Neutrophils die by apoptosis spontaneously within 12–24 h of their release from the bone marrow. The mechanism regulating entry of neutrophils into apoptosis at the end of their life-span is currently under debate. Our data suggest that neutrophil apoptosis involves a novel mechanism of caspase 8 activation that is indirectly regulated by accumulation of reactive oxygen species. We detected early activation of caspase 8 upstream of caspase 3 activation, suggesting death receptor signalling. The CD95 DISC (death-inducing signalling complex) was detected in neutrophils, but blocking antibodies to death receptors did not inhibit apoptosis, suggesting a novel mechanism for caspase 8 activation. Death receptor clustering in ceramide-rich lipid rafts is thought to be an early event in their signalling, so we investigated the role of ceramide generated by ASM (acid sphingomyelinase) in neutrophil apoptosis. Ceramide was generated early in neutrophil apoptosis, and ASM activity was required for neutrophil apoptosis. Moreover, neutrophil apoptosis was significantly delayed in ASM−/− mice compared with their wild-type littermates. CD95 DISC components were present in lipid rafts in neutrophils, and were progressively clustered in cultured neutrophils. Generation of ceramide was blocked by desferrioxamine, suggesting that hydroxyl radicals are important for the activation of ASM. This observation was in line with our earlier observation of a precipitous drop in reduced glutathione in the aging neutrophil.
Neutrophils enter apoptosis spontaneously within 12–24 h of their release from the bone marrow . The biological clock that determines the length of a neutrophil's life-span is not understood to date. We know that in most cell types, the apoptotic process is initiated through one of two pathways: the intrinsic pathway is regulated at the level of the mitochondrion, where lack of anti-apoptotic signals leads to loss of mitochondrial membrane integrity and release of pro-apoptotic factors. Among these is cytochrome c, which forms a complex termed the apoptosome with APAF (apoptotic protease-activating factor) and caspase 9 in the presence of ATP if released into the cytosol. This complex cleaves and activates caspase 9 . Caspase 9 then activates caspase 3, the central executioner of the apoptotic programme. On the other hand, the extrinsic pathway involves the active induction of apoptosis through ligation of members of the death receptor family. Clustering of the death receptors on the cell membrane during their interaction with their ligand leads to the formation of the DISC (death-inducing signalling complex). This complex comprises, in most cases, the death receptor, a linker molecule FADD (Fas-associated death domain) and procaspase 8. In the DISC complex, caspase 8 is activated and subsequently activates caspase 3 through both the mitochondrial, as well as the direct, route . Caspase 3 can be directly cleaved by caspase 8, but in cells with low levels of caspase 8 activation a second pathway has been shown to amplify the signal: caspase 8 cleaves Bid, a pro-apoptotic member of the bcl-2 family. Cleaved Bid has been shown to insert into the mitochondrial membrane, leading to release of cytochrome c, which then leads to an increase in apoptosome formation and additional caspase 3 activation .
Until recently, the entry into neutrophil spontaneous apoptosis was thought to be independent of death receptor signalling and regulated by the intrinsic pathway. Several groups have demonstrated that neutrophil spontaneous apoptosis is independent of the interaction between death receptors and their ligands [5,6]. The paradigm that death receptor signalling is not involved in neutrophil death became undermined when Daigle and Simon reported that the spontaneous apoptosis of neutrophils requires the activity of caspase 8 , an enzyme that is activated downstream of activation of death receptors. In the project presented here we propose a third pathway that combines features of both the intrinsic and the extrinsic pathway.
