Published online before print October 25, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Rheumatology and Inflammation Research, Göteborg University, Göteborg, Sweden
2 Correspondence: Department of Rheumatology and Inflammation Research, Göteborg University, Guldhedsgatan 10, 413 46 Göteborg, Sweden. E-mail: johan.bylund{at}rheuma.gu.se
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: acute-phase reactant • granulocyte • cell death • rheumatoid arthritis • nucleotide receptor
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In vitro, SAA has been ascribed numerous effects, most notably, induction of proinflammatory cytokine production [7 8 9 ] and chemotactic activity for various leukocytes [10 11 ]. Quite recently, SAA was also shown to act as an opsonin, facilitating phagocytosis of Gram-negative bacteria [12 ]. The chemotactic effect [13 ] and the cytokine production [14 ] have been shown to be mediated by the formyl peptide receptor-like 1 (FPRL1), a G-protein-coupled chemoattractant receptor similar to the FPR. FPRL1 has been shown to act as a specific receptor for numerous substances—synthetic and of microbial or endogenous origin (reviewed in ref. [15 ]). Just as FPRL1 is a promiscuous receptor that binds a variety of ligands, some of these ligands can also activate other receptors. One example, which displays large, functional similarities to SAA is the human cathelicidin LL-37. This antimicrobial peptide has been ascribed a multitude of immunomodulatory effects in addition to being bactericidal [16 ]. Although FPRL1 has been shown to mediate certain LL-37-induced effects [17 18 ], other receptors have also been implicated, such as the nucleotide receptor P2X7 [18 19 20 ], a ubiquitously expressed, ligand-gated ion channel and mediator of calcium transients and cellular activation [21 ].
In this report, we show that SAA, purified from plasma from RA patients and in recombinant form, delay spontaneous apoptosis of human neutrophils in vitro. This effect could not be abrogated by blocking FPRL1, using a specific receptor antagonist {WRWWWW (WRW4) [22 ]} or an intracellular inhibitor of FPRL1-mediated signaling {polyphosphoinositide-binding peptide 10 (PBP10) [23 ]}, indicating that the antiapoptotic effect of SAA was not mediated by FPRL1. Blocking the nucleotide receptor P2X7 using periodate-oxidized ATP (oxATP) [24 ], or Coomassie brilliant blue (CBB) G [25 ], abolished the antiapoptotic effect of SAA, suggesting that this receptor played a role in delaying apoptosis. SAA did not diminish antibody binding to P2X7, nor was SAA-mediated calcium signaling affected by oxATP, strongly implying that P2X7 was not a true SAA receptor on neutrophils. In addition, blocking P2X7 effectively counteracted the antiapoptotic effects of, e.g., LPS and GM-CSF and thus, did not inhibit the SAA effect specifically. We hypothesize that P2X7 is a general mediator of antiapoptotic signaling in neutrophils rather than being an actual receptor structure for SAA.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
0.1 ng/µg), high-density lipoprotein (HDL), apyrase, and the FITC-conjugated polyclonal 
-P2X7 antibody. Fura-2 AM was from Molecular Probes (Eugene, OR, USA), and the hexapeptide WKYMVM was synthesized and HPLC-purified by Alta Bioscience (University of Birmingham, UK). CBB G-250 was from Bio-Rad Laboratories (Sundbyberg, Sweden), RPMI-1640 medium was from PAA Laboratories (Pasching, Austria), and IL-8 ELISA was from R&D Systems Europe (UK).
Patients
After informed consent, peripheral blood was collected from patients with RA, as diagnosed according to the American College of Rheumatology’s classification criteria for RA [26
]. The study was performed in accordance with Ethical Permission No. S010-03, obtained from the Ethical Board of Göteborg University (Sweden). The Standard Clinical Immunological Laboratory (Sahlgrenska University Hospital, Sweden) determined the plasma levels of SAA by using the hSAA ELISA kit from Biosource (Nivelle, Belgium). Blood was collected in heparinized tubes and centrifuged at 1500 rpm for 7 min, and plasma was aliquoted and stored at –70°C. Plasma samples (0.1 µg total protein/sample) were separated on 15% SDS-PAGE gels under reducing conditions, transferred to polyvinylidene difluoride membranes, immunoblotted using an -SAA mAb (Abcam ab687, diluted 1:2400), followed by secondary HRP-conjugated -mouse antibody (Dako, Glostrup, Denmark; 1:1000), and developed with Vector Vip kit (Vector Laboratories Inc., Burlingame, CA, USA).
