Enediynes from aza-enediynes: C,N-dialkynyl imines undergo both aza-Bergman rearrangement and conversion to enediynes and fumaronitriles

Article abstract of DOI:10.1021/ol0600638

Aza-enediynes (C,N-dialkynyl imines) undergo thermal aza-Bergman rearrangement to β-alkynyl acrylonitriles through 2,5-didehydropyridine (2,5-ddp) intermediates. Certain aza-enediynes also undergo an alternative process affording enediynes and fumaronitri

Full text of DOI:10.1021/ol0600638

ORGANIC
LETTERS
2006
Vol. 8, No. 10
1983-1986
Enediynes from Aza-Enediynes:
C,N-Dialkynyl Imines Undergo Both
Aza-Bergman Rearrangement and
Conversion to Enediynes and
Fumaronitriles
Liping Feng, Aibin Zhang, and Sean M. Kerwin*
DiVision of Medicinal Chemistry, College of Pharmacy and Institute for Cellular and
Molecular Biology, The UniVersity of Texas at Austin, Austin, Texas 78712
skerwin@mail.utexas.edu
Received January 10, 2006
ABSTRACT
Aza-enediynes (C,N-dialkynyl imines) undergo thermal aza-Bergman rearrangement to
â-alkynyl acrylonitriles through 2,5-didehydropyridine
(2,5-ddp) intermediates. Certain aza-enediynes also undergo an alternative process affording enediynes and fumaronitriles. Studies employing
a specifically l3C-labeled aza-enediyne show that the conversion to enediyne is second order in aza-enediyne, proceeds by a “head-to-tail”
coupling, and affords the (Z)-enediyne.
Anticancer antibiotics containing the enediyne moiety (1,
Scheme 1) undergo Bergman cyclization to reactive 1,4-
didehydrobenzene (1,4-ddb) diradicals (2, Scheme 1).1 The
diradicals derived from these enediynes abstract hydrogen
atoms from the sugar phosphate backbone of DNA (2 f 4,
Scheme 1), leading to oxidative DNA strand scission and
ultimately cell death.1b,c The isolation of the enediyne
antitumor antibiotics has led to a renewed interest in the
Bergman cyclization in the design of cancer cell-selective
DNA cleavage agents2 and as an approach to the construction
of polycyclic molecules through free-radical cascade cy-
clizations.3 More recently, this has led to a search for
alternative diradical-generating cyclizations that might be
incorporated in the design of improved DNA-cleavage
agents4 or employed in the construction of heterocycles.5
Aza-enediynes, or C,N-dialkynyl imines (5, Scheme 1),
undergo a Bergman-type rearrangement to the corresponding
nitriles (7, Scheme 1), presumably through the 2,5-didehy-
dropyridine (2,5-ddp) intermediate (6, Scheme 1), which is
not trapped by hydrogen atom abstraction.6
(3) (a) Lewis, K. D.; Rowe, M. P.; Matzger, A. J. Tetrahedron 2004,
60, 7191-7196. (b) Mifsud, N.; Mellon, V.; Perera, K. P. U.; Smith, D.
W., Jr.; Echegoyen, L. J. Org. Chem. 2004, 69, 6124-6127. (c) Bowles,
D. M.; Palmer, G. J.; Landis, C. A.; Scott, J. L.; Anthony, J. E. Tetrahedron
2001, 57, 3753-3760. (b) Wang, K. K. Chem. ReV. 1996, 96, 207-222.
(4) (a) Myers, A. G.; Dragovich P. S.; Kuo, E. Y. J. Am. Chem. Soc.
1992, 114, 9369-9386. (b) Nagata, R.; Yamanaka, H.; Murahashi, E.; Saito,
I. Tetrahedron Lett. 1990, 31, 2907-2910. (c) Schmittel, M.; Maywald,
M.; Strittmatter, M. Synlett 1997, 165-166. (d) Feng, L.; Kumar, D.; Birney,
D. M.; Kerwin, S. M. Org. Lett. 2004, 6, 2059-2062. (e) Sullivan, R. W.;
Coghlan, V. M.; Munk, S. A.; Reed, M. W.; Moore, H. W. J. Org. Chem.
1994, 59, 2276-2278. (f) Braverman, S.; Zafrani, Y.; Gottlieb, H. E.
Tetrahedron 2001, 57, 9177-9185.
(1) (a) Xi, Z.; Goldberg, I. H. In ComprehensiVe Natural Products
Chemistry; Kool, E. T., Ed.; Elsevier: New York, 1999; pp 553-592. (b)
Thorson, J. S.; Sievers, E. L.; Ahlert, J.; Shepard, E.; Whitwam, R. E.;
Onwueme, K. C.; Ruppen, M. Curr. Pharm. Des. 2000, 6, 1841-1879. (c)
De Voss, J. J.; Townsend, C. A.; Ding, W. D.; Morton, G. O.; Ellestad, G.
A.; Zein, N.; Tabor, A. B.; Schreiber, S. L. J. Am. Chem. Soc. 1990, 112,
9669-9670.
(2) (a) Bhattacharyya, S.; Zaleski, J. M. Curr. Top. Med. Chem. 2004,
4, 1637-1654. (b) Jones, G. B.; Fouad, F. S. Curr. Pharm. Des. 2002, 8,
2415-2440. (c) Rawat, D. S.; Zaleski, J. M. Synlett 2004, 393-421
(5) (a) Lewis, K. D.; Wenzler, D. L.; Matzger, A. J. Org. Lett. 2003, 5,
2195-2197. (b) Bui, B. H.; Schreiner, P. R. Org. Lett. 2003, 5, 4871-
4874.
10.1021/ol0600638 CCC: $33.50
© 2006 American Chemical Society
Published on Web 04/19/2006

