Design and evaluation of 3,6-di(hetero)aryl imidazo[1,2-a]pyrazines as inhibitors of checkpoint and other kinases

Article abstract of DOI:10.1016/j.bmcl.2010.05.096

A range of 3,6-di(hetero)arylimidazo[1,2-a]pyrazine ATP-competitive inhibitors of CHK1 were developed by scaffold hopping from a weakly active screening hit. Efficient synthetic routes for parallel synthesis were developed to prepare analogues with improved potency and ligand efficiency against CHK1. Kinase profiling showed that the imidazo[1,2-a]pyrazines could inhibit other kinases, including CHK2 and ABL, with equivalent or better potency depending on the pendant substitution. These 3,6-di(hetero)aryl imidazo[1,2-a]pyrazines appear to represent a general kinase inhibitor scaffold.

Full text of DOI:10.1016/j.bmcl.2010.05.096

Bioorganic & Medicinal Chemistry Letters 20 (2010) 4045–4049
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design and evaluation of 3,6-di(hetero)aryl imidazo[1,2-a]pyrazines as
inhibitors of checkpoint and other kinases
a
c
a
c
c
Thomas P. Matthews a, , Tatiana McHardy , Suki Klair , Kathy Boxall , Martin Fisher , Michael Cherry ,
Charlotte E. Allen a, Glynn J. Addison c, John Ellard c, G. Wynne Aherne a, Isaac M. Westwood b,
Rob van Montfort b, Michelle D. Garrett a, John C. Reader c, Ian Collins a
*
a Cancer Research UK Centre for Cancer Therapeutics, The Institute of Cancer Research, 15 Cotswold Road, Sutton, Surrey SM2 5NG, United Kingdom
b Sections of Cancer Therapeutics and Structural Biology, The Institute of Cancer Research, 15 Cotswold Road, Sutton, Surrey SM2 5NG, United Kingdom
c Sareum Ltd, Unit 2A, Langford Arch, London Road, Pampisford, Cambridge CB22 3FX, United Kingdom
a r t i c l e i n f o
a b s t r a c t
Article history:
A range of 3,6-di(hetero)arylimidazo[1,2-a]pyrazine ATP-competitive inhibitors of CHK1 were developed
by scaffold hopping from a weakly active screening hit. Efficient synthetic routes for parallel synthesis
were developed to prepare analogues with improved potency and ligand efficiency against CHK1. Kinase
profiling showed that the imidazo[1,2-a]pyrazines could inhibit other kinases, including CHK2 and ABL,
with equivalent or better potency depending on the pendant substitution. These 3,6-di(hetero)aryl imi-
dazo[1,2-a]pyrazines appear to represent a general kinase inhibitor scaffold.
Received 6 April 2010
Revised 21 May 2010
Accepted 22 May 2010
Available online 27 May 2010
Keywords:
CHK1
Ó 2010 Elsevier Ltd. All rights reserved.
Kinase inhibitors
Imidazopyrazines
Protein kinase enzymes play vital roles in the control of cell cy-
cle progression, apoptosis, growth and metabolism and are well
established as potential targets for intervention in a number of dis-
eases including cancer.1 There is a need to identify new chemical
classes of kinase inhibitors, including novel heterocycles that fulfil
the pharmacophore requirements for occupying the ATP-binding
site of the enzymes.1,2 Checkpoint kinase 1 (CHK1) is an oncology
target of current interest as CHK1 inhibitors have the potential to
enhance the efficacy of cytotoxic treatments.3–5 CHK1 is a serine/
threonine kinase that is phosphorylated and activated in response
to DNA damage, initiating a signalling cascade culminating in cell
cycle arrest in the S or G2/M phases. Inhibition of CHK1 has been
shown to abrogate these checkpoints leading to enhanced tumour
cell death following DNA damage by a range of chemotherapeutics.
Tumour cells deficient in the tumour suppressor p53 and lacking
intact G1 checkpoints are particularly dependent on S and G2/M
checkpoints, and are therefore expected to be sensitised to
genotoxic treatment in the presence of a CHK1 inhibitor. Several
chemotypes of CHK1 inhibitor have been described, including
compounds in early clinical trials.3–6
azole (Fig. 1, 1) (26% inhibition at 250 l
M).7 We have investigated
the progression of a number of these hits to identify chemical series
with good prospects for optimisation into potent and selective leads.
