Discovery and Characterisation of Highly Cooperative FAK-Degrading PROTACs

Article abstract of DOI:10.1002/anie.202109237

Focal adhesion kinase (FAK) is a key mediator of tumour progression and metastasis. To date, clinical trials of FAK inhibitors have reported disappointing efficacy for oncology indications. We report the design and characterisation of GSK215, a potent, selective, FAK-degrading Proteolysis Targeting Chimera (PROTAC) based on a binder for the VHL E3 ligase and the known FAK inhibitor VS-4718. X-ray crystallography revealed the molecular basis of the highly cooperative FAK-GSK215-VHL ternary complex, and GSK215 showed differentiated in-vitro pharmacology compared to VS-4718. In mice, a single dose of GSK215 induced rapid and prolonged FAK degradation, giving a long-lasting effect on FAK levels (≈96 h) and a marked PK/PD disconnect. This tool PROTAC molecule is expected to be useful for the study of FAK-degradation biology in vivo, and our results indicate that FAK degradation may be a differentiated clinical strategy versus FAK inhibition for the treatment of cancer.

Full text of DOI:10.1002/anie.202109237

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Discovery and Characterisation of Highly Cooperative FAK-
Degrading PROTACs
Authors: Robert P. Law, Joao Nunes, Chun-wa Chung, Marcus
Bantscheff, Karol Buda, Han Dai, John P Evans, Adam
Flinders, Diana Klimaszewska, Antonia J Lewis, Marcel
Muelbaier, Paul Scott-Stevens, Peter Stacey, Christopher J
Tame, Gillian F Watt, Nico Zinn, Markus A Queisser, John D
Harling, and Andrew B Benowitz
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202109237
Link to VoR: https://doi.org/10.1002/anie.202109237

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
Discovery and Characterisation of Highly Cooperative FAK-
Degrading PROTACs
Robert P. Law,[a]* Joao Nunes,[a] Chun-wa Chung,[a] Marcus Bantscheff,[b] Karol Buda,[a] Han Dai,[a]
John P. Evans,[a] Adam Flinders,[a] Diana Klimaszewska,[a] Antonia J. Lewis,[a] Marcel Muelbaier,[b] Paul
Scott-Stevens,[a] Peter Stacey,[a] Christopher J. Tame,[a] Gillian F. Watt,[a] Nico Zinn,[b] Markus A.
Queisser,[a] John D. Harling[a] and Andrew B. Benowitz[a]
[a]
Dr. R. P. Law, Dr. J. Nunes, Dr. C. Chung, K. Buda, Dr. H. Dai, Dr. J. Evans, A. Flinders, D. Klimaszewska, A.J. Lewis, Dr. M. A. Queisser, Dr. P. Scott-
Stevens, Dr. P. Stacey, Dr. C. J. Tame, Dr. G.F. Watt, Dr. J. D. Harling, and Dr. A. B. Benowitz
GlaxoSmithKline
Gunnels Wood Road, Stevenage, Hertfordshire, SG1 2NY (UK)
E-mail: robert.p.law@gsk.com
[b]
Dr M. Bantscheff, Dr. M. Muelbaier, Dr. N. Zinn
Cellzome GmbH, a GSK company
Meyerhofstraße 1, 69117 Heidelberg (Germany)
Supporting information for this article is given via a link at the end of the document.
Abstract:
survival.[6] Nuclear FAK activity also promotes cancer cell immune
evasion by increasing levels of immunoprotective T-regs and
decreasing levels of cytotoxic CD8+ T-cells, thereby inducing an
immunosuppressive tumour microenvironment.[7]
Focal adhesion kinase (FAK) is a key mediator of tumour progression
and metastasis. To date, clinical trials of FAK inhibitors have reported
disappointing efficacy for oncology indications. We report the design
and characterisation of GSK215, a potent, selective, FAK-degrading
Proteolysis Targeting Chimera (PROTAC) based on a binder for the
VHL E3 ligase and the known FAK inhibitor VS-4718. X-ray
crystallography revealed the molecular basis of the highly cooperative
FAK-GSK215-VHL ternary complex, and GSK215 showed
differentiated in vitro pharmacology compared to VS-4718. In mice, a
single dose of GSK215 induced rapid and prolonged FAK degradation,
giving a long-lasting effect on FAK levels (~96 h) and a marked PK/PD
disconnect. This tool PROTAC molecule is expected to be useful for
the study of FAK degradation biology in vivo, and our results indicate
that FAK degradation may be a differentiated clinical strategy versus
FAK inhibition for the treatment of cancer.
To date, several ATP-competitive inhibitors targeting FAK kinase
activity have been developed. Phase I trials have been conducted
with PF-00562271,[8] GSK2256098,[9] defactinib (VS-6063)[10]
and BI853520[11] in several solid tumour types. Despite good
clinical safety, tolerability and pharmacokinetic profiles, the best
responses in these single-agent trials were limited to stable
disease, with some agents showing no effect on progression-free
survival compared to placebo.[12] These results have prompted
the evaluation of FAK inhibitors in combination settings. FAK
inhibitors have been found to be synergistic with checkpoint
immunotherapy agents in mouse models of pancreatic cancer,[13]
and the FAK inhibitor VS-6063 is currently under clinical
evaluation in combination with the anti-PD-1 antibody
pembrolizumab.[14]
Introduction
In addition to the signal transduction activity of the FAK kinase
domain, several reports indicate that FAK scaffolding functions
also have profound effects on cancer cell survival and progression.