Purification of hSAA from RA patients
Purification of hSAA was performed by hydrophobic interaction chromatography, eluted with an ethanol gradient (0–80%), followed by gel filtration, essentially as described by Smith and McDonald [27
]. In short, plasma diluted with Tris-NaCl buffer (50 mM Tris-HCl, 10 mM NaCl, pH 7.6) was absorbed on 2 x 5ml HiTrapTM Octyl FF columns (GE Healthcare Bio-sciences AB, Uppsala, Sweden) and eluted with an ethanol gradient (0–80%). Elution was monitored at 280 nm, and 25 µl from peak fractions was analyzed for SAA content by immunoblotting; SAA-positive fractions were pooled, dialyzed (3500 m.w. cut-off) overnight against two changes of distilled (d)H2O, and concentrated using polyethylene glycol 20,000. The concentrate was diluted with the same amount of 20% formic acid and injected onto a 120-ml gel filtration column HiPrepTM 16/60 SephacrylTM S-200 HR, (GE Healthcare Bio-sciences AB), which had been equilibrated with two columns of 10% formic acid. Fractions (2.5 ml) eluted with 10% formic acid were collected and dialyzed against two changes of PBS and two changes of dH2O. Samples (25 µl) from each fraction were diluted in reducing sample buffer, separated on SDS-PAGE gels (15%), and silver-stained or immunoblotted with -SAA.
Separation of neutrophils from buffy coats
Neutrophils were separated from buffy coats obtained from healthy individuals using dextran sedimentation and Ficoll-Paque gradient centrifugation [28
], after which, cells were washed and resuspended in Krebs-Ringer glucose (KRG) phosphate buffer (pH 7.3) containing glucose (10 mM), Ca2+(1 mM), and Mg2+ (1.5 mM) and stored on melting ice until use.
Neutrophil culture
Neutrophils were pelleted and resuspended in RPMI 1640 [supplemented with 10% FCS and 1% penicillin/streptomycin (PEST)] to a density of 5 x 106 cells/ml. Cells (450 µl) were added to polypropylene tubes and incubated at 37°C, 5% CO2, for 30 min. Stimuli, human plasma samples, or medium were added, and the samples were incubated further for 20 h, if not stated differently, before apoptosis was assessed. Samples treated with FPRL1 inhibitors (WRW4 or PBP10) or oxATP (900 µM) were preincubated for 30 min (37°C, 5% CO2) before stimuli were added. Samples treated with CBB (10 µM) were incubated for 10 min before addition of stimuli. For induction of apoptosis, 10 µg/ml azide-free -CD95 (FAS) mAb (Nordic BioSite, Täby, Sweden) was added to cells, and SAA was added 30 min after addition of the antibody. For hydrolysis of extracellular ATP, apyrase (10 U/ml) was added to cells immediately prior to stimulation with SAA or ATP. When polymyxin B was used, the stimulus was preincubated together with polymyxin B (10 µg/ml) at room temperature for 10 min before addition to cells.
