Correspondence José Luis Baronetti, Department of Pharmacy. Faculty of Chemical Sciences, National University of Córdoba, Haya de la Torre y Medina Allende, University Campus, 500 Córdoba, Argentina. Tel: +54 0351 4334163 ext. 104/105; fax: +54 0351 4334127; email: email@example.com
Shiga toxin-producing Escherichia coli are important food-borne pathogens. The main factor conferring virulence on this bacterium is its capacity to secrete Shiga toxins (Stxs), which have been reported to induce apoptosis in several cell types. However, the mechanisms of this apoptosis have not yet been fully elucidated. In addition, Stxs have been shown to stimulate macrophages to produce nitric oxide (NO), a well-known apoptosis inductor.The aim of this study was to investigate the participation of NO in apoptosis of rat peritoneal macrophages induced by culture supernatants or Stx2 from E. coli. Peritoneal macrophages incubated in the presence of E. coli supernatants showed an increase in the amounts of apoptosis and NO production. Furthermore, inhibition of NO synthesis induced by addition of aminoguanidine (AG) was correlated with a reduction in the percentage of apoptotic cells, indicating participation of this metabolite in the apoptotic process. Similarly, treatment of cells with Stx2 induced an increase in NO production and amount of apoptosis, these changes being reversed by addition of AG. In summary, these data show that treatment with E. coli supernatants or Stx2 induces NO-mediated apoptosis of macrophages.
Shiga toxin-producing Escherichia coli infections are responsible for widespread disease, including HUS, which is characterized by thrombocytopenia, microangiopathic hemolytic anemia and renal failure (1–3). During intestinal infection, E. coli secretes different products into the intestinal lumen, including LPS and its main virulence factor, Stx (2). The latter is a holotoxin composed of an enzymatic A subunit (StxA) in association with five B subunits (StxB) (4). This toxin translocates across the intestinal epithelial cell layer into the circulation, allowing it to act on different cell populations at distant sites, such as the kidney and the brain (5). Numerous effects on different cellular populations have been attributed to Stx, including cytokine production, secretion of ROI such as the superoxide anion (O2−), liberation of RNI such as NO, and apoptosis (6–9).
Apoptosis (programmed cell death) is a phenomenon that has been demonstrated under both physiological and pathological conditions. This process is characterized by changes in cell morphology, chromatin condensation, nuclear DNA fragmentation, and apoptotic body formation (10). Induction of apoptosis occurs through two main apoptotic pathways: the extrinsic, or death receptor, pathway and the intrinsic, or mitochondrial, pathway (11). The extrinsic pathway may be triggered by ligation of surface receptors such as tumor necrosis factor receptor 1 and Fas receptor, which are present in sensitive cells. On the other hand, different types of non-receptor-mediated stimuli, for example viral infection, hypoxia and free radicals, are responsible for activation of the mitochondrial pathway (12). Nevertheless, independent of stimulus type, both activation pathways converge to trigger a common execution phase of cell death. This execution pathway results in activation of caspase 3, which ultimately causes the apoptotic process (11).
Nitric oxide, a metabolite released by different cell types, mainly macrophages, is involved in a great variety of processes. For example, in infectious diseases, liberation of NO by cells of the innate immune system contributes to early control of infection (13). In addition, induction of this molecule has been reported to have beneficial effects in the treatment of acute myocardial infarction (14), to be necessary for the maintenance of homeostasis in the kidney (15) and to have cytotoxic effects on tumor cells (16). On the other hand, production of the same molecule at high concentrations or under certain specific conditions can have detrimental effects on health. Thus, overproduction of NO has been implicated in autoimmune disorders such as rheumatoid arthritis and multiple sclerosis, chronic inflammation and neurodegeneration (17). NO has also been shown to be involved in the apoptotic process, production of this metabolite inducing apoptosis of some cell types, including macrophages, neurons and human neutrophils (18–20).
Macrophages are very important in the control of infections. These cells are capable of phagocytosing and killing microorganisms and secrete an array of cytotoxic products such as hydrogen peroxide and NO (21). Macrophages also produce cytokines and chemokines, which are involved in regulation of both innate and adoptive immune responses against various infections. Macrophages also play other key roles in the immune system, since they can act as antigen presenting cells in secondary immune responses (22). Because of the fundamental role of this cell in defense against infections, a reduction in its viability or functions can result in impaired control of infectious diseases. For example, inadequate macrophage activation results in persistence of pathogenic microorganisms and exacerbation of different diseases such as leishmaniasis, schistosomiasis and cryptococcosis (23–25). Induction of apoptosis in macrophages by bacteria such as Salmonella typhimurium and Yersinia pestis has also been demonstrated (26, 27).
