An Aromatic Hydroxyamide Attenuates Multiresistant Staphylococcus aureus Toxin Expression

Article abstract of DOI:10.1002/chem.201503981

Methicillin-resistant Staphylococcus aureus (MRSA) causes severe infections with only few effective antibiotic therapies currently available. To approach this challenge, chemical entities with a novel and resistance-free mode of action are desperately needed. Here, we introduce a new hydroxyamide compound that effectively reduces the expression of devastating toxins in various S. aureus and MRSA strains. The molecular mechanism was investigated by transcriptome analysis as well as by affinity-based protein profiling. Down-regulation of several pathogenesis associated genes suggested the inhibition of a central virulence-related pathway. Mass spectrometry-based chemical proteomics revealed putative molecular targets. Systemic treatment with the hydroxyamide showed significant reduction of abscess sizes in a MRSA mouse skin infection model. The absence of resistance development in vitro further underlines the finding that targeting virulence could lead to prolonged therapeutic options in comparison to antibiotics that directly address bacterial survival. Disarming MRSA: A new hydroxyamide compound effectively reduces the expression of devastating toxins in various S. aureus and MRSA strains. The molecular mechanism was investigated by transcriptome analysis, as well as by affinity-based protein profiling. Systemic treatment with the hydroxyamide showed significant reduction of abscess sizes in an MRSA mouse skin infection model.

Full text of DOI:10.1002/chem.201503981

DOI: 10.1002/chem.201503981
Full Paper
&
Drug Discovery |Hot Paper|
An Aromatic Hydroxyamide Attenuates Multiresistant
Staphylococcus aureus Toxin Expression
Jan Vomacka,[b] Vadim S. Korotkov,[b] Bianca Bauer,[b] Franziska Weinandy,[b]
Martin H. Kunzmann,[a] Joanna Krysiak,[a] Oliver Baron,[c] Thomas Bçttcher,[d] Katrin Lorenz-
Baath,[b] and Stephan A. Sieber*[a]
Chem. Eur. J. 2016, 22, 1622 – 1630
1622
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Abstract: Methicillin-resistant Staphylococcus aureus (MRSA)
causes severe infections with only few effective antibiotic
therapies currently available. To approach this challenge,
chemical entities with a novel and resistance-free mode of
action are desperately needed. Here, we introduce a new hy-
droxyamide compound that effectively reduces the expres-
sion of devastating toxins in various S. aureus and MRSA
strains. The molecular mechanism was investigated by tran-
scriptome analysis as well as by affinity-based protein profil-
ing. Down-regulation of several pathogenesis associated
genes suggested the inhibition of a central virulence-related
pathway. Mass spectrometry-based chemical proteomics re-
vealed putative molecular targets. Systemic treatment with
the hydroxyamide showed significant reduction of abscess
sizes in a MRSA mouse skin infection model. The absence of
resistance development in vitro further underlines the find-
ing that targeting virulence could lead to prolonged thera-
peutic options in comparison to antibiotics that directly ad-
dress bacterial survival.
toxins[14,15] or cell adhesion.[16] Moreover, the efficacy of thera-
peutic antibodies that block diverse virulence determinants of
S. aureus has been reported from animal studies.[17,18] However,
immunotherapeutics against a single virulence factor were
only moderately effective in phase 1 or 2 clinical trials; there-
fore, current approaches are starting to combine multiple anti-
body formulations directed against a series of key staphylococ-
cal toxins.[18] Consequently, we focused our study on the dis-
covery of anti-virulence small molecules that inhibit not only
one but multiple staphylococcal virulence factors.
Introduction
The spread of methicillin-resistant Staphylococcus aureus
(MRSA) poses a major threat to public health and represents
one of the greatest challenges for clinicians today.[1] In addition
to the ability of MRSA to cause life-threatening diseases such
as endocarditis, pneumonia, or sepsis as well as severe skin
and soft tissue infections, its high mutation frequency by
which it rapidly accumulates multiple antibiotic resistances, is
of major concern.[2]
Alternative treatment strategies targeting bacterial pathoge-
nicity are emerging, that is, inhibiting virulence instead of bac-
terial viability.[3,4] Bacterial virulence is determined by a diverse
set of molecules, including toxins and adhesins, which are
needed to establish and propagate infection.[5] Inhibition of
central virulence pathways disarms bacteria without directly af-
fecting their viability, thus leading to less selective pressure
and decreased resistance development.[6–10] Several mouse
studies have already confirmed the success of this concept by
showing that anti-virulence compounds improve the outcome
of S. aureus induced sepsis, pneumonia or skin infections.[11–15]
These approaches addressed very different targets—from inter-
ference with cholesterol biosynthesis[11–13] to the inhibition of
S. aureus virulence is controlled by complex pathways such
as the accessory gene regulator (agr) two-component
system.[19] Small autoinducing peptides (AIPs), encoded by
agrD, are secreted and sensed by a growing population.[19]
Once a critical density is reached, AIPs bind to histidine kinase
AgrC on the cell surface and induces phosphorylation of an in-
tracellular response regulator (AgrA) to activate transcription
of RNAIII, which is an effector of a broad range of diverse viru-
lence factors.[20] Synthetic derivatives of AIPs have been shown
to effectively repress this sensing mechanism.[21–23] In addition,
staphylococcal accessory regulator (SarA) and homologous
proteins such as SarX independently control the transcription
of toxins through direct binding to promoter regions.[24,25]
Herein we report the identification of a new chemical viru-
lence inhibitor (R*,R*)-3 (Figure 1a), bearing an aromatic hy-
droxyamide functional group that blocks a variety of different
staphylococcal virulence factors. Systemic treatment with the
inhibitor in mice led to a significant reduction of MRSA in-
duced skin abscesses. Together with the lack of in vitro resist-
ance development of (R*,R*)-3 treated bacteria, the new com-
pound represents a promising starting point for further medici-
nal development.
