Published online before print August 26, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Université du Québec, INRS-Institut Armand-Frappier, Laval, QC, Canada
1 Correspondence: INRS-Institut Armand-Frappier, 531 Boul. des Prairies, Laval, QC H7V 1B7, Canada. E-mail: denis.girard{at}iaf.inrs.ca
ABSTRACT
Arsenic trioxide (ATO) is known for treating acute promyelocytic leukemia and for inducing apoptosis and mitogen-activated protein kinases (MAPKs) in promyelocytes and cancer cells. We recently reported that ATO induces neutrophil apoptosis. The aim of this study was to establish whether or not ATO recruits MAPKs in neutrophils, as well as to further investigate its agonistic properties. We found that ATO activates p38 and that, unlike H2O2, this response was not inhibited by exogenous catalase. Also, we demonstrated that ATO-induced p38 activation occurs before H2O2 generation and without a calcium burst. We next established that ATO recruits c-jun NH2-terminal (JNK) but not extracellular signal-regulated kinase 1 and 2 (Erk-1/2). Using pharmacological inhibitors, we found that the proapoptotic activity of ATO occurs by a MAPK-independent mechanism. In contrast, the ability of ATO to enhance adhesion, migration, phagocytosis, release, and activity of gelatinase and degranulation of secretory, specific, and gelatinase, but not azurophilic granules, is dependent upon activation of p38 and/or JNK. This is the first study establishing that ATO possesses important agonistic properties in human neutrophils. Given the central role of neutrophils in various inflammatory disorders, we propose that ATO might have broader therapeutic implications in clinics, especially for regulating inflammation.
Key Words: drug • inflammation • signalization
INTRODUCTION
Arsenic trioxide (As2O3 or ATO) is currently used as a cure against acute promyelocytic leukemia [1 2 3 ]. Its great efficiency in inducing cell differentiation or apoptosis in many cancerous cells has prompted clinicians to expand its use to other types of cancers such as multiple myeloma [2 ]. The mode of action of ATO is complex, probably largely cell specific, and is not fully understood. However, it is clear that this drug affects both intrinsic and extrinsic pathways of apoptosis, leading ultimately to caspase activation and disruption of cell machinery and cytoskeleton [3 ].
The vast majority of studies aimed at elucidating the mode of action of ATO have been performed in immature cancer cells. Only a few studies have examined the effect of this drug on primary mature cells of the immune system. Arsenic has been shown to alter the differentiation of monocytes to macrophages [4 ] and B cells to plasma cells [5 ]. In addition, ATO was found to alter human macrophage functions in vitro [6 ]. Also, induction of CD4 and CD8 T-cell apoptosis has been observed in response to ATO [7 ]. Recently, we found that ATO, at clinically achievable concentrations, induces neutrophil apoptosis by a H2O2- and de novo protein synthesis-dependent mechanism, leading to caspase activation [8 ]. Also, we found that ATO triggers de novo synthesis of many polypeptides in neutrophils, including annexin-I [1 ]. This protein was originally classified as anti-inflammatory due to its ability to inhibit cellular phospholipase A2 (cPLA2) and was recently identified as a chaperone-like protein. Unfortunately, we were not able to confirm a direct involvement of this protein in the proapoptotic activity of ATO, since the potency for inducing neutrophil apoptosis was not altered using cells isolated from annexin-I KO mice [1 ]. However, we demonstrated that ATO induced gene and protein expression of some heat shock proteins.
The three major mitogen-activated proteins (MAP) kinases have been identified in neutrophils: the extracellular signal-regulated kinase (Erk) 1 and 2, the p38 kinase and the c-Jun amino-terminal kinase (JNK, also referred as a stress-activated kinase or SAPK). The Erk-1/2 MAPK is involved in growth and differentiation. p38 and JNK MAPKs are typically associated with apoptosis and cell stress responses [9
]. These kinases are part of signaling pathways regulating activation of neutrophils by classical neutrophil agonists, including fMLP, tumor necrosis factor-
(TNF-), or platelet-activating factor (PAF) [10
, 11
]. In addition, p38 and JNK have been linked with many facets involved in the inflammatory response, including cell adhesion, degranulation, chemotaxis and release of cytokines [12
, 13
].
