Inhibition of human O6-alkylguanine-DNA alkyltransferase and potentiation of the cytotoxicity of chloroethylnitrosourea by 4(6)- (benzyloxy)-2,6(4)-diamino-5-(nitro or nitroso)pyrimidine derivatives and analogues

Article abstract of DOI:10.1021/jm970363i

A series of 4(6)-(benzyloxy)-2,6(4)-diamino-5-(nitro or nitroso)pyrimidine derivatives and analogues of which 4(6)-benzyloxy groups were replaced with a (2-, 3-, or 4-fluorobenzyl)oxy or (2-, 3-, or 4- pyridylmethyl)oxy group, was synthesized. The abilities of these compounds to inhibit human O⁶-alkylguanine-DNA alkyltransferase (AGAT) in vitro and to potentiate the cytotoxicity of 1-[(4-amino-2-methyl-5-pyrimidinyl)methyl]-3- (2-chloroethyl)-3-nitrosourea (ACNU) toward HeLa S3 cells were evaluated. 2,4-Diamino-6-[(2-fluorobenzyl)oxy]-5-nitropyrimidine (3) and 2,4-diamino-5- nitro-6-(2-pyridylmethoxy)pyrimidine (6), whose ortho positions of the 6- substituent are modified, were much weaker in terms of these abilities than the corresponding meta- or para-modified compounds. These results are consistent with those of our previous study using a series of O⁶- benzylguanine derivatives. All 5-nitrosopyrimidine derivatives examined exerted both stronger AGAT-inhibition and ACNU-enhancement abilities than the corresponding 5-nitro derivatives. Among a variety of compounds that we have examined to date, 2,4-diamino-6-[(4-fluorobenzyl)oxy]-5-nitrosopyrimidine (10) exhibited the strongest ability to inhibit AGAT, and its magnitude was 2.5 and 50 times those of 4-(benzyloxy)-2,6-diamino-5-nitrosopyrimidine (9) and O⁶-benzylguanine (1), respectively. A strong positive correlation was observed between the ability to inhibit AGAT and to potentiate the cytotoxicity of ACNU. This strongly indicates that 4(6)-(benzyloxy)pyrimidine derivatives and their analogues potentiate ACNU cytotoxicity by inhibiting AGAT activity. To characterize the reactivity of test compounds, alkyl- transfer reactions were also carried out using the biomimetic alkyl-transfer system.

Full text of DOI:10.1021/jm970363i

503
J . Med. Chem. 1998, 41, 503-508
In h ibition of Hu m a n O6-Alk ylgu a n in e-DNA Alk yltr a n sfer a se a n d P oten tia tion of
th e Cytotoxicity of Ch lor oeth yln itr osou r ea by 4(6)-(Ben zyloxy)-2,6(4)-d ia m in o-
5-(n itr o or n itr oso)p yr im id in e Der iva tives a n d An a logu es
Isamu Terashima and Kohfuku Kohda*
Faculty of Pharmaceutical Sciences, Nagoya City University, Tanabedori, Mizuho-ku, Nagoya 467, J apan
Received J une 2, 1997
A series of 4(6)-(benzyloxy)-2,6(4)-diamino-5-(nitro or nitroso)pyrimidine derivatives and
analogues of which 4(6)-benzyloxy groups were replaced with a (2-, 3-, or 4-fluorobenzyl)oxy or
(2-, 3-, or 4-pyridylmethyl)oxy group, was synthesized. The abilities of these compounds to
inhibit human O6-alkylguanine-DNA alkyltransferase (AGAT) in vitro and to potentiate the
cytotoxicity of 1-[(4-amino-2-methyl-5-pyrimidinyl)methyl]-3-(2-chloroethyl)-3-nitrosourea (ACNU)
toward HeLa S3 cells were evaluated. 2,4-Diamino-6-[(2-fluorobenzyl)oxy]-5-nitropyrimidine
(3) and 2,4-diamino-5-nitro-6-(2-pyridylmethoxy)pyrimidine (6), whose ortho positions of the
6-substituent are modified, were much weaker in terms of these abilities than the corresponding
meta- or para-modified compounds. These results are consistent with those of our previous
study using a series of O6-benzylguanine derivatives. All 5-nitrosopyrimidine derivatives
examined exerted both stronger AGAT-inhibition and ACNU-enhancement abilities than the
corresponding 5-nitro derivatives. Among a variety of compounds that we have examined to
date, 2,4-diamino-6-[(4-fluorobenzyl)oxy]-5-nitrosopyrimidine (10) exhibited the strongest ability
to inhibit AGAT, and its magnitude was 2.5 and 50 times those of 4-(benzyloxy)-2,6-diamino-
5-nitrosopyrimidine (9) and O6-benzylguanine (1), respectively. A strong positive correlation
was observed between the ability to inhibit AGAT and to potentiate the cytotoxicity of ACNU.
This strongly indicates that 4(6)-(benzyloxy)pyrimidine derivatives and their analogues
potentiate ACNU cytotoxicity by inhibiting AGAT activity. To characterize the reactivity of
test compounds, alkyl-transfer reactions were also carried out using the biomimetic alkyl-
transfer system.
In tr od u ction
derivatives including those modified at the 2-amino
group, 7-nitrogen, 9-nitrogen, or the para position of the
O6-benzyl group7,9,10 have been reported to be a more
potent AGAT inhibitor than 6BG, with the exception of
two 6BG derivatives that were modified at the 8-posi-
tion.11 We have previously described the AGAT-inhibit-
ing ability of a series of 6BG derivatives that were
modified at the O6-benzyl group. Modification at the
ortho position of the benzyl group markedly diminished
their ability to inhibit human AGAT activity, whereas
modification at the meta or para position had little
influence.12,13 Recently, Chae et al. reported that
4-(benzyloxy)-2,6-diamino-5-nitropyrimidine (NO2-BP)
and 4-(benzyloxy)-2,6-diamino-5-nitrosopyrimidine (NO-
BP), compounds with a pyrimidine moiety in the 6BG
structure, have a stronger ability to inhibit human
AGAT than 6BG.11 These 4-(benzyloxy)pyrimidine de-
rivatives are candidates as adjuvants in combination
chemotherapy with alkylating antitumor drugs.
