Synthesis and biological evaluation of butanoate, retinoate, and bis(2,2,2-trichloroethyl)phosphate derivatives of 5-fluoro-2'-deoxyuridine and 2',5-difluoro-2'-deoxyuridine as potential dual action anticancer prodrugs

Article abstract of DOI:10.1002/(SICI)1521-4184(19998)332:8<286::AID-ARDP286>3.0.CO;2-9

A group of 3'-O-butanoyl, 5'-O-butanoyl, and 3',5'-di-O-butanoyl esters of 5-fluoro-2'-deoxyuridine (FDU), and 2',5-difluoro-2'-deoxyuridine (DFDU), 3'-O-retinoyl, and 3',5'-di-O-retinoyl esters of FDU and 5'-O-bis(2,2,2- trichloroethyl)phosphoryl-FDU and

Full text of DOI:10.1002/(SICI)1521-4184(19998)332:8<286::AID-ARDP286>3.0.CO;2-9

286
Knaus and co-workers
Synthesis and Biological Evaluation of Butanoate, Retinoate,
and Bis(2,2,2-trichloroethyl)phosphate Derivatives of
5-Fluoro-2′-deoxyuridine and 2′,5-Difluoro-2′-deoxyuridine as Potential
Dual Action Anticancer Prodrugs
a)
a)
b)
a),
Zuping Xia , Leonard I. Wiebe , Gerald G. Miller , and Edward E. Knaus *
a) Faculty of Pharmacy and Pharmaceutical Sciences and b) The Noujaim Institute for Pharmaceutical Oncology Research, University of Alberta, Edmonton,
Alberta, Canada T6G 2N8
Key Words: 5-Fluoro-2′-deoxyuridines; butanoyl and retinoyl dual action prodrug esters; cell differentiation; tumor growth
delay; anticancer activity
is believed to be primarily due to inhibition of DNA biosyn-
thesis by blocking thymidylate synthase (TS), the enzyme
which catalyzes the methylation of 2′-deoxyuridine-5′-mono-
phosphate to thymidine-5′-monophosphate. FDU must be
phosphorylated intracellularly to afford 5-fluoro-2′-de-
Summary
A group of 3′-O-butanoyl, 5′-O-butanoyl, and 3′,5′-di-O-butanoyl
esters of 5-fluoro-2′-deoxyuridine (FDU), and 2′,5-difluoro-2′-de-
oxyuridine (DFDU), 3′-O-retinoyl, and 3′,5′-di-O-retinoyl esters
of FDU, and 5′-O-bis(2,2,2-trichloroethyl)phosphoryl-FDU and
its 3′-O-butanoyl ester, was synthesized. These compounds were
designed to act as double prodrugs that would serve as a depot to
release two active drugs that act through different mechanisms.
Thus, a nucleotide derivative of FDU or DFDU could act as a
competitive inhibitor for thymidylate synthase, whereas retinoic
acid and butyric acid would be expected to induce cell differentia-
tion. The in vitro anticancer activities for these prodrugs were
determined against a panel of nine tumor types (leukemia, non-
small cell lung, colon, CNS, melanoma, ovarian, renal, prostate,
breast) that encompassed about 60 human tumor cell lines. Struc-
ture-activity relationships indicate that O-butanoyl esters of FDU
are approximately equipotent to FDU, the O-butanoyl esters of
DFDU are less active than FDU, and the retinoyl and bis(2,2,2-
trichloroethyl)phosphate derivatives of FDU exhibit comparable
activity to FDU. In addition to their cytotoxic effect, 3′-O-retinoyl-
FDU (12) and 3′-O-butanoyl-5′-O-bis(2,2,2-trichloroethyl)phos-
phoryl-FDU (16) also induced in vitro cell differentiation of
promyelocytic leukemia HL60 cells. These combined cytotoxic
and cell differentiation effects exhibited by 12 and 16 produced
greater morphological drug-induced granulation and neutrophil
vacuolation, and more cell apoptosis, than observed upon exposure
to either retinoic acid or sodium butanoate. Dose-escalation studies
in mice showed that 12 or 16 did not induce any acute or chronic
toxicity, change in plasma clinical chemistry parameters, or gross
histapathological changes at 60 days following an initial dosage
regimen of 0.013 mmol/kg ip for 7-consecutive days. The in vivo
growth delay response of murine mammary EMT6 solid tumors
suggests that 3′-O-retinoyl-FDU (12) delays tumor growth relative
to the other prodrugs investigated, sodium butyrate, retinoic acid,
FDU, or a combination of retinoic acid and FDU. These prelimi-
nary results suggest that 3′-O-retinoyl-FDU (12) warrants further
in vivo investigation to determine its tissue biodistribution and
pharmacokinetic parameters that would be of value in assessing its
potential usefulness as an anticancer prodrug.
oxyuridine-5′-monophosphate (FDUMP), the specific com-
[2,3]
petitive inhibitor for TS, to exert its cytotoxic effect
. The
cellular anabolism of FDU to FDUMP by pyrimidine kinase
is offset by the catabolic effect of nucleoside phosphorylase,
which cleaves the glycosyl bond, to yield 5-fluorouracil
[4,5]
(FU)
. These observations provide, at least in part, an
explanation why FDU is more potent than FU in vitro, but not
in vivo. Resistance to FDU can result from the depletion of
[6]
[7]
activating kinases and uridine phosphorylase , or TS
[8]
overproduction and gene amplification . Rapid elimination
(plasma t = 7.7 min) also contributes to the low efficacy
[9]
1/2
of FDU.
A number of strategies have been investigated to overcome
the low efficacy and development of resistance to FDU. Since
ionic nucleotides penetrate cells poorly and are readily
dephosphorylated extracellulary, the development of lipo-
philic neutral masked nucleotide prodrugs that can cross the
cell membrane and potentially liberate the nucleotide in-
[10–15]
tracellularly has been investigated extensively
of FDU have also been widely investigated
. Esters
[16–18]
. It has been
reported that O-butanoyl esters are more effective anticancer
agents than FDU following oral administration to mice bear-
ing adenocarcinoma-755 tumors, where the relative potency
order was 3′,5′-di-O-butanoyl > 3′-O-butanoyl > 5′-O-bu-
tanoyl. The greater antitumor efficacy of these esters is due
in part to their slower rate of hydrolysis by non-specific
[16]
esterase
.
Alternatively, agents that induce cell differentiation or in-
hibit cell proliferation constitute another approach for the
[19]
treatment of cancer . All-trans-retinoic acid (RA) has been
shown to induce cell differentiation, or to suppress cell pro-
liferation, in numerous tumor cell lines such as melanoma,
neuroblastoma, leukemia, breast tumor, epithelial can-
[20,21]
[22]
cer
, and papillomavirus-induced cancer . However,
pharmacokinetic studies have shown, following administra-
tion of RA (intravenous bolus dosing on 8-consecutive days),
Introduction
that the drug maximum velocity (V ) of capacity-limited
max
[23]
5-Fluoro-2′-deoxyuridine (FDU) has been used extensively elimination increased several fold . This may constitute the
for the clinical treatment of carcinomas of the ovary, breast, reason why, after a few months of complete remission for
[1]
and gastrointestinal tract . The carcinostatic activity of FDU maintenance RA therapy, the majority of patients suffering
Arch. Pharm. Pharm. Med. Chem.
© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999
0365-6233/99/0808/0286 $17.50 +.50/0

Potential Dual Action Anticancer Prodrugs
287
[24]
acute promyelocytic leukemia relapsed . Although butyric derivative, and their in vitro anticancer activities, in vitro
acid (BA) also induces malignant cell differentiation in vi- ability to induce differentiation in human leukemia HL60
[25–29]
tro
, it is clinically ineffective due to its very short cells and delay the growth of murine EMT6 solid tumors in
half-life and large distribution volume, both of which pre- vivo.
clude an in vivo therapeutic concentration. BA concentrations
reported to be effective for in vitro studies were 10-fold
Chemistry
higher than blood concentrations observed in clinical stud-
ies . However, butyric acid monosaccharide ester prodrugs
[30]
3′-O-Butanoyl-5-fluoro-2′-deoxyuridine (1), 5′-O-butan-
oyl-5-fluoro-2′-deoxyuridine (2), and 3′,5′-di-O-butanoyl-5-
fluoro-2′-deoxyuridine (3) were synthesized using the
procedures described by Nishizawa et al. . Reaction of
2′,5-difluoro-2′-deoxyuridine (4)
(1.13 equiv.) in pyridine, and separation of the products by
silica gel column chromatography afforded 3′-O-butanoyl-
2′,5-difluoro-2′-deoxyuridine (5, 19%), 5′-O-butanoyl-2′,5-
difluoro-2′-deoxyuridine (6, 14%) and 3′,5′-di-O-butanoyl-
2′,5-difluoro-2′-deoxyuridine (7, 33%), respectively (see
Scheme 1). Alternatively, 3′-O-butanoyl-2′,5-difluoro-2′-de-
oxyuridine (5) was also synthesized in 35% yield by elabora-
were subsequently found to provide improved maintenance
[31]
of cell differentiation, antitumor protection
plasma concentrations of BA
and constant
[16]
[32]
in mice. Pivalyloxymethyl
[36]
with butyric anhydride
butyrate increased the survival of mice bearing Lewis lung
[33,34]
carcinoma
. More recently, 1,2,3-tri-O-butyrylglycerol
(tributyrin), a prodrug of BA, was reported to be 4-fold more
potent than BA for inducing differentiation ofHL60 cells, and
that a combination with RA synergistically induced cell dif-
[35]
ferentiation
.
