Synthesis and Biological Evaluation of D-Ring-Modified Taxanes:1 5(20)-Azadocetaxel Analogs

Article abstract of DOI:10.1021/jo9706842

Two 5(20)-aza analogs of docetaxel, N-20-benzyl-5(20)-azadocetaxel (5) and 5(20)-azadocetaxel (6), have been synthesized from 10-deacetylbaccatin III. The key steps of this synthesis involved the direct introduction of a C-5 leaving group while ring opening and the intramolecular nucleophilic attack of the C-20 amino group at C-5. Both compounds were inactive on the in vitro cytotoxic assay, and only the azadocetaxel 6 retains an antitubulin activity, but 16 times less than docetaxel.

Full text of DOI:10.1021/jo9706842

J . Org. Chem. 1997, 62, 6631-6637
6631
Syn th esis a n d Biologica l Eva lu a tion of D-Rin g-Mod ified Ta xa n es:1
5(20)-Aza d oceta xel An a logs
Raphae¨le Marder-Karsenti, J oe¨lle Dubois,* Laurent Bricard, Daniel Gue´nard, and
Franc¸oise Gue´ritte-Voegelein
Institut de Chimie des Substances Naturelles, Centre National de la Recherche Scientifique,
avenue de la Terrasse, 91198 Gif-sur-Yvette Cedex, France
Received April 16, 1997X
Two 5(20)-aza analogs of docetaxel, N-20-benzyl-5(20)-azadocetaxel (5) and 5(20)-azadocetaxel (6),
have been synthesized from 10-deacetylbaccatin III. The key steps of this synthesis involved the
direct introduction of a C-5 leaving group while ring opening and the intramolecular nucleophilic
attack of the C-20 amino group at C-5. Both compounds were inactive on the in vitro cytotoxic
assay, and only the azadocetaxel 6 retains an antitubulin activity, but 16 times less than docetaxel.
Since its isolation from the bark of the pacific yew tree,
Taxus brevifolia, at the end of the 1960’s,2 paclitaxel
(Taxol, 1a ) has been extensively studied and is now
clinically used in ovarian and breast cancer chemo-
therapy. This diterpenoid and related compounds pos-
sess an original mechanism of action on microtubules3
and constitute a new class of antimitotic agents. The
isolation of 10-deacetylbaccatin III (2) from the needles
of the European yew tree, Taxus baccata L.,4 permitted
the first semisynthesis of paclitaxel4b,5 and the discovery
of docetaxel (Taxotere, 1b),5 a clinically active agent, now
used against breast and non-small-cell lung cancers.
Based on the inactivity of D-secotaxol derivatives like
3,7 it has also been reported that the oxetane ring was
essential for interaction with microtubules and cytotox-
icity. This observation seemed unsatisfactory because
these compounds lacked the C-4 acetate group, known
to be important for activity.8 Furthermore they possessed
a C-5R hydroxyl moiety instead of a â-oxygen.
The oxetane ring might either act to rigidify ring C
and point the C-4 acetoxy group in the appropriate
direction for binding or be directly involved in the
interaction with microtubules by its oxygen atom. Thus,
replacement of this atom by another heteroatom or by a
methylene group might distinguish between conforma-
tional and electronic contributions. We first decided to
replace the oxetane by an azetidine ring. While pro-
gressing in our work, the synthesis of azetidine deriva-
tives 4 was described.9 But these compounds lacked the
C-1 and C-2 functionalities, the presence of the latter,
at least, being very important for the biological activity.
Thus, in order to evaluate the influence of the azetidine
ring on antitubulin activity, real docetaxel or paclitaxel
analogs were needed.
The availability of significant amounts of 2 secured the
long term supply of these anticancer drugs and facilitated
semisynthetic studies directed at the elucidation of the
paclitaxel pharmacophore. The number of analogs which
have been synthesized by different groups contributed
to the structure-activity relationship (SAR) studies.6
Thus, the C-13 phenylisoserine side chain and the C-2
and C-4 esters have been determined to be crucial for
biological activity.
X Abstract published in Advance ACS Abstracts, September 1, 1997.
(1) For general reviews on taxoid chemistry; see: (a) Taxol: Science
and Applications; Suffness, M., Ed.; CRC Press: New York, 1995. (b)
Taxane Anticancer Agents: Basic Science and Current Status; Georg,
I. G., Chen, T. T., Ojima, I., Vyas, D. M., Eds.; ACS Symposium Series
No. 583; American Chemical Society: Washington, D C, 1995. (c) The
Chemistry and Pharmacology of Taxol® and its Derivatives; Farina,
V., Ed.; Elsevier: New York, 1995.
(2) Wani, M. C.; Taylor, H. L.; Wall, M. E.; Coggon, P.; McPhail, A.
T. J . Am. Chem. Soc. 1971, 93, 2325.
(3) (a) Schiff, P. B.; Fant, J .; Horwitz, S. B. Nature 1979, 277, 665.
(b) Schiff, P. B.; Horwitz, S. B. Proc.Natl. Acad. Sci. U.S.A. 1980, 77,
1561.
(4) (a) Chauvie`re, G.; Gue´nard, D.; Picot, F.; Se´nihl, V.; Potier, P.
C. R. Acad. Sci. Paris 1981, 293, 501. (b) Denis, J . N.; Greene, A. E.;
Gue´nard, D.; Gue´ritte-Voegelein, F.; Mangatal, L.; Potier, P. J . Am.
Chem. Soc. 1988, 110, 5917.
(5) For review: Gue´nard, D.; Gue´ritte-Voegelein, F.; Lavelle, F.
Curr. Pharm. Design 1995, 1, 95.
(6) For review: Chen, S. H.; Farina, V. “Paclitaxel (Taxol®) Chem-
istry and Structure-Activity Relationships” In The Chemistry and
Pharmacology of Taxol® and its Derivatives; Farina, V., Ed.; Elsevi-
er: New York, 1995; pp 165-223.
(7) (a) Samaranayake, G.; Magri, N. F.; J itrangsri, C.; Kingston, D.
G. I. J . Org. Chem. 1991, 56, 5114. (b) Menichincheri, M.; Botta, M.;
Ceccarelli, W.; Ciomei, M.; Corelli, F.; D’Anello, M.; Fusar-Bassini, D.;
Mongelli, N.; Pesenti, E.; Pinciroli, V.; Tafi, A.; Vanotti, E. Med. Chem.
Res. 1995, 5, 534.
(8) (a) Marder, R.; Bricard, L.; Dubois, J .; Gue´nard, D.; Gue´ritte-
Voegelein, F. Tetrahedron Lett. 1996, 37, 1777. (b) Datta, A.; J ayas-
inghe, L. R.; Georg, G. I. J . Med. Chem. 1994, 37, 4258. (c) Neidigh,
K. A.; Gharpure, M. M.; Rimoldi, J . M.; Kingston, D. G. I.; J iang, Y.
Q.; Hamel, E. Tetrahedron Lett. 1994, 35, 6839.
S0022-3263(97)00684-1 CCC: $14.00 © 1997 American Chemical Society

6632 J . Org. Chem., Vol. 62, No. 19, 1997
Marder-Karsenti et al.
Sch em e 1
Sch em e 2a
As part of our SAR work, we wish to present herein
the synthesis and antitubulin activity evaluation of 5(20)-
azadocetaxel analogs 5 and 6.
a
Conditions: (i) TESCl/imidazole/DMF/rt, 48 h, 86%; (ii) Red-
Al (12 equiv)/THF/0 °C, 86%; (iii) (CCl3O)2CO/CH2Cl2-pyridine
(85-15)/-15 °C, 92%.