[a] Dr. M. H. Kunzmann, Dr. J. Krysiak, Prof. S. A. Sieber
Department of Chemistry, Chair of Organic Chemistry II
Center for Integrated Protein Science (CIPSM)
Institute of Advanced Studies (IAS), Technische Universität München (TUM)
Lichtenbergstrasse 4, 85747 Garching (Germany)
E-mail: stephan.sieber@tum.de
[b] J. Vomacka, Dr. V. S. Korotkov, Dr. B. Bauer, Dr. F. Weinandy,
Dr. K. Lorenz-Baath
Department of Chemistry, AVIRU, GO-Bio-project of the Federal Ministry of
Education and Research, FKZ: 031A131
Technische Universität München (TUM)
Lichtenbergstrasse 4, 85747 Garching (Germany)
[c] Dr. O. Baron
Results and Discussion
Department of Chemistry, Center for Integrated Protein Science (CIPSM)
Ludwig-Maximilians-Universität München (LMU)
Butenandtstrasse 5–13, 81377 Munich (Germany)
Hydroxyamide (R*,R*)-3 broadly inhibits virulence factors of
S. aureus in multiresistant clinical isolates
[d] Dr. T. Bçttcher
Department of Chemistry, Universität Konstanz
Universitätsstrasse 10, 78457 Konstanz (Germany)
In previous studies we identified b-lactones as specific inhibi-
tors of bacterial caseinolytic protease P (ClpP), which is a viru-
lence associated protein, that attenuates pathogenicity in bac-
terial cultures as well as in a murine infection model.[15,26]
Supporting information, including detailed experimental descriptions for all
experiments, and ORCID(s) from the author(s) for this article are available
on the WWW under http://dx.doi.org/10.1002/chem.201503981.
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1623
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Further progress was mainly limited by the low plasma sta-
bility of the b-lactones.[15] In the search for antivirulence com-
pounds with improved pharmacological properties, we opened
the lactone ring of 1 through hydrolysis or aminolysis to
obtain open-chain b-hydroxyacid 2 and b-hydroxyamide 3, re-
spectively, both as (R*,R*)-diastereomers (Figure 1a). Initially,
the biological activity of the newly synthesized substances was
investigated. To rule out an antibiotic mode of action and
focus on substances with antivirulence activity, we performed
a minimum inhibitory concentration (MIC) test with S. aureus
(NCTC 8325). Only compound 2 exhibited a MIC of 400 mm
(127 mgmLÀ1), and was thus excluded from further investiga-
tions (Figure 1a). Given that no inhibition of bacterial growth
was obtained after treatment with (R*,R*)-3 up to concentra-
tions of 1 mm (317 mgmLÀ1), we focused on this compound for
further studies.
Hemolysin alpha (hla), a pore-forming cytotoxin that causes
lysis of immune cells and erythrocytes, is one of the major
staphylococcal virulence factors. Remarkably, (R*,R*)-3 inhibited
hemolytic activity of two major S. aureus strains (NCTC 8325
Scheme 1. Two routes for the synthesis of hydroxyamide derivatives. Re-
agents and conditions: a) TBSCl, LHMDS, Et3N, THF, DMF, À78 to 08C, 2 h;
b) ZnCl2, R2CHO, CH2Cl2, RT, 4 d; c) NH3 in iPrOH, RT, 18 h; d) LHMDS, THF,
À788C, (no separation of diastereomers for compound 7; e) NaOH, H2O,
MeOH, RT, 18 h; f) NH4HCO3, Boc2O, py, CH3CN, RT, 18 h. TBS=tri-n-butylsilyl;
LHMDS=lithium hexamethyldisilazide.
and MRSA USA 300) with EC50 values of 13 mm (4.1 mgmLÀ1
;
Figure 1b) and 18 mm (5.7 mgmLÀ1; see the Supporting Infor-
mation, Figure S1a), respectively. Western blot analysis re-
vealed that reduced hemolysis was based on decreased pro-
duction of hla (Figure 1b and Figure S1a). Importantly, a broad
spectrum of 16 independent clinical isolates from different dis-
ease origins containing different resistance types (methicillin-
sensitive S. aureus (MSSA) and methicillin-resistant S. aureus
(MRSA)) were largely sensitive to treatment with (R*,R*)-3,
showing reduced hemolysis of erythrocytes (Figure 1c). Where-
as some isolates did not show effective hemolysis at all
(BK97296, VA402525, VA417350, VA12350), only two haemolytic
strains did not respond to the treatment with (R*,R*)-3. Impor-
tantly, the production of sepsis causing toxic shock syndrome
toxin (TSST-1) was also decreased with an EC50 of 7 mm
(2.22 mgmLÀ1) in TSST-1 producing strain Mu 50 (Figure 1d). In
addition, (R*,R*)-3 attenuated the production of other staphy-
lococcal virulence factors, including diarrhea causing entero-
toxins (SAE A-E; see the Supporting Information, Figure S1b) in
a concentration-dependent manner. Furthermore, the (R*,R*)-3
mediated inhibition of toxin secretion led to a protection of
eukaryotic, alveolar cells from cell death in an in vitro model of
S. aureus toxin-induced cell injury during lung infection (see
the Supporting Information, Figure S1c).
