Imidazo[5,1-f][1,2,4]triazin-2-amines as novel inhibitors of polo-like kinase 1

Article abstract of DOI:10.1016/j.bmcl.2008.09.100

The synthesis and biological activities of imidazo[5,1-f][1,2,4]triazin-2-amines (imidazotriazines) as novel polo-like kinase 1 inhibitors are reported.

Full text of DOI:10.1016/j.bmcl.2008.09.100

Bioorganic & Medicinal Chemistry Letters 18 (2008) 6214–6217
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Imidazo[5,1-f][1,2,4]triazin-2-amines as novel inhibitors of polo-like kinase 1
*
M. Cheung , K. W. Kuntz, M. Pobanz, J. M. Salovich, B. J. Wilson, C. W. Andrews III, L. M. Shewchuk,
A. H. Epperly, D. F. Hassler, M. A. Leesnitzer, J. L. Smith, G. K. Smith, T. J. Lansing, R. A. Mook Jr.
GlaxoSmithKline, Fire Moore Drive, Research Triangle Park, NC 27709, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
The synthesis and biological activities of imidazo[5,1-f][1,2,4]triazin-2-amines (imidazotriazines) as
novel polo-like kinase 1 inhibitors are reported.
Received 7 September 2008
Revised 26 September 2008
Accepted 29 September 2008
Available online 2 October 2008
Ó 2008 Elsevier Ltd. All rights reserved.
Keywords:
Imidazotriazine
Polo-like kinase inhibitor
Polo-like kinases (PLK) are evolutionarily conserved serine/
threonine kinases that regulate critical processes in the cell cycle.1
The PLK family includes PLK1, PLK2 (SNK), PLK3 (PRK/FNK), and
PLK4 (SAK). PLK1 plays key roles in the regulation of mitotic pro-
gression including mitotic entry, spindle formation, chromosome
segregation and cytokinesis.2 With the discovery of PLK1 as a crit-
ical regulator of mitotic progression, it has become a target of keen
interest for the treatment of cancer.3–5 Among all the PLK inhibi-
tors under clinical investigation, dihydropteridinone BI 2536 and
thiophene amide GSK461364 (Fig. 1) are the only two molecules
that have demonstrated inhibition of PLK kinase activity.5
In order to discover inhibitors of PLK1, we conducted a focused
screen of our kinase inhibitor collection and a high-throughput
screen (HTS) of the GSK compound collection. From these screens,
multiple chemical series were identified. Of particular interest
were hits from a thiophene amide series and hits from an imidazo-
triazine series, exemplified by compound 1 and 2, respectively
(Fig. 2). Our efforts toward optimizing the thiophene amide series
will be reported in due course.6 Herein, we report the synthesis
and biological activity of novel PLK1 inhibitors from the imidazo-
triazines series.7
ing mode shown in Figure 3.8 In this proposed model, the benzyl
amine N–H and triazine 4-N were believed to form key hydrogen
bond donating and accepting interactions with the hinge residue
of Cys133 in PLK1, with both benzylic and phenyl groups facing to-
wards the solvent region. Based on this model and success in other
programs that target these regions of kinases,9 we decided to ex-
N
N
O
O
N
N
S
N
HN
N
NH2
CF3
O
O
N
N
N
N
O
H
GSK461364
BI 2536
Figure 1. Structure of BI 2536 and GSK461364.
In order to guide our SAR exploration, we built a PLK1 homology
model based on the active confirmation of the protein PKA C-alpha.
Docking of compound 2 into this model led to the proposed bind-
O
N
S
4
N
5
NH2
N
N
7
N
O
2
N
N
H
Abbreviations: PLK, polo-like kinase; SNK, serum-inducible kinase; PRK, prolif-
eration-related kinase; FNK, fibroblast-growth-factor-inducible kinase; SAK, Snk
akin kinase; HTS, high-throughput screening; SAR, Structure–activity relationships;
CDK, cyclin-dependent kinase.
2
1
* Corresponding author. Tel.: +1 610 270 7782; fax: +1 610 270 6609.
E-mail address: mui.h.cheung@gsk.com (M. Cheung).
