ADP-MoA: a platform for screening antibiotic activity and their mechanism of action inPseudomonas aeruginosa

ADP-MoA: a platform for screening antibiotic1 activity and their mechanism of action in2 Pseudomonas aeruginosa3 Estela Ynés Valencia Morante,1,2 Viviane Abreu Nunes,3 Felipe S Chambergo,4 Beny4 Spira1∗5 1Department of Microbiology, Institute of Biomedical Sciences, University of São Paulo,6 São Paulo, Brazil7 2XYZ Molecular Target LTDA, São Paulo, Brazil8 3Laboratory of Skin Physiology and Tissue Bioengineering, School of Arts, Sciences and9 Humanities, University of São Paulo, São Paulo, SP , Brazil10 4Laboratory of Proteins and Biotechnology, School of Arts, Sciences and Humanities,11 University of São Paulo, São Paulo, SP , Brazil12 *Address correspondence to Beny Spira, [email protected]. 13 ABSTRACT14 The emergence and proliferation of multidrug-resistant bacteria pose a major threat to15 global public health. To address an imminent crisis, it is essential to identify and16 characterize new antibacterial molecules. With that in mind, we developed the ADP-MoA17 platform, that facilitates the discovery of new antibiotics and provides preliminary18 insights into their mechanisms of action. The basic idea is to simultaneously visualize19 antibiotic activity – growth inhibition, along with one of the three classic antibiotics20 mechanisms of action: DNA damage/inhibition of DNA replication, protein synthesis21 inhibition and cell wall damage. The platform consists of three different chromosomal22 1 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint fusions between the promoters of recA, ampC or armZ and the luxCDABE operon. The23 platform was constructed and hitherto tested in the pathogenic opportunistic bacterium24 Pseudomonas aeruginosa. As a proof of concept we showed that the promoter fusions were25 each activated by the expected antibiotics with known mechanisms of action. The26 armZ::luxCDABE fusion responded to antibiotics that inhibit protein synthesis27 (macrolides, chloramphenicol, tetracyclines and aminoglycosides), ampC::luxCDABE was28 induced by β-lactams and recA::luxCDABE was induced by quinolones. Interestingly,29 ciprofloxacin induced PampC and ParmZ as well, albeit at a lower level. The ADP-MoA30 platform offers a readily implementable, low-cost approach with significant potential for31 high-throughput screening of antimicrobials against P . aeruginosaand other bacterial32 species. 33 INTRODUCTION34 Multi-resistant pathogenic bacteria are one of the main threats to public health35 worldwide. An estimated 1.27 million deaths occur annually due to pathogenic bacteria36 resistant to available antibiotics [1] and that in 2050 they will cause 10 million deaths per37 year, making them more lethal than cancer. Although antibiotic resistance concerns all38 pathogenic bacteria, the WHO Bacterial Priority Pathogens List [2] named 15 families of39 antibiotic-resistant pathogens, which have a strong negative impact on human and animal40 health. Among them is the opportunistic bacterium Pseudomonas aeruginosa. In particular,41 P . aeruginosastrain PA14 (UCBPP-PA14), isolated from a patient burn wound [3], presents42 an arsenal of molecular mechanisms that confer intrinsic and acquired resistance to43 multiple classes of antibiotics and many virulence genes, mostly related to the secretion of44 toxins and biofilm formation [4, 5]. 45 The research and development of novel antibiotics by pharmaceutical companies has46 drastically decreased in recent years [6]. Preclinical antibiotic research (extraction,47 identification, functional analysis, molecular target validation, physicochemical48 characteristics, pharmacokinetic and pharmacodynamic studies) take approximately five49 years [7]. The simultaneous assessment of antibacterial activity and characterization of the50 2 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint mechanisms underlying this biological activity could shorten the pace of new antibiotics’51 discovery [8]. 52 In the present study, we report on the design and development of a bacterial genetic53 platform that provides a functional assay aimed at screening antimicrobial molecules54 (through growth inhibition) while simultaneously determining their mechanism of action55 (DNA damage/replication inhibition, ribosome inhibition, or cell wall damage). The set56 of biosensors that form the platform was named ADP-MoA, which stands for “Antibiotic57 discovery platform by mechanism of action”. At this stage, the platform was58 implemented in strain PA14 of P . aeruginosa, into which biosensors carrying specific59 promoters (recA, ampC or armZ) were fused to the luxCDABE operon (lux for the sake of60 brevity). These promoters were chosen based on their response to the presence of specific61 antibiotics with known mechanisms of action. The recA gene encodes a protein crucial for62 DNA repair and maintenance as part of the bacterial SOS response, with its transcription63 being induced by DNA damage or inhibition of DNA replication [9, 10]; ampC encodes a64 β-lactamase that is activated, via AmpR, in response to cell wall damage [11, 12] and armZ65 responds to ribosome-targeting antimicrobials [13, 14]. As a proof of concept, the three66 biosensors were tested with dozens of known antibiotics, validating their effectiveness67 and specificity. 68 MATERIALS AND METHODS69 Bacterial strains and growth conditions70 P . aeruginosaPA14 strain was used to host a chromosomal copy of each biosensor.71 Escherichia coli strains DH5α or DH10B were used for DNA manipulation. Bacteria were72 grown in lysogeny broth (LB) [15] at 37°C. When necessary, the medium was73 supplemented with 30 µg/mL gentamicin or 100 µg/mL ampicillin. Muller Hinton (MH)74 or MH-Agar media were used in the antimicrobial tests. 