Synthesis and biological activity of C-6 and C-7 modified paclitaxels

Article abstract of DOI:10.1016/S0040-4020(00)00611-6

Structure modification of paclitaxel at the C-6 and C-7 positions has been achieved using the readily available intermediate 6α-hydroxy-7-epipaclitaxel (2). While a single diastereomer of the cyclic sulfite of 2 was prepared at lower temperature, both diastereomers were obtained at room temperature. The cyclic sulfate was found to be inert toward nucleophilic attack, which confirms the great steric hindrance of the β-face of the C-6 and C-7 region of paclitaxel. Derivatization at the 6β-position with a nitrogen-containing function was accomplished using the alternate substrate 6α-trifluoromethanesulfonate (9), with the 7-epi hydroxyl group protected to avoid the formation of C-ring rearranged product. Hydrogenation of the 6β-azide was difficult but could be achieved in moderate yield under high pressure. Similar to other C-6 and C-7 modified analogs reported earlier, these new compounds displayed comparable in vitro cytotoxicity to paclitaxel against the HCT116 human colon cancer cell line and the A2780 human breast cancer cell line. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4020(00)00611-6

TETRAHEDRON
Pergamon
Tetrahedron 56 (2000) 6407±6414
Synthesis and Biological Activity of C-6 and C-7
Modi®ed Paclitaxels
Haiqing Yuan,a,² Craig R. Fairchild,b Xian Lianga and David G. I. Kingstona,
*
aDepartment of Chemistry, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061-0212, USA
bBristol-Myers Squibb Pharmaceutical Research Institute, P.O. Box 4000, Princeton, NJ 05843, USA
Received 23 May 2000; accepted 5 July 2000
AbstractÐStructure modi®cation of paclitaxel at the C-6 and C-7 positions has been achieved using the readily available intermediate 6a-
hydroxy-7-epipaclitaxel (2). While a single diastereomer of the cyclic sul®te of 2 was prepared at lower temperature, both diastereomers
were obtained at room temperature. The cyclic sulfate was found to be inert toward nucleophilic attack, which con®rms the great steric
hindrance of the b-face of the C-6 and C-7 region of paclitaxel. Derivatization at the 6b-position with a nitrogen-containing function was
accomplished using the alternate substrate 6a-tri¯uoromethanesulfonate (9), with the 7-epi hydroxyl group protected to avoid the formation
of C-ring rearranged product. Hydrogenation of the 6b-azide was dif®cult but could be achieved in moderate yield under high pressure.
Similar to other C-6 and C-7 modi®ed analogs reported earlier, these new compounds displayed comparable in vitro cytotoxicity to paclitaxel
against the HCT116 human colon cancer cell line and the A2780 human breast cancer cell line. q 2000 Elsevier Science Ltd. All rights
reserved.
Introduction
Paclitaxel has established itself as one of the most important
cancer chemotherapeutic agents in the past decades through
extensive chemical, biological, and medicinal research and
development.1 The chemical synthesis of various paclitaxel
analogs has not only provided critical information on the
structure±activity relationships of the molecule, but has also
rendered access to paclitaxel derivatives with desired
chemical and physical properties. Some of these analogs
have been utilized in helping to understand the interaction
between paclitaxel and its biological target at a molecular
level. The results of these studies may be useful in the
design and discovery of new generation anticancer drugs
that act in a similar way. Structure modi®cation of paclitaxel
at the C-6 position has been achieved by our group and
others, mainly via the 6,7-ole®n (1) and the derived
6,7-diol (2).2 A number of analogs have been synthesized,
The extant biological data suggest that modi®cations of
paclitaxel at the C-6/C-7 positions only cause slight changes
such as 6a-hydroxy-7-deoxy-paclitaxel,3 C-6 monoesters of
6a-hydroxy-7-epipaclitaxel,2a and the C-6/C-7 cyclic ester/
thioester of 6a-hydroxy-7-epipaclitaxel.2
in its cytotoxicity and tubulin-assembly activity.2,3 In con-
tinuation of our efforts to ®nd paclitaxel derivatives with
improved activity and favorable properties, we have made
some further studies of analogs at the C-6/C-7 position. In
this report, we detail the synthesis and biological evaluation
of these analogs.
Keywords: paclitaxel; taxoids; sul®tes; structure±activity; azides.
* Corresponding author. Tel.: 11-540-231-6570; fax: 11-540-231-7702;
e-mail: dkingston@vt.edu
Present address: 251 Nieuwland Science Hall, Department of Chemistry
and Biochemistry, University of Notre Dame, Notre Dame, IN 46556-5670,
USA.
Results and Discussion
²
A single diastereomer of 6a-hydroxy-7-epipaclitaxel 6,7-
O,O0-cyclosul®te (4) with respect to the orientation of the
0040±4020/00/$ - see front matter q 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(00)00611-6

6408
H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
Scheme 1.
lone pair electrons of the sulfur atom was synthesized
previously by treating 20-O-(tert-butyldimethylsilyl)-6a-
hydroxy-7-epipaclitaxel (3) with thionyl chloride and
4-dimethylaminopyridine (DMAP) at 08C.4 We envisioned
that the other diastereomer might exert somewhat different
biological properties due to its different local polarity
pro®le, and also that either the cyclic sul®te or the cyclic
sulfate would be good precursors to C-6 and/or C-7 pacli-
taxel derivatives through nucleophilic displacement. The
synthesis of 4 was thus revisited under conditions that
could give rise to the other diastereomer. Indeed, at room
temperature a mixture of the two cyclic sul®tes (4 and 5) in
about a 1.2 : 1 ratio was obtained and was separated by
careful multiple elution on preparative TLC (Scheme 1).
Deprotection gave the b diastereomer of the cyclic sul®te
(6).
Table 1 for comparison (the corresponding data for the
cyclic sulfate 7 are also included). Compounds 4 and 7
have very similar proton and carbon signals, and both are
considered typical as compared with paclitaxel. The unusual
down®eld chemical shift of the H-5 and H-14a protons in
compound 5 suggested that the lone pair electrons of the
sulfur atom in this compound were in close proximity to
these protons, which in turn established the orientation of
the lone pair electrons as being on the a-face. The opposite
orientation (b) was thus assigned to sul®te 4.
