Synthesis and microtubule binding of fluorescent paclitaxel derivatives

Article abstract of DOI:10.1016/S0960-894X(01)00453-X

The preparation of two new fluorescent derivatives of paclitaxel in which the fluorophore is bonded to paclitaxel at the C-10 position is reported. Both analogues, 10-deacetyl-10-(m-aminobenzoyl)paclitaxel (1, BTax) and 10-deacetyl-10-[7-(diethylamino) co

Full text of DOI:10.1016/S0960-894X(01)00453-X

Bioorganic& Medicinal Chemistry Letters 11 (2001) 2249–2252
Synthesis and Microtubule Binding of Fluorescent Paclitaxel
Derivatives
Erkan Baloglu,a David G. I. Kingston,a Pruthvi Patel,b Sabarni K. Chatterjeeb
and Susan L. Baneb,*
aDepartment of Chemistry, M/C 0212, Virginia Polytechnic Institute and State University, Blacksburg, VA 24061, USA
bDepartment of Chemistry, State University of New York at Binghamton, Binghamton, NY 13902-6016, USA
Received 8 February 2001; accepted 18 May 2001
Abstract—The preparation of two new fluorescent derivatives of paclitaxel in which the fluorophore is bonded to paclitaxel at the
C-10 position is reported. Both analogues, 10-deacetyl-10-(m-aminobenzoyl)paclitaxel (1, BTax) and 10-deacetyl-10-[7-(diethyla-
mino) coumarin-3-carbonyl]paclitaxel (2, CTax) retain good activity as promoters of in vitro tubulin assembly. Microtubule bind-
ing enhances the emission intensity of both probes. # 2001 Elsevier Science Ltd. All rights reserved.
The complex natural product paclitaxel (Taxol1) (3),
first isolated from Taxus brevifolia,1 is a member of a
large family of taxane diterpenoids.2 Paclitaxel is widely
used to treat solid tumors, particularly those of the
breast and ovaries. The drug exerts its pharmacological
effects by interacting with cellular microtubules.3
Microtubules are primarily composed of tubulin, a 100
kDal heterodimer that reversibly assembles to form the
core of the microtubule. Proper spatial and temporal
control of microtubule dynamics is essential for cellular
homeostasis. Paclitaxel disrupts the dynamic processes
of microtubules by binding to the polymer at a single
site on b-tubulin.4
BODIPY) and their fluorescence emission spectra
are relatively insensitive to their environment. These
fluorophores are most commonly attached to paclitaxel
through the C-7 position,9ꢀ13 the most synthetically
accessible site on the ‘northern hemisphere’ of paclitaxel
(C-6–C-12 positions). Structural changes in this portion
of the molecule appear to have less impact on its anti-
microtubule activity than modifications in the ‘southern
hemisphere’.14
We have been probing the nature of the paclitaxel–
tubulin interaction through synthesis and in vitro bio-
logical investigation of fluorescent paclitaxel derivatives.
In our previous studies, we prepared and investigated
paclitaxels in which the fluorophore is contained in the
‘southern hemisphere’ (i.e., at C-2 and C-30).7,8 We
chose the less utilized C-10 position to initiate our
investigations into the ‘northern hemisphere’ of pacli-
taxel. Moreover, we chose to use 7-(diethylamino)cou-
marin-3-carboxylic acid as one of the labels. This
fluorophore has not yet been used as a probe for the
paclitaxel–tubulin interaction. Other coumarin deriva-
tives have been used to label paclitaxel, but the resulting
probes had absorption maxima at high energy
(ꢁ370 nm).10,13 Inclusion of an electron withdrawing
group at the 3-position of 7-(diethylamino)coumarin
yields a probe that is excited by visible light (410–
430 nm) and is also environmentally sensitive.15 Repor-
ted herein are the synthesis of two C-10 modified fluor-
escent paclitaxels and their in vitro tubulin binding
properties.
