Home & Garden

Quorum Sensing and Virulence of Pseudomonas aeruginosa during Lung Infection of Cystic Fibrosis Patients

Description
Quorum Sensing and Virulence of Pseudomonas aeruginosa during Lung Infection of Cystic Fibrosis Patients
Categories
Published
of 10
2
Categories
Published
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Share
Transcript
  Quorum Sensing and Virulence of   Pseudomonas aeruginosa   during Lung Infection of Cystic FibrosisPatients Thomas Bjarnsholt 1 , Peter Østrup Jensen 2 , Tim Holm Jakobsen 1 , Richard Phipps 3 , Anne KirstineNielsen 1 , Morten Theil Rybtke 1 , Tim Tolker-Nielsen 1 , Michael Givskov 1 , Niels Høiby 1,2 , Oana Ciofu 1 * , theScandinavian Cystic Fibrosis Study Consortium 1 Institute for International Health, Immunology and Microbiology, University of Copenhagen, Copenhagen, Denmark,  2 Department of Clinical Microbiology, UniversityHospital, Rigshospitalet, Copenhagen, Denmark,  3 BioSys, Technical University of Denmark, Lyngby, Denmark  Abstract Pseudomonas aeruginosa  is the predominant microorganism in chronic lung infection of cystic fibrosis patients. The chroniclung infection is preceded by intermittent colonization. When the chronic infection becomes established, it is well acceptedthat the isolated strains differ phenotypically from the intermittent strains. Dominating changes are the switch to mucoidity(alginate overproduction) and loss of epigenetic regulation of virulence such as the Quorum Sensing (QS). To elucidate thedynamics of   P. aeruginosa  QS systems during long term infection of the CF lung, we have investigated 238 isolates obtainedfrom 152 CF patients at different stages of infection ranging from intermittent to late chronic. Isolates were characterized withregard to QS signal molecules, alginate, rhamnolipid and elastase production and mutant frequency. The genetic basis forchange in QS regulation were investigated and identified by sequence analysis of   lasR ,  rhlR, lasI   and  rhlI  . The first QS system tobe lost was the one encoded by  las  system 12 years (median value) after the onset of the lung infection with subsequent lossof the  rhl   encoded system after 17 years (median value) shown as deficiencies in production of the 3-oxo-C12-HSL and C4-HSLQS signal molecules respectively. The concomitant development of QS malfunction significantly correlated with the reducedproduction of rhamnolipids and elastase and with the occurrence of mutations in the regulatory genes  lasR  and  rhlR. Accumulation of mutations in both  lasR  and  rhlR  correlated with development of hypermutability. Interestingly, a highernumber of mucoid isolates were found to produce C4-HSL signal molecules and rhamnolipids compared to the non-mucoidisolates. As seen from the present data, we can conclude that  P. aeruginosa  and particularly the mucoid strains do not lose theQS regulation or the ability to produce rhamnolipids until the late stage of the chronic infection. Citation:  Bjarnsholt T, Jensen PØ, Jakobsen TH, Phipps R, Nielsen AK, et al. (2010) Quorum Sensing and Virulence of   Pseudomonas aeruginosa  during LungInfection of Cystic Fibrosis Patients. PLoS ONE 5(4): e10115. doi:10.1371/journal.pone.0010115 Editor:  James L. Kreindler, Abramson Research Center, United States of America Received  November 5, 2009;  Accepted  March 9, 2010;  Published  April 12, 2010 Copyright:    2010 Bjarnsholt et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the srcinal author and source are credited. Funding:  This study was supported by The Carlsberg Foundation and Lundbeck Foundation to TB and grants from the German Mukoviszidose e.V. and theDanish Strategic Research Council to MG. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of themanuscript. Competing Interests:  The authors have declared that no competing interests exist.* E-mail: ociofu@sund.ku.dk  Introduction The onset of the chronic lung infection with  Pseudomonas aeruginosa  in CF patients is preceded by intermittent colonization [1] usuallywith environmental strains [2]. The chain of events leading to theestablishment of a persistent infection is mainly due to the biofilmforming capacity of   P. aeruginosa   with important contributions fromindividual virulence factors such as elastase [3], LPS [4],rhamnolipids [5] and alginate [6]. We have demonstrated thatrhamnolipid plays a major role in the defense against the cellularcomponents of the immune system, especially against the polymor-phonuclear neutrophilic leukocytes (PMNs) which dominate theimmune response in the CF lung [7–9].  