New Evolutionary Insights into RpoA: A Novel Quorum Sensing Reprograming Factor in Pseudomonas aeruginosa

Abstract Quorum-sensing (QS) coordinates the expression of virulence factors in Pseudomonas aeruginosa, an opportunistic pathogen known for causing severe infections in immunocompromised patients. QS has a master regulator, the lasR gene, but in clinical settings, P. aeruginosa isolates have been found that are QS-active but LasR-null. In this study, we developed an experimental evolutionary approach to identify additional QS-reprogramming determinants. We began the study with a LasR-null mutant with an extra copy of mexT, a transcriptional regulator gene that is known to be able to reprogram QS in laboratory LasR-null strains. In this strain, spontaneous single mexT mutations are expected to have no or little phenotypic consequences. Using this novel method, which we have named “targeted gene duplication followed by mutant screening”, we identified QS-active revertants with mutations in genes other than mexT. One QS-active revertant had a point mutation in rpoA, a gene encoding the α-subunit of RNA polymerase. QS activation in this mutant was found to be associated with the downregulated expression of mexEF-oprN efflux pump genes. Our study therefore uncovers a new functional role for RpoA in regulating QS activity. Our results indicate that a RpoA-dependent regulatory circuit controlling the expression of the mexEF-oprN operon is critical for QS-reprogramming. In conclusion, our study reports on the identification of non-MexT proteins associated with QS-reprogramming in a laboratory strain, shedding light on possible QS activation mechanisms in clinical P. aeruginosa isolates.


Introduction
Pseudomonas aeruginosa is an opportunistic pathogen that causes severe acute and chronic human infections in cystic fibrosis (CF) patients with compromised immune systems (Gellatly and Hancock 2013;Klockgether and Tümmler 2017).A cohort of P. aeruginosa virulence factors is under the control of the quorum-sensing (QS) system (Lee and Zhang 2015;Papenfort and Bassler 2016).QS is a bacterial cell-cell communication system that regulates the expression levels of hundreds of genes in a cell densitydependent manner (Lee and Zhang 2015).Pseudomonas aeruginosa has two acyl-homoserine lactone (AHL) QS systems, the Las and Rhl QS systems.In the Las QS system, LasI catalyzes the diffusible QS signal N-3-oxododecanoyl homoserine lactone (3OC12-HSL), whereas the transcriptional regulator LasR binds to the 3OC12-HSL forming a protein-signal complex.The active 3OC12-HSL-LasR complex further activates Las-regulon genes.Similarly, Rhl QS contains an RhlI-RhlR pair, with RhlI synthesizing the butyryl-homoserine lactone (C4-HSL) and the C4-HSL-bound RhlR activating expression of the Rhl-regulon (Schuster and Greenberg 2006).These two AHL QS systems further interact with a so-called Pseudomonas quinolone signal (PQS) system, which is responsible for generating 2-heptyl-3-hydroxy-4-quinolone (PQS) and its precursor 2-heptyl-4-quinolone (HHQ).The pqs genes (operon pqsABCDE and pqsH) code for HHQ and PQS synthesis (Pesci et al. 1999;Diggle et al. 2006).In general, the Las QS is at the top of the QS hierarchy, and disruption of either LasI or LasR diminishes the expression of QS-activated genes (for more details on these QS systems, see Papenfort and Bassler 2016).
QS-active P. aeruginosa show QS-regulated responses such as production of extracellular proteases, C4-HSL, hydrogen cyanide, rhamnolipids, and pyocyanin (Bjarnsholt et al. 2010;Feltner et al. 2016;Cruz et al. 2020).Secretion of QS-regulated proteases allows the bacteria to be visualized on agar plates containing skim milk (Diggle et al. 2007;Sandoz et al. 2007).Thus, spontaneous QS-active revertants derived from QS-inactive mutants can be selected on these agar plates.In such evolution experiments, mutants reprogramming their QS have been obtained from a QS-inactive lasR mutant of the laboratory strain PAO1 (Oshri et al. 2018;Kostylev et al. 2019).Isolates with mutated lasR gene have been identified in CF patients with chronic P. aeruginosa infection (Smith et al. 2006;Hoffman et al. 2009;Feltner et al. 2016).However, although LasR is a master regulator of QS in P. aeruginosa, many LasR-null clinical isolates are QS-active (Bjarnsholt et al. 2010;Feltner et al. 2016;Cruz et al. 2020).Hence, QS-reprogramming of LasR-null mutants also appears to occur under natural conditions.
The P. aeruginosa genome encodes several resistancenodulation-division (RND) multidrug efflux pumps that transport small molecules into the extracellular environment.One of them, the MexEF-OprN efflux pump, is suggested to regulate the QS system by exporting the PQS precursor HHQ, which eventually leads to reduced intracellular PQS levels (Köhler et al. 2001).Furthermore, MexEF-OprN can export antibiotics such as chloramphenicol, fluoroquinolones, and trimethoprim, and thus contributes to antibiotic resistance (Lister et al. 2009).The mexT gene is localized upstream of the mexEF-oprN operon.The MexT protein, a LysR-type transcriptional regulator, positively regulates mexEF-oprN expression (Köhler et al. 1999).In addition to the mexEF-oprN operon, MexT regulates expression of more than 40 other genes (Tian, Mac Aogain, et al. 2009;Tian, Fargier, et al. 2009).Expression of mexT is negatively regulated by its neighboring gene mexS (Sobel et al. 2005;Richardot et al. 2016).Overexpression of the mexEF-oprN operon was found in various spontaneous P. aeruginosa mutants, and results in reduced QS-related responses, such as decreased synthesis of C4-HSL, hydrogen cyanide, and pyocyanin (Köhler, Michéa-Hamzehpour, Henze, et al. 1997;Köhler et al. 2001).
Mutations in the mexT gene were identified in clinical isolates from CF patients and non-CF patients with P. aeruginosa infections (Smith et al. 2006;Gilbert et al. 2012), as well as in P. aeruginosa laboratory strains (Luong et al. 2014;Cheng et al. 2022).In a QS-inactive LasR-null mutant of the laboratory strain PAO1, spontaneous mutations in mexT were found to be sufficient for QS-reprogramming (Oshri et al. 2018;Kostylev et al. 2019), indicating that reduced expression of the mexEF-oprN genes is associated with a functional QS (Köhler et al. 2001;Tian, Mac Aogain, et al. 2009).Likewise, disruption of the MexEF-OprN efflux pump in the LasR-null mutant background results in a QS-active phenotype with a functional Rhl QS as well as an active PQS system (Oshri et al. 2018;Kostylev et al. 2019).Deletion of genes of the PQS system (pqsA and/or pqsE) causes partially attenuated QS-dependent responses (Kostylev et al. 2019).Although mutations in mexT were frequently obtained in laboratory experiments, they are rare in clinical isolates (Smith et al. 2006).For example, a LasR-null isolate from a CF patient with chronic infection showed a QS-active phenotype although containing a functional MexT (Cruz et al. 2020).Therefore, QS-reprogramming in such clinical isolates is expected to be controlled by factors other than MexT inactivation (non-mexT mutations).
In this study, we developed a novel experimental evolution approach to identify non-mexT determinants involved in QS-reprogramming.A QS-inactive P. aeruginosa LasR-null mutant with an extra copy of mexT was generated and then used to screen for QS-active revertants.In this way, we identified a QS-active mutant with a spontaneous mutation in the transcriptional regulator gene rpoA.Further experiments showed that RpoA is involved in a regulatory circuit controlling the expression of the mexEF-oprN operon and thus is critical for bacterial antibiotic resistance and QS-reprogramming.Hence, our developed mutant screening method not only resulted in the discovery of a novel function of the RNA polymerase α-subunit protein RpoA but also highlighted its association with the regulation of MexEF-OprN efflux pump activity.

