Artificial Metallooxidases from Cyclodextrin Diacids

Article abstract of DOI:10.1002/chem.201702530

Unprecedented simple artificial metalloenzymes were made from cyclodextrin diacids. Six α- and β-cyclodextrin diacids were prepared and their metal-binding and acid properties were investigated. The diacids formed fairly stable complexes with copper, zinc, and iron, with dissociation constants of 0.4–8×10^?4 m. The iron complexes were found to catalyse a Fenton-like oxidation reaction of benzylic alcohols, which displayed Michaelis–Menten catalysis and rate accelerations up to 2700.

Full text of DOI:10.1002/chem.201702530

A Journal of
Accepted Article
Title: Artificial metallooxidases from cyclodextrin diacids
Authors: Mikael Bols and Bo Wang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201702530
Link to VoR: http://dx.doi.org/10.1002/chem.201702530
Supported by

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
Artificial metallooxidases from cyclodextrin diacids
Bo Wang and Mikael Bols*[a]
Abstract: Unprecedented simple artificial metalloenzymes were
made from cyclodextrin diacids. Six - and -cyclodextrin diacids
were prepared and their metalbinding and acid properties were
investigated. The diacids made fairly stable complexes with copper,
zinc and iron with dissociation constants of 0.4-8 x 10-4 M. The iron
complexes were found to catalyse a Fenton like oxidation reaction of
benzylic alcohols which displayed Michealis-Menten catalysis and
rate accelerations up to 2700.
based artificial enzymes being most prominent among classical
enzyme mimics.[10] Czarnik made the cobalt(III) complex 1 based
on a cyclodextrin host (Figure 1) which could catalyze ester
hydrolysis at neutral pH with a kcat/kuncat of 2900.[11] Cyclodextrin
amines or heteroamines complexed with Copper,[12] Nickel[13] or
Zinc[14] have similarly been shown to have esterase activity.
Metallocyclodextrins with two cyclodextrins have been found
excellent esterases. Recently Mao and collaborators made zinc-
complexes such as 2 (Figure 1) that gave rate accelerations
over 30.000[15] and were enatioselective.[16]
Figure 1. Cyclodextrin based artificial enzymes using a metal to assist
catalysis
Introduction
Figure 2. Cyclodextrin diacids and their proposed binding to divalent
metalions.
Enzymes are protein macromolecules that can accelerate the
rate of specific reactions tremendously.[1] The key to enzyme
function is the binding of substrates in an active site – a cavity in
the protein three dimensional structure. There the substrate is
binding in a reactive conformation while functional groups in the
enzyme facilitate reaction through proximity effects.[2] The
efficiency, mildness and selectivity of enzyme catalysis makes it
interesting to emulate and since most of the enzyme structure
appears superfluous or a necessity associated with creating a
cavity in the protein macrostructure enzyme models concentrate
on the active site.[3] Such active site models (artificial enzymes)
have been created from binders such as cyclodextrins,[4]
calixarenes,[5] cucurbiturils and metal ligand clusters[6] and can
become unique catalysts[7] or lead to a molecular understanding
of enzyme catalysis.
Breslow’s group have similarly demonstrated high esterase
activity
from
bipyridine-copper
cyclodextrins.[17]
More
organometallic type catalysts have also been made as
Armspach and collaborators have made cyclodextrin phosphines
such as
3
that can complex palladium and catalyse
hydroformylations (Figure 1),[18] while Monflier’s group have
made cyclodextrin-phophine rhodium and palladium complexes
that catalyze hydrogenation, hydroformylation and Suzuki
couplings.[19] Reinaud, Jabin and collaborators have studied a
series of metal-calixarene complexes with a higher structural
sophistication than has been seen in the above mentioned
cyclodextrin work and shown that the copper-complexes can act
as oxidases[5] and chatecholases in organic solvents.[20a] Also
palladium and nickel calixarene complexes have been reported
to catalyse crosscoupling reactions.[20b,20c]
Enzymes frequently use metal ions as part of their
catalysis, where the metal ion can act as a supercharge or a
reducing/oxidizing agent.[8] Consequently enzyme models
containing metals have also been devised,[9] with cyclodextrin
From the above work it is clear that relatively few
supramolecular metal catalysts have had two or more ligand
attachment sites especially when cyclodextrins are used as
supramolecular hosts. Also almost all supramolecular hosts
have had amine or phosphine attachment sites. Carboxylates
are however also fine metal ligands and this led us to the idea
that cyclodextrin diacids, some of which (4 and 4) are
known,[21] might work as unprecedented simple metalloenzymes
(Figure 2). We have therefore in this work synthesized the
diacids with variant chain length 4-6 and studied their binding to
divalent metalions of copper, zinc and iron. Secondly we have
[a]
Prof. Dr. M. Bols, Dr. B. Wang
Department of Chemistry
University of Copenhagen
Universitetsparken 5, 2100 Copenhagen O
E-mail: bols@chem.ku.dk
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
Results and Discussion
studied the catalytic properties of these complexes aiming at
establishing a Fenton oxidation like system as shown in Figure 3.
We find that the iron complexes function as artificial enzymes for
oxidation of alcohols giving rate accelerations of up to 2600 over
the uncatalyzed reaction.
Synthesis. - and -cyclodextrins with two acid groups attached
to C-6 at the A and D sugar residues were desired: Compounds
with a spacer of no (4 and 4), one (5 and 5) or two CH2
groups (6 and 6) between C-5 and the carboxylate were
made (Figure 2). The short diacids 4 and 4 were made from
the native cyclodextrins by benzylation, DIBAL promoted
debenzylation, oxidation and deprotection using published
methods.[21] The intermediate length acids 5 and 5were made
as outlined in scheme 2: Native - or -cyclodextrin was
converted to the divinyl derivatives 7 or 7, respectively by
benzylation, regioselective debenzylation, oxidation and wittig
olefination as has previously been reported.[22] Hydroboration
using 9-borabicyclo[3.3.1]nonane (9-BBN) followed by hydrogen
peroxide oxidation have the diol 8 or 8 in 77% and 74% yields,
respectively. Tempo oxidation followed by Pinnick oxidation
gave the diacids 9 or 9 in 86% and 90% yields, respectively.
Finally debenzylation by catalytic hydrogenolysis with Pd/C at
normal pressure gave 5 or 5 in 92% and 60% yield,
respectively.
Figure 3. Aimed mode of Fenton type catalysis of cyclodextrin diacids (L)
complexed to iron.
The longest diacids 6 and 6 were made from the
dialdehydes 10 or 10 that are intermediates in the preparation
of the other derivatives above (Scheme 2). Wittig reaction with
benzyl (triphenylphosphoranylidene)acetate in dichloromethane
gave the transalkenes 11 or 11 in 88% and 61% yield,
respectively. Hydrogenation and hydrogenolysis with Pd/C at
normal pressure gave 6 or 6 in 83% and 89% yield.
Scheme 2. Synthesis of long length diacids 6 and 6.
pKa determination. The six cyclodextrin diacids 4-6 were
subjected to potentiometric titration in order to determine acidity
constants. The pKa values cannot directly be seen from a
titration curve but they can be elucidated by fitting a simulated
titration curve to the experimental curve.[23] From analysis of
titration curves for compounds 4-6 (Figure S1) we obtain the
Scheme 1. Synthesis of diacids 4, 4, 5 and 5.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
acidity constants shown in Table 1. We see three general trends
in these pKa values:
determined as the cyclodextrin concentration at the half point of
titration. This gave dissociation constants to copper of ca 1 x 10-
4
1. Compound 4 and  have stronger acidic groups than
5 and  that have stronger acids than 6 and 
2. The difference between the pKa values of the two acids
in a compound grows in the series 4 to 5 to 6, but there
is essentially no difference between  and .
M and ca 2-8 x 10-4 M for Zn (Table 2). This means that 4-6
binds copper or zinc better than succinic acid (Kd(Cu): 2.5 x 10-3
M; Kd(Zn): 2.5 x 10-2 M) or simple organic acids.[24] There is very
little variation in the size of the binding constants which suggests
a certain flexibility in the interaction.
