Escherichia coli Cytotoxic Necrotizing Factor 1 Blocks Cell Cycle G2/M Transition in Uroepithelial Cells
http://www.100md.com
《感染与免疫杂志》
Department of Drug Research and Evaluation
Department of Technology and Health, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161, Rome, Italy
ABSTRACT
Evidence is accumulating that a growing number of bacterial toxins act by modulating the eukaryotic cell cycle machinery. In this context, we provide evidence that a protein toxin named cytotoxic necrotizing factor 1 (CNF1) from uropathogenic Escherichia coli is able to block cell cycle G2/M transition in the uroepithelial cell line T24. CNF1 permanently activates the small GTP-binding proteins of the Rho family that, beside controlling the actin cytoskeleton organization, also play a pivotal role in a large number of other cellular processes, including cell cycle regulation. The results reported here show that CNF1 is able to induce the accumulation of cells in the G2/M phase by sequestering cyclin B1 in the cytoplasm and down-regulating its expression. The possible role played by the Rho GTPases in the toxin-induced cell cycle deregulation has been investigated and discussed. The activity of CNF1 on cell cycle progression can offer a novel view of E. coli pathogenicity.
INTRODUCTION
It is largely recognized that bacterial pathogens use a variety of strategies, including the production of protein toxins, to manipulate host cell functions. In the past decade, evidence has accumulated on the ability of certain bacterial toxins to interfere with the eukaryotic cell cycle machinery, either promoting cell proliferation (30, 44) or inducing cell cycle arrest (7, 25). These protein toxins belong to the family of so-called "cyclomodulins," which actively deregulate the passage of cells throughout the host cell cycle (reviewed in reference 29). In this work, we show that a block in G2/M transition can be induced by cytotoxic necrotizing factor 1 (CNF1), a 110-kDa monomeric protein toxin frequently produced by pathogenic Escherichia coli strains associated with urinary tract infections. CNF1 is a chromosomally encoded toxin that activates the small GTP-binding proteins of the Rho family (Rho, Rac, and Cdc42), regulatory proteins that behave as molecular switches, cycling between an active GTP-bound state and an inactive GDP-bound state. CNF1 acts by deamidating a specific glutamine residue in the switch 2 domain that is crucial in GTP hydrolysis (glutamine 63 in Rho [18, 36, 37] or glutamine 61 in Rac and Cdc42 [24]), thus locking the G protein in the active state. This toxin activity confers new properties on epithelial cells, including the cytoskeleton-dependent ability to behave as professional phagocytes (13, 17).
Beside controlling the actin cytoskeleton organization (31), the Rho signaling pathways are also intimately involved in transducing mitogenic signals to the translational apparatus and play a pivotal role in different points of cell cycle regulation (28). Indeed, Rho, Rac, and Cdc42 GTPases have been demonstrated to participate in the control of G1/S progression, primarily through regulation of expression of cyclins and cyclin-dependent kinase inhibitors (42).
Hence, considering the Rho-dependent ability of CNF1 to interfere with cytokinesis, thus causing multinucleation (15), and to inhibit cell differentiation (41) in different mammalian cells, we asked whether this toxin could also somehow affect the cell cycle regulatory pathways. This question was addressed by using the human uroepithelial cell line T24, where CNF1 has previously been reported to enhance the transcription and release of proinflammatory cytokines (14). The results reported here allow the inclusion of CNF1 in the family of toxins that influence cell cycle progression and point out the role played by the Rho GTPases in this phenomenon.
MATERIALS AND METHODS
Cell cultures and treatments. T24 cells, a human epithelial cell line derived from a bladder carcinoma (American Type Culture Collection, Manassas, VA), were cultured in McCoy's 5A medium supplemented with 10% fetal calf serum, 1% nonessential amino acids, 100 μg/ml of streptomycin, and 100 U/ml of penicillin.
CNF1 was obtained from strain 392 ISS (kindly provided by V. Falbo, Rome, Italy) and purified as previously described (13). The plasmid coding for a nontoxic mutant of CNF1 that completely lacks the enzymatic activity (C866S) was kindly provided by E. Lemichez (Nice, France) and prepared as previously described (18). To rule out the possibility that the effects observed could be due to lipopolysaccharide contamination, a heat-inactivated (98°C for 10 min) CNF1 protein toxin preparation was used.
In all experiments, T24 cells were seeded at a concentration of 2 x 104/ml, and 24 h after seeding, cells were exposed for 6, 24, 48, 72, 96, 120, and 144 h to 10–10 M CNF1, heat-inactivated CNF1, or mutant CNF1 [CNF1(C866S)].
Rho GTPase transfections. T24 cells were seeded on 6-cm petri dishes and cotransfected with 2 μg of plasmid DNA encoding dominant-positive forms of either Rho, Rac, or Cdc42 Rho GTPases (RhoV14, RacV12, or Cdc42V12) and green fluorescent protein (GFP) (pEGFP; Clontech) by using Lipofectamine (Invitrogen) according to the manufacturer's instructions.
Fluorescence microscopy. Control, CNF1-treated, and transfected T24 cells were fixed with 3.7% paraformaldehyde in phosphate-buffered saline (PBS) (pH 7.4) for 10 min at room temperature. After a wash in the same buffer, cells were permeabilized with 0.5% Triton X-100 (Sigma) in PBS (pH 7.4) for 10 min at room temperature.
For F-actin visualization, cells were stained with fluorescein isothiocyanate-phalloidin (FITC) (Sigma; working dilution, 0.5 μg/ml) for 30 min at 37°C. For cyclin B1 detection, cells were stained with a monoclonal anti-human cyclin B1 antibody (BD Pharmingen; diluted 1:100). After a wash, an FITC-conjugated anti-mouse secondary antibody (Sigma; diluted 1:100) was added to the samples for 30 min at 37°C.
Finally, after extensive washing, coverslips were mounted with glycerol-PBS (2:1, vol/vol) and analyzed with an Olympus Bx51 Nikon Microphot fluorescence microscope.
Activated Rho GTPase pull-down assay. A pull-down assay was performed as previously described (41). Briefly, cells were lysed in the appropriate buffers, and the cleared lysates were incubated with 80 μg of either the glutathione S-transferase (GST)-PAK-CD (for Rac/Cdc42) or the GST-Rhotekin (for Rho) fusion protein (Cytoskeleton), bound to glutathione-coupled Sepharose beads (Amersham), for 40 min at 4°C. After three washes in the appropriate buffer, the bound proteins were eluted in sample buffer, subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (12% acrylamide), and electrically transferred to polyvinylidene difluoride membranes. Membranes were blocked with TBS-T (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 0.02% Tween 20) containing 2% skim milk (Bio-Rad) for 30 min at room temperature and were then incubated overnight at 4°C with primary antibodies diluted in the same buffer. The following antibodies were used: monoclonal anti-RhoA (Cytoskeleton; diluted 1:500), monoclonal anti-Rac1 (Transduction Laboratories; diluted 1:3,500), and monoclonal anti-Cdc42 (Santa Cruz Biotechnology; diluted 1:500). After extensive washing, immunocomplexes were detected with horseradish peroxidase-conjugated species-specific secondary antibodies (goat anti-mouse antibody, diluted 1:3,000 [Upstate]; goat anti-rabbit antibody, diluted 1:22,000 [Jackson]), followed by an enhanced chemiluminescence reaction (Pierce). Whole-cell lysates (2% of input) were analyzed in parallel.
Cell growth and viability. Control and CNF1-treated cells, detached from the culture dish with 10 mM EDTA (pH 7.4) and 0.25% trypsin in PBS (pH 7.4), were counted, and cell viability was determined by a trypan blue (80 μM; GIBCO BRL-Life Technologies) exclusion method. In each experiment, cell counts were conducted five times. Data represent means ± standard deviations (SD) from three separate experiments performed in duplicate. Growth inhibition was expressed as the percentage of the ratio between CNF1-treated and control cells.
Cytofluorimetric analysis. (i) Cell cycle analysis. T24 cells were trypsinized, washed twice in cold PBS, fixed with ice-cold 70% ethanol, and kept at 4°C for at least 2 h. Before the FACScan acquisition, cells were rehydrated in PBS and stained with a PBS solution containing 50 μg/ml propidium iodide (PI) and 10 μg/ml of RNase at room temperature for at least 20 min. A minimum of 10,000 events were acquired in linear mode with a FACScan (Becton Dickinson) and then analyzed with ModFit software. Transfected and control T24 cells were collected, washed twice in PBS, and fixed for 10 min with cold ethanol (70%, vol/vol) on ice. After this time, samples were centrifuged and resuspended in PBS containing PI at the final concentration of 30 μg/ml and 100 μg/ml of RNase. After 30 min of incubation at room temperature under dark conditions, a minimum of 10,000 GFP-positive events, gated by a dot plot of GFP versus forward scatter, were acquired and analyzed as described above.
(ii) Cyclin B1 detection. Control and treated T24 cells were trypsinized and washed twice in cold PBS. Cells were then fixed in 2% paraformaldehyde in PBS for 15 min and then permeabilized in 0.05% Triton X-100 in PBS for 10 min at 4°C. Samples were washed in cold PBS and incubated for 30 min at 4°C with a monoclonal anti-human cyclin B1 antibody (BD Pharmingen; diluted 1:100) After a wash, an FITC-conjugated anti mouse secondary antibody (Sigma; diluted 1:100) was added to the samples. Samples were washed twice, resuspended in PBS containing 5 μg/ml PI, and incubated for 1 h at 4°C. The fluorescence intensity of cyclin B1 was acquired in linear mode and plotted on the y axis versus the linear fluorescence intensity of PI plotted on the x axis. A minimum of 10,000 events were acquired by the FACScan (BD) and analyzed by CellQuest software (BD).
(iii) Apoptosis analysis with an annexin V-FITC kit. To detect translocation of phosphatidylserine from the inner face to the outer face of the plasma membrane in the first step of apoptosis, an annexin V-FITC kit (Medical & Biological Laboratories Co., Ltd.) was used (9, 26). Control and CNF1-treated cells were detached with EDTA and trypsin solutions. Cell suspensions were washed in ice-cold PBS, collected by centrifugation, and resuspended in annexin V binding buffer (10 mM HEPES-NaOH [pH 7.4], 140 mM NaCl, 2.5 mM CaCl2). Annexin V-FITC and propidium iodide were added to the binding buffer for 10 min at room temperature, and the cells were directly acquired by FACScan (BD). Emissions were collected at wavelengths of 530 nm for FITC and 650 nm for PI and acquired in log mode; a minimum of 10,000 events were acquired and analyzed with CellQuest software (BD).
