Synthesis and evaluation of 6-methylene-bridged uracil derivatives. Part 2: Optimization of inhibitors of human thymidine phosphorylase and their selectivity with uridine phosphorylase

Article abstract of DOI:10.1016/j.bmc.2004.04.046

A series of novel 6-methylene-bridged uracil derivatives have been optimized for clinical use as the inhibitors of human thymidine phosphorylase (TP). We describe their synthesis and evaluation. Introduction of a guanidino or an amidino group enhanced the in vitro inhibitory activity of TP comparing with formerly reported inhibitor 1. Their selectivity for TP based on uridine phosphorylase inhibitory activity was also evaluated. Compound 2 (TPI) has been selected for clinical evaluation based on its strong TP inhibition and excellent modulation of 2′-deoxy-5-(trifluoromethyl)uridine (F₃dThd) pharmacokinetics. As a result, TAS-102 (a combination of F₃dThd and TPI) is currently in phase 1 clinical studies.

Full text of DOI:10.1016/j.bmc.2004.04.046

Bioorganic & Medicinal Chemistry 12 (2004) 3443–3450
Synthesis and evaluation of 6-methylene-bridged uracil
derivatives. Part 2: Optimization of inhibitors of human
thymidine phosphorylase and their selectivity
with uridine phosphorylase
Shingo Yano,* Hideki Kazuno, Tsutomu Sato, Norihiko Suzuki, Tomohiro Emura,
Konstanty Wierzba, Jun-ichi Yamashita, Yukio Tada, Yuji Yamada,
Masakazu Fukushima and Tetsuji Asao
Hanno Research Center, Taiho Pharmaceutical Co., Ltd, 1-27 Misugidai, Hanno-shi, Saitama 357-8527, Japan
Received 5 March 2004; revised 28 April 2004; accepted 29 April 2004
Abstract—A series of novel 6-methylene-bridged uracil derivatives have been optimized for clinical use as the inhibitors of human
thymidine phosphorylase (TP). We describe their synthesis and evaluation. Introduction of a guanidino or an amidino group
enhanced the in vitro inhibitory activity of TP comparing with formerly reported inhibitor 1. Their selectivity for TP based on
uridine phosphorylase inhibitory activity was also evaluated. Compound 2 (TPI) has been selected for clinical evaluation based on
its strong TP inhibition and excellent modulation of 20-deoxy-5-(trifluoromethyl)uridine (F3dThd) pharmacokinetics. As a result,
TAS-102 (a combination of F3dThd and TPI) is currently in phase 1 clinical studies.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
TP, is a principal F3dThd cleaving enzyme in human.8;9
Therefore, to improve the anti-tumor effects of F3dThd
in human, there is necessity to design an inhibitor
directed toward TP rather than an inhibitor of UP.
20-Deoxy-5-(trifluoromethyl)uridine (F3dThd) is known
to have strong in vitro and in vivo anti-tumor activi-
ties.1;2 In addition, F3dThd was expected to be effective
against cancer cells resistant to 5-fluorouracil and/or 20-
deoxy-5-fluorouridine, characterized by amplification of
thymidylate synthase and/or deletion of orotate phos-
phoribosyltransferase.3 However, F3dThd is known to
be quickly cleaved into an inactive compound [5-(tri-
fluoromethyl)uracil] both in vitro and in vivo by
pyrimidine nucleoside phosphorylases.4–6 Therefore, it
did not provide satisfactory clinical anti-tumor effects.7
The cleavage of the glycosyl linkage of F3dThd is
mediated by one of two types of pyrimidine nucleoside
phosphorylases, uridine phosphorylase (UP) or thym-
idine phosphorylase (TP). The former, UP, mainly
occurs in rodents such as mice and rats, and the latter,
In our preceding report,10 human TP was revealed to
possess the tight limitation at the active site to fit
6-methylene-bridged uracil derivatives. Compound 1
was selected for further in vivo evaluation because of the
superior pharmacokinetic (PK) profile and the inhibi-
tory activity of human TP (Fig. 1).
In this paper, we describe the synthesis and biological
evaluation of a series of novel 6-methylene-bridged
O
Br
HN
N
O
N
H
Keywords: Thymidine phosphorylase inhibitor; TPI; 20-Deoxy-5-(tri-
fluoromethyl)uridine (F3dThd); TAS-102.
1
* Corresponding author. Tel.: +81-42-972-8900; fax: +81-42-972-0034;
e-mail: syano@taiho.co.jp
Figure 1.
0968-0896/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2004.04.046

3444
S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
uracil derivatives possessing a guanidino or an amidino
group. Several derivatives have been synthesized and
showed the enhanced in vitro TP inhibitory activity. The
selectivity of inhibitors to TP and UP is also discussed.
O
O
Cl
Cl
HN
HN
a
H
N
NH2
NHR
HCl
O
N
H
O
N
H
NR
3
4: R = H, H
5: R = CH3, H
6: R = CH2CH2
c
b
2. Chemistry
O
Cl
H
O
Scheme 1 illustrates the synthesis of 6-methylene-
bridged uracil derivatives possessing a guanidino, an
amidino, or a thioureido group. Compounds 3, 9, and
11 were prepared as previously described.10 Compounds
4–6 and 10 were prepared by treating 3 or 9 with
appropriate methylisothioureas in water.11 Compound 7
was prepared by treating 3 with 1-ethoxyethanimine
hydrochloride in DMF. Thiourea analogue 8 was pre-
pared by heating 3 with methyl isothiocyanate in DMF.
Isothiourea analogues 12–14 were prepared by heating
11 with appropriate thioureas in EtOH.
Cl
HN
HN
H
N
N
NHCH3
CH3
HCl
O
N
O
N
H
H
S
NH
8
7
O
O
Cl
Cl
Cl
a
HN
HN
CH3
N
NHCH3
NH2
O
N
H
O
N
H
NH
9
10
Scheme 2 illustrates the synthesis of 6-methylene-
bridged uracil derivatives possessing cyclic guanidines.