FAK knock-in mice with P878/881A mutations were shown to
exhibit reduced A2 endophilin binding to FAK, leading to a
reduction in tumour growth and metastasis.[15] Additionally,
disruption of VEGFR-3 interactions with the FAK FAT domain led
to apoptosis in breast cancer cells,[16] whereas EGF and PDGF
driven-cell motility was reduced by FAK knock-down but not
kinase inhibition.[17] FAK can also be phosphorylated by other
receptor tyrosine kinases, a potential resistance mechanism of
tumour cells against inhibition of Y397 autophosphorylation.[18]
The weight of evidence suggests that inhibition of FAK kinase
activity may not be sufficient to completely block FAK signaling,
and that targeting FAK scaffolding roles could prove to be a more
efficacious approach to treating solid tumours. However, known
inhibitors of FAK scaffolding functions show only weak potency,[19]
and retain the inherent limitations of occupancy-driven inhibitor
Focal Adhesion Kinase (FAK) is a non-receptor protein tyrosine
kinase that transduces signaling from integrins and growth factor
receptors to intracellular signaling pathways associated with cell
migration, invasion, survival and proliferation.[1] Consistent with
the importance of FAK in regulating such events, high levels of
FAK protein and mRNA have been reported in several solid
tumour types, and are associated with poor clinical outcomes.[1-2]
FAK drives cancer cell growth and invasion through kinase-
dependent phosphorylation of Y397 and by acting as a kinase-
independent scaffold for several signaling proteins.[1, 3] FAK is
localised to focal adhesions via binding of the FAT domain to LD
repeat motif-containing proteins,[4] while recruitment of integrins
and tyrosine kinases through the FAK FERM domain leads to
kinase activation via autophosphorylation of Y397.[5] In the
nucleus, the FAK FERM domain forms scaffolding interactions
with p53 and MDM2, degrading p53 to promote tumour cell
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
pharmacology. Therefore, addressing both the kinase and non-
kinase functions of FAK remains an intriguing prospect for anti-
cancer therapy.
A)
Proteolysis Targeting Chimeras (PROTACs) are an emerging
class of molecules that offer the possibility to modulate target
pharmacology that is inaccessible to molecules that possess only
inhibitory activity. PROTACs are heterobifunctional molecules
that redirect the ubiquitin proteasome system to degrade selected
proteins of interest (POI) by simultaneously recruiting both a POI
and an E3 ubiquitin ligase to promote ubiquitination and
subsequent degradation of the POI.[20] Due to the formation of
stabilising or destabilising protein-protein interactions, the binding
affinity of the PROTAC for the E3 ligase or POI may change once
bound to the other protein, an effect termed cooperativity. In
contrast to a target inhibition strategy, protein degradation has the
ability to modulate all functions of a particular POI, which is of
particular relevance to targets such as FAK which are known to
have important scaffolding functions. In addition, PROTACs can
induce a PD/PK disconnect that is dependent on the protein
synthesis rate,[21] and in principle they can target “undruggable”
proteins where only weak binders are known. Here we present
the discovery and characterization of a potent and selective FAK
PROTAC molecule, as well as the X-ray crystal structure of the
highly cooperative FAK-PROTAC-VHL ternary complex. We
show that the PROTAC has differentiated in vitro effects on
cancer cell growth, motility, and collagen I deposition in human
fibroblasts. We further demonstrate the rapid and prolonged
degradation of FAK in vivo after a single subcutaneous dose of
the PROTAC in mice. Collectively, this work highlights the ability
of FAK-degrading PROTACs to modulate non-enzymatic and
cancer-relevant functions of FAK, and describes FAK PROTAC
tool compounds for the further study of FAK in vitro and in vivo
biology.
B)
D)
Effect of linker length on pKD and log α
pDC50 correlates with log α
C)
10
8
10
9.0
Log

ternary
binary
8
6
4
2
0
8.5
8.0
7.5
7.0
6.5
6
4
0
5
10
15
0.5
1.0
1.5
2.0
2.5
Linker length
log alpha
SPR Titrations with FAK
SPR Titrations with FAK
E)
F)
10-6
10-7
10-8
10-9
100
80
60
40
20
0
- FAK
+
+
1
6
M FAK
M FAK
Results and Discussion
1
1
1
0
9
8
7
6
5
10-
10-
10-
10-
10-
10-
10-
Concentration (M)
PROTAC design and optimisation
FAK degradation using PROTACs was first reported using
promiscuous kinase binders.[22] Recently, more selective FAK
PROTACs were published separately by the Crews laboratory
(PROTAC-3),[23] Boehringer Ingelheim/University of Dundee (BI-
3663, BI-0319)[24] and Rao et al (FC-11)[25] (Scheme S1).
Although the reported molecules are potent degraders, their
effects on cell viability and detailed characterisation of the
requisite ternary complexes were not reported. Additionally, in
vivo FAK degradation has only been described using high doses
of PROTAC (50 mg/kg, or 20 mg/kg BID).[26] Here we report the
design of FAK PROTACs suitable for use as low-dose in vivo tool
molecules, and we reveal the structural determinants of the high
cooperativity () observed in the ternary complexes formed with
these compounds.
