Enantioselective syntheses of cryptocarya triacetate, Cryptocaryolone, and cryptocaryolone diacetate

Article abstract of DOI:10.1021/ol0345529

(Matrix presented) The enantioselective syntheses of three natural products from Cryptocarya latifolia have been achieved in 13-15 steps from ethyl sorbate. The route relies upon an enantio- and regioselective Sharpless dihydroxylation and a palladium-cat

Full text of DOI:10.1021/ol0345529

ORGANIC
LETTERS
2003
Vol. 5, No. 11
1959-1962
Enantioselective Syntheses of
Cryptocarya Triacetate, Cryptocaryolone,
and Cryptocaryolone Diacetate
Catherine M. Smith and George A. O’Doherty*
Department of Chemistry, West Virginia UniVersity, Morgantown, West Virginia 26506
george.o’doherty@mail.wVu.edu
Received March 30, 2003
ABSTRACT
The enantioselective syntheses of three natural products from Cryptocarya latifolia have been achieved in 13−15 steps from ethyl sorbate. The
route relies upon an enantio- and regioselective Sharpless dihydroxylation and a palladium-catalyzed reduction to establish the absolute
stereochemistry. The route also relies upon a highly (E)-selective olefin cross-metathesis reaction to form trans-δ-hydroxy-1-enoates. The
resulting δ-hydroxy-1-enoates were subsequently converted into cryptocarya triacetate, cryptocaryolone, and cryptocaryolone diacetate.
The leaves and bark of the South African plant, Cryptocarya
latifolia, have been long sought after for their legendary
magical and medicinal properties.1 These alleged properties
range from the treatment of headaches and morning sickness
to the treatment of cancer, pulmonary diseases, and various
bacterial and fungal infections.1 Motivated by these claims,
van Staden has tested crude extracts of C. latifolia and found
significant activity as cyclooxygenase inhibitors (COX-2/
COX-1).2 In a search to find the molecular origins of these
effects, Horn found a series of related 6-substituted 5,6-
dihydropyran-2-ones in the biologically active hexane and
acetone extracts, including cryptocarya diacetate 1 and
cryptocarya triacetate 2,3 along with two bicyclic pyranone/
polyol structures cryptocaryolone 3 and cryptocaryolone
diacetate 4 (Figure 1).3b The use of activity-guided fraction-
ation led to the discovery of more complex 1,3-polyol/5,6-
dihydropyran-2-one natural products from related trees4,5 such
as passifloricin A and 5,7-bis-epi-passifloricin A.5-7 Due to
the interest in their biological activities, several synthetic
approaches to these molecules have been reported by us8
and others.9
Horn has determined the absolute and relative stereochem-
istry of the cryptocarya acetates 1 and 2 by a combination
of Mosher ester analysis and Rychnovsky 13C NMR/
acetonide analysis.3b Finally, Nakata confirmed their results
by an enantioselective total synthesis of both 1 and 2 via
16- and 24-step routes, respectively, from the (S)-tert-butyl
3-hydroxybutyrate.10,11 While the absolute and relative ste-
reochemistries for both cryptocaryolone 3 and cryptocary-
(1) Sam, T. W.; Yeu, C. S.; Jodynis-Liebert, J.; Murias, M.; Bloszyk, E.
Planta Med. 2000, 66, 199.
(2) Zschocke, S.; van Staden, J. J. Ethnopharmacol. 2000, 71, 473.
(3) (a) Drewes, S. E.; Schlapelo, B. M.; Horn, M. M.; Scott-Shaw, R.;
Sandor, O. Phytochemistry 1995, 38, 1427. (b) Collett, L. A.; Cavies-
Coleman, M. T.; Rivett, D. E. A.; Drewes, S. E.; Horn, M. M. Phytochem-
istry 1997, 44, 935.
(4) Andrianaivoravelona, J. O.; Sahpaz, S.; Terreaux, C.; Hostettmann,
K.; Stoecki-Evans, H.; Rasolondramanitra, J. Phytochemistry 1999, 52, 265.
Figure 1.