Results
Simultaneous Occurrence of lasR and mexT Mutations Is Rare in P. aeruginosa CF Isolates LasR mutants are common in P. aeruginosa isolates from CF patients with chronic infection.These CF LasR-null isolates often contain an active Rhl QS (Bjarnsholt et al. 2010;Feltner et al. 2016).In evolution experiments with QS-inactive LasR-null mutants of the laboratory strain PAO1, QS-active revertants were screened on skim milk agar plates and identified to have mexT mutations (Oshri et al. 2018;Kostylev et al. 2019).However, it is unclear whether such QS-reprogramming also applies to clinical isolates.To address these concerns, we re-analyzed previously published whole genome sequencing (WGS) data of P. aeruginosa CF isolates with respect to lasR and mexT mutations (Smith et al. 2006).Among CF isolates from 29 patients examined, lasR mutations, including frameshift mutations, single nucleotide substitutions, and insertions, were identified in isolates from 22 patients (75.9%).However, only isolates from two patients (9.1%) simultaneously had nonsynonymous mutations in lasR and mexT (Fisher's exact test, P = 0.23, odds ratio = 0.27).Moreover, mexT mutations were additionally found in a number of isolates that contain an intact lasR (supplementary table S1, Supplementary Material online).Hence, mexT mutations are frequently not colocalized with lasR mutations in the examined WGS data.Therefore, inactivation of MexT might often not account for QS activity in clinical LasR-null isolates possessing a QS-active phenotype.In accordance with these findings, Cruz et al. (2020) reported on a QS-active clinical isolate, which was identified as a LasR-null CF isolate with a functional mexT gene (Cruz et al. 2020).
Transposon Mutagenesis Results in a QS-Active LasR-Null Mutant With Mutated mexF Gene As our analyzed WGS data and previous studies (Cruz et al. 2020)

MBE
uncharacterized QS factors.We first performed a traditional transposon mutagenesis experiment using a mariner-based transposon (pBT20) (Kulasekara et al. 2005) and a QS-inactive LasR-null mutant of strain P. aeruginosa PAO1.The constructed insertion mutant library was spread and screened on skim milk agar.In this screening approach, QS-active revertant mutants were selected basing on the secretion of QS-controlled proteases.A total of 40,000 mutant colonies, corresponding to a transposon density of about 156 bp per insertion in the PAO1 genome, were examined.Using this screening method, we obtained a single bacterial colony with a protease-positive phenotype.Sequencing analysis indicated that this mutant possesses a transposon insertion in the mexF gene, which codes for an essential component of the MexEF-OprN efflux pump.Given that QS-reprogramming in LasR-null mutants mostly relies on mutations in mexT (Oshri et al. 2018;Kostylev et al. 2019), it was not surprising to see that a disruption of MexT-regulated MexF results in a QS-active phenotype.However, no other genes involved in QS-reprogramming were identified using this transposon mutagenesis method.Therefore, traditional transposon mutagenesis approach is not an efficient way to identify new QS-reprogramming factors in the LasR-null mutant.
A Novel Experimental Evolution Method for Identification of QS-Reprogramming Genes Considering that P. aeruginosa has an intriguing capacity for adapting to environments through frequent genome variations (Huang et al. 2019), we expected that hitherto unknown genetic elements involved in QS-reprogramming can be identified in evolution experiments.In a classic approach, a QS-inactive LasR-null mutant is evolved in a casein medium and spontaneous mutants emerge.The QS-active phenotype of the protease-positive revertants can then be confirmed by measuring production of various QS-related metabolites such as pyocyanin.Using this approach, sequencing of QS-active mutants resulted only in mutations in the mexT gene (Oshri et al. 2018;Kostylev et al. 2019).We therefore developed a new approach in which the probability of obtaining mexT mutations was reduced by introducing an extra copy of mexT into a neutral site of the LasR mutant genome.We assumed that a deleterious mutation in one copy of the two mexT genes would still leave the other copy functional and therefore would have no or little phenotypic consequences in the evolution experiment.Using this strategy, mutants with single mexT mutations are expected to be largely filtered out, thereby facilitating the identification of non-mexT mutations.The method, termed "targeted gene duplication followed by mutant screening" (TGD-MS), is schematically illustrated in figure 1.
The TGD-MS Method Results in the Identification of a QS-Active LasR-Null Mutant With a Mutation in rpoA Using the established TGD-MS strategy, we screened for QS-active mutants of the engineered LasR-null mutant carrying two mexT copies.Mutants were further characterized with respect to QS-controlled extracellular protease activity using a skim milk agar assay and pyocyanin production.Four colonies were found to partially restore protease activity and show significantly increased pyocyanin production, providing clues that they contained a rewired active QS (fig.2).Three of these QS-active mutants were New QS Function of a RpoA Mutant • https://doi.org/10.1093/molbev/msad203MBE then subjected to whole genome re-sequencing (WGS).
The sequencing results revealed that these colonies possess a common single nucleotide mutation (nucleotide T4754641->C; amino acid residue T262->A) in rpoA (supplementary tables S2 and S3, Supplementary Material online).The rpoA gene encodes the α-subunit of RNA polymerase (RNAP) in P. aeruginosa.Since rpoA is essential for cell viability in P. aeruginosa (Lee et al. 2015), a deletion mutant of rpoA has not been reported in the literature.
To elucidate the role of rpoA in QS regulation, the same point mutation (T262->A) was introduced into the LasR-null mutant, generating the mutant LasR-RpoA-T262A (hereafter designated as LasR-RpoA* and the RpoA-T262A mutant as RpoA*).As expected, LasR-RpoA* displayed a typical QS-active phenotype with regard to extracellular protease activity as visualized on skim milk agar plates (supplementary fig.S1, Supplementary Material online).Furthermore, compared with the QS-inactive LasR-null mutant, LasR-RpoA* showed increased Rhl-and PQS-responsive activities as determined by reporter plasmids as well as elevated production of C4-HSL and PQS (fig.3).Overall, the QS-dependent activities and metabolites of the LasR-RpoA* mutant were higher than those of the LasR-null mutant, but lower compared with a constructed QS-active LasR-MexT double mutant (mexT deletion mutant of LasR-null).Complementation of the LasR-RpoA* mutant with an episomal copy of wild-type rpoA reduced the extracellular protease activity and pyocyanin production to levels determined for the LasR-null mutant carrying the empty vector (supplementary fig.S2, Supplementary Material online).In conclusion, application of the TGD-MS method resulted in the identification of RpoA as a novel regulator of QS-controlled processes, capable of rewiring QS activity in LasR-null.

