Inhibition of TNF-α-Induced Neutrophil Apoptosis by Crystals of Calcium Pyrophosphate Dihydrate Is Mediated by the Extracellular Signal-Regulated Kinase and Phosphatidylinositol 3-Kinase/Akt Pathways Up-Stream of Caspase 3¹

The inflammatory disease known as acute pseudogout arises from the deposition of calcium pyrophosphate dihydrate (CPPD)³ crystals in the synovial joints of humans (1). Both monoclinic and triclinic crystalline forms of CPPD are able to activate neutrophils in the joint, initiating the acute inflammatory responses such as the release of reactive oxygen species and degranulation, which are thought to be key factors in the pathophysiology of the disease (1, 2). In vitro, both uncoated and plasma- or synovial fluid-coated crystals have been shown to induce neutrophil activation, including superoxide anion production and lysosomal enzyme release (2, 3, 4).

During the resolution of inflammation, effete neutrophils are removed from inflammatory sites by a process of programmed cell death known as apoptosis, leading to their recognition and phagocytosis by macrophages (5). Any significant delay in neutrophil apoptosis can lead to excessive accumulation and damage to surrounding tissue (6). Although apoptosis is an intrinsic process, numerous inflammatory mediators such as LPS, GM-CSF, fMLP, or immune complexes (7, 8), and a variety of cytokines such as IL-1 and IL-6 (9, 10), are able to delay apoptosis. Although TNF-α has been shown to induce extensive apoptosis in neutrophils within 3 h (11, 12, 13), this cytokine may also inhibit apoptosis in cells via the early activation of the proinflammatory transcription factor NF-κB/rel (NF-κB), c-Jun N-terminal kinase (JNK), or p38 mitogen-activated protein (MAP) kinase (14, 15, 16). Early TNF-α-induced transient stimulation of NF-κB, JNK, and p38 is generally mediated independently of caspase activity (17).

Intracellular mediators of neutrophil apoptosis are currently under investigation. The stimulation of the stress-activated kinases concomitant with the inhibition of the extracellular signal-regulated kinase 1 (ERK1)/ERK2 MAP kinase pathway has been observed in a number of cell systems undergoing apoptosis (17, 18). ERK1 and ERK2 have also been shown to be directly involved in negatively regulating apoptosis (17, 19). Work done with HeLa cells demonstrated that the inhibition of ERK1 and ERK2 is implicated in the activation of the JNK and p38 pathways, resulting in apoptosis (17). Furthermore, Nolan et al. (19) showed that LPS-induced activation of ERK inhibited neutrophil apoptosis. Protein kinase B (Akt) has been reported to negatively regulate apoptosis in many cell systems (20). Akt is activated in part by the binding of the phosphatidylinositol 3,4,5-trisphosphate or phosphatidylinositol 3,4-bisphosphate to its amino-terminal pleckstrin homology domain, and by direct phosphorylation by the phosphatidylinositol-dependent kinase-1 (PDK1); PDK1 is also activated by these phospholipids. Thus, Akt and PDK1 are regulated by phosphatidylinositol 3-kinase (PI3-K), which phosphorylates the inositol rings of phosphatidylinositol 3,4-bisphosphate and phosphatidylinositol 3,4,5-trisphosphate at the 3′ position (21, 22). Akt is a direct effector of PI3-K-associated cell survival in vivo (20, 22, 23). Therefore, the ERK1/ERK2 and PI3-K→PDK1→Akt pathways appear to be involved in the regulation of the apoptosis signaling cascade.

We have recently demonstrated that opsonized crystals of monosodium urate monophosphate (MSUM) and CPPD inhibit TNF-α-induced and spontaneous apoptosis in human neutrophils (24), and that repression of apoptosis by these microcrystrals is mediated through inhibition of the proapoptotic cysteine protease, caspase 3. Akahoshi et al. (25) monitored DNA laddering on agarose gels and showed that low concentrations of nonopsonized MSUM crystals may inhibit apoptosis after extended incubations. However, in the same study, higher concentrations of unopsonized crystals caused cell lysis, as described by others (26, 27). The regulation of apoptosis in neutrophils is poorly understood and, to the best of our knowledge, there have been no reports on the mechanisms of crystal-associated repression of TNF-α-induced or spontaneous apoptosis. Because CPPD crystals have been shown to induce ERK1/ERK2 and Akt activation, while repressing TNF-α associated and spontaneous apoptosis, the objective of this work was to investigate the kinetics of CPPD crystal-induced activation of these enzymes and determine whether they mediate CPPD crystal-associated repression of neutrophil apoptosis. The results demonstrate that CPPD crystals induce a transient activation of ERK1/ERK2 and Akt in human neutrophils. Activation of both of these kinases was sustained in neutrophils that were stimulated with both TNF-α and CPPD crystals vs neutrophils incubated with TNF-α alone. Repression of either the ERK1/ERK2- or Akt-dependent pathways with pharmacological inhibitors of these pathways resulted in the reversal of the CPPD crystal-induced repression of TNF-α-associated apoptosis.

