Therapeutic reactivation of mutant p53 protein by quinazoline derivatives

Article abstract of DOI:10.1007/s10637-011-9744-z

Summary: Purpose The human tumour suppressor protein p53 is mutated in nearly half of human tumours and most mutant proteins have single amino acid changes. Several drugs including the quinazoline derivative 1 (CP-31398) have been reported to restore p53

Full text of DOI:10.1007/s10637-011-9744-z

Invest New Drugs (2012) 30:2035–2045
DOI 10.1007/s10637-011-9744-z
SHORT REPORT
Therapeutic reactivation of mutant p53 protein
by quinazoline derivatives
Hamish S. Sutherland & In Young Hwang & Elaine S. Marshall & Brent S. Lindsay &
William A. Denny & Catherine Gilchrist & Wayne R. Joseph & Debra Greenhalgh &
Emma Richardson & Philip Kestell & Angela Ding & Bruce C. Baguley
Received: 20 June 2011 /Accepted: 24 August 2011 /Published online: 13 September 2011
#
Springer Science+Business Media, LLC 2011
Summary Purpose The human tumour suppressor protein
p53 is mutated in nearly half of human tumours and most
mutant proteins have single amino acid changes. Several
drugs including the quinazoline derivative 1 (CP-31398)
have been reported to restore p53 activity in mutant cells.
The side chain of 1 contains a styryl linkage that
compromises its stability and we wished to explore the
activity of analogues containing more stable side chains.
Methods Reactivation of p53 function was measured by
flow cytometry as the ability to potentiate radiation-induced
G1-phase cell cycle arrest and by western blotting to
determine expression of p21WAF1. DNA binding was
measured by competition with ethidium and preliminary
pharmacological and xenograft studies were carried out.
Results Screening of analogues for potentiation of radia-
tion-induced G1-phase cell cycle arrest using NZOV11, an
ovarian tumour cell line containing a p53R248Q mutation,
demonstrated that the (2-benzofuranyl)-quinazoline deriva-
tive 5 was among the most active of the analogues.
Compound 5 showed similar effects in several other p53
mutant human tumour cell lines but not in a p53 null cell
line. 5 also potentiated p21WAF1 expression induced by
radiation. DNA binding affinity was measured and found to
correlate with p53 reactivation activity. Plasma concentra-
tions of 5 in mice were sufficient to suggest in vivo activity
and a small induced tumour growth delay (7 days) of
NZM4 melanoma xenografts was observed. Conclusion
Compound 5 restores p53-like function to a human tumour
cells lines expressing a variety of mutant p53 proteins, thus
providing a basis for the design of further new drugs.
.
.
Keywords TP53 mutation Quinazoline synthesis
.
.
Cell cycle arrest Flow cytometry Tumour growth delay
Introduction
The p53 pathway plays a key role in the response of normal
cells to DNA damage, hypoxia, metabolic stress and oncogene
activation, leading to its description as a “guardian of the
genome” and “a cellular gatekeeper” [1, 2]. In cells with wild-
type p53 function, a low cellular concentration of p53 protein
is maintained through a negative feedback system involving
marking of proteins for proteolysis by the mdm2 ubiquitin
ligase. Under conditions of cellular stress such as DNA
damage or hypoxia, p53 becomes modified, leading to
accumulation of protein. Consequent induction of gene
transcription lead to cell cycle arrest, apoptosis and DNA
repair [3]. Recent data also suggests that p53 can target
mitochondria; the p53 protein interacts with multiple mem-
bers of the bcl-2 family, potentially leading to a permeability
transition of the outer mitochondrial membrane, apoptosis
and necrosis [4]. This could explain the previously observed
activation of inflammatory and vascular responses that
contribute to tumour regression in a murine tumour following
induction of p53 by a tetracycline transactivator protein [5].
Loss of p53 function can occur through mutation or loss of
the TP53 gene, or through altered mdm2 function. Such
changes occur in a significant proportion of human cancers
Electronic supplementary material The online version of this article
(doi:10.1007/s10637-011-9744-z) contains supplementary material,
which is available to authorized users.
:
:
:
:
H. S. Sutherland I. Y. Hwang E. S. Marshall B. S. Lindsay
:
:
:
:
W. A. Denny C. Gilchrist W. R. Joseph D. Greenhalgh
E. Richardson P. Kestell A. Ding B. C. Baguley (*)
:
:
:
Auckland Cancer Society Research Centre, Faculty of Medical
and Health Sciences, The University of Auckland,
Private Bag 92019,
Auckland, New Zealand
e-mail: b.baguley@auckland.ac.nz

2036
Invest New Drugs (2012) 30:2035–2045
and are generally thought to be associated with poor prognosis
[1, 2]. Most TP53 gene mutations in human cancers involve
single amino acid changes in the p53 protein that destabilise
its conformation, leading to dominant-negative inhibition of
wild-type p53 function as well as gain-of-function changes
[6]. Reactivation of p53 function in cells expressing mutant
p53 protein is therefore an important goal in cancer therapy.
Therapeutic strategies aimed at restoring or augmenting
p53 have been a strong focus for cancer research [7].
