Triazoloacridin-6-ones as novel inhibitors of the quinone oxidoreductases NQO1 and NQO2

Article abstract of DOI:10.1016/j.bmc.2009.11.059

A range of triazoloacridin-6-ones functionalized at C5 and C8 have been synthesized and evaluated for ability to inhibit NQO1 and NQO2. The compounds were computationally docked into the active site of NQO1 and NQO2, and calculated binding affinities were

Full text of DOI:10.1016/j.bmc.2009.11.059

Bioorganic & Medicinal Chemistry 18 (2010) 696–706
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Triazoloacridin-6-ones as novel inhibitors of the quinone oxidoreductases
NQO1 and NQO2
Karen A. Nolan a, Matthew P. Humphries a, John Barnes a,b, Jeremy R. Doncaster a,b, Mary C. Caraher a,
Nicola Tirelli a, Richard A. Bryce a, Roger C. Whitehead b, Ian J. Stratford a,
*
a School of Pharmacy and Pharmaceutical Sciences, University of Manchester and Manchester Cancer Research Center, Manchester M13 9PT, UK
b School of Chemistry, University of Manchester and Manchester Cancer Research Center, Manchester M13 9PT, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
A range of triazoloacridin-6-ones functionalized at C5 and C8 have been synthesized and evaluated for
ability to inhibit NQO1 and NQO2. The compounds were computationally docked into the active site of
NQO1 and NQO2, and calculated binding affinities were compared with IC50 values for enzyme inhibition.
Excellent correlation coefficients were demonstrated suggesting a predictive QSAR model for this series
of structurally similar analogues. From this we have identified some of these triazoloacridin-6-ones to be
the most potent NQO2 inhibitors so far reported.
Received 16 September 2009
Revised 19 November 2009
Accepted 27 November 2009
Available online 21 December 2009
Keywords:
NQ01
Ó 2009 Elsevier Ltd. All rights reserved.
NQ02
Structure–activity relationship
Molecular modelling
Toxicity
1. Introduction
shown to activate the dinitrobenzene CB1954, which has led to
the renewed evaluation of this drug in cancer chemotherapy.14
The oxidoreductases, NAD(P)H:quinone oxidoreductase
1
A common property of the two enzymes is their apparent ability
to protect the tumour suppressor p53 against proteosomal degra-
dation, which leads to stabilization and activation of p53.15 Inhibi-
tion of NQO1 enzyme activity leads to destabilization of p53 and
similar NQO1-mediated effects have been observed for other
short-lived proteins including ornithine decarboxylase and
(NQO1, DT-diaphorase, EC 1.6.99.2) and NRH:quinone oxidoreduc-
tase 2 (NQO2, EC 1.10.99.2), are homodimeric, cytosolic flavopro-
teins that catalyse the two electron reduction of quinones to
hydroquinones.1,2 Human NOQ1 and NQO2 share 54% and 49%
cDNA and protein sequence identity, respectively.3 Although both
enzymes are expressed in many of the same tissues throughout
the body, the highest expression of NQO1 is found in kidney, gut
epithelium4 and importantly, in many types of solid tumour
including cancers of the colon, breast, lung and liver.5 In contrast
to NQO1, the NQO2 level in normal and malignant tissues is gener-
ally lower.6 However, there are some cell types such as haemopoi-
etic cells, where NQO2 is present and NQO1 is much reduced.7,8
Substrate specificity is different in the two enzymes, with NQO1
efficiently reducing 1,4-quinones,9 whereas NQO2 can reduce
ortho-quinones10 and aromatic nitro compounds11 as well as bind-
ing tyrosine kinase inhibitors such as imatinib.12 The reductase
properties of the enzymes have been exploited in cancer therapy
with NQO1 being used to selectively activate quinone-based drugs
such as mitomycin C, EO9 and RH1.13 In contrast, NQO2 has been
p73a 15–17
been shown to alter signaling via the NF-
.
Additional effects of modulating NQO2 activity have
j
B pathway.18 Thus, iden-
tification of novel, potent inhibitors of NQO1 and/or NQO2 could
provide routes to gaining a better understanding of the cellular
roles of these enzymes.
We recently applied a virtual screening strategy of the National
Cancer database to identify novel inhibitors of NQO1.19,20 This ap-
proach resulted in the discovery of the triazoleacridin-6-one,
NSC645827, as a submicromolar inhibitor of NQO1.19 We report
here a study where we have synthesised a series of compounds
structurally related to NSC645827 and evaluated their ability to in-
hibit NQO1. In addition, we have tested the ability of this series of
compounds to inhibit the function of NQO2. We have further used
computational molecular docking to identify structure/activity
relationships (SAR) for inhibition of both NQO1 and NQO2 by these
triazoloacridin-6-ones. We have then evaluated the compounds for
their ability to interact with DNA and related this to their toxicity
towards HCT116 colon cancer cells that show significant activity of
both NQO1 and NQO2.
* Corresponding author. Tel.: +44 (0) 161 275 2487; fax: +44 (0) 161 275 8342.
E-mail address: ian.stratford@manchester.ac.uk (I.J. Stratford).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.11.059

K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
697
2. Chemistry
-44
y = 0.4138x - 32.498
R2 = 0.87
Synthesis of a range of triazoloacridin-6-ones (Scheme 1),
wherein there are two loci for diversification (positions 5 and 8),
was carried out using an approach modelled on that described by
Cholody et al.23 The key intermediates in the syntheses were the
1-chloro-4-nitro acridones 2a–d which were prepared using gen-
eral methods detailed in the literature.21,22 Reduction of the ni-
tro-group of compounds 2a–d was readily accomplished using
stannous chloride in ethanolic HCl to furnish the corresponding
4-amino acridones 3a–d in good yield. Exposure of the latter com-
pounds to sodium nitrite in concd HCl at room temperature re-
sulted in sequential diazotisation and cyclisation to give the
triazoloacridin-6-ones 4a–d. In the first instance, nucleophilic
substitution of the chlorine atom at C5 of compounds 4a–d was
carried out, at slightly elevated temperature, using N,N-dimethyl-
amino-ethylenediamine (5) in dimethylacetamide to give the
5-amino-triazoloacridin-6-ones 5a–d. Subsequent oxidation of
the terminal side-chain nitrogen was then accomplished using
mCPBA in chloroform to give the N-oxides 10a–d. Similarly, the
remaining analogues (6a–d, 7a–d, 8a–d, 9a–d), bearing a variety
of different amino substituents at C5, were prepared via exposure
of compounds 4a–d to the appropriate amine (6–9) at elevated
temperature in dimethylacetamide.
-45
-46
-47
-48
-49
-50
-42
-40
-38
-36
-34
-32
-30
Δ
Gexp
-46
-48
-50
-52
-54
y = 0.8074x - 20.579
R2 = 0.93
3. Molecular modeling, biochemical and biological evaluation
Fifteen compounds from this series were assayed for their abil-
ity to inhibit purified recombinant human NQO1 and NQO2
(Table 1). The selection criterion for these compounds was based
solely on their solubility.
The compounds were computationally docked into the active
site of NQO1 and NQO2 using the X-ray crystal structures of hu-
-42
-40
-38
-36
-34
-32
Δ
Gexp
man NQO1 with bound FAD (PDB code 1D4A, resolution at
0
Figure 1. Calculated (DGcalc) and experimental (DGexp) binding affinities (kJ/mol)
1.7 ÅA)25 and human NQO2 with bound FAD (PDB code 1QR2, reso-
lution at 2 Å),26 respectively. After docking, the calculated binding
affinities (using the ChemScore27 scoring function in GOLD 4.1) and
experimental binding affinities, derived from IC50 values,28 were
compared for each enzyme (Table 1). Both enzymes demonstrate
excellent correlation coefficients of R2 = 0.87 for NQO1 and
R2 = 0.93 for NQO2 (Fig. 1A and B) suggesting a predictive QSAR
model for this series of structurally similar analogues.
are shown for (top) NQO1 as white circles and (bottom) for NQO2 as black circles.
D
Gexp is calculated from IC50 values using the Cheng–Prusoff equation.28
for NQO1, there is a more restricted range of calculated binding
affinities (5 kJ/mol) compared to experiment (9 kJ/mol). This is re-
flected in the near twofold change in the slope of the regression
lines for the two enzymes shown in Figure 1. This narrower calcu-
lated range of affinities for NQO1 may reflect the more limited ste-
ric fit of congeners modeled into the NQO1 active site compared to
NQO2 (see below). Furthermore, compounds that are good inhibi-
For NQO2, calculated binding affinities (
mental binding affinities derived from the observed IC50s (
D
Gcalc) and the experi-
Gexp
D
)
span similar ranges, of 7 kJ/mol in both cases (Table 1). However,
Table 1
Computationally-derived (DGcalc) and experimental (DGexp) binding affinities (kJ/mol) and experimental IC50 values (nM) SD for the interaction of the triazoloacridin-6-ones
with NQO1 and NQO2
ID
D
Gcalc NQO1
D
Gexp NQO1
NQO1 (IC50
)
D
Gcalc NQO2
D
Gexp NQO2
NQO2 (IC50
)
Ratio of IC50 NQO1/NQO2
5a
5b
5c
5d
6a
6c
6d
7c
8c
8d
9a
9c
ꢀ49.2
ꢀ49.2
ꢀ46.3
ꢀ47.5
ꢀ48.1
ꢀ47.9
ꢀ48.2
ꢀ47.4
ꢀ49.0
ꢀ49.0
ꢀ44.3
ꢀ45.5
ꢀ48.5
ꢀ48.7
ꢀ48.2
ꢀ39.9
ꢀ38.6
ꢀ33.0
ꢀ35.1
ꢀ38.4
ꢀ38.4
ꢀ39.3
ꢀ35.0
ꢀ40.4
ꢀ40.0
ꢀ31.1
ꢀ31.5
ꢀ37.3
ꢀ40.4
ꢀ36.2
111 25
188 10
1800 10
780 53
ꢀ48.7
ꢀ48.7
ꢀ47.7
ꢀ46.6
ꢀ49.6
ꢀ49.5
ꢀ48.7
ꢀ52.8
ꢀ49.4
ꢀ49.6
ꢀ50.5
ꢀ49.3
ꢀ53.9
ꢀ51.7
ꢀ53.0
ꢀ33.8
ꢀ35.1
ꢀ33.0
ꢀ33.5
ꢀ36.1
ꢀ36.3
ꢀ34.5
ꢀ40.3
ꢀ36.6
ꢀ35.1
ꢀ37.1
ꢀ35.9
ꢀ40.3
ꢀ38.9
ꢀ39.8
1333 577
783 375
1833 289
1483 525
517 126
483 104
987 121
98 13
427 108
767 153
350 132
567 208
98 10
0.08
0.24
0.98
0.53
0.40
0.42
0.15
8.24
0.22
0.14
11.24
5.79
3.35
0.56
4.33
206
203
5
6
145 59
808
92
8
3
107 15
3933 51
3283 29
10a
10c
10d
328
94
507
3
5
6
167 42
117 29
All experimental values are the mean of at least three independent experiments done in triplicate. Compounds are ordered in the table according firstly to substituent B and
then substituent A.

698
K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
tors of NQO1 are not necessarily good inhibitors of NQO2. This is
illustrated by comparing the ratio of IC50 for inhibition of NQO1
versus NQO2 for each compound (Table 1). For example, 5a is
one of the most potent inhibitors of NQO1 whereas it is a relatively
weak inhibitor of NQO2. In contrast, 9a is the poorest inhibitor of
NQO1 while it shows reasonable activity towards NQO2.