In regard to E. coli infection, different products of these bacteria have been shown to induce apoptosis in a variety of cell types. However, the precise relationship between NO synthesis and induction of apoptosis has not been thoroughly investigated. In this study, we have demonstrated that treatment with E. coli supernatants, or with the major virulence factor of this pathogen, Stx2, induces NO-mediated apoptosis of rat peritoneal macrophages. These results could contribute to a better understanding of the toxicity of Stx2 in HUS.
MATERIALS AND METHODS
Culture conditions and preparation of supernatants and shiga toxin
The clinically isolated E. coli O157:H7 strain was kindly provided by the Microbiology Laboratory of the Hospital of Niños de la Santísima Trinidad of Córdoba, Córdoba province, Argentina. This strain has been characterized as a Stx2 producer. Stock cultures were preserved at −80°C using glycerol 1% (v/v) as the cryoprotectant. This E. coli strain was grown in TSB for 48 hr at 37°C in an orbital rotator (150 rpm). The bacteria were then pelleted by centrifugation at 20,000 g for 30 min, and the cell-free culture supernatants sterilized through a 0.22 μm-diameter filter. Finally, this supernatant, adjusted to 100 μg/mL of total protein, was diluted 50-, 100- or 500-fold with PBS (dilutions 1/50, 1/100 and 1/500 respectively). Shiga toxin 2 was purified from culture supernatants by receptor-mediated affinity chromatography (28). Briefly, culture supernatants were passed through a small column (1 or 2 mL) of Octyl Sepharose CL-4B-Gb3 (Sigma, St Louis, MO, USA) five times. The column was then washed with 10 column volumes of PBS to eliminate the unabsorbed proteins, and the retained Stxs2 were eluted with 4.5 M MgCl2 in PBS (10 bed volumes) and dialyzed against PBS. The purified Stx2 was divided into aliquots and stored at −80°C until use. Toxin purity was assessed by SDS-PAGE electrophoresis with silver staining. The endotoxin content of the Stx2 preparation was below the level of detection, as measured by the Limulus amebocyte lysate assay. Serial dilutions of supernatants and Stx2 were performed in PBS.
Isolation of rat peritoneal macrophages
Female inbred Wistar rats (ages, 8 to 12 weeks) were used in this study. The animals were housed and cared for according to our institutional guidelines. They were killed following protocols approved by the Animal Experimentation Ethics Committee, Faculty of Chemical Sciences, National University of Córdoba. Resident peritoneal cells were obtained by lavage of the peritoneal cavity with 45 mL of ice-cold PBS containing 0.1% FCS. The macrophages were separated from other cell populations by using a discontinuous Percoll gradient (29). Briefly, peritoneal cells were centrifuged at 1500 rpm. for 10 min, re-suspended in 2 mL of Krebs Ringer phosphate dextrose buffer (pH 7.0) and slowly added to the Percoll gradient (2 mL of a solution of Percoll with a density of 1.090 g/mL and then a further 2 mL with a density of 1.080 g/mL, carefully overlaid). The tubes were centrifuged at 1500 rpm. for 30 min, and the macrophages collected from the interface between the Percoll layers. After being washed twice in RPMI 1640, the cells were cultured in RPMI 1640 with FCS 10% for 6 hr at 37°C in 5% CO2. Adherent cells were harvested and re-suspended in RPMI supplemented with 10% FCS. The viability of the cells was consistently more than 90%, as determined by a trypan blue exclusion method. May Grünwald-Giemsa staining showed that more than 90% of the cells had a macrophage morphology.
Rat peritoneal macrophages in RPMI 1640 containing 10% FBS were exposed to different dilutions of E. coli cell-free supernatants, Stx2 or their respective controls (TSB or PBS) for 24 hr at 37°C in a 5% CO2 humidified atmosphere. After incubation, numbers of apoptotic cells and NO production were evaluated by flow cytometry analysis and a colorimetric method, respectively. In some experiments, the cultures were performed in the presence of the iNOS inhibitor AG (1 mM), in order to examine the hypothesis that apoptosis is associated with induction of NO.
Measurement of nitrite production
To determine nitrite production, 50 μL of cell-free supernatant (diluted 1/50, 1/100 or 1/500 in PBS), Stx2 (400, 200 or 50 ng/mL) or LPS (10 μg/mL) were added to 950 μL of RPMI 1640 supplemented with 10% RPMI containing peritoneal macrophages (2 × 106/mL) and incubated for 24 hr at 37°C and 5% CO2, in the absence or presence of AG (1 mM). NO production by these cells was then determined spectrophotometrically by using the Griess reaction (30). Briefly, Griess reagent was prepared by mixing equal volumes of sulfanylamide (1.5% in 1 N HCl) and N-(1-naphthyl) ethylenediamide dihydrochloride (0.13% in H2O). A volume of 200 μL of Griess reagent was then mixed with 100 μL of culture supernatants, and this mixture incubated for 15 min in the dark. The absorbance at 540 nm was measured with an automated microplate reader (Bio-Rad, Hercules, CA, USA), and the concentration of nitrite calculated from a NaNO2 standard curve. No nitrite was detectable in the cell-free medium.
