Oxocarbenium ion cyclizations for C-branched cyclitols: Synthesis of a relay intermediate for fumagillin analogues

Article abstract of DOI:10.1021/jo050888f

A highly stereoselective oxocarbenium ion-alkene cyclization for synthesis of C-branched cylitols is described. This methodology was applied to 10S, a potentially versatile intermediate for side-chain analogues of the antiangiogenic agent fumagillin. Comp

Full text of DOI:10.1021/jo050888f

Oxocarbenium Ion Cyclizations for C-Branched Cyclitols:
Synthesis of a Relay Intermediate for Fumagillin Analogues
Fatoumata Camara, Johana Angarita, and David R. Mootoo*
Department of Chemistry, Hunter College, 695 Park Avenue, New York, New York 10021
dmootoo@hunter.cuny.edu
Received May 3, 2005
A highly stereoselective oxocarbenium ion-alkene cyclization for synthesis of C-branched cylitols
is described. This methodology was applied to 10S, a potentially versatile intermediate for side-
chain analogues of the antiangiogenic agent fumagillin. Compound 10S was subsequently converted
to diene 5. Because racemic 5 has been converted to racemic fumagillin, this synthesis of 5
constitutes a formal synthesis of the natural product.
Introduction
The natural product fumagillin 11 has attracted re-
newed interest in light of the discovery that it selectively
inhibits angiogenesis in tumor cells2 (Figure 1). This
activity has been related to disruption of endothelial cell
proliferation,3,4 which is believed to result from covalent
binding of 1 to, and subsequent inhibition of, methionine
aminopeptidase type 2 (MetAP-2).5 However, the link
between MetAP-2 inhibition and disruption of endothelial
cell growth has recently been questioned.6 Structure
activity studies have been generally confined to ana-
logues with different ester residues and modified C4
appendages, because of the accessibility of such struc-
tures from the natural product, and the finding that these
regions contain important contact points to MetAP-2.2,7-12
The most prominent derivative is TNP-470 2, which
FIGURE 1. Fumagillin and previously reported analogues.
contains an unnatural ester subunit.2,7 The development
of TNP-470 has, however, been stymied by poor phar-
macokinetic behavior and dose-limiting toxicity.13,14 More
recent studies suggest that the side-chain epoxide may
(1) (a) Elbe, T. E.; Hanson, F. R. Antibiot. Chemother. 1951, 1, 54.
(b) McCorkindale, N. J.; Sime, J. G. Proc. Chem. Soc. 1961, 331.
(2) Ingber, D.; Fujita, T.; Kishimoto, S.; Sudo, K.; Kanamaru, T.;
Brem, H.; Folkman, J. Nature 1990, 348, 555-557. Folkman, J. New
Engl. J. Med. 1995, 333, 1757-1763.
(3) Sin, N.; Meng, L.; Wang, M. Q. W.; Wen, J. J.; Bornmann, W.
G.; Crews, C. M. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 6099-6103.
(4) Griffith, E. C.; Su, Z.; Turk, B. E.; Chen, S.; Chang, Y.-H.; Wu,
Z.; Biemann, K.; Liu, J. O. Chem. Biol. 1997, 4, 461-471.
(5) Liu, S.; Widom, J.; Kemp, C. W.; Crews, C. M.; Clardy, J. Science
1998, 282, 1324-1327.
(6) Kim, S.; LaMontagne, K.; Sabio, M.; Sharma, S.; Versace, R. W.;
Yusuff, N.; Phillips, P. E. Cancer Res. 2004, 64, 2984-2987.
(7) Marui, S.; Itoh, F.; Kozai, Y.; Sudo, K.; Kishimoto, S. Chem.
Pharm. Bull. 1992, 40, 96-101.
(8) Han, C. K.; Ahn, H. S.; Choi, N. S.; Hong, R. K.; Moon, S. K.;
Chun, H. S.; Lee, S. J.; Kim, J. W.; Hong, C. I.; Kim, D.; Yoon, J. H.;
No, K. T. Bioorg. Med. Chem. Lett. 2000, 10, 39-43.
(9) Fardis, M.; Pyun, H.-J.; Tario, J.; Jin, H.; Kim, C. U.; Ruckman,
J.; Lin, Y.; Green, L.; Hicke, B. Bioorg. Med. Chem. 2003, 11, 5051-
5058.
(10) Pyun, H.-J.; Fardis, M.; Tario, J.; Yang, C. Y.; Ruckman, J.;
Jin, H.; Kim, C. U.; Ruckman, J.; Henninger, D.; Jin, H.; Kim, C. U.
Bioorg. Med. Chem. Lett. 2004, 14, 91-94.
(11) Rodeschini, V.; Boiteau, J.-G.; Van de Weghe, P.; Tarnus, C.;
Eustache, J. J. Org. Chem. 2004, 69, 357-373.
(12) Baldwin, J. E.; Bulger, P. G.; Marquez, R. Tetrahedron 2002,
58, 5441-5452.
(13) Kruger, E. A.; Figg, W. D. Expert Opin. Invest. Drugs 2000, 9,
1383-1396.
10.1021/jo050888f CCC: $30.25 © 2005 American Chemical Society
Published on Web 07/28/2005
6870
J. Org. Chem. 2005, 70, 6870-6875

Oxocarbenium Ion Cyclizations for C-Branched Cyclitols
SCHEME 1. Retrosynthesis of Fumagillin
Analogues
SCHEME 2. C-Branched Cyclitols via
Oxocarbenium Ion Cyclizations
SCHEME 3. Synthesis of 8R and 8Sa
not be critical for activity and that this region could be
extensively modified (cf., 3 and 49), without significant
decrease in binding to MetAP-2, as compared to TNP-
470.
