Total Synthesis of (+)-Zaragozic Acid C

Article abstract of DOI:10.1021/jo000700m

A total synthesis of (+)-zaragozic acid C is described. Key features of the synthesis are the use of a double Sharpless asymmetric dihydroxylation reaction of diene 6 to control stereochemistry at four contiguous stereocenters from C3 to C6; the introduction of the C1-side chain by reaction between the anion derived from the dithiane monosulfoxide 27 and the core aldehyde 12; a high yielding, acid-mediated simultaneous acetonide deprotection-dithiane removal-ketalization procedure leading exclusively to the 2,8-dioxabicyclo[3.2.1]octane core 34; and a novel triple oxidation procedure allowing installation of the tricarboxylic acid.

Full text of DOI:10.1021/jo000700m

7020
J . Org. Chem. 2000, 65, 7020-7032
Tota l Syn th esis of (+)-Za r a gozic Acid C
Alan Armstrong,* Paul A. Barsanti, Lyn H. J ones, and Ghafoor Ahmed
School of Chemistry, University of Nottingham, Nottingham NG7 2RD, U.K.
A.Armstrong@ic.ac.uk.
Received May 8, 2000
A total synthesis of (+)-zaragozic acid C is described. Key features of the synthesis are the use of
a double Sharpless asymmetric dihydroxylation reaction of diene 6 to control stereochemistry at
four contiguous stereocenters from C3 to C6; the introduction of the C1-side chain by reaction
between the anion derived from the dithiane monosulfoxide 27 and the core aldehyde 12; a high
yielding, acid-mediated simultaneous acetonide deprotection-dithiane removal-ketalization pro-
cedure leading exclusively to the 2,8-dioxabicyclo[3.2.1]octane core 34; and a novel triple oxidation
procedure allowing installation of the tricarboxylic acid.
In tr od u ction
chains, although some of the natural products show lower
oxidation levels at C6 and C7.
The development of drugs that inhibit cholesterol
biosynthesis has proved to be an effective strategy for
reducing the risk of hypercholesterolemia, and hence
coronary heart disease.1 The successful compounds cur-
rently on the market are inhibitors of HMGCoA reduc-
tase, and thus act early in the biosynthetic pathway.
Intervention at this point can interfere with the produc-
tion of other important steroids, leading to possible side
effects. In the pursuit of therapeutic agents that act at a
later point of the biosynthetic pathway, attention has
recently focused on inhibition of the enzyme squalene
synthase. Random screening procedures led to the inde-
pendent discovery by three groups2 of a family of natural
products possessing exceptionally potent squalene syn-
thase inhibitory properties. These compounds were named
by the Merck workers as the zaragozic acids, while the
Glaxo group termed them the squalestatins. An example
of these compounds, zaragozic acid C 1, illustrates the
key features common to most of the natural product
family: an intriguing and highly oxygenated 2,8-dioxa-
bicyclo[3.2.1]octane core, displaying a tricarboxylic acid
unit; an alkyl side chain at C1; and a C6-acyl unit.
Several members of the series have been reported,
commonly differing at the C1-alkyl and C6-acyl side
In view of the potent biological activity of these
compounds and their intriguing structures, it is not
surprising that interest in their synthesis has been
intense.3 As well as numerous approaches to the bicyclic
core,3,4 several total syntheses have been achieved: the
groups of Nicolaou5 and Heathcock6 have accomplished
(3) For an excellent review of the chemistry and biology of the
zaragozic acids, see: Nadin, A.; Nicolaou, K. C. Angew. Chem., Int.
Ed. Engl. 1996, 35, 1622.
(4) Recent studies in addition to those cited in ref 3 and ref 11b:
J otterand, N.; Vogel, P. Synlett 1999, 1883. J otterand, N.; Vogel, P.
Tetrahedron Lett. 1999, 40, 5499. Parsons, P. J .; Montagnon, T.; Giles,
M.; Hitchcock, P. Synlett 1999, 1407. Fallon, G. D.; J ones, E. D.;
Perlmutter, P.; Selajarern, W. Tetrahedron Lett. 1999, 40, 7435.
Taillefumier, C.; Lakhrissi, M.; Chapleur, Y. Synlett 1999, 697. Fraisse,
P.; Hanna, I.; Lallemand, J . Y.; Prange, T.; Ricard, L. Tetrahedron
1999, 55, 11819. Fraisse, P.; Hanna, I.; Lallemand, J . Y. Tetrahedron
Lett. 1998, 39, 7853. Koshimizu, H.; Baba, T.; Yoshimitsu, T.; Nagaoka,
H. Tetrahedron Lett. 1999, 40, 2777. Tomooka, K.; Kikuchi, M.; Igawa,
K.; Keong, P. H.; Nakai, T. Tetrahedron Lett. 1999, 40, 1917. Aggarwal,
V. K.; Masters, S. J .; Adams, H.; Spey, S. E.; Brown, G. R.; Foubister,
A. J . J . Chem. Soc., Perkin Trans. 1 1999, 155. Tsubuki, M.; Okita,
H.; Honda, T. Synlett 1998, 1417. Calter, M. A.; Sugathapala, P. M.
Tetrahedron Lett. 1998, 39, 8813. FreemanCook, K. D.; Halcomb, R.
L. Tetrahedron Lett. 1998, 39, 8567. Reid, A. M.; Steel, P. G. J . Chem.
Soc., Perkin Trans. 1 1998, 2795. Walker, L. F.; Connolly, S.; Wills,
M. Tetrahedron Lett. 1998, 39, 5273. DiFlorio, R.; Rizzacasa, M. A. J .
Org. Chem. 1998, 63, 8595. Mann, R. K.; Parsons, J . G.; Rizzacasa, M.
A. J . Chem. Soc., Perkin Trans. 1 1998, 1283. Maezaki, N.; Gijsen, H.
J . M.; Sun, L.-Q.; Paquette, L. A. J . Org. Chem. 1996, 61, 6685. Xu,
Y.; J ohnson, C. R. Tetrahedron Lett. 1997, 38, 1117. Brogan, J . B.;
Zercher, C. K. Tetrahedron Lett. 1998, 39, 1691. Kataoka, O.; Kitagaki,
S.; Watanabe, N.; Kobayashi, J .; Nakamura, S.; Shiro, M.; Hashimoto,
S. Tetrahedron Lett. 1998, 39, 2371.
* To whom correspondence should be addressed. Present address:
Department of Chemistry, Imperial College of Science, Technology and
Medicine, London SW7 2AY, UK. Tel: +44 (0)20 7594 5876. Fax: +44
(0)20 7594 5804.
(1) Reviews: Grundy, S. M.; Friedman, D. Curr. Probl. Cardiol.
1995, 20, 283. Yalpani, M. Chemistry and Industry 1996, 85. Alberts,
A. W. Drug Invest. 1990, 2, 9. Abe, I.; Tomesch, J . C.; Wattanasin, S.;
Prestwich, G. D. Nat. Prod. Rep. 1994, 11, 279. Procopiou, P. A.;
Watson, N. S. Prog. Med. Chem. 1996, 331.
(2) Bergstrom, J . D.; Kurtz, M. M.; Rew, D. J .; Amend, A. M.;
Karkas, J . D.; Bostedor, R. G.; Bansal, V. S.; Dufresne, C.; Van-
Middlesworth, F. L.; Hensens, O. D.; Liesch, J . M.; Zink, D. L.; Wilson,
K. E.; Onishi, J .; Milligan, J . A.; Bills, G.; Kaplan, L.; Nallin-Omstead,
M.; J enkins, R. G.; Huang, L.; Meinz, M. S.; Quinn, L.; Burg, R. W.;
Kong, Y. L.; Mochales, S.; Mojena, M.; Martin, I.; Pelaez, F.; Diez, M.
T.; Alberts, A. W. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 80. Dawson,
M. J .; Farthing, J . E.; Marshall, P. S.; Middleton, R. F.; O′Neill, M. J .;
Shuttleworth, A.; Stylli, C.; Tait, R. M.; Taylor, P. M.; Wildman, H.
G.; Buss, A. D.; Langley, D.; Hayes, M. V. J . Antibiot. 1992, 45, 639.
Baxter, A.; Fitzgerald, B. J .; Hutson, J . L.; McCarthy, A. D.; Motteram,
J . M.; Ross, B. C.; Sapra, M.; Snowden, M. A.; Watson, N. S.; Williams,
R. J .; Wright, C. J . Biol. Chem. 1992, 267, 11705. Hasumi, K.;
Tachikawa, K.; Sakai, K.; Murakawa, S.; Yoshikawa, N.; Kumazawa,
S.; Endo, A. J . Antibiot. 1993, 46, 689.
(5) Nicolaou, K. C.; Yue, E. W.; La Greca, S.; Nadin, A.; Yang, Z.;
Leresche, J . E.; Tsuri, T.; Naniwa, Y.; De Riccardis, F. Chem. Eur. J .
1995, 1, 467.
(6) (a) Stoermer, D.; Caron, S.; Heathcock, C. H. J . Org. Chem. 1996,
61, 9115. (b) Caron, S.; Stoermer, D.; Mapp, A. K.; Heathcock, C. H. J .
Org. Chem. 1996, 61, 9126.
10.1021/jo000700m CCC: $19.00 © 2000 American Chemical Society
Published on Web 09/16/2000

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7021
Sch em e 1a
a
Reagents and conditions: (a) LiI, AcOH, 70 °C, 1.5 h, 95%;
F igu r e 1. Retrosynthetic analysis.
(b) 2 mol % Pd2(dba)3, 4 mol % P(2-furyl)3, DMF, 65 °C, 4.5 d, 86%;
(c) Cu(2-thiophenecarboxylate), NMP, 2 h, rt, 87%; (d) DIBAL-H,
CH2Cl2, -30 °C, 94%.
the synthesis of zaragozic acid A (squalestatin 1) while
the groups of Carreira,7 Evans8 and Hashimoto9 have
prepared zaragozic acid C. Martin has also described a
synthesis of 6,7-dideoxysqualestatin H5.10
To accomplish our initial goal of 1,3-diene assembly,
we opted to employ the Stille coupling between a vinyl
stannane and vinyl iodide, in view of the high functional
group tolerance this process is known to display.12 An
appropriate vinyl stannane 4 (Scheme 1) was readily
obtained by a literature procedure involving palladium-
catalyzed cis-addition of Bu3SnH to an alkyne.13 We spent
some time investigating various methods of preparing a
suitable vinyl iodide, but best results were obtained by
addition of lithium iodide14 to acetylenic ester 2 in acetic
acid, which proceeded cleanly to give 3 as a single
stereoisomer according to 1H NMR analysis. After screen-
ing several alternative sets of reaction conditions, pal-
ladium-catalyzed cross coupling of 3 and 4 was found to
proceed reproducibly in 86% yield in the presence of
trifuryl phosphine. In our preliminary communication,
we stated that the addition of ZnCl2 was necessary in
this coupling step, but we have since found that this is
not required. More recently, we have found this coupling
can be effected in good yield at room temperature using
stoichiometric copper(I) thiophenecarboxylate as de-
scribed by Liebeskind.15 With this efficient preparation
of 5 established, DIBAL-H reduction then provided the
alcohol 6 which could be converted to several ether or
ester derivatives 7-9 under standard conditions.
In this paper, we describe in full11 our own work in
this area, which has also led to a total synthesis of 1.
Resu lts a n d Discu ssion
Our retrosynthetic analysis is shown in Figure 1. We
envisaged late introduction of the C1-alkyl and C6-acyl
side chains in order to allow a flexible route to several
members of the natural product family. Our first key
disconnection was to imagine the introduction of the
tricarboxylic acid moiety in 1 by oxidation of protected
hydroxymethyl groups at C3, C4, and C5 in I. This would
expand the range of anion chemistry that could be
performed at a late stage in the synthesis, as well as
greatly facilitating the alkene oxidation chemistry that
was to be a major feature of our strategy (vide infra).
We hoped that I would be formed as the thermodynami-
cally most stable of the possible ketal isomers upon
treatment of acyclic precursor II with acid. Protected
polyol II could itself be obtained by double Sharpless
asymmetric dihydroxylation of 1,3-diene III, with the C7-
stereocenter subsequently being introduced by addition
of an acyl anion equivalent to a C7-aldehyde. Diene III
would in turn arise from an sp2-sp2 coupling between
appropriately functionalized alkene precursors IV and
V.
The Sharpless asymmetric dihydroxylation (AD) is now
well established as one of the most reliable and selective
methods for asymmetric synthesis.16 Application of Sharp-
less’s simple mnemonic device16 to the diene 5, with the
overriding factor considered to be the need to place the
olefinic hydrogens in the unhindered “south east” quad-
rant, led to the prediction that both alkenes required the
â-series of ligand, derived from dihydroquinidine, to
provide the stereochemistry of the natural product. Thus,
At the outset of our work, there was no precedent for
several of the key aspects of this strategy. While our work
was underway, however, Nicolaou’s total synthesis of
zaragozic acid A was published which employs a concep-
tually similar route.5 However, our approach as described
in this paper effects several of the key oxidation steps
simultaneously at several sites in the molecule, allowing
a significant shortening of the overall synthesis.
(12) Farina, V.; Krishnamurthy, V.; Scott, W. J . Org. React. 1997,
50, 1.
(7) Carreira, E. M.; Du Bois, J . J . Am. Chem. Soc. 1995, 117, 8106.
(8) Evans, D. A.; Barrow, J . C.; Leighton, J . L.; Robichaud, A. J .;
Sefkow, M. J . J . Am. Chem. Soc. 1994, 116, 12111.
(9) Sato, H.; Nakamura, S.; Watanabe, N.; Hashimoto, S. Synlett
1997, 451.
(10) Martin, S. F.; Naito, S. J . Org. Chem. 1998, 63, 7592.
(11) (a) Armstrong, A.; Barsanti, P. A. Synlett 1995, 903. (b)
Armstrong, A.; J ones, L. H.; Barsanti, P. A. Tetrahedron Lett. 1998,
39, 3337.
(13) Zhang, H. X.; Guibe´, F.; Balavoine, G. J . Org. Chem. 1990, 55,
1857.
(14) Piers, E.; Wong, T.; Coish, P. D.; Rogers, C. Can. J . Chem. 1994,
72, 1816. Lu, X.; Ma, S. J . Chem. Soc., Chem. Commun. 1990, 1643.
(15) Allred, G. D.; Liebeskind, L. S. J . Am. Chem. Soc. 1996, 118,
2748.
(16) For an excellent review of the Sharpless AD reaction, see: Kolb,
H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem. Rev. 1994, 94,
2483.

