INFECTION AND IMMUNITY, June 2011, p. 2168–2181 0019-9567/11/$12.00 doi:10.1128/IAI.01304-10 Copyright © 2011, American Society for Microbiology. All Rights Reserved.

Vol. 79, No. 6

Lcl of Legionella pneumophila Is an Immunogenic GAG Binding Adhesin That Promotes Interactions with Lung Epithelial Cells and Plays a Crucial Role in Biofilm Formation䌤 Carla Duncan,1# Akriti Prashar,1,2# Jannice So,1,3 Patrick Tang,1 Donald E. Low,1,3,4 Mauricio Terebiznik,2 and Cyril Guyard1,3,4*

Received 10 December 2010/Returned for modification 20 January 2011/Accepted 11 March 2011

Legionellosis is mostly caused by Legionella pneumophila and is defined by a severe respiratory illness with a case fatality rate ranging from 5 to 80%. In vitro and in vivo, interactions of L. pneumophila with lung epithelial cells are mediated by the sulfated glycosaminoglycans (GAGs) of the host extracellular matrix. In this study, we have identified several Legionella heparin binding proteins. We have shown that one of these proteins, designated Lcl, is a polymorphic adhesin of L. pneumophila that is produced during legionellosis. Homologues of Lcl are ubiquitous in L. pneumophila serogroups but are undetected in other Legionella species. Recombinant Lcl binds to GAGs, and a ⌬lpg2644 mutant demonstrated reduced binding to GAGs and human lung epithelial cells. Importantly, we showed that the ⌬lpg2644 strain is dramatically impaired in biofilm formation. These data delineate the role of Lcl in the GAG binding properties of L. pneumophila and provide molecular evidence regarding its role in L. pneumophila adherence and biofilm formation. binds to lung epithelial cells (17), to extracellular matrix (87), and to alveolar macrophages (60). Intracellular replication of Lp1 within macrophages is the focus of multiple ongoing studies (55, 75). In contrast, infection of epithelial cells by Lp1 has received relatively little attention, although epithelial cells represent more than 95% of the surface area of the alveoli and constitute a niche for the replication and dissemination of Legionella in the lungs (16, 17, 29, 30, 54). Sulfated glycosaminoglycans (GAGs), which are expressed by all nucleated mammalian cells, are important docking platforms for bacteria (70). Indeed, exogenous heparin (a prototypical GAG) has been shown to specifically inhibit the binding of Lp1 to alveolar epithelial cells (81, 89). Furthermore, in a mouse infection model, preincubation of Lp1 with heparin decreased the mortality rate, protected the alveolar-capillary barrier, prevented systemic bacterial dissemination, and stimulated Th1 cytokine production (1). Taken together, these results suggest that Lp1 produces GAG binding adhesins that are essential to the pathogenesis of Legionella infections. Few mediators of Legionella adherence to host cells, such as type IV pilus, integrin analogue LaiA, Hsp60, structural toxin RtxA, and Lcl, have been reported (13, 18, 31, 77, 86). Nevertheless, none of these mediators has been tested for interaction with host cell GAGs. In this work, we have identified several heparin binding proteins of Lp1. We have obtained substantial genetic and biochemical data demonstrating that one of the identified proteins, Lcl, is a GAG binding adhesin specific to the pneumophila species. Moreover, we have established that an isogenic lpg2644 mutant is impaired in binding to GAGs and human lung epithelial cells and in biofilm formation. Importantly, we have demonstrated that Lcl is an immunogenic protein during

In the United States, it is estimated that 8,000 to 18,000 people contract Legionnaires’ disease every year (26). The severity of this disease ranges from a mild respiratory illness to a rapidly fatal pneumonia. Death occurs through progressive pneumonia with respiratory failure and/or shock and multiorgan failure (78). The case fatality rate of legionellosis ranges between 40 and 80% in untreated immunosuppressed patients but can be reduced to 5 to 30% with appropriate case management (2, 4). The causative agents of legionellosis are Gramnegative, non-spore-forming bacilli of the Legionella genus. Legionellosis is acquired by inhaling contaminated airborne water droplets (26). Legionella bacteria are found worldwide and can be detected in up to 80% of freshwater sites (27). While some Legionella species are frequently reported in cases of legionellosis, many others are only isolated from the environment. Among the 53 species of Legionella, L. pneumophila is the major cause of outbreaks (91.5%) and serogroup 1 (Lp1) is the predominant serotype (84.2%) (44, 90). To successfully establish infection, pathogenic bacteria depend on an arsenal of virulence factors which facilitate host colonization. The ability of bacteria to adhere to host cells and extracellular matrix is essential to escape mechanical clearance and to establish the focal point of an infection from which dissemination will occur (51, 63). Adhesins, involved in host cell binding, can also participate in biofilm formation, in the triggering of host cell signaling, and in cellular invasion (6, 38). During the early stages of the pulmonary infection, Lp1 * Corresponding author. Mailing address: Ontario Agency for Health Protection and Promotion, 81 Resources Road, Toronto, ON M9P 3T1, Canada. Phone: (416) 427-9060. Fax: (416) 235-6281. E-mail: cyril.guyard @oahpp.ca. # These authors participated equally in this work. 䌤 Published ahead of print on 21 March 2011. 2168

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

Ontario Agency for Health Protection and Promotion (OAHPP), Toronto, Ontario, Canada M9P 3T11; Department of Biological Sciences, University of Toronto at Scarborough, Toronto, Ontario, Canada M1C 1A42; Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada M5S 1A83; and Mount Sinai Hospital, Toronto, Ontario, Canada M5G 1X54

VOL. 79, 2011

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2169

TABLE 1. Legionella species and isolates used in experiments Code

a b

L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L. L.

pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila erythra feeleii erythra rubrilucens maceachernii anisa bozemanii bozemanii pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila pneumophila

Sga

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 3 4 5 6 8 10 12 1-9b NA NA NA NA NA NA 2 1 1 1 1 1 1 1 1 1

Designation/plasmid(s)

2006-LU2004 2006-LU1536 2006-LR1063 2002-LU0809 2003-LR0347 2006-LR1043 2006-LR0726 2005-LR1669 2005-LR1423 2005-LR1350 2005-LR1318 2005-LR1294 2005-LU1410 2004-LR1124 2004-LR0684 2003-LR1171 2003-LR1005 2002-LR1484 2002-LU1918 2002-LU1560 2002-LR0905 Lp02 LR1201 LU2270 LR0967 LR0929 LU2015 LR0162 LU0895 LR0933 LR1515 LR1359 LR1317

Lp02/pBH6119 Lp02 ⌬lpg2644/pBH6119 Lp02/pBH6119 Lp02/plpg2644 Lp02/pBH6119-IcmRp Lp02/plpg2644 Lp02/pBH6119-IcmRp Lp02 ⌬lpg2644/pBH6119-IcmRp

Source or reference

Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study 5 Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Human clinical, this study Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this Anthropogenic water, this This study This study This study This study This study This study This study This study

study study study study study study study study study

Sg, serogroup; NA, not applicable. Cross-reactive with serogroups 1 and 9.

legionellosis. Altogether, our data suggest that Lcl is an adhesin involved in Lp1 pathogenesis. MATERIALS AND METHODS Chemicals, bacterial strains and cultivation. All chemicals and antibodies were purchased from Sigma-Aldrich (Oakville, ON, Canada) unless otherwise noted. Restriction and cloning enzymes were purchased from New England BioLabs (Pickering, ON, Canada) and used according to the manufacturer’s recommendations. Legionella isolates and plasmid vectors used in this study are listed in Table 1. All Legionella isolates were cultured in buffered charcoal-yeast extract (BCYE) agar at 37°C and 5% CO2 and/or buffered yeast extract (BYE) broth at 37°C with shaking at 100 rpm (25). Cultures of L. pneumophila Lp02 strains were supplemented with 100 ␮g/ml thymidine (5). To obtain late-exponential-phase bacteria (optical density at 600 nm [OD600] of 3.0 to 3.5), overnight precultures of strain Lp02 were adjusted to an OD600 of 0.05 in BYE broth and were incubated at 37°C and 100 rpm. Once required ODs were reached, aliquots

were processed for analysis. Escherichia coli strains and plasmids are listed in Table 2. All strains were cultured in Luria-Bertani medium or RM medium (Invitrogen, Burlington, ON, Canada) for protein purification, and when appropriate, antibiotics were added to the medium at concentrations of 50 ␮g/ml kanamycin or 100 ␮g/ml carbenicillin. Purification and identification of heparin binding proteins. Late-exponentialphase Lp02 culture supernatant (Table 1, code 1) or E. coli lysate was passed through a heparin-agarose chromatography column (HiTrap heparin HP; GE Healthcare, Baie d’Urfe, QC, Canada) and eluted with a 0 to 500 mM NaCl gradient using an AKTA FPLC system (GE Healthcare). Eluted proteins were pooled, concentrated with Millipore Amicon Ultra-15 5K NMWL columns (Fisher Scientific, Ottawa, ON, Canada), and separated on 4 to 15% linear gradient Tris-HCl SDS-polyacrylamide gels (Bio-Rad, Mississauga, ON, Canada). Prior to mass spectrometry (MS) analysis of Legionella-secreted heparin binding proteins, bands were cut from the gel and in-gel digested according to the published procedure, with minor modifications (74). The reduction time using dithiothreitol was reduced to 45 min, and alkylation time using iodoacet-

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

a b c d e f g h i j k l m n o p q r s t u 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26

Species

2170

DUNCAN ET AL.

INFECT. IMMUN.

