A four-step total synthesis of radermachol

Article abstract of DOI:10.1021/ol500862w

Radermachol has been synthesized in four steps and an overall yield of 22% via key ytterbium triflate catalyzed furannulation and intramolecular nucleophilic acylation reactions.

Full text of DOI:10.1021/ol500862w

Letter
pubs.acs.org/OrgLett
A Four-Step Total Synthesis of Radermachol
Marco Buccini and Matthew J. Piggott*
School of Chemistry & Biochemistry, The University of Western Australia, Perth, Australia
S
* Supporting Information
ABSTRACT: Radermachol has been synthesized in four steps
and an overall yield of 22% via key ytterbium triflate catalyzed
furannulation and intramolecular nucleophilic acylation
reactions.
s part of investigations into the total synthesis of the
marine pyrroloacridine alpkinidine,1 we attempted to
generate its CE ring system by beginning with a Michael-type
addition of butyl vinyl ether (2) to the very electrophilic
(“activated”) quinone 1,2 with the view that the adduct 3, or
some related species, could be elaborated to isoquinolinetrione
4, and then on to the natural product (Scheme 1). In practice,
product synthesis.4 Several 5-hydroxybenzo[2,3-b]furan natural
products bearing an electron-withdrawing group at the 4-
position are known, and the methodology exemplified in
Scheme 1 provides a very rapid, regiospecific route to this
substructure. Thus, we set out to prove its utility in the total
synthesis arena.
A
Our first efforts focused on furomollugin (Scheme 2), which
was originally isolated5 from the rhizomes of Galium mollugo L.,
Scheme 1. Unexpected Formal [3 + 2] Cycloaddition of
Electron-Deficient Quinone 1 with Butyl Vinyl Ether (2),
and Subsequent Aromatization: A Net Furannulation
Scheme 2. A Rapid, Regiospecific Synthesis of Furomollugin
the reaction of 1 and 2 cleanly produced the dihydrobenzofuran
5, resulting from intramolecular interception of the initial
Michael adduct by the (formerly) quinonoid oxygen atom, a
formal [3 + 2] cycloaddition. A survey of the literature revealed
that similar reactions have been known for some time.3
It was apparent that a simple acid-catalyzed elimination of
BuOH from 5 would generate the benzofuran 6, and indeed the
expected outcome was achieved in good yield, either on
isolated acetal 5 or in one pot over the two steps. Such
“furannulations” have not been extensively exploited in natural
a European herbaceous annual from the Rubiaceae family,
commonly known as false baby’s breath. Furomullogin has
since been found in several Rubia species6 and has been shown
to exhibit activity against the hepatitis B virus6c and weak
cytotoxicity to a human colon carcinoma cell line (HT-29),
possibility mediated through topoisomerase inhibition.6e
Received: March 23, 2014
Published: April 21, 2014
© 2014 American Chemical Society
2490
dx.doi.org/10.1021/ol500862w | Org. Lett. 2014, 16, 2490−2493

