Structure guided optimization, in vitro activity, and in vivo activity of pan-PIM kinase inhibitors

Article abstract of DOI:10.1021/ml400307j

Proviral insertion of Moloney virus (PIM) 1, 2, and 3 kinases are serine/threonine kinases that normally function in survival and proliferation of hematopoietic cells. As high expression of PIM1, 2, and 3 is frequently observed in many human malignancies, including multiple myeloma, non-Hodgkins lymphoma, and myeloid leukemias, there is interest in determining whether selective PIM inhibition can improve outcomes of these human cancers. Herein, we describe our efforts toward this goal. The structure guided optimization of a singleton high throughput screening hit in which the potency against all three PIM isoforms was increased >10,000-fold to yield compounds with pan PIM K_is < 10 pM, nanomolar cellular potency, and in vivo activity in an acute myeloid leukemia Pim-dependent tumor model is described.

Full text of DOI:10.1021/ml400307j

Letter
pubs.acs.org/acsmedchemlett
Structure Guided Optimization, in Vitro Activity, and in Vivo Activity
of Pan-PIM Kinase Inhibitors
Matthew T. Burger,*,† Wooseok Han,† Jiong Lan,† Gisele Nishiguchi,† Cornelia Bellamacina,†
Mika Lindval,† Gordana Atallah,† Yu Ding,† Michelle Mathur,† Chris McBride,† Elizabeth L. Beans,†
Kristine Muller,† Victoriano Tamez,† Yanchen Zhang,† Kay Huh,† Paul Feucht,‡ Tatiana Zavorotinskaya,‡
Yumin Dai,‡ Jocelyn Holash,‡ Joseph Castillo,‡ John Langowski,‡ Yingyun Wang,‡ Min Y. Chen,‡
and Pablo D. Garcia‡
†Global Discovery Chemistry/Oncology & Exploratory Chemistry, Novartis Institutes for Biomedical Research, 4560 Horton Street,
Emeryville, California 94608, United States
‡Oncology Research, Novartis Institutes for Biomedical Research, 4560 Horton Street, Emeryville, California 94608, United States
S
* Supporting Information
ABSTRACT: Proviral insertion of Moloney virus (PIM) 1, 2,
and 3 kinases are serine/threonine kinases that normally
function in survival and proliferation of hematopoietic cells. As
high expression of PIM1, 2, and 3 is frequently observed in
many human malignancies, including multiple myeloma, non-
Hodgkins lymphoma, and myeloid leukemias, there is interest
in determining whether selective PIM inhibition can improve
outcomes of these human cancers. Herein, we describe our
efforts toward this goal. The structure guided optimization of a
singleton high throughput screening hit in which the potency against all three PIM isoforms was increased >10,000-fold to yield
compounds with pan PIM Kis < 10 pM, nanomolar cellular potency, and in vivo activity in an acute myeloid leukemia Pim-
dependent tumor model is described.
KEYWORDS: Proviral insetion site in Moloney murine leukemia virus kinases inhibitors, Pim1 kinase inhibitor, Pim2 kinase inhibitor,
Pim3 kinase inhibitor, pan-Pim kinase inhibitors
roviral insertion site of Moloney murine leukemia virus
Pkinases, or PIM 1, 2, and 3 kinases are constitutively active
serine/threonine kinases that normally function in the survival,
proliferation, and differentiation of hematopoietic cells in
response to growth factors and cytokines.1,2 PIM’s play
redundant roles in oncogenesis and, therefore, suggest that a
pan-PIM kinase inhibitor may be clinically useful.3 In human
disease, high expression and/or dysfunction of the three PIMs
has been implicated in the progression of hematopoetic and
solid tumor cancers.1,2 In addition to cancer, PIM kinases have
been reported to play a role in several autoimmune diseases.4
Not surprisingly, PIM kinases have emerged as attractive
therapeutic targets and have elicited several groups to
investigate and report novel inhibitors of PIM5−10 including
Figure 1. Pim clinical compounds and starting point 1a.
the clinical compounds SGI-17766 and AZD1208,7 Figure 1.