In this report we prepared NO2-BP and NO-BP
derivatives and analogues in which benzyloxy groups
were replaced with a (fluorobenzyl)oxy or (pyridyl-
methyl)oxy group, as we had also done with 6BG
derivatives, and tested their abilities to inhibit human
AGAT activity in vitro and to potentiate the cytotoxicity
of ACNU toward HeLa S3 cells. Among the pyrimidine
derivatives examined, 2,4-diamino-6-[(4-fluorobenzyl)-
oxy]-5-nitrosopyrimidine (10) showed the strongest
Treatment of cells with mutagenic and/or carcinogenic
alkylating agents results in alkylation at many nucleo-
philic sites in DNA.1 O6-Alkylguanine (6AG) is one of
the alkylated adducts formed, and 6AG is considered
to be the DNA modification most responsible for the
induction of cancer, mutation, and cell death.2 O6-
Alkylguanine-DNA alkyltransferase (AGAT) is a repair
enzyme which is found in a variety of organisms from
bacteria to mammalian cells. AGAT removes the O6-
alkyl group of 6AG by accepting it at a thiol group in
the enzyme’s active-site cysteine.3 Cells with low AGAT
activity are known to be sensitive to alkylating agents
which alkylate the O6-position of a guanine residue.4
Survival of brain tumor patients treated with carmus-
tine, a 2-chloroethylating agent, is strongly correlated
with AGAT levels.5
O6-Benzylguanine (6BG) is a well-known low-molec-
ular-weight inhibitor of mammalian AGAT.6 Pretreat-
ment of tumor cells with 6BG results in dramatic
potentiation of the cytotoxicity of methylating,7 2-chlo-
roethylating,6,7 and 2-fluoroethylating agents.8 The
application of these agents in combination cancer che-
motherapy is currently being employed in clinical trials.
To date, numerous 6BG derivatives and their related
compounds have been synthesized and tested for their
ability to inhibit AGAT activity. However, none of these
S0022-2623(97)00363-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/24/1998

504 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 4
Terashima and Kohda
F igu r e 1. Structures and identification numbers of O6-benzylguanine and 4(6)-(benzyloxy)pyrimidine derivatives.
Ta ble 1. Dealkylation Rates of 4(6)-(Benzyloxy)pyrimidine
Derivatives and Analogues
inhibition of human AGAT in vitro, and this compound
is the strongest AGAT inhibitor among compounds that
we have examined to date.
compd
dealkylation rate,a 104k (min-1
)
1
2
3
4
5
6
7
8
9
3.41 ( 0.45
13.7 ( 0.1
20.7 ( 1.5
14.9 ( 0.6
21.0 ( 0.1
10.2 ( 0.9
38.9 ( 0.9
21.2 ( 2.9
b
Resu lts a n d Discu ssion
Structures of the 12 test compounds employed in this
study are presented in Figure 1. O6-Benzylguanine
(6BG, 1) is well-known as a strong inhibitor of human
AGAT.6 4-(Benzyloxy)-2,6-diamino-5-nitropyrimidine
(NO2-BP, 2) and 4-(benzyloxy)-2,6-diamino-5-nitroso-
pyrimidine (NO-BP, 9) were recently shown to be
stronger inhibitors of human AGAT than 6BG.11 The
NO2-BP derivatives 2-5 and 2,4-diamino-5-nitro-6-
(pyridylmethoxy)pyrimidine derivatives 6-8 were syn-
thesized by reacting 4-chloro-2,6-diamino-5-nitropyri-
midine in tert-butyl alcohol with a corresponding sodium
benzyl oxide derivative or a sodium pyridylmethyl oxide
derivative, respectively. NO-BP derivatives 9 and 10
and an analogue (11) were prepared by reacting 4-chloro-
2,6-diaminopyrimidine with a corresponding sodium
arylmethyl oxide in tert-butyl alcohol followed by nit-
rosation at the 5-position of the pyrimidine ring.
10
11
12
b
b
1.12 ( 0.33
a
A mixture of test compound (0.244 µmol), thiophenol (1.75
mmol), and triethylamine (1.75 mmol) in MeOH (1 mL) was
incubated at 60 °C. Reactions other than dealkylation proceeded
(see text).
b
(9-11) were also degraded under these reaction condi-
tions; however, they were not dearylmethylated by
thiophenolate, and neither 2,4-diamino-6-hydroxy-5-
nitrosopyrimidine nor the corresponding arylmethyl
phenyl sulfide were detected among various products
that were not identified (data not shown). It was
reported that the reaction of nitrosoarene with glu-
tathione gave arylhydroxylamine and glutathione-
sulfinamide.15 Similar types of reactions may proceed
in the reactions of 5-nitroso compounds (9-11) with
thiophenolate.
The abilities of these test compounds to inhibit human
AGAT activity were examined in vitro. Purified recom-
binant human AGAT (60-80 fmol) was incubated with
each test compound for 20 min at 37 °C. Then, 3H-
methylated DNA was added, and the remaining AGAT
activity was measured.16-18 As an example, the dose-
response curves of compounds 9 and 10 are shown in
Figure 2. The IC50 value, the dose of test compound
required to produce 50% inhibition of AGAT activity,
was calculated from the curve. The IC50 values of all
test compounds obtained are summarized in Table 2.
The ability of NO2-BP (2) to inhibit AGAT activity was
2 times greater than that of 6BG (1). Compound 12,
which is compound 2 without the 5-nitro group, was 65
times less effective than 2. These results are consistent
with those reported previously.11 Compounds 3 and 6,
whose ortho positions of the 6(4)-substituents are modi-
fied, had 9.2- and 13.8-fold lower AGAT inhibitory
ability, respectively, than 2. Compounds 4, 7, and 8,
Enzymatic repair by AGAT acting on O6-alkylguanine
in cellular DNA involves a transfer of the O6-alkyl group
to a thiol group of an active-site cysteine of AGAT.3 To
understand the chemical reactivity of the test com-
pounds employed, their dealkylation rates with thiolate
treatment were examined using a biomimetic system,
as previously reported.12,14 Briefly, a test compound was
incubated at 60 °C in MeOH containing excess amounts
of thiophenol and triethylamine, and the products
formed were then analyzed using an HPLC apparatus
equipped with a UV detector. Except for the 5-nitroso
compounds (9-11), the amount of starting material
decreased with reaction time following pseudo-first-
order kinetics, and quantitative formation of both the
dearylmethylated product and the corresponding aryl-
methyl phenyl sulfide was observed (data not shown).