P. aeruginosa   respond to thepresence of PMNs by upregulating synthesis of a number of  virulence determinants including rhamnolipids, all of which are ableto cripple and eliminate cells of the host defense which support a‘launch a shield’ model by which rhamnolipids surround the biofilmbacteria and on contact eliminate incoming PMNs [9].Production of several  P. aeruginosa   virulence factors iscoordinated by a cell density monitoring mechanism termedQuorum Sensing (QS) [10–12].  P. aeruginosa   employ twodominating QS system the  las   and the  rhl   encoded system. Bothsystems feature specific signal molecules for separation of theprocesses, 3-oxo-C12-HSL and C4-HSL respectively. The basic AHL QS system is comprised of an I gene encoding the AHLsynthetase and a R gene encoding the receptor. During thegrowth of the bacteria, system specific signal molecules areproduced by the synthetase, the I protein. The signal moleculesproduced by the bacteria bind to the receptor, the R-protein,the AHL-responsive transcriptional activator. The regulatorproteins contain two functional domains. The signal moleculebinding region, which is located in the N-terminal portion of theprotein and a helix-turn-helix motif (HTH) located in the C-terminal, which is responsible for the protein binding to thetarget promoters [13–15]. Within these systems a thirdanalogous receptor, the QscR operates with 3-oxo-C(12)-HSL PLoS ONE | www.plosone.org 1 April 2010 | Volume 5 | Issue 4 | e10115  to modulate gene expression of a specific regulon which overlapswith the two other  las   and  rhl   regulons [16].  P. aeruginosa   has anadditional QS regulatory pathway termed the Pseudomonasquinolone signal (PQS) system [17].  In vitro  the QS systems of   P.aeruginosa   have been shown to be hierarchically arranged, withthe  las   system on top, controlling the  rhl   system [18] and thePQS system positioned as a mediator functionally positionedbetween the  las   and  rhl   systems. However, it has been proposedthat the  rhl   system can be activated independently of the  las  system, and it has been suggested that PQS system controls thisactivation [17]. This was further substantiated in a recent paper,where the authors provided evidence that  rhl   system is able toovercome the absence of the  las   system by activating specificLasR-controlled functions, including production of 3-oxo-C(12)-HSL and PQS [19].When the chronic lung infection in CF patients is establishedit is well recognized that  P. aeruginosa   isolated from the sputumdiffer phenotypically from the initial intermittent strains eventhough they produce similar pulse field gel electrophoresispatterns and therefore are considered isogenic [20,21]. Loss of epigenetic regulatory systems such as QS is one of thedominating changes that occur during the adaptive processof the bacteria in the CF lung [22]. Different modelsaccounting for the selection of non-functional QS systemshave been reported. One such model focuses on the specialnutrient availability in the CF lung which  P. aeruginosa   has toadapt to [22]. This model is supported by comparison of thegenomes obtained from different CF isolates [23] suggesting that  P. aeruginosa   has the potential to act in a range of environmental conditions. Furthermore, the authors suggestthat the bacterium acquires or discards genomic segments inorder to optimize its genomic repertoire for the present specificenvironment. Another model focuses on the fact that  P.aeruginosa   is exposed to oxygen radicals which in turn inducegenetic mutations [23,24]. We have recently demonstrated thatthe polymorphonuclear leukocytes (PMNs) are the majorcontributors of oxygen radicals in CF sputum [25]. It istherefore likely that oxygen radicals are derived from thePMNs. Recently, the cooperative behavior of mixed popula-tions of bacteria has been studied using populations including both QS wild-type and  lasR   mutants [26]. These studies haveintroduced the concept of ‘‘cheaters’’ (the QS mutants)exploiting the functional QS systems of other members of thepopulation [26,27]. It might be that in CF lungs, although  P.aeruginosa lasR-  mutants may accumulate, QS-active membersof the population are still maintained for the benefit of allmembers of the bacterial community.Based on these observations we aimed to correlate the changesthat occur in the QS systems with expression of virulence during stages of intermittent and chronic lung infections in CF patients.The capability to produce 3-oxo-C12-HSL and C4-HSL signalmolecules and the sequences of   lasR   and  rhlR   encoding thereceptor-transcriptional regulators as well as the  lasI   and  rhlI  encoding the synthethases were investigated in a large number of randomly collected CF isolates, (pairs of mucoid and non-mucoidif available) obtained from the intermittent or chronic stages of lung infection. The dynamics of