Pseudomonas aeruginosa - The Journal of Infectious Diseases

Novel Phosphorylcholine-Containing Protein ofPseudomonas aeruginosa Chronic Infection IsolatesInteracts with Airway Epithelial Cells

Mariette Barbier,1 Antonio Oliver,1,3 Jayasimha Rao,4 Sheri L. Hanna,4 Joanna B. Goldberg,4 and Sebastián Albertí1,2

1Institut Universitari d’Investigacions en Ciències de la Salut and 2Área de Microbiología, Departamento de Biología, Universitat de les IllesBalears, and 3Servicio de Microbiología, Hospital Universitario Son Dureta, Palma de Mallorca, Spain; 4University of Virginia, Charlottesville

Pseudomonas aeruginosa undergoes phase variation in the expression of the phosphorylcholine (ChoP) epitope,a structure crucial for the virulence of several respiratory pathogens. In this study, ChoP expression analysiscomparing organisms from acute and chronic infections revealed that expression of ChoP at 37°C was higheramong strains from chronic infections. Coimmunoprecipitation experiments and mass spectrometry analysisdemonstrated that ChoP was on the protein elongation factor Tu. The presence of ChoP at the surface was con-firmed by immunofluorescence and flow cytometry analysis of intact bacteria. Pretreatment of bronchial epithe-lial cells or mice with a platelet-activating factor receptor (PAFR) antagonist reduced adhesion and invasion of theChoP-positive P. aeruginosa isolates. Results of this study suggest that ChoP expression may represent a novelphenotype expressed by the chronic infection isolates that could mediate P. aeruginosa colonization of the epi-thelial airway by means of the interaction with the PAFR.

Phosphorylcholine (ChoP) is a prominent outer surface

component produced by a wide variety of microorgan-

isms, including 2 major human pathogens of the respi-

ratory tract, Streptococcus pneumoniae [1] and Hae-

mophilus influenzae [2]. Although both pathogens

express ChoP on their surface, they differ substantially

in its location. In S. pneumoniae, ChoP is incorporated

into the cell wall–associated teichoic acid and lipotei-

choic acid [1]; in H. influenzae, ChoP is present on the

lipooligosaccharide [2].

In both S. pneumoniae and H. influenzae, ChoP acts as

bacterial ligand for the adherence and invasion of the

airway epithelial cells through platelet-activating factor

receptor (PAFR) [3, 4]. This interaction is critical to the

pathogenesis of infection with both microorganisms,

contributing to the invasive capacity of S. pneumoniae

[3, 5] and to the persistence in the respiratory tract of H.

influenzae [6].

Pseudomonas aeruginosa, a major opportunistic

pathogen, is an important cause of nosocomial pneu-

monia as well as the chief cause of morbidity and mor-

tality in chronic lung infections such as cystic fibrosis

(CF) and bronchiectasis [7–9]. One of the most striking

features of the chronic lung infections by P. aeruginosa is

that the establishment of the infection correlates with

the display of a wide spectrum of colony variants. In

particular, P. aeruginosa isolated from chronic lung in-

fections includes otherwise isogenic variants that can be

mucoid, dwarf, nonmotile, nonflagellated, lipopolysac-

charide deficient, auxotrophic, or resistant to antibiotics

[10 –12]. It is likely that this wide range of phenotypes is

a result of the continuous adaptation of the microorgan-

ism to the changing and heterogeneous conditions of the

deteriorated lung tissue in these patients [13]. In addi-

tion, the high frequency of hypermutable (mutator) P.

aeruginosa strains in the lungs of chronically infected

patients [14, 15] may contribute to this wide spectrum

of phenotypes, particularly resistance to antibiotics

[14 –16].

Received 24 April 2007; accepted 28 August 2007; electronically published 9January 2008.

Potential conflicts of interest: none reported.Presented in part: 17th European Congress of Clinical Microbiology and Infec-

tious Diseases, Munich, 31 March–3 April 2007 (abstract O-296).Financial support: Fundación Respira (PII “Infecciones de la vias áreas bajas”);

Ministerio de Sanidad y Consumo; Instituto de Salud Carlos III-FEDER; SpanishNetwork for the Research in Infectious Diseases (grant REIPI RD06/0008); Fondode Investigaciones Sanitarias (grant PI04 –1829).

Reprints or correspondence: Dr. Sebastian Albertí, Universitat de les IllesBalears, Edificio Guillem Colom, CAMPUS-UIB, Crtra. Valldemossa, km 7.5, Palmade Mallorca 07122, Spain ([email protected]).

The Journal of Infectious Diseases 2008; 197:465–73© 2008 by the Infectious Diseases Society of America. All rights reserved.0022-1899/2008/19703-0020$15.00DOI: 10.1086/525048

M A J O R A R T I C L E

Role of ChoP in P. aeruginosa Infections ● JID 2008:197 (1 February) ● 465

Dow

nloaded from https://academ

ic.oup.com/jid/article/197/3/465/2908688 by guest on 19 January 2022

More than 9 years ago, Weiser et al. [17] demonstrated the

presence of ChoP on an unidentified 43-kDa protein of 12 P.

aeruginosa clinical isolates. It was unclear at that time whether

this modification contributed to the ability of P. aeruginosa to

colonize the airway tract via interaction with PAFR, because the

expression of ChoP was detected only at lower growth temper-

atures (�33.5°C) but not at 37°C [17].

