A Chemical Disruptor of the ClpX Chaperone Complex Attenuates the Virulence of Multidrug-Resistant Staphylococcus aureus

Article abstract of DOI:10.1002/anie.201708454

The Staphylococcus aureus ClpXP protease is an important regulator of cell homeostasis and virulence. We utilized a high-throughput screen against the ClpXP complex and identified a specific inhibitor of the ClpX chaperone that disrupts its oligomeric sta

Full text of DOI:10.1002/anie.201708454

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: A Chemical Disruptor of the ClpX Chaperone Complex
Attenuates Multiresistant Staphylococcus aureus Virulence
Authors: Christian Fetzer, Vadim S Korotkov, Robert Thänert, Kyu
Myung Lee, Martin Neuenschwander, Jens Peter von Kries,
Eva Medina, and Stephan Axel Sieber
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.201708454
Angew. Chem. 10.1002/ange.201708454
Link to VoR: http://dx.doi.org/10.1002/anie.201708454
http://dx.doi.org/10.1002/ange.201708454

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
A Chemical Disruptor of the ClpX Chaperone Complex Attenuates
Multiresistant Staphylococcus aureus Virulence
Christian Fetzer[a], Vadim S. Korotkov[a], Robert Thänert[b], Kyu Myung Lee[a], Martin
Neuenschwander[c], Jens Peter von Kries[c], Eva Medina[b] and Stephan A. Sieber*[a]
Abstract: The Staphylococcus aureus ClpXP protease is an
important regulator of cell homeostasis and virulence. Here we utilize
a high-throughput screen against the ClpXP complex and identify a
specific inhibitor of the ClpX chaperone that disrupts its oligomeric
state. Synthesis of 34 derivatives revealed that the molecular scaffold
is restrictive for diversification with only minor changes tolerated.
Subsequent analysis of the most active compound revealed strong
attenuation of S. aureus toxin production which was quantified via a
customized MS-based assay platform. Transcriptome and whole
proteome studies further confirmed the global reduction of virulence
and unraveled characteristic signatures of protein expression in
compound treated cells. Although these partially matched the pattern
of ClpX knockout cells, further depletion of toxins was observed
leading to the intriguing perspective that additional virulence pathways
may be directly or indirectly addressed by the small molecule.
beta-lactones, were shown to attenuate S. aureus virulence in
vitro and in vivo.[8] However, the facile ester bond limited
pharmacological applications and prompted the search for
alternative inhibitor scaffolds. In fact, reversible oxazole inhibitors
showed enhanced stability and potency but were rapidly ejected
from the ClpP active site upon chaperone binding.[9] Hence, we
here performed a high-throughput screen (HTS) against the
whole ClpXP complex and identified one potent hit molecule that
inhibits proteolysis by dissociation of ClpXP protein interactions
and globally reduces virulence of S. aureus and MRSA.
For screening of ClpXP inhibitors, an established assay based on
the degradation of green fluorescent protein (GFP) equipped with
a SsrA peptide-tag was used.[10] This native peptide sequence is
recognized by ClpX, which unfolds the tagged GFP under ATP
consumption. Subsequently, the linear peptide chain is digested
within the proteolytic ClpP barrel resulting in
a loss of
fluorescence signal. Putative inhibitors of proteolysis can either
target ClpP, ClpX, the interaction between these two components,
or the ATP regeneration system requiring careful validation of hits
in secondary assays. Given the high concentration of ATP in the
assay, identification of ATP binding site competitors in ClpX was
unlikely, while inhibitors of the GFP-SsrA substrate channel (low
µM KM in E. coli)[10] could be possible. We adapted the
fluorescence assay for HTS (see SI for details) and screened
about 40.000 compounds from the FMP library (Z-factor 0.69,
Figure 1A). 332 molecules were identified as primary hits
deviating from the normal distribution by three standard
deviations (z-score > 3). To select the most potent compounds
and exclude inhibitors of the ATP regenerating system, IC50
values were determined and molecules assayed against creatine
kinase in a secondary screen (Figure S1). Six compounds with
IC50 values ranging between 0.6 and 3.1 µM were selected for a
closer inspection of their mode of action.
For this we performed several assays focusing on either ClpP,
ClpX or the ClpXP complex. Interestingly, none of the hit
compounds inhibited ClpP peptidase activity up to a concentration
of 25 µM (25-fold excess), suggesting that these molecules
exhibit a novel inhibitory mechanism (Figure S2). Compounds
334 and 336, which encompass a similar structural core motif
(Figure 1B), blocked ClpX ATPase activity with an IC50 of 0.8 and
1.8 µM, respectively, while all other compounds were largely
inactive (Figure S3). Inhibition of chaperone activity is an
intriguing finding since specific ClpX inhibitors have not been
reported so far. Both hits did not alter ClpCP proteolysis
demonstrating selectivity solely for ClpX (Figure S4).
Staphylococcus aureus is a gram-positive pathogen responsible
for devastating infections of e.g. lungs, skin and bones.[1]
A
diverse arsenal of virulence factors including several proteases
and cytolysins facilitates the establishment of infection by
enabling the bacterium to thwart the immune response and
survive within the host.[2] The treatment and prevention of
staphylococcal infections with classical antibiotics has become
challenging due to the increasing prevalence of multiresistant S.
aureus strains such as methicillin-resistant S. aureus (MRSA).
Thus, alternative strategies that target bacterial virulence rather
than viability have been proposed.[3] Caseinolytic protease (ClpP)
was shown to play a crucial role in toxin expression[4] and its
genetic deletion resulted in a dramatic reduction in hemolysin
alpha (Hla) production.[5] ClpP associates with chaperones such
as ClpX that recognize and unfold substrate proteins.[6]
Accordingly, also ClpX represents a crucial target for anti-
virulence strategies.[5] While only little is known about chemical
inhibition of ClpX,[7] several specific ClpP inhibitors, foremost
[a]
C. Fetzer, V. S. Korotkov, K. M. Lee, S. A. Sieber
Center for Integrated Protein Science at the Department of
Chemistry, Technische Universität München
Lichtenbergstrasse 4, 85747 Garching (Germany)
E-mail: stephan.sieber@tum.de
[b]
[c]
R. Thänert, E. Medina
Infection Immunology Research Group, Helmholtz Centre for
Infection Research
Inhoffenstrasse 7, 38124 Braunschweig (Germany)
M. Neuenschwander, J. P. von Kries
Leibniz-Forschungsinstitut für Molekulare Pharmakologie
Robert-Roessle-Strasse 10, 13125 Berlin (Germany)
We further focused on compound 334 as the most potent ClpX
inhibitor and analyzed its mechanism of action. First, the 334
structure does not exhibit any obvious reactive electrophilic
moieties and accordingly no covalent modification of ClpX was
obtained by intact-protein mass spectrometry (Figure S5).
Second, ClpX hexamer stability was evaluated in presence and
* To whom correspondence should be addressed: S.A.S.:
stephan.sieber@tum.de
Supporting information for this article is available on the WWW under
http://dx.doi.org/
This article is protected by copyright. All rights reserved.

