SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Acinetobacter baumannii is an important opportunistic pathogen responsible for nosocomial infection. Despite considerable clinical and epidemiological data regarding the role of A. baumannii in nosocomial infection, the specific virulence factor or pathogenic mechanism of this organism has yet to be elucidated. This study investigated the molecular mechanism of apoptosis on the infection of human laryngeal epithelial HEp-2 cells with A. baumannii and examined the contribution of outer membrane protein 38 (Omp38) on the ability of A. baumannii to induce apoptosis of epithelial cells. A. baumannii induced apoptosis of HEp-2 cells through cell surface death receptors and mitochondrial disintegration. The Omp38-deficient mutant was not as able to induce apoptosis as the wild-type A. baumannii strain. Purified Omp38 entered the cells and was localized to the mitochondria, which led to a release of proapoptotic molecules such as cytochrome c and apoptosis-inducing factor (AIF). The activation of caspase-3, which is activated by caspase-9, degraded DNA approximately 180 bp in size, which resulted in the appearance of a characteristic DNA ladder. AIF degraded chromosomal DNA approximately 50 kb in size, which resulted in large-scale DNA fragmentation. These results demonstrate that Omp38 may act as a potential virulence factor to induce apoptosis of epithelial cells in the early stage of A. baumannii infection.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Apoptosis is an active process of gene-directed cellular self-destruction. It plays a central role not only during development and homeostasis, but also in the regulation of the host response during infection by microorganisms (Williams, 1994; Steller, 1995; Nagata, 1997; Zychlinski and Sansonetti, 1997). Apoptotic cell death is characterized by cellular shrinkage, blebbing of the cytoplasmic membrane, chromatin condensation, nuclear fragmentation and DNA cleavage (Kerr et al., 1972; Wyllie et al., 1984). Various apoptotic signals activate a family of cysteine proteases called caspases, which are initially produced as inactive zymogens (Salvesen and Dixit, 1997; Thornsberry and Lazebnik, 1998; Budihardjo et al., 1999). The complex of death receptors and bacterial ligands on the cell surface activates caspase-8, which is followed by activation of effector caspases such as caspase-3, -6 or -7, thereby inducing apoptosis. Some apoptotic signals also trigger changes in mitochondria, which lead to a translocation of cytochrome c from mitochondria into cytosol. The cytosolic cytochrome c binds to apoptosis-activating factor-1 (Apaf-1) and activates caspase-9, which in turn activates effector caspases (Li et al., 1997; Zou et al., 1999). Apoptosis-inducing factor (AIF) represents a mitochondrial flavoprotein that is able to induce apoptosis independently from caspases. This process has been implicated in caspase-independent mitochondrial pathways for the induction of apoptosis (Susin et al., 1999). Cytosolic AIF enters the nucleus, which then results in the breakdown of DNA. The induction of apoptosis through caspase-dependent pathway in host cells has been described for a variety of bacterial pathogens, but apoptotic cell death through the AIF-dependent pathway has been described for few bacterial pathogens (Braun et al., 2001; Jendrossek et al., 2003).

Bacteria of the genus Acinetobacter are ubiquitous microorganisms, which can be found in a variety of ecological niches including water and soil, and in clinical specimens of human and animal origins (Bergogne-Bérézin and Towner, 1996). In recent decades, Acinetobacter have emerged as important nosocomial pathogens, which can give rise to various infections including bacteraemia, meningitis, pneumonia, skin and wound infections, and urinary tract infections, mostly in severely ill patients. Of the 32 named and unnamed species currently known, Acinetobacter baumannii is the species with the highest prevalence in clinical specimens (Seifert et al., 1993; Vaneechoutte et al., 1995; Dijkshoorn et al., 1996; Nemec et al., 1999). Despite the fact that numerous researchers have reported the implication of A. baumannii in infections and outbreaks, the factors determining pathogenicity of this organism have yet to be elucidated. We previously demonstrated that live A. baumannii induced the apoptosis of epithelial cells in vitro (Lee et al., 2001). However, little is known of the bacterial factors or molecular mechanisms of apoptosis. As the apoptosis of host cells is highly regulated, the elucidation of the mechanisms associated with the apoptosis of epithelial cells may be the key to understanding the pathogenicity of A. baumannii in early infections.