Figure 2. Structure of thiophene amide 1 and imidazotriazine 2.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.09.100

M. Cheung et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6214–6217
6215
N
N
N
N
a
N
N
H2N
N
R1
N
N
H
12a-i
11
Scheme 2. Reagents and condition: (a) substituted aryl bromide or iodide, (S)-(ꢁ)-
2,20-bis(diphenylphosphino)-1,20-binaphthyl, NaO-t-Bu, Pd2(dba)3, microwave or
heating at 80 °C.
O
N
N
O
H
N
Figure 3. Imidazotriazine 2 in PLK1 homology model.
c,d
HN
N
a,b
6
plore the structure–activity relationships (SARs) around the phenyl
ring at position 7 and the benzyl group at position 2.
MeO
OMe
OMe
Synthetic access to the imidazotriazine ring system is rather lim-
ited.10 In order to explore the SAR of this series we needed to develop
a number of novel synthetic routes to allow better access to the tar-
geted compounds. Herein, we described novel syntheses of imidazo-
triazines (Scheme 1–3). Alkyl amine substituted analogs 10a–c were
synthesized using reactions described in Scheme 1. Racemic N-Boc
alanine 3 was coupled with N-methoxy-N-methylamine to form N-
methoxy-N-methyl amide 4 which was then displaced by lithiated
dithiane to form 5. Condensation of 5 with thiosemicarbazide fol-
lowed by cyclization in the presence of methyl iodide produced
tert-butyl 1-[3-(methylthio)-1,2,4-triazin-6-yl]ethylcarbamate (6)
and its regioisomer, tert-butyl 1-[3-(methylthio)-1,2,4-triazin-5-
yl]ethylcarbamate (7) in a 6.7:1 mixture with desired 6 obtained in
17% isolated yield. Though low yielding, the condensation and cycli-
zation step can be scaled up readily (600 g scale) to provide large
quantity (74 g) of intermediate 6. Deprotection of 6 followed by cou-
pling with benzoic acid gave compound 8 which was then cyclized
with POCl3 to give intermediate 9. Oxidation of the methylthio func-
tionality of 9 to the methyl sulfoxide using 1 equiv of m-chloroper-
benzoic acid (m-CPBA) and subsequent displacement with alkyl
amines gave compounds 10a–c (see Table 1).
13
O
N
N
N
H
R2
N
N
N
HN
N
HN
N
e
R2
MeO
OMe
MeO
OMe
OMe
OMe
14a-i
15
a-i
Scheme 3. Reagents and conditions: (a) m-CPBA, CH2Cl2, 3 h, 85%; (b) 3,4-5-
trimethoxyaniline, cat. TsOH, THF, rt; (c) 4 N HCl in dioxane, MeOH, 30 min; (d) for
aryloyl acid: substituted benzoic acid, Et3N, HATU, DMF; for aryloyl chloride:
substituted benzoyl chloride, Et3N, CH2Cl2, ꢁ78 °C, 1 h then rt; (e) 6 equiv, 1,2,4-
triazole, pyridine, 3 equiv, POCl3, rt, 25–46 h.
In contrast to alkyl amines, direct displacement of methyl sulf-
oxide group, obtained from mono-oxidation of intermediate 9,
with anilines was not fruitful. Instead, anilino substituted analogs
O
H
O
O
H
H
S
N
O
N
O
a
b
N
O
N
O
HO
S
O
O
O
5
3
4
O
O
S
N
N
e,f
N
O
N
N
H
O
c,d
H
+
N
N
O
S
N
7
6
N
N
R1
N
N
h,i
N
N
N
g
N
H
N
N
H
Ph
S
N
N
S
N
10a-c
9
8
Scheme 1. Reagents and conditions: (a) MeNHOMeꢀHCl, CDI, Et3N, CH2Cl2, 98%; (b) 1,3-dithiane, THF, ꢁ78 °C; then 2.5 M nBuLi, ꢁ15 to ꢁ25 °C, 2 h, then 4, 1.5 h, ꢁ15 °C, 77%;
(c) thiosemicarbazide, TsOH, 78 °C, 72 h, 100%; (d) MeI, CaCO3, CH3CN, 40 °C, 24 h, 17%; (e) 4 N HCl in dioxane, MeOH, 30 min; (f) PhCO2H, Et3N, HATU, DMF; (g) 6 equiv, 1,2,4-
triazole, pyridine, 3 equiv, POCl3, rt, 25–46 h; (h) 1 equiv, mCPBA, CH2Cl2, 3 h; (i) R1NH2, heat.

