Antitumor agents 216. Synthesis and evaluation of paclitaxel-camptothecin conjugates as novel cytotoxic agents

Article abstract of DOI:10.1016/S0968-0896(03)00040-3

Five conjugates (16-20) composed of a paclitaxel and a camptothecin derivative joined by an imine linkage were synthesized and evaluated as cytotoxic agents and as inhibitors of DNA topoisomerase I. All of the conjugates were potent inhibitors of tumor cell replication with improved activity relative to camptothecin. Significantly, compounds 16-18 were more active than paclitaxel and camptothecin against HCT-8 (colon adenocarcinoma) cell replication, and the spectrum of activity was different from a simple mixture of paclitaxel and camptothecin. All of the conjugates were significantly less potent than camptothecin as inhibitors of human topoisomerase I in vitro with 16, 18, and 19 showing only marginal activity at 50 μM. Based on activity against drug-resistant cell line replication, one could conclude that the conjugates are simply acting as 'weak taxanes', but the spectrum of activity, particularly against MCF-7 and HCT-8, strongly suggests that a novel mechanism of action has been achieved through conjugation.

Full text of DOI:10.1016/S0968-0896(03)00040-3

Bioorganic & Medicinal Chemistry 11 (2003) 1851–1857
Antitumor Agents 216. Synthesis and Evaluation of
Paclitaxel–Camptothecin Conjugates as Novel Cytotoxic Agents1
Hironori Ohtsu,a Yuka Nakanishi,a Kenneth F. Bastow,a
Fang-Yu Leeb and Kuo-Hsiung Leea,*
aNatural Products Laboratory, School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA
bYung-Shin Pharmaceutical Industrial Company, 1191, Section 1, Chung-Shan Road, Taichia, Taichung, Taiwan
Received 18July 2002; accepted 27 November 2002
Abstract—Five conjugates (16–20) composed of a paclitaxel and a camptothecin derivative joined by an imine linkage were
synthesized and evaluated as cytotoxic agents and as inhibitors of DNA topoisomerase I. All of the conjugates were potent inhi-
bitors of tumor cell replication with improved activity relative to camptothecin. Significantly, compounds 16–18 were more active
than paclitaxel and camptothecin against HCT-8(colon adenocarcinoma) cell replication, and the spectrum of activity was different
from a simple mixture of paclitaxel and camptothecin. All of the conjugates were significantly less potent than camptothecin as
inhibitors of human topoisomerase I in vitro with 16, 18, and 19 showing only marginal activity at 50 mM. Based on activity against
drug-resistant cell line replication, one could conclude that the conjugates are simply acting as ‘weak taxanes’, but the spectrum of
activity, particularly against MCF-7 and HCT-8, strongly suggests that a novel mechanism of action has been achieved through
conjugation.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
Previously, we reported the synthesis and evaluation of
two 40-O-demethyl epipodophyllotoxin-camptothecin
conjugates (3 and 4) as inhibitors of mammalian DNA
topoisomerases I and II.7 The most active conjugate
inhibited cell growth similarly to both topo I- and
II-inhibitory components (Fig. 1). These conjugates
were more cytotoxic than epipodophyllotoxin in several
cancer cell lines including HOP-62 leukemia, SW-620
colon cancer, MCF/ADR adriamycin-resistant breast
cancer, and A-498renal cancer. One conjugate was more
active than either etoposide or 2 against human KB
(nasopharnyx) and DU-145 (prostate) tumor cell growth
in nude mice. This novel topo I and topo II dual inhibi-
tory activity and the unique antitumor action of these
camptothecin-etoposide analogue conjugates prompted us
to investigate the bis-molecular model for generating
novel types of antitumor agents. Currently, there are two
reports describing conjugates between paclitaxel and
either daunorubicin or chlorambucil.8,9 DNA topo inhi-
bitor-paclitaxel hybrids have not been investigated.
Paclitaxel (1)2 and camptothecin (2)3 are both plant-
derived antitumor agents currently in clinical use.
Paclitaxel was approved by the FDA for treatment of
advanced ovarian cancer in 1992 and for treatment of
breast cancer in 1994. The mechanism of antitumor
effect of paclitaxel is antimitotic, specifically promoting
the irreversible assembly of tubulin into microtubules.4
The natural alkaloid, camptothecin (2) is another com-
pound that has undergone continual structural modifi-
cation aimed at developing more useful chemotherapeutic
agents. Its significant antitumor activity is attributable
to inhibition of DNA topoisomerase I (DNA topo I).5,6
Although both paclitaxel and camptothecin possess
potent antitumor activity, recent reports have shown
that treatment with these drugs often results in a number
of undesired side effects as well as multi-drug resistance.
Therefore, it remains essential to develop new anti-
cancer agents with fewer side effects and improved
activity against various classes of tumors.