All chemicals were obtained from Sigma (St. Louis, MO) unless stated otherwise.

CPPD crystals (triclinic form) were prepared and characterized as previously reported (27). The size distribution of the crystals was as follows: 33% < 10 μm, 58% between 10 and 20 μm and 9% > 20 μm. Plasma-opsonized crystals were used in all studies involving crystal-neutrophil incubations. Opsonization of crystals was conducted using 50% heparinized plasma and HBSS at 37°C for 30 min immediately before experiments. Briefly, 25 mg of CPPD crystals were weighed into 1.5-ml Eppendorf tubes followed by 0.5 ml of 50% fresh human plasma. The tubes were capped and placed in an Eppendorf that is strapped to a 30 rpm shaker at 37°C for long enough to ensure crystals were fully suspended (usually 2 min). The rotator was then stopped and the crystals allowed to “sitting” in the 50% plasma for 30 min. At 10-min intervals, the rack was rotated once to ensure good suspension of the crystals. Tubes were then centrifuged at 1000 × g, and crystals were washed in HBSS and centrifuged.

Neutrophils were prepared from freshly collected, human, citrated whole blood by dextran sedimentation and Ficoll Paque density centrifugation. Briefly, blood was mixed with enough 5% dextran T500 (Pharmacia LKB Biotechnology, Uppsala, Sweden) in HBSS to give a final concentration of 1%, and allowed to settle for 30 min. Plasma was collected continuously and 5 ml applied to 5 ml of Ficoll Paque (Pharmacia) in 15-ml polypropylene tubes (Corning Glass, Corning, NY). Following centrifugation at 500 × g for 30 min, the neutrophil pellets were washed free of erythrocytes by 15 s of hypotonic shock in distilled water. Neutrophils were resuspended in HBSS, kept on ice, and used for experiments within 4 h. Neutrophil incubation in HBSS has been used in other studies of neutrophil apoptosis (28, 29). Neutrophils prepared under these conditions yielded cell suspensions that contained over 95% neutrophil cell viability (assessed by trypan blue exclusion). To further confirm neutrophil viability, chemiluminescence studies were performed on a cell sample using a cell concentration of 5 × 10⁶ cells/ml in HBSS containing fMLP at 1 μM and cytochalasin B at 0.5 μM in 1.5-ml capped Eppendorf tubes, as described previously (30).

In all experiments, the final concentrations in 1.5-ml capped Eppendorf incubation tubes were: neutrophils at 5 × 10⁶ cells/ml, CPPD at 25 mg/ml, TNF-α at 10 ng/ml. Cells were incubated at 37°C with occasional tumbling. Preincubation conditions were maintained for 15 min before the addition of another antagonist (as appropriate) for 3.25 h, as indicated in the figure legend. Stock solutions (1000-fold) of the MEK1 and PI3-K inhibitors were prepared in DMSO to give the final concentrations (after addition to the cells): PD98059, 20 ng/ml; U0126, 250 nM; wortmannin, 100 nM; LY294002, 50 μM; z-Asp-Glu-Val-Asp (DEVD)-fmk (Calbiochem, La Jolla, CA), 100 μM.

Crystals were allowed to sediment under gravity for 30 s, and the cell rich supernatant was collected. Residual cells were further separated from crystals with two successive washings in HBSS and gravity sedimentation. Cells were then centrifuged at 500 × g for 5 min. Cells were harvested from incubations that did not contain crystals by centrifugation at 500 × g for 5 min, and washed once with HBSS. For the caspase 3 assay, cells were resuspended at 5 × 10⁷ cells/ml in hypotonic lysis buffer (25 mM HEPES (pH 7.5), 5 mM MgCl₂, 5 mM EDTA, 5 mM DTT, 2 mM PMSF, and 10 μg/ml each of pepstatin A and leupeptin) and subsequently lysed with four successive cycles of freezing in liquid nitrogen and thawing in a 37°C water bath. Cell lysates were centrifuged at 14,000 rpm for 20 min at 4°C, and the supernatant (cytosolic fraction) was collected. A sample was stored at −80°C for total protein concentration determination. Lysates for cytoplasmic histone-associated-DNA fragmentation assessment (Cell Death Detection ELISA^PLUS; Boehringer Mannheim, Indianapolis, IN) were obtained as described by the manufacturer. Briefly, cells were harvested as described above, and washed once in cold (4°C) PBS. Cells were resuspended in lysis buffer provided by the manufacturer to give a final concentration of 1 × 10⁴ cells/ml and incubated at room temperature for 30 min. The lysate was centrifuged at 200 × g for 10 min. The supernatant (cytosolic fraction) was then removed and analyzed immediately.

