Enantioselective total synthesis of the antitumor macrolide rhizoxin D

Article abstract of DOI:10.1021/jo034011x

The convergent, highly enantioselective synthesis of rhizoxin D, a natural product possessing potent antitumor and antifungal bioactivity, is described. The C(1)-C(9) fragment of the molecule was synthesized utilizing a threefold pseudosymmetric intermediate ultimately derived from γ-butyrolactone. The central core of rhizoxin D was prepared via a chiral resolution/asymmetric aldol protocol. Several methods for the generation of the polyene fragment were explored, and the side-chain was ultimately prepared from serine in six steps. The unification of the left and right wings of the molecule was achieved using a one-step olefination protocol, and the macrocyclization was carried out using a Horner-Emmons olefination at the C(2)-C(3) olefin.

Full text of DOI:10.1021/jo034011x

En a n tioselective Tota l Syn th esis of th e An titu m or Ma cr olid e
Rh izoxin D
J ennifer A. Lafontaine, David P. Provencal, Cristina Gardelli, and J ames W. Leahy*,1
Department of Chemistry, University of California, Berkeley, California 94720-1460
jleahy@exelixis.com
Received J anuary 5, 2003
The convergent, highly enantioselective synthesis of rhizoxin D, a natural product possessing potent
antitumor and antifungal bioactivity, is described. The C(1)-C(9) fragment of the molecule was
synthesized utilizing a threefold pseudosymmetric intermediate ultimately derived from γ-buty-
rolactone. The central core of rhizoxin D was prepared via a chiral resolution/asymmetric aldol
protocol. Several methods for the generation of the polyene fragment were explored, and the side-
chain was ultimately prepared from serine in six steps. The unification of the left and right wings
of the molecule was achieved using a one-step olefination protocol, and the macrocyclization was
carried out using a Horner-Emmons olefination at the C(2)-C(3) olefin.
In tr od u ction
In 1984, Iwasaki and co-workers reported the isolation
of a novel family of macrolactones that possess potent
biological activity.2 Rhizoxin (1) represents the most
extensively studied of these novel macrolides isolated
from the fungus Rhizopus chinensis, the causal agent of
rice seedling blight. In the original structure elucidation,
rhizoxin was cited as possessing potent antifungal activ-
ity. Since this time, several studies have established that
this alkaloid also exhibits other important medicinal
properties; it has been shown to have profound antitumor
activity, as well as antibiotic and antimicrobial activity.3
The discovery of rhizoxin was soon followed by the
isolation of several close structural analogues,4 including
rhizoxin D (2, Figure 1), which is believed to be a
biogenetic precursor of 1. Studies have shown that this
desepoxy homologue possesses bioactivity and antitumor
activities equivalent to those of rhizoxin, demonstrating
that the epoxide moieties are not needed for in vitro
activity.3b It seems probable that in vivo hydrolysis of the
epoxides may be responsible to some degree for the very
F IGURE 1.
short half-life of rhizoxin in the body, making compound
2 an intriguing potential therapeutic agent. In the years
since rhizoxin was isolated, several workers have pub-
lished progress toward its synthesis,5,6 and Ohno and co-
(1) Current address: Exelixis Pharmaceuticals, Inc., 170 Harbor
Way, P.O. Box 511, South San Francisco, CA 94083-0511.
(2) (a) Iwasaki, S.; Kobayashi, H.; Furukawa, J .; Namikoshi, M.;
Okuda, S. J . Antibiot. 1984, 37, 354. (b) Iwasaki, S.; Namikoshi, M.;
Kobayashi, H.; Furukawa, J .; Okuda, S.; Itai, A.; Kasuya, A.; Iitaka,
Y.; Sato, Z. J . Antibiot. 1986, 39, 424.
(3) (a) Tsuruo, T.; Oh-hara, T.; Iida, H.; Tsukagoshi, S.; Sato, Z.
Izumi, M.; Iwawsaki, S.; Okuda, S.; Shimizu, F.; Sasagawa, K.; Fukami,
M.; Fukuda, K.; Arakawa, M. Cancer Res. 1986, 46, 381. (b) Takahasi,
M.; Iwasaki, S.; Kobayashi, H.; Murai, T.; Sato, Y.; Haraguchi-Hiraoka,
T.; Nagano, H. J . Antibiot. 1987, 66. (c) Sullivan, A. S.; Prasad, V.;
Roach, M. C.; Takahashi, M.; Iwasaki, S.; Luduena, R. F. Cancer Res.
1990, 50, 4277. (d) Bisset, D.; Setanoians, A.; Chadwick, G. A.; Wilson,
P.; Koier, I.; Henrar, R.; Schwartsmann, G.; Cassidy, J .; Kaye, S. B.;
Kerr, D. J . Cancer Res. 1992, 52, 2894.
(4) (a) Iwasaki, S.; Namikoshi, M.; Kobayashi, H.; Furukawa, J .;
Okuda, S. Chem. Pharm. Bull. 1986, 34, 1387. (b) Kiyoto, S.; Kawai,
Y.; Kawakita, T.; Kino, E.; Okuhara, M.; Uchida, I.; Tanaka, H.;
Hashimoto, M.; Terano, H.; Kohsaka, M.; Aoki, H.; Imanaka, H. J .
Antibiot. 1986, 39, 762.
(5) (a) Boger, D. L.; Curran, T. J . Org. Chem. 1992, 57, 2235. (b)
Rao, A. V. R.; Sharma, G. V. M.; Bhanu, M. N. Tetrahedron Lett. 1992,
33, 3907. (c) Rao, A. V. R.; Bhanu, M. N.; Sharma, G. V. M. Tetrahedron
Lett. 1993, 34, 707. (d) Keck, G. E.; Park, M.; Krishnamurthy, D. J .
Org. Chem. 1993, 58, 3787. (e) Rao, S. P.; Murthy, V. S.; Rao, A. V. R.
Indian J . Chem. 1994, 33B, 820. (f) Keck, G. E.; Savin, K. A.; Weglarz,
M. A.; Cressman, E. N. K. Tetrahedron Lett. 1996, 37, 3291. (g) White,
J . D.; Nylund, C. S.; Green, N. J . Tetrahedron Lett. 1997, 38, 7329. (h)
White, J . D.; Holoboski, M. A.; Nylund, C. S.; Green, N. J . Tetrahedron
Lett. 1997, 38, 7333. (i) Burke, S. D.; Hong, J .; Mongin, A. P.
Tetrahedron Lett. 1998, 39, 2239. (j) Burke, S. D.; Hong, J .; Lennox,
J . R.; Mongin, A. P. J . Org. Chem. 1998, 63, 6952. (k) Davenport, R.
J .; Regan, A. C. Tetrahedron Lett. 2000, 41, 7619. (l) N’Zoutani, M.-
A.; Pancrazi, A.; Ardisson, J . Synlett 2001, 769.
10.1021/jo034011x CCC: $25.00 © 2003 American Chemical Society
Published on Web 04/19/2003
J . Org. Chem. 2003, 68, 4215-4234
4215

Lafontaine et al.
SCHEME 1
SCHEME 2
SCHEME 3 a
workers have successfully completed the task.7 In addi-
tion, six groups, including ours, have reported the suc-
cessful synthesis of 2.8 In this work, we describe our
enantioselective total synthesis of rhizoxin D.
Resu lts a n d Discu ssion
Retr osyn th etic An a lysis. As we began our retrosyn-
thetic analysis of rhizoxin D, we imagined a number of
possible cyclization strategies that could give rise to the
macrocycle, as shown in Scheme 1. We initially consid-
ered two complementary macrolactonization alternatives
(disconnection a): a standard macrolactonization wherein
activation of the carboxyl group would lead to retention
of configuration at the eventual C15 position and a
Mitsunobu macrolactonization.9 The latter cyclization,
which occurs via activation of the alcohol component and
inversion of configuration at C15, would provide an
opportunity to access the central portion of the molecule
in a much more facile manner. Given the potential
difficulties often associated with macrolactonizations, we
considered two additional macrocyclization strategies.
Specifically, both the C9-C10 and C2-C3 olefins (dis-
connections b and c, respectively) could potentially be
formed while closing the macrocycle. These options
seemed ideal since all were accessible without neces-
sitating a complete redesign of the synthesis. Instead,
a
Reaction conditions: (a) KOH, BnBr (98%); (b) EtOCOCl, Et3N
(88%); (c) NaHMDS, allyl bromide (73%); (d) LiBH4 (76%); (e) TsCl,
Et3N, DMAP (97%); (f) I2, PPh3, imidazole (70%).
simply rearranging the order in which the transforma-
tions would be conducted would allow us to redirect our
efforts. In practice, this flexibility proved to be valuable,
as we ultimately explored all of these options (vide infra).
Ma cr ola cton iza tion : In itia l Syn th etic Str a tegy.
Because the macrolactonization strategy represented two
distinctly different possibilities that would ideally require
very little resynthesis of the intermediates, we chose to
pursue this option at the outset. We thus embarked on a
synthetic approach to the requisite building blocks.
Our route to the C1-C9 portion of rhizoxin D was
based on a versatile synthetic intermediate, 7 (Scheme
2). The structural similarity in each of the three “arms”
of 7 about C5 gives the synthesis a degree of flexibility,
because any of the three arms could be converted into
one particular carbon chain in the natural product. In
this way, the intermediate can be mapped onto 6, the
appropriate intermediate for macrolactonization, in three
different ways by simply rotating about the C5 stereo-
center. Compound 7 would be prepared starting from
γ-butyrolactone.
(6) (a) Provencal, D. P.; Gardelli, C.; Lafontaine, J . A.; Leahy, J . W.
Tetrahedron Lett. 1995, 36, 6033. (b) Lafontaine, J . A.; Leahy, J . W.
Tetrahedron Lett. 1995, 36, 6029.
(7) (a) Nakada, M.; Kobayashi, S.; Shibasaki, M.; Iwasaki, S.; Ohno,
M. Tetrahedron Lett. 1993, 34, 1039. (b) Nakada, M.; Kobayashi, S.;
Iwasaki, S.; Ohno, M. Tetrahedron Lett. 1993, 34, 1035. (c) Nakada,
M. Yakugaku Zasshi 1997, 117, 486.
(8) (a) Kende, A. S.; Blass, B. E.; Henry, J . R. Tetrahedron Lett. 1995,
36, 4741. (b) Williams, D. R.; Werner, K. M.; Feng, B. Tetrahedron
Lett. 1997, 38, 6825 (c) Lafontaine, J . A.; Provencal, D. P.; Gardelli,
C.; Leahy, J . W. Tetrahedron Lett. 1999, 40, 4145. (d) Keck, G. E.;
Wager, C. A.; Wager, T.; Savin, K. A.; Covel, J . A.; McLaws, M. D.;
Krishnamurthy, D.; Cee, V. J . Angew. Chem., Int. Ed. 2001, 40 (1),
231. (e) Mitchell, I. S.; Pattenden, G.; Stonehouse, J . P. Tetrahedron
Lett. 2002, 43, 493 (f) White, J . D.; Blakemore, P. R.; Green, N. J .;
Hauser, E. B.; Holoboski, M. A.; Keown, L. E.; Nylund Kolz, C. S.;
Phillips, B. W. J . Org. Chem. 2002, 67 (22), 7750.
The synthesis of 6 began with the opening of γ-buty-
rolactone with potassium hydroxide in the presence of
benzyl bromide to afford the corresponding benzyl ether
8 (Scheme 3).10 The carboxylic acid was activated as the
mixed anhydride and added to the anion of oxazolidinone
(10) Szczepankiewicz, B. G.; Heathcock, C. H. Tetrahedron 1997,
53, 8853.
(9) Mitsunobu, O. Synthesis 1981, 1.
4216 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
SCHEME 4
SCHEME 5 a
9. The ketoimide formed (10) was then alkylated dia-
stereoselectively with allyl bromide, giving product 11.11
After removal of the chiral auxiliary to give alcohol 7,
we sought to extend the alcohol to give vinylstannane
15, which could eventually serve as the R,â-unsaturated
ester of our target. Alcohol 7 was therefore converted to
the tosylate (12) or the iodide (13), and both of these
compounds were explored as potential electrophiles.
Unfortunately, neither the tosylate nor the iodide was
displaced by the lithium anion of vinyl stannane 14. A
variety of cuprates derived from 14 were also prepared,
but all suffered from a lack of reactivity toward our
system.
a
Reaction conditions: (a) Dess-Martin oxidation (100%). (b)
Ph3PdCHOMe (91%); HCO2H (91%). (c) Bu2BOTf, i-Pr2EtN (80%).
(d) LiBH4 (99%). (e) BPSCl, imidazole, DMF (97%). (f) BnOC-
(NH)CCl3, TfOH (79%). (g) OsO4, NMO; NaIO4 (100%). (h)
EtO2CCH2PO(OEt)2, DBU, LiCl (80%). (i) HF, CH3CN (100%).
This setback was quickly addressed by re-examining
our approach in light of our versatile intermediate, 7
(Scheme 4). We devised a modified retrosynthesis that
was again based on this key intermediate; by remapping
this alcohol onto the right wing we could elaborate the
R-hydroxy carbon of 7 not as C4 of the natural product
but as C6. The first requirement of this revised approach
was the extension of the C6 arm of 7. Thus, oxidation to
the aldehyde (16),12 followed by a Wittig olefination/acidic
hydrolysis sequence, led to homologated aldehyde 17
(Scheme 5).13 After screening a large number of acids to
effect this hydrolysis of the methyl enol ether, we found
that formic acid in ether and water gave consistently high
yields of the desired aldehyde.14 This compound was now
employed in a syn-aldol reaction, which introduced the
C7 and C8 chiral centers in a completely stereospecific
manner.15 Reductive removal of the chiral auxiliary with
LiBH411 gave a diol that could be selectively protected at
the primary hydroxyl (21). Benzyl protection of the
remaining alcohol was accomplished with the use of
benzyl trichloroacetimidate16 in the presence of catalytic
triflic acid to provide the fully protected intermediate 22
in good yield.
SCHEME 6 a
a
Reaction conditions: (a) MsCl, Et3N, DMAP (96%). (b) NaSAr,
DMF, 65 °C; Mo(VI), H2O2 (80%, two steps).
ester 24. At this point, it remained only to convert the
silyl-protected ether to a benzothiazole sulfone, which
would be used for a J ulia olefination later in the
synthesis. Removal of the BPS protecting group pro-
ceeded smoothly, and the resulting alcohol was converted
to the corresponding sulfone as shown in Scheme 6. We
found the use of an ammonium molybdate-based oxida-
tion,19 in particular, was critical to providing the sulfone
in good yield. This transformation marked the completion
of the right wing (C1-C9) coupling partner of rhizoxin
D, so we then turned our efforts to the left-hand portion
(C10 to C26).
Oxidation of the terminal olefin to give aldehyde 23
was accomplished via a dihydroxylation/periodate cleav-
age sequence,17 and a Horner-Emmons olefination18 gave
(11) Evans, D. A.; Black, W. C. J . Am. Chem. Soc. 1993, 115, 4497.
(12) (a) Dess, D. B.; Martin, J . C. J . Org. Chem. 1983, 48, 4156. (b)
Dess, D. B.; Martin, J . C. J . Am. Chem. Soc. 1991, 113, 7277. (c)
Ireland, R. E.; Liu, L. J . Org. Chem. 1993, 58, 2899.
(13) Welch, S. C.; Chou, C.; Gruder, J . M.; Assercq, J . J . Org. Chem.
1985, 50, 2668.
(14) J enneskens, L. W.; Krul, A. H. P.; Kraakman, P. A.; Moene,
W.; de Wolf, W. H.; Bickelhaupt, F. J . Org. Chem. 1986, 51, 2162.
(15) Gage, J . R.; Evans, D. A. Org. Synth. 1989, 68, 83.
(16) Wessel, H.; Iversen, T.; Bundle, D. R. J . Chem. Soc., Perkin
Trans. 1 1985, 2247.
We felt from the outset that the best course of action
would be to perform any macrocyclization on a fully
(17) (a) J acobsen, E. N.; Marko, I.; Mungall, W. S.; Schroder, G.;
Sharpless, K. B. J . Am. Chem. Soc. 1988, 110, 1968. (b) Branalt, J .;
Kvarnstrom, I.; Svensson, S.; Classon, B.; Samuelsson, B. J . Org. Chem.
1994, 59, 4430.
(18) Blanchette, M. A.; Choy, W.; Davis, J . T.; Essenfield, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183.
J . Org. Chem, Vol. 68, No. 11, 2003 4217

