Aziridinyldinitrobenzamides: Synthesis and structure-activity relationships for activation by E. coli nitroreductase

Article abstract of DOI:10.1021/jm0498699

The 5-aziridinyl-2,4-dinitrobenzamide CB 1954 is a substrate for the oxygen-insensitive nitroreductase (NTR) from E. coli and is in clinical trial in combination with NTR-armed adenoviral vectors in a GDEPT protocol; CB 1954 is also of interest for selective deletion of NTR-marked cells in normal tissues. Since little further drug development has been carried out around this lead, we report here the synthesis of more soluble variants and regioisomers and structure-activity relationship (SAR) studies. The compounds were primarily prepared from the corresponding chloro(di)nitroacids through amide side chain elaboration and subsequent aziridine formation. One-electron reduction potentials [E(1)], determined by pulse radiolysis, were around -400 mV, varying little for aziridinyldinitrobenzamide regioisoiners. Cytotoxicity in a panel of NTR-transfected cell lines showed that in the CB 1954 series there was considerable tolerance of substituted CONHR side chains. The isomeric 2-aziridinyl-3,5-dinitrobenzamide was also selective toward NTR+ve lines but was approximately 10-fold less potent than CB 1954. Other regioisomers were too insoluble to evaluate. While CB 1954 gave both 2- and 4-hydroxylamine metabolites in NTR+ve cells, related analogues with substituted carboxamides gave only a single hydroxylamine metabolite possibly because the steric bulk in the side chain constrains binding within the active site. CB 1954 is also a substrate for the two-electron reductase DT-diaphorase, but all of the other aziridines (regioisomers and close analogues) were poorer substrates with resulting improved specificity for NTR. Bystander effects were determined in multicellular layer cocultures and showed that the more hydrophilic side chains resulted in a modest reduction in bystander killing efficiency. A limited number of analogues were tested for in vivo activity, using a single ip dose to CD-1 nude mice bearing WiDr-NTR^neo tumors. The most active of the CB 1954 analogues was a diol derivative, which showed a substantial median tumor growth delay (59 days compared with >85 days for CB 1954) in WiDr xenografts comprising 50% NTR+ve cells. The diol is much more soluble and can be formulated in saline for administration. The results suggest there may be advantages with carefully selected analogues of CB 1954; the weaker bystander effect of its diol derivative may be an advantage in the selective cell ablation of NTR-tagged cells in normal tissues.

Full text of DOI:10.1021/jm0498699

J . Med. Chem. 2004, 47, 3295-3307
3295
Azir id in yld in itr oben za m id es: Syn th esis a n d Str u ctu r e-Activity Rela tion sh ip s
for Activa tion by E. coli Nitr or ed u cta se
Nuala A. Helsby,* Graham J . Atwell, Shangjin Yang, Brian D. Palmer, Robert F. Anderson, Susan M. Pullen,
Dianne M. Ferry, Alison Hogg, William R. Wilson, and William A. Denny
Auckland Cancer Society Research Centre, School of Medical Sciences, Faculty of Medical and Health Sciences,
The University of Auckland, Private Bag 92019, Auckland, New Zealand
Received February 12, 2004
The 5-aziridinyl-2,4-dinitrobenzamide CB 1954 is a substrate for the oxygen-insensitive
nitroreductase (NTR) from E. coli and is in clinical trial in combination with NTR-armed
adenoviral vectors in a GDEPT protocol; CB 1954 is also of interest for selective deletion of
NTR-marked cells in normal tissues. Since little further drug development has been carried
out around this lead, we report here the synthesis of more soluble variants and regioisomers
and structure-activity relationship (SAR) studies. The compounds were primarily prepared
from the corresponding chloro(di)nitroacids through amide side chain elaboration and
subsequent aziridine formation. One-electron reduction potentials [E(1)], determined by pulse
radiolysis, were around -400 mV, varying little for aziridinyldinitrobenzamide regioisomers.
Cytotoxicity in a panel of NTR-transfected cell lines showed that in the CB 1954 series there
was considerable tolerance of substituted CONHR side chains. The isomeric 2-aziridinyl-3,5-
dinitrobenzamide was also selective toward NTR+ve lines but was approximately 10-fold less
potent than CB 1954. Other regioisomers were too insoluble to evaluate. While CB 1954 gave
both 2- and 4-hydroxylamine metabolites in NTR+ve cells, related analogues with substituted
carboxamides gave only a single hydroxylamine metabolite possibly because the steric bulk in
the side chain constrains binding within the active site. CB 1954 is also a substrate for the
two-electron reductase DT-diaphorase, but all of the other aziridines (regioisomers and close
analogues) were poorer substrates with resulting improved specificity for NTR. Bystander effects
were determined in multicellular layer cocultures and showed that the more hydrophilic side
chains resulted in a modest reduction in bystander killing efficiency. A limited number of
analogues were tested for in vivo activity, using a single ip dose to CD-1 nude mice bearing
WiDr-NTRneo tumors. The most active of the CB 1954 analogues was a diol derivative, which
showed a substantial median tumor growth delay (59 days compared with >85 days for CB
1954) in WiDr xenografts comprising 50% NTR+ve cells. The diol is much more soluble and
can be formulated in saline for administration. The results suggest there may be advantages
with carefully selected analogues of CB 1954; the weaker bystander effect of its diol derivative
may be an advantage in the selective cell ablation of NTR-tagged cells in normal tissues.
In tr od u ction
CB 1954 (1), is an antitumor agent that has activity
against Walker rat 256 tumor xenografts because of the
high levels of endogenous DT-diaphorase (NAD(P)H
quinone oxidoreductase, EC 1.6.99.2);7 however, CB
1954 is a poor substrate for the human form of the
enzyme (kcat ) 0.64 min-1).8 There is also interest in
the use of CB 1954 in combination with a variant of DT-
diaphorase, NQO2.9 CB 1954 is also a better substrate
for NTR (kcat ) 360 min-1)4 than it is for DT-diaphorase,
which led to interest in the use of CB 1954/NTR in
GDEPT. Expression of NTR in tumor cells provides a
large increase in sensitivity to NTR in cell culture10-13
and human tumor xenografts14,11,15 and is currently in
clinical trials in combination with NTR-armed adenovi-
ral vectors.16 CB 1954 is also of interest in combination
with NTR expression in transgenic animals as a selec-
tive cell ablation technique to provide models of human
disease.17-24
Gene-directed enzyme-prodrug therapy (GDEPT)
involves the specific expression of a prodrug-activating
enzyme in tumor cells, with utilization of this enzyme
to activate a prodrug selectively within a tumor.1 One
enzyme that has been explored extensively for GDEPT
is an oxygen-insensitive nitroreductase (NTR, EC
1.6.99.7) from E. coli (the nfsB/nfnb gene product),2,3
which possesses extensive (88-89%) homology with the
“classical” aerobic (oxygen-insensitive) nitroreductases
from Salmonella typhimurium and Enterobacter cloa-
cae.4 Recent crystal structures have shown E. coli NTR
to be a 48 kDa homodimer with two molecules of FMN
located within crevices formed at the dimer interface.5,6
It utilizes either NADPH or NADH as cofactors,4 ef-
ficiently reducing a variety of nitroaromatic substrates
to the corresponding hydroxylamines in a two-step bi-
bi mechanism.3,6 The 5-aziridinyl-2,4-dinitrobenzamide,
Either nitro group of CB 1954 can be reduced by NTR
to give an equimolar mixture of two hydroxylamines.25
This is consistent with recent crystal structure studies26
that show that 1 binds in nonidentical ways in the two
* To whom correspondence should be addressed. Telephone: +64
(0)9 373 7599 89831. E-mail: n.helsby@auckland.ac.nz. Fax: + 64 (0)9
373 7599 7502.
10.1021/jm0498699 CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/12/2004

3296 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
Sch em e 1a
Sch em e 2a
a
(i) Aziridine/MeCN or DMSO or dioxane, 20 or 35 °C, 1-48 h.
Sch em e 3a
a
(i) SOCl2/cat. DMF; (ii) R1R2NH; (iii) aziridine/EtOAc or
CH2Cl2 or THF, 20 °C, 3-18 h.
channels that provide access to the active site of the
enzyme; in channel A the 2-nitro group stacks above
the FMN cofactor, whereas in channel B the 4-nitro
group is positioned above the FMN.
Current vector delivery systems for GDEPT result in
heterogeneous gene expression in tumors, and hence,
bystander effects resulting from diffusion of activated
metabolite to nontransfected cells are important. Fol-
lowing activation by NTR, 1 has a demonstrable by-
stander effect against NTR-ve cells.13,15,27 It is widely
considered that the active cytotoxic metabolite of 1 is
the 4-hydroxylamine and that the potent cytotoxicity of
this metabolite is related to its further activation by
acetylation to form a cytotoxic DNA interstrand cross-
linking agent.28-30 However, recent studies31 have in-
dicated that the 2-amine metabolite of 1 (derived from
the unstable 2-hydroxylamine) is at least as potent
against human tumor cell lines and has superior diffu-
sion characteristics in tumor tissue, suggesting that it
is also an important bioactive metabolite of 1.
a
(i) 30% H2O/AcOH; (ii) conc H2SO4/fuming HNO3; (iii) SOCl2/
cat. DMF, then NH3; (iv) SOCl2/cat. DMF, then H2NCH2CH(OH)-
CH2OH; or NH4OH(v), aziridine/EtOAc, 20 °C, 16 h.
Sch em e 4a
Despite the wide interest in 1 in the above contexts,
little further drug development has been carried out
around this lead, and it is not clear that 1 is the optimal
aziridinyldinitrobenzamide prodrug for NTR. It is, for
example, relatively insoluble, requiring formulation in
N-methylpyrrolidone and poly(ethylene glycol) 300 for
clinical application16 and is only a moderately efficient
substrate for NTR (kcat ) 360 min-1).4 In addition, its
bystander effect is less efficient than the corresponding
nitrogen mustard, suggesting that this aspect could also
be improved.15 We report here a structure-activity
relationship (SAR) study of analogues of 1, including
both more soluble variants and regioisomers, and rela-
tive cytotoxicity in a panel of NTR-transfected cell lines.
We also report structure-activity relationships for the
activation of these compounds by DT-diaphorase. A
limited number of analogues were also tested for
antitumor activity in vivo, and the efficiency of the
bystander effect of these activated prodrugs was deter-
mined.
a
(i) Aziridine/EtOAc or CH2Cl2, 20 °C, 2-18 h.
amide33 (30) (Scheme 2); the fluoro compounds were
needed to ensure reaction of these less electron-deficient
benzamides with aziridine. The 4-(methylsulfonyl)-2-
nitro analogue 13 was similarly prepared from the
known34 5-fluoro-4-(methysulfonyl)-2-nitrobenzamide (31).
The isomeric 2-(methylsulfonyl)-4-nitro compound 10
has been reported,35 but its synthesis has not; it and
the diol analogue 11 were prepared from 5-chloro-2-
(methylsulfanyl)benzoic acid (32) by the route shown in
Scheme 3.
The 2-aziridinyl-3,5-dinitrobenzamides 14-16 were
similarly prepared from the known36 or prepared 2-chlo-
ro analogues 37-39, respectively (Scheme 4). Finally,
the 3-aziridinyl-2,6-dinitrobenzamide (17) and the 4-azir-
idinyl-3,5-dinitrobenzamide (18) were prepared from the
corresponding known50,51 chlorides 40 and 41 (Scheme
4).
The metabolite of 1, 5-(aziridinyl)-4-hydroxylamino-
2-nitrobenzamide 42, was prepared as reported previ-
ously.37
Ch em istr y
The 5-aziridinyl-2,4-dinitrobenzamide analogues 2-8
were prepared from 5-chloro-2,4-dinitrobenzoic acid32
(20) by reaction with SOCl2, followed by coupling of the
crude 5-chloro-2,4-dinitrobenzoyl chloride (21) with the
appropriate amines and subsequent displacement of the
5-chloro group with aziridine (Scheme 1). The mononi-
trobenzamides 9 and 12 were similarly prepared from
3-fluoro-4-nitrobenzamide (29) and 2-nitro-5-fluorobenz-