75 3 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint Construction of the biosensors76 The genomic DNA of strain PA14 was extracted using the Wizard™ Genomic DNA77 Purification Kit (Promega). DNA fragments corresponding to specific promoter regions78 were amplified by PCR with the help of the TransStart™ FastPfu DNA Polymerase kit79 (TransGen Biotech) and specific primers. The following amplicons were obtained80 (numbers correspond to the first and last base on the P . aeruginosaPA14 genome–81 GenBank number: CP034244.1): recA (1504655 to 1505155); ampC (933654 to 933962) and82 armZ (6432940 to 6432561). The amplicons were cloned in the BamHI/PstI sites of plasmid83 pUC18T-mini-Tn7T-lux-Gm [16] located upstream to the promoterless lux operon.84 Following ligation, the plasmids were electrotransformed in E. coli strains DH10B or85 DH5α, and then electrotransformed in strain PA14 together with the helper plasmid86 pTNS3 that carries the genes encoding the TnsABCD site-specific transposition proteins87 [17]. Integration of the promoter-lux fusions 25 nucleotides downstream of the glmS gene88 in PA14 chromosome was verified by PCR, with primers PTn7R 89 (5’-CACAGCATAACTGGACTGATTTC-3’) and PglmS-down 90 (5’-GCACATCGGCGACGTGCTCTC-3’), as recommended by [17]. 91 Antibacterial activity and luminescence detection on plates92 The Kirby-Bauer disc diffusion method [18] was used to test both antibacterial (growth93 inhibition) and biosensor activity in the presence of known antibiotics. Thirty eight94 different commercial disks (CECON, São Paulo-Brazil) were tested: amikacin (AMI, 3095 µg), ampicillin (AMP , 10 µg), ampicillin + sulbactam (SBA, 20 µg), azithromycin (AZI, 1596 µg), bacitracin (BAC, 10 µg), cefepime (CPM, 30 µg), cefoxitin (CFO, 30 µg), ceftriaxone97 (CRO, 30 µg), cefuroxime (CRX, 30 µg), chloramphenicol (CLO, 30 µg), Ciprofloxacin (CIP ,98 5 µg), clavulanic acid+ amoxicillin (AMC, 30 µg), clindamycin (CLI, 2 µg), doxycycline99 (DOX, 30 µg), enrofloxacin (ENO, 5µg), ertapenem (ETP , 10 µg), erythromycin (ERI, 15 µg),100 fosfomycin (FOS, 200 µg), gentamicin (GM, 10 µg), imipinem (IPM, 10 µg), kanamycin101 (CAN, 30 µg), levofloxacin (LVX, 5 µg), meropenem (MPM, 10 µg), nalidixic acid (NAL,102 30), neomycin (NEO, 30 µg), nitrofurantoin (NIT, 300 µg), norfloxacin, (NOR, 10 µg),103 4 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint ofloxacin (OFX, 5 µg), penicillin (PEN, 10 U), piperacillin + tazobactam (PPT, 110 µg),104 rifampicin (RIF, 5 µg), streptomycin (EST, 10 µg), sulfadiazine + trimethoprim (SDT, 25105 µg), sulfamethoxazole + trimethoprim (SUT, 25 µg), tetracycline (TET, 30 µg), ticarcillin106 (TIC, 75 µg), tobramycin (TOB, 10 µg) and trimethoprim (TRI, 5 µg). The disks were107 placed on MH agar plates, previously spread with 0.5 McFarland PA14 suspensions108 carrying each of the three biosensors. The plates were incubated for 24 h at 37°C, and then109 visualized using the ChemiDoc Imaging System (Bio-Rad, USA) to detect luciferase110 expression. 111 Determination of the minimum inhibitory concentration (MIC)112 We assessed the minimum inhibitory concentration of each antibiotic as per the113 Clinical and Laboratory Standards Institute 2020 protocol [18]. Bacteria grown in LB114 medium were diluted 100-fold in MH medium and further grown to an OD600 of ∼0.1.115 Then, 105 bacteria from each culture were added to each well of a 96-well plate containing116 two-fold serial dilutions in medium MH of each of the tested antibiotics, ranging from 16117 to 1024 µg/mL in the case of ampicillin; 0.03 to 4 µg/mL in the case of imipenem; 4 to 256118 µg/mL for chloramphenicol; 0.5 to 32 µg/mL for tetracycline; 0.007 to 2.5 µg/mL for119 ciprofloxacin .The plates were then incubated at 37ºC for 24 h. The MIC was obtained by120 determining the lowest antibiotic concentration that prevented bacterial growth in the121 plate wells [19, 20]. 122 Quantitative determination of luciferase123 Quantitative assays of luciferase activity were performed using flat bottom white124 96-well plates (Greiner, Cat. No. 655098). These plates are optimized for bioluminescence125 detection. Each well was filled with 105 bacteria (carrying the biosensor) suspended in126 MH medium containing two-fold serial dilutions of antibiotics, as described for the MIC127 determination. To quantify luciferase activity the plates were placed in the Synergy H1128 multimode microplate reader (BioTek, North Macedonia) at 37°C with stirring, in which129 both luminescence and OD600 were recorded every hour for 18-20 hours. As a positive130 5 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint control, wells filled with bacteria in the absence of antibiotics were added. Luminescence131 values were normalized against bacterial density in each well (Luminescence/OD600).132 RESULTS133 Susceptibility of P . aeruginosato selected antibiotics134 To determine the antibiotic susceptibility profile of PA14, a Kirby-Bauer disk diffusion135 assay with 38 antibiotic disks of different classes (see Materials and Methods) was136 performed. Figure 1 shows the antibiotics to which PA14 exhibits resistance: ampicillin,137 cefuroxime, cefoxitin and penicillin G that are associated with cell wall synthesis138 inhibition; clindamycin that inhibits protein synthesis; trimethoprim, an inhibitor of folic139 acid biosynthesis; rifampicin (RNA synthesis inhibition) and nitrofurantoin, whose exact140 mode of action is unknown [21, 22]. Sensitivity to gentamicin was also tested, as the141 biosensor backbone - pUC18T-mini-Tn7T-lux-Gm carries a gentamicin resistance gene.142 143 6 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint FIG 1 Profile of PA14 resistance to selected antibiotics. AMP (ampicillin); CLI (clindamycin); RIF (rifampicin); TRI (trimethoprim); NIT (nitrofurantoin); PEN (penicillin G); CFO (cefoxitin); CRX (cefuroxime); GEN (gentamicin). The ADP-MoA biosensor platform144 We developed the ADP-MoA platform as a tool to support the discovery of new145 antibiotics targeting P . aeruginosawhile simultaneously identifying their mechanism of146 action. The platform consists of a set of specific promoters each fused to a promoterless147 lux operon producing thus bioluminescence upon promoter induction. The chosen148 promoters were: PrecA, that responds to DNA damage or inhibition of replication [23];149 PampC, that as part of the ampR-ampC system, is activated in response