Our strategy (Scheme 1) was based on the work of
Ettouatti et al. for the construction of the oxetane ring.10
For the azetidine ring closure, a C-5R leaving group and
a C-20 amino moiety were needed. This functionality
could be introduced by reductive amination of the C-20
aldehyde obtained by oxidation of the corresponding
alcohol. The C-20 hydroxyl group might proceed from
oxetane ring opening, the C-5R leaving group being
introduced simultaneously as was done by Holton et al.11
in their Taxol total synthesis.
Sch em e 3
The mechanism of the oxetane ring opening catalyzed
by Lewis acids involves the intramolecular participation
of the C-4 acetyl group leading to the formation of an
orthoester between C-4, C-5, and C-20. The hydrolysis
of this orthoester afforded the C-5 and C-20 acetoxy de-
rivatives.12,13 These undesired acetylations can be avoided
by using a C-4 hydroxyl derivative rather than the C-4
acetoxy analog. To avoid intramolecular transacylations
between C-1, C-2, and C-20 oxygen and eventual A-ring
contraction during acid-catalyzed opening of the oxe-
tane,13 the protection by a C-1,C-2 cyclic carbonate was
required. As previously reported in paclitaxel total
syntheses,14 this protective group has additional ad-
vantages: it facilitates the introduction of the C-4 acetyl
function and can be easily transformed to the C-2 ben-
zoate derivative. Because of side reactions during forma-
tion of the cyclic carbonate, all the other hydroxyl groups
have to be protected. Thus, Lewis acid-promoted opening
of the oxetane ring was studied on compound 9. Its
synthesis is depicted in Scheme 2.
Starting from 10-deacetylbaccatin III (2), the C-7, C-10,
and C-13 hydroxyl groups were protected by chlorotri-
ethylsilane with imidazole in DMF. Complete protection
of the three alcohols required long reaction time (48 h to
yield 7 in 86% yield), the C-10 OH being sluggishly
silylated. Though Red-Al was described to selectively
deacylate the C-2 position when the C-13 hydroxyl group
was protected,15 a large excess of Red-Al (12 equiv)
removed both C-2 and C-4 acyl groups yielding 8 in 86%
yield. Similar results have been obtained by Georg et
al. on paclitaxel with tBuOK and KOH as deacylating
reagents.16 Formation of the C-1,C-2 cyclic carbonate was
easily achieved by addition of triphosgene to yield 9 in
92% yield.
Results on the opening of the oxetane ring promoted
by Lewis acid (Scheme 3) are summarized in Table 1.
Bromotrimethylsilane was described to open cyclic ethers
and to introduce a bromide and a trimethylsilyl ether.17
In our hands, no additional silyl ether was observed. On
the contrary, partial desilylation at C-13 was always
observed along with oxetane ring opening (entries 1-3)
likely due to the relative acidity of the reaction mixture.
Boron trifluoride etherate/iodide ion was described as a
mild ether-cleaving reagent.18 In order to correlate with
the previous experiments using bromotrimethylsilane,
bromide was used instead of iodide. Among the different
(9) Fenoglio, I.; Nano, G. M.; Vander Velde, D. G.; Appendino, G.
Tetrahedron Lett. 1996, 37, 3203.
(10) Ettouati, L.; Ahond, A.; Poupat, C.; Potier, P. Tetrahedron 1991,
47, 9823.
(11) Holton, R. A.; Kim, H.-B.; Somoza, C.; Liang, F.; Biediger, R.
J .; Boatman, P. D.; Shindo, M.; Smith, C. C.; Kim, S.; Nadizadeh, H.;
Suzuki, Y.; Tao, C.; Vu, P.; Tang, S.; Zhang, P.; Murthi, K. K.; Gentile,
L. N.; Liu, J . H. J . Am. Chem. Soc. 1994, 116, 1599.
(12) Chen, S. H.; Huang, S.; Wei, J .; Farina, V. Tetrahedron 1993,
49, 2805.
(13) (a) Wahl, A.; Gue´ritte-Voegelein, F.; Gue´nard, D.; Le Goff, M.-
T.; Potier, P. Tetrahedron 1992, 48, 6965. (b) Gue´ritte-Voegelein, F.;
Gue´nard, D.; Dubois, J .; Wahl, A.; Marder, R.; Mu¨ller, R.; Lund, M.;
Bricard, L.; Potier, P. In Taxane Anticancer Agents: Basic Science and
Current Status; Georg, I. G., Chen, T. T., Ojima, I., Vyas, D. M., Eds.;
ACS Symposium Series No. 583; American Chemical Society: Wash-
ington, DC, 1995; pp 189-202 and references therein.
(14) (a) Nicolaou, K. C.; Yang, Z.; Liu, J . J .; Ueno, H.; Nanternet,
P. G.; Guy, R. K.; Claiborne, C. F.; Renaud, J .; Couladouros, E. A.;
Paulvannan, K.; Sorensen, E. J . Nature 1994, 367, 630. (b) Holton,
R. A.; Somoza, C.; Kim, H.-B.; Liang, F.; Biediger, R. J .; Boatman, P.
D.; Shindo, M.; Smith, C. C.; Kim, S.; Nadizadeh, H.; Suzuki, Y.; Tao,
C.; Vu, P.; Tang, S.; Zhang, P.; Murthi, K. K.; Gentile, L. N.; Liu, J . H.
J . Am. Chem. Soc. 1994, 116, 1597. (c) Danishefsky, S. J .; Masters, J .
J .; Young, W. B.; Link, J . T.; Snyder, L. B.; Magee, T. V.; J ung, D. K.;
Isaacs, R. C. A.; Bornmann, W. G.; Alaimo, C. A.; Coburn, C. A.; Di
Grandi, M. J . J . Am. Chem. Soc. 1996, 118, 2843.
(15) Chen, S. H.; Farina, V.; Wei, J . M.; Long, B.; Fairchild, C.;
Mamber, S. W.; Kadow, J . F.; Vyas, D.; Doyle, T. W. Bioorg. Med. Chem.
Lett. 1994, 4, 479.
(16) Georg, G. I.; Ali, S. M.; Boge, T. C.; Datta, A.; Falborg, L.
Tetrahedron Lett. 1994, 48, 8931.
(17) Kriecheldorf, H. R.; Maber, G.; Regel, W. Synthesis 1981,
383.

D-Ring-Modified Taxanes
J . Org. Chem., Vol. 62, No. 19, 1997 6633
Sch em e 4a
Ta ble 1. Lew is Acid P r om oted Oxeta n e Rin g Op en in g
products (%)
entry
conditionsa
9
10
11
12
1
2
3
4
TMSBr (1 equiv), 0 °C, 15 min
TMSBr (1 equiv), 0 °C, 35 min
TMSBr (5 equiv), rt, 40 min
BF3‚OEt2 (0.9 equiv),
Et4NBr (3 equiv), rt, 3 min
BF3‚OEt2 (1.8 equiv),
Et4NBr (3 equiv), rt, 5 min
SnCl4 (1 equiv), Et4NBr
(1.5 equiv), rt, 2 h
BBr3 (1 equiv), Et4NBr
(3 equiv), 0 °C, 5 min
18
20
17
40
57
15
25
b
30
35
23
5
6
7
8
40
66
9
b
b
BF3‚OEt2 (1 equiv),
NaBr (3.5 equiv), rt, 5 min
a
b
All experiments were conducted in dry CH2Cl2. Formation
of very polar compounds not identified.
