Published online before print January 30, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Division of Medical Microbiology, Department of Molecular and Clinical Medicine, Faculty of Health Sciences, Linköping University, Linköping, Sweden; and
State Key Laboratory of Biocontrol, College of Life Sciences, Sun Yatsen (Zhongshan) University, Guangzhou, China
1 Correspondence: Division of Medical Microbiology (IMK), Department of Molecular and Clinical Medicine, Faculty of Health Sciences, Linköping University, SE-581 85, Linköping, Sweden. E-mail: robert.blomgran{at}imk.liu.se
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: bacteria • Bcl-2 family proteins • Escherichia coli • lysosomes • Mcl-1 • reactive oxygen species
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
ROS is an important component in the complex modulation of neutrophil apoptosis [5 6 7 8 ], and we recently showed that it is not the generation of ROS per se but the location of such production that is vital for initiation of apoptosis [4 ]. Using an adherent, Type 1-fimbriated Escherichia coli strain, which resisted phagocytosis, and also the complement-opsonized, isogenic mutant stain lacking FimH adhesins, we observed that phagosomal ROS, generated during receptor-mediated phagocytosis, did not lead to apoptosis, whereas intracellular, nonphagosomal ROS, produced in response to adherent bacteria, resulted in ROS-dependent apoptosis [4 ]. This finding argues against the general assumption that ROS always triggers phagocytosis-induced cell death and instead, puts more emphasis on the overall signaling property and effectors of different bacterial pathogens and foremost, the compartmentalization of such a ROS production according to the ability to triggerapoptosis. However, the precise underlying mechanism for the effects of ROS compartmentalization during apoptosis remains unclear.
Involvement of lysosomes and lysosomal proteases, such as cathepsins, is emerging as an additional point of entry into apoptosis in a variety of cell types. The stimuli known to induce lysosomal membrane permeabilization (LMP) and release of proteases include TNF-
receptor (TNF-R)-activated caspase 8 [9
]. However, ROS have been found to bring about a rapid and a more direct effect on the lysosomal membrane through a Fenton-like reaction, causing peroxidative damage to the lipids in the membrane [10
]. Although cathepsins may function as execution proteases in the late stages of apoptosis [11
], ROS-mediated translocation of cathepsins from the lysosomes to the cytosol occurs early during apoptosis, before loss of mitochondrial membrane potential (
m), release of cytochrome c, and caspase activation [12
, 13
]. Moreover, cathepsin-mediated activation of specific Bcl-2 family proteins such as Bax or Bid suggests that cathepsins induce apoptosis through the mitochondrial pathway [14
15
16
]. The regulation of LMP-induced apoptosis involves not only proapoptotic Bcl-2 proteins but also the antiapoptotic Bcl-2 family member, Bcl-2, which can inhibit oxidative stress-induced apoptosis by preventing secondary LMP [17
].
It is generally agreed that human neutrophils do not express the antiapoptotic protein Bcl-2 but instead, express the antiapoptotic proteins Mcl-1, A1, and Bcl-XL [18 , 19 ]. Mcl-1 contains targeting motifs for rapid turnover by the proteasome, although when survival is triggered, the level of Mcl-1 expression is increased greatly by de novo synthesis, which is correlated with the reduction in apoptosis [19 ]. Gradual loss of Mcl-1 during apoptosis releases Bax from its Mcl-1:Bax heterodimerization, which enables activation of Bax and translocation of that protein into the mitochondrial membrane, leading to membrane permeabilization and apoptosis [20 ]. This effect, together with the observation that Bid is the only Bcl-2 family member that functions as an agonist of the proapoptotic Bcl-2 proteins Bax and Bak [21 ], suggests that Bid and Mcl-1 are critical for balancing survival and apoptosis in human neutrophils.
Although the correlation between ROS-induced apoptosis and changes in the stability of the lysosomal membrane has been found in several types of cells [13 , 22 23 24 25 ], little is known about the role that lysosomes or azurophilic granules, which are the closest resemblance to primary lysosomes in neutrophils, play in microbe-induced apoptosis in human neutrophils. Here, we used a system with phagosomal versus nonphagosomal production of ROS [4 ] to study the potential role of LMP as a mechanism of the ROS-dependent apoptosis induced by Type 1-fimbriated, uropathogenic E. coli. We found that intracellular, nonphagosomal ROS, generated in response to the adherent bacteria, triggered LMP, accompanied by release of cysteine-cathepsins, which affected and targeted Bcl-2 proteins involved in the intrinsic apoptotic pathway. In contrast, compartmentalization of ROS to the phagosome sustained the integrity of the azurophilic granules and protected the neutrophils from apoptosis.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Antibodies and inhibitors
The primary antibodies anti-Bid (rabbit polyclonal) and antihuman Bax (rabbit polyclonal) were purchased from BD PharMingen (San Diego, CA, USA), and antibodies against Mcl-1 (S-19), sc-819 (rabbit polyclonal), and actin (I-19), sc-1616 (rabbit polyclonal) were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). The irreversible cysteine protease inhibitor (25,35)trans-epoxysuccinyl-L-leucylamindo-3-methylbutane ethyl ester (EST; E-64d), the reversible aspartic protease inhibitor pepstatin A, and the reversible caspase inhibitor z-Val-Ala-Asp-fluoromethylketone (z-VAD-fmk) were obtained from Calbiochem (San Diego, CA, USA). Stock solutions of EST and z-VAD-fmk (25 mM, dissolved in DMSO) were stored at –20°C and used at a concentration of 100 µM and 25 mM, respectively. Pepstatin A was made fresh daily (25 mM, dissolved in DMSO) and used at 100 µM. Stock solutions of DPI (5 mM, dissolved in 50% DMSO and 50% H2O) were stored at –20°C and used at 5 µM.
