Small molecule-induced degradation of the full length and V7 truncated variant forms of human androgen receptor

Article abstract of DOI:10.1016/j.ejmech.2018.08.059

The androgen receptor (AR) is a hormone-activated transcription factor that regulates the development and progression of prostate cancer (PCa) and represents one of the most well-established drug targets. Currently clinically approved small molecule inhib

Full text of DOI:10.1016/j.ejmech.2018.08.059

Accepted Manuscript
Small molecule-induced degradation of the full length and V7 truncated variant forms
of human androgen receptor
Kush Dalal, Helene Morin, Fuqiang Ban, Ashley Shepherd, Michael Fernandez, Kevin
J. Tam, Huifang Li, Eric LeBlanc, Nathan Lack, Helge Prinz, Paul S. Rennie, Artem
Cherkasov
PII:
S0223-5234(18)30729-3
10.1016/j.ejmech.2018.08.059
EJMECH 10670
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 2 May 2018
Revised Date: 5 August 2018
Accepted Date: 21 August 2018
Please cite this article as: K. Dalal, H. Morin, F. Ban, A. Shepherd, M. Fernandez, K.J. Tam, H. Li, E.
LeBlanc, N. Lack, H. Prinz, P.S. Rennie, A. Cherkasov, Small molecule-induced degradation of the full
length and V7 truncated variant forms of human androgen receptor, European Journal of Medicinal
Chemistry (2018), doi: 10.1016/j.ejmech.2018.08.059.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. 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.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Small Molecule-Induced Degradation of the Full Length and V7
Truncated Variant Forms of Human Androgen Receptor
Kush Dalal1, Helene Morin1, Fuqiang Ban1, Ashley Shepherd1, Michael Fernandez1, Kevin J. Tam1,
Huifang Li1, Eric LeBlanc1, Nathan Lack1, Helge Prinz2, Paul S. Rennie1 and Artem Cherkasov1, *
1Vancouver Prostate Centre (VPC), 2660 Oak Street, Vancouver, British Columbia, V6H3Z6, Canada.
2Institute of Pharmaceutical and Medicinal Chemistry, Westfälische Wilhelms-Universität Münster,
Corrensstr. 48, D-48149 Münster, Germany.
*Corresponding author: Artem Cherkasov University of British Columbia. 2660 Oak Street,
Vancouver, British Columbia V6H 3Z6, Canada. acherkasov@prostatecentre.com
ABSTRACT
The androgen receptor (AR) is a hormone-activated transcription factor that regulates the
development and progression of prostate cancer (PCa) and represents one of the most well-established
drug targets. Currently clinically approved small molecule inhibitors of AR, such as enzalutamide, are
built upon a common chemical scaffold that interacts with the AR by the same mechanism of action.
These inhibitors eventually fail due to the emergence of drug-resistance in the form of AR mutations and
1

ACCEPTED MANUSCRIPT
expression of truncated AR splice variants (e.g. AR-V7) that are constitutively active, signalling the
progression of the castration-resistant state of the disease. The urgent need therefore continues for novel
classes of AR inhibitors that can overcome drug resistance, especially since AR signalling remains
important even in the late-stage advanced PCa.
Previously, we identified a collection of 10-benzylidene-10H-anthracen-9-ones that effectively
inhibit AR transcriptional activity, induce AR degradation and display some ability to block recruitment
of hormones to the receptor. In the current work, we extended the analysis of the lead compounds, and
used methods of both ligand- and structure-based drug design to develop a panel of novel 10-
benzylidene-10H-anthracen-9-one derivatives capable of suppressing transcriptional activity and protein
expression levels of both full length- and AR-V7 truncated forms of human androgen receptor.
Importantly, the developed compounds efficiently inhibited the growth of AR-V7 dependent prostate
cancer cell-lines which are completely resistant to all current anti-androgens.
2

ACCEPTED MANUSCRIPT
1. INTRODUCTION
The androgen receptor (AR) is a hormone inducible transcription factor that regulates the
expression of genes driving the progression of prostate cancer (PCa) (1, 2). Structurally, the AR is
organized into an N-terminus activation domain (NTD) followed by a DNA binding domain (DBD) and
ligand binding domain (LBD, Fig. 1, panel A)) (3). Typically, the AR is activated by androgen steroids,
such as dihydrotestosterone, which interacts at the hormone binding site on the AR-LBD to stimulate
nuclear translocation of the receptor, followed by its binding to the DNA and by initiation of
transcription of AR-target genes (1).
Such multi-faceted activation presents numerous opportunities for inhibition of the AR at several
stages in its normal cycle of expression, hormone binding and transcriptional initiation. Conventional
AR antagonists, such as enzalutamide, bicalutamide, flutamide, nilutamide and recently developed
darolutamide all interact at the AR-LBD (4-7), but many eventually fail due to point mutations in the
AR-LBD that convert the compounds into agonists (8). Notably, darolutamide has been an exception
that continues to inhibit the full-length AR despite the presence of resistance-causing mutations such as
F877L and T878A in the AR-LBD (6). Advanced PCa patients may also overexpress the full-length AR
as well as constitutively active splice variants of the receptor, such as ARV7, which entirely lacks an
LBD and, therefore, are non-sensitive to antagonists (Fig. 1, panel A) (9-11). Collectively, these
resistance mechanisms define the castration resistant stage of PCa (CRPC), characterized by poor
outcomes for patients (12-15), although second generation enzalutamide derivatives, such as
apalutamide, are effective to impede the progression to metastatic disease for CRPC tumors (16).
All conventional PCa therapeutic agents share the same common structural motifs facilitating
small molecule binding to the LBD, which explains their susceptibility to CRPC (1, 2, 17). Hence,
3

