American Journal of Respiratory Cell and Molecular Biology

Neutrophil apoptosis is essential for the resolution of inflammation but is delayed by several inflammatory mediators. In such terminally differentiated cells it has been uncertain whether these agents can inhibit apoptosis through transcriptional regulation of anti-death (Bcl-XL, Mcl-1, Bcl2A1) or BH3-only (Bim, Bid, Puma) Bcl2-family proteins. We report that granulocyte/macrophage colony–stimulating factor (GM-CSF) and tumor necrosis factor (TNF)-α prevent the normal time-dependent loss of Mcl-1 and Bcl2A1 in neutrophils, and we demonstrate that they cause an NF-κB-dependent increase in Bcl-XL transcription/translation. We show that GM-CSF and TNF-α increase and/or maintain mRNA levels for the pro-apoptotic BH3-only protein Bid and that GM-CSF has a similar NF-κB-dependent effect on Bim transcription and BimEL expression. The in-vivo relevance of these findings was indicated by demonstrating that GM-CSF is the dominant neutrophil survival factor in lung lavage from patients with ventilator-associated pneumonia, confirming an increase in lung neutrophil Bim mRNA. Finally GM-CSF caused mitochondrial location of Bim and a switch in phenotype to a cell that displays accelerated caspase-9-dependent apoptosis. This study demonstrates the capacity of neutrophil survival agents to induce a paradoxical increase in the pro-apoptotic proteins Bid and Bim and suggests that this may function to facilitate rapid apoptosis at the termination of the inflammatory cycle.

Neutrophils are key effector cells of the innate immune response providing the first line of defense against invading microorganisms. However, persistent recruitment, activation, and aberrant survival of these cells at inflamed sites appears to be an important facilitator of neutrophil-mediated tissue damage (1). Neutrophils have a short circulating half-life (6–8 h) and die rapidly by apoptosis when aged in-vitro (2). This form of programmed cell death also occurs in-vivo (3). There is now a substantial body of data supporting the view that neutrophil apoptosis is vital for the safe clearance of senescent neutrophils under physiological conditions (4, 5). Importantly, several growth factors and cytokines, including granulocyte/macrophage colony–stimulating factor (GM-CSF) (6), granulocyte colony–stimulating factor (G-CSF) (7), IL-8 and tumor necrosis factor (TNF)-α (8) have been postulated to impede neutrophil apoptosis, thereby impairing the resolution of inflammation.

The mechanisms triggering neutrophil apoptosis are subject to debate and have been proposed to reflect oxidant-induced mitochondrial damage, a time-dependent critical decline in NF-κB activity, and changes in Bcl-2 family and cIAP protein expression (9). A number of survival factors are recognized to alter the expression, phosphorylation, activation state, cellular location, and binding partners of several Bcl-2-family members in neutrophils. The Bcl-2 family is a major regulator of mitochondrial integrity and mitochondria-initiated caspase activity (10). The family consists of both pro-apoptotic and anti-apoptotic members subdivided into three main classes defined by the homology within four conserved Bcl-2 domains (BH1–4) (10). The anti-apoptotic members (including Bcl-2, Bcl-XL, Bcl2A1, and Mcl-1) show sequence homology through BH1 to 4. The pro-apoptotic members are divided into Bax and Bak-like proteins and BH3-only proteins. The former group of proteins share homology in two or three BH domains. The BH3-only members (including Bim, Bid, and Puma) share homology in the BH3 region only.

Although the generality of the heterodimeric interactions that exist between the anti-apoptotic and pro-apoptotic Bcl-2 members has been well documented, the sequence of events leading to Bax/Bak activation remains controversial. Circulating mature human neutrophil lack Bcl-2, yet this cell expresses most other anti-apoptotic members, namely, Bcl-XL, Bcl2A1, and Mcl-1, as well as the pro-apoptotic proteins Bax, Bak, Bid, Bim, and Puma (1116). Previous reports have supported a dominant role for Bcl2A1 and Mcl-1 in regulating constitutive apoptosis in neutrophils. Hence, although neutrophils from mice deficient in Bcl2A1 develop normally, they exhibit enhanced rates of apoptosis when cultured in vitro (17). Likewise, neutrophils from Mcl-1−/− mice have a higher rate of spontaneous apoptosis compared with control mice; crucially, however, GM-CSF still has a profound survival effect in these cells (18). This suggests that the maintenance of Mcl-1 levels is unlikely to be the sole mechanism underlying the prosurvival effect of GM-CSF in neutrophils.

Bim is considered to be one of the most important BH3-only proteins in immune cells. Indeed, the life span of myeloid cells is substantially increased in Bim−/− mice, and blood neutrophil counts in these animals are increased by a factor of 2.5 (19). However, the mechanism of Bim activation is not entirely clear; whereas early reports suggested that inactive Bim is bound to the microtubule cytoskeleton and, once activated, translocates to the mitochondria to initiate apoptosis (20), it is now apparent that Bim is regulated at multiple levels. Therefore, certain death stimuli can up-regulate Bim transcription (21), phosphorylation (22), ubiquitylation, and proteasome-dependent degradation (23). The mechanism by which Bim regulates apoptosis is also uncertain, as some experimental data suggest that Bim can be up-regulated without inducing apoptosis (24), thus, it might be required but insufficient for cell death.

The specific aim of the current study was to characterize as fully as possible the effects of aging on the expression of the Bcl-2 family in human neutrophils and to determine the consequences of cytokine and growth factor–mediated survival-signaling on these proteins. We wanted to explore mechanisms aside from Mcl-1 that might explain the survival effects of GM-CSF and to reference such effects to inflammatory neutrophils in vivo. We report that GM-CSF and TNF-α cause a profound and agonist-selective alteration in the balance of pro-apoptotic and anti-apoptotic Bcl-2 family proteins in neutrophils, in particular, in maintaining the expression of Bcl-XL and enhancing the stability of Mcl-1 but also in increasing and maintaining the expression of certain pro-apoptotic proteins including Bim and Bid. Bim expression is increased in inflammatory neutrophils recovered from the lungs of patients with ventilator-associated pneumonia (VAP), with GM-CSF operating as an important survival factor. Moreover, in vitro GM-CSF appears to prime neutrophils for TNF-α–mediated killing. We speculate the latter effect may enable a rapid switch from a survival-prone to an apoptotic-prone phenotype when required, thus aiding the timely resolution of inflammation.

