Silencing of Salmonella typhimurium Pathogenesis: Atenolol Acquires Efficient Anti-Virulence Activities

The targeting of bacterial virulence is proposed as a promising approach to overcoming the bacterial resistance development to antibiotics. Salmonella enterica is one of the most important gut pathogens that cause a wide diversity of local and systemic illnesses. The Salmonella virulence is controlled by interplayed systems namely Quorum sensing (QS) and type three secretion system (T3SS). Furthermore, the Salmonella spy on the host cell via sensing the adrenergic hormones enhancing its virulence. The current study explores the possible anti-virulence activities of β-adrenoreceptor blocker atenolol against S. enterica serovar Typhimurium in vitro, in silico, and in vivo. The present findings revealed a significant atenolol ability to diminish the S. typhimurium biofilm formation, invasion into HeLa cells, and intracellular replication inside macrophages. Atenolol significantly downregulated the encoding genes of the T3SS-type II, QS receptor Lux analogs sdiA, and norepinephrine membranal sensors qseC and qseE. Moreover, atenolol significantly protected mice against S. typhimurium. For testing the possible mechanisms for atenolol anti-virulence activities, an in silico molecular docking study was conducted to assess the atenolol binding ability to QS receptor SdiA and norepinephrine membranal sensors QseC. Atenolol showed the ability to compete on the S. typhimurium targets. In conclusion, atenolol is a promising anti-virulence candidate to alleviate the S. typhimurium pathogenesis by targeting its QS and T3SS systems besides diminishing the eavesdropping on the host cells.


Introduction
Quorum sensing (QS) is a way that bacteria use to communicate with each other allowing specific processes to be regulated, such as virulence factors expression, biofilm formation, and secondary metabolites production in addition to stress adaptation mechanisms such as bacterial competition systems including secretion systems [1,2]. An interkingdom crosstalk was observed between bacteria and host cells, in which bacteria could sense the 3 of 16 Tryptic Soy Broth (TSB) provided with 0.001 µM AHL [12,19]. The cultures were adjusted to cell density 1 × 10 6 CFU/mL (OD600 = 0.4) before each experiment.

Chemicals and Microbiological Media
All the chemicals used in this study were of pharmaceutical grade. The microbiological media were ordered from Oxoid (Hampshire, UK). Atenolol (CAS number: 29122-68-7), N-hexanoyl-DL-homoserine lactone (AHL) (CAS Number: 106983-28-2), thiamine (CAS number: 67-03-8), DL-norepinephrine hydrochloride (CAS Number: 55-27-6), and Dulbecco's Modified Eagle's Medium (DMEM) medium were obtained from Sigma-Aldrich (St. Louis, MO, USA). It is worthy to mention that used atenolol was in form of hydrochloride salt and it was dissolved in water and kept in stocks prior to use in all the next experiments.

Determination of Minimum Inhibitory Concentrations (MICs) and Atenolol Effect on Bacterial Growth
The broth microdilution method was used to determine the atenolol MICs against E. coli K-12 and S. typhimurium according to the Clinical Laboratory and Standards Institute Guidelines (CLSI, 2020) [44].
The effect of atenolol at sub-MIC (1/5 MIC) on bacterial growth was evaluated as previously described [31,45]. Briefly, the bacterial viable counts of cultures provided or not with atenolol at 1/5 MIC at different time points were performed.

Evaluation of Adhesion and Biofilm Formation
E. coli K-12 and S. typhimurium were cultivated in the growth conditions mentioned above before assaying the bacterial adhesion and the biofilm formation in the presence or absence of atenolol at sub-MIC. As described previously, the absorbances of crystal violet staining adhered or biofilm forming bacterial cells in treated or untreated cultures after 1 h or 24 h were used to assess the effect on adhesion and biofilm formation [13,46]. Briefly, Overnight cultures were cultivated, diluted with tryptone soya broth (TSB) and the optical densities were adjusted to a cell density of 1 × 10 6 CFU/mL, 200 µL aliquots of the prepared bacterial suspensions were transferred into sterile 96-well polystyrene microplates and incubated at 37 • C overnight. The non-adherent cells were washed out and the, while the adherent cells were fixed with 99% methanol for 30 min, and stained with 1% crystal violet for 20 min. The unattached stain was washed off, and the plates were left to dry. The crystal violet dye was extracted by 33% glacial acetic and the optical densities were measured at 590 nm. The Atenolol inhibitory effect on biofilm formation was visualized by allowing the formation of biofilm coverslips in the presence or absence of atenolol at sub-MIC, as described before [11,12].

