N-substituted 2-(2,6-dinitrophenylamino)propanamides: Novel prodrugs that release a primary amine via nitroreduction and intramolecular cyclization

Article abstract of DOI:10.1021/jm960783s

A series of N-dinitrophenylamino acid amides [(4-CONHZ-2,6- diNO'2Ph)N(R)C(X,Y)CONHPhOMe] were prepared as potential bioreductive prodrugs and reduced radiolytically to study their rates of subsequent intramolecular cyclization. Compounds bearing a free NH group (R = H) underwent rapid cyclization in neutral aqueous buffers (t(1/2) < 1 min) following 4-electron reduction, with the generation of a N- hydroxydihydroquinoxalinone and concomitant release of 4-methoxyaniline. Amine release from analogous N-methyl analogues (R = Me) was relatively slow. These results are consistent with intramolecular cyclization of a monohydroxylamine intermediate. The high rates of cyclization/extrusion by these very electron-deficient hydroxylamines suggest that the process is greatly accelerated by the presence of an H-bonding 'conformational lock' between the anilino NH group and the adjacent o-nitro group (Kirk and Cohen, 1972). Changes in the phenylcarboxamide side chain or in C-methylation in the linking chain had little effect on the rate of cyclization. The model compounds had 1-electron reduction potentials in the range appropriate for cellular reduction (-373 mV for a measured example) and appeared suitable for development as prodrugs that release amine-based effectors following enzymic or radiolytic reduction. Prodrug examples containing 4-aminoaniline mustard and 5-amino-1-(chloromethyl)benz[e]indoline alkylating units were evaluated but were not activated efficiently by cellular nitroreductases. However, cell killing by the radiation-induced reduction of the latter prodrug was demonstrated.

Full text of DOI:10.1021/jm960783s

346
J . Med. Chem. 1999, 42, 346-355
N-Su bstitu ted 2-(2,6-Din itr op h en yla m in o)p r op a n a m id es: Novel P r od r u gs Th a t
Relea se a P r im a r y Am in e via Nitr or ed u ction a n d In tr a m olecu la r Cycliza tion
Bridget M. Sykes,† Graham J . Atwell,† Alison Hogg,‡ William R. Wilson,‡ Charmian J . O’Connor,§ and
William A. Denny*,†
Auckland Cancer Society Research Centre, Department of Pathology, and Department of Chemistry, The University of
Auckland, Private Bag 92019, Auckland, New Zealand
Received November 12, 1996
A series of N-dinitrophenylamino acid amides [(4-CONHZ-2,6-diNO2Ph)N(R)C(X,Y)CONHPhOMe]
were prepared as potential bioreductive prodrugs and reduced radiolytically to study their rates
of subsequent intramolecular cyclization. Compounds bearing a free NH group (R ) H)
underwent rapid cyclization in neutral aqueous buffers (t1/2 < 1 min) following 4-electron
reduction, with the generation of a N-hydroxydihydroquinoxalinone and concomitant release
of 4-methoxyaniline. Amine release from analogous N-methyl analogues (R ) Me) was relatively
slow. These results are consistent with intramolecular cyclization of a monohydroxylamine
intermediate. The high rates of cyclization/extrusion by these very electron-deficient hydroxyl-
amines suggest that the process is greatly accelerated by the presence of an H-bonding
“conformational lock” between the anilino NH group and the adjacent o-nitro group (Kirk and
Cohen, 1972). Changes in the phenylcarboxamide side chain or in C-methylation in the linking
chain had little effect on the rate of cyclization. The model compounds had 1-electron reduction
potentials in the range appropriate for cellular reduction (-373 mV for a measured example)
and appeared suitable for development as prodrugs that release amine-based effectors following
enzymic or radiolytic reduction. Prodrug examples containing 4-aminoaniline mustard and
5-amino-1-(chloromethyl)benz[e]indoline alkylating units were evaluated but were not activated
efficiently by cellular nitroreductases. However, cell killing by the radiation-induced reduction
of the latter prodrug was demonstrated.
The efficacy of most anticancer drugs is limited by
the damage they cause to populations of normal cycling
cells in the host. One approach to improving the utility
of such drugs is to develop less toxic prodrug forms
capable of selective release of the parent cytotoxins in
tumors. A number of possible tumor-selective reductive
mechanisms for the activation of prodrugs are known.
These include oxygen-inhibited reduction by endogenous
enzymes1-4 or therapeutic radiation5 and selective
reduction by introduced exogenous enzymes in antibody-
and gene-directed enzyme-prodrug therapy (ADEPT/
GDEPT) strategies.6-8 We have proposed a general
conceptual approach to the design of such prodrugs,
which divides the structure into three domains (trigger,
linker, and effector), allowing the independent optimiza-
tion of each.9 The trigger units control the selectivity of
activation, by undergoing reduction only in tumor cells.
We have recently reported the reductive release of
amines from nitroarylamide precursors, via amino10,11
and hydroxylamino12 amide cyclization-extrusion reac-
tions. Model compounds (e.g., 1a ) were shown to release
amines, via the hydroxylamine intermediate 1b, with
suitable rapidity (t1/2 < 1 min under physiological-type
conditions) following radiolytic reduction. However, the
related analogue 2 containing a cytotoxic effector (ni-
trogen mustard) did not show hypoxia-selective cyto-
toxicity in tumor cells in culture (W. R. Wilson, unpub-
lished data). A likely reason is that 2 still has a
1-electron reduction potential (-517 mV; R. F. Ander-
son, unpublished data) too low for rapid reduction by
endogenous enzymes. To raise the reduction potential
sufficiently in this design would require the use of
additional electron-withdrawing substituents on the
nitrophenyl ring, which in turn would slow the intramo-
lecular cyclization-extrusion process.10 Nitroimidazole
analogues do have suitable reduction potentials, but
model compounds (e.g., 3) do not undergo intramolecular
cyclization following chemical reduction, and mustard-
containing analogues (e.g., 4) were not hypoxia-selective
in cell culture.13 The different geometry of the 5-mem-
bered ring is the most likely reason, since previous
work10,14,15 had shown that the ground-state conforma-
tion of the molecule has a very large influence on rates
of cyclization.
† Auckland Cancer Society Research Centre.
‡ Department of Pathology.
In the present study we evaluate a new class of
reductively activated prodrugs based on a related
§ Department of Chemistry.
10.1021/jm960783s CCC: $18.00 © 1999 American Chemical Society
Published on Web 02/11/1999

N-Substituted 2-(Dinitrophenylamino)propanamides
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3 347
Sch em e 1a
bonding lock, while 16 and 17 compare different solu-
bilizing functions. Analogues 26 and 32, containing
different types of cytotoxic effectors, were evaluated in
cells to determine their utility as prodrugs.
Syn th etic Ch em istr y
Schemes 2 and 3 outline the synthetic routes to the
compounds. Diethyl cyanophosphonate (DECP)-induced
coupling of the known N-BOC amino acids 7a -7c with
4-methoxyaniline gave the intermediates 8a -8c, which
were deblocked to provide the amines 9a -9c. The
N-methylamine 9d was prepared from 9b via the
trifluoroacetyl intermediate 10 and subsequent N-
methyl derivative 11. Amine 9e was prepared from
N-CBZ-sarcosine (12) via amide 13. Reactions of amines
9a -9e with 4-chloro-N-(2-hydroxyethyl)-3,5-dinitro-
benzamide (18) (obtained from acid 14) gave the target
model prodrugs 19a -19e. Condensation of 9b with 14
to give 15, followed by amide formation, gave the
analogues 16 and 17.
The mustard prodrug 26 was prepared as shown in
Scheme 4. DECP-promoted reaction of 4-aminoaniline
mustard19 22 with N-BOC-alanine (7b) gave 23, which
after alanine deprotection was reacted with 4-chloro-
3,5-dinitrobenzoic acid (14). CDI-assisted amide forma-
tion from the resulting acid 25 gave 26 in good overall
yield (40% from 22).
a
(i) e- (aq).
N-(2,6-dinitrophenyl)amino trigger. These were de-
signed to both possess appropriate reduction potentials
and to undergo rapid subsequent cyclization-extrusion.
The second (symmetric) nitro group not only raises the
reduction potential but also, through H-bonded locking,
correctly positions the initial hydroxylamine or amine
for fast cyclization-extrusion (Scheme 1). This approach
has been used previously for the sequential cleavage of
peptide bonds.16-18 Coupling of an amino acid or peptide
with 3,5-dinitro-2-fluoroaniline to give a 3,5-dinitro-
anilide derivative (e.g., 6)16 results in immediate cy-
clization, via attack on the neighboring acid or amide
function by the amino group, to give a dihydroquinox-
alinone. The surprisingly facile cyclization of the very
electron-deficient 6 under alkaline conditions (prevent-
ing its isolation at pH 8) has been attributed17 to
conformational restriction of the side chain via H-
bonding to the o-nitro group. An analogue (5),17 more
electron-rich but not so constrained, was much more
stable (t1/2 5.8 h at pH 4.96 and >18 h at pH > 5).
The seco-cyclopropylindoline prodrug 32 was prepared
as shown in Scheme 5. Reaction of 14 with alanine ester
27 gave the substituted alanine 28, which was in turn
converted to the amide 29. Cleavage of the 4-methoxy-
benzyl ester group with anisole/TFA gave the acid 30,
which was coupled to 5-amino-seco-CBI-TMI20 (31) to
give the target compound.
Red u ction Ch em istr y of Mod el P r od r u gs
Anaerobic, aqueous solutions (pH 7.4, 0.01 M phos-
phate buffer, 0.5 M propan-2-ol) of 19a -19c, 16, and
17 were irradiated for times corresponding to the
introduction of 1-6 molar equiv of reducing species (i.e.,
Model compounds 19a -19e were made to explore the
geometry of the linker group between the trigger and
model effector units and the importance of the H-
e-
and (CH3)2C.OH). The concentration of starting
(aq)
material was 21 µM, with the exception of 19a , whose
low aqueous solubility allowed solutions of only 3 µM
Sch em e 2a
a
(i) 4-Methoxyaniline/DECP/1-methylimidazole/DMF/0 °C, then 20 °C/2 h; (ii) TFA/20 °C/1 h; (iii) TFAA/pyridine/CH2Cl2/0 °C;
(iv) NaH/DMF/20 °C/2 h, then MeI/20 °C/6 h; (v) Cs2CO3/MeOH/water/reflux/30 min; (vi) Pd/C/H2/EtOH/20 °C/h/60 psi.