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3297
Ta ble 1. Structures and Physicochemical Properties of Aziridinyldinitrobenzamides
a
compd
formula
X
R
log P
solb
1.6
22
1.8
3.3
1.6
5.6
1.1
E(1) c
1
2
3
4
5
6
7
8
Ia
Ia
Ia
Ia
Ia
Ia
Ia
Ia
Ib
Ib
Ib
Ic
Ic
II
NH2
0.14
-1.04
0.31
-0.26
-1.05
-0.48
-0.48
0.56
-399 ( 7
NHCH2CH(OH)CH2OH
NH(CH2)2Nmorpholine
N(Me)(CH2)2Nmorpholine
NH(CH2)3Nmorpholine
NH(CH2)4Nmorpholine
NH(CH2)2Nimidazole
NH(CH2)2CO2Me
NH2
-382 ( 9
-402 ( 9
-390 ( 9
0.8
-399 ( 8
-506 ( 8
-438 ( 8
9
H
0.47
0.43
0.88
4.15
3.4
0.78
1.3
2.5
0.35
2.7
10d
11
12
13
14
15
16
17
18
SO2Me
SO2Me
H
NH2
-0.41
-0.96
0.7e
-0.30
0.34
-0.15
0.11
-0.17
0.59
NHCH2CH(OH)CH2OH
NH2
-521 ( 8
-420 ( 8
-394 ( 9
-397 ( 8
SO2Me
NH2
NH2
II
II
III
IV
NHCH2CH(OH)CH2OH
NH(CH2)2Nmorpholine
NH2
-405 ( 8
-422 ( 9
NH2
0.04
a
b
Measured in n-octanol/water. Solubility limit (mM) in cell culture medium. c One-electron reduction potential versus NHE (mV).
Previously reported.35 e Value approximate because of drug instability.
d
Resu lts a n d Discu ssion
hydrophilic compounds (10 and 13). Bystander activity
following activation of prodrugs by NTR has been
demonstrated to correlate with lipophilicity15 presum-
ably because of the ability of lipophilic metabolites to
readily diffuse through the cell membrane. Hence, the
decrease in lipophilicity compared with 1 may compro-
mise the diffusion characteristics of the activated me-
tabolites of hydrophilic compounds such as 6 and 2.
CB 1954 (1) belongs to one of 16 possible regioisomer
families of aziridinyldinitrobenzamides. In the present
study, examples from four of these classes (denoted I-IV
in Table 1) preserving a meta relationship between the
nitro groups were prepared and evaluated. In the class
I series (5-aziridinyl-2,4-dinitrobenzamides), which in-
cludes CB 1954, a variety of carboxamide analogues
were prepared and either of the nitro groups were
replaced with hydrogen (compounds 9 and 12) or the
strongly electron-withdrawing methylsulfone (com-
pounds 10, 11, and 13).
Red u ction P oten tia l. One-electron reduction po-
tentials [E(1)] for the compounds were determined by
pulse radiolysis and are presented in Table 1. Potentials
were determined in aqueous solutions containing 2-pro-
panol (0.1 M) buffered at pH 7.0 (1 mM phosphate) by
measuring the equilibrium constant38 for electron trans-
fer between the radical anions of the compounds and
methyl viologen (-447 ( 7 mV) as reference standard.
The four carboxamide regioisomers (1, 14, 17, 18) had
E(1) values within 30 mV of each other. Reduction
potentials for analogues within class I also varied very
little, becoming more negative by only 20-40 mV even
on replacement of nitro by methylsulfone (compounds
1/10 and 1/13). However, the mononitrobenzamide
analogues 9 and 12, where the 4-nitro and 2-nitro
groups are replaced by hydrogen, have relatively low
reduction potentials (-506 and -521 mV, respectively).
Ch ar acter ization of NTR-Tr an sfected Cell Lin es.
NTR enzyme activity and protein levels were deter-
mined in four cell lines stably transfected with NTR.
The EMT6-NTRpuro transfected line has a bicistronic
expression cassette driven by the EF-1R promoter,39
while the others (V79-NTRpuro, WiDr-NTRneo, and Skov3-
NTRneo) use monocistronic cassettes expressed from the
CMV promoter. The relative expression of NTR protein
in these cell lines as determined by immunoblot analy-
sis indicated that the rank order of expression was
EMT6-NTRpuro > Skov3-NTRneo g WiDr-NTRneo . V79-
NTRpuro (Figure 1A). The expression of immunoreactive
Compounds 2, 11, and 15 were synthesized and
evaluated biologically as racemates, and no assessment
of potential differences in biological activity of the
enantiomers was attempted.
Solu bility a n d Lip op h ilicity. The potential to
increase the solubility of 1 by appending solubilizing
functionality on the carboxamide group was examined
in the class I series, using both neutral and basic
hydrophilic substituents (compounds 2-8). Compounds
2-6 had similar or greater solubility than 1. The most
soluble compound (in culture medium) was 2 (Table 1),
which was ca. 20-fold more soluble than 1. However,
the class II (2-aziridinyl-3,5-dinitrobenzamide) regio-
isomer 15 was considerably less soluble than 2 (solubil-
ity limit in culture medium was 2.5 vs 22 mM).
There was significant variation in the measured log P
values (determined in n-octanol/water) between the
carboxamide regioisomers, with the 3-aziridinyl-2,6-
dinitro (class III) derivative 17 being the most hydro-
philic (log P ) -0.17) and the 4-aziridinyl-3,5-dinitro
(class IV) compound 18 being the least (log P ) 0.59).
The increase in lipophilicity observed in the class II
isomer 14 with respect to 1 was consistent for the
substituted carboxamide analogues 15/2. Replacement
of either of the nitro groups of 1 with SO2Me gave more

3298 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
F igu r e 1. Relationship between metabolism of CB 1954 (1), NTR expression, and cytotoxicity across a panel of transfected cell
lines. (A) The relative expression of the immunoreactive NTR protein in the NTR transfected cell lines was determined by
immunoblot analysis of cell lysates (3 µg protein): (lane 1) Skov3-NTRneo; (lane 2) V79-NTRpuro; (lane 3) WiDr-NTRneo; (lane 4)
EMT6-NTRpuro. (B) NTR enzymatic activity was determined in intact cells by the kinetics of loss of the selective NTR substrate
6-bis(2-bromoethyl)amino-3,5-dinitrobenzamide from the extracellular medium in cell suspensions (105 cells/mL). Values are the
average and range of two determinations. The line is the log-log regression (r2 ) 0.98; slope is -1.34). (C) The growth inhibitory
activity (IC50) of CB 1954 (1) was determined following an 18 h drug exposure. Rate constants for CB 1954 metabolism by whole
cells (kmet) are from part B. (D) The growth inhibitory activity (IC50) of the synthetic 4-hydroxylamine metabolite (42) was determined
following a 4 h exposure.
NTR protein correlated with the activity of the enzyme
in intact cells as determined by the kinetics of loss of
the selective NTR substrate 6-bis(2-bromoethyl)amino-
3,5-dinitrobenzamide, with first-order rate constants of
1.89 h-1 for EMT6-NTRpuro, 0.71 ( 0.027 h-1 for Skov3-
NTRneo, 0.26 ( 0.02 h-1 for WiDr-NTRneo, and 0.124 (
0.017 h-1 for V79-NTRpuro. The rates of metabolism in
difference in sensitivity of the parental and NTR-
transfected cell pairs, indicating that the 4-hydroxyl-
amine is not itself a substrate for the NTR enzyme. The
Skov3 lines were least sensitive to the 4-hydroxylamine,
while the V79 lines (V79-NTRpuro and V79oub) were
approximately 8-fold more sensitive than Skov3 and
3-fold more sensitive than EMT6 or WiDr. The sensitiv-
ity of the Chinese hamster fibroblast background to the
4-hydroxylamine metabolite was confirmed by clono-
genic assay, with the concentration required for 1
log kill (18 h exposure, 106 cells/mL) being 221 ( 34 µM
for WiDr-NTRneo cells and 44 ( 12 µM for V79-NTRpuro
cells. Thus, the unusual sensitivity of V79-NTRpuro cells
to 1 appears to be due to its sensitivity to the 4-hy-
droxylamine metabolite rather than any differences in
prodrug activation. This might reflect differences in the
further bioactivation of this metabolite via N-acetyla-
tion,29 which results in formation of a potent DNA cross-
linking agent29 or differences in repair of the resulting
DNA adducts, in particular the putative C8-O6 ad-
duct.4 In this respect it may be significant that V79 cells
do not constitutively express O6-deoxyguanosine alkyl-
transferase.40
the parental cell lines were all less than 0.012 h-1
.
Cytotoxicity in NTR-Tr a n sfected Cell Lin es: CB
1954. Growth inhibitory potencies of the aziridinyldini-
trobenzamides were determined in each of the NTR-
expressing cell lines and compared with the correspond-
ing NTR-ve lines (Table 2). The cytotoxicity (IC50) of 1
in the NTR+ve lines and the ratios of IC50 values
relative to the parental lines correlated with NTR
expression and activity with the exception of the Chi-
nese hamster fibroblast line V79-NTRpuro. To determine
why the V79 line is so sensitive to 1, rates of metabolism
of 1 and sensitivity to the main cytotoxic metabolite 42
(the 4-hydroxylamine) were compared across the cell
lines. Rates of metabolism of 1 in whole cells were
closely correlated with NTR activity (Figure 1B), sug-
gesting that there is no other significant CB 1954
reductase activity in V79-NTRpuro cells. Plotting the IC50
against the rate of reduction of 1 (Figure 1C) demon-
strated the anomalous sensitivity of the V79 line to
NTR-activated CB 1954. The cytotoxicity of the syn-
thetic 4-hydroxylamine metabolite 42 was determined
across the cell line panel (Figure 1D). There was no
Cytotoxicity in NTR-Tr an sfected Cell Lin es: An a-
logu es. The mononitro analogues of 1 (the 2-H analogue
12 and 4-H analogue 9) showed little or no selective
toxicity toward the NTR-transfected cell lines (Table 2).
Given that there is no steric impediment in these
compounds, the lack of activation by NTR presumably