Organic Letters
Letter
The original isolators synthesized furomollugin in two steps
and 2% overall yield by cyclocondensation of glyoxal with 1,4-
dihydroxy-2-naphthoic acid, followed by selective esterification
with diazomethane.5 Furomollugin was produced as a minor
byproduct in Trauner’s elegant biomimetic synthesis of the
complex 5-hydroxybenzo[2,3-b]furan natural product rubicor-
difolin.7 The same group reported a second synthesis from the
related benzochromene natural product mollugin, via an
efficient sequence of oxidation, and base-induced rearrange-
ment and retro-Friedel−Crafts hydroxyalkylation, en route to a
total synthesis of rubioncolin B.8 A similar ring contraction of
mollugin was later reported by De Kimpe and co-workers.9
Our synthesis of furomollugin is outlined in Scheme 2. The
required quinone 710 was prepared by Fischer esterification of
1,4-dihydroxynaphthoic acid,11 followed by oxidative demethy-
lation. The reaction of 7 with butyl vinyl ether (2) gave acetal 8
in good yield. Acid-catalyzed aromatization then afforded
furomollugin. The one-pot procedure in toluene was
complicated by a cascading Friedel−Crafts alkylation of the
initial product with excess butyl vinyl ether, providing the dimer
9 in moderate yield. Switching solvent to MeOH before
acidification avoided the problem and gave furomollugin in a
slightly better yield than the two-pot process.
Scheme 3. Unsuccessful Approach to Radermachol
During the course of our work, Lee and Xia reported a
synthesis of furomollugin by a very similar method, using ceric
ammonium nitrate as a Lewis acid catalyst for the cyclization
reaction of 1 with ethyl vinyl ether, and tosic acid-catalyzed
elimination of EtOH to provide the natural product in excellent
yield over two steps.12 They have since used this methodology
to synthesize diverse furomollugin analogues and assess them
for antioxidant and antibacterial activities.13
The successful synthesis of furomollugin inspired us to tackle
a more challenging target, radermachol (Scheme 3). This bright
red pigment was first isolated14 from Radermachera xylocarpa K.
Schum, a plant from the Western Ghats of India. Its unique,
fused pentacyclic structure was established by spectroscopic
analysis and X-ray crystallography.14 It has since been found in
Tecomella undulata, another Indian plant from the Bignoniaceae
family.15 Although both plant sources are used as traditional
medicines, it seems that the biological activity of radermachol
has not been investigated. Two previous syntheses of
radermachol exist. The first, from the Pelletier group, afforded
the natural product in 14 steps and 7% overall yield.16 The
second, by Hauser and Yin,17 intercepted an intermediate in the
Pelletier’s synthesis, providing radermachol in 11 steps.
Our first approach to radermachol is outlined in Scheme 3.
Friedel−Crafts acylation of 1,4-dimethoxynaphthalene (10)
with phthalic acid monomethyl ester (11) proceeded smoothly
in refluxing TFAA/TFA,16b,18 without the need for Lewis acid
catalysis. CAN-mediated oxidative demethylation19 of 12 was
effective at low temperatures, giving quinone 13 in an
acceptable yield; at higher temperatures the overoxidation
product 14, presumably arising from conjugate addition of
water to 13 and reoxidation, predominated. To our delight, the
key furannulationreaction with silyl ether 15,20 derived in
one step from mesityl oxide, followed by in situ treatment with
TFAgave the desired isobutenylbenzofuran 16 in excellent
yield.
ester substituent, helping to direct lithiation of the proximal
furan β-carbon by the second equivalent of LDA. The lithiated
intermediate so formed could then attack the ester, giving
radermachol in one operation. In practice, only the starting
material was returned suggesting that there was no directed
lithiation.
We then resorted to intramolecular Friedel−Crafts acylation
of the activated furan β-position to effect the final ring closure.
Although challenging in the presence of the sensitive isobutenyl
substituent, we were encouraged by the already demonstrated
stability of this group to TFA. Ester 16 was first saponified to
give 17 (Scheme 3). Unfortunately, neither the in situ
generated trifluoroacetic mixed anhydride nor acid chloride
underwent the desired cyclization, despite many attempts and,
in the latter case, with various Lewis acids. In all cases,
intractable mixtures of products were formed. Based on the
hypothesis that the free phenol may contribute to these failures,
16 was methylated and then saponified. However, when 19 was
subjected to another raft of Friedel−Crafts acylation conditions,
no radermachol methyl ether (20) could be isolated from the
multitude of products formed.
It was obvious at this juncture that electrophilic ring closure
in this system was going to be very difficult. Accordingly, a
nucleophilic ring-closure strategy was devised (Scheme 4).
Acylation of 1,4-dimethoxynaphthalene (10) with 2-iodoben-
zoic acid (21) gave diarylketone 22. With this substrate,
oxidative demethylation proceeded cleanly, providing quinone
23 in excellent yield. Application of the furannulation protocol
required an enol ether such as 24, which presumably could be
derived from β-ketoester 25. However, it occurred to us that
The first efforts to achieve ring closure to form the
pentacyclic framework of radermachol began with treatment
of 16 with 2 equiv of LDA. The rationale here was that the first
equivalent would deprotonate the phenolic hydroxyl, deactivat-
ing the diarylketone to nucleophilic attack and, along with the
2491
dx.doi.org/10.1021/ol500862w | Org. Lett. 2014, 16, 2490−2493