The synthesis of compounds 15 and 16 was previously
reported.10 Compound 17 was prepared by treating 15
with a large excess of ethylenediamine in water. Treat-
ment of diamines 16 and 17 with cyanogen bromide12 in
water resulted in the formation of 2-iminoimidazolidine
derivatives 18 and 19, respectively. Cyclization of N-
alkylethylendiamines 20–22 with cyanogen bromide
in benzene,12 followed by alkylation of the resulting
1-alkyl-2-iminoimidazolidines 23–25 with 11 using NaOEt
in DMF,13 gave the desired 3-alkyl-2-iminoimidazol-
idine derivatives 26–28. 1-Alkyl-2-aminoimidazoles 31
and 32, prepared as free bases from 29 and 30 by the
Lawson’s method,14 were alkylated with 11 in water to
result in the formation of the desired 3-alkyl-2-imino-
imidazoline derivatives 33 and 34.13
O
O
Cl
d
HN
HN
Cl
S
NHR
HCl
O
N
H
11
O
N
H
NR
12: R = H, H
13: R = CH3, H
14: R = CH2CH2
Scheme 1. Reagents and conditions: (a) [CH3SC(@NH)NH2]2ÆH2SO4,
CH3SC(@NH)NHCH3ÆHI, or 2-methylthio-2-imidazoline hydro-
iodide, KOH, H2O; (b) CH3C(@NH)OCH2CH3ÆHCl, DMF; (c)
CH3NCS, DMF; (d) NH2C(@S)NH2;CH3NHC(@S)NH2, or 2-imid-
azolidinethione, EtOH.
Scheme 3 illustrates the synthesis of the compounds
possessing a 2-iminopyrrolidyl group. 2-Iminopyrrol-
O
O
X
O
X
X
HN
NH2
HN
HN
a
b
NH
HBr
NH
N
Cl
O
N
H
O
N
H
O
N
H
NH
11: X = Cl
16: X = Cl
18: X = Cl
15: X = Br
17: X = Br
19: X = Br
O
Cl
HN
c
d
NHR
HN
NR
HBr
H2N
NR
NH
N
O
N
H
11
NH
20: R = CH3
23: R = CH3
26: R = CH3
21: R = CH2CH3
22: R = CH(CH3)2
24: R = CH2CH3
25: R = CH(CH3)2
27: R = CH2CH3
28: R = CH(CH3)2
O
Cl
g
e, f
OR
HN
RN
N
NR
HCl
RHN
N
11
OR
O
N
H
NH2
NH
29: R = CH3
31: R = CH3
32: R = CH2CH3
33: R = CH3
34: R = CH2CH3
30: R = CH2CH3
Scheme 2. Reagents and conditions: (a) ethylenediamine, H2O; (b) BrCN, H2O; (c) BrCN, benzene; (d) NaOEt, DMF; (e) cyanamide, aq AcOH;
(f) concd HCl; (g) H2O.

S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
3445
1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in MeOH led
to the improvement of chemical yield (68% for com-
pound 2) as well as the ease of handling and simplicity of
the workup procedure because of no formation of
inorganic byproducts.
a
HN
Cl
CN
HCl
X
NH
35
36
O
O
X
b or c
HN
HN
N
3. Results and discussion
Cl
36
O
N
H
O
N
H
HCl
NH
In our previous report,10 compound 1 has been selected
for further in vivo evaluation. Careful evaluation of
compound 1 revealed a high death rate (4/6) and body
weight loss ()13.9%, n ¼ 2) in ICR mice (male, 5 weeks
old) given high doses (2000 mg/kg, po.) for 14 days.
Body weight loss ()6.9%, n ¼ 6) was also observed in
ICR mice (male, 5 weeks old) at lower doses (1000 mg/
kg, po.). These primarily disappointing results prompted
us to search for a new type of chemical class of com-
pounds. Hence, we focused our study on the design of
new inhibitors with strong TP inhibitory activity and
possessing negligible toxicity.
37: X = CH3
15: X = Br
11: X = Cl
38: X = CH3
39: X = Br
2: X = Cl
Scheme 3. Reagents and conditions: (a) NH3, MeOH; (b) NaOEt,
DMF; (c) DBU, MeOH.
idine hydrochloride (36) was prepared from 35 in a
previously described manner.15 6-(Chloromethyl)thy-
mine (37) was prepared according to the published
method.16 Alkylation of 36 with 37, 15, and 11 in the
presence of NaOEt in DMF resulted in the desired
2-iminopyrrolidine derivatives 38, 39, and 2, with
chemical yields of 22%, 13%, and 38%, respectively. The
reaction conditions were not satisfactory for a large-
scale preparation. The application of an alkylation with
In Table 1, we show the inhibitory effect of 6-methylene-
bridged uracil derivatives on human TP. We found that
the potency of guanidine derivative 4 was close to that of
Table 1. Inhibitory effect of 6-methylene-bridged uracil derivatives on human TP and rat UP
O
X
HN
R
O
N
H
a;b
a;c
Compd
R
X
TP IC50 (lM)
UP IC50 (lM)
1
N-Pyrrolidinyl
NHC(@NH)NH2
NHC(@NH)NHCH3
2-Imidazolin-2-ylamino
NHC(@NH)CH3
NHC(@S)NHCH3
NCH3C(@NH)NH2
SC(@NH)NH2
SC(@NH)NHCH3
2-Imidazolin-2-ylthio
N-(2-Iminoimidazolidinyl)
N-(2-Iminoimidazolidinyl)
N-(2-Imino-3-methylimidazolidinyl)
N-(3-Ethyl-2-iminoimidazolidinyl)
N-(2-Imino-3-isopropylimidazolidinyl)
N-(2-Imino-3-methylimidazolinyl)
N-(3-Ethyl-2-iminoimidazolinyl)
N-(2-Iminopyrrolidinyl)
N-(2-Iminopyrrolidinyl)
N-(2-Iminopyrrolidinyl)
NHCH3
Br
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Cl
Br
Cl
Cl
Cl
Cl
Cl
Cl
CH3
Br
Cl
Cl
Cl
Cl
0.51
0.27
0.087
31
14
390
4
5
230
6
NTd
>100
NTd
630
7
1.5
8
>100
0.031
0.35
0.15
25
10
12
13
14
18
19
26
27
28
33
34
2 (TPI)
38
39
9
610
NTd
NTd
>100
>100
>100
>100
>100
NTd
NTd
>100
NTd
NTd
7.8
0.013
0.030
0.046
0.36
4.0
0.24
6.9
0.035e
0.12
0.032
12
40
41
42
NHCH2CH3
N-Pyrrolidinyl
N-Imidazolyl
20
2.2
50
8.6
1.0
93
a Inhibitory effect is reported as an IC50 value, indicating the concentration of test compound required to inhibit enzyme activity by 50%. This assay
tests at least three concentrations of each compound.
