A cyclodextrin-based molecular reactor to template the formation of indigoid dyes

Article abstract of DOI:10.1016/S0040-4039(03)01403-5

N,N′-Bis(6^A-deoxy-β-cyclodextrin-6 ^A-yl)urea behaves as a molecular reactor to bias competing reactions of indoxyl anion and isatin-5-sulfonate in water, to give indigo and indirubin-5′-sulfonate. It appears that the cyclodextrin dimer increases the relative reactivity of the isatin-5-sulfonate, by selectively complexing the reactive form. The molecular host also aligns the isatinsulfonate with indoxyl anion to favour production of indirubin-5′-sulfonate, with the result that the ratio of indigo and indirubin-5′-sulfonate produced is altered by a factor of at least 3500, without a substantial loss of yield.

Full text of DOI:10.1016/S0040-4039(03)01403-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 5815–5818
A cyclodextrin-based molecular reactor to template the
formation of indigoid dyes
a,†
a,
b
Jason B. Harper, Christopher J. Easton * and Stephen F. Lincoln
a
Research School of Chemistry, Australian National University, Canberra ACT 0200, Australia
b
Department of Chemistry, University of Adelaide, Adelaide SA 5005, Australia
Received 2 May 2003; accepted 5 June 2003
A
A
Abstract—N,N%-Bis(6 -deoxy-b-cyclodextrin-6 -yl)urea behaves as a molecular reactor to bias competing reactions of indoxyl
anion and isatin-5-sulfonate in water, to give indigo and indirubin-5%-sulfonate. It appears that the cyclodextrin dimer increases
the relative reactivity of the isatin-5-sulfonate, by selectively complexing the reactive form. The molecular host also aligns the
isatinsulfonate with indoxyl anion to favour production of indirubin-5%-sulfonate, with the result that the ratio of indigo and
indirubin-5%-sulfonate produced is altered by a factor of at least 3500, without a substantial loss of yield.
©
2003 Elsevier Ltd. All rights reserved.
Cyclodextrins form inclusion complexes with small
organic compounds in water and catalyse reactions of the
Molecular templates for carbonꢀcarbon bond-forming
reactions have also been developed. Imidazole-substi-
tuted cyclodextrins have been used to control intramolec-
1
–5
included species. In this regard they may be considered
enzyme mimics. They have also been exploited as molec-
ular reactors, where they behave as nanoscale containers
to control the assembly of reactants and change the
12,13
ular aldol reactions,
dehydroalanine has been bound
to a cyclodextrin to bring about its reaction with indole
14
to give tryptophan, and dipolarophiles have been
attached to cyclodextrins to reverse the regioselectivity of
6–17
outcomes of chemical transformations.
Probably the
15,16
most straightforward examples of cyclodextrin-based
molecular reactors are those that involve a change in the
regioselectivity of reaction as a result of a substrate being
included in such a way as to restrict access of a
6–8
cycloadditions with nitrile oxides.
In another exam-
ple, we exploited the cyclodextrin dimer 2a as a molecular
reactor, to bias the competition between oxidative
dimerisation of the 1H-indol-3-ol anion (indoxyl anion)
6
–11
(
(
(
3) and its condensation with 1H-indoline-2,3-dione
reagent.
Early work in this area by Breslow et al.,
2,2%
isatin) (4a), to give D -biindoline-3,3%-dione (indigo)
showed that cyclodextrins alter the regioselectivity of
aromatic substitution. When anisole is included in a
cyclodextrin, the ortho positions are shielded from chlo-
rination with hypochlorous acid, while the para position
is still accessible.
2,3%
5) and
D
-biindoline-2%,3-dione (indirubin) (6a),
17
respectively (Scheme 1). Our rationale for using this
template was based on the particularly strong binding of
1
8
Tropaeolin 000 No. 2 (7) by the dimer 2a, and the
similarity between the orientation of the dimer 2a
required for efficient binding of the guest 7 and that
necessary to favour reaction of indoxyl anion (3) with
isatin (4a) to give indirubin (6a) (Fig. 1). In the event,
addition of the dimer 2a changed the ratio of formation
of indigo (5) to indirubin (6a) to ca. 1:30, from ca. 1:1
in either the absence of a cyclodextrin or in the presence
of the host 1 (Table 1). The cyclodextrin dimers 2b and
2
c bound the Tropaeolin 7 much less effectively and they
had little effect on the ratio of formation of the indigoid
dyes 5 and 6a.
Keywords: molecular reactors; templates; inclusion complexes;
cyclodextrins; dyes.
