Four orders of magnitude rate increase in artificial enzyme-catalyzed aryl glycoside hydrolysis

Article abstract of DOI:10.1021/jo050861w

(6^A6^DR)-6^A,6^D-Di-C-cyano- β-cyclodextrin (1) and 6^A,6^D-di-C-cyano-α- cyclodextrin (2) were synthesized and shown to catalyze hydrolysis of aryl glycosides into glucose and phenol with a reaction following Michaelis-Menten kinetics. At pH 8.0 and 59 °C hydrolysis of 4-nitrophenyl α-glucopyranoside was catalyzed by 1 with K_M = 10.5 ± 1.5 mM, k_cat = 1.42(±0.09) × 10^-4 s ^-1 and k_catk/_uncat = 7922, Catalysis was observed with a concentration of 1 as low as 10 μM. Hydrolysis of the other aryl glycosides containing stereochemical variation in the sugar-moiety and 4-nitro-, 2-nitro-, 2-aldehydo-, and 2,4-dinitro- were also catalyzed by 1 and 2 with k_cat/k_uncat ranging from 4 to 7100. Hydrolysis of a phenyl β-D-glucoside or the thioglycoside tolylthio β-D-glucoside was also catalyzed. From a series of prepared analogues of 1 it was found that the catalysis was associated with the hydroxyl groups α to the nitril groups. The monocyanohydrin 6-C-cyano-β-cyclodextrin (3) was also found to catalyze the hydrolysis of 4-nitrophenyl β-glucopyranoside with k _Cat/k_uncat = 1356. It was proposed that the cyclodextrin cyanohydrins 1-3 catalyze the hydrolysis by general acid catalysis on the bound substrate.

Full text of DOI:10.1021/jo050861w

Four Orders of Magnitude Rate Increase in Artificial
Enzyme-Catalyzed Aryl Glycoside Hydrolysis
Fernando Ortega-Caballero, Jeannette Bjerre, Line Skall Laustsen, and Mikael Bols*
Department of Chemistry, University of Aarhus, Langelandsgade 140, DK-8000 Aarhus C, Denmark
mb@chem.au.dk
Received April 28, 2005
(6AR,6DR)-6A,6D-Di-C-cyano-â-cyclodextrin (1) and 6A,6D-di-C-cyano-R-cyclodextrin (2) were synthe-
sized and shown to catalyze hydrolysis of aryl glycosides into glucose and phenol with a reaction
following Michaelis-Menten kinetics. At pH 8.0 and 59 °C hydrolysis of 4-nitrophenyl R-glucopy-
ranoside was catalyzed by 1 with KM ) 10.5 ( 1.5 mM, kcat ) 1.42((0.09) × 10-4 s-1, and kcat/kuncat
) 7922. Catalysis was observed with a concentration of 1 as low as 10 µM. Hydrolysis of the other
aryl glycosides containing stereochemical variation in the sugar-moiety and 4-nitro-, 2-nitro-,
2-aldehydo-, and 2,4-dinitro- were also catalyzed by 1 and 2 with kcat/kuncat ranging from 4 to 7100.
Hydrolysis of a phenyl â-D-glucoside or the thioglycoside tolylthio â-D-glucoside was also catalyzed.
From a series of prepared analogues of 1 it was found that the catalysis was associated with the
hydroxyl groups R to the nitril groups. The monocyanohydrin 6-C-cyano-â-cyclodextrin (3) was also
found to catalyze the hydrolysis of 4-nitrophenyl â-glucopyranoside with kcat/kuncat ) 1356. It was
proposed that the cyclodextrin cyanohydrins 1-3 catalyze the hydrolysis by general acid catalysis
on the bound substrate.
Introduction
A holy grail in bioorganic chemistry is the creation of
artificial enzymes with the ability to mimic the rate
enhancements of natural enzymes. This goal has been
pursued by investigation of modified proteins such as
genetically engineered enzymes or catalytic antibodies
or by synthesizing small organic molecules with the
ability to mimic the active site.1 As artificial enzymes
have been reported to achieve rate enhancements (kcat
/
kuncat) between 101 and 105, and natural enzymes typically
have kcat/kuncat values of 106-1017; this goal, though
difficult, does not appear too elusive.2
FIGURE 1. Hanes plot for 1-catalyzed hydrolysis of different
substrates at pH 7.4 and 59 °C.
Glycosidases are enzymes that catalyze hydrolysis of
glycosides employing a two-step mechanism involving
general acid catalysis and nucleophilic catalysis being
performed by two carboxylates in the active site.3 The
rate enhancements caused by glycosidases are among the
highest known among enzymes.4 Incorporation of two
carboxylates into a cyclodextrin creates a weak artificial
aryl glycosidase, but the catalysis does not occur by
protonation but rather by electrostatic stabilization of the
transition state.5
We have recently reported that the cyclodextrin dicy-
anohydrin 1 is an artificial glycosidase with a kcat/kuncat
of 1000 for the hydrolysis of 4-nitrophenyl-â-D-glucoside
(Figure 1).6 We have now found that 1 and its R-cyclo-
dextrin analogue 2 both, under optimal conditions,
catalyze this reaction with a kcat/kuncat of up to 8000. We
(1) Kirby, A. J. Angew. Chem., Int. Ed. Engl. 1994, 33, 551-553.
10.1021/jo050861w CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/04/2005
J. Org. Chem. 2005, 70, 7217-7226
7217

Ortega-Caballero et al.
SCHEME 1. Synthesis of Dicyanohydrins 1 and 2, and Monocyanohydrin 3
also report that a â-cyclodextrin with a single cyanohy-
drin group gives a kcat/kuncat of over 1000 as well as results
with other substrates.
and diols that can be prepared by the method of Pearce
and Sinay¨.7 Oxidation of these alcohols with Dess-
Martin’s reagent affords aldehydes that were reacted
with potassium cyanide giving cyanohydrins. Thus 1 and
2 were prepared by reaction of dialdehydes 48 and 69 with
potassium cyanide giving the dicyanohydrins 5 and 7 in
77% and 80% yield, respectively (Scheme 1). The dicy-
anohydrin 5 is essentially one stereoisomer, while 7 could
contain more than one stereoisomer. Hydrogenolysis of
5 and 7 with palladium on carbon catalysis gave 1 and 2
in quantitative yield.
Results and Discussion
Synthesis. The cyclodextrin cyanohydrins were syn-
thesized from the partially benzylated cyclodextrin mono-
(2) (a) Murakami, Y.; Kikuchi, J. I.; Hisaeda, Y.; Hayashida, O.
Chem. Rev. 1996, 96, 721-758. (b) Motherwell, W. B.; Bingham, M.
J.; Six, Y. Tetrahedron 2001, 57, 4663-4686. (c) Breslow, R. Acc. Chem.
Res. 1995, 28, 146-153. (d) Breslow, R.; Schmuck, C. J. Am. Chem.
Soc. 1996, 118, 6601-6605. (e) Breslow, R.; Zhang, B. J. Am. Chem.
Soc. 1992, 114, 5882-5883. (f) Akiike, T.; Nagano, Y.; Yamamoto, Y.;
Nakamura, A.; Ikeda, H.; Veno, A.; Toda, F. Chem. Lett. 1994, 1089-
1092. (g) Breslow, R.; Zhang, B. J. Am. Chem. Soc. 1994, 116, 7893-
7894. (h) Breslow, R.; Dong, S. D. Chem. Rev. 1998, 98, 1997-2011.
