Synthesis, hydrolytic activation and cytotoxicity of etoposide prodrugs.

Article abstract of DOI:10.1016/S0960-894X(01)00801-0

Two 4'-propylcarbonoxy derivatives (2,3) of etoposide (1), a topoisomerase II inhibitor, were synthesized and evaluated as potential prodrugs for anticancer therapy. Their activation via hydrolysis mechanisms was determined as a function of pH in buffer solutions, in human serum and in the presence of carboxyl ester hydrolase. Cytotoxicity was determined on various tumor cell lines and compared to the parent compound. On cell lines exhibiting resistance to etoposide we observed an enhanced cytotoxicity of the prodrugs of up to three orders of magnitude.

Full text of DOI:10.1016/S0960-894X(01)00801-0

Bioorganic & Medicinal Chemistry Letters 12 (2002) 557–560
Synthesis, Hydrolytic Activation and Cytotoxicity
of Etoposide Prodrugs
Wolf Wrasidlo,a,* Ulrike Schroder,a Kathrin Bernt,a Nicole Hubener,a Doron Shabat,b
Gerhard Gaedickea and Holger Lodea
a
´
Charite Children’s Hospital, Humboldt University, Augustenburger Platz 1, 13353 Berlin, Germany
bSchool of Chemistry, Tel-Aviv University, Tel-Aviv 69978, Israel
Received 6 June 2001; accepted 26 November 2001
Abstract—Two 40-propylcarbonoxy derivatives (2,3) of etoposide (1), a topoisomerase II inhibitor, were synthesized and evaluated
as potential prodrugs for anticancer therapy. Their activation via hydrolysis mechanisms was determined as a function of pH in
buffer solutions, in human serum and in the presence of carboxyl ester hydrolase. Cytotoxicity was determined on various tumor
cell lines and compared to the parent compound. On cell lines exhibiting resistance to etoposide we observed an enhanced cyto-
toxicity of the prodrugs of up to three orders of magnitude. # 2002 Elsevier Science Ltd. All rights reserved.
Etoposide (1) is a widely used, highly effective anti-
cancer drug against a broad spectrum of tumors
including pediatric cancers such as acute lymphatic
lymphomas, rhabdomyosarcomas and neuroblastomas,
as well as in most common adult cancers.1,2 It is also
used in bone marrow transplantation conditioning regi-
mens. However, the therapeutic use of etoposide is lim-
ited by toxicity involving mainly myelosuppression.3 A
major problem for the use of this drug (and other
topoisomerase inhibitors) is the development of multi-
drug resistance.4,5 Efforts are presently made to synthe-
size new derivatives of the natural podophyllotoxin to
overcome some of these shortcomings.6ꢀ8 An alternative
approach to new analogues of therapeutic agents is the
synthesis of prodrugs. In previous work9ꢀ12 we synthe-
sized prodrugs of various antitumor agents to improve
their bio-availability, phamacokinetics and aqueous
solubility. For paclitaxel11,12 we established that pro-
drugs incorporating hydrolytically cleavable moieties are
effective in lowering the systemic toxicity of the drug in
animal models and most importantly also in patients.13
Furthermore these prodrugs of paclitaxel revealed a
dramatically improved pharmacokinetics and could be
formulated for intravenous delivery with a minimum of
toxic organic vehicles. Based on these findings the
hydrolytic activation approach was applied to etoposide,
an important drug in pediatrics and adult oncology.
Synthesis
Compounds 2 and 3 were prepared as follows. Treat-
ment of a solution of 1 (Sigma, Germany) in dichloro-
methane, with one equivalent of soketal chloroformate
at ꢀ70 ꢁC followed by the addition of one equivalent of
pyridine and reacting for 6 h, lead to 2. The workup
consisted of extraction with brine, drying over magne-
sium sulfate, centrifugation and concentrating the
organic layer to one fifth of the reaction volume. The
product was isolated by HPLC purification (C18 reverse-
phase column with 50/50 vol% acetonitrile–water),
followed by lyophilization of the eluate (yield 83.5%,
purity 99.2%). Compound 3 was prepared from a THF
solution of 2 by the addition of 2 N HCl and reacting
for 2 h at ambient temperature. Workup consisted of
washing with brine, drying over magnesium sulfate,
reducing the volume to 25% of the reaction volume and
purification via HPLC under the conditions given for 2.
The yield of purified product was 85.5% and the purity
by HPLC was greater than 99%.
The structures of 2 and 3 were established by 1H NMR,
13C NMR and MS data.14,15
Solution stability
The chemical stability of solutions of 2 and 3 were
determined by dissolving the compounds in 50/50 vol%
chromophore–ethanol (1 mg/10 mL solvent mixture)
*Corresponding author. Fax: +49-30-450-559917; e-mail: wolf.
wrasidlo@charite.de
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00801-0

