It was envisaged that another natural product, 1-tigloyl-
3-acetyl-11-methoxyazadirachtinin8 10, could be prepared via
an analogous approach. Installation of the requisite methyl
acetal at C-11 was followed by selective C-23 acetal
exchange and skeletal rearrangement (Scheme 4). Finally,
oxidation/elimination proceeded smoothly to furnish 10 in
excellent yield for the first time,22 with spectral data identical
to that of the natural compound isolated by Kraus.8 Alterna-
tive strategies to prepare this natural product from either
azadirachtin (6) or 1-tigloyl-3-acetylazadirachtinin (9) proved
unsuccessful.23
Scheme 3. Synthesis of Isovepaol
Efforts were now directed toward the synthesis of 3-de-
sacetylazadirachtin (14).24 Accordingly, our common inter-
mediate (2) was first converted to diol 11 (Schemes 1 and
5). It was then necessary to temporarily protect the more
reactive C-3 hydroxyl group in 11 as the corresponding
acetate, thereby permitting installation of the C-1 tiglate
group. The C-3 acetate could then be cleanly and selectively
removed via methanolysis.
to exploit this rearrangement in the preparation of the natural
product 1-tigloyl-3-acetyl-azadirachtinin 9 (Scheme 4).20
During the synthesis of 3-desacetylazadirachtin (14), it was
necessary to effect a diastereoselective reduction of the C-7
carbonyl group in 12 to provide the axial alcohol required
for the natural product (14). In the case of azadirachtin (6),
which differs only by the presence of a tigloyl ester at C-1,
the inherent reactivity of the system provides a 1:1 dr,
employing cerium trichloride and sodium borohydride in
methanol.14 We were therefore surprised to find that under
identical conditions, ketone 12 underwent selective reduction
to the desired axial alcohol 13 in excellent yield (91%) with
exclusive diastereoselectivity.
Scheme 4. Synthesis of Azadirachtinins 9 and 10
Cleavage of the benzyl ether of 13 proceeded without
incident, and all that remained was the introduction of the
enol ether present in the natural product. Acetal exchange
followed by an oxidation/elimination protocol completed the
synthesis of 3-desacetylazadirachtin (14), which was identical
in all respects to the natural product.24,25
In summary, we have successfully prepared five natural
products isolated from the Indian neem tree from a common
precursor (2).26 The use of a selenoacetal to install the
Although 9 had been prepared previously within our group
by semisynthesis,19 at that time it was not reported as a
natural product. In 1996, Kumar20 reported the isolation of
9 from neem extract, and we therefore examined an improved
synthesis from 3. Treatment of 3 with benzene selenol and
sulfonic acid resin cleanly effected rearrangement of the
azadirachtin skeleton and concomitant acetal exchange. The
resulting epimeric selenides were immediately reacted with
hydrogen peroxide/pyridine and underwent oxidation fol-
lowed by syn elimination within 5 min at 0 °C to give
1-tigloyl-3-acetylazadirachtinin as a 3:1 epimeric mixture at
C-11 (9). While we are confident in the assignment of
synthetic material prepared, the NMR data differs substan-
tially from that reported previously.20 We can only therefore
conclude that the natural product these authors isolated was
not 1-tigloyl-3-acetylazadirachtinin but a closely related
compound whose identity is currently unknown.21
(21) Diastereomerically pure 9 underwent immediate epimerization when
dissolved in buffered CDCl3, which was not observed for the natural product.
This, combined with the significant difference between NMR spectra, has
led us to conclude that the two compounds are different.
(22) The synthesis of this compound has been reported in error, and we
would like to thank the reviewer for pointing out this misprint; see: Ley,
S. V.; Denholm, A. A.; Wood, A. Nat. Prod. Rep. 1993, 10, 109-157. In
this reference, the compounds 52/53 depicted on p 117 should be the
corresponding C-11 hemiacetals.
(23) Although selective C-11 methylation of azadirachtin (6) can be
achieved, attempts to then induce skeletal rearrangement led to decomposi-
tion. In the case of 1-tigloyl-3-acetylazadirachtinin (9), it was not possible
to effect C-11 methylation.
(24) Kraus, W.; Bokel, M.; Schwinger, M.; Vogler, B.; Soellner, R.;
Wendisch, D.; Steffens, R.; Wachendorff, U. Phytochemistry and Agricul-
ture; Oxford University Press: Oxford, 1993.
(25) It is worthy of note that 14 has been prepared previously from
azadirachtin; see: Butterworth, J. H.; Percy, G. R.; Morgan, E. D. J. Chem.
Soc., Perkin Trans. 1972, 1, 2445-2450. Yamasaki, R. B.; Klocke, J. A.
J. Agric. Food Chem. 1987, 35, 467-471.
(26) We are confident that these compounds are correctly assigned as
natural products and are not artifacts of isolation since treatment of
azadirachtin with methanol under the isolation conditions (see ref 8) gave
no trace of either 3 or 7. Similarly, 9 showed no conversion to 10 under
the isolation conditions; see ref 8.
(20) Kumar, C. S. S. R.; Srinivas, M.; Yakkundi, S. Phytochemistry 1996,
43, 451-455.
Org. Lett., Vol. 10, No. 4, 2008
571