Department of Pharmacology, University of California, San Diego, La Jolla, CA 92093-0721, USA;
PilT is required for PI(3,4,5)P3-mediated crosstalk between Neisseria gonorrhoeae and epithelial cells
Article first published online: 15 JUN 2005
Volume 7, Issue 9, pages 1271–1284, September 2005
How to Cite
Lee, S. W., Higashi, D. L., Snyder, A., Merz, A. J., Potter, L. and So, M. (2005), PilT is required for PI(3,4,5)P3-mediated crosstalk between Neisseria gonorrhoeae and epithelial cells. Cellular Microbiology, 7: 1271–1284. doi: 10.1111/j.1462-5822.2005.00551.x
- Issue published online: 15 JUN 2005
- Article first published online: 15 JUN 2005
- Received 7 March, 2005; revised 8 April, 2005; accepted 12 April, 2005.
The retractile type IV pilus participates in a number of fundamental bacterial processes, including motility, DNA transformation, fruiting body formation and attachment to host cells. Retraction of the N. gonorrhoeae type IV pilus requires a functional pilT. Retraction generates substantial force on its substrate (> 100 pN per retraction event), and it has been speculated that epithelial cells sense and respond to these forces during infection. We provide evidence that piliated, Opa non-expressing Neisseria gonorrhoeae activates the stress-responsive PI-3 kinase/Akt (PKB) pathway in human epithelial cells, and activation is enhanced by a functional pilT. PI-3 kinase inhibitors wortmannin and LY294002 reduce cell entry by 81% and 50%, respectively, illustrating the importance of this cascade in bacterial invasion. PI-3 kinase and its direct downstream effectors [PI(3,4,5)P3] and Akt are concentrated in the cell cortex beneath adherent bacteria, particularly at the periphery of the bacterial microcolonies. Furthermore, [PI(3,4,5)P3] is translocated to the outer leaflet of the plasma membrane. Finally, we show that [PI(3,4,5)P3] stimulates microcolony formation and upregulates pilT expression in vitro. We conclude that N. gonorrhoeae activation of PI-3 kinase triggers the host cell to produce a lipid second messenger that influences bacterial behaviour.
Type IV pili are produced by a wide variety of bacteria, including Myxococcus xanthus, Synechocystis spp., Vibrio cholerae, enteropathogenic and enterohaemorrhagic Escherichia coli, Pseudomonas aeruginosa, Neisseria gonorrhoeae (GC) and Neisseria meningitidis (see review, Strom and Lory, 1993). These structures play important roles in motility-associated behaviour such as fruiting body and biofilm formation, phototaxis, phage sensitivity and virulence (Wu et al., 1997; Comolli et al., 1999; Merz and So, 2000a; Bhaya et al., 2001).
The GC type IV pilus binds human complement regulatory protein CD46 in some cell types (Kallstrom et al., 1997). Piliated bacteria trigger tyrosine phosphorylation of the CD46 cytosolic tail 2 isoform in A431 epidermoid carcinoma cells (Lee et al., 2002). They induce the release of free Ca2+ from intracellular stores, which redirects trafficking of Lamp1 compartments and promotes bacterial intracellular survival (Kallstrom et al., 1997; Lin et al., 1997; Ayala et al., 2001). These activities of the GC type IV pilus strongly suggest that the structure modulates host cell signalling cascades that influence infectivity.
Type IV pili are retractile structures (Merz et al., 2000b; Skerker and Berg, 2001). Retraction of a single GC pilus often generates forces greater than 100 pN (Merz et al., 2000b; Maier et al., 2002). Retraction requires PilT, a member of a large family of putative NTPases from type II and type IV secretion systems (Wolfgang et al., 1998a,b; Herdendorf et al., 2002). GC pilT null mutants are piliated but cannot retract pili (Merz et al., 2000b). They are non-motile (Wolfgang et al., 1998b). They aggregate into microcolonies, though these structures have an aberrant appearance. They adhere to epithelial cells, but are unable to trigger the formation of cortical plaques – structures in the host cell cortex beneath adherent bacteria containing high concentrations of cytoskeletal components, transmembrane receptors and other signalling proteins (Merz et al., 1999). Finally, pilT mutants cannot be transformed with DNA (Wolfgang et al., 1998b).
How pilus retraction might influence infectivity is unknown. Mechanical forces less than 80 pN exerted on the plasma membrane of mammalian cells trigger Ca2+ fluxes and induce signalling cascades, promoting, among other events, cytoskeleton reorganization during cell–cell and cell–substrate adhesion and cellular adhesion plaque formation (Wang et al., 1993; Raucher et al., 2000; Berridge et al., 2003). Pilus retraction during bacterial attachment may therefore generate sufficient force on the epithelial membrane to trigger cellular signals that influence the course of infection.
One signal cascade that is activated through membrane tension is the phosphoinositide-3 (PI-3) kinase pathway (Miao et al., 2002). PI-3 kinases orchestrate many cellular events, including cell growth, motility, differentiation, survival and trafficking (Cantley, 2002). PI-3 kinases have been implicated in bacterial virulence. Efficient entry into host cells by Listeria monocytogenes, Helicobacter pylori and Chlamydia pneumoniae require activated PI-3 kinase (Ireton and Cossart, 1997; Coombes et al., 2002; Kwok et al., 2002). Invasion of Opa-expressing, non-piliated GC through the CEACAM-3 receptor is also stimulated by PI-3 kinase (Booth et al., 2003).
We tested the hypothesis that pilus retraction activates the PI-3 kinase pathway. We report that GC infection activates the PI-3 kinase/Akt pathway in a process that involves pilT. The cascade triggers the epithelial cell to produce a lipid second messenger that, in turn, affects bacterial motility behaviour.