The LasR-RpoA* Mutant Exhibits a QS-Active Transcriptome Profile
To investigate genome-wide gene expression changes in the LasR-RpoA* mutant, we performed an RNA-sequencing (RNA-seq) analysis and compared the transcriptome profile of LasR-RpoA* with that of the parent strain LasR-null.We found global changes in gene expression in the LasR-RpoA* mutant, with a total of 182 genes differentially  Based on the transcriptome profile obtained in this study, we hypothesized that the attenuated MexEF-OprN efflux pump might account for the QS-reprogramming observed in the LasR-RpoA* mutant.To address this possibility, we conducted a comparative transcriptome analysis between the LasR-MexT mutant and the LasR-null mutant.The transcriptome analysis revealed that a total of 297 differentially expressed genes were induced in the LasR-MexT mutant (supplementary table S5, Supplementary Material online).Consistent with previous findings (Tian, Fargier, et al. 2009), many of these induced genes were associated with the MexEF-OprN efflux pump in the LasR-MexT mutant.Since the genes substantially upregulated by MexT inactivation are known to be MexEF-OprN-dependent (Tian, Fargier, et al. 2009), we selected genes with at least a 4-fold change in expression in the LasR-MexT mutant for further analysis of expression shifts in the LasR-RpoA* mutant.The comparative transcriptome analysis demonstrated that genes significantly upregulated in the LasR-MexT mutant were also activated in the LasR-RpoA* mutant (P = 0.01, one-sided Kolmogorov-Smirnov (KS) test).However, we did not observe a similar transcription pattern for New QS Function of a RpoA Mutant • https://doi.org/10.1093/molbev/msad203MBE downregulated gene sets between the LasR-MexT mutant and the LasR-RpoA* mutant (P = 0.07, KS test).Given that disruption of the MexEF-OprN efflux pump is known to be responsible for QS-reprogramming in the LasR-MexT mutant (Oshri et al. 2018;Kostylev et al. 2019), we reasoned that, similar to LasR-MexT, the reduced expression of mexEF-oprN operon genes might also be associated with QS-reprogramming in LasR-RpoA*.
The RpoA* Protein Modulates the Transcription of mexEF-oprN Operon Genes We used a green fluorescent protein (GFP) based transcriptional fusion system (PmexE-GFP) to estimate the promoter activity of mexEF in PAO1 derivatives (core promoter of the mexEF-oprN operon).In line with the results obtained from RNA-seq and qRT-PCR analysis, the LasR-RpoA* mutant carrying the reporter construct caused a sharp reduction of fluorescent signals when compared with the LasR-null mutant, indicating the reduced expression of mexEF-oprN operon genes (fig.5A).An even stronger reduction of fluorescent signals was observed for the LasR-MexT mutant carrying the reporter construct (fig.5A).These findings are consistent with the observation that QS-dependent activities are weaker in LasR-RpoA* than in LasR-MexT (fig.3).We next examined how RpoA* attenuates the expression of mexEF-oprN operon genes.RpoA protein comprises α-NTD and α-CTD two domains, forming a core unit of the RNAP (Browning and Busby 2016).Given that the RpoA* mutation localizes in the α-CTD, a domain responsible for the contact of RNAP to promoter DNA, we hypothesized that the RpoA* protein may directly interfere with the binding of RNAP to the mexEF promoter.To test this hypothesis, we performed an electrophoretic mobility shift assay (EMSA).His-tagged RpoA or RpoA* was expressed in the LasR-RpoA* mutant.The recombinant RNAP proteins containing the His-tagged α-subunit were then extracted and purified by affinity chromatography.
The two different types of RNAP were then assayed for their ability to bind to a 243-bp mexEF promoter DNA probe (fig.5B).Both RNAP proteins bound to the test probe in the EMSA (fig.5C).Compared with RNAP containing the wild-type RpoA, however, a weaker signal was observed for RNAP containing the RpoA* variant, which indicates a reduced RNAP-probe interaction caused by RpoA*.We therefore concluded that RNAP containing RpoA* downregulates the expression of mexEF-oprN operon genes by reducing the direct contact of RNAP to the mexEF promoter DNA.
Previous findings showed that inactivation of the MexEF-OprN efflux pump upregulates QS-dependent activities partially by reducing the export of the PQS precursor HHQ (Köhler et al. 2001;Lamarche and Déziel 2011), which may contribute to increased PQS production in LasR-RpoA* and LasR-MexT mutants (fig.3).As pqsA codes for a co-enzyme ligase essential for PQS biosynthesis (Déziel et al. 2004), we deleted the pqsA gene in the LasR-RpoA* mutant (generating LasR-RpoA*-PqsA) and evaluated the mutant's ability to secrete proteases.The pqsA deletion caused a strong but not complete reduction of the QS-controlled extracellular protease activity (supplementary fig.S4, Supplementary Material online).Consistent with the observation of pqsA deletion in LasR-MexT (Kostylev et al. 2019), our obtained results suggest that QS-reprogramming in the LasR-RpoA* mutant also partially depends on increased PQS production.
