Mapping the ATP-binding domain of DNA-dependent protein kinase (DNA-PK) with coumarin- and isocoumarin-derived inhibitors

Article abstract of DOI:10.1016/j.bmcl.2010.04.102

Replacement of the core heterocycle of a defined series of chromen-4-one DNA-PK inhibitors by the isomeric chromen-2-one (coumarin) and isochromen-1-one (isocoumarin) scaffolds was investigated. Structure-activity relationships for DNA-PK inhibition were broadly consistent, albeit with a reduction of potency compared with the parent chromenone.

Full text of DOI:10.1016/j.bmcl.2010.04.102

Bioorganic & Medicinal Chemistry Letters 20 (2010) 3649–3653
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Mapping the ATP-binding domain of DNA-dependent protein kinase
(DNA-PK) with coumarin- and isocoumarin-derived inhibitors
Sara L. Payne a, , Sonsoles Rodriguez-Aristegui a, , Julia Bardos b, Céline Cano a, Bernard T. Golding a,
Ian R. Hardcastle a, Marcus Peacock b, Nahida Parveen b, Roger J. Griffin a,
*
a Northern Institute for Cancer Research, School of Chemistry, Bedson Building, Newcastle University, Newcastle Upon Tyne NE1 7RU, UK
b KuDOS Pharmaceuticals Ltd, 410 Cambridge Science Park, Milton Rd., Cambridge CB4 0PE, UK
a r t i c l e i n f o
a b s t r a c t
Article history:
Replacement of the core heterocycle of a defined series of chromen-4-one DNA-PK inhibitors by the iso-
meric chromen-2-one (coumarin) and isochromen-1-one (isocoumarin) scaffolds was investigated. Struc-
ture–activity relationships for DNA-PK inhibition were broadly consistent, albeit with a reduction of
potency compared with the parent chromenone.
Received 16 March 2010
Revised 20 April 2010
Accepted 21 April 2010
Available online 14 May 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Keywords:
DNA repair
DNA-PK
Kinase inhibitors
Chemopotentiation
Radiopotentiation
Anticancer drugs
R
R
O
O
N
O
O
N
R
O
R
O
O
O
N
S
O
O
O
N
N
7
8
OMe
1; R = H
O
N
OMe
O
O
2; R = Ph
Et
5; R = H
6; R =
O
N
N
3; R = thiophen-2-yl
4; R = biphenyl-3-yl
O
N
H
O
O
O
9
10
agents in the treatment of cancer.2,3 Perhaps more interestingly,
DDR defects are a common feature of cancer and the ‘mutator-phe-
notype’ associated with the resultant genomic instability may offer
a survival advantage over normal tissues, albeit that tumour
dependency on a remaining DNA-repair pathway can result. This
offers the exciting opportunity of achieving tumour selectivity
through pharmacological inhibition of an essential DNA repair
pathway, as elegantly demonstrated recently by the synthetic
lethality achieved with poly(ADP-ribose)polymerase-1 (PARP-1)
inhibitors in BRCA1 and BRCA2 deficient tumours (reviewed in
Refs. 1 and 3).
Detection and repair of the numerous, and potentially lethal,
DNA lesions arising in human cells daily is largely mediated by
an efficient system collectively termed the DNA-damage response
(DDR).1 Paradoxically, DNA-repair processes within cancer cells
also constitute a mechanism of resistance to DNA-damaging anti-
cancer therapies, and agents that impede DNA repair are thus of
potential therapeutic interest as chemo- and radio-sensitising
* Corresponding author. Fax: +44 191 2228591.
E-mail addresses: r.j.griffin@ncl.ac.uk, r.j.griffin@newcastle.ac.uk (R.J. Griffin).