PI-3 kinase and Akt are activated upon GC infection
We tested the hypothesis that pilus retraction during infection activates PI-3 kinase, as this pathway is known to be activated by membrane tension. Akt activity served as the indicator of PI-3 kinase activation, as Akt is a direct downstream effector of PI-3 kinase. A431 cells were infected for various lengths of time with wild-type (wt) GC strain MS11 or MS11pilT, a piliated, Opa non-expressing strain with a null mutation in pilT. This mutant adheres normally to cells, forms microcolonies, but fails to retract pili. Akt was immunoprecipitated from cell lysates and tested for its ability to phosphorylate the Akt substrate GSK-3α/β. Phospho-GSK-3α/β levels rose within 30 min after infection with wt MS11, suggesting Akt activation, and the levels remained high for 3 h (Fig. 1A; upper panel). In contrast, phospho-GSK-3α/β levels were significantly lower in cells infected with the pilT null mutant (Fig. 1B; upper panel). Control immunoblots of total cell lysates using an anti-Akt antibody showed that equal amounts of Akt were present in the samples (Fig. 1A and B, lower panels).
The influence of pilus retraction on Akt activation was further examined using MS11placZ-pilT, a strain in which pilT is under control of an IPTG-inducible promoter. A431 cells were infected for 30 min with wt MS11, MS11pilT, or MS11placZ-pilT, and Akt activity was assessed as described above. Like the previous experiment, phospho-GSK-3α/β levels were higher in cells infected with wt MS11 than in cells infected with either pilT mutant (Fig. 1C, upper panel). However, phospho-GSK-3α/β levels noticeably rose when IPTG was added to the cultures infected with MS11placZ-pilT. Immunoblots of the input, total cell lysates using an antibody recognizing phosphorylated and unphosphorylated Akt showed that variations in Akt activity were not resulting from loading error (middle panel). Additional immunoblots of input lysates with an anti-phospho-Akt Ser473 antibody showed that the appearance of phospho-Akt coincided with Akt activity on GSK-3α/β, confirming Akt activation in infected cells (lower panel).
The difference in intensity of the phospho-GSK-3α/β bands in Fig. 1B and C is a result of differences in exposure times of the immunoblots. Longer exposures of the blot shown in Fig. 1B revealed phospho-GSK-3α/β bands in infected cell lysates, these bands being more intense that those in the uninfected (0 h) lane (data not shown).
Finally, the PI-3 kinase inhibitor wortmannin blocked Akt activation in MS11-infected cells, verifying the dependence of Akt activation on PI-3 kinase (Fig. 1C, upper panel). Control experiments indicated that wortmannin did not affect bacterial viability or growth (data not shown).
Thus, infection of epithelial cells with piliated, Opa non-expressing GC activates the PI-3 kinase/Akt pathway in a process that is influenced by pilT.
Invasion of piliated GC is partially dependent on PI-3 kinase activation
PI-3 kinase contributes to cell invasion by several bacterial pathogens (Ireton and Cossart, 1997; Coombes et al., 2002; Kwok et al., 2002). We therefore determined whether the PI-3 kinase pathway also influenced GC cell entry. A431 cells were infected with wt MS11 for 6 h in the presence or absence of PI-3 kinase inhibitors wortmannin or LY294002. The bacterial invasion index was determined using the Gentamicin protection assay (Shaw and Falkow, 1988; Makino et al., 1991; Gomez-Duarte et al., 1997). This assay allows quantification of intracellular, or Gentamicin-protected, bacteria on a large scale. Each infection was performed in triplicate, and each experiment was repeated three times with similar results. Wortmannin and LY294002 reduced MS11 invasion by 81.6% (P = 0.005) and 50% (P = 0.001) respectively (Fig. 2A and B). The inhibitors did not affect bacterial viability, growth, or adhesion to cells (data not shown). These results suggest that entry of piliated, Opa non-expressing GC into epithelial cells requires, in part, the PI-3 kinase pathway.
We next determined whether GC invasion via the PI-3 kinase pathway requires a functional pilT. The invasion index of wt MS11 and MS11pilT was compared. MS11pilT adhered to A431 cells like the wt parent strain (data not shown). However, its invasiveness was reduced by 40% (P = 0.040; Fig. 2C). The same pilT mutation in another genetic background resulted in an identical invasion defect (data not shown). Although invasion was significantly reduced in the pilT mutant, it was not completely abolished. Thus, GC enters cells through pilT dependent and independent pathways.
Finally, we assessed the role of PI-3 kinase in cell entry. The invasion index of the MS11pilT mutant was determined in the presence or absence of PI-3 kinase inhibitor LY294002 (Fig. 2D). LY294002 did not affect the ability of MS11pilT to enter cells. Thus, PI-3 kinase is only involved in the pilT-dependent invasion pathway.
PI-3 kinase and its downstream effectors cluster at the base of microcolonies
Piliated Neisseria trigger the formation of cortical plaques in infected cells directly beneath sites of bacterial attachment. These structures, visible for 3–4 h after infection, are enriched in membrane and other signalling proteins and cytoskeletal components (Merz et al., 1999; Hoffmann et al., 2001a). We next determined the location of PI-3 kinase and Akt in infected cells. Cells were infected for 3 h with wt MS11 and processed for immunofluorescence microscopy using antibodies recognizing the PI-3 kinase p110 catalytic subunit, phospho-Akt Ser473, or ezrin, a cytoskeleton component present in Neisseria-induced cortical plaques. Fluorescence was absent in control experiments using primary or secondary antibodies alone (data not shown).