Furthermore, we complemented the LasR-RpoA* mutant by transferring an episomal copy of the mexEF-oprN operon fragment driven by an rrnB promoter.Expression of mexEF-oprN operon genes restored QS-controlled pyocyanin production (supplementary fig.S5, Supplementary Material online), confirming the reduction of MexEF-OprN activity responsible for QS-reprogramming in LasR-RpoA*.In addition, we analyzed the LasR-RpoA* mutant with respect to swarming motility, as inactivation of the MexEF-OprN efflux pump was reported to stimulate swarming motility (Lamarche and Déziel 2011).As expected, the LasR-RpoA* was found to exhibit considerably increased swarming motility when compared with the LasR-null parent strain (supplementary fig.S6, Supplementary Material online).
Through the regulation of the MexEF-OprN efflux pump, LasR-RpoA* thus gained a partially restored QS activity but a slightly reduced antibiotic resistance.We hypothesized that the arising LasR-RpoA* mutant from the neighboring LasR-null mutant would be fitter than the LasR-MexT mutant when the bacteria are exposed to an environment that requires both QS activation and antibiotic resistance.To test this hypothesis, we examined the ability of LasR-RpoA* or LasR-MexT to compete with LasR-null using the casein medium broth containing different concentrations of chloramphenicol.As expected, in the absence of chloramphenicol, LasR-MexT or LasR-RpoA* outcompeted LasR-null in the casein medium (fig.6B), consistent with the observation that QS-active LasR-null mutants are advantageous in the casein environment as previously reported (Oshri et al. 2018;Kostylev et al. 2019).In the presence of chloramphenicol, growth of LasR-RpoA* relative to LasR-null was reduced with increasing concentrations of chloramphenicol, indicating that chloramphenicol represses the growth advantage of LasR-RpoA*.Similarly, increasing chloramphenicol concentrations caused strongly reduced growth fitness of LasR-MexT relative to LasR-null.Notably, growth fitness reduction of LasR-MexT was even more pronounced than that of LasR-RpoA* (fig.6D).In conclusion, our findings indicate that the LasR-RpoA* is relatively fitter than the LasR-MexT mutant once it emerges from the LasR-null parent strain in the presence of chloramphenicol.
Trade-Off Between QS-Reprogramming and Bacterial Pathogenesis in LasR-RpoA* To investigate the impacts of LasR-RpoA* on bacterial pathogenesis, we first evaluated the QS-controlled virulence products in the LasR-RpoA* mutant.As shown in supplementary figure S7, Supplementary Material online, the LasR-RpoA* mutant exhibited higher levels of pyocyanin and hydrogen cyanide production compared with the LasR-null mutant, albeit lower than the LasR-MexT mutant.This intermediate level of QS-controlled virulence factor production in the LasR-RpoA* mutant is well in line with its QS-active phenotype.To further assess the changes in bacterial virulence brought by LasR-RpoA*, we conducted cell-FIG.6.Effects of chloramphenicol on the growth fitness of the LasR-RpoA* mutant.A) Sensitivity of the indicated mutants to chloramphenicol.The estimated 10 5 -10 6 CFU of each tested strain was spread onto LB plates containing paper disks loaded with 40 μg/ml chloramphenicol.The photograph was taken after 24 h of incubation at 37 °C.B) Quantification of diffusion diameters formed around the paper disk in (A).C) Experimental set-up of the competition assay.The LasR-RopA* and LasR-MexT mutants were labeled with a Gm resistance gene (pUC18 T-mini-Tn7 T-Gm).Each mutant was then cocultured with the LasR-null mutant at a start ratio of 5:95 in a PM-casein medium containing the indicated concentrations of chloramphenicol.After incubation for 48 h, colonies were counted and relative fitness (w) was estimated by determining the ratio of Malthusian growth parameters.D) Determination of the relative fitness of LasR-RpoA* or LasR-MexT against LasR-null using the competition assay.Data represent four biological replicates.Cai et al. • https://doi.org/10.1093/molbev/msad203MBE killing experiments using host cells of P. aeruginosa.Chinese hamster ovary (CHO) cells and human lung adenocarcinoma A549 cells were exposed to the test strains, and the induced cell death was quantified by measuring the release of cytosolic lactate dehydrogenase (LDH) from the cytosol.Consistent with its intermediate level of QS-dependent virulence factors, LasR-RpoA* caused an enhanced cell death compared with the LasR-null mutant, although it demonstrated lower cytotoxicity than the LasR-MexT mutant (fig.7).Overall, our results show that the non-MexT mutant LasR-RpoA* enables QS-reprogramming, albeit with a trade-off in compromised bacterial pathogenesis on host cells.

