M. L. Dougan et al. / Bioorg. Med. Chem. Lett. 14 (2004) 4153–4156
4155
Theoretical fits for the data in Figure 7 are not given,
since, as noted in the earlier reports,1;6 the rate depen-
dence on pD for cleavage of Kemp’s diacid amide
derivatives is not readily modeled.
In summary, we have demonstrated that the Kemp’s
triacid piperidyl amide derivative 7 and prolyl amide
derivative 8 undergo intramolecular acylolysis at rates
comparable to that of the previously reported pyrrolidyl
and methylphenethyl amide derivatives 1a and 5.
Molecular mechanics calculations suggest that relief of
pseudoallylic strain in these tertiary amide derivatives
does contribute to the rapid rate of reaction.
In our hands, pyrrolidyl amide 1a shows the same
reactivity as previously reported.1 Across the pD range
studied, piperidyl amide 7 cleaves approximately four-
times more quickly than 1a, whereas prolyl amide 8
cleaves approximately two-times more slowly (to yield
the acid anhydride 2a and proline––thus the prolyl
carboxylic acid functionality does not directly partici-
pate in the reaction). At pD 7.7, for example, piperidyl
amide 7 has a half-life of 3 min; at pD 7.8, prolyl amide 8
has half-life of 30 min, and the half-life for pyrrolidyl
amide 1a is 13 min. These results indicate that the
Kemp’s triacid scaffolding provides a general means of
activating tertiary amide derivatives.
Acknowledgements
We thank Richard Schowen for extremely helpful advice
on the use of pH paper for determining pD. We are
grateful to the National Institutes of Health (R15
GM63776) for financial support, as well as to the Am-
herst College Faculty Research Award Program, as
funded by the H. Axel Schupf ’57 Fund for Intellectual
Life. We also are grateful to the Pfizer Summer
Undergraduate Research Fellowship Program in Syn-
thetic Organic Chemistry for support of J.L.C. and
M.L.D. and to the Howard Hughes Medical Institute
for support of K.S. (through Amherst College’s
Undergraduate Biological Sciences Education Program
award).
We used molecular mechanics to explore whether, fol-
lowing the suggestion of Curran et al.,6 greater
pseudoallylic strain in piperidyl derivative 7 might be the
reason for its enhanced reactivity relative to the previ-
ously reported tertiary amide derivatives 1a and 5. The
minimized structures13 of the three species are shown in
Figure 8. The nonbonded interactions in the pyrrolidyl
and methylphenethyl amide derivatives 1a and 5 are
remarkably similar, each having three almost identical
hydrogen–hydrogen close contacts. However, due to the
pseudo-chair conformation of the piperidyl ring, deriv-
ative 7 uniquely has two nonbonded interactions with a
References and notes
ꢀ
hydrogen–hydrogen distance of only 2 A.
1. Menger, F. M.; Ladika, M. J. Am. Chem. Soc. 1988, 110,
6794–6796.
2. Kemp, D. S.; Petrakis, K. S. J. Org. Chem. 1981, 46, 5140–
5143.
3. Radzicka, A.; Wolfenden, R. J. Am. Chem. Soc. 1996, 118,
6105–6109.
4. Smith, R. M.; Hansen, D. E. J. Am. Chem. Soc. 1998, 120,
8910–8913.
While it is not possible to quantify accurately the relief
of strain upon formation of the rate-determining tran-
sition state(s), the above modeling results do support the
notion that ground-state pseudoallylic strain contributes
to the enhanced reactivity of tertiary amide derivatives.
Since the nonbonded interactions in the minimized
structure of prolyl derivative 8 (not shown) are identical
to those in 1a, its slower rate of cleavage likely arises
from a perturbation of the carboxylic acid group di-
rectly participating in the reaction by the nearby prolyl
carboxylic acid functionality.
5. Menger, F. M.; Ladika, M. J. Org. Chem. 1990, 55, 3006–
3007.
6. Curran, T. P.; Borysenko, C. W.; Abelleira, S. M.;
Messier, R. J. J. Org. Chem. 1994, 59, 3522–3529.
7. Johnson, F. Chem. Rev. 1968, 68, 375–413.
8. To a stirred suspension of anhydride, acid chloride 92
(889 mg, 3.44 mmol) in 120 mL of dry acetonitrile under
argon was added dry pyridine (8.7 mL). The mixture was
cooled to ꢁ40 °C (CO2/acetonitrile), and piperidine
(305 lL) was then added. A yellow color quickly devel-
oped. The mixture was allowed to warm to room
temperature and was stirred for 12 h. The solvent was
removed by evaporation under reduced pressure to yield a
bright yellow solid, which was purified by column chro-
matography (1:1 hexanes–ethyl acetate, Rf ¼ 0:36) to give
anhydride amide 10 as a white solid (116 mg, 11%). 1H
NMR (CDCl3, 400 MHz) d 3.49 (m, 4H), 2.80 (d,
J ¼ 12:8 Hz, 2H), 2.01 (dt, J ¼ 13:5, 1.8 Hz, 1H), 1.55–
1.65 (m, 6H), 1.35 (d, J ¼ 13:5 Hz, 1H), 1.35 (s, 6H), 1.31
(s, 3H), 1.23 (d, J ¼ 13:9 Hz, 2H). 13C NMR (CDCl3,
100 MHz) d 171.55, 171.51, 46.8, 43.1, 42.4, 40.5, 40.1,
28.64, 28.62, 25.76, 25.74, 25.1, 24.5. IR (CHCl3, cmꢁ1
)
Figure 8. Pseudoallylic strain in energy-minimized conformations of
derivatives 1a, 5, and 7. The distances of all hydrogen–hydrogen close
contacts are shown. The structures are truncated to emphasize the
nonbonded interactions.
1801, 1764, 1621, 1015. HRMS (TOF ESþ) calculated for
C17H26NO4 (MþH)þ 308.1862. Found 308.1859.
9. To a stirred suspension of anhydride, acid chloride 9
(404 mg, 1.56 mmol) in 50 mL dry acetonitrile under argon