The structural complexity and interesting biological
activity of fumagillin has resulted in considerable interest
in total synthesis. Such investigations are particularly
important because of the limited pool of structures that
is available from natural15 or semisynthetic sources.7-10,16
Following the first total synthesis by Corey and co-
workers,17 a number of total or formal syntheses of
racemic and enantiopure materials have been re-
ported.18-24 We envisaged that relatively simple C-
branched cyclitols 7R or 7S would be useful synthetic
intermediates for different side-chain analogues (Scheme
1). Methyl ketones 7R and 7S could be converted to a
C1′ epoxide (cf., 1, 2) and other analogues (cf., 3, 4) via
the known intermediates 518 and 6,10 previously prepared
from cyclohexadiene and the natural product 1, respec-
tively. Herein, we describe the synthesis of 7S, and its
conversion to optically active 5.
a Reagents and conditions: (a) PCC, CH2Cl2; (b) Ph3PdCH-
CO2Me, CH3CN, 65 °C, two steps, 88%; (c) AD-mix-â, t-BuOH-
H2O, MeSONH2, 90%; (d) (MeO)2CMe2, CSA, CH2Cl2; (e) THF-
aq KOH, two steps, 88%; (f) PhI(OAc)2, I2, CH2Cl2; (g) PhSH,
BF3‚OEt2, CH2Cl2, two steps, 80%; (h) DIBALH, CH2Cl2, -78 °C,
94%; (i) Swern’s Ox., 94%; (j) 2-methy-1-bromopropene, t-BuLi,
ether, -78 °C; (k) 2-methy-1-bromopropene, t-BuLi, ether, -78 °C,
ZnCl2, lithium (1S,2R)-N-methylephedrate, 0 °C, 83%; (l) TBDP-
SCl, imidazole, DMF, 50 °C.
Our synthetic plan relates 7R/S to the alkene precur-
sors 10R/S and follows from earlier studies on related
oxocarbenium ion-alkene cyclizations (Scheme 2). Thus,
treatment of the 1-phenylthio-1,2-O-isopropylidene ac-
(14) Griffith, E. C.; Su, Z.; Turk, B. E.; Chen, S.; Chang, Y.-H.; Wu,
Z.; Biemann, K.; Liu, J. O. Chem. Biol. 1997, 4, 461-471.
(15) For related natural products: Chu, M.; Mierzwa, R.; He, L.;
Xu, L.; Patel, M.; Patel, D.; Chan, T.-M. J. Antibiot. 2001, 54, 1096-
1099. Halasz, J.; Podanyi, B.; Vasvari-Debreczy, L.; Szabo, A.; Hajdu,
F.; Bocskei, Z.; Hegedus-Vajda, J.; Gyorbiro, A.; Hermecz, I. Tetrahe-
dron 2000, 56, 10081-10085. Kim, D.; Min, J.; Ahn, S. K.; Lee, H. W.;
Choi, N. S.; Moon, S. K. Chem. Pharm. Bull. 2004, 52, 447-450.
(16) Zhou, G.; Tsai, C. W.; Liu, J. O. J. Med. Chem. 2003, 46, 3452-
3454.
etal-alkene 11 with methyl triflate led to a single di-
astereomer 13, in which the isopropenyl appendage was
anti to the syn isopropylidene residue.25 This reaction
presumably proceeds through cyclization of the oxocar-
benium ion-alkene intermediate 12. Extension of this
concept to 10R/S calls for cyclization precursors 8R/S.
However, it was not clear how the configuration at the
allylic alcohol position would affect the stereochemical
outcome of the oxocarbenium ion cyclization. The cycliza-
tions of epimeric substrates 8R and 8S were therefore
first investigated.
The synthesis of 8R and 8S started from 4-pivaloyloxy-
1-butanol 1426 (Scheme 3). Alcohol 14 was first converted
to the isopropylidene-carboxylic acid 15. Thus, a straight-
forward alcohol oxidation-aldehyde olefination sequence
on 14 provided methyl (E)-1-pivaloyloxy-2-hexenoate,
(17) Corey, E. J.; Snider, B. B. J. Am. Chem. Soc. 1972, 94, 2549-
2550.
(18) Hutchings, M.; Moffat, D.; Simpkins, N. S. Synlett. 2001, 5,
661-663.
(19) Boiteau, J.-G.; Van de Weghe, P.; Eustache, J. Org. Lett. 2001,
3, 2737-2740.
(20) Vosburg, D. A.; Weiler, S.; Sorensen, E. J. Angew. Chem., Int.
Ed. 1999, 38, 971-974. Vosburg, D. A.; Weiler, S.; Sorensen, E. J.
Chirality 2003, 15, 156-166.
(21) Picoul, W.; Urchegui, R.; Haudrechy, A.; Langlois, Y. Tetrahe-
dron Lett. 1999, 40, 4797-4800.
(22) Taber, D. F.; Christos, T. E.; Rheingold, A. L.; Guzei, I. A. J.
Am. Chem. Soc. 1998, 121, 5589-5590.
(23) Kim, D.; Ahn, S. K.; Bae, H.; Choi, W. J.; Kim, H. S. Tetrahedron
Lett. 1997, 38, 4437-4440.
(25) Khan, N.; Cheng, X.; Mootoo, D. R. J. Am. Chem. Soc. 1999,
121, 4918-4919.
(26) Wu, Y.; Ahlberg, P. Acta Chem. Scand. 1995, 49, 364-374.
(24) Bedel, O.; Haudrechy, A.; Langlois, Y. Eur. J. Org. Chem. 2004,
18, 3813-3819.
J. Org. Chem, Vol. 70, No. 17, 2005 6871

Camara et al.
SCHEME 4. Cyclization of 8R and 8Sa
which was treated under standard Sharpless asymmetric
dihydroxylation conditions with AD-mix-â. Mosher ester
analysis indicated an ee of greater than 95% for the
resulting diol, the absolute stereochemistry of which was
based on the results for the dihydroxylation of closely
related E-disubstituted alkenes.27 Acetonation of the diol,
followed by selective hydrolysis of the methyl ester,
furnished 15, in 69% yield, over five steps from 14. The
synthesis of the 1-phenylthio-1,2-O-isopropylidene acetal
followed a variation of the Suarez radical fragmentation
of 1,2-O-isopropylidene sugars.28 Thus, exposure of 15 to
iodosobenzene diacetate under anhydrous conditions
provided an unseparated mixture of 1-acetoxy-1,2-O-
isopropylidene acetals 16 and 17. Boron trifluoride-
catalyzed acetal exchange on this mixture with thio-
phenol at low temperature led to a 9/1 mixture of anti/
syn acetal isomers 18 and 19, in 80% overall yield from
15. The stereochemistry of 18 and 19 was tentatively
assigned by examination of the proton chemical shifts of
a Reagents and conditions: (a) MeOTf, 2,6-di-tert-butyl-4-
methylpyridine, molecular sieves, CH2Cl2; 10R (45%) and 10S
(44%) from 8R/S (1/1); 10R (20%) and 10S (74%) from 8R/S (1/4).
