fourth internal hydrogen bond between the aromatic amide
NH of the m-phenyl spacer and the carbonyl group of the
isovaleryl amide.
using benzyl azide 4 as the dipole, the 1,3-dipolar cycload-
dition was performed by using diphenylnitrone 5 as the
dipole, and the conjugate addition was performed by using
thiophenol 6 as the nucleophile. We performed each of the
four reaction types under the same conditions in the presence
and in the absence of 20 mol % of 2. In each case, the starting
concentrations of 1 and the appropriate reaction partner were
50 mM and the progress of the reactions were monitored by
500 MHz 1H NMR spectroscopy for 16 h at 35 °C.
Deconvolution of the appropriate resonances arising from
the starting material and the product, from spectra recorded
at 30 min intervals, allowed concentration-time profiles for
each reaction to be extracted. Each reaction was performed
at least three times. These concentration-time profiles were
used as the basis for simulation and fitting of the data, which
allow kinetic parameters to be extracted. For each reaction,
we determined the rate constant for the reaction in the
absence of receptor 2 (kuncat) and the rate constant for the
reaction in the presence of catalyst 2 (kcat). These rate con-
stants11 were, in turn, used to assess the level of rate accel-
eration observed in each reaction through the kcat/kuncat ratio.
The measured kcat/kuncat ratios12 for the four reactions
were: Diels-Alder reaction, kcat/kuncat ) 1.0; azide cyclo-
This fourth hydrogen bond serves to increase the hydrogen-
bond-donating ability, as judged by the electrostatic potential
surface10 of 2 (Figure 2b), of the NH that binds to the
substrate. The calculations demonstrate (Figure 2c) that 1
and 2 are, indeed, complementary. It should be noted at the
outset that this receptor was designed to probe the hypothesis
set out above and was not designed to achieve large catalytic
effects on the reactions studied.
The presence of all of these hydrogen bonds was con-
firmed directly and indirectly by two methods. First, we per-
1
formed a H-15N HSQC experiment on a sample of 2 ([2]
) 15 mM) in CDCl3 at room temperature and on a mixture
of 1 and 2 ([1] ) [2] ) 15 mM) under the same conditions.
1
There are significant chemical shift changes in both the H
and 15N dimensions for all three NH resonances on moving
from receptor 2 to the complex [1‚2]. Second, we determined
the association constant (Ka) for the [1‚2] complex in CDCl3
1
at 35 °C by using the H NMR titration method. Nonlinear
curve fitting of the chemical shift change data for four
protons to a 1:1 binding model afforded a value for the Ka
of 750 ( 30 M-1. This value corresponds to a free energy
of binding of -17.0 kJ mol-1 at 35 °C. The observed Ka is
significantly higher than that measured for an isolated
amidopyridine-carboxylic acid complex under the same
conditions (95 M-1, -11.7 kJ mol-1). These results are
entirely consistent with the calculated structure for [1‚2].
With our receptor and substrate in place, we selected a
series of reactions whose rate-determining steps featured
transition states of varying charge separation (Figure 3).
addition, kcat/kuncat ) 1.5 nitrone cycloaddition, kcat/kuncat
)
1.8; conjugate addition, kcat/kuncat ) 37. Qualitatively, it is
obvious from these data that there is a significant difference
in the performance of catalyst 2 between the cycloadditions
which all have rather nonpolar transition states and the
conjugate addition, which develops significant negative
charge in the rate determining step. These data alone,
however, are not enough to establish the nature of the
relationship between transition-state charge and the rate
acceleration that can be developed by stabilizing that charge.
In an attempt to establish a more quantitative relationship,
we need to derive a metric that describes the change in charge
distribution as we move toward the transition state in the
rate-determining step of each of the four reactions. In other
words, having established a quantitative y-axis for the graph
shown in Figure 1 by determining the kcat/kuncat ratio, we now
needed to establish a quantitative x-axis.
In order to accomplish this task, we turned to electronic
structure calculations. Transition states were successfully
located at the B3LYP/6-31G(d) level of theory for all four
(9) Molecular mechanics calculations were performed by using the
AMBER* force field and GB/SA solvation model for CHCl3 as implemented
in Macromodel (Version 7.1, Schrodinger Inc., 2000).
(10) The lowest energy conformation of 2, located by the molecular
mechanics conformational search, was used as the basis for an electronic
structure calculation at the HF/6-31G(d) level of theory and the electrostatic
potential surface of 2 was computed and visualized by using data from this
calculation.
Figure 3. Reaction types and corresponding reagents selected for
comparison using catalyst 2.
(11) In the Diels-Alder reaction between 1 and 3, both possible
diastereoisomers, endo and exo, are formed in an approximately 1:1 ratio.
In analyzing the kinetic data for this reaction, we summed the concentrations
of these two products and fitted the formation of total cycloadduct to the
kinetic model. Similarly, we considered only total cycloadduct concentration
in the reaction between 1 and 5 which also affords two diastereoisomeric
products. The presence of catalyst 2 had an insignificant effect on the
diastereoselectivity of the dipolar cycloaddition between 1 and 5.
(12) Hamilton and co-workers (Fan, E. K.; Vicent, C.; Hamilton, A. D.;
New J. Chem. 1997, 21, 81-85) have reported a receptor, similar in structure
to 2, which is also capable of accelerating the reaction between a maleimide
and a thiolate anion.
The Diels-Alder reaction was performed by using furan
3 as the diene, the azide cycloaddition was performed by
(8) (a) Pearson, R. J.; Kassianidis, E.; Philp, D. Tetrahedron Lett. 2004,
43, 4777-4780. (b) Howell, S. J.; Spencer, N.; Philp, D. Org. Lett. 2002,
4, 273-276. (b) Rowe, H. L.; Spencer, N.; Philp, D. Tetrahedron Lett.
2000, 41, 4475-4479. (c) Bennes, R.; Philp, D.; Spencer, N.; Kariuki, B.
M.; Harris, K. D. M. Org. Lett. 1999, 1, 1087-1090. (d) Robertson, A.;
Philp, D.; Spencer, N. Tetrahedron 1999, 55, 11365-11384. (e) Booth, C.
A.; Philp, D. Tetrahedron Lett. 1998, 39, 6987-6990.
Org. Lett., Vol. 8, No. 23, 2006
5181