DNA ladder assay. For the DNA ladder assay, 5 x106 cells were scraped, washed in PBS, and collected by centrifugation. Cell pellets were resuspended in lysis buffer (10 mM Tris-HCl [pH 7.4], 1 mM EDTA, and 0.2% Triton X-100) containing 0.1 μg/ml proteinase K and then incubated at 37°C for 90 min. DNA was cleared from the lysates by centrifugation and then extracted by addition of 0.8 M NaCl and an equal volume of chilled isopropanol. Following an overnight incubation at –20°C, samples were centrifuged, and the pelleted DNA was resuspended in sterile water and treated with 0.5 μg/ml of RNase A at 65°C for 20 min. DNA was then analyzed by gel electrophoresis on a 1% agarose gel stained with ethidium bromide (0.5 μg/ml).
RESULTS
CNF1 reorganizes the actin cytoskeleton and modulates the Rho GTPases in T24 cells. We started our investigation by performing an analysis of the actin cytoskeleton organization of T24 cells challenged with CNF1 (Fig. 1a to e). As expected, CNF1 exposure (Fig. 1 b to e) induced a profound reorganization of the actin cytoskeleton and rendered cells more flattened and spread out than the control cells (Fig. 1a), with a consequent remarkable increase in the cell size. In particular, at early times of exposure, CNF1 induced the formation of stress fibers (Fig. 1b) that, as the incubation time with the toxin was prolonged, progressively disappeared, concomitantly with the promotion of membrane ruffles (Fig. 1c to e). It is worth noting that, in T24 cells, multinucleation, which is a typical feature of CNF1 activity in other epithelial cell lines (4), is a more limited phenomenon; no more than 30% of T24-treated cells underwent multinucleation after 72 h of toxin exposure. However, even if they remained mononucleated, most cells displayed nuclei larger than those observed in control cells (data not shown). Hence, since cytoskeleton-dependent phenomena are chiefly controlled by the Rho GTPases (12) and it is well known that CNF1 is able to activate/deactivate the Rho GTPases with different timing depending on the cell line used (11, 41), we verified by pull-down experiments the activation state of Rho GTPases in T24 cells following CNF1 exposure (Fig. 1f to h). CNF1 was able to activate Rho, Rac, and Cdc42 GTPases in a time-dependent manner. In particular, Rho activation, already evident after 6 h of CNF1 challenge, reached a maximum after 24 h of toxin exposure (Fig. 1f). The active form of Rac (Fig. 1g), faintly detectable following 6 h of CNF1 treatment, strongly increased at 24 h of toxin treatment and reached a maximum at 48 h. Similarly to Rac GTPase, Cdc42 became strongly activated following 24 h of CNF1 exposure and remained active until 48 h (Fig. 1 h). In keeping with results obtained for other cell types (11, 41), activation of Rho GTPases was transient. In fact, the active form of Rho strongly decreased starting from 48 h of toxin treatment, while both activated Rac and Cdc42 diminished at 72 h, although this was much more evident for Cdc42. As expected, by analyzing the activation state of Rho GTPases up to 144 h of incubation with CNF1, we observed that, starting from 96 h, the active forms of Rho, Rac, and Cdc42 were nearly comparable to the control levels (data not shown).
FIG. 1. CNF1 induces actin cytoskeleton reorganization and activation of Rho family proteins in T24 cells. (a to e) Fluorescence micrographs showing F-actin organization in T24 control cells (a) and in cells challenged with CNF1 for different times (b to e). CNF1 provokes a remarkable reorganization of F-actin into stress fibers and membrane ruffles. (f to h) Immunoblots showing the activation kinetics of RhoA (f), Rac1 (g), and Cdc42 (h) GTPases in control and CNF1-treated cells, detected by a pull-down assay. Activated forms of RhoA (Rho-GTP), Rac1 (Rac-GTP), and Cdc42 (Cdc42-GTP), as well as the total amounts of GTPases, are shown. CNF1 induces a time-dependent activation of Rho GTPases followed by their deactivation. Equal protein loading was confirmed by -tubulin detection. Bar, 10 μm.
Changes in actin cytoskeleton organization induced by CNF1 reflected the activation states of the different members of the Rho GTPase family. Indeed, while Rho-induced stress fibers were assembled at 6 h of CNF1 challenge and completely disappeared at 48 h, ruffles and filopodia, which are the consequences of Rac and Cdc42 activation, respectively, appeared only following 24 h of toxin treatment and became particularly evident starting from 48 h.
CNF1 influences cell cycle progression in T24 cells. It is well known that, besides controlling the actin cytoskeleton organization, Rho signaling pathways are also intimately involved in cell proliferation and cell cycle regulation (28). Hence, we wondered whether activation of CNF1-dependent Rho GTPases in T24 cells could somehow influence cell growth and the passage of cells through the cell cycle. For this purpose, control and CNF1-treated cells were detached, and cell viability was determined by a trypan blue exclusion test. We found that in T24 cells, CNF1 caused a time-dependent inhibition of cell growth that reached a maximum after 72 h of toxin exposure (Fig. 2a), apparently without induction of death, since dead cells represented 3 to 4% of the total population both in control and in treated cells (data not shown). To determine whether the decreased growth could be linked to an alteration of the cell cycle, we performed a flow cytometric analysis of DNA content. The results obtained are reported in Fig. 2b as a percentage of the ratio between treated and control cells, while two representative histograms of untreated and treated cells are shown in Fig. 2c and d, respectively. The distribution of cell cycle phases was significantly modified by the contact with CNF1 (Fig. 2b to d). In particular, the toxin was able to induce a progressive accumulation of cells in G2/M phase, and this effect, starting as early as after 24 h of CNF1 exposure, became strongly evident after 144 h of toxin treatment (Fig. 2b). In order to verify whether the perturbation of the cell cycle could be a more general response to CNF1, we evaluated this effect on a different uroepithelial cell line, the bladder cell line 5637, already used to investigate the activity of this toxin (27). Our results showed that in this cell type also, CNF1 caused a time- and dose-dependent accumulation of cells in G2/M phase (Table 1). It is worth noting that, when T24 cells were exposed to heat-inactivated CNF1 or to the CNF1(C866S) mutant, no accumulation of cells in G2/M phase was observed (data not shown). We thus wondered whether the activation of Rho GTPases by CNF1 could be involved in shifting cells to the G2/M phases. To answer this question, we cotransfected T24 cells with GFP and with plasmids encoding constitutively activated forms of the Rho GTPases, RhoV14, RacV12, and Cdc42V12, for 24 h, and then we analyzed the cell cycle phase distribution of transfected cells. We found that the expression of each single activated Rho GTPase caused a strong accumulation of cells in both the S and G2/M phases (Table 2), in accordance with previous reports showing the influence of Rho GTPases on the cell cycle (6, 29). Hence, these results support the role played by the Rho GTPases in the CNF1-induced block of cell cycle G2/M transition in T24 cells.
FIG. 2. CNF1 induces recruitment of T24 cells in the G2/M cell cycle phase. (a) Graph showing the effect of CNF1 on cell growth. Results are reported as percentages of growth inhibition, determined as the percentage of the ratio between CNF1-treated and control cells, and represent the means ± SD from three separate experiments performed in duplicate. The toxin induces a time-dependent inhibition of cell growth. (b) Graph showing the flow cytometric analysis of cell cycle phase distribution, expressed as the ratio between CNF1-treated and control T24 cells. Note that after 144 h of CNF1 exposure, a high percentage of treated cells are arrested in the G2 stage of the cell cycle. (c and d) Representative histograms showing the DNA content of control T24 cells (c) or cells challenged with CNF1 for 144 h (d).
CNF1 down-regulates and relocalizes cyclin B1 in T24 cells. It is largely acknowledged that progression from G2 to M phase is driven by activation of cyclin-dependent kinase 1 (Cdk1)/cyclin B1 complex and that cyclin B1 expression fluctuates throughout the cell cycle, peaking in G2/M (21, 38). Hence, we analyzed the effects of CNF1 on cyclin B1 expression by combined flow cytometric analysis of cyclin B1 and DNA (20). As reported in Fig. 3a and b, CNF1 induced a significant reduction of the cyclin B1 level with respect to control cells, whatever the time of exposure considered. The mitotic function of Cdc2/cyclin B1, however, is also triggered by spatial control (21). In fact, cyclin B1 is predominantly cytoplasmic throughout G2 phase, until its very rapid translocation into the nucleus, which precedes the nuclear envelope breakdown. Immunofluorescence investigations indicated that CNF1 was able to induce a cellular redistribution of this protein, since cyclin B1 was localized in the nuclei of control cells (Fig. 3c) and most of the signal was cytoplasmic in cells exposed to the toxin (Fig. 3d). Notably, cell cycle arrest and down-regulation of cyclin B1 were not detected in T24 cells treated with heat-inactivated CNF1 or with the CNF1(C866S) mutant (data not shown). In order to investigate the link between Rho GTPase activation and cyclin B1, we have analyzed the localization of cyclin B1 in transfected T24 cells with plasmids encoding constitutively activated forms of Rho GTPases, RhoV14, RacV12, and Cdc42V12. As shown in Fig. 3, in cells transfected with RhoV14, cyclin B1 was predominantly sequestered in the cytoplasm (Fig. 3h). The same results were obtained when T24 cells were transfected with RacV12 and Cdc42 V12; cyclin B1 remained predominantly in the cytosol (data not shown).
FIG. 3. CNF1 induces down-regulation and cytoplasmic sequestration of cyclin B1. (a and b) Graphs showing the fluorescence intensity of cyclin B1 in control and CNF1-treated T24 cells. The results were obtained from four different experiments and are expressed as arbitrary units. Cyclin B1 is down-regulated in cells exposed to the toxin. Note that, in order to avoid cell confluence, two separate experiments with different T24 cell seedings were performed, one for 24 to 72 h (a) and the other for 96 to 120 h (b). (c and d) Fluorescence micrographs of T24 cells that were either untreated (c) or treated with CNF1 for 72 h (d), stained for cyclin B1. (e to h) Fluorescence micrographs of T24 cells transfected with GFP (e) and stained for cyclin B1 (f) or cotransfected with RhoV14/GFP (g) and stained for cyclin B1 (h). CNF1 exposure, as well as RhoV14 transfection, causes the sequestration of cyclin B1 in the cytoplasm. Bar, 10 μm.
