Synthesis, biological evaluation, and modeling of a C,D-seco-taxoid

Article abstract of DOI:10.1016/j.tetlet.2005.05.071

A C,D-seco-paclitaxel derivative 12 was prepared and tested for biological activity. The key step in the synthesis was a free radical fragmentation of the bicyclic tertiary alcohol 7 under the conditions of the hypobromite reaction. The compound 12 showed

Full text of DOI:10.1016/j.tetlet.2005.05.071

Tetrahedron
Letters
Tetrahedron Letters 46 (2005) 5049–5052
Synthesis, biological evaluation, and modeling of a C,D-seco-taxoid
Zorana Ferjancic,a,b Radomir Matovic,b Zivorad Cekovic,b
James P. Snyderc and Radomir N. Saicica,b,
*
aFaculty of Chemistry, University of Belgrade, Studentski trg 16, PO Box 158, 11000 Belgrade, Serbia and Montenegro
bICTM-Centre for Chemistry, Njegoseva 12, 11000 Belgrade, Serbia and Montenegro
cDepartment of Chemistry, Emory University, 1515 Dickey Drive, Atlanta, GA 30322, USA
Received 9 April2005; revised 13 May 2005; accepted 17 May 2005
Available online 9 June 2005
Abstract—A C,D-seco-paclitaxel derivative 12 was prepared and tested for biological activity. The key step in the synthesis was a
free radical fragmentation of the bicyclic tertiary alcohol 7 under the conditions of the hypobromite reaction. The compound 12
showed no activity in the tubulin test.
Ó 2005 Elsevier Ltd. All rights reserved.
The advent of paclitaxel (taxol) 11 and docetaxel(taxo-
O
OR2
tere) 21,2 is considered as one of the most important
recent achievements in the development of anticancer
therapeutic agents (Fig. 1). In order to fully understand
the compoundsꢀ mechanism of action and determine the
structuralfeatures essentialfor their biool gicalactivity,
extensive structure–activity relationship (SAR) studies
on taxoids have been performed. These studies sug-
gested that the oxetane ring may be one of the crucial
structural units of biologically active taxoids; however,
itsꢀ exact role is still a matter of debate.3 The four-mem-
bered ring could act as a rigidifying element, which
imposes a proper conformationalbias to the taxane
core, thus holding the functional groups at C-2, C-4,
and C-13 at favorable positions for productive interac-
tions with tubulin receptor. Alternatively, electronic ef-
fects may be important, with the heteroatom acting as
a hydrogen bond acceptor, directly engaged in the inter-
action with microtubules. For both hypotheses experi-
mentalsupport exists, as wellas some contradictory
data. Thus, substituting the nitrogen atom for oxygen
in the four-membered ring results in reduced activity,
although the geometry of the azetidine derivatives does
not differ very much, with respect to the parent oxe-
tane.4 This finding infers the prevalence of the electronic
factors. SAR studies on thia-docetaxeland paciltaxel
R1
Ph
NH
O
O
OH
O
O
OH
HO
BzO
AcO
1 Paclitaxel R1 = Ph, R2 = Ac
2 Docetaxel R1 = OtBu, R2 = H
O
OAc
O
Ph
NH
O
OH
O
Ph
O
OH
HO
BzO
AcO
3 C,D-seco-Paclitaxel
Figure 1.
analogues point to a similar conclusion.5 On the other
hand, a recently synthesized taxoid containing cyclopro-
pane ring in place of the oxetane showed the activity
comparable to that of docetaxel, pointing to a purely
conformationaleffect of the D-ring. 6 The lack of activity
Keywords: Taxane antitumor agents; Radicals and radical reactions;
Medium sized rings; Natural products; Molecular modeling.
*
of severalD- seco-taxoids was explained by inappropri-
7,3b
Corresponding author. Tel.: +381 11 32 82 537; fax: +381 11 63 60
61; e-mail: rsaicic@chem.bg.ac.yu
ate orientation of the 4-acetylgroup.