The rate constants for the decrease in all test com-
pounds were calculated, and results are shown in Table
1. All NO2-BP derivatives (2-5) and analogues (6-8)
had 3-11 times greater rate constants than 6BG (1).
In contrast, one compound lacking the 5-nitro group (12)
had a 3-fold lower rate constant than 1. The higher
reactivity of compounds 2-8 may be attributed to the
5-nitro group which makes the 6(4)-arylmethyl group
more reactive to thiophenolate. 5-Nitroso compounds

Inhibition of Human AGAT
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 4 505
F igu r e 3. Correlation between the dealkylation rate and the
ability to inhibit AGAT activity. Dealkylation rates (Table 1)
were plotted against IC50 values (Table 2). Numbers in the
figure correspond to compound numbers in Figure 1.
F igu r e 2. Inactivation of AGAT by compounds 9 (b) and 10
(2). Purified AGAT was incubated with each test compound
for 20 min at 37 °C. The results are shown as percentages of
the AGAT activity remaining after incubation relative to that
of untreated control.
reactivity of 3 is greater than that of 2 while the
reactivity of 6 is similar to that of 2 (Table 1). These
results are consistent with our previous reports using
a series of 6BG derivatives.12,19 The ortho position of
the benzyl group may play an important role in the
interaction of AGAT with its substrate, and a modifica-
tion at this position may contribute to steric hindrance.
In terms of enzymatic benzyl transfer by AGAT, steric
features comprise the most important factor, and the
propensity of the substrate to lose its benzyl group
seems not to be particularly important. With regard
to 5-nitrosopyrimidine derivatives (9-11), thiophenolate
treatment resulted in chemical reactions other than
dealkylation. Although it is not clear if the same type
of AGAT inactivation can occur between NO-BP and
NO2-BP derivatives in cells, this finding suggests the
possibility that other mechanisms are involved in reac-
tions between AGAT and NO-BP derivatives. Further
experiments to resolve this issue will be required.
ACNU is a clinically employed antitumor agent. The
mechanism for its cytotoxicity includes 2-chloroethyla-
tion at the O6-position of guanine residues.20 Since
AGAT repairs this initial O6-modification, it is known
that the existence of AGAT in cells is responsible for
resistance to these antitumor alkylating agents.4 Pre-
treatment of cells with AGAT inhibitors such as 6BG
has been reported to remarkably potentiate the cyto-
toxicity of such chloroethylating antitumor agents.8,13
We therefore investigated the effect of this series of 4(6)-
(benzyloxy)pyrimidine derivatives and analogues on
potentiation of ACNU cytotoxicity. After HeLa cells
were pretreated with a test compound, the cytotoxicity
of ACNU was examined. As an example, the survival
curves of cells pretreated with several concentrations
of compound 10 are shown in Figure 4. To compare the
enhancing effects of test compounds, the “dose-modify-
ing factor (DMF)” (see Experimental Section) was
calculated as previously reported,8,12 and results are
shown in Table 3. The doses employed for pretreatment
were 0.01, 0.1, 1, and 10 µM, and the DMFs of compound
10 at 0.01, 0.1, and 1 µM were 1.4, 5.8, and 9.2,
respectively. The enhancing effect of compound 2 was
the same as that of 1, whereas it was reported that
the cytotoxicity of 1,3-bis(2-chloroethyl)-1-nitrosourea
(BCNU), the same 2-chloroethylating agent, was poten-
Ta ble 2. AGAT-Inhibitory Activity of
4(6)-(Benzyloxy)pyrimidine Derivatives and Analogues
compd
IC50 (µM)a
1
2
3
4
5
6
7
8
9
2.6 ( 0.1
1.3 ( 0.3
12 ( 0
2.5 ( 1.1
0.35 ( 0.01
18 ( 7
1.6 ( 0.1
1.9 ( 0.5
0.13 ( 0.03
0.051 ( 0.002
0.23 ( 0.08
170 ( 40
10
11
12
a
Dose required to produce 50% inhibition.
in which the meta or para positions of the 6-substituents
are modified, exhibited similar or slightly weaker activ-
ity than 2. In contrast, compound 5, the (p-fluoroben-
zyl)oxy derivative, was 3.7 times more effective than 2.
The effect of modification at the ortho, meta, and para
positions of the benzyl group on the ability to inhibit
AGAT activity was similar to what we had observed
with 6BG derivatives.19 In a series of 5-nitrosopyrimi-
dine derivatives (9-11), all had stronger ability to
inhibit AGAT activity than the corresponding 5-nitro-
pyrimidine derivatives. Compound 9 exhibited 10 times
greater AGAT inhibitory ability than 2 in our assay
system, while Chae et al. reported that both had the
same ability.11 2,4-Diamino-6-[(4-fluorobenzyl)oxy]-5-
nitrosopyrimidine (10) exhibited 2.5 times stronger
AGAT inhibitory ability than 9 (Figure 2 and Table 2),
and it was the strongest AGAT inhibitor among all
compounds that we have tested to date.
An examination of the relationship between the
dealkylation rate (Table 1) and the inhibition of AGAT
activity (Table 2) revealed that there was no obvious
correlation (Figure 3). However, the great difference
between 5-nitropyrimidine derivatives (2, 4, 5, 7, 8) and
compound 12 in their ability to inhibit AGAT may be
attributed to their chemical reactivities. Compounds 3
and 6, which have modifications at the ortho position
of the 6(4)-substituent, showed much weaker inhibition
of AGAT than did 2 (Table 2), although the chemical

506 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 4
Terashima and Kohda
F igu r e 5. Correlation between the ability to inhibit AGAT
activity and to enhance ACNU cytotoxicity. Dose-modifying
factors (at 1 µM, Table 3) were plotted against IC50 values
(Table 2); y ) -3.28 log x + 4.98, r ) 0.960 (data for compound
12 were omitted from this regression curve). Numbers in the
figure correspond to compound numbers in Figure 1.
