Formation of 9,10-unsaturation in the mitomycins: Facile fragmentation of β-alkyl-β-aryl-α-oxo-γ-butyrolactones

Article abstract of DOI:10.1021/ol000245g

(matrix presented) A facile fragmentation of β-alkyl-β-aryl-α-oxo-γ-butyrolactones is reported. A study to assist in the elucidation of the mechanism of the reaction is also revealed.

Full text of DOI:10.1021/ol000245g

ORGANIC
LETTERS
2000
Vol. 2, No. 23
3619-3621
Formation of 9,10-Unsaturation in the
Mitomycins: Facile Fragmentation of
â-Alkyl-â-aryl-r-oxo-γ-butyrolactones
Frederick E. Ziegler,* Michael Y. Berlin, Kyungae Lee, and Adam R. Looker
Sterling Chemistry Laboratory, Yale UniVersity, New HaVen, Connecticut 06511-8118
frederick.ziegler@yale.edu
Received August 25, 2000
ABSTRACT
A facile fragmentation of â-alkyl-â-aryl-r-oxo-γ-butyrolactones is reported. A study to assist in the elucidation of the mechanism of the
reaction is also revealed.
In studies directed toward the synthesis of mitomycin A (1),
we demonstrated that the enantiomerically pure tetracycle 2
could be formed by a highly stereoselective aziridinyl radical
cyclization. Subsequent transformations led to (+)-9a-
desmethoxymitomycin A.1 Eventual oxidative introduction
of the requisite C9a methoxy group requires suppression of
indole formation by protection of the C9 position with a
removable protecting group. A formyl group, removed by
decarbonylation, had served this end in a previous, related
study.2
propargyl alcohol, and the crude product 5b was subjected
to tandem radical cyclization to afford pentacycle 6 in 39%
overall yield from the aldehyde (Scheme 1).1 The cyclization
Scheme 1
To apply a similar strategy to a synthesis of mitomycin K
(4),3 a protecting group was sought, which upon liberation
would reveal the C9,10 double bond. This communication
details a mild method to achieve this goal.
The hydroxyl group of cis-3-indolyl alcohol 5a,4 prepared
by NaBH4 reduction of the aldehyde, was exchanged with
worked with the trans-bromide equally well. The stereo-
chemistry of 6 was inferred from the stereochemistry
obtained in the formation of tetracycle 2 wherein hydrogen
10.1021/ol000245g CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/25/2000

atom delivery occurred on the convex face of the intermediate
tetracyclic benzylic radical. Cyclization of a benzylic radical
with the acetylene to form 6 was expected to occur in a
similar fashion.5 Moreover, an NOE (6.7%) was observed
between the C1-H and the proximate C10-H.
To elucidate the mechanism of the reaction, the hypothesis
of Scheme 3 was tested. Intermediate dioxetane 14,9 as a
Scheme 3
Prior to exploring the introduction of oxygen at C9a,
prudence dictated that a method be developed first for
excising the atoms of the allyl ether moiety of 6 to produce
a styrene. Although the exocyclic double bond of 6 could
be readily isomerized to the endo cyclic position [cat. (Ph3P)3-
RhCl, DABCO, aqueous EtOH, 95 °C, 83%], efforts to
functionalize or cleave oxidatively the endo cyclic double
bond were unsuccessful. However, olefin 6 was converted
to ketone 8 by ozonolysis with the proviso that the basic
nitrogen of 6 was first protected as its N-oxide. Direct
ozonolysis led to decomposition. Not only did dimethyl
sulfide serve its usual role of reducing the ozonide, but it
also effected reduction of the N-oxide.
After several unsuccessful attempts to oxygenate the
carbon adjacent to the carbonyl group in spiro dihydrofura-
none 8, treatment of the furanone under the Gardner protocol6
[KHMDS, (EtO)3P, and O2] also failed to give any R-ketol
but rather surprisingly and rewardingly afforded the desired
styrene 7 in excellent yield! Moreover, the reaction proceeded
in the absence of the phosphite. Olefin 7 had been prepared
previously by Martin sulfurane dehydration (70%)7 where
other more conventional elimination techniques proved
unsuccessful.8
radical or anion, could fragment to the carboxylate anion or
radical 15, which could undergo elimination. The â-formyl-
oxy carboxylic acid of 15 was readily accessible from
â-hydroxy acid 9a. Treatment of the formate ester under
simulated reaction conditions led only to hydroxy acid 9b,
saponification occurring presumably as the result of adventi-
tious hydroxide.
The possibility of a radical decarboxylation of acyloxy
radical 15 to an intermediate benzyl radical prior to elimina-
tion was considered less likely because such radical species
are the product of acyloxy group migration from the benzylic
position to a primary radical.10 Nonetheless, the formyloxy
carboxylic acid was activated as its thiohydroxamate ester
and photolyzed with visible light.11 No styrene was observed,
but stereoisomeric benzyl dimers were identified along with
sulfur-containing products of radical origin.
The generality of the reaction was explored on the less
complex, racemic furanone 12, prepared from (S)-ibuprofen
(9a) as described in Scheme 2. The model retained a
An R-keto-γ-butyrolactone was considered as a likely
intermediate in the elimination procedure. To explore this
possibility, ketolactone 18 was prepared as outlined in
Scheme 4. Upon exposure of this material to K2CO3 in
aqueous THF at room temperature, R-methylstyrene and
oxalic acid were formed. Oxalic acid was identified by 13C
Scheme 2
1
NMR and by H NMR of its dimethyl ester.
McMurry has reported the formation of R-methylene
cyclohexanone from oxalocyclohexanone 20 (Scheme 5)
upon exposure of the latter compound to gaseous formal-
dehyde in aqueous NaHCO3 at 0 °C.12,13 The presumed
1
(4) New compounds were characterized by H and 13C NMR spectros-
copy, combustion analysis, and/or HRMS.
(5) For an example of stereocontrol in spiroalkylation, see: Stork, G.;
Danheiser, R. L., Ganem, B. J. Am. Chem. Soc. 1973, 95, 3414.
(6) (a) Gardner, J. N.; Carlon, F. E.; Gnoj, O. J. Org. Chem. 1968, 33,
3294. (b) Gardner, J. N.; Popper, T. L.; Carlon, F. E.; Gnoj, O.; Herzog, H.
L. J. Org. Chem. 1968, 33, 3695.
(7) Martin, J. C.; Arhart, R. J. J. Am. Chem. Soc. 1971, 93, 4327.
(8) For a related elimination, see: Jones, G. B.; Guzel, M.; Mathews, J.
E. Tetrahedron Lett. 2000, 41, 1123.
(9) For examples of the fragmentation of 1,2-dioxetanes, see: Singlet
Oxygen; Schaap, P. A., Zaklika, K. A., Eds.; Academic Press: New York,
1979; Vol. 40, Chapter 6, p 173.
quaternary carbon and an aromatic ring. The orange enolate
solution of 12 was bleached immediately by O2 to give olefin
11 in the absence of phosphite.
(10) Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. Chem. ReV.
1997, 97, 3273.
(1) Ziegler, F. E.; Berlin, M. Y. Tetrahedron Lett. 1998, 39, 2455.
(2) Ziegler, F. E.; Belema, M. J. Org. Chem. 1997, 62, 1083.
(3) For total syntheses of mitomycin K, see: (a) Benbow, J. W.; McClure,
K. F.; Danishefsky, S. J. J. Am. Chem. Soc. 1993, 115, 12305. (b) Wang,
Z.; Jimenez, L. S. Tetrahedron Lett. 1996, 37, 6049.
(11) Barton, D. H. R.; Crich, D.; Motherwell, W. B. Tetrahedron 1985,
41, 3901.
(12) Ksander, G. M.; McMurry, J. E.; Johnson, M. J. Org. Chem. 1977,
42, 1180.
(13) See also, Nield, C. H. J. Am. Chem. Soc. 1945, 67, 1145.
3620
Org. Lett., Vol. 2, No. 23, 2000

