Benzopyrimidodiazepinone inhibitors of TNK2

Article abstract of DOI:10.1016/j.bmcl.2020.126948

The SAR of a series of benzopyrimidodiazepinone inhibitors of TNK2 was developed, starting from the potent and selective compound XMD8-87. A diverse set of anilines was introduced in an effort to improve the in vivo PK profile and minimize the risk of quinone diimine formation.

Full text of DOI:10.1016/j.bmcl.2020.126948

Journal Pre-proofs
Benzopyrimidodiazepinone inhibitors of TNK2
Brian J. Groendyke, Chelsea E. Powell, Frederic Feru, Thomas W. Gero,
Zhengnian Li, Hilary Szabo, Kevin Pang, John Feutrill, Bailing Chen, Bin Li,
Nathanael S. Gray, David A. Scott
PII:
S0960-894X(20)30001-9
DOI:
https://doi.org/10.1016/j.bmcl.2020.126948
BMCL 126948
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
19 November 2019
20 December 2019
1 January 2020
Please cite this article as: 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, Bioorganic & Medicinal
Chemistry Letters (2020), doi: https://doi.org/10.1016/j.bmcl.2020.126948
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will
undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing
this version to give early visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
© 2020 Published by Elsevier Ltd.

Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com
Benzopyrimidodiazepinone inhibitors of TNK2
Brian J. Groendykea,b, Chelsea E. Powella,b, Frederic Ferua,b, Thomas W. Geroa,b, Zhengnian
Lia,b, Hilary Szaboc, Kevin Pangd, John Feutrille, Bailing Chene, Bin Lie, Nathanael S. Graya,b,
David A. Scott*,a,b
a Department of Cancer Biology, Dana-Farber Cancer Institute, Boston, MA 02115, USA.
b Department of Biological Chemistry & Molecular Pharmacology, Harvard Medical School, 360 Longwood Ave, Boston, MA 02115, USA.
c Vivid BioSciences, 50 Northern Ave, Boston, MA 02210, USA
d Northeastern University, 360 Huntington Ave, Boston, MA 02115, USA.
e SYNthesis med chem, 425 Changyang Street, Suzhou Industry Park, Suzhou, Jiangsu, China.
ARTICLE INFO
ABSTRACT
Article history:
Received
Revised
The SAR of a series of benzopyrimidodiazepinone inhibitors of TNK2 was developed,
starting from the potent and selective compound XMD8-87. A diverse set of anilines was
introduced in an effort to improve the in vivo PK profile and minimize the risk of quinone
diimine formation.
Accepted
Available online
2009 Elsevier Ltd. All rights reserved.
Keywords:
TNK2
Kinase inhibitor
Benzopyrimidodiazepinone