It was therefore of interest to design dual-action prodrugs
by chemical combination of anticancer nucleoside analogs
with a cell differentiation agent such as BA or RA. Since the tion of 2′,5-difluoro-2′-deoxyuridine (4) to the corresponding
malignant cell subpopulations of tumors are markedly hetero- 5′-O-trityl derivative, followed by reaction with butyric an-
geneous with respect to their rate of proliferation, a combina- hydride in pyridine, and then detritylation using 80% acetic
tion of two drugs acting by different mechanisms of action acid.
could have a synergistic effect to provide a broader spectrum
Reaction of all-trans-retinoic acid (8) with oxalyl chloride
[37]
of anticancer activity and enhanced clinical efficacy, relative in benzene yielded retinoyl chloride (9) , which, on con-
[16]
to the combined use of the individual drugs. It was envisaged densation with 5′-O-trityl-5-fluoro-2′-deoxyuridine (10)
that O-butanoyl and O-retinoyl prodrug esters would act as
depots to release the active drugs which could result in higher
plasma and target cell intracellular drug concentrations, and
improved biodistribution and pharmacokinetic properties
relative to the parent drugs. Previously, it was shown that
2′,5-difluoro-2′-deoxyuridine (DFDU), unlike FDU, was re-
sistant to glycosyl bond cleavage by phosphorylase due to the
stabilizing effect of the 2′-ribo F-substituent, and that it
provided a tumor-to-blood ratio of 10:3 at 4 h in BDF1 mice
[36]
bearing a Lewis lung tumor . We now report the synthesis
of3′-O-butanoyl, 5′-O-butanoyl and 3′,5′-di-O-butanoyl ester
derivatives of FDU and DFDU; the 3′-O-retinoyl and 3′,5′-
di-O-retinoyl ester derivatives of FDU; 5′-O-bis(2,2,2-
Scheme 1. Reagents and conditions: (i) butyric anhydride (1.13 equiv.),
pyridine, –10 °C (15 min) → –5 °C (48 h).
trichloroethyl)phosphoryl-FDU and its 3′-O-butanoyl ester
Scheme 2. Reagents and conditions: (i) oxalyl chloride, benzene, 25 °C, 1 h; (ii) 4-dimethylaminopyridine (DMAP), benzene, 0 °C (1 h) → 90 °C (3 h); (iii)
80% CH3CO2H, reflux, 10 min; (iv) DMAP, benzene, 0 °C (1 h) → reflux (3 h).
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

288
Knaus and co-workers
(cytotoxicity) , which were not sig-
nificantly different, showed a cyto-
toxicity profile that was more
similar to that of FDU than FU. In
contrast, the less cytotoxic and
equipotent butanoate esters of
DFDU (5–7) exhibited a cytotoxic-
ity profile more similar to the ref-
erence drug FU. Comparison of the
3′-O-retinoyl (12) and 3′,5′-di-O-
retinoyl (14) derivatives of FDU
showed that the potency order was
Scheme 3. Reagents and conditions: (i) DMAP, pyridine, benzene, (Cl3CCH2O)2P(=O)Cl, 25 °C (12 h) → reflux
(3 h); (ii) butyric anhydride, pyridine, 25 °C, 12 h.
3′-O-retinoyl- > 3′,5′-di-O-retinoyl
using 4-dimethylaminopyridine (DMAP) as catalyst, yielded
3′-O-retinoyl-5′-O-trityl-5-fluoro-2′-deoxyuridine (11,
79%). Detritylation of 11 using 80% acetic acid afforded
3′-O-retinoyl-5-fluoro-2′-deoxyuridine (12, 46%) as illus-
trated in Scheme 2. Condensation ofretinoyl chloride (9) with
in all tumor type panels, particularly at higher drug concen-
trations. Examination of the “mean” dose-response curves for
compounds 12 and 16 (see Figure 1) indicates that there is a
–5
concentration threshold around 10 M where the slope of the
growth inhibition curve increases rapidly. For example, when
FDU (13), using DMAP as catalyst, gave 3′,5′-di-O-retinoyl- the test compound concentration is lower than the “threshold”
–5
5-fluoro-2′-deoxyuridine (14, 65%).
(< 10 M), the cell growth inhibition profile is more similar
Treatment of FDU (13) with bis(2,2,2-trichloroethyl)phos- to that for FDU. In contrast, when the test drug concentration
–5
phorylchloridate using DMAP as catalyst afforded 5′-O- (12, 16) is greater than the “threshold” (> 10 M), there is a
rapid increase in the extent of cell growth inhibition which
was not observed for the reference drugs (FDU, FU). This
“threshold” effect may be due to the fact that there is a
required intracellular concentration for the parent ester (12,
16) and/or ester cleavage products which may have a lower
synergistic concentration relationship with respect to growth
bis(2,2,2-trichloroethyl)phosphoryl-5-fluoro-2′-deoxyurid-
ine (15, 35%). Subsequent reaction of 15 with butyric anhy-
dride yielded 3′-O-butanoyl-5′-O-bis(2,2,2-trichloroethyl)-
phosphoryl-5-fluoro-2′-deoxyuridine (16, 88%) (see Scheme
3).
[39]
inhibition . In the NCI screen, the 5′-O-bis(2,2,2-trichlo-
Results and Discussion
roethyl)phosphoryl-3′-O-butanoyl derivative of FDU (16, a
potential ester prodrug of a nucleotide derivative) provided
greater cell growth inhibition than the 5′-O-bis(2,2,2-trichlo-
roethyl)phosphoryl compound (15, a nucleotide derivative).
There is no evidence, to the best of our knowledge, that
5′-O-bis(2,2,2-trichloroethyl)phosphoryl derivatives of nu-
cleosides deliver the nucleoside monophosphate due to a
In Vitro Anticancer Screening
The butanoate, retinoate, and bis(2,2,2-trichloroethyl)phos-
phate esters of FDU and DFDU, and the reference drugs FU
and FDU, were evaluated in vitro against a diverse panel of
60 human tumor cell lines in culture derived from nine cancer
types (leukemia, non-small cell lung, colon, CNS, melanoma,
ovarian, renal, prostate, breast) by the Developmental Thera-
peutics Program, Division of Cancer Treatment, United
[14]
[40]
kinase-bypass process . Balzarini et al.
recently de-
scribed a class of cycloSal-nucleotide compounds that for the
first time delivers 2′,3′-didehydro-2′,3′-dideoxythymidine
monophosphate by a pH driven, chemically activated, tandem
reaction without the requirement for an enzymatic contribu-
[38]
States National Cancer Institute . This high-flux anticancer
drug screen has been designed to distinguish between broad
spectrum antitumor compounds and tumor- or subpanel-se-
lective agents. For each compound, dose-response curves for
each cell line were constructed from measurements at five
drug concentrations. In order to reduce the bulk raw data to
more interpretable categories, the data from each cell line of
a given tissue origin (for example, all breast lines) or tumor
type (for example, melanoma) were pooled, averaged and
plotted to produce a “mean” dose-response for each com-
pound to produce “mean activities”. The “mean” dose-re-
sponse curves for 3′-O-retinoyl-5-fluoro-2′-deoxyuridine
(12) and 5′-O-bis(2,2,2-trichloroethyl)phosphoryl-3′-O-bu-
–3
tion (kinase-bypass). Although 15 was less active (10 to
–8
10 M concentration range) than FDU, compound 16 was a
more potent inhibitor of cell growth at high drug concentra-
–5
tions (> 10 M), but a less active inhibitor at lower concen-
–5
tration (< 10 M), than FDU. A related study showed that
4
N -retinoyl derivatives of Ara-A were 25- to > 144-fold more
cytostatic, or equally active in the case of Ara-C, relative to
4
the parent nucleosides. These N -retinoyl conjugates were
stable to in vitro hydrolysis in human plasma and much more
[41]
lipophilic that the parent compounds
.
tanoyl-5-fluoro-2′-deoxyuridine (16), as well as the reference Drug Induced In Vitro Differentiation of Human Leukemia
drugs FDU and 5-fluorouracil (FU), for the nine major cancer
types that included 60 cell sublines are shown in Figure 1,
Panels A-I. The error bars represent the standard errors of the
“mean” data.