the desired compounds. The effect of the derivatives on hla in-
hibition was examined by hemolysis assay (Figure 1e). Interest-
ingly, diastereomer (R*,S*)-3 was almost inactive. In addition,
the orientation of both side chains was crucial for anti-hemo-
lytic activity, because substance 4, with exchanged substitu-
ents, showed a reduced effect in the hemolysis assay. All at-
tempts to replace the nonenyl chain, for example, by the intro-
duction of pentyl (5), phenylethyl (6), or 1-ethoxy-2-methoxy-
ethane (7), resulted in molecules with impaired activity com-
pared with (R*,R*)-3 (Figure 1e).
Unlike the parent b-lactones,[15,26] (R*,R*)-3 did not inhibit
ClpP activity in vitro (see the Supporting Information, Fig-
ure S2), suggesting a different antivirulence mode of action.
Transcriptome analysis revealed a strong impact of (R*,R*)-3
treatment on virulence factor regulation
To elucidate the inhibitory molecular mechanisms of (R*,R*)-3,
next generation sequencing was applied to analyze transcrip-
tional changes of NCTC 8325 cells after treatment with (R*,R*)-
3. Cells treated with dimethyl sulfoxide (DMSO) were used as
control. Significant (>1 or <À1log2 fold change, p<0.05) reg-
ulation of 197 genes (6.9% of all genes) was observed, with
149 genes being down-regulated and 48 being up-regulated
(Figure 2a). A DAVID database[29,30] gene cluster analysis based
on gene annotation terms revealed that a substantial number
of down-regulated genes clustered within the pyruvate metab-
olism, metal ion binding processes, and virulence associated
mechanisms (see the Supporting Information, Figure S3). Re-
markably, several enzymes connecting glycolysis with the cit-
rate acid cycle via pyruvate were significantly down-regulated,
suggesting a strong effect on central carbon metabolism. Con-
sistent with the antihemolytic effect, hla transcript levels were
To investigate structure-activity relationships we tested vari-
ous hydroxyamide derivatives. Several analogues featuring aro-
matic and aliphatic substituents as putative protein-binding li-
gands were synthesized according to two different routes
(Scheme 1). In brief, b-lactone ring-opening reactions with am-
monia in isopropanol stereoselectively led to the desired com-
pounds (Route I). trans-b-Lactones were obtained by ZnCl2-cat-
alyzed tandem Mukaiyama aldol-lactonization (TMAL)[27,28] from
aldehydes and ketene acetals. Alternatively, hydroxyamides
were synthesized by utilizing aldol condensation (Route II).
Given that the reaction is not stereoselective, a mixture of
both diastereomers was formed; these were successfully sepa-
rated by column chromatography and further converted into
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1624
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 1. (R*,R*)-3 strongly inhibits hemolysis in S. aureus. a) Ring opening of anti-virulence compound 1 through hydrolysis or aminolysis yields an open
chain b-hydroxyacid 2 and b-hydroxyamide 3, respectively, both as (R*,R*)-diastereomers. Antibiotic effects were determined by measuring minimum inhibito-
ry concentrations (MIC) up to 1 mm for three biological replicates. b) Hemolysis inhibition with (R*,R*)-3 in S. aureus NCTC 8325. The mean value of three tech-
nical replicates is shown. These findings were further confirmed by western blot analysis of hla (lower panel). Growth is indicated by colony-forming units
(CFU)/mL culture. c) Inhibition of hemolysis by 100 mm (R*,R*)-3 in clinical isolates originating from blood cultures (BK95395#, BK98122, BK98085, BK97358,
BK97472, BK97296#, BK98987), abscesses (VA423439, VA409044#, VA402525#, VA402923#, VA412350#) and pneumonia (VA418879#, IS050611#, IS050678#,
VA417350#). MRSA strains are marked by (#). The mean value of three biological replicates ÆSD is shown. Significant difference of compound treated com-
pared with DMSO control treated samples (one-sample Student’s t-test, * p<0.05, *** p<0.001). d) Toxic shock syndrome toxin 1 (TSST-1) reduction by
(R*,R*)-3 in S. aureus Mu 50 was detected by ELISA in three biological replicates. e) Structures of anti-virulence compounds and their derivatives used for struc-
ture activity relationship studies (1–7) and target identification (8). Antibiotic effects were determined by measuring minimum inhibitory concentrations (MIC)
up to 1 mm for three biological replicates. Antihemolytic activity of hydroxyamide derivatives (50 mm) 3–7: Measurements were normalized according to
DMSO control (0% activity) and (R*,R*)-3 (100% activity). The mean value of at least three biological replicates ÆSD is shown.