In order to further verify and identify each of the two
diastereomers, both sul®tes were oxidized to the sulfate
with ruthenium trichloride and sodium periodate. As
expected, both 4 and 5 were converted to the same 6a-
hydroxy-7-epipaclitaxel 6,7-O,O0-cyclosulfate (8) in
excellent yield after deprotection (Scheme 2).
The two isomers 4 and 5 had similar polarities, and they also
shared many common resonances in both 1H and 13C NMR
spectra. The signi®cant differences were the protons at the
C-5, C-7, C-14a, and C-20b positions. These are listed in
The 1,2 cyclic sulfate has been known as a useful synthon
with chemical reactivity similar to an epoxide,5 and in many
cases it is even better than an epoxide for nucleophilic
reactions.6 It was thus expected that nucleophilic substi-
tution at either the C-6 or the C-7 position of 7 would
give trans-disubstituted paclitaxel analogs.
Table 1. Comparison of selected proton chemical shifts (ppm) of
compounds 4, 5 and 7
Protons
4
5
7
However, when compound 7 was treated with sodium azide
and tetrabutylammonium azide,7 no detectable reaction was
observed. It was reasoned that the stability of the cyclic
sulfate 7 toward nucleophilic reaction was due to the steric
hindrance of the b face of the molecule caused by the C-19
H-5
H-7
H-14a
H-20b
4.76
4.89
2.40
4.35
5.86
4.39
3.08
5.04
5.03
4.86
2.47
4.46
Scheme 2.

H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
6409
Figure 1. Stereo view of the steric hindrance of the b face of 7 and related compounds.
angular methyl group and the oxetane ring (Fig. 1). This
same steric hindrance initially determined the stereo-
chemistry of the dihydroxylation of the 6,7-ole®n (1),
even in the presence of asymmetric induction that favors
the other (b) isomer, and uniformly gave the 6a,7a-diol
(Fig. 1).8 This result is also in accordance with the observed
stability of paclitaxel 6,7-epoxide. When the epoxide was
treated with a variety of reagents with the intention of
opening the epoxide ring, only starting material was
recovered, and under harsh conditions the epoxide ring
remained intact while the oxetane ring was often opened.9
excellent yield. However, treatment of 9 with sodium
azide induced a pinacol-type rearrangement, presumably
promoted by the weak base NaN3, which led to a product
with a contracted C-ring (10, Scheme 3) instead of the
desired substitution product. This competing reaction has
also been observed previously with other bases.2b,2c,4 To
avoid this undesired reaction it was necessary to eliminate
the retro-aldol pathway by protection of the 7-epi hydroxyl
group, but this was found to be sterically hindered and stable
toward a number of silylation and esteri®cation conditions.
Protection was ®nally achieved with dimethylsilyl chloride
to give 20-O-tert-butyldimethylsilyl-1,7a-bis-O-dimethyl-
silyl-6a-tri¯uoromethanesulfuryl-7-epipaclitaxel (11). In
the event, the C-1 hydroxyl group was also protected.10
Having protected the C-7 hydroxyl group, the azide substi-
tution reaction was effected by treating compound 11 with
sodium azide in anhydrous DMF at room temperature for
18 h, to give the desired 20-O-tert-butyldimethylsilyl-6b-
azido-7-epipaclitaxel (12) in 49% yield. The C-7 dimethyl-
silyl protecting group was removed during the reaction,
Since we were unable to open the cyclic sulfate 7 by nucleo-
philic substitution, we adopted an alternate approach by
converting the 6a hydroxyl group into a good leaving
group, which could be substituted directly by nucleophiles.
Treatment of 20-O-tert-butyldimethylsilyl-6a-hydroxy-7-
epipaclitaxel (3) with tri¯uoromethanesulfonyl chloride
and DMAP furnished the desired 20-O-tert-butyldimethyl-
silyl-6a-tri¯uoromethanesulfuryl-7-epipaclitaxel (9) in
Scheme 3.

6410
H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
Table 2. Selected coupling constants of paclitaxel and its known C-6 and C-
7 analogs
Table 3. Cytotoxicity of C-6/C-7 modi®ed paclitaxels
Compound
IC50 (mM)
Compound
H5±H6a
H5±H6b
H
6a±H7b
H
6b±H7b
5.0
HCT116
A2780
7-epi-Paclitaxel11
3
9.0±9.2
±
3.5±3.7
2.0±2.4
2.4
2.5±2.6
2.0±2.3
0
2.1
±
±
±
±
Paclitaxel
0.0021
0.0190
0.0034
0.0112
4.4±4.8
4.8
4.7±4.8
6a-Hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosul®te (6)
6a-Hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosulfate (8)
6b-Azido-7-epipaclitaxel (13)
6b-Amino-7-epipaclitaxel (15)
9
±
6a-O-Esters of 32a
Paclitaxel
±
9.5±9.6
±
.0.1073
.0.1073
±
±
6a-Hydroxypaclitaxel9b
±
0.00073
0.0802
0.0021
0.0460
which may account for the low yield of this reaction due to
the competing rearrangement discussed earlier. Desilylation
of 12 afforded 13 in 84% yield.
Scheme 4. Although monosubstituted aliphatic triazenes
are normally not isolable, being transient intermediates in
the hydrogenation of azides, they have been isolated
previously in at least one case.12
The stereochemistry at C-6 was determined both by a
NOESY experiment and by the determination of coupling
constants. The only through-space NOE correlation
observed for H-6 of the azide 13 was with 7-epi OH,
which in turn had correlation with H-3. This suggests an a
proton at the C-6 position. To ensure this stereochemistry
assignment, the coupling constants of the H-5, H-6, and H-7
protons of 13 were compared with those of C-6 and C-7
modi®ed analogs with known stereochemistry (Table 2).
The observed coupling constants of compound 13 (H5±H6
8.0±8.4, H6±H7 1.6±2.0) also supported an a proton at the
C-6 position. The structure of 13 was thus determined to be
6b-azido-7-epipaclitaxel. This stereochemistry is what
would be expected from an SN2 reaction.
A stoichiometric amount of Pd±C was then used along with
a much longer reaction time, in an effort to completely
reduce the azide. However, a complex reaction mixture
again resulted after 3 days. The major product, isolated in
62% yield, was identical with 14, suggesting that the tri-
azene 14 is relatively inert toward hydrogenation. Other
paclitaxel azides have proven to be resistant to hydrogena-
tion; thus the 7a-azide could not be hydrogenated at atmos-
pheric pressure,4 and the only reported hydrogenation of
paclitaxel azide in the literature required 48 psi hydrogen
for 72 h and gave only 50% yield.13 A hydrogenation
experiment under 50 psi hydrogen pressure was thus
conducted and yielded the desired 6b-amino-paclitaxel
(15) in an unoptimized 35% yield (Scheme 4).