The aim of this work was to investigate whether an
antitumor agent displaying multiple antitumor activity
or improved activity against drug resistant cells could be
prepared through conjugation between paclitaxel and
*Corresponding author. Tel.: +1-919-962-0066; fax: +1-919-966-
3893; e-mail: khlee@unc.edu
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00040-3

1852
H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
nucleus10,11 and recent SAR study of 7-substituted
camptothecin analogues.12,13 However, the C-20 hydroxyl
group of the paclitaxel side chain is normally the most
reactive and selective C-7 acylation can be achieved only
when C-20 is protected. The procedure of Nicolaou et al.11
was used for the preparation of C-7 amino-functionalized
paclitaxels (11–15). Thus, 20-carboxybenzyl (Cbz)-pacli-
taxel was treated with N-Cbz-amino acids in the pre-
sence of DCC and DMAP in CH2Cl2 to give 6–10 in
good yields. Subsequent reductive removal of both Cbz
groups by hydrogenation gave the desired C-7 amino-
functionalized paclitaxels (11–15). Five paclitaxel-
camptothecin conjugates (16–20) were obtained by con-
jugation of 11–15 with 7-formyl-camptothecin (5) in
benzene as shown in Scheme 1. Compounds 16–20 were
unstable on silica gel (the parental compounds were
regenerated), but could be purified by Sephadex LH-20
column chromatography. All conjugates (16–20)
showed the characteristic imine proton signal at d 9.30–
1
9.55 in their H NMR spectra.
Results and Discussion
The five conjugates (16–20) were tested for cytotoxicity
against a panel of human tumor cell lines (HTCL)
including paclitaxel-resistant and camptothecin-resis-
tant derivatives.14 The results are shown in Table 1
with the values for paclitaxel (1) and camptothecin (2)
and a combination of both drugs given for compar-
ison. Activity profiles for the conjugates against MCF-
7 (breast cancer) and HCT-8(colon adenocarcinoma)
cells are shown in Figure 2A and B, respectively. All
of the conjugates were potent inhibitors of tumor cell
replication with improved activity relative to 2, and
also showed better activity than 2 against campto-
thecin-resistant KB-CPT cells. Significantly, 16–18
were more active against HCT-8cell replication than
both 1 and 2, and their potency and spectrum of
activity were different from a simple mixture of 1 and
2 (Table 1 and Fig. 2). SAR consideration of the lin-
ker between the paclitaxel and camptothecin moieties
showed that 16–18, in which paclitaxel and camp-
tothecin were linked by aliphatic amino acids, showed
better cytotoxic activity than 19 and 20, in which
paclitaxel and camptothecin were linked by aromatic
amino acids.
Figure 1.
camptothecin derivatives, and how their antitumor
effects are affected by the conjugation. We report in this
paper the synthesis and evaluation of paclitaxel-
camptothecin conjugates as cytotoxic agents and DNA
topo I inhibitors.
Chemistry
The choice of the linkage position was based on the
known tolerance of the C-7 ester group in the taxane
Table 1. Activity of paclitaxel–camptothecin conjugates 16–20 as inhibitors of cancer cell replicationa
Compd
Cell line/ED50
b
50
KB
KB-CPT
1A9
1A9-PTX10 HCT8MCF-7
PC-3
Mean ED
Paclitaxel
Camptothecin
16
17
18
19
ꢀ12.1
ꢀ10.8
ꢀ12.2
ꢀ7.5
ꢀ11.1
ꢀ9.7
ꢀ8.1
ꢀ8.9
ꢀ8.1
ꢀ10.6
ꢀ12.2
ꢀ10.2
ꢀ11.7
ꢀ9.4
ꢀ7.1
ꢀ9.3
ꢀ13.9
ꢀ7.1
ꢀ10.3
ꢀ7.1
ꢀ7.0
ꢀ7.1
ꢀ7.5
ꢀ7.5
ꢀ11.1
ꢀ8.9
<ꢀ9.3
ꢀ9.0
<ꢀ11.0 (67)
ꢀ10.4
<ꢀ11.0 (76)
ꢀ10.2
ꢀ10.9
ꢀ10.5
ꢀ10.5
ꢀ8.4
ꢀ10.6
ꢀ10.0
ꢀ10.6
ꢀ8.1
ꢀ7.1
ꢀ7.1
ꢀ9.3
ꢀ10.0
ꢀ7.5
ꢀ9.0
ꢀ8.5
ꢀ8.1
ꢀ6.0
ꢀ7.6
20
Paclitaxel+Camptothecin (1:1)
ꢀ10.8
ꢀ8.3
ꢀ7.1
ꢀ7.4
>ꢀ4.0 (48)
>ꢀ5.0 (45)
ꢀ6.6
>ꢀ7.1
ꢀ6.9
ꢀ10.5
ꢀ9.0
ꢀ7.9
ꢀ6.4
ꢀ7.7
aEffects on tumor cell line replication were determined using a standard method.15 Cell line/ED50 in log10 M (replicates varied no more than 5%). If
inhibition <50% at highest test concentration or >50% at lowest, the observed percentage inhibition observed is the bracketed value.
bMean value is taken from HTCL data and does not include values for drug-resistant variants.

H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
1853
Scheme 1. (a) Benzylchroloformate, pyridine, CH2Cl2, 83%; (b) N-Cbz-NH-R-COOH, DCC, DMAP, CH2Cl2, 84–97%; (c) H2, 5% Pd–C, MeOH,
81–97%; (d) benzene, MS 4 A, reflux, 55–85%.
