HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: 285/2/C286    most recent 00070.2003v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (37) Citing Articles via Google Scholar Google Scholar Articles by Gudipaty, L. Articles by Dubyak, G. R. Search for Related Content PubMed PubMed Citation Articles by Gudipaty, L. Articles by Dubyak, G. R.

RECEPTORS AND SIGNAL TRANSDUCTION

Essential role for Ca²⁺ in regulation of IL-1 secretion by P2X₇ nucleotide receptor in monocytes, macrophages, and HEK-293 cells

Department of Physiology and Biophysics, Case Western Reserve University, Cleveland, Ohio 44106

Submitted 21 February 2003 ; accepted in final form 20 March 2003

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Interleukin (IL)-1 is a proinflammatory cytokine that elicits the majority of its biological activity extracellularly, but the lack of a secretory signal sequence prevents its export via classic secretory pathways. Efficient externalization of IL-1 in macrophages and monocytes can occur via stimulation of P2X₇ nucleotide receptors with extracellular ATP. However, the exact mechanisms by which the activation of these nonselective cation channels facilitates secretion of IL-1 remain unclear. Here we demonstrate a pivotal role for a sustained increase in cytosolic Ca²⁺ to potentiate secretion of IL-1 via the P2X₇ receptors. Using HEK-293 cells engineered to coexpress P2X₇ receptors with mature IL-1 (mIL-1), we show that activation of P2X₇ receptors results in a rapid secretion of mIL-1 by a process(es) that is dependent on influx of extracellular Ca²⁺ and a sustained rise in cytosolic Ca²⁺. Moreover, reduction in extracellular Ca²⁺ attenuates 90% of P2X₇ receptor-mediated IL-1 secretion but has no effect on enzymatic processing of precursor IL-1 (proIL-1) to mIL-1 by caspase-1. Similar experiments with THP-1 human monocytes and Bac1.2F5 murine macrophages confirm the unique role of Ca²⁺ in P2X₇ receptor-mediated secretion of IL-1. In addition, we report that cell surface expression of P2X₇ receptors in the absence of external stimulation also results in enhanced release of IL-1 and that this can be repressed by inhibitors of P2X₇ receptors. We clarify an essential role for Ca²⁺ in ATP-induced IL-1 secretion and indicate an additional role of P2X₇ receptors as enhancers of the secretory apparatus by which IL-1 is released.

interleukin-1; calcium ion; adenosine 5'-triphosphate; P2X₇ receptors

Previous reports established that perturbations of monovalent cation homeostasis, specifically the loss of intracellular K⁺ and gain of Na⁺, within monocytes and macrophages result in the activation of procaspase-1 and thereby accelerate the processing and release of IL-1 (24–26). However, the regulation of IL-1 processing by perturbation of monovalent cation homeostasis is less critical in cells that maintain high levels of the catalytically active form of caspase-1. For example, freshly isolated human blood monocytes primed with LPS can process and secrete significant amounts of IL-1 even in the absence of secondary stimuli that alter Na⁺ and K⁺ fluxes (34). Nonetheless, the rate of IL-1 secretion from these cells is markedly enhanced by P2X₇ receptor activation even in the absence of intracellular K⁺ loss and Na⁺ gain, suggesting additional roles for this receptor in the regulation of IL-1 secretion independent of caspase-1 activation (26).

The extremely rapid and highly coupled reactions that characterize P2X₇ receptor-induced IL-1 processing and secretion in monocytes and macrophages have complicated experiments aimed at dissociation of IL-1 secretion mechanisms from the upstream pathways of caspase-1 activation and proIL-1 cleavage. Previous studies aimed at dissociating IL-1 secretion from IL-1 processing have used a variety of experimental systems yielding disparate results. For example, Suttles et al. (34) reported that Ca²⁺ ionophores induce mIL-1 release from LPS-primed human monocytes but Perregaux et al. (24) were unable to observe similar effects of such ionophores in mouse peritoneal macrophages; this apparent discrepancy remains unresolved. Gardella et al. (8, 9) found that monocyte-derived dendritic cells stimulated with T cells could package IL-1 into vesicles resembling late, recycling endolysosomes and that this subcellular pool of IL-1 could be secreted via a Ca²⁺-dependent mechanism. Recently, MacKenzie et al. (18) demonstrated that 2',3'-O-(4-benzoyl) benzoyl-ATP (BzATP)-induced IL-1 release from human THP-1 monocytes occurs concurrently with microvesiculation, a phenomenon in which the plasma membrane sheds evaginated vesicles (0.5-µm diameter) within seconds of P2X₇ receptor activation. Those investigators further established that HEK-293 cells expressing P2X₇ receptors also exhibited a Ca²⁺-dependent microvesiculation response when stimulated with P2X₇ receptor agonists (18).

Given these observations, we hypothesized that P2X₇ receptors induce IL-1 secretion via Ca²⁺-dependent mechanisms, independent of their role in caspase-1 activation. As an experimental model for testing this hypothesis, we used human embryonic kidney cells (HEK-293 cells) that stably overexpress the human P2X₇ receptor as a vehicle for the transient transfection of expression plasmids encoding the mature (mIL-1) or precursor (proIL-1) forms of IL-1. These studies demonstrated that activation of P2X₇ receptors results in a rapid secretion of mIL-1 by a process(es) that is dependent on influx of extracellular Ca²⁺ and a sustained rise in cytosolic Ca²⁺. Experiments with THP-1 human monocytes and Bac1.2F5 murine macrophages established a similar essential role for increased cytosolic Ca²⁺ in the regulation of IL-1 release by natively expressed P2X₇ receptors. Finally, comparison of basal IL-1 secretion from wild-type HEK-293 (HEK-WT) cells vs. P2X₇ receptor-overexpressing HEK-293 (HEK-P2X₇) cells indicated that cell surface expression of P2X₇ receptors, even in the absence of stimulation by exogenously added ATP, results in an enhanced release of IL-1 that can be repressed by P2X₇ receptor antagonists. These results clarify an essential role for Ca²⁺ influx in P2X₇ receptor-induced IL-1 secretion and suggest an additional role for P2X₇ receptors as adapter proteins in the assembly or activation of the secretory apparatus by which IL-1 is exported from cells.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Reagents, plasmids, and antibodies. Cell culture reagents included poly-L-lysine-coated plates (BD-Biosciences), Dulbecco's modified Eagle's medium (DMEM; Sigma), RPMI 1640 (Sigma), human recombinant IFN- (Genentech), iron-supplemented newborn bovine calf serum (CS) (Hyclone), and Escherichia coli LPS (01101:B4 and K12-LCD25 sero-types; List Biologicals). Transfection reagents included a calcium phosphate transfection kit (Sigma), proIL-1 and mIL-1 cDNA in pRSV (gifts from Dr. Steven Mizel, Wake Forest University), and procaspase-1 subcloned into pcDNA3.1 (Invitrogen) from the original bacterial expression plasmid kindly provided by Dr. Douglas Miller (Merck). All nucleotides, A-23187, and ionomycin were purchased from Sigma. Human ELISA antibodies included M421B-E and M420B-B (Endogen). Murine IL-1 ELISA antibodies (PM-425B and MM-425B-B) were also from Endogen. The anti-IL-1 murine monoclonal antibody (3ZD) used for Western blot analyses was provided by the Biological Resources Branch of the National Cancer Institute-Frederick Cancer Research and Development Center; this antibody reacts with the precursor and processed forms of both human and murine IL-1. A rabbit anti-human caspase-1 (sc-515) antibody was from Santa Cruz Biotechnology, and a rabbit anti-P2X₇ receptor antibody was from Alamone. All horseradish peroxidase (HRP)-conjugated secondary antibodies were from Santa Cruz Biotechnology.