(i) Ikeda, H.; Horimoto, Y.; Nakata, M.; Ueno, A. Tetrahedron Lett.
2000, 41, 6483-6487. (j) Chou, D. T. H.; Zhu, J.; Huang, X. C.; Bennet,
A. J. J. Chem. Soc., Perkin Trans. 2001, 2, 83-89. (k) Kunishima, M.;
Yoshimura, K.;, Morigaki, H.; Kawamata, R.; Terao, K.; Tani, S. J.
Am. Chem. Soc. 2001, 123, 10760-10761. (l) Ren, X. J.; Xue, Y.; Zhang,
K.; Liu, J. Q.; Luo, G. M.; Zheng, J.; Mu, Y.; Shen, J. C. FEBS Lett.
2001, 507, 377-380. (m) Yu, J. X.; Zhao, Y. Z.; Holterman, M.; Venton,
D. L. Bioorg. Med. Chem. 2002, 10, 3291-3299. (n) Ren, X. J.; Xue,
Y.; Liu, J. Q.; Zhang, K.; Zheng, J.; Luo, G.; Guo, C. H.; Mu, Y.; Shen,
J. C. ChemBioChem 2002, 3, 356-363. (o) Liu, Y.; Li, B.; Li, L.; Zhang,
H. Y. Helv. Chim. Acta 2002, 85, 9-18. (p) Kataky, R.; Morgan, E.
Biosens. Bioelectron. 2003, 18, 1407-1417. (q) Chan, W. K.; Yu, W.
Y.; Che, C. M.; Wong, M. K. J. Org. Chem. 2003, 68, 6576-6582. (r)
Milovic, N. M.; Badjic, J. D.; Kostic, N. M. J. Am. Chem. Soc. 2004,
126, 696-697. (s) Rousseau, C.; Christensen, B.; Pedersen, T. E.; Bols,
M. Org. Biomol. Chem. 2004, 2, 3476-3482. (t) Fukudome, M.; Okabe,
Y.; Sakaguchi, M.; Morikawa, H.; Fujioka, T.; Yuan, D. Q.; Fujjta, K.
Tetrahedron Lett. 2004, 45, 9045-9048. (u) Dong, Z. Y.; Liu, J. Q.;
Mao, S. Z.; Huang, X.; Yang, B.; Ren, X. J.; Luo, G. M.; Shen, J. C. J.
Am. Chem. Soc. 2004, 126, 16395-16404. (v) Doug, T. H.; Chou, J. Z.;
Huang, X.; Bennet, A. J. J. Chem. Soc., Perkin Trans. 2 2001, 83-89.
(w) Ohe, T.; Kajiwara, Y.; Kida, T.; Zhang, W.; Nakatsuji, Y.; Ikeda, I.
Chem. Lett. 1999, 921-922.
Similarly the â-cyclodextrin monoaldehyde 810 was
reacted with KCN giving cyanohydrin 9 in 86% yield
(Scheme 1). This was followed by hydrogenolysis of the
benzyl protection groups of 9 giving 3 in 73% yield. The
cyanohydrin synthesis appears essentially stereoselective
and a single diastereomer is in any case obtained after
purification. The â-dicyanohydrin 1 was one diastere-
omer, while 2 and 3 at least had a major stereoisomer.
(3) Zechel, D. L.; Withers, S. G. Acc. Chem. Res. 2000, 33, 11-18.
(4) Wolfenden, R. Acc. Chem. Res. 2001, 34, 938-945.
(5) Rousseau, C.; Nielsen, N.; Bols, M. Tetrahedron Lett. 2004, 45,
8709-8711.
(6) Ortega-Caballero, F.; Rousseau, C.; Christensen, B.; Petersen,
T. E.; Bols M. J. Am. Chem. Soc. 2005, 127, 3238-3239.
(7) (a) Pearce, A. J.; Sinay¨, P. Angew. Chem., Int. Ed. 2000, 39,
3610-3612. (b) Lecourt, T.; Herault, A.; Pearce, A. J.; Sollogoub, M.;
Sinay¨, P. Chem. Eur. J. 2004, 10, 2960-2971.
(8) Hardlei, T.; Bols, M. J. Chem. Soc., Perkin Trans. 1 2002, 2880-
2885.
(9) Rousseau, C.; Christensen, B.; Bols, M. Eur. J. Org. Chem. 2005,
13, 2734-2739.
(10) Rousseau, C.; Ortega-Caballero, F.; Nordstrøm, L. U.; Chris-
tensen, B.; Petersen, T. E.; Bols, M. Chem. Eur. J. In press.
7218 J. Org. Chem., Vol. 70, No. 18, 2005

Artificial Enzyme-Catalyzed Aryl Glycoside Hydrolysis
SCHEME 2. Determination of the
Stereochemistry of 1 by Degradation to
D-glycero-D-gluco-Heptitol
and 21 were prepared (Schemes 3). Dinitrile 14 was
obtained by conversion of the corresponding known
benzylprotected A,D diol 11 into the diiodide 12 with
Ph3P, I2, and imidazole in 96% yield as has been reported
by Sinay¨.7 This diiodide was substituted with KCN giving
dinitrile 13 in 85% yield, and deprotected by hydro-
genolysis in EtOAc/MeOH giving dinitrile 14 in 99% yield
(Scheme 3). The dialdehyde 15 was obtained by similar
hydrogenolysis of the protected dialdehyde 48 (Scheme
3). NMR of 15 shows that in aqueous solution it is
exclusively on the hydrate form and it is therefore
interesting to compare its catalytic properties with those
of corresponding hydroxyl compounds. The di-6A,6D-C-
propyl-substituted derivates 17, 19, and 21 were pre-
pared by hydrogenolysis of the known stereoisomeric
benzylprotected allyl derivatives 16, 18, and 20 (Scheme
3) that have previously been reported by us.8 These
compounds have restricted conformational freedom of the
substituted C6’s disallowing substituents in the tg con-
formation.8 This means that the OH groups in 17 are in
the gt conformation pointing in toward the inner rim,
while those of 19 are in the gg conformation and are not.
Compound 21 is a mixture of diastereoisomers each
having one OH directed toward the inner rim.
Compound 1 was shown to be the R,R isomer by the
method outlined in Scheme 2. Hydrolysis of 1 with IR-
120 followed by NaBH4 reduction gave a 5:2 mixture of
D-glucitol and a heptitol 10 that by comparison with 13
C
NMR data11 was shown to have the D-glycero-D-gluco-
configuration. This stereochemistry is consistent with
attack by cyanide from the outside of the cyclodextrin
dialdehyde 4 and the cyanohydrin hydroxyl groups in 1
pointing toward the inner face.