Discussion
A novel approach, named TGD-MS, was used in the evolution experiments of this study to discover mutations in genes involved in QS-reprogramming in a LasR-null mutant of P. aeruginosa.In the present study, the TGD-MS approach was used for obtaining QS-active revertants.By constructing a strain containing an extra copy of mexT, QS-active revertants carrying a mutated mexT gene could be filtered out to a large extent in our screen for QS-reprogramming mutants.Using this increased screening stringency, a QS-active mutant with a single nucleotide substitution in rpoA was identified.It is worth noting in this context that it would have been impossible to identify this new QS-reprogramming function of RpoA by conventional transposon insertional mutagenesis, because rpoA is an essential gene for cell viability (Lee et al. 2015).More generally, the TGD-MS method described in this work allows us to elucidate genetic components of known pathways and even to reveal unknown metabolic pathways.Mutations in uncharacterized genes of pathways can be rapidly identified using phenotypic screening under defined selective conditions.The use of the TGD-MS method in evolution experiments will provide scientists with new opportunities to make fundamental discoveries in a variety of research areas.We hypothesize that the TGD-MS method can be employed to uncover genetic components of known and unexplored pathways in any transformable haploid organism.
Our study on a QS-inactive LasR-null mutant of P. aeruginosa uncovers a novel QS-reprogramming role for RpoA (fig.8A), which is an essential subunit of the bacterial RNAP.Adaptive mutations in rpoA have been found to be beneficial for bacterial growth in evolution experiments with different organisms (Le Gac et al. 2013;Rajaraman et al. 2016).Previous publications showed that mutations in rpoA of Bacillus subtilis and Escherichia coli promote the utilization of secondary carbon sources and ribosome synthesis (Murayama et al. 2015).The QS-related function of RpoA* found in the present study implicates that the T262 to A substitution of this protein variant leads to a structural modification in RpoA, which affects the interaction between RNAP and the promoter DNA of the mexEF-oprN operon (fig.5).These findings are in agreement with previous studies on bacteria that certain residues in the α-CTD domain of RpoA are important for promoter contact, whereas a subset of these residues is New QS Function of a RpoA Mutant • https://doi.org/10.1093/molbev/msad203MBE required for the interaction with transcriptional effectors (Gaal et al. 1996;Murakami et al. 1996;Finney et al. 2002).Overall, our results suggest that a minor structural change in RpoA caused by a single point mutation restores QS-dependent activities in a LasR-null mutant, thereby allowing adaptation to rapidly changing environments.LasR-MexT mutants of P. aeruginosa exhibit low expression of the mexEF-oprN operon, suggesting that weak or absent protein levels of the MexEF-OprN efflux pump lead to QS-reprogramming in evolution experiments (Oshri et al. 2018;Kostylev et al. 2019).Similarly, the QS-active LasR-RpoA* mutant in our study showed QS-related activities and transcriptome changes, including reduced expression of mexEF-oprN operon genes.Our transcriptome analysis, EMSA, and the QS-active phenotype of the LasR-RpoA* mutant collectively suggest that RpoA* affects the QS network through expression of mexEF-oprN operon genes.The MexEF-OprN efflux pump exports the PQS precursor HHQ and inactivation of this pump therefore increases PQS levels within the cell, resulting in induction of PQS-related QS responses (Köhler et al. 2001; Lamarche and Déziel However, deletion of key components of the PQS system, such as PqsA and/or PqsE (Déziel et al. 2004;Diggle et al. 2006), only partially reduced QS activity in a LasR-MexT mutant (Kostylev et al. 2019).Similarly, the LasR-RpoA*-PqsA mutant deficient in PQS synthesis was found to possess weak but detectable QS activity in the present study, indicating that the RpoA*-mediated QS activation in LasR-RpoA* is partially independent of the PQS system.
Multidrug-resistance pumps of the RND family play key roles in bacterial resistance against antibiotics (Schweizer 2003).The mexEF-oprN operon of P. aeruginosa codes for a typical RND efflux pump, which renders the bacterium largely resistant to certain antibiotics, such as chloramphenicol, fluoroquinolones, and trimethoprim (Lister et al. 2009).On the other hand, the MexEF-OprN efflux pump can also export HHQ, and overexpression of mexEF-oprN operon genes resulted in reduced QS-related activities (Köhler et al. 2001;Lamarche and Déziel 2011).Thus, protein levels of this efflux pump may ultimately affect QS-controlled bacterial virulence of clinical P. aeruginosa isolates.Previous studies reported that CF patients possess QS-active LasR-null strains in their lungs (Bjarnsholt et al. 2010;Feltner et al. 2016).Pseudomonas A) The LasR-RpoA* mutant shows downregulated expression of mexEF-oprN operon genes, resulting in the reduction of intercellular exportation of antibiotics and QS signal molecules.Thus, the accumulation of antibiotics and QS signal confers bacterial antibiotic susceptibility, although activating the QS circuit.B) The activity of the MexEF-OprN efflux pump positively correlates with antibiotic resistance whereas it negatively associates with QS activity.For example, the activation of MexEF-OprN enhances antibiotic resistance, but decreases QS activity and vice versa.Cai et al. • https://doi.org/10.1093/molbev/msad203MBE aeruginosa isolates from chronically infected CF patients often develop antibiotic resistance because they receive continuous high doses of antibiotics over a long period of time (Winstanley et al. 2016).In other words, the expression of efflux pump genes is required for antibiotic resistance but impairs QS circuits that control the production of bacterial virulence factors.This raises the question of how QS-inactive LasR-null mutants resolve the dilemma between antibiotic resistance and QS-reprogramming.The findings of this study highlight that the MexEF-OprN efflux pump is at the center of this trade-off between antibiotic resistance and QS-controlled virulence factor production (fig.8B).The single amino acid substitution in RpoA* caused reduced expression of the mexEF-oprN operon genes in our study.Accordingly, when compared with the LasR-MexT mutant, the LasR-RpoA* mutant was found to have relatively higher growth fitness in competing with the LasR-null parental strain when the bacteria were exposed to increasing concentrations of chloramphenicol.In conclusion, our work shows that a mutation in rpoA can fine-tune the efficiency of an RND efflux pump required for antibiotic resistance and efflux of HHQ, thereby modulating antibiotic resistance and the production of QS-dependent virulence factors under varying environmental conditions.
Pseudomonas aeruginosa possesses a complicated regulatory gene network to control the expression of virulence genes (Huang et al. 2019) and undergoes evolutionary changes in the phase of chronic CF infection (Winstanley et al. 2016).In our TGD-MS approach for example, when an extra copy of mexT was inserted into a LasR-null mutant, we observed that the bacterium can activate MexT-independent alternative QS-reprogramming pathways.Given that MexT controls not only the expression of mexEF-oprN operon genes but also the transcription of potentially important virulence genes (Tian, Mac Aogain, et al. 2009;Tian, Fargier, et al. 2009), it can be speculated that maintaining a non-modified MexT is critical for survival in chronically infected CF patients.Notably, our mexT sequence analysis and previous observations indicate that numerous LasR-null, QS-active P. aeruginosa isolates from CF patients retain a functional MexT (Smith et al. 2006;Cruz et al. 2020).Taken together, we report in this study on a novel experimental evolution approach that facilitated the identification of non-mexT genes involved in the formation of the MexEF-OprN efflux pump and QS-reprogramming.Future research is needed to identify similar mutations in clinical isolates and to clarify their effects on QS activities, virulence and, antibiotic resistance.