Figure 1. Investigating the effect of PROTAC linker length. a) Conceptual
design of an FAK-VHL PROTAC based on VS-4718; b) Linker identity, linker
length (number of atoms; #), FAK degradation pDC50, individual binary complex
and ternary complex binding of PROTACs to FAK and the VCB complex, and
ternary complex cooperativity (). For 3 and 4: pDC50 data generated using
formic acid salt and free base respectively, VHL and ternary data generated
using TFA salts; c) Comparison of linker length to VCB pKD (left axis, blue) and
cooperativity (right axis, red); d) correlation of cooperativity and degradation; e)
binding of GSK215 to VCB as measuring by SPR, showing increased affinity on
addition of FAK protein; f) KD values for binding of VHL binder (26) and PROTAC
GSK215 (± FAK Protein) to VCB, as measured by SPR.
We designed our PROTACs based on the clinical FAK inhibitor
VS-4718/PND-1186 (FAK pIC50 8.3)[27] and the von-Hippel Lindau
(VHL) E3 ligase ligand VHL-021.[28] We connected these binders
using a variety of linker lengths and chemotypes (See SI for
synthesis), and tested FAK degradation using an ELISA protein
quantification assay in A549 cells (Figure 1b). Recent reports
have suggested ternary complex kinetics and positively
cooperative binding can have significant effects on
degradation,[29] though increased cooperativity does not always
result in increased degradation.[30] To analyse ternary complex
formation, we established an FAK-VCB (VHL-ElonginC-ElonginB)
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
FRET assay capable of determining ternary KD, and calculated
Confirming proteasomal degradation
the ternary cooperativity α (α = binary KD / ternary KD).
To confirm that the mechanism of FAK degradation was
dependent upon E3 ligase complex recruitment, we demonstrated
reduced degradation of FAK upon separate addition of the FAK
inhibitor VS-4718 and the VHL ligand VHL-021 (Figure 3a),
indicating competitive displacement of PROTAC GSK215 from
the ternary complex. Epoxomicin treatment also hindered the
Encouragingly, we observed FAK degradation with a variety of
linkers, with all compounds showing strong positive cooperativity
(α > 1). Consistent with the complex SAR commonly seen with
PROTAC systems, there was no simple correlation between
linker length and VCB KD or cooperativity (Figure 1c). For example, ability of GSK215 to induce FAK degradation, as did separate
compounds 2 and 6 exhibited amongst the highest ternary
cooperativities reported to date despite having linkers with very
different lengths. However, the DC50 (half-maximal degradation)
values did correlate with α (Figure 1d), supporting the importance
of positive cooperativity in achieving potent degradation.
Interestingly, the most potent degradation was obtained with the
unusually short acetamide linker of 6 (hereafter referred to as
GSK215), and we selected this compound for further investigation.
treatment with MLN4924, a NEDD8 inhibitor that blocks cullin
scaffolding support for the E2 ubiquitin-conjugating enzyme
(Figure 3c). These experiments provided additional evidence for
proteasomal and neddylation dependent PROTAC-induced
substrate degradation.
Ternary complex formation is necessary
for FAK degradation
FAK Protac induced degradation is proteasome
and ubiquitin dependent
C)
A)
Epoxomicin
MLN4924
VS-4718
VHL-021
GSK215A (µM)
FAK
GSK215 (µM)
FAK
SPR analysis confirmed the ternary complex formation and high
cooperativity of GSK215 (Figure 1e-f), and a ~24-fold increase in
binding of this compound to VCB was measured in the presence
of saturating concentrations of FAK protein.
β-tubulin
p21
β-tubulin
B)
D)
DMSO
VS-4178
VHL-021
MLN4924
DMSO
150
125
100
75
150
125
100
75
A)
B)
Epoxomicin
PTK2
50
50
TNK1
25
25
RPS6KA1
PTK2B
0
0
DMSO
DMSO
GSK215 (µM)
GSK215 (µM)
Gene
Figure 3. Mechanistic experiments. a,b) Addition of competitive inhibitors
VS-4718 (1 μM) or VHL-021 (50 μM) reduces FAK degradation; c,d)
degradation is rescued by blocking proteasome activity (1 µM epoxomicin) or
neddylation (1 µM MLN4924).
C)
D)
Time (min)
150
100
50
GSK 215
150
100
50
ent-6
15
VS-4718
30
60
120
240
480
1440
DC50 1.3nM
Structure of PROTAC ternary complex
0
0
-11
-9
-8
-7
-6
-5
-10
-9
-8
-7
-6
-5
[GSK215] M
To better understand the structural determinants of the high
cooperativity observed in engagement of FAK and VHL-ElonginC-
ElonginB (VCB) by GSK215, a 2.2 Å structure of FAK-GSK215-
VCB was successfully elucidated by X-ray crystallography (Figure
4a, PDB ID: 7PI4). Crystallizing ternary complexes of PROTACs
bound to target proteins and E3 ligase complexes has proven to
be challenging, but several examples have been reported. The
largest number of VHL ternary complex structures have been
reported from bromodomain-containing proteins, such as
BRD4[29b] and SMARCA4,[31] with an additional structure of a non-
bromodomain protein also having been recently disclosed.[32]
Log [M] Compound
Figure 2. Structure and degradation profiling of FAK-VHL PROTAC GSK215.
a) Structure of GSK215; b) kinome binding selectivity of GSK215 as determined
by KinoBead profiling. Proteins with binding below the lower limit of
quantification are shown in grey; c) FAK levels in A549 cells as determined by
Western blot, after treatment with GSK215, enantiomeric control ent-6, and VS-
4718; d) time-course study of FAK degradation in A549 cells.