Apoptosis of peritoneal macrophages
To evaluate the amount of apoptosis, 50 μL of cell-free supernatant (diluted 1/50, 1/100 or 1/500 in PBS) or Stx2 (400, 200 or 50 ng/mL) were added to 950 μL of RPMI 1640 supplemented with 10% RPMI containing peritoneal macrophages (2 × 106/mL) and incubated for 24 hr at 37°C and 5% CO2, in the absence or presence of AG (1 mM). Following treatments, the cells were permeabilized with ice-cold 70% ethanol and stained with 750 μL PI (50 μg/mL in 0.1% Triton X-100/0.1% sodium citrate) for 30 min, and subjected to apoptosis analysis by flow cytometry (FACS) (31). Based on the PI staining, cells in the sub-G1 marker window (hypodiploid DNA) were considered to be apoptotic.
Data were expressed as means ± SEMs. The data were analyzed by a one-way analysis of variance with Tukey's post-hoc test to determine the statistical significance for all pairwise multiple comparison procedures. A P value of <0.05 was considered significant. All experiments were repeated, and equivalent results were obtained in each experiment.
Apoptosis of rat peritoneal macrophages by culture supernatants
To determine the effect of cell-free culture supernatants from our clinically isolated E. coli on rat peritoneal macrophage apoptosis, the cells were isolated from the cultures and the DNA labeled with PI and analyzed by flow cytometry (Fig. 1). Peritoneal macrophages were treated with different dilutions of culture supernatants, and the percentage of apoptotic cells at 24 hr post-treatment was determined. Figure 1 shows that treatment with dilutions of 1/50, 1/100 or 1/500 of the culture supernatants induced a significant increase in the percentage of apoptotic macrophages in comparison with those treated with TSB (control group).
Nitric oxide production by rat peritoneal macrophages
Nitric oxide production has been demonstrated to be involved in the induction of apoptosis in a variety of cell types. Therefore, NO production by rat peritoneal macrophages exposed to culture supernatants from E. coli was evaluated. Amounts of nitrite found in peritoneal macrophages treated with dilutions of 1/50, 1/100 or 1/500 of the culture supernatants were greater than those produced by macrophages treated with TSB (Fig. 2). LPS was used as positive control.
Contribution of nitric oxide production to macrophage apoptosis
Our previous results showed that treatment of peritoneal macrophages with different dilutions of culture supernatants results in induction of apoptosis and synthesis of NO in cell cultures. Therefore, we speculated that the apoptosis observed in the rat peritoneal macrophages treated with culture supernatants from E. coli could have been associated with the induction of NO. For this experiment, only the dilutions of 1/50 and 1/100 of supernatants were evaluated, since these had been shown to induce the highest percentages of apoptosis in peritoneal macrophages. The inhibition of NO synthesis induced by the addition of AG to the cultures (Fig. 3a) correlated with a reduction in the percentage of apoptotic cells in these cultures (Fig. 3b), suggesting participation of this metabolite in the apoptotic process. The concentration of AG used in these experiments significantly reduced NO production in cultures.
Nitric oxide-mediated apoptosis by Shiga toxin 2
In order to identify which of the products present in the culture supernatants was involved in these apoptotic phenomena, rat peritoneal macrophages were cultivated in the presence of different concentrations of Stx2, and NO production and the percentage of apoptotic cells at 24 hr post-treatment were determined. Similarly to our observations with culture supernatants, treatment of cells with 400 and 200 ng/mL of Stx2, but not with 50 ng/mL, induced an increase in NO production (Fig. 4a) and amount of apoptosis (Fig. 4b, top panels), in comparison to the cell cultures exposed to PBS. Moreover, the apoptosis of rat peritoneal macrophages induced by Stx2 (400 and 200 ng/mL) was reversed by the addition of AG to the cultures (Fig. 4b, bottom panels). The concentration of AG used in these experiments significantly reduced NO production in the cultures (Fig. 4a).
Previous findings by other authors have demonstrated that different products of E. coli are able to induce apoptosis in a variety of cell types, heat-labile enterotoxin, hemolysin and cytotoxic necrotizing factor from E. coli having been described as apoptosis inductors (32, 33). However, the most studied form has been E. coli's main virulence factor, Stx. This toxin induces apoptosis of cells such as macrophages, neutrophils and several other cell lines. Although different activation ways have been shown to be involved in this process, the precise mechanism remains unclear. Shiga toxin has also been implicated in the activation of cells to produce different products such as cytokines, O2− and NO (6–8). Nevertheless, in spite of the production of NO, a well-known apoptosis inductor, the association between synthesis of this metabolite and induction of apoptosis by Stx2 has not been studied extensively. Therefore, in this study, we investigated the participation of NO in the apoptosis of peritoneal macrophages induced by culture supernatants and Stx2 from E. coli.
