Synthesis and interactions of 7-deoxy-, 10-deacetoxy, and 10-deacetoxy-7-deoxypaclitaxel with NCI/ADR-RES cancer cells and bovine brain microvessel endothelial cells

Article abstract of DOI:10.1016/j.bmcl.2005.09.043

7-Deoxypaclitaxel, 10-deacetoxypaclitaxel and 10-deacetoxy-7- deoxypaclitaxel were prepared and evaluated for their ability to promote assembly of tubulin into microtubules, their cytotoxicity against NCI/ADR-RES cells and for their interactions with P-gl

Full text of DOI:10.1016/j.bmcl.2005.09.043

Bioorganic & Medicinal Chemistry Letters 16 (2006) 433–436
Synthesis and interactions of 7-deoxy-, 10-deacetoxy, and
10-deacetoxy-7-deoxypaclitaxel with NCI/ADR-RES cancer
cells and bovine brain microvessel endothelial cells
Haibo Ge,a Veena Vasandani,b Jacquelyn K. Huff,c Kenneth L. Audus,b
Richard H. Himes,c Anna Seeligd and Gunda I. Georga,*
aDepartment of Medicinal Chemistry, University of Kansas, 1251 Wescoe Hall Drive, Lawrence, KS 66045-7582, USA
bDepartment of Pharmaceutical Chemistry, University of Kansas, Lawrence, KS 66045-7582, USA
cDepartment of Molecular Biosciences, University of Kansas, Lawrence, KS 66045-7582, USA
dDepartment of Biophysical Chemistry, Biocenter of the University of Basel, Klingelbergstrasse 70, 4056 Basel, Switzerland
Received 16 August 2005; revised 13 September 2005; accepted 15 September 2005
Available online 3 November 2005
Abstract—7-Deoxypaclitaxel, 10-deacetoxypaclitaxel and 10-deacetoxy-7-deoxypaclitaxel were prepared and evaluated for their
ability to promote assembly of tubulin into microtubules, their cytotoxicity against NCI/ADR-RES cells and for their interactions
with P-glycoprotein in bovine brain microvessel endothelial cells. The three compounds were essentially equivalent to paclitaxel in
cytotoxicity against NCI/ADR-RES cells. They also appeared to interact with P-glycoprotein in the endothelial cells with the two
10-deacetoxy compounds having less interaction than paclitaxel and 7-deoxypaclitaxel.
Ó 2005 Elsevier Ltd. All rights reserved.
O
Paclitaxel (1, Fig. 1), a cytotoxic agent from the bark of
the Pacific yew (Taxus brevifolia),1 exerts its activity by
2
OR
O
1
R
NH
O
10
OH
O
altering tubulin dynamics.2 Paclitaxel as well as its semi-
synthetic analogue docetaxel (2, Fig. 1) are effective
agents for the treatment of ovarian and breast cancer,
KaposiÕs sarcoma, and non-small cell lung cancer.3
However, the continued evaluation of new analogues
is important, because drug resistance has developed,
and paclitaxel, like many other anticancer agents, does
not cross the blood–brain barrier (BBB). Active efflux
by P-glycoprotein (Pgp) is believed to be responsible
for the lack of brain uptake of many anticancer drugs,
leaving brain cancer patients with few viable treatment
options.4,5 PaclitaxelÕs limited oral bioavailability, lack
of accumulation in brain tissues, and efficacy in brain tu-
mors have been attributed to active efflux by Pgp.6,7 In
our efforts to develop paclitaxel analogues that can cross
the BBB,8 we are studying the influence of chemical
modifications of paclitaxel on Pgp in the BBB and are
2'
3'
13
O
7
5
2
OH
H
HO
BzO
AcO
Figure 1. Structures of paclitaxel (1, R1 = Ph, R2 = Ac) and docetaxel
(2, R1 = O-tert-Bu, R2 = H).
now reporting new results concerning structure–efflux
relationships of paclitaxel analogues.
A comparison of the structural features of over one
hundred known substrates and modulators for Pgp
suggested that clusters of hydrogen bond acceptors
(electron-donating groups), arranged in fixed spatial
distances from each other, are required for recognition
by Pgp-binding sites.9–11 The recognition elements are
formed either by two or by three electron-donating
groups. It was also hypothesized that the number and
strength of the hydrogen bonds present in a molecule
determine Pgp affinity. This implies that one could
remove recognition elements from the molecule that
are not necessary for biological activity and improve
Keywords: 7-Deoxypaclitaxel; 10-Deacetoxypaclitaxel; 10-Deacetoxy-
7-deoxypaclitaxel; Synthesis; Bovine brain microvessel endothelial ce-
lls; NCI/ADR-RES; P-glycoprotein interaction.
*
Corresponding author. Tel.: +1 785 864 4498; fax: +1 785 864
5836; e-mail: georg@ku.edu
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.043