Tyrosine Kinase Nonreceptor 2 (TNK2, also known as ACK1) is
a cytoplasmic kinase that was first identified based on its
binding to cell division control protein 42 homolog (CDC42), a
cell-cycle regulator.1 TNK2 acts as an effector of CDC42, which in
turn regulates cell migration.2 Over-expression of TNK2 in cancer
cell lines can increase cellular motility, invasiveness, and the
ability of cells to metastasize.3 TNK2 mutations and/or over-
expression are associated with numerous cancers, including
renal, lung, ovarian, gastric, prostate, and breast cancers.4 Small
molecule inhibitors of TNK2 from several structural classes have
been explored as potential treatments for some of these
indications, though none have yet reached the clinic.5-8 Recently,
Maxson et al. identified two point mutations, D163E and R806Q,
in TNK2 that act as driver mutations of acute myelogenous
leukemia (AML) suggesting that TNK2 inhibitors may have utility
as a therapy for AML.4
further optimization for specific targets through fine-tuning of
additional substituents on the core and the aniline.
O
O
H
N
O
N
N
N
N
N
N
N
N
HN
N
N
HN
N
HN
N
N
EtO
MeO
MeO
N
N
O
N
OH
N
XMD8-92
LRRK2-IN-1
XMD8-87
Figure 1. Benzopyrimidodiazepinone kinase inhibitors
Kinase selectivity profiling of multiple examples from the
benzopyrimidodiazepinone series led to the identification of
XMD8-87 as a potent and highly selective TNK2 inhibitor.4,19
XMD8-87 showed little activity against other kinases in both the
Kinativ20 and KinomeScan21 panels (SI – Fig 1) of 241 and 402
kinases respectively and is a useful pharmacological tool to
explore the effects of TNK2 inhibition. Based on our kinase
profiling data, benzopyrimidodiazepinones unsubstituted on the
central ring amide showed the greatest activity against TNK2. In
our work targeting other kinases with this series, we typically
observed increased potency for compounds with a methyl
substituent on the amide. For TNK2, modeling studies suggested
that the amide NH interacts with the Thr205 gatekeeper residue,
in a manner similar to that reported for a co-crystal structure
with TNK2 and an aminopyrazolopyrimidine.6
The tricyclic benzopyrimidodiazepinone series and related
scaffolds have delivered both highly selective tool compounds,4,
9-11
and advanced lead molecules for ERK5, (XMD8-92)12,13 and
LRRK2, (LRRK2-IN-1)14 (Fig 1). Most compounds derived from this
scaffold have an aniline group installed on the pyrimidine ring,
and for these inhibitors the overall kinase selectivity profile is
often enhanced by a methoxy substituent at the aniline 2-
position. The 2-MeO aniline group is present in other
pyrimidine-based kinase inhibitor scaffolds,15-18 but is not
sufficient by itself to confer good selectivity. The high degree of
overall
kinase
selectivity
achievable
in
the
benzopyrimidodiazepinone series is likely to be driven by the
combination of the 2-MeO aniline and the tricyclic core, with
H
N
O
H
N
O
NO2
NH
NO2
N
N
N
NO2
Cl
N
N
N
(a)
Cl
N
(c)
Cl
N
(b)
(d)
Cl
N
HN
R2
N
N
N
Cl
N
CO2Me
CO2Me
(d)
R1
R1
1
7
Me
R1
10 - 44
2
(i)
NH2
3
CO2Me
(k)
(e)
NO2
N
HN
NH2
N
F
N
4
CO2Me
(j)
HN
N
HN
N
N
NH2
(f), (g), (h)
CO2Me
CO2Me
R
CO2Me
6
R1
R2
R1
R2
R
8
9
5
i
i
Scheme 1. Synthesis of benzopyrimidodiazepinones. Reagents and conditions: (a) methyl anthranilate, Pr2NEt, PrOH; (b) NaH, MeI, DMF; (c) Fe, AcOH, 50°C;
(d) aniline, Pd2(dba)3, Xphos, K2CO3, tBuOH, 100°C; (e) MeI, K2CO3, acetone, 65°C; (f) (CF3CO)2O, iPr2NEt, DCM; (g) MeI, K2CO3, DMF; (h) NaOMe, MeOH; (i) 2,4-
dichloro-5-nitropyrimidine, iPr2NEt, dioxane, 50°C; (j) H2, Pd/C, MeOH; (k) TsOH, dioxane, MW, 150°C.
Disappointingly, XMD8-87 was very highly cleared in vivo in
rodent PK studies (Table 4). This was in contrast to several N-
methyl amide examples designed to target other kinases,12-14
and also XMD8-85, the N-methyl matched pair of XMD8-87, all
of which demonstrated good in vivo mouse PK profiles.