TABLE 2. E. coli strains and plasmids used in this study Strain designation

Plasmid(s)

Source or reference

2 43 51 15 16 45 18 25 29 57 61 88

TOP10 TOP10 TOP10 DH5␣ ␭ pir DH5␣ ␭ pir DH5␣ ␭ pir CR19 (Tra⫹) TOP10 TOP10 LMG-194 LMG-194 DH5␣ ␭ pir

None pBAD-HisB pBAD-HisB-lpg2644 None pSR47S pSR47S-⌬lpg2644 None pBH6119 pBH6119-icmRp None pBAD-HisB-lpg2644⌬repeats pSR47S-clpg2644

Invitrogen Invitrogen This study Stratagene 53 This study 53 24 This study Invitrogen This study This study

amide was reduced to 30 min. Peptides were extracted from the gel pieces with 5% (vol/vol) formic acid twice and then with 5% (vol/vol) formic acid in 30% (vol/vol) acetonitrile. Extracted peptides were subjected to liquid chromatography-tandem MS (LC–MS-MS) with back-to-back collision-induced dissociation and electron transfer dissociation (Agilent 1100 HPLC-Chip and 6340 ion trap MSD system; Agilent Technologies, Mississauga, ON, Canada). The HPLC-Chip system (Agilent Technologies) was used to separate peptides with a gradient of 3, 35, and 80% of 0.2% (vol/vol) formic acid and 100% acetonitrile. The MS scan range was from 300 to 1,300 m/z. Thirty-second dynamic exclusion was applied to the precursor previously selected for MS-MS twice. Raw MS-MS data files were searched against NCBInr using Spectrum Mill MS Proteomics Workbench (Agilent Technologies). E. coli lysates were prepared by resuspending cell pellets in binding buffer (0.16 M phosphate buffer, pH 7.4, 4 M NaCl, 10 mM imidazole, 2% Tween 20), sonicating at 6 W three times for 20 s at output setting 0.5 (Misonix S3000; VWR, Mississauga, ON, Canada), treating with 10 ␮g/ml DNase for 30 min at room temperature (RT), spinning at 5,000 ⫻ g for 15 min, and filtering the supernatant through a 0.45-␮m filter.

TABLE 3. Oligonucleotides used for PCR, qRT-PCR, sequencing, and cloning Code

Primer/probe

1 2 3 4 5

lpg2644 lpg2644 lpg2644 lpg2644 lpg2644

F R seq1 seq2 seq3

6 7 8 9 10

lpg2644 lpg2644 lpg2644 lpg2644 lpg2644

XhoI F EcoRI R ⌬repeat F ⌬repeat R probe F

11 12 13 14 15

lpg2644 probe R mip probe F mip probe R upstr BamHI F upstr XbaI R

16

dwnstr XbaI F

17 18 19 20

dwnstr BamHIR icmRP EcoRI F icmRP BamHI R lpg2644 XbaI F

21 22 23 24

lpg2644 SphI R gyrA F gyrA R gyrA probe

DNA amplified

lpg2644

Sequence (5⬘ to 3⬘)

AGACACGTGTTGAATCCACT CACCAAAAGCAATCCGGCCTCGCA CACCGCAGAAACGCTTGCAA TCCCCTCTGGACCTCAAGGA CACATGGTATTGAAATCGAG AGCTCGAGCAATCCGGCCTCGCAAGCC CGGAATTCCGGGTTGCGAGAGTTGGCTA GATGACGGCCAAGGTGTGCC TTGAGGTCCTTGAGGTCCAG TTGCCAAATCAAATGACCTC

mip Upstream lpg2644 Downstream lpg2644 icmR promoter lpg2644 gyrA

GGGAGTCTCCAGTCAGAATA GGATAAGTTGTCTTATAGCATTGG GGCTTCCCCTTTTACTTTATTTTC GGCGCGGATCCACTTATTCTAAAAATAGTTGTGCTT GCTAGTCTAGAACTTTATGCTCACTTGTCTC GCTAGTCTAGATTAAGCATGGCAAAACTTCA GGCGCGGATCCGATGGATTTTGCAGTGAACAATTTA GGAATTCGCCCAAGCCGTGATTATACT CGGGATCCTATTACCACTCCTGAGCTAA GGTCATCTAGAGAAATAAAGAATGATACATCGA GTGAGCGCATGCGCAAAGCGAATTTATGAACA GGCGGGCAAGGTGTTATTT GCAAGGAGCGGACCACTTT VIC-CATTTCGTTCAGTAACCTG-MGBNFQ

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

MR strain code

General DNA techniques. Total genomic DNAs were purified using a QIAamp DNA minikit (Qiagen, Mississauga, ON, Canada), and plasmid DNA was purified using a QIAprep spin miniprep kit (Qiagen). DNA was quantified by UV spectrophotometry, and 10 ng was used as a template for each PCR. Taq DNA polymerase, deoxynucleoside triphosphates, and buffer were used according to the recommendations of the manufacturer (Invitrogen, Burlington, ON, Canada). PCR amplifications for cloning were performed using Platinum Taq DNA polymerase high fidelity (Invitrogen). A list of primers used can be found in Table 3. Sequencing reactions were performed using a BigDye terminator cycle sequencing kit, version 3.1, purified using a BigDye X terminator purification kit, and run on a 3130xl genetic analyzer (Applied Biosystems, Streetsville, ON, Canada). All clones were verified by sequencing. Production and purification of His-Tag fusion proteins and polyclonal antibodies. The lpg2644 gene was amplified from L. pneumophila genomic DNA using primers 6 and 7 (Table 3). The PCR products were cloned into the pBAD-HisB vector according to the instructions of the manufacturer (Invitrogen) to obtain pBAD-HisB-lpg2644 (Table 2, MR51). E. coli TOP10 clones were tested for the expression of recombinant proteins after induction with 0.2% L-arabinose at 37°C for 4 h. His-tagged fusion proteins were purified under native conditions with a nickel-Sepharose high-performance chromatography column (HisTrap HP column) according to the instructions of the manufacturer (GE Healthcare). A recombinant Lcl (rLcl) protein lacking the collagen-like tandem repeats (Lcl ⌬repeats) was created using pBAD-HisB-lpg2644 as the template and primers 8 and 9 (Table 3). This was then blunt end cloned and transformed into E. coli LMG-194 (Table 2, MR61). Purified fusion proteins were dialyzed in phosphate-buffered saline (PBS) and suspended in a final volume of 10 ml of Freund’s complete adjuvant, and 500 ␮g was used to immunize rabbits by subcutaneous inoculation. Three boosts of 250 ␮g were carried out at 2-week intervals with recombinant protein suspended in incomplete Freund’s adjuvant. Three weeks after the last boost, the rabbits were bled and serum samples were collected. Immunofluorescence assays were performed according to standard procedures (8, 57). SDS-PAGE and immunoblot analysis. SDS-PAGE was performed according to the method of Laemmli (41). Immunoblotting was performed as described by Towbin et al. (85). Recombinant proteins were detected with anti-His mouse monoclonal antibody (Invitrogen) according to the instructions of the manufacturer. To detect specific Legionella proteins, cell lysates were prepared by heating

VOL. 79, 2011

2171

mix (Applied Biosystems) using 400 nM each specific primer and 200 nM probe. Amplification and detection of specific products were performed with an ABI Prism 7900 detection system (Applied Biosystems), with samples first incubated at 50°C for 2 min and then at 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and 60°C for 1 min. Cell culture and adherence assays. NCI-H292 (ATCC, CRL-1848) cells from human lung mucoepidermoid carcinoma were grown at 37°C with 5% CO2 in RPMI 1640 medium supplemented with 2 mM L-glutamine and 10% heatinactivated fetal bovine serum (RPMI⫹) (Invitrogen). For adherence assays, 96-well or 24-well flat bottom plates (Costar, Fisher Scientific) were seeded with 5 ⫻ 104 or 2.5 ⫻ 105 cells/well and monolayers were used after 18 h. Bacterial suspensions were prepared from 16 h broth cultures, washed twice with PBS, adjusted to OD600 of 2 (approximately 1.5 ⫻ 109 CFU/ml) and diluted 1:10 in culture media (RPMI ⫹). After wells were washed twice with PBS, cells were infected at a multiplicity of infection (MOI) of 300 bacteria per cell for 3 h at 37°C with 5% CO2. The monolayers were then washed three times with PBS and lysed with 0.1% Triton X-100 in PBS for 30 min at 37°C. Serial dilutions of the lysates were plated for counting viable bacteria. Assays were performed in triplicate and were repeated up to 3 times. Microscopic assessments of bacterial adhesion were performed in a similar manner with the exception that NCI-H292 cells were seeded onto glass coverslips in 24-well flat-bottom plates. Following 1 h of incubation with bacteria, the monolayers were washed three times with PBS and fixed in 4% paraformaldehyde (PFA) for 45 min. Extracellular bacteria were immunolabeled using anti-Lp1 antibodies (80), followed by Alexa Fluor 647 (Invitrogen) secondary antibody. Cells were permeabilized with 0.1% Triton X-100 for 30 min. Internalized bacteria were labeled using Alexa Fluor 555, and cells were stained using green phalloidin (Invitrogen). Attachment was quantified using a Leica DMI 600B microscope and Volocity software (Perkin Elmer). Biofilm formation and quantification. Biofilm assays using borosilicate glass tubes were performed according to the method of Piao et al. (62). For assays on polystyrene 96-well plate (Costar), Legionella strains were grown for 30 h in broth and diluted in fresh medium to give a final OD at 600 nm of 0.2 in a volume of 200 ␮l per well. For inhibition assays, the medium was supplemented with concentrations of glucose, mannose, chondroitin sulfate-C, or fucoidan ranging from 0 to 0.28 mg/ml. Plates were incubated at 37°C and 5% CO2 for 2 and 6 days. Biofilms were stained with 40 ␮l 0.25% crystal violet in each well for 15 min and washed three times in deionized water, and the crystal violet stain was solubilized in 95% ethanol. Absorbance was read at 600 nm. Three independent experiments were performed using 6 replicates. For microscopic examination of biofilms, cultures were prepared according to the procedure described above and added to a Lab-Tek II chambered coverglass (VWR). To obtain fluorescent Legionella bacteria, the Lp02 icmR promoter was cloned upstream of the green fluorescent protein (GFP) gene into pBH6119 (Table 3, primers 18 and 19). The resulting plasmid, pBH6119-IcmRp (Table 2, MR29), was transformed into Lp02 and Lp02 ⌬lpg2644 (Table 1, codes 25 and 26). After 2 to 6 days of incubation, chambered wells were washed three times with PBS. Confocal images were obtained with a Nikon Eclipse TE2000EZ inverted microscope (60⫻ Plan APO oil immersion differential interference contrast [DIC] N2 objective and Spectra-Physics 488-nm laser). Nucleotide sequence accession numbers. The GenBank accession numbers of lpg2644 homologues from clinical isolates described in this study are as follows: for L. pneumophila 2006-LU2004, GQ504706; for L. pneumophila 2006-LU1536, GQ504707; for L. pneumophila 2006-LR1063, GQ504708; for L. pneumophila 2002-LU0809, GQ504709; for L. pneumophila 2003-LR0347, GQ504710; and for L. pneumophila 2006-LR1043, GQ504711.