FACS analyses of cell death
Phosphatidyl serine exposure and membrane permeability
From each cultured sample, 200 µl was washed in 2 ml Annexin buffer [1 mM Hepes, 14 mM NaCl, 0.25 mM CaCl2, (pH 7.4)], and pellets were resuspended in 100 µl Annexin buffer, supplemented with 2 µl Annexin V-FLUOS. After 10 min incubation in the dark, another 400 µl Annexin buffer was added, and samples were subjected to FACS analysis using a FACScan (Becton Dickinson, Mountain View, CA, USA). When necrosis was measured, 5 µl 7-amino-actinomycin D (7-AAD) from BD Bioscience (Stockholm, Sweden) was also added to the samples before analysis. At least 10,000 events were acquired, and neutrophils were gated on the basis of side- and forward-scatter. Apoptosis was assessed on the basis of Annexin V-FLUOS fluorescence, as measured in the fluorescence 1 (FL1) channel, and necrosis was assessed on the basis of membrane permeability to 7-AAD, as measured in the FL3 channel; all data were analyzed using WinMDI 2.8 software. Typically, 20 h incubation in the absence of stimuli generated approximately 50% viable cells (Annexin V–/7-AAD–) and 50% apoptotic cells (Annexin V+/7-AAD–), whereas necrosis (Annexin V+/7-AAD+) was consistently below 5%. None of the stimuli/inhibitors used in this study altered the proportion of necrotic cells significantly, with the exception of FAS ligand (FASL) treatment. These low levels of necrosis were also corroborated by control experiments measuring the release of lactate dehydrogenase (LDH) to the supernatants using the LDH cytotoxic detection kit (Roche Diagnostics), according to the manufacturer’s instructions. Based on these low levels of necrosis, all Annexin V-positive cells are referred to as "apoptotic" unless otherwise noted.
Mitochondrial membrane potential
From each cultured sample, 200 µl was pelleted and resuspended in a 250-µl ApoAlert mitochondrial membrane sensor kit (Clontech, Mountain View, CA, USA), diluted 1:1000 in warm PBS. This fluorescent dye is incorporated in mitochondria of viable cells, where it forms aggregates exhibiting red fluorescence. Upon apoptosis, the dye cannot accumulate and aggregate in the mitochondria, as a result of depolarization of the mitochondrial membranes, and instead, the cytosolic monomers exhibit green fluorescence. After 30 min incubation at 37°C in the dark, cells were washed once, resuspended in warm PBS, and analyzed by FACS. Neutrophils were gated, and fluorescence was measured in the FL1 (green) and FL2 (red) channels. All data were analyzed using WinMDI 2.8 software.
Measurements of caspase-3 and -7 activities
Measurements of caspase-3 and -7 activities were performed with Caspase-Glo 3/7 assay (Promega, Madison, WI, USA) in white, 96-well plates in a Mithras LB 940 (Berthold Technologies, Germany) luminometer. Cells (105 cells/ml) were cultured for 4 h (as described above), and thereafter, 100 µl cells were mixed with 100 µl Caspase-Glo 3/7 substrate, and luminescence was measured after 30 min.
P2X7 expression
Purified neutrophils (106 cells) were incubated on ice for 30 min with 1 µg FITC-conjugated -P2X7 antibody or FITC-conjugated rabbit -sheep serum (Dako) as control. The cells were fixed with 2% ice-cold paraformaldehyde for 20 min and washed twice in PBS before analysis. In certain experiments, SAA (2 µM) or BzATP (1 mM) was added to cells on ice before addition of antibodies. Surface expression of P2X7 was assessed on the basis of FITC fluorescence and compared with cells stained with the control serum (background) using WinMDI 2.8 software.
Intracellular calcium measurements
Neutrophils (107/ml) were incubated in RPMI 1640 (supplemented with 10% FCS and 1% PEST) in the presence or absence of 900 µM oxATP at 37°C. After 2 h, cells were transferred to KRG buffer without calcium, supplemented with 0.1% BSA, and labeled with the fluorescent calcium indicator Fura-2 AM (2 µM) for 30 min at room temperature as described previously [29
]. In micro cuvettes, samples with 106 cells in KRG were equilibrated at 37°C for 3 min before addition of stimuli. The FPRL1 antagonist WRW4 (5 µM) was added to cells just before the 3-min equilibration. Fluorescence was followed (excitation at 340 and 380 nm; emission at 510 nm) using a luminescence spectrometer LS 50B (Perkin Elmer Corp., Wellesley, MA, USA), and intracellular calcium levels were plotted as the fluorescent ratio (340 nm/380 nm).