Cell culture. Human embryonic kidney (HEK-293) cells were stably transfected with the human P2X₇ receptor cDNA in the pIRES vector (HEK-P2X₇). The cells were maintained in DMEM supplemented with 10% CS, 1% PS (100 U/ml penicillin and 100 µg/ml streptomycin), and 250 µg/ml hygromycin. Control cells (HEK-WT) were stably transfected with the empty pIRES vector (to confer hygromycin resistance) and maintained under similar culture conditions. THP-1 human monocytic leukemia cells were cultured in RPMI 1640 supplemented with 10% heat-inactivated CS and 2 mM glutamine. Undifferentiated THP-1 cells were plated onto six-well poly-L-lysine-coated culture dishes at a density of 10⁶/ml and then cultured in the presence of human IFN- (1,000 U/ml) for 2 days to induce monocytic differentiation (12). The differentiated THP-1 monocytes were then primed with LPS (100 ng/ml of E. coli LPS, K12-LCD25) for 4 h to induce expression of proIL-1. Bac1.2F5 murine macrophage cells were cultured as described previously (13). For experiments, these cells were plated onto 12-well dishes at 10⁶/ml and stimulated with 1 µg/ml LPS (E. coli LPS, 0111:B4) for 4 h to induce proIL-1 expression.

Transfection protocols. HEK-WT or HEK-P2X₇ cells were plated onto 12-well poly-L-lysine-coated plates at 2.5 x 10⁵ cells/well. Twenty-four hours after plating, the cell culture medium was replaced with fresh DMEM containing 10% CS and 1% PS. The cells were then transfected with various combinations of cDNA expression plasmids including 0.5 µg of mIL-1, 0.5 µg of proIL-1, or 0.1 µg of procaspase-1 cDNA per well with standard calcium phosphate transfection protocols. Briefly, cDNA and 125 mM CaCl₂ were mixed into 0.1 ml of (in mM) 50 Na-HEPES, 280 NaCl, 1.5 Na₂HPO₄, pH 7.0, and allowed to coprecipitate for 30 min at 23°C before transfer into the 1-ml tissue culture volume. The cells were incubated at 37°C for 18 h after transfection before measurements of IL-1 secretion. In all experiments, the total amount of transfected cDNA was kept constant at 0.6 µg · ml^-1 · well^-1 with pcDNA3.1 vector.

Assay for IL-1 release. At 18 h after transfection (HEK-P2X₇ or HEK-WT cells) or 4 h after LPS stimulation (THP-1 monocytes, Bac1 macrophages), the culture medium was collected and the adherent cells were washed twice with prewarmed (37°C) phosphate-buffered saline (PBS). The cells were then bathed in 1 ml of HEPES-buffered saline (HBS) containing (in mM) 130 NaCl, 5 KCl, 20 HEPES, 1.0 CaCl₂, 1.0 MgCl₂, and 5 glucose with 0.1 mg/ml BSA, pH 7.4 (Ca²⁺-Mg²⁺-HBS) or in 1 ml of HBS containing (in mM) 130 NaCl, 5 KCl, 20 HEPES, 2.0 MgCl₂, and 5 glucose with 0.1 mg/ml BSA, pH 7.4 (Ca²⁺-free Mg²⁺-HBS). In some experiments (see Figs. 4 and 7), the extracellular divalent cation content was additionally varied to include 1) 0.3 mM CaCl₂ and 1.7 mM MgCl₂ or 2) 0.6 mM CaCl₂ and 1.4 mM MgCl₂; 130 mM NaCl was replaced with 130 mM KCl (KCl HBS) in other experiments (see Table 1). After washing and transfer to test saline solutions, the cells were further stimulated or not with various agonists (3 mM ATP, 300 µM BzATP, 100 µM ADP, 100 µM UTP, 1 µM ionomycin, or 1 µM A-23187) for up to 60 min at 37°C. At selected times, the conditioned medium was collected and centrifuged to pellet any detached cells; 5- to 25-µl aliquots were assayed for IL-1 by ELISA. The cells were washed once with PBS and lysed into 0.1–0.2 ml of lysis buffer (25 mM Na-HEPES, 300 mM NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 1% Triton X-100, 2 µg/ml leupeptin, and 100 µg/ml PMSF). One- to five-microliter aliquots of these lysates were also assayed for IL-1.

View larger version (17K): [in this window] [in a new window]   Fig. 4. Effects of cytosolic Ca2+ buffering on the kinetics of P2X7 receptor-induced IL-1 secretion. A: HEK-P2X7 cells were assayed for changes in cytosolic Ca2+ in response to 3 mM ATP, 100 µM ADP, and 100 µM UTP. B: HEK-P2X7 cells were preloaded with 30 µM BAPTA-AM and then assayed for changes in cytosolic Ca2+ in response to 3 mM ATP, 100 µM ADP, and 100 µM UTP. C: HEK-P2X7 cells were transiently transfected with 0.5 µg of mIL-1. Seventeen and a half hours after transfection, the cells were loaded with 30 µM BAPTA for 30 min. The cells were washed and then stimulated with ATP (3 mM) or not for 30 min. A similar plate of HEK-P2X7 cells transfected with 0.5 µg of mIL-1 and not loaded with BAPTA-AM served as controls. Supernatants from the various conditions were collected at 0, 5, 10, 15, 30, and 60 min after ATP stimulation and analyzed by ELISA.

View larger version (32K): [in this window] [in a new window]   Fig. 7. Equivalent IL-1 expression but decreased IL-1 release from wild-type HEK-293 (HEK-WT) cells vs. HEK-P2X7 cells transiently transfected with mIL-1. A: HEK-WT cells and HEK-P2X7 cells were transiently transfected with cDNA for procaspase-1 or mIL-1. The cell lysates were analyzed by Western blots probed for P2X7 receptor protein (top) caspase-1 (middle), and IL-1 (bottom). Asterisks indicate nonspecific bands. B: HEK-P2X7 cells or HEK-WT cells were transfected with 0.5 µg of mIL-1 cDNA. Eighteen hours after transfection, the cells were stimulated for 30 min at 37°C with ATP (3 mM), ionomycin (1 µM), or A-23187 (1 µM). Unstimulated IL-1 release (t = 30 min) from HEK-P2X7 cells was assigned a value of 100, and the data presented for HEK-WT cells and HEK-P2X7 cells are normalized to this value. Results represent n = 4 experiments with each experiment performed in duplicate. The unstimulated and ATP-stimulated data represent n = 14 experiments.