3. The -cyclodextrins are slightly stronger acids than 
Binding was also determined using isothermal calorimetry (ITC)
on compounds 4 and 4 (Table 2, Figure S2-S4). These results
which had to be carried out with corresponding metal acetates
rather than sulfates confirmed the stoichiometry and gave
dissociation constants of same magnitude or somewhat lower
presumably due to the different conditions of the ITC. Iron
binding was also investigated and Fe2+ was found to bind to 4
and 4 well. The 1:1 stoichiometry found in these studies is
consistent with the binding mode suggested in Figure 2. It was
also confirmed by Electrospray MS of the diacid 4 with Cu2+,
Zn2+ or Fe2+ showing in all three cases the monoprotonated
complex of metal ion and cyclodextrindicarboxylate.
Trend 1 is readily explained from the structure of the attached
acid groups by inductive effects. Compound 4 has the acid
directly attached to the glucose ring with all its
electronwithdrawing oxygen. In 5 one CH2 and in 6 two CH2
groups have been placed between the electronwithdrawing
sugar-ring resulting in less electronwithdrawing effect and a less
acidic carboxylate group.
Table 1. Acidity constants of cyclodextrin diacids
4
5
6
4
5
6
pKa(1)
pKa(2)
2.85
3.75
3.70
4.95
4.08
5.66
2.55
3.65
3.55
4.75
3.69
5.20
Trend 2 is actually the reverse of what one would expect
based on charge separation: For simple diacids the separation
of pKa values decrease as the distance between the acids
increase because it is more difficult to deprotonate a carboxylic
acid when another negative charge is nearby. The pKa for 4
and 4 is very similar to that of a long diacid, while pKa for 5
and 6 is more similar to succinic or even malonic acids. The
likely explanation is that the acid-groups in 5 and 6 is closer to
each other than in 4 which also structurally makes sense.
The most likely reason for trend 3 is that there are a
greater number of electronwithdrawing oxygen atoms in -
cyclodextrin than in -cyclodextrin which to a slightly higher
degree pulls electrons away from the carboxylic acids making
them more acidic.
Scheme 3. Oxidation reaction.
Catalysis. With comparatively strong metal complexation
established we examined the catalytic potential of the
complexes in oxidation. The basic reaction shown in Scheme 3
was performed: A substituted benzyl alcohol (2-20 mM) was
oxidized to benzaldehyde with the latter being monitored
spectrophotometrically.
Hydrogen
peroxide
was
the
stoichiometric oxidant and cyclodextrin and metal perchlorate
were present in catalytic amounts. The solvent was water with
20% acetonitrile to keep the substrate in solution. This reaction
has a slow background rate which is increased slightly if copper,
zinc and iron perchlorate is added (Figure S5). However for
Fe(ClO4)2 the oxidation rate increased 100-1000 times when a
cyclodextrin diacid was added - this was not the case for
Cu(ClO4)2 and Zn(ClO4)2. If succinic acid was added to the
Fe(ClO4)2 catalysed background reaction only a small rate
increase was seen (Figure S5), which clearly indicate that the
cyclodextrin cavity is important.
Binding studies. The binding of Cu2+ and Zn2+ ions to 4-6 were
determined using potentiometric titration: The voltage difference
between a copper or zinc electrode and a reference electrode
with a fixed amount of the corresponding metal sulfate (62 M)
was measured while the ion was titrated by addition of
cyclodextrin diacid. Metal concentration was calculated using the
Table 2. Dissociation constants of diacids with copper,zinc and iron (x10-4
M). The values in brackets were obtained by ITC; otherwise they were
determined electrochemically. – means not determined
4
5
6
4
5
6
The iron catalyzed reaction was optimally performed with
0.3 mM Fe(ClO4)2 and 0.6 mM of 4, 5 or 6 (Scheme 3). Under
these conditions most of the iron was complexed to cyclodextrin
and the oxidation of the substrate (2 to 20 mM) displayed
Michaelis-Menten kinetics giving the kinetic values shown in
Tables 3-5. The cyclodextrins display no catalysis when no iron
is present. Under these condition substrate conversion is kept
low (<10%), but it can readily be increased by increasing time
and concentration. Thus in an experiment of benzyl alcohol
(0.33 M), 4 and Fe(ClO4)2 (5 mol%) 95% conversion was
Cu2+
Zn2+
Fe2+
1.07
1.35
1.07
0.79
1.26
1.07
[0.39]
2.14
[0.79]
3.72
2.95
-
2.29
-
7.94
-
2.51
-
[0.53]
[0.60]
[0.69]
[0.72]
Nernst equation and was found to decrease upon cyclodextrin
addition. Plots of metal versus cyclodextrin concentration
showed 1:1 binding stoichiometry, and the Kd could directly be
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
obtained in 15 minutes, but at the expense of most of the
benzaldehyde being converted to benzoic acid and other
unidentified products.
Table 4. Kinetic parameters for oxidation of alcohols by cyclodextrin diacids
(CD). The substrates are substituted benzyl alcohols or 2-
hydroxymethylnaphtalene. Solvent: MeCN:water 1:4, [Substrate] = 2-20 mM;
[CD] = 0.6 mM, [Fe(ClO4)2] = 0.3 mM, [H2O2] = 50 mM.
[a]
[b]
Table 3. Kinetic parameters for oxidation of alcohols by cyclodextrin diacids
(CD). The substrates are substituted benzyl alcohols or 2-
hydroxymethylnaphtalene. Solvent: MeCN:water 1:4, [Substrate] = 2-20 mM;
[CD] = 0.6 mM, [Fe(ClO4)2] = 0.3 mM, [H2O2] = 50 mM.
CD
4 (Fe2+
Substrate
p-OMe
o-OMe
p-Me
kcat
Km
kcat/kuncat
807
)
)
)
)
)
)
)
1090±20
398±15
103±13
1183±37
340±1
22.7±0.2
16.5±1.1
1.9±0.3
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
892
[a]
[b]
CD
4 (Fe2+
Substrate
p-OMe
o-OMe
p-Me
kcat
Km
kcat/kuncat
481
343
)
)
)
)
)
)
)
650±10
191±3
11.5±0.1
6.4±0.3
2.2±0.4
49±3.1
8.1±0.4
6.9±0.2
21.1±7
p-Br
41.9±1.5
10.6±0.5
2.2±0.3
2622
974
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
4 (Fe2+
428
p-Cl
61±7
203
p-CN
110±3
239
p-Br
1210±27
200±12
117±2
2689
573
2-naphtyl
503±80
16.0±1.5
2117
p-Cl
[a] s-1 [b] mM
p-CN
254
trends such as p-bromobenzyl alcohol being
a superior
2-naphtyl
578±80
2408
substrate. With this substrate the increase in oxidation rate was
about 1700. The reason for the lower rate of catalysis is
probably because of the greater distance from the metal center
to the site of oxidation.
[a] s-1 [b] mM
In Table 3 we see the effect of the short -cyclodextrin
diacid 4 on the oxidation of various benzyl alcohols. The kcat is
the first order rate constant of reaction of substrate bound to
artificial enzyme and kcat/kuncat is how much higher this rate is
compared to the uncatalyzed reaction i.e. the rate increase of
the enzymatic reaction. The rate increase was 200-2700 over
the uncatalyzed reaction with Km values between 0.05 and 0.002
M. Since the Km value in a case like this essentially is equal to
the Kd value of the enzyme-substrate complex we see that these
substrates binds with affinity which is typical for binding of
aromatics to cyclodextrins but nevertheless with an affinity that
is 10 to 1000 times weaker than the binding of iron. The best
substrates were the p-bromobenzyl alcohol and 2-
hydroxymethylnaphtalene which gave the highest rate
accelerations within the cavity. The -cyclodextrin version of this
compound, 4, displayed similar efficient catalysis and substrate
preference (Table 4). The highest rates were those of p-
methoxybenzyl and p-bromobenzyl alcohol, while the best rate
accelerations were, similarly to 4, those of p-bromobenzyl
alcohol and 2-hydroxymethylnaphtalene.
S/V
S/V
400000
S/Vin
200000
S (mM)
0
0
10
20
Figure 4. Hanes plot for the oxidation of 4-chlorobenzyl alcohol by 4 (0.6
mM) in water-MeCN (4:1) with ( ■ ) and without ( ◆ ) the presence of
cyclopentanol (5 mM). Fe(ClO4)2 (0.3 mM) and H2O2 (50 mM) are also present.
