Stereoselective synthesis of advanced intermediates en route to Taxuspine U and X: a study of macrocyclization via ring closing metathesis to highly constrained twelve-membered rings

Article abstract of DOI:10.1016/j.tetlet.2006.11.167

The stereoselective synthesis of an advanced intermediate of Taxuspine U and X has been accomplished using a ring closing metathesis strategy. The feasibility of ring closing metathesis in synthesizing highly constrained and functionalized macrocycles has been demonstrated provided the appropriate substrate structure and substitution pattern are chosen.

Full text of DOI:10.1016/j.tetlet.2006.11.167

Tetrahedron Letters 48 (2007) 751–754
Stereoselective synthesis of advanced intermediates en route to
Taxuspine U and X: a study of macrocyclization via ring
closing metathesis to highly constrained twelve-membered rings
Elena Galletti, Stanislava I. Avramova, Michela L. Renzulli,
Federico Corelli and Maurizio Botta*
`
Dipartimento Farmaco Chimico Tecnologico, Universita degli Studi di Siena, 53100 Siena, Italy
Received 10 November 2006; revised 20 November 2006; accepted 29 November 2006
Available online 18 December 2006
Dedicated to Professor Fulvio Gualtieri, University of Firenze, on the occasion of his 70th birthday
Abstract—The stereoselective synthesis of an advanced intermediate of Taxuspine U and X has been accomplished using a ring clos-
ing metathesis strategy. The feasibility of ring closing metathesis in synthesizing highly constrained and functionalized macrocycles
has been demonstrated provided the appropriate substrate structure and substitution pattern are chosen.
ꢀ 2006 Elsevier Ltd. All rights reserved.
Taxuspines U (2) and X (3) have been isolated from the
stems of the Japanese yew Taxus cuspidata.1 Taxuspines
U (2) and X (3) are very rare bicyclic taxane-related
diterpenoids and their skeleton has been proposed as a
biogenetic precursor of taxanes1c (Fig. 1). Molecular
modeling studies from our group have revealed that
modified Taxuspines, such as 4 (Scheme 1) and analogs
thereof, can adopt a conformation similar to the bio-
active conformation of paclitaxel and can be well
accommodated within the pseudoreceptor model
proposed by us to predict the microtubule-stabilizing
activity for taxanes.2
Considering our recent work on the synthesis of Taxu-
spine U and X analogs via an original, convergent
approach, involving a ring closing metathesis (RCM)3 as
a key step,4 we focused our attention on the stereoselec-
tive synthesis of compound 5, an advanced intermediate
of Taxuspine U and X (Scheme 1). Our retrosynthetic
analysis (Scheme 1) suggested that the twelve-membered
macrocycle of 5 could be constructed by RCM of inter-
mediate 6, which is expected to result from the coupling
reaction between the suitably functionalized alkyne 74
and aldehyde 8. In this letter, we describe the prepara-
tion of chiral synthon 8 using the SAMP/RAMP-
hydrazones methodology,5 the stereoselectivity of the
alkynylation reaction, and our studies on the macro-
cyclization of substrate 6.
O
O
AcO
OH
Ph
NH
O
Ph
O
O
OH
OAc
AcO
H
HO
OAc
OCOPh
Taxol (1)
OAc
OH
OAc
AcO
AcO
AcO
O
The stereoselective synthesis of 8 is outlined in Scheme
2. 1,3-Dioxanone 9 has been obtained according to the
procedure described in the literature6 and transformed
into the corresponding 2,2-dimethyl-1,3-dioxan-5-one-
SAMP-hydrazone 10. Smooth reaction of the azaenol-
ate, generated by deprotonation with tert-buthyllithium
at low temperature, with ethyl bromoacetate afforded
ester 11 in a good yield. Cleavage of the chiral auxiliary
by ozonolysis gave a-substituted ketone 12 in an
OAc
O
H
H
Ph
OH
OAc
OH
OAc
Taxuspine X (3)
Taxuspine U (2)
Figure 1.