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (27K): [in a new window] |
Figure 1. Antiapoptotic effects of plasma from RA patients. (A) Immunoblot of plasma samples using an -SAA antibody showing highly elevated SAA levels in circulation of Patient #1 (P#1); positions of molecular weight (kD) markers are shown. (B) Human neutrophils were incubated for 20 h with indicated concentrations of plasma obtained from the two RA patients (Patient #1, gray bars; and Patient #5, open bars) and one healthy control donor (Ctrl., black bars), after which, apoptosis was quantified by Annexin V staining and compared with spontaneous apoptosis. Plasma from Patient #1 was more potently antiapoptotic than the other plasma samples. The graph depicts mean + SD from four experiments with cells from different, independent donors, and statistical analysis was performed using paired t-test, Patient #1 versus Patient #5; *, P < 0.05.
|
Purified SAA from patient plasma is antiapoptotic
Obviously, human plasma contains a wide variety of molecules that could potentially affect neutrophil apoptosis, and in an attempt to purify SAA from plasma of patients with RA, we used ethanol-eluted, hydrophobic interaction chromatography followed by gel filtration as described previously [27
]. After gel filtration, the protein content of the fractions was analyzed by SDS-PAGE and silver staining (Fig. 2A
), and fractions were also added to neutrophils in culture to assess antiapoptotic potential. Several fractions surrounding #41, delayed spontaneous neutrophil apoptosis, as determined by Annexin V staining (Fig. 2B)
; Fraction #41 also contained a protein band of approximately 12.5 kD on a silver-stained gel (Fig. 2A)
. Immunoblotting fractions obtained after gel filtration confirmed that the 12.5-kD band was indeed SAA and that this protein was also present in several fractions surrounding #41 (Fig. 2C)
.
 View larger version (23K): [in a new window] |
Figure 2. Antiapoptotic effect of SAA purified from RA plasma. SAA rich plasma from RA patients was subjected to hydrophobic interaction chromatography, followed by gel filtration, monitored at 280 nm. Certain fractions (indicated with a line in the elution profile) were separated on SDS-PAGE gels alongside rhSAA, silver-stained (A), and assessed for antiapoptotic activity on human neutrophils using Annexin V staining (B). Fraction #41 contained SAA and inhibited spontaneous apoptosis (Control). Fractions from another gel filtration were immunoblotted with the -SAA antibody and revealed SAA in fractions surrounding #41 (C). Figure shows data from two independent experiments with similar results.
|
 View larger version (19K): [in a new window] |
Figure 3. The antiapoptotic effects of rSAA, as measured by Annexin V staining after 20 h, was dose-dependent with an EC50 in the nanomolar range (A). The antiapoptotic effect of SAA (2 µM) was also seen using mitochondrial membrane potential to monitor apoptosis after 20 h (B). The fluorescent dye is incorporated in mitochondria of viable cells, where it forms aggregates exhibiting red fluorescence. Mitochondria of apoptotic cells are depolarized, preventing aggregation of the dye; the cytosolic dye in the monomeric form instead exhibits green fluorescence. SAA-induced antiapoptosis, as measured by Annexin V staining, was unaffected by polymyxin B (PxB; 10 µg/ml; C), and this concentration of polymyxin B diminished the antiapoptotic effect of LPS (100 ng/ml) potently. The active concentration range of SAA in terms of delaying neutrophil apoptosis coincided with the concentration range needed to induce secretion of IL-8 to the supernatant after 20 h (D). Shown are mean ± SD (A, n=5; D, n=3) with statistical analyses performed using paired t-tests for each concentration compared with untreated cells (**, P<0.005; ***, P<0.0005) or representative FACS plots from four (B and C) independent experiments.
|
We next corroborated the findings that SAA mediate IL-8 release by measuring the amounts of IL-8 in the cell-free supernatants obtained after neutrophil culture (Fig. 3D) . The IL-8 released after SAA stimulation seemed not to be responsible for the SAA-induced antiapoptotic effects, as addition of rIL-8 at concentrations up to 100 ng/ml (10 times more than the maximal IL-8 secretion induced by SAA) did not decrease apoptosis in our system (not shown).