Evaluation of S. typhimurium Invasion and Intracellular Replication
The S. typhimurium internalization within HeLa and macrophage cells in the presence or absence of atenolol at sub-MIC was performed by gentamicin protection assay [12,19,47]. The S. typhimurium cultures were grown in the presence or absence of atenolol at sub-MIC in the conditions described above in the presence of 0.001 µM AHL, and a master mix (1 × 10 5 bacteria/well) with multiplicity of infection (MOI 1) was prepared in DMEM in 24well plates. For invasion and intracellular replication assays, HeLa cells (5 × 10 5 cells/well) or RAW264.7 (2 × 10 5 cells/well) were used for seeding the DMEM medium. After half hour, the wells were washed with pre-warmed phosphate buffer saline (PBS). The gentamycin at 100 µg/mL was added for 1hr to kill the extracellular adhered bacterial cells. For invasion assay, HeLa cells were lysed with 0.1% Triton X-100 at 25 • C for 25 min. Viable counts for inoculum and lysates were performed. For evaluation of the intracellular replication, the infected macrophages were washed with PBS and lysed after 2 and 16 h with 0.1% TritonX-100 for 25 min. The viable counts were performed for inoculum and lysates, and the phagocytosed cell numbers/relative untaken cells (2 h against inoculum) and x-fold intracellular replication (16 h against 2 h) was calculated. The Salmonella-infected HeLa or macrophages were immunostained as described earlier (3,14,23). The cell lines infection was performed as described above on cover slips and then Infected cells were fixed with 2% paraformaldehyde for 25 min. Bovine serum albumin (BSA) (2%) was used as a blocking solution for 1 h. Rabbit anti-Salmonella O antigen (Difco, BD; San Joes, CA, USA) were added to fixed cells for 3 h. Next, anti-rabbit tagged with a green fluorescent protein (GFP) (Abcam; Eugene, OR, USA) was added, and the mixture was left for 1 h. The macrophages were counter-stained with blue fluorescent diamidino-2-phenylindole dye (DAPI) (Thermo Fisher Scientific; Bothell, WA, USA) for 1hr. A confocal laser scanning microscope LSM780 (Carl Zeiss, Jena, Germany) was used to take photos.
Briefly, S. typhimurium cultures were overnight cultured with atenolol at sub-MIC or norepinephrine (50 µg/mL) at 37 • C and then about 1 × 10 6 CFU/mL bacterial cells were suspended in PBS. Five mice groups were assigned, each comprises 10 female healthy (3-weeks old) mice. The first and second groups were either not injected or intraperitoneally injected with sterile PBS. The other three groups were intraperitoneally injected with 100 µL of untreated S. typhimurium or S. typhimurium treated with atenolol at sub-MIC or norepinephrine (50 µg/mL). The experiment was performed over 5 days and mice survival was plotted by Kaplan-Meier method.

Ligand and Protein Preparations
The 3D structure of atenolol was built on the MOE Builder within the Molecular Operating Environment (MOE 2019.012) using the SMILES obtained from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/) (accessed on 24 May 2022). The ligand was energy minimized and protonated at physiological pH 7.4.
The crystal structures of the target proteins (PDB ID: 4LFU, and 3JZ3 for E. coli SdiA, and E. coli QseC, respectively) were downloaded from the RCSB Protein Data Bank (https://www.rcsb.org/ accessed on 4 May 2022). The proteins were prepared for docking using a QuickPrep panel to minimize energy, protonate, fix, and tether atoms along with deleting unnecessary water molecules.