Scheme 4
Scheme 6
intermediate, nonenolic â-dicarbonyl 21 suffers retro-Claisen
cleavage and subsequent â-elimination. Undoubtedly, this
reaction can be considered of the E1cB type, owing to the
presence of the cyclohexanone carbonyl.
lated under vigorous acidic conditions. That the decarboxy-
lation conditions did not alter the labeling pattern in lactone
27a was confirmed by integration16 of the diastereotopic
methylene protons (1H NMR) of the derived vinylogous
amide 27b. Exposure of 27c to KHMDS or K2CO3 in THF
containing 1.5 equiv of water at room temperature gave rise
to a 4.67:1.00 mixture of (E)-R-methylstyrene 25 and its (Z)-
isomer, respectively, or an 82% “retention of configuration”.
Assuming that inversion occurs via a process that permits
free bond rotation, then 18% of the retained configuration
arises by bond dissociation. Thus, 64% of the product can
be demonstrated to form by a concerted, syn-elimination.
The product arising from isomerization may be the result
of a concerted elimination if the isomerization is prior to
the product determining step, or it may be the result of a
stepwise elimination. The clarification of this issue has not,
as yet, been addressed. No product of C-C bond cleavage
and protonation was detected.
Scheme 5
The aromatic ring clearly facilitates the elimination
because furanone 23, prepared by the method of Scheme 2
from tetrahydro-â-naphthoic acid, did not lead to olefin 24.
Paquette17 has observed both retention and inversion in
the Haller-Bauer (NaNH2, C6H6, reflux) cleavage (proto-
nolysis) of R,R-dialkyl deoxybenzoins and elimination to
R-methylstyrene with R-methyl-R-allyloxymethyl desoxy-
benzoin.18
The facility with which this fragmentation occurs is
reflected in the pyruvate nature of the carbonyl group and
the syn arrangement of the elimination. R-Ketolactones of
this type may serve as useful synthetic templates from which
R-alkyl styrenes may be synthesized.
A deuterium labeling experiment was designed to deter-
mine if the elimination reaction of ketolactone 18 was
concerted or stepwise. (E)-Styrene-d114 was hydroborated and
ultimately converted into diazomalonate 28b (Scheme 6).
Stereocontolled, rhodium-mediated C-C bond formation15
afforded a mixture of lactone esters, which was decarboxy-
Acknowledgment. This research was supported by PHS
grant GM-54499.
OL000245G
(14) Negishi, E.; Van Horn, D. E.; Yoshida, T. J. Am. Chem. Soc. 1985,
107, 6639.
(16) (E)- R-Methylstyrene-d1 (25) contained 16% d0-compound.
(17) Paquette, L. A.; Gilday, J. P.; Maynard, G. D. J. Org. Chem. 1989,
54, 5044.
(15) (a) Ledon, H.; Linstrumelle, G.; Julia, S. Tetrahedron Lett. 1973,
25. (b) Taber, D. F.; Petty, E. H.; Raman, K. J. Am. Chem. Soc. 1985, 107,
196.
(18) Paquette, L. A.; Maynard, G. D. J. Org. Chem. 1989, 54, 5054.
Org. Lett., Vol. 2, No. 23, 2000
3621