Research on small molecule drugs that reactivate the
function of mutant p53 has resulted in the identification of
a number of compounds including CP-31398 (1; structure
in Fig. 1), Prima-1 and PhiKan083 [8]. 1 was originally
selected from a library of compounds and found to activate
p53 function in vitro, as well as suppressing the growth of
A375.S2 melanoma (p53R249S) and DLD-1 colon carcino-
ma (p53S241F) tumours in vivo [9]. Subsequent studies
demonstrated that treatment with 1 reduced the incidence of
UV-induced skin cancer in mice, suggesting that mainte-
nance of p53 function could be important in preventing the
development of such cancers [10, 11]. The molecular action
of 1 is not yet understood but a biophysical study has
suggested that 1 acts by binding to DNA rather than to p53
protein itself [12]. 1 induced p53 reporter gene activity and
p21 expression in glioma cell lines with wild-type or
mutant p53, but not in p53null cells [13] and also appeared
to act in a p53-independent manner to increase expression
of bax [12].
Fig. 1 Structures of the quina-
zoline derivatives used in this
study

Invest New Drugs (2012) 30:2035–2045
2037
1 has been reported to be chemically unstable in solution
[12] and the styryl linkage is the most likely component of
the molecule that compromises its stability. We wished to
determine whether side chain analogues of 1 lacking a
styryl linkage might have equivalent activity but increased
stability. Here we report the synthesis of a series of
quinazoline derivatives with stable side chains and meas-
ures of their ability to reactivate p53 function using both
cellular and molecular assays. Several compounds, includ-
ing the (2-benzofuranyl)quinazoline derivative (5), were
found to have p53 reactivating ability equivalent to 1 and
we examined the ability of 5 to reactivate p53 function in
tumour cell lines with a variety of TP53 mutations. We also
measured the DNA binding constants of these analogues
and compared them to their ability to reactivate mutant p53
function. Finally, we examined the antitumour potential of 5
in vivo.
nones (2a–4a) were prepared by the Suzuki coupling of
the chloroquinazolinone (14) with an aryl or heteroaryl
boronic acid, while the quinazolinones (5a–11a) were
synthesised by reaction of anthranilamide (15) with an
aroyl chloride and subsequent cyclisation of the interme-
diate diamide using 5% aqueous KOH and ethanol.
Conversion of the quinazolinones (2a–11a) to the inter-
mediate chloroquinazolines was achieved using either
thionyl chloride and dimethyl formamide, or phosphorus
oxychloride and tetramethylammonium chloride. Chlori-
nation of benzofuran substituted quinazolinones using
phosphorus oxychloride instead of thionyl chloride
avoided the formation of small amounts of by-products
where chlorination had occurred on the benzofuran ring,
separation of these by-products from the desired product
generally required preparative HPLC. Reaction of the
intermediate chloroquinazolines with N,N-dimethyl-1,3-
propanediamine or N,N,2,2-tetramethyl-1,3-propanedi-
amine in refluxing dioxane gave aminoquinazolines (2–
11) and (12) respectively, which were isolated as their
hydrochloride salts. Detailed methods are provided in the
Supplementary Information.
Materials and methods
Synthesis of quinazoline analogues
The main schemes for synthesis analogues shown in
Fig. 1 are indicated in Fig. 2. The styryl quinazolinone
(1a) was synthesised by the sodium acetate mediated
condensation of 13 with 4-methoxybenzaldehyde [14].
Conversion of 1a to the intermediate chloroquinazoline
and subsequent amination [15] gave 1. The quinazoli-
Cell lines and culture conditions
Human tumour cell lines with characterised TP53 mutations
were either established in this centre [16] or were obtained
through the courtesy of the US National Cancer Institute
[17]. Most of the cell lines established in this centre have
Fig. 2 Scheme for synthesis of
quinazoline derivatives. See
also supporting information

2038
Invest New Drugs (2012) 30:2035–2045
been deposited with CellBank Australia. HCT116 cells were
kindly provided by Dr B. Vogelstein. Cell lines were cultured
in α-modified minimal essential medium (α-MEM) contain-
ing insulin, transferrin, selenite, penicillin and streptomycin as
previously described [17]. Growth medium was supple-
mented with 5% foetal bovine serum (FBS) for NZM4 and
NZOV11 cells and 10% FBS for SKMEL28 and HCT116
cells. Growth medium for NZOV11 was further supple-
mented with epidermal growth factor and oestradiol [17].
fractionated by sodium dodecyl sulphate-polyacrylamide
gel electrophoresis (SDS-PAGE; 10% gel) then transferred
to nitrocellulose membranes. The membranes were blocked
in 5% non-fat milk in 0.1% Tween 20 in PBS (blocking
buffer) for 1 h and probed overnight at 4°C with anti-p21
antibody (C-19, sc-397, Santa Cruz Biotechnology) or anti-
actin antibody (MAB1501R, Chemicon) in blocking buffer.
Cells were washed with 0.1% Tween 20 in PBS. Goat anti-
mouse or rabbit IgG horseradish peroxidise (HRP)-conju-
gate was applied as a secondary antibody diluted in
blocking buffer (1:2000; Santa Cruz) for 1 h. Membranes
were washed three times with PBS-Tween 20 and immu-
noreactivity was detected by subjecting the membranes to
chemiluminescence detection system (ECL; Pierce Biotech-
nology). The signal from the membrane was detected and
imaged by LAS-3000 imager (Fujifilm Corporation).