Pim kinases share a high level of sequence homology within the
properties and was recently used to establish a PK/PD efficacy
relationship in a PIM2 driven multiple myeloma xenograft
model.12 Here we also demonstrate efficacy in the AML EOL-1
xenograft model.
family (>61%) and all share the unique feature of being the
only kinases with a proline in the hinge,11 which results in only
one hydrogen bond interaction with ATP. As the ATP Km for
PIM2 is 10−100× lower than that for PIM1 and PIM3, cell
active pan PIM inhibitors have been more challenging to
identify than PIM 1/3 inhibitors. Herein, we describe potent
and selective cell active inhibitors of all three PIM kinases. A
representative of this compound series, 5c, has suitable PK
Received: August 6, 2013
Accepted: October 12, 2013
Published: October 15, 2013
© 2013 American Chemical Society
1193
dx.doi.org/10.1021/ml400307j | ACS Med. Chem. Lett. 2013, 4, 1193−1197

ACS Medicinal Chemistry Letters
Letter
The starting point for our discovery efforts was the singleton
high throughput screening hit 1a. While of modest pan PIM
potency, we followed up on it due to its low molecular weight
(344), good PIM1 ligand efficiency (LE = 0.41, FQ = 1.01) and
a presumed nonplanar ground state conformation for the
ortho-substituted acylaniline moiety, which we reasoned might
ultimately be beneficial with respect to physicochemical
properties.
acidic patch below and hydrophobic interactions to the glycine
rich loop above. Additionally, the NH-acetyl substituent
extended toward the hydrophobic lower hinge, making no
hydrogen bonds. Furthermore, the central amide in the
molecule made no hydrogen bonds as well, appearing to
serve only as a rigid linker connecting the phenyl and thiazole
rings.
Hit optimization efforts were initiated by variation of the
piperidine (A ring), phenyl (B ring), and N−Ac thiazole (C
ring) components of compound 1a. PIM1−3 kinase activity
was assessed initially in a Kinase-Glo assay, with [ATP] at or
below ATP Km for each isoform. As compound potency
increased, vide infra, the assay format was changed to a high
[ATP] Alphascreen format to extend the assay sensitivity.
Compound enzymatic data is presented as Kis to allow
comparison of activity of compounds run in the two assay
formats. Removal of the N−Ac from the thiazole 1a starting
point yielded 1b with >10× reduced potency, Table 1 . From
this weakly potent compound lacking any extensions to the
lower hinge, modifications in the C ring with heterocycles were
surveyed. Of note, aminopyrazine 1c increased potency relative
to 1b as well as demonstrating measurable PIM2 potency with
no extenstion to the lower hinge and with a half unit reduction
in cLogP. Modification of the phenyl B ring in 1a targeted
potential hydrogen bonding interactions with catalytic Lys67.
Multipe heterocycles and amino or hydroxy substituted
heterocycles gave marginal improvement in potency. However,
it was noted that B ring pyridine 2b maintained the potency of
the lead 1a, while having a 0.5 unit reduction in cLogP.
Combination of the aminopyrazine C ring and pyridine B ring
led to compound 3a with a clogP of 1.9, a 3× improved
potency against PIM1 and 3 and now measurable PIM2
potency (≥ 10× improvement). From 3a, variation of the
piperidine A ring with various amine and amide substituted
Prior to starting any synthetic chemistry efforts, a cocrystal
structure of compound 1a in PIM1 was obtained, Figure 2.13
Figure 2. Structure of 1a in PIM1.