ACCEPTED MANUSCRIPT
principally novel classes of AR inhibitors are required to effectively block transcriptional activity of AR
mutants or ARV7, either by targeting the compounds beyond the AR-LBD or influencing the expression
or activation of the receptor or its cofactors. ARV7, and other variants, pose additional challenges to the
development of small molecule inhibitors since androgens are not required for nuclear translocation of
these truncated receptors (9). Additionally, ARVs preferentially interact with open chromatin, lacking
nucleosomes, to drive a unique transcriptional program (18). This chromatin interaction with ARV7
occurs with much greater affinity than full-length AR in cultured PCa cells (19), possibly explaining the
difficulty to identify inhibitors that disrupt the high affinity interaction.
We have previously identified a series of 10-benzylidene-10H-anthracen-9-ones, including
compounds VPC-3033 and VPC-3045 (Fig. 1, panel B) that could weakly displace androgens from the
AR-LBD, yet could rather effectively inhibit the transcriptional activity of the full length AR and
demonstrated effective suppression of the tumor growth of AR-positive LNCaP xenografts in mice (20).
Beyond this DHT displacement activity, the compounds promoted degradation of AR across multiple
PCa cell lines expressing the full-length AR (16). In the current work we extended the analysis of these
compounds and employed computer aided drug discovery (CADD) methodology to design novel 10-
benzylidene-10H-anthracen-9-one derivatives with enhanced ability to degrade all forms of AR.
Consequently, the compounds have been synthesized and investigated for their ability to suppress
transcriptional activity and expression levels of wild type- and, importantly, drug resistant truncated
forms of AR.
4

ACCEPTED MANUSCRIPT
2. RESULTS
Degradation of target proteins with small molecule agents, or hybrid compounds linked to
proteolysis-enabling protein domains, represents a trending avenue in modern drug discovery (21, 22).
Attempts to therapeutically degrade AR have also been recently made but failed to advance beyond the
clinical trial phase (23, 24). In our previous work (20), we have developed a series of 31 substituted 10-
benzylidene-10H-anthracen-9-ones as potential AR LBD binders and reported that some of them, such
as compounds VPC-3033 and VPC-3045 (Fig. 1) in addition to displacing DHT from the AR-LBD, also
exhibit intriguing ability to degrade AR. Such AR degradation activity represents a prospective novel
approach to abolish the AR function, especially in the context of ARV truncated splice variants recently
emerged as critical targets in CRPC for which there are no effective drugs (10, 13-15).
In the current work we have employed methods of CADD to separately model DHT displacing-
and AR degrading abilities of 10-benzylidene-10H-anthracen-9-ones and to develop compounds for
which the former activity is minimized, while the latter is, in contrast, enhanced. Thus, the objective of
this study is to develop small molecules that are ‘pure AR degraders’ and thus, they can eliminate
truncated AR variants along with a full length target protein.
2.1 CADD modeling of 10-benzylidene-10H-anthracen-9-one derivatives
Firstly, we built classification Quantitate Structure-Activity Relationship (QSAR) models to
describe a pure degrader - with IC50 eGFP < 5 µM and DHT displacement activity in the range 10-40
µM or higher - electronic induction properties of the compounds to predict strong LNCaP-eGFP
transcription inhibition and weak DHT displacement from the purified AR-LBD, respectively (see
5

ACCEPTED MANUSCRIPT
Methods). By applying these constraints on eGFP and DHT displacement activities, we could predict a
subset of potent AR and ARV7 inhibitors without tendency to target the LBD, and validation using a
receiver operator curve (ROC) analysis (Fig. S2). Multilinear regression (MLR) in equation 1 illustrates
a relevant linear dependency of electronic induction properties of the training set molecules in Table S1
and the pure degrader activity with cross-validation area-under-the-curve (AUC) of 0.76 as depicted in
the receiver operator curve (ROC) in Figure S2.
degrading ability = − 0.288 + 1.503×Max_Ave_EO_Pos5.5Å + 0.328×Max_Ave_EO_Pos7.0Å
0.269×Max_Ave_Hardness_Pos10.5 Å − 0.540×Max_Ave_Hardness_Neg11.0 Å
+
Eq. 1
AUCtrain = 0.80, AUCval = 0.76, Ntrain = 28
Equation 1 reveals that the degrader activities increase with electro-negativitiesꢀofꢀatoms
withꢀpositiveꢀpartialꢀcharge at interatomic distance of 5.5 and 7.0 and the
hardnessesꢀofꢀatomsꢀwithꢀpositiveꢀpartialꢀcharge at interatomic distance of 10.5 Å (Max_Ave_EO_Pos5.5Å
Max_Ave_EO_Pos7.0Å, Max_Ave_Hardness_Pos10.5Å). In turn, the degrader ability decreases with
harnessesꢀofꢀatomsꢀwithꢀnegativeꢀpartialꢀcharge at interatomic distance of 11.0
Å
Å
,
Å
(Max_Ave_Hardness_Neg11.0 Å). A detailed description of the inductive autocorrelation descriptors
appears in the Supplementary Information.
As computed by our model, the descriptors that approximated the score (degrading power) are
related to electronegative properties and hardness of prospective substituents. The amenable accuracy of
the MLR model encouraged us to explore electronegative and hard substituents to enhance the degrading
score while eliminating DHT displacement. Thus, to explore the effect of electronegative and hard
substituents, we developed substitutions of two chlorines at 1,8-positions of the anthracen scaffold of the
6

ACCEPTED MANUSCRIPT
original compounds (shown on Figure 1) and degrader score were computed using the QSAR model in
equation 1.
In addition to the QSAR predictions, ligand-protein docking was performed for the series of
original 10-benzylidene-10H-anthracen-9-ones in the training set and their 1,8-dichloro-anthracen
derivatives. The predicted binding pose of compound VPC-3033, representing the most AR active
compound in the training set according to its eGFP IC50, is shown in Figure 2A. The meta-hydroxy-
phenyl of VPC-3033 has a favorable "sandwich hydrophobic interactions" with Met745 and Phe764 as
shown (Fig. 2, panel A), which primarily accounts for its DHT displacement activity of an IC50 = 0.625-
2.5 µM as previously reported (20). In turn, 23 compounds of the thirty-one 1,8-di-chloro-analogues set
were not docked to the LBD, while the remaining eight docked derivatives showed unrealistic
alternative binding poses. The docking result suggests that the two chlorines would cause significant
steric clashes with the pocket residues of Leu873 and Thr877 as illustrated using di-chloro-VPC3033 as
an example in Figure 2B. Hence, we hypothesized that the addition of electronegative and hard chlorine
substituents to the 1,8-positions of the anthracen moiety should eliminate DHT displacement for some
analogs and direct their 1,8-dichloro-substituted analogs towards a pure degrader mechanism of action.
With the findings from docking and the predictions from the QSAR model in equation 1 for the
di-chloral analog set, four compounds were selected as potential degraders and one as an weak degrader
compound for medicinal chemistry synthesis on the basis of their chemical feasibility. As it turned out in
Table 2, pure degrader ability is improved for the 1,8-dichloro-substituted analogs of VPC compounds
3035, 3038 and 3039, while the strong degrader ability of compounds 3042 was retained by its 1,8-
dichloro-substituted analog. At the same time, the derivative of VPC-3040 did not improve the poor
degrader profile of the original compound, as predicted by the QSAR model in equation 1.
7