Some general structure-activity cytotoxicity profiles have
been derived from the NCI screening data and the “mean”
dose-response curves shown in Figure 1. The abilities of the
HL60 Cells
The ability of the 3′-O-retinoyl (12) and 3′-O-butanoyl-5′-
O-bis(2,2,2-trichloroethyl)phosphoryl (16) derivatives of
FDU, which were designed as dual-action prodrugs to the
cytotoxic FDU or the phosphoryl derivative thereof and either
RA or BA, respectively to induce cell differentiation war-
butanoate esters of FDU (1–3) to inhibit in vitro cell growth ranted investigation. Compounds 12 and 16 were selected for
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

Potential Dual Action Anticancer Prodrugs
289
Figure 1. (A) Average growth inhibition for eight human breast cancer cell lines: MCF7, MCF7/ADR-RES, MDA-MB-231/ATCC, HS 578T, MDA-MB-435,
MDA-N; BT-549, T-47D; (B) Average growth inhibition for six human CNS cancer cell lines: SF-268, SF-295, SF-593, SNB-19, SNB-75, U251; (C) Average
growth inhibition for seven human colon cancer cell lines: COLO 205, HCC-2998, HCT-116, HCT-15, HT29, KM12, SW-620; (D) Average growth inhibition
for six human leukemia cell lines: CCRF-CEM, HL-60(TB), K-562, MOLT-4, SR, RPMI-8226; (E) Average growth inhibition for nine human non-small
cell lung cancer cell lines: A549/ATCC, EKVX, HOP-62, HOP-92, NCI-H226, NCI-H23, NCI-H322M, NCI-H460, NCI-H522; (F) Average growth
inhibition for eight human melanoma cancer cell lines: LOX-IMVI, MALME-3M, M14, SK-MEL-2, SK-MEL-28, SK-MEL-5, UACC-257, UACC-62;
(G) Average growth inhibition for six human ovarian cancer cell lines: IGROV1, OVCAR-3, OVCAR-4, OVCAR-5, OVCAR-8, SK-OV-3; (H) Average
growth inhibition for two human prostate cancer cell lines: PC-3, DU-145; (I) Average growth inhibition for eight human renal cancer cell lines: 786-0, A498,
ACHN, CAKI-1, RXF-393, SN12C, TK-10, UO-31.
the in vitro differentiation studies based on their superior at a dosage of 500 mg/kg/day for 10 days produced BA
–6
activity in the previously described NCI anticancer screen. plasma concentrations in the 30-to-59 × 10 M range, that
Examination of the results (see Figure 2) show that 12, 16,
are at least 10-fold lower than those reported to effectively
and the reference drugs RA and sodium butanoate (NaBu) induce in vitro differentiation. Furthermore, when the infu-
induced concentration-dependent differentiation in human sion was terminated, the plasma BA concentration decreased
leukemia HL60 cells upon incubation for 7 days at 37 °C. No
change in cell morphology was observed using a vehicle
control group (final DMSO concentration < 0.1%) which
indicates that DMSO did not change cell morphology. Ac-
cordingly, their relative ability to induce 50% cell differen-
rapidly (t = 6.1 ± 1.4 min) returning to pretreatment levels
within one hour . Changes in cell morphology upon incu-
bation with 12 or 16 produced a large amount of drug-induced
cell granulation and neutrophil vacuolation, and more exten-
sive apoptosis as indicated by flow cytometry studies (data
1/2
[30]
–6
–5
tiation (EC , M) was 12 (1 × 10 M) > RA (3.1 × 10 M) not shown) relative to RA or NaBu. These morphological
50
–5
–3
> 16 (1 × 10 M) > NaBu (6.8 × 10 M). The observation changes could be due, at least in part, to a synergistic effect
that the ester prodrug is much more potent than the acid (12 > of the cytotoxic nucleoside moiety and the differentiation
RA; 16 > NaBu) augurs well for the possibility that 12 and effect of the acid moiety. In contrast, it has been reported that
4
16 may be clinically superior to RA and BA, respectively, the N -retinoyl conjugates of Ara-A and Ara-C, which are
depending on their in vivo biodistribution and/or rate of ester stable to in vitro hydroylsis in human plasma, are at least 2 to
cleavage. This postulate (16 > NaBu) is based on a study 3 orders of magnitude less capable of inducing differentiation
[41]
which showed that intravenous infusion of NaBu into humans of HL-60 cells compared to retinoic acid
.
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

290
Knaus and co-workers
studies using RA alone, or RA in combination with FDU,
were suspended midway through the study due to acute, and
then chronic, toxicity. Although a lower body weight up to
30% was observed at Day 10 for the highest dosage-regimen
group (12, 16), the weight of these mice at 60 days was not
significantly different relative to a control group. The two
groups of mice that received the highest dosage-regimen
(0.013 mmol/kg × 7 days) of 12 and 16 were subjected to
clinical chemistry plasma analyses, and gross and his-
topathologic postmortem examination at Day 60. There were
no differences between the treated and control groups, which
indicates that 12 or 16 had no effect on the serum chemistry
parameters measured (see experimental section for parame-
ters measured), gross anatomy or microscopic anatomy (see
experimental section for organs examined).
Dose Escalation, Clinical Chemistry, and Gross His-
topathologic Postmortem Assessment
A dose-escalation study involving the administration of 12
or 16 to mice once-a-day for seven consecutive days, starting
from an ip dose of 0.00013 mmol/kg that was increased in
2-fold increments to a dose of 0.013 mmol/kg, was per-
formed. Daily observation of each dosage-regimen group (12
or 16) of mice up to 60 days from Day 1 did not indicate any
obvious acute or chronic toxicity. In contrast, dose escalation
Growth Delay Responses of EMT6 Mouse Mammary Tumors
Following Treatment with the Prodrugs 12 or 16, Compared
to RA, NaBu or FDU Alone, or in Combination
The in vivo anticancer efficacy of the prodrugs 12 and 16
was determined by comparison of the tumor volume for
several drug treatment groups relative to an untreated control
group. Tumor growth delay time (days) was calculated as the
increased time required to reach a specified tumor volume
3
(100, 200, 300, 400, 500 mm ), at doses of 0.0075 and
0.015 mmol/kg ip, for RA, NaBu, FDU, FDU + RA, FDU +
NaBu, 12 and 16 dosage regimens relative to a non-treated
Figure 2. Percentage HL60 cell differentiation after incubation with RA,
NaBu, 12 or 16 for 7 days at 37 °C.
Table 1. In vivo growth delay responses of EMT6 mouse mammary tumors upon treatment with the prodrugs 12 or 16 compared to RA, NaBu or FDU alone,
or in combination.
—————————————————————————————————————————————————————————————–
Dose
(mmol/
kg ip)
No. of
mice
Days ± SD for
tumor to grow
to 100 mm3
Days ± SD for
tumor to grow
to 200 mm3
Days ± SD for
tumor to grow
to 300 mm3
Days ± SD for
tumor to grow
to 400 mm3
Days ± SD for
tumor to grow
to 500 mm3
(Growth Delay)
(Growth Delay)
(Growth Delay)
(Growth Delay)
(Growth Delay)
—————————————————————————————————————————————————————————————–
Control
RA
30
7
9 ± 2
11 ± 2
12 ± 3
14 ± 4
15 ± 4
0.0075
0.015
9 ± 1 (0)
9 ± 1 (0)
9 ± 1 (0)
9 ± 1 (0)
8 ± 0 (–1)
8 ± 1 (–1)
14 ± 4 (5)
13 ± 5 (4)
10 ± 2 (1)
9 ± 1 (0)
19 ± 9 (10)
17 ± 4 (8)
13 ± 7 (4)
12 ± 3 (3)
10 ± 2 (–1)
10 ± 1 (–1)
10 ± 2 (–1)
10 ± 2 (–1)
10 ± 1 (–1)
13 ± 6 (2)
16 ± 4 (5)
17 ± 6 (6)
11 ± 3 (0)
10 ± 1 (–1)
22 ± 9 (11)
19 ± 6 (8)
14 ± 7 (3)
13 ± 3 (2)
12 ± 2 (0)
11 ± 1 (–1)
11 ± 2 (–1)
1 1 ± 2 (–1)
12 ± 1 (0)
15 ± 6 (3)
17 ± 5 (5)
20 ± 5 (8)
12 ± 3 (0)
11 ± 2 (–1)
24 ± 9 (12)
21 ± 7 (9)
16 ± 7 (4)
16 ± 3 (4)
13 ± 3 (–1)
12 ± 1 (–2)
12 ± 2 (–2)
12 ± 4 (–2)
14 ± 3 (0)
18 ± 6 (4)
18 ± 4 (4)
23 ± 6 (9)
1 4 ± 4 (0)
14 ± 4 (0)
26 ± 9 (12)
23 ± 7 (9)
17 ± 7 (3)
18 ± 4 (4)
15 ± 5 (0)
13 ± 2 (–2)
13 ± 3 (–2)
14 ± 5 (–1)
1 6 ± 3 (1)
19 ± 7 (4)
20 ± 5 (5)
24 ± 7 (9)
15 ± 4 (0)
14 ± 4 (–1)
30 ± 11 (15)[c]
24 ± 6 (9)[c]
19 ± 7(4)[c]
19 ± 5 (4)[c]
8
NaBu
0.0075
0.015
8
8
FDU
0.0075
0.015
6
6
FDU + RA
FDU + NaBu
12
0.0075
0.015
6
7[a]
0.0075
0.015
8
7[a]
13
6[a,b]
7[a]
14
0.0075
0.015
16
0.0075
0.015
—————————————————————————————————————————————————————————————–
[a] Complete tumor regression occurred in one mouse on treatment, which was excluded from the calculation for this treatment group.
