New insights into Quorum Sensing and other infection-related processes in Pseudomonas aeruginosa and Burkholderia cenocepacia

Abstract

Pseudomonas aeruginosa and Burkholderia cenocepacia, beside their relevance as a nosocomial pathogens, cause lethal chronic lung infections in the vast majority of cystic fibrosis (CF) patients. Once established, the CF lung infection is impossible to eradicate with traditional antimicrobial therapies. The main body of this project has been focused on P. aeruginosa. In this bacterium the Quorum Sensing (QS) system coordinates the production of virulence factors in a cell density-dependent manner via the secretion of specific signal molecules. During growth the bacteria secrete the signal molecules, which accumulate in the surrounding environment as the population density increases. When a threshold concentration is reached, the receptor is activated by the binding with the cognate signal molecule and triggers the expression of several genes, including virulence genes. P. aeruginosa has two QS systems relying on the production of acylated- homoserine lactone (acyl-HSL) as signal molecules: the las and rhl systems. The lasI gene encodes the LasI synthase producing the N-3-oxo-dodecanoyl-homoserine lactone (3-oxo-C12-HSL) signal molecule, that binds to the LasR receptor encoded by lasR. The rsaL and qscR genes encode repressors of lasI and constitute an integral part of the las QS system, since they respond to 3-oxo-C12-HSL. The rhl system is organized similarly to the las one: the rhlI and rhlR genes encode, respectively, for the RhlI synthase and the RhlR receptor of the N-butyryl-homoserin lactone (C4-HSL) signal molecule. The signal-activated LasR and RhlR receptors regulate transcription of hundreds of genes, including the genes encoding a third QS system, based on the production of a different signal molecule, and the vast majority of virulence genes. In P. aeruginosa many regulators have been found to affect the expression of the las and rhl genes, although their mechanism of action remains largely unknown. In particular, only the Vfr protein has been shown to directly activate lasR transcription. Besides las and rhl mutants, also a vfr mutant shows reduced virulence in murine models, suggesting that also the QS regulators could be feasible drug-targets. The main objective of this project has been the identification and functional characterization of transcriptional factors that directly regulate the las and rhl genes. To achieve this objective, six DNA fragments corresponding to the promoters of lasR, lasI, rsaL, qscR, rhlR and rhlI were independently conjugated to a chromatography resin and incubated with protein crude extracts prepared from P. aeruginosa cultures. The proteins able to specifically bind each DNA bait were recovered, separated by SDSPAGE and identified by MALDI-TOF mass spectrometry. Overall, 25 proteins bound on the promoter regions of the QS genes were identified. Out of these, fifteen factors were selected for further analysis as possible QS regulators. Noteworthy, the activator Vfr, already known to directly bind the lasR promoter, was recovered using this promoter as a bait. A set of fifteen P. aeruginosa mutants, each one deleted in one of the genes encoding the selected proteins, was generated. In each mutant the activity of correspondent target promoter was compared with that of the wild type, by means of transcriptional fusions between the promoter region of the target genes and the reporter system lux. In this system the production of bioluminescence is proportional to the promoter activity. Among the newly identified factors, we showed for the first time that the histon-like proteins MvaT and MvaU directly control the transcription of the las genes. In particular, MvaU is a repressor of both rsaL and lasI transcription, while MvaT represses lasI and qscR transcription and activates lasR transcription. MvaT and MvaU are global regulators of gene expression and on some promoters can also interact each other. The interplay between these two proteins in regulating QS genes is a complex issue that will be the object of future studies. Unexpectedly, mutations in the other genes here investigated did not cause evident effects on the activity of the corresponding target promoter under the tested laboratory conditions. A possible explanation for this result is that some of these factors are actually QS regulators, but they are not sufficiently expressed and/or active in the standard growth conditions we have used. To verify this hypothesis, we decided to test the activity of target promoters in recombinant P. aeruginosa strains over-expressing the putative QS regulators. We focused on the six factors retrieved on the lasR promoter. The results showed that PA3699 over-expression strongly repressed lasR promoter activity during the whole growth curve, without affecting cell growth. The ability of PA3699 to repress lasR expression was also confirmed in the heterologous host Escherichia coli. Moreover, the PA3699 protein was purified and its capability to bind in vitro the lasR promoter region was demonstrated by electrophoretic mobility shift assay. Overall, these results show that PA3699 is a repressor of lasR expression. Since PA3699 mutation has no effect on lasR expression under standard cultural conditions, it is likely that this regulator is activated by specific environmental/metabolic stimuli. This could be the case also for the other proteins identified in this study. Environmental and metabolic stimuli increasing the expression/activity of PA3699 and of the other putative regulators identified in this project are currently under investigation. A minor objective of this part of the project has been the screening of the Prestwick Chemical Library of FDA-approved drugs for the identification of possible secondary activities against P. aeruginosa QS. The screening was performed by using a biosensor able to detect 3-oxo-C12-HSL levels. Among the 1120 compounds tested, 9 displayed a QS-inhibitory effect. The anti-Pseudomonas activity of the most promising compound is currently under investigation in murine models of chronic lung infection. The second part of this project has been focused on B. cenocepacia, a member of the B. cepacia complex (Bcc). This group of strictly related bacteria is the second most important Gram negative pathogen, infecting the CF patients. Like many pathogens, including P. aeruginosa, B. cenocepacia is highly resistant to a wide range of antibiotics due to the production of drug-efflux pumps belonging to the Resistance-Nodulation-cell Division (RND) family. It is emerging the notion that, beside exogenous toxic compounds, RND efflux pumps are involved in the secretion of bacterial products important for virulence, including factors involved in biofilm formation. The biofilm is a bacterial community encased in a self-produced extracellular matrix and confers resistance to antibiotics and to the host immune systems, playing a major role in CF lung infection. The RND efflux pumps have been mainly studied for their effect on antibiotic resistance, while little is known about their impact on cell-physiology and virulence-related phenotypes. A preliminary transcriptomic study showed that mutants inactivated in the D4 and D9 RND- drug efflux systems of B. cenocepacia displayed altered expression of flagella-related genes. Since flagellar motility (a phenotype named swimming) is strictly connected to virulence and to biofilm formation, we tested the effect of these mutation on swimming and biofilm. Results showed that, in accordance with transcriptomic data, the D4 and D9 mutants showed increased and decreased swimming motility with respect to the wild type, respectively. Therefore the inactivation of distinct RND systems can have a dramatically different effect on a specific virulence-related phenotype. Interestingly, both the mutants showed increased biofilm formation with respect to the wild type. Since biofilm formation is a pleiotropic and complex phenotype it is not easy to correlate this result only to flagella genes expression. Many companies are developing Efflux Pumps Inhibitors (EPIs) as antibiotic therapy adjuvants. In this view, a relevant result of this research is that inactivation of efflux pumps can enhance biofilm formation and, sometimes, motility, raising serious concerns about the use of EPIs in therapy. Indeed, the use of EPIs could be, on one side positive for helping the antibiotic therapy, on the other side, it could promote biofilm formation and chronic infection. More detailed study on the effect of RND efflux pumps in virulence-related phenotypes and chronic infection are therefore strongly desirable.