ACCEPTED MANUSCRIPT
2.2 Medicinal chemistry
VPC-3033 was synthesized as described in our original work (20), whereas VPC-3045 synthesis
was detailed in 2003 (25). The 1,8-dichloro-substituted benzylidene-anthracenones were synthesized as
previously described (26). Briefly, commercially available 1,8-dichloroanthraquinone was reduced with
stannous chloride in boiling HCl and acetic acid to the 10H-anthracen-9-one, followed by an aldol-type
condensation reaction with appropriately substituted aromatic aldehydes under basic conditions in the
presence of pyridine/piperidine (26). The same protocol was applied for the synthesis of hydroxy-
substituted 130008 (27), which was not included in our previous paper. Additionally, 10-[2-(4-hydroxy-
phenyl)-2-oxo-ethylidene]-10H-anthracen-9-one (130006) was prepared by a Friedel-Crafts acylation
reaction with subsequent ether cleavage, as already described elsewhere (28).
2.3 AR/ARV7 inhibition profiles of 10-benzylidene-10H-anthracen-9-ones
The effects of the compounds on AR transcriptional activity were measured in a cell-based
reporter assay in AR-positive LNCaP-eGFP cells that stably incorporate the AR-specific probasin
promoter behind an eGFP cDNA sequence (29). In parallel, inhibition of ARV7 by the studied small
molecules was assessed on PC-3 cells that incorporate a doxycycline inducible ARV7 construct and
probasin based ARR3tk-Nanoluciferase reporter construct termed PC3V7_3TKNLuc (30). Table 1
displays the approximate IC50 values derived for each compound in the LNCaP-eGFP system (full
length AR) and PC3V7_3TKNLuc (ARV7) systems. The top five compounds identified across both
screening cell lines were VPC-3033, VPC-3045, VPC-130002, VPC-130004 and VPC-130005 ranging
from sub to low micromolar inhibition against the full length AR and ~10 micromolar against ARV7 in
8

ACCEPTED MANUSCRIPT
a 3 point concentration curve (Table 1 and Fig. S1). VPC-130008 treatment resulted in widespread cell
death even at the lowest concentration tested indicating extreme toxicity.
To confirm the inhibitory activity of 10-benzylidene-10H-anthracen-9-one derivatives against
the full-length AR (Fig. 3, panel A), the levels of secreted PSA, a known AR-target gene, were
measured from the media of the LNCaP-eGFP cells in response to compound-treatment (Fig. 3, panel
B). The results show consistent IC50 values between both PSA and eGFP assays, suggesting a specific
targeting of the AR signalling pathway. Inhibition of ARV7 was further quantified by a 12 point
concentration dose response curve in PC3V7_3TKNLuc cells which yielded IC50 values ranging 4-12
µM (Fig. 3, panel C). In all those experiments, the marketed AR antagonist enzalutamide was used as a
positive (full-length) or negative (ARV7) control to verify the fidelity of the reporter system. Compound
VPC-3045 emerged as the most potent inhibitor with IC50 values of 0.23, 0.10 and 4.2 micromolar in
eGFP, PSA and V7 assays, respectively.
The inhibition of growth of AR-positive LNCaP cells by VPC-3033 was previously reported by
us (20). In this study, we employ 22rv1 cells as a model system of ARV7 driven prostate cancer (10), to
establish a correlation between the transcriptional inhibition of ARV7 and cell viability. Whereas VPC-
3033 and VPC-130002 suppressed the growth of 22rv1 line (over 72 hours) at ~0.5 and 5 µM,
respectively, the remaining compounds exerted their effect at much lower concentration indicating
possible toxic effects (Fig. 3, panel D).
Similar effects on cell viability were observed with AR-negative PC-3 cells with the exception of
VPC-3033, which only inhibited growth at high compound concentration (Fig. 3, panel D). These results
suggest a specific effect of VPC-3033 on ARV7 driven growth of 22rv1 cells compared to AR-
independent PC-3 cell growth. The strong effect of the remaining 5 compounds may arise from their
ability to affect tubulin polymerization, as previously noted (20).
9

ACCEPTED MANUSCRIPT
2.4 Degraders reduce protein expression of full length AR and ARV7 in PCa cells
LNCaP and 22rv1 cells were used as model systems of AR or ARV7 driven PCa to establish the
ability of the prospective compounds to degrade all forms of the receptor. Given the apparent toxicity of
the compounds after their 72 hours administration during cell viability experiments, AR or ARV7
protein expression was analyzed after 5 hours for acute response to compound-treatment. In LNCaP,
VPC-3033, VPC-3045 and VPC-130002 reduced protein expression of the full-length AR progressively
in response to concentration, with lesser effects of the remaining compounds, and without any inhibitors
stimulating PARP cleavage, indicating that apoptosis was not induced (Fig. 4, panel A and C).
Diminished AR levels appear to occur at only the protein level, as RT-PCR experiments revealed no
change in total AR mRNA in response to compound-treatment with the two most potent degraders,
VPC-3033 and VPC-3045 (Fig. S3). Similar results were observed in 22rv1 cells, where VPC-3033,
VPC-3045, VPC-130002 and VPC-130005 could facilitate degradation of the AR or ARV7 (Fig, 4,
panel B and C).
Enzalutamide treatment also reduced AR and ARV7 levels in 22rv1 cells, but the protein
expression did not vary with concentration. Overall, the constitutively nuclear ARV7 was less
susceptible to degradation possibly because of its high expression level (31), tight binding to chromatin
(19), and possible evasion of cytosolic proteolytic machinery. These results provide evidence that the
panel of inhibitors block transcriptional activity through mechanisms beyond degradation of the
androgen receptor.
10