Example is
0040-4039/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.05.071

5050
Z. Ferjancic et al. / Tetrahedron Letters 46 (2005) 5049–5052
known, however, of a D-seco-taxoid, which retains high
activity in the tubulin test.8
procedure without affecting substantially the biological
10
activity of the finalproduct.
In order to estimate the effect of conformationalcon-
straint imposed by the oxetane ring, we set out to syn-
thesize C,D-seco-paclitaxel derivatives of type 3 and
evaluate their biological activity (Fig. 1). We decided
to perform our study on 7-deoxytaxoids, as the hydroxyl
group in position C-7 has been shown to be irrelevant
for the biological activity of paclitaxel.9 In addition,
some preliminary experiments indicated that the pres-
ence of a carbonyl group in position C-9 could compli-
cate some stages of the synthesis. Therefore, we planned
to install a 9-a-acetoxy group in place of the carbonyl, as
this modification would somewhat simplify the synthetic
The synthesis started with the cinnamoyltaxicine deriv-
atives 4a,b whose preparation from taxine B11—a pseudo-
alcaloid readily isolated from the renewable needles of
the European yew (Taxus baccata)—was described ear-
lier (Scheme 1).12 Selective cleavage of the cinnamoyl
side chain in 5, in the presence of two acetate units
and a cyclic carbonate, was effected using a previously
developed procedure:12 dihydroxylation of 5 with
OsO4 resulted in stereoselective introduction of two req-
uisite hydroxyl groups into the taxane core, while simul-
taneously activating (by intramolecular hydrogen
bonding) the intermediate 2,3-dihydroxy-phenylpropan-
OR1
OR2
AcO
OAc
Ref. 12
Ac2O, pyridine,
Taxine
O
O
CH2Cl2, DMAP
93%
O
OCinn
O
O
OCinn
O
O
5
O
4a R1= H; R2= Ac
b R1=Ac; R2=H
1. OsO4, THF,
H2O, t-BuOH
73%
OAc
2. KOAc, MeOH,
80 °C
1. TBDMSCl, DMAP,
Et3N, CH2Cl2
AcO
OAc
AcO
AcO
11
OAc
9
2
AgOAc, Br2,
PhH, hν
7
2. MsCl, pyridine
6
O
O
O13
3
5
X
O
3. TBAF, THF
4
O 1
O
O
O
O
OH
OH
O
O
O
HO
4. DIPEA, PhMe, ∆
O
HO
O
20
O
52% (over 4 steps)
8 X=Br
7
6
9 X=OAc
TBTH, PhH,
AIBN, hν
AcO
AcO
AcO
OAc
OAc
OAc
O
NaBH4
MeOH
1. PhLi, THF, -80°C
O
O
HO
2. Ac2O, CH2Cl2,
DMAP
HO
O
O
HO
O
O
O
O
O
72%
O
O
O
O
O
62%
Ph
11
Ph
10
61% (from 7)
PMP
O
BzN
Ph
O
OAc
O
1.
COOH
DCC, DMAP
PhCH3, 75 °C
OAc
NH
O
O
OH
HO
2. 5% pTsOH, MeOH
70%
O
BzO
12
Scheme 1.

Z. Ferjancic et al. / Tetrahedron Letters 46 (2005) 5049–5052
5051
oate side chain toward alcoholysis in a buffered methan-
olic solution (73% over two steps). The resulting triol 6
was converted into oxetane 7 by a slightly modified four-
step protocoldeveol ped by the CNRS group (52% over
four steps),12,13 thus setting the stage for the fragmenta-
tion reaction. We planned to effect the crucial skeletal
transformation—the scission of the centralcarbon–car-
bon bond of the C,D-ring system in 7—by using radical
methodology. To this aim several methods of tertiary
alkoxyl radical generation were tried with variable suc-
cess. Surprisingly, reagents such as HgO/I2, Ag2O/Br2,
or DIB/I2 under photolytic conditions gave mixtures
of halide 8 and the corresponding acetate 9 (with the lat-
ter predominating), as well as some unidentified prod-
ucts. After considerable experimentation, we found
that bromide 8 and acetate 9 could both be obtained
chemoselectively with AgOAc/Br2 under carefully con-
trolled conditions. When the reaction was performed
with 1.4–1.5 equiv of AgOAc and 2.0–2.5 equiv Br2 the
required bromide 8 could be isolated as the sole product.