to cell wall damage150 [11] and ParmZ, containing a leader peptide-encoding sequence (PA5471.1) upstream of151 armZ whose expression provides a sensor of ribosome function [13]. The amplicons152 containing the promoter regions were cloned in plasmid pUC18T-mini-Tn7T-lux-Gm [16]153 and transferred to P . aeruginosafor integration at the attTn7 site, located 25 nucleotides154 downstream of glmS [17]. The biosensor strains were evaluated for their responses155 (luminescence around the growth inhibition halos) to the presence of different antibiotic156 disks: quinolones (ciprofloxacin, levofloxacin, norfloxacin, ofloxacin, enrofloxacin and157 nalidixic acid), carbapenems (imipenem, ertapenem and meropenem), penicillins158 (ampicillin, ampicillin plus sulbactam, ticarcillin, piperacillin plus tazobactam),159 cephalosporins (cefepime and ceftriaxone), aminoglycosides (amikacin, streptomycin,160 neomycin, kanamycin, gentamicin and tobramycin) tetracyclines (doxycycline and161 tetracycline), as well as antibiotics that interact with the ribosome 50S subunit162 (erythromycin, azithromycin and chloramphenicol) and nitrofuran (nitrofurantoin). As163 expected, the PrecA::lux biosensor (Figure 2A) was induced by quinolones and by164 nitrofurantoin; the PampC::lux biosensor (Figure 2B) responded to a variety of β-lactam165 antibiotics and the ParmZ::lux biosensor (Figure 2C) responded to antibiotics that interact166 with the 30S (aminoglycosides, tetracyclines) or 50S ribosomal subunit (erythromycin,167 azithromycin and chloramphenicol). Overall, the three biosensors were found to be168 7 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint functional as they responded to all antibiotics of the corresponding class, and thereby169 could potentially be used for the screening of new antipseudomonal compounds.170 171 FIG 2 Effect of selected antibiotics on PA14 growth (top) and luciferase activity (bottom) in response to antibiotics that interfere with (A) DNA replication (biosensor PrecA::lux), (B) cell wall synthesis (biosensor PampC::lux), and (C) protein synthesis (biosensor ParmZ::lux). The plates were read in the ChemiDoc MP system detector (Bio-Rad, USA) to reveal the luminescence halos. The following antibiotic disks were used: CIP (ciprofloxacin), LVX (levofloxacin), NOR (norfloxacin), OFX (ofloxacin), ENO (enrofloxacin), NAL (nalidixic acid), NIT (nitrofurantoin), MER (meropenem), IPM (imipenem), ETP (ertapenem), AMP (ampicillin), SBA (ampicillin + sulbactam), PIT (piperacillin + tazobactam), CRO (ceftriaxone), CPM (cefepime), TIC (ticarcillin), AMI (amikacin), EST (streptomycin), DOX (doxycycline), TET (tetracycline), NEO (neomycin), CAN (kanamycin), GEM (gentamicin), TOB (tobramycin), ERI (erythromycin), AZI (azithromycin), CLO (chloramphenicol). 8 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint Biosensor specificity172 To determine the specificity of the promoter fusions, each biosensor was cross-tested173 with antibiotics that, according to the literature, should not induce its activity (Figure 3).174 As expected, ciprofloxacin induced bioluminescence only in bacteria carrying the175 PrecA::lux fusion, imipenem only induced the PampC::lux promoter and aminoglycosides176 that inhibit protein synthesis through interaction with the 30S ribosome subunit –177 amikacin, kanamycin and neomycin, induced only ParmZ::lux. However,178 chloramphenicol and azithromycin, which interact with the 50S ribosome subunit,179 induced both ParmZ::lux and PrecA::lux, albeit the latter not as strong as the former. It180 should be noted that all tested antibiotics inhibited the growth of PA14 irrespective of the181 constructs they carry – PrecA::lux, PampC::lux or ParmZ::lux, except, of course, for the182 antibiotics shown in Figure 1 and nitrofuran (see below). To further explore the specificity183 of the biosensors, longer exposition times (> 30 s) were applied to detect faint184 luminescence signals from weak promoter inductions (see Figure S1 in the supplement).185 Thirty-three antibiotics were tested: 12 of them affect the cell wall (fosfomycin, bacitracin,186 ticarcillin, cefepime, ceftriaxone, imipenem, ertapenem, meropenem, piperacillin +187 tazobactam, ampicillin + sulbactam, clavulanic acid + amoxicillin and ampicillin), 9188 inhibit DNA replication (ciprofloxacin, levofloxacin, norfloxacin, ofloxacin, enrofloxacin,189 nalidixic acid, nitrofurantoin, sulfamethoxazole + trimethoprim and sulfadiazine +190 trimethoprim), and 11 inhibit protein synthesis (clindamycin, erythromycin,191 chloramphenicol, azithromycin, doxycycline, tetracycline, kanamycin, streptomycin,192 amikacin, neomycin and tobramycin). Rifampicin was used here as a negative control193 (PA14 is resistant to rifampicin, see Figure 1). Figure S1A shows that with the exception of194 clindamycin, that did not induce luminescence around the inhibition halo, all antibiotics195 that interfere with protein synthesis induced ParmZ::lux activity, but not the other196 biosensors. PampC::lux, that responds to cell wall damage, was strongly induced by197 carbapenems (imipenem, meropenem and ertapenem), ampicillins (ampicillin +198 sulbactam, ampicillin, ticarcillin, clavulanic acid + amoxicillin, piperacillin + tazobactam)199 and cephalosporins (cefepime and ceftriaxone), but did not respond to fosfomycin or200 bacitracin and did not cross-induced other biosensors (Figure S1B). Finally, six antibiotics201 9 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint that affect DNA replication (ciprofloxacin, levofloxacin, norfloxacin, ofloxacin,202 enrofloxacin and nalidixic acid) induced only the PrecA::lux biosensor. Interestingly,203 nitrofuran also induced PrecA::lux but did not form a halo of growth inhibition around the204 disk. Of the antibiotics that affect DNA synthesis, only Sulfamethoxazole + trimethoprim205 (SUT) were unable to form a luminescent halo (Figure S1C). 