Because P. aeruginosa shares a common ecological niche with

S. pneumoniae and H. influenzae and exhibits a wide range of

phenotypes in the lungs of chronically infected patients, we hy-

pothesized that ChoP, a key factor for the colonization of the

lung by other respiratory pathogens, may be expressed by the

chronic infection isolates of P. aeruginosa at a more physiological

temperature (37°C) as a result of their adaptation to the deteri-

orated lungs.

To test this hypothesis, we monitored P. aeruginosa strains

from acute and chronic infections for expression of ChoP at

37°C, determined the identity of the 43-kDa ChoP-containing

protein of P. aeruginosa, and investigated whether the expression

of the ChoP epitope was involved in the pathogenesis of P.

aeruginosa respiratory infections.

MATERIALS AND METHODS

Bacterial strains. A collection of 92 well-characterized P.

aeruginosa clinical strains were used. All of the strains were iso-

lated from different patients and represented unique clones as

documented by pulsed-field gel electrophoresis [18, 19]. Half of

the isolates (n � 46) were collected from sputum samples from

chronically infected patients (18 with CF, 21 with bronchiectasis,

and 7 with chronic obstructive pulmonary disease). Only pa-

tients with at least 3 years of documented P. aeruginosa coloni-

zation were included. Twenty-three of these 46 isolates exhibited

the mutator phenotype, as shown in the previous studies. The

remaining 46 P. aeruginosa isolates were obtained from respira-

tory samples from different epidemiologically unrelated patients

with nonchronic infection admitted to the intensive care unit.

The laboratory strain P. aeruginosa PAO1 was also used in this

study.

Western blot analysis. Western blot analysis of bacterial

strains was performed using monoclonal antibody (MAb) HAS

(Statens Serum Institute) or MAb TEPC-15 (Sigma), both spe-

cific for the ChoP epitope, as described elsewhere [17]. Filters

were analyzed by densitometry using a Bio Image densitometer

and Whole Band software (version 3.1; Millipore). For the de-

tection of elongation factor Tu (EF-Tu), a goat anti–EF-Tu poly-

clonal antibody was used (provided by D. L. Miller, New York

State Institute for Basic Research in Developmental Disabilities,

Staten Island).

Two-dimensional gel electrophoresis. PAO1 was grown

overnight on trypticase soy agar (TSA) plates and whole-cell Tri-

ton X-114 extractions were done as described elsewhere [20].

Aliquots containing 100 or 150 �g of proteins were solubilized in

rehydration buffer (Bio-Rad) containing 9 mol/L urea, 2% Tri-

ton X-100, 2% Pharmalyte (pH 3–10; Pharmacia), 2%

�-mercaptoethanol, and bromophenol blue; additionally, tribu-

tylphosphine (Sigma) was added (20 mmol/L) [20]. Total pro-

teins were applied on immobilized pH gradient strips (isoelec-

tric point [pI], 3–10 NL; 11 cm) from Bio-Rad. Proteins were

first separated in the first dimension by isoelectric focusing (Pro-

tean IEF Cell; Bio-Rad). For the second dimension, proteins

were resolved on 8%–16% SDS-PAGE gel (Criterion Gel Sys-

tem) with the Dodeca cell system (Bio-Rad). Gels were stained

with Bio-Safe Coomassie stain (Bio-Rad). The cored spots from

Coomassie-stained gels were identified by microcapillary liquid

chromatography mass spectrometry (�LC/MS) and tandem

mass spectrometry (MS/MS), as described below.

For 2-dimensional Western blot analysis after 2-dimensional

gel electrophoresis, gels were transferred onto 0.2-�m nitrocel-

lulose membranes (Bio-Rad) and blocked with “killer filler” re-

agent (1.8 L of 1�PBS containing 200 mL of 0.1 mol/L NaOH,

10 g of casein, and 10 g of bovine serum albumin [BSA], adjusted

to pH 7.4; 0.2 g of phenol red; and 3.6 g of sodium azide) for 1 h

at room temperature and probed with MAb HAS diluted to

1:2500 in killer filler reagent overnight at room temperature.

Immunodetection was performed as described above.

Coimmunoprecipation experiments. PAO1 was grown on

TSA plates at room temperature, and cells were scraped, washed

3 times with ice-cold 1�PBS, lysed by sonication, and treated

with DNAse and RNAse (25 �g of each) for 30 min. The lysate

was then sedimented at 8000 rpm for 15 min at 4°C. The super-

natant was collected in a fresh 1.5-mL microcentrifuge tube, and

coimmunoprecipitation was done by adding 10 �L of human

ChoP–specific antibody (provided by J. N. Weiser, University of

Pennsylvania, Philadelphia) [21] to 500 �L of sample. Samples

were incubated by end-over-end mixing overnight at 4°C. After

incubation, the samples were treated with 100 �L of Protein G

resin beads (Pierce), and the samples were further incubated by

end-over-end mixing for 3– 4 h at 4°C. Finally, bound bead sam-

ples were washed 3 times with 1�PBS to remove all unbound

proteins, and 100 �L of 2� Laemmli sample buffer was added.