Scheme 1. Thermal Rearrangements of Enediynes and
Scheme 2. Synthesis of Aza-Enediynes and Their Conversion
Aza-Enediynes
to Enediynes17
Aza-enediynes have been the subject of a number of
theoretical7-11 and experimental studies.6,11-14 The 2,5-ddp
diradical is predicted to be a very short-lived, unreactive
species, and with one exception,11 efforts to trap this
intermediate have not been successful.6,12-14 As part of our
ongoing investigation into the unique chemistry of aza-
enediynes,12,14,15 we have uncovered a remarkable transfor-
mation of aza-enediynes to enediynes that can occur in
parallel with aza-Bergman rearrangement. Here we describe
this transformation and insights gained from the study of a
of the p-toluyl-substituted aza-enediyne 5b, mild thermolysis
in benzene affords the enediyne 11a and the toluylbisnitrile
1318 (Scheme 2).
l3
specifically C-labeled aza-enediyne.
Thermolysis of aza-enediyne [4-13C]-5a (120 mM) at 70
°C in benzene afforded [1,4-13C2]-enediynes (Z)-11a and (E)-
11a (1:3 ratio after chromatography, 60%), [5-13C]-7a (24%),
and unlabeled bisnitrile 12 (35%) (Scheme 3). The presence
The aza-enediynes 5a,b were prepared from 1,3-diphe-
nylpropynone (9) via the corresponding oxime mesitylate
(10), which undergoes cuprate coupling with higher order
alkynyl cuprates as shown in Scheme 2. These aza-enediynes
are isolated as predominantly a single imine double bond
isomer, presumably the more thermodynamically stable (Z)-
configuration.13 Aza-enediyne 5a is unstable when stored neat
for extended periods at -10 °C. In benzene (120 mM) at 55
°C over a course of days, 5a is converted to the enediyne
11a, isolated as a 1:1 E/Z mixture after chromatography, and
the diphenylfumaronitrile 12.16 However, under more dilute
conditions (0.45 mM) and higher temperature, 5a undergoes
aza-Bergman rearrangement to 7a (Scheme 2). In the case
Scheme 3
(6) David, W. M.; Kerwin, S. M. J. Am. Chem. Soc. 1997, 119, 1464-
1465.
(7) Cramer, C. J. J. Am. Chem. Soc. 1998, 120, 6261-6269.
(8) (a) Kraka, E.; Cremer, D. J. Am. Chem. Soc. 2000, 122, 8245-8264.
(b) Kraka, E.; Cremer, D. J. Comput. Chem. 2001, 22, 216-229.
(9) Schreiner, P. R.; Navarro-Vazquez, A.; Prall, M. Acc. Chem. Res.
2005, 38, 29-37.
of two 13C-labeled positions in both (Z)- and (E)-11a is
apparent from the 13C NMR spectra, in which the pair of
labeled carbons in each compound are coupled (4.1 and 4.6
Hz, respectively).19
The rate of disappearance of 5a monitored by H NMR
(120 mM 5a in benzene-d6, 70 ( 1 °C) does not follow either
(10) Schutz, M. J. Chem. Phys. 2002, 116, 8772-8785.
(11) Hoffer, J.; Schottelius, M. J.; Feichhtinger, D.; Chen, P. J. Am. Chem.
Soc. 1998, 120, 376-385.
(12) Feng, L.; Kerwin, S. M. Tetrahedron Lett. 2003, 44, 3463-3466.
(13) Feng, L.; Kumar, D.; Kerwin, S. M. J. Org. Chem. 2003, 68, 2234-
2242.
(14) (a) Nadipuram, A. K.; David, W. M.; Kumar, D.; Kerwin, S. M.
Org. Lett. 2002, 4, 4543-4546. (b) Kerwin, S. M.; Nadipuram, A. K. Synlett
2004, 8, 1404-1408.
(15) Nadipuram, A.; Kerwin, S. M. Tetrahedron 2006, 62, 3798-3808.
(16) Refat, H. M.; Waggenspack, J.; Dutt, M.; Zhang, H.; Fadda, A. A.;
Biehl, E. J. Org. Chem. 1995, 60, 1985-1989
1
(17) The reported yields of 11 and 12 assume a 2:1:1 stoichiometry of
starting aza-enediyne to enediyne to bisnitrile.
(18) Pochat, F. Tetrahedron Lett. 1978, 1055-1058.
1984
Org. Lett., Vol. 8, No. 10, 2006

