Synthesis of model BC bicycles of taxol using C10-C11 ring-closing metathesis strategy

Article abstract of DOI:10.1016/j.jorganchem.2006.08.078

Model BC ring-systems of taxol 23b and 28b, which lack an oxygenated substituent at C7, have been efficiently synthesized. The key step is a ring-closing metathesis (RCM) to form the 8-membered B ring between C10 and C11. Comparison of the metathesis reac

Full text of DOI:10.1016/j.jorganchem.2006.08.078

Journal of Organometallic Chemistry 691 (2006) 5438–5443
www.elsevier.com/locate/jorganchem
Synthesis of model BC bicycles of taxol using C10–C11
ring-closing metathesis strategy
a
a
b
a,
*
´
Stephanie Schiltz , Cong Ma , Louis Ricard , Joe¨lle Prunet
a
`
Laboratoire de Synthese Organique, UMR CNRS 7652, Ecole Polytechnique, DCSO, F-91128 Palaiseau, France
Laboratoire He´te´roe´le´ments et Coordination, UMR CNRS 7653, Ecole Polytechnique, DCPH, F-91128 Palaiseau, France
b
Received 5 July 2006; received in revised form 21 August 2006; accepted 21 August 2006
Available online 1 September 2006
Abstract
Model BC ring-systems of taxol 23b and 28b, which lack an oxygenated substituent at C7, have been efficiently synthesized. The key
step is a ring-closing metathesis (RCM) to form the 8-membered B ring between C10 and C11. Comparison of the metathesis reaction in
this route with the RCM used in a previous study with similar substrates is highlighted.
Ó 2006 Elsevier B.V. All rights reserved.
Keywords: Taxol; Metathesis; Cyclization; Ruthenium
TaxolÒ 1 and TaxotereÒ 2 are very effective therapeu-
tic agents for the treatment of breast and ovarian cancer
[1]. Six research groups have completed the total synthe-
sis of taxol, and numerous synthetic works have been
published since its isolation [2]. In the course of our
studies towards the synthesis of this molecule, we
planned a semi-convergent retrosynthesis of compound
3, which is an intermediate in Holton’s synthesis (Scheme
1) [2h,2i]. The A ring would be installed at a late stage
by an intramolecular aldol reaction on a diketone at
C13 and C11 (compound 4) [3]. Bicycle 5 would be
formed by a ring-closing metathesis (RCM) that would
close the B ring between C10 and C11. The metathesis
precursor 6 would be prepared by addition of the lithium
derivative of vinyl bromide 8 to aldehyde 7. We wish to
report here the synthesis of models of bicycle 5, which
lack the alkoxy group at C7 and the ketal function at
C13 [4].
dation with iodoxybenzoic acid [6], afforded ketone 10 in
excellent overall yield. Homologation of this ketone to
aldehyde 9 was effected by conversion to the corresponding
cyanohydrin 11 (TMSCN/ZnI2) and reduction by DIBAL-
H [7] (Scheme 2).
Vinyl bromide 12 was synthesized as an enantiopure
compound. The primary alcohol of known optically
active diol 13 [8] was protected as its trityl ether [9].
Other protecting groups such as triisopropylsilyl (TIPS)
or p-methoxybenzyl (PMB) ethers were also installed,
but the yield of coupling between 9 and 12 revealed very
dependent on this protecting group. We had originally
planned to convert ketone 14 into vinyl bromide 12 by
treatment of the corresponding hydrazone with NBS
[10], but formation of the hydrazone is plagued by an
important quantity of azine [11]. Ketone 14 was thus
transformed into vinyl triflate 15 in excellent yield,
and this compound afforded vinyl bromide 12 via the
corresponding trimethylvinyl stannane derivative [12]
(Scheme 3).
Model aldehyde 9 was prepared as a racemic mixture.
Barbier reaction between isoprenyl bromide and valeralde-
hyde according to Luche’s conditions [5], followed by oxi-
The coupling reaction of compounds 9 and 12 was then
undertaken. Halogen-metal exchange was best performed
by addition of a THF solution of bromide 12 into a solu-
tion of t-BuLi in THF. Subsequent addition of aldehyde
*
Corresponding author. Tel.: +33 1 69 33 48 73; fax: +33 1 69 33 38 51.
E-mail address: joelle.prunet@polytechnique.fr (J. Prunet).
0022-328X/$ - see front matter Ó 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2006.08.078