434
H. Ge et al. / Bioorg. Med. Chem. Lett. 16 (2006) 433–436
1
Based on these previous observations, we decided to ex-
6
plore additional chemical modifications of the putative
group 5 and 6 paclitaxel recognition elements with the
goal of decreasing the strength for Pgp binding and
thereby improving activity against MDR cancer cells
and enhancing BBB permeation. We therefore prepared
and evaluated paclitaxel analogues lacking the C7
hydroxy group and/or C10 acetoxy groups.
O
O
2
O
Group 1 Aa (1.3) weak
Group 2 AA (1.3)
Group 3 AA (1.3)
Group 4 AA (1.3)
Group 5 AA (1.5)
Group 6 AAA (1.4.6)
O
NH
O
10
OH
O
9
O
7
5
5
OH
H
HO
O
AcO
4
O
3
10-Deacetoxypaclitaxel was prepared in one step via
samarium diiodide-mediated deoxygenation.16,17 Treat-
ment of paclitaxel (1) with SmI2 and acetic acid provid-
ed 10-deacetoxypaclitaxel (3) in 91% yield (Scheme 1).
Figure 2. Analysis of paclitaxel for type I (1.3) and type II (1.5 and
1.4.6) interaction sites with Pgp.
activity against multi-drug resistant cancers and
achieve BBB penetration.
7-Deoxypaclitaxel was prepared, employing the Barton
deoxygenation reaction, as previously described by the
Kingston group (Scheme 2).18 Treatment of paclitaxel
(1) with tert-butyldimethylsilyl chloride, imidazole, and
4-(dimethylamino)pyridine furnished 20-(tert-butyldi-
methylsilyl)paclitaxel (4) in 85% yield. Compound 4
was treated with 2.2 equiv of sodium hydride in dry
THF, then excess carbon disulfide and iodomethane to
provide 5 in 65% yield. Treatment of 5 with tributyltin
We analyzed paclitaxel for Pgp recognition patterns and
determined six relevant hydrogen bond acceptor groups
(Fig. 2). Groups 1–4 (Fig. 2) are so-called type I units
that are formed by two hydrogen bond acceptor groups
˚
(A or a) with a spatial separation of 2.5 0.3 A. Strong
electron-donating groups, containing oxygens, are
denoted as ÔAÕ and less potent donors, such as tertiary
amines, halides, sulfides or p-electron-donating groups,
as Ôa.Õ In addition, two so-called type II units, groups 5
and 6, were identified. Type II units are formed either
by two hydrogen bond acceptor groups with a spatial
O
O
Ph
NH
O
OH
O
˚
separation of 4.6 0.6 A (group 5: AA (1.5) pattern be-
tween the 7-hydroxyl and the 5-ether group of the oxe-
tane moiety) or by three hydrogen bond acceptor
a
O
1
OH
HO
BzO
H
AcO
˚
groups separated from each other by 2.5 0.3 A, with
the outer two acceptor groups at a distance of
˚
4.6 0.6 A (group 6: AAA (1.4.6) relationship between
3
Scheme 1. Reagents: (a) SmI2, AcOH (91%).
the C9 carbonyl and the C10 acetate). The type I units
are found in all Pgp substrates and type II units are pres-
ent in all Pgp inducers and many Pgp substrates.
O
OAc
O
Ph
NH
O
OH
b
According to structure–activity relationship studies, nei-
ther the C7 hydroxy group nor the C10 acetyl group is
essential for bioactivity.12 Substitution at C7 and C10
is well tolerated,12 and it has been demonstrated that
these regions of the molecule are important for Pgp
affinity.13 Ojima and collaborators have shown that spe-
cific modifications at the C10 position (i.e., propionate,
instead of acetate, at O10) of 30-dephenyl-30-isobutyl-
paclitaxel and related analogues resulted in a significant
increase in cytotoxicity, especially against NCI/ADR-
RES cells, expressing the multi-drug resistant (MDR)
phenotype. This result was ascribed to an effective inhi-
bition of binding to Pgp by these analogues.13 The ef-
fects are complex, however, and replacement of the 10-
acetate with a 10-propionate group in paclitaxel does
not lead to an increase in cytotoxicity against MDR-ex-
pressing cancer cells.14 In another example, we have
recently shown that Tx-67, a paclitaxel analogue, carry-
ing a 10-O-succinyl monoester moiety, is able to evade
Pgp-mediated efflux and can permeate the BBB in vitro
and in situ.8 Of interest in this context are also recent re-
ports, disclosing that TXD-258 (7-O-methyl-10-O-meth-
yldocetaxel) is able to cross the BBB, and that RPR-
109881A (a 7,8-cyclopropyl docetaxel analogue) is effec-
tive against MDR-positive and taxane resistant human
tumor xenografts.15
a
O
1
OTBS
HO
BzO
H
O
AcO
4
O
OAc
S
O
Ph
NH
O
O
O
c
SMe
O
OTBS
HO
BzO
H
AcO
5
O
OAc
O
Ph
Ph
NH
O
d
O
OH
HO
BzO
H
AcO
O
6
O
O
NH
O
O
OH
HO
BzO
H
AcO
O
7
Scheme 2. Reagents: (a) TBSCl, imidazole, DMAP, DCM (85%); (b)
NaH, CS2, MeI, THF (65%); (c) AIBN, Bu3SnH, PhH; then HF-Py,
Py (48% overall); (d) SmI2, AcOH (82%).