View this table: [in this window] [in a new window]   Table 1. ATP- and Ca2+ ionophore-mediated IL-1 release from IFN-- and LPS-differentiated THP-1 monocytes vs. LPS-primed Bac1 macrophages

IL-1 ELISA. Sandwich ELISA protocols were used to assay for IL-1 in the extracellular medium samples and cell lysates. Briefly, a 96-well plate was coated with 1 µg/ml primary anti-human or anti-murine IL-1 by overnight incubation at 23°C and was then blocked with 4% BSA in PBS for 1 h. Plates were washed three times with ELISA buffer (50 mM Tris · HCl, pH 7.5, 0.2% Tween 20). Five- to twenty-five-microliter aliquots of medium samples or cell lysates diluted to fifty microliters with PBS were added to the blocked wells together with fifty microliters of the second, biotinylated anti-human or anti-murine IL-1 antibody (0.2 µg/ml). Other wells were supplemented with known amounts of human or murine IL-1 standards. The plates were incubated at 23°C for 2 h and subsequently washed three times with the ELISA buffer. The captured immune complexes were further incubated with streptavidin-HRP conjugate (0.l µg/ml), washed, and colorimetrically developed with tetramethyl benzidine (TMB) substrate for HRP, and absorbance measurements were taken with a Molecular Devices SoftMax Pro plate reader.

Western blot analysis. Proteins were precipitated from the conditioned medium by addition of trichloroacetic acid (TCA; 7.5% final concentration) and cholic acid (0.1% final concentration) to each medium sample. The precipitated proteins were washed twice with 100% acetone to extract residual TCA and then dissolved in SDS-PAGE sample buffer. Triton X-100 detergent-extracted cell lysates and TCA-precipitated proteins from the extracellular medium were separated via SDS-PAGE electrophoresis with 12% gels and then transferred to polyvinylidene difluoride (PVDF) membranes for Western blot analysis. IL-1 was probed with the 3ZD monoclonal antibody diluted to 5 µg/ml, caspase-1 was probed with the sc-515 antibody diluted to 1 µg/ml, and P2X₇ receptor protein was probed with Alamone anti-P2X₇ antibody diluted to 2 µg/ml.

Intracellular Ca²⁺ concentration measurements. Intracellular Ca²⁺ concentration was measured with trypsinized suspensions of HEK-293 cells loaded with fura 2 and incubated in a thermostatically controlled and stirred cuvette, exactly as described previously (14). In some experiments, the nonfluorescent Ca²⁺ buffer BAPTA-AM (final concentration 30 µM) was added to cells for 30 min at 37°C before experimentation.

Lactate dehydrogenase assay. HEK-P2X₇ cells were stimulated with ATP or not and incubated at 37°C for up to 60 min as described in Assay for IL-1 release. Equivalent proportions of extracellular medium or cell lysates were assayed for lactate dehydrogenase (LDH) activity with a Cytotoxicity Detection Kit from Boehringer Mannheim. All values are expressed as LDH released as a percentage of the total LDH activity measured in combined extracellular supernatants plus whole cell lysates.

Quantitative methods. The extracellular concentration of mIL-1 release (in pg/ml) was calculated from ELISA-based standard curves with known quantities of human or murine recombinant IL-1. In some experiments, we determined the fraction of expressed mIL-1 released to the extracellular compartment during various test stimuli. The percent release was calculated by expressing the measured IL-1 content in the extracellular supernatants as a percentage of the total IL-1 protein within both the supernatant and cell lysates. As indicated in the Fig. 1 time course, we consistently noted a significant amount of immunoreactive IL-1 in the extracellular medium of transfected HEK cells (corresponding to 5–7% of the total cellular IL-1 content) at the nominal zero time (t) point of test incubations, i.e., immediately after the three washes with PBS and transfer to the test medium. This may reflect either unintended cell lysis or mechanical activation of the secretory machinery by the fluid shear stresses of the PBS washes and medium transfers. This IL-1 content at t = 0 was routinely subtracted from the total IL-1 content measured in the extracellular medium sampled at successive time points during unstimulated or stimulated test incubations.

View larger version (29K): [in this window] [in a new window]   Fig. 1. Stimulation of heterologously expressed P2X7 receptors induces mature (m) interleukin (IL)-1 secretion from HEK-293 cells. A: HEK-P2X7 cells were transiently transfected with varying concentrations of mIL-1 cDNA. The cell lysates were analyzed by Western blot probed with a monoclonal antibody to IL-1. B: kinetics of ATP-induced mIL-1 and lactate dehydrogenase (LDH) release. HEK-P2X7 cells were transfected with 0.5 µg mIL-1, and 18 h after transfection the cells were further stimulated or not with 3 mM ATP. Extracellular medium samples were collected and analyzed for IL-1 and LDH release. Left y-axis depicts % IL-1 release; right y-axis depicts % LDH release. The results are representative of 2 experiments, each performed in duplicate.

Data presentation. The pooled data for Table 1 were obtained from n 4 experiments with each experiment performed in duplicate. Except where noted, Figs. 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 show the results from single representative experiments. However, all experiments were performed two to three times with similar results.

View larger version (14K): [in this window] [in a new window]   Fig. 2. P2X7 receptor agonists and Ca2+ ionophores induce similar increases in IL-1 secretion. P2X7 receptor-overexpressing HEK-293 (HEK-P2X7) cells were transiently transfected with 0.5 µg of mIL-1, and 18 h after transfection the cells were stimulated for 30 min at 37°C with nucleotide agonists [3 mM ATP, 300 µM 2',3'-O-(4-benzoyl)benzoyl-ATP (BzATP), 100 µM UTP, and 100 µM ADP; A] or ATP (3 mM), ionomycin (1 µM) or A-23187 (1 µM) (B). mIL-1 release was assayed by ELISA protocols. Results are from 1 experiment performed in duplicate but are representative of 4–6 similar experiments.

View larger version (20K): [in this window] [in a new window]   Fig. 3. Reduction in extracellular Ca2+ strongly attenuates P2X7 receptor-induced IL-1 secretion. HEK-P2X7 cells were transfected with 0.5 µg of mIL-1. The cells were stimulated with 3 mM ATP or remained unstimulated for 30 min in the presence of varying concentrations of extracellular Ca2+. IL-1 release was measured by assaying extracellular medium samples by ELISA protocols (A) and by Western blot analysis of the trichloroacetic acid (TCA)-precipitated proteins probed for IL-1 (B).

View larger version (17K): [in this window] [in a new window]   Fig. 5. Heterologous expression of precursor IL-1 (proIL-1) and procaspase-1 proteins. HEK-P2X7 cells were transiently transfected with the indicated amounts of cDNA for procaspase-1 (A) or proIL-1 (B). The cell lysates were analyzed by Western blot probed with a polyclonal antibody to caspase-1 (A) or a monoclonal antibody to IL-1 (B).