Cell-based assay of p53 function using flow cytometry
Cells were grown in culture dishes (106 per dish) for at least
18 h prior to drug treatment. Unless otherwise stated the cells
were radiated at 9 Gy to induce a DNA damage response and
paclitaxel (200 nM) was used to inhibit cell division. Treated
cells were harvested by trypsinisation, washed in PBS
containing 1% FCS as blocking buffer, resuspended in ice-
cold methanol and stored at −20°C. Prior to analysis,
samples were washed and resuspended in PBS containing
3% FCS and treated with RNase (0.1 mg/ml, Roche Applied
Science, IN, USA) and propidium iodide (0.02 mg/ml,
Sigma-Aldrich) prior to analysis for DNA content on a flow
cytometer (FACScan; Becton Dickinson, CA, USA).
DNA damage was estimated by γ-phosphorylation of
histone H2AX. Following drug exposure, cells were
harvested by trypsinisation, washed in PBS, resuspended
in ice cold ethanol (70%) and stored for at least 18 h at
−20°C. Prior to analysis, samples were washed in blocking
buffer and incubated on a shaker platform for at least 2 h at
225 rpm. Blocking buffer was removed γ-H2AX primary
antibody (200 μl, diluted 1:500 in blocking buffer), was
added and cells were incubated on a shaker platform (2 h,
225 rpm). Samples were washed once in blocking buffer
and incubated in secondary antibody (Alexa fluor 488)
diluted in blocking buffer in the dark on a shaker platform.
Cells were then washed five times, resuspended in blocking
buffer (1 ml) and stored in the dark at 4°C until analysis on
the same day. Ribonuclease and propidium iodide were
added in the dark 10 min prior to analysis to stain cellular
DNA. Fluorescence associated with γ-H2AX (green) and
propidium iodide (red) was measured on a flow cytometer
and results were analysed using Modfit LT software.
Cell proliferation (IC50) assays
Cells were cultured in 96-well plates in triplicate with
drug added in 2-fold serial dilutions. Cultures were
incubated for 48 h and [3H-methyl] labeled thymidine
(20 Ci/mM, 0.04 μCi per well), together with unlabelled
thymidine (TdR; 0.1 μM final concentration) and 5-fluoro-
2′-deoxyuridine (FUdR; 0.1 μM final concentration) were
added to each well over the last 6 h. Cells were gently
dislodged with 0.5 mg/mL pronase in 2 mM EDTA (150 μL
per well) at 37°C, harvested and radioactivity measured in a
liquid scintillation counter (Wallac, OY, Finland).
Pharmacokinetics
Female C57 mice were treated intraperitoneally with 5
(100 mg/kg). Blood was collected from terminally anaes-
thetised mice 1–4 h later (3 mice per time point) in 1.5 ml
microfuge tubes containing 7.5% EDTA (20 μl). Tubes were
centrifuged (5,000g; 10 min) and plasma samples were
removed and deproteinised by addition of 1 ml of ice-cold
acetonitrile:water (3:1) to 100 μl plasma. Compound 9
(0.1 μM) was added as an internal standard. After mixing
and centrifugation (3,000g; 15 min; 4°C), the supernatants
were concentrated in a centrifugal vacuum concentrator, re-
constituted in 100 μl mobile phase (80% acetonitrile:45 mM
formate buffer (35:63), injected (5 μl) into a LUNA C18 5 μ
50×0.5 mm stainless steel (Phenomenex) and run at room
temperature (initial flow rate 15 μl/min) on an Agilent
capillary liquid chromatograph with ion trap detection.
Determination of p53 function by western immunoblotting
Total cellular protein extracts of the treated cells were
prepared using a buffer containing 10 mM Tris-HCl pH 7.5,
1 mM EGTA, 150 mM NaCl, 0.5% deoxycholate, 1% NP-
40, 1 mM Na3VO4, 1 mM NaF, 0.1% SDS and protease
inhibitor cocktail (P1860, Sigma-Aldrich, MO, USA). The
resulting supernatants containing the whole cell protein
extracts were collected. Protein samples (30 μg) were
Xenograft studies
Rag1 immunodeficient mice were bred and housed at
constant temperature and humidity with sterile bedding
and food (Vernon Jansen Unit, the University of Auckland).

Invest New Drugs (2012) 30:2035–2045
2039
All experiments were approved by the University of
Auckland Animal Ethics Committee and conformed to
international guidelines [18]. NZM3 cell lines were grown
in culture and inoculated subcutaneously (107 cells/mouse)
in one flank. Tumours were allowed to grow to a diameter
of 4–6 mm before treatment with 5, which was dissolved in
water and injected intraperitoneally. Tumour size was
measured twice weekly and tumour volumes were calcu-
lated as 0.52a2b, where a and b are the minor and major
axes of the tumour, respectively. Tumour growth delay was
determined as the difference in the number of days required
for the control versus treated tumours to increase to ten
times the initial volume.
