Approach toward the total synthesis of 5-hydroxyaloin a

Article abstract of DOI:10.1021/ol102318k

The synthesis of a thiomethyl analogue of 5-hydroxyaloin A has been achieved using benzyne and naphthyne [4 + 2] cycloadditions with substituted furans. A regiocontrolled cycloaddition was achieved using a silicon tether, and a regioselective ring opening

Full text of DOI:10.1021/ol102318k

ORGANIC
LETTERS
2010
Vol. 12, No. 24
5632-5635
Approach Toward the Total Synthesis of
5-Hydroxyaloin A
Kristen J. Procko, Hui Li, and Stephen F. Martin*
Department of Chemistry and Biochemistry, The UniVersity of Texas at Austin, Austin,
Texas 78712, United States
sfmartin@mail.utexas.edu
Received September 26, 2010
ABSTRACT
The synthesis of a thiomethyl analogue of 5-hydroxyaloin A has been achieved using benzyne and naphthyne [4 + 2] cycloadditions with
substituted furans. A regiocontrolled cycloaddition was achieved using a silicon tether, and a regioselective ring opening was accomplished
using a sulfide as a directing group.
Over the past three decades, there have been numerous synthetic
approaches to the C-aryl glycoside class of natural products,
owing to both their significant anticancer properties and the
challenges associated with forming the C-glycosidic linkage.1
Toward addressing this problem, we have developed a unified
strategy for preparing the four major classes of C-aryl glyco-
sides.2 The approach features benzyne-furan [4 + 2] cycload-
ditions and has been applied to a formal synthesis of galtamy-
cinone3 and total syntheses of vineomycinone B2 methyl ester4
and isokidamycin.5 We also envisioned that this methodology
Figure 1. Anthrone C-glycosides.
might be extended to enable facile access to the novel class of
anthrone C-glycosides as represented by 5-hydroxyaloin A (1),6
aloin (2),7 and cassialoin (3) (Figure 1).8,9 These C-glycosides,
which are not technically categorized as C-aryl glycosides,
contain a sugar moiety bonded to C(10) of an anthrone ring
via a C-C bond. We thus became interested in the aloe-
derived glycoside 5-hydroxyaloin A (1), which has not yet
been prepared by total synthesis.
(1) For reviews on the synthesis of C-aryl glycosides, see: (a) Jaramillo,
C.; Knapp, S. Synthesis 1994, 1–20. (b) Suzuki, K. Pure Appl. Chem. 1994,
66, 2175–2178. (c) Levy, D. E.; Tang, C. In The Chemistry of C-Glycosides;
Elsevier Science: Tarrytown, NY, 1995. (d) Du, Y. G.; Linhardt, R. J.;
Vlahov, I. R. Tetrahedron 1998, 54, 9913–9959. (e) Liu, L.; McKee, M.;
Postema, M. H. D. Curr. Org. Chem. 2001, 5, 1133–1167.
(2) (a) Kaelin, D. E.; Lopez, O. D.; Martin, S. F. J. Am. Chem. Soc.
2001, 123, 6937–6938. (b) Martin, S. F. Pure Appl. Chem. 2003, 75, 63–
70. (c) Kaelin, D. E., Jr.; Sparks, S. M.; Plake, H. R.; Martin, S. F. J. Am.
Chem. Soc. 2003, 125, 12994–12995.
(6) (a) Bisrat, D.; Dagne, E.; van Wyk, B. E.; Viljoen, A. Phytochemistry
2000, 55, 949–952. (b) Holzapfel, C. W.; Wessels, P. L.; VanWyk, B. E.;
Marais, W.; Portwig, M. Phytochemistry 1997, 45, 97–102.
(3) Apsel, B.; Bender, J. A.; Escobar, M.; Kaelin, D. E., Jr.; Lopez,
O. D.; Martin, S. F. Tetrahedron Lett. 2003, 44, 1075–1077.
(4) (a) Chen, C. L.; Sparks, S. M.; Martin, S. F. J. Am. Chem. Soc.
2006, 128, 13696–13697. (b) Sparks, S.; Chen, C.; Martin, S. Tetrahedron
2007, 63, 8619–8635.
(7) Birch, A. J.; Donovan, F. W. Aust. J. Chem. 1955, 8, 523–528.
(8) Krenn, L.; Pradhan, R.; Presser, A.; Reznicek, G.; Kopp, B. Chem.
Pharm. Bull. 2004, 52, 391–393
(9) Koyama, Y.; Yamaguchi, R.; Suzuki, K. Angew. Chem., Int. Ed.
2008, 47, 1084–1087
.
(5) O’Keefe, B. M.; Mans, D. M.; Kaelin, D. E., Jr.; Martin, S. F. J. Am.
Chem. Soc. 2010, 132, 15528–15530.
.
10.1021/ol102318k  2010 American Chemical Society
Published on Web 11/17/2010