Keywords: Taxuspine; Macrocyclization; Ring closing metathesis.
Corresponding author. Tel.: +39 0577 234306; fax: +39 0577
234333; e-mail: botta@unisi.it
*
0040-4039/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2006.11.167

752
E. Galletti et al. / Tetrahedron Letters 48 (2007) 751–754
esterification
coupling
OAc
OH
O
Ring Closing Metathesis
R'
O
A
H
B
OH
OAc
OBz
4
oxygenation
OR
OR
8
OR
9
RO
OR
O
3
4
O
OR
OR
OR
H
H
H
O
OR
OR
OR
OR
5
6
7
8
R = protecting group
Scheme 1.
O
AcO
N
N
N
N
O
H
i
H3CO
H3CO
ii
i
COOEt
O
O
O
O
O
11
O
O
H
O
OTBDPS
(+/-)
10
9
7
8
O
iii
iv
COOEt
COOEt
AcO
AcO
OR
OR
7
7
O
O
O
O
12
13
O
O
H
H
v
CHO
[α]D20 = -9.15
OTBDPS
OTBDPS
O
O
O
O
14a R = H
15a R = Ac
[α]D20 = -16.6
14b R = H
15b R = Ac
[α]D20 = -22.3
ii
ii
8
Scheme 2. Reagents and conditions: (i) SAMP, benzene, reflux, 12 h,
95%; (ii) t-BuLi, BrCH2COOEt, THF, ꢀ78 ꢁC to rt, 12 h, 75%; (iii) O3,
CH2Cl2, ꢀ78 ꢁC, 4 h, 95%; (iv) Ph3P(Br)CH3, t-ButOK, THF, 0 ꢁC to
reflux, 1 h, 60%; (v) DIBAH, CH2Cl2, ꢀ78 ꢁC, 30 min, 65%.
Scheme 3. Reagents and conditions: (i) LiHMSA, THF, ꢀ78 ꢁC to rt,
1 h, 50–65%; (ii) Ac2O, Et3N, DMAP, CH2Cl2, rt, 1 h, 95%.
favoured
Nu
excellent enantiomeric excess (ee = 89%, determined by
HPLC with chiral column Chiralcel OD). Wittig olefin-
ation of the carbonyl moiety and partial reduction of the
ester with DIBAH, led to the desired chiral aldehyde 8.7
Nu
OR2 OH
1,3-anti
R3
R3
O M
O
H
R2
Reaction of the organolithium derivative of racemic
alkyne 7 with the enantiomerically pure aldehyde 8 gave
a 1:1 mixture of two diasteroisomeric alcohols 14a and
14b (Scheme 3).
R2
O
O M
Nu
R3
H
OR2 OH
1,3-syn
R3
disfavoured
Nu
The diastereoisomers have been separated by chroma-
tography. HPLC/MS and NMR analysis on the O-acet-
ylated compounds 15a8 and 15b helped us in
determining the relative stereochemistry of the newly
formed stereocenter at C-7; COSY and NOESY experi-
ments, performed at room and low temperature,
Scheme 4. Prechair transition-state model for bidentate 1,3-chelation.
revealed that the stereocenter at position 7 has the
same configuration in both derivatives. The oxygen at

E. Galletti et al. / Tetrahedron Letters 48 (2007) 751–754
753
AcO
AcO
Several RCM assays have been performed on inter-
mediates 15a and 15b, in the different conditions of solvent
(CH2Cl2, C2H4Cl2, toluene), dilution (1.0–3.5 mM), tem-
perature (rt–60 ꢁC), catalyst loading (10–20%), using
both ruthenium (Grubbs I or Grubbs II catalyst) and
molybdenum (Schrock catalyst) catalysts (Scheme 5).