Having successfully prepared 6b-azido-7-epipaclitaxel
(13), its reduction to the amino analog was the next immedi-
ate goal in order to evaluate the effect of an amino group at
C-6 on the anticancer activity as well as to afford an analog
with improved water solubility in acidic environments.
Compound 13 was hydrogenated in methanol catalyzed by
10% palladium on activated carbon under atmospheric
pressure. TLC analysis after 1.5 h indicated complete
consumption of the starting material and formation of
very polar products (Rf,0 in 7:3 EtOAc:hexanes). The
major component of the reaction mixture isolated in 40%
yield was tentatively identi®ed as 6b-triazeno-7-epipaclit-
1
The structure of compound 15 was assigned by H, 13C,
TOCSY, and HMQC spectra. The 6a proton was found to
shift from 4.12 to 3.35 ppm, while the C-6 carbon was
shifted from 62.3 to 54.6 ppm. Both changes are consistent
with the expected reduction of an azide to an amine. The
molecular formula was con®rmed by HRFABMS.
Compounds 6, 8, 13 and 15 were evaluated in an in vitro
cytotoxicity assay against the human colon cancer cell line
HCT116 and the breast cancer cell line A2780. The results
are summarized in Table 3. Interestingly, while the cyclic
sul®te 6 showed modest activity (about tenfold less active
than paclitaxel in the HCT116 cell line), the cyclic sulfate 8
was essentially inactive. The aminopaclitaxel 15 was also
signi®cantly less active than paclitaxel itself, but the azido
analog 13 was 2±3 times more cytotoxic than paclitaxel.
1
axel (14). The H NMR spectrum of 14 features an up®eld
shift of the C-6 proton from 4.12 to 3.0 ppm, in accordance
with the reduction of the azide functionality. The
HRFABMS revealed, however, that compound 14 had a
molecular formula of C47H52O14N4, corresponding to the
unusual partially hydrogenated triazeno analog, as in
Scheme 4.

H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
6411
Since the amino derivative 15 was not suf®ciently active for
it to be of interest for drug development, we did not pursue
the synthesis of additional analogs in this area.
20-O-(tert-Butyldimethylsilyl)-6a-hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosul®te (a) (5). 1H NMR d 8.02 (d, Jˆ7.2 Hz,
2H), 7.75 (d, Jˆ7.2 Hz, 2H), 7.58±7.28 (m, 11H), 7.11 (d,
Jˆ9.6 Hz, 1H), 6.79 (s, 1H), 6.33 (t, Jˆ8.8 Hz, 1H), 5.85 (s,
1H), 5.80 (m, 2H), 5.04 (d, Jˆ8.4 Hz, 1H), 4.94 (d, Jˆ
6.8 Hz, 1H), 4.65 (d, Jˆ1.6 Hz, 1H), 4.39 (d, Jˆ6.4 Hz,
1H), 4.21 (d, Jˆ8.4 Hz, 1H), 3.80, (d, Jˆ7.2 Hz, 1H),
3.07 (m, 1H), 2.52 (s, 3H), 2.22 (s, 3H), 2.10 (m, 1H),
1.98 (s, 3H), 1.91 (s, 3H), 1.21 (s, 3H), 1.17 (s, 3H), 0.78
(s, 9H), 20.06 (s, 3H), 20.33 (s, 3H). 13C NMR d 204.4,
171.2, 169.7, 169.6, 167.2, 166.9, 141.5, 138.3, 134.2,
133.8, 131.9, 131.8, 129.9, 129.1, 128.9, 128.8, 128.7,
127.9, 127.0, 126.3, 88.0, 84.2, 81.7, 80.1, 78.9, 77.2,
75.6, 75.2, 74.6, 70.8, 55.5, 54.2, 42.6, 42.0, 35.8, 25.8,
25.5, 22.9, 21.4, 20.8, 18.1, 15.4, 15.0, 25.3, 25.9.
HRFABMS calculated for C53H63NO16SSi (M1H)1
1030.3716, found 1030.3701, error 21.4 ppm.
Experimental
General methods
Unless otherwise noted, all materials were used as received
from a commercial supplier without further puri®cation. All
anhydrous reactions were performed in oven-dried glass-
ware under argon. Anhydrous THF and Et2O were distilled
from sodium/benzophenone. Anhydrous toluene was
distilled from sodium. CH2Cl2 was distilled from CaH2.
All reactions were monitored by E. Merck analytical thin
layer chromatography (TLC) plates (silica gel 60 GF,
aluminum back) and analyzed with 254 nm UV light and/
or vanillin/sulfuric acid spray. Silica gel for column
chromatography was purchased from E. Merck (230±400
mesh). Preparative thin layer chromatography (PTLC)
plates (silica gel 60 GF) were purchased from Analtech.
1H and 13C NMR spectra were obtained in CDCl3 on a
Varian Unity 400 spectrometer (operating at 399.951 MHz
for 1H and 100.578 MHz for 13C) or a Bruker WP 360 spec-
trometer (operating at 360.140 MHz for 1H and 90.562 MHz
for 13C), and were assigned by comparison of chemical
shifts and coupling constants with those of related
compounds and by appropriate 2D-NMR techniques
(TOCSY, HMQC, HMBC, NOESY). All 2D-NMR spectra
were obtained on the Varian Unity 400 spectrometer.
Chemical shifts were reported as d-values using residual
CHCl3 as internal reference, and coupling constants were
reported in Hertz. Mass spectra (LRFABMS/HRFABMS)
were obtained at the Nebraska Center for Mass Spectro-
metry, University of Nebraska.
6a-Hydroxy-7-epipaclitaxel 6,7-O,O0-cyclosul®te (a) (6).
