Total synthesis of (-)-xyloketal A

Article abstract of DOI:10.1021/ol060266w

The first total synthesis of the C₃-symmetric and biologically active natural product, (-)-xyloketal A, has been accomplished in one step from phloroglucinol (1,3,5-trihydroxybenzene) and (4R)-3-hydroxymethyl-2,4-dimethyl- 4,5-dihydrofuran. Thi

Full text of DOI:10.1021/ol060266w

ORGANIC
LETTERS
2006
Vol. 8, No. 7
1427-1429
Total Synthesis of (−)-Xyloketal A
Jeremy D. Pettigrew and Peter D. Wilson*
Department of Chemistry, Simon Fraser UniVersity, 8888 UniVersity DriVe,
Burnaby, British Columbia V5A 1S6, Canada
pwilson@sfu.ca
Received January 31, 2006
ABSTRACT
The first total synthesis of the C3-symmetric and biologically active natural product, (−)-xyloketal A, has been accomplished in one step from
phloroglucinol (1,3,5-trihydroxybenzene) and (4R)-3-hydroxymethyl-2,4-dimethyl-4,5-dihydrofuran. This remarkably direct process involved an
exceedingly facile and diastereoselective boron trifluoride diethyl etherate-promoted triple electrophilic aromatic substitution reaction that
was coupled to three bicyclic acetal formation reactions.
The isolation and structural characterization of seven closely
related natural products, the xyloketals, from a mangrove
fungus of the Xylaria species have been reported recently
in a series of papers by Lin and co-workers.1-3 Of these
natural products, (-)-xyloketal A (1) has a unique and
aesthetically pleasing C3-symmetric molecular structure that
was elucidated by detailed spectroscopic studies and by X-ray
crystallography (Figure 1).1 In the solid state, it was revealed
that this substance adopts a symmetric bowl-shaped confor-
mation that is presumably reinforced by anomeric effects.
The absolute stereochemistry of this novel secondary me-
tabolite was assigned by interpretation of the CD spectrum.1
(-)-Xyloketal A (1) has also been shown to be a potent
inhibitor of acetylcholine esterase and to have L-calcium
channel blocking activity.1,2 Thus, (-)-xyloketal A (1)
represents an important lead compound for the treatment of
cardio- and cerebrovascular conditions as well as neurological
diseases.1,2 In view of the structure and biological properties
of this particular substance, we have been actively engaged
in studies toward the total synthesis of the xyloketal natural
products.4,5 In this Letter, we report the first total synthesis
of (-)-xyloketal A (1).6
It was realized, before the outset of our earlier synthetic
studies,4,5 that (-)-xyloketal A (1) could be prepared by a
triple electrophilic aromatic substitution reaction of phloro-
glucinol (1,3,5-trihydroxybenzene) 2 and the reactive inter-
mediate 4 (Figure 1). The product of these initial electrophilic
substitution processes would then be expected to react, under
appropriate acidic conditions, to afford the thermodynami-
(4) We have previously reported a racemic synthesis of the two possible
diastereoisomers of a tris-demethyl analogue of xyloketal A and a racemic
synthesis of a demethyl analogue of xyloketal D. We have also reported
the first total synthesis of (()-xyloketal D which is, in a structural sense,
one of the simpler members of this family of natural products, see: (a)
Pettigrew, J. D.; Bexrud, J. A.; Freeman, R. P.; Wilson, P. D. Heterocycles
2004, 62, 445. We have also described an asymmetric total synthesis of
(-)-xyloketal D and its enantiomer, see: (b) Pettigrew, J. D.; Freeman, R.
P.; Wilson, P. D. Can. J. Chem. 2004, 82, 1640. These syntheses featured
the cycloaddition reactions of appropriately functionalized o-quinone
methides and dihydrofurans as a key step. The further application of this
chemistry, to complete the total synthesis of xyloketal A (1), has not been
successful.
(1) Lin, Y.; Wu, X.; Feng, S.; Jiang, G.; Luo, J.; Zhou, S.; Vrijmoed, L.
L. P.; Jones, E. B. G.; Krohn, K.; Steingro¨ver, K.; Zsila, F. J. Org. Chem.
2001, 66, 6252.
(2) Wu, X. Y.; Liu, X. H.; Lin, Y. C.; Luo, J. H.; She, Z. G.; Houjin, L.;
Chan, W. L.; Antus, S.; Kurtan, T.; Elsa¨sser, B.; Krohn, K. Eur. J. Org.
Chem. 2005, 4061.
(3) Wu, X.; Liu, X.; Jiang, G.; Lin, Y.; Chan, W.; Vrijmoed, L. L. P.
Chem. Nat. Compd. 2005, 41, 27.
(5) We have recently described an alternative approach to the synthesis
of the xyloketal A (1) based on a novel phenylboronic acid-mediated triple
condensation reaction of R,â-unsaturated aldehydes and phloroglucinol,
see: Pettigrew, J. D.; Cadieux, J. A.; So, S. S. S.; Wilson, P. D. Org. Lett.
2005, 7, 467.
10.1021/ol060266w CCC: $33.50
© 2006 American Chemical Society
Published on Web 03/01/2006