ACCEPTED MANUSCRIPT
2.5 Degraders do not block AR nuclear translocation or strongly displace DHT from the AR-
LBD
To measure the effect of the panel of inhibitors on AR nuclear transport, yellow fluorescence
protein (YFP) tagged full length AR was over-expressed in HEK293T cells, followed by R1881
stimulation and visualization of the sub-cellular localization by confocal microscopy. Following R1881
stimulation and 4 hours of compound treatment (10 µM) or vehicle control, no acute effect of any
compound to block AR translocation to the nucleus was observed, with the exception of enzalutamide
(Fig. 5). These results imply that inhibition of full-length AR transcriptional activity do not solely stem
from competition for steroids at the LBD
To directly test for DHT displacement from the LBD, the interaction between fluorescent-labeled
DHT and purified recombinant AR-LBD was determined (Polar Screen assay) in the presence and
absence of inhibitors. Whereas both enzalutamide and bicalutamide, two known AR antagonists,
significantly displaced DHT from the AR-LBD at 1 and 10 µM compound concentrations, the panel of
degraders showed less ability to elicit the same response (Fig. 6A). Notably, VPC-3033 was previously
reported to displace DHT with an IC50 between 0.625-2.5 µM as calculated by a dose response curve
(20), whereas the activity in Figure 6A is only an estimated based on two concentrations to facilitate
screening of the entire panel of inhibitors.
Given that the studied compounds inhibited ARV7 transcriptional activity, our results confirm
that the small molecules could operate by other/additional mechanisms of actions compared to
conventional AR antagonists that target the AR-LBD.
11

ACCEPTED MANUSCRIPT
2.6 Degraders are selective for the AR at moderate concentration.
Cross-reactivity against related receptors were ruled out with reporter assays using full-length AR, ER,
GR and PR. Reporter constructs were employed that contained the appropriate response regions for each
receptor with ARR3tk for AR/GR/PR and estrogen response element for ER. Whereas VPC-3033 and
VPC-3045 treatment inhibited the AR, no inhibition was observed for ER and GR, with only minimal
effect on PR (Fig. 6B).
12

ACCEPTED MANUSCRIPT
3. DISCUSSION
In this study, we have used a combination of CADD methodology and medicinal chemistry to
build upon a previous generation of 10-benzylidene-10H-anthracen-9-ones that degrade the AR and
inhibit its transcriptional activity without affecting nuclear trans-localization of the receptor. Further
study of VPC-3033 and its derivatives revealed their ability to target ARV7. Among the most active
compounds under investigation (VPC-3033, -3045, -130002, -130004, -130005), only VPC-3033 could
selectively inhibit the growth of ARV-driven 22rv1 cells without exhibiting overt cytotoxicity toward
AR negative PC-3 cells. These results indicate that non-specific toxicity may represent at least a partial
component of the AR degrading mechanism of action. Differential effects of the compounds on the
degradation of AR and ARV7 were also observed in LNCaP and 22rv1 cell lines with moderate to
strong reduction in protein levels of each AR isoform. It should be noted, however, that although the
CADD-designed ‘pure degraders’ VPC-130002, -130004, -130005 demonstrated comparable V7
inhibition effect on ARV7 as their parental molecule VPC-3033 (which could actually correspond to the
sensitivity limit of the assay), they demonstrated much reduced DHT displacement from the AR-LBD
and thus, they should be much less prone to agonism-converting AR-LBD mutations.
Degradation of transcription factors or chromatin readers is an expanding field of study that
offers an alternative to targeting a protein's specific molecular function. Much like the small molecules
presented here, the compound galeterone displayed multiple activities to interfere with steroid synthesis
and degrade the AR or ARV7 through a complex mechanism including cytochrome C release and
initiation of apoptosis (23, 24). In addition, proteolysis targeting chimera molecules (PROTAC), which
are bifunctional compounds containing ligands for both E3-ubiquitin ligase and a protein of interest (21,
22, 32, 33), have been used to selectively ubiquitinate and degrade androgen and estrogen receptors as
13

ACCEPTED MANUSCRIPT
potential therapies for prostate and breast cancers (34). PROTAC methods have also been used to target
bromodomain proteins, which are important chromatin readers that facilitate AR and ARV7 binding to
DNA (35), as another means to suppress the growth of castration-resistant prostate cancers (36).
Conjugating a receptor ligand to a specific and nongenetic inhibitor of apoptosis protein (SNIPERs) is
another way to recruit a E3 ubiquitin ligase, as has been to used to selectively degrade the AR (37).
Beyond the recruitment of proteolytic machinery, other compounds can act as selective AR down-
regulators (SARDS) via direct interactions with the AR-LBD (38, 39), some of which have clinical
potential to treat CRPC tumors (39). Finally, small molecules have been developed that target the
transcriptional regulators of the AR itself, such as the transcription factor for ROR-gamma, which
results in suppression of the growth of AR-positive xenografts (40). Our small molecules that promote
degradation of the AR could be developed to work in combination with any of these prospective
therapeutic options following addition medicinal chemistry to eliminate toxicity and increase potency.
The exact mechanism of action that is exerted by the developed degraders remains mostly
unclear, although the levels of full-length AR mRNA were not affected in LNCaP cells. At least some
the compounds were derived from chemical scaffolds that could inhibit tubulin polymerization in cells
(25), and VPC-3033 itself weakly blocked assembly of tubulin monomers in an in vitro assay (20, IC50 ~
10 µM). It is also well established that taxanes, such as docetaxel and cabazitaxel, affect the microtubule
network in cells which blocks the full length AR transport into the nucleus (41, 42). AR splice variants
such as ARV7 and ARV567es, however, can evade taxane treatment and continue to propagate
androgen signalling (42), a mechanism that confers drug resistance to taxane treated tumors. The
degrader compounds developed here continue to inhibit AR splice variants and, for this reason, may
offer a promising alternative to taxane chemotherapies to bypass the observed resistance mechanism.
Our future work will focus on increasing potency and metabolic stability of the degraders with the goal
14