To a solution of 20-O-(tert-butyldimethylsilyl)-6a-hydroxy-
7-epipaclitaxel 6,7-O,O0-cyclosul®te (a) (5, 3.2 mg,
0.0031 mmol) in dry THF (100 mL) was added HF-pyridine
(70%, 20 mL) and the solution was stirred at room tempera-
ture for 4 h. The reaction mixture was diluted with EtOAc
and washed with dilute sodium bicarbonate and dilute HCl
(1 N), the organic layers were combined and washed with
water and brine, dried over anhydrous sodium sulfate, and
concentrated under reduced pressure. The crude product
was puri®ed by preparative TLC (silica gel, 500m, EtOA-
c:hexanes 6:4) to afford 6a-hydroxy-7-epipaclitaxel 6,7-
1
O,O0-cyclosul®te (a) (6, 2.8 mg, 98%). H NMR d 7.98
(d, Jˆ8.4 Hz, 2H), 7.78 (dd, Jˆ8.0 Hz, 1.2, 2H), 7.60±
7.32 (m, 11H), 7.13 (d, Jˆ9.6 Hz, 1H), 6.72 (s, 1H), 6.17
(t, Jˆ8.0 Hz, 1H), 5.81 (m, 2H), 5.80 (s, 1H), 5.00 (d,
Jˆ8.4 Hz, 1H), 4.94 (d, Jˆ6.4 Hz, 1H), 4.78 (m, 1H),
4.36 (d, Jˆ6.8 Hz, 1H), 4.16 (d, Jˆ8.0 Hz, 1H), 3.77 (d,
Jˆ6.8 Hz, 1H), 3.43 (d, Jˆ5.2 Hz, 1H), 3.02 (m, 1H), 2.36
(s, 3H), 2.33 (m, 1H), 2.23 (s, 3H), 1.89 (s, 3H), 1.72 (s, 3H),
1.18 (s, 3H), 1.17 (s, 3H). 13C NMR d 203.6, 172.3, 169.6
(2C), 167.1, 166.7, 140.7, 138.1, 133.9, 133.8, 132.8, 131.9,
129.8, 129.1, 128.9 (2C), 128.7, 128.3, 127.0 (2C), 87.6,
84.2, 81.7, 80.3, 78.4, 77.2, 75.4, 74.1, 73.4, 72.3, 54.8,
54.4, 42.4, 42.0, 35.6, 26.5, 22.6, 20.8, 20.5, 15.7, 14.8.
HRFABMS calculated for C47H49NO16S (M1H)1
916.2851, found 916.2844, error 20.7 ppm.
20-O-(tert-Butyldimethylsilyl)-6a-hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosul®te (b) (4) and 20-O-(tert-butyldi-
methylsilyl)-6a-hydroxy-7-epipaclitaxel 6,7-O,O0-cyclo-
sul®te (a) (5). To
a
solution of 20-O-(tert-
butyldimethylsilyl)-6a-hydroxy-7-epipaclitaxel (3, 40 mg,
0.041 mmol) in dry CH2Cl2 (2 mL) was added DMAP
(20 mg, 0.16 mmol) and SOCl2 (9 mL, 0.12 mmol) and the
mixture was stirred at room temperature for 1.5 h. The
reaction mixture was then diluted with EtOAc and washed
with water and brine, dried over anhydrous sodium sulfate,
and concentrated under reduced pressure. The residue was
puri®ed by preparative TLC (silica gel, 1000m, EtOAc:
hexanes 4:6) to afford 27.8 mg of a mixture of 20-O-(tert-
butyldimethylsilyl)-6a,7a-O-cyclosulfonyl(a)-paclitaxel (5)
and 20-O-(tert-butyldimethylsilyl)-6a,7a-O-cyclosulfonyl
(b)-paclitaxel (4) along with starting material (3, 10 mg).
The mixture was further separated by preparative TLC
(silica gel, 1000m, EtOAc:hexanes 2.5:7.5, multiple elution)
to afford 20-O-(tert-butyldimethylsilyl)-6a-hydroxy-7-
epipaclitaxel 6,7-O,O0-cyclosul®te (a) (5, 9 mg, 37%
based on unrecovered staring material) and 20-O-(tert-butyl-
dimethylsilyl)-6a-hydroxy-7-epipaclitaxel 6,7-O,O0-cyclo-
sul®te (b) (4, 11 mg, 46% based on unrecovered staring
material) along with starting material (3, 6 mg, decomposed
on PTLC plate).
20-O-(tert-Butyldimethylsilyl)-6a-hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosulfate (7). Route A Ð from 4: To a solution
of 20-O-(tert-butyldimethylsilyl)-6a-hydroxy-7-epipacli-
taxel 6,7-O,O0-cyclosul®te (b) (4, 4.6 mg, 0.0045 mmol)
in CCl4/H2O/CH3CN (120 mL/240 mL/120 mL) was added
sodium periodate (NaIO4, 4.8 mg, 0.022 mmol) and
ruthenium trichloride (RuCl3´H2O, 1.2 mg, 0.0046 mmol)
and the mixture was stirred at room temperature for 2 h.
The reaction mixture was then ®ltered through a pad of
silica gel and rinsed with EtOAc (1 mL£2). The ®ltrate
was concentrated under reduced pressure to afford 20-O-
(tert-butyldimethylsilyl)-6a-hydroxy-7-epipaclitaxel 6,7-
O,O0-cyclosul®te (7, 4.6 mg, 98%).
Route BÐfrom 5: To a solution of 20-O-(tert-butyl-
dimethylsilyl)-6a-hydroxy-7-epipaclitaxel 6,7-O-O0-cyclo-
sul®te (a) (5, 23 mg, 0.022 mmol) in CCl4/H2O/CH3CN

6412
H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
(0.6 mL/1.2 mL/0.6 mL) was added sodium periodate
(NaIO4, 24 mg, 0.11 mmol) and ruthenium trichloride
(RuCl3´H2O, 6 mg, 0.023 mmol) and the mixture was stirred
at room temperature for 1.5 h. The reaction mixture was
then ®ltered through a pad of silica gel and rinsed with
EtOAc (2 mL£2). The ®ltrate was concentrated under
reduced pressure to afford 20-O-(tert-butyldimethylsilyl)-
(d, Jˆ8.9 Hz, 1H), 3.97 (d, Jˆ7.3 Hz, 1H), 3.92 (dd, Jˆ4.6,
12.0 Hz, 1H), 2.71 (s, 3H), 2.40 (dd, Jˆ9.6, 15.2 Hz, 1H),
2.20 (s, 3H), 2.11 (dd, Jˆ9.6, 15.2 Hz, 1H), 1.91 (s, 3H),
1.68 (s, 3H), 1.20 (s, 3H), 1.13 (s, 3H), 0.78 (s, 9H), 20.05
(s, 3H), 20.30 (s, 3H). HRFABMS calculated for
C53H63NO14SiNa (M1Na-CF3SO3H)1 988.3915, found
988.3913, error 20.2 ppm.