Preparation and treatment of human neutrophils
Gradient centrifugation was performed essentially as described previously to isolate neutrophils from heparinized whole blood obtained from healthy human donors [26
]. Briefly, neutrophils in the interphase between Polymorph PrepTM and LymphoprepTM were collected and washed in PBS, after which, contaminating erythrocytes were removed by a brief hypotonic lysis, and neutrophils were washed in KRG without Ca2+. Neutrophils of 
98% purity were then resuspended in RPMI-1640 medium supplemented with 10% heat-inactivated FCS and 2 mM L-glutamine, here, simply referred to as RPMI medium.
In experiments using an inhibitor, that substance was present during stimulation and subsequent incubation of neutrophils. The NADPH-oxidase activity of neutrophils was inhibited by preincubating the cells with DPI for 15 min at 37°C [27 ]. The concentration of DPI used did not affect phagocytosis of serum-opsonized bacteria [4 , 28 ]. The times and concentrations of pretreatments with the protease inhibitors were optimized to 45–60 min, as neutrophils are short-lived cells that undergo constitutive apoptosis.
Bacterial strains, growth conditions, and preparations
Wild-type E. coli Strain 1177 (Serotype O1:K1:H7), expressing Type 1 fimbriae isolated from children with their first episode of acute pyelonephritis, and the derived isogenic, Type 1 fimbriae mutant CN1016 (here referred to as FimH–) and add-back mutant CN1018 (here referred to as FimH+) were kindly provided by Dr. Catharina Svanborg (Department of Laboratory Medicine, Lund University Hospital, Lund, Sweden) [29
]. All the strains were hemolysin-negative and expressed P-fimbriae [29
]. The bacteria were grown statically overnight in Luria broth supplemented with appropriate antibiotics selective for the expression of Type 1 fimbriae, as described elsewhere [29
]. The bacterial concentration was estimated by spectrophotometry (Hitachi U-1100 spectrophotometer, Hitachi Ltd., Tokyo, Japan) and verified by microscopy and assay of CFU. Aliquots of the bacteria culture were washed in PBS (1400 g for 10 min at 4°C) and then resuspended in RPMI medium and placed on ice until used. Before performing experiments, Type 1 fimbrial expression and/or integrity of the FimH adhesin were confirmed by a mannose-sensitive yeast cell agglutination assay, as described previously [4
].
To opsonize the bacteria, aliquots of the culture were washed and resuspended in KRG, supplemented with 20% AB serum from healthy human donors, and then incubated for 30 min at 37°C. Thereafter, the bacteria were washed again and resuspended in RPMI medium.
Neutrophil-bacteria interaction
Freshly isolated neutrophils were suspended in RPMI medium and plated in a 24-well tissue-culture plate, after which, they were prewarmed for 10–15 min at 37°C in a humidified CO2 incubator (5% CO2, 95% air) and subsequently exposed to prewarmed bacteria or medium as a control. Infection was stopped after 40 min by adding gentamicin (50 µg/ml), and the incubation was then continued at 37°C in 5% CO2. At the time-points indicated in the figure legends, the cells were washed with PBS and used for further analysis.
Lysosomal membrane stability assays
AO is a lysosomotropic, weak base (pKa=10.3), which through the process of proton trapping, is retained in its charged form (AOH+) inside the acidic vacuolar compartment. To determine how various treatments affected lysosomal membrane stability in the neutrophils, the cells were incubated with RPMI medium supplemented with AO (5 µg/ml for 15 min at 37°C) at the end of all experiments, and the cells that displayed a reduced number of intact AO-accumulating lysosomes, so called "pale cells", were analyzed as having decreased red fluorescence (the AO uptake method) [17
, 24
]. Red fluorescence (FL-3 channel) was measured by flow cytometry (FACSCalibur, BD Biosciences, San Jose, CA, USA) and evaluated using CellQuest software.
To further determine the level of lysosomal membrane damage in neutrophils, we measured the activity of the 250-kDa azurophilic granule protein, ß-N-acetylglucosaminidase (NAG), in digitonin-extracted cytosols. In short, 1 h after stimulation, neutrophils were treated with an extraction buffer (250 mM sucrose, 20 mM Hepes, 10 mM KCl, 1.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, and 1 mM Pefabloc, pH 7.5) containing 20 µg/ml digitonin (or 100 µg/ml digitonin to determine total cellular NAG activity) for 10 min on ice, inverting the tubes every 2 min. NAG activity in different cytosol extracts (1x106 neutrophils/sample) was estimated by adding three volumes of 0.2 M sodium citrate buffer (pH 4.5) containing 300 µg/ml 4-methylumbelliferyl-2-acetamido-2-deoxy-ß-D-glucopyranoside. The Vmax of the liberation of methylumberliferyl (excitation, 360/40 nm; emission, 460/40 nm) was measured for 40 min at 37°C in a Synery HT multidetection microplate reader from BioTek Instruments, Inc. (Winooski, VT, USA). The digitonin concentration and incubation times were optimized to achieve maximum cytosolic lactate dehydrogenase (LDH) activity without disruption of lysosomes. LDH activity was assayed in the presence of 0.34 mM sodium pyruvate and 0.23 mM ß-NADH in 28 mM sodium phosphate buffer (pH 7.4) by measuring the decrease in absorbance at 340 nm.