ACCEPTED MANUSCRIPT
of improving activity against AR splice variants and demonstrating pre-clinical efficacy in mouse
xenografts of ARV7 positive cell lines.
15

ACCEPTED MANUSCRIPT
CONCLUSIONS
We have previously identified a series of 10-benzylidene-10H-anthracen-9-ones as compounds
that could effectively inhibit full length human androgen receptor by two seemingly independent
mechanisms – through competition with androgens binding to receptor’s ligand binding domain AR-
LBD (the conventional mode of action for all current anti-androgen) and by reduction of the expression
levels of AR, resulting from protein’s degradation caused by the compounds.
In years following our original report, the role of splice variants of AR, lacking the entire AR-
LBD portion has emerged as one of defining factors of progressing prostate cancer into lethal castration
resistant stage. The activity of such truncated forms of AR became a significant therapeutics concern
and called for an entirely novel way of targeting the AR, and specifically ARVs. The emerging resistant
mutations in the AR-LBD portion of the receptors further supported the quest for small molecule
therapeutics directed beyond AR-LBD (19, 31).
Thus, we have revisited our original results on 10-benzylidene-10H-anthracen-9-ones and
modeled several derivatives with reduced AR-LBD affinity and enhanced ability to degrade AR. We
anticipated that such degradation effect will enable elimination of both full length and truncated forms
of AR. A series of modified 10-benzylidene-10H-anthracen-9-ones have been synthesised and these
compounds demonstrated effective inhibition of transcriptional activity of ARV isoforms, reduced their
expression levels and suppressed the growth of AR-V7 dependent cell-lines.
16

ACCEPTED MANUSCRIPT
4. METHODS
4.1 Compounds
VPC-3033 and VPC-3045 were purchased from Life Chemicals (Niagra, ON). VPC-130001 to 130008
were provided by Helge Prinz (Institute of Pharmaceutical and Medicinal Chemistry, Westfälische
Wilhelms-Universität, Münster, Germany). Enzalutamide (MDV-3100) was obtained from Haoyuam
Chemexpress. R1881 synthetic androgen was purchased from Sigma.
4.2 Plasmids constructs and luciferase reporter assay.
Full-length human AR (hARWT), ER, GR, and PR were encoded on a pcDNA3.1 expression plasmid.
pGL3-ARR3tk-NLuc consists of the ARR3tk 3 x probasin promoter followed by Nanoluciferase cDNA
on a pGL3 basic vector backbone. The ARR3tk promoter was replaced with a 3 x ERE (Estrogen
response element) to create a pGL3-ERE-NLuc reporter construct. Following steroid starvation in RPMI
media with 5% charcoal stripped serum (RPMI+CSS) for 2 days, PC-3 (5000 cells/well, 96 well,
RMPI+CSS) were transfected with 50 ng/well of hARWT plasmid, pGL3-ARR3tk-NLuc (5 ng) and 3µL
TT20/µg DNA (48 h). Cells were treated with compounds and 0.1 nM R1881 androgen (24 h), followed
by cell lysis with Nanoglo reagent (Promega, N1120) and luminescence measurement (TECAN
m200Pro). Transfections with estrogen (ER), glucocorticoid (GR) or progesterone (PR) receptor (50
ng/well) were performed as above using pGL3-ARR3tk-NLuc for GR/PR and pGL3-ERE-NLuc for ER.
4.3 PCa cell lines
17

ACCEPTED MANUSCRIPT
Prostate cancer cell lines LNCaP (ATCC, CRL-1740), 22Rv1 (ATCC, CRL-2505) and PC-3 (ATCC,
CRL-1435) were obtained in 2013 and authenticated in 2014 by IDEXX Laboratories. Testing for
mycoplasma contamination was performed weekly.
4.4 eGFP and PSA assays
AR positive LNCaP-eGFP cells with incorporated AR2PB-eGFP reporter construct were previously
described (29). LNCaP-eGFP cells were androgen deprived in RPMI 1640 + 5% charcoal stripped
media (RPMI+CSS) for at least 2 days prior to experiments. Inhibitory effects of degrader compounds
on expression of the eGFP reporter gene and secreted PSA were measured as previously described (20).
4.5 PC3_iV7 + 3TKNLuc cell line generation
Prostate cancer cell line PC3V7 (doxycycline inducible ARV7) was generated with lentivirus derived
from the pLIX402-ARV7 plasmid encoding doxycycline inducible ARV7 cDNA as previously
described (30), and applying virus onto PC-3 target cells cultured in DMEM + 5% FBS (DMEM+FBS)
and 1 µg/mL puromycin selection. pLU-ARR3tk-NLuc reporter lentivirus construct was used to deliver
the AR specific Nanoluciferase reporter into PC3V7 cells to generate PC3V7_3TKNLuc population
under puromycin (1 µg/mL) and blasticidin (4 µg/mL) selection (DMEM+FBS) as previously described
(30). Nanoluciferase reporter assays were performed by culturing PC3V7_3TKNLuc in DMEM + 5%
CSS media (DMEM+CSS, 96 well), induction of ARV7 with 200 ng/mL doxycycline treatment with
18

ACCEPTED MANUSCRIPT
compounds overnight, and cell lysis with Nanoglo (Promega) lysis buffer/ substrate mixture (5 minutes,
room temp.). Luminescence measurements were carried out on a TECAN m200pro luminometer.
4.6 PrestoBlue Cell Viability Assays
Cell viability tests were performed on 22Rv1 and PC-3 by seeding 5000 cells/well (96-well plates, 24
hours) in androgen-deprived RPMI+CSS (22Rv1) or DMEM+CSS (PC-3) prior to treatment with
compounds and R1881 for 72 hours. Plates were incubated with 30 µL/well PrestoBlue™ Cell Viability
Reagent (1 hr) followed by fluorescent measurement (535 excitation, 612 emission) on a TECAN F200
fluorometer.
4.7 Western Blot
LNCaP and 22Rv1 were grown in 6-well dishes of RPMI 1640 with 5% fetal bovine serum (FBS) and
treated with compounds for 6 hours at 2µM, 10µM and 20µM in 5% DMSO. Cells were harvested in 1X
phosphate buffered saline (PBS) and lysed in 1X Passive Lysis Buffer (Promega). Protein concentration
was normalized using Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific) and analyzed on an
Epoch 96-well spectrophotometer (BioTek) at 562 nm. Between 30-50µg of protein per lane was loaded
from each treatment and resolved on a 10% SDS polyacrylamide gel. The membrane was probed with
anti-H-280 (Santa Cruz BioTechnology, sc-13062), anti-actin (Sigma, A2066) and anti-PARP1 (Sigma,
P7605) rabbit polyclonal antibodies, followed by secondary antibody Alexa Fluor® 680 donkey anti-
rabbit (Life Technologies, A10043), to resolve AR (full-length and V7),actin and total PARP proteins.
Densities of western blot bands were measured using ImageJ software.
19