Figure 2. Compounds were tested as inhibitors of human tumor cell line replication and ED50 values interpolated from dose–response activity data
are given in Table 1. Graphs show activity profiles of compounds against MCF-7 (A) and HCT-8cell replication (B).
Compounds 16–20 were also tested as inhibitors of
human DNA topo I activity in vitro (Fig. 3).15 All of the
conjugates were significantly less potent than 2 in this
assay. Compounds 16, 18, and 19 showed only marginal
activity at 50 mM, while 17 and 20 were inactive. None
of the conjugates inhibited DNA relaxation.
Based on activity against drug-resistant cell line repli-
cation, one could conclude that the conjugates are sim-
ply acting as ‘weak taxanes’, but the spectrum of
activity, particularly against MCF-7 and HCT-8( Fig. 2),
is not consistent with this conclusion. The latter results
indicate that, compared with 1 and 2, the cytotoxicity

1854
H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
Figure 3. Compounds were tested as human DNA topoisomerase I inhibitors and analyzed using an ethidium bromide-containing agarose gel as
described.15 Lane 1 is the pBR322 DNA substrate. Lane 2 is enzyme control. Lanes 3–10 are compound treated (50 mM) as follows: lanes 3 and 10,
campthothecin; lane 4, paclitaxal; lane 5, compound 16; lane 6, compound 18; lane 7, compound 19; lane 8, compound 17; lane 9, compound 20.
profiles of 16–18 are quite distinctive. In addition,
because the results from the DNA topo I inhibition
assay show conclusively that the conjugates, unlike 2,
are not DNA topo I inhibitory in vitro, it is likely that a
novel mechanism of action has been achieved through
conjugation.
was performed using pre-coated silica gel on aluminum
plates (Aldrich, Inc.). Elemental analyses were per-
formed by Atlantic Microlab Inc., Norcross, GA.
N-Cbz-animo acids were prepared from the appropriate
amino acids using standard methodology16 using benzyl-
chloroformate, except for N-Cbz-b-alanine (Sigma).
20-O-Cbz-paclitaxel was synthesized by a literature
method.11
Conclusions
General procedure for 7-O-acylation of 20-O-Cbz-pacli-
taxel. A solution of 20-O-Cbz-paclitaxel and 4-(N,N-
dimethylamino)pyridine (1.0 equiv) in dry dichloro-
methane (5 mL) under nitrogen was treated with the
appropriate N-Cbz-amino acid (5.0 equiv) and dicyclo-
hexylcarbodiimide (DCC, 5.0 equiv) overnight at rt. The
mixture was diluted with dichloromethane (5 mL) and
filtered to remove the urea precipitate. The solvent was
evaporated in vacuo and the resulting residue was pur-
ified by silica gel column chromatography using a
n-hexane–EtOAc solvent system (2:1 to 1:1) to afford
the desired 20-O-Cbz-7-O-acyl paclitaxel as the sole
product.
In summary, we have synthesized five paclitaxel-
camptothecin conjugates joined by an ester and an
imine linkage. When evaluated for cytotoxicity against a
panel of human tumor cell lines, all of the conjugates
showed potent cytotoxic activity against tumor cell
replication. Interestingly, compounds 16–18 showed
improved activity against HCT-8cell replication com-
pared with either 1 or 2. In-depth mechanistic studies
and the development of new paclitaxel-camptothecin
conjugates are actively underway in our laboratory.
Experimental
20-O-Cbz-7-(N-Cbz-ꢀ-alanyl)-paclitaxel (6). Yield 90.5%
(starting with 59.3 mg of 20-O-Cbz-paclitaxel); white
powder, [a]D ꢀ44.7ꢁ (c 0.76, CHCl3); 1H NMR
(CDCl3): d 1.16 (3H, s), 1.22 (3H, s), 1.81 (3H, s), 2.00
(3H, s), 2.06 (3H, s), 2.46 (3H, s), 3.46 (2H, m), 3.96
(1H, d. J=7.0 Hz), 4.20 (1H, d, J=8.5 Hz), 4.33 (1H, d,
J=8.5 Hz), 4.95 (1H, d, J=8.5 Hz), 5.11 (2H, m), 5.17
(2H, m), 5.46 (1H, d. J=2.5 Hz), 5.64 (1H, dd, J=7.2,
9.8Hz), 5.70 (1H. d. J=7.0 Hz), 5.77 (1H, t, J=5.1 Hz),
5.98(1H, d, J=9.0 Hz), 6.26 (1H, t, J=9.3 Hz), 6.29
(1H, s), 6.94 (1H, d, J=9.0 Hz), 7.30–7.42 (band, 17H),
7.44–7.54 (3H, m), 7.62 (1H, m), 7.74 (2H, d,
J=8.5 Hz), 8.13 (2H, d, J=8.5 Hz); FAB-MS m/z 1193
[M+1]+. All data are in agreement with reported lit-
erature values.3
General procedures
Paclitaxel was obtained from Yung Shin Pharmaceu-
tical Ind. Co., Ltd., Taiwan. Preparation of 7-formyl-
camptothecin was reported in our previous paper.7 All
reagents and solvents were reagent grade or were puri-
1
fied by standard methods before use. H NMR spectra
were obtained on a Varian Gemini-300 spectrometer
using TMS as the internal standard. Chemical shifts (d
values) and coupling constants (J values) are given in ppm
and Hz, respectively. Optical rotations were measured
with a JASCO DIP-1000 digital polarimeter. Column
chromatography was carried out on silica gel 60 (Merck
230–400 mesh), and thin layer chromatography (TLC)

H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
1855
20 -O-Cbz-7-(N-Cbz-4-aminobutyroyl)-paclitaxel (7).