[S] is 4-chlorobenzyl alcohol concentration. V is starting velocity. The plot
shows that cyclopentanol is a competitive inhibitor with Ki = 2.6 mM.
To confirm whether the cyclodextrin-iron complex was in
fact behaving as a metalloenzyme, inhibition experiments were
performed with cyclopentanol as an inhibitor. In the presence of
5 mM cyclopentanol the rate of oxidation catalyzed by 4 or 4
was clearly impeded and a Hanes plot of the data showed that
cyclopentanol was a competitive inhibitor with Ki values of 2.6
and 4.3 mM, respectively (Figure 4, Figure S6).
The iron complexes of the longer cyclodextrin diacids 5
and 6 also catalyzed the reaction though in general somewhat
less efficient (Table 5). The catalysis displayed Michaelis-
Menten saturation kinetics in accordance with a behavior of 5
and 6 as artificial enzymes and with some of the same specificity
Table 5. Kinetic parameters for oxidation of alcohols by cyclodextrin diacids
(CD). The substrates are substituted benzyl alcohols or 2-
hydroxymethylnaphtalene. Solvent: MeCN:water 1:4, [Substrate] = 2-20 mM;
[CD] = 0.6 mM, [Fe(ClO4)2] = 0.3 mM, [H2O2] = 50 mM.
[a]
[b]
CD
5 (Fe2+
Substrate
p-OMe
p-Cl
kcat
Km
kcat/kuncat
230
)
)
)
310±20
101±9
70±0.7
90±15
37.1±4.0
18±1.2
5 (Fe2+
5 (Fe2+
289
2-naphtyl
o-OMe
4.5±0.3
12.7±0.5
292
5 (Fe2+
)
202
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
5 (Fe2+
6 (Fe2+
6 (Fe2+
6 (Fe2+
6 (Fe2+
6 (Fe2+
)
)
)
)
)
)
2-naphtyl
p-OMe
o-OMe
p-Br
83±0.5
290±10
133±2
12.0±0.4
15.7±0.3
8.4±0.2
47±2.4
346
215
298
1669
65
Future work should focus on exploiting this effect possibly with
new metals or new ligands.
Experimental Section
751±14
30±1.4
116±6
General information
p-CN
7.16±0.2
6.3±0.4
Dry solvents were tapped from a solvent purification system. (PE is
petroleum ether with boiling point 40-65C). Reactants were purchased
from commercial sources and used without further purification.
Potentiometric titration and pH titration were recorded by PHM210
Standard pH Meter. HRMS were recorded on a Bruker Solarix XR mass
spectrometer analyzing TOF. NMR spectra were recorded on a Brüker
500 MHz spectrometer. Chemical shifts (δ) are reported in ppm relative
to the residue solvent signals (CDCl3: δ = 7.26 for 1H-NMR and 77.16 for
13C-NMR. D2O: δ = 4.79 for 1H-NMR). Assignments were aided by COSY
and HSQC experiments. The oxidation experiments were monitored by
Epoche 2 Microplate spectrophotometer with a 96-well quarts microplate.
2-naphtyl
479
[a] s-1 [b] mM
This hypothesis is confirmed by inspection of models. A
model constructed of 4 binding an octahedral iron atom with 3
water molecules and a benzyl alcohol molecule as ligand (Figure
5) show that the substrate is well accommodated in the -
cyclodextrin cativity with the iron atom situated at the narrow end.
According to this model there is plenty of space for access of
hydrogen peroxide to replace a water molecule ligand and
undergo reaction with the alcohol. Longer acids places the iron
atom further from the cavity which could be the reason for the
lower rate of oxidation of 5 and 6.
6A,6D-dideoxy-6A,6D-di(hydroxymethyl)-hexadeca-O-benzyl-α-
cyclodextrin (8):
To
a solution of 6A,6D-dideoxy-6A,6D-di-vinyl-hexadeca-O-benzyl-α-
cyclodextrin 7[22] (1.1 g, 0.457 mmol, 1 equiv.) in THF (25 mL) was
added 9-BBN (7.3 mL, 4.570 mmol, 10 equiv., 0.5 M in THF). The
reaction mixture was stirred at room temperature for 4 h, then the
reaction mixture was cooled down to 0°C, water (1 mL) was added to
decompose the excess 9-BBN. 3 M NaOH (2 mL) and 35% H2O2 (2 mL)
were added and the mixture was stirred overnight. EtOAc was added and
the aqueous layer was extracted with EtOAc (3×25mL). The combined
organic layers were dried over MgSO4, filtered and concentrated in
vacuum. Silica gel flash chromatography of the residue (PE/EtOAc: 4/1,
the 3/1) afforded -cyclodextrin 8 (850 mg, 77%) as
a white
foam. [훂]ퟐ퐃ퟓ= +25.8 (c 1.0, CHCl3). Rf = 0.67 (PE/EtOAc: 2/1). 1H NMR
(CDCl3, 500 MHz): δ 7.34-7.06 (m, 40 H, H-Ar), 5.38 (d, 1 H, 2J = 10.5 Hz,
CHPh), 5.29 (d, 1 H, 2J = 10.9 Hz, CHPh), 5.28 (d, 1 H, 3J1,2 = 3.0 Hz, H-
2
1), 4.96 (d, 1 H, 2J = 10.1 Hz, CHPh), 4.94 (d, 1 H, J = 11.5 Hz, CHPh),
3
4.86-4.84 (m, 3 H, H-1, 2×CHPh), 4.72 (d, 1 H, J1,2 = 3.1 Hz, H-1), 4.63
(d, 1 H, 2J = 11.1 Hz, CHPh), 4.52-4.38 (m, 8 H, 8×CHPh), 4.31 (d, 1 H,
2J = 12.2 Hz, CHPh), 4.27-4.02 (m, 7 H, 3×H-3, 3×H-5, H-6), 3.96 (dd,
2
1H, 3J5,6 = 4.6 Hz, J6,6 = 9.0 Hz, H-6), 3.88-3.75 (m, 4 H, 2×H-4, 2×H-6),
Figure 5. Model of a 1:1 complex of 4, iron (black), with benzyl alcohol
3.52 (dd, 1H, 3J1,2 = 3.5 Hz, 3J2,3 = 9.9 Hz, H-2), 3.44-3.36 (m, 4 H, 2×H-2,
H-4, CH2CHOH), 3.19 (m, 1 H, CH2CHOH), 2.16 (m, 1 H, CHCH2OH),
1.40 (m, 1 H, CHCH2OH). 13C NMR (CDCl3, 125 MHz): δ 139.88, 139.70,
139.57, 138.86, 138.56, 138.31, 138.16, 137.90 (8×C-Arquat.), 128.41-
126.40 (40×C-Artert.), 100.34, 99.97, 98.57 (3×C-1), 84.96, 82.36, 81.64
(3×C-4), 80.74, 80.67, 80.53 (3×C-3), 80.21, 78.61, 78.34 (3×C-2), 76.28,
75.79, 74.43, 73.67, 73.54, 72.98, 72.69, 72.23 (8×CH2Ph), 71.41, 71.23
(2×C-5), 70.10, 69.53 (2×C-6), 67.65 (C-5), 57.54 (CH2CH2OH), 35.60
(green) and three water-molecules (blue) coordinating to iron.
Conclusions
The cyclodextrin diacids 4-6 are comparatively strong acids with
lower pKa values when the acid is close to the cyclodextrin rim
due to the electron withdrawing effect from the oxygen atoms in
the saccharide rings. The difference in pKa values increase with
longer acid suggesting that hydrogen bonding between occur
more efficiently in monoprotonated 5 and 6 than in 4.
The cyclodextrin diacid form 4-6 relatively strong 1:1
complexes with Cu2+, Zn2+ and Fe2+ ions which are roughly of
the same strength irrespectively of the size or chainlenght of the
cyclodextrin diacids.
