A R T I C L E S
Baroudi et al.
Table 4. Summary of the Geometrical Differences in TS3 of the
Four Enediynes and the Activation Barriers for the Fragmentation
Step. Data in Parentheses Correspond to the Respective
Monoradicals
dimethyl substitution in 5 and 6. Although the calculated
dihedral angles in 5 and 6 still allow substantial C-O and
π-system overlap, benzylic stabilization in these systems does
not assist to the C-O bond cleavage to its full extent. While
hyperconjugative assistance from the methyl groups in 5 and 6
compensates for the partial loss of benzylic stabilization, the
overall activation barriers seem to result from the tug-of-war
of the electronic and steric effects. For example, the lower γ
value observed for enediyne 6 could explain the slight increase
in the activation barrier (∼1 kcal/mol) obtained for the
fragmentation step of this molecule.
Summary of Computational Analysis. The first, highly
endothermic, step of the cascade transformation in Scheme 13
(the Bergman cyclization) is thermodynamically unfavorable
and, unless a suitable trapping step is available, rapidly
reversible. However, each subsequent transition state is lower
in absolute energy than the previous one, and overall sequence
of reactions culminates in the final products which are 20-25
kcal/mol lower than the starting materials, thus providing a
strong driving force for the overall transformation. Overall, this
potential energy surface can be classified as an example of a
“slippery slope” reaction energy profile.41
Conformational Control of Bergman Cyclization and Resistance
to Natural Antibiotics by the Producing Organisms. Together with
the available literature data, our results fit well into the following
logical progression of facts and observations: (a) Esperamicins
are produced by microorganisms and exported outside of the
cell in a full, unfragmented form; (b) Esperamicins fragment
into the aromatized and carbohydrate parts upon their activation
and subsequent Bergman cyclization; (c) compounds 3-6,
designed to mimic esperamicins, also undergo fragmentation
after their Bergman cyclization; (d) chemical similarity between
our molecules and esperamicins suggests that these fragmenta-
tions proceed via a similar mechanism; and (e) this “confor-
mationally gated” mechanism depends on the orientation of the
carbohydrate moiety because its proximity to the p-benzyne
radical is required to enable the intramolecular H-abstraction
step. When all of the above points are considered together from
a more general perspective, they can be combined in an
intriguing hypothesis outlined in this section.
entry
(enediyne)
dihedral
angle γ
Ea,
kcal/mol
∆E,
kcal/mol
dTS3
Å
1 (3)
2 (4)
3 (5)
4 (6)
94.14 (94.75) 1.750 (1.760) 9.7 (10.0) -16.9 (-16.3)
93.82 (94.89) 1.718 (1.720) 7.6 (7.7)
61.15 (69.77) 1.750 (1.750) 5.4 (5.8)
50.34 (55.41) 1.710 (1.710) 6.9 (7.0)
-22.5 (-21.9)
-18.9 (-21.2)
-24.7 (-24.0)
5. In addition, the activation barriers for the fragmentation of 5
and 6 are lower with respect to those of 3 and 4. All activation
barriers for the fragmentation step, C-O incipient bond
distances and the dihedral angles between the breaking C-O
bond and the plane of the naphthalene moiety are summarized
in Table 4.
The higher stability of γ-butyrolactone relative to δ-valero-
lactone could be the reason behind the higher exothermicity of
the BC/Fragmentation of enediynes 4 and 6. On the basis of
the correlation between reaction energy and activation barriers,40
this would contribute to the lowering of the activation barriers
of the fragmentation step.
Since both enediynes 5 and 6 have the gem-dimethyl
substituents at the benzylic position in TS2, one would expect
a higher stabilization for the incipient radical in the transition
state. Such stabilization could be the reason behind lower
activation barriers and higher exothermicities for the fragmenta-
tion step of the two gem-dimethyl substituted enediynes.