We were intrigued by several aspects of the binding interactions
(or lack thereof), which suggested multiple avenues to increase
potency and, importantly, not necessarily increase the size or
lipophilicity. First, there was no H bonding interaction to the
hinge (or any other part of the protein). Second, the phenyl
group was positioned in proximity to the catalytic Lys67. Third,
the piperidine was in a chair conformation with well-defined
vectors to access potential hydrogen bonding interactions in the
Table 1. Selected SAR for A, B, and C Ring Variations of 1a
compd
A ring
B ring
X,Y,Z = CH
C ring
PIM1 Ki (nM) PIM2 Ki (nM) PIM3 Ki (nM) cLogP
1a
1b
1c
1d
1e
1g
1h
2a
2b
2c
2d
2e
3a
3b
3c
3d
3e
3f
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperidyl
1-piperazinyl
2-NHAc-4-thiazolyl
4-thiazolyl
92
>12,500
>12,500
7,000
230
2.8
2.9
2.5
3.1
4.0
2.8
2.4
2.3
2.3
1.6
1.6
1.9
1.9
0.5
−0.2
0.0
0.5
0.6
1.3
X,Y,Z = CH
1,800
460
>12,500
1,500
12,000
>12,500
11,000
>12,500
140
X,Y,Z = CH
3-NH2, 2-pyrazinyl
3-NH2, 2-pyridyl
3-OH, 2-pyridyl
X,Y,Z = CH
3,300
12,300
2,300
>12,500
59
>12,500
>12,500
>12,500
>12,500
6,300
X,Y,Z = CH
X,Y,Z = CH
6-NH2, 2-pyridyl
2-imidazoyl
X,Y,Z = CH
Y = N, X,Z = CH
Z = N, X,Y = CH
X,Z = N, Y = CH
X,Z = N, Y = CNH2
X = CH,Z = N,Y = CNH2
Z = N, X,Y = CH
Z = N, X,Y = CH
Z = N, X,Y = CH
Z = N, X,Y = CH
Z = N, X,Y = CH
Z = N, X,Y = CH
Z = N, X,Y = CH
2-NHAc-4-thiazolyl
2-NHAc-4-thiazolyl
2-NHAc-4-thiazolyl
2-NHAc-4-thiazolyl
2-NHAc-4-thiazolyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyrazinyl
3-NH2, 2-pyridyl
61
>12,500
>12,500
>12,500
>12,500
1,600
220
>12,500
>12,500
1,000
30
>12,500
>12,500
1,700
90
810
>12,500
2,000
1,000
490
4-(2-aminoacetyl)piperazin-1-yl
3-aminopropanoyl)piperazin-1-yl
( )-3-(aminomethyl)piperidin-1-yl
(S)-3-aminopiperidin-1-yl
122
56
680
180
11
700
50
16
70
20
3g
(S)-3-aminopiperidin-1-yl
18
110
20
1194
dx.doi.org/10.1021/ml400307j | ACS Med. Chem. Lett. 2013, 4, 1193−1197

ACS Medicinal Chemistry Letters
Letter
hetereocycles, which could potentially interact with the protein
side chain carboxylates and backbone amides in the ribose
patch (Asp128 and Glu171), was targeted. While most
modifications offered little if any potency advantage over 3a,
the 3-S amino piperidine compound 3f stood out with
increased potency while further lowering the cLogP to 0.6.