Proteins were separated on a 10% Tris-tricine SDS-PAGE gel

and transferred onto 0.2-�m nitrocellulose membranes. Blots

were probed with anti–EF-Tu. Parallel gel was silver stained, and

a selected band was subjected to MS/MS, as described below.

Protein identification by �LC/MS and MS/MS. The se-

lected protein spots were excised from the gels, trypsin digested,

and identified by MS/MS, by the W. M. Keck Biomedical Mass

Spectrometry Laboratory at the University of Virginia Biomed-

ical Research Facility, as described elsewhere [22]. Data were

analyzed using the SEQUEST (Thermo Finnigan) search algo-

rithm against P. aeruginosa genome deposited with the National

Center for Biotechnology Information and Scaffold software

(version 1.6; Proteome Software).

466 ● JID 2008:197 (1 February) ● Barbier et al.

Dow



Immunofluorescence microscopy. Bacteria were spotted to

0.2-�m filter membranes, and nonspecific sites were blocked for

2 h with PBS–1% BSA. Membranes were sequentially incubated

for 1 h with anti-ChoP MAb, TEPC-15, or anti–EF-Tu and then

with a goat anti–mouse IgA–fluorescein isothiocyante conjugate

(Caltag) or a rabbit anti– goat IgG–phycoerythrin (PE) conju-

gate (Sigma), respectively. Membranes were then washed 3 times

with PBS and stained with 4’-6-diamidino-2-phenylindole.

Flow cytometry analysis. Bacterial cells were washed once

with PBS, sequentially incubated with MAb TEPC-15 and rabbit

anti–mouse IgG-PE conjugate, and, finally, resuspended in iso-

tonic buffer. The analyses were done in an Epics XL flow cytom-

eter using Expo32 software. A minimum of 6000 cells were ana-

lyzed in each experiment.

Cell culture assays. Monolayers of human bronchoepithe-

lial immortalized cells (16HBE14o-) were grown to confluence

in Earle’s modified Eagle medium supplemented with 10% fetal

bovine serum plus penicillin and streptomycin in 24-well tissue

culture plates (�5 � 105 cells/well). Before the adhesion and

internalization assays, monolayers were pretreated for 30 min

with the PAFR antagonist 1-O-hexadecyl-2-acetyl-sn-glycerol-

3-phospho-(N,N,N-trimethyl)-hexanolamine (Calbiochem) and

maintained with the PAFR antagonist during the infection. The

adhesion and invasion assays were performed as described else-

where [23].

Murine model of lung infection. Male ICR-CD1 mice

(20 –25 g; Harlan Ibérica) were anesthetized and inoculated in-

tratracheally with 1 � 107 cfu of P. aeruginosa by means of a

blunt-end feeding needle. Bacteria were inoculated with saline

or with saline containing PAFR antagonist. At 24 h, animals

(n � 6 per group) were killed, and lung homogenates were asep-

tically obtained and plated for quantitative bacterial cultures. All

animal experiments were done according to institutional and

national guidelines and were approved by the Experimental An-

imal Committee of the institution.

RESULTS

Expression of ChoP epitope among P. aeruginosa isolates

from acute or chronic infections. Whole-cell lysates from an

equivalent number of cells of 92 genetically unrelated P. aerugi-

nosa clinical isolates (46 from acute infections and 46 from

chronic infections), grown to stationary phase in Luria-Bertani

broth at 22°C and 37°C, were examined by Western blot analysis

with the ChoP-specific MAb TEPC-15. A band of 43 kDa was

present in all clinical isolates of P. aeruginosa examined when

they were grown at 22°C, although the expression was variable

among different strains (data not shown). In most of the isolates,

no detectable expression of the ChoP epitope was apparent in

cells grown at 37°C (figure 1A). However, we detected the ex-

pression of the ChoP epitope at 37°C in some strains (figure 1A).

We refer to this phenotype as ChoP�. To quantify this for each

strain, expression of ChoP epitope at 37°C was normalized for

the expression at 22°C. Overall, the ChoP expression was higher

at 22°C than at 37°C, but the expression of ChoP at 37°C was

higher among strains from chronic infections that among those

from acute infections (P � .004; 2-tailed t test) (figure 1B).

Identification of the 43-kDa ChoP-containing protein. We

subjected whole-cell extracts of PAO1 to 2-dimensional gel analysis.

