Highly Selective PTK2 Proteolysis Targeting Chimeras to Probe Focal Adhesion Kinase Scaffolding Functions

Article abstract of DOI:10.1021/acs.jmedchem.8b01826

Focal adhesion tyrosine kinase (PTK2) is often overexpressed in human hepatocellular carcinoma (HCC), and several reports have linked PTK2 depletion and/or pharmacological inhibition to reduced tumorigenicity. However, the clinical relevance of targeting

Full text of DOI:10.1021/acs.jmedchem.8b01826

Subscriber access provided by MIDWESTERN UNIVERSITY
Article
Highly Selective PTK2 Proteolysis Targeting Chimeras (PROTACs)
to Probe Focal Adhesion Kinase Scaffolding Functions
Johannes Popow, Heribert Arnhof, Gerd Bader, Helmut Berger, Alessio Ciulli, David Covini, Christian
Dank, Teresa Gmaschitz, Peter Greb, Jale Karolyi-Oezguer, Manfred Koegl, Darryl McConnell, Mark
Pearson, Maria Rieger, Jörg Rinnenthal, Vanessa Roessler, Andreas Schrenk, Markus Spina, Steffen
Steurer, Nicole Trainor, Elisabeth Traxler, Corinna Wieshofer, Andreas Zoephel, and Peter Ettmayer
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b01826 • Publication Date (Web): 09 Feb 2019
Downloaded from http://pubs.acs.org on February 10, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 52
Journal of Medicinal Chemistry
Popow et al.
1
1
2
3
4
5
6
7
8
9
Highly Selective PTK2 Proteolysis Targeting
Chimeras (PROTACs) to Probe Focal Adhesion
Kinase Scaffolding Functions.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
1
1
1
2
Johannes Popow , Heribert Arnhof , Gerd Bader , Helmut Berger , Alessio Ciulli , David
1
1
1
1
1
Covini , Christian Dank , Teresa Gmaschitz , Peter Greb , Jale Karolyi-Özguer , Manfred
1
1
1
1
Koegl , Darryl B. McConnell , Mark Pearson , Maria Rieger, Joerg Rinnenthal , Vanessa
1
1
1
1
2
Roessler , Andreas Schrenk , Markus Spina , Steffen Steurer , Nicole Trainor , Elisabeth
1
1
1
1*
Traxler , Corinna Wieshofer , Andreas Zoephel , Peter Ettmayer
1
2
Boehringer Ingelheim RCV GmbH & Co KG, 1221 Vienna, Austria
Division of Biological Chemistry and Drug Discovery, School of Life Sciences, James Black
Centre, University of Dundee, Dow Street, DD1 5EH, Dundee, Scotland, United Kingdom
*corresponding author, E-mail: peter.ettmayer@boehringer-ingelheim.com
KEYWORDS
PROTACs, protein degradation, E3 ubiquitin ligases, kinases, von Hippel-Lindau, Cereblon,
CRBN
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 52
Popow et al.
2
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 3 of 52
Journal of Medicinal Chemistry
Popow et al.
3
1
2
3
4
5
6
7
8
9
ABSTRACT
The focal adhesion tyrosine kinase (PTK2) is often over-expressed in human hepatocellular
carcinoma (HCC) and several reports have linked PTK2 depletion and/or pharmacological
inhibition to reduced tumorigenicity. However, the clinical relevance of targeting PTK2 still
remains to be proven. Here we present two highly selective and functional PTK2 PROTACs
utilizing VHL and cereblon ligands to hijack E3 ligases for PTK2 degradation. BI-3663
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(cereblon-based) degrades PTK2 with a median DC of 30 nM to > 80 % across a panel of
50
eleven HCC cell lines. Despite effective PTK2 degradation, these compounds did not phenocopy
the reported anti-proliferative effects of PTK2 depletion in any of the cell lines tested. By
disclosing these compounds, we hope to provide valuable tools for the study of PTK2
degradation across different biological systems.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 52
Popow et al.
4
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Introduction
Focal adhesion tyrosine kinase (PTK2) is a cytoplasmic protein tyrosine kinase that is
overexpressed and activated in many types of advanced-stage solid cancers. PTK2 is reported to
play an important role in adhesion, spreading, motility, invasion, metastasis, survival,
angiogenesis, epithelial to mesenchymal transition (EMT), cancer stem cells and the tumour
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
,2
microenvironment . Overexpression and activation of PTK2 is associated with several human
3
4
5
malignant diseases, including colorectal , ovarian , esophageal and hepatocellular carcinoma
6
7, 8
(
HCC) and is correlated with poor overall patient survival . Hepatocellular carcinoma (HCC)
is the most common type of liver cancer in humans, accounting for 70–85 % of primary liver
malignancies, and is the third leading cause of cancer-related death worldwide9, 10. HCC is
associated with an extremely poor prognosis since it is often diagnosed at advanced stages,
restricting the currently available therapeutic options to either surgical resection or liver
transplantation11, 12. Several reports link PTK2 depletion or pharmacological inhibition of its
kinase activity to reduction of in vitro and in vivo tumorigenicity in HCC models13, 14. However,
the disconnect between modulation of intracellular PTK2 autophosphorylation and growth
inhibition as well as the often suboptimal selectivity profile of the inhibitors used makes it
difficult to link the reported blockade of HCC tumour initiation and maintenance to PTK2
inhibition.
We previously described BI-853520, a novel ATP-competitive inhibitor distinguished by high
potency and selectivity15, 16. BI-853520 inhibits PTK2 autophosphorylation in cancer cell lines
and blocks anchorage-independent proliferation with single digit nmol/L potency. In contrast,
cells grown in conventional surface (2-D) culture were found to be 1,000-fold less sensitive to
BI-853520. These findings are in keeping with the described role of PTK2 in integrin mediated
ACS Paragon Plus Environment