The antiapoptotic effect of SAA does not involve FPRL1
SAA reportedly mediate its chemotactic activity through the chemoattractant receptor FPRL1 in monocytes and transfected human embryo kidney 293 cells [13
]. FPRL1 has also been implicated in SAA-induced IL-8 secretion from neutrophils, a response correlating with intracellular calcium signaling and activation of the transcription factor NF-
B [14
]. To investigate the possible involvement of FPRL1 in the antiapoptotic effect of SAA, we used two different inhibitors of this receptor. The specific FPRL1 antagonist WRW4 [22
] did not affect spontaneous apoptosis, and SAA inhibited neutrophil apoptosis to a similar extent, regardless of the presence of WRW4 (Fig. 4A
). The cell-permeable peptide PBP10 has been shown to be a potent blocker of FPRL1-mediated neutrophil activation [23
] but was unable to block SAA-induced antiapoptosis (Fig. 4A)
. Taken together, these data indicated that FPRL1 was not responsible for mediating the antiapoptotic effects of SAA. In line with this, the specific FPRL1 agonist WKYMVM [30
], at concentrations ranging from 10–10 M to 10–6 M, had no effect on neutrophil apoptosis (data not shown), further supporting the notion that stimulation of FPRL1 did not constitute an antiapoptotic signal. Moreover, WRW4 blocked intracellular calcium signaling completely in WKYMVM-stimulated neutrophils (Fig. 4B)
, showing that the concentration used in our apoptosis set-up (5 µM) was indeed effective in blocking signaling through FPRL1.
 View larger version (20K): [in a new window] |
Figure 4. Effects of FPRL1 inhibitors. (A) Neutrophils were preincubated for 30 min with medium (Ctrl), the FPRL1 antagonist WRW4 (5 µM), or the intracellular inhibitor of FPRL1-mediated signaling PBP10 (1 µM) and then cultured in the absence (open bars) or presence of SAA (2 µM; solid bars) for 20 h. Neither of the FPRL1 inhibitors diminished the antiapoptotic effect of SAA. (B) The FPRL1 antagonist WRW4 (5 µM) completely blocked calcium signaling in neutrophils stimulated with the FPRL1 agonist WKYMVM (0.1 µM). Shown are (A) mean + SEM (*, statistically significant differences; P<0.05, as analyzed by paired t-tests comparing each inhibitor in the presence or absence of SAA stimulation; n=5) and (B) one representative experiment out of 10; arrow indicates addition of WKYMVM.
|
-FAS antibody increased neutrophil cell death, apoptosis and necrosis, potently (Fig. 5
). When SAA was added to cells 30 min after FAS treatment, apoptosis and necrosis were reduced after an additional 20 h in culture (Fig. 5)
. These data indicate that SAA decreased apoptosis effectively, not only when cell death occurred spontaneously but also when apoptosis was actively induced.
 View larger version (18K): [in a new window] |
Figure 5. SAA counteracts the proapoptotic effect of FAS ligation. Apoptosis was induced by incubating cells in the presence of -FAS (10 µg/ml) for 30 min before addition of SAA (2 µM; right panel) or medium (left panel), and incubation continued for 20 h before cell death was assessed by Annexin V in combination with 7-AAD. -FAS induced apoptosis (Annexin V+/7-AAD–) and necrosis (Annexin V+/7-AAD+); SAA decreased both types of cell death. The graph depicts mean + SD (n=3), and the FACS plots are representative from five repeats with cells from three different donors.
|
-P2X7 antibody and found that this receptor was indeed expressed on the surface of these cells (Fig. 6A
). Next, we used periodate oxATP, a Schiff-base forming reagent that has been shown to exhibit antagonism toward P2X7 without affinity for related P2Y receptors [31
]. We pretreated neutrophils with oxATP before stimulation with SAA and found that the P2X7 antagonist abrogated the antiapoptotic effect of SAA completely (Fig. 6B)
. This implied that SAA mediated its antiapoptotic effects directly by ligating P2X7 or that the antiapoptotic signal required a functional P2X7.