No case cyclization has been observed; the unaltered
starting material has been recovered when the reactions
have been run at room temperature, whereas degradation
occurred at a higher temperature, in accordance with
the outcome of the RCM reactions in our previous
studies.4
OAc
OAc
i
O
O
O
O
H
H
OTBDPS
OTBDPS
16
15a
AcO
OAc
ii
AcO
OAc
O
O
H
i
O
19
It is known that RCM is strongly substrate-dependent
mainly due to steric hindrance, complexation, and con-
formational preorganization of the acyclic dienic precur-
sor;10 therefore, ring closing metathesis reactions have
been studied on modified substrates. After cleavage of
silyl ether 15a, the resulting less hindered allylic alcohol
17 was treated with 20% Grubbs II catalyst in refluxing
dichloromethane; HPLC/MS analysis of the reaction
mixture revealed the presence of the desired macrocycle
18 in 11% yield, whereas methyl ketone 1911 has been
isolated as the major compound (Scheme 5). Competi-
tive fragmentation reactions occur when the ring closure
is slow12 and it has been suggested that molecular strains
in 17 favor this type of side reactions.
O
O
AcO
OAc
H
OH
17
O
O
H
OH
18
[α]D20 = -7.9
Scheme 5. Reagents and conditions: (i) RCM; (ii) TBAF, THF, 40 ꢁC,
24 h, 80–95%.
the stereocenter in 8 plays a crucial role in controlling
the stereochemistry of the reaction and our results are
in agreement with the literature data concerning the
alkynylation of chiral aldehydes.9
No improvement has been achieved when less hindered
protecting groups for the allylic alcohol have been
employed.
Due to the lithium chelation, in the prechair transition
state (Scheme 4) the attack by the incoming nucleophile
on the less hindered side of the aldehyde favors the for-
mation of the 1,3-anti isomer, while the 1,3-syn reaction
product is disfavored. Furthermore, thanks to this sub-
strate-controlled reaction, the starting racemic mixture
of alkyne 7 has been separated.
To enhance the flexibility of the acyclic precursor and to
promote the approach of the terminal double bonds, the
acetal functionality has been cleaved (Scheme 6) under
acidic condition to afford the deprotected triol 20,
OAc
OAc
AcO
AcO
OAc
i
+
OR
OR
OR
O
O
H
H
ii
H
OR
20 R = H
OR
OR
OH
17
21 R = H
ii
23 R = Ac
22 R = Ac
AcO
AcO
OAc
OAc
iii
iii
OAc
OAc
OAc
OAc
H
H
OAc
24
OAc
22
OAc
OAc
OAc
OAc
OAc
H
H
OAc
OAc
25
[α]D20 = -4.6
OAc
23
Scheme 6. Reagents and conditions: (i) CH3COOH 80%, 0 ꢁC, 1 h; (ii) Ac2O, Et3N, DMAP, CH2Cl2, rt, 1 h, 95%.; (iii) RCM.

754
E. Galletti et al. / Tetrahedron Letters 48 (2007) 751–754
Chem., Int. Ed. 2000, 39, 2073–2077; (h) Furstner, A.
Angew. Chem., Int. Ed. 2000, 39, 3013–3043; (i) Jorgensen,
M.; Hadwiger, P.; Madsen, R.; Stutz, A. E.; Wrodnigg, T.
M. Curr. Org. Chem. 2000, 4, 565–588; (j) Wright, D. L.
Curr. Org. Chem. 1999, 3, 211–240; (k) Grubbs, R. H.;
Chang, S. Tetrahedron 1998, 54, 4413–4450; (l) Arm-
strong, S. K. J. Chem. Soc., Perkin Trans. 1 1998, 371–388;
(m) Blechert, S.; Schuster, M. Angew. Chem., Int. Ed.
together with the unexpected product 21 deriving from
deacetylation and dehydratation at C-10 position.
Although no other examples of this reaction have been
described, we noticed that it occurs very easily in our
C-13 unsubstituted substrates.
RCM on the O-acetylated derivative 22 proved to be
unsuccessful, probably due to entropic reasons. Interest-
ingly, when 23 was treated with 20% Schrock catalyst in
toluene at 60 ꢁC, bicyclic compound 2513 has been
isolated in 23% yield. Based on molecular mechanics
calculations, this result may be interpreted in terms of
molecular constraints of 23, which most likely create the
appropriate distance and relative orientation for giving
the ring closure between the two terminal double bonds.
