CD134 (OX40) is a member of the tumor necrosis factor (TNF) receptor superfamily expressed on activated T cells. Here, we show that human peripheral blood neutrophils express CD134. Activationof CD134 by soluble CD134 ligand (OX40 ligand/gp34) resulted in delayed caspase-3 activation and consequently in delayed neutrophil apoptosis in vitro. Moreover, CD134 ligand, like G-CSF, maintained anti-apoptotic Mcl-1 levels and inhibited cleavage of the pro-apoptotic Bcl-2 family members Bid and Bax in these cells, suggesting that CD134-mediated signals block apoptosis pathways proximal to mitochondria activation. In conclusion, CD134 regulates neutrophil survival, suggesting that this molecule does not only contribute to adaptive but also to innate immune responses.
CD134 (OX40) is a 50-kDa cell surface protein expressed on activated CD4+ and CD8+ T cells 1–5. Molecular cloning of its cDNA revealed that CD134 is a type I transmembrane protein belonging to the tumor necrosis factor (TNF) receptor family 2–4. The ligand (L) of CD134 (CD134L/OX40L/gp34) has also been cloned and identified as a 34-kDa type II transmembrane protein belonging to the TNF superfamily 6, 7. CD134L is expressed on activated T cells, B cells, macrophages, dendritic cells, and endothelial cells 8. All studies carried out to understand CD134 function were performed in T cells, which are activated via this surface molecule. For instance, CD134 signals mediate differentiation processes that enhance memory T cell formation 9, reverse T cell tolerance 10, induce both Th1 and Th2 cytokines 11, and prolong T cell survival 12. Molecular CD134L/CD134 interactions have also been proposed to play important roles in anti-tumor immunity 13 and in the development of autoimmune diseases 14. Since only active, but not resting, T cells express CD134, it has been suggested that this molecule may be a suitable drug target in chronic T cell-mediated diseases 8.
Neutrophils are important players within the innate immune system. Their numbers are controlled by rates of generation and apoptosis. Apoptosis is the most common physiological cell death of neutrophils both in vitro and in vivo. Delayed neutrophil apoptosis has been associated with several infectious diseases 15. Besides granulocyte colony-stimulating factor (G-CSF) and granulocyte/macrophage CSF (GM-CSF) 16, we have recently identified macrophage migration inhibitory factor (MIF) as a survival factor 17, which may contribute to prolonged neutrophil survival at the site of inflammation. Potential intracellular mechanisms that block neutrophil apoptosis by the mentioned cytokines include the preventionof the following: Bid and Bax activation, mitochondrial cytochrome c release and the formation of active caspase-3 17, 18.
During immunophenotyping of patients with cystic fibrosis (CF) and control individuals, we realized that human blood neutrophils constitutively express CD134. Since it has been believed that cell activation is a requisite for CD134 expression in primary cells, we decided to investigate potential functions of this molecule in neutrophils. Here, we report that CD134 engagement results in delayed neutrophil apoptosis in vitro. Analyzing pro-apoptotic processes in neutrophils revealed that the mechanisms by which CD134 mediates its anti-apoptotic function are, at least partially, the same as those mediated via other cytokine receptors.
2.1 CD134 is constitutively expressed by neutrophils
We measured CD134 mRNA and protein levels in freshly isolated human blood neutrophils from normal donors. Neutrophils expressed CD134 mRNA (Fig. 1A). Freshly purified peripheral blood mononuclear cells (PBMC), in vitro activated PBMC, and Jurkat cells served as controls. As expected, activated but not resting PBMC expressed CD134 mRNA. Jurkat cells were CD134 mRNA negative. To determine whether the expression of CD134 mRNA correlates with the expression of CD134 protein, we performed flow cytometric and immunoblotting studies. Blood neutrophils and activated lymphocytes, isolated from patients with atopic dermatitis, but not blood monocytes or Jurkat cells demonstrated evidence for CD134 surface expression (Fig. 1B). Moreover, CD134 protein expression of neutrophils was confirmed by immunoblotting (Fig. 1C). CF is an inflammatory disease associated with high expression of GM-CSF and G-CSF 16, as well as MIF 17. CD134 protein expression did not differ between neutrophils obtained from CF patients and normal control individuals (Fig. 1D). We also did not see changes in CD134 expression following in vitro stimulation of neutrophils with GM-CSF or G-CSF up to 24 h (data not shown). To demonstrate CD134 expression in neutrophils under in vivo conditions, we analyzed neutrophils in tissue sections of patients with acute appendicitis by a double immunofluorescence technique. Infiltrating neutrophils were identified using an anti-CD15 mAb. All detected neutrophils expressed CD134, demonstrated by a ring-like staining pattern, consistent with its expression on the cell surface (Fig. 1E).