S. Schiltz et al. / Journal of Organometallic Chemistry 691 (2006) 5438–5443
5439
O
9
R2O
10
TESO
OH
7
OBOM
R1HN
Ph
O
C
C
O
3
5
B
13
B
A
1
A
D
O
TBSO
2
O
H
OBz
H
O
HO
OAc
O
OH
3
1 Taxol® (paclitaxel)
R1 = C6H5CO, R2 = Ac
O
Holton's intermediate
Aldol
2 Taxotere® (docetaxel) R1 = t-BuOCO, R2 = H
TESO
O
OR1
OR1
C
10
7
11
11
B
C
B
13
O
OR3
OR3
OR2
OR2
13
O
O
5
4
RCM
PO
OR1
C
OR1
C
10
11
CHO
O
Br
OTMS
O
O
OR3
OR2
7
8
O
6
Scheme 1.
9 furnished the desired coupling product as a mixture of
two diastereomers. Deprotection of the trimethylsilyl ether
then gave diols 16a and 16b, which were easily separated by
column chromatography (Scheme 4). The coupling reac-
tion is very stereoselective, favoring the trans diols [13]
(two isomers are obtained since the starting aldehyde is
racemic). This diastereoselectivity had already been
observed for similar coupling reactions performed in our
laboratory during previous studies on taxol precursors
[14]. The structures of both coupling products were secured
by X-ray crystallographic analysis of diol 16b for the
desired diastereomer (Scheme 4), and of the derived triol
20a for the other one [15].
1)
, Zn, NH4Cl
O
Br
2) IBX
98% (2 steps)
H
O
10
97%
TMSCN, ZnI2
DIBAL-H
85%
O
CN
13
OTMS
OTMS
9
11
Scheme 2.
We decided to convert both isomers to substrates for the
metathesis reaction, in order to compare the influence of
the diol stereochemistry on the RCM. The primary trityl
group had to be converted into a terminal olefin. Diols
16a and 16b were first protected as their carbonate deriva-
tives 17a and 17b, respectively (Scheme 5). Treatment with
sodium hydride and carbonyldiimidazole [4b] only gave the
desired products in 10% yield, along with some decom-
posed material. Fortunately, 17a and 17b could be
obtained by heating in toluene at reflux in the presence of
carbonyldiimidazole only. The trityl group in 17a and
17b was first deprotected with formic acid, and the corre-
sponding formate esters were dissolved in dioxane and
hydrolyzed with 1 N aqueous sodium hydroxide, leading
to alcohols 18a and 18b in 78% and 93% yield, respectively.
OH
OTr
1) DMAP•TrCl
2) IBX
HO
O
13
14
77% (2 steps)
KHMDS, HMPA
PhN(Tf)2
quant.
OTr
OTr
1) Pd(PPh3)4
(Me3Sn)2, LiCl
7
2) Br2
Br
TfO
12
15
78% (2 steps)
Scheme 3.

5440
S. Schiltz et al. / Journal of Organometallic Chemistry 691 (2006) 5438–5443
Scheme 4.
OTr
OH
OTr
OTr
Amberlyst-15
Im2CO
Toluene, Δ
20a 88%
20b 80%
HO OH
HO OH
70%
17b 60%
17a
O
HO OH
O
16a
16b
16a
16b
O
CSA 2,2-Dimethoxypropane
OH
Amberlyst-15
NO2
SeCN
1)
OH
NO2
SeCN
1)
PBu3
2) H2O2
PBu3
2) H2O2
71%
89%
18a
18b 94%
19a
19b 75%
O
O
O
O
22a 63%
22b 71%
21a 94%
21b quant.
O
O
O
O
O
O
Scheme 6.
Scheme 5.
These compounds were obtained in a more direct fashion
by treatment with Amberlyst-15 resin in methanol. The pri-
mary alcohols present in 18a and 18b were eliminated using
Grieco’s method [16], leading to the metathesis precursors
19a and 19b.
Metathesis precursors 22a and 22b where the diol moi-
ety is protected as an acetonide were also synthesized
(Scheme 6). Diols 16a and 16b were transformed into the
corresponding triols 20a and 20b with Amberlyst-15 resin
in methanol, and treatment of these compounds with 2,2-
dimethoxypropane and CSA furnished acetonides 21a
and 21b in good overall yields (83% and 80%, respectively).
Grieco’s method was then used as before, leading to the
desired products 22a and 22b.
The ring-closing metathesis (RCM) experiments were
first tested with carbonate 19a, which possesses the wrong
stereochemistry for taxol at C1 and C2. No cyclized prod-
uct was observed with Grubbs 1 [17] or Grubbs 2 [18] cata-
lyst in 1,2-dichloroethane at reflux, and prolonged reaction
times (several days) with the latter complex only led to
decomposition products. However, when diastereomer
19b was submitted to 30 mol% of Grubbs 1, after several
days at ambient temperature [19], the corresponding cyc-
looctene 23b was produced in 65% yield as the Z isomer
exclusively (Scheme 7). Use of second-generation catalysts
Grubbs 2 or [RuIMes] [20] furnished the same product in
comparable yields (69% and 72%, respectively), but the
reaction was much faster (1 h only).