Unfortunately the use of the cyclodextrins was associated
with very substantial decreases in the combined yields of
the dyes 5 and 6a, from 29% in the absence of a
cyclodextrin, to 5% in the presence of b-cyclodextrin (1)
and ca. 1% in the presence of the dimer 2a. This can
*
Corresponding author. Tel.: +61 2 6125 8201; fax: +61 2 6125 8114;
e-mail: easton@rsc.anu.edu.au
Present address: School of Chemical Sciences, University of New
South Wales, Sydney NSW 2052, Australia.
†
0
040-4039/$ - see front matter © 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)01403-5

5
816
J. B. Harper et al. / Tetrahedron Letters 44 (2003) 5815–5818
Table 1. Percentage yields of the dyes 5 and 6a from
reactions of indoxyl anion (3) and isatin (4a) at ambient
17
temperature in 0.01 M borate buffer at pH 10.0
Cyclodextrin
Product
5
6a
–
16
13
1
2.5
2.5
2
2
2
a
b
c
0.03
0.2
0.5
1.0
0.6
0.7
be attributed to the inclusion of the substrates 3 and 4a
within each of the cyclodextrins increasing their effec-
tive steric bulk and therefore decreasing the frequency
of their productive collisions to give the dyes 5 and 6a.
At the same time the rate of hydrolytic decomposition
of the substrates 3 and 4a is largely unaffected, so these
processes become more dominant. To overcome this, it
is necessary to avoid decreasing the reactivity of one or
both of the substrates 3 and 4a towards formation of
the dyes 5 and 6a as a result of complex formation and
we are now able to report that this has been accom-
plished by using 2,3-dioxo-2,3-dihydro-1H-indole-5-sul-
fonate (4b) in place of isatin (4a). In this case the
urea-linked cyclodextrin dimer 2a templates the reac-
tion to give indirubin-5%-sulfonate (6b), without signifi-
cantly reducing the yield. Further, the product bias
displayed by the cyclodextrin 2a is much greater and
changes the ratio of formation of the dyes 5 and 6a by
a factor of at least 3500.
Scheme 1.
The reactions of indoxyl anion (3) and the isatinsul-
fonate 4b were carried out as described previously for
17
the non-sulfonated isatin (4a). The anion 3 was gener-
ated in situ through hydrolysis of the corresponding
acetate (1 mM) in 0.01 M aqueous borate buffer at pH
1
0.0, containing a ca. 6 molar excess of the isatinsul-
fonate 4b and either no cyclodextrin, b-cyclodextrin (1)
(
20 mM) or one of the linked species 2a–c (10 mM).
After 16 h, the reactions were quenched through acidifi-
cation, and the mixtures were then partitioned with
chloroform. HPLC analysis on an Alltech Econosil
column (5 mM silica, 4.6×250 mm), eluting with ace-
tone–chloroform and monitoring at 550 nm using a
system calibrated with authentic samples, showed that
the organic phase contained all the indigo (5) while all
the indirubinsulfonate 6b remained in the aqueous
phase. The separate phases were therefore easily
analysed using UV spectroscopy, based on the molar
extinction coefficients of indigo (5) and the indiru-
−
1
−1
binsulfonate 6b of 14000 and 7890 M cm , in chloro-
form and water, respectively, to determine the yields of
the dyes 5 and 6b, and the results are shown in Table 2.
Addition of b-cyclodextrin (1) changes the ratio of
indigo (5) to the sulfonate 6b produced, from ca. 18:1
to 1:7; that is, by a factor of ca. 120. This effect is most
probably related to the fact that in solution the sulfo-
Figure 1. Schematic representation of the orientation of the
cyclodextrin dimer 2a required to (a) bind the dye 7 and (b)
template the synthesis of indirubin (6a) and the sulfonate 6b.

J. B. Harper et al. / Tetrahedron Letters 44 (2003) 5815–5818
5817
Table 2. Percentage yields of the dyes 5 and 6b from
reactions of indoxyl anion (3) and isatin-5%-sulfonate (4b)
at ambient temperature in 0.01 M borate buffer at pH
mM), but under otherwise identical conditions, the
same resonances are observed for the isatin 4b and the
hydrate 8 but in a ratio of ca. 4:1. The cyclodextrin 1
biases the ratio by complexing the non-hydrated form
10.0
4
b in preference to the hydrate 8 (Scheme 2).