install the aniline substituent at the pyrimidine C2 position via
Pd coupling chemistry. In some cases, the second aniline was
incorporated prior to the cyclization step to generate the
With the goal of developing a TNK2 inhibitor from the
benzopyrimidodiazepinone series, we sought to explore the SAR
and improve the in vivo PK profile. We introduced substituents
to the phenyl ring of the tricyclic core and explored a broader
range of anilines and aniline replacements at the C2 position of
the pyrimidine ring. We were keen to retain the exquisite kinase
selectivity of XMD8-87 to help confirm the relevance of TNK2 as
a therapeutic target.
bisanilinopyrimidines
8,
and
a
two-step
nitro
reduction/cyclization sequence was employed to generate the
final compounds. Characterization data for the final compounds
is included in the Supplementary Data.
TNK2 enzyme IC50s and cell data for compounds with
modifications to the phenyl ring are shown in Table 1. None of
the substituents we introduced to the phenyl ring led to
significantly improved potency in either the TNK2 enzyme
assay,22 or TNK2 D163E cell assay.23 The majority of compounds
showed reduced TNK2 activity relative to XMD8-87 in both the
enzyme and cell, but a methyl group adjacent to the central ring
amine was tolerated (13) and fluorine could be introduced at
two positions (15, 17) without much impact on potency. We also
profiled XMD8-85, the N-methyl amide analog of XMD8-87, and
confirmed that the loss of the amide NH led to reduced TNK2
potency in the enzyme and cell assays.
Compounds were prepared using chemistry similar to that
previously described (Scheme 1), with modifications to the
reaction sequence depending upon starting material availability
and differences in intermediate reactivity for some steps.12
Anthranilic esters were installed at the 4-position of 2,4-
dichloro-5-nitropyrimidine, either as the primary anilines to give
2, with a subsequent methylation step, or as the N-methyl
anilines, to give 3. The N-methylanilines 6 were commercially
available or prepared by methylation of either the primary
aniline or trifluoroacetamide. A one-pot reduction/cyclization
was achieved with iron in acetic acid and moderate heating for
most examples to give the tricyclic core 7, with a final step to
16
17
18
19
-
-
-
-
9-F
308
101
723
2480
1.01
0.23
3.59
1.39
10-F
8-CN
9-CN
Table 1. Phenyl ring SAR; compounds 10-19
R1
O
N
7
N
8
R2
HN
N
N
9
MeO
10
N
N
Table 2. Aniline SAR, 2- and 3-positions; compounds 20-27
H
N
O
Compound R1
R2
TNK2 IC50
(nM)a
Cell TNK2 D163E (uM)
N
HN
N
N
R1
R2
XMD8-87
-
-
-
44
0.19
0.56
1.63
1.92
0.26
0.13
0.38
0.09
N
N
XMD8-85
Me
959
1610
363
223
73*
217
26*
10
11
12
13
14
15
-
-
-
-
-
-
7-Me
8-Me
9-Me
10-Me
7-F
Compound R1
R2
TNK2 IC50
(nM) a
Cell TNK2 D163E
(uM)
XMD8-87
OMe
H
H
44
0.19
0.04
20
H
16*
8-F