Fluorescence spectroscopy is a powerful and versatile
tool for the study of biological systems and has been
extensively applied to the study of paclitaxel–micro-
tubule interactions. Small, environmentally sensitive
probes based on the aminobenzoate fluorophore have
been used to explore the local environment of the pacli-
taxel binding site.5ꢀ8 This fluorophore, however, must
be excited with ultraviolet light, which limits its useful-
ness. Probes that are exited by visible light are more
desirable. They are best suited for visualization of
microtubule-bound paclitaxel using fluorescence micro-
scopy and can be easily detected in plate readers. These
structures tend to be large (rhodamine, fluorescein, and
*Corresponding author. Tel.: +1-607-777-2927; fax: +1-607-777-
4478; e-mail: sbane@binghamton.edu
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00453-X

2250
E. Baloglu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2249–2252
Results
are both in the visible region of the electromagnetic
spectrum. The emission intensity is very strongly affec-
ted by environmental polarity, while the emission
energy is less so (Table 1). Microtubule binding by
CTax is accompanied by a significant increase in emis-
sion intensity (Fig. 2B). The appearance of fluorescence
coincides with assembly of the protein (Fig. 3), which
further demonstrates that the fluorescence changes that
occur in CTax in the presence of tubulin are due to
microtubule binding.
Synthesis of fluorescent paclitaxels modified at the C-10
position
The sequence leading to the required fluorescent pacli-
taxels is shown in Scheme 1. The known paclitaxel
derivative (4)16 was esterified at the C-10 position with
3-nitrobenzoic acid, employing the carbodiimide-based
coupling protocol to yield the paclitaxel analogue 5.
Concomitant removal of both silyl protecting groups,
followed by hydrogenation, gave the desired amine (1,
BTax) in good yield. CTax (2) was prepared from the
common intermediate 4 by acylation at C-10 as before
utilizing 7-(diethylamino)coumarin-3-carboxylic acid,
followed by simultaneous removal of the protecting
groups.17
Microtubule activity
The activities of the ligands were assessed by their abil-
ities to induce purified tubulin to assemble in vitro. Both
Btax and CTax were nearly equipotent to paclitaxel in
promoting tubulin assembly, although CTax did not
promote assembly quite as rapidly as paclitaxel (Fig. 1).
The absorption, excitation and emission spectra of the
two fluorescent ligands were obtained in a variety of
solvents and bound to microtubules. The effect of
microtubule binding on the emission spectra of the
compounds is illustrated in Figure 2. The emission
intensity of BTax increased and underwent a small blue
shift upon microtubule binding (Fig. 2A). Since the
absorption maximum of the ligand in buffer is near
320 nm, ultraviolet radiation is required for excitation
of this fluorophore. CTax possesses photochemical
properties that will make it more useful than BTax as a
fluorescent probe. The absorption and emission maxima
Figure 1. Promotion of tubulin assembly by paclitaxel, BTax and
CTax. Tubulin in PMEG buffer (0.1 M Pipes, 1 mM MgSO4, 2 mM
EGTA and 0.1 mM GTP, pH 6.9) was equilibrated to 37 ꢂC prior to
addition of the ligand. Assembly was monitored by apparent light
scattering (absorption at 350 nm). Solid curve: 10 mM BTax; dashed
curve: 10 mM paclitaxel. Dash dot dot curve: 5 mM paclitaxel; dotted
curve: 5 mM CTax.
Scheme 1. Synthesis of fluorescent paclitaxel derivatives.

E. Baloglu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2249–2252
2251
Figure 3. Promotion of tubulin assembly by CTax. Tubulin (10 mM) in
PMEG buffer was equilibrated to 37 ꢂC prior to addition of 10 mM
CTax. Assembly was monitored by apparent light scattering (dotted
curve) or by emission at 465 nm (solid curve; excitation at 420 nm).
known that tubulin has at least two arginine residues
located on the exterior of the paclitaxel binding site,19
and at least one of these is in close proximity to the C-7
position of the tubulin-bound ligand.20 The spectro-
scopic data for microtubule-bound Btax and CTax are
also consistent with the presence of a charged amino
acid in close proximity to the fluorophore.