1
the two methyl groups of the acetonide. Following H
NMR trends for anti/syn pairs of 1,2-diol acetonides, the
isomer in which the two methyl groups appeared as a
6H singlet (δ 1.46) was assigned as anti (i.e., 18), and
the other which showed resolved 3H singlets (δ 1.38 and
1.63 ppm) was assigned as syn (i.e., 19)29 The subsequent
reactions were carried out on the major anti isomer, 18.
DIBALH reduction of the pivalate in 18 and oxidation of
the resulting alcohol gave aldehyde 20. Treatment of 20
with 2-propenyllithium provided a 1/1 inseparable mix-
ture of allylic alcohols, which was converted to the
corresponding silyl ether derivatives 8R and 8S, in 61%
yield from 20. Attempted separation of mixture 8R/8S
was also not successful. The configuration at the newly
formed carbinol carbon was assigned in later derivatives
(vide infra). As the results of the subsequent cyclization
studies unfolded, it became necessary to develop a
stereoselective synthesis of 8S. Accordingly, 20 was
added to a mixture of 2-propenyllithium, zinc bromide,
and lithium (1S,1R)-N-methylephedrate, following the
Oppolzer procedure for the addition of chiral alkenyl zinc
reagents to aldehydes (Scheme 3).30 Silylation of the
reaction product gave an approximate 1/4 ratio of 8R/8S
in 83% overall yield from 20.
The cyclization of 8R and 8S was next investigated
(Scheme 4). Treatment of the 1/1 mixture of 8R/S with
methyl triflate and 4-methyl-2,6-di-tert-butylpyridine in
anhydrous dichloromethane led to a 1/1 ratio of two
chromatographically separable, cyclization products 10R
and 10S, in a combined yield of 89%. Under similar
conditions, the 1/4 mixture of 8R/S afforded 10R and 10S
in respective yields of 20% and 74%. The absence of any
other diastereomeric cycloadducts suggests that the
cyclization process was essentially, completely stereo-
selective; that is, 8R and 8S produced 10R and 10S,
respectively. The stereochemistry of 10R and 10S was
assigned by analysis of J values for vicinal protons on
the cyclohexane ring. For 10R, J3,4, J4,5, and J5,6 values
FIGURE 2. Chairlike transition states for oxocarbenium ion
cyclizations.
of 9.9, 3.5, and 5.3 are in agreement with a trans diaxial
arrangement between H3 and H4, and axial-equatorial
relationships between H4 and H5, and H5 and H6. For
10S, the corresponding J values were 10.0, 10.0, and 4.8,
which are consistent with a mutually trans-diaxial type
arrangement for H3, H4, and H5, and a cis axial-
equatorial relationship between H5 and H6.
In the absence of additional data, the stereochemical
results appear to be controlled by conformational effects
in the cyclization of the intermediate oxocarbenium ions
9R/S (Scheme 2). For both 8R and 8S, transition states
leading to a cis fused isopropylidene are expected to be
favored over ones leading to the more strained trans
fused system. Therefore, four chairlike31 transition states
R1-R4 and S1-S4 may be considered for 8R or 8S,
respectively, corresponding to “flip-chair” conformations
and an R- or â-orientation of the C4 substituent (Figure
2). In the case of 8S, S1 in which the eventual C3, C4,
and C5 substituents on the cyclohexane are mutually
trans and all pseudoequatorial appears to be a reasonable
pathway to 10S. For 8R, the inverted chairlike confor-
(27) Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless, K. B. Chem.
Rev. 1994, 94, 2483-2547.
(28) De Armas, P.; Francisco, C. G.; Suarez, E. Angew. Chem., Int.
Ed. Engl. 1992, 31, 772-774.
(29) Rieser, J. M.; Fang, X.-P.; Anderson, J. E.; Miesbauer, L. R.;
Smith, D. L.; McLaughlin, J. L. Helv. Chim. Acta 1993, 6, 2433-2444.
(30) Oppolzer, W.; Radinov, R. N. Tetrahedron Lett. 1991, 32, 5777-
5780.
(31) More distorted chair and boat geometries are conceivable. For
simplicity, chairlike structures are illustrated.
6872 J. Org. Chem., Vol. 70, No. 17, 2005

Oxocarbenium Ion Cyclizations for C-Branched Cyclitols
SCHEME 5. Transformation of 10S to Fumagillin
isomer in the starting mixture of 23E/Z was observed.
This result is presumably due to chromatographic re-
moval of the epoxide derivative from 23E prior to the
reaction with Ph2PLi. The 1H and 13C NMR data for 23E
were essentially identical to that for previously prepared
Relay Compound 5a
5
18 (Supporting Information). Because racemic 5 has been
converted to racemic fumagillin, the synthesis described
here constitutes a formal synthesis of the natural prod-
uct.
Conclusion
In summary, the synthesis of 10S, a potentially
versatile precursor for side-chain analogues of fumagillin,
has been developed. The utility of 10S was illustrated
by its conversion to the known fumagillin relay compound
5. On a more general note, the oxocarbenium ion cycliza-
tion strategy appears suitable for the preparation of
unusual, C-branched cyclitols, such as required as probes
in inositol-related biochemical mechanisms.34 Investiga-
tions along these lines are underway.
a Reagents and conditions: (a) O3, CH2Cl2-MeOH, -78 °C, then
Ph3P, 85%; (b) 21, LiHMDS, THF, -78 °C, 87%; (c) n-Bu4NF, THF,
quant.; (d) VO(acac)2, TBHP, CHCl2; (e) Ph2PLi, THF, then MeI,
two steps, 67%.
Experimental Section
mation R1 in which the C3, C4, and C5 substituents
adopt pseudoequatorial, pseudoequatorial, and pseudo-
axial orientations, respectively, are in agreement with
the observed result. Thus, the configurations at C3 and
C6 in the precursor apparently dictate the stereochem-
istry at the newly formed centers at C4 (trans to C3) and
C5 (cis to C6), respectively.