[b] One mouse in this treatment group, which did not develop a palpable tumor, was excluded from the calculation for this group.
[c] Statistical evaluation (unpaired T-test) was performed, for tumors which grew to a size of 500 mm3, where the effect of the prodrug 12 or 16 was compared
to that of a mixture of its two prodrug moieties.
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

Potential Dual Action Anticancer Prodrugs
291
control group (see Table 1). The EMT6 tumor model was mine its in vivo biodistribution and pharmacokinetic parame-
selected since EMT6 cells are responsive to either RA or ters which would be of value to predict its potential utility as
NaBu in vitro. In this study, both complete regression (CR), an anticancer prodrug.
and mice that did not develop a transplanted tumor, were
excluded from the tumor growth delay calculation.
For example, 12 showed P values (at 0.0075 and
Acknowledgments
0.015 mmol/kg, respectively) of 0.00022 and 0.08773 when
compared to FDR alone, P values of 0.00035 and 0.00348
when compared to RA alone, and P values of 0.00428 and
0.41893 (if the animal that was cured had been included in
this analysis, this P value would have been much lower) when
compared to a mixture of FDU + RA. Although the P values
for 16 (at 0.0075 and 0.015 mmol/kg, respectively) compared
to FDU were not significant (0.15810 and 0.48521, the P
values compared with NaBu (0.03692 and 0.01894), and with
FDU + NaBu (0.09688 and 0.02472), were significant.
A number of the drug treatment regimens (FDU + RA,
0.015 mmol/kg; FDU + NaBu, 0.015 mmol/kg; 12,
0.015 mmol/kg; 16, 0.0075 mmol/kg) produced one tumor-
free survivor (complete tumor regression) at 60 days in the
group. In contrast, no instance of complete tumor regression
was observed for the RA, NaBu or FDU single drug regimen
groups. Furthermore, the extent of tumor growth delay (days)
induced by the prodrugs 12 and 16 was greater, in most cases,
than that observed for either single prodrug moiety (RA,
NaBu, FDU) alone, or in combination. This latter observation
could be due, at least in part, to superior biodistribution
properties and/or pharmacokinetic parameters associated
with the prodrugs 12 and 16 relative to RA, NaBu or FDU
singly, or in combination. Although RA or NaBu did not
prevent tumor growth, FDU (0.015 mmol/kg) provided a
modest 4-day growth delay in development of larger tumors
We are grateful to the Alberta Cancer Board (Grant No. R-266) for
financial support of this research, and to the United States National Cancer
Institute for performing the in vitro cytotoxicity screen. We also thank Dr.
Nick Nation, University of Alberta Health Sciences Laboratory and Services,
for performing the serum clinical chemistry and histological studies.
Experimental Section
All solvents and reagents used were of reagent grade quality. Melting
points were determined on a Thomas Hoover capillary apparatus and were
uncorrected. Nuclear magnetic resonance spectra (1H NMR, 13C NMR, 19
F
NMR) were determined on a Bruker AM-300 spectrometer using
{D]CF3CO2H as an internal standard (19F NMR). The assignments of all
exchangeable protons (OH, NH) were confirmed by addition of [D2]H2O.
13C NMR spectra were acquired using the J-modulated spin echo technique
where methyl and methine carbon resonances appear as positive peaks and
methylene and quaternary carbons appear as negative peaks. Microanalyses
were within ± 0.4% of theoretical values for all elements listed, unless
otherwise noted. Silica gelcolumnchromatographywasperformedusingEM
SCIENCE silica gel 60 (70–230 mesh). 5-Fluoro-2′-deoxyuridine (13),
bis(2,2,2-trichloroethyl)phosphorochloridate and all-trans-retinoic acid
were purchased from the Sigma Chemical Co. and butyric anhydride was
purchased from the Eastman Kodak Co. The retinoyl compounds were
synthesized under an argon atmosphere in the dark or with the lights dimmed.
3′-O-Butanoyl-5-fluoro-2′-deoxyuridine (1)[16], 5′-O-butanoyl-5-fluoro-2′-
deoxyuridine (2)[16], 3′,5′-di-O-butanoyl-5-fluoro-2′-deoxyuridine (3)[16]
,
2′,5-difluoro-2′deoxyuridine (4)[36] and 5′-O-trityl-5-fluoro-2′-deoxyuridine
(10)[16] were synthesized according to the literature procedures. HL60 (origi-
nating from the American Type Tissue Culture Collection) and EMT6/Ed
cells were obtained from the Cross Cancer Institute, University of Alberta,
Edmonton, Alberta. Female BDF1 mice were purchased from the Charles
River, Montreal, Canada. Female Balb/c mice were purchased from the
Health Sciences Laboratory Animal Services Unit at the University of
Alberta and were 17–21 g in weight when studies were initiated.
3
(400–500 mm ). A combination of RA + FDU was more
effective than FDU alone. Although it may appear that the
3′-O-retinoyl prodrug delayed tumor growth (all tumor vol-
3
umes from 100–500 mm ) more effectively for a
0.0075 mmol/kg compared to a higher 0.015 mmol/kg dos-
age regimen, the differences in tumor growth delay were not
statistically significant. Similarly, the 3′-O-butanoyl prodrug
16 was approximately equipotent for either a 0.0075 or
0.015 mmol/kg dosage regimen.
3′-O-Butanoyl-2′,5-difluoro-2′-deoxyuridine 5, 5′-O-Butanoyl-2′,5-di-
fluoro-2′-deoxyuridine
oxyuridine 7
6
and 3′,5′-Di-O-butanoyl-2′,5-difluoro-2′-de-
A solution of 4 (184 mg, 0.7 mmol) in dry pyridine (20 ml) was stirred for
15 minutes at –10 °C and then butyric anhydride (130 ml, 0.79 mmol) was
added. The reaction mixture was maintained at –5 °C for 48 h with stirring,
the solvent was removed in vacuo, and the residue was co-evaporated with
toluene twice. Thisresidue was purifiedbysilica gel columnchromatography
using a gradient of 0–2.5% v/v methanol in chloroform. Three fractions were
separated and collected which exhibited Rf values (micro TLC,
MeOH:CHCl3, 1:20 v/v as development solvent) of 0.66 (5), 0.38 (6), and
0.27 (7), respectively (66% total yield).
Product 5 was obtained as a white solid upon precipitation from EtOAc-
n-hexane (44 mg, 19%): mp 155–156 °C; 1H NMR (CHCl3-d1): δ 0.98 (t, J
= 7.4 Hz, 3H, CH2CH3), 1.68 (sextet, J = 7.4 Hz, 2H, CH2CH2CH3), 2.40 (t,
J = 7.4 Hz, 2H, COCH2CH2), 3.13 (br s, 1H, OH), 3.82 (dd, Jgem = 12.4, J4′,5′
= 1.8 Hz, 1H, H-5′), 4.05 (dd, Jgem = 12.4, J4′,5” = 1.8 Hz, H-5”), 4.27 (d,
Conclusions
A group of potential dual-action prodrug esters was de-
signed to act as depots for the in vivo release of two active
drugs that include FDU, DFDU or 5′-O-bis(2,2,2-trichlo-
roethyl)phosphoryl-FDU (inhibitors of thymidylate syn-
thase), and either BA or RA (induce cell differentiation).