bromide ions (entries 4 and 8) and Lewis acids (entries
4, 6, and 7) used, the best results were obtained with
Et4NBr and boron trifluoride etherate. Greater amounts
of this later reagent increased desilylation (entry 5), and
variation in the amount of bromide ion had little influ-
ence on the opening rate (data not shown). Increasing
reaction times only allowed additional removal of silyl
ethers. This side reaction could not be avoided by lower
temperature but was minimized by using dry methylene
chloride as the solvent for this reaction. Acetonitrile,
ethyl ether, and DMF gave predominantly desilylated
compounds (data not shown).
a
Conditions: (i) TFAA/DMSO in toluene then Et3N, -60 °C,
13 (82%); (ii) BnNH2 (4 equiv), ZnCl2 (1 equiv) in CH3CN, rt, 14
(27% isolated yield); (iii) NaBH3CN (3.5 equiv), AcOH, MeOH, 0
°C, 15 (61%); (iv) BnNH2 (4 equiv), ZnCl2 (1.1 equiv) in CH3CN,
rt then NaBH3CN (3.5 equiv), AcOH, 0 °C; (v) CH2Cl2, rt, 16 (56%
from 13).
triethylsilyl ethers and the cyclic carbonate. Transfor-
mation of the C-13 hydroxyl moiety into a ketone seemed
to us of valuable interest, since it could be easily reduced
to the starting hydroxyl group22 and would not trouble
acetylation.
Under those different conditions, the oxetane ring was
always opened with the same regioselectivity, with the
bromide being introduced at the C-5 R-position. In the
literature both reagents (TMSBr or BF3‚OEt2/Et4NBr)
were known to yield the less hindered bromide.17,18 Our
results, as well as those of Holton with TMSCl,11 could
be explained by the steric hindrance at the C-20 position
due to the presence of the C-4 hydroxyl group on the
R-face and the C-19 methyl group on the â-face.
Two derivatives bearing a C-5R leaving group and a
free C-20 hydroxyl group (10 and 11) were thus available.
Though the subsequent steps of the synthesis, i.e.,
oxidation, reductive amination, and azetidine ring forma-
tion, have been realized on compound 1019 and optimized
to 54% overall yield, further acetylation and selective
removal of the C-13 protective group appeared trouble-
some. Indeed, the C-4 OH was known to be hard to
acylate20 especially when a bulky group such as trieth-
ylsilyl was in the C-13 position. Moreover, preliminary
studies on 7,10,13-tris(triethylsilyl) derivatives showed
that the C-13 protective group could best be removed by
HF/pyridine complex. Because of the sensitivity of the
azetidine ring toward mineral acids,21 starting from the
triprotected derivative 10 did not appear suitable and we
chose to continue the synthesis on compound 11.
Scheme 4 summarizes the synthetic route to the
azetidine derivative 16. Dess-Martin periodinane23 first
gave the desired compound, but the reaction was not
reproducible and yields were only moderate (no more
than 50%). On the contrary to results of Fenoglio et al.,9
Swern oxidation was totally ineffective. Finally, com-
pound 13 was obtained in good yield using trifluoroacetic
anhydride-activated DMSO24 in toluene, both C-13 and
C-20 alcohol being oxidized simultaneously. Reductive
amination on 13 was achieved with benzylamine and
NaBH3CN under acidic conditions. This reaction could
not be realized in one step as usually described.25 The
imine 14 was obtained in acetonitrile with an excess of
amine, and the yield was improved by addition of ZnCl2
(it would serve as dehydrating agent as it was described
with TiCl4 in the same type of reaction26). The C-13
ketone was unreactive toward benzylamine at room
temperature. This was predictable from recent results
on the oximation of this ketone that could only be realized
at elevated temperature.20b,27 Compound 14 was isolated
before reduction, but amine 15 was obtained in a better
yield by direct reduction of the reaction mixture with
sodium cyanoborohydride under acidic conditions. The
azetidine ring closure occurred spontaneously in meth-
ylene chloride at room temperature. Thus, compound 15
led to the aza derivative 16. The overall yield of this
synthetic pathway could only be improved by carrying
We needed a small C-13 protective group that would
be removed under mild conditions in the presence of the
(18) (a) Mandal, A. K.; Soni, N. R.; Ratnam, K. R. Synthesis 1985,
274. (b) Yadav, V. K.; Fallis, A. G. J . Am. Chem. Soc. 1986, 51,
3372.
(19) Marder-Karsenti, R. Re´activite´ chimique des taxo¨ıdes: Modi-
fications structurales des cycles A et D Ph.D. Dissertation, Universite´
Paris VI, Gif-sur-Yvette, France, 1996.
(22) Se´nihl, V.; Gue´ritte, F.; Gue´nard, D.; Colin, M.; Potier, P. C.
R. Acad. Sci. Paris 1984, 299, 1039.
(23) (a) Dess, D. B.; Martin, J . C. J . Org. Chem. 1983, 48, 4155. (b)
Dess, D. B.; Martin, J . C. J . Am. Chem. Soc. 1991, 113, 7277.
(24) Manusco, A. J .; Swern, D. Synthesis 1981, 165 and references
therein.
(20) (a) Samaranayake, G.; Neidigh, K. A.; Kingston, D. G. I. J . Nat.
Prod. 1993, 56, 884. (b) Chen, S. H.; Farina, V.; Vyas, D. M.; Doyle,
T. W. J . Org. Chem. 1996, 61, 2065.
(21) Davies, D. E.; Storr, R. C. Azetidines, Azetines and Azetes. In
Comprehensive Heterocyclic Chemistry; Katritsky, A. P., Rees, C. W.,
Eds.; Pergamon Press: New York, 1984; Vol. 7, pp 237-243.
(25) For example, see review: Lane, C. F. Synthesis 1975, 135.
(26) Weingarten, H.; Chupp, J . P.; White, W. A. J . Org. Chem. 1967,
32, 3246.
(27) Menichincheri, M.; Ceccarelli, W.; Fusar-Bassini, D.; Vanotti,
E.; Ciomei, M.; Mongelli, N.; Pinciroli, V.; Vulpetti, A. Med. Chem. Res.
1996, 6, 264.

6634 J . Org. Chem., Vol. 62, No. 19, 1997
Marder-Karsenti et al.
Ta ble 2. Resu lts of Biologica l Eva lu a tion of
5(20)-Aza d oceta xel An a logs
Sch em e 5
microtubule disassembly
inhibitory activitya
IC50/IC50(paclitaxel)
cytotoxicity against
KB cell lineb
IC50 (nM)
compd
1a
1b
5
1
0.5
6
3.2
inactive
inactive
inactivec
8
6
a
IC50 is the concentration required to inhibit 50% of the rate
of microtubule disassembly. The ratio IC50/IC50(paclitaxel) gives
b
the activity with respect to paclitaxel. IC50 measures the drug
concentration required for the inhibition of 50% cell proliferation
after 72 h incubation. c Only 15% inhibition at 75 µM.
to note that molecular modeling studies and examination
of related examples from the Cambridge Data Bank show
that the azetidine ring induces almost the same con-
straint on the C-ring than the oxetane, even if its
structural parameters are very slightly different. More-
over, the replacement of an oxygen by a nitrogen should
only bring small electronic modifications in this area of
the molecule. Indeed, the pyramidal N-inversion can
easily take place, the two isomers being very similar in
energy as shown by molecular modeling analysis. Thus,
in order to explain the large difference of activity between
docetaxel (2) and its aza analog 6, we can suppose the
involvement of a specific interaction of the heteroatom
with the protein, probably troubled by the presence of
the hydrogen of the amine. The total inactivity of
compound 5 supports the hypothesis of a sterically
restricted binding site for this part of the molecule. But,
an ether has been replaced by an amine, a more basic
function likely protonated at biological pH, so we cannot
rule out the hypothesis of a charge repulsion at the
binding site. Preliminary studies on the N-acetyl deriva-
tive of 6, a nonprotonable amide devoid of antitubulin
activity (data not shown), are not in favor of this latest
hypothesis. Additional studies of other small amide
derivatives are under current investigation and will be
reported in due time.