The docking calculations predicted only one dominant binding
mode for NQO1: the triazoloacridin-6-one scaffold is held in the
active site of NQO1 by the formation of several strong hydrogen
bonds (Fig. 2). Using 5a as an example (Fig. 2A), the Od of Tyr126
forms a hydrogen bond with either N1 (2.6 Å) or N2 (2.4 Å) of
the triazole ring and the OH of Tyr1280 forms a strong hydrogen
bond with the ketone group of C6 (1.5 ÅA). The hydrogen atom on
compound has a methoxy substituent at C8 which is able to form
0
an additional polar contact with His161 (3.1 ÅA, Fig. 2A). In addition
to the lipophilic effect of the methoxy group with Phe 178, this
may account for the 17-fold increased inhibitory potency over 5c
which has an IC50 of 1.8 lM and a H atom at C8. The methoxy sub-
stituent on 6a is unable to make this hydrogen bond contact as a
consequence of forming a polar contact with Gly149 (compare
6c, Fig. 2B). Likewise, 10a is predicted to form a polar contact with
the FAD side chain (compare 10c, Fig. 2C) resulting in a slight twist
of the pharmacophore scaffold. Addition of fluorine (5b) and bro-
mine (5d, 6d, 8d (Fig. 2D) and 10d) substituents at C8 is well tol-
erated allowing favourable lipophilic active site interactions.
Despite the active sites of NQO1 and NQO2 being comparable
(see Supplementary data), there are three notable variations: the
Tyr126, Tyr128 and Met131 residues are replaced in NQO2 with
Phe126, Ile128 and Phe131. This makes the NQO2 active site
slightly larger and more hydrophobic than in NOQ1. These differ-
ences and the lack of a C-terminal domain may be part of the rea-
the N5 side chain of the ligand also makes a hydrogen bond contact
0
with the Od of Tyr128 (2.9 ÅA). This proton may also form an inter-
nal hydrogen bond with C6 if the N5 side chain is rotated. The li-
gand is positioned in the centre of the NQO1 binding pocket by
p
stacking interactions with the isoalloxazine ring of FAD, Trp105
and Phe178 (Fig. 2A). Thus, all ligands are able to make good
hydrophobic contacts with active site residues and this is reflected
in the lipophilic contribution to the docking score. However, the
calculated binding affinity is compromised by clash penalties ap-
plied to the docking score as a consequence of steric hindrance
with Phe106 forming part of internal wall of the NQO1 active site.
Additional clashes are also observed with ligands which have a
phenyl side chain (7c, 9a and 9c) forming unfavourable interac-
tions with the glycerol moiety of FAD, Gly149 and Gly150. The
N5 side chain has to project up and out of the active site and this
is largely parallel to the FAD cofactor. This steric stress is reflected
in both a lower calculated binding affinity and a reduced inhibitory
capacity for these compounds (Table 1).
son why NQO2 requires
a
different cofactor to NQO1.11,29
Furthermore, they are proposed to account for the observed differ-
ences in substrate and inhibitor specificity between the two
enzymes.
In contrast to NQO1, the docking studies suggest that triazolo-
acridin-6-ones are able to bind into the active site of NQO2 in sev-
eral energetically favourable ways. Compounds 5a (Fig. 3A), 5b and
5c do not form any polar contacts with active site residues and
seem to be held in the binding pocket by lipophilic interactions
with the isoalloxazine ring of FAD, Trp105, Phe106, Phe126,
Ile128 and Phe178. These ligands form an internal hydrogen bond
between the N5 proton and the ketone oxygen at C6 (2.0 Å,
Fig. 3A); this limits rotation of the N5 side chain and holds the tri-
azole scaffold in a planar conformation. The N-oxides, 10a (Fig. 3B),
10c and 10d are the most potent inhibitors of NQO2 with IC50
Compound 5a is a potent inhibitor of NQO1 in with an IC50 of
111 nM and a calculated binding affinity of ꢀ49.2 kJ/mol. This
Cl
O
Cl
O
Cl
Cl
O
R1
R1
CO2H
R1
ii)
iii)
i)
OH
OH
NH2
a : R1 = OMe
b : R1 = F
c : R1 = H
d : R1 = Br
N
H
N
H
Cl
Cl
NO2
NO2
NO2
1a-d
2a-d
iv)
N
N
O
Cl
O
Cl
O
NH
O
NH
O
R1
R1
R1
R1
v)
vi)
vi)
vii)
N
H
N
N
N
N
N
N
NH2
N
N
N
4a-d
3a-d
5a-d
10a-d
R2
O
R1
N
N
N
6a-d: R2 = NH(CH2)2N(CH3)2
7a-d: R2 = NHPh-p-OCH3
8a-d: R2 = NH(CH2)2CH(CH3)2
9a-d: R2 = NH(CH2)2Ph
Scheme 1. General synthesis of the triazoloacrid-6-ones. Reagents and conditions: (i) H2SO4, HNO3, 70 °C, 1 h21; (ii) N,N-dimethylaniline, 110 °C, 18 h, N222; (iii)
dichloroethane, N,N-dimethylaniline, POCl3, 70 °C, 1 h22,23; (iv) concd HCl, EtOH, SnCl2, reflux, 6 h23; (v) NaNO2, concd HCl, 4 h, rt23; (vi) dimethylacetamide, amine 5–9, 60 °C,
4 h23; (vii) mCPBA, CHCl3, 0 °C, 24 h.24

K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
699
Gly150
His161
Gly150
His161
2.0
Tyr128
Tyr128
Phe178
3.1
1.5
3.0
Phe178
1.5
5a
2.9
Gly149
Gly149
6c
Tyr126
FAD
2.6
Tyr126
2.6
FAD
Phe106
2.5
2.4
Phe106
A
B
Trp105
Trp105
Gly150
His161
His161
Gly150
Tyr128
Gly149
Tyr128
Phe178
1.5
2.9
1.5
Phe178
Gly149
FAD
2.0
10c
8d
2.3
3.0
2.7
Tyr126
Tyr126
FAD
Phe106
Phe106
Trp105
Trp105
C
D
Figure 2. Compounds (A) 5a, (B) 6c (C) 10c, (D) 8d docked into the active site of NQO1. Polar contacts are shown as dashed lines and
p–p
stacking interactions are shown as
0
solid arrows in A. Distances are given in ÅA.
values of 98, 167 and 117 nM, respectively. The N5 proton forms a
hydrogen bond with Glu122 (2.5 Å) which causes the N5 side chain
to rotate and the oxygen forms a hydrogen bond with a hydrogen
on the FAD chain (2.7 Å, Fig. 3B). Compounds 7c (Fig. 3C), and 9c
(Fig. 3D) fit into the binding pocket without any clash penalties
arising from unfavourable interactions. The methoxy group on 7c
(Fig. 3C) forms a close hydrogen bond contact with Thr71 (2.4 Å)
which causes the acridine scaffold to rotate slightly. The triazoloac-
ridin-6-ones with fluorine and bromine substituents at C8 do not
appear to play any part in affecting potency suggesting that elec-
tronic interactions are not important in binding to NQO2.
We have measured the toxicity of the triazoloacridin-6-ones in
HCT116 colon cancer cells. Examples of dose response curves for
toxicity are given in Figure 4 and from data like this we have de-
rived values of IC50, the concentration required to reduce cell sur-
vival by 50% relative to untreated controls, for each of the
compounds. The values of IC50 obtained following either 24 or
96 h drug exposure are given in Table 2. These cells show signifi-
cant activity of both NQO1 and NQO2 (1076 70 and
493 74 nmol DCPIP reduced/min/mg protein). At first sight there
is no obvious relationship between inhibitory potency towards
either, or both enzymes, and toxicity. Thus it is likely that there
are other underlying mechanisms of toxicity.
agents have shown significant anti-tumour activity in pre-clinical
models.23,30 The planar structure of the triazoloacridin-6-ones to-
gether with the known mechanism of action of related imidazo-
acridinones and anthracenediones, strongly suggests that
interaction with DNA could be the underlying mechanism of toxic-
ity of the triazoloacridin-6-ones in cells. Thus, we evaluated the
ability of the compounds to bind to DNA by measuring changes
in the melting temperature, Tm, of salmon sperm DNA (20
lg/ml)
when treated with 10 M of each of the triazoloacridin-6-ones.
l
Values of Tm obtained at two different salt concentrations are given
in Table 2. Consistent with recent studies by Koba and Kanopa31
and Lemke et al.32 we show that the triazoloacridin-6-ones cause
a change in DNA melting temperature, Tm, but only at low salt con-
centrations. At physiologically relevant salt concentrations, little if
any change in Tm is observed. Some examples of the DNA melting
curves are given in Figure 5. In the figure, 10 lM adriamycin is
used as a positive control and under these low-salt conditions adri-
amycin results in an 8 °C change in Tm. Nearly all the triazoloacri-
din-6-ones (except the 10 series) show changes in Tm, but the
maximum change is no more than only 4 °C, suggesting they have
only weak intercalating ability. Koba and Kanopa31 went on to
show, using DNA viscometry, that intercalation in DNA was unli-
kely to be the major mechanism of toxicity of these compounds.
Some evidence for DNA cross-linking by the triazoloacridin-6-ones
was demonstrated but this did not correlate with biological activ-
A substantial number of triazoloacridin-6-ones have already
been synthesized by Konopa and co-workers and some of these

700
K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
Gly150
Asn161
Ile128
Ile128
Gly150
Gly149
Gly149
Asn161
2.0
5a
Phe126
Phe178
2.5
FAD
10a
2.7
Glu122
Phe126
Gly174
FAD
Phe106
Trp105
Phe106
Trp105
A
B
Thr71
Gly150
Gly150
Asn161
Asn161
Gly149
Ile128
Gly149
2.0
9c
Phe106
Gly174
Ile128
2.0
FAD
Phe178
Phe106
Gly174
Phe178
FAD
7c
Phe126
Phe126
2.4
Trp105
Thr71
C
D
Trp105
Figure 3. Compounds (A) 5a, (B) 10a (C) 7c, (D) 9c docked into the active site of NQO2. Polar contacts are shown as dashed lines and
p–
p
stacking interactions are shown as
0
solid arrows in A. Distances are shown in ÅA.
Figure 4. Survival of HCT116 cells when exposed to a variety of selected triazoloacridin-6-ones for 96 h. Data points are derived from at least three individual experiments.