The presence of the ChoP epitope was detected by Western blot at

�43 kDa and pIs of 4.5 and 5.1, and the corresponding spots were

excised from a parallel gel (figure 2). These proteins were subjected

to MS analysis and were found to correspond to P. aeruginosa EF-

Tu. Antibodies to EF-Tu were used to detect the expression of this

protein in PAO1. As anticipated, EF-Tu was detected as a 43-kDa

band. EF-Tu was expressed at both 22°C and 37°C, indicating that,

unlike the ChoP modification, the expression of EF-Tu was not

temperature dependent in PAO1 (data not shown).

To verify the identification of the 43-kDa ChoP-containing

protein, we conducted coimmunoprecipitation using a human

antibody specific for ChoP. Immunocomplexes were applied to

SDS-PAGE and subjected to Western immunoblotting with

anti–EF-Tu antibodies (figure 3). A faint specific protein of 43

Figure 1. Expression of the phosphorylcholine (ChoP) epitope amongPseudomonas aeruginosa isolates from acute and chronic infections. A,Representative Western blot analysis of ChoP epitope expression in P.aeruginosa acute and chronic airway infection isolates grown at 22°Cand 37°C. B, Expression of ChoP epitope at 37°C/22°C among P. aerugi-nosa isolates from acute and chronic infections. Expression of ChoP wasdetermined by densitometric analysis of the 43-kDa ChoP-containingprotein band detected by Western blot analysis, as shown in panel A.Molecular size marker (in kilodaltons) is indicated on the left. Linesrepresent the means of each group. Data represent at least 3 indepen-dent experiments. P � .004 for the comparison between both groups ofisolates (2-tailed t test).


Dow



kDa was noted in coimmunoprecipitation that was not seen with

secondary antibody alone (compare lanes 1 and 2). A parallel

silver-stained gel revealed a faint protein at this same molecular

weight. This band was subjected to MS/MS, and the major pro-

tein detected was EF-Tu.

Surface location of ChoP epitope. EF-Tu, an essential

component in protein synthesis, is generally considered to be a

cytoplasmic protein. However, ChoP is a cell surface–associated

epitope in a wide range of pathogens. The unexpected location of

the ChoP epitope on EF-Tu led us to examine its surface location

in P. aeruginosa. Anti–ChoP MAb TEPC-15 and anti–EF-Tu

polyclonal antibodies were used to examine intact cells of the

ChoP� chronic infection isolate PAHM4 and PAO1 by immu-

nofluorescence microscopy. Positive immunostaining for both

antibodies was observed on the outer surface of both strains

growth at 22°C (figure 4B and 4C, respectively), whereas at 37°C

only the chronic infection isolate, PAHM4, expressed the ChoP

epitope on the outer surface (figure 4B). Interestingly, EF-Tu

was observed on the outer surface of both strains grown at 37°C

(figure 4C).

These immunofluorescence microscopy results were further

confirmed by flow cytometry analysis of intact P. aeruginosa cells

incubated with the MAb TEPC-15 (figure 4D). Both PAHM4

and PAO1 grown at 22°C had equivalent rates of positive reac-

tion: 21.6% � 4.1% and 24.9% � 3.3% of the bacterial cells

screened reacted with the ChoP-specific antibody, respectively.

In contrast, only PAHM4 grown at 37°C reacted with the specific

antibody (24% � 1.1%), whereas PAO1 grown at 37°C was

negative (2.85% � 0.7%).

Effect of a PAFR antagonist on the adhesion to and inva-

sion of bronchial epithelial cells by P. aeruginosa. To eval-

uate the involvement of PAFR in the adherence to and invasion

of bronchial epithelial cells by P. aeruginosa ChoP� strains at

37°C, we tested the effects of a PAFR antagonist, as described

elsewhere [4]. As shown in figure 5A, treatment of cells with 100

nmol/L PAFR antagonist significantly inhibited the adhesion of

the ChoP� P. aeruginosa strain PAHM4 (P � .003 vs. control

without antagonist; 2-tailed t test). The inhibitory effect of the

PAFR antagonist was also observed in invasion assays (figure

5B). Thus, invasion was reduced by up to 42% and 45% with 50

and 100 nmol/L concentrations of the antagonist, respectively

(P � .03 and P � .01 vs. control without antagonist, respec-

tively; 2-tailed t test).

We then determined the relationship between ChoP expres-

sion at 37°C and the inhibitory effect of the PAFR antagonist.

Bronchial epithelial cells were treated with 50 nmol/L of the

PAFR antagonist, and the invasion of different P. aeruginosa

clinical strains, isolated from either acute or chronic infections

and expressing different amounts of ChoP at 37°C, was deter-

mined. We found that the level of ChoP expression at 37°C was

correlated with greater inhibition of the cell invasion by PAFR

antagonist and showed a positive linear trend (P � .007; linear

regression) (figure 5C).