Lafontaine et al.
SCHEME 7
SCHEME 9 a
a
Reaction conditions: (a) TBSCl, Et3N (100%). (b) O3; PPh3
(100%). (c) PPh3dC(CH3)CHO (84%). (d) NaOH, H2O2 (92%, 88%
ee). (e) NaH, BnBr, Bu4NI (96%). (f) O3 (1.3 equiv); PPh3 (57%).
SCHEME 8 a
equiv of aldehyde 38. Wittig olefination with 2-(triphen-
ylphosphoranylidene)propionaldehyde gave the R,â-
unsaturated aldehyde 39 as a single isomer. Using the
chemistry of Brown,23 39 was allylated to yield the enan-
tioenriched (88% ee) alcohol 40, which was then benzy-
lated to provide diene 41. Selective ozonolysis of the
terminal olefin was accomplished by the careful addition
of 1 equiv of ozone to afford aldehyde 42.
a
Although this selective ozonolysis gave a reasonable
recovery of the desired product, it was operationally
difficult and gave mixtures of starting material, product,
and over-oxidized material. While we investigated alter-
natives to this transformation, we were unable to ad-
equately improve on these conditions. In addition, we had
noticed that the Brown asymmetric allylation, which had
reliably given 41 in high ees on a smaller scale, gave
diminished enantioselectivity on the large scale required
for our efforts. These two issues led us to explore
alternative routes to aldehyde 42. We believed that we
could find a solution to both of these issues in the use of
acetate aldol chemistry. Though we did explore asym-
metric variants, the most selective methods required very
low reaction temperatures (-115 °C) and were therefore
not ideally suited for scale-up. Instead, we turned to the
use of a kinetic resolution of an achiral aldol product.
We were aware that such a route would have the obvious
disadvantage that half of our material would be sacri-
ficed, but we believed that this would be a reliable route
to useful quantities of essentially enantiopure material,
at a stage sufficiently early in the synthesis to be worth
the compromise.
Reaction conditions: (a) ethyl acetimidate HCl, Et3N (80%);
(b) CuBr2, DBU, HMTA (85%); (c) LiAlH4; (d) Swern oxidation
(61%, two steps); (e) Ph3PdC(CH3)CHO (91%); (f) CHl3, CrCl2
(58%).
elaborated substrate in order to make the synthesis as
convergent as possible. In that regard, we initially chose
to construct the C10-C26 portion such that the triene
portion would be made as late as possible, and we opted
for a Negishi-type carbometalation20 strategy in the hopes
of avoiding undesired olefin isomers (Scheme 7). We also
chose a synthetic route that would require inversion of
the C15 stereocenter in the Mitsunobu macrocyclization
step, as this would allow for facile access to the central
core through the reliable and precedented syn aldol
reaction (vide infra).
The synthesis of the oxazole side chain began from
serine methyl ester, which was treated with ethyl ace-
timidate to cleanly afford the corresponding oxazoline
(31, Scheme 8). Subsequent oxidation with cupric bro-
mide21 produced the oxazole in excellent yield. The
methyl ester was then reduced to the aldehyde in a two-
step protocol, and Wittig olefination gave the homolo-
gated aldehyde as a single isomer. Finally, a Takai22
reaction on 35 produced our Negishi coupling partner,
vinyl iodide 28.
The substrate to be resolved was prepared by treat-
ment of aldehyde 39 with the sodium enolate of methyl
acetate (Scheme 10), giving racemic alcohol 43 in good
yield on a 50 g scale. This racemic aldol product was then
treated with standard Sharpless asymmetric epoxidation
conditions, but in the presence of only 0.6 equiv of the
Turning to the synthesis of the central core, we began
from cis-2-buten-1,4-diol (36, Scheme 9), which was bis-
silylated and then subjected to ozonolysis to afford 2
(19) Bellingham, R.; J arowicki, K.; Kocienski, P.; Martin, V. Syn-
thesis 1996, 285.
(20) Van Horn, D. E.; Negishi, E. J . Am. Chem. Soc. 1978, 100, 2252.
(21) Barrish, J . C.; Singh, J .; Spergel, S. H.; Han, W. C.; Kissick, T.
P.; Kronenthal, D. R.; Mueller, R. H. J . Org. Chem. 1993, 58, 4494.
(22) Takai, K.; Nitta, K.; Utimoto, K. J . Am. Chem. Soc. 1986, 108,
7408.
(23) J adhav, P. K.; Bhat, K. S.; Perumal, P. T.; Brown, H. C. J . Org.
Chem. 1986, 51, 432.
4218 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
SCHEME 10 a
SCHEME 11 a
a
Reaction conditions: (a) MeOAc, NaHMDS, THF, -78 °C
(78%); (b) (-)-DIPT, Ti(OiPr)4, TBHP (40%); (c) benzyl trichloro-
acetimidate, TfOH (76%) or TIPSOTf, 2,6-lutidine (99%); (d)
DIBAL-H, 1 equiv.
stoichiometric oxidant, tert-butylhydroperoxide (TBHP).24
The reaction was quenched when approximately 55% of
the olefinic material had been consumed (as judged by
1H NMR), and we were able to recover 40% of the desired
homochiral 43. Using both Mosher’s ester analysis25 and
chiral shift reagents, we were unable to detect any
undesired isomer. Importantly, this product was readily
converted to aldehyde 42, which allowed us to avoid the
difficult ozonolysis. Ultimately, we found it necessary to
modify our protecting group strategy (vide infra), and this
was accomplished simply by silylation of the resolved
alcohol to provide 46 in good overall yield.
a
Reaction conditions: (a) Bu2BOTf, i-Pr2NEt, 18 (89%); (b)
MeO(Me)NH, Me3Al (95%); (c) ethynylmagnesium bromide (66%);
(d) Et2BOMe, NaBH4 (84%); (e) NaH, MeI (83%); (f) PMBOC-
(NH)CCl3, TfOH (70%).
SCHEME 12
With a facile route to the large-scale preparation of the
desired aldehyde, we turned to the installation of the C15
and C16 chiral centers. To this end, a syn Evans aldol15
reaction was performed on 42, giving a product with the
(desired) unnatural stereochemistry at C15 (Scheme 11).
This hydroxyl would be inverted in the planned Mit-
sunobu macrocyclization. Oxazolidinone 47 was con-
verted to the Weinreb amide,26 and an acetylenic unit
was introduced via the Grignard reagent. The resulting
ketone was reduced to the syn-1,3-diol using diethyl-
methoxyborane.27 We were pleased to find that selective
methylation of the propargylic C17 hydroxyl was surpris-
ingly facile in the presence of the free C15 hydroxyl.
Protection of that remaining alcohol as the p-methoxy-
benzyl ether gave 29, the substrate to be coupled to the
oxazole side chain.
Unfortunately, we found that attempted carbometala-
tion of the alkyne followed by treatment with the vinyl
iodide side chain failed to produce triene 52, even under
forcing conditions (Scheme 12). Attempts to add various
cuprates to the alkyne to form a vinyl cuprate were also
unsuccessful. The difficulty in these reactions may be the
result of the propargylic methoxy substituent present in
29. Though there is precedent for carboaluminations on
similar substrates,28 propargylic substituents appear to
make addition to the alkyne more difficult.29
With the failure of the Negishi coupling, we redesigned
our approach to the installation of the triene. While we
investigated a series of alternatives for the generation
of this sensitive region of the molecule, ultimately only
one proved to be fruitful. In our revised approach, we
planned to employ a Horner-Emmons olefination to
unite the oxazole side chain 53 with a ketophosphonate
central core (54, Scheme 13). A good portion of the
chemistry we had already developed could be employed
without the added risk of using Lewis acidic metals on
highly functionalized intermediates.
The revised synthesis of oxazole fragment 53 was
straightforward from previously prepared intermediate
35 (Scheme 14). Horner-Emmons olefination afforded
ethyl ester 55, which was reduced with DIBALH to the
corresponding allylic alcohol 56. This unstable alcohol
(24) Gao, Y.; Hanson, R. M.; Klunder, J . M.; Ko, S. Y.; Masamune,
H.; Sharpless, K. B. J . Am. Chem. Soc. 1987, 109, 5765.
(25) Ohtani, I.; Kusumi, T.; Kashman, Y.; Kakisawa, H. J . Am.
Chem. Soc. 1991, 113, 4092.
(26) Basha, A.; Lipton, M.; Weinreb, S. M. Tetrahedron Lett. 1977,
4171.
(28) (a) Negishi, E.; Van Horn, D. E.; Yoshida, T. J . Am. Chem. Soc.
1985, 107, 6639 (b) Rand, C. L.; Van Horn, D. E.; Moore, M. W.;
Negishi, E. J . Org. Chem. 1981, 46, 4093.
(29) (a) Normant, J . F.; Cahiez, G.; Bourgain, M.; Chuit, C.; Villieras,
J . Bull. Soc. Chim. Fr. 1974, 1656. (b) Marfat, A.; McGuirk, P. R.;
Helquist, P. Tetrahedron Lett. 1978, 19, 1363.
(27) Chen, K.-M.; Hardtmann, G. E.; Prasad, K.; Repic, O.; Shapiro,
M. J . Tetrahedron Lett. 1987, 28, 155.
J . Org. Chem, Vol. 68, No. 11, 2003 4219

Lafontaine et al.
SCHEME 13
SCHEME 15 a
SCHEME 14 a
a
Reaction conditions: (a) (EtO)2P(O)CH2CO2Et, LiCl, DBU,
CH3CN (93%); (b) DIBAL-H (97%); (c) Dess-Martin oxidation
(75%).
was quickly carried into the following oxidation step to
afford aldehyde 53, which proved to be a stable, white,
crystalline solid that could be stored for several months
without significant decomposition.
a
Reaction conditions: (a) TMSCl, Et3N (100%). (b) DIBAL-H
(93%). (c) (EtO)2P(O)Et, t-BuLi (89%); Dess-Martin oxidation
(97%). (d) 53, Ba(OH)2‚8H2O (65%). (e) citric acid (89%). (f)
Et2BOMe, NaBH4 (76%). (g) NaH, MeI (63%). (h) PNBCl, pyr.
(82%). (i) TBAF (87%). (j) Dess-Martin oxidation (87%).
With aldehyde 53 in hand, we turned to the synthesis
of the revised central core fragment, starting from
Weinreb amide 48 (Scheme 15). We were surprised to
note that the anion of ethyl diethylphosphonate could not
be added directly into 48 under any conditions. We
postulated that the remaining alcohol functionality was
interfering with the reactivity of the Weinreb amide.
However, simple protection of the alcohol as a trimeth-
ylsilyl ether still did not allow for the addition to occur.30
Thus, we were forced to prepare the more reactive
aldehyde 58 as an intermediate, which underwent smooth
addition of the phosphonate. Subsequent oxidation af-
forded the requisite diastereomeric mixture of â-keto-
phosphonates 54 in two steps.
Efforts were next directed toward the Horner-Em-
mons olefination in order to introduce the oxazole side
chain. Roush-Masamune31 conditions failed to give 59,
but the use of barium hydroxide in aqueous THF worked
well to install the triene.32 Importantly, only the desired
trans-olefin product was observed. With the triene se-
curely in place, the trimethylsilyl group was readily
removed with mild citric acid to reveal the C15 alcohol
60. This alcohol was crucial for the introduction of the
remaining stereocenter at C17. From the unnatural
orientation at C15 in 60, we required a chelation-
controlled reduction of the ketone in order to generate
the desired syn-1,3-diol 61, which Prasad’s method
readily provided.27 The allylic alcohol was then selectively
methylated, and the remaining C15 alcohol was protected
as the corresponding p-nitrobenzoate (PNB) ester 63.
Liberation of the TBS-protected alcohol with tetrabutyl-
ammonium fluoride was followed by Dess-Martin oxida-
tion to give aldehyde 27, which constitutes the entire
C10-C26 fragment of rhizoxin, epimeric at only the
macrolactonization center.
With the left and right wing coupling partners, 6 and
27, respectively, in hand, we moved on to the installation
of the C9-C10 olefin to unite the entire skeleton of our
target. Early plans to employ a traditional J ulia olefi-
nation33 were redirected when yields of coupled products
in model systems were consistently low. We instead
(30) Addition of the anion of methyl dimethylphosphonate to a
Weinreb amide is known: Maugras, I.; Poncet, J .; J ouin, P. Tetrahe-
dron 1990, 46, 2807. There are no reports of the use of ethyl
diethylphosphonate in this context, however.
(31) (a) Blanchette, M. A.; Choy, W.; Davis, J . T.; Essenfeld, A. P.;
Masamune, S.; Roush, W. R.; Sakai, T. Tetrahedron Lett. 1984, 25,
2183. (b) Rathke, M. W.; Nowak, M. J . J . Org. Chem. 1985, 50, 2624.
(32) Paterson, I.; Yeung, K.; Smaill, J . B. Synlett 1993, 774.
(33) (a) J ulia, M.; Paris, J . M. Tetrahedron Lett. 1973, 4833. (b)
Kocienski, P. Phosphorus Sulfur 1985, 24, 97.
4220 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
SCHEME 16 a
SCHEME 17
the result of steric hindrance at the hydroxyl center, can
typically be avoided by changing the reaction conditions
or by using a bulkier azodicarboxylate. Extensive efforts
to optimize the cyclization were undertaken. The choice
of solvent (benzene, toluene, THF), diazocarboxylate
(DEAD, DIAD, ADDP), and order of addition of reagents
was varied, but compounds such as 67 were the major
products under all protocols examined.
There are other macrocyclization protocols that occur
with inversion at the hydroxyl such as displacement of a
mesylate with a cesium carboxylate salt. These methods,
however, require activation of the alcohol prior to depro-
tection of the acid. Our approach, which released the
alcohol and the acid in the same step, could not accom-
modate such procedures. It therefore seemed that mac-
rocyclization with inversion of configuration was not a
viable option in our case without significant revision of
our efforts. However, we had initially anticipated this
potential problem. We were cognizant that the C15
stereocenter could alternatively be controlled by the
appropriate choice of protective conditions. Compound 62
would serve as the lynchpin for this plan. We had already
demonstrated that standard esterification conditions led
to the product with unnatural stereochemistry at C15 (as
in Scheme 15). Similarly, we envisioned that protection
under Mitsunobu conditions would yield the epimer with
the natural configuration (68, Scheme 17). Much to our
chagrin, treatment of 62 with p-nitrobenzoic acid under
a variety of conditions failed to afford any of the desired
product. In fact, only unreacted starting materials and
a mixture of elimination products were returned. We also
undertook efforts to invert the C15 stereocenter via an
oxidation/reduction protocol. While a variety of methods
were examined in this pursuit, no better than a 5:2 ratio
of isomers favoring the desired product could be obtained.
We were thus forced to install the desired stereochem-
istry at C15 at a much earlier juncture in our synthesis.
Ma cr ola cton iza tion : Revised Syn th etic Str a tegy.
With an efficient route to the C15 episeco acid already
developed, it was our great desire to utilize as much of
the work in place to complete our goal. We therefore set
out to prepare the requisite J ulia precursor with the
natural (S)-stereochemistry at C15 (Scheme 18). We
already had a tractable source of 53 in hand. We hoped
that 70 could be generated by building on our previous
success, the critical difference being that an efficient anti-
aldol reaction was needed to install the natural stereo-
chemistry.
a
Reaction conditions: (a) LiHMDS, 27 (50%); (b) TMSO-K+
(69%); (c) DIAD, PPh3.
examined the use of a modified J ulia olefination protocol,
first introduced by Sylvestre J ulia in 1991.34 This pro-
cedure utilizes not a phenyl sulfone but a benzothiazole
sulfone and represents a marked improvement over the
traditional reaction because it requires only one synthetic
operation to access the olefin, as opposed to three discrete
steps. In addition, we found the modified olefination
reaction to be milder and more tolerant of diverse
functionality than the three-step protocol, as it was
compatible with both the R,â-unsaturated ester of 6 and
the sensitive triene moiety of 27.
Treatment of a solution of sulfone 6 and aldehyde 27
with base led to the exclusive formation of 65 (Scheme
16). Compound 65 represents the complete (C1-C26)
carbon skeleton of our target, and from this compound
we could access the seco acid required for cyclization in
one step. Traditional saponification protocols gave sig-
nificant decomposition of our substrate, but we found that
potassium trimethylsilanolate effectively cleaved both the
ethyl ester and PNB ester in one step to afford 66.35
Attempts to close the ring through a Mitsunobu reaction
were then undertaken. Unfortunately, our efforts were
consistently plagued by amide formation, the result of
azodicarboxylate addition into the phosphine-activated
carboxylate of our substrate. This problem, most likely
A second-generation synthesis of the left wing, then,
began from aldehyde 46, which was studied extensively
as the substrate for an anti-aldol reaction (Scheme 19).
Our initial efforts met with little success, as the protocols
of Paterson,36 Heathcock,37 Duthaler,38 and Oppolzer39 all
failed to give the desired aldol adduct in good yield and/
(34) Baudin, J . B.; Hareau, G.; J ulia, S. A.; Ruel, O. Tetrahedron
Lett. 1991, 32, 1175.
(35) Laganis, E. D.; Chenard, B. L. Tetrahedron Lett. 1984, 25, 5831.
J . Org. Chem, Vol. 68, No. 11, 2003 4221

Lafontaine et al.
SCHEME 18
SCHEME 20
SCHEME 19 a
a
Combined yield of both isomers, after separation by chroma-
tography.
well on our substrate, allowing us to install the C15 and
C16 stereocenters with excellent diastereoselectivity (90%
de) and in good yield. Protection of the resultant free hy-
droxyl, followed by reductive removal of the chiral auxili-
ary afforded 74. The alcohol was then oxidized to the cor-
responding aldehyde, which was converted in two steps
to the â-ketophosphonate 70. As before, a barium hy-
droxide-mediated Horner-Emmons olefination was used
to unite the phosphonate with the oxazole side chain (53),
giving rise to a single isomer at the newly formed olefin.
With all of the carbons again in place for the revised
C10-C26 fragment, we were poised for what seemed to
be the straightforward elaboration to 69 following the
pathway previously developed with the epimer. However,
the seemingly trivial differences between these two
substrates created several difficult situations that needed
to be addressed. Removal of the C15 TMS protecting
group of 77 was the first example. Many of the typical
deprotection conditions led to the formation of both 78
and an undesired side product (Scheme 20). This con-
taminant was shown by both one- and two-dimensional
NOE studies to be 79, the product of isomerization at
the C22-C23 olefin. We found, however, that this
isomerization could be minimized by employing fluoro-
silicic acid in isopropyl alcohol at reduced temperature
to effect the deprotection.41 Importantly, neither of the
other silyl protecting groups were affected under these
conditions, allowing access to 78 in high yield.
a
Reaction conditions: (a) Et3N (81%); (b) TMSCl, imidazole
(86%); (c) DIBAL-H (100%); (d) Dess-Martin oxidation (90%); (e)
(EtO)2P(O)Et, t-BuLi (100%); (f) Dess-Martin oxidation (96%); (g)
53, Ba(OH)2‚H2O (85%).
or diastereoselectivity. In particular, we found that loss
of the primary silyl protecting group or elimination of
the protected hydroxyl â to the carbonyl were recurring
problems.
Fortunately, during the course of our studies, Masam-
une and co-workers introduced novel aldol methodology
that utilizes a norephedrine-derived auxiliary (71) to give
anti-aldol products.40 This new approach worked quite
We were now in position to install the remaining
stereocenter at C17. While we knew that it would be
possible to effect a directed reduction in order to continue
along our previously elaborated pathway, we were also
aware of another unexpected consequence of the inver-
sion at C15. As a result of our previous synthetic efforts,
we were able to cull sufficient amounts of 1,3-anti-diol
80 from byproducts of our main route to perform model
(36) Paterson, I; Wallace, D. J .; Velazquez, S. M. Tetrahedron Lett.
1994, 35, 9083.
(37) Walker, M. A.; Heathcock, C. H. J . Org. Chem. 1991, 56, 5747.
(38) Duthaler, R. O.; Hafner, A.; Alsters, P. L.; Bold, G.; Rihs, G.;
Rothe-Streit, P.; Wyss, B. Inorg. Chim. Acta 1994, 222, 95.
(39) Oppolzer, W.; Starkemann, C.; Rodriguez, I.; Bernardinelli, G.
Tetrahedron Lett. 1991, 32, 61.
(40) Abiko, A.; Liu, J .; Masamune, S. J . Am. Chem. Soc. 1997, 119,
2586.
(41) Pilcher, A. S.; DeShong, P. J . Org. Chem. 1993, 58, 5130.
4222 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
SCHEME 21 a
SCHEME 23 a
a
Reaction conditions: (a) TIPSCl, imidazole, DMAP (95%). (b)
a
Reaction conditions: (a) MeI, NaH.
Na/NH3 (98%). (c) Ag2CO3/Celite (86%). (d) DIBAL (100%). (e)
MeOH, Amberlyst 15 (87%). (f) TBAF (89%). (g) MsCl, NEt3. (h)
ArSNa (90%). (i) Mo(VI), H2O2 (91%). (j) OsO4, NMO; NaIO4 (83%).
(k) EtO2CCH2PO(OEt)2, LiCl, DBU (85%).
SCHEME 22 a
Selective removal of the TBS protecting group in the
presence of the TIPS group proceeded smoothly with
fluorosilicic acid, and oxidation to the R,â-unsaturated
aldehyde 86 completed the synthesis of the C10-C26
portion of rhizoxin D.
At this point, we were poised to couple aldehyde 86 to
the right wing of the molecule via the J ulia olefination.
Because the modified J ulia reaction had proven to be
tolerant of such diverse functionality, we decided to
further explore the limits of the reaction by using a more
advanced sulfone coupling partner, 91 (Scheme 23). In
this new synthon, the six-membered ring is installed
prior to coupling, allowing for a more convergent syn-
thesis. In addition, the presence of the ring would remove
some of the free rotation associated with our original
macrocyclization precursor, more closely approximating
the true seco acid of the natural product. The modified
synthesis of the sulfone began from 20, which was
silylated and debenzylated to afford diol 88 in excellent
yield. Fetizon oxidation44 smoothly gave the correspond-
ing δ-lactone 89. Partial reduction of the lactone and acid-
catalyzed addition of methanol gave the acetal as a
mixture of unseparable diastereomers. The primary
alcohol was then liberated and, following mesylation,
converted into the requisite benzothiazolyl sulfone 90.
Oxidation of the olefin and Horner-Emmons olefination
gave the pyran acetal analogue of 6. To our delight, the
J ulia coupling of 86 and 91 was very successful, providing
a good yield of material that once again contained all of
the carbons necessary for the natural product. Saponi-
fication of both of the esters with potassium trimethyl-
silanolate as before gave the seco acid, now with the
natural stereochemistry at the C15 position. Unfortu-
a
Reaction conditions: (a) p-nitrobenzaldehyde, SmI2 (83%); (b)
MeI, Ag2O, ultrasound (77%); (c) H2SiF6, 4:1 CH3CN/t-BuOH
(94%); (d) Dess-Martin oxidation (82%).
studies germane to our goal. Surprisingly, although the
methylation of 61 had proceeded readily to give exclu-
sively the desired product (Scheme 15), the natural
isomer failed to undergo selective alkylation (Scheme 21).
While we felt that this would bode well for the reactivity
at C15 in terms of our eventual macrocyclization, we
needed a way to distinguish between C15 and C17.
The intramolecular Tishchenko reaction proved to be
a valuable transformation in our efforts. Treatment of
78 with p-nitrobenzaldehyde and samarium diiodide gave
exclusively allylic alcohol 83 (Scheme 22).42 Methylation
under strongly basic conditions led to intramolecular
scrambling of the ester, but we were able to introduce
the methyl group by using methyl iodide and silver
oxide.43 While this reaction proceeded very slowly under
the standard conditions, we found that the rate increased
dramatically if the reaction mixture was sonicated.
(42) Evans, D. A.; Hoveyda, A. H. J . Am. Chem. Soc. 1990, 112, 6447.
(43) Finch, N.; Fitt, J . J .; Hau, I. H. S. J . Org. Chem. 1975, 40, 206.
(44) (a) Girault, J . P.; Dana, G. Tetrahedron Lett. 1970, 4135. (b)
Fetizon, M.; Golfier, M. C. R. Acad. Sc. Paris (C) 1968, 267, 900. (c)
McKillop, A.; Young, D. W. Synthesis 1979, 401.
J . Org. Chem, Vol. 68, No. 11, 2003 4223