RESULTS Lcl is a heparin binding protein of L. pneumophila. Surfaceexposed heparin binding adhesins of pulmonary pathogens such as Bordetella pertussis or Mycobacterium tuberculosis were successfully affinity purified from spent supernatants in the past (48, 50). Using a similar experimental strategy, we passed a late-exponential culture supernatant from Lp02 (Table 1, code 1) over a heparin-agarose chromatography column and eluted bound proteins with a NaCl gradient (49, 52, 59). SDSpolyacrylamide gels of eluted fractions revealed a total of 7 detectable species that were subsequently submitted to mass spectrometry analysis (Fig. 1). Based on MS-MS scores, protein coverage, and spectral intensity, we could identify a total

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

(95°C for 10 min) washed broth cultures (108 cells/ml) in PBS mixed with an equal volume of Laemmli loading buffer containing 5% 2-mercaptoethanol. Bound Lcl (1:20,000) antibodies were detected with peroxidase-linked anti-rabbit IgG (1:160,000). ELISAs. All enzyme-linked immunosorbent assays (ELISAs) were performed in Immulon 2-HB 96-well plates (VWR). Plates were coated overnight at 37°C unless otherwise stated. Pierce TMB (3,3⬘,5,5⬘-tetramethyl benzidine) ELISA substrate (Fisher Scientific) was used as the substrate for HRP (horseradish peroxidase), the reactions were stopped after 30 min at RT with 2 M sulfuric acid, and the absorbance was determined at 450 nm using a BioTek Powerwave XS plate reader. All samples were tested in triplicate. Heparin was biotinylated as described previously (58, 59). To examine the immunogenicity of rLcl, plates were coated with 2.5 ␮g of protein/well in PBS and blocked for 1 h at 37°C with 5% bovine serum albumin (BSA) in PBS–0.05% Tween 20 (PBST). Human sera from 10 convalescent legionellosis patients and 2 nonimmune controls were diluted 1:10,000 in PBS, and secondary anti-human IgG-HRP antibody (GE Healthcare) was diluted 1:5,000. Primary and secondary antibodies were both incubated for 1 h at 37°C. To monitor the binding of heparin sulfate to rLcl, plates were coated with 2.5 ␮g protein/well in carbonate buffer (pH 9.0). The wells were blocked at 37°C and washed with 1% BSA–PBST. Coated wells were incubated with 20 ␮g of biotinylated heparin for 2 h at 37°C, and unbound heparin was removed with 1% BSA–PBST–150 mM NaCl followed by a 1-h incubation at room temperature with 0.1 ␮g of streptavidin-HRP (Pierce, Fisher Scientific) per well. To examine the binding of rLcl to GAGs, heparin, heparan sulfate, dextran sulfate, chondroitin sulfate A, B, and C, fucoidan, and dextran were used to coat plates (59). Recombinant Lcl (0.2 ␮g/ml in 1% BSA–PBS) was added to each well and incubated at room temperature for 1 h. Wells were washed with PBS–0.5% Tween 20 and probed with rabbit anti-His-rLcl diluted 1:10,000 in 1% BSA–PBS, followed by anti-rabbit antibody conjugated to horseradish peroxidase diluted 1:40,000 in 1% BSA–PBS. Southern blot analyses. Genomic DNA samples from Legionella isolates (Table 1, codes 2 to 18) were examined for lpg2644 and mip homologues by Southern blot analysis under low-stringency conditions (72, 76). The hybridization probes were designed from conserved regions of the respective genes and amplified from genomic DNA of L. pneumophila Lp02 with a PCR digoxigenin (DIG) probe synthesis kit (Roche Applied Science, Laval, QC, Canada) using primers 10 and 11 and primers 12 and 13 (Table 3). Generation of lpg2644 knockout and complementation. An in-frame deletion of lpg2644 was created using suicide vector pSR47s (Kanr sacB) containing genomic regions flanking lpg2644. Flanking regions were amplified from genomic DNA using primers 14 and 15 and 16 and 17 (Table 3). These fragments were digested with XbaI, ligated, and reamplified using primers 14 and 17. This fragment was then digested with BamHI and cloned into pSR47s. The resulting plasmid, pSR47S-⌬lpg2644 (Table 2, MR45), was transformed into E. coli DH5␣ ␭ pir. Primary integration into the Lp02 chromosome via homologous recombination was selected for by kanamycin resistance. The second recombination event (excision of plasmid) was selected using sucrose resistance. A chromosomal complementation of lpg2644 was created with the same technique, using primers 14 and 17 to amplify lpg2644 and flanking regions. This fragment, containing a unique intact open reading frame (lpg2644), was cloned into pSR47s to obtain pSR47S-clpg2644 (Table 2, MR88). To construct a plasmid for complementation of lpg2644 knockout and overexpression of rLcl, the icmR promoter of L. pneumophila Lp02 was PCR amplified from genomic DNA using primers 18 and 19 (Table 3), digested with EcoRI and BamHI, and cloned into the pBH6119 vector (MR25, Table 2) by using a quick ligation kit (New England BioLabs). ORF lpg2644 of L. pneumophila Lp02 was amplified using primers 20 and 21 (Table 3) and TA cloned into pCR2.1 (Invitrogen). This was then digested with XbaI and SphI and cloned downstream of the icmR promoter in pBH6119. The resulting plasmid (plpg2644) was transformed into Lp02 and Lp02 ⌬lpg2644. Deletion and complementation of lpg2644 were confirmed by Southern blotting, anti-Lcl immunoblot analyses, and sequencing. Quantification of Legionella bound to coated GAG using qPCR. To evaluate the effect of deleting lpg2644 on Lp02 GAG binding properties, 5 ␮g of heparan sulfate and fucoidan (Sigma) were used to coat 96-well BD heparin binding plates according to the manufacturer’s recommendations (BD Biosciences, Mississauga, ON, Canada). Coated wells were incubated for 1 h at 37°C and 5% CO2 with 100 ␮l of Legionella suspension (OD600 of 2) in PBS. After three washes in PBS, DNA from adhering cells was purified directly from each well with a DNeasy 96 blood and tissue kit according to the manufacturer’s instructions (Qiagen). Genome copy numbers present in each well were next quantified by quantitative PCR (qPCR) using oligonucleotides and probe to gyrA (Table 3, codes 22 to 24). Quantitative PCR was performed with Universal PCR master

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2172

DUNCAN ET AL.

INFECT. IMMUN.

of 7 proteins (Fig. 1). All of them were predicted to contain a signal peptide, suggesting that they might be exported. Two proteins of 80 kDa and 39 kDa (Fig. 1) were identified as a chitinase (ChiA) and a zinc metalloprotease (ProA), respectively. These proteins were reported in the past to be involved in L. pneumophila pathogenicity (20, 68). IcmX was identified within the species migrating at an estimated molecular mass of 28 kDa (Fig. 1). This protein is required for the biogenesis of the intracellular phagosome, but its precise function is unknown (7, 47). Mass spectrometry analysis of the species migrating at an estimated molecular mass of 20 to 25 kDa (Fig. 1) identified the major outer membrane protein (MOMP), encoded by lpg2961. MOMP may be involved in adhesion to phagocytic cells and plays a role in virulence in a chicken embryo model (40). CelA (Fig. 1), encoded by lpg1918, was shown to be a secreted endoglucanase (61). LapA (Fig. 1), encoded by lpg2814, was shown to be a secreted aminopeptidase (20, 68). No additional experiments were done to confirm the heparin binding properties of these proteins. Finally, we identified a 52-kDa protein which had been annotated as phage tail fiber protein and is encoded by lpg2644 of Lp1 Philadelphia strain (Fig. 1). Peak elution of this protein was observed at 380 mM NaCl, which is comparable with previously described heparin binding adhesins from B. pertussis and M. tuberculosis (50, 52). Homologues of lpg2644 were found in the genome sequences of the Lens (lpl2569), Paris (lpp2697), Alcoy (lpa0387), and Corby (lpc0495) strains (19, 20, 33). Gene lpg2644 was shown to be distributed largely in environmental and clinical Lp1 strains (86). The product of lpg2644 (Lcl) was previously identified as a soluble type II secreted protein found both in L. pneumophila culture super-

FIG. 2. (A) Hydropathy profile of deduced amino acid sequence of lpg2644. The double arrow indicates the hydrophilic repeated sequences. (B) Multiple tandem repeats of Lcl. Gray boxes highlight the conserved KGD domains present throughout the Lcl predicted amino acid (AA) sequence.