Cell lysis and Western blotting
Immunoblotting experiments were performed in polypropylene Eppendorf tubes. Neutrophils (2x106 per sample) were allowed to equilibrate for 10 min (or 45 min when inhibitors were used) at 37°C before being stimulated. The reactions were terminated by adding ice-cold PBS, supplemented with 1 mM Na3VO4 and 1 mM Pefabloc, and then subjecting the samples to rapid centrifugation. Each pellet was resuspended in lysis buffer containing 0.1% SDS, 1% Triton X-100, 150 mM NaCl, 50 mM Tris (pH 7.4), 10 mM NaF, 10 mM EGTA, 1 mM Na3VO4, 10 µg/ml aprotinin, 2 µg/ml leupeptin, 2 µg/ml pepstatin, and 1 mM Pefabloc, and the suspension was incubated on ice for 20 min and centrifuged at 12,000 g for 5 min. The resulting lysate was dissolved in Laemmli sample buffer [30
] and boiled for 5 min. Equal amounts of proteins were separated by 8–16% SDS-PAGE and electrotransferred onto nitrocellulose membranes, which were blocked with 5% milk powder in PBS Tween-20 (0.075%) and incubated with primary antibodies recognizing Bid (1:2000), Bax (1:1000), or Mcl-1 (1:250) or with actin (1:2000) as a loading control. If the molecular weight of the protein of interest overlapped with the weight of another protein, the blots were stripped and reprobed; otherwise, membranes from the same blot were split according to the molecular standard and then used to detect several proteins. The specific proteins were detected with a commercial ECL kit (Amersham, Little Chalfont, UK). Densitometric analysis of bands was performed using Quantity One (Bio-Rad, Hercules, CA, USA).
Immunoprecipitation
For immunoprecipitation, 1 x 107 neutrophils per sample were used. After termination of the reactions, the pelleted cells were resuspended in 1 ml sonication buffer, composed of 150 mM NaCl, 50 mM Tris (pH 7.4), 10 mM NaF, 10 mM EGTA, 1 mM Na3VO4, 10 µg/ml aprotinin, 2 µg/ml leupeptin, 2 µg/ml pepstatin, and 1 mM Pefabloc, and lysed through pulse sonication (duration: 8x5 s; amplitude: 6 µm). The lysates were treated with 1 µl Benzonase® for 30 min and thereafter, with 50 µl Protein G Plus/Protein A-Agarose beads for 20 min (at rotation at 4°C). From the precleared lysates (500 g for 3 min at 4°C), Bax was immunoprecipitated using 6.4 µg antihuman Bax mAb (Clone YTH-2D2 from Nordic Biosite, Täby, Sweden) or a mouse IgG1 control antibody, and the immuncomplex was captured with 50 µl Protein G Plus/Protein A-Agarose beads, which were washed four times in sonication buffer containing inhibitors, and the complexes were removed by boiling in Laemmli sample buffer and analyzed on an 8–16% gradient SDS-PAGE gel.
m
m was assessed by use of tetramethylrhodamine ethyl ester (TMRE), which accumulates in the mitochondrial matrix. Decreased m was indicated by a reduction in the intensity of TMRE-induced red fluorescence, detected at the end of experiments by flow cytometric analysis. Neutrophils were collected and loaded with RPMI medium supplemented with 1 µM TMRE; this was done for 15 min in the dark at 37°C. Thereafter, the cells were analyzed by performing flow cytometry, using an activating wavelength of 488 nm and monitoring the red fluorescence in the FL-3 channel (650 LP filter).
Assessment of apoptosis
As an indicator of early apoptosis, we analyzed the exposure of phosphatidylserine (PS) on the surface of apoptotic neutrophils by staining with FITC-conjugated annexin V, according to the protocol from the manufacturer (R&D Systems). Nonspecific binding of annexin V to the studied bacteria was excluded. Neutrophils were washed once in binding buffer (150 mM NaCl, 5 mM KCl, 1 mM MgCl2, 1.8 mM CaCl2, and 10 mM HEPES, pH 7.4), and the specific binding of annexin V was achieved by incubating 106 neutrophils for 15 min at 4°C in the dark in 60 µl binding buffer containing a saturated concentration of FITC-annexin V. To discriminate between early apoptosis and necrosis, prior to analysis, the cells were stained simultaneously with annexin V and propidium iodide (PI). The binding of FITC-annexin V (FL1) and PI (FL2) to the cells was measured by FACS and analyzed using CellQuest software [28
]. A rise in FITC fluorescence (FITC+/PI–), corresponding to increased exposure of PS, was considered to reveal early apoptotic neutrophils, whereas elevation of FITC and PI fluorescence was regarded as indicating late apoptotic or necrotic cells [31
]. At least 10,000 cells were counted in each sample, and a gate based on forward- and side-scatter was set to exclude cell debris.