(CH2CH2OH), HRMS (ESP): Calcd. For C150H160NaO30 [M+Na]1+
2465.8898; Found: 2465.0880
:
6A,6D-dideoxy-hexadeca-O-benzyl-α-cyclodextrin-6A,6D-
dimethylenecarboxylic Acid (9):
To a solution of cyclodextrin 8 (2.6 g, 1.065 mmol, 1 equiv.) in acetone
(50 mL) was added aq. NaHCO3 (6 mL), then NaBr (55 mg, 0.533 mmol,
0.5 equiv.) and TEMPO (17 mg, 0.107 mmol, 0.1 equiv.) were added at
0°C. Following slow addition of TCCA (990 mg, 4.26 mmol, 4 equiv.) at
0°C, the reaction mixture was stirred at room temperature overnight.
EtOAc was added and the aqueous layer was extracted with EtOAc
(3×25mL). The combined organic layers were dried over MgSO4, filtered
The iron complexes were found to function as artificial
oxidases for the oxidation of benzyl alcohols to aldehydes with a
clear enzyme-like rate enhancing effect of the cyclodextrin cavity.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
2
and concentrated in vacuum. The residue was dissolved in a mixture of
tBuOH (25 mL), THF (25 mL) and 2-methyl-2-butene (2.26 mL, 21.3
mmol, 20 equiv.), and NaClO2 (1.44 g, 15.975 mmol, 15 equiv.) and
NaH2PO4 (2.49 g, 15.975 mmol, 15 equiv.) in water (6 mL) were added.
The reaction mixture was stirred overnight and then quenched with 1
mol/L aq. HCl (10 mL) and extracted with EtOAc (3 × 20 mL). The
organic phase was dried (MgSO4), filtered, and the organic solvent was
removed in vacuum. The residue was purified by chromatography
(DCM/MeOH: 98/2 containing 1% HCOOH) afforded dicarboxylic acid--
cyclodextrin 9 (2.26 g, 86%) as a white foam. [훂]ퟐ퐃ퟓ = +35.6 (c 0.55,
CHCl3), Rf = 0.55 (DCM/MeOH: 95/5), 1H NMR (CDCl3, 500 MHz): δ
7.25-6.69(m, 40 H, H-Ar), 5.25 (d, 1 H, 2J = 10.4 Hz, CHPh), 5.13 (d, 1 H,
1), 5.37 (d, 1 H, J = 11.5 Hz, CHPh), 5.35 (d, 1 H, 2J =10.9 Hz, CHPh),
5.19 (d, 1 H, 2J = 12.4 Hz, CH=CHCOOCHPh), 5.18 (d, 1 H, 3J1,2 =3.5 Hz,
H-1), 5.05 (d, 1 H, 2J = 11.4 Hz, CHPh), 5.01 (d, 1 H, 2J =11.4 Hz, CHPh),
4.96 (d, 1 H, 2J = 12.4 Hz, CH=CHCOOCHPh), 4.95-4.92 (m, 3 H, H-1,
2×CHPh), 4.72 (d, 1 H, 2J = 11.6 Hz, CHPh), 4.67 (dd, 1 H, 3J5,6 =4.1 Hz,
2J6,6 = 9.7 Hz, H-6), 4.61 (d, 1 H, 2J = 12.0 Hz, CHPh), 4.57-4.47 (m, 6 H,
2
2
6×CHPh), 4.42 (d, 1 H, J = 11.6 Hz, CHPh), 4.36 (d, 1 H, J = 12.1 Hz,
CHPh), 4.29-4.12 (m, 6 H, 3×H-3, 2×H-4, H-5), 4.06-3.98 (m, 2 H, 2×H-5),
3.67 (dd, 1 H, 3J1,2 = 3.7 Hz, 3J2,3 = 9.6 Hz, H-2), 3.63-3.53 (m, 5 H, H-2,
H-4, 3×H-6), 3.67 (dd, 1 H, J1,2 =3.1 Hz, J2,3 = 9.9 Hz, H-2), 13C NMR
(CDCl3, 125 MHz): δ 165.80 (CH=CHCOOBn), 145.62 (CH=CHCOOBn),
139.57, 139.49, 139.39, 138.82, 138.53, 138.36, 138.25, 138.22, 136.20
(9×C-Arquat.), 128.64-126.72 (45×C-Artert.), 121.58 (CH=CHCOOBn),
99.25, 99.19, 97.81 (3×C-1), 83.26 (C-4), 81.34, 81.07, 80.92 (3×C-3),
80.35 (C-4), 79.62 (C-2), 79.38 (C-4), 78.98, 78.85 (2×C-2), 76.18, 75.70,
74.93, 73.48, 73.43, 73.32, 72.88, 72.63 (8×CH2Ph), 71.84, 71.77, 69.71
(3×C-5), 69.10, 69.01 (2×C-6), 66.06 (CH=CHCOOCH2Ph), HRMS
3
3
3
2
2J = 11.3 Hz, CHPh), 4.94 (d, 1 H, J1,2 = 3.6 Hz, H-1), 4.90 (d, 1 H, J =
3
11.4 Hz, CHPh), 4.84 (d, 1 H, 2J = 11.4 Hz, CHPh), 4.80 (d, 1 H, J1,2
=
3.6 Hz, H-1), 4.75 (d, 1 H, 2J = 11.5 Hz, CHPh), 4.72 (d, 1 H, 2J = 11.6
Hz, CHPh), 4.61 (d, 1 H, J1,2 = 3.6 Hz, H-1), 4.51-4.21 (m, 11 H,
3
10×CHPh, H-5), 4.09-4.04 (m, 3 H, 2×H-3, H-5), 3.94-3.85 (m, 4 H, H-3,
2
3
H-4, 2×H-5), 3.72 (d, 1 H, J6,6 = 11.1 Hz, H-6), 3.57 (dd, 1H, J5,6 = 6.4
Hz, 2J6,6 = 11.5 Hz, H-6), 3.45 (t, 1 H, 3J3,4 = 3J4,5 = 9.3 Hz, H-4), 3.35 (dd,
(ESP): Calcd. For C166H168NaO32 [M+Na]1+
2698.1386
:
2698.1278; Found:
1H, 3J1,2 = 3.4 Hz, J2,3 = 10.6 Hz, H-2), 3.33 (dd, 1H, J1,2 = 3.2 Hz, J2,3
= 9.5 Hz, H-2), 3.28-3.18 (m, 4 H, H-2, H-6, H-4, CHCOOH), 2.23 (m, 1 H,
CHCOOH). 13C NMR (CDCl3, 125 MHz): δ 174.42 (CH2COOH), 139.72,
139.68, 139.51, 138.86, 138.32, 138.15(2C), 137.48 (8×C-Arquat.),
128.67-126.77 (40×C-Artert.), 101.12, 100.98, 97.73 (3×C-1), 85.80, 82.42,
81.14 (3×C-4), 80.94, 80.58, 80.04 (3×C-3), 79.87, 78.27, 78.06 (3×C-2),
76.53, 75.26(2C), 73.85, 73.37, 73.01, 72.53, 72.39 (8×CH2Ph), 72.03,
71.32 (2×C-5), 69.92 (C-6), 68.87 (C-5), 68.76 (C-6), 38.15 (CH2COOH).
3
3
3
6A,6D-dideoxy-α-cyclodextrin-6A,6D-diethylenecarboxylic acid (6):
To a solution of cyclodextrin 11 (2.21 g, 0.826 mmol, 1 equiv.) in a
mixture of MeOH/EtOAc (50 mL) were added Pd/C (2.2 g) and TFA (cat.).