DNA flow cytometry profile of untreated cells was typical
of a proliferating population. Radiation (9 Gy) was used to
induce a DNA damage response and paclitaxel to inhibit
cell division [17, 19]. Radiation at 9 Gy followed by
exposure to paclitaxel for 24 h caused movement of most
G1-phase cells to G2/M-phase, with no cells dividing and a
small proportion entering another cell cycle without dividing
(Fig. 3b). Addition of 1 together with radiation and
paclitaxel, followed by culture for 24 h, increased the
proportion of cells that remained in G1-phase; this proportion
was dependent on the concentration of 1 (Fig. 3c–e).
Addition of 1 alone to untreated cultures did not increase
the proportion of G1- or G2-phase cells (data not shown).
Screening of quinazoline derivatives in a cell-based assay
Results
Derivatives with differing side chains at the 2-position
(Fig. 2) were synthesised and tested in the above assay
using NZOV11 at concentrations of 2.2, 6.7 and 20 μM. An
advantage of the cell-based assay system was that other
drug effects on cell growth, including the induction of cell
death, could also be monitored by flow cytometry. The
results are shown in Fig. 4. Compounds 5 and 7 were the
Development of a cellular assay for reactivation of mutant
p53
An ovarian cancer cell line (NZOV11) expressing mutant
p53 protein with a single amino acid change (p53R248Q
)
was used to screen for activity. As shown in Fig. 3a, the
A
B
G1
G2
G1
G2
G1
G2
C
D
G1
G2
E
G1
G2
Fig. 3 Flow cytometry profiles of NZOV11. Red fluorescence from
propidium staining is used as an indicator of DNA content (positions
of G1- and G2-phase cell populations are marked). a Untreated cells.
B. Irradiation at 9 Gy followed by culture in the presence of paclitaxel
(200 nM) for 24 h, showing induction of G2/M-phase arrest. C, D, E.
Accumulation of cells in G1- and S-phases following addition of 1 at
concentrations of 2.2, 6.7 and 20 μM, respectively, in addition to
irradiation and addition of paclitaxel

2040
Invest New Drugs (2012) 30:2035–2045
Fig. 4 Comparison of the abil-
ity of quinazoline derivatives at
concentrations of 2.2 μM (white
bars), 6.7 μM (gray bars) and
20 μM (black bars) to retain
NZOV11 cells in G1-phase 24 h
after irradiation (9 Gy) and
40
30
20
10
0
addition of paclitaxel (200 nM)
1
2
3
4
5
6
7
8
9
10 11 12
Compound number
most effective of the analogues in the ability to induce
retention of cells in G1-phase. Compound 5 was selected
for further study.
in the negative controls (i.e. treated with radiation and
paclitaxel alone) varied among the cell lines, and the results
were scored as percentage increases in G1-phase content. As
shown in Table 1, 5 caused an increase in G1-phase content
in all TP53 mutant cell lines tested except for the line that
had a frameshift mutation. A similar experiment using
HCT116 p53−/− cells showed that 5 did not induce G1-phase
arrest, also suggesting that the effect was p53-dependent.
Induction of p53 transcription products
Since p21WAF1, a transcription product of p53, is thought to
be responsible for the induction of G1-phase arrest, its
cellular concentration was measured by western blotting.
As shown in Fig. 5, both 1 and 5 induced p21 expression in
NZOV11 cells in a concentration-dependent manner.
Increases in p21 paralleled increases in the proportion of
G1-phase cells measured by flow cytometry.
Effects of compound 5 on NZOV11 cells in the absence
of radiation
Addition of 5 in the presence of paclitaxel but without
irradiation, followed by culture for 24 h, also increased the
proportion of G1-phase cells in a dose-dependent manner,
although the effects were smaller than those observed in the
presence of radiation (Fig. 6b). Addition of 1 under similar
conditions also increased the proportion of G1-phase cells
in a dose dependent manner (data not shown).
Effects of compound 5 on other p53 mutant lines
We next determined whether 5 could reactivate p53 in cell
lines expressing other mutant p53 proteins. Because cell
lines grew at different rates, the proportion of G1-phase cells
Fig. 5 Expression of p21 in
NZOV11 cells treated for 24 h
with 1 and 5, at the indicated
concentrations, after irradiation
NZOV11
Compound 1
Compound 5
at 9 Gy. The loading control was
β-actin
p21
p21
β-actin
β-actin

Invest New Drugs (2012) 30:2035–2045
2041
40
30
20
10
0
Table 1 Properties of cell lines used in this study and their responses
to 5. Cells in each case were irradiated to 9 Gy and grown in the
presence of paclitaxel (200 nM) 5 (20 μM) for 24 h before
measurement of G1-phase cell content
A
B
Cell line
Tumour
type
p53 status
Induced
G1-arrest %
NZM4
Melanoma
Melanoma
Melanoma
Melanoma
Ovarian
Ovarian
Ovarian
Ovarian
Ovarian
Ovarian
Ovarian
Breast
p53S241T
p53S241T
p53G245S
p53C145V
p53H178T
p53Y220D
p53S215G
p53Y220F
p53R248Q
p53Y126_K132del
p53H179R
p53L194F
p53R273H
p53R273H
p53null
50
22
20
26
16
14
17
11
7
NZM7
SKMEL2
SKMEL28
NZOV1
NZOV2
NZOV8
NZOV10
OVCAR3
OVCAR8
SKOV3
T47D
0
4
17
5
HT29
Colorectal
Colon
SW620
HCT116 p53−/−
2
Colon
0
Fig. 6 a Comparison of the ability of 5 at concentrations of 2.2 μM
(white bars), 6.7 μM (gray bars) and 20 μM (black bars) to retain
NZM3 melanoma cells (p53 wild-type) in G1-phase 24 h after
radiation (2 Gy) and addition of paclitaxel (200 nM). The effect of
adding nutlin-3 (10 μM) to the cultures is indicated by the label (+N).