Outer membrane proteins (Omps) of Gram-negative bacteria are known to be key players in bacterial adaptation and pathogenesis in host cells (Lin et al., 2002). Porins play a variety of roles depending on the bacterial species, including the maintenance of cellular structural integrity, bacterial conjugation and bacteriophage binding, antimicrobial resistance and pore formation to permit the penetration of small molecules (Beher et al., 1980; Vordermeier et al., 1990; Weiser and Gotschlich, 1991; Hauschildt and Kleine, 1995; Saint et al., 2000). Porins are also found in mitochondria. The major band of Omps of A. baumannii on SDS-PAGE is a 38 kDa porin (Omp38). Jyothisri et al. (1999) structurally defined this Omp, which is a trimeric porin with a pore size of 1.3 nm and which acts as a general diffusion pore. However, the role of Omp38 in the pathogenesis of A. baumannii infection has not yet been explored.

In this study, we investigated the molecular mechanisms of apoptosis on the infection of human laryngeal epithelial HEp-2 cells with live A. baumannii and on the treatment of the cells with purified Omp38 from A. baumannii. Here we report that Omp38 localizes to the mitochondria and induces a release of cytochrome c and AIF into cytosol, which mediates caspase-dependent and AIF-dependent apoptosis in epithelial cells.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Acinetobacter baumannii induces apoptosis of HEp-2 cells

As the respiratory tract is the most common site associated with A. baumannii infection, the cytotoxic effects of A. baumannii were investigated in human laryngeal HEp-2 cells. A. baumannii infection induced morphological changes of HEp-2 cells. Cell death such as cellular shrinkage, membrane blebbing, round-up and detachment from the culture plate, and nuclear condensation and fragmentation was clearly identified by their distinct morphology after 12 h of infection (Fig. 1A). An increase of hypo-diploid cell population was observed (Fig. 1B). DNA laddering is one of the most distinctive features of caspase-dependent apoptosis. The chromosomal DNA of HEp-2 cells infected with A. baumannii was extensively fragmented into oligonucleosomes, as demonstrated by the characteristic DNA ladder (Fig. 1C). These findings indicate that A. baumannii induces apoptosis of HEp-2 cells in vitro.

image

Figure 1. Acinetobacter baumannii directly induces apoptosis of HEp-2 cells. A. Morphological changes of HEp-2 cells. Cells were infected with live A. baumannii at an moi of 100 for the indicated time or left uninfected. Cellular morphology was observed by phase contrast microscope (top) and nuclei were stained with DAPI (bottom). B. Flow cytometric analysis of the DNA content of cells after staining with PI. The 104 cells per sample were analysed. The percentage of cells in the hypo-diploid population is indicated in the diagrams. C. Analysis of internucleosomal DNA fragmentation. Extracted DNA was electrophoresed on 1.8% agarose gel and visualized by ethidium bromide staining. Lane M, 100 bp DNA ladder.

Download figure to PowerPoint

Caspases are activated during A. baumannii-induced apoptosis

The induction of apoptosis is mediated through an ordered series of events, such as the activation of caspases followed by its substrate degradation (Budihardjo et al., 1999). We identified caspase-activating cascades in the HEp-2 cells infected with A. baumannii. The pro-form of caspase-3 and poly[ADP-ribose] polymerase (PARP) was cleaved and the amount of the active form of both molecules increased with the duration of the infection (Fig. 2A). The cleavage of effector caspase-3 was mediated by the activation of upstream initiator such as caspase-8 and caspase-9. Immunoblot analysis revealed that the pro-form of caspase-8 and caspase-9 clearly decreased and that the active form of both caspases increased after infection (Fig. 2B). These findings indicate that A. baumannii induces the apoptosis of HEp-2 cells through caspase-dependent cascades, which are mediated by cell surface signalling and mitochondrial disintegration.

image

Figure 2. Cleavage of caspases and PARP during A. baumannii-induced apoptosis. HEp-2 cells were infected with live A. baumannii at an moi of 100 for the indicated time. Cell lysates were resolved on 10% or 12% SDS-polyacrylamide gel and immunoblotted with specific antibodies. β-Actin was used as a loading control. A. Immunoblot analysis of caspase-3 and PARP. B. Immunoblot analysis of caspase-8 and -9.