These authors contributed equally to this work.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.04.102

3650
S. L. Payne et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3649–3653
The phosphatidylinositol-3-kinase related kinase (PIKK) family
bonyl oxygen functions are thought to make key hydrogen bond
interactions within the ATP-binding domain, with the pendant aryl
substituent occupying a putative hydrophobic pocket. This ‘3-point
binding’ interaction is thought to determine the overall orientation
and positioning of the inhibitor. Accordingly, a superimposition of
the chromenone of 2 with the isomeric coumarin and isocoumarin
heterocyclic systems provided an opportunity to investigate a scaf-
fold-hopping strategy, through the introduction of aryl substitu-
ents at the coumarin 6- or 7-positions and the isocoumarin 5-
position (Figure 1). This was supported by a previous observation
that 6-methoxycoumarin (10) is approximately equipotent with
the isomeric 8-methoxychromenone (9) (DNA-PK, IC50 values of
member DNA-dependent protein kinase (DNA-PK) plays a key role
in the DDR, via the non-homologous end-joining pathway of DNA
double-strand break (DSB) repair.4 Importantly, inhibition of this
kinase has been demonstrated to potentiate the cytotoxicity of
DNA DSB-inducing anticancer therapies,5 and there is evidence
that DNA-PK is over-expressed in a number of tumours. The devel-
opment of clinically useful ATP-competitive DNA-PK inhibitors is a
major goal of our research. In the absence of suitable structural
information for DNA-PK, we have conducted structure-activity
relationship (SAR) studies around the 2-morpholino-4H-chro-
men-4-one pharmacophore (1) from which the non-selective PIKK
inhibitor LY294002 (2) was derived.6 These studies resulted in the
identification of a number of interesting DNA-PK inhibitors, includ-
ing the thiophen-2-yl (3)7 and 8-biphenyl derivatives (4),8 as well
as NU7441 (5; IC50 = 30 nM),7,9 elaboration of which has led to the
highly potent water-soluble DNA-PK inhibitor KU-0060648 (6;
IC50 = 5 nM).10 Importantly, preclinical proof-of-principle studies
with 5 and 6 have demonstrated activity in vitro as chemo- and
radio-potentiators in a range of human tumour cell lines, and
initial in vivo investigations with these agents have been
promising.9,10
1.8 lM and 1.2 l
M, respectively).7 Given the high potency of 5
and 6, it was also of interest to examine the impact on activity of
the subtle structural changes imposed by moving the ring oxygen
atoms.
In this Letter, we describe the synthesis and DNA-PK-inhibitory
activity of a defined series of coumarins and isocoumarins, and
demonstrate SARs that generally parallel those observed for the
corresponding 8-substituted 2-morpholin-4-yl-chromen-4-ones.
The surprising loss of activity for the coumarins and isocoumarins
is discussed in terms of conformational effects.
With a view to identifying new DNA-PK inhibitors, we have
investigated alternative heterocyclic scaffolds, and have previously
demonstrated that sub-micromolar DNA-PK inhibitory activity also
resides in suitably substituted pyran-2-ones (7) and pyran-4-ones
(8).11,12 These studies imply that the ring oxygen of chromenone-
and pyranone-based DNA-PK inhibitors does not contribute di-
rectly to inhibitor binding. By contrast, the morpholin-4-yl and car-
The target 6- and 7-substituted coumarins 16–27 were readily
prepared as summarised in Scheme 1. Thus, thermal condensation
of commercially available 5-substituted-2-hydroxyacetophenones
(11a–11c) with diethyl carbonate afforded the corresponding 6-
substituted-4-hydroxycoumarins (13a–13c) in high yield. 4-Bro-
mo-2-hydroxyacetophenone (11d) was prepared in good overall
yield from 3-bromophenol by sequential O-acetylation–Fries
Coumarin
Chromenone
Isocoumarin
hydrophobic
pocket
6
Ar
hydrophobic
pocket
hydrophobic
pocket
7
Ar
O
Ar
O
O
5
O
O
N
N
N
O
O
O
O
Figure 1. Comparison of the putative relative binding orientations of the coumarin, chromenone and isocoumarin core templates within the ATP-binding domain of DNA-PK
(? = hydrogen bond interaction).