Using 4 equiv of AgOAc and 2 equiv of Br2 reverses the
chemoselectivity of the reaction and leads to the exclu-
sive formation of acetate 9. Bromide 8 was obtained as
an equimolar mixture of diastereoisomers, which were
further reduced (without separation) with Bu3SnH to
furnish the C,D-seco-derivative 10 (61% yield, over
two steps). Subsequent transformations involved instal-
lation of a benzoate ester into the position C-2 (for sim-
plicity and consistency, the taxane numeration, as
shown in the structuralformual of 8, is maintained after
the fragmentation step) (excess PhLi, 62%, after reacet-
ylation) and reduction of the C-13 carbonyl group
(NaBH4, 72%). At this stage we planned the reduction
of the C-4 carbonylgroup in 11 followed by acetylation.
Surprisingly, this ketone proved unreactive toward a
variety of reduction agents, such as NaBH4 (20 equiv,
rt), NaBH4/CeCl3, BH3ÆMe2S (rt), Li-Selectride
(ꢀ70 ! 0 °C), Red-Al( ꢀ60 ! +50 °C), or Noyori
transfer–hydrogenation (performed separately with both
(R,R)- and (S,S)-enantiomers of TsDPEN ligand). Un-
der reducing conditions with Et3SiH/BF3, only products
derived from Lewis acid-catalyzed rearrangement were
observed. With LiAlH4 a complex mixture of products
was obtained. At low temperatures, excess DIBALH
induced only the reductive hydrolysis of the acetate
groups, while a very complex mixture of products was
formed at rt. The attempted Noyori reduction of 10
was very slow and resulted only in reductive deprotec-
tion of the cyclic carbonate unit. With other reducing re-
agents occasionalreduction of the C-13 carbonylgroup
was observed, but the C-4 carbonylremained intact.
Apparently, fragmentation of the central bond in the
condensed C,D-system brings about a major conforma-
tionalchange in the taxane core and renders the newyl
formed carbonylgroup (C-4) inaccessibel to reducing
agents (in paclitaxel it is the carbonyl group at C-9 that
is resistant even to LiAlH4).
Figure 2.
ergy conformer is a crown form at 3.1 kcal/mol, while
two twist–chair structures fall higher. Boltzmann popu-
lations at 298 K are calculated to be 99.5%, 0.5%, and
<0.1%, respectively. The calculated relative stability of
15
conformers is identicalto that for cycol octanone
and
suggests that attempts to reduce the C-4 carbonylin
11 operate on the BC conformation shown in as a
space-filling model in Figure 2. The C-4 carbonyl(cen-
ter, yellow arrow) is surrounded and sequestered by
the C-2 phenylgroup (elft), the C-8 CH group (right)
2
and the C-18 Me group (i.e., Me-19, top, magenta ar-
row). The eight-membered ring oxygen is highlighted
by the red arrow. The buried carbonylcarbon is steri-
cally inaccessible on both p-faces to either nucleophile
or reducing agent.
Thus, introduction of an acetate group into position
C-4—a structuralfeature which is known to be impor-
tant for the biological activity of taxoids—turned out
to be unfeasible for C,D-seco-derivative 11. However,
we proceeded further with the appendage of the side
chain and the conversion of 4-oxo-derivative 11 into a
C,D-seco-taxoid 12 suitable for testing of biological
activity. This transformation proceeded uneventfully,
according to the previously described protocol.16 It
should be noted, however, that prolonged reaction times
in the deprotection step lead to the skeletal rearrange-
ment of 12 to the 11(15-1)-abeotaxane.