Lafontaine et al.
SCHEME 24 a
SCHEME 25 a
a
Reaction conditions: (a) HF, MeCN (95%). (b) MsCl, NEt3,
DMAP (96%). (c) ArSNa (94%). (d) Mo(VI), H2O2 (91%). (e) DIBAL
(80%). (f) TBSCl, imidazole (85%). (g) OsO4, NMO; NaIO4 (87%).
SCHEME 26 a
a
Reaction conditions: (a) 86, LiHMDS (81%); (b) TMSO-K+
(65%).
nately, all attempts to effect this macrolactonization were
unsuccessful, leading to decomposition or recovered
starting material. The triene portion of the molecule
seemed particularly susceptible to decomposition when
exposed to a variety of cyclization conditions. Fearing
that steric hindrance at C15 and the sensitivity of the
triene functionality made this approach untenable, we
made the strategic decision to focus on other approaches
to closing the macrocycle.
Ma cr ocycliza tion via Olefin a tion : Rin g Closu r e
a t C9-C10. With the disappointing outcome of the first
two cyclization strategies, we were forced to examine the
alternative possibilities for closing the macrocycle. Of our
remaining options, we were most intrigued by the pos-
sibility of closing the ring via a J ulia olefination at
C9-C10 (disconnection b, Scheme 1). This option repre-
sented a novel macrocyclization strategy that would occur
at a site remote from our earlier failed efforts. Such a
plan would also require minimal re-engineering of our
earlier synthetic efforts. Since we were constantly striv-
ing to improve on our earlier work, we chose to explore
methods that would remove the isomeric mixtures of
acetals that had resulted from our cyclization of the
δ-lactone. In that regard, desilylation of 89 followed by
elaboration of the terminus to the aryl sulfone using the
same protocol as before gave 96 (Scheme 25). Partial
reduction again gave the hemiacetal, and we were
pleased to see that silylation of this lactol gave 98
exclusively as the anomer in which the silyl group is the
least sterically encumbered. Oxidative cleavage of the
olefin then afforded aldehyde 99.
a
Reaction conditions: (a) DIBAL (77%); (b) (EtO)2POCH2COCl,
pyr. (80%); (c) 99, LiCl, DBU (83%); (d) H2SiF6, tert-amyl alcohol
(88%); (e) Dess-Martin oxidation (72%).
tions, which gave 102 in good yield. The newly formed
olefin was generated exclusively in the desired trans
geometry.
The other component of our J ulia macrocyclization
strategy also required some synthetic manipulation. The
ester of 84 was reductively cleaved, and acylation with
diethylphosphonoacetyl chloride45 gave 101 (Scheme 26).
With the two coupling partners in hand, we exposed 99
and 101 to standard Horner-Emmons olefination condi-
(45) (a) Cooke, M. P.; Biciunas, K. P. Synthesis 1981, 283. (b)
Haseltine, J . N.; Cabal, M. P.; Mantlo, N. B.; Iwasawa, N.; Yamashita,
D. S.; Coleman, R. S.; Danishefsky, S. J .; Schulte, G. K. J . Am. Chem.
Soc. 1991, 113, 3850.
4224 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
SCHEME 27 a
The next challenge involved conversion of the primary
TBS ether of 102 to the R,â-unsaturated aldehyde
required for the J ulia olefination. This matter was
complicated by the fact that the deprotection of the C10
hydroxyl had to be carried out in the presence of our TBS
acetal. In fact, exposure of 102 to typical conditions such
as HF‚pyridine led to approximately equal amounts of
the desired product (103) as well as the corresponding
hemiacetal. Critically, no material that retained the
primary silyl ether was obtained. Returning to conditions
that had been successful on the central core synthesis,
we found that the use of fluorosilicic acid in t-amyl alcohol
gave excellent results. DeShong has studied the use of
this acid for selective silyl deprotections46 and has found
that the best selectivity is obtained using bulky alcohols
(such as t-BuOH) as the solvent. We found that t-amyl
alcohol has the added advantage that it could be cooled
to -15 °C, which, in our studies, conferred an additional
degree of selectivity to the deprotection.
The final step prior to cyclization involved oxidation
of the primary hydroxyl of 103 to the aldehyde required
for the J ulia. This reaction proceeded smoothly with the
Dess Martin periodinane,47 and we turned to the mac-
rocyclization. Sulfone 104 was subjected to the standard
reactions conditions used for the intermolecular J ulia
coupling, but the material isolated from this reaction was
not the cyclized product. Along with significant decom-
position, a compound that lacked the aldehyde moiety,
but still possessed the sulfone, was recovered from the
reaction. This compound, to which we were unable to
assign a complete structure, had also undergone migra-
tion of the C2-C3 olefin, likely the result of deprotona-
tion at C4. It was clear that treatment of 104 with base
had not led to the desired results and had instead caused
degradation of other functionality in our starting mate-
rial.
Ma cr ocycliza t ion via Olefin a t ion : Cou p lin g a t
C2-C3. With the failure of our J ulia macrocyclization
tactic, we were left with the one strategy we had yet to
explore: cyclization via Horner-Emmons olefination at
C2-C3 (disconnection c, Scheme 1). This plan would not
require manipulation of aldehyde 86, but we were once
again faced with the prospect of tweaking our sulfone
coupling partner (Scheme 27). To this end, we reduced
99 and protected the resultant alcohol as triethylsilyl
ether 106. We then turned to the J ulia olefination with
aldehyde 86, which proceeded well to couple the right
and left portions of our molecule in improved yields
compared with previous substrates. As usual, only the
desired trans-olefin geometry was observed. Subse-
quently, the lone ester was reductively cleaved to expose
the C15 alcohol.
a
Reaction conditions: (a) NaBH4, MeOH (88%); (b) TESCl,
Et3N, DMAP, (96%); (c) 86, LiHMDS (2.3 equiv) (79%); (d)
DIBAL-H (90%).
primary alcohol led us to the familiar position of having
a macrocyclization precursor. We began our cyclization
studies by employing Roush-Masamune Horner-Em-
mons conditions. These conditions had worked well
during our previous efforts to form the C2-C3 olefin, so
we imagined that they would be compatible with this
system. When we treated phosphonate 111 to standard
reactions conditions, however, we found that we gratify-
ingly obtained cyclized material, but only in low yield.
We therefore turned to a protocol that had previously
proven to be successful for the installation of the triene:
the use of barium hydroxide in aqueous THF. Running
this reaction at high dilution and with 20 equiv of barium
hydroxide, we found that this method reliably provided
the desired macrocycle, 112. Because this procedure
employs hydroxide to effect the Horner-Emmons, a side
product of the reaction resulted from saponification of
the acyl phosphonate to give the free C15 hydroxyl. Thus,
the closure of the ring proceeded in moderate yield, but
we were nonetheless elated to have the macrocycle in
hand.
We were finally within grasp of our target, rhizoxin D
(2). We continued the synthesis through selective removal
of the TBS acetal protecting group in the presence of a
TIPS ether with a very concentrated solution of HF‚
pyridine. The Dess-Martin reaction that had been so
reliable throughout the synthesis worked well to oxidize
hemiacetal 113 on a very small scale, but on scale-up,
the reaction gave only acetylation of the hemiacetal; this
side reaction was the result of a slow oxidation and the
presence of adventitious acetic acid in the reagent. The
use of catalytic TPAP with NMO as the oxidant,48
however, gave satisfactory results and afforded lactone
114 in reasonable yield.
We were able to introduce the phosphonate ester at
C15 using a procedure similar to the one previously
optimized for 101 (Scheme 28). Now faced with the
selective deprotection of compound 109, we returned to
the reagent that had given excellent results at other
points in our synthesis. We found that, once again,
fluorosilicic acid (this time in 2-propanol at -40 °C)
cleaved our TES protecting group without any observed
loss of the TBS group. Dess Martin oxidation of this
The final transformation required to access 2 was
removal of the TIPS protecting group at C13. We first
employed standard HF‚CH3CN deprotection conditions
(46) Pilcher, A. S.; DeShong, P. J . Org. Chem. 1993, 58, 5130.
(47) Dess, D. B.; Martin, J . C. J . Am. Chem. Soc. 1991, 113, 7277.
(48) Griffith, W. P.; Ley, S. V.; Whitcombe, G. P.; White, A. D. J .
Chem. Soc., Chem. Commun. 1987, 1625.
J . Org. Chem, Vol. 68, No. 11, 2003 4225