348 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3
Sykes et al.
Sch em e 3a
Sch em e 5a
a
(i) SOCl2/DMF, then H2N(CH2)2OH/Me2CO/water/0 °C, then
HCl; (ii) 9a -9e/THF/(iPr)2NEt/20 °C/8 h; (iii) 9b/THF/(iPr)2NEt/
20 °C/12 h, then 50 °C/30 min; (iv) CDI/DMF/0 °C, then
H2N(CH2)2NMe2/0 °C; (v) CDI/DMF/20 °C, then H2NCH2CH(OH)-
CH2OH/0 °C.
Sch em e 4a
(i) (iPr)2NEt/THF/25 °C/30 h; (ii) CDI/DMF/20 °C/1 h, then
H2N(CH2)2NMe2; (iii) anisole/TFA/20 °C/15 min; (iv) EDCI HCl/
DMA/20 °C/4 h.
a
a
(i) DECP/DMF/0 °C, then (iPr)2NEt/20 °C/1.5 h; (ii) TFA/20
°C/2 h; (iii) 4-chloro-3,5-dinitrobenzoic acid (14)/(iPr)2NEt/20 °C/
12 h; (iv) CDI/DMF/25 °C/1 h, then H2N(CH2)2NMe2/25 °C/5 min.
time and spectrum. An exception was 16, for which no
corresponding N-hydroxydihydroquinoxalinone coprod-
uct was detected at pH 7.4.
in buffer containing 1 M 2-propanol, 1 M acetone.
Radiolytic reductions of 19a were therefore carried out
in steps of 2 stoichiometric equiv (the addition of 1
reducing equiv to a 3 µM solution required an irradia-
tion time of only 7.5 s).
The yield of 4-methoxyaniline was not altered by
irradiating solutions of 19 a -19c, 16, and 17 with >4
reducing equiv, but the N-hydroxydihydroquinoxali-
nones were consumed (e.g., irradiation of 19b with up
to 14 reducing equiv gave two new unidentified products
which maximized at 8-fold stoichiometry; data not
shown). The absorbance spectrum of solutions of 19b
and 19c did not change with time up to 120 min after
reduction with 4-fold stoichiometry (irradiation time 4
min), and the concentrations of released 4-methoxy-
aniline also showed no increase when HPLC analysis
of irradiated solutions was delayed. Thus release of
4-methoxyaniline from the N-hydroxydihydroquinoxali-
none intermediates is too fast to detect kinetically at
these low dose rates, and the half-life for this reaction
is estimated at <1 min.
Figure 1A depicts the changes in the composition of
radiolysis solutions of 19b as the stoichiometry of
reduction was varied and typifies the radiolytic reduc-
tion chemistry of 19a -19c, 16, and 17. The nitro
compound 19b was consumed with 4-fold stoichiometry,
consistent with either monoreduction of a nitro group
to the hydroxylamine or reduction of both nitro groups
to the dinitroso compound. The two major products
formed in parallel were 4-methoxyaniline (by compari-
son of retention time and spectrum with an authentic
sample) and N-hydroxydihydroquinoxalinone (21) (iden-
tified by mass spectrometry of the mixture of products
formed after 4-fold reduction). The latter arises from
cyclization of the initially formed monohydroxylamine
(20) (Scheme 1). 4-Methoxyaniline was likewise formed
from reduced 19a , 19c, 16, and 17 along with a
coproduct, assumed to be the corresponding N-hydroxy-
dihydroquinoxalinone on the basis of similar retention
The N-methyl analogues 19d and 19e at pH 7 were
too hydrolytically unstable for their reduction to be
studied at pH 7, but this was done at pH 4 (together
with their desmethyl counterparts 19a and 19b). Figure
1B shows the changes in composition of 19e (19d was
similar, and 19a and 19b gave similar results to those

N-Substituted 2-(Dinitrophenylamino)propanamides
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3 349
Ta ble 1. Comparison of the Observed Half-Life for and
Percent of 4-Methoxyaniline Released from 4-fold Reduced
19a -19c, 16, and 17 (pH 7.4, 10 mM phosphate buffer, 0.5 M
2-propanol, rt) and 19a , 19b, 19d , and 19e (pH 4.0, 10 mM
formate buffer, 0.5 M 2-propanol, rt)
4-fold reduced
released
8-fold reduced
t1/2 released
t1/2
no.
(min)
4-methoxyaniline (min) 4-methoxyaniline
pH 7.4
19a <1
19b <1
19c <1
69%
77%
90%
89%
92%
a
a
a
a
a
16
17
<1
<1
pH 4.0
19a <1
19b <1
83%
97%
80 ( 10%
80 ( 20%
a
a
19d
19e
7.2 ( 5
12 ( 1
4 ( 1
15 ( 1
100 ( 2%
94 ( 2%
a
Percent release of 4-methoxyaniline was not significantly
different from that released by 4-fold reduction, and no evidence
was found for formation of a cyclizable, di-NHOH form of the
prodrug.
analogues containing cytotoxic primary amine-based
effectors were prepared (26, 32). The trigger investi-
gated was based on the rapidly cyclizing 19b, but with
a basic 4-side chain to improve aqueous solubility. The
cytotoxicities of the prodrugs and effectors, as deter-
mined by growth inhibition (IC50 assays) and plating
efficiency (clonogenic assays), are summarized in Table
2. The aniline mustard 22 was 33-fold more toxic (IC50
assay) than the corresponding prodrug 26 against the
UV4 cell line, which is defective in DNA cross-link
repair.21 This indicates that the toxicity of the effector
is masked effectively in the prodrug form. However, the
repair-competent cell lines (AA8, EMT6, SKOV3) showed
similar sensitivity to 22 and 26. The lack of hypersen-
sitivity of UV4 to 26, relative to 22, clearly demonstrates
that the toxicity of the prodrug in repair-competent cells
is not due to release of 22.
A much more potent cytotoxic amine effector was
therefore investigated, in an attempt to avoid the
toxicity of the prodrug form seen in the nitrogen
mustard example. The DNA minor groove alkylator
5-amino-seco-CBI-TMI20 (31) had IC50 values e 1 nM
against all four cell lines, and the corresponding prodrug
32 was much less toxic (87-206-fold) in all lines. The
cytotoxicity of these compounds against aerobic UV4
cells in clonogenic assays confirmed these conclusions,
demonstrating the high potency of 31 relative to 32.
Under N2 the potency of 32 (but not 31) was increased
1.8-fold; this selectivity for hypoxia was statistically
significant and was greater than that for the corre-
sponding mustard prodrug 26 (which showed a dif-
ferential no greater than its amine effector 22). How-
ever, the increase in activity under hypoxia is minor
relative to the large change theoretically possible (140-
fold if all the 32 were converted rapidly to 31), suggest-
ing either that 32 is a poor substrate for reduction by
reductases in these cells or that reduction intermediates
such as the nitroso or hydroxylamine are intercepted
before intramolecular cyclization occurs.
F igu r e 1. Changes in solution composition with changing
stoichiometry of radiolytic reduction. A: Changes to 19b, 21
µM at pH 7.4, 0.01 M phosphate, room temperature. B:
Changes to 19e, 25 µM at pH 4.0, 0.01 M formate, room
temperature. Symbols refer to model prodrug (b), 4-methoxy-
aniline (9), ring-closed N-hydroxydihydroquinoxalinone (21)
([), the monohydroxylamino reduction product (tR ) 4.4 min)
(1), the dihydroxylamino reduction product (tR ) 3.0 min) (2).
at pH 7.4; Figure 1A). All were reduced with 4-fold
stoichiometry at pH 4.0, but product formation differed
between the N-methyl and desmethyl counterparts.
Both 19d (retention time 6.4 min) and 19e were
converted to a more polar product that maximized at
4-fold stoichiometry, and this was then converted to a
second product at 8-fold stoichiometry. These patterns
are consistent with sequential reduction of both nitro
groups to the hydroxylamines.
4-Methoxyaniline was observed in irradiated solutions
of 19d and 19e, but yields immediately after irradiation
were low after 4-fold reduction stoichiometry and ap-
peared to increase further with dose (Figure 1B).
However, the detection of hydroxylamino intermediates
in the case of the N-methyl compounds suggested that
this apparent increase at higher doses might reflect slow
cyclization kinetics of the latter. The spectra of solutions
of 19d and 19e reduced with 4- and 8-fold stoichiometry
changed with time following irradiation and passed
through isosbestic points. The observed pseudo-first-
order rate coefficients for reaction of the 4- and 8-fold
reduced forms were measured, and values of the corre-
sponding reaction half-lifes (t1/2 ) ln 2/kobs) are pre-
sented in Table 1. The product solutions eventually
showed essentially 100% release of 4-methoxyaniline
and loss of the hydroxylamino intermediates (Table 1).
Biologica l Stu d ies
Although not activated efficiently by enzymatic re-
duction in hypoxic cells, the extreme potency of 31
suggested that it might be possible to reduce enough of
the prodrug 32 with ionizing radiation to achieve
In Vitr o Biologica l Eva lu a tion : Cytotoxicity a n d
Activa tion by Ra d ia tion . To study the usefulness of
2-(2,6-dinitrophenylamino)propanamides as prodrugs,

350 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3
Sykes et al.
Ta ble 2. Cytotoxicity Data for Prodrugs 26 and 32 and Their Precursors 22 and 31
growth inhibition (IC50, nM)a
UV4 clonogenic assay
c
no.