7022 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
Sch em e 2a
we hoped to effect a double asymmetric dihydroxylation
to introduce simultaneously the four stereocenters at C3
to C6. Our initial attempts to effect dihydroxylation of
diene 5 under standard AD conditions were not promis-
ing. Exposure of 5 to the commercial AD-Mix â-reagent
(stoichiometric co-oxidant K3Fe(CN)6) in the presence of
1 equiv MeSO2NH2 resulted in very low conversion.
Bitmann had reported that the use of K2S2O8 as ad-
ditional co-oxidant allowed the use of reduced volumes
of solvent and enhanced rates.17 We therefore exposed 5
again to standard AD-mix conditions at higher concen-
tration (0.34 M in alkene), and with 1 eq K2S2O8 added.
Pleasingly, almost all of the starting material had disap-
peared by TLC after 2 h, affording what was assigned
a
Reagents and conditions: see Table 1.
Ta ble 1. Dou ble Asym m etr ic Dih yd r oxyla tion of Dien es
6-9
ratio of
1
by H NMR analysis of the crude product as a mixture
entry diene conditionsa diastereomersb eec (%) yield (%)
of regioisomeric diols. However, even after several days
under forcing conditions, with several equivalents of
reagents, very little further reaction was observed. The
small amounts of fully dihydroxylated products that were
formed underwent partial cyclization to give a mixture
of lactones, which complicated product analysis. To avoid
this problem, and to render the diene more electron rich,
we turned our attention to the alcohol 6 and its deriva-
tives. We were pleased to find that 6 did indeed appear
qualitatively to be more reactive than 5, being converted
to an inseparable mixture of regioisomeric triol products
in relatively short times. However, even after several
days at higher concentration and with larger quantities
of osmium and ligand, only small amounts (ca. 10%) of
pentaols were observed.
At this point, we became aware of some important
background work by Sharpless, who had examined the
dihydroxylation of E,E-1,4-diphenyl-1,3-butadiene.18 Sharp-
less found that dihydroxylation of this diene using AD-
mix in the two-phase tBuOH/H2O solvent system stopped
cleanly at the diol, resulting from reaction at only one of
the two alkenes. However, under monophasic conditions
using NMO in acetone/water as the co-oxidant, the same
diene was converted directly into the tetraol. Sharpless
explained this difference in reactivity in the two co-
oxidant systems in terms of a “two-cycle” mechanism. In
either solvent system, interaction of the alkene with the
osmium gives rise to an osmate ester. In the biphasic
system, this osmate ester undergoes hydrolysis to give
the product diol. In the monophasic system, however,
further contact with the co-oxidant effects oxidation of
the osmate ester to a trisoxo osmium (VIII) glycolate
species which is capable of effecting alkene dihydroxy-
lation, albeit without enantioselectivity since the chiral
ligand is not involved. Sharpless suggested that hydrogen
bonding to this intemediate trisoxo osmium (VIII) gly-
colate was responsible for the rate enhancement of the
second dihydroxylation event involving the ene diol inter-
mediate. We were therefore attracted to this monophasic
system as a possible means of effecting exhaustive
dihydroxylation of our 1,3-dienes. Initial results with
alcohol 6 were highly encouraging: a good yield of the
desired pentaol 10 was obtained with high diastereose-
lectivity (ca. 8:1) (Scheme 2 and Table 1, entry 1). To
minimize future protecting group manipulations, we also
performed dihydroxylation on the protected derivatives
7 and 8. However, these substrates afforded lower
1
2
3
4
5
6
7
8
9
6
A
A
A
A
B
8:1
2:1
3:1
24
d
d
29
76
74
37
59
76
45
d
>9:1
a
Conditions: (A) 1 mol % OsO4, 5 mol % (DHQD)2-PHAL, 3
equiv of NMO, 5:1 acetone/H2O, rt; (B) (i) AD-Mix â, 1 mol % OsO4,
5 mol % (DHQD)2-PHAL, 2 equiv of CH3SO2NH2, 2 equiv of
K2S2O8, 1:1 BuOH/H2O, 0 °C to rt, 4 days; (ii) 1 mol % OsO4, 5
t
mol % (DHQD)2-PHAL, 2 equiv of NMO, 5:1 acetone/H2O, rt.
Estimated by integration of crude 13C NMR spectrum with pulse
b
delay 10 s. c Measured by NMR analysis of the Mosher’s ester
d
derivative of compound 11. Not determined.
diastereoselectivity (entries 2 and 3). This was surprising,
since the C7-hydroxyl group was expected to lie in the
“north eastern” quadrant of the Sharpless mnemonic,
generally regarded as a “sterically neutral” region.16
While these results with the monophasic system were
highly encouraging in terms of yield and diastereoselec-
tivity, the eventual determination of the enantioselec-
tivity (vide infra) afforded disappointing results: dihy-
droxylation of 6 had proceeded in only 24% ee (Table 1,
entry 1). In work with the conventional AD-mix BuOH/
H2O system, Corey had reported that enhanced enanti-
oselectivity could be obtained in the AD of allylic alcohols
by converting them first to their para-methoxybenzoate
esters.19 However, no such improvement was observed
using 9 with the NMO acetone/H2O procedure (entry 4).
It seemed likely that the low enantioselectivities in the
monophasic NMO system were due to the intervention
of the trisoxo osmium (VIII) glycolate in Sharpless’
“second catalytic cycle”. We therefore decided to adopt a
new procedure in which the exhaustive dihydroxylation
of 6 was performed in two steps. In the first, 6 was
subjected to dihydroxylation with Super AD-Mix (com-
mercial AD-Mix supplemented with ligand (5 mol %) and
osmium tetroxide (1 mol %)). The resulting mixture of
regioisomeric triols (78% yield) was then converted to the
pentaols 10 (58% yield) on exposure to 1 mol % OsO4, 5
mol % (DHQD)2-PHAL and 2 equiv. NMO in acetone/
water. The product was obtained with good diastereose-
lectivity (>9:1) and with much improved enantioselec-
tivity (76% ee) (Table 1, entry 5). This procedure allowed
us access to multigram quantities of the key intermediate
10, possessing the correct natural product stereochem-
istry at C3-C6. However, the overall procedure required
a total reaction time of several days. In attempting to
speed up the first dihydroxylation step, we found that
(17) Byun, H.-S.; Ravi Kumar, E.; Bittman, R. J . Org. Chem. 1994,
59, 2630.
(18) Park, C. Y.; Kim, B. M.; Sharpless, K. B. Tetrahedron Lett. 1991,
32, 1003.
(19) Corey, E. J .; Guzman-Perez, A.; Noe, M. C. J . Am. Chem. Soc.
1994, 116, 12109.

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7023
Sch em e 3a
replacing the K2S2O8 with the more soluble sodium salt,
in conjunction with higher loadings of osmium and
ligand, resulted in greatly improved rates (complete
consumption of starting material in 18 h rather than 4
days), although at the expense of a slight decrease in
enantioselectivity (68% ee).
Since the first dihydroxylation step introduces asym-
metry, the question arises as to why we needed chiral
ligand in the second one. In practical terms, this is
because we found that reaction rates were greater using
the (DHQD)2-PHAL ligands than with achiral quinu-
clidine. An additional intriguing possibility was that if
the second dihydroxylation were performed using a chiral
reagent, the enantioselectivity of the final product might
be improved relative to that of the intermediate triols,
since the minor enantiomer from the first dihydroxylation
could be converted to the minor diastereomer of 10. We
tested this possibility by dividing a sample of the
intermediate triols from the first dihydroxylation step
into two portions, submitting one to the second, monopha-
sic dihydroxylation in the presence of (DHQD)2-PHAL
ligand, and one in the presence of quinuclidine. The
pentaols from the two reactions had equal enantiomeric
excesses, suggesting that the species responsible for the
second dihydroxylation was likely to be achiral, such as
the trisoxo osmium (VIII) glycolate suggested by Sharp-
less.
While our work was underway, the Nicolaou group
reported their results on the dihydroxylation of a range
of similar 1,3-dienes en route to the first total synthesis
of zaragozic acid A.5 Our overall process compares well
to the Nicolaou work, which provided material of 78% ee
but in low yield (20%; 40% based on recovered starting
material). Nicolaou went on to perform a second achiral
dihydroxylation later in the synthesis. As will be seen, a
major advantage of carrying out the two dihydroxylations
consecutively, as we have done, is the resulting reduction
in the number of synthetic steps due to the possibility of
carrying out subsequent manipulations of protection
group and oxidation state at several sites simultaneously.
With pentaol 10 in hand, differential protection of the
hydroxyl groups was accomplished by a sequence of
selective primary pivalate esterification, bis-acetonide
formation and reductive removal of the pivalate ester
(Scheme 3). The minor diastereomer from the dihydroxy-
lation could be removed from the resulting alcohol 11 by
column chromatography, and the enantiomeric excess
determined at this stage by formation of the Mosher’s
ester derivative. Clearly, for an eventual total synthesis,
we desired enantiomerically pure 11. We discovered that
two successive recrystallizations of 11 from cold methanol
provided racemic crystals but an enantiomerically en-
riched mother liquor with 96% ee (61% recovery of the
desired enantiomer). At this point, however, the stereo-
chemical assignment to our dihydroxylation products
rested solely on the Sharpless mnemonic device. Using
material of ca. 76% ee, we decided to proceed to bicyclic
ketal derivatives in order to facilitate stereochemical
assignments. Oxidation to the aldehyde 12 was ac-
complished under Swern conditions, although care was
required in order to avoid R-epimerization: it was neces-
sary, 1 h after addition of triethylamine at -78 °C, to
quench the reaction with saturated aqueous NH4Cl at
that temperature before allowing the reaction to warm
to room temperature. We now needed to add an acyl
anion equivalent to the C7-aldehyde 12. To give the
a
Reagents and conditions: (a) PivCl, pyridine, CH2Cl2, 5 mol
% DMAP, 22 h, 77%; (b) 2-methoxypropene, cat. p-TsOH, DMF,
rt, 17 h, 77%; (c) DIBAL-H, CH2Cl2, -78 °C, 91%; (d) (i) (COCl)2/
DMSO, CH2Cl2, -78 °C; (ii) Et3N, -78 °C, 1 h, 75%; (e) 2-lithio-
2-methyl-1,3-dithiane, THF, -78 °C, 20 min, 97%; 53% of 14 after
flash chromatography; (f) Hg(ClO4)2, CaCO3, 5:1 THF/H2O, 30 min,
92%.
correct stereochemistry at C7, we required this reaction
to proceed with chelation control by the R-oxygen of the
acetonide ring. However, we were aware of the possibility
of alternative modes of attack which may lead to the
wrong C7-epimer. We selected 2-lithio-2-methyl-1,3-
dithiane as a simple acyl anion model for the C1-side
chain, and in the event this reacted with 12 to afford the
diastereomeric alcohols 13 and 14 in essentially quan-
titative yield, but as a readily separable 1:1.2 mixture.
Attempts to improve this ratio by adding chelating metal
salts (MgBr2 or ZnCl2) were not successful, however.
Removal of the 1,3-dithiane protecting group from the
correct diastereomer 14 provided the ketone 15, setting
the scene for the key acid ketalization event. We were
aware that, depending on which two of the three hy-
droxyls at C3, C4 and C5 underwent cyclization, there
were three possible isomeric ketals, two of which are
shown in Scheme 4. The third possibility, a dioxabicyclo-
[2.2.0]heptane system, was considered to be the least
thermodynamically favorable alternative due to inherent
ring strain. At this time, there was no precedent to
suggest which of the isomer(s) would be formed, but we
hoped that the natural system would be thermodynami-
cally preferred. We found that heating 15 with 2% HCl
in methanol did indeed effect cyclization, providing a
mixture of two isomeric ketals 16 (45%) and 18 (47%).
Subsequently, we discovered that the same two com-
pounds were obtained using TFA/CH2Cl2/H2O (38% 16,
38% 18). The structure and stereochemistry of ketal 16
1
was assigned by extensive H NMR studies on 16 itself
and on the derived bisacetate 17. The observed J H6-
H7 value of 2.8 Hz is characteristic of the desired ring