Successive 0.1 µm thick vertical (Z)-sections of infected cells from the apical to the basolateral membrane revealed high concentrations of PI-3 kinase, phospho-Akt and ezrin at the cell cortex, particularly at the periphery of the microcolonies (Fig. 3A–F). Image overlays demonstrated the colocalization of PI-3 kinase and Akt with ezrin. Three-dimensional reconstructions of the same stack of Z-section images supported this interpretation. Ezrin concentrated beneath the microcolony. They also clustered within the microvilli at this site (Fig. 3G–I), giving the cell surface a brush-like appearance. In contrast, neither PI-3 kinase nor phospho-Akt clustered beneath adherent MS11pilT(Fig. 4A). Thus, GC clusters PI-3 kinase p110 and phospho-Akt at sites of infection, and this process involves pilT.
[PI(3,4,5)P3] is translocated to the outer leaflet of infected cell membranes
Activated PI-3 kinase phosphorylates [PI(4,5)P2], generating [PI(3,4,5)P3]. This lipid second messenger recruits Akt to the membrane, leading to its subsequent activation. The clustering of activated PI-3 kinase and Akt at sites of GC attachment predicts that these sites will also have high levels of [PI(3,4,5)P3]. Immunofluorescence microscopy of MS11-infected cells using anti-[PI(3,4,5)P3] antibodies revealed accumulations of this phospholipid around microcolonies (Fig. 4B).
Infected unpermeabilized cells were also stained with the [PI(3,4,5)P3] antibody (Fig. 4B). In these experiments, saponin was omitted from the fixation protocol and all processing steps were performed at 4°C to minimize antibody entry. Under the same conditions, infected cells did not stain with an antibody to PARP, a cytosolic protein (Fig. 4B). This differential staining of [PI(3,4,5)P3] suggests that the phospholipid is surface exposed. In contrast, [PI(3,4,5)P3] did not cluster nor become surface exposed when cells were infected with the MS11pilT mutant (data not shown). Thus, GC clusters [PI(3,4,5)P3] in a pattern resembling PI-3 kinase p110 and phospho-Akt at sites of infection, and this process involves pilT.
A number of pathogens trigger rapid externalization of phosphatidylserine (PS) in infected cells (Fujimoto et al., 1998; Gao and Abu Kwaik, 1999; Goth and Stephens, 2001; Yilmaz et al., 2004). PS is normally sequestered on the cytosolic face of the plasma membrane, but can be rapidly and transiently translocated to the outer leaflet of the membrane. Infected cells were stained with a monoclonal antibody to PS (Chemicon) to examine membrane translocation of this lipid species during the course of infection. Some PS externalized to the outer membrane of A431 cells upon infection by MS11 (Fig. 4C, upper panel). In contrast to [PI(3,4,5)P3], externalized PS did not cluster around adherent bacteria. In non-permeabilized cells, costaining with the known plaque component phosphoERM(ezrin-radixin-moesin) was absent, confirming the specific staining of outer membrane species in our immunofluorescence assays (Fig. 4C, upper panel). This experiment was repeated with the addition of the membrane permeabilizing agent saponin (Fig. 4C, lower panel). Under these conditions, phosphoERM is shown clustering beneath adherent microcolonies. Furthermore, staining of PS in the nucleus after permeabilization was pronounced. Taken together, these data support the finding that lipid translocation takes place upon infection by wt GC expressing a functional PilT protein. Furthermore, [PI(3,4,5)P3], but not PS, translocates to areas directly beneath microcolony formation, implying that the lipid externalization by GC infection is specific for [PI(3,4,5)P3] and is not a general phenomenon of lipid membrane depolarization.
A marker of PS translocation is the phosphorylation of α-adducin at Ser 724 in host cells (Chapline et al., 1993; Dong et al., 1995). To lend further support for lipid translocation, infected cells were assayed for the presence of phospho-adducin.
A431 cells were infected for various lengths of time with MS11 or MS11pilT and the lysates were probed for phospho-adducin at Ser 724. Phospho-adducin levels were noticeably increased after 2 h of infection with MS11 or TPA treatment (Fig. 4D). This lends additional support to the observation that lipid translocation occurs in infected cell membranes, and is in direct support of our immunofluorescence data. Phospho-adducin levels were unchanged in MS11pilT-infected cells, strongly suggesting that pilT is required for lipid translocation. Reprobing of the same blots with an anti-β-tubulin antibody showed equal sample loading (Fig. 4D, lower panels).
We also determined whether infected cells released [PI(3,4,5)P3] into the medium. ELISA was used to determine total [PI(3,4,5)P3] concentrations in infected and uninfected cell supernates. In numerous assays, nearly identical amounts of [PI(3,4,5)P3] were detected in supernates of infected and uninfected cells (26.4 ± 1.7 SD pmol). Thus, [PI(3,4,5)P3] is anchored to the infected cell membrane and not released into the medium.
[PI(3,4,5)P3] modulates bacterial behaviour
The proximity of [PI(3,4,5)P3] to adherent GC led us to determine whether this phospholipid affected bacterial behaviour, in particular, microcolony formation and twitching motility. An aggregation assay was first performed. Wild-type MS11 cells were incubated for 1 h with 10 µg of [PI(3,4,5)P3] or vehicle alone, and microcolony formation was assessed by live-cell imaging and Differential Interference Contrast microscopy. Bacteria incubated with [PI(3,4,5)P3] or vehicle were motile. However, bacteria exposed to [PI(3,4,5)P3] aggregated into microcolonies more quickly than those in the presence of vehicle (Fig. 5). Moreover, microcolonies formed in the presence of the lipid were much larger than those formed in the presence of the vehicle. As an additional control, bacteria were exposed to the lipid species [PI(4,5)P2] under identical conditions. Interestingly, bacteria incubated with [PI(4,5)P2] did not aggregate into microcolonies as quickly as [PI(3,4,5)P3] (data not shown). These results suggest that [PI(3,4,5)P3] stimulates microcolony formation.
