Discovery of a Pyrimidothiazolodiazepinone as a Potent and Selective Focal Adhesion Kinase (FAK) Inhibitor

Article abstract of DOI:10.1021/acsmedchemlett.0c00338

Focal adhesion kinase (FAK) is a tyrosine kinase with prominent roles in protein scaffolding, migration, angiogenesis, and anchorage-independent cell survival and is an attractive target for the development of cancer therapeutics. However, current FAK inhibitors display dual kinase inhibition and/or significant activity on several kinases. Although multitargeted activity is at times therapeutically advantageous, such behavior can also lead to toxicity and confound chemical-biology studies. We report a novel series of small molecules based on a tricyclic pyrimidothiazolodiazepinone core that displays both high potency and selectivity for FAK. Structure-activity relationship (SAR) studies explored modifications to the thiazole, diazepinone, and aniline "tail,"which identified lead compound BJG-03-025. BJG-03-025 displays potent biochemical FAK inhibition (IC50 = 20 nM), excellent kinome selectivity, activity in 3D-culture breast and gastric cancer models, and favorable pharmacokinetic properties in mice. BJG-03-025 is a valuable chemical probe for evaluation of FAK-dependent biology.

Full text of DOI:10.1021/acsmedchemlett.0c00338

pubs.acs.org/acsmedchemlett
Letter
Discovery of a Pyrimidothiazolodiazepinone as a Potent and
Selective Focal Adhesion Kinase (FAK) Inhibitor
Brian J. Groendyke,# Behnam Nabet,# Mikaela L. Mohardt, Haisheng Zhang, Ke Peng, Eriko Koide,
Calvin R. Coffey, Jianwei Che, David A. Scott, Adam J. Bass, and Nathanael S. Gray*
Cite This: https://dx.doi.org/10.1021/acsmedchemlett.0c00338
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Focal adhesion kinase (FAK) is a tyrosine kinase
with prominent roles in protein scaffolding, migration, angio-
genesis, and anchorage-independent cell survival and is an
attractive target for the development of cancer therapeutics.
However, current FAK inhibitors display dual kinase inhibition
and/or significant activity on several kinases. Although multi-
targeted activity is at times therapeutically advantageous, such
behavior can also lead to toxicity and confound chemical-biology
studies. We report a novel series of small molecules based on a
tricyclic pyrimidothiazolodiazepinone core that displays both high
potency and selectivity for FAK. Structure−activity relationship
(SAR) studies explored modifications to the thiazole, diazepinone, and aniline “tail,” which identified lead compound BJG-03-025.
BJG-03-025 displays potent biochemical FAK inhibition (IC50 = 20 nM), excellent kinome selectivity, activity in 3D-culture breast
and gastric cancer models, and favorable pharmacokinetic properties in mice. BJG-03-025 is a valuable chemical probe for evaluation
of FAK-dependent biology.
KEYWORDS: FAK, selective inhibitors, structure−activity relationships, aminothiazole, anticancer agents
biochemical IC50 values in the single-digit nanomolar range
ocal adhesion kinase (FAK, also known as protein tyrosine
Fkinase 2 [PTK2]) is a tyrosine kinase with important roles but display either dual kinase inhibition or significant activity
against other kinases. Several bicyclic scaffolds, namely, 7-
azaindoles,11 pyrrolopyrimidines,12 and imidazotriazines,13
have delivered potent (IC50 < 50 nM) FAK inhibitors, but
the full kinome profiles of these series are not reported.
Although off-target activity is at times therapeutically advanta-
geous,14,15 off-target effects from nonselective probes can
confound chemical-biology studies. In this study, we describe
the identification and characterization of a highly selective FAK
inhibitor, BJG-03-025. BJG-03-025 is based on a privileged
benzopyrimidodiazepinone core scaffold which we have
elaborated previously to target kinases including: ERK5,16
PI3Kδ,17 TNK2,18 and nonkinases such as BRD4.19 BJG-03-
025 was identified following a systematic structure−activity
relationship campaign after observing serendipitous selectivity
for this scaffold class from kinome-wide profiling studies. We
observe that BJG-03-025 displays antiproliferative activity in
three-dimensional (3D)-breast and gastric cancer models and
in reorganizing actin and focal adhesions, coordinating PI3K-
AKT and integrin signaling networks, and regulating nuclear
gene expression programs. In the context of cancer, FAK plays
prominent kinase-dependent and kinase-independent roles in
coordinating migration, angiogenesis, and anchorage-inde-
pendent cell survival.1 FAK is commonly overexpressed or
amplified in many cancers, including ovarian, breast, colorectal,
and pancreatic cancers. We recently demonstrated that FAK
coordinates YAP/TAZ, PI3K-AKT, and β-catenin signaling in
diffuse gastric cancer.2 For these reasons, targeted disruption of
FAK has been of great interest and a number of potent FAK
inhibitors have been developed and advanced to clinical trials.
While FAK inhibition has been tolerable and leads to cytostatic
effects, single-agent clinical efficacy of FAK inhibitors is
limited.3 Combination approaches with FAK inhibitors are
under active investigation. For example, FAK inhibitors have
been shown to sensitize pancreatic cancer to chemotherapy or
checkpoint inhibitors,4 leading to several ongoing clinical
trials.5
Received: June 18, 2020
Accepted: November 18, 2020
Many of the existing FAK inhibitors, such as defactinib (VS-
6063),6 VS-4718 (PND-1186),7 PF-573228,8 CEP-37440,9
and TAE226,10 share a 2,4-diaminopyrimidine or 2,4-
diaminopyridine motif (Figure 1). These compounds are
highly potent ATP-competitive kinase inhibitors, with
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
© XXXX American Chemical Society
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX
A

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Figure 1. Existing FAK inhibitors are potent but display multikinase activity.