For comparison with the cyanohydrin compounds 1-3,
a number of the â-cyclodextrin analogues 14, 15, 17, 19,
Catalysis. Compound 1 was found by NMR to catalyze
the conversion of 4-nitrophenyl â-D-glucopyranoside into
SCHEME 3. Synthesis of â-Cyclodextrin Derivative 14, 15, 17, 19, and 21
J. Org. Chem, Vol. 70, No. 18, 2005 7219

Ortega-Caballero et al.
glucose and nitrophenol. Thus after leaving 1 for several
days at 60 °C with 4-nitrophenyl â-D-glucopyranoside in
phosphate buffer/D2O, R- and â-glucose and 4-nitrophenol
can be observed. The catalysis can be inhibited by
addition of cyclopentanol confirming that the cyclodextrin
cavity is involved in the process. No catalysis is observed
with either â-cyclodextrin or mandelonitrile showing that
the supramolecular positioning of binding cavity and
cyanohydrin group is essential for catalysis. By following
the formation of phenolate in this reaction by UV
spectroscopy the process can be quantified. For one
substrate, the phenyl glucoside 30, the reaction was also
followed by determining the amount of formed glucose
using oxidation with 3,5-dinitrosalicylic acid, which gave
essentially the same result as obtained by following
phenolate formation (Table 1). The catalyzed reaction,
which is obtained by subtraction of parallel uncatalyzed
reactions, follows Michaelis-Menten kinetics as is seen
in the Hanes plot (Figure 1). These kinetic experiments
were carried out with substrate in concentrations of 1-50
mM and cyclodextrin catalyst in concentrations from 0.01
to 0.42 mM with the latter considerably smaller.
kuncat becomes 1200. All together these results show that
the catalyzed reaction depends little on the leaving group
ability.
Remarkably the dideoxy derivative 29 is not a sub-
strate despite it being a much more reactive glycoside than
its structural parent 23. The binding mode may be dis-
rupted by the larger lipophilic surface this substrate has.
Interestingly the thioglycoside 32 is a substrate as well
giving a kcat of 1.8 × 10-4 s-1 and a kcat/kuncat of 19 (Table
1). Thiophenol has a pKa of 6.6 and is thus between
4-nitrophenol and 2-chloro-4-nitrophenol in leaving group
ability.
In Table 2 is shown the kinetic data obtained with the
R-cyclodextrin dicyanohydrin 2. This catalyst has a
similar pH profile as was observed with 1: The optimum
catalysis is found at pH 8.0 in phosphate buffer. Increas-
ing the buffer concentration increases the catalytic rate
so that a kcat/kuncat value of 6-7000 is obtained for
hydrolysis of 22 and 23 in 0.5 M phosphate buffer. A
study of other buffers showed that catalysis in borate,
Hepes, or glycin buffers were less efficient giving a kcat
kuncat value of 157-955. Compound 2 also catalyzed the
hydrolysis of helicin 31, quite effectively in terms of kcat
/
,
The kinetic constants for the catalysis by 1 of the
hydrolysis of different nitrophenyl glycosides under dif-
ferent conditions are shown in Table 1. The kinetic
constants were calculated by nonlinear least-squares
fitting of Vcat vs S at four different substrate concentra-
tions chosen in the interval S ) 0.2 to 5 Km (typically
1-50 mM). In 50 mM phosphate buffer at 59 °C, a Km of
1-10 mM is obtained for reaction of the â-glucoside 22,
with an optimum kcat at pH 8.0 of 5.9 × 10-5 s-1, which
is about 3000 times faster than the uncatalyzed reaction.
Hydrolysis of the stereoisomeric substrates 23, 24, 25,
and 26 (an ortho isomer) was also catalyzed with efficacy,
at pH 7.4, based on a kcat value of 26 (â-galacto) > 22
(â-gluco) > 23 (R-gluco) > 25 (R-galacto) > 24 (R-manno)
or based on a kcat/kuncat value of 22 > 23 > 26 > 25 > 24.
The variation in catalyzed rate is relatively small, about
4-fold, while the variation in kcat/kuncat is up to 10-fold.
By increasing the phosphate concentration the catalytic
rate increases somewhat, so that a kcat/kuncat value of 7922
is obtained for the hydrolysis of 23 (Table 1). However,
from a series of experiments of increasing phosphate
buffer concentration it is seen that no simple relationship
between phosphate concentration and kcat exists for
hydrolysis of 22 (Table 1, see also Figure S1 in the
Supporting Information) so it is most plausible the
increase is mainly due the change in ion strenght.
but the background is very high as well leading to a
mediocre kcat/kuncat value of 95.
As with â-cyclodextrin dicyanohydrin 1 the catalysis
performed by 2 of hydrolysis of 2-chloro-4-nitrophenyl
glucoside 27 and 2,4-dinitrophenyl glucoside 28 is not
enhanced compared to the 4-nitrophenyl glycosides de-
spite the much higher leaving group ability of these
groups. As the background hydrolysis rate is much higher
for these substrates kcat/kuncat is comparatively low par-
ticularly for 28. These substrates are bound with a
similar strength at 22, however, suggesting that poor fit
is not the reason for the drop in kcat/kuncat. For the dideoxy
sugar 29 no catalysis was observed, and it is possible that
any catalysis is overshadowed by the much higher
background hydrolysis of this substrate.
The â-cyclodextrin monocyanohydrin 3 was also found
catalytic on substrates 22-23 and 25-26 (Table 3).
Depending on the substrate the catalysis is 2-20 times
slower than that by 1. Nevertheless it is clear that one
cyanohydrin group is sufficient for catalysis and that two
merely increase catalysis by increasing the chance for
productive binding of substrate.
The catalytic power of other CD derivatives toward
4-nitrophenyl â-D-glucopyranoside (22) hydrolysis is shown
in Table 4. The propyl analogues 17, 19, and 21 afford
no catalysis, similarly to â-cyclodextrin and regardless
of whether the two OH groups point toward or away from
the cavity. This shows that the cyano groups are es-
sential. Dinitrile 14 is catalytic, but with a 250 times
lower catalytic power, showing that the cyanohydrin OH
groups are very important for the catalysis. These
experiments show that the role of the cyano group must
be in its electron withdrawing capacity increasing the
acidity of the hydroxyl group. The dihydrate 15 is a
comparatively good catalyst with a kcat/kuncat value about
60 times lower than that of 1.
The reaction in D2O with pD ) 8.0 was about twice as
fast as the reaction at pH 8.0 showing an solvent
deuterium isotope effect of kH/kD ) 1.5.
Adding electron withdrawing substituents in the phen-
yl group in the substrates 27 and 28 did not change the
catalyzed hydrolysis rate very much, but because the
uncatalyzed rate was increased the kcat/kuncat drops down
to 51 and 6-fold, respectively. The Km is essentially
unchanged. Even though the substrate is still bound
quite tightly the binding may be altered by the introduc-
tion of substituents. The phenyl derivative 30 is a
The above results can be summated to the following
main points: (1) A single cyanohydrin at the primary rim
is sufficient to promote catalysis. (2) The cyano group
itself promotes little catalysis. (3) The acidity of the 6-OH
group appears to be important, since unmodified cyclo-
somewhat slower substrate. For this substrate the kcat
/
(11) Angyal, S. J.; Le Fur, R. Carbohydr. Res. 1984, 126, 15-26.
7220 J. Org. Chem., Vol. 70, No. 18, 2005

Artificial Enzyme-Catalyzed Aryl Glycoside Hydrolysis
TABLE 1. Kinetic Parameters for the Dicyanohydrin-â-cyclodextrin (1)-Catalyzed Hydrolysis of Various Glycosides at
Different pH and 59 °Ca
a The reactions were followed by measuring absorption at 400 nm. *: at 25 °C. †: at 90 °C; #: followed by measuring absorption at 290
nm. §: followed by measuring glucose formation. @: in D2O. Concentration of 1 was 0.42 mM.