Docking Experiments
The binding pocket of the targets was determined by the MOE Site Finder and the position of the co-crystallized ligand. The Computed Atlas for Surface Topography of Proteins (CASTp; http://sts.bioe.uic.edu/castp/index.html, accessed on 28 May 2022) was used to calculate the pocket area/volume of the proteins [51]. Re-docking of the co-crystallized ligand was performed to validate the use of the protein in structure-based drug design.
Docking for the ligands in the active site was performed using Molecular Operating Environment (MOE) 2019.0102 with Alpha triangle placement, Amber10:EHT forcefield, refinement with forcefield and scoring using Affinity dG as the default settings.

Statistical Analysis
The statistical significance was examined by the student's t-test (unless mentioned), where a p value < 0.05 is considered significant (GraphPad Prism Software, v.8, San Diego, CA, USA). The experiments including effect on bacterial growth, biofilm formation, internalization in HeLa cells or macrophages and quantification of the expression of tested genes were conducted in triplicate, and the data are presented as the means ± standard error.

Determination of MICs and Its Effect on Bacterial Growth
The MICs were considered the lowest atenolol concentration that can inhibit bacterial growth. Atenolol inhibited the growth of both S. typhimurium and E. coli K-12 at 2 mg/mL.
To exclude the effect of atenolol on bacterial growth, the antivirulence activities were evaluated at sub-MIC (1/5 MIC). For further testing, the bacterial growth was evaluated in the presence of atenolol at sub-MIC and compared with the growth of untreated bacterial cultures. There were no significant differences between viable counts of S. typhimurium and E. coli K-12 in the presence or absence of atenolol at sub-MIC ( Figure 1).

Effect on Bacterial Adhesion and Biofilm Formation
The inhibitory effect of atenolol on S. typhimurium and E. coli K-12 adhesion and biofilm formation was assessed using the crystal violet method. The adhered or biofilm forming bacterial cells were stained with crystal violet after 2 h or 24 h, respectively. Captured light microscope images of formed biofilms on cover slips showed the marked effect of atenolol on diminishing the biofilm formation ( Figure 2A). The absorbances of the extracted crystal violet staining the bacterial cells treated with atenolol were photometrically measured and compared to the absorbances of untreated cultures. Atenolol significantly decreased bacterial adhesion and biofilm formation; the data are presented as percentage changes from untreated control ( Figure 2B). growth. Atenolol inhibited the growth of both S. typhimurium and E. coli K-12 at 2 mg/mL.
To exclude the effect of atenolol on bacterial growth, the antivirulence activities were evaluated at sub-MIC (1/5 MIC). For further testing, the bacterial growth was evaluated in the presence of atenolol at sub-MIC and compared with the growth of untreated bacterial cultures. There were no significant differences between viable counts of S. typhimurium and E. coli K-12 in the presence or absence of atenolol at sub-MIC ( Figure 1).

Effect on Bacterial Adhesion and Biofilm Formation
The inhibitory effect of atenolol on S. typhimurium and E. coli K-12 adhesion and biofilm formation was assessed using the crystal violet method. The adhered or biofilm forming bacterial cells were stained with crystal violet after 2 h or 24 h, respectively. Captured light microscope images of formed biofilms on cover slips showed the marked effect of atenolol on diminishing the biofilm formation ( Figure 2A). The absorbances of the extracted crystal violet staining the bacterial cells treated with atenolol were photometrically measured and compared to the absorbances of untreated cultures. Atenolol significantly decreased bacterial adhesion and biofilm formation; the data are presented as percentage changes from untreated control ( Figure 2B).

Effect on S. typhimurium Invasion and Internalization
The antivirulence activity of atenolol at sub-MIC on S. typhimurium invasion and intracellular replication was evaluated by employing a gentamicin protection assay. Significantly, atenolol attenuated the invasiveness of S. typhimurium in HeLa cells ( Figure 3A,B) and also diminished the numbers of intracellularly replicated bacterial cells inside the raw The experiment was performed in triplicate and results are expressed as the means ± standard deviation with p value < 0.05 considered statistically significant.