7024 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
Sch em e 4a
Sch em e 5a
a
Reagents and conditions: (a) BzCl, DMAP, pyridine, 50 °C,
100%; (b) H2, 10% Pd-C (cat.), 68%; (c) 3.5 equiv of (COCl)2, 7.0
equiv of DMSO, CH2Cl2, -78 °C, then 10.5 equiv of NEt3; (d)
NaClO2, 5:1.2 tBuOH/â-isoamylene, KH2PO4, 10 °C; (e) N,N′-
diisopropyl-O-tert-butylisourea, CH2Cl2, rt, 24 h, 45% from 20.
a
Reagents and conditions: (a) 2% HCl, MeOH, rt, 10.5 h, then
50 °C, 5 h, 45% of 16 and 47% of 18; (b) 20:10:1 CH2Cl2/TFA/H2O,
rt, 16 h, 38% of 16 and 38% of 18; (c) Ac2O, pyridine, CH2Cl2, 1.5
d (73%); (d) 20:10:1 CH2Cl2/TFA/H2O, rt, 16 h, 78%.
After these model ketalization studies, Hashimoto also
suggested that formation of a mixture of ketal isomers
could be attributed to differential rates of protecting
group hydrolysis.9 Whatever the true mechanistic expla-
nation, this modification represents a substantial im-
provement on the original synthesis of the model core
because formation of the undesired ketal isomer 18 has
been eliminated. Additionally, since mercury(II)-medi-
ated deprotection of the dithiane moiety is no longer
required, the synthesis of the core is shortened by one
step.
system and of the correct 6R, 7R relative stereochemistry.
Important cross-peaks in the NOESY spectrum of 17
were observed between the C1-methyl group and H7, and
between H6 and H3. This last interaction was confirmed
in a nOe difference spectrum (14% enhancement of H3
upon irradiation of H6), providing strong evidence for the
stereochemistry assigned and thus for the sense of
diastereoselectivity of the dihydroxylation process.
While we were encouraged that our dihydroxylations
appeared to have provided the correct relative stereo-
chemistry, and that the correct ketal isomer could be
secured, we were intrigued by the formation of two ketal
isomers, and speculated on how the reaction might be
coerced into providing exclusively the desired ketal. Over
the past few years, this question of the formation of these
isomeric ketals has been the subject of much study, and
several factors have been implicated.20 Depending on the
substrate, the ketalization can proceed under kinetic or
thermodynamic control. In our own case, we believed that
we were observing the kinetic product ratio, since we
found that re-submitting the separate ketal isomers to
the reaction conditions did not result in their intercon-
version over a 16 h time period (the length of time taken
for the ketalization reaction to reach completion). We
speculated that the relative rates of hydrolysis of the two
acetonides present in 15 might have an effect on the ketal
ratio. Conceivably, if the C3-C4 acetonide was removed
first, then formation of a six-membered ring through
closure of the C4 hydroxyl onto the C1 carbonyl might
lead eventually to 18. Conversely, initial hydrolysis of
the C5-C6 acetonide might be followed by rapid cycliza-
tion of the C5 hydroxyl group onto the C1 carbonyl,
leading to a five-membered ring which may then remain
closed until hydrolysis of the second acetonide and
subsequent closure of the second ring occurred. To test
our hypothesis, we sought to remove this additional
complication. We would therefore require deprotection of
both acetonides before unmasking the C1 carbonyl car-
bon. This was attempted by treatment of the dithiane
14 with TFA/H2O. To our delight, this reaction effected
not only acetonide hydrolysis, but also dithiane removal
and cyclization, giving the desired ketal 16 as the only
observed isomer in 78% yield. Although dithianes are
generally resistant to acid hydrolysis, in this system
intramolecular attack of the C5 hydroxyl group onto the
presumed thionium ion intermediate would facilitate its
removal.
The remaining major challenge in our synthetic ap-
proach to the zaragozic acids was the oxidation to the
tricarboxylic acid level. We hoped from the outset that
this transformation could be performed simultaneously
at all three sites. Encouraging observations by Carreira7
had previously shown that a trialdehyde, obtained by
simultaneous oxidations at the C3 and C5 methanols and
ozonolysis of an exocyclic alkene at C4, could be converted
to the triacid using sodium chlorite. We therefore pro-
ceeded to perform initial investigations on our model
bicyclic ring system 16 (Scheme 5). Benzoate protection
was selected for the C6 and C7 hydroxyl groups since
preliminary studies showed that C6-acetates were prone
to migration onto the C5-hydroxymethyl. Benzyl ether
deprotection provided cleanly the tetraol 20 (Scheme 5).
Oxidation of tetraol 20 suffered the risk that initial
oxidation at one of the three primary alcohols would lead
to lactol/lactone formation and hence incomplete oxida-
tion. Indeed, attempted oxidation of 20 to the triacid
using J ones reagent or potassium perruthenate,21 fol-
lowed by treatment of the crude reaction product with
N,N′-diisopropyl-O-tert-butylisourea, gave a complex mix-
ture of products, none of which was the desired triester.
We reasoned that success was more likely if we employed
an activated DMSO oxidation system,22 where the prod-
uct aldehyde is not liberated until the addition of base
at the end of the reaction. Indeed, Ley and co-workers
had previously reported use of this strategy to overcome
problems of lactol/lactone formation during oxidation of
a 1,4-diol intermediate in their synthesis of the antifeed-
ant polygodial.23 Although treatment of the tetraol 20
with 3.5 equiv of the Swern reagent provided a mixture
of compounds by TLC analysis, presumably due to facile
hydrate formation, we were pleased to find that the crude
(21) Green, G.; Griffith, W. P.; Hollinshead, D. M.; Ley, S. V.;
Schro¨der, M. J . Chem. Soc., Perkin Trans. 1 1984, 681.
(22) Tidwell, T. T. Org. React. 1990, 39, 297.
(23) Hollinshead, D. M.; Howell, S. C.; Ley, S. V.; Mahon, M.;
Ratcliffe, N. M.; Worthington, P. A. J . Chem. Soc.; Perkin Trans. 1
1983, 1579.
(20) Dominey, A. P.; Goodman, J . M. Organic Lett. 1999, 1, 473,
and references therein.

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7025
Sch em e 6a
a
Reagents and conditions: (a) TBDPSCl, imidazole, DMF, 100 °C, 24 h, 93%; (b) BCl3‚SMe2, CH2Cl2, rt, 1 h, 94%; (c) (COCl)2, DMSO,
CH2Cl2, -78 °C then NEt3, 0 °C, 97%; (d) HS(CH2)3SH, BF3‚OEt2, CH2Cl2, rt, 1 h, 92%; (e) mCPBA, CH2Cl2, 0 °C, 78%, trans/cis 5:1; (f)
3 equiv of 27, 3.1 equiv of nBuLi, THF, -78 °C, 15 min then 1 equiv of 12, -78 °C, 15 min.; (g) 0.55 equiv of P2I4, 1 equiv of NEt3, CH2Cl2,
rt, dark, 15 min, 59% from 12; (h) TBAF, THF, 80 °C, 36% 28 and 32% 30; (i) Ac2O, DMAP (cat.), pyridine, 80 °C, 12 h, 85%; (j) Ac2O,
DMAP (cat.), pyridine, 60 °C, 3 h, 85%; (k) Ac2O, DMAP (cat.), pyridine, 80 °C, 26 h, 86%.
1H NMR spectrum showed three aldehydic protons.
Pinnick (sodium chlorite) oxidation24 of the crude trial-
dehyde followed by tert-butyl esterification provided the
triester 21 in pleasing overall yield (45% over three steps,
Scheme 5). Interestingly, use of larger excesses of the
Swern reagent (9 eq.) led to facile formation of a meth-
ylthiomethyl (MTM) ether on the C4 tertiary hydroxyl,
which could not be removed following treatment with
mercury(II) chloride. However, demonstration of the
successful triple oxidation allowed a highly concise
synthesis of the fully functionalized natural product core
and is likely to be of interest to other workers in the field
as it could be utilized in the progression of other reported
model syntheses.
Having optimized the acid-catalyzed ketalization reac-
tion and shown that the triple oxidation to the triacid
was a viable approach, we were now ready to focus on
the total synthesis of zaragozic acid C. This aim required
synthesis of the 1,3-dithiane 26 corresponding to the full
C1-side chain. Synthesis of 26 began with the alcohol 22,
prepared according to the route described by Carreira,7
employing Evans aldol methodology to control relative
and absolute stereochemistry. Silyl ether protection of
22 provided 23 in 93% yield (Scheme 6). Debenzylation
using BCl3‚SMe2 in CH2Cl225 followed by Swern oxidation
of the resulting alcohol 24 provided aldehyde 25. Protec-
tion with 1,3-propanedithiol in the presence of catalytic
BF3‚OEt2 furnished the fully functionalized C1 side chain
dithiane 26.
1,3-dithianes which have oxygenation in the δ-position
have been noted previously,26 but have been circum-
vented by use of, for example, sodium bases.26b In our
case, these modifications did not prove successful. These
difficulties are surprising since Nicolaou had already
reported the successful metalation of a structurally
similar dithiane using nBuLi.5 We briefly explored the
use of alternative acyl anion equivalents derived from
aldehyde 25 (e.g., the triethylsilyl cyanohydrin27), to no
avail. We considered that the problems in deprotonating
26 could be solved by conversion to the more acidic
monosulfoxide, a strategy that had previously proved
successful for Soll and Seitz.28 Treatment of dithiane 26
with mCPBA at 0 °C provided the monosulfoxide 27 as
a mixture of four possible diastereomers that appeared
as two major separable spots by TLC (more polar, 63%;
less polar, 15%). Based on literature precedent by Carey,29
we assigned the major fraction as being a mixture of two
diastereomers having the trans-relationship between the
alkyl chain and the sulfoxide oxygen, with the minor
fraction being the cis-isomers. Carey has shown that
peracid oxidation of simpler monosubstituted 1,3-dithianes
indeed gives predominantly the trans-isomer. Addition-
ally, Carey observed that the trans-compounds all ex-
hibited a one-proton multiplet centered between 3.2 and
(26) For example: (a) J ones, A. B.; Villalobos, A.; Linde, R. G., II;
Danishefsky, S. J . J . Org. Chem. 1990, 55, 2786. (b) Lipshutz, B. H.;
Garcia, E. Tetrahedron Lett. 1990, 31, 7261. (c) Khandekar, G.;
Robinson, G. C.; Stacey, N. A.; Steel, P. G.; Thomas, E. J .; Vather, S.
J . Chem. Soc., Chem. Commun. 1987, 877. (d) Ohmori, K.; Suzuki, T.;
Miyazawa, K.; Nishiyama, S.; Yamamura, S. Tetrahedron Lett. 1993,
34, 4981. (e) Toshima, H.; Goto, T.; Ichihara, A. Tetrahedron Lett. 1994,
35, 4361. (f) Park, P.; Broka, C. A.; J ohnson, B. F.; Kishi, Y. J . Am.
Chem. Soc. 1987, 109, 6205.
With the dithiane 26 now in hand, we proceeded to
investigate its coupling with the core aldehyde 12.
However, despite examining a range of bases, reaction
times, temperatures and additives, we were unable to
effect clean metalation of 26. Difficulties in metalating
(27) Lampe, T. F. J .; Hoffmann, H. M. R. Tetrahedron Lett. 1996,
37, 7695.
(28) Soll, R. M.; Seitz, S. P. Tetrahedron Lett. 1987, 28, 5457.
(29) (a) Carey, F. A.; Dailey, O. D., J r.; Hernandez, O.; Tucker, J .
R. J . Org. Chem. 1976, 41, 3975. (b) Carey, F. A.; Dailey, O. D., J r.;
Hernandez, O. J . Org. Chem. 1976, 41, 3979. (c) Carlson, R. M.;
Helquist, P. M. J . Org. Chem. 1968, 33, 2596.
(24) Bal. B. S.; Childers, W. E.; Pinnick, H. W. Tetrahedron 1981,
37, 2090.
(25) Congreve, M. S.; Davison, E. C.; Fuhry, M. A. M.; Holmes, A.
B.; Payne, A. N.; Robinson, R. A.; Ward, S. E. Synlett 1993, 663.