Isolation and Culture of Human Neutrophils

Human peripheral blood neutrophils were obtained from healthy nonmedicated adult donors as previously detailed (see the online supplement) (25). All human studies were approved by the local Research Ethics Committee (06/Q0108/281; 08/H0306/17). Neutrophils were cultured as previously detailed (6) in the presence and absence of GM-CSF (10 ng/ml) or TNF-α (10ng/ml) with and without prior incubation (30 min) with the NF-κB inhibitor BAY11–7082 (10 μM), PI3-kinase inhibitor LY294002 (10 μM), MEK inhibitor U0126 (10 μM), or JNK inhibitor SP600125 (20 μM).

To assess the susceptibility of GM-CSF and G-CSF primed neutrophils to subsequent TNF-α–mediated apoptosis, cells were cultured ± GM-CSF (10 ng/ml) or G-CSF (10 ng/ml) for 6 hours before incubation with TNF-α (10 ng/ml) for an additional 12 hours. Where used, the pan-caspase inhibitor z-VAD-fmk (30 μM), caspase-8 inhibitor IETD-CHO (3 μM), or caspase-9 inhibitor LEHD-CHO (3 μM), were added, as detailed, 30 minutes before TNF-α.

Assessment of Neutrophil Apoptosis

Neutrophils were harvested at the time points indicated, cyto-centrifuged, fixed in methanol, stained with May-Grünwald-Giemsa (Merck Ltd., Nottingham, UK), and analyzed as previously described (6) with the observer blinded to the experimental variables. Apoptosis was also assessed by flow cytometry using (1) FITC-labeled recombinant human Annexin-V(AnV)/propidium iodide (PI) staining (6) and (2) the fluorescent cationic dye JC-1.

Real-Time PCR

Total RNA (1 μg) isolated with TRI-reagent (Sigma, St. Louis, MO) was used to generate cDNA (high capacity cDNA kit, Applied Biosystems, Carlsbad, CA). Relative gene expression was determined by quantitative PCR (iCycler, Bio-Rad, Hertfordshire, UK) using Sybr-green master-mix (Sigma) and relevant primers from Qiagen (West Sussex, UK). Relative gene expression was determined by correcting cycle threshold for the target gene against five house-keeping genes (B2M, GAPDH, β-actin, YWHAZ, and UBC) using genNORM (http://medgen.ugent.be/∼jvdesomp/genorm). Relative gene expression (fold change) is expressed as 2-ΔΔCT.

Immunoblotting

Neutrophils were lysed at the time points indicated (0, 4, 8 or 12 hours) in a hypotonic buffer as previously detailed (6). Lysates were separated on 15% SDS-PAGE and electro-transferred to polyvinylidene fluoride (PVDF) membranes. The membranes were incubated overnight in PBS-Tw20–0.25% dried milk powder with polyclonal antibodies to Bcl-XL, Mcl-1 (Santa Cruz Biotechnology Inc., Santa Cruz, CA), Bim (Chemicon International, Watford, UK), Bid, PUMA, Bak (Cell signaling, Danvers, MA) or Bax (BD Pharmingen, Oxford, UK). Membranes were washed in PBS-Tw20 then incubated with peroxidase-conjugated secondary antibody (1:5000) in PBS-Tw20–0.25% dried milk powder for 1 hour. Detection was performed by chemiluminescence using an ECL-kit (Amersham Life Science, Buckinghamshire, UK).

Confocal Microscopy

Neutrophils were preincubated with MitoTracker (Molecular Probes, Invitrogen, Paisley, UK) 30 minutes before adherence to poly- l-lysine coated cover slips (15 min, room temperature). Cells were fixed (3.7% paraformaldehyde, 20 min, RT) then washed (HBSS 2 × 5 min) and permeabilized (RPMI1640 containing 10% goat sera, 0.1% TX-100, 10 mM glycine, and 10 mM HEPES [permeabilizing solution], 30 min at RT) before immune staining. Coverslips were incubated at 4°C overnight in permeabilizing solution with anti-Bim (1:250; Cell Signaling) or nonimmune IgG control then washed with permeabilizing solution and incubated with Alexa fluor 488 goat anti-rabbit IgG diluted in permeabilizing solution (1:200, 60 min, RT). Coverslips were mounted onto glass slides with Mowiol and 4'-6-Diamidino-2-phenylindole (DAPI) (Sigma, St. Louis, MO). Cells were photographed using an ultraviolet confocal microscope (TCS Leica, Milton Keynes, UK) and images captured using ImagePro Plus 4.1(Media Cybernetics, Bethesda, MD). The images shown in Figure 6 are composite photomicrographs created by stacking 10 images taken through the z-plan.

Isolation of Peripheral Blood and Broncho-Alveolar Lavage Neutrophils from Patients with VAP

Bronchoalveolar lavage fluid (BALF; 0.9% NaCl) was obtained from patients with VAP, placed on ice, and strained through sterile gauze. The BALF suspension was centrifuged (310 × g for 5 min at 4°C) and the cells resuspended at 50 × 106/ml in RoboSep buffer (StemCell Technologies, Vancouver, Canada). Peripheral venous blood was drawn simultaneously into sodium citrate–containing tubes; red cell sedimentation was achieved using 6% Hetastarch (StemCell Technologies, Vancouver, Canada), the upper layer centrifuged (310 × g for 5 min at 4°C). The supernatants were removed and stored at −80°C and the cell pellet washed and resuspended in RoboSep buffer at 50 × 106 cells/ml.

