General and expedient synthesis of 1,4-dioxygenated xanthones

Article abstract of DOI:10.1021/ol201910v

A facile entry to 1,4-dioxygenated xanthones having a variety of substitution patterns and substituents was developed that features a novel application of the Moore cyclization using substrates that were readily assembled in a highly convergent fashion by an acetylide stitching process. The practical utility of the methodology was demonstrated by an efficient synthesis of a naturally occurring xanthone and correction of the structure of dulcisxanthone C.

Full text of DOI:10.1021/ol201910v

ORGANIC
LETTERS
2011
Vol. 13, No. 17
4696–4699
General and Expedient Synthesis of
1,4-Dioxygenated Xanthones
Alexander L. Nichols, Patricia Zhang, and Stephen F. Martin*
Department of Chemistry and Biochemistry, The University of Texas at Austin,
1 University Station-A5300, Austin, Texas 78712, United States
sfmartin@mail.utexas.edu
Received July 14, 2011
ABSTRACT
A facile entry to 1,4-dioxygenated xanthones having a variety of substitution patterns and substituents was developed that features a novel
application of the Moore cyclization using substrates that were readily assembled in a highly convergent fashion by an acetylide stitching
process. The practical utility of the methodology was demonstrated by an efficient synthesis of a naturally occurring xanthone and correction of
the structure of dulcisxanthone C.
The xanthone core 1 is a common motif found in
numerous naturally occurring compounds having biologi-
cal activity (Figure 1).1 An important xanthone subclass
incorporates oxygenation at the 1- and 4-positions as
illustrated in 2. Many 1,4-dioxygenated xanthones elicit
useful biological properties. For example, atroviridin (3),2
which has been synthesized by the groups of Theodorakis3
and Suzuki,4 is used in traditional medicine as a remedy for
earaches. The more complex IB-00208 (4) is a hexacyclic,
angularly fused quinone xanthone that has 1 nM activity
against the P388D1, A-549, HT-29, and SK-MEL-28 cell
lines as well as antimicrobial activity against Staphylococcus
aureus and Bacillus subtilis.5 The tetramethoxy xanthone 5is
reported to inhibit the growth of osteoclasts in vitro.6
Figure 1. 1,4-Dioxygenated xanthone natural products.
Numerous strategies have been developed to prepare
xanthones, but there is a relative paucity of methods that
xanthones.7 These typically suffer from a number of sig-
are readily applicable to the synthesis of 1,4-dioxygenated
nificant limitations, including lengthy sequences, low
(1) (a) Vieira, L. M. M.; Kijjoa, A. Curr. Med. Chem. 2005, 12, 2413–
2446. (b) Pinto, M. M. M.; Sousa, M. E.; Nascimento, M. S. J. Curr.
Med. Chem. 2005, 12, 2517–2538.
(2) Kosin, J.; Ruangrungsi, N.; Ito, C.; Furukawa, H. Phytochem-
istry 1998, 47, 1167–1168.
(3) Tisdale, E. J.; Kochman, D. A.; Theodorakis, E. A. Tetrahedron
Lett. 2003, 44, 3281–3284.
(4) Suzuki, Y.; Fukuta, Y.; Ota, S.; Kamiya, M.; Sato, M. J. Org.
Chem. 2011, 76, 3960–3967.
yields, harsh conditions, restricted substitution patterns,
narrow substrate scope, and the production of isomeric
products.8 Such procedures are not well-suited for the
efficient synthesis of xanthones with diverse substitution
patterns or high levels of structural complexity such as 4.
A robust and reproducible method is needed that will
enable broad access to 1,4-dioxygenated xanthones found
in nature and biologically active compounds.
ꢀ
ꢀ
ꢀ
(5) Malet-Cascon, L.; Romero, F.;Espliego-Vazquez, F.; Gravalos, D.;
ꢀ
Fernandez-Puentes, J. L. J. Antibiot. 2003, 56, 219–225. (b) Rodrıguez,
´
J. C.; Puentes, J. L. F.; Baz, J. P.; Canedo, L. M. J. Antibiot. 2003, 56,
318–321.
~
(6) Zhang, J.; Ahn, M.-J.; Sun, Q. S.; Kim, K.-Y.; Hwang, Y. H.;
Ryu, J. M.; Kim, J. Arch. Pharm. Res. 2008, 31, 850–855.
(7) For a review of methods to prepare xanthones, see: Sousa, M. E.;
Pinto, M. M. M. Curr. Med. Chem. 2005, 12, 2447–2479.
r
10.1021/ol201910v
Published on Web 08/03/2011
2011 American Chemical Society

