Selectivity in an encapsulated cycloaddition reaction

Article abstract of DOI:10.1021/ol0168115

(Equation Presented) A 1,3-dipolar cycloaddition takes place within a reversibly formed, self-assembled capsule. The reaction proceeds through an unsymmetrically loaded encapsulation complex with absolute regioselectivity.

Full text of DOI:10.1021/ol0168115

ORGANIC
LETTERS
2
002
Vol. 4, No. 3
27-329
Selectivity in an Encapsulated
Cycloaddition Reaction
3
Jian Chen and Julius Rebek, Jr.*
The Skaggs Institute for Chemical Biology and Department of Chemistry, The Scripps
Research Institute, 10550 North Torrey Pines Road, La Jolla, California 92037
jrebek@scripps.edu
Received September 24, 2001
ABSTRACT
A 1,3-dipolar cycloaddition takes place within a reversibly formed, self-assembled capsule. The reaction proceeds through an unsymmetrically
loaded encapsulation complex with absolute regioselectivity.
Synthetic receptors that apply molecular recognition to
catalysis aim at elusive, moving targetsstransition states.
Cycloaddition reactions are popular in this context, perhaps
because there are so few enzymatic examples or because their
transition states resemble stationary targetssthe products.
The latter feature makes product inhibition inevitable, but
1
2
3
cyclodextrins, cucurbituril, multimeric porphyrins, and
4
reversibly formed encapsulation complexes have all shown
promise in accelerating these reactions. We report here a
synthetic chamber that accelerates a 1,3-dipolar cycloaddi-
tion. Although the system does not solve the turnover
problem, it allows the direct observation of the “Michaelis
complex”.
Figure 1. Line drawing of the resorcinarene subunit (left) and ball-
and-stick model of the dimeric capsule (center). Two toluene
molecules are encapsulated in the energy-minimized complex
Earlier we described a capsule with a roughly cylindrical
5
cavity in which two different aromatic guest molecules can
be accommodated simultaneously. Their orientation is con-
strained to edge-to-edge approaches, and only their peripheral
substituents make contact. Figure 1 uses the interaction of
the methyl groups of two encapsulated toluene molecules to
(right).
illustrate this point. The arrangement hints at reactions
between such substituents, anchored in the capsule by their
respective aromatic nuclei.
(
(
1) Breslow, R.; Guo, T. J. Am. Chem. Soc. 1988, 110, 5613-5617.
2) Mock, W. L.; Irra, T. A.; Wepsiec, J. P.; Manimaran, T. L. J. Org.
Chem. 1983, 48, 3619-3620. Jon, S. Y.; Ko, Y. H.; Park, S. H.; Kim,
The cycloaddition involves phenylacetylene 3 and phenyl
azide 4 (Figure 2). These compounds react very slowly to
give roughly equal amounts of two regioisomeric triazoles
H.-J.; Kim, K. Chem. Commun. 2001, 1938-1939.
(3) Marty, M.; Clyde-Watson, Z.; Twyman, L. J.; Nakash, M.; Sanders,
J. K. M. Chem. Commun. 1998, 2265-2266.
(
4) Kang, J.; Rebek, J., Jr. Nature 1997, 385, 50-52.
6
(5) Heinz, T.; Rudkevich, D. M.; Rebek, J., Jr. Nature 1998, 394, 764-
5 and 6. At ambient temperature in mesitylene solvent the
7
1
66. Tucci, F. C.; Rudkevich, D. M.; Rebek, J., Jr. J. Am. Chem. Soc. 1999,
-9
-1 -1
rate constant is 4.3 × 10
M
s
and corresponds to a
21, 4928-4929. K o¨ rner, S. K.; Tucci, F. C.; Rudkevich, D. M.; Heinz,
T.; Rebek, J., Jr. Chem.sEur. J. 1999, 6, 187-195.
half-life (at 1 M each component) of several years. At the
1
0.1021/ol0168115 CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/08/2002