3.29 fold reduced (1.72 log2 fold). Several major virulence-asso-
ciated transcriptional activators such as SarA and SarR[31] were
among the strongest down-regulated genes (Figure 2b and
Figure S3in the Supporting Information). In contrast, SarX,[25]
a negative regulator of agr promoted virulence, was found to
be the only virulence-associated up-regulated gene. Thus, tran-
scriptional analysis supports the global antivirulence properties
of (R*,R*)-3.
(R*,S*)-3 with an alkyne handle at the terminal end of the alkyl
chain, and replaced the phenyl ring with a benzophenone
photocrosslinker (Scheme 2). Synthesis started with the Heck
reaction of the protected 4-bromobenzophenone with acrylic
acid. The product obtained was hydrogenated and converted
into the corresponding aldehyde. Its aldol reaction with TMS-
protected methyl undec-10-ynoate yielded a mixture of two
diastereomeric 3-hydroxyesters. Diastereomers of the 3-hydrox-
yacid obtained after deprotection steps were separated by
column chromatography and converted into the desired
photo-probes (R*,R*)-8 and (R*,S*)-8.
Chemical proteomics reveal target proteins
To identify protein target(s) of (R*,R*)-3, we synthetically
equipped the active and inactive core structures (R*,R*)-3 and
The functionalized affinity-based protein profiling (AfBPP)
probe[32] (R*,R*)-8 retained anti-hemolytic properties, albeit
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1625
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Next generation sequencing of NCTC 8325 cells treated with
100 mm (31.7 mgmLÀ1) (R*,R*)-3 compared with DMSO control. a) Fraction of
up- and down-regulated genes (>1 or <À1log2 fold change, p-value
<0.05). b) Up- and down-regulated virulence associated genes. Virulence
genes were identified by searching the DAVID[29,30] functional annotation
table for the terms: virulence, pathogenesis, cell killing and cytolysis. For
regulated DAVID database gene clusters, see the Supporting Information
(Figure S3). A–E: gene is part of the following DAVID gene clusters (enrich-
ment score): A: metal ion binding (2.27); B: virulence (2.26); C: virulence reg-
ulation (0.80). D: DNA repair, stress response (0.43); E: nucleotide binding
(0.41). See also the supporting transcriptome analysis.xlsx file.
Scheme 2. Synthesis of the affinity-based protein profiling (AfBPP) probes.
Reagents and conditions: a) acrylic acid, Pd(OAc)2, Bu3N, P(2-MeC6H4)3, o-
xylene, H2O, 1508C, 5 h, 74%; b) H2/Pd/C, MeOH, RT, 16 h, 97%; c) LiAlH4,
THF, 808C, 2 h, 50%; d) PCC, CH2Cl2, RT, 4 h, 80%; e) LHMDS, THF, À788C to
RT, 4 h, 28%; f) trifluoroacetic acid, RT, 20 h; g) NaOH, THF, H2O, MeOH, sepa-
ration of diastereomers; h) NH4HCO3, Boc2O, py, CH3CN, RT, 22%; i) NH4HCO3,
Boc2O, py, CH3CN, RT, 22%.
with reduced potency compared with the parent structure
(Figure 3a). As expected, the (R*,S*)-8 diastereomer was almost
inactive. Both probes were incubated with live S. aureus (NCTC
8325 and USA 300 Lac) cells and irradiated with UV light to es-
tablish a covalent link between the target protein and the pho-
tolinker. Cells were lysed and the probe-labeled proteome was
modified with fluorescent biotin tags through the use of click
chemistry.[33–35] Avidin bead enrichment, SDS-PAGE analysis and
fluorescent scanning revealed two intense bands at approxi-
mately 90 and 35 kDa (Figure 3b) that were analyzed by LC-
MS/MS following tryptic digest. The upper band was identified
as 3-hydroxyacyl-CoA dehydrogenase (see the Supporting In-
formation, Figure S4) and the lower band as MntC, which is
a binding domain for a manganese membrane transporter
(MntABC; see the Supporting Information, Figure S5).
and heavy-isotope tags to peptides after enrichment and
digest.[36] Differentially labeled peptides derived from cells
treated with the active probe (R*,R*)-8, the inactive probe
(R*,S*)-8 or DMSO, respectively, were mixed and analyzed by
MS (see the Supporting Information, Figure S8). Results were
ranked according to highest enrichment ratios derived from
the active probe in comparison to the DMSO control (Fig-
ure 3c) and to the inactive probe (Figure 3d). Only proteins
with enrichment higher than one-fold standard deviation (SD)
and a p value <0.01 were considered. The four hits through-
out all experiments were MntC, an uncharacterized esterase,
a proline dehydrogenase, and the FtsH metalloprotease (Fig-
ure 3e, Supporting Information ABPP_analysis.xlsx file). The
90 kDa 3-hydroxyacyl-CoA dehydrogenase was not enriched by
using this method, suggesting that it is an unspecific target of
gel-based labeling.
Importantly, labeling with the inactive probe (R*,S*)-8 re-
vealed comparable intensity of the upper band but reduced in-
tensity of the lower band (see the Supporting Information, Fig-
ure S6). Labeling in S. aureus NCTC 8325 was dose-dependent
and the lower band could be detected down to a concentra-
tion of 3.5 mm (1.5 mgmLÀ1) of (R*,R*)-8 (see the Supporting In-
formation, Figure S7).