Download figure to PowerPoint

Acinetobacter baumannii induces mitochondrial damage and induces the release of cytochrome c and AIF

Activation of caspase-9 suggests the release of cytochrome c from mitochondria into cytosol. We tested A. baumannii-induced mitochondrial damage using a mitochondrion-selective fluorescent dye, the uptake of which depends on intact mitochondrial membrane potential. The mitochondrial uptake of MitoTrackerTM clearly decreased after 4 h of infection (Fig. 3A). Reduction of the mitochondrial membrane potential preceded all morphological and biochemical signs of apoptosis. Loss of mitochondrial membrane potential can result in the release of proapoptotic molecules such as cytochrome c and AIF into the cytosol. Apoptosis can be initiated by mitochondrial damage without any significant caspase activation in some cell death models (Lorenzo et al., 1999; Susin et al., 1999). In the current study, immunoblot analysis showed that A. baumannii infection induced the release of cytochrome c and AIF into cytosol (Fig. 3B). To determine whether A. baumannii-induced apoptosis was inhibited by the pan-caspase inhibitor, Z-VAD-fmk was treated for 1 h before the application of bacteria and maintained during the infection. Pan-caspase inhibitor did not completely inhibit apoptosis of HEp-2 cells infected with A. baumannii, but partially inhibited apoptosis of HEp-2 cells (Fig. 3C). This finding suggests that both caspase-activating cascades and the AIF-dependent pathway mediate apoptosis of HEp-2 cells.

image

Figure 3. Acinetobacter baumannii induced mitochondrial damage and induced the release of cytochrome c and AIF. A. A. baumannii infection decreased mitochondrial membrane potential. Cells infected with A. baumannii were stained with MitotrackerTM (green). B. Immunoblot of AIF and cytochrome c. Mitochondrial lysates and cytosolic fractions were resolved on 12% SDS-polyacrylamide gel and immunoblotted with anti-AIF and anti-cytochrome c antibody. C. Partial inhibition of apoptosis by pan-caspase inhibitor. Cells were infected with A. baumannii for 16 h (b), treated with 100 µM of Z-VAD-fmk for 1 h before application of bacteria and during infection (c) or left uninfected (a). Cells were stained with DAPI.

Download figure to PowerPoint

Wild-type A. baumannii induces apoptosis more efficiently than isogenic Omp38-deficient mutant

To identify bacterial factors inducing apoptosis of HEp-2 cells, random transposon mutagenesis was performed in A. baumannii ATCC19606T. One mutant strain, KS37, showed a significant decrease of the rate and extent of cell death. The wild-type strain expressed 38 kDa of Omp, while the KS37 mutant completely abolished the expression of this Omp, as demonstrated by SDS-PAGE (Fig. 4A) and transmission electron microscopy (Fig. 4B). We designated this Omp as Omp38 according to the molecular mass in SDS-PAGE gel. All HEp-2 cells infected with the wild-type strain died within 20 h of infection, while 55% of the cells infected with the Omp38-deficient mutant were viable (Fig. 4C). An apoptotic DNA ladder was observed in the HEp-2 cells infected with a wild-type strain within 12 h of infection, while the isogenic Omp38-deficient mutant showed no signs of a DNA ladder up to 20 h of infection (Fig. 4D). These findings indicate that Omp38 is responsible for apoptosis of HEp-2 cells in A. baumannii infection.

image

Figure 4. Characteristics of Omp38-deficient mutant. A. Outer membrane protein profiles of A. baumannii ATCC19606T (lane 1) and its isogenic Omp38-deficient mutant KS37 (lane 2). Outer membrane fractions were electrophoresed on 10% SDS-polyacrylamide gel. Lane M, molecular weight size marker. Arrow indicates the position of a 38 kDa. B. Transmission electron micrographs of A. baumannii ATCC19606T (a) and isogenic Omp38-deficient mutant (b); original magnification: ×40 000. Inner and outer membranes were shown in A. baumannii ATCC19606T, while only inner membrane was shown in isogenic Omp38-deficient mutant. C. Omp38-deficient mutant was not as cytotoxic as the wild-type strain. Cells were infected with A. baumannii ATCC19606T (blank bar) and its isogenic Omp38-deficient mutant (black bar) for the indicated time and stained with trypan blue. Trypan blue-stained and -unstained cells were counted by a haemocytometer. D. Analysis of internucleosomal DNA fragmentation in A. baumannii ATCC19606T-infected cells (a) and its isogenic Omp38-deficient mutant-infected cells (b). Extracted DNA was electrophoresed on 1.8% agarose gel and visualized by ethidium bromide staining. Lane M, 100 bp DNA ladder.