The reaction mixture was stirred at room temperature for 2 days under
hydrogen atmosphere. Filtration through Celite and evaporation of the
solvent gave cyclodextrin 6 (0.72 g, 83%) as a white foam. [훂]퐃ퟐퟓ
=
HRMS (ESP): Calcd. For C150H157O32 [M+H]1+
2471.0381
: 2471.8740; Found:
1
3
+108.8 (c 0.5, H2O), H NMR (D2O, 500 MHz): δ 5.09 (d, 1 H, J1,2 = 3.0
Hz, H-1), 5.05 (d, 1 H, 3J1,2 = 3.1 Hz, H-1), 5.01 (d, 1 H, 3J1,2 = 3.0 Hz, H-
1), 4.00-3.80 (m, 10 H, 3×H-3, 3×H-5, 4×H-6), 3.64-3.60 (m, 5 H, 3×H-2,
2×H-4), 3.36 (t, 1 H, J3,4
3
=
3J4,5 = 8.8 Hz, H-4), 2.61-2.36 (m, 3 H,
6A,6D-dideoxy-α-cyclodextrin-6A,6D-dimethylenecarboxylic
(5):
Acid
2×CH2CHCOOH, CHCH2COOH), 1.74 (m, 1 H, CHCH2COOH), 13C NMR
(D2O, 125 MHz): δ 177.82 (CH2CH2COOH), 101.68, 101.55, 101.04
(3×C-1), 85.78, 81.11, 80.88 (3×C-4), 73.37, 73.32, 73.13 (3×C-3),
71.86(2C) (2×C-5), 71.75, 71.72, 71.59 (3×C-2), 70.22 (C-5), 60.19,
60.07 (2×C-6), 29.81 (CH2CH2COOH), 26.17 (CH2CH2COOH), HRMS
(ESP): Calcd. For C40H65O32 [M+H]1+: 1057.3458; Found: 1057.3481,
Calcd. For C40H64O32Na [M+Na]1+: 1079.3278; Found: 1079.3297.
To a solution of cyclodextrin 9 (2.6 g, 1.065 mmol, 1 equiv.) in a mixture
of MeOH/EtOAc (50 mL) were added Pd/C (2.4 g) and TFA (cat.). The
reaction mixture was stirred at room temperature overnight under
hydrogen atmosphere. Filtration through Celite and evaporation of the
solvent gave the cyclodextrin 5 (1.0 g, 92%) as a white foam. [훂]퐃ퟐퟓ
=
1
3
+101.0 (c 0.6, H2O), H NMR (D2O, 500 MHz): δ 5.12 (d, 1 H, J1,2 = 3.0
Hz, H-1), 5.08 (d, 1 H, 3J1,2 = 3.2 Hz, H-1), 5.01 (d, 1 H, 3J1,2 = 2.9 Hz, H-
1), 4.23 (t, 1 H, H-5), 4.03-3.98 (m, 3 H, 3×H-3), 3.91-3.79 (m, 9 H, 2×H-
5, 4×H-6), 3.69-3.63 (m, 4 H, 3×H-2, H-4), 3.59-3.47 (m, 2 H, 2×H-4),
6A,6D-dideoxy-6A,6D-di(hydroxymethyl)-nonadeca-O-benzyl--
cyclodextrin (8):
2
3.22 (d, 1 H, J6,6 = 14.7 Hz, CHCOOH), 2.56 (dd, 1 H, CHCOOH), 13C
To
a solution of 6A,6D-dideoxy-6A,6D-di-vinyl-nonadeca-O-benzyl--
NMR (D2O, 125 MHz): δ 175.67 (CH2COOH), 101.64, 101.43, 100.56
(3×C-1), 84.80, 81.45, 80.58 (3×C-4),73.28, 73.01, 72.88 (3×C-3), 71.88,
71.81 (2×C-5), 71.63, 71.58, 71.48 (3×C-2), 68.98 (C-5), 60.39, 60.05
(2×C-6), 37.21 (CH2COOH), HRMS (ESP): Calcd. For C38H61O32
cyclodextrin 7ref (1.6 g, 0.564 mmol, 1 equiv.) in THF (25 mL) was
added 9-BBN (11.3 mL, 5.64 mmol, 10 equiv., 0.5 M in THF). The
reaction mixture was stirred at room temperature for 4 h, then the
reaction mixture was cooled down to 0°C, water (1 mL) was added to
decompose the excess 9-BBN. 3 M NaOH (2 mL) and 35% H2O2 (2 mL)
were added and the mixture was stirred overnight. EtOAc was added and
the aqueous layer was extracted with EtOAc (3×25mL). The combined
organic layers were dried over MgSO4, filtered and concentrated in
vacuum. Silica gel flash chromatography of the residue (PE/EtOAc: 4/1,
[M+H]1+
: 1029.3145; Found: 1029.3167, Calcd. For C38H60O32Na
[M+Na]1+: 1051.2965; Found: 1051.2988.
6A,6D-dideoxy-hexadeca-O-benzyl-α-cyclodextrin-6A,6D-
di(ethylidenecarboxylic acid benzylester) (11):
the 3/1) afforded -cyclodextrin 8 (1.2 g, 74%) as a white foam. [훂]퐃ퟐퟓ
=
To a solution of 6A,6D-dialdehydo-hexadeca-O-benzyl-α-cyclodextrin
+35.2 (c 1.0, CHCl3), Rf = 0.65 (PE/EtOAc: 2/1), 1H NMR (CDCl3, 500
ref
3
10
(1.0 g, 0.414 mmol, 1 equiv.) in dichloromethane (20 mL) was
MHz): δ 7.29-7.01 (m, 95 H, H-Ar), 5.63 (d, 1 H, J1,2 = 3.7 Hz, H-1),
5.26-5.16 (m, 7 H, 2×H-1, 5×CHPh), 5.00 (d, 1 H, J = 11.4 Hz, CHPh),
2
added Ph3PCHCOOBn (876 mg, 2.070 mmol, 5 equiv.). The reaction
mixture was stirred at room temperature overnight, then water was added
and the aqueous phase was extracted with EtOAc (3×25mL). The
combined organic layers were dried over MgSO4, filtered and
concentrated in vacuum. The residue was purified by chromatography
(PE/EtOAc: 4/1) afforded -cyclodextrin 11 (960 mg, 88%) as a white
4.98 (d, 1 H, 3J1,2 = 3.5 Hz, H-1), 4.93 (d, 1 H, 3J1,2 = 3.3 Hz, H-1), 4.85 (d,
1 H, 2J = 11.4 Hz, CHPh), 4.83 (d, 1 H, 3J1,2 = 3.4 Hz, H-1), 4.80-4.70 (m,
10 H, H-1, 9×CHPh), 4.67 (d, 1 H, 2J = 10.6 Hz, CHPh), 4.60 (d, 1 H, 2J =
11.4 Hz, CHPh), 4.55-4.31 (m, 20 H, 20×CHPh), 4.17-3.82 (m, 19 H,
7×H-3, 5×H-4, 2×H-5, 5×H-6), 3.79-3.70 (m, 4 H, H-4, 3×H-6), 3.64 (d,
2
foam. [훂]ퟐ퐃ퟓ
=
+47.5 (c 0.8, CHCl3), Rf = 0.66 (PE/EtOAc: 2/1), 1H NMR
2J6,6 = 10.2 Hz, H-6), 3.58 (d, J6,6 = 10.0 Hz, H-6), 3.55-3.31 (m, 12 H,
(CDCl3, 500 MHz): δ 7.44-7.18 (m, 46 H, 45×H-Ar, CH=CHCOOBn), 6.09
(d, 1 H, 3Jtrans = 15.3 Hz, CH=CHCOOBn), 5.38 (d, 1 H, 3J1,2 = 3.2 Hz, H-
7×H-2, H-4, 4×CH2CHOH), 2.30 (m, 1 H, CHCH2OH), 2.10 (m, 1 H,
CHCH2OH), 1.55 (m, 1 H, CHCH2OH), 1.35 (m, 1 H, CHCH2OH). 13C
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
NMR (CDCl3, 125 MHz): δ 139.59(2C), 139.54, 139.49, 139.47, 139.43,
139.33, 138.85, 138.80, 138.63, 138.59, 138.47, 138.42, 138.36, 138.32,
138.30, 138.08, 138.04, 137.96 (19×C-Arquat.), 128.49-126.53 (95×C-
Artert.), 99.40, 99.26, 99.18, 98.21, 98.15, 98.03, 97.73 (7×C-1), 83.06 (C-
4), 81.57, 81.14, 81.10, 80.92(2C), 80.85(2C) (7×C-3), 80.81, 80.68,
80.47, 80.30, 80.20 (5×C-4), 80.12, 79.51, 78.91, 78.70, 78.59, 78.51
(6×C-2), 78.01 (C-4), 77.91 (C-2), 76.50, 76.09, 75.82, 75.60, 75.08,
74.98, 74.18, 73.76, 73.62, 73.56, 73.54, 73.52, 73.24, 73.07, 72.91,
72.82, 72.80, 72.61, 72.42 (19×CH2Ph), 71.96, 71.91, 71.73, 71.63,
71.14 (5×C-5), 69.84, 69.81, 69.77, 69.58, 69.01 (5×C-6), 67.89, 67.79
(2×C-5), 58.47, 58.07 (2×CH2CH2OH), 35.49, 34.84 (2×CH2CH2OH).