View larger version (23K): [in this window] [in a new window]   Fig. 6. P2X7 receptor activation induces the Ca2+-dependent secretion of precursor and mature IL-1 from HEK-P2X7 cells. A and B: HEK-P2X7 cells were transfected with 0.5 µg of proIL-1. The cells were stimulated with 3 mM ATP or remained unstimulated for 30 min in the presence of varying concentrations of extracellular Ca2+. IL-1 release was measured by assaying extracellular medium samples by ELISA protocols (A) and by Western blot analysis of the TCA-precipitated proteins probed for IL-1 (B). C and D: HEK-P2X7 cells were cotransfected with 0.5 µg of proIL-1 and 0.1 µg of procaspase-1. Eighteen hours after transfection, the cells were stimulated with 3 mM ATP or remained unstimulated for 30 min in the presence of varying concentrations of extracellular Ca2+. IL-1 release was measured by assaying extracellular medium samples by ELISA protocols (C) and by Western blot analysis of the TCA-precipitated proteins probed for IL-1 (D, top). The cell lysates were also assayed by Western blots probed with anti-IL-1 (D, bottom).

View larger version (13K): [in this window] [in a new window]   Fig. 8. Increased rates of basal IL-1 secretion from HEK-P2X7 cells vs. HEK-WT cells. A: HEK-P2X7 cells and HEK-WT cells were transfected with 0.5 µg of mIL-1 cDNA. Fourteen hours after transfection, 100 µM oxidized ATP (oATP) was added to a subset of wells on the transfected 12-well plate. Eighteen hours after transfection, the extracellular medium conditioned for 18 h was collected and assayed for IL-1 release. The results are representative of 2 experiments performed in triplicate. B: the oATP-treated HEK-P2X7 cells described in A were washed, and fresh Ca2+-Mg2+ HEPES-buffered saline (HBS) was added to the cells. The cells were further stimulated with 3 mM ATP or not for 30 min at 37°C. The results were assayed by ELISA protocols.

View larger version (21K): [in this window] [in a new window]   Fig. 9. IL-1 secretion in THP-1 monocytes is Ca2+ dependent. THP-1 monocytes were differentiated with 1,000 U/ml IFN (2 days) and 100 ng/ml lipopolysaccharide (LPS; 4 h) A: cells were preloaded with 30 µM BAPTA-AM for 30 min at 37°C or not, washed, bathed in Ca2+-Mg2+-HBS or Ca2+-free Mg2+-HBS, and then stimulated with or without 3 mM ATP for 30 min. IL-1 release was measured by assaying the extracellular medium with human IL-1 ELISA protocols. The results represent data from n = 4 experiments. B: representative Western blot analysis of the cell lysates (bottom) or the TCA-precipitated proteins from the extracellular medium samples (top) from THP-1 monocytes.

View larger version (14K): [in this window] [in a new window]   Fig. 10. IL-1 secretion in Bac1 murine macrophages is Ca2+ dependent. Bac1 murine macrophages were primed with 1 µg/ml LPS (4 h) A: cells were preloaded with 30 µM BAPTA-AM for 30 min at 37°C or not, washed, bathed in Ca2+-Mg2+-HBS or Ca2+-free Mg2+-HBS, and further stimulated with or without 3 mM ATP for 30 min. IL-1 release was measured by assaying the extracellular medium with murine IL-1 ELISA protocols. The results represent data from n = 4 experiments. B: representative Western blot analysis of the cell lysates (bottom) or the TCA-precipitated proteins from the extracellular medium samples (top).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Stimulation of heterologously expressed P2X₇ receptors induces mIL-1 secretion from HEK-293 cells. To determine the specific role of P2X₇ receptors in the secretion of mIL-1, cDNA encoding mIL-1 as a fully processed and biologically active 17-kDa cytokine was transiently expressed in an HEK-293 line engineered to stably express the human P2X₇ receptor. These HEK-P2X₇ cells were transiently transfected with various concentrations of mIL-1 cDNA, incubated for 18 h, and then extracted for Western blot analysis. Analysis of the cell lysates verified that these cells do not natively express mIL-1 and that the cellular levels of mIL-1 could be readily manipulated (Fig. 1A). Similar transient transfections of HEK-P2X₇ cells with cDNA encoding green fluorescent protein (GFP) indicated 30–50% transfection efficiency (data not shown). Timed incubations of these IL-1-expressing cells in fresh medium revealed a slow basal release of IL-1 that was markedly accelerated in the presence of 3 mM extracellular ATP (Fig. 1B). Concentration-response analyses indicated that 3 mM ATP was an optimal concentration to induce the accelerated IL-1 secretion (data not shown). This requirement for millimolar ATP is consistent with the unusual agonist pharmacology of P2X₇ receptors (21) and with previous analyses of P2X₇ receptor-dependent IL-1 secretion from monocyte/macrophages that natively express this receptor (6, 10, 25). As noted in MATERIALS AND METHODS, these timed incubation experiments revealed that HEK-P2X₇ cells immediately released 7% of their total IL-1 content in response to the washing and medium exchange transfers that immediately preceded the test incubations. This "zero-time" release may reflect acute cell lysis and/or a transient increase in basal secretion by the mechanical shear stresses that accompany medium replacement. The initial response to ATP was a very rapidly induced (within the first minute) release of mIL-1 that amounted to 5% of the total IL-1 content measured at t = 0 (Fig. 1B). This initial burst of IL-1 release was followed by a 4- to 5-min plateau phase that preceded a sustained phase of IL-1 secretion over the next 20–30 min. By 30 min, the stimulated IL-1 secretion reached a near maximal value corresponding to 28% of the total expressed IL-1 (after subtraction of the zero-time IL-1 signal). In contrast, IL-1 release from the unstimulated cells increased by only 3% over this same 30-min interval. This ninefold difference in IL-1 release between ATP-stimulated and unstimulated cells was not correlated with cell lysis as measured by release of cytosolic LDH. The analysis of LDH activity shown in Fig. 1B revealed similar increases in extracellular LDH (2% of total cellular LDH) in control and ATP-stimulated cells during the initial 30 min of incubation. As with IL-1, extracellular LDH (corresponding to 3% of the total content) was present even in the zero-time samples after medium replacement. These results suggest that minimal cell lysis occurs during the initial 30 min of ATP stimulation and cannot explain the elevated rate of IL-1 release. In contrast, the additional small increase in IL-1 release observed when the ATP stimulus was prolonged from 30 to 60 min was correlated with a modest increase in extracellular LDH during this time interval. These data are consistent with, and comparable to, previous findings in which P2X₇ receptor-activated THP-1 monocytes and mouse microglial cells released 30% of total IL-1 and <5% of total LDH in 30 min (18, 29).