2.2 Agonistic stimulation of CD134 delays neutrophil apoptosis
Since CD134 delivers anti-apoptotic signals in T cells 12, we investigated whether CD134 stimulation delays apoptosis of neutrophils that spontaneously occurs following culturing these cells 15. Soluble CD134L (sCD134L) delayed neutrophil death in vitro in a dose-dependent manner (Fig. 2A). Optimal concentrations of sCD134L and G-CSF had similar anti-death potencies on neutrophils up to 30 h in cultures (Fig. 2B). At later time points, sCD134L demonstrated decreased efficacy compared to G-CSF (data not shown). sCD134L-mediated anti-death effects were not blocked in the presence of 15 μg/ml polymyxin B (data not shown), excluding any potential nonspecific effect via LPS. Similarly to G-CSF, neutrophils had to be continuously exposed to sCD134L to observe a significant survival effect after 25 h (data not shown). We next investigated whether the anti-death effect mediated by sCD134L was due to inhibition of apoptosis. Indeed, CD134 activation resulted in reduced redistribution of phosphatidylserine (PS), a characteristic feature of apoptotic neutrophils 19 (Fig. 2C, D).
2.3 Agonistic stimulation of CD134 inhibits caspase-3 activation
Caspase-3 is a critical effector caspase in neutrophil apoptosis 17, 18, 20. We investigated caspase-3 activation by immunoblotting and an enzymatic assay. Freshly isolated blood neutrophils expressed the 32-kDa proform of caspase-3 (Fig. 3A). Culturing the cells for 8 h resulted in the appearance of the active 17-kDa form and cleavage of caspase-3 was accelerated in anti-CD95 (Fas, Apo-1 21) antibody-treated neutrophils. In contrast, both sCD134L and G-CSF prevented the occurrence of the 17-kDa form, suggesting that both survival cytokines blocked caspase-3 processing at this time point. Furthermore, both sCD134L and G-CSF suppressed caspase-3-like DEVDase activity in neutrophils, whereas CD95 activation resulted in enhanced enzymatic activity (Fig. 3B). Thus, the presence of the 17-kDa fragment within neutrophil lysates correlated well with an increased enzymatic caspase activity, and both were blocked by sCD134L, furthermore suggesting that this member of the TNF superfamily represents a novel anti-apoptotic factor for neutrophils.
2.4 Agonistic stimulation of CD134 inhibits Bid and Bax cleavage
Further experiments were performed to elucidate the mechanisms of sCD134L-mediated inactivation of caspase-3 in neutrophils. The pro-apoptotic Bcl-2 family member Bid 22, 23 demonstrated a detectable 15-kDa cleavage product in neutrophils after short-term cultures (Fig. 4A). This cleavage product was not seen (4.5-h cultures) or at lower levels (6-h cultures) in sCD134L- or G-CSF-treated neutrophils. In these experiments, CD95-activated Jurkat cells served as positive controls. Culturing the cells for 8 h revealed reduced levels of full-length Bid in untreated and anti-CD95-treated neutrophils compared to neutrophils incubated with sCD134L and G-CSF (data not shown).
Bax, another pro-apoptotic member of the Bcl-2 family, is cleaved into an 18-kDa fragment during spontaneous neutrophil apoptosis, a process which has recently been shown to be calpain-1 mediated 18. Both sCD134L and G-CSF prevented Bax cleavage in 8-h cultures (Fig. 4B, upper panel). In 18-h cultures, the inhibitory effect of G-CSF on Bax cleavage appeared to be stronger compared to sCD134L. Nevertheless, sCD134L-treated neutrophils had markedly reduced truncated-Bax levels compared to untreated or anti-CD95-stimulated neutrophils (Fig. 4B, lower panel).
2.5 Agonistic stimulation of CD134 maintains Mcl-1 levels
Besides inhibition of pro-apoptotic members, the sCD134L-mediated delay of neutrophil apoptosis may also involve anti-apoptotic members of the Bcl-2 family. In agreement with previously published work, we observed Mcl-1 24 but no detectable Bcl-2 25 expression in freshly purified mature human blood neutrophils as assessed by immunoblotting (Fig. 5). Bcl-xL was detectable but required enrichment by immunoprecipitation before immunoblotting (data not shown and 18). In untreated or anti-CD95-stimulated neutrophils, Mcl-1 was no longer detectable in 8-h cultures. The proteolysis of Mcl-1 associated with spontaneous neutrophil apoptosis 24 was inhibited by both sCD134L and G-CSF (Fig. 5). Both anti-apoptotic factors were unable to induce detectable Bcl-2 expression in neutrophils.