Preliminary in vitro anticancer screen results indicate that the
3′-O-retinoyl esters of FDU and 5′-O-bis(2,2,2-trichlo-
roethyl)phosphoryl-3′-O-butanoyl-5-fluoro-2′-deoxyuridine
are more potent than FDU at a test compound concentration
–5
J3′,4′ = 5.9 Hz, 1H, H-4′), 5.16–5.40 (m, 2H, H-2′, H-3′), 6.01 (dd, J1′,F
=
≥10 M. In vitro cell differentiation and in vivo EMT6 tumor
15.9, J1′,2′ = 2.2 Hz, 1H, H-1′), 8.05 (d, J5F,6 = 6.2 Hz, 1H, H-6), 9.80 (s, 1H,
NH); 13C NMR (CHCl3-d1): δ 13.5 (CH2CH3), 18.3 (CH2CH3), 35.6
(COCH2), 60.5 (C-5′), 69.2 (d, J2′F,3′ = 15.4 Hz, C-3′), 82.4 (C-4′), 88.8 (d,
growth delay studies suggest that the 3′-O-retinoyl ester
prodrug 12 is the most promising (potent) anticancer agent
that exhibits a broad spectrum of anticancer activities relative
to other prodrug esters investigated, as well as to FDU, RA
or BA alone or in physical combination. Accordingly, these
preliminary results suggest that 3′-O-retinoyl-5-fluoro-2′-de-
oxyuridine (12) warrants further in vivo evaluation to deter-
J1′,2′F = 34 Hz, C-1′), 91.1 (d, J2′,2′F = 194 Hz, C-2′), 125.1 (d, J5F,6 = 34.6
Hz, C-6), 140.6 (d, J5,5F = 119 Hz, C-5), 148.9 (C-2 C=O), 157.2 (d, J = 27.0
Hz, C-4 C=O), 173.1 (CO2); 19F NMR (CHCl3-d1, CF3CO2H-d1 internal
standard): δ –88.4 (d, J5F,6 = 6.2 Hz, F-5), –127.6 (ddd, J2′,F = 52.6, J1′,F
14.7, J3′,F =14.7 Hz, F-2′). Anal. (C13H16F2N2O6) C, H, N.
=
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

292
Knaus and co-workers
Product 6 was obtained as a white solid upon precipitation from EtOAc-
which was precipitated from acetone-water to afford 12 as a white solid
(100 mg, 46%): mp 79–80 °C; 1H NMR (CHCl3-d1): δ 1.04 (s, 6H, cyclo-
hexene C-6 Me′s), 1.46–1.67 (m, 4H, cyclohexene H-4 and H-5), 1.72 (s, 3H,
cyclohexene C-2 Me), 2.01–2.11 (m, 5H, retinoyl C-7 Me and cyclohexene
H-3), 2.28–2.54 (m, 5H, H-2′ and retinoyl C-3 Me), 4.0–4.07 (m, 2H, H-5′),
4.16–4.22 (m, 1H, H-4′), 5.36–5.42 (m, 1H, H-3′), 5.79 (s, 1H, retinoyl H-2),
6.13–6.37 (m, 5H, H-1′ and retinoyl H-4, H-6, H-8 and H-9), 7.06 (dd, J =15,
J = 11.4 Hz, 1H, retinoyl H-5), 8.08 (d, J5F,6 = 6.3 Hz, 1H, H-6), 9.54 (br s,
1H, NH); 13C NMR (CHCl3-d1): δ 13.0 (retinoyl C-7 Me), 14.1 (retinoyl C-3
Me), 19.3 (cyclohexene C-4), 21.8 (cyclohexene C-2 Me), 29.0 (cyclohexene
C-6 Me′s), 33.2 (cyclohexene C-3), 34.3 (cyclohexene C-6), 38.0 (C-2′), 39.7
(cyclohexene C-5), 62.7 (C-5′), 73.7 (C-3′), 85.4 (C-1′), 85.8 (C-4′), 116.8
(retinoyl C-2), 124.5 (d, J5F,6 = 34 Hz, C-6), 129.2 and 129.3 (retinoyl C-6
and C-9), 130.2 (cyclohexene C-2), 132.0 (retinoyl C-5), 134.5 (retinoyl
C-4), 137.1 (retinoyl C-8), 137.6 (cyclohexene C-1), 140.4 (retinoyl C-7),
140.7 (d, J5,5F = 240 Hz, C-5), 148.5 (C-2 C=O), 155.1 (retinoyl C-3), 156.7
(d, J4,5F = 26.2 Hz, C-4 C=O), 166.6 (CO2). The 1H and 13C NMR resonances
assigned to the retinoic ester moiety resonances are in agreement with those
previously assigned to retinoic acid[42] and methyl retinoate[43]. Anal.
(C29H37FN2O6) H, N; C: calcd, 65.83; found, 64.42.
n-hexane (34 mg, 14%): mp 58–60 °C; 1H NMR (CHCl3-d1): δ 0.96 (t, J =
7.4 Hz, 3H, CH2CH3), 1.67 (sextet, J = 7.4 Hz, 2H, CH2CH2CH3), 2.37 (t, J
= 7.4 Hz, 2H, COCH2CH2), 3.37 (br s, 1H, OH), 4.23–4.38 (m, 2H, H-3′,
H-4′), 4.45 and 4.52 (two dd, Jgem = 13.1, J4′,5′ = 1.8 Hz, 1H each, H-5′
hydrogens), 5.08 (d, J2′,F = 52.6 Hz, 1H, H-2′), 5.91 (J1′,2′F = 16.7 Hz, 1H,
H-1′), 7.78 (d, J5F,6 = 6.0 Hz, 1H, H-6), 9.90 (s, 1H, NH); 13C NMR
(CHCl3-d1): δ 13.5 (CH2CH3), 18.3 (CH2CH3), 35.8 (COCH2), 61.9 (C-5′),
68.8 (d, J2′F,3′ = 16.7 Hz, C-3′), 80.8 (C-4′), 88.9 (d, J1′,2′F = 35.4 Hz, C-1′),
93.3 (d, J2′,2′F = 188 Hz, C-2′), 124.2 (d, J5F,6 = 34.6 Hz, C-6), 140.6 (d, J5,5F
= 238 Hz, C-5), 148.7 (C-2 C=O), 157.0 (d, J4,5F = 26.8 Hz, C-4 C=O), 173.3
(CO2); 19F NMR (CHCl3-d1, CF3CO2H-d1 internal standard): δ –88.7 (d,
J5F,6 = 6.0 Hz, F-5), -126.6 (ddd, J2′,2′F = 52.6, J3′,F = 21.1, J1′,F = 17.4 Hz,
F-2′). Anal. (C13H16F2N2O6) C, H, N.
Product 7 was obtained as a white solid upon precipitation from CHCl3-n-
hexane (99 mg, 33%): mp 87–88 °C; 1H NMR (CHCl3-d1): δ 0.97 (t, J =7.4
Hz, 6H, CH2CH3), 1.69 (sextet, 4H, CH2CH2CH3), 2.37 and 2.40 (two
overlapping t, J = 7.4 Hz, 2H each, COCH2CH2), 4.35–4.50 (m, 3H, H-4′,
H-5′), 5.07 (ddd, J2′F,3′ = 18.6, J3′,4′ = 8.2, J2′,3′ = 4.5 Hz, 1H, H-3′), 5.26 (dd,
J2′,2′F = 51.6, J2′,3′ = 4.5 Hz, 1H, H-2′), 5.96 (d, J1′,2′F = 16.8 Hz, 1H, H-1′),
7.73 (d, J5F,6 = 6.1 Hz, 1H, H-6), 9.15 (s, 1H, NH); 13C NMR (CHCl3-d1): δ
13.43 and 13.52 (two CH2CH3), 18.3 (CH2CH3), 35.5 and 35.8 (two
COCH2), 61.4 (C-5′), 68.8 (d, J2′F,3′ = 15.8 Hz, C-3′), 78.8 (C-4′), 89.4 (d,
J1′,2′F = 37.5 Hz, C-1′), 90.9 (d, J2′,2′F = 194 Hz, C-2′), 123.8 (d, J5F,6 = 34
3′,5′-Di-O-Retinoyl-5-fluoro-2′-deoxyuridine 14
Oxalyl chloride (60 µl, 0.674 mmol) was added to a solution of 8 (134 mg,
0.446 mmol) in dry benzene (8 ml) with stirring at 25 °C and the reaction
was allowed to proceed for 1 h under an atmosphere of argon. The solvent
was removed in vacuo, the residue was dissolved in dry benzene (5 ml), and
this solution was added to a solution of 13 (53 mg, 0.215 mmol) and 4-di-
methylaminopyridine (84 mg, 0.681 mmol) in dry benzene (5 ml) at 0 °C
under an argon atmosphere. The reaction was allowed to proceed at 0 °C for
1 h prior to heating at reflux for 3 h. Removal of the solvent in vacuo and
purification of the residue by silica gel column chromatography using a
gradient of 0–15% EtOAc in CHCl3 afforded 14 [114 mg, 65%; Rf 0.64 using
EtOAc-CHCl3 (1:3, v/v) as development solvent] as a yellow solid upon
precipitation using acetone-water: mp 98–100 °C; 1H NMR (CHCl3-d1): δ
1.04 (s, 12H, cyclohexene C-6 Me′s), 1.46–1.70 (m, 8H, cyclohexene H-4
and H-5), 1.73 (s, 6H, cyclohexene C-2 Me′s), 2.02–2.23 (m, 11H, H-2′,
retinoyl C-7 Me′s, cyclohexene H-3), 2.37 and 2.40 (two s, 3H each, retinoyl
C-3 Me′s), 2.50–2.60 (m, 1H, H-2′′), 4.30–4.36 (m, 1H, H-4′), 4.37–4.55 (m,
2H, H-5′), 4.26–4.32 (m, 1H, H-3′), 5.75 and 5.78 (two s, 1H each, retinoyl
H-2), 6.13–6.24 (m, 9H, retinoyl H-4, H-6, H-8 and H-9, H-1′), 7.0–7.12 (m,
2H, retinoyl H-5), 7.80 (d, J5F,6 = 6.0 Hz, 1H, H-6); 13C NMR (CHCl3-d1):
δ 13.0 (retinoyl C-7 Me′s), 14.1 (retinoyl C-3 Me′s), 19.3 (cyclohexene
C-4′s), 21.8 (cyclohexene C-2 Me′s), 29.0 (cyclohexene C-6 Me′s), 33.2
(cyclohexene C-3′s), 34.3 (cyclohexene C-6′s), 38.3 (C-2′), 39.7 (cyclo-
hexene C-5′s), 62.8 (C-5′), 73.3 (C-3′), 83.2 (C-1′), 85.5 (C-4′), 116.0 and
116.6 (retinoyl C-2′s), 123.5 (d, J5F,6 = 34.9 Hz, C-6), 129.2 and 129.3
(retinoyl C-6′s and C-9′s), 130.2 (cyclohexene C-2′s), 132.1 and 132.2
(retinoyl C-5′s), 134.4 and 134.5 (retinoyl C-4′s), 137.1 (retinoyl C-8′s),
137.7 (cyclohexene C-1′s), 140.5 (retinoyl C-7′s), 140.7 (d, J5,5F = 238 Hz,
Hz, C-6), 140.7 (d, J5,5F = 239 Hz, C-5), 148.3 (C-2 C=O), 156.4 (d, J4,5F
=
27 Hz, C-4 C=O), 172.5 and 172.8 (two CO2); 19F NMR (CHCl3-d1,
CF3CO2H-d1 internal standard): δ –88.2 (d, J5F,6 = 6.1 Hz, F-5), 125.2 (ddd,
J2′,2′F = 52, J1′,2′F = 17.7, J3′,2′F = 17.7 Hz, F-2′). Anal. (C17H22F2N2O7) C,
H, N.