558
W. Wrasidlo et al. / Bioorg. Med. Chem. Lett. 12 (2002) 557–560
Scheme 1. (a) Soketal* chloroformate, CH2Cl2; (b) THF, 2 N HCl, (c) PBS, pH>8.0; or carboxylesterase, pH 7.3; or serum (d) carboxyl esterase,
pH 7.3; or serum (*2,2-dimethyl-4-hydroxymethyl-1,3-dioxolan).
and adding 90 mL of 1ꢂ PBS, pH 7.3 to obtain
concentrations of about 14 mM. Solutions were also
prepared in 50 vol% acetonitrile–water. These solutions
were stored under ambient laboratory conditions with
average temperatures of about 22ꢁC. Samples, 2 mL,
were periodically removed and analyzed by HPLC over a
period of 3 weeks.
Cytotoxicity
Anti-proliferative activities of 1–3 were determined
using the XTT assay.16 Briefly, 10,000 tumor cells/well
were added to 96-well plates and incubated under stan-
dard tissue culture conditions (37 ꢁC, 5% CO2) for 24 h
before addition of drug at concentrations ranging from
100 mM to 100 fM by direct addition from stock solu-
tions in DMSO. After 72 h, XTT (Aldrich Chemical
Co., Germany) was added to a final concentration of
250 mg/well. The plates were then incubated for an
additional 4 h and the OD determined at 450 nm using
a microplate reader. The cell viability–drug dilution
profiles were obtained from sigma plots, and drug
concentrations which inhibited growth by 50% were
calculated from duplicate runs (IC50).
Hydrolysis studies
Hydrolysis rates were determined by dissolving the
compounds in acetonitrile–aqueous buffer solutions (20/
80 vol%) at different pH, in human serum or with car-
boxylic ester hydrolase (CEH) solutions (1 mM, from
pig liver, Sigma Co., Germany). In addition, human
serum was fractionated via HPLC size exclusion chro-
matography (G200), resulting in four peaks (ranging in
Mr cutoffs between 350,000 and 10,000) and fractions
containing these peaks were used in hydrolysis experi-
ments. All experiments were conducted at 37 ꢁC at drug
concentrations of about 6 mM. Serum samples were
pre-separated using Sep-Pac C18 cartridges. Samples
were worked up by removing 10 mL aliquots, adding 10
mL of acetonitrile, centrifuging at 14,000 rpm and using
10 mL of the clear supernatant for HPLC analysis. The
data were quantified from peak integrated areas of the
HPLC chromatograms.
Protein binding
Serum affinity binding experiments were conducted
by dissolving the prodrug in a minimum of methanol
and adding human serum to a final concentration of
about 100 mM. Samples were ultrafiltered using the
microcon YM-10 (Amicon Corp.) membrane centri-
fugation system and analyzed by HPLC. The values
obtained for 2 and 3 were 97ꢃ4% and 35ꢃ7%,
respectively.