 View larger version (20K): [in a new window] |
Figure 6. Expression of P2X7 and the impact of oxATP on the antiapoptotic effect of SAA. Purified neutrophils were assayed for surface expression of P2X7 (A); a representative histogram from three independent experiments is shown. (B) Preincubation of cells in the presence of oxATP (900 µM) before cultivation with medium (left panel) or SAA (2 µM; right panel). oxATP blocked the antiapoptotic effects of SAA without affecting necrosis, as measured by Annexin V staining in combination with 7-AAD. Representative plots from five experiments with cells from three independent donors are shown.
|
 View larger version (12K): [in a new window] |
Figure 7. P2X7-induced calcium flux: the effects of oxATP on calcium signaling and competition for antibody binding to P2X7. (A) Neutrophils were preincubated with medium (solid lines) or oxATP (900 µM; dashed lines), labeled with Fura-2 AM, and stimulated with the specific P2X7 agonist BzATP (left; 400 µM) or SAA (right, 10 µM). Intracellular calcium concentrations were followed kinetically and are expressed as fluorescent ratio (340/380 nm). Blocking P2X7 with oxATP abrogated BzATP-induced calcium transients completely but had no effect on SAA stimulation. Arrows indicate addition of stimuli; curves are representative experiments out of at least three. (B) Cells were incubated on ice in the presence of buffer (filled histogram), BzATP (1 mM; gray line), or SAA (2 µM; black line) before addition of P2X7 antibody to all samples. BzATP potently decreased antibody binding to P2X7, whereas SAA did not; one representative experiment out of three is shown.
|
 View larger version (19K): [in a new window] |
Figure 8. Impact of P2X7 blocking with respect to antiapoptotic signaling and the role of extracellular ATP. P2X7 was blocked by preincubation of cells with (A) oxATP (900 µM for 30 min) or (B) CBB (10 µM for 10 min) prior to the addition of medium (spontaneous) or SAA (2 µM), LPS (100 ng/ml), or GM-CSF (100 ng/ml) for 20 h, after which, apoptosis was quantified on the basis of Annexin V binding. The blocking of P2X7 inhibited antiapoptotic activity in general. (C) Cultured neutrophils in the presence (solid bars) or absence (open bars) of apyrase (10 U/ml) were stimulated with SAA (2 µM) or ATP (5 mM) and incubated for 20 h, after which, apoptosis was assessed by Annexin V binding. Apyrase decreased the antiapoptotic effect of ATP but not by SAA. Graphs depict mean + SD with statistical analyses performed using paired t-tests to compare each antiapoptotic factor in the presence or absence of inhibitor; ns, not significant. *, P < 0.05 (n=3–5, A; n=3, B; n=5, C).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
B [39
]. In addition, Raynes and co-workers [12] demonstrated that SAA, also when complexed to HDL, can bind Gram-negative bacteria and function as an opsonin that facilitates phagocytosis by neutrophils and macrophages. We investigated whether HDL was able to block the antiapoptotic effect of SAA by mixing the two substances before addition to cells. HDL (up to 100 µg/ml) did not block the antiapoptotic effect of 2 µM SAA in our systems. However, the HDL preparation used (from Sigma Chemical Co.) possessed neutrophil-activating potential (e.g., induction of calcium transients and chemotaxis) by itself, which suggests impurities in the preparation. Although HDL alone did not alter spontaneous apoptosis at the concentrations used, no decisive conclusions could be drawn regarding the ability of HDL to block the antiapoptotic effect of SAA. With this report, we add another important SAA feature to the list of proinflammatory functions and show that SAA has the ability to delay neutrophil apoptosis. The inhibition was manifest on cultured cells (spontaneous apoptosis) as well as on cells that were stimulated with proapoptotic stimuli (FAS-induced apoptosis). We also show that SAA suppressed caspase-3/7 activity, providing a logical link to the increased neutrophil survival.