The CNF1-induced G2/M arrest does not bring about apoptosis. It is known that accumulation in the G2/M phase of the cell cycle is one of the key events that precede apoptosis (38), and since the ability of CNF1 to act as a pro- or antiapoptotic factor is under debate (16, 27), we have evaluated the possibility that toxin-treated T24 cells could respond to the cell cycle accumulation in G2/M with apoptotic cell death. Control and CNF1-treated cells were thus stained with annexin V and analyzed by flow cytometry. Challenge with the toxin did not induce the exposure of annexin V binding sites, and furthermore, the percentages of cells that showed positive fluorescence for both annexin V and PI were similar for control and treated cells (Fig. 4a and b). To further confirm the inability of CNF1 to induce apoptosis in T24 cells, we analyzed the eventual DNA fragmentation in T24 cells exposed to CNF1. As shown in Fig. 4c, no DNA fragmentation was induced by CNF1 exposure even after 144 h of toxin treatment, thus demonstrating that CNF1 could not cause apoptosis in T24 cells despite the impairment in cell cycle progression. It is noteworthy that T24 cells were able to undergo DNA fragmentation when challenged with a proapoptotic factor such as staurosporine (10 mM), used as a positive control (Fig. 4c, lane D). Moreover, we used here, as a positive control, the bladder cell line 5637, which has previously been reported to undergo apoptosis when exposed to CNF1 (27). We found that 5637 cells underwent apoptosis, in terms of DNA fragmentation (data not shown), only at doses of CNF1 higher (3 x 10–10 M) than those commonly used to induce a cell response (3, 11). In T24 cells, the same higher dose of CNF1 remained unable to cause cell death.
FIG. 4. The CNF1-induced G2/M accumulation does not bring about apoptosis. (a and b) Cytofluorimetric analysis of T24 control cells (a) and of cells treated with CNF1 for 144 h (b), stained with annexin V. (c) Analysis of DNA fragmentation by DNA laddering in T24 cells. Staurosporine (10 μM) was used as a positive control for DNA fragmentation. No signs of apoptosis were observed following CNF1 exposure, whatever the method of detection used.
DISCUSSION
In this work, we provide evidence that the Rho-activating protein toxin CNF1 was able to induce, in T24 uroepithelial cells, an accumulation of cells in the G2/M phase of the cell cycle, as well as to confine the cyclin B1 in the cytoplasm and significantly decrease its expression. Our findings are in accordance with the work of De Rycke and coworkers, who reported that HeLa cells infected with E. coli strains producing CNF1, but not with those defective in CNF1 production, were totally blocked at stage G2 of the cell cycle (10).
Nowadays, the abilities of several bacterial toxins to interfere with the eukaryotic cell cycle through specific effectors and mechanisms are rapidly emerging (reviewed in reference 29). Indeed, it is well known that toxins belonging to the cytolethal distending toxin (CDT) family, produced by a broad array of pathogenic bacteria, cause irreversible cell cycle arrest or death in the target cells by using different strategies (reviewed in reference 40). For example, the Haemophilus ducreyi CDT causes cell cycle arrest and apoptosis via the DNA damage checkpoint pathways, with an effect that is cell specific (7), whereas the Campylobacter jejuni CDT blocks the host cell cycle at the G2/M transition through a complex mechanism that leads to the inactivation of the Cdc25 phosphatase and, as a consequence, to the maintenance of the cyclin B1-Cdk1 complex in the phosphorylated (inactive) form (43). In a similar fashion, the effector protein Cif, recently identified in enteropathogenic and enterohemorrhagic Escherichia coli strains, acts by inducing a sustained phosphorylation of Cdk1 and cell cycle arrest in G2 (25). On the other hand, the Helicobacter pylori homomultimeric exotoxin VacA and the polyketide mycolactone, secreted by Mycobacterium ulcerans, trigger G1 cell cycle block by mechanisms that remain to be deciphered (8, 19), while the Pasteurella multocida toxin and H. pylori CagA act on the mitogen-activated protein kinase cascade, leading to up-regulation of cyclin D1 and moving the cell into the S phase (30, 44).
Concerning the mechanism by which CNF1 causes the accumulation of cells in the G2/M phase, it appeared evident that the enzymatic activity on the Rho GTPases was crucial in the perturbation of the cell cycle, a result not reachable with the mutant CNF1 that bears an alteration of the active site (37). In fact, although principally known for their role in signal transduction processes leading to cytoskeleton-dependent responses, the Rho GTPases are also involved in a huge number of other cellular functions, including cell cycle regulation. It has been demonstrated, for example, that Rho, Rac, and Cdc42 promote G1/S transition by influencing the expression levels of cyclin D1 and those of the cyclin-dependent kinase inhibitors p27 and p21 (reviewed in reference 6). In accordance, in the present work, we showed that transfection of T24 cells with plasmids encoding the single constitutively active forms of Rho, Rac, and Cdc42 G proteins caused a strong accumulation of cells in the S and G2/M phases. On the other hand, the whole activation of both Rho and Rac/Cdc42 GTPases by CNF1 resulted in a significant increase in the percentage of cells only in the G2/M phase; we did not observe any effect on the G1/S transition. This apparent discrepancy can be explained if we consider that the simultaneous activation of the three Rho subfamilies, as in the case of CNF1, could trigger a cellular response that diverges from that induced by the single Rho proteins. In fact, the molecular pathways triggered by the different Rho family members antagonize each other (34, 35), and a dynamic balance between Rho and the Rac/Cdc42 subfamily seems to be crucial for the occurrence of different biological processes. For example, in fibroblasts, activation of Rac has been shown to down-regulate Rho activity (35), and whereas Cdc42 and Rac promote membrane protrusions, Rho promotes membrane retraction through contractile actin and myosin filaments (12). Not only do Rho- and Rac/Cdc42-induced phenotypes appear mutually exclusive, but also Rho- and Rac/Cdc42-mediated pathways regulate each other. Such a phenotypic antagonism reflects their activation states in various cell types, where the levels of active Rac/Cdc42 are inversely correlated with the levels of active Rho. This is true, for example, in epithelial cells that are growing or undergoing cadherin-dependent cell-cell contact formation (35). Finally, we have recently shown that the maintenance of a dynamic balance between Rho and Rac/Cdc42 activities is crucial for correct muscle cell differentiation (41).
This last finding appears in line with the ability of CNF1 to interfere with cell cycle progression, since the irreversible exit from the cell cycle is an essential prerequisite for the differentiation program onset. On the other hand, the ability of CNF1 to block the cell cycle G2/M transition in uroepithelial cells is consistent with a strategy that favors the induction of limited damage to the host as a means to stimulate specific cellular responses rather than rapid cell death. Hence, by inducing cell cycle arrest, CNF1 could influence the renewal and developmental processes of the epithelium, prolonging E. coli colonization through inhibition of cell shedding. Moreover, the lack of epithelial renewal would produce epithelial lesions that might transiently alter the functional capacity of the tissues or facilitate the bacterial invasion of underlying tissues. We should also consider that different bacterial protein toxins able to interfere with the host cell cycle represent a long-term risk of carcinogenesis (25) and that the possible link between bacterial infection and tumorigenesis is an emerging theme in bacterial pathogenesis (5). Helicobacter pylori is now recognized as an important etiological cofactor of gastric cancers, and several other bacteria, such as Salmonella enterica serovar Typhi, Citrobacter rodentium, and Bartonella spp., are suspected to be involved in carcinogenesis (23). In this context, it is possible to speculate that the genotoxic activity of cyclomodulins such as CDTs or Cif could act as a promoting factor for the development of cancer and similarly that CNF1 could be considered a potential promoter in cell transformation.
Although we are aware that the majority of studies on CNF1 have been conducted on cancer cells and not on primary cultures, cell lines remain invaluable means to dissect the pathways controlled by toxins. The hypothesis of CNF1 as a possible transforming factor, in fact, has been suggested by evidence obtained with cell lines that disclose the ability of CNF1 to (i) induce the transactivation of nuclear factor B (NF-B) (3), a transcription factor whose activity has been related to cell transformation; (ii) stimulate the expression of Cox-2, a typical marker of cancer (22); and (iii) protect epithelial cells from apoptotic stimuli (16). Concerning apoptosis, it is worth noting that, in contrast to the effect of H. ducreyi CDT (7), the cell spreading and block in G2 due to CNF1 were not followed by cell death, and in our hands, the uroepithelial cell line 5637, previously reported to undergo apoptosis when exposed to CNF1 (27), displayed time- and dose-dependent accumulation of cells in G2/M phase but underwent DNA fragmentation only at doses higher than those commonly used to trigger a cell response (3, 11, 16). All this further supports our view of CNF1 as a prosurvival factor that might possibly contribute to the long-term risk of transformation.
Moreover, it is now largely accepted that chronic inflammation also represents a contributing factor for cancer development (2), and recently we have reported that CNF1 is able to enhance the transcription and release of proinflammatory cytokines in T24 uroepithelial cells (14). All this is consistent with in vivo reports on the contribution of the toxin to the infection and the inflammation processes of uroepithelial tissues (1, 32, 33). In this scenario, we hypothesize that CNF1, by promoting inflammation and perturbing the cell cycle without condemning cells to death, can represent a long-term risk for cell transformation and tumor development in infected uroepithelial tissue. This is in accordance with animal studies indicating urinary tract infections as a significant risk factor for the development of bladder cancer (39).
FOOTNOTES
L.F. and P.F. contributed equally to this work.
REFERENCES
1. Andreu, A., A. E. Stapleton, C. Fennell, H. A. Lockman, M. Xercavins, F. Fernandez, and W. E. Stamm. 1997. Urovirulence determinants in Escherichia coli strains causing prostatitis. J. Infect. Dis. 176:464-469.
2. Baldassarre, G., M. S. Nicoloso, M. Schiappacassi, E. Chimienti, and B. Belletti. 2004. Linking inflammation to cell cycle progression. Curr. Pharm. Des. 10:1653-1666.
3. Boyer, L., S. Travaglione, L. Falzano, N. C. Gauthier, M. R. Popoff, E. Lemichez, C. Fiorentini, and A. Fabbri. 2004. Rac GTPase instructs nuclear factor-B activation by conveying the SCF complex and IkB to the ruffling membranes. Mol. Biol. Cell 15:1124-1133.
4. Caprioli, A., V. Falbo, L. G. Roda, F. M. Ruggeri, and C. Zona. 1983. Partial purification and characterization of an Escherichia coli toxic factor that induces morphological cell alterations. Infect. Immun. 39:1300-1306.
