GW5074

Proteomic analysis of bacterial response to a 4-hydroXybenzylidene indolinone compound, which re-sensitizes bacteria to traditional antibiotics
Clement Opoku-Temenga,b, Kenneth Ikenna Onyedibeb, Uma K. Aryalc, Herman O. Sintimb,⁎
a Graduate Program in Biochemistry, University of Maryland, College Park, MD 20742, USA
b Chemistry Department, Institute for Drug Discovery, Purdue University, West Lafayette, IN 47907, USA
c Purdue Proteomics Facility, Bindley Bioscience Center, Purdue University, West Lafayette, IN 47907, USA

A R T I C L E I N F O A B S T R A C T

Keywords:

Halogenated

4-hydroXybenzylidene

indolinones have been shown to re-sensitize methicillin-resistant

S. aureus global proteomics Quorum sensing inhibition Purine biosynthesis inhibition Lytic transglycosylase
SceD

Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecalis (VRE) to methicillin and vanco- mycin respectively. The mechanism of antibiotic re-sensitization was however not previously studied. Here, we probe the scope of antibiotic re-sensitization and present the global proteomics analysis of S. aureus treated with GW5074, a 4-hydroXybenzylidene indolinone compound. With a minimum inhibitory concentration (MIC) of 8 μg/mL against S. aureus, GW5074 synergized with beta-lactam antibiotics like ampicillin, carbenicillin and cloXacillin, the DNA synthesis inhibitor, ciprofloXacin, the protein synthesis inhibitor, gentamicin and the folate
acid synthesis inhibitor, trimethoprim. Global proteomics analysis revealed that GW5074 treatment resulted in significant downregulation of enzymes involved in the purine biosynthesis. S. aureus proteins involved in amino acid metabolism and peptide transport were also observed to be downregulated. Interestingly, anti-virulence targets such as AgrC (a quorum sensing-related histidine kinase), AgrA (a quorum sensing-related response regulator) as well as downstream targets, such as hemolysins, lipases and proteases in S. aureus were also downregulated by GW5074. We observed that the peptidoglycan hydrolase, SceD was significantly upregulated. The activity of GW5074 on S. aureus suggests that the compound primes bacteria for the antibacterial action of ineffective antibiotics.
Significance: Antibiotic resistance continues to present significant challenges to the treatment of bacterial in- fections. Given that antibiotic resistance is a natural phenomenon and that it has become increasingly difficult to discover novel antibiotics, efforts to improve the activity of existing agents are worth pursuing. A few small molecules that re-sensitize resistant bacteria to traditional antibiotics have been described but the molecular details that underpin how these compounds work to re-sensitize bacteria remain largely unknown. In this report, global label-free quantitative proteomics was used to identify changes in the proteome that occurs when GW5074, a compound that re-sensitize MRSA to methicillin, is administered to S. aureus. The identification of pathways that are impacted by GW5074 could help identify novel targets for antibiotic re-sensitization.

1. Introduction

Shortly after the introduction of antibiotics into clinical use, bac- teria that were resistant to them were also identified [1]. Antibiotic resistance is now widespread and consequently infections caused by antibiotic-resistant bacteria and deaths related to such infections con- tinue to surge, particularly in healthcare settings. Currently, deaths due to infections caused by drug-resistant bacteria claim the lives of
~700,000 people yearly and this is estimated to reach 10 million by 2050 [2,3]. Clearly, there is a need for measures to be put in place to delay if not stop the onset of the ‘post-antibiotic era’ [4]. The discovery

of novel antibiotics has significantly decreased, probably due to the closure of antibiotic discovery programs at major pharmaceutical companies and/or the fact that most “low hanging” antibacterial targets have now been targeted [5,6]. In the absence of bacterial resistance, many traditional antibiotics are capable of clearing infections. There- fore, antibiotic adjuvants, compounds that have little to no bactericidal or static properties themselves but enhance the activity of established antibiotics could improve bacterial infection management [7,8]. Sig- nificant efforts have been directed towards the identification of small molecules with antibiotic adjuvant properties [7,9]. An example of successful application of an antibiotic adjuvant is the combination of β-

⁎ Corresponding author.
E-mail address: [email protected] (H.O. Sintim).
https://doi.org/10.1016/j.jprot.2019.04.018
Received 13 December 2018; Received in revised form 21 March 2019; Accepted 23 April 2019
Availableonline25April2019
1874-3919/©2019ElsevierB.V.Allrightsreserved.

lactams with β-lactamase inhibitors, such as Augmentin (amoXycillin + clavulanic acid) [7]. Drug resistance in bacteria could also be thwarted by using antibiotic combination therapy [10,11]. The synergistic effect
obtained from combining two or more antibiotics stems from the fact that simultaneous mutation of two or more targets in bacteria is slower than mutation of a single target [11]. The combination of a novel an- tibiotic with an established drug leverages the pre-existing pharmaco- logical profiles of the approved antibiotic and reduces the risk of translating two novel compounds with unknown pharmacological and toXicological properties [12].
Although several small molecules have been found to synergize with traditional antibiotics against drug-resistant bacteria, oftentimes the molecular details of the synergistic combination is not fully elucidated [13–16]. We recently reported that GW5074, an inhibitor of cyclic di- adenylate monophosphate (c-di-AMP) synthase, re-sensitized methi- cillin-resistant Staphylococcus aureus (MRSA) and vancomycin-resistant Enterococcus faecalis (VRE) to methicillin and vancomycin respectively [17]. Efforts that delineate the mechanism(s) of compounds, which re- sensitize bacteria to traditional antibiotics, could reveal new tactics to deal with the challenge of antibiotic resistance. Here, we present the effect of GW5074 treatment on changes in global protein abundances in
S. aureus using label free quantitative shotgun proteomics approach. A total of 1322 proteins/protein families were identified in control sam- ples and 1354 proteins/protein families were identified in the GW5074 treated samples, of which 1231 proteins/protein families were common. The differences in protein expression patterns allowed us to fill some of the gaps in our knowledge regarding how GW5074 re- sensitizes MRSA to methicillin and provides potential strategies to po- tentiate the effects of β-lactams.

2. Materials and methods

2.1. Chemicals and culture methods

GW5074 was purchased from Cayman chemicals (Ann Arbor, MI, USA) and a 10 mg/mL DMSO stock solution was prepared. S. aureus ATCC 25923 was obtained from American Type Culture Collection (ATCC). MRSA ATCC 33592 purchased from ATCC was a kind gift of Dr. William Wuest (Emory University). Bacteria were routinely cultured in BD™ Tryptic Soy Broth (TSB) at 37 °C with 250 rpm shaking unless otherwise stated.

2.2. Minimum inhibitory concentration and synergy checkerboard assay

The minimum inhibitory concentration (MIC) values of GW5074 and various antibiotics against S. aureus ATCC 25923 and MRSA ATCC 33592 were determined as previously described [17]. The checker- board assay was used to determine antibiotics-compounds interactions against S. aureus ATCC 25923 and MRSA ATCC 33592. Briefly, anti- biotics and GW5074 at 10 mg/mL were prepared in appropriate di- luents. Compounds were diluted serially (1:2) along the ordinate whilst antibiotics were similarly diluted along the abscissa of 96-well micro- titer plates. Bacteria standardized using the 0.5 McFarland standard was diluted (1:100) and added to aliquoted into respective wells. The plates were incubated in a static incubator at 37 °C for 18–20 h before the MIC was determined. The fractional inhibitory concentration (FIC) index was calculated for each combination and the lowest FIC index was selected [18].
For the effect of combining GW5074 with an antibiotic, the FIC of either agent was calculated as:
FIC MIC of GW5074 in combination MIC of GW5074 alone
FIC MIC of antibiotic in combination

The cumulative fractional inhibitory concentration index ∑FICI was the calculated as:
∑ FIC = FICGW 5074 + FICantibiotic
The calculated ∑FIC indexes were interpreted as follows: synergistic (∑FIC: ≤0.5), additive (∑FIC: > 0.5 and ≤ 1), indifferent (∑FIC: > 1 and ≤ 2), antagonistic (∑FIC: > 2).