AA8
UV4
EMT6
SKOV3
C10 (nM)b
air/N2
1.4
1.3
22
26
31
32
5700 ( 1100
10500 ( 900
0.46 ( 0.05
95 ( 6
90 ( 30
1200 ( 100
1300 ( 200
0.27 ( 0.03
41 ( 8
2400 ( 300
1300 ( 200
1.0 ( 0.1
87 ( 2
220
3000 ( 600
0.29 ( 0.2
53 ( 1
2100
3.4 ( 0.4d
490 ( 20d
1.1 ( 0.1d
1.8 ( 0.3d
a
IC50, concentration of drug to reduce cell numbers to 50% of controls in a growth inhibition assay (4-h aerobic drug exposure). See
b
text for description of cell lines. Values are mean ( SEM for 3-5 experiments. C10, drug concentration to reduce plating efficiency of
c
d
UV4 cells to 10% of controls (1-h exposure) under aerobic conditions (20% O2). C10 in 20% O2/C10 in N2. Values are mean ( SEM for
3 experiments.
masses at doses as low as 13.3 µmol/kg. The maximum
nontoxic dose of 32 was approximately 7.5 µmol/kg.
Compound 32 was tested against the RIF-1 tumor at
133 µmol/kg (i.e., well above the dose which caused
delayed toxicity) by excising tumors for clonogenic assay
18 h after treatment. The compound showed no activity
as a single agent, and no increase in cell killing above
that for radiation only (15 Gy) was seen when 32 was
administered either 30 min before or 5 min after
irradiation.
Discu ssion
The 2-(2,6-dinitrophenylamino)propanamide prodrugs
were designed to exploit a proposed17 H-bonding “lock”
to accelerate the kinetics of reductively induced cycliza-
tion. Examples 19a -19c explored the geometry of the
linker group between the trigger and model effector
units. Examples 19d and 19e, with an N-methyl group
preventing any H-bonding to the ring nitro group, were
prepared for direct comparison with the corresponding
“lockable” NH derivatives 19a and 19b, and 16 and 17
were made to examine the effects of different solubiliz-
ing side chains.
The results show that radiolytic reduction under
physiological conditions (aqueous solution at pH 7)
F igu r e 2. A: Increased cytotoxic activity of the prodrug 32
(1 µM) following γ-irradiation in duplicate (b, [) in anoxic
results in initial reduction of only one nitro group to
the hydroxylamine (4-fold stoichiometry). The most
likely alternative possibility (reduction of both nitro
groups to a noncyclizable dinitroso species) is incompat-
ible with the release of the ring-closed N-hydroxydihy-
droquinoxalinone species and 4-methoxyaniline. Cy-
clization of the monohydroxylamines from 19b, 19c, 16,
and 17 is rapid, going to completion during the irradia-
tion. Given a time of 56 s for reduction of one-fourth of
19b (addition of 1 reducing equiv), and a total time of
approximately 4 min from commencement of the reduc-
tion to HPLC assay, the t1/2 for cyclization appears to
be <1 min (pH 7.4 and 4.0, room temperature). This is
a much faster release rate than observed for any
monohydroxylamino aryl amide system12 (e.g., 33 and
34) (Figure 3).
N-Methylation of the proposed H-donor site slowed
cyclization significantly, although this could not be
quantified due to the very rapid cyclization of the
“locked” analogues 19a and 19b (Table 1). However, a
significant role for H-bonding is suggested by previous
work.10-14 For the “locked” compounds (19a -19c),
quantitative evaluation of the effects of C-substituents
in the linker chain was not possible, since all cyclized
with t1/2 values < 1 min. However, a comparison of the
slower-reacting “unlocked” compounds 19d and 19e
suggests that C-methylation does slow the cyclization
reaction slightly, via both the mono- and dihydroxyl-
culture medium containing 5% FCS. The IC50 of irradiated
solutions was determined by bioassay against the UV4 cell
line. B: Yield of effector 31 in irradiated solutions, calculated
from the measured IC50 values26 assuming the cytotoxic effects
of 31 and 32 are independent and additive.
significant activation at clinically relevant radiation
doses in biological systems. Irradiation of 32 (1 µM, close
to its solubility limit of 1.3 µM) in anoxic culture
medium containing 5% fetal calf serum at doses up to
10.7 Gy resulted in a significant activation as deter-
mined by bioassay against UV4 cells (Figure 2). No
increase in cytotoxicity was observed upon radiolysis of
the corresponding mustard prodrug 26 (50 µM, doses
up to 200 Gy, data not shown), as expected given the
low potency of the resulting effector.
In Vivo Biologica l Eva lu a tion : Toxicity a n d
An titu m or Activity. Animals treated ip with 31 at
doses in the range 5.6-10 µmol/kg died within 5-18
days; at doses lower than this, benign masses developed
within the peritoneum after several weeks, resulting in
distension of the abdomen, bowel obstruction, with
sporadic deaths after about 4 weeks. The intraperitoneal
masses involved the omentum and intestines, with
histological evidence of fatty necrosis. The highest
nontoxic dose was 1.78 µmol/kg. The 2-(2,6-dinitrophen-
ylamino)propanamide prodrug 32 showed no acute
toxicity at doses up to at least 178 µmol/kg, but delayed
toxicity was again seen with development of abdominal

N-Substituted 2-(Dinitrophenylamino)propanamides
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3 351
times the IC50 of 31 against the least sensitive tumor
cell line tested (SKOV3).
Despite these promising characteristics in vitro, 32
lacks significant activity against hypoxic cells in RIF-1
tumors (as demonstrated by failure to kill additional
cells when combined with ionizing radiation), even at
prodrug doses high enough to eventually cause chronic
toxicity to the host. These experiments were designed
to detect either bioreductive activation by endogenous
enzymes in hypoxic regions (increased killing when drug
is administered before or after radiation) or radiolytic
activation (increased killing when drug is administered
before radiation only). Many requirements must be met
simultaneously for activity against hypoxic cells in vivo.
The lack of tumor activity of 32 could reflect unfavorable
pharmacokinetics of the prodrug, inefficient radiolytic
and enzymatic activation in tissue, alternate pathways
for removal of the hydroxylamine in competition with
cyclization, or lack of activity of the released effector in
the tumor microenvironment. Nevertheless, the N-(2,6-
dinitrophenyl)amino trigger system described here pro-
vides a novel and well-defined chemistry for reductive
activation of prodrugs and may provide therapeutically
useful agents if these issues can be addressed.
F igu r e 3. Comparison of the half-life for rate of cyclization
(pH 7.4, room temperature) of the hydroxylamine intermedi-
ates derived from 19c with two previously investigated12 model
prodrugs.
amino intermediates (Table 1). Changes to the carbox-
amide side chain increased the solubility significantly
(relative solubilities 17 < 19b < 16) without affecting
rates of reductive cyclization, as expected.
The kinetic investigation shows that NH analogues
possessing an H-bonding “lock” capability cyclize rapidly
following reduction to the monohydroxylamine. As 19b
has a 1-electron reduction potential of -373 ( 10 mV
(measured by pulse radiolysis, R. F. Anderson, unpub-
lished data), these compounds should be efficiently
reduced by cellular nitroreductases, suggesting this as
a suitable system for the development of hypoxia-
activated prodrugs bearing amine-based effectors.9 To
evaluate the utility of this approach, analogues 26 and
32 were prepared as prodrugs of 4-aminoaniline mus-
tard19 (22) and 5-amino-seco-CBI-TMI20-23 (31), respec-
tively. Previous studies of both effectors have shown
that such prodrugs should be markedly less cytotoxic
than the corresponding amine effectors. The results
listed in Table 2 show that while the mustard 22 was
not consistently deactivated as expected, the indoline
31 was.
Exp er im en ta l Section
Syn th etic Ch em istr y. Analyses were performed by the
Microchemical Laboratory, University of Otago, Dunedin.
Melting points were determined on an Electrothermal model
9200 digital melting point apparatus and are as read. NMR
spectra were measured on a Bruker AM-200 or AM-400
spectrometer and referenced to Me4Si.
P r ep a r a tion of P r otected Am in es. 2-[(ter t-Bu tyloxy-
ca r bon yl)a m in o]-N-(4-m eth oxyp h en yl)-2-m eth ylp r op a n -
a m id e (8c). A stirred solution of 2-[(tert-butyloxycarbonyl)-
amino]-2-methylpropanoic acid (7c) (3.50 g, 17.2 mmol),
4-methoxyaniline (2.33 g, 18.9 mmol), and 1-methylimidazole
(1.70 g, 20.7 mmol) in DMF (15 mL) was treated dropwise at
0 °C with diethyl cyanophosphonate (93%, 3.32 g, 18.9 mmol).
The mixture was stirred at 20 °C for 24 h and then diluted
with excess aqueous Na2CO3. The resulting solid was recrys-
tallized twice from MeOH/H2O to give 8c (3.89 g, 73%): mp
201 °C; 1H NMR (CDCl3) δ 8.11 (br s, 1 H, CONHPh), 7.42 (d,
J ) 9.0 Hz, 2 H, H-2′,6′), 6.85 (d, J ) 9.0 Hz, 2 H, H-3′,5′),
4.91 (br s, 1 H, OCONH), 3.90 (s, 3 H, OCH3), 1.56 (s, 6 H,
C(CH3)2), 1.44 (s, 9 H, C(CH3)3). Anal. (C16H24N2O4) C, H, N.
[(ter t-Bu tyloxyca r bon yl)a m in o]-N-(4-m eth oxyp h en yl)-
a ceta m id e (8a ). Reaction of N-(tert-butyloxycarbonyl)glycine
(7a ) and 4-methoxyaniline as above gave 8a (82%): mp
(MeOH/H2O) 143-144 °C; 1H NMR (CDCl3) δ 8.10 (br s, 1 H,
CONHPh), 7.40 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.85 (d, J ) 9.0
Hz, 2 H, H-3′,5′), 5.33 (br s, 1 H, NHCH2), 3.92 (d, J ) 5.9 Hz,
2 H, CH2), 3.79 (s, 3 H, OCH3), 1.48 (s, 9 H, C(CH3)3). Anal.
(C14H20N2O4) C, H, N.