Lafontaine et al.
SCHEME 28 a
Con clu sion s
Despite the many detours encountered in reaching our
target, we believe that our strategy makes a positive
contribution to the existing synthetic knowledge about
this natural macrolide. Several aspects of our approach
differ significantly from other published efforts and
represent a nice complement to those endeavors. First,
by introducing the oxazole side chain much earlier than
in any of the other approaches, our synthesis provides
insight regarding the compatibility of the triene with a
wide variety of chemical transformations, while also
making the synthesis more convergent. In addition, the
problem of setting the four stereocenters of the C10-C19
portion of rhizoxin was successfully addressed in our
approach, largely thanks to the use of anti-aldol meth-
odology recently developed by Masamune. Finally, our
synthesis benefits from a highly efficient method for the
installation of the C9-C10 olefin: a modified, one-step
J ulia reaction allows us to assemble the framework of
our target in high yield.
Exp er im en ta l Section
(3S)-[2-(Ben zyloxyeth yl)]-5-h exen a l (17). A suspension
of (methoxymethyl)-triphenylphosphonium chloride (66.6 g,
194 mmol) in THF (300 mL) was cooled to -78 °C, and
NaHMDS (1 M in THF, 188 mL, 188 mmol) was added to the
mixture dropwise. The resulting orange solution was stirred
for 1 h, and then aldehyde 16 (6.50 g, 29.0 mmol) in THF (100
mL) was added via syringe. After stirring for an additional 2
h at -78 °C, the mixture was allowed to warm to room
temperature for 30 min, and then the reaction was quenched
by the addition of saturated aqueous NaHCO3. The material
was extracted with Et2O, and the combined organic extracts
were dried over MgSO4, filtered, and concentrated in vacuo.
This crude material was then run through a plug of silica gel
in order to separate nonpolar impurities, and the resulting
mixture of two enol ethers ((E)- and (Z)-olefin isomers) was
then carried directly into the following hydrolysis reaction.
The enol ethers from the above reaction (7.6 g, 29 mmol)
were dissolved in 110 mL of Et2O. To this solution were added
7.5 mL of water and 75 mL of formic acid, sequentially. The
mixture was stirred at room temperature for 10 h, brought to
neutral pH with 3 N NaOH, and extracted with Et2O. The
combined organic extracts were dried over MgSO4, filtered, and
concentrated to afford 5.9 g (82% from aldehyde 16) of
homologated aldehyde 17: [R]23D +2.74 (c 0.4, CHCl3); IR (thin
a
Reaction conditions: (a) (EtO)2POCH2COCl, pyr. (92%); (b)
1
film) 2919, 2859, 2760, 1728 cm-1; H NMR δ 1.49 (m, 1H),
H2SiF6, 2-propanol, -40 °C (88%); (c) Dess-Martin oxidation
(78%); (d) Ba(OH)2, THF, H2O (49%); (e) HF‚pyridine (80%); (f)
TPAP, NMO, 3 Å sieves (61%); (g) TBAF (73%).
1.61 (m, 1H), 1.94 (m, 1H), 2.06 (m, 1H), 2.18 (m, 1H), 2.29
(m, 2H), 3.40 (t, 2H, J ) 6.3 Hz), 4.37 (s, 2H), 4.93 (m, 2H),
5.63 (m, 1H), 7.21 (m, 5H), 9.62 (t, 1H, J ) 1.9 Hz); 13C NMR
δ 30.0, 33.8, 38.7, 47.9, 67.9, 72.9, 117.3, 127.58, 127.64, 128.4,
135.9, 138.3, 202.6. Anal. Calcd for C15H20O2: C, 77.55; H, 8.68.
Found: C, 77.57; H, 8.74.
to this end and were disappointed to find rapid and
complete decomposition of our substrate. Using a much
weaker solution of the same reagent did afford the
desired product, but in only trace quantities and ac-
companied by two minor side products. Similar results
were observed when the experiment was run in a
polypropylene reaction vessel or when fluorosilicic acid
was employed. As a final effort, we turned to TBAF, a
reagent we were reluctant to try because its basicity
had contributed to unsatisfactory results at earlier points
in the synthesis. We were excited to find, though, that
this reaction was extremely facile and much cleaner than
the acidic methods we had tested. The protocol gave a
73% yield of 2. The synthetic material compared favor-
ably in all respects with data provided for the natural
product.4a
[4S,3(2S,3R,5S)]-4-Ben zyl-3-(5-a llyl-6-ben zyloxy-3-h y-
d r oxy-2-m eth yl-1-h ep ta n oyl)-2-oxa zolid in on e (19). (4S)-
3-Propionyl-4-(benzyl)-2-oxazolidinone (5.13 g, 22.0 mmol) (18)
was dissolved in CH2Cl2 (55 mL) and cooled to 0 °C. Bu2BOTf
(4.8 mL, 22.2 mmol) was then added via syringe, dropwise.
After the mixture was stirred for 15 min, i-Pr2EtN (4.22 mL,
24.2 mmol) was added to the solution, which was stirred for
an additional 40 min. The mixture was then cooled to -78 °C,
and aldehyde 17 (5.64 g, 24.2 mmol) in CH2Cl2 (70 mL) was
added via cannula over 10 min. The resulting solution was
stirred at -78 °C for 3 h, and the reaction was then quenched
by the addition of 35 mL of pH 7 buffer and 45 mL of MeOH.
The mixture was allowed to warm to room temperature, and
an additional 50 mL of MeOH and 35 mL of 30% H2O2 were
added. After stirring vigorously for 3 h, the mixture was
4226 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
allowed to stand overnight. The volatiles were then removed
in vacuo, and the resulting paste was dissolved in saturated
aqueous NaHCO3 and extracted with Et2O. The combined
organic extracts were washed with brine, dried over MgSO4,
filtered, and concentrated. The crude material was purified
by column chromatography (30:1 hexanes/EtOAc) to afford 8.15
g (80%) of aldol adduct 19, along with 1.07 g (19%) of recovered
aldehyde 17: [R]23D +49.2 (c 1.04, CHCl3); IR (thin film) 3521,
Hz), 2.23 (m, 1H), 2.33 (m, 2H), 3.44 (t, 2H, J ) 6.4 Hz), 3.55
(dd, 1H, J ) 6.4, 10.0 Hz), 3.68 (m, 1H), 3.73 (dd, 1H, J ) 7.0,
10.0 Hz), 4.39 (d, a of ab quartet, 1H, J ) 11.5 Hz), 4.41 (s,
2H), 4.46 (d, b of ab quartet, 1H, J ) 11.4 Hz), 7.24-7.40 (m,
17H), 7.63 (m, 3H), 9.60 (s, 1H); 13C NMR δ 11.5, 19.2, 26.9,
27.51, 27.55, 34.4, 36.2, 38.7, 48.4, 65.5, 68.0, 71.6, 72.6, 127.4,
127.5, 127.62, 127.66, 127.69, 128.2, 128.3, 129.6, 133.8, 135.6,
138.4, 138.9, 202.6. Anal. Calcd for C40H50O4Si: C, 77.13; H,
8.09. Found: C, 76.77; H, 8.25.
1
2860, 1781, 1694 cm-1; H NMR δ 1.16 (d, 3H, J ) 7.0 Hz),
1.28 (ddd, 1H, J ) 3.6, 7.8, 11.5 Hz), 1.62 (m, 4H), 2.03 (m,
2H), 2.70 (dd, 1H, J ) 9.4, 13.3 Hz), 2.82 (br s, 1H), 3.16 (dd,
1H, J ) 3.3, 13.4 Hz), 3.45 (dt, 2H, J ) 1.8, 6.7 Hz), 3.63 (dq,
1H, J ) 3.0, 7.0 Hz), 4.00 (dt, 1H, J ) 3.4, 9.1 Hz), 4.08 (m,
2H), 4.42 (s, 2H), 4.60 (m, 1H), 4.95 (m, 2H), 5.68 (m, 1H),
7.11-7.29 (m, 10H); 13C NMR δ 10.6, 31.5, 33.6, 37.7, 37.8,
37.9, 42.6, 55.0, 66.0, 68.3, 69.4, 72.8, 116.6, 127.3, 127.4, 127.5,
128.2, 128.9, 129.3, 135.0, 136.4, 138.4, 152.9, 177.1. Anal.
Calcd for C28H35NO5: C, 72.23; H, 7.58; N, 3.01. Found: C,
72.12; H, 7.53; N, 2.96.
Eth yl (2E,5S,7R,8R)-9-[[t-Bu tyld ip h en ylsilyl]oxy]-8-
m e t h yl-7-b e n zyloxy-5-[2-(b e n zyloxy)e t h yl]-2-n on e n o-
a te (24). Triethylphosphonoacetate (99.7 µL, 0.500 mmol) was
dissolved in CH3CN (3.0 mL), and LiCl (16.9 mg, 0.399 mmol)
and DBU (55.0 µL, 0.366 mmol) were added at room temper-
ature, sequentially. This solution was stirred for 20 min, and
then aldehyde 23 (207 mg, 0.333 mmol) in CH3CN (0.3 mL)
was added to the reaction mixture via syringe. After stirring
at room temperature overnight, the solution had turned
yellow-orange in color and contained some precipitate. The
reaction was quenched with water and extracted with CH2-
Cl2, and the combined organic extracts were dried over MgSO4,
filtered, and concentrated to a yellow-brown oil. This crude
material was purified by column chromatography (49:1 hex-
(2R,3R,5S)-(t-Bu tyl)-[2-m eth yl-3-(ben zyloxy)-5-[2-(ben -
zyloxy)eth yl]-7-octen yl]oxy]d ip h en ylsila n e (22). Alcohol
21 (223 mg, 0.420 mmol) was dissolved in a mixture of
CH2Cl2 (1.0 mL) and cyclohexane (2.0 mL) and cooled to 0 °C.
Benzyl trichloroacetimidate33 (0.158 mL, 0.840 mmol) was then
added dropwise, followed by triflic acid (3.7 µL, 40 µmol). The
reaction mixture immediately became cloudy with precipitate
upon addition of the acid, and stirring was continued at 0 °C
for an additional 12 h. The reaction was then quenched by
the addition of saturated aqueous NaHCO3, and the mixture
was extracted with CH2Cl2. The combined organic extracts
were washed with brine, dried over MgSO4, filtered, and
concentrated in vacuo. The resulting crude material was
purified by column chromatography (petroleum ether to 99:1
petroleum ether/EtOAc) to afford 207 mg of benzyl ether 22
anes/EtOAc) to give 177 mg (80%) of ester 24: [R]23 +3.24 (c
D
1.45, benzene); IR (thin film) 2930, 2857, 2361, 1718, 1652
cm-1; 1H NMR δ 0.80 (d, 3H, J ) 6.9 Hz), 0.98 (s, 9H), 1.21 (t,
3H, J ) 7.1 Hz), 1.29 (m, 1H), 1.44-1.56 (m, 3H), 1.72 (m,
1H), 1.85 (ddd, 1H, J ) 2.9, 6.8, 9.7 Hz), 2.10 (m, 2H), 3.46 (t,
2H, J ) 6.8 Hz), 3.55 (dd, 1H, J ) 6.5, 9.9 Hz), 1.73 (m, 2H),
4.18 (q, 2H, J ) 7.1 Hz), 4.43 (m, 4H), 5.78 (d, 1H, J ) 15.6
Hz), 6.86 (dt, 1H, J ) 7.5, 15.6 Hz), 7.23-7.44 (m, 16H), 7.61-
7.66 (m, 4H); 13C NMR δ 11.3, 14.3, 19.3, 26.9, 31.5, 33.8, 35.6,
39.0, 60.2, 65.3, 68.1, 71.9, 73.0, 76.8, 123.1, 127.4, 127.5,
127.59, 127.62, 127.66, 128.3, 128.4, 129.6, 133.8, 135.6, 138.5,
139.0, 147.3, 166.4. Anal. Calcd for C44H56O5Si: C, 76.26; H,
8.14. Found: C, 76.19; H, 7.91.
(79%): [R]23 +3.08 (c 1.04, CHCl3); IR (thin film) 3068, 2929,
D
1
1427 cm-1; H NMR δ 0.96 (d, 3H, J ) 6.9 Hz), 1.14 (s, 9H),
1.32-1.81 (m, 5H), 2.00 (ddd, 1H, J ) 2.9, 6.8, 13.6 Hz), 2.14
(t, 2H, J ) 5.9 Hz), 3.56 (t, 2H, J ) 6.9 Hz), 3.64 (dd, 1H, J )
6.4, 9.9 Hz), 3.83 (m, 2H), 4.54 (d, 4H, J ) 4.6 Hz), 5.05 (m,
4-(4-Iod o-2-m et h yl-b u t a -1,3-d ien yl)-2-m et h yl-oxa zole
(28). A solution of aldehyde 35 (0.542 g, 3.58 mmol) and
iodoform (2.83 g, 7.57 mmol) in THF (15 mL) was added
dropwise to a stirred suspension of chromium (II) chloride (2.66
g, 21.6 mmol) in THF (35 mL) at 0 °C. The reaction mixture
was stirred at 0 °C for 15 h and then poured onto water (75
mL). The aqueous layer was extracted with ether, and the
organic layers were dried, filtered, and concentrated. The crude
residue was purified by flash chromatography to give 28 (0.570
g, 58%) as a crystalline solid. IR (thin film) 2991, 2342, 1582,
1441 cm-1; 1H NMR δ 2.09 (s, 3 H), 2.46 (s, 3 H), 6.19 (s, 1 H),
6.41 (d, 1 H, J ) 14.6 Hz), 7.16 (d, 1 H, J ) 14.6 Hz), 7.55 (s,
1 H); 13C NMR δ 13.7, 13.8, 76.1, 81.3, 121.0, 136.6, 137.9,
149.3, 161.1.
Alk yn e 29. To a stirred solution of alcohol 51 (0.110 g, 0.247
mmol) and PMBOC(NH)CCl3 (0.160 g, 0.567 mmol) in dry
ether (3 mL) at room temperature was added triflic acid (0.1
µL, 0.001 mmol). The reaction mixture was stirred for 30 min
and then partitioned between water (2 mL) and ether (20 mL).
The organic layer was washed with saturated aqueous sodium
bicarbonate and brine, dried, filtered, and concentrated.
Purification by flash chromatography gave 29 (98.9 mg, 70%)
as a colorless oil: [R]25D +0.7 (c 0.73); IR (thin film) 3295, 2952,
2857, 2359, 1613 cm-1; 1H NMR δ 0.089 (s, 6 H), 0.91 (s, 9 H),
1.03 (d, 3 H, J ) 6.9 Hz), 1.62 (s, 3 H), 1.72 (dt, 1 H, J ) 14.3,
6.0 Hz), 1.94 (m, 1 H), 2.02 (m, 1 H), 2.41 (d, 1 H, J ) 1.5 Hz),
3.39 (S, 3 H), 3.79 (S, 3 H), 3.83 (M, 2 H), 4.02 (dd, 1 H, J )
1.7, 6.9 Hz), 4.30 (m, 3 H), 4.41 (d, 1 H, J ) 7.1 Hz), 4.45 (d,
2 H, J ) 8.2 Hz), 5.54 (t, 1 H, J ) 5.9 Hz), 6.84 (d, 2 H, J )
8.4 Hz), 7.22 (d, 2 H, J ) 8.4 Hz), 7.24 (m, 5 H); 13C NMR δ
-5.0, 10.1, 11.0, 18.4, 26.0, 35.7, 41.5, 55.3, 56.6, 59.9, 70.1,
71.3, 73.2, 74.5, 77.0, 81.9, 82.4, 113.7, 127.4, 127.9, 128.3,
129.2, 129.3, 131.2, 135.2, 138.7, 159.0. Anal. Calcd for
2H), 5.68 (m, 1H), 7.29-7.50 (m, 16H), 7.71-7.75 (m, 4H); 13
C
NMR δ 11.2, 19.2, 26.9, 31.6, 33.7, 35.7, 38.1, 39.1, 65.8, 68.4,
71.9, 72.8, 83.5, 116.4, 127.2, 127.4, 127.51, 127.52, 127.6,
128.2, 128.3, 129.5, 133.8, 135.6, 136.5, 138.6, 139.2. Anal.
Calcd for C41H52O3Si: C, 79.31; H, 8.44. Found: C, 79.50; H,
8.42.
(3S,5R,6R)-7-[[t-Bu t yld ip h en ylsilyl]oxy]-6-m et h yl-5-
ben zyloxy-3-[2-(ben zyloxy)eth yl]-h ep ta n a l (23). Olefin 22
(207 mg, 0.334 mmol) was dissolved in a mixture of THF/water
(3:1, 9.2 mL total) and cooled to 0 °C. To this solution were
added N-methylmorpholine N-oxide (78.0 mg, 0.667 mmol) and
OsO4 (4 wt % in water, 0.105 mL, 0.017 mmol), sequentially.
The mixture was warmed to room temperature and stirred
overnight. The reaction was then quenched by the addition of
solid sodium bisulfite (325 mg) to afford a dark red solution,
which was then concentrated in vacuo to remove all volatiles.
The resulting paste was partitioned between 1 N HCl and
EtOAc and then extracted with Et2O and EtOAc. The com-
bined organic extracts were washed with saturated aqueous
NaHCO3, dried over K2CO3, filtered, and concentrated. This
crude diol was used directly in the oxidative cleavage step that
follows.
The crude dihydroxylated material was dissolved in THF/
water (3:1, 9.2 mL total), and NaIO4 (142 mg, 0.667 mmol)
was added to this solution in one portion. A white precipitate
formed after several minutes, and after 8 h the reaction was
judged to be complete by TLC analysis. The volatiles were
removed in vacuo, and the resulting paste was dissolved in
brine and extracted with Et2O. The combined organic extracts
were dried over MgSO4, filtered, and concentrated to afford
212 mg of aldehyde 23 (100%): IR (thin film) 2931, 2861, 1718
C
34H50O5Si: C, 72.04; H, 8.89. Found: C, 72.34; H, 9.11.
1
cm-1; H NMR δ 0.89 (d, 3H, J ) 6.9 Hz), 1.07 (s, 9H), 1.39
Ald eh yd e 34. To a solution of oxalyl chloride (0.460 mL,
5.15 mmol) in CH2Cl2 (15 mL) at -60 °C was added a solution
(m, 1H), 1.58-1.68 (m, 3H), 1.94 (ddd, 1H, J ) 2.9, 6.8, 13.6
J . Org. Chem, Vol. 68, No. 11, 2003 4227