1997, 36, 2036–2056; (n) Furstner, A.; Langemann, K.
¨
Synthesis 1997, 792–803; (o) Furstner, A.; Langemann, K.
¨
J. Org. Chem 1996, 61, 3942–3943.
4. (a) Renzulli, M.; Rocheblave, L.; Avramova, S.; Corelli,
F.; Botta, M. Tetrahedron Lett. 2004, 45, 5155–5158; (b)
Renzulli, M. L.; Rocheblave, L.; Avramova, S. I.; Galletti,
E.; Castagnolo, D.; Tafi, A.; Corelli, F.; Botta, M.
Tetrahedron 2007, 63, 497–509.
5. Job, A.; Janeck, C. F.; Bettray, W.; Peters, R.; Enders, D.
Tetrahedron 2002, 58, 2253.
In summary, we have applied the ring closing metathesis
to synthesize highly constrained and functionalized macro-
cycles; so far, no example of twelve-membered carbocy-
cle, obtained via RCM, has been reported in the litera-
ture prior to our studies.4 Moreover, an original
methodology to achieve the bicyclic core of Taxuspine
U and X analogs has been developed. The reported
study is part of a wider project aimed at designing and
synthesizing new microtubule-stabilizing antimitotic
compounds.
6. Enders, D.; Voith, M.; Iince, S. J. Synthesis 2002, 12, 1775.
7. Compound 8 has been characterized: IR (film): m 1725,
1110 cmꢀ1. Electrospray MS m/z 203 [M+Na]+. 1H NMR
(200 MHz, CDCl3) d 9.8 (b, 1H), 4.83 (d, J = 1.2 Hz, 2H),
4.8 (m, 1H), 4.14 (dd, 1H, J = 7.0, 1.6 Hz), 2.77 (m, 2H),
1.55, 1.38 (2s, 6H). Anal. Calcd for C9H14O3: C, 63.51; H,
20
8.29. Found: C, 63.53; H, 8.30; ½aꢁD ꢀ9.15 (c 0.185,
CHCl3).
8. Compound 15a has been characterized: Electrospray MS
m/z 750 [M+Na]+. 1H NMR (400 MHz, CDCl3) d 7.7–
7.26 (m, 10H), 5.99 (s, 1H), 5.84 (ddd, 1H, J = 18.0, 10,
8.4 Hz), 5.65 (dd, 1H, J = 9.7, 3 Hz), 4.84 (d, 2H,
J = 7.5 Hz), 4.61 (d, 1H, J = 10 Hz ), 4.56 (d, 1H,
J = 17 Hz,), 4.35 (d, 1H, J = 8.7 Hz), 4.38 (d, 1H, J =
9.7 Hz), 4.26 (s, 2H,), 2.27 (m, 1H6a), 2.2–2.0 (m, 2H), 2.06
(s, 3H), 2.04 (s, 3H), 1.94 (m, 1H6b), 1.87 (s, 3H), 1.65–1.4
(m, 2H), 1.36 (s, 3H), 1.35 (dd, 1H, J = 10, 3 Hz), 1.25 (s,
3H), 1.01 (s, 3H), 0.95 (s, 9H), 0.87 (s, 3H). Anal. Calcd
for C44H58O7Si: C, 72.69; H, 8.04. Found: C, 72.74; H,
Acknowledgments
Financial support provided by the Fondazione Monte
dei Paschi di Siena is gratefully acknowledged. We also
thank Dott. F. Bartocci (Galenica Senese) for providing
us with a very practical handmade ozone generator.
20
8.02; ½aꢁD ꢀ16.6 (c 0.217, CHCl3).
´
9. Guillame, S.; Ple, K.; Banchet, A.; Liard, A.; Haudrechy,
A. Chem. Rev. 2006, 106, 2355–2403.
References and notes
10. (a) Van de Weghe, P.; Eustache, J.; Cossy, J. Curr. Top.
