Total synthesis of sanguiin H-5

Article abstract of DOI:10.1021/ol8009545

(Chemical Equation Presented) Using an atropdiastereoselective oxidative biaryl coupling as the key step, the total synthesis of the ellagitannin natural product sanguiin H-5 is reported. Both organomagnesium and organozinc based metalation methodologies

Full text of DOI:10.1021/ol8009545

ORGANIC
LETTERS
2008
Vol. 10, No. 12
2593-2596
Total Synthesis of Sanguiin H-5
Xianbin Su, David S. Surry, Richard J. Spandl, and David R. Spring*
Department of Chemistry, UniVersity of Cambridge, Lensfield Road,
Cambridge CB2 1EW, UK
drspring@ch.cam.ac.uk
Received April 25, 2008
ABSTRACT
Using an atropdiastereoselective oxidative biaryl coupling as the key step, the total synthesis of the ellagitannin natural product sanguiin H-5
is reported. Both organomagnesium and organozinc based metalation methodologies were used to efficiently construct the strained medium-
ring core of the natural product.
Sanguiin H-5 is a member of the ellagitannin class of
hydrolyzable plant polyphenols.1 In addition to industrial
applications,2 ellagitannins display a range of useful biologi-
cal properties including antiviral3 and antitumor activities.4
The structural features of sanguiin H-5 make it a challenging
molecular target.5 In addition to the ꢀ-glycosidic link at the
anomeric center, the characteristic hexahydrodiphenoyl
(HHDP) moiety1c common to all ellagitannins is part of a
strained medium ring with an (S)-configuration about the
biaryl bond (Scheme 1).
generated the desired (S)-atropoisomer of the HHDP group,
the complete reaction pathway suffered from somewhat
moderate yields, illustrating the complexity of the target
molecule. Since there are only a few reports of the efficient
synthesis of biaryl-containing medium rings, we sought to
exploit our organocuprate oxidation protocols7 to forge the
key biaryl bond of sanguiin H-5. These coupling methodolo-
gies utilized the following general sequence: halogen-metal
exchange (either iodine-magnesium7a or bromine-zinc;7b
copper salt mediated transmetalation; and finally, organo-
cuprate oxidation and biaryl bond formation.
We envisaged a strategy whereby the globally protected
sanguiin H-5 precursor 1 could be accessed in an atropdi-
astereoselective intramolecular biaryl coupling from the
appropriate diaryl halide 2 or 3 (Scheme 1).
The precursors 2 and 3 could potentially be synthesized
by the diacylation of the diol 4 with the gallic acid derivatives
5 or 6. Provided the pyransose 4 could be synthesized with
a ꢀ-configuration at the anomeric center and that the
halogenation of 7 is possible, compounds 7, 8, and 9 should
serve as readily available starting materials (Scheme 1).
The only previous synthesis of sanguiin H-5 was reported
by Feldman and Sambandam and involved an elegant
oxidative coupling of pendant aryl groups attached to a
central pyranose scaffold.6 Although the key coupling step
(1) (a) Tanaka, T.; Nonaka, G. I.; Nishioka, I. J. Chem. Res. Synop.
1985, 176. (b) Khanbabaee, K.; van Ree, T. Nat. Prod. Rep. 2001, 18, 641–
649. (c) Quideau, S.; Feldman, K. S. Chem. ReV. 1996, 96, 475–503, and
references therein.
(2) (a) Hemingway, R. W.; Laks, P. E. Basic Life Science. Plant
Polyphenols, Synthesis, Properties, Significance; Plenum Press: New York,
1992. (b) Haslam, E. Plant Polyphenols. Vegetable Tannins reVisited;
Cambridge University Press: Cambridge, 1989.
(3) (a) Nakashima, H.; Murakami, T.; Yamamoto, N.; Sakagami, H.;
Tanuma, S.; Hatano, T.; Yoshida, T.; Okuda, T. AntiViral Res. 1992, 18,
91–103. (b) Xie, L.; Xie, J. X.; Kashiwada, Y.; Cosentino, L. M.; Liu, S. H.;
Pai, R. B.; Cheng, Y. C.; Lee, K. H. J. Med. Chem. 1995, 38, 3003–3008.
(4) (a) Kashiwada, Y.; Nonaka, G.; Nishioka, I.; Chang, J. J.; Lee, K. H.
J. Nat. Prod. 1992, 55, 1033–1043. (b) Kashiwada, Y.; Nonaka, G.;
Nishioka, I.; Lee, K. J. H.; Bori, I.; Fukushima, Y.; Bastow, K. F.; Lee,
K. H. J. Pharm. Sci. 1993, 82, 487–492.
(6) Feldman, K. S.; Sambandam, A. J. Org. Chem. 1995, 60, 8171–
8178.
(7) (a) Surry, D. S.; Su, X.; Fox, D. J.; Franckevicius, V.; Macdonald,
S. J. F.; Spring, D. R. Angew. Chem., Int. Ed 2005, 44, 1870–1873. (b) Su,
X.; Fox, D. J.; Blackwell, D. T.; Tanaka, T.; Spring, D. R. Chem. Commun.
2006, 3883–3885. (c) Surry, D. S.; Fox, D. J.; Macdonald, S. J. F.; Spring,
D. R. Chem. Commun. 2005, 2589–2590.
(5) Khanbabaee, K.; van Ree, T. Synthesis 2001, 1585–1610.
10.1021/ol8009545 CCC: $40.75
Published on Web 05/28/2008
2008 American Chemical Society