6a-hydroxy-7-epipaclitaxel
6,7-O,O0-cyclosulfate
(7,
1
23 mg, 98%). H NMR d 8.15 (dd, Jˆ9.2, 1.6 Hz, 2H),
7.73 (dd, Jˆ9.2, 1.6 Hz, 2H), 7.62±7.32 (m, 11H), 7.08
(d, Jˆ10.0 Hz, 1H), 6.69 (s, 1H), 6.30 (t, Jˆ9.6 Hz, 1H),
5.80 (2d, 2H), 5.03 (s, 1H), 5.01 (d, Jˆ7.6 Hz, 1H), 4.86 (d,
Jˆ7.6 Hz, 1H), 4.69 (d, Jˆ2.4 Hz, 1H), 4.46 (d, Jˆ9.6 Hz,
1H), 4.24 (d, Jˆ9.6 Hz, 1H), 4.03 (d, Jˆ7.6 Hz, 1H), 2.64
(s, 3H), 2.48 (m, 1H), 2.21 (s, 3H), 2.13 (m, 1H), 1.99 (br s,
3H), 1.90 (s, 3H), 1.21 (s, 3H), 1.17 (s, 3H), 0.78 (s, 9H),
20.04 (s, 3H), 20.32 (s, 3H).
1-O-Dimethylsilyl-20-O-(tert-butyldimethylsilyl)-6a-tri-
¯uoromethanesulfuryl-7-O-dimethylsilyl-7-epipaclitaxel
(11). To a solution of 20-O-(tert-butyldimethylsilyl)-6a-
tri¯uoromethanesulfuryl-7-epipaclitaxel (9, 68 mg, 0.060
mmol) in dry CH2Cl2 (3 mL) was added imidazole
(80 mg, 1.2 mmol) and chlorodimethylsilane (DMSCl,
35 mL, 0.31 mmol) and the solution was stirred at room
temperature for 1.75 h. The reaction mixture was then
directly subjected to column chromatography (silica gel,
EtOAc:hexanes 3:7) to afford 1-O-dimethylsilyl-20-O-
(tert-butyldimethylsilyl)-6a-tri¯uoromethanesulfuryl-7-O-
dimethylsilyl-7-epipaclitaxel (11, 71 mg, 95%). 1H NMR d
8.16 (d, Jˆ8.4 Hz, 2H), 7.75 (m, 2H), 7.61±7.32 (m, 11H),
7.05 (d, Jˆ9.2 Hz, 1H), 6.39 (s, 1H), 6.36 (t, Jˆ8.4 Hz, 1H),
5.91 (dd, Jˆ9.2 Hz, 2.0, 1H), 5.64 (d, Jˆ7.2 Hz, 1H), 5.33
(dd, Jˆ6.4 Hz, 2.4, 1H), 5.02 (d, Jˆ6.4 Hz, 1H), 4.84 (m,
1H), 4.71 (d, Jˆ2.4 Hz, 1H), 4.65 (d, Jˆ8.4 Hz, 1H), 4.32
(m, 1H), 4.19 (2d, 2H), 4.04 (d, Jˆ2.0 Hz, 1H), 2.69 (s, 3H),
2.50 (m, 1H), 2.35 (m, 1H), 2.19 (s, 3H), 1.91 (br s, 3H),
1.69 (s, 3H), 1.15 (s, 3H), 1.07 (s, 3H), 0.78 (s, 9H), 0.36 (d,
Jˆ2.8 Hz, 3H), 0.33 (d, Jˆ2.8 Hz, 3H), 20.04 (s, 3H),
20.18 (d, Jˆ2.8 Hz, 1H), 20.30 (s, 3H), 20.47 (d, Jˆ
2.8 Hz, 3H).
6a-Hydroxy-7-epipaclitaxel 6,7-O,O0-cyclosulfate (8). To
a solution of 20-O-(tert-butyldimethylsilyl)-6a-hydroxy-7-
epipaclitaxel 6,7-O,O0-cyclosulfate (7, 4.6 mg, 0.0044
mmol) in dry THF (250 mL) was added HF-pyridine
(70%, 40 mL) and the solution was stirred at room tempera-
ture for 4 h. The reaction mixture was diluted with EtOAc
and washed with dilute sodium bicarbonate and dilute HCl
(1 N), the organic layers were combined and washed with
water and brine, dried over anhydrous sodium sulfate, and
concentrated under reduced pressure. The crude product
was puri®ed by preparative TLC (silica gel, 500m, EtOAc:
hexanes 6:4) to afford 6a-hydroxy-7-epipaclitaxel 6,7-
1
O,O0-cyclosulfate (8, 3.7 mg, 90%). H NMR d 8.13 (d,
Jˆ7.2 Hz, 2H), 7.74 (d, Jˆ7.6 Hz, 2H), 7.64±7.35 (m,
11H), 7.01 (d, Jˆ9.2 Hz, 1H), 6.65 (s, 1H), 6.22 (t,
Jˆ8.8 Hz, 1H), 5.82 (2d, 2H), 4.99 (d, Jˆ7.2 Hz, 1H),
4.98 (s, 1H), 4.84 (br s, 1H), 4.83 (d, Jˆ7.2 Hz, 1H), 4.43
(d, Jˆ8.0 Hz, 1H), 4.21 (d, Jˆ8.0 Hz, 1H), 4.03 (d,
Jˆ7.2 Hz, 1H), 3.59 (br d, Jˆ4.4 Hz, 1H), 2.45 (s, 3H),
2.44 (m, 1H), 2.33 (m, 1H), 2.21 (s, 3H), 1.88 (s, 3H),
1.86 (s, 3H), 1.21 (s, 3H), 1.18 (s, 3H). 13C NMR d 202.1,
172.7, 169.5, 169.1, 167.1, 166.9, 141.5, 138.0, 134.0,
132.0, 131.9, 130.2, 129.0, 128.9, 128.7, 128.3, 127.0,
84.8, 82.5, 80.3, 80.0, 78.3, 77.3, 74.1, 73.1, 71.8, 55.2,
54.7, 42.5, 39.2, 35.9, 25.8, 22.3, 21.1, 20.6, 15.1, 13.5.
