First Total Synthesis of the Marine Alkaloids (±)-Fasicularin and (±)-Lepadiformine Based on Stereocontrolled Intramolecular Acylnitroso-Diels−Alder Reaction

Abstract

The first total synthesis of tricyclic marine alkaloids (()-fasicularin (2) and (()-lepadiformine (5) was accomplished. The key common strategic element for the synthesis is the stereocontrolled intramolecular hetero-Diels-Alder reaction of an N-acylnitroso moiety to an exocyclic diene with or without bromine substitution to control the syn-facial or anti-facial selectivity, respectively, leading to the trans-or cis-fused decahydroquinoline ring systems 21 or 39 involving the simultaneous introduction of the nitrogenated quaternary center in a single step. On further elaboration of the six-membered or five-membered ring A, the trans-fused adduct 21 provided either (()-fasicularin (2) or (()-lepadiformine ( 5). The hydrochloride salt of synthetic (()-5 was found to be identical with the isolated natural sample of lepadiformine; however, the tricyclic amino alcohol 4 having the proposed structure of lepadiformine in a nonzwitterionic form, derived from the cis-fused adduct 39, was found to be different from lepadiformine by spectral comparison. These results thus unambiguously established the relative stereochemistry of lepadiformine, formerly assigned incorrectly, to be 3R*,5S*,7aR*,11aR* shown by 5.

Development of New Synthetic Methods and Its Application to Total Synthesis of Nitrogen-Containing Bioactive Natural Products

“…Formation of fasicularin was thus attained in remarkably high combined yield of 90% from the tricyclic amino alcohol 88. Fasicularin so obtained, having spectral properties in agreement with those previously reported, 60,63) proved to be enantiomerically pure by chiral HPLC analysis (Dicel Chiralpak AD column) in comparison with (Ϯ)-57 previously obtained by us, 63) and was found to have [a] D 21 Ϫ4.4 (MeOH). The present fasicularin formation can be rationalized by considering the initial formation of the aziridinium ion 89 that undergoes nucleophilic attack of thiocyanate ion with subsequent expansion reaction of the aziridine.…”

“…The resultant amino alcohol 81 was subjected to cyclodehydration using CBr 4 and PPh 3 13) to give the tricyclic amine 82 with complete inversion of the configuration at C-3Ј. Hydrogenolytic removal of the benzyl protecting group of 82 furnished lepadiformine (59) whose spectral properties were identical in all respects with those of an authentic sample of racemic lepadiformine (Ϯ)-59 previously prepared 63) by us. The optical rotation of synthetic alkaloid 59 was measured: [a] D 28 Ϫ15.0°(cϭ0.37, MeOH) for the free base (oil) and [a] D 26 ϩ2.6°(cϭ0.54, CHCl 3 ) for the hydrochloride salt (colorless gum).…”

“…We have achieved the first total synthesis of racemic fasicularin and lepadiformine, the latter of which led to a revision of the proposed structure 58 of lepadiformine to 59. 62,63) We also have achieved the enantioselective synthesis of (Ϫ)-lepadiformine 64,65) that allowed us to assign the 3S,5R,7aS,11aS configuration for the natural product. 66,67) …”

“…The optical rotation of synthetic alkaloid 59 was measured: [a] D 28 Ϫ15.0°(cϭ0.37, MeOH) for the free base (oil) and [a] D 26 ϩ2.6°(cϭ0.54, CHCl 3 ) for the hydrochloride salt (colorless gum). Although comparison of the optical rotation of our synthetic sample with that of the natural product was impossible since the rotation of the natural product (actually the hydrochloride salt 63) ) had been reported [49][50][51][52][53][54] to be zero, synthetic (Ϫ)-59 proved to be identical with natural lepadiformine, kindly provided by Professor Biard, on the basis of their chromatographic behavior on the HPLC chiral phase. 64) This allowed the absolute configuration of the natural product to be assigned as 3R,5S,7aR,11aR.…”

“…63) After this report, the second synthesis of (Ϯ)-57 using a 2-amidoacrolein Diels-Alder cycloaddition has been published by Funk and Maeng, 93) and more recently the formal construction of 57 starting from (S)-5-hydroxy-2-piperidone has been reported by the Dake group. 94) However, these syntheses 63,93) suffered from very poor overall yields (0.9% and 2.4%), mainly due to difficulty in incorporation of the thiocyanato group in the final step which was performed by a Mitsunobu procedure (HSCN, Ph 3 P, DEAD) that we have exploited previously 63) or an SN2 displacement of the mesylate by Bu 4 NSCN 93) resulting in very low yield (20%) of fasicularin in each case. Thus, there is a great need to develop a new thiocyanation method that can overcome such problem.…”