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3299
Ta ble 2. Growth-Inhibitory Activity (IC50 in µM) of Aziridinyldinitrobenzamides in NTR-Transfected versus Parental Cell Lines
Chinese hamstera
human colonb
IC50
human ovarianc
IC50
mouse mammaryd
IC50
e
IC50
compd
NTR+ve
ratiof
NTR+ve
ratio
NTR+ve
ratio
NTR+ve
ratio
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
0.26 ( 0.02
0.71 ( 0.21
0.15 ( 0.025
1.8 ( 0.45
0.048 ( 0.012
0.21 ( 0.053
0.31 ( 0.09
0.33 ( 0.07
61.5
2090 ( 212
381 ( 79
4160 ( 974
235 ( 81
6430 ( 1306
1700 ( 608
1960 ( 832
1920 ( 345
1.4
1.14 ( 0.05
3.7 ( 0.73
0.72 ( 0.02
8.1 ( 0.36
0.24 ( 0.04
0.31 ( 0.06
1.1
2.6
62.4 ( 6.9
26 ( 3
281 ( 10
89 ( 3.7
22 ( 1.6
13 ( 0.8
12.5 ( 0.8
18 ( 1.5
58.2 ( 9.3
>20
51.1 ( 2.2
49 ( 6.2
183 ( 16
21 ( 3
0.64 ( 0.04
2.3 ( 0.31
0.45 (0.11
13 ( 6
0.15 ( 0.013
0.30 ( 0.8
0.58 ( 0.005
2.5 ( 0.24
61
317 ( 21
83.7 ( 30.9
1100 ( 444
46 ( 10.1
732 ( 157
513 ( 57
105 ( 4
270 ( 15
1.6
0.098 ( 0.01
1.1 ( 0.25
0.053 ( 0.01
1.8 ( 1.2
947 ( 126
86.5 ( 12.8
4020 ( 1056
309 ( 191
3310 ( 346
12170 ( 669
278 ( 114
1520 ( 291
2.8 ( 0.25
125 ( 14
32 ( 9.2
122 ( 11
93 ( 3
0.04 ( 0.01
0.06 ( 0.009
0.29 ( 0.06
0.31 ( 0.032
41.4 ( 5.6
3.6 ( 0.08
84.5 ( 24.5
67 ( 2.2
41.5
63.6
1.6 ( 0.01
33 ( 4
10 ( 0.2
1.4 ( 0.2
10 ( 2
85 ( 9.4
71 ( 5.5
32
8.8 ( 1.9
132 ( 12
81 (1.9
87 ( 24
>36
6.3 ( 0.5
160 ( 21
26
1.2
160 ( 5
1.5 ( 0.14
12 ( 0.9
865 ( 103
208 ( 11
>63
103 ( 5
1.3 ( 0.32
25 ( 1.7
40 ( 4
21.8 ( 2.2
6.1 ( 1.0
7.3 ( 0.6
11 ( 4.1
65.8 ( 2.6
>20
19 ( 4.4
281 ( 245
185 ( 44
>72
11 ( 2
1.2 ( 0.2
6.1 ( 0.36
7.3
1.3 ( 0.2
627 ( 54
252 ( 38
177
4.4 ( 0.4
3.4 ( 0.7
135 ( 54
>20
3.8 ( 0.4
5.9 ( 0.2
9.6
27 ( 2.1
9.2 ( 0.9
>20
a
b
Chinese hamster fibroblast: wild-type (NTR-ve) is V79oua; transfected (NTR+ve) is V79-NTRpuro
.
Human colon: wild-type is WiDr;
transfected is WiDr-NTRneo
.
c Human ovarian: wild-type is Skov-3; transfected is Skov3-NTRneo
.
Mouse mammary: wild-type is EMT6;
d
transfected is EMT6-NTRpuro e IC50 (µM) for an 18 h drug exposure. f Ratio ) IC50(NTR-)/IC50(NTR+).
.
reflects a two-electron reduction potential that is too low
for efficient reduction by NTR; this is consistent with
the lowering of E(1) by at least 100 mV relative to 1 in
both these compounds (Table 1). Replacement of either
nitro group by the nearly isoelectronic SO2Me (10 and
13) gave E(1) values only slightly less than that of 1
(Table 1) and provided NTR IC50 ratios between those
of 1 and the RdH analogues (Table 2). Of the SO2Me
analogues, the 4-NO2 regioisomer 10 was the more
potent and selective, but the 2-NO2 compound 13
showed ratios in the range 10- to 27-fold. This is
consistent with the recent demonstration that reduction
of the 2-NO2 group of 1 can contribute to its activity as
an NTR prodrug.31 The relative cytotoxicities of the
regioisomers of 1 were also tested (Table 2). The class
II compound (2-aziridinyl-3,5-dinitrobenzamide, 14) was
approximately 10-fold less potent than 1 against the
NTR+ve lines but showed selectivity relative to the
NTR-ve lines that was similar to 1. However, the class
III regioisomer 17 was much less potent and selective
against NTR+ve cells (Table 2). The activity of the class
IV regioisomer 18 could not be evaluated because of poor
solubility in culture medium (Table 1).
lines were the C2 and C3 morpholides 3 and 5, although
the C4 morpholide (6), the C2 imidazole (7), and the C2
ester (8) showed broadly comparable activity. These
results show that while both nitro groups are important
for high activity, both the nature of the carboxamide
and its disposition can be varied.
NTR Meta bolism a n d Id en tifica tion of Meta bo-
lit es. The cytotoxicity of the aziridinyldinitrobenz-
amides will be related to both the relative expression
of NTR and kinetics (particularly the intrinsic clearance,
kcat/KM) of reduction of the compound for the enzyme.
However, the identity of the metabolite(s) formed will
also have a role because both the 4- and 2-nitro
reduction products of 1 are potent cytotoxins.31 The
unusual sensitivity of V79-NTRpuro cells to 1 appears
to be due to its sensitivity to the 4-hydroxylamine
metabolite 42, and similarly, this cell line was more
sensitive to the analogues 2-8, 10, 11, and 14-16
(Table 2), indicating that these analogues may be
activated to a similar 4-hydroxylamine metabolite.
Formation of the hydroxylamine metabolite(s) was
determined by LC/MS analysis following incubation of
the analogues with purified NTR enzyme in the pres-
ence of NADPH. Metabolism of 1 gave two products with
retention times of 6.6 and 9.9 min (Table 3). Both were
identified as the previously reported25 hydroxylamines
by their m/z values (14 mass units less than the parent
dinitro compounds). Comparison of their UV spectra
with the literature30,41 and with authentic standards of
the related 2- and 4-hydroxylamine (42) derivatives of
1 allowed assignment of the 6.6 min peak (λmax ) 261
nm) as the 4-hydroxylamine and the 9.9 min peak (λmax
) 246 nm) as the 2-hydroxylamine. The increased
retention time of the 2-hydroxylamine metabolite is
considered to be a consequence of intramolecular H-
bonding to the adjacent carboxamide, an interaction
unavailable to the 4-hydroxylamine derivative.41 Com-
pound 2 also gave two hydroxylamines with similar
mass spectra, λmax changes, and retention times relative
to the dinitro compound (Table 3) and were assigned
similarly. In contrast, metabolism of compounds 3-8
A primary goal was to explore the bulk tolerance
around the carboxamide group because it is a potential
position from which to append solubilizing functionality.
Compounds 2-8 examine this in the class I (5-aziridi-
nyl-2,4-dinitrobenzamide) series, using both neutral and
basic hydrophilic substituents. All these analogues were
much more cytotoxic in the NTR+ve cell lines, and
hence, there is a considerable tolerance of substituted
CONHR side chains. Some of the basic side chains
increased IC50 ratios over that for 1 and neutral hydro-
philic side chains (hydroxy, ester groups) were also
acceptable; in particular, analogues 3 and 5-8 had IC50
ratios as high or higher than 1. The class II analogues
with modified carboxamide side chains (15, 16) were
also bioactivated but, as for the unsubstituted carbox-
amide, were less potent than the class I compounds and
had lower IC50 ratios than 14. The only compounds with
ratios clearly superior to those of 1 across all the cell

3300 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
Ta ble 3. Metabolism of Aziridinyldinitrobenzamides by the NTR Enzyme
prodrug
metabolite X
metabolite Y
a
b
a
b
a
b
compd
T
λmax
MW
m/z c
T
λmax
m/z c
T
λmax
m/z c
1
2
3
4
5
6
7
8
15
11.6
10.6
12.9
13.9
14.4
16.6
13.5
13.8
11.1
270/330
270/330
271/330
274/330
271/330
270/330
271/330
270/330
322
252.2
326.3
365.3
379.4
379.4
393.4
346.3
338.3
326.3
251
6.6
6.1
261
261
237.1
311.0
350.3
364.0
364.1
378.1
331.2
323.0
311.0
9.9
9.4
246
247
237.1
311.0
325.0
364.2
378.1
378.0
392
345.2
338
325.0
10.4
11.1
11.9
14.5
10.7
10.7
7.5
261.5
262.5
261.5
261
261.5
261
248
a
b
Retention time in minutes. Absorbance maximum (nm). c APCI negative mode molecular ion.
resulted in the formation of a single hydroxylamine
metabolite, with a distinctive mass spectrum 14 mass
units lower than the corresponding parent compound
(Table 3). While the retention times of these metabolites
differed on reverse-phase HPLC because of their more
lipophilic (neutral form) side chains, they all had λmax
values of 261-262.5 nm, suggesting they are 4-hydroxyl-
amines. This assignment is consistent with the marked
sensitivity of V79-NTRpuro cells as noted above. Forma-
tion of a single metabolite for compounds 3-8 may be
due to the steric bulk in the side chain, which may
constrain the orientations within the active site, en-
abling only the 4-nitro group of these compounds to
align with the N1 of the FMN in the substrate binding
pocket.6,26
Given that the 2-amino metabolite of CB 1954 ap-
pears to be important as a mediator of bystander
effects,31 analogues (such as 2) that retain the ability
to be reduced at the 2-NO2 group may be preferable.
The class II (2-aziridinyl-3,5-dinitrobenzamide) com-
pound 15 also gave a single hydroxylamine metabolite,
but in the absence of further information it is not clear
whether this is the 3- or 5-hydroxylamine.
Activa tion by DT-d ia p h or a se (DTD). One of the
potential drawbacks to 1 as a nitroreductase-activated
prodrug is that it is also a substrate for DT-diaphorase.
Reduction of 1 by human DTD is less efficient than by
rat DTD;8 however, this widely distributed nitroreduc-
tase enzyme42,43 provides a potential route for nonselec-
tive activation of the prodrugs in normal human tissues,
an undesirable characteristic for a GDEPT prodrug. The
activation of 1 by DTD was confirmed in the present
work using the Chinese hamster DTD-133 cell line,
which expresses rat DTD with a total DTD activity 133-
fold higher than the parental CHO-K1 line. 44 IC50 ratios
of 23.5 ( 0.9 were shown by 1 in DTD-133 relative to
parental Chinese hamster ovary (CHO) cells. However,
all of the other aziridines (even close analogues of 1)
were much less affected, with IC50 ratios close to 1
(Supporting Information). Thus, the SAR for activation
by DTD is narrower than for NTR, and the analogues
show improved specificity for NTR relative to DTD.
Bysta n d er Activity in Mu lticellu la r La yer Cu l-
tu r es. Addition of solubilizing moieties to the carbox-
amide group in class I analogues can provide high
selectivity for NTR as well as improved formulation in
comparison to 1. However, an important consideration
for activity of these compounds in GDEPT is the ability
to produce a bystander effect due to diffusion of acti-
vated metabolite to nontransfected tumor cells. We have
recently shown that bystander effects of dinitrobenz-
amide aziridines and mustards increase with the lipo-
philicity of the prodrug,15 which presumably correlates
with the lipophilicity of the corresponding nitro reduc-
tion products that mediate the bystander effect. The
effect of side chain substitution on the bystander effect
was determined for a subset of the analogues in mul-
ticellular layer (MCL) cocultures, which provide a model
for the extravascular compartment in tumors.15 This
method utilizes the differential antibiotic sensitivity of
the parental and NTR transfected cell lines to determine
the prodrug concentration required to reduce cell sur-
vival to 10% (C10) for both the activator (NTR express-
ing) and target (NTR-ve) cells, as illustrated for 1 in
MCLs containing 3% or 10% V79-NTRpuro activators
(balance V79oua targets) in Figure 2A. This shows that
the C10 for target cells in the cocultures (TC) is lowered
markedly relative to that for target cells in MCLs with
no activators (T), thus demonstrating a strong bystander
effect. The shift in TC was less pronounced for the diol
analogue 2 (Figure 2B), indicating a significant but
weaker bystander effect.
The C10 values are summarized for the analogues in
Table 4. Compounds 1-3 and 8 all showed clear
bystander effects in V79 MCLs, while the class II
regioisomer 14 showed no bystander killing at its
solubility limit. The results suggest that the more
hydrophilic side chains in 2 and 3 do compromise
bystander killing to some extent. Thus, 1-3 and 8 all
have similar potency in MCLs comprising NTR-ve
V79oua target cells only, and 2 and 3 are at least as
potent as the more lipophilic 1 and 8 against V79-
NTRpuro activator cells in the cocultures but are ca. 2-
to 3-fold less potent in killing target cells in the
cocultures.
Compounds 1-3 also showed bystander effects in
WiDr MCLs (Table 4), but these were generally less
marked than in the V79 system because of the greater
potency of the prodrugs against NTR-ve WiDr cells (in
the absence of NTR+ve activators). Overall, the data
suggest only a modest reduction in bystander killing
efficiency for prodrugs with more hydrophilic side
chains.
In Vivo Activity. A subset of compounds (Table 5)
were selected for preliminary in vivo testing on the basis
of potency and/or NTR selectivity in the IC50 panel (3),
improved aqueous solubility (2, 3), replacement of the
2-NO2 group with a nonmetabolizable electron-with-
drawing moiety (10), or chemical novelty (14-16). The
compounds were administered as a single ip dose to
CD-1 nude mice bearing WiDr-NTRneo tumors at dose
levels determined from preliminary determinations of

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3301
F igu r e 2. Bystander killing by CB 1954 (A) and a more soluble dihydroxy analogue 2 (B) in V79 multicellular layer (MCL)
cocultures containing mixtures of NTR+ve cells (activators) and NTR-ve cells (targets). Representative clonogenic survival curves
are illustrated: (O) activators in MCLs grown from 3% activators/97% targets; (4) activators in MCLs grown from 10% activators/
97% targets; (b) targets in MCLs grown from 3% activators/97% targets; (2) targets in MCLs grown from 10% activators/97%
targets; ([) MCLs comprising target cells only.
Ta ble 4. Bystander Effects in Multicellular Layer (MCL) Culturesa
V79 MCLb
WiDr MCLc
compd
AC (µM)d,g
TC (µM)e,g
T (µM)f,g
AC (µM)d,g
TC (µM)e,g
T (µM)f,g
1
2
3
23 ( 3 (3)
26 ( 12 (2)
11.5 ( 2.5 (2)
20
173( 27 (3)
840 ( 310 (2)
470 ( 70 (2)
200
1400 ( 180 (4)
1595 ( 550 (4)
1600
33 ( 6 (3)
28 ( 4 (2)
8.5
252 ( 104 (3)
513 ( 13 (2)
320
105
nd
540
680
nd
8
1900
nd
14
450
>2000
>2000
nd
nd
nd
a
MCLs comprising NTR-ve cells (targets) only, or cocultures of NTR+ve (activator) and target cells, were exposed to prodrugs at a
range of concentrations for 5 h under 95% O2 (to suppress any bioreductive activation by one-electron reductases under hypoxia). Clonogenic
survival of both cell populations was assessed by plating in selective media (see Experimental Section). Targets V79oua cells; cocultures
b
comprised 3% V79-NTRpuro (activators) and 97% targets. c Targets WiDr cells; cocultures comprised 10% WiDr-NTRneo (activators) and
d
e
f
90% targets. C10 (concentration for 10% cell survival) for activators in MCL cocultures. C10 for targets in MCL cocultures. C10 for
targets in MCLs grown from targets only. Values are the mean ( range or SEM (number of experiments are in parentheses).
g
Ta ble 5. In Vivo Activity of Selected Aziridinyldinitrobenzamides against WiDr-NTRneo Tumors
median tumor
compd
dosea (µmol/kg)
vehicle
growth delayb (days)
cures
deaths
1
2
3
10
14
15
16
200
1000
300
178
750
DMA/PEG/water (1:4:5 v/v/v)
saline
0.2 M acetate buffer (pH 4)/DMSO (9:1 v/v)
DMSO/PEG/water (1:4:5)
DMA/PEG(2:8 v/v)
>87c
>86
35
17
47
8
-1
31/53
2/7
1/7
1/7
1/7
12/53
1/7
0/7
2/7
0/7
1000
178
water
0/6
0/3
1/7
4/7
0.2 mM acetate buffer (pH 4)
a
b
Dose was selected on the basis of preliminary investigations of maximal tolerated dose in C3H/HeN male mice. Median growth
delay of WiDr-NTRneo tumors. c Combined data from 12 individual experiments. This compound was used as the standard for comparison
of activity of the analogues.
maximum tolerated dose in C3H/HeN mice (Table 5).
The increased solubility in comparison to 1 enabled
improvements in formulation of compounds for in vivo
administration so that compound 2 and 15 could be
administered in saline or water, with 3 and 16 formu-
lated in acetate buffer at pH 4. Details of the vehicles
for analogues tested in vivo are shown in Table 5. The
vehicle chosen for CB 1954 was as reported previously45
with addition of PEG 400 (Table 5).
Compound 2 showed activity similar to that of 1
against WiDr-NTRneo tumors, with a median time to the
tumor endpoint (13 mm diameter) of >100 days (>86
and >87 day growth delay relative to vehicle-treated
controls) with 2/7 animals free of palpable tumor at the
100 day termination. The class II regioisomer 14 and
the class I morpholide 3 also showed considerable
antitumor activity (median growth delays of 47 and 35
days) but were less active than 1. Compound 15, the