S. Schiltz et al. / Journal of Organometallic Chemistry 691 (2006) 5438–5443
5441
Catalyst
10
11
1,2-Dichloroethane
O
O
O
O
19b
23b
O
O
N
N
Mes
Cl
Cl
Mes
Ph
Yield
Catalyst
Temperature
Time
Ru
PCy3
Grubbs 1 (30 mol%)
Grubbs 2 (10 mol%)
[RuIMes] (5 mol%)
20°C
80°C
80°C
5 days
1 h
1 h
65%
69%
72%
[RuIMes]
Scheme 7.
These results are in sharp contrast with those obtained
during our previous route to BC ring-systems of taxol,
where the RCM took place between the C9 and the C10
carbons instead of C10–C11 on otherwise similar sub-
strates [4]. In those cases, the diastereomer presenting the
wrong stereochemistry at the diol stereocenters (24a) did
not cyclize with Grubbs 1, but produced the expected cyc-
looctene 25a with [RuIMes] (Scheme 8) [21]. The other dia-
stereomer (24b) led to the kinetic trans cyclooctene E-25b
with Grubbs 1, and to the cis 8-membered ring Z-25b with
[RuIMes]. It seems that metathesis is much more difficult
for the wrong diastereomer when the ring-closing reaction
occurs at C10–C11 (this paper) rather than at C9–C10 (pre-
vious study), and much easier for the desired diastereomer.
Of course, one cannot directly compare the two reactions,
because the first carbene is formed on opposite sides of the
molecules in each case. In the first route, it is formed on
C10, which is attached to the left-hand side of the molecule,
and the reaction proceeds to close onto C9. In this work,
the first carbene is formed on C10, but this carbon is
attached to the right-hand side of the precursor, and
ring-closure occurs at C11.
10
9
O
O
O
O
Finally, the acetonide derivatives were subjected to
Grubbs 2 catalyst. After 30 min in CH2Cl2 at reflux, diaste-
reomer 22a only led to dimeric products 26 and 27 in a 3:1
ratio, favoring the symmetrical product (Scheme 9). Each
dimer was produced as a 3:1 mixture of diasteromers.
Cyclization of the other diastereomer 22b was exception-
ally fast, and bicycle 28b was obtained in quantitative yield
after only 5 min in 1,2-dichloroethane at reflux.
24a
24b
24a
10 mol% Grubbs 1
O
O
O
Benzene
reflux
8 days
O
O
O
O
E-25b
O
Here also, the same comparison can be made with the
RCM reactions of acetonides 29a and 29b from the previ-
ous study, which both cyclized with [RuIMes] after 12 h in
1,2-dichloroethane at reflux (Scheme 10). In the second
route, cyclization with the wrong diastereomer is not possi-
ble, and much faster for the isomer which leads to model
BC ring-systems of taxol.
In conclusion, we have synthesized highly functionalized
cyclooctenes that are model precursors of taxol, using a
ring-closing metathesis at C10–C11 to form the B ring.
Direct comparison of this route with a previous study
involving RCM at C9–C10 shows that the metathesis step
works much more efficiently for the diastereomers useful
for the synthesis using the C10–C11 strategy. We are cur-
rently preparing metathesis precursors with all the required
functionalities for taxol.
O
O
O
O
O
O
O
O
24a
24b
25a
5 mol% [RuIMes]
O
O
O
O
1,2-Dichloroethane
reflux
12 h
Z-25b
Scheme 8.