Organic Letters
Letter
In conclusion, radermachol was prepared in four linear steps,
from commercially available starting materials, and in an overall
yield of 22%, a marked improvement over existing syntheses.
This rapid and efficient synthesis provides a means to
investigate the biological activity of radermachol and perhaps
elucidate its ecological role.
Scheme 4. Total Synthesis of Radermachol
ASSOCIATED CONTENT
* Supporting Information
■
S
General experimental details and specific procedures, character-
ization of new compounds, including 1H and 13C NMR spectra.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: matthew.piggott@uwa.edu.au.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We acknowledge the facilities as well as scientific and technical
assistance of Drs. Lindsay Byrne and Anthony Reeder, of the
Australian Microscopy and Microanalysis Research Facility at
the Centre for Microscopy, Characterisation, and Analysis, The
University of Western Australia, a facility funded by the
University, State and Commonwealth Governments. M.B. is the
recipient of an Australian Postgraduate Award.
the enol tautomer of 25 might itself be sufficiently nucleophilic
for our requirements. These musings led us to the work of De
Kimpe and co-workers, who showed that ytterbium triflate
catalyzes the efficient one-pot furannulation of activated
quinones with β-ketoesters.21 A trial reaction of butyl vinyl
ether (2) and quinone 1 (Scheme 1) with this catalyst gave the
formal cycloaddition product 5 at room temperature, and more
efficiently than the thermal procedure, giving us the confidence
to apply this method to radermachol.
β-Ketoester 25 (Scheme 4) was prepared by Claisen-like
condensation of mesityl oxide with Mander’s reagent.22
Pleasingly, in the presence of Yb(OTf)3, 25 reacted with
quinone 23 to give benzofuran 26 in good yield. In situ
deprotonation of the phenol with NaH, followed by lithium−
iodine exchange, allowed intramolecular nucleophilic acylation
involving the methyl ester, providing radermachol in modest
yield, accompanied by some starting material and the
deiodinated product. The NMR spectroscopic data obtained
for radermachol were essentially identical to those previously
reported for the naturally derived material.15 The 13C NMR
spectrum reported by Pelletier16b for synthetic radermachol is
at slight variance with what we and Sing15 observed; the
resonance at 117.3 ppm is missing from Pelletier’s data, and an
additional (spurious) peak at 129.3 ppm was reported. All other
data concur with our spectrum.
REFERENCES
■
(1) Buccini, M.; Jeow, S. Y.; Byrne, L.; Skelton, B. W.; Nguyen, T.
M.; Chai, C. L. L.; Piggott, M. J. Eur. J. Org. Chem. 2013, 3232.
(2) Matsumoto, K.; Kozmin, S. A. Adv. Synth. Catal. 2008, 350, 557.
(3) (a) Eugster, C. H.; Kuser, P. Chimia 1964, 18, 358. (b) Kuser, P.;
Frauenfelder, E. F.; Eugster, C. H. Helv. Chim. Acta 1971, 54, 969.
(c) Cameron, D. W.; Feutrill, G. I.; Sefton, M. A. Aust. J. Chem. 1978,
31, 2099. (d) Cameron, D. W.; Feutrill, G. I.; Patti, A. F.; Perlmutter,
P.; Sefton, M. A. Aust. J. Chem. 1982, 35, 1501. (e) Noland, W. E.;
Kedrowski, B. L. Org. Lett. 2000, 2, 2109. (f) Alvey, L.; Prado, S.;
Huteau, V.; Saint-Joanis, B.; Michel, S.; Koch, M.; Cole, S. T.;
Tillequin, F.; Janin, Y. L. Bioorg. Med. Chem. 2008, 16, 8264.
(4) (a) Brimble, M. A.; Brimble, M. T.; Gibson, J. J. J. Chem. Soc.,
Perkin Trans. 1 1989, 179. (b) Brimble, M. A.; Spicer, J. A. Aust. J.
Chem. 1991, 44, 197. (c) Brimble, M. A.; Neville, D.; Duncalf, L. J.
Tetrahedron Lett. 1998, 39, 5647. (d) Uno, H.; Murakami, S.;
Fujimoto, A.; Yamaoka, Y. Tetrahedron Lett. 2005, 46, 3997.
(5) Schildknecht, H.; Straub, F. Justus Liebigs Ann. Chem. 1976, 1772.
(6) (a) Itokawa, H.; Qiao, Y.; Takeya, K. Phytochemistry 1991, 30,
637. (b) Wu, L.; Wang, S.; Hua, H.; Li, X.; Zhu, T.; Miyase, T.; Ueno,
A. Phytochemistry 1991, 30, 1710. (c) Ho, L.-K.; Don, M.-J.; Chen, H.-
C.; Yeh, S.-F.; Chen, J.-M. J. Nat. Prod. 1996, 59, 330. (d) Gutpa, P. P.;
Srimal, R. C.; Verma, N.; Tandon, J. S. Pharmaceutical Biol. 1999, 37,
46. (e) Son, J. K.; Jung, S. J.; Jung, J. H.; Fang, Z.; Lee, C. S.; Seo, C.
S.; Moon, D. C.; Min, B. S.; Kim, M. R.; Woo, M. H. Chem. Pharm.
Bull. 2008, 56, 213. (f) Hou, B.; Wang, S. Zhongcaoyao 2000, 31, 492.
(7) Lumb, J.-P.; Trauner, D. J. Am. Chem. Soc. 2005, 127, 2870.
(8) Lumb, J.-P.; Choong, K. C.; Trauner, D. J. Am. Chem. Soc. 2008,
130, 9230.
It is possible that the yield of the final cyclization could be
improved by prior protection of the phenol in 26. Then again,
as mentioned previously, the corresponding phenoxide serves
to deactivate the ketone carbonyl to nucleophilic attack, and no
protecting group can achieve this. Moreover, given the brevity
of the route in Scheme 4, and the good yields of the preceding
steps, additional protection and deprotection steps did not
seem warranted.
(9) Sastry, M. N. V.; Claessens, S.; Habonimana, P.; De Kimpe, N. J.
Org. Chem. 2010, 75, 2274.
(10) Brimble, M. A.; Spicer, J. A. Aust. J. Chem. 1991, 44, 197.
(11) Jung, H. W.; Oh, J. S.; Lee, S. H.; Liang, J. L.; Kim, D. H.;
Rahman, A. F. M. M.; Jahng, Y. Bull. Korean Chem. Soc. 2007, 28, 1863.
(12) Xia, L.; Lee, Y. R. Org. Biomol. Chem. 2013, 11, 6097.
2492
dx.doi.org/10.1021/ol500862w | Org. Lett. 2014, 16, 2490−2493