Profiling of GSK215 using a cell lysate kinobead assay showed
that the PROTAC maintained high selectivity for binding to the
FAK kinase domain. It should be noted that although the FAK
binding affinity of GSK215 is significantly reduced compared to
VS-4718 (pIC50 7.8 vs 8.3, Table S1), the catalytic mechanism of
the PROTAC allows it to induce potent FAK degradation. We
profiled FAK degradation in A549 non-small-cell lung cancer cells
using Western blot and determined a DC50 of 1.3 nM (Figure 2c)
and maximal degradation (Dmax) near the limit of the assay
quantification (Dmax 99%). To confirm degradation was PROTAC-
mediated, we synthesised the enantiomeric control ent-6 which is
unable to bind to VHL. Neither this compound nor the parent
inhibitor VS-4718 caused any decrease in FAK protein levels
(Figure 2c), indicating the critical role of E3 ligase recruitment for
degradation. We also observed a strong time-dependence effect
on degradation (Figure 2d), with a reduction in FAK levels
detectable after 30 min treatment with 0.1 μM GSK215, which
increased to >90% degradation after 2 hours.
The high ternary complex cooperativity induced by GSK215 can
be explained by many neo-interactions between FAK and VHL
that occur at both ends of the PROTAC molecule (Figure 4b). As
expected, the FAK binding moiety of the PROTAC is completely
enclosed by the FAK protein. However, beyond the FAK ligand
binding site, the VHL and FAK proteins make many neo-non-
cognate direct protein-protein interactions (PPIs; Figure 4c).
These contacts include Van-der-Waals interactions, as well as
direct H-bonds, such as that between the sidechain of N67 (VHL)
and the backbone carbonyl of C427 (FAK), and a salt bridge
between R69 (VHL) and E430 (FAK). At the VHL binding end of
the PROTAC ligand, the C-terminal helix of FAK 506-516 lies on
top of the VHL ligand, which now sits in an enclosed pocket rather
than in an exposed site normally seen with VHL binders.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
The total buried surface upon ternary complex formation is
2661 Å2, with the largest VHL:target protein neo-PPI interface
reported to date (721 Å2, Table S4), which is consistent with the
high measured cooperativity.
proteome dynamics profiling (mPDP). mPDP is a mass-
spectrometry-based approach, combining dynamic-SILAC
labelling with isobaric mass tagging for multiplexed analysis of
protein degradation and synthesis.[35] In mature proteins, GSK215
showed dose dependent degradation of FAK, with selective
degradation observed up to 10 nM GSK215 after 6 h (Figure 6a).
At concentrations above 100 nM levels of additional mature
proteins were reduced, primarily kinases CDK7, RPS6KA3, MET
and GAK. Interestingly, all members of the CAK complex (CDK7,
MNAT1 and CCNH) were degraded. These results are in contrast
to previously reported FAK degraders. PROTAC-3 primarily
showed reduction of the androgen receptor, p-Akt and p-SRc,[23]
while FC-11 did not degrade CDK7,[26c] highlighting the potential
for degraders of differing structure to produce distinct phenotypic
effects. GSK215 showed binding to RPS6KA3 in the KinoBead
assay (Table S2), but binding to MET and GAK was below the
lower limit of quantification.
We were unable to obtain a X-ray crystal structure of 2, which also
demonstrated high cooperativity, bound to the FAK-VCB ternary
complex. Given the sharp turn created by the acetamide of
GSK215, and the close proximity of the FAK- and VHL-binding
portions, it is likely that longer linkers would produce different
ternary complex structures with alternative positioning of the two
proteins. Longer linkers can produce flexible ternary complexes
which adopt multiple conformations;[30, 33] we hypothesise that
while 2 can induce strongly cooperative ternary binding, the rigid
GSK215 structure is a key factor in its more effective degradation
of FAK. A recently reported FGFR-VHL PROTAC includes a
similar acetylpiperazine linker,[34] suggesting commonalities in
their binding mode and the possibility of transferrable SAR
between kinase PROTACs. This work reinforces the importance
of thoroughly surveying linker length and rigidity in PROTAC
design, and the effect small modifications can have on
cooperativity and DC50.
In addition, levels of nascent proteins were also reduced, such as
FAK and the CAK complex, presumably driven by PROTAC-
induced degradation of both mature and nascent protein. For
other proteins such as FOSL1, reduction was only seen on the
nascent level, indicative of a downstream regulatory effect (Figure
6b). After 24 h treatment (Figure 6d) the selectivity of degradation
for GSK215 was maintained up to 10 nM while at higher
concentrations additional targets of induced degradation were
observed, with pronounced downstream regulatory effects on
nascent protein levels at 100 nM.
