Asymmetric total synthesis of the proposed structure of the medicinal alkaloid jamtine using the chiral base approach.

Article abstract of DOI:10.1021/ol027447s

[reaction: see text] A highly step-economic asymmetric synthesis of the tetracyclic structure assigned to the medicinal alkaloid jamtine has been accomplished using a chiral lithium amide base desymmetrization of a ring-fused imide. The structure synthesi

Full text of DOI:10.1021/ol027447s

ORGANIC
LETTERS
2003
Vol. 5, No. 4
535-537
Asymmetric Total Synthesis of the
Proposed Structure of the Medicinal
Alkaloid Jamtine Using the Chiral Base
Approach
Nigel S. Simpkins* and Christopher D. Gill
School of Chemistry, UniVersity of Nottingham, Nottingham, United Kingdom
nigel.simpkins@nottingham.ac.uk
Received December 11, 2002
ABSTRACT
A highly step-economic asymmetric synthesis of the tetracyclic structure assigned to the medicinal alkaloid jamtine has been accomplished
using a chiral lithium amide base desymmetrization of a ring-fused imide. The structure synthesized appears to be different from that of the
natural product originally reported.
As part of our ongoing program aimed at developing the
chemistry of chiral lithium amide bases, we have recently
described highly enantioselective desymmetrization reactions
of certain types of cyclic imide.1,2 Typically, reaction of a
ring-fused imide, e.g. cyclopropane derivative 1, with chiral
base 3 gave substituted products such as 2 in high yield and
enantioselectivity (Scheme 1).
A particularly important aspect of this type of desymme-
trization is that the silyl or alkyl groups installed completely
control the regiochemistry of subsequent imide reaction.
Thus, reaction of silylimide 2 with DIBAL occurred only at
the carbonyl function distal to the silicon substituent, leading
to hydroxylactam 4.3 This approach allows efficient enan-
tioselective access to a range of substituted lactams and
related products.
In seeking to apply this chemistry to interesting alkaloid
targets we were attracted to a recent report from the group
of Padwa,4 describing the first synthesis of an unusual
alkaloid called jamtine 5. This compound was originally
reported, in the form of an N-oxide, as one of a small group
of isoquinoline alkaloids isolated from the climbing shrub
Cocculus hirsutus.5 This plant is commonly found in parts
of Pakistan and its various parts are reputed to have
therapeutic properties according to local folk medicine.6
Scheme 1
(1) Adams, D. J.; Blake, A. J.; Cooke, P. A.; Gill, C. D.; Simpkins, N.
S. Tetrahedron 2002, 58, 4603 (part of a special issue dedicated to chiral
lithium amides).
(2) For a review, see: O’Brien, P. J. Chem. Soc., Perkin Trans. 1 1998,
1439.
(3) Speckamp, W. N.; Hiemstra, H. Tetrahedron 1985, 41, 4367.
10.1021/ol027447s CCC: $25.00 © 2003 American Chemical Society
Published on Web 01/21/2003