The RoboSep was used according to manufacturer's instructions and allowed simultaneous isolation of neutrophils from blood and BALF samples. This protocol generated peripheral blood neutrophils that were more than 99% pure and viable and BALF neutrophils that were more than 97% pure and more than 99% viable. Both cell suspensions were centrifuged (310 × g for 5 min at 4°C) and the pellets processed for total RNA. Neutrophils from healthy volunteers were processed in an identical manner to provide references for the qPCR data. This isolation procedure takes approximately 90 minutes compared with the 120 minutes for the standard method.

BALF Immuno-Deletion Apoptosis Assay

The BALF samples were preincubated with anti-G-CSF 2 μg/ml or GM-CSF 2 μg/ml antibody for 30 minutes at RT, then diluted 1:1 (vol/vol) with complete Iscove's Modified Dulbecco's Media (IMDM) medium containing 107 neutrophils/ml (final in well 5 × 106) for 18 hours and apoptosis was assessed using Annexin-V-FITC binding (BD pharmingen, Oxford, UK). (See online supplement for detailed Methods).

Measurement of BALF Inflammatory Mediators

BALF and serum concentrations of cytokines and growth factors were measured using commercially available ELISA kits (proinflammatory 9-plex human cytokine assay kit (Meso Scale Discovery, MD; Duo-ELISA kit, R&D Systems, Abingdon, UK). These data were corrected to μg total protein present in the BALF or serum sample. All samples were analyzed for endotoxin contamination (Charles River, Endosafe-IPT, Charles River Laboratories International, Wilmington, MA); all data points derived from BALF samples had an EU reading below 0.2/ml protein.

Statistical Analysis

All data represent the mean (± SEM) of (n) independent experiments unless otherwise stated. Differences between groups were assessed using one-way analysis of variance (ANOVA) and post hoc analysis with Tukey's multiple comparisons. P < 0.05 was considered significant.

Effects of GM-CSF and TNF-α on Neutrophil Survival

Assessment of apoptosis using An-V-FITC demonstrated that GM-CSF reduces the rate of spontaneous apoptosis in neutrophils from 61.0 ± 6.0% to 35.5 ± 1.8% (P < 0.01) at 18 hours (Figure 1A). TNF-α also induced neutrophil survival with 46.6 ± 3.9% (P < 0.05) apoptosis at this time point. Representative flow cytometry data from neutrophils incubated with GM-CSF or TNF-α for 18 hours are shown in Figure 1B. Assessment of mitochondrial membrane depolarization using JC-1 produced near identical data, with GM-CSF and TNF-α both able to preserve mitochondrial membrane potential compared with time-matched controls (Figure 1A). Morphological quantification of apoptotic and nonapoptotic neutrophils confirmed the survival responses to GM-CSF and TNF-α defined by An-V-FITC and JC-1 staining (Figure 1C).

Transcriptional and Translational Regulation of the Bcl-2 Family in Human Neutrophils

Using this system we conducted a comprehensive investigation of the transcriptional regulation of Bcl-2-family member genes in aged and survival factor–stimulated human neutrophils. The quality of total RNA was assessed throughout using NanoDrop (Thermo Fisher Scientific Ltd., Northumberland, UK) and separating the samples by agarose gel electrophoresis to check for degradation (see online supplement). We investigated the transcriptional regulation of eight members of the Bcl-2-family, the anti-apoptotic factors Mcl-1, Bcl-XL, Bcl2A1, the BH3-only members Bid, Bim and Puma (Figure 2A), and the pro-apoptotic members Bax and Bak (see Figure E1A in the online supplement).

The anti-apoptotic factor Mcl-1 has previously been studied and proposed to be a major regulator of neutrophil survival (26). Our data show that mRNA levels for Mcl-1 diminished markedly over time, reducing 10-fold by 12 hours when compared with freshly isolated cells. GM-CSF and TNF-α caused a minor reduction in Mcl-1 message at 2 hours but did not modify the subsequent time-dependent loss of mRNA. This contrasts to the ability of GM-CSF to stabilize Mcl-1 protein and prevent its degradation (Figure 2B, Figure E2).

The presence of Bcl-XL mRNA and protein in human neutrophils has been controversial (12). Our data indicate abundant and relatively stable levels of Bcl-XL mRNA (Figure 2A) and the ability of GM-CSF, but not TNF-α, to increase Bcl-XL mRNA expression beyond 6 hours (P < 0.01) (Figure 2A). Both survival factors increased the amount of Bcl-XL protein compared with time-matched controls at 4 to 12 hours (P < 0.01, n = 3) (Figure 2B, Figure E2). Bcl2A1 mRNA levels declined in a time-dependent fashion (e.g., by 49-fold at 18 h) and GM-CSF and TNF-α were both effective at preventing this decline (Figure 2A). We have been unable to detect Bcl2A1 at protein level using the available antibodies.

There is a consensus view that the BH3-only family is central to promoting mitochondrial-dependent apoptosis in the neutrophil. We examined Bid, Bim, and Puma, which are known to be expressed in neutrophils. We show that whereas levels of Bid and Bim mRNA both decline over time in untreated cells, GM-CSF caused a seemingly paradoxical rapid increase in mRNA abundance of both of these BH3-only members, which was maintained to 18 hours (Figure 2A), with TNF-α increasing Bid but not Bim mRNA expression (Figure 2A). Previous reports investigating the expression of Bid protein in the neutrophil have described a prominent 22 kD band on Western blotting, which diminishes with caspase-dependent apoptosis, resulting in a 15 kD Bid fragment (27, 28). Bid protein (22 kD) was clearly present in freshly isolated neutrophils, diminishing rapidly over time with the appearance of a corresponding 15 kD band; in contrast, in the presence of GM-CSF or TNF-α Bid expression was maintained throughout the time course (Figure 2B, Figure E3). Of the three major isoforms of Bim (BimEL 23 kD, BimL 15 kD, and BimS 12 kD) only the former two splice-variants were detected in neutrophil lysates (Figure 2B). BimL and BimEL increased over time with further increases observed in the presence of GM-CSF or TNF-α at 8 hours (both P < 0.05) (Figure 2B, Figure E3). mRNA levels for Puma were stable and unaffected by GM-CSF or TNF-α (Figure 2A) and we could detect no consistent differences in Puma protein expression either over time or with GM-CSF or TNF-α (Figure 2B, Figure E3).