Scheme 1. Total Synthesis of (-)-Xyloketal A
Figure 1. Molecular structure and retrosynthetic analysis of
xyloketal A.
cally stable cis-fused bicyclic acetal moieties of the natural
product. It was also appreciated that the stereochemistry of
the acetal formation reactions would be directed by the steric
influence of the stereogenic C4-methyl substituent of the
reactive intermediate 4. In principle, this reactive intermediate
could be generated on ionization of the corresponding chiral
nonracemic alcohol 3.7 To demonstrate this synthetic objec-
tive, a method was developed for the preparation of a
precursor to the chiral nonracemic alcohol 3. Thus, the
methyl ester 11 was prepared from the known oxazolidinone
5 (Scheme 1).8
Deprotonation of the oxazolidinone 5 in a mixture of
tetrahydrohydrofuran and hexamethylphosphoramide (HMPA,
∼10%) with lithium diisopropylamide (LDA, 1.5 equiv), and
on subsequent reaction with propargyl bromide (4 equiv) at
-78 °C for 20 h, afforded the oxazolidinone 6 in good yield
(77%) and as a single diastereoisomer.9 This oxazolidinone
was then reduced with lithium aluminum hydride to afford
the known chiral nonracemic alcohol 7.10,11 This acetylenic
alcohol was in turn converted to the known endo-cyclic
dihydrofuran 8 on heating with a substoichiometric amount
of sodium amide followed by thermal isomerization of the
corresponding exo-cyclic dihydrofuran.4b,12
To prepare the required methyl ester 11, the dihydrofuran
8 was reacted (based on literature procedures) with trichlo-
roacetyl chloride and pyridine.13,14 This afforded a readily
separable mixture of the desired trichloroketone 9 (45%
yield) and a substantial quantity of the regioisomeric trichlo-
roketone 10 (46% yield).15 Presumably, the latter compound
was formed via isomerization of the endo-cyclic dihydrofuran
8 to the corresponding exo-cyclic dihydrofuran under the
reaction conditions. To suppress the formation of the
regioisomeric trichloroketone 10, the reaction was repeated
at -78 °C. This resulted in the isolation of the trichloroketone
9 in very good yield (93%). Subsequent methanolysis of the
trichloroketone 9 afforded the methyl ester 11 in excellent
(6) Krohn and co-workers have attempted the synthesis of xyloketal A
(1) from racemic 5-hydroxy-4-methyl-3-methylenepentan-2-one and phlo-
roglucinol. In this instance, the desired tris-adducts were isolated in 6%
yield as a mixture of eight diastereoisomers, see: (a) Krohn, K.; Riaz, M.;
Flo¨rke, U. Eur. J. Org. Chem. 2004, 1261. In this paper, an asymmetric
synthesis of (-)-xyloketal D via a conjugate addition reaction of 2,4-
dihydroxyacetophenone to the (4R)-enantiomer of the aforementioned ketone
was also described. In addition, Krohn and co-workers reported the synthesis
of demethyl analogues of the xyloketal natural products in this paper. For
a preliminary communication of this work, see: (b) Krohn, K.; Riaz, M.
Tetrahedron Lett. 2004, 45, 293.
(7) We have established, in simplified model systems, that this novel
triple electrophilic aromatic substitution reaction and subsequent bicyclic
acetal formation process can be promoted with a variety of acidic reagents
in the presence of anhydrous magnesium sulfate, see: Pettigrew, J. D.;
Wilson, P. D. J. Org. Chem. 2006, 71, 1620.
(12) The final reaction product was contaminated with a small amount
of the corresponding exo-cyclic dihydrofuran (<7%).
(13) (a) Hojo, M.; Masuda, R.; Sakaguchi, S.; Takagawa, M. Synthesis
1986, 1016. (b) Colla, A.; Martins, M. A. P.; Clar, G.; Krimmer, S.; Fischer,
P. Synthesis 1991, 483.
(14) A direct one-step procedure for the synthesis of dihydrofuran esters
from the corresponding dihydrofurans has been reported, see: Stetter, H.;
Lorenz, G. Chem. Ber. 1985, 118, 1115. However, in our hands, a complex
mixture of reaction products was obtained when the dihydrofuran 8 was
employed as a reaction substrate.
(8) Gage, J. R.; Evans, D. A. Org. Synth. 1990, 68, 83.
(9) For reports on the use of HMPA in alkylation reactions of derivatives
of oxazolidinone chiral auxiliaries, see: (a) Fadel, A. Tetrahedron:
Asymmetry 1994, 5, 531. (b) Versteeg, M.; Bezuidenhoudt, B. C. B.;
Ferreira, D.; Swart, K. J. J. Chem. Soc., Chem. Commun. 1995, 1317. (c)
Versteeg, M.; Bezuidenhoudt, B. C. B.; Ferreira, D. Tetrahedron 1999, 55,
3365.
(10) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737.
(11) For an alternative synthesis (employing a resolution procedure) and
proof of absolute stereochemistry of the chiral nonracemic alcohol 7, see
ref 4b.
(15) The trichloroketone 10 was isolated as a single geometrical isomer.
The geometry of the double bond was not determined but it is reasonable
to assume, based on electronic and steric considerations, that it is trans.
1428
Org. Lett., Vol. 8, No. 7, 2006