Figure 2. The reaction of the acetylene and azide gives comparable
amounts of the two regioisomers.
Figure 4. Selective formation of the 1,4-isomer. (A) Addition of
DMF-d to a solution of authentic encapsulated 1,4-isomer in
mesitylene-d12. (B) Addition of DMF-d to the mesitylene-d12
solution obtained from phenyl azide (25 mM) and phenylacetylene
(50 mM) in the presence of the capsule (5 mM). (C)Addition of
7
millimolar concentrations used in the experiments below, the
rate is negligible on the human time scale.
A solution of 3 (50 mM), 4 (25 mM), and 2 (5 mM) in
deuterated mesitylene shows the encapsulated product within
a few days (Figure 3A-D). The product triazole gives a
7
7
DMF-d to solution of authentic 1,5-isomer in mesitylene-d12. (b)
Released 1,4-isomer; (3) 1,5 isomer.
Further evidence for an encapsulated transition state (rather
than, say, regioselective catalysis by functional groups of
the capsule) involves size and shape selectivity. Neither
1
-naphthyl azide 7 nor 4-biphenyl azide 8 show rate
accelerations in their cycloadditions with 3 when in the
presence of the capsule. The naphthyl azide is not encap-
sulated, while the biphenyl azide is quickly displaced from
the capsule by two phenyl acetylenes.
The various encapsulated species were identified as
follows. The capsule in the presence of an excess of 4 alone
gives a single symmetrical species. The aryl CH resonances
are as shown in Figure 5 and were assigned from their
Figure 3. Formation of the 1,4-isomer inside the capsule from
phenyl azide (25 mM) and phenylacetylene (50 mM) in the presence
of the capsule (5 mM) in mesitylene-d12. (A) t ) 0; (B) t ) 1540
min; (C) t ) 4320 min; (D) t ) 8500 min.
unique signal at 5.76 ppm as it is formed (Figure 3). The
initial rate of product formation, determined over 6 days, is
-
9
-1
1
.3 × 10 M s .
Only the 1,4-isomer is formed. Control experiments
Figure 5. The disproportionation equilibria of encapsulated
reactants. Selected chemical shifts are given in ppm.
established that only the 1,4-isomer is encapsulated in
mesitylene and this compound is liberated by the addition
of DMF (Figure 4A). The product obtained from the capsule
on treatment with DMF is shown in Figure 4B; the
appropriate regions of the spectrum of the 1,5-isomer are
shown for comparison in Figure 4C.
relative shifts and multiplicities. The only orientation con-
sistent with these shifts is that shown in 9. The arrangement
is consistent with the preference of polar groups to be near
the seam of hydrogen bonds.
The assignments for encapsulated acetylene were com-
plicated by a pesky impurity in the commercial material that
could not be removed by distillation. Excess acetylene gave
(6) Kirmse, W.; Horner, L. Liebigs Ann. 1958, 614, 1. For discussions,
see the following: Fleming, I. Frontier Orbitals and Organic Chemical
Reactions; Wiley: New York, 1977; p 155. Kolb, H. C.; Finn, M. G.;
Sharpless, K. B. Angew. Chem., Int. Ed. 2001, 40, 2004-2021.
328
Org. Lett., Vol. 4, No. 3, 2002