Hit validation was performed by using selective transposon
mutants (TN mut.) of S. aureus NCTC 8325 that were obtained
by phage transduction from USA 300 Lac TN mutant strains
from the Nebraska transposon mutant library (NTML, University
of Nebraska Medical Center). Fluorescent gel analysis following
labeling with (R*,R*)-8 in the MntC-deficient strain revealed
To comprehensively identify and confirm putative target pro-
teins through quantitative proteomics, we utilized a gel-free
approach with dimethyl labeling to introduce light-, medium-,
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1626
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 4a). To examine the impact of all putative hits revealed by
gel-free MS analysis (Figure 3c–e), TN mutations of MntC, the
uncharacterized esterase, the proline dehydrogenase, and FtsH
were tested for their ability to reduce hemolysis in S. aureus
USA 300 Lac and NCTC 8325 (Figure 4b). Whereas FtsH and
the esterase showed no or only a slight reduction in hemolysis
production, respectively, a significant effect was observed for
proline dehydrogenase and MntC in the NCTC 8325 strain. It is
thus possible that multiple targets are involved in the ob-
served phenotype.
Hydroxyamide (R*,R*)-3 does not induce resistance in vitro
Targeting nonessential virulence pathways is believed to de-
crease selective pressure and limit the development of resis-
tances.[10] To test this concept, the antivirulence compound
(R*,R*)-3 and a control antibiotic ofloxacin were compared in
a resistance development assay (Figure 5). Bacterial S. aureus
cultures (NCTC 8325) were treated with either 16 mm
(5.2 mgmLÀ1) (R*,R*)-3 (EC50 concentration in the hemolysis
assay) or 3.5 mm (1.5 mgmLÀ1) ofloxacin (no growth inhibition
at this concentration based on OD600) over 12 passages. Impor-
tantly, whereas treatment with ofloxacin resulted in an in-
creased EC50 already after four passages, no change in hemoly-
sis EC50 could be observed with the antivirulence compound
(R*,R*)-3 (Figure 5). In an independent long-term experiment
with (R*,R*)-3, no resistance development could be observed
even after 45 passages (see the Supporting Information, Fig-
ure S9).
In vivo efficacy in methicillin-resistant S. aureus (MRSA) skin
infections
Figure 3. MS based target identification. a) Structure and antihemolytic ac-
tivity of 50 mm photo-probes (R*,R*)-8 and (R*,S*)-8. Measurements were nor-
malized according to DMSO control (0% activity) and (R*,R*)-3 (100% activi-
ty). The mean value of at least three biological replicates ÆSD is shown.
b) Labeling of NCTC 8325 wt with 50 mm (21.0 mgmLÀ1) photo-probe (R*,R*)-
8. c), d) Volcano plots showing p value (Àlog10 transformed) and enrichment
factor (log2 transformed) for 35 mm (R*,R*)-8 vs. DMSO control (c) and for
(R*,R*)-8 (35 mm) vs. the inactive diastereomer photo-probe (R*,S*)-8 control
(35 mm) (d). SD represents the enrichment factor standard deviation over all
proteins. Black dots represent proteins that were enriched over both con-
trols. 10 independent biological experiments (five replicates NCTC 8325 and
five replicates USA 300 Lac) were performed. Both strains were analyzed to-
gether. Separate analysis of each strain can be found in a separate excel file
(ABPP_analysis.xlsx). To be considered a hit, the mean of both enrichment
factors (active diastereomer (R*,R*)-8 over DMSO and active diastereomer
(R*,R*)-8 over inactive diastereomer (R*,S*)-8) had to be greater than the re-
spective average enrichment factor of all proteins by at least one standard
deviation (SD). In addition, the enrichment had to be significant according
to a one-sample Student’s t-test for enrichment greater than 1 (one-tailed,
p<0.01). e) Hits enriched over DMSO and inactive diastereomer photo-
probe (R*,S*)-8 control: Uniprot[37] accession numbers and protein names
from fasta entries are shown. Uncharacterized proteins were identified by
using blast (uniprot). Number of experiments (out of 10 replicates) in which
the protein was identified (N), average (av.) enrichment factors and corre-
sponding p values are shown for both controls.
In contrast to b-lactones, which undergo rapid hydrolysis,[15]
the related b-hydroxyamide (R*,R*)-3 displayed significantly im-
proved stability in both mouse and human plasma, indicating
that it may be suitable for pharmacological application (Fig-
ure 6a). We thus evaluated the phase-1 metabolism of (R*,R*)-
3 in mouse and human microsomes and observed by LC-MS
analysis a dihydroxylation product as the major metabolite.
This assignment was confirmed by chemical synthesis of the
expected metabolite 9, which exhibited an identical fragmen-
tation pattern (see the Supporting Information, Figure S10). In
vivo pharmacokinetics were performed by using a single dose
of 0.42 mg (R*,R*)-3 by intravenous (i.v.) administration in Balb/
cOlaHsd mice. Rapid elimination with a half-life of t1/2 =7.7 min
was observed. Further studies, however, showed that intraperi-
toneal (i.p.) administration of 3.17 mg of (R*,R*)-3 prolonged
the elimination half-life to 4.3 h (Figure 6b). By using a repeti-
tive dosing regimen (three times during 18 h), animals were
tightly monitored for clinical symptoms during an explorative
toxicity assessment and finally this delivery route was utilized
to determine the efficacy of (R*,R*)-3 in an in vivo S. aureus
mouse abscess model by systemic (i.p.) compound administra-
tion. We induced dorsal skin abscesses by subcutaneous infec-
tion with MRSA strain USA 300 BAA 1717 in Crl:SKH1-Hrhr hair-
less mice. To ensure optimal compound availability, we admin-
a similar labeling pattern to that of wild type (wt), except for
the loss of the strong fluorescent band at 35 kDa, confirming
the previous gel-based assignment of this band as MntC (Fig-
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1627
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
duration for 13 days. The abscess size was significantly de-
creased to about 60% compared with abscesses obtained in
vehicle treated control animals (Figure 6c).