Figure 2. Identification of 1 (BJG-01-181), a highly selective FAK inhibitor and chemical structures of related analogues.
impacts cellular signaling and migration. Our selective FAK
inhibitor is a valuable probe for evaluation of FAK-specific
biology.
must be in the “top” position, adjacent to the diazepinone
carbonyl (Figure 2). We surmised that this nitrogen may
provide an important hydrogen-bonding interaction. The
closely related pyrazole 3 displayed similar FAK potency
(IC50 = 132 nM), lending further support to this hypothesis.
For other members of the pyrimidodiazepinone series, amide
N-methylation often leads to increased potency due to a
beneficial interaction with the gatekeeper residue.20,21
Furthermore, the N-methylated compounds generally possess
better pharmacokinetic (PK) properties than the secondary
amides.18 Compound 4 (BJG-03-025), an N-methyl matched-
pair with 1, was synthesized and showed a 3-fold improvement,
Previous work in our group has shown the tricyclic
benzopyrimidodiazepinone core to be a privileged scaffold
for the design of potent and selective kinase inhibitors. During
a campaign to replace the benzene of the benzopyrimidodia-
zepinone core with heteroaryl rings, KINOMEscan profiling
revealed 1 (BJG-01-181) as an exceptionally selective FAK
inhibitor, displaying >65% inhibition of only 5 out of 402
kinases at a testing concentration of 10 μM (Figures 2 and S1,
Table S1). A biochemical kinase assay confirmed 1 as a
moderately potent FAK inhibitor (IC50 = 62.2 nM); although
less potent than other FAK inhibitors, 1 is notably more
selective (Table S2). From this starting point, we explored the
structure−activity relationships (SAR) of this novel scaffold to
improve FAK potency and better understand the source of its
kinome selectivity.
with a FAK IC50 of 20.2
1.9 nM. Further optimization
focused on the N-methylated tricyclic core. SAR studies to
investigate the effects of different aniline “tails” and
substitution at the tricyclic core were conducted in parallel
(Table 1). To enable direct comparisons, aniline screening was
conducted using the 2-methyl thiazole core 10. Similarly, to
interrogate the effects of substitution patterns on the tricyclic
A comparison of 1 with a structurally similar thiazole isomer
(2, FAK IC50 > 10 000 nM) revealed that the thiazole nitrogen
B
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Table 1. Structure−Activity Relationships of the Aniline Component
a
b
IC50 measured using the Z’LYTE assay (ThermoFisher Scientific), as duplicate experiments. IC50 of viability of MDA-MB-231 cells in ultra-low
adherent 3D-spheroid suspensions. Values are derived from data presented in Figures 5B and S8 and are representative of n = 3 independent
c
experiments. N.D., not determined. n = 4, two independent batches each tested in duplicate.
core, the 2-methoxy-4-(4-(4-methylpiperazinyl)piperidinyl)
aniline was used in all examples.
cyclization and subsequent methylation with iodomethane to
provide the tricyclic core 10. The final compounds were
prepared by Buchwald−Hartwig amination with the corre-
sponding aniline or heterocyclic amine.
Tricyclic core 10 can be obtained via the synthetic scheme
shown in Scheme 1. Other analogues were prepared using the
same synthetic sequence, or with slight modifications (see the
Supporting Information (SI) for details). Commercially
available ethyl 2-methylthiazole-4-carboxylate 5 was bromi-
nated with N-bromosuccinimide (NBS) to afford bromothia-
zole 6. Palladium-catalyzed amination with benzophenone
imine, followed by hydrolysis provided primary amine 7 in
good yield. Deprotonation of the primary amine and treatment
with iodomethane provided N-methyl 8. All attempts to forge
the thiazole C5−N bond via nitration of 5 or SNAr reaction of
6 were unsuccessful. N-Arylation of 8 with 2,4-dichloro-5-
nitropyrimidine under NaH conditions provided intermediate
9, which was subjected to an iron-promoted reductive
Modification of the solvent-exposed side chain aimed to
improve the potency of 4 while maintaining high kinome
selectivity. We targeted several key residues in the solvent-
exposed channel, including Arg426, Gln438, and Glu506
(Figure 3), by incorporating hydrogen-bond donors or
acceptors into the aniline tail. A variety of modifications
yielded inhibitors with biochemical IC50s in the 30−60 nM
range (Table 1), but did not improve upon the potency of 4. In
contrast, several side chains, such as hydroxyethyl ether 19,
tetrahydroquinolinone 22, and pyrazole 27, were not tolerated
and resulted in lower biochemical potency.
C
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
a
Scheme 1. Synthesis of 4 and 11−28
a
Reagents and conditions: (a) NBS, acetonitrile, 40 °C, 61%. (b) (i) Pd2 (dba)3, BINAP, Cs2CO3, benzophenone imine, toluene, 80 °C, (ii) THF/
1 M aq HCl (4:1 mixture), rt, 68% over two steps. (c) NaH, MeI, THF, 0 °C to rt, 41%. (d) NaH, 2,4-dichloro-5-nitropyrimidine, THF, 0 °C to rt,
59%. (e) Fe, glacial acetic acid, 50 °C. (f) NaH, MeI, DMF, 0 °C to rt, 48% over two steps. (g) Pd2 (dba)3, XPhos, K2CO3, amine, t-BuOH 100 °C.
Table 2. Structure−Activity Relationships of the Tricyclic
Core
a
b
compd
R1
R2
CH3
R3
FAK IC50 (μM)
cell IC50 (μM)
c
4
CH3
H
Et
i-Pr
Ph
CF3
CH3
CH3
CH3
CH3
CH3
CH3
H
0.020 0.002
0.510 0.035
0.032 0.001
0.059 0.001
>10
0.134 0.001
0.019 0.001
0.023 0.001
0.068 0.002
0.047 0.001
0.525 0.016
3.6
Figure 3. Docking of 4 with FAK (PDB 6I8Z). (A) Key interactions
of 4 with active-site residues, notably the critical hydrogen-bond of
the thiazole nitrogen with Asp564. (B) FAK surface map shows the
U-shaped active site, with the 2-methyl group extending into a
hydrophobic pocket.