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
Figure 1.
HTS of 40,000 compounds revealed potent inhibitors of ClpX and ClpXP. A) Initial screen with 10 µM compound concentration yielded 332 compounds
for further validation. B) Two potent compounds, 334 and 336, share the same structural motif. C) 334 and 336 are potent inhibitors of both ClpX ATPase activity
and ClpXP protease activity (mean ± standard deviation of at least three experiments). D) Size-exclusion chromatography experiments show the disruption of ClpX-
hexamer and ClpXP-complex upon treatment with 100 µM 334.
absence of 334 via analytical size-exclusion chromatography
(Figure 1D). Importantly, a dramatic disruption of the oligomeric
state to dimeric/trimeric species was observed upon compound
addition. Third, even the whole ClpXP proteolytic complex
collapsed in response to 334 binding (Figure 1D). Fourth, 334-
induced deoligomerization of ClpX was associated with a
decrease in the melting temperature of 2.8 K as obtained in
thermal-shift assays indicating destabilization of the complex
(Figure S6). No disruption of protein-protein interactions could be
observed with the other four hit compounds implying a distinctive
mode of action.
To dissect the molecular prerequisites for 334 inhibitory activity
we prepared 31 derivatives varying in their aromatic ring
substitutions. The synthetic strategy was initiated by
condensation of indan-1,3-dione with the corresponding
benzaldehydes. The desired dihydrothiazepines were obtained by
the reaction of 2-benzylideneindan-1,3-diones with 2-
aminothiophenol (Scheme 1).
introduction of additional hydroxy- (347), bromo- (358) or
methoxy- (344) groups. However, positioning of the phenolic
hydroxy-group in meta (parent 334) and para (343) was crucial for
activity while ortho (351) resulted in a significant drop of potency.
Other substituents at meta-position including fluorine (348),
methoxy (341), amino (359) and hydroxymethyl (367) showed
reduced activity similar to the unsubstituted benzene (340).
Together with the observation that bulky substituents (365, 350)
were not tolerated, we conclude, that the A-benzene ring is like
the B-ring crucial for enzyme binding. A certain degree of
structural flexibility was observed for the C-ring where chlorine
(366) and alkynyl (356) substituents were tolerated. Oxidation of
the thioether to a sulfone (e.g. 355) resulted in a tenfold drop of
potency. Of note, some compounds, although exhibiting
comparable IC50 values, largely varied in the extent of maximum
inhibition ranging from 100 % (e.g. 334) to 8 % (e.g. 375)
Introduction of a methoxy substituent at 5 position in the upper B-
benzene ring (345) almost completely abolished inhibition of
ClpXP suggesting that this site is less suited for structural
modifications (Figure 2A). An exchange of the lower phenol ring
A by thiophene (336) retained potency while a replacement by
pyridine (352) resulted in a significantly increased IC50 value.
Interestingly, the phenol ring turned out to be amenable for the
Scheme 1. Synthesis of dihydrothiazepines. Reagents and conditions: a)
HOAc, NaOAc, reflux, 3 h; b) L-proline, MeOH, r. t., 16 h; c) 2-aminothiophenol,
EtOH, r. t., 18 h; d) 2-aminothiophenol, iPrOH/HOAc, r. t., 18 h.
This article is protected by copyright. All rights reserved.

