Synthesis and evaluation of novel benzotropolones as Atg4B inhibiting autophagy blockers

Article abstract of DOI:10.1016/j.bioorg.2019.03.021

Autophagy is an intracellular degradation/recycling pathway that provides nutrients and building blocks to cellular metabolism and keeps the cytoplasm clear of obsolete proteins and organelles. During recent years, dysregulated autophagy activity has been reported to be a characteristic of many different disease types, including cancer and neurodegenerative disorders. This has created a strong case for development of autophagy modulating compounds as potential treatments for these diseases. Inhibitors of autophagy have been proposed as a therapeutic intervention in, e.g., advanced cancer, and inhibiting the cysteine protease Atg4B has been put forward as a main strategy to block autophagy. We recently identified and demonstrated -both in vitro and in vivo - that compounds with a benzotropolone basic structure targeting Atg4B, can significantly slow down tumor growth and potentiate the effect of classical chemotherapy. In this study we report the synthesis and inhibition profile of new benzotropolone derivatives with additional structural modifications at 6 different positions. To obtain a solid inhibition profile, all compounds were evaluated on three levels, including two cell-based assays to confirm autophagy and intracellular Atg4B inhibition and an SDS-PAGE-based experiment to assess in vitro Atg4B affinity. Several molecules with a promising profile were identified.

Full text of DOI:10.1016/j.bioorg.2019.03.021

Accepted Manuscript
Synthesis and evaluation of novel benzotropolones as Atg4B inhibiting autoph-
agy blockers
Muhammet Tanc, Matthias Cleenewerck, Ammar Kurdi, Ria Roelandt, Wim
Declercq, Guido De Meyer, Koen Augustyns, Wim Martinet, Pieter van der
Veken
PII:
S0045-2068(18)31336-1
DOI:
https://doi.org/10.1016/j.bioorg.2019.03.021
YBIOO 2847
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
16 November 2018
4 March 2019
11 March 2019
Please cite this article as: M. Tanc, M. Cleenewerck, A. Kurdi, R. Roelandt, W. Declercq, G. De Meyer, K.
Augustyns, W. Martinet, P. van der Veken, Synthesis and evaluation of novel benzotropolones as Atg4B inhibiting
autophagy blockers, Bioorganic Chemistry (2019), doi: https://doi.org/10.1016/j.bioorg.2019.03.021
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.

Synthesis and evaluation of novel benzotropolones as Atg4B inhibiting autophagy
blockers.
Muhammet Tanc, aᵻ Matthias Cleenewerck, aᵻ Ammar Kurdi,b Ria Roelandt,c, d Wim Declercq, c, d Guido De
Meyer, b Koen Augustyns, a Wim Martinet,b* Pieter van der Veken.a*
a
Laboratory of Medicinal Chemistry (UAMC), Department of Pharmaceutical Sciences, University of Antwerp, Antwerp,
Belgium
bLaboratory of Physiopharmacology, Department of Pharmaceutical Sciences, University of Antwerp, Antwerp, Belgium
cVIB Inflammation Research Center, Technologiepark 927, B-9052, Ghent, Belgium
dDepartment of Biomedical Molecular Biology, Ghent University, Technologiepark 927, B-9052, Ghent, Belgium
Abstract:
Autophagy is an intracellular degradation/recycling pathway that provides nutrients and
building blocks to cellular metabolism and keeps the cytoplasm clear of obsolete proteins and
organelles. During recent years, dysregulated autophagy activity has been reported to be a
characteristic of many different disease types, including cancer and neurodegenerative disorders. This
has created a strong case for development of autophagy modulating compounds as potential
treatments for these diseases. Inhibitors of autophagy have been proposed as a therapeutic
intervention in, e.g., advanced cancer, and inhibiting the cysteine protease Atg4B has been put forward
as a main strategy to block autophagy. We recently identified and demonstrated -both in vitro and in
vivo - that compounds with a benzotropolone basic structure targeting Atg4B, can significantly slow
down tumor growth and potentiate the effect of classical chemotherapy. In this study we report the
synthesis and inhibition profile of new benzotropolone derivatives with additional structural
modifications at 6 different positions. To obtain a solid inhibition profile, all compounds were
evaluated on three levels, including two cell-based assays to confirm autophagy and intracellular
Atg4B inhibition and an SDS-PAGE-based experiment to assess in vitro Atg4B affinity. Several
molecules with a promising profile were identified.
Keywords : Autophagy, Atg4B, Benzotropolone, UAMC-2526
______
*Corresponding authors. E-mail: pieter.vanderveken@uantwerpen.be (P. Van der Veken) ; wim.martinet@uantwerpen.be
(W. Martinet)

ᵻ Authors contributed equally.