Page 5 of 52
Journal of Medicinal Chemistry
Popow et al.
5
1
2
3
4
5
6
7
8
9
signaling2, 17. We reasoned that ligands derived from BI-853520 would be suitable to develop
PROTACs (PROteolysis TArgeting Chimeras) and compare the potential anti-oncogenic effects
1
8
of PTK2 depletion as compared to inhibition of its kinase activity .
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
PROTACs are bifunctional degrader molecules composed of a ligand for the target protein
linked to a module that recruits an E3 ligase19, 20. PROTACs represent an emerging therapeutic
strategy to use small molecules to deplete a protein by repurposing the ubiquitin-proteasome
system21, 22. Upon formation of a ternary complex target:PROTAC:E3 ligase23-25, the protein of
interest is ubiquitinated and subsequently degraded by the proteasome. Compared to
pharmacological inhibition of the kinase function, protein degradation more closely resembles
genetic approaches to interfere with PTK2 expression in HCC models. Moreover, PROTAC
molecules can add a layer of target selectivity beyond that expected from the constitutive binding
ligands, thus providing highly selective degraders with potentially reduced off-target effects25-28
.
The most common E3 ligases currently recruited using PROTACs are the von Hippel-Lindau
VHL) protein complex CRL2VHL and the cereblon (CRBN) complex CRL4CRBN. PROTACs
(
consisting of the same target ligand but employing either VHL or CRBN ligands can exhibit
different degradation selectivity and efficacy26, 29-31. Emerging evidence suggests that it might be
beneficial to develop parallel chemical series hijacking different E3 ligases. First, expression of
the recruited E3 ligase and intrinsic activity may be context-dependent and vary widely amongst
3
2
cell- and tissue types . Second, resistance mechanisms could potentially arise from loss of the
recruited E3 ligase, as demonstrated by the correlation between level of CRBN and response to
33
CRBN-recruiting drugs in multiple myeloma . Flexible choice of the recruited E3 ligase may
therefore aid targeted protein degradation.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 52
Popow et al.
6
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Recently, the first PTK2 PROTACs have been reported. Conjugating the known pyrimidine-
based ALK inhibitors, TAE684 and LDK378, to the cereblon ligand pomalidomide led to the
discovery of the first small molecule degraders of ALK that concomitantly degrade, PTK2,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
34
Aurora A, FER, and RPS6KA1 . A PROTAC based on a highly promiscuous kinase inhibitor
3
5
was also reported to degrade PTK2 amongst 27 additional kinases . While both studies
demonstrate that PTK2 can be targeted by a PROTAC approach, they also clearly highlight the
need for more selective PTK2 PROTACs to address its role in cancer. Motivated by the potential
of targeting kinase-independent functions of PTK2 with a low molecular weight modality, we
initiated a PROTAC medicinal chemistry campaign to artificially recruit PTK2 to two different
E3 ligases.
Here, we show the synthesis and characterization of the first probe-quality PROTACs targeting
PTK2. Structure-guided conjugation of a highly selective PTK2 inhibitor to either a CRBN or
VHL ligand led to selective PTK2 degraders. We profiled both PROTACs for degradation
efficacy and anti-proliferative activity in a panel of HCC, tongue squamous cell carcinoma,
melanoma, pancreatic ductal adenocarcinoma and non-small cell lung cancer cell lines several of
which were recently shown to depend on expression of PTK2 by functional RNA interference
3
6
screening . Despite potent and complete degradation of PTK2, both degraders failed to show the
anticipated growth reduction beyond the effect of a PTK2 kinase inhibitor in these models.
Results and Discussion
We designed a set of PTK2 PROTACs recruiting the two E3 ubiquitin ligases, VHL and
CRBN, to test whether PTK2 can be targeted by targeted proteolysis. Making use of available E3
ligase ligands, we aimed at maximizing the chance of developing an active PTK2 degrader tool
compound. We selected the highly selective ATP competitive inhibitor BI-4464 (1, Scheme 1), a
ACS Paragon Plus Environment

Page 7 of 52
Journal of Medicinal Chemistry
Popow et al.
7
1
2
3
4
5
6
7
8
9
15
close analogue of the 2-aminophenyl-4-phenoxypyrimidine BI-853520 currently being tested in
clinical trials as a PTK2 ligand. The choice of BI-4464 as a target ligand was driven by its high
binding affinity and exquisite selectivity across a large kinase panel (see also Figure 3). We
believe that this exceptional selectivity originates from a unique binding mode which seems to be
less tolerated by other kinases. To design a first generation of degraders, we inspected the crystal
structure of BI-4464 bound to the PTK2 kinase domain to identify suitable attachment points and
vectors for linker conjugation (Figure 1). The structure shows binding of the trifluoro-
aminopyrimidine moiety to the backbone nitrogen and carbonyl of cysteine 502 in the hinge
region via two hydrogen bonds. In addition, the oxygen of the dihydroindenone moiety forms
another hydrogen bond to the backbone nitrogen of aspartic acid 564. The N-methyl piperidine
group of BI-4464 was identified as solvent-exposed and not involved in any interactions with the
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
7
38
protein. We further selected 2 and pomalidomide 3, (Scheme 1) as VHL and CRBN ligands,
respectively. Their terminal amino groups (Table 1 and Scheme 1) were conjugated via an amide
bond to the linker consisting of up to five PEG units without perturbing the interaction with the
E3 ligases, as previously demonstrated28, 31, 39, 40
.
Scheme 1. Chemical structures of BI-4464 ligand 1, VHL ligand 237, 41, 42 and CRBN ligand 338.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 52
Popow et al.
8
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structure of BI-4464 bound to PTK2 (PDB ID 6I8Z). Hydrogen bonds to cysteine 502
and aspartic acid 564 are depicted in blue.
To obtain the described compounds 6 and 8 (Table 1), a convergent synthesis was developed to
connect 1a to each E3 ligase ligand 2 and 3. PEG linkers 1-5 (Scheme 2 shows an example of
linker length three) were synthesized and profiled. The synthesis of ligand 1a was carried out as
outlined in Scheme 2. Aromatic nucleophilic substitution of the 4-chloropyrimidine 4 with
7
‐hydroxy‐2,3‐dihydroindan‐1‐one produced the carboxylic acid 1a. Amide coupling of azido-
PEG-3-propanoic acid and azido-PEG-3-acetic acid to the E3 ligase ligand 2 and 3 followed by
catalytic hydrogenation produced the amines 5b, 7b and 10b in good yields. The final PROTACs
6
, 8 and the cis-VHL analogue 11 were synthesized following amide coupling of the ligase linker
conjugates 5b, 7b and 10b with 1a.
ACS Paragon Plus Environment