Statistical analysis
All data are given as means ± SEM, unless otherwise indicated. The significance of the differences was analyzed by Student’s t test (paired).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (18K): [in a new window] |
Figure 1. ROS and protease-dependent apoptosis in Type 1 fimbria-stimulated neutrophils. (A) Neutrophils were prewarmed for 15 min at 37°C and then exposed to Type 1-fimbriated E. coli (FimH+; bacteria:neutrophil ratio, 20:1) or medium for 40 min, followed by subsequent cultivation in the presence of gentamicin (total times with bacteria are given). The levels of apoptosis were determined by flow cytometric analysis of annexin V binding. The illustrated results represent means ± SEM for three separate experiments. (B) Neutrophils were incubated in medium alone (–) or with 100 µM EST (cysteine-cathepsin inhibitor), 25 µM z-VAD-fmk (general caspase inhibitor), or 5 µM DPI (NADPH oxidase inhibitor) for 1 h at 37°C and then exposed to bacteria (bacteria:neutrophil ratio, 20:1) or medium for 40 min. The neutrophils were subsequently cultured in the presence of gentamicin for 5 h, after which, the levels of apoptosis were determined as in A. The illustrated results represent means ± SEM for four separate experiments. *, Significant differences at P < 0.05. Only 1–3% of the neutrophils (exposed or not exposed to the bacteria) exhibited counterstaining with PI, showing that the treatment did not cause necrosis.
|
 View larger version (24K): [in a new window] |
Figure 2. LMP and protease release induced in neutrophils by Type 1-fimbriated E. coli. Neutrophils were stimulated with nonfimbriated bacteria (FimH–), serum-opsonized, nonfimbriated bacteria (ops-FimH–), or fimbriated bacteria (FimH+) at a bacteria:neutrophil ratio of 20:1. Flow cytometric analysis of lysosomal integrity achieved by monitoring decrease in red fluorescence (pale cells), detected by the AO-uptake method (A and B). (A) Representative histograms of neutrophils stimulated with FimH– (thin line) and FimH+ (bold line), where M1 is set to evaluate the percentage of neutrophils, which have lost their AO-accumulating lysosomes, so-called, pale cells. (B) The illustrated data represent the indicated time-points and means ± SEM of the percentage of pale cells from three independent experiments. *, **, Significant differences compared with neutrophils treated with medium alone (*, P<0.05; **, P<0.01). (C) The activity of the lysosome marker NAG, measured as the Vmax of liberation of methylumbelliferyl from the NAG substrate in neutrophil cytosol extracts prepared with 20 µg digitonin/ml 1 h after exposure to bacteria. The NAG activity in the samples was determined by comparison with the total NAG activity (using 100 µg digitonin/ml). The NAG activity in medium-stimulated control neutrophils was set to 100% (12.8% of the total NAG activity, solid line). The illustrated results represent means + SEM from three to five independent experiments (n=3 for CN1016 and ops-CN1016; n=5 for medium and CN1018). *, **, Significant differences compared with medium-stimulated cells (*, P<0.05; **, P <0.01).
|
40% higher in neutrophils exposed to bacteria expressing Type 1 fimbriae than in control, whereas FimH-stimulated neutrophils showed an 20% decrease in such activity compared with controls. These data show that Type 1-fimbriated E. coli induced LMP with concomitant release of the contents of azurophilic granules. In contrast, the isogenic, Type 1 fimbriae mutant actually protected against LMP.
Type 1-fimbriated E. coli induced LMP through oxygen-dependent mechanisms
We used the NADPH-oxidase inhibitor DPI to test if the oxidative burst and intracellular release of ROS are responsible for triggering LMP in bacteria-stimulated human neutrophils. After 3 and 21 h of stimulation, Type 1 fimbria-induced LMP was inhibited significantly by DPI (Fig. 3
). However, the high level of LMP seen after 21 h in the spontaneously apoptotic control cells was not related to production of ROS, showing that different mechanisms give rise to Type 1 fimbria-induced (rapid) and spontaneously occurring (slow) LMP in human neutrophils. Further, the high DNA fragmentation induced by the fimbriated bacteria (40%) shows that bacteria-induced cell death, triggered through the action of LMP, results in apoptosis and no necrosis (data not shown).
 View larger version (9K): [in a new window] |
Figure 3. Type 1 fimbria-induced LMP mediated by the production of ROS. Neutrophils were pretreated for 15 min at 37°C, with or without the NADPH-oxidase inhibitor DPI (5 µM) and were subsequently stimulated with nonfimbriated bacteria (FimH–), serum-opsonized, nonfimbriated bacteria (ops-FimH–), or fimbriated bacteria (FimH+) at a bacteria:neutrophil ratio of 20:1. The degree of LMP was analyzed by the AO-uptake method. The illustrated results are means ± SEM of the percentage of neutrophils that had lost their AO-accumulating lysosomes in three to four independent experiments (n=3 for data collected after 3 h, and n=4 for data collected after 21 h of exposure to the bacteria). *, Significant difference compared with neutrophils stimulated with FimH+ in the absence of DPI for the same amount of time (P<0.05).
