Identification of a highly potent and selective DNA-dependent protein kinase (DNA-PK) inhibitor (NU7441) by screening of chromenone libraries

Article abstract of DOI:10.1016/j.bmcl.2004.09.060

A solution-phase multiple-parallel synthesis approach was employed for the preparation of 6-, 7- and 8-aryl-substituted chromenone libraries, which were screened as inhibitors of the DNA repair enzyme DNA-dependent protein kinase (DNA-PK). These studies r

Full text of DOI:10.1016/j.bmcl.2004.09.060

Bioorganic & Medicinal Chemistry Letters 14 (2004) 6083–6087
Identification of a highly potent and selective
DNA-dependent protein kinase (DNA-PK) inhibitor (NU7441)
by screening of chromenone libraries
Justin J. J. Leahy,a Bernard T. Golding,a Roger J. Griffin,a,* Ian R. Hardcastle,a
Caroline Richardson,b Laurent Rigoreaub and Graeme C. M. Smithb
aNorthern Institute for Cancer Research, School of Natural Sciences-Chemistry, Bedson Building,
University of Newcastle, Newcastle Upon Tyne NE17RU, UK
bKuDOS Pharmaceuticals Ltd, 327 Cambridge Science Park, Milton Rd, Cambridge CB4 4WG, UK
Received 23 July 2004; revised 17 September 2004; accepted 21 September 2004
Available online 28 October 2004
Abstract—A solution-phase multiple-parallel synthesis approach was employed for the preparation of 6-, 7- and 8-aryl-substituted
chromenone libraries, which were screened as inhibitors of the DNA repair enzyme DNA-dependent protein kinase (DNA-PK).
These studies resulted in the identification of 8-dibenzothiophen-4-yl-2-morpholin-4-yl-chromen-4-one (NU7441) as a highly potent
and selective DNA-PK inhibitor (IC50 = 14nM), exhibiting ATP-competitive inhibition kinetics.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
stress responses, which also includes the mammalian
proteins ATM, ATR, mTOR and hSMG1.3 The intact
enzyme apparatus comprises a catalytic unit (DNA-
PKcs) and a heterodimeric regulatory factor (Ku70/
Ku80). Detection and binding of Ku to DNA DSBs is
thought to precede recruitment of DNA-PKcs to gener-
ate the active serine/threonine kinase. Once bound to the
site of a DNA DSB, DNA-PK is believed to act as a
scaffold for the other components of the NHEJ path-
way. The kinase activity of DNA-PK appears essential
for DNA DSB repair, as DNA-PK defective cell lines
The ability of cancer cells to repair DNA damage is an
important determinant of their susceptibility to chemo-
or radio-therapy in the treatment of cancer.1 The
DNA-dependent protein kinase (DNA-PK) plays an
important role in the detection and repair of DNA dou-
ble-strand breaks (DSBs) via the non-homologous end-
joining (NHEJ) pathway.2 DNA-PK is a member of the
phosphatidylinositol (PI) 3-kinase related kinase
(PIKK) family involved in the signalling of cellular
R
O
O
O
O
R
8
X
N
O
O
N
O
O
N
N
N
R 7
N
6
O
O
1; R = 8-Ph
7; R = H
2; X = O
3; X = S
4; R = H
5; R = Me
6
Keywords: Chromenones; DNA repair; DNA-PK; Inhibitors.
*
Corresponding author. Tel./fax: +44 191 222 8591; e-mail: r.j.griffin@ncl.ac.uk
0960-894X/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.09.060

6084
J. J. J. Leahy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 6083–6087
are hypersensitive to DNA-damaging agents.2,4 Impor-
tantly, inhibition of DNA-PK has been demonstrated
to elicit chemo- and radio-sensitization, and to potenti-
ate the cytotoxicity of ionizing radiation and a number
of DSB-inducing agents in vitro.5,6
hydroxyacetophenone(8)byamodificationofthemethod
of Vlahos et al. (Scheme 1).15 Thus, cyclocondensa-
tion of 8 with CS2 gave the chromene-2-thione (9), with
subsequent conversion into the S-ethyl derivative (10)
enabling direct introduction of the 2-morpholine substit-
uent to furnish 11 in good yield. The corresponding 7-
and 8-bromo-2-hydroxyacetophenenones were not read-
ily accessible, and an alternative approach utilizing the
7- and 8-chromenone triflates (19, 20) was employed,
as shown in Scheme 2. Dihydroxybenzoate esters (13,
14), prepared from the corresponding phenols by
standard carboxylation–esterification methods, were
converted into the monotriflate esters (15, 16) on treat-
ment with triflic anhydride under basic conditions. The
regioselectivity of the sulfonation of esters (13, 14) is
presumably due to deactivation of the 2-hydroxyl func-
tion by the neighbouring carbomethoxy group.