ATO has been shown to activate all three MAPKs in a variety of cancer cells. Trivalent arsenic species inhibit dephosphorylation of JNK in HeLa cells [14 ]. In addition, ATO induced sustained phosphorylation of JNK in multiple myeloma cells such as ILKM-3 and U266, and this activation was correlated with apoptosis induction [15 ]. Phosphorylation of p38 has also been demonstrated in a variety of cancer cells, including K562, LNKAP, and MCF-7. In these cells, the addition of ascorbic acid potentiates the response of ATO by increasing the GTP-bound form of Rac1, a downstream effector of p38 [16 ]. This indicates that reactive oxygen species, especially H2O2, could be implicated in the ATO-induced phosphorylation of p38. ATO-induced Erk-1/2 phosphorylation has been reported in BEAS bronchial epithelial cells [17 ], but not in K562 cells [18 ], demonstrating that the effect of ATO on the MAPKs can vary in different cells. Indeed, the pattern of MAPK phosphorylation induced by ATO was reported to vary according to cell types, doses, and the cellular environment.
In the present study, we were interested in establishing the involvement of the MAPKs in ATO-treated human neutrophils. We postulated that H2O2 could trigger the phosphorylation of the stress kinase p38, as this relationship has been established in other systems [19 ]. This phosphorylation could, in turn, control many of the neutrophil functions. Surprisingly, our results demonstrated that ATO activated p38 in an H2O2-independent fashion. Also, JNK, but not Erk-1/2, was activated by ATO, and MAPKs are not involved in ATO-induced neutrophil apoptosis. However, we show here that ATO possesses neutrophil agonistic properties never before reported, including adhesion, migration, phagocytosis, release, and activity of gelatinase and degranulation of secretory, specific, and gelatinase, but not azurophilic granules, and that all of these functions are dependent upon activation of p38 and/or JNK. ATO should now be considered as a potent activator of the neutrophil functions. A model of activation is presented.
MATERIALS AND METHODS
Chemicals, agonists, and Abs
Arsenic trioxide (As2O3), catalase, EGTA, LPS, PD98059, SB203580, A23187, and TNF- were all purchased from Sigma-Aldrich (St. Louis, MO, USA), and technical toxaphene was from Ultra Scientific (North Kingstown, RI, USA). IL-8 was purchased from R&D Systems (Minneapolis, MN, USA) and GM-CSF was obtained from PeproTech (Rocky Hill, NJ). Anti-phosphospecific p38 (pTpY180/182), Erk-1/2 (pTpY185/187), and JNK1-2/SAPK (pTpY183/185) were from BioSource (Camarillo, CA, USA). Antibodies against the nonphosphorylated form of p38 (sc-535) were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA), polyclonal Erk-1/2 was obtained from Upstate Biotechnology (Lake Placid, NY, USA), and polyclonal SAPK/JNK was obtained from Cell Signaling Technology (Danvers, MA, USA). FITC-CD16 conjugated antibody, monoclonal CD35 (clone E11), and CD63 (clone H5C6) were purchased from BD PharMingen (San Diego, CA, USA). The anti-CD66b antibody (clone 80H3) was from AbD Serotec (Raleigh, NC, USA). Fluo-3-AM was acquired from Molecular Probes (Eugene, OR, USA), and the JNK inhibitor II was purchased from Calbiochem (Madison, WI, USA).
Neutrophil isolation
Neutrophils were isolated from venous blood of healthy volunteers by dextran sedimentation followed by centrifugation over Ficoll–Hypaque (Pharmacia Biotech, Inc., Quebec, Canada), as described previously [20
]. Blood donations were obtained from informed and consenting individuals according to institutionally approved procedures. Cell viability was monitored by trypan blue exclusion, and the purity (>98%) was verified by cytology from cytocentrifuged preparations colored by Diff-Quick staining (Fisher Scientific, Ottawa, Canada). Cell viability was systematically evaluated before and after each treatment.
