Novel Mps1 kinase inhibitors: From purine to pyrrolopyrimidine and quinazoline leads

Article abstract of DOI:10.1016/j.bmcl.2013.10.008

Mps1, also known as TTK, is a mitotic checkpoint protein kinase that has become a promising new target of cancer research. In an effort to improve the lead-likeness of our recent Mps1 purine lead compounds, a scaffold hopping exercise has been undertaken.

Full text of DOI:10.1016/j.bmcl.2013.10.008

Bioorganic & Medicinal Chemistry Letters 23 (2013) 6829–6833
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel Mps1 kinase inhibitors: From purine to pyrrolopyrimidine and
quinazoline leads
a
a
a
b
Matthew G. Bursavich a, , David Dastrup , Mark Shenderovich , Kraig M. Yager , Daniel M. Cimbora ,
⇑
Brandi Williams b, D. Vijay Kumar a
a Medicinal Chemistry, Myrexis, Inc., 305 Chipeta Way, Salt Lake City, UT 84108, United States
b In Vitro Pharmacology, Myrexis, Inc., 305 Chipeta Way, Salt Lake City, UT 84108, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Mps1, also known as TTK, is a mitotic checkpoint protein kinase that has become a promising new target
of cancer research. In an effort to improve the lead-likeness of our recent Mps1 purine lead compounds, a
scaffold hopping exercise has been undertaken. Structure-based design, principles of conformational
restriction, and subsequent scaffold hopping has led to novel pyrrolopyrimidine and quinazoline Mps1
inhibitors. These new single-digit nanomolar leads provide the basis for developing potent, novel
Mps1 inhibitors with improved drug-like properties.
Received 9 September 2013
Revised 30 September 2013
Accepted 4 October 2013
Available online 11 October 2013
Keywords:
Mps1
Ó 2013 Elsevier Ltd. All rights reserved.
TTK
Protein kinase inhibitors
Scaffold hopping
Conformational restriction
Structure-based design
Cancer
Purines
Pyrrolopyrimidines
Quinazoline
Mps1, also known as TTK, is a dual specificity protein kinase
that phosphorylates tyrosine, serine or threonine residues with a
critical role during mitosis. Mps1 facilitates chromosomal
alignment during metaphase, as well as proper attachment of the
bipolar microtubules to the kinetochores by eliminating misat-
tachments.1,2 It is also required for the full assembly of the spindle
checkpoint proteins at the kinetochore and activation of this
complex.3
Mps1 is dynamic kinase expressed only in proliferating cells
and activated via phosphorylation during mitosis. It is over
expressed in various human tumors and necessary for cellular pro-
liferation. Mps1 inhibition has been shown to cause premature
mitotic exit and gross aneuploidy, which is ultimately associated
with cell death. It has been hypothesized that mitotic checkpoints
are necessary to sustain cancer cellular proliferation in the pres-
ence of aneuploidy. Thus, Mps1 inhibition has become a promising
new target of cancer research.
O
O
S
HN
N
O
HN
N
N
N
N
N
N
N
N
N
H
N
H
N
H
N
H
R
OMe
1a: R = Me
2
1b: R = OMe
Figure 1. Myrexis Mps1 lead structures.
Mps1 (Fig. 1). Purine 1a has been shown to disrupt the spindle
assembly checkpoint, resulting in chromosome segregation defects
and aneuploidy.4,5 Purine 1a also demonstrated cytotoxicity across
a broad panel of tumor cell lines and exhibited antitumor activity
in nude mice bearing human tumor xenografts.4 Since we initiated
our efforts, several promising Mps1 kinase inhibitors have been
published (Fig. 2).6
We have recently published our efforts developing purine based
lead structures 1a, 1b and 2 as potent, selective, novel inhibitors of
Due to the high molecular weight (MW) and polar surface area
(TPSA) of these leads (Table 1), a de novo design effort was under-
taken. Herein we report those efforts leading to new pyrrolopyrim-
idine and quinazoline inhibitors of Mps1. These new analogs
⇑
Corresponding author. Address: EnVivo Pharmaceuticals, 500 Arsenal Street,
Watertown, MA 02472, United States. Tel.: +1 617-225-4219.
