First total synthesis of acerogenin C and aceroside IV

Full text of DOI:10.1021/jo9714324

7544
J . Org. Chem. 1997, 62, 7544-7545
Sch em e 1
F ir st Tota l Syn th esis of Acer ogen in C a n d
Acer osid e IV
Gabriela Islas Gonzalez and J ieping Zhu*
Institut de Chimie des Substances Naturelles, CNRS,
91198 Gif-sur-Yvette, France
Received August 4, 1997
The stem bark of Acer nikoense Maxim (Aceraceae), a
tree indigenous to J apan, has been used in folk medicine
as a remedy for hepatic disorder and for eyewash.1 Since
the isolation of acerogenin A (1a ) by Nagai et al. in 1976,2
several dozens of structurally related products with an
endocyclic biaryl ether bond have been identified from
the same plant.3 They all belong to a growing family of
metabolites called diarylheptanoids, which are character-
ized by the presence of two aromatic rings connected by
an oxygenated seven-carbon aliphatic chain.4 No total
synthesis5 of acerogenin-type compounds has yet ap-
peared in the literature probably due to the lack of an
efficient ring closure reaction. Previously, an intramo-
lecular Wurtz reaction6 and Wittig reaction7 have been
employed by Nogradi et al.6,7 as the key cyclization step
for the total synthesis of related natural products garu-
gamblin 1 and garuganin III.4
Sch em e 2a
The intramolecular SNAr reaction developed recently
in this laboratory has proved to be an efficient methodol-
ogy for the construction of polypeptide macrocycles with
endo aryl-aryl8 and aryl-alkyl ether9 bond(s). We have
attributed the success of this remarkable cycloetherifi-
cation to an intramolecular recognition phenomenon.8-11
Several structural elements found in our previously
studied substrates could indeed help their preorganiza-
tion12 in such a way that a folded conformation was a
predominant low energy one, thus favoring the desired
cyclization.13 To evaluate the influence of intramolecular
H-bonding14 on the outcome of cyclization and to further
a
i
Key: (a) PrBr, K2CO3, DMF; (b) LAH-THF, 95%; (c) TsCl,
Py; (d) NaI, Me2CO, 87%; (e) methyl acetoacetate, LDA, then iodide
9, 80%; (f) NaH, THF, then 4-fluoro-3-nitrobenzyl iodide (6), 75%;
(g) BCl3, CH2Cl2; (h) 6 N HCl, 90%.
expand the generality of this methodology, we were
interested in investigating the cyclization of a linear
compound wherein the two reactive sites are linked by
an aliphatic hydrocarbon chain (e.g., compound 3). The
acerogenin-type natural diarylheptanoids seemed to be
appropriate synthetic targets for this purpose. The
successful implementation of this strategy as exemplified
by the first total synthesis of acerogenin C (2a ) and
aceroside IV (2b)3c is the subject of the present paper.
(1) Nagumo, S.; Kaji, N.; Inoue, T.; Nagai, M. Chem. Pharm. Bull.
1993, 41, 1255-1257.
(2) Nagai, M.; Kubo, M.; Fujita, M.; Inoue, T.; Matsuo, M. J . Chem.
Soc., Chem. Commun. 1976, 338-339.
(3) (a) Nagai, M.; Kubo, M.; Fujita, M.; Inoue, T.; Matsuo, M. Chem.
Pharm. Bull. 1978, 26, 2805-2810. (b) Kubo, M.; Inoue, T.; Nagai, M.
Chem. Pharm. Bull. 1980, 28, 1300-1303. (c) Kubo, M.; Nagai, M.;
Inoue, T. Chem. Pharm. Bull. 1983, 31, 1917-1922. (d) Nagai, M.;
Kubo, M.; Takahashi, K.; Fujita, M.; Inoue, T. Chem. Pharm. Bull.
1983, 31, 1923-1928. (e) Nagumo, S.; Ishizawa, S.; Nagai, M.; Inoue,
T. Chem. Pharm. Bull. 1996, 44, 1086-1089.
(4) Keseru¨, G. M.; Nogradi, M. The Chemistry of Natural Diaryl-
heptanoids. In Studies in Natural Products Chemistry; Atta-ur-
Rahman, Ed.; Elsevier Science B.V.: New York, 1995; Vol. 17, pp 357-
394.
(5) The oxidative coupling of phenol promoted by thallium tris-
(trifluoroacetate) has been reported to give macrocyclic diaryl ethers;
see: Whiting, D. A.; Wood, A. F. J . Chem. Soc., Perkin Trans. 1 1980,
623-628.