3302 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
F igu r e 3. In vivo evaluation of selected aziridinyldinitrobenzamides in mixed WiDr/WiDr-NTRneo (50:50%) tumors. Growth delay
of established (ca. 300 mg) tumors following a single ip dose of 1 at 200 µmol/kg in DMA/PEG/water (1:4:5 v/v/v) or 2 at 1000
µmol/kg in saline in CD-1 nude mice. Filled symbols are vehicle-only controls.
class II regioisomer of 2, and the 2-SO2Me compound
10 had poor in vivo activity.
(e.g., 1 and 14) are very selective. The sole example of
class III (17) showed significant but low selectivity,
while the only example of class IV (18) was too insoluble
for proper evaluation. Thus, we consider we have
delineated the regioisomers of the aziridinyldinitrobenz-
amides that are of greatest interest as prodrugs for
NTR/GDEPT.
To assess bystander activity in vivo, 1 and 2 were also
tested for activity against mixed WiDr/WiDr-NTRneo (50:
50% or 90:10%) tumors (Figure 3). Compound 2, which
had the highest MTD (1000 µmol/kg), again provided a
large median tumor growth delay (59 days versus >85
days for 1) in tumors with 50% WiDr-NTRneo cells, with
some long-term survivors (1/6 versus 16/31 animals).
However, lower expression of NTR (10% WiDr-NTRneo
cells) in the tumors resulted in 6 and 17 day growth
delay. These data indicate 2 to be only slightly less
active than 1 in NTR-expressing WiDr tumors. Com-
pounds 2, 3, 10, and 15 also had limited in vivo
bystander activity in tumors with low expression of NTR
(EMT6/EMT6-NTRpuro, 95:5%). Tumor cell kill was
determined by clonogenic assay following excision of the
tumor 18 h after drug treatment. Only compound 1 was
active against the activator cells EMT6-NTRpuro (greater
than 2 logs of cell kill compared with less than 1 log of
cell kill for 2, 3, 10, and 15), and only 1 showed
bystander activity against target EMT6 cells (<1 log cell
kill, range 0.06-0.53; data not shown).
Within the class I analogues, considerable structural
change can be accommodated while retaining selectivity
of activation to cytotoxic metabolites in NTR-transfected
tumor cells in vitro. The resulting analogues have
improved selectivity for NTR over DT-diaphorase rela-
tive to 1, including examples with higher aqueous
solubility. However, those analogues that appear to be
the most promising in relation to these parameters (e.g.
2, 3) have less efficient bystander effects in MCLs
probably because their metabolites are more hydro-
philic. They also lack bystander effects in vivo at low
activator cell density (EMT6/EMT6-NTRpuro 95:5% mixed
tumors). All the compounds tested in vivo except for 2
had inferior activity compared with 1 against WiDr-
NTRneo tumors. The diol analogue 2, which had the
highest MTD, however provided a substantial tumor
growth delay (59 days versus >85 days for 1) against
mixed WiDr/WiDr-NTRneo (50:50%) tumors with some
long-term survivors. This compound is activated simi-
larly to 1 by NTR, with the generation of two hydroxyl-
amine metabolites, but is much more soluble (and can
be formulated in saline for administration to mice).
Thus, the formulation problems seen with CB 1954 (1)
can potentially be avoided by use of carefully selected
analogues. The somewhat weaker bystander effect of 2
relative to 1 may be an advantage for its use in selective
cell ablation therapy of NTR-tagged cells in normal
Con clu sion s
While much work has been carried out with the
5-aziridinyl-2,4-dinitrobenzamide CB 1954 (1), there has
been little study of the analogues. While there are 16
possible regioisomers of the parent molecule, our previ-
ous work34,46 with related dinitrobenzamide mustards
suggested to us that only compounds with nitro groups
meta to each other would have reduction potentials in
the appropriate range. In addition, examination of the
NTR crystal structure suggests that, of the remaining
regioisomers, only those that also have the two nitro
groups ortho and para to the aziridine have the ap-
propriate cross-sectional area to allow binding suf-
ficiently far into the groove to allow reduction of the
critical 4-nitro group. Only 3 of the 16 regioisomers meet
both these criteria, and we were able to prepare
representatives of these (formulas I-III in Table 1),
together with a representative of the regioisomer class
IV where the 2-nitro group and carboxamide are
switched. As predicted, compounds of classes I and II
17-24
tissues
Exp er im en ta l Section
Ch em istr y. Analyses were carried out in the Microchemical
Laboratory, University of Otago, Dunedin, New Zealand.
Melting points were recorded on an Electrothermal melting
point apparatus. NMR spectra were obtained on a Bruker
1
DRX-400 spectrometer (400 MHz for H and 100 MHz for 13C)
and are referenced to Me4Si. Column chromatography was
carried out on Merck 300-400 mesh silica gel (flash) or
alumina. Petroleum ether refers to the fraction boiling at 40-