5. Clavers, H. 2004. At the crossroads of inflammation and cancer. Cell 118:671-674.
6. Coleman, M. L., C. J. Marshall, and M. F. Olsen. 2004. Ras and Rho GTPases in G1-phase cell-cycle regulation. Nat. Rev. Mol. Cell Biol. 5:355-366.
7. Cortes-Bratti, X., C. Karlsson, T. Lagergrd, M. Thelestam, and T. Frisan. 2001. The Haemophilus ducreyi cytolethal distending toxin induces cell cycle arrest and apoptosis via the DNA damage checkpoint pathways. J. Biol. Chem. 276:5296-5302.
8. Cover, T. L., U. S. Krishna, D. A. Israel, and R. M. Peek, Jr. 2003. Induction of gastric epithelial cell apoptosis by Helicobacter pylori vacuolating cytotoxin. Cancer Res. 63:951-957.
9. Darzynkiewicz, Z., S. Bruno, G. Del Bino, W. Gorczyca, M. A. Hotz, P. Lassota, and F. Traganos. 1992. Features of apoptotic cells measured by flow cytometry. Cytometry 13:795-808.
10. De Rycke, J., P. Mazars, J.-P. Nougayrede, C. Tasca, M. Boury, F. Herault, A. Valette, and E. Oswald. 1996. Mitotic block and delayed lethality in HeLa epithelial cells exposed to Escherichia coli BM2-1 producing cytotoxic necrotizing factor type 1. Infect. Immun. 64:1694-1705.
11. Doye, A., A. Mettouchi, G. Bossis, R. Clement, C. Boisson-Touati, G. Flatau, L. Gagnoux, M. Piechaczyk, P. Boquet, and E. Lemichez. 2002. CNF1 exploits the ubiquitin-proteasome machinery to restrict Rho GTPase activation for bacterial host cell invasion. Cell 111:553-564.
12. Etienne-Manneville, S., and A. Hall. 2002. Rho GTPases in cell biology. Nature 12:629-635.
13. Falzano, L., C. Fiorentini, G. Donelli, E. Michel, C. Kocks, P. Cossart, L. Cabanié, E. Oswald, and P. Boquet. 1993. Induction of phagocytic behaviour in human epithelial cells by E. coli cytotoxic necrotizing factor type 1. Mol. Microbiol. 9:1247-1254.
14. Falzano, L., M. G. Quaranta, S. Travaglione, P. Filippini, A. Fabbri, M. Viora, G. Donelli, and C. Fiorentini. 2003. Cytotoxic necrotizing factor 1 enhances reactive oxygen species-dependent transcription and secretion of proinflammatory cytokines in human uroepithelial cells. Infect. Immun. 71:4178-4181.
15. Fiorentini, C., G. Arancia, A. Caprioli, V. Falbo, F. M. Ruggeri, and G. Donelli. 1988. Cytoskeletal changes induced in HEp-2 cells by the cytotoxic necrotizing factor of Escherichia coli. Toxicon 26:1047-1056.
16. Fiorentini, C., P. Matarrese, E. Straface, L. Falzano, G. Donelli, P. Boquet, and W. Malorni. 1998. Rho-dependent cell spreading activated by E. coli cytotoxic necrotizing factor 1 hinders apoptosis in epithelial cells. Cell Death Differ. 5:921-929.
17. Fiorentini, C., L. Falzano, A. Fabbri, A. Stringaro, A. Logozzi, S. Travaglione, S. Contamin, G. Arancia, W. Malorni, and S. Fais. 2001. Activation of Rho GTPases by cytotoxic necrotizing factor 1 induces macropinocytosis and scavenging activity in epithelial cells. Mol. Biol. Cell 12:2061-2073.
18. Flatau, G., E. Lemichez, M. Gauthier, P. Chardin, S. Paris, C. Fiorentini, and P. Boquet. 1997. Rho GTPase activation by bacterial toxin-induced glutamine deamidation. Nature 387:729-733.
19. George, K. M., L. Pascopella, D. M. Welty, and P. L. Small. 2000. A Mycobacterium ulcerans toxin, mycolactone, causes apoptosis in guinea pig ulcers and tissue culture cells. Infect. Immun. 68:587-593.
20. Gong, J., F. Traganos, and Z. Darzynkiewicz. 1995. Discrimination of G2 and mitotic cells by flow cytometry based on different expression of cyclin A and B1. Exp. Cell Res. 220:226-231.
21. Hagting, A., M. Jackman, K. Simpson, and J. Pines. 1999. Translocation of cyclin B1 to the nucleus at prophase requires a phosphorylation-dependent nuclear import signal. Curr. Biol. 9:680-689.
22. Hahn, A., H. Barth, M. Kress, P. R. Mertens, and M. Goppelt-Struebe. 2002. Role of Rac and Cdc42 in lysophosphatidic acid-mediated cyclo-oxygenase-2 gene expression. Biochem. J. 362:33-40.
23. Lax, A. J., and W. Thomas. 2002. How bacteria could cause cancer: one step at a time. Trends Microbiol. 10:293-299.
24. Lerm, M., J. Selzer, A. Hoffmeyer, U. R. Rapp, K. Aktories, and G. Schmidt. 1999. Deamidation of Cdc42 and Rac by Escherichia coli cytotoxic necrotizing factor 1: activation of c-Jun N-terminal kinase in HeLa cells. Infect. Immun. 67:496-503.
25. Marches, O., T. N. Ledger, M. Boury, M. Ohara, X. Tu, F. Goffaux, J. Mainil, I. Rosenshine, M. Sugai, J. De Rycke, and E. Oswald. 2003. Enteropathogenic and enterohaemorrhagic Escherichia coli deliver a novel effector called Cif, which blocks cell cycle G2/M transition. Mol. Microbiol. 50:1553-1567.
26. Martin, S. J., C. P. M. Reutelingsperger, A. J. McGahon, J. A. Rader, R. C. A. A. Van Schie, D. M. LaFace, and D. R. Green. 1995. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med. 182:1545-1556.
27. Mills, M., K. C. Meysick, and A. D. O'Brien. 2000. Cytotoxic necrotizing factor type 1 of uropathogenic Escherichia coli kills cultured human uroepithelial 5637 cells by an apoptotic mechanism. Infect. Immun. 68:5869-5880.
28. Olson, M. F., A. Ashworth, and A. Hall. 1995. An essential role for Rho, Rac and Cdc42 GTPases in cell cycle progression through G1. Science 269:1270-1272.
29. Oswald, E., J.-P. Nougayrède, F. Taieb, and M. Sugai. 2005. Bacterial toxins that modulate host cell-cycle progression. Curr. Opin. Microbiol. 8:83-91.
30. Peek, R. M., Jr., M. J. Blaser, D. J. Mays, M. H. Forsyth, T. L. Cover, S. Y. Song, U. Krishna, and J. A. Pietenpol. 1999. Helicobacter pylori strain-specific genotypes and modulation of the gastric epithelial cell cycle. Cancer Res. 59:6124-6131.
31. Ren, X. D., W. B. Kiosses, and M. A. Schwartz. 1999. Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J. 18:578-585.
32. Rippere-Lampe, K. E., A. D. O'Brien, R. Conran, and H. A. Lockman. 2001. Mutation of the gene encoding cytotoxic necrotizing factor type 1 (cnf1) attenuates the virulence of uropathogenic Escherichia coli. Infect. Immun. 69:3954-3964.
33. Rippere-Lampe, K. E., M. Lang, H. Ceri, M. Olson, H. A. Lockman, and A. D. O'Brien. 2001. Cytotoxic necrotizing factor type 1-positive Escherichia coli causes increased inflammation and tissue damage to the prostate in a rat prostatitis model. Infect. Immun. 69:6515-6519.
34. Sander, E. E., S. vanDelft, J. P. ten Clooster, T. Reid, R. A. van der Cammen, F. Michiels, and J. G. Collard. 1998. Matrix-dependent Tiam1/Rac signaling in epithelial cells promotes either cell-cell adhesion or cell migration and is regulated by phosphatidylinositol 3-kinase. J. Cell Biol. 143:1385-1398.
35. Sander, E. E., and J. G. Collard. 1999. Rho-like GTPases: their role in epithelial cell-cell adhesion and invasion. Eur. J. Cancer 35:1905-1911.
36. Schmidt, G., P. Sehr, M. Wilm, J. Selzer, M. Mann, and K. Aktories. 1997. Gln 63 of Rho is deamidated by Escherichia coli cytotoxic necrotizing factor-1. Nature 387:725-729.
37. Schmidt, G., J. Selzer, M. Lerm, and K. Aktories. 1998. The Rho-deamidating cytotoxic necrotizing factor 1 from Escherichia coli possesses transglutaminase activity: cysteine 866 and histidine 881 are essential for enzyme activity. J. Biol. Chem. 273:13669-13674.
38. Smits, V. A. J., and R. H. Medema. 2001. Checking out the G2/M transition. Biochim. Biophys. Acta 1519:1-12.
39. Tamatani, T., P. Turk, S. Weitzman, and R. Oyasu. 1999. Tumorigenic conversion of a rat urothelial cell line by human polymorphonuclear leukocytes activated by lipopolysaccharide. Jpn. J. Cancer Res. 90:829-836.
40. Thelestam, M., and T. Frisan. 2004. Cytolethal distending toxins. Rev. Physiol. Biochem. Pharmacol. 152:111-133.
41. Travaglione, S., G. Messina, A. Fabbri, L. Falzano, A. M. Giammarioli, M. Grossi, S. Rufini, and C. Fiorentini. 2005. Cytotoxic necrotizing factor 1 hinders skeletal muscle differentiation in vitro by perturbing the activation/deactivation balance of Rho GTPases. Cell Death Differ. 12:78-86.
42. Welsh, C. F. 2004. Rho GTPases as key transducers of proliferative signals in G1 cell cycle regulation. Breast Cancer Res. Treat. 84:33-42.
43. Whitehouse, C. A., P. B. Balbo, E. C. Pesci, D. L. Cottle, P. M. Mirabito, and C. L. Pickett. 1998. Campylobacter jejuni cytolethal distending toxin causes a G2-phase cell cycle block. Infect. Immun. 66:1934-1940.
