precipitation with petro-
leum ether or by ultracentrifugation. The template was
removed afterwards. Since the nanogels are completely
soluble in THF, CHCl3, DMF, CH2Cl2, DMSO, and in
mixtures of water and acetonitrile, they could be investigated
by typical analytical methods for polymers, such as gel
permeation chromatography (GPC) and membrane osmom-
etry, and kinetic experiments could be conducted in homoge-
neous solution.
[*] Prof. Dr. G. Wulff, Dr. B.-O. Chong
Institute of Organic and Macromolecular Chemistry
Heinrich Heine University Düsseldorf
Universitätsstrasse 1, 40225 Düsseldorf (Germany)
Fax: (+49)211-811-5840
E-mail: wulffg@uni-duesseldorf.de
Dr. U. Kolb
Institute of Physical Chemistry
University of Mainz
Welderweg 11, 55099 Mainz (Germany)
[**] We thank Dr. P. Casper for the GPC measurements. This work was
supported by the Deutsche Forschungsgemeinschaft and the Fonds
der Chemischen Industrie. B.-O.C. acknowledges a scholarship
from the DFG Graduierten-Kolleg “Molecular physiology: Sub-
stance and energy transformation”.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2006, 45, 2955 –2958
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2955

Communications
Table 1: Preparation and molecular weights of imprinted nanogels ING1–ING7.[a]
Nanogel
cmonomer [%]
Method[b]
Yield [%]
Mn (”103)
Mw/Mn
Mn,abs. (”104)
Mn,abs./Mn
ING1
ING2
ING3
ING4
ING5
ING6
ING7
1.0
1.0
1.0
1.5
0.5
0.1
0.1
standard
A
53.6
81.3
24.5
99.5
8.91
1.34
31.4
23.1
29.5
6.0
42.2
3.6
8.5
10.4
62.4
18.6
26.1
4.43
1.54
38.1
49.2
25.5
92.4
29.3
33.0
3.90
16.5
16.7
B, C83.7
A, B, C, D
B, C61.1
B, C27.6
B, C, E
21.9
1.54
1.30
30.1
[a] Standard conditions: The monomer mixture consists of 80 wt% of ethylene dimethacrylate, 11 wt% of methyl methacrylate, 9 wt% of 1; addition of
azobis(isobutyronitrile) (3 wt% of the monomer mixture). The mixture was heated in cyclopentanone at the stated concentration under nitrogen to
808Cfor 4 d. After evaporation to one-third of the volume, the product was isolated by precipitation by adding to of five times the volume of petroleum
ether (b.p. 60–808C), ultrafiltration, and drying. The template was removed by dissolving the nanogel in chloroform and shaking the solution three
times with aqueous 0.05n NaOH solution. The organic layer was washed with neutral water, and the nanogel was isolated as before by precipitation.
Apparent molecular weights (Mn, Mw) were determined in THF by GPCusing linear polystyrenes as standards (for details see the Supporting
Information). Mn,abs. values were obtained by membrane osmometry in chloroform using regenerated cellulose membranes with pore sizes of 5 to
10 nm. [b] Variations: A) cross-linking content 90%. B) Stepwise polymerization at 608C(144 h), 70 8C(96 h), and 80 8C(96 h) with additional 1%
initiator at 708Cand 80 8C. C) In the “postdilution method” the monomer mixture was dissolved in the same volume of cyclopentanone and heated to
608Cfor 120 min. (This time was determined in advance from heating experiments until macrogelation occurred.) The mixture was then diluted with
cyclopentanone to the desired monomer concentration (0.1–1.5%) and polymerized further in the stepwise temperature mode. D) Cross-linker
trimethylolpropane trimethacrylate (TRIM). E) Template monomer 1 at half concentration (4.5%).
The preparation of the first nanogel (ING1, Table 1) was
analogous to previous syntheses.[10,11] ING1 has an apparent
number-average molecular weight Mn of 23100 (by GPC) and
a high polydispersity of Mw/Mn = 8.5. The GPC analysis
clearly shows that the nanogels are completely soluble.