Two quantitative methods were used for the determination of neutrophil apoptosis; cytoplasmic histone-associated-DNA fragmentation assessment (Cell Death Detection ELISA^PLUS; Boehringer Mannheim) and endogenous caspase 3 substrate (Ac-DEVD-7-amino-4-methyl coumarin (AMC)) cleavage (Fluorometric CaspACE Assay System; Promega, Madison, WI). In all apoptosis experiments, the protein concentration in the samples was determined by Bradford analysis and experiments were normalized to these concentrations.

Caspase 3 (CPP-32β) activity was assessed by endogenous cleavage of the caspase 3-specific substrate peptide DEVD labeled with the fluorochrome AMC provided with the CaspACE Assay System (TB248; Promega). Activity was determined by observing the fluorescence of the cleaved substrate after subtraction of the fluorescence units (FU) obtained in the presence of the tetrapeptide inhibitor DEVD-CHO, as reported previously (39). The nonspecific component of FU is dependent upon the “gain” setting of the fluorometer, which was consistent throughout this study, and was typically observed to be ∼1200 FU. Caspase 3 activity was monitored at 37°C and determined following a 2-h incubation with substrate (and substrate and inhibitor). The rate of fluorescence units released was still in the linear range during this period (as determined in this work).

For cytoplasmic histone-associated-DNA fragmentation assessment (Cell Death Detection ELISA^PLUS; Boehringer Mannheim), 20 μl of extract were used, and the enrichment of nucleosomes in the cytoplasm was quantitated as described by the manufacturer. Briefly, lysate was added to streptavidin-coated wells of 96-well microtiter plates, to which was added a mixture of anti-histone-biotin and anti-DNA-peroxidase Ab. Following a 2-h incubation and washing, the amount of cytoplasmic nucleosome was quantified by the peroxidase retained in the immunocomplex, which was determined spectrophotometrically with 2,2′-azino-di[3-ethylbenzthiazolin-sulfonat] as substrate at an absorbance of 405–490 nm.

ERK1 was assayed at 30°C using the substrate myelin basic protein (MBP; 1 mg/ml). Akt was assayed using either the Akt-peptide (1.5 mg/ml) (Upstate Biotechnology, Lake Placid, NY), or histone H2B (1.5 mg/ml) (Sigma). Substrates were prepared in assay dilution buffer: 20 mM MOPS (pH 7.2), 25 mM β-glycerophosphate, 5 mM EGTA, 2 mM EDTA, 20 mM MgCl₂, 2 mM sodium orthovanadate, 1 mM DTT, 500 nM cAMP-dependent protein kinase inhibitor peptide.

Three hundred micrograms of total protein (bicinchoninic acid assay; Pierce, Rockford, IL) from extract supernatants were diluted into immunoprecipitation buffer (50 mM Tris-HCl, pH 7.5, 150 mM NaCl, 0.5% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% SDS, 5% glycerol, 10 mM sodium fluoride, 5 mM EGTA, 1 mM EDTA, 30 mM β-glycerolphosphate), and the solutions were incubated with the appropriate Abs to MAPK: anti-ERK1-CT, (Upstate Biotechnology) and Akt: anti-Akt1 (Santa Cruz Biotechnology, Santa Cruz, CA) for 4 h at 4°C. A 50:50 mixture of protein A-Sepharose and protein G-Sepharose beads (40 μl) (Pharmacia) that were preincubated in cold (4°C) immunoprecipitation buffer were added, and the samples were further incubated for 1 h at 4°C. The beads were washed twice with immunoprecipitation buffer, and twice with KII buffer (12.5 mM MOPS, pH 7.5, 12 mM β-glycerophosphate, 5 mM EGTA, 7.5 mM MgCl₂, 50 mM NaF, and 0.25 mM DTT). The beads were resuspended in KII buffer supplemented with 10 mM MgCl₂ and 5% glycerol and 10 μl of substrate, and the reactions were initiated with 10 μl [γ-³²P]ATP (∼2000 cpm/pmol) (in assay dilution buffer) in a final volume of 40 μl and incubated for 20 min at 30°C. For the MBP and Akt-peptide p81 paper assays, the reaction was stopped by spotting 20-μl aliquots onto a 1.5-cm × 2-cm p81 phosphocellulose paper, which was then washed extensively with 1% phosphoric acid with 10 changes, after which the adsorbed radioactivity was quantified by liquid scintillation counting in a Packard (Meriden, CT) TriCarb 4530 instrument. Otherwise, the reactions were terminated with the addition of 5× SDS-sample buffer (125 mM Tris-HCl, pH 6.8, 4% SDS, 0.01% bromophenol blue, 10% β-mercaptoethanol, and 20% glycerol), boiled for 5 min, and loaded onto a SDS-PAGE gel. Following transfer of protein as described above, the membrane was immunoblotted for the appropriate protein, exposed to film, and the bands representing MBP for the ERK1 assays, or histone H2B for the Akt assays were excised and subjected to Cerenkov counting.