In summary, 5(20)-aza derivatives of docetaxel (5 and
6) have been prepared from 10-deacetylbaccatin III.
These are the first tetracyclic docetaxel analogs bearing
a modified D-ring. The key steps of this synthesis were
the direct introduction of a C-5 leaving group while the
ring was opened and the intramolecular nucleophilic
attack of the C-20 amino group at C-5. This strategy can
be applied to the preparation of other modified D-ring
analogs that could also help to determine the precise role
of the oxetane ring in the interaction of taxoids with
microtubules.
a
Conditions: (i) DMAP (20 equiv), Ac2O (25 equiv), CH2Cl2, rt,
17 (71%); (ii) PhLi (5 equiv), THF, -72 °C, 18 (91%); (iii)
NaBH4 (14 equiv), EtOH/THF (1/3), rt, 19 (74%); (iv) 20 (3
equiv), DCO (3 equiv), DMAP (1.5 equiv), toluene, rt, 21 (99%);
(v) PTSA, (3 equiv), MeOH, rt, 5 (72%); (vi) H2, Pd/C, MeOH, rt,
6 (61%).
out the reaction without isolating neither the imine 14
nor the amine 15.
Afterwards, the following steps of the synthesis (Scheme
5) were uneventful. Acetylation of compound 16 was
achieved at room temperature, affording 17 in 70% yield.
As previously described,28 the C-1,C-2 carbonate was
readily opened by phenyllithium in excellent yield (18,
90%) without affecting the C-13 ketone. This function
was then reduced with NaBH4 to afford the C-13R isomer
19 in 70% yield. Esterification was realized with the 2-(4-
OMe)phenyl-1,3-oxazolidine of N-Boc-phenylisoserine
(20),29 DCC, and DMAP in toluene at room temperature
in excellent yield (21, 99%). Deprotection of 21 with
PTSA in MeOH afforded 5, which was fully deprotected
by hydrogenolysis to afford 6, the azetidine isostere of
the oxetane ring of docetaxel.
The activity of compounds 5 and 6 was evaluated in
vitro on the disassembly process of microtubules into
tubulin30 and on the cytotoxicity against KB cell line.31
As shown in Table 2, the replacement of the oxetane
oxygen atom by a nitrogen greatly affects the interaction
with microtubules. Compound 6 is 16 times less active
than docetaxel, and compound 5 is inactive. It is worthy
Exp er im en ta l Section
Gen er a l Meth od s. General methods were the same as
previously described.32 Standard workup means extraction
with a suitable solvant (CH2Cl2 unless otherwise specified),
washing the extract with H2O or brine, drying over Na2SO4,
and evaporation in vacuo. 10-Deacetylbaccatin III was ex-
tracted from T. baccata leaves,4a and the docetaxel acid side
chain 22 was a gift from Rhoˆne-Poulenc Rorer S.A. Microtu-
bular proteins were purified from bovine brain as previously
described.33 Molecular modeling studies were performed using
MacroModel on a Silicon Graphics 4D/35 workstation.
(28) (a) Nicolaou, K. C.; Couladouros, E. A.; Nantermet, P. G.;
Renaud, J .; Guy, R. K.; Wrasidlo, W. Angew. Chem., Int. Ed. Engl.
1994, 33, 1581. (b) Nicolaou, K. C.; Renaud, J .; Nantermet, P. G.;
Couladouros, E. A.; Guy, R. K.; Wrasidlo, W. J . Am. Chem. Soc. 1995,
117, 2409.
(29) Didier, E.; Fouque, E.; Taillepied, I.; Commerc¸on, A. Tetrahe-
dron Lett. 1994, 35, 2349.
(30) Lataste, H.; Se´nihl, V.; Wright, M.; Gue´nard, D.; Potier, P. Proc.
Natl. Acad. Sci. U.S.A. 1984, 81, 4090.
(31) Da Silva, A. D.; Machado, A. S.; Tempeˆte, C.; Robert-Gero, M.
Eur. J . Med. Chem. 1994, 29, 149.
(32) Marder, R.; Dubois, J .; Gue´nard, D.; Gue´ritte-Voegelein, F.;
Potier, P. Tetrahedron 1995, 51, 1985.
(33) (33) ) Dubois, J .; Le Goff, M. T.; Gue´ritte-Voegelein, F.; Gue´nard,
D.; Tollon, Y.; Wright, M. Bioorg. Med. Chem. 1995, 3, 1357.

D-Ring-Modified Taxanes
J . Org. Chem., Vol. 62, No. 19, 1997 6635
P r ep a r a tion of Com p ou n d 7. To a solution of 2 (4.6 g,
8.5 mmol) in 13 mL of dry DMF was added imidazole (1.85 g,
27 mmol, 3.2 equiv). Then triethylsilyl chloride (7 mL, 42.5
mmol, 5 equiv) was added dropwise at room temperature. The
solution was stirred for 48 h. After removal of the solvant,
100 mL of AcOEt and 75 mL of water were added. After
standard workup the residue was purified on silica gel (hep-
tane/AcOEt: 95/5) to yield pure 7 (6.5 g, 86%): 1H NMR (250
MHz, CDCl3) δ 0.62 (18H, m), 1.01 (27H, m), 1.12 (3H, s), 1.20
(3H, s), 1.65 (3H, s), 1.80 (1H, m), 1.99 (3H, s), 2.17 (2H, m,
2H), 2.29 (3H, s), 2.51 (1H, m), 3.87 (1H, d, J ) 7 Hz), 4.14
(1H, d, J ) 8 Hz), 4.30 (1H, d, J ) 8 Hz), 4.42 (1H, dd, J ) 11
Hz, J ′ ) 7 Hz), 4.93 (1H, m), 4.95 (1H, m), 5.20 (1H, s), 5.62
(1H, d, J ) 7 Hz), 7.45, 7.59, 8.10 (5H, m); 13C NMR (75 MHz,
CDCl3) δ 5.0, 5.4, 5.9, 7.0, 7.1, 10.8, 14.8, 20.9, 22.6, 26.6, 37.6,
40.1, 43.2, 47.0, 60.0, 68.5, 72.8, 75.6, 75.9, 76.8, 79.7, 80.9,
84.2, 128.7, 129.8, 130.2, 133.6, 136.6, 139.6, 167.3, 170.1,
205.8; IR (CHCl3) 1725 cm-1; UV (EtOH) λmax (ꢀ) 203.2 (1727),
214 (1727), 240.4 (6933), 270.4 (1550) nm; MS (FAB+) m/z 909
(MNa+).