Effect of PAFR antagonist on P. aeruginosa lung infection

in vivo. To examine the effect of the PAFR antagonist on P.

aeruginosa lung colonization, mice infected with the ChoP�

chronic infection isolate PAHM4 were treated with PAFR antag-

onist or saline. As a control, mice were infected with the chronic

Figure 2. Identification of phosphorylcholine (ChoP)– containing protein from Pseudomonas aeruginosa. Proteins from P. aeruginosa PAO1 wereextracted with Triton X-114 and subjected to 2-dimensional gel electrophoresis and Coomassie staining (left), followed by Western blot analysis withanti–ChoP antibody HAS (right). Spots from gels were excised as indicated by arrows and subjected to tandem mass spectrometry. Sequences of spotscorrespond to P. aeruginosa elongation factor Tu: spot A, 11 peptides, 34% coverage, and 135 of 397 amino acids; spot B, 15 peptides, 54% coverage,and 216 of 397 amino acids. Molecular size markers (in kilodaltons) and isoelectric points (pIs) are indicated on the left and top of the figure, respectively.


Dow



infection isolate PAHM9, which does not express ChoP at 37°C.

After 24 h, the number of P. aeruginosa in lung homogenates was

quantified. Lung colonization by the ChoP� strain PAHM4 was

attenuated at 24 h to almost 50% by PAFR antagonist administered

at a physiologically relevant concentration (0.25 or 0.5 �g/mouse)

(figure 6A). Furthermore, both doses reduced significantly the bac-

terial load in the infected lungs (P � .05 vs. control without an-

tagonist; 2-tailed t test). In contrast, the bacterial load in the

lungs of the animals infected with the chronic infection isolate

that does not express ChoP at 37°C, PAHM9, was unaffected by

the PAFR antagonist (figure 6B), suggesting that the effect of the

antagonist in vivo is specific for the ChoP� strains.

DISCUSSION

Expression of ChoP undergoes phase variation in some micro-

organisms [6, 17, 24]. In P. aeruginosa, expression of ChoP is

modulated by temperature: at 22°C expression is high, whereas

at 37°C expression is negligible. The ability of P. aeruginosa to

down-regulate ChoP expression at 37°C may be crucial in its

capacity to cause invasive acute infections. It is likely that in these

infections the absence of ChoP contributes to the virulence of

the microorganism evading humoral clearance mechanisms,

such as the serum killing mediated by C-reactive protein that

occurs in H. influenzae [6]. Our results are consistent with this

hypothesis, because the ChoP expression at 37°C was undetect-

able in most of the isolates collected from acute infections. In

contrast, we found that the expression of ChoP at 37°C was

higher among the chronic infection isolates. Interestingly,

within the group of the chronic infection isolates, the highest

rate of ChoP� isolates at 37°C was found in the group of the

mutator strains (data not shown). In these strains, which are

defective in mismatch repair systems, the rate of mutation is very

high, and changes allowing for the adaptation to the lung envi-

ronment appear frequently [25]. Altogether, these results sug-

gest that ChoP expression may favor the persistence of the mi-

croorganism in the airway tract, as in H. influenzae, where ChoP

contributes to survival in the respiratory tract [6], and could

represent an important adaptation to the environment of the

deteriorated lung encountered by the microorganism during the

chronic infection process.

Our results, obtained using different approaches, have dem-

onstrated that the 43-kDa ChoP-containing protein is EF-Tu.

Unlike the bacterial structures that have been shown to contain

ChoP in other respiratory pathogens, such as S. pneumoniae [1]

and H. influenzae [2], this epitope is found on a protein in P.

aeruginosa, as well as in pathogenic Neisseria [17]. The presence

of the ChoP epitope on EF-Tu was unexpected, because this

would be the first example of a cytoplasmic protein containing

ChoP and is in contrast to the surface localization of the ChoP

epitope in other pathogens [1, 2, 17]. However, our immunode-

tection assays using intact cells detected the presence of ChoP on

the outer surface of P. aeruginosa. Furthermore, using specific

antibodies against EF-Tu, we also detected the presence of this

protein on the bacterial surface. Surface location of EF-Tu, a

component of the protein synthesis machinery, is unusual

among microorganisms, although it has been reported that un-

der stress conditions EF-Tu becomes membrane associated in

Escherichia coli. Furthermore, the EF-Tu molecule, originally

thought to be restricted to the cytoplasm of bacteria, has also

been shown to be associated with the cell envelopes of many

microorganisms [26 –30]. Interestingly, we found that EF-Tu is

surface exposed at either 22°C or 37°C in all P. aeruginosa strains

tested, suggesting that the incorporation of choline into EF-Tu is

modulated by an unknown mechanism that operates at 37°C in

the chronic infection isolates but not in the acute isolates. In H.

influenzae, the locus involved in the regulation of the ChoP ex-

pression is lic1 [31]. Homologues to the lic1 genes have been

identified in S. pneumoniae [32] and Neisseria [33], but we have

Figure 3. Coimmunoprecipitation of phosphorylcholine (ChoP) and elon-gation factor Tu (EF-Tu) on Pseudomonas aeruginosa. Human antibodies toChoP were used to precipitate P. aeruginosa PAO1. Complexes were sepa-rated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis andblotted with anti–EF-Tu antibodies (lane 1) or secondary antibody only (lane2). Parallel gel was silver stained (lane 3). A selected band (arrow) wasexcised from silver-stained gel and subjected to tandem mass spectrometry.The sequence of the band corresponded to P. aeruginosa EF-Tu: 16 peptides,58% coverage, and 232 of 397 amino acids. H and L indicate the migrationof the heavy and light chains of immunoglobulin. Molecular size markers (inkilodaltons) are indicated on the left.