10.1021/ol0345529 CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/03/2003

olone diacetate 4 have been assumed on the basis of the
structure of crytocarya triacetate, no structural proof has been
offered. Only the relative stereochemistry between C-7 and
C-9 and the relative stereochemistry of the bicyclic portion
of the cryptocaryolones are known.
thuslactone and cryptocarya diacetate 1), we found that the
benzylidene-protected syn-ethyl-3,5-dihydroxyhexanoate 6
was a useful precursor.8a Accordingly, 6 was easily prepared
from ethyl sorbate 7 in either optically enriched form in four
steps and ∼36% yield.
To determine the exact pharmacological role of the polyol
and pyranone portions of these molecules, we have endeav-
ored to prepare this class of natural products as well as both
stereoisomeric and homologous analogues. An important
element to our synthetic strategy is the use of asymmetric
catalysis to establish the initial asymmetry in addition to the
use of efficient diastereoselective reactions to install the
remaining stereochemistry.12 Herein, we describe our suc-
cessful implementation of this strategy toward the syntheses
of the all-syn 1,3-polyol containing natural products cryp-
tocarya triacetate 2, cryptocaryolone 3, and cryptocaryolone
diacetate 4 (Figure 1).
Scheme 2
Scheme 1
Turning our attention to the stereochemically more com-
plex natural products 2-4, we envisioned a protected all-
syn bis-benzylidene-protected tetraol intermediate 8, which
contained all the stereochemical information needed for their
synthesis. On the basis of our previous results, we believed
that the second dioxane ring in 8 could be prepared by the
Evans acetal-forming reaction from 9, which in turn could
be prepared by a cross-metathesis reaction between ethyl
acrylate and 5. Previously we have found 5 to be easily and
enantioselectively prepared from ethyl sorbate in seven steps
and 25% yield. Thus, the problem was reduced to an efficient
synthesis of trans-δ-hydroxy-1-enoates 9.14,15 Herein, we
describe our approach to the synthesis of these key building
blocks via an efficient asymmetric and diastereoselective
reaction sequence.
Inspired by the diverse range of structurally related 1,3-
polyol-substituted 5,6-dihydropyran-2-ones, we have been
interested in the development of practical and concise
enantioselective approaches to syn-1,3-polyol functional-
ities.13 As part of our successful efforts toward the total
synthesis of the diol and triol natural products (tarchonan-
Typically, a three-step protection/oxidative cleavage/Wittig
reaction sequence was used for the homologation of ho-
moallylic alcohol 5 to 9.16 We anticipated that a much simpler
procedure would result from a transition metal-mediated
cross-metathesis coupling reaction of homoallylic alcohols
and acrylates providing trans-δ-hydroxy-1-enoates (i.e., 10
plus 11 to yield 12).17 Previously, Crowe has shown that
the Schrock molybdenum catalyst (Mo(CHCMe2Ph) (NAr)
[OCMe(CF3)2]2) cross couples acrylonitrile and protected
homoallylic alcohols to give good yields of a ∼5:1 cis:trans
ratio of double-bond isomers.18 More recently, Grubbs has
(5) (a) Echeverri, F.; Arango, V.; Quinones, W.; Torres, F.; Escobar,
G.; Rosero, Y.; Archbold, R. Phytochemistry 2001, 56, 881. (b) Herz, W.;
Ramakrishnan, G. Phytochemistry 1978, 17, 1327.
(6) (a) Jodynis-Liebert, J.; Murias, M.; Bloszyk, E. Planta Med. 2000,
66, 199. (b) Meyer, B. N.; Ferrigni, N. R.; Putnam, J. E.; Jacobsen, L. B.;
Nicholson, D. E. Planta Med. 1982, 45, 31.
(7) Drewes, S. E.; Schlapelo, B. M.; Horn, M. M.; Scott-Shaw, R.;
Sandor, O. Phytochemistry 1995, 38, 1427.
(8) (a) Hunter, T. J.; O’Doherty, G. A. Org. Lett. 2001, 3 (17), 2777.