the functionality of QS systems inthe clinical strains were correlated to rhamnolipids and elastaseproduction as well as to the mutational frequencies of the isolates.Our results show that functionality of the  rhl   encoded system ismaintained longer than the  las   system during the chronic infection,especially in the mucoid isolates, providing evidence for thepossible role of QS inhibitors in the treatment of early as well aslate stages of   P. aeruginosa   infections. Results and Discussion QS functionality and duration of infection Loss of QS regulation is generally considered a hallmark of chronic virulence and has been described for several  P. aeruginosa  CF isolates [22,28]. To investigate the dynamics of the QS loss atdifferent stages of the  P. aeruginosa   lung infection, we determinedthe production of QS signal molecules of isolates collected frompatients with intermittent colonization as well as chronic lung infection at different time points.The  P. aeruginosa   CF isolates from the intermittently colonizedpatients showed significantly higher frequency of strains withsimultaneous production of both QS molecules (  x 2 test p , 0.0001)and higher levels of rhamnolipid production (median [ranges]=15.3[0–26.4]  m g/ml) compared to the isolates from the group of chronically infected patients (median [ranges]=2.4[0–72.8]  m g/ml)(Mann-Whitney, p , 0.0001) (figure 1).The CF isolates were divided in four groups according to theirability to produce QS molecules: a group producing both 3-oxo-C12-HSL and C4-HSL (n=58), a group producing only C4-HSL(n=63), a group producing only 3-oxo-C12-HSL (n=16) and agroup not producing QS molecules (n=73). A significantdifference was found between the duration of the chronic lung infection of the CF patients harboring the isolates belonging to thedifferent groups (table 1). The majority of the CF isolates (63) werenot producing 3-oxo-C12-HSL after 12 years of infection whileonly a small proportion (16 isolates) were 3-oxo-C12-HSLproducers but lost the ability to produce C4-HSL. Importantly,the lost of both QS molecules was found first after 17 years of infection. This shows that the abilities to produce 3-oxo-C12-HSLand C4-HSL signal molecules are lost at different time pointsduring the chronic lung infection and particularly interesting is thefinding of C4-HSL molecules in isolates from the late stages of theinfection. This indicates that the  rhl   system is functional even in thelate phases of the chronic lung infection and suggests that the Las-independent regulation of   rhl   system is maintained during thechronic lung infection. These data emphasize that the shielding through rhamnolipid production might play an important roleduring the first 17 years of infection. Figure 1. Distribution of   in vitro   rhamnolipid production in  P. aeruginosa   isolates from CF patients at different stages of lunginfection.  The box and whisker plots represent the median (thick lineinside the box), 10, 25, 75 and 90 centiles of the  in vitro  rhamnolipidproduction ( m g/ml) of   P. aeruginosa  isolates from intermittently colonizedand chronically infected patients; Mann-Whitney test was used toinvestigate the significance of the difference between the groups.doi:10.1371/journal.pone.0010115.g001QS and  P. aeruginosa  in CFPLoS ONE | www.plosone.org 2 April 2010 | Volume 5 | Issue 4 | e10115  Significant differences in the level of rhamnolipid (table 1) andelastase (table 1) were found between the four groups of QS signalmolecule producers concurring that these virulence factors requirea functional QS system for expression. QS and mucoidity Early occurrence of mucoid  P. aeruginosa   in the sputum of CFpatients has been correlated to a poor prognosis [5,29]. Mucoidityhas been shown to be selected for in the CF lung due to theprotective role of alginates against oxygen radicals from activatedPMNs [30]. In addition, as judged from flow-cell and animalexperiments, mucoid isolates form more robust biofilms[31,32].Investigations of QS functionality and connected phenotypesexpressed by mucoid and non-mucoid isolates that were obtainedfrom the chronically infected patients, showed that a significantlyhigher proportion of mucoid isolates produced C4-HSL comparedto their non-mucoid counterparts (  x 2 test, p=0.02). Furthermore,this number was found to correlate with significantly higheramounts of rhamnolipids (median [ranges]=4.5 [0–72.8]  m g/ml)produced by mucoid compared with non-mucoid isolates (median[ranges]=0[0–48]), (p=0.02, Mann-Whitney) (Figure 2).Thus, mucoid isolates may be protected against the antimicro-bial properties of the PMNs not only by alginate but also due tothe production of rhamnolipids. This is in accordance with