Organic Letters
Letter
(13) Xia, L.; Idhayadhulla, A.; Lee, Y. R.; Kim, S. H.; Wee, Y.-J. Med.
Chem. Res. 2014, DOI: 10.1007/s00044-014-0929-9.
(14) Joshi, B. S.; Gawad, D. H.; Pelletier, S. W.; Kartha, G.; Bhandary,
K. Tetrahedron Lett. 1984, 25, 5847.
(15) Singh, P.; Khandelwal, P.; Hara, N.; Asai, T.; Fujimoto, Y. Indian
J. Chem., Sect. B 2008, 47B, 1865.
(16) (a) Jiang, Q.; Joshi, B. S.; Pelletier, S. W. Tetrahedron Lett. 1991,
32, 5283. (b) Joshi, B. S.; Jiang, Q.; Rho, T.; Pelletier, S. W. J. Org.
Chem. 1994, 59, 8220.
(17) Hauser, F. M.; Yin, H. Org. Lett. 2000, 2, 1045.
(18) De Koning, C. B.; Giles, R. G. F.; Engelhardt, L. M.; White, A.
H. J. Chem. Soc., Perkin Trans. 1 1988, 3209.
(19) Brimble, M. A.; Nairn, M. R. J. Chem. Soc., Perkin Trans. 1 1990,
169.
(20) Cazeau, P.; Duboudin, F.; Moulines, F.; Babot, O.; Dunogues, J.
Tetrahedron 1987, 43, 2089.
(21) Mudiganti, N. V. S.; Claessens, S.; De Kimpe, N. Tetrahedron
2009, 65, 1716.
(22) (a) Crabtree, S. R.; Chu, W. L. A.; Mander, L. N. Synlett 1990,
169. (b) Snider, B. B.; Patricia, J. J. J. Org. Chem. 1989, 54, 38.
NOTE ADDED AFTER ASAP PUBLICATION
■
The reference to Hauser and Yin was missing in the version
published ASAP April 21, 2014; the new reference 17 was
added and reposted April 22, 2014.
2493
dx.doi.org/10.1021/ol500862w | Org. Lett. 2014, 16, 2490−2493