Lafontaine et al.
of dimethyl sulfoxide (0.760 mL, 10.7 mmol) in CH2Cl2 (2 mL).
The solution was stirred for 2 min followed by the addition of
alcohol 33 (0.562 g, 4.97 mmol) in CH2Cl2 (7 mL) dropwise over
5 min. After 15 min, triethylamine (3.45 mL, 24.8 mmol) was
added and the reaction mixture was stirred at -60 °C for 5
min, warmed to room temperature, and then stirred for 10
min. Water (25 mL) was added, and the mixture was extracted
with CH2Cl2. The combined organic extracts were dried,
filtered, and concentrated. The crude residue was purified by
flash chromatography (ether) to afford 34 (0.522 g, 94%) as a
light yellow solid: IR (thin film) 3008, 2844, 1701 (s), 1599,
1564 cm-1; 1H NMR δ 2.54 (s, 3 H), 8.18 (s, 1 H), 9.91 (s, 1 H);
13C NMR δ 13.7, 140.9, 144.5, 163.0, 183.7.
En a l 35. To a solution of aldehyde 34 (0.228 g, 2.05 mmol)
in benzene (12 mL) was added 2-(triphenyl-phosphoranylidene)-
propionaldehyde (0.797 g, 2.50 mmol) at room temperature.
After 3 days, water (10 mL) was added and the mixture was
extracted with ether. The combined organics were dried,
filtered, and concentrated. Purification by flash chromatogra-
phy yielded 35 (0.281 g, 91%): IR (thin film) 3003, 1688 (s),
1640, 1589 cm-1; 1H NMR δ 2.09 (s, 3 H), 2.52 (s, 3 H), 7.07 (s,
1 H), 7.83 (s, 1 H), 9.54 (s, 1 H); 13C NMR δ 11.1, 13.8, 137.4,
137.8, 138.5, 139.7, 162.0, 194.4.
was added rapidly via cannula. The reaction mixture was
stirred for 10 min, and triphenylphosphine (2.87 g, 10.9 mmol)
was added followed by warming to room temperature and
stirring for 4 h. The reaction mixture was concentrated to give
a crude white solid. Purification by flash chromatography gave
the starting diene 41 (0.769 g, 2.22 mmol) and the desired
aldehyde 42 (1.72 g, 57%, 76% BORSM) as a colorless oil:
[R]25 -19.9 (c 1.17); IR (thin film) 2954, 2856, 1728 (s) cm-1
;
D
1H NMR δ 0.086 (s, 3 H), 0.089 (s, 3 H), 0.92 (s, 9 H), 1.63 (d,
3 H, J ) 1.0 Hz), 2.42 (ddd, 1 H, J ) 2.0, 4.2, 16.1 Hz), 2.77
(ddd, 1 H, J ) 2.7, 9.3, 16.1 Hz), 4.27 (m, 3 H), 4.29 (d, 1 H, J
) 11.4 Hz), 4.50 (d, 1 H, J ) 11.7 Hz), 5.64 (t, 1 H, J ) 5.7
Hz), 7.30 (m, 5 H), 9.72 (t, 1 H, J ) 2.3 Hz); 13C NMR δ -5.1,
11.2, 18.3, 25.9, 47.7, 59.7, 70.0, 79.0, 127.7, 127.8, 128.4, 129.7,
133.7, 137.9, 200.9.
Meth yl (4E)-6-[(t-Bu tyld im eth ylsilyl)oxy]-4-m eth yl-3-
h yd r oxy-4-h exen oa t e (r a c-43). A solution of diisopropyl-
amine (6.96 mL, 49.7 mmol) in THF (36 mL) was cooled to
-78 °C, and n-BuLi (2.5 M, 18.2 mL, 45.7 mmol) was added
dropwise via syringe. The resulting solution was stirred for
15 min, after which a solution of methyl acetate (4.03 mL, 50.7
mmol) in THF (25 mL) was added dropwise. This mixture was
stirred for 1 h, and aldehyde 39 (1.88 g, 9.90 mmol) in THF
(10 mL) was then added to the enolate solution. After 5 min,
all of the aldehyde had been consumed, and the reaction was
quenched by the addition of saturated aqueous NH4Cl; the
mixture was then extracted with Et2O. The organic extracts
were dried over MgSO4, filtered, and concentrated in vacuo.
The resulting crude oil was purified by column chromatogra-
phy (hexanes to 9:1 hexanes/EtOAc) to provide 2.00 g (78%)
of the racemic aldol product 43: IR (thin film) 3479, 2952, 1731
Ald eh yd e 38. A solution of 37 (21.3 g, 67.3 mmol) in dry
CH2Cl2 (400 mL) was cooled to -78 °C, and ozone was bubbled
through until the solution turned blue. Nitrogen was then
bubbled through the solution until the reaction mixture turned
colorless. Triphenylphosphine (20.7 g, 78.9 mmol) was added,
and the mixture was warmed to room temperature and stirred
under nitrogen for 1.5 h. The solvent was removed in vacuo,
and the residue was purified by flash chromatography to yield
38 (23.6 g, 100%) as a colorless oil: IR (thin film) 3458, 2952,
1
cm-1; H NMR δ 0.04 (s, 6H), 0.87 (s, 9H), 1.61 (s, 3H), 2.48
2858, 1740 (s) cm-1 1H NMR δ 0.11 (s, 6 H), 0.93 (s, 9 H),
;
(dd, 1H, J ) 4.0, 15.5 Hz), 2.54 (dd, 1H, J ) 9.0, 16.0 Hz),
2.90 (br s, 1H), 3.68 (s, 3H), 4.19 (d, 2H, J ) 5.5 Hz), 4.41 (dd,
1H, J ) 3.5, 8.5 Hz), 5.58 (t, 1H, J ) 6.0 Hz); 13C NMR δ -5.2,
12.1, 18.3, 25.9, 39.8, 51.7, 59.8, 72.7, 126.4, 136.3, 172.8. Anal.
Calcd for C14H28O4Si: C, 58.29; H, 9.78. Found: C, 58.06; H,
9.56.
4.22 (d, 2 H, J ) 0.7 Hz), 9.70 (t, 1 H, J ) 0.7 Hz); 13C NMR
δ -5.4, 18.3, 25.7, 69.6, 202.3.
En a l 39. A solution of aldehyde 38 (3.43 g, 19.7 mmol) in
dry benzene (100 mL) was cooled to 0 °C, and 2-(triphenyl-
phosphoranylidene)-propionaldehyde (7.54 g, 23.7 mmol) was
added. The suspension was warmed to room temperature and
stirred for 2 days. Water (80 mL) was added, and the mix-
ture was extracted with ether. The combined organic ex-
tracts were dried, filtered, and concentrated. Flash chroma-
tography provided 39 (3.57 g, 84%) as a colorless oil: IR (thin
film) 2954, 2857, 1692 (s) cm-1; 1H NMR δ 0.062 (s, 6 H), 0.88
(s, 9 H), 1.68 (q, 3 H, J ) 1.2 Hz), 4.46 (dq, 2 H, J ) 5.3, 1.1
Hz), 6.48 (tq, 1 H, J ) 5.3, 1.3 Hz), 9.39 (s, 1 H); 13C NMR δ
-5.4, 9.2, 18.2, 25.7, 60.4, 137.7, 152.9, 194.3. Anal. Calcd for
Meth yl (4E,3S)-6-[(t-Bu tyld im eth ylsilyl)oxy]-4-m eth yl-
3-h yd r oxy-4-h exen oa te ((-)-43). In a three-necked round-
bottomed flask equipped with a thermometer, (-)-diisopropyl
tartrate (DIPT) (62.5 mg, 0.260 mmol) and crushed 3 Å sieves
(122 mg) were stirred in CH2Cl2 (6 mL) at -20 °C. This
solution was treated with Ti(Oi-Pr)4 (52.5 µL, 0.178 mmol) and
TBHP (predried over sieves, 4.0 M in CH2Cl2, 0.33 mL, 1.33
mmol) sequentially, and the resulting mixture was stirred for
30 min and then cooled to -30 °C. Racemic ester 43 (490 mg,
1.78 mmol) in 3 mL of CH2Cl2 was then added to the solution.
This mixture was stirred at -20 °C for 20 h; at this time, NMR
analysis of the crude reaction mixture revealed 55% conversion
to the epoxide. The reaction was then quenched by the addition
of FeSO4/citric acid solution (2.5 g/0.85 g in 7.5 mL of water),
and the mixture was diluted with Et2O and extracted. The
combined organic extracts were dried over MgSO4, filtered, and
concentrated in vacuo. The crude material was then purified
by column chromatography (19:1 hexanes/EtOAc) to afford 198
C
11H22O2Si: C, 61.63; H, 10.35. Found: C, 61.50; H, 10.12.
Alcoh ol 40. A stirred solution of (-)-B-methoxy diisopi-
nocampheylborane (5.06 g, 16.0 mmol) in dry ether (20 mL)
was cooled to -78 °C, and allylmagnesium bromide (16.0 mL,
1.0 M in ether, 16.0 mmol) was added dropwise. The reaction
mixture was stirred at -78 °C for 15 min and then warmed
to room temperature for 1 h. The solution was cooled to -78
°C, and a solution of aldehyde 39 (3.36 g, 15.7 mmol) in dry
ether (10 mL) was added dropwise. The solution was stirred
at -78 °C for 1 h and then warmed to room temperature for
1 h. Sodium hydroxide (12.0 mL, 3 M, 36.0 mmol) and
hydrogen peroxide (30%, 8.0 mL) were added, and the resulting
mixture was heated to reflux for 2 h. The organic layer was
washed with water, dried, filtered, and concentrated. Purifica-
tion by flash chromatography produced 40 (3.70 g, 92%) as a
colorless oil: [R]25D -8.26 (c 1.00); IR (thin film) 3389 (b), 2930,
mg (40%) of olefin 43: [R]23 -5.65 (c 0.92, CHCl3).
D
(4E,3S)-6-[(t-Bu tyld im eth ylsilyl)oxy]-4-m eth yl-3-[(tr i-
isop r op ylsilyl)oxy]-4-h exen a l (46). In a three-necked flask
equipped with a thermometer, ester 45 (71.7 mg, 0.166 mmol)
was dissolved in CH2Cl2 (1.66 mL) and cooled to -95 °C (N2
(l)/EtOH). DIBAL-H (1.5 M in toluene, 0.143 mL, 0.215 mmol)
was then added dropwise to this solution. The resulting
mixture was stirred for 15 min, and then the reaction was
quenched by the addition of MeOH. The solution was then
warmed and diluted with aqueous Rochelle’s salt solution and
Et2O and stirred vigorously for 2 h. The aqueous phase was
then extracted with Et2O, and the combined organic extracts
were dried over MgSO4, filtered, and concentrated to afford a
colorless oil. This crude material was then purified by column
chromatography (19:1 hexanes/EtOAc) to give 59.2 mg (89%)
of aldehyde 46: [R]23D -2.04 (c 1.4, CHCl3); IR (thin film) 2930,
1
2857, 1669 cm-1; H NMR δ 0.07 (s, 6 H), 0.90 (s, 9 H), 1.63
(d, 3 H, J ) 1.0 Hz), 2.02 (s, 1 H), 2.31 (m, 2 H), 4.06 (t, 1 H,
J ) 6.5 Hz), 4.23 (d, 2 H, J ) 6.1 Hz), 5.11 (m 2 H), 5.56 (tt,
1 H, J ) 1.2, 6.1 Hz), 5.77 (m, 1 H); 13C NMR δ -5.2, 12, 18.3,
25.9, 39.6, 59.5, 75.9, 117.6, 126.1, 134.6, 137.6. Anal. Calcd
for C14H28O2Si: C, 65.57; H, 11.00. Found: C, 65.37; H, 10.86.
Ald eh yd e 42. A solution of 41 (3.025 g, 8.73 mmol) in dry
CH2Cl2 (20 mL) was cooled to -78 °C and a saturated solution
of ozone (0.04 M in CH2Cl2, 290 mL, 11.6 mmol) at -78 °C
4228 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
2865, 1727 cm-1; 1H NMR δ 0.04 (s, 6H), 0.88 (s, 9H), 1.02 (s,
21H), 1.61 (s, 3H), 2.54 (ddd, 1H, J ) 2.7, 5.9, 15.6 Hz), 2.61
(ddd, 1H, J ) 2.6, 5.9, 15.5 Hz), 4.19 (d, 2H, J ) 5.6 Hz), 4.62
(t, 1H, J ) 5.9 Hz), 5.62 (t, 1H, J ) 5.3 Hz), 9.74 (t, 1H, J )
2.6 Hz); 13C NMR δ -5.3, 11.9, 12.2, 17.9, 18.0, 18.2, 25.8, 49.9,
59.7, 73.3, 126.6, 201.7.
79.4, 81.0, 84.9, 127.8, 127.9, 128.5, 129.6, 134.4, 137.6, 189.6.
Anal. Calcd for C25H38O4Si: C, 69.72; H, 8.89. Found: C, 69.80;
H, 8.81.
Diol 50. A stirred solution of 49 (0.323 g, 0.749 mmol) in
THF (5 mL) was cooled to -78 °C under nitrogen, and
diethylmethoxyborane (0.90 mL, 1.0 M in THF, 0.90 mmol)
was added dropwise. The reaction mixture was stirred for 15
min followed by the addition of sodium borohydride (36.8 mg,
0.974 mmol). The solution was stirred at -78 °C for 3.5 h and
then warmed to room temperature. Acetic acid (0.6 mL) and
ethyl acetate (40 mL) were added, and the organic layer was
washed with saturated aqueous NaHCO3 and brine, dried,
filtered, and concentrated. Purification by flash chromatogra-
Ald ol P r od u ct 47. A solution of (4S)-3-propionyl-4-(benzyl)-
2-oxazolidinone (2.69 g, 11.5 mmol) in CH2Cl2 (35 mL) was
cooled to 0 °C, and dibutylboron triflate (2.88 mL, 11.5 mmol)
was added dropwise. The reaction mixture was stirred for 10
min followed by the addition of diisopropylethylamine (2.00
mL, 11.5 mmol). The solution was stirred at 0 °C for 15 min
and then cooled to -78 °C. A solution of 42 (2.68 g, 7.69 mmol)
in CH2Cl2 (15 mL) was added dropwise followed by stirring
for 3 h. The reaction was warmed to 0 °C, and 1 M sodium
acetate in 9:1 MeOH/water (55 mL) and hydrogen peroxide
(30%, 22 mL) were added. The cloudy mixture was stirred at
0 °C for 90 min and then concentrated. The crude residue was
diluted with water (50 mL) and extracted with CH2Cl2. The
combined organic extracts were washed with saturated aque-
ous NaHCO3 and brine, dried over sodium sulfate, filtered,
and concentrated. Purification by flash chromatography yielded
phy yielded 50 (0.272 g, 84%) as a colorless oil: [R]25 -40.6
D
(c 0.50); IR (thin film) 3428 (b), 3296 (b), 2856, 2360 cm-1; 1H
NMR δ 0.089 (s, 3H), 0.092 (s, 3 H), 0.92 (s, 9 H), 1.09 (d, 3 H,
J ) 7.0 Hz), 1.47 (d, 1 H, J ) 14.6 Hz), 1.64 (s, 3 H), 1.73 (m,
1 H), 2.01 (dt, 1 H, J ) 14.5, 10.2 Hz), 2.43 (d, 1 H, J ) 2.1
Hz), 3.52 (s, 1 H), 3.98 (dd, 1 H, J ) 2.9, 10.5 Hz), 4.11 (d, 1
H, J ) 10.0 Hz), 4.26 (d, 1 H, J ) 11.7 Hz), 4.27 (m, 3 H), 4.49
(d, 1 H, J ) 11.4 Hz), 4.65 (t, 1 H, J ) 3.2 Hz), 5.59 (t, 1 H, J
) 5.8 Hz), 7.31 (m, 5 H); 13C NMR δ -5.1, 0.0, 7.1, 11.2, 18.3,
25.9, 38.6, 43.5, 59.8, 66.8, 70.1, 72.8, 75.3, 85.6, 127.9, 128.0,
128.6, 129.4, 134.6, 137.6. Anal. Calcd for C25H40O4Si: C, 69.40;
H, 9.32. Found: C, 69.19; H, 9.11.
47 (3.97 g, 89%) as a colorless oil: [R]25 -22.7 (c 1.02); IR
D
(thin film) 3505 (b), 2928, 2856, 1781 (s), 1695 (s) cm-1 1H
;
NMR δ 0.089 (s, 6 H), 0.92 (s, 9 H), 1.25 (d, 3 H, J ) 6.9 Hz),
1.58 (m, 1 H), 1.63 (s, 3 H), 1.93 (dt, 1 H, J ) 14.2, 9.3 Hz),
2.76 (dd, 1 H, J ) 9.5, 13.3 Hz), 3.27 (dd, 1 H, J ) 3.4, 10.2
Hz), 3.67 (d, 1 H, J ) 1.7 Hz), 3.81 (m, 1 H), 4.01 (dd, 1 H, J
) 4.3, 9.8 Hz), 4.09 (m, 1 H), 4.15 (m, 2 H), 4,26 (d, 1 H, J )
11.4 Hz), 4.27 (m, 2 H), 4.46 (d, 1 H, J ) 11.4 Hz), 4.67 (m, 1
H), 5.61 (t, 1 H, J ) 5.7 Hz), 7.26 (m, 10 H); 13C NMR δ -5.1,
11.0, 11.5, 18.3, 25.9, 37.8, 37.9, 42.9, 55.3, 59.8, 66.0, 70.0,
71.3, 84.6, 127.3, 127.6, 127.9, 128.4, 128.9, 129.4, 129.6, 134.5,
135.2, 137.9, 153.1, 175.8. Anal. Calcd for C33H47NO6Si: C,
68.12; H, 8.14; N, 2.41. Found: C, 67.88; H, 7.96; N, 2.34.
P h osp h on a te 54. To a solution of ethyl diethylphosphonate
(0.161 g, 0.966 mmol) at -78 °C in THF (5 mL) was added
tert-butyllithium (0.570 mL, 1.7 M in pentane, 0.970 mmol)
dropwise. After 45 min, a solution of aldehyde 58 (0.303 g,
0.633 mmol) in THF (2 mL) was added and the mixture was
stirred for 15 min. Saturated aqueous NH4Cl was added, and
the mixture was extracted with ether. The combined organic
extracts were dried, filtered, and concentrated. The crude oil
was purified by flash chromatography to provide the product
alcohols (0.363 g, 89%) as a mixture of diastereomers. See
Supporting Information for the 1H NMR spectrum: IR (thin
Wein r eb Am id e 48. To a stirred suspension of N,O-
dimethylhydroxylamine hydrochloride (0.777 g, 7.97 mmol) in
dry THF (25 mL) at 0 °C was added dropwise trimethylalu-
minum (4.00 mL, 2.0 M in hexanes, 8.00 mmol). The resulting
solution was stirred at 0 °C for 30 min followed by the addition
of 47 (1.02 g, 1.75 mL) in THF (10 mL). The reaction mixture
was warmed to room temperature and stirred for 12 h, cooled
to 0 °C, and added via cannula to a cooled (0 °C) mixture of
CH2Cl2 (50 mL) and 0.5 M HCl (40 mL). After stirring at 0 °C
for 1 h, the mixture was extracted with CH2Cl2, dried over
sodium sulfate, filtered, and concentrated. Flash chromatog-
raphy afforded 48 (0.771 g, 95%) as a colorless oil: [R]25D -16.4
film) 3427 (b), 2955, 2857 cm-1
.
Dess-Martin periodinane (0.253 g, 0.596 mmol) was added
to a solution of the products from above (0.252 g, 0.391 mmol)
in CH2Cl2 (5 mL) at 0 °C. The reaction mixture was warmed
to room temperature and stirred for 20 min. Ether was added,
and the organic layer was washed with 1 N sodium hydroxide
and water, dried, filtered, and concentrated to give phospho-
nate 54 (0.244 g, 97%) as a colorless oil. The crude product,
an expected mixture of diastereomers, was not purified further
and was used directly in the next step. See Supporting
1
Information for the H NMR spectrum: IR (cm-1) 2955, 2857,
1710 (s).
(c 1.09); IR (thin film) 3475 (b), 2930, 2857, 1739 (s), 1657 cm-1
;
(1′E,3′E)-4-(4′-(Eth oxyca r bon yl)-2′-m eth yl-1′,3′-bu ta d i-
en yl)-2-m eth yloxazole (55). Triethylphosphonoacetate (0.692
mL, 3.47 mmol) was dissolved in CH3CN (20 mL), and LiCl
(118 mg, 2.78 mmol) and DBU (380 mL, 2.54 mmol) were
added at room temperature, sequentially. This solution was
stirred for 20 min, and then aldehyde 35 (350 mg, 2.31 mmol)
in CH3CN (3 mL) was added to the reaction mixture via
syringe. After stirring at room temperature for 3 days, the
solution had turned yellow in color, and the reaction was
judged to be complete by TLC analysis. The reaction was
quenched with water and extracted with Et2O (three times)
and CH2Cl2 (one time), and the combined organic extracts were
dried over MgSO4, filtered, and concentrated to an oil. This
crude material was purified by column chromatography (19:1
hexanes/EtOAc) to give 478 mg (93%) of ester 55 as a white
solid: IR (thin film) 2360, 1670, 1610 cm-1; 1H NMR δ 1.29 (t,
3H, J ) 7.0 Hz), 2.15 (s, 3H), 2.46 (s, 3H), 4.20 (q, 2H, J ) 7.0
Hz), 5.97 (d, 1H, J ) 15.5 Hz), 6.46 (s, 1H), 7.43 (d, 1H, J )
15.5 Hz), 7.62 (s, 1H); 13C NMR δ 13.8, 14.0, 14.3, 60.3, 117.8,
126.9, 134.7, 137.7, 138.2, 148.8, 161.3, 167.2. Anal. Calcd for
1H NMR δ 0.085 (s, 6 H), 0.91 (s, 9 H), 1.19 (d, 3 H, J ) 7.0
Hz), 1.59 (m, 1 H), 1.61 (s, 3 H), 1.87 (dt, 1 H, J ) 14.1, 9.1
Hz), 2.91 (bs, 1 H), 3.16 (s, 3 H), 3.65 (s, 3 H), 3.91 (m, 1 H),
3.99 (m, 1 H), 4.01 (s, 1 H), 4.26 (m, 2 H), 4.26 (d, 1 H, J )
11.5 Hz), 4.47 (d, 1 H, J ) 11.5 Hz), 5.59 (t, 1 H, J ) 5.6 Hz),
7.29 (m, 5 H); 13C NMR δ -5.1, 11.0, 12.5, 18.3, 25.9, 31.9,
38.0, 40.3, 59.8, 61.4, 70.0, 71.4, 84.3, 127.6, 127.8, 128.4, 129.6,
134.5, 138.0, 151.2. Anal. Calcd for C25H43NO5Si: C, 64.48;
H, 9.31; N, 3.01. Found: C, 64.30; H, 9.08; N, 2.95.
Alk yn e 49. To a solution of 48 (94.4 mg, 0.203 mmol) in
THF (3 mL) at room temperature was added ethynylmagne-
sium bromide (4.0 mL, 0.5 M in THF, 2.0 mmol). The reaction
mixture was stirred for 15 h and then filtered through a plug
of silica gel, which was flushed with ethyl acetate. The filtrate
was concentrated and the residue purified by flash chroma-
tography to give 49 (57.9 mg, 66%): [R]25 -21.8 (c 0.92); IR
D
(thin film) 3484 (b), 3261, 2359 (w), 2090, 1679 cm-1; 1H NMR
δ 0.092 (s, 6 H), 0.92 (s, 9 H), 1.22 (d, 3 H, J ) 7.0 Hz), 1.54
(m, 1 H), 1.63 (s, 3 H), 1.90 (dt, 1 H, J ) 14.3, 10.0 Hz), 2.59
(m, 1 H), 3.26 (s, 1 H), 3.64 (s, 1 H), 4.01 (dd, 1 H, J ) 3.5,
10.0 Hz), 4.26 (m, 3 H), 4.27 (d, 1 H, J ) 11.3 Hz), 4.49 (d, 1
H, J ) 11.4 Hz), 5.60 (t, 1 H, J ) 5.9 Hz), 7.31 (m, 5 H); 13C
NMR δ -5.1, 10.0, 11.1, 18.3, 25.9, 38.4, 54.0, 59.8, 70.1, 71.1,
C
12H15NO3: C, 65.14; H, 6.83; N, 6.33. Found: C, 65.32; H,
6.70; N, 6.42.
Ald eh yd e 58. To a solution of amide 57 (1.65 g, 3.06 mmol)
in THF (30 mL) at -78 °C was added diisobutylaluminum
J . Org. Chem, Vol. 68, No. 11, 2003 4229