CD134 has been regarded as a molecule associated with T cell activation 8–14. In agreement with this view, CD134+ T cells were observed under several inflammatory conditions, for instance in joint fluids derived from patients with rheumatoid arthritis 26, in gastrointestinal tissues of patients with immune-mediated intestinal diseases 27, as well as in tumor tissues 28. Therefore, CD134 has been suggested as an ideal target for the treatment of T cell-mediated diseases 8. The fact that also mature neutrophils express CD134, even in the absence of an inflammatory response, clearly changes our view in this respect. Targeting CD134 would mean also targeting neutrophils, and this may cause deficiencies in anti-bacterial and anti-fungal defense mechanisms, particularly when cell-depleting CD134-specific mAb or fusion proteins are to be administered. Anti-viral responses, however, did not appear to be affected when CD134/CD134L molecular interactions were blocked 29.
Similar to T cells, CD134 delivers anti-apoptotic signals in neutrophils. Our previous studies suggested that spontaneous neutrophil apoptosis involves activation of caspase-8 20 and calpain-1 18, which cleave the pro-apoptotic Bcl-2 family members Bid and Bax, respectively. We have also shown that these events lead to the release of cytochrome c and Smac from mitochondria; the latter appears to be essential for caspase-3 activation in neutrophils 18. We observed in this study that CD134L, similar to G-CSF, inhibits Bid and Bax cleavage and maintains Mcl-1 levels in cultured neutrophils, suggesting that both cytokines delay apoptosis by inhibiting the pro-apoptotic activation of mitochondria. This view is in agreement with previously published work, suggesting that G-CSF prevents the mitochondrial death pathway in neutrophils 30.
Cleavage of Bid in cultured neutrophils and its inhibition by caspase-8 inhibitors implies a role of caspase-8 in spontaneous neutrophil apoptosis 17, 18. How caspase-8 becomes activated is unclear, but CD95 has been excluded as a potential candidate for triggering this process 31. However, CD95 activation additionally activates caspase-8, which generates more truncated Bid, resulting in higher caspase-3 activity and accelerated apoptosis. CD134 stimulation did not block CD95-mediated neutrophil death and had no effect on c-Flip levels in these cells (data not shown).
How can CD134 on neutrophils be activated under in vivo conditions? The first CD134-mediated activation might occur when neutrophils migrate into inflamed tissues. Endothelial cells have been reported to express CD134L 32. Molecular CD134L/CD134 interactions have been demonstrated between endothelial cells and pathogenic T cells in the process of T cell recruitment 33. Following recruitment of neutrophils to the site of inflammation, cognate interactions with CD134L-expressing cells, such as activated T cells 6 or dendritic cells 34, may contribute to prolonged survival of neutrophils and accumulation of these cells at inflammatory sites. On the other hand, ligation of CD134L on dendritic cells induced the production of pro-inflammatory cytokines 34. Thus, perhaps, neutrophils stimulate dendritic cells via CD134, a scenario that would fit into the concept seeing neutrophils as important immunoregulatory cells 15. Clearly, additional studies are required to address the role of CD134 on neutrophils under in vivo conditions.
4 Materials and methods
Neutrophils and PBMC were isolated from peripheral blood of healthy donors by Ficoll-Hypaque centrifugation 35, 36. In some experiments, we purified neutrophils from CF patients. The resulting cell populations contained >95% neutrophils and PBMC, respectively. Written informed consent was obtained from all donors, and the study was approved by the local ethics committee. To obtain activated T cells, PBMC were stimulated with 10 μg/ml PHA for 5 h. Jurkat cells were purchased from ATCC (Rockville, MD).
Complete culture medium was RPMI 1640 (Life Technologies, Inc., Basel, Switzerland) supplemented with 2 mM L-glutamine, 200 IU/ml penicillin, 100 μg/ml streptomycin, and 10% fetal bovine serum (all from Life Technologies). Mouse anti-human CD134 mAb (BerAct35) was from Ancell Corporation (distributed by Alexis Biochemicals, Läufelfingen, Switzerland). Recombinant human sCD134L was purchased from Apotech (Apotech Corporation, Epalinges, Switzerland; batch 109-26; endotoxin level <0.04 ng/μg purified protein) and G-CSF from R&D Systems (distributed by Bühlmann AG, Basel, Switzerland). The agonistic anti-CD95 mAb CH11 was purchased from Beckman Coulter International S.A. (Nyon, Switzerland). Anti-CD15 mAb, anti-Mcl-1 mAb, and rabbit polyclonal anti-human caspase-3 and Bax Ab were from Becton Dickinson Biosciences (Basel, Switzerland). Goat anti-human Bid Ab was obtained from R&D Systems. Anti-human β-actin mAb was from Sigma (Buchs, Switzerland). Mouse and rabbit horseradish peroxidase (HRP)-conjugated secondary antibodies were from Amersham Pharmacia Biotech (Dübendorf, Switzerland). Anti-Bcl-2 mAb, control IgG1 mAb, and mouse anti-goat HRP-conjugated secondary Ab were from DAKO (Zug, Switzerland). Goat anti-mouse Ab and goat F(ab′)2 anti-mouse R-phycoerythrin (R-PE)-conjugated secondary Ab were purchased from Biosource International (Camarillo, CA). Tetramethylrhodamine isothiocyanate (TRITC)- and fluorescein (FITC) isothiocyanate-conjugated donkey anti-mouse secondary Ab were obtained from Jackson ImmunoResearch Laboratories (Milan Analytica, La Roche, Switzerland). Anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mAb was from Chemicon International, Inc. (Temecula, CA). Unless stated otherwise, all other reagents were from Sigma.