1. Introduction
Autophagy is an intracellular recycling/degradation process that is remarkably evolutionary
conserved from yeast to higher eukaryotes. It targets long-lived cytosolic proteins and organelles to
subsidise cellular homeostasis and has also been demonstrated to be involved in host-pathogen
interactions with viruses, bacteria and parasitic protozoa [1,2]. The process is characterized by the
formation of double walled membrane vesicles called autophagosomes that engulf obsolete proteins
and cellular components. Breakdown of the contents is ensured by fusing of autophagosomes with a
lysozome. Autophagy is typically induced as a response to different stressors such as starvation or
hypoxia [3]. In addition, dysregulated autophagy (i.e., either defective or upregulated autophagy) has
recently been identified as a hallmark of multiple human disease types, including cancer,
neurodegenerative disorders and cardiovascular disease [4]. As a consequence, there is a strong and
growing demand for autophagy modulating compounds, both as research tools and as potential drugs.
Inhibitors of autophagy can be expected to find main applications in oncology: it has been convincingly
shown that cancer cells in advanced tumors draw heavily on autophagy to survive in the harsh
conditions of the tumor micro environment. Furthermore, autophagy is known to protect cancer cells
against oncolytic drugs by effectively limiting exchange of materials with the extracellular
environment. [5,6].
To date, both direct and indirect strategies to block autophagy have been reported. Exemplary for the
indirect approaches, lysosomal inhibitors like chloroquine and its derivatives have shown efficacy in a
number of animal oncology models, and several clinical trials have meanwhile been initiated.
Nonetheless, chloroquine derivatives and other molecules with similar activity (e.g., bafilomycin) most
likely only interfere with lysozomal acidification and do not have direct effects on the cell’s autophagy
machinery. In addition, the extracellular matrix acidification that is present in many advanced tumors,
has been claimed to render chloroquine clinically inefficient [7]. Conversely, direct targeting of
proteins involved in the autophagy cascade has been proposed as a potentially more interesting
alternative. In this framework, Atg4B has several attractive features as a potential target for directly
blocking autophagy. Atg4B is an enzyme with two well-defined activity types that have key roles
during autophagosome formation and maturation. Both activity types are catalyzed by the same active
center. First, Atg4B proteolytically primes its substrate LC3B relying on cysteine-type protease activity
and generating LC3B-I. Further downstream in the autophagy cascade, LC3B-I is lipidated with
phosphatidylethanolamine, yielding LC3B-II. The latter is essential for anchoring the so-called
isolation membrane during autophagosome formation. Afterwards, Atg4B recycles LC3B-I by
delipidating LC3B-II. In spite of significant effort by multiple groups, targeting Atg4B activity has so far
proven daunting. Inhibitor discovery has been hindered by two factors. 1) First, the conformational
dynamics of Atg4B in solution remain uncharacterized. Crystal structures have shown that the enzyme
can adopt at least two conformations: an active (‘open’) and a putatively catalytically inert (‘closed’)