Yield 97.2% (starting with 69.1 mg of 20-O-Cbz-pacli-
content was filtered to remove the catalyst and evapo-
rated to dryness in vacuo to give the desired deprotected
compounds 11–15.
ꢁ
1
taxel); white powder, [a]D ꢀ31.5 (c 0.46, CHCl3); H
NMR (CDCl3): d 1.16 (3H, s), 1.21 (3H, s), 1.81 (3H, s),
1.99 (3H, s), 2.09 (3H, s), 2.45 (3H, s), 3.19–3.40 (2H,
m), 3.94 (1H, d. J=6.9 Hz), 4.20 (1H, d, J=8.5 Hz),
4.33 (1H, d, J=8.5 Hz), 4.95 (1H, d, J=8.5 Hz), 5.08
(2H, s), 5.17 (2H, m), 5.45 (1H, d. J=2.4 Hz), 5.57 (1H,
dd, J=6.9, 10.2 Hz), 5.71 (1H. d, J=6.9 Hz), 5.73 (1H,
t, J=5.1 Hz), 5.98(1H, dd, J=2.4, 9.0 Hz), 6.24 (1H, s),
6.26 (1H, t, J=9.3 Hz), 6.92 (1H, d, J=9.0 Hz), 7.27–
7.46 (band, 16H), 7.48–7.54 (3H, m), 7.62 (1H, m), 7.73
(2H, d, J=8.5 Hz), 8.13 (2H, d, J=8.5 Hz); ESI-MS m/z
1230 [M+Na]+.
7-O-ꢀ-Alanylpaclitaxel (11). Yield 81.2% (starting with
47.2 mg of 6); white film, [a]D ꢀ39.1ꢁ (c 0.34, MeOH);
1H NMR (CD3OD): d 1.10 (3H, s), 1.16 (3H, s), 1.78
(3H, s), 1.88 (3H, s), 2.16 (3H, s), 2.37 (3H, s), 3.12–3.22
(2H, m), 3.90 (1H, d, J=7.0 Hz), 4.20 (2H, dd, J=7.0,
14.0 Hz), 4.73 (1H, d, J=6.0 Hz), 5.00 (1H, m), 5.64
(3H, m), 6.15 (1H, br t, J=6.2 Hz), 6.23 (1H, s), 7.28–
7.68(band, 11H), 7.85 (2H, d, J=8.5 Hz), 8.10 (2H, d,
J=8.5 Hz); FAB-MS m/z 925 [M+1]+. All data are in
agreement with reported literature values.3
20 -O-Cbz-7-(N-Cbz-6-aminohexanoyl)-paclitaxel (8).
Yield 94.7% (starting with 78.3mg of 20-O-Cbz-pacli-
7-O-(4-Aminobutyroyl)paclitaxel (12). Yield 92.7%
(starting with 59.1 mg of 7); white film, [a]D ꢀ29.7ꢁ (c
taxel); white powder, [a]D ꢀ44.0ꢁ (c 1.04, CHCl3); H
0.37, MeOH); H NMR (CD3OD): d 1.11 (3H, s), 1.16
1
1
NMR (CDCl3): d 1.16 (3H, s), 1.21 (3H, s), 1.81 (3H, s),
2.00 (3H, s), 2.13 (3H, s), 2.45 (3H, s), 3.19 (2H, dd, J=6.3,
12.9 Hz), 3.96 (1H, d. J=6.6 Hz), 4.19 (1H, d, J=8.5 Hz),
4.32 (1H, d, J=8.5Hz), 4.96 (1H, d, J=8.7Hz), 5.08 (2H,
s), 5.16 (2H, m), 5.45 (1H, d. J=2.4 Hz), 5.60 (1H, dd,
J=7.2, 10.5 Hz), 5.71 (1H. d, J=6.9 Hz), 5.98(1H, dd,
J=3.0, 9.5 Hz), 6.26 (1H, t, J=9.0Hz), 6.28(1H, s), 6.95
(1H, d, J=9.3 Hz), 7.26–7.43 (band, 16H), 7.46–7.53 (3H,
m), 7.61 (1H, m), 7.73 (2H, d, J=8.5 Hz), 8.13 (2H, d,
J=8.5Hz); FAB-MS m/z 1236 [M+1]+.