From only limited kinase profiling of 3f it was apparent that the
aminopyrazine was fairly promiscuos with its donor and
acceptor hinge motif14 as relative to PIM inhibition several
non-PIM kinases were inhibited with equivalent or greater
potency.15 At this time, the unique nature of the PIM hinge was
exploited by switching to the donor only C ring aminopyridine
3g, which maintained comparable PIM potencies relative to 3f,
but with a >100 potency reduction in the non-PIM kinases
tested.16
From compound 3g (cLogP = 1.3, cLogD7.4 = −0.7), the
lower hinge was explored with an eye toward improving
potency even if it came at the expense of increased lipophilicity
as an ultimate logD7.4 of 1−3 was targeted. The chemistry was
amenable to extensive variation (via a C ring 6-bromo-3-
aminopyridine) with oxygen, nitrogen, and carbon linkages
explored, Table 2. Potency gains were realized as many
compounds reached the limits of detection in the low [ATP]
assay conditions. Ultimately, the 2,6 difluorophenyl proved
highly potent with compound 4j exhibiting picomolar17 pan
PIM potencies while retaining a reasonable lipophilicity (cLogP
= 3.7, cLogD7.4 = 1.5). Notably, the panPIM potency increase
of 4j relative to 3g did not translate into a similar potency
increase for the non-PIM kinases tested, as the selectivity
window actually widened.18
Further optimization of 4j focused on improving overall
drug-like properties, Table 3. Incorporation of a trans hydroxy
on the piperidine, 5a, reduced the lipophilicity while
maintaining picomolar pan Pim potency. Attempts to increase
potency further with minimal increase in lipophilicity included
evaluation of a 5-fluoro substituent on the carboxypyridyl ring
as well as targeting a hydrophobic dimple in the glycine loop
with additional substituents on the piperidine ring. Incorpo-
ration of both of these modifications in the methyl-, hydroxy-,
and amino-substitued piperidine 5b did indeed increase the
panPim potency. With the hydroxy addition to the amino-
piperidine, it was noted that the PSA, 130, of 5b as well as
number of hydrogen bond donors, six, was increasing. A
reduction in both properties was then pursued for potential
improvement of overall drug like properties. As the original lead
lacked any hinge interaction with PIM kinases, we proceeded to
ask the question of whether from these optimized ligands the
hinge amino donor interaction could be removed without
compromising potency. The hinge donor interaction could
indeed be removed without a dramatic loss in potency, as
compound 5c maintained picomolar enzymatic activity and low
nanomolar cellular activity, vide infra. When compared
alongside the clinical compounds SGI-1776 and AZD1208,
5c is more potent in enzymatic and cellular, vide infra, assays.
As may be expected with no potential hinge interacting moiety,
kinase profiling of 5c using the KINOME scan binding
displacement assay indicated high kinase selectivity [S(35) =
0.026 at 1 μM, 442 kinases). With a measured logD7.4 of 2.2
and Kis of 1−2 picomolar for PIMs 1−3, the LLE of 5c is quite
high (8.4 with respect to cLogP and 9.2 with respect to
measured logD7.4).
Table 2. Selected SAR for R1 Analogues of 3g
PIM1 Ki
(nM)
PIM2 Ki
(nM)
PIM3 Ki
(nM)
compd
R1
cLogP
3g
4a
4b
4c
4d
4e
hydrogen
phenyoxy
2-pyridyl
3-pyridyl
4-pyridyl
18
4
110
122
6
20
15
3
1.3
3.4
2.2
2.0
2.0
1.2
a
<1
a
<1
12
6
4
a
<1
5
2-NH2-4-
pyrimidyl
1
16
3
a
a
4f
4g
4h
4i
2-CH3O-phenyl
2-CF3-phenyl
2-Cl-phenyl
<1
8
7
<1
2.8
4.3
3.8
3.5
3.7
The crystal structure of 5c in PIM-1 was obtained, Figure
3.19 The overall binding mode is the same as the original lead
1a with the central amide making no hydrogen bonds and
displaying the two sides of the molecule toward the hinge/
lower hinge and catalytic Lys region/ribose patch. The amino
group on the piperidine ring makes two hydrogen bonds to the
Asp128 side chain carboxylate and the Glu171 backbone
3
3
a
a
a
<1
<1
<1
a
a
a
2-F-phenyl
<1
<1
<1
b
b
b
4j
2,6-diF-phenyl
0.002
0.004
0.006
a
b
KinaseGlow assay detection limit. Data from high [ATP]
Alphascreen assay.