44. Wilson, B. A., X. Zhu, M. Ho, and L. Lu. 1997. Pasteurella multocida toxin activates the inositol triphosphate signaling pathway in Xenopus oocytes via Gq-coupled phospholipase C-1. J. Biol. Chem. 272:1268-1275(Loredana Falzano Perla Fi)
Department of Technology and Health, Istituto Superiore di Sanità, Viale Regina Elena 299, 00161, Rome, Italy
ABSTRACT
Evidence is accumulating that a growing number of bacterial toxins act by modulating the eukaryotic cell cycle machinery. In this context, we provide evidence that a protein toxin named cytotoxic necrotizing factor 1 (CNF1) from uropathogenic Escherichia coli is able to block cell cycle G2/M transition in the uroepithelial cell line T24. CNF1 permanently activates the small GTP-binding proteins of the Rho family that, beside controlling the actin cytoskeleton organization, also play a pivotal role in a large number of other cellular processes, including cell cycle regulation. The results reported here show that CNF1 is able to induce the accumulation of cells in the G2/M phase by sequestering cyclin B1 in the cytoplasm and down-regulating its expression. The possible role played by the Rho GTPases in the toxin-induced cell cycle deregulation has been investigated and discussed. The activity of CNF1 on cell cycle progression can offer a novel view of E. coli pathogenicity.
INTRODUCTION
It is largely recognized that bacterial pathogens use a variety of strategies, including the production of protein toxins, to manipulate host cell functions. In the past decade, evidence has accumulated on the ability of certain bacterial toxins to interfere with the eukaryotic cell cycle machinery, either promoting cell proliferation (30, 44) or inducing cell cycle arrest (7, 25). These protein toxins belong to the family of so-called "cyclomodulins," which actively deregulate the passage of cells throughout the host cell cycle (reviewed in reference 29). In this work, we show that a block in G2/M transition can be induced by cytotoxic necrotizing factor 1 (CNF1), a 110-kDa monomeric protein toxin frequently produced by pathogenic Escherichia coli strains associated with urinary tract infections. CNF1 is a chromosomally encoded toxin that activates the small GTP-binding proteins of the Rho family (Rho, Rac, and Cdc42), regulatory proteins that behave as molecular switches, cycling between an active GTP-bound state and an inactive GDP-bound state. CNF1 acts by deamidating a specific glutamine residue in the switch 2 domain that is crucial in GTP hydrolysis (glutamine 63 in Rho [18, 36, 37] or glutamine 61 in Rac and Cdc42 [24]), thus locking the G protein in the active state. This toxin activity confers new properties on epithelial cells, including the cytoskeleton-dependent ability to behave as professional phagocytes (13, 17).
Beside controlling the actin cytoskeleton organization (31), the Rho signaling pathways are also intimately involved in transducing mitogenic signals to the translational apparatus and play a pivotal role in different points of cell cycle regulation (28). Indeed, Rho, Rac, and Cdc42 GTPases have been demonstrated to participate in the control of G1/S progression, primarily through regulation of expression of cyclins and cyclin-dependent kinase inhibitors (42).
Hence, considering the Rho-dependent ability of CNF1 to interfere with cytokinesis, thus causing multinucleation (15), and to inhibit cell differentiation (41) in different mammalian cells, we asked whether this toxin could also somehow affect the cell cycle regulatory pathways. This question was addressed by using the human uroepithelial cell line T24, where CNF1 has previously been reported to enhance the transcription and release of proinflammatory cytokines (14). The results reported here allow the inclusion of CNF1 in the family of toxins that influence cell cycle progression and point out the role played by the Rho GTPases in this phenomenon.
MATERIALS AND METHODS
Cell cultures and treatments. T24 cells, a human epithelial cell line derived from a bladder carcinoma (American Type Culture Collection, Manassas, VA), were cultured in McCoy's 5A medium supplemented with 10% fetal calf serum, 1% nonessential amino acids, 100 μg/ml of streptomycin, and 100 U/ml of penicillin.
CNF1 was obtained from strain 392 ISS (kindly provided by V. Falbo, Rome, Italy) and purified as previously described (13). The plasmid coding for a nontoxic mutant of CNF1 that completely lacks the enzymatic activity (C866S) was kindly provided by E. Lemichez (Nice, France) and prepared as previously described (18). To rule out the possibility that the effects observed could be due to lipopolysaccharide contamination, a heat-inactivated (98°C for 10 min) CNF1 protein toxin preparation was used.
In all experiments, T24 cells were seeded at a concentration of 2 x 104/ml, and 24 h after seeding, cells were exposed for 6, 24, 48, 72, 96, 120, and 144 h to 10–10 M CNF1, heat-inactivated CNF1, or mutant CNF1 [CNF1(C866S)].
Rho GTPase transfections. T24 cells were seeded on 6-cm petri dishes and cotransfected with 2 μg of plasmid DNA encoding dominant-positive forms of either Rho, Rac, or Cdc42 Rho GTPases (RhoV14, RacV12, or Cdc42V12) and green fluorescent protein (GFP) (pEGFP; Clontech) by using Lipofectamine (Invitrogen) according to the manufacturer's instructions.
Fluorescence microscopy. Control, CNF1-treated, and transfected T24 cells were fixed with 3.7% paraformaldehyde in phosphate-buffered saline (PBS) (pH 7.4) for 10 min at room temperature. After a wash in the same buffer, cells were permeabilized with 0.5% Triton X-100 (Sigma) in PBS (pH 7.4) for 10 min at room temperature.
For F-actin visualization, cells were stained with fluorescein isothiocyanate-phalloidin (FITC) (Sigma; working dilution, 0.5 μg/ml) for 30 min at 37°C. For cyclin B1 detection, cells were stained with a monoclonal anti-human cyclin B1 antibody (BD Pharmingen; diluted 1:100). After a wash, an FITC-conjugated anti-mouse secondary antibody (Sigma; diluted 1:100) was added to the samples for 30 min at 37°C.
Finally, after extensive washing, coverslips were mounted with glycerol-PBS (2:1, vol/vol) and analyzed with an Olympus Bx51 Nikon Microphot fluorescence microscope.
Activated Rho GTPase pull-down assay. A pull-down assay was performed as previously described (41). Briefly, cells were lysed in the appropriate buffers, and the cleared lysates were incubated with 80 μg of either the glutathione S-transferase (GST)-PAK-CD (for Rac/Cdc42) or the GST-Rhotekin (for Rho) fusion protein (Cytoskeleton), bound to glutathione-coupled Sepharose beads (Amersham), for 40 min at 4°C. After three washes in the appropriate buffer, the bound proteins were eluted in sample buffer, subjected to sodium dodecyl sulfate-polyacrylamide gel electrophoresis (12% acrylamide), and electrically transferred to polyvinylidene difluoride membranes. Membranes were blocked with TBS-T (20 mM Tris-HCl [pH 7.4], 150 mM NaCl, 0.02% Tween 20) containing 2% skim milk (Bio-Rad) for 30 min at room temperature and were then incubated overnight at 4°C with primary antibodies diluted in the same buffer. The following antibodies were used: monoclonal anti-RhoA (Cytoskeleton; diluted 1:500), monoclonal anti-Rac1 (Transduction Laboratories; diluted 1:3,500), and monoclonal anti-Cdc42 (Santa Cruz Biotechnology; diluted 1:500). After extensive washing, immunocomplexes were detected with horseradish peroxidase-conjugated species-specific secondary antibodies (goat anti-mouse antibody, diluted 1:3,000 [Upstate]; goat anti-rabbit antibody, diluted 1:22,000 [Jackson]), followed by an enhanced chemiluminescence reaction (Pierce). Whole-cell lysates (2% of input) were analyzed in parallel.
Cell growth and viability. Control and CNF1-treated cells, detached from the culture dish with 10 mM EDTA (pH 7.4) and 0.25% trypsin in PBS (pH 7.4), were counted, and cell viability was determined by a trypan blue (80 μM; GIBCO BRL-Life Technologies) exclusion method. In each experiment, cell counts were conducted five times. Data represent means ± standard deviations (SD) from three separate experiments performed in duplicate. Growth inhibition was expressed as the percentage of the ratio between CNF1-treated and control cells.
Cytofluorimetric analysis. (i) Cell cycle analysis. T24 cells were trypsinized, washed twice in cold PBS, fixed with ice-cold 70% ethanol, and kept at 4°C for at least 2 h. Before the FACScan acquisition, cells were rehydrated in PBS and stained with a PBS solution containing 50 μg/ml propidium iodide (PI) and 10 μg/ml of RNase at room temperature for at least 20 min. A minimum of 10,000 events were acquired in linear mode with a FACScan (Becton Dickinson) and then analyzed with ModFit software. Transfected and control T24 cells were collected, washed twice in PBS, and fixed for 10 min with cold ethanol (70%, vol/vol) on ice. After this time, samples were centrifuged and resuspended in PBS containing PI at the final concentration of 30 μg/ml and 100 μg/ml of RNase. After 30 min of incubation at room temperature under dark conditions, a minimum of 10,000 GFP-positive events, gated by a dot plot of GFP versus forward scatter, were acquired and analyzed as described above.
(ii) Cyclin B1 detection. Control and treated T24 cells were trypsinized and washed twice in cold PBS. Cells were then fixed in 2% paraformaldehyde in PBS for 15 min and then permeabilized in 0.05% Triton X-100 in PBS for 10 min at 4°C. Samples were washed in cold PBS and incubated for 30 min at 4°C with a monoclonal anti-human cyclin B1 antibody (BD Pharmingen; diluted 1:100) After a wash, an FITC-conjugated anti mouse secondary antibody (Sigma; diluted 1:100) was added to the samples. Samples were washed twice, resuspended in PBS containing 5 μg/ml PI, and incubated for 1 h at 4°C. The fluorescence intensity of cyclin B1 was acquired in linear mode and plotted on the y axis versus the linear fluorescence intensity of PI plotted on the x axis. A minimum of 10,000 events were acquired by the FACScan (BD) and analyzed by CellQuest software (BD).
(iii) Apoptosis analysis with an annexin V-FITC kit. To detect translocation of phosphatidylserine from the inner face to the outer face of the plasma membrane in the first step of apoptosis, an annexin V-FITC kit (Medical & Biological Laboratories Co., Ltd.) was used (9, 26). Control and CNF1-treated cells were detached with EDTA and trypsin solutions. Cell suspensions were washed in ice-cold PBS, collected by centrifugation, and resuspended in annexin V binding buffer (10 mM HEPES-NaOH [pH 7.4], 140 mM NaCl, 2.5 mM CaCl2). Annexin V-FITC and propidium iodide were added to the binding buffer for 10 min at room temperature, and the cells were directly acquired by FACScan (BD). Emissions were collected at wavelengths of 530 nm for FITC and 650 nm for PI and acquired in log mode; a minimum of 10,000 events were acquired and analyzed with CellQuest software (BD).