In our retrosynthetic analysis, we envisioned that 5-hy-
droxyaloin A (1) would be formed by desulfurization of 4,
which would arise from desilylation and acid-catalyzed ring
opening of 5 (Scheme 1). The intermediate 5 would be
naphthyne as a regiocontrol element. Although [4 + 2]
cycloadditions involving benzynes and furans can proceed
with good regioselectivity,10 examination of the literature
suggests that the polarization of a naphthyne generated from
a derivative of 9 and a 2-thiomethylfuran would likely
proceed preferentially in the undesired regiochemical sense
(Figure 2). In an exploratory experiment, we found that this
Scheme 1. Retrosynthetic Analysis of 5-Hydroxyaloin A (1)
Figure 2. Predicted preferential regioselectivity of bimolecular [4
+ 2] naphthyne-furan cycloaddition.
intermolecular cycloaddition did in fact deliver a mixture of
cycloadducts. To address such problems of low regioselec-
tivity in intermolecular benzyne cycloadditions, we had
previously shown that silicon tethers can be nicely exploited
as disposable linkers to circumvent this problem by rendering
the reaction intramolecular.2c
We initiated our efforts with the preparation of the
naphthol derivative 9. In the event, the benzyne precursor
13 was formed in 81% yield upon treatment of chlorohyd-
roquinone (12) with BnBr and NaH (Scheme 2). Deproto-
Scheme 2. Synthesis of Naphthol 9
formed from an intramolecular [4 + 2] cycloaddition of a
furanyl naphthyne that would be generated from 6, which
in turn would be assembled by etherification of 9 with 7.
The bromosilane 7 would then be accessed in three steps
from commercially available 3-furanmethanol (8). The [4
+ 2] cycloaddition of benzyne 10 and glycosyl furan 11
would lead to the substituted C-aryl glycoside 9.
There are several issues associated with the regiochemical
outcomes of reactions in our retrosynthetic analysis that merit
additional comment. The first of these involves the regiochem-
istry of the ring opening of 5 to give 4. Given the known
propensity of related compounds lacking an electron-donating
group at the bridgehead to undergo acid-catalyzed ring opening
in the opposite regiochemical sense,2a,10c the placement of an
electron-donating group at the bridgehead to direct the ring
opening was deemed essential. Indeed, the importance of
using such a directing group for the ring opening of
compounds related to 5 was validated in preliminary model
studies. Either an alkoxy or a thioalkyl substituent could serve
this purpose, but a thioalkyl, in particular, a thiomethyl, group
was viewed as being easier to remove at a late stage in the
synthesis. A second feature of the approach is the use of an
intramolecular [4 + 2] cycloaddition of a furan and a
nation of 13 with s-BuLi at -95 °C and addition of the
known glycosyl furan 7,11 followed by warming to -10 °C,
provided cycloadduct 14 in 53% yield (84% BRSM). Ring
(10) For example: (a) Matsumoto, T.; Hosoya, T.; Katsuki, M.; Suzuki,
K. Tetrahedron Lett. 1991, 32, 6735–6736. (b) Giles, R. G. F.; Hughes,
A. B.; Sargent, M. V. J. Chem. Soc., Perkin Trans. 1 1991, 1581–1587. (c)
Baker, R. W.; Baker, T. M.; Birkbeck, A. A.; Giles, R. G. F.; Sargent,
M. V.; Skelton, B. W.; White, A. H. J. Chem. Soc., Perkin Trans. 1 1991,
1589–1600.
(11) Czernecki, S.; Ville, G. J. Org. Chem. 1989, 54, 610–612.
Org. Lett., Vol. 12, No. 24, 2010
5633

opening of the oxabicycloheptadiene ring was achieved upon
treating 14 with ZnCl2 to give an intermediate phenol that
was regioselectively chlorinated by the action of N-chloro-
succinimide to give 9 in 64% overall yield.
Scheme 4. Etherification and Naphthyne Cycloaddition
The next stage of the synthesis required the preparation
of a suitably substituted furan that would eventually give
rise to the unsymmetrically substituted C ring of 1. Regi-
oselective lithiation of 3-furanmethanol (15) at C(2), followed
by reaction with dimethyl disulfide, gave 16 (Scheme 3).12
Scheme 3. Synthesis of the Furan Precursor of the C Ring
With the cycloadduct 5 in hand, it remained to cleave the
silicon tether, open the oxabicycloheptadiene, and remove
the benzyl protecting groups and the thiomethyl moiety. In
prior work, we had developed an effective two-step procedure
for removing the silicon tether with TBAF in DMF, followed
by opening the oxabicyclo ring with a Brønsted or Lewis
acid to give a naphthol.2c In the present instance, however,
we found that when 5 was simply treated with camphorsul-
fonic acid (CSA) the anthrone 21 was obtained as a single
diastereomer in 80% yield (Scheme 5).
Treatment of sulfide 16 with TBSCl and imidazole furnished
silyl ether 17 in about 60% overall yield from 15. The
conversion of 17 into 18 was achieved by sequential lithiation
of 17 at C(5) and treatment with bromomethyl chlorodim-
ethylsilane. Unfortunately, we discovered that the etherifi-
cation of the naphthol 9 with 18 under a variety of conditions
was problematic, giving complex mixtures of compounds.
This difficulty was not completely unexpected as we had
previously encountered problems with similar alkylations.2c
Indeed, during the course of those investigations, we had
also developed a solution to this problem wherein a Mit-
sunobu reaction was employed to prepare phenolic ethers.
Accordingly, bromide 18 was converted to the acetate 19
by treatment with NaOAc and TBAI; subsequent reduction
of 19 with LiAlH4 then provided the requisite alcohol 20.
Naphthol 9 and alcohol 20 were then coupled by a
Mitsunobu etherification to give 6 in 72% yield (Scheme
4). Deprotonation of 6 with s-BuLi led to naphthyne
generation and intramolecular cycloaddition to furnish 5
in 34% yield (75% BRSM). Unfortunately, it was not
possible to optimize this reaction by using excess s-BuLi
because Si-C bond cleavage intervened as a major side
reaction. Fortunately, however, it was possible to recover
and recycle 6.
Scheme 5
.
Synthesis of the Thiomethyl Analogue of
5-Hydroxyaloin A
Attempts to desulfurize 21 using a variety of reagents,
including deactivated Raney-Ni,13 nickel boride,14 and
NiCRAs,15 were uniformly unsuccessful, giving either no
reaction or intractable mixtures. We thus postponed the
desulfurization step and effected global deprotection of the
(12) For a similar reaction, see: Goldsmith, D.; Liotta, D.; Saindane,
M.; Waykole, L.; Bowen, P. Tetrahedron Lett. 1983, 24, 5835–5838.
5634
Org. Lett., Vol. 12, No. 24, 2010