Effect on S. typhimurium Invasion and Internalization
The antivirulence activity of atenolol at sub-MIC on S. typhimurium invasion and intracellular replication was evaluated by employing a gentamicin protection assay. Significantly, atenolol attenuated the invasiveness of S. typhimurium in HeLa cells ( Figure 3A,B) and also diminished the numbers of intracellularly replicated bacterial cells inside the raw macrophage ( Figure 3C,D). Atenolol significantly decreased the intracellular replication of S. typhimurium (*** p < 0.0001). The experiment was performed in triplicate and results are expressed as the means ± standard deviation with p value < 0.05 considered statistically significant.

Effect on the Expression of Virulence Encoding Genes
The expressions of the encoding genes of different virulence factors in S. typhimurium were quantified with RT-PCR in the presence of atenolol at sub-MIC and compared to untreated controls. The expression of TTSS-II, sdiA QS receptor, and sensor kinases qseC

Effect on the Expression of Virulence Encoding Genes
The expressions of the encoding genes of different virulence factors in S. typhimurium were quantified with RT-PCR in the presence of atenolol at sub-MIC and compared to untreated controls. The expression of TTSS-II, sdiA QS receptor, and sensor kinases qseC and qseE encoding genes were determined in the presence or absence of atenolol. Interestingly, atenolol showed a significant ability to downregulate the tested genes when the genes' expressions were normalized to gyrB gene ( Figure 4). For further confirmation the tested genes' expressions were normalized to another housekeeping gene 16S rRNA. Atenolol significantly downregulated the expression of tested genes ( Figure S1).
Microorganisms 2022, 10, x FOR PEER REVIEW 9 of 18 genes' expressions were normalized to gyrB gene (Figure 4). For further confirmation the tested genes' expressions were normalized to another housekeeping gene 16S rRNA. Atenolol significantly downregulated the expression of tested genes ( Figure S1). The expression of the QS encoding gene sdiA and the norepinephrine membranal sensor encoding genes, qseC and qseE, and TTSS encoding genes ssrB, sifA, sifB, sseF, sseJ. sseI, ssaE, ssaJ, sscA, steC, and pipB were normalized to gyrB housekeeping gene in the presence of atenolol in comparison to untreated controls. Atenolol significantly downregulated the expression of S. typhimurium virulence encoding genes (*** p < 0.0001). The experiment was performed in triplicate and results are expressed as the means ± standard deviation with p value < 0.05 considered statistically significant.

The Effect on the In Vivo Pathogenesis
In vivo mice protection assay was employed to test antivirulence activity of atenolol at sub-MIC. For more convincing, a mice group was injected with norepinephrine. All mice groups that were injected with PBS or kept un-injected survived. Meanwhile, only five out of ten mice survived in the mice group injected with untreated S. typhimurium. Interestingly, norepinephrine enhanced the S. typhimurium pathogenesis, six deaths out of ten (60%) were recorded. On the other hand, atenolol significantly protected mice against S. typhimurium pathogenesis, recording only 2 deaths out of ten (20%) (log rank test for trend p = 0.0034) ( Figure 5).  The expression of the QS encoding gene sdiA and the norepinephrine membranal sensor encoding genes, qseC and qseE, and TTSS encoding genes ssrB, sifA, sifB, sseF, sseJ. sseI, ssaE, ssaJ, sscA, steC, and pipB were normalized to gyrB housekeeping gene in the presence of atenolol in comparison to untreated controls. Atenolol significantly downregulated the expression of S. typhimurium virulence encoding genes (*** p < 0.0001). The experiment was performed in triplicate and results are expressed as the means ± standard deviation with p value < 0.05 considered statistically significant.

The Effect on the In Vivo Pathogenesis
In vivo mice protection assay was employed to test antivirulence activity of atenolol at sub-MIC. For more convincing, a mice group was injected with norepinephrine. All mice groups that were injected with PBS or kept un-injected survived. Meanwhile, only five out of ten mice survived in the mice group injected with untreated S. typhimurium. Interestingly, norepinephrine enhanced the S. typhimurium pathogenesis, six deaths out of ten (60%) were recorded. On the other hand, atenolol significantly protected mice against S. typhimurium pathogenesis, recording only 2 deaths out of ten (20%) (log rank test for trend p = 0.0034) ( Figure 5). On the other hand, atenolol protected mice and decreased dea rate to 20%, which indicates the significant atenolol ability to reduce the S. typhimurium capacity kill mice (log rank test p = 0.0034). Mice survival in each group was recorded every day over 5 da plotted using the Kaplan-Meier method and significance (p < 0.05) was calculated using the L rank test. ** p < 0.01.