Download figure to PowerPoint

Purified Omp38 is sufficient to induce the apoptosis of HEp-2 cells

To assess whether Omp38 is sufficient to induce apoptosis of host cells, HEp-2 cells were treated with purified Omp38 from A. baumannii ATCC19606T(Fig. 5A) and cytotoxicity was determined by measuring formazan dye in the viable HEp-2 cells. Six micrograms per millilitre or more concentrations of purified Omp38 induced the cytotoxic effects (Fig. 5B). The dose-dependent increase in the cytotoxic effect was not linear. This finding suggests that Omp38 induced the cytotoxic effects at a certain threshold. HEp-2 cells treated with elution buffer for gel filtration or 3 µg ml−1 purified Omp38 for 12 h had no effect on the DNA content, but HEp-2 cells treated with 6 and 9 µg ml−1 purified Omp38 showed 23.8% and 98.7% of hypo-diploid cell population respectively (Fig. 5C). The morphology of Omp38-treated cells (Fig. 5D) such as cellular shrinkage, membrane blebbing and chromatin condensation was indistinguishable from A. baumannii-infected cells (Fig. 1A). HEp-2 cells treated with purified Omp38 showed fragmentation of chromosomal DNA into oligonucleosomes (Fig. 5E). A similar apoptotic effect was achieved by treatment with TNFα in combination with cycloheximide. Purified Omp38 also induced apoptosis of human bronchial epithelial NCI-H292 cells and human monocyte THP-1 cells (data not shown).

image

Figure 5. Purified Omp38 induces apoptosis of HEp-2 cells. A. A Coomassie-Blue-stained gel of purified Omp38 from A. baumannii ATCC19606T (lane 1). Samples were boiled before electrophoresis. B. Cytotoxicity of HEp-2 cells treated with purified Omp38. HEp-2 cells were seeded at the concentration of 2.0 × 105 ml−1 in 96-well microplates. Omp38 was added to the wells of the microplate at a range of 3–30 µg ml−1 for the indicated time. Cellular growth was measured by using WST1. C. Flow cytometric analysis of the DNA content of purified Omp38-treated cells after staining with PI. Cells were treated with elution buffer (10 µl), TNFα (10 ng ml−1) and cycloheximide (20 µg ml−1), or purified Omp38 (3, 6, 9 µg ml−1) for 12 h. The percentage of cells in the hypo-diploid population is indicated in the diagrams. D. Morphological changes of purified Omp38-treated cells. Cells were treated with 6 µg ml−1 purified Omp38 (c and d) or left untreated (a and b) for 12 h and stained with DAPI (b and d). E. Fragmentation of chromosomal DNA into oligonucleosomes after treatment of TNFα and cycloheximide (lane 2), 6 µg ml−1 purified Omp38 (lane 3) or left untreated (lane 1) for 12 h.

Download figure to PowerPoint

Omp38 localizes to the mitochondria and induces the release of proapoptotic molecules

As mitochondrial damage was observed in the HEp-2 cells infected with A. baumannii (Fig. 3A), we investigated whether purified Omp38 localized to the mitochondria. HEp-2 cells were treated with purified Omp38 for 12 h and stained with a MitoTrackerTM (green) and polyclonal anti-rabbit Omp38 antibody, followed by Alexa Fluor® 568-conjugated goat anti-rabbit IgG antibody (red). Confocal microscopy showed that Omp38 and mitochondria colocalized and a yellow colour, which resulted from the combination of green and red colours, appeared in the merged image (Fig. 6). Next, we investigated whether localization of Omp38 to the mitochondria was capable of releasing proapoptotic molecules. Cytochrome c and AIF in the mitochondrial fraction clearly decreased in the HEp-2 cells treated with Omp38 (Fig. 7A). A nuclear appearance of AIF (Fig. 7B) and DNA fragments with high molecular weights (Fig. 7C) were observed in the Omp38-treated HEp-2 cells. The pro-form of caspase-9 and PARP decreased and an active form of both molecules increased upon treatment with Omp38 in HEp-2 cells (Fig. 7D). The activation of caspase-9 suggests that cytochrome c is released into cytosol and functions as a cofactor with Apaf-1 to activate caspase-9. However, the pro-form of caspase-8 was not cleaved by treatment with Omp38. Furthermore, a specific caspase-8 inhibitor did not inhibit the cleavage of caspase-9 and PARP in the HEp-2 cells treated with Omp38. These results indicate that Omp38 induces apoptosis of HEp-2 cells through caspases-dependent and AIF-dependent pathways, which are mediated by mitochondrial damage.

image

Figure 6. Omp38 and mitochondria colocalize. HEp-2 cells were treated with 6 µg ml−1 purified Omp38 for 12 h. Mitochondria were stained with MitoTrackerTM (green). Omp38 was labelled with polyclonal anti-rabbit Omp38 antibody, followed by Alexa Fluor® 568 secondary antibody (red).