F igu r e 4. Enhancing effect of pretreatment with compound
10 on ACNU cytotoxicity. HeLa S3 cells were treated with
compound 10 for 2 h and then with increasing concentrations
of ACNU: no addition (O), 0.01 µM compound 10 (b), 0.1 µM
(2), 1 µM (9).
Ta ble 3. Enhancing Effect of 4(6)-(Benzyloxy)pyrimidine
Derivatives and Analogues on ACNU Cytotoxicitya
Exp er im en ta l Section
dose-modifying factor
Ma ter ia ls a n d Meth od s. 1H NMR spectra were recorded
on a J EOL EX 270, GSX 400, or ALPHA 500 spectrometer
(Tokyo, J apan). Samples were dissolved in DMSO-d6, and
chemical shifts are reported in ppm using Me4Si as the internal
standard. Mass spectra were obtained with a J EOL DX-300
spectrometer (Tokyo). UV spectra were recorded on a Shi-
madzu UV-2100 spectrophotometer (Kyoto, J apan). Melting
points were measured with a Yanagimoto micromelting point
apparatus (Kyoto) and are uncorrected. HPLC analyses were
carried out using a Shimadzu LC10AD apparatus equipped
with a photodiode array UV detector, SPD-M6A (Kyoto). For
analyses of the purity of prepared compounds, a Merck
LiChrospher 100 RP-18 (e) (4 × 250 mm) column was used
and was eluted with a 1/15 M phosphate buffer (pH 6.8)-
MeOH system at a flow rate of 1 mL/min.
compd
0.01 µM
0.1 µM
1 µM
10 µM
1
2
3
4
5
6
7
8
9
1.3 ( 0.0
1.2 ( 0.0
1.0 ( 0.2
1.2 ( 0.3
1.5 ( 0.2
1.0 ( 0.1
1.2 ( 0.2
1.2 ( 0.1
4.3 ( 1.0
5.8 ( 2.6
2.9 ( 0.7
4.4 ( 1.1
4.0 ( 0.2
1.1 ( 0.2
4.8 ( 1.1
5.1 ( 0.9
1.0 ( 0.1
3.6 ( 0.5
3.9 ( 0.4
8.8 ( 1.7
9.2 ( 1.6
7.3 ( 1.3
7.6 ( 1.3
b
b
b
b
b
b
b
b
1.2 ( 0.1
1.4 ( 0.1
1.0 ( 0.1
10
11
12
b
b
1.0 ( 0.0
a
The enhancing effect is expressed in terms of a dose-modifying
factor. For experimental details, see Experimental Section. Tox-
b
6BG (1),22 4-(benzyloxy)-2,6-diamino-5-nitropyrimidine (2),23
4-(benzyloxy)-2,6-diamino-5-nitrosopyrimidine (9),24 4-(benz-
yloxy)-2,6-diaminopyrimidine (12),24 and 4-chloro-2,6-diamino-
5-nitropyrimidine25 were synthesized as previously reported.
icity appeared at this dose.
tiated by compound 2 to a greater extent than 1 using
another tumor cell line system.21 Compounds 4, 5, 7,
and 8 potentiated ACNU cytotoxicity as well as 2.
Compounds 3 and 6, which have modifications at the
ortho position of the 6-substituent, however, did not
enhance ACNU cytotoxicity whatsoever. Three 5-ni-
trosopyrimidine derivatives (9-11) were apparently
more effective than 2, and these compounds remarkably
potentiated ACNU cytotoxicity, even after 0.1 µM pre-
treatment. At a 10 µM dose, all 5-nitropyrimidine
derivatives (2-8) and 5-nitrosopyrimidine derivatives
(9-11) were cytotoxic by themselves. Compound 12 did
not exhibit any ability to enhance ACNU cytotoxicity,
nor was it cytotoxic on its own even at a 10 µM dose
(Table 3).
2,4-Dia m in o-6-[(2-flu or ob e n zyl)oxy]-5-n it r op yr im i-
d in e (3). 4-Chloro-2,6-diamino-5-nitropyrimidine (100 mg,
0.527 mmol) was added to a mixture of sodium tert-butoxide
(60 mg, 0.632 mmol) and 2-fluorobenzyl alcohol (84 µL, 0.791
mmol) in tert-butyl alcohol (10 mL). The reaction mixture was
kept standing at 30 °C for 5 h with stirring. The solvent was
evaporated, and the product was recrystallized twice with H2O/
MeOH to obtain 20.6 mg (14%) of yellow microcrystals: mp
222-224 °C; UV λmax nm (ꢀ) (H2O/MeOH ) 1/1) 233 (sh)
(13 900), 262 (4700), 269 (5100), 334 (16 500), (0.01 N NaOH)
233 (sh) (13 800), 262 (4800), 269 (5100), 334 (16 300), (0.1 N
HCl) 228 (sh) (18 800), 264 (7600), 317 (13 700); 1H NMR δ
5.47 (s, 2 H, CH2), 7.25 (m, 2 H, 3,5-ArH), 7.28 (br d, 2 H,
NH2), 7.42 (m, 1 H, 6-ArH), 7.66 (td, 1 H, 4-ArH, J ) 1.6, 7.8
Hz), 7.93 (br s, 2 H, NH2). Anal. (C11H10FN5O3) C, H, N.
2,4-Dia m in o-6-[(3-flu or ob e n zyl)oxy]-5-n it r op yr im i-
d in e (4). Compound 4 was synthesized following a procedure
similar to that described for compound 3 using 3-fluorobenzyl
alcohol (87 µL, 0.791 mmol). Recrystallization of the product
from H2O/MeOH gave 55.5 mg (38%) of a yellow filamentous
solid: mp 200-204 °C; UV λmax nm (ꢀ) (H2O/MeOH ) 1/1) 232
(sh) (14 100), 270 (sh) (4800), 334 (16 000), (0.01 N NaOH) 232
(sh) (13 900), 270 (sh) (4800), 334 (16 000), (0.1 N HCl) 228
(sh) (18 600), 263 (6800), 267 (sh) (6600), 318 (13 100); 1H NMR
δ 5.43 (s, 2 H, CH2), 7.15 (td, 2 H, 5-ArH, J ) 2.4, 8.5 Hz),
7.25 (br d, 2 H, NH2), 7.31-7.34 (m, 2 H, 2,6-ArH), 7.44 (dd,
1 H, 4-ArH), 7.93 (br s, 2 H, NH2). Anal. (C11H10FN5O3‚1/4H2O)
C, H, N.