The mRNA expression profile for Bax and Bak was stable throughout the same time course and only marginally affected by GM-CSF and TNF-α at very late times (Figure E1). The protein expression of Bax and Bak was also very stable and relatively unaffected by GM-CSF or TNF-α (Figure E1B, E2). These findings support the view that the major time-dependent loss in Mcl-1 and Bcl2A1 described above does not reflect nonspecific loss of mRNA occurring as a consequence of constitutive apoptosis.

cIAP-1, cIAP-2, XIAP, and survivin are all expressed in the neutrophil and are known to block caspase-dependent apoptosis (29). We investigated the expression of IAP mRNAs and protein over the same time course as above and identified only modest early increases in mRNA abundance of cIAP-1 and cIAP-2 with TNF-α treatment and preservation of cIAP expression with GM-CSF at later times (Figure E1A).

Neutrophil Isolation from Peripheral Blood and BALF from Patients with VAP

We considered that the most interesting aspect of our observations was the ability of GM-CSF to cause a seemingly paradoxical increase in Bim mRNA abundance and BimEL protein expression in neutrophils in vitro. Thus we wanted to define whether enhanced Bim expression could be observed in inflammatory neutrophils in vivo under conditions where GM-CSF may be responsible for aberrant neutrophil survival and to explore the potential physiological importance of this effect. We sought to examine Bim expression in highly pure lung neutrophils derived from patients with VAP, which we hypothesized would display GM-CSF and G-CSF-dependent neutrophil survival.

This is the first description of the successful isolation of high purity BALF neutrophils. The initial BALF samples (n = 8) contained 50 to 70% neutrophils and these were enhanced to more than 98% purity using the RoboSep isolation protocol (Figure 3A). An identical protocol was followed to isolate neutrophils from the peripheral blood from these patients and from healthy volunteers to compare gene expression.

Determination of Inflammatory Mediators in the BALF of Patients with VAP

We measured the expression of proinflammatory mediators present in BALF from VAP patients and age-matched controls undergoing bronchoscopy for investigation of cough. This revealed significantly elevated concentrations of GM-CSF (0.037 pg/ml/μg protein; P < 0.05), G-CSF (0.40 pg/ml/μg protein; P < 0.05), IL-6 (0.68 pg/ml/μg protein; P < 0.005), IL-8 (7.19 pg/ml/μg protein; P < 0.005) and TNF-α (0.21 pg/ml/μg protein; P < 0.005) in the BALF from VAP subjects compared with control subjects (Figure 3B). The mean values for control-subject concentrations of IL-6 and IL-8 were 0.030 pg/ml/μg protein and 0.55 pg/ml/μg protein, respectively. GM-CSF, G-CSF, and TNF-α were undetectable in the BALF from control subjects.

BALF Neutrophils from Patients with VAP Express High Concentrations of Bim mRNA

We analyzed mRNA expression of the Bcl-2-family members of interest in neutrophils isolated from BALF and peripheral blood of VAP patients and compared these to healthy cells from control subjects. Bim mRNA expression was significantly higher in neutrophils isolated from the blood of patients with VAP compared with healthy control subjects with a further increase in Bim expression observed in the inflammatory BALF neutrophils (Figure 3C). Furthermore, Bcl-XL mRNA was also increased in a similar pattern by 2.3-fold and 13-fold in blood and BALF, respectively, when compared with healthy control subjects (Figure 3C). Of interest, Mcl-1 and PUMA mRNA expression was significantly lower in VAP patient neutrophil samples compared with samples from healthy control subjects (Figure 3C). The remaining Bcl-2-family members (Bax, Bak, Bid, Bcl2A1) were unchanged when compared with healthy control subjects (Figure 3, Figure E4). c-IAP2 mRNA expression was also higher (8.3-fold) in the BALF neutrophils from VAP patients compared with control subjects, although no changes were seen in c-IAP1, XIAP, or survivin expression when compared with healthy control subjects (Figure E4).

Effect of Immuno-Depletion of GM-CSF and G-CSF in BALF on In Vitro Neutrophil Apoptosis

Previous reports have identified a potential dual role for BALF G-CSF and GM-CSF in delaying neutrophil apoptosis (7). We used a similar approach to determine the relative importance of these growth factors in the survival effect of BALF from VAP patients using isolated neutrophils in vitro. The concentration and specificity of the neutralization antibody were predetermined using a checkerboard analysis incorporating GM-CSF (10 ng/ml), G-CSF (10 ng/ml) and varying concentrations of neutralizing or isotype-matched control antibodies (Figure 4A, Figure E5). After active immuno-depletion of GM-CSF and/or G-CSF or vehicle only control, the BALF supernatants from eight VAP patients were incubated with healthy donor neutrophils for 18 hours and the percentage of apoptosis determined. Figure E6 depicts the individual patient data and Figure 4 depicts the combined patient data. Anti-GM-CSF reversed the neutrophil survival effect of the BALF by 48.4% (range 20–75%; P < 0.005), anti-G-CSF by 49.4% (range 32.5–68.3%; P < 0.005) and combination by 84% (range 62.9–98.9%; P < 0.005) (Figure 4B, Figure E5). These percentage values were calculated using the apoptosis rates measured in the absence of added BALF (Figure 4).

These data indicate that BALF from patients with VAP contain significant amounts of active GM-CSF and G-CSF, both of which contribute toward the survival effect of this BALF on neutrophils and may be responsible for the elevated levels of Bim mRNA observed in the inflammatory BALF neutrophils. However, given that the expression of Bim, BclXL, and Bcl2A1 was found to be significantly higher in GM-CSF–stimulated cells compared with G-CSF–stimulated cells (P < 0.05, n = 3; Figure 5A) we predicted that GM-CSF was the main driver of elevated Bim levels in the inflammatory BALF neutrophils and proposed that neutrophils exposed to GM-CSF would undergo accelerated cell death after exposure to an independent death ligand.