Download figure to PowerPoint

image

Figure 7. Omp38 triggers the release of proapoptotic molecules from the mitochondria and induces apoptosis of HEp-2 cells. A. Release of AIF and cytochrome c from mitochondria. Cells were treated with 6 µg ml−1 purified Omp38 for 12 h (lane 2) or left untreated (lane 1). Proteins in mitochondrial fractions were separated on 12% SDS-polyacrylamide gel and immunoblotted with anti-AIF and anti-cytochrome c antibodies. β-Actin was used as a loading control. B. Nuclear localization of AIF. HEp-2 cells were treated with 6 µg ml−1 purified Omp38 for 12 h. Nuclei were stained with DAPI (blue). AIF was labelled with polyclonal anti-rabbit AIF antibody, followed by Alexa Fluor® 568 secondary antibody (red). Arrows indicate the nuclear localization of AIF. C. Macrodigestion of chromosomal DNA. The genomic DNA of cells was separated on a 1.0% agarose gel by pulsed-field gel electrophoresis. Cells were treated with 6 µg ml−1 purified Omp38 for 12 h (lane 2) or left untreated (lane 1). Lane M is size marker. D. Immunoblot analysis of caspase-8, caspase-9 and PARP. Cells were treated with 6 µg ml−1 purified Omp38 for 12 h. Proteins were separated on 10% or 12% SDS-polyacrylamide gel and immunoblotted with anti-caspase-8, anti-caspase-9 and anti-PARP. For inhibition assay, HEp-2 cells were treated with 50 µM Z-IETD-FMK for 1 h before the application of Omp38 and maintained during incubation.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Our results provide a novel insight into the molecular mechanism of Omp38 regarding the induction of apoptosis in A. baumannii infection. Purified Omp38 from A. baumannii entered the cells and was localized to the mitochondria. Proapoptotic molecules such as cytochrome c and AIF were released into cytosol as a consequence of mitochondrial disintegation. This event activated caspase-3, which was followed by the degradation of DNA approximately 180 bp in size, which is characteristic of apoptotic cells. On the other hand, AIF activated caspase-independent apoptosis and degraded chromosomal DNA approximately 50 kb in size, which resulted in large-scale DNA fragmentation. Live A. baumannii and purified Omp38 elicited identical effects with respect to the induction of apoptosis in epithelial cells. A. baumannii-induced apoptosis thus contrasts with that reported for a variety of bacterial pathogens, which induced caspase-dependent host cell apoptosis (Chen et al., 1996; Ruckdeschel et al., 1997; Tsai et al., 1999; Nakagawa et al., 2001; Jarvelainen et al., 2003).

The present study demonstrates unusual mechanistic features of apoptosis by inducing mitochondrial damage and the release of the proapoptotic molecule AIF. Live A. baumannii and purified Omp38 induced the release of AIF into cytosol. AIF is a soluble protein of unknown function localized in the intermembrane space of mitochondria in healthy cells and is thought to be an important executioner of apoptosis (Susin et al., 1999). Pneumococcal toxins, pneumolysin and H2O2, trigger the release of AIF from mitochondria and induce apoptosis of brain cells (Braun et al., 2002). In addition, the redox proteins of Psedomonas aeruginosa, azurin and cytochrome C551, induce apoptosis of host cells through the release of proapoptotic molecules such as cytochrome c, AIF, Smac/DIABLO, endonuclease G and Omi/HtrA2 (Punj and Chakrabarty, 2003). Additional evidence of an AIF-dependent pathway was provided by the apoptosis of Chang cells infected with piliated P. aeruginosa strains (Jendrossek et al., 2003). The current results indicate an important role for AIF in the apoptosis of epithelial cells induced by A. baumannii infection, although the precise mechanism by which this occurs has not yet been resolved.