b Comparison of the ability of 5 at concentrations of 2.2 μM (white
bars), 6.7 μM (gray bars) and 20 μM (black bars) to retain NZOV11
ovarian cancer cells (p53 mutant) in G1-phase 24 h after after radiation
(2 Gy) and addition of paclitaxel (200 nM). The effect of adding
nutlin-3 (10 μM) to the cultures is indicated by the label (+N)
Effects of compound 5 on cells expressing wild-type p53
One interpretation of the above results is that 5 stabilises
p53 in an active conformation, leading to increased
downstream effects. If this were the case, 5 might also
stabilise wild-type p53, leading to enhanced p53 responses.
Cell cycle responses were therefore measured using the
NZM3 melanoma line, which expresses wild-type p53
protein [16]. Because radiation at 9 Gy alone induced a
high degree of G1-phase arrest (45%) in this cell line, a
lower radiation dose (2 Gy) was selected. Radiation and
paclitaxel alone resulted in 8.5% G1-phase content, and
addition of 5 induced a dose-dependent increase in G1-
phase content (Fig. 6a), consistent with an effect on wild-
type p53 function. The drug nutlin-3 is known to enhance
the activity of wild-type p53 expression by inhibiting
mdm2-mediated degradation of p53 [20], and its effect
was investigated. At a concentration of 10 μM, nutlin alone
induced a high degree of G1-phase arrest (36%), consistent
with its enhancement of a p53 response to DNA damage.
Addition of 5 as well as nutlin did not induce further arrest
(Fig. 6a).
after 24 h was found to be 6% with or without the addition
of nutlin-3 at the time of irradiation (data not shown). The
effect of 5 was most evident at a concentration of 6.7 μM
(Fig. 6b). The proportion of G1-phase cells in NZM11 cells
was 8% in the absence of nutlin-3, but the addition of
nutlin-3 (10 μM) increased the degree of G1-phase arrest
induced by 5 to 15% (Fig. 6b), suggesting increased p53
activity.
Relationship between p53 reversion activity and DNA
binding
Since 1 is reported to bind DNA [12], it is possible that its
effects are mediated by DNA. To determine whether DNA
binding affinity was related to the biological effects of the
quinazoline derivatives, DNA binding was measured by
competition with ethidium as previously described [21].
DNA binding affinity for both poly(dAT).poly(dAT) was
found to correlate (r=0.84; p<0.01) with the induction of
p53 G1-phase arrest in the cell-based assay (Fig. 7). A
The results of previous sections suggests that 5 stabilises
mutant p53, and if this then behaved like wild-type p53
protein one might expect nutlin-3 to further enhance its
stability. The effect of addition of nutlin-3 in addition to 5
on the NZOV11 p53 mutant cell line was therefore
examined. Following radiation at 2 Gy and growth in
paclitaxel, the proportion of G1-phase cells in NZM3 cells

2042
Invest New Drugs (2012) 30:2035–2045
40
topoisomerase II was also tested utilising Jurkat, a TP53
null human leukaemia cell line, together with two derived
cell sub-lines (JLA and JLD) that had been selected for
resistance to amsacrine and doxorubicin, respectively [24]
and expressed low amounts of topoisomerase II protein
[25]. As shown in Table 2, neither line was cross-resistant
to 1 or 5, consistent with the hypothesis that they are not
topoisomerase poisons. It was of interest that 5 showed
slightly greater activity against the resistant sub-lines than
against the parental line, but the reason for this was not
investigated.
30
20
10
0
In vivo activity of 5
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
DNA binding affinity
We determined whether plasma concentrations of 5 that
were active in reactivating p53 could be achieved in vivo.
An analytical procedure for the determination of 5
concentrations in mouse plasma was developed using
high performance liquid chromatography and ion trap
mass spectrometry. 5 was well tolerated in C57Bl6 mice
with a maximum tolerated daily intraperitoneal dose of
100 mg/kg. The maximum plasma concentration of
8.5 μM (Fig. 9) was in the range of concentrations found
to be active in the cell-based assays. We next investigated
the ability of 5 to affect the growth of the NZM4 human
melanoma line (p53S241T) as a xenograft growing in
immunodeficient mice. Treatment with 5 alone (100 mg/kg)
had no effect on tumour growth while radiation (2 Gy)
resulted in a growth delay of 1.5 days. Administration of
radiation (2 Gy) and 5 in combination resulted in a small
tumour growth delay of 7 days (Fig. 9).
Fig. 7 Relationship between DNA binding affinity and p53 reactiva-
tion. DNA binding was measured by competition with propidium
iodide for binding sites on the double-stranded synthetic DNA poly
(dAT).poly(dAT)
slightly lower correlation with binding to poly(dGC).poly
(dGC) was also observed (data not shown).
