An efficient asymmetric synthesis of manzacidin C

Article abstract of DOI:10.1021/ol8011869

(Chemical Equation Presented) A brief synthesis of manzacidin C based on a chiral silane-promoted diastereo- and enatioselective acylhydrazone-alkene [3 + 2] cycloaddition reaction has been achieved. This synthesis is the first synthesis of any of the man

Full text of DOI:10.1021/ol8011869

ORGANIC
LETTERS
2008
Vol. 10, No. 14
3165-3167
An Efficient Asymmetric Synthesis of
Manzacidin C
Kristy Tran, Pamela J. Lombardi, and James L. Leighton*
Department of Chemistry, Columbia UniVersity, New York, New York 10027
leighton@chem.columbia.edu
Received May 23, 2008
ABSTRACT
A brief synthesis of manzacidin C based on a chiral silane-promoted diastereo- and enatioselective acylhydrazone-alkene [3 + 2] cycloaddition
reaction has been achieved. This synthesis is the first synthesis of any of the manzacidins wherein the C(4) and C(6) stereocenters are
established in a single highly stereoselective step.
In 1991 Kobayashi and co-workers reported the isolation
of manzacidins A, B, and C,1 bromopyrrole alkaloids
possessing an unusual 3,4,5,6-tetrahydropyrimidine ring, and
in 1997, the isolation of manzacidin D, the debromo-N-
methyl derivative of manzacidin A was reported (Scheme
1).2 Because of the scarcity of these compounds, their full
manzacidin A and C that served to confirm the relative and
establish the absolute stereochemistry of these compounds.5
There have been several additional syntheses of manzacidins
A,6 C,5,7 and D,8 and very recently the first synthesis of
manzacidin B was achieved that revised and established its
relative and absolute stereostructure.9
Our interest in the synthesis of the manzacidins was
elicited (1) by the observation that in none of the reported
syntheses were the C(4) and C(6) stereocenters both estab-
Scheme 1
(1) Kobayashi, J.; Kanda, F.; Ishibashi, M.; Shigemori, H. J. Org. Chem.
1991, 56, 4574.
(2) Jahn, T.; Ko¨nig, G. M.; Wright, A. D. Tetrahedron Lett. 1997, 38,
3883.
(3) For a summary and lead references, see: Faulkner, D. J. Nat. Prod.
Rep. 1998, 15, 113.
(4) Hashimoto, T.; Maruoka, K. Org. Biomol. Chem. 2008, 6, 829.
(5) Namba, K.; Shinada, T.; Teramoto, T.; Ohfune, Y. J. Am. Chem.
Soc. 2000, 122, 10708.
(6) (a) Wehn, P. M.; Du Bois, J. J. Am. Chem. Soc. 2002, 124, 12950.
(b) Kano, T.; Hashimoto, T.; Maruoka, K. J. Am. Chem. Soc. 2006, 128,
2174. (c) Wang, Y.; Liu, X.; Deng, L. J. Am. Chem. Soc. 2006, 128, 3928.
(d) Sibi, M. P.; Stanley, L. M.; Soeta, T. Org. Lett. 2007, 9, 1553.
(7) (a) Lanter, J. C.; Chen, H.; Zhang, X.; Sui, Z. Org. Lett. 2005, 7,
5905. (b) Wang, B.; Wu, F.; Wang, Y.; Liu, X.; Deng, L. J. Am. Chem.
pharmacological profiles initially proved difficult to establish,
but it has been shown that they possess activity as R-adreno-
ceptor blockers, antagonists of serotonergic receptors, and
actomyosin ATPase activators.3
Because of their unusual structures and a desire to obtain
significant quantities of them for more comprehensive
biological profiling, the manzacidins have proven to be
popular synthetic targets.4 Ohfune reported syntheses of
Soc. 2007, 129, 768
.
(8) Drouin, C.; Woo, J. C. S.; MacKay, D. B.; Lavigne, R. M. A.
Tetrahedron Lett. 2004, 45, 7197.
(9) Shinada, T.; Ikebe, E.; Oe, K.; Namba, K.; Kawasaki, M.; Ohfune,
Y. Org. Lett. 2007, 9, 1765.
10.1021/ol8011869 CCC: $40.75
Published on Web 06/24/2008
2008 American Chemical Society