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
(Figure 2B). This phenomenon could be caused by varying
degrees of conformational changes induced by the compound.
Especially a hydroxy group in meta position seems to be a crucial
trigger for full inhibition, however, the exact mechanism is not yet
understood.
These initial structure-activity relationship (SAR) studies revealed
flexibility for compound functionalization and guided the design of
probes for in situ target validation via affinity-based protein
profiling (AfBPP).[11] However, due to UV-lability of the core
structure and SAR restrictions for attaching the photoreactive
moiety, we concluded that development of a functional probe is,
despite massive synthetic efforts, not feasible (see SI for detailed
discussion).
Alternatively, we utilized a multidisciplinary strategy based on
toxin secretion assays, RNA sequencing and full proteome
analysis via LC-MS/MS to characterize the phenotypic effects of
334 and establish putative links with ClpX. Seminal studies by
Frees et al. already demonstrated a global reduction of toxin
secretion in a S. aureus clpX knockout strain.[5] One predominant
trait of S. aureus virulence is the secretion of hemolysins. We thus
treated S. aureus with various concentrations of 334 overnight
and applied the bacterial supernatants to sheep-blood agar plates
to assess the level of hemolysin production. As expected, a
concentration dependent reduction of hemolysis activity with an
IC50 of 3 µM was observed (Figure S7). Moreover, overall
proteolysis, an additional hallmark of staphylococcal virulence,
Figure 2. Structure-activity relationship studies of 334 analogues in ClpXP-protease assay. A) Synthesized derivatives of compound 334. B) Inhibition data of all
compounds in the ClpXP-assay. Shown are IC50 values (black bars) and the degree of inhibition (blue bars). * only tested in the HTS validation; ‡ compounds
exhibited an IC50 > 50 µM. Mean ± standard deviation from at least three experiments.
This article is protected by copyright. All rights reserved.