2.3. Global proteomics analysis: Sample preparation for LC-MS/MS

EXponentially growing S. aureus ATCC 25923 cells were treated with 2 μg/mL of GW5074 (MIC against S. aureus =8 μg/mL) or an equivalent amount of DMSO for 3 h. The cells were pelleted by cen-
trifugation and washed twice with PBS. The cell pellets were homo- genized in 8 M urea with Precellys® 24 Bead Mill Homogenizer (Bertin Corp., Rockville, MD, USA) and centrifuged at 14,000 rpm for 15 min at 4 °C. Precipitation of proteins was achieved by adding five equivalents (v/v) of pre-chilled acetone to the supernatants and incubating at
−20 °C overnight. After precipitation, the protein pellets were dis- solved in 8 M urea and the Pierce™ BCA Protein Assay Kit (ThermoFisher Scientific, Waltham, MA, USA) was used to determine the protein concentration. Protein (50 μg) was reduced with 10 mM dithiothreitol (DTT) at 55 °C for 45 min followed by cysteine alkylation
with 20 mM iodoacetamide at room temperature under dark for 45 min and an additional 5 mM DTT for 20 min at 37 °C. Trypsin/Lys-C MiX (Promega, Madison, WI, USA) at 1:25 (w/w) enzyme-protein ratio was used to digest the protein at 37 °C overnight and passed through C18 silica micro spin columns (The Nest Group Inc., Southborough, MA, USA). The peptides were then eluted with 0.1% formic acid (FA) in 80% acetonitrile (ACN). The eluted peptides were vacuum dried and re- suspended in 0.1% FA in 3% ACN. Peptide concentration was de- termined with the BCA assay as above and adjusted to 0.2 μg/μL.

2.4. LC-MS/MS data acquisition

A reverse-phase HPLC-ESI-MS/MS system composed of an UltiMate™ 3000 RSLCnano system coupled to a Q-EXactive (QE) High Field (HF) Hybrid Quadrupole-Orbitrap™ mass spectrometer (Thermo Fisher Scientific, Waltham, MA) and a Nano-spray Flex™ ion source (Thermo Fisher Scientific) was used to analyze the samples standard data-dependent mode. A 98% purified water/2% ACN/0.01% FA sol-
vent system was used to wash purified peptides loaded onto a trap column (300 μm ID × 5 mm, 5 μm 100 Å PepMap C18 medium) at a 5 μL/min flowrate. After 5 min, the trap column was switched in-line with the Acclaim™ PepMap™ RSLC C18 (75 μm × 15 cm, 3 μm 100 Å PepMap C18 medium, Thermo Fisher Scientific) analytical column for peptide separation. Each run constituted loading a 1 μg total peptide onto the trap column followed by a 0.3 μL/min flowrate of 0.1% formic acid (FA) in water (solvent A) and 0.1% FA in 80% ACN (solvent B) for
120 min for peptide separation at 35 °C. A 5–30% linear gradient of solvent B was run for 80 min, followed by 11 min of 45% solvent B and 2 min of 100% solvent B with an additional 7 min of isocratic flow. Solvent A was then applied at 95% for 20 min for column equilibration. A Top20 data-dependent MS/MS scan method was used to acquire the MS data. Injection time was set to 100 ms, resolution to 120, 000 at 200 m/z, spray voltage of 2-eV and an AGC target of 1 × 106 for a full MS spectra scan with a range of 400–1650 m/z. Precursor ions were fragmented at a normalized collision energy of 27 eV using a high-en- ergy C-trap dissociation. Acquisition of MS/MS scans were done at a resolution of 15,000 at m/z 200. To avoid repeated scanning of iden- tical peptides, we set the dynamic exclusion at 30 s.

2.4.1. Data analysis

antibiotic =

MIC of antibiotic alone

The MaxQuant software (v. 1.6.0.16) [19–21] with the Andromeda search engine was used to analyze the LC-MS/MS data. For protein

identification and relative quantification, the spectra were searched against the S. aureus sequences downloaded from NCBI database on May 18, 2018, with a minimal length of siX amino acids. A precursor

Table 1
Sequence of primers used in RT-PCR.

Primer name Sequence (5′-3′) Source

mass tolerance of 10 ppm, MS/MS fragment ion tolerance of 20 ppm

and enzyme specificity for trypsin and LysC (for up to two missed cleavages) were used to the database search. Also, methionine oXida- tion (M), acetyl (protein N-term), carbamyl (protein N-term) and car- bamyl lysine was set as variable modifications whilst cysteine carba-

PurL forward GTGAAGGTGCAGGGGTAGTC This study
PurL reverse ATGATTCCACCAACGCCTGT This study sarX forward GGGGTGCAACATTTTGAATACTGA This study sarX reverse TCTTTGCAATGCTTCATCGTT This study
agrA forward AACTGCACATACACGCTTACA Thaenert et al.

midomethylation (C) was set as a fiXed

modification. Peptide

[24]

quantification was performed using the ‘unique plus razor peptides’. A 1% false discovery rate (FDR) was set for both peptides spectral match and proteins identification. The Perseus software [22] was used for bioinformatics analysis. Since each treatment was done in triplicates, proteins identified in at least two out of the three replicates and with at least 2 MS/MS counts were included for further analysis. Differential expression analysis was performed using LFQ intensities. After Log2 transformation of the intensities and filtering of the data, a two-sample Student’s t-test was used to determine differentially abundant proteins using a 5% permutation-based FDR filter. Scatter plots were used to determine the correlation between replicates. The Z-score normalized data was used to perform hierarchical clustering and to generate the heat map analysis. The Log2FC values (Student’s t-test difference be- tween Log2 intensities of GW5074 and DMSO samples) and the -Log p- values were used to generate volcano plot in OriginPro 2017 Software (OriginLab, Massachusetts, USA).

2.5. STRING analysis for protein partners of select targets

The interacting partners of AgrA and SarR were searched for using the STRING 10.5 online database [23] against S. aureus Mu50. Network edges were set to display in the evidence mood and all default active interaction sources were visualized (textmining, experiments, data- bases, co-expression, neighborhood, gene fusion, co-occurrence). To limit false positive interactions, the minimum required interaction score with set to high confidence (0.7) and the maximum number of inter- actors was set to ≥10.

2.6. Total RNA isolation and RT-PCR

EXponentially growing S. aureus was incubated with 2 μg/mL GW5074 or DMSO for 3 h at 37 °C in triplicates. The cells were then pelleted by centrifugation at 5000 rpm for 5 min at 4 °C and re-
suspended in 1 mL TRIzol (Invitrogen, Carlsbad, CA) for total RNA isolation according to the manufacturer’s protocol. Residual genomic DNA was then removed by treating isolated RNA with the Turbo DNA-
free kit (Ambion, Austin, TX). The isolated RNA (1 μg) was then reverse- transcribed using the Superscript II Reverse Transcriptase
(ThermoFisher Scientific). The resulting cDNA were analyzed and quantified using gene-specific primers (Table 1) and the QuantiTect SYBR Green PCR Kit (Qiagen, Germantown, MD) on a BioRad CFX96™ Touch Real-Time PCR Detection System following the manufacturer’s protocol. PCR primers were either designed using Primer-BLAST or obtained from the referenced literature. The data were normalized against 16S rRNA and the p-values from student’s t-test showed
* ≤ 0.05, ** ≤ 0.01 and *** ≤ 0.001.

3. Results

3.1. GW5074 synergizes with different classes of antibiotics

We previously determined that GW5074 and related 4-hydro- Xybenzylidene indolinone compounds could potentiate the activity of methicillin against MRSA and VRE faecalis respectively [17]. Con- sidering that both methicillin and vancomycin are cell wall-targeting antibiotics, we wondered whether the synergistic activity was limited to cell wall targeting antibiotics. The checkerboard assay [18] was used to

agrA reverse GGCAATGAGTCTGTGAGATTT Thaenert et al. [24]
SceD forward GCAGTAGGTTTAGGAATCGTAGCAGGAAAT Dubrac et al. [25] SceD reverse CTGATTCAAAGTGATAAGTAAACCCTTCAT Dubrac et al. [25] 16S forward CGGTCCAGACTCCTACGGGAGGCAGCA Thaenert et al.
[24]
16S reverse GCGTGGACTACCAGGGTATCTAATCC Thaenert et al.
[24]

probe interactions between GW5074 and different antibiotics against S. aureus ATCC 25923 and MRSA ATCC 33592. Generally, the antibiotic- GW5074 combinations resulted in synergistic, additive and indifferent interactions (Table 2). Beta-lactam antibiotics such as ampicillin, car- benicillin, cloXacillin and methicillin appeared to preferentially sy- nergize with GW5074, particularly against the MRSA strain. Other cell wall-targeting antibiotics like bacitracin and fosfomycin were additive. The DNA synthesis inhibitor, ciprofloXacin was synergistic with GW5074 in both S. aureus and MRSA strains whilst gentamicin, a pro- tein synthesis inhibitor and trimethoprim, a folate acid synthesis in- hibitor were observed to synergize with the compound in the tested MRSA strain (Table 2).