Cyclodextrin
Product
In equivalent solutions buffered at pH 9.0, some peak
5
6b
1.4
11
22
36
16
1
broadening occurs in the H NMR spectra, presumably
due to the formation of the anion 9 and rapid intercon-
version of the indolines 4b, 8 and 9 on the NMR
time-scale. Analogous solutions buffered at pH 9.5 and
–
1
25
1.6
a
2a
2b
2c
<0.1
1
1.8
6.0
above gave rise to H NMR spectra in which peak
broadening is too extensive for it to be practical to
assign resonances to individual species. This broadening
occurs both in the presence and absence of the
cyclodextrin 1, but the extent of the broadening is less
when the host 1 is present.
a
Below the detection limit of the method.
nate 4b exists partially as the hydrate 8, which has a
pKa of 9.55 and deprotonates to give the anion 9
Scheme 2). The cyclodextrin 1 appears to complex
the isatin 4b and the corresponding hydrate 8 in prefer-
ence to the anion 9. Evidence for this was obtained
using H NMR spectroscopy.
1
9
The effect of the cyclodextrin 1 is difficult to quantify
from these results and decomposition of the isatin 4b at
(
20
high pH prevents a more thorough analysis. Never-
1
theless, the broadening in the H NMR spectra of
1
solutions containing the cyclodextrin 1 and buffered at
pH 10.0 is less than that seen in solutions buffered at
pH 9.5 that did not contain any cyclodextrin. This
At neutral pH and below, buffered aqueous solutions
prepared using the sodium salt of the sulfonate 4b (5
mM) showed two discrete sets of resonances in their H
NMR spectra. These were observed at l 7.09 (d, J=7
Hz, C7-H), 7.95 (s, C4-H) and 7.99 (d, J=7 Hz, C6-H),
and at l 7.07 (d, J=7 Hz, C7-H), 7.77 (s, C4-H) and
implies that the cyclodextrin 1 increases the pK value
1
a
of the hydrate 8 by more than 0.5 units. This is
consistent with complexation of the isatin 4b and the
hydrate 8 in preference to the anion 9, as might be
expected due to the increased ionic character of the
latter. There is ample literature precedent for such an
7
.81 (d, J=7 Hz, C6-H), corresponding to the isatin 4b
19
and the hydrate 8, respectively. Integration of these
signals shows that the species 4b and 8 are present in a
ratio of ca. 2:1. In the presence of b-cyclodextrin (1) (10
effect, since cyclodextrins are known to alter the pK
values of phenols and benzoic acids through selec-
tive complexation of the acid forms.
a
21
22
A consequence of the complexation of the non-
hydrated isatin 4b in preference to the hydrate 8, and
the change in the pK value of the hydrate 8, is that at
a
high pH the cyclodextrin 1 substantially increases the
proportion of the isatin 4b present in the mixture. Since
it is this species that reacts with indoxyl anion (3) to
form the sulfonated indirubin 6b, the addition of the
host 1 increases the amount available so that more of
the dye 6b is produced. A similar effect is observed with
each of the dimeric hosts 2b and 2c.
Thus, the reactivity of indoxyl anion (3) and isatin (4a)
is reduced by complexation in either b-cyclodextrin (1)
or the dimers 2b and 2c, and the yields of the reactions
to give indigo (5) and indirubin (6a) are reduced as a
consequence (Table 1). By comparison, the hosts 1 and
2
4
b,c do not reduce the reactivity of the isatinsulfonate
b, at least to the same extent. Any loss of reactivity as
a result of inclusion is offset, and may even be out-
weighed, by selective complexation of the reactive form
4
b. As a result, the cross-coupling of indoxyl anion (3)
with the isatinsulfonate 4b, to give the indirubinsul-
fonate 6b, is favoured by the enhanced reactivity of the
sulfonate 4b relative to that of the anion 3. Further, the
combined yields of the dyes 5 and 6b are not signifi-
cantly decreased (Table 2). The effects are different
with isatin (4a) and the isatinsulfonate 4b because only
the latter is hydrated in aqueous solutions to any
meaningful extent.
Scheme 2. Complexation of the isatin 4b and the correspond-
ing hydrate 8 within the cyclodextrin 1 limits formation of the
dianion 9.