Page 9 of 52
Journal of Medicinal Chemistry
Popow et al.
9
1
2
3
4
5
6
7
8
9
Table 1. Binary affinities of compounds 1, 6, 8 for PTK2 and the respective PTK2 degradation
data in A549 cells.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
E3
A549 cells, 18h**
PTK2*
pIC50
Code
N
X
ligase
ligand
pDC50
Dmax [%]
1
6
(BI-4464)
(BI-3663)
-
-
-
7.8 ± 0.1
7.7 ± 0.1
>10,000
7.6 ± 0.1
-
3
CONH
POMA
95 ± 4
O-CH
CONH
Thermo Fisher selectScreen Kinase Profiling Services, Z´-Light, ATP@Km, pIC50
2
-
8
(BI-0319)
2
2
7.7 ± 0.1
6.7 ± 0.4
80 ± 9
*
± STDEV ** Degradation activity is reported as concentration needed to achieve 50 %
PTK2 protein degradation (pDC50 ± STDEV) and maximal achievable protein
degradation (Dmax) relative to DMSO. PTK2 levels were determined by protein
capillary electrophoresis and normalized to GAPDH. (N = 3)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 52
Popow et al.
10
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Scheme 2. Synthesis of the PTK2 degraders.a
F
NH2
F
F
F
O
F
F
F
OH
O
Cl
N
F
O
F
N
HN
N
O
N
b
N
N
N
O
O
N
O
NH
a
HN
O
NH
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
O
O
OH
O
OH 1a
O
4
O
1 (BI-4464)
H
N
H
O
N
O
N
O
F
F
F
O
O
O
NH
O
N
N
+
N
O
O
O
H
O
N
N
O
NH
N
O
N
O
O
OH
1a
N
O
O
HN
c,d
e
H
5
a, R = N3
N
O
O
O
NH2
O
O
O
O
5b, R = NH2
O
3
6 (BI-3663)
O
HO
R
O
O
O
HO
F
F
F
N
HN
HO
NH
O
O
N
+
N
O
N
O
NH
N
O
OH
N
N
O
HN
O
O
N
O
1a
HN
O
HN
NH2
O
H
N
h
O
O
f,g
S
O
O
O
N
2
O
8 (BI-0319)
O
S
7a, R = N3
O
N
S
7
b, R = NH2
N
HO
R
HO
NH
O
F
F
F
O
N
HO
NH
N
+
N
O
O
N
HN
N
O
OH
O
HN
O
O
N
N
O
NH2
O
N
O
1a
O
O
HN
O
HN
i,j
k
O
H
N
O
O
O
S
9
O
11 (BI-4206)
O
S
N
O
N
S
10a, R = N3
10b, R = NH2
O
N
R
aReaction Conditions: (a) Cs
T3P, pyridine, DMF, 80°C, 3h, 76%; (d) H
rt, 18 h, 10%; (f) HATU, DIPEA, DMF, rt, 1 h, 63%; (g) H
, dioxane, 80°C, 16 h, 78%; (b) HATU, DIPEA, DMF, rt, 2 h, 75%; (c)
(6 bar), Pd/C (10%), MeOH, rt, 3 h, 90%; (e) 1a, HATU, DIPEA, DMF,
(5 bar), Pd/C (10%), MeOH, rt, 2.5 h, 82%; (h) 1a,
2
3 4
CO , MgSO
2
2
2
HATU, DIPEA, DMF, rt, 24 h, 34%; (i) HATU, DIPEA, DMF, rt, 2 h, 47%; (j) H (6 bar), Pd/C (10%), MeOH, rt, 3
h, 98%; (k) 1a, HATU, DIPEA, DMF, rt, 16 h, 69%.
ACS Paragon Plus Environment

Page 11 of 52
Journal of Medicinal Chemistry
Popow et al.
11
1
2
3
4
5
6
7
8
9
We next tested PTK2 degradation by all synthesized PROTACs (sampling linker lengths of up
to five ethylene glycol units) after 16 hours treatment of the human lung adenocarcinoma cell
line A549. PROTACs 6 (BI-3663) and 8 (BI-0319) were identified as the best degraders in the
CRBN and VHL series and characterized in greater detail. The CRBN-based PROTAC 6 (BI-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
663) degraded PTK2 potently (DC = 27 nM) with a maximally obtainable degradation at the
50
chosen experimental conditions of 95% (Dmax) while the best VHL-based PROTAC 8 (BI-0319)
only achieved partial degradation (Dmax = 80 %) and was less potent (DC = 243 nM) in A549
50
cells (Table 1 and Figure 2). Further, both PROTACs show a small apparent hook effect in A549
cells at 25 µM. Consistent with VHL dependency, which requires the substituents of the
2
8
pyrrolidine in the VHL ligand in a trans configuration , degradation of PTK2 by 8 (BI-0319)
was abolished by switching the stereochemistry of the pyrrolidine substituents to the cis
configuration (11, BI-4206) (Figure 2A). Moreover, as predicted for neddylation-dependent E3
4
3
ligases such as VHL and CRBN, the NEDD8 inhibitor MLN4924 (MLN) inhibited degradation
of PTK2 by 6 (BI-3663) and 8 (BI-0319) (Figure 2A and B). Importantly, the PTK2 ligand 1
itself did not affect PTK2 levels.
We assessed PTK2 target engagement in vitro to understand the effect of the linker and exit
vector on PTK2 inhibition of PROTACs 6 and 8. As predicted from the structure-based design,
both PROTACs and the PTK2 ligand inhibited PTK2 in vitro with virtually identical potencies
(
IC = 18 and 19 versus 17 nM for the PTK2 inhibitor 1).
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 52
Popow et al.
12
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 2. Characterization of BI-0319 and BI-3663 in A549 cells. A) A549 cells were treated
with the indicated concentrations of BI-0319 in presence or absence of the NEDD8 inhibitor
MLN4924 (MLN, 3 µM) or an inactive stereoisomer (BI-4206) for 18 h. PTK2 levels were
determined by protein capillary electrophoresis and normalized to GAPDH. Values are stated as
percentages of DMSO controls (POC). B) A549 cells were treated with the indicated
concentrations of BI-3663 in presence or absence of the NEDD8 inhibitor MLN4924 (3 µM) or
with the PTK2 inhibitor BI-4664 for 18 h. PTK2 levels were measured and are stated as in A.
In contrast to the previously published less selective PTK2 degraders34, 35, we believe that a
defining aspect of the PTK2 degraders described here is their high selectivity, which should
enable validation of the relevance of PTK2 scaffolding functions in biologically relevant
contexts. Indeed, of the 397 kinases tested in a kinase panel screening only 2 were inhibited by
more than 90 % at 1 µM (Figure 3). Interestingly, BI-0319 (8) is more selective than the already
highly selective PTK2 TKI 1 (BI-4464). We assume that the kinase selectivity panel for BI-3663
(6) might be comparable to BI-0319 since both exit vector and linker are identical for a distance
of nine atoms from the piperidine moiety of the PTK2 ligand
ACS Paragon Plus Environment

Page 13 of 52
Journal of Medicinal Chemistry
Popow et al.
13
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Kinase selectivity panels of the PTK2 ligand BI-4464 and VHL PROTAC BI-0319.
Bars indicate percent inhibition at 1 µM of either compound.
We employed multiplexed isobaric tagging mass spectrometry to assess the cellular selectivity
of BI-3663 (6) and BI-0319 (8) for PTK2 degradation and identify potential degradation off-
targets in a quantitative and unbiased manner. Amongst the 6,008 proteins quantified in this
analysis in A549 cells, PTK2 showed a distinct and significant change in abundance upon
treatment with either PROTAC. (Figure 4 and Supplemental table 3). Neither BI-3663 (6) nor
BI-0319 (8) induced any significant changes in abundance of other detectable kinases, thus
confirming the high selectivity of both degraders within the kinase family. Of note, the two most
prominent kinase off-targets of the inhibitor were not detected in this dataset. Interestingly BI-
0
319 (8) – but not BI-3663 (6) – also induced a significant change of PDE6D levels (Figure 4A),
a finding corroborated by immunoblot in A549 cells (Supplemental figure 1).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 52
Popow et al.
14
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Total proteome analysis of A549 cells treated with 3 µM of BI-0319 (A) or BI-3663
B) for 18 h and compared to DMSO controls. Samples were run in biological triplicates and
(
analyzed by mass spectrometry. Volcano plot displays log2 of fold-change in abundance versus –
log10 of adjusted p value (N = 3).
Having established the high selectivity of both PROTACs, in particular of the CRBN-based
BI-3663, we tested their potential to degrade PTK2 in a panel of eleven HCC cell lines (Table 3
and Supplemental figure 2 and 3). Both PROTACs show comparable degradation potencies and
efficacies in the 11 HCC cell lines tested (BI-3663 mean pDC50 = 7.45±0.60 versus BI-0319
mean pDC50 = 7.08±0.52). Whereas PTK2 levels in cell lines such as SNU-398 were equally
sensitive to treatment with both PROTACs, other cell lines, such as Hep3B2.1-7 exhibited a >10
fold difference of the potencies of 6 and 8 in cell-based PTK2 degradation assays, as observed
for A549 cells during the profiling experiments (Figure 5 and Table 2).
It is not unexpected that heterobifunctional molecules unavoidably result in high molecular
weight compounds, making it difficult to balance their physicochemical properties in a manner
ACS Paragon Plus Environment