29
30
31
32
33
34
35
36
37
CH3
CH3
CH3
CH3
CH3
CH3
Et
CH2CF3
n-Pr
CH3
ND
ND
ND
ND
ND
0.93
>10
ND
ND
>10
CH3
CH3
CH3
CH3
Et
CH3
CH3
CH3
CH3
Our observation that the opposite thiazole isomer (i.e., 2)
was inactive suggested that the steric and electronic properties
of the tricyclic core had a significant effect on the ability to
bind FAK. We turned to molecular modeling to gain further
insight into the binding mode of 4 and how its potency could
be improved (Figure 3). Docking of 4 with FAK (PDB: 6I8Z)
revealed two hydrogen-bonding interactions: binding to hinge
Cys502 via the aminopyrimidine and binding to the side-chain
of Asp564, part of the DFG motif, via the diazepinone carbonyl
and thiazole nitrogen. The loss of potency observed for
thiazole isomer 2 is likely due to the loss of this hydrogen-
bonding interaction. Providing further support for this
hypothesis, related benzopyrimidodiazepinones such as
XMD8-87 (Figure 2) do not engage FAK.18 The N-methyl
amide extends toward the top of the FAK cavity, potentially
interacting with Ala452 (Figure 3A); however, it was uncertain
whether further modifications would be tolerated. The thiazole
2-methyl extends into the back pocket of the FAK active site,
and we hypothesized that longer substituents at this position
might benefit from hydrophobic interactions (Figure 3B).
Similarly, we proposed that longer substituents on the
“bottom” nitrogen of the central ring could extend out of the
back end of the U-shaped binding site, potentially gaining
beneficial hydrophobic contacts (Figure 3B). Docking analyses
of 3, 20, and 26 predicted the same binding mode as 4 (Figure
S6).
d
38
a
IC50 measured using the Z’LYTE assay (ThermoFisher Scientific), as
b
duplicate experiments. IC50 of viability of MDA-MB-231 cells in
ultra-low adherent 3D-spheroid suspensions. Values are derived from
data presented in Figures 5B and S8 and are representative of n = 3
independent experiments. ND, not determined. n = 4, two
independent batches each tested in duplicate. Oxazole instead of
c
d
thiazole.
potency roughly 8-fold compared to 1 (IC50 510 vs 62 nM).
However, larger substituents at the thiazole 2-position were
met with mixed success: ethyl- and isopropyl-substituted
analogues 30 and 31 retained activity against FAK, but a
phenyl group was not tolerated (32). Whereas N-ethylation at
either position of the central ring (34, 35) retained nearly
identical potency to 4, other substitutions of the bottom
nitrogen led to a slight loss in potency (36, 37). Introduction
of an electron-withdrawing group (2-CF3 33, oxazole 38)
sharply decreased potency, further highlighting the importance
of the hydrogen bond between the thiazole and Asp564.
Interestingly, while 38 is predicted to bind in the same manner
as 4, docking of 37 revealed a puckered conformation of the
central ring to accommodate the longer N-propyl group
(Figure S6).
A summary of SAR studies on the tricyclic core is presented
in Table 2. Removal of the thiazole 2-methyl (29) decreased
D
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
With 4 selected as the lead compound, its kinome selectivity
was profiled by KINOMEscan screening to confirm that amide
N-methylation does not ablate the selectivity initially observed
with 1 (Figure 4, Table S1). Although a few additional kinase
including breast, colorectal, and gastric cancer models (Cancer
Dependency Map)23 (Figure S7). We selected MDA-MB-231
cells as a versatile cell line to explore FAK-dependent
phenotypes due to their high genetic dependence and
overexpression of PTK2 (Figure S7D). Prior studies also
indicate that these MDA-MB-231 cells display FAK-dependent
migration and invasion potential.24,25 For high-throughput
evaluation of the biological effects of our compound series, we
compared antiproliferative effects of defactinib, VS-4718, and
PF-573228 to 17 representative compounds from our series
upon treatment of MDA-MB-231 cells in two-dimensional
(2D)-monolayer cultures and ultra-low adherent (ULA) three-
dimensional (3D)-spheroid suspensions. We and others have
also previously shown that 3D-spheroids serve as better model
systems to evaluate drug sensitivity and in vivo tumor
physiology.20,26−28 In addition, published FAK agents have
limited activity in 2D-monolayer cultures and studies
demonstrate improved activity in 3D-culture and in vivo
models potentially due to FAK’s kinase-dependent and kinase-
independent roles in integrin signaling, adhesion and protein
scaffolding.7,29,30
Across the compound series, we observed that MDA-MB-
231 cells were more sensitive to FAK inhibition in 3D-
spheroids, consistent with prior studies (Figures 5A,B and S8;
Tables 1 and 2). VS-4718, defactinib, and PF-573228, which
display the most pronounced biochemical potency on FAK,
also had the most pronounced antiproliferative effects likely
due to a combination of direct FAK inhibitory activity
combined with potential multikinase activities. Of our series,
4, 11, 12, 17, and 34 displayed the most pronounced
antiproliferative effects in 3D-spheroids, with little to no
effects in 2D-monolayer cultures.
In general, the SAR trends observed for biochemical potency
of our series largely translate to activity in 3D-spheroids. For
example, 24 and 28 are ∼4-fold less potent than 4 and are >3-
fold less potent in the 3D assay. Likewise, 22 shows only
modest biochemical potency and does not inhibit proliferation.
One exception is the matched pair 20 and 21, where the 2-
methoxy 21 is more potent biochemically but demonstrates
weaker antiproliferation. This weaker phenotype may stem
from a reduced off-target profile that is frequently observed for
2-methoxyanilines.18 Notably, 34, 4, and 35 elicit different
biological activity despite possessing nearly identical bio-
chemical potencies. These three compounds have similar
kinetic and thermodynamic solubilities, show moderate
permeability in a Caco-2 assay, and are potential substrates
of P-glycoprotein (P-gp; Tables S8−S10). The improved
antiproliferative effect of 34 relative to 4 may stem from the
better permeability (Papp,A→B 83 vs 58 nm/s) and lower efflux
ratio (2.2 vs 4.1, respectively) of 34. At present, it remains
unclear as to why several compounds, such as 14, 16, 23, and
35 display little activity in 3D-spheroids despite being potent
biochemical inhibitors.