not only the symmetrically loaded capsule 10 but also 20-
0% (depending on the other components present) of an
K
D
to be 4. Six determinations with various mole fractions
4
of the azide and acetylene showed K ) 9 ( 3. The exchange
D
unsymmetrically filled capsule. This species was identified
by the addition of excess dioxane as a capsule containing
one acetylene and one dioxane molecule.
of these small guests is fast. They equilibrate on mixing;
the system is at a steady state. Perhaps the space in 11 is
better occupied or there is a weak attractive force between
the occupants. In any case, the preference for 11 is only a
few tenths of a kcal/mol.
At these concentrations, chosen for convenience of moni-
toring by NMR, the capsule is saturated. Increasing the
external concentrations of the reactants does not increase the
concentration of the heterocapsule, but it increases the rate
of the cycloaddition reaction outside. The rate inside is fixed
A new, unsymmetrical complex appeared when a mixture
of azide and acetylene were present, and it was the most
abundant species. Fortunately, its NH resonances were well
separated from those of the other capsule species (Figure 6)
and allowed clean integration and quantification of this
“
Michaelis complex.”7
at saturation, and comparing the kcat to the outside rate k
2
suffers all of the ambiguities that arise when a bimolecular
rate is compared to a unimolecular one.
The calculated concentrations inside are a matter of
3
volume: the capacity is ∼450 Å , and the two reactants enjoy
a 3.7 M engagement for a matter of seconds when they are
encapsulated. The rate inside, assuming the appropriate
-
8
orientation can be achieved, would then be ∼6 × 10
M
-
1
s , a figure larger than the initial rate actually observed.
One might well ask, why is the reaction not faster inside?
Perhaps they are positioned in a way that is not ideal for the
transition state; we have no reason to believe that the
transition state is bound better than the reactants. At these
concentrations the reaction outside the capsule is calculated
-
12
-1
to proceed at a rate of 5.4 × 10 M s , some 240 times
Figure 6. Encapsulated species in a solution of capsule 2, phenyl
azide, and phenylacetylene in mesitylene-d12. (a) NH resonances
of the capsule with acetylene and dioxane; (b) NH resonances of
capsule with azide and dioxane; (c) NH resonances of “Michaelis”
complex 11; (d) NH resonance of capsule complex 9; (e) NH
resonance of capsule complex 10; (f) NH resonances of product
complex 12; (g) ortho-protons of the phenyl rings of encapsulated
slower than inside the capsule.
The direct observation of the Michaelis complex is, to our
knowledge, unique to the case at hand. While it simplifies
the analysis, one thing is clear: the product is the best guest
in the system, gradually the capsule is filled with it, and the
reaction, slowed by product inhibition, grinds to a halt.
Overcoming the general problem of product inhibition
1,4-isomer 5 in complex 12 (see assignments in Figure 5). (A) At
t ) 0 incubating 3 (50 mM), 4 (25 mM), and capsule 2 (5 mM) in
8
remains a challenge for the future. In the meantime, the
exquisite selectivity for the extended regioisomer 5 augurs
mesitylene-d12. (B) t ) 8500 min for the same system as in (A).
(C) Encapsulation of authentic 1,4-isomer 5 with 2 in mesitylene-
well for controlling reactions within reversibly formed
d .
12
9
capsules.
Acknowledgment. We are grateful to the Skaggs Re-
search Foundation and the National Institutes of Health for
support. J.C. is a Skaggs Postdoctoral Fellow. Professors
Dmitry Rudkevich, JoAnne Stubbe, Jamie Williamson, and
Dr. Steffi K o¨ rner gave us invaluable advice.
We studied the disproportionation equilibrium of Figure
. A purely statistical distribution of the two guests predicts
5
twice as much heterocapsule 11 as either homocapsule, since
there are two ways to fill the former and only one way to
fill either of the latter (given the orientation preferences
discussed above). The effect of dioxane, the spectator guest,
was comparable on both homocapsules; i.e., no great
OL0168115
(8) For a discussion, see: Nakash, M.; Clyde-Watson, Z.; Feeder, N.;
preference for homocapsules exists. For the equilibrium K
D
Davies, J. E.; Teat, S. J.; Sanders, J. K. M. J. Am. Chem. Soc. 2000, 122,
5286-5293 and references therein.
2
)
[11] /[9][10], the statistical distribution would predict
(
9) For examples with metal-ligand capsules, see: Zeigler, M.; Brum-
aghim, J. L.; Raymond, K. N. Angew. Chem., Int. Ed. 2000, 39, 4119-
4121. Caulder, D. L.; Raymond, K. N. J. Chem. Soc., Dalton Trans. 1999,
1185-1200. Umemoto, K.; Yamaguchi, K.; Fujita, J. J. Am. Chem. Soc.
2000, 122, 7150-7151. Fujita, M.; Umemoto, K.; Yoshizawa, M.; Fujita,
N.; Kusukawa, T.; Biradha, K. Chem. Commun. 2001, 509-518.
(
7) Because both reactants form symmetrical complexes (e.g., 9 and 10)
the system shows substrate inhibition and does not obey Michaelis-Menten
kinetics; for a discussion, see: Fersht, A. Enzyme Structure and Mechanism,
nd ed.; W. H. Freeman: New York, 1985; p 114.
2
Org. Lett., Vol. 4, No. 3, 2002
329