206 In addition, we confirmed in the present study that the expanded regulatory region of207 armZ, that includes the leader peptide [13] responds to antimicrobials that interfere with208 protein synthesis, either by interacting with the 30S subunit (aminoglycosides and209 tetracycline) or with the 50S subunit (erythromycin, azithromycin and chloramphenicol)210 (Figure 2, Figure 3 and Figure S1A). Also, this biosensor was weakly induced by some211 quinolones (Figure S1C). 212 213 FIG 3 Specificity of the biosensors towards known antibiotics. Bacteria carrying PrecA::lux (1), PampC::lux (2) or ParmZ::lux (3) were exposed to disks of each of the following antibiotics: CIP (ciprofloxacin), IPM (imipenem), AMI 10 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint (amikacin), CAN (kanamycin), NEO (neomycin), CLO (chloramphenicol) or AZI (azithromycin). Images at the top show growth inhibition halos caused by the antibiotics. Images at the bottom show bioluminescence halos elicited by induction of the lux fusions. (A) Inhibition of DNA replication by the quinolone ciprofloxacin; (B) inhibition of cell wall synthesis by the β-lactam imipenem; (C) inhibition of protein synthesis by the aminoglycosides amikacin, kanamycin and neomycin; (D) inhibition of protein synthesis by chloramphenicol and azithromycin. Quantification of biosensor response214 To obtain a quantitative assessment of the biosensors response, bacteria carrying each215 of the three constructions (PrecA::lux, PampC::lux or ParmZ::lux) were grown in 96-well216 plates containing MH medium supplemented with subinhibitory concentrations of each217 of the following antibiotics: ampicillin (64 µg/mL), imipenem (1 µg/mL), tetracycline (4218 µg/mL), chloramphenicol (32 µg/mL) or ciprofloxacin (0.03 µg/mL). Bioluminescence219 was assessed at several point intervals throughout growth (Figure 4). The time-points220 varied according to the antibiotic used, some induced earlier responses than others. The221 PampC::lux biosensor was respectively induced by ampicillin and imipenem by more than222 4-fold and 20-fold (compared to last time-point before induction) (Figure 4A and 4B). The223 other biosensors (PrecA::lux and ParmZ::lux) were not significantly induced by the224 β-lactam antibiotics. Tetracycline (Figure 4C) and chloramphenicol (Figure 4D) specifically225 induced ParmZ::lux by 86-fold and 60-fold, respectively. PrecA::lux was 12-fold induced by226 ciprofloxacin (Figure 4E). Some antibiotics induced the activity of non-related biosensors,227 but this response was relatively minor. For instance, PrecA::lux activity increased 4-fold in228 the presence of chloramphenicol (Figure 4D) and 5-fold in the presence of tetracycline229 (Figure 4C) after 17 h of treatment, without further increase after that. PampC::lux was230 induced by 6-fold and 8-fold in the presence of tetracycline and chloramphenicol,231 respectively, at 17 h treatment onward (Figure4C and Figure 4D). The response to232 ciprofloxacin was less specific, while PrecA::lux activity was induced by 12-fold,233 PampC::lux and ParmZ::lux were 4-fold induced by this antibiotic (Figure 4E).234 11 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint 235 FIG 4 Quantitative analysis of the effect of antibiotics on the biosensors PrecA::lux, PampC::lux and ParmZ::lux. Bacteria were exposed to ampicillin (64 µg/mL) (A), imipenem (1 µg/mL) (B), tetracycline (4 µg/mL) (C), chloramphenicol (32 µg/mL) (D) or ciprofloxacin 0,03 µg/mL (E). Luminescence values were normalized by the optical density of the cultures (lux/OD600) and the results represent the fold-change compared to the control (no antibiotic). 12 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint ANOVA analysis followed by Tukey: ∗ indicate significant differences (p < 0.05); ∗∗ indicates p < 0.004 compared with other biosensors. Error bars represent the mean ± standard deviation of three independent biological replicates. DISCUSSION236 The ADP-MoA platform was designed to aid in the search for potential antibacterial237 compounds by detecting antibiotic-induced growth inhibition and, simultaneously,238 identifying the underlying mechanism of action. The platform was hitherto implemented239 and tested in the PA14 strain of P . aeruginosa. We showed that the platform works out in240 both liquid and solid media and is amenable to high-throughput screening by growing241 the bacteria in 96 or 384 microplates, providing a low-cost biosensor platform adequate242 for the screening of molecule libraries of natural or synthetic origin. Table S1 summarizes243 the effect of all antibiotics tested in this study on the ADP-MoA platform.244 The PampC::lux biosensor responded to a variety of β-lactams, including carbapenems,245 penicillins and cephalosporins. All β-lactam antibiotics used in this study activated the246 expression of ampC, including ampicillin, that did not inhibit bacterial growth, but caused247 a strong bioluminescent response. Carbapenems are resistant to most β-lactamases248 including ampC and extended-spectrum β-lactamases (ESBL) [24]. A key factor in the249 efficacy of carbapenems is their ability to bind different PBPs, in particular PBP2, PBP4,250 and PBP5/6 [25, 26]. Cephalosporins, on the other hand, preferentially target PBP1a and251 PBP3 and penicillin preferentially interacts with PBP3 and PBP4 [26]. We suspect that the252 strong effect of the carbapenems meropenem, erthapenem and imipinem on ampC::lux253 induction observed here is related to their ability to strongly bind multiple different PBPs.254 ParmZ was used with the aim of obtaining a broad-spectrum sensor that responds to255 molecules interfering with protein synthesis. The armZ gene (also known as PA5471)256 forms an operon with PA5470, which encodes a putative peptide chain release factor [27].257 Just upstream to armZ, a small ORF (PA5471.1) encodes a 13 amino acid leader peptide,258 which is freely expressed in the absence of antibacterial compounds. In the presence of259 antibiotics, ribosome stalls on the PA5471.1 mRNA resulting in alternate mRNA folding260 13 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint enhancing armZ expression [13]. In essence, the PA5471.1 peptide acts as a sensor for261 ribosomal function, inducing the expression of armZ in response to