GM-CSF Primes Neutrophils for TNF-α–Mediated Apoptosis

To test the hypothesis that inflammatory neutrophils are “primed to die” secondary to GM-CSF–mediated increases in Bim expression, healthy cells were incubated with GM-CSF for 6 hours followed by TNF-α incubation for 12 hours. Under these conditions, TNF-α reverted from being a prosurvival agent to one capable of inducing apoptosis, an effect confirmed to be caspase-dependent. This effect was not seen with G-CSF (Figure 5B). Figure 5C also demonstrates that a similar uplift in apoptosis was observed when neutrophils were cocultured with GM-CSF and TNF-α from the outset and indicates the ability of these agents alone or together to elevate Bid and Bim protein expression (Figure 5C) or mRNA (data not shown). Further evidence supporting the establishment of a new Bim-dependent apoptosis pathway in GM-CSF–treated neutrophils was obtained using carefully titrated concentrations of caspase-8 and caspase-9 inhibitors, which identified TNF-α–stimulated apoptosis to be predominantly caspase-9 (mitonchondrial)–dependent in GM-CSF–primed neutrophils, compared with a predominantly caspase-8–dependent mechanism in cells stimulated with TNF-α alone (Figure 5D) (7, 30).

Role of PI3-Kinase, MAPK, and NF-κB in the Transcriptional Regulation of Bcl-2-Like Genes

Neutrophil apoptosis is controlled by a complex network of signaling pathways. We have previously published data highlighting the importance of the PI3-kinase, MAPK, and NF-κB signaling pathways in GM-CSF and TNF-α–stimulated survival (8). Therefore, we used previously optimized concentrations of inhibitors of these pathways to dissect their importance in the transcriptional regulation of the Bcl-2 family. The expression of Bcl-XL has been previously shown to be regulated by NF-κB (31). Inhibition of this signaling pathway using BAY 11–7082 confirmed this finding in neutrophils (Figure E7). However, this is the first report to describe the role of NF-κB in regulating the transcription of BH3-only Bcl-2-family members. We report that inhibition of NF-κB signaling reduced the basal expression of Bid in aged neutrophils by 4.3-fold at 6 hours (P < 0.05) (Figure E7) and that the TNF-α–stimulated increase in the expression of Bid is NF-κB–dependent (P < 0.005) (Figure E7). The increase in Bim mRNA expression observed following GM-CSF stimulation, which is the focus of this article, was also significantly attenuated by inhibiting NF-κB signaling (Figure E7). Puma and Mcl-1 expression were unaffected by NF-κB inhibition (Figure E7).

Activation of the PI3-kinase/Akt pathway has been shown to repress Bim expression through phosphorylation of the transcription factor FOXO, removing the factor from its consensus binding site (32). We predicted and observed that inhibition of the PI3-kinase pathway would augment Bim expression above that seen with GM-CSF stimulation alone. We observed a fourfold increase in Bim mRNA in cells preincubated with LY294002 and stimulated with GM-CSF compared with GM-CSF alone (Figure E8). Inhibition of the PI3-kinase pathway did not alter the basal expression or survival factor–modulated expression of Mcl-1, Bcl-XL, Bcl2A1, Bax, Bak, Bid, or Puma (Figure E8).

The MAPK inhibitor U0126 and JNK inhibitor SP600125 had no effect on the time-dependent or survival factor–mediated changes in mRNA expression of any of the Bcl-2 family genes under study (Figure E9).

Fluorescent Distribution of Bim

Finally, confocal microscopy demonstrated diffuse and largely cytoplasmic staining of Bim in control neutrophils and only minimal colocalization with Mitotracker (Figure 6). In contrast, GM-CSF stimulation produced pyknotic Bim staining with a large degree of colocalization with Mitotracker, which was more evident following the addition of TNF-α. These data suggest that GM-CSF, in addition to its effects on Bim expression, initiates the redistribution of Bim from the cytosol to the mitochondria.

Apoptosis has been proposed as one of the cardinal determinants of whether neutrophilic inflammation resolves or persists. Hence, defining the mechanisms whereby this process is impeded by inflammatory cytokines or growth factors is essential. Although a number of studies have examined the presence or absence of individual components of the Bcl-2 family in human neutrophils, to our knowledge, this is the first comprehensive study of the transcriptional and translational regulation of pro-apoptotic and anti-apoptotic Bcl-2-family proteins following inflammatory cytokine and growth factor stimulation. We provide evidence that the survival factors GM-CSF and TNF-α induce a major shift in the balance of Bcl-2-family members at both mRNA and protein level, increasing the relative abundance of anti-apoptotic Bcl-2-members (principally Mcl-1, Bcl-XL, and potentially Bcl2A1), which appear to counteract, at least initially, the observed increase in the expression of the pro-apoptotic BH3-only members.

Interestingly, we found that GM-CSF and TNF-α reduce spontaneous apoptosis in neutrophils while increasing the expression of Bim (GM-CSF) and Bid (GM-CSF, TNF-α). Thus, in the presence of such survival, agonist levels of Bim and Bid were inversely correlated with the overall proportion of neutrophils undergoing apoptosis. We speculate that the enhanced expression of BclXL (GM-CSF), Bcl2A1 (GM-CSF, TNF-α) and the maintenance of Mcl-1 protein expression (GM-CSF, TNF-α) is able to compensate and dominate at early time points in setting the overall apoptotic threshold of the neutrophil. It appears that enhanced expression of Bim and Bid can be tolerated in these cells if matched by enhanced expression of anti-apoptotic Bcl-2 members. This conclusion is supported by recent data (33). We believe that these studies are particularly instructive as they map the outcome of physiologically relevant and fully integrated changes in Bcl-2 expression in cells lacking Bcl-2. Moreover, our studies in inflammatory neutrophils extracted from the lungs of patients with VAP indicate that GM-CSF is a major prosurvival agent in this context and that up-regulation of Bim is also observed in vivo. Pretreatment of neutrophils with GM-CSF in vitro generates a cell that has acquired susceptibility to TNF-α–mediated killing through a mitochondrial and caspase-9–dependent pathway. We propose that this reveals the true physiological function underlying the up-regulation of BH3-only proteins, namely, to accelerate neutrophil clearance once the initial beneficial function(s) of these cells are complete.