Page 15 of 52
Journal of Medicinal Chemistry
Popow et al.
15
1
2
3
4
5
6
7
8
9
consistent with acceptable permeability and solubility. While both PROTACs have a comparable
low aqueous solubility and high tPSA (241,243 Å2), PROTAC 6 has one hydrogen bond donor
(HBD) more than PROTAC 8. The negative impact on permeability for 6 might be
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
compensated by the higher lipophilicity (ClogP 6.9 vs 3.8) however, the higher lipophilicity of 6
also translates in higher PPB and lower free fraction (6: 4.0%, 8: 11.3; Supplemental table 1).
As assessed by the Caco2 assay, both PROTACs exhibit low permeability and significant
efflux (Supplemental table 1). Therefore, high efflux might compromise cellular exposures
achieved by both PROTACs. We reasoned that blocking multidrug efflux pumps (such as the
Multidrug-Resistance-Protein 1, PGP) might boost the potency and efficacy of both PROTACs
in cell lines with a large discrepancy between biochemical and intracellular activity.
Cyclosporine A (CsnA) has been described as a substrate of several drug transporters including
PGP44 and can therefore be used to saturate these transporters at high concentrations.
Consequently, we assessed PTK2 degradation by both PROTACs upon addition of CsnA to
saturate drug transport in Hep3B2.1-7 (Figure 6A and B). Indeed, we observed a comparable
shift in the potency of PTK2 degradation for both PROTACs (Table 2), consistent with efflux
contributing to their reduced degradation efficacy in this cell line. Of note, CsnA only had a
minor effect on PTK2 levels (Figure 6C).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 52
Popow et al.
16
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 5. Cumulative analysis of half-maximal effective concentrations (DC , A) and maximal
50
degree of PTK2 degradation (Dmax, B) achieved by BI-0319 or BI-3663. pDC and Dmax at 18
5
0
h were determined as indicated in supplemental figures 2 and 3.
ACS Paragon Plus Environment

Page 17 of 52
Journal of Medicinal Chemistry
Popow et al.
17
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 6. Analysis of PTK2 degradation in Hep3B2.1-7 cells. A) cells were treated with the
indicated concentrations of BI-0319 for 18 h in presence (red triangles) or absence (blue circles)
of 10 µM cyclosporine A (CsnA). B) cells were treated for 18 h with the indicated concentrations
of BI-3663 in presence (red triangles) or absence (green circles) of 10 µM CsnA. C) cells were
treated with 10 µM CsnA for 18 h. D) Time course analysis of PTK2 levels in cells treated with
3
µM BI-0319. E) Time course analysis of PTK2 levels in cells treated with 3 µM BI-3663.
Values state PTK2 protein levels normalized to GAPDH relative to DMSO (or CsnA only)
controls.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 52
Popow et al.
18
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 2. PTK2 degradation data of BI-3663 and BI-0319 in A549 and Hep3B2.1-7 cells.
A549, 18 h*
Hep3B2.1-7, 18 h *
Hep3B2.1-7, 18 h *
10 µM CsnA
+
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pDC50
D
max
pDC50
D
max [%]
pDC50
D
max [%]
[%]
6
8
(BI- 3663)
(BI -0319)
7.6 ± 0.1
6.7 ± 0.4
95 ± 4
80 ± 9
7.9
6.2
96
59
9.0
7.4
94
88
*
Degradation activity is reported as concentration needed to achieve 50 % PTK2 protein
degradation (pDC50 ± STDEV ) and maximal achievable protein degradation (Dmax) relative to
DMSO.
PROTACs 6 and 8 differ considerably with respect to their cellular degradation rate in
Hep3B2.1-7 cells. The CRBN-based PROTAC 6 achieved complete degradation of PTK2 after
five hours whereas the VHL-based PROTAC 8 achieved maximal degradation only after 20
hours (Figure 6D and E). Of note, the CRBN-based PROTAC 6 was considerably less stable in
cell assay buffer containing 10% FCS (M+18 and +32 observed) than the VHL-based PROTAC
8
(Supplemental Table 1). PROTAC 6 was found to be stable as a solid and in DMSO stock
solution (>3 month, data not shown). Despite this previously reported instability of CRBN based
4
5
PROTACs PROATC 6 as well as PROTAC 8 showed comparable maximal degradation of
PTK2 after 18 h and 72h days incubation (Supplemental figure 6). Taken together, this indicates
that the both PROTACs 6 and 8 are suitable probes to further study the role of PTK2 in a broad
range of tumour cell lines.
ACS Paragon Plus Environment