Atenolol binding on SdiA and QseC
A two-step docking protocol, consisting of a preliminary rigid receptor approach a a further induced-fit docking, was applied to explore the interactions of atenolol with ba terial targets regulating virulence genes; the E. coli SdiA (PDB ID: 4LFU), and the E. c QseC receptor (PDB ID: 3JZ3).
The protein structure of E. coli SdiA which has been solved at 2.26 Å is composed a ligand-binding domain (LBD) represented by residues 5−167 at the N-terminus and DNA-binding domain (DBD) represented by residues 184-240 located at the C-terminu Both of them are connected together by a 16 amino acid linker. The topology of the ligan binding domain is basically a sandwich of a central antiparallel five-stranded β-sheet between two α-helices. The ligand-binding site with the co-crystallized tetraethylene g col molecule can be found as a concave surface on the β-sheet. The binding pockets of t biotargets were defined by matching the position of the co-crystallized ligand with t sites obtained from MOE Site Finder.
The E. coli QseC contains a total of 310 amino acids divided into two chains, w chain B more structured than chain A in the 2.5 Å crystal structure. The final structure which carries a short N-terminal β strand (β1) and an α helix (α1), and the catalytic d main. The E. coli QseC had no co-crystallized ligand, and the active site selection d pended upon MOE Site Finder and CASTp predictions. The 3D protein structure and t putative binding pockets with the calculated Richards' solvent accessible surface area (S and volume (V) of the biotargets as estimated using the Computed Atlas for Surface T pography of Proteins (CASTp) (Figure 6). However, all the mice survived in the negative control groups, and only 50% of mice survived in the positive control mice group injected with untreated S. typhimurium. Furthermore, norepinephrine increased the S. typhimurium pathogenesis causing a death rate of 60%. On the other hand, atenolol protected mice and decreased death rate to 20%, which indicates the significant atenolol ability to reduce the S. typhimurium capacity to kill mice (log rank test p = 0.0034). Mice survival in each group was recorded every day over 5 days, plotted using the Kaplan-Meier method and significance (p < 0.05) was calculated using the Log rank test. ** p < 0.01.

Atenolol Binding on SdiA and QseC
A two-step docking protocol, consisting of a preliminary rigid receptor approach and a further induced-fit docking, was applied to explore the interactions of atenolol with bacterial targets regulating virulence genes; the E. coli SdiA (PDB ID: 4LFU), and the E. coli QseC receptor (PDB ID: 3JZ3).
The protein structure of E. coli SdiA which has been solved at 2.26 Å is composed of a ligand-binding domain (LBD) represented by residues 5-167 at the N-terminus and a DNA-binding domain (DBD) represented by residues 184-240 located at the C-terminus. Both of them are connected together by a 16 amino acid linker. The topology of the ligandbinding domain is basically a sandwich of a central antiparallel five-stranded β-sheet in between two α-helices. The ligand-binding site with the co-crystallized tetraethylene glycol molecule can be found as a concave surface on the β-sheet. The binding pockets of the biotargets were defined by matching the position of the co-crystallized ligand with the sites obtained from MOE Site Finder.
The E. coli QseC contains a total of 310 amino acids divided into two chains, with chain B more structured than chain A in the 2.5 Å crystal structure. The final structure of which carries a short N-terminal β strand (β1) and an α helix (α1), and the catalytic domain. The E. coli QseC had no co-crystallized ligand, and the active site selection depended upon MOE Site Finder and CASTp predictions. The 3D protein structure and the putative binding pockets with the calculated Richards' solvent accessible surface area (SA) and volume (V) of the biotargets as estimated using the Computed Atlas for Surface Topography of Proteins (CASTp) (Figure 6).