SDS-PAGE of the neutrophil lysates was performed on 1.5-ml-thick gels. A 12% separating gel and a 4% stacking gel were used. Samples were boiled for 5 min in the presence of 5× SDS-sample buffer and electrophoresed for ∼12 h at 10 mA. Subsequently, the separating gel was soaked in transfer buffer (25 mM Tris, 192 mM glycine, 20% methanol) for 10 min and then sandwiched with a nitrocellulose membrane. Proteins were transferred for 3 h at 300 mA. For Akt Western analysis, the membranes were blocked with TBS (20 mM Tris-HCl, pH 7.4, 0.25 M NaCl) containing 5% BSA for 3 h at room temperature, and then washed three times in TBS containing 0.05% Tween (TBST) for 15 min. Anti-phosphoERK blots were treated with PBS instead of TBS throughout. The membranes were incubated overnight at room temperature with primary Ab (phosphoERK; New England Biolabs, Beverly, MA; Akt, anti-Akt1; Santa Cruz Biotechnology). Membranes were then washed with TBST three times before 45 min of incubation at room temperature with secondary Ab (goat anti-rabbit IgG coupled to HRP in 5% BSA/PBS). The membranes were then rinsed three times in TBST and once in TBS before (unless when developing the anti-phosphoERK blot). For detection of HRP-conjugated secondary Abs, membranes were washed as described previously and subjected to enhanced chemiluminescence (Amersham, Arlington Heights, IL).

Statistical significance was determined using the Student’s t test at p < 0.05.

Induction of apoptosis in neutrophils mediated by TNF-α and spontaneous apoptotic signaling has been shown to converge upon the activation of caspase 3 (10, 31). Endogenous neutrophil caspase 3 activity was determined by monitoring the cleavage of the caspase 3-specific peptide DEVD labeled with the fluorochrome AMC with and without the inhibitory tetrapeptide DEVD-CHO to subtract nonspecific cleavage. Neutrophils incubated in buffer for 3 h showed a 2-fold increase in caspase 3 activity relative to freshly isolated neutrophils (control), indicative of spontaneous apoptosis (Ref. 24 , and references within) (Fig. 1 A). Neutrophil stimulation with TNF-α resulted in a 5-fold increase in caspase 3 activity compared with untreated cells or an ∼10-fold increase over control cells. Incubation with CPPD crystals in the absence or presence of TNF-α reduced the caspase 3 activity to control levels. Neutrophil incubation with the MEK1 inhibitor PD98059, or the PI3-K inhibitor wortmannin resulted in caspase activity that was similar to the basal levels observed in neutrophils incubated in buffer for 3 h. We have demonstrated that wortmannin inhibits PI3-K at the concentration shown here (32), and that no change in caspase 3 activity can be observed even at a 10-fold higher concentration (data not shown). Preincubation with the MEK1 inhibitor U0126 or the PI3-K inhibitor LY294002 gave similar results (data not shown). These results indicate that repression of the ERK1/ERK2 and PI3-K/Akt pathways may not affect caspase 3 activation associated with spontaneous apoptosis.

Preincubation of TNF-α-stimulated neutrophils with the MEK1 inhibitor PD98059 had no effect on TNF-α-induced caspase 3 activity (Fig. 1,B). Because MEK1 functions directly upstream of ERK1 and ERK2, any effects that these MEK1 inhibitors have on neutrophils can be assumed to be regulated by these MAP kinases. Fig. 1 B demonstrates that inhibiting ERK1 and ERK2 by preincubation with PD98059 or U0126 in neutrophils costimulated with CPPD crystals and TNF-α resulted in caspase 3 activity levels significantly greater than that observed in neutrophils coincubated with CPPD crystals and TNF-α alone. Therefore, repression of the ERK1/ERK2 pathways resulted in an abrogation of the repressive effect CPPD crystals have on TNF-α-induced caspase 3 activation. Pretreatment with U0126, which is a more potent MEK1 and MEK2 inhibitor than PD98059, resulted in caspase 3 activity that exceeded levels obtained from neutrophils stimulated with TNF-α.

Preincubation of TNF-α-stimulated neutrophils with the PI3-K inhibitor wortmannin did not inhibit TNF-α-induced caspase 3 activity (Fig. 1,C), but resulted in an ∼25% increase in caspase 3 activation over that in neutrophils stimulated with TNF-α alone. Fig. 1 C demonstrates that inhibiting PI3-K by preincubation with wortmannin or LY294002 in neutrophils costimulated with CPPD crystals and TNF-α resulted in caspase 3 activity levels significantly greater than that observed in neutrophils coincubated with CPPD crystals and TNF-α alone. Therefore, both wortmannin and LY294002 pretreatment resulted in an abrogation of the repressive effect that CPPD crystals have on TNF-α-induced caspase 3 activation. These results indicate that the ERK1/ERK2 and PI3-K pathways are both important for crystal-associated repression of TNF-α-induced caspase 3 activation.