Production of reactive oxygen species
Cells were washed once and resuspended in HBSS-0.1% BSA, containing 5 µM of CM-H2DCFDA (Molecular Probes, Eugene, OR, USA), at a cell density of 1 x 106 cells/ml for 15 min at 37°C. Cells were then incubated in the presence or absence of ATO from 5 to 180 min. The shift of fluorescence in FL1 was recorded using a FACScan.
Cell culture
The human lung epithelial A549 cell line was purchased from the American Type Culture Collection (Rockville, MD, USA). Cells were grown in RPMI-1640 (GIBCO BRL, Grand Island, NY, USA) supplemented with 10% FCS and antibiotics. Cell viability was systematically evaluated by trypan blue exclusion assay before and after each treatment, and mortality never exceeded 5%.
Migration assay
In vitro migration was performed in a 48-well microchemotaxis chamber (Neuro Probe, Gaithersburg, MD, USA) using a 3-µm polycarbonate membrane filter as previously published [21
]. The bottom wells were loaded with buffer or agonists (final volume of 28 µl), the membrane was placed over the wells, and the top layer of the chamber was added over the membrane. Cells [50 µl from a suspension of 1x106 cells/ml in RPMI-1640] were then added to the top chamber wells. The chamber was incubated at 37°C for 60 min in a humidified incubator in the presence of 5% CO2. After incubation, the top of the chamber was removed, and the upper side of the membrane was carefully wiped with a rubber scraper. The membrane was then fixed in methanol, colored with Diff-Quick staining kit, mounted on a glass slide, and examined under oil immersion at x400. The number of cells in five representative fields was counted, and the results were expressed as indices (number of cells from tested group/number of cells from the corresponding control). IL-8 was used as a positive control [22
].
Gelatin zymography
Neutrophils were incubated from 0 to 60 min under various conditions. Cells were then centrifuged at 13,000 rpm for 10 min, and pellets were discarded. The supernatants (10 µl corresponding to 0.1x106 cells) were mixed with a nonreducing buffer (40% glycerol, Tris-HCl 1M pH 6.8, SDS 8%), and run on 10% acrylamide gels containing 0.2% gelatin. Gels were washed for 30 min twice with 2.5% Triton X-100 and incubated overnight in digestion buffer (Tris-HCl 50 mM pH 7.4, NaCl 150 mM, CaCl2 5 mM). Gels were stained with Coomassie blue 0.1% and destained. Densitometric analysis was performed with a Fluor S MultiImager, Multi Analyst ver. 1.1 program (Bio-Rad, Hercules, CA, USA).
Adhesion of neutrophils
Neutrophils (treated or not with ATO 5 µM for 2 h, previously found to be optimal) were labeled for 30 min with 5-µM calcein-AM (Molecular Probes), and adhesion onto A549 cells was measured essentially as we have previously published [23
]. The number of adherent neutrophils was calculated by counting the number of fluorescent cells from five randomly selected high-power fields (x400) observed with a photomicroscope Leica DMRE equipped with an ebq 100 dc epifluorescent condenser. Images were taken with a Cooke Sensicam high-performance camera coupled to the Image Pro-plus (ver. 4.0) program.
Phosphorylation events
Neutrophils (40x106 cells/ml RPMI-HEPES-P/S) were stimulated from 15 s to 30 min with the indicated agonists. In some experiments, cells were preincubated for 30 min with inhibitors for p38 (SB203580, at 10 µM), Erk-1/2 (PD98059, at 10 µM) or JNK (JNKi, at 5 µM). Cells were lysed in 2x Laemmli’s sample buffer, and aliquots corresponding to 1 x 106 cells were loaded onto 10% SDS-PAGE and transferred to nitrocellulose (p38) or polyvinylidene difluoride membranes (Erk-1/2 and JNK). Membranes were blocked for 1 h at room temperature in Tris-buffered saline (TBS)-Tween containing 5% nonfat dry milk (Carnation, Don Mills, Ontario, Canada). After washing, the antiphosphospecific p38, Erk-1/2, or JNK antibody, or the antibody directed against the corresponding unphosphorylated form were added at a final dilution of (1:1000) for phosphospecific antibody or 1:400 for the unphosphorylated form, respectively, in TBS-Tween. The membranes were kept overnight at 4°C and were then washed with TBS-Tween and incubated for 1 h at room temperature with a goat anti-rabbit horseradish peroxidase (HRP) secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) at 1:20,000 in TBS-Tween +5% BSA or a goat anti-mouse HRP secondary antibody (Jackson ImmunoResearch Laboratories) at 1:20,000 in TBS-Tween + 5% nonfat dry milk followed by several washes. Protein expression was revealed using an enhanced chemiluminescence Western blot analysis detection system (Amersham Biosciences, Piscataway, NJ, USA).