form, each with distinct pocket structures in the active center [8]. It is currently unclear whether the
solution structure of Atg4B indeed corresponds to the closed conformation. 2) In addition, developing
a straightforward and reliable Atg4B-activity assay to support inhibitor screening has been shown to
be challenging. A readily accessible, standardized Atg4B-assay would also be most useful to compare,
and benchmark reported Atg4B-inhibitory potencies of compounds. So far, both rational design
strategies and several types of high throughput screening have only delivered compounds with
affinities in the micromolar range. Remarkably, none of the compounds identified possess a typical,
druglike ‘decorated scaffold’ architecture (Figure 1). Additionally, Roche’s fluoromethyl ketone FMK
9a was shown later to phenotypically induce autophagy (instead of inhibiting the process) [9-11]. Our
own effort in this domain has been focusing on benzotropolone-derived molecules. Apparent Atg4B-
affnities in an SDS-page based assay (vide infra) are typically also in the micromolar range, but these
molecules nonetheless strongly suppress both Atg4B activity and autophagy in cellulo at 10 µM
concentrations. In addition, we have demonstrated that compound UAMC-2526 is capable of potently
inhibiting autophagy in vivo and significantly reduces tumor growth in a HT29-mouse model of
colorectal cancer. Noteably, UAMC-2526 also significantly potentiates the effect of the reference
chemotherapeutic oxaliplatin [12,13]. This study reports the synthesis and evaluation of 29 novel
benzotropolone derivatives. It aims to provide a more profound insight in the Structure-Activity
Relationship compound class, a topic which so far, we investigated superficially.
Figure 1: Relevant examples of reported, Atg4B-targeting autophagy inhibitors (A) FMK 9a [14], (B) UAMC
2526 [13], (C) aurintricarboxylic acid [15], (D:) compound 4-28 [10].
2. Results and Discussion
2.1 Chemistry
The generic benzotropolone structure covering all the new compounds in the manuscript has six
variable positions (R1-R6) and is shown in Figure 2. A first aim of this study was to investigate the
impact of the R1 substituent. Specifically, we were interested whether UAMC 2526’s butyl ester
function at this position can be removed or replaced by a potentially more metabolically stable
carboxamide group. Earlier preliminary work from us had indicated that the amide alternative could
be worth investigating in more depth. [13] Likewise, we wanted to obtain more insight in the scope of
O- or N- alkyl substituents that are tolerated at this position. In addition, we aimed to explore the effect
of halogenation at position R2. Halogenation was considered interesting, since it has the potential to

increase overall lipophilicity and, possibly, cellular permeability. Reference UAMC 2526 already
carries an R3-bromine group and while earlier work had shown this to be preferable over an R3
fluorine or methoxy substituent in terms of cellular potency, open questions remained regarding the
halogenation of the R2-position [13]. In the same framework, we decided to evaluate the introduction
of an R2-methoxy and tert-butyl group. At the R4-R6 positions, the compounds in this study in most
cases carry a hydrogen atom. This is again based on the foregoing manuscript where it was shown that
the phenol and enol-type alcohol functions of the benzotropolone basic skeleton are essential for
autophagy blocking activity. Three compounds in the series nonetheless are derivatized with acetyl
groups at these positions. The latter molecules were included as potential ester prodrugs that could
have increased cellular permeability.
Figure 2: Overview of compounds prepared for this study.
Noteworthy, the choice of substituents was also governed by their compatibility with the
benzotropolone condensation reaction required for construction of all final products. The latter
consists of a KIO3-mediated oxidative reaction of a gallic acid derivative and a catechol derivative (vide
infra). Practically, the synthesis of the target molecules was initiated with the preparation of the gallic
acid building blocks. The carboxylate function in these gallate building blocks corresponds to the R1-
carboxylate group of the benzotropolone target compounds. (Scheme 1, Entry A). The required
gallate esters( 1a-1c),were obtained via Fisher-type, acid-catalyzed reaction between free gallic acid
and one of the selected alcohols. Synthesis of the gallic acid amides, 5a – 5h, was initiated by
acetylation of 1 using acetic anhydride and a catalytic amount of H2SO4 to yield the protected 3,4,5-
triacetoxybenzoic acid 2. The free carboxylate function of 2 was then converted to acyl chloride 3
using thionyl chloride. In the following step, a selected amine was added to obtain the protected
amides 4a – 4h. Deacetylation with hydrazine afforded the corresponding gallamides 5a – 5h.

Scheme 1: Synthetic preparation of gallic acid-based intermediates (Entry A);final products (Entry B); and
synthesis of compound 36 (Entry C). Reagents and conditions: (a) Alcohol, H2SO4 (cat.), reflux, overnight (b)
Ac2O, H2SO4 (cat.), 80°C, 15’; (c) Dry toluene, SOCl2, reflux, 1.5h; (d) Amine , dry DCM, DIPEA, rt, 1h; (e) ACN,
NH2NH2.H2O, rt, 0.5 h.
Compounds 8 and 35 were purchased from Sigma-Aldrich. Compound 19 was derived from methyl
3,4,6-trihydroxy-5-oxo-5H-benzo[7]annulene-8-carboxylate [13] via acetylation of the R4- R6
positions by using acetic anhydride in basic conditions. The same synthetic strategy was used to obtain
compound 20 from 3,4,6-trihydroxy-5-oxo-5H-benzo[7]annulene-8-carboxylic acid [13]. Compound
21 was derived from methyl 3,4,6-trihydroxy-5-oxo-5H-benzo[7]annulene-8-carboxylate [13] using
ammonia in MeOH.