Lafontaine et al.
1
hydride (2.90 mL, 1.5 M in toluene, 4.35 mmol). After the
reaction was stirred for 45 min, a saturated aqueous solution
of sodium potassium tartrate (100 mL) was added and the
mixture was extracted with ether. The combined organic layers
were dried, filtered, and concentrated. Purification by flash
chromatography afforded alcohol 58 (1.37 g, 93%) as a colorless
cm-1; H NMR δ 0.075 (s, 3 H), 0.078 (s, 3 H), 0.82 (d, 3 H, J
) 7.0 Hz), 0.90 (s, 9 H), 1.44 (d, 1 H, J ) 14.2 Hz), 1.60 (m, 1
H), 1.63 (s, 3 H), 1.72 (s, 3 H), 1.98 (dt, 1 H, J ) 14.4, 10.0
Hz), 2.11 (s, 3 H), 2.43 (s, 3 H), 3.71 (s, 1 H), 3.95 (s, 1 H), 3.99
(dd, 1 H, J ) 2.9, 10.3 Hz), 4.06 (d, 1 H, J ) 9.6 Hz), 4.25 (m,
2 H), 4,26 (d, 1 H, J ) 11.5 Hz), 4.31 (s, 1 H), 4.49 (d, 1 H, J
) 11.5 Hz), 5.57 (t, 1 H, J ) 5.4 Hz), 6.21 (s, 1 H), 6.31 (d, 1
H, J ) 11.1 Hz), 6.39 (d, 1 H, J ) 15.0 Hz), 6.57 (dd, 1 H, J )
11.1, 15.0 Hz), 7.30 (m, 5 H), 7.49 (s, 1 H); 13C NMR δ -5.1,
5.1, 11.2, 13.8, 14.4, 14.6, 18.3, 25.9, 39.2, 40.4, 59.8, 70.0, 76.1,
79.7, 85.4, 119.7, 124.5, 124.8, 127.8, 127.9, 128.5, 129.2, 134.7,
135.6, 136.5, 137.2, 137.6, 138.4, 138.8, 160.8.
oil: [R]25D -2.2 (c 1.26); IR (thin film) 2955, 2857, 1732 (s) cm-1
;
1H NMR δ 0.08 (s, 9 H), 0.091 (s, 3 H), 0.093 (s, 3 H), 0.92 (s,
9 H), 0.99 (d, 3 H, J ) 7.0 Hz), 1.61 (s, 3 H), 1.67 (m, 1 H),
1.88 (m, 1 H), 2.24 (m, 1 H), 3.71 (dd, 1 H, J ) 5.4, 8.1 Hz),
4.20 (d, 1 H, J ) 11.9 Hz), 4.28 (m, 3 H), 4.48 (d, 1 H, J ) 11.9
Hz), 5.56 (t, 1 H, J ) 5.6 Hz), 7.30 (m, 5 H), 9.62 (d, 1 H, J )
0.7 Hz); 13C NMR δ -5.1, 0.4, 7.5, 11.0, 18.3, 25.9, 38.7, 50.7,
59.8, 68.7, 69.6, 80.7, 127.5, 127.9, 128.3, 129.4, 134.7, 138.4,
204.6. Anal. Calcd for C26H46O4Si2: C, 65.22; H, 9.68. Found:
C, 64.99; H, 9.49.
Ester 72.40 The norephedrine-based auxiliary 71 (260 mg,
0.543 mmol) and triethylamine (0.264 mL, 1.90 mmol) were
dissolved in CH2Cl2 (4.0 mL), and the mixture was cooled to
-78 °C. Dicyclohexylboron triflate (1 M in hexanes, 1.62 mL,
1.62 mmol) was then added to the solution via syringe, and
the resulting mixture was stirred for 2 h at -78 °C. Aldehyde
46 (248 mg, 0.597 mmol) in CH2Cl2 (2.0 mL) was added to the
solution dropwise via cannula, and the mixture was stirred
at -78 °C for 1 h and then warmed to 0 °C for an additional
2 h. The reaction was then quenched by the addition of 1 M
NaOAc in 9:1 MeOH/H2O (10 mL) and 30% H2O2 (4 mL), and
this mixture was stirred vigorously for 12 h. The volatiles were
then removed in vacuo, and the resulting paste was partitioned
between Et2O and water and then extracted with Et2O. The
organic layer was dried over MgSO4, filtered, and concentrated
to an oil. The crude material was purified by column chroma-
tography (hexanes to 40:1 hexanes/EtOAc) to afford 391 mg
Tr ien e 59. Barium hydroxide octahydrate (dried at 110 °C
for 18 h, 0.730 g, 2.31 mmol) was added to a solution of
phosphonate 54 (1.55 g, 2.41 mmol) in THF (10 mL), and the
resulting suspension was stirred at ambient temperature for
30 min. A solution of aldehyde 53 (0.478 g, 2.70 mmol) in 40:1
THF/water (5 mL) was added, and the reaction mixture was
stirred for 16.5 h. The reaction mixture was diluted with
CH2Cl2, and the organic layer was washed with saturated
aqueous sodium bicarbonate and then dried, filtered, and
concentrated. Purification by flash chromatography provided
59 (1.04 g, 65%) as a yellow oil: [R]25 -18.7 (c 0.61); IR (thin
D
film) 2954, 2856, 1652 cm-1; H NMR δ 0.069 (s, 6 H), 0.092
1
(s, 9 H), 0.90 (s, 9 H), 1.10 (d, 3 H, J ) 6.8 Hz), 1.53 (s, 3 H),
1.61 (dt, 1 H, J ) 5.4 Hz), 1.80 (m, 1 H), 1.85 (s, 3 H), 2.14 (s,
3 H), 2.48 (s, 3 H), 3.49 (quint, 1 H, J ) 7.2 Hz), 3.91 (t, 1 H,
J ) 6.5 Hz), 3.98 (m, 1 H), 4.25 (d, 1 H, J ) 11.9 Hz), 4.26
(m, 2 H), 4.41 (d, 1 H, J ) 11.9 Hz), 5.53 (t, 1 H, J ) 6.2 Hz),
6.29 (s, 1 H), 6.54 (m, 2 H), 7.07 (d, 1 H, J ) 10.3 Hz), 7.25
(m, 5 H), 7.59 (s, 1 H); 13C NMR δ -5.1, 0.6, 10.9, 11.9,
13.8, 14.2, 15.4, 18.3, 25.9, 39.8, 45.1, 59.8, 69.5, 72.2, 81.1,
123.7, 124.5, 127.1, 127.5, 128.1, 129.3, 135.2, 135.3, 136.4,
136.8, 138.6, 138.6, 138.8, 144.1, 161.1, 204.5. Anal. Calcd for
of aldol adduct 72 (81%): [R]23 -1.77 (c 0.6, CDCl3); IR (thin
D
film) 3452, 2930, 1730 cm-1; H NMR δ 0.06 (s, 6Η), 0.89 (s,
1
9H), 1.04 (m, 23H), 1.08 (d, 3H, J ) 7.5 Hz), 1.11 (d, 3H, J )
7.0 Hz), 1.52 (s, 3H), 1.63 (ddd, 1H, J ) 4.5, 10.0, 14.5 Hz),
1.68 (ddd, 1H, J ) 2.0, 5.5, 14.5 Hz), 2.45 (s, 3H), 2.50 (s, 8H),
3.38 (s, 1H), 3.91 (t, 1H, J ) 8.0 Hz), 4.01 (m, 1H), 4.21 (m,
1H), 4.41 (t, 1H, J ) 4.0 Hz), 4.60 (d, 1H, J ) 16.5 Hz), 4.82
(d, 1H, J ) 16.5 Hz), 5.67 (t, 1H, J ) 6.0 Hz), 5.73 (d, 1H, J )
4.0 Hz), 6.79 (d, 2H, J ) 7.0 Hz), 6.90 (s, 2H), 7.22 (m, 6H);
13C NMR δ -5.29, -5.25, 12.2, 13.1, 13.2, 13.4, 17.98, 18.00,
18.2, 20.9, 22.9, 25.8, 38.5, 45.9, 48.2, 56.7, 59.7, 70.1, 75.7,
77.9, 125.7, 125.9, 127.0, 127.6, 127.7, 128.2, 128.3, 132.1,
133.5, 136.7, 138.4, 138.9, 140.3, 142.4, 173.7. Anal. Calcd for
C
38H59NO5Si2: C, 68.53; H, 8.93; N, 2.10. Found: C, 68.50; H,
9.13; N, 1.72.
Alcoh ol 60. To a solution of silyl ether 59 (0.433 g, 0.650
mmol) in THF (8 mL) at room temperature was added a
solution of citric acid in methanol (11.0 mL, 1 M, 11.0 mmol).
After 1 h, the reaction mixture was neutralized with saturated
aqueous sodium bicarbonate and extracted with ether. The
combined organic layers were dried, filtered, and concentrated.
Purification by flash chromatography afforded alcohol 60
C
50H79NO7SSi2: C, 67.14; H, 8.90; N, 1.57. Found: C, 67.15;
H, 8.74; N, 1.47.
(2E,4S,6S,7R,8RS)-1-[(t-Bu tyld im eth ylsilyl)oxy]-9-(d i-
eth ylp h osp h on o)-3,7-d im eth yl-4-[(tr iisop r op ylsilyl)oxy]-
6-[(tr im eth ylsilyl)oxy]-2-d eca en -8-ol (76). Ethyl diethyl
phosphonate (0.609 g, 9.17 mmol) was dissolved in THF (12.2
mL) and cooled to -78 °C, and t-BuLi (1.5 M in pentane, 6.11
mL, 9.17 mmol) was added dropwise. This solution was stirred
for 45 min, and then a solution of aldehyde 75 (1.75 g, 3.21
mmol) in 12.2 mL of THF was added via syringe. After 10 min,
the reaction was quenched by the addition of saturated
aqueous NH4Cl. The resulting mixture was diluted and
extracted with Et2O, and the combined organic extracts were
dried over MgSO4, filtered, and concentrated in vacuo. The
crude material was purified by column chromatography (19:1
to 3:1 hexanes/EtOAc) to afford 2.28 g (100%) of hydroxy
phosphonates 76 as a mixture of four isomers: IR (thin film)
3420, 2928 cm-1. Anal. Calcd for C34H75O7PSi3: C, 57.42; H,
10.63. Found: C, 57.35; H, 10.46. See Supporting Information
for the 1H NMR spectrum.
(2E,4S,6S,7S)-1-[(t-Bu t yld im et h ylsilyl)oxy]-9-(d iet h -
ylp h osp h on o)-3,7-d im et h yl-4-[(t r iisop r op ylsilyl)oxy]-6-
[(tr im eth ylsilyl)oxy]-2-d eca en -8-on e (70). Hydroxy phos-
phonates 76 (125 mg, 0.175 mmol) were dissolved in CH2Cl2
(1.75 mL) and cooled to 0 °C. Dess-Martin periodinane (111
mg, 0.262 mmol) was added in one portion, and the resulting
solution was stirred at 0 °C for 30 min. The reaction was then
quenched with 1 N NaOH and then extracted with Et2O. The
organic extracts were combined and dried over MgSO4, and
the solution was filtered and concentrated in vacuo. The crude
(0.334 g, 89%): [R]25 -66.8 (c 1.16); IR (thin film) 3494 (b),
D
2953, 2856, 1646 (s) cm-1; 1H NMR δ 0.075 (s, 6 H), 0.90 (s, 9
H), 1.17 (d, 3 H, J ) 7.0 Hz), 1.50 (dq, 1 H, J ) 14.0, 2.3 Hz),
1.59 (s, 3 H), 1.79 (dt, 1 H, J ) 14.2, 9.3 Hz), 1.94 (s, 3 H),
2.20 (s, 3 H), 2.48 (s, 3 H), 3.28 (quint, 1 H, J ) 6.6 Hz), 3.86
(d, 1 H, J ) 1.0 Hz), 3.98 (m, 2 H), 4.25 (m, 2 H), 4.26 (d, 1 H,
J ) 11.5 Hz), 4.47 (d, 1 H, J ) 11.5 Hz), 5.58 (t, 1 H, J ) 5.7
Hz), 6.42 (s, 1 H), 6.68 (m, 2 H), 7.15 (d, 1 H, J ) 9.0 Hz), 7.29
(m, 5 H), 7.61 (s, 1 H); 13C NMR δ -5.1, 11.0, 11.9, 13.82, 13.84,
14.2, 18.3, 25.9, 38.2, 44.3, 59.8, 70.0, 72.0, 84.3, 124.1, 124.3,
127.6, 127.9, 128.4, 129.6, 135.2, 136.3, 137.0, 137.9, 138.5,
139.3, 144.9, 161.2, 205.5. Anal. Calcd for C35H51NO5Si: C,
70.79; H, 8.66; N, 2.36. Found: C, 71.17; H, 8.72; N, 1.98.
Diol 61. Diethylmethoxyborane (1.55 mL, 1.0 M in THF,
1.55 mmol) was added dropwise to a solution of hydroxy ketone
60 (0.668 g, 1.24 mmol) in THF (12 mL) and methanol (2 mL)
at -78 °C. After 30 min, sodium borohydride (94.6 mg, 2.50
mmol) was added. After 3 h, the reaction was treated with
acetic acid (2.0 mL) and ethyl acetate (150 mL). The organic
layer was washed with saturated aqueous NaHCO 3 and brine
and then dried, filtered, and concentrated. The crude residue
was dissolved in methanol and concentrated three times.
Purification by flash chromatography gave 61 (0.561 g, 76%)
as a yellow oil. The unstable diol was used directly in the next
step: [R]25 -44.4 (c 0.93); IR (thin film) 3421 (b), 2931, 2856
D
4230 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
material was purified by column chromatography (14:1 to 6:1
hexanes/EtOAc) to afford 120 mg (96%) of ketophosphonate
extracts were dried over MgSO4, filtered, and concentrated in
vacuo. The crude oil was purified by column chromatography
(14:1 to 6:1 hexanes/EtOAc) to give 30.7 mg (83%) of product
83. This compound was found to be extremely unstable and
70 as a mixture of two isomers: IR (thin film) 2941, 1717 cm-1
.
Anal. Calcd for C34H73O7PSi3: C, 57.58; H, 10.38. Found: C,
1
57.39; H, 10.16. See Supporting Information for the H NMR
was used immediately in the following reaction: [R]23 +38.0
D
spectrum.
(c 1.8, CH2Cl2); IR (thin film) 3433, 2943, 2864, 1721 cm-1; 1H
NMR δ 0.00 (s, 6H), 0.85 (s, 9H), 0.97 (m, 24H), 1.57 (s, 3H),
1.86 (s, 3H), 1.98 (m, 2H), 2.12 (s, 3H), 2.14 (m, 1H), 2.46 (s,
3H), 3.84 (dd, 1H, J ) 5.5, 12.5 Hz), 4.00 (dd, 2H, J ) 6.5,
13.5 Hz), 4.18 (dd, 1H, J ) 12.5, 13.5 Hz), 5.13 (m, 1H), 5.39
(t, 1H, J ) 5.5 Hz), 6.22 (m, 2H), 6.36 (d, 1H, J ) 15.0 Hz),
6.55 (dd, 1H, J ) 11, 15 Hz), 7.53 (s, 1H), 8.17 (d, 2H, J ) 9.0
Hz), 8.29 (d, 2H, J ) 9.0 Hz); 13C NMR δ -5.4, 9.7, 11.3, 12.4,
13.2, 13.8, 14.3, 18.0, 18.1, 18.2, 25.8, 37.0, 39.7, 59.5, 75.4,
76.0, 76.9, 120.3, 123.5, 124.3, 126.6, 130.6, 135.6, 135.8, 136.9,
137.4, 137.7, 137.9, 138.7, 150.5, 160.8, 164.5.
(2′-Meth yloxa zol-4′-yl)-4-[(tr iisop r op ylsilyl)oxy]-6-[(tr i-
m et h ylsilyl)oxy]-2,9,11,13-t et r a d eca t et r a en -8-on e (77).
Phosphonate 70 (880 mg, 1.28 mmol) and Ba(OH)2 (dried at
110 °C for 24 h, 405 mg, 1.28 mmol) were stirred in THF (3.2
mL) at room temperature for 30 min. Aldehyde 53 (227 mg,
1.28 mmol) was then dissolved in a THF/H2O mixture (40:1,
3.2 mL total) and added to the phosphonate via syringe. The
resulting mixture was stirred for 48 h and then quenched by
the addition of saturated aqueous NH4Cl. The mixture was
extracted with Et2O, dried over MgSO4, filtered, and concen-
trated in vacuo, and the resulting crude oil was purified by
column chromatography (hexanes to 19:1 hexanes/EtOAc) to
(2E,4S,6S,7S,8R,9E,11E,13E)-1-[(t-Bu tyld im eth ylsilyl)-
oxy]-3,7,9,13-tetr a m eth yl-8-m eth oxy-14-(2′-m eth yl-4′-ox-
a zolyl)-6-(p-n itr oben zoa te)-4-[(tr iisop r op ylsilyl)oxy]-2,9,-
11,13-tetr a d eca tetr a en e (84). In a flame-dried flask, crushed
3 Å sieves (110 mg) and Ag2O (112 mg, 0.483 mmol) were
stirred in dry benzene (0.1 mL) for 5 min. Methyl iodide (60.2
mL, 0.967 mmol) was added to the solution, which was stirred
for another 10 min. Alcohol 83 (26.3 mg, 0.032 mmol) in
benzene (0.1 mL) was added to the mixture, and the solution
was sonicated for 12 h. The reaction mixture was then loaded
directly onto a flash column and purified (hexanes to 6:1
hexanes/EtOAc) to afford 20.7 mg (77%) of the methylated
afford 769 mg (85%) of triene 77 as one olefin isomer: [R]23
D
-5.75 (c 0.4, CH2Cl2); IR (thin film) 2945, 2864, 1655, 1599
1
cm-1; H NMR δ 0.05 (s, 6H), 0.08 (s, 9H), 0.88 (s, 9H), 1.01
(m, 24H), 1.58 (s, 3H), 1.61 (m, 1H), 1.68 (m, 1H), 1.93 (s, 3H),
2.19 (s, 3H), 2.47 (s, 3H), 3.51 (m, 1H), 4.07 (m, 1H), 4.20 (m,
3H), 5.47 (t, 1H, J ) 5.0 Hz), 6.41 (s, 1H), 6.67 (m, 2H), 7.17
(d, 1H, J ) 9.0 Hz); 7.60 (s, 1H); 13C NMR δ -5.30, -5.28, 0.6,
11.4, 12.2, 12.3, 12.6, 13.8, 14.2, 18.1, 18.2, 18.3, 25.9, 41.5,
46.0, 59.8, 72.1, 75.6, 123.7, 124.5, 126.0, 136.0, 136.4, 136.8,
138.5, 138.6, 138.9, 144.1, 161.1, 203.7. Anal. Calcd for C40H73
-
material 84: [R]23 +22.2 (c 2.44, CDCl3); IR (thin film) 2930,
NO5Si3: C, 65.61; H, 10.05; N, 1.91. Found: C, 65.33; H, 10.03;
N, 1.99.
D
1723 cm-1; 1H NMR δ 0.22 (s, 6H), 0.84 (s, 9H), 0.95 (s, 21H),
1.03 (d, 3H, J ) 6.6 Hz), 1.55 (s, 3H), 1.81 (ddd, 1H, J ) 1.5,
7.0, 14.5 Hz), 1.91 (s, 3H), 1.94 (ddd, 1H, J ) 6.5, 10.5, 14.5
Hz), 2.13 (s, 3H), 2.27 (ddd, 1H, J ) 1.0, 4.0, 13.5 Hz), 2.47 (s,
3H), 3.18 (s, 3H), 3.27 (d, 1H, J ) 8.1 Hz), 3.80 (ddd, 1H, J )
0.5, 5.4, 13.0 Hz), 3.97 (dd, 1H, J ) 6.6, 13.0 Hz), 4.15 (t, 1H,
J ) 6.6 Hz), 5.06 (ddd, 1H, J ) 1.5, 4.0, 10.5 Hz), 5.35 (t, 1H,
J ) 5.5 Hz), 6.09 (d, 1H, J ) 10.8 Hz), 6.24 (s, 1H), 6.36 (d,
1H, J ) 15.3 Hz), 6.58 (dd, 1H, J ) 10.5, 15.0 Hz), 7.54 (s,
1H), 8.15 (d, 2H, J ) 9.0 Hz), 8.27 (d, 2H, J ) 9.0 Hz); 13C
NMR δ -5.4, 1.0, 10.4, 11.2, 11.9, 12.4, 13.8, 14.3, 18.0, 18.1,
25.8, 35.3, 38.9, 56.2, 59.6, 74.4, 76.0, 88.9, 120.6, 123.5, 124.1,
126.6, 129.3, 130.4, 135.8, 135.9, 136.0, 136.8, 137.5, 136.8,
138.7, 150.4, 160.8, 163.7. Anal. Calcd for C45H72N2O8Si2: C,
65.45; H, 8.79; N, 3.39. Found: C, 65.29; H, 9.02; N, 3.48.
(2E,4S,6S,7S,9E,11E,13E)-1-[(t-Bu tyld im eth ylsilyl)oxy]-
6-h ydr oxy-3,7,9,13-tetr a m eth yl-14-(2′-m eth yl-4′-oxa zolyl)-
4-[(t r iisop r op ylsilyl)oxy]-2,9,11,13-t et r a d eca t et r a en -8-
on e (78). Triene 77 (769 mg, 1.05 mmol) was dissolved in
2-propanol (15 mL) and cooled to -40 °C (CH3CN/dry ice bath),
and fluorosilicic acid (25% in water, 0.25 mL) was added
dropwise. The mixture was allowed to stir for 1 h, and the
reaction was then quenched by the addition of saturated
aqueous NaHCO3. The aqueous phase was extracted with
Et2O, and the combined organic extracts were dried over
MgSO4, filtered, and concentrated in vacuo. The crude oil was
purified by column chromatography (19:1 to 6:1 hexanes/
EtOAc) to afford 584 mg of the desired product 78 (86%)
along with an additional 37.2 mg (4%) of the isomerized
material 79: [R]23 -43.3 (c 0.6, CH2Cl2); IR (thin film) 3470,
D
(2R,3R,5S)-1-[(Tr iisop r op ylsilyl)oxy]-2-m et h yl-5-[(2-
h yd r oxy)eth yl]-7-octen e-3-ol (88). A three-necked flask was
equipped with a mechanical stirrer and a dry ice/acetone
condenser and cooled to -78 °C. Anhydrous NH3 gas (ca. 200
mL) was then condensed into the flask, and 2.0 g (87 mmol)
of sodium, broken into small pieces, was added to form a dark
blue solution. Alcohol 87 (2.5 g, 5.7 mmol) was then dissolved
in THF (50 mL) and added to the Na/NH3 solution via syringe.
The resulting mixture was stirred for 30 min at -78 °C, and
the reaction was then quenched by the addition of solid
NH4Cl until the blue color dissipated. The reaction was allowed
to warm to room temperature over several minutes, causing
most of the ammonia to evolve and giving a thick white paste.
This paste was dissolved in water, diluted with Et2O, and
extracted. The combined organic extracts were dried over K2-
CO3, filtered, and concentrated in vacuo, and the resulting
crude material was purified by column chromatography (9:1
to 2:1 hexanes/EtOAc). Diol 88 (2.0 g, 98%) was recovered as
a colorless oil: [R]23D +19.3 (c 1.2, CH2Cl2); IR (thin film) 3340,
2941 cm-1; 1H NMR δ 0.89 (d, 3H, J ) 7.0 Hz), 1.04 (m, 21H),
1.23 (m, 1H), 1.40 (m, 1H), 1.51 (ddd, 1H, J ) 4.5, 10.0, 14.0
Hz), 1.68 (m, 2H), 2.07 (m, 1H), 3.23 (br s, 2H), 3.61 (m, 1H),
3.68 (ddd, 1H, J ) 5.5, 8.0, 11.0 Hz), 3.73 (dd, 1H, J ) 3.5,
10.0 Hz), 3.80 (dd, 1H, J ) 3.5, 10.0 Hz), 3.87 (dt, 1H, J ) 3.0,
10.5 Hz), 4.97 (m, 2H), 5.74 (m, 1H); 13C NMR δ 10.6, 11.7,
17.86, 17.87, 31.0, 37.16, 37.23, 39.1, 39.6, 60.1, 68.6, 73.4,
116.3, 136.7. Anal. Calcd for C20H42O3Si: C, 66.98; H, 11.80.
Found: C, 67.01; H, 11.60.
2927, 1654 cm-1; 1H NMR δ 0.04 (s, 6H), 0.88 (s, 9H), 1.03 (m,
21H), 1.14 (d, 3H, J ) 4.5 Hz), 1.56 (s, 3H), 1.64 (ddd, 1H, J
) 4.5, 10.0, 14.5 Hz), 1.70 (ddd, 1H, J ) 2.5, 6.0, 14.0 Hz),
1.94 (s, 3H), 2.19 (s, 3H), 2.47 (s, 3H), 3.37 (m, 1H,), 3.49 (d,
1H, J ) 5.0 Hz), 3.95 (br s, 1H), 4.20 (m, 2H), 4.41 (t, 1H, J )
5.0 Hz), 5.62 (t, 1H, J ) 6.0 Hz), 6.41 (s, 1H), 6.69 (m, 2H),
7.19 (d, 1H, J ) 10.5 Hz), 7.60 (s, 1H); 13C NMR δ -5.29,
-5.26, 11.8, 12.3, 12.6, 13.8, 14.2, 14.8, 18.02, 18.04, 18.2, 25.8,
39.9, 44.2, 59.7, 70.8, 75.5, 124.3, 125.6, 135.5, 136.4, 136.9,
137.6, 1385, 139.3, 144.8, 161.2, 205.9. Anal. Calcd for C37H65
-
NO5Si2: C, 67.32; H, 9.93; N, 2.12. Found: C, 66.97; H, 10.18;
N, 2.28.
(2E,4S,6S,7S,8R,9E,11E,13E)-1-[(t-Bu tyld im eth ylsilyl)-
oxy]-3,7,9,13-tetr a m eth yl-14-(2′-m eth yl-4′-oxa zolyl)-6-(p-
n it r ob en zoa t e)-4-[(t r iisop r op ylsilyl)oxy]-2,9,11,13-t et -
r adecatetr aen -8-ol (83). In a flame-dried flask, diiodomethane
(0.033 mL, 0.41 mmol) was dissolved in THF (3.0 mL), and
this solution was added dropwise to samarium metal (73 mg,
0.48 mmol). The resulting mixture was cooled to 0 °C in the
dark for 45 min to generate a 0.1 M solution of SmI2. In a
separate flask, ketone 78 (30.2 mg, 0.046 mmol) was dissolved
in dry THF (0.15 mL) and cooled to -10 °C (MeOH/ice), and
p-nitrobenzaldehyde (34.6 mg, 0.229 mmol) was added in one
portion. The SmI2 solution (0.136 mL, 13.5 mmol) was then
added dropwise, and the mixture was stirred for 3.5 h. The
reaction was quenched with saturated aqueous NaHCO3, and
the mixture was extracted with Et2O. The combined organic
J . Org. Chem, Vol. 68, No. 11, 2003 4231