P r ep a r a tion of Com p ou n d 8. To a THF solution (50 mL)
of 7 (5 g, 5.6 mmol) was added Red-Al (3.5 M) in toluene (21
mL, 73.5 mmol, 13 equiv) at 0 °C. The solution was stirred at
0 °C for 25 min, and the reaction was stopped by careful
addition of 15 mL of a saturated potassium-sodium tartrate
solution. After standard workup with AcOEt the residue
was purified by silica gel chromatography (cyclohexane/
AcOEt: 70/30) to afford 8 (3.6 g, 86%): 1H NMR (250 MHz,
CDCl3) δ 0.71 (18H, m), 1.03 (27H, m), 1.11 (3H, s), 1.56 (3H,
s), 1.92 (3H, s), 2.02 (1H, m), 2.18 (1H, m), 2.48 (1H, m), 2.50
(1H, m), 3.26 (1H, s), 3.28 (1H, d, J ) 6 Hz), 3.48 (1H, d,
J ) 11 Hz), 3.78 (1H, dd, J ) 11 Hz, J ′ ) 6 Hz), 4.02 (1H, dd,
J ) 11 Hz, J ′ ) 6 Hz), 4.19 (1H, s), 4.40 (1H, d, J ) 8 Hz),
4.62 (1H, m), 4.67 (1H, dd, J ) 9.5 Hz, J ′ ) 4 Hz), 4.75 (1H,
d, J ) 8 Hz), 5.26 (1H, s); 13C NMR (75 MHz, CDCl3) δ 5.3,
5.8, 6.6, 7.4, 7.5, 10.6, 17.9, 18.4, 30.6, 38.4, 39.2, 41.8, 51.8,
59.6, 70.8, 74.0, 74.2, 76.4, 77.1, 77.6, 80.5, 86.8, 135.7, 141.3,
206.6; IR (CHCl3) 3493, 1712, 1593 cm-1; MS (FAB+) m/z 763
(MNa+).
P r ep a r a tion of Com p ou n d 9. A solution of 8 (1.18 g, 1.6
mmol) in 20 mL of a mixture of CH2Cl2/pyridine (85/15) was
cooled to -15 °C. Then triphosgene (0.85 g, 2.9 mmol, 1.8
equiv) was added, and the solution was stirred for 15 min at
-15 °C. The reaction was stopped by addition of a saturated
sodium bicarbonate solution. After standard workup, the
residue was purified on silica gel (heptane/AcOEt: 85/15) to
yield pure 9 (1 g, 92%): 1H NMR (300 MHz, CDCl3) δ 0.61
(18H, m), 0.98 (27H, m), 1.03 (3H, s), 1.10 (3H, s), 1.55 (3H,
s), 1.91 (4H, m), 2.43 (2H, m), 2.57 (1H, m), 3.02 (1H, d, J )
5 Hz), 4.07 (1H, dd, J ) 11 Hz, J ′ ) 7 Hz), 4.25 (1H, d, J ) 5
Hz), 4.42 (1H, d, J ) 8 Hz), 4.48 (1H, d, J ) 8 Hz), 4.53 (1H,
m), 4.70 (1H, dd, J ) 10 Hz, J ′ ) 1 Hz), 5.14 (1H, s); 13C NMR
(75 MHz, CDCl3) δ 4.5, 5.7, 6.7, 6.7, 9.9, 17.3, 18.5, 28.1, 35.7,
37.8, 40.3, 18.7, 60.3, 68.9, 73.4, 73.5, 77.3, 78.5, 80.4, 87.3,
89.8, 137.2, 139.3, 153.2, 206.1; IR (CHCl3) 3700, 1800, 1712,
1600 cm-1; MS (FAB+) m/z 789 (MNa+).
91.0, 135.0, 142.8, 153.4, 205.7; IR (CHCl3) 3495, 1798, 1718
cm-1; MS (FAB+) m/z 869-871 (MNa+); HRMS calcd for
C39H71O9Si3BrNa (MNa+) 869.4039, found 869.3455.
Com p ou n d 11: 1H NMR (300 MHz, CDCl3) δ 0.49, 0.55
(12H, m), 0.90 (18H, m), 1.05 (3H, s), 1.09 (3H, s), 1.13 (3H,
s), 2.14 (1H, m), 2.23 (3H, s), 2.32 (1H, m), 2.59 (1H, dd, J )
16 Hz, J ′ ) 9 Hz), 2.88 (1H, dd, J ) 16 Hz, J ′ ) 4 Hz), 3.01
(1H, bs), 3.59 (1H, d, J ) 11 Hz), 3.69 (1H, d, J ) 5 Hz), 3.81
(1H, s), 4.11 (1H, d, J ) 11 Hz), 4.21 (1H, d, J ) 5 Hz), 4.41
(1H, dd, J ) 11 Hz, J ′ )4 Hz), 4.65 (2H, m), 5.28 (1H, s); 13C
NMR (75 MHz, CDCl3) δ 5.1, 5.8, 6.9, 13.2, 17.1, 19.1, 27.4,
36.1, 37.2, 40.8, 45.8, 61.2, 61.9, 62.7, 68.4, 70.0, 74.0, 76.9,
81.9, 91.6, 135.8, 142.5, 154.0, 205.7; IR (CHCl3) 3518, 1787,
1706 cm-1; MS (FAB+) m/z 755-757 (MNa+), 677 (M - Br +
Na+).
Com p ou n d 12: 1H NMR (300 MHz, CDCl3) δ 0.61 (12H,
m), 0.98 (18H, m), 1.12 (3H, s), 1.18 (3H, s), 1.60 (3H, s), 1.96
(1H, m), 2.05 (3H, s), 2.53 (1H, m), 2.62 (2H, m), 3.16 (1H, d,
J ) 5 Hz), 3.49 (1H, s), 4.15 (1H, dd, J ) 11 Hz, J ′ ) 7 Hz),
4.39 (1H, d, J ) 5 Hz), 4.50 (1H, d, J ) 8 Hz), 4.62 (1H, d, J
) 8 Hz), 4.62 (1H, m), 4.85 (1H, dd, J ) 10 Hz, J ′ ) 1 Hz),
5.21 (1H, s); 13C NMR (75 MHz, CDCl3) δ 5.3, 6.1, 7.0, 7.2,
10.4, 17.4, 18.9, 28.5, 35.3, 38.0, 40.7, 48.2, 60.4, 68.2, 72.7,
74.5, 77.8, 79.7, 80.9, 87.9, 91.2, 137.0, 140.8, 154.1, 206.5; IR
(CHCl3) 3362, 1806, 1725 cm-1 MS (CI) m/z 653 (M + H)+,
635 (M - H2O + H)+.
P r ep a r a tion of Com p ou n d 13. A dry toluene solution (6
mL) of DMSO (1.22 mL, 17 mmol) was added dropwise at -70
°C to a toluene solution (6 mL) of TFAA (1.8 mL, 13 mmol).
After stirring at -65 °C for 15 min, a toluene solution (12 mL)
of 11 (633 mg, 0.86 mmol) was added dropwise, and the
solution was stirred between -60 and -50 °C for 45 min. Then
Et3N (2.4 mL, 17 mmol) was added dropwise, and the solution
was stirred below -40 °C for 15 min. The reaction was stopped
by addition of water (5 mL), and the solution was allowed to
warm up. After standard extraction the residue was purified
on silica gel (heptane/AcOEt: 82/18) to yield 13 (520 mg,
82%): 1H NMR (300 MHz, CDCl3) δ 0.57 (12H, m), 0.90 (18H,
m), 1.13 (3H, s), 1.19 (3H, s), 1.28 (3H, s), 2.22 (3H, s), 2.28
(2H, m), 2.66 (1H, d, J ) 19 Hz), 3.25 (1H, s), 3.61 (1H, d, J )
19 Hz), 3.62 (1H, d, J ) 5 Hz), 4.40 (1H, dd, J ) 11Hz, J ′ ) 4
Hz), 4.43 (1H, d, J ) 5 Hz), 4.81 (1H, t, J ) 3 Hz), 5.41 (1H,
s), 9.78 (1H, s); 13C NMR (75 MHz, CDCl3) δ 5.4, 6.1, 7.1, 7.2,
13.3, 15.4, 19.0, 32.2, 37.6, 41.2, 42.5, 45.3, 56.7, 62.4, 70.3,
77.5, 78.1, 80.8, 89.2, 139.7, 151.9, 155.2, 195.8, 197.4, 204.0;
IR (CHCl3) 1806, 1720, 1685 cm-1; MS (LSIMS+) m/z 751-
753 (MNa+).