b A 0.6 mM concentration of [6-3H]dThd is used as the substrate for the TP reactions.
c A 0.6 mM concentration of [5-3H]Urd is used as the substrate for the UP reactions.
d Not tested.
e The Ki value for TPI determined by using recombinant human TP was 0.017 lM.17

3446
S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
Table 2. Oral absorption of compounds 1, 2, 4, 5, 12, 18, 19, 26, and
27 in micea
the formerly reported candidate 1. Furthermore, com-
pound 5 as a N-methyl derivative of 4 showed more
potent activity. The substitution of the end of guanidine
with imidazoline function (Compound 6) significantly
abolished the inhibitory effects. This may indicate that a
nonsubstituted imino group was probably necessary to
enhance the inhibitory effect. After introduction of the
smallest amidino group, methylamidino group, the po-
tency was not restored (compound 7) up to the level of
compound 4. With replacement of imine by thione, the
resulting thiourea derivative (compound 8) dramatically
diminished the inhibitory potency. Notably compound
10 showed close inhibitory activity to its isomer 5. Alkyl
substitution at the amino group, containing the non-
substituted imino-group in guanidine, could be impor-
tant in enhancing the inhibitory activity. Evaluation of
corresponding isothiourea derivatives (12–14) of guan-
idine derivatives (4–6) showed close inhibitory activity,
however the isothiourea derivatives appeared to be less
stable in acidic or basic conditions. The above results
indicated that compounds containing the guanidine
group seemed to play essential roles in the development
of clinically important chemical entities.
b
c
d
Compd
Cmax (lM)
Tmax (h)
AUC0-8 (lM h)
1
2 (TPI)
80
11
0.25
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
130
17
4
3.1
5.0
7.4
12
7.1
5.1
12
5
12
18
19
26
27
26
6.9
3.1
1.8
5.8
3.4
1.3
a Compounds 1, 2, 4, 5, 12, or 18 (0.169 mmol/kg) with F3dThd
(0.169 mmol/kg) dissolved in 0.5% hydroxypropylmethylcellulose
(HPMC) were orally administered to ICR mice (male, 6 weeks old,
n ¼ 3). Compounds 19, 26, or 27 (0.169 mmol/kg) with F3dThd
(0.169 mmol/kg) dissolved in 0.5% HPMC was orally administered to
BALB/c nu/nu nude mice (male, 6 weeks old, n ¼ 3).
b Maximum plasma concentration after oral dosing.
c Time to Cmax
.
d Area under the concentration–time curve for 0–8 h after oral dosing.
further in vivo biological evaluation. Therefore, the PK
studies of F3dThd given with compounds (2, 4, 5, 12, 18,
19, 26, and 27) were performed in mice after oral
administration. The maximum plasma concentration
(Cmax) and an AUC value of each compound are shown
in Table 2. To improve the anti-tumor effect of F3dThd,
the clinical candidate should have a similar ratio of the
Cmax and the IC50 values like compound 1. Considering
the IC50 value of compound 1 against human TP, the
Cmax and AUC0–8 values of compounds 2, 18, and 19
should have been similar to those of compound 1 to
exert similar pharmacological effect. Linear guanidine or
amidine derivatives (4, 5, and 12) showed moderate
values of Cmax and AUC, but they did not satisfy our
criteria. In contrast, the Cmax and AUC0–8 values of
cyclic guanidine derivative 18 were higher than those of
linear compounds. Considering its IC50 value, one may
expect this compound to be worth of further pharma-
cological evaluation. Bromo and alkyl derivatives of 18
(compounds 19, 26, and 27) were also evaluated, how-
ever, their values were not superior to those of com-
pound 18. The synthesis of compound 18 required a
highly toxic reagent, cyanogen bromide. In order not to
violate the safety of workers, the development of this
compound should not be enforced. Compound 2 (TPI)
was finally evaluated, since its values of Cmax and AUC
met our criteria of a potent pharmacological agent. In
addition, its synthesis was characterized by a good yield
and safer synthetic procedure than that of 18. To con-
firm its in vivo potency, TPI (9.42 mg/kg) was adminis-
tered orally with an equimolar amount of F3dThd
(10 mg/kg) to monkeys (n ¼ 3) to evaluate its ability to
elevate the plasma concentration of F3dThd. Plasma
F3dThd levels were increased very markedly, with a Tmax
of 1 h, a Cmax of 15.2 lg/mL, and an AUC value of
28.5 lg h/mL, compared with F3dThd administration
alone, with a Tmax of 1 h, a Cmax of 0.23 lg/mL, and an
AUC value of 0.28 lg h/mL.17 In addition, the combined
oral administration of TPI significantly potentiated the
anti-tumor activity of F3dThd against AZ-521 human
stomach cancer xenografts in nude mice.17
In our preceding report,10 we have found that pyrrol-
idine and imidazole best matched the active site of
human TP. In addition, the best biological results were
observed, when bromide was substituted at the 5-posi-
tion of the uracil molecule. Among the iminoimidazo-
lidinyl derivatives (18, 19, and 26–28), compound 18
showed the best inhibitory effect in vitro. Unexpectedly
the bromide substituted derivative 19 did not show more
potent activity. It was suggested that the conformation
of the inhibitors would be changed the fitting to the
region that surround the active site of human TP.
N-alkyl derivatives of 18 were evaluated because they
were expected to improve the inhibitory activity based
on the effects of compounds 4 and 5. Methyl derivative
26 showed diminished potency. Furthermore, ethyl and
isopropyl derivatives (27 and 28) showed much dimin-
ished inhibitory activity, probably caused by too large
molecular size. The iminoimidazolinyl derivatives (33
and 34) appeared to be almost 10-fold weaker than the
corresponding imidazolidinyl derivatives (compounds
26 and 27). N-Alkyl substitution appeared to be not
effective in enhancing the inhibitory activity of TP,
because of tight limitation of the active site of human TP
as we described previously.10 Furthermore, introduction
of a double bond in the 5-membered heterocycle lead to
the reduction of the inhibitory activity, therefore the
compounds without a double bond appeared to be
important to maximize the activity. Based on these
results, the 2-iminopyrrolidinyl derivatives were evalu-
ated. Although compound 7 showed only moderate
activity, compound 2 possessed strong inhibitory activ-
ity, indicating the 2-iminopyrrolidinyl group matching
well at the active site of human TP. Methyl derivative 38
showed weaker activity than 2. The potency of bromo
derivative 39, which has a similar structure to that of the
former clinical candidate 1, was close to that of 2.