OH
Br
OH
O
O
R4
R7
a, b
c
5
6
O
OH
11a 11b
12
; R = H
,
; R = OMe
11c, 11d; R = Br
13a, 13b; R = OMe
13c 13d
,
; R = Br
O
O
O
O
R7
R7
d
e
6
6
N
O
N
O
10; R = 6-OMe
16-27
14
; R = H
15a, 15b; R = Br
15c; R = 7-OMe
Scheme 1. Reagents and conditions: (a) Ac2O, H2SO4, 25 °C, 1 h; (b) AlCl3, 170 °C, 3 h, 78–80%; (c) Na, CO(OEt)2; Dowtherm A, 160 °C, 2 h, 50–60%; (d)
triisopropylbenzenesulfonyl chloride, Et3N, THF, 120 °C, 10 min, MW, followed by morpholine, 25 °C, 18 h, 50–70%; (e) ArB(OH)2, Pd(PPh3)4, K2CO3, dioxane, reflux, 18 h,
40–70%.

S. L. Payne et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3649–3653
3651
rearrangement,13 with subsequent conversion into the require 7-
bromocoumarin (13d) proceeding smoothly. Introduction of the
4-morpholinyl function was achieved by a previously optimised
one-pot procedure entailing activation of the phenol as the tri-
isopropylbenzenesulfonyl ester, followed by displacement with
morpholine, to give coumarins 10, and 15a–15c in excellent
yields.7 The parent 4-morpholinylcoumarin (14) was similarly pre-
pared from 4-hydroxycoumarin (12). Final Suzuki–Miyaura cou-
pling of bromocoumarins 15a and 15b with the appropriate
arylboronic acids under standard conditions,14 gave the target 6-
and 7-arylcoumarins (16–27).
The synthetic route employed for the preparation of the
isocoumarins is shown in Scheme 2. Commercially available 3-
methoxy- and 3-bromo-2-methylbenzoic acids afforded the corre-
sponding ethyl homophthalates (28a and 28b), respectively, on
treatment with LDA-diethyl carbonate, and subsequent acid-cata-
lysed cyclisation to the desired 5-bromohomophthalic anhydrides
(29b and 29c) proceeded in high yield.15 The use of a hindered base
(LDA) at À50 °C avoided unwanted halogen-metal exchange
reactions involving the bromo group.16 Reaction with morpholine,
and conversion of the resulting 2-(2-morpholin-4-yl-2-
oxoethyl)benzoic acids (30b and 30c) into the key 5-substituted-
3-morpholin-4-yl-isochromen-1-one (31b and 31c), proceeded
smoothly in good overall yield. The parent isocoumarin (31a)
was readily prepared from commercially available homophthalic
anhydride (29a) under essentially identical conditions. Final palla-
dium-catalysed cross-coupling of 31c with arylboronic acids was
achieved under standard conditions (Pd[PPh3]4, Cs2CO3, dioxane),
to afford the target 5-aryl-3-morpholin-4-yl-isochromen-1-ones
(32–35) in good overall yields (Scheme 2). Although these reac-
tions proved sluggish under conventional reflux conditions, the
use of microwave (MW) heating dramatically shortened reaction
times, with complete consumption of 31c occurring within 40–
90 min.
inhibitory activity (compare 9 with 10, 15c, and 31b and 2 with
16, 22 and 32), consistent with similar binding orientations. Inter-
estingly, this was not the case for the parent heterocycles, and
while the coumarin (14) exhibited low micromolar potency, the
corresponding chromenone (1) and isochromenone (31a) deriva-
tives were essentially devoid of activity.
Introduction of a 2-thienyl group at the coumarin 7-position
(23) conferred sub-micromolar potency comparable with that of
the chromenone (3), whereas the corresponding 6-substituted
coumarin (17) was slightly less potent. However, the most dra-
matic differences in potency were observed for larger aryl sub-
stituents. Thus, the 6-(3-phenylphenyl)coumarin (18) proved
some 10-fold less potent than the analogous chromen-4-one
(4), with the 7-substituted coumarin (24) proving inactive. These
difference in potency were even more pronounced for the benzo-
thiophen-2-yl, dibenzothiophen-1-yl and dibenzofuran-1-yl ana-
logues, and while the 7-substituted coumarins 26 and 27 were
70-fold less potent than 5, the other derivatives (19–21, 25) were
inactive. The proposed binding mode for the 6-substituted cou-
marin scaffold (B) positions the fused phenyl ring within the
putative hydrophobic pocket occupied by substituents at the
chromenone 8-position (A) and the isocoumarin 5-position (D).