Page 19 of 52
Journal of Medicinal Chemistry
Popow et al.
19
1
2
3
4
5
6
7
8
9
Reasoning that additional targeting of the scaffolding functions of PTK2 by PROTAC-
mediated degradation might result in enhanced efficacy as compared to inhibition of its kinase
activity, we next tested the effect of PROTAC-mediated PTK2 degradation on proliferation of
the same HCC cell lines. Surprisingly, efficient depletion of PTK2 by both PROTACs (Table 3
and Supplemental figures 2 and 3) did not affect proliferation of the tested cell lines more
severely than the PTK2 inhibitor BI-4464 in standard culture conditions or in anchorage-
independent growth assays (Table 3, Supplemental figures 4 and 5). Notably, sensitivity to
genetic depletion of PTK2 has been described for several of the tested cell lines (Supplemental
table 4). One explanation may be that effects on proliferation may require long-term depletion of
PTK2. To further address this, we treated cell lines with reported high sensitivity to genetic
depletion of PTK236, 46 (Supplemental table 4) for 21 days with either PROTAC 6 (BI-3663) or 8
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(BI-0319) or the PTK2 inhibitor 1 (BI-4464). Despite the effective depletion of PTK2 even after
21 days, we did not observe any pronounced anti-proliferative effects for any of the tested cell
lines (Supplemental figure 6). Notably, CRISPR-based genetic validation in cultured HuH-1
cells also resulted in a weak dependency on PTK2 (Supplemental figure 7). Taken together, these
data questions the scaffolding function of PTK2 being required for in vitro proliferation of the
tested cell lines beyond the effect of inhibition of its kinase activity. Nevertheless, our
observations indicate that both probes can be used to effectively reduce PTK2 protein levels in
long term experiments in cultured cells.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 52
Popow et al.
20
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 3. Degradation characteristics and effect on proliferation of BI-3663 (6) and BI-0319 (8)
in HCC lines.
BI-3663
Dmax
[%]
BI-0319
Dmax
BI-4464
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Cell line
,
pIC50
(proliferation)
<4.6
,
pIC50
pIC50
pDC50
pDC50
[%] (proliferation) (proliferation)
SNU-387
HUH-1
7.6
6.6
7.9
7.5
7.5
6.4
8.5
7.9
6.8
7.3
7.9
90.0
50.0
96.0
89.0
89.0
30.0
95.0
90.0
79.0
93.0
93.0
7.5
7.1
6.2
6.2
6.9
7.4
8.0
6.9
7.6
7.0
7.1
98.0
79.0
59.0
65.0
87.0
66.0
99.0
86.0
85.0
79.0
87.0
5.3
4.7
5.2
5.4
5.3
5.5
5.2
5.4
5.2
5.2
5.1
5.4
5.4
4.6
Hep3B2.1-7
HepG2
4.6
5.2
<4.6
<4.6
5.1
SK-Hep-1
HLF
5.8
<4.6
<4.6
5.1
SNU-398
HUCCT1
HLE
<4.6
<4.6
5.2
<4.6
5.0
HuH-7
<4.6
5.4
SNU-423
4.7
5.2
Conclusion
We describe the development of two highly selective and functional PROTACs to degrade the
PTK2 protein, a kinase of significant relevance to cancer research, in particular, in the area of
hepatocellular carcinoma. Structure-guided conjugation of a highly selective PTK2 inhibitor BI-
4
464 to either a CRBN or VHL ligand via polyethylene glycol linkers led to the selective PTK2
degraders BI-0319 and BI-3663, respectively, using orthogonal E3 ligase binders. Both
PROTACs were characterized with respect to in vitro PTK2 engagement, ligase dependence
ACS Paragon Plus Environment

Page 21 of 52
Journal of Medicinal Chemistry
Popow et al.
21
1
2
3
4
5
6
7
8
9
selectivity (kinase inhibition as well as whole cell proteome) and degradation efficacy (DC and
5
0
Dmax) in twelve cell lines (one lung cancer and eleven hepatocellular carcinoma cell lines). Both
PROTACs are highly selective E3 ligase-dependent PTK2 degraders and show overall
comparable potencies and efficacies of PTK2 degradation in the 12 cell tested lines. Whereas
PTK2 levels in cell lines such as SNU-398 were equally sensitive to treatment with both
PROTACs, other cell lines, such as Hep3B2.1-7 exhibited a more than ten-fold difference in
cellular potency. This difference in degradation efficacy in Hep3B2.1-7 cells was associated with
a higher degradation rate of BI-3663. Despite the efficient depletion of PTK2, treatment with
either PROTAC did not affect proliferation of the tested cell lines in either short or long term
assays in vitro (6 vs 21 days) beyond the effect achieved by a PTK2 kinase inhibitor. Taken
together, these data suggest that PROTACs 6 (BI-3663) and 8 (BI-0319) mediated PTK2
depletion is likely insufficient to affect proliferation in vitro under the conditions tested beyond
the effect of inhibition of its kinase activity. Nevertheless, both probes provide valuable tools to
effectively reduce PTK2 protein levels in experimental conditions that might be better suited to
reveal and differentiate between kinase- dependent and -independent functions of PTK2. BI-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
663, BI-0319 and its inactive control BI-4206 will be made available to all scientists via our
open innovation portal opnMe.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 52
Popow et al.
22
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
EXPERIMENTAL SECTION
Chemistry. Synthesis. Unless otherwise indicated, all reactions were carried out in standard
commercially available glassware using standard synthetic chemistry methods. Air-sensitive and
moisture-sensitive reactions were performed under an atmosphere of dry nitrogen or argon with
dried glassware. Commercial starting materials were used without further purification. Solvents
used for reactions were of commercial “dry”- or “extra-dry” or “analytical” grade. All other
solvents used were reagent grade.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
The thin layer chromatography is carried out on ready-made silica gel 60 TLC plates on glass
(
with fluorescence indicator F-254) made by Merck. The preparative high pressure
chromatography (RP-HPLC) is carried out on Gilson systems with columns made by Waters
names: Sunfire™ Prep C18, OBD™ 10 µm, 50x150 mm or XBridge™ Prep C18, OBD™ 10
µm, 50x150 mm) and YMC (names: Actus-Triart Prep C18, 5 µm, 30x50 mm). Different
(
gradients of MeCN/H O are used to elute the compounds, for acidic conditions 0.1 % HCOOH is
2
added to the water. For the chromatography under basic conditions the water is made alkaline as
follows: 5 mL NH HCO solution (158 g in 1 L H O) and 2 mL NH (28 % in H O) are
4
3
2
3
2
replenished to 1 L with H O.
2
All compounds had a purity >95% acording to HPLC and NMR analysis acquired with the
systems and parameters stated in the following. HPLC samples were analyzed on an Agilent
1
200 series LC system coupled with an Agilent 6140 mass spectrometer. Purity was determined
via UV detection with a bandwidth of 170nm in the range from 230-400nm. LC parameters were
as follows: Waters Xbridge C18 column, 2.5µm particle size, 2.1 x 20mm. Total run time 3.1
minutes, flow 1ml/min, column temperature 60°C and 5µl injections. Solvent A (20mM
ACS Paragon Plus Environment