We describe here new imidazo[1,2-a]pyrazines (Fig. 1, 2) and the
development of an efficient synthetic procedure to make analogues
of this scaffold. Improvements in CHK1 potency were achieved,
while kinome screening indicated the new compounds to have
wider applicability as a kinase inhibitor chemotype.
A literature search indicated that benzisoxazole kinase inhibitors
were already documented.8,9 Indazoles with strikingly similar
substitution patterns have been reported as JNK inhibitors.10,11 The
design of an alternative 6,5-bicyclic template that would provide
increased affinity for CHK1 was therefore pursued and a novel
3,6-disubstituted imidazo[1,2-a]pyrazine core 2 was envisaged.
Although imidazo[1,2-a]pyrazines bearing heteroatom 3- and
6-substituents have been reported as kinase inhibitors12 there are
We have previously reported a screening strategy that identified
several low molecular weight compounds, or templates, as ATP-
competitive CHK1 inhibitors, including the weakly active benzisox-
* Corresponding author. Tel.: +44 2086438901.
E-mail address: tom.matthews@icr.ac.uk (T.P. Matthews).
Figure 1. Benzisoxazole 1 and the imidazo[1,2-a]pyrazine core 2.
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.05.096

4046
T. P. Matthews et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4045–4049
only limited examples of similar 3,6-di(hetero)aryl imidazo[1,2-
a]pyrazines in the patent literature.13,14
A small array of 3-phenyl imidazo[1,2-a]pyrazines were synthes-
ised which displayed encouraging CHK1 activity (Table 1). a-Brom-
ophenylacetaldehyde dimethyl acetal 4 was deprotected with HBr
followed by condensation of the resulting aldehyde with 2-amino-
5-bromopyrazine 3 to afford 6-bromo-3-phenyl imidazopyrazine 5
(Scheme 1).15 Suzuki–Miyaura coupling (or in the case of 18 a Stille
coupling) installed the 6-substituent. Of the nine examples synthes-
ised, the 4-pyrazolyl and 4-pyridyl analogues 13 and 20 were the
most potent against CHK1. The pyrrole 16, indole 17 and 3-pyrazole
analogue 12, which most closely mirrors the substitution in the ori-
ginal hit 1, were less active. N-Substitution of the 4-pyrazole (14, 15)
led to a reduction in potency as did replacing the 4-pyridyl with the
2- or 3- regioisomers (18, 19).
The binding mode of 13 in CHK1 was determined by X-ray crys-
tallography, through soaking of the ligand into crystals of apo-
CHK17 (PDB code 2XF0,Fig. 2, Panel A). This showed a hydrogen bond
from N1 to the amide NH of Cys87 with the suggestion of a second
favourable contact with the hinge region in the form of a C(Ar)H. . .O
interaction between C2 and the carbonyl of Cys87. The pyrazole ring
pointed towards the interior pocket of the enzyme which has been
targeted to achieve selectivity for CHK1.16 The phenyl 3-substituent
was directed towards the ribose pocket and solvent.
Scheme 1. Reagents and conditions: (a) (i) 4, 48% HBr/H2O, 105 °C, 10 min,
(ii) NaHCO3, IPA, then 3, 150 °C, 20 min, wave; 44% yield; (b) RB(OH)2, Pd(PPh3)4,
Na2CO3, MeCN/H2O, 150 °C, 20 min wave; 35–94%; (c) (i) 7, 48% HBr/H2O, 100 °C,
lwave;
l
The ligand efficiencies (L.E.)17 for these first analogues, calcu-
lated using the compact formula L.E. = [À1.4 Â log10(IC50 (M))]/
(number of heavy atoms)17, were modest (Table 1). As the 3-sub-
stituent of the imidazo[1,2-a]pyrazine followed a vector towards
the polar environment of the ribose pocket and solvent exposed re-
gion, further analogues were synthesised based on the most active
6-(4-pyrazolyl) analogue 13 but with more hydrophilic 3-substitu-
ents. As shown in Scheme 1, the synthesis allowed the point of
diversity on the scaffold to be elaborated last. After treatment with
acid, bromoacetaldehyde dimethyl acetal 7 was condensed with 2-
amino-5-bromopyrazine 3 to form the 6-bromo-imidazole bicycle
8. Palladium-mediated coupling with pyrazole-4-boronic acid
pinacol ester installed the preferred substituent in the 6-position
(9), before bromination gave the advanced intermediate 10 for
the final Suzuki–Miyaura reactions.