H. Ge et al. / Bioorg. Med. Chem. Lett. 16 (2006) 433–436
435
Table 1. Biological evaluation of deoxypaclitaxel analogues in a
tubulin assembly assay and for antiproliferative activity against MCF7
and NCI/ADR-RES cell lines14
litaxel against HCT 116 cells, while 7 was slightly less
active.19,20
We assessed the analogues for rhodamine 123 uptake
into bovine brain microvessel endothelial cells
(BMECs).8 The intracellular concentration of rhoda-
mine 123, itself a substrate for Pgp, is enhanced by com-
pounds that compete for binding with Pgp in BMECs.
Analogues 3, 6, and 7 were compared to paclitaxel at
two concentrations and to cyclosporin A, as positive
controls (Fig. 3). All compounds increased the level of
rhodamine 123 uptake, indicating that they interact with
Pgp. At 5 lM, compounds 3 and 7 interacted less than
paclitaxel and 6.
O
1
R
O
Ph
NH
O
2
R
O
OH
HO
BzO
H
AcO
O
Compound21 R1
R2
MTa MCF7b NCI/ADR-RESc
3
6
7
H
OH 2.2
0.87
30.0
9.0
0.79
1.0
OAc
H
H
H
1.4
2.0
0.21
The lack of large differences between paclitaxel 3, 6, and 7
as cytotoxic agents against NCI/ADR-RES cells indicates
that removal of the group 5 and 6 (type II) recognition ele-
ments in paclitaxel did not reduce interactions between
the paclitaxel analogues and Pgp in the MDR cancer cell
line. Apparently, the interactions of the remaining group
1–4 recognition elements of the paclitaxel analogues 3, 6,
and 7 with Pgp are sufficient enough for effective binding
to cause efflux from MDR cancer cells.
a Microtubule assembly ED50/ED50(Paclitaxel)
b MCF7ED50/ED50(Paclitaxel)
c NCI/ADR-RES22 ED50/ED50(Paclitaxel)
.
.
.
hydride and AIBN in toluene at 80 °C, followed by
deprotection of 20-(tert-butyldimethylsilyl), yielded 7-
deoxypaclitaxel (6) in 48% from 5.
In the course of these studies, we also developed an
efficient method to convert paclitaxel into 10-deacet-
oxy-7-deoxypaclitaxel (7, Scheme 2). Treatment of
7-deoxypaclitaxel (6) with SmI2 in the presence of AcOH
furnished 7 in 82% yield. This method is superior to the
one reported earlier for the synthesis of 7, which
produced a low overall yield (5%).19
The effect in BMECs is concentration dependent.
Paclitaxel and the three analogues increased rhodamine
123 uptake into BMECs at 10 lM. However, at 5 lM,
compounds 3 and 7, which both lack the C10 acetoxy
group, had reduced uptake of rhodamine 123 in com-
parison to paclitaxel and compound 6, which lacks the
C7 hydroxyl group but carries the C10 acetoxy group.
The results demonstrate the importance of the group 6
recognition element for paclitaxel interaction with Pgp
in BMECs. Compounds 3 and 7 are nevertheless sub-
strates for Pgp because uptake of rhodamine 123 is
increased in comparison to the rhodamine 123 control.
As was the case with the MDR cancer cell line, the
group 1–4 recognition elements are sufficient to increase
rhodamine 123 uptake in comparison to the control.
Overall, our original hypothesis, that removal of Pgp
recognition elements could decrease the strength for
Pgp binding, was verified, because rhodamine 123
uptake was clearly reduced for compounds 3 and 7 in
BMECs.
The three deoxygenated analogues 3, 6, and 7 were test-
ed in a tubulin assembly assay and as inhibitors of pro-
liferation against the MCF7 and NCI/ADR-RES breast
cancer lines (Table 1).14 In the tubulin assembly assay,
analogue 6 was just slightly less active than paclitaxel,
while 3 and 7 were about half as active as paclitaxel.
Compound 3 had activity similar to paclitaxel against
MCF7 proliferation but 6 and 7 had significantly re-
duced activity. Differences between the three analogues
and paclitaxel against the resistant cell line NCI/ADR-
RES were minimal. In a previous report, it was found
that 3 and 6 had essentially the same cytotoxicity as pac-
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Con CsA
Tx
3
6
7
Tx
3
6
7
10 µM
5 µM
Figure 3. Rhodamine 123 (5 lM) uptake into BMECs alone (Con = control) and in the presence of cyclosporin A (CsA) at 5 lM, and paclitaxel (Tx)
and paclitaxel analogues 3, 6, and 7 at 10 and 5 lM.8