Numerous studies have demonstrated the participation of different caspases and kinases on the apoptotic phenomena induced by Stx. Related to this, Smith and co-workers have demonstrated that treatment of intestinal epithelial cells with Stx results in the activation of stress-activated protein kinases, JNKs, and p38 MAP kinases. This activation may be related to the cleavage of caspase-3 and finally the death by apoptosis of these cells (34). On the other hand, Stx also induces apoptosis in HeLe cells. However, in this case cell death is caused by a different mechanism than that observed with intestinal epithelial cells (35). Exposure of HeLa cells to Stx results in activation of caspase-6, -8 and -3, indicating a death receptor pathway. Moreover, this treatment also induces activation of components of the mitochondrial pathway, such as caspase-9, permeabilization of mitochondria and the release of cytochrome c. Nevertheless, alteration of these variables is a secondary effect, and not a mechanism of apoptosis induction. Another type of cell that undergoes apoptosis during treatment with Stx is human brain microvascular endothelial cells. In this case, apoptosis is mediated by CHOP, and the intrinsic and extrinsic pathways are involved (36). Therefore, the mechanism of apoptosis induced by Stx is dependent on the type of cell involved. In macrophage-like THP-1 cells, treatment with Stx results in production of proteins and proinflammatory cytokines, associated with delayed apoptosis (24 hr). Furthermore, Stx triggers the activation of a great variety of mediators involved in the apoptotic phenomenon, such as caspases and cytochrome c (37–39).
It is well known that NO can induce apoptosis in many cell types. In relation to this, in articular chondrocytes, the apoptosis induced by NO is mediated by activation of p38 kinases, Bax, caspase-3 and the accumulation of p53 (40). Furthermore, treatment of microglial cells with LPS plus interferon-γ results in induction of NO and apoptosis (41). In this process, apoptosis occurs through CHOP and the endoplasmic reticulum stress pathway, but is independent of p53. NO has also been found to induce apoptosis in macrophages, h cytochrome c, CHOP and the ER stress pathway all playing important roles in this NO-mediated apoptosis (42). In addition, studies by Jun et al. have demonstrated that treatment of murine RAW 264.7 macrophages with sodium nitroprusside, a NO-generating agent, results in activation of JNKs and apoptosis of these cells (43). On the other hand, production of NO has also been associated with anti-apoptotic effects. Thus, NO can act as a survival factor and inhibit apoptosis (44). Moreover, NO can react with O2− to form peroxynitrite, which has been shown to induce apoptosis in different types of cells (45–47). In addition, although the mechanism by which this metabolite triggers apoptosis has not been fully elucidated, induction of ROI and activation of caspases have been found to be involved in this phenomenon (48–50).
In the present study, we found that rat peritoneal macrophages incubated in the presence of E. coli supernatants showed increases in the amount of apoptosis and NO production. Furthermore, inhibition of NO synthesis induced by the addition of AG at cultures was correlated with a reduction in the percentage of apoptotic cells in these cultures, indicating the participation of this metabolite in the apoptotic process. In order to identify which of the products present in the culture supernatants is involved in these phenomena, rat peritoneal macrophages were cultivated in the presence of different concentrations of Stx2. In a similar way to that observed with the culture supernatants, the treatment of cells with Stx2 induced an increase in NO production and amount of apoptosis. Moreover, the changes in these variables were reversed by addition of AG to the cultures. Therefore, we have demonstrated that treatment with E. coli supernatants, or with the major virulence factors of this pathogen, Stx2, induces NO-mediated apoptosis of rat peritoneal macrophages. On the other hand, the possibility that peroxynitrite or ROI, and activation of caspases and cytochrome c, also participate in this apoptosis cannot be discounted.
Apoptosis of macrophages, vital cells of the innate immune system which are involved in defense against different microorganisms, has been implicated in the pathogenesis of a variety of infectious diseases, including the HUS caused by Shiga toxin-producing E. coli. Therefore, these results could contribute to a better understanding of the immunopathology of E. coli infection.
This work was supported by grants from the National Bureau of Scientific and Technical Research (CONICET), Secretariat of Science and Technology of the National University of Córdoba (SECyT-UNC) and FONCYT PICTO-UNC 36163. We would like to thank native speaker, Dr. Paul Hobson, for revision of the manuscript. We are also very grateful to Dr. Ricardo D. Lardone for his technical support, Dr María E. Suarez for providing the E. coli strain and Drs Padola NL and Parma AE, Laboratory of Immunochemistry and Biotechnology, Faculty of Veterinary Sciences, University Nacional del Centro, for strain characterization.