Our studies have centred on the development of potent
and selective DNA-PK inhibitors, suitable for clinical
evaluation as agents to enhance the cytotoxicity of
DNA-damaging anticancer therapies. We have previ-
ously utilized the chromen-4-one LY294002 (1), a non-
specific ATP-competitive PIKK inhibitor (IC50 for
DNA-PK = 1.2lM),7,10 as a template for DNA-PK
inhibitor design, and have identified novel pyranone
(2), thiopyranone (3), chromenone (4, 5) and pyrimid-
oisoquinolone (6) pharmacophores.8–10 A number of
these inhibitors, including NU7026 (4; IC50 = 0.23lM)
and NU7163 (5; IC50 = 0.19lM), exhibit good selec-
tivity for DNA-PK over other PIKK family members,
and have been demonstrated to act as radio- and
chemo-sensitizers in several human tumour cell lines in
vitro.11,12
Reaction with N-acetylmorpholine gave the b-keto-
amides (17, 18), and ring closure to the required 7-,
and 8-chromen-4-one triflate esters (19, 20) occurred
smoothly on treatment with triflic anhydride.16 The 6-,
7-, and 8-aryl-substituted chromen-4-one libraries (12,
21, 22) were synthesized by Suzuki–Miyaura reactions
of the 6-bromochromenone (11) and the 7- and 8-
chromenone triflates (19, 20), with 64 commercially
available arylboronic acids of diverse structure.17 Reac-
tions were conducted in a solution-phase multiple-paral-
With the objective of improving activity further, we
wished to investigate the introduction of functionality
at the 6-, 7-, and 8-positions on the chromenone tem-
plate (7), as the ATP-binding domain of DNA-PK had
previously been found10 to tolerate substitution at these
positions. In this paper we describe the use of a solution-
phase multiple-parallel synthesis approach for the prep-
aration of 6-, 7-, and 8-aryl-substituted chromenone
libraries, and the identification of NU7441 as a highly
potent and selective DNA-PK inhibitor.
lel format using
a
GreenhouseTM reactor station
(Radleys). The reaction progress was monitored by
LC–MSanalysis, and product purification achieved by
semi-preparative HPLC.
3. Results and discussion
2. Chemical synthesis
Our previous studies have demonstrated that the chro-
men-4-one template (7) can serve as a versatile platform
for the design of selective DNA-PK inhibitors, and
structure–activity relationships are beginning to emerge
for this pharmacophore (Fig. 1).8–10 A core six-mem-
bered heterocyclic ring bearing a 4-carbonyl or thiocar-
The required chromen-4-one libraries were prepared by
palladium catalyzed cross-coupling (Suzuki–Miyaura)
reactions13,14 with suitably functionalized chromen-4-
one derivatives. 6-Bromochromen-4-one (11) was readily
prepared from commercially available 5-bromo-2-
OH
O
O
S
SEt
i
ii
Me
Br
Br
N
Br
O
O
OH
(8)
(9)
(10)
O
O
iv
O
N
O
iii
6
Ar
Br
O
O
(11)
(12)
Scheme 1. Reagents and conditions: (i) CS2, KOtBu, toluene, 25°C, 40–50%; (ii) EtI, K2CO3, acetone, reflux, 64%; (iii) morpholine, HOCH2CH2OH,
120°C, 63%; (iv) cat. Pd(PPh3)4, K2CO3, ArB(OH)2, dioxane, 90°C, 10–30%.