Since compound 1 had toxic effects, the compounds with
more diverse chemical structured were preferred for

S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
3447
Despite the presence of strong TP inhibitory effects and
good PK profile, one more factor seemed to be impor-
tant for the selection of therapeutically important
compound. This additional factor was selectivity, in
other words, the compound should have a weak inhib-
itory effect to UP.
tion of F3dThd and TPI) is currently in the phase 1
clinical study for colorectal and breast cancer patients in
the USA.
5. Experimental
5.1. Chemistry
Table 1 shows the inhibitory effect of typical compounds
on rat UP and comparison with those on human TP.
Compound 1 had moderate activity against UP. How-
ever, it was about 30-fold weaker than TP. In contrast to
the case of TP, introduction of an imino group did not
enhance the inhibitory effect of UP (compounds 4, 5, 7,
10, 12, 18, 19, 26, 27, 28, and 2). All the compounds
showed diminished inhibitory activity when compared
with 1. It appeared that an amidino or a guanidino
group at the 6-position did not match at the active site
of UP. In contrast, compounds with no imino group
(9,41)10 had close inhibitory effect on UP when com-
pared with compound 1. Compounds 40 and 4210
showed weaker activity than 1. Three types of com-
pounds were found among the 6-methylene-bridged
uracil derivatives. Typically, compounds 9, 40, and 41
were dual inhibitors of TP and UP, compounds 1 and 42
were selective inhibitors of TP, and compound 2 was a
specific inhibitor of TP. Compound 1 showed a high
enough Cmax value in mice to exercise the inhibition of
UP in vivo. In contrast, specific TP inhibitor 2 should
not affect the function of UP in vivo. To avoid unex-
pected adverse effects, it was clear that selection of the
specific TP inhibitor was a better choice. Therefore, TPI
has been chosen as a candidate being able to provide an
excellent pharmacological profile when clinically used.
All melting points were determined on a Yanagimoto
micromelting point apparatus and are uncorrected. The
proton nuclear magnetic resonance (1H NMR) spectra
were recorded on a JEOL JNM-EX270 (270 MHz)
spectrometer. Chemical shifts are given in d values
(ppm) using tetramethylsilane as the internal standard.
All J values are given in hertz. Mass spectra were
recorded on a JEOL JMS-SX102A spectrometer. Ele-
mental analyses were carried out with a Yanagimoto C
H N Corder MT-5. Reagents and solvents were
obtained from commercial sources and used without
further purification. Flash column chromatography was
performed with Merck silica gel 60 (art. 9385; 230–400
mesh), and reaction progress was monitored by thin-
layer chromatography (TLC) analysis on silica gel 60
F254 plates (Merck). Visualization was done with UV
light (254 nm) or iodine. Yields are based on purified
material and were not optimized.
5.1.1. Amino{[5-chloro-2,6-dioxo(1,3-dihydropyrimidin-4-
yl)]methyl}carboxamidine hydrochloride (4). A mixture
of 600 mg (3.42 mmol) of 3 and 455 mg (3.27 mmol) of
methylisothiourea sulfate in 33 mL of 0.1 M aqueous
KOH was heated at 75 °C for 2 h. The reaction mixture
was allowed to stand at room temperature. An insoluble
material was collected by filtration and suspended in
5 mL of 2 M aqueous HCl at room temperature for 2 h.
The precipitate was collected by filtration to give 287 mg
(33%) of 4 as a brown powder: mp 255 °C dec. 1H NMR
(DMSO-d6): d 4.29 (d, 2H, J ¼ 5:0 Hz), 7.45 (s, 3H),
7.99 (t, 1H, J ¼ 5:0 Hz), 11.45 (s, 1H), 11.64 (s, 1H).
FAB-MS m=z: 218 (M+H)HCl)þ. Anal. Calcd for
C6H8ClN5O2ÆHClÆ0.1H2O: C, 28.16; H, 3.62; N, 27.37.
Found: C, 28.37; H, 3.67; N, 27.13.
4. Conclusion
In conclusion, we evaluated novel 6-methylene-bridged
uracil derivatives with an amidino or a guanidino group
as potent human TP inhibitors. The introduction of an
imino group has created more potent human TP inhib-
itors than former pre-clinical candidate 1. The specific
inhibitors of TP were prepared as a result of a mismatch
with the active site of UP.
Compound 2 (TPI) enhanced plasma concentration of
F3dThd in monkeys and potentiated the in vivo anti-
tumor activity of F3dThd in mice after combined oral
administration. Furthermore, neither significant body
weight loss nor animal death was observed in ICR mice
(male, 5 weeks old, n ¼ 6) even after administration of
TPI at high dose (2000 mg/kg, po.). Based on these
biological results, TPI has been selected as a candidate
for clinical trials. Furthermore, human TP has been
found to be identical to Platelet Derived Endothelial
Cell Growth Factor (PD-ECGF), known as an endo-
genous angiogenic factor.18 The results of biological
studies on TPI indicated that TPI may also inhibit the
process of angiogenesis, which is closely associated with
malignancy of solid tumors.19;20 Since the process of
angiogenesis is closely related to metastatic ability, TPI
may have a significant impact on the treatment of
patients with advanced cancers. TAS-102 (a combina-
5.1.2. 5-Chloro-6-({[imino(methylamino)methyl]amino}-
methyl)uracil hydrochloride (5). A mixture of 500 mg
(2.85 mmol) of 3 and 1.32 g (5.70 mmol) of 1,2-dimethyl-
2-thiopseudourea hydroiodide in 19 mL of 0.3 M aque-
ous KOH was heated at 40 °C for 3 days. An insoluble
material was collected by filtration and dissolved in
20 mL of 0.2 M aqueous HCl at room temperature. The
reaction mixture was filtered to remove insoluble
material and the filtrate was concentrated under a vac-
uum. The residue was recrystallized from water and
MeOH to give 320 mg (42%) of 5 as yellow crystals: mp
172 °C dec. 1H NMR (DMSO-d6): d 2.75 (d, 3H,
J ¼ 4:3 Hz), 4.30 (d, 2H, J ¼ 5:6 Hz), 7.60 (s, 2H), 7.74–
7.82 (m, 2H), 11.35 (s, 1H), 11.67 (s, 1H). FAB-MS m=z:
232 (M+H)þ. Anal. Calcd for C7H10ClN5O2ÆHClÆH2O:
C, 29.65; H, 4.23; N, 24.17. Found: C, 29.39; H, 4.58; N,
24.48.