Figure 4. KINOMEscan profile of 4.
targets are registered at a 10 μM screening concentration
(Figure S2, Table S1), 4 displays “complete” selectivity for
FAK at a screening concentration of 1 μM, with an S(35) score
of 0.01. In comparison, at an identical screening concentration
the S(35) score for defactinib is 0.24.22 Biochemical IC50s were
obtained for all nonmutant kinases registering <35% control in
the 10 μM KINOMEscan assay (Figure S2), demonstrating
that 4 is approximately 100−300-fold selective for FAK. PLK1
proved to be a false positive, with an enzyme IC50 of 8.3 μM,
and no cell-cycle disruption was observed during in vitro
profiling.
Seven compounds from the 2-methylthiazole series were
evaluated in a mouse-liver microsome stability assay (Table
S6). All compounds demonstrated good to excellent micro-
some stability, with half-lives ranging from 25.5 to >120 min.
Lead compound 4 was then submitted for further pharmaco-
kinetic (PK) profiling in mice, with IV (2 mg/kg) and PO (10
mg/kg) doses administered. As shown in Table 3, 4 exhibits
Table 3. Mouse Pharmacokinetic Parameters for 4
dose
parameter
4
2 mg/kg IV
CL (mL/min/kg)
t1/2 (h)
18.5
5.29
4.07
0.42
4.00
1.82
Vss (L/kg)
Cmax (μM)
Tmax (h)
10 mg/kg PO
AUCinf (h·μg/mL)
F (%)
18.3
We further evaluated the effects of 4 on the ability of MDA-
MB-231 cells to migrate in response to chemoattraction
(serum-containing media). Treatment with 4 led to a
statistically significant albeit modest effect on migration of
MDA-MB-231 cells. These effects were more pronounced
upon treatment with defactinib and nocodazole, an agent that
disrupts microtubule polymerization31 (Figures 5C and S9).
Together, these data indicate that selective FAK inhibition
with 4 leads to modest loss of viability in 3D-suspensions and
migratory potential of MDA-MB-231 cells.
low clearance (18.5 mL/min/kg) and is metabolically stable,
displaying an average t1/2 of 5.29 h under IV administration.
For both routes, detectable levels of drug were observed 24 h
after dosing (Tables S7, S8; Figures S4, S5). However, oral
bioavailability is limited (%F = 18).
We next aimed to explore the biological activity of our FAK
inhibitor series. PTK2, the gene that encodes FAK, is a
significant genetic dependency in pan-cancer cell line analysis
E
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Figure 5. FAK inhibition decreases proliferation in 3D-spheroid suspensions and migration of breast cancer cells. (A) Scatterplot depicting
biochemical IC50 compared to cellular IC50 upon treatment of MDA-MB-231 cells cultured as 2D-adherent monolayers (left plot) or ultra-low
adherent 3D-spheroid suspensions (right plot). VS-4718 (VS), defactinib (Def), and PF-573228 (PF) are noted in gray, 4 is noted in red, and the
pyrimidothiazolodiazepinone series is noted in black. (B) DMSO-normalized antiproliferation of MDA-MB-231 cells treated with the indicated
compounds for 120 h. Cells were cultured as 2D-adherent monolayers (left plot) or ultra-low adherent 3D-spheroid suspensions (right plot). Data
are presented as mean SD of n = 4 biologically independent samples and are representative of n = 3 independent experiments. (C) DMSO-
normalized migration of MDA-MB-231 cells after 24 h under the indicated conditions (nocodazole, Noco). Representative images are provided in
Figure S9. Data are presented as mean SD of n = 3 biologically independent samples and are representative of n = 3 independent experiments. P
values were derived from a two-tailed Student’s t test compared to DMSO control are noted (*P < 0.05, **P < 0.01).
Figure 6. FAK inhibition decreases aberrant signaling and proliferation of gastric cancer organoids. (A) Immunoblot analysis of CDH1−/−
RHOAY42C/+ gastric organoids treated with DMSO, 4, or defactinib at the indicated concentrations for 24 h. Data are representative of n = 3
independent experiments. (B) Relative proliferation of CDH1−/− RHOAY42C/+ gastric organoids treated with DMSO, 4, or defactinib (Def) at the
indicated concentrations for the indicated time-course. Data are presented as mean SD of n = 7−10 biologically independent samples and are
representative of n = 3 independent experiments. P values were derived from a two-tailed Student’s t test compared to DMSO control are noted
(*P < 0.05, ***P < 0.001).
F
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
To corroborate these findings, we evaluated the effects of
FAK inhibition in HT-29 cells, a colorectal adenocarcinoma
cell line with PTK2 copy number alterations and elevated
expression levels (Figures S7D and S10A). Treatment with 4
led to little or no activity in 2D-monolayer or 3D-suspensions,
while more pronounced antiproliferative activity was observed
upon treatment with VS-4718, defactinib, and PF-573228
(Figure S10B). While the modest results achieved with 4 are
consistent with a prior study evaluating FAK inhibition in HT-
29 cells in 2D-monolayer culture and in vivo,32 further testing
of potential drug combination strategies may be necessary to
achieve enhanced viability effects. For example, combination
studies with vemurafenib may be warranted, which has been
shown to synergize with FAK inhibitors in these BRAF mutant
cells.32
To explore the effects of FAK inhibition in an additional
translationally relevant biological context, we assayed the
consequences of FAK inhibition in a gastric cancer CDH1−/−
RHOAY42C/+ organoid model. We recently demonstrated that
FAK, through activation of YAP/TAZ, PI3K/AKT and β-
catenin signaling, coordinates diffuse gastric cancer progres-
sion.2 Treatment of gastric organoids with 4 led to loss of
active, phosphorylated FAK (pFAK Y397), which was more
pronounced with defactinib after 24 h (Figure 6A). Defactinib
treatment resulted in changes in active, phosphorylated AKT
(pAKT S473) and total YAP/TAZ levels, consistent with our
prior findings within 24 h,2 while modest effects on pAKT
S473 and active, nonphosphorylated YAP were observed with
4 after 48 h treatment (Figures 6A and S11). Differences in
pathway responses corresponded with levels of antiproliferative
activity. Treatment with 4 led to dose-dependent activity on
gastric organoid proliferation, which was more pronounced
with defactinib treatment (Figure 6B). Together, these results
confirm the potential of targeted FAK inhibition in diffuse
gastric cancer.