3′-O-Butanoyl-2′,5-difluoro-2′-deoxyuridine 5
A solution containing 4 (92 mg, 0.348 mmol) and trityl chloride (117 mg,
0.411 mmol) was stirred at 25 °C for 6 h, followed by heating at 100 °C for
3 h. The reaction mixture was cooled to 25 °C, butyric anhydride (285 µl,
2.2 mmol) was added and the reaction was allowed to proceed for 12 h at
25 °C with stirring. Removal of the solvent in vacuo gave a residue to which
toluene (10 ml) was added and immediately removed by evaporation (two
times). Acetic acid (4 ml of 80% w/v) was added to the residue and the
solution was heated at reflux for 10 min. Removal of the solvent in vacuo
gave a residue which was purified by silica gel column chromatography using
a gradient of 0–50% EtOAc in CHCl3 as eluent to yield 5 as a white solid
upon precipitation using EtOAc-n-hexane (41 mg, 35%). Product 5 was
identical (mp, 1H NMR) to that described in the previous procedure.
3′-O-Retinoyl-5′-O-trityl-5-fluoro-2′-deoxyuridine 11
Oxalyl chloride (50 µl, 0.922 mmol) was added to a solution of all-trans-
retinoic acid(8, 185 mg, 0.615 mmol) in dry benzene (10 ml) and the reaction
mixture was stirred for 1 h at 25 °C. Removal of the solvent in vacuo gave a
residue which was dissolved in dry benzene (5 ml). A solution of 10 (201 mg,
0.410 mmol) and 4-dimethylaminopyridine (79 mg. 0.64 mmol) in dry ben-
zene (10 ml) was added, and the reaction mixture was maintained at 0 °C for
1 h followed by heating at 90 °C for 3 h under an argon atmosphere. The
reaction mixture was cooled to 25 °C and washed with water (3 × 15 ml)
which was discarded. Removal of the solvent from the organic fraction in
vacuo gave a residue which was purified by elution from a silica gel column
using EtOAc-CHCl3 (1:9, v/v) as eluent (performed in the dark). Collection
of the fractions having Rf 0.62 (micro TLC, n-hexane-EtOAc, 1:1, v/v as
development solvent) yielded 11 (250 mg, 79%) that was used immediately
in the subsequent reaction.
C-5), 148.5 (C-2 C=O), 155.2 and 155.6 (retinoyl C-3′s), 156.3 (d, J4,5F
=
27.2 Hz, C-4 C=O), 166.0 and 166.2 (CO2′s). Anal. (C49H63FN2O7) C,H, N.
5′-O-Bis(2,2,2-trichloroethyl)phosphoryl-5-fluoro-2′-deoxyuridine 15
A solution of bis(2,2,2-trichoroethyl)phosphorochloridate (312 mg,
0.820 mmol) in dry benzene (5 ml) was added with stirring to a solution of
13 (101 mg, 0.409 mmol) and 4-dimethylaminopyridine (88 mg,
0.713 mmol) in a mixture of pyridine (5 ml) and benzene (5 ml). The result-
ing suspension was stirred for 12 h at 25 °C prior to heating at reflux for 3 h.
Removal of the solvent in vacuo gave a residue to which toluene (10 ml) was
added and immediately evaporated (two times). This residue was purified by
silica gel column chromatography using a gradient of 1–5% MeOH in CHCl3.
The fractions which exhibited Rf 0.43 (micro TLC using EtOAc as develop-
ment solvent) were combined and recrystallized from CHCl3 to afford 15 as
a white solid (85 mg, 35%): mp 148–149 °C; 1H NMR (CH3OH-d4): δ
2.20–2.40 (m, 2H, H-2′), 4.06–4.15 (m, 1H, H-4′), 4.38–4.52 (m, 3H, H-3′,
H-5′), 4.76–4.84 (m, 4H, OCH2CCl3), 6.23 (dd, J1′,2′ = J1′,2′′ = 6.7 Hz, 1H,
H-1′), 7.83 (d, J5F,6 = 6.6 Hz, 1H, H-6); 13C NMR (CH3OH-d4): δ 40.5 (C-2′),
3′-O-Retinoyl-5-fluoro-2′-deoxyuridine 12
A solution of 11 (250 mg, 0.324 mmol) in 80% acetic acid (10 ml) was
heated at reflux for 10 min under an argon atmosphere. Removal of the
solvent in vacuo gave a residue which was purified by silica gel column
chromatography using a gradient of 0–3% MeOH in CHCl3 as eluent (per-
formed in the dark). Collection of the fractions having Rf 0.39 (micro TLC
using EtOAc-CHCl3, 1:1, v/v as development solvent) yielded a residue
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

Potential Dual Action Anticancer Prodrugs
293
RA and NaBu. No change in cell morphology was observed using a vehicle
control group.
69.9 (d, J5′,P = 5.9 Hz, C-5′), 71.6 (C-3′), 78.5 (d, JCH2O,P = 4.3 Hz, CH2OP),
86.1 (d, J4′,P = 7.6 Hz, C-4′), 87.2 (C-1′), 95.9 (d, JCl3C,P = 10.8 Hz, CCl3),
125.9 (d, J5F,6 = 34.9 Hz, C-6), 141.9 (J5,5F = 234 Hz, C-5), 150.6 (C-2 C=O),
159.3 (d, J4,5F = 25.1 Hz, C-4). Anal. (C13H14FN2O8P) C, H, N.
Dose Escalation, Clinical Chemistry, Gross and Histopathologic Postmor-
tem Assessment, Studies
Compounds 12 and 16 were evaluated in a dose escalation study where a
solution of the test compound in 10% DMSO in polyethyleneglycol 400 :
Hanks’ buffer (1:1, v/v) was administered ip once a day for seven consecutive
days to female BDF1 mice (17–21 g, n = 3) starting with a dose of
0.00013 mmol/kg that was increased in 2-fold increments to the highest dose
of 0.015 mmol/kg. Each group of mice administered a particular dose was
observed daily for 60 days starting from Day 1 (first dose) to determine any
acute or chronic toxicities, including restlessness, lethargy, inappropriate
aggression, abnormal gait, or weight loss. Each group of mice (n = 3)
receiving the highest doseof 12 or 16 (0.013 mmol/kg on 7-consecutive days)
were sacrificed at Day 60 by CO2 euthanasia and a cardiac bleed was
performed. Blood was collected into a serum tube, allowed to clot, centri-
fuged, and tested using a clinical chemistry panel [alanine aminotransferase
(iu/l); alkaline phosphatase (iu/l); total bilirubin (µmol/l); glucose (mmol/l);
total protein (g/l); albumin (g/l); globulin (g/l); blood urea nitrogen (mmol/l);
creatinine (µmol/l); lipase (iu/l); creatine kinase (iu/l); osmality (milliosmo-
les/kg), sodium (mmol/l); potassium (mmol/l); chloride (mmol/l); bicarbon-
ate (mmol/l); calcium (mmol/l) and phosphorous (mmol/l)].
