Isolation of a cyclopropane-containing product from the rearrangement of a 3-aza-3-ene-1,5-diyne under acid catalysis

Article abstract of DOI:10.1016/S0040-4039(03)00675-0

A 3-aza-enediyne that undergoes rapid aza-Bergman rearrangement was treated with trifluoromethanesulfonic acid in the presence of 1,4-cyclohexadiene in an attempt to trap the putative 2,5-didehydropyridinium aza-Bergman intermediate. No pyridine products were detected; rather, a cyclopropane derivative of 1,4-cyclohexadiene derived from a 5-oxazolylcarbene was isolated.

Full text of DOI:10.1016/S0040-4039(03)00675-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3463–3466
Isolation of a cyclopropane-containing product from the
rearrangement of a 3-aza-3-ene-1,5-diyne under acid catalysis
Liping Feng and Sean M. Kerwin*
Division of Medicinal Chemistry, College of Pharmacy, Institute for Cellular and Molecular Biology,
University of Texas at Austin, Austin, TX 78712, USA
Received 24 February 2003; revised 10 March 2003; accepted 12 March 2003
Abstract—A 3-aza-enediyne that undergoes rapid aza-Bergman rearrangement was treated with trifluoromethanesulfonic acid in
the presence of 1,4-cyclohexadiene in an attempt to trap the putative 2,5-didehydropyridinium aza-Bergman intermediate. No
pyridine products were detected; rather, a cyclopropane derivative of 1,4-cyclohexadiene derived from a 5-oxazolylcarbene was
isolated. © 2003 Elsevier Science Ltd. All rights reserved.
Enediynes (Scheme 1, A, X=CH) undergo Bergman
cyclization1 to reactive 1,4-didehydrobenzene intermedi-
ates (B, X=CH), and this process has been the subject
of much theoretical,2 synthetic,3 and biological inter-
est.4 The chemistry of 3-aza-enediynes (Scheme 1, A,
X=N) has not been as widely studied. We have shown
that 3-aza-enediynes undergo a rapid aza-Bergman
rearrangement through a 2,5-didehydropyridine inter-
mediate (B, X=N) to afford the isomeric b-alkynyl
acrylonitrile products (C, X=N).5 Unlike the Bergman
cyclization of enediynes, the aza-Bergman rearrange-
ment of 3-aza-enediynes does not afford any products
corresponding to trapping of the intermediate diradi-
cals. Theoretical studies6 have predicted that the parent
2,5-didehydropyridine intermediate (B, X=N, R1=
R4=R6=H) can undergo a rapid retro-aza-Bergman
reaction to the nitrile (C, X=N, R1=R4=R6=H).
This, together with the large singlet-state stabilization7
of the 2,5-didehydropyridine diradical, conspire to
make the 2,5-didehydropyridine intermediate an unre-
active, very short-lived species. In contrast, the proto-
nated 2,5-didehydropyridinium diradical (Scheme 1, B,
X=NH+) is predicted to be more resistant to retro-aza-
Bergman rearrangement and have a smaller singlet–
triplet gap, indicating that this intermediate would
more readily undergo trapping reactions.6 These predic-
tions have led to the proposal that these aza-enediynes
might serve as pH-dependent DNA cleavage agents in
which the pH-dependence is mediated by the effect of
protonation on the reactivity of the intermediate dirad-
ical.6 We recently reported that sterically unencumbered
acyclic 3-aza-3-ene-1,5-diynes undergo aza-Bergman
rearrangement spontaneously at room temperature and
below.8 This provided an opportunity to examine the
effect of protonation on the reactivity of the intermedi-
ates formed by aza-Bergman cyclization of these 3-aza-
3-ene-1,5-diynes. Here we report our studies on the
effect of added acid on the rearrangement chemistry of
3-aza-3-ene-1,5-diynes. Instead of products derived
from trapping of the protonated 2,5-didehydropyridine
diradical intermediate, one such 3-aza-3-ene-1,5-diyne
affords a cyclopropane product apparently derived
from a carbene intermediate.