Omps of several human pathogens such as P. aeruginosa and Neisseria gonorrheae have been demonstrated to induce apoptosis of host cells. Purified Omps from P. aeruginosa have induced apoptosis of rat seminal vesicle epithelial cells (Buommino et al., 1999). Müller et al. (1999) reported that native PorB of N. gonorrheae translocated into mitochondria, increased cytosolic Ca2+ and induced apoptosis of epithelial cells through Ca2+-dependent protease calpain and caspase family. In contrast with this result, Massari et al. (2003) demonstrated that purified PorB or live N. meningitidis inhibited apoptosis of B cells, Jurkat cells and HeLa cells. The differences between the results reported by these two groups may result from the absence of fetal bovine serum in the culture medium, because cells are more susceptible to cellular stress and apoptosis in serum-deprived conditions (Gnesutta et al., 2001). Moreover, OmpAEscherichia coli infection induced apoptosis of macrophages, while OmpA+E. coli were resistant to apoptosis of macrophages induced by staurosporine (Sukumaran et al., 2004). However, the current study clearly demonstrated apoptosis of epithelial cells induced by Omp38 from A. baumannii, which was characterized by distinct morphological and biochemical changes in apoptosis. As porins are highly conserved among Gram-negative bacteria, it would be interesting to investigate whether porins or some other kinds of Omps among Gram-negative bacteria induce apoptosis of host cells.

Omp38 is not the only trigger factor for A. baumannii-induced apoptosis. We found that A. baumannii activated caspase-8 (Fig. 2B). This suggested that live A. baumannii induced apoptosis of HEp-2 cells through a receptor-mediated pathway on the cell surface. We previously reported that apoptosis of HeLa cells was induced by the culture supernatants of A. baumannii, but not formalin-killed bacteria (Lee et al., 2001). This finding suggested that the secreted products or cell wall-associated components, but not cell surface-exposed components, likely mediated apoptosis of epithelial cells. As live A. baumannii, Omp38-deficient mutant and formalin-killed bacteria adhered to the epithelial cells (data not shown), close contact between bacteria and target cell membranes did not seem to be associated with apoptosis of host cells. The effects of lipopolysaccharide (LPS) on apoptosis could be excluded because epithelial cells did not express CD14. Experiments with purified Omp38 were carried out after neutralization of LPS with polymyxin B. Although mitochondrial damage could be explained by the effect of Omp38 in live A. baumannii-induced apoptosis, the bacterial factor associated with cell surface signalling, which activated caspase-8, was not identified.

In summary, we have demonstrated that Omp38 is a potent cytotoxin that induces apoptosis of epithelial cells in A. baumannii infection. Omp38 was localized in the mitochondria, followed by release of proapoptotic molecules such as cytochrome c and AIF, which then induced apoptosis of epithelial cells. It is concluded that Omp38 plays a central role in apoptosis of epithelial cells, which was previously an unrecognized role of Omp38 in A. baumannii. Apoptosis of epithelial cells may disrupt the mucosal lining and allow for the access of bacteria or bacterial products to the deep tissues. In this regard, the outcome of the infection with A. baumannii may depend on the induction of apoptosis in epithelial cells.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Bacterial strains

Acinetobacter baumannii ATCC19606T and its isogenic Omp38-deficient mutant KS37 were used in this study. Before each experiment, a single colony from a MacConkey agar plate (Difco Laboratories) was inoculated on a nutrient agar plate (Difco Laboratories) and was grown at 37°C for 20 h. The bacterial cells were suspended in the cell culture medium without antibiotics at a density of 2.0 × 108 cfu ml−1 for the infection.

Omp38 knockout mutant

The Omp38 knockout mutant was constructed by random transposon mutagenesis. A. baumannii ATCC19606T was mutagenized using EZ::TN <DHFR-1> Tnp Transposomes (Epicentre Technologies) according to the manufacturer's instructions. The transposon-inserted clones were selected by plating onto Mueller-Hinton agar (Difco Laboratories) plates containing 50 µg ml−1 trimethoprim. The mutant strains were co-cultured with HEp-2 cells for 20 h in 24-well microplates and the viabilities and morphological changes of HEp-2 cells were compared with the wild-type strain. Of the 1500 mutant strains, A. baumannii KS37 was not as able to induce cell death as the wild-type strain. Polymerase chain reaction (PCR) amplification with dfrA1-specific primers (forward primer: 5′-ACG GAT CCT GGC TGT TGG TTG GAC GC-3′; reverse primer: 5′-CGG AAT TCA CCT TCC GGC TCG ATG TC-3′) and Southern hybridization with dfrA1 probe were performed to determine the single-locus insertion of transposon in the chromosomal DNA. The transposon-inserted gene was directly sequenced with the following primers: forward primer: 5′-GGC GGA AAC ATT GGA TGC GG-3′ and reverse primer: 5′-GAC ACT CTG TTA TTA CAA ATC G-3′. Sequencing analysis showed that transposon was inserted into the omp38 gene (GenBank Accession No. AY485227).