We detected early activation of caspase 8, before activation of caspase 3 was detectable, suggesting an activation of the extrinsic pathway . However, blocking of death receptor/ligand interaction, in line with the literature, did not affect neutrophil spontaneous apoptosis. We were therefore looking for an alternative route to caspase 8 activation. Previous work from our group has shown that, in T cells, components of the DISC complex can be found in lipid rafts, and that efficient Fas signalling is dependent upon intact lipid rafts . We investigated the role of lipid rafts in neutrophil spontaneous apoptosis and found that raft disruptors such as methyl-β-cyclodextrin and nystatin significantly delayed apoptosis. Fas localization to lipid rafts was detected both by sucrose gradients and immunofluorescence. Pre-formed DISC complexes were detected even in freshly isolated neutrophils, reminiscent of the pre-aggregation of DISC complexes detected in germinal centre B cells by Thierry De France and colleagues . Interestingly, both of these cell types are highly dependent upon continuous survival signals: in their absence, they spontaneously and rapidly enter apoptosis. Immunofluorescence co-staining of Fas and GM1 on fresh and cultured neutrophils revealed clustering of FAS containing lipid rafts in aged neutrophils. In the meantime, the groups of Gulbins and Kolesnick had shown that clustering of death receptors in lipid rafts is a prerequisite for efficient activation of the DISC complex. Clustering is driven by ceramide, a sphingolipid metabolite that is released from membrane lipids by ASM (acid sphingomyelinase) . We then extensively studied the roles of ASM and ceramide in neutrophil spontaneous apoptosis . Addition of exogenous ceramide, as well as addition of ASM, accelerated neutrophil apoptosis in a caspase 8-dependent manner. Inhibition of ASM activity, either by synthetic inhibitors or by genetic deletion in ASM−/− mice, delayed apoptosis in neutrophils. We conclude from these observations that ceramide plays an important role in the initiation of the death signal in neutrophils.
In the next set of experiments, we investigated the mechanisms of ASM activation in neutrophils . Qui et al.  described a potential activation of ASM by oxidation of cysteine residues in the C-terminal domain. We and others [12,13] had earlier made the observation that in neutrophils the intracellular redox balance changes over time. The levels of reduced glutathione drops early in neutrophil apoptosis. We had previously anticipated that these changes were a consequence of apoptosis, but was it also possible that they were the driving force behind the apoptotic process? Antioxidants, such as N-acetylcysteine as well as desferrioxamine, an iron chelator that inhibits hydroxyl radical generation, inhibit neutrophil spontaneous apoptosis . We tested if ceramide generation and clustering of Fas in lipid rafts in neutrophils were regulated by reactive oxygen species. Both Fas clustering and ceramide generation were diminished in neutrophils cultured in the presence of desferrioxamine.
Taking all these data together, we propose that in neutrophils the apoptotic programme is initiated by the activation of ASM by reactive oxygen species, leading to an increase of ceramide in the membrane (Figure 1). By changing local membrane fluidity, ceramide increases coalescence of lipid rafts . This leads to the activation of preformed DISCs, which are present constitutively in the lipid rafts of neutrophils. Caspase 8 is activated in these complexes, and then triggers the entry into the execution phase of apoptosis.
The proposed pathway of initiation of neutrophil spontaneous apoptosis by reactive oxygen species-dependent ceramide generation, and clustering of death receptors in lipid rafts
Because of the enormous noxious potential of the intracellular enzymes in the neutrophil, the tight regulation of neutrophil apoptosis is highly important for homoeostasis. It is therefore likely that there is more than one pathway involved. This would be well in line with the observation that neutrophil apoptosis is delayed, but not abolished, in ASM-deficient mice. The next step from here will be to investigate the mechanisms regulating the redox balance in neutrophils, and to study how anti-apoptotic cytokines can interfere with the ceramide-induced activation of death receptor signalling.
Lipids, Rafts and Traffic: A Focus Topic at BioScience2004, held at SECC Glasgow, U.K., 18–22 July 2004. Edited by G. Banting (Bristol, U.K.), N. Bulleid (Manchester, U.K.), C. Connolly (Dundee, U.K.), S. High (Manchester, U.K.) and K. Okkenhaug (Babraham Institute, Cambridge, U.K.)