Table 3. Selected SAR for Analogues of 4j and Comparison to SGI-1776 and AZD1208
compd
4j
R1
R2
R3
R4
PIM1 Ki (nM)
PIM2 Ki (nM)
PIM3 Ki (nM)
cLogP
H
H
NH2
NH2
NH2
H
H
H
F
0.002
0.002
0.001
0.001
16
0.004
0.006
0.002
0.002
610
0.006
0.007
0.003
0.001
24
3.7
2.4
3.1
3.1
4.6
3.2
5a
OH
OH
OH
H
5b
CH3
CH3
5c
F
SGI-1776
AZD1208
0.017
0.16
0.23
1195
dx.doi.org/10.1021/ml400307j | ACS Med. Chem. Lett. 2013, 4, 1193−1197

ACS Medicinal Chemistry Letters
Letter
well tolerated, as judged by minimal body weight loss (within a
margin of 5% of initial weight) observed for both groups.
Figure 3. Structure of 5c in PIM1.
carbonyl in the ribose patch. Additionally, the methyl group fills
a hydrophobic dimple in the glycine rich loop above. The B
ring pyridine nitrogen makes a hydrogen bond with the
catalytic Lys 67. The C ring fluoropyridine fills the Pro123
hinge region with no hydrogen bond interactions and the
difluorophenyl ring extends into the lower hinge region.
The picomolar enzymatic PIM activity of 5a−5c translates to
nanomolar cellular activity in mechanism target modulation and
antiproliferative cell assays for cell lines dependent on PIM1, 2,
or 3. For example, in the multiple myeloma derived KMS-11
luc cell line with high expression levels of PIM2, compound 5c
Ser112
inhibits phosphorylation of pBAD
(a direct substrate) at
Ser240/4
10 nM, inhibits phosphorylation of pS6RP
(a down-
stream substrate) at 38 nM and inhibits cell proliferation at 17
nM.12 The large upward potency shift in going from the
enzyme to cells was consistent with the low ATP Km for
PIM2.20 In comparison, SGI1776 and AZD1208 inhibit cell
proliferation in the KMS-11luc cell line at 4500 and 680 nM,
respectively. In the acute myeloid leukemia derived EOL-1 cell
line,21 compound 5c inhibits cell proliferation at <40 nM.
As a highly kinase selective and cell active pan-PIM inhibitor,
compound 5c was next assessed for it is potential to serve as a
tool to study the effect of PIM kinase inhibition in vivo. The
close focus on controling lipophilicity during potency
optimization proved beneficial as 5c did indeed possess suitable
in vivo properties. Upon dosing 5 mg/kg IV and 10 mg/kg PO
of 5c to mice, a CL of 25 mL/min/kg, a bioavailabity of 55%
and a Vss of 4 L/kg were determined. Upon dosing 10 mg/kg
IV and 20 mg/kg PO of 5c to rats, a CL of 13 mL/min/kg, a
bioavailabity of 44%, and a Vss of 3.4 L/kg were determined.
The favorable physical properties of 5c were further reflected in
the ability to conduct the in vivo studies using acetate buffer
formulations. At higher oral doses in both species (up to 100
mg/kg) the plasma exposure increased in a roughly propor-
tional manner relative to the screening PK studies, vide infra. As
such, free plasma levels of 5c22 in excess of the cellular target
modulation and antiproliferative EC50s can be achieved and
maintained in rodents.
Figure 4. Antitumor activity and plasma exposure of 5c in EOL-1
acute myeloid leukemia subcutaneous tumor model.
In summary, the structure guided optimization of singleton
screening hit 1a in which the potency against all three PIM
isoforms was increased >10,000-fold, with no increase in
lipophilicity, to yield compound 5c that exhibits single digit
picomolar pan PIM potency, high kinase selectivity, and in vivo
properties suitable to achieve efficacy in an acute myeloid
leukemia PIM-dependent tumor model has been described.
During the optimization, a focus was initially placed on
increasing potency while simultaneously reducing the lipho-
philicity, as enthalpic contributors to potency were identified.
From this lowered clogP space, a hydrophobic interaction was
optimized ultimately yielding compound 5c. Our continued
efforts toward bringing forward selective and potent pan PIM
kinase inhibitors to benefit cancer patients will be reported in
due course.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details for the synthesis and characterization of all
compounds, biological assays, and pharmacology model
procedures. This material is available free of charge via the
Internet at http://pubs.acs.org.