(1R,2R)- and (1S,2R)-1-O-Acetyl-1,2-O-isopropylidene-
5-pivaloyloxy-pentane-hemiacetal (16 and 17). To a solu-
tion of carboxylic acid 15 (10.3 g, 35.8 mmol) in dry dichlo-
romethane (150 mL) were added diacetoxyiodobenzene (13.8
g, 42.9 mmol) and iodine (9.0 g, 35.8 mmol). The reaction
mixture was stirred under argon for 4 h and then poured into
10% aqueous Na2SO3. The mixture was extracted with ether,
and the organic phase was dried (Na2SO4), filtered, and
concentrated under reduced pressure. FCC of the residue
provided an unsepararated mixture of 16 and 17 as a colorless
oil (9.5 g, 88%): Rf ) 0.60 (20% EtOAc/petroleum ether). For
major isomer: 1H NMR (500 MHz) δ 1.21 (s, 9H), 1.49 (s, 6H),
1.62-1.82 (m, 4H), 2.11 (s, 3H), 4.07-4.11 (m, 2H), 4.18-4.22
(m, 1H), 5.99 (d, 1H, J ) 2.4 Hz); 13C NMR (125 MHz) δ 21.5,
24.7, 26.9, 27.4, 27.9, 29.6, 38.9, 63.9, 81.8, 99.2, 112.6, 170.7,
178.7. For minor isomer: 1H NMR (500 MHz) δ 1.21 (s, 9H),
1.40 (s, 3H), 1.52 (s, 3H), 1.62-1.82 (m, 4H), 2.11 (s, 3H), 4.07-
The extension of the side chain in 10S, the cycloadduct
required for fumagillin, was next investigated (Scheme
5). An olefin metathesis approach was initially explored.
However, neither a cross metathesis on 10S nor a ring-
closing strategy on unsaturated ester derivatives of the
alcohol derived from desilylation of 10S was successful.
Alkene 10S was therefore transformed to the methyl
ketone 7S, and elaboration of 7S was pursued. Unsta-
bilized or stabilized Wittig, or conventional Julia re-
agents, resulted in no reaction. The Kocienski modifica-
tion32 of the Julia reaction using benzothiazole sulfone
21 was more successful but unfortunately led to a
mixture of 22 E/Z (E/Z: ca. 1/6), with the undesired Z
isomer as the predominant product. All attempts to
improve this result by varying the base, solvent, and
reaction temperature led to no significant change in
stereoselectivity. The isomerization of 22E/Z was there-
fore investigated. Direct, thiol-mediated procedures un-
der photochemical or thermal conditions were unsuccess-
ful, leading to intractable product mixtures. An eventual
solution was found in the Vedejs’ two-step isomerization
protocol.33 Thus, VO(acac)2 promoted, regiospecific ep-
oxidation of the mixture of homoallylic alcohols 23E/Z
led to a mixture of epoxides, which was chromatographi-
cally separated. Reaction of the major product with Ph2-
PLi, followed by treatment of the product with methyl
iodide, led to 23E as a single E-isomer, in 67% overall
yield from the original mixture of homoallylic alcohols
23E/Z (E/Z: 1/6). None of the Z isomer that would have
resulted from stereospecific inversion of the minor E
4.11 (m, 2H), 4.12-4.17 (m, 1H), 6.23 (d, 1H, J ) 3.3 Hz); 13
NMR (125 MHz) δ 21.4, 25.2, 25.5, 25.9, 28.4, 38.8, 64.1, 80.0,
94.5, 111.4, 170.7, 178.7. For mixture: MS(ESI) m/z 320.2 [M
+ NH4].
C
(1S,2R)- and (1R,2R)-1-(S)-phenylthio-1,2-O-isopro-
pylidene-5-pivaloyloxy-pentane-(S)-phenyl-monothio-
hemiacetal (18 and 19). BF3‚OEt2 (3.69 mL, 29.1 mmol) was
slowly added at -78 °C, under argon, to a solution of mixture
16/17 (8.8 g, 29.1 mmol) and thiophenol (5.98 mL, 58.3 mmol),
in anhydrous CH2Cl2 (100 mL). The temperature was warmed
to -40 °C, and stirring was continued at this temperature for
1 h (or until the TLC indicated complete disappearance of the
starting material). The reaction mixture was quenched by
addition of triethylamine (5 mL), then poured into saturated
aqueous NaHCO3 (150 mL) and extracted with ether. The
organic extract was washed with brine (150 mL), dried (Na2-
SO4), filtered, and concentrated in vacuo. FCC of the residue
gave 18 (8.45 g, 82%) and 19 (0.95 g, 9%). For 18: Rf ) 0.54
(10% EtOAc/petroleum ether); [R]D +113 (c 0.7, CHCl3); 1H
NMR (500 MHz) δ 1.19 (s, 9H), 1.46 (s, 6H), 1.62-1.85 (m,
4H), 4.05 (m, 1H), 4.10 (t, 2H, J ) 5.9 Hz), 5.06 (d, 1H, J )
7.3 Hz), 7.26-7.52 (m, 5H); 13C NMR (125 MHz) δ 25.3, 26.0,
27.4, 27.8, 29.5, 39.0, 64.1, 80.3, 88.6, 111.1, 127.6, 129.2, 131.7,
134.2, 178.2. HRMS(ESI) calcd for C19H29O4S (M + H) 353.1787,
found 353.1788. For 19: Rf ) 0.43 (10% EtOAc/petroleum
(32) Bellingham, R.; Jarowicki, K.; Kocienski, P.; Martin, V. Syn-
thesis 1996, 285-296.
(33) Vedejs, E. J. Org. Chem. 1973, 38, 1178-1183. Zhang, H.;
Mootoo, D. R. J. Org. Chem. 1995, 60, 8134-8135.
(34) Potter, B. V. L.; Lampe, D. Angew. Chem., Int. Ed. Engl. 1995,
34, 1933-1972. Rosenberg, H. J.; Riley, A. M.; Correa, V.; Taylor, C.
W.; Potter, B. V. L. Carbohydr. Res. 2000, 329, 7-16.