ACCEPTED MANUSCRIPT
4.8 Confocal microscopy
HEK293T cells (DMEM+CSS) were seeded into Falcon 8-chamber glass microscopy slides at 20,000
cells/well for 24 hours. Transfections were performed with 25 ng of YFP-AR (48 h, 0.15 µL/well
TransIT20/20 reagent - Mirus) followed treatment with 10 nM R1881 and 10 µM compounds for 4
hours. Cells were fixed in paraformaldehyde (PFA) overnight and mounted with DAPI containing
Vectashield mounting medium. Visualization performed on Zeiss Observer Z1 780 Confocal
Microscope for FRET analysis for DAPI stain (405 nm, Laser power = 9, pinhole = 600, gain = 500) or
YFP (514 nm, Laser power = 20, pinhole = 600, gain = 500) fluorescence channels (digital offset = 1,
digital gain = 0.3).
4.9 DHT displacement
Degrader compounds were tested for their ability to displace DHT from the purified, recombinant AR-
LBD supplied by the Polar Screen Androgen Receptor Competitor Green Assay Kit (Thermofisher) as
per instructions from the manufacturer.
4.10 Modelling of the potential degrader and DHT displacement activities
The geometry of the molecules in the dataset (VPC-3022 to VPC-3049 in Fig. S2 and novel compounds
from Fig. 1) were optimized using PM3 semi-empirical method prior inductive autocorrelation
calculation. To explore the structure-activity relationship, the molecular descriptors were correlated to
20

ACCEPTED MANUSCRIPT
the activity classes using multilinear regression (MLR) models and genetic algorithm (GA). Based on
the predicted degrader response and the experimental classes, the ROC curve plots the fraction of true
‘pure’ degrader compounds among all ‘pure’ degraders (TPR = true positive rate or sensitivity) versus
the fraction of false ‘pure’ degraders compounds among the ‘non-pure’ degraders (FPR = false positive
rate or 1-specificity). The best possible prediction is 100% sensitivity and 100% specificity with area
under the curve (AUC) of 1, while completely random guessing is depicted by the diagonal dashed line
from the bottom left to the top right corners with an AUC value of 0.5 on the ROC curve (as shown in
Fig. S2).
The docking of VPC3033 and its derivatives into the AR-LBD was done with a crystal structure
resolved at 1.65 Å (PDB: 2PNU). The virtual screening was done with Glide docking program and the
extra precision mode as previously described (20, 43, 44).
4.11 Statistical analysis
Data are reported as the mean ± SD for the indicated number of replicates in each experiment. Where
specified, statistical significance of the difference between two data sets was determined by paired, two-
tailed t-test where p-value of < 0.05 (*) and p-value of < 0.01 (**) were considered significant.
21

ACCEPTED MANUSCRIPT
ACKNOWLEDGEMENTS
This work was funded by a Canadian Institutes of Health Research (CIHR, grant# 123460), a
Prostate Cancer Canada-Translational Acceleration Grant (PCC-TAG, grant# TAG2014-05) and a Terry
Fox New Frontiers Program Project grant (TFRI Project 1062).
22

ACCEPTED MANUSCRIPT
REFERENCES
1. N. Lallous, K. Dalal, A. Cherkasov, P.S. Rennie, Targeting alternative sites on the androgen receptor
to treat castration-resistant prostate cancer, Int. J. Mol. Sci. 14 (2013) 12496-12519.
2. S.M. Green, E.A. Mostaghel, P.S. Nelson, Androgen action and metabolism in prostate cancer, Mol.
Cell. Endocrinol. 360 (2012) 3-13.
3. G. Jenster, H.A. van der Korput, C. van Vroonhoven, T.H. van der Kwast, J. Trapman, A.O.
Brinkmann, Domains of the human androgen receptor involved in steroid binding, transcriptional
activation, and subcellular localization, Mol. Endocrinol. 5 (1991) 1396-1404.
4. C. Tran, S. Ouk, N.J. Clegg, Y. Chen, P.A. Watson, V. Arora, J. Wongvipat, P.M. Smith-Jones, D.
Yoo, A. Kwon, T. Wasielewska, D. Welsbie, C.D. Chen, C.S. Higano, T.M. Beer, D.T. Hung, H.I.
Scher, M.E. Jung, C.L. Sawyers, Development of a second-generation antiandrogen for treatment of
advanced prostate cancer, Science. 324 (2009) 787-790.
5. N. Lallous, S.V. Volik, S. Awrey, E. Leblanc, R. Tse, J. Murillo, K. Singh, A.A. Azad, A.W. Wyatt,
S. LeBihan, K.N. Chi, M.E. Gleave, P.S. Rennie, C.C. Collins, A. Cherkasov, Functional analysis of
androgen receptor mutations that confer anti-androgen resistance identified in circulating cell-free
DNA from prostate cancer patients, Genome Biol. 17 (2016) 10-015-0864-1.
6. H. Borgmann, N. Lallous, D. Ozistanbullu, E. Beraldi, N. Paul, K. Dalal, L. Fazli, A. Haferkamp, P.
Lejeune, A. Cherkasov, M.E. Gleave, Moving Towards Precision Urologic Oncology: Targeting
Enzalutamide-resistant Prostate Cancer and Mutated Forms of the Androgen Receptor Using the
Novel Inhibitor Darolutamide (ODM-201), Eur. Urol. 73 (2018) 4-8.
7. D.J. Osguthorpe, A.T. Hagler, Mechanism of androgen receptor antagonism by bicalutamide in the
treatment of prostate cancer, Biochemistry. 50 (2011) 4105-4113.
23