Bacterial Strains and Growth Conditions
Pseudomonas aeruginosa PAO1 and mutant derivatives (supplementary table S6, Supplementary Material online) were grown in Luria Bertani (LB) broth (containing 10 mg/ml tryptone, 5 mg/ml yeast extract, 10 mg/ml NaCl) at 37 °C.LB broth was buffered with 50 mM 3-(N-morpholino) propanesulfonic acid, pH 7.0 (LB-MOPS broth).The photosynthetic medium (PM) (Kim and Harwood 1991) supplemented with 1% sodium caseinate (Sigma Aldrich, St. Louis, MO, USA) as the sole carbon source (PM-casein medium) was used for evolution experiments at 37 °C.Unless otherwise specified, P. aeruginosa strains were cultured in 14 ml Falcon tubes (Corning Inc., Corning, NY, USA) containing a 3 ml medium, with shaking (220 rpm) at 37 °C.Escherichia coli was grown in LB broth at 37 °C.Details on bacterial strains and plasmids used in this study are shown in supplementary table S6, Supplementary Material online.

Construction of P. aeruginosa Mutants
Construction of P. aeruginosa mutants (LasR-MexT with mexT deletion, LasR-RpoA* mutant with a rpoA point mutation (T262→A), LasR-RpoA*-PqsA with a pqsA deletion) was performed according to a homologous recombination approach described previously (Rietsch et al. 2005).Briefly, about 500-1000 bp of DNA flanking the targeted single nucleotide substitution or the full length gene of interest were PCR-amplified and cloned into the pEXG-2 vector containing a gentamycin (Gm) resistance gene (Stover et al. 2000;Rietsch et al. 2005) using the Vazyme ClonExpress II One Step Cloning Kit (Vazyme Biotech, Nanjing, China).The constructed plasmids were then mobilized into P. aeruginosa (LasR-null and LasR-RpoA*, respectively) by triparental mating with the help of E. coli carrying the helper plasmid pRK2013 (Figurski and Helinski 1979).Mutants were first selected on Pseudomonas Isolation Agar containing Peptic digest of animal tissue 20.0 g/l, magnesium chloride 1.4 g/l, potassium sulfate 10.0 g/l, triclosan (Irgasan) 0.025 g/l, and agar 13.6 g/l supplemented with 100 μg/ml Gm and then on LB agar containing 10% (w/v) sucrose.All mutants were confirmed by PCR amplification and subsequent DNA Sanger sequencing.The primers used in this study are listed in supplementary table S7, Supplementary Material online.

Complementation of the LasR-RpoA* Mutant
The rpoA gene was cloned into the pJN105 expression vector with the Vazyme ClonExpress II One Step Cloning Kit, generating plasmid pJN105-rpoA.The constructed plasmid and the empty vector pJN105 as a control were then mobilized into the LasR-RpoA* mutant by triparental mating.The bacteria were grown in LB-MOPS broth at 37 °C.Expression of rpoA in the complementation strain was induced by addition of 1 mM L-arabinose, and the bacteria were harvested 20 h later.

Evolution Experiments
Three independent colonies of the LasR-null mutant carrying a miniTn7-mexT construct at a neutral site of the genome were separately inoculated into a 3 ml LB-MOPS broth to obtain an overnight culture.One hundred microliters of the bacterial suspension was then transferred into a 3 ml PM-casein medium in 14 ml Falcon tubes (Corning).

MBE
One passage was performed every 5 days by transferring 100 μl of bacterial suspension into a 3 ml fresh PM-casein medium.When required, water was added to compensate for an evaporated medium.The bacterial suspensions were then spread on LB agar plates.Finally, to visualize extracellular protease activity of the obtained colonies, the bacteria were subjected to a skim milk assay.An initial evolution experiment with a non-modified LasR-null mutant population (carrying a single mexT gene copy) was performed in a similar way.Furthermore, an iterative evolution experiment was performed with a LasR-null mutant carrying miniTn7-mexT and an additional miniTn7-rpoA construct.

Skim Milk Assay
Total extracellular proteolytic activity of P. aeruginosa strains was evaluated using skim milk agar plates on which the bacteria form a protease-catalyzed clearing zone surrounding each colony.Individual colonies were grown on LB agar and then spotted on the skim milk agar plates (25% (v/v) LB, 4% (w/v) skim milk, 1.5% (w/v) agar).Colonies with extracellular proteolytic activity formed clearing zones after incubation at 37 °C for 24 h.The area of transparent zones reflecting extracellular proteolytic activity was quantified from captured photographs.

Pyocyanin Measurement
Overnight cultures of P. aeruginosa grown in LB-MOPS broth were diluted into 4 ml Pseudomonas P broth (20 g/l pancreatic digest of gelatin, 1.4 g/l magnesium chloride, 10 g/l potassium sulfate) to reach a starting OD 600 ≈ 0.02.The bacteria were cultured at 37 °C for 20 h.The cells were then centrifuged at 16,000 g × 2 min.The culture supernatants were collected, and their OD values at 695 nm were photometrically determined at 695 nm.Pyocyanin production was estimated by determining OD 695 /OD 600 values over time.

Hydrogen Cyanide Measurement
A test paper method was used to detect hydrogen cyanide produced by P. aeruginosa strains.A circle of a Whatman 3MM chromatography paper corresponding to the size of an agar plate was soaked in the HCN detection reagent consisting of 5 mg copper (II) ethyl acetoacetate 5 mg 4,4′-methylenebis-(N,N-dimethylaniline) mixed with chloroform (1 to 2 ml).Pseudomonas aeruginosa cells grown overnight in LB-MOPS broth were used to spot-inoculate 2% (w/ v) peptone agar plates.After incubation at 37 °C for 12 h, the plates were overlaid with the prepared test paper and incubated at 37 °C for additional 12 h.The color of the paper turned blue when it was exposed to hydrogen cyanide.ImageJ software (https://imagej.nih.gov/ij) was used to quantify the production of hydrogen cyanide.

Labeling of miniTn7-Gm
The pUC18T-mini-Tn7T-Gm (NCBI accession number: AY599232) (Choi 2006) was integrated into a neutral site of the PAO1 genome by triparental mating using the helper plasmid pTNS2 (NCBI accession number: AY884833).The integration event was confirmed by PCR amplification and DNA sequencing.The excision of the Gm resistance gene was performed with the pFLP2 plasmid (NCBI accession number: AF048702) (Choi 2006) and selection on LB agar containing 5% (w/v) sucrose.