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
Figure 3.
Treatment with 334 reduces the transcription and synthesis of toxins and proteases by S. aureus. A) Secretome analysis of 334 (10 µM) treated S.
aureus NCTC 8325 and USA300 reveals lower levels of secreted toxins in comparison to DMSO treated cells. Red dots in volcano plots represent toxins listed in
table. B) Titration of 334 in MS experiments leads to concentration dependent LFQ detection of toxin abundances in NCTC 8325. Mean ± standard deviation from
three experiments. C) Changes in the expression levels of selected genes encoding virulence factors or regulators by S. aureus wt in response to 334 treatment.
Mean ± standard error from three experiments. D) Heatmap showing the changes in protein abundance induced by 334 treatment in wt S. aureus in comparison
with DMSO control (left lane), ΔclpX in comparison with S. aureus in DMSO control,(middle lane) and 334-treated ΔclpX compared to S. aureus in DMSO control
(right lane). A red-blue color scale depicts protein expression levels (blue: high, red: low). E) Venn diagram showing the numbers of proteins with changes in
abundance that are shared or unique between 334-treated S. aureus wt and S. aureus ΔclpX, or between DMSO-treated S. aureus ΔclpX and 334-treated S. aureus
ΔclpX.
was significantly reduced (Figure S7). To globally monitor the
expression of virulence proteins that are affected by 334, we
established an MS-based whole secretome analysis platform. S.
aureus NCTC 8325 as well as MRSA strain USA300 were
incubated overnight in the presence of 10 µM 334 or DMSO. The
bacterial supernatant was collected, proteins digested and
modified for quantitative analysis. Importantly, the visualization of
334/DMSO protein ratios of NCTC 8325 and MRSA in the
respective volcano plots (Figure 3A) shows a dramatic and global
down-regulation of major virulence factors including toxins
(hemolysins alpha and gamma, leukotoxin) and diverse proteases
(serine proteases A-F, staphopain). In addition, our novel MS-
based virulence assay allowed the individual determination of IC50
values for each detected toxin. For this purpose, LC-MS/MS
analysis was performed via label free quantification at various 334
concentrations. All IC50 values are in the low µM range
corroborating the results of the hemolysis assay (Figure 3B and
S8).
To gain more detailed insights into the inhibitory activity of 334,
we analyzed changes in the global gene expression of S. aureus
induced by treatment with 334 using RNA-seq. Strikingly, the
reduced expression of genes encoding toxins (hla, hlgA/B/C, hld,
lukS) and the proteases (aur) matched the secretome data
(Figure 3C) and are in accordance with previous studies reporting
reduced extracellular virulence factor expression in absence of
ClpX.[5] Notably, 334 treatment had a remarkable impact on the
expression of important regulatory systems such as RNAIII, Sae,
and TRAP, which control the expression of a large number of
virulence factors including several toxins and adhesins (Figure
3C). The secretome and transcriptome data demonstrate the
significant impact of 334 on S. aureus expression and production
of virulence factors.
To determine, if this effect was mediated by the 334-targeting of
ClpX, whole proteome analysis was performed on S. aureus wt
after treatment with 334 or DMSO as well as on a S. aureus ClpX
knock out (ΔclpX)[5] after incubation with 334 or DMSO. Treatment
This article is protected by copyright. All rights reserved.