(b) Garaas, S.; Hunter, T. J.; O’Doherty, G. A. J. Org. Chem. 2002, 67 (8),
2682.
(9) (a) Nakata, T.; Hata, N.; Iida, K.; Oishi, T. Tetrahedron Lett. 1987,
28, 5661. (b) Mori, Y.; Suzuki, M. J. Chem. Soc., Perkin Trans. 1 1990,
1809. (c) Mori, Y.; Kageyama, H.; Suzuki, M. Chem. Pharm. Bull. 1990,
38, 2574. (d) Solladie´, G.; Gressot-Kempf Tetrahedron: Asymmetry 1996,
7 (8), 2371. (e) Jorgensen, K. B.; Suenaga, T.; Nakata, T. Tetrahedron Lett.
1999, 40, 8855. (f) Reddy, M. V. R.; Yucel, A. J.; Ramachandran, P. V. J.
Org. Chem. 2001, 66, 2512. (g) Gosh, A. K.; Bilcer, G. Tetrahedron Lett.
2000, 41, 1003. (h) Boger, D. L.; Ichikawa, D.; Zhong, W. J. Am. Chem.
Soc. 2001, 123, 4161. (i) Reddy, M. V. R.; Rearick, J. P.; Hoch, N.;
Ramachandran, P. V. Org. Lett. 2001, 3, 19. (j) Smith, A. B.; Brandt, B.
M. Org. Lett. 2001, 3, 1685.
(10) Jorgensen, K. B.; Suenaga, T.; Nakata, T. Tetrahedron Lett. 1999,
40, 8855.
(11) Aldrich Chemical Co. provides (S)-tert-butyl 3-hydroxybutyrate at
the cost of $41/mL.
(12) At the outset of this project, we set the goal for efficiency as three
steps/stereocenter of the target molecule.
(13) Hunter, T. J.; O’Doherty, G. A. Org. Lett. 2001, 3 (17), 1049. Also
see, ref 8.
(14) We have found that for the successful implementation of the Evans
acetal-forming reaction, the δ-hydroxy-1-enoates must be in their trans form
to prevent lactonization.
(15) For vinylogous aldol approaches to δ-hydroxy-1-enoates, see: (a)
Fleming, I. Bull. Soc. Chem. Fr. 1981, 2, 7. (b) Barloy-Da Silva, C.;
Benkouider, A.; Pale, P. Tetrahedron Lett. 2000, 41, 3077. (c) Albaugh-
Robertson, P.; Katzenellenbogen, J. A. J. Org. Chem. 1983, 48, 5288. For
aldol/Wittig approaches, see ref 16 and: (d) Keck, G. E.; Palani, A.;
McHardy, S. F. J. Org. Chem. 1994, 59, 3113. (e) Solladie, G.; Gressot,
L.; Colobert, F. Eur. J. Org. Chem. 2000, 357.
(16) For examples of the Wittig approach to trans-δ-hydroxy-1-enoates
from homoallylic alcohols, see: Diez-Martin, D.; Kotecha, N. R.; Ley, S.
V.; Mantegani, S.; Menendez, C. J. Tetrahedron 1992, 48 (37), 7899.
1960
Org. Lett., Vol. 5, No. 11, 2003

presence of 14 (2-5%) for 16 h. These conditions consis-
tently gave good yields (55-90%) of enoates 12a-g.21
Unfortunately, we found that the Weinreb amide of acrylic
acid failed to cross-couple with 10a under these same
conditions.22
Scheme 3
Having established the viability of cross-coupling homo-
allylic alcohols 10 and acrylates 11 to form δ-hydroxy enoate
12, we turned our attention to the synthesis of 2-4 from
homoallylic alcohol syn-5. Previously, we have demonstrated
that syn-5 can be prepared from either 12g (Scheme 4), which
Scheme 4
shown that ω-hydroxy-R-olefins smoothly undergo the cross-
coupling reaction with various acrylic esters and amides.19
Table 1a.