A)
GSK215
Elongin C
VHL
Since FAK and the CAK complex subsequently influence cell-
cycle progression, we explored non-catalytic FAK biological roles.
To simultaneously discover potential differentiating effects of FAK
degradation versus kinase inhibition, we also tested GSK215
alongside VS-4718 and the corresponding non-degrading
enantiomeric control compound ent-6 in phenotypic assays.
GSK215 repressed 2D cell proliferation in A549 and MCF-7 cells
in a concentration dependent manner, but not in BT474 cells
(Figure 5a, Figure S5).
FAK
Elongin B
B)
FAK
Y516
S508
V513
R107
Interestingly, we obtained a different result in an anchorage
independent cell growth assay, in which inhibition of 3D
proliferation upon treatment with GSK215 was not observed in
A549 or MCF-7 cells (Figure S5), but was clearly seen in BT474
cells (Figure 5b). No inhibition of 3D proliferation was seen in any
of these cells upon treatment with ent-6 or VS-4718. These
results support previous findings that indicate diverse roles for
FAK in different growth formats, as well as significant variability
between cell lines.
Y98
C427
N67
R69
VHL
C)
D)
K515
D435
C427
V513
Y98
Y516
E430
E430
R69
A)
B)
R426
N67
BT474 Anchorage Independent Growth
BT474 2D proliferation
GSK215
VS-4718
GSK215
VS-4718
ent-6
D92
R107
T109
150
125
100
75
ent-6
125
100
75
50
25
0
F91
P99
50
Figure 4. 2.2Å X-ray crystal structure of the FAK-GSK215-VCB ternary
25
complex (PDB ID: 7PI4).
0
-9
-8
-7
-6
-5
-10
-9
-8
-7
-6
-5
Log [M] Compound
Log [M] Compound
GSK215 reveals differential effects of FAK degradation over
FAK inhibition
To further profile the PROTAC and discover potential targets of
induced degradation, we profiled GSK215 by multiplexed
Figure 5. Effect of GSK215, enantiomeric control ent-6 and VS-4718 on a) 2D
proliferation and b) anchorage independent growth in 3D culture media in
BT474 cells.
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
app
A
C
Kinase
Compound
VS-4718
GSK215
VS-4718
GSK215
VS-4718
GSK215
VS-4718
GSK215
VS-4718
GSK215
Mean pKD
5.6
NA
CDK7
< 5.2
< 5.2
6.5
GAK
MET
< 5.0
8.9
PTK2
7.0
6.2
RPS6KA3
5.2
B
D
Figure 6. Proteomics profiling of GSK215. a,b) Line chart of mature (a) and nascent (b) protein abundance changes with increasing concentration of GSK215
after 6 h of treatment. Red lines indicate significantly regulated proteins; c) Kinobeads affinity of GSK215 and VS-4718 against kinases degraded as mature
protein after 6 h; d) mPDP Scatterplot after 24 h treatment with 10 and 100 nM of GSK215 for mature and nascent proteins.
A)
B)
DMSO
ent-6
GSK215
VS-4718
VHL-021
100
80
60
40
20
0
600
400
200
0
0h
h8
48 1
-
S
h
48
48h
8
h
48
h
5
h
468
h
h
428
h
48
h
48
h
48
O
S
5
O
S
1
1
1
48
2
K
481
7
7
2
2
K
1
6
4
M
D
-
0
4
M
D
-
0
t
-
t
-
L
H
-
n
e
L
H
S
G
n
e
S
V
V
S
G
V
V
C)
D)
125
100
75
50
25
0
ent-6
VS-4718
GSK215
GSK215
ent-6
VS-4718
-10
-9
-8
-7
-6
-5
-4
Log [M] Compound
Figure 7. GSK215 inhibits migration, invasion and collagen deposition. a) Representative imaging of wound closure in A549 cell culture following compound
treatment (100 nM); b) Bar chart of wound closure assay results with GSK215 or inhibitor controls (100 nM, n=3); c) Representative imaging of fibroblast collagen
deposition following compound treatment; d) effect of GSK215, enantiomeric control ent-6 and VS-4718 on fibroblast collagen deposition.
To further probe the biological effects of FAK degradation versus
FAK inhibition, we studied A549 cell motility in a wound scratch
model system upon treatment with GSK215. We observed that
GSK215 treatment was better able to suppress cell motility when
compared to treatment with FAK kinase inhibitors. Indeed, only
treatment with GSK215 was able to repress simulated wound
closure, whereas treatment with ent-6 or VS-4718 gave results
similar to those seen with untreated cells (Figure 7a,b).