Scheme 1. Proposed Retrosynthesis of Sanguiin H-5
Using the copper-mediated oxidation of an organomag-
nesium halide, a model system of the strained medium ring
core of sanguiin H-5 (no hydroxyl substitutions on the aryl
rings) had previously been synthesized.7a We were therefore
keen to test this methodology in the total synthesis of the
more sterically hindered and demanding target (i.e., sanguiin
H-5). As metalation reaction requires aryl iodides, the gallic
acid derivative 5 was needed to facilitate the synthesis of
the precursor.
from D-glucose (15). Using the procedure reported by Murai
and co-workers,10 D-glucose (15) was converted to the
Scheme 2. Synthesis of the Benzoic Acids 5 and 6
Intriguingly, no previous synthesis of the iodide 5 has been
disclosed. Although the bromide 6 could be accessed from
methyl gallate 10 (via initial protection to give 7, bromination
to furnish 11 and subsequent ester cleavage (Scheme 2)8),
neither 7 nor the benzyl ether-protected gallic acid 9 could
be iodinated.9 However, a stepwise approach from 7 was
found to be productive. After the reduction of 7 with LiAlH4,
the resulting alcohol 12 was iodonated in the presence of
iodine and silver trifluoroacetate to yield 13. The iodo alcohol
13 was then converted via a Swern oxidation to the aldehyde
14; a Pinnick oxidation then furnished the desired benzoic
acid 5 (Scheme 2).
Following the synthesis of 5, in order to access the
ꢀ-glucopyranose 4 with stereocontrol at the anomeric center,
a series of protecting group manipulations were performed
alcohol 16 in five steps in an overall yield of 57% (Scheme
3). In a similar method to that used by Feldman and
Lawlor,11 the R-trichloroacetimidate 17 was synthesized from
16 using Cl3CCN and DBU. This allowed the exploitation
(8) Feldman, K. S. Personal communication. Feldman, K. S.; Quideau,
S.; Appel, H. M. J. Org. Chem. 1996, 61, 6656.
(9) (a) Methods examined: (i) Electrophilic iodination: Molander, G. A.;
George, K. M.; Monovich, L. G. J. Org. Chem. 2003, 68, 9533. (b) Orito,
K.; Hatakeyama, T.; Takeo, M.; Suginome, H. Synthesis 1995, 1273. (c)
van Laak, K.; Scharf, H. D. Tetrahedron 1989, 45, 5511(ii) Suzuki’s
oxidative conditions: (d) Suzuki, H. Bull. Chem. Soc. Jpn. 1971, 44, 2871(iii)
Directed ortho-lithiation and iodine quench: (e) Gohier, F.; Mortier, J. J.
Org. Chem. 2003, 68, 2030.
(10) Tanaka, H.; Kawai, K.; Fujiwara, K.; Murai, A. Tetrahedron 2002,
58, 10017–10031.
(11) Feldman, K. S.; Lawlor, M. D., J. Am. Chem. Soc. 2000, 122, 7396–
7397, NB: NaH was used as the base in this reported synthesis.
2594
Org. Lett., Vol. 10, No. 12, 2008