5442
S. Schiltz et al. / Journal of Organometallic Chemistry 691 (2006) 5438–5443
O
O
O
O
5 mol%
Grubbs 2
CH2Cl2
reflux
30 min
O
O
22a
O
O
O
O
87%
26
27
E/Z = 3:1 26/27 = 3:1 E/Z = 3:1
5 mol% Grubbs 2
1,2-Dichloroethane
80°C, 5 min
O
O
O
O
quant.
22b
28b
Scheme 9.
(f) K.C. Nicolaou, Z. Yang, J.J. Liu, P.G. Nantermet, C.F.
Claiborne, J. Renaud, R.K. Guy, K. Shibayama, J. Am. Chem.
Soc. 117 (1995) 645;
(g) K.C. Nicolaou, H. Ueno, J.J. Liu, P.G. Nantermet, Z. Yang, J.
Renaud, K. Paulvannan, R. Chadha, J. Am. Chem. Soc. 117 (1995)
653;
O
O
O
O
O
O
O
O
29a
29b
30a
30b
5 mol% [RuIMes]
(h) R.A. Holton, C. Somoza, H.B. Kim, F. Liang, R.J. Biediger,
P.D. Boatman, M. Shindo, C.C. Smith, S. Kim, H. Nadizadeh, Y.
Suzuki, C. Tao, P. Vu, S. Tang, P. Zhang, K.K. Murthi, L.N.
Gentile, J.H. Liu, J. Am. Chem. Soc. 116 (1994) 1597;
(i) R.A. Holton, H.B. Kim, C. Somoza, F. Liang, R.J. Biediger, P.D.
Boatman, M. Shindo, C.C. Smith, S. Kim, H. Nadizadeh, Y. Suzuki,
C. Tao, P. Vu, S. Tang, P. Zhang, K.K. Murthi, L.N. Gentile, J.H.
Liu, J. Am. Chem. Soc. 116 (1994) 1599;
1,2-Dichloroethane
reflux
12 h
(j) J.J. Masters, J.T. Link, L.B. Snyder, W.B. Young, S. Danishefsky,
Angew. Chem., Int. Ed. Engl. 34 (1995) 1723;
Scheme 10.
(k) P.A. Wender, N.F. Badham, S.P. Conway, P.E. Floreancig, T.E.
Glass, C. Gra¨nicher, J.B. Houze, J. Ja¨nichen, D. Lee, D.G. Marquess,
P.L. McGrane, W. Meng, T.P. Mucciaro, M. Mu¨hlebach, M.G.
Natchus, H. Paulsen, D.B. Rawlins, J. Satkofsky, A.J. Shuker, J.C.
Sutton, R.E. Taylor, K. Tomooka, J. Am. Chem. Soc. 119 (1997) 2755;
(l) P.A. Wender, N.F. Badham, S.P. Conway, P.E. Floreancig, T.E.
Glass, J.B. Houze, N.E. Krauss, D. Lee, D.G. Marquess, P.L.
McGrane, W. Meng, M.G. Natchus, A.J. Shuker, J.C. Sutton, R.E.
Taylor, J. Am. Chem. Soc. 119 (1997) 2757;
Acknowledgements
Financial support was provided by the CNRS and the
´ ´
Ecole Polytechnique. S.S. acknowledges the Delegation
´ ´
Generale pour l’Armement (DGA) for a fellowship. We
thank Dr. Ange Pancrazi for helpful discussion.
(m) T. Mukaiyama, I. Shiina, H. Iwadare, M. Saitoh, T. Nishimura,
N. Ohkawa, H. Sakoh, H. Nishimura, Y.-I. Tani, M. Hasegawa, K.
Yamada, K. Saito, Chem. Eur. J. 5 (1999) 121;
References
(n) K. Morihara, R. Hara, S. Kawahara, T. Nishimori, N. Nakamura,
H. Kusama, I. Kuwajima, J. Am. Chem. Soc. 120 (1998) 12980.
[3] C.S. Swindell, B.P. Patel, J. Org. Chem. 55 (1990) 3.
[4] For previous work from this laboratory on syntheses of model BC
ring-systems of taxol involving RCM between C9 and C10, see: (a) D.
Bourgeois, A. Pancrazi, L. Ricard, J. Prunet, Angew. Chem., Int. Ed.
Engl. 39 (2000) 725;
[1] N.H. Oberlies, D.J. Kroll, J. Nat. Prod. 67 (2004) 129.
[2] For reviews on the chemistry of taxol, see: (a) K.C. Nicolaou, W.-M.
Dai, R.K. Guy, Angew. Chem., Int. Ed. Engl. 33 (1994) 15;
(b) D.G.I. Kingston, Chem. Commun. (2001) 867;
For the total syntheses, see: (c) K.C. Nicolaou, Z. Yang, J.J. Liu, H.
Ueno, P.G. Nantermet, R.K. Guy, C.F. Claiborne, J. Renaud, E.A.
Couladouros, K. Paulvannan, E.J. Sorensen, Nature 367 (1994) 630;
(d) K.C. Nicolaou, P.G. Nantermet, H. Ueno, R.K. Guy, E.A.
Couladouros, E.J. Sorensen, J. Am. Chem. Soc. 117 (1995) 624;
(e) K.C. Nicolaou, J.J. Liu, Z. Yang, H. Ueno, E.J. Sorensen, C.F.
Claiborne, R.K. Guy, C.K. Hwang, M. Nakada, P.G. Nantermet, J.
Am. Chem. Soc. 117 (1995) 634;
(b) D. Bourgeois, J. Mahuteau, A. Pancrazi, S.P. Nolan, J. Prunet,
Synthesis (2000) 869.
´
[5] C. Petrier, J.-L. Luche, J. Org. Chem. 50 (1985) 12.
[6] M. Frigerio, M. Santagostino, S. Sputore, J. Org. Chem. 64 (1999)
4537.
[7] For a similar homologation, see Ref [4b] and: B. Muller, F. Delaloge,
´ ´
M. den Hartog, J.-P. Ferezou, A. Pancrazi, J. Prunet, J.-Y.