Assays With Strains Containing Reporter Plasmids
Plasmids with a rhlΑ promoter-GFP fusion (Feltner et al. 2016) and a constructed pqsΑ promoter-GFP fusion were used to quantify the RhlR-and PQS-responsive activities, respectively.A plasmid containing a mexE promoter-GFP construct was used to assess the expression of the mexEF-oprN operon.The reporter plasmids were mobilized into P. aeruginosa strains and selected on LB agar plates (containing Gm).PAO1 strains carrying QS reporter plasmids were grown in LB-MOPS broth containing 50 mg/ml Gm for 12 h.The bacterial suspensions were then diluted to LB-MOPS broth containing 50 mg/ml Gm (OD 600 ≈ 0.02) and grown for 18 h to stationary phase.Finally, the bacteria were transferred to 96-well plates (200 µl/well) with three technical replicates.Fluorescence (excitation 488 nm, emission 525 nm) and optical density (OD 600 ) values of the samples were determined using a microplate reader (Synergy H1MF, BioTek Instruments, Winooski, VT, USA).

Quantification of C4-HSL and PQS
Pseudomonas aeruginosa strains were cultured overnight in LB-MOPS broth.Bacteria were then grown in a 4 ml LB broth at 37 °C for 18 h (starting OD 600 ≈ 0.02).AHLs were extracted with an equal amount of ethyl acetate.A bioassay with the reporter E. coli strain containing the pECP61.5 with a rhlA-lacZ fusion and an IPTG-inducible rhlR (Feltner et al. 2016) was used to quantify C4-HSL in the ethyl acetate extract.Precise quantification of PQS was performed by liquid chromatography-mass spectrometry (LC-MS) as described previously (Déziel et al. 2004).Ten microliters of the ethyl acetate phase was subjected to LC-MS analysis.The detection system (Q-Exactive Focus/Ultimate 3000; Thermo Fisher Scientific, Waltham, MA, USA) was equipped with a 100 × 2.1 mm, a 1.7 µm ACQUITY UPLC HSS T3 chromatographic column (Waters, Milford, MA, USA).An acidified (1% glacial acetic acid by volume) water/methanol gradient was used as the mobile phase (0.4 ml/min flow rate; 40 °C).PQS was quantified by measuring the area of PQS peaks in chromatograms from different samples, and values were standardized according to the concentration of an added internal standard.Full length of His-tagged RpoA or the RpoA* variant was cloned into pJN105 expression vector.The plasmid was then mobilized into the LasR-RpoA* mutant, and protein expression was induced by growth in LB broth containing 1 mM L-arabinose at 37 °C for 18 h.Cells were pelleted at 13,000 g × 5 min and resuspended in a lysis buffer (PBS buffer supplemented with 1 × Protease Inhibitor Cocktail [Bimake, Houston, TX, USA], 1 mg/ml lysozyme).Resuspended cells were lysed in a high pressure cell homogenizer machine (Stansted Fluid Power Ltd, Essex, United Kingdom) at 420 MPa.The lysed cells were centrifuged at 13,000 g × 5 min (4 °C).The supernatant was then incubated with Ni-NTA beads (Qiagene, Shanghai, China) at 4 °C for 2 h.The mixture was loaded on a chromatographic column, washed with a washing buffer (500 mM NaCl, 20 mM NaH 2 PO 4 , 20 mM imidazole, pH 8.0), and finally eluted with a buffer containing high concentrations of imidazole (500 mM NaCl, 20 mM NaH 2 PO 4 , 250 mM imidazole, pH 8.0).Collected fractions were assessed by Native PAGE analysis.Fractions containing recombinant protein were pooled and concentrated with an ultrafiltration device (MilliporeSigma, Burlington, MA, USA) and finally eluted with a PBS buffer.Samples were stored at −80 °C.

Electrophoretic Mobility Shift Assay (EMSA)
The promoter sequence of the mexEF-oprN operon (PmexE) was used as probe for EMSA.Probe DNA was labeled according to our published method (Cai et al. 2023).Briefly, the probe DNA was labeled with Biotin-11-UTP (Jena Bioscience, Jena, Germany) through PCR amplification using T4 DNA polymerase (Novoprotein, Suzhou, China).The labeled probe (30 ng) was incubated with 0.2 μg purified His-tagged RpoA or RpoA* protein in a binding buffer (10 mM Tris-HCl, 50 mM KCl, 1 mM DTT, 1% (v/v) glycerin, 10 mM MgCl 2 , 1 μg/μl poly(deoxyinosinic-deoxycytidylic) acid, pH 7.5) at room temperature for 30 min.The samples containing a given DNΑ-protein complex were subjected to electrophoresis on a 5% polyacrylamide gel in a 0.5× Tris-borate-EDTA buffer at 100 V for 180 min.DNA subsequently was transferred to a nylon membrane and incubated with a milk blocking buffer (50 g/l milk powder in a TBST buffer (50 mM Tris, 150 mM NaCl, 0.05% Tween-20, pH 7.5) for 30 min with shaking at room temperature.The membrane was then incubated with streptavidinhorseradish peroxidase conjugate (Thermo Fisher Scientific) in a blocking buffer for 30 min at room temperature.The membrane was subsequently washed with a TBST buffer for four times, 5 min each.DNA was visualized using the Immobilon Western kit (Millipore) and photographed by a Tanon 5200 Multi-Imaging System (Tanon, Shanghai, China).

Mammalian Cell Cytotoxicity Assay
Human lung cancer A549 cells and CHO cells were cultured in DMEM medium (GIBCO, Thermo Fisher Scientific, China) and RPMI1640 medium (GIBCO, Thermo Fisher Scientific, China) supplemented 10% FBS (GIBCO), respectively, at 5% CO 2 and 37 °C.Exponentially growing P. aeruginosa bacteria cultured in LB broth (OD 600 ≈ 0.6) were collected by centrifugation (4,000 g × 5 min) and washed with PBS.Next, the cells were added to near-confluent CHO and A549 cells at a starting multiplicity of infection ratio of 5:1.After incubation at 37 °C for 6 h, the extent of cell killing was determined by quantification of the release of LDH into the cell culture supernatant using the LDH Cytotoxicity Detection Kit (Beyotime, Nantong, China).

Competition Experiment
The P. aeruginosa strains (LasR-null, LasR-RpoA*, and LasR-MexT) were grown in a PM medium supplemented with 0.5% (w/v) casein amino acids (Sangon, Shanghai, China) overnight and then transferred to a liquid PM-casein medium (adjusted to OD 600 ≈ 0.02) with or without chloramphenicol.Each strain tagged with a pUC18T-mini-Tn7T-Gm construct was cocultured with the non-tagged strain at a starting ratio of 5:95.The mixed cells were then grown at 37 °C for 48 h.Cell counts were determined by spreading the bacteria onto LB agar plates supplemented with or without 10 μg/ml Gm.The relative fitness of each strain was deduced from the ratio of Malthusian growth parameters (w) = ln(X 1 /X 0 )/ln(Y 1 /Y 0 ), as defined previously (Lenski et al. 1991).The experiment was repeated four times with similar results.