mol % reaction
yield
E/Z ratio
of 12
10
11
12
of 14
time (h) of 12
1
10a
11a
11a
11a
11a
11b
11b
11b
12a
12b
12c
12d
12e
12f
12g
2
2
2
5
5
2
2
16
16
24
16
24
16
24
70%
65%
83%
90%
55%
63%
75%
20:1
20:1
10:1
>25:1
>20:1
>20:1
>20:1
2b 10b
3
4
5
6
7
10c
10d
10b
10d
10e
in turn was prepared from 10e (Scheme 3), or in an
asymmetric fashion from ethyl sorbate 7 (Scheme 1).8
a Typical reaction conditions were performed with a 1:2 ratio of 10 to
11 at room temperature and in a 0.2 M benzene solution. b This reaction
Applying the optimized cross-coupling conditions to the
more complex homoallylic alcohol syn-5 and ethyl acrylate
11b (Scheme 5) provided near quantitative yields of 9 in a
was run in a 0.2 M CH2Cl2 solution.
To test the feasibility of a cross-metathesis sequence with
homoallylic alcohols, we decided to start with 4-phenyl-
butenol 10c, which can be prepared in a racemic form from
benzaldehyde and allyl-Grignard. Our initial reactions using
Grubbs’ bis-tricyclohexylphosphine catalyst 13 were entirely
unsuccessful. For example, when a benzene solution of 10a
and 11a (0.2 M, 1:5 ratio of 10a to 11a) was stirred in the
presence of 5% 14 at room temperature for 4 days, no
indication of the formation of enoate 12a was observed.
Similarly, no product was detected after heating an identical
solution at 80 °C for 3 days. Success was achieved upon
switching to the new, more reactive Grubbs’ phosphine/
carbene ligand catalyst 14. We found that using carbene 14
for the cross-coupling reaction of 10a and 11a provided a
good yield of enoate 12a with excellent double-bond
control.20 Typical reaction conditions involved stirring a
room-temperature solution (0.2 M in benzene) of a 1:2 ratio
of homoallylic alcohol (10a-d) and acrylate (11a-b) in the
Scheme 5
20:1 trans:cis ratio of double-bond isomers (96%). The final
stereocenter of cryptocarya triacetate was installed by expos-
ing 9 to the Evans acetal-forming reaction. Reacting the
δ-hydroxy enoate 9 with 4 equiv of benzaldehyde and a
catalytic amount of KOt-Bu led to a 55% yield of the bis-
(17) As this methodology was being completed, Cossy reported similar
cross-metathesis acetyl-protected homoallylic alcohols and enoates; see: (a)
BouzBouz, S.; Cossy, J. Org. Lett. 2001, 3, 1451. (b) Cossy, J.; Bargiggia,
F.; BouzBouz, S. Org. Lett. 2003, 5, 459. For a related example, see: (c)
Dreher, S. D.; Leighton, J. L. J. Am. Chem. Soc. 2001, 123, 341.
(18) Crowe, W. E.; Goldberg, D. R. J. Am. Chem. Soc. 1995, 117, 5162.
(19) (a) Chatterjee, A. K.; Morgan, J. P.; Scholl, M.; Grubbs, R. H. J.
Am. Chem. Soc. 2000, 122, 3783. (b) Choi, T. L.; Chatterjee, A. K.; Grubbs,
R. H. Angew. Chem., Int. Ed. 2001, 40, 1277.
(20) All levels of double-bond selectivity were determined by 1H NMR
analysis (at 500 MHz).
(21) All new compounds were identified and characterized by 1H NMR,
13C NMR, FTIR, HRMS, and/or elemental analysis.
(22) This result was somewhat surprising given that Grubbs et al. have
shown that ω-hydroxy-R-olefins smoothly undergo the cross-coupling
reaction with the Weinreb amide of acrylic acid; see ref 19.