K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
701
Table 2
Toxicity of the triazoloacridin-6-ones towards HCT116 colon cancer cell line following 24 and 96 h exposure
ID
IC50
(l
M) 24 h
IC50
(lM) 96 h
Tm DNA (high salt)
Tm DNA (low salt)
5a
5b
5c
5d
6a
6c
6d
7c
8c
8d
9a
9c
1
0.1
0.333 0.148
0.465 0.042
12.4 3.2
0.495 0.036
0.627 0.136
0.630 0.180
0.715 0.111
0.630 0.140
0.815 0.262
0.08 0.47
0.03 0.31
0.33 0.52
0.20 0.64
0.19 0.54
1.48 1.56
0.42 0.00
0.19 0.00
0.23 0.27
0.09 0.47
0.08 0.49
0.31 0.16
0.30 0.07
0.22 0.04
0.45 0.29
2.9 0.73
2.6 0.45
1.7 0.34
2.6 0.41
3.2 0.44
1.2 0.8
20.5 1.8
1.6 0.09
1.5 0.5
0.99 0.3
2.8 0.60
1.9 0.22
3
3
0.13
0.12
0.2 0.40
0.31
3.2
1
2
27 2.2
12.5 2.2
3.5 1.1
57 2.8
80 4.4
48 3.8
7
0.495
1.9 0.05
1.7 0.39
1.7 0.36
0.15 0.41
0.5 0.44
0.1 0.12
0.711 0.278
0.803 0.272
18.8 0.138
23.8 0.290
12 0.126
10a
10c
10d
Toxicity was measured using the MTT assay. Changes in DNA melting temperature,
D
Tm, in °C are given for salmon sperm DNA exposed to 10
lM of each compound in the
presence of physiological (high) or low salt concentrations.31,32 Each value SD is derived from three independent experiments.
and NQO2 and is not an outlier in the binding affinity structure–
activity relationships. It is not an N-oxide therefore this low toxic-
ity would not be predicted. In a previously reported small series of
triazoloacridin-6-ones31, 5c (designated C-1233) was shown to
have the lowest toxicity both in vitro and in vivo and further it
caused one of the smallest changes in Tm under low-salt condi-
tions, which is consistent with the present findings.
4. Conclusions
We have shown that these triazoloacridin-6-ones are novel
inhibitors of NQO1 and NQO2. In our previous study of inhibitors
of NQO120 we demonstrated that the widely used inhibitor of
NQO1, dicoumarol, gave an IC50 for enzyme inhibition of ꢁ3 nM.
Thus, although we have demonstrated a highly significant struc-
ture–activity relationship between binding affinity and inhibitory
potency for this enzyme, it is unlikely that the compounds re-
ported here will be useful as pharmacological inhibitors of NQO1.
In contrast, this is not the case for NQO2. Not only have we dem-
onstrated a clear structure–activity relationship, but we have also
shown that those compounds in the 10 series are highly potent
inhibitors (ꢁ100 nM). In the literature, resveratrol is used as stan-
dard inhibitor of NQO2. Using the enzyme assay reported here the
IC50 for resveratrol is 450 150 nM (unpublished results). Hence,
some of the triazoloacridin-6-ones represent the most potent
NQO2 inhibitors so far reported. Further, N-oxidation of the ter-
tiary amines confers reduced toxicity and increased water solubil-
ity while maintaining inhibitory potency. This provides direction
for future drug development as well as yielding compounds that
may have useful pharmacological activity.
Figure 5. DNA melting curves for salmon sperm DNA in the presence of 10
adriamycin (Doxorubicin), 5a or 10a.
lM
ity.31 Nevertheless, from these observations it can be inferred that
the triazoloacridin-6-ones have the potential for interacting with
DNA.
The most striking observation from the toxicity results is the
finding that the N-oxides (designated 10) are consistently less toxic
than the other triazoloacridin-6-ones. Each of the 10 compounds
show a ꢁ20-fold reduction in toxicity compared their parent, 5
compounds. This can be rationalized on the basis that the N-oxides
have no effect on Tm whereas the parent compounds do interact
with DNA. There are other examples in the literature which show
planar, aromatic, N-oxides to be less toxic than their unsubstituted
analogues. The best characterised example is the prodrug, banox-
antrone (AQ4 N), which is substantially less toxic than its product
of reductive activation, AQ4.33–35 Interestingly, this has also been
associated with decreased interaction of the N-oxide with DNA
compared with the reduced analogue.33,36 Inspection of the ability
of the N-oxides to act as inhibitors of NQO1 and NQO2 shows that
the N-oxides are consistently the most efficient inhibitors of NQO2,
which is not the same for NQO1. The fact that these compounds are
also the least toxic compounds in the series, suggests they might
be useful pharmacological tools to study the biological role of
NQO2 in cells.
5. Experimental
5.1. Chemistry
Chemicals were purchased from Sigma–Aldrich Co., and Alfa
Aesar. Melting points were measured using a Sanyo Gallenkamp
MPD350 heater. Elemental analyses was carried out by the mi-
cro-analytical Laboratory, at the University of Manchester and is
within 0.3% of predicted values. Infrared spectra were recorded
in the solid state using a Perkin–Elmer FT-IR instrument. Absorp-
tion maxima (m
max) are recorded in wavenumbers (cmꢀ1). 1H and
One other compound in the series of triazoloacridin-6-ones
shows a much reduced toxicity following both 24 and 96 h drug
exposure and this is 5c. It is a modest inhibitor of both NQO1
13C NMR spectra were aquired using Bruker Avance 500, Bruker
Avance 400, and Bruker Avance 300 spectrometers. 1H and 13C
assignments were supported by 1H COSY, DEPT and HMQC.

702
K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
Chemical shifts are quoted in parts per million (ppm) to the nearest
0.01 ppm and referenced to the solvent peak. Mass spectroscopy
was carried out by the Mass Spectrometry Laboratory, School of
Chemistry, at the University of Manchester. Molecular ions, frac-
tions from molecular ions and other major peaks are reported as
mass/charge (m/z) ratios and are within 5 ppm mass units for
electrospray and 10 ppm for HRMS. Solvents were evaporated
on a Buchi RE111 rotaryevaporator equipped with a Buchi 461
water bath. Column chromatography was performed using silica
gel (Sigma–Aldrich) 40–63 lm 60 Å. The eluents are specified in
individual procedures. Distilled water was obtained from a Milli-
pore Elix 5.
Synthesis of 2-(4-methoxyphenylamino)-6-chloro-3-nitrobenzoic
acid (1a).22 A solution of 4-methoxyaniline (0.700 g, 5.68 mmol)
in N,N-dimethylaniline (2 mL) was added to a flask containing
Synthesis of 2-(4-bromophenylamino)-6-chloro-3-nitrobenzoic
acid (1d): 4-bromoaniline (7.29 g, 42.4 mmol, 5 equiv) was heated
to 110 °C until completely melted. 2,6-Dichloro-3-nitrobenzoic
acid (2.00 g, 8.47 mmol, 1 equiv) was then added in small portions
over 20 min with stirring. Heating was continued and after 3 h a
solid mass had formed which was kept at 110ꢀC for a further 4 h.
Once cooled it was thoroughly crushed in 2 M NaOH (30 mL) and
left to stir for 30 min. The unreacted aniline was filtered off and
washed once with 2 M NaOH (5 mL) and twice with water
(2 ꢂ 5 mL). The filtrate was then acidified with dilute HCl to pH 3
which gave a yellow precipitate. This was collected and dried to
give 1d (1.89 g, 60%) as a yellow powder. mp 198.8–202.0 °C;
m
max/cmꢀ1 3305w (N–H), 3090w, 2818 m (C–H), 1703s (C@O),
1596s, 1574s, 1528 m; dH (500 MHz; DMSO-d6) 13.87 (1H, s,
C(7)O2H), 8.54 (1H, s, NH), 8.06 (1H, d, J 8.9, C(4)H), 7.51 (1H, d, J
8.9, C(5)H), 7.30 (2H, d, J 8.8, C(10)H, C(12)H), 6.71 (2H, d, J 8.8,
C(9)H, C(13)H); dC (125 MHz; DMSO-d6) 165.0 (q, C(7)), 143.4 (q),
143.0 (q), 135.2 (q), 134.5 (q), 133.5 (q), 131.5 (C(10)H, C(12)H),
127.3 (C(4)H), 125.2 (C(5)H), 118.7 (C(9)H, C(13)H), 112.0 (q,
C(11)); m/z (ꢀES) 373 (25, [MꢀH]ꢀ, 37Cl81Br), 371 (100, [MꢀH]ꢀ,
37Cl79Br, 35Cl81Br), 369 (80, [MꢀH]ꢀ, 35Cl79Br). 368.9282 ([MꢀH]ꢀ)
found by ꢀES, required 368.9283, error 0.3 ppm.
2,6-dichloro-3-nitrobenzoic
acid
(1.00 g,
4.24 mmol)
in
N,N-dimethylaniline (3 mL). The mixture was heated to 120 °C
and stirred for 18 h. Once cooled it was diluted with CHCl3
(20 mL), extracted with 1 M NaOH (20 mL), and washed twice
with 1 M NaOH (2 ꢂ 10 mL). The combined extracts were acidi-
fied with dilute HCl to pH 3 which gave a yellow precipitate. This
was collected and dried to give 1a (0.930 g, 68%): mp 211–213 °C
(Lit.,22 210–212 °C);
m
max/cmꢀ1 3297w (N–H), 2848w (C–H),
Synthesis of 1-chloro-7-methoxy-4-nitroacridin-9(10H)-one (2a).22
A mixture of 1a (1.00 g, 3.10 mmol), POCl3 (1.88 mL, 20.2 mmol)
and N,N-dimethylaniline (0.09 mL) in dichloroethane (10 mL) was
heated under reflux for 2 h. The resulting red precipitate was filtered,
washed with dichloroethane (2.0 mL), and dried to give 2a (0.891 g,
1698s (C@O), 1589s, 1574s, 1510s; dH (300 MHz; DMSO-d6) 8.69
(1H, s, NH), 8.04 (1H, d, J 8.9, C(4)H), 7.17 (1H, d, J 8.9, C(5)H),
6.91 (2H, d, J 8.9, C(10)H, C(12)H), 6.79 (2H, d, J 8.9, C(9)H,
C(13)H), 3.70 (3H, s, C(14)H3); dC (75 MHz; DMSO-d6) 165.0 (q,
C(7)), 155.6 (q), 153.3 (q), 138.4 (q), 136.5 (q), 134.7 (q), 134.0
(q), 127.6 (C(4)H), 122.6 (CH, Ph), 120.9 (C(5)H), 114.0 (CH, Ph),
55.2 (OCH3); m/z (ꢀES) 323 (10, [MꢀH]ꢀ, 37Cl), 321 (30, [MꢀH]ꢀ,
35Cl); 321.0290 ([MꢀH]ꢀ) found by ꢀES, required 321.0278, error
3.7 ppm.
87%) as a red needles. mp 263–264 °C (Lit.,22 262–263 °C);
mmax/cm
ꢀ1 3288 m (N–H), 3113w, 2839w (C–H), 1644s (C@O), 1606s, 1589s,
1565s; dH (300 MHz; DMSO-d6) 11.64 (1H, s, NH), 8.55 (1H, d, J 8.9,
C(3)H), 8.02 (1H, d, J 9.1, C(5)H), 7.59 (1H, d, J 2.9, C(8)H), 7.47 (1H,
dd, J 9.1, 2.9, C(6)H), 7.38 (1H, d, J 8.9, C(2)H), 3.16 (3H, s, C(14)H3);
m/z (ꢀES) 305 (35, [MꢀH]ꢀ, 37Cl), 303 (100, [MꢀH]ꢀ, 35Cl); (+ES)
329 (20, [M+Na]+, 37Cl), 327 (60, [M+Na]+, 35Cl); 303.0180 ([MꢀH]ꢀ)
found by ꢀES, required 303.0173, error 2.4 ppm.
Synthesis of 2-(4-fluorophenylamino)-6-chloro-3-nitrobenzoic acid
(1b):
A solution of 4-fluoroaniline (0.53 mL, 5.50 mmol) in
N,N-dimethylaniline (1.5 mL) was added to a flask containing 2,6-
dichloro-3-nitrobenzoic acid (1.18 g, 5.00 mmol) in N,N-dimethyl-
aniline (3.0 mL). The mixture was heated to 110 °C with stirring
under nitrogen for 18 h. Once cooled it was diluted with CHCl3
(20 mL), extracted with 1 M NaOH (20 mL), and washed twice with
1 M NaOH (2 ꢂ 10 mL). The combined extracts were acidified with
dilute HCl to pH 3 resulting in formation of a yellow precipitate.