HRFABMS calculated for C47H49NO17S (M1H)1
932.2800, found 932.2813, error 1.4 ppm.
1-O-Dimethylsilyl-20-O-(tert-butyldimethylsilyl)-6b-azido-
7-epipaclitaxel (12). To a solution of 1-O-dimethylsilyl-20-
O-(tert-butyldimethylsilyl)-6a-tri¯uoromethanesulfuryl-7-
O-dimethylsilyl-7-epipaclitaxel (11, 71 mg, 0.057 mmol) in
DMF (1.5 mL) was added sodium azide (NaN3, 100 mg,
1.5 mmol) and the solution was stirred at room temperature
for 18 h. The reaction mixture was then diluted with EtOAc
and washed with water and brine, dried over anhydrous
sodium sulfate, and concentrated under reduced pressure.
The crude product was puri®ed by preparative TLC (silica
gel, 1000m, EtOAc:hexanes 4:6) to afford 1-O-dimethyl-
silyl-20-O-(tert-butyldimethylsilyl)-6b-azido-7-epipaclitaxel
(12, 25 mg, 49%, based on unrecovered starting material),
the starting material (11, 11 mg) along with other products
that were not fully characterized. 1H NMR d 8.18 (d,
Jˆ8.0 Hz, 2H), 7.72 (m, 2H), 7.60±7.31 (m, 11H), 7.03
(d, Jˆ9.6 Hz, 1H), 6.79 (s, 1H), 6.32 (t, Jˆ8.8 Hz, 1H),
5.87 (dd, Jˆ9.6, 2.4 Hz, 1H), 5.78 (d, Jˆ7.6 Hz, 1H),
5.05 (d, Jˆ8.0 Hz, 1H), 4.68 (d, Jˆ2.4 Hz, 1H), 4.57 (d,
Jˆ11.2 Hz, 1H), 4.36 (s, 2H), 4.26 (m, 1H), 4.13 (d,
Jˆ9.2 Hz, 1H), 3.86 (d, Jˆ7.2 Hz, 1H), 3.68 (dd, Jˆ10.8,
1.2 Hz, 1H), 2.72 (s, 3H), 2.31 (m, 2H), 2.19 (s, 3H), 2.04 (s,
3H), 1.88 (s, 3H), 1.79 (s, 3H), 1.13 (s, 3H), 1.11 (s, 3H),
0.76 (s, 9H), 20.05 (s, 3H), 20.15 (d, Jˆ2.4 Hz, 1H),
20.31 (s, 3H), 20.49 (d, Jˆ2.8 Hz, 3H). 13C NMR d
206.0, 172.6, 170.9, 169.5, 166.8, 165.6, 139.5, 138.0,
133.9, 133.4, 132.8, 131.7, 130.4, 129.9, 128.75, 128.69,
128.6, 127.9, 127.1, 126.4, 82.7, 81.9, 80.7, 78.6, 77.8,
75.7, 75.4, 70.5, 62.1, 57.5, 55.4, 43.5, 39.9, 34.8, 26.8,
25.5, 22.7, 22.1, 20.9, 18.2, 14.8, 14.7, 0.3, 20.4, 25.3,
26.0. FT-IR 2104.5 cm21 (strong).
20-O-(tert-Butyldimethylsilyl)-6a-tri¯uoromethanesulfuryl-
7-epipaclitaxel (9). 20-O-(tert-Butyldimethylsilyl)-6a-hy-
droxy-7-epipaclitaxel (3, 170 mg) was dissolved in dry
CH2Cl2 (2 mL) and treated with 4-dimethylaminopyridine
(102 mg, 5 eq.) and tri¯uoromethanesulfonyl chloride
(47 mL, 2 eq.) at 08C with stirring for 1 h. The reaction
mixture was then diluted with EtOAc (4.0 mL) and the
precipitate ®ltered off on Celite. The resulting solution
was evaporated and the residue puri®ed by preparative
TLC (silica gel, 1:1 EtOAc:hexanes) to furnish 20-O-(tert-
butyldimethylsilyl)-6a-tri¯uoromethanesulfuryl-7-epipacli-
1
taxel (9, 176 mg, 96%). H NMR d 8.15 (d, 2H), 7.70 (d,
2H), 7.62 (t, 1H), 7.64±7.26 (m, 10H), 7.07 (d, Jˆ9.2 Hz,
1H), 6.77 (s, 1H), 6.29 (t, Jˆ8.6 Hz, 1H), 5.80 (dd, Jˆ8.8,
2.4 Hz, 1H), 5.73 (d, Jˆ7.3 Hz, 1H), 5.28 (dd, Jˆ4.6,
2.7 Hz, 1H), 4.93 (d, Jˆ2.8 Hz, 1H), 4.90 (d, Jˆ12.0 Hz
1H) 4.67 (d, Jˆ2.3 Hz, 1H), 4.47 (d, Jˆ8.9 Hz, 1H), 4.38

H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
6413
6b-Azido-7-epipaclitaxel (13). To a solution of 1-O-
dimethylsilyl-20-O-(tert-butyldimethylsilyl)-6b-azido-7-epi
paclitaxel (12, 25 mg, 0.023 mmol) in dry THF (0.5 mL)
was added HF-pyridine (70%, 100 mL) and the solution
was stirred at room temperature for 3 h. The reaction
mixture was then diluted with EtOAc and washed with
dilute sodium bicarbonate and dilute HCl (1 N), the organic
layers were combined and washed with water and brine,
dried over anhydrous sodium sulfate, and concentrated
under reduced pressure. The crude product was puri®ed
by preparative TLC (silica gel, 1000m, EtOAc:hexanes
6:4) to afford 6b-azido-7-epipaclitaxel (13, 17.6 mg,
57.8, 54.7, 42.8, 42.6, 40.5, 35.7, 25.8, 22.5, 20.9, 20.8,
17.5, 14.7. LRFABMS calculated for C47H52N4O14
(M1H)1 897.8, found 897.4; (M1Na)1 919.3, found
919.4; (M1Li)1 903.4, found 903.4.