E-mail address: mbursavich@envivopharma.com (M.G. Bursavich).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.10.008

6830
M. G. Bursavich et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6829–6833
O
O
CN
O
S
N
O
O
N
N
H
N
N
N
HN
N
N
HN
N
F3CO
N
N
N
N
H
N
H
N
NH
OMe
N
OH
N
H
N
H
H2NO2S
NMS-P715
AZ3146
O
NH
N
OMe
MPS-1-IN-1
Shionogi
Figure 2. Published Mps1 kinase inhibitors.
Table 1
Structure–activity relationships of pyrimidines
O
O
R2
S
N
N
N
R1
N
N
N
H
OMe
Compd
R1
R2
Mps1a IC50 (nM)
HCT116a IC50
(l
M)
MW
TPSA
LEb
1a
2
6
7
8
—
—
H
H
Me
Me
—
—
H
Me
H
5
2
>100
35
28
0.023
0.011
—
2.0
1.5
424
501
461
475
475
489
100
137
108
99
108
99
0.27
0.25
—
0.23
0.23
—
9
Me
>100
—
a
Values are means of two experiments, standard deviations are 10%.
LE = ÀLog(Mps1 IC50)/# of heavy atoms.
b
O
O
S
R2
R2
N
N
Cl
N
a
b
R1
N
R1
R1
N
N
N
N
Cl
N
N
H
N
Cl
OMe
3
4
5
Scheme 1. Reagents and conditions: (a) amine, TEA, THF 50 °C; (b) aniline, cat.
TsOH, dioxane, 170 °C, MW.
Figure 3. Binding model overlay with 5-methyl pyrimidine structure (pink) and
Mps1 lead structure 2.
demonstrate potent Mps1 activity with significantly reduced TPSA
while maintaining their ligand efficiency (LE).7
In an effort to reduce the MW and TPSA of the lead compounds,
a series of pyrimidines were first designed and modeled. The bind-
ing model overlay of the 5-methyl pyrimidine scaffold with purine
lead compound 2 is shown in Figure 3.8
Figure 4. Binding model overlay with N-cyclohexyl pyrrolopyrimidine structure
(pink) and Mps1 lead structure 2.
The synthesis of the diaminopyrimidine inhibitors is shown in
Scheme 1.9 The synthesis begins with commercially available
2,4-dichloropyrimidines 3 (R1 = H or Me). Heating cyclohexyl
amine or N-methylcyclohexylamine with 3 and triethylamine in
THF at 50 °C provided compound 4 (R1 = H or Me; (R2 = H or Me).
Reaction of 4 with the desired aniline under catalytic acid condi-
tions in the microwave provided the diaminopyrimidines 5.10

M. G. Bursavich et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6829–6833
6831
R1
N
R1
HN
Cl
HN
N
a
b
Br
EtO
Br
N
N
Cl
N
N
Cl
X
N
Cl
10
11
12
R1
N
R1
N
N
c
d or e
N
N
Cl
N
N
H
R2
13a : R1 = H
14
13b : R1 = cyclohexyl
Scheme 2. Reagents and conditions: (a) ammonia, THF or cyclohexylamine, THF,
50 °C (b) B(OH)2CHCHOEt, NaHCO3, Pd(PPh3)4, DME, reflux; (c) AcOH, 120 °C; (d) (i)
Ar/HetArBr, CuI, K3PO4, ( )-trans-1,2-diaminocyclohexane, 90 °C; (ii) aniline, Cs2-
CO3, Pd(OAc)2, BINAP, toluene, reflux; (e) aniline, Cs2CO3, Pd(OAc)2, BINAP, toluene,
reflux.
Figure 5. Binding model overlay with quinazoline structure (pink) and Mps1 lead
structure 1b.
X
X
X
Kinase inhibition activity was determined as previously
described using full-length Mps1 enzyme at 2xKm ATP concentra-
tions. Cellular proliferation activity was determined by monitoring
cell growth densities in HCT116 cell cultures.4
a
b
c
N
N
NO2
N
OH
N
Cl
H
O
The importance of combined R1, R2 substituents was immedi-
ately evident. The unsubstituted compound (6: R1, R2 = H) was
shown to be inactive, but substitution of an R2 methyl group (7)
provided a potent starting point (IC50 = 35 nM). Incorporating an
R1 methyl group on the pyrimidine ring with an unsubstituted R2
group (8) provided a slightly more potent analog (IC50 = 28 nM).