Correlation between the ability to inhibit AGAT
activity and to enhance ACNU cytotoxicity was studied,
and the results are shown in Figure 5. There was a
strong positive correlation between these abilities (r )
0.960). These results clearly indicate that the enhanc-
ing effect of ACNU cytotoxicity was caused by inhibition
of intracellular AGAT. We previously reported a similar
strong correlation with a series of 6BG derivatives.19
5-Nitrosopyrimidine derivatives, especially compound
10, show great potential as candidates for potentiating
alkylating antitumor agents.

Inhibition of Human AGAT
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 4 507
2,4-Dia m in o-6-[(4-flu or ob e n zyl)oxy]-5-n it r op yr im i-
d in e (5). Compound 5 was synthesized following a procedure
similar to that described for compound 3 using 4-fluorobenzyl
alcohol (87 µL, 0.791 mmol). Recrystallization of the product
from H2O/MeOH gave 50.3 mg (34%) of yellow microcrystals:
mp 196-199 °C; UV λmax nm (ꢀ) (H2O/MeOH ) 1/1) 234 (sh)
(12 400), 262 (4100), 269 (4500), 333 (15 500), (0.01 N NaOH)
234 (sh) (12 400), 262 (4200), 269 (4500), 333 (15 400), (0.1 N
HCl) 228 (sh) (17 700), 263 (7100), 268 (sh) (6800), 316
(10 mL), and the mixture was heated under reflux for 1 day.
The reaction mixture was then neutralized with acetic acid
and evaporated. After H2O (20 mL) was added to the residue,
the product was extracted with CHCl3 (10 mL, performed
twice). Recrystallization of the product from benzene/hexane
gave 112.7 mg (35%) of a white solid: mp 96-97 °C; UV (H2O/
MeOH ) 1/1) λmax nm 231 (sh), 265, (0.01 N NaOH) 231 (sh),
1
264, (0.1 N HCl) 228 (sh), 278; H NMR δ 5.07 (s, 1 H, H-5),
5.18 (s, 2 H, CH2), 5.86 (br s, 2 H, NH2), 6.02 (br s, 2 H, NH2),
1
(13 000); H NMR δ 5.39 (s, 2 H, CH2), 7.22 (t, 2 H, 3-ArH, J
7.17 (t, 2 H, 3-ArH, J ) 9.0 Hz), 7.44 (dd, 2 H, 3-ArH, J 2,F
)
) 8.9 Hz), 7.25 (br s, 2 H, NH2), 7.54 (dd, 2 H, 2-ArH, J ) 5.4
5.9 Hz, J 2,3 ) 8.3 Hz); HRMS m/z calcd for C11H11FN4O
234.0917 (M+), found 234.0912; analytical HPLC (a 1/15 M
phosphate buffer (pH 6.8)-MeOH system, 30% MeOH for
0-10 min followed by a linear gradient of 30-100% MeOH
for 10-60 min) tR ) 33.0 min (95% pure).
Hz), 7.91 (br s, 2 H, NH2). Anal. (C11H10FN5O3‚1/3H2O) C, H,
N.
2,4-Dia m in o-5-n itr o-6-(2-p yr id ylm eth oxy)p yr im id in e
(6). 4-Chloro-2,6-diamino-5-nitropyrimidine (100 mg, 0.527
mmol) was added to a mixture of sodium tert-butoxide (50 mg,
0.527 mmol) and 2-pyridylmethanol (51 µL, 0.527 mmol) in
tert-butyl alcohol (10 mL). The reaction mixture was kept
standing at 30 °C for 7 h with stirring. The reaction mixture
was neutralized with acetic acid and evaporated to dryness.
The residue was washed with hot H2O/MeOH (10/1, 50 mL)
and dried to obtain 45.9 mg (33%) of a white solid. Using these
procedures, an analytical sample was obtained without further
purification: mp 256-260 °C; UV λmax nm (ꢀ) (H2O/MeOH )
1/1) 238 (sh) (12 800), 259 (8300), 266 (6600), 333 (16 600),
(0.01 N NaOH) 238 (sh) (12 800), 259 (8300), 266 (6600), 333
(16 600), (0.1 N HCl) 260 (11 500), 268 (sh) (10 000), 332
2,4-Dia m in o-6-[(4-flu or oben zyl)oxy]-5-n itr osop yr im i-
d in e (10). A solution of sodium nitrite (8.9 mg, 0.128 mmol)
in H2O (100 µL) was added to a solution of 2,4-diamino-6-[(4-
fluorobenzyl)oxy]pyrimidine (20 mg, 92.9 µmol) in 30% acetic
acid (0.5 mL) with stirring at 70 °C. After 5 min, the resulting
precipitate was filtered, washed with water-MeOH, and dried
to obtain 21.4 mg (88%) of purple powder. Using this
procedure, a more than 99% pure product was obtained;
however, an analytical sample was not obtained after several
trials of repeated recrystallization: mp 221-224 °C dec; UV
(H2O/MeOH ) 1/1) λmax nm (ꢀ) 233 (6700), 332 (22 100), (0.01
N NaOH) 233 (6600), 332 (22 100), (0.1 N HCl) 234 (5400),
331 (19 200); 1H NMR δ 5.56 (s, 2 H, CH2), 7.24 (tt, 2 H, 3-ArH,
J ) 2.5, 8.8 Hz), 7.61 (dd, 2 H, 2-ArH, J 2,F ) 2.9 Hz, J 2,3 ) 8.5
Hz), 7.82 (br s, 1 H, NH), 7.88 (br s, 1 H, NH), 8.02 (d, 1 H,
NH, J ) 3.9 Hz), 10.04 (d, 1 H, NH); HRMS m/z calcd for
1
(16 100); H NMR δ 5.48 (s, 2 H, CH2), 7.25 (br d, 2 H, NH2),
7.34 (dd, 1 H, 4-ArH, J 3,4 ) 4.9 Hz), 7.59 (d, 1 H, 6-ArH, J )
7.9 Hz), 7.86 (td, 1 H, 5-ArH, J 4,5 ) 7.6 Hz, J 5,6 ) 7.9 Hz), 7.95
(br d, 2 H, NH2), 8.56 (d, 1 H, 3-ArH). Anal. (C10H10N6O3‚
1/3CH3OH) C, H, N.