(3H, s), 1.78(3H, s), 1.87 (3H, s), 2.16 (3H, s), 2.37 (3H,
s), 3.01 (2H, t, J=7.5 Hz), 3.90 (1H, d, J=6.9 Hz), 4.20
(2H, br t, J=8.7 Hz), 4.75 (1H, d, J=5.4 Hz), 5.00 (1H,
d, J=9.3 Hz), 5.59–5.66 (3H, m), 6.15 (1H, br t,
J=6.2 Hz), 6.21 (1H, s), 7.26–7.69 (band, 11H), 7.85
(2H, d, J=8.5 Hz), 8.10 (2H, d, J=8.5 Hz); ESI-MS m/z
939 [M+1]+.
7-O-(6-Aminohexanoyl)paclitaxel (13). Yield 81.1%
(starting with 67.1 mg of 8); white film, [a]D ꢀ26.0ꢁ (c
1
0.53, MeOH); H NMR (CD3OD): d 1.11 (3H, s), 1.16
20 -O-Cbz-7-(N-Cbz-4-methylaminobenzoyl)-paclitaxel
(9). Yield 89.1% (starting with 70.4 mg of 20-O-Cbz-
paclitaxel); white powder, [a]D ꢀ28.2ꢁ (c 0.99, CHCl3);
1H NMR (CDCl3): d 1.19 (3H, s), 1.21 (3H, s), 1.95 (3H,
s), 1.97 (3H, s), 2.04 (3H, s), 2.48(3H, s), 4.05 (1H, d.
J=6.9 Hz), 4.24 (1H, d, J=8.5 Hz), 4.37 (1H, d,
J=8.5 Hz), 4.42 (1H, d, J=6.0 Hz), 5.01 (1H, d,
J=8.4 Hz), 5.10–5.22 (6H, m), 5.48 (1H, d. J=2.7 Hz),
5.74–5.81 (2Hm), 5.99 (1H, dd, J=2.7, 9.2 Hz), 6.27
(1H, t, J=9.0 Hz), 6.41 (1H, s), 6.97 (1H, d, J=9.3 Hz),
7.29–7.42 (band, 19H), 7.46–7.54 (3H, m), 7.62 (1H, m),
7.74 (2H, d, J=8.5 Hz), 7.89 (2H, d, J=8.4 Hz). 8.15
(2H, d, J=8.5 Hz); ESI-MS m/z 1277 [M+Na]+.
(3H, s), 1.77 (3H, s), 1.88 (3H, s), 2.14 (3H, s), 2.37 (3H,
s), 2.92 (2H, m), 3.90 (1H, d, J=6.9 Hz), 4.20 (2H, m),
4.75 (1H, d, J=5.4 Hz), 4.99 (1H, d, J=9.0 Hz), 5.58
(1H, dd, J=7.8, 10.5 Hz), 5.60–5.70 (2H, m), 6.15 (1H,
br t, J=6.2 Hz), 6.24 (1H, s), 7.26–7.72 (band, 11H),
7.85 (2H, d, J=8.5 Hz), 8.10 (2H, d, J=8.5 Hz); FAB-
+
MS m/z 968[M+1]
.
7-O-(4-Methylaminobenzoyl)paclitaxel
(14).
Yield
96.7% (starting with 58.1 mg of 9); white film, [a]D
ꢀ12.9ꢁ (c 0.38, MeOH); 1H NMR (CD3OD): d 1.13
(3H, s), 1.15 (3H, s), 1.91 (9H, s), 2.40 (3H, s), 3.90 (1H,
d, J=6.9 Hz), 4.18–4.30 (4H, m), 4.76 (1H, d,
J=5.7 Hz), 5.05 (1H, m), 5.65 (1H, d, J=5.1 Hz), 5.70
(1H, d, J=7.2 Hz), 5.76 (1H, dd, J=7.2, 10.5 Hz), 6.16
(1H, br t, J=6.2 Hz), 6.35 (1H, s), 7.28–7.71 (band,
13H), 7.85 (2H, d, J=8.5 Hz), 7.94 (2H, d, J=8.5 Hz),
8.12 (2H, d, J=8.5 Hz); ESI-MS m/z 1009 [M+Na]+.
20-O-Cbz-7-(N-Cbz-4-(4-aminophenyl)-butyroyl)-pacli-
taxel (10). Yield 83.8% (starting with 66.8 mg of 20-O-
Cbz-paclitaxel); white powder, [a]D ꢀ46.5ꢁ (c 0.40,
1
CHCl3); H NMR (CDCl3): d 1.17 (3H, s), 1.21 (3H, s),
1.81 (3H, s), 2.01 (3H, s), 2.13 (3H, s), 2.45 (3H, s), 3.96
(1H, d. J=6.9 Hz), 4.19 (1H, d, J=8.5 Hz), 4.33 (1H, d,
J=8.5 Hz), 4.96 (1H, d, J=8.1 Hz), 5.12–5.21 (4H, m),
5.46 (1H, d. J=2.7 Hz), 5.60 (1H, dd, J=7.2, 10.5 Hz), 5.70
(1H. d, J=7.2 Hz), 5.98(1H, dd, J=2.7, 9.2 Hz), 6.26 (1H,
t, J=9.0 Hz), 6.29 (1H, s), 6.66 (1H, s), 6.94 (1H, d,
J=9.3 Hz), 7.12 (2H, d, J=8.7 Hz), 7.26–7.42 (band, 19H),
7.46–7.53 (3H, m), 7.61 (1H, m), 7.73 (2H, d, J=8.5 Hz),
8.13 (2H, d, J=8.5 Hz); FAB-MS m/z 1306 [M+Na]+.