We have previously reported that a 2-min incubation of neutrophils with CPPD crystals results in the activation of ERK1/ERK2 (30). These MAP kinases have been shown to undergo biphasic activation, depending on the type of stimuli and cells, and it has been suggested that the duration of their activation can determine whether they function to regulate apoptosis (33, 34). Therefore, we determined the effects of CPPD crystals and TNF-α on ERK1 phosphotransferase activity at different times over the period of caspase 3 activation. Incubation of neutrophils with CPPD crystals resulted in a rapid induction of ERK1 activity over basal levels that were sustained for at least 60 min (Fig. 2,A). Incubation with TNF-α resulted in a 50% repression of ERK1 activity compared with untreated cells at 5 min and 3 h, which may be due to enhanced apoptosis in TNF-α-stimulated cells vs untreated cells. Costimulation of neutrophils with CPPD crystals and TNF-α rescued the TNF-α-associated repression of ERK1 activity, and resulted in an increasing and sustained activation of ERK1 (Fig. 2,A). At 3 h of incubation with both TNF-α and CPPD crystals, an ∼100% increase of ERK1 activity over basal levels at 0 min was observed. Fig. 2 B demonstrates that preincubation with PD98059 resulted in the inhibition of CPPD crystal-induced activation of ERK1. These results are consistent with the notion that CPPD crystal-dependent activation of ERK1 in untreated or TNF-α-stimulated neutrophils may be responsible for suppression of TNF-α-induced and spontaneous apoptotic activation of caspase 3.

We have observed in our laboratory that neither ERK1, ERK2, nor Akt undergo changes in expression for at least 4 h (the period we have conducted this analysis) (data not shown). To determine whether both ERK1 and ERK2 are affected similarly by CPPD crystals, we employed a qualitative method of determining ERK1 and ERK2 activity by assessing the level of phosphorylation of these enzymes by Western blot analysis with Abs that recognize the phosphorylated (active) forms of ERK1 and ERK2 (Fig. 3). Fig. 3 illustrates that in both untreated and TNF-α-stimulated cells, phosphorylation of p44 ERK1 (upper band) and p42 ERK2 (lower band) is minimal, with lower levels following extended periods of incubation. CPPD crystal-induced phosphorylation of ERK1 and ERK2 in the absence or presence of TNF-α is significantly greater than cells incubated alone or with TNF-α, respectively. Furthermore, the phosphorylation of ERK1 and particularly ERK2 is sustained in the presence of CPPD crystals. These results are consistent with the MBP phosphotransferase analysis described in Fig. 2, and indicate that the ERK1/ERK2 signaling pathway is positively regulated by crystals and may have a critical role in promoting crystal-dependent neutrophil survival.

PI3-K has been demonstrated to function upstream of Akt because wortmannin can block Akt activation (35). Platelet-derived growth factor mutants that fail to activate PI3-K also fail to activate Akt (36, 37). Because repression of PI3-K with wortmannin inhibited the prosurvival effect of CPPD crystals, we investigated the effects of TNF-α and CPPD crystals in the absence or presence of TNF-α on endogenous neutrophil Akt activity (Figs. 4 and 5). Minimal Akt-associated histone H2B phosphotransferase activity was observed following 30 min and 5 min of incubation in buffer without or with TNF-α, respectively, and was negligible compared with neutrophils incubated with CPPD crystals (Fig. 4,A). Furthermore, incubation of TNF-α-stimulated neutrophils with CPPD crystals resulted in an induction of Akt activity vs neutrophils stimulated with TNF-α alone. Preincubation with wortmannin was shown to inhibit CPPD crystal-induced Akt activity. Interestingly, pretreatment with PD98059 resulted in an enhanced activation of Akt following 30 min of CPPD crystal incubation (Fig. 4,A). Additional controls are shown in Fig. 5 B.

Fig. 5 demonstrates a time course analysis of endogenous Akt activity by measuring the phosphotransferase activity of immunoprecipitated Akt with an Akt-specific substrate peptide in the presence of [γ-³²P]ATP. Crystal-induced Akt activation was observed to be sustained for ∼60 min with no apparent biphasic activation (Fig. 6 A). Activation of Akt in cells costimulated with CPPD crystals and TNF-α also resulted in a transient activation of Akt. These results indicate that CPPD crystal-induction of Akt may function to promote neutrophil survival, and that only a short, transient activation is probably required.