J. Org. Chem, Vol. 70, No. 18, 2005 7221

Ortega-Caballero et al.
TABLE 2. Kinetic Parameters for the Dicyanohydrin-r-cyclodextrin 2-Catalyzed Hydrolysis of Various Glycosides at
Different pH and 59 °Ca
a The reactions were followed by measuring absorption at 400 nm. *: at 25 °C. #: followed by measuring absorption at 376 nm.
Concentration of (2) was 0.49 mM.
dextrins or propyl cyclodextrins 17, 19, and 21 not are
catalytic, while 15 is. (4) The catalytic rate does not follow
leaving group ability. Thus a good leaving group such as
2,4-dinitrophenol (pKa ) 4.0), 2-chloro-4-nitrophenol (pKa
) 5.45), and tolylthiophenol (pKa ) 6.6) in the substrate
does not give a faster hydrolysis than 4-nitrophenol (pKa
) 7.15). Nor were the 4-nitrophenol glycosides signifi-
cantly better substrates than 30 even though phenol (pKa
) 9.9) is an almost 1000-fold worse leaving group. (5) A
solvent deuterium isotope effect of 1.5 is observed. (6) No
clear pH dependency is observed, but in phosphate buffer
the catalytic rate increases with increasing pH. (7) There
is no clear phosphate dependence though the catalytic
rate in general increases with increasing the phosphate
concentration.
Points 1-3 indicate that the hydroxyl group and its
acidity is important for the catalysis. Two roles of the OH
group can be imagined: protonation of the exocyclic glyco-
side oxygen by general acid catalysis or nucleophilic cata-
lysis by the alcoholate substituting the aryloxy group. The
lack of importance of leaving group ability (point 4) and
the solvent isotope effect (point 2) suggest that this role
7222 J. Org. Chem., Vol. 70, No. 18, 2005

Artificial Enzyme-Catalyzed Aryl Glycoside Hydrolysis
TABLE 3. Kinetic Parameters for the Cyanohydrin-â-cyclodextrin (3)-Catalyzed Hydrolysis of Various Glycosides at
Different Phosphate Concentrations and 59 °Ca
a The reactions were followed by measuring absorption at 400 nm. Concentration of 3 was 0.27 mM.
TABLE 4. Kinetic Parameters for the Catalysis by Different Cyclodextrin Derivatives of the Hydrolysis of
4-Nitrophenyl-â-D-glucoside at pH 7.4 and 59 °C in 50 mM Phosphate Buffera
a The reactions were followed by measuring absorption at 400 nm. The catalyst concentration was 2.2 mM.
is general acid catalysis as shown in Figure 2, since this
will have a rate determining step that involves deuterium
transfer and is independent of leaving group ability.
The pH dependency (point 6) and the phosphate
dependency (point 7) suggest that the mechanism may
be more complicated than can be elucidated from the
present data. The pH profile appear to be closely associ-
ated to the ionization of phosphate since no rate increase
as a function of pH is seen in other buffers. A small
amount of phosphate clearly has a positive effect, since
the rate is much higher in phosphate buffer, and the most
obvious reason for that could be nucleophilic catalysis by
secondary phosphate. However, the kcat increase obtained
by increasing the phosphate concentration 10-20 times
is only 2-3. Perhaps this behavior can be explained by
a change in rate determining step, so that when phos-
phate is not present the cleavage of the C-OAryl bond
becomes rate determining.
The stereochemistry of the cyanohydrin is likely to be
important though this has not been demonstrated. Previ-
FIGURE 2. Proposed mechanism for the catalysis.
J. Org. Chem, Vol. 70, No. 18, 2005 7223

Ortega-Caballero et al.
69 (404 mg, 0.17 mmol) in ether/MeOH (12 mL/9 mL). The
reaction mixture was stirred overnight at room temperature.
The organic solvent was removed in a vacuum and the water
phase was extracted with CH2Cl2. The organic layer was
washed, dried (MgSO4), filtered, and concentrated in vacuo.
The residue was purified by chromatography (eluent gradient,
EtOAc/pentane 1:3 f 1:2) to afford dicyanohydrin 7 (327 mg,
80%) as a white foam: [R]D +34.4 (c 0.25, CHCl3); IR (KBr)
3335, 3031, 2928, 2865, 2247 (CN), 1496, 1454, 1355, 1208,
1165 cm-1; 1H NMR (400 MHz, CDCl3) δ (ppm) 7.42-7.14 (m,
90H, aromatic-H), 5.48 (d, 1H, 3J1,2 ) 3.6 Hz, H-1), 5.42-5.37
ous work on 6-C-substituted cyclodextrins showed that
these derivatives have very restricted conformational
freedom along the C5-C6 bond as both OH and alkyl
substituents shun the tg conformation.4 Therefore an
important feature in 1, 2, and 3 could be that the R
isomers having the cyanohydrin 6-OH groups are fixed
in the gt conformation pointing toward the binding site.
In summary cyclodextrin cyanohydrins have been
shown to be remarkably efficient artificial enzymes
catalyzing hydrolysis of aryl glycosides with rate in-
creases close to 104. A single cyanohydrin is required for
the catalysis. The catalysis is believed to be general acid
catalysis with the cyanohydrin OH delivering a proton
to the exocyclic oxygen of the glycoside. Phosphate
appears to have an influence and may assist in displacing
the aryloxy group in the second step. Thus cyclodextrin
cyanohydrins mimic part of mechanistic apparatus of
natural glycosidases as the cyanohydrin mimics a pro-
tonating carboxylate. Including other parts of this ap-
paratus such as nucleophilic catalysis may improve the
catalysis further.
3
(m, 2H), 5.31-5.26 (m, 3H), 5.22 (d, 1H, J1,2 ) 3.2 Hz, H-1),
3
5.17 (d, 1H, J1,2 ) 3.2 Hz, H-1), 5.13-5.09 (m, 2H), 5.03-
4.64 (m, 10H), 4.60-4.36 (m, 20H), 4.32-3.93 (m, 10H), 3.87-
3.47 (m, 10H); 13C NMR (100 MHz, CDCl3) δ (ppm) 139.8-
137.5 (Cipso), 129.0-126.7 (CH aromatic), 119.8 (CN), 119.7
(CN), 117.8 (CN), 101.1 (C-1), 101.0 (C-1), 100.2 (C-1), 99.8
(C-1), 99.3 (C-1), 99.1 (C-1), 98.4 (C-1), 97.8 (C-1), 82.5, 82.4,
81.9, 81.7, 81.3, 81.2, 81.1, 81.0, 80.9, 80.8, 80.5, 80.3, 79.6,
79.5, 79.3, 78.7, 76.7, 76.6, 76.2, 76.1, 75.9, 75.7, 75.3, 75.2,
73.9, 73.8, 73.7, 73.5, 73.4, 73.3, 73.2, 73.1, 73.0, 72.9, 72.8,
72.7, 72.6, 72.2, 72.0, 71.8, 71.7, 71.6, 70.4, 70.0, 69.8, 69.5,
69.4, 62.9 (CH2, CH), 60.4 (CH(OH)CN), 60.3 (CH(OH)CN);
MALDI-TOF-MS m/z calcd for C150H154O30N2Na 2486.0484,
found 2486.973.