2×CHCOOH). 13C NMR (D2O, 125 MHz):
δ
177.64, 177.62
(2×CH2COOH), 102.09 (3C), 102.06(2C), 101.48, 101.41 (7×C-1), 84.99,
84.89 (2×C-4), 81.43, 81.41, 81.34, 81.29, 80.83, 80.80, 80.76 (7×C-2),
73.18(2C), 73.06, 73.02, 72.94, 72.92, 72.88, 72.86, 72.07, 72.03, 71.97,
71.95, 71.89, 71.81(2C), 71.80(2C) (7×C-3, 5×C-4, 5×C-5), 69.03, 68.95
(2×C-5), 60.15(2C), 59.79, 59.76, 59.60 (5×C-6), 36.70, 36.67
(2×CH2COOH). HRMS (ESP): Calcd. For C44H70O37Na [M+Na]1+
1213.3493; Found: 1213.3512
:
6A,6D-dideoxy-hexadeca-O-benzyl--cyclodextrin-6A,6D-
di(ethylidenecarboxylic acid benzylester) (11):
To a solution of dialdehydo--cyclodextrin 10ref (2.0 g, 0.703 mmol, 1
equiv.) in dichloromethane (30 mL) was added Ph3PCHCOOBn (1.5 g,
3.515 mmol, 5 equiv.). The reaction mixture was stirred at room
temperature overnight, then water was added and the aqueous phase
was extracted with EtOAc (3×30mL). The combined organic layers were
dried over MgSO4, filtered and concentrated in vacuum. The residue was
purified by chromatography (PE/EtOAc: 4/1) afforded -cyclodextrin 11
6A,6D-dideoxy-nonadeca-O-benzyl--cyclodextrin-6A,6D-
dimethylenecarboxylic acid (9):
To a solution of cyclodextrin 8 (1.1 g, 0.383 mmol, 1 equiv.) in acetone
(40 mL) was added aq. NaHCO3 (6 mL), then NaBr (20 mg, 0.192 mmol,
0.5 equiv.) and TEMPO (6.1 mg, 0.038 mmol, 0.1 equiv.) were added at
0°C. Following slow addition of TCCA (356 mg, 1.532 mmol, 4 equiv.) at
0°C, the reaction mixture was stirred at room temperature overnight.
EtOAc was added and the aqueous layer was extracted with EtOAc
(3×25mL). The combined organic layers were dried over MgSO4, filtered
and concentrated in vacuum. The residue was dissolved in a mixture of
tBuOH (20 mL), THF (20 mL) and 2-methyl-2-butene (0.81 mL, 7.66
mmol, 20 equiv.), and NaClO2 (0.52 g, 5.75 mmol, 15 equiv.) and
NaH2PO4 (0.896 g, 5.75 mmol, 15 equiv.) in water (5 mL) were added.
The reaction mixture was stirred overnight and then quenched with 1
mol/L aq. HCl (10 mL) and extracted with EtOAc (3 × 20 mL). The
organic phase was dried (MgSO4), filtered, and the organic solvent was
removed in vacuum. The residue was purified by chromatography
(DCM/MeOH: 98/2 containing 1% HCOOH) afforded dicarboxylic acid--
cyclodextrin (1.0 g, 90%) as a white foam. [훂]ퟐ퐃ퟓ= +36.9 (c 0.45, CHCl3),
Rf = 0.55 (DCM/MeOH: 95/5). 1H NMR (CDCl3, 500 MHz): δ 7.34-6.98(m,
95 H, H-Ar), 5.64 (d, 1 H, 3J1,2 = 4.0 Hz, H-1), 5.36 (d, 1 H, 3J1,2 = 3.3 Hz,
H-1), 5.23 (t, 2 H, 2J = 10.4 Hz, 2×CHPh), 5.14 (d, 1 H, 2J = 10.6 Hz,
(1.34 g, 61%) as a white foam. [훂]퐃ퟐퟓ
= +58.2 (c 1.0, CHCl3), Rf = 0.68
(PE/EtOAc: 2/1), 1H NMR (CDCl3, 500 MHz): δ 7.28-6.87 (m, 107 H,
3
105×H-Ar, 2×CH=CHCOOBn), 5.96 (d, 1H, Jtrans
= 15.6 Hz,
CH=CHCOOBn), 5.91 (d, 1 H, 3Jtrans = 15.5 Hz, CH=CHCOOBn), 5.39 (d,
1 H, 3J1,2 = 3.7 Hz, H-1), 5.24 (d, 1 H, 2J = 11.4 Hz, CHPh), 5.16 (d, 1 H,
2J = 10.3 Hz, CHPh), 5.15 (d, 1 H, 3J1,2 = 3.7 Hz, H-1), 5.13 (d, 1 H, J1,2
3
3
= 3.6 Hz, H-1), 5.10-5.02 (m, 3 H, 3×CHPh), 4.98 (d, 1 H, J1,2 = 3.3 Hz,
H-1), 4.94 (d, 1 H, J1,2 = 3.5 Hz, H-1), 4.90 (d, 1 H, 2J = 13.3 Hz,
3
CH=CHCOOCHPh), 4.88 (d, 1 H, 3J1,2 = 3.7 Hz, H-1), 4.85 (d, 1 H, 3J1,2
=
3.1 Hz, H-1), 4.81 (d, 1 H, 2J = 12.7 Hz, CH=CHCOOCHPh), 4.80 (d, 1 H,
2J = 11.2 Hz, CHPh), 4.76 (d, 1 H, 2J = 10.8 Hz, CHPh), 4.76-4.66 (m, 7
H, 2×CH=CHCOOCHPh, 5×CHPh), 4.61 (d, 1 H, 2J = 10.7 Hz, CHPh),
4.58-4.48 (m, 4 H, 2×CHPh, 2×H-5), 4.43-4.19 (m, 24 H, 22×CHPh, 2×H-
2
5), 4.14 (d, 1 H, J = 11.7 Hz, CHPh), 4.01-3.76 (m, 20 H, 7×H-3, 5×H-4,
3×H-5, 5×H-6), 3.48-3.24 (m, 14 H, 7×H-2, 2×H-4, 5×H-6). 13C NMR
(CDCl3, 125 MHz): δ 165.63, 165.55 (2×CH=CHCOOBn), 145.97, 145.92
(2×CH=CHCOOBn), 139.63, 139.59, 139.58, 139.50, 139.35, 139.32,
139.10, 138.84, 138.73, 138.71, 138.63, 138.50, 138.43, 138.30, 138.27,
138.25, 138.21, 138.11, 138.08, 136.06, 136.01 (21×C-Arquat.), 128.64-
126.72 (105×C-Artert.), 122.01, 121.44 (2×CH=CHCOOBn), 99.25, 98.98,
98.88, 98.63, 98.44, 98.15, 98.07 (7×C-1), 81.90, 81.23 (2×C-4), 81.12,
81.09, 80.97(2C), 80.88 (5×C-3), 80.26 (C-4), 80.02, 79.97 (2×C-3),
79.82, 79.31, 79.19 (3×C-4), 79.12, 79.07, 78.95(2C), 78.91, 78.52(2C)
(7×C-2), 77.87 (C-4), 76.30, 76.19, 76.14, 75.93, 75.38, 75.13, 74.51,
73.73, 73.57, 73.46, 73.35(2C), 73.30, 73.22, 73.12, 72.88, 72.64, 72.49,
72.46 (19×CH2Ph), 72.02, 71.81, 71.61, 71.46, 71.43, 69.96 (6×C-5),
69.44 (C-6), 69.41 (C-5), 69.38, 68.87(2C), 68.73 (4×C-6), 66.01, 65.83
(2×CH=CHCOOCH2Ph).