Figure 2. Emission spectra of BTax and CTax in the presence and
absence of tubulin. (A) Solid curve: BTax (10 mM) in PMEG buffer
was incubated with 10 mM tubulin at 37 ꢂC prior to collection of the
emission spectrum. Dashed curve: Emission spectrum of 10 mM BTax
in PMEG buffer. The excitation wavelength was 320 nm. (B) Solid
curve: CTax (0.4 mM) in PMEG buffer was incubated with 5 mM
tubulin at 37 ꢂC prior to collection of the emission spectrum. Dashed
curve: Emission spectrum of 0.4 mM CTax in PMEG buffer. The
excitation wavelength was 420 nm.
Conclusion
The C-10 position can be used to prepare fluorescent
derivatives of paclitaxel that retain good in vitro
assembly-promoting activity. The small change in emis-
sion intensity when BTax binds to microtubules limits
its utility as a probe. The 7-(diethylamino)coumarin-3-
carbonyl fluorophore, which has not been previously
used to fluorescently label paclitaxel, produces a probe
with many possible utilities. Since the probe can be
excited with visible light, the fluorescent paclitaxel-
microtubule association can be readily observed in
standard plate readers and possibly in live cells. Unlike
the rhodamine and fluorescein paclitaxel derivatives,13
the emission intensity of the diethylaminocoumarin
fluorophore in CTax undergoes a large change in emis-
sion intensity in upon microtubule binding. Thus,
background fluorescence due to unbound ligand will be
low when CTax is used as a probe.
Table 1. Absorption and emission maxima and relative emission
intensity of CTax in solvent and bound to microtubules
Solvent
Absorption
Emission
Relative
maximum (nm) maximum (nm) fluorescence
intensitya
Dioxane
413
415
427
422
423
424
428b
430c
447
452
469
464
461
462
462b
468
42
40
18
Ethyl acetate
Dimethylsulfoxide
Acetonitrile
Ethanol
Methanol
2% DMSO/water
Microtubules
1.0
0.11
0.27
0.06
4.8d
aFor solvents: intensity at emission maximum, normalized to acetoni-
trile value. The absorptivity of the solvent samples at the excitation
wavelength (420 nm) was equal. For microtubules: intensity of bound
drug relative to free drug.
bWithout DMSO cosolvent and at higher concentrations the ligand
undergoes self-association. The absorption maximum of aggregated
samples is 434 nm and the emission maximum is 530 nm.
cApproximate value due to turbidity of microtubule-containing solu-
tion.
dConcentration of free drug at 0.4 mM, concentration of bound drug at
0.4 mM in presence of 5 mM tubulin.
Acknowledgements
We thank Sergey Osipov for technical assistance, Dr.
Barbara Poliks for performing the molecular modeling
and the Nebraska Center for Mass Spectrometry at the
University of Nebraska for mass spectrometric assis-
tance.
The small change in emission energy observed when
BTax and CTax bind to microtubules indicates that the
substituent is solvent accessible. This observation is
consistent with the structure–activity data for C-10
analogues of paclitaxel.18 Based on their spectroscopic
observations, Evangelio et al.13 and Sengupta et al.6
hypothesized that the C-7 and C-10 position of pacli-
taxel may be in a cationic microenvironment. These
predictions were made before the electron crystal-
lographicstructure of tubulin was published. 19 It is now
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. Baloglu, E.; Kingston, D. G. I. J. Nat. Prod. 1999, 62,
1448.
3. Horwitz, S. B. Ann. Oncol. 1994, 5, S3.

2252
E. Baloglu et al. / Bioorg. Med. Chem. Lett. 11 (2001) 2249–2252
4. Jordan, M. A.; Wilson, L. Curr. Opin. Cell Biol. 1998, 10, 123.
5. Sengupta, S.; Boge, T. C.; Georg, G. I.; Himes, R. H. Bio-
chemistry 1995, 34, 11889.