3448
S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
5.1.3. 5-Chloro-6-[(2-imidazolin-2-ylamino)methyl]uracil
hydrochloride (6). Compound 6 was prepared in 25%
yield from 3 and 2-methylthio-2-imidazoline hydro-
iodide by a method similar to that described for 5, as
brown crystals: mp 235 °C dec. 1H NMR (DMSO-d6): d
3.62 (s, 4H), 4.38 (d, 2H, J ¼ 5:6 Hz), 8.45 (s, 1H), 8.69
(t, 1H, J ¼ 5:6 Hz), 11.44 (s, 1H), 11.67 (s, 1H). FAB-
MS m=z: 244 (M+H)HCl)þ. Anal. Calcd for
C8H10ClN5O2ÆHClÆ0.5H2O: C, 33.23; H, 4.18; N, 24.22.
Found: C, 33.26; H, 4.09; N, 24.13.
4.35 (s, 2H), 9.46 (s, 4H), 11.57 (s, 1H), 11.70 (1H, s).
FAB-MS m=z: 235 (M+H)HCl)þ. Anal. Calcd for
C6H7ClN4O2SÆHCl: C, 26.58; H, 2.97; N, 20.66. Found:
C, 26.93; H, 3.05; N, 20.31.
5.1.8. 5-Chloro-6-({[(methylamino)thioxomethyl]amino}-
methyl)uracil hydrochloride (13). Compound 13 was
prepared in 93% yield from 11 and N-methylthiourea by
a method similar to that described for 12, as a white
powder: mp 205 °C dec. 1H NMR (DMSO-d6): d 2.90 (s,
3H), 4.32 (s, 2H), 9.40 (s, 1H), 9.77 (s, 1H) 10.20 (s, 1H),
11.51 (s, 1H), 11.67 (s, 1H). FAB-MS m=z: 249
(M+H)HCl)þ. Anal. Calcd for C7H9ClN4O2SÆHCl: C,
29.49; H, 3.53; N, 19.65. Found: C, 29.61; H, 3.60; N,
19.70.
5.1.4. 5-Chloro-6-{[(iminoethyl)amino]methyl}uracil hydro-
chloride (7). A mixture of 500 mg (2.85 mmol) of 3 and
705 mg (5.70 mmol) of ethyl acetimidate hydrochloride
in 12 mL of DMF was stirred at room temperature for
14 h. The precipitate was collected by filtration and
washed with 10% HCl in MeOH to give 190 mg (26%) of
7 as a white powder: mp 220 °C dec. 1H NMR (DMSO-
d6): d 2.19 (s, 3H), 4.40 (d, 2H, J ¼ 5:0 Hz), 9.13 (s, 1H),
9.63 (s, 1H), 9.86 (t, 1H, J ¼ 5:0 Hz), 11.53 (s, 1H),
11.73 (s, 1H). FAB-MS m=z: 217 (M+H)HCl)þ. Anal.
Calcd for C7H9ClN4O2ÆHClÆ0.6H2O: C, 31.86; H, 4.28;
N, 21.23. Found: C, 31.68; H, 4.13; N, 21.49.
5.1.9. 5-Chloro-6-(2-imidazolin-2-ylthiomethyl)uracil hy-
drochloride (14). Compound 14 was prepared in 80%
yield from 11 and 2-imidazolidinetaahione by a method
similar to that described for 12, as a white powder: mp
220 °C dec. 1H NMR (DMSO-d6): d 3.34 (s, 4H), 4.41 (s,
2H), 10.60 (s, 1H), 11.70 (s, 1H). FAB-MS m=z: 261
(M+H)HCl)þ. Anal. Calcd for C8H9ClN4O2SÆHCl: C,
32.34; H, 3.39; N, 18.85. Found: C, 32.46; H, 3.40; N,
18.98.
5.1.5. 5-Chloro-6-({[(methylamino)thioxomethyl]amino}-
methyl)uracil (8). A mixture of 500 mg (2.85 mmol) of
3 and 219 mg (3.00 mmol) of methyl isothiocyanate in
DMF was heated at 70 °C for 2 h. The reaction mixture
was allowed to stand at room temperature and diluted
with 50 mL of water. The precipitate was collected by
filtration and washed with water and MeOH to give
5.1.10. 6-{[2-(Aminoethyl)amino]methyl}-5-bromouracil
(17). A mixture of 45.0 g (0.188 mol) of 15 and 56.4 g
(0.940 mol) of ethylenediamine in 188 mL of water was
stirred at room temperature for 2 h. The precipitate was
collected by filtration and washed with water and
MeOH to give 22.9 g (46%) of 17 as a white powder: mp
1
435 mg (61%) of 8 as a red powder: mp 195 °C dec. H
NMR (DMSO-d6): d 2.82 (s, 3H), 4.52 (s, 2H), 7.86 (s,
2H), 10.90 (s, 1H), 11.60 (s, 1H). FAB-MS m=z: 247
(M)H)ꢀ. Anal. Calcd for C7H9ClN4O2SÆ0.4H2O: C,
32.86; H, 3.86; N, 21.89. Found: C, 32.81; H, 3.58; N,
21.83.
1
168 °C dec. H NMR (DMSO-d6): d 2.63–2.75 (m, 4H),
3.58 (s, 2H). FAB-MS m=z: 263, 265 (M+H)þ. Anal.
Calcd for C7H11BrN4O2: C, 31.96; H, 4.21; N, 21.30.
Found: C, 31.73; H, 4.31; N, 21.32.