3.2. Effect of GW5074 on global proteomics in S. aureus

GW5074 possesses some antibacterial activity (MIC of 8 μg/mL) against S. aureus, although not potent (Table 2). The array of interac- tions observed from the combination of GW5074 with various anti- biotics suggested that sub-inhibitory concentrations of the compound
compromised bacterial fitness in a manner that afforded ineffective antibiotics like methicillin to be active against MRSA. To identify pathways and proteins that are impacted by GW5074 treatment, we embarked on a global proteomics study. EXponentially growing S. aureus cells were exposed to GW5074 or DMSO (control) and the re- sulting proteomics changes were profiled. From the analysis of the proteomics data, we identified a total of 12,967 peptides matched to 1472 proteins with molecular weight ranging from 3.5 kDa to 163 kDa. Of the 12,967 peptides, 11 peptides were identified as 2-carbamyl ly- sine modification, and 1184 peptides were identified as 1-carbamyl lysine modification. Out of the 1472 proteins, we found that 27 proteins were detected in only one of the 3 biological replicates (removed from analysis) whilst 1445 proteins were identified in at least 2 biological replicates. From the 1445 proteins, 1231 proteins (~85.2%) were ob- served to be shared by both DMSO control and GW5074 whilst 91 proteins (6.3%) were only identified in DMSO control and 123 proteins (8.5%) were only identified in GW5074 (Fig. 1A). From correlation analysis using scatter plots, we observed good correlation of the bio- logical replicates (Fig. 1B & S1) as well as between samples (Fig. S2).
Hierarchical clustering of the 1231 proteins (85.2%) found in both DMSO and GW5074 samples revealed that the control and treatment samples clustered into two groups (Fig. 2A). In the presence of GW5074, 64 proteins were observed to be downregulated (p ≤ .05 and Log2FC ≤ −2) whilst 29 proteins were upregulated (p ≤ .05 and Log2FC ≥ 2) from the 1231 commonly expressed proteins (Fig. 2A & 2B). A quick glance at the downregulated proteins revealed that the enzymes involved in de novo purine biosynthesis were highly re- presented (Fig. 2B, 2C & 2D, and Table 3). Treatment of S. aureus with GW5074 caused the downregulation of PurC, PurE, PurD and PurH with

Table 2
Antibacterial activity (MIC in μg/mL) and results of GW5074-antibiotic combinations.

Antibiotics S. aureus ATCC 25923 MRSA ATCC 33592
MIC alone Combination MIC ∑FIC index Interpretation MIC alone Combination MIC ∑FIC index Interpretation
Antibiotic GW5074 Antibiotic GW5074 Antibiotic GW5074 Antibiotic GW5074
Ampicillin 0.125 8 0.03125 2 0.50 Syn 256 8 32 1 0.25 Syn
Bacitracin 64 8 2 4 0.53 Add > 128 8 64 4 0.75 Add
Carbenicillin 1 8 0.25 2 0.50 Syn > 128 8 8 2 0.28 Syn
Chloramphenicol 32 8 16 4 1.00 Add 64 8 64 4 1.50 Ind
CiprofloXacin 0.5 8 0.125 2 0.50 Syn 0.25 8 0.03125 1 0.25 Syn
CloXacillin 0.25 8 0.03125 4 0.63 Add 16 8 0.5 2 0.28 Syn
Fosfomycin 16 8 8 4 1.00 Add 512 8 256 2 0.75 Add
Gentamicin 0.25 8 0.03125 4 0.63 Add 1 8 0.125 1 0.25 Syn
Linezolid 2 8 0.25 4 0.63 Add 2 8 1 4 1.00 Add
Methicillin 2 8 1 2 0.75 Add 128 8 4 2 0.28 Syn
Trimethoprim 2 8 1 4 1.00 Add 256 8 1 4 0.50 Syn
The ∑FIC indices were interpreted as synergistic (Syn, ∑FIC: ≤0.5), additive (Add, ∑FIC: > 0.5 and ≤ 1), indifferent (Ind, ∑FIC: > 1 and ≤ 2), antagonistic (Ant,
∑FIC: > 2).

PurC being the most downregulated of the four enzymes(Log2FC =
−5.9, p-value 0.000325) (Figs. 2B & 2C, Table 3). Additionally, PurL, PurQ, PurS, PurK, PurM, PurF and PurN were only present in the DMSO samples implying that these were strongly downregulated in the pre- sence of GW5074 (Fig. 2D).
The highest downregulated protein was Map (Log2FC = −9.2) (Figs. 2B& 2C), which has been implicated in virulence. Other S. aureus virulence factors such as the type VII secretion protein EsaA (Log2FC =
−6.4), gamma-hemolysin subunits A and B, the response regulator AgrA (Log2FC = −2.2), and staphopain A (ScpA) were also down- regulated by GW5074 treatment (Fig. 2C & Table 3). Beta hemolysin (also called sphingomyelin phosphodiesterase, Hlb) was only present in the DMSO-treated sample (Fig. 2D). The histidine kinase AgrC and the glutamyl endopeptidase SspA were observed only in the DMSO-treated sample (Fig. 2D).

Aside purine biosynthesis, the downregulated proteins belonged to a range of functional classifications (Table 3). The relative abundance of several components of membrane transporters responsible for solutes (eg. OpuCC), metal ions (eg. NikA) and peptide transport (eg. OppA, MetN, MetQ) were also downregulated (Table 3). We also found that proteins involved in amino acid biosynthesis and metabolism, like SerA, a critical enzyme in L-serine biosynthesis and DapE, an enzyme in L- lysine biosynthesis pathway were downregulated significantly. Others included MetE and PatA as well as the ribosomal protein S1, RpsA were observed to be significantly downregulated. The ATP-dependent Clp protease ATP-binding subunit ClpX was also downregulated (Table 3). The ClpX ATPase, a chaperone protein interacts with ClpP protease to perform protein refolding and degradation functions (Table 3) [26,27]. One cell wall-associated protein, a LysM domain containing protein, which is involved in cell wall degradation [28] was found to be

Fig. 1. Global proteomics analysis of S. aureus cells treated with GW5074. S. aureus was treated in triplicates with GW5074 and the levels of proteins assessed. A. Venn diagram for comparison of proteins identified in DMSO-treated cells alone, GW5074-treated cells alone and in both treatments. B. Representative correlation plots for the replicates of (i) DMSO and (ii) GW5074. Additional correlation plots are presented in Fig. S1-S2.

(caption on next page)

Fig. 2. Analysis of differentially expressed protein from S. aureus cells treated with GW5074. A. Heatmap analysis of global proteomics data showing differentially expressed proteins between replicates of DMSO-treated and GW5074-treated (GW) S. aureus. B. Volcano plot of global proteomics data showing the statistical p-value (y-axis) vs the relative abundance ratio, Log2 fold change (Log2FC, X-axis). Insert represents key to the differently coloured dots. Select proteins have been labeled. Dotted horizontal line represents the adjusted p-value threshold of 0.05 whilst the dotted vertical lines represent the Log2FC cut-offs. Significant differential ex- pression was defined as either p ≤ .05 and Log2FC ≥ 2 for upregulated proteins or p ≤ .05 and Log2FC ≤ −2 for downregulated proteins. C. Bar chart representation of the top 20 proteins which were downregulated, D. Table of select proteins that were identified to be present in DMSO-treated samples only. E. Bar chart representation of top 20 upregulated proteins. The Log2FC values were plotted for the corresponding proteins using OriginPro 2017 Software (OriginLab, Massachusetts, USA). Data analysis was performed using the Perseus software [22]. Volcano plot was generated with OriginPro 2017 Software (OriginLab, Mas- sachusetts, USA).

downregulated (Log2FC =−4.4). Other downregulated proteins in- cluded those involved in carbohydrate metabolism (AdhP, LdhD, ButA etc), transcription (CspC) and translation (RpsA). A significant number of hypothetical/uncharacterized proteins, making up ~27% of identi- fied downregulated proteins were also observed (Table 3). The HTH- type transcriptional regulators SarR and SarV were also downregulated (Log2FC of −2.4 and − 2.2 respectively) by GW5074 treatment (Table 3).
The highest upregulated protein was the lytic transglycosylase, SceD (Log2FC = 8.0, p-value = 2.37 × 10−6) (Fig. 2E). Three out of the 29 upregulated proteins are involved in urea degradation. Urease subunit alpha, UreC, one of three structural urease proteins was observed to be upregulated (Fig. 2E). The other two structural proteins, UreA and UreB were overexpressed and found only in the GW5074-treated samples. Two urease accessory proteins UreE and UreG were observed to be upregulated upon treatment with GW5074 (Fig. 2E). Also, upregulated were the DNA-templated-DNA polymerase, PolX as well as the DNA helicase, DnaB. EXpression of SarX, a transcriptional regulator for the agr locus was upregulated (Log2FC = 2.09, p-value = .00042).