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3303
60 °C, and hexanes refer to the fraction boiling at 60-65 °C.
N,N-Dimethylacetamide (DMA) and DMF were dried over
molecular sieves. THF and Et2O were dried over sodium/
benzophenone.
2.39, 2-27-2.21 (2 m, 10 H, CH2N(CH2)CH2, aziridine-H).
Anal. (C16H21N5O6) C, H. HRMS (EI) C16H21N5O6 requires M+
379.1492. Found 379.1496.
5-(Azir id in -1-yl)-N-[3-(4-m or p h olin o)p r op yl]-2,4-d in i-
tr oben za m id e (5). Similar reaction of 21 in Et2O with 4-(3-
aminopropyl)morpholine (2 equiv) in water gave 5-chloro-N-
[3-(4-morpholino)propyl]-2,4-dinitrobenzamide (25) (81%): mp
5-(Azir id in -1-yl)-N-(2,3-d ih yd r oxyp r op yl)-2,4-d in itr o-
ben za m id e (2). A solution of N-(2,3-dihydroxypropyl)-5-
chloro-2,4-dinitrobenzamide46 (22) (0.50 g, 1.56 mmol) and
aziridine (1.00 g, 23 mmol) in EtOAc (100 mL) was stirred at
room temperature for 18 h. After being washed with water,
the solution was dried and concentrated under reduced pres-
sure to give 2 (0.43 g, 84%): mp (EtOAc/petroleum ether) 145-
1
(CH2Cl2/petroleum ether) 167-168 °C; H NMR [(CD3)2SO] δ
8.85 (t, J ) 5.5 Hz, 1 H, NH), 8.83 (s, 1 H, H-3), 8.13 (s, 1 H,
H-6), 3.57 (t, J ) 4.6 Hz, 4 H, CH2OCH2), 3.32-3.23 (m, 2 H,
NHCH2), 2.24-2.29 (m, 6 H, CH2N(CH2)CH2), 1.67 (pent, J )
7.0 Hz, 2 H, CH2CH2CH2). Anal. C14H17ClN4O6) C, H, N.
1
147 °C; H NMR [(CD3)2SO] δ 8.74 (t, J ) 5.7 Hz, 1 H, NH),
8.64 (s, 1 H, H-3), 7.43 (s, 1 H, H-6), 4.85 (d, J ) 4.9 Hz, 1 H,
OH), 4.58 (t, J ) 5.7 Hz, 1 H, OH), 3.62 (m, 1 H, CHOH), 3.44-
3.36 (m, 2 H, CH2OH), 3.39-3.36 (m, 1 H, CONHCHH), 3.16-
3.09 (m, 1 H, CONHCHH), 2.48 (s, 4 H, aziridine-H); 13C NMR
δ 164.19 (s), 152.98 (s), 139.70 (s), 138.89 (s), 137.53 (s), 124.57
(d), 122.58 (d), 70.02 (d), 63.70 (t), 42.73 (t), 29.91 (t). Anal.
(C12H14N4O7) C, H, N.
Reaction of 25 (200 mg, 0.54 mmol) in CH2Cl2 (10 mL) with
aziridine (112 µL, 2.16 mmol) for 3 h at 20 °C, followed by
partition between more CH2Cl2 and water, gave 5 (114 mg,
56%): mp (CH2Cl2/EtOAc/petroleum ether) 151-152 °C; 1H
NMR [(CD3)2SO] δ 8.71 (t, J ) 5.4 Hz, 1 H, NH), 8.65 (s, 1 H,
H-3), 7.41 (s, 1 H, H-6), 3.57 (t, J ) 4.5 Hz, 4 H, CH2OCH2),
3.30-3.23 (m, 2 H, NHCH2), 2.48 (s, 4 H, aziridine-H), 2.41-
2.30 (m, 6 H, CH2N(CH2)CH2), 1.68 (pent, J ) 7.0 Hz, 2 H,
CH2CH2CH2). Anal. (C16H21N5O6) C, H, N.
5-(Azir id in -1-yl)-N-[2-(4-m or p h olin o)et h yl]-2,4-d in i-
tr oben za m id e (3). A mixture of 5-chloro-2,4-dinitrobenzoic
acid (20) (2.00 g, 8.11 mmol) and SOCl2 (30 mL) containing
DMF (2 drops) was refluxed under nitrogen for 2 h before
concentration to dryness. The resulting crude 5-chloro-2,4-
dinitrobenzoyl chloride (21) was dissolved in dry Et2O (100
mL), and the solution was cooled to 0 °C and treated in one
portion with a solution of 4-(2-aminoethyl)morpholine (1.96 g,
15 mmol) in Et2O (20 mL). After the mixture was stirred at
this temperature for 15 min, the resultant solid was filtered
off, dissolved in water (50 mL), and treated with an excess of
saturated aqueous NaHCO3. The mixture was extracted with
EtOAc and the extract was worked up to give 5-chloro-N-[2-
(4-morpholino)ethyl]-2,4-dinitrobenzamide (23) (1.44 g, 49%):
mp (EtOAc/petroleum ether) 124 °C (dec); 1H NMR [(CD3)2-
SO] δ 8.83 (t, J ) 5.4 Hz, 1 H, NH), 8.83 (s, 1 H, H-3), 8.10 (s,
1 H, H-6), 3.58 (t, J ) 4.55 Hz, 4 H, CH2O), 3.39-3.30 (m, 2
H, CONHCH2), 2.47 (t, J ) 6.7 Hz, 2 H, CH2Nmorph), 2.41
(br s, 4 H, NCH2); 13C NMR δ 162.58 (s), 147.09 (s), 144.99 (s),
136.18 (s), 132.33 (d), 130.33 (s), 122.11 (d), 66.09 (t), 56.61
(t), 53.13 (t), 36.40 (t). Anal. (C13H15ClN4O6) C, H, N.
A solution of 23 (0.50 g, 1.39 mmol) and aziridine (1.00 g,
23 mmol) in EtOAc (80 mL) was stirred at room temperature
for 18 h. After being washed with water, the solution was dried
over Na2SO4 and concentrated under reduced pressure to ca.
20 mL. Petroleum ether was added until a slight cloudiness
persisted and the solution was chilled at -20 °C to give 3 as
coarse yellow needles (0.37 g, 73%): mp 159 °C; 1H NMR
[(CD3)2SO] δ 8.70 (t, J ) 5.6 Hz, 1 H, NH), 8.66 (s, 1 H, H-3),
7.40 (s, 1 H, H-6), 3.57 (t, J ) 4.6 Hz, 4 H, CH2(CH2)O), 3.40-
3.29 (m, 2 H, NHCH2), 2.50-2.37 (m, 10 H, CH2N(CH2)CH2,
aziridine-H); 13C NMR δ 163.97 (s), 153.03 (s), 139.72 (s),
138.82 (s), 137.49 (s), 124.42 (d), 122.68 (d), 66.12 (t), 56.68
(t), 53.17 (t), 36.40 (t), 29.90 (t). Anal. (C15H19N5O6) C, H, N.
5-(Azir id in -1-yl)-N-m eth yl-N-[2-(4-m or p h olin o)eth yl]-
2,4-d in itr oben za m id e (4). Similar reaction of 21 in Et2O
with 4-[2-(methylamino)ethyl]morpholine (2 equiv) in water,
followed by chromatography of the product on alumina-90,
eluting with EtOAc, gave 5-chloro-N-methyl-N-[2-(4-morpholi-
no)ethyl]-2,4-dinitrobenzamide (24) (48%): mp (EtOAc/iPr2O)
123-123.5 °C; 1H NMR [(CD3)2SO] (mixture of rotamers) δ
8.94, 8.93 (2 s, 1 H, H-3), 8.12, 8.08 (2 s, 1 H, H-6), 3.62-3.55,
3.53-3.48, 3.27-3.19 (3 m, 6 H, CH2(CH2)O, CONCH2), 3.05,
2.89 (2 s, 3 H, CH3), 2.61-2.54, 2.49-2.39, 2.28-2.22 (3 m, 6
H, CH2N(CH2CH2)). Anal. (C14H17ClN4O6) C, H, N.
A stirred solution of 24 (200 mg, 0.54 mmol) in CH2Cl2 (15
mL) was treated with aziridine (112 µL, 2.16 mmol) at room
temperature for 3 h. After being washed with water (2×), the
solution was dried and evaporated under reduced pressure.
The residue was dissolved in EtOAc, filtered through a column
of alumina-90, and then diluted with petroleum ether to
precipitate 4 (147 mg, 72%) as an unstable gum: 1H NMR
[(CD3)2SO] (mixture of rotamers) δ 8.77, 8.76 (2 s, 1 H, H-3),
7.43, 7.35 (2 s, 1 H, H-6), 3.63-3.55, 3.52-3.47, 3.24-3.17 (3
m, 6 H, CH2(CH2)O, CONCH2), 3.05, 2.85 (2 s, 3 H, CH3), 2.62-
5-(Azir id in -1-yl)-N-[4-(4-m or p h olin o)bu t yl]- 2,4-d in i-
tr oben za m id e (6). Similar reaction of 21 in Et2O with 4-(4-
aminobutyl)morpholine (2 equiv) in water, followed by chro-
matography of the product on alumina-90 and elution with
CH2Cl2/EtOAc (1:3), gave 5-chloro-N-[4-(4-morpholino)butyl]-
2,4-dinitrobenzamide (26) (58%): mp (EtOAc/iPr2O) 116-117
1
°C; H NMR [(CD3)2SO] δ 8.84 (t, J ) 5.5 Hz, 1 H, NH), 8.83
(s, 1 H, H-3), 8.12 (s, 1 H, H-6), 3.56 (t, J ) 4.5 Hz, 4 H, (CH2)-
CH2O), 3.25 (q, J ) 6. 0 Hz, 2 H, NHCH2), 2.33 (br s, 4 H,
N(CH2)CH2), 2.28 (t, J ) 6.6 Hz, 2 H, CH2Nmorph), 1.44 (m,
4 H, NHCH2(CH2)2). Anal. (C15H19NClN4O6) C, H, N.
Reaction of 26 with aziridine in CH2Cl2 as above, followed
by chromatography of the product on alumina-90 and elution
with EtOAc, gave 6 (62%): mp (EtOAc/petroleum ether) 118-
1
122 °C; H NMR [(CD3)2SO] δ 8.71 (t, J ) 5.5 Hz, 1 H, NH),
8.65 (s, 1 H, H-3), 7.40 (s, 1 H, H-6), 3.56 (t, J ) 4.6 Hz, 4 H,
(CH2OCH2), 3.24 (q, J ) 6.0 Hz, 2 H, NHCH2), 2.48 (s, 4 H,
aziridine-H), 2.34 (br s, 4 H, N(CH2)CH2), 2.29 (t, J ) 6.8 Hz,
2 H, CH2Nmorph), 1.59-1.45 (m, 4 H, NHCH2(CH2)2). Anal.
(C17H23N5O6) C, H, N.
5-(Azir id in -1-yl)-N-[2-(im id a zol-1-yl)eth yl]-2,4-d in itr o-
ben za m id e (7). Similar reaction of 21 in Et2O with N-[-2-
(aminoethyl)]imidazole (2 equiv) in water, followed by direct
recrystallization of the product from EtOAc and then from
EtOAc/MeOH, gave 5-chloro-N-[2-(imidazol-1-yl)ethyl]-2,4-
1
dinitrobenzamide (27) (49%): mp >300 °C; H NMR [(CD3)2-
SO] δ 9.04 (t, J ) 5.6 Hz, 1 H, NH), 8.44 (s, 1 H, H-3), 8.03 (s,
1 H, H-6), 7.69, 7.24, 6.92 (3 s, 3 H, imidazole-H), 4.15 (t, J )
5.8 Hz, 2 H, NHCH2CH2), 3.57 (q, J ) 5.8 Hz, 2 H, NHCH2).
Anal. (C12H10ClN5O6) C, H, N.
A stirred suspension of 27 (150 mg, 0.44 mmol) in THF (40
mL) was treated with aziridine (91 µL, 1.76 mmol) at room
temperature for 4 h, and then additional aziridine (91 µL) was
added. After a further 4 h, the mixture was concentrated under
reduced pressure below 25 °C, and the residue was partitioned
between EtOAc and saturated NaCl. Evaporation of the
organic layer gave a product that was triturated with EtOAc
and then recrystallized from MeCN/EtOAc/petroleum ether to
give 7 (56 mg, 37%): mp >250 °C; 1H NMR [(CD3)2SO] δ 8.91
(t, J ) 5.6 Hz, 1 H, NH), 8.66 (s, 1 H, H-3), 7.67 (s, 1 H,
imidazole-H), 7.29 (s, 1 H, H-6), 7.24, 6.93 (2 s, 2 H, imidazole-
H), 4.16 (t, J ) 5.9 Hz, 2 H, NHCH2CH2), 3.62-3.51 (m, 2 H,
NHCH2), 2.49 (s, 4 H, aziridine-H). Anal. (C14H14N6O5) C, H,
N.
5-(Azir idin -1-yl)-N-[2-(m eth oxycar bon yl)eth yl]-2,4-din i-
tr oben za m id e (8). Similar reaction of 21 in Et2O with a
vigorously stirred suspension of methyl 3-aminopropanoate
hydrochloride (2 equiv) and Et3N (3 equiv) in Et2O for 30 min
gave 5-chloro-N-[2-(methoxycarbonyl)ethyl]-2,4-dinitrobenz-
amide (28) (47%): mp (EtOAc/petroleum ether) 139-141 °C;
1H NMR [(CD3)2SO] δ 8.98 (t, J ) 5.6 Hz, 1 H, CONH), 8.83
(s, 1 H, H-3), 8.09 (s, 1 H, H-6), 3.63 (s, 3 H, CH3), 3.47 (dt, J