J. Org. Chem, Vol. 70, No. 17, 2005 6873

Camara et al.
ether); 1H NMR (300 MHz) δ 1.21 (s, 9H), 1.38 (s, 3H), 1.63 (s,
3H), 1.75-1.95 (m, 4H), 4.05-4.20 (m, 2H), 4.30 (m, 1H), 5.58
(d, 1H, J ) 4.8 Hz), 7.24-7.53 (m, 5H); 13C NMR (75 MHz) δ
25.8, 26.5, 27.4, 28.3, 39.0, 64.1, 79.1, 88.8, 110.5, 127.3, 129.2,
132.9, 135.0, 178.4. MS(ESI) m/z 370.2 [M + NH4].
°C for 4 h. The reaction mixture was then diluted with water
and extracted with ether. The combined organic phase was
washed with brine, dried (Na2SO4), filtered, and evaporated
under reduced pressure. The residue was purified by FCC to
give an inseparable mixture of 8R/S (3.00 g, ca. 100% from
alcohol mixture) as a yellow oil. The isomer ratio was esti-
mated at 1/4 based on the ratio of the respective signals for
isomeric carbinol (81.0 vs 80.9 ppm) and isomeric acetal (88.7
vs 88.9 ppm) carbons, and the stereochemistry was assigned
by correlation with the products of the subsequent cyclization
reaction (vide infra). Rf ) 0.80 (10% EtOAc/petroleum ether);
1H NMR (300 MHz) δ 1.03 (s, 9H), 1.12 (s, 3H), 1.43 (s, 6H),
1.53 (s, 3H), 1.57-1.73 (m, 4H), 3.93 (m, 1H), 4.36 (m, 1H),
4.99 (t, 1H, J ) 7.3 Hz), 5.13 (br d, 1H, J ) 8.1 Hz), 7.25-7.70
(m, 15H); 13C NMR (75 MHz) δ 18.2, 19.5, 25.8, 26.0, 27.2,
27.8, 28.3, 28.6, 34.5 (minor), 34.6 (major), 70.3 (major), 70.4
(minor), 80.9 (major), 81.0 (minor), 88.7 (minor), 88.9 (major),
110.8, 127.4, 127.6, 128.2, 129.1, 129.5, 129.6, 131.6, 132.7,
132.8, 134.7, 134.9, 136.1, 136.2. HRMS(FAB) calcd for C34H43O3-
Si (M - H) 559.2702, found 559.2700.
Cyclitols (10R and 10S). A portion of the mixture of 8R/S
from the previous step (720 mg, 1.28 mmol), 2,6-di-tert-butyl-
4-methylpyridine (2.37 g, 11.6 mmol), and freshly activated,
powdered 4A molecular sieves (2.0 g) in anhydrous dichlo-
romethane (15 mL) were stirred for 15 min, at room temper-
ature, under an argon atmosphere, and then cooled to 0 °C.
Methyl triflate (1.00 mL, 8.99 mmol) was introduced, and the
reaction was warmed to room temperature and stirred for an
additional 18 h, at which time Et3N (1.6 mL) was added. The
mixture was diluted with ether, washed with saturated
aqueous NaHCO3 and brine, dried (Na2SO4), filtered, and
evaporated under reduced pressure. FCC afforded 10R (113
mg, 20%) and 10S (433 mg, 74%).
Aldehyde (20). A 1 M solution of DIBALH in heptane (45.9
mL, 45.9 mmol) was added dropwise over 10 min to a solution
of 18 (7.70 g, 21.9 mmol) in dry dichloromethane (80 mL), at
-78 °C, under an atmosphere of nitrogen. The reaction was
allowed to warm to room temperature, stirred at this temper-
ature for 1 h, and then poured into saturated aqueous NH4Cl
and extracted with ether. The organic phase was washed with
saturated aqueous NaHCO3, dried (Na2SO4), filtered, and
concentrated in vacuo. The residue was purified by FCC to
afford the derived primary alcohol (5.53 g, 94%) as a colorless
oil: Rf ) 0.24 (20% EtOAc/petroleum ether); 1H NMR (300
MHz) δ 1.46 (s, 6H), 1.65-1.95 (m, 4H), 3.68 (t, 2H, J ) 5.9
Hz), 4.00 (m, 1H), 5.07 (d, 1H, J ) 7.3 Hz), 7.26-7.52 (m, 5H);
13C NMR (75 MHz) δ 26.7, 28.3, 29.8, 30.2, 63.2, 81.3, 89.2,
111.4, 128.0, 129.5, 132.4, 135.0. HRMS(ESI) calcd for C14H21O3S
(M + H) 269.1211, found 269.1210.
DMSO (6.80 mL, 97.0 mmol) was added dropwise to a
solution of oxalyl chloride (4.20 mL, 48.8 mmol) in CH2Cl2 (60
mL), at -78 °C under an argon atmosphere. After the mixture
was stirred at this temperature for 20 min, a solution of the
alcohol from the previous step (5.20 g, 19.5 mmol) in CH2Cl2
was slowly introduced. The reaction was stirred for 25 min at
-78 °C, at which time Et3N (20.8 mL) was added. The solution
was warmed to room temperature, stirred for an additional
10 min at this temperature, and then diluted with ether. The
mixture was washed with saturated aqueous NaHCO3, and
the organic phase was dried (Na2SO4) and concentrated in
vacuo. FCC of the residue afforded aldehyde 20 (4.90 g, 94%
from alcohol) as a yellow oil: Rf ) 0.60 (20% EtOAc/petroleum
For 10R: Rf ) 0.68 (10% EtOAc/petroleum ether); 1H NMR
(500 MHz) δ 1.06 (s, 9H), 1.28 (s, 3H), 1.38 (s, 3H), 1.82 (s,
3H), 2.48 (dd 1H, J ) 3.5, 9.9 Hz), 4.06-4.14 (m, 2H), 4.21
(dd, 1H, J ) 3.5, 5.3 Hz), 4.92 (s, 1H), 4.96 (s, 1H), 7.40-7.85
(m, 10H); 13C NMR δ (125 MHz) δ 19.5, 21.9, 25.7, 26.9, 27.2,
28.1, 30.7, 53.4, 69.4, 74.4, 78.6, 108.2, 113.9, 127.5, 127.6,
129.6, 129.8, 134.1, 135.4, 136.1, 145.3. HRMS(ESI) calcd for
C28H39O3Si (M + H) 451.2668, found 451.2690.