Furthermore, potent inhibition of collagen I deposition occurred
following treatment of fibroblasts with GSK215, with FAK inhibitor
controls showing significantly lower efficacy (Figure 7c,d). This
effect was independent of cell viability (Figure S6), supporting the
role of FAK in collagen I formation, and providing further evidence
for the differentiated biology of FAK degradation compared to FAK
inhibition.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
In vivo profile of FAK PROTAC GSK215
The pharmacokinetic and pharmacodynamic profile of GSK215
Acknowledgements
was determined in male CD1 mice following
a single
We are grateful to A. Cooney and K. Elverson for technical
support with compound synthesis, and to A. Konstantinou and
J. Eden for assistance running cell assays.
subcutaneous 8 mg/kg injection (Figure 8). A composite study
design, consisting of serial and terminal blood sampling was used
to determine the concentration-time profile of GSK215 and
facilitate measurement of FAK degradation in liver up to 96 h post
dose. Endogenous FAK levels in liver tissues were measured and
compared to untreated control animals to assess the amount of
FAK degradation over time. Figure 8 shows a simulated GSK215
concentration profile (see Supplementary Information). Following
SC injection, GSK215 was rapidly absorbed from the site of
administration with a measured mean Cmax of 526 ng/mL in blood
(Tmax 0.33 h), with GSK215 concentrations falling to below the
lower level of quantification (LLQ: 1 ng/mL) after 48 h. GSK215
caused a rapid and profound degradation of FAK in liver over time,
with a maximal degradation of ~85% being achieved within 18 h.
Endogenous FAK was found to still be reduced by ~60% at 96 h
post-dose, at which time the GSK215 exposure was below the
lower limit of quantification. This result highlights the potential for
protein degradation therapies to have extended PK/PD effects,
which may be due to the slow synthesis rate of the protein.[21]
Keywords: cancer • drug design • medicinal chemistry • protein
degradation • proteolysis-targeting chimeras (PROTACs)
References
[1]
[2]
[3]
F. J. Sulzmaier, C. Jean, D. D. Schlaepfer, Nat. Rev.
Cancer 2014, 14, 598-610.
V. M. Golubovskaya, F. A. Kweh, W. G. Cance, Histol.
Histopathol. 2009, 24, 503-510.
a) W. G. Cance, E. Kurenova, T. Marlowe, V.
Golubovskaya, Sci, Signaling 2013, 6, pe10; b) M. Roy-
Luzarraga, K. Hodivala-Dilke, Clin. Cancer Res. 2016, 22,
3718-3724.
[4]
[5]
M. D. Schaller, J. Cell Sci. 2010, 123, 1007-1013.
M. C. Frame, H. Patel, B. Serrels, D. Lietha, M. J. Eck, Nat.
Rev. Mol. Cell Biol. 2010, 11, 802-814.
[6]
[7]
S. T. Lim, X. L. Chen, Y. Lim, D. A. Hanson, T. T. Vo, K.
Howerton, N. Larocque, S. J. Fisher, D. D. Schlaepfer, D.
Ilic, Mol. Cell 2008, 29, 9-22.
A. Serrels, T. Lund, B. Serrels, A. Byron, R. C. McPherson,
A. von Kriegsheim, L. Gomez-Cuadrado, M. Canel, M. Muir,
J. E. Ring, E. Maniati, A. H. Sims, J. A. Pachter, V. G.
Brunton, N. Gilbert, S. M. Anderton, R. J. Nibbs, M. C.
Frame, Cell 2015, 163, 160-173.
120
100
80
60
40
20
0
500
400
300
200
100
0
Untreated
PD
PK
[8]
[9]
J. R. Infante, D. R. Camidge, L. R. Mileshkin, E. X. Chen,
R. J. Hicks, D. Rischin, H. Fingert, K. J. Pierce, H. Xu, W.
G. Roberts, S. M. Shreeve, H. A. Burris, L. L. Siu, J. Clin.
Oncol. 2012, 30, 1527-1533.
J. C. Soria, H. K. Gan, S. P. Blagden, R. Plummer, H. T.
Arkenau, M. Ranson, T. R. Evans, G. Zalcman, R. Bahleda,
A. Hollebecque, C. Lemech, E. Dean, J. Brown, D. Gibson,
V. Peddareddigari, S. Murray, N. Nebot, J. Mazumdar, L.
Swartz, K. R. Auger, R. A. Fleming, R. Singh, M. Millward,
Ann. Oncol. 2016, 27, 2268-2274.
***
*
***
***
***
***
**
**
0
24
48
72
96
Time after administration (hours)
[10]
[11]
[12]
S. F. Jones, L. L. Siu, J. C. Bendell, J. M. Cleary, A. R.
Razak, J. R. Infante, S. S. Pandya, P. L. Bedard, K. J.
Pierce, B. Houk, W. G. Roberts, S. M. Shreeve, G. I.
Shapiro, Invest. New Drugs 2015, 33, 1100-1107.
M. J. A. de Jonge, N. Steeghs, M. P. Lolkema, S. J. Hotte,
H. W. Hirte, D. A. J. van der Biessen, A. R. Abdul Razak, F.
De Vos, R. B. Verheijen, D. Schnell, L. C. Pronk, M. Jansen,
L. L. Siu, Targeted Oncology 2019, 14, 43-55.
a) A. Mohanty, R. R. Pharaon, A. Nam, S. Salgia, P.
Kulkarni, E. Massarelli, Expert Opin. Invest. Drugs 2020,
29, 399-409; b) G. Mak, J. C. Soria, S. P. Blagden, R.
Plummer, R. A. Fleming, N. Nebot, J. Zhang, J. Mazumdar,
D. Rogan, A. Gazzah, I. Rizzuto, A. Greystoke, L. Yan, J.
Tolson, K. R. Auger, H. T. Arkenau, Br. J. Cancer 2019,
120, 975-981.
