.Certain soil microbes produce enediyne compounds, such as calicheamicin (above), which are toxic to bacteria and cancer cells. The extremely potent cellular toxicity exhibited by these naturally occurring enediynes is believed to result from the ability of the enediyne portion (in red, above) to undergo a Bergman cyclization (Scheme 1, X = CH) to a reactive 1,4-didehydrobenzene (1,4-ddb) diradical (B, X= CH, Scheme 1), which abstracts hydrogen atoms from the sugar phosphate backbone of DNA, leading to DNA scission and cell death.

Although these naturally occurring enediynes are among the most potent cytotoxic agents ever investigated, the lack of tumor specificity of these compounds has limited their use in fighting cancer. We have undertaken an approach to achieve cancer-cell specific targeting by enediynes involving the redesign of the enediyne core.
We are currently investigating a class of re-designed enediynes, the 3-aza-enediynes (Scheme 1, A, X = N). We have found that these aza-enediynes (for example aza-enediyne 2, Scheme 2) undergo a facile aza-Bergman cyclization to generate 2,5-didehydropyridine (2,5-ddp) diradicals (Scheme 1, B, X = N and Scheme 2, 3). Unlike the 1,4-ddb diradical intermediates of Bergman cyclization of enediynes, the 2,5-ddp intermediates of the aza-Bergman rearrangement of aza-enediynes are not trapped under normal conditions. Instead, these 2,5-ddp intermediates undergo a rapid retro-aza-Bergman rearrangement to the nitriles (Scheme 1, C, X = N; Scheme 2, 2 –> 4, 89% yield).

Interestingly, protonation of the nitrogen atom, such as might occur in the acidic environment of a tumor has been predicted to increase the reactivity of 2,5-ddp diradicals formed from aza-Bergman rearrangement (Scheme 1, X = NH+). We carried out the thermolysis of the aza-enediyne 2 under acidic conditions in the presence of a radical trapping reagent 1,4-cyclohexdiene (Scheme 3). Instead of pyridine product, a cyclopropane derivative (5) was isolated along with retro-aza-Bergman rearrangement product nitrile 4. We proposed that this oxazole compound 5 is derived from a 5-oxazolylcarbene intermediate which results from cyclization of the amide generated from acid-catalyzed addition of water to the aza-enediyne 2 as shown in Scheme 3. In the presence of H2O18 and acid, oxazolyl compound 5 incorporates an O18 label, demonstrating that this oxygen is derived from water. We are currently investigating the generality of this reaction.

For more information on aza-enediynes, see the following papers:

David, W. M.; Kerwin, S. M. Synthesis and thermal rearrangement of C,N-dialkynyl imines: a potential aza-Bergman route to 2,5-didehydropyridine. J. Am. Chem. Soc. 1997, 119, 1464-1465.

David, W.M.; Kumar, D.; Kerwin, S.M. Synthesis of a heterocyclic aza-enediynes and its DNA-cleavage properties. Bioorg. Med. Chem. Lett. 2000, 10, 2509-2512.

Feng, L.; Kumar, D.; Kerwin, S.M. An extremely facile aza-Bergman rearrangement of sterically unencumbered acyclic 3-aza-3-ene-1,5-diynes. J. Org. Chem. 2003, 68, 2234-2242.

Feng, L.; Kerwin, S.M. Isolation of a cyclopropane-containing product from the rearrangement of a 3-aza-3-ene-1,5-diyne under acid catalysis. Tetrahedron Lett. 2003, 44, 3463-3466.

College of Pharmacy at UT Austin