Docking Simulations on E. coli SdiA QS Receptor
Docking of atenolol in E. coli SdiA resulted in generally comparable binding position, score, and interactions with the co-crystallized ligand (TEG) which was re-docked for validation and comparison as summarized in Table 2. The simply root-mean-square deviation (RMSD) for docking results of the co-crystallized was below 2.00 Å; indicating a valid and reliable docking procedure. Atenolol showed -6.4145 kcal/mol score which is slightly lower than that of the co-crystallized ligand TEG; -6.1143 Kcal/mol, indicating a good binding affinity of the ligand. RMSD, simply root-mean-square deviation.
Both atenolol and TEG exhibited numerous hydrophobic interactions with Tyr63, Tyr71, Trp95, Phe100, Leu106, Trp107, Ala110, Arg111, Leu115, Arg116 and Arg117 amino acid residues, representing the hydrophobic lining of the binding pocket. The acceptor oxygen of amide group in atenolol formed a hydrogen bond with the basic Arg111 and the hydroxyl group interacted with Asp80 through H-bond formation. Moreover, a protonated amino group formed an ionic bond with Asp80 (Figure 7).

Docking Simulations on E. coli SdiA QS Receptor
Docking of atenolol in E. coli SdiA resulted in generally comparable binding position, score, and interactions with the co-crystallized ligand (TEG) which was re-docked for validation and comparison as summarized in Table 2. The simply root-mean-square deviation (RMSD) for docking results of the co-crystallized was below 2.00 Å; indicating a valid and reliable docking procedure. Atenolol showed −6.4145 kcal/mol score which is slightly lower than that of the co-crystallized ligand TEG; −6.1143 Kcal/mol, indicating a good binding affinity of the ligand. RMSD, simply root-mean-square deviation.
Both atenolol and TEG exhibited numerous hydrophobic interactions with Tyr63, Tyr71, Trp95, Phe100, Leu106, Trp107, Ala110, Arg111, Leu115, Arg116 and Arg117 amino acid residues, representing the hydrophobic lining of the binding pocket. The acceptor oxygen of amide group in atenolol formed a hydrogen bond with the basic Arg111 and the hydroxyl group interacted with Asp80 through H-bond formation. Moreover, a protonated amino group formed an ionic bond with Asp80 (Figure 7).

Docking Simulations on E. coli QseC Receptor
QseC autophosphorylation occurs in response to epinephrine/norepinephrine (EPI/NEI) detection as a part of a complex pathway [9]. Norepinephrine (NE) was used in the docking process of atenolol into QseC for comparison as a reference compound as there was no co-crystallized ligand. Docking score for atenolol was -5.7045 Kcal/mol, while for norepinephrine was -4.4487. This suggests that atenolol is better than norepinephrine in binding affinity towards the active site pocket. Details of the docking results are illustrated in Table 3. Several key amino acid residues played a pivotal role in anchoring atenolol within the E. coli QseC pocket. Amidic and hydroxyl groups showed hydrogen bond interactions with Gln278 and Asp323, respectively, as well as pi-H interaction of phenyl ring with

Docking Simulations on E. coli QseC Receptor
QseC autophosphorylation occurs in response to epinephrine/norepinephrine (EPI/NEI) detection as a part of a complex pathway [9]. Norepinephrine (NE) was used in the docking process of atenolol into QseC for comparison as a reference compound as there was no cocrystallized ligand. Docking score for atenolol was −5.7045 Kcal/mol, while for norepinephrine was −4.4487. This suggests that atenolol is better than norepinephrine in binding affinity towards the active site pocket. Details of the docking results are illustrated in Table 3. Several key amino acid residues played a pivotal role in anchoring atenolol within the E. coli QseC pocket. Amidic and hydroxyl groups showed hydrogen bond interactions with Gln278 and Asp323, respectively, as well as pi-H interaction of phenyl ring with His280.