l
2 h; (ii) NaHCO3, IPA, then 3, 100 °C, 3 h, 76%; (d) 4-(4,4,5,5-tetramethyl-1,3,2-
dioxaborolan-2-yl)-1H-pyrazole, Pd(PPh3)4, Na2CO3, MeCN/H2O, 150 °C, 20 min
l
wave, 73%; (e) NBS, DMF, rt, 2.5 h, 37%; (f) R1B(OH)2, Pd(PPh3)4, 0.5 M Na2CO3,
DMF, 150 °C, 30 min. 7–20%.
cies of 22 and 24 were significantly improved. With the exception
of the highly lipophilic benzyl ether 25 and the 3-pyridyl analogue
21, the other compounds displayed weak CHK1 activity. X-ray
crystallography showed 24 bound in an identical manner to 13
(PDB code 2XEZ, Fig. 2, Panel B), with the H-bond from N1 and
A number of the compounds showed increased potency against
CHK1 (Table 2), notably the bis-pyrazolyl analogues 22 and 24.
While still inhibiting in the micromolar range, the ligand efficien-
Table 1
CHK1 activity of 3-phenylimidazo[1,2-a]pyrazines
a
No.
R
CHK1 IC50
>100
(lM)
L.E.b
12
13
14
15
16
17
18
19
20
1H-Pyrazol-3-yl
1H-Pyrazol-4-yl
1-Methylpyrazol-4-yl
1-Benzylpyrazol-4-yl
1H-Pyrrol-2-yl
1H-Indol-3-yl
2-Pyridyl
n.d.d
0.32
0.28
0.19
n.d.
n.d.
n.d.
0.26
0.29
30c (28, 31)
72
208
>100
>100
>100
139
3-Pyridyl
4-Pyridyl
52
a
Single determination in DELFIA assay.7 Standard inhibitor staurosporine gave
Figure 2. (A) Crystal structure of hit 13 bound to the CHK1active site. 2Fo À Fc map
IC50 = 2.1 ( 1.8) nM.
contoured at 1.0
r represented by blue mesh (PDB code 2XF0); (B) Crystal structure
b
Ligand efficiency (kcal molÀ1 non-H atomÀ1
.
of hit 24 bound to the CHK1active site. 2Fo À Fc map contoured at 1.0
r represented
c
Mean of two independent determinations, individual values in parentheses.
Not determined.
by blue mesh (PDB code 2XEZ); (C) Overlay of 13, 24 and 31g relative to the hinge
d
region of CHK1, cyan = 13, red = 24, blue = 31g (PDB code 2XEY), green = CHK1.

T. P. Matthews et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4045–4049
4047
Table 2
Table 3
CHK1 activity of 6-(1H-pyrazol-4-yl)imidazo[1,2-a]pyrazines
CHK1 activity of selected 3,6-disubstituted imidazo[1,2-a]pyrazines prepared by
parallel synthesis
L.E.b
a
No.
R
CHK1 IC50
>100c
(lM)
21
22
23
24
3-Pyridyl
n.d.d
0.39
0.33
0.43
a
1H-Pyrazol-4-yl
1-Methylpyrazol-4-yl
1H-Pyrazol-3-yl
5.0 (2.2, 7.7)
22 (21, 23)
1.5 (1.4, 1.6)
Compd
R
Chk1 IC50
>100b
(lM)
31a
25
>100c
n.d.
31b
31c
>100b
>100b
26
27
46 (34, 58)
23c
0.28
0.24
31d
26 (12, 40)
28
22 (15, 31)
0.24
a
Mean of two independent determinations unless otherwise stated, individual
31e
31f
96b
values in parentheses. Standard inhibitor staurosporine gave IC50 = 2.1 ( 1.8) nM.
b
Ligand efficiency (kcal molÀ1 non-H atomÀ1).