3304 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
) 6.7, 5.6 Hz, 2 H, CONHCH2), 2.61 (t, J ) 6.7 Hz, 2 H, CH2-
CO2CH3). Anal. (C11H10ClN3O7) C, H, N, Cl.
(0.51 g, 81%): mp 219-220 °C; 1H NMR [(CD3)2SO] δ 8.45 (s,
1 H), 8.21 (s, 1 H), 7.87 (s, 1 H), 7.42 (s, 1 H), 3.41 (s, 3 H),
2.47 (s, 4 H); HRMS (EI) calcd for C10H11N3O5S [M+] m/z
285.0419, found 285.0415.
A stirred solution of 28 (315 mg, 0.95 mmol) in CH2Cl2 (35
mL) was treated with aziridine (197 µL, 3.81 mmol) at room
temperature for 8 h. The mixture was partitioned between
more CH2Cl2 and water, and the organic layer was dried and
concentrated under reduced pressure. The residue was dis-
solved in EtOAc and filtered through a short column of silica
gel to give 8 (269 mg, 84%): mp (CH2Cl2/iPr2O) 137 °C; 1H
NMR [(CD3)2SO] δ 8.86 (t, J ) 5.6 Hz, 1 H, NH), 8.66 (s, 1 H,
H-3), 7.40 (s, 1 H, H-6), 3.63 (s, 3 H, Me), 3.51-3.43 (m, 2 H,
NHCH2), 2.61 (t, J ) 6.8 Hz, 2 H, CH2CO), 2.48 (s, 4 H,
aziridine-H). Anal. (C13H14N4O7) C, H, N.
5-(Azir id in -1-yl)-N-(2,3-d ih yd r oxyp r op yl)-2-(m et h yl-
su lfon yl)-4-n itr oben za m id e (11). A mixture of 34 (2.21 g,
7.9 mmol) and SOCl2 (20 mL) containing DMF (2 drops) was
refluxed for 1 h, then evaporated to dryness under reduced
pressure. The resulting crude acid chloride was dissolved in
anhydrous Me2CO (40 mL), and the solution was cooled to -5
°C and treated in one portion with a cold solution of 3-ami-
nopropane-1,2-diol (1.48 g, 1.62 mmol) in water (20 mL). The
mixture was shaken at room temperature for 15 min, then
diluted with water (20 mL) and concentrated under reduced
pressure to ca. 30 mL. After addition of solid NaCl, the mixture
was extracted with EtOAc (3×), and the extracts were worked
up to give a solid. This was dissolved in EtOAc and filtered
through a column of silica gel to give 5-chloro-N-(2,3-dihy-
droxypropyl)-2-(methylsulfonyl)-4-nitrobenzamide (36) (2.54 g,
3-(Azir id in -1-yl)-4-n it r oben za m id e (9). 3-Fluoro-4-ni-
trobenzoic acid47 was converted to the acid chloride (SOCl2/
DMF) and then reacted with concentrated NH4OH in Et2O to
give 3-fluoro-4-nitrobenzamide (29) (92%): mp (EtOAc/petro-
1
leum ether) 156-166 °C; H NMR [(CD3)2SO] δ 8.31 (br s, 1
H, NHH), 8.26 (t, J ) 8.1 Hz, 1 H, H-5), 7.98 (dd, J ) 12.1, 1.7
Hz, 1 H, H-2), 7.89 (dd, J ) 8.6, 1.1 Hz, 1 H, H-6), 7.86 (br s,
1 H, NHH). Anal. (C7H5FN2O3) C, H, N.
1
91%): mp (EtOAc/iPr2O) 141-142 °C; H NMR [(CD3)2SO] δ
8.83 (t, J ) 5.7 Hz, 1 H, NH), 8.55 (s, 1 H, H-3), 8.08 (s, 1 H,
H-6), 4.81 (d, J ) 5.0 Hz, 1 H, CHOH), 4.55 (t, J ) 5.8 Hz, 1
H, CH2OH), 3.70-3.58 (m, 1 H, CHOH), 3.45 (s, 3 H, CH3),
3.44-3.35 (m, 3 H, CH2OH and CONHCHH), 3.17-3.06 (m, 1
H, CONHCHH). Anal. (C11H13ClN2O7S) C, H, N, S.
A stirred solution of 29 (100 mg, 0.54 mmol) in dry MeCN
(4 mL) was treated with aziridine (168 µL, 3.24 mmol) at room
temperature for 4 h. The resulting precipitate was collected,
washed with cold water, and crystallized twice from EtOAc to
give 3-(aziridin-1-yl)-4-nitrobenzamide (9) (46 mg, 41%): mp
223-224 °C; 1H NMR [(CD3)2SO] δ 8.21 (br s, 1 H, NHH), 7.99
(d, J ) 8.5 Hz, 1 H, H-5), 7.69 (d, J ) 1.8 Hz, 1 H, H-2), 7.67
(br s, 1 H, NHH), 7.56 (dd, J ) 8.5, 1.8 Hz, 1 H, H-6), 2.30 (s,
4 H, aziridine-H). Anal. (C9H9N3O3) C, H, N.
5-(Azir id in -1-yl)-2-(m et h ylsu lfon yl)-4-n it r ob en za m -
id e (10) (Sch em e 3). A mixture of 5-chloro-2-(methylsulfanyl)-
benzoic acid48 (32) 2.03 g, 0.01 mol), 30% H2O2 (4.52 mL, 0.04
mol), and AcOH (10 mL) was stirred at 80 °C for 3 h, then
treated with additional 30% H2O2 (2.26 mL, 0.02 mol) and
heated at 80 °C for a further 3 h. The mixture was cooled,
stirred with Pt for 1 h, then filtered. The solution was
concentrated under reduced pressure and the residue was
crystallized from benzene to give 5-chloro-2-(methylsulfonyl)-
benzoic acid (33) (1.81 g, 77%): mp 155-160 °C; 1H NMR
(CDCl3) δ 8.11 (d, J ) 8.4 Hz, 1 H), 7.84 (d, J ) 2.2 Hz, 1 H),
7.67 (dd, J ) 8.5, 2.2 Hz, 1 H), 3.40 (s, 3 H). CO2H not observed.
Anal. (C8H7ClO4S) C, H, N.
A solution of 36 (0.84 g, 2.38 mmol) and aziridine (0.74 mL,
14.3 mmol) in EtOAc (20 mL) was stirred at room temperature
for 16 h. The reaction mixture was worked up and the residue
was chromatographed on silica gel, eluting with EtOAc/MeOH
(4:1) to give 11 (0.65 g, 76%): mp (EtOAc) 120-123 °C; 1H
NMR [(CD3)2SO] δ 8.71 (t, J ) 5.8 Hz, 1 H, NH), 8.45 (s, 1 H,
H-3), 7.44 (s, 1 H, H-6), 4.77 (d, J ) 5.0 Hz, 1 H, CHOH), 4.54
(t, J ) 5.8 Hz, 1 H, CH2OH), 3.70-3.61 (m, 1 H, CHOH), 3.45-
3.36 (m, 3 H, CH2OH, CONHCHH), 3.41 (s, 3 H, CH3), 3.17-
3.07 (m, 1 H, CONHCHH), 2.47 (s, 4 H, aziridine-H). Anal.
(C13H17N3O7S‚0.5H2O) C, H, N.
5-(Azir id in -1-yl)-2-n itr oben za m id e (12). A stirred solu-
tion of 2-nitro-5-fluorobenzamide49 (30) (480 mg, 2.61 mmol)
in dry DMSO (2 mL) was treated with aziridine (0.54 mL, 10.4
mmol) at room temperature under N2 for 48 h, then poured
into cold water (50 mL). Prolonged cooling at -5 °C gave a
crystalline product that was collected and filtered through a
column of silica gel in EtOAc to give 12 (256 mg, 47%): mp
1
(EtOAc) 172-174 °C; H NMR [(CD3)2SO] δ 8.01, 7.61 (2 s, 2
A solution of 33 (3.98 g, 17 mmol) in concentrated H2SO4
(18 mL) was treated with HNO3 (d ) 1.52, 3.9 mL, 94 mmol)
and heated at 100 °C for 1 h. The cooled mixture was poured
into a limited amount of crushed ice, and following prolonged
cooling, the separated solid was collected and washed with a
little ice-cold water. Crystallization from EtOAc/petroleum
ether gave 5-chloro-2-(methylsulfonyl)-4-nitrobenzoic acid (34)
H, CONH2), 7.93 (d, J ) 8.9 Hz, 1 H, H-3), 7.17 (dd, J ) 8.8,
2.4 Hz, 1 H, H-4), 7.09 (d, J ) 2.3 Hz, 1 H, H-6), 2.24 (s, 4 H,
aziridine-H). Anal. (C9H9N3O3) C, H, N.
5-(Azir id in -1-yl)-4-(m et h ylsu lfon yl)-2-n it r ob en za m -
id e (13). A stirred suspension of 5-fluoro-4-(methysulfonyl)-
2-nitrobenzamide34 (31) (262 mg, 1.00 mmol) in dry dioxane
(15 mL) was treated with aziridine (207 µL, 4.00 mmol) at 35
°C for 1 h. The solvent was then removed under reduced
pressure, and the residue was shaken with water and recrys-
tallized from MeOH to give 13 (217 mg, 76%): mp 222-224
1
(2.88 g, 61%): mp 222-223 °C; H NMR [(CD3)2SO] δ 14.3 (v
br, 1 H), 8.59 (s, 1 H), 8.23 (s, 1 H), 3.46 (s, 3 H). Anal. (C8H6-
ClNO6S) C, H, N.
A suspension of 34 (2.68 g, 4.58 mmol) in SOCl2 (40 mL)
containing DMF (1 drop) was heated under reflux for 1 h and
then concentrated under reduced pressure and re-evaporated
after the addition of benzene. The residue was dissolved in
acetone (40 mL) and treated at 0 °C with cold concentrated
aqueous NH3 (10 mL). The mixture was shaken at room
temperature for 5 min, then concentrated to a small volume
under reduced pressure below 35 °C and diluted with water.
The solid was collected and washed with water to give 5-chloro-
2-(methylsulfonyl)-4-nitrobenzamide (35) (2.42 g, 91%): mp
1
°C; H NMR [(CD3)2SO] δ 8.39 (s, 1 H, H-3), 8.14 and 7.86 (2
× br s, 2 H, CONH2), 7.33 (s, 1 H, H-6), 3.41 (s, 3 H, CH3),
2.57 (s, 4 H, aziridine-H). Anal. (C10H11N3O5S) C, H, N.
2-(Azir id in -1-yl)-3,5-d in itr oben za m id e (14). Reaction of
2-chloro-3,5-dinitrobenzamide36 (37) in EtOAc with excess
aziridine as above, but for only 2 h, gave 14 (64%): mp (EtOAc/
1
petroleum ether) 200 °C; H NMR [(CD3)2SO] δ 8.74 (d, J )
2.6 Hz, 1 H, H-4), 8.35 (d, J ) 2.6 Hz, 1 H, H-6), 8.10, 7.93 (2
br s, 2 H, CONH2), 2.40 (s, 4 H, aziridine-H); 13C NMR δ 165.76
(s), 151.10 (s), 141.70 (s), 139.07 (s), 132.89 (s), 126.92 (d),
121.96 (d), 30.60 (t). Anal. (C9H8N4O5) C, H, N.
1
(EtOAc/petroleum ether) 255-256 °C; H NMR [(CD3)2SO] δ
8.55 (s, 1 H), 8.31 (s, 1 H), 8.06 (s, 1 H), 8.01 (s, 1 H), 3.46 (s,
3 H). Anal. (C8H7ClN2O5S) C, H, N).
2-(Azir id in -1-yl)-N-(2,3-d ih yd r oxyp r op yl)-3,5-d in itr o-
ben za m id e (15). A stirred solution of 2-chloro-N-(2,3-dihy-
droxypropyl)-3,5-dinitrobenzamide36 (38) (670 mg, 2.10 mmol)
in EtOAc (35 mL) was treated with aziridine (325 µL, 6.28
mmol) at room temperature for 3 h, then concentrated under
reduced pressure. The residue was dissolved in warm EtOAc
(220 mL) and filtered though a short column of silica gel to
give 15 (304 mg, 44%): mp (EtOAc/petroleum ether) 154-155
°C; 1H NMR [(CD3)2SO] δ 8.74 (d, J ) 2.7 Hz, 1 H, H-4), 8.57
A solution of 35 (0.70 g, 2.51 mmol) in EtOAc/MeCN (1:1,
25 mL) was treated with aziridine (0.52 mL, 10 mmol) and
stirred at room temperature for 4 h. Additional aziridine (0.26
mL, 5 mmol) was added, and the mixture was stirred for a
further 2 h and then concentrated under reduced pressure.
The residue was shaken with aqueous KHCO3, then dissolved
in warm EtOAc and filtered through a short column of silica
gel. The eluate was concentrated to a small volume to give 10

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3305
(t, J ) 5.6 Hz, 1 H, NH), 8.36 (d, J ) 2.7 Hz, 1 H, H-6), 4.88
(d, J ) 5.3 Hz, 1 H, CHOH), 4.64 (t, J ) 5.7 Hz, 1 H, CH2OH),
3.73-3.63 (m, 1 H, CHOH), 3.52-3.42 (m, 1 H, NHCHH),
3.42-3.34 (m, 2 H, CH2OH), 3.26-3.15 (m, 1 H, NHCHH), 2.39
(s, 4 H, aziridine-H). Anal. (C12H14N4O7) C, H, N.
varied from 50 to 125 V. The parent molecular ion and
hydroxylamine were both detected as [M - H]- ions. The LC
column was an Alltima C8 5µ (3.6 mm × 150 mm). The mobile
phase was (A) 80% MeCN (in water) and (B) water at 0.5 mL/
min, using the following gradient (A/B %) and run time: 0-2
min 5:95; 2-15 min 70:30; 15-18 min 5:95; a run time of 20
min with diode array detection (253 nm).
2-(Azir id in -1-yl)-N-[2-(4-m or p h olin o)et h yl]-3,5-d in i-
tr oben za m id e (16). A solution of 2-chloro-3,5-dinitrobenzoyl
chloride36 (2.50 g, 9.43 mmol) in dry Et2O (30 mL) was cooled
to -5 °C and treated in one portion with a cold solution of
4-(2-aminoethyl)morpholine (2.58 g, 19.8 mmol) in water (30
mL). After vigorous shaking at 10 °C for 5 min, the resultant
solid was collected and crystallized from EtOAc to give
2-chloro-N-[2-(4-morpholino)ethyl]-3,5-dinitrobenzamide (39)
(3.01 g, 89%): mp 188-190 °C; 1H NMR [(CD3)2SO] δ 9.00 (d,
J ) 2.6 Hz, 1 H, H-4), 8.00 (t, J ) 5.3 Hz, 1 H, NH), 8.51 (d,
J ) 2.6 Hz, 1 H, H-6), 3.58 (t, J ) 4.5 Hz, 4 H, O(CH2)2), 3.45-
3.36 (m, 2 H, NHCH2), 2.53-2.46 (m, partially obscured, 2 H,
Cell Lin es a n d P r olifer a tion Assa ys. Compounds were
evaluated for cytotoxic activation by NTR by comparing IC50
values following an 18 h drug exposure in four pairs of NTR-
expressing and NTR-ve cell lines. The Chinese hamster
fibroblast NTR-ve line (V79puro) is a V79 Chinese hamster
fibroblast transfected with an empty vector, and the corre-
sponding stable NTR transfectant is V79-NTRpuro; these lines
were originally named T78-1 and T79-A3, respectively.46 WiDr-
NTRneo (originally named WC14) is an NTR transfectant of
the parental human colon carcinoma line WiDr,13 Skov3-
NTRneo (originally named SC313) is from the human ovarian
carcinoma line Skov3, and EMT6-NTRpuro is from EMT6.53 The
log-phase cells were seeded in 96-well plates in 50 µL RMEM
containing 5% fetal bovine serum (FBS, 50-800 cells/well).
After growth for 18 h, prodrugs were added from stock
solutions in DMSO, which were diluted into culture medium
immediately prior to addition to the cell cultures. Prodrug
exposure was terminated by washing with fresh medium after
18 h, and cultures were grown for a further 3 days (rodent
lines) or 4 days (human lines) before staining with sulfo-
rhodamine B54 to assess cell density. The IC50 was determined
as the interpolated drug concentration required to reduce
absorbance to 50% of that of controls on the same plate.
NHCH2CH2), 2.42 (br s, 4 H, N(CH2)CH2). Anal. (C13H15
-
ClN4O6) C, H, N.
A stirred solution of 39 (660 mg, 1.84 mmol) in CH2Cl2 (80
mL) was treated with aziridine (286 µL, 5.52 mmol) at room
temperature for 3 h. The solution was then washed with water
(2×), dried, and evaporated, and the residue was crystallized
(2×) from CH2Cl2/iPr2O to give 16 (548 mg, 82%): mp 176-
1
177 °C; H NMR [(CD3)2SO] δ 8.75 (d, J ) 2.7 Hz, 1 H, H-4),
8.58 (t, J ) 5.5 Hz, 1 H, NH), 8.34 (d, J ) 2.7 Hz, 1 H, H-6),
3.58 (t, J ) 4.5 Hz, 4 H, CH2OCH2), 3.43 (q, J ) 6.2 Hz, 2 H,
NHCH2), 2.54-2.48 (m, partially obscured, 2 H, NHCH2CH2),
2.46-2.36 (m, 8 H, N(CH2)CH2, aziridine-H). Anal. (C15H19N5O6)
H, N. C, found, 48.8; calcd, 49.3%.
3-(Azir id in -1-yl)-2,6-d in itr oben za m id e (17). A solution
of 3-chloro-2,6-dinitrobenzamide50 (40) (0.50 g, 2.04 mmol) and
aziridine (1.00 g, 0.023 mol) in EtOAc (80 mL) was stirred at
room temperature for 18 h. After being washed with water,
the solution was dried over Na2SO4 and concentrated to about
20 mL under reduced pressure. Petroleum ether was added
until a slight cloudiness persisted and the solution was chilled
NTR Exp r ession in Tr a n sfected Cell Lin es. The relative
expression of NTR was determined by immunoblot analysis
of cell lysates (3 µg of protein) and purified NTR (0.1 µg),
electrophoresed on a 10% polyacrylamide gel under reducing
conditions (Mini-Protean-II electrophoresis cell, BioRad). The
proteins were transferred onto PVDF membrane by electro-
phoretic transfer for 1 h (Mini-transblot cell, BioRad). The
membrane was blocked with casein (1%) and then incubated
with a polyclonal sheep antibody for 30 min (1:4000 dilution;
Cobra Therapeutics, U.K.). Following a series of washes with
phosphate-buffered saline containing Tween 20 (0.1%), the
secondary antibody (biotinylated antisheep IgG, Vecta-Stain
Elite ABC kit, Vector Laboratories) was incubated with the
immobilized proteins (1 h, room temperature). Following a
series of washes, the avidin/biotinylated horseradish peroxi-
dase complex was added and detection was through enhanced
chemiluminescence with luminol as a substrate (ECL Western
blotting detection reagent; Amersham, U.K.).
NTR Activity in Tr a n sfected Cell Lin es a n d Kin etics
of CB 1954 Meta bolism by In ta ct Cells. The relative
activity of NTR in the transfected cell lines was determined
by assay of the selective NTR substrate 6-bis(2-bromoethyl)-
amino-3,5-dinitrobenzamide.55 Enzymatic activity was deter-
mined from the kinetics of loss of the substrate (100 µM) in
stirred single-cell suspensions (5 × 105 cells/mL in RMEM, gas
phase 5% CO2/air) at 37 °C. Samples of the culture media were
removed at intervals and centrifuged. The extracellular me-
dium was analyzed by HPLC using an Altima C8 5µ 2.1 mm
× 150 mm column (Alltech Associates, Inc.) and a mobile phase
of 80% acetonitrile in water (A) and formate buffer (0.45 M,
pH 4.5) (B) with a gradient of 10% A for 0-8 min, 10-90% A
between 8 and 13 min, 90% A for 13-18 min, and 90-10% A
between 18 and 23 min. The first-order rate constant for
metabolism of the NTR substrate (kmet) was fitted and cor-
rected to 105 cells/mL. The rate of metabolism of CB 1954
across the cell line panel was determined similarly, as reported
previously.31
1
at -20 °C to give 17 (68%): mp 206-210 °C (dec); H NMR
[(CD3)2SO] δ 8.26 (br s, 1 H, CONH), 8.24 (d, J ) 9.0 Hz, 1 H,
H-5), 7.92 (br s, 1 H, CONH), 7.44 (d, J ) 9.0 Hz, 1 H, H-4),
2.35 (s, 4 H, aziridine-H); 13C NMR δ 162.67 (s), 150.63 (s),
141.82 (s), 139.32 (s), 128.17 (s), 127.77 (d), 123.31 (d), 28.41
(t). Anal. (C9H8N4O5) C, H, N.
4-(Azir id in -1-yl)-3,5-d in itr oben za m id e (18). Reaction of
4-chloro-3,5-dinitrobenzamide51 (41) with aziridine in EtOAc
as above for 2 h gave 18 (57%): mp (THF/petroleum ether)
1
228-231 °C; H NMR [(CD3)2SO] δ 8.68 (s, 2 H, H-2,6), 8.33,
7.75 (2 br s, 2 H, CONH2), 2.36 (s, 4 H, aziridine-H); 13NMR δ
163.77 (s), 145.16 (s), 143.40 (s), 128.43 (d), 125.89 (s), 30.61
(t). Anal. (C9H8N4O5) C, H, N.
Red u ction P oten tia ls. Pulse radiolysis experiments were
carried out on a 1.8 MV Linac, delivering ca. 3 Gy in 0.2 µs to
a 2 cm path length cell containing the compounds as 3 mM
solutions in phosphate buffer at pH 7, using either 2-propanol
(typically 0.2 M) or 2-propanol/acetone mixtures as cosolvent,
as reported previously,52 with methyl viologen as a redox
indicator.
An a lysis of Hyd r oxyla m in e Meta bolites of Azir id in yl-
d in itr oben za m id es. Aziridinyldinitrobenzamides (200 µM)
were incubated at 25 °C with NADPH generating system
(glucose 6-phosphate, 2 µmol; MgCl2, 2 µmol; NADP, 0.2 µmol;
glucose 6-phosphate dehydrogenase, 0.4 units in 100 µL) in a
final volume of 1 mL of 10 mM phosphate buffer, pH 7. The
reaction was initiated by addition of purified NTR, purchased
from Biotherapeutic Product Development, CAMR, Porton
Down, Salisbury, U.K. (0.5 µg). After 30 min an aliquot (100
µL) was analyzed by LC/MS, using a single-stage quadrupole
Agilent LC/MSD (series A) coupled to an HP 1100 series binary
pump, vacuum degasser, and autosampler. Ionization was by
negative mode atmospheric pressure chemical ionization (APCI),
and the MS detector conditions were the following: capillary
voltage 4000 V, nebulizer pressure 55 psig, drying gas flow 5
L/min at 350 °C, corona current 5 µA, and vaporizer temp 350
°C. The fragmentation voltage (optimized for each compound)
Mu lticellu la r La yer (MCL) Assa ys. MCL cocultures were
grown as described previously.15,53 Briefly, Teflon microporous
membranes of Millicell CM cell culture inserts (Millipore,
Bedford, MA) were coated with collagen to facilitate cell
attachment and seeded with trypsinized single cell suspensions
(106 cells/0.5 mL R-MEM with 5% FBS) of 3% V79-NTRpuro
(NTR+ve activators) and 97% V79oua (NTR-ve targets) or 10%