To determine whether [PI(3,4,5)P3] associates with the aggregating bacteria, MS11 was incubated for 60 min with [PI(3,4,5)P3] convalently attached to fluorescent beads. Bacteria were visualized by Differential Interference Contrast microscopy, and the beads were visualized using the FITC channel (Fig. 6A). As bacteria moved along the surface, they aggregated, actively acquiring lipid-coated beads into the developing microcolonies. The diffuse fluorescence of the [PI(3,4,5)P3] beads in these images (Fig. 6A, upper panel) is resulting from diffraction of fluorescence emission by bacteria. The same FITC image viewed by Differential Interference Contrast revealed the presence of microcolonies at these sites (Fig. 6A, lower panel). As observed in the previous assay, bacteria exposed to mock-coated beads aggregated to a lesser extent than those exposed to [PI(3,4,5)P3]-coated beads, and their microcolonies were much smaller. Bacteria rarely incorporated mock-coated beads into their microcolonies, even though they came into physical contact with the beads (Fig. 6B).
Once a lipid-coated bead was ‘engulfed’ by a microcolony, it remained within the structure and moved with it. Time-lapsed video microscopy (see Supplementary material) revealed the microcolony-incorporated beads in constant motion resembling twitching behaviour. Mock-coated beads remained attached to the coverslip even if they had been touched by a passing microcolony. Taken together, these imaging studies demonstrate that host cell [PI(3,4,5)P3] enhances GC microcolony formation.
[PI(3,4,5)P3] upregulates pilT
GC twitching motility and pilus retraction require pilT. The effect of [PI(3,4,5)P3] on GC movement and microcolony formation led us to determine whether this phospholipid affected pilT expression. Wild-type MS11 was incubated for 2 h with soluble [PI(3,4,5)P3], [PI(4,5)P2], or vehicle alone, and the level of pilT message was determined by quantitative real-time polymerase chain reaction (PCR). Expression of pilT increased 1.8-fold (SD ± 0.2) in cultures incubated with [PI(4,5)P3]. This response is specific to [PI(3,4,5)P3], as neither vehicle nor [PI(4,5)P2] altered pilT expression. [PI(4,5)P2] differs from [PI(3,4,5)P3] by the absence of a phosphate group in the inositol ring. Thus, [PI(3,4,5)P3] not only affects motility behaviour, it also affects expression of a gene that is involved in twitching motility and pilus retraction.
The N. gonorrhoeae type IV pilus modulates host cell functions during infection (Kallstrom et al., 1998; Merz and So, 2000; Lee et al., 2002). We tested the hypothesis that pilus retraction can also influence GC interactions with the epithelial cell. We presented evidence that GC activates the PI3-kinase/Akt signalling pathway in human epithelial cells, and that this activation requires, in part, a functional pilT (Fig. 1). The PI-3 kinase pathway plays a role in invasion, as wortmannin and LY294002 reduce invasion by 81.6% (P = 0.005) and 50.0% (P = 0.001) respectively (Fig. 2).
A number of bacterial pathogens use the PI-3 kinase pathway to trigger cytoskeleton remodelling and/or modulate phagocytic events that promote invasion (Ireton and Cossart, 1997; Coombes et al., 2002; Kwok et al., 2002). How these pathogens activate PI-3 kinase is not known. Non-piliated GC can activate PI-3 kinase through Opa binding of CEACAM-3 (Booth et al., 2003). Our study indicates that the GC type IV pilus can also activate this cascade.
Two different approaches show that [PI(3,4,5)P3], the direct product of activated PI-3 kinase, is exposed on the outer leaflet of the infected cell membrane. Although phosphoinositides are usually localized to the cytoplasmic leaflet of the eucaryotic plasma membrane, they and other lipid moieties can be translocated across the membrane bilayer and/or released into the medium (Daleke, 2003). Physiological levels of exogenous [PI(3,4,5)P3], but not [PI(3,4)P(2)], PI[(4,5)P(2)] or PI[(3)P], induce Ca2+ fluxes in T cells, implying that extracellular [PI(3,4,5)P3] has an important role in cellular communication (Hsu et al., 2000). That [PI(3,4,5)P3] is present on the outer face of the infected cell membrane, in close proximity to the attached microcolony, suggests that the colonizing bacteria may have access to this pool of lipids.
What might be the function(s) of [PI(3,4,5)P3] in GC–host cell interactions? H. pylori, Chlamydia trachomatis and Campylobacter upsaliensis bind host lipids with high affinity (Lingwood et al., 1992; Huesca et al., 1996; Sylvester et al., 1996; Busse et al., 1997; Barnett Foster et al., 1999), leading to the hypothesis that lipids promote bacterial attachment. In this model, [PI(3,4,5)P3] binding would strengthen physical contact between the bacterium and epithelial cell.
Alternatively, [PI(3,4,5)P3] may act as a chemoattractant, its accumulation at the site of infection creating a gradient that is recognized by the bacteria. Several studies support this hypothesis. A number of lipids affect bacterial twitching motility behaviour. Certain phosphoethanolamine (PE) species are chemoattractants for M. xanthus, stimulating type IV pilus-dependent social motility (Kearns and Shimkets, 2001). Others enhance the twitching velocity of P. aeruginosa type IV pili (Kearns and Shimkets, 1998). PE also induces aggregation of enteropathogenic and enterohaemorrhagic E. coli (Barnett Foster et al., 1999). Consistent with these findings, we observe that [PI(3,4,5)P3] affects GC aggregation/microcolony formation (Fig. 5) and upregulates transcription of pilT, the gene encoding the presumptive retraction motor. That [PI(4,5)P2] had no effect on pilT expression indicates the specificity of the latter response and lends further support for this hypothesis. PilT protein levels were not measured in this study. However, a 1.8-fold increase in the level of a transcript could result in a much higher-fold increase in its translated product.