6216
M. Cheung et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6214–6217
Table 1
Table 3
PLK1 activity for compounds 2, 11, 10a–c.
PLK1 activity for compounds 12i, 15a–i.
OMe
N
MeO
MeO
N
N
R1
N
N
N
N
N
H
N
N
H
R2
PLK1 enzyme IC50 (lM)
a
Compound
R1
PLK1 enzyme IC50 (lM)
a
Compound
R2
H
2-OMe
2-NO2
2-Br
3-Br
3-CF3
4-CF3
2
11
10a
10b
10c
–CH2Ph
–H
–CH2CH2-p-pyridine
–CH2CH2CH2-morpholine
–Cyclohexyl
1.5
>10
>10
>10
>10
12i
0.04
5.66
1.98
2.67
0.07
0.03
1.06
0.41
0.50
0.08
15a
15b
15c
15d
15e
15f
15g
15h
15i
a
Values are means of P3 experiments.
3-C(O)Ph
3-O-Ph
4-F
(compounds 12a–i, see Table 2) were prepared by coupling of aryl
halides such as aryl bromides or aryl iodides to 5-methyl-7-pheny-
limidazo[5,1-f][1,2,4]triazin-2-amine (11) using standard palla-
dium coupling chemistry (Scheme 2).
a
Values are means of P3 experiments.
In order to explore the SAR around the phenyl substituent at po-
sition 7, the synthesis shown in Scheme 3 was developed. Oxida-
tion of thiomethyl derivative 6 followed by treatment with 2,3,4-
trimethyoxyaniline yielded compound 13. Deprotection of the
tert-butyoxycarbonyl group under acidic conditions and subse-
quent coupling with either aryl acids or aryl acid chlorides pro-
vided compounds 14a–i. Treatment of 14 with conventional
POCl3 cyclization conditions resulted in poor yields (0–20%) of
the desired imidazotriazine derivatives. Interestingly, addition of
6 equiv of 1,2,4-triazole significantly improved the POCl3 cycliza-
tion reaction to give the imidazotriazines 15a–i in 30–50% yield.
The PLK1 enzyme inhibitory activity of the compounds pre-
pared is shown below (Tables 1–3). As shown in Table 1, removal
of the benzyl group (compound 11) resulted in significant reduc-
tion in the PLK1 activity.11 Extending the side chain (compounds
10a, 10b) or replacing the benzyl group with a cyclohexyl group
(compound 10c) also resulted in reduction of the PLK1 activity.
Interestingly, when the benzyl group at position 2 was replaced
by phenyl (compound 12a), PLK1 potency was maintained (Table
2). Substitution of the phenyl group with various functional groups
was tolerated. Compounds with electron donating groups (12b–d,
12h) showed higher PLK1 activity than compounds with electron-
withdrawing groups (12e–g). Incorporation of the 3,4,5-trime-
thoxyphenyl substitution (compound 12i), led to improvement in
PLK1 potency by 25-fold over compound 2.
Given the enhanced potency observed with the 3,4,5-trimeth-
oxylaniline substituent (12i), this group was held constant as we
explored the SAR of the aryl ring at the 7-position (Table 3). Com-
pounds with both electron donating (15a) and electron withdraw-
ing (15b–c) groups at the ortho-position resulted in a drop in PLK1
potency when compared to parent compound (12i). Meta-substitu-
tion was better tolerated than ortho- and para-substitution as evi-
denced by 3-Br compound (15d) being 38-fold more potent against
PLK1 than the 2-Br compound (15c), and 3-CF3 compound (15e)
being 35-fold more potent than 4-CF3 compound (15f). Com-
pounds with large substituents at the meta-position (15g–h) and
small substituents at the para-position (15i) were also tolerated.
The most active PLK1 inhibitors were evaluated in cellular as-
says for their ability to inhibit proliferation of a variety of tumor
cell lines.11 As shown in Table 4, compounds 12i, 15d, and 15e
were modest inhibitors of proliferation in the tumor cell lines
examined.