5.1.6. Amino{[5-chloro-2,6-dioxo(1,3-dihydropyrimidin-4-
yl)]methyl}methylcarboxamidine (10).
5.1.11. 5-Chloro-6-[(2-iminoimidazolidinyl)methyl]uracil
hydrobromide (18). To a solution of 3.42 g (32.3 mmol)
of cyanogen bromide in 32 mL of water was added
7.00 g (32.0 mmol) of 16 and stirred for 3 h at room
temperature. The reaction mixture was cooled with an
ice-bath. The precipitate was collected by filtration and
washed with water, DMF, and MeOH to give 3.58 g
(34%) of 18 as a beige powder: mp 242 °C dec. 1H NMR
(DMSO-d6): d 3.49–3.63 (m, 4H), 4.45 (s, 2H), 8.03 (s,
1H), 8.07 (s, 2H), 11.31 (s, 1H), 11.67 (s, 1H). FAB-MS
A mixture of
300 mg (1.58 mmol) of 9 and 661 mg (4.75 mmol) of
methylisothiourea sulfate in 10 mL of 0.4 M aqueous
KOH was stirred at room temperature for 3 days. An
insoluble material was collected by filtration and sus-
pended in 5 mL of 28% aqueous NH4OH at room
temperature for 0.5 h. The precipitate was collected by
filtration to give 281 mg (77%) of 4 as a white powder:
1
mp 200 °C dec. H NMR (DMSO-d6): d 3.00 (s, 3H),
4.25 (s, 2H). FAB-MS m=z: 232 (M+H)þ. Anal. Calcd
for C7H10ClN5O2ÆH2O: C, 33.68; H, 4.84; N, 28.05.
Found: C, 33.96; H, 4.97; N, 27.92.
m=z:
244
(M+H)HBr)þ.
Anal.
Calcd
for
C8H10ClN5O2ÆHBr: C, 29.61; H, 3.42; N, 21.58. Found:
C, 29.67; H, 3.51; N, 21.62.
5.1.7. {[5-Chloro-2,6-dioxo(1,3-dihydropyrimidin-4-yl)]-
methyl}thiocarboxamidine hydrochloride (12). A mix-
ture of 300 mg (1.54 mmol) of 11 and 140 mg
(1.84 mmol) of thiourea in EtOH was heated under re-
flux for 6 h. The precipitate was collected by filtration
and washed with EtOH to give 337 mg (81%) of 12 as a
5.1.12. 5-Bromo-6-[(2-iminoimidazolidinyl)methyl]uracil
hydrobromide (19). Compound 19 was prepared in 25%
yield from 17 and cyanogen bromide by a method sim-
ilar to that described for 18, as a white powder: mp
248 °C dec. H NMR (DMSO-d6): d 3.50–3.61 (m, 4H),
4.42 (s, 2H), 8.02 (s, 1H), 8.05 (s, 2H), 11.31 (s, 1H),
11.62 (s, 1H). FAB-MS m=z: 288, 290 (M+H)HBr)þ.
1
1
white powder: mp 220 °C dec. H NMR (DMSO-d6): d

S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
3449
Anal. Calcd for C8H10BrN5O2ÆHBr: C, 26.04; H, 3.00;
N, 18.98. Found: C, 25.97; H, 3.04; N, 18.91.
scribed for 26, as a white solid: mp 220 °C dec. 1H NMR
(DMSO-d6): d 1.15 (d, 6H, J ¼ 6:3 Hz), 3.48–3.62 (m,
4H), 4.04 (sept, 1H, J ¼ 6:3 Hz), 4.26 (s, 2H). FAB-MS
m=z: 286 (M+H)þ. Anal. Calcd for C11H16ClN5
O2Æ0.8H2O: C, 44.02; H, 5.91; N, 23.33. Found: C, 44.07;
H, 5.90; N, 23.37.
5.1.13. 2-Imino-1-methylimidazolidine hydrobromide (23).
To a solution of 10.0 g (135 mmol) of 25 in 80 mL of
benzene cooling with an ice-bath was added a solution
of 15.8 g (149 mmol) of cyanogen bromide in 20 mL of
benzene and stirred for 18 h at room temperature. The
precipitate was collected by filtration and washed with
benzene to give 23.0 g (95%) of 23 as a white powder. 1H
NMR (DMSO-d6): d 2.90 (s, 3H), 3.45–3.64 (m, 4H),
7.74 (s, 1H), 7.91 (s, 2H). EI-MS (70 eV) m=z: 99 (Mþ).
5.1.19. 2-Amino-1-methylimidazole (31). To a solution of
5.20 g (124 mmol) of cyanamide in 40 mL of 50% aque-
ous AcOH was added 5.00 g (42.0 mmol) of 29 and
heated at 100 °C for 1 h. The reaction mixture was
evaporated under a vacuum. The residue was dissolved
in 20 mL of concd HCl and heated at 100 °C for 0.25 h.
The pH of the reaction mixture was adjusted to 11 by
addition of KOH. The reaction mixture was extracted
with CHCl3, dried over MgSO4, and evaporated under a
5.1.14. 1-Ethyl-2-iminoimidazolidine hydrobromide (24).
Compound 24 was prepared in 99% yield from 21 and
cyanogen bromide by a method similar to that described
for 23, as a white powder. 1H NMR (DMSO-d6): d 1.09
(t, 3H, J ¼ 7:3 Hz), 3.33 (q, 2H, J ¼ 7:3 Hz), 3.47–3.66
(m, 4H), 7.73 (s, 1H), 7.91 (s, 2H). EI-MS (70 eV) m=z:
113 (Mþ).
1
vacuum to give 687 mg (17%) of 31 as a yellow oil. H
NMR (DMSO-d6): d 3.29 (s, 3H), 5.20 (s, 2H), 6.31 (d,
1H, J ¼ 2:6 Hz), 6.50 (d, 1H, J ¼ 2:6 Hz). EI-MS
(70 eV) m=z: 97 (Mþ).
5.1.20. 2-Amino-1-ethylimidazole (32). Compound 32
was prepared in 41% yield from 30 and cyanamide by a
method similar to that described for 31, as a yellow oil.