In summary, we describe the identification and character-
ization of a selective FAK inhibitor, 4, and we disclose a novel
tricyclic core bearing a substituted thiazole. We observe that
FAK inhibition leads to loss of viability in breast and gastric
cancer 3D-culture model systems, as well as effects on cellular
signaling and migration. These effects are milder compared to
published FAK inhibitors, in which enhanced effects may be
due to additional activities on other kinases. Despite its milder
phenotype, 4 may be particularly beneficial as a chemical probe
when used in tandem with another kinase inhibitor to study
the signaling effects and drug synergy upon dual inhibition,
where a clean kinome profile is imperative. As 4 possesses
suitable PK properties for studies in mice, further evaluation is
required to examine whether the effects in 3D-culture models
translate to efficacy in vivo. Our chemical probe serves as a
valuable tool to explore the consequences of FAK inhibition
and identify potential combination approaches in cancer.
AUTHOR INFORMATION
Corresponding Author
■
Nathanael S. Gray − Department of Cancer Biology, Dana-
Farber Cancer Institute, Boston, Massachusetts 02115,
United States; Department of Biological Chemistry and
Molecular Pharmacology, Harvard Medical School, Boston,
Massachusetts 02115, United States; orcid.org/0000-
0001-5354-7403; Email: nathanael_gray@
dfci.harvard.edu
Authors
Brian J. Groendyke − Department of Cancer Biology, Dana-
Farber Cancer Institute, Boston, Massachusetts 02115,
United States; Department of Biological Chemistry and
Molecular Pharmacology, Harvard Medical School, Boston,
Massachusetts 02115, United States; orcid.org/0000-
0002-3860-8308
Behnam Nabet − Department of Cancer Biology, Dana-Farber
Cancer Institute, Boston, Massachusetts 02115, United
States; Department of Biological Chemistry and Molecular
Pharmacology, Harvard Medical School, Boston,
Massachusetts 02115, United States; orcid.org/0000-
0002-6738-4200
Mikaela L. Mohardt − Department of Cancer Biology, Dana-
Farber Cancer Institute, Boston, Massachusetts 02115,
United States
Haisheng Zhang − Department of General Surgery, Nanfang
Hospital, Southern Medical University, 510515 Guangzhou,
China; Division of Molecular and Cellular Oncology, Dana-
Farber Cancer Institute, Harvard Medical School, Boston,
Massachusetts 02115, United States
Ke Peng − Division of Molecular and Cellular Oncology,
Dana-Farber Cancer Institute, Harvard Medical School,
Boston, Massachusetts 02115, United States
Eriko Koide − Department of Cancer Biology, Dana-Farber
Cancer Institute, Boston, Massachusetts 02115, United States
Calvin R. Coffey − Department of Cancer Biology, Dana-
Farber Cancer Institute, Boston, Massachusetts 02115,
United States
Jianwei Che − Department of Cancer Biology, Dana-Farber
Cancer Institute, Boston, Massachusetts 02115, United
States; Department of Biological Chemistry and Molecular
Pharmacology, Harvard Medical School, Boston,
Massachusetts 02115, United States
David A. Scott − Department of Cancer Biology, Dana-Farber
Cancer Institute, Boston, Massachusetts 02115, United
States; Department of Biological Chemistry and Molecular
Pharmacology, Harvard Medical School, Boston,
Massachusetts 02115, United States; orcid.org/0000-
0003-3243-528X
Adam J. Bass − Division of Molecular and Cellular Oncology,
Dana-Farber Cancer Institute, Harvard Medical School,
Boston, Massachusetts 02115, United States; Broad Institute
of MIT and Harvard, Cambridge, Massachusetts 02142,
United States
Complete contact information is available at:
https://pubs.acs.org/10.1021/acsmedchemlett.0c00338
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acsmedchemlett.0c00338.
Author Contributions
#B.J.G. and B.N. contributed equally to this work. B.J.G., B.N.,
and N.S.G. conceived of the study. B.J.G. led the medicinal
chemistry study and performed compound synthesis and
optimization. B.N. led and supervised the biological studies.
Details of biochemical and biological assays, synthetic
procedures, and analytical data (PDF)
G
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
Kinase (FAK) Inhibitor in Japanese Patients with Advanced Solid
Tumors. Cancer Chemother. Pharmacol. 2016, 77 (5), 997−1003.
(7) Tanjoni, I.; Walsh, C.; Uryu, S.; Tomar, A.; Nam, J. O.; Mielgo,
A.; Lim, S. T.; Liang, C.; Koenig, M.; Patel, N.; Kwok, C.; McMahon,
G.; Stupack, D. G.; Schlaepfer, D. D. PND-1186 FAK Inhibitor
Selectively Promotes Tumor Cell Apoptosis in Three-Dimensional
Environments. Cancer Biol. Ther. 2010, 9 (10), 764−77.
(8) Slack-Davis, J. K.; Martin, K. H.; Tilghman, R. W.; Iwanicki, M.;
Ung, E. J.; Autry, C.; Luzzio, M. J.; Cooper, B.; Kath, J. C.; Roberts,
W. G.; Parsons, J. T. Cellular Characterization of a Novel Focal
Adhesion Kinase Inhibitor. J. Biol. Chem. 2007, 282 (20), 14845−52.