|
-activated cells [33
]. No Bid cleavage could be discerned in control neutrophils or those exposed to 100 µg/ml LPS for 180 min (data not shown). Furthermore, no change could be seen in the proapoptotic protein Bax over the same period, although it is possible that Bax was activated through conformational changes, not detected by Western blotting. Mcl-1 is an antiapoptotic protein, which has a safeguarding function in neutrophils similar to the protective role that Bcl-2 plays in other types of cells. Stimulation of the phagocytes with the Type 1-fimbriated E. coli diminished the expression of Mcl-1 over time, whereas levels of this protein were somewhat increased in control neutrophils after 90 min, indicating that up-regulation of Mcl-1 may promote survival in neutrophils. Although the activity of proapoptotic and antiapoptotic Bcl-2 proteins is important, it is the ratio of pro- and antiapoptotic molecules within this family that sets the threshold of susceptibility toward apoptosis in neutrophils [34
]. We, therefore, formed the antiapoptotic Mcl-1:t-Bid ratio using band densitometry measurements of the total amounts of Mcl-1 expression and Bid-cleavage product, generated at the indicated time-points (Fig. 4B)
. The ratio remained constant in neutrophils incubated in medium alone, whereas cells exposed to fimbriated bacteria showed a reduction in this ratio already after 30 min, indicating a shift toward apoptosis.
 View larger version (18K): [in a new window] |
Figure 4. Type 1-fimbriated E. coli induced early cleavage of the Bcl-2 protein Bid and down-regulation of Mcl-1. Neutrophils were stimulated with medium alone or with the Type 1-fimbriated E. coli (FimH+) at a ratio of 20 bacteria:1 neutrophil. Thereafter, the neutrophils were lysed, and nucleus-free supernatants were run on SDS-PAGE, and the blots were probed with anti-Bid, anti-Bax, anti-Mcl-1, and antiactin (as loading control). (A) Representative blots of four separate experiments, where the OD measurements of the Mcl-1 bands are included (normalized mean values, n=4). (B) Antiapoptotic Mcl-1:t-Bid ratios. The relative band density of total Mcl-1 protein expression and the Bid cleavage product (t-Bid) were corrected against loading controls, and the resulting values were used to calculate the antiapoptotic Mcl-1:t-Bid ratio. The illustrated results are means ± SEM from four independent experiments. *, **, Significant differences compared with medium-stimulated neutrophils (*, P<0.05; **, P<0.01). (C) Coimmunoprecipitation of Mcl-1 with Bax. Neutrophils were stimulated with medium alone or with the Type 1-fimbriated bacteria. The cells were lysed in a detergent-free buffer through pulse sonication, and extracts were immunoprecipitated with anti-Bax (IgG1) or IgG1 mouse control protein (IP). Western blot analyses (WB) were performed with rabbit anti-Mcl-1. The amount of Bax in immunoprecipitates was determined by analysis with rabbit anti-Bax antibody. Representative blots from three separate experiments are shown, where the normalized values for the OD are shown as means, and the highest mean values are set to 1.0.
|
Cleavage of Bid requires cathepsins but not caspases
As Bid cleavage is an important event in the lysosomal protease pathway to apoptosis [15
], we used the cathepsin inhibitor EST to inhibit cysteine proteases (mainly cathepsin B and L) and pepstatin A to inhibit cathepsin D specifically (aspartic proteases). The differential effects of the two cathepsin inhibitors on apoptosis coincided with their impacts on Bid cleavage: Pepstatin A had no influence on splitting the protein, whereas EST (blocks cathepsins B and L) protected against the cleavage significantly. In addition, the cleavage of Bid was not affected by the general caspase inhibitor z-VAD-fmk (Fig. 5
). These findings provide further evidence that the cleavage of Bid was mediated by released proteases and not through caspase-dependent mechanisms.
 View larger version (38K): [in a new window] |
Figure 5. Cysteine-cathepsin-dependent cleavage of Bid. Neutrophils were incubated in medium alone (–) or were pretreated with pepstatin A (Pep), EST, or z-VAD-fmk (Z), as described in the legend of Figure 1
, and were then exposed for 90 min to the Type 1-fimbriated E. coli (FimH+) at a bacteria:neutrophil ratio of 20:1 or to medium alone. Proteins from nucleus-free lysates were run on SDS-PAGE, and blots were probed for anti-Bid. (A) A blot representative of four separate experiments. (B) Results of band densitometric analysis of the specific Bid cleavage product t-Bid expressed as means ± SEM of four independent experiments. Significant differences compared with medium-stimulated neutrophils and stimulated with fimbriated bacteria in the absence of inhibitor are, respectively, indicated as follows: **, P < 0.01; ##, P< 0.01.
|
m in neutrophilsm by using the cationic and lipophilic dye TMRE (Fig. 6
). Twenty-one hours after stimulation, the nonfimbriated bacteria, opsonized and unopsonized, induced less mitochondrial damage than seen in control cells. Similar to LMP, maintenance of the m by the nonfimbriated bacteria at 21 h correlated with a low level of apoptosis.