7026 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
1
3.5 ppm in the H NMR, assigned to the C6 equatorial
(DIAD, PPh3, nitrobenzoic acid, THF). In an alternative
approach, 29 could be oxidized to the corresponding
ketone (Dess-Martin periodinane, 62%), but reduction
using NaBH4/EtOH or Zn(BH4)2 both returned the un-
desired C7-epimer 29. Attempted reduction using L-
Selectride (lithium tri-sec-butylborohydride) resulted in
no reaction.
proton in the ring (dithiane numbering), and that the
corresponding signal appeared at higher field in the cis-
series. For the two compounds in the major fraction of
27, we observed a one-proton multiplet at 3.40-3.35 ppm
1
in the H NMR spectrum, which is therefore indicative
of the trans-isomers. The corresponding multiplet is
shifted upfield (masked by other dithiane ring protons)
in the minor fraction of 27, which is accordingly assigned
the cis-stereochemistry.
With 27 in hand, we were pleased to find that it indeed
underwent clean metalation and subsequent reaction
with simple electrophiles. Continuing with the total
synthesis, the major, trans-monosulfoxide 27 was treated
When the process of C1-side chain addition to the core
aldehyde 12 was repeated on a smaller scale (0.22 mmol
aldehyde) we obtained a rather surprising result. Addi-
tion of the metalated monosulfoxide 27 to 12 proceeded
smoothly as before, to give a mixture of inseparable
diastereomers. Reduction of the sulfoxide moiety to the
1,3-dithiane gave, by TLC, two compounds that cor-
responded to the expected adducts. However, when the
mixture was treated with TBAF to deprotect the C4′-
hydroxyl, the less polar ‘incorrect” C7 epimer 28 was
obtained, but none of the desired diastereomer 30 could
be detected. Instead, a new compound, more polar than
30, was isolated. Spectroscopic analysis suggested that
this new material was isomeric with 30, but further
elucidation of the structure proved difficult owing to the
n
with BuLi at -78 °C for 15 min, followed by addition of
the core aldehyde 12 to the resulting anion. TLC analysis
of the addition reaction mixture showed complete con-
sumption of the starting aldehyde 12 after 15 min, and
column chromatography allowed separation of unreacted
side chain. Deoxygenation of the mixture of adducts to
regenerate the 1,3-dithiane moiety was accomplished28,30
using P2I4, crucially in the presence of NEt3, providing
an inseparable mixture of C7-epimers in a ratio of 1:1
by NMR, and in 59% overall yield from the aldehyde 12
(Scheme 6). It was decided at this stage to remove the
C4′ TBDPS ether, since it was likely that the TBDPS
group would not withstand the subsequent acidic ketal-
ization conditions.5 The resulting diols 28 (less polar
isomer) and 30 were readily separable by column chro-
matography and were isolated in 36% and 32% yield,
respectively. At this stage, the assignment of C7-stere-
ochemistry was based on similarities in the 1H NMR
spectra of each isomer with the corresponding model
compounds previously synthesized (13 and 14). In par-
ticular, the shift, multiplicity and coupling constants
associated with H7 for the pairs of C7 epimers seemed
to concur. These stereochemical assignments were con-
firmed by eventual conversion of 30 to the natural
product.
1
complexity of the H NMR spectrum. Following acetyla-
tion of the C4′ hydroxyl, we were able to tentatively
assign this new material the structure 33 resulting from
acetonide migration from the C5-C6 to the C6-C7
1
position. The H NMR spectra of 33 and 31 were very
similar, except for an apparent shift in H7 from 4.78 ppm
for 31 to 5.09 ppm for 33. This observation could be a
result of H7 being part of a five-membered isopropylidene
ring in 33 and therefore experiencing a downfield shift.
We were unable to ascertain at what stage in the smaller
scale synthesis the acetonide migration had occurred, or
indeed why it had transpired. We can only speculate that
an acid-catalyzed rearrangement may have taken place
(possibly during the deoxygenation step due to the
incomplete quenching of phosphorus oxyacids) to form
the more thermodynamically stable trans-C6-C7 ac-
etonide 32.32 Since our successful model synthesis relied
upon deprotection of both acetonides before ketalization,
we hoped that conversion of either 31 or 33 to the bicyclic
core should be possible. We were pleased to find that
treatment of either 31 or 33 with TFA/H2O indeed
furnished the same ketal 34 in excellent yield (Scheme
7), thus supporting our original postulate that only
acetonide migration had taken place. Therefore, from a
synthetic viewpoint, this was not considered to be a
problem.
The scene was now set for the final major challenge in
the synthesis: oxidation to the tricarboxylic acid level.
Following the protocol already established with the model
system, benzoate protection and subsequent removal of
the benzyl ethers provided 36 (Scheme 7). We were
pleased to find that treatment of 36 with 3.5 equiv of the
Swern reagent followed by Pinnick oxidation and tert-
butyl esterification provided the triester 37 in good yield
over three steps (33%). Pleasingly, selective deprotection
of the C6- and C7-benzoates proceeded remarkably
smoothly using 0.2% K2CO3 in MeOH. Synthesis of 38
completes a formal synthesis of (+)-zaragozic acid C by
intersecting a late intermediate in the Carreira synthe-
sis.7 All that was required now was to follow the Carreira
precedent in order to complete the total synthesis of
To improve the stereoselectivity of the coupling reac-
tion we wondered if the use of a single monosulfoxide
diastereomer might make a difference. In speculative
studies, we endeavored to synthesize the monosulfoxide
side chain in diastereomerically enriched form by asym-
metric oxidation of 26 under Kagan conditions.31 Thus,
dithiane 26 was treated with Ti(OiPr)4 and BuOOH in
t
the presence of (+)-diethyl tartrate ((+)-DET) and 1
molar equivalent of water at -30 °C. These conditions
provided only the trans-monosulfoxide 27 in 41% yield
after 3 days. Although we did not determine the diaster-
eomeric purity of the sulfoxide at this stage, we found
that coupling with the core aldehyde 12 again led through
to a ca. 1:1 mixture of the C7 epimers 28 and 30. The
process was repeated with (-)-DET as the chiral ligand
in the oxidation step, but the sequence again led to a 1:1
mixture of adducts.
Since we were not able to effect a diastereoselective
addition of the side chain to the core aldehyde using this
method, we hoped that it may be possible to recycle the
undesired C7-epimer. Thus, selective acetylation of the
C4′-hydroxyl group of 28 provided 29 and similarly, the
desired epimer 30 afforded 31. However, hindered alcohol
29 proved to be unreactive to Mitsunobu inversion
(30) Denis, J . N.; Krief, A. Tetrahedron Lett. 1979, 20, 3995.
(31) Pitchen, P.; Kagan, H. B. Tetrahedron Lett. 1984, 25, 1049.
(32) Mulzer, J .; de Lasalle, P.; Friessler, A. Liebigs Ann. Chem. 1986,
1152.

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7027
Sch em e 7a
a
Reagents and conditions: (a) 20:10:1 CH2Cl2/TFA/H2O, 30 min, 90%; (b) BzCl, DMAP (cat.), pyridine, rt, 24 h, 97%; (c) H2, 10% Pd/C
(cat.), 89%; (d) 3.5 equiv of (COCl)2, 7 equiv of DMSO, CH2Cl2, -78 °C then 10.5 equiv of NEt3; (e) NaClO2, pH 3.5 aqueous phosphate
t
buffer, 5:1.2 BuOH/â-isoamylene; (f) N,N′-diisopropyl-O-tert-butylisourea, CH2Cl2, 23 h, 33% (three steps); (g) 0.2% K2CO3 in MeOH, 30
min, 75%; (h) (Boc)2O, Et3N, CH2Cl2, 4-pyrrolidinopyridine (cat.), 70%; (i) DCC, DMAP (cat.), CH2Cl2, 87%; (j) 25% TFA, CH2Cl2, 100%.
dinone from calcium hydride; and dimethylformamide from
calcium hydride or from anhydrous magnesium sulfate. Petrol
refers to petroleum ether bp 40-60 °C which was distilled prior
to use. Other solvents and reagents were purified by standard
procedures as necessary.35 Flash chromatography was per-
formed on Merck Kieselgel 60 (230-400 mesh) unless other-
wise stated. Analytical thin-layer chromatography was per-
formed using precoated glass-backed plates (Merck Kieselgel
60 F254). Chemical shifts for 1H and 13C NMR are expressed
in ppm relative to internal CHCl3, (7.26 ppm) or CDCl3 (77.0
ppm, t). J values are measured in Hertz. Multiplicities in 13C
spectra were determined by DEPT experiments and are
expressed as q (CH3), t (CH2), d (CH), and s (C). IR spectra
were recorded as thin films on NaCl plates unless otherwise
stated.
zaragozic acid C. Thus, selective Boc-protection of the C7
hydroxyl group, introduction of the C6-acyl side chain,33
and final simultaneous hydrolysis of the Boc-group and
the tert-butyl esters afforded (+)-zaragozic acid C (1),
[R]24D +9.6 (c 1.0, EtOH) (lit.34 +9.6 (c 0.3, EtOH)), HRMS
(FAB) M+Na 777.3166 (C40H50O14Na requires 777.3098),
identical to an authentic sample of the natural product
by 1H NMR, 13C NMR, IR and TLC. The successful
completion of this total synthesis served to confirm the
stereochemical assignment of the double Sharpless AD
process and the dithiane addition steps.
Con clu sion s
(Z)-Eth yl 4-(ben zyloxy)-3-iod o-2-bu ten oa te (3). To a
stirred solution of 236 (9.22 g, 42.30 mmol) in glacial acetic
acid (40 mL) was added LiI (6.23 g, 46.53 mmol). The reaction
was heated at 70 °C for 1.5h and then allowed to cool to room
temperature. The mixture was evaporated in vacuo (azeotro-
ping with toluene) and the residue was taken up in Et2O and
filtered to remove the inorganics. Additional Et2O was added
and washed with saturated aqueous NaHCO3 (×5). The
aqueous layers were combined and back extracted with Et2O
(×3), and the ethereal layers combined and washed with
saturated aqueous sodium thiosulfate (×3), then water (×2)
and saturated aqueous brine (×3). The organics were dried
(MgSO4), filtered and evaporated under reduced pressure. The
residue was purified by flash chromatography (10% Et2O/
petrol) to give iodide 3 (14.67 g, 95%) as a pale yellow liquid:
IR (film) 1726, 1632, cm-1; 1H NMR (400 MHz, CDCl3) 7.30-
7.16 (m, 5H), 6.73 (t, 1H, J 1.8), 4.48 (s, 2H), 4.16 (d, 2H, J
2.1), 4.15 (q, 2H, J 7.0), 1.22 (t, 3H, J 7.0); 13C NMR (100 MHz,
CDCl3) 164.3, 137.0, 128.4, 127.9, 127.6, 123.7, 115.8, 78.9,
72.4, 60.6, 14.1; m/z (EI) 346 (M+). Found: C, 45.5; H, 4.6.
We have achieved a relatively concise, stereoselective
synthesis of the natural product (+)-zaragozic acid C (1).
A key feature of the synthesis was the use of a double
Sharpless asymmetric dihydroxylation reaction to control
stereochemistry at four contiguous stereocenters from C3
to C6, further exemplifying the power of the Sharpless
AD reaction in synthesis. Other notable features include
the introduction of the C1-side chain using a dithiane
monosulfoxide anion mediated coupling reaction between
27 and the core aldehyde 12; a high yielding, acid-
mediated simultaneous acetonide deprotection-dithiane
removal-ketalization procedure leading exclusively to
the 2,8-dioxabicyclo[3.2.1]octane core 34; and a novel
triple oxidation procedure allowing installation of the
tricarboxylic acid. This strategy is likely to be of value
in the advancement of several of the model syntheses
reported by other groups. Our approach should be suf-
ficiently flexible to allow extension to other members of
the natural product family, investigation of which is
currently underway.
C
13H15IO3 requires C, 45.1; H, 4.4.
(2Z,4E)-Eth yl 6-ben zyloxy-(3,4-bis-(ben zyloxym eth yl))-
2,4-h exa d ien oa te (5). (a ) P d Ca ta lysis. Vinyl iodide 3 (16.50
g, 47.69 mmol) was dissolved in dry DMF (170 mL). The
solution was degassed by ultrasonication under argon flow
before the addition of P(2-furyl)3 (332 mg, 1.91 mmol) and Pd2-
(dba)3 (665 mg, 0.95 mmol). The solution was stirred under
argon for 25 min before the addition of the vinyl stannane 4
(31.93 g, 57.22 mmol) neat via cannula, rinsing in with DMF
(10 mL). The mixture was heated at 65 °C for 20h and then
allowed to cool to room temperature. The mixture was then
Exp er im en ta l Section
Gen er a l P r oced u r es. Diethyl ether and tetrahydrofuran
solvents were distilled from sodium-benzophenone ketyl;
toluene from sodium; dichloromethane and N-methylpyrolli-
(33) Santini, C.; Ball, R. G.; Berger, G. D. J . Org. Chem. 1994, 59,
2261.
(34) Dufresne, C.; Wilson, K. E.; Zink, D.; Smith, J .; Bergstrom, J .
D.; Kurtz, M.; Rew, D.; Nallin, M.; J enkins, R.; Bartizal, K.; Trainor,
C.; Bills, G.; Meinz, M.; Huang, L.; Onishi, J .; Milligan, J .; Mojena,
M.; Pelaez, F. Tetrahedron 1992, 48, 10221.
(35) Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemical, 2nd ed.; Pergamon: Oxford, 1988.
(36) Heathcock, C. H.; Stafford, J . A.; Clark, D. L. J . Org. Chem.
1992, 57, 2575.