7-O-(4-(4-Aminophenyl)butyroyl)paclitaxel (15). Yield
81.3% (starting with 67.1 mg of 10); white film, [a]D
ꢀ45.3ꢁ (c 0.15, MeOH); 1H NMR (CD3OD): d 1.11
(3H, s), 1.14 (3H, s), 1.77 (3H, s), 1.89 (3H, s), 2.09 (3H,
s), 2.36 (3H, s), 2.46 (2H, t, J=5.7 Hz), 3.89 (1H, d,
J=6.9 Hz), 4.19 (2H, brs), 4.75 (1H, d, J=5.4 Hz), 4.97
(1H, d, J=9.6 Hz), 5.56 (1H, dd, J=7.8, 10.5 Hz), 5.63–
5.66 (2H, m), 6.15 (1H, br t, J=6.2 Hz), 6.26 (1H, s),
6.67 (2H, d, J=8.4 Hz), 6.93 (2H, d, J=8.4 Hz), 7.28
General procedure for removal of the benzyloxycarbonyl
(Cbz) protecting group. A solution of taxoid (6–10) in
methanol under nitrogen was treated with 50 wt.% of
5% palladium on activated carbon (Degussa type
E101 NO/W) and placed under hydrogen (40 psi). The
mixture was shaken in a Parr apparatus for 7 h then the
(1H, m), 7.38–7.68 (band, 10H), 7.84 (2H, d, J=8.5 Hz),
+
8.10 (2H, d, J=8.5 Hz); FAB-MS m/z 1038[M+Na]
.
General procedure for the synthesis of paclitaxel–camp-
tothecin conjugates (16–20). A solution of taxoid (11–
15) and 7-formylcamptothecin (1.2 equiv) in dry ben-

1856
H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
zene (10 mL) was refluxed over 4 A molecular sieves over-
night. The mixture was filtered and evaporated in vacuo to
give crude paclitaxel–camptothecin conjugates (16–20).
The product was purified by Sephadex LH-20 column
chromatography with CHCl3–MeOH (1:1) as eluent.
20: Yield 55.1% [starting with 23.3 mg (0.023 mmol) of 15];
pale yellow amorphous solid, [a]D ꢀ11.5 (c 0.27, CHCl3),
1H NMR (CDCl3): d 1.06 (3H, t, J=7.5 Hz), 1.18(3H, s),
1.25 (3H, s), 1.84 (3H, s), 1.85 (3H, s), 2.17 (3H, s), 2.39
(3H, s), 3.94 (1H, d, J=6.6 Hz), 4.20 and 4.33 (each 1H, d,
J=8.4 Hz), 4.81 (1H, brs), 4.96 (1H, d, J=9.3 Hz), 5.33
(1H, d, J=16.5Hz), 5.58 (1H, m), 5.70 (1H, d, J=5.7 Hz),
5.73 (2H, s), 5.81(1H, d, J=5.7 Hz), 6.19 (1H, t,
J=9.0 Hz), 6.26 (1H, s), 7.08(1H, d, J=9.3 Hz), 7.27–7.80
(18H, m), 7.88 (1H, t, J=7.2 Hz), 8.12 (2H, d, J=7.2 Hz),
8.33 (1H, d, J=8.4 Hz), 8.56 (1H, d, J=8.4 Hz), 9.55 (1H,
s); ESI-MS m/z 1368[M+Na] +. Anal. (C78H78N4O18):
theory: C, 68.91; H, 5.78. Found C, 68.86; H, 5.77.