J. J. J. Leahy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 6083–6087
6085
3
3
3
OH
OH
OH
4
4
4
ii
O
i
HO
TfO
TfO
N
CO2Me
CO2Me
O
O
(13, 14)
(15, 16)
(17, 18)
O
O
8
8
O
N
O
O
N
iv
7
7
iii
TfO
Ar
O
(19, 20)
(21); 7-Aryl series
(22); 8-Aryl series
Scheme 2. Reagents and conditions: (i) pyridine, cat. DMAP, Tf2O, DCM, 0°C, 37–40%; (ii) N-acetylmorpholine, LDA, THF, ꢁ78 ! 25°C,
35–40%; (iii) Tf2O, DCM, 0°C, 16h, 41%; (iv) cat. Pd(PPh3)4, K2CO3, ArB(OH)2, dioxane, 90°C, 10–30%.
initial concentration of 0.5lM.18 Interestingly, only
modest inhibitory activity (ꢀ40–60% inhibition) was
observed at this concentration for members of the 6-
and 7-aryl-substituted chromen-4-one libraries (12 and
O
X
Y
N
21), with no compounds proving more active than the
parent 2-morpholinylchromen-4-one (7, IC50 = 1.3lM).
By contrast, eight members of the 8-aryl-substituted
chromen-4-one library (22a–h) were found to exhibit
promising activity at 0.5lM, and IC50 values were
determined for these hits (Table 1).
Me
Z
X = O, S, or N (NH)
Y = O or S; Z = C, N
Four of the library members (22a–c,f) were of compara-
ble potency with IC50 values of 0.6–0.9lM, while the 2-
thienyl derivative (22d) was some twofold more potent
(IC50 = 0.3lM). However, the structurally similar di-
benzothiophene-1-yl and dibenzofuran-1-yl derivatives
(22g and 22h) were found to be extremely potent
DNA-PK inhibitors, giving IC50 values of approxi-
mately 20 and 40nM, respectively. Confirmation of this
activity was achieved following the independent synthe-
sis, purification and full characterization of each com-
pound,19 with respective IC50 values of 14 1 and
41 9nM being observed against DNA-PK for 22g
and 22h. ATP-competitive inhibition kinetics were
established for the more potent dibenzothiophen-1-yl
derivative 22g, which was assigned the house number
NU7441. Crucially, NU7441 (22g) retained excellent
selectivity for DNA-PK over other PIKK family mem-
bers compared with NU7026 (4), with only a modest in-
crease in potency being observed against PI 3-K and
mTOR, and no significant activity against ATM or
ATR (Table 2). The high DNA-PK selectivity of 22g
was also confirmed against a commercial panel of 60 di-
verse kinases, with no inhibitory activity being observed
at an inhibitor concentration of 10lM (data not shown).
Figure 1. Putative structure–activity relationships for inhibition of
DNA-PK by chromen-4-one derivatives. The minimum pharmaco-
phore is shown in bold, with dotted lines indicating regions tolerant to
structural modification. For a detailed discussion see Ref. 10.
bonyl group and a 2-morpholinyl substituent, appear to
be a prerequisite for significant inhibitory activity. Elabo-
ration to a chromen-4-one or pyrimidoisoquinolone is
clearly favourable, as witnessed by the activities of 4
and 6, and alkyl group substitution at the morpholine
2-position as for 5 is tolerated. Notably, the Ônorth–westÕ
position of the pharmacophore was found to be very tol-
erant to aromatic group substitution, and we wished to
probe this region of the ATP-binding pocket further,
with the objective of improving potency while retaining
selectivity for DNA-PK over other PIKK family
enzymes.