While 7 and 8 are not as potent as purine 2, these analogs possess
reduced MW, reduced TPSA, and only slightly reduced LE as com-
pared with purine lead 2. Interestingly, the disubstituted com-
pound (9: R1, R2 = Me) was shown to be inactive, presumably due
to a unproductively biased conformation.
22
23
21
R1
N
X
Y
Y
d
N
N
N
N
N
N
H
N
H
OMe
OMe
24
25
Scheme 3. Reagents and conditions: (a) (i) Fe, AcOH, HCl, EtOH, 70 °C; (ii) urea,
180 °C; (b) POCl3, 110 °C; (c) aniline–HCl, IPA, 100 °C; (d) Ar/HetArB(OH)2, Na2CO3,
Pd(PPh3)4, DME or cyclohexylamine, Cs2CO3, Pd(OAc)2, BINAP, toluene, reflux.
Encouraged by the diaminopyrimidine results, and realizing the
potential for conformational restriction, a series of pyrrolopyrimi-
dines were designed and modeled. The binding model overlay of
the N-cyclohexyl pyrrolopyrimidine scaffold with purine lead com-
pound 2 is shown in Figure 4.8
(R1 = H) into 14 required an initial Buchwald reaction to function-
alize the indole NH and introduce the R1 substituent followed by a
second Buchwald–Hartwig reaction to incorporate the desired ani-
line.12,13 Conversion of 13b (R1 = cyclohexyl) into 14 required the
same Buchwald–Hartwig reaction to install the desired aniline.13
In the case of the pyrrolopyrimidines (Table 2) all the R1 = cyclo-
hexyl analogs (15–17) were shown to be potent Mps1 inhibitors.
The 2-methyl-morpholine analog 16 (R2 = Me, X = O) provided an
attractive starting point (IC50 = 22 nM) with a MW under 400, TPSA
The synthesis of the pyrrolopyrimidine inhibitors9 (Scheme 2)
begins with reaction of commercially available 5-bromo-2,4-
dichloropyrimidine 10 with either ammonia or cyclohexylamine
to provide 11. Suzuki reaction with the vinyl ether borane reagent
provided the pyrrolopyrimidine precursor 12.11 Cyclization under
acidic conditions afforded both 13a and 13b. Conversion of 13a
Table 2
Structure–activity relationships of pyrrolopyrimidines
R1
N
X
N
N
N
N
H
R2
Compd
R1
R2
X
Mps1a IC50 (nM)
HCT116a IC50
(l
M)
MW
TPSA
LEb
15
16
17
18
19
20
Cyclohexyl
Cyclohexyl
Cyclohexyl
Phenyl
2-Pyridyl
3-Pyridyl
OMe
Me
OMe
OMe
OMe
OMe
O
O
46
22
20
>100
32
7.5
8.6
0.75
—
10
—
408
392
485
479
480
480
64
55
101
101
114
114
0.24
0.26
0.23
—
0.22
—
NSO2Me
NSO2Me
NSO2Me
NSO2Me
>100
a
Values are means of two experiments, standard deviations are 10%.
LE = ÀLog(Mps1 IC50)/# of heavy atoms.
b

6832
M. G. Bursavich et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6829–6833
Table 3
Structure–activity relationships of quinazolines
R1
R1
N
Y
O
N
N
N
N
N
H
N
N
H
OMe
OMe
I
II: Y = O
III: Y = NMSO2Me
Compd
Scaffold
R1
Mps1a IC50 (nM)
HCT116a IC50
(lM)
MW
TPSA
LEb
26
27
28
29
30
31
32
I
I
Phenyl
N-Cyclohexyl
Phenyl
N-Cyclohexyl
Phenyl
3-Pyridyl
N-Cyclohexyl
>100
>100
12
9
5
8
—
—
412
434
412
434
490
491
511
60
72
60
72
96
—
—
II
II
III
III
III
0.85
6.32
1.74
1.58
0.81
0.26
0.25
0.25
0.24
0.25
109
108
2
a
Values are means of two experiments, standard deviations are 10%.