Lafontaine et al.
[4R,6R,6(1R)]-4-Allyl-6-[1-m eth yl-2-((tr iisop r op yl-sily-
l)oxy)eth yl]-4H-p yr a n -2-on e (89). In 500 mL round-
to 0 °C. To this solution were added N-methylmorpholine
N-oxide (46.1 mg, 0.393 mmol) and OsO4 (4 wt % in water,
62.0 µL, 9.80 µmol) sequentially. The mixture was warmed to
room temperature and stirred overnight. The reaction was
then quenched by the addition of solid sodium bisulfite (325
mg) to afford a dark red solution that was then concentrated
in vacuo to remove all volatiles. The resulting paste was
partitioned between 1 N HCl and EtOAc and then extracted
with Et2O (two times) and EtOAc (two times). The combined
organic extracts were washed with saturated aqueous
NaHCO3, dried over K2CO3, filtered, and concentrated. This
crude diol was used directly in the oxidative cleavage step that
follows.
a
bottomed flask equipped with a Dean-Stark trap and a
condenser, diol 88 (2.0 g, 5.6 mmol) was dissolved in 250 mL
of dry benzene. To this solution was added Ag2CO3 on Celite
(approximately 1.0 mmol/0.57 g of Celite, 32 g), and the
resulting slurry was protected from light with foil, stirred
vigorously, and heated to 80 °C for 12 h. The reaction mixture,
which had turned from green to black in color over the 12 h,
was then cooled to room temperature and filtered through
sintered glass to remove all solids. The solids were washed
sequentially with benzene (two times) and CH2Cl2 (two times),
and the combined yellow filtrate was concentrated to an oil in
vacuo. This crude oil was purified by column chromatography
(hexanes to 19:1 hexanes/EtOAc) to give 1.7 g (86%) of lac-
The crude dihydroxylated material was dissolved in THF/
water (3:1, 5.6 mL total), and NaIO4 (84.2 mg, 0.393 mmol)
was added to this solution in one portion. A white precipitate
formed after several minutes, and after 1 h the reaction was
judged to be complete by TLC analysis. The volatiles were
removed in vacuo, and the resulting paste was dissolved in
brine and extracted with Et2O. The combined organic extracts
were dried over MgSO4, filtered, and concentrated to afford
tone 89 as a colorless oil: [R]23 -18.4 (c 0.64, CH2Cl2); IR
D
(thin film) 2942, 1742 cm-1
;
1H NMR δ 0.96 (d, 3H, J )
6.7 Hz), 1.03-1.08 (m, 21H), 1.36 (m, 1H), 1.81 (m, 1H),
1.91 (d, 1H, J ) 14.3 Hz), 2.05-2.12 (m, 4H), 2.67 (ddd, 1H, J
) 1.9, 5.3, 17.5 Hz), 3.65 (dd, 1H, J ) 5.1, 9.7 Hz), 3.71 (dd,
1H, J ) 7.2, 9.8 Hz), 4.47 (ddd, 1H, J ) 2.9, 3.6, 12.1 Hz),
5.08 (m, 2H), 5.71 (m, 1H); 13C NMR δ 11.0, 11.9, 17.6, 18.0,
31.5, 32.3, 36.0, 40.5, 40.6, 64.6, 80.4, 117.6, 134.6, 171.6. Anal.
Calcd for C20H38O3Si: C, 67.74; H, 10.80. Found: C, 68.14; H,
10.99.
84.6 mg of aldehyde 99 (87%): [R]23 +4.2 (c 0.5, CH2Cl2); IR
D
(thin film) 2927, 1723 cm-1; H NMR δ 0.07 (s, 6H), 0.86 (s,
1
9H), 0.96 (m, 1H), 1.06 (m, 1H), 1.14 (d, 3H, J ) 7.0 Hz), 1.51
(d, 1H, J ) 11.0 Hz), 1.82 (d, 1H, J ) 12.5 Hz), 2.20 (m, 1H),
2.38 (d, 2H, J ) 7.0 Hz), 2.46 (m, 1H), 3.40 (dd, 1H, J ) 7.0,
14.0 Hz), 3.62 (d, 1H, J ) 11.5 Hz), 3.78 (dd, 1H, J ) 5.0, 14.0
Hz), 4.74 (d, 1H, J ) 9.0 Hz), 7.60 (t, 1H, J ) 7.0 Hz), 7.64 (t,
1H, J ) 7.5 Hz), 8.01 (d, 1H, J ) 8.0 Hz), 8.21 (d, 1H, J ) 8.0
Hz), 9.75 (d, 1H, J ) 1.5 Hz); 13C NMR δ -5.1, -3.9, 14.6,
17.9, 25.7, 28.6, 32.7, 33.1, 40.1, 50.1, 57.9, 76.1, 96.5, 122.3,
125.4, 127.6, 128.0, 136.7, 152.7, 166.3, 201.0. Anal. Calcd for
Su lfon e 96. In a round-bottomed flask equipped with a
condenser, 2-mercaptobenzothiazole (455 mg, 2.72 mmol) was
dissolved in DMF (2.0 mL) and cooled to 0 °C. NaH (60%
dispersion in oil, 108.8 mg, 2.72 mmol) then was added to the
solution in one portion, and the resulting resulting dark yellow
mixture was cooled to 0 °C and stirred for 20 min. Mesylate
95 (250 mg, 0.907 mmol) in 0.11 mL of DMF was then added
to the anion via syringe, and the cloudy mixture was heated
to 65 °C for 2 h. The reaction mixture was then passed directly
through a column of silica gel (hexanes to 4:1 hexanes/EtOAc),
C
23H35NO5S2Si: C, 55.50; H, 7.09; N, 2.81. Found: C, 55.77;
H, 6.96; N, 2.98.
Alcoh ol 105. To a solution of aldehyde 99 (47.4 mg, 95.0
µmol) in 1 mL of MeOH at 0 °C was added sodium borohydride
(7.2 mg, 0.19 mmol). The reaction was judged to be complete
by TLC analysis after 5 min of stirring and then quenched by
the addition of 0.5 M aqueous Rochelle’s salt. The mixture was
diluted with Et2O, stirred for several min, and then extracted
with Et2O. The combined organic extracts were dried over
MgSO4, filtered, and concentrated in vacuo to afford 41.8 mg
affording 295 mg (94%) of the desired sulfide: [R]23 -58.8 (c
D
0.41, CH2Cl2); IR (thin film) 2917, 1733 cm-1; 1H NMR δ 1.13
(d, 3H, J ) 7.0 Hz), 1.35 (m, 1H), 1.85 (dq, 1H, J ) 1.5, 13.5
Hz), 2.06-2.11 (m, 4H), 2.21 (ddd, 1H, J ) 3.0, 7.0, 14.0 Hz),
2.69 (ddd, 1H, J ) 1.5, 5.0, 17.0 Hz), 3.43 (dd, a of ab quartet,
1H, J ) 6.5, 13.5 Hz), 3.47 (dd, b of ab quartet, 1H, J ) 6.5,
13.5 Hz), 4.54 (dt, 1H, J ) 2.5, 12.0 Hz), 5.07 (m, 2H), 5.70
(m, 1H), 7.29 (ddd, 1H, J ) 1.0, 7.5, 8.5 Hz), 7.04 (ddd, 1H, J
) 1.0, 7.5, 8.5 Hz), 7.74 (d, 1H, J ) 8.0 Hz), 7.84 (d, 1H, J )
8.0 Hz); 13C NMR δ 13.4, 31.3, 32.0, 36.0, 36.4, 37.6, 40.3, 81.0,
117.8, 121.0, 121.5, 124.3, 126.0, 134.4, 135.2, 153.1, 166.4,
171.1. Anal. Calcd for C18H21NO2S2: C, 62.22; H, 6.09; N, 4.03.
Found: C, 62.03; H, 5.95; N, 4.08.
(88%) of the desired alcohol 105: [R]23 +3.68 (c 1.44,
D
CH2Cl2); IR (thin film) 3440, 2930, 1472 cm-1; 1H NMR δ 0.06
(s, 6H), 0.86 (s, 9H), 0.90-1.05 (m, 2H), 1.12 (d, 3H, J ) 7.0
Hz), 1.50 (m, 4H), 1.77 (m, 2H), 2.45 (m, 1H), 3.40 (dd, 1H, J
) 7.5, 14.5 Hz), 3.55 (d, 1H, J ) 13.0 Hz), 3.67 (t, 2H, J ) 6.5
Hz), 3.78 (dd, 1H, J ) 5.5, 14.0 Hz), 4.70 (dd, 1H, J ) 2.0, 9.0
Hz), 7.58 (dt, 1H, J ) 1.0, 7.5 Hz), 7.63 (dt, 1H, J ) 1.0, 7.5
Hz), 8.01 (d, 1H, J ) 8.0 Hz), 8.20 (d, 1H, J ) 8.0 Hz); 13C
NMR δ -5.0, -3.9, 14.7, 17.9, 25.7, 30.6, 32.8, 33.3, 39.1, 40.3,
58.1, 60.4, 76.5, 96.9, 122.3, 125.4, 127.6, 128.0, 136.7, 152.6,
166.4. Anal. Calcd for C23H37NO5S2Si: C, 55.28; H, 7.46; N,
2.80. Found: C, 55.68; H, 7.29; N, 2.99.
Ammonium molybdate (272 mg, 0.204 mmol) was added to
30% H2O2 (0.340 mL, 3.36 mmol) at 0 °C to give a bright yellow
solution. In a separate flask, the thioether from above (295
mg, 0.851 mmol) was dissolved in 5.1 mL of absolute EtOH
and cooled to 0 °C. The molybdate solution was then added
by syringe to the thioether, resulting in a cloudy yellow
mixture that was stirred for 24 h. The solution was diluted
with saturated aqueous NaHCO3 solution and extracted with
Et2O, and the combined organic extracts were then dried over
MgSO4, filtered, and concentrated. The resulting crude mate-
rial was purified by column chromatography (19:1 to 2:1
hexane/EtOAc) to give 234 mg (91%) of sulfone 96: [R]23D -20.0
Oxa zole 107. A solution of sulfone 106 (90.0 mg, 0.146
mmol) and aldehyde 86 (104 mg, 0.146 mmol) in 5.7 mL of
THF was cooled to -78 °C, and LiHMDS (1.0 M in THF, 0.337
mL) was added dropwise. The resulting dark yellow mixture
was stirred for 2 h at -78 °C, during which time it became
progressively darker in color. The solution was then allowed
to warm to room temperature and was stirred for 10 min after
reaching ambient temperature. The reaction was judged to be
complete by TLC analysis and was quenched by the addition
of saturated aqueous NH4Cl. The aqueous phase was extracted
with Et2O, and the combined organic extracts were dried over
MgSO4, filtered, and concentrated in vacuo. The crude material
was then purified by column chromatography (hexanes to 14:1
hexanes/EtOAc) to afford 128 mg (79%) of the coupled product
(c 0.58, CH2Cl2); IR (thin film) 2918, 1731 cm-1 1H NMR δ
;
1.11 (d, 3H, J ) 7.0 Hz), 1.27 (m, 1H), 1.89 (d, 1H, J ) 11.9
Hz), 2.09 (m, 4H), 2.66 (m, 2H), 3.50 (dd, 1H, J ) 6.4, 14.5
Hz), 3.84 (dd, 1H, J ) 6.1, 14.5 Hz), 4.60 (dt, 1H, J ) 2.6, 11.9
Hz), 5.10 (m, 2H), 5.70 (m, 1H), 7.60 (dt, 1H, J ) 1.4, 7.2 Hz),
7.65 (dt, 1H, J ) 1.5, 7.2 Hz), 8.02 (m, 1H), 8.21 (m, 1H); 13C
NMR δ 13.6, 31.2, 31.5, 32.5, 35.8, 40.3, 57.4, 81.2, 118.0, 122.3,
125.5, 127.7, 128.1, 134.2, 136.5, 152.5, 166.0, 170.6. Anal.
Calcd for C18H21NO4S2: C, 56.97; H, 5.58; N, 3.69. Found: C,
56.71; H, 5.75; N, 3.65.
(107) as a yellow foam: [R]23 +41.2 (c 0.17, CH2Cl2); IR (thin
D
film) 2954, 2866, 1724, 1530 cm-1; 1H NMR δ 0.09 (s, 3H), 0.11
(s, 3H), 0.58 (q, 6H, J ) 7.9 Hz), 0.88 (s, 9H), 0.89 (m, 9H),
0.95 (m, 21H), 1.03 (d, 3H, J ) 6.9 Hz), 1.04 (m, 3H), 1.26-
Ald eh yd e 99. Olefin 98 (97.0 mg, 0.196 mmol) was dis-
solved in a mixture of THF/water (3:1, 5.6 mL total) and cooled
4232 J . Org. Chem., Vol. 68, No. 11, 2003