In leukocytes, the receptor commonly agreed on to mediate SAA signaling in neutrophils is FPRL1 [13 14 42 ], a chemotactic receptor with a wide variety of agonists [15 ]. In contrast to earlier work, we found that the antiapoptotic effect of SAA was not mediated by FPRL1. However, during the preparation of this manuscript, a paper was published, suggesting that SAA may indeed mediate antiapoptotic effects via FPRL1 [43 ]. This conclusion was drawn from experiments showing that the SAA effects were inhibited partially by the receptor antagonist Boc-2. This antagonist, however, preferentially inhibits another receptor (FPR) as opposed to FPRL1, and at the high concentration used (50 µM), it also exerts inhibitory effects on other neutrophil receptors not belonging to the FPR family [44 ]. No proper conclusions can thus be drawn from these experiments regarding the identity of the receptor responsible for the SAA effects. In the current study, we used the FPRL1-specific antagonist WRW4 [22 ] and the intracellular inhibitor of FPRL1-mediated signal transduction PBP10 [23 ]. The presence of either of these molecules did not counteract the antiapoptotic action of SAA, strongly suggesting that the effect was independent of FPRL1. In the above-mentioned study, Lipoxin A4 (LXA4) counteracted the antiapoptotic effects of SAA, even when added up to 4 h after SAA [43 ]. Taken together with previous reports showing that LXA4 does not interfere with SAA receptor binding/signaling [13 ], this implies that the inhibitory effects of LXA4 are executed most probably by mechanisms other than competition for receptor binding. LXA4 has been ascribed a multitude of anti-inflammatory effects, some of which are reportedly mediated by FPRL1 [45 ], but also other receptors have been implicated in the biological activity of LXA4 [46 ]. The finding that LXA4 counteracts certain SAA effects is interesting but does not shed any direct light on the receptor used by SAA to delay neutrophil apoptosis. Regarding SAA and its potential affinity for FPRL1, we have found recently that although SAA may in fact signal through FPRL1 when expressed on transfected HL-60 cells, this is not the receptor used for mediating calcium transients or the release of ROS from primary human neutrophils [47 ]. Together with the present report, our findings heavily implicate one or several receptor(s) distinct from FPRL1 as mediators of at least certain SAA-induced neutrophil activities such as delay of apoptosis (this report), calcium signaling, and release of oxygen radicals [47 ].
Another receptor investigated in this report was the nucleotide receptor P2X7, a ubiquitous receptor, shown originally to function as a ligand-gated ion channel responsive to extracellular nucleotides such as ATP [21
]. Among leukocytes, P2X7 has been studied most extensively in monocytes/macrophages (and related cell lines), and the cellular effects induced by ATP activation include intracellular calcium transients [48
], pore formation [49
], processing various cytokines [50
], and activation of NF-B [51
]. In neutrophils, ATP stimulation has been shown to result in calcium oscillation [52
], pore formation, and ROS formation [53
]. Although ATP has the ability to bind and activate several nucleotide receptors of the P2 family [21
], a direct involvement of P2X7 in pore formation and ROS formation has been proposed [53
]. Cells obtained from P2X7-deficient mice are unresponsive to several activation features normally associated with ATP stimulation [54
], including calcium signaling [55
], implying a direct ligand receptor relationship. Recently, P2X7 was implicated as a neutrophil receptor for LL-37, mainly based on the fact that the effects of LL-37 could be counteracted by oxATP [18
19
]. In agreement with these observations, we found that oxATP blocked the antiapoptotic effect of SAA completely. However, oxATP did not block SAA-induced calcium signaling, strongly suggesting that P2X7 is not the bona fide receptor structure for SAA. In support of this, SAA was not able to decrease antibody binding to P2X7, and although this could be explained potentially by binding to different epitopes, the fact that a polyclonal antibody was used indicates that no direct interaction between SAA and P2X7 is occurring. In addition, oxATP was equally effective at blocking antiapoptotic signals generated by LPS or GM-CSF. These proinflammatory substances are well-characterized and mediate their antiapoptotic effects by activating TLR4 and GM-CSFR, respectively. It seems unlikely that P2X7 is the actual receptor for such a diverse array of proinflammatory mediators including SAA, LPS, GM-CSF, LL-37 [18
19
20
], IL-1β, and TNF- [56
]. We propose instead that the function of P2X7 is important for proinflammatory signaling in general. The finding that P2X7-deficient mice were markedly less affected than wild-type mice when used in a collagen-induced arthritis model [52
] lends support to this proposition. As so many different functions are governed by P2X7, we can, at present, only speculate about the details of this process. One possibility is that extracellular ATP is a second messenger of more profound importance than realized previously and that various inflammatory mediators induce ATP release. We added ATP to cells in culture and found that concentrations over 5 mM were needed to inhibit apoptosis (data not shown), which is far more ATP than released from stimulated neutrophils [18
]. In addition, hydrolysis of extracellular ATP by apyrase did not alter the ability of SAA to inhibit spontaneous apoptosis, making a crucial role for extracellular ATP (as an endogenous second messenger) in antiapoptotic signaling unlikely. Another possibility is that oxATP blocks some antiapoptotic signaling pathway independently of P2X7; dampening effects of oxATP on proinflammatory signaling have in fact been described in P2X7-deficient mice [57
] and in cell lines devoid of any known P2 receptors [58
]. Ongoing studies in our laboratory address these issues in more detail.