As in the case for other bacteria, the details of phospholipids signalling between GC and the host cell remain to be defined. In light of the above observations, it is tempting to speculate that [PI(3,4,5)P3] alters the pilus retraction process, perhaps by increasing its rate of retraction, its force of retraction, or both. One direct consequence of increased pilT expression might be an increase in the number of retractile motors, which in turn would increase the overall rate of retraction.
Both models of [PI(3,4,5)P3] function require the translocation of this phospholipid across the plasma membrane. The molecular basis for this reaction is unclear, though precedents exist for the translocation of other lipids. A number of enzymes are known to translocate PE and PI (phosphatidylinositols) to the outer leaflet of the lipid bilayer (Daleke, 2003). Lipids can be externalized by stress (van Engeland et al., 1998) and by bacterial infection (Fujimoto et al., 1998; Gao and Abu Kwaik, 1999; Goth and Stephens, 2001; Hoffmann et al., 2001b), and the latter event is postulated to stimulate phagocytosis.
The membrane translocation of [PI(3,4,5)P3] and PS does not necessarily indicate that the infected cell is undergoing apoptosis. Indeed, there is evidence to suggest the opposite. Chlamydia-triggered translocation of PS correlates with anti-apoptotic effects in the cell (Greene et al., 2004). Porphyromonas gingivalis infection stimulates early translocation of PS while activating PI3K/Akt to promote host survival (Yilmaz et al., 2004). PS externalization and clustering on phagocyte membranes may serve to stimulate microbial engulfment (Fadok et al., 2000). Finally, recent studies in our lab indicate that pilus retraction activates cytoprotective cascades that override apoptosis signalling in infected cells (Howie et al., submitted).
Our study shows that communication of GC with the epithelial cell via the type IV pilus is bi-directional, involving an extended dialogue between bacterial and host components. This mode of bacteria-host cell communication has gained attention by the recent discovery that the eucaryotic hormones epinephrine and norepinephrine induce quorum sensing and regulate virulence gene expression among EHEC (Sperandio et al., 2003). Further studies should provide new insight into the role of interspecies signalling in bacterial pathogenesis.
Reagents, cell lines, bacterial strains and infections
Wortmannin, LY294002, soluble [PI(3,4,5)P3] and [PI(4,5)P2] were purchased from Calbiochem; TPA (Phorbol-12 Myristate-13–Acetate) from Cell Signaling Technologies. Antibodies were purchased from the following: PI-3 kinase p110 subunit and phospho-Akt from Cell Signaling Technologies; PS from Chemicon; phospho-adducin from Upstate Biotechnologies; ezrin from BD Transduction Laboratories; [PI(3,4,5)P3] from Echelon Biochemical; β-tubulin clone E7 from the Developmental Studies Hybridoma Bank; Alexa-488- and Alexa-572-labelled secondary antibodies from Molecular Probes. The PI-3 kinase detection ELISA kit was purchased from Echelon Biochemical.
A431 cells were maintained in DMEM with 10% heat-inactivated FBS at 37°C (Life Technologies) and 5% CO2. GC wt strain MS11 (Segal et al., 1986) and an isogenic GC pilT mutant was used for infection experiments. The piliation and Opa status of all strains and inocula were monitored by microscopy observation of colony morphology, immunoblotting with antibodies directed against pilin (Virji et al., 1989) and Opa using the pan-Opa antibody 4B12 from M. Blake(Blake et al., 1995). The MS11 pilT null mutant was constructed by transforming chromosomal DNA from strain GT103 (N400 P+, Opa–, pilT::mTnErm) (Wolfgang et al., 1998b) into MS11 and selecting transformants for erythromycin resistance. pilE expression was confirmed by Western blotting and piliation was determined by immunofluorescence microscopy using a rabbit polyclonal antibody raised against total GC proteins. Loss of pilT function was determined by loss of competence for DNA transformation. One pilT candidate was selected and chromsomal DNA isolated and used to backcross (retransform) the same stock of MS11, with tests repeated. Transformation was used to transfer the isopropyl β-D-thiogalactopyranoside (IPTG)-inducible pilT gene in strain MW4 (Wolfgang et al., 1998b) into MS11 to yield the MS11 inducible pilT mutant used in this study. Piliation and Opa phenotypes were monitored by colony morphology.
A431 cells were grown to 80% confluency in 12 well Plate 24 h prior to infection. Adhesion was measured as described previously (Shaw and Falkow, 1988; Waldbeser et al., 1994). Cells were washed three times with serum-free DMEM prior to infection and incubated for 1 h in serum-free medium containing LY294002 (50 µM), wortmannin (10 nM), or medium alone. Bacteria were then added to the cultures at a multiplicity of infection (moi) of 50, and the cultures were incubated for 3 h in the presence of inhibitor. Supernatants were collected and saved for plating. Infected cultures were washed five times in PBS and the washes also saved and plated (see below). Washed cells were then lysed in liquid GCB medium with saponin (0.5%, wt/vol). Serial dilutions of the lysates were plated on supplemented GCB agar, the plates were grown for 48 h at 37°C, 5% CO2, and colony-forming units (cfu) were counted. Adhesion index was measured as the number of cell-associated bacteria divided by the total number of bacteria in the well at the end of the infection.