21
22
23
24
OEt
Me
Cl
H
88*
97
63
4
0.40
0.37
0.38
0.012
25
26
27
H
H
H
OEt
Me
Cl
4
0.016
0.05
0.10
H
6
H
11
a Data was generated at BioDuro or ThermoFisher (*).
H
OMe
a Data was generated at BioDuro or ThermoFisher (*).
stability in mouse microsomes and hepatocytes. In particular, the
in vitro data for XMD8-87 and N-methyl analog XMD8-85 was
similar (Table 4), despite different rates of in vivo clearance. The
contrasting PK profiles of XMD8-87 and XMD8-85 suggested
that the poor PK of the former derived from the tricyclic core
rather than the aniline. However, in vitro metabolite ID data did
not suggest a metabolic pathway that directly involved the
unsubstituted amide. Instead, an oxidation step appeared to be the
major metabolic pathway for XMD8-87, with the only significant
metabolite corresponding to a molecular weight one less than the
parent (Table 5 and SI Figure 2). This was also the major
metabolite in rat urine in an in vivo metabolite ID study, and we
speculated that it corresponded to the quinone diimine. The
corresponding GSH adduct was observed in rat bile from the
same study. The in vitro metabolite ID data for the N-methyl
analog XMD8-85 showed a similar metabolite, but with a
reduced rate of conversion (SI Figure 3).
We felt that small modifications on the phenyl ring of the
tricyclic core were unlikely to significantly impact the PK profile
of the series, and therefore focused our efforts on the aniline
substituent. We prepared a set of compounds combining small
substituents at the aniline 2- and 3-positions with a 4-N-methyl
piperazine group (Table 2). Compound 20, lacking the 2-MeO
group, was more potent than XMD8-87, but our archive of
kinase profiling data (including Kinativ data for 29)4 suggested
that compounds of this type were less selective. A small
substituent at the aniline 3-position led to enhanced TNK2
potency in both enzyme and cells (e.g. 24) but reduced
selectivity in the KinomeScan panel (SI – Figure 1), and we
concluded that 2-MeO gave the best balance of potency and
kinase selectivity.
Examples 31-34 were profiled for IV PK in mice, and we
observed high levels of clearance, similar to those seen with
XMD8-87. This was especially disappointing for the piperidine
examples, as an in vitro metabolite ID study with 33 showed a
reduced rate of conversion to the MW-1 oxidation product (SI
Figure 4).
We also prepared a set of compounds with structurally
diverse cyclic amines at the 4-position, designed to deliver
acceptable physical properties (Table 3). The 1,4-diamino
substitution pattern on a phenyl ring is a perceived risk in terms
of formation of a reactive quinone diimine. This concern has
been mitigated previously for kinase inhibitors by replacing a
piperazine substituent at the aniline 4-position with
a
piperidine.24-26 In addition to compounds such as 28, 30, 31 and
32, we therefore prepared compounds with a carbon-linked
piperidine (33, 34). Compounds with the 2-MeO substituent and
a piperazine or N-linked piperidine at the aniline 4-position
typically showed good potency in our cell assay, while the C-
linked piperidines (33, 34) were less potent. 3-MeO anilines (35 -
37) delivered consistently improved potency in both the enzyme
and cell assays, but we assumed that their overall kinase
selectivity profiles would resemble that of 24. Moving the
piperazine substituent to the 3-position (38) led to a 10-fold
reduction in enzyme and cell potency. We also installed bicyclic
anilines as an alternative strategy to mitigate potential quinone
diimine formation, following the approach adopted in the
development of tetrahydroisoquinoline pyrimidine ALK
inhibitors.27 As with the monocyclic aniline examples,
compounds without the 2-MeO group (39, 41, 43) were more
potent in the enzyme and cell.
Table 3. Aniline SAR, amines; compounds 28-44
H
N
O
N
HN
N
N
R1
R2
R3
Compound R1
R2
R3 or aniline
replacement
TNK2
IC50
(nM) a
Cell
TNK2
D163E
(uM)
XMD8-87
OMe
H
H
H
H
N-Me piperazine
4-OH piperidine
4-OH piperidine
morpholine
44
0.19
0.14
0.08
0.19
28
29
30
OMe
H
37*
7*
We did not see a good correlation between in vitro and in vivo
PK for this series, with compounds frequently showing good
OMe
41

H
N
31
32
OMe
OMe
H
H
37
29
0.08
0.06
39
40
-
-
-
-
26
91
0.30
0.85
N
N
N
N
HO
HN
H
N
N
OMe
H
N
41
42
-
-
-
-
18
50
0.11
0.26
33
34
OMe
OMe
H
H
62
66
0.25
0.26
N
N
H
N
N
N
HO
N
N
OMe
35
36
H
H
OMe
OMe
7
4
0.013
0.013
H
N
43
44
-
-
-
-
28
0.12
0.11
N
HO
HN
H
N
140
N
N
OMe
37
38
H
-
OMe
-
13
0.14
1.10
a Data was generated at BioDuro or ThermoFisher (*).
N
N
254
H
N
N
OMe
Table 4. Mouse microsome and hepatocyte stability, and in vivo mouse PKa data (3mpk IV, 10mpk PO)
Compound ID
Mouse microsomes
Mouse hepatocytes
(T1/2
IV Cl
(ml/min/kg)
IV T1/2
(hr)
0.7
4.4
2.0
0.9
1.7
1.8
10
F %
Auc inf (PO)
(T1/2
)
)
(ng*hr/ml)
XMD8-87
28
22
139
-
347 mins
150
25
5
56
2000
-
XMD8-85
>500
46
-
31
32
33
34
42
44
75
-
140
130
100
130
120
86
11
-
141
-
94
-
64
-
-
-
-
-
7
125
525
-
-
1.4
22
a PK studies were conducted in male CD-1 mice, and all values are an average from three animals