The result of the tubulin test17 showed 12 to be devoid of
microtubule stabilizing activity. At 11.9 mM concentra-
tion, 12 effected only 7% inhibition of the microtubule
disassembly process (for comparison, taxol showed
50% inhibition at 1 lM concentration). This finding
confirms the importance of both the oxetane ring and
the C,D-ring system in taxoids for the biological activ-
ity. However, the fact that compound 12 lacks the C-4
acetate group makes this result somewhat inconclusive,
as it is difficult to estimate, which of the two structural
features of 12 (the conformationalchange, or the ab-
sence of the C-4 acetate) contributes more to the loss
of the microtubule stabilizing properties. Another factor
concerns the relocation of the oxetane oxygen and the
adjacent carbons by the BC conformation placing them
deep within the hydrophobic tubulin pocket. The corre-
sponding position of the CH2–O–CH2 fragment within
In order to examine any potentialconformationalreor-
ganization arising from the ring cleavage, we performed
a 10,000 step MMFF/GBSA/H2O14 conformational
search for ketone 11 within 7.0 kcal/mol from the
boat–chair (BC) global minimum. The second lowest en-

5052
Z. Ferjancic et al. / Tetrahedron Letters 46 (2005) 5049–5052
2. Guenard, D.; Gueritte-Voegelin, F.; Potier, P. Acc. Chem.
Res. 1993, 26, 160–167.
3. (a) Wang, M.; Cornett, B.; Nettles, J.; Liotta, D. C.;
Snyder, J. P. J. Org. Chem. 2000, 65, 1059–1068; (b) Boge,
T. C.; Hepperle, M.; Vander Velde, D. G.; Gunn, C. W.;
Grunewald, G. L.; Georg, G. I. Bioorg. Med. Chem. Lett.
1999, 9, 3041–3046.
4. Marder-Karsenti, R.; Dubois, J.; Bricard, L.; Guenard,
D.; Gueritte, F. J. Org. Chem. 1997, 62, 6631–6637.
5. (a) Merckle, L.; Dubois, J.; Place, E.; Thoret, S.; Gueritte,
F.; Guenard, D.; Poupat, C.; Ahond, A.; Potier, P. J. Org.
Chem. 2001, 66, 5058–5065; (b) Gunatilaka, A. L.;
Ramdayal, F. D.; Sarragiotto, M. H.; Kingston, D. G.
I.; Sackett, D. L.; Hamel, E. J. Org. Chem. 1999, 64, 2694–
2703.
Figure 3.
6. Dubois, J.; Thoret, S.; Gueritte, F.; Guenard, D. Tetra-
hedron Lett. 2000, 41, 3331–3334. Recently, it was
reported, however, that although highly active in the
tubulin test, the cyclopropyl-derivative is much less
cytotoxic than paclitaxel.
the concave pocket of the tricyclic core of 11 is illus-
trated at the bottom of Figure 3a. By contrast the same
atoms in the oxetane ring of the tetracyclic baccatin
moiety are directed outward from the convex face of
the molecule (Fig. 3b). Thus, steric conflicts between
12 and residues in the binding site may also contribute
to the lack of activity as speculated for low-activity tax-
anes that can otherwise achieve the T-taxolbinding
conformation.18
7. (a) Barboni, L.; Data, A.; Gutta, D.; Georg, G. I.; Vander
Velde, D. G.; Himes, R. H.; Wang, M.; Snyder, J. P. J.
Org. Chem. 2001, 66, 3321–3329; (b) Deka, V.; Dubois, J.;
Thoret, S.; Gueritte, F.; Guenard, D. Org. Lett. 2003, 5,
5031–5034; (c) Beusker, P. H.; Veldhuis, H.; Brinkhorst,
J.; Hetterscheid, D. G. H.; Feichter, N.; Bugaut, A.;
Scheeren, H. W. Eur. J. Org. Chem. 2003, 689–705; (d)
Samaranayake, G.; Magri, N. F.; Jitrangsrei, C.; Kings-
ton, D. G. I. J. Org. Chem. 1991, 56, 5114–5119.
8. Barboni, L.; Giarlo, G.; Ricciutelli, M.; Ballini, R.; Georg,
G. I.; Vander Velde, D. G.; Himes, R. H.; Wang, M.;
Lakdawala, A.; Snyder, J. P. Org. Lett. 2004, 4, 461–464.
9. (a) Chaudhary, A. G.; Rimoldi, J. M.; Kingston, D. G. I.