Org. Lett., Vol. 5, No. 11, 2003
1961

benzylidene-protected tetraol 8 along with 23% recovered
starting material.23
Scheme 7
With the establishment of an enantioselective route to the
bis-benzylidene acetal 8, we decided to address the instal-
lation of the pyranone functionality of cryptocarya triacetate
(Scheme 6).24 To this end, we planned to convert ester 8
Scheme 6
7). In fact, allowing the mixture of 18/19 to stir longer in
the TsOH/benzene solution (16 instead of 3 h) gave high
conversions of the bicyclic natural product 3. Thus, a 44%
yield of cryptocaryolone 3 was isolated after the two-step
aqueous acetic acid, TsOH/benzene procedure. As reported
by Drewes et al. in their isolation paper,3 3 was easily
peracylated with excess Ac2O/Pyr to give excellent yields
of cryptocaryolone diacetate 4 (97%).
Comparison of the spectral data of synthetic 3, 5, and 6
with the reported spectral data (IR, 1H NMR, and 13C NMR
spectra) for cryptocarya triacetate, cryptocaryolone, and
cryptocaryolone diacetate showed complete agreement.3
Thus, for the first time, both the absolute and relative
stereochemistries of cryptocaryolone and cryptocaryolone
diacetate were established as drawn in Scheme 7.
In conclusion, three short and enantioselective syntheses
of three natural products isolated from C. latifolia have been
presented. The route to cryptocarya triacetate is significantly
shorter than the previous approach, requiring only 13 steps
instead of 24 steps.9 The approach also allows for the first
time the syntheses of two related natural products, crypto-
caryolone and cryptocaryolone diacetate, which occurred
with similar efficiency (13 and 14 steps, respectively). This
highly enantio- and diastereocontrolled route illustrates the
utility of our recently developed use of Grubbs cross-
metathesis and Os/Pd route to benzylidene-protected syn-
1,3-diols. The syntheses provide cryptocarya triacetate in
11% overall yield and cryptocaryolone and cryptocaryolone
diacetate both in 4% overall yields.
into the cis-enoate 17, which possesses all of the desired
carbon atoms and the correct double-bond stereochemistry
of 2. Aldehyde 15 was easily prepared in an excellent yield
(90%) by exposure of a THF solution of ester 8 with 1.1
equiv of DibalH at -78 °C. Treatment of 15 with the
potassium salt of 16 and 18-crown-6 afforded a 75% yield
of 17 with a 4:1 double-bond cis:trans stereoselectivity.25
Next, we looked to investigate the potential for a depro-
tection and lactonization reaction sequence to convert enoate
17 to pyranone 19 and not bicyclic lactone 3 (Schemes 6
and 7). In the course of our tarchonanthuslactone and
cryptocarya diacetate 1 syntheses, we found that the ben-
zylidene protecting group could be easily removed by
refluxing in 80% aqueous acetic acid. Exposing 17 to these
conditions for ∼1 h gave primarily tetraol 18 with a small
amount of lactone 19. In vacuo removal of solvent followed
by the addition of benzene and 1% TsOH gave good
conversion to triol 19 with minimal amounts of 3 after only
3 h. Because of our concerns with bicycle formation, crude
19 was peracylated with excess Ac2O/Pyr and catalytic
DMAP providing good yields of cryptocarya triacetate.
Not surprisingly, a similar acid-catalyzed deprotection
strategy was used to prepare the cyrptocaryolones (Scheme
Acknowledgment. We thank both the Arnold and Mabel
Beckman Foundation and the National Institute of General
Medical Sciences (1R01 GM63150-01A1) for their generous
support of our program.
Supporting Information Available: Complete experi-
mental procedures and spectral data for all new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(23) Evans, D. A.; Gauchet-Prunet, J. A. J. Org. Chem. 1993, 58, 2446.
(24) For other approaches to pyranone synthesis, see refs 8b and 9d
and: (a) Hayakawa, H.; Miyashita, M. Tetrahedron Lett. 2000, 41, 707.
(b) Tosaki, S.-Y.; Nemoto, T.; Ohshima, T.; Shibasaki, M. Org. Lett. 2003,
5, 495.
(25) Still, W. C.; Gennari, C. Tetrahedron Lett. 1983, 24, 4405.
OL0345529
1962
Org. Lett., Vol. 5, No. 11, 2003