2.2. Compound evaluation.
All compounds were evaluated on 3 complementary levels (Table 1). 1) First, a luciferase
reporter assay reported by Ketteler was implemented for assessment of intracellular Atg4B inhibition.
[16] HEK293 cells transfected with a [Gaussia luciferase--actine-LC3] construct and treated with
autophagy-inducing everolimus, are incubated with an autophagy inhibitor at 10µM concentration.
Luminescence quantification is used as a measure of intracellular Atg4B inhibition. The lower a
compound scores in this assay, the stronger it suppresses cellular Atg4B activity. In addition, the
luminescence scores in Table 1 are normalized, with luminescence in everolimus-treated controls
equated to 1. 2) Second, inhibition of everolimus-induced autophagy in Jurkat cells (10 µM inhibitor
concentration) was used as a phenotypical screen to assess autophagy blocking potential. Read-out in
this assay, published earlier by Stankov, consists of FACS-based fluorescence quantification of
autophagy vacuoles in cells after treatment with the fluorescent CytoID reagent. Scores represented in
Table
1 represent the mean fluorescence intensities (based on 10.000 individual cell
measurements/compound). In this assay, lower scores represent lower autophagosome fluorescence
and, hence, stronger autophagy inhibition. Scores have also been normalized: fluorescence in
everolimus-treated controls was equated to 1. Non-everolimus treated controls had a mean
fluorescence score of 0.37. 3) Third, a previously published gel-based assay was used for Atg4B
inhibitory potency determination [12]. The latter determines the % inhibition (500 µM compound
concentration) of Atg4B-mediated processing of LC3-GST, a fusion protein between LC3 and
glutathione-S-transferase (GST). The read-out in this assay consists of densitometric quantification of
remaining LC3-GST, separated from the other assay components using SDS-PAGE. Earlier
benchmarking effort with literature Atg4B inhibitors has shown that the experimental parameters of
this assay (e.g., the high enzyme and substrate concentrations required for reliable densitometric
quantification) produce IC50-values that are systematically higher than in other reported assays.
Compounds therefore appear to be less potent than they would in assay types where lower enzyme
and substrate concentrations can be used. [14] We nonetheless also include results for the new
compounds in the gel-based assay for the sake of completeness.

Table 1: Screening results of the benzotropolone library.
Luciferase
% Atg4B
Compound R1
3-MA(c)
R2
R3
R4
R5
R6
reporter CYTO-ID assay(b)
assay(a)
inhibition
N.D.(d)
0.57
0.55
N.D.
UAMC-
2526(e)
H
H
H
H
H
Br
H
H
H
H
H
H
H
H
H
H
H
H
0.43
49.97
6
7
H
H
H
H
H
0.89
N.D.(e)
0.67
N.D.
0.76
N.D.
0.46
1.01
0.99
42.25
9.67
OCH3
OCH3
OCOCH3
8
19.79
0.00
9
10
H
H
H
Br
Br
Br
Cl
H
H
H
H
H
H
H
H
0.67
30.30
11
12
13
H
H
H
0.85
0.17
0.24
0.94
0.52
0.35
16.69
18.03
24.66
H
H
14
15
16
17
Cl
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
N.D.
N.D.
0.18
0.44
1.26
0.93
0.29
0.60
54.45
30.59
12.46
2.62
OCH3
Cl
C(CH3)3