As shown in the retro-synthetic scheme (Scheme 1),
our plan calls for the macrocyclization of substrate 3 as
a key ring closure step via an intramolecular SNAr
reaction. While many different strategies may be envis-
aged for the synthesis of type 3 linear diarylheptanoids,
a convergent and flexible approach via a “3 + 3 + 1”
strategy taking advantage of the dianion chemistry of
methyl acetoacetate15 was attempted in this study.
The linear cyclization precursor 3 was synthesized
without event following standard procedures (Scheme 2).
Cyclization of 3 occurred smoothly in DMF (0.01 M, room
(6) Vermes, B.; Keseru¨, G. M.; Mezey-Vandor, G.; Nogradi, M.; Toth,
G. Tetrahedron 1993, 49, 4893-4900.
(7) (a) Keseru¨, G. M.; Dienes, Z.; Nogradi, M.; Kajtar-Peredy, M. J .
Org. Chem. 1993, 58, 6725-6728. (b) Keseru¨, G. M.; Nogradi, M.;
Kajtar-Peredy, M. Liebigs Ann. Chem. 1994, 361-364.
(8) (a) Zhu, J . Synlett 1997, 133-144. (b) Burgess, K.; Lim, D.;
Martinez, C. I. Angew. Chem., Int. Ed. Engl. 1996, 35, 1077-1078.
(9) Zhu, J .; La¨ıb, T.; Chastanet, J .; Beugelmans, R. Angew. Chem.,
Int. Ed. Engl. 1996, 35, 2517-2519.
(10) Beugelmans, R.; Bourdet, S.; Zhu, J . Tetrahedron Lett. 1995,
36, 1279-1282.
(11) Abramovitch, R. A.; Shi, Q. Heterocycles 1994, 38, 2147-2151.
(12) Cram, D. J . Angew. Chem., Int. Ed. Engl. 1986, 25, 1039-1057.
(13) (a) Menger, F. M. Acc. Chem. Res. 1985, 18, 128-134. (b)
Mandolini, L. Bull. Soc. Chim. Fr. 1988, 173-176.
(14) For an example of an intramolecular hydrogen-bond-directed
macrocyclization, see: Carver, F. J .; Hunter, C. A.; Shannon, R. J . J .
Chem. Soc., Chem. Commun. 1994, 1277-1280.
(15) Huckin, S.; Weiler, L. J . Am. Chem. Soc. 1974, 96, 1082-1087.
(16) Bois-Choussy, M.; Neuville, L.; Beugelmans, R.; Zhu, J . J . Org.
Chem. 1996, 61, 9309-9322.
S0022-3263(97)01432-1 CCC: $14.00 © 1997 American Chemical Society

Communications
J . Org. Chem., Vol. 62, No. 22, 1997 7545
Sch em e 3
Sch em e 4a
temperature, 3 h) using cesium fluoride16 as base, provid-
ing the desired macrocycle 12 in almost quantitative yield
1
(based on the H NMR spectrum of the crude product,
Scheme 3). The cyclic structure of 12 was evident from
the characteristic high-field shift of the H-6 proton in the
1H NMR spectrum, due to the anisotropic effect of the
aromatic B ring. Potassium carbonate17 was also found
to be effective in promoting this cyclization, though a
relatively longer time (10 h) was required. The cycloet-
herification rate could be accelerated by heating the
reaction mixture to 60 °C without diminishing the yield
of cyclized product. To examine the effect of concentra-
tion on the outcome of the cyclization (Scheme 3),
compound 3 was treated with CsF in DMF at 0.01, 0.05,
0.1, and 1 M. We observed that even at high concentration
(1 M of 3 in DMF), the analytically pure macrocycle 12
could still be obtained in 45-50% isolated yield.18 This
provides experimental evidence that the intramolecular
reaction of compound 3 is indeed facile and highly
competitive with the alternative intermolecular process.
This result is quite instructive considering the fact that
the aliphatic carbon chain usually adopts an extended
conformation.19 We reasoned that the presence of electron-
rich and electron-poor aromatic rings at the two chain
terminals alters the conformational properties of 3 lead-
ing to a folded conformer conducive to cyclization.20-22
We anticipate that, in the future, the proper use of the
principle of intramolecular recognition phenomena8-11,14
will help to design new macrocyclization techniques.