5
818
J. B. Harper et al. / Tetrahedron Letters 44 (2003) 5815–5818
In the case of the urea-linked dimer 2a, superimposed
on the effect of the host to increase the proportion of
the isatinsulfonate 4b present in the mixture is a tem-
plating effect analogous to that seen in the reactions of
indoxyl anion (3) with isatin (4a). This is apparent from
comparison of the effects of b-cyclodextrin (1) and the
dimer 2a on the yields of the dyes 5 and 6a (Table 1),
and 5 and 6b (Table 2). The preferred geometry of
alignment of the cyclodextrin annuli of the dimer 2a
further favours formation of the isatinsulfonate 6b and
limits the production of indigo 5 (Fig. 1). The net effect
of the molecular reactor 2a is therefore to change the
ratio of formation of the dyes 5 and 6b by a factor of
at least 3500 without a substantial loss of yield.
4. Breslow, R.; Dong, S. D. Chem. Rev. 1998, 98, 1997.
5. Barr, L.; Easton, C. J.; Lee, K.; Lincoln, S. F.; Simp-
son, J. S. Tetrahedron Lett. 2002, 43, 7797.
6. Breslow, R.; Campbell, P. J. Am. Chem. Soc. 1969, 91,
3085.
7. Breslow, R.; Campbell, P. Bioorg. Chem. 1971, 1, 140.
8. Breslow, R.; Kohn, H.; Siegel, B. Tetrahedron Lett.
1
976, 17, 1645.
. Ch eˆ nevert, R.; Ampleman, G. Can. J. Chem. 1987, 65,
07.
0. Komiyama, M.; Hirai, H. J. Am. Chem. Soc. 1983, 105,
018.
1. Komiyama, M.; Hirai, H. J. Am. Chem. Soc. 1984, 106,
74.
2. Desper, J. M.; Breslow, R. J. Am. Chem. Soc. 1994, 116,
2081.
9
3
1
1
1
2
1
In conclusion, by using the isatinsulfonate 4b in place
of isatin (4a), it has been possible to overcome the loss
of yield resulting from the decrease in reactivity associ-
ated with inclusion in the cyclodextrin dimer 2a. As a
result the effect of the host 2a to bias the ratio of
formation of indigoid dyes in water at ambient temper-
ature is also greatly magnified. Therefore the dimer 2a
is demonstrated to be a very efficient molecular reactor.
1
1
1
3. Breslow, R. Acc. Chem. Res. 1995, 28, 146.
4. Weiner, W.; Winkler, J.; Zimmerman, S. C.; Czarnik, A.
W.; Breslow, R. J. Am. Chem. Soc. 1985, 107, 4093.
5. Meyer, A. G.; Easton, C. J.; Lincoln, S. F.; Simpson, G.
W. Chem. Commun. 1997, 1517.
1
1
1
1
6. Meyer, A. G.; Easton, C. J.; Lincoln, S. F.; Simpson, G.
W. J. Org. Chem. 1998, 63, 9069.
7. Easton, C. J.; Harper, J. B.; Lincoln, S. F. New J.
Chem. 1998, 22, 1163.
8. Haskard, C. A.; May, B. L.; Kurucsev, T.; Lincoln, S.
F.; Easton, C. J. J. Chem. Soc., Faraday Trans. 1997,
Acknowledgements
Support from the Australian Research Council and the
award of the Shell Australia Postgraduate Scholarship
to J.B.H. are gratefully acknowledged.
93, 279.
1
2
9. Stunzi, H. Aust. J. Chem. 1981, 365.
0. Isbell, A. F.; Henze, H. R. J. Am. Chem. Soc. 1944, 66,
2096.
21. (a) Werner, T. C.; Iannacone, J. L.; Amoo, M. N. J.
Inclusion Phenom. Mol. Recognit. Chem. 1996, 25, 77;
References
(
2
b) Hamai, S.; Satoh, N. Carbohydr. Res. 1997, 304,
29.
1
2
3
. VanEtten, R. L.; Clowes, G. A.; Sebastian, J. F.; Ben-
der, M. L. J. Am. Chem. Soc. 1967, 89, 3253.
. VanEtten, R. L.; Sebastian, J. F.; Clowes, G. A.; Ben-
2
2. (a) Brown, S. E.; Coates, J. H.; Duckworth, P. A.;
Lincoln, S. F.; Easton, C. J.; May, B. L. J. Chem. Soc.,
Faraday Trans. 1993, 89, 1035; (b) Hendrickson, K.;
Easton, C. J.; Lincoln, S. F. Aust. J. Chem. 1995, 48,
1125.
der, M. L. J. Am. Chem. Soc. 1967, 89, 3242.
. Easton, C. J.; Lincoln, S. F. Modified Cyclodextrins.
Scaffolds and Templates for Supramolecular Chemistry;
Imperial College Press: London, 1999.