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
Vomacka, C. Fetzer, M. Lakemeyer, A. Fux, S. A. Sieber, J Proteome
Res 2017, 16, 1180-1192.
of wt S. aureus with 334 resulted in reduced expression of
extracellular toxins and proteases and in an increased production
of adhesins (Figure 3D). The effect of 334 on the level of virulence
factor expression was largely comparable to that observed after
genetic deletion of ClpX (Figure 3D). Therefore, the significant
overlap of 130 proteins with changes in overall abundance (of
which 84 are regulated in the same direction) induced by 334 with
those induced by the genetic deletion of ClpX suggested, at least
in part, an inhibitory effect on ClpX (Figure 3E). Interestingly, 334
also significantly affected the expression of virulence factors and
other proteins in the ΔclpX mutant (Figure 3D and 3E). These
findings indicate that 334 could address additional targets that
further enhance the inhibitory effect on the production of
extracellular virulence factors by ClpX. Due to the restrictions of
334 for application in chemical proteomics as discussed above,
assessing the direct effect of the compound on key regulators
within the agr system will constitute the most straightforward way
for the identification of these putative targets in future studies.
In conclusion, we introduced the first reversible inhibitor of ClpX,
which retained activity against the whole proteolytic ClpXP
complex, a superior trait compared to previous reversible ClpP
binders. In addition, global transcriptome and proteome analysis
pointed towards targeting of ClpXP also in living cells. Accordingly,
a reduction of virulence expression was detected via classical
assays and more comprehensively with a MS platform allowing
reliable in situ toxin monitoring. Given the susceptibility of the
ΔclpX mutant strain towards 334 we anticipate that other cellular
virulence pathways may be directly or indirectly addressed. This
opens an intriguing perspective of a multifaceted virulence
reduction effective also for MRSA strains.
[5]
[6]
[7]
D. Frees, S. N. Qazi, P. J. Hill, H. Ingmer, Mol Microbiol 2003, 48, 1565-
1578.
a) R. T. Sauer, T. A. Baker, Annu Rev Biochem 2011, 80, 587-612; b) Z.
Maglica, K. Kolygo, E. Weber-Ban, Structure 2009, 17, 508-516.
a) L. Cheng, T. A. Naumann, A. R. Horswill, S. J. Hong, B. J. Venters, J.
W. Tomsho, S. J. Benkovic, K. C. Keiler, Protein Sci 2007, 16, 1535-
1542; b) S. M. McGillivray, D. N. Tran, N. S. Ramadoss, J. N. Alumasa,
C. Y. Okumura, G. Sakoulas, M. M. Vaughn, D. X. Zhang, K. C. Keiler,
V. Nizet, Antimicrob Agents Chemother 2012, 56, 1854-1861.
a) T. Böttcher, S. A. Sieber, J Am Chem Soc 2008, 130, 14400-14401;
b) F. Weinandy, K. Lorenz-Baath, V. S. Korotkov, T. Bottcher, S. Sethi,
T. Chakraborty, S. A. Sieber, ChemMedChem 2014, 9, 710-713.
A. Pahl, et al., Angew Chem Int Ed Engl 2015, 54, 15892-15896.
[8]
[9]
[10] Y. I. Kim, R. E. Burton, B. M. Burton, R. T. Sauer, T. A. Baker, Mol Cell
2000, 5, 639-648.
[11] P. P. Geurink, L. M. Prely, G. A. van der Marel, R. Bischoff, H. S.
Overkleeft, Top Curr Chem 2012, 324, 85-113
[12] J. A. Vizcaíno, et al., Nucleic Acids Res 2016, 44, D447–D456.
Acknowledgements
This work was funded by the DFG (SI1096/8-1) and the Center
for Integrated Protein Science Munich. K. M. L. thanks the Korea
Research Institute of Chemical Technology. We thank Dorte
Frees for providing 8325-4 wt and ΔclpX strains. We also thank
Carola Seyffarth and Jessica Pryzgodda (FMP Berlin) for
realization of the HTS, Katja Bäuml and Mona Wolff for excellent
technical assistance, Melina Vollmer and Theresa Rauh for help
with experiments and Markus Lakemeyer and Matthias Stahl for
critical proofreading. The MS data have been deposited to the
ProteomeXchange Consortium via the PRIDE[12] partner
repository with the dataset identifier PXD007259.
Keywords: Antivirulence • high-throughput screening •
proteolysis • proteomics • Staphylococcus aureus
[1]
[2]
[3]
F. D. Lowy, N Engl J Med 1998, 339, 520-532.
A. Zecconi, F. Scali, Immunol Lett 2013, 150, 12-22.
a) A. E. Clatworthy, E. Pierson, D. T. Hung, Nat Chem Biol 2007, 3, 541-
548; b) C. I. Liu, G. Y. Liu, Y. Song, F. Yin, M. E. Hensler, W. Y. Jeng, V.
Nizet, A. H. Wang, E. Oldfield, Science 2008, 319, 1391-1394; c) D. A.
Rasko, V. Sperandio, Nat Rev Drug Discov 2010, 9, 117-128.
a) D. Frees, K. Sorensen, H. Ingmer, Infect Immun 2005, 73, 8100-8108;
b) A. Michel, F. Agerer, C. R. Hauck, M. Herrmann, J. Ullrich, J. Hacker,
K. Ohlsen, J Bacteriol 2006, 188, 5783-5796; c) J. Krysiak, M. Stahl, J.
[4]
This article is protected by copyright. All rights reserved.

10.1002/anie.201708454
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Christian Fetzer, Vadim S. Korotkov,
Robert Thänert, Kyu Myung Lee, Martin
Neuenschwander, Jens Peter von Kries,
Eva Medina, Stephan A. Sieber*
Page No. – Page No.
A Chemical Disruptor of the ClpX
Chaperone Complex Attenuates
Multiresistant Staphylococcus aureus
Virulence
A high-throughput screen against the ClpXP protease revealed the first non-
covalent inhibitors of the chaperone ClpX. Binding of the inhibitors induces
disruption of the ClpX hexamer and even of the whole ClpXP proteolytic complex.
Treatment of Staphylococcus aureus with the compounds leads to a global down-
regulation of a plethora of secreted virulence factors.
This article is protected by copyright. All rights reserved.