DNA ladder assay. For the DNA ladder assay, 5 x106 cells were scraped, washed in PBS, and collected by centrifugation. Cell pellets were resuspended in lysis buffer (10 mM Tris-HCl [pH 7.4], 1 mM EDTA, and 0.2% Triton X-100) containing 0.1 μg/ml proteinase K and then incubated at 37°C for 90 min. DNA was cleared from the lysates by centrifugation and then extracted by addition of 0.8 M NaCl and an equal volume of chilled isopropanol. Following an overnight incubation at –20°C, samples were centrifuged, and the pelleted DNA was resuspended in sterile water and treated with 0.5 μg/ml of RNase A at 65°C for 20 min. DNA was then analyzed by gel electrophoresis on a 1% agarose gel stained with ethidium bromide (0.5 μg/ml).
RESULTS
CNF1 reorganizes the actin cytoskeleton and modulates the Rho GTPases in T24 cells. We started our investigation by performing an analysis of the actin cytoskeleton organization of T24 cells challenged with CNF1 (Fig. 1a to e). As expected, CNF1 exposure (Fig. 1 b to e) induced a profound reorganization of the actin cytoskeleton and rendered cells more flattened and spread out than the control cells (Fig. 1a), with a consequent remarkable increase in the cell size. In particular, at early times of exposure, CNF1 induced the formation of stress fibers (Fig. 1b) that, as the incubation time with the toxin was prolonged, progressively disappeared, concomitantly with the promotion of membrane ruffles (Fig. 1c to e). It is worth noting that, in T24 cells, multinucleation, which is a typical feature of CNF1 activity in other epithelial cell lines (4), is a more limited phenomenon; no more than 30% of T24-treated cells underwent multinucleation after 72 h of toxin exposure. However, even if they remained mononucleated, most cells displayed nuclei larger than those observed in control cells (data not shown). Hence, since cytoskeleton-dependent phenomena are chiefly controlled by the Rho GTPases (12) and it is well known that CNF1 is able to activate/deactivate the Rho GTPases with different timing depending on the cell line used (11, 41), we verified by pull-down experiments the activation state of Rho GTPases in T24 cells following CNF1 exposure (Fig. 1f to h). CNF1 was able to activate Rho, Rac, and Cdc42 GTPases in a time-dependent manner. In particular, Rho activation, already evident after 6 h of CNF1 challenge, reached a maximum after 24 h of toxin exposure (Fig. 1f). The active form of Rac (Fig. 1g), faintly detectable following 6 h of CNF1 treatment, strongly increased at 24 h of toxin treatment and reached a maximum at 48 h. Similarly to Rac GTPase, Cdc42 became strongly activated following 24 h of CNF1 exposure and remained active until 48 h (Fig. 1 h). In keeping with results obtained for other cell types (11, 41), activation of Rho GTPases was transient. In fact, the active form of Rho strongly decreased starting from 48 h of toxin treatment, while both activated Rac and Cdc42 diminished at 72 h, although this was much more evident for Cdc42. As expected, by analyzing the activation state of Rho GTPases up to 144 h of incubation with CNF1, we observed that, starting from 96 h, the active forms of Rho, Rac, and Cdc42 were nearly comparable to the control levels (data not shown).
FIG. 1. CNF1 induces actin cytoskeleton reorganization and activation of Rho family proteins in T24 cells. (a to e) Fluorescence micrographs showing F-actin organization in T24 control cells (a) and in cells challenged with CNF1 for different times (b to e). CNF1 provokes a remarkable reorganization of F-actin into stress fibers and membrane ruffles. (f to h) Immunoblots showing the activation kinetics of RhoA (f), Rac1 (g), and Cdc42 (h) GTPases in control and CNF1-treated cells, detected by a pull-down assay. Activated forms of RhoA (Rho-GTP), Rac1 (Rac-GTP), and Cdc42 (Cdc42-GTP), as well as the total amounts of GTPases, are shown. CNF1 induces a time-dependent activation of Rho GTPases followed by their deactivation. Equal protein loading was confirmed by -tubulin detection. Bar, 10 μm.
Changes in actin cytoskeleton organization induced by CNF1 reflected the activation states of the different members of the Rho GTPase family. Indeed, while Rho-induced stress fibers were assembled at 6 h of CNF1 challenge and completely disappeared at 48 h, ruffles and filopodia, which are the consequences of Rac and Cdc42 activation, respectively, appeared only following 24 h of toxin treatment and became particularly evident starting from 48 h.
CNF1 influences cell cycle progression in T24 cells. It is well known that, besides controlling the actin cytoskeleton organization, Rho signaling pathways are also intimately involved in cell proliferation and cell cycle regulation (28). Hence, we wondered whether activation of CNF1-dependent Rho GTPases in T24 cells could somehow influence cell growth and the passage of cells through the cell cycle. For this purpose, control and CNF1-treated cells were detached, and cell viability was determined by a trypan blue exclusion test. We found that in T24 cells, CNF1 caused a time-dependent inhibition of cell growth that reached a maximum after 72 h of toxin exposure (Fig. 2a), apparently without induction of death, since dead cells represented 3 to 4% of the total population both in control and in treated cells (data not shown). To determine whether the decreased growth could be linked to an alteration of the cell cycle, we performed a flow cytometric analysis of DNA content. The results obtained are reported in Fig. 2b as a percentage of the ratio between treated and control cells, while two representative histograms of untreated and treated cells are shown in Fig. 2c and d, respectively. The distribution of cell cycle phases was significantly modified by the contact with CNF1 (Fig. 2b to d). In particular, the toxin was able to induce a progressive accumulation of cells in G2/M phase, and this effect, starting as early as after 24 h of CNF1 exposure, became strongly evident after 144 h of toxin treatment (Fig. 2b). In order to verify whether the perturbation of the cell cycle could be a more general response to CNF1, we evaluated this effect on a different uroepithelial cell line, the bladder cell line 5637, already used to investigate the activity of this toxin (27). Our results showed that in this cell type also, CNF1 caused a time- and dose-dependent accumulation of cells in G2/M phase (Table 1). It is worth noting that, when T24 cells were exposed to heat-inactivated CNF1 or to the CNF1(C866S) mutant, no accumulation of cells in G2/M phase was observed (data not shown). We thus wondered whether the activation of Rho GTPases by CNF1 could be involved in shifting cells to the G2/M phases. To answer this question, we cotransfected T24 cells with GFP and with plasmids encoding constitutively activated forms of the Rho GTPases, RhoV14, RacV12, and Cdc42V12, for 24 h, and then we analyzed the cell cycle phase distribution of transfected cells. We found that the expression of each single activated Rho GTPase caused a strong accumulation of cells in both the S and G2/M phases (Table 2), in accordance with previous reports showing the influence of Rho GTPases on the cell cycle (6, 29). Hence, these results support the role played by the Rho GTPases in the CNF1-induced block of cell cycle G2/M transition in T24 cells.
FIG. 2. CNF1 induces recruitment of T24 cells in the G2/M cell cycle phase. (a) Graph showing the effect of CNF1 on cell growth. Results are reported as percentages of growth inhibition, determined as the percentage of the ratio between CNF1-treated and control cells, and represent the means ± SD from three separate experiments performed in duplicate. The toxin induces a time-dependent inhibition of cell growth. (b) Graph showing the flow cytometric analysis of cell cycle phase distribution, expressed as the ratio between CNF1-treated and control T24 cells. Note that after 144 h of CNF1 exposure, a high percentage of treated cells are arrested in the G2 stage of the cell cycle. (c and d) Representative histograms showing the DNA content of control T24 cells (c) or cells challenged with CNF1 for 144 h (d).
CNF1 down-regulates and relocalizes cyclin B1 in T24 cells. It is largely acknowledged that progression from G2 to M phase is driven by activation of cyclin-dependent kinase 1 (Cdk1)/cyclin B1 complex and that cyclin B1 expression fluctuates throughout the cell cycle, peaking in G2/M (21, 38). Hence, we analyzed the effects of CNF1 on cyclin B1 expression by combined flow cytometric analysis of cyclin B1 and DNA (20). As reported in Fig. 3a and b, CNF1 induced a significant reduction of the cyclin B1 level with respect to control cells, whatever the time of exposure considered. The mitotic function of Cdc2/cyclin B1, however, is also triggered by spatial control (21). In fact, cyclin B1 is predominantly cytoplasmic throughout G2 phase, until its very rapid translocation into the nucleus, which precedes the nuclear envelope breakdown. Immunofluorescence investigations indicated that CNF1 was able to induce a cellular redistribution of this protein, since cyclin B1 was localized in the nuclei of control cells (Fig. 3c) and most of the signal was cytoplasmic in cells exposed to the toxin (Fig. 3d). Notably, cell cycle arrest and down-regulation of cyclin B1 were not detected in T24 cells treated with heat-inactivated CNF1 or with the CNF1(C866S) mutant (data not shown). In order to investigate the link between Rho GTPase activation and cyclin B1, we have analyzed the localization of cyclin B1 in transfected T24 cells with plasmids encoding constitutively activated forms of Rho GTPases, RhoV14, RacV12, and Cdc42V12. As shown in Fig. 3, in cells transfected with RhoV14, cyclin B1 was predominantly sequestered in the cytoplasm (Fig. 3h). The same results were obtained when T24 cells were transfected with RacV12 and Cdc42 V12; cyclin B1 remained predominantly in the cytosol (data not shown).
FIG. 3. CNF1 induces down-regulation and cytoplasmic sequestration of cyclin B1. (a and b) Graphs showing the fluorescence intensity of cyclin B1 in control and CNF1-treated T24 cells. The results were obtained from four different experiments and are expressed as arbitrary units. Cyclin B1 is down-regulated in cells exposed to the toxin. Note that, in order to avoid cell confluence, two separate experiments with different T24 cell seedings were performed, one for 24 to 72 h (a) and the other for 96 to 120 h (b). (c and d) Fluorescence micrographs of T24 cells that were either untreated (c) or treated with CNF1 for 72 h (d), stained for cyclin B1. (e to h) Fluorescence micrographs of T24 cells transfected with GFP (e) and stained for cyclin B1 (f) or cotransfected with RhoV14/GFP (g) and stained for cyclin B1 (h). CNF1 exposure, as well as RhoV14 transfection, causes the sequestration of cyclin B1 in the cytoplasm. Bar, 10 μm.