antimicrobials that262 disrupt ribosome activity. In the absence of specific antibiotics, the leader sequence263 inhibits the transcription of armZ through a mechanism of translation attenuation. The264 induction of armZ by antibiotics requires the presence of the entire 367 bp regulatory265 region located between the PA5472 and armZ genes [13]. We showed that ParmZ::lux was266 strongly induced by all antibiotics that inhibit protein synthesis irrespective of their267 specific mechanism. For instance, macrolides (azithromycin and erythromycin) and268 amphenicol (chloramphenicol) that binds to the 50S ribosomal subunit, as well as269 tetracyclines (tetracycline, doxycycline) and aminoglycosides (amikacin, neomycin,270 tobramycin, streptomycin and kanamycin) that bind to the 30S ribosomal subunit all271 induced ParmZ.272 Recent evidence suggests that in addition to their known mechanism of action273 macrolides, amphenicol and tetracycline, may respectively interact with the V domain of274 23S rRNA [28], inhibit the dissociation of the 70S ribosome [29] and bind to the initiation275 complex comprising the 70S ribosome linked to P-site tRNAfMet and mRNA [30]. These276 interactions may explain the strong induction of ParmZ by these antibiotics. On the other277 hand, aminoglycosides elicited a mild ParmZ::lux response, possibly because278 aminoglycosides cause mRNA misreading, but not ribosome arrest [31]. This result279 demonstrates that our sensor is broad-spectrum for ribosome arrest, irrespective of280 whether the inhibition occurs at the 30S or 50S subunit. Both chloramphenicol and281 tetracycline induced ParmZ::lux, as expected, but also activated PampC::lux and PrecA::lux282 though at substantially lower levels. Interestingly, it has been reported that283 chloramphenicol-derived compounds inhibit the early stage of peptidoglycan284 biosynthesis in S. aureus [32]. Also, the tetracycline chelocardin has been shown to exhibit285 two mechanisms of action: in sub-inhibitory concentrations it inhibits protein synthesis,286 while at high concentrations it causes both cell wall damage and protein synthesis287 inhibition [33]. These data might explain why PampC::lux is slightly induced in response288 to these antibiotics. 289 14 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint The PrecA::lux biosensor was induced by all tested quinolones, including290 nitrofurantoin, whose exact mode of action is not entirely understood [22]. It has been291 shown that nitrofurantoin inhibits DNA replication by inducing recA, which in turn292 activates the SOS response [34, 35]. Ciprofloxacin induces the SOS response by interfering293 with the enzymes DNA gyrase or topoisomerase. Interaction with ciprofloxacin generates294 double-strand breaks (DSBs), resulting in single-strand DNA (ssDNA), which also295 induces the SOS response [36, 37]. In addition, ciprofloxacin has also been shown to296 increase the levels of intracellular ROS [38]. Here, we confirmed that ciprofloxacin297 induces recA. However, we also observed that this antibiotic induced other biosensors that298 respond to cell wall damage and ribosome arrest. Accordingly, in addition to DNA299 replication inhibition and chromosome fragmentation [39], fluoroquinolones have been300 shown to cause cytoplasmic condensation by damaging the membrane, leading to301 cytoplasmic leakage [40]. Another study reported that norfloxacin and ciprofloxacin302 treatment resulted in some degree of membrane damage [41]. Metabolomics [42] and303 metabolomic-proteomic [43] studies have identified alterations in transcription,304 translation, and cell wall synthesis as part of the ciprofloxacin mechanism of action305 against Mycobacterium tuberculosis [42, 44] and E. coli [43]. Collectively, these studies306 suggest that quinolones may have a secondary mechanism of action related to bacteria307 wall disruption, which might explain the induction of PampC by these antibiotics. In any308 case, our findings confirm that quinolones operate mainly through recA induction.309 Finally, the PrecA::luxsensor was slightly induced by chloramphenicol and310 tetracycline. As mentioned above, some antibiotics in addition of interacting with their311 specific target, generate ROS that contribute to cell killing [45, 46, 47]. Free radicals also312 damage DNA [48] and consequently activate the SOS system via recA [49].313 The use of promoter fusions with fluorescence or luminescence-encoding genes to314 study the interplay between drugs and gene expression under different conditions is315 well-established. For instance, Bollenbach et al. [50] analyze a library of ∼110 different E.316 coli promoters fused to the green fluorescent protein (GFP), showing that ribosomal genes317 are not directly regulated by DNA stress, leading to an imbalance between cellular DNA318 and protein content. Similarly, Elad et al. [51] evaluated the toxicity and mechanism of319 15 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint action of 420 FDA-approved drugs using 15 different bacterial reporters fused to the lux320 operon, associated with oxidative stress, DNA damage, heat shock, and surplus metal321 efflux. Valencia et al. [52] used a DNA fusion between the recA promoter and lux in P .322 aeruginosa PAO1 to show that amikacin prevents ciprofloxacin induction of the SOS323 regulon. Also, Higuera-Llantén et al. [53] used the concept of biosensors to identify the324 mechanism of action of new antibiotics. 