6A,6D-Di-C-cyano-R-cyclodextrin (2). Compound 7 (327
mg, 0.13 mmol) was dissolved in a mixture of MeOH/EtOAc
(1:1) (12 mL). Then Pd/C (33 mg) and TFA (cat) were added
and the mixture was stirred overnight under hydrogen atmo-
sphere. Filtration through Celite and evaporation of the
solvent gave 2 (133 mg, 100%) as a white solid: [R]D +84.6 (c
0.5, CHCl3); IR (KBr) 3401, 2936, 2252 (CN), 1681, 1436, 1203,
1151, 1033 cm-1; 1H NMR (400 MHz, D2O) δ (ppm) 5.15 (br s,
1H, H-1), 5.03 (br s, 1H, H-1), 4.93 (br s, 4H, H-1), 3.95-3.39
(m, 30H); 13C NMR (100 MHz, CDCl3) δ (ppm) 118.2 (CN),
115.4 (CN), 101.7 (C-1), 101.6 (C-1), 81.8, 81.4, 80.1, 73.2, 72.5,
72.3, 71.9, 71.6, 60.8, 60.6, 59.6 (CH2, CH); MALDI-TOF-MS
m/z calcd for C38H58O30N2Na 1045.2972, found 1045.2972.
Experimental Section
6A,6D-Di-C-cyano-2A-G,3A-G,6B,C,E-G-nonadecakis-O-benzyl-
â-cyclodextrin (5). A mixture of potassium cyanide (649 mg,
9.96 mmol) and ammonium chloride (802 mg, 15 mmol) in
water (10 mL) was added at 0 °C to a solution of 48 (200 mg,
0.07 mmol) in ether/MeOH (1:1) (10 mL). The reaction mixture
was stirred overnight at room temperature. The organic
solvent was removed in a vacuum and the water phase was
extracted with CH2Cl2. The organic layer was washed, dried
(MgSO4), filtered, and concentrated in vacuo. The residue was
purified by chromatography (eluent gradient, EtOAc/pentane
1:4 f 1:3) to afford 5 (156 mg, 77%) as a white foam: [R]D
+41.8 (c 1.0, CHCl3); IR (KBr) 3410, 3029, 2924, 2867, 2242
(CN), 1496, 1453, 1358, 1208, 1095, 1040 cm-1; 1H NMR (400
MHz, CDCl3) δ (ppm) 7.50-7.01 (m, 95H, aromatic-H), 5.78
6A-C-Cyano-2A-G,3A-G,6B-G-eiocosakis-O-benzyl-â-cy-
clodextrin (9). A mixture of potassium cyanide (4.08 g, 63
mmol) and ammonium chloride (5.05 g, 94 mmol) in water (63
mL) was added to a solution of 810 (1.22 g, 0.42 mmol) in Et2O/
MeOH (1:1) (63 mL). The reaction mixture was stirred
overnight at room temperature. The organic solvent was
removed in a vacuum and the water phase was extracted with
CH2Cl2. The organic layer was washed with water, dried
(MgSO4), filtered, and concentrated in vacuo. The residue was
purified by chromatography (eluent gradient, EtOAc/pentane
1:4 f 1:3) to afford 9 (1.06 g, 86%) as an oil: [R]D +32.6 (c 1.0,
CDCl3); IR (film) 3364, 3030, 2925, 2868, 2230 (CN), 1496,
1453, 1356, 1207, 1094, 1040, 1028 cm-1; 1H NMR (400 MHz,
CDCl3) δ (ppm) 7.32-6.86 (m, 100H, aromatic-H), 5.59 (d, 1H,
3J1,2 ) 3.2 Hz, H-1), 5.30-5.16 (m, 3H), 5.15-4.97 (m, 6H),
3
3
(d, 1H, J1,2 ) 4.0 Hz, H-1), 5.75 (d, 1H, J1,2 ) 3.6 Hz, H-1),
5.40-5.31 (m, 4H), 5.20-5.15 (m, 2H), 5.11-4.97 (m, 6H),
4.90-4.21 (m, 44H), 4.18-3.97 (m, 14H), 3.95-3.84 (m, 5H),
3.79-3.47 (m, 15H); 13C NMR (100 MHz, CDCl3) δ (ppm)
139.9-138.0 (Cipso), 128.9-126.8 (CH aromatic), 120.3 (CN),
119.5 (CN), 100.8 (C-1), 100.2 (C-1), 99.5 (C-1), 98.3 (C-1), 98.0
(C-1), 97.5 (C-1), 97.3 (C-1), 82.3, 82.1, 81.9, 81.7, 81.4, 81.2,
81.1, 80.9, 80.6, 80.2, 79.9, 79.8, 78.9, 78.4, 76.7, 76.3, 76.2,
75.6, 75.4, 74.9, 74.6, 74.5, 74.0, 73.9, 73.7, 73.6, 73.5, 73.4,
73.2, 73.1, 73.0, 72.9, 72.7, 72.5, 72.4, 72.2, 72.1, 72.0, 71.8,
71.6, 71.1, 70.0, 69.8, 69.1, 66.1 (CH2, CH), 60.4 (CH(OH)CN),
59.9 (CH(OH)CN); MALDI-TOF-MS m/z calcd for C177H182O35N2-
Na 2918.2420, found 2918.1184.
3
4.96-4.92 (t, 2H, J ) 3.2 Hz), 4.91-4.88 (d, 1H, J1,2 ) 3.2
Hz, H-1), 4.85-4.82 (d, 1H, 3J1,2 ) 4.0 Hz, H-1), 4.76-4.24 (m,
36H), 4.08-3.77 (m, 22H), 3.76-3.38 (m, 15H), 3.40 (dd, 1H,
3J1,2 ) 3.2 Hz, 3J2,3 ) 9.6 Hz, H-2); 13C NMR (100 MHz, CDCl3)
δ (ppm) 139.7-138.0 (Cipso), 129.3-125.5 (CH aromatic), 118.8
(CN), 100.0 (C1), 99.1 (C1), 98.6 (C1), 98.2 (C1), 98.0 (C1), 97.5
(C1), 81.8, 81.5, 81.4, 81.3, 81.2, 80.4, 80.0, 79.9, 79.8, 79.3,
79.2, 79.1, 78.9, 78.7, 77.8, 76.6, 76.2, 76.0, 75.6, 75.5, 74.8,
74.6, 73.9, 73.7, 73.6, 73.5, 73.5, 73.3, 73.1, 73.0, 72.9, 72.8,
72.7, 72.4, 72.3, 72.1, 72.0, 71.8, 71.6, 71.4, 71.3, 70.0, 69.9,
69.5, 69.4, 69.3, 69.2; MALDI-TOF-MS m/z calcd for C183H189O35-
NNa 2983.2937, found 2983.4310.
