An efficient, asymmetric synthesis of (+)-euphococcinine

Article abstract of DOI:10.1016/S0040-4039(00)00665-1

The enantioselective synthesis of (+)-euphococcinine (1), a homotropane alkaloid, has been achieved in five steps from bicyclic lactam 5. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00665-1

Tetrahedron Letters 41 (2000) 4339±4342
An ecient, asymmetric synthesis of (+)-euphococcininey
Mark F. Mechelke and A.I. Meyers*
Department of Chemistry, Colorado State University, Fort Collins, CO 80523, USA
Received 13 March 2000; accepted 18 April 2000
Abstract
The enantioselective synthesis of (+)-euphococcinine (1), a homotropane alkaloid, has been achieved in
®ve steps from bicyclic lactam 5. # 2000 Elsevier Science Ltd. All rights reserved.
Euphococcinine (1) was ®rst isolated from an Australian coastal plant, Euphorbia atoto, in
1967.1 Since then, it has also been found in the animal kingdom, speci®cally in the chemical
defense systems of both ladybugs2 and Mexican bean beetles.3 Along with its pentyl analog,
(^)-adaline, which was also isolated from ladybug secretions,2,4 (+)-euphococcinine has been
shown to be a potent feeding deterrent to both spiders and ants.
Although not particularly interesting from a biological standpoint, euphococcinine's unique
bicyclic structure has drawn considerable synthetic interest. A variety of racemic syntheses of
adaline and euphococcinine have been published in the past 40 years and most recently, two
enantioselective syntheses of euphoccinine have been reported by Husson in 19925a and by Murahashi
this year.5b Due to our previous asymmetric routes to cis-2,6-disubstituted piperidines from chiral
bicyclic thiolactams6 and our continuing e€ort in expanding the versatility of this chiral molecule,
we anticipated that (+)-euphococcinine could be prepared in an asymmetric manner from an
appropriately substituted 5,6-bicyclic lactam.
As shown retrosynthetically in Scheme 1, the preparation of (+)-euphococcinine (1) can be
envisioned through an intramolecular Mannich reaction between the enol form of ketone 3 and
the latent iminium ion 2 generated upon opening of the oxazolidine ring system. Ketone 3 may be
accessed from bicyclic thiolactam 4 through an Eschenmoser sul®de contraction7 using bromo-
acetone as the electrophile followed by a diastereoselective reduction of the resultant a,b-unsatu-
rated ketone.
In preliminary studies toward the preparation of ketone 3, we employed the gem-dimethyl-
phenylglycinol derived bicyclic lactam 6.8 Treatment of the latter with Belleau's reagent9 provided
* Corresponding author.
y
This paper is dedicated to Dr. Gurnes Jones for his long list of contributions to heterocyclic chemistry and on the
occasion of his 70th birthday.
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00665-1

4340
Scheme 1.
thiolactam 7 in quantitative yield. It was planned to convert thiolactam 7 into the vinylogous
amide 11 in a single step via a sul®de contraction with bromoacetone. A similar transformation
using methyl bromoacetate was previously applied in our group to prepare the corresponding
vinylogous urethane 10 in an 83% yield.10 The addition of freshly prepared bromoacetone11 to
thiolactam 7, however, resulted in the unexpected formation of thiophene 12, instead of the
desired vinylogous amide, 11 (Scheme 2). Similar thiophene products have been previously
reported in sul®de contractions using a-bromoketones as substrates and are known to be a result
of proton abstraction from the intermediate thioiminium salt (i.e. 9).7
Scheme 2.
An alternative route considered to prepare 3 involved a sul®de contraction with the a-bromo
Weinreb amide 14. The latter was prepared from commercially available bromoacetyl bromide
and N,O-dimethylhydroxylamine hydrochloride (13) in moderate yield (Scheme 3).12 Further-
more, the chiral auxiliary in Scheme 3 was changed from a gem-dimethylphenylglycinol-derived
bicyclic lactam (6) to a cis-1-amino-2-indanol derived lactam (5).13 Both of these chiral bicyclic
lactams (5 and 6) were carried through the same sequence, but as will be seen (vide supra), the
indanol derived lactam 5 gave the more satisfactory stereochemical and synthetic results.
Thiolactam 4, prepared in quantitative yield from lactam 5 via Belleau's reagent,9 was treated
with freshly prepared bromide 14 to a€ord the desired intermediate thioiminium salt, which, on