C
11H10FN5O2 263.0819 (M+), found 263.0819; analytical HPLC
2,4-Dia m in o-5-n itr o-6-(3-p yr id ylm eth oxy)p yr im id in e
(7). 4-Chloro-2,6-diamino-5-nitropyrimidine (200 mg, 1.05
mmol) was added to a mixture of sodium tert-butoxide (100
mg, 1.05 mmol) and 3-pyridylmethanol (102 µL, 1.05 mmol)
in tert-butyl alcohol (10 mL). The reaction mixture was kept
standing at 30 °C with stirring overnight. Additional sodium
tert-butoxide (10 mg) and 3-pyridylmethanol (10 µL) were
added, and the mixture was stirred overnight. The reaction
mixture was neutralized with acetic acid and evaporated to
dryness. The residue was washed with hot H2O/MeOH (10/1,
50 mL) and dried to obtain 47.1 mg (17%) of a yellow solid.
Following these procedures, an analytical sample was ob-
tained: mp 220-224 °C; UV λmax nm (ꢀ) (H2O/MeOH ) 1/1)
234 (sh) (14 100), 260 (sh) (6800), 266 (sh) (6000), 333 (17 000),
(0.01 N NaOH) 233 (sh) (13 900), 260 (sh) (6700), 266 (sh)
(6000), 334 (17 000), (0.1 N HCl) 229 (sh) (16 800), 260 (sh)
(a 1/15 M phosphate buffer (pH 6.8)-MeOH system, 30%
MeOH for 0-10 min followed by a linear gradient of 30-100%
MeOH for 10-60 min) tR ) 31.4 min (99.5% pure).
2,4-Diam in o-6-(4-pyr idylm eth oxy)pyr im idin e. 4-Chloro-
2,6-diaminopyrimidine (200 mg, 1.38 mmol) was added to a
solution of sodium (95 mg, 4.15 mmol) in 4-pyridylmethanol
(3 g), and the mixture was kept at 75 °C with stirring. After
19 h, the reaction mixture was neutralized with acetic acid
and evaporated to dryness. The product was separated on a
Sephadex LH-20 column (2.5 × 40 cm) which was eluted with
MeOH to obtain 100.6 mg (34%) of a white solid: mp 197-
198 °C; UV (H2O/MeOH ) 1/1) λmax nm (ꢀ) 232 (7600), 263
(9500), (0.01 N NaOH) 232 (sh) (7500), 263 (9700), (0.1 N HCl)
224 (sh) (9800), 260 (7700), 280 (13 600); 1H NMR δ 5.15 (s, 1
H, H-5), 5.27 (s, 2 H, CH2), 5.90 (br s, 2 H, NH2), 6.08 (br s, 2
H, NH2), 7.34 (d, 2 H, 2-ArH, J 2,3 ) 5.5 Hz), 8.53 (d, 2 H,
3-ArH); HRMS m/z calcd for C10H11N5O 217.0964 (M+), found
217.0964; analytical HPLC (1/15 M phosphate buffer (pH 6.8)-
30% MeOH) tR ) 10.4 min (99.8% pure).
2,4-Dia m in o-5-n it r oso-6-(4-p yr id ylm et h oxy)p yr im i-
d in e (11). A solution of sodium nitrite (18.4 mg, 0.266 mmol)
in H2O (200 µL) was added to a solution of 2,4-diamino-6-(4-
pyridylmethoxy)pyrimidine (42 mg, 0.193 mmol) in 30% acetic
acid (1 mL) with stirring at 70 °C. After 5 min, the reaction
mixture was evaporated and the residue was washed with cold
H2O (2 mL) and dried to obtain 41.3 mg (87%) of a purple solid.
Without further purification, an analytical sample was ob-
tained: mp 232-235 °C dec; UV (H2O/MeOH ) 1/1) λmax nm
(ꢀ) 232 (9100), 259 (sh) (4500), 330 (25 000), (0.01 N NaOH)
235 (7100), 257 (sh) (5000), 330 (25 900), (0.1 N HCl) 240 (sh)
(8900), 253 (9500), 331 (19 800); 1H NMR δ 5.64 (s, 2 H, CH2),
7.49 (d, 2 H, 2-ArH, J 2,3 ) 6.5 Hz), 7.85 (br s, 2 H, NH2), 8.07
(br s, 1 H, NH), 8.60 (d, 2 H, 3-ArH), 10.03 (br s, 2 H, NH).
Anal. (C10H10FN6O2) C, H, N.
1
(11 500), 321 (14 500); H NMR δ 5.45 (s, 2 H, CH2), 7.27 (br
d, 2 H, NH2), 7.43 (dd, 1 H, 5-ArH, J 4,5 ) 4.9 Hz), 7.91 (dt, 1
H, 6-ArH, J 2,6 ) 1.8 Hz), 7.92 (br s, 2 H, NH2), 8.54 (dd, 1 H,
4-ArH), 8.71 (d, 1 H, 2-ArH). Anal. (C10H10N6O3‚1/4H2O) C,
H, N.