18
19
C(CH3)3
H
H
H
H
H
0.54
1.02
0.86
N.D.
3.86
H
COCH3 COCH3 COCH3
COCH3 COCH3 COCH3
13.00
20
21
H
H
H
H
0.87
0.82
N.D.
0.53
11.57
63.83
H
H
H
H
H
H
22
23
H
H
H
H
0.96
0.62
0.90
1.10
47.71
67.34
H
H
H
24
25
26
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
0.46
0.18
0.27
1.01
0.98
0.62
68.75
31.88
67.95
27
28
H
H
H
H
H
H
H
H
H
0.16
0.75
0.91
0.85
28.09
63.28
Br
29
30
31
32
H
H
H
H
Br
Br
H
H
H
H
H
H
H
H
H
H
H
H
0.18
0.21
N.D.
1.00
0.94
0.73
N.D.
0.99
60.17
45.05
34.50
33.68
OCH3
Br
33
H
Br
H
H
H
0.21
0.60
25.97

34
Br
H
H
H
H
N.D.
0.85
63.69
35
36
H
H
OH
H
H
H
H
H
H
N.D.
N.D.
1.05
1.02
1.75
9.20
OCOCH3
(a) Luciferase activity in everolimus-treated and control group was respectively 1 (normalized) and 0.40. Compounds were tested at 10 µM
concentration.
(b) Cyto-ID quantified autophagy in everolimus–treated and control group was respectively 1 (normalized) and 0.37. Compounds were
tested at 10 µM concentration.
(c) 3-MA: 3-Methyladenine (assay concentration 10 mM)
(d) N.D.: value was not determined
(e) UAMC-2526 was reported earlier by us in [13] and is used here as a reference.
When inspecting the data in Table 1, it should be noted first that results obtained in cellular assays
must be interpreted with caution. First of all, compound performance is strongly dependent on a
molecule’s cellular permeability and/or tendency for intracellular accumulation. In addition, the data
have been obtained from cells treated with everolimus. The latter stimulates autophagy in a largely
homogenous manner throughout the full population. While this approach is generally recognized to be
the most adequate one for reproducibly and reliably evaluating autophagy inhibitors, obtained results
do not necessarily correlate with hypothetical compound potencies in normal, unstimulated cells.
Autophagic flux in these normal cell populations and autophagosome pool turnover rates can be
expected to be significantly different. Nonetheless, it can reasonably be anticipated that the inhibitory
response to the compounds in non-everolimus primed cells will be relatively more potent. This has for
example been illustrated by ex vivo histology on liver tissue from UAMC-2526-treated animals [13]
A general observation when looking at overall autophagy blocking potential (determined in the Cyto-
ID assay), is that the most potent benzotropolones (e.g., 12-14) at 10 µM inhibit autophagy at least as
effectively as 3-methyladenine (3-MA), used at 10 mM. 3-MA is a reference autophagy blocker used as
a positive control in the experiment. The latter compound does not act through Atg4B-inhibition, but
targets type III phosphatidylinositol 3-kinases (PI-3K). Of note, the Cyto-ID scores and Luciferase-
reporter assay scores tend to correlate well within the the ester series (6-20), but are not in
agreement for some representatives of the amide series (21-34). This can be illustrated by comparing
both parameters for 23-25, 27 or 29. A possible explanation could be that only Atg4B endopeptidase
activity is measured in the luciferase reporter assay. However, as Atg4B is also involved in autophagy
through LC3B-II deconjugase activity, the latter feature could account for different scores in both assay
types in some cases. Although this is only a tentative rationale for the discrepancy, the more efficient
autophagy blocking potential of the esters in our opinion seems to favour the latter compounds over
their amide counterparts. Compound 32 (the amide analogue of UAMC-2526) is a relevant example of
this observation: compared to its ester parent compound, the amide loses most of its Atg4B- and
autophagy blocking potency. A noteable exception nonetheless is amide 26, which is both a strong
autophagy inhibitor and one of the most effective inhibitors of intracellular Atg4B-activity in Table 1.