The triisopropylsilyl-substituted aza-enediyne 2 was
prepared as previously reported.8 Briefly, the propy-
none 1 was converted to the oxime, which was activated
as the sulfonate ester followed by addition of a cuprate
reagent derived from phenylacetylene to afford the
aza-enediyne 2 in modest yield (Scheme 2). The aza-
enediyne 2 can be purified by chromatography and
isolated as relatively stable yellow oil. However, when a
solution of aza-enediyne 2 in chlorobenzene is heated at
150°C for a period of days, the aza-enediyne undergoes
Scheme 1.
* Corresponding author. E-mail: skerwin@mail.utexas.edu
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00675-0

3464
L. Feng, S. M. Kerwin / Tetrahedron Letters 44 (2003) 3463–3466
Scheme 2.
aza-Bergman rearrangement to afford the nitrile 4
(Scheme 2). Initial attempts to convert the triisopropyl-
silyl-substituted aza-enediyne 2 to the desilylated aza-
enediyne 3 by treatment with tetrabutylammonium
fluoride (TBAF) followed by aqueous work-up and
chromatography led instead to the isolation of the
aza-Bergman rearrangement product 5 as the sole
product in 89% yield. However, treatment of 2 with
TBAF followed by rapid aqueous work-up and imme-
1
diate H NMR analysis of the reaction mixture demon-
strates the presence of the aza-enediyne
3
as
Scheme 3.
predominantly the (Z)-isomer, which undergoes con-
version to the nitrile 5 with a first-order half-life of 75
min at 20°C.
Desilylation of aza-enediyne 2 was carried out at −78°C
for 5–10 min, and the reaction mixture subjected to
rapid aqueous work-up. The organic extracts were
cooled in a dry ice bath, dried over Na2SO4, and passed
through a plug of silica gel to remove the tetra-
butylammonium species. After evaporation of the sol-
vent, the resulting oil consisting of aza-enediyne 3 along
with variable amounts of the nitrile 5 was dissolved in
CH2Cl2 and treated with catalytic trifluoromethanesul-
fonic acid (TfOH) in the presence of an excess of
1,4-cyclohexadiene (1,4-chd) at −10°C for 4 days. Mon-
itoring of the progress of the reaction by TLC demon-
strated the disappearance of aza-enediyne 3 and the
appearance of two products, the aza-Bergman rear-
rangement product 5 and a more polar product 6
(Scheme 3). These products were isolated after evapora-
tion of the reaction mixture and chromatography. The
structural characterization of the cyclopropane 6 was
carried out by NMR and single-crystal X-ray diffrac-
tion studies (Fig. 1).9
Figure 1. X-Ray structure of 6 showing the atom labeling
scheme. Displacement ellipsoids are scaled to the 30% proba-
bility level. The hydrogen atoms are drawn to an arbitrary
size.
along with a 15% yield of the nitrile 5. In the absence of
added acid, only nitrile 5 was isolated in 89% yield.
Re-subjecting nitrile 5 to the reaction conditions
resulted only in recovered starting material; there was
no evidence of conversion of nitrile 5 to the cyclo-
propane 6 under these reaction conditions.
The cyclopropanyl oxazole 6 appears to be the trapping
product of an intermediate carbene 8 (Scheme 4).