The CNF1-induced G2/M arrest does not bring about apoptosis. It is known that accumulation in the G2/M phase of the cell cycle is one of the key events that precede apoptosis (38), and since the ability of CNF1 to act as a pro- or antiapoptotic factor is under debate (16, 27), we have evaluated the possibility that toxin-treated T24 cells could respond to the cell cycle accumulation in G2/M with apoptotic cell death. Control and CNF1-treated cells were thus stained with annexin V and analyzed by flow cytometry. Challenge with the toxin did not induce the exposure of annexin V binding sites, and furthermore, the percentages of cells that showed positive fluorescence for both annexin V and PI were similar for control and treated cells (Fig. 4a and b). To further confirm the inability of CNF1 to induce apoptosis in T24 cells, we analyzed the eventual DNA fragmentation in T24 cells exposed to CNF1. As shown in Fig. 4c, no DNA fragmentation was induced by CNF1 exposure even after 144 h of toxin treatment, thus demonstrating that CNF1 could not cause apoptosis in T24 cells despite the impairment in cell cycle progression. It is noteworthy that T24 cells were able to undergo DNA fragmentation when challenged with a proapoptotic factor such as staurosporine (10 mM), used as a positive control (Fig. 4c, lane D). Moreover, we used here, as a positive control, the bladder cell line 5637, which has previously been reported to undergo apoptosis when exposed to CNF1 (27). We found that 5637 cells underwent apoptosis, in terms of DNA fragmentation (data not shown), only at doses of CNF1 higher (3 x 10–10 M) than those commonly used to induce a cell response (3, 11). In T24 cells, the same higher dose of CNF1 remained unable to cause cell death.
FIG. 4. The CNF1-induced G2/M accumulation does not bring about apoptosis. (a and b) Cytofluorimetric analysis of T24 control cells (a) and of cells treated with CNF1 for 144 h (b), stained with annexin V. (c) Analysis of DNA fragmentation by DNA laddering in T24 cells. Staurosporine (10 μM) was used as a positive control for DNA fragmentation. No signs of apoptosis were observed following CNF1 exposure, whatever the method of detection used.
DISCUSSION
In this work, we provide evidence that the Rho-activating protein toxin CNF1 was able to induce, in T24 uroepithelial cells, an accumulation of cells in the G2/M phase of the cell cycle, as well as to confine the cyclin B1 in the cytoplasm and significantly decrease its expression. Our findings are in accordance with the work of De Rycke and coworkers, who reported that HeLa cells infected with E. coli strains producing CNF1, but not with those defective in CNF1 production, were totally blocked at stage G2 of the cell cycle (10).
Nowadays, the abilities of several bacterial toxins to interfere with the eukaryotic cell cycle through specific effectors and mechanisms are rapidly emerging (reviewed in reference 29). Indeed, it is well known that toxins belonging to the cytolethal distending toxin (CDT) family, produced by a broad array of pathogenic bacteria, cause irreversible cell cycle arrest or death in the target cells by using different strategies (reviewed in reference 40). For example, the Haemophilus ducreyi CDT causes cell cycle arrest and apoptosis via the DNA damage checkpoint pathways, with an effect that is cell specific (7), whereas the Campylobacter jejuni CDT blocks the host cell cycle at the G2/M transition through a complex mechanism that leads to the inactivation of the Cdc25 phosphatase and, as a consequence, to the maintenance of the cyclin B1-Cdk1 complex in the phosphorylated (inactive) form (43). In a similar fashion, the effector protein Cif, recently identified in enteropathogenic and enterohemorrhagic Escherichia coli strains, acts by inducing a sustained phosphorylation of Cdk1 and cell cycle arrest in G2 (25). On the other hand, the Helicobacter pylori homomultimeric exotoxin VacA and the polyketide mycolactone, secreted by Mycobacterium ulcerans, trigger G1 cell cycle block by mechanisms that remain to be deciphered (8, 19), while the Pasteurella multocida toxin and H. pylori CagA act on the mitogen-activated protein kinase cascade, leading to up-regulation of cyclin D1 and moving the cell into the S phase (30, 44).
Concerning the mechanism by which CNF1 causes the accumulation of cells in the G2/M phase, it appeared evident that the enzymatic activity on the Rho GTPases was crucial in the perturbation of the cell cycle, a result not reachable with the mutant CNF1 that bears an alteration of the active site (37). In fact, although principally known for their role in signal transduction processes leading to cytoskeleton-dependent responses, the Rho GTPases are also involved in a huge number of other cellular functions, including cell cycle regulation. It has been demonstrated, for example, that Rho, Rac, and Cdc42 promote G1/S transition by influencing the expression levels of cyclin D1 and those of the cyclin-dependent kinase inhibitors p27 and p21 (reviewed in reference 6). In accordance, in the present work, we showed that transfection of T24 cells with plasmids encoding the single constitutively active forms of Rho, Rac, and Cdc42 G proteins caused a strong accumulation of cells in the S and G2/M phases. On the other hand, the whole activation of both Rho and Rac/Cdc42 GTPases by CNF1 resulted in a significant increase in the percentage of cells only in the G2/M phase; we did not observe any effect on the G1/S transition. This apparent discrepancy can be explained if we consider that the simultaneous activation of the three Rho subfamilies, as in the case of CNF1, could trigger a cellular response that diverges from that induced by the single Rho proteins. In fact, the molecular pathways triggered by the different Rho family members antagonize each other (34, 35), and a dynamic balance between Rho and the Rac/Cdc42 subfamily seems to be crucial for the occurrence of different biological processes. For example, in fibroblasts, activation of Rac has been shown to down-regulate Rho activity (35), and whereas Cdc42 and Rac promote membrane protrusions, Rho promotes membrane retraction through contractile actin and myosin filaments (12). Not only do Rho- and Rac/Cdc42-induced phenotypes appear mutually exclusive, but also Rho- and Rac/Cdc42-mediated pathways regulate each other. Such a phenotypic antagonism reflects their activation states in various cell types, where the levels of active Rac/Cdc42 are inversely correlated with the levels of active Rho. This is true, for example, in epithelial cells that are growing or undergoing cadherin-dependent cell-cell contact formation (35). Finally, we have recently shown that the maintenance of a dynamic balance between Rho and Rac/Cdc42 activities is crucial for correct muscle cell differentiation (41).
This last finding appears in line with the ability of CNF1 to interfere with cell cycle progression, since the irreversible exit from the cell cycle is an essential prerequisite for the differentiation program onset. On the other hand, the ability of CNF1 to block the cell cycle G2/M transition in uroepithelial cells is consistent with a strategy that favors the induction of limited damage to the host as a means to stimulate specific cellular responses rather than rapid cell death. Hence, by inducing cell cycle arrest, CNF1 could influence the renewal and developmental processes of the epithelium, prolonging E. coli colonization through inhibition of cell shedding. Moreover, the lack of epithelial renewal would produce epithelial lesions that might transiently alter the functional capacity of the tissues or facilitate the bacterial invasion of underlying tissues. We should also consider that different bacterial protein toxins able to interfere with the host cell cycle represent a long-term risk of carcinogenesis (25) and that the possible link between bacterial infection and tumorigenesis is an emerging theme in bacterial pathogenesis (5). Helicobacter pylori is now recognized as an important etiological cofactor of gastric cancers, and several other bacteria, such as Salmonella enterica serovar Typhi, Citrobacter rodentium, and Bartonella spp., are suspected to be involved in carcinogenesis (23). In this context, it is possible to speculate that the genotoxic activity of cyclomodulins such as CDTs or Cif could act as a promoting factor for the development of cancer and similarly that CNF1 could be considered a potential promoter in cell transformation.
Although we are aware that the majority of studies on CNF1 have been conducted on cancer cells and not on primary cultures, cell lines remain invaluable means to dissect the pathways controlled by toxins. The hypothesis of CNF1 as a possible transforming factor, in fact, has been suggested by evidence obtained with cell lines that disclose the ability of CNF1 to (i) induce the transactivation of nuclear factor B (NF-B) (3), a transcription factor whose activity has been related to cell transformation; (ii) stimulate the expression of Cox-2, a typical marker of cancer (22); and (iii) protect epithelial cells from apoptotic stimuli (16). Concerning apoptosis, it is worth noting that, in contrast to the effect of H. ducreyi CDT (7), the cell spreading and block in G2 due to CNF1 were not followed by cell death, and in our hands, the uroepithelial cell line 5637, previously reported to undergo apoptosis when exposed to CNF1 (27), displayed time- and dose-dependent accumulation of cells in G2/M phase but underwent DNA fragmentation only at doses higher than those commonly used to trigger a cell response (3, 11, 16). All this further supports our view of CNF1 as a prosurvival factor that might possibly contribute to the long-term risk of transformation.
Moreover, it is now largely accepted that chronic inflammation also represents a contributing factor for cancer development (2), and recently we have reported that CNF1 is able to enhance the transcription and release of proinflammatory cytokines in T24 uroepithelial cells (14). All this is consistent with in vivo reports on the contribution of the toxin to the infection and the inflammation processes of uroepithelial tissues (1, 32, 33). In this scenario, we hypothesize that CNF1, by promoting inflammation and perturbing the cell cycle without condemning cells to death, can represent a long-term risk for cell transformation and tumor development in infected uroepithelial tissue. This is in accordance with animal studies indicating urinary tract infections as a significant risk factor for the development of bladder cancer (39).
FOOTNOTES
L.F. and P.F. contributed equally to this work.
REFERENCES
1. Andreu, A., A. E. Stapleton, C. Fennell, H. A. Lockman, M. Xercavins, F. Fernandez, and W. E. Stamm. 1997. Urovirulence determinants in Escherichia coli strains causing prostatitis. J. Infect. Dis. 176:464-469.
2. Baldassarre, G., M. S. Nicoloso, M. Schiappacassi, E. Chimienti, and B. Belletti. 2004. Linking inflammation to cell cycle progression. Curr. Pharm. Des. 10:1653-1666.
3. Boyer, L., S. Travaglione, L. Falzano, N. C. Gauthier, M. R. Popoff, E. Lemichez, C. Fiorentini, and A. Fabbri. 2004. Rac GTPase instructs nuclear factor-B activation by conveying the SCF complex and IkB to the ruffling membranes. Mol. Biol. Cell 15:1124-1133.
4. Caprioli, A., V. Falbo, L. G. Roda, F. M. Ruggeri, and C. Zona. 1983. Partial purification and characterization of an Escherichia coli toxic factor that induces morphological cell alterations. Infect. Immun. 39:1300-1306.
5. Clavers, H. 2004. At the crossroads of inflammation and cancer. Cell 118:671-674.
6. Coleman, M. L., C. J. Marshall, and M. F. Olsen. 2004. Ras and Rho GTPases in G1-phase cell-cycle regulation. Nat. Rev. Mol. Cell Biol. 5:355-366.