The measurement of the cytoplasmic DNA concentrations allows for the quantitation of this late apoptotic event in neutrophils. We assessed the effects of the PD98059 and wortmannin on CPPD crystal-mediated repression of TNF-α and spontaneous associated DNA fragmentation in neutrophils to determine whether CPPD crystal-induced induction of ERK1/ERK2 and Akt and repression of caspase 3 results in the inhibition of neutrophil apoptosis (Fig. 6). Cells incubated in buffer alone for 4 h had an enhanced level of DNA fragmentation vs cells that were isolated from human blood and immediately assessed for DNA fragmentation, indicative of spontaneous apoptosis. TNF-α-stimulated cells showed amplified levels of DNA fragmentation of ∼3-fold over cells incubated in buffer alone. Similar to the caspase 3 activity results, coincubation of neutrophils with CPPD crystals and TNF-α resulted in a repression of TNF-α-induced DNA fragmentation to levels comparable to untreated cells at t = 0. Preincubation with either PD98059 or wortmannin in cells costimulated with CPPD crystals and TNF-α resulted in the repression of the inhibitory effect that CPPD crystals have on TNF-α-induced DNA fragmentation. This result indicates that induction of the ERK1/ERK2 and PI3-K/Akt pathways may regulate CPPD crystal repression of TNF-α-associated apoptosis. Incubation with PD98059 or wortmannin alone resulted in a repression of DNA fragmentation to levels lower than cells incubated in buffer alone for 4 h.

Neutrophil preincubation with PD98059 in CPPD crystal-stimulated cells resulted in an induction of DNA fragmentation, whereas wortmannin preincubation had no effect. This indicates that, contrary to TNF-α-induced apoptosis, repression of crystal-induced PI3-K does not result in the inhibition of spontaneous apoptosis, although caspase 3 activity is suppressed. To illustrate that crystal induction of the ERK1/ERK2 and PI3-K/Akt pathways regulate neutrophil survival through suppression of caspase 3 in TNF-α-stimulated cells, we coincubated neutrophils treated with TNF-α, CPPD crystals, and TNF-α in the absence or presence of PD98059 or wortmannin, with the caspase 3 inhibitor z-DEVD-fmk (38). Inhibition of caspase 3 resulted in DNA fragmentation levels significantly lower than levels observed in untreated cells under all conditions, except for cells costimulated with CPPD crystals and TNF-α that were preincubated with wortmannin. Therefore, PI3-K-mediated suppression of TNF-α-associated apoptosis functions through an alternate pathway, in addition to suppressing caspase 3 activity. The multifunctional role of PI3-K in response to TNF-α was also apparent when TNF-α-stimulated neutrophils were preincubated with wortmannin. DNA fragmentation was observed to be inhibited, which indicates that TNF-α-associated PI3-K can have a proapoptotic effect (Fig. 6).

Studies in this laboratory have focused on elucidating the signal transduction pathways involved in the microcrystal-induced neutrophil oxidative and degranulation responses. Recently, we showed that neutrophil activation is mediated by CPPD crystal-induced activation of PKC (39) and PI3-K (30). Two inhibitors of PI3-K, wortmannin and LY294002, were shown to completely suppress neutrophil activation induced by plasma-coated CPPD crystals at concentrations lower than the known IC₅₀ of these inhibitors for PI3-K. The involvement of PI3-K activity in chemoattractant-stimulated neutrophils has been reported previously (40, 41). Neutrophil activation by both fMLP and CPPD crystals has been demonstrated to activate both p44 ERK1 and p42 ERK2 forms of MAP kinase (30, 42). We reported a possible link between CPPD crystal-induced ERK1/ERK2 activation and neutrophil chemiluminescence, superoxide anion generation, and degranulation. We demonstrated that the anti-cancer, and putative anti-arthritic drug paclitaxel (43, 44) effectively disrupted ERK1/ERK2 and neutrophil activation stimulated by CPPD crystals (30). In our investigations with AGM-1470, which has been demonstrated to inhibit rat adjuvant and collagen-induced arthritis (43, 45), we found that AGM-1470 inhibited both fMLP and PMA induction of ERK1/ERK2 and neutrophil activation, but had no effect on CPPD crystal-induced MAP kinase and neutrophil stimulation (32). Furthermore, we found a PI3-K-independent correlation between ERK1/ERK2 and neutrophil activation.

It is now thought likely that the extended inflammation associated with disease states such as rheumatoid arthritis and inflammatory bowel disease are characterized by the accumulation of neutrophils in the joint or bowel, respectively. Because the rapid accumulation of neutrophils in the synovial joints of patients with MSUM or CPPD crystal induced inflammation is well-documented (1), we have addressed in the current study whether CPPD crystal induction of the ERK1/ERK2 and/or PI3-K/Akt signaling pathways are responsible for crystal-induced neutrophil survival under TNF-α and spontaneous apoptotic conditions. CPPD crystals were observed to induce a robust and transient activation of the MAP kinases and Akt. In the presence of TNF-α, CPPD crystals induced ERK1 and Akt activation that was not present in cells stimulated with TNF-α alone, and this activation was sustained compared with cells incubated with CPPD crystals alone. Direct inhibition of MEK1 or PI3-K with specific pharmacological antagonists resulted in the repression of ERK1/ERK2 and Akt activities, respectively, associated with CPPD crystal incubation, in the absence or presence of TNF-α. Furthermore, inhibition of the ERK1/ERK2 or PI3-K pathways with their respective inhibitors resulted in the repression of CPPD crystal-associated suppression of TNF-α-induced neutrophil apoptosis as determined by caspase 3 activity and DNA fragmentation.