3306 J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12
Helsby et al.
WiDr-NTRneo (NTR+ve activators) and 90% WiDr (NTR-ve
targets). MCLs were grown submerged in a stirred reservoir
of R-MEM with 5% FBS under 5% CO2/air (3 days for V79 and
6 days for WiDr MCLs) and then exposed to prodrugs in 10
mL of the same medium for 5 h at 37 °C under 5% CO2/95%
O2 with magnetic stirring. The MCLs were then trypinized,
centrifuged, resuspended in fresh medium, and plated to
determine clonogenic cell survival as described previously.15
Cells were grown in nonselective medium (total cells) or in
medium containing 1 mM ouabain (V79oua cells), 15 µM
puromycin (V79-NTRpuro cells), 0.3 mg/mL G418 (WiDr-NTRneo
cells). After 6 days (V79 lines) or 14 days (WiDr lines), plates
were stained with methylene blue and colonies containing >50
cells were counted.
NAD(P)H dehydrogenase (quinone) (EC 1.6.99.2). Biochem.
Pharmacol. 1988, 37, 4671-4677.
(8) Boland, M. P.; Knox, R. J .; Roberts, J . J . The differences in
kinetics of rat and human DT diaphorase result in a differential
sensitivity of derived cell lines to CB 1954 (5-(aziridin-1-yl)-2,4-
dinitrobenzamide). Biochem. Pharmacol. 1991, 41, 867-875.
(9) Knox, R. J .; J enkins, T. C.; Hobbs, S. M.; Chen, S.; Melton, R.
G.; Burke, P. J . Bioactivation of 5-(aziridin-1-yl)-2,4-dinitrobenz-
amide (CB 1954) by human NAD(P)H quinone oxidoreductase
2: a novel co-substrate-mediated antitumor prodrug therapy.
Cancer Res. 2000, 60, 4179-4186.
(10) Bridgewater, J . A.; Springer, C. J .; Knox, R. J .; Minton, N. P.;
Michael, N. P.; Collins, M. K. Expression of the bacterial
nitroreductase enzyme in mammalian cells renders them selec-
tively sensitive to killing by the prodrug CB1954. Eur. J . Cancer
1995, 31A, 2362-2370.
(11) McNeish, I. A.; Green, N. K.; Gilligan, M. G.; Ford, M. J .;
Mautner, V.; Young, L. S.; Kerr, D. J .; Searle, P. F. Virus directed
enzyme prodrug therapy for ovarian and pancreatic cancer using
retrovirally delivered E. coli nitroreductase and CB1954. Gene
Ther. 1998, 5, 1061-1069.
(12) Green, N. K.; Youngs, D. J .; Neoptolemos, J . P.; Friedlos, F.;
Knox, R. J .; Springer, C. J .; Anlezark, G. M.; Michael, N. P.;
Melton, R. G.; Ford, M. J .; Young, L. S.; Kerr, D. J .; Searle, P.
F. Sensitization of colorectal and pancreatic cancer cell lines to
the prodrug 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB1954) by
retroviral transduction and expression of the E. coli nitroreduc-
tase gene. Cancer Gene Ther. 1997, 4, 229-238.
(13) Friedlos, F.; Court, S.; Ford, M.; Denny, W. A.; Springer, C. Gene-
directed enzyme prodrug therapy: quantitative bystander cy-
totoxicity and DNA damage induced by CB1954 in cells express-
ing bacterial nitroreductase. Gene Ther. 1998, 5, 105-112.
(14) Djeha, A. H.; Hulme, A.; Dexter, M. T.; Mountain, A.; Young, L.
S.; Searle, P. F.; Kerr, D. J .; Wrighton, C. J . Expression of
Escherichia coli B nitroreductase in established human tumor
xenografts in mice results in potent antitumoral and bystander
effects upon systemic administration of the prodrug CB1954.
Cancer Gene Ther. 2000, 7, 721-731.
(15) Wilson, W. R.; Pullen, S. M.; Hogg, A.; Helsby, N. A.; Hicks, K.
O.; Denny, W. A. Quantitation of bystander effects in nitrore-
ductase suicide gene therapy using three-dimensional cell
cultures. Cancer Res. 2002, 62, 1425-1432.
(16) Chung-Faye, G.; Palmer, D.; Anderson, D.; Clark, J .; Downes,
M.; Baddeley, J .; Hussain, S.; Murray, P. I.; Searle, P.; Seymour,
L.; Harris, P. A.; Ferry, D.; Kerr, D. J . Virus-Directed, Enzyme
Prodrug Therapy with Nitroimidazole Reductase: A Phase I and
Pharmacokinetic Study of Its Prodrug, CB1954. Clin. Cancer
Res. 2001, 7, 2662-2668.
(17) Clark, A. J .; Iwobi, M.; Cui, W.; Crompton, M.; Harold, G.; Hobbs,
S.; Kamalati, T.; Knox, R.; Neil, C.; Yull, F.; Gusterson, B.
Selective cell ablation in transgenic mice expression E. coli
nitroreductase. Gene Ther. 1997, 4, 101-110.
(18) Cui, W.; Allen, N. D.; Skynner, M.; Gusterson, B.; Clark, A. J .
Inducible ablation of astrocytes shows that these cells are
required for neuronal survival in the adult brain. Glia 2001, 34,
272-282.
(19) Cui, W.; Gusterson, B.; Clark, A. J . Nitroreductase-mediated cell
ablation is very rapid and mediated by a p53-independent
apoptotic pathway. Gene Ther. 1999, 6, 764-770.
(20) Cui, W.; Gusterson, B. A.; Clark, A. J . Inhibition of myc-
dependent breast tumor formation in transgenic mice. Breast
Cancer Res. Treat. 2002, 71, 9-20.
(21) Gusterson, B.; Cui, W.; Iwobi, M.; Crompton, M. R.; Harold, G.;
Hobbs, S.; Kamalati, T.; Knox, R.; Neil, C.; Yull, F.; Howard,
B.; Clark, A. J . Selective cell ablation in the mammary gland of
transgenic mice. Endocr.-Relat. Cancer 1997, 4, 67-74.
(22) Isles, A. R.; Ma, D.; Milsom, C.; Skynner, M.; Cui, W.; Clark, J .;
Keverne, E. B.; Allen, N. D. Conditional ablation of neurones in
transgenic mice. J . Neurobiol. 2001, 47, 183-193.
(23) Felmer, R.; Cui, W.; Clark, A. J . Inducible ablation of adipocytes
in adult transgenic mice expressing the E. coli nitroreductase
gene. J . Endocrinol. 2002, 175, 487-498.
(24) Ma, D.; Allen, N. D.; Van Bergen, Y. C.; J ones, C. M.; Baum, M.
J .; Keverne, E. B.; Brennan, P. A. Selective ablation of olfactory
receptor neurons without functional impairment of vomeronasal
receptor neurons in OMP-ntr transgenic mice. Eur. J . Neurosci.
2002, 16, 2317-2323.
(25) Knox, R. J .; Friedlos, F.; Sherwood, R. F.; Melton, R. G.;
Anlezark, G. M. The bioactivation of 5-(aziridin-1-yl)-2,4-dini-
trobenzamide (CB1954). II. A comparison of an Eschericia coli
nitroreductase and Walker DT diaphorase. Biochem. Pharmacol.
1992, 44, 2297-2301.
In Vivo Assa ys. Tumors were grown in CD-1 homozygous
nude mice of either sex by injection of 107 cells (grown as
monolayers in culture) sc on the dorsum, using WiDr-NTRneo
,
50:50% or 90:10% mixtures of WiDr/WiDr-NTRneo cell lines.
Mice were individually ear-tagged and randomized to treat-
ment when the mean of the two largest orthogonal tumor
diameters reached 8 mm. Drugs were administered as single
ip doses at the maximum tolerated dose (MTD, defined as the
dose causing no drug-related deaths in a group of six mice with
mean body weight loss at day 5 of <10%) as determined in
separate experiments with non-tumor-bearing C3H/HeN mice.
The injection volume corresponded to 1 µL/g of DMSO or 10
µL/g of aqueous vehicle. For determination of tumor growth
inhibition, the two largest tumor diameters were measured
twice weekly after treatment. The end point was time to
regrowth of the tumors to 15 mm in mean diameter. Animals
with no evidence of tumor 100 days after treatment were
classed as cures.
Ack n ow led gm en t. This work was carried out under
Grant DC1003/0102 from the U.K. Cancer Research
Campaign and Grant 01/276 from the Health Research
Council of New Zealand.
Su p p or tin g In for m a tion Ava ila ble: Growth inhibitory
activity of aziridinyldinitrobenzamides in a cell line overex-
pressing DT-diaphorase (DTD-133) and parental Chinese
hamster cells (CHO-K1), and elemental analysis results of all
new compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
Refer en ces
(1) Greco, O.; Dachs, G. U. Gene directed enzyme/prodrug therapy
of cancer: historical appraisal and future prospectives. J . Cell.
Physiol. 2001, 187, 22-36.
(2) Michael, N. P.; Brehm, J . K.; Anlezark, G. M.; Minton, N. P.
Physical characterisation of the Escherichia coli B gene encoding
nitroreductase and its over-expression in Escherichia coli K12.
FEMS Microbiol. Lett. 1994, 124, 195-202.
(3) Zenno, S.; Koike, H.; Tanokura, M.; Saigo, K. Gene cloning,
purification, and characterization of NfsB, a minor oxygen-
insensitive nitroreductase from Escherichia coli, similar in
biochemical properties to FRase I, the major flavin reductase
in Vibrio fischeri. J . Biochem. (Tokyo) 1996, 120, 736-744.
(4) Anlezark, G. M.; Melton, R. G.; Sherwood, R. F.; Coles, B.;
Friedlos, F.; Knox, R. J . The bioactivation of 5-(aziridin-1-yl)-
2,4-dinitrobenzamide (CB1954). I. Purification and properties
of a nitroreductase enzyme from Escherichia colisa potential
enzyme for antibody-directed enzyme prodrug therapy (ADEPT).
Biochem. Pharmacol. 1992, 44, 2289-2295.
(5) Lovering, A. L.; Hyde, E. I.; Searle, P. F.; White, S. A. The
structure of Escherichia coli nitroreductase complexed with
nicotinic acid: three crystal forms at 1.7, 1.8, and 2.4
Å
resolution. J . Mol. Biol. 2001, 309, 203-213.
(6) Parkinson, G. N.; Skelly, J . V.; Neidle, S. Crystal structure of
FMN-dependent nitroreductase from Escherichia coli B:
a
prodrug-activating enzyme. J Med. Chem. 2000, 43, 3624-3631.
(7) Knox, R. J .; Boland, M. P.; Friedlos, F.; Coles, B.; Southan, C.;
Roberts, J . J . The nitroreductase enzyme in Walker cells that
activates 5-(aziridin-1-yl)-2,4-dinitrobenzamide (CB 1954) to
5-(aziridin-1-yl)-4-hydroxylamino-2-nitrobenzamide is a form of
(26) J ohansson, E.; Parkinson, G.; Denny, W. A.; Neidle, S. Studies
on the Nitroreductase Prodrug-Activating System. Crystal Struc-
tures of Complexes with the Inhibitor Dicoumarol and Dini-
trobenzamide Prodrugs and of the Enzyme Active Form. J . Med.
Chem. 2003, 46, 4009-4020.