3
CHPh), 5.10 (d, 1 H, J1,2 = 3.6 Hz, H-1), 5.09 (d, 1 H, 2J = 10.6 Hz,
3
CHPh), 4.92 (d, 1 H, J1,2 = 3.3 Hz, H-1), 4.84-4.66 (m, 13 H, 3×H-1,
2
10×CHPh), 4.61-4.25 (m, 24 H, 2×H-5, 24×CHPh), 4.19 (d, 2 H, J6,6
=
11.3 Hz, 2×H-6), 4.07-3.69 (m, 23 H, 7×H-3, 7×H-4, 5×H-5, 4×H-6), 3.61-
2
3.30 (m, 11 H, 4×H-6, 7×H-2), 3.10 (d, 1 H, J6,6 = 13.7 Hz, CHCOOH),
2
3
2.97 (d, 1 H, J6,6 = 16.2 Hz, CHCOOH), 2.42 (dd, 1 H, J5,6 = 8.3 Hz,
2J6,6 = 16.2 Hz, CHCOOH), 2.26 (m, 1 H, CHCOOH). 13C NMR (CDCl3,
125 MHz): δ 193.05, 183.08 (2×CH2COOH), 139.62, 139.49, 139.40,
139.38, 139.22, 139.18, 139.00, 138.78, 138.75, 138.50, 138.49, 138.26,
138.13, 138.03, 138.02, 137.98(2C), 137.95, 137.90 (19×C-Arquat.),
128.37-126.40 (95×C-Artert.), 100.07, 99.37(2C), 98.98, 98.05, 98.03,
97.15 (7×C-1), 81.68-78.50 (7×C-2, 7×C-3, 7×C-4), 76.36, 76.23, 75.83,
75.17, 74.92, 74.49, 74.01, 73.50, 73.40, 73.32(3C), 73.23, 73.05,
72.91(2C), 72.62, 72.25, 72.18 (19×CH2Ph), 72.05, 71.75, 71.44, 71.39,
71.36 (5×C-5), 69.72, 69.44, 69.30, 68.74(3C, 6×C-6) 68.67, 67.25 (2×C-
5), 38.24, 37.81 (2×CH2COOH). HRMS (ESP): Calcd. for C177H183NaO37
[M+Na]1+: 2924.2414; Found: 2924.2349
6A,6D-dideoxy--cyclodextrin-6A,6D-diethylenecarboxylic acid (6):
To a solution of cyclodextrin 11 (2.28 g, 0.734 mmol, 1 equiv.) in a
mixture of MeOH/EtOAc (50 mL) were added Pd/C (2.0 g) and TFA (cat.).
The reaction mixture was stirred at room temperature for 2 days under
hydrogen atmosphere. Filtration through Celite and evaporation of the
6A,6D-dideoxy--cyclodextrin-6A,6D-dimethylenecarboxylic
(5):
Acid
solvent gave cyclodextrin 6 (0.793 g, 89%) as a white foam. [훂]퐃ퟐퟓ
=
+125.9 (c 0.6, H2O). 1H NMR (D2O, 500 MHz): δ 5.05-4.95 (m, 7 H, 7×H-
To a solution of cyclodextrin 9 (2.4 g, 0.827 mmol, 1 equiv.) in a mixture
of MeOH/EtOAc (50 mL) were added Pd/C (2.4 g) and TFA (cat.). The
reaction mixture was stirred at room temperature for 2 days under
hydrogen atmosphere. Filtration through Celite and evaporation of the
1), 3.92-3.66 (m, 24 H, 7×H-3, 7×H-5, 10×H-6), 3.60-3.49 (m, 12 H, 7×H-
2, 5×H-4), 3.31-3.27 (m,
2
H, 2×H-4), 2.53-2.47 (m,
2
H,
2×CH2CHCOOH), 2.41-2.35 (m, 2 H, 2×CH2CHCOOH), 2.30-2.24 (m, 2
H, 2×CHCH2COOH), 1.73-1.65 (m, 2 H, 2×CHCH2COOH). 13C NMR
(D2O, 125 MHz): δ 177.64, 177.62 (2×CH2CH2COOH), 102.07(2C),
101.88, 101.84(2C), 101.34, 101.22 (7×C-1), 85.52(2C), 81.11, 81.00,
80.96, 80.63, 80.55 (7×C-4), 73.18(2C), 73.11(2C), 73.09, 72.88, 72.83
(7×C-3), 72.17, 72.14, 72.09(2C), 72.01, 71.96, 70.90 (7×C-2), 71.78(2C),
solvent gave cyclodextrin 5 (0.530 g, 60%) as a white foam. [훂]퐃ퟐퟓ
=
+119.0 (c 0.4, H2O). 1H NMR (D2O, 500 MHz): δ 5.12-5.03 (m, 7 H, 7×H-
1), 4.20 (br, 2 H, 2×H-5), 3.94-3.43 (m, 36 H, 7×H-2, 7×H-3, 7×H-4, 5×H-
5, 10×H-6), 3.20-3.10 (m, 2 H, 2×CHCOOH), 2.65-2.51 (m, 2 H,
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
71.73, 71.55, 71.53, 70.34(2C) (7×C-5), 60.23, 60.07(2C), 59.98, 59.94
(5×C-6), 29.76, 29.75 (2×CH2CH2COOH), 26.10(2C) (2×CH2CH2COOH).
HRMS (ESP): Calcd. For C146H74O37Na [M+Na]1+: 1241.3806; Found:
1241.3830
squares nonlinear regression fitting to the vmax vs. [S] curve. kcat was
calculated as vmax/[cyclodextrin]. kuncat was determined as the slope from
a plot of vuncat vs. [S].
Inhibition Assay:
Determination of pKa values of dicarboxylate-cyclodextrins:
In a platereader 6-7 samples (0.25 mL each) of the appropriate substrate
at different concentrations (5-18 mM) in water/CH3CN (4/1) solution
containing cyclodextrindicarboxylate (0.6 mM) and Fe(ClO4)2 (0.3 mM)
or nothing (as control) reactions were simultaneously started by addition
of H2O2 to a final concentration of 50 mM. The reactions were followed at
37ꢀ°C using UV absorption at 286 nm and typically monitored for 2 h.
This was performed with and without cyclopentanol (5 mM). From the
obtained data, Km (without cyclopentanol) and K’m (with cyclopentanol)
could be determined by the Hanes plot, then the Ki value could be
calculated (using Ki = [I]/(K’m/Km-1).
A solution of 5.0×10-5 M cyclodextrin in 50 mL 0.1 M KCl was prepared.
Then 2 mL 0.04 M HCl was added. The solution was titrated by dropwise
addition 0.04 M KOH. All solutions were prepared using boiling deionized
water to avoid the errors caused by CO2 in water. After each addition the
total volume of added KOH and electrode potential (mV) were noted. The
obtained data was plotted and fitted to give pKa value of different
dicarboxylate-cyclodextrins (Figure S1 and Table 1).
Determination of binding constant of different metals with
cyclodextrins by means of potentiometric titration:
Different stock solution were prepared: A) CuSO4 (6.2×10-5 M) in 0.1 M
Na2SO4; B) ZnSO4 (6.2×10-5 M) in 0.1 M Na2SO4 and C) cyclodextrins 4,
5, 6, 4, 5, or 6 in concentrations 2×10-3 M or 1×10-2 M. The cell
consisted of a 50-mL beaker, a saturated calomel electrode and a piece
of well-polished metal sheet (copper or zinc, it was washed with diluted
nitric acid to remove the oxide on the surface). 40 mL of solution A or B
was taken for a measurement and stirred continuously by a magnetic
stirrer. The potential was varied by dropwise addition of solution C. The
obtained data were plotted and fitted with nonlinear regression to give the
binding constant of the metal with cyclodextrin.
Acknowledgements
We thank Preben Graa Sørensen and Axel Hunding for advice
regarding the titrations and thank the danish national research
council for supporting this project.
Keywords: artificial enzyme • oxidase • ironenzyme •
cyclodextrin • enzyme kinetics
[1]
Depending on the process enzymatic reactions have been estimated to
run 106 to 1018 times faster than the corresponding uncatalysed
reaction: R. Wolfenden, M. Snider Acc. Chem. Res. 2001, 34, 938-945.
A. J. Kirby, F. Hollfelder From Enzyme Models to Model Enzymes, RSC
publishing, 2009.