Relationship between p53 reversion activity
and topoisomerase II poisoning
Many DNA binding drugs, particularly acridine derivatives,
interact with the enzyme topoisomerase II and induce DNA
damage in the form of double-stranded breaks, with
consequent cytotoxicity [22]. Furthermore, several active
analogues of 1 are acridine derivatives [9]. The DNA
binding ability of 5 raised the question of whether it also
induced DNA damage. The DNA damage response of
NZOV11 cells was assessed by measuring γ-
phosphorylation of histone H2AX by two dimensional
flow cytometry, with propidium staining of DNA in the
second dimension (Fig. 8). As a positive control, the
topoisomerase II poison amsacrine [23] was tested. H2AX
phosphorylation was evident in the case of amsacrine but
not with 5 or 1, which instead each induced depletion of S-
phase cells. The activity of 1 and 5 as poisons of
Discussion
A cell-based assay measuring the potential of drugs to
potentiate the induction of G1-phase arrest in a human p53
mutant cell line was used to evaluate a series of quinazoline
derivatives related to 1. This screen identified several
analogues of 1 that lacked the relatively unstable styryl
Untreated
Compound 1 (10 µM)
Compound 5 (10 µM)
Amsacrine (0.3 µM)
G1G2
G1G2
G1G2
G1G2
Fig. 8 Flow cytometry profiles of NZOV11 cells either untreated or
exposed to 1, 5 or amsacrine at the indicated concentrations. The x-
axis indicates red fluorescence from propidium staining as an indicator
of DNA content (positions of G1- and G2-phase cell populations are
marked). The y-axis indicates green staining from antibody to
phosphorylated histone H2AX as a measure of DNA damage

Invest New Drugs (2012) 30:2035–2045
2043
Table 2 Growth inhibitory concentrations of 1 and 5 on the Jurkat
human leukaemia cell line (JL) and on two sub-lines (JLA and JLD)
that express low amounts of topoisomerase II [25]
drugs to reactivate mutant p53 function and suggests that
other active, chemically stable analogues of 5 may be
found.
IC50 values (μM)
JL
JLA
JLD
Since the aim of this study was to identify drugs that
potentiated a p53 responses to radiation-induced DNA
damage, it was surprising to find in a control experiments
that addition of 5 or 1 to cultures in the absence of ionising
radiation also induced G1-phase arrest, although of smaller
magnitude than when used in combination with radiation at
2 Gy or 9 Gy (Fig. 6). A likely explanation for this effect is
that 5 potentiates the p53 response to endogenous DNA
damage; many tumour cell lines have short telomeres or
other chromosomal characteristics that induce an endoge-
nous DNA damage response [26]. The positive relationship
between induced G1-phase arrest and DNA binding affinity
(Fig. 7), together with the known ability of many DNA
binding drugs to induce DNA damage through effects on
the enzyme topoisomerase II [27], raises the question of
whether these drugs themselves induce DNA damage as part
of their cellular cycle effects. However, phosphorylation of
histone H2AX, a standard indicator of DNA damage, was
not induced by 5 or 1 but was induced by the DNA binding
topoisomerase II poison amsacrine (Fig. 8).
1
3.5
3.0
2.0
5
8.0
5.4
5.5
Amsacrine
Doxorubicin
0.027
0.007
0.26
0.028
0.30
0.112
side chain but retained comparable activity (Fig. 4). Some
structure-activity relationships can be gleaned from these
data. The poor activity of 2, 3, 4 and 6 attest to a clear
preference for benzofuran or benzothiophene side chains,
which place the phenyl group of the aromatic side chain in
approximately the same position as that of 1. Compounds 7
and 5 were the most active with the former showing more
activity at 20 μM and the latter more activity at 6.7 μM
(Fig. 3). The activity of compounds 8–11 suggested
minimal bulk tolerance around the benzofuran while lack
of activity of 12 suggested reduced tolerance around the
aliphatic 3-(dimethylamino)propyl side chain.
Compound 5 was investigated further and shown to be
capable of potentiating G1-phase arrest in several cell lines
containing different TP53 mutations, but not in a TP53 null
line or in a line containing a frameshift mutation of the
TP53 gene (Table 1). Addition of 1 or 5 to NZOV11 cells
increased expression of p21WAF1, a cyclin-dependent kinase
inhibitor product of a p53-target gene, consistent with p53
mediation of G1-phase arrest. Taken together with previous
data for 1, the results suggest that 5 acts to restore the
function of a variety of mutant p53 proteins. The cell-based
screen provides a rapid means of assessing the ability of
As shown in Table 1, 5 potentiated the p53 response to
radiation in a variety of cell lines with differing mutations
in the TP53 gene, but not in a cell line lacking the TP53
gene. Unexpectedly, 5 also potentiated the p53 response to
100
Control
Radiation alone (2 Gy)
Radiation (2 Gy) + 5 at 0 hr
Radiation (2 Gy) + 5 at 4 hr
10
8
10
6
4
1
2
0
20
40
60
Time (days)
0
0
1
2
3
4
Fig. 10 Effect of radiation and 5 on the growth of NZM4 melanoma
xenografts in immunodeficient mice. Mice were either untreated (•),
treated with radiation at 2 Gy (○), treated with radiation plus 5
(100 mg/kg) simultaneously (Δ) or treated with radiation and 4 h later
with 5 (100 mg/kg) (∇)
Time (hours)
Fig. 9 Plasma pharmacokinetics of 5 in mice following a single i.p.