A series of pyrrolotriazine kinase inhibitors containing a 1,4-
diamino substituted aniline were reported to give unusually high
levels of GSH adducts via formation of the reactive quinone
diimine.28 This was mitigated by the replacement of the
piperazine substituent with a piperidine.25 The data suggested
that relative to other chemical series, the pyrrolotriazine
scaffold greatly increased the susceptibility of the 1,4-diamino
aniline to oxidation, perhaps through interactions with the P450
In vitro metabolite ID data for examples with a bicyclic aniline
41 and 44 (SI Figure 5) showed a distinct profile relative to the
other compounds, suggesting an alternative metabolic pathway.
For both compounds, there was no evidence of the MW-1
metabolite, with the main routes of metabolism being
conversion to an oxidation product (+16) or demethylation (-14).
For 44, the rate of conversion of the parent appeared relatively
slow. Disappointingly, however, the in vivo PK data for bicyclic
anilines 42 and 44 showed only a limited improvement over
XMD8-87, with high IV Cl for both compounds.
active
site.
Similarly,
it
is
possible
that
benzopyrimidodiazepinones such as XMD8-87 may be
susceptible to much higher rates of oxidation to the quinone
diimine than the corresponding methylated scaffold. This could
result in greatly increased in vivo clearance, although we did not
observe a correlation between the rate of metabolite formation
in our in vitro metabolite ID studies and mouse IV Cl.
In conclusion, we have developed the SAR of the
benzopyrimidodiazepinone series in regard to TNK2 inhibition
and pursued several approaches to mitigate the unexpectedly
high in vivo clearance for examples of this scaffold. Our further
efforts to optimize this series will be reported in due course.
Table 5. In vitro metabolite ID studies in mouse liver microsomes
Compound
XMD8-87
XMD8-85
33
Parent (m/z)+
446.2 (2%)
Major metabolite (m/z)+
445.2 (98%)
Other metabolites (m/z)+
Incubation time (mins)
< 1%
60
460.3 (40%)
445.2 (75%)
387.2 (17%)
417.2 (64%)
459.3 (53%)
446.2 (6%)
< 2%
60
444.2 (23%)
120
120
120
41
403.2 (73%)
373.2 (10%)
403.2 (11%), 415.2 (10%)
44
433.2 (11%)
Acknowledgments
5.
6.
7.
8.
9.
Jin, M.; Wang, J.; Kleinberg, A. et al. Bioorg Med Chem Lett.
2013, 23, 979.
Kopecky, D. J.; Hao, X.; Chen, Y. et al. Bioorg Med Chem Lett.
2008, 18, 6352.
Jiao, X.; Kopecky, D. J.; Liu, J. et al. Bioorg Med Chem Lett.
2012, 22, 6212.
Lawrence, H. R.; Mahajan, K.; Luo, Y. et al. J Med Chem. 2015,
58, 2746.
Ba/F3 cells expressing TNK2 D163E were generously
provided by Jeffrey W. Tyner (Oregon Health and Science
University, Portland, OR). In vitro and in vivo PK studies were
conducted at BioDuro, Shanghai, under the supervision of Cindy
Liang and Ying Zhou. We thank Vivid BioSciences for funding
support.
Fan, F.; He, Z.; Kong, L. L. et al. Sci Transl Med. 2016, 8,
352ra108.
References and notes
10. Ferguson, F. M.; Ni, J.; Zhang, T. et al. ACS Med Chem Lett.
2016, 7, 908.
1.
2.
3.
4.
Manser, E.; Leung, T.; Salihuddin, H. et al. Nature, 1993, 363,
364.
Modzelewska, K.; Newman, L. P.; Desai, R.; Keely, P. J. J Biol
Chem. 2006, 281, 37527.
van der Horst, E. H.; Degenhardt, Y. Y.; Strelow, A. et al. Proc
Natl Acad Sci USA. 2005, 102, 15901.
Maxson, J. E.; Abel, M. L.; Wang, J. et al. Cancer Res. 2016, 76,
127.
11. Banerjee, S.; Buhrlage, S. J.; Huang, H. T. et al. Biochem J. 2014,
457, 215.
12. Deng, X.; Yang, Q.; Kwiatkowski, N. et al. ACS Med Chem Lett.
2011, 2, 195.
13. Deng, X.; Elkins, J. M.; Zhang, J. et al. Eur J Med Chem. 2013,
70, 758.
14. Deng, X.; Dzamko, N.; Prescott, A. et al. Nat Chem Biol. 2011, 7,
203.