P r ep a r a tion of Com p ou n d 14. To a CH3CN solution (2
mL) of 13 (59 mg, 81 µmol) were added ZnCl2 (11 mg, 81 µmol)
and benzylamine (35 µL, 32 µmol). The solution was stirred
at room temperature for 6 h. After standard workup, the
residue was purified on preparative TLC (CH2Cl2/acetone: 97/
3) to yield 14 (18 mg, 27%): 1H NMR (300 MHz, CDCl3) δ 0.56
(6H, m), 0.65 (6H, m), 0.94 (18H, m), 1.17 (3H, s), 1.26 (3H, s),
1.38 (3H, s), 2.31 (3H, s), 2.31 (2H, m), 2.89 (1H, d, J ) 19
Hz), 3.71 (1H, d, J ) 5 Hz), 3.84 (1H, d, J ) 19 Hz), 4.42 (1H,
d, J ) 5 Hz), 4.50 (1H, dd, J ) 4 Hz, J ′ ) 10 Hz), 4.68 (2H,
qAB, J ) 14 Hz), 5.10 (1H, t, J ) 3 Hz), 5.48 (1H, s), 7.32 (5H,
m), 8.16 (1H, s); 13C NMR (75 MHz, CDCl3) δ 5.1, 5.9, 6.9, 13.2,
15.1, 18.6, 31.9, 37.4, 41.0, 42.2, 44.9, 59.8, 62.6, 64.2, 70.4,
74.7, 77.6, 80.9, 88.9, 127.6, 128.1, 128.8, 137.5, 139.4, 152.5,
154.7, 163.3, 194.4, 204.3; IR (CHCl3) 1810, 1717, 1685, 1605
cm-1; MS (LSIMS+) m/z 818-820 (MH+).
P r ep a r a tion of Com p ou n d 15. To a CH3CN solution (3
mL) of 14 (49 mg, 60 µmol) were added AcOH and NaBH3CN
(12.5 mg, 0.2 mmol, 3.3 equiv). The solution was stirred at 0
°C for 2 h and then hydrolyzed with saturated NaHCO3. After
standard workup, the residue was purified by preparative TLC
(CH2Cl2/acetone: 97/3) to yield 15 (20 mg, 40%) along with 16
(9.7 mg, 22%): 1H NMR (250 MHz, CDCl3) δ 0.53 (6H, m),
0.64 (6H, m, 6H), 0.98 (18H, m), 1.22 (6H, s), 1.31 (3H, s), 2.08
(1H, m), 2.23 (1H, m), 2.27 (3H, s), 2.62 (1H, d, J ) 13 Hz),
2.80 (1H, d, J ) 19 Hz), 3.25 (1H, d, J ) 13 Hz), 3.66 (1H, d,
J ) 5 Hz), 3.85 (2H, qAB, J ) 13 Hz), 4.03 (1H, t, J ) 2 Hz),
4.14 (1H, d, J ) 19 Hz), 4.30 (1H, d, J ) 5 Hz), 4.46 (1H, dd,
J ) 4 Hz, J ′ ) 10 Hz), 5.44 (1H, s), 7.29 (5H, m); 13C NMR
Oxeta n e Rin g Op en in g. Experimental conditions are
described in Table 1. Dry CH2Cl2 and freshly distilled BF3‚
OEt2 were required. As a typical procedure entry 5 is
described:
To a solution of 9 (155 mg, 0.2 mmol) in 10 mL dry of CH2Cl2
were added Et4NBr (127 mg, 0.6 mmol, 3 equiv) and BF3‚OEt2
(22 µL, 0.18 mmol, 0.9 equiv). The solution was stirred at room
temperature for 3 min and then hydrolyzed. After standard
workup, the residue was purified on preparative TLC (hep-
tane/AcOEt/MeOH: 50/50/2) to yield pure 10 (60 mg, 35%)
along with 11 (37 mg, 25%) and 12 (40 mg, 30%).
Com p ou n d 10: 1H NMR (300 MHz, CDCl3) δ 0.59 (18H,
m), 0.94 (27H, m), 1.08 (3H, s), 1.12 (3H, s), 1.17 (3H, s), 2.08
(3H, s), 2.14 (1H, m), 2.26 (1H, m), 2.34 (1H, dd, J ) 15 Hz, J ′
) 9 Hz), 2.88 (1H, dd, J ) 15 Hz, J ′ ) 5 Hz), 2.95 (1H, s), 3.58
(1H, d, J ) 12 Hz), 3.68 (1H, d, J ) 5 Hz), 4.18 (1H, d, J ) 12
Hz), 4.21 (1H, d, J ) 5 Hz), 4.39 (1H, dd, J ) 10 Hz, J ′ ) 4
Hz), 4.80 (1H, m), 4.88 (1H, bs), 5.28 (1H, s); 13C NMR (75
MHz, CDCl3) δ 5.0, 5.2, 5.8, 6.9, 7.0, 13.3, 17.1, 19.5, 26.8, 37.1,
37.4, 40.9, 46.5, 61.4, 62.3, 63.2, 68.4, 70.0, 74.9, 76.7, 82.1,

6636 J . Org. Chem., Vol. 62, No. 19, 1997
Marder-Karsenti et al.
(62.5 MHz, CDCl3) δ 5.1, 5.8, 6.8, 6.9, 13.5, 15.0, 18.7, 31.7,
36.9, 41.1, 42.0, 44.2, 51.4, 54.3, 62.7, 63.0, 70.1, 71.1, 77.7,
81.1, 88.9, 127.9, 128.4, 128.8, 138.6, 139.1, 153.0, 154.1, 197.7,
204.7; IR (CHCl3) 3402, 1802, 1715, 1682 cm-1; MS (LSIMS+)
m/z 820-822 (MH+), 740 (MH+ of 16).
(6H, q, J ) 7 Hz), 0.75 (6H, m), 0.87 (9H, t, J ) 7 Hz), 1.01
(9H, t, J ) 7 Hz), 1.05 (3H, s), 1.18 (3H, s), 1.48 (1H, m), 1.80
(3H, s), 1.98 (3H, s), 2.15 (1H, m), 2.23 (2H, m), 2.25 (3H, s),
2.90 (1H, d, J ) 11 Hz), 3.48 (1H, bd, J ) 9 Hz), 3.52 (1H, d,
J ) 11 Hz), 3.52 (1H, d, J ) 13 Hz), 3.65 (1H, d, J ) 13 Hz),
3.77 (1H, d, J ) 7 Hz), 4.36 (1H, dd, J ) 7 Hz, J ′ ) 10 Hz),
4.81 (1H, m, 1H), 5.18 (1H, s), 5.59 (1H, d, J ) 7 Hz), 7.18
(5H, m), 7.49 (2H, t, J ) 7 Hz), 7.60 (1H, t, J ) 7 Hz), 8.15
(2H, t, J ) 7 Hz); 13C NMR (75 MHz, CDCl3) δ 5.7, 6.5, 7.4,
7.4, 10.5, 15.1, 20.0, 23.6, 27.2, 38.4, 39.1, 43.1, 48.1, 59.4,
62.3, 63.0, 68.5, 68.8, 74.3, 75.8, 76.3, 79.5, 79.8, 127.5, 128.6,
128.9, 129.2, 130.5, 130.6, 133.7, 137.1, 138.0, 138.3, 167.8,
171.2, 206.9; IR (CHCl3) 1718 cm-1; MS (LSIMS+) m/z 862
(MH+).