Conclusions
The dramatic increase in bacterial resistance to commonly
used antibiotics has been paralleled by a decrease in antibiotic
drug development, leading to a poor prognosis for future
treatment of infectious diseases. As current antibiotics address
only a limited spectrum of essential targets such as cell wall
biosynthesis, DNA replication and translation, viability-inde-
pendent pathways that significantly limit bacterial pathogene-
sis represent attractive alternatives. Here, we present hydroxya-
mide (R*,R*)-3 as a potent inhibitor of S. aureus virulence. The
global reduction of toxin levels such as hla, enterotoxin SAE A-
D, and TSST-1, suggests that (R*,R*)-3 hits major target(s) that
are responsible for virulence regulation. This observation was
further validated by transcriptome analysis of hydroxyamide-
treated S. aureus cells. Parallel down-regulation of common vir-
ulence associated transcriptional activators and up-regulation
of the virulence repressor protein SarX was observed. Target
identification revealed four protein hits. Among these, proline
dehydrogenase and MntC revealed the most significant down-
regulation of hemolysis in the NCTC 8325 strain. In fact, MntC
is an essential component of the MntABC Mn2+ transport
system, which consists of an ATP-binding protein (MntA), an in-
tegral membrane transporter (MntB), and the metal-binding
lipoprotein (MntC).[38,39] A recent study demonstrated severe in-
hibition of murine systemic infection with a deletion of solely
MntC in S. aureus USA 300.[40] However, S. aureus also encodes
MntH, a second manganese uptake system related to the
Nramp metal transport family. The individual contribution of
both transporters to pathogenesis remains elusive.[41,42]
Figure 4. Validation of (R*,R*)-3 targets. a) Labeling of NCTC 8325 wt and
transposon mutants (TN mut.) with 50 mm (21.0 mgmLÀ1) photo-probe
(R*,R*)-8. b) Hemolysis assay of wild type (wt) and transposon mutants (TN
mut.) in NCTC 8325 and USA 300 Lac background. * p<0.05 (one sample t
test); four biological replicates (three biological replicates for esterase TN
mut.).
Future studies will thus investigate a putative synergistic
role of other identified targets for the broad antivirulence
mechanism. Given the inhibition of devastating toxin expres-
sion in various S. aureus isolates, the corresponding beneficial
effect in a systemic mouse abscess model with an MRSA (USA
300) strain and the lack of resistance development in vitro, aro-
matic hydroxyamides represent promising starting points for
antivirulence drug development.
Figure 5. Resistance development assay with antihemolytic compound
(R*,R*)-3 (16 mm (5.2 mgmLÀ1)) and ofloxacin (3.5 mm (1.5 mgmLÀ1)). Relative
EC50 values in hemolysis inhibition of (R*,R*)-3 and growth inhibition of
ofloxacin are plotted against the number of passages. Relative EC50: EC50 of
long-term compound treated bacteria/EC50 of long-term vehicle-treated bac-
teria (DMSO control for (R*,R*)-3 or 0.1 N NaOH control for ofloxacin). The
curves for two biological replicates are shown. An independent long-term
experiment with compound (R*,R*)-3 with no resistance development over
45 passages is shown in the Supporting Information (Figure S9).
Experimental Section
Phenotype assays: Bioactivity assays (inhibition of hemolysis, en-
terotoxins, and TSST) were performed with supernatants from
S. aureus cultures grown for 20 h at 378C. Shortly thereafter, to
monitor hemolysis, heparinized sheep blood was added and lysis
of erythrocytes was monitored. Enterotoxin and toxic shock syn-
drome toxin levels in supernatants were determined by ELISA (Ri-
dascreen from R-Biopharm and antibodies from Toxin Technology).
Detection of a-hemolysin was performed by Western Blot analysis
using ab50536 (Abcam) as the primary antibody. For the long-term
resistance assay, cultures were passaged daily with an inoculum of
each 3.4”104 colony-forming units. Animal experiments were per-
formed with Balb/cOlaHsd (Harlan) and Crl:SKH1-Hrhr (Charles
River) mice.
istered the first dose of 3.17 mg (159 mg per kg body weight)
(R*,R*)-3 one hour prior to infection and repeated the same
dosing six and 18 h after infection (Figure 6c). Abscess lesions
were fully developed after one to two days in untreated mice.