Compounds 12i and 15e were evaluated against a panel of 40
kinases (including PLK3) to assess kinase selectivity. Compounds
12i and 15e were found to inhibit PLK3 with an IC50 of 0.09
lM
and 0.03 M, respectively, and thus are to be considered dual
l
inhibitors of PLK1 and PLK3. Compound 15e was at least 10-fold
selective for PLK1 and PLK3 versus the other kinases tested. In gen-
eral, compound 12i was less selective.
Pharmacokinetic properties of compounds 12i and 15e in the
mouse were also evaluated (Table 5). Compound 15e had high
clearance (81 mL/min/kg) and modest oral bioavailability (18%F),
while compound 12i had moderate clearance (53 mL/min/kg) and
a lower oral bioavailability (10%F).
Table 2
PLK1 activity for compounds 12a–i.
N
N
N
R1
N
H
N
To build confidence in the binding mode used in the homology
model, an X-ray structure of 12i bound in CDK2/cyclin A was ob-
tained.12 As shown in Figure 4, the binding mode of 12i in the
CDK2/cyclin A is consistent with the proposed binding mode in
the PLK homology model. The inhibitor hydrogen bonds to the
backbone NH and C@O of Leu83 in the CDK2/cyclin A hinge region.
It was interesting to note that one of the methoxy groups is within
hydrogen bonding distance of the side chain of Asp86 in the CDK2
structure.
In summary, we have discovered a novel series of polo-like kinase
inhibitors based on an imidazotriazine template. A PLK1 homology
model was employed to identify key areas to explore the struc-
ture–activity relationships of the series. Modifications led to an
12a-i
a
Compound
R1
H
–3-OMe
–4-OMe
PLK1 enzyme IC50 (lM)
12a
12b
12c
12d
12e
12f
12g
12h
12i
1.2b
0.3b
0.68b
0.38
>10
2.81
2.74
0.09
0.04
–4-O-CH2CH2N(Me)2
–3-CF3
–4-NO2
–4-SO2NH2
3,4-DiOMe
3,4,5-TriOMe
a
Values are means of P3 experiments.

M. Cheung et al. / Bioorg. Med. Chem. Lett. 18 (2008) 6214–6217
6217
Table 4
PLK1 enzymatic and anti-proliferative activities for compounds 12i, 15d, and 15e.
a
Compound
IC50 (lM)
PLK1 enzyme
RKO (colon)
HCT116 (colon)
H460 (lung)
MCF7 (breast)
PC3 (prostate)
12i
15d
15e
0.04
0.07
0.03
3.7
7.6
3.2
5.3
7.8
3.9
6.5
5.8
5.4
5.2
11
4.8
12
26
7.8
a
Values are means of P 3 experiments for PLK1 enzyme assay and values are means of 2–4 experiments for anti-proliferative assays.
Table 5
Pharmacokinetic for compounds 12i and 15e in mouse.
Compound
Dose (iv) (mg/kg)
Cl (mL/min/kg)
Vdss (L/kg)
AUCpo (ng h/mL)
%F
12i
15e
5a
53
81
1.6
2.3
301
264
10
18
2.9b
a
Compound 12i was formulated in 10% DMSO in solutol solution.
Compound 15e was formulated in 95:5 intralipid/DMSO for iv dosing and 0.5% HPMC/0.1% Tween 80 for po dosing.
b
References and notes
1. Dai, W. Oncogene 2005, 24, 215.
2. Barr, F. A.; Silljé, H. H. W.; Nigg, E. A. Nat. Rev. Mol. Cell Biol. 2004, 5, 429.
3. Lansing, T. J.; McConnell, R. T.; Duckett, D. R.; Spehar, G. M.; Knick, V. B.;
Hassler, D. F.; Noro, N.; Furuta, M.; Emmitte, K. A.; Gilmer, T. M.; Mook, R. A., Jr.;
Cheung, M. Mol. Cancer Ther. 2007, 6, 450.
4. McInnes, C.; Mezna, M.; Fischer, P. M. Curr. Top. Med. Chem. 2005, 5, 181.
5. Warner, S. L.; Stephens, B. J.; Von Hoff, D. D. Curr. Oncol. Rep. 2008, 10, 122.
6. Andrews, C. W., III; Cheung, M.; Davis-Ward, R. G.; Drewry, D. H.; Emmitte, K.
A.; Hubbard, R. D.; Kuntz, K. W.; Linn, J. A.; Mook, R. A.; Smith, G. K.; Veal, J. M.