7028 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
poured into a separating funnel containing saturated aqueous
NH4Cl (150 mL) and Et2O (200 mL). The ethereal layer was
separated and the aqueous layer extracted with Et2O (4 × 150
mL). The organics were combined and washed with saturated
aqueous sodium thiosulfate (2 × 100 mL), H2O (1 × 100 mL)
and saturated brine (3 × 100 mL), then dried (MgSO4), filtered
and evaporated under reduced pressure. The residue was
purified by flash chromatography (15% EtOAc/petrol) to give
the dienoate 5 (19.93 g, 86%) as a pale yellow oil: ΙR 1716,
of the triols (11.75 g, 24.58 mmol) in acetone (47 mL) and H2O
(9 mL) was added (DHQD)2PHAL (957 mg, 1.23 mmol) and
OsO4 (63 mg, 0.25 mmol) followed by NMO (5.76 g, 49.16
mmol). After 3.5d solid Na2S2O5 was added, followed by CH2-
Cl2 (100 mL). After a further 1 h, the mixture was filtered,
evaporated under reduced pressure, and azeotroped with
benzene. The residual black oil was purified by flash chroma-
tography (75-100% EtOAc/petrol) to give the pentaol 10 as a
>9:1 diastereomeric mixture (7.30 g, 58%) as a clear oil: [R]19
D
1
1
1636 cm-1; Η ΝΜR (270 MHz, CDCl3) 7.25-7.17 (m, 15H),
+4.5 (c 1.0 in CHCl3) at ca. 76% ee; IR (film) 3423 cm-1; H
6.00 (d, 1H, J 1.8), 5.56 (t, 1H, J 6.4), 4.48 (s, 2H), 4.43 (s,
2H), 4.37 (s, 2H), 4.15 (s, 2H), 4.10-3.99 (m, 6H, m), 1.16 (t,
3H, J 7.0); 13C NMR (100 MHz, CDCl3) 165.8 (s), 156.6 (s),
138.1 (s), 137.8 (s), 128.9 (d), 128.4 (d), 128.3 (d), 127.9 (d),
127.7 (d), 127.64 (d), 127.59 (d), 115.4 (d), 73.2 (t), 72.6 (t),
72.2 (t), 68.0 (t), 65.9 (t), 59.9 (t), 14.2 (q); m/z (CI, isobutane)
487 (M+). Found: C, 76.4; H, 7.2. C31H34O5 requires C, 76.5;
H, 7.0.
NMR (270 MHz, CDCl3) 7.38-7.21 (m, 15 H), 4.57-3.40 (m,
18H), 2.67 (br s, 1H), 1.27-1.07 (m, 2H); δC (100 MHz, CDCl3)
137.6, 136.9, 136.2, 128.9, 128.8, 128.7, 128.6, 128.5, 128.4,
128.3, 128.2, 128.13, 128.05, 128.0, 127.9, 78.7, 77.7, 74.1, 73.9,
73.5, 73.2, 71.5, 71.4, 70.5, 70.4, 63.2; m/z (CI, isobutane) 513
(M + H). Found: C, 67.6; H, 7.1. C29H36O8 requires C, 68.0;
H, 7.1.
(2R,3R,4R,5S)-6-Ben zyloxy-(3,4-bis(ben zyloxym eth yl))-
1-p iva loyloxym eth yl-h exa n -2,3,4,5-tetr a ol. To a cooled (0
°C) stirred solution of pentaol 10 (14.58 g, 28.48 mmol) in CH2-
Cl2 (100 mL) was added pivaloyl chloride (4.14 mL, 34.17
mmol) followed by DMAP (173 mg, 1.42 mmol) and pyridine
(3.46 mL, 42.7 mmol). The reaction was then allowed to come
to room temperature and stirred for 16 h, then quenched by
addition of saturated aqueous NH4Cl (50 mL). The organic
layer was separated and the aqueous extracted with CH2Cl2
(4 × 50 mL). The organics were combined then washed with
2M HCl (3 × 50 mL), H2O (2 × 50 mL), saturated aqueous
NaHCO3 (2 × 50 mL), H2O (1 × 50 mL), saturated aqueous
brine (2 × 50 mL), then dried (MgSO4), filtered and evaporated
under reduced pressure. The residue was purified by flash
chromatography (60% Et2O/petrol) to give the pivalate ester
(13.07 g, 77%) as a pale yellow oil: [R]19D +7.4 (c 1.0 in CHCl3)
(b) Cu -Med ia ted Cou p lin g.15 To a solution of vinyl iodide
3 (29.8 g, 0.086 mol) and stannane 4 (40 g, 0.072 mol) in NMP
(280 mL) at 0 °C under N2 was added Cu(I) thiophencarboxy-
late (20.28 g, 0.108 mol). The mixture was allowed to warm to
room temperature and stirred for 2 h, then diluted with ether
and filtered though an alumina pad, washing through with
ether. The combined ether washings were washed successively
with water and brine, dried (K2CO3) and evaporated. Column
chromatoraphy on silica gel (5-20% ether-petrol) gave the
product (30.39 g, 87%) as an oil; data as above.
(2Z,4E)-6-Ben zyloxy-(3,4-bis-(b en zyloxym et h yl))-2,4-
h exa d ien ol (6). To a cooled (-40 °C), stirred solution of dienic
ester 5 (32.44 g, 66.75 mmol) in dry CH2Cl2 (150 mL) was
added DIBAL-H (140 mL, 1.0 M solution in CH2Cl2, 140 mmol)
dropwise under argon over 30 min. The temperature was
raised to -20 °C and maintained for 1h, then quenched at -10
°C by careful addition of dry MeOH (10 mL). The mixture was
then stirred at 0 °C and then a solution of Rochelle’s salt (33
mL of saturated solution in 200 mL H2O) was cautiously added
in 4 portions. The mixture was then allowed to warm to room
temperature and stirred for 30 min. The organic phase was
separated and the aqueous phase was acidified with 2M HCl
until all the aluminum salts had dissolved, then extracted with
CH2Cl2 (4 × 150 mL). The organics were combined, then
washed with 2M HCl (2 × 100 mL), H2O (1 × 100 mL),
saturated aqueous brine (2 × 100 mL), then dried (MgSO4),
filtered and evaporated under reduced pressure. The residual
yellow oil was purified by flash chromatography (50-70%
Et2O/petrol) to give alcohol 6 (27.86 g, 94%) as a clear oil: IR
1
at ca. 76% ee, IR (film) 3442, 1724 cm-1; H NMR (400 MHz,
CDCl3) 7.37-7.17 (m, 15H), 4.53-4.18 (m, 10H, includes 4.52
(d, 1H, J 11.2)), 4.50 (s, 4H), 4.40 (dd, 2H, J 19.3 and 6.5),
4.35 (d, 1H, J 8.8), 4.32 (d, 1H, J 11.7), 4.24 (d, 1H, J 11.2),
4.17 (s, 1H), 4.07 (s, 1H), 3.78 (dd, 2H, J 16.6 and 10.3), 3.66-
3.63 (m, 3H, includes 3.64 (d, 1H, J 8.8)), 3.54 (br s, 1H), 3.45
(d, 1H, J 10.3), 1.20 (s, 9H); 13C NMR (100 MHz, CDCl3) 178.9
(s), 137.6 (s), 136.9 (s), 136.3 (s), 128.6 (d), 128.5 (d), 128.5 (d),
128.3 (d), 128.2 (d), 128.0 (d), 127.9 (d), 127.7 (d), 78.2 (s), 77.1
(s), 74.0 (t), 73.8 (t), 73.5 (t), 72.7 (d), 71.6 (d), 71.2 (t), 70.7 (t),
70.4 (t), 66.2 (t), 38.7 (s), 27.2 (q), 27.2 (q), 27.2 (q). Found: C,
68.5; H, 7.6. C34H44O9 requires C, 68.4; H, 7.4.
(2R,3R,4S,5S)-2,3,4,5-Bis(d i-O-isop r op ylid en e)-6-b en -
zyloxy-(3,4-bis(ben zyloxym eth yl))-1-(p iva loyloxym eth yl-
)h exa n e. To a stirred solution of the above tetraol (13.00 g,
21.81 mmol) in DMF (125 mL) containing p-TsOH (400 mg,
catalytic amount) was added dropwise 2-methoxypropene (7.84
mL, 87.25 mmol). After 13h the reaction was quenched by
addition of saturated aqueous NaHCO3 (100 mL), then ex-
tracted with Et2O (4 × 100 mL). The organics were combined
and washed with saturated aqueous NaHCO3 (1 × 100 mL),
H2O (1 × 100 mL), saturated aqueous brine (2 × 100 mL),
then dried (MgSO4), filtered and evaporated under reduced
pressure. The residue was purified by flash chromatography
(30% Et2O/petrol) to give the bis-acetonide (13.07 g, 77%) as a
1
3421 cm-1; H NMR (270 MHz, CDCl3) 7.35-7.24 (m, 15H),
5.90 (t, 1H, J 7.1), 5.64 (t, 1H, J 6.4), 4.48 (s, 2H), 4.45 (s, 2H),
4.43 (s, 2H,), 4.20-4.09 (m, 4H), 4.05 (s, 4H), 2.04 (t, 1H, J
6.4); 13C NMR (100 MHz, CDCl3) 140.9, 138.1, 138.0, 137.5,
136.6, 130.7, 129.1, 128.5, 128.4, 128.3, 127.9, 127.8, 127.7,
127.6, 73.0, 72.8, 72.2, 71.8, 66.1, 65.6, 59.1; m/z (+FAB) 467
(M + Na), 445 (M + H). Found: C, 78.7; H, 7.4. C29H32O4
requires C, 78.4; H, 7.3.
(2R,3R,4R,5S)-2,3,4,5-Tet r a h yd r oxy-6-b en zyloxy-(3,4-
bis(ben zyloxym eth yl))h exa n -1-ol (10). Double dihydroxy-
lation procedure (entry 5, Table 1). A 1 L three-necked flask
was charged with BuOH (130 mL) and H2O (160 mL). To this
pale orange oil: [R]18 -5.1 (c 1.0 in CHCl3) at ca. 76% ee; IR
t
D
1
was added AD-mix-â (44 g) followed by (DHQD)2PHAL (1.23
g, 1.58 mmol), OsO4 (80 mg, 0.32 mmol), K2S2O8 (17.05 g, 63.06
mmol), and CH3SO2NH2 (5.99 g, 63.06 mmol). This mixture
was stirred well for 20 min, then cooled to 0 °C before dropwise
addition of dienic alcohol 6 (14.00 g, 31.53 mmol) in tBuOH
(30 mL). The reaction came to room temperature after a few
hours, and stirring was continued for an additional 4 d at room
temperature, after which time solid Na2SO3 (45 g) and EtOAc
(300 mL) were added and the mixture stirred for 1 h. The
organic layer was separated and the aqueous layer extracted
with EtOAc (3 × 150 mL). The combined organics were washed
sequentially with 2 M KOH (3 × 100 mL), H2O (2 × 100 mL)
and brine (3 × 100 mL), then dried (MgSO4), filtered and
evaporated under reduced pressure. The residue was purified
by flash chromatography (80% Et2O/petrol) to give a mixture
of the intermediate triols (11.75 g, 78%). To a stirred solution
(film) 1730 cm-1; H NMR (270 MHz, CDCl3) 7.40-7.20 (m,
15H), 4.72-4.34 (m, 10H), 3.86-3.68 (m, 4H), 3.62 (d, 1H, J
10.8), 3.54 (d, 1H, J 10.8), 1.42 (s, 3H), 1.35 (s, 6H), 1.32 (s,
3H), 1.21 (s, 9H); m/z (+FAB) 677 (M + H). Found: C, 71.0;
H, 7.8.C40H52O9 requires C, 71.0; H, 7.7.
(2R,3R,4S,5S)-2,3,4,5-Bis(d i-O-isop r op ylid en e)-6-b en -
zyloxy-(3,4-bis(ben zyloxym eth yl))h exa n -1-ol (11). To a
cooled (-78 °C), stirred solution of pivaloate ester (10.68 g,
15.79 mmol) in dry CH2Cl2 (85 mL) was added DIBAL-H (39.5
mL, 1.0 M solution in CH2Cl2, 39.5 mmol) dropwise under
argon over 20 min. After 45 min the reaction was quenched
by careful addition of dry MeOH (5 mL). The mixture was
allowed to warm to 0 °C and stirred at this temperature for
10 min before careful addition of a solution of Rochelle’s salt
(33 mL of a saturated solution in 200 mL H2O) in 3 portions.
The mixture was then allowed to warm to room temperature