Although the origin of this carbene is uncertain, a
possible route involves the acid-catalyzed addition of
When the reaction was carried out in the presence of
stoichiometric TfOH, complete consumption of the aza-
enediyne was observed in 30 min at −78°C, and the
cyclopropane derivative 6 was isolated in 30% yield,

L. Feng, S. M. Kerwin / Tetrahedron Letters 44 (2003) 3463–3466
3465
Scheme 4.
water to the ynamine functionality10 of the 3-aza-
enediyne 3 to afford an intermediate N-acyl-C-alkynyl
imine 7. The presence of adventitious water in the crude
desilylated 3-aza-enediyne 3 would be expected, given
the procedures employed in the work-up of the desilyla-
tion reactions, which involved storing solutions of 3 on
dry-ice during work-up and isolation in order to mini-
mize conversion to the nitrile 5. The proposed cycliza-
The substitution of nitrogen for carbon in the enediyne
system can have a profound impact on the facility of
the Bergman rearrangement and on the nature of the
intermediates that are involved. Here we demonstrate
that certain 3-aza-enediynes can also undergo reaction
under acidic aqueous conditions to afford carbene
intermediates that can be efficiently trapped. The gener-
ality of this reaction and its application to pH-depen-
dent DNA cleavage reactions are under investigation.
tion of the N-acyl-C-alkynyl imine
7
to the
5-oxazolylcarbene 8 is related to the observations of
Schecter and co-workers,11 who found that certain 5-
oxazolylcarbenes, generated by thermolysis or photoly-
sis of the corresponding diazo compounds, undergo
CꢀH insertion and cyclopropanation reactions as well
as ring-opening to N-acyl-C-alkynyl imines. The facile
ring opening of related 2-furylcarbenes is well known,12
and there are examples of the reverse reaction in which
a heteroatom-substituted dieneyne system undergoes
thermal cyclization to a carbene intermediate.13 We
propose that the reversible formation of the carbene 8
from the N-acyl imine 7 followed by addition of 8 to
the double bond of 1,4-chd leads to consumption of the
N-acyl imine species and formation of 6 (Scheme 4).
Acknowledgements
We thank Vincent Lynch of the Department of Chem-
istry for the single-crystal X-ray diffraction data for
compound 6. This work was supported by grants from
the Robert Welch Foundation (F-1298) and the
Petroleum Research Fund, administered by the Ameri-
can Chemical Society (AC-34115).
References
1. Bergman, R. G. Acc. Chem. Res. 1973, 6, 25–31.
2. (a) Schreiner, P. R. J. Am. Chem. Soc. 1998, 120, 4184–
4190; (b) Cramer, C. J.; Nash, J. J.; Squires, R. R. Chem.
Phys. Lett. 1997, 277, 311–320.
3. Jones, G. B.; Fouad, F. S. Curr. Pharm. Des. 2002, 8,
2415–2440.
4. (a) Dai, W. M.; Lai, K. W.; Wu, A.; Hamaguchi, W.;
Lee, M. Y. H.; Zhou, L.; Ishii, A.; Nishimoto, S. J. Med.
Chem. 2002, 45, 758–761; (b) Liu, W.; Christenson, S. D.;
Standage, S.; Shen, B. Science 2002, 297, 1170–1173.
5. David, W. M.; Kerwin, S. M. J. Am. Chem. Soc. 1997,
119, 1464–1465.
6. (a) Hoffner, J.; Schottelius, M. J.; Feichtinger, D.; Chen,
P. J. Am. Chem. Soc. 1998, 120, 376–385; (b) Cramer, C.
J. J. Am. Chem. Soc. 1998, 120, 6261–6269; (c) Kraka, E.;
Cremer, D. J. Comput. Chem. 2001, 22, 216–229.
7. Logan, C. F.; Chen, P. J. Am. Chem. Soc. 1996, 118,
2113–2114.
The role of adventitious water in the formation of 6
was established by carrying out the isolation of crude 3
and the TfOH-catalyzed rearrangement in the presence
of 1,4-chd under strictly anhydrous conditions and in
the presence of 4 A molecular sieves. Under these
conditions, no cyclopropane 6 is observed; instead, only
nitrile 5 is isolated in 87% yield.