3.3. GW5074 affects the mRNA levels of target proteins

From the proteomics analysis, we observed the differential regula- tion of several proteins following GW5074 treatment. We performed real-time RT-PCR analysis of select targets to validate the observed differential protein levels (Fig. 3A-D). Given the extensive effect of GW5074 on purine biosynthesis, we determined the mRNA levels of purL. The mRNA levels of sceD that encodes the most upregulated protein, SceD was also analyzed. Additionally, given the effect of GW5074 on the agr locus and virulence production, we also quantified the mRNA levels of agrA and sarX, which encode the downregulated response regulator AgrA and the upregulated transcriptional regulator SarX respectively. In line with the observations from the global pro- teomics analysis, we observed decreased purL and agrA mRNA levels and increased sceD and sarX mRNA levels (Fig. 3A-D). These observa- tions suggest that GW5074 regulates the target mRNA expression, which leads to differential protein abundance.

4. Discussion

S. aureus has demonstrated the capacity to develop resistance to antibiotic therapy and hence continues to be a major health threat in both hospital and community settings [29,30]. We previously reported that the 4-hydroXybenzylidene indolinone, GW5074 and related com- pounds could inhibit c-di-AMP synthesis [17]. With modest anti- bacterial activities, the compounds were observed to synergize with methicillin against MRSA and vancomycin against VRE [17]. These prompted us to evaluate the mechanism of antibacterial action of GW5074 and related compounds against S. aureus.
The scope of antibiotic interaction of GW5074 is not limited to methicillin and vancomycin. We found that against MRSA, GW5074 could synergize with the ampicillin, carbenicillin and cloXacillin in
addition to methicillin. β-lactam antibiotics inhibit cell wall synthesis by binding to and inhibiting the transpeptidase activity of penicillin- binding proteins [31]. Resistance to methicillin and all β-lactam anti- biotics is conferred by the mecA gene present on the staphylococcal

cassette chromosome mec (SCCmec) [31]. Interactions with non-β- lactam cell wall-targeting antibiotics like bacitracin and fosfomycin were only additive. However, the re-sensitization activity of GW5074 was not limited to β-lactam antibiotics or cell wall-targeting antibiotics.
We found synergistic interactions with ciprofloXacin, gentamicin and
trimethoprim which target DNA, protein and folic acid syntheses re- spectively. The observation that GW5074 synergizes with antibiotics with different mechanisms of action suggests that the compound con- tributes significant fitness cost to bacteria, enhancing the susceptibility of resistant bacterial pathogens to other antibiotics.
The effect of GW5074 on S. aureus nucleotide metabolism was im- mediately apparent from the global proteomic analysis. Enzymes en- coded by the pur-operon (purEKCSQLFMNHD) were observed to be significantly downregulated either among the differentially expressed proteins or were completely degraded and hence found only in the DMSO control. These enzymes sequentially catalyze the de novo synthesis of inosine monophosphate (IMP) from phosphoribosyl pyr- ophosphate (Fig. 4). Downregulation of some of the enzymes following treatment with antibacterial agents is often observed in S. aureus [32–34]. Purine biosynthesis enzymes have been implicated in viru- lence, persistence and tolerance to stresses such as antibiotics in S. aureus [35,36]. For example, purC mutants were growth-defective [37]. In another study, PurL, PurN, PurM and PurE were found to be critical for S. aureus fitness in abscess formation [38]. Furthermore, purH mu- tant S. aureus was found to have significantly decreased virulence and in vivo survival compared to wildtype [39]. The global proteomics data revealed that GW5074 affected all enzymes encoded by the pur-operon (PurL, PurC, PurD, PurE and PurH were downregulated whilst PurF, PurM, PurN, PurK, PurQ and PurS were only present in the DMSO- treated sample). Such an extensive effect could be caused by the com- pound potentially modulating the transcription of the operon. The PurR transcriptional repressor of the pur-operon was however not sig- nificantly affected. Therefore, it is possible that PurR is allosterically activated upon GW5074-treatment, given that purL mRNA expression was also decreased. Varying effectors act on the expression of purR. In
E. coli, hypoXanthine and guanine co-repress PurR whilst adenine ac- tivates PurR to repress the pur-operon in B. subtilis [40,41]. The purine biosynthesis pathway is important for normal growth of S. aureus as well as effective pathogenesis. Hence a complete shut-down of the de novo purine biosynthesis pathway, such as caused by GW5074-treat- ment will have fitness costs.
From the proteomics analysis, several virulence-related proteins were downregulated upon treatment with GW5074. The highest downregulated protein, Map (MHC class II analog protein, Log2FC = − 9.2), is a cell surface protein that has been shown to bind to fibrinogen, fibronectin, vitronectin and other extracellular matriX (ECM) compo- nents of host cells [42]. Although map− mutant S. aureus was not de- ficient in adhering to ECM components [43], the mutant strains were observed to have decreased pathogenicity as abscess formation, osteo- myelitis and arthritis levels were reduced compared to wildtype strains [44]. In nude mice however, no significant difference in the virulence of
S. aureus was observed between mutant and wildtype strains, an in- dication that Map may function to modulate the immune system to facilitate S. aureus survival and virulence [44].
EsaA, an essential component of the ESAT-6 like Secretion system (ESS), was also downregulated (Log2FC = − 6.4). The ESS is a Type VII

Table 3
Functional characterization of proteins downregulated in S. aureus after treatment with GW5074.
ID Protein Classification Log2FC p-value

Nucleotide metabolism
gi|685631236 SasH Multifunctional 2′,3′-cyclic-nucleotide 2′-phosphodiesterase/5′-nucleotidase/3′-nucleotidase −6.62 0.000361
gi|685632154 PurC Phosphoribosylaminoimidazole-succinocarboXamide synthase −5.93 0.000325
gi|1145680706 PurE 5-(carboXyamino)imidazole ribonucleotide mutase −5.19 0.001552
gi|685632162 PurD Phosphoribosylamine–glycine ligase −4.05 7.2E-05
gi|685632161 PurH PhosphoribosylaminoimidazolecarboXamide formyltransferase −4.00 0.000245
Transporters
gi|685633599 NikA Nickel ABC transporter substrate-binding protein −5.47 0.000538
gi|685631614 GmpC Methionine ABC transporter substrate-binding protein −4.40 0.000126
gi|685631881 OpuCC_2 Glycine/betaine ABC transporter permease −3.40 0.0002
gi|685632054 OppA_2 Peptide ABC transporter substrate-binding protein −2.99 0.002874
gi|1145683425 ABC transporter permease −2.71 1.86E-06
gi|685631972 MetN Methionine ABC transporter ATP-binding protein −2.59 0.000233
gi|685631503 TatA Preprotein translocase subunit TatA −2.52 0.003655
gi|685633686 CopZ Copper chaperone CopZ −2.31 0.008546
gi|685633492 LctP L-lactate permease −2.13 0.000417
gi|685633639 EcsA Lantibiotic ABC transporter ATP-binding protein −2.08 4.21E-05
Virulence
gi|685633121 Map Protein map −9.19 0.000203
gi|685631438 EsaA Type VII secretion protein EsaA −6.36 4.99E-05
gi|685632243 Fib_1 Fibrinogen-binding protein −6.00 0.00034
gi|685631476 Geh Glycerol ester hydrolase;Lipase (EC 3.1.1.3) −5.66 0.001074
gi|685631499 EfeO Efem/EfeO family lipoprotein −5.23 9.83E-05
gi|685633122 LukS-PV Gamma-hemolysin subunit B −4.90 0.00042
gi|1145682818 Hel 5-nucleotidase, lipoprotein e(P4) family −3.59 0.000122
gi|685631950 EXtracellular matriX protein-binding protein emp −3.11 0.00278
gi|685633549 HlgC Gamma-hemolysin subunit A −2.96 0.002967
gi|685631717 SdrD Hydrolase −2.95 0.000146
gi|685633090 SspC Cysteine protease staphopain A −2.71 0.000661
Amino acid biosynthesis
gi|685632900 SerA D-3-phosphoglycerate dehydrogenase −5.94 0.00158
gi|685633123 DapE Succinyl-diaminopimelate desuccinylase −4.52 0.000274
gi|685632899 Aminotransferase class V −3.49 0.001886
gi|685631513 MetE 5-methyltetrahydropteroyltriglutamate–homocysteine methyltransferase −2.89 0.001276
gi|685633124 DapE Succinyl-diaminopimelate desuccinylase −2.42 4.99E-06
gi|685632136 DapC N-acetyl-L,L-diaminopimelate aminotransferase −2.10 0.000904
gi|685632551 Ald Alanine dehydrogenase −2.07 0.108511
Carbohydrate metabolism
gi|685631305 AdhE Acetaldehyde dehydrogenase −4.63 0.084893
gi|685631760 AdhP Ethanol-active dehydrogenase/acetaldehyde-active reductase −3.88 0.004166
gi|685631372 PlfB Formate acetyltransferase −3.31 0.004722
gi|685633649 LdhD Lactate dehydrogenase −2.43 0.000681
gi|685632476 4-oXalocrotonate tautomerase −2.00 0.000337
gi|685631284 ButA Acetoin reductase −2.00 0.000637
Cell envelope
gi|685631268 Peptigoglycan-binding protein LysM −4.40 0.000358
Folding, sorting and degradation
gi|685633675 ClpX Clp protease ClpX −2.90 0.000159
gi|685631411 ScdA Iron-sulfur cluster repair protein ScdA −2.74 0.027149
Signal transduction
gi|685633160 AgrA DNA-binding response regulator −2.21 0.000218
gi|685632905 DegP Serine protease −2.11 0.000801
Transcription
gi|685631953 CspC Cold shock protein −2.93 0.008146
gi|685633415 SarR MarR family transcriptional regulator −2.39 3.46E-05
gi|685633386 SarV MarR family transcriptional regulator −2.20 0.000465
Translation
gi|685632586 RpsA 30S ribosomal protein S1 −2.06 9.89E-05
Hypothetical/uncharacterized proteins
gi|685631327 SrpF Alpha-helical coiled-coil protein −4.77 0.01058
gi|685633613 Hypothetical protein −4.07 0.000503
gi|685632805 Hypothetical protein −3.98 0.001317
gi|685631975 Hypothetical protein −3.84 0.009208
gi|685633308 Hypothetical protein −3.52 0.000742
gi|685631865 Hypothetical protein −3.49 8.35E-05
gi|685631435 Hypothetical protein −3.27 0.000801
gi|685633494 Hypothetical protein −2.92 0.000139
gi|685633773 Hypothetical protein −2.67 0.001662
gi|685631768 Hypothetical protein −2.61 0.001602
(continued on next page)