H. Jiang, S. Hegde, B. L. Knolhoff, Y. Zhu, J. M. Herndon,
M. A. Meyer, T. M. Nywening, W. G. Hawkins, I. M. Shapiro,
D. T. Weaver, J. A. Pachter, A. Wang-Gillam, D. G.
DeNardo, Nat. Med. 2016, 22, 851-860.
ClinicalTrials.gov, "NCT02758587, Study of FAK
(Defactinib) and PD-1 (Pembrolizumab) Inhibition in
Advanced Solid Malignancies (FAK-PD1)", can be found
Figure 8. PK/PD of GSK215 (HCl salt) in naive CD1 mice following a single
8 mg/kg subcutaneous dose.
Conclusion
In summary, we have developed GSK215 as a potent and
selective in vitro and in vivo FAK degrader tool molecule which
shows a differentiated phenotypic response compared to FAK
inhibition in cancer cells. The high degradation potency of
GSK215 is derived from an unusually short and rigid linker, which
generates a highly cooperative ternary complex which has been
characterized by SPR and X-ray crystallography. GSK215 was
found to degrade FAK at low doses in vivo in mouse liver, and
produced a PK/PD disconnect, with >50% reduction in FAK levels
still observed up to 96 hours post-dose, long after the compound
has been cleared from blood. We anticipate that GSK215 will find
further use as an in vitro and in vivo tool compound to probe the
effects of FAK degradation, and to explore the potential
therapeutic utility of FAK degraders.
[13]
[14]
under
https://clinicaltrials.gov/ct2/show/NCT02758587,
2020.
[15]
[16]
[17]
H. Fan, X. Zhao, S. Sun, M. Luo, J. L. Guan, J. Biol. Chem.
2013, 288, 3322-3333.
C. A. Garces, E. V. Kurenova, V. M. Golubovskaya, W. G.
Cance, Cancer Res. 2006, 66, 1446-1454.
D. J. Sieg, C. R. Hauck, D. Ilic, C. K. Klingbeil, E. Schaefer,
C. H. Damsky, D. D. Schlaepfer, Nat. Cell Biol. 2000, 2,
249-256.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
[18]
[19]
[20]
T. A. Marlowe, F. L. Lenzo, S. A. Figel, A. T. Grapes, W. G.
Cance, Mol. Cancer Ther. 2016, 15, 3028-3039.
Ponthier, M. Ishoey, T. Zhang, J. D. Mancias, N. S. Gray, J.
E. Bradner, E. S. Fischer, Nat. Chem. Biol. 2018, 14, 706-
714.
G. Du, J. Jiang, Q. Wu, N. J. Henning, K. A. Donovan, H.
Yue, J. Che, W. Lu, E. S. Fischer, N. Bardeesy, T. Zhang,
N. S. Gray, Angew. Chem. Int. Ed. Engl. 2021, 60, 15905-
15911
M. M. Savitski, N. Zinn, M. Faelth-Savitski, D. Poeckel, S.
Gade, I. Becher, M. Muelbaier, A. J. Wagner, K. Strohmer,
T. Werner, S. Melchert, M. Petretich, A. Rutkowska, J.
Vappiani, H. Franken, M. Steidel, G. M. Sweetman, O.
Gilan, E. Y. N. Lam, M. A. Dawson, R. K. Prinjha, P. Grandi,
G. Bergamini, M. Bantscheff, Cell 2018, 173, 260-274, e25.
E. V. Kurenova, D. L. Hunt, D. He, A. T. Magis, D. A. Ostrov,
W. G. Cance, J. Med. Chem. 2009, 52, 4716-4724.
a) G. M. Burslem, C. M. Crews, Cell 2020, 181, 102-114; b)
M. Pettersson, C. M. Crews, Drug Discovery Today:
Technol. 2019, 31, 15-27; c) D. P. Bondeson, A. Mares, I.
E. Smith, E. Ko, S. Campos, A. H. Miah, K. E. Mulholland,
N. Routly, D. L. Buckley, J. L. Gustafson, N. Zinn, P. Grandi,
S. Shimamura, G. Bergamini, M. Faelth-Savitski, M.
Bantscheff, C. Cox, D. A. Gordon, R. R. Willard, J. J.
Flanagan, L. N. Casillas, B. J. Votta, W. den Besten, K.
Famm, L. Kruidenier, P. S. Carter, J. D. Harling, I.
Churcher, C. M. Crews, Nat. Chem. Biol. 2015, 11, 611-
617.
[34]
[35]
[21]
[22]
A. Mares, A. H. Miah, I. E. D. Smith, M. Rackham, A. R.
Thawani, J. Cryan, P. A. Haile, B. J. Votta, A. M. Beal, C.
Capriotti, M. A. Reilly, D. T. Fisher, N. Zinn, M. Bantscheff,
T. T. MacDonald, A. Vossenkamper, P. Dace, I. Churcher,
A. B. Benowitz, G. Watt, J. Denyer, P. Scott-Stevens, J. D.
Harling, Commun. Biol. 2020, 3, 140.
H. T. Huang, D. Dobrovolsky, J. Paulk, G. Yang, E. L.
Weisberg, Z. M. Doctor, D. L. Buckley, J. H. Cho, E. Ko, J.