6. Sengupta, S.; Boge, T. C.; Liu, Y.; Hepperle, M.; Georg,
G. I.; Himes, R. H. Biochemistry 1997, 36, 5179.
7. Han, Y.; Chaudhary, A. G.; Chordia, M. D.; Sackett, D. L.;
Perez-Ramirez, B.; Kingston, D. G.; Bane, S. Biochemistry
1996, 35, 14173.
8. Li, Y.; Poliks, B.; Cegelski, L.; Poliks, M.; Gryczynski, Z.;
Piszczek, G.; Jagtap, P. G.; Studelska, D. R.; Kingston,
D. G. I.; Schaefer, J.; Bane, S. Biochemistry 2000, 39, 281.
9. Dubois, J.; Le, G. M. T.; Gueritte-Voegelein, F.; Guenard,
D.; Tollon, Y.; Wright, M. Bioorg. Med. Chem. 1995, 3, 1357.
10. Souto, A. A.; Acuna, A. U.; Andreu, J. M.; Barasoain, I.;
Abal, M.; Amat-Guerri, F. Angew. Chem., Int. Ed. Engl. 1995,
34, 2710.
11. Guy, R.; Scott, Z.; Sloboda, R.; Nicolaou, K. Chem. Biol.
1996, 3, 1021.
12. Rao, C. S.; Chu, J. J.; Liu, R. S.; Lai, Y. K. Bioorg. Med.
Chem. 1998, 6, 2193.
13. Evangelio, J. A.; Abal, M.; Barasoain, I.; Souto, A. A.;
Lillo, M. P.; Acuna, A. U.; Amat-Guerri, F.; Andreu, J. M.
Cell Motil. Cytoskeleton 1998, 39, 73.
14. Kingston, D. G. I. J. Nat. Prod. 2000, 63, 726.
15. Bangar Raju, B.; Varadarajan, T. S. J. Phys. Chem. 1994,
98, 8903.
16. Paclitaxel derivative 4 was prepared by protecting the
known 10-deacetyl-20-(tert-butyldimethylsilyl) paclitaxel as its
triethylsilyl ether at C-7 position. Datta, A.; Hepperle,
M. I. G. G. J. Org. Chem. 1995, 60, 761.
17. General procedure for esterification at the C-10 position: 1
equivalent of the carboxylic acid was stirred in toluene at
room temperature with 1 equivalent of dicyclohexyl carbodi-
imide (DCC) or 1-[3-(dimethylamino)propyl-3-ethylcarbodi-
imide hydrochloride (EDC) for 15 min. 0.01 equivalent of 4-
(dimethylamino)pyridine (DMAP) was added and stirred for
5 min. 0.1 equivalent of 10-deacetyl-20-(tert-butyldimethylsi-
lyl)-7-(triethylsilyl) paclitaxel was then introduced and stirred
at rt for 24–48 h. In order to overcome the solubility problem
of 7-(diethylamino)coumarin-3-carboxylic acid, CH2Cl2/DMF
(10:1) solvent system was used for 2. The progress of the
reaction was monitored by thin layer chromatography and
when the reaction was complete, the reaction mixture was
diluted with EtOAc. The organic layer washed with water
(ꢃ2), NaHCO3 (ꢃ2), brine and dried over sodium sulfate. The
crude product was applied on a preparative thin layer chro-