This was collected and dried to give 1b (0.285 g, 18%): mp
Synthesis of 1-chloro-7-fluoro-4-nitroacridin-9(10H)-one (2b): A
mixture of 1b (248 mg, 0.80 mmol), POCl3 (0.50 mL, 5.25 mmol)
and N,N-dimethylaniline (0.23 lL) in dichloroethane (25 mL) was
heated under reflux for 2 h. The resulting red precipitate was fil-
tered, washed with dichloroethane (2.0 mL), and dried to give 2b
(135 mg, 58%) as orange needles. mp 269.5–271.0 °C; m
max/cmꢀ1
3305 m (N–H), 3078w (C–H), 1647s (C@O), 1624 m, 1589s,
1569s, 1525w, 1484s; dH (300 MHz; DMSO-d6) 11.68 (1H, NH),
8.55 (1H, d, J 8.8, C(3)H), 8.12 (1H, dd, J 8.9, 4.6, C(8)H), 7.82 (1H,
dd, J 8.9, 3.0, C(5)H), 7.73 (1H, app.dt, J 8.9, 3.0, C(6)H), 7.41 (1H,
d, J 8.8, C(2)H); m/z (ꢀES) 293 (35, [MꢀH]ꢀ, 37Cl), 291 (100,
[MꢀH]ꢀ, 35Cl); (+ES) 317 (35, [M+Na]+, 37Cl), 315 (100, [M+Na]+,
35Cl); 314.9960 ([M+Na]+) found by +ES, required 314.9943, error
5.3 ppm.
176.5–177.0 °C; m
max/cmꢀ1 3348w (N–H), 2980br (O–H), 2881w,
(C–H), 1704s (C@O), 1588s, 1569s, 1530 m, 1504s; dH (300 MHz;
DMSO-d6) 10.73 (1H, s, C(7)O2H), 8.57 (1H, s, NH), 8.05 (1H, d, J
8.9, C(4)H), 7.35 (1H, d, J 8.9, C(5)H), 7.02 (2H, app.t, J 8.8,
C(10)H, C(12)H), 6.88 (2H, dd, J 8.7, 4.8, C(9)H, C(13)H); dC
(75 MHz; DMSO-d6) 165.0 (q), 164.0 (q), 140.9 (q), 139.3 (q),
136.5 (q), 135.8 (q), 131.1 (q), 129.8 (q), 127.5 (C(4)H), 123.1
(C(5)H), 120.6 (d, J 8.2, C(9)H, C(13)H), 115.5 (d, J 7.1, C(10)H,
C(12)H); m/z (ꢀES) 311 (10, [MꢀH]ꢀ, 37Cl), 309 (30, [MꢀH]ꢀ,
35Cl). 309.0090 ([MꢀH]ꢀ) found by ꢀES, required 309.0084, error
2.0 ppm.
Synthesis of 1-chloro-4-nitroacridin-9(10H)-one (2c).22 A similar
procedure was adopted as above using 1c to give 2c (1.23 g, 82%)
as a bright red powder: mp 248–249 °C (Lit.,22 252–255 °C);
m
max/cmꢀ1 3307 m (N–H), 1644s (C@O), 1587, 1569 (N@O); dH
Synthesis of 6-chloro-3-nitro-2-(phenylamino)benzoic acid (1c).22
A similar procedure was adopted as above using aniline to give
1c (0.552 g, 46%) as a fine, yellow powder: mp 203–205 °C (Lit.,1
(500 MHz; DMSO-d6) 11.47 (1H, s, NH), 8.55 (1H, d, J 8.8, C(3)H),
8.18 (1H, d, J 8.0, C(5)H), 7.99 (1H , d, J 8.2, C(8)H), 7.89–7.82
(1H, m, C(6)H), 7.40 (1H, d, J 8.8, C(2)H), 7.38–7.41 (1H, m,
C(7)H); dC (100 MHz, DMSO-d6) 174.9 (q, C(9)), 141.8 (q), 138.4
(q), 137.0 (q), 133.9 (C(3)H), 130.1 (C(6)H), 125.6 (C(8)H), 123.2
(C(7)H), 122.5 (C(5)H), 121.8 (q), 118.6 (q), 118.3 (C(2)H), 99.2
(q); m/z (ꢀES) 275.1 (32, [MꢀH]ꢀ, 37Cl), 273.1 (100, [MꢀH]ꢀ,
35Cl). 297.0051 ([M+Na]+) found by +ES, required 297.00543, error
2.7 ppm.
199–204 °C);
m
max/cmꢀ1 3333w (N–H), 3040w (C–H), 1704s
(C@O), 1583s, 1568s; dH (500 MHz; DMSO-d6) 8.47 (1H, s, NH),
8.02 (1H, d, J 8.8, C(4)H), 7.41 (1H, d, J 8.8, C(5)H), 7.14 (2H, app.t,
J 7.9 C(10)H, C(12)H), 6.85 (1H, app.t, J 7.4, C(9)H, C(11)H), 6.77
(2H, d, J 7.6, C(9)H, C(13)H); dC (125 MHz; DMSO-d6) 165.1 (q,
C(7)), 143.4 (q), 142.1 (q), 135.3 (C(4)H), 132.5 (q), 129.0 (q),
128.6 (C(10)H, C(12)H), 127.7 (C(11)H), 124.1 (q), 121.0 (C(9),
C(13)H), 116.9 (C(5)H); m/z (ꢀES) 293 (8, [MꢀH]ꢀ, 37Cl), 291 (24,
[MꢀH]ꢀ, 35Cl); 291.0178 ([MꢀH]ꢀ) found by ꢀES, required
291.0178, error 0.0 ppm.
Synthesis of 7-bromo-1-chloro-4-nitroacridin-9(10H)-one (2d): A
similar procedure was adopted as above using 1d to give 2d
(1.61 g, 94%) as a bright red powder: mp 310.0–310.8 °C;
m
max/cm
ꢀ1
3291 m (N–H), 3074 m (C–H), 1658s (C@O), 1613, 1571s; dH

K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
703
(500 MHz; DMSO-d6) 8.50 (1H, d, J 8.8, C(3)H), 8.24 (1H, s, C(8)H),
7.98 (1H, d, J 8.6, C(5)H), 7.93 (1H, d, J 8.6, C(6)H), 7.39 (1H, d, J 8.8,
C(2)H); m/z (ꢀES) 355 (25, [MꢀH]ꢀ, 37Cl81Br), 353 (100, [MꢀH]ꢀ,
37Cl79Br, 35Cl81Br), 351 (80, [MꢀH]ꢀ, 35Cl79Br); (+ES) 378 (3,
[M+Na]+, 37Cl81Br), 376 (12, [M+Na]+, 37Cl79Br, 35Cl81Br), 374 (9,
[M+Na]+, 35Cl79Br); 350.9178 ([MꢀH]+) found by ꢀES, required
350.9178, error 0.1 ppm.
7.09 (1H, d, J 8.4, C(2)H), 7.03 (1H, d, J 8.4, C(3)H), 5.04 (2H, br s,
NH2); dC (75 MHz; DMSO-d6) 175.8 (q, C(9)), 139.5 (q), 136.7 (q),
136.5 (C(6)H), 132.5 (q), 128.7 (C(8)H), 125.1 (q), 123.3 (C(2)H),
120.8 (q), 120.1 (C(5)H), 117.7 (C(3)H), 117.0 (q), 114.3 (q, C(7));
m/z (ꢀES) 325.0 (25, [MꢀH]ꢀ, 37Cl81Br), 323.0 (100, [MꢀH]ꢀ,
37Cl79Br, 35Cl81Br), 321.0 (80, [MꢀH]ꢀ, 35Cl79Br); (+ES) 349.0 (12,
[M+Na]+, 37Cl81Br), 347.0 (60, [M+Na]+, 37Cl79Br, 35Cl81Br), 345.0
(45, [M+Na]+, 35Cl79Br), 327.0 (6, [M+H]+, 37Cl81Br), 325.0 (25,
[M+H]+, 37Cl79Br, 35Cl81Br), 323.0 (18, [M+Na]+, 35Cl79Br).
344.9397 ([M+Na]+) found by +ES, required 344.9301, error
ꢀ1.1 ppm.
Synthesis of 4-amino-1-chloro-7-methoxyacridin-9(10H)-one
(3a).23 A solution of SnCl2 (2.24 g, 11.8 mmol) in concd HCl
(2.5 mL) was added to a suspension of finely powdered 2a
(0.800 g, 2.63 mmol) in ethanol–concd HCl (2 mL: 2 mL). The mix-
ture was heated to reflux and stirred vigorously for 24 h and then
allowed to cool. The HCl salt of the product was filtered from the
solution as a peach coloured precipitate. This was added to water
(100 mL), made basic with NaOH, and stirred for 30 min. The sus-
pension was then filtered and washed with water to give 3a
(0.760 g, 86%) as a yellow powder: mp 215–218 °C (dec) (Lit.,23
Synthesis of 5-chloro-8-methoxy-6H-[1,2,3]triazolo[4,5,1-de]acri-
din-6-one (4a).23 A suspension of 3a (700 mg, 2.55 mmol) in concd
HCl (8.5 mL) was stirred at room temperature for 30 min. A solu-
tion of NaNO2 (235 mg, 3.40 mmol) in water (5.0 mL) was then
added gradually and the mixture was stirred for 4 h. The grey pre-
cipitate was collected and washed with water to give 4a (564 mg,
215–217 °C (dec));
m
max/cmꢀ1 3263 m (N–H), 3113w, 3004w,
78%) as a grey power: mp 261–262 °C (Lit.,23 263–266 °C);
mmax/cm
ꢀ1
2942w, 2840w (C–H), 1644s (C@O), 1605s, 1589s, 1566s; dH
(300 MHz; DMSO-d6) 10.58 (1H, s, NH), 7.67 (1H, d, J 9.0, C(5)H),
7.56 (1H, d, J 3.0, C(8)H), 7.37 (1H, dd, J 9.0, 3.0, C(6)H), 6.97 (1H,
d, J 8.1, C(2)H), 6.90 (1H, d, J 8.1, C(3)H), 3.84 (3H, s, OMe); dC
(75 MHz; DMSO-d6) 175.6 (q, C(9)), 154.2 (q), 136.1 (q), 134.6
(q), 131.6 (q), 123.8 (C(6)H), 123.5 (C(2)H), 122.0 (q), 119.2
(C(5)H), 119.0 (q), 116.1 (q), 114.9 (C(3)H), 105.1 (C(8)H), 55.3
(OCH3); m/z (ꢀES) 275 (30, [MꢀH]ꢀ, 37Cl), 273 (100, [MꢀH]ꢀ,
35Cl); (+ES) 299 (10, [M+Na]+, 37Cl), 297 (30, [M+Na]+, 35Cl), 277
(40, [M+H]+, 37Cl), 275 (100, [M+H]+, 35Cl); 275.0579 ([M+H]+)
found by +ES, required 275.0582, error 1.0 ppm.