LE = ÀLog(Mps1 IC50)/# of heavy atoms.
b
of 55, and a LE slightly better than purine lead 2 (LE = 0.26 vs 0.25).
The 2-methoxy-4-sulfonamide analog 17 (R2 = Me, X = NSO2Me)
also provided a potent starting point (IC50 = 20 nM) demonstrating
submicromolar cellular toxicity, reduced MW and TPSA compared
to purine 2 (TPSA = 101 vs 137). In the case of the R1 = aryl or het-
eroaryl groups (18–20), both the phenyl and 3-pyridyl analogs
were shown to be substantially less active. The 2-pyridyl analog
19 was shown to be slightly less active against Mps1 than the
cyclohexyl analog (compare 17 vs 19), but maintained acceptable
TPSA and LE values, while the phenyl and 3-pyridyl analogs dem-
onstrated IC50s >100 nM.
conformational restriction and scaffold hopping led to new pyrrol-
opyrimidine and quinazoline inhibitors. These new scaffolds have
been shown to be potent Mps1 inhibitor starting points with low-
ered MW, TPSA, and good ligand efficiencies compared to the pur-
ine leads. Some of the analogs also possess submicromolar cellular
cytotoxicity. These new leads provide the basis for developing
more potent, novel inhibitors of Mps1 with drug-like properties.
Acknowledgement
We thank the Myrexis Analytical Department for their help
with compound purification and characterization.
Based on the encouraging results with the pyrrolopyrimidines, a
series of quinazolines were designed and modeled. The binding
model overlay of the phenyl quinazoline scaffold with purine lead
compound 1b is shown in Figure 5.8
References and notes
The synthesis of the quinazolines inhibitors9 (Scheme 3) begins
with commercially available halo-nitrophenyl aldehydes 21 (X = 6-
Br or 5-Cl). Iron reduction of the nitro group of 21 and subsequent
cyclization with urea provided intermediate 22. Treatment of 22
with POCl3 afforded the chloroquinazoline 23. Displacement of
the pyrimidinyl chloride with desired aniline under acidic condi-
tions provided compound 24.10 Finally, a Buchwald reaction12 to
install an R1 amine substituent or Suzuki reaction to incorporate
an R1 aryl or heteroaryl group provided the desired quinazoline
inhibitors 25.
While scaffold I analogs (26 and 27) were shown to be inactive,
the regioisomeric scaffold II and III analogs were much more prom-
ising (Table 3). The R1 = phenyl analog 28 provided a potent start-
ing point (IC50 = 12 nM) with submicromolar cellular toxicity,
reduced MW and TPSA (TPSA = 60 vs 137), and similar LE com-
pared to purine lead compound 1b. The R1 = N-cyclohexyl analog
29 provided a further boost in biochemical potency (IC50 = 9 nM)
with a slightly increased MW, reduced TPSA, and similar LE
compared to purine lead compound 1b. All three of the scaffold
III compounds possessing the aniline sulfonamide substituent
(30–32) showed single digit nanomolar Mps1 potency with slightly
decreased or similar MW, reduced TPSA, and similar LE compared
to purine lead compound 2. Compound 32 also demonstrated
submicromolar cellular toxicity.
1. Jelluma, N.; Brenkman, A. B.; van den Broek, N. J.; Cruijsen, C. W.; van Osch, M.
H.; Lens, S. M.; Medema, R. H.; Kops, G. J. Cell 2008, 132, 233.
2. Huang, H.; Hittle, J.; Zappacosta, F.; Annan, R. S.; Hershko, A.; Yen, T. J. J. Cell
Biol. 2008, 183, 667.
3. Liu, S. T.; Chan, G. K.; Hittle, J. C.; Fujii, G.; Lees, E.; Yen, T. J. Mol. Biol. Cell 2003,
14, 1638.