16: Yield 85.4% [starting with 14.2 mg (0.015 mmol) of
11]; pale yellow amorphous solid, [a]D ꢀ33.9ꢁ (c 0.18,
1
CHCl3), H NMR (CDCl3): d 1.02 (3H, t, J=7.5 Hz),
1.18 (3H, s), 1.25 (3H, s), 1.80 (3H, s), 1.85 (3H, s), 2.19
(3H, s), 2.37 (3H, s), 3.92 (1H, d, J=6.6 Hz), 4.14 and
4.29 (each 1H, d, J=8.4 Hz), 4.80 (1H, brs), 4.90 (1H, d,
J=9.3 Hz), 5.21 (1H, d, J=16.5 Hz), 5.40 (2H, s), 5.58–
5.70 (3H, m), 5.80 (1H, dd, J=8.9, 2.1 Hz), 6.19 (1H, t,
J=9.0 Hz), 6.24 (1H, s), 7.27–7.86 (16H, m), 8.09 (2H,
d, J=7.2 Hz), 8.27 (1H, d, J=8.4 Hz), 8.45 (1H, d,
J=8.4 Hz), 9.32 (1H, s); FAB-MS m/z (rel. int.) 1284
[M+1]+ (35), 1224 (15), 940 (51), 449 (77), 289 (98),
240 (100). Anal. (C71H72N4O18): theory: C, 67.18; H,
5.72. Found C, 67.12; H, 5.69.
Human tumor cell replication assay. Compounds were
tested as inhibitors of cell growth against a limited panel
of solid tumor lines including KB (epidermoid carci-
noma, CL 17) and a camptothecin-resistant sub-line
called KB-CPT; HCT-8(ileocecal adenocarcinoma,
CCL 244); MCF-7 (breast adenocarcinoma, HTB 22);
PC-3 (prostate adenocarcinoma, CRL 1435); 1A9
(ovarian carcinoma) and a plactixel resistant sub-line
called PTX10 with mutated b-tubulin. Drug-resistant
cell lines were generous gifts of Dr. Y. C. Cheng (Yale)
and Dr. P. Giannakakou (NIH, Bethesda, MD). Cell
lines were adapted for growth in RPMI-1640 medium
supplemented with 25 mM Hepes, 2% NaHCO3, 10%
(v/v) fetal calf serum and 100 mg/mL kanamycin. Cul-
tures were maintained in a 5% CO2 atmosphere at
37 ꢁC. The sulforhodamine B microtiter plate assay
described by Rubenstein et al.14 was used to evaluate
compounds as inhibitors of cell replication. The ED50
value, the concentration of compound that inhibited cell
line replication by 50% relative to control after three
days of continuous exposure was interpolated from
dose–response graphs fitted to data using Prizm
(GraphPad software, San Diego, CA).
17: Yield 60.7% [starting with 27.0 mg (0.029 mmol) of
12]; pale yellow amorphous solid, [a]D ꢀ17.2ꢁ (c 0.47,
1
CHCl3), H NMR (CDCl3): d 1.04 (3H, t, J=7.5 Hz),
1.16 (3H, s), 1.20 (3H, s), 1.81 (3H, s), 1.83 (3H, s), 2.04
(3H, s), 2.38(3H, s), 3.81–3.93 (4H, m), 4.18and 4.31
(each 1H, d, J=8.4 Hz), 4.81 (1H, dd, J=6.2, 2.7 Hz),
4.93 (1H, d, J=8.7 Hz), 5.30 (1H, d, J=16.5 Hz), 5.50
(2H, s), 5.58(1H, dd, J=10.2, 6.9 Hz), 5.67 (1H, d,
J=6.9Hz), 5.74 (1H, d, J=16.5 Hz), 5.80 (1H, dd, J=8.9,
2.4 Hz), 6.17 (1H, t, J=9.0Hz), 6.23 (1H, s), 7.14 (1H, d,
J=9.3Hz), 7.30–7.87 (15H, m), 8.10 (2H, d, J=7.2Hz),
8.28 (1H, d, J=8.4Hz), 8.50 (1H, d, J=8.4 Hz), 9.32 (1H,
s); ESI-MS m/z 1320 [M+Na]+. Anal. (C72H74N4O18):
theory: C, 67.38; H, 5.81. Found C, 67.30; H, 5.79.
18: Yield 69.3% [starting with 27.8mg (0.029 mmol) of
13]; pale yellow amorphous solid, [a]D ꢀ41.3 (c 0.15,
1
CHCl3), H NMR (CDCl3): d 1.03 (3H, t, J=7.5 Hz),
DNA topoisomerase I activity assay. Compounds were
tested as inhibitors of enzyme catalyzed plasmid DNA
relaxation. Reactions contained 0.5 mg pBR322 DNA
and 0.5 U human DNA topoisomerase I (TopoGen,
Columbus, OH), compounds at 50 mM and buffer com-
ponents described previously.15 After 15 min at 37 ꢁC,
reactions were terminated by adding proteinase K
(0.1 mg/mL) and 1% (w/v) SDS. After 1 h at 50 ꢁC, DNA
was resolved on a 1% (w/v) agarose gel containing 0.5 mg/
mL ethidium bromide in standard TBE buffer, and
photographed under UV light using polaroid 667 film.
1.15 (3H, s), 1.25 (3H, s), 1.57 (3H, s), 1.76 (3H, s), 2.10
(3H, s), 2.28(3H, s), 3.80 (1H, d, J=6.6 Hz), 4.13 and
4.27 (each 1H, d, J=8.4 Hz), 4.79 (1H, d, J=2.7 Hz), 4.85
(1H, d, J=8.1Hz), 5.31 (1H, d, J=16.5 Hz), 5.46 (1H, dd,
J=10.2, 7.2 Hz), 5.55 (2H, s), 5.61 (1H, d, J=6.9Hz), 5.74
(2H, m), 6.07 (1H, t, J=9.0Hz), 6.12 (1H, s), 7.18(1H, d,
J=9.3Hz), 7.32–7.88 (15H, m), 8.08 (2H, d, J=7.2Hz),
8.30 (1H, d, J=8.4Hz), 8.52 (1H, d, J=8.4 Hz), 9.30 (1H,
s); ESI-MS m/z 1348[M+Na] +. Anal. (C74H78N4O18):
theory: C, 67.77; H, 5.99. Found C, 67.69; H, 5.98.