Despite this, and the fact that the prodrug 32 has a
high enough reduction potential to expect rapid cellular
reduction, it showed little hypoxic selectivity in cell
culture. Previous studies24 have shown that the rates
of reduction of ortho-substituted nitro compounds is
lower than predicted from their 1-electron reduction
potentials, but it is not known if this is a factor here. It
is also possible that intermediates of bioreduction such
as the nitroso or hydroxylamine are intercepted before
intramolecular cyclization occurs. However, Figure 2
shows that 32 is significantly activated by ionizing
radiation at low radiation doses. The IC50 data cor-
respond to a yield of effector [G(31) ) 4.5 ( 0.3 nmol
J -1],25 which is only 1.6% of the G value for e-(aq), or
2-[(ter t-Bu tyloxycar bon yl)am in o]-N-(4-m eth oxyph en yl)-
p r op a n a m id e (8b). Reaction of N-(tert-butyloxycarbonyl)-
alanine (7b) and 4-methoxyaniline as above gave 8b (77%):
mp (MeOH/H2O) 139-140 °C; 1H NMR (CDCl3) δ 8.34 (br s, 1
H, CONHPh), 7.41 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.84 (d, J )
9.0 Hz, 2 H, H-3′,5′), 5.16 (br s, 1 H, OCONH), 4.31 (br s, 1 H,
CHCO), 3.78 (s, 3 H, OCH3), 1.46 (s, 9 H, C(CH3)3), 1.42 (d, J
) 7.0 Hz, 3 H, CHCH3). Anal. (C15H22N2O4) C, H, N.
6.4% of the theoretical yield if all e-
were captured
(aq)
and 4-fold stoichiometry of activation was required. This
low yield is not surprising, given the low concentration
of 32 (1 µM) which must compromise its ability to
compete for e-(aq). However, the result demonstrates
that even inefficient radiolytic activation might be
exploited to activate prodrugs if a sufficiently potent
effector is released. In this instance, a clinically relevant
dose of 2 Gy is calculated to release 9 nM effector, 9
N-(4-Meth oxyp h en yl)-2-(tr iflu or oa cetyla m in o)p r op a n -
a m id e (10). Reaction of trifluoroacetic anhydride with 9b in
CH2Cl2 containing pyridine gave 10 (78%): mp (MeOH/H2O)
1
190-191 °C; H NMR [(CD3)2SO] δ 10.01 (s, 1 H, CONHPh),
9.70 (s, 1 H, CF3CONH), 7.49 (d, J ) 9.1 Hz, 2 H, H-2′,6′),
6.89 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 4.46 (br s, 1 H, CH), 3.72 (s,
3 H, OCH3), 1.40 (d, J ) 7.2 Hz, 3 H, CHCH3). Anal.
(C12H13F3N2O3) C, H, N.

352 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3
Sykes et al.
N-(4-Meth oxyph en yl)-2-[N-(tr iflu or oacetyl)m eth ylam i-
n o]p r op a n a m id e (11). A solution of 10 (1.20 g, 4.13 mmol)
in DMF (6 mL) was treated with NaH (0.173 g, 4.33 mmol,
60% dispersion in mineral oil), and the mixture was stirred
at 20 °C for 2 h. Iodomethane (0.615 g, 4.33 mmol) was then
added, and the mixture was stirred for a further 8 h at 20 °C
and then poured into 0.1 N aqueous AcOH. Prolonged cooling
gave a solid which was collected and chromatographed on silica
gel. Elution with CH2Cl2 removed the mineral oil, and elution
with CH2Cl2/EtOAc (19:1) gave material which was recrystal-
lized from MeOH/H2O to give 11 (0.72 g, 57%): mp 116-117
°C; 1H NMR [(CD3)2SO] δ 9.97 (s, 1 H, CONHPh), 7.48 (d, J )
9.1 Hz, 2 H, H-2′,6′), 6.89 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 4.93 (q,
J ) 7.2 Hz, 1 H, CHCO), 3.72 (s, 3 H, OCH3), 3.11 (s, 3 H,
NCH3), 1.44 (d, J ) 7.2 Hz, 3 H, CHCH3). Anal. (C13H15F3N2O3)
C, H, N.
dinitrobenzoic acid (14) (2.12 g, 8.6 mmol), amine 9b (2.00 g,
10.3 mmol), and diisopropylethylamine (2.34 g, 18.1 mmol) in
THF (20 mL) was stirred at 20 °C for 12 h and then heated at
50 °C for 30 min. The mixture was concentrated under reduced
pressure, and the residue was extracted with warm dilute
aqueous ammonia. The filtrate was acidified with concentrated
HCl, and the resulting solid was recrystallized twice from
1
EtOH/H2O to give 15 (2.97 g, 85%): mp 219-221 °C dec; H
NMR [(CD3)2SO] δ 13.60 (br s, 1 H, COOH), 10.22 (s, 1 H,
CONHPh), 9.30 (d, J ) 7.5 Hz, 1 H, PhNHCH2), 8.66 (s, 2 H,
H-3,5), 7.47 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.90 (d, J ) 9.0 Hz,
2 H, H-3′,5′), 4.15 (quint, J ) 6.8 Hz, 1 H, CHCH3), 3.73 (s, 3
H, OCH3), 1.37 (d, J ) 6.7 Hz, 3 H, CHCH3). Anal. (C17H16N4O8)
C, H, N.
2-[[4-[N-[2-(Dim eth yla m in o)eth yl]ca r ba m oyl]-2,6-d in i-
tr oph en yl]am in o]-N-(4-m eth oxyph en yl)pr opan am ide (16).
A stirred solution of acid 15 (1.20 g, 2.97 mmol) in DMF (8
mL) was treated with 1,1′-carbonyldiimidazole (0.577 g, 3.56
mmol), at 20 °C for 30 min, then cooled to 0 °C, and treated
with N,N-dimethylethylenediamine (0.367 g, 4.16 mmol). The
mixture was stirred for a further 15 min at 20 °C and then
diluted with dilute KHCO3. The precipitated solid was dis-
solved in dilute aqueous AcOH, and the filtered solution was
basified with dilute KHCO3. Crystallization of the precipitate
from aqueous MeOH followed by EtOAc/petroleum ether gave
16 (0.98 g, 70%): mp 163-164 °C; 1H NMR [(CD3)2SO] δ 10.21
(s, 1 H. CONHPh), 9.10 (d, J ) 7.5 Hz, 1 H, PhNHCH), 8.82-
8.73 (m, 3 H, PhCONH, H-3,5), 7.46 (d, J ) 8.9 Hz, 2 H,
H-2′,6′), 6.90 (d, J ) 9.0 Hz, 2 H, H-3′,5′), 4.13 (quint, J ) 6.8
Hz, 1 H, CHCH3), 3.72 (s, 3 H, OCH3), ca. 3.4 (under H2O,
CH2), 2.42 (t, J ) 6.6 Hz, 2 H, CH2), 2.19 (s, 6 H, N(CH3)3),
1.37 (d, J ) 6.7 Hz, 3 H, CHCH3). Anal. (C21H26N6O7) C, H, N.
[N-(Ben zyloxyca r bon yl)m eth yla m in o]-N-(4-m eth oxy-
p h en yl)a ceta m id e (13). Reaction of N-(benzyloxycarbonyl)-
sarcosine (12) and 4-methoxyaniline following the procedure
for the preparation of 8c gave 13 (69%): mp (MeOH/H2O) 135
1
°C; H NMR (CDCl3) δ 8.01 (br s, 1 H, CONHPh), 7.41-7.26
(m, 7 H, ArH, H-2′,6′), 6.84 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 5.19
(s, 2 H, CH2O), 4.03 (s, 2 H, CH2CO), 3.79 (s, 3 H, OCH3), 3.08
(s, 3 H, NCH3). Anal. (C18H20N2O4) C, H, N.
P r ep a r a tion of Am in es 9a -9e. 2-Am in o-N-(4-m eth oxy-
p h en yl)-2-m eth ylp r op a n a m id e (9c). A solution of 8c (3.60
g, 11.7 mmol) in CF3COOH (20 mL) was stirred at 20 °C for 1
h and then concentrated under reduced pressure below 30 °C.
The residue was shaken with saturated aqueous Na2CO3 and
then evaporated to dryness under reduced pressure. The
resulting solid was extracted with hot EtOAc, and the filtered
extract was washed with saturated aqueous NaCl, dried, and
concentrated under reduced pressure. The resulting oil was
chromatographed on alumina 90, eluting with CH2Cl2, and the
resulting product was recrystallized from petroleum ether to
2-[[4-[N-(2,3-Dih yd r oxyp r op yl)ca r ba m oyl]-2,6-d in itr o-
p h en yl]a m in o]-N-(4-m eth oxyp h en yl)p r op a n a m id e (17).
A stirred solution of acid 15 (1.60 g, 3.96 mmol) in DMF (10
mL) was treated with 1,1′-carbonyldiimidazole (0.707 g, 4.36
mmol) at 20 °C for 30 min, then cooled to -5 °C, and treated
in one portion with 3-amino-1,2-propanediol (0.469 g, 5.15
mmol). The mixture was concentrated under reduced pressure
below 50 °C, and the residue was shaken with dilute KHCO3.
The resulting solid was crystallized from aqueous MeOH
followed by MeOH/EtOAc/iPr2O to give 17 (1.21 g, 64%): mp
1
give 9c (2.03 g, 84%): mp 51 °C; H NMR (CDCl3) δ 9.43 (br
s, 1 H, CONH), 7.52 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.86 (d, J )
9.0 Hz, 2 H, H-3′,5′), 3.79 (s, 3 H, OCH3), 1.45 (s, 6 H, C(CH3)2).
Anal. (C11H16N2O2) C, H, N.
Am in o-N-(4-m et h oxyp h en yl)a cet a m id e (9a ). Similar
treatment of 8a gave 9a (89%): mp (EtOAc/iPr2O) 90-91 °C;
1H NMR (CDCl3) δ 9.24 (br s, 1 H, CONH), 7.50 (d, J ) 9.0
Hz, 2 H, H-2′,6′), 6.87 (d, J ) 9.0 Hz, 2 H, H-3′,5′), 3.79 (s, 3
H, OCH3), 3.45 (s, 2 H, CH2). Anal. (C9H12N2O2) C, H, N.