Three focused libraries (12, 21 and 22) totalling 152 aryl-
substituted chromen-4-ones (48, 54 and 50 compounds,
respectively) were successfully synthesized from 11, 19
and 20. Reactions were not optimized, and products
were generally isolated in modest yields. However, with
the exception of a small number of compounds that
proved inseparable from palladium- and triphenylphos-
phine-derived contaminants, product isolation was
readily achieved by semi-preparative HPLC. Thus,
136 compounds satisfied the minimum purity criteria
required (P85%) for biological evaluation, and were
pre-screened for DNA-PK inhibitory activity at an
In summary, the synthesis and screening of focused chro-
men-4-one libraries has resulted in the identification of
NU7441 (22g) as a highly potent and selective inhibitor
of DNA-PK. In addition to the potential therapeutic
value of NU7741 as an adjunct to the chemo- and
radio-therapy of cancer, a potent and selective small

6086
J. J. J. Leahy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 6083–6087
Table 1. 8-Aryl-substituted chromen-4-one library members (22)
selected from the DNA-PK inhibition pre-screena
Table 2. Inhibition of PIKK family kinases by NU7441 (22g),
NU7026 (4) and LY294002 (1)a
Compound
Kinase inhibition (IC50 lM)
DNA-PKa PI 3-Kb ATMc ATRd mTORe
O
Ar
O
N
NU7441 (22g) 0.014
NU7026 (4) 0.23
LY294002 (1) 1.2
5.0
13
2.3
>100
>100
>100
>100
>100
>100
1.7
6.4
2.5
a IC50 values were determined in accordance with Ref. 18. See Ref. 11
for details of the other assays procedures.
b P110a.
O
Compound number
Ar
OMe
IC50 (lM)b
c Mutated in ataxia telangiectasia protein.
d Ataxia telangiectasia related protein.
e Mammalian target of rapamycin.
22a
0.9
and the results of these studies will be published
subsequently.
HO
22b
0.6
Acknowledgements
The authors would like to thank Dr. Keith Menear and
Dr. Marc Hummersone for their helpful contributions,
Sheila Garman, Adrian Moore and Penny Moore for
technical work, and Cancer Research UK for financial
support.
S
22c
22d
0.7
0.3
S
References and notes
Ac
1. Jackson, S. P. Carcinogenesis 2002, 23, 687–696.
2. Smith, G. C. M.; Jackson, S. P. Gene. Dev. 1999, 13, 916–
934.
3. Smith, G. C. M.; Jackson, S. P. In Handbook of Cell
Signaling; Elsevier Academic, 2003; Vol. 1, pp 557–
561.
4. Kurimasa, A.; Kumano, S.; Boubnov, N. V.; Story, M.
D.; Tung, C. S.; Peterson, S. R.; Chen, D. J. Mol. Cell.
Biol. 1999, 19, 3877–3884.
22e
22f
0.8
0.8
S
S
5. Rosenzweig, K. E.; Youmell, M. B.; Palayoor, S. T.; Price,
B. D. Clin. Cancer Res. 1997, 3, 1149–1156.
6. Boulton, S.; Kyle, S.; Yalcintepe, L.; Durkacz, B. W.
Carcinogenesis 1997, 17, 2285–2290.
7. Izzard, R. A.; Jackson, S. P.; Smith, G. C. M. Cancer Res.
1999, 59, 2581–2586.
8. Hollick, J. J.; Golding, B. T.; Hardcastle, I. H.; Martin,
N.; Richardson, C.; Rigoreau, L. J. M.; Smith, G. C. M.;
Griffin, R. J. Bioorg. Med. Chem. Lett. 2003, 13, 3083–
3086.
9. Griffin, R. J.; Calvert, A. H.; Curtin, N. J.; Durkacz, B.
W.; Golding, B. T., et al. Proc. Amer. Assoc. Cancer Res.
2002, 43, 4210.
22g
22h
0.02
0.04
S
O
10. Griffin, R. J.; Fontana, G.; Golding, B. T.; Guiard, S.;
Hardcastle, I. R.; Leahy, J. J. J.; Martin, N.; Richardson,
C.; Rigoreau, L. J. M.; Stockley, M.; Smith, G. C. M.,
J. Med. Chem., in press.
11. Veuger, S. J.; Curtin, N. J.; Richardson, C. J.; Smith, G.
C. M.; Durkacz, B. W. Cancer Res. 2003, 63, 6008–6015.
12. Willmore, E.; de Caux, S.; Sunter, N. J.; Tilby, M. J.;
Jackson, G. H.; Austin, C. A.; Durkacz, B. W. Blood 2004,
103, 4659–4665.
13. Suzuki, A. J. Organomet. Chem. 1999, 576, 147–168.
14. Littke, A. F.; Dai, C. Y.; Fu, G. C. J. Am. Chem. Soc.
2000, 122, 4020–4028.