lished in a single, experimentally straightforward, and
diastero- and enantioselective step and (2) by the possibility
to accomplish exactly that suggested by our recently reported
chiral silane Lewis acid (1) promoted acylhydrazone-enol
ether [3 + 2] cycloaddition methodology (Scheme 2).10
Scheme 3
Scheme 2
Retrosynthetic analysis along these lines led in a straight-
forward fashion to diamide 2 and then to pyrazolidine 3,
which would arise from a cycloaddition between a gly-
oxylate-derived hydrazone and a methallyl alcohol ether or
synthetic equivalent. It should be noted that the Maruoka6b
and Sibi6d syntheses employ conceptually related enantio-
selective diazoester-methacrolein/methacrylate [3 + 2] cyclo-
addition reactions. While these reactions do establish the C(6)
stereocenter, the C(4) stereocenter is established in a
subsequent operation that is only moderately diastereoselective.
At the outset it seemed likely that methallylsilane deriva-
tives (Scheme 2, X ) SiR3) might possess the requisite
reactivity to engage in the [3 + 2] cycloaddition reaction,
while also providing a functional equivalent for the requisite
alcohol by way of a Tamao oxidation. Indeed, silane 1 was
found to promote cycloaddition between hydrazone 4 and
methallylsilane 5 to give 6 with promising enantioselectivity,
albeit with no diastereoselectivity (Scheme 3). Our recently
reported second-generation silane 711 promoted the same
reaction with at least measurable diastereoselectivity and
excellent enantioselectivity (93% ee) for the major dia-
stereomer.12 Before taking on the optimization of these
cycloadditions we became enamored of the idea that we
might develop an effective cycloaddition using a simple
methallyl alcohol as the dipolarophile. Such a cycloaddition
would be “ideal” in that both stereocenters would be set in
a single reaction, with every relevant carbon atom in the
correct oxidation state, requiring no further manipulations
other than a final global deprotection. In line with our
expectations, however, treatment of hydrazone 4 with tert-
butyldiphenylsilyl (TBDPS) methallyl alcohol (8) and silane
1 or 7 under a variety of conditions gave no cycloaddition
product. In an effort to increase the activity of the silane
Lewis acids, preactivation with AgOTf was considered, and
indeed, treatment of (S,S)-7 with AgOTf led to a Lewis acid
that successfully promoted the reaction of 4 with 8 to give
9 in 38% yield and 94% ee (Scheme 3). Examination of the
reaction mixture revealed the presence of at least one other
product that was produced in significant (∼20%) amounts:
ene product 10. Assuming a stepwise mechanism,13 this
observation suggested that after initial attack of the alkene
on the silane-hydrazone complex, the resulting carbocation
partitions between ring closure to give 9 and elimination to
give 10. It seemed plausible in turn that modification of the
benzoylhydrazone to a more electron-rich aroylhydrazone
would increase the rate of ring closure while not impacting
the rate of elimination. Gratifyingly, this line of reasoning
led to the use of thienylhydrazone 11, which upon cyclo-
addition with 8 and (S,S)-7/AgOTf gave 12 as a single
diastereomer in 78% yield and 93% ee.
It will be noted that 12 is the opposite enantiomer to that
required for a synthesis of manzacidin C.14 Indeed, the
relative and absolute stereochemistry of 12 (and of 9) was
assigned by its conversion to ent-manzacidin C by the route
described below. The cycloaddition of 11 and 8 was therefore
carried out with the enantiomeric silane (R,R)-7 and AgOTf
to provide ent-12 in 73% yield as a single diastereomer in
(13) We have established a stepwise mechanism for the acylhydrazone-
enol ether cycloaddition reaction. See ref 10.
(10) Shirakawa, S.; Lombardi, P. J.; Leighton, J. L. J. Am. Chem. Soc.
2005, 127, 9974.
(14) This result was unexpected and is a reversal of the absolute sense
of induction observed when hydrazone 4 is engaged in Friedel-Crafts
reactions (see : Shirakawa, S.; Berger, R.; Leighton, J. L. J. Am. Chem.
Soc. 2005, 127, 2858). The mechanistic basis for this reversal, possibly
associated with the use of AgOTf, is under investigation.
(11) Notte, G. T.; Leighton, J. L. J. Am. Chem. Soc. 2008, 130, 6676.
(12) The relative and absolute stereochemistry of the diastereomers of
6 has not been rigorously established.
3166
Org. Lett., Vol. 10, No. 14, 2008

94% ee (Scheme 4). Formylation of hydrazide ent-12
proceeded smoothly to give 13 and provided the final carbon
atom for the synthesis of the tetrahydropyrimidine ring.
Hydrazide reduction with SmI2 was straightforward and gave
bisamide 14 in 73% yield over two steps. The task of
devising a dehydrative cyclization of 14 proved nontrivial
and required extensive screening of dehydrating agents,
bases, and reaction conditions.15 Among the problems
encountered were a propensity for epimerization of the C(4)
stereocenter, cyclization onto the thienyl amide instead of
the formamide, and instability of the desired product to silica
gel chromatography. Eventually it was found that the
combination of PCl5 and 4-dimethylaminopyridine (DMAP)
in CHCl3 was effective in promoting the desired cyclization
with minimal epimerization to give 15. Subjection of 15
directly to global deprotection with concentrated HCl gave
16 and was followed by installation of the bromopyrrole side
chain according to the procedure of Ohfune5 to give
manzacidin C in 49% yield over the three-step sequence from
14. Full spectral comparison (1H and 13C NMR, IR, MS,
optical rotation) confirmed the identity of our synthetic
material.
Scheme 4
A brief, efficient, and stereocontrolled synthesis of manz-
acidin C has been achieved (6 steps from 11 and 8 in 26%
overall yield). The synthesis inspired the discovery that the
AgOTf-modified silane 7 has significantly greater activity,
which in turn led to a significant expansion in the scope of
our acylhydrazone-alkene [3 + 2] cycloaddition and made
possible the cycloaddition between 11 and 8, a reaction that
establishes both stereocenters of the target in a single highly
diastereo- and enantioselective step.
Acknowledgment. Financial support was provided by the
National Science Foundation (CHE-04-53853) and a Focused
Funding Award from Johnson & Johnson. We thank Amgen
and Merck Research Laboratories for unrestricted support.
Supporting Information Available: Experimental pro-
cedures, characterization data, and stereochemical proofs.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(15) A similar cyclization was carried out by Du Bois and Wehn, and
they describe similar observations. See ref 6a.
OL8011869
Org. Lett., Vol. 10, No. 14, 2008
3167