PCT Int. Appl. WO04015899, 2004.
7. Cheung, M.; King, N. P.; Kuntz, K. W.; Mook, R. A., Jr.; Pobanz, M. A.; Salovich, J.
M.; Wilson, B. J. PCT Int. Appl. WO04087652, 2004.
8. The crystal structure of the catalytic domain of polo-like kinase was not
available when we initiated our lead optimization effort. The structure was
published in 2007 Kothe, M.; Kohls, D.; Low, S.; Coli, R.; Cheng, A. C.; Jacques, S.
L.; Johnson, T. L.; Lewis, C.; Loh, C.; Nonomiya, J.; Sheils, A.; Verdries, K. A.;
Wynn, T. A.; Kuhn, C.; Ding, Y.-H. Biochemistry 2007, 46, 5960.
9. (a) Harris, P. A.; Boloor, A.; Cheung, M.; Kumar, R.; Crosby, R. M.; Davis-Ward, R.
G.; Epperly, A. H.; Hinkle, K. W.; Hunter, R. N., III; Johnson, J. H.; Knick, V. B.;
Laudeman, C. R.; Luttrell, D. K.; Mook, R. A.; Nolte, R. T.; Rudolph, S. K.;
Szewczyk, J. R.; Truesdale, A. T.; Veal, J. M.; Wang, L.; Stafford, J. A. J. Med. Chem.
2008, 51, 4632; (b) Harris, P. A.; Cheung, M.; Hunter, R. N., III; Brown, M. L.;
Veal, J. M.; Nolte, R. T.; Wang, L.; Liu, W.; Crosby, R. M.; Johnson, J. H.; Epperly,
A. H.; Kumar, R.; Luttrell, D. K.; Stafford, J. Am. J. Med. Chem. 2005, 48, 1610.
10. Clarke, R. W.; Garside, S. C.; Lunts, L. H. C.; Hartley, D.; Hornby, R.; Oxford, A. W.
J. C. S. Perkin Trans. 1: Org. Bioorg. Chem. 1989, 4, 1120.
Figure 4. X-ray structure of 12i bound in CDK2/Cyclin A.
11. Description of assay conditions for PLK1 kinase assay and in vitro 72 h growth
3
inhibition assays can be found in Ref.
.
12. X-ray coordinate of 12i in CDK2/cyclinA has been deposited with the RCSB
Protein Data Bank (PDB) database with PDB deposition number of 3EOC.
improvement in the PLK1 potency by ꢂ50-fold. In addition, we have
developed novel synthetic routes that allow preparation of multiple
derivatives to explore the SAR of the imidazo[5,1-f][1,2,4]triazin-2-
amines series, a novel template for kinase inhibition.
Compound 12i inhibited CDK2/cyclinA with an IC50 of 0.23 lM. CDK2/cyclinA
was expressed, purified, and crystallized as previously described.13 Crystals
were cross-linked with 25% glutaraldehyde for 15 min, then soaked with 1 mM
compound for 3 days. Prior to data collection, glycerol was added to 25% and
the crystals were flash-frozen in liquid nitrogen. Data were collected [99%
complete to 3.2 Å resolution (reflections/observations, 33,305/207,344; Rmerge
,
11%)] on an RAXISIV detector and processed with HKL2000.14 The structure
was solved by molecular replacement using Refmac15 and coordinates 1FVV
from the Protein Data Bank. The structure was refined to an Rfactor/Rfree of 21/
23% at 3.2 Å resolution.
Acknowledgments
We thank Philip Harris for reviewing this manuscript and Scott
Dickerson for providing Figure 3. We acknowledge C. Mckernan for
providing compound 11, V. Bordas for providing compounds 12a–
c, and L.C. Miller for providing compounds 10c, 12d, and 12h.
13. Jeffrey, P. D.; Russo, A. A.; Polyak, K.; Gibbs, E.; Hurwitz, J.; Massagué, J.;
Pavletich, N. P. Nature 1995, 376, 313.
14. Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307.
15. Murshudov, G. N.; Vagin, A. A.; Dodson, E. J. Acta Crystallogr. D 1997, 53, 240.