4. Tardiff, K. D.; Roger, A.; Cassiano, J.; Roth, B. L.; Cimbora, D. M.; McKinnon, R.;
Peterson, A.; Douce, T.; Robinson, R.; Dorweiler, I.; Davis, T.; Hess, M. A.;
Ostanin, K.; Papac, D. I.; Baichwal, V.; McAlexander, I.; Willardsen, J. A.;
Saunders, M.; Hoarau, C.; Kumar, D. V.; Wettstein, D. A.; Carlson, R. O.;
Williams, B. L. Mol. Cancer Ther. 2011, 10, 2267.
5. Kumar, D. V.; Hoarau, C.; Bursavich, M.; Slattum, P.; Gerrish, D.; Yager, K.;
Saunders, M.; Shenderovich, M.; Roth, B.; Mckinnon, R.; Chan, A.; Cimbora, D.;
Reeves, L.; Patton, S.; Bradford, C.; Papac, D.; Williams, B.; Carlson, R. O. Bioorg.
Med. Chem. Lett. 2012, 22, 4377.
6. (a) Caldarelli, M.; Angiolini, M.; Disingrini, T.; Donati, D.; Guanci, M.; Nuvoloni,
S.; Posteri, H.; Quartieri, F.; Silvagni, M.; Colombo, R. Bioorg. Med. Chem. Lett.
2011, 21, 4507; (b) Kwiatkowski, N.; Jelluma, N.; Fillippakopouls, P.;
Soundararajan, M.; Manak, M. S.; Kwon, M.; Choi, H. G.; Sim, T.; Deveraux, Q.
L.; Rottmann, S.; Pellman, D.; Shah, J. V.; Kops, G. J. P. L.; Knapp, S.; Gray Nat, N.
S. Chem. Biol. 2010, 6, 359; (c) Hewit, L.; Tighe, A.; Santaguida, S.; White, A. M.;
Jones, C. D.; Musacchio, A.; Green, S.; Taylor, S. S. J. Cell Biol. 2010, 190, 25; (d)
Kusakabe, K.; Ide, N.; Daigo, Y.; Tachibana, Y.; Itoh, T.; Yamamoto, T.;
Hashizume, H.; Hato, Y.; Higashino, K.; Okano, Y.; Sato, Y.; Inoue, M.; Iguchi,
M.; Kanazawa, T.; Ishioka, Y.; Dohi, K.; Kido, Y.; Sakamoto, S.; Yasuo, K.; Maeda,
M.; Higaki, M.; Ueda, K.; Yoshizawa, H.; Baba, Y.; Shiota, T.; Murai, H.;
Nakamura, Y. J. Med. Chem. 2013, 56, 4343.
7. Abad-Zapatero, C.; Metz, J. T. Drug Discovery Today 2005, 10, 464.
8. Molecular Docking Studies were Performed Using Schrodinger Program Glide:
Glide5.0 User Manual; Schrodinger Press, Schrodinger, L.L.C.: New York, NY,
2008; Crystal structure of Mps1 complex with a small molecule inhibitor (PDB
ID 3H9F, Ref. 6) solved with resolution 2.60 Å was used as a target for docking.
Docking model of Mps1 complex with compound 1 was further refined using
Schrodinger induced fit docking protocol: Sherman, W.; Day, T.; Jacobson, M.
In conclusion, we have presented a series of novel Mps1 inhib-
itors identified via de novo design efforts on a purine-based lead
scaffold. Structure design using molecular modeling, followed by

M. G. Bursavich et al. / Bioorg. Med. Chem. Lett. 23 (2013) 6829–6833
6833
P.; Friesner, R. A.; Farid, R. J. Med. Chem. 2006, 49, 534; The Images were made
with Accelrys Discovery Studio Visualizer, Release 3.2; Accelrys Software Inc.: San
Diego, USA, 2011.
10. Bursavich, M. G.; Lombardi, S.; Gilbert, A. M. Org. Lett. 2005, 7, 4113.
11. Whelligan, D. K. T.; Thomson, D. W.; Taylor, D.; Hoelder, S. J. Org. Chem. 2010,
75, 11.
9. All newly prepared compounds were purified by reverse phase HPLC and
characterized by LCMS, NMR and HRMS.
12. Surry, D. S.; Buchwald, S. L. Chem. Sci. 2010, 1, 13.
13. Wolfe, J. P.; Buchwald, S. L. J. Org. Chem. 2000, 65, 1144.