As such, the loss of activity observed for larger groups at the cou-
marin 6-position (e.g., 20 and 21) may reflect steric constraints
within the hydrophobic pocket. Interestingly, the isocoumarin
series proved more tolerant to substitution at the 5-position,
with only a fourfold loss of activity being observed for the 5-
(3-phenylphenyl)isocoumarin (33) compared with 4. Similarly,
isocoumarins bearing a dibenzothiophen-1-yl or dibenzofuran-
1-yl group at the 5-position (34 and 35) retained very reasonable
DNA-PK inhibitory activity (IC50 ꢀ300 nM) compared with the
parent chromenone (5).
The accentuation of the difference in potency between the
chromenones (e.g., 5) and isocoumarins (e.g., 34) with a larger sub-
stituent at the 8-position (in 5) or 5-position (in 34) could also be
due to the effect of the 4-H atom in the isochromenones (vs O in
The overall objective of this study was to investigate the effect
upon DNA-PK inhibitory activity of replacing the core chromenone
heterocycle by coumarin and isocoumarin templates, whilst retain-
ing the relative positioning of key groups, notably a ring carbonyl
and a morpholinyl substituent. Our previous SAR studies have also
highlighted the importance of an appropriate 8-aryl substituent for
chromenone-based inhibitors, and the nature and position of aryl
groups on the coumarins and isocoumarins was directed by these
observations. Methoxy or phenyl group substitution on the couma-
rin and isocoumarin rings, at positions analogous to the chrome-
none 8-position, afforded compounds with comparable DNA-PK
the chromenones) on the preferred conformation about the
r-
bond connecting the substituent to the other heteroaromatic sys-
tem. This suggestion is consistent with the relative poor activity
of structural classes B and C (Table 1), which are conformationally
constrained at the C–C bond corresponding to that connecting the
two (hetero)aromatic systems of A and D. Current studies are
exploring the influence of such conformational factors on the ste-
reochemistry and activities of chromen-4-ones and isochromen-
1-ones.
R
R
R
Me
O
CO2Et
CO2H
a
b
c
O
CO2H
O
R = OMe, Br
28a; R = OMe
28b; R = Br
29a; R = H
29b; R = OMe
29c
; R = Br
R
O
R
O
Ar
O
N
N
N
e
d
O
O
O
R = Br
CO2H
O
O
30a
31a
; R = H
32-35
; R = H
30b; R = OMe
30c
31b; R = OMe
31c
; R = Br
; R = Br
Scheme 2. Reagents and conditions: (a) (i) LDA, THF, À50 °C, 63%; (ii) ClCO2Et; (b) CSA, toluene, reflux, 85%; (c) morpholine, toluene, 80 °C, 70–90%; (d) ClCO2Et, Et3N,
toluene, À78 °C, 65–80%; (e) K2CO3, Pd(PPh3)4, dioxane, ArB(OH)2, MW 175 °C, 40–90 min, 35–80%.

3652
S. L. Payne et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3649–3653
Table 1
DNA-PK inhibitory activity of chromenones, coumarins, and isocoumarins
R
6
7
R
R
O
R
O
O
O
8
5
O
O
N
N
N
N
O
O
O
O
O
O
A
B
C
D
a
Compd
Structure
R
IC50
>10
(lM)
1
2
A
A
H
Ph
1.6b
S
3
A
0.7c
Ph
4
A
0.18b
0.03
5
A
S
9
A
B/C
B
MeO
H
MeO
Ph
1.8
4.8
1.7
1.8
14
10
16
B
S
17
B
1.2
Ph
18
19
B
B
1.6
>10
S
20
B
>10
S
21
B
>10
O
15c
22
C
C
OMe
Ph
5.4
1.7
S
23
C
0.6
Ph
24
25
C
C
>10
>10
S
26
27
C
C
2.1
2.1
S
O

S. L. Payne et al. / Bioorg. Med. Chem. Lett. 20 (2010) 3649–3653
3653
M)
Table 1 (continued)
a
Compd
Structure
R
IC50 (l
31a
31b
32
D
D
D
H
MeO
Ph
>10
3.2
2.6
Ph
33
D
0.68
34
D
0.34
S
35
D
0.33
O
a
IC50 values were determined in accordance with Ref. 7.
