Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell Microbiol. Author manuscript; available in PMC 2009 March 1.
Published in final edited form as:
PMCID: PMC2579332

Legionella pneumophila EnhC is required for efficient replication in tumor necrosis factor α-stimulated macrophages


Legionella pneumophila enhC mutants were originally identified as being defective for uptake into host cells. In this work, we found that the absence of EnhC resulted in defective intracellular growth when dissemination of intracellular bacteria to neighboring cells was expected to occur. No such defect was observed during growth within the amoeba Dictyostelium discoideum Culture supernatants containing the secreted products of infected macrophages added to host cells restricted the growth of the ΔenhC strain, while tumor necrosis factor α (TNF-α), at concentrations similar to those found in macrophage culture supernatants, could reproduce the growth restriction exerted by culture supernatants on L. pneumophila ΔenhC. The absence of EnhC also caused defective trafficking of the Legionella-containing vacuole in TNF-α treated macrophages. EnhC was shown to be an envelope-associated protein largely localized to the periplasm, with its expression induced in post-exponential phase, as is true for many virulence-associated proteins. Furthermore, the absence of EnhC appeared to affect survival under stress conditions, as the ΔenhC mutant was more susceptible to H2O2 treatment than the wild type strain. EnhC, therefore, is a unique virulence factor that is required for growth specifically when macrophages have heightened potential to restrict microbial replication.


Legionella pneumophila is a Gram-negative intracellular pathogen that is the causative agent of Legionnaires’ pneumonia (Burnsed et al., 2007; McDade et al., 1977). The major natural reservoir for L. pneumophila is freshwater amoebae (Fields et al., 1984; Wadowsky et al., 1988). The disease in humans is initiated by inhalation of contaminated water sources, and progresses as a result of intracellular replication of L. pneumophila in alveolar macrophages (Horwitz and Silverstein, 1980; McDade et al., 1977). Replication of the microorganism during the disease process can be reproduced in culture, as the microorganism can replicate in a wide array of both phagocytic and normally nonphagocytic mammalian cells (Yamamoto et al., 1988; Daisy et al., 1981; Horwitz and Silverstein, 1980).

The cellular events required for the intracellular growth of L. pneumophila have been characterized in detail at the microscopic level (Payne and Horwitz, 1987; Horwitz, 1984; Nash et al., 1984; Horwitz and Silverstein, 1981). After L. pneumophila entry into host cells, bacteria reside in a membrane-bound compartment, called the Legionella containing vacuole (LCV) (Horwitz, 1983b). The LCV acquires mitochondria, ribosomes and small vesicles derived from the endoplasmic reticulum (ER) until it transforms into a rough ER-like compartment in which L. pneumophila initiates replication (Tilney et al., 2001; Horwitz, 1983b). During this process, the LCV avoids association with the endocytic network, as observed by the lack of association of this compartment with the late endosomal/lysosomal protein marker LAMP-1 (Clemens et al., 2000; Roy et al., 1998; Horwitz, 1983a). The ability of the LCV to evade fusion with the lysosome is crucial for the replication of L. pneumophila in the hostile host cell. The first round of replication in the LCV is completed 16–24 hours after initial uptake, resulting in release of bacteria into the culture medium and initiation of a second round of infection in neighboring cells. Most of the characterized mutants of L. pneumophila altered in intracellular growth rates exhibit their defects within a few hours of the initiation of replication (Laguna et al., 2006; Vincent and Vogel, 2006; Bardill et al., 2005; Ninio et al., 2005; Sexton et al., 2004a; VanRheenen et al., 2004; Andrews et al., 1998; Segal et al., 1998; Berger and Isberg, 1993).

The L. pneumophila Dot/Icm proteins, which form a Type IVB protein translocation system, are required for biogenesis of the replication vacuole and intracellular replication (VanRheenen et al., 2004; Andrews et al., 1998; Segal et al., 1998; Vogel et al., 1998; Segal and Shuman, 1997; Vogel et al., 1996; Berger and Isberg, 1993). An impressive array of bacterial protein substrates of the Dot/Icm system have been characterized, and there may be over 100 such proteins translocated to host cells (Machner and Isberg, 2006; Murata et al., 2006; Weber et al., 2006; Derre and Isberg, 2005; Chen et al., 2004; Luo and Isberg, 2004; Conover et al., 2003; Nagai et al., 2002; Huang and Isberg, unpublished data). Although Dot/Icm is absolutely required for intracellular replication, most of the substrates appear dispensable for intracellular growth.

A genetic screen for L. pneumophila proteins that enhance uptake into normally nonphagocytic cells identified a number of genes that may modulate this process (Cirillo et al., 2000). One of these “enhanced entry” (enh) loci was enhC (lpg2639; Chien et al., 2004). An inframe deletion of this gene in the L. pneumophila AA100 strain results in decreased entry into a pair of cell lines (Cirillo et al., 2000). A second screen also identified enhC based on an independent phenotype, in which insertion mutations were isolated that resulted in L. pneumophila mutants having low plating efficiency in the presence of overproduced Dot/Icm protein DotA (Conover et al., 2003). This second screen, which resulted in the identification of several genes encoding proteins involved in membrane integrity, suggested that EnhC may be involved in biogenesis of some components of the cell envelope rather than being a ligand for host cell recognition.

In response to Legionella infections in either cell cultures or animals, host cells will produce proinflammatory factors including tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), interferon-γ (IFN-γ), interleukin-6 (IL-6) and interleukin-12 (IL-12) (Losick and Isberg, 2006; McHugh et al., 2000b; Brieland et al., 1998; Susa et al., 1998; Skerrett and Martin, 1996; Brieland et al., 1995; Matsiota-Bernard et al., 1993; Blanchard et al., 1987). Among these cytokines, TNF-α appears to be pivotal for the activation of phagocytes and the resolution of pneumonic infection (Skerrett et al., 1997; Blanchard et al., 1989; Blanchard et al., 1988). Treating genetically susceptible A/J mouse macrophages in vitro with TNF-α makes the macrophages more resistant to Legionella infection (McHugh et al., 2000c). Furthermore, the inhibitory effects of other cytokines, such as IL-12 and IFN-γ, on the replication of L. pneumophila are at least partially dependent on TNF-α (Brieland et al., 1998; Skerrett and Martin, 1996; Matsiota-Bernard et al., 1993). Given the capacity of TNF-α to restrict intracellular growth of the microorganism, L. pneumophila may have proteins that allow it to tolerate this aspect of host defense. Here we report that EnhC is one such protein, as mutants lacking this protein show specific defects in intracellular replication when there is enhanced TNF-α production by macrophages.


The absence of EnhC results in a delayed defect in intracellular growth

The enhC mutant was isolated in a screen for L. pneumophila mutants that showed low viability in the presence of overproduced DotA (Conover et al., 2003). An in-frame deletion of enhC in LP02 (a derivative of the Philadelphia-1 isolate) that expresses wild type levels of DotA, however, showed levels of viability indistinguishable from the parental strain LP02, so this ΔenhC strain was directly tested for uptake by primary bone marrow-derived macrophages (BMDM) from A/J mice (Fig. 1). Using our standard low multiplicity of infection (MOI) conditions and the gentamicin protection assay, there was no uptake defect observed for ΔenhC at 2 hours post-infection (hpi) (Fig. 1; Experimental Procedures). As the previous published uptake assays of enhC mutant used immortalized HEp-2 cells (Cirillo et al., 2000), uptake into the HEp-2 cell line was analyzed at a variety of MOI, as well. Once again, no defect was observed for ΔenhC, at MOIs ranging from 5 to 100 bacteria per cell (Fig. 1). We conclude that the absence of EnhC has no consequence on uptake of L. pneumophila Philadelphia-1 under our infection conditions.

Fig. 1
EnhC is dispensable for uptake of Legionella pneumophila Philadelphia-1 by host cells

To determine if loss of EnhC alters infection kinetics at later time points, L. pneumophila ΔenhC was introduced onto BMDM monolayers for 72 hours and its replication efficiency was monitored relative to the wild type LP02 strain. The ΔenhC strain grew at a similar rate compared to wild type during the first 24 hours incubation (Fig. 2A). During the next 24 hours period, however, there was a five-fold reduction in the replication rate of the ΔenhC strain relative to wild type. This later period corresponded to the time at which bacteria released after an initial round of intracellular replication are predicted to disseminate to neighboring cells and undergo a second round of intracellular growth. Plasmid-encoded EnhC could complement the growth defect of the mutant, although there was lower uptake by both this plasmid-harboring strain as well as a vector control strain relative to plasmid-free strains (Fig. 2B).

Fig. 2
EnhC is required for persistent replication within macrophages, but not Dictyostelium discoideum

The restriction of intracellular growth that occurs during the second day of growth was not observed when the amoeba Dictyostelium discoideum was challenged with the ΔenhC mutant (Fig. 2C). Within amoebae, the growth of the mutant was indistinguishable from the parental strain throughout 72 hours incubation. Therefore, there may be restrictive conditions that arise within macrophages after a 24 hr. replication period that are not apparent during growth within this amoebal species.