injection (100 mg/kg; female C57BL mice)

2044
Invest New Drugs (2012) 30:2035–2045
radiation in a cell line (NZM3) that expresses wild-type p53
protein (Fig. 4). However, these observations are all
consistent with the hypothesis that 1 and 5 stabilise active
conformations of both wild-type and mutant p53 proteins,
allowing them to bind more productively to consensus p53
binding sites. Wild-type p53 protein, although more stable
than mutant p53 proteins, is still intrinsically unstable,
particularly in the presence of mdm2 [28]. These observa-
tions are also consistent with reports 1 potentiates the
effects of both wild-type and mutant p53 [29]. The
cytoplasmic form of the wild-type p53 protein is normally
maintained in a tetrameric form with all four subunits
geometrically equivalent [30] but in response to stress
signals the protein is modified and the p53 proteins then
bind as a linear tetramer (“dimer of dimers”) to consecutive
pairs of half-promoter sites on the DNA [31].
The receptor for 1 is not known and previous studies
have failed to detect a physical interaction between 1 and
the p53 core domain [12]. 1 binds to DNA [12] and the
positive relationship between DNA binding affinity and p53
reactivation of these analogues (Fig. 7) suggests that DNA
binding could be involved in the action of these com-
pounds. Alternatively, the drugs could stabilise the p53
protein indirectly by interaction with proteins that maintain
an active p53 conformation. Further work is in progress to
investigate these relationships using a larger group of
compounds.
Following administration of 5 to mice at a non-toxic
dose, plasma drug concentrations were found to be in the
same range as those causing the induction of G1-phase
arrest in the cell-based system (Figs. 4 and 9), suggesting
that the drug could have in vivo activity. A preliminary
study of the effect of 5 on a tumour xenograft of the p53
mutant melanoma line NZM3 suggested the presence of a
low level of activity (Fig. 10). The original demonstration
of in vivo activity of 1 was carried out in the absence of an
added DNA damaging agent [9] but in another experiment
(not shown) 5 was shown to lack activity against NZM3
cells in the absence of radiation.
In conclusion, this study has provided examples of new
small molecule drugs with good chemical stability and the
ability to reactivate mutant p53. The flow cytometry
approach has an advantage in screening procedures because
it allows estimation of both induced G1-phase and of other
induced cellular changes including the induction of apo-
ptosis. The p53 protein has a complex action that includes
effects on both gene transcription and mitochondrial
function [3, 4] and reactivation by small molecule drugs
remains as an important goal of medicinal chemistry. More
than a decade has passed since the publication of the first
drugs that can reactivate mutant p53 and although many
candidate molecules have been described, clinical efficacy
has not yet been demonstrated [7]. Further research may
require integrated combination of molecular, cellular and
whole animal studies.
Acknowledgement The authors are grateful to the Auckland Cancer
Society for financial support.
References
1. Lane DP (1992) Cancer. p53, guardian of the genome. Nature
358:15–16
2. Levine AJ (1997) p53, the cellular gatekeeper for growth and
division. Cell 88:323–331
3. Sengupta S, Harris CC (2005) p53: traffic cop at the
crossroads of DNA repair and recombination. Nat Rev Mol
Cell Biol 6:44–55
4. Han J, Goldstein LA, Hou W, Gastman BR, Rabinowich H (2010)
Regulation of mitochondrial apoptotic events by p53-mediated
disruption of complexes between antiapoptotic Bcl-2 members
and Bim. J Biol Chem 285:22473–22483
5. Xue W, Zender L, Miething C, Dickins RA, Hernando E,
Krizhanovsky V, Cordon-Cardo C, Lowe SW (2007) Senescence
and tumour clearance is triggered by p53 restoration in murine
liver carcinomas. Nature 445:656–660
6. Cadwell C, Zambetti GP (2001) The effects of wild-type p53
tumor suppressor activity and mutant p53 gain-of-function on cell
growth. Gene 277:15–30
7. Cheok CF, Verma CS, Baselga J, Lane DP (2010) Translating p53
into the clinic. Nature Rev Clin Oncol 8:25–37
8. Brown CJ, Lain S, Verma CS, Fersht AR, Lane DP (2009)
Awakening guardian angels: drugging the p53 pathway. Nature
Rev Cancer 9:862–873
9. Foster BA, Coffey HA, Morin MJ, Rastinejad F (1999) Pharma-
cological rescue of mutant p53 conformation and function.
Science 286:2507–2510
10. Tang X, Zhu Y, Han L, Kim AL, Kopelovich L, Bickers DR, Athar
M (2007) CP-31398 restores mutant p53 tumor suppressor
function and inhibits UVB-induced skin carcinogenesis in mice.