2-Am in o-N-(4-m eth oxyp h en yl)p r op a n a m id e (9b). Simi-
lar treatment of 8b, followed by purification of the product by
chromatography on alumina 90, eluting with CH2Cl2/EtOAc
1
203 °C dec; H NMR [(CD3)2SO] δ 10.21 (s, 1 H. CONHPh),
9.09 (d, J ) 7.4 Hz, 1 H, PhNHCH), 8.84-8.75 (m, 3 H,
PhCONH, H-3,5), 7.46 (d, J ) 8.8 Hz, 2 H, H-2′,6′), 6.90 (d, J
) 8.8 Hz, 2 H, H-3′,5′), 4.85 (d, J ) 4.9 Hz, 1 H, CHOH), 4.59
(t, J ) 5.7 Hz, 1 H, CH2OH), 4.13 (quint, J ) 6.5 Hz, 1 H,
CHCH3), 3.72 (s, 3 H, OCH3), 3.68-3.60 (m, 1 H, CHOH),
3.46-3.30 (m, 2 H, CH2), 3.24-3.13 (m, 2 H, CH2), 1.37 (d, J
) 6.6 Hz, 3 H, CHCH3). Anal. (C20H23N5O9) C, H, N.
1
(1:1), gave 9b (85%): mp 64-65 °C; H NMR (CDCl3) δ 9.33
(br s, 1 H, CONH), 7.50 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.86 (d,
J ) 9.0 Hz, 2 H, H-3′,5′), 3.79 (s, 3 H, OCH3), 3.61 (q, J ) 7.0
Hz, 1 H, CH), 1.42 (d, J ) 7.0 Hz, 3 H, CHCH3). Anal.
(C10H14N2O2) C, H, N.
4-Ch lor o-N-(2-h ydr oxyeth yl)-3,5-din itr oben zam ide (18).
A mixture of 14 (4.66 g, 18.9 mmol), SOCl2 (25 mL), and DMF
(0.2 mL) was heated under reflux for 30 min; then excess
reagent was removed under reduced pressure. Traces of SOCl2
were removed by azeotroping with benzene, to give a crude
acid chloride, which was dissolved in Me2CO (35 mL). The
solution was cooled to 0 °C and added in one portion to a
stirred solution of 2-aminoethanol (2.53 g, 41.4 mmol) in water
(45 mL) at 0 °C. The mixture was stirred vigorously for a
further 30 s and then immediately treated with concentrated
HCl (1 mL). Dilution with water (400 mL) gave a yellow solid,
which was stirred as a suspension in 10% aqueous KHCO3
(50 mL) and then collected and recrystallized sequentially from
EtOAc and MeOH/H2O/trace concentrated HCl to give 18 (4.06
N-(4-Met h oxyp h en yl)-2-(m et h yla m in o)p r op a n a m id e
(9d ). A mixture of 11 (0.62 g, 2.04 mmol) and Cs2CO3 (1.63 g)
in 50% aqueous MeOH (5 mL) was heated under reflux for 30
min and then evaporated to dryness under reduced pressure.
The resulting residue was extracted with EtOAc (2 × 25 mL),
and the combined extract was filtered and evaporated. Re-
crystallization of the product from petroleum ether gave 9d
1
(0.36 g, 86%): mp 77 °C; H NMR (CDCl3) δ 9.15 (br s, 1 H,
CONH), 7.51 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.87 (d, J ) 9.0 Hz,
2 H, H-3′,5′), 3.79 (s, 3 H, OCH3), 3.15 (q, J ) 6.9 Hz, 1 H,
CH), 2.46 (s, 3 H, NCH3), 1.38 (d, J ) 7.0 Hz, 3 H, CHCH3).
Anal. (C11H16N2O2) C, H, N.
1
g, 74%): mp 167-168 °C; H NMR [(CD3)2SO] δ 9.02 (t, J )
N-(Meth oxyp h en yl)(m eth yla m in o)a ceta m id e (9e). A
suspension of 13 in EtOH was hydrogenated over 5% Pd/C at
60 psi for 6 h to give 9e (96%): mp (iPr2O/petroleum ether)
53-54 °C; 1H NMR (CDCl3) δ 9.13 (br s, 1 H, CONH), 7.50 (d,
J ) 9.0 Hz, 2 H, H-2′,6′), 6.87 (d, J ) 9.0 Hz, 2 H, H-3′,5′),
3.79 (s, 3 H, OCH3), 3.34 (s, 2 H, CH2), 2.50 (s, 3 H, NCH3).
Anal. (C10H14N2O2) C, H, N.
P r ep a r a tion of (2,6-Din itr op h en yl)a m in o Acid Der iva -
tives. 2-[(4-Ca r boxy-2,6-d in itr op h en yl)a m in o]-N-(4-m eth -
oxyp h en yl)p r op a n a m id e (15). A mixture of 4-chloro-3,5-
5.3 Hz, 1 CONH), 8.80 (s, 2 H, H-2,6), 4.80 (t, J ) 5.7 Hz, 1 H,
OH), 3.54 (q, J ) 5.8 Hz, 2 H, CH2), 3.37 (q, J ) 5.8 Hz, 2 H,
CH2). Anal. (C9H8ClN3O6) C, H, N.
2-[[2,6-Din itr o-4-[N-(2-h ydr oxyeth yl)car bam oyl]ph en yl]-
a m in o]-N-(4-m eth oxyp h en yl)p r op a n a m id e (19b). A mix-
ture of 18 (1.00 g, 3.45 mmol) and amine 9b (1.01 g, 5.20 mmol)
in THF (12 mL) was treated with diisopropylethylamine (0.45
g, 3.48 mmol). The reaction was stirred at 20 °C for 8 h and
then diluted with 0.5 N HCl. The precipitated solid was

N-Substituted 2-(Dinitrophenylamino)propanamides
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3 353
recrystallized twice from MeOH/H2O to give 19b (1.36 g,
88%): mp 183-184 °C; H NMR [(CD3)2SO] δ 10.20 (s, 1 H,
corresponding to the N-hydroxydihydroquinoxalinone 21 (calcd
for C12H15N4O6, 311.0992; found, 311.0987).
1
CONHPh), 9.09 (d, J ) 7.5 Hz, 1 H, PhNHCH), 8.85-8.76 (m,
3 H, H-3,5, PhCONH), 7.46 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.90
(d, J ) 9.0 Hz, 2 H, H-3′,5′), 4.76 (br s, 1 H, OH), 4.13 (quint,
J ) 7.1 Hz, 1 H, CHCH3), 3.73 (s, 3 H, OCH3), 3.52 (t, J ) 6.0
Hz, 2 H, CH2), 3.38-3.30 (m, 2 H, CH2), 1.37 (d, J ) 6.7 Hz,
3 H, CHCH3). Anal. (C19H21N5O8) C, H, N.
N-[4-[Bis(2-ch lor oeth yl)a m in o]p h en yl]-2-[[4-[N-[2-(d i-
m et h yla m in o)et h yl]ca r b a m oyl]-2,6-d in it r op h en yl]a m i-
n o]p r op a n a m id e (26) (Sch em e 4). A stirred mixture of
N-(tert-butyloxycarbonyl)alanine (7b) (1.89 g, 10 mmol) and
4-[bis(2-chloroethyl)amino]aniline hydrochloride19 (22) (2.70 g,
10 mmol) in DMF (10 mL) was treated at 0 °C with diethyl
cyanophosphonate (1.84 g of 93%, 10.5 mmol) and then
dropwise with diisopropylethylamine (2.71 g, 21 mmol). The
mixture was stirred at 20 °C for 1.5 h and then diluted with
10% aqueous Na2CO3. The precipitated semisolid was dissolved
in CH2Cl2, washed with water, and dried. The solution was
concentrated under reduced pressure below 30 °C to ca. 20
mL and then diluted with petroleum ether to give N-[4-[bis-
(2-chloroethyl)amino]phenyl]-2-(tert-butyloxycarbonylamino)-
propanamide (23) (3.79 g, 94%): mp (CH2Cl2/iPr2O) 61-63 °C;
1H NMR [(CD3)2SO] δ 9.62 (s, 1 H, CONHPh) 7.42 (d, J ) 9.0
Hz, 2 H, H-2′,6′), 6.96 (d, J ) 7.3 Hz, 1 H, NHCH) 6.71 (d, J
) 9.1 Hz, 2 H, H-3′,5′), 4.07 (quintet, J ) 7.1 Hz, 1 H, NHCH)
3.69 (t, J ) 3.2 Hz, 8 H, N(CH2CH2Cl)2), 1.38 (s, 9 H, C(CH3)3),
1.23 (d, J ) 7.1 Hz, 3 H, CHCH3). Anal. (C18H27Cl2N3O3‚
0.5C6H14O) C, H, N.
[[2,6-Din itr o-4-[N-(2-h yd r oxyeth yl)ca r ba m oyl]p h en yl]-
a m in o]-N-(4-m eth oxyp h en yl)a ceta m id e (19a ). Reaction of
18 with amine 9a by the above procedure gave 19a (91%): mp
1
(DMF/MeOH) 238-239 °C; H NMR [(CD3)2SO] δ 10.22 (s, 1
H, CONHPh), 9.37 (s, 1 H, PhNHCH2), 8.80 (s, 3 H, H-3,5,
PhCONH), 7.47 (d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.90 (d, J ) 9.0
Hz, 2 H, H-3′,5′), 4.76 (t, J ) 5.6 Hz, 1 H, OH), 3.85 (d, J )
3.8 Hz, 2 H, PhNHCH2), 3.72 (s, 3 H, OCH3), 3.52 (q, J ) 5.8
Hz, 2 H, CH2), 3.38-3.27 (m, 2 H, CH2). Anal. (C18H19N5O8)
C, H, N.
2-[[2,6-Din itr o-4-[N-(2-h ydr oxyeth yl)car bam oyl]ph en yl]-
am in o]-N-(4-m eth oxyph en yl)-2-m eth ylpr opan am ide (19c).
Reaction of 18 with amine 9c (for 60 h) by the above procedure
gave 19c (35%): mp (MeOH/H2O) 199-200 °C; 1H NMR
[(CD3)2SO] δ 9.64 (s, 1 H, CONHPh), 8.85 (t, J ) 5.4 Hz, 1 H,
PhCONH), 8.78 (s, 2 H, H-3,5), 8.11 (s, 1 H, NHC(CH3)2), 7.52
(d, J ) 9.0 Hz, 2 H, H-2′,6′), 6.91 (d, J ) 9.0 Hz, 2 H, H-3′,5′),
4.77 (t, J ) 5.7 Hz, 1 H, OH), 3.74 (s, 3 H, OCH3), 3.52 (q, J
) 5.9 Hz, 2 H, CH2), 3.40-3.30 (m, 2 H, CH2), 1.45 (s, 6 H,
C(CH3)2. Anal. (C20H23N5O8) C, H, N.