ACCEPTED MANUSCRIPT
8. M.E. Taplin, G.J. Bubley, T.D. Shuster, M.E. Frantz, A.E. Spooner, G.K. Ogata, H.N. Keer, S.P.
Balk, Mutation of the androgen-receptor gene in metastatic androgen-independent prostate cancer,
N. Engl. J. Med. 332 (1995) 1393-1398.
9. S.M. Dehm, L.J. Schmidt, H.V. Heemers, R.L. Vessella, D.J. Tindall, Splicing of a novel androgen
receptor exon generates a constitutively active androgen receptor that mediates prostate cancer
therapy resistance, Cancer Res. 68 (2008) 5469-5477.
10. S.C. Chan, Y. Li, S.M. Dehm, Androgen receptor splice variants activate androgen receptor target
genes and support aberrant prostate cancer cell growth independent of canonical androgen receptor
nuclear localization signal, J. Biol. Chem. 287 (2012) 19736-19749.
11. Y. Li, S.C. Chan, L.J. Brand, T.H. Hwang, K.A. Silverstein, S.M. Dehm, Androgen receptor splice
variants mediate enzalutamide resistance in castration-resistant prostate cancer cell lines, Cancer
Res. 73 (2013) 483-489.
12. J.C. Park, M.A. Eisenberger, Advances in the Treatment of Metastatic Prostate Cancer, Mayo Clin.
Proc. 90 (2015) 1719-1733.
13. A. Anantharaman, T.W. Friedlander, Targeting the androgen receptor in metastatic castrate-resistant
prostate cancer: A review, Urol. Oncol. 34 (2016) 356-367.
14. T. Seisen, M. Roupret, F. Gomez, G.G. Malouf, S.F. Shariat, B. Peyronnet, J.P. Spano, G. Cancel-
Tassin, O. Cussenot, A comprehensive review of genomic landscape, biomarkers and treatment
sequencing in castration-resistant prostate cancer, Cancer Treat. Rev. 48 (2016) 25-33.
15. J. Luo, Development of AR-V7 as a putative treatment selection marker for metastatic castration-
resistant prostate cancer, Asian J. Androl. 18 (2016) 580-585.Anantharaman, A., and Friedlander,
T.W. (2016) Targeting the androgen receptor in metastatic castrate-resistant prostate cancer: A
review. Urol.Oncol. 34, 356-367.
24

ACCEPTED MANUSCRIPT
16. M.R. Smith, F. Saad, S. Chowdhury, S. Oudard, B.A. Hadaschik, J.N. Graff, D. Olmos, P.N.
Mainwaring, J.Y. Lee, H. Uemura, A. Lopez-Gitlitz, G.C. Trudel, B.M. Espina, Y. Shu, Y.C. Park,
W.R. Rackoff, M.K. Yu, E.J. Small, SPARTAN Investigators, Apalutamide Treatment and
Metastasis-free Survival in Prostate Cancer, N. Engl. J. Med. 378 (2018) 1408-1418.
17. C.E. Bohl, D.D. Miller, J. Chen, C.E. Bell, J.T. Dalton, Structural basis for accommodation of
nonsteroidal ligands in the androgen receptor, J. Biol. Chem. 280 (2005) 37747-37754.
18. Y. He, J. Lu, Z. Ye, S. Hao, L. Wang, M. Kohli, D.J. Tindall, B. Li, R. Zhu, L. Wang, H. Huang,
Androgen receptor splice variants bind to constitutively open chromatin and promote abiraterone-
resistant growth of prostate cancer, Nucleic Acids Res. 46 (2018) 1895-1911.
19. K. Dalal, M. Che, N.S. Que, A. Sharma, R. Yang, N. Lallous, H. Borgmann, D. Ozistanbullu, R.
Tse, F. Ban, H. Li, K.J. Tam, M. Roshan-Moniri, E. LeBlanc, M.E. Gleave, D.T. Gewirth, S.M.
Dehm, A. Cherkasov, P.S. Rennie, Bypassing Drug Resistance Mechanisms of Prostate Cancer with
Small Molecules that Target Androgen Receptor-Chromatin Interactions, Mol. Cancer. Ther. 16
(2017) 2281-2291.
20. H. Li, M.D. Hassona, N.A. Lack, P. Axerio-Cilies, E. Leblanc, P. Tavassoli, N. Kanaan, K. Frewin,
K. Singh, H. Adomat, K.J. Bohm, H. Prinz, E.T. Guns, P.S. Rennie, A. Cherkasov, Characterization
of a new class of androgen receptor antagonists with potential therapeutic application in advanced
prostate cancer, Mol. Cancer. Ther. 12 (2013) 2425-2435.
21. B.Z. Stanton, E.J. Chory, G.R. Crabtree, Chemically induced proximity in biology and medicine,
Science. 359 (2018) 10.1126/science.aao5902.
22. H. Lebraud, T.D. Heightman, Protein degradation: a validated therapeutic strategy with exciting
prospects, Essays Biochem. 61 (2017) 517-527.
25