recentdata from an animal model of chronic lung infection whichshowed that persistence against the host defense was maintained inmucoid but lost in nonmucoid isolates during the chronic lung infection of one CF patient [33]. This difference in thefunctionality of the QS system between mucoid and non-mucoidisolates strongly support that different adaptation strategies areemployed by the two phenotypes [5,31]. QS and clonal distribution The typing analysis showed that three previously identifiedbacterial clones entitled DK-1 (39 isolates), DK-2 (29 isolates) andNO (26 isolates) were represented among the 238 CF isolates.Clones DK-1 or ‘‘red’’ and DK-2 or ‘‘blue’’ are two dominantclones in the Copenhagen CF Center and clone NO is a cloneidentified among the Norwegian isolates[32–34]. The rest of theisolates were considered non-clonally related.Significant differences in the functionality of the QS systemswere found among the various clonal groups. While functionalityloss of both the  las   and  rhl   systems was found in 75% of the DK-2strains, this phenotype was encountered in only 46% and 33% of the NO and DK-1 isolates, respectively. These differences in theability of the isolates to produce QS signal molecules wereassociated with significant differences in the ability of the isolates toproduce rhamnolipids (Figure 3). Importantly, we always saw apositive correlation between production of C4-HSL and rhamno-lipids but no correlation between 3-oxo-C12-HSL and rhamno-lipid production. This is in accordance with findings by us [35]and others who showed that PAO1 do not require a functional  las  system for expression of   rhl   and  pqs   controlled genes.Several isolates belonging to the DK-2 clone did not harbor a  lasR  gene as shown by the lack of gene amplification which in turn suggeststhat this particular mutant of the DK-2 clone might have spreadamong CF patients after the apparent loss of QS signal recognition, orthat a deletion hotspot exists at the particular chromosomal position.However, this second option is statistically very unlikely. Similarresults were obtained with isolates belonging to the NO and DK-1 Table 1.  Duration of the chronic lung infection of CF patients harboring  P. aeruginosa  isolates producing both C4-HSL and 3-oxo-C12-HSL, either C4-HSL or 3-oxo-C12-HSL or none of the QS molecules and distribution of the rhamnolipid and elastase levels in CF P. aeruginosa  isolates producing both, one or none of the QS molecules. C4-HSL   3-oxo-C12-HSL   (n=58)C4-HSL   3-oxo-C12-HSL -(n=63)C4-HSL –3-oxo-C12-HSL   (n=16)C4-HSL –3-oxo-C12-HSL -(n=72) Duration of chronic infection(years)Median[ranges]7.5*[intermittent-29]12**[intermittent-32]13[1–32]17* , **[intermittent-31]Rhamnolipid ( m g/ml)Median[ranges]16.6 a,b,d [0–53.8]10.7 a,c [0–72.8]1.7 d [0–11]0 b,c [0–48]Elastase activity (mU)Median[ranges]48 1 [0–276]40.3[0–180]46.7[0–109]36.8 1 [0–104.8]*p=0.0002.**p=0.016 (Mann-Whitney). a p=0.008. b p , 0.0001. c p , 0.0001. d p , 0.0001 (Mann-Whitney). 1 p=0.01 (Mann-Whitney).doi:10.1371/journal.pone.0010115.t001 Figure 2. Distribution of rhamnolipid production in mucoid andnonmucoidisolates.  Box and whisker plots represent the median (thick line inside the box), 10, 25, 75 and 90 centiles of the  in vitro  rhamnolipidproduction ( m g/ml) of mucoid and non-mucoid  P. aeruginosa  isolatesfrom chronically infected CF patients. Mann-Whitney test was used toinvestigate the significance of the difference between the groups.doi:10.1371/journal.pone.0010115.g002QS and  P. aeruginosa  in CFPLoS ONE | www.plosone.org 3 April 2010 | Volume 5 | Issue 4 | e10115  clones, although these isolates were found to harbor a functional  rhl  system. This suggests that dissemination and establishment within acommunity of CF patients does not require a functional LasR-system. Alternatively, LasR proficient bacterial subpopulations mighthave been present in the initial infection but these subpopulationswere not identified in our study. However, it is important to mentionthat we identified two CF patients that harbored QS-proficient DK-2bacteria, suggesting that several evolutionary lineages develop in theCF population as recently published by Wilder [36]. QS signal molecules and sequence of the QS genes To investigate the cause of QS loss we performed sequenceanalysis of the genes  lasR   and  rhlR   encoding the QS-regulatorsLasR and RhlR as well as of the genes  lasI   and  rhlI   encoding thesignal molecule synthetases LasI and RhlI, respectively. Theanalysis showed that the wild-type sequences of   lasI   wereconserved among the CF  P. aeruginosa   isolates. From 238 isolates,we only found a single occurrence of a loss of function mutation in lasI   gene and intact  