1
ether); H NMR (300 MHz) δ 1.44 (s, 6H), 1.94 (m, 1H), 2.15
(m, 1H), 2.64 (t, 2H, J ) 7.0 Hz), 4.02 (m, 1H), 5.06 (d, 1H, J
) 7.3 Hz), 7.26-7.52 (m, 5H), 9.80 (s, 1H); 13C NMR (125 MHz)
δ 25.2, 26.0, 27.6, 40.1, 79.6, 88.4, 111.2, 127.7, 129.2, 131.9,
134.1, 201.4. HRMS(ESI) calcd for C14H19O3S (M + H) 267.1055,
found 267.1067.
Thioacetal Alkenes (8R/S). tert-Butyllithium (1.7 M in
hexane, 15.5 mL, 26.3 mmol) was added dropwise to a solution
of 2-methyl-1-bromopropene (1.60 mL, 15.8 mmol) in dry ether,
at -78 °C, under an argon atmosphere. The mixture was
warmed to -35 °C and stirred at this temperature for 2 h. A
solution of ZnBr2 (0.6 M in ether, 15.8 mmol) was then added,
and the mixture was warmed to 0 °C and maintained at this
temperature for an additional 1 h. A solution of lithium
(1S,2R)-N-methylephedrate was prepared by addition of n-
BuLi (1.6 M in hexanes, 6.50 mL, 15.8 mmol) to (-)-N-
methylephedrine (2.80 g, 15.8 mmol) in toluene (60 mL) at 0
°C, in a separate reaction vessel, and then slowly added to
the reaction mixture. The resulting solution was stirred for 1
h at 0 °C, at which time a solution of aldehyde 20 (700 mg,
2.60 mmol) in dry ether (10 mL) was added dropwise. After
an additional 1 h at 0 °C, the reaction was quenched by
addition of saturated aqueous NH4Cl, the organic phase was
separated, and the aqueous layer was extracted with ether.
The organic extract was washed with a second portion of
aqueous NH4Cl, dried (Na2SO4), and concentrated under
reduced pressure. FCC of the residue gave a mixture of allylic
alcohol products (700 mg, 83%) as an inseparable mixture. For
mixture: Rf ) 0.27 (20% EtOAc/petroleum ether); 1H NMR
(300 MHz, C6D6) δ 1.40 (s, 3H), 1.44 (s, 6H), 1.48 (1.48, 3H),
1.70-2.0 (m, 4H), 4.15-4.22 (m, 1H), 4.23-4.32 (m, 1H), 5.16
(d, 1H, J ) 8.4 Hz), 5.23 (d, 1H, J ) 7.0 Hz), 5.24 (d, 1H, J )
7.0 Hz); 13C NMR (75 MHz, C6D6) δ 19.0, 26.4, 26.7, 28.3, 29.5,
34.5, 69.0, 81.5 (major), 81.6 (minor), 89.3 (major), 89.2 (minor),
111.3, 129.5, 135.0. MS(ESI) m/z 345.1 [M + Na].
For 10S: Rf ) 0.60 (10% EtOAc/petroleum ether); [R]D +14.0
1
(c 1.0, CHCl3); H NMR (300 MHz,) δ 1.01 (s, 9H), 1.24-1.65
(m, 3H), 1.31 (s, 3H), 1.57 (s, 3H), 1.67 (s, 3H), 1.98 (m, 1H),
2.40 (t, 1H, J ) 10.0 Hz), 3.39 (dt, 1H, J ) 3.8, 10.0 Hz), 3.82
(dd, J ) 4.8, 10.0 Hz, 1H), 4.05 (m, 1H), 4.88 (s, 1H), 4.99 (s,
1H), 7.35-7.80 (m, 10H); 13C NMR (125 MHz) δ 19.5, 20.3,
23.9, 26.5, 27.1, 28.8, 30.0, 57.2, 71.6, 72.8, 77.6, 108.7, 114.9,
127.5, 127.7, 129.6, 129.9, 133.8, 135.2, 136.2, 143.7; HRMS-
(FAB) calcd for C28H37O3Si (M - H) 449.2512, found 449.2511.
Methyl Ketone (7S). Alkene 10S (350 mg, 0.77 mmol) was
dissolved in a 5/1 mixture of CH2Cl2/MeOH (19 mL). The
solution was cooled to -78 °C and treated with a stream of O3
in O2 until TLC indicated the complete disappearance of the
starting material. The reaction was then purged with N2, and
triphenylphosphine (700 mg) was added. The mixture was
warmed to room temperature, stirred for 1 h at this temper-
ature, and concentrated under reduced pressure. FCC of the
residue afforded 7S (300 mg, 85%) as a colorless oil: Rf ) 0.32
(10% EtOAc/petroleum ether); [R]D ) -39.0 (c 0.35, CHCl3);
1H NMR (500 MHz) δ 0.99 (s, 9H), 1.25-1.40 (m, 2H), 1.31 (s,
3H), 1.60 (m, 1H), 1.62 (s, 3H), 1.95 (m, 1H), 2.36 (s, 3H), 3.02
(t, 1H, J ) 9.9 Hz), 3.74 (dt, 1H, J ) 3.30, 9.9 Hz), 3.99-4.08
(m, 2H), 7.30-7.80 (m, 10H); 13C NMR (125 MHz) δ 19.5, 23.7,
26.5, 27.0, 28.8, 29.1, 29.9, 34.7, 61.9, 72.5 (two peaks), 77.6,
109.1, 127.6, 127.9, 129.8, 130.1, 132.8, 134.7, 136.0, 136.1,
211.8. HRMS(ESI) calcd for C27H37O4Si [M + H] 453.2461,
found 453.2480.
Benzothiazole Sulfone (21).35 A mixture of mercaptoben-
zothiazole (500 mg, 2.98 mmol), NaH (299 mg of a 60%
A sample of the alcohol mixture from the previous step (1.70
g, 5.28 mmol), TBDPSCl (1.78 mL, 6.86 mmol), imidazole (719
mg, 10.6 mmol), in anhydrous DMF (15 mL) was stirred at 50
(35) For a similar procedure: Lafontaine, J. A.; Provencal, D. P.;
Gardelli, C.; Leahy, J. W. J. Org. Chem. 2003, 68, 4215-4234.