6b-Amino-7-epipaclitaxel (15). To a solution of 6b-azido-
7-epipaclitaxel (13, 34 mg, 0.038 mmol) in EtOAc (10 mL)
was added catalytic amount of palladium on activated
carbon (5%). This suspension was stirred under 50 psi
hydrogen atmosphere for 24 h. The suspension was ®ltered
through Celite and the ®ltrate was concentrated under
reduced pressure. The residue was puri®ed by preparative
TLC (silica gel, 1000m, MeOH:CH2Cl2 8:92) to furnish 6b-
1
84%). H NMR d (8.16 (d, Jˆ7.6 Hz, 2H), 7.71 (m, 2H),
1
7.62±7.35 (m, 11H), 7.05 (d, Jˆ8.8 Hz, 1H), 6.77 (s, 1H),
6.20 (t, Jˆ8.8 Hz, 1H), 5.80 (dd, Jˆ8.8, 2.0 Hz, 1H), 5.76
(d, Jˆ7.6 Hz, 1H), 5.03 (d, Jˆ8.0 Hz, 1H), 4.79 (m, 1H),
4.49 (d, Jˆ10.8 Hz, 1H), 4.42 (d, Jˆ8.8 Hz, 1H), 4.27 (d,
Jˆ8.8 Hz, 1H), 4.12 (m, 1H), 3.86 (d, Jˆ7.2 Hz, 1H), 3.68
(br s, 1H), 3.66 (m, 1H), 2.50 (s, 3H), 2.35 (m, 1H), 2.26 (m,
1H), 2.18 (s, 3H), 1.77 (s, 3H), 1.76 (s, 3H), 1.17 (s, 3H),
1.13 (s, 3H). 13C NMR d 205.6, 172.6, 172.4, 169.5, 167.1,
167.07, 139.6, 137.9, 133.8, 133.6, 133.3, 132.0, 130.2,
129.1, 129.0, 128.8, 128.7, 128.3, 127.0, 126.9, 82.6, 80.5,
79.0, 78.4, 78.1, 77.2, 75.0, 73.2, 72.0, 62.3, 57.7, 54.9,
42.6, 40.1, 36.0, 25.8, 22.3, 21.1, 20.8, 14.76, 14.73.
HRFABMS calculated for C47H50N4O14 (M1H)1
895.3402, found 895.3401, error 20.1 ppm.
amino-7-epipaclitaxel (15, 12 mg, 35%). H NMR d 8.17
(dd, Jˆ7.2, 1.6 Hz, 2H), 7.71 (dd, Jˆ6.8, 1.6 Hz, 2H),
7.62±7.34 (m, 11H), 7.11 (d, Jˆ9.2 Hz, 1H), 6.81 (s, 1H),
6.19 (t, Jˆ8.8 Hz, 1H), 5.79 (dd, Jˆ8.8, 2.4 Hz, 1H), 5.75
(d, Jˆ7.6 Hz, 1H), 4.90 (d, Jˆ8.4 Hz, 1H), 4.78 (d, Jˆ
2.4 Hz, 1H), 4.39 (d, Jˆ8.0 Hz, 1H), 4.17 (d, Jˆ8.4 Hz,
1H), 3.85 (d, Jˆ7.6 Hz, 1H), 3.52 (br s, 1H), 3.35 (d,
Jˆ8.0 Hz, 1H), 2.47 (s, 3H), 2.36 (m, 1H), 2.23 (m, 1H),
2.1±2.2 (br, 3H), 2.17 (s, 3H), 1.75 (s, 3H), 1.73 (s, 3H),
1.16 (s, 3H), 1.13 (s, 3H). 13C NMR d 206.5, 172.6, 172.1,
169.5, 167.2, 167.1, 139.4, 138.0, 133.7, 133.6, 133.2,
131.9, 130.2, 129.3, 129.0, 128.8, 128.6, 128.3, 127.1,
126.9, 83.5, 83.0, 80.7, 79.1, 78.0, 76.2, 75.2, 73.2, 72.0,
58.0, 54.9, 54.6, 42.5, 40.2, 36.0, 25.8, 22.3, 21.1, 20.9,
14.8, 14.5. HRFABMS calculated for C47H52N2O14
(M1H)1 869.3497, found 869.3504, error 0.8 ppm.
Hydrogenation of 6b-azido-7-epipaclitaxel (13). To a
solution of 6b-azido-7-epipaclitaxel (13, 10 mg, 0.011
mmol) in MeOH (1 mL) was added a catalytic amount of
palladium on activated carbon (10%). Hydrogen gas was
bubbled through a needle into the suspension for 1.5 h.
TLC analysis showed complete disappearance of the
starting material and formation of a very polar spot. The
suspension was ®ltered through Celite and the ®ltrate was
concentrated under reduced pressure. The residue was
puri®ed by preparative TLC (silica gel, 500m,
MeOH:CH2Cl2 8:92) to give a complex reaction mixture.
The major component (4.0 mg, 40%) was isolated and
identi®ed as 6b-triazeno-7-epipaclitaxel (14).
Acknowledgements
Financial support by the National Cancer Institute, NIH
(Grant Number CA-69571) is gratefully acknowledged, as
is a gift of paclitaxel from Bristol-Myers Squibb Pharma-
ceutical Research Institute, Wallingford. High resolution
mass spectra were obtained at the Nebraska Center for
Mass Spectrometry. We thank Kathy Johnston, Russ
Peterson and Judy MacBeth (Bristol-Myers Squibb,
Princeton) for carrying out the in vitro cytotoxicity assays.
6b-Triazeno-7-epipaclitaxel (14). To a solution of 6b-
azido-7-epipaclitaxel (13, 7.8 mg, 0.0088 mmol) in MeOH
(2 mL) was added excess amount of palladium on activated
carbon (10%). This suspension was stirred under hydrogen
atmosphere for 72 h. The suspension was ®ltered through
Celite and the ®ltrate was concentrated under reduced
pressure. The residue was puri®ed by preparative TLC
(silica gel, 500m, MeOH:EtOAc 5:95) to afford 6b-tri-
References
1. For reviews of the chemistry, structure±activity relationships of
paclitaxel, and the development and clinical utility of the taxane
class of anticancer agents, see: (a) Kingston, D. G. I. Pharmac.
Ther. 1991, 52, 1±34. (b) Kingston, D. G. I. Trends Biotechnol.
1994, 12, 222±227. (c) Chen, S.-H.; Farina, V. Paclitaxel
structure±activity relationships and core skeletal rearrangements.