EnhC is required for efficient growth in restrictive macrophages

To determine the reason for the defective intracellular growth of the ΔenhC mutant in BMDM, a number of models were tested. Presumably, the defective replication during the second day of growth could be due to: 1) delayed lysis after the first round of intracellular replication; 2) reduced replication potential of the ΔenhC strain after it is released from the first round of intracellular replication; or 3) changes in the macrophage monolayers as a result of L. pneumophila infection that make the macrophages selectively restrictive to the mutant.

To test the delayed lysis model, ΔenhC and wild type LP02 were incubated with macrophage monolayers for 24 hours to allow one round of intracellular growth and liberation from macrophages. Bacteria that remained macrophage-associated or found in the culture supernatant were then assessed for colony forming units (CFU). The absence of EnhC had no effect on liberation of bacteria from macrophages, as the fraction of bacteria that remained macrophage-associated was unaffected in the ΔenhC mutant (Fig. 3A).

Fig. 3
EnhC is required for growth in macrophages that have been previously exposed to L. pneumophila

We next determined if a round of intracellular growth rendered the ΔenhC strain less competent than wild type at establishing further rounds of growth within macrophages. To this end, the bacterial progeny released after the first round of intracellular replication were collected from the culture supernatant of the infected monolayers at 24 hpi and applied to freshly prepared macrophage monolayers that never encountered bacteria. Intracellular growth was then monitored over the next 72 hours to test if the ΔenhC progeny from the first round were defective relative to the wild type progeny in initiating and maintaining a new round of replication. The result of this reinfection assay (Fig. 3B) indicated that the ΔenhC progeny liberated by macrophages had the same potential as wild type in terms of establishing a new round of growth during the first 24 hours (Fig. 3B). On the other hand, after 24 hours, a strong block in replication was observed again for the ΔenhC strain. This suggests that changes in the nature of the infected monolayer over time may attenuate intracellular growth of the mutant, and it is unlikely that one round of intracellular growth makes the mutant less competent than wild type at establishing new growth. Therefore, the monolayer appears to be restrictive for the mutant after 24 hours of replication.

The intracellular growth curves showed apparent resumption of replication of the ΔenhC mutant after 48 hours of incubation with macrophages (Fig. 2 and Fig. 3B). This could be explained by a restriction at 24 hours that results in a pause in replication of the mutant. Alternatively, there may exist a fraction of the population of the ΔenhC mutant that is incapable of initiating replication when the monolayer becomes restrictive, with the replication-competent fraction growing at kinetics that are nearly identical to the wild type strain. The following experiments are consistent with the latter explanation, as restriction appears to result in replicating and non-replicating populations of the ΔenhC mutant.

To provide evidence that macrophage monolayers become selectively restrictive for the ΔenhC strain after 24 hours of infection, a super-infection assay was performed. BMDM were incubated with either the wild type or the mutant first (called “pre-infection”). 24 hours later, the monolayers were challenged with GFP-expressing strains that harbor either wild type or ΔenhC alleles (called “super-infection”). The super-infections were allowed to proceed for 14 hours, fixed and the number of GFP-expressing bacteria in individual LCV was determined by microscopic observation (Fig. 3C). In macrophage monolayers pre-infected with L. pneumophila, the percentage of vacuoles containing single GFP ΔenhC bacterium was increased two-fold compared to vacuoles containing single GFP LP02 bacterium (Fig. 3C). Furthermore, the percentages of medium sized (5~10 bacteria) and large sized vacuoles (11~20 bacteria) harboring the GFP ΔenhC bacteria were correspondingly decreased. These results were independent of the strain used in the pre-infection, as monolayers pre-infected with either LP02 or ΔenhC were equally effective at restricting GFP ΔenhC (Fig. 3C). The defective replication of the strain lacking EnhC observed in the monolayers pre-infected with L. pneumophila did not occur in the freshly prepared monolayers that had never encountered bacteria (Fig. 3D). This difference indicates that loss of EnhC results in enhanced sensitivity to the restrictive macrophage monolayers generated by prior exposure to L. pneumophila, with a fraction of the population unable to initiate replication in the presence of restrictive conditions.

Soluble factors from infected monolayers contribute to restriction of intracellular growth by ΔenhC

The restriction generated by pre-infection did not seem to require that the newly introduced bacteria and bacteria from pre-infection enter into identical macrophages, as many of the small vacuoles containing single GFP-bacterium were present in macrophages without bacteria from pre-infection (data not shown). The restriction conferred by infected monolayers might be due to some transmissible factors.

To test whether the restriction observed in infected monolayers was transmissible, macrophage culture supernatants from infected monolayers were tested for their ability to compromise intracellular growth of ΔenhC in freshly prepared macrophages that never encountered bacteria (called “naïve” macrophages). Culture supernatants were collected from infected monolayers (MOI = 0.05 for 24 hours) and then subjected to filtration to remove bacteria in the supernatants. Culture supernatants generated in this fashion were called “conditioned media.” Naïve macrophages pretreated with media conditioned by infections with either LP02 or the ΔenhC mutant strain were able to selectively restrict growth of the ΔenhC strain but not LP02 (Fig. 4A). Conditioned media from macrophages infected with either the ΔenhC strain or LP02 were equally effective at restriction (Fig. 4A). To confirm that conditioned medium decreased the growth rate in the absence of EnhC, the spectrum of LCV sizes was determined in monolayers pretreated with conditioned medium (Fig. 4B). Compared to wild type LP02, the distribution of vacuoles containing the mutant was skewed away from large vacuoles (>20 bacteria) (Fig. 4B). Once again, the source of the conditioned medium was not important, as culture supernatants from macrophages infected with either wild type or the ΔenhC strain were equally competent at blocking intracellular growth of the mutant (Fig. 4B).

Fig. 4
EnhC is required for intracellular growth in the presence of conditioned medium from infected macrophage monolayers

Since conditioned medium was generated after filtration (Experimental Procedures), it implied that soluble factors, such as soluble cytokines, might play a role in transmitting the restriction from infected monolayers to naïve monolayers. It should be pointed out that the conditioned medium did not restrict the growth of the mutant to the same extent as was observed in standard growth curves (compare Fig. 4 to Fig. 2), suggesting that additional cell-associated, matrix or unstable factors may contribute to the growth restriction.

EnhC is associated with resistance of L. pneumophila to restriction conferred by TNF-α treatment of macrophages

As pro-inflammatory cytokines including TNF-α, IL-6, and IL-1β have been reported to be secreted by macrophages challenged by L. pneumophila (McHugh et al., 2000b; Arata et al., 1993; Blanchard et al., 1987), cytokine concentrations were determined in the conditioned media. As most of the growth restriction experiments (Fig. 2 and Fig. 4) were performed under low MOI conditions to avoid the cytotoxicity effect of L. pneumophila (MOI = 0.05), cytokine levels were determined under similar conditions. Detectable signals for tumor necrosis factor-α (TNF-α) and macrophage chemotactic protein-1 (MCP-1) were observed in the conditioned media from macrophages infected with either LP02 or the ΔenhC strain for 24 hours (Fig. 5A). Levels of these cytokines were almost identical for media from macrophages challenged by these two strains (Fig. 5A), as expected from the previous restriction experiments (Fig. 3D and Fig. 4). The levels of four other cytokines-interleukin12 p70 (IL-12p70), IL-10, IL-6 and interferon-γ (IFN-γ) were not above the background levels (Fig. 5A; dashed line). In addition, there was no detectable IL-1β in the conditioned media (data not shown). Although IL-6 and IL-1β have been reported in macrophage culture supernatants in response to L. pneumophila infections (McHugh et al., 2000a; McHugh et al., 2000c; Skerrett and Martin, 1996; Arata et al., 1993; Blanchard et al., 1987), such responses were observed using MOIs that are 20–200 times higher than used in our experiments.

Fig. 5
EnhC is required for efficient intracellular growth in the presence of TNF-α

As MCP-1 and TNF-α were detected in conditioned medium, each was added to naïve macrophage monolayers to determine if they caused enhanced restriction of the ΔenhC mutant relative to LP02. The addition of recombinant mouse MCP-1 to BMDM at a variety of concentrations ranging from 0 ng/ml to 700 ng/ml resulted in no restriction of either L. pneumophila strain during intracellular replication (data not shown). In contrast, at low concentrations recombinant mouse TNF-α interfered with growth of the ΔenhC strain (Fig. 5B). The defect observed in the mutant could be rescued completely by a plasmid-encoded EnhC (Fig. 5C). As the effective concentration of TNF-α was very close to the concentration of TNF-α detected in the conditioned media (70 pg/ml), and the extent of restriction of the mutant was similar to that observed with conditioned media (Fig. 4A), much of the restriction observed with conditioned media may be due to the presence of TNF-α secreted in response to L. pneumophila (McHugh et al., 2000c; McHugh et al., 2000b; Skerrett and Martin, 1996).