P2X₇ receptor agonists and Ca²⁺ ionophores induce similar increases in IL-1 secretion. P2X₇ receptors have high selectivity for ATP, with most other physiological nucleotides having little or no activity. However, BzATP is a synthetic nucleotide analog that is 10-fold more potent than ATP in activating these receptors (10, 22). To verify that the ATP-stimulated IL-1 secretion was due to the heterologously expressed P2X₇ receptor system, HEK-P2X₇ cells were also challenged with BzATP. Figure 2A shows that 300 µM BzATP stimulated IL-1 release to the same extent (8- to 10-fold) as 3 mM ATP. In contrast, 100 µM UTP and 100 µM ADP, which act as maximally efficacious agonists for the P2Y₂ and P2Y₁ receptors that are natively expressed by HEK-293 cells (30), induced only twofold increases in IL-1 secretion (Fig. 2A) during similar 30-min test incubations. These increases in IL-1 release triggered by the Ca²⁺-mobilizing P2Y receptors were similar in magnitude to the early burst of cytokine secretion observed during the initial 60 s of P2X₇ receptor activation (Fig. 1B). Increasing the test concentrations of ADP or UTP to 1 mM did not induce further increases in IL-1 secretion (data not shown). To address the role of increased cytosolic Ca²⁺ in IL-1 secretion, we compared IL-1 secretion from HEK-P2X₇ cells stimulated with ATP (3 mM), ionomycin (1 µM), or A-23187 (1 µM). Figure 2B shows that the A-23187-stimulated release of mIL-1 (3,000 pg · ml^-1 · 30 min^-1) was comparable in magnitude to that produced by P2X₇ receptor activation (3,200 pg · ml^-1 · 30 min^-1) whereas ionomycin was a less efficacious secretagogue (2,000 pg · ml^-1 · 30 min^-1). Elevating the ionomycin concentration to 10 µM did not induce a further increase in IL-1 release (data not shown).

Reduction of extracellular Ca²⁺ strongly attenuates P2X₇ receptor-induced IL-1 secretion. To characterize the role of Ca²⁺ influx in the IL-1 release response to P2X₇ receptor activation, HEK-P2X₇ cells were bathed in test media containing increasing concentrations (0, 0.3, 0.6, and 1.0 mM) of added extracellular CaCl₂. The total concentration of extracellular divalent cations was maintained at 2 mM by replacing the CaCl₂ with equimolar MgCl₂. Consistent with the function of P2X₇ receptors as ligand-gated Ca²⁺ channels, the magnitude of ATP-stimulated mIL-1 secretion was progressively reduced as the extracellular Ca²⁺ was decreased from the normal level of 1 mM down to no added Ca²⁺ (10 µM background; Fig. 3A). In the nominally Ca²⁺-free medium, ATP-induced IL-1 release was reduced to near-basal levels (450 pg · ml^-1 · 30 min^-1). Western blot analysis of the same medium samples verified that this progressive decrease in the ATP-induced release of ELISA-reactive IL-1 protein was due to reduced extracellular levels of the 17-kDa mIL-1 (Fig. 3B). We also noted a progressive decrease in the basal rate of IL-1 secretion from these HEK-P2X₇ cells as the extracellular Ca²⁺ levels in the bathing medium were reduced. It should be stressed that reduction or elimination of extracellular Ca²⁺ does not attenuate the ability of ATP to activate P2X₇ receptors (22, 23). Indeed, we observed that 3 mM ATP triggered a similar release of intracellular K⁺ from HEK-P2X₇ cells incubated in Ca²⁺-free or Ca²⁺-containing test media (data not shown). To rule out the possibility that increased Mg²⁺ (2 mM Mg²⁺ in the Ca²⁺-free HBS vs. 1 mM Mg²⁺ in standard HBS) rather than decreased Ca²⁺ is affecting either P2X₇ receptors or ATP-induced IL-1 release process, we performed IL-1 release studies in Ca²⁺-free HBS containing the usual 1 mM Mg²⁺ and observed similar inhibition of IL-1 release (data not shown). Moreover, the use of 3 mM ATP to maximally activate P2X₇ receptors minimized changes in receptor activity due to alterations in total divalent cation concentration.

Effects of cytosolic Ca²⁺ buffering on the kinetics of P2X₇ receptor-induced IL-1 secretion. To further investigate the temporal relationship between increased cytosolic Ca²⁺ and the stimulation of IL-1 release during P2X₇ receptor activation, HEK-P2X₇ cells transfected with mIL-1 were additionally loaded with 30 µM BAPTA-AM to buffer rapid changes in cytosolic free Ca²⁺. Figure 4, A and B, shows that the UTP- and ADP-induced Ca²⁺ transients mediated by the Ca²⁺-mobilizing P2Y receptors were effectively eliminated by BAPTA loading. BAPTA loading also significantly retarded the initial rate at which cytosolic Ca²⁺ increased in response to P2X₇ receptor activation but was unable to prevent the sustained increase in Ca²⁺ that accompanied stimulation of these receptors. This indicates that the influx of extracellular Ca²⁺ through the nondesensitizing P2X₇ receptor channels rapidly overwhelms the buffering capacity of BAPTA. Comparison of the kinetics of ATP-induced IL-1 release in control vs. BAPTA-loaded cells (Fig. 4C) revealed complete inhibition of the initial, small-magnitude burst of IL-1 release (measured at 1 min) but only modest reduction in the extent of cytokine secretion during the second, sustained phase of IL-1 export. A minor diminution of basal IL-1 release was also observed in the BAPTA-loaded cells compared with their unbuffered counterparts (Fig. 4C).

P2X₇ receptor activation induces Ca²⁺-dependent secretion of precursor cytokine proIL-1 from HEK-P2X₇ cells. Previous studies reported the selective release of mIL-1 but not proIL-1 from murine P388D₁ macrophages transfected with IL-1 expression plasmids and then stimulated with Ca²⁺ ionophores (31, 32). We designed similar experiments to test whether the Ca²⁺-dependent IL-1 secretion machinery observed in our HEK-P2X₇ model system could effectively discriminate between the mature and precursor forms of the cytokine. Western blots of cell lysates from HEK-P2X₇ cells transfected with increasing amounts of human proIL-1 or human procaspase-1 cDNA confirmed that these cells do not natively express proIL-1 or its processing enzyme but can rapidly accumulate large amounts of these proteins within 18 h after transfection (Fig. 5). We transiently expressed proIL-1 in HEK-P2X₇ cells in the absence (Fig. 6, A and B) or the presence (Fig. 6, C and D) of cotransfected caspase-1 and then measured basal vs. ATP-stimulated IL-1 secretion in the presence of various concentrations of extracellular Ca²⁺ as described above. Both ELISA (Fig. 6A) and Western blot (Fig. 6B) analysis indicated that the P2X₇-induced IL-1 secretion response was not specific for mIL-1 in that proIL-1 was also released in a Ca²⁺-dependent manner. It is important to note that the antibodies used for the ELISA measurements recognize both proIL-1 and mature IL-1 but have a significantly higher affinity for mIL-1. Under routine conditions, the raw ELISA data [absorbance at 450 nm (Abs₄₅₀)] were calibrated relative to ELISA signals from standard amounts of recombinant mature IL-1. Because of the lack of recombinant proIL-1 standards for calibration of the ELISA in terms of proIL-1 mass, the ELISA-based secretion data in Fig. 6, A and C, are expressed as relative Abs₄₅₀ units.