yield.16 In preliminary experiments, it was found that the
methyl ester 11 could be cleanly reduced to the required
alcohol 3 with lithium aluminum hydride. However, the latter
compound was found to be somewhat unstable to isolation
and purification and so it was used directly in subsequent
experiments. Thus, according to preliminary model studies,
a mixture of a two-fold excess (per phenolic reaction site of
phloroglucinol 2) of the alcohol 3 and phloroglucinol 2 (1
equiv) in ether at 0 °C was stirred with boron trifluoride
diethyl etherate (1 equiv) and anhydrous magnesium sulfate.7
Within 20 min, an exceptionally clean synthetic transforma-
tion occurred that led to the isolation of a chromatographi-
cally inseparable mixture of xyloketal A (1) and 2,6-epi-
xyloketal A (2,6-epi-1) in 85% yield (dr ) 5:2).17,18 Although
the diastereoselectivity of the overall process was not
exceptional, the diastereoselectivity of each of the three
individual ring formation reactions was certainly respectable
(dr ) ∼9:1). Moreover, the 1H NMR spectrum of the crude
reaction mixture was dominated by signals that corresponded
to the C3-symmetric natural product.1 To improve the overall
diastereoselectivity of this remarkable process, the above
reaction was repeated at -78 °C.19 This led to the isolation
of a mixture of xyloketal A (1) and 2,6-epi-xyloketal A (2,6-
epi-1) in similar yield (79% yield). However, the diastereo-
selectivity of the overall process was significantly increased
(dr ) 4:1). In this case, the diastereoselectivity of each of
the three individual ring formation reactions had reached an
impressive level (dr ) ∼19:1). An analytically pure sample
of synthetic (-)-xyloketal A (1) was subsequently obtained
on crystallization from petroleum ether. The 1H and 13C NMR
spectra of this material were identical with those recorded
for the natural product.1 Moreover, the optical rotation (in
both sense and magnitude) as well as the melting point of
the synthetic material were in close agreement with those
reported in the original isolation paper.1 Thus, the absolute
stereochemistry of the natural product is now firmly estab-
lished.
In conclusion, the first total synthesis of (-)-xyloketal A
(1) has been accomplished from phloroglucinol 2 and the
chiral nonracemic alcohol 3. The key step of this total
synthesis involved an exceedingly facile and diastereoselec-
tive boron trifluoride diethyl etherate-promoted triple elec-
trophilic aromatic substitution reaction that was coupled to
three bicyclic acetal formation reactions. An efficient five-
step asymmetric synthesis of (4R)-methyl-2,4-dimethyl-4,5-
dihydrofuran-3-carboxylate (11), the direct precursor of
alcohol 3, was also developed from the readily available
oxazolidinone 5. The application of this potentially biomi-
metic synthetic method toward the total synthesis of the
remaining members of the xyloketal family of natural
products will be reported in due course.
Acknowledgment. We are grateful to the Natural Sci-
ences and Engineering Research Council of Canada (NSERC)
and Simon Fraser University (SFU) for financial support.
We also wish to acknowledge the Canadian Foundation for
Innovation (CFI), the British Columbia Knowledge Develop-
ment Fund (BCKDF), and AstraZeneca Canada, Inc. for
support of our research program. J.D.P. thanks SFU for a
special graduate entrance scholarship, as well as NSERC for
postgraduate scholarships (PGS-A and PGS-D).
(16) For a related procedure, see: de Buyck, L.; de Pooter, H.; Schamp,
N. Bull. Soc. Chim. Belg. 1988, 97, 371.
(17) Herein, the numbering scheme is based on that described by Lin
and co-workers (see ref 1).
(18) The stereochemistry of the single minor diastereoisomer from this
reaction was tentatively assigned as to that of the 2,6-epimer of xyloketal
A. In addition, the diastereomeric ratio of the products from this reaction
was determined by integration of the partially resolved signals in the
Supporting Information Available: General experimen-
tal details, experimental procedures, and full product char-
acterization data for all of the compounds synthesized, as
1
downfield region of the crude H NMR spectrum.
1
well as H and 13C NMR spectra for compounds 6, 9-11,
(19) The reaction was also repeated at 0 °C with 4 equiv of the alcohol
3. In this instance, a mixture of xyloketal A (1) and 2,6-epi-xyloketal A
(2,6-epi-1) was isolated in 14% yield (dr ) 5:2). In addition, no evidence
for the formation of mono- or bis-addition products was recorded. This
reflects the relative instability of the alcohol 3 and the increased reactivity
of the more electron-rich mono- and bis-addition products toward electro-
philic aromatic substitution processes.
1, and 2,6-epi-1. This material is available free of charge
via the Internet at http://pubs.acs.org.
OL060266W
Org. Lett., Vol. 8, No. 7, 2006
1429