J Clin Invest 117:3753–3764
11. El Deiry WS (2007) Targeting mutant p53 shows promise for
sunscreens and skin cancer. J Clin Invest 117:3658–3660
12. Rippin TM, Bykov VJ, Freund SM, Selivanova G, Wiman KG,
Fersht AR (2002) Characterization of the p53-rescue drug CP-
31398 in vitro and in living cells. Oncogene 21:2119–2129
13. Wischhusen J, Naumann U, Ohgaki H, Rastinejad F, Weller M
(2003) CP-31398, a novel p53-stabilizing agent, induces p53-
dependent and p53-independent glioma cell death. Oncogene
22:8233–8245
14. Ram VJ, Farhanullah, Tripathi BK, Srivastava AK (2003)
Synthesis and antihyperglycemic activity of suitably functional-
ized 3H-quinazolin-4-ones. Bioorg Med Chem 11:2439–2444
15. Coffey HA, Connell RD, Foster BA, Rastinejad F. Methods and
compositions using hydrophobic group- and cationic group-
containing compounds for restoring conformational stability of a
protein of the p53 family. (PCT Int. Appl.CODEN: PIXXD2 WO
2000032175 A2 20000608 CAN 133:26844 AN 2000:383903
CAPLUS), 1–76. 2000
16. Parmar J, Marshall ES, Charters GA, Holdaway KM, Shelling
AN, Baguley BC (2000) Radiation-induced cell cycle delays and
p53 status of early passage melanoma cell lines. Oncol Res
12:149–155
17. Baguley BC, Marshall ES, Whittaker JR, Dotchin MC, Nixon J,
McCrystal MR, Finlay GJ, Matthews JH, Holdaway KM, van Zijl

Invest New Drugs (2012) 30:2035–2045
2045
P (1995) Resistance mechanisms determining the in vitro
sensitivity to paclitaxel of tumour cells cultured from patients
with ovarian cancer. Eur J Cancer 31A:230–237
24. Finlay GJ, Baguley BC, Snow K, Judd W (1990) Multiple
patterns of resistance of human leukemia cell sublines to
amsacrine analogues. J Natl Cancer Inst 82:662–667
18. Workman P, Aboagye EO, Balkwill F, Balmain A, Bruder G,
Chaplin DJ, Double JA, Everitt J, Farningham DA, Glennie MJ,
Kelland LR, Robinson V, Stratford IJ, Tozer GM, Watson S,
Wedge SR, Eccles SA (2010) Guidelines for the welfare and use
of animals in cancer research. Br J Cancer 102:1555–1577
19. Basse B, Baguley BC, Marshall ES, Joseph WR, van Brunt B,
Wake GC, Wall DJN (2004) Modelling cell death in human
tumour cell lines exposed to the anticancer drug paclitaxel. J Math
Biol 49:329–357
20. Vassilev LT, Vu BT, Graves B, Carvajal D, Podlaski F, Filipovic Z,
Kong N, Kammlott U, Lukacs C, Klein C, Fotouhi N, Liu EA
(2004) In vivo activation of the p53 pathway by small-molecule
antagonists of MDM2. Science 303:844–848
21. Baguley BC, Denny WA, Atwell GJ, Cain BF (1981) Potential
antitumor agents. 34. Quantitative relationships between DNA
binding and molecular structure for 9-anilinoacridines substituted
in the anilino ring. J Med Chem 24:170–177
22. Denny WA, Baguley BC (1994) New anticancer acridines. In:
Neidle S, Waring MJ (eds) Molecular aspects of anticancer drug-
DNA interaction. London, pp 270–311
25. Drummond CJ, Finlay GJ, Broome L, Marshall ES, Richardson E,
Baguley BC (2010) Action of SN 28049, a new DNA binding
topoisomerase II-directed antitumour drug: comparison with
doxorubicin and etoposide. Invest New Drugs in press
26. Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K,
Guldberg P, Sehested M, Nesland JM, Lukas C, Orntoft T, Lukas
J, Bartek J (2005) DNA damage response as a candidate anti-
cancer barrier in early human tumorigenesis. Nature 434:864–870
27. Baguley BC (1991) DNA intercalating anti-tumour agents.
Anticancer Drug Des 6:1–35
28. Sasaki M, Nie L, Maki CG (2007) MDM2 binding induces a
conformational change in p53 that is opposed by heat-shock
protein 90 and precedes p53 proteasomal degradation. J Biol
Chem 282:14626–14634
29. Xu J, Timares L, Heilpern C, Weng Z, Li C, Xu H, Pressey JG,
Elmets CA, Kopelovich L, Athar M (2010) Targeting wild-type
and mutant p53 with small molecule CP-31398 blocks the growth
of rhabdomyosarcoma by inducing reactive oxygen species-
dependent apoptosis. Cancer Res 70:6566–6576
30. McLure KG, Lee PW (1998) How p53 binds DNA as a tetramer.
EMBO J 17:3342–3350
31. Kitayner M, Rozenberg H, Kessler N, Rabinovich D, Shaulov L,
Haran TE, Shakked Z (2006) Structural basis of DNA recognition
by p53 tetramers. Mol Cell 22:741–753
23. Nelson EM, Tewey K, Liu LF (1984) Mechanism of antitumor
drug action: poisoning of mammalian DNA topoisomerase II on
DNA by 4′-(9-acridinylamino)-methanesulfon-m-anisidide. Proc
Natl Acad Sci U S A 81:1361–1365