The necrotic areas were measured over the entire experimental
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1628
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 6. a) In vitro plasma levels of (R*,R*)-3 remain stable for >4 h, whereas the positive control compound procaine is rapidly disintegrated. Three technical
replicates are shown. b) In vivo pharmacokinetic parameters were determined following a single-dose i.p. injection of 3.17 mg (R*,R*)-3 in corn oil. The group
size comprised four mice (n=4) using three groups and an alternating blood withdrawal scheme. c) Systemic treatment of skin infections with compound
(R*,R*)-3 significantly (*p<0.05) decreased the size of lesions induced by S. aureus USA 300 BAA-1717 (antihemolytic activity of (R*,R*)-3 in this strain is
shown in the Supporting Information, Figure S1a). Graphs represent abscess sizes normalized to control animals at day one for control- and (R*,R*)-3-treated
animals monitored over 13 days post-infection. All results are shown as mean values ÆSEM for n=32 animals from four independent experiments. During
the course of the experiments with S. aureus BAA-1717, six animals died prematurely (two control animals and four (R*,R*)-3 treated animals).
Mode of action analysis: Transcriptome analysis was performed
Keywords: click chemistry · drug discovery · proteomics ·
medicinal chemistry · toxicology
by RNA-seq analysis described by Dugar et al.[43] Target identifica-
tion was performed with AfBPP[32] using photo-probes and quanti-
tative dimethyl-labeling for mass-spectrometry, as previously de-
scribed.[36] Validation experiments include functional studies of
transposon mutants (Nebraska Transposon Mutant Library).
[1] CDC Antibiotic resistance threats in the United States, U.S. Department
of Health and Human Services, Centers for Disease Control and Preven-
tion, 2013.
[2] M. M. Mwangi, S. W. Wu, Y. Zhou, K. Sieradzki, H. de Lencastre, P. Ri-
chardson, D. Bruce, E. Rubin, E. Myers, E. D. Siggia, A. Tomasz, Proc. Natl.
Acad. Sci. USA 2007, 104, 9451–9456.
Conflict of interest: S. Sieber, T. Bçttcher, K. Lorenz-Baath, O.
Baron, V. Korotkov, F. Weinandy are named inventors on a patent
(“Beta-O/S/N fatty acids and derivatives as anti-virulence drugs”) re-
lated to this study. S. Sieber, O. Baron, T. Bçttcher founded a compa-
ny (AVIRU).
[3] A. E. Clatworthy, E. Pierson, D. T. Hung, Nat. Chem. Biol. 2007, 3, 541–
548.
[4] D. A. Rasko, V. Sperandio, Nat. Rev. Drug Discovery 2010, 9, 117–128.
[5] S. Ruer, N. Pinotsis, D. Steadman, G. Waksman, H. Remaut, Chem. Biol.
Drug Des. 2015, 86, 379–399.
Acknowledgements
[6] D. T. Hung, E. A. Shakhnovich, E. Pierson, J. J. Mekalanos, Science 2005,
310, 670–674.
We thank the Network on Antimicrobial Resistance in Staphylo-
coccus aureus (NARSA) for the supply of the Nebraska Transpo-
son Mutant Library (NTML). Furthermore, we thank Professor
Markus Gerhard for excellent scientific advice and the permis-
sion to perform animal experiments in his group at the Insti-
tute of Medical Microbiology, Immunology and Hygiene (Tech-
nische Universität München) as well as for the supply of clinical
isolates. Animal studies were approved by the Government of
Upper Bavaria under the accreditation number 55.2-1/54-2531-
114/10. The work was funded by an ERC starting grant
(250924-antibacterials) and by the Federal Ministry of Educa-
tion and Research, Germany, FKZ: 031A131. The authors ac-
count for the content of the work. We would also like to thank
Josef Schachtner for contribution to the synthesis of photo-
reactive chemical probes and Dr. Malte Gersch for an introduc-
tion to ABPP. Finally, we thank Ernst Bernges, Heike Hofmann,
Mona Wolff, Burghard Cordes, and Katja Bäuml for excellent
technical support.
[7] D. A. Rasko, C. G. Moreira, D. R. Li, N. C. Reading, J. M. Ritchie, M. K.
Waldor, N. Williams, R. Taussig, S. Wei, M. Roth, D. T. Hughes, J. F. Hunt-
ley, M. W. Fina, J. R. Falck, V. Sperandio, Science 2008, 321, 1078–1080.
[8] V. Khodaverdian, M. Pesho, B. Truitt, L. Bollinger, P. Patel, S. Nithianan-
tham, G. Yu, E. Delaney, E. Jankowsky, M. Shoham, Antimicrob. Agents
Chemother. 2013, 57, 3645–3652.
[9] F. Tang, W.-H. Li, X. Zhou, Y.-H. Liu, Z. Li, Y.-S. Tang, X. Kou, S.-D. Wang,
M. Bao, L.-D. Qu, M. Li, B. Li, Microb. Drug Resist. 2014, 20, 357–363.
[10] R. C. Allen, R. Popat, S. P. Diggle, S. P. Brown, Nat. Rev. Microbiol. 2014,
12, 300–308.
[11] G. Y. Liu, A. Essex, J. T. Buchanan, V. Datta, H. M. Hoffman, J. F. Bastian, J.
Fierer, V. Nizet, J. Exp. Med. 2005, 202, 209–215.
[12] 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.
[13] Y. Song, C. I. Liu, F. Y. Lin, J. H. No, M. Hensler, Y. L. Liu, W. Y. Jeng, J. Low,
G. Y. Liu, V. Nizet, A. H. Wang, E. Oldfield, J. Med. Chem. 2009, 52, 3869–
3880.