Gentamicin protection assay
Gentamicin protection assays (Shaw and Falkow, 1988) were performed as described previously (Waldbeser et al., 1994). A431 cells were grown to 80% confluency in 12 well Plate 24 h prior to infection. Cells were washed three times with serum-free DMEM prior to infection and incubated in serum-free medium with LY294002 or wortmannin as described above. Cells were infected for 6 h with MS11 at an moi of 50. Cells were then incubated with Gentamicin (100 µg ml−1) for 1 h at 37°C to kill extracellular and attached bacteria. After washing five times in PBS to remove non-adherent bacteria, the cells were lysed. The lysates were collected in GCB plus saponin (0.5%, wt/vol) and serially diluted and spread onto supplemented GCB plates. The plates were incubated for 48 h at 37°C, 5% CO2, and colony forming units were determined. Invasion index is calculated as the number of Gentamicin-resistant bacteria divided by the number of cell-associated (adhered and invaded) bacteria in the well at the end of the infection.
Akt kinase assay
A431 cells were infected with GC wt strain MS11 and an isogenic GC pilT mutant at an moi of 50 for various lengths of time. Akt was immunoprecipitated from infected cells using an Akt kinase assay kit (Cell Signaling Technology). Akt activity was assayed using an Akt kinase assay kit (Cell Signaling Technology). Akt activity was measured by using a phospho-GSK-3α/β antibody to detect kinase activity of immunoprecipitated Akt after incubation with a GSK fusion peptide. Phospho-Akt antibodies were used to confirm levels of Akt activity. Polyclonal antibodies to Akt were used to confirm total levels of Akt in lysates.
A431 cells were grown on coverslips to 30–50% confluence and infected for 3 h with MS11 at an moi of 100. Coverslips were washed three times in PBS, fixed for 20 min at room temperature in 4% paraformaldehyde, and blocked for 1 h in isotonic PBS containing normal goat serum (3%, v/v) and saponin (0.02%, wt/vol). To determine the presence of [PI(3,4,5)P3] on the outer leaflet of the membrane, saponin was omitted from all blocking and incubation steps. Primary antibody was diluted as specified above in blocking buffer, added to samples, and incubated overnight at 4°C in a moist chamber. The coverslips were rinsed in PBS and incubated with an Alexa conjugated secondary antibody diluted 1:500 in blocking buffer for 1 h at 25°C. Samples to visualize outer membrane proteins and phospholipids were all incubated at 4°C for all steps. The cells were also incubated with the DNA stain DAPI at 1:1000 for 10 min at 25°C to visualize the bacteria and nuclei. Samples were rinsed extensively in PBS before mounting in Fluoromount-G (Fisher Scientific). Negative control samples were processed identically except for the omission of primary antibodies. Optical sections in the z-axis plane were obtained with a Deltavision Restoration Microscope (Applied Precision Instruments, Inc.) fitted with a Nikon 60x oil immersion objective, and the images were processed at a Silicon Graphics workstation with accompanying API software. The images were subsequently exported to Adobe Photoshop and Adobe Illustrator for manuscript preparation.
PI-3 kinase ELISA
ELISA to determine levels of [PI(3,4,5)P3] in A431 cell supernatants was performed using a PI-3 kinase assay kit (Echelon BioSciences) per manufacturer's instructions except the supernatants from infected and uninfected A431 cells were incubated with the [PI(3,4,5)P3] detection reagent at 4°C for 18 h. Cells were infected with GC wt strain MS11 at an moi of 50 for 3 h, and the supernatants collected for analysis. Control wells containing either uninfected cells or cells incubated with bacteria alone were used to eliminate cross-reactivity. The reaction mixtures were then transferred to the detection plate and all other steps performed as indicated. [PI(3,4,5)P3] levels were detected using a MicroMax plate reader using absorbance at 450 nm. Levels of [PI(3,4,5)P3] were quantified by comparing values from wells containing supernatants from cells to the values in a standard curve run concomitantly with test wells.
Adducin phosphorylation following GC infection was determined using a phospho-adducin (Ser 724) antibody (Upstate Biotechnologies). Monoclonal antibodies to β-tubulin were used to normalize protein levels in lysate samples. As a positive control, uninfected cells were treated for 1 h with 0.2 µM TPA (Phorbol-12 Myristate-13–Acetate).
Bacterial aggregation (microcolony formation) assay
Wide-type MS11 bacteria were diluted to approximately 5 × 108 cfu ml−1 in supplemented GCB medium in a total volume of 1 ml. 10 µg of soluble [PI(3,4,5)P3] or [PI(4,5)P2] (Calbiochem) was added at a concentration of 1µg/µl and the samples vortexed for 1 min. 100 µl of the culture was added to a coverdish (Costar) and bacterial aggregation was monitored in real time at room temperature by Differential Interference Contrast microscopy using the Deltavision Restoration Microscope (API).
[PI(3,4,5)P3] bead aggregation assays
1.0 µm yellow-green fluorescent polystyrene beads (Fluospheres, Molecular probes) were covalently coupled with soluble [PI(3,4,5)P3] (Calbiochem) using water-soluble caboiimide (EDAC) as recommended by the supplier. The beads were washed thoroughly in PBS and adsorbed onto a clean coverslip dish in the presence of PBS. BSA (bovine serum albumin) was used to coat mock beads for controls. GC cultures were added as stated for aggregation assays. Differential Interference Contrast and fluorescence microscopy was used to visualize aggregating bacteria and beads respectively. Time-lapsed movies were produced in Quicktime format using a SiliconGraphics workstation with accompanying API software.