The target modulation and antiproliferative activity of 5c in a
multiple myeloma (KMS-11.luc) tumor xenograph model has
been recently reported.12 Continual modulation of target, as
judged by pS6RP inhibition, was associated with maximal
efficacy. In addition to antitumor activity in the multiple
myeloma model, here we demonstrate efficacy in the EOL-1
acute myeloid leukemia xenograph model. Administration of 5c
at 30 and 100 mg/kg daily yields significant dose-dependent
reduction in tumor growth, Figure 4. Both doses tested were
AUTHOR INFORMATION
Corresponding Author
*(M.T.B.) E-mail: matthew.burger@novartis.com.
■
Notes
The authors declare no competing financial interest.
1196
dx.doi.org/10.1021/ml400307j | ACS Med. Chem. Lett. 2013, 4, 1193−1197

ACS Medicinal Chemistry Letters
Letter
Structural basis of constitutive activity and a unique nucleotide binding
mode of human Pim-1 kinase. J. Biol. Chem. 2005, 280, 6130−6137.
(12) Lu, J.; Zavorotinskaya, T.; Dai, Y.; Niu, X.; Castillo, J.; Sim, J.;
Yu, J.; Wang, Y.; Langowski, J. L.; Holash, J.; Shannon, K.; Garcia, P.
D. Pim2 is required for maintaining multiple myeloma cell growth
through modulating TSC2 phosphorylation. Blood 2013, 122, 1610−
1620.
(13) Structure of 1a in PIM1 submitted under PDB accession code
4N6Y.
(14) The hinge binding potential of 3f was confirmed by the X-ray
cocrystal in Pim1, submitted under PDB accession code 4N6Z.
(15) PDGF(0.009 μM), PKCε(0.014 μM), PKCδ(0.032 μM), and
PKA (0.15 μM).
ACKNOWLEDGMENTS
■
We thank Weiping Jia, Dahzi Tang, and Gavin Dollinger for
analytical chemistry support.
ABBREVIATIONS
■
AML, acute myeloid luekemia; ATP, adenosine triphospate;
[ATP], ATP concentration; BAD, pro-apoptotic protein of the
BCl2 family; cLogP, log of the calculated octanol−water
partition coefficient of a compound; CL, clearance; FQ, fit
quality = LE/LE_Scale; JAK/STAT, Janus Kinase and Signal
Transducer and Activator of Transcription; Km, Michaelis
constant; LE, ligand efficiency = ΔG/number of heavy atoms;
LLE, ligand lipophilicity efficiency = pKi-clogP; logD7.4, log of
the octanol−water distribution coefficient at pH 7.4 of ionized
and unionized compound; PD, pharmacodynamics; PIM,
proviral insertion of Moloney virus; PK, pharmacokinetics;
PSA, polar surface area; S6RP, ribosomal protein S6; S(35),
number of nonmutant kinases tested with >/65% inhibition/
number nonmutant kinases tested
(16) PDGF(1.7 μM), PKCε(3.9 μM), PKCδ(14 μM), and PKA (>25
μM).
(17) With compound potency at the limit of detection, assay format
changed to high [ATP]. See Supporting Information.
(18) PDGF(0.30 μM), PKCε(9.3 μM), and PKA (8.9 μM).
(19) Structure of 5c in PIM1 submitted under PDB accession code
4N70.
(20) Pim1, 2, and 3 exhibit 400, 4, and 40 μM ATP Kms. On the basis
of Cheng−Prusoff equation, a ca. 12.5-, 1250-, or 125-fold shift,
respectively (assuming 5 mM cellular ATP concentration), is to be
expected from enzymatic Kis to target modulation EC50s.
(21) Mayumi, M. EoL-1, a human eosinophlic cell line. Leuk.
Lymphoma 1992, 7, 243−250.