4341
Scheme 3.
heating to re¯ux in triethylamine and trimethylphosphite, provided the sul®de contraction product
15 in 69% yield. Preliminary attempts at catalytic hydrogenation, using the conditions optimized
for reduction of the corresponding methyl ester 10 (1 atm H2, Pt/C), led only to starting material.
The unsaturated amide 15, required 60 psi of hydrogen (approximately 4 atm) to produce a high
yield of the saturated amide 16. Once obtained, amide 16 was treated with MeLi at ^78ꢀC to
a€ord ketone 3 in 82% yield over the two steps. At this point, the ®rst major di€erence between
the two chiral auxiliaries, (lactams 5 and 6), was observed.14 In the case of the gem-dimethyl-
phenylglycinol bicyclic lactam 6, a 7:1 mixture of exo:endo diastereomers was observed after
hydrogenation, whereas hydrogenation of lactam 5, gave a single diastereomer of 16.15 The poor
diastereoselectivity observed with lactam 6 may be attributed to an increase in steric bulk on the
endo face arising from the gem-dimethyl substituent.
To complete the synthesis of (+)-euphococcinine (1), an intramolecular Mannich reaction of 3
to 17 was required. A variety of acidic conditions were employed in an attempt to open the
bicyclic ring system, thus exposing the latent iminium ion 2, but all of these experiments met with
failure, giving only recovered starting material or an unidenti®able mixture of products. However,
the Mannich cyclization proceeded smoothly when amphoteric conditions were utilized. Ketone 3
was dissolved in 1:1 acetic acid:ethanol, treated with 10 equivalents of ammonium acetate, placed
in a 75ꢀC oil bath, and stirred overnight. After work-up (concentration followed by a Na2CO3(aq)/
CH2Cl2 extraction) and puri®cation by column chromatography (9:1 CH2Cl2:MeOH) the reaction
mixture provided euphococcinine in 91% yield. Column chromatography (9:1 CH2Cl2:MeOH),
was necessary to a€ord (+)-euphococcinine (1). As shown in Scheme 4, this procedure not only
transformed 3 into the natural product's bicyclic framework, but also resulted in concomitant
loss of the chiral auxiliary to a€ord euphococcinine in a single step.
In conclusion, the synthesis of (+)-euphococcinine (1) was achieved in ®ve steps, in 51%
overall yield, from bicyclic lactam 5. Spectral analysis (1H and 13C NMR, IR, and MS) of the
synthetic sample was identical in all respects to the natural product.5 The [ꢀ]D optical rotation
of our sample averaged +5.7 (¯uctuating between +3.8 and +6.6), yet was well within range of the
literature value, [ꢀ]D +6.0.2 Finally, it is interesting to note that when the synthetic sample of (+)-
euphococcinine was treated with (R)-Mosher's acid chloride, no reaction was observed, however
when it was treated with (S)-Mosher's acid chloride, euphococcinine was completely converted
to a single Mosher amide. This observation not only proves that the present synthetic sample is

4342
Scheme 4.
of high enantiomeric purity but also demonstrates that by simply treating synthetic racemic
euphococcinine with (R)- or (S)-Mosher's acid chloride, kinetic resolution may potentially be
achieved to access the single enantiomer.
Acknowledgements
The authors thank Merck for a post-doctoral fellowship (to M.F.M.) and the National Institutes
of Health for ®nancial support. We also thank Prof. Frank R. Stermitz for providing a natural
sample of (+)-euphococcinine.
References
1. Hart, N. K.; Johns, S. R.; Lamberton, J. A. Aust. J. Chem. 1967, 20, 561.
2. Brown, W. V.; Moore, B. P. Aust. J. Chem. 1982, 35, 1255.
3. Eisner, T.; Goetz, M.; Aneshausley, D.; Ferstandig-Arnold, G.; Meinwald, J. Experienta 1986, 42, 204.
4. Tursch, B.; Braekman, J.-C.; Daloze, D.; Hootele, C.; Losman, D.; Karlsson, R.; Pasteels, J.-M. Tetrahedron Lett.
1973, 201.
5. (a) Yue, C.; Royer, J.; Husson, H.-P. J. Org. Chem. 1992, 57, 4211, and references cited therein. (b) Murahashi, S.-
I.; Sun, J.; Kurosawa, H.; Imada, Y. Heterocycles 2000, 52, 557.
6. Munchhof, M. J.; Meyers, A. I. J. Am. Chem. Soc. 1995, 117, 5399.
7. Roth, M.; Dubs, P.; Gotschi, E.; Eschenmoser, A. Helv. Chim. Acta 1971, 54, 710.
8. Lemieux, R. M.; Devine, P. N.; Mechelke, M. F.; Meyers, A. I. J. Org. Chem. 1999, 64, 3585.
9. Lajoie, G.; Lepine, F.; Maziak, L.; Belleau, B. Tetrahedron Lett. 1983, 24, 3815.
10. Mechelke, M. F.; Meyers, A. I., unpublished results.
11. Levene, P. A. Org. Synth., Coll. Vol. 1947, 2, 88.
12. Nuzillard, J.-M.; Boumendjel, A.; Massiot, G. Tetrahedron Lett. 1989, 30, 3779.
13. Senanayake, C. H.; Roberts, F. E.; DiMichele, L. M.; Ryan, K. M.; Liu, J.; Fredenburgh, L. E.; Foster, B. S.;
Douglas, A. W.; Larsen, R. D.; Verhoeven, T. R.; Reider, P. J. Tetrahedron Lett. 1995, 36, 3993.
14. Another di€erence in lactams 5 and 6 was observed in the ®nal cyclization step (3 to 1) where a 91% yield was
achieved with 3 and only a 61% yield was obtained with the ketone derived from 6.
15. The exo stereochemistry was initially assigned based on results found in Ref. 6 and eventually proven by the
synthesis of (+)-euphococcinine.