EnhC is required for efficient avoidance of the endocytic pathway in TNF-α treated macrophages

Formation of the L. pneumophila replication vacuole (LCV) requires bypass of entry into the endocytic pathway, as assayed by lack of colocalization of the LCV with the late endosome/lysosome protein LAMP-1 (Roy et al., 1998). To determine if loss of EnhC affects formation of the LCV in the presence of TNF-α, macrophages incubated in the presence or absence of TNF-α were challenged with L. pneumophila for one hour and colocalization with LAMP-1 was detected by immunofluorescence microscopy. In untreated naïve macrophages, the absence of EnhC had no effect on colocalization of LAMP-1 with the LCV (Figs. 6A and 6B). In contrast, the addition of TNF-α to naïve macrophages resulted in a 3–4 fold increase in the percentage of vacuoles containing the ΔenhC strain that colocalized with LAMP-1 (Fig. 6B). TNF-α caused no such increase in colocalization with LAMP-1 for vacuoles bearing LP02 (Fig. 6B). The results obtained for the addition of conditioned medium were almost identical to those observed for TNF-α treatment of macrophages (Figs. 6B and 6C). These results are consistent with the model that the presence of TNF-α in conditioned medium restricts the replication vacuole formation by bacteria lacking EnhC.

Fig. 6
EnhC is required for efficient bypass of the endocytic pathway in restrictive macrophages

As the L. pneumophila Dot/Icm protein translocation system is required for LCV formation (Roy et al., 1998), it is possible that the absence of EnhC reduces the translocation efficiency of Dot/Icm system. This was not the case, at least for naïve macrophages, as the translocation efficiencies of several known substrates (SidC, LidA and SidM) were not affected by loss of EnhC (data not shown). Furthermore, loss of EnhC did not affect the translocation efficiency of SidC in macrophages treated with TNF-α (data not shown). Although it is possible that EnhC is required for the translocations of specific substrates in restrictive macrophages, the absence of EnhC does not appear to cause wholesale defects on Dot/Icm system. Alternatively, EnhC may be a Dot/Icm substrate, and its absence could cause the TNF-α-specific effects. Immunofluorescence staining of infected macrophages and purified phagosomes showed no evidence for translocation of EnhC by Dot/Icm system (data not shown) and, as will be shown below, the localization of EnhC in L. pneumophila is inconsistent with this latter model.

Post-exponentially induced and envelope-associated EnhC localizes primarily in the periplasmic space

A number of L. pneumophila proteins associated with intracellular growth are up-regulated in post-exponential phase (VanRheenen et al., 2006; Vincent and Vogel, 2006). Using anti-EnhC antibody, levels of EnhC were monitored by Western analysis at different phases of growth in AYE broth, to determine if EnhC was regulated in a similar fashion. Steady state levels of EnhC were clearly increased as the bacteria entered post-exponential phase (A600 = 3.8), when compared to the loading control of isocitrate dehydrogenase (ICDH) (Fig. 7A), consistent with previously published micro-array results of bacterial in vivo transcription (Bruggemann et al., 2006).

Fig. 7
The expression of EnhC in L. pneumophila is growth phase dependent and a major fraction of EnhC is associated with the bacterial envelope

As EnhC does not have any well-characterized orthologs to allow the prediction of its function at molecular level, an initial characterization of the localization of this protein was performed to attempt to gain insight into its site of action. To this end, post-exponential phase bacteria were fractionated to identify the compartment enriched in EnhC (Experimental Procedures). Spheroplasts formed by lysozyme treatment resulted in the release of about 40% of EnhC from bacteria (Fig. 7B, LP02, lane P), indicating that at least 40% of EnhC may be localized in the periplasmic space. As a control for a cytoplasmic protein, ICDH was not released by lysozyme treatment of wild type LP02 (Fig. 7B, LP02, lane P). Interestingly, a fraction of ICDH was released by lysozyme treatment of a strain lacking EnhC, indicating that the absence of EnhC may destabilize the bacterial envelope (Fig. 7B, ΔenhC, lane P). The spheroplasts were then lysed by sonication, and membrane associated proteins were pelleted. A small portion of EnhC was found in the supernatants (Fig. 7B, LP02, Lane S), indicating that there may be a small amount of the protein in the cytoplasm, or a fraction is present in the periplasm that could not be released by lysozyme treatment. The remainder of EnhC was found in the Triton X-100 insoluble membrane fraction, so a portion of EnhC may associate with the outer-membrane or be assembled into a Triton X-100-resistant protein complex in the periplasmic space (Fig. 7B, LP02, lane O). Although the fractionation assay could not demonstrate a single fraction containing EnhC, the results indicated that the majority of EnhC was probably localized in the bacterial envelope and, in particular, the periplasmic space.

To further confirm the association of EnhC with the envelope, intact LP02 grown to post-exponential phase was probed by using a membrane impermeant amino reactive biotinylation reagent (Sulfo-NHS-LC-biotin) to label bacterial proteins (Experimental Procedures). This reagent is small enough to penetrate the outer membrane porins of intact L. pneumophila, so proteins having amino groups exposed in the outer membrane, periplasm or periplasmic face of the inner membrane will be covalently modified with biotin. After biotinylation, bacteria were lysed and biotin-modified proteins were pelleted using streptavidin-linked beads (Experimental Procedures). EnhC could be labeled by the biotinylation reagent and pulled down by the streptavidin beads as could DotG, an inner membrane protein thought to have a large periplasmically-localized domain (Fig. 7C) (Vincent et al., 2006). In contrast, the cytoplasmic protein ICDH was not pelleted by the streptavidin beads (Fig. 7C). The results from this assay further confirmed the association of EnhC with the bacterial envelope.

The absence of EnhC results in increased sensitivity to stress conditions

Fractionation of the ΔenhC strain resulted in some leakage of the cytoplasmic ICDH after lysozyme treatment in hyperosmotic buffer, consistent with a dysfunctional envelope in post-exponential phase (Fig. 7B, ΔenhC, lane P). Such a defect could result in enhanced sensitivity of post-exponential bacteria to a variety of stress conditions, at least partially explaining the lowered replication potential of the ΔenhC mutant in the presence of enhanced restriction during the second day of intracellular growth. Therefore, the ΔenhC mutant was grown in the absence of host cells and was tested for its resistance to a pair of stress conditions.

The presence of NaCl in bacteriological medium reduces the viability of L. pneumophila harboring an intact Dot/Icm system, presumably because the Dot/Icm translocator disrupts Na+ homeostasis (Sadosky et al., 1993). In some strain backgrounds, mutations in the dotL gene result in enhanced sensitivity to this cation, perhaps due to misregulation of envelope proteins (Buscher et al., 2005). The absence of EnhC similarly resulted in enhanced sensitivity to NaCl, with more than 10 fold lower viability in the presence of 0.65% NaCl relative to the wild type LP02 strain (Supplemental Figure 1). To further pursue this result, sensitivity to H2O2 in culture medium was analyzed, as resistance to this reagent is a better-defined and more specific measure of the bacterial stress response (Gonzalez-Flecha and Demple, 2000). Post-exponential bacteria were incubated at 25°C in culture medium containing increasing amounts of H2O2 (Experimental Procedures), and bacterial viability was determined by CFU. The mutant lacking EnhC was 3–4 X more sensitive to 5 mM H2O2 relative to wild type and about 10 X more sensitive to 15 mM H2O2 than wild type (Fig. 8A and 8B). This defect could be complemented by the presence of a plasmid-encoded enhC gene (Fig. 8B), indicating that EnhC contributes to the resistance of post-exponential L. pneumophila to H2O2 stress. Therefore, by a pair of measures, the mutant appeared to have enhanced stress sensitivity, perhaps due to a partially defective bacterial envelope. This may potentiate the sensitivity of the mutant to the increased restriction that occurs during the second day of growth in macrophage culture.

Fig. 8
EnhC allows enhanced survival of L. pneumophila in presence of peroxide (H2O2) stress

EnhC is associated with the bacterial envelope in a punctate pattern

Using anti-EnhC antibody, immunofluorescence (IF) staining was performed on L. pneumophila grown to exponential phase (A600=1.8) and post-exponential phase (A600=3.7), to visualize the localization of EnhC in bacteria. We were unable to detect EnhC on the surface of bacteria by IF staining in the absence of fixation (data not shown). The localization of EnhC in fixed post-exponential phase bacteria revealed pronounced staining of EnhC around the periphery of bacteria (Fig. 9A). The staining for EnhC appeared to be punctate in many bacteria (Fig. 9A, enlarged images). When IF images were analyzed (Experimental Procedures), 82 ± 11% of post-exponential phase bacteria gave a positive signal above the threshold, whereas 42±11% of exponential phase bacteria stained positively.

Fig. 9
Growth phase regulation and punctate localization of EnhC in L. pneumophila

As the percentage of exponential phase bacteria that stained positively for EnhC was slightly higher than expected from the results of Western analysis (Fig. 7A), the IF intensity in each positively stained bacterium was determined individually and the data were plotted as a histogram (Fig. 9B, Experimental Procedures). The IF intensity of EnhC in exponential phase bacteria ranged from 8 units of pixel intensity per pixel (PIPP) to 18 PIPP. In contrast, the intensity of EnhC in post-exponential phase bacteria ranged from 63 PIPP to 120 PIPP. The higher intensity of EnhC in post-exponential phase bacteria was also verified by the fact that the mean intensity of EnhC for post-exponential phase bacteria (85 ± 5 PIPP) was about 6 fold higher than that for exponential phase bacteria (15 ± 5 PIPP). The quantification results for IF staining combined with the results of Western analysis (Fig. 7A) suggested that the association of EnhC with the bacterial envelope was greatly enhanced during post-exponential phase.


Here we show that L. pneumophila EnhC is required for persistent intracellular growth in macrophages. The delayed growth defect observed in a ΔenhC mutant (Fig. 2) is easily distinguishable from the previously observed phenotype of most L. pneumophila intracellular growth mutants, which show lowered growth rates relative to wild type beginning at the earliest time points of the replication cycle (Laguna et al., 2006; Vincent and Vogel, 2006; Bardill et al., 2005; Ninio et al., 2005; Sexton et al., 2004b; VanRheenen et al., 2004; Andrews et al., 1998; Segal and Shuman, 1997; Berger and Isberg, 1993). Loss of EnhC did not result in altered intracellular growth if macrophages had not been previously exposed to bacteria, but only appeared when macrophage cultures had started to accumulate cytokines (Fig. 3 and Fig. 4). Consistent with a defect dependent on the production of cytokine, no such depression in intracellular growth was observed when bacteria were grown within Dictyostelium discoideum amoebae (Fig. 2C). Previous work (McHugh et al., 2000b; Arata et al., 1993; Blanchard et al., 1987) indicated that cytokines, especially TNF-α, could restrict the replication of L. pneumophila. At the MOI conditions used here (MOI = 0.05), TNF-α only accumulates to a low concentration, but this level of cytokine was sufficient to selectively interfere with the replication of ΔenhC (Fig. 5). Although it may be argued that any protection conferred by EnhC is limited to rather low levels of TNF-α, it should be emphasized that in the absence of external macrophage activation, the level of TNF-α in macrophage cultures never reaches the concentrations sufficient to interfere with the intracellular growth of wild type L. pneumophila. This supports the model that EnhC is required for L. pneumophila to resist the restriction conferred by TNF-α found in macrophage cultures after 24 hours exposure to bacteria.

The observed trafficking defect of vacuoles containing the ΔenhC mutant in restrictive macrophages (Fig. 6) combined with the observation that the presence of conditioned medium (Fig. 4) or TNF-α (Fig. 5) caused selective inhibition of intracellular growth by the mutant, is consistent with the model that the delayed growth defect of the mutant is at least partially due to improper targeting of the vacuole to the late endosome. Improper trafficking does not appear to be caused by a general defect in Dot/Icm translocation, as lack of EnhC did not affect the translocation efficiency of SidC (a known substrate of Dot/Icm) in both naïve macrophages and macrophages treated with TNF-α (data not shown). However, these results do not exclude the possibility that a specific set of unknown substrates required for LCV formation are inefficiently translocated in the ΔenhC mutant.

It is unclear how TNF-α stimulation leads to selective restriction of bacteria lacking EnhC, nor why this treatment leads to improper trafficking of the LCV. TNF-α is a pleiotropic proinflammatory cytokine important for a wide range of host responses, including cell apoptosis, proliferation, differentiation, activation of NF-κB as well as stimulating production of a number of additional cytokines (Aggarwal, 2003). Furthermore, it has the ability to induce the productions of reactive oxygen species (ROS) and reactive nitrogen species (RNS) (Bubici et al., 2006; Dutta et al., 2006). Among the responses predicted from these divergent signaling pathways activated by TNF-α, ROS and RNS have been suggested to be involved in the inhibition of L. pneumophila (Skerrett and Martin, 1996; Brieland et al., 1995). In fact, we observed that the ΔenhC mutant showed increased sensitivity to H2O2 relative to the wild type strain, suggesting that reactive oxygen species produced downstream of TNF-α signaling may be partially responsible for the reduced growth rates occurring in macrophages starting at the 24 hr. timepoint. Attack by these species on the ΔenhC mutant after contact with TNF-α stimulated macrophages may cause increased targeting of the more susceptible mutant to the endocytic compartment. The observation that production of spheroplasts from ΔenhC resulted in release of the cytoplasmic protein ICDH (Fig. 7B), as well as the increased sensitivity of the ΔenhC mutant to NaCl in bacteriological medium (Supplemental Figure 1) indicate that the ΔenhC strain has a fragile bacterial envelope that may contribute to this susceptibility. These clues about the mutant combined with the observation that EnhC is associated with the bacterial envelope (Fig. 7 and Fig. 9) suggest that EnhC may be involved in stability of the bacterial envelope. L. pneumophila lacking EnhC may sensitize the bacterium to antimicrobial tactics used by restrictive macrophages, such as ROS-dependent killing.

Although enhC was originally identified as a gene required for maximal uptake of L. pneumophila into cell lines (Cirillo et al., 2000), our studies indicated there was no defect in uptake of a ΔenhC mutant (Fig. 1 and Fig. 2). The inability to reproduce the published defect does not fully eliminate the possibility that EnhC plays a role in bacterial uptake, as the growth conditions and the parental bacterial strain analyzed in this study differ from the previous work. Conflicting results on uptake of L. pneumophila have been observed previously (Hilbi et al., 2001; Watarai et al., 2001). However, our studies do argue that EnhC primarily localizes in the periplasm (Fig. 7B), making it unlikely that it is a ligand for a mammalian cell surface receptor. We were also unable to detect EnhC on the surface of bacteria by IF staining in the absence of fixation (data not shown), which is inconsistent with the protein being a surface exposed adhesin.

Although EnhC allows resistance to low levels of TNF-α as well as to conditioned media (Fig. 4 and Fig. 5), neither treatment generates the level of restriction observed during a standard growth curve (Fig. 2), so soluble mediators cannot fully explain the phenotype of the ΔenhC mutant. Restriction of the mutant may also include cell-associated responses that cannot be transferred by conditioned media, or which synergize with TNF-α to inhibit replication of the microorganism. Further analysis of the role of EnhC in supporting intracellular growth will require identifying the cell-associated factors involved in restriction, as well as understanding how the response to TNF-α results in restriction in the absence of EnhC.

The observation that the mutant lacking enhC exhibited a defect in intracellular growth only after accumulation of cytokine in the culture raises a problem for determining the source of the selective pressures that maintain this gene. As there is no evidence for transmission of L. pneumophila between animal hosts, intracellular growth in the presence of low levels of cytokines is an unlikely source of selection for maintenance of the enhC gene. Rather, selective pressures probably result from passage between amoebae or survival of the bacterium in water supplies. Although the replication efficiencies of the mutant and wild type L. pneumophila strains were indistinguishable in the amoebal species D. discoideum (Fig. 2C), this result does not exclude the model that there are unicellular hosts that can mimic the restrictive properties observed in a macrophage culture. Furthermore, the fact that the absence of enhC results in sensitivity of the bacterium to a variety of chemical treatments (Fig. 8 and Fig. S1) raises the possibility that selection outside of a host could cause retention of a gene that is only important when a macrophage has increased potential to restrict survival of microorganisms. Identification of the molecular target of EnhC should provide a key to understanding how this protein supports survival of the microorganism in a variety of environments.

Experimental Procedures

Bacterial strains, plasmids, reagents and cell culture

All bacterial strains and plasmids used in this work were listed in Table 1. All PCR primers used in this work were listed in Supplemental Table 1. Charcoal Yeast Extract agar (CYE) or Charcoal Yeast Extract Thymidine agar(CYET) plates and ACES buffered Yeast Extract (AYE) broth were used to cultivate Legionella pneumophila in vitro as described (Berger and Isberg, 1993; Feeley et al., 1979). Thymidine was added at a concentration of 100 µg/ml when needed. L. pneumophila strain LP02 (thyA hsdR rpsL) (Berger et al., 1994), a thymine-auxotrophic derivative of the Philadelphia-1 isolate, was used as wild type in all experiments. The ΔenhC strain (MLL101), an in frame deletion of enhC, was constructed in the LP02 background as described previously (Rankin et al., 2002). Briefly, suicide vector pSR47s (pML101) containing genomic regions flanking enhC was integrated into the chromosome via homologous recombination and integration events were selected by kanamycin resistance followed by selecting for a second crossover using sucrose resistance. Recombinants were screened for the presence of the deletion by PCR and Western analysis. The EnhC+ complementation plasmid (pML201) was constructed by introducing enhC into the vector pJB908 (Table 1). Bacterial strains (MLL401, MLL501) expressing green fluorescence protein (GFP) were constructed by introducing GFP expressing plasmid pAM239 (Table 1) into bacteria.

Table 1
Bacterial strains and plasmids used in this work

Mouse bone marrow derived macrophages (BMDM) were prepared from the femurs of female A/J mice (Jackson Laboratories) (Swanson and Isberg, 1995) and cultivated in RPMI 1640 (Invitrogen-Gibco) with 10% heat-inactivated fetal bovine serum(FBS) (Gibco). HEp-2 cells (ATCC CCL23) were passaged in DMEM media (Invitrogen-Gibco) containing 10% heat-inactivated FBS (Gibco). The chemical reagents used in this paper were from Sigma (St. Louis, MO) if not stated otherwise.

Gentamicin protection assay to measure bacterial internalization

Bone marrow macrophages (BMDM) and HEp-2 cells were seeded in 24 well tissue culture plates (Falcon) at a density of 4 × 105 cells per well and 2.5 × 105 cells per well, respectively. Post-exponential phase (A600 = 3.7~3.9) motile bacteria cultures were diluted and added into cell monolayers at desired multiplicities of infection (MOI). SpectraMax A5 model (Molecular Devices Co.) spectrophotometer was used for all A600 readings of bacteria in this paper. The HEp-2/BMDM cultures were placed in a plate carrier in the Hermile Z 360 K model (National Labnet Co.) centrifuge and spun at 1000 RPM for 5 min. The plates were moved to 37°C 5% CO2 for 2 hours (hrs) before washing 3 times (3X) with PBS. The cells were incubated in culture medium containing 100µg/ml gentamicin for 2 hrs to kill the extracellular bacteria. After gentamicin treatment, the cells were washed with PBS again before they were lysed in presence of 0.02% saponin. After lysis, the internalized bacteria were measured by plating for colony forming units (CFU) on CYET plates. Uptake percentage was determined by the percentage of the inoculum that survived gentamicin killing.

Intracellular growth assays

The intracellular growth of L. pneumophila in BMDM was measured by change in CFU (Berger et al., 1994). For all studies monitoring the intracellular growth of L. pneumophila in BMDM by following CFU, BMDM were seeded in 24 well tissue culture plates (Falcon) at 4 × 105 cells per well. Motile bacteria grown to post-exponential phase (A600 = 3.7~3.9) in AYE broth were diluted in RPMI and added to BMDM at an MOI = 0.05. After 2 hrs, all wells were washed with RPMI1640 to synchronize the infection. Some wells were then lysed with 0.02% saponin and incubated on CYET or CYE plates for the initial cell-associated CFUs. The remaining wells were lysed at 24 hour-post-infection (hpi), 48 hpi and 72 hpi sequentially to monitor the bacterial intracellular growth over 3 days. Standard deviations were calculated based on CFUs from triplicate infected wells.

The intracellular growth of L. pneumophila in Dictyostelium discoideum was measured by change in CFU as described previously (Li et al., 2005). D. discoideum were seeded in 24 well tissue culture plates (Falcon) at 5 × 105 cells per well. The cells were equilibrated at 25°C for 3 hrs and then infected by post-exponential phase L. pneumophila at MOI = 0.05. After 2 hrs, all wells were washed to synchronize the infection. Some wells were lysed with 0.02% saponin and plated for the initial cell-associated CFUs. The remaining wells were lysed at 24 hpi, 48 hpi and 72 hpi sequentially to monitor the bacterial intracellular growth over 3 days in D. discoideum.

For all studies in this paper requiring immunofluorescence (IF) assay to visualize L.pneumophila in macrophages, BMDMs were seeded onto circular glass cover-slips (Fisher Co.) in 24 well Falcon dishes at 2 ×105 cells per cover-slip. For super-infection, the BMDM monolayer was first infected at MOI = 0.05 for 24 hrs by LP02 or MLL101 (ΔenhC), followed by super-infection by MLL401 (GFP –LP02) or MLL501 (GFP –ΔenhC) at MOI=1 for 14 hrs. After super-infection, the cells were fixed and blocked in goat serum (Roy et al., 1998). Before permeabilization of BMDM, extracellular bacteria were stained by anti-L. pneumophila rabbit serum (1:10,000) (Pocono Rabbit Farm and Laboratory Inc.) and secondary anti-rabbit-Cascade blue (1:500) (Invitrogen). After permeabilizing BMDM with cold methanol, total bacteria were stained by anti-L. pneumophila rabbit serum and secondary anti-rabbit-Texas red (1:500) (Invitrogen). When observed under microscope, only phagosomes containing GFP-bacteria stained by Texas red but not Cascade blue were scored. For IF experiments without performing super-infection, MOI 1 was used. The intracellular bacteria were scored at 14 hpi.

Reinfection assay on BMDM

BMDMs were incubated with L. pneumophila (MOI 0.05) for 24 hrs at 37°C, 5% CO2. The bacterial progeny released from macrophages and present in the macrophage culture supernatant were collected and applied to freshly prepared macrophages to initiate new infections with an MOI approximately equal to 0.05. Subsequent growth was followed by determining CFU over time. If the lysis efficiency of bacteria at 24 hpi in primary infection was measured, the macrophage culture supernatant was collected at 24 hpi and plated out for the CFU released into the supernatant. The remaining macrophage monolayer was lysed by 0.02% saponin and plated out for the CFU associated with macrophages.

Assay for growth of L. pneumophila in BMDM monolayers treated with conditioned medium

BMDMs were incubated with LP02 or MLL101 (ΔenhC) at MOI = 0.05 for 24 hrs. The culture media for the infected monolayers were collected and passed through Steri-flip filters (0.22µm) (Millipore) to remove bacteria and generate bacteria-free “conditioned media.” Freshly prepared BMDMs were incubated with LP02 infection-conditioned media or MLL101 infection-conditioned media for ~16 hrs at 37°C, 5% CO2. The treated BMDMs were then incubated with LP02 or MLL101 for 2 hrs at 37°C, 5% CO2 before washing with RPMI and addition of fresh RPMI containing 10% FBS to BMDMs. To determine CFU, MOI = 0.05 was used; for microscopic assay to determine the distribution of phagosome sizes, MOI = 1 was used. To determine CFU, BMDMs were lysed at 2 hpi and 24 hpi. To determine phagosome size, BMDMs were fixed at 17 hpi.

Measurement of the concentrations of cytokines in the conditioned medium

The concentrations for IL-12p70, TNF-α, IFN-γ, MCP-1, IL-10, and IL-6 were determined by ELISA based cytometric bead array kit (BD Biosciences). A slightly modified protocol based on the manufacturer’s instructions was used (Auerbuch and Isberg, 2007). The concentration for IL-1β in the conditioned medium was determined by mouse IL-1β ELISA kit (eBioscience).

Determination of the intracellular growth of L. pneumophila in BMDM treated by TNF-α or MCP-1

Freshly prepared BMDMs were treated by either E.coli-expressed recombinant mouse TNF-α (R&D Systems) or mouse MCP-1 (R&D Systems) in RPMI + 10% FBS at 37°C, 5% CO2 for at least three hours. The ED50 of TNF-α determined by TNF mediated cytotoxicity in the mouse L-929 cell line in the presence of the metabolic inhibitor actinomycin D is about 50 pg/ml (R&D Systems). The ED50 of MCP-1 determined by its ability to chemoattract hCCR2A transfected mouse BaF/3 cells is about 4 ng/ml (R&D Systems). Post-exponential phase L. pneumophila cultures were appropriately diluted and added to macrophages at a MOI = 0.05. Two hours later, macrophages were washed by RPMI to synchronize the infection and then fresh RPMI containing 10% FBS was added to BMDMs. At 2 hpi and 24 hpi, two sets of macrophages were lysed by 0.02% saponin and incubated on CYE or CYET plates to determine initial CFU associated with macrophages and the CFU obtained after 24 hrs of intracellular growth. The CFU obtained at 24 hpi was divided by the CFU at 2 hpi to determine fold growth of L. pneumophila after 24 hrs of intracellular growth.

Assay for co-localization of LAMP-1 with the L. pneumophila-containing vacuoles (LCV)

Freshly prepared BMDMs were treated by E.coli-expressed recombinant mouse TNF-α (R&D Systems) in RPMI + 10% FBS at 37°C, 5% CO2 for three hours. Untreated BMDMs or TNF-treated macrophages were then incubated with post-exponential phase bacteria at MOI = 1. After one hour of infection, macrophages were fixed and blocked (Roy et al., 1998). Extracellular bacteria and intracellular bacteria were distinguished as described in intracellular growth assays (above). LAMP-1 protein was detected by rat anti-LAMP-1(1D4B) primary monoclonal antibody (Hybridoma Bank at the University of Iowa) using a dilution of 1:100 and secondary anti-rat FITC (Invitrogen).

Anti-EnhC antibody production

The amino-terminal signal sequence of EnhC was replaced by six histidines to generate a His6-tagged version of EnhC that could be purified by using Ni-nitrilotriacetic acid chromatography (Qiagen). The purified His6-EnhC was injected into rabbits to raise anti-EnhC serum (Pocono Rabbit Farm and Laboratory Inc.). Rabbit anti-EnhC polyclonal antibody was purified by affinity-purification (Laguna et al., 2006). Briefly, Affigel-10 beads (BioRad) were linked with purified His6-EnhC. Serum was added to an equal volume of beads, incubated overnight at 4°C, and anti-EnhC antibody was eluted using Glycine/HCl buffer (pH = 2.5). Purified anti-EnhC antibody recognized a single protein with predicted size from LP02 (WT) extracts, but did not recognize any protein from ΔenhC extracts.

Assays for translocations of Dot/Icm substrates in the strain lacking EnhC or WT bacteria

BMDMs were seeded onto glass cover-slips in 24 well Falcon dishes at 2 × 105 cells per cover-slip and incubated with motile post-exponential phase bacteria (A600 = 3.7~3.9) for 2 hrs or 14 hrs at MOI = 1. If TNF-α treatment was analyzed, macrophages were treated for three hours before infection. After infection, macrophages were fixed and blocked (Roy et al., 1998). Using rat anti-Legionella serum (Pocono Rabbit Farm and Laboratory Inc.), extracellular bacteria and intracellular bacteria were distinguished as described in intracellular growth assays (above). The translocation of SidC or EnhC was detected by rabbit anti-SidC antibody (Luo and Isberg, 2004) or rabbit anti-EnhC antibody (Pocono Rabbit Farm and Laboratory Inc.), using a dilution of 1:500, followed by probing with anti-rabbit FITC (Invitrogen). If the digitonin extraction procedure was used on bulk cultures to assess translocations of SidC, Sid M or LidA, published procedures were used (Machner and Isberg, 2006; Derre and Isberg, 2005; Luo and Isberg, 2004). Immunofluorescence staining on purified phagosomes to test the translocation of EnhC, was as described (Conover et al., 2003).

Bacterial fractionation

The protocol for subcellular fractionation of L. pneumophila was slightly modified from previous protocols (Roy and Isberg, 1997). About 20 mls of post-exponential L. pneumophila grown in AYE medium (A600 = 3.7~3.9) were pelleted at 5,000 Xg for 20 min. The bacterial pellet was resuspended in 0.5 ml of 200 mM Tris-HCl (pH = 8.0) and 0.5 ml sucrose buffer (1M sucrose , 50 mM Tris-HCl [pH = 8.0]) was then added. Then 10 µl of 0.5 M EDTA (pH = 8.0), 10 µl of 10 mg/ml lysozyme and 1 ml of H2O were sequentially added to the bacterial suspension. The suspension was then incubated on ice for 30 min to form spheroplasts. MgCl2 was added to a final concentration = 20 mM and spheroplasts were pelleted for 20 min at 5,000 Xg. The supernatant after centrifugation was collected as Periplasmic (P) fraction (~ 2 ml). The spheroplasts were resuspended in 5 ml of 50 mM Tris-HCl (pH = 8.0) and lysed on ice by sonication with 3 sonic bursts at 40% intensity. Cell disruptor 200 (Branson Co.) was used for sonication. Unlysed spheroplasts were pelleted at 5,000 X g for 20 min. 50% of the supernatant (~ 2.5 ml) was saved as fraction T, representing both soluble and membrane-associated proteins. The other half of the supernatant was subjected to ultracentrifugation at 100,000 X g for 1 hr. The supernatant (~2.5 ml) was saved as fraction S, representing soluble cytoplasmic proteins. The pellet (membrane fraction) was resuspended in 1 ml of 50 mM Tris-HCl (pH = 8.0) before 10% Triton X-100 was added to a final concentration of 1%. The membrane fraction was extracted on ice for 30 min before another ultracentrifugation step (100,000 X g, 30 min). The supernatant was saved as TritonX-100 soluble proteins (fraction I ~1 ml). The pellet containing TritonX-100 insoluble proteins was resuspended in 1 ml of 50 mM Tris-HCl (pH = 8.0) (fraction O). Fractions were analyzed by SDS-PAGE followed by immunoblotting (Roy and Isberg, 1997). The gel loading volumes for different fractions were equivalent to the samples extracted from 3 × 108 bacteria. EnhC was detected by probing with rabbit anti-EnhC diluted 1: 10,000. Isocitrate Dehydrogenase (ICDH) was detected by using rabbit anti-ICDH at 1: 30,000 dilution (gift from Dr. Linc. Sonenshein, Department of Microbiology, Tufts University).

Bacterial surface labeling assay

2 × 109 post-exponential phase L. pneumophila bacteria were pelleted at 8,000 Xg for 15 min. The bacteria were washed once with cold PBS and then resuspended in 1 ml of cold PBS. Freshly made Sulfo-NHS-LC-Biotin (Pierce Chemical Co.) stock solution (6.5 mg/ml in PBS) was added to the bacterial suspension to a final concentration = 370 µg/ml to label free amino groups on the proteins exported across face of bacterial inner membrane. The bacterial suspension was left on ice for 30 min. The labeling process was stopped by addition of NH4Cl to final concentration = 50 mM and incubating the suspension on ice for 15 min. Bacterial cells were washed 3 times in cold PBS then lysed in 100 µl of 1% SDS, 10 mM Tris-HCl (pH = 8.0), 1 mM EDTA. Bacterial cells were boiled for 2 min and then cooled down to room temperature prior to the addition of 900 µl ice cold buffer containing 2% Triton X-100, 50 mM Tris-HCl (pH = 8.0) 1 mM EDTA and 150 mM NaCl. The diluted lysate was cleared by centrifugation (15,000 Xg for 10 min, 4°C). The supernatant after centrifugation was split into two halves. One half was saved and subjected to methanol-chloroform precipitation (Derre and Isberg, 2005) and labeled as T (Total input for pull-down). The other half was incubated with 50 µl gel bed volume of streptavidin-agarose beads (Pierce Chemical Co.) overnight on rolling drum at 4°C. After overnight adsorption, the beads were pelleted and the supernatant was subjected to methanol-chloroform precipitation and labeled as S (Supernatant after pull-down). The agarose beads were washed at least three times with cold PBS before SDS sample buffer was added to the beads. The sample was labeled as B (beads) after the beads were boiled in sample buffer for 5 min. Samples were then fractionated on SDS-PAGE and revealed by immunoblotting. DotG was detected by anti-DotG at dilution = 1: 12,000 (Vincent et al., 2006).

Assay for the resistance of L. pneumophila to H2O2 stress

Motile, post-exponential phase bacteria (A600 = 3.7~3.9) were pelleted at 8,000 Xg for 15 min. The supernatant was removed and the pellet was resuspended in fresh AYE broth containing varying concentrations of H2O2 (0 mM, 2 mM, 5 mM, and 15 mM). A small portion of the resuspended culture was diluted and plated for input CFU before exposing to H2O2. The remaining culture was slowly shaken on a platform shaker (Model: Innova 2000, New Brunswick Scientific) at 28 RPM, room temperature, for 16 hrs. After 16 hrs, the culture was diluted and plated for output CFU surviving H2O2 containing medium. The percentage of surviving bacteria was determined by dividing the output CFUs by the input CFUs.

Assay for the resistance of L. pneumophila to NaCl stress

Post-exponential phase (A600 = 3.7~3.9) motile bacteria were diluted and incubated on CYET plates for input CFU and on CYET plates with 0.65% NaCl for CFU resistant to NaCl. The percentage of survivors in presence of NaCl was determined by determing the fraction of CFU resistant to NaCl relative to input CFUs.

Immunofluorescence localization of EnhC in L. pneumophila

The fixative was freshly prepared by mixing 0.2 µl of 25% gluteraldehyde to 100 µl of 16% paraformaldehyde on ice. 40 µl of 0.5 M NaH2PO4 was then added to the fixative prior to slow addition of 0.5 ml of post-exponential phase L. pneumophila broth culture. Then the mixture was left at room temperature for at least 15 min without aeration before it was incubated on ice for at least 30 min. After the fixation step, the bacterial cells were washed 3 times with PBS. The bacterial cells were resuspended in 0.5 ml GTE buffer (50 mM Glucose, 25 mM Tris-HCl, pH = 8.0, 10 mM EDTA). The fixed bacteria were either spotted onto the cover-slips immediately or stored at 4 °C for up to 3 days prior to spotting on cover-slips, using the following protocol.

As the fixation process was taking place, the cover-slips (Fisher. Co) were coated with 1 mg/ml poly-L-Lysine for 20~30 min at room temperature. Then the cover-slips were washed with water 3 times. 20 µl fixed bacteria in GTE buffer was diluted in 400 µl GTE buffer and added to the poly-L-lysine coated cover-slip. The bacteria were allowed to settle on the cover-slip by 1,000 RPM centrifugation using Hermle Z 360 K model (National Labnet Co.) for 5 min. Then the cover-slips coated with the fixed bacteria were left at 37°C for at least 15 min before it was allowed to sit at room temperature for an additional 15 min. The cover-slips were then washed 3 times with PBS before proceeding to immunofluorescence (IF) staining or alternatively, the cover-slips were stored in PBS at 4°C until probing. Before IF staining, fixed bacteria were permeabilized by 0.5 µg/ml of lysozyme in GTE buffer at room temperature for 5 min. EnhC in L. pneumophila was detected by rabbit anti-EnhC using a dilution of 1:500 and secondary anti-rabbit IgG-Texas Red (Invitrogen) with dilution 1:500. Bacterial DNA was stained by DAPI to localize bacteria on the cover-slips (Swanson and Isberg, 1996).

Image analysis was performed using IPLab Spectrum® software (BD Biosciences-Scanalytics). To generate the mean fluorescence intensity value (pixel intensity per pixel or PIPP) for a bacterial population, the total pixel intensity for all bacterial images in a region of interest (ROI) of the picture was divided by the total number of the pixels for all bacterial images in the ROI. The distribution histogram showing bacterial counts at different signal intensity levels was performed by measuring the intensity (PIPP) of IF signal on every single bacterium for the bacterial population in the ROI. Based on these measurements, a histogram was generated by graphing the number of bacteria as a function of different intensities of IF signals on bacteria.

Supplementary Material

Supp Fig 1

Fig. S1. EnhC is required for enhanced resistance of L. pneumophila to NaCl:

The lack of EnhC causes enhanced sensitivity to NaCl relative to the wild type LP02 strain, with more than 10 X lower viability on CYET solid medium containing 0.65% NaCl relative to the wild type. The values shown on Y-axis are negative, so the larger bars represent the lower values. LP03 (dotA) served as a positive control for the strain resistant to NaCl (Experimental procedures). Student’s t-test analysis on LP02 and ΔenhC: P < 0.007. Strains used are described in Table 1. Mean ± standard deviation from triplicate samples shown. Data represent typical experiment, repeated 3 times.



We thank Matthias Machner, Tamara O’Connor, Matt Heidtman, Molly Bergman, Elizabeth Creasey and Vicki Auerbuch Stone for careful review of the manuscript. This work was supported by NIDDK Center Grant P30 DK34928 as well as the Howard Hughes Medical Institute. R. I. is an investigator of the Howard Hughes Medical Institute.


  • Aggarwal BB. Signalling pathways of the TNF superfamily: a double-edged sword. Nat Rev Immunol. 2003;3:745–756. [PubMed]
  • Andrews HL, Vogel JP, Isberg RR. Identification of linked Legionella pneumophila genes essential for intracellular growth and evasion of the endocytic pathway. Infect Immun. 1998;66:950–958. [PMC free article] [PubMed]
  • Arata S, Newton C, Klein TW, Yamamoto Y, Friedman H. Legionella pneumophila induced tumor necrosis factor production in permissive versus nonpermissive macrophages. Proc Soc Exp Biol Med. 1993;203:26–29. [PubMed]
  • Auerbuch V, Isberg RR. Growth of Yersinia pseudotuberculosis in mice occurs independently of Toll-like receptor 2 expression and induction of interleukin 10. Infect Immun. 2007;75:3561–3570. [PMC free article] [PubMed]
  • Bardill JP, Miller JL, Vogel JP. IcmS-dependent translocation of SdeA into macrophages by the Legionella pneumophila type IV secretion system. Mol Microbiol. 2005;56:90–103. [PubMed]
  • Berger KH, Isberg RR. Two distinct defects in intracellular growth complemented by a single genetic locus in Legionella pneumophila. Mol Microbiol. 1993;7:7–19. [PubMed]
  • Berger KH, Merriam JJ, Isberg RR. Altered intracellular targeting properties associated with mutations in the Legionella pneumophila dotA gene. Mol Microbiol. 1994;14:809–822. [PubMed]
  • Blanchard DK, Friedman H, Klein TW, Djeu JY. Induction of interferon-gamma and tumor necrosis factor by Legionella pneumophila: augmentation of human neutrophil bactericidal activity. J Leukoc Biol. 1989;45:538–545. [PubMed]
  • Blanchard DK, Djeu JY, Klein TW, Friedman H, Stewart WE., 2nd Induction of tumor necrosis factor by Legionella pneumophila. Infect Immun. 1987;55:433–437. [PMC free article] [PubMed]
  • Blanchard DK, Djeu JY, Klein TW, Friedman H, Stewart WE., 2nd Protective effects of tumor necrosis factor in experimental Legionella pneumophila infections of mice via activation of PMN function. J Leukoc Biol. 1988;43:429–435. [PubMed]
  • Brieland JK, Remick DG, LeGendre ML, Engleberg NC, Fantone JC. In vivo regulation of replicative Legionella pneumophila lung infection by endogenous interleukin-12. Infect Immun. 1998;66:65–69. [PMC free article] [PubMed]
  • Brieland JK, Remick DG, Freeman PT, Hurley MC, Fantone JC, Engleberg NC. In vivo regulation of replicative Legionella pneumophila lung infection by endogenous tumor necrosis factor alpha and nitric oxide. Infect Immun. 1995;63:3253–3258. [PMC free article] [PubMed]
  • Bruggemann H, Hagman A, Jules M, Sismeiro O, Dillies MA, Gouyette C, et al. Virulence strategies for infecting phagocytes deduced from the in vivo transcriptional program of Legionella pneumophila. Cell Microbiol. 2006;8:1228–1240. [PubMed]
  • Bubici C, Papa S, Dean K, Franzoso G. Mutual cross-talk between reactive oxygen species and nuclear factor-kappa B: molecular basis and biological significance. Oncogene. 2006;25:6731–6748. [PubMed]
  • Burnsed LJ, Hicks LA, Smithee LM, Fields BS, Bradley KK, Pascoe N, et al. A large, travel-associated outbreak of legionellosis among hotel guests: utility of the urine antigen assay in confirming Pontiac fever. Clin Infect Dis. 2007;44:222–228. [PubMed]
  • Buscher BA, Conover GM, Miller JL, Vogel SA, Meyers SN, Isberg RR, et al. The DotL protein, a member of the TraG-coupling protein family, is essential for viability of Legionella pneumophila strian Lp02. J Bacteriol. 2005;187:2927–2938. [PMC free article] [PubMed]
  • Chen J, de Felipe KS, Clarke M, Lu H, Anderson OR, Segal G, Shuman HA. Legionella effectors that promote nonlytic release from protozoa. Science. 2004;303:1358–1361. [PubMed]
  • Chien M, Morozova I, Shi S, Sheng H, Chen J, Gomez SM, et al. The genomic sequence of the accidental pathogen Legionella pneumophila. Science. 2004;305:1966–1968. [PubMed]
  • Cirillo SL, Lum J, Cirillo JD. Identification of novel loci involved in entry by Legionella pneumophila. Microbiology. 2000;146:1345–1359. [PubMed]
  • Clemens DL, Lee BY, Horwitz MA. Mycobacterium tuberculosis and Legionella pneumophila phagosomes exhibit arrested maturation despite acquisition of Rab7. Infect Immun. 2000;68:5154–5166. [PMC free article] [PubMed]
  • Conover GM, Derre I, Vogel JP, Isberg RR. The Legionella pneumophila LidA protein: a translocated substrate of the Dot/Icm system associated with maintenance of bacterial integrity. Mol Microbiol. 2003;48:305–321. [PubMed]
  • Daisy JA, Benson CE, McKitrick J, Friedman HM. Intracellular replication of Legionella pneumophila. J Infect Dis. 1981;143:460–464. [PubMed]
  • Delhalle S, Blasius R, Dicato M, Diederich M. A beginner's guide to NF-kappaB signaling pathways. Ann N Y Acad Sci. 2004;1030:1–13. [PubMed]
  • Derre I, Isberg RR. LidA, a translocated substrate of the Legionella pneumophila type IV secretion system, interferes with the early secretory pathway. Infect Immun. 2005;73:4370–4380. [PMC free article] [PubMed]
  • Dutta J, Fan Y, Gupta N, Fan G, Gelinas C. Current insights into the regulation of programmed cell death by NF-kappaB. Oncogene. 2006;25:6800–6816. [PubMed]
  • Feeley JC, Gibson RJ, Gorman GW, Langford NC, Rasheed JK, Mackel DC, Baine WB. Charcoal-yeast extract agar: primary isolation medium for Legionella pneumophila. J Clin Microbiol. 1979;10:437–441. [PMC free article] [PubMed]
  • Fields BS, Shotts EB, Jr, Feeley JC, Gorman GW, Martin WT. Proliferation of Legionella pneumophila as an intracellular parasite of the ciliated protozoan Tetrahymena pyriformis. Appl Environ Microbiol. 1984;47:467–471. [PMC free article] [PubMed]
  • Gonzalez-Flecha B, Demple B. Genetic responses to free radicals. Homeostasis and gene control. Ann N Y Acad Sci. 2000;899:69–87. [PubMed]
  • Hilbi H, Segal G, Shuman HA. Icm/dot-dependent upregulation of phagocytosis by Legionella pneumophila. Mol Microbiol. 2001;42:603–617. [PubMed]
  • Horwitz MA. The Legionnaires' disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J Exp Med. 1983a;158:2108–2126. [PMC free article] [PubMed]
  • Horwitz MA. Formation of a novel phagosome by the Legionnaires' disease bacterium (Legionella pneumophila) in human monocytes. J Exp Med. 1983b;158:1319–1331. [PMC free article] [PubMed]
  • Horwitz MA. Phagocytosis of the Legionnaires' disease bacterium (Legionella pneumophila) occurs by a novel mechanism: engulfment within a pseudopod coil. Cell. 1984;36:27–33. [PubMed]
  • Horwitz MA, Silverstein SC. Legionnaires' disease bacterium (Legionella pneumophila) multiples intracellularly in human monocytes. J Clin Invest. 1980;66:441–450. [PMC free article] [PubMed]
  • Horwitz MA, Silverstein SC. Interaction of the Legionnaires' disease bacterium (Legionella pneumophila) with human phagocytes. I. L. pneumophila resists killing by polymorphonuclear leukocytes, antibody, and complement. J Exp Med. 1981;153:386–397. [PMC free article] [PubMed]