Antitumor Macrolide Rhizoxin D
1.29 (m, 3H), 1.43-1.52 (m, 3H), 1.66 (s, 3H), 1.71 (m, 2H),
1.79 (m, 2H), 1.88 (s, 3H), 1.96 (m, 1H), 2.14 (s, 3H), 2.22 (m,
1H), 2.47 (s, 3H), 3.04 (dd, 1H, J ) 8.1, 9.2 Hz), 3.17 (s, 3H),
3.28 (d, 1H, J ) 8.3 Hz), 3.64 (t, 2H, J ) 6.4 Hz), 4.63 (dd, 1H,
J ) 1.4, 9.0 Hz), 5.08 (dd, 1H, J ) 2.7, 9.9 Hz), 5.47 (dd, 1H,
J ) 7.6, 15.0 Hz), 5.79 (d, 1H, J ) 10.9 Hz), 5.88 (dd, 1H, J )
10.2, 15.0 Hz), 6.10 (d, 1H, J ) 10.7 Hz), 6.26 (s, 1H), 6.36 (d,
1H, J ) 15.1 Hz), 6.60 (dd, 1H, J ) 11.1, 15.3 Hz), 7.55 (s,
1H), 8.11 (d, 2H, J ) 8.6 Hz), 8.25 (d, 2H, J ) 8.6 Hz); 13C
NMR δ -5.2, -4.0, 4.4, 6.8, 10.5, 11.7, 12.0, 12.4, 13.8, 14.4,
16.2, 18.0, 18.1, 25.8, 30.7, 34.7, 35.1, 35.6, 39.1, 39.5, 40.4,
42.0, 56.2, 60.0, 74.4, 76.4, 79.0, 89.0, 97.1, 120.6, 123.4, 124.2,
125.3, 125.9, 129.3, 130.5, 135.9, 136.0, 136.2, 136.7, 136.9,
137.56, 137.65, 138.7, 150.3, 160.9, 163.8; HRMS (FAB) m/z
calcd for C61H102N2O10Si3 (M+) 1106.6842, found 1106.6849.
P h osp h on a te 109. To a solution of diethyl phosphonoacetic
acid (100 mg, 0.509 mmol) in benzene (2 mL) was added oxalyl
chloride (133 µL, 1.52 mmol) dropwise, followed by DMF (8
µL). The visible evolution of gas ceased after ca. 10 min, and
the clear solution was stirred for an additional 40 min. The
mixture was then concentrated in vacuo to remove all volatiles,
leaving a pale yellow oil, which was dissolved in 1 mL of THF
to give a 0.5 M solution of diethylphosphonoacetyl chloride.45b
2H), 2.36 (d, 1H, J ) 11.7 Hz), 2.46 (s, 3H), 2.79 (dd, 1H, J )
9.7, 10.3 Hz), 3.15 (s, 3H), 3.20 (m, 1H), 3.99 (dd, 1H, J ) 2.8,
10.7 Hz), 4.44 (dd, 1H, J ) 2.8, 10.7 Hz), 4.62 (d, 1H, J ) 7.9
Hz), 5.06 (dd, 1H, J ) 10.1, 15.1 Hz), 5.53 (d, 1H, J ) 15.6
Hz), 5.67 (d, 1H, J ) 11.0 Hz), 6.04 (d, 1H, J ) 11.0 Hz), 6.13
(dd, 1H, J ) 10.9, 15.0 Hz), 6.24 (s, 1H), 6.33 (d, 1H, J ) 15.2
Hz), 6.61 (dd, 1H, J ) 10.8, 15.2 Hz), 6.75 (ddd, 1H, J ) 4.6,
11.5, 15.9 Hz), 7.55 (s, 1H); 13C NMR δ -5.3, -4.0, 10.1, 11.1,
11.6, 12.4, 13.8, 14.4, 17.3, 18.0, 18.1, 18.2, 25.8, 31.9, 34.2,
37.6, 38.8, 40.7, 44.9, 56.0, 73.6, 78.2, 78.6, 89.9, 97.0, 120.4,
123.3, 124.3, 125.0, 128.9, 129.3, 134.8, 135.8, 137.05, 137.11,
137.4, 138.7, 139.0, 147.4, 154.9, 166.2; HRMS (FAB) m/z calcd
for C50H83NO7Si2 (M+) 865.5708, found 865.5715.
Hem ia ceta l 113. To a solution of silyl ether 112 (34.0 mg,
0.390 mmol) in THF (0.5 mL) was added 400 µL of a mixture
of HF (49%) and pyridine (1:1 v/v) at room temperature. The
resulting solution was stirred for 9 h and then diluted with
saturated aqueous NaHCO3 and extracted with Et2O. The
combined organic extracts were dried over MgSO4, filtered, and
concentrated, and the resulting crude material was purified
by column chromatography (hexanes to 3:1 hexanes/EtOAc).
Product 113 (23.3 mg, 80%) was obtained as a mixture (∼1:1)
of isomers at the acetal position. Data for the mixture of
isomers: [R]23 +144.7 (c 0.30, CH2Cl2); IR (thin film) 3418,
D
In a separate flask, alcohol 108 (91.8 mg, 95.9 µmol) was
dissolved in THF (1 mL) and cooled to 0 °C, and pyridine (46.6
µL, 57.5 µmol) was added. A portion of the acid chloride
solution (0.85 mL) was then added to the mixture dropwise.
The resulting cloudy solution was stirred for 10 min, and the
reaction was then quenched by the addition of saturated
aqueous NH4Cl. The mixture was extracted with Et2O, and
the combined organic extracts were dried over MgSO4, filtered,
and concentrated in vacuo. Purification of the crude material
by column chromatography (9:1 to 3:1 hexanes/EtOAc) afforded
2927, 2862, 1716 cm-1; 1H NMR δ 0.29 (m, 2H), 1.01 (m, 48H),
1.07 (d, 3H, J ) 6.4 Hz), 1.11 (d, 3H, J ) 6.4 Hz), 1.26 (m,
4H), 1.57-1.70 (m, 8H), 1.74 (s, 6H), 1.91 (s, 6H), 2.04 (s, 6H),
2.22-2.28 (m, 4H), 2.40 (m, 2H), 2.45 (s, 6H), 2.87 (dd, 1H, J
) 9.6, 9.8 Hz), 2.91 (br s, 1H), 3.16 (s, 6H), 3.20 (d, 2H, J )
9.4 Hz), 3.33 (d, 1H, J ) 9.2, 10.1 Hz), 3.98 (dt, 2H, J ) 2.8,
10.7 Hz), 4.27 (m, 1H), 4.54 (d, 2H, J ) 10.7 Hz), 4.69 (d, 1H,
J ) 7.9 Hz, O-CH-OH), 5.06 (dd, 2H, J ) 9.8, 15.0 Hz), 5.36
(s, 1H, O-CH-OH), 5.51 (d, 2H, J ) 15.7 Hz), 5.56 (d, 2H, J
) 15.7 Hz), 5.66 (d, 2H, J ) 10.3 Hz), 5.69 (d, 2H, J ) 10.8
Hz), 6.04 (d, 2H, J ) 11.8 Hz), 6.11 (dd, 2H, J ) 11.4, 14.7
Hz), 6.14 (dd, 2H, J ) 11.4, 14.7 Hz), 6.24 (s, 2H), 6.33 (d, 2H,
J ) 15.2 Hz), 6.61 (dd, 2H, J ) 11.1, 14.8 Hz), 6.75 (ddd, 2H,
J ) 4.6, 10.7, 15.7 Hz), 7.55 (s, 2H).
phosphonate 109 (102 mg, 92%): [R]23 +37.3 (c 0.55,
D
CH2Cl2); IR (thin film) 2932, 2865, 1733 cm-1; 1H NMR δ 0.09
(s, 3H), 0.11 (s, 3H), 0.57 (q, 6H, J ) 7.9 Hz), 0.89 (s, 9H),
0.96 (m, 30H), 1.00 (m, 1H), 1.08 (d, 3H, J ) 6.7 Hz), 1.25 (m,
2H), 1.31 (dt, 6H, J H-H ) 7.1 Hz, J P-H ) 1.8 Hz), 1.42-1.51
(m, 3H), 1.65-1.79 (m, 6H), 1.69 (s, 3H), 1.82 (s, 3H), 2.12 (s,
3H), 2.13 (m, 1H), 2.30 (m, 1H), 2.46 (s, 3H), 2.72 (dd, 1H,
J H-H ) 14.2 Hz, J P-H ) 21.7 Hz), 2.83 (dd, 1H, J H-H ) 14.2
Hz, J P-H ) 21.7 Hz), 3.10 (m, 1H), 3.16 (s, 3H), 3.22 (d, 1H, J
) 8.3 Hz), 3.62 (t, 2H, J ) 6.5 Hz), 4.08 (t, 1H, J ) 6.5 Hz),
4.13 (m, 4H), 4.64 (dd, 1H, J ) 2.0, 9.0 Hz), 4.80 (dd, 1H, J )
2.5, 9.5 Hz), 5.54 (dd, 1H, J ) 7.8, 15.2 Hz), 5.71 (d, 1H, J )
11.2 Hz), 6.06 (d, 1H, J ) 10.3 Hz), 6.14 (dd, 1H, J ) 10.9,
15.0 Hz), 6.23 (s, 1H), 6.34 (d, 1H, J ) 15.2 Hz), 6.55 (dd, 1H,
J ) 10.8, 15.5 Hz), 7.53 (s, 1H); 13C NMR δ -5.3, -4.0, 4.3,
6.7, 10.0, 11.7, 11.9, 12.3, 12.4, 13.8, 14.3, 16.29 (d, J C-P ) 6.0
Hz), 16.32 (d, J C-P ) 6.0 Hz), 16.6, 18.00, 18.05, 18.1, 25.8,
30.7, 34.5 (d, J C-P ) 134.0 Hz), 35.28, 35.30, 38.6, 39.4, 40.3,
42.2, 56.1, 60.0, 62.39 (d, J C-P ) 6.3 Hz), 62.43 (d, J C-P ) 6.3
La cton e 114. To a solution of hemiacetal 113 (12.3 mg, 16.3
µmol) in CH2Cl2 (0.3 mL) were added 3 Å sieves (15 mg) and
tetrapropylammonium perruthenate (TPAP) (2.0 mg, 5.9
µmol) at room temperature. After stirring for 10 min, the
solution was cooled to 0 °C, and N-methylmorpholine N-oxide
(2.0 mg, 17 µmol) was added to the reaction mixture, which
quickly turned from pale to dark green. The mixture was
stirred for 5 min and was then loaded directly onto a silica
gel flash column. The column was flushed with 3:1 hexanes/
EtOAc to afford 7.5 mg (61%) of the oxidized product 114:
1
[R]23 +156.9 (c 0.36, CDCl3); H NMR δ 0.67 (ddd, 1H, J )
D
11.8, 12.0, 14.1 Hz), 1.01 (m, 24H), 1.19 (d, 3H, J ) 6.4 Hz),
1.66 (dd, 1H, J ) 2.7, 14.4 Hz), 1.72 (m, 1H), 1.76 (s, 3H), 1.76
(m, 1H), 1.90 (s, 3H), 2.00 (m, 1H), 2.12 (m, 1H), 2.14 (s, 3H),
2.15 (m, 1H), 2.27 (m, 2H), 2.47 (s, 3H), 2.52 (m, 1H), 2.76
(ddd, 1H, J ) 2.1, 4.9, 17.8 Hz), 3.16 (s, 3H), 3.21 (d, 1H, J )
9.3 Hz), 3.67 (ddd, 1H, J ) 2.6, 9.4, 9.5 Hz), 3.91 (dd, 1H, J )
2.8, 10.8 Hz), 4.54 (dd, 1H, J ) 2.9, 11.2 Hz), 5.09 (dd, 1H, J
) 9.4, 15.2 Hz), 5.61 (d, 1H, J ) 16.1 Hz), 5.69 (d, 1H, J )
10.1 Hz), 6.04 (d, 1H, J ) 10.0 Hz), 6.21 (dd, 1H, J ) 10.8,
14.8 Hz), 6.25 (s, 1H), 6.33 (d, 1H, J ) 15.2 Hz), 6.62 (dd, 1H,
J ) 10.8, 15.1 Hz), 6.75 (ddd, 1H, J ) 4.9, 11.0, 15.5 Hz), 7.55
(s, 1H); 13C NMR δ 10.0, 11.0, 11.5, 12.3, 13.7, 14.1, 16.6, 18.0,
18.2, 29.7, 34.2, 34.6, 36.8, 38.1, 38.7, 45.3, 56.1, 73.8, 78.5,
89.7, 120.0, 123.2, 124.4, 124.5, 124.6, 129.2, 130.0, 133.2,
135.8, 137.2, 137.3, 137.4, 138.2, 140.1, 145.8, 161.2, 165.6,
170.4.
Hz), 73.9, 76.2, 79.0, 88.8, 97.1, 120.3, 124.3, 125.5 (d, J C-P
3.8 Hz), 129.0, 135.7, 136.2, 136.7, 136.9, 137.3, 138.2, 138.7,
160.8, 164.8.
)
E st er 112. To a solution of aldehyde 111 (74.1 mg, 71.5
µmol) in THF (143 mL) and water (1.8 mL) was added
Ba(OH)2 (dried at 110 °C, 452 mg, 1.43 mmol). The cloudy
reaction mixture was then allowed to stir at room temperature
for 4 days, after which TLC analysis revealed the consumption
of all starting material. Solid NaHCO3 was added to the
mixture, which was then concentrated to remove most of the
THF. The resulting white residue was taken up in water and
extracted with Et2O, and the combined organic extracts were
dried over MgSO4, filtered, and concentrated in vacuo. Puri-
fication of the crude material by column chromatography
(hexanes to 14:1 hexanes/EtOAc) afforded 30.9 mg (49%) of
Rh izoxin D (2). A solution of lactone 114 (3.8 mg, 5.06
µmol) in THF (0.3 mL) was cooled to -78 °C. In a separate
flask, a dilute solution of TBAF was prepared (0.20 mL of 1.0
M TBAF in THF dissolved in an additional 10.0 mL of THF),
and 280 µL (5.6 µmol) of this solution was added to the lactone.
The reaction mixture was gradually warmed to room temper-
ature over 2.5 h and then allowed to stir at room temperature
macrocycle 112: [R]23 +118.4 (c 1.03, CH2Cl2); IR (thin film)
D
2957, 2863, 1715; 1H NMR 0.10 (s, 3H), 0.12 (s, 3H), 0.28 (m,
1H), 0.90 (s, 9H), 1.01 (m, 24H), 1.07 (d, 3H, J ) 6.3 Hz), 1.08
(m, 1H), 1.26 (m, 1H), 1.42 (m, 1H), 1.64 (m, 2H), 1.73 (s, 3H),
1.69-1.77 (m, 2H), 1.90 (s, 3H), 2.14 (s, 3H), 2.13-2.27 (m,
J . Org. Chem, Vol. 68, No. 11, 2003 4233

Lafontaine et al.
for 14.5 h. The solution then was diluted with Et2O, and the
reaction was quenched by the addition of pH 7 buffer. The
aqueous phase was extracted with Et2O, and the combined
organic extracts were dried over MgSO4, filtered, and concen-
trated in vacuo. The crude material was then purified by
column chromatography (hexanes to 1:1 hexanes/EtOAc) to
afford 2.2 mg (73%) of didesepoxyrhizoxin as a white solid,
which compared favorably with the natural product in all
respects.
10.8 Hz), 5.15 (dd, 1H, J ) 9.6, 15.2 Hz), 5.61 (d, 1H, J ) 16.0
Hz), 5.81 (d, 1H, J ) 11.0 Hz), 6.12 (d, 1H, J ) 10.8 Hz), 6.23
(dd, 1H, J ) 11.0, 15.2 Hz), 6.27 (s, 1H), 6.41 (d, 1H, J ) 15.2
Hz), 6.63 (dd, 1H, J ) 10.8, 15.2 Hz), 6.75 (ddd, 1H, J ) 5.0,
11.0, 16.0 Hz), 7.55 (s, 1H); HRMS (EI) m/z calcd for C35H47
-
NO7 ([M]+) 593.3414, found 593.3382.
Ack n ow led gm en t. Helpful discussions regarding
the J ulia olefination with Prof. Philip Kocienski and Dr.
Richard Brown are gratefully acknowledged, and in-
valuable experimental details for the preparation of
dicyclohexylboron triflate were communicated to us by
Prof. Ian Paterson. NMR spectra for rhizoxin D were
generously provided by Professors Andrew Kende and
David Williams. We wish to thank the National Science
Foundation (predoctoral fellowship to J .A.L. and NSF
CHE-9502507), Eastman Kodak (predoctoral fellowship
to D.P.P.), the Research Corporation (Cottrell Scholar
Award to J .W.L.), and the University of California
Cancer Research Coordinating Committee for funding
this endeavor.
Data for synthetic rhizoxin D: [R]23 +285 (c 0.16, MeOH);
D
1H NMR δ 0.68 (ddd, 1H, J ) 12.0, 12.0, 14.5 Hz), 1.00 (d, 3H,
J ) 7.0 Hz), 1.19 (d, 3H, J ) 6.5 Hz), 1.70 (m, 1H), 1.74 (m,
1H), 1.75 (m, 1H), 1.79 (s, 3H), 1.89 (s, 3H), 1.98 (ddd, 1H, J
) 2.0, 2.0, 14.2 Hz), 2.08 (dd, 1H, J ) 11.0, 17.9 Hz), 2.14 (m,
1H), 2.15 (s, 3H), 2.27 (m, 2H), 2.46 (s, 3H), 2.53 (m, 1H), 2.76
(ddd, 1H, J ) 2.5, 5.5, 18.0 Hz), 3.17 (s, 3H), 3.26 (d, 1H, J )
9.2 Hz), 3.67 (ddd, 1H, J ) 2.7, 9.3, 11.8 Hz), 3.91 (dd, 1H, J
) 2.8, 10.8 Hz), 4.58 (dd, 1H, J ) 3.0, 10.8 Hz), 5.15 (dd, 1H,
J ) 9.7, 15.4 Hz), 5.61 (d, 1H, J ) 16.2 Hz), 5.81 (d, 1H, J )
11.0 Hz), 6.12 (d, 1H, J ) 10.8 Hz), 6.23 (dd, 1H, J ) 10.9,
15.3 Hz), 6.27 (s, 1H), 6.41 (d, 1H, J ) 15.3 Hz), 6.63 (dd, 1H,
J ) 10.9, 15.2 Hz), 6.77 (ddd, 1H, J ) 4.8, 11.1, 15.9 Hz), 7.55
(s, 1H). HRMS (FAB) m/z calcd for C35H48NO7 ([M + H]+):
594.3431. Found: 594.3432.
Su p p or tin g In for m a tion Ava ila ble: Complete experi-
mental descriptions of transformations not included in the
Experimental Section and characterization for all new com-
pounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
Data for natural rhizoxin D:4a,49 [R]24 +287.1 (c 0.48,
D
MeOH); 1H NMR δ 0.68 (ddd, 1H, J ) 12.0, 12.0, 14.4 Hz),
1.00 (d, 3H, J ) 7.0 Hz), 1.19 (d, 3H, J ) 6.6 Hz), 1.70 (m,
1H), 1.74 (m, 1H), 1.75 (m, 1H), 1.79 (s, 3H), 1.89 (s, 3H), 1.98
(dddd, 1H, J ) 2.4, 2.4, 2.8, 14.4 Hz), 2.08 (dd, 1H, J ) 11.6,
18.0 Hz), 2.14 (m, 1H), 2.15 (s, 3H), 2.27 (m, 2H), 2.46 (s, 3H),
2.53 (m, 1H), 2.76 (ddd, 1H, J ) 2.4, 5.4, 18.0 Hz), 3.17 (s,
3H), 3.26 (d, 1H, J ) 9.2 Hz), 3.67 (ddd, 1H, J ) 2.8, 9.6, 12.0
Hz), 3.90 (dd, 1H, J ) 3.0, 11.0 Hz), 4.58 (dd, 1H, J ) 3.0,
J O034011X
(49) Portions of these data were forwarded to us by Prof. Andrew
Kende, University of Rochester.
4234 J . Org. Chem., Vol. 68, No. 11, 2003