Single determination.
Not determined.
c
43 (34, 52)
d
31g
6.5 (5.9,7.1)
32a
32b
25 (16, 34)
17 (14, 21)
Scheme 2. Reagents and conditions: (a) NCS, DCM, reflux, 5.5 h, 96% yield; (b)
32c
15 (12, 18)
R1B(OH)2, Pd(PPh3)4, 0.5 M Na2CO3, MeCN, 120 °C, 1 h,
l
wave, 42–75%; (c)
R2B(OH)2, Pd(PPh3)4, 0.5 M Na2CO3, DMF, 150 °C, 20 min,
lwave, 26–31%.
32d
32e
15 (14, 17)
28 (19, 37)
C(Ar)H. . .O interaction from C2 to Cys87. A better interaction of the
nitrogen on the pendant 6-(1H-pyrazol-4-yl) group with Lys38 was
seen for 24 compared to 13 (2.7 Å vs 3.5 Å). The 3-(1H-pyrazol-3-
yl) substituent of 24 did not make any additional H-bonds to the
protein, with no atoms on either side within reasonable H-bonding
distances or angles of the ligand. The increase in potency of 24 over
the phenyl-substituted 13 may therefore be due to the improved
interaction with Lys38. The 3-pyrazole ring was modelled with
the nitrogen atoms pointing out towards solvent though the data
are not of sufficiently high resolution to discriminate this from
the alternative orientation with the pyrazole rotated by 180°.
To enable elaboration of the bicyclic scaffold at both the 3- and
6-positions, a sequential coupling strategy was sought. Thus chlo-
rination of 8 gave the 3-chloro-6-bromoimidazo[1,2-a]pyrazine 29
(Scheme 2) and Suzuki–Miyaura coupling of phenylboronic acid
was attempted under various conditions. Most of these led to mix-
tures of the desired 3-chloro-6-phenyl product and the related des-
chloro byproduct. However, specific conditions were identified
32f
32g
32h
32i
3.5 (3.2, 3.7)
2.0 (1.6, 2.3)
29 (19, 39)
39 (21, 57)
a
Mean of two independent determinations unless otherwise stated, individual
values in parentheses. Standard inhibitor staurosporine gave IC50 = 2.1 ( 1.8) nM.
b
Single determination.
(Pd(PPh3)4, Na2CO3, MeCN/water, 120 °C, 60 min,
both excellent selectivity for reaction at the 6-position and left
l
wave) that gave
the 3-chloro group in place. Using the same base and catalyst
combination but changing the solvent to DMF and raising the

4048
T. P. Matthews et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4045–4049
Table 4
Kinase profiling for selected compounds by mobility shift assay and IC50 determinations
a % Inhibition of kinases in a mobility shift assay21at 10
lM concentration of test compound and [ATP] = Km,ATP for each individual kinase. Single point determination. Orange
>80% inhibition, yellow = 50–80% inhibition.
b
IC50 determined in DELFIA assay.7 Mean of two independent determinations unless otherwise stated.
IC50 determined in Z’-Lyte assay.23
c
d
e
Single determination.
Mean ( SEM)% inhibition for internal standard H89, a pan kinase inhibitor,24 n = 15 determinations.
reaction temperature to 150 °C allowed the 3-chloro substituent to
be subsequently derivatised. By carrying out the two reactions in
tandem and using automated HPLC purification for the final com-
pounds, a larger number of compounds were prepared (Table 3).
Compound 31g was shown by crystallography to bind identically
to the previous examples (Fig. 2C).
in changes in orientation of the core between kinase targets. It
may however be possible to modulate the selectivity profile of
the imidazo[1,2-a]pyrazines by specific choice of the 3- and
6-substituents.
In summary, scaffold hopping from a very weak benzisoxazole
hit generated novel 3,6-di(hetero)aryl imidazo[1,2-a]pyrazine
ATP-competitive inhibitors of CHK1. By developing synthetic
routes that allowed late stage derivatisation of the scaffold, we
quickly synthesised analogues with improved potency and ligand
efficiency, leading to micromolar inhibitors of CHK1. Kinase profil-
ing showed that several of the imidazo[1,2-a]pyrazines inhibited
other kinases with equivalent or better potency. The selectivity
profile of the compounds varied with the nature of the pendant
substituents. The 3,6-di(hetero)aryl imidazo[1,2-a]pyrazine scaf-
fold may therefore be generally useful for the development of
new kinase inhibitors.