7. Cortes-Bratti, X., C. Karlsson, T. Lagergrd, M. Thelestam, and T. Frisan. 2001. The Haemophilus ducreyi cytolethal distending toxin induces cell cycle arrest and apoptosis via the DNA damage checkpoint pathways. J. Biol. Chem. 276:5296-5302.
8. Cover, T. L., U. S. Krishna, D. A. Israel, and R. M. Peek, Jr. 2003. Induction of gastric epithelial cell apoptosis by Helicobacter pylori vacuolating cytotoxin. Cancer Res. 63:951-957.
9. Darzynkiewicz, Z., S. Bruno, G. Del Bino, W. Gorczyca, M. A. Hotz, P. Lassota, and F. Traganos. 1992. Features of apoptotic cells measured by flow cytometry. Cytometry 13:795-808.
10. De Rycke, J., P. Mazars, J.-P. Nougayrede, C. Tasca, M. Boury, F. Herault, A. Valette, and E. Oswald. 1996. Mitotic block and delayed lethality in HeLa epithelial cells exposed to Escherichia coli BM2-1 producing cytotoxic necrotizing factor type 1. Infect. Immun. 64:1694-1705.
11. Doye, A., A. Mettouchi, G. Bossis, R. Clement, C. Boisson-Touati, G. Flatau, L. Gagnoux, M. Piechaczyk, P. Boquet, and E. Lemichez. 2002. CNF1 exploits the ubiquitin-proteasome machinery to restrict Rho GTPase activation for bacterial host cell invasion. Cell 111:553-564.
12. Etienne-Manneville, S., and A. Hall. 2002. Rho GTPases in cell biology. Nature 12:629-635.
13. Falzano, L., C. Fiorentini, G. Donelli, E. Michel, C. Kocks, P. Cossart, L. Cabanié, E. Oswald, and P. Boquet. 1993. Induction of phagocytic behaviour in human epithelial cells by E. coli cytotoxic necrotizing factor type 1. Mol. Microbiol. 9:1247-1254.
14. Falzano, L., M. G. Quaranta, S. Travaglione, P. Filippini, A. Fabbri, M. Viora, G. Donelli, and C. Fiorentini. 2003. Cytotoxic necrotizing factor 1 enhances reactive oxygen species-dependent transcription and secretion of proinflammatory cytokines in human uroepithelial cells. Infect. Immun. 71:4178-4181.
15. Fiorentini, C., G. Arancia, A. Caprioli, V. Falbo, F. M. Ruggeri, and G. Donelli. 1988. Cytoskeletal changes induced in HEp-2 cells by the cytotoxic necrotizing factor of Escherichia coli. Toxicon 26:1047-1056.
16. Fiorentini, C., P. Matarrese, E. Straface, L. Falzano, G. Donelli, P. Boquet, and W. Malorni. 1998. Rho-dependent cell spreading activated by E. coli cytotoxic necrotizing factor 1 hinders apoptosis in epithelial cells. Cell Death Differ. 5:921-929.
17. Fiorentini, C., L. Falzano, A. Fabbri, A. Stringaro, A. Logozzi, S. Travaglione, S. Contamin, G. Arancia, W. Malorni, and S. Fais. 2001. Activation of Rho GTPases by cytotoxic necrotizing factor 1 induces macropinocytosis and scavenging activity in epithelial cells. Mol. Biol. Cell 12:2061-2073.
18. Flatau, G., E. Lemichez, M. Gauthier, P. Chardin, S. Paris, C. Fiorentini, and P. Boquet. 1997. Rho GTPase activation by bacterial toxin-induced glutamine deamidation. Nature 387:729-733.
19. George, K. M., L. Pascopella, D. M. Welty, and P. L. Small. 2000. A Mycobacterium ulcerans toxin, mycolactone, causes apoptosis in guinea pig ulcers and tissue culture cells. Infect. Immun. 68:587-593.
20. Gong, J., F. Traganos, and Z. Darzynkiewicz. 1995. Discrimination of G2 and mitotic cells by flow cytometry based on different expression of cyclin A and B1. Exp. Cell Res. 220:226-231.
21. Hagting, A., M. Jackman, K. Simpson, and J. Pines. 1999. Translocation of cyclin B1 to the nucleus at prophase requires a phosphorylation-dependent nuclear import signal. Curr. Biol. 9:680-689.
22. Hahn, A., H. Barth, M. Kress, P. R. Mertens, and M. Goppelt-Struebe. 2002. Role of Rac and Cdc42 in lysophosphatidic acid-mediated cyclo-oxygenase-2 gene expression. Biochem. J. 362:33-40.
23. Lax, A. J., and W. Thomas. 2002. How bacteria could cause cancer: one step at a time. Trends Microbiol. 10:293-299.
24. Lerm, M., J. Selzer, A. Hoffmeyer, U. R. Rapp, K. Aktories, and G. Schmidt. 1999. Deamidation of Cdc42 and Rac by Escherichia coli cytotoxic necrotizing factor 1: activation of c-Jun N-terminal kinase in HeLa cells. Infect. Immun. 67:496-503.
25. Marches, O., T. N. Ledger, M. Boury, M. Ohara, X. Tu, F. Goffaux, J. Mainil, I. Rosenshine, M. Sugai, J. De Rycke, and E. Oswald. 2003. Enteropathogenic and enterohaemorrhagic Escherichia coli deliver a novel effector called Cif, which blocks cell cycle G2/M transition. Mol. Microbiol. 50:1553-1567.
26. Martin, S. J., C. P. M. Reutelingsperger, A. J. McGahon, J. A. Rader, R. C. A. A. Van Schie, D. M. LaFace, and D. R. Green. 1995. Early redistribution of plasma membrane phosphatidylserine is a general feature of apoptosis regardless of the initiating stimulus: inhibition by overexpression of Bcl-2 and Abl. J. Exp. Med. 182:1545-1556.
27. Mills, M., K. C. Meysick, and A. D. O'Brien. 2000. Cytotoxic necrotizing factor type 1 of uropathogenic Escherichia coli kills cultured human uroepithelial 5637 cells by an apoptotic mechanism. Infect. Immun. 68:5869-5880.
28. Olson, M. F., A. Ashworth, and A. Hall. 1995. An essential role for Rho, Rac and Cdc42 GTPases in cell cycle progression through G1. Science 269:1270-1272.
29. Oswald, E., J.-P. Nougayrède, F. Taieb, and M. Sugai. 2005. Bacterial toxins that modulate host cell-cycle progression. Curr. Opin. Microbiol. 8:83-91.
30. Peek, R. M., Jr., M. J. Blaser, D. J. Mays, M. H. Forsyth, T. L. Cover, S. Y. Song, U. Krishna, and J. A. Pietenpol. 1999. Helicobacter pylori strain-specific genotypes and modulation of the gastric epithelial cell cycle. Cancer Res. 59:6124-6131.
31. Ren, X. D., W. B. Kiosses, and M. A. Schwartz. 1999. Regulation of the small GTP-binding protein Rho by cell adhesion and the cytoskeleton. EMBO J. 18:578-585.
32. Rippere-Lampe, K. E., A. D. O'Brien, R. Conran, and H. A. Lockman. 2001. Mutation of the gene encoding cytotoxic necrotizing factor type 1 (cnf1) attenuates the virulence of uropathogenic Escherichia coli. Infect. Immun. 69:3954-3964.
33. Rippere-Lampe, K. E., M. Lang, H. Ceri, M. Olson, H. A. Lockman, and A. D. O'Brien. 2001. Cytotoxic necrotizing factor type 1-positive Escherichia coli causes increased inflammation and tissue damage to the prostate in a rat prostatitis model. Infect. Immun. 69:6515-6519.
34. Sander, E. E., S. vanDelft, J. P. ten Clooster, T. Reid, R. A. van der Cammen, F. Michiels, and J. G. Collard. 1998. Matrix-dependent Tiam1/Rac signaling in epithelial cells promotes either cell-cell adhesion or cell migration and is regulated by phosphatidylinositol 3-kinase. J. Cell Biol. 143:1385-1398.
35. Sander, E. E., and J. G. Collard. 1999. Rho-like GTPases: their role in epithelial cell-cell adhesion and invasion. Eur. J. Cancer 35:1905-1911.
36. Schmidt, G., P. Sehr, M. Wilm, J. Selzer, M. Mann, and K. Aktories. 1997. Gln 63 of Rho is deamidated by Escherichia coli cytotoxic necrotizing factor-1. Nature 387:725-729.
37. Schmidt, G., J. Selzer, M. Lerm, and K. Aktories. 1998. The Rho-deamidating cytotoxic necrotizing factor 1 from Escherichia coli possesses transglutaminase activity: cysteine 866 and histidine 881 are essential for enzyme activity. J. Biol. Chem. 273:13669-13674.
38. Smits, V. A. J., and R. H. Medema. 2001. Checking out the G2/M transition. Biochim. Biophys. Acta 1519:1-12.
39. Tamatani, T., P. Turk, S. Weitzman, and R. Oyasu. 1999. Tumorigenic conversion of a rat urothelial cell line by human polymorphonuclear leukocytes activated by lipopolysaccharide. Jpn. J. Cancer Res. 90:829-836.
40. Thelestam, M., and T. Frisan. 2004. Cytolethal distending toxins. Rev. Physiol. Biochem. Pharmacol. 152:111-133.
41. Travaglione, S., G. Messina, A. Fabbri, L. Falzano, A. M. Giammarioli, M. Grossi, S. Rufini, and C. Fiorentini. 2005. Cytotoxic necrotizing factor 1 hinders skeletal muscle differentiation in vitro by perturbing the activation/deactivation balance of Rho GTPases. Cell Death Differ. 12:78-86.
42. Welsh, C. F. 2004. Rho GTPases as key transducers of proliferative signals in G1 cell cycle regulation. Breast Cancer Res. Treat. 84:33-42.
43. Whitehouse, C. A., P. B. Balbo, E. C. Pesci, D. L. Cottle, P. M. Mirabito, and C. L. Pickett. 1998. Campylobacter jejuni cytolethal distending toxin causes a G2-phase cell cycle block. Infect. Immun. 66:1934-1940.
44. Wilson, B. A., X. Zhu, M. Ho, and L. Lu. 1997. Pasteurella multocida toxin activates the inositol triphosphate signaling pathway in Xenopus oocytes via Gq-coupled phospholipase C-1. J. Biol. Chem. 272:1268-1275(Loredana Falzano Perla Fi)