3070w, 1672s (C@O), 1594s , 1506s; dH (300 MHz; CDCl3) 8.47
(1H, d, J 8.9, C(10)H), 8.37 (1H, d, J 8.5, C(3)H), 7.96 (1H, d, J 2.7,
C(7)H), 7.72 (1H, d, J 8.5, C(4)H), 7.50 (1H, dd, J 8.9, 2.7, C(9)H),
4.00 (3H, s, OMe); m/z (+ES) 310 (25, [M+Na]+, 37Cl), 208 (70,
[M+Na]+, 35Cl). 321.9960 ([M+Cl]ꢀ) found by ꢀES, required
321.9964, error 0.4 ppm.
Synthesis of 5-chloro-8-fluoro-6H-[1,2,3]triazolo[4,5,1-de]acridin-
6-one (4b): A suspension of 3b (75 mg, 0.286 mmol) in concd HCl
(1.0 mL) was stirred at room temperature for 30 min. A solution
of NaNO2 (25.6 mg, 0.371 mmol) in water (0.60 mL) was then
added gradually and the mixture was stirred for 4 h. The precipi-
tate was collected and washed with water to give 4b (65 mg,
Synthesis of 4-amino-1-chloro-7-fluoroacridin-9(10H)-one (3b): A
solution of SnCl2 (350 mg, 1.85 mmol) in concd HCl (1.0 mL) was
83%) as
a grey power: mp 291.0–302.0 °C (dec); m
max/cmꢀ1
added to
a
suspension of finely powdered 2b (120 mg,
3101 m (C–H), 1662s (C@O), 1638 m, 1595s, 1505s, 1478s; dH
(300 MHz; DMSO-d6) 8.68 (1H, d, J 8.5, C(3)H), 8.63–8.58 (1H, m,
C(10)), 8.10–8.06 (1H, m, C(7)H), 7.96 (1H, app.dt, J 8.6, 3.4,
C(9)H), 7.89 (1H, d, J 8.5, C(4)H); m/z (+ES) 298 (30, [M+Na]+,
37Cl), 296 (100, [M+Na]+, 35Cl).
0.410 mmol) in ethanol–concd HCl (1 mL: 1 mL). The mixture
was heated to reflux and stirred vigorously for 24 h and then al-
lowed to cool. The HCl salt of the product was filtered from the
solution as a peach coloured solid. This was added to water
(15 mL), basified with NaOH, and stirred for 30 mins. The suspen-
sion was then filtered and washed with water to give 3b (85 mg,
Synthesis of 5-chloro-6H-[1,2,3]triazolo[4,5,1-de]acridin-6-one
(4c).23 A similar procedure was adopted as above using 3c to give
4c (0.732 g, 70%) as a grey power: mp 221–222 °C (Lit.,23 222–
79%) as
a yellow powder: mp 270–280 °C (dec); m
max/cmꢀ1
3330 m, 3292 m, 3232w (N–H), 1570s (C@O), 1498s, 1461s; dH
(300 MHz; DMSO-d6) 7.74 (1H, dd, J 10.3, 3.2, C(8)H), 7.62–7.57
(1H, m, C(5)H), 7.37–7.33 (1H, m, C(6)H), 6.70 (1H, d, J 7.9,
C(2)H), 6.61 (1H, d, J 7.9, C(3)H), 5.55 (2H, s br, NH2); m/z (ꢀES)
263 (35, [MꢀH]ꢀ, 37Cl), 261 (100, [MꢀH]ꢀ, 35Cl); (+ES) 287 (30,
[M+Na]+, 37Cl), 285 (100, [M+Na]+, 35Cl). 285.0197 ([M+Na]+) found
by +ES, 285.0201, error 1.5 ppm.
223 °C); m
max/cmꢀ1 3076w (C–H), 1661s (C@O), 1634w, 1606s,
1594s, 1504s; dH (500 MHz; DMSO-d6) 8.63 (1H, d, J 9.0, C(3)H),
8.49 (1H, d, J 7.7, C(7)H), 8.34 (1H, d, J 7.7, C(10)H), 8.04 (1H, app.t,
J 7.7, C(9)H), 7.84 (1H, d, J 9.0, C(4)H), 7.71 (1H, app.t, J 7.7, C(8)H);
dC (75 MHz, DMSO-d6) 175.7 (q, C(6)), 143.5 (q), 135.4 (C(3)H),
134.7 (q), 129.5 (C(9)H), 128.4 (C(7)H), 127.7 (C(10)H), 127.0
(C(4)H), 124.5 (q), 115.3 (C(8)H); m/z (+ES) 280 (16, [M+Na]+,
37Cl), 278 (42, [M+Na]+, 35Cl), 258 (10, [M+H]+, 37Cl) 256 (30,
[M+H]+, 35Cl); 256.0274 ([M+H]+) found by +ES, required
256.0272, error 0.7 ppm.
Synthesis of 4-amino-1-chloroacridin-9(10H)-one (3c).23 A similar
procedure was adopted as above using 2c to give 3c (1.07 g, 92%) as
a yellow powder: mp 222–232 °C (dec) (Lit.,23 232–234 °C (dec));
m
max/cmꢀ1 3364w, 3264w (N–H), 1621s (C@O), 1566s, 1526s; dH
Synthesis of 8-bromo-5-chloro-6H-[1,2,3]triazolo[4,5,1-de]acridin-
6-one (4d): A similar procedure was adopted as above using 3d to
give 4d (0.725 g, 95%) as a grey power. This powder is insoluble in
(500 MHz; DMSO-d6) 10.49 (1H, s, NH), 8.15 (1H, dd, J 8.1, 2.2,
C(8)H), 7.70 (1H, m, C(6)H), 7.65 (1H, d, J 7.9, C(5)H), 7.23–7.26
(1H, m, C(7)H), 7.00 (1H, d, J 8.2, C(2)H), 6.94 (1H, d, J 8.2,
C(3)H), 5.65 (2H, s, NH2); dC (125 MHz; DMSO-d6) 176.4 (q, C(9)),
139.9 (q), 136.3 (q), 133.1 (q), 131.8 (C(6)H), 126.5 (C(8)H), 125.8
(q), 124.1 (q), 121.4 (C(3)H), 119.1 (q), 117.5 (C(2)H), 117.0
(C(5)H), 115.7 (C(7)H); m/z (ꢀES) 245 (32, [MꢀH]ꢀ, 37Cl), 243
(100, [MꢀH]ꢀ, 35Cl); (+ES) 269 (30, [M+Na]+, 37Cl), 267 (100,
[M+Na]+, 35Cl), 147 (20, [M+H]+, 37Cl), 145 (60, [M+H]+, 35Cl);
245.0476 ([M+H]+) found by +ES, required 245.0476, error 0.1 ppm.
Synthesis of 4-amino-7-bromo-1-chloroacridin-9(10H)-one (3d):
A similar procedure was adopted as above using 2d to give 3d
all typical NMR solvents: mp 291.0–291.6 °C; m
max/cmꢀ1 3087 m
(C–H), 1654s (C@O), 1633 m, 1603s, 1591s, 1558s; dH (500 MHz;
DMSO-d6) 8.69 (1H, s), 8.45 (2H, m), 8.25 (1H, s), 7.90 (1H, s).
Synthesis of 5-(2-(dimethylamino)ethylamino)-8-methoxy-6H-
[1,2,3]triazolo[4,5,1-de]acridin-6-one (5a).23 N,N-dimethylethylen-
diamine (185 mg, 2.10 mmol) was added to a suspension of 4a
(200 mg, 0.701 mmol) in anhydrous dimethylacetamide (DMA)
(3.0 mL) and the mixture was heated to 60 °C for 24 h. The reaction
mixture was allowed to cool, then EtOH (2.0 mL) was added and
the flask was left in a refrigerator for 2 days. The resulting yellow
crystals were collected and washed twice with water (2 ꢂ 10 mL)
and once with MeOH (10 mL) to give 5a (175 mg, 74%): mp
(0.825 g, 98%) as a yellow powder: mp 285–294 °C (dec); mmax
/
cmꢀ1 3335 m, 3285 m (N–H), 1617s (C@O), 1597s, 1577s, 1519s;
dH (500 MHz; DMSO-d6) 10.88 (1H, s, NH), 8.27 (1H, d, J 2.3,
C(8)H), 7.88 (1H, dd, J 8.9, 2.3, C(6)H), 7.74 (1H, d, J 8.9, C(5)H),
173–175 °C (Lit.,23 173–174 °C);
m
max/cmꢀ1 3294w (N–H), 2990w,
2951w, 2823w, 2772w (C–H), 1662s (C@O), 1597s, 1548s, 1573s;

704
K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
dH (300 MHz; DMSO-d6) 9.44 (1H, s br, NH), 8.40 (1H, d, J 9.1,
C(10)H), 8.30 (1H, d, J 9.3, C(3)H), 7.80 (1H, d, J 2.9, C(7)H), 7.57
(1H, dd, J 9.1, 2.9, C(9)H), 7.11 (1H, d, J 9.3, C(4)H), 3.94 (3H, s,
OMe), 3.59 (2H, app.q, J 5.5, C(12)H2), 2.61 (2H, t, J 5.9, C(13)H),
2.27 (6H, s, C(14)H3, C(15)H3); dC (75 MHz; DMSO-d6) 176.2 (q,
C(6)), 157.7 (q), 152.2 (q), 148.2 (q), 134.7 (q), 131.0 (q), 128.8
(C(3)H), 125.9 (q), 122.1 (C(9)H), 116.9 (C(10)H), 112.1 (C(4)H),
108.9 (C(7)H), 98.7 (q), 57.0 (C(13)H), 55.7 (OMe), 45.0 (C(14)H3,
C(15)H3), 40.2 (C(12)H); m/z (+ES) 360.0 (40, [M+Na]+), 338.1
(100, [M+H]+); 338.1621 ([M+H]+) found by +ES, required
338.1612, error 2.8 ppm.
(6H, s, NMe2); dC (75 MHz; DMSO-d6) 176.6 (q, C@O), 152.1 (q), 134.7
(q), 134.6 (q), 134.0 (C(9)H), 131.7 (q), 128.7 (C(3)H), 127.4
(C(8)H), 126.6 (C(7)H), 124.6 (q), 115.2 (C(10)H), 112.1 (C(4)H),
99.0 (q), 57.1 (C(13)H2), 45.1 (C(14)H3, C(15)H3), 40.3 (C(12)H2);
m/z (ꢀES) 306 (100, [MꢀH]ꢀ); (+ES) 330 (100, [M+Na]+), 308 (30,
[M+H]+); 330.1325 ([M+Na]+) found by +ES, required 330.1325, er-
ror 0.1 ppm.