In Taxane Anticancer Agents: Basic Science and Current Status;
Georg, G. I., Chen, T. T., Ojima, I., Vyas, D. M., Eds.; ACS
Symposium Series 583, American Chemical Society: Washington,
DC, 1994; pp 247±261. (d) Georg, G. I.; Boge, T. C.; Cheruvallath,
Z. S.; Clowers, J. S.; Harriman, G. C. B.; Hepperle, M.; Park, H.
The medicinal chemistry of taxol. In Taxol: Science and Appli-
cations; Suffness, M., Ed.; CRC: Boca Raton, FL, 1995; pp 317±
375. (e) Kingston, D. G. I. Recent advances in the chemistry and
structure±activity relationships of paclitaxel. In Taxane Anticancer
Agents: Basic Science and Current Status; George, G. I., Chen,
T. T., Ojima, I., Vyas, D. M., Eds.; ACS Symposium Series 583,
American Chemical Society: Washington, DC, 1994; pp 203±216.
1
azeno-7-epipaclitaxel (14, 4.8 mg, 62%). H NMR d 8.12
(dd, Jˆ8.0 Hz, 1.2, 2H), 7.77 (dd, Jˆ8.0, 1.2 Hz, 2H),
7.62±7.34 (m, 11H), 7.17 (d, Jˆ9.6 Hz, 1H), 6.80 (s, 1H),
6.15 (t, Jˆ8.4 Hz, 1H), 5.83 (dd, Jˆ8.8, 2.4 Hz, 1H), 5.80
(d, Jˆ6.0 Hz, 1H), 5.15 (d, Jˆ3.2 Hz, 1H), 4.80 (d, Jˆ
2.4 Hz, 1H), 4.39 (d, Jˆ8.0 Hz, 1H), 4.19 (d, Jˆ7.6 Hz,
1H), 4.03 (d, Jˆ6.0 Hz, 1H), 3.82 (d, Jˆ6.4 Hz, 1H), 3.80
(m, 1H), 3.61 (br s, 1H), 3.01 (m, 1H), 2.41 (s, 3H), 2.33 (m,
1H), 2.23 (m, 1H), 2.21 (s, 3H), 1.91 (s, 3H), 1.75 (s, 3H),
1.20 (s, 3H), 1.18 (s, 3H). 13C NMR d 203.8, 171.9, 170.9,
169.4, 167.0, 166.8, 139.3, 138.1, 133.9, 133.8, 133.7,
131.9, 130.1, 129.1, 128.9, 128.7, 128.2, 127.02, 127.05,
82.5, 80.8, 79.3, 78.3, 77.2, 75.3, 73.8, 73.4, 71.8, 69.1,

6414
H. Yuan et al. / Tetrahedron 56 (2000) 6407±6414
(f) Nicolaou, K. C.; Dai, W.-M.; Guy, R. K. Angew. Chem. Int. Ed.
Engl. 1994, 33, 15±44. (g) Vyas, D. M.; Kadow, J. F. In Progress
in Medicinal Chemistry; Ellis, G. P., Luscombe, D. K., Eds.;
Elsevier: Amsterdam, 1995; Vol. 32, pp 289±337. (h) Guenard,
D.; Gueritte-Voegelein, F.; Lavelle, F. Curr. Pharm. Design 1995,
1, 95±112. (i) Rowinsky, E. K. Ann. Rev. Med. 1997, 48, 353±374.
2. (a) Liang, X.; Kingston, D. G. I.; Lin, C. M.; Hamel, E. Tetra-
hedron Lett. 1995, 36, 2901±2904. (b) Chen, S.-H.; Huang, S.;
Roth, G. P. Tetrahedron Lett. 1995, 36, 8933±8936. (c) Liang,
X.; Kingston, D. G. I.; Long, B. H.; Fairchild, C. A.; Johnston,
K. A. Tetrahedron 1997, 53, 3441±3456. (d) Johnson, R. A.; Nidy,
E. G.; Dobrowolski, P. J.; Gebhard, I.; Qualls, S. J.; Wicnienski,
N. A.; Kelly, R. C. Tetrahedron Lett. 1994, 35, 7893±7896.
3. Wittman, M. D.; Alstadt, T. J.; Kadow, J. F.; Vyas, D. M.;
Johnson, K.; Fairchild, C.; Long, B. Tetrahedron Lett. 1999, 40,
4943±4946.
7. Prepared from the reaction of tetrabutylammonium hydroxide
and sodium azide, see: Brandstrom, A.; Lamm, B.; Palmertz, I.
Acta. Chem. Scand. B 1974, 28, 699.
8. Liang, X.; Kingston, D. G. I. Unpublished results.
9. The epoxide was initially assigned as b and was considered a
versatile intermediate to a number of interesting analogs. However,
it proved extremely stable toward various nucleophiles. The stereo-
chemistry of the epoxide was later revised to a, its stability was
thus explainable on the basis of hindered nucleophilic attack as
shown in Fig. 1. See: (a) Ref. 2a. (b) Alstadt, T. J.; Gao, Q.;
Wittman, M. D.; Kadow, J. F.; Vyas, D. M. Tetrahedron Lett.
1998, 39, 4965±4966. (c) Yuan, H.; Kingston, D. G. I. Tetrahedron
Lett. 1998, 39, 4967±4970.
10. Chen, S. H.; Kadow, J. F.; Farina, V. J. Org. Chem. 1994, 59,
6156±6158.
11. Chaudhary, A. G.; Rimoldi, J. M.; Kingston, D. G. I. J. Org.
Chem. 1993, 58, 3798±3799.
4. Liang, X. Dissertation: Studies on Ring C Modi®ed Paclitaxel
Analogs, Virginia Polytechnic Institute and State University, 1996.
5. For a review, see: Lohray, B. B. Synthesis 1992, 11, 1035.
6. (a) Gao, Y.; Sharpless, K. B. J. Am. Chem. Soc. 1988, 7538±
7539. (b) Kim, B. M.; Sharpless, K. B. Tetrahedron Lett. 1989, 30,
655±658.
12. Gaoni, Y. J. Org. Chem. 1994, 59, 6853±6855.
13. Menichincheri, M.; Botta, M.; Ceccarelli, W.; Ciomei, M.;
Correli, F.; D'Attello, M.; Fusar-Bassini, D.; Mongelli, N.; Pesenti,
E.; Pinciroli, V.; Ta®, A.; Vanotti, E. Med. Chem. Res. 1995, 534±
555.