Dow



not detected homologues to these genes in the genome of P.

aeruginosa. Ongoing investigations will be required to character-

ize the genetic basis for incorporation of choline into EF-Tu in P.

aeruginosa and to understand the different mechanisms that op-

erate in acute and chronic infection isolates.

The ability to adhere to and invade the epithelial lining is

thought to be an important step in the respiratory pathogenesis

of P. aeruginosa. A number of bacterial ligands and host recep-

tors have been associated with adherence to and invasion of ep-

ithelial cells by P. aeruginosa [34, 35]. Our results show that the

adhesion or invasion of the ChoP� P. aeruginosa strains is re-

duced by the PAFR antagonist, suggesting that PAFR is a cellular

receptor for the ChoP� strains. However, the contribution of

this interaction to P. aeruginosa airway chronic pathogenesis is

still unclear. It has been proposed that ChoP is a determinant of

the ability of H. influenzae to colonize and persist within the

nasopharyngeal environment, perhaps by mediating bacterial

adherence to and invasion of the host epithelia. Evidence for this

Figure 4. Surface location of the phosphorylcholine (ChoP) epitope on Pseudomonas aeruginosa. Intact P. aeruginosa cells of the ChoP� chronicinfection isolate PAHM4 and the laboratory strain PAO1 were grown at 22°C or at 37°C and labeled with 4'-6-diamidino-2-phenylindole (A), with theanti-ChoP monoclonal antibody TEPC-15 and goat anti–mouse IgA–fluorescein isothiocyante– conjugated polyclonal antibodies (B), or with anti–elongation factor Tu antibody and rabbit anti– goat IgG–phycoerythrin– conjugated antibodies (C). Cells were analyzed by immunofluorescencemicroscopy (A, B, and C ) or flow cytometry (D). The lines on the histograms plots indicate the gate used for the anti-ChoP� cells. Representative resultsof at least 3 independent experiments are shown.


Dow



hypothesis includes the finding that H. influenzae isolated from

human respiratory secretions are enriched for variants that

would be predicted to express ChoP [6]. On the other hand, it

has been reported that ChoP promotes the establishment of sta-

ble biofilm communities of nontypeable H. influenzae [36]. One

could speculate that the presence of ChoP� variants among the

P. aeruginosa chronic infection isolates might represent evidence

for the role of ChoP in the colonization and persistence in the

airway tract, promoting the biofilm formation in the lungs of

chronically infected patients. It is predicted that there would be a

selection for the ChoP� variants that are able to infect and evade

the early host defense mechanisms in the initial steps of the in-

fection. These ChoP� variants, which initially would represent a

small part of the population, would be able to adhere to the

epithelial cells and promote biofilm formation and chronic per-

sistence in the airway tract. Our results from the in vivo experi-

ments support this hypothesis. PAFR antagonist reduced signif-

icantly the bacterial loads in lung homogenates of mice infected

Figure 5. Effect of a platelet-activating factor receptor (PAFR) antagonist on the adhesion to and invasion of bronchial epithelial cells byPseudomonas aeruginosa. Shown is dose-dependent inhibition of the adherence (A) and invasion (B) of the phosphorylcholine (ChoP)� chronic infectionisolate PAHM4 by the PAFR antagonist. Different physiologically relevant concentrations (50 and 100 nmol/L) of the PAFR antagonist were added tomonolayers 30 min before and during the infection. Adherent or intracellular bacteria were quantified as described in Materials and Methods. Resultsthat are significantly different from those in controls are denoted by an asterisk (2-tailed t test). C, Correlation between PAFR antagonist inhibition ofP. aeruginosa invasion and ChoP expression. Monolayers were treated with the PAFR antagonist (50 nmol/L) as described and were infected withdifferent clinical isolates expressing different amounts of ChoP from acute (white circles) or chronic infections (black circles). The effect of PAFRantagonist on invasion is expressed as the percentage of untreated controls. Data represent at least 3 independent experiments.

Figure 6. Effect of platelet-activating factor receptor (PAFR) antagonist on the Pseudomonas aeruginosa lung infection in vivo. Mice (n � 6 pergroup) were intratracheally inoculated with the phosphorylcholine (ChoP)� strain PAHM4 (A) or the ChoP� strain PAHM9 (B). Both strains wereadministered with saline (control) or saline solutions of PAFR antagonist (0.25 or 0.5 �g), and the nos. of bacterial cells in lung homogenates weredetermined at 24 h. Results that are significantly different from those in untreated controls are denoted by an asterisk (2-tailed t test).


Dow



with a ChoP� chronic infection isolate. This observation was not

due to the potential protective effects of the antagonist during

airway infection, because we did not observe a reduction of the

bacterial loads in lung homogenates of mice infected with a

ChoP� isolate.