RNA extraction and real-time RT-PCR assays
Bacterial RNA was extracted from MS11 using the RNAeasy kit (Qiagen) per manufacturer's instructions. Briefly, wt MS11 was diluted in supplemented GCB medium and incubated with 20 µg of [PI(3,4,5)P3], [PI(4,5)P2] or a vehicle control for 2 h before collection. Cells were lysed in lysozyme (0.4 µg µ−1l wt/vol) containing Tris-EDTA buffer. DNA contamination was removed by treatment with DNA-free (Ambion) and samples were quantified by spectrophotometric analysis (Beckman DU 600). Complementary DNA was generated using the iScript cDNA Synthesis Kit (Bio-Rad) as recommended by the manufacturer. Oligonucleotides complementary to pilT were designed using the primer express software (PE Applied biosystems) to obtain amplicons of the same size. Oligonucleotides to pilT (pilTin.630, 5′-GTC GACCGTATCGTGGACGTATT-3′; pilTout.730, 5′-TTCAGCAG GTTTTGGGAGATGAC-3′) were purchased from Invitrogen.
Real-time PCR using SYBR GREEN PCR Master Mix (Applied Biosystems) was carried out on the ABI Prism 7000 Sequence Detector System (Applied Biosystems). Amplification plots were analysed with the ABI Prism SDS Software package (Version 1.0) and the data were processed using TaqMan Turbocrunch (96) v3. Relative quantification of gene expression was performed by the comparative Ct (threshold cycle) method according to the manufacturer's instructions and as published (Schmittgen et al., 2000). The parameter Ct is defined as the cycle number at which fluorescence passes the fixed threshold. Two independent real-time reverse transcription polymerase chain reaction (RT-PCR) experiments were performed. Dilutions of the reverse transcription product from each experiment were loaded in triplicate for each oligonucleotide couple in all experiments.
We thank Michael Sheetz at Columbia University for critical comments. We thank Patricia Ayala, Heather Howie, Robert Bonnah, Jason Larson and George Niemann for their expert technical assistance; all of the members of the So Laboratory for helpful comments on the manuscript; and J. Scott Wilbur and Chris Langford for their help in preparing the manuscript. This work was funded by NIH Grants AI32493 and AI4997301 to M. So.
- 2001) The pilus-induced Ca2+ flux triggers lysosome exocytosis and increases the amount of Lamp1 accessible to Neisseria IgA1 protease. Cell Microbiol 3: 265–275. , , , , , and (
- 1999) Phosphatidylethanolamine recognition promotes enteropathogenic E. coli and enterohemorrhagic E. coli host cell attachment. Microb Pathog 27: 289–301. , , , , , and (
- 2003) Calcium signalling: dynamics, homeostasis and remodelling. Nat Rev Mol Cell Biol 4: 517–529. , , and (
- 2001) Light regulation of type IV pilus-dependent motility by chemosensor-like elements in Synechocystis PCC6803. Proc Natl Acad Sci USA 98: 7540–7545. , , and (
- 1995) Gonococcal opacity: lectin-like interactions between Opa proteins and lipooligosaccharide. Infect Immun 63: 1434–1439. , , , and (
- 2003) Phosphatidylinositol 3-kinases in carcinoembryonic antigen-related cellular adhesion molecule-mediated internalization of Neisseria gonorrhoeae. J Biol Chem 278: 14037–14045. , , , , , , and (
- 1997) Receptor affinity purification of a lipid-binding adhesin from Haemophilus influenzae. J Infect Dis 175: 77–83. , , and (
- 2002) The phosphoinositide 3-kinase pathway. Science 296: 1655–1657. (
- 1993) Interaction cloning of protein kinase C substrates. J Biol Chem 268: 6858–6861. , , , and (
- 1999) Pili binding to asialo-GM1 on epithelial cells can mediate cytotoxicity or bacterial internalization by Pseudomonas aeruginosa. Infect Immun 67: 3207–3214. , , , and (
- 2002) Strategic targeting of essential host–pathogen interactions in chlamydial disease. Curr Drug Targets Infect Disord 2: 201–216. , , and (
- 2003) Regulation of transbilayer plasma membrane phospholipid asymmetry. J Lipid Res 44: 233–242. (
- 1995) 35H, a sequence isolated as a protein kinase C binding protein, is a novel member of the adducin family. J Biol Chem 270: 25534–25540. , , , , , , and (
- 1998) Annexin V-affinity assay: a review on an apoptosis detection system based on phosphatidylserine exposure. Cytometry 31: 1–9. , , , , and (
- 2000) A receptor for phosphatidylserine-specific clearance of apoptotic cells. Nature 405: 85–90. , , , , , and (
- 1998) Co-expression of Fas and Fas-ligand on the surface of influenza virus-infected cells. Cell Death Differ 5: 426–431. , , , and (
- 1999) Apoptosis in macrophages and alveolar epithelial cells during early stages of infection by Legionella pneumophila and its role in cytopathogenicity. Infect Immun 67: 862–870. , and (
- 1997) Binding of vitronectin to opa-expressing Neisseria gonorrhoeae mediates invasion of HeLa cells. Infect Immun 65: 3857–3866. , , , , , and (
- 2001) Rapid, transient phosphatidylserine externalization induced in host cells by infection with Chlamydia spp. Infect Immun 69: 1109–1119. , and (
- 2004) Chlamydia-infected cells continue to undergo mitosis and resist induction of apoptosis. Infect Immun 72: 451–460. , , , , and (
- 2002) Aquifex aeolicus PilT, homologue of a surface motility protein, is a thermostable oligomeric NTPase. J Bacteriol 184: 6465–6471. , , and (
- 2001a) Activation of ErbB2 receptor tyrosine kinase supports invasion of endothelial cells by Neisseria meningitidis. J Cell Biol 155: 133–143. , , , , and (
- 2001b) Phosphatidylserine (PS) induces PS receptor-mediated macropinocytosis and promotes clearance of apoptotic cells. J Cell Biol 155: 649–659. , , , , , , et al. (
- 2000) Novel function of phosphoinositide 3-kinase in T cell Ca2+ signaling. A phosphatidylinositol 3,4,5-trisphosphate-mediated Ca2+ entry mechanism. J Biol Chem 275: 16242–16250. , , , , , , and (
- 1996) Acidic pH changes receptor binding specificity of Helicobacter pylori: a binary adhesion model in which surface heat shock (stress) proteins mediate sulfatide recognition in gastric colonization. Infect Immun 64: 2643–2648. , , , and (
- 1997) Host–pathogen interactions during entry and actin-based movement of Listeria monocytogenes. Annu Rev Genet 31: 113–138. , and (
- 1997) Membrane cofactor protein (MCP or CD46) is a cellular pilus receptor for pathogenic Neisseria. Mol Microbiol 25: 639–647. , , , and (
- 1998) Cell signaling by the type IV pili of pathogenic Neisseria. J Biol Chem 273: 21777–21782. , , , and (
- 1998) Chemotaxis in a gliding bacterium. Proc Natl Acad Sci USA 95: 11957–11962. , and (
- 2001) Lipid chemotaxis and signal transduction in Myxococcus xanthus. Trends Microbiol 9: 126–129. , and (
- 2002) Specific entry of Helicobacter pylori into cultured gastric epithelial cells via a zipper-like mechanism. Infect Immun 70: 2108–2120. , , , , and (
- 2002) CD46 is phosphorylated at tyrosine 354 upon infection of epithelial cells by Neisseria gonorrhoeae. J Cell Biol 156: 951–957. , , , , , and (
- 1997) The Neisseria type 2 IgA1 protease cleaves LAMP1 and promotes survival of bacteria within epithelial cells. Mol Microbiol 24: 1083–1094. , , , , , , et al. (
- 1992) The glycerolipid receptor for Helicobacter pylori (and exoenzyme S) is phosphatidylethanolamine. Infect Immun 60: 2470–2474. , , and (
- 2002) Single pilus motor forces exceed 100 pN. Proc Natl Acad Sci USA 99: 16012–16017. , , , , , and (
- 1991) Phase variation of the opacity outer membrane protein controls invasion by Neisseria gonorrhoeae into human epithelial cells. EMBO J 10: 1307–1315. , , and (
- 2000a) Interactions of pathogenic neisseriae with epithelial cell membranes. Annu Rev Cell Dev Biol 16: 423–457. , and (
- 1999) Type IV pili of pathogenic neisseriae elicit cortical plaque formation in epithelial cells. Mol Microbiol 32: 1316–1332. , , and (
- 2000b) Pilus retraction powers bacterial twitching motility. Nature 407: 98–102. , , and (
- 2002) Role of Cbl in shear-activation of PI 3-kinase and JNK in endothelial cells. Biochem Biophys Res Commun 292: 892–899. , , , , (
- 2000) Phosphatidylinositol 4,5-bisphosphate functions as a second messenger that regulates cytoskeleton-plasma membrane adhesion [In Process Citation]. Cell 100: 221–228. , , , , , , et al. (
- 2000) Quantitative reverse transcription-polymerase chain reaction to study mRNA decay: comparison of endpoint and real-time methods. Anal Biochem 285: 194–204. , , , , , and (
- 1986) Antigenic variation of gonococcal pilus involves assembly of separated silent gene segments. Proc Natl Acad Sci USA 83: 2177–2181. , , , and (
- 1988) Model for invasion of human tissue culture cells by Neisseria gonnorrhoeae. Infect Immun 56: 1625–1632. , and (
- 2001) Direct observation of extension and retraction of type IV pili. Proc Natl Acad Sci USA 98: 6901–6904. , and (
- 2003) Bacteria-host communication: the language of hormones. Proc Natl Acad Sci USA 100: 8951–8956. , , , , and (
- 1993) Structure-function and biogenesis of the type IV pili. Annu Rev Microbiol 47: 565–596. , and (
- 1996) Adherence to lipids and intestinal mucin by a recently recognized human pathogen, Campylobacter upsaliensis. Infect Immun 64: 4060–4066. , , , , and (
- 1989) Identification of epitopes recognized by monoclonal antibodies SM1 and SM2 which react with all pili of Neisseria gonorrhoeae but which differentiate between two structural classes of pili expressed by Neisseria meningitidis and the distribution of their encoding sequences in the genomes of Neisseria spp. J Gen Microbiol 135: 3239–3251. , , , , and (
- 1994) The opaH locus of Neisseria gonorrhoeae MS11A is involved in epithelial cell invasion. Mol Microbiol 13: 919–928. , , , , , , and (
- 1993) Mechanotransduction across the cell surface and through the cytoskeleton. [See comments]. Science 260: 1124–1127. , , and (
- 1998a) Suppression of an absolute defect in type IV pilus biogenesis by loss-of-function mutations in pilT, a twitching motility gene in Neisseria gonorrhoeae. Proc Natl Acad Sci USA 95: 14973–14978. , , , , and (
- 1998b) PilT mutations lead to simultaneous defects in competence for natural transformation and twitching motility in piliated Neisseria gonorrhoeae. Mol Microbiol 29: 321–330. , , , , , and (
- 1997) The Myxococcus xanthus pilT locus is required for social gliding motility although pili are still produced. Mol Microbiol 23: 109–121. , , and (
- 2004) Activation of the phosphatidylinositol 3-kinase/Akt pathway contributes to survival of primary epithelial cells infected with the periodontal pathogen Porphyromonas gingivalis. Infect Immun 72: 3743–3751. , , , and (