6-C-Cyano-â-cyclodextrin (3). Compound 9 (1.06 g, 0.36
mmol) was dissolved in 2-methoxyethanol (25 mL). Then Pd/C
(107 mg) and TFA (cat) were added and the mixture was
stirred at room temperature under hydrogen atmosphere until
completion. Filtration through Celite and evaporation of the
solvent gave 3 (303 mg, 73%) as a white solid: [R]D +49.5 (c
0.1, H2O); IR (film) 3364, 2938, 2079, 1684, 1203, 1141, 1054,
(6AR,6DR)-6A,6D-Di-C-cyano-â-cyclodextrin (1). Com-
pound 5 (415 mg, 0.14 mmol) was dissolved in a mixture of
MeOH/EtOAc (1:1) (12 mL). Then Pd/C (42 mg) and TFA (cat)
were added and the mixture was stirred overnight under
hydrogen atmosphere. Filtration through Celite and evapora-
tion of the solvent gave 1 (169 mg, 100%) as a white solid:
[R]D +89.4 (c 1.0, H2O); IR (KBr) 3363, 2932, 2258 (CN), 1678,
1425, 1203, 1156, 1030 cm-1; 1H NMR (400 MHz, D2O) δ (ppm)
3
5.05 (br s, 3H, H-1), 4.95 (br s, 4H, H-1), 4.00 d, 1H, J ) 8.8
Hz), 3.91-3.42 (m, 30H); 13C NMR (100 MHz, D2O) δ (ppm)
119.1 (CN), 102.1 (C-1), 102.0 (C-1), 81.6, 81.5, 81.4, 81.3, 81.2,
80.0, 73.2, 73.1, 72.5, 72.3, 72.1, 72.0, 71.9 (CH), 60.7, 60.5,
59.6 (CH2, CH(OH)CN); MALDI-TOF-MS m/z calcd for
C44H68O35N2Na 1207.3500, found 1207.3739.
6A,6D-Di-C-cyano-2A-F,3A-F,6B,C,E,F-hexadecakis-O-benzyl-
R-cyclodextrin (7). A mixture of potassium cyanide (1.54 g,
23.57 mmol) and ammonium chloride (2.15 g, 40.17 mmol) in
water (24 mL) was added at 0° C to a solution of dialdehyde
7224 J. Org. Chem., Vol. 70, No. 18, 2005

Artificial Enzyme-Catalyzed Aryl Glycoside Hydrolysis
1
3
1033 cm-1; H NMR (400 MHz, D2O) δ (ppm) 5.07-5.03 (m,
4.99 (d, 5H, J1,2 ) 3.6 Hz, H-1), 4.96 (s, 2H, H-1), 4.02 (br t,
1H, H-1), 5.0-4.92 (m, 6H, H-1), 4.00-3.36 (m, 41 H); MALDI-
TOF-MS m/z calcd for C43H69O35NNa 1182.3548, found
1183.0570.
1H, 3J ) 7.4 Hz, H-5), 3.78-3.68 (m, 22H), 3.61-3.47 (m, 18H),
3.44 (t, 2H, 3J ) 9.2 Hz), 3.37 (t, 2H, 3J ) 9.4 Hz), 3.08 (br d,
2
2H, H-6A or 6D), 2.83 (dd, 1H, J6,6′ ) 17.4 Hz, H-6D or 6A),
2.82 (dd, 1H, 2J6,6′ ) 16.8 Hz, H-6′D or 6′A); 13C NMR (100 MHz,
D2O) δ (ppm) 118.8 (CN), 102.2 (C-1), 102.1 (C-1), 101.9 (C-1),
84.6, 81.5, 81.4, 73.2, 72.7, 72.2, 72.0, 67.5, 60.5 (CH), 20.7
(CH2CN); MALDI-TOF-MS m/z calcd for C44H68O33N2Na
1175.3602, found 1175.3444.
6A,6D-Dideoxy-6A,6D-diiodononadecakis-O-benzyl-â-cy-
clodextrin (12). A mixture of 6A, 6D-diol-nonadecakis-O-
benzyl-â-cyclodextrin (11)7 (1.40 g, 0.49 mmol), iodine (749 mg,
2.95 mmol), triphenylphosphine (774 mg, 2.95 mmol), and
imidazole (402 mg, 5.90 mmol) in toluene (70 mL) was
vigorously stirred at 75 °C for 16 h. To reaction mixture was
added an equal volume of sat. NaHCO3 and the mixture was
stirred 5 min. An excess of iodine was removed by the addition
of aqueous sat. Na2S2O3. The organic layer was diluted with
EtOAc (200 mL) and washed with water (80 mL), dried
(MgSO4), filtered, and concentrated in vacuo. The residue was
purified by chromatography (eluent, EtOAc/pentane 1:4 f 1:3)
to afford 12 (1.45 g, 96%) as a white foam: [R]D +34.4 (c 1.0,
CHCl3); 1H NMR (400 MHz, CDCl3) δ (ppm) 7.22-6.96 (m,
6A,6D-Dialdehydo-â-cyclodextrin (15). Compound 48 (196
mg, 0.07 mmol) was dissolved in a mixture of MeOH/EtOAc
(1:1) (5 mL). Then Pd/C (20 mg) and TFA (cat) were added
and the mixture was stirred overnight under hydrogen atmo-
sphere. Filtration through Celite and evaporation of the
solvent gave 15 (80 mg, 100%) as a white solid: [R]D +83.6 (c
1.0, H2O); IR (KBr) 3343, 2937, 1681, 1438, 1206, 1153, 1031
cm-1; H NMR (400 MHz, D2O) δ (ppm) 5.29 (br s, 2H, OH),
1
5.01 (br s, 2H, OH), 4.94 (br s, 7H, H-1), 3.84-3.75 (m, 31H),
3.54-3.47 (m, 20H); 13C NMR (100 MHz, D2O) δ (ppm) 102.0
(C-1), 87.4, 82.3, 81.2, 73.2, 72.9, 72.2, 71.9 (CH), 60.4 (C-6);
MALDI-TOF-MS m/z calcd for C42H70O37Na 1189.3494, found
1189.3035 (dihydrate); for C42H68O36Na 1171.3388, found
1171.2986 (monohydrate); for C42H66O35Na 1153.3282, found
1153.2978 (dialdehyde).
3
95H, aromatic-H), 5.24 (d, 1H, J1,2 ) 4.0 Hz, H-1), 5.18 (d,
1H, J1,2 ) 3.6 Hz, H-1), 5.10 (d, 1H, 2J ) 12.8 Hz, CHPh),
3
2
3
5.08 (d, 1H, J ) 10.8 Hz, CHPh), 5.06 (d, 1H, J1,2 ) 3.2 Hz,
H-1), 5.04 (d, 1H, 2J ) 11.2 Hz, CHPh), 5.03 (d, 1H, 3J1,2 ) 3.6
2
2
Hz, H-1), 5.10 (d, 1H, J ) 10.8 Hz, CHPh), 4.91 (d, 1H, J )
11.2 Hz, CHPh), 4.89 (d, 1H, 3J1,2 ) 3.6 Hz, H-1), 4.86 (d, 1H,
3J1,2 ) 3.6 Hz, H-1), 4.82 (d, 1H, 2J ) 11.2 Hz, CHPh), 4.77 (d,
(6AS,6DS)-6A,6D-Di-C-propyl-â-cyclodextrin (17). Com-
pound 168 (500 mg, 0.37 mmol) was dissolved in a mixture of
MeOH/EtOAc (1:1) (50 mL). Then Pd/C (100 mg) and TFA (cat)
were added and the mixture was stirred overnight under
hydrogen atmosphere. Filtration through Celite and evapora-
tion of the solvent gave 17 (203 mg, 100%) as a white foam.