[14] J. Qiu, X. Niu, J. Dong, D. Wang, J. Wang, H. Li, M. Luo, S. Li, H. Feng, X.
Deng, J. Infect. Dis. 2012, 206, 292–301.
[15] F. Weinandy, K. Lorenz-Baath, V. S. Korotkov, T. Bottcher, S. Sethi, T. Chak-
raborty, S. A. Sieber, ChemMedChem 2014, 9, 710–713.
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1629
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[16] J. Zhang, H. Liu, K. Zhu, S. Gong, S. Dramsi, Y. T. Wang, J. Li, F. Chen, R.
Zhang, L. Zhou, L. Lan, H. Jiang, O. Schneewind, C. Luo, C. G. Yang, Proc.
Natl. Acad. Sci. USA 2014, 111, 13517–13522.
[17] M. Otto, Expert Opin. Biol. Ther. 2010, 10, 1049–1059.
[18] G. Y. C. Cheung, M. Otto, Expert Opin. Ther. Targets 2012, 16, 601–612.
[19] E. A. George, T. W. Muir, ChemBioChem 2007, 8, 847–855.
[20] R. P. Novick, Mol. Microbiol. 2003, 48, 1429–1449.
[21] E. A. George, R. P. Novick, T. W. Muir, J. Am. Chem. Soc. 2008, 130, 4914–
4924.
[33] R. Huisgen, Proc. Chem. Soc. 1961, 357–396.
[34] V. V. Rostovtsev, L. G. Green, V. V. Fokin, K. B. Sharpless, Angew. Chem.
Int. Ed. 2002, 41, 2596–2599; Angew. Chem. 2002, 114, 2708–2711.
[35] C. W. Tornøe, C. Christensen, M. Meldal, J. Org. Chem. 2002, 67, 3057–
3064.
[36] J. L. Hsu, S. Y. Huang, N. H. Chow, S. H. Chen, Anal. Chem. 2003, 75,
6843–6852.
[37] UniProt, Nucleic Acids Res. 2015, 43, D204–D212.
[38] H. J. Tseng, Y. Srikhanta, A. G. McEwan, M. P. Jennings, Mol. Microbiol.
2001, 40, 1175–1186.
[39] K. M. Papp-Wallace, M. E. Maguire, Annu. Rev. Microbiol. 2006, 60, 187–
209.
[40] B. A. Diep, Q. Phung, S. Date, D. Arnott, C. Bakalarski, M. Xu, G. Naka-
mura, D. L. Swem, M. K. Alexander, H. N. Le, T. T. Mai, M.-W. Tan, E. J.
Brown, M. Nishiyama, J. Infect. Dis. 2014, 209, 1533–1541.
[41] M. J. Horsburgh, S. J. Wharton, A. G. Cox, E. Ingham, S. Peacock, S. J.
Foster, Mol. Microbiol. 2002, 44, 1269–1286.
[42] T. E. Kehl-Fie, Y. Zhang, J. L. Moore, A. J. Farrand, M. I. Hood, S. Rathi,
W. J. Chazin, R. M. Caprioli, E. P. Skaar, Infect. Immun. 2013, 81, 3395–
3405.
[22] Y. Tal-Gan, D. M. Stacy, M. K. Foegen, D. W. Koenig, H. E. Blackwell, J. Am.
Chem. Soc. 2013, 135, 7869–7882.
[23] E. J. Murray, R. C. Crowley, A. Truman, S. R. Clarke, J. A. Cottam, G. P.
Jadhav, V. R. Steele, P. O’Shea, C. Lindholm, A. Cockayne, S. R. Chhabra,
W. C. Chan, P. Williams, J. Med. Chem. 2014, 57, 2813–2819.
[24] Y. Chien, A. C. Manna, S. J. Projan, A. L. Cheung, J. Biol. Chem. 1999, 274,
37169–37176.
[25] A. C. Manna, A. L. Cheung, J. Bacteriol. 2006, 188, 4288–4299.
[26] T. Bçttcher, S. A. Sieber, J. Am. Chem. Soc. 2008, 130, 14400–14401.
[27] E. Zeiler, V. S. Korotkov, K. Lorenz-Baath, T. Bottcher, S. A. Sieber, Bioorg.
Med. Chem. 2012, 20, 583–591.
[28] S. W. Cho, D. Romo, Org. Lett. 2007, 9, 1537–1540.
[29] D. W. Huang, B. T. Sherman, R. A. Lempicki, Nat. Protoc. 2008, 4, 44–57.
[30] D. W. Huang, B. T. Sherman, R. A. Lempicki, Nucleic Acids Res. 2009, 37,
1–13.
[43] G. Dugar, A. Herbig, K. U. Forstner, N. Heidrich, R. Reinhardt, K. Nieselt,
C. M. Sharma, PLoS Genet. 2013, 9, e1003495.
[31] A. L. Cheung, A. S. Bayer, G. Zhang, H. Gresham, Y. Q. Xiong, FEMS Im-
munol. Med. Microbiol. 2004, 40, 1–9.
Received: October 5, 2015
[32] M. J. Evans, B. F. Cravatt, Chem. Rev. 2006, 106, 3279–3301.
Published online on January 8, 2016
Chem. Eur. J. 2016, 22, 1622 – 1630
www.chemeurj.org
1630
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim