当前位置: 首页 > 医学版 > 期刊论文 > 基础医学 > 感染与免疫杂志 > 2005年 > 第9期 > 正文
编号:11254082
Components of the Legionella pneumophila Flagellar Regulon Contribute to Multiple Virulence Traits, Including Lysosome Avoidance and Macroph
     Department of Microbiology and Immunology, University of Michigan Medical School, Ann Arbor, Michigan 48109-0620

    ABSTRACT

    Legionella pneumophila is a motile intracellular pathogen of macrophages and amoebae. When nutrients become scarce, the bacterium induces expression of transmission traits, some of which are dependent on the flagellar sigma factor FliA (28). To test how particular components of the L. pneumophila flagellar regulon contribute to virulence, we compared a fliA mutant with strains whose flagellar construction is disrupted at various stages. We find that L. pneumophila requires FliA to avoid lysosomal degradation in murine bone marrow-derived macrophages (BMM), to regulate production of a melanin-like pigment, and to regulate binding to the dye crystal violet, whereas motility, flagellar secretion, and external flagella or flagellin are dispensable for these activities. Thus, in addition to flagellar genes, the FliA sigma factor regulates an effector(s) or regulator(s) that contributes to other transmissive traits, notably inhibition of phagosome maturation. Whether or not the microbes produced flagellin, all nonmotile L. pneumophila mutants bound BMM less efficiently than the wild type, resulting in poor infectivity and a loss of contact-dependent death of BMM. Therefore, bacterial motility increases contact with host cells during infection, but flagellin is not an adhesin. When BMM contact by each nonmotile strain was promoted by centrifugation, all the mutants bound BMM similarly, but only those microbes that synthesized flagellin induced BMM death. Thus, the flagellar regulon equips the aquatic pathogen L. pneumophila to coordinate motility with multiple traits vital to virulence.

    INTRODUCTION

    Legionella pneumophila is a ubiquitous gram-negative motile bacterium that replicates inside protozoa in aquatic environments (27, 77, 86). When aerosols from contaminated water sources are inhaled by humans, L. pneumophila can also replicate within phagocytic alveolar macrophages (43), causing the life-threatening pneumonia Legionnaires' disease (60). Like many intracellular pathogens, L. pneumophila alternates between two distinct life cycle phases: an intracellular replicative phase and an infectious transmissive phase (reviewed in references 65 and 76). Transmissive microbes express a number of virulence-associated traits absent in replicative microbes, prominent among them flagellar production and subsequent motility.

    In addition to motility, two other transmissive-phase virulence traits are the ability to (i) isolate the microbial vacuole from the endosomal network of phagocytes to avoid lysosomal degradation (lysosome avoidance) and (ii) induce rapid, replication-independent macrophage death (cytotoxicity). One L. pneumophila structure essential for both traits is the type IV secretion system encoded by the dot/icm loci (11, 21, 51, 78, 82, 93). In addition, a link between flagellation and virulence-associated traits is well established (15, 17, 38, 72), and regulators that control flagellation also affect both lysosome avoidance and cytotoxicity (7, 36, 66; also see below). However, the specific mechanisms that link flagellation and virulence are not known and are the subject of this report.

    Motility is conferred by the flagellum, a large protein complex encoded by flagellar regulon genes whose temporal expression pattern is highly coordinated (Fig. 1) (reviewed in references 2 and 59). The ancestor of type III secretion systems, the flagellar secretion apparatus, exports the majority of flagellar components to the bacterial periplasm or external milieu, including rod, hook, flagellin, and capping proteins. The basal body stabilizes the flagellum, and the rotational forces generated by the Mot ion channels are transmitted from the basal body through the hook to the external filament, causing filament rotation. In addition to the structural genes necessary to form the flagellum, a number of regulators are dedicated to L. pneumophila flagellin (flaA) expression, and several of these also govern virulence (reviewed in references 40 and 65). These include the flagellar sigma factor FliA (36, 38, 39) and the LetA/S two-component system (29, 36, 58). LetA/S indirectly induces motility and a host of other transmission traits by inhibiting the global regulator CsrA, resulting in the expression of fliA and flaA (26, 65, 66). As in Pseudomonas and Vibrio sp. flagellar regulation (23, 71), the L. pneumophila master flagellar regulators are RpoN and the coactivator FleQ (48), whereas FliA (28) may independently govern several flagellar elements, including flaA expression.

    To establish a replication niche, many virulent bacteria utilize flagella at multiple stages of infection. Flagella not only mediate chemotaxis toward favorable environments but also contribute to various virulence traits, including microbial adhesion or invasion (3, 33, 47, 52, 80, 90) and toxin delivery (31, 54, 84, 100). Further, flagellar production is often coregulated with other virulence systems (reviewed in references 40 and 70). L. pneumophila coordinately regulates flagellation and virulence-associated traits, including intracellular replication. Flagella per se are dispensable for replication in macrophages or amoebae (24, 36, 48, 61). However, L. pneumophila requires the regulator FliA for motility, for intracellular growth in the social amoeba Dictyostelium discoideum (38), and for efficient infection and cytotoxicity in murine bone marrow-derived macrophages (BMM) (36, 38, 66). Accordingly, the flagellar regulon may directly influence lysosome avoidance and/or cytotoxicity, two early events during macrophage infection that are not easily detected solely by determining bacterial CFU. Although the flaA locus is one target of FliA induction (36, 38, 39), other targets remain unknown but likely include genes within the flagellar regulon such as flhB and fliD (reviewed in reference 40). Because fliA mutants have multiple defects, FliA may also regulate other genes outside the flagellar system. To test which flagellar components contribute to the L. pneumophila transmission phenotype, including lysosome avoidance and cytotoxicity, we systematically constructed and analyzed a group of mutants defective at various stages of flagellar development (Fig. 1).

    MATERIALS AND METHODS

    Bacterial strains, media, and reagents. L. pneumophila Lp02 (thyA hsdR rpsL; MB110), a virulent thymine auxotroph derived from the serogroup 1 clinical isolate Philadelphia 1 (10), was the parent for all mutant strains analyzed (Table 1). All mutants are indicated by both strain number and the lpg locus designation from the L. pneumophila Philadelphia 1 genome (20). L. pneumophila was cultured at 37°C in N-(2-acetamido)-2-aminoethanesulfonic acid (ACES; Sigma)-buffered yeast extract (AYE) broth in 5-ml aliquots on a rotating wheel or on ACES-buffered charcoal yeast extract (CYE) agar, both supplemented with 100 μg ml–1 thymidine (AYET and CYET, respectively). Bacteria from frozen stocks were streaked onto CYET plates with appropriate antibiotics, and <2-week-old colonies were cultured in broth overnight to mid-exponential phase, then subcultured in fresh AYET for an additional 16 to 24 h prior to experimentation. Exponential (E)-phase cultures were defined as having an optical density at 600 nm (OD600) of 0.4 to 2.0, and post-exponential (PE) cultures as having an OD600 of 3.5 to 4.5 and exhibiting full motility and a compact, coccoid shape. We noted that a minority of nonmotile wild-type (WT) E-phase microbes are at least partially flagellate (see Fig. 2), are cytotoxic to BMM (see Fig. 5), and avoid degradation in BMM (see Fig. 3C). Although plate-grown microbes are cultured in broth for >24 h to obtain synchronous E-phase cultures for analysis, a variable minority of WT E-phase microbes do maintain a subset of PE-phase transmission traits, including flagellation, perhaps from either their prior growth on agar or the lag phase in broth. Indeed, variability was observed for the cytotoxicity of WT E microbes, in part dependent on the optical density of the culture (Fig. 5B). Antibiotics were not used in broth culture, unless noted. To ascertain CFU, serial dilutions of bacteria were incubated on duplicate CYET plates for 4 days and resultant colonies counted. The proton ionophore carbonyl cyanide m-chlorophenylhydrazone (10 μM; Sigma) or the sodium-channel-blocking agent phenamil (0.8 mM, Sigma) was used to block motility by coincubation with microbes for 10 min.

    Mutant and plasmid construction and complementation. All strains and plasmids (Tables 1 and 2) were constructed using standard techniques, as described elsewhere (66). All cloning was done in the Escherichia coli strain DH5. At least two independent isolates of each L. pneumophila mutant were analyzed in all assays, but results for only one isolate are presented.

    To construct a flaA flagellin mutant (fliC, lpg1340), the flaA 1,900-bp genomic locus was amplified from Lp02 genomic DNA using primers flaAUP and flaADOWN, then ligated to create pGEM-flaA. A 65-bp N-terminal region of the flaA open reading frame (ORF) was removed by digestion with HindIII, and the remaining pGEM-flaA was blunted with Klenow fragment and dephosphorylated with calf intestinal phosphatase (New England Biolabs). The 1.3-kb kanamycin resistance cassette from pUC4K or the 1.7-kb gentamicin resistance cassette from pGEMT-Gent was purified by EcoRI digestion, blunted with Klenow fragment, and ligated into the digested pGEM-flaA plasmid to create pGEM-flaA-kan or pGEM-flaA-gent. The disrupted flaA locus was amplified by PCR (Roche; High Fidelity), transferred to L. pneumophila Lp02 by natural competence as described previously (8, 87), and selected for by antibiotic resistance. Multiple putative flaA::kan and flaA::gent mutants were screened by PCR for mutant-sized loci, and two independent isolates (MB534, MB532) were tested and found similar in all phenotypic assays; MB534 data are shown. To construct a flaA complementing plasmid that expressed FlaA from its native promoter, a 3.0-kb genomic region including flaA was amplified with primers flaA-Sac1 and flaA-Kpn1, by adding SacI and KpnI sites to the respective ends, and ligated into the multicloning site of digested pMMBGent-mob, creating pMMBGent-flaA. This plasmid was electroporated into flaA::kan (MB557) cells, then selected by gentamicin resistance.

    To construct a motAB motility ion channel mutant (lpg1780 to lpg1781), a 1.6-kb motAB genomic region, immediately 3' to the FliA ORF, was amplified with primers motA8510 and motB8511 and ligated to create pGEM-motAB. pGEM-motAB was digested with StuI and ClaI, dropping out a 129-bp fragment including the 3' terminus of motA and the 5' terminus of motB, blunted, and the 1.7-kb Gentr fragment was digested with EcoRI, blunted, and ligated to create pGEM-motAB-gent. Two independent isolates of this plasmid were obtained, with the Gentr fragment in opposite orientations. Each plasmid was used as a template for PCR, and the resulting PCR product was transferred to Lp02 by natural competence. Multiple motAB::gent mutants were isolated and verified as described above. Two independent motAB::gent isolates (MB560, MB561), one from each PCR product, were tested and found similar in all assays; MB560 data are shown. To complement the mutant, the 1.7-kb motAB ORF amplified with primers BamHI-motA5 and SalI-motB3 was ligated into the pMMB206mob plasmid to create the isopropyl--D-thiogalactopyranoside (IPTG)-inducible plasmid pMMB206-motAB, and the resultant plasmid was electroporated into the motAB::gent mutant to create MB562.

    To create a fliD flagellar cap mutant (Hap2, lpg1338), a 3.4-kb fliD genomic region was amplified by PCR with primers FliD_UP and FliD_DOWN and ligated to create pGEM-fliD. pGEM-fliD was digested with EcoRI to disrupt the fliD ORF; then either the Kanr or the Gentr cassette was ligated to create pGEM-fliD-kan or pGEM-fliD-gent. Multiple independent fliD mutants in Lp02 were isolated by natural competence and verified as described above, and two isolates (MB552, MB553) were tested in all assays and found similar; MB552 data are shown.

    To create the flhB flagellar secretion-defective mutant (lpg1786), a 3.4-kb flhB genomic region was amplified by PCR with primers flhB_UP and flhB_DOWN and ligated to create pGEM-flhB. To add an EcoRV site while simultaneously deleting a 800-bp portion of the ORF, we performed PCR from plasmid pGEM-flhB outward from the ORF with primers flhB-Int-Down-EcoRV and flhB-Int-Up, digested the PCR product with EcoRV, and religated to create pGEM-flhB-EcoRV. pGEM-flhB-EcoRV was digested with EcoRV, and either a blunted Kanr or Gentr cassette was ligated to create pGEM-flhB-gent or pGEM-flhB-kan. Multiple independent flhB mutants in Lp02 were isolated by natural competence and verified as described above, and a majority of assays were performed with two independent isolates (MB554, MB555); MB554 data are shown. It is possible that downstream ORFs were also disrupted in this strain, including the flagellar secretion gene flhA.

    A fliA (28) transposon mutant was previously isolated and characterized (36). To complement this mutant, we followed two approaches. First, a 940-bp EcoRI fragment from pGEM-fliA containing the FliA ORF and putative promoter was ligated to create the noninducible complementing vector pMMBGent-FliA, which was electroporated into the fliA mutant to create MB510. Second, a 750-bp region with the fliA ORF and ribosomal binding site was amplified with primers fliAUP2-EcoRI and fliADOWN-BamHI and ligated to create the IPTG-inducible vector pregFliA, which was then electroporated into the fliA mutant to create MB550.

    Macrophage cultures. BMM were isolated from the femur exudates of female A/J mice (Jackson Laboratory, Bar Harbor, Maine) and cultured in L-cell supernatant-conditioned medium for 1 week before replating onto 24- or 96 well plates as described elsewhere (66, 89). Macrophages were cultured in RPMI medium plus 10% heat-inactivated fetal bovine serum (FBS), a medium that does not support extracellular L. pneumophila growth.

    Quantification of pigment production, CV binding, sodium sensitivity, and heat resistance. The soluble melanin-like pigment produced during late-PE phase was measured by spectrophotometry (97). Broth cultures maintained in the PE phase for 2 to 4 days were centrifuged at 16,000 x g for 10 min; then the OD550 of culture supernatants was determined. Crystal violet (CV; Sigma) binding was quantified by taking aliquots of broth cultures, adjusting to an OD600 of 1.0 in 1-ml volumes, then incubating with 5 μg/ml of CV for 10 min at 37°C. After centrifugation for 5 min at 16,000 x g, supernatants were measured at OD600 and results expressed as the percentage of total CV remaining in the supernatant (not bound by microbes). PE cultures used for CV binding had an OD600 of >4.0 (mid- to late-PE phase), since inconsistent results were obtained with lower culture densities. Sodium sensitivity was calculated by plating 10-fold serial dilutions of broth cultures in phosphate-buffered saline (PBS) onto CYET agar with or without 100 mM NaCl, then enumerating CFU after a 5- to 6-day incubation, as described elsewhere (17, 66). Heat resistance was determined by incubating microbes for 20 min at 57°C and plating for CFU as described elsewhere (36).

    CFPs. To isolate a crude flagellar preparation (CFP) (81), 400 ml of a PE-phase broth culture of L. pneumophila was gently pelleted at 6,000 x g for 20 min at 4°C. Bacterial pellets were resuspended in 40 ml of PBS and vortexed vigorously for 5 min to dislodge surface structures, including flagella. Loss of motility was confirmed by microscopy where appropriate. Bacteria were centrifuged again at 6,000 x g for 20 min at 4°C, and the PBS supernatant containing flagellar fragments was passed through 0.45-μm filters and collected by centrifugation at 100,000 x g for 3 h at 4°C. CFPs were suspended in 500 μl of PBS, aliquoted, and stored at –20°C until use.

    Acid washing of microbes to remove flagella. To dissociate external flagella, we employed a brief acid wash as described previously (68). Microbes were collected by gentle centrifugation at 5,000 x g for 5 min, resuspended in PBS, pH 2.0, for 5 min at 37°C, and then washed twice with sterile PBS, pH 7.4. Complete loss of motility was confirmed by microscopy where appropriate, and experiments determined no loss of viability for every PE-phase L. pneumophila strain tested, as noted previously (66). Control experiments verified that PBS washes without acid did not affect motility or other virulence traits of WT PE-phase microbes (data not shown).

    Percentage of intact L. pneumophila organisms in BMM. The percentage of intact microbes after a 2-h incubation in macrophages was determined by immunofluorescence (IF) microscopy, as described previously, with minor modifications (8, 66). In brief, BMM were infected with microbes at a multiplicity of infection (MOI) of 1, incubated 10 min, washed three times to remove extracellular microbes, and then incubated for an additional 2 h. Fixation, antibody staining, and mounting were performed as described previously (66), and duplicate coverslips were scored for intact versus degraded L. pneumophila, as gauged by anti-Legionella staining of "intact" rod shapes versus dispersed or rounded "degraded" particles (66). Results from four or more experiments were expressed as means ± standard deviations (SD), except for flhB, which was analyzed twice. To detect microbial delivery to lysosomes, we prelabeled the lysosomal network with Texas Red-ovalbumin (TRov; 50 μg/ml; Molecular Probes); infection, fixation, and antibody straining were performed as described above, and colocalization between L. pneumophila staining and TRov was scored (49). To detect colocalization with the endosomal/lysosomal protein LAMP-1, fixed monolayers were incubated with rat monoclonal antibody 1D4B against LAMP-1 at 1:1,000 and subsequently with an Oregon Green-conjugated goat anti-rat antibody and visualized by IF microscopy as described elsewhere (49).

    Percentage of BMM bound. To determine the ability of microbes to contact and adhere to BMM over time, we determined the fraction bound (percent infected BMM). Infection of BMM in 24-well plates was initiated by centrifugation at 400 x g for 10 min; then cultures were incubated for an additional 10 min or 2 h. At the indicated time postinfection, coverslips were washed, fixed, and microbes enumerated with an anti-Legionella antibody by microscopy as described previously (66). Duplicate coverslips were scored by counting total BMM in five or more independent fields and the total anti-Legionella-stained particles in those fields, including both microbes that were intact and those that were degraded. The actual MOI, calculated by determining CFU of inoculates, varied from 0.5 to 5. To pool data from multiple experiments, results were expressed as mean percent BMM bound ± SD, and all data were normalized to an MOI of 1.

    Infectivity and intracellular growth. Infectivity is a gauge of the ability of L. pneumophila to bind, enter, and survive inside BMM during a 2-h incubation, as previously described (17, 66). In brief, BMM were infected with strains at an MOI of 1, with or without an initial centrifugation at 400 x g for 10 min, as indicated. Cells were incubated for 2 h at 37°C, washed three times with fresh RPMI, and lysed into 1 ml of PBS; then intracellular CFU were enumerated. Results are expressed as the percentage of the inoculated CFU recovered from lysed BMM at 2 h postinfection. To gauge intracellular growth, the pooled BMM supernatant and lysate were plated for CFU at various times postinfection, as described elsewhere (66).

    Quantifying flagellation and motility. In preliminary experiments, we determined that the fragile flagella are best preserved when microbes are internalized in BMM phagosomes. To determine flagellation of microbial strains, BMM on coverslips were briefly infected with microbes at an MOI of 10 by an initial centrifugation of 5 min at 400 x g and were fixed immediately thereafter. The primary mouse monoclonal antibody 2A5 against Legionella flagella (a gift of N. C. Engleberg, Ann Arbor, MI) was used at 1:100, and a secondary Oregon Green-conjugated anti-mouse antibody (Molecular Probes) was used at 1:1,000, as described elsewhere (17). Microbes were counted by 4',6'-diamidino-2-phenylindole (DAPI) staining, and colocalization with punctate, filamentous anti-flagellin staining was scored. Flagellin levels of crude flagellar preparations or whole microbes were also gauged by Western blot analysis, essentially as described elsewhere (48), using a 1:100 dilution of the antiflagellar monoclonal antibody 2A5. Whole microbes were normalized by optical density before Western blot analysis. We were unable to detect flagellin in concentrated supernatants from WT or fliD microbes, perhaps due to high flagellin levels required for detection by antibody 2A5. To gauge flagellation by transmission electron microscopy (TEM), broth-grown PE-phase microbes were fixed in 1% phosphotungstic acid and applied to grids for examination. To gauge motility qualitatively, microbes cultured to the PE phase in broth were examined by inverted microscopy as wet mounts, with relative motility levels based on observations of rapid, directed motion of many microbes.

    Macrophage cell death (cytotoxicity): Alamar Blue reduction. To quantify BMM viability after L. pneumophila infection, microbes were coincubated with BMM in RPMI-10% FBS in 96-well plates, with or without an initial centrifugation at 400 x g for 10 min as indicated. Duplicate wells were coincubated for an additional 1 h with twofold dilutions of microbes spanning a range of MOIs; then the medium was removed, RPMI-FBS plus 10% Alamar Blue (AccuMed) was added for 6 to 12 h, and viability was determined by spectrophotometer, as described previously (17, 35). The actual MOI of the inoculum was determined by enumerating CFU. Because L. pneumophila thymine auxotrophs were analyzed, the potential contribution of intracellular replication to cytotoxicity was eliminated by omitting exogenous thymidine.

    RESULTS

    Flagellar regulon mutants, flagellation, and motility. We first determined the levels of flagellation and motility for our panel of flagellar regulon mutants. As the transmissive- and replicative-phase prototypes in each assay, we used broth-grown WT PE and WT E L. pneumophila, respectively (17, 65, 88). WT PE microbes were highly motile and 75% had a single, monopolar flagellum, whereas WT E microbes were completely nonmotile and 10% flagellate (Fig. 2B), with minimal flagellin protein detected by Western blot analysis (Fig. 2D) (17). As an additional reference strain, we tested dotA mutants (MB460) that lack an integral component of the Dot/Icm type IV secretion system. dotA mutants were as motile and flagellate as WT PE organisms (Fig. 2). Both WT PE and dotA PE whole-microbe lysates and crude flagellar preparations possessed an abundant 48-kDa protein that was identified as flagellin by Western blot analysis (Fig. 2D and data not shown).

    To eliminate production of flagellin, we disrupted flaA, the gene that encodes the major component of the external flagellar filament (24, 37). As expected, the flaA mutant (MB534) was nonmotile, and its motility defect was fully complemented by the pMMBGent-FlaA vector. Further, flaA PE crude flagellar preparations and whole-microbe lysates lacked flagellin, and TEM revealed microbes with no external flagella (Fig. 2D and data not shown). As an alternate means of removing the external flagella, microbes were briefly washed with PBS (pH 2.0), a treatment that dissociates the external flagella but does not kill PE-phase microbes (data not shown) (68). Like flaA PE bacteria, WT PE acid-washed microbes were nonmotile and completely lacked external flagella (Fig. 2B), although Western blot analysis of whole-microbe lysates revealed detectable but reduced levels of total flagellin protein (Fig. 2D), which presumably represent an intracellular pool.

    To disrupt motility but not flagellin production or secretion, we constructed a mutant lacking FliD (Hap2), the flagellar capping protein required for assembly of secreted flagellin into functional flagella (45, 46, 53, 55, 83). fliD mutants (MB552) were completely nonmotile and also lacked intact external flagella (Fig. 2B and C), although moderate amounts of flagellin were detected in whole-microbe lysates (Fig. 2D). Further, fliD PE microbes produced flagellin at levels comparable to those of WT PE microbes; Western blot analysis of both acid-washed strains revealed similar flagellin levels (Fig. 2E). Rare fliD microbes whose surfaces stained diffusely with an anti-flagellin antibody were observed by microscopy, a pattern not observed for any other strain examined. In other bacterial species, removal of the flagellar cap does not disrupt flagellin production or secretion but instead perturbs its polymerization (46, 83), likely accounting for the continued production, secretion, and subsequent adhesion of flagellin to the surfaces of L. pneumophila fliD mutants.

    To block flagellar secretion, we disrupted the integral component of the flagellar secretion apparatus, FlhB (62, 63), or the required flagellar ATPase, FliI (61, 91). Because flagellar secretion is required for building the rod, hook, and flagellar filament (Fig. 1) (reviewed in reference 13), flhB (MB554) and fliI (MB564) mutants were nonmotile and nonflagellate, validating the likely impairment of flagellar secretion (Fig. 2 and data not shown), as described previously for the L. pneumophila fliI mutant (61). Nevertheless, flagellar secretion-deficient strains of L. pneumophila accumulated flagellin protein, since the flaA promoter is still highly active in mutants that lack FliI or FlhB (B. Hammer and A. Molofsky, unpublished), and Western blot analysis detected flagellin associated with whole-microbe lysates (Fig. 2D). When the secretion mutants were first acid washed to determine intracellular protein levels, their flagellin levels were only modestly decreased relative to those of WT or fliD microbes (Fig. 2E). Thus, fliI and flhB mutants contain flagellin that is neither secreted by flagellar secretion nor assembled into a filament.

    In an attempt to disrupt motility without eliminating external flagella, we mutated motAB (MB560), the locus that encodes the ion channels required to drive flagellar motion (25, 102). L. pneumophila has two sets of nonhomologous motAB genes, similar to the arrangement in Pseudomonas aeruginosa (25); we disrupted the lpg1780-to-lpg1781 operon adjacent to fliA. As expected, only 1 to 5% of motAB microbes displayed directed motion, and expression of exogenous motA and motB from the pMMB206-motAB plasmid restored motility to wild-type levels. However, only 25% of the mutant cells bore intact flagella, and whole-microbe flagellin levels were modestly decreased compared with those of WT PE or dotA PE microbes (Fig. 2D). Likewise, both motility and flagellation were significantly decreased when L. pneumophila was treated with either of two drugs reported to block flagellar motion but not assembly by other bacteria, namely, the proton ionophore carbonyl cyanide m-chlorophenylhydrazone or the sodium channel-blocking agent phenamil (5, 34, 41, 67). Therefore, both genetic and pharmacological data indicate that nonmotile flagella are not as stable as motile flagella on L. pneumophila under the conditions tested. Here the motAB mutant is a tool to examine the impact of an intermediate level of flagellation coupled with a very low level of motility; since a significant percentage of the mutant microbes retain intact but immobile flagella, the strain allows the contributions of motility and flagella to other traits to be separated.

    To discern the functions of the two primary flagellar regulators previously implicated in virulence, we analyzed mutants that lack the two-component response regulator LetA or the flagellar sigma factor FliA (29, 36, 37, 58). Both letA (MB413) and fliA (MB410) mutants lack flaA promoter activity and flagellin protein and are completely nonflagellate and nonmotile (Fig. 2) (36, 39), defects that are partially complemented by the exogenous expression of pMMBGent-LetA or pMMBGent-FliA, respectively (36) (data not shown). Using this panel of flagellar regulon mutants (Fig. 1), we examined the contributions of external flagella, flagellin protein, motility, flagellar secretion, and flagellar regulators to L. pneumophila virulence-associated traits.

    Motility, but not flagellin protein, contributes prominently to infectivity toward BMM. An intracellular pathogen infects a host cell in several steps: (i) the host cell is contacted, (ii) the microbe adheres intimately, (iii) the microbe is engulfed, and (iv) the microbe survives intracellularly. As a sum of these steps, we measure the infectivity of L. pneumophila by incubating BMM with microbes at an MOI of 1 for 2 h, then enumerating the CFU of intracellular microbes as a percentage of the initial inoculum. A correlation exists between the transmissive-phase traits of motility and infectivity (17, 36, 66). Here we specifically tested the contributions of particular flagellar regulon genes to each step of BMM infection.

    We first examined binding to BMM microscopically, an assay that measures only (i) microbial contact and (ii) intimate adherence. At 10 min postinfection, motile WT PE microbes associated with BMM slightly better than all nonmotile strains (Fig. 3A). In contrast, by 2 h, motile WT PE microbes bound BMM 6-fold more often than all nonmotile strains (Fig. 3A). At either a low or a high MOI, no differences in BMM binding or total infectivity were detected for nonmotile strains, whether the cells lacked external flagellin protein (flaA PE and acid-washed WT PE microbes), failed to polymerize flagellin (fliD), or formed a nonfunctional flagellum (motAB) (Fig. 3A; also data not shown). If polymerized flagellin promoted BMM binding, we would likely detect increased binding by the motAB mutants, since 25% of the microbes remain flagellate (Fig. 2). Thus, we conclude, in agreement with the results of others (24), that flagellin does not contribute significantly to BMM adhesion. Instead, motility likely promotes numerous BMM contacts during the 2-h infection period.

    A similar pattern was observed when infectivity was measured by enumerating cell-associated CFU after a 2-h incubation. Even when cell contact was promoted by initiating the assay with a mild centrifugation, nonmotile microbes did not infect as efficiently as the WT (Fig. 3B) (data not shown). Consistent with previous results (17), motile WT PE microbes were 15% infectious, a proportion similar to that of motile dotA PE microbes (data not shown), whereas all nonmotile strains tested were <5% infectious. Although nonmotile strains (acid-washed WT PE microbes, flaA, fliI, motAB, and fliD) infected BMM poorly, the microbes that did enter and survive the initial infection replicated efficiently, as did several nonmotile, nonflagellate strains described previously: WT E, letA PE, fliI PE, and fliA PE organisms (36, 61, 66; A. Molofsky, unpublished). Thus, motility but not flagellin likely promotes contacts with BMM and increases infectivity, but an intact flagellum, flagellar secretion, or motility is not required for subsequent microbial replication.

    FliA promotes lysosome avoidance in BMM independently of the flagellar regulon. L. pneumophila avoids degradation in macrophage lysosomes, a trait integral to its survival and pathogenesis in humans (42, 88). Therefore, we examined how efficiently each flagellar regulon mutant avoided lysosomal degradation. By scoring the morphology of each intracellular microbe by IF microscopy, trafficking was analyzed independently of BMM binding efficiency (see Materials and Methods) (66). After a 2-h incubation, approximately 80% of intracellular WT PE microbes were intact, as judged by their rod shape, whereas only 20% of WT E microbes were intact (Fig. 3C), confirming the effect of growth phase on lysosome avoidance (17). Similarly to letA mutants (36), fliA PE mutants were degraded twice as frequently as WT PE controls (Fig. 3C), a phenotype largely complemented by expression of pMMBGent-FliA (data not shown). fliA mutants also colocalized with lysosomal TRov (49) five times more frequently than did WT PE microbes (data not shown). Thus, the flagellar sigma factor FliA induces expression of factors that protect L. pneumophila from degradation by macrophages.

    To ascertain how FliA contributes to lysosome avoidance, we tested other flagellar regulon mutants, including the known or likely targets of FliA regulation: flaA, fliD, and flhB (40). Compared to the fliA or letA mutants, significantly more of each of the other intracellular nonmotile microbes were intact, including acid-washed WT PE, flaA, fliD, flhB, and motAB bacteria, although each nonmotile strain was degraded slightly more frequently than WT PE microbes (Fig. 3C). Further, flaA and fliD PE mutants avoid acquisition of the late endosomal/lysosomal protein LAMP-1 with an efficiency similar to that of WT PE microbes (data not shown). Thus, intact flagella, flagellin protein, and flagellar secretion are not required for lysosome avoidance; motility may play a minor role. In contrast, the flagellar sigma factor FliA (28) is required for proper trafficking. Therefore, L. pneumophila has coopted FliA to regulate genes beyond the flagellar regulon, namely, virulence factors that promote lysosome avoidance.

    FliA regulates melanin-like pigment production and crystal violet binding independently of the flagellar regulon. To investigate whether additional transmission traits are regulated by FliA, we next assayed pigment production. As it reaches the transmissive PE phase, L. pneumophila produces a soluble, melanin-like pigment that is believed to protect the microbe from UV damage in nature but is dispensable for efficient infection and replication in eukaryotic cells (95, 97, 98). There is a strong correlation between expression of transmissive-phase virulence traits and pigment production: the letA, letS, letE, and relA transmission mutants are all defective for pigmentation (7; reviewed in references 65, 66, and 103). In contrast, fliA (28) mutants are hyperpigmented (Fig. 4A) (66), a phenotype significantly suppressed when fliA expression is restored by the pregFliA plasmid (data not shown). To discern if the effect of FliA on pigmentation is dependent on its induction of other flagellar genes, we quantified the soluble pigment produced by the panel of strains. External flagella/flagellin, flagellar secretion, motility, and type IV secretion did not influence pigmentation: flaA, fliD, flhB, fliI, motAB, and dotA single mutants and flaA fliD (MB567) and fliI fliD (MB565) double mutants pigmented similarly to WT PE microbes (Fig. 4A; also data not shown). Further, flagellar secretion did not contribute to hyperpigmentation by the fliA mutant; a fliA flhB (MB573) double mutant maintained excess pigmentation (data not shown).

    To test if FliA regulates other nonflagellar properties, we measured the propensity of broth-grown L. pneumophila to bind the dye crystal violet (14), a surface-dependent property that is elevated in E-phase relative to PE-phase microbes (E. Fernandez-Moreira, unpublished) (Fig. 4B). letA and fliA PE-phase mutants bind excess CV compared with WT PE microbes, although to different degrees (Fig. 4B). Like the pigmentation phenotype, CV binding was not dependent on external flagella/flagellin, flagellar secretion, motility, or type IV secretion: PE-phase flaA, fliD, fliI, motAB, and dotA mutants bound CV similarly to WT PE microbes (Fig. 4B). Further, removal of flagella and other surface proteins by acid washing of WT PE microbes did not significantly alter microbial CV binding (data not shown). Since pigment production/secretion, crystal violet binding, and lysosomal avoidance are not dependent on the flagellar regulon, FliA must regulate these traits independently of its control of flagellation and motility.

    The flagellar regulon does not influence heat resistance or type IV secretion-dependent sodium sensitivity. As additional transmissive-phase phenotypes, we tested if FliA and the flagellar regulon impacted L. pneumophila resistance to heat or sensitivity to sodium chloride. Heat resistance and salt sensitivity are dependent on both the PE phase and regulators of transmission, including LetA and LetE (7, 66), whereas salt sensitivity also requires the Dot/Icm type IV secretion system (17, 79, 94). The flagellar regulon was not required for heat tolerance; fliA, flaA, and fliI mutants were as heat resistant as WT PE microbes (data not shown). Since the flagellar pore secretes proteins, we reasoned it may also be a conduit for small ions, thereby conferring sodium sensitivity. As reported previously, WT E microbes or PE-phase mutants lacking the type IV secretion gene dotA, the response regulator letA, or the alternate sigma factor rpoS (MB379) were defective for sodium sensitivity, efficiently forming colonies on agar that contained 100 mM sodium (Fig. 4C; also data not shown) (8, 36). Like WT PE microbes, all PE flagellar regulon mutants tested were sensitive to sodium, including fliA, flaA, fliD, fliI, and motAB. Thus, unlike the type IV Dot/Icm secretion machinery, the type III flagellar secretion apparatus does not confer sodium sensitivity. Furthermore, defects in the flagellar regulon do not appear to impact type IV secretion, as judged by sodium sensitivity, endosomal avoidance, and intracellular growth (Fig. 4C and data not shown).

    Legionella induction of BMM cell death requires motility and flagellin. As L. pneumophila becomes motile, it expresses cytolytic activity against red blood cells and macrophages. Since cytotoxicity is contact dependent (44, 51), we determined whether the poor BMM binding of the nonmotile strains (Fig. 3A) affected their cytotoxicity. As predicted, the WT E, flaA PE, fliD PE, and fliA PE nonmotile strains were completely noncytotoxic at all MOIs tested, whereas motile WT PE microbes were highly cytotoxic to BMM (Fig. 5A; supplemental Fig. 1C; data not shown). When contact of the nonmotile strains with BMM was promoted by mild centrifugation, both WT E and fliD PE microbes killed significantly more BMM; nevertheless, flaA microbes remained noncytotoxic at all culture densities tested (Fig. 5A and B), a defect fully complemented by exogenous expression of pMMBGent-FlaA (Fig. 5A). Additional evidence indicated a requirement for flagellin in BMM cell death: all nonmotile strains tested that lacked flagellin were similarly defective for BMM cytotoxicity, including PE-phase letA, fliA, flaA, and acid-washed WT microbes (A. Molofsky, unpublished). The nonmotile, flagellin-producing motAB, flhB, and fliI strains displayed intermediate levels of BMM killing, slightly greater than that of WT E microbes (A. Molofsky, unpublished). Further, the degree of BMM killing correlated with growth phase-dependent flagellin expression by L. pneumophila cultured in broth (Fig. 5B) (66). Therefore, BMM cytotoxicity was absolutely dependent not only on motility to initiate host-cell contacts, but also on the production of flagellin protein.

    DISCUSSION

    In nature, L. pneumophila is prevalent in fresh-water sources, where it replicates after infecting single-celled protozoa, a lifestyle likely dependent on motility. Indeed, a vast majority of Legionella species express flagellin (FlaA) and are motile (15, 37, 69), indicating that this trait confers fitness in aquatic environments. Flagellate L. pneumophila organisms are also observed in lung tissue of patients with Legionnaires' disease (18), where the microbes can infect and replicate inside alveolar macrophages. L. pneumophila is not flagellate during intracellular replication but becomes so during the transmissive phase, when the microbes lyse their spent host and search for a new replication niche (17, 65, 76). Here we report that components of the flagellar regulon also contribute to several virulence-associated transmissive traits, namely, macrophage binding, lysosome avoidance, pigment production, surface modifications, and cytotoxicity.

    The flagellar sigma factor FliA (28) and the response regulator LetA promote efficient lysosome avoidance in macrophages, the hallmark of L. pneumophila pathogenesis. LetA likely contributes to lysosome avoidance by relieving CsrA repression of FliA expression (6, 66). Motility appears to play a minor role in lysosome avoidance (Fig. 3C), perhaps by orienting a polar type IV secretion apparatus toward the BMM (9, 75) or by promoting contact and subsequent binding of microbes with active transmission programs. However, FliA also induces lysosome avoidance independently of its targets in the flagellar regulon, as evidenced by the fact that fliA mutants are degraded more frequently than other nonmotile strains (Fig. 3C). Indeed, FliA contributes to at least two other transmission traits that are independent of motility, namely, pigmentation (Fig. 4A) and crystal violet binding (Fig. 4B). Likewise, FliA regulates microbial growth in the social amoeba D. discoideum independently of the flaA locus (38). Accordingly, the classic definition of FliA as a flagellum-specific sigma factor is not accurate for L. pneumophila. FliA must cooperate with other regulators to govern the cohort of transmission genes required for efficient lysosome avoidance, since both letA and fliA mutants evade degradation more frequently than replicative-phase WT E microbes do (Fig. 3C). Identification of the targets of FliA regulation, including those outside the flagellar regulon, will yield insight into how this pathogen avoids destruction in macrophages, sentinels of innate immunity.

    FliA apparently induces lysosome avoidance by a mechanism that is coordinated with, but distinct from, the Dot/Icm effectors that immediately isolate L. pneumophila from the endosomal network. First, all flagellar regulon mutants (fliA, flaA, fliD, and fliI) have a functional Dot/Icm secretion system, as gauged by their replication in BMM (data not shown) and sodium sensitivity (Fig. 4C). Second, >50% of fliA PE mutants are degraded in BMM, whereas dotA mutants remain intact in a late endosomal vacuole (data not shown) (49). Finally, unlike dotA mutants, fliA and letA cells have altered crystal violet binding, an assay that reflects surface properties (14) and correlates to evasion of lysosomes (E. Fernandez Moreira and M. Swanson, unpublished). Therefore, by a mechanism that requires FliA but is independent of type IV secretion, L. pneumophila modifies or exports to its surface virulence factors that delay or alter its delivery to degradative lysosomes (49). One potential target of FliA regulation is either synthesis or modification of the pathogen's unusually hydrophobic lipopolysaccharide (32, 57, 101), a process known to be coregulated with flagellation in E. coli (56).

    The flagellar regulon also promotes binding to host cells (Fig. 3A), a trait vital to both infection and cytotoxicity. Flagellin itself does not significantly mediate BMM adherence; all nonmotile strains bind BMM similarly to WT PE microbes at 10 min postinfection, whether or not they produce or polymerize flagellin (Fig. 2D and 3A). Instead, the occurrence of fewer contacts with BMM during the 2-h infection period likely accounts for the poor binding (Fig. 3A) and infectivity (Fig. 3B) of nonmotile bacteria. Once L. pneumophila has bound to macrophages, its motility, flagellin, and flagellar secretion apparatus are largely dispensable for either survival (Fig. 3C) or replication (data not shown). Thus, our study corroborates and extends the previous report that L. pneumophila flaA mutants are defective for early stages of infection of amoebae and macrophages (24).

    A hallmark of Legionnaires' disease is extensive inflammation with lysis of inflammatory cells and necrosis of the alveolar epithelium, a pathology that has been linked in mice to a bacterial cytotoxin(s) (4, 12, 16, 50). The selective advantage for L. pneumophila cytotoxicity was thought to be escape from amoebae (30, 64), though egress may also occur by vacuolar exocytosis mediated by two Dot/Icm substrates, LepA and LepB (19). In vitro, lysis of red blood cells and certain eukaryotic hosts, including amoebae and macrophages, is mediated by the contact- and type IV secretion system-dependent pore-forming activity of L. pneumophila (4, 30, 44, 51). Several other L. pneumophila factors (ProA, Lly, or PlaB) can promote hemolysis, but mutations in each cause little or no defect in macrophage cytotoxicity (28, 73, 99). In contrast, the RtxA protein is partially required for macrophage cytotoxicity, by promoting either the pore-forming activity or macrophage adhesion (22). Our analysis expands the limited list of L. pneumophila factors required for cytotoxicity against macrophages to include flagellin.

    There are several plausible models by which flagellin could act in concert with type IV secretion to promote BMM cell death. Flagellin does not contribute directly to pore formation, as flagellin is dispensable for lysis of red blood cells (A. Molofsky and C. Madigan, unpublished). Instead, flagellin protein could promote cytotoxicity by orienting the pathogen's type IV secretion system toward the host cell, as both the type IV secretion system and flagellum are likely located at its poles (9). However, this model seems unlikely, since unlike dot/icm mutants, flagellin-deficient strains traffic and replicate in BMM as efficiently as WT microbes (Fig. 3C and data not shown). A number of observations are consistent with the idea that flagellin could target L. pneumophila to regions of the BMM surface that are susceptible to either delivery or activity of its cytotoxin(s): Cohorts of toxins are cholesterol dependent and active in "lipid rafts," microdomains rich in glycosphingolipids, GM1 gangliosides, and cholesterol (reviewed in reference 92); transmissive L. pneumophila enters macrophages at lipid-raft-rich regions (96); and Pseudomonas aeruginosa flagellin binds asialo-GM1 ganglioside (1), a component of lipid rafts. It is also possible that flagellin itself triggers BMM death, since Pseudomonas syringae flagellin elicits a hypersensitive reaction that kills tomato cells, a death that is enhanced with a fliD mutant that hypersecretes flagellin (83). However, since CFPs were not cytotoxic to BMM, and purified flagellin failed to complement the defective cytotoxicity of nonflagellate mutants (data not shown), we hypothesize a requirement for type IV secretion to function either directly, to deliver flagellin to the host cell cytoplasm, or indirectly, to create a vacuolar pore by which excess surface flagellin might leak intracellularly. Understanding of the interplay between microbial virulence factors and macrophage defense mechanisms can be obtained by testing these models directly.

    In addition to knowledge of how components of the L. pneumophila flagellar regulon contribute to virulence, we also gained insight into their coordinate regulation. Since inactive flagella (motAB PE) are somewhat unstable on L. pneumophila, either nonmotile flagella are more susceptible to physical stress or a feedback loop may promote the shedding of nonfunctional flagella. Secondly, flagellar secretion is not absolutely required for FliA function; flhB or fliI secretion mutants are not defective for pigmentation, crystal violet binding, or lysosome avoidance (Fig. 3C and 4), a situation that differs from flagellar regulation by enteric bacteria (reviewed in reference 2). Secretion mutants (flhB, fliI) as well as the capping mutant (fliD) do contain less total flagellin than whole WT PE microbes (Fig. 2D), but this likely reflects the large amount of external polymerized flagellin of WT PE microbes. In support of this hypothesis, when microbes were acid washed to remove external flagellin (Fig. 2D and E), WT PE microbes had levels of flagellin similar to those of both the capping mutant (fliD) and the motility mutant (motAB), and only slightly greater than those of the secretion mutants (flhB, fliI).

    L. pneumophila produces a pigment in the transmissive PE phase (65), a trait sensitive to FliA activity. Unlike the letA regulatory mutant, which is defective for expression of all transmission traits tested, fliA mutants are as sodium sensitive (Fig. 4C) and heat resistant as the WT, but they are hyperpigmented (Fig. 4A). Strikingly, excess pigmentation was also observed when wild-type or mutant strains were cultured either in a stringent chemically defined medium (74), or with gentamicin for plasmid maintenance, or in the absence of csrA expression (data not shown) (66). We hypothesize that a variety of metabolic stresses induce pigment production, perhaps due to activation of gluconeogenic pathways that synthesize aromatic amino acids, the precursors of the melanin-like pigment (85). Since fliA mutants grow similarly to the WT in broth culture, most likely some accessory metabolic pathway, such as aromatic amino acid synthesis, accounts for hyperpigmentation.

    In summary, our analysis demonstrates that the L. pneumophila flagellar regulon genes impact multiple virulence traits in different manners. Motility promotes contact with host cells, the prelude to subsequent pathogen-phagocyte interactions, including lysosome avoidance or host cell death (cytotoxicity). In addition to its role in motility, the flagellar sigma factor FliA is required for lysosome avoidance, pigmentation, and crystal violet binding by mechanisms that are independent of flagellin production, secretion, or assembly. Finally, flagellin protein itself directly or indirectly promotes the death of macrophages. Thus, flagellar regulation has evolved to promote not only motility but also key elements of L. pneumophila pathogenesis, thereby providing the intracellular pathogen with a mechanism to coordinate expression of transmission traits.

    ACKNOWLEDGMENTS

    We thank J. D. Sauer for scientific discussions, M. O'Riordan for insight and guidance, especially during the sabbatical of M.S.S., and C. A. Madigan for excellent experimental assistance and discussions. Also thanks to David Friedman and his lab for sharing equipment and reagents.

    This project was supported by NIH grant AI 44212-01. A.B.M. was supported by NIH National Research Service Award 56-T32-GM07544 from the National Institute of General Medical Sciences and a University of Michigan Frederick G. Novy Fellowship.

    REFERENCES

    1. Adamo, R., S. Sokol, G. Soong, M. I. Gomez, and A. Prince. 2004. Pseudomonas aeruginosa flagella activate airway epithelial cells through asialoGM1 and toll-like receptor 2 as well as toll-like receptor 5. Am. J. Respir. Cell Mol. Biol. 30:627-634.

    2. Aldridge, P., and K. T. Hughes. 2002. Regulation of flagellar assembly. Curr. Opin. Microbiol. 5:160-165.

    3. Allen-Vercoe, E., and M. J. Woodward. 1999. The role of flagella, but not fimbriae, in the adherence of Salmonella enterica serotype Enteritidis to chick gut explant. J. Med. Microbiol. 48:771-780.

    4. Alli, O. A. T., L.-Y. Gao, L. L. Pedersen, S. Zink, M. Radulic, M. Doric, and Y. Abu Kwaik. 2000. Temporal pore formation-mediated egress from macrophages and alveolar epithelial cells by Legionella pneumophila. Infect. Immun. 68:6431-6440.

    5. Atsumi, T., S. Sugiyama, E. J. Cragoe, Jr., and Y. Imae. 1990. Specific inhibition of the Na+-driven flagellar motors of alkalophilic Bacillus strains by the amiloride analog phenamil. J. Bacteriol. 172:1634-1639.

    6. Bachman, M. A., and M. S. Swanson. 2004. Genetic evidence that Legionella pneumophila RpoS modulates expression of the transmission phenotype in both the exponential phase and the stationary phase. Infect. Immun. 72:2468-2476.

    7. Bachman, M. A., and M. S. Swanson. 2004. The LetE protein enhances expression of multiple LetA/LetS-dependent transmission traits by Legionella pneumophila. Infect. Immun. 72:3284-3293.

    8. Bachman, M. A., and M. S. Swanson. 2001. RpoS co-operates with other factors to induce Legionella pneumophila virulence in the stationary phase. Mol. Microbiol. 40:1201-1214.

    9. Bardill, J. P., J. L. Miller, and J. P. Vogel. 2005. IcmS-dependent translocation of SdeA into macrophages by the Legionella pneumophila type IV secretion system. Mol. Microbiol. 56:90-103.

    10. 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.

    11. Berger, K. H., J. J. Merriam, and R. I. Isberg. 1994. Altered intracellular targeting properties associated with mutations in the Legionella pneumophila dotA gene. Mol. Microbiol. 14:809-822.

    12. Blackmon, J. A., M. D. Hicklin, and F. W. Chandler. 1978. Legionnaires' disease. Pathological and historical aspects of a ‘new’ disease. Arch. Pathol. Lab. Med. 102:337-343.

    13. Blocker, A., K. Komoriya, and S. Aizawa. 2003. Type III secretion systems and bacterial flagella: insights into their function from structural similarities. Proc. Natl. Acad. Sci. USA 100:3027-3030.

    14. Bohach, G. A., and I. S. Snyder. 1983. Characterization of surfaces involved in adherence of Legionella pneumophila to Fischerella species. Infect. Immun. 42:318-325.

    15. Bosshardt, S. C., R. F. Benson, and B. S. Fields. 1997. Flagella are a positive predictor for virulence in Legionella. Microb. Pathog. 23:107-112.

    16. Brieland, J., P. Freeman, R. Kunkel, C. Chrisp, M. Hurley, J. Fantone, and C. Engleberg. 1994. Replicative Legionella pneumophila lung infection in intratracheally inoculated A/J mice. Am. J. Pathol. 145:1537-1546.

    17. Byrne, B., and M. S. Swanson. 1998. Expression of Legionella pneumophila virulence traits in response to growth conditions. Infect. Immun. 66:3029-3034.

    18. Chandler, F. W., B. M. Thomason, and G. A. Hebert. 1980. Flagella on Legionnaires' disease bacteria in the human lung. Ann. Intern. Med. 93:715-716.

    19. Chen, J., K. S. de Felipe, M. Clarke, H. Lu, O. R. Anderson, G. Segal, and H. A. Shuman. 2004. Legionella effectors that promote nonlytic release from protozoa. Science 303:1358-1361.

    20. Chien, M., I. Morozova, S. Shi, H. Sheng, J. Chen, S. M. Gomez, G. Asamani, K. Hill, J. Nuara, M. Feder, J. Rineer, J. J. Greenberg, V. Steshenko, S. H. Park, B. Zhao, E. Teplitskaya, J. R. Edwards, S. Pampou, A. Georghiou, I. C. Chou, W. Iannuccilli, M. E. Ulz, D. H. Kim, A. Geringer-Sameth, C. Goldsberry, P. Morozov, S. G. Fischer, G. Segal, X. Qu, A. Rzhetsky, P. Zhang, E. Cayanis, P. J. De Jong, J. Ju, S. Kalachikov, H. A. Shuman, and J. J. Russo. 2004. The genomic sequence of the accidental pathogen Legionella pneumophila. Science 305:1966-1968.

    21. Christie, P. J., and J. P. Vogel. 2000. Bacterial type IV secretion: conjugation systems adapted to deliver effector molecules to host cells. Trends Microbiol. 8:354-360.

    22. 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.

    23. Dasgupta, N., M. C. Wolfgang, A. L. Goodman, S. K. Arora, J. Jyot, S. Lory, and R. Ramphal. 2003. A four-tiered transcriptional regulatory circuit controls flagellar biogenesis in Pseudomonas aeruginosa. Mol. Microbiol. 50:809-824.

    24. Dietrich, C., K. Heuner, B. C. Brand, J. Hacker, and M. Steinert. 2001. Flagellum of Legionella pneumophila positively affects the early phase of infection of eukaryotic host cells. Infect. Immun. 69:2116-2122.

    25. Doyle, T. B., A. C. Hawkins, and L. L. McCarter. 2004. The complex flagellar torque generator of Pseudomonas aeruginosa. J. Bacteriol. 186:6341-6350.

    26. Fettes, P. S., V. Forsbach-Birk, D. Lynch, and R. Marre. 2001. Overexpresssion of a Legionella pneumophila homologue of the E. coli regulator csrA affects cell size, flagellation, and pigmentation. Int. J. Med. Microbiol. 291:353-360.

    27. Fields, B. S. 1996. The molecular ecology of legionellae. Trends Microbiol. 4:286-290.

    28. Flieger, A., K. Rydzewski, S. Banerji, M. Broich, and K. Heuner. 2004. Cloning and characterization of the gene encoding the major cell-associated phospholipase A of Legionella pneumophila, plaB, exhibiting hemolytic activity. Infect. Immun. 72:2648-2658.

    29. Gal-Mor, O., and G. Segal. 2003. The Legionella pneumophila GacA homolog (LetA) is involved in the regulation of icm virulence genes and is required for intracellular multiplication in Acanthamoeba castellanii. Microb. Pathog. 34:187-194.

    30. Gao, L. Y., and Y. A. Kwaik. 2000. The mechanism of killing and exiting the protozoan host Acanthamoeba polyphaga by Legionella pneumophila. Environ. Microbiol. 2:79-90.

    31. Ghelardi, E., F. Celandroni, S. Salvetti, D. J. Beecher, M. Gominet, D. Lereclus, A. C. Wong, and S. Senesi. 2002. Requirement of flhA for swarming differentiation, flagellin export, and secretion of virulence-associated proteins in Bacillus thuringiensis. J. Bacteriol. 184:6424-6433.

    32. Girard, R., T. Pedron, S. Uematsu, V. Balloy, M. Chignard, S. Akira, and R. Chaby. 2003. Lipopolysaccharides from Legionella and Rhizobium stimulate mouse bone marrow granulocytes via Toll-like receptor 2. J. Cell Sci. 116:293-302.

    33. Giron, J. A., A. G. Torres, E. Freer, and J. B. Kaper. 2002. The flagella of enteropathogenic Escherichia coli mediate adherence to epithelial cells. Mol. Microbiol. 44:361-379.

    34. Gosink, K. K., and C. C. Hase. 2000. Requirements for conversion of the Na+-driven flagellar motor of Vibrio cholerae to the H+-driven motor of Escherichia coli. J. Bacteriol. 182:4234-4240.

    35. Hammer, B. K., and M. S. Swanson. 1999. Co-ordination of Legionella pneumophila virulence with entry into stationary phase by ppGpp. Mol. Microbiol. 33:721-731.

    36. Hammer, B. K., E. S. Tateda, and M. S. Swanson. 2002. A two-component regulator induces the transmission phenotype of stationary-phase Legionella pneumophila. Mol. Microbiol. 44:107-118.

    37. Heuner, K., L. Bender-Beck, B. C. Brand, P. C. Luck, K. H. Mann, R. Marre, M. Ott, and J. Hacker. 1995. Cloning and genetic characterization of the flagellum subunit gene (flaA) of Legionella pneumophila serogroup 1. Infect. Immun. 63:2499-2507.

    38. Heuner, K., C. Dietrich, C. Skriwan, M. Steinert, and J. Hacker. 2002. Influence of the alternative 28 factor on virulence and flagellum expression of Legionella pneumophila. Infect. Immun. 70:1604-1608.

    39. Heuner, K., J. Hacker, and B. C. Brand. 1997. The alternative sigma factor 28 of Legionella pneumophila restores flagellation and motility to an Escherichia coli fliA mutant. J. Bacteriol. 179:17-23.

    40. Heuner, K., and M. Steinert. 2003. The flagellum of Legionella pneumophila and its link to the expression of the virulent phenotype. Int. J. Med. Microbiol. 293:133-143.

    41. Heytler, P. G., and W. W. Prichard. 1962. A new class of uncoupling agents—carbonyl cyanide phenylhydrazones. Biochem. Biophys. Res. Commun. 7:272-275.

    42. Horwitz, M. A. 1983. The Legionnaires' disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J. Exp. Med. 158:2108-2126.

    43. Horwitz, M. A., and S. C. Silverstein. 1980. Legionnaires' disease bacterium (Legionella pneumophila) multiplies intracellularly in human monocytes. J. Clin. Investig. 66:441-450.

    44. Husmann, L. K., and W. Johnson. 1994. Cytotoxicity of extracellular Legionella pneumophila. Infect. Immun. 62:2111-2114.

    45. Ikeda, T., K. Oosawa, and H. Hotani. 1996. Self-assembly of the filament capping protein, FliD, of bacterial flagella into an annular structure. J. Mol. Biol. 259:679-686.

    46. Ikeda, T., S. Yamaguchi, and H. Hotani. 1993. Flagellar growth in a filament-less Salmonella fliD mutant supplemented with purified hook-associated protein 2. J. Biochem. (Tokyo) 114:39-44.

    47. Inglis, T. J., T. Robertson, D. E. Woods, N. Dutton, and B. J. Chang. 2003. Flagellum-mediated adhesion by Burkholderia pseudomallei precedes invasion of Acanthamoeba astronyxis. Infect. Immun. 71:2280-2282.

    48. Jacobi, S., R. Schade, and K. Heuner. 2004. Characterization of the alternative sigma factor 54 and the transcriptional regulator FleQ of Legionella pneumophila, which are both involved in the regulation cascade of flagellar gene expression. J. Bacteriol. 186:2540-2547.

    49. Joshi, A. D., S. Sturgill-Koszycki, and M. S. Swanson. 2001. Evidence that Dot-dependent and -independent factors isolate the Legionella pneumophila phagosome from the endocytic network in mouse macrophages. Cell Microbiol. 3:99-114.

    50. Katz, S. M., and S. Hahemi. 1982. Electron microscopic examination of the inflammatory response to Legionella pneumophila in guinea pigs. Lab. Investig. 46:24-32.

    51. Kirby, J. E., J. P. Vogel, H. L. Andrews, and R. R. Isberg. 1998. Evidence of pore-forming ability by Legionella pneumophila. Mol. Microbiol. 27:323-336.

    52. Kirov, S. M., M. Castrisios, and J. G. Shaw. 2004. Aeromonas flagella (polar and lateral) are enterocyte adhesins that contribute to biofilm formation on surfaces. Infect. Immun. 72:1939-1945.

    53. Komoriya, K., N. Shibano, T. Higano, N. Azuma, S. Yamaguchi, and S. I. Aizawa. 1999. Flagellar proteins and type III-exported virulence factors are the predominant proteins secreted into the culture media of Salmonella typhimurium. Mol. Microbiol. 34:767-779.

    54. Konkel, M. E., J. D. Klena, V. Rivera-Amill, M. R. Monteville, D. Biswas, B. Raphael, and J. Mickelson. 2004. Secretion of virulence proteins from Campylobacter jejuni is dependent on a functional flagellar export apparatus. J. Bacteriol. 186:3296-3303.

    55. Kutsukake, K., Y. Ohya, and T. Iino. 1990. Transcriptional analysis of the flagellar regulon of Salmonella typhimurium. J. Bacteriol. 172:741-747.

    56. Landini, P., and A. J. Zehnder. 2002. The global regulatory hns gene negatively affects adhesion to solid surfaces by anaerobically grown Escherichia coli by modulating expression of flagellar genes and lipopolysaccharide production. J. Bacteriol. 184:1522-1529.

    57. Luneberg, E., N. Zetzmann, D. Alber, Y. A. Knirel, O. Kooistra, U. Zahringer, and M. Frosch. 2000. Cloning and functional characterization of a 30 kb gene locus required for lipopolysaccharide biosynthesis in Legionella pneumophila. Int. J. Med. Microbiol. 290:37-49.

    58. Lynch, D., N. Fieser, K. Gloggler, V. Forsbach-Birk, and R. Marre. 2003. The response regulator LetA regulates the stationary-phase stress response in Legionella pneumophila and is required for efficient infection of Acanthamoeba castellanii. FEMS Microbiol. Lett. 219:241-248.

    59. Macnab, R. M. 2003. How bacteria assemble flagella. Annu. Rev. Microbiol. 57:77-100.

    60. McDade, J. E., C. C. Shepard, D. W. Fraser, T. R. Tsai, M. A. Redus, and W. R. Dowdle. 1977. Legionnaires' disease: isolation of a bacterium and demonstration of its role in other respiratory diseases. N. Engl. J. Med. 297:1197-1203.

    61. 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.

    62. Minamino, T., T. Iino, and K. Kutuskake. 1994. Molecular characterization of the Salmonella typhimurium flhB operon and its protein products. J. Bacteriol. 176:7630-7637.

    63. Minamino, T., and R. M. Macnab. 1999. Components of the Salmonella flagellar export apparatus and classification of export substrates. J. Bacteriol. 181:1388-1394.

    64. Molmeret, M., O. A. Alli, M. Radulic, M. Susa, M. Doric, and Y. A. Kwaik. 2002. The C-terminus of IcmT is essential for pore formation and for intracellular trafficking of Legionella pneumophila within Acanthamoeba polyphaga. Mol. Microbiol. 43:1139-1150.

    65. Molofsky, A. B., and M. S. Swanson. 2004. Differentiate to thrive: lessons from the Legionella pneumophila life cycle. Mol. Microbiol. 53:29-40.

    66. Molofsky, A. B., and M. S. Swanson. 2003. Legionella pneumophila CsrA is a pivotal repressor of transmission traits and activator of replication. Mol. Microbiol. 50:445-461.

    67. Nishihara, T., and E. Freese. 1975. Motility of Bacillus subtilis during growth and sporulation. J. Bacteriol. 123:366-371.

    68. Okuda, S., R. Igarashi, Y. Kusui, Y. Kasahara, and H. Morisaki. 2003. Electrophoretic mobility of Bacillus subtilis knockout mutants with and without flagella. J. Bacteriol. 185:3711-3717.

    69. Ott, M., P. Messner, J. Heesemann, R. Marre, and J. Hacker. 1991. Temperature-dependent expression of flagella in Legionella. J. Gen. Microbiol. 137:1955-1961.

    70. Ottemann, K. M., and J. F. Miller. 1997. Roles for motility in bacterial-host interactions. Mol. Microbiol. 24:1109-1117.

    71. Prouty, M. G., N. E. Correa, and K. E. Klose. 2001. The novel 54- and 28-dependent flagellar gene transcription hierarchy of Vibrio cholerae. Mol. Microbiol. 39:1595-1609.

    72. Pruckler, J. M., R. F. Benson, M. Moyenuddin, W. T. Martin, and B. S. Fields. 1995. Association of flagellum expression and intracellular growth of Legionella pneumophila. Infect. Immun. 63:4928-4932.

    73. Quinn, F. D., and L. S. Tompkins. 1989. Analysis of a cloned sequence of Legionella pneumophila encoding a 38 kD metalloprotease possessing haemolytic and cytotoxic activities. Mol. Microbiol. 3:797-805.

    74. Ristroph, J. D., K. W. Hedlund, and S. Gowda. 1981. Chemically defined medium for Legionella pneumophila growth. J. Clin. Microbiol. 13:115-119.

    75. Rohde, M., J. Puls, R. Buhrdorf, W. Fischer, and R. Haas. 2003. A novel sheathed surface organelle of the Helicobacter pylori cag type IV secretion system. Mol. Microbiol. 49:219-234.

    76. Rowbotham, T. J. 1986. Current views on the relationships between amoebae, legionellae and man. Isr. J. Med. Sci. 22:678-689.

    77. Rowbotham, T. J. 1980. Preliminary report on the pathogenicity of Legionella pneumophila for freshwater and soil amoebae. J. Clin. Pathol. 33:1179-1183.

    78. Roy, C. R., and R. R. Isberg. 1997. Topology of Legionella pneumophila DotA: an inner membrane protein required for replication in macrophages. Infect. Immun. 65:571-578.

    79. Sadosky, A. B., L. A. Wiater, and H. A. Shuman. 1993. Identification of Legionella pneumophila genes required for growth within and killing of human macrophages. Infect. Immun. 61:5361-5373.

    80. Scharfman, A., S. K. Arora, P. Delmotte, E. Van Brussel, J. Mazurier, R. Ramphal, and P. Roussel. 2001. Recognition of Lewis x derivatives present on mucins by flagellar components of Pseudomonas aeruginosa. Infect. Immun. 69:5243-5248.

    81. Scherer, D. C., I. DeBuron-Connors, and M. F. Minnick. 1993. Characterization of Bartonella bacilliformis flagella and effect of antiflagellin antibodies on invasion of human erythrocytes. Infect. Immun. 61:4962-4971.

    82. Segal, G., M. Purchell, and H. A. Shuman. 1998. Host cell killing and bacterial conjugation require overlapping sets of genes within a 22-kb region of the Legionella pneumophila genome. Proc. Natl. Acad. Sci. USA 95:1669-1674.

    83. Shimizu, R., F. Taguchi, M. Marutani, T. Mukaihara, Y. Inagaki, K. Toyoda, T. Shiraishi, and Y. Ichinose. 2003. The fliD mutant of Pseudomonas syringae pv. tabaci, which secretes flagellin monomers, induces a strong hypersensitive reaction (HR) in non-host tomato cells. Mol. Genet. Genomics 269:21-30.

    84. Song, Y. C., S. Jin, H. Louie, D. Ng, R. Lau, Y. Zhang, R. Weerasekera, S. Al Rashid, L. A. Ward, S. D. Der, and V. L. Chan. 2004. FlaC, a protein of Campylobacter jejuni TGH9011 (ATCC 43431) secreted through the flagellar apparatus, binds epithelial cells and influences cell invasion. Mol. Microbiol. 53:541-553.

    85. Steinert, M., M. Flugel, M. Schuppler, J. H. Helbig, A. Supriyono, P. Proksch, and P. C. Luck. 2001. The Lly protein is essential for p-hydroxyphenylpyruvate dioxygenase activity in Legionella pneumophila. FEMS Microbiol. Lett. 203:41-47.

    86. Steinert, M., U. Hentschel, and J. Hacker. 2002. Legionella pneumophila: an aquatic microbe goes astray. FEMS Microbiol. Rev. 26:149-162.

    87. Stone, B. J., and Y. Abu Kwaik. 1999. Natural competence for DNA transformation by Legionella pneumophila and its association with expression of type IV pili. J. Bacteriol. 181:1395-1402.

    88. Swanson, M. S., and B. K. Hammer. 2000. Legionella pneumophila pathogenesis: a fateful journey from amoebae to macrophages. Annu. Rev. Microbiol. 54:567-613.

    89. Swanson, M. S., and R. R. Isberg. 1995. Association of Legionella pneumophila with the macrophage endoplasmic reticulum. Infect. Immun. 63:3609-3620.

    90. Tasteyre, A., M. C. Barc, A. Collignon, H. Boureau, and T. Karjalainen. 2001. Role of FliC and FliD flagellar proteins of Clostridium difficile in adherence and gut colonization. Infect. Immun. 69:7937-7940.

    91. Thomas, J., G. P. Stafford, and C. Hughes. 2004. Docking of cytosolic chaperone-substrate complexes at the membrane ATPase during flagellar type III protein export. Proc. Natl. Acad. Sci. USA 101:3945-3950.

    92. Tweten, R. K., M. W. Parker, and A. E. Johnson. 2001. The cholesterol-dependent cytolysins. Curr. Top. Microbiol. Immunol. 257:15-33.

    93. Vogel, J. P., H. L. Andrews, S. K. Wong, and R. R. Isberg. 1998. Conjugative transfer by the virulence system of Legionella pneumophila. Science 279:873-876.

    94. Vogel, J. P., C. Roy, and R. R. Isberg. 1996. Use of salt to isolate Legionella pneumophila mutants unable to replicate in macrophages. Ann. N. Y. Acad. Sci. 797:271-272.

    95. Warren, W. J., and R. D. Miller. 1979. Growth of Legionnaires' disease bacterium (Legionella pneumophila) in chemically defined medium. J. Clin. Microbiol. 10:50-55.

    96. Watarai, M., I. Derre, J. Kirby, J. D. Growney, W. F. Dietrich, and R. R. Isberg. 2001. Legionella pneumophila is internalized by a macropinocytotic uptake pathway controlled by the Dot/Icm system and the mouse Lgn1 locus. J. Exp. Med. 194:1081-1096.

    97. Wiater, L. A., A. B. Sadosky, and H. A. Shuman. 1994. Mutagenesis of Legionella pneumophila using Tn903 dlllacZ: identification of a growth-phase-regulated pigmentation gene. Mol. Microbiol. 11:641-653.

    98. Wintermeyer, E., M. Flugel, M. Ott, M. Steinert, U. Rdest, K. H. Mann, and J. Hacker. 1994. Sequence determination and mutational analysis of the lly locus of Legionella pneumophila. Infect. Immun. 62:1109-1117.

    99. Wintermeyer, E., U. Rdest, B. Ludwig, A. Debes, and J. Hacker. 1991. Characterization of legiolysin (lly), responsible for hemolytic activity, colour production, and fluorescence of Legionella pneumophila. Mol. Microbiol. 5:1135-1143.

    100. Young, G. M., D. H. Schmiel, and V. L. Miller. 1999. A new pathway for the secretion of virulence factors by bacteria: the flagellar export apparatus functions as a protein-secretion system. Proc. Natl. Acad. Sci. USA 96:6456-6461.

    101. Zahringer, U., Y. A. Knirel, B. Lindner, J. H. Helbig, A. Sonesson, R. Marre, and E. T. Rietschel. 1995. The lipopolysaccharide of Legionella pneumophila serogroup 1 (strain Philadelphia 1): chemical structure and biological significance. Prog. Clin. Biol. Res. 392:113-139.

    102. Zuberi, A. R., C. W. Ying, H. M. Parker, and G. W. Ordal. 1990. Transposon Tn917lacZ mutagenesis of Bacillus subtilis: identification of two new loci required for motility and chemotaxis. J. Bacteriol. 172:6841-6848.

    103. Zusman, T., O. Gal-Mor, and G. Segal. 2002. Characterization of a Legionella pneumophila relA insertion mutant and roles of RelA and RpoS in virulence gene expression. J. Bacteriol. 184:67-75.(A. B. Molofsky, L. M. She)