In conclusion, the present study demonstrates that the acute-phase reactant SAA, in recombinant form and purified from plasma from patients with RA, potently suppressed neutrophil apoptosis, a finding completely in line with the emerging view of SAA displaying proinflammatory cytokine-like properties. In contrast to earlier reports regarding the interaction between SAA and neutrophils, we found no evidence for the involvement of the chemotactic receptor FPRL1. We also show that the P2X7 inhibitors oxATP and CBB were effective at abrogating the antiapoptotic signal, not only when activated by SAA but also in general. We thus propose that rather than being a receptor structure for SAA, P2X7 serves a more general function in antiapoptotic signaling. Future studies hopefully will unravel the mechanistic details of this pathway.
Received May 3, 2007; revised September 12, 2007; accepted September 14, 2007.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, interleukin-1β, and interleukin-8 by human blood neutrophil Biochem. Biophys. Res. Commun. 268,405-408[CrossRef][Medline] (TNF-) in human neutrophils Immunol. Lett. 91,33-37[CrossRef][Medline]B and proinflammatory gene expression in human and murine intestinal epithelial cells Eur. J. Immunol. 35,718-726[CrossRef][Medline]B through the P2Z purinoreceptor by selectively targeting NF-B p65 J. Cell Biol. 139,1635-1643B in osteoclasts J. Bone Miner. Res. 19,642-651[CrossRef][Medline] signaling by P2 purinergic receptors in human fetal astrocytes J. Neurosci. 20,5292-5299This article has been cited by other articles:
 |
|
 |
|
  S. E. Evans, B. L. Scott, C. G. Clement, D. T. Larson, D. Kontoyiannis, R. E. Lewis, P. R. LaSala, J. Pawlik, J. W. Peterson, A. K. Chopra, et al. Stimulated Innate Resistance of Lung Epithelium Protects Mice Broadly against Bacteria and Fungi Am. J. Respir. Cell Mol. Biol., January 1, 2010; 42(1): 40 - 50. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. D. Ye, F. Boulay, J. M. Wang, C. Dahlgren, C. Gerard, M. Parmentier, C. N. Serhan, and P. M. Murphy International Union of Basic and Clinical Pharmacology. LXXIII. Nomenclature for the Formyl Peptide Receptor (FPR) Family Pharmacol. Rev., June 1, 2009; 61(2): 119 - 161. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Bjorstad, G. Askarieh, K. L. Brown, K. Christenson, H. Forsman, K. Onnheim, H.-N. Li, S. Teneberg, O. Maier, D. Hoekstra, et al. The Host Defense Peptide LL-37 Selectively Permeabilizes Apoptotic Leukocytes Antimicrob. Agents Chemother., March 1, 2009; 53(3): 1027 - 1038. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. El Kebir, L. Jozsef, and J. G. Filep Opposing regulation of neutrophil apoptosis through the formyl peptide receptor-like 1/lipoxin A4 receptor: implications for resolution of inflammation J. Leukoc. Biol., September 1, 2008; 84(3): 600 - 606. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