325 Our set of biosensors display several improvements: a stable single chromosomal copy326 of the fusion, which result in a more reliable response by reducing cell-to-cell variability327 and eliminates the requirement of antibiotics to maintain the plasmid [54]. Another328 important feature is the fact that in our system the promoters used in the biosensors are329 derived from the same species and strain as the host – P . aeruginosaPA14, which is330 pathogenic bacterium of clinical interest. In addition, our platform showed a great331 sensitivity as it allowed the detection of antibacterial compounds that did not cause332 growth inhibition, as was the case of ampicillin and nitrofurantoin. Another advantage of333 this platform is that the molecule must penetrate the cell to exert its effect, excluding thus334 molecules that potentially possess antibacterial activity but cannot go through the335 membrane barrier. Finally, the ADP-MoA platform can be adapted for the identification of336 antimicrobials against other pathogens from the ESKAPE group (Enterococcus faecium,337 Staphylococcus aureus, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas338 aeruginosa and Enterobacter), Mycobacterium tuberculosis, and others.339 Conclusion340 The ADP-MoA platform offers a readily implementable, low-cost approach with341 significant potential for high-throughput screening of new antimicrobials against P .342 aeruginosa and other bacterial species. The search for new antibiotics is being actively343 pursued by major projects such as the CARB-X program (https://carb-x.org/) and by the344 National Institute of Allergy and Infectious Diseases and the National Cancer Institute345 (USA) [55]. Hopefully, our platform will also be able to contribute to the achievement of346 this important goal. 347 16 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint ACKNOWLEDGMENTS348 This work was supported by the Conselho Nacional de DesenvolvimentoCientífico e349 Tecnológico, Brazil (CNPq; grant nº 104461/2019-5) and the São Paulo Research350 Foundation (FAPESP) PIPE-TC program, process nº 2023/04848-6 and CEPID B3 (process351 nº 2021/10577-0). Valencia EY was funded by a postdoctoral fellowship from CNPq352 process nº 104461/2019-5 and fellowship from FAPESP process nº 2024/02524-1. V . Nunes;353 B. Spira and F. Chambergo are CNPq research fellows. 354 DATA A V AILABILITY STATEMENT355 Raw data (spreadsheets) are available upon request to the authors.356 FUNDING357 This work was supported by the Conselho Nacional de DesenvolvimentoCientífico e358 Tecnológico, Brazil (CNPq; grant nº 104461/2019-5) and the São Paulo Research359 Foundation (FAPESP) PIPE-TC program, process nº 2023/04848-6 and CEPID B3 (process360 nº 2021/10577-0). Valencia EY was funded by a postdoctoral fellowship from CNPq361 process nº 104461/2019-5 and fellowship from FAPESP process nº 2024/02524-1. V . Nunes;362 B. Spira and F. Chambergo are CNPq research fellows. 363 CONFLICTS OF INTEREST364 Valencia EY, Chambergo FS, Nunes VA and Spira B, are inventors of University of São365 Paulo, patents on the use of platforms to identify antibiotics. Valencia EY is founder of366 startup XYZ Molecular Target Ltda. The authors disclose the following patent filing:367 Expression System for Identifying molecules with antimicrobial activity, process for368 construction and use of said system. Provisional Application No. BR102020026097 9, filed369 December 18, 2020, National Institute of Industrial Property (INPI), Brazil370 17 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint References371 [1] Okeke IN, Kraker MEA, Boeckel TP , Kumar CK, Schmitt H, Gales AC, Bertagnolio372 S, Sharland M, Laxminarayan R. 2024. The scope of the antimicrobial resistance373 challenge. Lancet 1;403(10442):2426-2438. doi:10.1016/S0140-6736(24)00876-6.374 [2] WHO. 2024. WHO Bacterial Priority Pathogens List, 2024: bacterial pathogens of375 public health importance to guide research, development and strategies to prevent376 and control antimicrobial resistance. World Health Organization, Geneva Licence: CC377 BY-NC-SA 3.0 IGO. 378 [3] Grace A, Sahu R, Owen DR, Dennis V A. 2022.Pseudomonas aeruginosa reference379 strains PAO1 and PA14: A genomic, phenotypic, and therapeutic review. Front380 Microbiol 13 (1023523). doi:10.3389/fmicb.2022.1023523. PMID: 36312971; PMCID:381 PMC9607943. 382 [4] Driscoll JA, Brody SL, Kollef MH. 2007. The epidemiology, pathogenesis and383 treatment of Pseudomonas aeruginosa infections. Drugs 67 (3):351–68. doi:384 10.2165/00003495-200767030-00003. PMID: 17335295. 385 [5] Moradali MF, Ghods S, Rehm BH. 2017. Pseudomonas aeruginosa Lifestyle: A386 Paradigm for Adaptation, Survival, and Persistence. Front Cell Infect Microbiol 7 (39).387 doi:10.3389/fcimb.2017.00039. PMID: 28261568;PMCID: PMC5310132.388 [6] Li WX, Tong X, Yang PP , Zheng Y, Liang JH, Li GH, Liu D, Guan DG, Dai SX. 2022.389 Screening of antibacterial compounds with novel structure from the FDA approved390 drugs using machine learning methods. Aging (Albany NY) 14 (3):1448.391 [7] Monserrat-Martinez A, Gambin Y, Sierecki E. 2019. Thinking Outside the Bug:392 Molecular Targets and Strategies to Overcome Antibiotic Resistance. Int J Mol Sci393 13;20(6):1255. doi:10.3390/ijms20061255. PMID: 30871132; PMCID: PMC6470534.394 [8] Rütten A, Kirchner T, Musiol-Kroll EM. Oct 2022. Overview on Strategies and395 Assays for Antibiotic Discovery. Pharmaceuticals 15 (10):1302. doi:10.3390/ph15101302.396 [9] Butala M, gur Bertokb DZ, Busbya SJW. 2009. The bacterial LexA transcriptional397 repressor. Cell. Mol 66 (1):82–93.398 18 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint [10] Kaushik V , Tiwari M, Tiwari V. 2022. Interaction of RecA mediated SOS response399 with bacterial persistence, biofilm formation, and host response. Int J Biol Macromol400 217:931–943. doi:10.1016/j.ijbiomac.2022.07.176. PMID: 35905765.401 [11] Lister PD, Wolter DJ, Hanson ND. 2009. Antibacterial-resistant Pseudomonas402 aeruginosa: clinical impact and complex regulation of chromosomally encoded403 resistance mechanisms. Clin Microbiol Rev Oct;22(4):582-610. doi: 404 10.1128/CMR.00040-09. PMID: 19822890; PMCID: PMC2772362. 405 [12] Dik DA, Kim C, Madukoma CS, Fisher JF, Shrout JD, Mobashery S. 2020.406 Fluorescence Assessment of the AmpR-Signaling Network of Pseudomonas aeruginosa407 to Exposure to β-Lactam Antibiotics. ACS Chem Biol 15;15(5):1184-1194. doi:408 10.1021/acschembio.9b00875. Epub 2020 Feb 10. PMID: 31990176; PMCID:409 PMC7980316.410 [13] Morita Y, Gilmour C, Metcalf D, Poole K. 2009. Translational control of the411 antibiotic inducibility of the PA5471 gene required for mexXY multidrug efflux gene412 expression in Pseudomonas aeruginosa. Journal of bacteriology 191 (15):4966–4975.413 [14] Morita Y, Tomida J, Kawamura Y. 2014. Responses ofPseudomonas aeruginosa to414 antimicrobials. Front Microbiol 4 (422). doi:10.3389/fmicb.2013.00422. PMID: 24409175;415 PMCID: PMC3884212. 