The concept of increased expression of pro-apoptotic Bcl-2–related proteins in priming neutrophils leading to more rapid apoptosis when inflammation resolves may have precedents in other systems. In the testes, for example, it is thought that male germ cells may be primed to undergo a wave of apoptosis during the first cycle of spermatogenesis by de novo expression of Bax and Bad in the period immediately leading up to this event (34).

The expression profile of the anti-apoptotic Bcl-2 family member Mcl-1 in myeloid cells has been well documented (35). Mcl-1 has a short half-life and is rapidly degraded by the ubiquitin-proteasome pathway in response to cytokine deprivation and other death stimuli. Agents that delay apoptosis have been shown to enhance the stability of Mcl-1 via signaling mechanisms, including PI3-kinase, GSK, and ERK (26, 3537). Our study supports the observation that GM-CSF maintains Mcl-1 levels within the neutrophil. Furthermore, we show an initial increase in Mcl-1 protein levels with TNF-α incubation followed by an identical rate of loss to that observed under control conditions. These data counter the view that the early pro-apoptotic effect of TNF-α on neutrophils is due to a caspase-dependent cleavage of Mcl-1 (38). We also report a rapid loss of Mcl-1 mRNA in aged neutrophils, which is not stabilized by GM-CSF or TNF-α stimulation. Although there is little doubt that Mcl-1 plays an important role in controlling apoptotic thresholds in the neutrophil (18), it is important to recognize that GM-CSF still acts as a powerful survival signal in Mcl-1–deficient neutrophils suggesting that GM-CSF can use alternative routes to prevent neutrophil apoptosis, as shown in our study.

Neutrophils also express Bcl-XL and Bcl2A1, two key anti-apoptotic Bcl-2-members. Enhanced neutrophil survival in septic rats has been associated with increased expression of Bcl-XL occurring via a C5a-dependent mechanism linked to PI3-kinase and MAPK signaling (14, 15). Likewise, the enhanced numbers of neutrophils seen in the pleural cavity in a carrageenan-induced animal model of pleurisy has been attributed to increased expression of Bcl-XL. (39). These reports suggest that Bcl-XL may also regulate neutrophil survival in many inflammatory conditions. We report that Bcl-XL is subject to considerable transcriptional regulation, with GM-CSF able to increase and maintain high Bcl-XL expression. Our data obtained for Bcl2A1 encompass mRNA expression only but show a surprising degree of transcriptional regulation. Hence, GM-CSF and TNF-α enhanced and maintained levels of mRNA for Bcl2A1, compared with rapidly diminishing levels within control cells. Together, these data support the involvement of three members of the anti-apoptotic Bcl-2-family, Mcl-1, Bcl-XL, and Bcl2A1 in promoting neutrophil survival in an inflamed environment.

Bim is one of the most effective BH3-only proteins for promoting cell death and analysis of Bim-deficient mice shows that Bim is a key determinant of the life span of neutrophils. We report a marked increase in the expression of Bim mRNA in neutrophils in response to GM-CSF followed by significant increases in BimEL and BimL. The levels of Bim are thus inversely correlated with the propensity of these cells to undergo apoptosis. These data follow closely the findings of Bauer and colleagues (40) who showed that LPS stimulation of mouse bone marrow neutrophils reduced spontaneous apoptosis but at the same time caused a clear increase in Bim expression. We speculate that the increased expression of Bim is countered by the increased expression of Bcl-XL, Bcl2A1, and the stabilization of Mcl-1. This supports the view offered by Andina and colleagues that neutrophil hamatopoietins initiate a pro-apoptotic counterregulation (33). It is therefore possible that neutrophils can tolerate the increase in Bim expression because of the corresponding increase in anti-apoptotic Bcl-2-members. The increase in Bim expression we report in neutrophils following stimulation with GM-CSF differs markedly to that seen in fibroblasts and epithelial cells where Bim is expressed de novo following the withdrawal of survival factors (22). Increases in Bim mRNA have been shown to result from inactivation of the Akt pathway leading to the activation of FOXO transcription factors (32) or by increased stabilization of the most abundant BimEL isoform following inactivation of the ERK1/2 pathway. Intriguingly, inhibition of PI3-kinase/Akt signaling augmented GM-CSF stimulation of Bim mRNA levels. This indicates that although the Akt/FOXO pathway operates in neutrophils, GM-CSF uses other pathways to induce Bim expression. Our results suggest for the first time that activation of NF-κB represents one such alternative pathway.

Bid is unique among BH3-only members in its ability to link the extrinsic caspase-dependent death pathway with the intrinsic mitochondrial-dependent pathway through a caspase-dependent cleavage of Bid generating truncated Bid (tBid) (41). tBid has been shown to translocate to the mitochondria to induce the oligomerisation of Bax and Bak and ultimately the release of cytochrome C (42). We report here that Bid is also under significant transcriptional regulation and that GM-CSF and TNF-α increase the expression of Bid maintaining its expression as control levels diminish with time. Inhibition of the NF-κB pathway reversed this effect. Very few data exist regarding NF-κB transcriptional regulation of Bid. Protein analysis showed maintenance of Bid expression following GM-CSF and TNF-α stimulation compared with rapidly diminishing values under control conditions. We were unable to detect tBid in neutrophil extracts; possibly due to the rapid turnover, protein instability, or low abundance in our samples. We are aware that the diminishing amount of Bid under control conditions may represent caspase-dependent cleavage of Bid during spontaneous apoptosis. However, the stabilization of Bid expression may be countered by a corresponding increase in expression of anti-apoptotic Bcl-2-members, enabling the neutrophil to tolerate otherwise increased levels of activator BH3-only proteins.