Purification of Omp38

Bacteria were grown in Luria–Bertani (LB) broth supplemented with 100 mM NaCl at 30°C for 16 h with constant shaking. Bacterial cells were harvested, sonicated and centrifuged at 1700 g for 20 min. The supernatant was centrifuged at 100 000 g for 1 h at 4°C. The pellet containing cell envelope was resuspended in 10 mM Hepes buffer with 2% sodium lauryl sarcosine, and incubated for 30 min at room temperature to solubilize the inner membrane. Then the suspension was centrifuged at 100 000 g for 1 h at 4°C. The outer membrane fractions were loaded onto a Sephacryl S-300 (Pharmacia) column equilibrated with elution buffer (0.1% SDS, 0.1 mM EDTA, 0.2 M LiCl2, 50 mM Tris-HCl pH 7.5) and eluted with the same buffer. The fractions were concentrated using a Centricon 100 centrifugal filter unit (Millipore) and massively dialysed against PBS. To remove the effects of SDS, the purified Omp38 was diluted at least 200-fold and applied to the cells. The purified porin preparations were pre-incubated for 1 h at room temperature with 10 µg ml−1 polymyxin B to neutralize possible contamination with LPS. A pool of protein-free gel filtration fractions was used as a negative control in the experiments.

Cell culture and induction of apoptosis

HEp-2 cells from human laryngeal epithelial cells were grown in Dulbecco's modified Eagle medium (DMEM; Gibco BRL) supplemented with 10% fetal bovine serum (HyClone), 2 mM l-glutamine, 1000 U ml−1 penicillin G and 50 µg ml−1 streptomycin at 37°C in 5% CO2. The confluent growth was obtained in 100 mm-diameter dishes and the cells were routinely passaged every 3 days. Cells were seeded in new 100 mm-diameter dishes or 24-well plates to infect bacteria or treat porin. Before the induction of apoptosis, cells in 80% confluency were washed three times with pre-warmed PBS and inoculated with live bacteria at a multiplicity of infection (moi) of 100 or treated with different concentrations of purified Omp38. TNFα (10 ng ml−1) combined with cycloheximide (20 µg ml−1) was used as a positive control in the experiments. For an inhibition assay, HEp-2 cells were treated with 100 µM of pancaspase inhibitor (Z-VAD-fmk; Calbiochem) and 50 µM of caspase-8 inhibitor (Z-IETD-FMK; Calbiochem) for 1 h before the application of bacteria and Omp38 respectively.

Determination of cell growth

The growth inhibition of HEp-2 cells was measured using the Premix WST1 cell proliferation assay system (TaKaRa Shuzo). WST1 is a type of tetrazolium salt that cleaves to formazan dye by the succinate-tetrazolium reductase, which exists in the mitochondrial respiratory chain and is active only in viable cells. The dark red formazan dye formed by metabolically active cells was quantified by measuring its absorbance. HEp-2 cells were seeded at a concentration of 2.0 × 105 ml−1 in a 96-well microplate. Purified Omp38 was added to the wells of the microplate at a range of 3–30 µg ml−1. The growth inhibition of the cells was measured at 450 nm 3 h after treatment with WST1.

Subcellular fractionation and immunoblotting

Cells were treated with 6 µg ml−1 purified Omp38. The adherent and detached cells were harvested and the mitochondrial fraction was isolated using a Mitochondria/Cytosol fractionation kit (BioVision) according to the manufacturer's instructions. The samples were separated by electrophoresis with 12% SDS-PAGE and an immunoblot was performed. Antibodies used were polyclonal anti-rabbit AIF (Santa Cruz Biotechnology) and monoclonal anti-mouse cytochrome c (Santa Cruz Biotechnology).