1H NMR (CDCl3): d 1.33 (t, 3H, J ¼ 7:3 Hz), 3.73 (q,
2H, J ¼ 7:3 Hz), 6.53 (d, 1H, J ¼ 2:6 Hz), 6.65 (d, 1H,
J ¼ 2:6 Hz). EI-MS (70 eV) m=z: 111 (Mþ).
5.1.15. 2-Imino-1-isopropylimidazolidine hydrobromide
(25). Compound 25 was prepared in 96% yield from
22 and cyanogen bromide by a method similar to that
1
described for 23, as a beige powder. H NMR (DMSO-
d6): d 1.14 (d, 6H, J ¼ 6:6 Hz), 3.45–3.61 (m, 4H), 4.03
(sept, 1H, J ¼ 6:6 Hz), 7.70 (s, 1H), 7.91 (s, 2H). EI-MS
(70 eV) m=z: 127 (Mþ).
5.1.21. 5-Chloro-6-{[2-imino-3-methyl(4-imidazolinyl)]-
methyl}uracil hydrochloride (33). A mixture of 250 mg
(1.28 mmol) of 11 and 350 mg (3.61 mmol) of 31 in
10 mL of water was heated at 100 °C for 1 h. The pre-
cipitate was collected by filtration and extracted with
2 M aqueous HCl. An insoluble material was filtered out
and the filtrate was evaporated under a vacuum. The
residue was washed with MeOH to give 80 mg (24%) of
33 as a beige powder: mp 250 °C dec. 1H NMR (DMSO-
d6): d 3.46 (s, 3H), 5.04 (s, 2H), 7.04 (d, 1H, J ¼ 2:6 Hz),
7.08 (d, 1H, J ¼ 2:6 Hz), 8.11 (s, 2H), 11.38 (s, 1H),
11.70 (s, 1H). EI-MS (70 eV) m=z: 255 (Mþ). Anal. Calcd
for C9H10ClN5O2Æ0.25H2O: C, 36.44; H, 3.91; N, 23.61.
Found: C, 36.41; H, 3.86; N, 23.55.
5.1.16.
methyl]uracil (26). To
5-Chloro-6-[(2-imino-3-methylimidazolidinyl)-
suspension of 1.84 g
a
(10.2 mmol) of 23 in 20 mL of DMF was added 1.04 g
(15.4 mmol) of NaOEt and stirred for 0.5 h at room
temperature. To the reaction mixture was added 1.00 g
(5.12 mmol) of 11 and stirred for 18 h at room temper-
ature. The reaction mixture was diluted with 100 mL of
MeOH. The precipitate was collected by filtration and
washed with 0.1% aqueous AcOH and MeOH to give
290 mg (22%) of 26 as a brown solid: mp 215 °C dec. 1H
NMR (DMSO-d6): d 2.92 (s, 3H), 3.57 (s, 4H), 4.27 (s,
2H). FAB-MS m=z: 258 (M+H)þ. Anal. Calcd for
C9H12ClN5O2: C, 41.95; H, 4.69; N, 27.18. Found: C,
41.64; H, 4.75; N, 26.80.
5.1.22.
5-Chloro-6-{[3-ethyl-2-imino(4-imidazolinyl)]-
methyl}uracil hydrochloride (34). Compound 34 was
prepared in 7% yield from 32 and 11 by a method similar
to that described for 33, as a beige powder: mp 235 °C
5.1.17.
5-Chloro-6-[(3-ethyl-2-iminoimidazolidinyl)-
methyl]uracil (27). Compound 27 was prepared in 11%
yield from 24 and 11 by a method similar to that de-
scribed for 26, as a beige solid: mp 205 °C dec. 1H NMR
(DMSO-d6): d 1.11 (t, 3H, J ¼ 7:3 Hz), 3.35 (q, 2H,
J ¼ 7:3 Hz), 3.59 (s, 4H), 4.26 (s, 2H), 9.76 (s, 1H).
FAB-MS m=z: 272 (M+H)þ. Anal. Calcd for
C10H14ClN5O2Æ0.2H2O: C, 43.63; H, 5.27; N, 25.44.
Found: C, 43.69; H, 5.23; N, 25.51.
1
dec. H NMR (DMSO-d6): d 1.26 (t, 3H, J ¼ 7:3 Hz),
3.89 (q, 2H, J ¼ 7:3 Hz), 5.06 (s, 2H), 7.07 (d, 1H,
J ¼ 2:6 Hz), 7.07 (d, 1H, J ¼ 2:6 Hz), 8.19 (s, 2H), 11.47
(s, 1H), 11.72 (s, 1H). FAB-MS m=z: 270 (M+H)HCl)þ.
Anal. Calcd for C10H12ClN5O2ÆH2O: C, 37.05; H, 4.66;
N, 21.60. Found: C, 37.33; H, 4.33; N, 21.52.
5.1.23. 6-[(2-Iminopyrrolidinyl)methyl]thymine hydro-
chloride (38). To a suspension of 1.00 g (5.73 mmol) of
35 in 10 mL of DMF was added 1.16 g (17.0 mmol) of
NaOEt and stirred for 0.5 h at room temperature. To the
5.1.18. 5-Chloro-6-[(2-imino-3-isopropylimidazolidinyl)-
methyl]uracil (28). Compound 28 was prepared in 15%
yield from 25 and 11 by a method similar to that de-

3450
S. Yano et al. / Bioorg. Med. Chem. 12 (2004) 3443–3450
reaction mixture was added 1.00 g (5.73 mmol) of 36 and
stirred for 16 h at room temperature. The precipitate
was collected by filtration, washed with water and
MeOH, and extracted with 2 M aqueous HCl. An
insoluble material was filtered out and the filtrate was
evaporated under a vacuum. The residue was washed
with EtOH to give 320 mg (22%) of 38 as a white pow-
5.2.3. Oral absorption in mice. This study was performed
as previously described.10
Acknowledgements
We are grateful to Prof. Kei Takeda of Hiroshima
University for giving us helpful advice.
1
der: mp 250 °C dec. H NMR (DMSO-d6): d 1.76 (s,
3H), 2.02 (quintet, 2H, J ¼ 7:6 Hz), 2.84 (t, 2H,
J ¼ 7:6 Hz), 3.51 (t, 2H, J ¼ 7:6 Hz), 4.55 (s, 2H). FAB-
MS m=z: 223 (M+H)HCl)þ. Anal. Calcd for
C10H14N4O2ÆHClÆ0.5H2O: C, 44.86; H, 6.02; N, 20.93.