natant and in outer membrane vesicles (20, 28). More recently, using cellular fractionation, Lcl was also shown to be an outer membrane protein (86). Importantly, polyclonal anti-Lcl antibodies were shown to partially block Lp1 cytoadherence (86). Consistent with previous reports (20, 86), in silico analyses showed that Lcl (calculated mass of 49.6 kDa) has similarities to eukaryotic collagen and contains multiple collagen-like tandem repeats, which are frequently associated with binding to proteins during processes such as protein transport and regulation. Furthermore, Lcl contains an RGD motif from amino acids 62 to 64. This motif is known to mediate adhesion to integrins (69), which suggests that Lcl might also bind to this host cell receptor. Hydropathy analysis of Lcl predicted that the repeat region is highly hydrophilic, and thus, this domain is likely to be exposed on the surface of the protein (Fig. 2A). At pH 7.0, the calculated charges of Lcl (pI 5.28) and its hydrophilic repeat region (pI 4.96) are ⫺12.38 and ⫺6.19, respectively. This suggests that electrostatic interactions may play only a minor role in the binding of Lcl with negatively charged heparin. Sequence analysis of the Lcl tandem repeats showed that it contains 17 KGD motifs. These motifs are alternative tripeptides to RGD (Fig. 2B), and they are also able to bind integrins (3, 65). Interestingly, the GC content of Lp1 is 38% (12), while that of lpg2644 is 51.3%. The GC contents of open reading frames (ORFs) lpg2643 and uvrC, respectively upstream and downstream of lpg2644, are both 39%. This suggests that this particular gene might have been acquired by

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 1. SDS-PAGE analysis of proteins purified from L. pneumophila culture supernatant by heparin affinity chromatography. Purified proteins were separated using 7% polyacrylamide gel followed by silver staining. Molecular masses are indicated in kDa on the left. Locus tags and genes encoding proteins identified by matrix-assisted laser desorption ionization–time of flight (MALDI-TOF) MS analysis are indicated on the right.

VOL. 79, 2011

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2173

horizontal transfer and could potentially be absent from other Legionella species. ORFs lpg2643 and lpg2644 have the same orientation but are separated by 167 bp that include the functional promoter of lpg2644 (data not shown). ORFs lpg2644 and uvrC have opposite orientations. None of these ORFs seems to be part of an operon. Altogether, the in silico and preliminary data suggested that Lcl might be a GAG binding adhesin, and we decided to focus our study on its functional characterization. Lcl polymorphic protein is produced in vitro. Previous studies showed that lpg2644 is a polymorphic gene in Lp1 (23, 86). Based on this initial finding, we sought to evaluate whether polymorphic Lcl proteins are produced in vitro in a panel of circulating clinical Lp1 strains. A PCR analysis was performed with 21 clinical isolates of L. pneumophila serogroup 1 (Table 1, codes a to u) using a set of primers that flanked lpg2644 (Table 3, primers 1 and 2). All PCR amplifications of lpg2644 homologues gave single amplicons (data not shown). As expected, six different estimated sizes of amplicons ranging from 1,829 to 2,595 bp were observed (Fig. 3A). DNA sequencing from selected isolates (Table 1, codes a to f; Table 3, primers 1 to 5) confirmed that the polymorphism of lpg2644 was directly related to the number of predicted repeats. Interestingly, it also revealed that there was heterogeneity in the repeat sequences among these isolates (data not shown). Anti-recombinant Lcl (rLcl) polyclonal antibodies were produced in rabbits immunized with the purified rLcl. Immunoblot analysis with anti-rLcl antibodies showed reactivity with rLcl. With a strain Lp02 lysate, Lcl antiserum reacted strongly with a 52kDa protein, corresponding with the estimated size of Lcl, and weakly with a 40-kDa protein (Fig. 3B). No reactivity was observed with rLcl and Lp02 lysate when immunoblot analyses were performed with preimmune rabbit sera. Although antiLcl antibodies did not react with intact live, fixed, and permeabilized bacteria in immunofluorescence assays, they were

highly specific in Western blots and ELISAs. The presence and the size variation of Lcl were then examined in Lp1 clinical isolates selected in our PCR analysis (Table 1, codes a to f). All isolates grown in vitro contained proteins that reacted with anti-Lcl antiserum. Proteins recognized by anti-Lcl antiserum exhibited a marked polymorphism (33.9 to 58.6 kDa) which correlated with the genetic polymorphism of the lpg2644 gene from the same isolates (Fig. 3A and C). Lcl is an immunogenic protein that is produced in vivo. To determine if Lcl is immunogenic and is produced in vivo, the reactivities of sera from 10 convalescent legionellosis patients were tested against rLcl by ELISA and immunoblot analysis. First, to control for seroconversion of convalescent patients to Lp1, sera were tested in ELISAs against an Lp02 lysate. As expected, immune sera showed more reactivity with an Lp02 lysate (Fig. 4A, sera 1 through 10, black bars) than did nonimmune sera (P ⬍ 0.05) (Fig. 4A, sera 11 and 12, black bars). In ELISAs against rLcl, 7 of the 10 immune sera showed significantly higher reactivities than did the nonimmune control serum 11. All immune sera showed significantly higher reactivities against rLcl than nonimmune control serum 12 (Fig. 4A, sera 1 through 10, gray bars). In immunoblot analysis, 9 of the 10 sera (90%) reacted with rLcl (Fig. 4B, lanes 1 through 10), while none of the control sera reacted (Fig. 4B, lanes 11 and 12). These results suggest that Lcl is produced and is antigenic in human infections. Because Lcl appears to elicit various immune responses in humans and its repeats show antigenic polymorphism, we next investigated the role of the repeats of Lcl in its immunogenicity. A purified variant of Lcl lacking collagen-like tandem repeats, rLcl⌬repeats, did not exhibit significant reactivity with patient sera compared to its reactivity with the controls (Fig. 4A, white bars). In all cases, rLcl⌬repeats was significantly less recognized than full-length rLcl. Thus, Lcl immunogenicity is partially driven by the collagen-like tandem repeats.

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 3. Lcl is a polymorphic protein in clinical isolates. (A) PCR analysis of lpg2644 genes in L. pneumophila serogroup 1 clinical isolates. Codes corresponding to the isolates (a to f) are indicated at the top of each lane. The positions of the molecular size markers (M) are indicated in kb on the left side of the gel. (B) Immunoblot analysis of polyclonal antibodies against rLcl and Lp02 lysate using anti-rLcl serum from immune rabbits (i) and preimmune rabbit serum (p). (C) Immunoblot analysis of clinical isolates (lanes a to f) using anti-rLcl antibodies. The molecular masses are indicated to the left of the blot. Proteins recognized by anti-rLcl antibodies showed polymorphism, with molecular masses ranging from 33.9 kDa to 58.6 kDa.

2174

DUNCAN ET AL.

INFECT. IMMUN.

Lcl is only detected in L. pneumophila species. Southern blot analyses detected lpg2644 homologues in L. pneumophila isolates of serogroups 2, 3, 4, 5, 6, 8, 10, and 12 (Fig. 5A, Table 1, codes 2 to 9). Immunoblot analyses with anti-Lcl antiserum detected Lcl homologues in all L. pneumophila serogroups (Fig. 5B). Thus, the lpg2644 gene and the Lcl protein seem to be conserved across the pneumophila species. Because lpg2644 is absent from the recently published Legionella longbeachae genomes (11, 39) and the high GC% of the lpg2644 locus suggested that it might have been recently acquired by the L. pneumophila species, the presence of homologous genes was next tested in a panel of Legionella environmental species that are rarely found in patients. All these isolates were collected from anthropogenic water as part of outbreak investigations in hospital and health care centers. Southern blot analysis under low-stringency conditions did not detect lpg2644 homologues in isolates of L. erythra, L. feeleii, L. rubrilucens, L. maceachernii, L. anisa, L. bozemanii of serogroup 2, or L. bozemanii of serogroup 1 (Fig. 5C; Table 1, codes 10 to 18). In a control Southern blot analysis, despite weak reactivities with L. erythra LR1359 and L. erythra LR1317 DNA, all strains were positive for the presence of the broadly conserved mip gene. Anti-Lcl immunoblots were negative with lysates of these environmental isolates (Fig. 5D). Although this could be caused by a lack of cross-reactivity of anti-Lcl with other species, it correlated with the absence of detection of the lpg2644 gene with Southern blot analysis. These results support the hypothesis that Lcl might be unique to the L. pneumophila species.

An L. pneumophila lpg2644 deletion mutant is impaired in binding to GAGs and in adherence to lung epithelial cells. Collagenlike proteins such as Lcl are rare in prokaryotic organisms, and those described are often involved in adherence (45, 64). Exogenous GAGs were shown to specifically inhibit the binding of Lp1 to lung epithelial cells using in vitro (82, 89) and in vivo (1) experimental models. Taking into account that anti-Lcl polyclonal antibodies were shown to partially block the adherence of Lp1 (86), we hypothesized that an lpg2644 deletion mutant would be impaired in adherence to GAGs and lung epithelial cells. Using the suicide plasmid pSR47S⌬lpg2644 (Table 2, MR45), an Lp02 ⌬lpg2644 isogenic mutant was created (Table 1, code 19). Deletion of lpg2644 was confirmed by Southern blot analysis with an lpg2644 probe, DNA sequencing (data not shown), and anti-Lcl immunoblot analysis (Fig. 6A). Analysis of the growth curves of Lp02 ⌬lpg2644 in BYE broth at 37°C compared to those of wild-type (WT) Lp02 showed that there is no growth defect in the lpg2644 mutant (data not shown). Lp02 ⌬lpg2644 was used to evaluate the contribution of Lcl to the GAG binding properties of Lp1. WT and mutant strains were incubated in wells of microtiter plates coated with GAGs. Binding of Legionella was quantified using qPCR. We showed that Lp02 ⌬lpg2644 bound significantly less to heparin and fucoidan than the Lp02 WT strain (Fig. 6B and C). To determine whether reintroduction of lpg2644 in Lp02 ⌬lpg2644 could restore the WT phenotype, we complemented the lpg2644 mutation using plpg2644 (Table 1, code 20). The ex-

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 4. Lcl is an immunogenic protein produced during legionellosis. (A) ELISA analysis of rLcl and human sera from convalescent legionellosis patients (1 to 10). Nonimmune sera from healthy individuals (11 and 12) were used as controls. Plates were coated with the same concentration of proteins from Lp02 cell lysate (black bars), rLcl (gray bars), and rLcl⌬repeats (white bars). Values represent means ⫾ standard deviations from three individual experiments. ⴱ, P ⬍ 0.05 versus nonimmune control serum 11. (B) Immunoblot analysis showing reactivity between human sera and rLcl.