We recently devised an effective strategy to prepare
quinone natural products using variants of the Moore
cyclization.9,10 Wethereforequeried whetherthe chemistry
we had discovered might be more generally applied to
synthesizing 1,4-dioxygenated xanthones. The fundamen-
tal features of the plan are highlighted in a retrosynthetic
format in Scheme 1. The squarate 9 and a readily available
aryl aldehyde 10 are first joined together by an acetylide
stitching process to give the key intermediate 8, which is
heated to induce a Moore cyclization that produces the
quinone 7. Oxidation of 7 followed by cyclization via 1,4-
addition then provided the desired xanthone 6. We now
report the successful implementation of this new metho-
dology for xanthone synthesis and the application to a
facile synthesis of the natural product 5.
60À83% yield. The benzylic alcohols 15aÀi were
oxidized to furnish ketones 16aÀi in 89À99% yield.
Scheme 2. Preparation of Substituted Benzoyl Quinones via
Moore Cyclization
Scheme 1. Retrosynthetic Entry to 1,4-Dioxygenated
Xanthones
The stage was then set for removal of the PMB protect-
ing group so the intermediate phenol could undergo
cyclization to deliver the desired xanthone. Although
reaction of 16aÀi with oxidants such as DDQ and CAN
largely returned recovered starting material, treatment
of these ketones with TFA in deoxygenated CH2Cl2
cleanly removed the PMB group allowing cyclization to
occur. The regioselectivity of the cyclization of the
intermediate phenol was found to be dependent upon
substitution on the aromatic ring. For example, depro-
tection of 16a furnished the xanthone 18a in 95%
yield (Scheme 3, entry a). Deprotection/cyclization of
16bÀd,g (entries bÀd,g) also provided xanthones
18bÀd,g. On the other hand, deprotection and cycliza-
tion of 16e,f,i gave inseparable mixtures of spirocyclic
ketones 17e,f,i and xanthones 18e,f,i, whereas 16h
afforded spirocyclic ketone 17h as the sole product.
Fortunately, we discovered that the spirocyclic ketones
17e,f,h,i underwent facile rearrangement to the corre-
sponding xanthones 18e,f,h,i upon treatment with
K2CO3, conditions much milder than those reported
for similar rearrangements.11
The commercially available salicylaldehydes 11aÀi were
first protected as their corresponding p-methoxyben-
zyl (PMB) ethers 12aÀi in 71À98% yield (Scheme 2).
Ethynylation of 12aÀi with ethynyl magnesium
bromide afforded the propargylic alcohols 13aÀi in
87À99% yields. The dianions of 13aÀi, which were
generated using 2.2 equiv of n-BuLi, were then allowed
to react with dimethoxysquarate (9) to give the ad-
ducts 14aÀi in 52À65% yield. Because 14aÀi were
only moderately stable, they were quickly purified
and then heated under reflux in toluene to induce the
Moore cyclization and give the quinones 15aÀi in
The disparate behavior of the variously substituted,
intermediate phenolic ketones derived from 16aÀi toward
cyclization to deliver 17 or 18 or mixtures thereof begs
rationalization. Based upon the observed substituent
effects, we believe a combination of electronic and steric
effects are at play, and we have formulated tentative
hypothesesthatareconsistent withthe extant experimental
observations. For substrates 16aÀd lacking substituents at
(8) For leading references to prepare 1,4-dioxygenated xanthones:
(a) Ellis, R. C.; Whalley, W. B.; Ball, K. Chem. Commun. 1967, 803–804.
(b) Stout, G. H.; Balkenhol, W. J. Tetrahedron 1969, 25, 1947–1960.
(c) Jain, A. C.; Khanna, V. K.; Seshadri, T. R. Indian J. Chem. 1970, 8,
667–669. (d) Simoneau, B.; Brassard, P. J. Chem. Soc., Perkin Trans. I
1984, 1507–1510. (e) Bekaert, A.; Andrieux, J.; Plat, M. Tetrahedron
Lett. 1992, 33, 2805–2806. (f) Elix, J. A.; Gaul, K. L.; Jiang, H. Aust. J.
Chem. 1993, 46, 95–110. (g) Sun, L.; Liebeskind, L. S. J. Am. Chem. Soc.
1997, 118, 12473–12474. (h) Hauser, F. M.; Dorsch, W. A. Org. Lett.
2003, 5, 3753–3754. (i) Masuo, R.; Ohmori, K.; Hintermann, L.; Suzuki,
K. Angew. Chem., Int. Ed. 2009, 48, 3462–3465.
(9) Knueppel, D.; Martin, S. F. Angew. Chem., Int. Ed. 2009, 48,
2569–2571.
(10) (a) Karlsson, J. O.; Nguyen, N. V.; Foland, L. D.; Moore, H. W.
J. Am. Chem. Soc. 1985, 107, 3392–3393. (b) Foland, L. D.; Karlsson,
J. O.; Perri, S. T.; Schwabe, R.; Xu, S. L.; Patil, S.; Moore, H. W. J. Am.
Chem. Soc. 1989, 111, 975–989.
(11) For other reports of rearrangements of spirocyclic ketones
related to 17 to xanthones, see: (a) Lewis, J. R.; Paul, J. G. J. Chem.
Soc., Perkin Trans. 1 1981, 770–775. (b) Koning, C. B.; Giles, R. G. F.;
Engelhardt, L. M.; White, A. H. J. Chem. Soc., Perkin Trans. 1 1988,
3209–3216.
Org. Lett., Vol. 13, No. 17, 2011
4697