6874 J. Org. Chem., Vol. 70, No. 17, 2005

Oxocarbenium Ion Cyclizations for C-Branched Cyclitols
dispersion in mineral oil, 7.47 mmol), and Bu4NI (110 mg, 0.29
mmol) in dry DMF (5 mL) was stirred at room temperature
for 30 min, under argon. 5-Bromo-2-methyl-2-pentene (0.80
mL, 5.98 mmol) was then added to the reaction mixture, and
stirring was continued for 1.5 h. The reaction was quenched
with water, and the mixture was extracted with ether. The
organic phase was washed with brine, dried (Na2SO4), filtered,
and concentrated in vacuo. The residue was purified by FCC
to give the derived thioether as a colorless oil (720 mg, 97%):
EtOAc/petroleum ether). The mixture was determined to be
an E/Z ratio of 1/6 based on integration of 1H NMR signals at
δ 5.40 and 5.62 ppm, respectively. For 23Z: 1H NMR (300
MHz) δ 1.26 (s, 1H), 1.36 (s, 3H), 1.56 (s, 3H), 1.64 (s, 3H),
1.69 (s, 3H), 1.74-1.79 (m, 2H), 1.86-1.92 (m, 1H), 2.23-2.27
(m, 1H), 2.71 (t, 1H, J ) 10.0 Hz), 2.73-2.92 (m, 2H), 3.42 (t,
1H, J ) 9.0 Hz), 4.09 (dd, 1H, J ) 5.5, 9.5 Hz), 4.26-4.28 (br
s, 1H), 5.13 (t, 1H, J ) 7.3 Hz), 5.62 (t, 1H, J ) 7.0 Hz); 13C
NMR (125 MHz) δ 18.0, 19.1, 24.1, 25.8, 26.4, 27.1, 28.3, 28.6,
50.8, 68.3, 73.0, 76.0, 108.8, 123.0, 130.6, 132.3, 132.5. For 23E,
see below.
E Homoallylic Alcohol 23E. tert-Butyl hydroperoxide in
decane (0.08 mL of a 5-6 M solution, ca. 0.5 mmol) was added
to a solution of 23 E/Z (80 mg, 0.28 mmol) and vanadyl
acetylacetonate (8 mg, 0.03 mmol), in dry CH2Cl2 (27 mL), at
-10 °C under argon. The mixture was stirred at this temper-
ature for 1 h, then quenched with dimethyl sulfide (0.2 mL),
and stirred at 0 °C for an additional 30 min. The volatiles were
then evaporated in vacuo, and the residue was purified by FCC
to yield a single epoxide derivative (75 mg, 88%) as a colorless
oil: Rf ) 0.35 (20% EtOAc/petroleum ether). 1H NMR (500
MHz) δ 1.33 (s, 3H), 1.38 (s, 3H), 1.47 (s, 3H), 1.55-1.80 (m,
4H), 1.62 (s, 3H), 1.71 (s, 3H), 2.09-2.17 (m, 2H), 2.47-2.52
(m, 1H), 2.74 (dd, 1H, J ) 4.03, 8.12 Hz), 3.49 (s, 1H), 3.77-
3.81 (m, 1H), 3.98 (dd, 1H, J ) 5.64, 9.37 Hz), 4.22-4.24 (m,
1H), 5.24-5.28 (m, 1H); 13C NMR (125 MHz) δ 18.2, 19.5, 23.8,
25.9, 26.1, 28.1, 28.3, 28.4, 49.1, 62.9, 64.4, 69.4, 73.1, 75.6,
109.2, 119.7, 134.6; MS(ESI) m/z 297.2 [M + H].
A 0.5 M stock solution of Ph2PLi was prepared by the
addition of a hexane solution of n-butyllithium (1.0 mL, 1.6
M) to a solution of Ph2PH (0.3 mL, 1.68 mmol) in dry THF (2
mL) at room temperature under an argon atmosphere, followed
by stirring for an additional 1 h. An aliquot of the red solution
of Ph2PLi (1.95 mL, 1.00 mmol) was added at room tempera-
ture to a THF solution (0.8 mL), of a portion of the epoxide
derivative (25 mg, 0.08 mmol), from the previous step. The
reaction mixture was stirred for 2 h, at which time freshly
distilled MeI (0.06 mL, 1.00 mmol) was introduced, and
stirring was continued for an additional 1 h. The mixture was
then cooled to -78 °C, and n-butyllithium (1.0 mL, 1.6 M) was
added until the red-yellow color persisted. The mixture was
then diluted with ether (10 mL) and filtered through Celite.
The filtrate was concentrated in vacuo, and the residue was
purified by FCC to afford 23E as a light-yellow oil (18 mg,
76% from epoxide): Rf ) 0.47 (20% EtOAc/petroleum ether).
[R]D -107.2 (c 0.7, CHCl3); 1H NMR (500 MHz) δ 1.37 (s, 3H),
1.52 (s, 3H), 1.54-1.62 (m, 1H), 1.64 (s, 3H), 1.70 (s, 6H), 1.72-
1.78 (m, 1H), 1.79 (br s, 1H), 1.86-1.90 (m, 1H), 2.14 (t, 1H,
J ) 10.1 Hz), 2.20-2.24 (m, 1H), 2.75-2.88 (m, 2H), 3.41 (dt,
1H, J ) 3.7, 10.6 Hz), 4.04 (dd, 1H, J ) 4.8, 9.5 Hz), 4.21-
4.24 (m, 1H), 5.13 (t, 1H, J ) 7.1 Hz), 5.40 (t, 1H, J ) 6.6 Hz);
13C NMR (125 MHz) δ 13.2, 18.0, 23.9, 25.9, 26.6, 27.4, 28.0,
28.7, 59.0, 68.1, 73.1, 76.8, 108.8, 122.8, 131.1, 131.3, 132.2;
MS(ESI) m/z 281.2 [M + H]. HRMS(ESI) calcd for C17H27O3
[M - H] 279.1960, found 279.1966.