2,4-Dia m in o-5-n itr o-6-(4-p yr id ylm eth oxy)p yr im id in e
(8). 4-Chloro-2,6-diamino-5-nitropyrimidine (200 mg, 1.05
mmol) was added to a mixture of sodium tert-butoxide (100
mg, 1.05 mmol) and 4-pyridylmethanol (138 µL, 1.05 mmol)
in tert-butyl alcohol (10 mL). The reaction mixture was kept
standing at 30 °C with stirring overnight. The solvent was
evaporated, and the product was separated by means of silica
gel column chromatography (Merck silica gel 60, 70-230 mesh,
63-200 µm, eluted with 0-10% MeOH in CHCl3). The
fractions collected were evaporated and then washed with cold
MeOH (4 mL) to obtain 60.2 mg (22%) of a pale-brown solid:
mp 188-191 °C; UV λmax nm (ꢀ) (H2O/MeOH ) 1/1) 233 (sh)
(12 700), 334 (15 900), (0.01 N NaOH) 233 (sh) (12 700), 333
(15 900), (0.1 N HCl) 252 (sh) (19 800), 260 (sh) (17 800), 320
1
(27 000); H NMR δ 5.47 (s, 2 H, CH2), 7.25 (br d, 2 H, NH2),
Dea lk yla tion Rea ction . Dealkylation rates of test com-
pounds were measured as previously reported.12 Briefly, a
solution containing a test compound (0.244 µmol), thiophenol
(180 µL, 1.75 mmol), triethylamine (240 µL, 1.75 mmol), and
MeOH (1 mL) in a sealed tube was incubated at 60 °C. After
appropriate times, an aliquot of the reaction mixture was
taken out and diluted with MeOH. This solution was subjected
7.46 (d, 2 H, 2-ArH, J 1,2 ) 5.8 Hz), 7.96 (br d, 2 H, NH2), 8.58
(d, 3 H, 3-ArH). Anal. (C10H10N6O3‚1/4CH3OH) C, H, N.
2,4-Diam in o-6-[(4-flu or oben zyl)oxy]pyr im idin e. 4-Chlo-
ro-2,6-diaminopyrimidine (200 mg, 1.38 mmol) was added to
a solution of sodium tert-butoxide (438 mg, 4.56 mmol) and
4-fluorobenzyl alcohol (498 µL, 4.56 mmol) in tert-butyl alcohol

508 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 4
Terashima and Kohda
(6) Dolan, M. E.; Moschel, R. C.; Pegg, A. E. Depletion of mammalian
O6-alkylguanine-DNA alkyltransferase activity by O6-benzylgua-
nine provides a means to evaluate the role of this protein in
protection against carcinogenic and therapeutic alkylating agents.
Proc. Natl. Acad. Sci. U.S.A. 1990, 87, 5368-5372.
(7) Dolan, M. E.; Mitchell, R. B.; Mummert, C.; Moschel, R. C.; Pegg,
A. E. Effect of O6-benzylguanine analogues on sensitivity of
human tumor cells to the cytotoxic effects of alkylating agents.
Cancer Res. 1991, 51, 3367-3372.
(8) Terashima, I.; Kohda, K. Cytotoxicity of fluoroethylating agents
is potentiated by O6-benzylguanine. Biol. Pharm. Bull. 1996, 19,
956-961.
(9) Moschel, R. C.; McDougall, M. G.; Dolan, M. E.; Stine, L.; Pegg,
A. E. Structural features of substituted purine derivatives
compatible with depletion of human O6-alkylguanine-DNA alkyl-
transferase. J . Med. Chem. 1992, 35, 4486-4491.
(10) Chae, M.-Y.; McDougall, M. G.; Dolan, M. E.; Swenn, K.; Pegg,
A. E.; Moschel, R. C. Substituted O6-benzylguanine derivatives
and their inactivation of human O6-alkylguanine-DNA alkyl-
transferase. J . Med. Chem. 1994, 37, 342-347.
(11) Chae, M.-Y.; Swenn, K.; Kanugula, S.; Dolan, M. E.; Pegg, A.
E.; Moschel, R. C. 8-Substituted O6-benzylguanine, substituted
6(4)-(benzyloxy)pyrimidine, and related derivatives as inactiva-
tors of human O6-alkylguanine-DNA alkyltransferase. J . Med.
Chem. 1995, 38, 359-365.
(12) Kohda, K.; Terashima, I.; Koyama, K.; Watanabe, K.; Mineura,
K. Potentiation of the cytotoxicity of chloroethylnitrosourea by
O6-arylmethylguanines. Biol. Pharm. Bull. 1995, 18, 424-430.
(13) Mineura, K.; Izumi, I.; Watanabe, K.; Kawada, M.; Kohda, K.;
Koyama, K.; Terashima, I.; Ikenaga, M. Enhancing effect of O6-
alkylguanine derivatives on chloroethylnitrosourea cytotoxicity
toward tumor cells. Int. J . Cancer 1994, 58, 706-712.
(14) Kohda, K.; Yasuda, M.; Sawada, N.; Itano, K.; Kawazoe, Y.
Dealkylation rates of O6-alkyldeoxyguanosine, O4-alkylthymi-
dine and related compounds in an alkyl-transfer system. Chem.-
Biol. Interact. 1991, 78, 23-32.
(15) Eyer, P. Reactions of oxidatively activated arylamines with
thiols: Mechanisms and biologic implications. An overview.
Environ. Health Perspect. 1994, 102 (Suppl. 6), 123-132.
(16) Kawate, H.; Ihara, K.; Kohda, K.; Sakumi, K.; Sekiguchi, M.
Mouse methyltransferase for repair of O6-methylguanine and
O4-methylthymine in DNA. Carcinogenesis 1995, 16, 1595-1602.
(17) Nakabeppu, Y.; Kondo, H.; Kawabata, S.; Iwanaga, S.; Sekiguchi,
M. Purification and structure of the intact Ada regulatory
protein of Escherichia coli K12, O6-methylguanine-DNA meth-
yltransferase. J . Biol. Chem. 1985, 260, 7281-7288.
to product analysis using an HPLC apparatus equipped with
a UV detector.
P r ep a r a tion of AGAT. Purification of human AGAT was
performed as described16,26 with slight modifications. The
plasmid PHH2427 carrying human AGAT cDNA was intro-
duced into the AGAT-deficient Escherichia coli strain, KT233
(ada-, ogt-). The cells were grown at 37 °C in LB medium
containing 50 µg/mL ampicillin and 1 mM isopropyl â-D-
thiogalactopyranoside (IPTG). Using the harvested cells, a
crude extract was prepared by sonication. After ammonium
sulfate precipitation, human AGAT was purified by column
chromatography on DEAE-Sephacel and MonoS columns
obtained from Pharmacia LKB Biotechnology Inc. Purification
was monitored by measuring AGAT activity. The MonoS
fraction was used for the AGAT-inhibition assay.