Table 3 (continued)

ID Protein Classification Log2FC p-value
gi|685632106 Hypothetical protein −2.35 0.008093
gi|685632424 LSM domain protein −2.23 0.010147
gi|1145681833 Hypothetical protein −2.15 0.002061
gi|685632405 Hypothetical protein −2.12 0.001311
gi|685633778 Phage infection protein −2.12 0.000752
gi|1145683535 DUF2648 domain-containing protein −2.08 0.006552

secretion system that plays a critical role in S. aureus pathogenesis [45,46]. EsaA, EssA, EssB and EssC form the membrane components of the ESS that cooperate to transport the secreted virulence factors EsXA, EsXC, EsXB, EsXD, and EsaD. Consequently, EsaA is proposed to interact with the components of the ESS. Indeed, deletion of EsaA revealed its essential role in the transport of EsXA and EsXC [47]. Hence, down- regulation of EsaA will affect S. aureus virulence.
GW5074 also downregulated AgrA and appeared to completely abolish expression of the histidine kinase AgrC. Downregulation of AgrA was further observed to be due to decreased expression of agrA mRNA. The agr and sar loci (Figs. 5 and 6) regulate virulence in S. aureus [48,49]. The former is composed of two divergent transcripts, RNAII and RNAIII under the control of P2 and P3 promoters respec- tively [50]. The RNAII transcript encodes the components of the agr operon (agrBDCA) [50]. The autoinducer peptide (AIP) produced from AgrD is proteolytically processed by the membrane component, AgrB. The resulting AIP intermediate cleaved a second time before being exported. The exact mechanism of export is however unclear [49,51]. When a quorum is reached, AIP increases in concentration, binds and causes the autophosphorylation of AgrC histidine kinase, which in turn phosphorylates and activates AgrA response regulator. AgrA then di- rects transcription from the RNAII and RNAIII transcripts. RNAIII is a

major regulator of S. aureus virulence factor production which represses cell surface virulence factors and enhances exoproteases and toXins (Fig. 6) [49,50,52]. Downregulation of AgrA would therefore imply decreased virulence since agrA mutation resulted in the depletion of RNAIII mRNA [53]. Consequently, beta-hemolysin, gamma-hemolysin, lipase and fibrinogen-binding protein were observed to be down- regulated alongside AgrA [54]. Efforts have been made to identify in- hibitors of the agr system as potential anti-virulence therapies. An array of molecules including analogs of AIP and small molecules have been identified (reviewed in [55]).
The observation that, SarX an HTH-type transcriptional regulator was upregulated in both proteomics and RT-PCR analyses, further va- lidated the effect of GW5074 on inhibiting S. aureus agr locus. SarX is one of at least 10 homologs of SarA, a major S. aureus virulence reg- ulator. SarA is needed for the transcription of RNAII and RNAIII from P2 and P3 promoters respectively. Partial characterization of some SarA homologs including SarR, SarS, SarX, SarV, SarU, SarT, Rot and MgrA has been done [56]. SarX was found to repress the agr locus (Fig. 6) [57]. It was demonstrated that sarX mutant S. aureus had increased levels of hemolysins and proteases due to enhanced activity of agr promoters [57]. Since SarX represses agr, upregulating SarX will lead to decreased levels of virulence genes, such as those that encode

Fig. 3. Relative mRNA expression of select targets from global proteomics analysis. The effect of GW5074 treatment on the transcription of A. purL, B. agrA, C. sarX and D. sceD. Total RNA isolated from S. aureus treated with either DMSO or 2 μg/mL GW5074 was reversed transcribed and cDNAs were quantified by qRT-PCR using target-specific primers. The data represents the mean ± SD of triplicate experiments normalized with 16S RNA. Statistically significant differences between DMSO-
treatment and GW5074-treatment as determined by Student’s t-test analysis (unpaired, two-tailed) is represented as *p ≤ .05, **p ≤ .01 and ***p ≤ .001.

Fig. 4. Purine biosynthesis pathway. The enzymes encoded by the pur-operon sequentially catalyze the conversion of PRPP to IMP. PurQ and PurL are known to participate in the same step, and PurS is thought to play a role in this step. The enzymes shown in green were differentially downregulated or were only found in DMSO-treated samples. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 5. Predicted functional protein-protein association networks for select downregulated proteins. Predicted association of A. AgrA, the response regulator interacts with other components of agr system encoded by the agrBDCA operon and B. SarR transcriptional regulator with other SarA homologs. Network edges are represented by the differently colored lines in the evidence mode. Known interactions from fully curated databases (light blue lines) and experimentally determined interactions (magenta lines) are shown. Predicted interactions showing evidence of gene neighborhood (green lines), gene fusions (red lines) and gene co-occurrences (blue lines) have also been shown. Other interactions showing textmining evidence (yellow lines) as well as co-expression evidence (black lines) are shown. Figures were generated by STRING v 10.5 online database [23]. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 6. The S. aureus agr system. RNAIII modulates the expression of several virulence factors (upregu- lated factors are indicated with red arrow and downregulated factors are indicated with the green arrow). Regulation of agr by select SarA homologs is depicted as well. AgrA, AgrC, SarR and SarV were downregulated whilst SarX was upregulated in S. aureus when treated with GW5074. (For interpreta- tion of the references to colour in this figure legend, the reader is referred to the web version of this ar- ticle.)

hemolysins and proteases [57]. Additionally, the transcriptional reg- ulators SarR and SarV were downregulated. In late exponential and stationary phase cultures, SarR functions as a repressor for sarA tran- scription [58]. Analysis of a double sarA/sarR mutant at early growth phase revealed that SarR positively regulates the expression of genes encoded by the agr locus (Fig. 6), implying that down-regulating SarR would result in decreased secreted virulence factors [59]. The tran- scriptional regulator SarV, which was also downregulated by GW5074- treated regulates autolysis and some virulence factor production. Northern blot and transcriptional fusion assays demonstrated that sarV mutant had decreased RNAII and RNAIII transcripts [60]. The study also revealed that the levels of scdA was lower in the sarV mutant. The scdA gene encodes the peptidoglycan hydrolyzing protein ScdA and is implicated in autolysis [61]. Consistent with these observations, we also saw decreased abundance in ScdA (Log2FC = −2.7).
The S. aureus ClpX chaperone forms the ClpXP protease complex by interacting with ClpP protease to direct protein re-folding and de- gradation of damaged proteins. Findings from clpX mutants in S. aureus point to a function for ClpX in cell growth as clpX mutants have de- fective growth relative to wildtype [62]. In a murine skin abscess model, clpX mutants were observed to have attenuated virulence [63]. Additionally, the transcription of RNAIII, the agr effector molecule, and hence activity of AIP, were observed to be reduced in the absence of ClpX [63]. Taken together, the downregulation of AgrA, ClpX, EsaA and other virulence-related proteins could be a major contributing factor to observed anti-staphylococcal activity of GW5074.
GW5074 downregulated enzymes from different amino acid bio- synthesis pathways. In the phosphorylated pathway of L-serine bio- synthesis, the initial committed and rate limiting step is catalyzed by SerA [64]. The cellular pool of L-serine is largely controlled by the SerA- mediated pathway. The biosynthesis of pathways of purines, trypto- phan, glycine, phospholipids and cysteine utilize L-serine as a precursor [65,66]. GW5074 downregulated DapC and DapE, two successive en- zymes that catalyze the second and third steps respectively in the succinylase route of lysine biosynthesis [67,68]. DapE is involved in the meso-diaminopimelate (m-DAP), an essential component of pepti- doglycan [69]. In Helicobacter pylori, Mycobacterium smegmatis and M.