Page 23 of 52
Journal of Medicinal Chemistry
Popow et al.
23
1
2
3
4
5
6
7
8
9
NH HCO / NH pH 9), solvent B (MS grade acetonitrile). Start 10% B, gradient 10% - 95% B
4
3
3
from 0.0 - 1.5min, 95% B from 1.5 - 2.0min, gradient 95% - 10% B from 2.0 – 2.1min.
NMR experiments were recorded on a Bruker Avance HD 500 MHz spectrometer equipped
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
with a TCI cryoprobe at 298 K. Samples were dissolved in 600 μL DMSO-d and TMS was
6
1
added as an internal standard. 1D H spectra were acquired with 30° excitation pulses and an
1
3
interpulse delay of 4.2 sec with 64k data points and 20 ppm sweep width. 1D C spectra were
acquired with broadband composite pulse decoupling (WALTZ16) and an interpulse delay of 3.3
sec with 64 k data points and a sweep width of 240 ppm. Processing and analysis of 1D spectra
was performed with Bruker Topspin 3.2 software. No zero filling was performed and spectra
were manually integrated after automatic baseline correction. Chemical shifts are reported in
ppm on the δ scale.
HSQC spectra were recorded on all samples to aid the interpretation of the data and to identify
signals hidden underneath solvent peaks. Spectra were acquired with sweep widths obtained by
automatic sweep width detection from 1D reference spectra in the direct dimension with 1k
datapoints and with 210 ppm, and 256 datapoints in the indirect dimension.
For HRMS a LTQ Orbitrap XL (Thermo Scientific) coupled with a Triversa Nanomate
Nanospray ion source (ADVION Bioscience Inc.) was used.
3
-methoxy-4-({4-[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-
yl}amino)benzoic
acid
(1a).
4-{[4-chloro-5-(trifluoromethyl)pyrimidin-2-yl]amino}-3-
47
methoxybenzoic acid (4 , 1.00g, 1eq) and 7-hydroxy-2,3-dihydro-1H-inden-1-one (879 mg, 2
eq) were taken up in dioxane (10 mL), then Cs CO (4.686 g, 5 eq) was added. The reaction
2
3
mixture was stirred at 80°C for 16 h. The reaction mixture was diluted with H O and acetonitrile
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 52
Popow et al.
24
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
and concentrated under reduced pressure. The residue was purified by preparative RP-HPLC
under basic conditions using MeCN/H2O as eluents in a gradient from 20:80 to 65:35 over 12
min. (column: XBridge™ Prep C18, OBD™ 10 µm, 50x150 mm; flow: 110 mL/min). Product
containing fractions were freeze dried to give 3-methoxy-4-({4-[(3-oxo-2,3-dihydro-1H-inden-4-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
yl)oxy]-5-(trifluoromethyl)pyrimidin-2-yl}amino)benzoic acid (1a, 1.03 g, 78% yield) as brown
1
solid. H NMR (500 MHz, DMSO-d ) δ = 12.85 (br s, 1H), 8.86 (s, 1H), 8.74 (s, 1H), 7.80 (t,
6
J=7.88 Hz, 1H), 7.60 (d, J=7.88 Hz, 1H), 7.41 (d, J=0.95 Hz, 1H), 7.32 (br d, J=1.00 Hz, 1H),
7
.24 (d, J=7.88 Hz, 1H), 7.15 (br d, J=1.00 Hz, 1H), 3.80 (s, 3H), 3.05-3.17 (m, 2H), two protons
1
3
under DMSO. C NMR (125 MHz, DMSO-d ) δ = 203.4, 167.3, 166.4, 161.0, 158.4, 157.8,
6
149.7, 148.0, 137.0, 131.7, 128.9, 126.5, 125.5, 121.8, 121.1, 123.9, 120.4, 111.7, 101.5, 56.3,
+
3
6.7, 25.8. HRMS (m/z): [M+H] calculated for C H F N O , 459.10421; found, 459.10411.
2
2
16
3
3
5
t
HPLC-MS R = 1.00 min.
3
-methoxy-N-(1-methylpiperidin-4-yl)-4-({4-[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-
(trifluoromethyl)pyrimidin-2-yl}amino)benzamide (1, BI-4464). In a glass vial, 3-methoxy-4-
{4-[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-yl}amino)benzoic
(
acid (1a, 100 mg, 1 eq) was dissolved in DMF (1 mL). DIPEA (86 µl, 3 eq) and HATU (114 mg,
1.5 eq) were added. After the reaction mixture was stirred at rt for 5 min, 1-methylpiperidin-4-
amine (46 mg, 52 µl, 2 eq) was added. The mixture was stirred at rt for 2 h. The reaction mixture
was diluted with MeCN and H O and filtered through a syringe filter prior to purification via RP-
2
HPLC under basic conditions using MeCN/H2O as eluents in a gradient from 30:70 to 98:2 over
8
min (column: YMC Actus-Triart Prep C18, 5 µm, 30x50 mm; flow: 50 mL/min). Product
containing fractions were freeze dried to give 3-methoxy-N-(1-methylpiperidin-4-yl)-4-({4-[(3-
oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-yl}amino)benzamide
(1,
ACS Paragon Plus Environment