 View larger version (15K): [in a new window] |
Figure 6. Type 1-fimbriated E. coli induced early cysteine-cathepsin-dependent mitochondrial damage in neutrophils. Flow cytometry was used to analyze changes in m by monitoring decreased red fluorescence of TMRE after 1.5, 3, 5, and 21 h of stimulation with the E. coli strains (explained in legend of Fig. 2
). (A) These illustrated results were obtained from the identical preparation of cells and run in parallel, as results illustrated in Figure 2B
, and are results from three independent experiments shown as means ± SEM for the percentage of neutrophils with decreased m. *, Significant differences compared with medium-stimulated neutrophils (P<0.05). (B) Neutrophils were untreated (–) or were pretreated with the indicated inhibitors, as described in the legend of Figure 1
, and were subsequently exposed to the Type 1-fimbriated bacteria. The illustrated results represent means ± SEM of the percentage of neutrophils with decreased m 3 h after stimulation observed in four independent experiments. Significant differences compared with unstimulated neutrophils and with neutrophils stimulated with FimH+ in the absence of inhibitors are, respectively, as follows: *, P < 0.05; #, P < 0.05.
|
m (Fig. 6A)
, and to examine the role of LMP and lysosomal proteases in that effect, we preincubated neutrophils with the cathepsin inhibitors and recorded the m after 3 h (Fig. 6B)
. The fimbriated bacteria caused a cysteine-cathepsin-dependent decrease in the m (P<0.05, compared with and without the cysteine-cathepsin inhibitor EST), whereas no significant change was seen when we used the cathepsin D inhibitor pepstatin A (neutrophils not pretreated and pretreated with the inhibitor had values of 26.6%±3.5% and 30.6%±3.7%, respectively; mean±SEM, n=4). Also, the NADPH-oxidase inhibitor DPI impeded the bacteria-induced reduction in m significantly, indicating that LMP is required for mitochondrial damage. Similar to the results regarding Bid cleavage, we found that the decrease in m was caspase-independent, as the caspase inhibitor z-VAD-fmk had no effect on the potential. As shown in Figures 2B
and 6A
(experiment run in parallel from identical cell preparations), the fimbriated bacteria led to similar kinetics for LMP and mitochondrial damage with the exception of a slower rate of mitochondrial membrane permeabilization after exposure to the bacteria for 1.5 h (17.5%±3.8% and 29.4%±1.7% for cells with lost m and AO-accumulating lysosomes, respectively). Inasmuch as neither the cathepsin inhibitors nor the caspase inhibitor affected the bacteria-induced LMP (results not shown), the data shown in Figure 6B
suggest that the mitochondrial damage occurred downstream of the LMP in Type 1 fimbria-stimulated neutrophils.
In contrast to bacteria-induced apoptosis, spontaneous apoptosis measured at 21 h did not depend on cysteine-cathepsins for induction of mitochondrial damage (Fig. 7A
). Instead, this decrease in m was dependent only on the activity of caspases. However, the late LMP in spontaneous apoptotic neutrophils had a synergistic effect on apoptosis, indicating that cysteine-cathepsins and caspases contribute to the induction and cellular demise through separate pathways (Fig. 7B)
.
 View larger version (12K): [in a new window] |
Figure 7. Involvement of cysteine-cathepsins and caspases in neutrophil spontaneous apoptosis. Neutrophils were cultured for 21 h in medium alone (–), with 100 µM EST, with 20 µM z-VAD-fmk, or with a combination of 100 µM EST and 20 µM z-VAD-fmk, and thereafter, the percentage of cells with decreased m (A) and the level of apoptosis (B) were analyzed as described in the legend of Figures 6
and 1
, respectively. (A and B) The illustrated results represent means ± SEM of five separate experiments, which were run in parallel. *, **, ***, Significant differences compared with neutrophils incubated in medium alone (*, P<0.05; **, P<0.01; ***, P<0.001).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
m, and caspase activation. These findings present a novel pathway for apoptosis in activated neutrophils.
Although the caspase inhibitor did not suppress changes in membrane stability in neutrophil lysosomes or mitochondria, it was still as effective as the NADPH oxidase and cysteine-cathepsin inhibitor in reducing apoptosis induced by the fimbriated bacteria. This finding indicates that postmitochondrial involvement of caspases also provides a critical checkpoint for apoptosis signaling in Type 1 fimbria-stimulated neutrophils. The identification of effector caspase-3 and -7 as key mediators of the loss of m through a possible amplification loop promoting further mitochondrial damage [43
] was not seen in the bacteria-induced lysosomal pathway to apoptosis in neutrophils, where the LMP-generated loss in m occurred without the activity of caspases. Furthermore, other investigators studying TNF--induced apoptosis in tumor cells have reported that cathepsin B activity was indispensable for the final phase of execution and exposure of the scavenger marker PS [11
]. This digestive power of lysosomal proteases has also been observed in rat hepatocytes undergoing bile, salt-induced apoptosis [44
]. However, we did not see this effect in human neutrophils but instead, found that cysteine-cathepsins played a more direct role in the intrinsic, apoptotic pathway via the specific targeting of the Bcl-2 family protein Bid and induction of mitochondrial damage.