Found: C, 44.55; H, 6.14; N, 20.72.
We thank Dr. Kazuhiro Kobayashi, Mr. Tomio
Yamazaki, and Mr. Toshiharu Mutoh for performing
mass spectra and elemental analysis studies.
References and notes
5.1.24.
5-Bromo-6-[(2-iminopyrrolidinyl)methyl]uracil
hydrochloride (39). Compound 39 was prepared in 13%
yield from 15 and 36 by a method similar to that de-
scribed for 38, as a white powder: mp 180 °C dec. H
NMR (DMSO-d6): d 2.05 (quintet, 2H, J ¼ 7:4 Hz), 2.86
(t, 2H, J ¼ 7:4 Hz), 3.59 (t, 2H, J ¼ 7:4 Hz), 4.63 (s,
2H), 9.29 (s, 1H), 9.68 (s, 1H), 11.44 (s, 1H), 11.69 (s,
1H). FAB-MS m=z: 287, 289 (M+H)HCl)þ. Anal. Calcd
for C9H11BrN4O2ÆHClÆ1.25H2O: C, 31.23; H, 4.22; N,
16.19. Found: C, 31.23; H, 4.31; N, 16.16.
1. Umeda, M.; Heidelberger, C. Cancer Res. 1968, 28, 2529.
2. Heidelberger, C.; Anderson, S. W. Cancer Res. 1964, 24,
1979.
3. Murakami, Y.; Kazuno, H.; Emura, T.; Tsujimoto, H.;
Suzuki, N.; Fukushima, M. Int. J. Oncol. 2000, 17, 277.
4. Nakayama, C.; Wataya, Y.; Meyer, R. B.; Santi, D. V.;
Saneyoshi, M.; Ueda, T. J. Med. Chem. 1980, 23, 962.
5. Rogers, W. I.; Hartman, A. C.; Palm, P. E.; Okstein, C.;
Kensler, C. J. Cancer Res. 1969, 29, 953.
1
6. Dexter, D. L.; Wolberg, W. H.; Ansfield, F. J.; Helson, L.;
Heidelberger, C. Cancer Res. 1972, 32, 247.
7. Ansfield, F. J.; Ramirez, G. Cancer Chemother. Rep. 1971,
55, 205.
8. el Kouni, M. H.; el Kouni, M. M.; Naguib, F. N. M.
Cancer Res. 1993, 53, 3687.
9. Maehara, Y.; Sakaguchi, Y.; Kusumoto, T.; Kusumoto,
H.; Sugimachi, K. J. Surg. Oncol. 1989, 42, 184.
10. Yano, S.; Kazuno, H.; Suzuki, N.; Emura, T.; Wierzba,
K.; Yamashita, J.; Tada, Y.; Yamada, Y.; Fukushima, M.;
Asao, T. Bioorg. Med. Chem. 2004, 12, see doi:10.1016/
j.bmc.2004.04.036.
5.1.25.
5-Chloro-6-[(2-iminopyrrolidinyl)methyl]uracil
hydrochloride (2). To a solution of 116 g (0.962 mol) of
35 and 293 g (1.92 mol) of DBU in 1.25 L of MeOH was
added 125 g (0.641 mol) of 11 and heated under reflux
for 2 h. The precipitate was collected by filtration,
washed with MeOH, and dissolved in 0.4 L of 2 M
aqueous HCl at 90 °C. To the reaction mixture was
added 1.6 L of EtOH and allowed to stand at room
temperature, and the precipitate was collected by
filtration to give 121.5 g (68%) of 2 as white crystals: mp
11. Ferrario, F.; Levi, S.; Sala, A.; Trupiano, F. Synth.
Commun. 1991, 21, 99.
12. Ishikawa, F.; Kosasayama, A.; Nakamura, S.; Konno, T.
Chem. Pharm. Bull. 1978, 26, 3658.
1
245 °C dec. H NMR (DMSO-d6): d 2.04 (quintet, 2H,
J ¼ 7:6 Hz), 2.87 (t, 2H, J ¼ 7:6 Hz), 3.59 (t, 2H,
J ¼ 7:6 Hz), 4.69 (s, 2H), 9.40 (s, 1H), 9.75 (s, 1H), 11.46
(s, 1H), 11.73 (s, 1H). FAB-MS m=z: 243 (M+H)HCl)þ.
Anal. Calcd for C9H11ClN4O2ÆHCl: C, 38.73; H, 4.33;
N, 20.07. Found: C, 38.76; H, 4.38; N, 20.01.
13. Miller, L. F.; Bambury, R. E. J. Med. Chem. 1972, 15, 415.
14. Lawson, A. J. Chem. Soc. 1956, 307.
15. Moriconi, E. J.; Cevasco, A. A. J. Org. Chem. 1968, 33,
2109.
16. Johnson, T. B.; Chernoff, L. H. J. Am. Chem. Soc. 1913,
35, 585.
17. Fukushima, M.; Suzuki, N.; Emura, T.; Yano, S.;
Kazuno, H.; Tada, Y.; Yamada, Y.; Asao, T. Biochem.
Pharmacol. 2000, 59, 1227.
5.2. Biology
18. Furukawa, T.; Yoshimura, A.; Sumizawa, T.; Haraguchi,
M.; Akiyama, S.; Fukui, K.; Ishizawa, M.; Yamada, Y.
Nature 1992, 356, 668.
19. Matsushita, S.; Nitanda, T.; Furukawa, T.; Sumizawa, T.;
Tani, A.; Nishimoto, K.; Akiba, S.; Miyadera, K.;
Fukushima, M.; Yamada, Y.; Yoshida, H.; Kanzaki, T.;
Akiyama, S. Cancer Res. 1999, 59, 1911.
5.2.1. Assay of human TP activity. TP was extracted
from human placenta. This study was performed as
previously described.17
5.2.2. Assay of rat UP activity. UP was extracted from
rat liver. This study was performed as previously
described.17
20. Takao, S.; Akiyama, S.; Nakajo, A.; Yoh, H.; Kitazono,
M.; Natsugoe, S.; Miyadera, K.; Fukushima, M.;
Yamada, Y.; Aikou, T. Cancer Res. 2000, 60, 5345.