In conclusion, our results suggest the involvement of at least three members of the anti-apoptotic Bcl-2-family, Mcl-1, Bcl-XL, and Bcl2A1 in GM-CSF and TNF-α–mediated neutrophil survival. This occurs in the context of a seemingly paradoxical increase in Bid and Bim. The abundance of Bim appears to be finely tuned both at a transcriptional and post-translational modification level and may determine the cellular localization and activation state of the different Bim isoforms. The precise mechanism responsible for increasing Bim and Bid expression following GM-CSF stimulation remains uncertain, but our data suggest a prominent role for NF-κB activation. Up-regulation of Bim was also observed in inflammatory neutrophils exposed to GM-CSF in-vivo; this increase may serve to prepare the cell for rapid apoptosis and removal at the termination of the inflammatory cycle.

The authors are grateful to Dr. Doris Rassl for her help with the immuno-fluorescence work and Keith Burling (NIHR Cambridge BRC, Core Biochemical Assay Laboratory) for his help running the multiplex cytokine array.

1. Jeffery PK. Remodeling and inflammation of bronchi in asthma and chronic obstructive pulmonary disease. Proc Am Thorac Soc 2004;1:176–183.
2. Lee A, Whyte MK, Haslett C. Inhibition of apoptosis and prolongation of neutrophil functional longevity by inflammatory mediators. J Leukoc Biol 1993;54:283–288.
3. Renshaw SA, Loynes CA, Trushell DM, Elworthy S, Ingham PW, Whyte MK. A transgenic zebrafish model of neutrophilic inflammation. Blood 2006;108:3976–3978.
4. Martin C, Burdon PC, Bridger G, Gutierrez-Ramos JC, Williams TJ, Rankin SM. Chemokines acting via CXCR2 and CXCR4 control the release of neutrophils from the bone marrow and their return following senescence. Immunity 2003;19:583–593.
5. Perl M, Lomas-Neira J, Chung CS, Ayala A. Acute lung injury: neutrophils, epithelial cell death & inflammation: a unifying hypothesis. Mol Med 2008.
6. Cowburn AS, Cadwallader KA, Reed BJ, Farahi N, Chilvers ER. Role of PI3-kinase-dependent bad phosporylation and altered transcription in cytokine-mediated neutrophil survival. Blood 2002;100:2607–2616.
7. Matute-Bello G, Liles WC, Radella F, Steinberg KP, Ruzinski JT, Hudson LD, Martin TR. Modulation of neutrophil apoptosis by granulocyte colony-stimulating factor and granulocyte/macrophage colony-stimulating factor during the course of acute respiratory distress syndrome. Crit Care Med 2000;28:1–7.
8. Cowburn AS, Deighton J, Walmsley SR, Chilvers ER. The survival effect of TNF-α in human neutrophils is mediated via NF-κB-dependent IL-8 release. Eur J Immunol 2004;34:1733–1743.
9. Luo HR, Loison F. Constitutive neutrophil apoptosis: mechanisms and regulation. Am J Hematol 2008;83:288–295.
10. Cory S, Adams JM. The Bcl2 family: regulators of the cellular life-or-death switch. Nat Rev Cancer 2002;2:647–656.
11. Kobayashi SD, Voyich JM, Whitney AR, DeLeo FR. Spontaneous neutrophil apoptosis and regulation of cell survival by granulocyte macrophage-colony stimulating factor. J Leukoc Biol 2005;78:1408–1418.
12. Moulding DA, Akgul C, Derouet M, White MR, Edwards SW. BCL-2 family expression in human neutrophils during delayed and accelerated apoptosis. J Leukoc Biol 2001;70:783–792.
13. Ge Y, Yoshiie K, Kuribayashi F, Lin M, Rikihisa Y. Anaplasma phagocytophilum inhibits human neutrophil apoptosis via upregulation of Bfl-1, maintenance of mitochondrial membrane potential and prevention of caspase 3 activation. Cell Microbiol 2005;7:29–38.
14. Guo RF, Sun L, Gao H, Shi KX, Rittirsch D, Sarma VJ, Zetoune FS, Ward PA. In vivo regulation of neutrophil apoptosis by C5a during sepsis. J Leukoc Biol 2006;80:1575–1583.
15. Sano J, Oguma K, Kano R, Yazawa M, Tsujimoto H, Hasegawa A. High expression of Bcl-XL in delayed apoptosis of canine neutrophils induced by lipopolysaccharide. Res Vet Sci 2005;78:183–187.
16. Weinmann P, Gaehtgens P, Walzog B. Bcl-Xl- and Bax-alpha-mediated regulation of apoptosis of human neutrophils via caspase-3. Blood 1999;93:3106–3115.
17. Hamasaki A, Sendo F, Nakayama K, Ishida N, Negishi I, Nakayama K, Hatakeyama S. Accelerated neutrophil apoptosis in mice lacking A1-a, a subtype of the Bcl-2-related A1 gene. J Exp Med 1998;188:1985–1992.
18. Dzhagalov I, St John A, He YW. The antiapoptotic protein Mcl-1 is essential for the survival of neutrophils but not macrophages. Blood 2007;109:1620–1626.
19. Bouillet P, Metcalf D, Huang DC, Tarlinton DM, Kay TW, Kontgen F, Adams JM, Strasser A. Proapoptotic Bcl-2 relative bim required for certain apoptotic responses, leukocyte homeostasis, and to preclude autoimmunity. Science 1999;286:1735–1738.
20. Puthalakath H, Huang DC, O'Reilly LA, King SM, Strasser A. The proapoptotic activity of the Bcl-2 family member bim is regulated by interaction with the dynein motor complex. Mol Cell 1999;3:287–296.