  • Laguna RK, Creasey EA, Li Z, Valtz N, Isberg RR. A Legionella pneumophila-translocated substrate that is required for growth within macrophages and protection from host cell death. Proc Natl Acad Sci U S A. 2006;103:18745–18750. [PubMed]
  • Li Z, Solomon JM, Isberg RR. Dictyostelium discoideum strains lacking the RtoA protein are defective for maturation of the Legionella pneumophila replication vacuole. Cell Microbiol. 2005;7:431–442. [PubMed]
  • Losick VP, Isberg RR. NF-kappaB translocation prevents host cell death after low-dose challenge by Legionella pneumophila. J Exp Med. 2006;203:2177–2189. [PMC free article] [PubMed]
  • Luo ZQ, Isberg RR. Multiple substrates of the Legionella pneumophila Dot/Icm system identified by interbacterial protein transfer. Proc Natl Acad Sci U S A. 2004;101:841–846. [PubMed]
  • Machner MP, Isberg RR. Targeting of host Rab GTPase function by the intravacuolar pathogen Legionella pneumophila. Dev Cell. 2006;11:47–56. [PubMed]
  • Matsiota-Bernard P, Lefebre C, Sedqui M, Cornillet P, Guenounou M. Involvement of tumor necrosis factor alpha in intracellular multiplication of Legionella pneumophila in human monocytes. Infect Immun. 1993;61:4980–4983. [PMC free article] [PubMed]
  • McDade JE, Shepard CC, Fraser DW, Tsai TR, Redus MA, Dowdle WR. Legionnaires' disease: isolation of a bacterium and demonstration of its role in other respiratory disease. N Engl J Med. 1977;297:1197–1203. [PubMed]
  • McHugh S, Yamamoto Y, Klein TW, Friedman H. Differential expression of IL-1 and TNF receptors in murine macrophages infected with virulent vs. avirulent Legionella pneumophila. Can J Microbiol. 2000a;46:885–891. [PubMed]
  • McHugh SL, Yamamoto Y, Klein TW, Friedman H. Murine macrophages differentially produce proinflammatory cytokines after infection with virulent vs. avirulent Legionella pneumophila. J Leukoc Biol. 2000b;67:863–868. [PubMed]
  • McHugh SL, Newton CA, Yamamoto Y, Klein, TW, Friedman H. Tumor necrosis factor induces resistance of macrophages to Legionella pneumophila infection. Proc Soc Exp Biol Med. 2000c;224:191–196. [PubMed]
  • Murata T, Delprato A, Ingmundson A, Toomre DK, Lambright DG, Roy CR. The Legionella pneumophila effector protein DrrA is a Rab1 guanine nucleotide-exchange factor. Nat Cell Biol. 2006;8:971–977. [PubMed]
  • Nagai H, Kagan JC, Zhu X, Kahn RA, Roy CR. A bacterial guanine nucleotide exchange factor activates ARF on Legionella phagosomes. Science. 2002;295:679–682. [PubMed]
  • Nash TW, Libby DM, Horwitz MA. Interaction between the legionnaires' disease bacterium (Legionella pneumophila) and human alveolar macrophages. Influence of antibody, lymphokines, and hydrocortisone. J Clin Invest. 1984;74:771–782. [PMC free article] [PubMed]
  • Ninio S, Zuckman-Cholon DM, Cambronne ED, Roy CR. The Legionella IcmS-IcmW protein complex is important for Dot/Icm-mediated protein translocation. Mol Microbiol. 2005;55:912–926. [PubMed]
  • Payne NR, Horwitz MA. Phagocytosis of Legionella pneumophila is mediated by human monocyte complement receptors. J Exp Med. 1987;166:1377–1389. [PMC free article] [PubMed]
  • Rankin S, Li Z, Isberg RR. Macrophage-induced genes of Legionella pneumophila: protection from reactive intermediates and solute imbalance during intracellular growth. Infect Immun. 2002;70:3637–3648. [PMC free article] [PubMed]
  • Roy CR, Isberg RR. Topology of Legionella pneumophila DotA: an inner membrane protein required for replication in macrophages. Infect Immun. 1997;65:571–578. [PMC free article] [PubMed]
  • Roy CR, Berger KH, Isberg RR. Legionella pneumophila DotA protein is required for early phagosome trafficking decisions that occur within minutes of bacterial uptake. Mol Microbiol. 1998;28:663–674. [PubMed]
  • Sadosky AB, Wiater LA, Shuman HA. Identification of Legionella pneumophila genes required for growth within and killing of human macrophages. Infect Immun. 1993;61:5361–5373. [PMC free article] [PubMed]
  • Segal G, Shuman HA. Characterization of a new region required for macrophage killing by Legionella pneumophila. Infect Immun. 1997;65:5057–5066. [PMC free article] [PubMed]
  • Segal G, Purcell M, Shuman HA. 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 U S A. 1998;95:1669–1674. [PubMed]
  • Sexton JA, Miller JL, Yoneda A, Kehl-Fie TE, Vogel JP. Legionella pneumophila DotU and IcmF are required for stability of the Dot/Icm complex. Infect Immun. 2004a;72:5983–5992. [PMC free article] [PubMed]
  • Sexton JA, Pinkner JS, Roth R, Heuser JE, Hultgren SJ, Vogel JP. The Legionella pneumophila PilT homologue DotB exhibits ATPase activity that is critical for intracellular growth. J Bacteriol. 2004b;186:1658–1666. [PMC free article] [PubMed]
  • Skerrett SJ, Martin TR. Roles for tumor necrosis factor alpha and nitric oxide in resistance of rat alveolar macrophages to Legionella pneumophila. Infect Immun. 1996;64:3236–3243. [PMC free article] [PubMed]
  • Skerrett SJ, Bagby GJ, Schmidt RA, Nelson S. Antibody-mediated depletion of tumor necrosis factor-alpha impairs pulmonary host defenses to Legionella pneumophila. J Infect Dis. 1997;176:1019–1028. [PubMed]
  • Susa M, Ticac B, Rukavina T, Doric M, Marre R. Legionella pneumophila infection in intratracheally inoculated T cell-depleted or -nondepleted A/J mice. J Immunol. 1998;160:316–321. [PubMed]
  • Swanson MS, Isberg RR. Association of Legionella pneumophila with the macrophage endoplasmic reticulum. Infect Immun. 1995;63:3609–3620. [PMC free article] [PubMed]
  • Swanson MS, Isberg RR. Analysis of the intracellular fate of Legionella pneumophila mutants. Ann N Y Acad Sci. 1996;797:8–18. [PubMed]
  • Tilney LG, Harb OS, Connelly PS, Robinson CG, Roy CR. How the parasitic bacterium Legionella pneumophila modifies its phagosome and transforms it into rough ER: implications for conversion of plasma membrane to the ER membrane. J Cell Sci. 2001;114:4637–4650. [PubMed]
  • VanRheenen SM, Dumenil G, Isberg RR. IcmF and DotU are required for optimal effector translocation and trafficking of the Legionella pneumophila vacuole. Infect Immun. 2004;72:5972–5982. [PMC free article] [PubMed]
  • VanRheenen SM, Luo ZQ, O'Connor T, Isberg RR. Members of a Legionella pneumophila family of proteins with ExoU (phospholipase A) active sites are translocated to target cells. Infect Immun. 2006;74:3597–3606. [PMC free article] [PubMed]
  • Vincent CD, Vogel JP. The Legionella pneumophila IcmS-LvgA protein complex is important for Dot/Icm-dependent intracellular growth. Mol Microbiol. 2006;61:596–613. [PubMed]
  • Vincent CD, Friedman JR, Jeong KC, Buford EC, Miller JL, Vogel JP. Identification of the core transmembrane complex of the Legionella Dot/Icm type IV secretion system. Mol Microbiol. 2006;62:1278–1291. [PubMed]
  • Vogel JP, Roy C, Isberg RR. Use of salt to isolate Legionella pneumophila mutants unable to replicate in macrophages. Ann N Y Acad Sci. 1996;797:271–272. [PubMed]
  • Vogel JP, Andrews HL, Wong SK, Isberg RR. Conjugative transfer by the virulence system of Legionella pneumophila. Science. 1998;279:873–876. [PubMed]
  • Wadowsky RM, Butler LJ, Cook MK, Verma SM, Paul MA, Fields BS, et al. Growth-supporting activity for Legionella pneumophila in tap water cultures and implication of hartmannellid amoebae as growth factors. Appl Environ Microbiol. 1988;54:2677–2682. [PMC free article] [PubMed]
  • Watarai M, Derre I, Kirby J, Growney JD, Dietrich WF, Isberg RR. Legionella pneumophila is internalized by a macropinocytotic uptake pathway controlled by the Dot/Icm system and the mouse Lgn1 locus. J Exp Med. 2001;194:1081–1096. [PMC free article] [PubMed]
  • Weber SS, Ragaz C, Reus K, Nyfeler Y, Hilbi H. Legionella pneumophila exploits PI(4)P to anchor secreted effector proteins to the replicative vacuole. PLoS Pathog. 2006;2:e46. [PMC free article] [PubMed]
  • Yamamoto Y, Klein TW, Newton CA, Widen R, Friedman H. Growth of Legionella pneumophila in thioglycolate-elicited peritoneal macrophages from A/J mice. Infect Immun. 1988;56:370–375. [PMC free article] [PubMed]