Med. Chem. 2005, 5, 1461–1472, and references cited
therein; (b) Gaich, T.; Mulzer, J. Curr. Top. Med. Chem.
2005, 5, 1473–1494; (c) Van de Weghe, P.; Eustache, J.
Curr. Top. Med. Chem. 2005, 5, 1495–1519.
11. Compound 19 has been characterized: Electrospray MS
m/z 281, 497 [M+Na]+. 1H NMR (400 MHz, CDCl3) d
6.10 (s, 1H), 5.59 (m, 1H), 4.84 (d, 2H, J = 1.2 Hz), 4.58
(m, 1H), 4.28 (m, 2H), 2.6–2.4 (br s, 1H), 2.15 (s, 3H), 2.0–
1.6 (br s, 6H), 2.05, 2.04 (2 · s, 6H), 1.90 (s, 3H), 1.44, 1.38
(2 · s, 6H), 1.07 (s, 3H), 099 (s, 3H). Anal. Calcd for
C27H38O7: C, 68.33; H, 8.21.
1. (a) Hosoyama, H.; Inubushi, A.; Katsui, T.; Shigemori,
H.; Kobayashi, J. Tetrahedron 1996, 52, 13145; (b)
Shigemori, H.; Wang, X.-X.; Yoshida, N.; Kobayashi, J.
Chem. Pharm. Bull. 1997, 45, 1205; (c) Shigemori, H.;
Kobayashi, J. J. Nat. Prod. 2004, 67, 245–256.
2. (a) Manetti, F.; Maccari, L.; Corelli, F.; Botta, M. Curr.
Top. Med. Chem. 2004, 4, 203–217; (b) Maccari, L.;
Manetti, F.; Corelli, F.; Botta, M. Il Farmaco 2003, 58,
659–668; (c) Wang, M.; Xia, X. Y.; Kim, Y.; Hwang, D.;
Jansen, J. M.; Botta, M.; Liotta, D. C.; Snyder, J. P. Org.
Lett. 1999, 1, 43–46.
3. For recent reviews see: (a) Gradillas, A.; Perez-Castells, J.
Angew. Chem., Int. Ed. 2006, 45, 6086–6101; (b) Nicolaou,
K. C.; Bulger, P. G.; Sarlah, D. Angew. Chem., Int. Ed.
2005, 44, 4490–4527; (c) Prunet, J. Angew. Chem., Int. Ed.
2003, 42, 2826–2830; (d) schrock, R. R.; Hoveyda, A. H.
Angew. Chem., Int. Ed. 2003, 42, 4592–4633; (e) Trnka, T.
M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18–29; (f)
Sanford, M. S.; Love, J. A.; Grubbs, R. H. J. Am. Chem.
Soc. 2001, 123, 6543–6554; (g) Maier, M. E. Angew.
12. (a) Schmidt, B. Eur. J. Org. Chem. 2004, 1865–1880; (b)
Hoye, T. R.; Zhao, H. Org. Lett. 1999, 1, 1123–1125.
13. Compound 25 has been characterized: Electrospray MS
1
m/z 509 [M+Na]+. H NMR (400 MHz, CDCl3) d 6.02–
5.89 (m, 3H), 5.42 (m, 1H), 5.29–5.24 (m, 1H), 5.21–5.17
(m, 1H), 4.46 (br s, 2H), 2.56–2.44 (m, 2H), 2.12 (s, 3H),
2.09 (s, 3H), 2.07 (s, 3H), 2.06–2.03 (m, 1H), 2.01 (s, 3H)
1.94 (s, 3H), 1.54, 1.43 (m, 2H), 1.26 (s, 3H), 1.16 (s, 3H).
Anal. Calcd for C27H34O8: C, 66.65; H, 7.04. Found: C,
20
66.57; H, 7.06; ½aꢁD ꢀ4.6 (c 0.175, CHCl3).