416 [15] MacWilliams MP , Liao MK. 2006. Luria broth (LB) and Luria agar (LA) media and417 their uses protocol. ASM MicrobeLibrary. American Society for Microbiology v:1–4,.418 [16] Damron FH, McKenney ES, Barbier M, Liechti GW, Schweizer HP , Goldberg JB.419 2013. Construction of mobilizable mini-Tn7 vectors for bioluminescent detection of420 gram-negative bacteria and single-copy promoter lux reporter analyses. Appl Environ421 Microbiol 79 (13):4149–53. 422 [17] Choi KH, Schweizer HP. 2006. mini-Tn7 insertion in bacteria with single attTn7 sites:423 example Pseudomonas aeruginosa. Nat Protoc 1 (1):153–61. 424 [18] I CLS. 2020. Performance Standards for antimicrobial susceptibility tests 30th ed.425 CLSI supplement M100, Clinical and Laboratory Standards Institute, Wayne, PA.426 19 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint [19] Valencia EY, Moraes Gomes F, Ospino K, Spira B. 2024. RpoS role in antibiotic427 resistance, tolerance and persistence in E. coli natural isolates. BMC Microbiol428 5;24(1):72. doi:10.1186/s12866-024-03222-7. PMID: 38443813; PMCID: PMC11323705.429 [20] Silva RTP , Petri MV , Valencia EY, Camargo PHC, Torresi SIC, Spira B. 2020. Visible430 light plasmon excitation of silver nanoparticles against antibiotic-resistant431 Pseudomonas aeruginosa. Photodiagnosis Photodyn Ther 31 (101908). doi:432 10.1016/j.pdpdt.2020.101908. PMID: 32634655. 433 [21] Recht J, Chansamouth V , White NJ, Ashley EA. 2022. Nitrofurantoin and434 glucose-6-phosphate dehydrogenase deficiency: a safety review. JAC Antimicrob Resist435 3;4(3):dlac045. doi:10.1093/jacamr/dlac045. PMID: 35529053; PMCID: PMC9070801.436 [22] Munoz-Davila MJ. 2014. Role of Old Antibiotics in the Era of Antibiotic Resistance.437 Highlighted Nitrofurantoin for the Treatment of Lower Urinary Tract Infections.438 Antibiotics (Basel 10;3(1):39-48. doi:10.3390/antibiotics3010039. PMID: 27025732;439 PMCID: PMC4790343. 440 [23] Cirz RT, O’Neill BM, a HJ, Head R SR, E F. 2006. Defining thePseudomonas441 aeruginosa SOS response and its role in the global response to the antibiotic442 ciprofloxacin. J. Bacteriol 188:7101–10.443 [24] Zhanel GG, Wiebe R, Dilay L, Thomson K, Rubinstein E, Hoban DJ, Noreddin444 AM, Karlowsky JA. 2007. Comparative review of the carbapenems. Drugs 67445 (7):1027–52. doi:10.2165/00003495-200767070-00006. PMID: 17488146.446 [25] Papp-Wallace KM, Endimiani A, Taracila MA, Bonomo RA. 2011. Carbapenems:447 past, present, and future. Antimicrob Agents Chemother 55 ((11)):4943–60. doi:448 10.1128/AAC.00296-11. PMID: 21859938; PMCID: PMC3195018. 449 [26] Montaner M, Lopez-Argüello S, Oliver A, Moya B. 2023. PBP target profiling by450 β-lactam and β-lactamase inhibitors in intact Pseudomonas aeruginosa: Effects of the451 intrinsic and acquired resistance determinants on the periplasmic drug availability.452 Microbiology spectrum 11 (1):e03038–22.453 [27] Winsor GL, Griffiths EJ, Lo R, Dhillon BK, Shay JA, Brinkman FSL. Nov 2015.454 Enhanced annotations and features for comparing thousands of Pseudomonas genomes455 20 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint in the Pseudomonas genome database. Nucleic Acids Research 44 (D1):D646–D653. doi:456 10.1093/nar/gkv1227. 457 [28] Xiong L, Korkhin Y, Mankin AS. 2005. Binding site of the bridged macrolides in the458 Escherichia coli ribosome. Antimicrob Agents Chemother Jan;49(1):281-8. doi:459 10.1128/AAC.49.1.281-288.2005. PMID: 15616307; PMCID: PMC538896.460 [29] Pathak BK, Mondal S, Barat C. 2017. Inhibition of Escherichia coli ribosome subunit461 dissociation by chloramphenicol and Blasticidin: a new mode of action of the462 antibiotics. Lett Appl Microbiol Jan;64(1):79-85. doi:10.1111/lam.12686. Epub 2016 Nov463 24. PMID: 27739094. 464 [30] Nguyen F, Starosta AL, Arenz S, Sohmen D, Dönhöfer A, Wilson DN. 2014.465 Tetracycline antibiotics and resistance mechanisms. Biol Chem May;395(5):559-75. doi:466 10.1515/hsz-2013-0292. PMID: 24497223. 467 [31] Krause KM, Serio AW, Kane TR, Connolly LE. 2016. Aminoglycosides: An468 Overview. Cold Spring Harb Perspect Med 1;6(6):a027029. doi: 469 10.1101/cshperspect.a027029. PMID: 27252397; PMCID: PMC4888811.470 [32] Zada SL, Green KD, Shrestha SK, Herzog IM, Garneau-Tsodikova S, Fridman M.471 2018. Derivatives of Ribosome-Inhibiting Antibiotic Chloramphenicol Inhibit the472 Biosynthesis of Bacterial Cell Wall. ACS Infect Dis 13;4(7):1121-1129. doi:473 10.1021/acsinfecdis.8b00078. Epub 2018 May 1. PMID: 29714997; PMCID:474 PMC6536003. 475 [33] Stepanek JJ, Lukežiˇ c T, Teichert I, Petkovi´ c H, Bandow JE. 2016. Dual mechanism of476 action of the atypical tetracycline chelocardin. Biochim Biophys Acta Jun;1864(6):645-654.477 doi:10.1016/j.bbapap.2016.03.004. Epub 2016 Mar 9. PMID: 26969785.478 [34] Revitt-Mills SA, Wright EK, Vereker M, O’Flaherty C, McPherson F, Dawson C,479 Oijen AM, Robinson A. 2022. Defects in DNA double-strand break repair resensitize480 antibiotic-resistant Escherichia coli to multiple bactericidal antibiotics. Microbiology open481 Oct;11(5):e1316. doi:10.1002/mbo3.1316. PMID: 36314749; PMCID: PMC9500592.482 [35] Ribeiro ACDS, WMBDS M, Silva AAD, Gales AC, Rando DGG, Minarini LADR.483 2020. Exposure to sub-inhibitory ciprofloxacin and nitrofurantoin concentrations484 21 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint increases recA gene expression in uropathogenic Escherichia coli: The role of RecA485 protein as a drug target. Eur J Pharm Sci 146 (105268). doi:10.1016/j.ejps.2020.105268.486 Epub 2020 Feb 17. PMID: 32081832.487 [36] Drlica K, Malik M, Kerns RJ, Zhao X. 2008. Quinolone-mediated bacterial death.488 Antimicrob Agents Chemother Feb;52(2):385-92. doi:10.1128/AAC.01617-06. PMID:489 17724149; PMCID: PMC2224783.490 [37] Little JW. 1991. Mechanism of specific LexA cleavage: autodigestion and the role of491 RecA coprotease. Biochimie Apr;73(4):411-21. doi:10.1016/0300-9084(91)90108-d.492 PMID: 1911941.493 [38] Rodríguez-Rosado AI, Valencia EY, Rodríguez-Rojas A, Costas C, Galhardo RS,494 Rodríguez-Beltrán J, Blázquez J. 2019. N-acetylcysteine blocks SOS induction and495 mutagenesis produced by fluoroquinolones in Escherichia coli. J AntimicrobChemother496 1;74(8):2188-2196. doi:10.1093/jac/dkz210. PMID: 31102529.497 [39] Bush NG, Diez-Santos I, Abbott LR, Maxwell A. 2020. Quinolones: Mechanism,498 Lethality and Their Contributions to Antibiotic Resistance. Molecules 1;25(23):5662.499 doi:10.3390/molecules25235662. PMID: 33271787; PMCID: PMC7730664.500 [40] Wong F, Stokes JM, Cervantes B, Penkov S, Friedrichs J, Renner LD, Collins JJ.501 2021. Cytoplasmic condensation induced by membrane damage is associated with502 antibiotic lethality. Nat Commun 19;12(1):2321. doi:10.1038/s41467-021-22485-6. PMID:503 33875652; PMCID: PMC8055701. 