The syntheses of acerogenin C (2a ) and aceroside IV
(2b) were accomplished as shown in Scheme 4. Reduc-
tion of the nitro group of 12 was carried out by hydro-
genolysis under conventional conditions (Pd/C, MeOH,
1 atm) to afford the amino compound 13, which was
converted into compound 14 (90% overall yield) employ-
a
t
Key: (a) H2, Pd/C, MeOH; (b) BuONO, DMF, 90%; (c) AlCl3,
CH2Cl2, 88%; (d) 2,3,4,6-tetrabenzoylglucopyranosyl bromide,
CH2Cl2-NaOH, Bu4NBr, 85%; (e) MeOH-H2O, NaOH, 92%.
ing Doyle’s one-step deamination procedure.23 The O-
demethylation of 14 with AlCl3 in refluxing CH2Cl2 gave
acerogenin C (2a ), whose spectral data were identical in
all respects to the literature values.3c
Glycosidation of 2a was first attempted with 2,3,4,6-
R-D-tetrabenzylglucopyranose under Mitsunobu condi-
tions.24 While the reaction proceeded cleanly to give the
desired aryl glucoside, the conversion was low (<30%)
even under forced conditions (e.g., excess of glycosyl
donor, heating to 60 °C in different solvents). The desired
transformation was finally realized using a two-phase
glycosidation methodology.25 Thus, 2,3,4,6-R-D-tetraben-
zoylglucopyranosyl bromide was allowed to react with
acerogenin C (2a ) in the presence of tetrabutylammonium
bromide (CH2Cl2-aqueous NaOH) to afford stereospe-
cifically the â-aryl glucoside 15 in 85% yield. Subsequent
saponification (MeOH-H2O, NaOH) of the benzoyl groups
gave the aceroside IV (2b) in 92% isolated yield. The
spectral data of our synthetic material were identical
with those of the natural product.
In summary, we have accomplished the first total
synthesis of acerogenin C and aceroside IV26 featuring a
high-yielding cycloetherification reaction as the key ring-
closure step. The synthesis is convergent and flexible.
It should be easily amenable to other members of this
family and should allow for the generation of libraries of
cyclic diarylheptanoids for biological investigations.
(17) Zhu, J .; Beugelmans, R.; Bourdet, S.; Chastanet, J .; Roussi, G.
J . Org. Chem. 1995, 60, 6389-6396.
(18) A mixture of cyclic and noncyclic dimers was also isolated, which
accounts for the total mass balance of the reaction.
(19) Go¨ttlich, R.; Kahrs, B. C.; Kru¨ger, J .; Hoffmann, R. W. J . Chem.
Soc., Chem. Commun. 1997, 247-251 and references therein.
(20) Molecular modeling [MM2 (Allinger, N. L. J . Am. Chem. Soc.
1977, 99, 8127-8134), water set)] of compound 3 showed that the
lowest energy conformation was indeed the folded one. Similar
computational studies have been carried out by Abramovitch’s group;
see ref 11.
Ack n ow led gm en t. We thank Professor Nagai for
kindly sending us the NMR spectra of natural acero-
genin C and aceroside IV. Financial support from the
CNRS and a SFERE-CONACYT doctoral fellowship
funded jointly by the Mexican and French governments
to G. Islas Gonzalez are gratefully acknowledged.
Su p p or tin g In for m a tion Ava ila ble: Experimental pro-
cedure and complete characterization of natural products
acerogenin C (2a ) and aceroside IV (2b) as well as the key
synthetic intermediates 3 and 10-15 (7 pages).
(21) At the present time, we have failed to observe a CT absorption
band in the UV-vis spectrum of compound 14 (1 × 10-4 M, MeCN).
However, further studies are required before drawing any conclusion.
Detailed NMR studies are in progress.
(22) As suggested by one of the reviewers, it would be interesting
to check the outcome of an alternative synthetic strategy, i.e., formation
of the biaryl ether followed by macrocyclization via intramolecular
alkylation of the keto ester. The cyclization results will provide a direct
evidence of our hypothesis as intramolecular recognition phenomena
will be absent in this case. This study will be incorporated in a future
full account.
J O9714324
(24) Roush, W. R.; Lin, X. F. J . Org. Chem. 1991, 56, 5740-5742.
(25) Dess, D.; Kleine, H. P.; Weinberg, D. V.; Kaufman, R. J .; Sidhu,
R. S. Synthesis 1981, 883-885.
(26) We have been informed that Professor Nogradi has also
accomplished, very recently, a total synthesis of acerogenin C.
(23) Doyle, M. P.; Dellaria, J . F.; Siegfried, J r. B.; Bishop, S. W. J .
Org. Chem. 1977, 42, 3494-3498.