(9) Ott, G. R.; Cheng, M.; Learn, K. S.; Wagner, J.; Gingrich, D. E.;
Lisko, J. G.; Curry, M.; Mesaros, E. F.; Ghose, A. K.; Quail, M. R.;
Wan, W.; Lu, L.; Dobrzanski, P.; Albom, M. S.; Angeles, T. S.; Wells-
Knecht, K.; Huang, Z.; Aimone, L. D.; Bruckheimer, E.; Anderson, N.;
Friedman, J.; Fernandez, S. V.; Ator, M. A.; Ruggeri, B. A.; Dorsey, B.
D. Discovery of Clinical Candidate CEP-37440, a Selective Inhibitor
of Focal Adhesion Kinase (FAK) and Anaplastic Lymphoma Kinase
(ALK). J. Med. Chem. 2016, 59 (16), 7478−96.
B.N., M.L.M, H.Z., K.P., and E.K. performed biological assays.
C.R.C. contributed to compound synthesis. J.C. performed
docking modeling. D.A.S., A.J.B. and N.S.G. supervised the
study. B.J.G. and B.N. wrote the manuscript, with edits from
N.S.G. and D.A.S. All authors read and approved the
manuscript.
Notes
The authors declare the following competing financial
interest(s): FAK inhibitors noted in this study are subject to
patent applications filed by Dana-Farber Cancer Institute.
N.S.G. is a Scientific Founder, member of the SAB and equity
holder in C4 Therapeutics, Syros, Soltego (board member),
B2S, Allorion, Gatekeeper and Petra Pharmaceuticals. J.C. is a
consultant for Soltego and equity holder of M3 bioinformatics
& technology Inc. The Gray lab receives or has received
research funding from Novartis, Takeda, Astellas, Taiho,
Janssen, Kinogen, Voroni, Arbella, Deerfield and Sanofi. A.J.B
receives funding from Bayer, Merck and Novartis and is a co-
founder of Signet Therapeutics.
(10) Shi, Q.; Hjelmeland, A. B.; Keir, S. T.; Song, L.; Wickman, S.;
Jackson, D.; Ohmori, O.; Bigner, D. D.; Friedman, H. S.; Rich, J. N. A
Novel Low-Molecular Weight Inhibitor of Focal Adhesion Kinase,
TAE226, Inhibits Glioma Growth. Mol. Carcinog. 2007, 46 (6), 488−
96.
ACKNOWLEDGMENTS
■
We thank members of the Gray laboratory and S. Nabet for
helpful discussions. This work was supported by an American
Cancer Society Postdoctoral Fellowship PF-17-010-01-CDD
(B.N.), Claudia Adams Barr Program in Innovative Basic
Cancer Research Award (B.N.), Katherine L. and Steven C.
Pinard Research Fund (N.S.G. and B.N.), V Foundation
Translational Award (H.Z., K.P., and A.J.B.), NIH grant
R01CA224428 (H.Z., K.P., and A.J.B.).
(11) Heinrich, T.; Seenisamy, J.; Emmanuvel, L.; Kulkarni, S. S.;
̈
Bomke, J.; Rohdich, F.; Greiner, H.; Esdar, C.; Krier, M.; Gradler, U.;
Musil, D. Fragment-Based Discovery of New Highly Substituted 1H-
Pyrrolo[2,3-b]- and 3H-Imidazolo[4,5-b]-Pyridines as Focal Adhe-
sion Kinase Inhibitors. J. Med. Chem. 2013, 56 (3), 1160−1170.
(12) Choi, H.-S.; Wang, Z.; Richmond, W.; He, X.; Yang, K.; Jiang,
T.; Karanewsky, D.; Gu, X.-j.; Zhou, V.; Liu, Y.; Che, J.; Lee, C. C.;
Caldwell, J.; Kanazawa, T.; Umemura, I.; Matsuura, N.; Ohmori, O.;
Honda, T.; Gray, N.; He, Y. Design and synthesis of 7H-pyrrolo[2,3-
d]pyrimidines as focal adhesion kinase inhibitors. Part 2. Bioorg. Med.
Chem. Lett. 2006, 16 (10), 2689−2692.
ABBREVIATIONS
■
AUCinf, area under the curve extrapolated to infinity; CL,
clearance; Cmax, maximum concentration; F, bioavailability; IV,
intravenous; PO, per os; t1/2, half-life; Tmax, time at Cmax; Vss,
volume of distribution at steady state.
(13) Dao, P.; Smith, N.; Tomkiewicz-Raulet, C.; Yen-Pon, E.;
Camacho-Artacho, M.; Lietha, D.; Herbeuval, J.-P.; Coumoul, X.;
Garbay, C.; Chen, H. Design, Synthesis, and Evaluation of Novel
Imidazo[1,2-a][1,3,5]triazines and Their Derivatives as Focal
Adhesion Kinase Inhibitors with Antitumor Activity. J. Med. Chem.
2015, 58 (1), 237−251.
(14) Ferguson, F. M.; Gray, N. S. Kinase Inhibitors: the Road
Ahead. Nat. Rev. Drug Discovery 2018, 17 (5), 353−377.
(15) Lin, A.; Giuliano, C. J.; Palladino, A.; John, K. M.; Abramowicz,
C.; Yuan, M. L.; Sausville, E. L.; Lukow, D. A.; Liu, L.; Chait, A. R.;
Galluzzo, Z. C.; Tucker, C.; Sheltzer, J. M. Off-Target Toxicity is a
Common Mechanism of Action of Cancer Drugs Undergoing Clinical
Trials. Sci. Transl. Med. 2019, 11 (509), eaaw8412.
(16) Deng, X.; Yang, Q.; Kwiatkowski, N.; Sim, T.; McDermott, U.;
Settleman, J. E.; Lee, J. D.; Gray, N. S. Discovery of a Benzo[e]-
pyrimido-[5,4-b][1,4]diazepin-6(11H)-one as a Potent and Selective
Inhibitor of Big MAP Kinase 1. ACS Med. Chem. Lett. 2011, 2 (3),
195−200.