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7029
and stirred for 30 min. The organic phase was separated and
the aqueous phase was acidified with 2 M HCl until all the
aluminum salts had dissolved, then extracted with CH2Cl2 (4
× 100 mL). The organics were combined, then washed with
2M HCl (1 × 100 mL), H2O (1 × 100 mL), brine (2 × 100 mL),
then dried (MgSO4), filtered and evaporated under reduced
pressure. The residual yellow oil was purified by flash chro-
matography (50% Et2O/petrol) to give the title compound 11
(5R,6R)-5-(ter t-Bu tyld ip h en ylsilyloxy)-6-m eth yl-7-p h e-
n ylh ep ta n -1-ol (24). To a stirred solution of benzyl ether 23
(6.4 g, 11.6 mmol) in CH2Cl2 (100 mL) under N2 was added
BCl3‚SMe2 (40 mL, 2.0 M solution in CH2Cl2, 81.3 mmol). The
resulting brown mixture was stirred at room temperature for
1 h before the cautious addition of saturated aqueous NaHCO3.
The mixture was then extracted with Et2O (×4) and the
ethereal layers were combined then dried (MgSO4), filtered and
evaporated under reduced pressure. The residual yellow/brown
oil was purified by flash chromatography (20-50% EtOAc-
petrol) to give the title compound 24 (5.0 g, 94%) as a colorless
(8.51 g, 91%) as a clear oil: [R]18 -9.1 (c 1.0 in CHCl3) at ca.
D
76% ee; IR (film) 3491 cm-1; 1H NMR (500 MHz, CDCl3) 7.37-
7.20 (m, 15H), 4.68 (dd, 1H, J 8.0 and 2.6), 4.61 (d, 1H, J 12.3),
4.58-4.47 (m, 5H, includes 4.50 (d, 1H, J 12.6)), 4.39 (d, 1H,
J 11.9), 4.03-3.96 (m, 1H), 3.92-3.84 (m, 1H), 3.83-3.72 (m.
5H, includes 3.81 (d, 1H, J 19.6) and 3.75 (d, 1H, J 19.6)), 3.50
(d, 1H, J 11.0), 2.55 (dd, 1H, J 7.7 and 6.0), 1.46 (s, 3H), 1.37
(s, 3H), 1.35 (s, 3H), 1.34 (s, 3H); 13C NMR (100 MHz, CDCl3)
138.2 (s), 137.8 (s), 137.6 (s), 128.4 (d), 128.3 (d), 128.3 (d),
127.8 (d), 127.6 (d), 127.5 (d), 110.5 (s), 108.7 (s), 86.2 (s), 85.0
(s), 81.0 (d), 79.9 (d), 73.9 (t), 73.6 (t), 73.2 (t), 71.6 (t), 70.6 (t),
69.4 (t), 60.7 (t), 27.9 (q), 27.1 (q), 26.1 (q), 25.9 (q); HRMS
(FAB) M + Na, 615.2936. C35H44NaO8 requires M + Na,
615.2934.
oil: Rf 0.23 (20% EtOAc/petrol); [R]23 +8.9 (c 0.5, CHCl3); IR
D
(CHCl3 solution) 3625 cm-1; δH (270 MHz, CDCl3) 7.72-7.68
(m, 4H), 7.43-7.33 (m, 6H), 7.25-7.12 (m, 3H), 7.05-7.02 (m,
2H), 3.72-3.65 (m, 1H), 3.37 (t, 2H, J 6.3), 2.88 (dd, 1H, J
13.5, 5.0), 2.42 (dd, 1H, J 13.5, 9.9), 1.89-1.85 (m, 1H), 1.52-
1.13 (m, 6H), 1.09 (s, 9H), 0.81 (d, 3H, J 6.9); 13C NMR (68
MHz, CDCl3) 141.7 (s), 136.1 (d), 136.0 (d), 134.9 (s), 134.3
(s), 129.6 (d), 129.4 (d), 129.1 (d), 128.1 (d), 127.6 (d), 127.5
(d), 127.4 (d), 125.6 (d), 76.9 (d), 62.7 (t), 39.6 (d), 39.2 (t), 33.4
(t), 32.5 (t), 27.0 (q), 22.1 (t), 19.6 (s), 13.5 (q); m/z (FAB+) 461
(M + H); HRMS obsd 403.2109. C26H31SiO2 (M - tBu) requires
403.2093.
(2R,3R,4S,5S)-2,3,4,5-Bis(d i-O-isop r op ylid en e)-6-b en -
zyloxy-(3,4-bis(ben zyloxym eth yl))h exa n -1-a l (12). To a
cooled (-78 °C), stirred solution of oxalyl chloride (67 µL, 0.76
mmol in dry CH2Cl2 (1.7 mL) was added a solution of
anhydrous DMSO (107 µL, 1.52 mmol) in CH2Cl2 (1.5 mL)
dropwise via cannula over 4 min. After 14 min a solution of
alcohol 11 (300 mg, 0.51 mmol) in CH2Cl2 (1.4 mL) was added
dropwise over 5 min and stirred at -78 °C for 25 min. Then
Et3N (354 µL, 2.53 mmol) was added over 4 min. After 1h, the
reaction was quenched at -78 °C by addition of saturated
aqueous NH4Cl (0.8 mL) and the mixture was then allowed to
warm to room temperature. The organics were separated and
the aqueous layer extracted with CH2Cl2 (3 × 5 mL). The
organics were combined and washed with saturated aqueous
NH4Cl (2 × 5 mL), H2O (1 × 5 mL), saturated aqueous brine
(1 × 5 mL), then dried (MgSO4), filtered and evaporated under
reduced pressure. The residual yellow oil was purified by flash
chromatography (30% Et2O/petrol) to give the aldehyde 12 (248
(5R,6R)-5-(ter t-Bu tyld ip h en ylsilyloxy)-6-m eth yl-7-p h e-
n ylh ep ta n a l (25). To a cooled (-78 °C), stirred solution of
oxalyl chloride (27 µL, 0.30 mmol) in dry CH2Cl2 (1.0 mL) was
added anhydrous DMSO (43 µL, 0.61 mmol) slowly dropwise.
Following gas evolution, the mixture was stirred for a further
10 min at -78 °C before a solution of alcohol 24 (70 mg, 0.15
mmol) in CH2Cl2 (1.0 mL) was added dropwise over 45 min.
The resulting white mixture was stirred for a further 30 min.
at -78 °C. NEt3 (127 µL, 0.91 mmol) was then added dropwise
and the mixture stirred for 20 min. After warming to 0 °C the
reaction mixture was quenched with aqueous 1.0 M KH2PO4
and extracted with Et2O (×4). The ethereal layers were
combined and dried (MgSO4), filtered and evaporated under
reduced pressure. The residual orange oil was purified by flash
chromatography (12% EtOAc/petrol) to give the aldehyde 25
(68 mg, 97%) as a colorless oil: Rf 0.57 (20% EtOAc/petrol);
[R]23D +12.8 (c 0.48, CHCl3); IR (film) 1722 cm-1; 1H NMR (500
MHz, CDCl3) 9.49 (t, 1H, J 1.7), 7.71-7.65 (m, 4H), 7.44-7.40
(m, 2H), 7.38-7.35 (m, 4H), 7.23-7.21 (m, 2H), 7.17-7.14 (m,
1H, m), 7.03 (d, 2H, J 7.2), 3.68-3.65 (m, 1H), 2.89 (dd, 1H, J
13.3, 4.9), 2.40 (dd, 1H, J 13.3, 9.9), 2.04-2.00 (m, 2H), 1.92-
1.87 (m, 1H), 1.51-1.30 (m, 4H), 1.10 (s, 9H), 0.80 (d, 3H, J
6.8); 13C NMR (125 MHz, CDCl3) 202.3 (d), 141.4 (s), 136.0
(d), 134.6 (s), 134.1 (s), 129.6 (d), 129.5 (d), 129.0 (d), 128.1
(d), 127.5 (d), 127.4 (d), 125.6 (d), 76.1 (d), 43.4 (t), 39.6 (d),
38.9 (t), 33.0 (t), 27.1 (q), 19.6 (s), 18.5 (t), 13.7 (q); m/z (EI+)
459 (M + H). Found: C, 78.6; H, 8.7. C30H38SiO2 requires C,
78.5; H, 8.4.
mg, 83%) as a clear oil: [R]27 -15.1 (c 0.94 in CHCl3) at 68%
D
1
ee, IR (film) 1729 cm-1; H NMR (500 MHz, CDCl3) 9.70 (s,
1H), 7.36-7.23 (m, 15H), 4.86 (s, 1H), 4.61 (d, 1H, J 12.3),
4.55 (d, 1H, J 11.9), 4.51 (d, 1H, J 11.9), 4.46-4.41 (m, 3H),
4.31 (d, 1H, J 12.1), 3.90 (d, 1H, J 10.5), 3.85 (d, 1H, J 10.5),
3.71 (dd, 1H, J 10.5 and 1.8), 3.60 (dd, 1H, J 10.5 and 8.3),
3.52 (d, 1H, J 9.9), 3.48 (d, 1H, J 9.9), 1.41 (s, 3H), 1.39 (s,
3H), 1.37 (s, 3H), 1.35 (s, 3H); 13C NMR (125 MHz, CDCl3)
195.5, 138.0, 137.5, 128.4, 128.3, 127.9, 127.8, 127.8, 127.7,
127.6, 110.1, 109.9, 86.9, 85.2, 83.5, 79.5, 73.8, 73.3, 73.1, 70.6,
69.4, 68.9, 28.6, 27.0, 26.4, 26.0. Found C, 71.3; H, 7.2.
C
35H42O8 requires C, 71.2; H, 7.2.
(4′R,5′R)-2-[4′-(ter t-Bu t yld ip h en ylsilyloxy)-5′-m et h yl-
6′-p h en ylh ex-1′-yl][1,3]-d ith ia n e (26). To a stirred solution
of aldehyde 25 (4.74 g, 10.3 mmol) in dry CH2Cl2 (40 mL) under
N2 was added propane-1,3-dithiol (1.03 mL, 10.3 mmol) and
BF3‚OEt2 (258 µL, 2.1 mmol). The mixture was stirred at room
temperature for 1 h before the addition of saturated aqueous
NaHCO3. The mixture was extracted with CH2Cl2 (×4), dried
(MgSO4), filtered, and evaporated under reduced pressure. The
residual yellow oil was purified by flash chromatography (6%
EtOAc/petrol) to give the dithiane 26 (5.17 g, 92%) as a viscous
(5R,6R)-O-Ben zyl-5-(ter t-bu tyldiph en ylsilyloxy)-6-m eth -
yl-7-p h en ylh ep ta n -1-ol (23). To a stirred solution of alcohol
227 (3.19 g, 12.5 mmol) in DMF (30 mL) under N2 was added
imidazole (3.74 g, 55 mmol) followed by TBDPSCl (7.8 mL,
30.0 mmol). The mixture was stirred at 100 °C for 24 h before
dilution with Et2O. The organics were separated and washed
with aqueous 1.0 M HCl then water, dried (MgSO4), filtered
and evaporated under reduced pressure. The residual orange
oil was purified by flash chromatography (8% EtOAc/petrol)
to give the silyl ether 23 (6.40 g, 93%) as a yellow oil: Rf 0.69
colorless oil: Rf 0.35 (6% EtOAc/petrol); [R]23 +4.4 (c 0.48,
D
(20% EtOAc/petrol); [R]23 +8.0 (c 0.50, CHCl3); IR (CHCl3
CHCl3); IR (CHCl3 solution) 1589 cm-1; δH (500 MHz, CDCl3)
7.72-7.69 (m, 4H), 7.45-7.37 (m, 6H), 7.27-7.21 (m, 2H),
7.17-7.15 (m, 1H), 7.05-7.03 (m, 2H), 3.78 (t, 1H, J 7.0), 3.71-
3.68 (m, 1H), 2.85 (dd, 1H, J 13.5, 4.8), 2.81-2.76 (m, 4H),
2.43 (dd, 1H, J 13.5, 10.2), 2.11-2.06 (m, 1H), 1.89-1.72 (m,
2H); 1.53-1.25 (m, 6H), 1.11 (s, 9H), 0.81 (d, 3H, J 6.8); 13C
NMR (68 MHz, CDCl3) 141.5 (s), 136.0 (d), 136.0 (d), 134.8
(s), 134.1 (s), 129.5 (d), 129.4 (d), 129.0 (d), 128.0 (d), 127.6
(d), 127.5 (d), 127.3 (d), 125.5 (d), 76.3 (d), 47.2 (d), 39.3 (d),
35.0 (t), 33.3 (t), 30.2 (t), 27.2 (q), 25.9 (t), 22.9 (t), 19.5 (s),
D
solution) 1589 cm-1; 1H NMR (270 MHz, CDCl3) 7.72-7.67 (m,
4H), 7.41-7.11 (m, 14H), 7.04-7.00 (m, 2H), 4.39 (s, 2H),
3.72-3.66 (m, 1H), 3.23 (t, 2H, J 7.5), 2.86 (dd, 1H, J 13.5,
5.0), 2.41 (dd, 1H, J 13.5, 9.9), 1.87 (1H, m, CH(CH3)), 1.52-
1.27 (m, 6H), 1.09 (s, 9H), 0.78 (d, 3H); 13C NMR (68 MHz,
CDCl3) 141.7 (s), 136.0 (s), 136.0 (d), 134.9 (s), 134.3 (s), 129.5
(d), 129.4 (d), 129.1 (d), 128.3 (d), 128.1 (d), 127.5 (d), 127.5
(d), 127.3 (d), 125.5 (d), 76.7 (d), 72.7 (t), 70.1 (t), 39.5 (d), 39.2
(t), 33.5 (t), 29.5 (t), 27.2 (q), 22.5 (t), 19.6 (s), 13.4 (q); m/z
(EI+) 551 (M + H). Found: C, 80.4; H, 8.5. C37H46SiO2 requires
C, 80.7; H, 8.4.
13.3 (q); m/z (EI+) 548 (M+); HRMS obsd 491.1892. C29H35
-
t
OSiS2 (M - Bu) requires 491.1899.