Membrane osmometry was employed to obtain information
on the absolute number-average molecular weight (Mn,abs.) of
the nanogel. It can be seen that Mn,abs. is 16 times greater than
Mn (GPC). This is not surprising since highly cross-linked
nanogels dissolved in good solvents possess a much more
densely packed structure than the linear polystyrenes with the
same molecular weight used as standards in GPC. The ratio
Mn,abs./Mn therefore gives a good indication of the density of
the nanogels.
pseudo-first-order rate constants (kimpr.) were calculated. The
efficiency of the catalyst was expressed by the ratio of the rate
constant in the presence of the nanogel catalyst to that in
HEPES–AcCN solution (kimpr./ksol.). The ratio of the rate
constants in the presence of the imprinted nanogel to that of a
corresponding control nanogel (kimpr./kcontr.) reflects effects
connected to the imprinting in the nanogel (the imprinting
effect[6a]).
The catalytic activity of the nanogels was determined by
investigating the rate of hydrolysis of 2 [Eq. (1)] in HEPES
buffer (pH 7.3)/acetonitrile (1:1). The kinetic data is given in
Table 2. The kinetics were studied at low conversion, and
Control nanogels were prepared in analogy to the
imprinted ones (ING1–ING7), but instead of the template
diphenyl phosphate, formic acid was used (CNG1–CNG7; see
the Supporting Information). The standard nanogel ING1
shows clear catalytic activity, but it is lower than that of the
corresponding insoluble macroporous polymers.[14] To obtain
more rigid nanogels with better stabilization of the active site
by more efficient cross-linking, we employed new methods of
preparation (see the Supporting Information); some repre-
sentative data is shown in Tables 1 and 2.
Higher monomer concentrations and higher degrees of
cross-linking resulted in higher molecular weights and
considerably higher polydispersities Mw/Mn, but at the same
time the catalytic activity of the nanogels also increased. The
potential for increasing the monomer concentration is limited
because at concentrations above the critical concentration cm
(in the present case at around 1.5 wt%) macrogelation takes
place. A further considerable improvement of the properties
was achieved when the polymerization was performed in
three stages with increasing temperatures (60, 70, and 808C).
The most effective increase in the rigidity of the nanogel
structure was obtained by a novel method we have called the
“postdilution method”. It is based on the observation that the
Table 2: Catalytic activity of the imprinted nanogels ING1–ING7.
[a]
[b]
[c]
Nanogel kimpr.
kimpr./ksol.
kimpr./kcontr.
Available
cavities
Cavities
per
particle
(”10À7
)
]
[minÀ1
[mmolgÀ1 [d]
]
ING1
ING2
ING3
ING4
ING5
ING6
ING7
3.59
6.16
12.8
70.9
3.77
3.48
3.91
14.8
25.3
52.7
291.4
15.5
14.3
16.1
1.42
–
0.067
0.097
0.073
0.102
0.070
0.041
0.027
25.5
47.7
45.7
94.5
18.2
1.81
1.03
5.52
18.5
3.20
2.58
2.43
[a] Hydrolysis of 2 in a solution of 50 mm HEPES buffer (pH 7.3)/MeCN
(1:1) at 108Cwith 2 equiv of cavities to 1 equiv of substrate. (HEPES =2-
[4-(2-hydroxyethyl)-1-piperazine]ethanesulfonic acid). kimpr.: pseudo-first-
order rate constant in presence of the imprinted nanogel. [b] ksol. : rate
constant in HEPES buffer (pH 7.3)/MeCN (1:1). [c] kcontr. : rate constant
in presence of the control nanogel. [d] Available cavities with amidinium
groups determined by potentiometric titration in aqueous 0.1n NaCl/
MeCN (1:1) solution with 0.02n aqueous HCl.
2956 www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2955 –2958

Angewandte
Chemie
in the homopolymerization at high EDMA concentration the
cross-linking degree of the system reaches its maximum at a
point just before macrogelation takes place.[15] To avoid
macrogelation we stopped the polymerization at high con-
centration just prior to macrogelation and then diluted
extensively with cyclopentanone to keep the concentration
of the polymerization solution below cm (0.1–1.5 wt.%). In
this way further polymerization results in an increase of the
ratio Mn,abs./Mn, a measure of the density of the particles, from
around 16 to values of 26–33 with EDMA as cross-linker (see
Table 1). At the same time the polydispersity became lower,
and most importantly, the catalytic activity was greatly
improved.
weight contain, on average, a higher number of active sites
per particle. To obtain nanogels with an average of one active
site per particle, we reduced the amount of template
monomer 1 in the preparation of ING7. Thus we obtained
for the first time soluble nanogels with molecular weights
similar to those of natural enzymes, around 40 kDalton, very
low polydispersity, and on average only one active site per
particle.