The stimulation of hemopoietic cells by cytokines leads to the rapid activation of PI3-K (13, 21). The lipid products of PI3-K serve to localize and activate downstream signaling molecules, such as Akt, in the vicinity of the cell membrane (35, 37, 40). The signal transduction pathway initiated by growth factors and cytokines involving PI3-K and Akt seems to be important for cell survival in a number of cell systems (13, 21, 22). The role of PI3-K and Akt in neutrophil survival in response to neutrophil-CPPD crystal interactions or microcrystal phagocytosis has not been previously delineated. We have shown a correlation between CPPD crystal-induced activation of the PI3-K/Akt pathway and neutrophil survival in TNF-α-stimulated cells. Inhibition of Akt with inhibitors of PI3-K in cells costimulated with TNF-α and CPPD crystals resulted in an activation of caspase 3 and DNA fragmentation associated with TNF-α-stimulated apoptosis. The observation that caspase 3 inhibition with z-DEVD-fmk in wortmannin preincubated neutrophils stimulated with CPPD crystals and TNF-α resulted in the induction of DNA fragmentation (Fig. 6) indicates that PI3-K functions to regulate neutrophil survival through an alternate pathway, in addition to suppression of caspase 3. Because preincubation of CPPD crystal-stimulated cells with wortmannin did not result in the stimulation of DNA fragmentation, CPPD crystal-associated repression of spontaneous apoptosis is not mediated solely through the PI3-K/Akt pathway even though caspase 3 is repressed.

Although we have demonstrated an activation of Akt in neutrophils within 5 min of stimulation with CPPD crystals (39), a time course of Akt activity following crystal incubation has not been done. Activation of Akt in cells incubated with CPPD crystals, or costimulated with crystals and TNF-α resulted in a transient activation, which diminished to basal levels after 30 min (Figs. 4 and 5). PI3-K has been shown to contribute to the activation of the ERK1 and ERK2 (38). Dominant negative forms of PI3-K were used to show that selective inhibition of Raf isoforms, upstream effecters of ERK1 and ERK2, may explain the ability of the PI3-K pathway to affect these MAP kinases in some cell systems, and not others (46). CPPD crystal-stimulated neutrophils preincubated with wortmannin showed an ∼30% increase in crystal-induced ERK1/ERK2 activity. Therefore, the ability of wortmannin and LY294002 to inhibit the prosurvival effect of CPPD crystals in TNF-α-stimulated cells is not a result of ERK1 inhibition in neutrophils.

Because CPPD crystals were able to activate ERK1 and ERK2 in neutrophils incubated in buffer alone or with TNF-α over basal levels, it was possible that this MAP kinase pathway could mediate the signal for neutrophil survival. We show here that CPPD crystal-dependent cell survival in TNF-α-stimulated neutrophils is dependent on ERK1/ERK2 activation: inhibition of MEK1 with PD98059 or U0126 abrogated the anti-apoptotic effect of crystals (Fig. 2). The transient ERK1 and ERK2 activation in CPPD crystal-incubated neutrophils stimulated with TNF-α indicates that the additional CPPD crystal-dependent activation of ERK1/ERK2 over basal levels could be involved in promoting neutrophil survival. It could be suggested that the transient activation of this MAP kinase pathway induces the expression of genes that are anti-apoptotic. Previous work in our laboratory is consistent with this. We showed that the suppression of TNF-α induced and spontaneous apoptosis mediated through caspase 3 following crystal incubation was inhibited by preincubation with cycloheximide (24). Although the activation of the ERK1/ERK2 pathway in TNF-α-stimulated cells was sustained in response to crystals over levels observed in neutrophils treated with crystals alone, these results indicate that only the initial activation of ERK1/ERK2, leading to the expression of prosurvival genes is necessary for neutrophil survival. We have identified in our laboratory that cycloheximide preincubation of CPPD crystal-treated neutrophils resulted in the selective repression of CPPD crystal-induced ERK1 activity, with no effect on endogenous Akt activity (data not shown). We previously reported the existence of a pathway in TNF-α-associated apoptosis that is repressed by crystal treatment independent of protein synthesis, which probably does not function in neutrophil spontaneous apoptosis (24). The results presented in this report suggest that the PI3-K→Akt pathway selectively represses TNF-α vs spontaneous associated apoptosis signaling, whereas the ERK1/ERK2 pathway represses both. Therefore, it is possible that the cycloheximide-dependent suppressive effects of CPPD crystals on neutrophil apoptosis may be mediated in part through the ERK1/ERK2 pathway, independent of PI3-K→Akt. This is currently under investigation in our laboratory.