tuberculosis, dapE deletion was observed to be lethal [70–72]. Conse- quently, DapE has been suggested as a potential antibiotic target given that lysine biosynthesis pathways are not present in human cells [69]. We also saw the upregulation of SceD (Log2FC = 8.0), which was potentially caused by an increase in sceD mRNA expression as seen in the RT-PCR analysis. The S. aureus SceD is a putative lytic transglyco- sylases [73]. Lytic transglycosylases (LTs) are peptidoglycan hydrolases that cleave peptidoglycan to allow for biosynthesis and recycling of peptidoglycan as well as cell division [74]. The cleavage activity of LTs also creates space for membrane components such as secretion systems [74]. The LT activity of SceD in S. aureus was demonstrated by Sta- pleton et al. in 2007 [73]. The authors revealed that sceD mutation hindered cell separation. EXpression of SceD is positively regulated by the two-component system WalKR (also called YycFG), sigma factor B and agr whilst SarA, LytSR and SaeRS serve as negative regulators [25,73]. However, little is known about the effects of increased LT activity. A continued and uncontrolled LT activity is detrimental to the cell as it could result in autolysis [75,76]. Others have pursued the use of peptidoglycan hydrolases as potential antibacterial strategy [75,77]. For example, the bacteriophage phi MR11-derived endolysin, MV-L was shown to completely lyse drug-resistant S. aureus strains such as MRSA and vancomycin-resistant S. aureus [78]. MV-L could synergize with vancomycin against vancomycin-intermediate resistant S. aureus [78].
We observed that GW5074 preferentially synergized with β-lactam
antibiotics which target the cell wall. The observed upregulation of SceD following GW5074 treatment potentially contributes to the sy- nergy observed with cell wall-targeting antibiotics [17].

5. Conclusion

The modest antibacterial activity of GW5074 suggests that the compound presents a fitness challenge to bacteria that potentially af- fords ineffective antibiotics the opportunity to exert antibacterial ac- tion. A search for the impact of GW5074 on S. aureus viability revealed that the compound affects various bacterial processes. An extensive downregulation of the enzymes of the de novo purine biosynthesis pathway, which comprises protein encoded by the purEKCSQLFMNHD

operon was observed. Individually, these enzymes are non-essential but downregulating the entire operon is bound to have some fitness costs to bacteria. We determined by RT-PCR that the effect of GW5074 was at the mRNA level at least in the case of PurL.
GW5074 also downregulated S. aureus virulence. The decrease in virulence factor expression levels was characterized by downregulation of the AgrC-AgrA two-component system, transcriptional regulators SarR and SarV, and a corresponding decrease in abundance of down- stream targets, such as hemolysins, lipases and proteases. Future studies will further explore the effect of GW5074 and related 4-hydro- Xybenzylidene indolinone compounds on S. aureus quorum sensing.
The peptidoglycan is essential in maintaining cell integrity. The
upregulation of the peptidoglycan hydrolase SceD is another exciting effect of the treatment of S. aureus with GW5074. The enzymatic ac- tivity of SceD involves a regulated cleavage of the peptidoglycan to allow for cell wall remodeling. Understandably, upregulating SceD expression could result in an enhanced and uncontrolled activity of the peptidoglycan hydrolase with a potentially detrimental effect on bac- terial cell wall integrity. This effect of GW5074 probably contributes to the susceptibility of bacteria to cell wall targeting-antibiotics like the β-
lactams ampicillin, carbenicillin, cloXacillin and methicillin.
In summary, through proteomics analysis we showed that the modest antibacterial activity of GW5074 against S. aureus presents significant fitness challenge which enhances susceptibility to anti- biotics. The proteomics data have been deposited to the MassIVE (https://massive.ucsd.edu/) with identifier MSV000083170 and might be useful for future studies for investigating antibiotic resistance of S. aureus and other bacteria. Additionally, future studies will focus on the analysis of posttranslational modifications of differentially expressed proteins as well as endogenous protein complexes and protein-protein interactions.

Author contributions

H.O.S. supervised the project. Research was designed by C.O.T. and H.O.S.; growth and synergy experiments were performed by C.O.T.; K.I·O performed synergy experiments; LC-MS/MS experiments were performed by U.K.A; data analysis was performed by C.O.T. and U.K.A.; RT-PCR experiments were performed by C.O.T. The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

Funding sources

This work was supported by Purdue University.

Conflict of interest

The authors declare that there was no conflict of interest.

Acknowledgement

We thank Victoria Hedrick of Purdue Proteomics Facility for support in LC-MS/MS data collection and for critical reading of the manuscript. All the mass spectrometry experiments were performed at the Purdue Proteomics Facility. The Q EXactive Orbitrap HF mass spectrometer used for LC-MS/MS analysis was purchased through the funding pro- vided by the Purdue Office of the EXecutive Vice President for Research and Partnership under the large equipment grants.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https:// doi.org/10.1016/j.jprot.2019.04.018.