The ester analogue of 26 was reported earlier by us in reference [13], but was not among the most
promising molecules identified in that report. An additional study goal relating to the R1-chain was to
determine whether the O- or N-substituents in the esters and/or amides can be optimized. When
comparing data for compound subsets with different R1-chains (but otherwise identically decorated
benzotropolone rings, e.g., 10-12 and 21-24), there is no clear indication that the O- or N- substituent
chain plays a pivotal role in target binding: this is generally reflected by the SDS-PAGE data and
cellular potency data. Therefore, this part of the compound could therefore be suited for finetuning
lipophilicity during the lead optimization phase, without affecting potency. Regardless of the
foregoing, the carboxyl function present in most R1-groups of the series does seem to be involved in
target binding. If it is removed (as in compounds 35 and 36), most of the Atg4B and autophagy
blocking potential is abolished.
Investigating the effect of functionalisation at the R2-position (as in 15-18 and 34) was another goal of
the study. The introduction of a Br-, Cl-, methoxy or tert-butyl substituent at R2 is tolerated, but
nonetheless generally decreases compound performance on all three parameters investigated. This
can be illustrated by comparing the results for 6 (R2= -H) and 16 (R2= -Cl). The opposite trend
however seems to be present for the introduction of a halogen atom at the R3-position. When
comparing the corresponding pairs 21/28, 23/29 and 24/30, the halogenated compounds overall
have better scores than analogues with R3=H. Lastly, the series included three derivatives with
acetylated –OH groups at R4-R6 or R3 (19-20 and 36, respectively) that were prepared as potential
prodrugs with increased cellular permeability. Intracellular hydrolysis would then be required to
release the active molecule. None of these compounds however displayed noteable activity.
Compounds 19 and 20 are the direct analogues of two molecules that we reported in reference [13]
(as UAMC-1356 and UAMC-1409, respectively) and were found to have a satisfactory autophagy
inhibiting profile. The lack of activity for 19 and 20 therefore seems to imply that the hypothesized
intracellular hydrolysis of the acetate prodrugs is not taking place at appreciable rates.
3.Conclusion
In this study, we report 29 novel autophagy inhibitors with a benzotropolone-based structure.
The benzotropolone core of the molecules was installed relying on a one-pot oxidative condensation of
a gallate and a catechol-derived precursor. For all prepared molecules, activity was determined on
three complementary levels: intracellular Atg4B-inhibition, autophagy inhibitory potency and in vitro
inhibition of Atg4B. At 10 µM concentration, the best compounds in the series block autophagy with
comparable potency as reference 3-MA used at 10 mM. Within the set of novel molecules, compounds
bearing an ester function at R1 generally performed better than their amide counterparts. A notable

exception is amide 26, which outperforms reference UAMC-2526 in all three assay types and is among
the most promising compounds identified in this study. In addition, this work also indicates that the O-
or N-substituent at R1 is not critically involved in target binding. Therefore, this position can offer an
opportunity for lipophilicity tuning of inhibitors, which is a potentially interesting feature for
compounds with an obligatory intracellular mode-of-action. Finally , this manuscript shows that for
introduction of a lipohilic substituent, the R3-position is generally more favoured than the R2 position.
Acetylation of the OH-groups at R4-R6 is another modification type that can be used to increase
lipophilicity and/or cellular permeability. The potential prodrugs that are obtained in this way,
however, do not show activity. The latter indicates that either their cellular permeability is
unexpectedly low or that they are not efficiently hydrolysed into the active benzotropolone inside the
cells.
Acknowledgments
We are grateful to the EU H2020-MSCA-IF actions (project ONCOPHAGY) and the Fund for Scientific
Research Vlaanderen (FWO Vlaanderen) for supporting this work with a research grant. The Hercules
Foundation is acknowledged for heavy equipment grants (NMR, preparatory purification
infrastructure).
References
[1] N. Mizushima, M. Komatsu; Autophagy: renovation of cells and tissues. Cell 147, (2011) 728–741.
[2] F. Randow, R.J. Youle; Self and nonself: how autophagy targets mitochondria and bacteria. Cell Host
Microbe 15, (2014) 403–411.
[3] Y. Ohsumi; Historical landmarks of autophagy research, Cell Res. 24, (2014) 9–23.
[4] W. Martinet, P. Agostinis, B. Vanhoecke, M. Dewaele, G.R.Y. De Meyer, Autophagy in disease: a
double-edged sword with therapeutic potential, Clin. Sci. 116 (2009) 697-712.
[5] X. Sui, R. Chen, Z. Wang, Z. Huang, N. Kong, M. Zhang, et al., Autophagy and chemotherapy
resistance: a promising therapeutic target for cancer treatment, Cell Death Dis. 4 (2013), 838.
[6] G. Manic, F. Obrist, G. Kroemer, I. Vitale, L. Galluzzi; Chloroquine and hydroxychloroquine for
cancer therapy, Mol. Cell Oncol. 1 (2014), 29911. Taylor & Francis.
[7] P. Pellegrini, A. Strambi, C. Zipoli, M. Hägg-Olofsson, M. Buoncervello, S. Linder, A. De Milito; Acidic
extracellular pH neutralizes the autophagy-inhibiting activity of chloroquine, Autophagy 10 (2014),
562-571.