VOL. 79, 2011

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2175

pression of rLcl in L. pneumophila was confirmed by anti-Lcl immunoblots (Fig. 6A). Complementation of the ⌬lpg2644 by transformation with the plpg2644 plasmid resulted in partial recovery of heparin (Fig. 6B) and fucoidan (Fig. 6C) binding properties. To avoid plasmid instability encountered in nonselective media and/or toxicity due to overexpression of a recombinant protein in subsequent assays, we also complemented the lpg2644 mutation by integrating a WT copy of lpg2644 in the chromosome to obtain Lp02 ⌬lpg2644/clpg2644 (Table 1, code 23). Although not fully restored, the levels of binding of Lp02 ⌬lpg2644/clpg2644 to immobilized heparin and fucoidan were significantly higher than those of Lp02 ⌬lpg2644 (P ⬍ 0.05) (Fig. 6B and C). Interestingly, the overexpressing strain Lp02 plpg2644 (Table 1, code 24) did not show increased binding to coated GAGs. The binding of Lp02 ⌬lpg2644/clpg2644 to coated fucoidan was not significantly different from that of WT Lp02 plpg2644 (P ⬎ 0.05). Lp02 ⌬lpg2644/clpg2644 bound significantly more to fucoidan than did Lp02 ⌬lpg2644 (1.1 ⫻ 107 versus 2.5 ⫻ 106 genome copies/well, P ⬍ 0.05) (Fig. 6C). The contribution of Lcl to the Lp1 adherence to pulmonary epithelial cells was evaluated using in vitro adhesion assays with NCI-H292 human lung mucoepidermoid cells (35). Adhesion assays showed that the adherence of the isogenic lpg2644 mutant to NCI-H292 cells is significantly reduced compared with the adherence of the parental strain Lp02 (Fig. 6D). To confirm the phenotype observed with CFU assays, the adherence

of Lp02 and the isogenic lpg2644 mutant was also compared in microscopy assays by counting the average numbers of attached bacteria per 250 cells. In six independent assays, the isogenic lpg2644 mutant (51 ⫾ 4.9 [mean ⫾ standard deviation] bacteria/250 cells) showed a significant (28%) decrease in binding (P ⬍ 0.05) compared to the binding of the WT (70.83 ⫾ 6.6 bacteria/250 cells). In CFU assays, the complemented mutant regained binding to NCI-H292 cells (P ⬎ 0.05) and adhered significantly more than Lp02 ⌬lpg2644 (Fig. 6D). To evaluate the potential role of Lcl in cell invasion, gentamicin protection assays were performed. The ratios between CFU of adherent and intracellular bacteria did not differ between the WT and Lp02 ⌬lpg2644, suggesting that Lcl might not play a role in cellular entry (data not shown). GAG binding properties of rLcl. Taking into account the finding that Lp02 ⌬lpg2644 is deficient in binding to immobilized GAGs and lung epithelial cells, we sought to further explore the GAG binding properties of Lcl using recombinant proteins produced in E. coli. Native Lcl could be affinity purified from an Lp1 culture supernatant using heparin affinity chromatography. To ascertain that a recombinant Lcl retains Lcl binding properties, lysates of E. coli expressing rLcl (Table 2, MR51) were passed through a heparin-agarose chromatography column. Recombinant Lcl was eluted at 500 mM NaCl, suggesting that it retains the heparin binding properties of native Lcl (Fig. 7A). This result was further confirmed by

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 5. Lcl is expressed by all clinical L. pneumophila serogroups tested. (A and C) Southern blot analyses of lpg2644 homologues among clinical non-Lp1 serogroups (A) and non-pneumophila Legionella isolates (C) were performed. Serogroups of clinical isolates (2 to 12) and Legionella species are indicated at the top of each lane. A control hybridization pattern using a conserved mip probe was obtained with non-pneumophila Legionella isolates (C). The positions of the molecular size markers are indicated on the left side of each gel in kb. (B and D) Immunoblot analyses of cell lysates of different serogroups (B) and non-pneumophila Legionella isolates (D) using anti-rLcl antibodies. Molecular masses are marked on the left side of the gel in kDa. sg, serogroup(s).

2176

DUNCAN ET AL.

INFECT. IMMUN.

ELISAs showing specific binding of biotinylated heparin to immobilized rLcl (Fig. 7B). Heparin binding assays performed with rLcl lacking collagen repeats showed lower OD values than assays with full-length Lcl. These results suggest that the collagen-like tandem repeats play a critical role in its binding properties. We next examined the GAG binding specificity of rLcl. Binding to GAGs was detected using anti-Lcl-specific antiserum. Recombinant Lcl exhibited significant affinity for heparin, heparan sulfate, dextran sulfate, chondroitin sulfate A, and fucoidan compared to its affinity for immobilized BSA (P ⬍ 0.05). In contrast, rLcl did not exhibit significant affinity for dextran or chondroitin sulfate B or C (P ⬎ 0.05) (Fig. 7C).

The higher affinities were observed with chondroitin sulfate A (OD ⫽ 0.33) and fucoidan (OD ⫽ 2.18). Lcl is involved in L. pneumophila biofilm formation. To explain the high incidence of Lp1 compared to that of other Legionella species in legionellosis patients, it has been hypothesized that Lp1 is more fit for survival in aquatic environments (62). One of the strategies used by Lp1 to survive in an anthropogenic aqueous environment is to form biofilms, a process that requires adhesin(s) to fulfill the initial step of attachment (21, 22, 67). As it was shown that GAG binding adhesins play a role in the adherence of Lp1 to epithelial cells, we hypothesized that exogenous GAGs might also specifically in-

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 6. Deletion of lpg2644 reduced the binding of L. pneumophila to GAGs and human lung epithelial cells. (A to C) WT (Lp02), an lpg2644 deletion mutant (Lp02 ⌬lpg2644), complemented deletion mutant strains (Lp02 ⌬lpg2644 clpg2644 and Lp02 ⌬lpg2644 plpg2644), and an overexpressing strain (Lp02 plpg2644) were analyzed by immunoblot with anti-rLcl polyclonal antibodies (A) and qPCR to measure binding to wells coated with heparin (B) and fucoidan (C). (D) Adhesion assays to quantify the role of Lcl in L. pneumophila binding to lung epithelial cell line NCI-H292. The binding of the Lp02 ⌬lpg2644 deletion mutant was significantly reduced compared to that of WT Lp02 and a complemented mutant. Values are means ⫾ standard deviations from three individual experiments. ⴱ, P ⬍ 0.05 for Lp02 ⌬lpg2644 versus all other strains by two-tailed Student’s t test.

VOL. 79, 2011

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2177

DISCUSSION FIG. 7. Lcl binds glycosaminoglycans. (A) Immunoblot of Lcl purified from cell lysate of Lcl-expressing E. coli using heparin-agarose affinity chromatography. Purified rLcl (lane 1), cell lysate of Lcl-expressing E. coli (lane 2), a flowthrough fraction (lane 3), protein fraction eluted at 150 mM NaCl (lane 4), and peak of elution obtained at 500 mM NaCl (lanes 5 to 8). Lcl was detected using anti-His-tagged antibodies. Coomassie-stained gel shows equal loading of the proteins. Arrows point to bands corresponding to the estimated size of Lcl. Molecular masses in kDa are indicated on the left side. (B) ELISA analysis of binding between immobilized rLcl or rLcl⌬repeats and biotinylated heparin. HRP-streptavidin was used to measure OD450 values. BSA-coated wells were used as controls. Values are means ⫾ standard deviations from three individual experiments. ⴱ, P ⬍ 5.4 ⫻ 10⫺6 versus BSA by two-tailed Student’s t test. (C) Binding of rLcl to immobilized GAGs. ELISA analysis was performed using anti-rLcl antibodies. HRP-conjugated anti-rabbit antibodies were used to measure OD450 values. BSA-coated wells were used as controls. Values are means ⫾ standard deviations from three individual experiments done in triplicates. ⴱ, P ⬍ 0.05 versus BSA by two-tailed Student’s t test.