Ref. 7.
Ref. 8.
b
c
8. Desage-El Murr, M.; Cano, C.; Golding, B. T.; Hardcastle, I. R.; Hummersome, M.;
Frigerio, M.; Curtin, N. J.; Menear, K.; Richardson, C.; Smith, G. C. M.; Griffin, R. J.
Bioorg. Med. Chem. Lett. 2008, 18, 4885.
9. Zhao, Y.; Thomas, H. D.; Batey, M. A.; Cowell, I. G.; Richardson, C. J.; Griffin, R. J.;
Calvert, A. H.; Newell, D. R.; Smith, G. C. M.; Curtin, N. J. Cancer Res. 2006, 66,
5354.
10. Saravanan, K.; Albertella, M.; Cano, C.; Curtin, N. J.; Frigerio, M.; Golding, B. T.;
Haggerty, K.; Hardcastle, I. R.; Hummersone, M.; Menear, K.; Newell, D. R.;
Rennison, T.; Richardson, C.; Rigoreau, L.; Rodriguez-Aristegui, S.; Smith, G. C.
M.; Griffin, R. J. Proc. Am. Assoc. Cancer Res. 2008, 4156.
11. Hollick, J. J.; Golding, B. T.; Hardcastle, I. R.; Martin, N.; Richardson, C.;
Rigoreau, L. J. M.; Smith, G. C. M.; Griffin, R. J. Bioorg. Med. Chem. Lett. 2003, 13,
3083.
12. Hollick, J. J.; Rigoreau, L. J. M.; Cano, C.; Cockcroft, X.; Curtin, N. J.; Frigerio, M.;
Golding, B. T.; Guiard, S.; Hardcastle, I. R.; Hickson, I.; Hummersone, M. G.;
Menear, K. A.; Martin, N. M. B.; Matthews, I.; Newell, D. R.; Ord, R.; Richardson,
C. J.; Smith, G. C. M.; Griffin, R. J. J. Med. Chem. 2007, 50, 1958.
13. Lanier M.; Moorjani M.; Tellew J. E.; William J. P.; Patent No. PCT/US2005/
037295.
14. Miyaura, N.; Suzuki, A. Chem. Rev. 1995, 95, 2457.
15. Wawzonek, S.; Hansen, G. R. J. Org. Chem. 1975, 40, 2974.
16. Gohier, F.; Mortier, J. J. Org. Chem. 2003, 68, 2030.
Acknowledgements
The authors thank Cancer Research UK for financial support.
The use of the EPSRC Mass Spectrometry Service at the University
of Wales (Swansea) is also gratefully acknowledged.
References and notes
1. Jackson, S. P.; Bartek, J. Nature 2009, 461, 1071.
2. Madhusudan, S.; Middleton, M. R. Cancer Treat. Rev. 2005, 31, 603.
3. O’Connor, M. J.; Martin, N. M. B.; Smith, G. C. M. Oncogene 2007, 26, 7816.
4. Smith, G. C. M.; Jackson, S. P.. In Handbook of Cell Signalling; Elsevier Academic
Press, 2003; Vol. 1.
5. Boulton, S.; Kyle, S.; Yalcintepe, L.; Durkacz, B. W. Carcinogenesis 1997,
17, 2285.
6. Izzard, R. A.; Jackson, S. P.; Smith, G. C. M. Cancer Res. 1999, 59, 2581.
7. Hardcastle, I. R.; Cockcroft, X.; Curtin, N. J.; Desage El-Murr, M.; Leahy, J. J. J.;
Stockley, M.; Golding, B. T.; Rigoreau, L. J. M.; Richardson, C.; Smith, G. C. M.;
Griffin, R. J. J. Med. Chem. 2005, 48, 7829.