15. Finlay, M. R.; Anderton, M.; Ashton, S. et al. J Med Chem. 2014,
57, 8249.
16. Galkin, A. V.; Melnick, J. S.; Kim, S. et al. Proc Natl Acad Sci
USA. 2007, 104, 270.
17. Huang, W. S.; Liu, S.; Zou, D. et al. J Med Chem. 2016, 59, 4948.
18. Rudolph, D.; Steegmaier, M.; Hoffmann, M. et al Clin Cancer
Res. 2009, 15, 3094.
19. Miduturu, C. V.; Deng, X.; Kwiatkowski, N. et al Chem Biol.
2011, 18, 868.
20. Patricelli, M. P.; Szardenings, A. K.; Liyanage, M. et al.
Biochemistry. 2007, 46, 350.
21. Davis, M. I.; Hunt, J. P.; Herrgard, S. et al. Nat Biotechnol. 2011,
29, 1046.
22. Enzyme testing was conducted at BioDuro or at ThermoFisher,
using an HTRF LanthaScreen assay. For compounds tested at both
sites, there was an excellent correlation between the two assays,
with the ThermoFisher assay typically showing slightly increased
sensitivity (compounds up to 2-fold more potent).
23. We used a cell viability assay based on Ba/F3 cells expressing
TNK2 D163E. Cells were plated at a volume of 50 µL with a
density of 105 cells/mL in white 384-well plates (Corning 3570).
100 nL of compound in DMSO from stock plates was added by
pin transfer using a Janus Workstation (PerkinElmer). After 72 hr,
cell viability was evaluated using the CellTiter-Glo Luminescent
Cell Viability Assay (Promega) following the manufacturer’s
standards. IC50s were modeled from 3 biological replicates using
GraphPad Prism 8 software.
24. Marsilje, T. H.; Pei, W.; Chen, B. et al. J Med Chem. 2013, 56,
5675.
25. Mesaros, E. F.; Thieu, T. V.; Wells, G. J. et al. J Med Chem. 2012,
55, 115.
26. Weinberg, L. R.; Albom, M. S.; Angeles, T. S. et al. Bioorg Med
Chem Lett. 2011, 21, 7325.
27. Achary, R.; Yun, J. I.; Park, C. M. et al. Bioorg Med Chem. 2016,
24, 207.
28. Wells-Knecht, K. J.; Ott, G. R. Cheng, M. et al. Chem Res
Toxicol. 2011, 24, 1994.
Supplementary Material

Declaration of interests
X The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
☐The authors declare the following financial interests/personal relationships which may be considered
as potential competing interests:
Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions
Leave this area blank for abstract info.
Benzopyrimidodiazepinone inhibitors of
TNK2
Brian J. Groendyke, Chelsea E. Powell, Frederic Feru, Thomas W. Gero, Zhengnian Li, Hilary Szabo, Kevin
Pang, John Feutrill, Bailing Chen, Bin Li, Nathanael S. Gray, David A. Scott
H
N
O
N
H
N
O
N
HN
N
N
MeO
HN
N
N
N
MeO
N
N