We have demonstrated previously that the adherent, Type 1-fimbriated E. coli and the serum-opsonized fimbrial mutant induced marked production of NADPH oxidase-dependent ROS in neutrophils within the first hour of phagocyte-bacteria interaction [4 ]. This activation was correlated here with early lysosomal damage in the neutrophils exposed to the fimbriated bacteria. ROS showed a selective, deleterious effect on lysosomes, as indicated by the observation that the cysteine-cathepsin inhibitor EST, which did not influence the level of LMP, was able to prevent mitochondrial damage in the presence of ROS. Bidere et al. [14 ] investigated staurosporine-treated T lymphocytes and noted a correlation between limited lysosomal destabilization and increased apoptosis. Those authors found that the staurosporine-induced release of cathepsins B, L, and D from lysosomes was not accompanied by liberation of the 250-kDa protein NAG or FITC-dextran molecules with a molecular mass greater than 70 kDa. Therefore, the induced LMP in neutrophils can be regarded as more pronounced, as the activity of NAG in digitonin-extracted cytosols was increased significantly in Type 1 fimbria-stimulated cells. Exposure of cells to H2O2 to mimic oxidative stress-induced apoptosis [10 , 13 , 22 23 24 25 ] has shown that a quantitative relationship exists in which a low level of stress with limited LMP leads to apoptosis, and a high level of stress with complete lysosomal rupture results in cellular necrosis. Using a system exhibiting strong endogenous ROS production without cell necrosis has allowed us to show in a more physiological way that oxidative stress is an important mediator of LMP and apoptosis.
Cathepsins B, L, and D (i.e., cysteine and aspartic cathepsins) are released simultaneously during oxidative stress and in response to other stimuli [14
, 23
]. Roberg and colleagues [12
] noted that lysosomal translocation of cathepsin D in cultured fibroblasts led to decreases in the m and the release of cytochrome c and apoptosis, and in another study [9
], it was observed that cathepsin B was more active in hepatocytes. Selective sensitivity to different cathepsins may be a result of the level of expression of these proteins but also to the availability and expression of other proteins important for the apoptosis machinery. Krajewska et al. [45
] found that the expression of the Bcl-2 family member Bid varied widely in nontumor tissues and different tumor cell lines, and they reported that the level of Bid expression correlated with the susceptibility to apoptosis in a tumor cell line model. Moreover, neither Bid alone nor lysosomal extracts in the absence of Bid had any effect on cytochrome c release, unless Bid was cleaved proteolytically first by exposure to lysosomal extracts, which indicates that Bid is a substrate for lysosomal enzymes [15
]. Accordingly, the level of Bid expression may be an important factor in determining which group of lysosomal proteases and/or pathways will be active during LMP-dependent apoptosis. We observed that Bid cleavage and a decrease in m were dependent on cysteine-cathepsins in human neutrophils and that this route of apoptosis could not occur via alternative pathways during ROS-dependent LMP. The cathepsin D inhibitor pepstatin A had no effect on apoptosis and no significant impact on Bid cleavage or mitochondrial damage, which suggests that cathepsin D does not play an important role in bacteria-induced apoptosis in human neutrophils.
Kobayashi et al. [8 ] have shown that phagocytosis of opsonized latex beads triggers ROS-related expression of apoptosis regulatory genes and induction of apoptosis in human neutrophils. However, although intracellular ROS production triggered by adherent, Type 1-fimbriated bacteria was a prerequisite for early LMP and apoptosis, we found that the phagosomal production of ROS during complement-mediated phagocytosis did not lead to lysosomal damage or apoptosis. Similar results were obtained with opsonized Salmonella bacteria (ref. [46 ] and R. Blomgran et al., unpublished data). This might, in part, be explained by the following: the confinement of ROS to phagolysosomal vacuoles, phagosomal fusion rather than lysosomal damage, and phagocytosis of opsonized bacteria also triggering survival signals through MAPK/ERK- and PI-3K/Akt-dependent pathways [6 , 46 ]. Derouet et al. [47 ] found that in human neutrophils treated with GM-CSF, the enhanced stability of Mcl-1 and delayed apoptosis were related to activation of PI-3K/Akt and ERK, whereas Kuo et al. [48 ] showed that only PI-3K/Akt-dependent pathways were required for the rapid up-regulation of Mcl-1 in IL-6-treated hepatocytes. PI-3K/Akt could also induce serine phosphorylation of Bax, thereby promoting the heterodimerization of Bax with Mcl-1 [20 ]. Besides being able to regulate the functions of proteins involved in preserving the integrity of the mitochondrial membrane, Akt mediates phosphorylation of caspase-9, which also provides an important, antiapoptotic target [49 ]. In a previous investigation [46 ], we noted that complement-opsonized bacteria triggered activation of the serine-threonine kinase Akt and that when Akt was silenced by antisense, it no longer protected macophages from undergoing phagocytosis-induced cell death. It remains to be determined whether Akt confers the same protection against such death in human neutrophils.
Zhao and colleagues [17
] have proposed that lysosome-dependent apoptosis involves a positive feed-back mechanism between LMP and mitochondrial damage, in which early lysosomal rupture causes mitochondrial damage, resulting in leakage of proteins that can, in turn, increase lysosomal damage and subsequent apoptosis. It has also been suggested that in TNF-stimulated NIH3T3 cells, mitochondrial depolarization and production of mitochondrial ROS constitute the initial step for lysosomal release of cysteine-cathepsins and subsequent apoptosis [50
]. In human neutrophils, we found that the early and extensive LMP, triggered by the fimbriated bacteria, was sufficient to initiate the intrinsic apoptotic cascade. However, the LMP, observed after 20 h of spontaneous apoptosis in the absence of NADPH oxidase-generated ROS, might represent delayed permeabilization, in which mitochondrial proteins or other molecules have targeted the lysosomes.