15. Vlahos, C. J.; Matter, W. F.; Hui, K. Y.; Brown, R. F.
J. Biol. Chem. 1994, 269, 5241–5248.
a Library compounds were pre-screened against DNA-PK at 0.5lM
concentration.
b IC50 values were determined in accordance with Ref. 18.
molecule DNA-PK inhibitor will represent a useful tool
for elucidating the role of DNA-PK in the signalling and
modulation of DNA DSBs. Preliminary studies have
demonstrated that NU7441 effectively sensitizes a
human tumour cell line (HeLa) to the cytotoxic effects
of both ionizing radiation and the topoisomerase II
inhibitor etoposide at submicromolar concentrations,20

J. J. J. Leahy et al. / Bioorg. Med. Chem. Lett. 14 (2004) 6083–6087
6087
16. Morris, J.; Wishka, D. G.; Fang, Y. Synth. Commun. 1994,
24, 849–858.
17. A general procedure for the MPSof aryl-substituted
chromen-4-ones is as follows: a mixture of the bromo-
chromenone (11) or chromenone triflate (19, 20) (20mg,
ability of compounds to inhibit this phosphorylation event
was monitored over a concentration range with IC50
values generated from these results. See Ref. 10 for full
details of the assay.
19. The independent synthesis of compound 22g was con-
ducted as follows: to a solution of 20 (0.15g, 0.40mmol) in
dioxane (5mL) was added K2CO3 (0.11g, 0.79mmol),
dibenzothiophene-4-boronic acid (0.11g, 0.48mmol) and
tetrakis(triphenylphosphine)palladium(0) (23mg, 0.02
mmol), and the mixture was stirred at 90°C for 16h.
The reaction mixture was cooled to 25°C, diluted with
DCM (15mL), washed with water (10mL) and the organic
fraction was evaporated to dryness in vacuo. Purification
by column chromatography employing DCM–MeOH
(10:1) as eluent, gave 22g as an off-white solid (59mg,
0.14mmol, 35%), mp 220–221°C. 1H NMR (300MHz,
CDCl3): d 3.02–3.05 (4H, m, CH2NCH2); 3.43–3.48 (4H,
m, CH2OCH2); 5.45 (1H, s, 3-H); 7.40–7.46 (4H, m,
Ar-H); 7.50–7.52 (1H, m, Ar-H); 7.70–7.74 (2H, m, Ar-
H); 8.15–8.19 (2H, m, Ar-H); 8.21–8.23 (1H, m, Ar-H);
MS(ES+) m/z 414 [M+H+]+; IR (film) 3055, 2948, 2855,
1622, 1562cmꢁ1; found: C, 72.51; H, 4.69; N, 3.35;
C25H19NO3Srequires: C, 72.62; H, 4.63; N, 3.39. Com-
pound 22h was prepared in a similar manner, and
exhibited spectral (1H NMR, IR, UV) and analytical
(elemental analysis and LC–MS) data fully consistent with
the assigned structure.
0.053mmol),
the
appropriate
arylboronic
acid
(0.06mmol), and powdered potassium carbonate
(14.6mg, 0.106mmol) was purged with nitrogen, and
dioxane (0.5mL, degassed by sonication–purging with
nitrogen) was added. A solution of tetrakis(triphenylphos-
phine)palladium(0) (3.1mg) in degassed dioxane (0.3mL)
was added, and the mixture was stirred under nitrogen for
18h at 90°C. The reaction mixture was cooled, and filtered
through silica (isolute Si 0.5g cartridge) eluting with
DCM–MeOH (7:3 v/v, 8mL). Confirmation of product
formation and approximate purity was determined by
LC–MS/UV analysis, and further purification was
achieved by semi-preparative HPLC, furnishing the
required chromenones in 10–30% yields.
18. The DNA-PK used for in vitro assays was purified from
HeLa cell nuclear extract. The known ability of DNA-PK
to phosphorylate the serine-15 residue of a p53 peptide in
vitro was exploited in a classic ELISA style assay using an
antibody that only recognises the p53 serine-15 site when
phosphorylated (Cell Signalling Technology). The primary
antibody to p53 phosphoserine-15 was detected using an
HRP conjugated goat anti-rabbit antibody (Pierce) with
ECL reagent (NEN) being used for the readout. The
20. Smith, G. C. M., et al., unpublished results.