21. Biswas SC, Shi Y, Sproul A, Greene LA. Pro-apoptotic Bim induction in response to nerve growth factor deprivation requires simultaneous activation of three different death signaling pathways. J Biol Chem 2007;282:29368–29374.
22. Ewings KE, Hadfield-Moorhouse K, Wiggins CM, Wickenden JA, Balmanno K, Gilley R, Degenhardt K, White E, Cook SJ. ERK1/2-dependent phosphorylation of BimEL promotes its rapid dissociation from Mcl-1 and Bcl-XL. EMBO J 2007;26:2856–2867.
23. Ley R, Balmanno K, Hadfield K, Weston C, Cook SJ. Activation of the ERK1/2 signaling pathway promotes phosphorylation and proteasome-dependent degradation of the BH3-only protein, Bim. J Biol Chem 2003;278:18811–18816.
24. Bouillet P, Purton JF, Godfrey DI, Zhang LC, Coultas L, Puthalakath H, Pellegrini M, Cory S, Adams JM, Strasser A. BH3-only Bcl-2 family member Bim is required for apoptosis of autoreactive thymocytes. Nature 2002;415:922–926.
25. Haslett C, Guthrie LA, Kopaniak MM, Johnston RB, Henson PM. Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacterial lipopolysaccharide. Am J Pathol 1985;119:101–110.
26. Edwards SW, Derouet M, Howse M, Moots RJ. Regulation of neutrophil apoptosis by Mcl-1. Biochem Soc Trans 2004;32:489–492.
27. Chou JJ, Li H, Salvesen GS, Yuan J, Wagner G. Solution structure of BID, an intracellular amplifier of apoptotic signaling. Cell 1999;96:615–624.
28. Gross A, Yin XM, Wang K, Wei MC, Jockel J, Milliman C, Erdjument-Bromage H, Tempst P, Korsmeyer SJ. Caspase cleaved BID targets mitochondria and is required for cytochrome c release, while BCL-XL prevents this release but not tumor necrosis factor-R1/Fas death. J Biol Chem 1999;274:1156–1163.
29. O'Neill AJ, Doyle BT, Molloy E, Watson C, Phelan D, Greenan MC, Fitzpatrick JM, Watson RW. Gene expression profile of inflammatory neutrophils: alterations in the inhibitors of apoptosis proteins during spontaneous and delayed apoptosis. Shock 2004;21:512–518.
30. Cowburn AS, White JF, Deighton J, Walmsley SR, Chilvers ER. Z-VAD-Fmk augmentation of TNF alpha-stimulated neutrophil apoptosis is compound specific and does not involve the generation of reactive oxygen species. Blood 2005;105:2970–2972.
31. Chen C, Edelstein LC, Gelinas C. The Rel/NF-κB family directly activates expression of the apoptosis inhibitor Bcl-x(L). Mol Cell Biol 2000;20:2687–2695.
32. Dijkers PF, Medema RH, Lammers JW, Koenderman L, Coffer PJ. Expression of the pro-apoptotic Bcl-2 family member bim is regulated by the forkhead transcription factor FKHR-L1. Curr Biol 2000;10:1201–1204.
33. Andina N, Conus S, Schneider EM, Fey MF, Simon HU. Induction of Bim limits cytokine-mediated prolonged survival of neutrophils. Cell Death Differ 2009;16:1248–1255.
34. Yan W, Suominen J, Samson M, Jegou B, Toppari J. Involvement of Bcl-2 family proteins in germ cell apoptosis during testicular development in the rat and pro-survival effect of stem cell factor on germ cells in vitro. Mol Cell Endocrinol 2000;165:115–129.
35. Moulding DA, Quayle JA, Hart CA, Edwards SW. Mcl-1 expression in human neutrophils: regulation by cytokines and correlation with cell survival. Blood 1998;92:2495–2502.
36. Leuenroth SJ, Grutkoski PS, Ayala A, Simms HH. Suppression of PMN apoptosis by hypoxia is dependent on Mcl-1 and MAPK activity. Surgery 2000;128:171–177.
37. Saffar AS, Dragon S, Ezzati P, Shan L, Gounni AS. Phosphatidylinositol 3-kinase and P38 mitogen-activated protein kinase regulate induction of Mcl-1 and survival in glucocorticoid-treated human neutrophils. J Allergy Clin Immunol 2008;121:492–498.
38. Cross A, Moots RJ, Edwards SW. The dual effects of TNFα on neutrophil apoptosis are mediated via differential effects on expression of Mcl-1 and Bfl-1. Blood 2008;111:878–884.
39. Sawatzky DA, Willoughby DA, Colville-Nash PR, Rossi AG. The involvement of the apoptosis-modulating proteins ERK 1/2, Bcl-XL and Bax in the resolution of acute inflammation in vivo. Am J Pathol 2006;168:33–41.
40. Bauer A, Kirschnek S, Hacker G. Inhibition of apoptosis can be accompanied by increased Bim levels in T lymphocytes and neutrophil granulocytes. Cell Death Differ 2007;14:1714–1716.
41. Li H, Zhu H, Xu CJ, Yuan J. Cleavage of BID by caspase 8 mediates the mitochondrial damage in the Fas pathway of apoptosis. Cell 1998;94:491–501.
42. Youle RJ, Strasser A. The BCL-2 protein family: opposing activities that mediate cell death. Nat Rev Mol Cell Biol 2008;9:47–59.
Correspondence and requests for reprints should be addressed to Andrew S. Cowburn, B.Sc., M.Sc., Ph.D., Respiratory Medicine Division, Department of Medicine, University of Cambridge School of Clinical Medicine, Level 5, Box 157, Addenbrooke's Hospital, CUHNHSFT, Hills Road, Cambridge, CB2 2QQ, United Kingdom. E-mail:

Related

No related items
American Journal of Respiratory Cell and Molecular Biology
44
6

Click to see any corrections or updates and to confirm this is the authentic version of record