An Unusually Broad Series of Seven Cyclombandakamines, Bridged Dimeric Naphthylisoquinoline Alkaloids from the Congolese Liana Ancistrocladus ealaensis

Abstract

A series of seven unusual dimeric naphthylisoquinoline alkaloids was isolated from the leaves of the tropical liana Ancistrocladus ealaensis J. Léonard, named cyclombandakamine A (1), 1- epi -cyclombandakamine A (2), and cyclombandakamines A 3–7 (3–7). These alkaloids have a chemically thrilling structural array consisting of a twisted dihydrofuran-cyclohexenone-isochromene system. The 1′″-epimer of 4, cyclombandakamine A 1 (8), had previously been discovered in an unidentified Ancistrocladus species related to A. ealaensis . Both lianas produce the potential parent precursor, mbandakamine A (9), but only A. ealaensis synthesizes the corresponding cyclized form, along with a broad series of slightly modified analogs. The challenging isolation required, besides multi-dimensional chromatography, the use of a pentafluorophenyl stationary phase. Featuring up to six stereocenters and two types of chiral axes, their structures were elucidated by means of 1D and 2D NMR, HRESIMS, in combination with oxidative chemical degradation experiments as well as chiroptical (electronic circular dichroism spectroscopy) and quantum chemical calculations. Compared to the ‘open-chain’ parent compound 9, these dimers displayed rather moderate antiplasmodial activities.

Spirombandakamine A3 and Cyclombandakamines A8 and A9, Polycyclic Naphthylisoquinoline Dimers, with Antiprotozoal Activity, from a Congolese Ancistrocladus Plant

“…The other northwestern molecular half consisted of a tetralone subunit linked to an isoquinoline part by a 5‴,8″-axis and by an oxygen bridge between C-1′′ and C-6′′′ (Figure 4A). NMR data evidencing such connections were identical to those already observed in the other, previously characterized cyclombandakamines, 11,13 among them joint HMBC correlations from H-7″, H-7‴, and H eq -4‴ to C-5″′, which indicated a 5′′′,8′′-coupling, and ROESY interactions between H-7′′ and both diastereotopic protons at C-4′′′ and between H-7′′′ and The connections between the two molecular halves, via the 6′,1″-axis and the C-2′′−O-5′−C-5′ oxygen bridge, to form an entire cyclombandakamine-type scaffold, were evident from HMBC cross-peaks between C-1′′ and both H b -3″ and H-6′ and from the chemical shift of C-2′′ (δ C 92.4) indicating it to be an acetal, together with ROESY interactions between CH 3 -3‴, in the northwestern isoquinoline portion, and H-7′, H-7, OCH 3 -8, and CH 3 -1 in the southeastern naphthylisoquinoline moiety (Figure 4A). Hence, the second isolated dimer, compound 8, was constitutionally different from its isomeric and co-occurring congener 3 only by its N-methyl pattern.…”

“…The second new dimer, compound 8, was obtained as a yellowish amorphous solid, possessing the molecular formula C 49 H 53 N 2 O 8 (HRESIMS: m/z 797.380 95, [M + H] + ), which indicated that it was an isomer of the above-described spirombandakamine A 3 (7) and of the previously known and co-occurring dimers spirombandakamine A 2 (6) and cyclombandakamine A 1 (3). From its UV curve and from the observation of a 13 C NMR signal at δ C 196.2, which was reminiscent of the chemical shift of the carbonyl C atom of cyclombandakamines A 1 (3) and A 2 (4), 11 this metabolite was soon recognized to be a cyclombandakamine-type dimer.…”

“…Its UV spectrum, however, was most similar to those of spirombandakamines, 10 rather than to the ones of cyclombandakamines, 11,13 hence suggesting the compound to be a spirombandakamine-type alkaloid. This assumption was further supported by the observation of a resonance signal at δ C 203.4 in the 13 C NMR spectrum, which is characteristic of the carbonyl carbon atom in spirombandakamines. 10 The 1 H NMR spectrum (Table 1) displayed a full set of signals, including those of seven aromatic protons, three methylene groups with diastereotopic protons, one N-methyl group, four methoxy functions, four methine groups, and six methyl groups.…”

“…It was, thus, isomeric to the co-occurring dimers cyclombandakamine A 1 (3) and spirombandakamine A 2 (6). Its UV spectrum, however, was most similar to those of spirombandakamines, 10 rather than to the ones of cyclombandakamines, 11,13 hence suggesting the compound to be a spirombandakamine-type alkaloid. This assumption was further supported by the observation of a resonance signal at δ C 203.4 in the 13 C NMR spectrum, which is characteristic of the carbonyl carbon atom in spirombandakamines.…”

“…Two of them were identified to be Table 1. 1 H (600 MHz) and 13 C (151 MHz) NMR Data of Spirombandakamine A 3 (7) and Cyclombandakamines A 8 (8) and A 9 (9) in MeOH-d 4 (δ in ppm) mbandakamines C (10) and D (11), which had previously been discovered in another Congolese Ancistrocladus liana, A. ealaensis. 12 The remaining three compounds were as yet unknown; their structural elucidation is described in the following.…”