These findings suggest a close correlation between the involvement of ROS in the initiation of LMP and the release of proteases, which give rise to a cysteine-cathepsin-dependent cleavage of the Bcl-2 family protein Bid, subsequent mitochondrial damage, and apoptosis (Fig. 8 ). During immune receptor-mediated phagocytosis of bacteria, the formation of a phagolysosome and compartmentalization of ROS could protect the cell from ROS-inflicted damage and apoptosis. However, during activation of the phagocytes by certain stimuli, intracellular ROS, not confounded in the phagosome, can trigger apoptosis by targeting the lysosomal granules. This may represent a strategy by which certain pathogens evade the innate immune response.
 View larger version (22K): [in a new window] |
Figure 8. Model of the different effects of phagosomal and nonphagosomal ROS production on apoptosis in human neutrophils. 1. Nonphagosomal ROS triggers LMP and release of cysteine-cathepsins, which induce Bid cleavage into t-Bid and mitochondrial membrane permeabilization, leading to release of cytochrome c, caspase activation, and apoptosis. In parallel, Mcl-1 protein levels are decreased, and Bax is liberated, all of which allows t-Bid to activate Bax. 2. Receptor-mediated phagocytosis and phagosomal ROS production do not lead to lysosomal or mitochondrial damage and subsequent apoptosis. In parallel, during phagocytosis, Akt is activated and inhibits apoptosis by increasing Mcl-1:Bax interaction and decreasing caspase activation.
|
Received May 27, 2006; revised December 20, 2006; accepted December 27, 2006.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-mediated hepatocyte apoptosis by promoting mitochondrial release of cytochrome c J. Clin. Invest. 106,1127-1137[Medline]-mediated regulation of apoptosis of human neutrophils via caspase-3 Blood 93,3106-3115This article has been cited by other articles:
 |
|
 |
|
  S. Anguissola, B. Kohler, R. O'Byrne, H. Dussmann, M. D. Cannon, F. E. Murray, C. G. Concannon, M. Rehm, D. Kogel, and J. H. M. Prehn Bid and Calpains Cooperate to Trigger Oxaliplatin-Induced Apoptosis of Cervical Carcinoma HeLa Cells Mol. Pharmacol., November 1, 2009; 76(5): 998 - 1010. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Turk and V. Turk Lysosomes as "Suicide Bags" in Cell Death: Myth or Reality? J. Biol. Chem., August 14, 2009; 284(33): 21783 - 21787. [Full Text] [PDF]
|
|
|
|
|
|
H. Autefage, V. Albinet, V. Garcia, H. Berges, M.-L. Nicolau, N. Therville, M.-F. Altie, C. Caillaud, T. Levade, and N. Andrieu-Abadie Lysosomal Serine Protease CLN2 Regulates Tumor Necrosis Factor-{alpha}-mediated Apoptosis in a Bid-dependent Manner J. Biol. Chem., April 24, 2009; 284(17): 11507 - 11516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Q. Remijsen, T. Vanden Berghe, E. Parthoens, B. Asselbergh, P. Vandenabeele, and J. Willems Inhibition of spontaneous neutrophil apoptosis by parabutoporin acts independently of NADPH oxidase inhibition but by lipid raft-dependent stimulation of Akt J. Leukoc. Biol., March 1, 2009; 85(3): 497 - 507. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Droga-Mazovec, L. Bojic, A. Petelin, S. Ivanova, R. Romih, U. Repnik, G. S. Salvesen, V. Stoka, V. Turk, and B. Turk Cysteine Cathepsins Trigger Caspase-dependent Cell Death through Cleavage of Bid and Antiapoptotic Bcl-2 Homologues J. Biol. Chem., July 4, 2008; 283(27): 19140 - 19150. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. J. Hwang, S.-J. Lee, T.-Y. Kim, J.-H. Cho, and J.-Y. Koh Zinc and 4-Hydroxy-2-Nonenal Mediate Lysosomal Membrane Permeabilization Induced by H2O2 in Cultured Hippocampal Neurons J. Neurosci., March 19, 2008; 28(12): 3114 - 3122. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Conus, R. Perozzo, T. Reinheckel, C. Peters, L. Scapozza, S. Yousefi, and H.-U. Simon Caspase-8 is activated by cathepsin D initiating neutrophil apoptosis during the resolution of inflammation J. Exp. Med., March 17, 2008; 205(3): 685 - 698. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Prince, S. M. Bianchi, K. M. Vaughan, M. A. Bewley, H. M. Marriott, S. R. Walmsley, G. W. Taylor, D. J. Buttle, I. Sabroe, D. H. Dockrell, et al. Subversion of a Lysosomal Pathway Regulating Neutrophil Apoptosis by a Major Bacterial Toxin, Pyocyanin J. Immunol., March 1, 2008; 180(5): 3502 - 3511. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. M. Cooper, S. Radom-Aizik, C. Schwindt, and F. Zaldivar Jr. Dangerous exercise: lessons learned from dysregulated inflammatory responses to physical activity J Appl Physiol, August 1, 2007; 103(2): 700 - 709. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. E. Steinberg and S. Grinstein Unconventional Roles of the NADPH Oxidase: Signaling, Ion Homeostasis, and Cell Death Sci. Signal., March 27, 2007; 2007(379): pe11 - pe11. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