W. Wrasidlo et al. / Bioorg. Med. Chem. Lett. 12 (2002) 557–560
559
Results and Discussion
The reaction scheme for the synthesis of prodrugs 2 and
3 and their activation to the parent compound 1 is
shown in Scheme. 1. One-step reactions resulted in these
compounds in high yield (>80%) and both compounds
were easily obtained in high purity (99%) by conven-
tional HPLC methods. Both prodrugs show, surpris-
ingly, exceptional solution stabilities for several weeks
in formulations (surfactant–ethanol, 1% saline combi-
nations) and under conditions encountered in clinical
applications (intravenous delivery). Under acidic condi-
tions (1% HCl) compound 2 converts to 3 within 2 h
with some degradation (about 10%) of the glycoside
moiety.
In preliminary experiments at neutral pH in PBS buf-
fered solutions, the prodrugs were very stable, but in the
presence of plasma or serum both 2 and 3 hydrolyzed to
1, with the elimination of soketal and glycerine, respec-
tively. When serum was fractionated by size exclusion
chromatography (G-200 column) it was found that all
the conversion to the parent drug occurred in a fraction
of molecular weights between 80 and 40 kD, which was
within the molecular weight range of carboxyl esterases.
The kinetics of prodrug hydrolysis were then determined
at 37 ꢁC in PBS buffer at various pH, in human serum
and in buffer solutions containing porcine liver esterase.
Figure 1. Transition state complex of ProVP-16 I and carboxyl
esterase. *HPD, hydrophobic positioning domain. See ref 17 for anion
hole.
Table 1. Kinetic parameters for hydrolysis of etoposide prodrugs
Compd
Media (37 ꢁCꢃ0.5)
pH (ꢃ0.1)
kobs (10ꢀ3 minꢀ1 a
)
t1/2 (min)
Prodrug I
2
PBS buffer
5.0–10.0
11.8
7.3
No conversion
8.333ꢃ0.50
—
83.16
750.8
14.17
—
216.49
97.03
6.93
Human serum esteraseb
PBS buffer
0.923ꢃ0.046
48.89ꢃ3.62
7.3
Prodrug II
3
5.0–7.3
8.0
8.8
10.5
7.3
No conversion
3.201ꢃ0.428
7.142ꢃ0.351
100.01ꢃ12.72
12.36ꢃ0.965
1.348ꢃ0.081
Human serum esterase
56.07
514.10
7.3
aMean ꢃSD.
bPorcine liver carboxyl ester hydrolase.
Table 2. Cytotoxicity of etoposide and prodrugs (XTT assay)
Cell line
Type
MDR-1
IC50 (mM)a
Etoposide
Prodrug I
Prodrug II
1
2
3
Molt 3
T-cell leukemia, parental
VP16 resistant Molt 3
T-lymphoblastoide VC res.
Colon carcinoma
Colon adenocarcinoma res.
Ovarian carcinoma parental
Doxo resistant A2780
ꢀ
++
+
0.065
7.55
15.5
10b
0.010
0.012
3.2
0.0035
0.11c
0.09
0.035
0.050
7.6
0.0057
0.11c
0.58
MOVP-3
VCR100
HT-29
SW480
A2780
ADR5000
ꢀ
+
ꢀ
50c
65
100
++
6.5
38
aAverage of duplicate determinations.
bOnly 35% of cells were affected.
c
ꢄ50% cells affected by drug.