The initial attachment to host cells and surfaces is a crucial event in the pathogenesis of most infectious agents (15). In this study, we identified 7 potential Lp1 heparin binding proteins using affinity chromatography, and we revealed the GAG binding properties of Lcl. Using an lpg2644 isogenic mutant, we showed that Lcl plays a key role in the attachment of L. pneumophila. To evaluate the clinical relevance of Lcl, we tested for the presence of lpg2644 homologues in clinical isolates and nonpneumophila Legionella species. Our data suggest that lpg2644 is unique to the L. pneumophila species, which is an indicator that it might have been recently acquired though horizontal transfer. This hypothesis is consistent with the high GC content of lpg2644 in comparison to the GC content of the Lp1 genome and with its absence from the L. longbeachae genome. We propose that the recent acquisition of this gene may provide an advantage to L. pneumophila by making it more fit for survival

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

hibit biofilm formation (81). Lp02 was grown in the presence or absence of mannose, glucose, chondroitin sulfate C, and fucoidan. These carbohydrates were nontoxic at all concentrations tested (up to 0.25 mg/ml). Chondroitin sulfate C was selected due to its low binding to Lcl in ELISAs, and mannose and glucose were used as negative controls. Lp02 formed normal biofilms in the presence of mannose, glucose, and chondroitin sulfate C, but a dose-dependent inhibition of biofilm formation was observed in the presence of fucoidan (Fig. 8A). Based on the finding that Lcl binds strongly to fucoidan, we next hypothesized that Lcl could also play a role in the biofilm formation of Lp1. Static biofilm assays were performed on WT Lp02, the Lp02 ⌬lpg2644 isogenic mutant, the Lp02 ⌬lpg2644 complemented strains, and the Lp02 plpg2644 overexpressing strain. After 2 and 6 days of incubation, WT Lp02 was able to form a biofilm as assessed by crystal violet-staining assay (Fig. 8B). In contrast, Lp02 ⌬lpg2644 was dramatically impaired in forming biofilms (Fig. 8B). The complemented mutants, Lp02 ⌬lpg2644 clpg2644 and Lp02 ⌬lpg2644 plpg2644, showed biofilm formation similar to that of the WT after 2 and 6 days of incubation (Fig. 8B). The overexpressing strain, Lp02 plpg2644, did not show a significant increase in biofilm formation. Comparable results were obtained when borosilicate glass tubes were used instead of polystyrene microtiter plates (data not shown). Interestingly, Legionella species which were negative for the presence of lpg2644 homologues (Fig. 5D) showed very low levels of biofilm formation after 2 and 6 days of incubation in comparison to the biofilm formation of Lp02 (Fig. 8C). To look further into the deficiency of the lpg2644 deletion mutant in producing biofilm, we also used confocal laser scanning microscopy to compare the thickness and structure of biofilms grown on glass-bottom chambers immersed in BYE broth at 37°C, using GFP-expressing WT and isogenic ⌬lpg2644 strains (Table 1, codes 25 and 26). The isogenic mutant was unable to produce uniform biofilms, and the thickness of the structure produced was considerably decreased compared to the thickness of the WT strain’s biofilm (Fig. 9A and B). Altogether, these data strongly suggest that Lcl is crucial for the production of biofilm in Lp1.

2178

DUNCAN ET AL.

in aquatic environments and/or participate in its increased virulence compared to the virulence of other Legionella species. Sera from legionellosis patients showed reactivity with rLcl, which suggests that it is produced during human infections. Moreover, the immunogenicity of Lcl is driven in part by its collagen-like repeats. This has also been shown with Scl collagen-like proteins in Streptococcus. Like Lcl, Scl proteins are immunogenic and exhibit structural similarity to human collagen (GXY repeats) (36). Our data suggest that the antigenic polymorphism of Lcl may elicit various immune responses in infected patients. Therefore, it could also participate in evasion from preexisting immunity acquired during prior infection with different strains. In addition to heparin, we show that rLcl binds to highly sulfated GAG analogs like chondroitin sulfate A, which is a major component of lung polysaccharides (37, 71). Although the rLcl protein was eluted from a heparin (5 to 30 kDa) chromatography column at 500 mM NaCl, ELISA binding assay values with heparin (4 to 6 kDa) were relatively low compared to those of other GAGs. This difference may be explained by the use of heparin with different molecular weights, potential inhomogeneous degrees of sulfation, and/or the low coating efficiency of heparin (83). The highest OD values of our GAG binding assays were obtained with fucoidan. Like chondroitin sulfate A, fucoidan is a highly sulfated polysaccharide. While not present in mammalians, fucoidan is predominantly composed of sulfated fucose, which is a critical component of the extracellular matrix of the lung epithelium (91). Increased levels of fucosyl residues are associated with inflammatory processes and various cancers, conditions which lead to a higher risk of Legionella infections (32, 46). In other respiratory pathogens, such as Pseudomonas aeruginosa and Haemophilus influenzae, host fucosylated proteoglycans are crucial primary adhesion receptors (66). Fucosylation of host E-cadherin and ␣3␤1 integrin plays an essential role in their signaling functions (79). Additional work is needed to determine whether Lcl can bind to such fucosylated proteoglycan receptors and trigger cell signaling. Polymers of sulfated fucose are major constituents of brown seaweed species and extracellular polymeric substances produced by cyanobacteria (43). Legionella bacteria have been isolated from algal biofilm communities in aquatic environments (73). Lcl may serve as an anchoring point for the formation of community-associated biofilms with algae. To evaluate the contribution of Lcl to the GAG binding properties of Lp1, we constructed an isogenic lpg2644 mutant of Lp02. The binding of Lp02 ⌬lpg2644 to GAGs is significantly reduced compared to that of the WT. This finding suggests that Lcl plays a role in the GAG binding properties of Lp1. Conversely, the binding of Lp1 to GAGs was not completely abolished in the lpg2644 mutant; thus, it is possible that other bacterial molecules contribute to Lp1’s GAG binding properties. This result is consistent with the identification of additional Lp1 heparin binding proteins in this study. Importantly, an isogenic lpg2644 mutant of Lp1 was also significantly impaired in binding to lung epithelial cells. This result correlates with the finding that preincubation with exogenous heparin and anti-Lcl antibodies can inhibit the binding of Lp1 to lung epithelial cells (81, 86). It is also consistent with in vivo data showing that heparin can prevent bacterial dissemination

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 8. Lcl promotes biofilm formation. (A) Production of biofilm by L. pneumophila Lp02 in the presence of mannose (triangles), glucose (circles), chondroitin sulfate C (squares), and fucoidan (diamonds) after 2 days of incubation. (B and C) L. pneumophila strains (Lp02, Lp02 ⌬lpg2644, Lp02 ⌬lpg2644 clpg2644, Lp02 plpg2644, and Lp02 ⌬lpg2644 plpg2644) (B) and non-pneumophila Legionella species grown in BYE broth were added to polystyrene 96-well plates and grown at 37°C and 5% CO2 (C). Bacteria were incubated for 2 days (gray bars) and 6 days (white bars) before OD600 was measured to quantify biofilm formation. Error bars represent the standard deviations of results from 4 to 6 replicates; data are representative of 3 independent experiments.

INFECT. IMMUN.

VOL. 79, 2011

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN

2179

and the inflammatory response to Lp1 in a mouse model (1). Like Lcl, other members of the collagen-like superfamily were shown to participate in the adherence to human lung epithelial cells. The collagen-like proteins in group A Streptococcus, Scls, play a role in soft tissue pathology, and their collagen-like repeats mediate internalization in epithelial cells via binding to the human collagen receptors integrin ␣2␤1 and ␣11␤1 (9, 10, 45). One of the strategies used by Lp1 to colonize aqueous environments is to form biofilms, but little is known about the genetic determinant involved in this process. Here, we demonstrate that exogenous sulfated carbohydrates showing a high affinity for Lcl can also specifically inhibit the formation of Lp1 biofilm in a dose-dependent manner. Deletion of lpg2644 dramatically affects the production of Lp1 biofilm. Thus, the finding that Lcl is ubiquitous in the L. pneumophila species but undetected in other tested species may explain the high prevalence of L. pneumophila in human legionellosis. This is also consistent with the low level of biofilm production that we observed with species that lack lpg2644 homologues. From the data presented in our study, we suspect that Lcl may play a role in the initial attachment step of biofilm formation, but it could also be involved in cell-to-cell interactions. Further experiments are needed to confirm these hypotheses. In other bacteria, adherence mechanisms that mediate adhesion to host cells can also mediate biofilm formation (34, 56, 84). Although there is no evidence that it is the case for Lp1, biofilm formation can also be involved in host colonization by promoting stronger adhesion to host cells. Typical examples are the fibronectin binding proteins of methicillin-resistant Staphylococcus aureus (FnBPA and FnBPB), which mediate adherence to host cells but also participate in biofilm formation (34, 56). More recently, an outer membrane lectin of P. aeruginosa, designated LecB, was shown to mediate adherence to lung epithelial cells and to be involved in bacterial dissemination (14). Similar to Lcl, LecB has a high affinity for fucose analogs (88) and is involved in biofilm formation (84). In summary, we have revealed the GAG binding properties of Lcl, a polymorphic adhesin of L. pneumophila that is con-