1
Rf ) 0.63 (20% EtOAc/petroleum ether); H NMR (300 MHz)
δ 1.63 (s, 3H), 1.69 (s, 3H), 2.49 (q, 2H, J ) 7.3 Hz), 3.31 (t,
2H, J ) 7.3 Hz), 5.19 (t, 1H, J ) 7.3 Hz), 7.20-7.80 (m, 4H);
13C NMR (75 MHz) δ 18.3, 26.0, 28.4, 34.1, 121.1, 121.7, 121.8,
124.3, 126.1, 134.7, 135.4, 153.6, 167.3.
Ammonium molybdate (1.90 g) and hydrogen peroxide (7.72
mL) were combined at 0 °C and stirred for 15 min. The
resulting bright yellow solution was added dropwise at 0 °C
over 90 min, to a solution of the product from the previous
step (200 mg, 0.80 mmol) in EtOH (7.40 mL). The progress of
the reaction was carefully monitored via TLC. Upon disap-
pearance of the starting material, the mixture was diluted with
water (20 mL) and extracted with ether. The organic extract
was washed with brine, dried (Na2SO4), filtered, and concen-
trated in vacuo. The residue was purified by FCC to provide
sulfone 21 (187 mg, 83% from thioether) as a white solid, and
another product that was presumed to be the sulfoxide
derivative (17 mg, 7% from thioether), as a colorless oil: Rf
0.63 and 0.53 (20% EtOAc/petroleum ether), respectively. For
21: 1H NMR (300 MHz) δ 1.55 (s, 6H), 2.55 (q, 2H, J ) 8.0
Hz), 3.50 (t, 2H, J ) 7.7 Hz), 4.99 (t, 1H, J ) 8.8 Hz), 7.50-
8.20 (m, 4H); 13C NMR (125 MHz) δ 17.9, 21.6, 25.7, 54.7,
119.0, 122.5, 125.6, 127.8, 128.2, 135.9, 136.9, 152.9, 166.2.
MS(ESI) m/z 282 [M + H]. HRMS(ESI) calcd for C13H16NO2S2
[M + H] 282.0622, found 282.0627. For sulfoxide: 1H NMR
(300 MHz) δ 1.60 (s, 3H), 1.65 (s, 3H), 2.30-2.49 (m, 1H), 2.57
(m, 1H), 3.10-3.30 (m, 2H), 5.10 (t, 1H, J ) 7.5 Hz), 7.40-
7.59 (m, 2H), 7.90-8.09 (m, 2H); 13C NMR (75 MHz) δ 18.2,
20.9, 25.9, 60.0, 120.0, 122.4, 124.1, 126.2, 127.0, 135.6, 136.2,
154.1, 178.1.
Diene Mixture (22E/Z). LiHMDS (0.49 mL of a 1.0 M
solution in THF, 0.49 mmol) was added dropwise, under argon,
to a solution of sulfone 21 (140 mg, 0.49 mmol) and methyl
ketone 7S (50 mg, 0.11 mmol) in anhydrous THF (5 mL), at
-78 °C. The resulting light yellow solution was stirred for 1.5
h, then warmed to room temperature, and stirred for 10 min
at this temperature. The reaction was then quenched by the
addition of saturated aqueous NH4Cl, and the mixture was
extracted with ether. The organic extract was dried (Na2SO4),
filtered, and concentrated in vacuo. FCC of the residue afforded
an inseparable 1/6 mixture of 22E/Z (50 mg, 87%) as a colorless
oil: Rf ) 0.60 (10% EtOAc/petroleum ether). The E/Z ratio was
determined by integration of the 1H NMR signals at δ 5.16
and 5.43 ppm, respectively. For 22Z: 1H NMR (500 MHz) δ
1.01 (s, 9H), 1.31 (s, 3H), 1.49-1.74 (m, 3H), 1.56 (s, 3H), 1.60
(s, 3H), 1.66 (s, 3H), 1.70 (s, 3H), 1.96-2.02 (m, 1H), 2.82-
2.90 (m, 2H), 2.96-3.02 (m, 1H), 3.38 (dt, 1H, J ) 3.5, 10.0
Hz), 3.88 (dd, 1H, J ) 5.5, 9.5 Hz), 4.08 (m, 1H), 5.23 (t, 1H,
Acknowledgment. This investigation was sup-
ported by NIH grant GM 57865. “Research Centers in
Minority Institutions” award RR-03037 from the Na-
tional Center for Research Resources of the NIH, which
supports the infrastructure and instrumentation of the
Chemistry Department at Hunter College, and the
MBRS/RISE program are also acknowledged. We thank
Professor Simpkins for providing spectral data for (()
5.
J ) 7.0 Hz), 5.43 (t, 1H, J ) 7.0 Hz), 7.28-7.70 (m, 10H); 13
C
NMR δ (75 MHz) δ 18.1, 19.6, 24.4, 25.9, 26.5, 27.2, 27.5, 28.7,
30.3, 50.9, 71.2, 72.8, 77.5, 108.6, 124.1, 127.5, 127.6, 129.6,
129.7, 129.8, 131.2, 132.3, 134.1, 135.3, 136.1. For 22E
(selected signals): 1H NMR (500 MHz) δ 2.31 (t, 1H, J ) 10.5
Hz), 2.76-2.81 (m, 1H), 3.82 (dd, 1H, J ) 5.0, J ) 9.5 Hz),
4.02-4.06 (m, 1H), 5.16 (t, 1H, J ) 6.5 Hz), 5.23 (t, 1H, J )
7.0 Hz). HRMS(ESI) mixture calcd for C33H47O3Si [M + H]
519.3294, found 519.3317.
Supporting Information Available: Experimental pro-
Homoallylic Alcohol Mixture (23E/Z). A mixture of
22E/Z (120 mg, 0.23 mmol) in THF (1.5 mL) and Bu4NF (2.3
mL of a 1.0 M solution in THF, 0.23 mmol) was stirred at room
temperature for 18 h, and then concentrated in vacuo. The
residue was purified by FCC to give an inseparable mixture
of 23E/Z as a colorless oil (65 mg, quant.): Rf ) 0.25 (20%
1
cedures and physical data for 15. H and 13C NMR charts for
7S, 8R/S, 10S, 16/17, 18-21, 22Z/E, 23E/Z, 23E, and 23E-
acetate. This material is available free of charge via the
Internet at http://pubs.acs.org.
JO050888F
J. Org. Chem, Vol. 70, No. 17, 2005 6875