A solution of 23 (3.39 g, 8.4 mmol) in trifluoroacetic acid
(15 mL) was stirred at 20 °C for 2 h and then poured slowly
into excess ice-cold 10% Na2CO3 solution. The resulting
mixture was extracted with EtOAc, washed twice with water,
and dried, and the organic layer was evaporated under reduced
pressure below 30 °C to provide 2-amino-N-[4-[bis(2-chloro-
ethyl)amino]phenyl]propanamide (24) (2.37 g, 93%) as an
unstable colorless oil, which was used directly: 1H NMR
[(CD3)2SO] δ 9.56 (br s, 1 H, CONH), 7.46 (d, J ) 9.1 Hz, 2 H,
H-2′,6′), 6.70 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 3.69 (t, J ) 3.7 Hz,
8 H, N(CH2CH2Cl)2), 3.36 (q, J ) 6.9 Hz, 1 H, CHCO), 1.19 (d,
J ) 6.9 Hz, 3 H, CH3).
[N-[2,6-Din itr o-4-[N-(2-h yd r oxyeth yl)ca r ba m oyl]p h en -
yl]m eth yla m in o]-N-(4-m eth oxyp h en yl)a ceta m id e (19d ).
Reaction of 18 with amine 9e by the above procedure for 2 h
gave a crude product which was purified by chromatography
on silica gel. Elution with CH2Cl2/EtOAc (1:3) followed by two
recrystallizations from MeOH/H2O containing a trace of AcOH
(the compound is extremely base-sensitive, giving deep-purple
solutions at pH > 7) gave 19d (63%): mp 98-101 °C; 1H NMR
[(CD3)2SO, 200 MHz] δ 9.68 (s, 1 H, CONHPh), 8.84 (t, J )
5.4 Hz, 1 H, PhCONH), 8.59 (s, 2 H, H-3,5), 7.46 (d, J ) 9.1
Hz, 2 H, H-2′,6′), 6.89 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 4.77 (t, J
) 5.6 Hz, 1 H, OH), 3.72 (s, 5 H, OCH3, CH2CO), 3.53 (q, J )
5.6 Hz, 2 H, CH2), 3.40-3.30 (m, 2 H, CH2), 2.89 (s, 3 H,
NCH3). Anal. (C19H21N5O8‚H2O) C, H, N.
2-[N-[2,6-Din itr o-4-[N-(2-h ydr oxyeth yl)car bam oyl]ph en -
yl]m eth ylam in o]-N-(4-m eth oxyph en yl)pr opan am ide (19e).
Reaction of 18 with amine 9d by the above procedure for 48 h
gave a crude product which was purified by chromatography
on silica gel. Elution with EtOAc followed by recrystallization
from MeOH/H2O containing a trace of AcOH (the compound
is extremely base-sensitive, giving deep-purple solutions at pH
> 7) gave 19e (74%): mp 192-193 °C; 1H NMR [(CD3)2SO,
200 MHz] δ 9.70 (s, 1 H, CONHPh), 8.83 (t, J ) 5.5 Hz, 1 H,
PhCONH), 8.85 (s, 2 H, H-3,5), 7.42 (d, J ) 9.0 Hz, 2 H,
H-2′,6′), 6.86 (d, J ) 9.1 Hz, 2 H, H-3′,5′), 4.75 (br s, 1 H, OH),
3.87 (q, J ) 6 7 Hz, 1 H, CHCH3), 3.71 (s, 3 H, OCH3), 3.58-
3.47 (m, 2 H, CH2), 3.41-3.38 (m, 2 H, CH2), 2.81 (s, 3 H,
NCH3), 1.27 (d, J ) 6.7 Hz, 3 H, CHCH3). Anal. (C20H23N5O8)
C, H, N.
N1-Hyd r oxy-7-[N-(2-h yd r oxyeth yl)ca r ba m oyl]-3-m eth -
yl-5-n itr o-3,4-d ih yd r oqu in oxa lin -2(1H)-on e (21) (Cycliza -
tion P r od u ct of 19b). Radiolytic reductions revealed that
ring-closed (dihydroquinoxaline) products underwent subse-
quent reductions when excess (>4-fold stoichiometry) reducing
equivalents were added. This propensity to undergo further
reductive reactions thwarted attempts to prepare them by
chemical reduction, and a sample had to be obtained by
purification of a radiolysis reaction. A stirred solution of 19b
(22 µM), 2-propanol (1 M), and acetone (0.3 M) in Milli-Q was
evacuated and irradiated for 4.5 h (60Co, 0.01 Gy s-1). The
resultant solution was assayed by HPLC to verify formation
of 4-methoxyaniline and a cyclized product. Solvents were then
removed by rotary evaporation under reduced pressure. The
mass spectrum of the resultant product mixture (FAB, sample
solubilized in 2-nitrobenzyl alcohol) gave an MH+ peak at 311,
A mixture of 24 (2.19 g, 7.2 mmol) and 4-chloro-3,5-
dinitrobenzoic acid (1.48 g, 6.00 mmol) in THF (12 mL) was
treated with diisopropylethylamine (1.71 g, 13.2 mmol) and
stirred at 20 °C for 12 h. The mixture was poured into excess
in aqueous AcOH, and the precipitated semisolid was taken
up in EtOAc. The organic solution was washed successively
with 1 N aqueous AcOH and water, then dried, concentrated
under reduced pressure below 30 °C to ca. 20 mL, and diluted
with petroleum ether. The resulting solid was chromato-
graphed on silica gel. EtOAc eluted N-[4-[bis(2-chloroethyl)-
amino]phenyl]-2-[(4-carboxy-2,6-dinitrophenyl)amino]propan-
amide (25) (2.19 g, 71%), which crystallized from EtOAc/
CH2Cl2/petroleum ether as green-brown prisms: mp 192-194
1
°C; H NMR [(CD3)2SO] δ 13.60 (br s, 1 H, CO2H), 10.11 (s, 1
H, CONH), 9.31 (d, J ) 6.9 Hz, 1 H, NHCH), 8.66 (s, 2 H,
H-3,5), 7.39 (d, J ) 9.1 Hz, 2 H, H-2′,6′), 6.73 (d, J ) 9.1 Hz,
2 H, H-3′,5′), 4.14 (quintet, J ) 6.8 Hz, 1 H, CHCO), 3.70 (s,
8 H, N(CH2CH2Cl)2), 1.36 (d, J ) 6.7 Hz, 3 H, CH3). Anal.
(C20H21Cl2N5O7) C, H, N, Cl.
A stirred solution of 25 (0.86 g, 1.67 mmol) in DMF (4 mL)
was treated with 1,1′-carbonyldiimidazole (0.30 g, 1.84 mmol)
at 25 °C for 1 h, then cooled to 0 °C, and treated with N,N-
dimethylethylenediamine (0.19 g, 2.16 mmol). The mixture was
allowed to warm to 25 °C over a 5-min period and was then
diluted with aqueous KHCO3. The resulting solid was purified
by chromatography on neutral Al2O3 (activity I). Elution with
CH2Cl2 removed nonpolar impurities. Further elution with
EtOAc/MeOH (20:1) gave an eluent which was concentrated
under reduced pressure below 30 °C to ca. 10 mL and then
diluted with petroleum ether to afford 26 (0.62 g, 62%): mp
(EtOAc/petroleum ether) 95-98 °C; 1H NMR [(CD3)2SO] δ
10.10 (s, 1 H, CHCONH), 9.13 (d, J ) 7.5 Hz, 1 H, NHCHCO),
8.78 (s, 2 H, H-3,5), 8.76 (t, J ) 5.6 Hz, 1 H, NHCH2), 7.38 (d,
J ) 9.1 Hz, 2 H, H-2′,6′), 6.73 (d, J ) 9.1 Hz, 2 H, H-3′,5′),
4.11 (quintet, J ) 6.8 Hz, 1 H, CHCO), 3.70 (s, 8 H, N(CH2-
CH2Cl)2), 3.28-3.42 (m, 2 H, NHCH2), 2.39 (t, J ) 6.7 Hz, 2
H, CH2N(CH3)2), 2.17 (s, 6 H, N(CH3)2), 1.35 (d, J ) 6.6 Hz, 3
H, CH3). Anal. (C24H31Cl2N7O6) C, H, N, Cl. Treatment with

354 J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3
Sykes et al.
EtOAc/HCl (1:1 equiv) followed by addition of petroleum ether
gave the hydrochloride salt.
>200 °C; 1H NMR [(CD3)2SO] δ 11.49 (s, 1 H, indole NH), 10.43
(s, 1 H, NHCOCH), 9.08 (t, J ) 7.6 Hz, 1 H, NHCH), 8.81 (s,
2 H, H-3′′′,5′′′), 8.78 (t, J ) 5.7 Hz, 1 H, CONHCH2), 8.59 (br
s, 1 H, H-4′), 8.03-7.94 (m, 2 H, H-6′,9′), 7.60 (t, J ) 7.6 Hz,
1 H, H-8′), 7.49 (t, J ) 7.6 Hz, 1 H, H-7′), 7.11 (d, J ) 1.7 Hz,
1 H, H-3′′), 6.98 (s, 1 H, H-4′′), 4.81 (t, J ) 10.1 Hz, 1 H, H-2′),
4.54 (dd, J ) 11.0, 1.8 Hz, 1 H, H-2′), 4.47-4.33 (m, 2 H, H-1′,
CHHCl), 4.08 (dd, J ) 11.1, 2.9 Hz, 1 H, CHHCl), 3.99-3.96
(m, 1 H, CHCH3), 3.93 (s, 3 H, OCH3), 3.82 (s, 3 H, OCH3),
3.80 (s, 3 H, OCH3), 3.38 (q, J ) 6.3 Hz, 2 H, NHCH2), 2.42 (t,
J ) 6.5 Hz, 2 H, NHCH2CH2), 2.19 (s, 6 H, N(CH3)2), 1.55 (dd,
J ) 6.5, 1.7 Hz, 3 H, CHCH3). Treatment of the free base with
N-[1-(Ch lor om et h yl)-3-[(5,6,7-t r im et h oxyin d ol-2-yl)-
ca r bon yl]-1,2-d ih yd r o-3H-b en z[e]in d ol-5-yl]-2-[[4-[N-[2-
(dim eth ylam in o)eth yl]car bam oyl]-2,6-din itr oph en yl]am i-
n o]p r op a n a m id e (32) (Sch em e 5). A mixture of 14 (0.59 g,
2.4 mmol), alanine 4-methoxybenzyl ester (27) (0.55 g, 2.6
mmol), and diisopropylethylamine (0.63 g, 4.9 mmol) in THF
(10 mL) was stirred at 25 °C for 30 h and then concentrated
under reduced pressure below 30 °C. The residue was shaken
with dilute HCl, and the resulting yellow solid was twice
recrystallized from EtOAc/iPr2O to give 4-methoxybenzyl 2-[(4-
carboxy-2,6-dinitrophenyl)amino]propanoate (28) (0.51 g,
51%): mp 156.5-157 °C; 1H NMR [(CD3)2SO] δ 13.6 (v br s, 1
H, CO2H), 8.60 (s, 2 H, H-3′,5′), 8.59 (d, J ) 8.4 Hz, partially
obscured, 1 H, NH), 7.25 (d, J ) 8.7 Hz, 2 H, H-2′′, 6′′), 6.89
(d, J ) 8.7 Hz, 2 H, H-3′′,5′′), 5.09 (d, J ) 11.9 Hz, 1 H, Ph
CHH), 5.02 (d, J ) 11.9 Hz, 1 H, PhCHH), 4.05-3.95 (m, 1 H,
CHCH3), 3.75 (s, 3 H, OCH3), 1.41 (d, J ) 7.0 Hz, 3 H, CHCH3).
Anal. (C18H17N3O9) C, H, N.
i
MeOH/EtOAc/HCl (1:2 equiv) followed by dilution with Pr2O
gave the hydrochloride salt: mp 210-215 °C dec. Anal. (C39H41
ClN8O10‚HCl‚3H2O) C, H, N, Cl.
-
Ra d iolytic Red u ction s. These were carried out by 60Co
γ-radiation (ca. 0.7 Gy s-1) in 10 mM sodium formate or
phosphate buffers at pH range 4.0 or 7.4, respectively, under
anaerobic conditions. Reducing conditions were realized by the
addition of either 1 M 2-propanol and 1 M acetone or 0.5 M
‚
2-propanol to the anaerobic buffer, to convert oxidizing OH
A mixture of 28 (252 mg, 0.60 mmol) and 1,1′-carbonyldi-
imidazole (117 mg, 0.72 mmol) in DMF (4 mL) was stirred at
20 °C for 1 h and then warmed to 50 °C for 5 min. The mixture
was cooled to 0 °C and treated with N,N-dimethylethylenedi-
amine (79 mg, 0.90 mmol). After being stirred at 20 °C for 5
min, the mixture was diluted with excess dilute KHCO3, and
the resulting oily precipitate was partitioned between CH2Cl2
and dilute KHCO3. The organic layer was washed with NaCl
solution, dried, and concentrated under reduced pressure. The
radicals (formed on γ-radiolysis of water) to the reducing
(CH3)2C‚OH radical. The 2-propanol/acetone mixed system was
used for compounds 19a and 17 to increase solubility. All
radiolytic reductions, and studies of subsequent reactions, were
carried out at room temperature. HPLC and spectrophotom-
etry was carried out immediately on removing samples from
the γ-source. Samples were introduced to the spectrophotom-
eter by pouring the solution through a sidearm into a pre-
attached quartz cuvette. The observed pseudo-first-order rate
coefficients (kobs) were determined using a modified Marquardt
iterative curve-fitting procedure.
HPLC assays employed an Econosphere C-18 column (5-
mm particle size, 250-mm × 4.6-mm i.d.), injecting 50 µL of
sample through a 20-µL sample loop. Both isocratic and
gradient elutions were used, with mobile phases comprising
methanol and either phosphate buffer (pH 6.5, 0.01 M) or
formate buffer (pH 4.0, 0.01 M) to assay radiolysis products
formed at pH 7.4 and 4.0, respectively, at a flow rate of 0.8
mL min-1. An HP1040M SeriesII diode array detector allowed
the eluent to be monitored at several analytical wavelengths
(internally blanked against 590 nm) as well as the collection
of spectra (in the range 200-400 nm).
Cell Cu ltu r e Gr ow th In h ibition Assa ys. Cell cultures
were initiated in 96-well plates in 50 µL of culture medium
(RMEM with 5% fetal bovine serum, 100 units/mL penicillin
and 100 µg/mL streptomycin) at 200 (AA8), 300 (UV4), 75
(EMT6), or 600 (SKOV3) cells/well and grown for 24 h before
exposure to drugs. Drug stock solutions were prepared in
DMSO and diluted into culture medium (final DMSO concen-
tration < 1%), and the pH was adjusted to 7.4 if necessary
under 5% CO2. Cells were exposed under aerobic conditions
for 4 h, drugs were removed by washing cultures three times
with fresh medium, and cultures were incubated in 150 µL of
medium for a further 4-5 days before staining with methylene
blue.26 IC50 values were calculated as the drug concentration
providing 50% inhibition of growth relative to controls on the
same plate.
Deter m in a tion of Hyp oxia -Selective Cytotoxicity in
Vitr o by Clon ogen ic Assa y. UV4 cells were grown to early
plateau phase (5 × 105 cell/mL) in spinner flasks and resus-
pended in fresh medium. Cell suspensions and drug solutions
in growth medium were preequilibrated with magnetic stirring
under 5% CO2 in air or N2 for 1 h prior to mixing to ensure
essentially complete anoxia throughout the period of drug
contact in hypoxic cultures. After exposure of cells to drug for
1 h at 106/mL with continuous gassing, clonogenic assays were
performed as detailed elsewhere.27 The ratios of the concentra-
tion for a surviving fraction of 10% (C10) for the aerobic and
hypoxic survival curves were used as the measure of hypoxic
selectivity.
i
residue was triturated with Pr2O, providing a solid that was
twice recrystallized from CH2Cl2/EtOAc to give 4-methoxy-
benzyl 2-[[4-[N-[2-(dimethylamino)ethyl]carbamoyl]-2,6-dini-
trophenyl]amino]propanoate (29) (218 mg, 74%): mp 107-109
°C; 1H NMR [(CD3)2SO] δ 8.76 (t, J ) 5.5 Hz, partially
obscured, 1 H, CONH), 8.74 (s, 2 H, H-3′,5′), 8.45 (d, J ) 8.4
Hz, 1 H, PhNH), 7.24 (d, J ) 8.6 Hz, 2 H, H-2′′,6′′), 6.89 (d, J
) 8.6 Hz, 2 H, H-3′′,5′′), 5.08 (d, J ) 11.9 Hz, 1 H, Ph CHH),
5.02 (d, J ) 11.9 Hz, 1 H, Ph CHH), 4.02 (m, 1 H, CHCH3),
3.75 (s, 3 H, OCH3), 3.37 (q, J ) 6.3 Hz, 2 H, NHCH2), 2.40 (t,
J ) 6.7 Hz, 2 H, NHCH2CH2), 2.18 (s, 6 H, N(CH3)2), 1.40 (d,
J ) 6.9 Hz, 3 H, CHCH3). Anal. (C22H27N5O8) C, H, N.
The benzyl ester 29 (1.75 g, 3.58 mmol) was treated with
anisole (0.58 mL) followed by ice-cold CF3CO2H (15 mL). The
resulting solution was stirred at 20 °C for 15 min and then
concentrated under reduced pressure below 30 °C. The residue
i
was triturated repeatedly with Pr2O to provide a semisolid
which was dissolved in EtOAc. The solution was clarified by
filtration, cooled to -5 °C, treated with dry HCl/EtOAc, and
then diluted with iPr2O. The precipitated yellow solid was
collected and washed with EtOAc/iPr2O to give 2-[[4-[N-[2-
(dimethylamino)ethyl]carbamoyl]-2,6-dinitrophenyl]amino]-
propanoic acid hydrochloride (30) (1.34 g, 92%) as a hygro-
scopic solid, suitable for further use. A sample recrystallized
from MeOH/EtOAc/iPr2O had mp 92-96 °C; 1H NMR [(CD3)2-
SO] δ 13.5 (v br s, 1 H, CO2H), 10.35 (s, 1 H, N+H), 9.23 (t, J
) 5.5 Hz, 1 H, CONH), 8.83 (s, 2 H, H-3′,5′), 8.76 (d, J ) 7.7
Hz, 1 H, Ph NH), 3.92 (quin, J ) 7.1 Hz, 1 H, CHCH3), 3.65
(q, J ) 5.2 Hz, 2 H, CONHCH2), 3.27 (q, J ) 5.5 Hz, 2 H,
CH2N+H(CH3)2), 2.81 (d, J ) 4.5 Hz, 6 H, N(CH3)2), 1.36 (d, J
) 6.8 Hz, 3 H, CHCH3). Anal. (C14H20ClN5O7‚H2O) C, H, N,
Cl.
A mixture of 30 (172 mg, 0.42 mmol), 5-amino-1-(chloro-
methyl)-3-[(5,6,7-trimethoxyindol-2-yl)carbonyl]-1,2-dihydro-
3H-benz[e]indole20 (31) (165 mg, 0.35 mmol) and EDCI‚HCl
(201 mg, 1.05 mmol) in DMA (2 mL) was stirred at 20 °C for
4 h and then diluted with dilute KHCO3. The precipitated solid
was collected, washed with water, then extracted with 0.1 N
AcOH, and filtered. The clarified solution was treated with
solid KHCO3 and extracted twice with EtOAc. The combined
extracts were washed with water, dried (Na2SO4), concentrated
to a small volume under reduced pressure below 30 °C, and
i
then diluted with Pr2O. The resulting product was purified
Deter m in a tion of Cytotoxicity of Ra d iolytica lly Acti-
va ted P r od r u gs. Compounds were dissolved in culture
medium containing 5% fetal bovine serum and deoxygenated
by precipitation from a MeOH/EtOAc solution at 20 °C with
iPr2O to give 32 (130 mg, 45%) as a light-sensitive solid: mp

N-Substituted 2-(Dinitrophenylamino)propanamides
J ournal of Medicinal Chemistry, 1999, Vol. 42, No. 3 355
antibody-directed enzyme-prodrug therapy (ADEPT). J . Med.