7030 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
1,3-Dith ia n e m on osu lfoxid es (27). To a cooled (0 °C)
stirred solution of dithiane 26 (805 mg, 1.47 mmol) in CH2Cl2
(10.0 mL) under N2 was added mCPBA (362 mg of 70% purity,
1.47 mmol). The reaction mixture was stirred for 1 min. before
the addition of aqueous sodium sulfite solution. The mixture
was extracted with CH2Cl2 (×4), dried (MgSO4), filtered and
evaporated under reduced pressure. The residual white foam
was purified by flash chromatography (100% EtOAc) to give a
separable mixture of the trans and cis pairs of diastereomers
(27 trans: 538 mg, 65%; 27 cis: 105 mg, 13%) both as white
foams, which gave complex NMR spectra indicating the
presence of mixtures of diastereomers. Data for trans: Rf 0.54
(10% MeOH/EtOAc); HRMS obsd 564.2542. C33H44S2SiO2
requires 564.2552. Data for cis isomers: Rf 0.62 (10% MeOH/
EtOAc); HRMS obsd 564.2523, C33H44S2SiO2 requires 564.2552.
(4′R,5′R,1′′R,2′′R,3′′R,4′′S,5′′S)-2-[6′′-Ben zyloxy-3′′,4′′-bis-
(ben zyloxym eth yl)-2′′,3′′,4′′,5′′-bis(d i-O-isop r op ylid en e)-
1′′-h ydr oxyh ex-1′′-yl]-2-[4′-h ydr oxy-5′-m eth yl-6′-ph en ylh ex-
1′-yl][1,3]d ith ia n e (30) a n d (4′R,5′R,1′′S,2′′R,3′′R,4′′S,5′′S)-
2-[6′′-Ben zyloxy-3′′,4′′-b is(b en zyloxym et h yl)-2′′,3′′,4′′,5′′-
b is(d i-O-isop r op ylid e n e )-1′′-h yd r oxyh e x-1′′-yl]-2-[4′-
h yd r oxy-5′-m eth yl-6′-p h en ylh ex-1′-yl][1,3]d ith ia n e (28).
To a cooled (-78 °C) stirred solution of monosulfoxide 27 (3.29
g, 5.83 mmol) in THF (17.0 mL) under Ar was added nBuLi
(2.33 mL, 2.5 M solution in THF, 5.83 mmol) slowly dropwise.
After 15 min., a solution of aldehyde 12 (1.27 g, 2.15 mmol) in
THF (8.0 mL) was added dropwise. The reaction mixture was
stirred for 15 min. before the addition of water and was then
allowed to warm to room temperature. Saturated aqueous NH4-
Cl was added and the mixture extracted with Et2O (×4), dried
(MgSO4), filtered and evaporated under reduced pressure. The
residual pale yellow foam was purified by flash chromatogra-
phy (40-100% EtOAc-petrol) to separate the adduct as a
mixture of diastereomers (2.87 g) from the excess monosul-
foxide 27 (1.77 g, 3.14 mmol).
7.36 (m, 2H), 7.34-7.22 (m, 15H), 7.18-7.15 (m, 3H), 5.04 (d,
1H, J 9.1), 4.74 (dd, 1H, J 7.8, 2.2), 4.65-4.59 (m, 3H), 4.56-
4.47 (m, 3H), 4.44 (d, 1H, J 12.1), 4.27 (d, 1H, J 3.6), 3.97 (d,
1H, J 11.0), 3.82-3.74 (m, 4H), 3.55 (d, 1H, J 11.0), 3.47-
3.44 (m, 1H), 3.28 (ddd, 1H, J 13.5, 10.0, 3.0), 3.13 (ddd, 1H,
J 13.5, 10.0, 3.0), 2.77 (dd, 1H, J 13.5, 6.2), 2.68 (ddd, 1H, J
13.5, 6.5, 3.0), 2.57 (ddd, 1H, J 13.5, 6.5, 3.0), 2.42 (dd, 1H, J
13.5, 8.7), 2.04-1.97 (m, 1H), 1.95-1.83 (m, 3H), 1.80-1.73
(m, 1H), 1.50 (s, 3H), 1.36-1.31 (m, 10H), 1.21 (s, 3H), 0.83
(d, 3H); 13C NMR (68 MHz, CDCl3) 141.2 (s), 138.1 (s), 138.0
(s), 137.5 (s), 129.1 (d), 128.4 (d), 128.2 (d), 127.9 (d), 127.7
(d), 127.6 (d), 127.5 (d), 125.7 (d), 110.9 (s), 108.2 (s), 86.2 (s),
85.4 (s), 80.7 (d), 78.5 (d), 77.2 (d), 74.0 (t), 73.6 (d), 73.5 (t),
73.2 (t), 70.7 (t), 70.2 (t), 69.0 (t), 57.1 (s), 40.4 (d), 39.9 (t),
37.8 (t), 34.9 (t), 27.7 (t), 27.7 (q), 27.0 (q), 26.1 (q), 25.6 (q),
25.1 (t), 21.2 (t), 13.1 (q); HRMS (FAB+) obsd 901.4382,
C
52H69S2O9 (M + H) requires 901.4383.
(4′R,5′R,1′′R,2′′R,3′′R,4′′S,5′′S)-2-[4′-Acetoxy-5′-m eth yl-
6′-p h en yl-h ex-1′-yl]-2-[6′′-ben zyloxy-3′′,4′′-bis(ben zyloxy-
m eth yl)-2′′,3′′,4′′,5′′-bis(d i-O-isop r op ylid en e)-1′′-h yd r ox-
yh ex-1′′-yl][1,3]d ith ia n e (31). To a stirred solution of 30 (210
mg, 0.23 mmol) in pyridine (15.0 mL) under N2 was added
acetic anhydride (133 µL, 1.40 mmol) and DMAP (cat.). The
reaction mixture was stirred at 60 °C for 3 h before the
addition of water. The mixture was extracted with Et2O (×4),
washed with saturated aqueous CuSO4, dried (MgSO4), filtered
and evaporated under reduced pressure. The residual yellow
oil was purified by flash chromatography (40% Et2O/petrol)
to give the acetate 31 (186 mg, 85%) as a white foam: Rf 0.38
(50% Et2O/petrol); [R]26 -5.6 (c 0.71, CHCl3); IR (film) 3448,
D
1733 cm-1; H NMR (500 MHz, CDCl3) 7.37-7.23 (m, 16H),
1
7.20-7.15 (m, 2H), 7.13-7.12 (m, 2H), 4.93 (s, 1H), 4.92-4.88
(m, 1H), 4.78 (d, 1H, J 8.1), 4.67 (d, 1H, J 12.6), 4.58-4.47
(m, 4H), 4.44-4.41 (m, 2H, including 4.42 (d, 1H, J 11.9)), 4.15
(d, 1H, d, J 9.5), 3.91-3.75 (m, 4H), 3.55 (d, 1H, d, J 11.0),
3.01 (m, 1H), 2.78-2.73 (m, 1H), 2.75 (dd, 1H, J 13.5, 5.0),
2.73-2.57 (m, 1H), 2.57-2.55 (m, 1H), 2.31-2.26 (m, 2H), 2.28
(dd, 1H, J 13.5, 9.7), 2.07 (s, 3H), 2.04-1.96 (m, 1H), 1.80-
1.51 (m, 7H), 1.48 (s, 3H), 1.37 (s, 6H), 1.34 (s, 3H), 0.84 (d,
3H, J 6.8); 13C NMR (68 MHz, CDCl3) 170.8 (s), 140.6 (s), 138.6
(s), 138.1 (s), 137.8 (s), 129.1 (d), 128.2 (d), 128.1 (d), 127.9
(d), 127.8 (d), 127.6 (d), 127.4 (d), 127.3 (d), 125.8 (d), 110.2
(s), 108.8 (s), 86.3 (s), 81.1 (d), 76.5 (d), 75.6 (d), 73.9 (t), 73.5
(t), 73.0 (t), 71.3 (t), 69.1 (t), 66.1 (d), 59.3 (s), 39.3 (t), 38.4 (d),
31.6 (t), 28.3 (q), 27.2 (q), 26.1 (q), 26.0 (q), 25.6 (t), 24.2 (t),
21.2 (q), 20.5 (t), 13.9 (q); m/z (FAB+) 943 (M+). Found: C,
68.9; H, 7.8. C54H70S2O10 requires C, 68.8; H, 7.5.
To a stirred solution of P2I4 (780 mg, 1.37 mmol) in CH2Cl2
(40.0 mL) under Ar in the dark was added NEt3 (347 µL, 2.49
mmol) followed by a solution of monosulfoxide adducts (2.87
g, 2.49 mmol) in CH2Cl2 (40.0 mL) via cannula. The reaction
mixture was stirred for 5 min. before the addition of aqueous
sodium thiosulfate. The mixture was extracted with CH2Cl2
(×4), washed with aqueous sodium thiosulfate (×2), dried
(MgSO4), filtered and evaporated under reduced pressure. The
residual pale yellow foam was purified by flash chromatogra-
phy (24% EtOAc-petrol) to give the 1,3-dithiane adducts (1.43
g, 59% from aldehyde 12) as a white foam and an inseparable
mixture of C7-epimers in a ∼1:1 ratio.
To a stirred solution of these adducts (720 mg, 0.80 mmol)
in THF (20.0 mL) under N2 was added TBAF (4.04 mL, 1.0 M
solution in THF, 4.04 mmol). The reaction mixture was stirred
for 16 h at 80 °C before the addition of water. The mixture
was extracted with Et2O (×4), dried (MgSO4), filtered and
evaporated under reduced pressure. The residual yellow oil
was purified by flash chromatography (40-60% Et2O-petrol)
to give a readily separable mixture of C7-epimers (30: 230
mg, 32%; 28: 260 mg, 36%) as colorless oils.
(4′R,5′R,1′′R,2′′R,3′′R,4′′S,5′′S)-2-[4′-Acetoxy-5′-m eth yl-
6′-p h en ylh ex-1′-yl]-2-[6′′-ben zyloxy-3′′,4′′-bis(ben zyloxy-
m eth yl)-1′′,2′′,4′′,5′′-bis(d i-O-isop r op ylid en e)-3′′-h yd r oxy-
h ex-1′′-yl][1,3]d ith ia n e (33). To a cooled (-78 °C) stirred
solution of trans-monosulfoxide 27 (373 mg, 0.66 mmol) in THF
(1.7 mL) under Ar was added nBuLi (297 µL, 2.3 M solution
in THF, 0.68 mmol) slowly dropwise. After 15 min., a solution
of aldehyde 12 (130 mg, 0.22 mmol) in THF (0.8 mL) was added
dropwise. The reaction mixture was stirred for 15 min. before
the addition of water and the mixture was then allowed to
warm to room temperature. Saturated aqueous NH4Cl was
added and the mixture extracted with Et2O (×4), dried
(MgSO4), filtered and evaporated under reduced pressure. The
residual pale yellow foam was purified by flash chromatogra-
phy (40-100% EtOAc/petrol) to separate the adduct as a
mixture of diastereomers (194 mg) from the excess monosul-
foxide 27 (147 mg, 0.26 mmol).
To a stirred solution of P2I4 (36 mg, 0.062 mmol) in CH2Cl2
(2.0 mL) under N2 in the dark was added NEt3 (9 µL, 0.062
mmol) followed by a solution of the intermediate mixture of
monosulfoxide adducts (131 mg, 0.114 mmol) in CH2Cl2 (2.0
mL) via syringe. The reaction mixture was stirred for 5 min.
before adding aqueous sodium thiosulfate solution. The mix-
ture was extracted with Et2O, dried (MgSO4), filtered and
evaporated under reduced pressure. The residual viscous
yellow oil was purified by flash chromatography (24% EtOAc/
petrol) to give an inseparable mixture of 1,3-dithianes (57 mg,
34% from aldehyde 12) as a colorless oil.
Data for 30: Rf 0.27 (50% Et2O/petrol); [R]18D +17.0 (c 0.74,
CHCl3); IR (film) 3456 cm-1; 1H NMR (500 MHz, CDCl3) 7.37-
7.28 (m, 17H), 7.23-7.16 (m, 3H), 4.93 (s, 1H), 4.76 (d, 1H, J
8.4), 4.65 (d, 1H, J 12.6), 4.57-4.51 (m, 4H), 4.48 (m, 1H), 4.43
(d, 1H, J 11.9), 4.17 (d, 1H, J 11.0), 3.91-3.76 (m, 4H), 3.56
(d, 1H, J 11.0), 3.51 (m, 1H), 2.83-2.81 (m, 1H), 2.78 (dd, 1H,
J 13.3, 6.3), 2.72-2.70 (m, 1H), 2.57 (m, 1H), 2.45 (dd, 1H, J
13.3, 8.6), 2.30 (m, 1H), 1.85-1.73 (m, 5H), 1.68-1.55 (m, 4H),
1.47 (s, 3H), 1.37 (s, 3H), 1.36 (s, 3H), 1.34 (s, 3H), 0.85 (d,
3H, J 6.9); 13C NMR (68 MHz, CDCl3) 141.1 (s), 138.5 (s), 138.1
(s), 137.8 (s), 129.1 (d), 128.2 (d), 128.2 (d), 128.0 (d), 128.0
(d), 127.8 (d), 127.6 (d), 127.4 (d), 127.3 (d), 125.8 (d), 110.2
(s), 108.8 (s), 86.4 (s), 81.0 (d), 77.2 (s), 75.5 (d), 73.7 (t), 73.6
(d), 73.5 (t), 73.0 (t), 71.3 (t), 69.1 (t), 66.1 (d), 59.4 (s), 40.3
(d), 39.9 (t), 35.2 (t), 34.3 (t), 28.2 (q), 27.2 (q), 26.1 (q), 26.0
(q), 25.6 (t), 24.3 (t), 21.3 (t), 13.1 (q); HRMS (FAB) obsd
923.4227, C52H68S2O9Na (M + Na) requires 923.4203.
Data for 28: Rf 0.42 (50% Et2O/petrol); [R]23 -3.8 (c 0.71,
D
CHCl3); IR (film) 3456 cm-1; 1H NMR (500 MHz, CDCl3) 7.39-