A postmortem examination was performed on each mouse to detect gross
postmortem abnormalities and a set of tissues consisting of heart, lung, liver,
spleen, kidney, brain, gonad, intestine, and sternal bone marrow was obtained
and fixed in10% neutral buffered formalin. Following fixation, sectionswere
processed into paraffin blocks, sectioned at 5 microns and stained with
hematoxylin and eosin using standard methods. Histopathologic sections
were examined by a board certified veterinary pathologist to detect any
histopathologic organ abnormality.
5′-O-Bis(2,2,2-trichloroethyl)phosphoryl-3′-O-butanoyl-5-fluoro-2′-deoxy-
uridine 16
Butyric anhydride (96 µl, 0.587 mmol) was added to a solution of 15
(80 mg. 0.136 mmol) in dry pyridine (10 ml) with stirring and the reaction
was allowed to proceed at 25 °C for 12 h. Removal of the solvent in vacuo
gave a residue to which toluene (10 ml) was added and immediately evapo-
rated (two times). The residue was purified by silica gel column chromatog-
raphy using a gradient of 0–3% MeOH in CHCl3 as eluent. The fractions
which exhibited Rf 0.41 [micro TLC using MeOH-CHCl3 (1:19, v/v) as
development solvent] yielded 16 as white needles after recrystallization from
ethanol-water (79 mg, 88%): mp 118–119 °C; 1H NMR (CHCl3-d1): δ 0.98
(t, J = 7.4 Hz, 3H, CH2CH3), 1.68 (sextet, 2H, CH2CH2CH3), 2.10–2.13 (m,
1H, H-2′), 2.36 (t, J = 7.4 Hz, 2H, CH2CO), 2.50–2.60 (m, 2H, H-2′′),
4.22–4.28 (m, 1H, H-4′), 4.44–4.60 (m, 2H, H-5′), 4.61–4.78 (m, 4H,
OCH2CCl3), 5.30–5.36 (m, 1H, H-3′), 6.31–6.40 (m, 1H, H-1′), 7.73 (d, J5F,6
= 6.1 Hz, 1H, H-6), 8.9 (br s, 1H, NH); 13C NMR (CHCl3-d1): δ 13.6
(CH2CH3), 18.3 (CH2CH3), 35.9 (COCH2), 37.4 (C-2′), 68.3 (d, J5′,P = 6.5
Hz, C-5′), 73.8 (C-3′), 77.4 (CH2OP, overlaps with solvent CDCl3), 83.1 (d,
J
4′,P = 7.6 Hz, C-4′), 85.2 (C-1′), 94.4 (d, JCl3C,P = 10.9 Hz, CCl3), 123.3 (d,
J5F,6 = 34.9 Hz, C-6), 140.7 (d, J5,5F = 240 Hz, C-5), 148.6 (C-2 C=O), 156.2
(d, J4,5F = 27.2 Hz, C-4 C=O), 173.1 (CO2). Anal. (C17H20Cl6FN2O9P) C,
H, N.
In Vitro Cytotoxic Screen
The in vitro human tumor cell line screen was performed by the National
Cancer Institute[38]. Briefly, a total of about 60 human tumor cell lines
derived from nine cancer types (leukemia, non-small cell lung cancer, colon
cancer, CNS cancer, melanoma, ovarian cancer, renal cancer, prostate can-
cer, breast cancer) were used in this screen. A stock solution of each test
compound, prepared in dimethyl sulfoxide (DMSO), was tested at five
different concentrations (10-fold dilution) against each cell lineincell culture
medium. The cells were plated in 96-well plates at densities between 5,000
to 40,000 cells/100 µl in each well. After pre-incubation for 24 h, the test
compound was added, the cells were exposed to the test compound for 48 h,
the cells were fixed with trichloroacetic acid and then stained with sulforho-
damine B (SRB) which binds to the basic amino acids of the cellular
macromolecules. The unbound SRB was removed by washing with 1% acetic
acid, the bound dye was solubilized in Tris buffer and measured spectro-
photometrically using an automatic plate reader, and the GI50, TGI, and
Tumor Growth Delay Studies
EMT6 murine mammary tumor cells were cultivated as a monolayer in
RPMI 1640 medium supplemented with 10% of Fetal Bovine Serum
(GIBCO) at 37 °C in a 5% CO2 humidified atmosphere, and then 106 cells
were implanted subcutaneously into the flank of female mice (17–21 g).
Solutions of FDU and sodium butyrate were prepared by dissolution in
Hanks’ buffer, whereas solutions of RA, 12 and 16 were prepared by
dissolution in DMSO-H2O (1:10, v/v) followed by subsequent dilution with
50% v/v polyethylene glycol 400 in Hanks’ buffer to provide the desired test
compound concentration. Twenty-four h after tumor cell implantation, a
solution of the test compound, either alone or in combination, was injected
into mice (0.0075 mmol/kg and 0.015 mmol/kg ip dose, respectively) for
seven consecutive days. Tumor size was measured (L × W × H) using a caliper
once every other day starting from Day 7 after tumor cell implantation, until
the tumor volume reached 500 mm3. Tumor volume was calculated using the
volume formula for a hemiellipsoid: V = 1/2(4/3)(L/2)(W/2)(H)
LC50 values were calculated as reported[38]
.
In Vitro Determination of Drug Induced Human Leukemia HL60 Cell
Differentiation
References
HL60 human promyelocytic leukemia cells were cultivatedasasuspension
in RPMI 1640 Medium supplemented with 10% Fetal Bovine Serum
(GIBCO, Burlington, Canada). The cells were grown at 37 °C in a humidified
5% CO2 atmosphere. Cell density was determined using a Coulter counter
(Coulter Electronics, Ltd.), and cell viability was estimated by trypan blue
dye exclusion. Cells were harvested by centrifugation, and resuspended in
the growth medium at a density of 2 × 105 cells/ml prior to dispensing
1 ml/well onto 24-well culture plates. After a 24 h incubation, a freshly
prepared solution of the test compound in DMSO was added having various
concentrations in the 10–8 M to 10–5 M range for compounds 12, 16 and RA,
and in the 10–6 M to 10–3 M range for NaBu, in 10-fold concentration
increments. The final concentration of DMSO was less than 0.1%. Culture
plates were then incubated for 7 days at 37 °C in a humidified 5% CO2
atmosphere prior to morphological analysis using light microscopy. Cell
differentiation was assessed morphologically by counts on Wright-stained
cytospin preparations. Differential counts were always scored blind using at
least 200 cells per slide, except when the number of cells on the plate was
less than 200 when a high concentration of the test drug was used. Morpho-
logical changes induced by the prodrugs were compared to that induced by
[1] C. Heidelberger, Fluorinated Pyrimidines and Their Nucleosides. In
Handbook of Experimental Pharmacology, Vol. 38, Part 2, Springer-
Verlag, Berlin, 1975, 193–231.
[2] R. J. Langenbach, P. V. Danenberg, C. Heidelberger, Biochem. Bio-
phys. Res. Commun. 1972, 48, 1565–1571.
[3] D. V. Santi, C. S. McHenry, H. Sommer, Biochemistry 1974, 13,
471–481.
[4] N. K. Chaudhuri, K. L. Mukherjee, C. Heidelberger, Biochem. Phar-
macol. 1959, 1, 328–341.
[5] M. Uchida, N. Brown, D. H. W. Ho, Anticancer Res. 1990, 10, 779–783.
[6] A. R. Bapat, C. Zarow, P. V. Danenberg, J. Biol. Chem. 1983, 258,
4130–4136.
[7] H. M. Pinedo, G. F. J. Peters, J. Clin. Oncol. 1988, 6, 1653–1644.
[8] S. H. Berger, C.-H. Jenh, L. F. Johnson, F. G. Berger, Mol. Pharmacol.
1985, 28, 461–467.
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)

294
Knaus and co-workers
[9] F. P. LaCreta, J. Chromatogr. 1987, 414, 197–201.
[26] A. Leder, P. Leder, Cell 1975, 5, 319–322.
[10] K. M. Fries, C. Joswig, R. F. Borch, J. Med. Chem. 1995, 38, 2672–
[27] C. Augeron, C. L. Laboisse, Cancer Res. 1984, 44, 3961–3969.
2680.
[28] S. P. Langdon, M. M. Hawkes, F. G. Hay, S. S. Lawrie, D. J. Schol, J.
Hilgers, R. C. F. Leonard, J. F. Smyth, Cancer Res. 1988, 48, 6161–
6165.