lasR   gene. However, the vast majority of theisolates presented point mutations in  rhlI   (data not shown)especially C249A leading to D83E which interestingly has alsobeen found in isolates from CF patients attending the OregonHealth and Science University [36]. The measurements of C4-HSL were not affected by these point mutations. This indicatesthat these point mutations have no effect on the functionality of the gene and its encoded product. Mutations preferentiallyoccurred in the genes encoding the regulatory proteins, inaccordance with previous observations [37].CF isolates with mutations in the regulatory genes  lasR   or  rhlR  produced significantly less 3-oxo-C12-HSL (  x 2 p , 0.0001) andC4-HSL (  x 2 p , 0.0001) respectively and lower levels of rhamno-lipids compared to the isolates with wild-type genes (Figure 4Aand 4B).The type of mutations identified in  lasR   and  rhlR   genes arepresented in Figure 5 and 6. Mutations in  lasR   were identified inboth the signal-binding N-terminal domain and the DNA-binding domain (C-terminal). The mutations observed (Figure 5) wereinsertions and deletions leading to frame shifts and pointmutations (both transitions and transversions) resulting in eitherstop codons or substitutions in conserved, semi-conserved, ornon-conserved amino acids [38]. The complementation assaysshowed that the identified point mutations were responsible forthe phenotypes (marked in yellow in figure 5). In the signal-binding domain, particular interesting are the mutations causing a Tyr56 to Cys exchange and a Thr75 to Lys exchange as bothTyr56 and Thr75 have been shown to be important for thebinding of signal molecules to LasR [39]. In addition, Pro74, Ala105 and Gly113 were all amino acids that have beendescribed as important for the multimerization and function of LasR (marked by squares in figure 5) [38]. Several of themutations described in this study have been found by otherinvestigators in  lasR   mutants of   P. aeruginosa   obtained under  in vitro evolution experiments (encircled in figure 5) [22,27,38,40]. This Figure 3. Distribution of rhamnolipid production in isolatesbelonging to different clones.  Box and whisker plots represent themedian (thick line inside the box), 10, 25, 75 and 90 centiles of the  invitro  rhamnolipid production ( m g/ml) of   P. aeruginosa  isolates belongingto different clones included in the study. The clones DK1 and DK2 arethe two dominating clones in Denmark. Clone NO is a dominating clonein Norway and the non-clonal isolates have no clonal relationship.Mann-Whitney test was used to investigate the significance of thedifference between the groups.doi:10.1371/journal.pone.0010115.g003 Figure 4. Distribution of rhamnolipid production in isolates with or without mutations in the QS regulatory genes.  Box and whiskerplots represent the median (thick line inside the box), 10, 25, 75 and 90 centiles of the  in vitro  rhamnolipid production ( m g/ml) of   P. aeruginosa  isolates(A) with ( rhlR 2 ) or without mutations ( rhlR + ) in the regulatory gene  rhlR ; (B) with ( lasR 2 ) or without mutations ( lasR + ) in the regulatory gene  lasR Mann-Whitney test was used to investigate the significance of the difference between the groups.doi:10.1371/journal.pone.0010115.g004QS and  P. aeruginosa  in CFPLoS ONE | www.plosone.org 4 April 2010 | Volume 5 | Issue 4 | e10115  reflects a level of similarity between the  in vitro  and  in vivo  bacterialevolution and suggest a possible selective advantage of these kindsof mutations  in vivo .The  lasR   gene could not be amplified in 32 out of 39 DK-2isolates suggesting the deletion of this gene in this particularclone at the time of investigation. Loss of function mutations in Figure 5. Mutations in the  lasR   gene (A) and  rhlR   gene (B) of   P. aeruginosa   isolates from CF patients.  The nucleotide sequence alterationswere identified by alignment with the PAO1 sequence. On top of the PAO1 sequence the nucleotide substitutions, insertions (ins) or deletions ( D ) areindicated. Under the amino acid sequence, frame shifts (fs), stop codons(*) or amino acid deletions ( D ) or changes are indicated in bold. The alterednucleotides are shown in yellow if the mutation was complemented with a plasmid containing the wild-type  lasR  or in grey if not complemented.Amino acid changes that have been previously shown to impair the LasR function [37,38] are marked in squares and amino acid changes that havebeen previously described in  in vitro  studies [22,39,48] are encircled.doi:10.1371/journal.pone.0010115.g005QS and  P. aeruginosa  in CFPLoS ONE | www.plosone.org 5 April 2010 | Volume 5 | Issue 4 | e10115
Search
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x