which is present in the [A‚B] complex, may also persist in
the final product and might therefore have a profound
influence on the stability of the products of the reaction. This
stabilizing effect on the product ground state will manifest
itself when the reaction in question is under thermodynamic
control and the selective stabilization of one product with
respect to another will serve to facilitate7 the formation of
the more stable product. Therefore, unlike enzymes or
enzyme mimics, our methodology is ideally suited for
reactions which have unfavorable equilibrium constants under
normal conditions, i.e. reactions which suffer from a
thermodynamic disadvantage as opposed to a kinetic disad-
vantage. Here, we report the recognition-mediated facilitation
of the reaction between a 2-phenylfuran derivative and a
maleimide and the characterization and analysis of the origins
of this facilitation by kinetic simulation, 1H NMR spectros-
copy, and X-ray crystallography.
2-Phenylfuran derivatives are notoriously poor dienes in
Diels-Alder cycloaddition reactions as a direct result of their
high level of π conjugation, which is destroyed by the
formation of a Diels-Alder cycloadduct. Thus, when a 100
mM solution of diene 1 and maleimide 2 in CDCl3 was
heated at 50 °C (Scheme 1), only very small amounts of the
Figure 1. Rate profiles for (a) the reaction between 1 and 2 in
CDCl3 at 50 °C and (b) the reaction between 4 and 5 in CDCl3 at
50 °C. In both cases, the starting concentrations of the reactants
were 100 mM. In both plots, the filled squares represent the
concentration of the appropriate exo cycloadduct. The solid lines
represent the best fit of the appropriate kinetic model to the
experimental data. For clarity, error bars are omitted from the
graphs; however, errors in concentration are estimated to be (4%.
Scheme 1
value of the equilibrium constant (Keq) for this reaction of
0.18 M-1. To investigate the recognition-mediated facilitation
of this cycloaddition reaction, diene 4, bearing an amidopy-
ridine recognition site, and maleimide 5, bearing a comple-
mentary carboxylic acid recognition site, were prepared10
from readily available starting materials. When a 100 mM
solution of diene 4 and maleimide 5 in CDCl3 was heated at
50 °C (Scheme 1), a significant quantity of the exo
1
cycloadduct 6 could be detected by 500 MHz H NMR
spectroscopy after 15 h. Kinetic simulation and fitting of
the reaction profile (Figure 1b) allowed the extraction of a
value of the equilibrium constant (Keq) for this reaction of
15.3 M-1.
From these results, it is clear that the introduction of the
recognition elements within the diene and dienophile has
increased the extent of reaction dramatically. To assess the
1
exo cycloadduct could be detected8 by 500 MHz H NMR
spectroscopy after 15 h. Kinetic simulation and fitting9 of
the reaction profile (Figure 1a) allowed the extraction of a
(8) In all cases, only the exo cycloadduct was detected in solution. This
is consistent with the fact that the exo adduct is normally the thermodynamic
product of furan cyclodaddition reactions.
(9) Kinetic simulation and fitting was carried out using Gepasi (version
3.21): Mendes, P. Comput. Appl. Biosci. 1993, 9, 563. Mendes, P. Trends
Biochem. Sci. 1997, 22, 361. Mendes P.; Kell, D. B. Bioinformatics 1998,
14, 869).
(6) Page, M. I.; Jencks, W. P. Proc. Natl. Acad. Sci. USA 1971, 68, 1678.
Page, M. I. Chem. Soc. ReV. 1973, 2, 295. Kirby, A. J. AdV. Phys. Org.
Chem. 1980, 17, 183. Page, M. I. In The Chemistry of Enzyme Action; Page,
M. I., Ed.; Elsevier: Amsterdam, 1984; p 1. Menger, F. M. Acc. Chem.
Res. 1985, 18, 128. Page, M. I. Philos. Trans. R. Soc. London B 1991, 332,
149. Kirby, A. J. Philos. Trans. R. Soc. London A 1993, 345, 67. Menger,
F. M. Acc. Chem. Res. 1993, 26, 206. Bruice, T. C.; Lightstone, F. C. Acc.
Chem. Res. 1999, 32, 127.
(7) To distinguish between transition state effects and ground-state effects,
we will use the description accelerated when the rate of a reaction is
enhanced by the lowering of the energy of the transition state by a
recognition event and the description facilitated when the extent of reaction
is enhanced by the lowering of the energy of the product by a recognition
event.
(10) Selected spectroscopic data for 4: 1H NMR (300 MHz, CDCl3) δH
3
8.71 (1H, s), 8.21 (1H, d, JHH ) 7.3 Hz), 8.19 (1H, s), 7.85-7.78 (2H,
m), 7.65 (1H, t, 3JHH ) 7.3 Hz), 7.49 (1H, d, 4JHH ) 0.7 Hz), 7.48 (1H, t,
3JHH ) 7.7 Hz), 6.92 (1H, d, 3JHH ) 7.3 Hz), 6.74 (1H, d, 3JHH ) 3.3 Hz,
4JHH ) 0.7 Hz), 6.49 (1H, dd, 3JHH ) 1.5, 3.3 Hz), 2.45 (3H, s); 13C NMR
(75 MHz, CDCl3) δC 165.7, 156.8, 152.7, 151.0, 142.6, 138.7, 135.0, 131.4,
129.1, 127.0, 126.0, 122.6, 119.5, 111.8, 111.3, 106.1, 23.8; m/z (EIMS)
[M]+ 278 (55), 249 (88), 171 (100). Selected spectroscopic data for 5: 1H
NMR (300 MHz, CDCl3) δH 6.7 (2H, s), 3.82 (2H, t, 3JHH ) 6.9 Hz), 2.69
(2H, t, 3JHH ) 6.9 Hz); 13C NMR (75 MHz, CDCl3) δC 176.6, 170.4, 134.3,
33.3, 32.6; m/z (EIMS) [M]+ 169 (9), 123 (51), 44 (100).
1088
Org. Lett., Vol. 1, No. 7, 1999