References

[1] T. Frieden, Antibiotic Resistance Threats in the United States, 2013, Centers for Disease Control and Prevention, Atlanta, GA, USA, 2013, p. 114.
[2] C. Nathan, O. Cars, Antibiotic resistance – problems, progress, and prospects, N. Engl. J. Med. 371 (19) (2014) 1761–1763.
[3] J. O’Neill, Review on Antimicrobial Resistance: Tackling a Crisis for the Health and Wealth of Nations, United Kingdom, London, 2014.
[4] G.D. Wright, Solving the antibiotic crisis, ACS Infect. Dis. 1 (2) (2015) 80–84.
[5] I.M. Gould, A.M. Bal, New antibiotic agents in the pipeline and how they can help overcome microbial resistance, Virulence 4 (2) (2013) 185–191.
[6] C.L. Ventola, The antibiotic resistance crisis: part 1: causes and threats, P T 40 (4) (2015) 277–283.
[7] G.D. Wright, Antibiotic adjuvants: rescuing antibiotics from resistance, Trends Microbiol. 24 (11) (2016) 862–871.
[8] L. Kalan, G.D. Wright, Antibiotic adjuvants: multicomponent anti-infective strate- gies, EXpert Rev. Mol. Med. 13 (2011) e5.
[9] R.J. Melander, C. Melander, Antibiotic adjuvants, in: J.F. Fisher, S. Mobashery,
M.J. Miller (Eds.), Antibacterials: Volume I, Springer International Publishing, Cham, 2017, pp. 89–118.
[10] A. Ahmed, A. Azim, M. Gurjar, A.K. Baronia, Current concepts in combination antibiotic therapy for critically ill patients, Indian J. Crit. Care Med. 18 (5) (2014) 310–314.
[11] J.J. Rahal, Novel antibiotic combinations against infections with almost completely resistant Pseudomonas aeruginosa and Acinetobacter species, Clin. Infect. Dis. 43 (2006) S95–S99.
[12] C.R. Chong, D.J. Sullivan, New uses for old drugs, Nature 448 (7154) (2007) 645–646.
[13] L.J. Bessa, S. Buttachon, T. Dethoup, R. Martins, V. Vasconcelos, A. Kijjoa,
P. Martins da Costa, Neofiscalin A and fiscalin C are potential novel indole alkaloid alternatives for the treatment of multidrug-resistant Gram-positive bacterial infec- tions, FEMS Microbiol. Lett. 363 (15) (2016).
[14] N. Tharmalingam, R. Rajmuthiah, W. Kim, B.B. Fuchs, E. Jeyamani, M.J. Kelso,
E. Mylonakis, Antibacterial properties of four novel hit compounds from a methi- cillin-resistant Staphylococcus aureus-Caenorhabditis elegans high-throughput screen, Microb. Drug Resist. 24 (5) (2018) 666–674.
[15] S.P. Ekambaram, S.S. Perumal, A. Balakrishnan, N. Marappan, S.S. Gajendran,
V. Viswanathan, Antibacterial synergy between rosmarinic acid and antibiotics against methicillin-resistant Staphylococcus aureus, J. Intercult. Ethnopharmacol. 5 (4) (2016) 358–363.
[16] S.F. Ishak, A.R. Ghazali, N.M. Zin, D.F. Basri, Pterostilbene enhanced anti-methi- cillin resistant Staphylococcus aureus (MRSA) activity of oXacillin, Am. J. Infect. Dis. 12 (1) (2016) 1–10.
[17] C. Opoku-Temeng, N. Dayal, J. Miller, H.O. Sintim, HydroXybenzylidene-in- dolinones, c-di-AMP synthase inhibitors, have antibacterial and anti-biofilm activ- ities and also re-sensitize resistant bacteria to methicillin and vancomycin, RSC Adv. 7 (14) (2017) 8288–8294.
[18] EUCAST, Terminology relating to methods for the determination of susceptibility of bacteria to antimicrobial agents, Clin. Microbiol. Infect. 6 (9) (2000) 503–508.
[19] J. CoX, M.Y. Hein, C.A. Luber, I. Paron, N. Nagaraj, M. Mann, Accurate proteome- wide label-free quantification by delayed normalization and maximal peptide ratio extraction, termed MaxLFQ, Mol. Cell. Proteomics 13 (9) (2014) 2513–2526.
[20] J. CoX, M. Mann, MaxQuant enables high peptide identification rates, in- dividualized p.p.b.-range mass accuracies and proteome-wide protein quantifica- tion, Nat. Biotechnol. 26 (12) (2008) 1367–1372.
[21] J. CoX, N. Neuhauser, A. Michalski, R.A. Scheltema, J.V. Olsen, M. Mann, Andromeda: a peptide search engine integrated into the MaxQuant environment, J. Proteome Res. 10 (4) (2011) 1794–1805.
[22] S. Tyanova, T. Temu, P. Sinitcyn, A. Carlson, M.Y. Hein, T. Geiger, M. Mann, J. CoX, The Perseus computational platform for comprehensive analysis of (prote)omics data, Nat. Methods 13 (9) (2016) 731–740.
[23] D. Szklarczyk, J.H. Morris, H. Cook, M. Kuhn, S. Wyder, M. Simonovic, A. Santos,
N.T. Doncheva, A. Roth, P. Bork, L.J. Jensen, C. von Mering, The STRING database in 2017: quality-controlled protein-protein association networks, made broadly accessible, Nucleic Acids Res. 45 (D1) (2017) D362–D368.
[24] R. Thänert, O. Goldmann, A. Beineke, E. Medina, Host-inherent variability influ- ences the transcriptional response of Staphylococcus aureus during in vivo infection, Nat. Commun. 8 (2017) 14268.
[25] S. Dubrac, I.G. Boneca, O. Poupel, T. Msadek, New insights into the WalK/WalR (YycG/YycF) essential signal transduction pathway reveal a major role in control- ling cell wall metabolism and biofilm formation in Staphylococcus aureus, J. Bacteriol. 189 (22) (2007) 8257–8269.
[26] D. Frees, K. Savijoki, P. Varmanen, H. Ingmer, Clp ATPases and ClpP proteolytic complexes regulate vital biological processes in low GC, Gram-positive bacteria, Mol. Microbiol. 63 (5) (2007) 1285–1295.
[27] M. Zolkiewski, A camel passes through the eye of a needle: protein unfolding ac- tivity of Clp ATPases, Mol. Microbiol. 61 (5) (2006) 1094–1100.
[28] G. Buist, A. Steen, J. Kok, O.P. Kuipers, LysM, a widely distributed protein motif for binding to (peptido)glycans, Mol. Microbiol. 68 (4) (2008) 838–847.
[29] R.V. Rasmussen, V.G. Fowler, R. Skov, N.E. Bruun, Future challenges and treatment of Staphylococcus aureus bacteremia with emphasis on MRSA, Future Microbiol 6 (1) (2011) 43–56.
[30] H. Grundmann, M. Aires-de-Sousa, J. Boyce, E. Tiemersma, Emergence and re- surgence of meticillin-resistant Staphylococcus aureus as a public-health threat, Lancet 368 (9538) (2006) 874–885.

[31] P.D. Stapleton, P.W. Taylor, Methicillin resistance in Staphylococcus aureus: me- chanisms and modulation, Sci. Prog. 85 (Pt 1) (2002) 57–72.
[32] B.P. Conlon, E.S. Nakayasu, L.E. Fleck, M.D. LaFleur, V.M. Isabella, K. Coleman,
S.N. Leonard, R.D. Smith, J.N. Adkins, K. Lewis, Activated ClpP kills persisters and eradicates a chronic biofilm infection, Nature 503 (7476) (2013) 365–370.
[33] D. Subramanian, J. Natarajan, Network analysis of S. aureus response to ramoplanin reveals modules for virulence factors and resistance mechanisms and characteristic novel genes, Gene 574 (1) (2015) 149–162.
[34] W.R. Schwan, R. Polanowski, P.M. Dunman, S. Medina-Bielski, M. Lane, M. Rott,
L. Lipker, A. Wescott, A. Monte, J.M. Cook, D.D. Baumann, V.V.N.P. Tiruveedhula,
C.M. Witzigmann, C. Mikel, M.T. Rahman, Identification of Staphylococcus aureus cellular pathways affected by the stilbenoid lead drug SK-03-92 using a microarray, Antibiotics (Basel) 6 (3) (2017).
[35] A. Kriegeskorte, D. Block, M. Drescher, N. Windmüller, A. Mellmann, C. Baum,
C. Neumann, N.I. Lorè, A. Bragonzi, E. Liebau, P. Hertel, J. Seggewiss, K. Becker,
R.A. Proctor, G. Peters, B.C. Kahl, Inactivation of thyA in Staphylococcus aureus attenuates virulence and has a strong impact on metabolism and virulence gene expression, MBio 5 (4) (2014) e01447–14.
[36] R. Yee, P. Cui, W. Shi, J. Feng, Y. Zhang, Genetic screen reveals the role of purine metabolism in Staphylococcus aureus persistence to rifampicin, Antibiotics (Basel) 4 (4) (2015) 627–642.
[37] Y. Ji, B. Zhang, S.F. Van, Y. Horn, P. Warren, G. Woodnutt, M.K. Burnham,
M. Rosenberg, Identification of critical staphylococcal genes using conditional phenotypes generated by antisense RNA, Science 293 (5538) (2001) 2266–2269.
[38] M.D. Valentino, L. Foulston, A. Sadaka, V.N. Kos, R.A. Villet, J. Santa Maria,
D.W. Lazinski, A. Camilli, S. Walker, D.C. Hooper, M.S. Gilmore, Genes contributing to Staphylococcus aureus fitness in abscess- and infection-related ecologies, MBio 5 (5) (2014) e01729–14.
[39] L. Lan, A. Cheng, P.M. Dunman, D. Missiakas, C. He, Golden pigment production and virulence gene expression are affected by metabolisms in Staphylococcus aureus, J. Bacteriol. 192 (12) (2010) 3068–3077.
[40] L.M. Meng, P. Nygaard, Identification of hypoXanthine and guanine as the co-re- pressors for the purine regulon genes of Escherichia coli, Mol. Microbiol. 4 (12) (1990) 2187–2192.
[41] M. Weng, P.L. Nagy, H. Zalkin, Identification of the Bacillus subtilis pur operon re- pressor, Proc. Natl. Acad. Sci. U. S. A. 92 (16) (1995) 7455–7459.
[42] K. Jönsson, D. McDevitt, M.H. McGavin, J.M. Patti, M. Höök, Staphylococcus aureus expresses a major histocompatibility complex class II analog, J. Biol. Chem. 270 (37) (1995) 21457–21460.
[43] B. Kreikemeyer, D. McDevitt, A. Podbielski, The role of the map protein in Staphylococcus aureus matriX protein and eukaryotic cell adherence, Int. J. Med. Microbiol. 292 (3–4) (2002) 283–295.
[44] L.Y. Lee, Y.J. Miyamoto, B.W. McIntyre, M. Höök, K.W. McCrea, D. McDevitt,
E.L. Brown, The Staphylococcus aureus map protein is an immunomodulator that interferes with T cell-mediated responses, J. Clin. Invest. 110 (10) (2002) 1461–1471.
[45] M.J. Pallen, The ESAT-6/WXG100 superfamily – and a new Gram-positive secretion system? Trends Microbiol. 10 (5) (2002) 209–212.
[46] M.L. Burts, W.A. Williams, K. DeBord, D.M. Missiakas, EsXA and EsXB are secreted by an ESAT-6-like system that is required for the pathogenesis of Staphylococcus aureus infections, Proc. Natl. Acad. Sci. U. S. A. 102 (4) (2005) 1169–1174.
[47] K.A. Aly, M. Anderson, R.J. Ohr, D. Missiakas, Isolation of a membrane protein complex for type VII secretion in Staphylococcus aureus, J. Bacteriol. 199 (23) (2017).
[48] Y. Chien, A.L. Cheung, Molecular interactions between two global regulators, sar and agr, in Staphylococcus aureus, J. Biol. Chem. 273 (5) (1998) 2645–2652.
[49] B. Wang, T.W. Muir, Regulation of virulence in Staphylococcus aureus: molecular mechanisms and remaining puzzles, Cell Chem. Biol. 23 (2) (2016) 214–224.
[50] R.P. Novick, Autoinduction and signal transduction in the regulation of staphylo- coccal virulence, Mol. Microbiol. 48 (6) (2003) 1429–1449.
[51] L. Zhang, L. Gray, R.P. Novick, G. Ji, Transmembrane topology of AgrB, the protein involved in the post-translational modification of AgrD in Staphylococcus aureus, J. Biol. Chem. 277 (38) (2002) 34736–34742.
[52] R.L. Koenig, J.L. Ray, S.J. Maleki, M.S. Smeltzer, B.K. Hurlburt, Staphylococcus aureus AgrA binding to the RNAIII-agr regulatory region, J. Bacteriol. 186 (22) (2004) 7549–7555.
[53] D. Reyes, D.O. Andrey, A. Monod, W.L. Kelley, G. Zhang, A.L. Cheung, Coordinated regulation by AgrA, SarA, and SarR to control agr expression in Staphylococcus aureus, J. Bacteriol. 193 (21) (2011) 6020–6031.
[54] B. Gray, P. Hall, H. Gresham, Targeting agr- and agr-like quorum sensing systems for development of common therapeutics to treat multiple Gram-positive bacterial infections, Sensors (Basel) 13 (4) (2013) 5130–5166.
[55] C.P. Gordon, P. Williams, W.C. Chan, Attenuating Staphylococcus aureus virulence