Synthesis of 5-(3-dydroxypropylamino)-6H-[1,2,3]triazolo[4,5,1-
de]acridin-6-one (6c): A similar procedure was adopted as above
using 4c and 3-amino-1-propanol to give 6c (0.076 g, 65%) as green
crystals: mp 184–186 °C; m
max/cmꢀ1 3310br (O–H), 3278w (N–H),
Synthesis of 5-(3-hydroxypropylamino)-8-methoxy-6H-[1,2,3]triaz-
olo[4,5,1-de]acridin-6-one (6a): A similar procedure was adopted as
above using 3-amino-1-propanol to give 6a (32.2 mg, 56%): mp
2875w (C–H), 1662s (C@O), 1599s, 1567s, 1501s; dH (500 MHz;
DMSO-d6) 9.39 (1H, t, J 5.8, NH), 8.48 (1H, d, J 8.1, C(10)H), 8.39
(1H, dd, J 7.9, 1.5, C(7)H), 8.31 (1H, d, J 9.4, C(3)H), 8.00 (1H, app.dt,
J 7.8, 1.5, C(9)H), 7.70 (1H, app.t, J 7.6, C(8)H), 7.16 (1H, d, J 9.5,
C(4)H), 4.76 (1H, t, J 5.1, OH), 3.62 (2H, app.q, J 6.3, C(14)H2),
3.58 (2H, app.t, J 5.9, C(12)H2), 1.85 (2H, app.p, J 6.3, C(13)H2); dC
(75 MHz; DMSO-d6) 176.38 (q, C(6)), 152.09 (q), 134.51 (q),
134.29 (q), 133.79 (C(8)H), 131.33 (q), 128.40 (CH, C(9)), 127.02
(C(10)H), 126.38 (C(7)H), 124.26 (q), 115.03 (C(3)H), 111.40
(C(4)H), 98.77 (q), 58.14 (C(15)H2), 39.85 (C(13)H2), 31.76
(C(14)H2); m/z (ꢀES) 293 (100, [MꢀH]ꢀ); (+ES) 317 (100,
[M+Na]+); 293.1042 ([MꢀH]ꢀ) found by ꢀES, required 293.1044,
error ꢀ0.7 ppm.
173–174 °C; m
max/cmꢀ1 3339w (O–H), 3254 (N–H), 3046w, 2943w,
2866w (C–H), 1667s (C@O), 1601s, 1578s, 1556s, 1503s; dH
(300 MHz; DMSO-d6) 9.36 (1H, s br, NH), 8.38 (1H, d, J 8.9, C(10)H),
8.29 (1H, d, J 9.2, C(3)H), 7.75 (1H, s, C(7)H), 7.57 (1H, d, J 8.9,
C(9)H), 7.12 (1H, d, J 9.2, C(4)H), 4.75 (1H, t, J 5.0, OH), 3.93 (3H, s,
OMe), 3.56–3.64 (4H, m, C(12)H2, C(14)H2), 1.86 (2H, app.quin, J 6.3,
C(13)H2); dC (75 MHz; DMSO-d6) 176.3 (q, C(6)), 157.7 (q), 152.5 (q),
139.6 (q), 134.6 (q), 128.9 (C(3)H), 125.9 (q), 121.9 (C(9)H), 116.9
(C(10)H), 111.7 (C(4)H), 108.97 (C(7)H), 106.5 (q), 98.7 (q), 58.1
(C(14)H2), 55.7 (C(15)H3), 40.0 (C(12)H2), 31.7 (C(13)H2); m/z (ꢀES)
323 (100, [MꢀH]ꢀ); (+ES) 347 (100, [M+Na]+), 325 (40, [M+H]+);
325.1287 ([M+H]+) found by +ES, required 325.1295, error 2.5 ppm.
Synthesis of 8-methoxy-5-(phenethylamino)-6H-[1,2,3]triazol-
o[4,5,1-de]acridin-6-one (9a): A similar procedure was adopted as
above using 2-phenylethylamine to give 9a (47.5 mg, 73%): mp
Synthesis of 5-(4-methoxyphenylamino)-6H-[1,2,3]triazolo[4,5,1-
de]acridin-6-one (7c): A similar procedure was adopted as above
using 4c and p-anisidine to give 7c (0.116 g, 87%) as green crystals:
mp 234–235 °C;
m
max/cmꢀ1 3242w (N–H), 2836w (C–H), 1664s
(C@O), 1627 m, 1603s, 1584s, 1543s; dH (500 MHz; DMSO-d6)
10.72 (1H, s, NH), 8.55 (1H, d, J 8.2, C(10)H), 8.46 (1H, dd, J 7.8,
1.5, C(7)H), 8.31 (1H, d, J 9.3, C(3)H), 8.06 (1H, app.dt, J 7.9, 1.5,
C(9)H), 7.75 (1H, app.t, J 7.8 C(8)H), 7.44 (2H, d, J 8.9, C(14)H,
C(16)H), 7.14 (1H, d, J 9.3, C(4)H), 7.09 (2H, d, J 8.9, C(13)H,
C(17)H), 3.82 (3H, s, CH3); dC (100 MHz; DMSO-d6) 178.1 (q,
C(6)), 158.5 (q), 151.2 (q), 136.5 (q), 135.5 (q), 134.9 (q), 132.3
(C(9)H), 131.1 (q), 129.2 (q), 128.0 (C(3)H), 127.31 (C(10)H),
126.5 (C(8)H), 125.0 (C(14)H, C(16)H), 115.9 (C(7)H), 115.7
(C(13)H, C(17)H), 113.0 (C(4)H), 100.8 (q), 56.0 (C(18)H3); m/z
(ꢀES) 341 (100, [MꢀH]ꢀ); (+ES) 365 (100, [M+Na]+), 343 (20,
[M+H]+); 343.1194 ([M+Na]+) found by +ES, required 343.1190, er-
ror 1.3 ppm.
195–197 °C;
m
max/cmꢀ1 3280w (N–H), 2988w (C–H), 1657s
(C@O), 1601s, 1579s, 1556s, 1503s; dH (300 MHz; CDCl3) 9.42
(1H, s br, NH), 8.47 (1H, d, J 9.1, C(10)H), 8.14 (1H, d, J 9.2,
C(3)H), 7.93 (1H, d J 2.9, C(7)H), 7.46 (1 h, dd, J 9.0, 2.9, C(9)H),
7.30–7.39 (5H, m, Ph), 6.88 (1H, d, J 9.2, C(4)H), 4.00 (3H, s,
OMe), 3.77 (2H, app.q, J 6.9, C(12)H2), 3.11 (2H, t, J 7.3, C(13)H2)
dC (75 MHz; CDCl3) 178.0 (q, C(6)), 158.2 (q), 152.8 (q), 138.0 (q),
135.6 (q), 131.5 (q), 129.8 (q), 128.9 (C(3)H), 128.8 (2CH, Ph),
128.8 (2CH, Ph), 126.9 (CH, Ph), 126.4 (q), 122.7 (C(9)H), 117.2
(C(10)H), 110.6 (C(4)H), 108.8 (C(7)H), 99.9 (q), 55.9 (OMe), 44.9
(C(12)H2), 35.8 (C(13)H2); m/z (ꢀES) 369 (50, [MꢀH]ꢀ); (+ES) 393
(100, [M+Na]+), 371 (60, [M+H]+); 371.1515 ([M+H]+) found by
+ES, required 371.1503, error 3.4 ppm.
Synthesis of 5-(isopentylamino)-6H-[1,2,3]triazolo[4,5,1-de]acri-
din-6-one (8c): A similar procedure was adopted as above using
4c and 3-methylbutylamine to give 8c (0.09 g, 76%) as green crys-
Synthesis of 5-(2-(dimethylamino)ethylamino)-8-fluoro-6H-[1,2,3]
triazolo[4,5,1-de]acridin-6-one (5b): A similar procedure was adopted
as above using 4b and N,N-dimethylethylendiamine to give 5b
tals: mp 144–145 °C;
m
max/cmꢀ1 3273w (N–H), 3066w, 2950w,
(55 mg, 92%) as yellow crystals: mp 187.7–188.0 °C;
m
max/cmꢀ1
2864w (C–H), 1657s (C@O), 1623 m, 1599s, 1582s, 1502s; dH
(500 MHz; DMSO-d6) 9.30 (1H, t, J 5.7, NH), 8.50 (1H, d, J 7.6,
C(10)H), 8.40 (1H, dd, J 8.2, 1.2, C(7)H), 8.33 (1H, d, J 9.2, C(3)H),
8.01 (1H, app.dt, J 7.8, 1.2, C(9)H), 7.71 (1H, app.t, J 7.8, C(8)H),
7.16 (1H, d, J 9.2, C(4)H), 3.57 (2H, app.q, J 6.8, C(12)H2), 1.76
(1H, app.quin, J 6.7C(14)H), 1.61 (2H, app.t, J 7.2, C(13)H2), 0.98
(6H, d, J 6.6, C(15)H3, C(16)H3); dC (100; MHz, CDCl3) 178.2 (q,
C(6)), 152.9 (q), 135.4 (q), 135.3 (q), 133.7 (C(9)H), 132.3 (q),
128.9 (C(3)H), 127.7 (C(7)H), 126.5 (C(8)H), 125.1 (q), 115.7
(C(10)H), 110.9 (C(4)H), 100.0 (q), 41.5 (C(12)H2), 37.9 (C(13)H2),
25.8 (C(14)H), 22.5 (C(15)H3, C(16)H3); m/z (ꢀES) 305 (100,
[MꢀH]ꢀ); (+ES) 329 (100, [M+Na]+), 307 (10, [M+H]+); 329.1380
([M+Na]+) found by +ES, required 329.1373, error 2.2 ppm.
3283w (N–H), 2947w, 2774w (C–H), 1660s (C@O), 1629 m, 1609s,
1579s, 1548s, 1501s; dH (300 MHz; CDCl3) 9.50 (1H, s br, NH), 8.55
(1H, dd, J 9.0, 4.5, C(10)H), 8.21 (1H, dd, J 9.0, 2.8, C(7)H), 8.19 (1H,
d, J 9.2, C(3)H), 7.63 – 7.57 (1H, m, C(9)H), 6.93 (1H, d, J 9.2, C(4)H),
3.58 (2H, app.q, J 8.9, C(12)H2), 2.73 (2H, t, J 6.3, C(13)H2), 2.39 (6H,
s, C(14)H3, C(15)H3); dC (75 MHz; CDCl3) 177.1, 162.5, 159.3, 152.8,
135.5, 132.1, 131.8, 129.2, 127.2, 127.1, 121.7, 121.3, 117.8, 117.7,
113.9, 113.5, 111.1, 99.9, 57.6, 45.4, 41.1; m/z (+ES) 326 (100,
[M+H]+); 348.1241 ([M+Na]+) found by +ES, required 348.1231 error
2.8 ppm.
Synthesis of 5-(2-(dimethylamino)ethylamino)-6H-[1,2,3]triazol-
o[4,5,1-de]acridin-6-one (5c).23 A similar procedure was adopted
as above using 4c and N,N-dimethylethylendiamine to give 5c
(0.0473 g, 39.3%) as yellow crystals: mp 129–130 °C (Lit.,23 128–
Synthesis of 5-(phenethylamino)-6H-[1,2,3]triazolo[4,5,1-de]acri-
din-6-one (9c): A similar procedure was adopted as above using
4c and 2-phenylethylamine to give 9c (0.102 g, 77%) as dark, green
crystals: mp 153–155 °C. C21H16N4O requires C, 74.10; H, 4.74; N,
129 °C); m
max/cmꢀ1 3293w (N–H), 2943w, 2767w, 1662s (C@O),
1600s, 1567s, 1542s, 1502s; dH (500 MHz; DMSO-d6) 9.45 (1H, t,
J 4.8, NH), 8.50 (1H, d, J 7.9, C(10)H), 8.42 (1H, dd, J 7.9, 1.3,
C(7)H), 8.33 (1H, d, J 9.2, C(3)H), 8.01 (1H, app.dt, J 7.7, 1.3,
C(9)H), 7.71 (1H, app.t, J 7.6, C(8)H), 7.15 (1H, d, J 9.2, C(4)H),
3.61 (2H, app.q, J 5.7, C(12)H2), 2.61 (2H, t, J 6.0, C(13)H2), 2.27
16.46. Found: C, 73.85; H, 4.77; N, 16.36. m
max/cmꢀ1 3280w (N–
H), 2926w, 2863w (C–H), 1658s (C@O), 1623 m, 1599s, 1567s; dH
(500 MHz; DMSO-d6) 9.27 (1H, t, J 5.8, NH), 8.44 (1H, d, J 7.9,
C(10)H), 8.34 (1H, d, J 8.0, C(7)H), 8.26 (1H, d, J 9.2, C(3)H), 7.98

K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
705
(1H, app.t, J 7.7, C(9)H), 7.67 (1H, app.t, J 7.7, C(8)H), 7.37 (2H, d, J
7.3, C(15)H, C(19)H), 7.33 (2H, app.t, J 7.3, C(16)H, C(18)H), 7.23
(1H, app.t, J 7.3, C(17)H), 7.15 (1H, d, J 9.2, C(4)H), 3.79 (2H, app.q,
J 6.7, C(12)H2), 3.01 (2H, t, J 7.3, C(13)H2); dC (75 MHz; DMSO-d6)
176.8 (q, C(6)), 152.2 (q), 138.5 (q), 134.8 (q), 134.6 (q), 134.1
(C(8)H), 131.6 (q), 128.9 (C(17)H, C(19)H), 128.8 (C(9)H), 128.5
(C(16)H, C(20)H), 127.3 (C(10)H), 126.7 (C(7)H), 126.5 (C(18)H),
124.5 (q), 115.3 (C(3)H), 111.9 (C(4)H), 99.1 (q), 44.0 (C(13)H2),
34.8 (C(14)H2); m/z (ꢀES) 339 (80, [MꢀH]ꢀ); (+ES) 363 (100,
[M+Na]+); 363.1214 ([M+Na]+) found by +ES, required 363.1216,
error 0.6 ppm.
crude 10a (182 mg) which was then dissolved in CHCl3 (5 mL)
and filtered to remove alumina residues and reduced under vac-
uum to give pure 10a (65 mg, 62%) as a yellow powder: mp 147–
148 °C;
m
max/cmꢀ1 3320 m (N–H), 1663 (C@O), 1602s, 1580s,
1565s; dH (500 MHz; D2O) 7.17 (1H, s), 6.66 (1H, d, J 6.6), 6.42
(1H, d, J 7.2), 6.23 (1H, d, J 8.8), 5.78 (1H, s), 3.52 (2H, s,
C(13)H2), 3.42 (2H, s, C(14)H2), 3.23 (3H, s, OMe), 3.19 (6H, s,
Me2); m/z (ꢀES) 352 (60, [MꢀH]ꢀ); (+ES) 376 (100, [M+Na]+), 354
(15, [M+H]+); 354.1563 ([M+H]+) found by +ES, required
354.1561 error 0.7 ppm.
Synthesis
of
5-(2-(dimethylamino)ethylamino)-8-fluoro-6H-
Synthesis
of
5-(2-(dimethylamino)ethylamino)-8-bromo-6H-
[1,2,3]triazolo[4,5,1-de]acridin-6-one N-oxide (10b): A similar proce-
dure was adopted as above using 5b to give 10b (27 mg, 52%). as a
[1,2,3]triazolo[4,5,1-de]acridin-6-one (5d): A similar procedure was
adopted as above using 4d and N,N-dimethylethylendiamine to
give 5d (94.9 mg, 82%) as yellow crystals: mp 200.8–201.2 °C;
yellow powder: mp 130–140 °C (dec); m
max/cmꢀ1 3248br (N–H),
1663s (C@O), 1630s, 1578s, 1500s; dH (300 MHz; D2O) 7.34 (1H,
d, J 9.5, C(3)H), 7.20–7.08 (2H, m, C(9)H, C(10)H), 6.75 (1H, dd, J
9.4, 2.2, C(7)H), 6.38 (1H, d, J 9.5, C(4)H), 3.63 (2H, t, J 6.1,
C(12)H2), 3.49 (2H, t, J 6.1, C(13)H2), 3.23 (6H, s, Me2); m/z (ꢀES)
340 (100, [MꢀH]ꢀ); (+ES) 364 (100, [M+Na]+); 340.1216 ([MꢀH]ꢀ)
found by ꢀES, required 340.1215 error 0.2 ppm.
m
max/cmꢀ1 3251w (N–H), 2942w, 2793w (C–H), 1661s (C@O),
1625s, 1599s, 1574s and 1562s; dH (500 MHz; DMSO-d6) 9.39
(1H, s, NH), 8.59 (1H, d, J 2.0, C(7)H), 8.35 (1H, d, J 8.6, C(10)H),
8.12 (1H, d, J 9.0, C(3)H), 7.90 (1H, dd, J 8.6, 2.0, C(9)H), 6.92 (1H,
d, J 9.0, C(4)H), 3.63 (2H, br s, C(12)H2), 2.77 (2H, br s, C(13)H2),
2.41 (6H, s, Me2); dC (75 MHz; CDCl3) 176.6 (q, C(6)), 152.7 (q),
136.5 (C(9)H), 135.5 (q), 134.1 (q), 132.3 (q), 130.8 (C(7)H), 129.2
(C(3)H), 126.6 (q), 120.3 (q), 117.4 (C(10)H), 111.2 (C(4)H), 100.2
(q), 57.5 (C(13)H2), 45.4 (C(14)H3, C(15)H3), 41.1 (C(12)H2); m/z
(+ES) 388 (100, [M+H]+, 81Br), 386 (100, [M+H]+, 79Br); 386.0611
([M+H]+) found by +ES, required 386.0611 error 0.0 ppm.
Synthesis of 5-(2-(dimethylamino)ethylamino)-6H-[1,2,3]triazol-
o[4,5,1-de]acridin-6-one N-oxide (10c): A similar procedure was
adopted as above using 5c to give 10c (28 mg, 46%) as a yellow
powder: mp 145–146 °C;
m
max/cmꢀ1 3271w (N–H), 2925w,
2863w (C–H), 1661s (C@O), 1624s, 1599s, 1601s, 1570s; dH
(300 MHz; CDCl3) 9.90 (1H, br s, NH), 8.51–8.57 (2H, m, C(7)H,
C(10)H), 8.25 (1H, d, J 9.1, C(3)H), 7.90 (1H, app.t, J 7.7, C(9)H),
7.62 (1H, app.t, J 7.7, C(8)H), 7.15 (1H, d, J 9.1, C(4)H), 4.24 (2H,
app.q, J 5.9, C(12)H2), 3.70 (2H, t, J 5.9, C(13)H2), 3.39 (6H, s,
NMe2); m/z (ꢀES) 322 (30, [MꢀH]ꢀ); (+ES) 346 (100, [M+Na]+);
324.1452 ([M+H]+) found by +ES, required 324.1455 error
ꢀ0.9 ppm.
Synthesis of 5-(3-hydroxypropylamino)-8-bromo-6H-[1,2,3]triaz-
olo[4,5,1-de]acridin-6-one (6d): A similar procedure was adopted
as above using 4d and 3-amino-1-propanol to give 6d (92.6 mg,
83%) as yellow crystals: mp 210.8–211.0 °C;
m
max/cmꢀ1 3438 m
(O–H), 3254w (N–H), 3076w, 2929w, 2873w (C–H), 1665s (C@O),
1622s, 1598s, 1566s; dH (300 MHz; CDCl3) 9.36 (1H, s, NH), 8.58
(1H, d, J 2.2, C(7)H), 8.37 (1H, d, J 8.7, C(10)H), 8.12 (1H, d, J 9.3,
C(3)H), 7.91 (1H, dd, J 8.7, 2.2, C(9)H), 6.93 (1H, d, J 9.3, C(4)H),
3.84 (2H, t, J 5.8, C(14)H2), 3.63 (2H, app.q, J 6.4, C(12)H2), 2.00
(2H, app.p, J 6.3, C(13)H2); dC (75 MHz; DMSO-d6) 175.2 (q, C(6)),
152.5 (q), 136.5 (C(9)H), 134.6 (q), 133.5 (C(7)H), 131.7 (q), 129.4
(C(3)H), 129.1 (q), 126.1 (q), 119.3 (q), 117.7 (C(10)H), 111.8
(C(4)H), 99.1 (q), 58.1 (C(12)H), 40.1 (C14)H), 31.7 (C(13)H); m/z
(ꢀES) 373 (100, [MꢀH]ꢀ, 81Br), 371 (100, [MꢀH]ꢀ, 79Br); (+ES)
397 (50, [M+Na]+, 81Br), 395 (50, [M+Na]+, 79Br); 371.0150
([MꢀH]ꢀ) found by ꢀES, required 371.0150 error 0.2 ppm.
Synthesis
of
5-(2-(dimethylamino)ethylamino)-8-bromo-6H-
[1,2,3]triazolo[4,5,1-de]acridin-6-one N-oxide (10d): A similar proce-
dure was adopted as above using 5d to give 10d (50 mg, 50%). as a
yellow powder: mp 143–144 °C; m
max/cmꢀ1 3253br (N–H), 2956w
(C–H), 1660s (C@O), 1622s, 1599s, 1565s; dH (300 MHz; CDCl3)
10.06 (1H, s, NH), 8.64 (1H, d, J 2.2, C(7)H), 8.43 (1H, d, J 8.7,
C(10)H), 8.24 (1H, d, J 9.2, C(3)H), 7.98 (1H, dd, J 8.7, 2.2, C(9)H),
7.15 (1H, d, J 9.2, C(4)H), 4.32 (2H, app.q, J 6.0, C(12)H2), 3.68
(2H, t, J 6.1, C(13)H2), 3.36 (6H, s, Me2); m/z (ꢀES) 402 (100,
[MꢀH]ꢀ, 81Br), 400 (100, [MꢀH]ꢀ, 79Br); (+ES) 426 (100, [M+Na]+,
81Br), 424 (100, [M+Na]+, 79Br); 402.0563 ([M+H]+) found by +ES,
required 402.0560 error 0.8 ppm.
Synthesis of 5-(4-methoxyphenylamino)-8-bromo-6H-[1,2,3]triaz-
olo[4,5,1-de]acridin-6-one (7d): A similar procedure was adopted
as above using 4d and p-anisidine to give 7d (104 mg, 83%) as an
orange powder: mp 240–245 °C (dec);
m
max/cmꢀ1 3236w (N–H),
5.2. Molecular modeling
1658s (C@O), 1621s, 1599s, 1574s, 1559s, 1508s; dH (300 MHz;
CDCl3) 10.71 (1H, s br, NH), 8.68 (1H, d, J 2.2, C(7)H), 8.46 (1H, d,
J 8.7, C(10)H), 8.10 (1H, d, J 9.2, C(3)H), 8.01 (1H, dd, J 8.7, 2.2,
C(9)H), 7.34 (2H, d, J 8.8, C(14)H, C(16)H), 7.14 (1H, d, J 9.2,
C(4)H), 7.03 (2H, d, J 8.8, C(13)H, C(17)H), 3.88 (3H, s, OMe,
C(18)H3), 1.58 (3H, s, OMe, C(19)H3); dC (75 MHz; CDCl3) 177.1
(q, C(6)), 158.5 (q), 151.7 (q), 136.9 (C(9)H), 136.4 (q), 134.3 (q),
132.2 (q), 130.8 (C(7)H), 130.4 (q), 129.0 (C(3)H), 126.5 (C(14)H,
C(16)H), 126.4 (q), 120.5 (q), 117.5 (C(10)H), 115.1 (C(13)H,
C(17)H), 112.5 (C(4)H), 100.7 (q), 55.6 (C(18)H3); m/z (ꢀES) 421
(50, [MꢀH]ꢀ, 81Br), 419 (50, [MꢀH]ꢀ, 79Br); 419.0139 ([MꢀH]ꢀ)
found by ꢀES, required 419.0138 error 0.2 ppm.