served among clinical isolates and serogroups and remains undetected in other Legionella species. We have also shown here that Lcl is an immunogenic antigen in the human host, which mediates the binding of Lp1 to GAGs and human lung epithelial cells. We have revealed that Lcl plays a major role in biofilm formation, a process that is believed to be critical for the proliferation and dissemination of this waterborne pathogen (42). ACKNOWLEDGMENTS This work was supported by the Ontario Agency for Health Protection and Promotion and the Canadian Institutes of Health Research (grant MOP-102514). A. Prashar and M. Terebiznik are supported by grants from NSERC Canada. We thank M. S. Swanson, R. R. Isberg, and R. S. Gardun ˜o for sharing the L. pneumophila strain Lp02, P. S. Hoffman for generously providing plasmid pBH6119, and C. R. Roy for sharing plasmid pSR47S. We thank P. S. Hoffman, G. Mallo, and T. G. Schwan for reviewing the manuscript. We are grateful to F. D. Menozzi who inspired this work. REFERENCES 1. Ader, F., et al. 2008. In vivo effect of adhesion inhibitor heparin on Legionella pneumophila pathogenesis in a murine pneumonia model. Intensive Care Med. 34:1511–1519. 2. Bartram, J., Y. Chartier, J. V. Lee, K. Pond, and S. Surman-Lee (ed.). 2007. Legionella and the prevention of legionellosis. World Health Organization, WHO Press, Geneva, Switzerland. 3. Behera, A. K., et al. 2008. Borrelia burgdorferi BBB07 interaction with integrin alpha3beta1 stimulates production of pro-inflammatory mediators in primary human chondrocytes. Cell. Microbiol. 10:320–331. 4. Benin, A. L., R. F. Benson, and R. E. Besser. 2002. Trends in legionnaires disease, 1980-1998: declining mortality and new patterns of diagnosis. Clin. Infect. Dis. 35:1039–1046. 5. Berger, K. H., and R. R. Isberg. 1993. Two distinct defects in intracellular growth complemented by a single genetic locus in Legionella pneumophila. Mol. Microbiol. 7:7–19. 6. Boyle, E. C., and B. B. Finlay. 2003. Bacterial pathogenesis: exploiting cellular adherence. Curr. Opin. Cell Biol. 15:633–639. 7. Brand, B. C., A. B. Sadosky, and H. A. Shuman. 1994. The Legionella pneumophila icm locus: a set of genes required for intracellular multiplication in human macrophages. Mol. Microbiol. 14:797–808. 8. Cabanes, D., et al. 2005. Gp96 is a receptor for a novel Listeria monocytogenes virulence factor, Vip, a surface protein. EMBO. J. 24:2827–2838. 9. Caswell, C. C., et al. 2008. Identification of the first prokaryotic collagen sequence motif that mediates binding to human collagen receptors, integrins alpha2beta1 and alpha11beta1. J. Biol. Chem. 283:36168–36175.

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

FIG. 9. Confocal scanning laser microscopy (CSLM) micrographs (A) and average thicknesses (B) of biofilms produced by GFP-expressing Lp02 and isogenic ⌬lpg2644 strains. A representative CSLM image is shown for each sample. Error bars represent the standard deviations of thicknesses from 10 representative microscopic fields. Data are representative of 3 independent experiments.

2180

DUNCAN ET AL.

38. Konto-Ghiorghi, Y., et al. 2009. Dual role for pilus in adherence to epithelial cells and biofilm formation in Streptococcus agalactiae. PLoS Pathog. 5:e1000422. 39. Kozak, N. A., et al. 2010. Virulence factors encoded by Legionella longbeachae identified on the basis of the genome sequence analysis of clinical isolate D-4968. J. Bacteriol. 192:1030–1044. 40. Krinos, C., A. S. High, and F. G. Rodgers. 1999. Role of the 25 kDa major outer membrane protein of Legionella pneumophila in attachment to U-937 cells and its potential as a virulence factor for chick embryos. J. Appl. Microbiol. 86:237–244. 41. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680–685. 42. Lau, H. Y., and N. J. Ashbolt. 2009. The role of biofilms and protozoa in Legionella pathogenesis: implications for drinking water. J. Appl. Microbiol. 107:368–378. 43. Li, B., F. Lu, X. Wei, and R. Zhao. 2008. Fucoidan: structure and bioactivity. Molecules 13:1671–1695. 44. Lu ¨ck, C. P., et al. 2010. Legionella dresdenensis sp. nov. isolated from river water. Int. J. Syst. Evol. Microbiol. 60:2557–2562. 45. Lukomski, S., et al. 2000. Identification and characterization of the scl gene encoding a group A Streptococcus extracellular protein virulence factor with similarity to human collagen. Infect. Immun. 68:6542–6553. 46. Matsumura, K., et al. 2007. Carbohydrate binding specificity of a fucosespecific lectin from Aspergillus oryzae: a novel probe for core fucose. J. Biol. Chem. 282:15700–15708. 47. Matthews, M., and C. R. Roy. 2000. Identification and subcellular localization of the Legionella pneumophila IcmX protein: a factor essential for establishment of a replicative organelle in eukaryotic host cells. Infect. Immun. 68:3971–3982. 48. Menozzi, F. D., R. Bischoff, E. Fort, M. J. Brennan, and C. Locht. 1998. Molecular characterization of the mycobacterial heparin-binding hemagglutinin, a mycobacterial adhesin. Proc. Natl. Acad. Sci. U. S. A. 95:12625– 12630. 49. Menozzi, F. D., C. Gantiez, and C. Locht. 1991. Identification and purification of transferrin- and lactoferrin-binding proteins of Bordetella pertussis and Bordetella bronchiseptica. Infect. Immun. 59:3982–3988. 50. Menozzi, F. D., C. Gantiez, and C. Locht. 1991. Interaction of the Bordetella pertussis filamentous hemagglutinin with heparin. FEMS Microbiol. Lett. 62:59–64. 51. Menozzi, F. D., et al. 2002. Enhanced bacterial virulence through exploitation of host glycosaminoglycans. Mol. Microbiol. 43:1379–1386. 52. Menozzi, F. D., et al. 1996. Identification of a heparin-binding hemagglutinin present in mycobacteria. J. Exp. Med. 184:993–1001. 53. Merriam, J. J., R. Mathur, R. Maxfield-Boumil, and R. R. Isberg. 1997. Analysis of the Legionella pneumophila fliI gene: intracellular growth of a defined mutant defective for flagellum biosynthesis. Infect. Immun. 65:2497– 2501. 54. Mody, C. H., et al. 1993. Legionella pneumophila replicates within rat alveolar epithelial cells. J. Infect. Dis. 167:1138–1145. 55. Newton, H. J., D. K. Ang, I. R. van Driel, and E. L. Hartland. 2010. Molecular pathogenesis of infections caused by Legionella pneumophila. Clin. Microbiol. Rev. 23:274–298. 56. O’Neill, E., et al. 2008. A novel Staphylococcus aureus biofilm phenotype mediated by the fibronectin-binding proteins, FnBPA and FnBPB. J. Bacteriol. 190:3835–3850. 57. Ozeri, V., et al. 2001. De novo formation of focal complex-like structures in host cells by invading Streptococci. Mol. Microbiol. 41:561–573. 58. Parveen, N., M. Caimano, J. D. Radolf, and J. M. Leong. 2003. Adaptation of the Lyme disease spirochaete to the mammalian host environment results in enhanced glycosaminoglycan and host cell binding. Mol. Microbiol. 47: 1433–1444. 59. Parveen, N., and J. M. Leong. 2000. Identification of a candidate glycosaminoglycan-binding adhesin of the Lyme disease spirochete Borrelia burgdorferi. Mol. Microbiol. 35:1220–1234. 60. Payne, N. R., and M. A. Horwitz. 1987. Phagocytosis of Legionella pneumophila is mediated by human monocyte complement receptors. J. Exp. Med. 166:1377–1389. 61. Pearce, M. M., and N. P. Cianciotto. 2009. Legionella pneumophila secretes an endoglucanase that belongs to the family-5 of glycosyl hydrolases and is dependent upon type II secretion. FEMS Microbiol. Lett. 300:256–264. 62. Piao, Z., C. C. Sze, O. Barysheva, K. Iida, and S. Yoshida. 2006. Temperature-regulated formation of mycelial mat-like biofilms by Legionella pneumophila. Appl. Environ. Microbiol. 72:1613–1622. 63. Pizarro-Cerda, J., and P. Cossart. 2006. Bacterial adhesion and entry into host cells. Cell 124:715–727. 64. Rasmussen, M., M. Jacobsson, and L. Bjorck. 2003. Genome-based identification and analysis of collagen-related structural motifs in bacterial and viral proteins. J. Biol. Chem. 278:32313–32316. 65. Reiss, S., et al. 2006. Inhibition of platelet aggregation by grafting RGD and KGD sequences on the structural scaffold of small disulfide-rich proteins. Platelets 17:153–157. 66. Rhim, A. D., L. Stoykova, M. C. Glick, and T. F. Scanlin. 2001. Terminal