In summary, we have shown that another important respira-

tory pathogen uses the ChoP moiety to interact with airway ep-

ithelium via PAFR. In P. aeruginosa, this moiety is particularly

predominant among the chronic infection isolates that may rep-

resent an adaptation of the pathogen to the inflamed airway tract

where the expression of PAFR is elevated, such as those found in

patients with CF, bronchiectasis, or chronic obstructive pulmo-

nary diseases, and may provide a mechanism to promote persis-

tence in the airway tract. This indicates that the ChoP/PAFR

interaction may be of particular importance in the elevated in-

cidence of P. aeruginosa in these patients and may be a new target

for the development of future therapies.

Acknowledgments

We thank Teresa de Francisco (Serveis Cientificotècnis, Universitat de lesIlles Balears, Palma de Mallorca, Spain) and Catalina Crespí (Hospital SonDureta, Palma de Mallorca, Spain) for their assistance in the animal andWestern blot experiments, respectively.

References

1. Mosser JL, Tomasz A. Choline-containing teichoic acid as a structuralcomponent of pneumococcal cell wall and its role in sensitivity of lysisby an autolytic enzyme. J Biol Chem 1970; 245:287–98.

2. Weiser JN, Shchepetov M, Chong STH. Decoration of lipopolysaccha-ride with phosphorylcholine: a phase-variable characteristic of Hae-mophilus influenzae. Infect Immun 1997; 65:943–50.

3. Cundell DR, Gerard NP, Gerard C, Idanpaanheikkila I, Tuomanen EI.Streptococcuspneumoniae anchor to activated human cells by the recep-tor for platelet-activating factor. Nature 1995; 377:435– 8.

4. Swords WE, Buscher BA, Li KVS, et al. Non-typeable Haemophilus in-fluenzae adhere to and invade human bronchial epithelial cells via aninteraction of lipooligosaccharide with the PAF receptor. Mol Microbiol2000; 37:13–27.

5. Rijneveld AW, Weijer S, Florquin S, et al. Improved host defense againstpneumococcal pneumonia in platelet-activating factor receptor– defi-cient mice. J Infect Dis 2004; 189:711– 6.

6. Weiser JN, Pan N, McGowan KL, Musher D, Martin A, Richards J. Phos-phorylcholine on the lipopolysaccharide of Haemophilus influenzaecontributes to persistence in the respiratory tract and sensitivity to se-rum killing mediated by C-reactive protein. J Exp Med 1998; 187:631–40.

7. Vincent JL. Nosocomial infections in adult intensive-care units. Lancet2003; 361:2068 –77.

8. Lyczak JB, Cannon CL, Pier GB. Lung infections associated with cysticfibrosis. Clin Microbiol Rev 2002; 15:194 –222.

9. Evans SA, Turner SM, Bosch BJ, Hardy CC, Woodhead MA. Lung func-tion in bronchiectasis: the influence of Pseudomonas aeruginosa. EurRespir J 1996; 9:1601– 4.

10. Mahenthiralingam E, Campbell ME, Speert DP. Nonmotility and phag-ocytic resistance of Pseudomonas aeruginosa isolates from chronicallycolonized patients with cystic fibrosis. Infect Immun 1994; 62:596 – 605.

11. Speert D, Campbell M, Puterman ML, et al. A multicenter comparisonof methods for typing strains of Pseudomonas aeruginosa predominantlyfrom patients with cystic fibrosis. J Infect Dis 1994; 169:134 – 42.

12. Govan JRW, Deretic V. Microbial pathogenesis in cystic fibrosis: mu-coid Pseudomonas aeruginosa and Burkholderia cepacia. Microbiol Rev1996; 60:539 –74.

13. Nguyen D, Singh PK. Evolving stealth: genetic adaptation of Pseudomo-nas aeruginosa during cystic fibrosis infections. Proc Natl Acad Sci USA2006; 103:8305– 6.

14. Oliver A, Canton R, Campo P, Baquero F, Blazquez J. High frequency ofhypermutable Pseudomonas aeruginosa in cystic fibrosis lung infection.Science 2000; 288:1251–3.

15. Macia MD, Blanquer D, Togores B, Sauleda J, Perez JL, Oliver A. Hy-permutation is a key factor in development of multiple-antimicrobialresistance in Pseudomonas aeruginosa strains causing chronic lung infec-tions. Antimicrob Agents Chemother 2005; 49:3382– 6.

16. Hogardt M, Hoboth C, Schmoldt S, Henke C, Bader L, Heesemann J.Stage-specific adaptation of hypermutable Pseudomonas aeruginosa iso-lates during chronic pulmonary infection in patients with cystic fibrosis.J Infect Dis 2007; 195:70 – 80.

17. Weiser JN, Goldberg JB, Pan N, Wilson L, Virji M. The phosphorylcho-line epitope undergoes phase variation on a 43-kilodalton protein inPseudomonas aeruginosa and on pili of Neisseria meningitidis and Neis-seria gonorrhoeae. Infect Immun 1998; 66:4263–7.