Immunoblotting

Cells were lysed in lysis buffer (10 mM Tris pH 7.4, 5 mM EDTA, 130 mM NaCl, 1% Triton X-100, 10 µg ml−1 PMSF, 10 µg ml−1 aprotinin, 10 µg ml−1 leupeptin, 5 mM phenanthroline and 28 mM benzamidine-HCl) for 30 min on ice. The lysates were cleared by centrifugation and then quantified using a Bradford assay. A portion (30 µg of protein) of each sample was separated with 10% or 12% SDS-PAGE, followed by electrotransfer onto the nitrocellulose membranes (Hybond-ECL; Amersham Pharmacia Biotech). The blots were blocked in 5% non-fat skim milk and incubated with primary antibodies. Proteins were visualized by incubation with horseradish peroxidase-conjugated secondary antibodies, followed by enhanced chemilluminescence (ECL plus; Amersham Pharmacia Biotech) according  to  the  manufacturer’s  instructions.  Primary  antibodies against PARP (Oncogene), caspase-3 (Calbiochem), caspase-8 (Calbiochem), caspase-9 (Calbiochem) and β-actin (Santa Cruz Biotechnology) were applied at the optimized concentrations.

DNA fragmentation analysis

The adherent and detached cells were collected, washed with ice-cold PBS and resuspended in lysis buffer (0.5% Triton X-100, 5 mM Tris, 20 mM EDTA, pH 7.4). The lysates were collected in ice and centrifuged at 4°C. The supernatants were incubated with RNase A (5 µg ml−1) at 37°C for 1 h, followed by 1 h incubation of proteinase K (200 µg ml−1) at 50°C. After two extractions with phenol/chloroform, the DNA was precipitated with ethanol, dried and resolved in TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0). Fragmented DNAs were analysed by 1.8% agarose gel and visualized by ethidium bromide staining under ultraviolet light.

Pulsed-field gel electrophoresis

The genomic DNA of HEp-2 cells treated with purified Omp38 was separated on a 1.0% agarose gel using a contour-clamped homogeneous-field apparatus (CHEF DRIII systems; Bio-Rad Laboratories) in a 0.5× TBE buffer. The conditions for electrophoresis were 6 V cm−1 for 15 h with an increasing pulse time from 5 s to 20 s. A lambda DNA ladder, comprised of 48.5 kb concatemers (Bio-Rad Laboratories), was used as the size standard.

Quantification of hypo-diploid cells

The DNA contents were determined on a single-cell level using a flow cytometer (Beckman Coulter). The adherent and detached cells were collected, washed twice in PBS and resuspended in 300 µl of PBS containing 5% fetal bovine serum. Cells were suspended in 300 µl of PBS containing 50 µg ml−1 propidium iodide (PI) and 0.5 mg ml−1 RNase A, and incubated in the dark for 20 min. After two more washes with PBS, the DNA content of 10 000 cells per sample was assessed on a flow cytometer by plotting the PI fluorescent intensity in a histogram plot.

Fluorescent and confocal microscopic analysis

Cells were seeded in glass coverslips and infected with bacteria or treated with Omp38 as previously described. Nuclear changes, such as chromatin condensation and nuclear fragmentation, were analysed by staining with 4′,6-diamidino-2-phenyllindole dihydrochloride (DAPI; Molecular Probes). After the cells were treated for the indicated times, the cells were fixed in 3% paraformaldehyde and stained with DAPI (10 µg ml−1) for 10 min in the dark. The stained cells were observed using a fluorescent microscope (Carl Zeiss). To observe mitochondrial damage, the culture medium was removed and the pre-warmed medium containing 100 nM of MitoTrackerTM (Molecular Probes) was added. Cells were incubated for 10 min and the pre-warmed medium was replaced by fresh medium. To determine the localization of AIF, cells were treated with purified Omp38 for 12 h and stained with DAPI (10 µg ml−1) and polyclonal anti-rabbit AIF antibody. The subcellular localization of Omp38 was analysed using a confocal microscope system (Carl Zeiss). After the cells were treated with purified Omp38 for 12 h, mitochondria were first stained with 100 nM MitoTrackerTM. The cells were immersed in the solution of 3.5% paraformaldehyde and 0.2% Triton X-100 for 10 min. Omp38 was stained with polyclonal anti-rabbit Omp38 antibody and Alexa Fluor® 568-conjugated goat anti-rabbit IgG antibody (Molecular Probes).

Transmission electron microscopy

Bacteria were grown in LB broth and then washed twice with PBS. After centrifuging in a microcentrifuge tube, bacterial cells were fixed with 2.5% glutaraldehyde. Then the samples were dehydrated in a series of ethanol concentrations and embedded in Epon. Thin sections were cut using an ultramicrotome with a diamond knife. Bacteria were examined with a transmission electron microscope (Hitachi 7000).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This study was supported by a grant from the Korea Science and Engineering Foundation, Republic of Korea (Project No. R05-2003-000-10067-0).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References