We visualized and characterized these nanogels by scan-
ning transmission electron microscopy (STEM).[16] ING6 was
prepared without drying the material in order to avoid any
additional aggregation and was subsequently analyzed. RuO4-
stained particles of ING6 in dilute chloroform solution were
applied to a STEM copper grid covered by fenestrated carbon
film. The STEM overview picture in Figure 1a shows a
Since the trifunctional cross-linker trimethylolpropane-
trimethacrylate (TRIM) had a positive influence on the
catalytic properties of the nanogels, it was used in the
preparation of ING4. Here, all possibilities for improvement
were combined, and a very efficient catalyst system was
obtained with the highest kimpr./ksol value of 291 and a kimpr.
/
kcontr. value of 18.5. Like natural enzymes, ING4 shows typical
Michaelis–Menten kinetics for the catalysis of carbonate
hydrolysis. A plot of initial reaction rates versus the substrate
concentration shows saturation kinetics (see the Supporting
Information). From this data the Michaelis constant Km =
1.82 mm and the turnover number kcat = 7.27 ” 10À5 minÀ1
can be calculated. The kcat./kuncat. ratio reached the remarkable
value of 2990. A control polymer (CNG4) showed consid-
erably lower catalytic activity, and no saturation phenomena
were observed (nearly a straight line in the plot in Figure SI-
2).
These results clearly indicate that the catalysis is occurring
inside the active site. Thus, it was possible for the first time to
prepare a soluble imprinted nanogel with catalytic activity
similar to that of insoluble imprinted polymers. A further
improvement due to the solubility of the catalysts was not
observed. In the preparation of the nanogels the yields and
the number of obtained active sites were nearly quantitative.
This procedure, therefore, should also be suitable for more
sophisticated and efficient catalytic systems with rate
enhancements of more than 100000 fold, as has been
described previously.[7] In addition it should also be applicable
to the frequently used noncovalent imprinting protocols
based on weaker interactions[5b] since during the imprinting
no water or other protic solvents are present.
After obtaining nanogels with high rigidity and good
catalytic activity, we tried again to reduce the heterogeneity
by controlling the molecular weight and the polydispersity of
the catalysts. This could be achieved by using higher dilution
during the preparation in the final step of the “postdilution
method”. ING3, ING5, and ING6 were prepared under
identical conditions, except the dilution was 1.0, 0.5, and
0.1%. As can be seen in Table 1 the Mn,abs. dropped from 624
to 261 and to 44.3 kDalton. At the same time, the polydis-
persity Mw/Mn dropped from 6.0 to 3.6 and 1.54. The reason is
a much lower aggregation of the primary particles. A
polydispersity of 1.54 is an extremely good value for a radical
polymerization. A calculation based on the experimental
results for ING6 show the existence of on average 1.8 active
sites per particle. As expected, particles of higher molecular
Figure 1. STEM pictures of ING6. a) Overview: RuO4-stained nano-
particles (scale bar: 200 nm). The size distribution is small. b) At
higher magnification the structure of the spherical particles can be
seen (scale bar: 20 nm). The diameter of particles of 10–20 nm can be
distinguished. The investigation was performed with a Philips Tecnai
F30 analytical TEM instrument in STEM mode through a “high-angular
annular dark field” detector (HAADF). The presence of Ru was
confirmed through energy-dispersive X-ray spectroscopy with an
Oxford Pentafet (Si/Li) EDX detector. Ru is homogeneously distributed
on the nanogel and forms nanosized clusters on the surface of the
polymer. Through a tilt series in the measurement of approximately
Æ408 the shape and the size of the particles did not change.
narrow size distribution, and at higher magnification (Fig-
ure 1b) spherical particles with diameters between 10 and
20 nm can be clearly seen. These particles represent single,
intramolecularly cross-linked macromolecules that do not
possess a marked fractal structure. The particles are rigid and
do not collapse onto the support film. With the available data
still no safe conclusions about the fine structure of the
nanogels can be obtained from these images.
Since these soluble particles of 40 kDalton can be
prepared, on average, with one active catalytic site per
particle and show Michaelis–Menten kinetics, a high analogy
to natural enzymes was achieved. These catalysts can be
analyzed and handled like enzymes but they are by far more
stable. As with enzymes, it should also be possible to purify
and enrich these nanogels by enzyme methodology, for
example, by affinity chromatography. In addition to their
importance as enzyme mimics, these novel soluble nanogels
combining low molecular weight and low polydispersty with
Angew. Chem. Int. Ed. 2006, 45, 2955 –2958
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2957

Communications
densely packed structures might find widespread use in
different areas of science and technology.
Received: November 5, 2005
Revised: January 22, 2006
Published online: March 28, 2006
Keywords: enzyme models · homogeneous catalysis ·
.
molecular imprinting · nanogels · reaction kinetics
[1] For reviews see: a) A. J. Kirby, Angew.Chem. 1996, 108, 770 –
790; Angew.Chem.Int.Ed.Engl.
1996, 35, 707 – 724; b) W. B.
Motherwell, M. J. Bingham, Y. Six, Tetrahedron 2001, 57, 4663 –
4686.
[2] a) D. J. Cram, Nature 1992, 356, 1009 – 1020; b) R. Breslow, S. D.
Dong, Chem.Rev. 1998, 98, 1997 – 2011; c) C. Liang, J. M. J.
FrØchet, Prog.Polym.Sci. 2005, 30, 385 – 402.
[3] W. Jencks, Catalysis in Chemistry and Enzymology, McGraw-
Hill, New York, 1969.
[4] a) R. A. Lerner, S. J. Benkovic, P. G. Schultz, Science 1991, 252,
659 – 667; b) D. Hilvert, Annu.Rev.Biochem. 2000, 69, 751 – 793.
[5] For reviews see: a) G. Wulff, Angew.Chem. 1995, 107, 1958 –
1979; Angew.Chem.Int.Ed.Engl.
1995, 34, 1812 – 1832;
b) Molecularly Imprinted Materials.Science and Technology
(Eds.: M. Yan, O. Ramstroem), Marcel Dekker, New York, 2005.
[6] For reviews, see: a) G. Wulff, Chem.Rev. 2002, 102, 1 – 28;
b) M. E. Davis, A. Katz, W. R. Ahmad, Chem.Mater. 1996, 8,
1820 – 1839.
[7] a) J.-Q. Liu, G. Wulff, Angew.Chem. 2004, 116, 1307 – 1311;
Angew.Chem.Int.Ed.
2004, 43, 1287 – 1290; b) J.-Q. Liu, G.
Wulff, J.Am.Chem.Soc. 2004, 126, 7452 – 7453.
[8] S. J. Zimmerman, M. S. Wendland, N. A. Rakow, I. Zharov, K. S.
Suslick, Nature 2002, 418, 399 – 403.
[9] W. Funke, O. Okay, B. Joos-Müller, Adv.Polym.Sci. 1998, 136,
139 – 234.
[10] N. B. Graham, A. Cameron, Pure Appl.Chem. 1998, 70, 1271 –
1275.
[11] A. Biffis, N. B. Graham, G. Siedlaczek, S. Stalberg, G. Wulff,
Macromol.Chem.Phys. 2001, 2002, 163 – 171.
[12] S. C. Maddock, P. Pasetto, M. Resmini, Chem.Commun. 2004,
536 – 537.
[13] M. Antonietti, C. Rosenauer, Macromolecules 1991, 24, 3434 –
3442.
[14] A. G. Strikovsky, D. Kaspar, M. Grün, B. S. Green, J. Hradil, G.
Wulff, J.Am.Chem.Soc. 2000, 122, 6295 – 6296.
[15] Y. Y. Chiu, L. J. Lee, J.Polym.Sci.Part A
1995, 33, 257 – 267.
[16] a) Z. L. Wang, Adv.Mater. 2003, 15, 1497 – 1514; b) R. Xie, U.
Kolb, J. Li, T. BaschØ, A. Mews, J.Am.Chem.Soc. 2005, 127,
7480 – 7488; B. Rückert, U. Kolb, MICRON 2005, 36, 247 – 260.
2958
www.angewandte.org
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2006, 45, 2955 –2958