Caspase-1-mediated processing of proIL-1 to mIL-1 is independent of P2X₇ receptor-induced changes in cytosolic Ca²⁺. The experiments in Fig. 6, A and B, indicated that the Ca²⁺-dependent secretion machinery triggered by P2X₇ receptor activation can readily recognize and export the precursor form of IL-1 in the absence of competition from mIL-1. Previous studies showed that ectopic expression of procaspase-1 in HEK-293 cells induces autocatalytic generation of the active form of this protease, which then facilitates processing of coexpressed proIL-1 to mIL-1 (20). Accordingly, HEK-P2X₇ cells were cotransfected with proIL-1 and procaspase-1 and then stimulated with 3 mM ATP under the same conditions described for the experiments in Fig. 6, A and B. Significantly, these cells secreted approximately equal amounts of mature IL-1 and proIL-1 when stimulated in the standard medium containing 1 mM extracellular Ca²⁺ (Fig. 6D). The similar extracellular levels of proIL-1 and mature IL-1 contrasted with the high ratio of intracellular proIL-1 to mature IL-1 observed in the corresponding cell lysates. This suggests that the secretory machinery can very efficiently recognize and export mIL-1 as it is generated via the caspase-1-mediated processing of the much more abundant proIL-1 precursor. When these proIL-1- and caspase-1-expressing HEK-P2X₇ cells were challenged with ATP in media containing progressively reduced levels of extracellular Ca²⁺, the export of both mIL-1 and proIL-1 decreased in parallel whereas the cell lysates were characterized by a progressive accumulation of intracellular mIL-1 (Fig. 6D). This enhanced accumulation of intracellular mIL-1 despite the reduced export of mIL-1 indicated that the ability of caspase-1 to convert proIL-1 to the readily released mIL-1 was not attenuated by the lowering of extracellular Ca²⁺. This suggests a unique role for Ca²⁺ in the secretion of IL-1 (mature or precursor) but not in the upstream processing of proIL-1 that is catalyzed by active caspase-1.

Expression of P2X₇ receptors enhances IL-1 release in absence of receptor activation by exogenous nucleotide agonists. In the course of optimizing the HEK-P2X₇ cell system for investigation of P2X₇ receptor-dependent IL-1 secretion, we also analyzed the secretion of transfected IL-1 from control HEK-293 cells that lack P2X₇ receptor expression. Figure 7A shows that these wild-type cells (HEK-WT), as well as their P2X₇ receptor-expressing counterparts, accumulate similar levels of mIL-1 or caspase-1 when identically transfected with IL-1 or caspase-1 expression plasmids. Figure 7B compares the basal, ATP-stimulated, and Ca²⁺ ionophore-stimulated IL-1 secretion from HEK-WT and HEK-P2X₇ cells measured during identical 30-min incubations in standard test medium containing 1 mM Ca²⁺. As expected, stimulation of the HEK-WT cells with ATP did not increase mIL-1 export relative to the basal secretion rate observed in unstimulated cells. However, these comparative analyses revealed that the basal rate of mIL-1 release from the HEK-WT cells (120 pg · ml^-1 · 30 min^-1) was fourfold lower than basal secretion measured in the HEK-P2X₇ cells (450 pg · ml^-1 · 30 min^-1). Although stimulation of the HEK-WT cells with ionomycin or A-23187 elicited >10-fold increases in IL-1 secretion, the absolute magnitude of the ionophore-induced effects in these cells was 30–40% lower than corresponding responses to Ca²⁺ ionophores observed in the HEK-P2X₇ cells. These results indicated that the Ca²⁺-dependent machinery used for the secretion of mIL-1 is a general phenotypic characteristic of HEK-293 cells regardless of the presence or absence of P2X₇ receptors. However, these experiments also suggested that the presence of P2X₇ receptors, even in the absence of deliberate activation by exogenously added ATP, provides a stimulatory signal to the IL-1 export machinery that mediates basal or constitutive secretion of this cytokine. To further analyze this apparent role of P2X₇ receptors in constitutive IL-1 export, we compared the absolute amounts of mIL-1 present in the tissue culture media conditioned by either HEK-WT cells or HEK-P2X₇ cells during the 18-h incubations after transfection with mIL-1 expression vectors. Figure 8A shows that HEK-P2X₇ cells constitutively secreted threefold more IL-1 than HEK-WT cells during this 18-h test period. To test whether this increased rate of constitutive IL-1 secretion might reflect a basal or constitutive activity of the expressed P2X₇ receptors, parallel wells of the transfected cells (both HEK-WT and HEK-P2X₇) were supplemented with the covalent P2X₇ antagonist oxidized ATP (oATP) during the last 4 h of the 18-h medium-conditioning incubation (Fig. 8A). Inclusion of oATP in the culture medium did not affect the overall extent of mIL-1 constitutively secreted from the HEK-WT cells but produced an 40% decrease in the amount of IL-1 present in the medium conditioned by the HEK-P2X₇ cells. To verify that the P2X₇ receptors were maximally and irreversibly inhibited by this oATP treatment protocol, wells containing untreated or oATP-treated HEK-P2X₇ cells were supplemented with freshly added HBS test medium after removal of the conditioned culture medium and then challenged with or without 3 mM ATP for an additional 30 min (Fig. 8B). Although ATP stimulation produced the usual greater than eightfold increase in IL-1 release from the untreated HEK-P2X₇ cells, the oATP-treated cells showed no increase in IL-1 export in response to ATP. Together, the data presented in Figs. 7 and 8 suggest a role for P2X₇ receptors as enhancers of IL-1 secretion independent of their activation by exogenously added agonist. The nature of this nominally constitutive action of expressed P2X₇ receptors remains to be determined.