Repression of CPPD crystal-induced survival was induced by the preincubation of neutrophils with either the MEK1 inhibitors or the PI3-K inhibitors in TNF-α-stimulated cells. These results, coupled with the observation that the MEK1 and PI3-K inhibitors suppressed ERK1/ERK2 or Akt pathways, respectively, would indicate that the induction of both the ERK1/ERK2 and PI3-K/Akt pathways is necessary to mediate neutrophil survival. It is interesting to note that cells preincubated with wortmannin for 5 min resulted in a ∼30% increase in CPPD crystal-induced ERK1 activation (Fig. 3,B), and cells preincubated with PD98059 for 30 min resulted in a ∼95% increase in CPPD crystal-induced Akt activation (Fig. 5 B). This would suggest that cross-talk between these pathways exist in neutrophils, where one pathway actually functions to suppress the other. Our laboratory and others have demonstrated that both ERK1/ERK2 and PI3-K/Akt are involved in neutrophil activation (30, 32, 47). Further, reactive oxygen species production in neutrophils has been shown to mediate apoptosis in neutrophils stimulated with Fas/CD95 (48, 49). It is possible that repression of basal levels of the ERK1/ERK2 and PI3-K, because of their coinhibitory effects on each other during normal neutrophil activation, results in the suppression of prosurvival signaling, because subsequent inhibition of ERK1/ERK2 and PI3-K pathways would result in the reduction of their suppressive effects on caspase 3. Because the levels of ERK1/ERK2 and Akt activity in neutrophils are significantly greater subsequent to CPPD crystal treatment, it is likely that any repression is masked by the immense inductive signals mediated by the crystals.

Another explanation for the observation that either inhibition of the ERK1/ERK2 or PI3-K pathways results in the induction of caspase 3 in CPPD crystal-stimulated neutrophils is that these pathways may converge to regulate survival. For instance, it has been demonstrated that the Bcl-2 family proteins regulate caspase 3 in neutrophils (50). Both of the anti-apoptotic bcl-X₁ gene and the death-promoting bax-α gene were shown to be expressed in neutrophils. Bcl-X₁ and bax-α are known to form homodimers and heterodimers, respectively (Ref. 51 , and references within). Any shift in the balance of the bax-α/bcl-X₁ ratio, which can be achieved by the up-regulation or down-regulation of both interacting members, has previously been shown to determine whether a cell will undergo apoptosis (51). ERK1/ERK2 can activate the Elk-1 and cAMP response element-binding protein, CREB. Activated CREB binds to the cAMP response elements (CRE), specific sites on the cAMP responsive genes. The Bcl-2 promoter contains a functional CRE, and Bcl-2 expression in B cells is also dependent on CREB phosphorylation (52). Therefore, induction of ERK1/ERK2 by CPPD crystals may shift the balance of bax-α and bcl-X₁ in favor of the prosurvival bcl-X₁ gene. Bad family proapoptotic proteins, including Bax-α, can be repressed posttranscriptionally by phosphorylation (51, 52). Phosphorylation of the Bad family proteins has been demonstrated to be mediated by Akt (53). Collectively, these results, coupled with these presented here, indicate that CPPD crystal-induction of the ERK1/ERK2 and PI3-K pathways, may cooperate to repress caspase 3 activity in neutrophils by regulating the expression of genes that are anti-apoptotic, and may also induce posttranslational modification to inactivate the neutrophil-intrinsic death machinery. In the presence of CPPD crystals, the induction of other signaling pathways, such as ERK1/ERK2, probably function in concert with PI3-K to tilt the balance of the survival/apoptosis signals in favor of a prosurvival signal in neutrophils.

Our findings with respect to CPPD crystal-associated neutrophil survival being mediated upstream of caspase 3 through activation of the ERK1/ERK2 and PI3-K/Akt pathways is illustrated in Fig. 7. The specific mechanism(s) remain to be elucidated and are currently under investigation in our laboratory. For instance, we are now investigating the regulation of bcl-X₁ and bax-α expression in response to CPPD crystals in neutrophils. Together, our results indicate that CPPD crystals function to induce acute inflammatory responses through ERK1/ERK2 and PI3-K/Akt-mediated repression of apoptosis in neutrophils. Furthermore, we suggest that mediators of the ERK1/ERK2 pathway may have therapeutic implications in treating diseases, where accumulation of neutrophils occurs because their apoptosis machinery has become dysfunctional is symptomatic, including arthritis.