Discussion
Antibiotic misuse is considered one of the main causes of the development of bacterial resistance. The decreased numbers of newly discovered efficient antibiotics cannot resolve this issue and bacterial resistance development is augmented to be one of the major public health problems [52,53]. In this context, several approaches have been proposed; targeting bacterial resistance is a promising one [16,31,54,55]. This approach confers antivirulence agents that can be used in addition to antibiotics in particular in aggressive resistant bacterial infections [2,56]. The main concept of this approach is weakening bacterial virulence to be easily killed by host immune cells, which occurs without affecting growth and hence does not stress them to develop resistance [10,29]. The potent anti-virulence activities of several chemical compounds, drugs, and natural products proved the legibility of this approach [10,14,15]. To exclude any effect of atenolol on bacterial growth, all the performed experiments were performed using sub-MIC concentration. Moreover, the effect of atenolol at sub-MIC was evaluated on bacterial growth. It was shown that atenolol at sub-MIC did not have any significant effect on bacterial survival.
In addition to its clinical importance, Salmonella recruits several interplayed systems that control its pathogenesis [18,24]. These systems control the overall virulence behavior of Salmonella and their targeting could guarantee its virulence mitigation. The most important system is quorum sensing (QS) which is the key regulator of the production of Salmonella virulence factors, motility, and biofilm formation [22,57,58]. Indeed, Salmonella utilizes Lux-analogs SdiA to sense a variety of AHLs autoinducers [19,22,59]. The QS roles are very important for Salmonella adhesion and biofilm formation, as the sdiA mutants were not able to adhere to abiotic surfaces and did not form strong biofilms similar to those formed by wild-type Salmonella [19]. Meanwhile, atenolol virtually showed a marked ability to compete on SdiA, as it decreased the expression of its encoding gene. Furthermore, atenolol significantly decreased the E. coli and S. typhimurium adhesion and biofilm formation.
In addition to the main roles of QS in controlling the production of virulence factors, QS activates the expression of different secretion systems including the T3SS [60,61]. Salmonella employs two T3SS injectosomes to regulate its invasion into the host cells and intracellular replication inside immune cells [18,26]. The arrangement of several virulence genes in specific loci on the bacterial chromosome is called pathogenicity islands. To our