THP-1 human monocytes secrete IL-1 via Ca²⁺-dependent mechanisms. We used THP-1 human monocytic leukemia cells as a model system to characterize the role of Ca²⁺ in P2X₇ receptor-dependent IL-1 release from cells that natively express and secrete this cytokine. THP-1 cells were differentiated for 2 days with 1,000 U/ml IFN- to upregulate P2X₇ receptor expression (12) and then primed for 4 h with 100 ng/ml LPS to induce expression of proIL-1. Stimulation of these primed THP-1 monocytes with ATP induced a >10-fold increase in IL-1 release when assayed in standard Ca²⁺-containing saline (Fig. 9A and Table 1). Removal of extracellular CaCl₂ from the test medium reduced this ATP-stimulated IL-1 secretion by 85%, whereas buffering of intracellular Ca²⁺ with 30 µM BAPTA-AM produced a 50% decrease in IL-1 release. When combined, removal of extracellular Ca²⁺ plus BAPTA loading inhibited ATP-induced IL-1 secretion by 90%. Western blot analysis of the TCA-precipitated extracellular medium indicated that ATP induced release of both proIL-1 and mIL-1 and that removal of extracellular Ca²⁺ repressed the secretion of both forms of the cytokine (Fig. 9B). In contrast, BAPTA loading predominantly attenuated secretion of processed mIL-1. Western blot analysis of the cell lysates revealed an intracellular accumulation of mature IL-1 when the THP-1 monocytes were stimulated with ATP in Ca²⁺-free test saline (Fig. 9B, bottom). Both A-23187 and ionomycin acted as IL-1 secretagogues in the LPS-primed THP-1 monocytes, but neither ionophore was as efficacious as activated P2X₇ receptors (Table 1).

Bac1 murine macrophages secrete IL-1 via Ca²⁺-dependent mechanisms. Although Ca²⁺ ionophores can stimulate IL-1 secretion from LPS-primed human monocytes (34) or THP-1 cells (Table 1), previous studies indicated that such ionophores cannot potentiate IL-1 secretion from LPS-primed mouse macrophages (24). Thus the general role of Ca²⁺ in IL-1 secretion from macrophages (mouse or human) remains unclear. Moreover, specific roles for Ca²⁺ in the ability of P2X₇ receptor to stimulate IL-1 release from macrophages have not been evaluated. We previously (4, 13) used the Bac1.2F5 murine macrophage line as a model system to study multiple facets of P2X₇ receptor signaling in a macrophage background. When primed with 1 µg/ml LPS for 4 h, these cells accumulated massive amounts of proIL-1 (Fig. 10B, bottom) but secreted very minor amounts of mature IL-1 when incubated under basal conditions (Fig. 10; Table 1). Indeed, the basal rate of secretion of IL-1 from these macrophages was at least 5- to 10-fold lower than that observed in THP-1 monocytes, even though both cell types accumulated equivalent amounts of intracellular proIL-1 in response to LPS priming. Stimulation of the Bac1 macrophages with ATP in the standard Ca²⁺-containing saline elicited a >200-fold increase in IL-1 secretion (Fig. 10A; Table 1), with proIL-1 and mIL-1 being released in approximately equivalent amounts (Fig. 10B). Removal of extracellular CaCl₂ decreased the total ATP-stimulated IL-1 secretion by 60–70% (Fig. 10A; Table 1), primarily because of reduced export of mIL-1 rather than proIL-1 (Fig. 10B). Likewise, loading the Bac1 macrophages with the BAPTA Ca²⁺ buffer before ATP stimulation produced a 50–55% attenuation in total IL-1 release, with the predominant effect on secretion of the processed form of the cytokine. The combination of BAPTA loading and removal of extracellular Ca²⁺ reduced P2X₇ receptor-activated IL-1 secretion by 75%. In contrast to the HEK-P2X₇ cells (Fig. 6D) and THP-1 monocytes (Fig. 9B), inhibition of ATP-induced IL-1 secretion by removal of extracellular Ca²⁺ did not lead to an enhanced intracellular accumulation of mature IL-1 in the Bac1 macrophages (Fig. 10B, bottom).

Thus, as observed in the HEK-P2X₇ and THP-1 cell models, perturbation of Ca²⁺ homeostasis also strongly attenuates the ability of activated P2X₇ receptors to stimulate IL-1 secretion from macrophages. However, in contrast to our findings with HEK-P2X₇ cells and THP-1 monocytes but consistent with a previous study by Perregaux et al. (24), stimulation of the Bac1 macrophages with either A-23187 or ionomycin produced no detectable increase in IL-1 secretion over the very low basal rate (Table 1). This indicates that increased cytosolic Ca²⁺ per se, in the absence of the other ionic perturbations triggered by P2X₇ receptor activation, is not a sufficient signal for activation of the IL-1 export machinery in macrophages. Although the P2X₇ receptor-induced export of mature IL-1 secretion is similarly Ca²⁺ dependent in both THP-1 monocytes and Bac1 macrophages, the other ionic signals for regulating production of the mature cytokine clearly differ between these inflammatory cell types. We tested this by monitoring ATP-induced IL-1 secretion from THP-1 or Bac1 cells under conditions that eliminate the normal transmembrane Na⁺ and K⁺ gradients (by replacing the 130 mM NaCl in the standard test saline with equimolar KCl). In this modified saline, activation of the nonselective P2X₇ cation channels does not result in significantly decreased intracellular K⁺ or increased Na⁺ but does increase cytosolic Ca²⁺ given the presence of normal (1 mM) extracellular CaCl₂ (data not shown). Significantly, ATP-induced secretion of IL-1 was completely inhibited when Bac1 macrophages were incubated in this high-K⁺ saline but release of the cytokine from THP-1 monocytes was only slightly attenuated (Table 1). Consistent with the likely role of perturbed K⁺/Na⁺ homeostasis in the generation of mature IL-1, rather than the export of the processed cytokine, incubation of HEK-P2X₇ cells expressing mIL-1 in the high-K⁺ test saline had no inhibitory effect on P2X₇ receptor-activated IL-1 secretion in that model system (data not shown).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
In this study, we coexpressed P2X₇ receptors with the cytokine IL-1 in HEK-293 cells to determine the specific role of Ca²⁺ in P2X₇ receptor-induced IL-1 secretion. With this model system, we demonstrate that 1) activation of P2X₇ receptors can elicit secretion of IL-1 in cells other than the monocyte/macrophages in which P2X₇ receptors and IL-1 are natively expressed; 2) calcium ionophores can induce secretion but not processing of IL-1; 3) reduction of extracellular Ca²⁺ markedly retards P2X₇ receptor-induced IL-1 release; 4) cells transfected with proIL-1 also release this procytokine, whereas coexpression of proIL-1 and procaspase-1 results in the corelease of proIL-1 and mIL-1 in a Ca²⁺-dependent manner; and 5) cell surface expression of P2X₇ receptors, even in the absence of stimulation by added extracellular ATP, results in an enhanced constitutive release of IL-1. Parallel experiments with THP-1 human monocytes and Bac1.2F5 murine macrophages indicated that buffering of intracellular Ca²⁺ or removal of extracellular Ca²⁺ also significantly attenuates ATP-induced IL-1 release in cells that natively coexpress P2X₇ receptors and IL-1. Together, these data support a pivotal role for Ca²⁺ in the regulation of IL-1 secretion by both native and heterologously expressed P2X₇ receptors.