first-order or second-order (with respect to 5a) kinetics, but
can be fit to a rate equation for parallel first- and second-
order reactions (Table 1).20 Data obtained from 13C NMR
Scheme 4
Table 1. Rate Constants for the First-Order Aza-Bergman
Rearrangement of 5a and the Second-Order Transformation of
5a to 11a at 70 °C
first-order rate
constantb (s-1
second-order rate
constantb (L mol-1 s-1
)
solventa
)
c
benzene-d6
THF-d8
(4.4 ( 0.3) × 10-5
(4.1 ( 0.2) × 10-4
d
(1.8 ( 1.3) × 10-5
(2.1 ( 0.8) × 10-4
a Initial 5a or [1-13C]-5a concentration of 120 mM. b The rate of
disappearance of 5a determined from NMR integrals versus an internal
1
standard (2,5-dimethylfuran for H NMR and solvent for 13C NMR) was
fit to simultaneous, parallel first- and second-order processes. For details
see the Supporting Information. c Average values from both 1H and 13C
NMR data. d Values from 13C NMR data.
studies of the disappearance of [4-13C]-5a were also fit to
parallel first- and second-order processes, which match the
appearance of 13C NMR signals for 12 and (Z)-11a,
respectively. These 13C NMR kinetic studies show that
enediyne 11a is formed predominantly as the (Z)-isomer,
which undergoes isomerization during extended heating21 to
afford mixtures of (E)- and (Z)-11a. NMR kinetic studies of
[4-13C]-5a were also carried out in THF-d8 (Table 1).
Previous studies have shown that the rate of aza-Bergman
rearrangement of 6-unsubstituted-3-aza-hex-3-ene-1,5-diynes
shows a slight solvent dependence; proceeding more slowly
in more polar solvents.13 Both the first- and second-order
components of the rate of disappearance of aza-enediyne 5a
show a similar decrease in the more polar solvent (Table 1).
Thermolysis of aza-enediyne 5a carried out in neat 1,4-
cyclohexadiene in attempts to trap the 2,5-ddp 6a or other
intermediates, such as 1,3-diphenylpropargyl carbene,21 that
might be involved in a dissociative process leading to 11a
failed to afford either the pyridine 8a or any other trapping
products; only 7a, 11a, and 12 were obtained. This, together
with the labeling studies and the observation that the rate of
conversion of 5a to 11a is second order in 5a leads to the
proposal that the conversion of aza-enediynes to enediynes
proceeds through an initial head-to-tail coupling. One pos-
sibility involves a concerted [2+2] dimerization to an
azacyclobutene22 (e.g., 14, Scheme 4), ring opening of which
produces an intermediate (15) similar to that involved in the
oxidative formation of alkynes from 1,2-hydrazones.23 In the
present case, the formation of the alkyne moiety is ac-
companied by elimination of the bisnitrile 12 (or in the case
of 15b the bisnitrile 13), which provides a strong thermo-
dynamic driving force for the conversion. This may involve
initial cyclization (to 16 or 1724) and elimination to produce
the enediyne 11 and 12 directly (from 16), or after diaza-
Bergman rearrangement of the azoacetylene 1825 derived
from 17 (Scheme 4). While 12 is predicted to be formed
initially as the (Z)-isomer, isomerization may occur during
thermolysis or chromatographic isolation,26 the later being
complicated by trace amounts of an unidentified product of
similar mobility, which may explain the nonequivalence of
the isolated yields of 12 and 11. There is some evidence for
these proposed intermediates. ESI-MS analysis of samples
of 5a stored at -10 °C in benzene demonstrate initial
formation of dimer, followed by formation of enediyne over
the course of days. Careful chromatography of these samples
affords trace amounts of colored compounds that upon mass
spectrometric analysis produce ions corresponding to dimers
of 5a; however, the small amount obtained and conversion
of these compounds to 11a and 12 has prevented their
complete characterization. Further evidence for the azacy-
clobutene 14 comes from the 13C NMR kinetic studies of
[4-13C]-5a, which demonstrate a low-intensity peak whose
chemical shift (54.6 ppm) and appearance and disappearance
over time is commensurate with such an intermediate.
(19) (a) Krivdin, L. B.; Della, E. W. Prog. NMR Spectrosc. 1991, 23,
301-610. (b) Porwoll, J. P.; Leete, E. J. Labelled Compd. Radiopharm.
1985, 22, 257-271. (c) Lambert, J.; Klessinger, M. Magn. Reson. Chem.
1987, 25, 456-461.
(20) The kinetic model is for two parallel, irreversible reactions of 5a,
one first-order in 5a and the other second-order in 5a: 5a f 7a (k1); 2(5a)
f 11a + 12 (k2).
(21) Shimizu, T.; Miyasaka, D.; Kamigata, N. Org. Lett. 2000, 2, 1923-
1925.
(24) We thank one of the reviewers for suggesting this possibility.
(25) For a discussion of the difficulties in preparing azoacetylenes, see:
Denonne, F.; Seiler, P.; Diederich, F. HelV. Chim. Acta 2003, 86, 3096-
3117.
(26) Yeh, H.-C.; Wu, W.-C.; Wen, Y.-S.; Dai, D.-C.; Wang, J.-K.; Chen,
C.-T. J. Org. Chem. 2004, 69, 6455-6462 and references therein.
(22) For an apparent [2+2] cycloaddition of a propyne iminium salt to
afford an azacyclobutene see: Nikolai, J.; Schlegel, J.; Regitz, M.; Maas,
G. Synthesis 2002, 497-504.
(23) Tsuji, J.; Kezuka, H.; Toshida, Y.; Takayanagi, H.; Yamamoto, K.
Tetrahedron 1983, 39, 3279-3282.
Org. Lett., Vol. 8, No. 10, 2006
1985