504 [41] AM KES, I AN, MK AEG, AS A, J P , AB S, M W, FA O. 2023. N4-Substituted505 Piperazinyl Norfloxacin Derivatives with Broad-Spectrum Activity and Multiple506 Mechanisms on Gyrase, Topoisomerase IV , and Bacterial Cell Wall Synthesis.ACS Bio507 Med Chem Au 30;3(6):494-506. doi:10.1021/acsbiomedchemau.3c00038. PMID:508 38144255; PMCID: PMC10739246. 509 [42] Knoll KE, Lindeque Z, Adeniji AA, Oosthuizen CB, Lall N, Loots DT. 2021.510 Elucidating the Antimycobacterial Mechanism of Action of Ciprofloxacin Using511 Metabolomics. Microorganisms 28;9(6):1158. doi:10.3390/microorganisms9061158.512 PMID: 34071153; PMCID: PMC8228629. 513 22 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint [43] Li W, Zhang S, Wang X, Yu J, Li Z, Lin W, Lin X. 2018. Systematically integrated514 metabonomic-proteomic studies of Escherichia coli under ciprofloxacin stress. J515 Proteomics doi:10.1016/j.jprot.2018.03.002. 15;179:61-70.516 [44] Shariati A, Arshadi M, Khosrojerdi MA, Abedinzadeh M, Ganjalishahi M, Maleki517 A, Heidary M, Khoshnood S. 2022. The resistance mechanisms of bacteria against518 ciprofloxacin and new approaches for enhancing the efficacy of this antibiotic. Front519 Public Health 10 (1025633). doi:10.3389/fpubh.2022.1025633. PMID: 36620240; PMCID:520 PMC9815622. 521 [45] Kohanski MA, Dwyer DJ, Wierzbowski J, Cottarel G, Collins JJ. 2008.522 Mistranslation of membrane proteins and two-component system activation trigger523 antibiotic-mediated cell death. Cell 14;135(4):679-90. doi:10.1016/j.cell.2008.09.038.524 PMID: 19013277; PMCID: PMC2684502.525 [46] Hong Y, Zeng J, Wang X, Drlica K, Zhao X. 2019. Post-stress bacterial cell death526 mediated by reactive oxygen species. Proc Natl Acad Sci U S A 14;116(20):10064-10071.527 doi:10.1073/pnas.1901730116. Epub 2019 Apr 4. PMID: 30948634; PMCID:528 PMC6525477. 529 [47] Dwyer DJ, Belenky PA, Yang JH, MacDonald IC, Martell JD, Takahashi N, Chan530 CT, Lobritz MA, Braff D, Schwarz EG, Ye JD, Pati M, Vercruysse M, Ralifo PS,531 Allison KR, Khalil AS, Ting AY, Walker GC, Collins JJ. 2014. Antibiotics induce532 redox-related physiological alterations as part of their lethality. Proc Natl Acad Sci U S533 A 20;111(20):E2100-9. doi:10.1073/pnas.1401876111. Epub 2014 May 6. PMID:534 24803433; PMCID: PMC4034191. 535 [48] Salles B, Germanier M, Defais M. 1987. A bacterial strain for detecting agents that536 produce free radical-mediated DNA strand breaks. Mutat Res May;183(3):213-7. doi:537 10.1016/0167-8817(87)90002-2. PMID: 2437447. 538 [49] Konola JT, Sargent KE, Gow JB. 2000. Efficient repair of hydrogen peroxide-induced539 DNA damage by Escherichia coli requires SOS induction of RecA and RuvA proteins.540 Mutat Res 28;459(3):187-94. doi:10.1016/s0921-8777(99)00073-7. PMID: 10812330.541 23 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint [50] Bollenbach T, Quan S, Chait R, Kishony R. 2009. Nonoptimal microbial response to542 antibiotics underlies suppressive drug interactions. Cell 13;139(4):707-18. doi:543 10.1016/j.cell.2009.10.025. 544 [51] Elad T, Seo HB, Belkin S, Gu MB. 2015. High-throughput prescreening of545 pharmaceuticals using a genome- wide bacterial bioreporter array. BiosensBioelectron546 68:699–704. doi:10.1016/j.bios.2015.01.067. 547 [52] Valencia EY, Esposito F, Spira B, Blázquez J, Galhardo RS. 2017.548 Ciprofloxacin-Mediated Mutagenesis Is Suppressed by Subinhibitory Concentrations549 of Amikacin in Pseudomonas aeruginosa. Antimicrob Agents Chemother 23;61(3):e02107-16.550 doi:10.1128/AAC.02107-16. PMID: 28031197; PMCID: PMC5328548.551 [53] Higuera-Llantén S, Alcalde-Rico M, Vasquez-Ponce F, Ibacache-Quiroga C,552 Blazquez J, Olivares-Pacheco J. 2024. A whole-cell hypersensitive biosensor for553 β-lactams based on the AmpR-AmpC regulatory circuit from the Antarctic554 Pseudomonas sp. IB20. MicrobBiotechnol Jan;17(1):e14385. doi:10.1111/1751-7915.14385.555 Epub 2024 Jan 10. PMID: 38197486; PMCID: PMC10832568. 556 [54] Saleski TE, Chung MT, Carruthers DN, Khasbaatar A, Kurabayashi K, Lin XN.557 2021. Optimized gene expression from bacterial chromosome by high-throughput558 integration and screening. Sci Adv 12;7(7):eabe1767. doi:10.1126/sciadv.abe1767.559 PMID: 33579713; PMCID: PMC7880599.560 [55] Martínez-Fructuoso L, Arends SJR, Freire VF, Evans JR, et al. 2023. Screen for New561 Antimicrobial Natural Products from the NCI Program for Natural Product Discovery562 Prefractionated Extract Library. ACS Infect Dis 9;9(6):1245-1256. doi:563 10.1021/acsinfecdis.3c00067. 564 AUTHOR BIOGRAPHIES565 Author biographies (≤150 words each) and photos (black and white, passport size)566 can be inserted here. They are required for reviews in CMR and MMBR but optional for567 24 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint minireviews in other journals and Gems in JVI. Do not include them for primary-research568 articles. See the specific instructions for individual journals for more information.569 Use \begin{authorbios}...\end{authorbios} environment to place your author570 biographies section, commands to do this: output will not come in the pdf for double571 blind mode. 572 25 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted November 11, 2024. ; https://doi.org/10.1101/2024.11.08.622684doi: bioRxiv preprint