We next applied the Moore cyclization approach to the
preparation of naturally occurring xanthones (Scheme 4).
This goal was readily achieved by the regioselective methy-
lation of xanthones 18b and 18e to afford the tetramethoxy
xanthones 5 and 22 in 84 and 99% yield, respectively.
The present synthesis of 5 in seven steps and in 22%
overall yield represents a significant improvement over
Scheme 3. Phenol Deprotection, Cyclization, and Spirocycle
RearrangementaÀd
1
the prior art.8b Although the H NMR spectral data for
synthetic 5 are consistent with those found in the
literature,8b,12 there are some notable differences in the
13C NMR data that have been reported for 5. A compound
that was named dulcisxanthone C and assigned the struc-
ture 22 has been isolated,13 but the spectral data for
synthetic 22, the structure of which was also verified by
X-ray crystallography, do not match those reported.14
1
Indeed, the H and 13C NMR data reported for dulcis-
xanthone C are consistent with those of synthetic 5.
a Inseparable mixture; ratios determined by 1H NMR. bOverall yield
from 15aÀi. cReaction required 24 h. dEthyl acetate was used as solvent
for rearrangement; yield in acetone was 30%.
Scheme 4. Regioselective Phenol Methylation
C(4) or C(6), xanthones 18aÀd are obtained as the only
isolable products; there are no significant substituent
effects. For 4-substituted compounds 16eÀg, the re-
giochemistry varies with the electron-donating ability
of the R-substituent. Electron release by a 4-methoxy
group might be expected to decrease the ability of the
ketone moiety to activate the double bond in the
quinone toward nucleophilic attack through resonance
as shown in 19, thereby leading to increased amounts of
the spirocycle 17e (Figure 2). The inductive effect of a
4-chloro substituent would have the opposite effect,
and only the xanthone 18g is formed. The weak elec-
tron-releasing ability of a 4-methyl substituent is re-
flected by an increased preference for forming the
xanthone 18f. Substituents at the 6-position of 16h,i
appear to exert both electronic and steric effects. The
presence of a substituent at the 6-position should dis-
favor the conformation 20 that is required for cycliza-
tion to the xanthone because of unfavorable steric
interactions that are relieved in the alternate conforma-
tion 21 that is predisposed to cyclize to a spirocyclic
ketone. Given that the effective sizes of chlorine and
methoxy groups are comparable, electronic effects may
also contribute to the observed preference for forming
17h, as the methoxy group would lessen the activating
effect of the ketone moiety toward a mode of 1,4-addition
that would generate the xanthone.
In summary, we have developed a general and efficient
strategy for the synthesis of 1,4-dioxygenated xanthones
that does not suffer the disadvantages of other known
procedures. We established the utility of the method by
its application to a significantly improved synthesis of
the naturally occurring xanthone 5 and the revision of
the structure of a compound that had been previously
reported to be dulcisxanthone C and misassigned the
structure 22. We also set forth a hypothesis involving
the interplay of electronic and steric factors for the
divergent regioselectivity that is observed in cycliza-
tions of 2-hydroxy aroyl quinones of the general struc-
ture 16 to form spirocycles or xanthones. Applications
of this methodology to the syntheses of more complex
1,4-dioxygenated xanthones are in progress, and the
results of these investigations will be reported in due
course.
Acknowledgment. We thank the National Institutes of
Health (GM31077), the Robert A. Welch Foundation
(F-652) for generous support of this research. We would
(12) (a) Wang, A.-X.; Zhang, Q.; Jia, Z.-J. Pharmazie 2004, 59, 807–
811. (b) Shi, G.-F.; Lu, R.-H.; Yang, Y.-S.; Li, C.-L.; Yang, A.-M.; Cai,
L.-X. Chin. J. Struct. Chem. 2004, 23, 1164–1168. (c) Mikhailova, T. M.;
Tankhaeva, L. M.; Shul’ts, E. E.; Tolstikov, G. A. Chem. Nat. Compd.
2005, 41, 513–515.
(13) Deachathai, S.; Mahabusarakam, W.; Phongpaichit, S.; Taylor,
W. C.; Zhang, Y.-J.; Yang, C.-R. Phytochemistry 2006, 67, 464–469.
(14) CCDC 830944 contains the supplementary crystallographic
data for 22. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
Figure 2. Spirocycle versus fused: Hypothesis for divergent
regioselectivity.
4698
Org. Lett., Vol. 13, No. 17, 2011

also like to thank Vincent Lynch (The University of Texas)
for obtaining X-ray data for compound 22.
well as X-ray information for compound 22 and com-
1
parison of H and 13C NMR spectral data of dulcis-
xanthone C and synthetic and natural 5. This material
is available free of charge via the Internet at http://
pubs.acs.org
Supporting Information Available. Experimental pro-
cedures and spectral data for all new compounds as
Org. Lett., Vol. 13, No. 17, 2011
4699