4.3 Reverse transcriptase PCR
CD134 mRNA expression was determined by reverse transcrriptase (RT)-PCR as previously described 17. Primer sequences were as followed: 5′-AAT AGC TCG GAC GCA ATC TGT-3′ and 5′-CAT GGC ATA CGT AAG CAG AGA G-3′. CD134 (524 bp) and β-actin (450 bp) PCR products were separated on 1% agarose gels and visualized by ethidium bromide staining.
4.4 Flow cytometry
Staining of anti-CD134 mAb on neutrophils was measured by single flow cytometry (FACSCalibur, Becton Dickinson). Isotype-matched control mAb were used as negative controls.
4.5 Confocal laser scanning microscopy
Immunofluorescence stainings were carried out on 5-μm-thick paraformaldehyde-fixed paraffin-embedded tissue sections from appendicitis patients as previously described 37. Briefly, slides were dried at 52°C for 2 h and deparaffinized using NeoClear solution (Merck, Darmstadt, Germany), ethanol (100%, 90%, 80%, 60%, and 40%), and water at room temperature. Followingmicrowave treatment in TE buffer (10 mM Tris, pH 8.0, 1 mM EDTA), slides were washed in water, blocked, and stained with anti-CD15 (1:20) and anti-CD134 Ab (1:50). Control Ab were used at the same concentrations in each experiment. Following incubation with primary Ab, tissues were incubated with appropriate TRITC- and FITC-conjugated secondary Ab (1:100) in the dark at room temperature for 1 h. The anti-fading agent Mowiol (Calbiochem, Juro Supply GmbH, Lucerne, Switzerland) was added. Slides were covered by coverslips and analyzed by confocal laser scanning microscopy (LSM 510, Carl Zeiss, Heidelberg, Germany) equipped with Ar and HeNe lasers.
4.6 Cell viability and apoptosis measurements
Neutrophils were cultured in complete culture medium in the presence and absence of sCD134L (0.75 μg/ml or indicated concentration), G-CSF (25 ng/ml), or anti-CD95 (1 μg/ml) for the indicated times. Neutrophil death was assessed by uptake of 1 μM ethidium bromide and flow cytometric analysis (FACSCalibur) 17, 18. To determine whether cell death was apoptosis, redistribution of phosphatidylserine in the absence of propidium iodide was measured 17, 18.
Immunoblotting using anti-Bid, anti-Bax, anti-caspase-3, anti-Mcl-1, anti-Bcl-2, anti-β-actin, and anti-GAPDH Ab was performed as previously described 17, 18, 37. In the CD134 immunoblots, cells were lysed in a buffer containing 10 mM Hepes pH 7.4, 142.5 mM KCl, 5 mM MgCl2, 1 mM EGTA, and 1.75% Triton X-100 supplemented with a protease inhibitor cocktail from Sigma. The filters were incubated with anti-CD134 mAb (1:500), followed by goat anti-mouse Ab (1:1,000) and HRP-conjugated mouse anti-goat Ab (1:2,000).
4.8 Enzymatic caspase-3 assay
Caspase-3-like activity was measured using a commercial kit (QuantiZyme caspase-3 cellular activity assay kit; Biomol, Plymouth Meeting, PA) according to the manufacturer's instructions and as previously described 17, 18.
4.9 Statistical analysis
Statistical analysis was performed by using the ANOVA test. If mean levels are presented, the standard errors of the mean (SEM), the numbers of independent experiments (n), and statistical differences (p values) are additionally indicated.
We greatly appreciate the technical support provided by I. Schmid (caspase-3 activity assay) and E. Kozlowski (confocal microscopy) as well as the help in cystic fibrosis blood sampling from Drs. C. Casaulta and M. H. Schöni (both Department of Pediatrics, Inselspital, Bern). This work was supported by grants from the Swiss National Science Foundation (grant no. 31-58916.99), OPO-Foundation (Zurich), Bonizzi-Theler-Foundation (Lucerne), and the Bernische Krebsliga (Bern).