Extraction and Sequencing of RNA
Pseudomonas aeruginosa strains were grown in LB-MOPS broth at 37 °C for 18 h (OD 600 ≈ 3.8-4.0).Total RNA was isolated using TRIzol Reagent (Thermo Fisher Scientific).The quantity of extracted total RNA was measured using a NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific).Extracted RNA was treated with DNase (RQ1 RNase-free DNase I, Promega, Beijing, China) to remove traces of genomic DNA.The obtained total RNA (two independent RNA samples per strain) was sent to Novagen (Tianjin, China) for stranded paired-end mRNA-seq sequencing using the Illumina Novaseq Platform.
Quantitative Real-Time PCR (qRT-PCR) Total RNA of P. aeruginosa strains was reverse transcribed using the HiScript II 1st Strand cDNA Synthesis Kit (Vazyme Biotech) following the manufacturer's instructions.The obtained cDNA was then used for qRT-PCR analysis.Reactions containing the SYBR qPCR Master Mix (Vazyme Biotech) were prepared in 96-well plates and run in a StepOnePlus Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) as recommended by the supplier.Primers used for qRT-PCR are listed in supplementary table S7, Supplementary Material online.The measured values for a given gene were normalized to the expression level of the proC gene.Reactions were performed in triplicate.

Analysis of Illumina HiSeq Short Reads
Raw short reads were subjected to quality control including removing adapters using Cutadapt by Novagen New QS Function of a RpoA Mutant • https://doi.org/10.1093/molbev/msad203MBE

FIG. 1 .
FIG. 1.Scheme illustrating the novel TGD-MS strategy.A) Illustration of the engineering of the recombinant LasR-null mutant.An extra copy of mexT was inserted into a neutral site of the LasR-null mutant genome, generating the recombinant LasR-null mutant carrying two copies of mexT.B) The overall procedure of the TGD-MS strategy developed in our study aims to discover mutations in non-mexT genes.The evolution experiment is initiated with a QS-inactive recombinant LasR-null mutant with an extra copy of mexT.The bacteria are spread on skim milk agar plates to select QS-active revertants secreting QS-dependent proteases.This screening method results in the identification of mutations in genes other than mexT.

FIG. 2 .
FIG. 2.Identification of QS-active mutants obtained from a LasR mutant with an extra copy of mexT.A) Culture tube of the LasR-null mutant carrying an extra copy of mexT, which was inoculated into a PM-casein medium and photographed at different time points.The culture of the engineered mutant is slightly turbid at day 30 (d30) due to the emergence of QS revertants.B) Frequency of QS revertant colonies obtained from three TGD-MS experiments with the LasR mutant carrying an extra copy of mexT.QS revertant colonies were selected based on their proteasepositive phenotypes on the skim milk agar plate.C) Bacterial colonies were examined for extracellular protease activity at different time points using skim milk agar plates (control strains: WT, wild-type strain PAO1; LasR-null, LasR-null mutant used as a negative control; LasR-MexT, LasR-MexT employed as a positive control).D) Pyocyanin production in identified protease-positive colonies.Pyocyanin production (OD 695 /OD 600 values) was quantified for each strain (QS revertant-1/2/3/4, obtained mutants displaying protease-positive phenotype and significantly increased pyocyanin production).Data are presented as means ± SD (n = 3).***P < 0.001 (t-test).

FIG. 4 .
FIG. 4.Transcriptome analysis of the LasR-RpoA* mutant.A) Volcano plots showing the magnitude of differential gene expression induced by the identified T262->A mutation in rpoA (transcriptome comparison of LasR-RpoA* compared with the parent strain LasR-null).Each dot represents one annotated sequence with detectable expression.Thresholds for defining a gene significantly differentially expressed (log 2 (fold change) ≧ |1.0|, P value ≦ 0.05) are shown as dashed lines.B) KEGG pathway analysis of differentially expressed genes.Fold enrichment shows the enrichment of differentially expressed genes in the corresponding pathway (differentially expressed gene number to the total gene number in a certain pathway).The size and color of the bubble represent the amount of differentially expressed genes enriched in the pathway and enrichment significance, respectively.C, D) Expression levels of genes encoding QS system components (C) and MexEF-OprN efflux pump and its regulators (D).A positive log 2 (fold change) value indicates upregulation, and a negative value indicates downregulation of a given gene in the LasR-RpoA* mutant compared with LasR-null.Data are indicated by means ± SD.
FIG. 5. RpoA* modulates expression of mexEF-oprN operon genes.A) The LasR-RpoA* mutant shows reduced transcription of mexEF-oprN operon genes as estimated with the help of PmexE-GFP, a mexEF promoter-GFP fusion construct.Data of indicated strains are presented as means ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 (t-test).B) Illustration of the probe DNA used for the performed EMSA with recombinant RpoA of wildtype (WT) bacteria or RpoA*.C) EMSA results showing that RNAP containing RpoA forms a DNΑ-protein complex.A weaker signal is seen for the sample with recombinant RpoA*.The unbound probe DNA is seen at the bottom of the blot.

FIG. 7 .
FIG. 7.The RpoA* mutation significantly enhances host cell cytotoxicity.A) Illustration of P. aeruginosa cell killing on host cells.B, C) The killing assay with human lung cancer A549 cells (B) and Chinese hamster ovary (CHO) cells (C ) was performed with equal amounts of the indicated strains.After incubation for 6 h, the release of cytosolic lactate dehydrogenase (LDH) from infected cells was quantified.The released amount of LDH inoculated with the wild-type strain (WT) was set to 100%.Data represent means ± SD (n = 3).*P < 0.05, **P < 0.01, ***P < 0.001 (t-test).
FIG.8.Schematic representation of the reduced activity of MexEF-OprN in LasR-RpoA*.A) The LasR-RpoA* mutant shows downregulated expression of mexEF-oprN operon genes, resulting in the reduction of intercellular exportation of antibiotics and QS signal molecules.Thus, the accumulation of antibiotics and QS signal confers bacterial antibiotic susceptibility, although activating the QS circuit.B) The activity of the MexEF-OprN efflux pump positively correlates with antibiotic resistance whereas it negatively associates with QS activity.For example, the activation of MexEF-OprN enhances antibiotic resistance, but decreases QS activity and vice versa.