matography plate, developed with solvent system of 40%
EtOAc/hexane to give the desired products. 1H NMR of 1
(400 MHz, CDCl3) d 8.12–8.14 (d, 2H), 7.72–7.48 (d, 2H), 7.61
(t, 1H), 7.34–7.53 (m, 12H), 7.24 (t, 1H), 7.02–7.05 (d, 1H),
6.88–6.90 (dd, 1H), 6.49 (s, 1H), 6.24 (t, 1H), 5.77–5.80 (dd,
1H), 5.70–5.72 (d, 1H), 4.95–4.97 (d, 1H), 4.80 (s, 1H), 4.47
(m, 1H), 4.30–4.32 (d, 1H), 4.19–4.21 (d, 1H), 3.85–3.87 (d,
1H), 3.82 (bs, 2H), 3.62 (bs, 1H), 2.66–2.67 (d, 1H), 2.57 (m,
1H), 2.39 (s, 3H), 2.34 (m, 1H), 1.89 (m, 1H), 1.81 (s, 3H), 1.69
(s, 3H), 1.66 (s, 3H), 1.30 (s, 3H), 1.24 (s, 3H); 13C NMR of 1
(100 MHz) d 203.45, 172.66, 170.38, 167.05, 166.97, 166.57,
146.56, 142.07, 137.89, 133.72, 133.59, 133.24, 131.96, 130.19,
129.97, 129.41, 129.13, 129.01, 128.72, 128.69, 128.37, 127.02,
120.17, 120.08, 115.99, 84.44, 81.18, 79.04, 76.49, 75.85, 74.95,
73.17, 72.38, 72.25, 58.66, 55.03, 45.72, 43.22, 35.72, 35.68,
29.68, 27.04, 22.63, 22.07, 14.88, 9.57. HR-FABMS of 1 m/z
found 953.343872 (M+Na)+, calcd 953.3473 (M+Na)+; LR-
FABMS m/z found 953.5 (M+Na)+. 1H NMR of
2
(400 MHz, CDCl3) d 8.48 (s, 1H), 8.12–8.14 (d, 2H), 7.74–7.76
(d, 2H), 7.61 (t, 1H), 7.34–7.53 (m, 11H), 7.05–7.07 (d, 1H),
6.60–6.63 (dd, 1H), 6.52 (s, 1H), 6.45–6.46 (d, 1H), 6.25 (t,
1H), 5.78–5.80 (dd, 1H), 5.70–5.72 (d, 1H), 4.95–4.97 (d, 1H),
4.801–4.807 (d, 1H), 4.40–4.48 (m, 1H), 4.30–4.32 (d, 1H),
4.20–4.22 (d, 1H), 3.86–3.88 (d, 1H), 3.43–3.49 (q, 4H), 2.57
(m, 1H), 2.40 (s, 3H), 2.30–2.34 (m, 3H), 1.87–1.94 (m, 2H),
1.82 (s, 3H), 1.71 (s, 3H), 1.32 (s, 3H), 1.23–1.28 (m, 15H); 13
C
NMR of 2 (100 MHz) d 203.65, 172.56, 170.40, 166.98, 164.32,
158.80, 153.23, 150.22, 141.79, 137.89, 133.68, 133.63, 133.18,
131.92, 131.47, 130.19, 129.16, 128.99, 128.68, 128.34, 127.04,
109.73, 107.84, 107.37, 69.77, 84.44, 81.18, 79.04, 76.49, 75.73,
74.96, 73.20, 72.38, 72.19, 58.69, 55.01, 45.79, 45.19, 43.27,
35.74, 35.66, 26.57, 22.63, 21.66, 14.92, 12.43, 9.58 LR-
FABMS (M+H)+ m/z found 1055.6.
18. Kant, J.; O’Keeffe, W. S.; Chen, S.-H.; Farina, V.; Fair-
child, C.; Johnson, K.; Kadow, J. F.; Long, B. H.; Vyas, D.
Tetrahedron Lett. 1994, 35, 5543.
19. Nogales, E.; Wolf, S. G.; Downing, K. H. Nature 1998,
391, 199.
20. Rao, S.; He, L. F.; Chakravarty, S.; Ojima, I; Orr, G. A.;
Horwitz, S. B. J. Biol. Chem. 1999, 274, 37990.