[R]D +94.4 (c 1.0, H2O); 1H NMR (400 MHz, D2O, 80 °C) δ
(ppm) 5.65-5.45 (m, 7H, H-1), 4.50-4.21 (m, 23H), 4.20-3.98
(m, 18H), 2.16-2.04 (m, 2H), 2.02-1.91 (m, 2H), 1.88-1.70
(m, 4H), 1.34 (br s, 6H, CH3); 13C NMR (100 MHz, D2O) δ (ppm)
102.0 (C-1), 81.2, 81.1, 80.9, 79.9, 73.7, 73.6, 73.4, 73.2, 72.9,
72.6, 72.2, 72.1, 71.9, 67.6, 67.5, 60.3 (CH), 60.1, 59.9, 59.8
(C-6), 35.2, 18.9, 18.7, 18.6 (CH2), 13.3 (CH3); MALDI-TOF-
MS m/z calcd for C48H82O35Na 1241.4534, found 1241.4782.
(6AR,6DR)-6A,6D-Di-C-propyl-â-cyclodextrin (19). Com-
pound 188 (109 mg, 0.37 mmol) was dissolved in a mixture of
MeOH/EtOAc (1:1) (20 mL). Then Pd/C (40 mg) and TFA (cat)
were added and the mixture was stirred overnight under
hydrogen atmosphere. Filtration through Celite and evapora-
tion of the solvent gave 19 (43 mg, 100%) as a white foam.
[R]D +55.2 (c 1.0, H2O); 1H NMR (400 MHz, D2O, 80 °C) δ
(ppm) 5.65-5.45 (m, 7H, H-1), 4.50-4.21 (m, 23H), 4.20-3.98
(m, 18H), 2.16-2.04 (m, 2H), 2.02-1.91 (m, 2H), 1.88-1.70
(m, 4H), 1.34 (br s, 6H, CH3); 13C NMR (100 MHz, D2O) δ (ppm)
102.0 (C-1), 81.2, 81.1, 80.9, 79.9, 73.7, 73.6, 73.4, 73.2, 72.9,
72.6, 72.2, 72.1, 71.9, 67.6, 67.5, 60.3 (CH), 60.1, 59.9, 59.8
(C-6), 35.2, 18.9, 18.7, 18.6 (CH2), 13.3 (CH3); MALDI-TOF-
MS m/z calcd for C48H82O35Na 1241.4534, found 1241.4789.
(6ARS,6DSR)-6A,6D-Di-C-propyl-â-cyclodextrin (21). Com-
pound 208 (350 mg, 0.37 mmol) was dissolved in a mixture of
MeOH/EtOAc (1:1) (20 mL). Then Pd/C (50 mg) and TFA (cat)
were added and the mixture was stirred overnight under
hydrogen atmosphere. Filtration through Celite and evapora-
tion of the solvent gave 21 (141 mg, 100%) as a white foam.
[R]D +91.9 (c 1.0, H2O); 1H NMR (400 MHz, D2O, 80 °C) δ
(ppm) 5.65-5.45 (m, 7H, H-1), 4.50-4.21 (m, 23H), 4.20-3.98
(m, 18H), 2.16-2.04 (m, 2H), 2.02-1.91 (m, 2H), 1.88-1.70
(m, 4H), 1.34 (br s, 6H, CH3); 13C NMR (100 MHz, D2O) δ (ppm)
102.0 (C-1), 81.2, 81.1, 80.9, 79.9, 73.7, 73.6, 73.4, 73.2, 72.9,
72.6, 72.2, 72.1, 71.9, 67.6, 67.5, 60.3 (CH), 60.1, 59.9, 59.8
(C-6), 35.2, 18.9, 18.7, 18.6 (CH2), 13.3 (CH3); MALDI-TOF-
MS m/z calcd for C48H82O35Na 1241.4534, found 1241.4660.
Hydrolysis of Compound 1. To a solution of 1 (82 mg,
0.07 mmol) in water (6 mL) was added 20 mL of Amberlite
IR-120 (H+), and the mixture was stirred at 100 °C for 48 h.
The resin was removed by filtration and NaBH4 (132 mg, 3.5
mmol) was added to the filtrate. Then the reaction mixture
was stirred for 30 min at room temperature and Amberlite
2
1H, J ) 10.8 Hz, CHPh), 4.66-4.60 (m, 8H), 4.52-4.26 (m,
25H), 4.15 (t, 2H, 3J ) 8.2 Hz), 3.93-3.82 (m, 20H), 3.70-
3.60 (m, 7H), 3.50-3.26 (m, 15H); 13C NMR (100 MHz, CDCl3)
δ (ppm) 139.5-139.2 (Cipso), 138.7-138.2 (Cipso), 128.6-127.1
(CH aromatic), 99.3 (C-1), 98.9 (C-1), 98.6 (C-1), 98.5 (C-1),
98.1 (C-1), 83.7, 81.7, 81.1, 81.0, 80.8, 80.5, 80.3, 80.1, 79.7,
79.6, 79.5, 79.0, 78.7, 78.5, 78.0, 76.2, 75.8, 75.3, 75.1, 74.9,
73.8, 73.7, 73.6, 73.2, 73.0, 72.9, 72.8, 72.7, 72.0, 71.7, 71.6,
71.2, 70.5, 69.9, 69.5, 69.4 (CH2, CH), 9.9 (CH2I), 9.3 (CH2I);
MALDI-TOF-MS m/z calcd for C175H182O33I2Na 3088.0550,
found 3088.0424.
6A,6D-Di-C-cyano-6A,6D-dideoxynonadecakis-O-benzyl-
â-cyclodextrin (13). Potassium cyanide (457 mg, 7.01 mmol)
was added to a solution of 12 (1.02 g, 0.33 mmol) in DMF (25
mL). The reaction mixture was stirred at 80 °C for 17 h. The
mixture was cooled and water (30 mL) and EtOAc (60 mL)
were added. The aqueous layer was washed with EtOAc (30
mL) and the combined organic layer was washed with water
(40 mL), dried (MgSO4), filtered, and concentrated in vacuo.
The residue was purified by chromatography (eluent, EtOAc/
pentane 1:3 f 1:2) to afford 13 (817 mg, 85%) as a white
foam: [R]D +35.7 (c 1.0, CHCl3); IR (KBr) 3482, 2924, 2867,
2252 (CN), 1496, 1453, 1356, 1208, 1094, 1040 cm-1; 1H NMR
(400 MHz, CDCl3) δ (ppm) 7.29-6.89 (m, 95H, aromatic-H),
3
5.27 (d, 1H, J1,2 ) 3.6 Hz, H-1), 5.09-4.97 (m, 7H), 5.00 (d,
1H, 3J1,2 ) 3.6 Hz, H-1), 4.84 (d, 1H, 3J1,2 ) 3.0 Hz, H-1), 4.80
3
(d, 1H, J1,2 ) 3.2 Hz, H-1), 4.71-4.12 (m, 36H), 4.23-3.74
3
(m, 26H), 3.61 (t, 2H, J ) 11 Hz), 3.51-3.31 (m, 10H), 3.27
(dd, 1H, 3J2,3 ) 9.8 Hz, 3J1,2 ) 3.0 Hz, H-2), 2.81(dd, 2H, 2J6,6′
3
2
) 14.4 Hz, J5,6 ) 7.6 Hz, H-6A or 6D), 2.53 (dd, 1H, J6,6′
)
3
2
17.2 Hz, J5,6) 7.2 Hz, H-6D or 6A), 2.47 (dd, 1H, J6,6′ ) 17.2
Hz, 3J5,6 ) 7.8 Hz, H-6′D or 6′A); 13C NMR (100 MHz, CDCl3) δ
(ppm) 139.5-138.1 (Cipso), 128.8-127.0 (CH aromatic), 118.1
(CN), 117.8 (CN), 99.2 (3 × C-1), 98.8 (C-1), 98.5 (C-1), 98.3
(C-1), 98.0 (C-1), 82.1, 81.0, 80.8, 80.4, 80.2, 80.0, 79.6, 79.4,
79.0, 76.2, 75.8, 75.1, 74.5, 73.7, 73.5, 73.2, 73.0, 72.9, 72.8,
72.0, 71.8, 69.8, 69.5, 69.0, 67.9, 67.6 (CH2, CH), 22.3
(CH2CN), 21.8 (CH2CN); MALDI-TOF-MS m/z calcd for
C
177H182O33N2Na 2886.2522, found 2886.2307.