Aziridinyldinitrobenzamides
J ournal of Medicinal Chemistry, 2004, Vol. 47, No. 12 3307
(27) Bridgewater, J . A.; Knox, R. J .; Pitts, J . D.; Collins, M. K.;
Springer, C. J . The bystander effect of the nitroreductase/
CB1954 enzyme/prodrug system is due to a cell-permeable
metabolite. Hum. Gene Ther. 1997, 8, 709-717.
(28) Knox, R. J .; Friedlos, F.; J arman, M.; Robert, J . J . A new
cytotoxic, DNA interstrand crosslinking agent, 5-(aziridin-1-yl)-
4-hydrozylamino-2-nitrobenzamide, is formed from 5-(aziridin-
1-yl)-2,4-dinitrobenzamide (CB1954) by nitroreductase enzyme
in Walker carcinoma cells. Biochem. Pharmacol. 1988, 37, 4661.
(29) Knox, R. J .; Friedlos, F.; Marchbank, T.; Roberts, J . J . Bioacti-
vation of CB 1954: reaction of the active 4-hydroxylamino
derivative with thioesters to form the ultimate DNA-DNA
interstrand crosslinking species. Biochem. Pharmacol. 1991, 42,
1691-1697.
(30) Helsby, N. A.; Wheeler, S. J .; Pruijn, F. B.; Palmer, B. D.; Yang,
S.; Denny, W. A.; Wilson, W. R. Effect of Nitroreduction on the
Alkylating Reactivity and Cytotoxicity of the 2,4-Dinitrobenz-
amide-5-aziridine CB 1954 and the Corresponding Nitrogen
Mustard SN 23862: Distinct Mechanisms of Bioreductive Ac-
tivation. Chem. Res. Toxicol. 2003, 16, 469-478.
(31) Helsby, N. A.; Ferry, D. M.; Patterson, A. V.; Pullen, S. M.;
Wilson, W. R. 2-Amino metabolites are key mediators of CB 1954
and SN 23862 bystander effects in nitroreductase GDEPT. Br.
J . Cancer 2004, 90, 1084-1092.
(32) Goldstein, H.; Stamm, R. On 5-chloro-2,4-dinitrobenzoic acid.
Helv. Chim. Acta 1952, 35, 1330-1333.
(33) J ohnson, W. O. Nitro- and trifluoromethyl-substituted diphenyl
ethers. Chem. Abstr. 1981, 94, 15653h.
(34) Atwell, G. J .; Boyd, M.; Palmer, B. D.; Anderson, R. F.; Pullen,
S. M.; Wilson, W. R.; Denny, W. A. Synthesis and evaluation of
4-substituted analogues of 5-[N,N-bis(2-chloroethyl)amino]-2-
nitrobenzamide as bioreductively activated prodrugs using an
Escherichia coli nitroreductase. Anti-Cancer Drug Des. 1996, 11,
553-567.
(35) Somani, H. H.; Wilman, D. E. Novel analogues of CB 1954: Their
potential use in an antibody directed enzyme prodrug therapy
(ADEPT) strategy. Br. J . Cancer 1994, 69, 38.
(36) Palmer, B. D.; Wilson, W. R.; Anderson, R. F.; Boyd, M.; Denny,
W. A. Hypoxia-selective antitumor agents. 14. Synthesis and
hypoxic cell cytotoxicity of regioisomers of the hypoxia-selective
cytotoxin 5-[N,N-bis(2-chloroethyl)amino]-2,4-dinitrobenzamide.
J . Med. Chem. 1996, 39, 2518-2528.
chloroethyl)amino]-2,4-dinitrobenzamide. J . Med. Chem. 1995,
38, 1229-1241.
(42) Skelly, J . V.; Knox, R. J .; J enkins, T. C. Aerobic nitroreduction
by flavoproteins: enzyme structure, mechanisms and role in
cancer chemotherapy. Mini-Rev. Med. Chem. 2001, 1, 293-306.
(43) Ross, D.; Kepa, J . K.; Winski, S. L.; Beall, H. D.; Anwar, A.;
Siegel, D. NAD(P)H:quinone oxidoreductase 1 (NQO1): chemo-
protection, bioactivation, gene regulation and genetic polymor-
phisms. Chem. Biol. Interact. 2000, 129, 77-97.
(44) Belcourt, M. F.; Hodnick, W. F.; Rockwell, S.; Sartorelli, A. C.
Bioactivation of mitomycin antibiotics by aerobic and hypoxic
Chinese hamster ovary cells overexpressing DT-diaphorase.
Biochem. Pharmacol. 1996, 51, 1669-1678.
(45) Cobb, L. M. Toxicity of the selective antitumor agent 5-aziridino-
2,4-dinitrobenzamide in the rat. Toxicol. Appl. Pharmacol. 1970,
17, 231-238.
(46) Friedlos, F.; Denny, W. A.; Palmer, B. D.; Springer, C. J . Mustard
prodrugs for activation by Escherichia coli nitroreductase in
gene-directed enzyme prodrug therapy. J . Med. Chem. 1997, 40,
1270-1275.
(47) Schmelkes, F. C.; Rubin, M. Substituted p-aminobenzoic acids.
J . Am. Chem. Soc. 1944, 66, 1631-1632.
(48) Ruff, F.; Kapovits, I.; Rabai, J .; Kucsman, A. Neighboring group
participation in reactions of sulfides with chloramine-T. Tetra-
hedron 1978, 34, 2767-2773.
(49) Protiva, M.; J ilek, J .; Rajsner, M.; Pomykacek, J .; Ryska, M.;
Holubek, J .; Svatek, E.; Metysova, J . Fluorinated tricyclic
neuroleptics with prolonged action. 7-Fluoro-11-[4-(2-hydroxy-
ethyl)piperazino]-2-isopropyl-10,11-dihydrodibenzo[b,f]thiepin.
Collect. Czech. Chem. Commun. 1986, 51, 698-722.
(50) Denny, W. A.; Atwell, G. J .; Palmer, B. D.; Wilson, W. R.
Nitrophenylaziridine compounds and their uses as prodrugs.
U.S. Patent 6,517,836, 2003
(51) Hostetler, W.; Reinheimer, J . D. Aromatic nucleophilic substitu-
tion. Specific salt effects on rate constants with benzyltrimethyl-
ammonium methoxide in methanol. J . Org. Chem. 1968, 33,
3510-3513.
(52) Denny, W. A.; Atwell, G. J .; Roberts, P. B.; Anderson, R. F.; Boyd,
M.; Lock, C. J .; Wilson, W. R. Hypoxia-selective antitumor
agents. 6. 4-(Alkylamino)nitroquinolines: a new class of hypoxia-
selective cytotoxins. J . Med. Chem. 1992, 35, 4832-4841.
(53) Wilson, W. R.; Pullen, S. M.; Hogg, A.; Hobbs, S. M.; Pruijn, F.
B.; Hicks, K. O. In vitro and in vivo models for evaluation of
GDEPT: quantifying bystander killing in cell cultures and
tumors. In Suicide Gene Therapy: Methods and Reviews;
Springer, C. J ., Ed.; Humana Press: Totowa, NJ , 2003; pp 403-
432.
(37) J arman, M.; Melzack, D. H.; Ross, W. C. J . The metabolism of
the anti-tumour agent 5-(1-aziridinyl)-2,4-dinitrobenzamide (CB
1954). Biochem. Pharmacol. 1976, 25, 2475-2478.
(38) Patel, K. B.; Willson, R. L. Semiquinone free radicals and oxygen.
Pulse radiolysis study of one-electron transfer equilibria. J .
Chem. Soc., Faraday Trans. 1973, 69, 814-819.
(39) Hobbs, S.; J itrapakdee, S.; Wallace, J . C. Development of a
bicistronic vector driven by the human polypeptide chain elonga-
tion factor 1alpha promoter for creation of stable mammalian
cell lines that express very high levels of recombinant proteins.
Biochem. Biophys. Res. Commun. 1998, 252, 368-372.
(40) Morten, J . E.; Margison, G. P. Increased O6-alkylguanine
alkyltransferase activity in Chinese hamster V79 cells following
selection with chloroethylating agents. Carcinogenesis 1988, 9,
45-49.
(54) Skehan, P.; Storeng, R.; Scudiero, D.; Monks, A.; McMahon, J .;
Vistica, D.; Warren, J . T.; Bokesch, H.; Kenney, S.; Boyd, M. R.
New colorimetric cytotoxicity assay for anticancer-drug screen-
ing. J . Natl. Cancer Inst. 1990, 82, 1107-1112.
(55) Pruijn, F. B.; Hogg, A.; Helsby, N. A.; Pullen, S. M.; Hicks, K.
O.; Denny, W. A.; Wilson, W. R. SN 26209, a novel orally active
NTR-GDEPT prodrug: Pharmacokinetics and in vivo activity
against xenografts expressing E. coli nitroreductase. Cancer
Chemother. Pharmacol., submitted.
(41) Palmer, B. D.; van Zijl, P.; Denny, W. A.; Wilson, W. R. Reductive
chemistry of the novel hypoxia-selective cytotoxin 5-[N,N-bis(2-
J M0498699