P r ep a r a tion of Com p ou n d 21. To a toluene solution (4
mL) of 19 (30 mg, 35 µmol) were added 4-DMAP (6.5 mg, 53
µmol, 1.5 equiv), DCC (21.3 mg, 103 µmol, 3 equiv), and 20
(45 mg, 113 µmol, 3.2 equiv). The solution was stirred for 40
min at room temperature. After standard workup the resi-
due was purified on preparative TLC (heptane/AcOEt: 70/30)
to yield 21 (43 mg, 99%): 1H NMR (250 MHz, CDCl3) δ 0.42
(6H, q, J ) 7 Hz), 0.60 (6H, m), 0.87 (9H, t, J ) 7 Hz), 0.98
(9H, m), 1.06 (9H, s), 1.16 (3H, s), 1.18 (3H, s), 1.46 (4H, m),
1.61 (3H, s), 1.73 (3H, s), 2.09 (3H, m), 2.81 (1H, d, J ) 11
Hz), 3.29 (1H, bd, J ) 9 Hz), 3.35 (1H, d, J ) 11 Hz), 3.36
(1H, d, J ) 13 Hz), 3.57 (1H, d, J ) 7 Hz), 3.60 (1H, d, J ) 13
Hz), 3.81 (3H, s), 4.23 (1H, dd, J ) 7, 10 Hz), 4.60 (1H, d,
J ) 5 Hz), 5.04 (1H, s), 5.40 (1H, bd, J ) 5 Hz), 5.58 (1H, d,
J ) 7 Hz), 6.09 (1H, m), 6.42 (1H, m), 6.92 (2H, d, J ) 9
Hz), 7.14 (2H, m), 7.42 (10H, m), 7.50 (2H, t, J ) 7 Hz), 7.62
(1H, t, J ) 7 Hz), 8.05 (2H, d, J ) 7 Hz); 13C NMR (75 MHz,
CDCl3) δ 5.3, 6.2, 7.0, 7.2, 10.2, 13.4, 20.5, 22.1, 26.5, 27.9,
35.5, 37.9, 43.1, 47.4, 55.3, 58.7, 61.8, 62.6, 63.9, 68.4, 71.9,
73.6, 75.2, 75.6, 79.1, 79.5, 80.9, 83.6, 92.7, 113.9, 126.6, 127.1,
128.0, 128.2, 128.3, 128.5, 128.8, 129.0, 130.0, 130.2, 133.4,
134.1, 137.4, 137.8, 137.39, 137.8, 160.5, 167.2, 169.6, 169.7,
205.9; IR (CHCl3) 1717, 1707 cm-1; MS (LSIMS+) m/z 1243
(MH+).
P r ep a r a tion of Com p ou n d 16 fr om 13. To a CH3CN
solution (5 mL) of 13 (108 mg, 0.15 mmol) were added
benzylamine (70 µL, 0.64 mmol, 4 equiv) and zinc chloride
(21.6 mg, 0.16 mmol, 1.1 equiv). The solution was stirred
overnight at room temperature. The solution was cooled to 0
°C, and 0.9 mL of acetic acid and NaBH3CN (29.3 mg, 0.47
mmol, 3.1 equiv) were added. The solution was stirred at 0
°C for 30 min and then hydrolyzed by saturated NaHCO3. After
standard workup, the residue was allowed to stand at room
temperature in CH2Cl2 for 2 days and then was purified on
silica gel (CH2Cl2/acetone: 97/3) to yield 16 (62 mg, 56%): 1H
NMR (300 MHz, CDCl3) δ 0.47 (6H, m), 0.65 (6H, m), 0.95
(18H, m), 1.26 (3H, s), 1.31 (3H, s), 1.62 (1H, m), 1.79 (3H, s),
2.04 (3H, s), 2.17 (1H, s), 2.62 (1H, d, J ) 5 Hz), 2.78 (1H, d,
J ) 19 Hz), 2.92 (1H, d, J ) 11 Hz), 3.26 (1H, bd, J ) 8.5 Hz),
3.61 (1H, d, J ) 19 Hz), 3.74 (2H, qAB, J ) 13 Hz), 3.88 (1H,
d, J ) 11 Hz), 4.02 (1H, dd, J ) 7 Hz, J ′ ) 10 Hz), 4.45 (1H,
d, J ) 5 Hz), 5.29 (1H, s), 7.29 (5H, m); 13C NMR (75 MHz,
CDCl3) δ 5.2, 6.0, 6.9, 7.0, 10.1, 14.6, 18.2, 31.6, 38.0, 40.7,
41.8, 48.8, 60.8, 62.5, 64.7, 70.5, 73.8, 78.8, 81.0, 89.0, 127.2,
128.2, 128.9, 137.7, 138.0, 153.3, 154.8, 198.1, 205.4; IR
(CHCl3) 1800, 1712, 1687 cm-1; MS (LSIMS+) m/z 740 (MH+),
712 (MH+ - carbonate).
P r ep a r a tion of Com p ou n d 17. To a CH2Cl2 solution (8
mL) of 16 (60 mg, 81 µmol) were added 4-DMAP (200 mg, 1.6
mmol, 20 equiv) and acetic anhydride (190 µL, 2 mmol, 25
equiv). The solution was stirred for 3 h at room temperature
and then hydrolyzed by aqueous NaHCO3. After standard
workup, the residue was purified by preparative TLC (hep-
tane/AcOE: 70/30) to yield 17 (46 mg, 71%): 1H NMR (250
MHz, CDCl3) δ 0.47 (6H, m), 0.65 (6H, m), 0.94 (18H, m), 1.25
(3H, s), 1.29 (3H, s), 1.52 (1H, m), 1.87 (3H, s), 2.04 (3H, s),
2.06 (3H, s), 2.28 (1H, m), 2.83 (2H, qAB, J ) 19 Hz), 3.14
(1H, d, J ) 11 Hz), 3.39 (1H, bd, J ) 9 Hz), 3.42 (1H, d,
J ) 5 Hz), 3.68 (2H, qAB, J ) 11 Hz), 3.73 (1H, d, J ) 11
Hz), 4.34 (1H, dd, J ) 8 Hz, J ′ ) 10 Hz), 4.45 (1H, d, J ) 5
Hz), 5.31 (1H, s), 7.32 (5H, m); 13C NMR (62.5 MHz, CDCl3) δ
5.1, 6.0, 6.8, 7.0, 9.9, 14.3, 18.5, 22.1, 31.8, 38.2, 40.1, 41.6,
43.8, 61.0, 61.4, 62.7, 68.4, 73.0, 77.6, 77.9, 80.5, 88.7, 127.4,
128.4, 128.9, 137.6, 138.0, 152.7, 155.6, 170.4, 196.6, 204.9;
IR (CHCl3) 1806, 1722, 1686 cm-1; MS (LSIMS+) m/z 782
(MH+).