ACCEPTED MANUSCRIPT
23. Z. Yu, C. Cai, S. Gao, N.I. Simon, H.C. Shen, S.P. Balk, Galeterone prevents androgen receptor
binding to chromatin and enhances degradation of mutant androgen receptor, Clin. Cancer Res. 20
(2014) 4075-4085.
24. A.K. Kwegyir-Afful, S. Ramalingam, P. Purushottamachar, V.P. Ramamurthy, V.C. Njar,
Galeterone and VNPT55 induce proteasomal degradation of AR/AR-V7, induce significant
apoptosis via cytochrome c release and suppress growth of castration resistant prostate cancer
xenografts in vivo, Oncotarget. 6 (2015) 27440-27460.
25. H. Prinz, Y. Ishii, T. Hirano, T. Stoiber, J.A. Camacho Gomez, P. Schmidt, H. Dussmann, A.M.
Burger, J.H. Prehn, E.G. Gunther, E. Unger, K. Umezawa, Novel benzylidene-9(10H)-anthracenones
as highly active antimicrotubule agents. Synthesis, antiproliferative activity, and inhibition of tubulin
polymerization, J. Med. Chem. 46 (2003) 3382-3394.
26. H. C. Nickel, P. Schmidt, K. J. Böhm, S. Baasner, K. Müller, M. Gerlach, E. Unger, E.G. Günther,
H. Prinz, Synthesis, antiproliferative activity and inhibition of tubulin polymerization by 1,5- and
1,8-disubstituted 10H-anthracen-9-ones bearing a 10-benzylidene or 10-(2-oxo-2-phenylethylidene)
moiety, Eur. J. Med. Chem., 45 (2010) 3420-3438.
27. S. M. M. El-Shafie, Synthesis of 10-arylidene-9-anthrones & related compounds, Indian J. Chem.,
16B (1978) 828-830.
28. H. Prinz, P. Schmidt, K.J. Böhm, S. Baasner, K. Müller, E. Unger, M. Gerlach, E.G. Günther, 10-(2-
oxo-2-Phenylethylidene)-10H-anthracen-9-ones as Highly Active Antimicrotubule Agents:
Synthesis, Antiproliferative Activity, and Inhibition of Tubulin Polymerization, J. Med. Chem., 52
(2009) 1284-1294.
26

ACCEPTED MANUSCRIPT
29. P. Tavassoli, R. Snoek, M. Ray, L.G. Rao, P.S. Rennie, Rapid, non-destructive, cell-based screening
assays for agents that modulate growth, death, and androgen receptor activation in prostate cancer
cells, Prostate. 67 (2007) 416-426.
30. K. Dalal, F. Ban, H. Li, H. Morin, M. Roshan-Moniri, K.J. Tam, A. Shepherd, A. Sharma, J.
Peacock, M.L. Carlson, E. LeBlanc, C. Perez, F. Duong, C.J. Ong, A. Cherkasov, P.S. Rennie,
Selectively targeting the dimerization interface of the androgen receptor with small-molecules to
treat castration-resistant prostate cancer. Cancer Letters. (2018). Accepted.
31. K. Dalal, M. Roshan-Moniri, A. Sharma, H. Li, F. Ban, M.D. Hassona, M. Hsing, K. Singh, E.
LeBlanc, S. Dehm, E.S. Tomlinson Guns, A. Cherkasov, P.S. Rennie, Selectively targeting the
DNA-binding domain of the androgen receptor as a prospective therapy for prostate cancer, J. Biol.
Chem. 289 (2014) 26417-26429.
32. K.M. Sakamoto, K.B. Kim, A. Kumagai, F. Mercurio, C.M. Crews, R.J. Deshaies, Protacs: chimeric
molecules that target proteins to the Skp1-Cullin-F box complex for ubiquitination and degradation,
Proc. Natl. Acad. Sci. U. S. A. 98 (2001) 8554-8559.
33. K.M. Sakamoto, K.B. Kim, R. Verma, A. Ransick, B. Stein, C.M. Crews, R.J. Deshaies,
Development of Protacs to target cancer-promoting proteins for ubiquitination and degradation, Mol.
Cell. Proteomics. 2 (2003) 1350-1358.
34. A. Rodriguez-Gonzalez, K. Cyrus, M. Salcius, K. Kim, C.M. Crews, R.J. Deshaies, K.M. Sakamoto,
Targeting steroid hormone receptors for ubiquitination and degradation in breast and prostate cancer,
Oncogene. 27 (2008) 7201-7211.
35. S.C. Chan, L.A. Selth, Y. Li, M.D. Nyquist, L. Miao, J.E. Bradner, G.V. Raj, W.D. Tilley, S.M.
Dehm, Targeting chromatin binding regulation of constitutively active AR variants to overcome
prostate cancer resistance to endocrine-based therapies, Nucleic Acids Res. 43 (2015) 5880-5897
27

ACCEPTED MANUSCRIPT
36. 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, PROTAC-induced BET protein degradation
as a therapy for castration-resistant prostate cancer, Proc. Natl. Acad. Sci. U. S. A. 113 (2016) 7124-
7129.
37. N. Shibata, K. Nagai, Y. Morita, O. Ujikawa, N. Ohoka, T. Hattori, R. Koyama, O. Sano, Y. Imaeda,
H. Nara, N. Cho, M. Naito, Development of Protein Degradation Inducers of Androgen Receptor by
Conjugation of Androgen Receptor Ligands and Inhibitor of Apoptosis Protein Ligands, J. Med.
Chem. 61 (2018) 543-575.
38. R.H. Bradbury, N.J. Hales, A.A. Rabow, G.E. Walker, D.G. Acton, D.M. Andrews, P. Ballard, N.A.
Brooks, N. Colclough, A. Girdwood, U.J. Hancox, O. Jones, D. Jude, S.A. Loddick, A.A. Mortlock,
Small-molecule androgen receptor downregulators as an approach to treatment of advanced prostate
cancer, Bioorg. Med. Chem. Lett. 21 (2011) 5442-5445.A. Omlin, R.J. Jones, R. van der Noll, T.
Satoh, M. Niwakawa, S.A. Smith, J. Graham, M. Ong, R
39. D. Finkelman, J.H. Schellens, A. Zivi, M. Crespo, R. Riisnaes, D. Nava-Rodrigues, M.D. Malone, C.
Dive, R. Sloane, D. Moore, J.J. Alumkal, A. Dymond, P.A. Dickinson, M. Ranson, G. Clack, J. de
Bono, T. Elliott, AZD3514, an oral selective androgen receptor down-regulator in patients with
castration-resistant prostate cancer - results of two parallel first-in-human phase I studies, Invest.
New Drugs. 33 (2015) 679-690.
40. J. Wang, J.X. Zou, X. Xue, D. Cai, Y. Zhang, Z. Duan, Q. Xiang, J.C. Yang, M.C. Louie, A.D.
Borowsky, A.C. Gao, C.P. Evans, K.S. Lam, J. Xu, H.J. Kung, R.M. Evans, Y. Xu, H.W. Chen,
ROR-gamma drives androgen receptor expression and represents a therapeutic target in castration-
resistant prostate cancer, Nat. Med. 22 (2016) 488-496.
28