(17) Ferguson, F. M.; Ni, J.; Zhang, T.; Tesar, B.; Sim, T.; Kim, N.
D.; Deng, X.; Brown, J. R.; Zhao, J. J.; Gray, N. S. Discovery of a
Series of 5,11-Dihydro-6H-benzo[e]pyrimido[5,4-b][1,4]diazepin-6-
ones as Selective PI3K-delta/gamma Inhibitors. ACS Med. Chem. Lett.
2016, 7 (10), 908−912.
(18) Groendyke, B. J.; Powell, C. E.; Feru, F.; Gero, T. W.; Li, Z.;
Szabo, H.; Pang, K.; Feutrill, J.; Chen, B.; Li, B.; Gray, N. S.; Scott, D.
A. Benzopyrimidodiazepinone Inhibitors of TNK2. Bioorg. Med.
Chem. Lett. 2020, 30 (4), 126948.
(19) Wang, J.; Erazo, T.; Ferguson, F. M.; Buckley, D. L.; Gomez,
N.; Munoz-Guardiola, P.; Dieguez-Martinez, N.; Deng, X.; Hao, M.;
Massefski, W.; Fedorov, O.; Offei-Addo, N. K.; Park, P. M.; Dai, L.;
DiBona, A.; Becht, K.; Kim, N. D.; McKeown, M. R.; Roberts, J. M.;
Zhang, J.; Sim, T.; Alessi, D. R.; Bradner, J. E.; Lizcano, J. M.;
Blacklow, S. C.; Qi, J.; Xu, X.; Gray, N. S. Structural and
Atropisomeric Factors Governing the Selectivity of Pyrimido-
REFERENCES
■
(1) Sulzmaier, F. J.; Jean, C.; Schlaepfer, D. D. FAK in Cancer:
Mechanistic Findings and Clinical Applications. Nat. Rev. Cancer
2014, 14 (9), 598−610.
(2) Zhang, H.; Schaefer, A.; Wang, Y.; Hodge, R. G.; Blake, D. R.;
Diehl, J. N.; Papageorge, A. G.; Stachler, M. D.; Liao, J.; Zhou, J.; Wu,
Z.; Akarca, F.; De Klerk, L. K.; Derks, S.; Pierobon, M.; Hoadley, K.
A.; Wang, T. C.; Church, G.; Wong, K. K.; Petricoin, E. F.; Cox, A.
D.; Lowy, D. R.; Der, C. J.; Bass, A. J. Gain-of-Function RHOA
Mutations Promote Focal Adhesion Kinase Activation and Depend-
ency in Diffuse Gastric Cancer. Cancer Discovery 2020, 10 (2), 288−
305.
(3) Mohanty, A.; Pharaon, R. R.; Nam, A.; Salgia, S.; Kulkarni, P.;
Massarelli, E. FAK-Targeted and Combination Therapies for the
Treatment of Cancer: an Overview of Phase I and II Clinical Trials.
Expert Opin. Invest. Drugs 2020, 29 (4), 399−409.
(4) Jiang, H.; Hegde, S.; Knolhoff, B. L.; Zhu, Y.; Herndon, J. M.;
Meyer, M. A.; Nywening, T. M.; Hawkins, W. G.; Shapiro, I. M.;
Weaver, D. T.; Pachter, J. A.; Wang-Gillam, A.; DeNardo, D. G.
Targeting Focal Adhesion Kinase Renders Pancreatic Cancers
Responsive to Checkpoint Immunotherapy. Nat. Med. 2016, 22 (8),
851−60.
(5) Symeonides, S. N.; Anderton, S. M.; Serrels, A. FAK-Inhibition
Opens the Door to Checkpoint Immunotherapy in Pancreatic Cancer.
J. Immunother. Cancer 2017, 5, 17.
(6) Shimizu, T.; Fukuoka, K.; Takeda, M.; Iwasa, T.; Yoshida, T.;
Horobin, J.; Keegan, M.; Vaickus, L.; Chavan, A.; Padval, M.;
Nakagawa, K. A First-in-Asian Phase 1 Study to Evaluate Safety,
Pharmacokinetics and Clinical Activity of VS-6063, a Focal Adhesion
H
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX

ACS Medicinal Chemistry Letters
pubs.acs.org/acsmedchemlett
Letter
benzodiazipinones as Inhibitors of Kinases and Bromodomains. ACS
Chem. Biol. 2018, 13 (9), 2438−2448.
(31) De Brabander, M. J.; Van de Veire, R. M.; Aerts, F. E.; Borgers,
M.; Janssen, P. A. The Effects of Methyl (5-(2-thienylcarbonyl)-1H-
benzimidazol-2-yl) Carbamate, (R 17934; NSC 238159), a New
Synthetic Antitumoral Drug Interfering with Microtubules, on
Mammalian Cells Cultured in vitro. Cancer Res. 1976, 36 (3), 905−
16.
(32) Chen, G.; Gao, C.; Gao, X.; Zhang, D. H.; Kuan, S. F.; Burns,
T. F.; Hu, J. Wnt/beta-Catenin Pathway Activation Mediates
Adaptive Resistance to BRAF Inhibition in Colorectal Cancer. Mol.
Cancer Ther. 2018, 17 (4), 806−813.
(20) Ferguson, F. M.; Nabet, B.; Raghavan, S.; Liu, Y.; Leggett, A. L.;
Kuljanin, M.; Kalekar, R. L.; Yang, A.; He, S.; Wang, J.; Ng, R. W. S.;
Sulahian, R.; Li, L.; Poulin, E. J.; Huang, L.; Koren, J.; Dieguez-
Martinez, N.; Espinosa, S.; Zeng, Z.; Corona, C. R.; Vasta, J. D.; Ohi,
R.; Sim, T.; Kim, N. D.; Harshbarger, W.; Lizcano, J. M.; Robers, M.
B.; Muthaswamy, S.; Lin, C. Y.; Look, A. T.; Haigis, K. M.; Mancias, J.