However, the activation barrier for enediyne 5 is ∼1 kcal/mol
lower than that of 6, which is not consistent with the result
obtained for enediynes 3 and 4. In order to understand the
reasons for this discrepancy, we examined the alignment of the
breaking C-O bond with the naphthalene π-system (the OCCC
dihedral angle, γ in Table 4). A value of γ ) 90° would
correspond to a maximum interaction (orbital overlap) between
the C-O bond and the π-system which would provide maxi-
mum stabilization to the incipient radical center. Table 4
illustrates that this geometry is achieved in the TS3 of enediynes
3 and 4. On the other hand, the γ values are lower in the case
of enediynes 5 and 6 (50-70°). This deviation in the γ values
is probably a result of the steric interaction exerted by the gem-
The microorganisms which produce natural antibiotics must
protect themselves from the lethal effects of their own toxins
as these molecules are produced, transported, and exported.7b
Medicinal chemists who develop thermally activated prodrugs
face the same problem when designing drugs which are nontoxic
and shelf-stable until they reach their clinical target.
Results outlined in this paper illustrated how conformational
effects can be used in conjunction with intramolecular reactions
for control of enediyne reactivity through shifting the enediyne/
p-benzyne equilibrium toward the radical species. Because
Bergman cyclizations are often endothermic, equilibrium con-
(37) (a) Hoffman, R.; Imamura, A.; Hehre, W. J. J. Am. Chem. Soc. 1968,
90, 1499. Hoffman, R. Acc. Chem. Res. 1971, 4, 1. (b) Paddon-Row,
M. N. Acc. Chem. Res. 1982, 15, 245. (c) Gleiter, R.; Schafer, W.
Acc. Chem. Res. 1990, 23, 369. (d) Brodskaya, E. I.; Ratovskii, G. V.;
Voronkov, M. G. Russ. Chem. ReV. 1993, 62, 975.
(40) (a) Marcus, R. A. J. Chem. Phys. 1956, 24, 966. (b) Marcus, R. A.
Annu. ReV. Phys. Chem. 1964, 15, 155. (c) Marcus, R. A. J. Phys.
Chem. 1968, 72, 891. Application of Marcus theory to pericyclic
reactions: (d) Alabugin, I. V.; Manoharan, M.; Breiner, B.; Lewis, F.
J. Am. Chem. Soc. 2003, 125, 9329. Radical cyclizations: (e) Alabugin,
I. V.; Manoharan, M. J. Am. Chem. Soc. 2005, 127, 12583. (f)
Alabugin, I. V.; Manoharan, M. J. Am. Chem. Soc. 2005, 127, 9534.
(g) Wu, C. W.; Ho, J. J. J. Org. Chem. 2006, 71, 9595. (h) Yu, Y.-Y.;
Fu, Y.; Xie, M.; Liu, L.; Guo, Q.-X. J. Org. Chem. 2007, 72, 8025.
(I) Alabugin, I. V.; Timokhin, V. I.; Abrams, J. N.; Manoharan, M.;
Ghiviriga, I.; Abrams, R. J. Am. Chem. Soc. 2008, 30, 10984.
(41) (a) Rempala, P.; Kroulik, J.; King, T. B. J. Am. Chem. Soc. 2004,
126, 15002. (b) Engler, E. M.; Farcasiu, M.; Sevin, A.; Cense, J. M.;
Schleyer, P. V. R. J. Am. Chem. Soc. 1973, 95, 5769.
(38) (a) Logan, C. F.; Chen, P. J. Am. Chem. Soc. 1996, 118, 2113. (b)
Schottelius, M. J.; Chen, P. J. Am. Chem. Soc. 1996, 118, 4896.
(39) (a) Squires, R. R.; Cramer, C. J. J. Phys. Chem. A 1998, 102, 9072.
(b) Kraka, E.; Cremer, D. J. Am. Chem. Soc. 2000, 122, 8245. (c)
Pickard, F. C., IV; Shepherd, R. L.; Gillis, A. E.; Dunn, M. E.; Feldgus,
S.; Kirschner, K. N.; Shields, G C.; Manoharan, M.; Alabugin, I. V.
J. Phys. Chem. A 2006, 110, 2517. For a general analysis of orbital
interactions involving R-bonds, see: (d) Alabugin, I. V.; Zeidan, T. A.
J. Am. Chem. Soc. 2002, 124, 3175. (e) Alabugin, I. V. J. Org. Chem.
2000, 65, 3910.
9
978 J. AM. CHEM. SOC. VOL. 132, NO. 3, 2010