gene regulation: a medicinal chemistry perspective, J. Med. Chem. 56 (4) (2013) 1389–1404.
[56] A.L. Cheung, A.S. Bayer, G. Zhang, H. Gresham, Y.Q. Xiong, Regulation of virulence determinants in vitro and in vivo in Staphylococcus aureus, FEMS Immunol. Med. Microbiol. 40 (1) (2004) 1–9.
[57] A.C. Manna, A.L. Cheung, EXpression of SarX, a negative regulator of agr and exoprotein synthesis, is activated by MgrA in Staphylococcus aureus, J. Bacteriol. 188 (12) (2006) 4288–4299.
[58] A. Manna, A.L. Cheung, Characterization of sarR, a modulator of sar expression in
Staphylococcus aureus, Infect. Immun. 69 (2) (2001) 885–896.
[59] A.C. Manna, A.L. Cheung, Transcriptional regulation of the agr locus and the identification of DNA binding residues of the global regulatory protein SarR in Staphylococcus aureus, Mol. Microbiol. 60 (5) (2006) 1289–1301.
[60] A.C. Manna, S.S. Ingavale, M. Maloney, W. van Wamel, A.L. Cheung, Identification
of sarV (SA2062), a new transcriptional regulator, is repressed by SarA and MgrA (SA0641) and involved in the regulation of autolysis in Staphylococcus aureus, J. Bacteriol. 186 (16) (2004) 5267–5280.
[61] E.W. Brunskill, B.L. de Jonge, K.W. Bayles, The Staphylococcus aureus scdA gene: a novel locus that affects cell division and morphogenesis, Microbiol 143 (Pt 9) (1997) 2877–2882.
[62] A.O. Olivares, T.A. Baker, R.T. Sauer, Mechanistic insights into bacterial AAA+ proteases and protein-remodelling machines, Nat. Rev. Microbiol. 14 (1) (2016) 33–44.
[63] D. Frees, S.N. Qazi, P.J. Hill, H. Ingmer, Alternative roles of ClpX and ClpP in Staphylococcus aureus stress tolerance and virulence, Mol. Microbiol. 48 (6) (2003) 1565–1578.
[64] E. Okamura, M.Y. Hirai, Novel regulatory mechanism of serine biosynthesis asso- ciated with 3-phosphoglycerate dehydrogenase in Arabidopsis thaliana, Sci. Rep. 7 (1) (2017) 3533.
[65] G.A. Grant, Contrasting catalytic and allosteric mechanisms for phosphoglycerate dehydrogenases, Arch. Biochem. Biophys. 519 (2) (2012) 175–185.
[66] J.W. Locasale, A.R. Grassian, T. Melman, C.A. Lyssiotis, K.R. Mattaini, A.J. Bass,
G. Heffron, C.M. Metallo, T. Muranen, H. Sharfi, A.T. Sasaki, D. Anastasiou,
E. Mullarky, N.I. Vokes, M. Sasaki, R. Beroukhim, G. Stephanopoulos, A.H. Ligon,
M. Meyerson, A.L. Richardson, L. Chin, G. Wagner, J.M. Asara, J.S. Brugge,
L.C. Cantley, M.G. Vander Heiden, Phosphoglycerate dehydrogenase diverts gly- colytic fluX and contributes to oncogenesis, Nat. Genet. 43 (9) (2011) 869–874.
[67] A.J. Triassi, M.S. Wheatley, M.A. Savka, H.M. Gan, R.C. Dobson, A.O. Hudson, L,L- diaminopimelate aminotransferase (DapL): a putative target for the development of narrow-spectrum antibacterial compounds, Front. Microbiol. 5 (2014) 509.
[68] F. Grant Pearce, A.O. Hudson, K. Loomes, R.C.J. Dobson, Dihydrodipicolinate synthase: structure, dynamics, function, and evolution, Subcell. Biochem. 83 (2017) 271–289.
[69] D.M. Gillner, D.P. Becker, R.C. Holz, Lysine biosynthesis in bacteria: a metallode- succinylase as a potential antimicrobial target, J. Biol. Inorg. Chem. 18 (2) (2013) 155–163.
[70] M. Karita, M.L. Etterbeek, M.H. Forsyth, M.K. Tummuru, M.J. Blaser, Characterization of Helicobacter pylori dapE and construction of a conditionally lethal dapE mutant, Infect. Immun. 65 (10) (1997) 4158–4164.
[71] M.S. Pavelka, W.R. Jacobs, Biosynthesis of diaminopimelate, the precursor of lysine and a component of peptidoglycan, is an essential function of Mycobacterium smegmatis, J. Bacteriol. 178 (22) (1996) 6496–6507.
[72] C.M. Sassetti, D.H. Boyd, E.J. Rubin, Genes required for mycobacterial growth defined by high density mutagenesis, Mol. Microbiol. 48 (1) (2003) 77–84.
[73] M.R. Stapleton, M.J. Horsburgh, E.J. Hayhurst, L. Wright, I.M. Jonsson,
A. Tarkowski, J.F. Kokai-Kun, J.J. Mond, S.J. Foster, Characterization of IsaA and SceD, two putative lytic transglycosylases of Staphylococcus aureus, J. Bacteriol. 189 (20) (2007) 7316–7325.
[74] D.A. Dik, D.R. Marous, J.F. Fisher, S. Mobashery, Lytic transglycosylases: concinnity in concision of the bacterial cell wall, Crit. Rev. Biochem. Mol. Biol. 52 (5) (2017) 503–542.
[75] P. Szweda, M. Schielmann, R. Kotlowski, G. Gorczyca, M. Zalewska, S. Milewski, Peptidoglycan hydrolases-potential weapons against Staphylococcus aureus, Appl. Microbiol. Biotechnol. 96 (5) (2012) 1157–1174.
[76] E. Scheurwater, C.W. Reid, A.J. Clarke, Lytic transglycosylases: bacterial space- making autolysins, Int. J. Biochem. Cell Biol. 40 (4) (2008) 586–591.
[77] A.K. Sharma, S. Kumar, H. K, D.B. Dhakan, V.K. Sharma, Prediction of pepti- doglycan hydrolases- a new class of antibacterial proteins, BMC Genomics 17 (2016) 411.
[78] M. Rashel, J. Uchiyama, T. Ujihara, Y. Uehara, S. Kuramoto, S. Sugihara, K. Yagyu,
A. Muraoka, M. Sugai, K. Hiramatsu, K. Honke, S. Matsuzaki, Efficient elimination of multidrug-resistant Staphylococcus aureus by cloned lysin derived from bacter- iophage phi MR11, J. Infect. Dis. 196 (8) (2007) 1237–1247.