Homotropic cooperativity of cyclodextrin dimer as an artificial hydrolase

Article abstract of DOI:10.1016/j.molcata.2010.06.005

Three kinds of cyclodextrin (CD) dimer (one homo-dimer of β-CD and two types of hetero-dimer of α-CD and β-CD) were synthesized as artificial hydrolases. The initial rates of the cleavage reaction of p-nitrophenyl methoxyethoxyethoxyacetate (8) by each of the three dimers were measured under the condition of excess of substrate at 25 °C in a pH 7.4 phosphate buffer solution. In a plot of initial rate against concentration of the substrate, the reaction with the homo-dimer (βCβH 5) gave a sigmoidal curve, whereas the reaction with either of the hetero-dimers gave a simple hyperbolic curve. Only βCβH 5 showed homotropic cooperativity, with a Hill constant of 1.8.

Full text of DOI:10.1016/j.molcata.2010.06.005

Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Editor’s Choice paper
Homotropic cooperativity of cyclodextrin dimer as an artificial hydrolase
Hiroshi Ikeda∗, Satoshi Nishikawa, Yukinori Yamamoto, Akihiko Ueno
Department of Bioengineering, Graduate School of Bioscience and Biotechnology, Tokyo Institute of Technology, 4259-B44 Nagatsuta-cho, Midori-ku, Yokohama 226-8501, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Three kinds of cyclodextrin (CD) dimer (one homo-dimer of ␤-CD and two types of hetero-dimer of
␣-CD and ␤-CD) were synthesized as artificial hydrolases. The initial rates of the cleavage reaction of
p-nitrophenyl methoxyethoxyethoxyacetate (8) by each of the three dimers were measured under the
condition of excess of substrate at 25 ◦C in a pH 7.4 phosphate buffer solution. In a plot of initial rate
against concentration of the substrate, the reaction with the homo-dimer (␤C␤H 5) gave a sigmoidal
curve, whereas the reaction with either of the hetero-dimers gave a simple hyperbolic curve. Only ␤C␤H
5 showed homotropic cooperativity, with a Hill constant of 1.8.
Received 2 March 2010
Received in revised form 3 June 2010
Accepted 4 June 2010
Available online 11 June 2010
Keywords:
Cooperative effects
Cyclodextrins
© 2010 Elsevier B.V. All rights reserved.
Enzyme models
Host–guest systems
Inclusion compounds
1. Introduction
homotropic allosteric enzyme changes its tertiary structure on sub-
strate binding, which affects the structure and chemical activity of
Natural enzymes have mechanisms to control their activities
in response to the concentration of substrates, products or other
effectors [1,2]. In contrast, there have not been many reports of
artificial enzymes that have mechanisms to control their activities,
although many successful artificial enzymes with high catalytic
activity or high substrate selectivity have been reported [3–15].
Thus, an important area of research in the field of supramolecular
chemistry is the development of chemical methods to control the
activities of artificial enzymes [16,17].
Allosteric regulation is a sophisticated way in which the activity
of an enzyme can be controlled [18]. Some artificial heterotropic
allosteric enzymes, with a mechanism to control their activity
in response to an effector, have been reported, but few artificial
homotropic allosteric enzymes, with a mechanism to control their
activity towards the substrate itself, have been described [19,20].
Allosteric enzymes are made up of several subunits and are classi-
fied as either heterotropic or homotropic enzymes. A heterotropic
allosteric enzyme has a binding site for an effector as well as a
substrate-binding site. Binding of an effector to the effector-binding
site in the subunit of a heterotropic allosteric enzyme changes the
affinity of the substrate-binding site in the neighboring subunit. By
contrast, a homotropic allosteric enzyme consists of some identical
subunits with identical binding sites located in symmetrical posi-
tions. Binding of a first substrate to a homotropic allosteric enzyme
facilitates the binding of a second substrate. Each subunit of a
its neighboring subunits. This structural change causes a sigmoidal
dependence of the reaction rate on the substrate concentration.
The Monod–Wyman–Changeux (MWC) model provides a sim-
ple description of allosteric proteins [18]. This model minimizes the
number of intermediate states, and thus is only an approximation
of reality; however, this simplification is also its virtue. The MWC
model provides a simple framework to rationalize experiments or
explain phenomena and can also assist in the molecular design of
artificial allosteric proteins. The MWC model makes the follow-
ing assumptions. (1) Allosteric proteins are oligomers composed
of identical monomers (or ‘protomers’) that are associated in such
a way that they all occupy equivalent positions. (2) Only one site on
each protomer binds to each ligand that is able to form a stereospe-
cific complex. (3) The protein exists in either of two conformational
states, the T or R state. The two states are in equilibrium, but the
T state predominates when the protein is unliganded. They differ
in the energies and numbers of bonds between the subunits, with
the T state being more constrained than the R state. (4) The T state
has a lower affinity for ligands than the R state. (5) All binding sites
in each state are equivalent and have identical binding constants.
When the protein switches from one state to another, its molecular
symmetry is conserved (the symmetry assumption).
The MWC model for a dimeric enzyme is illustrated in Fig. 1.
Its homotropic activity can be explained by an equilibrium argu-
ment. In the absence of a substrate the enzyme is mainly in the
T state; however, the conformational equilibrium is shifted to the
right (the R state) when the substrate binds to the enzyme, because
the binding ability of the R state is larger than that of the T state.
This structural change induced by the substrate binding causes
∗
Corresponding author.
E-mail address: hikeda@bio.titech.ac.jp (H. Ikeda).
1381-1169/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2010.06.005

2
H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
2.2.1. Synthesis of 6-carboxymethylthio-6-deoxy-˛-cyclodextrin,
˛C, 1
Methyl sodium sulfidoacetate (2.27 g, 17.7 mmol) was added to
a solution of 6-deoxy-6-iodo-␣-cyclodextrin (6.33 g, 5.85 mmol) in
DMF (90 mL). The resulting mixture was stirred at 80 ◦C for 72 h.
After being cooled to room temperature, the reaction mixture was
poured into acetone, and the precipitate was collected and dried in
vacuo. A solution of the crude product in water (30 mL) was mixed
with 1 M NaOH (10 mL), and was stirred at 0 ◦C for 1 h. After being
warmed to room temperature, the reaction mixture was charged
on SP Sephadex C-25 (H+ form). The eluent was charged on QAE
Sephadex A-25 (OH− form). The eluent containing pure product
was obtained with a linear gradient of 0–0.5 M of NH4HCO3 aque-
ous solution (2.20 g, 36% yield). 1H NMR (500 MHz, D2O, 25 ◦C):
ı = 2.84 (dd, 1H, S-CH2(6b)), 3.12 (dd, 1H, S-CH2(6a)), 3.27 (d, 1H,
S-CH2-CO), 3.34 (d, 1H, S-CH2-CO), 4.96–5.06 ppm (m, 6H, CH(1));
MS (TOFMS, m/z): [M+Na]+ calcd 1069.3, found 1069.2; Anal. calcd
for C38H62O31S.3H2O: C 41.45, H 6.23, S 2.91. Found: C 41.21, H
6.52, S 2.66.
2.2.2. Synthesis of 6-carboxymethylthio-6-deoxy-ˇ-cyclodextrin,
ˇC, 2
Fig. 1. MWC model for an allosteric enzyme.
This compound was prepared by the condensation of 6-deoxy-
6-iodo-␤-cyclodextrin (13.4 g, 10.8 mmol) and methyl sodium
sulfidoacetate (4.15 g, 32.4 mmol) by the same method used for
1 (8.22 g, 63% yield). 1H NMR (500 MHz, D2O, 25 ◦C): ı = 2.85 (dd,
1H, S-CH2(6b)), 3.08 (dd, 1H, S-CH2(6a)), 3.34 (s, 2H, S-CH2-CO),
4.97–5.11 ppm (m, 7H, CH(1)); MS (TOFMS, m/z): [M+Na]+ calcd
1231.3, found 1231.3; Anal. calcd for C44H72O36S.5H2O: C 40.68, H
6.36, S 2.47. Found: C 40.94, H 6.61, S 2.64.
sigmoidal dependence of the initial rate on the substrate concen-
tration.
In terms of the MWC model, a host molecule that has two equal-
sized binding sites at the symmetrical positions is expected to show
an allosteric effect, if the binding of the first substrate to a binding
site can improve the affinity for the second substrate. We previ-
ously prepared a cyclodextrin (CD) dimer (designated ␤C␤H) that
has a catalytic site located between two ␤-cyclodextrins (␤-CDs)
and acts as an artificial hydrolase [21]. ␤C␤H shows a large acceler-
ation ability and substrate specificity for the acyl chain length of the
substrate, when p-nitrophenyl alkanoates are used as the substrate.
This result indicates that there is some cooperativity between the
two CD cavities of the CD dimer. New water-soluble substrates and
two kinds of CD hetero-dimers were prepared and the catalytic
activities of the CD homo-dimer and two kinds of CD hetero-dimers
were studied. Here we have investigated the homotropic allosteric-
like effect of ␤C␤H, in comparison with cyclodextrin hetero-dimers
that have a catalytic site located between a ␤-CD unit and an ␣-CD
unit.
2.2.3. Synthesis of 6-deoxy-6-(l-histidylamino)-˛-cyclodextrin,
˛H, 3
N␣-tert-Butyloxycarbonyl-Nim-tosyl-l-histidine
(3.30 g,
a solution of 6-amino-6-deoxy-␣-
8.06 mmol) was added to
cyclodextrin (3.45 g, 3.55 mmol) in DMF (20 mL). After the solution
was cooled below 0 ◦C, N,N-dicyclohexylcarbodiimide (900 mg,
4.36 mmol) and 1-hydroxybenzotriazole (343 mg, 2.54 mmol)
were added. This resulting mixture was stirred at 0 ◦C for 2 h and
then at room temperature for 2 days. The insoluble materials
were removed by filtration, the filtrate was poured into acetone,
and the precipitate was collected and dried in vacuo. The crude
product was dissolved in a mixed solvent of methanol/water
(5:1) and was refluxed for 2 h. The reaction mixture was poured
into acetone, and the precipitate was collected and dried in
2. Experimental
vacuo. The crude product, 6-(N␣-tert-butyloxycarbonyl-Nim
-
2.1. General
tosyl-l-histidylamino)-6-deoxy-␣-cyclodextrin, was dissolved
with DMF (10 mL) and cooled to 0 ◦C. This solution was reacted
with 1 M NaOH aqueous solution (5 mL) at 0 ◦C for 1 h. The
reaction mixture was neutralized with 1 M HCl aqueous solu-
tion and poured into acetone, and then the precipitate was
collected and dried in vacuo. The crude product, 6-(N␣-tert-
butyloxycarbonyl-l-histidinylamino)-6-deoxy-␣-cyclodextrin,
was dissolved with 30 mL of CF3COOH/CH2Cl2 (9:1) and stirred
at 0 ◦C for 1 h. After evaporation of CF3COOH and CH2Cl2, the
crude product was dissolved in water (1 L) and charged on
SP Sephadex C-25 (H+ form). After washing with water (5 L),
the crude product was eluted with 1 M NH3 aqueous solution
(750 mL). The crude product was purified by column chro-
matography on CM-Sephadex C-25 (NH4+ form) with a linear
gradient of 0–0.09 M of NH4HCO3 aqueous solution. The elu-
ent containing the pure product was concentrated to give the
desired product (1.30 g, 33% yield). 1H NMR (500 MHz, D2O,
25 ◦C): ı = 2.85 (m, 2H, Im-CH2), 3.04 (t, 1H, CH(4ꢀ)), 4.94–5.02
(m, 6H, CH(1)), 6.90 (s, 1H, Im), 7.66 ppm (s, 1H, Im); MS (TOFMS,
m/z): [M+Na]+ calcd 1131.4, found: 1131.3; Anal. calcd for
Thin-layer chromatography (TLC) was carried out with silica gel
60 F254 (Merck Co.). Absorption spectra were recorded on a Shi-
madzu UV-3100 spectrometer. NMR spectra were recorded in D2O
at 25 ◦C on Varian VXR-500S and UNITY plus-400 spectrometers
operating at 499.843 MHz and 399.973 MHz, respectively for 1H.
HDO (ı = 4.70 ppm) and CDCl3 (ı = 7.26 ppm) were used as inter-
nal standards. Mass spectrometry was performed on a Shimadzu
MALDI III mass spectrometer (TOFMS). Elemental analyses were
performed by the Analytical Division in Research Laboratory of
Resources Utilization of Tokyo Institute of Technology.
2.2. Materials
␣- and ␤-Cyclodextrin were kind gifts from Nihon Shokuhin
Kako Co., Ltd. All chemicals were of reagent grade and were used
without further purification unless otherwise noted. Distilled water
and acetonitrile used as solvents for spectroscopy were special flu-
orometry grade (Uvasol) from Kanto Chemicals. Deuterium oxide,
with an isotopic purity of 99.95%, was purchased from Merck Co.

H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
3
C42H68O30N4.3H2O: C 43.37, H 6.41, N 4.82. Found: C 43.64, H 6.66,
N 4.91.
formed in the reaction mixture was filtered off. After evaporation
of the solvent, the product was purified with a column chromatog-
raphy on silica gel (elution with chloroform/ethyl acetate (10:1))
to give the desired product (9.25 g, 86% yield). 1H NMR (500 MHz,
CDCl3, 25 ◦C): ı = 3.38 (s, 3H, OCH3), 3.56 (m, 2H, OCH2), 3.66 (m,
2H, OCH2), 3.74 (m, 2H, OCH2), 3.84 (m, 2H, OCH2), 4.46 (s, 2H,
COCH2), 7.33 (d, 2H, Ar), 8.28 ppm (d, 2H, Ar); Anal. calcd for
2.2.4. Synthesis of 6-deoxy-6-(l-histidylamino)-ˇ-cyclodextrin,
ˇH, 4
This compound was prepared from 6-amino-6-deoxy-␤-
cyclodextrin (4.50 g, 3.97 mmol) by the same method used for
3 (1.82 g, 36% yield). 1H NMR (500 MHz, D2O, 25 ◦C): ı = 2.78
(m, 2H, Im-CH2), 3.10 (t, 1H, H(4ꢀ)), 3.23 (dd, 1H, NH-CH2(6b)),
4.92–5.00 (m, 7H, CH(1)), 6.77 (s, 1H, Im), 7.57 ppm (s, 1H, Im); MS
(TOFMS, m/z): [M+Na]+ calcd 1293.4, found: 1293.4; Anal. calcd for
C48H78O35N4.5H2O: C 42.35, H 6.52, N 4.12. Found: C 42.25, H 6.24,
N 4.07.
C13H17O7N: C 52.17, H 5.73, N 4.68. Found: C 52.15, H 5.70, N 4.68.
2.2.9. Synthesis of m-nitrophenyl 2-[2-(2-methoxyethoxy)
ethoxy]acetate, 9
m-Nitrophenol (5.0 g, 35.9 mmol), 2-[2-(2-methoxyethoxy)
ethoxy]acetic
acid
(8.3 g,
46.6 mmol),
and
1,3-
dicyclohexylcarbodiimide (14.8 g, 71.7 mmol) were reacted by
the same method used for 8 (9.0 g, 84% yield). 1H NMR (500 MHz,
CDCl3, 25 ◦C): ı = 3.39 (s, 3H, OCH3), 3.57 (m, 2H, OCH2), 3.68 (m,
2H, OCH2), 3.74 (m, 2H, OCH2), 3.85 (m, 2H, OCH2), 4.43 (s, 2H,
COCH2), 7.49 (m, 1H, Ar), 7.57 (t, 1H, Ar), 8.02 (t, 1H, Ar), 8.12 ppm
(m, 1H, Ar); Anal. calcd for C13H17O7N: C 52.17, H 5.73, N 4.68.
Found: C 52.16, H 5.74, N 4.66.
2.2.5. Synthesis of 6-{N-[6-(6-deoxy-ˇ-cyclodextrinylthio)
acetyl]-l-histidylamino}-6-deoxy-ˇ-cyclodextrin, ˇCˇH, 5
N,N-Dicyclohexylcarbodiimide (80.3 mg, 0.389 mmol) and 1-
hydroxybenzotriazole (52.7 mg, 0.390 mmol) were added to a
solution of 6-deoxy-6-(l-histidylamino)-␤-cyclodextrin (450 mg,
0.354 mmol) and 6-carboxymethylthio-6-deoxy-␤-cyclodextrin
(855 mg, 0.707 mmol) in DMF (10 mL) at 0 ◦C. This solution was
stirred at 0 ◦C for 2 h and then at room temperature for 48 h. The
insoluble materials were removed by filtration, the filtrate was
poured into acetone, and the precipitate was collected and dried
in vacuo. The crude product was purified by column chromatogra-
phies on CM-Sephadex C-25 (H+ form and NH4+ form) to give
the desired product (573 mg, 66% yield). 1H NMR (500 MHz, D2O,
25 ◦C): ı = 2.71 (dd, 1H, S-CH2(6b)), 2.90 (dd, 1H, S-CH2(6a)), 2.96
(dd, 1H, Im-CH2), 3.03 (dd, 1H, Im-CH2), 3.19 (t, 1H, CH(4ꢀ)), 3.29 (d,
1H, S-CH2-CO), 3.34 (d, 1H, S-CH2-CO), 3.37 (dd, 1H, NH-CH2(6b)),
4.90–5.04 (m, 14H, CH(1)), 6.92 (s, 1H, Im), 7.67 ppm (s, 1H, Im); MS
(TOFMS, m/z): [M+Na]+ calcd 2483.8, found: 2483.7; Anal. calcd for
2.2.10. Synthesis of p-nitrophenyl {2-[2-(2-methoxyethoxy)
ethoxy]ethyl} ether, 10
2-[2-(2-Methoxyethoxy)ethoxy]ethyl p-toluensulfonate (5.7 g,
17.9 mmol), p-nitrophenol (2.49 g, 17.9 mmol), and potassium car-
bonate (4.96 g, 35.9 mmol) were refluxed in acetone (130 mL) for
24 h. After potassium carbonate was filtered off and the solvent was
evaporated, the product was purified with a column chromatog-
raphy on silica gel (elution with chloroform/ethyl acetate (7: 1))
to give the desired product (7.08 g, 90% yield). 1H NMR (500 MHz,
CDCl3, 25 ◦C): ı = 3.38 (s, 3H, OCH3), 3.55 (m, 2H, OCH2), 3.65 (m, 2H,
OCH2), 3.68 (m, 2H, OCH2), 3.89 (m, 2H, OCH2), 4.22 (s, 2H, COCH2),
6.98 (d, 2H, Ar), 8.20 ppm (d, 2H, Ar); Anal. calcd for C13H19O6N: C
54.73, H 6.71, N 4.91. Found: C 54.72, H 6.71, N 4.91.
C92H148O70N4S.15H2O: C 40.44, H 6.57, N 2.05, S 1.17. Found: C
40.36, H 6.54, N 1.98, S 1.13.
2.3. Kinetics
2.2.6. Synthesis of 6-{N-[6-(6-deoxy-˛-cyclodextrinylthio)
acetyl]-l-histidylamino}-6-deoxy-ˇ-cyclodextrin, ˛C␤H, 6
Hydrolysis reactions were followed by monitoring the appear-
ance of p-nitrophenol or m-nitrophenol spectrophotomerically
using a Shimadzu UV-3100 spectrometer. The reaction was con-
ducted in a quartz cell in the water-jacketed cell holder of UV-3100.
Temperature was maintained at 25 ◦C by a HAAKE F3 circulating
water bath. The reaction was initiated by adding a stock solution of
an ester in acetonitrile to a buffer solution in the quartz cell using
a HAMILTON microliter syringe. The total amount of acetonitrile in
the cell was adjusted to be same in all runs by adding an amount of
acetonitrile before addition of the substrate. The pH of the reaction
mixture did not change during the course of the reaction. The rates
used in the calculation of rate constants were averages of at least
three determinations that agreed within 3%.
This compound was prepared by the condensation of 1 (489 mg,
0.467 mmol) and 4 (300 mg, 0.236 mmol) by the same method
used for 5 (419 mg, 77% yield). 1H NMR (500 MHz, D2O, 25 ◦C):
ı = 2.73 (dd, 1H, S-CH2(6b)), 3.01 (dd, 1H, S-CH2(6a)), 2.95 (dd,
1H, Im-CH2), 3.03 (dd, 1H, Im-CH2), 3.17 (t, 1H, CH(4ꢀ)), 3.30 (d,
1H, S-CH2-CO), 3.36 (d, 1H, S-CH2-CO), 3.39 (dd, 1H, NH-CH2(6b)),
4.89–5.04 (m, 13H, CH(1)), 6.92 (s, 1H, Im), 7.67 ppm (s, 1H, Im); MS
(TOFMS, m/z): [M+Na]+ calcd 2321.7, found: 2321.9; Anal. calcd for
C86H138O65N4S.14H2O: C 40.47, H 6.56, N 2.20, S 1.26. Found: C
40.44, H 6.30, N 2.26, S 1.44.
2.2.7. Synthesis of 6-{N-[6-(6-deoxy-ˇ-cyclodextrinylthio)
acetyl]-l-histidylamino}-6-deoxy-˛-cyclodextrin, ˇC˛H, 7
This compound was prepared by the condensation of 2 (436 mg,
0.361 mmol) and 3 (300 mg, 0.271 mmol) by the same method used
for 5 (439 mg, 71% yield). 1H NMR (500 MHz, D2O, 25 ◦C): ı = 2.69
(dd, 1H, S-CH2(6b)), 2.91 (dd, 1H, S-CH2(6a)), 2.94 (dd, 1H, Im-CH2),
3.05 (dd, 1H, Im-CH2), 3.12 (t, 1H, CH(4ꢀ)), 3.29 (d, 1H, S-CH2-CO),
3.34 (d, 1H, S-CH2-CO), 4.85–5.02 (m, 13H, CH(1)), 6.96 (s, 1H,
Im), 7.66 ppm (s, 1H, Im); MS (TOFMS, m/z): [M+Na]+ calcd 2321.7,
found: 2321.6; Anal. calcd for C86H138O65N4S.10H2O: C 41.65, H
6.42, N 2.26, S 1.29. Found: C 41.70, H 6.12, N 2.27, S 1.06.
3. Results and discussion
3.1. Catalytic activities of artificial enzymes (4–7)
In our previous results, cooperativity between the two CD cavi-
ties of the CD homo-dimer (␤C␤H) was implied by the dependence
of the reactivity of the homo-dimer on the acyl chain length of the
substrate in the cleavage reaction of p-nitrophenyl alkanoates [21],
however, the cooperativity of this reactivity with the alkanoate
esters could not be examined in detail owing to the low solubility
of the alkanoate esters in water. We therefore synthesized a new
substrate ester 8 (Chart 1) that is more soluble in water, and stud-
ied the catalytic activity of the CD homo-dimer (␤C␤H) under the
condition of high substrate concentration. We also synthesized two
kinds of hetero-dimer (␣C␤H 6 and ␤C␣H 7) by the same method
2.2.8. Synthesis of p-nitrophenyl 2-[2-(2-methoxyethoxy)
ethoxy]acetate, 8
p-Nitrophenol (5.0 g, 35.9 mmol), 2-[2-(2-methoxyethoxy)
ethoxy]acetic
acid
(8.3 g,
46.6 mmol),
and
1,3-
dicyclohexylcarbodiimide (14.8 g, 71.7 mmol) were reacted in
tetrahydrofuran (100 mL) at 0 ◦C overnight. A precipitate that

4
H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
Chart 1. New water-soluble substrates and potential inhibitor for hydrolysis reac-
tion.
Fig. 2. Dependence of the initial rate on the substrate concentration for the cleavage
of the substrate 8 by ␤C␤H 5 (4.8 × 10−5 M) in pH 7.4 phosphate buffer (0.1 M) at
25 ◦C.
Fig. 3. Dependence of the initial rate on the substrate concentration for the cleavage
of the substrate 8 by ␤H 4 (4.8 × 10−5 M) in pH 7.4 phosphate buffer (0.1 M) at 25 ◦C.
the homo-dimer ␤C␤H 5 against substrate concentration gave a
sigmoidal curve, as shown in Fig. 2, whereas the plot of the ini-
tial rate of the cleavage reaction by ␤H 4 gave a simple hyperbolic
curve, as shown in Fig. 3. Thus, the observed sigmoidal behaviour
is due to the reactivity of ␤C␤H 5 rather than to that of the ester
8, because ␤H 4 shows simple saturation behaviour for the same
substrate.
Scheme 1. Syntheses of the homo-dimer (␤C␤H 5) and the hetero-dimers (␣C␤H
6 and ␤C␣H 7).
used for the homo-dimer (Scheme 1) and their reactivities were
compared with the homo-dimer (␤C␤H 5).
The initial rates for the cleavage reaction of the p-nitrophenyl
ester 8 catalyzed by a CD monomer (␤H 4) and the CD homo-dimer
(␤C␤H 5) were measured under the condition of excess substrate
(a 5- to 50-fold excess over the artificial enzymes) at 25 ◦C in a pH
7.4 phosphate buffer solution. The reaction was followed by mon-
itoring the appearance of p-nitrophenolate at 400 nm. The plot of
the initial rate of the cleavage reaction of the substrate ester 8 by
The kinetic data for ␤C␤H 5 were fitted to the Hill equation (Eq.
(1)) [22,23] as shown in Fig. 2, to give a Hill constant of 1.8 (Table 1).
This result means that high cooperativity exists between the two ␤-
CD cavities of the homo-dimer ␤C␤H 5, and the Hill constant of 1.8
indicates that a 1:2 complex of the homo-dimer and the substrate
8 is the most reactive intermediate. The kinetic data for ␤H 4 were
fitted to the Michaelis–Menten equation (Eq. (2)) and were also
Table 1
Michaelis–Menten’s parameters and Hill’s parameters for the cleavage of the ester (8) by ␤H, ␤C␤H, ␣C␤H, or ␤C␣Ha.
Michaelis–Menten’s Parameters
Hill’s Parameters
kcat [10−2 s−1
kcat [10−2 s−1
]
Km [10−3 M]
r
]
Kꢀ [10−4 M]
5.92 0.22
0.0656 0.0012
12.5 0.8
n
r
␤H
1.10 0.10
–
1.48 0.56
0.837 0.026
2.43 0.40
–
5.08 0.26
4.77 0.20
0.998
–
0.997
0.997
0.900 0.238
1.05 0.11
0.824 0.048
0.612 0.053
1.16 0.25
0.999
0.999
0.998
0.998
␤C␤H
␣C␤H
␤C␣H
1.80 0.22
1.15 0.53
1.13 0.47
13.0 0.6
a
At 25 ◦C in pH 7.4 phosphate buffer (0.1 M); [8] = 2.5 × 10−4–2.5 × 10−3 M, [␤H] = [␤C␤H] = [␣C␤H] = [␤C␣H] = 4.8 × 10−5 M.

H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
5
Scheme 2. Cooperative binding of substrates to the enzyme.
fitted to the Hill equation (Table 1). The Hill constant for ␤H 4 is
1.1, which means that the 1:2 complex does not form in the case of
␤H 4.
kcat[E]0[S]n
V0
V0
=
(1)
(2)
Kꢀ + [S]n
kcat[E] [S]
Km + 0[S]
=
The dissociation constant of the homo-dimer and the second
substrate is given by aKs, and Kꢀ of Eq. (1) is given by an−1Ksn, where
Ks is the dissociation constant of the homo-dimer for the first sub-
strate and a is the cooperativity coefficient (Scheme 2) [22,23]. If
Ks for ␤C␤H 5 is equal to Km for ␤H 4, i.e., 2.43 × 10−3, then a is
calculated to be 0.25 for ␤C␤H 5. This value shows that the second
dissociation constant (aKs) of ␤C␤H 5 is 0.25 times the value of its
first dissociation constant (Ks); in other words, the second binding
constant (1/aKs) of ␤C␤H 5 is four times larger than the first binding
constant (1/Ks). Thus, the cooperative behaviour of the two cavities
makes the binding ability of ␤C␤H greater for the second substrate
than for the first substrate. The following experiments were per-
formed to elucidate the mechanism of this allosteric-like reaction
behaviour of ␤C␤H.
The initial rates of the cleavage reaction of the p-nitrophenyl
ester 8 by hetero-dimers ␣C␤H 6 and ␤C␣H 7 were measured
under the same conditions used for ␤H 4 and ␤C␤H 5. The hetero-
dimers ␣C␤H 6 and ␤C␣H 7 showed simple saturation behaviour
rather than sigmoidal behaviour (Fig. 4). Table 1 shows the kinetic
parameters of 4–7 for the cleavage reactions, which were obtained
by fitting the kinetic data to ether the Hill equation (Eq. (1)) or the
Michaelis–Menten equation (Eq. (2)). The Hill constants of ␣C␤H
6 and ␤C␣H 7 are approximately 1. The kcat values of 4–7 do not
differ much, but the Hill constant of 5 is twice as large as that of
the others. The overall apparent binding ability (1/Kꢀ) of ␤C␤H 5
is 200 times larger than that of ␣C␤H 6 or ␤C␣H 7. Among the
artificial enzymes, only ␤C␤H 5 has a large binding ability for the
second substrate and demonstrates high cooperativity for substrate
binding. During the reaction, the p-nitrophenyl moiety is located
in one of the two cavities and the ethylene glycol chain is located
in the other cavity of the dimer. The ethylene glycol chain of the
Fig. 5. Effect of p-nitrophenol on the initial rates of cleavage of the substrate 8
(4.3 × 10−5 M) by the CD dimers 5–7 (4.0 × 10−5 M) in pH 7.4 phosphate buffer
(0.1 M) at 25 ◦C.
ester acts as a spacer to stabilize the complex of the homo-dimer
with the second substrate. Because the ␣-CD cavity is too small
to hold both the phenyl moiety and the ethylene glycol chain, the
hetero-dimer cannot make an inclusion complex with the second
substrate. The important point of these results is that two ␤-CD
cavities caused the sigmoidal behaviour observed for the cleavage
reaction of the ester 8, in contrast, ␣- and ␤-CD cavities did not
cause such a behaviour. This means that the formation of the 1:2
host/guest complex is important for the sigmoidal behaviour. kcat
of ␣C␤H 6 or ␤C␣H 7 are not same, because the imidazole moi-
ety is not located at the center of the dimer 6 or 7. The reactivities
of the reaction intermediates depend on both the distance and the
angle between the imidazole moiety and the ester carbonyl group
of the substrate 8. Therefore, the reactivity of the dimer 6 is differ-
ent from that of the dimer 7, and their kcat are not same, because
kcat depends on the reactivity of the reaction intermediate. On the
other hand, the binding ability of the dimer mainly depends on its
cavity size. So, Km of the dimer 6 is similar to that of the dimer 7.
The Hill constant is correlated to the binding ability in this case and
the Hill constant of the dimer 6 is similar to that of the dimer 7.
We carried out an inhibition experiment using p-nitrophenol to
confirm that both two cavities in the dimer are involved in the reac-
tion. The initial rate of the cleavage reaction of the p-nitrophenyl
ester 8 catalyzed by ␤C␤H 5 linearly decreased with an increas-
ing concentration of p-nitrophenol (Fig. 5). The initial rate of the
reaction catalyzed by ␣C␤H 6 or ␤C␣H 7 scarcely changed at low
concentrations of p-nitrophenol, but linearly decreased at high con-
centrations of p-nitrophenol (Fig. 5). These results indicate that
both cavities in the dimers take part in the reactions and that the
two cavities in ␤C␤H 5 participate in the reaction to the same
degree. These results also show that the binding abilities of the two
cavities for p-nitrophenol are not equivalent in the hetero-dimers,
whereas the two cavities of ␤C␤H 5 have same binding ability for p-
nitrophenol. This can be explained as follows. pKa of p-nitrophenol
is 7.08 and p-nitrophenol already dissociates in the experiment
condition. It is reported that the binding constants of ␣-CD and ␤-
CD for p-nitrophenolate are 3550 and 944 M−1, respectively [24].
Therefore, occupation of the ␤-CD cavity of the hetero-dimer by
the substrate will not be affected by a low concentration of p-
nitrophenol and the reaction rate will not be decreased. When a
larger quantity of p-nitrophenol is added, however, the ␤-CD cavity
is filled with p-nitrophenol and the reaction rate is slowed. By con-
trast, the rate of the reaction catalyzed by the homo-dimer is slowed
even by low concentrations of p-nitrophenol, because the forma-
Fig. 4. Dependence of the initial rate on the substrate concentration for the cleavage
of the substrate 8 by ␣C␤H 6 (4.8 × 10−5 M) and ␤C␣H 7 (4.8 × 10−5 M) in pH 7.4
phosphate buffer (0.1 M) at 25 ◦C.

6
H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
Fig. 7. Dependence of the initial rate on the substrate concentration for the cleavage
of the substrate 9 by the CD dimers 5–7 (4.8 × 10−5 M) in pH 7.4 phosphate buffer
(0.1 M) at 25 ◦C.
Fig. 6. Dependence of the initial rate on the substrate concentration for the cleavage
of p-nitrophenyl acetate by ␤C␤H 5 (4.8 × 10−5 M) in pH 7.4 phosphate buffer (0.1 M)
at 25 ◦C.
Table 3
Michaelis–Menten’s parameters for the cleavage of the ester (9) by ␤C␤H, ␣C␤H,
or ␤C␣Ha.
tion of the major reaction intermediate, i.e., the 1:2 host/guest
complex, is inhibited by low concentrations of p-nitrophenol.
By comparing the reaction behaviour of the ester 8 with ␤C␤H 5
and that of p-nitrophenyl acetate (pNPA) with ␤C␤H 5, it is possible
to determine whether the ethylene glycol chain is essential for the
sigmoidal behaviour. The plot of the dependence of the initial rates
for the cleavage reaction of pNPA by the homo-dimer ␤C␤H 5 gave
a simple saturation curve rather than a sigmoidal curve, as shown
in Fig. 6. The kinetic data were fitted to the Michaelis–Menten
equation (Table 2). This result suggests that the ethylene glycol
chain is required for the sigmoidal behaviour, and that both the
p-nitrophenyl moiety and the ethylene glycol chain of the ester 8
occupy the cavities of the dimer during the reaction. Unfortunately,
it is difficult to determine whether the ethylene glycol chain has a
specific property that gives rise to cooperative behaviour, because
the reactivity of the ester 8 cannot be compared in the same reac-
tion conditions with that of simple alkanoate esters, owing to the
low solubility of alkanoate esters in water.
Comparing the reaction behaviour of the p-nitrophenyl ester 8
with that of the m-nitrophenyl ester 9 yields information about the
structure of the inclusion complex that the dimer forms with the
substrate during the reaction, because the m-nitrophenyl moiety
and the ethylene glycol chain of the m-nitrophenyl ester 9 can-
not both occupy each cavity of the same dimer molecule at the
same time, whereas the p-nitrophenyl moiety and the ethylene gly-
col chain of the p-nitrophenyl ester 8 can simultaneously occupy
each cavity. When the m-nitrophenyl ester 9 was used as a sub-
strate, the plot of the dependence of the initial rate of the cleavage
reaction catalyzed by dimers 5–7 on the substrate concentration
gave a simple saturation curve, as shown in Fig. 7. The kinetic data
could be fitted to the Michaelis–Menten equation (Table 3). This
result therefore strongly indicates that the sigmoidal behaviour of
␤C␤H 5 arises from the inclusion of both the nitrophenyl moiety
and the ethylene glycol chain of the ester 8 in each cavity of the
same homo-dimer molecule.
kcat [10−3 s−1
]
Km [10−3 M]
r
␤C␤H
␣C␤H
␤C␣H
2.28 0.41
5.42 0.97
3.05 0.38
0.470 0.028
4.13 0.10
1.18 0.36
0.999
0.997
0.990
a
At 25 ◦
C
in pH 7.4 phosphate buffer (0.1 M); [9] = 2.5 × 10−4–2.5 × 10−3 M,
[␤C␤H] = [␣C␤H] = [␤C␣H] = 4.8 × 10−5 M.
3.2. Estimation of the reaction mechanism
A mechanism for the observed sigmoidal behaviour was derived
on the basis of the above data as shown in Fig. 8. The homo-dimer
can take various conformations in aqueous solution in the absence
of a guest. In order to simplify the mechanism, therefore, two main
conformations were assumed: a cis conformation and a trans con-
formation. In the cis conformation (C Form) the two CD cavities of
the homo-dimer face each other, whereas in the trans conforma-
tion (T Form) the two CD cavities of the homo-dimer face away
from each other. Our computational conformational studies indi-
cate that if the linker is long enough to rotate the CD cavity, then
the T Form is more stable than the C Form in aqueous solution in
Table 2
Michaelis–Menten’s parameters for the cleavage of pNPA by ␤C␤Ha.
kcat [10−3 s−1
]
Km [10−3 M]
r
␤C␤H
2.99 0.95
8.45 0.32
0.995
a
At 25 ◦
C
in pH 7.4 phosphate buffer (0.1 M); [␤C␤H] = 4.8 × 10−5 M,
Fig. 8. Schematic representation of proposed mechanism for the reaction of ␤C␤H
5 with the substrate 8.
[pNPA] = 2.5 × 10−4–2.5 × 10−3 M.

H. Ikeda et al. / Journal of Molecular Catalysis A: Chemical 328 (2010) 1–7
7
the absence of a guest, because hydroxy groups of CD tend to make
Acknowledgements
hydrogen bonds with waters in aqueous solution and the dimer in
the T Form can make hydrogen bonds with more waters than can
the dimer in the C Form [14]. If the CD cavity in the C Form accom-
modates the first substrate, the dimer immediately proceeds to the
state of a 1:2 host/guest complex, because the first substrate fixes
the conformation as the C Form and the guest in the CD dimer in
the C Form acts a spacer against the binding of a second guest. The
first guest to bind in a CD cavity in the T Form slightly affects the
binding of the second guest in the CD cavity in the T Form. There-
fore, the formation of this inclusion complex shifts the equilibrium
from the T Form to the C Form, and this conformational change
causes the observed sigmoidal behaviour.