ITC measurement:
An isothermal calorimeter (ITC200, MicroCal Inc., USA) was used for
simultaneously determining the binding constant and the inclusion
enthalpy of the studied complexes at 298K. Degassed solutions were
used in both cell (1.8 mL, 2 mM) and syringe (280 L, 20 mM). For all
protocols, experiments were implemented as follows. After the addition of
an initial aliquot of 5 μL, 28 aliquots of 10 μL of the syringe solution were
delivered (over 10 s for each injection). The time interval between two
consecutive injections was 240 s and the agitation speed was 307 rpm
for all experiments. The resulting heat flow was recorded as a function of
time. In addition, the heat of dilution (for each partner) was eliminated by
subtracting the raw signal obtained for the corresponding blank titrations
(i.e., only one partner in cell or syringe, the other compartment being
filled with buffer). The peak area following each addition was obtained by
integration of the resulting signal and was expressed as the heat effect
per injection. Binding constants and inclusion enthalpies were finally
determined by nonlinear regression analysis of the binding isotherms.
[2]
[3]
a) E. Kuah, S. Toh, J. Yee, Q. Ma, Z. Gao Chem. Eur. J. 2016, 22,
8404-8430. b) C. Deraedt, D. Astruc, Coord. Chem. Rev. 2016, 324,
106-122. c) A. Meister, R. Breslow Artificial Enzymes and Enzyme
Models, J. Wiley & sons, 2006.
[4]
a) L. Marinescu, M. Bols Trends in Glycoscience and Glycotechnology
2009, 21, 309-323. b) R. Breslow, S. D. Dong, Chem. Rev. 1998, 98,
1997-2011
[5]
[6]
[7]
N. Le Poul, Y. Le Mest,, I. Jabin, O. Reinaud Acc. Chem. Res. 2015, 48,
2097−2106.
C. J. Brown, F. D. Toste, R. G. Bergman, K. N. Raymond Chem. Rev.
2015, 115, 3012-3035.
For artificial enzymes that can detoxify nerve gasses see a) C.
Schneider, A. Bierwisch, M. Koller, F. Worek, S. Kubik Angew. Chem.
Int. Ed. 2016, 55, 12668-12672. b) S. Letort, D. Mathiron, T. Grel, C.
Albaret, S. Daulon, F. Djedani-Pilard, G. Gouhiera, F. Estour Chem.
Commun. 2015, 51, 2601-2604.
Oxidation assay:
[8]
[9]
R. B. Silverman Organic Chemistry of Enzyme-Catalyzed Reactions,
2nd Ed. Academic Press, 2002.
In a platereader 10 samples (0.25 mL each) of the appropriate substrate
at different concentrations(2-20 mM) in water/CH3CN (4/1) solution
containing cyclodextrindicarboxylate (0.6 mM) and Fe(ClO4)2 (0.3 mM)
or nothing (as control) was simultaneously started by addition of 50 mM
H2O2. The reactions were followed at 37ꢀ°C using UV absorption at an
appropriate wavelength[25] and typically monitored for 2 h. Velocities were
determined as the slope of the progress curve of each reaction. The
velocities of the uncatalyzed reactions were obtained directly from the
control samples, those of the catalyzed reactions were calculated by
subtracting the uncatalyzed rate from the total rate of the appropriate
cyclodextrin-containing sample. The vcat values were used to construct
Hanes plots ([S]/v vs. [S]) to ensure that the reaction follows Michaelis–
Menten kinetics. In that case Km and vmax were determined using least-
O. Bistri, O. Reinaud Org. Biomol. Chem., 2015, 13, 2849-2865.
[10] D. Prochowicza, A. Kornowicza, I. Justyniak, J. Lewinski Coord. Chem.
Rev. 2016, 306, 331–345
[11] E. U. Akkaya, A. W. Czarnik, J. Am. Chem. Soc. 1988, 110, 8553-8554.
[12] H.-J. Schneider, F. Xiao, J. Chem. Soc. Perkin Trans. 2, 1992, 387-391.
[13] R. Breslow, L. E. Overman, J. Am. Chem. Soc. 1970, 92, 1075-1077.
[14] K. D. Zoh, S. H. Lee, J. Suh, Bioorg. Chem. 1994, 22, 242-252. b) S. M.
Kim, I. S. Hong, J. Suh, Bioorg. Chem. 1998, 26, 51-60. c) J. Suh, W. J.
Kwon, Bioorg. Chem. 1998, 26, 103-117.
[15] Y.-H. Zhou, M. Zhao, Z. -W. Mao and L. -N. Ji, Chem. Eur. J. 2008, 14,
7193-7201.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
[16] S.-S. Xue, M. Zhao, J. -X. Lan, R. -R. Ye, Y. Li, L. -N. Ji, Z. -W. Mao, J.
Mol. Cat. A. Chem. 2016, 424, 297–303.
[17] B. Zhang, R. Breslow, J. Am. Chem. Soc. 1997, 119, 1676-1681.
[18] a) D. Armspach, D. Matt, C. R. Chim. 2011, 14, 135-148. b) D.
Armspach, D. Matt, Chem. Commun. 1999, 1073-1074. c) M. Jouffroy,
R. Gramage-Doria, D. Sémeril, D. Armspach, D. Matt, W. Oberhauser,
L. Toupet, Beilstein J. Org. Chem. 2014, 10, 2388–2405. d) M. Jouffroy,
D. Armspach, D. Matt Dalton Trans., 2015, 44, 12942–12969. e) M.
Jouffroy, R. Gramage-Doria, D. Armspach, D. Sémeril, W. Oberhauser,
D. Matt, L. Toupet, Angew. Chem. Int. Ed. 2014, 53, 3937–3940.
[19] a) C. Machut-Binkowski, F.-X. Legrand, N. Azaroual, S. Tilloy, E.
Monflier, Chem. Eur. J. 2010, 16, 10195-10201. b) F.-X. Legrand, N.
Six, C. Slomianny, H. Bricout, S. Tilloy, E. Monflier, Adv. Synth. Catal.
2011, 353, 1325-1334. c) D. N. Tran, F.-X. Legrand, S. Menuel, H.
Bricout, S. Tilloy, E. Monflier, Chem. Commun. 2012, 48, 753-755. d)
F.-X. Legrand, M. Ménand, M. Sollogoub, S. Tilloy, E. Monflier, New J.
Chem. 2011, 35, 2061-2065. d) M. Guitet, F. Marcelo, S. A. de
Beaumais, Y. Zhang, J. Jiménez-Barbero, S. Tilloy, E. Monflier, M.
Ménand, M. Sollogoub Eur. J. Org. Chem. 2013, 3691–3699.
[20] a) O. Sꢁnꢂque, M. Campion, N. Douziech, M. Giorgi, E. Riviꢂre, Y.
Journaux, Y. Le Mest, O. Reinaud, Eur. J. Inorg. Chem. 2002, 2007-
2014. b) L. Monnereau, D. Sémeril, D. Matt, L. Toupet, Chem. Eur. J.
2010, 16, 9237–9247. c) L. Monnereau, D. Sémeril, D. Matt, Chem.
Comm. 2011, 47, 6626-6628.
[21] C. Rousseau, F. Ortega-Caballero, L. U. Nordstrøm, B. Christensen, T.
E. Petersen, M. Bols, Chem. Eur. J. 2005, 11, 5094-5101
[22] O. Bistri, P. Sinay, J. Jiminez-Barbero, M. Sollogoub, Chem. Eur. J.
2007, 13, 9757-9774.
[23] V. Tomišić, T. Gopurenko, K. Majorinc, V. Simeon Croatia Chem. Acta.
2006, 79, 613-618.
[24] M. Yasuda, K. Yamasaki, H. Ohtaki, Bull. Chim. Japan 1960, 33, 1067-
1070.
[25] L. G. Marinescu, M. Bols, Angew. Chem. Int. Ed. 2006, 45, 4590-4593.
This article is protected by copyright. All rights reserved.

10.1002/chem.201702530
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Iron cyclodextrin complexes are
artificial oxidases of benzylethers.
Bo Wang and Mikael Bols*
Page No. – Page No.
Artificial metallooxidases from
cyclodextrin diacids
Layout 2:
FULL PAPER
Author(s), Corresponding Author(s)*
((Insert TOC Graphic here; max. width: 11.5 cm; max. height: 2.5 cm))
Page No. – Page No.
Title
Text for Table of Contents
This article is protected by copyright. All rights reserved.