Examination of other aza-enediynes demonstrates that
there are certain structural requirements in order for the
conversion of aza-enediynes to enediynes to compete with
aza-Bergman rearrangement. Aza-enediyne 5c6 (Figure 1)
enediynes, presumably due to steric effects on the initial
dimerization to 14. However, for aza-enediynes bearing
terminal aryl substituents (e.g., 5a,b), the conversion to
enediynes can predominate over aza-Bergman rearrangement,
particularly at high concentrations of aza-enediyne or lower
temperatures.
The remarkable conversion of aza-enediynes to enediynes
adds another facet to the distinct chemistry12,14,15 of this class
of compounds, whose design was inspired by analogy to the
enediyne Bergman cyclization. Aza-enediynes have now
come full circle, serving as synthetic precursors to the source
of inspiration for their conception. Benzenoid diradicals
derived from Bergman cyclization of enediynes can be
efficiently trapped through hydrogen atom abstraction
reactions.1 There have been a number of attempts to detect
similar chemistry from aza-enediyne-derived 2,5-didehydro-
pyridines.6,11-13 The conversion of aza-enediynes to ene-
diynes reported here, which can occur in parallel with the
thermal aza-Bergman rearrangement of these compounds,
should inform future studies of the thermal chemistry of aza-
enediynes. For example, reactive diradicals from aza-
enediynes may arise from Bergman cyclization of the
enediynes derived from them rather than aza-Bergman-
derived 2,5-didehydropyridines.
Figure 1. Aza-enediyne structural effects on the conversion to
enediynes.
also apparently undergoes this transformation; chromatog-
raphy of a sample of 5c stored neat at -10 °C for 2 years
afforded the corresponding enediyne27 (14% yield, ∼1:1
mixture of E/Z isomers) and the bisnitrile 12 (35% yield).
For sterically unencumbered aza-enediynes 5d,e (Figure 1),
which undergo rapid aza-Bergman rearrangement at subam-
bient temperatures,3,10 no enediyne formation is observed.
Similarly, aza-enediynes that have terminal triisopropylsilyl
substituents (5f,g) which undergo aza-Bergman rearrange-
ment more slowly,10 also do not undergo conversion to
Acknowledgment. This work was supported by grants
from the Robert Welch Foundation (F-1298) and from the
National Institute of Environmental Health Sciences, NIH
(P30 ES07784). The technical assistance of Gwendolyn
Marriner and Sarah Pierce is gratefully acknowledged.
(27) The structure of the enediyne was assigned based on UV, IR, HRMS,
and comparison of the 1H NMR spectrum to that of 1,6-dimethyl-3,4-
diphenyl-3-hexene-1,5-diyne (Hayashi, M.; Saigo, K. Tetrahedron Lett.
1997, 38, 6241-6244) and 3,4-dimethyl-1,6-diphenyl-3-hexene-1,5-diyne
(Blackwell, J. M.; Figueroa, J. S.; Stephens, F. H.; Cummins, C. C.
Organometallics 2003, 22, 3351-3353).
Supporting Information Available: Experimental details
and kinetic plots. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL0600638
1986
Org. Lett., Vol. 8, No. 10, 2006