[11] D. Farquahar, S. Khan, D. N. Scrivastva, P. P. Saunders, J. Med. Chem.
1994, 37, 3902–3909.
[12] J. Imai, P. F. Torrence, J. Org. Chem. 1981, 46, 4015–4021.
[29] L. Gamet, D. Daviaud, C. Denis-Pouxviel, C. Remesy, J. Murat, Int. J.
Cancer 1992, 52, 286–289.
[13] C. McGuigan, H. M. Sheeka, N. Mahmood, A. Hay, Bioorg. Med.
Chem. Lett. 1993, 3, 1203–1206.
[30] A. A. Miller, E. Kurschel, R. Osieka, C. G. Schmidt, Eur. J. Cancer
Clin. Oncol. 1987, 23, 1283–1287.
[14] C. McGuigan, D. Kinchington, S. R. Nickolls, C. Nickson, T. J.
O’Connor, Bioorg. Med. Chem. Lett. 1993, 3, 1207–1210.
[31] P. Pouillart, L. Cerutti, G. Ronco, P. Villa, C. Chany, Int. J. Cancer
1991, 49, 89–95.
[15] C. McGuigan, R. N. Pathirana, J. Balzarini, E. DeClercq, J. Med. Chem.
1993, 36, 1048–1052.
[32] P. Pouillart, G. Ronco, I. Cerutti, J. H. Trouvin, F. Pieri, P. Villa, J.
Pharm. Sci. 1992, 81, 241–244.
[16] Y. Nishizawa, J. E. Casida, S. W. Anderson, C. Heidelberger, Biochem.
Pharmacol. 1965, 14, 1605–1619.
[33] A. Rephaeli, E. Rabizadeh, A. Aviram, M. Shaklai, M. Ruse, A.
[17] T. Kawaguchi, S. Fukushima, Y. Hayashi, M. Nakano, Pharm. Res.
1988, 5, 741–744.
Nudelman, Int. J. Cancer 1991, 49, 66–72.
[34] A. Nudelman, M. Ruse, A. Aviram, E. Rabizadeh, M. Shaklai, Y.
[18] J. Yamashita, S. Takeda, H. Matsumoto, N. Unemi, M. Yasumoto, J.
Med. Chem. 1989, 32, 136–139.
Zimrah, A. Rephaeli, J. Med. Chem. 1992, 35, 687–694.
[35] Z.-X. Chen, T. R. Breitman, Cancer Res. 1994, 54, 3494–3499.
[19] (a) R. E. Scott, M. Edens, D. N. Estervig, M. Filipak, B. J. Hoerl, B. M.
Hsu, P. B. Maercklein, P. Minoo, C. Y. Tzen, M. R. Wilke, M. A.
Zschunke, Cellular Differentiation and the Prevention and Treatment
of Cancer. In The Status of Differentiation Therapy of Cancer,
Waxman, S.; Rossi, G. B.; Takaku, K. (Eds.). Raven Press, New York,
1988, 3–16. (b) L. Sachs, Cancer Res. 1987, 47, 1981–1986. (c) M.
Reiss, C. Gamba-Vitalo, A. C. Sartorelli, Cancer Treatment Reports,
1986, 70, 201–218.
[36] J. R. Mercer, E. E. Knaus, L. I. Wiebe, J. Med. Chem. 1987, 30,
670–675.
[37] S. K. Aggarwal, S. R. Gogu, S. R. S. Rangan, K. C. Agrawal, J. Med.
Chem. 1990, 33, 1505–1510.
[38] A. Monks, D. Scudiero, P. Skehan, R. Shoemaker, K. Paull, D. Vistica,
C. Hose, J. Langley, P. Cronise, A. Vaigro-Wolff, M. Gray-Goodrich,
H. Campbell, J. Mayo, M. R. Boyd, J. Natl. Cancer Inst. 1991, 83,
757–766.
[20] M. A. Smith, D. R. Parkinson, B. D. Cheson, M. A. Friedman, J. Clin.
Oncol. 1992, 10, 839–864.
[39] T. R. Breitman, R. He, Cancer Res. 1990, 50, 6268–6273.
[21] R. P. Warrell, Seminars in Hematology 1994, 31, 1–48.
[22] M. A. Khan, G. R. Jenkins, W. H. Tolleson, K. E. Creek, L. Pirisi,
[40] C. Meier, M. Lorey, E. De Clercq, J. Balzarini, J. Med. Chem. 1998,
41, 1417–1427.
Cancer Res. 1993, 53, 905–909.
[23] P. C. Adamson, J. F. Boylan, F. M. Balis, R. F. Murphy, K. A. Godwin,
[41] S. Manfredini, D. Simoni, R. Ferroni, R. Bazzanini, S. Vertuani, S.
L. J. Gudas, D. G. Poplack, Cancer Res. 1993, 53, 472–476.
Hatse, J. Balzarini, E. De Clercq, J. Med. Chem. 1997, 40, 3851–3857.
[24] L. Degos, H. Dombret, C. Chomienne, M. T. Daniel, J. M. Miclea, C.
[42] D. V. Waterhous, D. D. Muccio, Mag. Res. Chem. 1990, 28, 223–226.
[43] R.W. Curley, J. W. Fowble, Mag. Res. Chem. 1989, 27, 707–709.
Received: April 30, 1999 [FP391]
Chastang, P. Fenaux, Blood, 1995, 85, 2643–2653.
[25] J. Leavitt, J. C. Barrett, D. B. Crawford, P. O. P. Ts’o, Nature 1978,
271, 262–265.
© WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1999 – Printed in the Federal Republic of Germany
Alle Rechte, insbesondere die der Übersetzung in fremde Sprachen, vorbehalten. Kein Teil dieser Zeitschrift darf ohne schriftliche Genehmigung des Verlages in irgendeiner Form
– durch Fotokopie, Mikrofilm oder irgendein anderes Verfahren – reproduziert oder in eine von Maschinen, insbesondere von Datenverarbeitungsmaschinen verwendbare Sprache
übertragen oder übersetzt werden.– All rights reserved (including those of translation into foreign languages). No part of this issue may be reproduced in any form – photoprint,
microfilm, or any other means – nor transmitted or translated into a machine language without the permission in writing of the publishers.– Von einzelnen Beiträgen oder Teilen
von ihnen dürfen nur einzelne Vervielfältigungsstücke für den persönlichen und sonstigen eigenen Gebrauch hergestellt werden. Die Weitergabe von Vervielfältigungen, gleichgültig
zu welchem Zweck sie hergestellt werden, ist eine Urheberrechtsverletzung. Der Inhalt dieses Heftes wurde sorgfältig erarbeitet. Dennoch übernehmen Autoren, Herausgeber,
Redaktion und Verlag für die Richtigkeit von Angaben, Hinweisen und Ratschlägen sowie für eventuelle Druckfehler keine Haftung. This journal was carefully produced in all its
parts. Nevertheless, authors, editors and publishers do not warrant the information contained therein to be free of errors. Readers are advised to keep in mind that statements, data,
illustrations, procedural details or other items may inadvertently be inaccurate.– Die Wiedergabe von Gerbrauchsnamen, Handelsnamen,Warenbezeichnungen u. dgl. in dieser
Zeitschrift berechtigt nicht zu der Annahme, daß solcheNamen ohne weiteres vonjedermann benutzt werden dürfen. Es handelt sich häufig um gesetzlich eingetragene Warenzeichen,
auch wenn sie in dieser Zeitschrift nicht als solche gekennzeichnet sind. Druck und Buchbinder: Druck Partner Rübelmann GmbH, D-69502 Hemsbach.– Unverlangt zur Rezension
eingehende Bücher werden nicht zurückgesandt.
For USA and Canada: Archiv der Pharmazie (ISSN 0365-6233) is published monthly.– Printed in the Federal Republic of Germany.
US Postmaster: Periodicals postage paid at Jamaica, NY 11431. Air freight and mailing in the USA by Publications Expediting Services Inc., 200 Meacham Ave., Elmont NY 11003.
Send address changes to: “Archiv der Pharmazie”, c/o Publications Expediting Services Inc., 200 Meacham Ave., Elmont NY 11003
Valid for users in the USA: The appearance of the code at the bottom of the first page of an article in this journal (serial) indicates the copyright owner’s consent that copies of the
article may be made for personal or internal use, or for the personal or internal use of specific clients. This consent is given on the condition, however, that copier pay the stated per
copy fee throught the Copyright Clearance Center, Inc., for copying beyond that permitted by Sections 107 for 108 of the U.S. Copyright Law. This consent does not extend to
other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective work, or for resale. For copying from back
volumes of this journal see »Permissions to Photo-Copy: Publisher’s Fee List« of the CCC.
Printed on chlorine- and acid-free paper/Gedruckt auf säurefreiem und chlorfrei gebleichtem Papier
Arch. Pharm. Pharm. Med. Chem. 332, 286–294 (1999)