We are grateful to Nihon Shokuhin Kako Co. Ltd. for providing
cyclodextrins. This work was supported by a Grant-in-Aid for Sci-
entific Research (C) from Japan Society for the Promotion of Science.
References
[1] A. Fersht, Structure and Mechanism in Protein Science, W.H. Freeman and Com-
pany, New York, 1999.
[2] D. Voet, J. Voet, Biochemistry, 3rd ed., John Wiley & Sons, New York, 2004.
[3] R. Breslow, Artificial Enzymes, Wiley-VCH, Weinheim, 2005.
[4] W.B. Motherwell, M.J. Bingham, Y. Six, Tetrahedron 57 (2001) 4663–4686.
[5] R. Breslow, S.D. Dong, Chem. Rev. 98 (1998) 1997–2011.
[6] R. Breslow, Acc. Chem. Res. 28 (1995) 146–153.
To verify this proposed mechanism, a non-reactive material (the
ether 10) that would act as a spacer was added to the reaction sys-
tem. The initial rates for the cleavage reaction of the p-nitrophenyl
ester 8 catalyzed by the CD homo-dimer (␤C␤H 5) were increased
1.2- and 1.3-folds by the addition of the ether 10 at 5 × 10−4 and
1 × 10−3 M, respectively under the conditions of the substrate 8 at
5 × 10−4 M and ␤C␤H 5 at 4.7 × 10−5 M in pH 7.4 phosphate buffer
(0.1 M) at 25 ◦C. This result indicates that one of the substrates in
the 1:2 complex of ␤C␤H 5 and the substrate 8 acts as a spacer or
a potential inhibitor [20,25,26].
[7] J. Yang, R. Breslow, Angew. Chem. Int. Ed. 39 (2000) 2692–2694.
[8] R.R. French, P. Holzer, M.G. Leuenberger, W.-D. Woggon, Angew. Chem. Int. Ed.
39 (2000) 1267–1269.
[9] M.J. Han, K.S. Yoo, J.Y. Chang, T.-K. Ha, Angew. Chem. Int. Ed. 39 (2000) 347–349.
[10] R. Breslow, B. Zhang, J. Am. Chem. Soc. 114 (1992) 5882–5883.
[11] R. Ueoka, Y. Matsumoto, K. Harada, H. Akahoshi, Y. Ihara, Y. Kato, J. Am. Chem.
Soc. 114 (1992) 8339–8340.
[12] H. Tsutsumi, H. Ikeda, H. Mihara, A. Ueno, Bioorg. Med. Chem. Lett. 14 (2004)
723–726.
[13] H. Nakajima, Y. Sakabe, H. Ikeda, A. Ueno, Bioorg. Med. Chem. Lett. 14 (2004)
1783–1786.
[14] H. Ikeda, Y. Horimoto, M. Nakata, A. Ueno, Tetrahedron Lett. 41 (2000)
6483–6487.
[15] H. Ikeda, T. Ikeda, F. Toda, Bioorg. Med. Chem. Lett. 2 (1992) 1581–1584.
[16] A.D. Hamilton (Ed.), Supramolecular Control of Structure and Reactivity, Per-
spective in Supramolecular Chemistry, vol. 3, John Wiley & Sons, Chichester,
UK, 1997.
[17] J. Rebek Jr., Acc. Chem. Res. 17 (1984) 258–264.
[18] M.F. Perutz, Mechanisms of Cooperativity and Allosteric Regulation in Proteins,
Cambridge University Press, Cambridge, 1990.
[19] J.-L. Pierre, G. Gagnaire, P. Chautemps, Tetrahedron Lett. 33 (1992) 217–220.
[20] O.S. Tee, Adv. Phys. Org. Chem. 29 (1994) 1–85.
[21] T. Akiike, Y. Nagano, Y. Yamamoto, A. Nakamura, H. Ikeda, A. Ueno, F. Toda,
Chem. Lett. 23 (1994) 1089–1092.
[22] I.H. Segel, Biochemical Calculations, 2nd ed., John Wiley & Sons, New York,
1976.
4. Conclusions
The cooperativity of the two ␤-CD cavities in the homo-dimer
molecule gives rise to the sigmoidal behaviour of the dependence
of the initial rate for the ester cleavage on the substrate concentra-
tion. This cooperativity is caused by the formation of a 1:2 complex
between the homo-dimer and the substrate. In this 1:2 complex,
both the phenyl moiety and the ethylene glycol chain occupy the
two cavities of the same dimer molecule, with the ethylene glycol
chain of one substrate and the phenyl moiety of the other substrate
together in same cavity. This 1:2 complex is the most reactive inter-
mediate formed during the reaction. Such behaviour is similar to
that of a natural allosteric enzyme, which is made up of identical
subunits.
[23] G.P. Royer, Fundamentals of Enzymology, John Wiley & Sons, New York, 1982.
[24] T.K. Korpela, J.-P. Himanen, J. Chromatogr. 290 (1984) 351–361.
[25] H. Ikeda, S. Nishikawa, J. Takaoka, T. Akiike, Y. Yamamoto, A. Ueno, F. Toda, J.
Inclusion Phenom. Mol. Recognit. Chem. 25 (1996) 133–136.
[26] O.S. Tee, M. Bozzi, J. Am. Chem. Soc. 112 (1990) 7815–7816.