Elsevier

Tetrahedron

Volume 61, Issue 5, 31 January 2005, Pages 1187-1198
Tetrahedron

Enantioselective syntheses of two 5, 9E diastereomers of 223V, an alkaloid from the poison frog Dendrobates pumilio

https://doi.org/10.1016/j.tet.2004.11.060Get rights and content

Abstract

Enantioselective syntheses of two 5, 9E diastereomers (1 and 2) of 223V (3) are described. Neither corresponded on GC–MS and GC-FTIR analyses to alkaloid 223I, previously tentatively proposed to be a 5,8-disubstituted indolizidine of the unusual 5, 9E relative stereochemistry. Synthetic (−)-(5, 9Z)-5-n-propyl-8-n-butylindolizidine (3) corresponds on GC–MS and GC-FTIR analyses to the natural indolizidine 223V found in a pumilio from ‘Split Hill’, Panama.

Introduction

5,8-Disubstituted indolizidines represent a major class of alkaloids found in skins of poison frogs.1 Over 60 such alkaloids have been proposed. Some structures are tentative, being based only on their GC–MS, dominated by a base peak due to α-cleavage of the 5-substituent, followed by a retro-Diels–Alder loss to yield a characteristic fragment at m/z 96.2 Several of the 5,8-disubstituted indolizidines have been isolated from frog skin in sufficient quantities to allow structure confirmation by NMR spectral analysis. These include (−)-203A,13 (−)-205A,14 (−)-207A,15 233D,13 (−)-235B and (+)-235B (formerly 235B).14, 15 Structures are shown in Figure 1. The structures of (−)-207A, (−)-235B, and (+)-235B have been confirmed by enantioselective synthesis.3, 16, 17 The relative stereochemistry of 205A is depicted on the basis of comparison with synthetic racemic material.18 The structure and absolute stereochemistry of natural 209I15 was confirmed (unpublished results) by comparison to synthetic racemic material,19 and the synthetic (−)-unnatural enantiomer.20 Several laboratories have reported syntheses of (−)-209B.18, 21, 22 Virtually all alkaloids of this class possess a 5, 9Z structure as shown by a characteristic sharp and intense Bohlmann band near 2790 cm−1 in their GC-FTIR spectra.2 Only two alkaloids have been tentatively proposed to be (5, 9E)-5,8-disubstituted indolizidines, based on GC–MS and a weak absorbance in the Bohlmann band region on GC-IR.1 One of these is alkaloid 259B from one population of Dendrobates pumilio with an EI-MS showing the α-cleavage expected of a 5-C9H13-8-CH3-indolizidine, followed by retro-Diels–Alder cleavage of the fragment at m/z 138 to yield a significant diagnostic ion at m/z 96. The second was alkaloid 223I from another population of the poison frog D. pumilio, tentatively proposed to have a (5, 9) E-5-propyl-8-butylindolizidine structure even though the diagnostic peak in EI-MS at m/z 96 was much weaker than expected.

In this paper, we would like to report the enantioselective syntheses of two 8-epimers of (5, 9E) 5-propyl-8-butylindolizidine (1, 2) and comparison to alkaloid 223I. In addition, a previously synthesized (−)-(5, 9Z) 5-propyl-8-butylindolizidine3 (3) has now been shown to be identical in GC–MS and GC-FTIR to alkaloid 223V from yet another population of the same poison frog, D. pumilio from ‘Spilit Hill’, Panama.

Section snippets

Results and discussion

The stereoselective synthesis of 3 has been described.3 The synthesis of 1 began with the enaminoester 4,4 which was treated with lithium dibutylcuprate to afford the adduct 5 as a single isomer.5 The stereoselectivity of this addition reaction can be explained by the stereoelectronic effect6 and Cieplak's hypothesis7 as shown below (Scheme 1).

The carbon chain at the α-position of 5 was elongated by two Arndt–Eistert reactions to provide the two-carbon homologated ester 7, which was converted

Conclusion

The synthesis and properties of indolizidine 3 have been reported.3 It proved on GC–MS {223 (M+, 1), 222 (1), 180 (100), 166 (1), 138 (1), 136 (1), 126 (1), 124 (2), 110 (2), 108 (1), 96 (12), 70 (9), 55 (4)} and GC-FTIR (2968, 2938, 2880, 2787, 1459, 1377, 1133 cm−1) analysis to be identical to natural indolizidine 223V.

Synthesis of the 5, 9E-indolizidines 1 and 2 provided clear proof that alkaloid 223I was not a 5, 9E-indolizidine. Both 1 and 2 had an appreciable retro-Diels–Alder fragment at m

General

Melting points were determined with a Yanaco micro melting point apparatus and are uncorrected. 1H and 13C NMR spectra were taken on a Varian Gemini 300 or Unity Plus 500 spectrometer. 1H NMR spectra were recorded at the indicated field strength as solutions in CDCl3 unless otherwise indicated. Chemical shifts are given in parts per million (ppm, δ) downfield from TMS and are referenced to CHCl3 (7.26 ppm) as an internal standard. Splitting patterns are designated as s, singlet; d, doublet; t,

Acknowledgements

This work was supported in part by The Research Foundation for Pharmaceutical Sciences.

References and notes (22)

  • J.W. Daly et al.

    Alkaloids from Amphibian Skin

  • T. Tokuyama et al.

    Tetrahedron

    (1991)
  • T. Tokuyama et al.

    Tetrahedron

    (1987)
  • N. Toyooka et al.

    Tetrahedron

    (1997)
  • Y. Song et al.

    Tetrahedron Lett.

    (2002)
  • N. Toyooka et al.

    J. Org. Chem.

    (2002)
  • P. Deslougchamps

    Stereoelectronic Effects in Organic Chemisry

    (1983)
  • T. Okita et al.

    Synlett

    (1994)
    T. Okita et al.

    Tetrahedron

    (1995)
    C.J. Foti et al.

    J. Org. Chem.

    (1995)
    K.C. Nicolaou et al.

    Chem. Commun.

    (1998)
  • H.M. Garraffo et al.

    Rapid Commun. Mass Spectrom.

    (1999)
  • M.W. Edwards et al.

    J. Nat. Prod.

    (1988)
  • T. Momose et al.

    J. Org. Chem.

    (1994)
  • Cited by (22)

    • Simple indolizidine and quinolizidine alkaloids

      2016, Alkaloids: Chemistry and Biology
      Citation Excerpt :

      Most of the published research relates to the total synthesis of this small subset of compounds, and is described below. It was previously reported in this series of volumes1 that the postulated structure rel-2217 for quinolizidine 207I, obtained from the Madagascan frog M. baroni, had been disproved by synthesis, and that the 1,4-cis-disubstituted structure 1728 had been suggested instead.835 Michel and coworkers subsequently undertook a synthesis of rac-1728 in order to validate the proposal (Scheme 282).842,891

    • Enantiodivergent synthesis of the quinolizidine poison frog alkaloid 195C

      2013, Tetrahedron
      Citation Excerpt :

      In the conformer A, hydrogenation would occur from the less hindered β-face to give rise to (−)-195C as a major product. On the other hand, in the confomer B, the propenyl or n-propyl substituent on the iminium moiety would occupy the upper face due to the steric repulsion with methyl group at the 4-position, and hydrogenation would occur from the α-face to afford 14 as the major product (Fig. 2).12 This study represents the first chiral synthesis of the quinolizidine alkaloid 195C.

    View all citing articles on Scopus
    View full text