Jang, K. Shi, H. G. Choi, J. D. Griffin, Y. Li, S. P. Treon, E.
S. Fischer, J. E. Bradner, L. Tan, N. S. Gray, Cell Chem.
Biol. 2018, 25, 88-99 e86.
[23]
[24]
P. M. Cromm, K. T. G. Samarasinghe, J. Hines, C. M.
Crews, J. Am. Chem. Soc. 2018, 140, 17019-17026.
J. Popow, H. Arnhof, G. Bader, H. Berger, A. Ciulli, D.
Covini, C. Dank, T. Gmaschitz, P. Greb, J. Karolyi-Ozguer,
M. Koegl, D. B. McConnell, M. Pearson, M. Rieger, J.
Rinnenthal, V. Roessler, A. Schrenk, M. Spina, S. Steurer,
N. Trainor, E. Traxler, C. Wieshofer, A. Zoephel, P.
Ettmayer, J. Med. Chem. 2019, 62, 2508-2520.
[25]
[26]
H. Gao, Y. Wu, Y. Sun, Y. Yang, G. Zhou, Y. Rao, ACS
Med. Chem. Lett. 2020, 11, 1855-1862.
a) C. M. Crews, P. M. Cromm, A. P. Crew (Yale University,
Arvinas Operations Inc.), WO/2020/023851, 2020; b) N. S.
Gray, B. Jiang, B. Nabet, T. Zhang, M. Hao (Dana-Farber
Cancer Institute, Inc), WO/2020/069117, 2020; c) H. Gao,
C. Zheng, J. Du, Y. Wu, Y. Sun, C. Han, K, Kee, Y. Rao,
Protein Cell 2020, 11, 534–539.
[27]
[28]
[29]
I. Tanjoni, C. Walsh, S. Uryu, A. Tomar, J. O. Nam, A.
Mielgo, S. T. Lim, C. Liang, M. Koenig, C. Sun, N. Patel, C.
Kwok, G. McMahon, D. G. Stupack, D. D. Schlaepfer,
Cancer Biol. Ther. 2010, 9, 764-777.
K. Raina, J. Lu, Y. Qian, M. Altieri, D. Gordon, A. M. Rossi,
J. Wang, X. Chen, H. Dong, K. Siu, J. D. Winkler, A. P.
Crew, C. M. Crews, K. G. Coleman, Proc. Natl. Acad. Sci.
U. S. A. 2016, 113, 7124-7129.
a) M. J. Roy, S. Winkler, S. J. Hughes, C. Whitworth, M.
Galant, W. Farnaby, K. Rumpel, A. Ciulli, ACS Chem. Biol.
2019, 14, 361-368; b) M. S. Gadd, A. Testa, X. Lucas, K. H.
Chan, W. Chen, D. J. Lamont, M. Zengerle, A. Ciulli, Nat.
Chem. Biol. 2017, 13, 514-521.
[30]
[31]
J. Schiemer, R. Horst, Y. Meng, J. I. Montgomery, Y. Xu, X.
Feng, K. Borzilleri, D. P. Uccello, C. Leverett, S. Brown, Y.
Che, M. F. Brown, M. M. Hayward, A. M. Gilbert, M. C. Noe,
M. F. Calabrese, Nat. Chem. Biol. 2021, 17, 152-160.
W. Farnaby, M. Koegl, M. J. Roy, C. Whitworth, E. Diers, N.
Trainor, D. Zollman, S. Steurer, J. Karolyi-Oezguer, C.
Riedmueller, T. Gmaschitz, J. Wachter, C. Dank, M. Galant,
B. Sharps, K. Rumpel, E. Traxler, T. Gerstberger, R.
Schnitzer, O. Petermann, P. Greb, H. Weinstabl, G. Bader,
A. Zoephel, A. Weiss-Puxbaum, K. Ehrenhofer-Wolfer, S.
Wohrle, G. Boehmelt, J. Rinnenthal, H. Arnhof, N.
Wiechens, M. Y. Wu, T. Owen-Hughes, P. Ettmayer, M.
Pearson, D. B. McConnell, A. Ciulli, Nat. Chem. Biol. 2019,
15, 672-680.
[32]
[33]
C. W. Chung, H. Dai, E. Fernandez, C. P. Tinworth, I.
Churcher, J. Cryan, J. Denyer, J. D. Harling, A. Konopacka,
M. A. Queisser, C. J. Tame, G. Watt, F. Jiang, D. Qian, A.
B. Benowitz, ACS Chem. Biol. 2020, 15, 2316-2323.
R. P. Nowak, S. L. DeAngelo, D. Buckley, Z. He, K. A.
Donovan, J. An, N. Safaee, M. P. Jedrychowski, C. M.
7
This article is protected by copyright. All rights reserved.

10.1002/anie.202109237
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
FAK
VHL
GSK215
FAK PROTAC
DC50 1.3 nM
A PROTAC with an unusually short linker potently degrades FAK. SPR and X-ray crystallography revealed a highly cooperative FAK-
PROTAC-VCB ternary complex, and FAK degradation showed enhanced effects on 3D cell growth compared to FAK inhibitors. In
mice, GSK215 induced rapid and sustained degradation of FAK with a profound PK/PD disconnect.
8
This article is protected by copyright. All rights reserved.