Despite a variety of functionality introduced onto the 3- and
6-substituents, the activity for these compounds at CHK1 was not
improved over that of 24. In addition, counterscreening against the
structurally distinct enzyme CHK2, also a potential oncology tar-
get,18 indicated little selectivity. For example, compounds 24, 31g
and 32g gave CHK2 IC50 values7 of 0.48
lM, 1.3 lM and 0.12 lM,
respectively (Table 4). Further selectivity profiling revealed other
significant kinase inhibitory activities associated with the scaffold.
Aselection of11analoguesweretestedat10 lM inhibitor concen-
tration against a panel of 24 kinases using a mobility shift assay20 (Ta-
ble 4, [ATP] = Km,ATP for each kinase,21,22 see Supplementary data). At
this concentration none of the selected imidazo[1,2-a]pyrazines
showed greater than 50% inhibition against MAPKAPK2, PKC, PRAK,
MET, ERK2, PKA, AKT2, INSR, p38a, AKT1 and MSK1. However activity
was observed against the remaining 13 kinases by one or more of the
compounds. A group of structurally similar analogues (32g, 32h, 32i)
Acknowledgements
We thank Dr. A. Mirza, M. Richards and Dr. M. Liu for assistance
in the spectroscopic characterisation of test compounds and Dr M.
Lamers for the preparation of CHK1 protein. This work was sup-
ported by Cancer Research UK [CUK] Grant Numbers C309/
A2187, C309/A8274 and C309/A8365 and by The Institute of Can-
cer Research. We acknowledge NHS funding to the NIHR Biomedi-
cal Research Centre.
stronglyinhibited(>94%@10 lM)thetyrosinekinasesABL,FYN, LYN,
LCK and SRC. Compounds 22, 24, 31g and again 32i were inhibitors of
AurA. To validate the screening results, IC50 determinations were
made for the most potent inhibitors against CHK2, ABL1 and GSK3b,23
and good agreement was seen between these data and the single-
point inhibitions. Notably, sub-micromolar potency was observed
for inhibition of ABL1 and CHK2 with these compounds.
Supplementary data
The elaboration of other template screen hits, which were eval-
uated in parallel with 1, led to potent and selective CHK1 inhibitors
which were able to bind in the interior pocket of the enzyme.19 In
view of this, and given the low CHK1 potency and selectivity seen
with the compounds described here, the imidazo[1,2-a]pyrazines
were not progressed further against CHK1. Our experience shows
how scaffold morphing from fragment-like hits can generate new
inhibitor scaffolds, but also highlights the benefit of monitoring ki-
nase selectivity during the development of the new scaffolds. We
speculate that the single explicit hydrogen bond from the core imi-
dazo[1,2-a]pyrazine to the hinge region may result in an increased
importance of the interactions formed by the two pendant aryl
groups to determine affinity and selectivity, and may also result
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.05.096.
References and notes
1. Cohen, P. Nat. Rev. Drug Disc. 2002, 1, 309.
2. Pitt, W. R.; Parry, D. M.; Perry, B. G.; Groom, C. R. J. Med. Chem. 2009, 52, 2952.
3. Blasina, A.; Hallin, J.; Chen, E.; Arango, M. E.; Kraynov, E.; Register, J.; Grant, S.;
Ninkovic, S.; Chen, P.; Nichols, T.; O’Connor, P.; Anderes, K. Mol. Cancer Ther.
2008, 7, 2394.
4. Ashwell, S.; Zabludoff, S. Clin. Cancer Res. 2008, 14, 4032.
5. Bucher, N.; Britten, C. D. Br. J. Cancer 2008, 98, 523.
6. Janetka, J. W.; Ashwell, S.; Zabludoff, S.; Lyne, P. Curr. Opin. Drug Discov. Devel.
2007, 10, 473.