For docking purposes, the crystallographic coordinates of hu-
man NQO1 (PDB code 1D4A; resolution 1.7 Å) and NQO2 (PDB code
1QR2; resolution 2 Å) were obtained from the Brookhaven Data-
base. Hydrogen atoms were added to these structures which in-
cluded the protein and FAD allowing for appropriate ionization at
physiological pH. The protonated complexes were then minimized
within SYBYL 7.3 whilst holding all heavy atoms stationary. The li-
gand was then removed leaving the enzyme active site available
for subsequent docking studies. Inspection of the active sites of
both enzymes showed a reasonable internal hydrogen bond net-
work within the protein.
For preparation of ligand structures, fragments from SYBYL 7.3
were used to construct the compounds. Ligands were subject to
1000 iterations of energy minimization using the steepest descent
algorithm using the Tripos force field. Flexible ligand docking was
performed using the GOLD 4.1 software in combination with the
ChemScore27 scoring function. The FAD fragment was re-atom-
typed to avoid underestimation by ChemScore of lipophilic/aro-
matic interactions. The active site was defined as being any volume
Synthesis of 5-(2-(dimethylamino)ethylamino)-8-methoxy-6H-
[1,2,3]triazolo[4,5,1-de]acridin-6-one N-oxide (10a): Compound 5a
(100 mg, 0.296 mmol) was stirred in CHCl3 (5 mL) until dissolved.
The mixture was cooled to below 5 °C before 50% meta-chloroper-
benzoic acid (mCPBA) (113 mg, 0.326 mmol) was added. The solu-
tion was stirred for 24 h at room temperature. The reaction
mixture was added straight to a column of alumina for purification
by flash chromatography using MeOH/CHCl3 (1:10). This gave

706
K. A. Nolan et al. / Bioorg. Med. Chem. 18 (2010) 696–706
within 15 Å of N5 of the FAD cofactor. The top ten solutions for
each ligand were retained and analyzed for favourable interactions
within the active site of NQO1 and NQO2, including low protein–li-
gand clash, ligand distortion energies, lipophilic contacts and
hydrogen bonding interactions.
8.0 software. Curve fitting was done using the sigmoidal function
which gave a Boltzmann distribution and regression analysis of
the data. Values of Tm were derived from at least three indepen-
dent determinations.
Acknowledgments
5.3. Inhibition of NQO1 and NQO2
This work was funded by an MRC program grant to I.J.S.
(G0500366) and a project grant from the Association for Interna-
tional Cancer Research to K.A.N., R.A.B., R.C.W. and I.J.S.
Recombinant human NQO1 and NQO2 were obtained from Sig-
ma and diluted in 50 mM phosphate buffer to give an enzyme
activity that would result in a change in optical absorbance of sub-
strate of approximately 0.1 per minute. The enzyme reaction was
Supplementary data
started by adding 5
phate buffer at pH 7.4 containing either 200
for NQO1 or NQO2, respectively, together with 40
l
L of this solution to 495
l
l
L of 50 mM phos-
M NADH or NRH
M DCPIP
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2009.11.059.
l
(dichlorophenolindophenol) and various concentrations of the po-
tential inhibitor dissolved in DMSO (final concentration 1.0% v/v).
Reactions were carried out at 37 °C and reduction of DCPIP was
monitored at 600 nm in a Beckman DU 650 spectrophotometer.
IC50 values were determined using nonlinear curve fitting as imple-
mented in the program Excel for which a 50% reduction of the ini-
tial rate was attained. Each measurement was made in triplicate
and the experiments carried out three times. Values of IC50 were
converted to experimental binding affinities using the Cheng–Prus-
References and notes
1. Ernster, L.; Navazio, F. Biochim. Biophys. Acta 1957, 26, 408.
2. Liao, S.; Williams-Ashman, H. G. Biochem. Biophys. Res. Commun. 1961, 4, 208.
3. Bianchet, M. A.; Faig, M.; Amzel, L. M. Methods Enzymol. 2004, 382, 144.
4. Jaiswal, A. K.; McBride, O. W.; Adesnik, M.; Nebert, D. W. J. Biol. Chem. 1988,
263, 13572.
5. Schlager, J. J.; Powis, G. Int. J. Cancer 1990, 45, 403.
6. Long, D. J., 2nd; Jaiswal, A. K. Chem. Biol. Interact. 2000, 129, 99.
7. Long, D. J., 2nd; Iskander, K.; Gaikwad, A.; Arin, M.; Roop, D. R.; Knox, R.;
Barrios, R.; Jaiswal, A. K. J. Biol. Chem. 2002, 277, 46131.
8. Iskander, K.; Barrios, R. J.; Jaiswal, A. K. Clin. Cancer Res. 2009, 15, 1534.
9. Phillips, R. M.; Jaffar, M.; Maitland, D. J.; Loadman, P. M.; Shnyder, S. D.; Steans,
G.; Cooper, P. A.; Race, A.; Patterson, A. V.; Stratford, I. J. Biochem. Pharmacol.
2004, 68, 2107.
off equation28, where
D
Gcalc = ꢀRT ln(IC50/{1 + ([S]/Km)} where [S]
is the substrate concentration and Km is the Michaelis constant.
Regression analyses were derived using Excel.
10. Gaikwad, N. W.; Yang, L.; Rogan, E. G.; Cavalieri, E. L. Free Radical Biol. Med.
2009, 46, 253.
11. Knox, R. J.; Jenkins, T. C.; Hobbs, S. M.; Chen, S.; Melton, R. G.; Burke, P. J. Cancer
Res. 2000, 60, 4179.
5.4. Toxicity assays
HCT116 colon cancer carcinoma cells were routinely main-
tained in monolayer culture in RPMI 1640, supplemented with
10% FCS (foetal calf serum) and 2 mM L-glutamine. Chemosensitiv-
ity was determined by MTT assay37. 2 ꢂ 103 cells were plated per
well in a 96-well culture plate and incubated at 37° in an atmo-
sphere of air plus 5% CO2. Cells were than exposed continuously
to drug for 4 days at 37 °C. Alternatively, 24 h drug exposure was
used after which drug was removed and 0.2 ml of fresh medium
was added and cells allowed to grow for a further 3 days. Subse-
quently, MTT was added and cells incubated for a further 4 h. After
this time, culture medium and MTT solution were removed and
formazan crystals were dissolved in DMSO. Optical density was
12. Winger, J. A.; Hantschel, O.; Superti-Furga, G.; Kuriyan, J. B. M. C. Struct. Biol.
2009, 9, 7.
13. Stratford, I. J.; Workman, P. Anticancer Drug Des. 1998, 13, 519.
14. Knox, R. J.; Burke, P. J.; Chen, S.; Kerr, D. J. Curr. Pharm. Des. 2003, 9, 2091.
15. Gong, X.; Kole, L.; Iskander, K.; Jaiswal, A. K. Cancer Res. 2007, 67, 5380.
16. Asher, G.; Bercovich, Z.; Tsvetkov, P.; Shaul, Y.; Kahana, C. Mol. Cell. 2005, 17,
645.
17. Asher, G.; Tsvetkov, P.; Kahana, C.; Shaul, Y. Gene Dev. 2005, 19, 316.
18. Ahn, K. S.; Gong, X.; Sethi, G.; Chaturvedi, M. M.; Jaiswal, A. K.; Aggarwal, B. B.
Cancer Res. 2007, 67, 10004.
19. Nolan, K. A.; Timson, D. J.; Stratford, I. J.; Bryce, R. A. Bioorg. Med. Chem. Lett.
2006, 16, 6246.
20. Nolan, K. A.; Zhao, H.; Faulder, P. F.; Frenkel, A. D.; Timson, D. J.; Siegel, D.; Ross,
D.; Burke, T. R.; Stratford, I. J., Jr.; Bryce, R. A. J. Med. Chem. 2007, 50, 6316.
21. Cuyegkeng, M. A.; Mannschreck, A. Chem. Ber. 1987, 120, 803.
22. Capps, D. B.; Dunbar, J.; Kesten, S. R.; Shillis, J.; Werbel, L. M. J. Med. Chem. 1992,
35, 4770.
23. Cholody, W. M.; Martellli, S.; Konopa, J. J. Med. Chem. 1990, 33, 2852.
24. Cymerman Craig, J.; Purushothaman, K. K. J. Org. Chem. 1969, 35, 1721.
25. Faig, M.; Bianchet, M. A.; Talalay, P.; Chen, S.; Winski, S.; Ross, D.; Amzel, L. M.
Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 3177.
measured at 550 nM on a BioTek—lQuant multiwell spectropho-
tometer using Gen5 software. Values of IC50 are the drug concen-
trations required to reduce optical density by 50% relative to
untreated and/or vehicle only treated cells. All experiments were
repeated at least three times.
26. Foster, C. E.; Bianchet, M. A.; Talalay, P.; Zhao, Q.; Amzel, L. M. Biochemistry
1999, 38, 9881.
27. Verdonk, M. L.; Cole, J. C.; Hartshorn, M. J.; Murray, C. W.; Taylor, R. D. Proteins
2003, 52, 609.
5.5. Measurement of melting temperature
28. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
29. Jamieson, D.; Tung, A. T.; Knox, R. J.; Boddy, A. V. Br. J. Cancer 2006, 95, 1229.
30. Kus´nierczyk, H.; Chołody, W. M.; Paradziej-Lukowicz, J.; Radzikowski, C.;
Konopa, J. Arch. Immunol. Ther. Exp. (Warsz). 1994, 42, 415.
31. Koba, M.; Konopa, J. Acta Biochim. Pol. 2007, 54, 297.
32. Lemka, K.; Poindessous, V.; Skladanowski, A.; Larsen, A. K. Mol. Pharmacol.
2004, 66, 1035.
33. Patterson, L. H. Cancer Metastasis Rev. 1993, 12, 119.
34. Wilson, W. R.; Denny, W. A.; Pullen, S. M.; Thompson, K. M.; Li, A. E.; Patterson,
L. H.; Lee, H. H. Br. J. Cancer Suppl. 1996, 27, S43.
DNA stability was studied by establishing the melting tempera-
ture of SSDNA (Salmon Sperm DNA) in the presence of 10 lM of
the triazoloacridin-6-ones in high and low salt phosphate buffers
(High salt—150 mM NaCl, 15 mM sodium citrate, pH 7.0. Low
salt—7.5 mM NaH2PO4, 1 mM EDTA, pH 7.0). Curettes were heated
over the range 37–98 °C and absorbance at 260 nm was read every
degree, using a Perkin–Elmer™ Precisely Lambda 25 UV/vis Spec-
trophotometer. Temperature was controlled by a Perkin-Elmer™
Peltier Temperature Programmer (PTP-6 Peltier System). Determi-
nation of the Tm was read from the mid-point of the hypochromic
transition curves. Relative absorbance was plotted using OriginÒ
35. Mehibel, M.; Singh, S.; Chinje, E. C.; Cowen, R. L.; Stratford, I. J. Mol. Cancer Ther.
2009, 8, 1261.
36. Smith, P. J.; Desnoyers, R.; Blunt, N.; Giles, Y.; Patterson, L. H.; Watson, J. V.
Cytometry 1997, 27, 43.
37. Stratford, I. J.; Stephens, M. A. Int. J. Radiat. Oncol. 1989, 16, 973.