benzyl and silyl ethers on 21 using BBr3 to furnish 22 as a
single diastereomer in 65% yield. Once again, however, all
efforts to effect desulfurization of 22 to give 5-hydroxyaloin
A (1) using Raney nickel, nickel boride, and NiCRAs were
unsuccessful. We also explored the possibility of desulfur-
izing the sulfoxide and the sulfone derived from 22 using
Raney nickel, Na/Hg amalgam in the presence of dibasic
sodium phosphate,16 and SmI2,17 but these attempts were
also fruitless. We were therefore reluctantly compelled to
accept the synthesis of the thiomethyl 5-hydroxyaloin A
derivative 22 as the terminal goal for these efforts. Although
both 21 and 22 were isolated as single diastereomers, we
were unable to determine the relative stereochemistry at
C(10) in either. Efforts to obtain crystals suitable for X-ray
analysis were unsuccessful, and NOE experiments were
ambiguous.
In summary, we have applied a variant of our benzyne-
furan [4 + 2] cycloaddition strategy for the synthesis of a
thiomethyl analogue of 5-hydroxyaloin A. Although the
thiomethyl group served admirably in its appointed task of
directing the regiochemistry of the ring opening of the
cycloadduct 5, difficulties encountered in its removal pre-
cluded us from completing the synthesis of 5-hydroxyaloin
A. An alternate plan must now be developed. Nevertheless,
we did discover a useful one-step procedure for converting
cycloadducts such as 5 into naphthol derivatives, thus
improving upon our prior art. Further applications of our
general strategy for the synthesis of C-aryl glycosides are
in progress and will be reported in due course.
Acknowledgment. We wish to thank the National Insti-
tutes of Health General Medical Sciences (GM 31077), the
Robert A. Welch Foundation (F-0652), Pfizer, Inc., Merck
Research Laboratories, and Boehringer-Ingelheim Pharma-
ceuticals for their generous support of this research.
(13) (a) Barun, O.; Nandi, S.; Kausik Panda; Ila, H.; Junjappa, H. J.
Org. Chem. 2002, 67, 5398–5401. (b) Nandi, S.; Kumar, U. K. S.; Ila, H.;
Junjappa, H. J. Org. Chem. 2002, 67, 4916–4923. (c) Boger, D. L.;
Wolkenberg, S. E. J. Org. Chem. 2000, 65, 9120–9124.
(14) Back, T. G.; Yang, K. X.; Krouse, H. R. J. Org. Chem. 1992, 57,
1986–1990.
Supporting Information Available: Experimental pro-
(15) (a) Becker, S.; Fort, Y.; Vanderesse, R.; Caubere, P. J. Org. Chem.
1989, 54, 4848–4853. (b) Becker, S.; Fort, Y.; Caubere, P. J. Org. Chem.
1990, 55, 6194–6198.
cedures, chromatographic data, and characterization and
1
copies of H and 13C NMR spectra for all new compounds.
(16) Mirsadeghi, S.; Rickborn, B. J. Org. Chem. 1985, 40, 4340–4345.
(17) SmI2 generation: Concello, J. M.; Rodry´guez-Solla, H.; Bardales,
E.; Huerta, M. Eur. J. Org. Chem. 2003, 2003, 1775–1778. For an example
of desulfurization using SmI2, see: Adrio, J.; Carretero, J. C.; Garc´ıa, J. L.;
Ruano, L. M.; Cabrejas, M. Tetrahedron: Asymmetry 1997, 8, 1623–1631.
This material is available free of charge via the Internet at
http://pubs.acs.org.
OL102318K
Org. Lett., Vol. 12, No. 24, 2010
5635