T. P. Matthews et al. / Bioorg. Med. Chem. Lett. 20 (2010) 4045–4049
4049
7. Matthews, T. P.; Klair, S.; Burns, S.; Boxall, K.; Cherry, M.; Fisher, M.; Westwood, I.
M.; Walton, M. I.; McHardy, T.; Cheung, K.-M. J.; Van Montfort, R.; Williams, D.;
Aherne, G. W.; Garrett, M. D.; Reader, J.; Collins, I. J. Med. Chem. 2009, 52, 4810.
8. Ledeboer, M.; Davies, R. J.; Messersmith, D.; Moon, Y.-C.; Mullican, M. D. Patent
Appl. WO2004058749, 2004; Chem. Abstr. 2004, 141, 106455.
14. Sato, Y.; Kurihara, H.; Kamijo, K.; Onozaki, Y.; Tsujino, T.; Sugimoto, T.;
Watanabe, A.; Mitsuya, M.; Komatani, H. Patent Appl. WO2006049339, 2006;
Chem. Abstr. 2006, 144, 468203.
15. Lumma, W. C.; Randall, W. C.; Cresson, E. L.; Huff, J. R.; Hartman, R. D.; Lyon, T.
F. J. Med. Chem. 1983, 26, 357.
9. Moon, Y. C.; Green, J.; Davies, R.; Choquette, D.; Pierce, A.; Ledeboer, M. Patent
Appl. WO2002102800, 2002; Chem. Abstr. 2002, 138, 39275.
16. Foloppe, N.; Fisher, L. M.; Francis, G.; Howes, R.; Kierstan, P.; Potter, A. Bioorg.
Med. Chem. 2006, 14, 1792.
10. Bhagwat, S. S.; Satoh, Y.; Sakata, S. T.; Buhr, C. A.; Albers, R.; Sapienza, J.;
Plantevin, V.; Chao, Q.; Sahasrabudhe, K.; Ferri, R.. Patent U.S. Patent Appl. Publ.
2004127536; Chem. Abstr. 2004, 141, 89085.
11. Bhagwat, S. S.; Satoh, Y.; Sakata, S. T.; Buhr, C. A.; Albers, R.; Sapienza, J.;
Plantevin, V.; Chao, Q.; Sahasrabudhe, K.; Ferri, R.; Narla, R. K. U.S. Pat. Appl.
Publ. 2005009876; Chem. Abstr. 2004, 142, 134589.
17. Hopkins, A. L.; Groom, C. R.; Alex, A. Drug Discovery Today 2004, 9, 430.
18. Antoni, L.; Sodha, N.; Collins, I.; Garrett, M. D. Nat. Rev. Cancer 2007, 7, 925.
19. Walton, M. I.; Eve, P. D.; Hayes, A.; Valenti, M.; De Haven Brandon, A.; Box, G.;
Boxall, K. J.; Aherne, G. W.; Eccles, S. A.; Raynaud, F. I.; Williams, D. H.; Reader, J.
C.; Collins, I.; Garrett, M. D. Mol. Cancer Ther. 2010, 9, 89.
20. Dunne, J.; Reardon, H.; Trinh, V.; Li, E.; Farinas, J. Assay Drug Devel. Technol.
2004, 2, 121.
12. Harris, J.; Church, N.; Proud, A.; Kling, M.; Vickery, B. D. U.K. Patent Appl.
2400101, 2004; Chem. Abstr. 2004, 141, 314349.
21. Smyth, L. A.; Collins, I. J. Chem. Biol. 2009, 3, 131.
13. Berdini, V.; Besong, G. E.; Callaghan, O.; Carr, M. G.; Congreve, M. S.; Gill, A. L.;
Griffiths-Jones, C. M.; Madin, A.; Murray, C. W.; Nijjar, R. K.; O’Brien, M. A.; Pike,
A.; Saxty, G.; Taylor, R. D.; Vickerstaffe, E. Patent Appl. WO2008078091, 2008;
Chem. Abstr. 2008, 149, 128825.
22. Knight, Z. A.; Shokat, K. M. Chem. Biol. 2005, 12, 621.
23. SelectScreen, Invitrogen (Scotland) Ltd (www.invitrogen.com).
24. Lochner, A.; Moolman, J. A. Cardiovasc. Drug Rev. 2006, 24, 261.