560
W. Wrasidlo et al. / Bioorg. Med. Chem. Lett. 12 (2002) 557–560
In all experiments conversion–time curves exhibited
first-order kinetics from which linear log conversion–
time plots were obtained. The rate constants, kobs, and
half-lives, t1/2, were calculated from the slopes of these
plots. The data are summarized in Table 1.
Acknowledgements
We are very grateful to Dr. Muegge, Chemistry
Department, for the determination of the NMR spectra
and Dr. Janek, Biochemistry Department, for the mass
spectra. This work was supported by the DFG, Emmy-
Noether Program Lo 635/2-2.
In PBS buffer, 2 is surprisingly stable over a broad
range of pH from 5 to 10. Only above pH 11, this
prodrug hydrolyzes to give etoposide at relatively fast
rates with a half life of 83 min at pH 11.8. This contrasts
with reaction characteristics of 3 which undergoes base-
catalyzed hydrolysis above pH 8 at rates which rapidly
increase with increasing pH. We attribute the unusual
hydrolytic stability of 2 to the hydrophobic nature of
the entire southern region of this molecule and, to a
lesser extent, also to steric hindrance from the two ortho
methoxy groups surrounding the carbonate moiety. For
the esterase-catalyzed conversion, the rate constant for
2 is nearly 40 times that of 3. This increase in substrate
specificity of 2 over 3 for the enzyme may be due to the
presence of a hydrophobic positioning domain next to
the oxyanion hole within the catalytic region, with a
high affinity for the isopropylideneglyceryl side chain, as
shown in Figure 1. In human serum, both compounds
exhibit prodrug behavior. Interestingly, the activation
rate of 3 in human serum is 14 times faster than that of
2. Based on the dramatically increased affinity of 2 to
serum proteins, we suggest that a reduction of free 2 in
human serum (due to binding competition of various
proteins components) may contribute to its lower
hydrolysis rate. The relatively large differences in HPLC
elution times (C18 column) of 7.1 min for 3 compared to
18.5 min for 2 (lipophilic index) support the above
observations. In any event, the kinetic data presented
here clearly show the significance of specific structural
motives (other than those directly involved in the
catalytic process, i.e., sp2 carbonyl) in determining
the transformation of the prodrug to the parent
compound.
References and Notes
1. Hande, K. R. Eur. J. Cancer 1998, 34, 1514.
2. Slevin, M. L.; Clark, P. I.; Joel, S. P. et al. J. Clin. Oncol.
1989, 7, 1333.
3. Kobayashi, K.; Ratain, M. J. Cancer Chemother. Pharma-
col. 1994, 43, 120.
4. Gottesmann, M. M. Cancer Res. 1993, 53, 747.
5. Borst, P.; Evers, R.; Kool, M.; Wijnholds, J. Biochim. Bio-
phys. Acta 1999, 1461, 347.
6. Tze-Sing, H.; Chun-Chun, L.; Yee, C.; CH-Hung, S.; Li-
Tzong, C.; Min-Hsiung, C.; Chiou-Chung, Y.; Whang-Peng,
J. Pharmaceutical Res. 1999, 16, 997.
7. Zhang, Y. L.; Guo, X.; Cheng, Y. C.; Lee, K. H. J. Med.
Chem. 1994, 37, 446.
8. Daley, L.; Guminski, Y.; Demerseman, P.; Kruczynski, A.;
Etievant, C.; Imbert, T.; Hill, B.; Monneret, C. J. Med. Chem.
1998, 41, 4475.
9. Nicolaou, K. C.; Li, T.; Nakada, M.; Hummel, C. W.;
Hiatt, A.; Wrasidlo, W. Angew. Chem., Int. Ed. 1994, 33, 183.
10. Nicolaou, K. C.; Dai, M. W.; Tsay, S. C.; Estevez, V. A.;
Wrasidlo, W. Science 1992, 256, 1172.
11. Nicolaou, K. C.; Riemer, C.; Kerr, M. A.; Rideout, D.;
Wrasidlo, W. Nature 1993, 364, 464.
12. Niethammer, A.; Gaedicke, G.; Lode, H. N.; Wrasidlo,
W. Bioconj. Chem. 2001, 3, 414.
13. Wrasidlo W. Unpublished results.
14. 40-(200,300-Isopropylideneglyceryl carbonoxy)-etoposide: 1H
NMR (300 MHz, DMSO-d6) d 7.01 (s, 1H, H5), 6.54 (s, 1H,
0
H8), 6.28 (s, 2H, H2 , H6 ), 6.02 (s,1H, O-CH-O), 5.23 (d, 1H,
0
0
H1 ), 4.94 (d, 1H, H4), 4.72 (m, 2H, H2,H7 ), 4.58 (t, 2H, H1),
00
00
00
4.28 (d, 1H, H9a), 4.20–4.02, (m, 3H, H9b,H5 , H6 ), 3.60 (m,
3H, OCH3), 3.32 (m, 1H, H500), 3.28 (d, 1H, H2), 3.15 (d, 2H,
00 00
3 ,H4 ), 2.48 (s, 3H, CH3), 1,26 (m, 3H, C(2 ,3 ),CH3). 13C
00 00
H
Next, we tested the cytotoxicity of 1–3 on a number of
parental and drug-resistant tumor cell lines (Table 2).
Both 2 and 3 exhibit no initial cytotoxicity compared to
1. Their IC50 values after 72 h in cell culture (the drug
concentration of 50% cell viability) were significantly
below the parent drug, etoposide. With cell lines which
strongly expressed the MDR-1 gene (data not shown),
we observed significant differences of up to three orders
of magnitude. The nearly identical IC50 values for the
prodrugs between the parental cell line Molt-3 and its
etoposide-resistant clone (MOVP-3) are direct evidence
that these compounds completely circumvent drug
resistance mechanisms.
NMR 174.58, 152.33, 150.69, 147.83, 146.35, 139.11, 131.96,
128.93, 127.01, 109.77, 108.97, 107.28,. 101.49, 101.36, 98.57,
80.10, 74.37, 72.94, 72.69, 71.71, 68.52, 67.32, 65.75, 64.98,
55.80, 43.18, 40.34, 40.05, 39.78, 39.50, 39.23, 38.95, 38.67,
26.36, 25.32, 20.31. UV max 248 nm (e 1942, 271 nm (e 1504).
FAB LRMS m/e 746.85 (M+H)+ calc. for C36H42O17
746.70. HPLC elution time (C18 column, 50/50 CAN–H2O,
flow 0.5 mL/min), 18.46 min (etoposide 8.42 min).
15. 40-(200,300-dihydroxypropyl carbonoxy)-etoposide: FAB
LRMS m/e (M+H)+ 706.91, calcd for C33H38O17 706.64.
UV max 246 nm (e 1704), 273 nm (e 1355). HPLC elution time
(C18 column, 50 vol% ACN–H2O, flow 0.5 mL/min) 7.1 min.
16. Scudiero, D. A.; Shoemaker, R. H.; Paull, K. D.; Monks,
A.; Tierney, S.; Nofziger, T. H.; Currens, M. J.; Seniff, D.;
Boyd, M. R. Cancer Res. 1988, 48, 4827.
Detailed preclinical investigations are presently being
done and will be reported shortly.
17. Satoh, T.; Hosokawa, M. Annu Rev. Pharmacol. Toxicol.
1998, 38, 257.