[8] K. Satoo, N. N. Noda, H. Kumeta, Y. Fujioka, N. Mizushima, Y. Ohsumi, F. Inagaki, The structure of
Atg4B–LC3 complex reveals the mechanism of LC3 processing and delipidation during autophagy,
EMBO J. 28 (2009) 1341.
[9] T. Maruyama, N. N. Noda, Autophagy-regulating protease Atg4: structure, function, regulation and
inhibition, J. Antibiot. 71 (2018), 72–78.
[10] D. Bosc, L. Vezenkov, S. Bortnik, J. An, J. Xu, C. Choutka, A. M. Hannigan, S. Kovacic, S. Loo, P. G. K.
Clark, G. Chen, R. N. Guay-Ross, K. Yang, W. H. Dragowska, F. Zhang, N. E. Go, A. Leung, N. S. Honson, T.
A. Pfeifer, M. Gleave, M. Bally, S. J. Jones, S. M. Gorski, R. N. Young; A new quinoline-based chemical
probe inhibits the autophagyrelated cysteine protease ATG4B, Sci. Rep. 8 (2018), 11653
[11] J.Chu, Y. Fu, J. Xu, X. Zheng, Q. Gu, X. Luo, Q. Dai, S. Zhang, P. Liu, L. Hong, M. Li; ATG4B inhibitor
FMK-9a induces autophagy independent on its enzyme inhibition, Arch. Biochem. Biophys. 644 (2018) ,
29-36.
[12] M. Cleenewerck, M.O.J. Grootaert, R. Gladysz, Y. Adriaenssens, R. Roelandt, J. Joossens, A.M.
Lambeir, G. R.Y. De Meyer, W. Declercq, K. Augustyns, W. Martinet, P. V. Der Veken; Inhibitor screening
and enzymatic activity determination for autophagy target Atg4B using a gel electrophoresis-based
assay, Eur. J. Med. 123 (2016) 631–638.
[13] A. Kurdi, M. Cleenewerck, C. Vangestel, S. Lyssens, W. Declercq, J.P. Timmermans, S. Stroobants, K.
Augustyns, G. R.Y. De Meyer, P. V. Der Veken, W. Martinet; ATG4B inhibitors with a benzotropolone
core structure block autophagy and augment efficiency of chemotherapy in mice Biochem. Pharmacol.
138 (2017), 150-162.
[14] Z . Qiu, B. Kuhn, J. Aebi, X. Lin, H. Ding, Z. Zhou, Z. Xu, D. Xu, L. Han, C. Liu, H. Qiu, Y. Zhang, W. Haap,
C. Riemer, M. Stahl, N. Qin, H.C. Shen, G. Tang; Discovery of Fluoromethylketone-Based
Peptidomimetics as Covalent ATG4B (Autophagin-1) Inhibitors. ACS Med Chem Lett. 8 (2016), 802-806
[15] T. G. Nguyen, N. S. Honson, S. Arns, T. L. Davis, S. Dhe-Paganon, S. Kovacic, N. S. Kumar, T. A.
Pfeifer, R. N. Young ; Development of Fluorescent Substrates and Assays for the Key Autophagy-
Related Cysteine Protease Enzyme, ATG4B Assay Drug Dev. Techn. 12 (2014), 176-189
[16] R. Ketteler, B. Seed, Quantitation of autophagy by luciferase release assay. Autophagy, 4 (2008), 801-
806.

Blocking
Target
Outcome
ATG4B
Autophagic Vacuoles
Benzotropolones
Analysis
Atg4B inhibitory potency determination:
SDS-PAGE
Cellular Atg4B activity determination:
Luciferase Reporter, CYTO-ID

Highlights





Cancer cells rely on autophagy for survival in the tumor micro-environment
Inhibitors of autophagy could therefore be used in oncotherapy
Atg4B is one of the most interesting targets for autophagy inhibition
We report novel benzotropolone-based Atg4B-inhibitors
The most promising molecules potently inhibit autophagy at 10 µM