Total Synthesis of (+)-Zaragozic Acid C
J . Org. Chem., Vol. 65, No. 21, 2000 7031
To a stirred solution of the intermediate mixture of TBDPS
ethers (57 mg, 0.05 mmol) in THF (1.5 mL) under N2 was
added TBAF (150 µL, 1.0 M solution in THF, 0.15 mmol). The
reaction mixture was stirred for 12 h at 80 °C before the
addition of water. The mixture was extracted with Et2O, dried
(MgSO4), filtered and evaporated under reduced pressure. The
residual pale yellow oil was purified by flash chromatography
(40% Et2O-petrol) to give a readily separable mixture of 28
and 32 (28: 20 mg, 44%; 32: 15 mg, 33%) as colorless oils.
To a stirred solution of alcohol 32 (110 mg, 0.12 mmol) in
pyridine (5.0 mL) under N2 was added acetic anhydride (69
µL, 0.73 mmol) and DMAP (cat.). The reaction mixture was
stirred at 80 °C for 26 h before the addition of water. The
mixture was extracted with Et2O (×4), washed with saturated
aqueous CuSO4, dried (MgSO4), filtered and evaporated under
reduced pressure. The residual yellow oil was purified by flash
chromatography (32-40% Et2O/petrol) to give the acetate 33
J 12.0), 4.35 (d, 1H, J 12.0), 4.32 (d, 1H, J 11.5), 4.20 (d, 1H,
J 11.5), 4.04 (s, 1H), 3.94-3.90 (m, 2H), 3.75-3.70 (m, 3H),
3.60 (d, 1H, J 10.3), 2.72 (dd, 1H, J 13.5, 4.8), 2.25 (dd, 1H, J
13.5, 9.5), 1.96 (s, 3H), 1.94-1.85 (m, 3H), 1.70-1.48 (m, 4H),
0.80 (d, 3H, J 6.8); 13C NMR (68 MHz, CDCl3) 170.8 (s), 165.3
(s), 164.9 (s), 140.6 (s), 138.4 (s), 137.6 (s), 136.8 (s), 133.3 (d),
133.2 (d), 130.0 (d), 129.8 (d), 129.6 (s), 129.2 (s), 129.0 (d),
128.4 (d), 128.2 (d), 128.2 (d), 127.9 (d), 127.8 (d), 127.5 (d),
127.4 (d), 125.8 (d), 105.0 (s), 85.7 (s), 80.8 (d), 78.1 (d), 76.8
(d), 74.2 (d), 74.0 (t), 73.4 (t), 73.2 (t), 71.8 (s), 69.8 (t), 69.7 (t),
69.2 (t), 39.2 (t), 38.1 (d), 35.7 (t), 30.9 (t), 21.0 (q), 19.1 (t),
13.9 (q); m/z (FAB+) 985 (M + Na). Found: C, 73.6; H, 6.6.
C
59H62O12 requires C, 73.6; H, 6.6.
(1S,3R,4S,5R,6R,7R,4′R,5′R)-1-[4′-Acetoxy-5′-m eth yl-6′-
p h en ylh ex-1′-yl]-6,7-bis(ben zoyloxy)-4-h yd r oxy-3,4,5-tr i-
h yd r oxym eth yl-2,8-d ioxa bicyclo[3.2.1]octa n e (36). To a
stirred solution of tris-benzyl ether 35 (116 mg, 0.12 mmol) in
EtOAc (6.0 mL) was added Pd-C 10% (cat.) and 2.0 M aqueous
HCl (20 µL). The mixture was stirred at room temperature
under 1 atm of H2 for 5 h before being filtered through Celite
and evaporated under reduced pressure. The residual white
foam was purified by flash chromatography (100% EtOAc) to
give the alcohol 36 (74 mg, 89%) as a white foam: Rf 0.55
(99 mg, 86%) as a white foam: Rf 0.47 (50% Et2O/petrol); [R]22
D
-3.0 (c 0.99, CHCl3); IR (film) 3361, 1731 cm-1; 1H NMR (270
MHz, CDCl3) 7.30-7.12 (m, 18H), 7.08 (m, 2H), 5.09 (d, 1H, J
6.6), 4.85-4.83 (m, 2H), 4.65 (d, 1H, J 12.5), 4.54 (d, 1H, J
6.6), 4.47-4.38 (m, 5H), 4.16 (s, 1H), 3.87 (d, 1H, J 9.9), 3.78-
3.75 (m, 2H), 3.70-3.63 (m, 2H), 3.55 (d, 1H, J 9.9), 2.89 (m,
1H), 2.71 (dd, 1H, J 13.5, 4.8), 2.59-2.47 (m, 3H), 2.25 (dd,
1H, J 13.5, 9.6), 2.01 (s, 3H), 1.99-1.53 (m, 9H), 1.45 (br s,
6H), 1.38 (s, 3H), 1.35 (s, 3H), 0.79 (d, 3H, J 6.6); m/z (FAB+)
944 (M + H); HRMS obsd 941.4413. C54H69S2O10 (M - H)
requires 941.4332.
(100% EtOAc); [R]27 +87.2 (c 0.26, CHCl3); IR (film) 3425,
D
1
1726, 1602 cm-1; H NMR (500 MHz, CDCl3) 8.10-8.06 (m,
2H), 8.05-8.01 (m, 2H), 7.61-7.57 (m, 2H), 7.48-7.44 (m, 4H),
7.26-7.21 (m, 2H), 7.18-7.10 (m, 1H), 7.08-7.05 (m, 2H), 5.96
(d, 1H, J 2.9), 5.58 (d, 1H, J 2.9), 4.86 (dt, 1H, J 8.2, 4.0), 4.26
(dd, 1H, J 5.1, 3.9), 4.01-3.88 (m, 7H), ∼3.80-2.80 (br s, 3H),
2.70 (dd, 1H, J 13.4, 5.2), 2.29 (dd, 1H, J 13.4, 9.3), 2.03 (s,
3H), 1.97-1.82 (m, 3H), 1.71-1.47 (m, 4H), 0.83 (d, 3H, J 6.8);
13C NMR (125 MHz, CDCl3) 171.3 (s), 166.1 (s), 165.2 (s), 140.5
(s), 133.7 (d), 133.6 (d), 133.5 (d), 130.1 (d), 129.9 (d), 129.9
(s), 129.0 (s), 128.8 (d), 128.6 (d), 128.6 (d), 128.4 (d), 128.2
(d), 125.9 (d), 104.8 (s), 86.2 (s), 80.7 (d), 77.8 (d), 76.6 (d), 74.3
(d), 72.5 (s), 61.7 (t), 61.5 (t), 60.2 (t), 39.4 (t), 38.3 (d), 35.1 (t),
30.9 (t), 21.1 (q), 18.9 (t), 13.8 (q); m/z (FAB+) 715 (M + Na),
693 (M + H); HRMS obsd 693.2911. C38H45O12 (M + H)
requires 693.2911.
(1S,3R,4S,5R,6R,7R,4′R,5′R)-1-[4′-Acetoxy-5′-m eth yl-6′-
p h en ylh ex-1′-yl]-34,5-tr is(ben zyloxym eth yl)-4,6,7-tr ih y-
d r oxy-2,8-d ioxa bicyclo[3.2.1]octa n e (34). To a stirred so-
lution of 31 (181 mg, 0.19 mmol) in CH2Cl2 (4.0 mL) was added
water (200 µL) and TFA (2.0 mL). The mixture was stirred
for 30 min. before the addition of saturated aqueous sodium
bicarbonate. The mixture was extracted with CH2Cl2 (×4),
dried (MgSO4), filtered and evaporated under reduced pres-
sure. The residual pale yellow oil was purified by flash
chromatography (30% EtOAc/petrol) to give the ketal 34 (130
mg, 90%) as a colorless oil: Rf 0.61 (50% EtOAc/petrol); [R]21
D
+12.6 (c 0.92, CH2Cl2); IR (film) 3458, 1730 cm-1
;
1H NMR
(1S,3R,4S,5R,6R,7R,4′R,5′R)-1-[4′-Acetoxy-5′-m eth yl-6′-
p h en yl-h ex-1′-yl]-6,7-b is(ben zoyloxy)-3,4,5-t r is(2,2-d im -
e t h y lp r o p io n y l)-4-h y d r o x y -2,8-d io x a b ic y c lo [3.2.1]-
octa n e (37). To a cooled (-78 °C) stirred solution of oxalyl
chloride (31 µL, 0.35 mmol) in CH2Cl2 (1.3 mL) under N2 was
added anhydrous DMSO (51 µL, 0.71 mmol). The mixture was
stirred for 10 min. at -78 °C before a solution of 36 (70 mg,
0.10 mmol) in CH2Cl2 (1.3 mL) was added dropwise. The
resulting white mixture was stirred for 45 min before the
dropwise addition of NEt3 (148 µL, 1.06 mmol). The clear
solution was stirred for a further 30 min. before the addition
of water. The mixture was extracted with CH2Cl2 (×5), washed
with aqueous 2.0 M HCl (×1), dried (MgSO4), filtered and
evaporated under reduced pressure to give 71 mg of a colorless
oil. The product trialdehyde was used without further purifica-
tion.
To a cooled (10 °C) stirred solution of the intermediate
trialdehyde (71 mg, assume 0.10 mmol from 36) in 5:1.2
tBuOH/2-methyl-2-butene (14.8 mL) was added an ice-cold
buffered 1.1 M aqueous solution of NaClO2 (1.71 mL, 1.52
mmol) slowly dropwise. The reaction mixture was stirred for
3 h and then quenched with pH 2 KH2PO4-HCl buffer. The
mixture was extracted with EtOAc (×6) and then extracted
into saturated aqueous NaHCO3. This solution was then
reacidified with 2.0 M aqueous HCl, extracted with EtOAc
(×6), dried (MgSO4), filtered and evaporated under reduced
pressure to give 53 mg of a colorless oil. The product triacid
was used without further purification.
(500 MHz, CDCl3) 7.31-7.23 (m, 15H), 7.17-7.15 (m, 3H),
7.10-7.09 (m, 2H), 4.85-4.83 (m, 1H), 4.76 (br s, 1H), 4.54-
4.43 (m, 4H), 4.38 (m, 1H), 4.32-4.27 (m, 2H), 3.98 (br s, 1H),
3.90 (d, 1H, J 9.6), 3.77 (dd, 1H, J 10.7, 3.0), 3.59-3.56 (m,
3H), 3.46-3.45 (m, 2H), 3.32 (s, 1H), 2.71 (dd, 1H, J 13.5, 4.8),
2.30 (br s, 1H), 2.28 (dd, 1H, J 13.5, 9.6), 2.02 (s, 3H), 1.95 (m,
1H), 1.82-1.45 (m, 6H), 0.82 (d, 3H); 13C NMR (68 MHz,
CDCl3) 170.9 (s), 140.5 (s), 138.0 (s), 137.6 (s), 136.7 (s), 129.0
(d), 128.5 (d), 128.3 (d), 128.2 (d), 128.1 (d), 127.8 (d), 127.6
(d), 127.5 (d), 125.8 (d), 105.1 (s), 86.6 (s), 83.5 (d), 79.4 (d),
76.8 (d), 73.8 (t), 73.4 (t), 73.1 (t), 72.8 (d), 70.9 (s), 69.5 (t),
69.3 (t), 68.4 (t), 39.3 (t), 38.2 (d), 35.2 (t), 31.0 (t), 21.1 (q),
19.1 (t), 13.9 (q); HRMS (FAB+) obsd 755.3850, C45H55O10 (M
+ H) requires 755.3795.
Repeating the above procedure starting from 33 gave
identical results.
(1S,3R,4S,5R,6R,7R,4′R,5′R)-1-[4′-Acetoxy-5′-m eth yl-6′-
p h en ylh ex-1′-yl]-6,7-bis(ben zoyloxy)-3, 4, 5-tr is(ben zy-
loxym eth yl)-4-h ydr oxy-2,8-dioxabicyclo[3.2.1]octan e (35).
To a stirred solution of triol 34 (111 mg, 0.15 mmol) in pyridine
(3.0 mL) under N2 was added benzoyl chloride (68 µL, 0.59
mmol) and DMAP (cat.). The reaction mixture was stirred at
room temperature for 24 h before the addition of water. The
mixture was extracted with Et2O (×4), washed with saturated
aqueous CuSO4 (×4) then water (×1), 2.0 M aqueous HCl (×1)
and again with water (×1), dried (MgSO4), filtered and
evaporated under reduced pressure. The residual yellow oil
was purified by flash chromatography (40-50% Et2O/petrol)
to give the benzoate 35 (138 mg, 97%) as a colorless oil: Rf
To a solution of the intermediate triacid (53 mg) in CH2Cl2
(2.5 mL) under N2 was added N,N′-diisopropyl-O-tert-butyli-
sourea (400 µL). The reaction mixture was stirred for 23 h
during which time formation of a white precipitate was
observed. The mixture was diluted with pentane, filtered and
evaporated under reduced pressure. The residual yellow oil
was purified by flash chromatography (32% EtOAc/petrol) to
give the title compound 37 (30 mg, 33% from tris-1°-alcohol
0.41 (50% Et2O/petrol); [R]25 +53.7 (c 0.26, CHCl3); IR (film)
D
3473, 1727, 1602 cm-1; 1H NMR (500 MHz, CDCl3) 8.09-8.03
(m, 4H), 7.59 (t, 1H, J 7.4), 7.52 (t, 1H, J 7.4), 7.45 (t, 2H, J
7.7), 7.31-7.13 (m, 18H), 7.07-7.06 (m, 4H), 6.42 (d, 1H, J
2.3), 5.50 (d, 1H, J 2.3), 4.86-4.84 (m, 1H), 4.75 (dd, 1H, J
5.8, 2.2), 4.61 (d, 1H, J 12.0), 4.53 (d, 1H, J 12.0), 4.46 (d, 1H,

7032 J . Org. Chem., Vol. 65, No. 21, 2000
Armstrong et al.
36) as a white foam: Rf 0.28 (20% EtOAc/petrol); [R]28 +54.2
1.92 (m, 3H), 1.62-1.31 (m, 4H), 1.59 (s, 9H), 1.49 (s, 9H), 1.44
(s, 9H), 0.85 (d, 3H, J 6.6 Hz). These data concur with those
reported by Carreira.7
Con ver sion of 38 to 1. This was performed according to
the procedure of Carreira.7 Full details are given in the
Supporting Information.
D
(c 0.82, CHCl3); IR (film) 1760, 1730 cm-1; 1H NMR (500 MHz,
CDCl3) 8.08-8.03 (m, 2H), 8.04-8.03 (m, 2H, Ph), 7.65-7.57
(m, 2H), 7.51-7.42 (m, 4H), 7.25-7.24 (m, 2H), 7.18-7.15 (m,
1H), 7.11 (d, 2H, J 7.0), 6.82 (d, 1H, J 2.0), 5.43 (d, 1H, J 2.0),
5.14 (s, 1H), 4.90-4.87 (m, 1H), 4.16 (s, 1H), 2.76 (dd, 1H, J
13.4, 4.8), 2.28 (dd, 1H, J 13.4, 9.9), 2.16-2.06 (m, 2H,), 2.05-
2.01 (m, 1H), 2.00 (s, 3H), 1.78-1.65 (m, 4H), 1.63 (s, 9H), 1.48
(s, 9H), 1.28 (s, 9H), 0.83 (d, 3H, J 6.8); 13C NMR (125 MHz,
CDCl3) 170.8 (s), 168.6 (s), 165.6 (s), 165.1 (s), 164.3 (s), 163.5
(s), 140.8 (s), 133.7 (d), 133.5 (d), 130.1 (d), 129.8 (s), 129.1 (s),
129.0 (d), 128.9 (d), 128.6 (d), 128.4 (d), 128.2 (d), 125.8 (d),
104.4 (s), 90.2 (s), 86.3 (s), 84.4 (s), 83.6 (s), 81.4 (d), 77.0 (d),
76.4 (d), 75.5 (d), 74.0 (s), 39.3 (t), 38.0 (d), 36.0 (t), 30.9 (t),
28.1 (q), 28.0 (q), 27.9 (q), 21.1 (q), 19.1 (t), 13.8 (q); HRMS
(FAB) obsd 925.3884, C50H62O15Na (M + Na) requires 925.3986.
(1S,3R,4S,5R,6R,7R,4′R,5′R)-1-[4′-Acetoxy-5′-m eth yl-6′-
p h en ylh ex-1′-yl]-3,4,5-tr is(2,2-d im eth ylp r op ion yl)-4,6,7-
tr ih yd r oxy-2,8-d ioxa bicyclo[3.2.1]octa n e (38). Bis-ben-
zoate 37 (7 mg, 7.75 µmol) was treated with 0.2% K2CO3
methanolic solution (1.2 mL). The reaction mixture was stirred
for 5 h at room temperature before the reaction was quenched
with 0.3 M aqueous KH2PO4 solution. The mixture was
extracted with Et2O (×5), dried (MgSO4), filtered and evapo-
rated under reduced pressure. The residual colorless oil was
purified by flash chromatography (50% EtOAc/hexanes) to give
the title compound 38 (4 mg, 75%) as a colorless oil: Rf 0.28
Ack n ow led gm en t. We thank the EPSRC (GR/
L37588), the University of Nottingham (studentship to
L.H.J .), and Pfizer and Zeneca (unrestricted grants) for
their support of this work. We are grateful to Prof. K.
B. Sharpless for helpful suggestions on the AD chem-
istry, to Drs. C. Dufresne and I. Davies (Merck, Rahway)
for spectra and samples of the natural product, and to
Prof. E. Carreira for supplying copies of spectra. We
thank the EPSRC Mass Spectrometry Centre for ac-
curate mass determinations. We wish to acknowledge
the use of the EPSRC’s Chemical Database Service at
Daresbury.
Su p p or tin g In for m a tion Ava ila ble: Preparation of and
data for 2, 4, and 7-9; typical procedure for monophasic
dihydroxylation of 6-9; preparation of and data for Mosher’s
ester of 11, 13-21, 29, and conversion of 38 to 1; 13C NMR
spectrum of 7; 1H NMR spectra of 9, 11; 1H and 19F NMR
1
spectra of Mosher’s ester of 11; H NMR spectra of 14, 16-
21, 24, 26, 27, 28-31, 33, 34, and 36-38; intermediates in
the conversion of 38 to 1. This material is available free of
charge via the Internet at http://pubs.acs.org.
(50% EtOAc/hexanes); [R]23 +9.6 (c 0.27, CHCl3); δH (270
D
MHz, CDCl3) 7.30-7.11 (m, 5H), 5.02 (br s, 1H), 4.91 (br s,
1H), 4.89-4.86 (m, 1H), 3.91 (br s, 1H), 2.75 (dd, 1H, J 13.5,
4.6), 2.35 (dd, 1H, J 13.5, 9.9), 2.07 (s, 1H), 2.05 (s, 3H), 1.99-
J O000700M