P r ep a r a tion of Com p ou n d 18. To a dry THF solution (3
mL) of 17 (30 mg, 39 µmol) cooled to -72 °C was added under
argon a hexane solution of phenyllithium (1.6 M, 120 µL, 192
µmol, 4.8 equiv). The solution was stirred for 30 min at -72
°C and then poured on a mixture of CH2Cl2 and saturated
ammonium chloride and stirred for 1 min at room tempera-
ture. After standard workup, the residue was purified on
preparative TLC (heptane/AcOEt: 70/30) to yield 18 (30 mg,
91%): 1H NMR (300 MHz, CDCl3) δ 0.51 (6H, q, J ) 7 Hz),
0.76 (6H, q, J ) 7 Hz), 0.96 (9H, t, J ) 7 Hz), 1.09 (9H, t, J )
7 Hz), 1.26 (3H, s), 1.34 (3H, s), 1.50 (1H, m), 1.86 (3H, s),
2.11 (3H, s), 2.22 (3H, s), 2.24 (1H, m), 2.67 (1H, d, J ) 19
Hz), 2.98 (1H, d, J ) 11 Hz), 2.99 (1H, d, J ) 19 Hz), 3.45
(1H, bd, J ) 9 Hz), 3.46 (1H, d, J ) 11 Hz), 3.48 (1H, d, J )
13 Hz), 3.71 (1H, d, J ) 13 Hz), 3.84 (1H, d, J ) 7 Hz), 4.41
(1H, dd, J ) 7 Hz, J ′ ) 10 Hz), 5.42 (1H, s), 5.74 (1H, d, J )
7 Hz), 7.26 (5H, m), 7.56 (2H, t, J ) 7 Hz), 7.69 (1H, t, J ) 7
Hz), 8.16 (2H, d, J ) 7 Hz); 13C NMR (75 MHz, CDCl3) δ 5.3,
6.2, 6.9, 7.1, 9.9, 13.5, 17.9, 22.3, 33.1, 38.0, 42.6, 43.5, 46.9,
59.1, 61.6, 62.6, 73.5, 73.9, 76.9, 79.1, 127.5, 128.6, 129.0, 129.2,
129.9, 130.5, 133.9, 136.2, 138.1, 158.2, 167.3, 170.0, 199.2,
204.7; IR (CHCl3) 1725, 1667 cm-1; MS (LSIMS+) m/z 860
(MH+).
P r ep a r a tion of Com p ou n d 5. To a MeOH solution (3 mL)
of 21 (43 mg, 35 µmol) was added APTS (20 mg, 0.1 mmol, 3
equiv). The solution was stirred at room temperature for 5 h
and then diluted with an excess of ethyl acetate. After
standard workup, the residue was purified on preparative TLC
(CH2Cl2/MeOH: 95/5) to yield 5 (22 mg, 72%): 1H NMR (250
MHz, CDCl3) δ 1.11 (3H, s), 1.22 (3H, s), 1.34 (9H, s), 1.47
(1H, m), 1.82 (3H, s), 1.90 (3H, s), 2.24 (2H, m), 2.29 (3H, s),
2.33 (1H, m), 2.91 (1H, d, J ) 11 Hz), 3.43 (1H, bd, J ) 9 Hz),
3.47 (1H, d, J ) 11 Hz), 3.56 (2H, qAB, J ) 12 Hz), 3.79 (1H,
d, J ) 7 Hz), 4.15 (1H, dd, J ) 7 Hz, J ′ ) 10 Hz), 4.62 (1H,
bs), 5.19 (1H, s), 5.28 (1H, bd, J ) 9 Hz), 5.50 (1H, d, J ) 9
Hz), 5.65 (1H, d, J ) 7 Hz), 6.19 (1H, bt, J ) 8 Hz), 7.09, 7.21,
7.32 (5H, m), 7.39 (5H, m), 7.50 (2H, t, J ) 7 Hz), 7.62 (1H, t,
J ) 7 Hz), 8.11 (2H, d, J ) 7 Hz);
13C NMR (75 MHz, CDCl3)
δ 10.0, 14.5, 20.6, 23.0, 27.6, 28.3, 35.9, 37.6, 43.1, 46.9, 58.1,
58.5, 61.8, 62.2, 68.4, 72.5, 73.0, 74.0, 74.6, 75.4, 79.2, 79.3,
80.2, 126.9, 127.3, 128.0, 128.4, 128.7, 128.7, 128.8, 129.7,
130.3, 133.5, 137.5, 135.9, 137.2, 155.4, 167.3, 170.9, 176.7,
212.0; IR (CHCl3
) 1718 cm-1; MS (LSIMS+) m/z 897 (MH+),
919 (MNa+); HRMS calcd for C50H61N2O13 (MH+) 897.4174,
found 897.4182.
P r ep a r a tion of Com p ou n d 6. A MeOH solution (3 mL)
of 5 (22 mg, 24 µmol) with traces of acetic acid was stirred
over Pd on 10% charcoal (12 mg) under hydrogen for 5 h. The
solution was filtered on Celite, and the Celite was washed
several times with MeOH and CH2Cl2. After concentration
the residue was purified on preparative TLC (CH2Cl2/MeOH:
90/10) to yield 6 (12 mg, 61%): 1H NMR (250 MHz, CDCl3 +
CD3OD) δ 1.10 (3H, s), 1.21 (3H, s), 1.32 (9H, s), 1.68 (3H, s),
1.71 (1H, m), 1.75 (3H, s), 2.21 (2H, m), 2.32 (3H, s), 2.42 (1H,
m), 3.48 (2H, qAB, J ) 12 Hz), 3.82 (1H, d, J ) 7 Hz), 4.09
(1H, dd, J ) 6 Hz, J ′ ) 11 Hz), 3.43 (1H, dd, J ) 4 Hz, J ′ )10
Hz), 4.57 (1H, bs, 1H), 5.16 (2H, m), 5.61 (1H, d, J ) 7 Hz),
5.89 (1H, d, J ) 9 Hz), 6.15 (1H, bt, J ) 8 Hz), 7.35 (5H, m),
7.47 (2H, t, J ) 7 Hz), 7.59 (1H, t, J ) 7 Hz), 8.07 (2H, d, J )
7 Hz); 13C NMR (75 MHz, CDCl3 + CD3OD) δ 10.7, 14.6, 20.7,
23.1, 26.5, 28.3, 36.0, 38.6, 43.1, 46.7, 55.1, 56.2, 57.9, 61.3,
P r ep a r a tion of Com p ou n d 19. To a THF/EtOH solution
(22/78, 7.7 mL) of 18 (30 mg, 35 µmol) was added NaBH4 (18
mg, 480 µmol, 14 equiv). The solution was stirred at room
temperature overnight. After standard workup, the residue
was purified on preparative TLC (heptane/AcOEt: 60/40) to
yield 19 (22.5 mg, 74%): 1H NMR (300 MHz, CDCl3) δ 0.44

D-Ring-Modified Taxanes
J . Org. Chem., Vol. 62, No. 19, 1997 6637
72.5, 73.9, 74.7, 75.2, 79.0, 80.0, 83.2, 126.9, 128.1, 128.8, 128.9,
129.6, 130.3, 135.9, 133.7, 138.5, 155.4, 167.2, 170.5, 172.0,
212.0; IR (CHCl3) 1719 cm-1; MS (LSIMS+) m/z 829 (MNa+),
807 (MH+), 548 (MNa+ - side chain); HRMS calcd for
C43H55N2O13 (MH+) 807.3704, found 807.3724.
assays, Christiane Gaspard for cytotoxicity assays, and
J ean-Franc¸ois Gallard and Marie-The´re`se Martin for
performing NMR spectra.
Su p p or tin g In for m a tion Ava ila ble: 1H NMR (300 or 250
MHz) spectra for compounds 5-13, 16-19, and 21 and full
lisiting of NMR data, accompanied by subjective peak assign-
ments, for compounds 5-19 and 21 (21 pages). This material
is contained in libraries on microfiche, immediately follows this
article in the microfilm version of the journal, and can be
ordered from the ACS; see any current masthead page for
ordering information.
Ack n ow led gm en t. The authors thank Prof. Pierre
Potier for his constant interest in this work. We
gratefully acknowledge financial support from Rhoˆne-
Poulenc Rorer S.A. and the European Community
(AIR3-CT94-1979). We thank Dr. Alain Commerc¸on for
a gift of the docetaxel acid side chain. Sylviane Thoret
is also acknowledged for the microtubule disassembly
J O9706842