,
The isolation of the cyclopropyloxazole 6 from aza-
enediyne 3 under acidic conditions contrasts with the
results of Chen and co-workers,6a who report that the
aza-enediyne 9 (Scheme 1, A, X=N, R1=R6=Ph, R4=
Me) affords primarily the hydrolysis products 4-phenyl-
but-3-yn-2-one and phenylacetonitrile under acidic
conditions. When aza-enediyne 2 is subjected to similar
reaction conditions as employed in the conversion of 3
to 6, the hydrolysis product 1-phenyl-3-triisoproylsilyl-
propynone is obtained. No pyridine products corre-
sponding to trapping of a reactive 2,5-didehydro-
8. Feng, L.; Kumar, D.; Kerwin, S. M. J. Org. Chem. 2003,
68, 2234–2242.
1
pyridinium species were detected in the crude H NMR
9. The product 6 was re-crystallized in hexanes and afforded
1
spectra of any of the acidic reactions involving either 2
or 3, which is also in contrast with Chen’s report of a
pyridine product from the thermolysis of aza-enediyne
9 under acidic conditions.6a We note, however, that our
analysis would not be able to detect the extremely low
yields (ꢀ0.05%) of pyridine product reported by Chen.
colorless lathes and needles: mp=85.0–86.4°C; H NMR
(500 MHz, CDCl3) l 7.88 (dd, J=8.5, 1.4 Hz, 2H,
H21/25), 7.37 (t, J=8.1 Hz, 2H, H16/18), 7.29 (dd,
J=7.1, 1.0 Hz, 2H, H15/19) 7.28–7.15 (m, 4H, H22/24
and H23 and H17), 4.89 (s(br), 2H, H9/10), 4.04 (s, 2H,
H13), 2.22 (d(br), J=19.8 Hz, 2H, Heq8/11), 2.18–2.00
(m, 3H, Hax8/11 and H6), 1.60 (dd, J=7.7, 4.5 Hz, 2H,

3466
L. Feng, S. M. Kerwin / Tetrahedron Letters 44 (2003) 3463–3466
H7/12); 13C NMR (CDCl3) l 13.77 (C6), 15.79 (C7/12),
data for this structure has been deposited with the Cam-
bridge Crystallographic Data Centre as supplementary
publication number CCDC 204592 and can be obtained,
free of charge, from CCDC at deposit@ccdc.cam.ac.uk.
10. Feustel, M.; Himbert, G. Tetrahedron Lett. 1983, 24,
2165–2168.
11. Hayashi, S-I.; Nair, M.; Houser, D. J.; Shechter, H.
Tetrahedron Lett. 1979, 20, 2961–2964.
12. (a) Birney, D. M. J. Am. Chem. Soc. 2000, 122, 10917–
10925; (b) Khasanova, T.; Sheridan, R. S. J. Am. Chem.
Soc. 1998, 120, 233–234.
20.78 (C8/11), 34.85 (C13), 123.15 (C9/10), 126.00 (C21/
25), 126.83 (C17 or C22/24), 126.84 (C17 or C22/24),
128.39 (C15/19 or C16/18), 128.45 (C15/19 or C16/18),
128.86 (C23), 132.77 (C4), 135.67 (C20), 136.01 (C14),
145.72 (C5), 160.45 (C2); HRMS m/z 328.1711 (calcd
328.1702, C23H21NO). The data crystal was cut from a
long lathe and had approximate dimensions; 0.27×0.17×
0.04 mm. The data were collected on a Nonius Kappa
CCD diffractometer using a graphite monochromator
,
with MoKa radiation (u=0.71073 A). A total of 400
13. (a) Kimball, D. B.; Herges, R.; Haley, M. M. J. Am.
Chem. Soc. 2002, 124, 1572–1573; (b) Frey, L. F.; Tillyer,
R. D.; Ouellet, S. G.; Reamer, R. A.; Grabowski, E. J. J.;
Reider, P. J. J. Am. Chem. Soc. 2000, 122, 1215–1216.
frames of data were collected using v-scans with a scan
range of 1° and a counting time of 136 seconds per frame.
The data were collected at 153 K using an Oxford
Cryostream low temperature device. Crystallographic