D.; Wolpin, B. M.; Aguirre, A. J.; Hahn, W. C.; Westover, K. D.; Gray,
N. S. Discovery of a Selective Inhibitor of Doublecortin Like Kinase 1.
Nat. Chem. Biol. 2020, 16, 635−643.
(21) Deng, X.; Elkins, J. M.; Zhang, J.; Yang, Q.; Erazo, T.; Gomez,
N.; Choi, H. G.; Wang, J.; Dzamko, N.; Lee, J. D.; Sim, T.; Kim, N.;
Alessi, D. R.; Lizcano, J. M.; Knapp, S.; Gray, N. S. Structural
Determinants for ERK5 (MAPK7) and Leucine Rich Repeat Kinase 2
Activities of Benzo[e]pyrimido-[5,4-b]diazepine-6(11H)-ones. Eur. J.
Med. Chem. 2013, 70, 758−67.
(22) Cromm, P. M.; Samarasinghe, K. T. G.; Hines, J.; Crews, C. M.
Addressing Kinase-Independent Functions of FAK via PROTAC-
Mediated Degradation. J. Am. Chem. Soc. 2018, 140 (49), 17019−
17026.
(23) Tsherniak, A.; Vazquez, F.; Montgomery, P. G.; Weir, B. A.;
Kryukov, G.; Cowley, G. S.; Gill, S.; Harrington, W. F.; Pantel, S.;
Krill-Burger, J. M.; Meyers, R. M.; Ali, L.; Goodale, A.; Lee, Y.; Jiang,
G.; Hsiao, J.; Gerath, W. F. J.; Howell, S.; Merkel, E.; Ghandi, M.;
Garraway, L. A.; Root, D. E.; Golub, T. R.; Boehm, J. S.; Hahn, W. C.
Defining a Cancer Dependency Map. Cell 2017, 170 (3), 564−576.
(24) Taliaferro-Smith, L.; Oberlick, E.; Liu, T.; McGlothen, T.;
Alcaide, T.; Tobin, R.; Donnelly, S.; Commander, R.; Kline, E.;
Nagaraju, G. P.; Havel, L.; Marcus, A.; Nahta, R.; O’Regan, R. FAK
Activation is Required for IGF1R-Mediated Regulation of EMT,
Migration, and Invasion in Mesenchymal Triple Negative Breast
Cancer Cells. Oncotarget 2015, 6 (7), 4757−72.
(25) Pan, M. R.; Hou, M. F.; Ou-Yang, F.; Wu, C. C.; Chang, S. J.;
Hung, W. C.; Yip, H. K.; Luo, C. W. FAK is Required for Tumor
Metastasis-Related Fluid Microenvironment in Triple-Negative Breast
Cancer. J. Clin. Med. 2019, 8 (1), 38.
(26) Janes, M. R.; Zhang, J.; Li, L. S.; Hansen, R.; Peters, U.; Guo,
X.; Chen, Y.; Babbar, A.; Firdaus, S. J.; Darjania, L.; Feng, J.; Chen, J.
H.; Li, S.; Li, S.; Long, Y. O.; Thach, C.; Liu, Y.; Zarieh, A.; Ely, T.;
Kucharski, J. M.; Kessler, L. V.; Wu, T.; Yu, K.; Wang, Y.; Yao, Y.;
Deng, X.; Zarrinkar, P. P.; Brehmer, D.; Dhanak, D.; Lorenzi, M. V.;
Hu-Lowe, D.; Patricelli, M. P.; Ren, P.; Liu, Y. Targeting KRAS
Mutant Cancers with a Covalent G12C-Specific Inhibitor. Cell 2018,
172 (3), 578−589.e17.
(27) Santana-Codina, N.; Chandhoke, A. S.; Yu, Q.; Malachowska,
B.; Kuljanin, M.; Gikandi, A.; Stanczak, M.; Gableske, S.;
Jedrychowski, M. P.; Scott, D. A.; Aguirre, A. J.; Fendler, W.; Gray,
N. S.; Mancias, J. D. Defining and Targeting Adaptations to
Oncogenic KRAS(G12C) Inhibition Using Quantitative Temporal
Proteomics. Cell Rep. 2020, 30 (13), 4584−4599.
(28) Han, K.; Pierce, S. E.; Li, A.; Spees, K.; Anderson, G. R.;
Seoane, J. A.; Lo, Y. H.; Dubreuil, M.; Olivas, M.; Kamber, R. A.;
Wainberg, M.; Kostyrko, K.; Kelly, M. R.; Yousefi, M.; Simpkins, S.
W.; Yao, D.; Lee, K.; Kuo, C. J.; Jackson, P. K.; Sweet-Cordero, A.;
Kundaje, A.; Gentles, A. J.; Curtis, C.; Winslow, M. M.; Bassik, M. C.
CRISPR Screens in Cancer Spheroids Identify 3D Growth-Specific
Vulnerabilities. Nature 2020, 580 (7801), 136−141.
(29) Howe, G. A.; Xiao, B.; Zhao, H.; Al-Zahrani, K. N.; Hasim, M.
S.; Villeneuve, J.; Sekhon, H. S.; Goss, G. D.; Sabourin, L. A.;
Dimitroulakos, J.; Addison, C. L. Focal Adhesion Kinase Inhibitors in
Combination with Erlotinib Demonstrate Enhanced Anti-Tumor
Activity in Non-Small Cell Lung Cancer. PLoS One 2016, 11 (3),
No. e0150567.
(30) Tiede, S.; Meyer-Schaller, N.; Kalathur, R. K. R.; Ivanek, R.;
Fagiani, E.; Schmassmann, P.; Stillhard, P.; Hafliger, S.; Kraut, N.;
Schweifer, N.; Waizenegger, I. C.; Bill, R.; Christofori, G. The FAK
Inhibitor BI 853520 Exerts Anti-Tumor Effects in Breast Cancer.
Oncogenesis 2018, 7 (9), 73.
I
https://dx.doi.org/10.1021/acsmedchemlett.0c00338
ACS Med. Chem. Lett. XXXX, XXX, XXX−XXX