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

10. Caswell, C. C., E. Lukomska, N. S. Seo, M. Hook, and S. Lukomski. 2007. Scl1-dependent internalization of group A Streptococcus via direct interactions with the alpha2beta(1) integrin enhances pathogen survival and reemergence. Mol. Microbiol. 64:1319–1331. 11. Cazalet, C., et al. 2010. Analysis of the Legionella longbeachae genome and transcriptome uncovers unique strategies to cause Legionnaires’ disease. PLoS Genet. 6:e1000851. 12. Cazalet, C., et al. 2004. Evidence in the Legionella pneumophila genome for exploitation of host cell functions and high genome plasticity. Nat. Genet. 36:1165–1173. 13. Chang, B., F. Kura, J. Amemura-Maekawa, N. Koizumi, and H. Watanabe. 2005. Identification of a novel adhesion molecule involved in the virulence of Legionella pneumophila. Infect. Immun. 73:4272–4280. 14. Chemani, C., et al. 2009. Role of LecA and LecB lectins in Pseudomonas aeruginosa-induced lung injury and effect of carbohydrate ligands. Infect. Immun. 77:2065–2075. 15. Chen, Y., M. Gotte, J. Liu, and P. W. Park. 2008. Microbial subversion of heparan sulfate proteoglycans. Mol. Cells 26:415–426. 16. Chrystal, R. G., J. B. West, and P. J. Barnes. 1997. The lung: scientific foundations. Lippincott-Raven Press, Philadelphia, PA. 17. Cianciotto, N. P., J. K. Stamos, and D. W. Kamp. 1995. Infectivity of Legionella pneumophila mip mutant for alveolar epithelial cells. Curr. Microbiol. 30:247–250. 18. Cirillo, S. L., L. E. Bermudez, S. H. El-Etr, G. E. Duhamel, and J. D. Cirillo. 2001. Legionella pneumophila entry gene rtxA is involved in virulence. Infect. Immun. 69:508–517. 19. D’Auria, G., N. Jimenez-Hernandez, F. Peris-Bondia, A. Moya, and A. Latorre. 2010. Legionella pneumophila pangenome reveals strain-specific virulence factors. BMC Genomics 11:181. 20. DebRoy, S., J. Dao, M. Soderberg, O. Rossier, and N. P. Cianciotto. 2006. Legionella pneumophila type II secretome reveals unique exoproteins and a chitinase that promotes bacterial persistence in the lung. Proc. Natl. Acad. Sci. U. S. A. 103:19146–19151. 21. Declerck, P. 2010. Biofilms: the environmental playground of Legionella pneumophila. Environ. Microbiol. 12:557–566. 22. Declerck, P., J. Behets, V. van Hoef, and F. Ollevier. 2007. Replication of Legionella pneumophila in floating biofilms. Curr. Microbiol. 55:435–440. 23. Duncan, C., et al. 2009. LHBP1, a novel immunogenic gylcosaminoglycan binding protein of Legionella pneumophila, is produced during legionellosis and contributes to alveolar epithelial cells adhesion, abstr. P63, p. 145. Legionella 2009, Institut Pasteur, Paris, France. 24. Ewann, F., and P. S. Hoffman. 2006. Cysteine metabolism in Legionella pneumophila: characterization of an L-cystine-utilizing mutant. Appl. Environ. Microbiol. 72:3993–4000. 25. Feeley, J. C., et al. 1979. Charcoal-yeast extract agar: primary isolation medium for Legionella pneumophila. J. Clin. Microbiol. 10:437–441. 26. Fields, B. S., R. F. Benson, and R. E. Besser. 2002. Legionella and Legionnaires’ disease: 25 years of investigation. Clin. Microbiol. Rev. 15:506–526. 27. Fliermans, C. B., et al. 1981. Ecological distribution of Legionella pneumophila. Appl. Environ. Microbiol. 41:9–16. 28. Galka, F., et al. 2008. Proteomic characterization of the whole secretome of Legionella pneumophila and functional analysis of outer membrane vesicles. Infect. Immun. 76:1825–1836. 29. Gao, L. Y., B. J. Stone, J. K. Brieland, and Y. Abu Kwaik. 1998. Different fates of Legionella pneumophila pmi and mil mutants within macrophages and alveolar epithelial cells. Microb. Pathog. 25:291–306. 30. Gao, L. Y., M. Susa, B. Ticac, and Y. Abu Kwaik. 1999. Heterogeneity in intracellular replication and cytopathogenicity of Legionella pneumophila and Legionella micdadei in mammalian and protozoan cells. Microb. Pathog. 27:273–287. 31. Garduno, R. A., E. Garduno, and P. S. Hoffman. 1998. Surface-associated hsp60 chaperonin of Legionella pneumophila mediates invasion in a HeLa cell model. Infect. Immun. 66:4602–4610. 32. Ginevra, C., et al. 2009. Host-related risk factors and clinical features of community-acquired Legionnaires disease due to the Paris and Lorraine endemic strains, 1998-2007, France. Clin. Infect. Dis. 49:184–191. 33. Glockner, G., et al. 2008. Identification and characterization of a new conjugation/type IVA secretion system (trb/tra) of Legionella pneumophila Corby localized on two mobile genomic islands. Int. J. Med. Microbiol. 298:411– 428. 34. Heilmann, C., et al. 2004. Staphylococcus aureus fibronectin-binding protein (FnBP)-mediated adherence to platelets, and aggregation of platelets induced by FnBPA but not by FnBPB. J. Infect. Dis. 190:321–329. 35. Hierholzer, J. C., E. Castells, G. G. Banks, J. A. Bryan, and C. T. McEwen. 1993. Sensitivity of NCI-H292 human lung mucoepidermoid cells for respiratory and other human viruses. J. Clin. Microbiol. 31:1504–1510. 36. Hoe, N. P., E. Lukomska, J. M. Musser, and S. Lukomski. 2007. Characterization of the immune response to collagen-like proteins Scl1 and Scl2 of serotype M1 and M28 group A Streptococcus. FEMS Microbiol. Lett. 277: 142–149. 37. Horwitz, A. L., and R. C. Crystal. 1975. Content and synthesis of glycosaminoglycans in the developing lung. J. Clin. Invest. 56:1312–1318.

INFECT. IMMUN.

VOL. 79, 2011

67.

68.

69. 70. 71.

73.

74.

75. 76. 77.

78. 79.

Editor: J. N. Weiser

2181

80. Tang, P. W., and S. Toma. 1986. Broad-spectrum enzyme-linked immunosorbent assay for detection of Legionella soluble antigens. J. Clin. Microbiol. 24:556–558. 81. Thomas, R., and T. Brooks. 2004. Common oligosaccharide moieties inhibit the adherence of typical and atypical respiratory pathogens. J. Med. Microbiol. 53:833–840. 82. Thomas, R. J., and T. J. Brooks. 2004. Oligosaccharide receptor mimics inhibit Legionella pneumophila attachment to human respiratory epithelial cells. Microb. Pathog. 36:83–92. 83. Thompson, L. D., M. W. Pantoliano, and B. A. Springer. 1994. Energetic characterization of the basic fibroblast growth factor-heparin interaction: identification of the heparin binding domain. Biochemistry 33:3831–3840. 84. Tielker, D., et al. 2005. Pseudomonas aeruginosa lectin LecB is located in the outer membrane and is involved in biofilm formation. Microbiology 151: 1313–1323. 85. Towbin, H., T. Staehelin, and J. Gordon. 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl. Acad. Sci. U. S. A. 76:4350–4354. 86. Vandersmissen, L., E. De Buck, V. Saels, D. A. Coil, and J. Anne. 2010. A Legionella pneumophila collagen-like protein encoded by a gene with a variable number of tandem repeats is involved in the adherence and invasion of host cells. FEMS Microbiol. Lett. 306:168–176. 87. Wagner, C., et al. 2007. Collagen binding protein Mip enables Legionella pneumophila to transmigrate through a barrier of NCI-H292 lung epithelial cells and extracellular matrix. Cell. Microbiol. 9:450–462. 88. Wu, A. M., et al. 2006. Interactions of the fucose-specific Pseudomonas aeruginosa lectin, PA-IIL, with mammalian glycoconjugates bearing polyvalent Lewis(a) and ABH blood group glycotopes. Biochimie 88:1479–1492. 89. Yaradou, D. F., et al. 2007. Zinc-dependent cytoadherence of Legionella pneumophila to human alveolar epithelial cells in vitro. Microb. Pathog. 43:234–242. 90. Yu, V. L., et al. 2002. Distribution of Legionella species and serogroups isolated by culture in patients with sporadic community-acquired legionellosis: an international collaborative survey. J. Infect. Dis. 186:127–128. 91. Zhao, Y., et al. 2008. Branched N-glycans regulate the biological functions of integrins and cadherins. FEBS J. 275:1939–1948.

Downloaded from http://iai.asm.org/ on February 20, 2016 by Penn State Univ

72.

glycosylation in cystic fibrosis (CF): a review emphasizing the airway epithelial cell. Glycoconj. J. 18:649–659. Rogers, J., A. B. Dowsett, P. J. Dennis, J. V. Lee, and C. W. Keevil. 1994. Influence of temperature and plumbing material selection on biofilm formation and growth of Legionella pneumophila in a model potable water system containing complex microbial flora. Appl. Environ. Microbiol. 60:1585–1592. Rossier, O., J. Dao, and N. P. Cianciotto. 2008. The type II secretion system of Legionella pneumophila elaborates two aminopeptidases, as well as a metalloprotease that contributes to differential infection among protozoan hosts. Appl. Environ. Microbiol. 74:753–761. Ruoslahti, E., and M. D. Pierschbacher. 1987. New perspectives in cell adhesion: RGD and integrins. Science 238:491–497. Salmivirta, M., K. Lidholt, and U. Lindahl. 1996. Heparan sulfate: a piece of information. FASEB J. 10:1270–1279. Sampson, P. M., R. B. Boyd, G. G. Pietra, and A. P. Fishman. 1984. Glycosaminoglycan biosynthesis in the isolated perfused rat lung. J. Appl. Physiol. 57:1648–1654. Schwan, T. G., et al. 2005. Phylogenetic analysis of the spirochetes Borrelia parkeri and Borrelia turicatae and the potential for tick-borne relapsing fever in Florida. J. Clin. Microbiol. 43:3851–3859. Sheehan, K. B., J. M. Henson, and M. J. Ferris. 2005. Legionella species diversity in an acidic biofilm community in Yellowstone National Park. Appl. Environ. Microbiol. 71:507–511. Shevchenko, A., M. Wilm, O. Vorm, and M. Mann. 1996. Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels. Anal. Chem. 68: 850–858. Shin, S., and C. R. Roy. 2008. Host cell processes that influence the intracellular survival of Legionella pneumophila. Cell. Microbiol. 10:1209–1220. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol. 98:503–517. Stone, B. J., and Y. Abu Kwaik. 1998. Expression of multiple pili by Legionella pneumophila: identification and characterization of a type IV pilin gene and its role in adherence to mammalian and protozoan cells. Infect. Immun. 66:1768–1775. Stout, J. E., and V. L. Yu. 1997. Legionellosis. N. Engl. J. Med. 337:682–687. Takahashi, M., Y. Kuroki, K. Ohtsubo, and N. Taniguchi. 2009. Core fucose and bisecting GlcNAc, the direct modifiers of the N-glycan core: their functions and target proteins. Carbohydrate Res. 344:1387–1390.

Lcl IS AN L. PNEUMOPHILA GAG BINDING ADHESIN