6A,6D-Di-C-cyano-6A,6D-deoxy-â-cyclodextrin (14). Com-
pound 13 (1.12 g, 0.39 mmol) was dissolved in a mixture of
MeOH/EtOAc (1:1) (30 mL). Then Pd/C (112 mg) and TFA (cat)
were added and the mixture was stirred overnight under
hydrogen atmosphere. Filtration through Celite and evapora-
tion of the solvent gave 14 (445 mg, 99%) as a white solid:
[R]D +86.3 (c 1.0, H2O); IR (KBr) 3405, 2929, 2258 (CN), 1676,
1420, 1156, 1079, 1033 cm-1; 1H NMR (400 MHz, D2O) δ (ppm)
J. Org. Chem, Vol. 70, No. 18, 2005 7225

Ortega-Caballero et al.
IR-120 was added until pH was acid. The resin was removed
by filtration and the solvent was removed. The residue was
coevaporated with MeOH several times to remove boronic acid,
to give a residue containing 10 and D-glucitol. 10: 13C NMR
(100 MHz, D2O) δ (ppm) 72.7, 72.5, 71.6, 71.5, 69.4, 62.7, 62.3.
This is in agreement with the D-glyco-D-gluco-heptitol.11
4-Nitrophenyl 2,3-Dideoxy-R-D-glucopyranoside (29).
This compound was made from 1,4,6-tri-O-acetyl-2,3-dideoxy-
R,â-D-glucopyranose12 and 4-nitrophenol, using a general
procedure previosuly described for other arylglycosides.13 The
product nitrophenyl 2,3-dideoxy-4,6-di-O-acetyl-R-D-glucopy-
ranoside was purified by Flash chromatography in pentane-
ether (1:1) and was obtained in 25% yield in an unseparable
mixture containing 10-20% of the â-anomer. 1H NMR (400
different concentrations mixed with 1 mL of phosphate or other
buffer containing either cyclodextrin derivative (0.025-5 mg)
or nothing as control. The reactions were followed continuously
at 25 or 59 °C, using UV absorption at 400 nm for the
nitrophenyl substrates or 376 nm for 31. The reaction of 30
was monitored by taking out samples of the reaction at 90 °C,
diluting 4 times with pH 10 buffer before measuring absorption
at 290 nm. Alternatively glucose formation was determined
by heating samples with 3,5-dinitrosalicylic acid at 100 °C and
measuring absorption at 540 nm.15 The reactions were moni-
tored for 3-18 h. Velocities were determined as the slope of
the progress curve of each reaction. Uncatalyzed velocities
were obtained directly from the control samples. Catalyzed
velocities were calculated by subtracting the uncatalyzed
velocity from the velocity of the appropriate cyclodextrin
containing sample. The catalyzed velocities were used to
construct a Hanes plot ([S]/V vs [S]) from which Km and Vmax
were determined. kcat was calculated as Vmax/[cyclodextrin].
kuncat was determined as the slope from a plot of Vuncat versus
[S]. The inhibition experiments were made by adding cyclo-
pentanol (15 µL) to an catalyzed sample.
3
MHz, CDCl3) δ (ppm) 8.21 (d, 2H, J ) 9.2 Hz, aromatic-H),
3
7.18 (d, 2H, J ) 9.2 Hz, aromatic-H), 5.66 (s, 1H, H-1), 4.83
(dt, 1H, 3J ) 5.2 Hz, H-4), 4.24 (dd, 1H, 3J5,6 ) 5.2 Hz, 2J6,6′
)
3
3
12.0 Hz, H-6), 4.03 (dd, 1H, J5,6′ ) 2.0 Hz, J6,6′ ) 12.0 Hz,
3
3
3
H-6′), 3.90 (ddd, 1H, J5,6′ ) 2.0 Hz, J5,6 ) 5.2 Hz, J4,5 ) 5.6
Hz, H-5), 2.19-2.00 (m, 4H, H-2,2′,3,3′), 2.06 (s, 3H, CH3CO),
2.00 (s, 3H, CH3CO).
The acetates were removed by conventional Zemplen deacety-
lation14 giving an essentially quantitative yield of the product
29 in an unseparable mixture containing 10-20% of the
Acknowledgment. This work has been supported
by the Lundbeck foundation, The Danish National
Science Research Council, and the Ramo´n Areces Foun-
dation (F.O.C.). We also thank Annette Nordestgaard
for Technical Assistance.
1
â-anomer. H NMR (400 MHz, CD3OD) δ (ppm) 8.21 (d, 2H,
3
3J ) 9.2 Hz, aromatic-H), 7.28 (d, 2H, J ) 9.2 Hz, aromatic-
H), 5.72 (s, 1H, H-1), 3.71 (dd, 1H, 3J5,6 ) 2.8 Hz, 2J6,6′ ) 12.0
3
3
Hz, H-6), 3.65 (dd, 1H, J5,6′ ) 5.2 Hz, J6,6′ ) 12.0 Hz, H-6′),
3
3
3.60-3,63 (m, 1H, H-4), 3.47 (ddd, 1H, J5,6′ ) 2.8 Hz, J5,6
5.2 Hz, H-5), 2.09-1.93 (m, 4H, H-2,2′,3,3′); HRMS m/z calcd
for C12H15O6NNa 292.0797, found 292.0801.
)
Supporting Information Available: Copies of 13C or 1H
NMR spectra of new compounds and a plot of kcat vs phosphate
concentration for the hydrolysis of 22 catalyzed by 1. This
material is available free of charge via the Internet at
http://pubs.acs.org.
Procedure for Determining the Rate of Hydrolysis.
Each assay was performed on 2 mL samples prepared from 1
mL aqueous solutions of the appropriate aryl glycoside at
JO050861W
(12) Jiang, Y.; Isobe, M. Tetrahedron 1996, 52, 2877-92.
(13) Smits, E.; Engberts, J. B. F. N.; Kellogg, R. M.; van Doren, H.
A. J. Chem. Soc., Perkin Trans. I 1996, 2873-2877.
(14) Shafizadeh, F.; Stacey, M. J. Chem. Soc. 1957, 4612-4615.
(15) Miller, G. L. Anal. Chem. 1959, 31, 426.
7226 J. Org. Chem., Vol. 70, No. 18, 2005