REFERENCES
■
(1) Brault, L.; Gasser, C.; Bracher, F.; Huber, K.; Knapp, S.;
Schwaller, J. PIM serine/threonine kinases in the pathogenesis and
therapy of hematologic malignancies and solid cancers. Haematologica
2010, 95, 1004−1015.
(22) Mouse and rat plasma protein binding of 5c are 98% and 97%,
respectively.
(2) Nawijn, M. C.; Alendar, A.; Berns, A. For better or for worse: the
role of Pim oncogenes in tumorigenesis. Nat. Rev. Cancer 2011, 11,
23−34.
(3) Blanco-Aparicio, C.; Carnero, A. Pim kinases in cancer:
diagnostic, prognostic and treatment opportunities. Biochem. Pharma-
col. 2013, 85, 629−643.
(4) Jackson, L. J.; Pheneger, J. A.; Pheneger, T. J.; Davis, G.; Wright,
A. D.; Robinson, J. E.; Allen, S.; Munson, M. C.; Carter, L. L. The role
of PIM kinases in human and mouse CD4+ T cell activation and
inflammatory bowel disease. Cell. Immunol. 2012, 272, 200−213.
(5) Morwick, T. Pim kinase inhibitors: a survey of the patent
literature. Expert Opin. Ther. Pat. 2010, 20 (2), 193−212 and
references therein.
(6) Mumenthaler, S. M.; Ng, P. Y.; Hodge, A.; Bearss, D.; Berk, G.;
Kanekal, S.; Redkar, S.; Taverna, P.; Agus, D. B.; Jain, A.
Pharmacologic inhibition of Pim kinases alters prostate cancer cell
growth and resensitizes chemoresistant cells to taxanes. Mol. Cancer
Ther. 2009, 8, 2882−2893.
(7) Dakin, L. A.; Block, M. H.; Chen, H.; Code, E.; Dowling, J. E.;
Feng, X.; Ferguson, A. D.; Green, I.; Hird, A. W.; Howard, T.
Discovery of novel benzylidene-1,3-thiazolidine-2,4-diones as potent
and selective inhibitors of the PIM-1, PIM-2, and PIM-3 protein
kinases. Bioorg. Med. Chem. Lett. 2012, 22, 4599−4604.
(8) Haddach, M.; Michaux, J.; Schwaebe, M. K.; Pierre, F.; O’Brien,
S. E.; Borsan, C.; Tran, J.; Raffaele, N.; Ravula, S.; Drygin, D.
Discovery of CX-6258. A potent, selective, and orally efficacious pan-
PIM kinase inhibitor. ACS Med. Chem. Lett. 2012, 3, 135−139.
(9) Nishiguchi, G. A.; Atallah, G.; Bellamacina, C.; Burger, M. T.;
Ding, Y.; Feucht, P. H.; Garcia, P. D.; Han, W.; Klivansky, L.; Lindval,
M. Discovery of novel 3,5-disubstituted indole derivatives as potent
inhibitors of Pim-1, Pim-2, and Pim-3 protein kinases. Bioorg. Med.
Chem. Lett. 2011, 21, 6366−6369.
(10) Wang, X.; Magnuson, S.; Pastor, R.; Fan, E.; Hu, H.; Tsui, V.;
Deng, W.; Murray, J.; Steffek, M.; Wallwever, H.; Moffat, J.;
Drummond, J.; Chan, G.; Harstad, E.; Eben, A. J. Discovery of
novel pyrazolo[1,5-a]pyrimidines as potent pan-Pim inhibitors by
structure- and property-based drug design. Bioorg. Med. Chem. Lett.
2013, 23, 3149−3153.
(11) Qian, K. C.; Wang, L.; Hickey, E. R.; Studts, J.; Barringer, K.;
Peng, C.; Kronkaitis, A.; Li, J.; White, A.; Mische, S.; Farmer, B.
1197
dx.doi.org/10.1021/ml400307j | ACS Med. Chem. Lett. 2013, 4, 1193−1197