J. Org. Chem. 1993, 58, 3798–3799; (b) Chen, S.-H.;
Huang, S.; Kant, J.; Fairchild, C.; Wei, J.; Farina, V. J.
Org. Chem. 1993, 58, 5028–5029.
To summarize, the described chemistry allows for the
efficient preparation of C,D-seco-taxoids. The scission
of the C4–C5 bond in the taxane core brings about
important modifications of both the chemicalreactivity
and the biological activity of the starting taxoid.
10. Ref. 1c, pp 165–253.
Acknowledgements
11. (a) Lucas, H. Arch. Pharm. 1856, 14, 438; (b) Appendino,
G. In Alkaloids: Chemical and Biological Perspectives;
Pelletier, S. W., Ed.; Pergamon: Oxford, 1996; Vol. 11, pp
237–268; (c) Jenniskens, L. H. D.; van Rosendaal, E. L.
M.; van Beek, T. A.; Wiegerinck, P. H. G.; Scheeren, H.
W. J. Nat. Prod. 1996, 59, 117–123; (d) Ettouati, L.;
Ahond, A.; Poupat, C.; Potier, P. J. Nat. Prod. 1991, 54,
1455–1458.
12. (a) Saicic, R. N.; Matovic, R. J. Chem. Soc., Perkin Trans.
1 2000, 59–65; (b) Matovic, R.; Saicic, R. N. J. Chem.
Soc., Chem. Commun. 1998, 1745–1746.
13. Ettouati, L.; Ahond, A.; Poupat, C.; Potier, P. Tetrahe-
dron 1991, 47, 9823–9838.
This work was financially supported by the Ministry of
Science, Technology, and Development (Project No.
1719). R.N.S. is gratefulto Dr. Francoise Gueritte and
Dr. DanielGuenard (Institut de Chimie des Substances
Naturelles, CNRS, France) for stimulating discussions,
to Dr. Sylviane Thoret for performing the tubulin test
on 12, as well as to Dr. Christiane Poupat and Dr. Alain
Ahond (from the same Institute) for a generous gift of
taxine extract.
14. Halgren, T. A. J. Comput. Chem. 1999, 20, 730–748, cf.
http://www.schrodinger.com/Products/macromodel.html.
15. Langley, C. H.; Lii, J.-H.; Allinger, N. L. J. Comput.
Chem. 2001, 22, 1451–1475.
16. Kanazawa, A. M.; Denis, J.-N.; Greene, A. E. J. Chem.
Soc., Chem. Commun. 1994, 2591–2592.
17. Lataste, H.; Senihl, V.; Wright, M.; Guenard, D.; Potier,
P. Proc. Natl. Acad. Sci. U.S.A. 1984, 81, 4090–4094.
18. (a) Barboni, L.; Lambertucci, C.; Appendino, G.; Vander
Velde, D. G.; Himes, R. H.; Bombardelli, E.; Wang, M.;
Snyder, J. P. J. Med. Chem. 2001, 44, 1576–1587; (b)
Metaferia, B. B.; Hoch, J.; Glass, T. E.; Bane, S. L.;
Chatterjee, S. K.; Snyder, J. P.; Lakdawala, A.; Cornett,
B.; Kingston, D. G. I. Org. Lett. 2001, 3, 2461–2464; (c)
For a generaldiscussion of the phenomenon, see: Kings-
ton, D. G. I.; Bane, S.; Snyder, J. P. Cell Cycle 2005, 279–
289.
Supplementary data
Supplementary data associated with this article can be
found, in the online version, at doi:10.1016/
j.tetlet.2005.05.071.
References and notes
1. (a) Kingston, D. G. I.; Jagtap, P. G.; Yuan, H.; Samala, L.
Prog. Chem. Org. Nat. Prod. 2002, 84, 53–225; (b)
Gueritte, F. Curr. Pharm. Des. 2001, 7, 933–953; (c)
The Chemistry and Pharmacology of Taxol and its
Derivatives, Pharmacochemistry library, Farina, V., Ed.;
Elsevier: Amsterdam, 1995; Vol. 22.