436
H. Ge et al. / Bioorg. Med. Chem. Lett. 16 (2006) 433–436
11. For review: Seelig, A.; Landwojtowicz, W.; Fisher, H.;
Acknowledgments
Blatter, X. L. Methods Princ. Med. Chem. 2003, 18,
461.
The authors thank the National Cancer Institute for
financial support of this research (NIH CA82801) and
Tapestry Pharmaceuticals (Boulder, CO) for a generous
gift of paclitaxel.
12. For review: Georg, G. I.; Boge, T. C.; Cheruvallath,
Z. S.; Clowers, J. S.; Harriman, G. C. B.; Hepperle,
M.; Park, H. In Taxol Scienceâ and Applications;
Suffness, M, Ed.; CRC Press: Boca Raton, FL, 1995;
p 317.
13. Ojima, I.; Slater, J. C.; Michaud, E.; Kuduk, S. D.;
Bounaud, P.-Y.; Vrignaud, P.; Bissery, M. C.; Veith, J.
M.; Pera, P.; Bernacki, R. J. J. Med. Chem. 1996, 39,
3889.
14. Liu, Y.; Boge, T. C.; Victory, S.; Ali, S. M.; Zygmunt, J.;
Georg, G. I.; Marquez, R. T.; Himes, R. H. Comb. Chem.
High Throughput Screening 2002, 5, 39.
References and notes
1. Wani, M. C.; Taylor, H. L.; Wall, M. E.; Coggon, P.;
McPhail, A. T. J. Am. Chem. Soc. 1971, 93, 2325.
2. For review: Zhou, J.; Giannakakou, P. Curr. Med. Chem.
Anticancer Agents 2005, 5, 65.
3. Spencer, C. M.; Faulds, D. Drugs 1994, 48, 794.
4. Schinkel, A. H.; Smit, J. J.; van Tellingen, O.; Beijnen, J.
H.; Wagenaar, E.; van Deemter, L.; Mol, C. A.; van der
Valk, M. A.; Robanus-Maandag, E. C.; te Riele, H. P.
Cell 1994, 77, 491.
5. Schinkel, A. H.; Wagenaar, E.; Mol, C. A.; van Deemter,
L. J. Clin. Invest. 1996, 97, 2517.
6. Sparreboom, A.; van Asperen, J.; Mayer, U.; Schinkel, A.
H.; Smit, J. W.; Meijer, D. K.; Borst, P.; Nooijen, W. J.;
Beijnen, J. H.; van Tellingen, O. Proc. Natl. Acad. Sci.
U.S.A. 1997, 94, 2031.
15. For review: Becker, T.; Mahboobi, S. Drugs Future 2003,
28, 767.
16. Georg, G. I.; Cheruvallath, Z. S. J. Org. Chem. 1994, 59,
4015.
17. Holton, R. A.; Somoza, C.; Chai, K. B. Tetrahedron Lett.
1994, 35, 1665.
18. Chaudhary, A. G.; Rimoldi, J. M.; Kingston, D. G. I. J.
Org. Chem. 1993, 58, 3798.
19. Chen, S. H.; Huang, S.; Kant, J.; Fairchild, C.; Wei, J.;
Farina, V. J. Org. Chem. 1993, 58, 5028.
20. Chen, S. H.; Fairchild, C.; Mamber, S. W.; Farina, V. J.
Org. Chem. 1993, 58, 2927.
21. All compounds showed spectroscopic properties in agree-
ment with their structures.
22. This cell line has been renamed NCI/ADR-RES as DNA
fingerprinting has shown it is not derived from the MCF7
cell line: Scudiero, D. A.; Monks, A.; Sausville, E. A. J.
Natl. Cancer Inst. 1998, 90, 862.
7. Brouty-Boye, D.; Kolonias, D.; Wu, C. J.; Savaraj, N.;
Lampidis, T. J. Cancer Res. 1995, 55, 1633.
8. Rice, A.; Liu, Y.; Michaelis, M. L.; Himes, R. H.; Georg,
G. I.; Audus, K. J. Med. Chem. 2005, 48, 832.
9. Seelig, A. Eur. J. Biochem. 1998, 251, 252.
10. Seelig, A.; Li Blatter, X.; Wohnsland, F. Int. J. Clin.
Pharmacol. Ther. 2000, 38, 111.