AGAT Assa y. Methyltransferase activity remaining after
the enzyme was treated with a test compound was measured
as previously reported.16-18 Briefly, the reaction mixture (100
µL) containing 70 mM Hepes-KOH (pH 7.8), 1 mM DTT, 5 mM
EDTA, various amounts of test compounds, and 60-80 fmol
of AGAT was incubated at 37 °C for 20 min. Then, 1 µg of
3H-methylated calf thymus DNA prepared by treatment with
N-[3H]methyl-N-nitrosourea was added, and the mixture was
incubated at 37 °C for another 15 min. The reaction was
terminated by addition of 300 µL of 2 M perchloric acid and
200 µL of 1 mg/mL BSA. The mixture was heated at 70 °C
for 60 min to hydrolyze DNA, and the [3H]methyl-accepted
protein was collected by centrifugation at 4 °C and washed
twice with 1 M perchloric acid. The pellet was dissolved in
10 µL of 0.1 N NaOH and then neutralized with 10 µL of 0.1
N HCl. Radioactivity was determined in a liquid scintillation
counter.
Cell Assa y. HeLa S3 cervical tumor cells were provided
by Dr. K. Mineura of Akita University, J apan. The AGAT
activity was 418 fmol/mg of protein. The cytotoxicity assay
was carried out using the same method as reported previously.8
The dose-modifying factor (DMF) was calculated as the ratio
of the IC50 value of cells treated with ACNU alone versus the
IC50 of cells treated with a test compound followed by ACNU
treatment.
(18) Sassanfar, M.; Dosanjh, M. K.; Essigmann, J . M.; Samson, L.
Relative effeciencies of the bacterial, yeast, and human DNA
methyltransferases for the repair of O6-methylguanine and O4-
methylthymine. J . Biol. Chem. 1991, 266, 2767-2771.
(19) Mineura, K.; Fukuchi, M.; Kowada, M.; Terashima, I.; Kohda,
K. Differential inactivation of O6-methylguanine-DNA methyl-
transferase activity by O6-arylmethylguanines. Int. J . Cancer
1995, 63, 148-151.
(20) Ludlum, D. B. DNA alkylation by the haloethylnitrosoureas:
Nature of modifications produced and their enzymatic repair or
removal. Mutat. Res. 1990, 233, 117-126.
(21) Pegg, A. E.; Swenn, K.; Chae, M.-Y.; Dolan, M. E.; Moschel, R.
C. Increased killing of prostate, breast, colon, and lung tumor
cells by the combination of inactivators of O6-alkylguanine-DNA
alkyltransferase and N,N′-bis(2-chloroethyl)-N-nitrosourea. Bio-
chem. Pharmacol. 1995, 50, 1141-1148.
Ack n ow led gm en t. We are grateful to Professor Y.
Kawazoe of Nagoya City University for helpful advice
and encouragement and to Drs. H. Kawate, K. Sakumi,
and M. Sekiguchi of Kyushu University for their help
in preparation of human AGAT. We thank Dr. K.
Mineura of Akita University for providing HeLa S3 cells.
We also thank Mr. A. Kobayashi for his help with HPLC
analyses. One of the authors (I.T.) gratefully acknowl-
edges a postgraduate fellowship from the J apan Society
for the Promotion of Science.
(22) Balsiger, R. W.; Montgomery, J . A. Synthesis of potential
anticancer agents. XXV. Preparation of 6-alkoxy-2-amino-
purines. J . Org. Chem. 1960, 25, 1573-1575.
(23) Kosary, J .; Diesler, E.; Matyus, P.; Kasztreiner, E. Preparation
of pyrimidine derivatives with potential cardiotonic activity. Acta
Pharm. Hung. 1989, 49, 241-247.
(24) Pfleiderer, W.; Lohrmann, R. Synthesis of 2-amino-4-alkoxy-
pteridine. Chem. Ber. 1961, 94, 12-18.
(25) O’Brien, D. E.; Cheng, C. C.; Pfleiderer, W. Pyrimidines. XVIII.
2,4-Diamino-5-nitro-6-arylaminopyrimidines. Nitration study of
2,4-diamino-6-chloropyrimidine and a literature correction. J .
Med. Chem. 1966, 9, 573-575.
(26) Koike, G.; Maki, H.; Takeya, H.; Hayakawa, H.; Sekiguchi, M.
Purification, structure, and biochemical properties of human O6-
methylguanine-DNA methyltransferase. J . Biol. Chem. 1990,
265, 14754-14762.
(27) Chueh, L.-L.; Nakamura, T.; Nakatsu, Y.; Sakumi, K.; Hay-
akawa, H.; Sekiguchi, M. Specific amino acid sequences required
for O6-methylguanine-DNA methyltransferase activity: Analy-
ses of three residues at or near the methyl acceptor site.
Carcinogenesis 1992, 13, 837-843.
Refer en ces
(1) Singer, B. The chemical effects of nucleic acid alkylation and
their relation to mutagenesis and carcinogenesis. Prog. Nucleic
Acid Res. Mol. Biol. 1975, 15, 219-284.
(2) Loveless, A. Possible relevance of O-6 alkylation of deoxygua-
nosine to the mutagenicity and carcinogenicity of nitrosamines
and nitrosamides. Nature 1969, 223, 206-207.
(3) Lindahl, T.; Sedgwick, B.; Sekiguchi, M.; Nakabeppu, Y. Regula-
tion and expression of the adaptive response to alkylating agents.
Annu. Rev. Biochem. 1988, 57, 133-157.
(4) Scudiero, D. A.; Meyer, S. A.; Clatterbuck, B. E.; Mattern, M.
R.; Ziolkowski, C. H. J .; Day, R. S., III. Sensitivity of human
cell strains having different abilities to repair O6-methylguanine
in DNA to inactivation by alkylating agents including chloro-
ethylnitrosoureas. Cancer Res. 1984, 44, 2467-2474.
(5) Belanich, M.; Pastor, M.; Randall, T.; Guerra, D.; Kibitel, J .; Alas,
L.; Li, B.; Citron, M.; Wasserman, P.; White, A.; Eyre, H.;
J aeckle, K.; Schulman, S.; Rector, D.; Prados, M.; Coons, S.;
Shapiro, W.; Yarosh, D. Retrospective study of the correlation
between the DNA repair protein alkyltransferase and survival
of brain tumor patients treated with carmustine. Cancer Res.
1996, 56, 783-788.
J M970363I