Page 25 of 52
Journal of Medicinal Chemistry
Popow et al.
25
1
2
3
4
5
6
7
8
9
1
(
BI-4464), 83 mg, 75% yield) as off-white lyophilizate. H NMR (500 MHz, DMSO-d ) δ = 8.83
6
(
br s, 1H), 8.69 (s, 1H), 8.13 (d, J=7.57 Hz, 1H), 7.77 (dd, J=7.57, 7.88 Hz, 1H), 7.55 (d, J=7.57
Hz, 1H), 7.38 (d, J=0.95 Hz, 1H), 7.22 (d, J=7.88 Hz, 1H), 7.27 (br s, 1H), 7.12 (br s, 1H), 3.78
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
s, 3H), 3.66-3.75 (m, 1H), 3.04-3.15 (m, 2H), 2.78 (br d, J=11.35 Hz, 2H), 2.16 (s, 3H), 1.85-
1
3
1
.98 (m, 2H), 1.69-1.80 (m, 2H), 1.49-1.65 (m, 2H), two protons under DMSO. C NMR (125
MHz, DMSO-d ) δ = 201.3, 164.2, 163.2, 159.1, 156.2, 155.7, 148.0, 145.8, 134.8, 128.8, 127.8,
6
1
26.8, 123.3, 118.8, 121.9, 117.5, 108.1, 99.0, 54.2, 53.0, 45.0, 44.4, 34.6, 29.9, 23.7, one carbon
+
not detected. HRMS (m/z): [M+H] calculated for C H F N O , 555.20934; found, 555.20824.
2
8
28
3
5
4
t
HPLC-MS R = 1.39 min.
3-{2-[2-(2-azidoethoxy)ethoxy]ethoxy}-N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-
dihydro-1H-isoindol-4-yl]propanamide (5a). In a 50 mL round-bottom flask, 4-amino-2-(2,6-
dioxo-3-piperidyl)isoindoline-1,3-dione (500 mg, 1 eq) was dissolved in DMF (5 mL) and
cooled to 0°C in an ice bath. 3-{2-[2-(2-azidoethoxy)ethoxy]ethoxy}propanoic acid (905 mg, 2
eq) was added dropwise. Then N-propylphosphonic acid anhydride, cyclic trimer (7.0 g, 6.5 mL,
6
eq) and pyridine (1.5 mL, 10 eq) were added. The mixture was stirred at 80°C for 3h. The
reaction mixture was diluted with ACN/H O. The solution was filtered through a syringe filter
2
and purified by preparative RP-HPLC under basic conditions using MeCN/H O as eluents in a
2
gradient from 10:90 to 60:40 over 9 min. (column: XBridge™ Prep C18, OBD™ 10 µm, 50x150
mm; flow: 150 mL/min). Product containing fractions were freeze dried to give 3-{2-[2-(2-
azidoethoxy)ethoxy]ethoxy}-N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-1H-isoindol-
1
4
-yl]propanamide (5a, 697 mg, 76% yield) as yellow lyophylizate. H NMR (500 MHz, DMSO-
d6) δ = 11.15 (s, 1H), 9.88 (s, 1H), 8.54 (d, J=8.31 Hz, 1H), 7.83 (dd, J=7.26, 8.31 Hz, 1H), 7.62
(
d, J=7.26 Hz, 1H), 5.14 (dd, J=5.52, 12.77 Hz, 1H), 3.74 (t, J=5.83 Hz, 2H), 3.52-3.61 (m, 6H),
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 52
Popow et al.
26
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
3
.47-3.52 (m, 4H), 3.34-3.37 (m, 2H), 2.84-2.96 (m, 1H), 2.70 (t, J=5.99 Hz, 2H), 2.53-2.66 (m,
1
3
2
H), 2.00-2.12 (m, 1H). C NMR (125 MHz, DMSO-d ) δ = 173.2, 170.9, 170.2, 168.1, 167.1,
6
136.9, 136.6, 131.9, 126.5, 118.7, 117.1, 70.2, 70.2, 70.1, 70.1, 69.7, 66.6, 50.4, 49.4, 38.0, 31.4,
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
+
2
2.4. HRMS (m/z): [M+H] calculated for C H N O , 502.18121; found, 502.18114. HPLC-
2
2
26
6
8
t
MS R = 1.05 min.
3
-{2-[2-(2-aminoethoxy)ethoxy]ethoxy}-N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-
dihydro-1H-isoindol-4-yl]propanamide (5b). In Parr-reactor 3-{2-[2-(2-
azidoethoxy)ethoxy]ethoxy}-N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-1H-isoindol-
-yl]propanamide (5a, 190 mg, 1 eq) was dissolved in methanol (5 mL) and Pd/C (10%, 40 mg,
0.1 eq) was added. The reactor was flushed with N and filled with H (6 bar). The reaction
a
4
2
2
mixture was stirred at rt for 3 h. A spoon of celite was added to the reaction mixture, the catalyst
was filtered off through a celite pad and rinsed with DCM and MeOH. The filtrate was
concentrated under reduced pressure. 3-{2-[2-(2-aminoethoxy)ethoxy]ethoxy}-N-[2-(2,6-
dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-1H-isoindol-4-yl]propanamide (5b, 162 mg, 90%
yield) was obtained as yellow resin. The product was taken to the next step without further
1
purification. H NMR (500 MHz, DMSO-d ) δ = ppm 9.89 (br s, 1 H), 8.54 (d, J=8.20 Hz, 1 H),
6
8.36 (s, 1 H), 7.84 (dd, J=8.20, 7.25 Hz, 1 H), 7.62 (br d, J=7.25 Hz, 1 H), 5.14 (dd, J=12.93,
5
.36 Hz, 1 H), 3.71 - 3.77 (m, 2 H), 3.45 - 3.60 (m, overlapped), 2.85 - 2.96 (m, 2 H), 2.82 (br t,
1
3
J=5.36 Hz, 2 H), 2.71 (br t, J=5.83 Hz, 2 H), 2.02 - 2.11 (m, 1 H). C NMR (125 MHz, DMSO-
d6) δ = 173.3, 170.9, 170.3, 168.1, 167.1, 136.9, 136.7, 131.9, 126.5, 118.8, 117.2, 70.2, 70.1,
+
7
0.1, 70.0, 69.2, 66.6, 49.4, 39.8, 38.0, 31.4, 22.4. HRMS (m/z): [M+H] calculated for
t
C H N O , 476.19071; found, 476.19041. HPLC-MS R = 0.81 min.
2
2
28
4
8
ACS Paragon Plus Environment

Page 27 of 52
Journal of Medicinal Chemistry
Popow et al.
27
1
2
3
4
5
6
7
8
9
N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-1H-isoindol-4-yl]-3-{2-[2-(2-{[3-
methoxy-4-({4-[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-
yl}amino)phenyl]formamido}ethoxy)ethoxy]ethoxy}propanamide (6, BI-3663). In a glass
vial, 3-methoxy-4-({4-[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-
yl}amino)benzoic acid (2, 170 mg, 1 eq) was dissolved in DMF (2 mL). DIPEA (150 µL, 3 eq)
and HATU (155 mg, 1.2 eq) were added. The reaction mixture was stirred at rt for 5 min, then 3-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
{
2-[2-(2-aminoethoxy)ethoxy]ethoxy}-N-[2-(2,6-dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-
H-isoindol-4-yl]propanamide (5b, 162 mg, 1eq) was added. The mixture was stirred at rt
1
overnight. The reaction mixture was diluted with MeCN and H O and filtered through a syringe
2
filter prior to purification via RP-HPLC under acidic conditions using MeCN/H2O as eluents in a
gradient from 30:70 to 98:2 over 8 min (column: YMC Actus-Triart Prep C18, 5 µm, 30x50 mm;
flow: 50 mL/min). Product containing fractions were freeze dried to give N-[2-(2,6-
dioxopiperidin-3-yl)-1,3-dioxo-2,3-dihydro-1H-isoindol-4-yl]-3-{2-[2-(2-{[3-methoxy-4-({4-
[(3-oxo-2,3-dihydro-1H-inden-4-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-
yl}amino)phenyl]formamido}ethoxy)ethoxy]ethoxy}propanamide (6 (BI-3663), 32 mg, 10%
1
yield) as off-white lyophilizate. H NMR (500 MHz, DMSO-d ) δ = 11.13 (br s, 1H), 9.86 (s,
6
1
H), 8.79 (br s, 1H), 8.70 (s, 1H), 8.53 (d, J=8.51 Hz, 1H), 8.43 (t, J=5.52 Hz, 1H), 7.81 (dd,
J=7.25, 8.51 Hz, 1H), 7.77 (t, J=7.88 Hz, 1H), 7.60 (d, J=7.25 Hz, 1H), 7.56 (d, J=7.88 Hz, 1H),
.40 (s, 1H), 7.21 (d, J=7.88 Hz, 1H), 7.25 (br s, 1H), 7.10 (br s, 1H), 5.14 (dd, J=5.52, 12.77
Hz, 1H), 3.78 (s, 3H), 3.71 (t, J=5.99 Hz, 2H), 3.43-3.61 (m, 10H), 3.36-3.41 (m, 2H), 3.07-3.15
m, 2H), 2.84-2.95 (m, 1H), 2.68 (t, J=5.99 Hz, 2H), 2.52-2.65 (m, 2H), 2.02-2.11 (m, 1H), two
7
(
1
3
protons under DMSO. C NMR (125 MHz, DMSO-d ) δ = 203.4, 173.2, 170.9, 170.3, 168.1,
6
167.1, 166.4, 165.9, 161.1, 158.3, 157.8, 149.9, 147.9, 136.9, 136.9, 136.6, 131.9, 130.4, 130.0,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 52
Popow et al.
28
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
6
28.9, 126.4, 125.5, 120.9, 124.0, 119.5, 118.7, 117.1, 110.2, 101.2, 70.2, 70.2, 70.1, 70.0, 69.5,
6.6, 56.3, 49.4, 39.6, 38.0, 36.8, 31.4, 25.8, 22.4, one carbon not detected. HRMS (m/z):
+
t
[M+H] calculated for C H F N O , 917.28435; found, 917.28471. HPLC-MS R = 1.35 min.
4
4
42
3
7
12
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(2S,4R)-1-[(2S)-2-(2-{2-[2-(2-azidoethoxy)ethoxy]ethoxy}acetamido)-3,3-dimethyl-
butanoyl]-4-hydroxy-N-{[4-(4-methyl-1,3-thiazol-5-yl)phenyl]methyl}pyrrolidine-2-
carboxamide (7a).
(2S,4R)-1-[(2S)-2-amino-3,3-dimethyl-butanoyl]-4-hydroxy-N-[[4-(4-
methylthiazol-5-yl)phenyl]methyl]pyrrolidine-2-carboxamide hydrochloride (2, 1.000 g, 1 eq)
was dissolved in DMF (5 mL). HATU (855 mg, 1.05 eq) and DIPEA (1.28 mL, 3.5 eq) were
added. The reaction mixture was stirred for 15 min, then 2-[2-[2-(2-
azidoethoxy)ethoxy]ethoxy]acetic acid (524 mg, 1.05 eq) was added. The reaction mixture was
stirred at rt for 1 h. The reaction mixture was diluted with EtOAc and washed with a saturated
solution of NaHCO and then with brine. The organic layer is dried over MgSO and evaporated.
3
4
The residue was purified via RP-HPLC under acidic conditions using MeCN/H O as eluents in a
2
gradient from 10:90 to 98:2 over 10 min (column: Sunfire™ Prep C18, OBD™ 10 µm, 50x150
mm; flow: 120 mL/min). Product containing fractions were freeze dried to give (2S,4R)-1-[(2S)-
2
-(2-{2-[2-(2-azidoethoxy)ethoxy]ethoxy}acetamido)-3,3-dimethylbutanoyl]-4-hydroxy-N-{[4-
(4-methyl-1,3-thiazol-5-yl)phenyl]methyl}pyrrolidine-2-carboxamide (7a, 870 mg, 63% yield).
1
H NMR (500 MHz, DMSO-d ) δ = 8.98 (s, 1H), 8.60 (t, J=5.99 Hz, 1H), 7.35-7.47 (m, 5H),
6
4
.67-5.61 (m, 1H), 4.57 (d, J=9.77 Hz, 1H), 4.32-4.48 (m, 3H), 4.21-4.29 (m, 1H), 3.97 (s, 2H),
3.64-3.69 (m, 1H), 3.53-3.64 (m, 12H), 2.44 (s, 3H), 2.02-2.09 (m, 1H), 1.85-1.94 (m, 1H), 0.93-
13
0
1
.96 (m, 8H), 0.94 (s, 1H), one proton under water. C NMR (125 MHz, DMSO-d ) δ = 172.2,
6
69.6, 169.1, 152.0, 148.2, 139.9, 131.6, 130.2, 129.2, 127.9, 70.9, 70.4, 70.2, 70.1, 70.1, 69.7,
ACS Paragon Plus Environment