Discussion
Antibiotic misuse is considered one of the main causes of the development of bacterial resistance. The decreased numbers of newly discovered efficient antibiotics cannot resolve this issue and bacterial resistance development is augmented to be one of the major public health problems [52,53]. In this context, several approaches have been proposed; targeting bacterial resistance is a promising one [16,31,54,55]. This approach confers anti-virulence agents that can be used in addition to antibiotics in particular in aggressive resistant bacterial infections [2,56]. The main concept of this approach is weakening bacterial virulence to be easily killed by host immune cells, which occurs without affecting growth and hence does not stress them to develop resistance [10,29]. The potent anti-virulence activities of several chemical compounds, drugs, and natural products proved the legibility of this approach [10,14,15]. To exclude any effect of atenolol on bacterial growth, all the performed experiments were performed using sub-MIC concentration. Moreover, the effect of atenolol at sub-MIC was evaluated on bacterial growth. It was shown that atenolol at sub-MIC did not have any significant effect on bacterial survival.
In addition to its clinical importance, Salmonella recruits several interplayed systems that control its pathogenesis [18,24]. These systems control the overall virulence behavior of Salmonella and their targeting could guarantee its virulence mitigation. The most important system is quorum sensing (QS) which is the key regulator of the production of Salmonella virulence factors, motility, and biofilm formation [22,57,58]. Indeed, Salmonella utilizes Lux-analogs SdiA to sense a variety of AHLs autoinducers [19,22,59]. The QS roles are very important for Salmonella adhesion and biofilm formation, as the sdiA mutants were not able to adhere to abiotic surfaces and did not form strong biofilms similar to those formed by wild-type Salmonella [19]. Meanwhile, atenolol virtually showed a marked ability to compete on SdiA, as it decreased the expression of its encoding gene. Furthermore, atenolol significantly decreased the E. coli and S. typhimurium adhesion and biofilm formation.
In addition to the main roles of QS in controlling the production of virulence factors, QS activates the expression of different secretion systems including the T3SS [60,61]. Salmonella employs two T3SS injectosomes to regulate its invasion into the host cells and intracellular replication inside immune cells [18,26]. The arrangement of several virulence genes in specific loci on the bacterial chromosome is called pathogenicity islands. To our knowledge, the virulence genes are arranged on 22 Salmonella pathogenicity islands (SPIs) [62]. Particularly, SPI1 and SPI2 encode the two types of T3SS; the first type T3SS-SPI1translocates its effectors in the early stages to modify the cytoskeleton modulating the Salmonella invasion into the host cells. Later, when Salmonella is engulfed inside the immune cell's phagosomes, T3SS-SPI2 works in the surrounding drastic conditions translocating its effectors to the cytoplasm to ensure the Salmonella survival and even replication in the phagosomes [25,46]. Interestingly, atenolol downregulated the expression of the genes encoding the T3SS-SPI2 structure, function and regulation (Table 2). It was shown that the Salmonella sdiA mutants were found to be deficient in invasion and intracellular replication in comparison to wild-type [19]. Bearing in mind the significant anti-QS atenolol activity that in turn diminishes the expression of T3SS in addition to decreasing the expression of its encoding genes, it can be said that atenolol could interfere with the T3SS. For attesting to these findings, the effect of atenolol at sub-MIC on the Salmonella invasion into HeLa cells and intracellular replication inside macrophages was investigated. In compliance with interference with T3SS, atenolol significantly diminished the Salmonella invasion and decreased its replication in macrophages.
While bacteria can communicate with each other using chemical signals as known for QS, they can inter-species talk. For instance, Salmonella SdiA QS receptor can sense autoinducers of another species [5,19]. More interestingly, bacteria can hear our stress as bacteria can sense the adrenergic hormones using special sensors on their membranes [7,9]. The adrenergic hormones crosstalk with the autoinducers of QS activating the same signaling pathway, and enhancing the virulence [6,8,9]. Parallelly, autoinducers stimulate the norepinephrine surge in host cells resulting in augmenting the virulence [7,8]. The bacterial eavesdrop is observed mainly among the gut pathogen; E. coli K-12 as well as Salmonella spy on the host cells using membranal sensors QseC and QseE [3,6,8,9]. The β-adrenoceptor blockers could counteract the norepinephrine compromising effect on the host defense [5]. Atenolol showed considered ability to compete with norepinephrine on QseC, and significantly decreased the expression of norepinephrine sensors encoding genes qseC and qseE. These results indicate the possible atenolol's diminishing effects on Salmonella espionage on host cells resulting in mitigation of bacterial virulence.
To sum up the anti-virulence effects of atenolol, a mice protection assay was performed against S. typhimurium. Atenolol at sub-MIC significantly mitigated the Salmonella virulence and decreased its capacity to kill mice. The current data document the potent anti-virulence activities of atenolol via targeting QS, interfering with T3SS and diminishing bacterial espionage. However, further detailed pharmacological studies have to be conducted to avoid any predicted adverse effects.

Conclusions
S. enterica is a clinically important pathogen and develops resistance to antibiotics. The drug repurposing approach for attenuating bacterial virulence is supposed to be an efficient approach to overcome resistance development. In the current study, atenolol was evaluated in silico, in vitro and in vivo to be repositioned as an anti-virulence adjuvant to antibiotics. Atenolol significantly diminished Salmonella adhesion, biofilm formation, invasion and intracellular replication. Atenolol downregulated the T3SS encoding genes and protected mice against Salmonella. Furthermore, it showed the ability to bind to Qs receptor SdiA and norepinephrine sensors QseC and QseE, in addition to down regulation of the encoding genes. These findings suppose the atenolol as a potent anti-virulence candidate to be used besides antibiotics in the treatment of resistant infections. Future clinical studies are warranted to confirm findings from this study in the treatment of infections caused by Salmonella spp.
Supplementary Materials: The following supporting information can be downloaded at: https://www. mdpi.com/article/10.3390/microorganisms10101976/s1, Figure S1: Effect of atenolol at sub-MIC on the expression of S. typhimurium virulence encoding genes.