18. Gutiérrez O, Juan C, Pérez JL, Oliver A. Lack of association betweenhypermutation and antibiotic resistance development in Pseudomonasaeruginosa isolates from intensive care unit patients. Antimicrob AgentsChemother 2004; 48:3573–5.

19. Macia MD, Borrell N, Perez JL, Oliver A. Detection and susceptibilitytesting of hypermutable Pseudomonas aeruginosa strains with the Etestand disk diffusion. Antimicrob Agents Chemother 2004; 48:2665–72.

20. Bordier C. Phase separation of integral membrane proteins in TritonX-114 solution. J Biol Chem 1981; 256:1604 –7.

21. Goldenberg HB, McCool TL, Weiser JN. Cross-reactivity of human im-munoglobulin G2 recognizing phosphorylcholine and evidence for pro-tection against major bacterial pathogens of the human respiratory tract.J Infect Dis 2004; 190:1254 – 63.

22. Hanna SL, Sherman NE, Kinter MT, Goldberg JB. Comparison of pro-teins expressed by Pseudomonas aeruginosa strains representing initialand chronic isolates from a cystic fibrosis patient: an analysis by 2-D gelelectrophoresis and capillary column liquid chromatography-tandemmass spectrometry. Microbiology 2000; 146:2495–508.

23. Fleiszig SMJ, Evans DJ, Do N, Vallas V, Shin S, Mostov KE. Epithelial cellpolarity affects susceptibility to Pseudomonas aeruginosa invasion andcytotoxicity. Infect Immun 1997; 65:2861–7.

24. Kim JO, Weiser JN. Association of intrastrain phase variation in quan-tity of capsular polysaccharide and teichoic acid with the virulence ofStreptococcus pneumoniae. J Infect Dis 1998; 177:368 –7.

25. Oliver A, Baquero F, Blazquez J. The mismatch repair system (mutS,mutL and uvrD genes) in Pseudomonas aeruginosa: molecular character-ization of naturally occurring mutants. Mol Microbiol 2002; 43:1641–50.

26. Granato D, Bergonzelli GE, Pridmore RD, Marvin L, Rouvet M,Corthesy-Theulaz IE. Cell surface-associated elongation factor Tu me-diates the attachment of Lactobacillus johnsonii NCC533 (La1) to humanintestinal cells and mucins. Infect Immun 2004; 72:2160 –9.

27. Jacobson GR, Rosenbusch JP. Abundance and membrane association ofelongation-factor Tu in Escherichia coli. Nature 1976; 261:23– 6.

28. Dallo SF, Kannan TR, Blaylock MW, Baseman JB. Elongation factor Tuand E1 beta subunit of pyruvate dehydrogenase complex act as fibronec-tin binding proteins in Mycoplasma pneumoniae. Mol Microbiol 2002;46:1041–51.

29. Harding SV, Sarkar-Tyson M, Smither SJ, et al. The identification ofsurface proteins of Burkholderia pseudomallei. Vaccine 2007; 25:2664 –72.

30. Marques MAM, Chitale S, Brennan PJ, Pessolani MCV. Mapping andidentification of the major cell wall-associated components of Mycobac-terium leprae. Infect Immun 1998; 66:2625–31.

31. Weiser JN, Lindberg AA, Manning EJ, Hansen EJ, Moxon ER.Identification of a chromosomal locus for expression of lipopolysac-


Dow



charide epitopes in Hemophilus-influenzae. Infect Immun 1989; 57:3045–52.

32. Zhang JR, Idanpaan-Heikkila I, Fischer W, Tuomanen EI. Pneumococ-cal licD2 gene is involved in phosphorylcholine metabolism. Mol Mi-crobiol 1999; 31:1477– 88.

33. Serino L, Virji M. Phosphorylcholine decoration of lipopolysaccharidedifferentiates commensal Neisseriae from pathogenic strains: identifica-tion of licA-type genes in commensal Neisseriae. Mol Microbiol 2000;35:1550 –9.

34. Lyczak JB, Cannon CL, Pier GB. Establishment of Pseudomonas aerugi-

nosa infection: lessons from a versatile opportunist. Microbes Infect2000; 2:1051– 60.

35. Sadikot RT, Blackwell TS, Christman JW, Prince AS. Pathogen-host in-teractions in Pseudomonas aeruginosa pneumonia. Am J Resp Crit CareMed 2005; 171:1209 –13.

36. Hong WZ, Mason K, Jurcisek J, Novotny L, Bakaletz LO, Swords WE.Phosphorylcholine decreases early inflammation and promotes the es-tablishment of stable biofilm communities of nontypeable Haemophilusinfluenzae strain 86 – 028NP in a chinchilla model of otitis media. InfectImmun 2007; 75:958 – 65.


Dow



Pseudomonas aeruginosa - The Journal of Infectious Diseases

Documents

Transcript of Pseudomonas aeruginosa - The Journal of Infectious Diseases