Page 29 of 52
Journal of Medicinal Chemistry
Popow et al.
29
1
2
3
4
5
6
7
8
9
+
6
9.3, 59.2, 57.0, 56.1, 50.4, 42.1, 38.4, 36.2, 26.6, 16.4. HRMS (m/z): [M+H] calculated for
t
C H N O S, 645.29447; found, 645.29442. HPLC-MS R = 1.16 min.
3
0
43
7
7
(
2S,4R)-1-[(2S)-2-(2-{2-[2-(2-aminoethoxy)ethoxy]ethoxy}acetamido)-3,3-
dimethylbutanoyl]-4-hydroxy-N-{[4-(4-methyl-1,3-thiazol-5-yl)phenyl]methyl}pyrrolidine-
-carboxamide (7b). (2S,4R)-1-[(2S)-2-(2-{2-[2-(2-azidoethoxy)ethoxy]ethoxy}acetamido)-3,3-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
dimethylbutanoyl]-4-hydroxy-N-{[4-(4-methyl-1,3-thiazol-5-yl)phenyl]methyl}pyrrolidine-2-
carboxamide (7a, 1.475 g, 1 eq) was dissolved in MeOH (10 mL), Pd/C (250 mg, 10 %) was
added and the mixture was hydrogenated (5 bar H pressure) at rt for 2.5 h. The catalyst was
2
filtered off and the filtrate was purified via RP-HPLC under basic conditions using MeCN/H O
2
as eluents in a gradient from 10:90 to 90:10 over 10 min. (column: XBridge™ Prep C18, OBD™
1
0 µm, 50x150 mm; flow: 140 mL/min). Product containing fractions were freeze dried to give
(
2S,4R)-1-[(2S)-2-(2-{2-[2-(2-aminoethoxy)ethoxy]ethoxy}acetamido)-3,3-dimethylbutanoyl]-4-
hydroxy-N-{[4-(4-methyl-1,3-thiazol-5-yl)phenyl]methyl}pyrrolidine-2-carboxamide (7b, 1.166
1
g, 82% yield). H NMR (500 MHz, DMSO-d ) δ = 9.00 (s, 1H), 8.64 (t, J=6.00 Hz, 1H), 7.37-
6
7
.49 (m, 5H), 5.17 (br s, 1H), 4.58 (br d, J=9.46 Hz, 1H), 4.21-4.48 (m, 4H), 3.99 (s, 2H), 3.65-
.71 (m, 1H), 3.51-3.65 (m, 9H), 3.46 (t, J=5.50 Hz, 1H), 2.79 (t, J=5.67 Hz, 1H), 2.45 (s, 3H),
3
1
3
2.02-2.11 (m, 1H), 1.86-1.96 (m, 1H), 0.95 (s, 9H), two protons not detected (NH ). C NMR
2
(
125 MHz, DMSO-d ) δ = 172.2, 169.6, 169.1, 152.0, 148.2, 139.9, 131.6, 130.2, 129.2, 127.9,
6
7
0.9, 70.4, 70.2, 70.1, 70.1, 70.0, 69.3, 59.2, 57.1, 56.2, 42.1, 38.4, 36.2, 26.7, 16.4, one carbon
+
not detected. HRMS (m/z): [M+H] calculated for C H N O S, 619.30397; found, 619.30333.
3
0
45
5
7
t
HPLC-MS R = 1.00 min.
(
2S,4R)-4-hydroxy-1-[(2S)-2-(2-{2-[2-(2-{[3-methoxy-4-({4-[(3-oxo-2,3-dihydro-1H-inden-
4
-yl)oxy]-5-(trifluoromethyl)pyrimidin-2-yl}amino)phenyl]formamido}ethoxy)
ACS Paragon Plus Environment