19: Yield 64.3% [starting with 28.2 mg (0.029 mmol) of
14]; pale yellow amorphous solid, [a]D ꢀ8.0ꢁ (c 0.40,
Elemental analysis data for compounds 16–20
1
CHCl3), H NMR (CDCl3): d 1.02 (3H, t, J=7.5 Hz),
1.19 (3H, s), 1.25 (3H, s), 1.95 (3H, s), 2.17 (3H, s), 2.40
(3H, s), 2.63 (3H, s), 4.00 (1H, d, J=6.3 Hz), 4.23 and
4.33 (each 1H, d, J=8.1 Hz), 4.82 (1H, d, J=2.7 Hz),
4.98(1H, d, J=8.7 Hz), 5.14 (2H, brs), 5.28 (1H, d,
J=16.2 Hz), 5.52 (2H, s), 5.67–5.76 (2H, m), 5.81 (1H,
dd, J=8.7, 2.1 Hz), 6.19 (1H, t, J=9.0 Hz), 6.37 (1H, s),
7.21–7.96 (20H, m), 8.12 (2H, d, J=7.2 Hz), 8.27 (1H,
d, J=8.4 Hz), 8.42 (1H, d, J=8.4 Hz), 9.39 (1H, s); ESI-
MS m/z 1396 [M+Na]+. Anal. (C76H74N4O18): theory:
C, 68.56; H, 5.60. Found C, 68.40; H, 5.56.
Compd
Calculated
Found
C
Formula
C
H
H
16
17
18
19
20
C71H72N4O18
67.185.72
67.3185.8 67.30
67.12
5.69
C72H74N4O18
C74H78N4O18
C76H74N4O18
C78H78N4O18
5.79
67.77
68.56
68.91
5.99
5.60
5.78
67.69
68.40
68.86
5.98
5.56
5.77

H. Ohtsu et al. / Bioorg. Med. Chem. 11 (2003) 1851–1857
1857
Acknowledgements
7. Bastow, K. F.; Wang, H. K.; Cheng, Y. C.; Lee, K. H.
Bioorg. Med. Chem. 1997, 5, 1481.
8. Kar, A. K.; Braun, P. D.; Wandless, T. J. Bioorg. Med.
Chem. Lett. 2000, 10, 261.
9. Wittman, M. D.; Kadow, J. F.; Vyas, D. M.; Lee, F. L.;
Rose, W. C.; Long, B. H.; Fairchild, D.; Johnston, K. Bioorg.
Med. Chem. Lett. 2001, 11, 811.
This research was supported by a grant from the
National Cancer Institute (CA 17625) awarded to K. H.
Lee.
10. Nicolaou, K. C.; Dai, W. M.; Guy, R. K. Angew. Chem.,
Intl. Ed. Engl. 1994, 106, 15.
References and Notes
11. Guy, R. K.; Scott, Z. A.; Sloboda, R. D.; Nicolaou, K. C.
Chem. Biol. 1996, 3, 1021.
12. Wang, H. K.; Liu, S. Y.; Hwang, K. M.; Taylor, G.; Lee,
K. H. Bioorg. Med. Chem. 1994, 2, 1397.
1. For a previous paper in this series see, Han, S.; Hamel, E.,
Bastow, K. F.; McPhail, A. T.; Brossi, A.; Lee, K. H. Bioorg.
Med. Chem Lett., in press.
2. Wani, M. C.; Taylor, H. L.; Wall, M. E.; Coggon, P.;
McPhail, A. T. J. Am. Chem. Soc. 1971, 93, 2325.
3. Wall, M. E.; Wani, M. C.; Cook, C. E.; Palmer, K. H.;
McPhail, A. T.; Sim, G. A. J. Am. Chem. Soc. 1966, 88,
3888.
4. Schiff, P. B.; Fant, J.; Horwitz, S. B. Nature 1979, 277,
665.
5. Hsiang, Y. H.; Hertzberg, R.; Hecht, S.; Liu, L. F. J. Biol.
Chem. 1985, 260, 14873.
6. Wang, J. C. Biochim. Biophys. Acta 1987, 909, 1.
13. Wang, Z. Q.; Kuo, Y. H.; Schnur, D.; Bowen, J. P.; Liu,
S. Y.; Han, F. S.; Chang, J. Y.; Cheng, Y. C.; Lee, K. H.
J. Med. Chem. 1990, 33, 2660.
14. Rubinstein, C. V.; Shoemaker, R. H.; Paull, K. D.; Simon,
R. M.; Tosini, S.; Skehan, P.; Scudiero, D. A.; Monks, A.;
Boyd, M. R. J. Natl. Cancer Inst. 1990, 82, 1113.
15. Krishnan, P.; Bastow, K. F. Anticancer Drug Des. 2000,
15, 255.
16. Berger, A.; Kurtz, J.; Katchslski, J. J. Am. Chem. Soc.
1954, 76, 5552.