intramolecular cuprate oxidation in the presence of the
oxidant 19,7 gave access to the benzyl-protected sanguiin
H-5 1 (Scheme 4). The reaction proceeded with complete
diastereoselectivity and in good isolated yield (65%); pleas-
ingly no dimer side products were observed. This observed
selectivity for the (S)-atropoisomer in the biaryl coupling
step as has been discussed previously,1c is thought to be a
result of the structural restraints imposed by the galloylated
sugar ring core. Our efforts to determine if this selectivity
was as a result of a kinetic or thermodynamic effects were
unsuccessful; heating the biaryl 1 led to decomposition and
isomerization was not observed.
Scheme 3. Synthesis of the ꢀ-Glucopyranose 4
After demonstrating the effectiveness of the above meth-
odology, the copper catalyzed oxidative organozinc biaryl
coupling was examined.7b The milder conditions allow the
more readily available aryl bromide coupling precursor to
be used in the key C-C bond-forming reaction. Once
synthesized,8 the bromobenzoic acid 6 was coupled to the
ꢀ-glucopyranose 4 to give the coupling precursor 3. After
treatment of 3 with Rieke zinc (Zn*), transmetalation, and
oxidation (as before) facilitated the formation of the cyclized
product 1 in an optimized 70% isolated yield (Scheme 4).
The reaction was highly sensitive to moisture and rigorously
anhydrous reaction conditions were required to minimize the
formation of the debrominated byproduct.
of trichloroacetimidate acylation chemistry to furnish the
desired ꢀ-anomer primarily. Thus, in the presence of the
protected gallic acid 9, the R-trichloroacetimidate 17 was
converted to the ester 18 as a mixture of R and ꢀ-anomers
1
(R/ꢀ ) 1:4 by H NMR analysis) in an 80% yield. The
To complete our total synthesis from the globally protected
compound 1 formed via either of the above routes, a Pd/C
induced hydrogenolytic deprotection followed by filtration
through Celite furnished sanguiin H-5 (Scheme 4). Sanguiin
H-5 was found to be hydrolytically unstable on silica and
alumina, making further purification problematic.12 The
spectroscopic data obtained matched that reported previously
for the natural product.6
desired ꢀ-galloylglucose product 4 was isolated in 72% yield
by flash column chromatography after desilylation (Scheme
3).
In the presence of DCC, DMAP and the benzoic acid 5,
the ꢀ-galloylglucose 4 was converted to the cyclization
precursor 2 via a double esterification at O(2) and O(3) in
an 84% yield. With 2 in hand, the key orangocuprate
oxidative intramolecular biaryl bond-forming reaction was
attempted. Treatment of 2 with isopropylmagnesium chloride,
followed by transmetalation with CuBr·SMe2 and subsequent
These routes based on organocuprate oxidative biaryl bond
formation constitute efficient methods to access sanguiin H-5
Scheme 4. Intermolecular Biaryl Formation and the Synthesis of Sanguiin H-5
Org. Lett., Vol. 10, No. 12, 2008
2595

and potentially other ellagitannins. In one step after either
the initial formation of an organomagnesium or organozinc
intermediate, intramolecular oxidative coupling of the result-
ing diarylcuprate allows diastereoselective and concomitant
biaryl bond and medium ring formation. As a result, this
approach constitutes a robust and general procedure for the
synthesis of natural and unnatural products containing this
strained motif. The successful performance of these meth-
odologies in the total synthesis of sanguiin H-5 represents a
significant advance to complement existing biaryl-coupling
strategies.
Acknowledgment. We thank EPSRC, BBSRC, MRC, and
the Augustus and Harry Newman Foundation for financial
support.
Supporting Information Available: Experimental pro-
cedures and spectral data for all new compounds, including
NMR spectra. This material is available free of charge via
the Internet at http://pubs.acs.org.
(12) Choromatography using reversed-phase (C-18) silica could be used
if required; however, the filtered, concentrated reaction mixture gave material
of ca. 95% purity.
OL8009545
2596
Org. Lett., Vol. 10, No. 12, 2008