Chem. 1995, 38, 5051-5065.
by stirring under a stream of 5% CO2 in N2. Solutions were
irradiated (2.2 Gy min-1 60Co) and immediately frozen for
,
(7) Hay, M. P.; Denny, W. A. Antibody-directed enzyme prodrug
therapy (ADEPT). Drugs Future 1996, 21, 917-931.
(8) Trinh, Q. T.; Austin, E. A.; Murray, D. M.; Knick, V. C.; Huber,
B. E. Enzyme/prodrug gene therapy: comparison of cytosine
deaminase/5-fluorocytosine versus thymidine kinase/ganciclovir
enzyme/prodrug systems in a human colorectal carcinoma cell
line. Cancer Res. 1995, 55, 4808-4812.
subsequent bioassay against UV4 cells using the above growth
inhibition assay. The IC50 of authentic effector was determined
in the same experiments. The concentration of cytotoxic
effector, [E], was calculated from the change in apparent IC50
of the irradiated solution using the following relationship:
IC50,E
IC50
IC50,E
(9) Denny, W. A.; Wilson, W. R.; Hay, M. P. Recent developments
in the design of bioreductive drugs. Br. J . Cancer, Suppl. 1996,
74, 281-294.
× [P]0
-
× [P]0
(
) (IC
)
50,P
[E] )
(10) Atwell, G. J .; Sykes, B. M.; O’Connor, C. J .; Denny, W. A.
Relationships between structure and kinetics of cyclization of
2-aminoaryl amides: Potential prodrugs of cyclization-activated
aromatic mustards. J . Med. Chem. 1994, 37, 371-380.
(11) Sykes, B. M.; Atwell, G. J .; Denny, W. A.; McLennan, D. J .;
O’Connor, C. J . Kinetics and mechanism of the cyclization of
substituted N-phenyl-2-methyl-2-(2-aminophenyl)-propanamides
and analogues. J . Chem. Soc., Perkin Trans. 2 1995, 337-342.
(12) Sykes, B. M.; Atwell, G. J .; Denny, W. A.; O’Connor, C. J .
Radiolytic studies of reductive cyclization of 2-nitroarylamides:
cyclization via hydroxylamine intermediates. J . Phys. Org.
Chem. 1995, 8, 587-596.
(13) Hay, M. P.; Wilson, W. R.; Denny, W. A. Nitroimidazole-based
“extruded mustards” designed as reductively activated hypoxia-
selective cytotoxins. Anti-Cancer Drug Des. 1996, 11, 383-402.
(14) Amsberry, K. L.; Borchardt, R. T. The lactonization of 2′-
hydroxyhydrocinnamic acid amides: A potential prodrug for
amines. J . Org. Chem. 1990, 55, 5867-5877.
(15) Amsberry, K. L.; Borchardt, R. T. Amine prodrugs which utilize
hydroxy amide lactonization. I. A potential redox-sensitive amide
prodrug. Pharm. Res. 1991, 8, 323-330.
(16) Kirk, K. L.; Cohen, L. A. Participation of the anilino group in
peptide bond cleavage. The use of tert-butyl 3,5-dinitro-2-
fluorocarbanilate as a peptide reagent. J . Org. Chem. 1969, 34,
395-397.
(17) Kirk, K. L.; Cohen, L. A. Intramolecular aminolysis of amides.
Effects of electronic variation in the attacking and leaving
groups. J . Am. Chem. Soc. 1972, 94, 8142-8147.
(18) J ohnstone, R. A. W.; Povall, T. J .; Entwistle, I. D. Methods of
peptide sequencing. Part II. Cyclisation of N-2-amino-6-nitro-
phenyl and N-3-amino-2-pyridyl derivatives of amino acids and
peptides. J . Chem. Soc., Perkin Trans. I 1975, 1424-1427.
(19) Palmer, B. D.; Wilson, W. R.; Pullen, S. M.; Denny, W. A.
Hypoxia-selective antitumor agents. 3. Relationships between
structure and cytotoxicity against cultured tumor cells for
substituted N,N-bis(2-chloroethyl)anilines. J . Med. Chem. 1990,
33, 112-121.
(20) Atwell, G. J .; Wilson, W. R.; Denny, W. A. Synthesis and
cytotoxicity of amino analogues of the potent DNA alkylating
agent seco-CBI-TMI. Bioorg. Med. Chem. Lett. 1997, 7, 1493-
1496.
IC50,E
1 -
IC50,P
where IC50,E and IC50,P are the IC50 values of the pure effector
and (unirradiated) prodrug, respectively, and IC50 is the IC50
value of the irradiated mixture, expressed relative to the initial
concentration of the prodrug, [P]0. This calculation assumes
that the cytotoxicity of prodrug and effector is independent
and additive. Full details of this testing methodology, including
derivation of the above equation, will be reported in full
elsewhere.25
In Vivo Testin g. All studies used C3H/HeN mice bred in
this institution, and all experiments were approved under The
University of Auckland Animal Ethics Committee. Compounds
were formulated in dimethylacetamide, poly(ethylene glycol)
(MW 400), and water (10/40/50, v/v/v) and administered ip at
0.01 mL/g of body weight. Toxicity was evaluated using 1.33-
fold dose increments, with an observation time of 60 days. The
maximum tolerated dose was defined at the highest dose
causing no deaths or significant morbidity in a group of 6
animals. Evaluation of antitumor activity was performed
essentially as described by Twentyman et al.28 Briefly, RIF-1
tumors were grown im to 0.5 g (leg + tumor diameter 10 mm),
and mice were treated with single doses of drug and/or
radiation (15 Gy, 60Co, 2 Gy min-1), the latter by restraining
unanesthetized mice in boxes and extending the leg into the
radiation field using a clip to tether the ankle. Tumors were
excised 18 h after the last treatment, dissociated enzymati-
cally, and plated in vitro to determine the number of surviving
clonogenic cells/tumor.
Ack n ow led gm en t. B.M.S. gratefully acknowledges
the financial support of an Auckland University Re-
search Committee Postdoctoral Fellowship. Support was
also provided by the Auckland Division of the Cancer
Society of New Zealand and the Health Research
Council of New Zealand. We are grateful to Dr. R. F.
Anderson, Department of Chemistry, The University of
Auckland, for determination of the reduction potential
of compound 19a . Biological testing was supported by
Contract NO1-CM 47019 from the National Cancer
Institute, Bethesda, MD.
(21) Hoy, C. A.; Salazar, E. P.; Thompson, L. H. Rapid detection of
DNA-damaging agents using repair-deficient CHO cells. Mutat.
Res. 1984, 130, 321-332.
(22) Tercel, M.; Denny, W. A.; Wilson, W. R. Nitrogen and sulfur
analogues of the seco-CI alkylating agent: synthesis and cyto-
toxicity. Bioorg. Med. Chem. Lett. 1996, 6, 2735-2740.
(23) Boger, D. L.; J ohnson, D. S. CC-1065 and the duocarmycins:
understanding their biological function through mechanistic
studies. Angew. Chem., Int. Ed. Engl. 1996, 35, 1438-1474.
(24) Siim, B. G.; Wilson, W. R. Efficient redox cycling of nitroquinoline
bioreductive drugs due to aerobic nitroreduction in Chinese
Hamster cells. Biochem. Pharmacol. 1995, 50, 75-82.
(25) Wilson, W. R.; Moselen, J . M.; Tercel, M.; Denny, W. A.;
Anderson, R. F. Nitroarylmethyl quaternary (NMQ) prodrugs
as radiation-activated cytotoxins: evaluation of nitrogen mus-
tard release using a bioassay method. Manuscript in prepara-
tion.
(26) Finlay, G. J .; Baguley, B. C.; Wilson, W. R. A semiautomated
microculture method for investigating growth inhibitory effects
of cytotoxic compounds on exponentially growing carcinoma cells.
Anal. Biochem. l984, 139, 272-277.
(27) Siim, B. G.; Atwell, G. J .; Wilson, W. R. Oxygen dependence of
the cytotoxicity and metabolic activation of 4-alkylamino-5-
nitroquinoline bioreductive drugs. Br. J . Cancer 1994, 70, 596-
603.
Refer en ces
(1) Wilson, W. R. Tumour hypoxia: challenges for cancer chemo-
therapy. In Cancer Biology and Medicine; Waring, M. J ., Ponder,
B. A. J ., Eds.; Kluwer Academic Publishers: Lancaster, 1992;
Vol. 3, pp 87-131.
(2) Rockwell, S. Use of hypoxia-directed drugs in therapy of solid
tumors. Semin. Oncol. 1992, 19, 29-40.
(3) Workman, P.; Stratford, I. J . The experimental development of
bioreductive drugs and their role in cancer therapy. Cancer
Metastasis Rev. 1993, 12, 73-82.
(4) Brown, J . M.; Siim, B. G. Hypoxia specific cytotoxins in cancer
therapy. Semin. Radiat. Oncol. 1996, 6, 22-36.
(5) (a) Wilson, W. R.; Ferry, D. M.; Tercel, M.; Anderson, R. F.;
Denny, W. A. Reduction of nitroarylmethyl quaternary am-
monium prodrugs of mechlorethamine by radiation. Radiat. Res.
1998, 149, 237-246.
(28) Twentyman, P. R.; Brown, J . M.; Gray, J . W.; Franko, A. J .;
Scoles, M. A.; Kallman, R. F. A new mouse tumor model system
(RIF-1) for comparison of end-point studies. J . Natl. Cancer Inst.
1980, 64, 595-604.
(6) Springer, C. J .; Dowell, R.; Burke, P. J .; Davies, H. D.; Blakey,
D. C.; Melton, R. G.; Niculescu-Duvaz, I. Optimization of
alkylating agent prodrugs derived from phenol and aniline
mustards:
a
new clinical candidate prodrug (ZD2767) for
J M960783S