S. Schiltz et al. / Journal of Organometallic Chemistry 691 (2006) 5438–5443
5443
´
Lallemand, A. Neuman, T. Prange, Tetrahedron Lett. 37 (1996)
3313.
(b) P. de Armas, F. Garcia-Tellado, J.J. Marrero-Tellado, Eur. J.
Org. Chem. (2001) 4423.
[8] J. Boukouvalas, Y.-X. Cheng, J. Robichaud, J. Org. Chem. 63 (1998)
228.
[17] (a) P. Schwab, M.B. France, J.W. Ziller, R.H. Grubbs, Angew.
Chem., Int. Ed. Engl. 34 (1995) 2039;
[9] O. Hernandez, S.K. Chaudhary, R.H. Cox, J. Porter, Tetrahedron
Lett. 22 (1981) 1491.
(b) P. Schwab, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 118
(1996) 100.
[10] A. Pross, S. Sternbell, Aust. J. Chem. 23 (1970) 989.
[11] A. Pross, S. Sternbell, Aust. J. Chem. 24 (1971) 1437.
[12] L.A. Paquette, S. Liang, H.-L. Wang, J. Org. Chem. 61 (1996) 3268.
[13] Diols 16a and 16b would be trans if the B ring were closed.
[14] D. Bourgeois, G. Maiti, A. Pancrazi, J. Prunet, Synlett (2000) 323.
[15] Crystallographic data for the structural analysis has been deposited
with the Cambridge Crystallographic Data Centre, CCDC No.
613208 for compound 16b and No. 613209 for compound 20a.
[16] (a) P.A. Grieco, S. Gilman, M. Nishizawa, J. Org. Chem. 41 (1976)
1485;
[18] M. Scholl, S. Ding, C.W. Lee, R.H. Grubbs, Org. Lett. 1 (1999) 953.
[19] Reaction was started with 10 mol% catalyst, and 5 mol% additional
catalyst was added every day.
[20] (a) J. Huang, E.D. Stevens, S.P. Nolan, J.L. Petersen, J. Am. Chem.
Soc. 121 (1999) 2674;
(b) M. Scholl, T.M. Trnka, J.P. Morgan, R.H. Grubbs, Tetrahedron
Lett. 40 (1999) 2247.
[21] The RCM reactions depicted on Scheme 8 were performed on
mixtures of 24a and 24b, as these diastereomers are inseparable by
chromatography on silica gel.