IL-1 lacks the secretory signal sequence, and the rate of secretion of this cytokine from LPS-primed macrophages is slow and inefficient until activated by a secondary stimulus such as extracellular ATP. ATP targets P2X₇ receptor cation channels that mediate influx of extracellular Na⁺ and Ca²⁺ and efflux of intracellular K⁺ (21, 33). Our studies revealed that P2X₇ receptor-dependent IL-1 secretion requires Ca²⁺ influx because reduction in extracellular Ca²⁺ levels markedly retarded the ATP-stimulated release of this cytokine. Biphasic secretion of IL-1 was observed such that a minor but rapid burst of the cytokine was observed immediately after ATP application and this was followed by a slower yet sustained rate of IL-1 release. Activation of Ca²⁺-mobilizing P2Y receptors triggered IL-1 secretion that was equal in magnitude to that of the initial burst observed with P2X₇ receptor activation. Moreover, the total amount of secreted mIL-1 was reduced in cells loaded with the Ca²⁺-chelating BAPTA buffer, such that the initial burst of IL-1 secretion was abolished although the slower and more sustained phase was maintained under these conditions. Activation of P2X₇ receptors induced rapid and sustained elevation of cytosolic Ca²⁺, whereas active P2Y receptors were unable to sustain the elevated cytosolic Ca²⁺ and an initial lag was observed in the cytosolic Ca²⁺ increase in BAPTA-buffered cells. Therefore, these experiments suggest that the initial magnitude and duration of stimulus-induced increases in intracellular Ca²⁺ are correlated with the rate and extent of IL-1 export.

The results from our study reveal an essential role for a sustained increase in cytosolic Ca²⁺ for the initiation of IL-1 secretion not only in our HEK-293 model system but also in cell types that natively express P2X₇ receptors such as THP-1 monocytes and Bac1.2F5 murine macrophages. P2X₇ receptor-induced changes in ionic homeostasis (Na⁺ influx, K⁺ efflux) additionally modulate proIL-1 processing (25). Efflux of intracellular K⁺ promotes procaspase-1 activation, thereby providing active caspase-1 tetramers that accelerate the rate of mIL-1 production (25). However, LPS priming of freshly isolated blood monocytes is sufficient to induce processing and secretion of IL-1, even in the absence of secondary stimuli (e.g., ATP), primarily because LPS stimulation alone activates procaspase-1 to an enzymatically active form. Activation of P2X₇ receptors further potentiates the rate of procaspase-1 activation and thereby IL-1 processing and release. We observed ATP-induced IL-1 secretion from LPS-primed THP-1 monocytes regardless of whether the cells were assayed in test medium that allows for ATP-induced K⁺ efflux or not (Table 1). These data indicate that ATP-induced K⁺ efflux is not an obligatory signal for induction of IL-1 processing in THP-1 cells or blood monocytes because LPS priming of the cells is sufficient for accumulation of active caspase-1. However, P2X₇ receptor-mediated efflux of K⁺ and influx of Ca²⁺ accelerate an ongoing process of cytokine production and secretion.

In contrast to blood monocytes or THP-1 cells, macrophages stimulated with LPS alone or in conjunction with Ca²⁺ ionophores neither process nor secrete IL-1 (Table 1). A potential difference may lie in the regulation of caspase-1 activity. LPS-primed macrophages require activation of procaspase-1 (by efflux of intracellular K⁺; Ref. 28) In macrophages, P2X₇ receptor stimulation both activates procaspase-1 by altering cytosolic K⁺ concentrations and signals to the secretory machinery by increasing cytosolic Ca²⁺. Ca²⁺ ionophores alone are insufficient to induce IL-1 processing, because the loss of intracellular K⁺ does not occur under these conditions. Interestingly, proIL-1 was not secreted from these cells on stimulation with the Ca²⁺ ionophores whereas activation of P2X₇ receptors with ATP induced release of proIL-1 and mIL-1 (Table 1, Fig. 10). This suggests that macrophages contain an additional level of regulation of IL-1 secretion that prevents release of the procytokine without procaspase-1 activation. Consistent with this model are the previous observations by Siders et al. (31, 32) that transfection of precursor or mature IL-1 into P388D₁ macrophages induced selective release of mIL-1 only and not proIL-1. In contrast, our results in a non-macrophage cell type (HEK-P2X₇ cells) demonstrate ATP-mediated, Ca²⁺-dependent release of proIL-1. In light of these findings, we conclude that, unlike HEK cells or fresh blood monocytes, macrophages actively inhibit proIL-1 secretion.

One potential mechanism for a macrophage-specific regulation of IL-1 release process might involve a recently identified complex of proteins termed the "inflammasome" by Martinon et al. (19). Among the multiple proteins comprising the inflammasome are several that contain CARD interaction domains including procaspase-1. It is therefore conceivable that inflammasome formation in macrophages is the rate-limiting step for procaspase-1 activation and mIL-1 production. In such a model, K⁺ efflux triggered by activation of P2X₇ receptors may favor the formation of procaspase-1-activating inflammasome complexes.

Previous studies (5, 25, 28) of IL-1 secretion used cells that natively expressed P2X₇ receptors and IL-1 (blood monocytes and mature macrophages) or cells that transiently expressed IL-1 but had not been characterized for P2X₇ receptor expression (P388D₁ macrophages) (31, 32). Our use of HEK-293 cells to study secretion of IL-1 in the presence (HEK-P2X₇ cells) or absence (HEK-WT cells) of P2X₇ receptors has revealed yet another mechanism by which P2X₇ receptors enhance IL-1 secretion. Comparison of basal IL-1 secretion between HEK-WT cells and HEK-P2X₇ cells indicated that HEK-P2X₇ cells secreted fourfold greater IL-1 than HEK-WT cells. This difference in the nominally basal rate of IL-1 secretion implies that P2X₇ receptors generate signals involved in IL-1 secretion even in the absence of exogenously added ATP. Moreover, preincubation of HEK-P2X₇ cells with oATP, a P2X₇ receptor antagonist, markedly reduced this constitutive release of mIL-1. oATP covalently modifies the P2X₇ receptor and presumably locks the conformation of the receptor in an inactive state. Our observations suggest that the unoccupied P2X₇ receptors may have low intrinsic activity, which is markedly enhanced by ATP binding but reversed by interactions with antagonists such as oATP. Grahames et al. (10) demonstrated that LPS-primed monocytes can also secrete IL-1 in the absence of P2X₇ receptor activation and observed that these cells did not release endogenous stores of ATP to the extracellular medium to amounts sufficient to activate P2X₇ receptors. Moreover, addition of soluble apyrase in the extracellular medium did not reduce the total IL-1 secreted from LPS-activated isolated blood monocytes (10). We also observed no effects on IL-1 secretion when soluble apyrase was added to the extracellular media of HEK-P2X₇ cells (data not shown). These data suggest that if endogenous ATP released from HEK-P2X₇ cells is acting via autocrine mechanisms to promote P2X₇ receptor activation, the ATP must be proximal and concentrated at the cell surface (2) such that it remains undetectable by soluble apyrase.

One potential mechanism for an ATP-independent but P2X₇ receptor-dependent regulation of IL-1 secretion might involve physical interaction of these receptors with the secretory apparatus by which IL-1 is released. Using HEK-293 cells expressing recombinant P2X₇ receptors, Surprenant and colleagues