We were interested in the asymmetric synthesis of jamtine,
starting from a chiral imide 6, in which the substituent X
would become the carbomethoxy substituent present in the
natural product (Scheme 2).
results with base 3, and indeed imide 8 gave similarly poor
chemical yield and enantioselectivity under our usual condi-
tions. Instead, use of base 11, a monolithiated diamine base,
proved much more effective and allowed highly enantio-
selective carboxymethylation to give (-)-10 on quenching
the reaction with Mander’s reagent.7
With imide 10 available in essentially enantiopure form
we next probed possibilities for regioselective imide reduc-
tion as a prelude to cyclization to form the isoquinoline.
Efficient and completely regioselective reduction was achieved
simply by reaction of 10 with NaBH4 in EtOH (Scheme 4).8
Scheme 2
Scheme 4a
This idea presented two specific issues to be overcome.
First, in our previous work succinimides having a fused
cyclohexane ring proved to be the sole substrates that did
not give good results on reaction with base 3. Second, the
elaboration of 6 toward jamtine requires regioselective
reaction at the imide carbonyl proximal to the installed group
X. This complementary mode of reaction, compared to that
seen in reduction of 2 to give 4, has little precedent, but
appeared viable through electronic or chelation modes of
activation.
Herein we describe the successful application of this
strategy to a new and very concise asymmetric synthesis of
alkaloid 5, and we conclude that this is probably not the
correct structure of the alkaloid originally isolated.
Our synthesis started with the construction of an ap-
propriately substituted meso-imide 8 by the straightforward
combination of commercially available anhydride 7 and the
amine 9 (Scheme 3). As mentioned above, this type of system
(i.e. with NMe or NPh substituents) had not given good
a Reagents: (a) NaBH4, EtOH, -5 °C, 89%; (b) camphorsulfonic
acid, toluene, 80 °C, 69%; (c) KH, PhSeSePh, THF, 50 °C, then
H2O2, py, CH2Cl2, rt, 65% overall; (d) Me3OBF4, 2,6-di-tert-butyl-
4-methylpyridine, CH2Cl2, rt, then NaBH4, MeOH, 0 °C, 69%.
Scheme 3a
Since metal chelation is unlikely under such conditions
we attribute the high regiocontrol to the inductive effect of
the ester enhancing the electrophilicity of the proximal imide
carbonyl.
With the two key steps accomplished, the synthesis of
alkaloid 5 from hydroxylactam 11 proved quite straightfor-
ward. Thus, completely stereoselective cyclization of 11 was
effected under standard N-acyliminium ion conditions to give
the complete alkaloid skeleton in the form of lactam (+)-
12. Dehydrogenation using a selenoxide syn-elimination gave
(4) Padwa, A.; Danca, M. D. Org. Lett. 2002, 4, 715.
(5) Ahmad, V. U.; Rahman, A.; Rasheed, T.; Rehman, H. Heterocycles
1987, 26, 1987. See also: Rasheed, T.; Khan, M. N. I.; Zhadi, S. S. A.;
Durrani, S. J. Nat. Prod. 1991, 54, 582. Ahmad, V. U.; Iqbal, S. Nat. Prod.
Lett. 1993, 2, 105.
(6) Chopra, R. N.; Chopra, I. C.; Handa, K. L.; Kapoor, L. D. Indigenous
Drugs of India; U. N. Dhar and Sons Pvt. Ltd.; Calcutta, India, 1958.
(7) Base 11, in either mono- or dilithiated form always gives enantio-
complementary results to base 3. At present our assignment of absolute
stereochemistry for 10 rests on this fact, along with a firmly established
pattern of enantioselectivity for base 3 (see ref 1).
(8) For related examples of regioselective imide reduction, see: Pilli,
R. A.; Russowsky, J. Org. Chem. 1996, 61, 3187. Hsu, R.-T.; Cheng, L.-
M.; Chang, N.-C.; Tai, H.-M. J. Org. Chem. 2002, 67, 5044.
a Reagents: (a) AcOH, reflux, 87%; (b) base 11, THF, -78 °C
then MeO2CCN, 85%, 95 to 98% ee.
536
Org. Lett., Vol. 5, No. 4, 2003

13, which is an intermediate in the Padwa synthesis of 5.
Finally, we used a method for selective lactam reduction
reported by Martin and co-workers to transform 13 into
jamtine 5.9
and Danca.10 That the natural product “jamtine oxide” could
be the opposite N-oxide diastereomer to the one that we have
prepared appears a remote possibility. A more likely
explanation is that the original structural assignment is
incorrect.
On exposure to mCPBA in CH2Cl2 at low temperature 5
In conclusion, we have demonstrated the utility of the
chiral base desymmetrization of imides in the synthesis of a
simple alkaloid. The route delivers the tetracyclic isoquino-
line structure 5 in six steps from commerical materials and
in an overall yield of around 20%. Further applications of
this approach are underway.
was converted into the corresponding N-oxide 14 (Scheme
5).
Scheme 5
Acknowledgment. We thank the Engineering and Physi-
cal Sciences Research Council (EPSRC) for a Fellowship
to C.D.G. We also thank Professor Albert Padwa, Emory
University, for helpful exchanges of information.
Supporting Information Available: Description of the
key chiral base reaction leading to 10, and NMR and [R]D
data for all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
At this stage it became clear that there are substantial
differences between the NMR data for 14 and those reported
for the natural product “jamtine oxide”. That our synthesis
has delivered the structure 5 is beyond doubt, given the
excellent agreement between our data and those of Padwa
OL027447S
1
(10) The H NMR spectra obtained by us for compounds 13, 5, and 14
are virtually identical with those kindly forwarded to us by Prof. Padwa. In
a personal communication Prof. Padwa has also indicated that the structure
of “jamtine” 5 has been secured by X-ray crystallography. A full report by
this group, in which they reach similar conclusions to ours, has been recently
published; see: Padwa, A.; Danca, M. D.; Hardcastle, K. I.; McClure, M.
S. J. Org. Chem. 2003, 68, 929.
(9) Martin, S. F.; Clark, C. W.; Ito, M.; Mortimore, M. J. Am. Chem.
Soc. 1996, 118, 9805.
Org. Lett., Vol. 5, No. 4, 2003
537