Assessment of neutrophil apoptosis
Apoptosis was evaluated by monitoring CD16 shedding. Briefly, cells were incubated on ice for 30 min with PBS + 20% autologous serum to prevent nonspecific binding via Fc receptors. Cells were washed and 2 µl FITC-CD16 antibody was added to 1 x 106 cells for 30 min on ice (light protected). Flow cytometric analysis (10,000 events) was performed using a FACScan (BD Biosciences, San Jose, CA, USA), and apoptotic neutrophils were evaluated by measuring the number of CD16-negative cells.
Phagocytosis of sheep erythrocytes
Sheep red blood cells (SRBCs) (Quelab, Montreal, Quebec, Canada) were opsonized with a 1:200 dilution of a rabbit IgG anti-SRBC antibody (Sigma) for 45 min at 37°C. Neutrophils (2x106 cells in RPMI-1640) pretreated 15 min with the indicated agonists were incubated with 50 x 106 opsonized SRBCs for 30 min. The samples were centrifuged at 200 g for 10 min at 4°C. Supernatants were discarded, and an osmotic shock was performed on the pellets by treating them with 300 µl H2O for 20 s, followed by the addition of 4.5 ml of PBS. The samples were washed, and the final pellets were suspended in 200 µl of PBS. Phagocytosis was measured and expressed as the percentage of neutrophils ingesting at least one opsonized SRBC. Toxaphene was used herein as a positive control for evaluating chemical-induced phagocytosis as previously reported [24
].
Cell surface expression of CD35, CD63, and CD66b
Cell surface expression of CD35, CD63, and CD66b was monitored for assessing degranulation of secretory, specific, and gelatinase or azurophilic granules, respectively, by flow cytometry essentially as previously published [25
]. Nonspecific binding of the antibodies was prevented by incubating the cells with PBS+20% autologous serum for 30 min on ice. Then, after several washes, primary antibodies or an IgG1 isotypic control antibody were added at a concentration of 0.5 µg/ml for 30 min on ice. Cells were washed twice and incubated with 10 µg/ml of a FITC-conjugated goat anti-mouse secondary antibody for 30 min at 4°C. Analysis was performed with a FACScan (BD Biosciences).
Calcium mobilization
Calcium mobilization was assessed using the fluorescent probe Fluo-3 AM(Molecular Probes) as described previously [26
]. Neutrophils (5x106 cells/ml) were incubated for 30 min at 37°C with 3 µM Fluo-3-AM in PBS-1% BSA. Cells were washed three times and resuspended at 5 x 106 cells/ml in PBS-1% BSA. An aliquot of 50 µl was transferred to a FACS tube containing 450 µl of PBS-1% BSA + 1 mM CaCl2. Samples were loaded onto a FACScan flow cytometer, and fluorescence in FL1 was recorded for a period of 
20 s. Then, 5 µl of 100x agonist was added to the tube and fluorescence was recorded during an additional 3 min. The calcium ionophore A23187 was used at a final concentration of 10 µM. HBSS was used as negative control. A gate based on forward- and side-scatter was used to exclude debris.
Statistics
Statistical analysis was performed with SigmaStat for Windows ver. 3.0 using Student’s t test. Statistical significance was established at P < 0.05.
RESULTS
ATO induces p38 phosphorylation in an H2O2- and calcium-independent manner
We have previously found that induction of apoptosis by ATO occurs by a H2O2-dependent mechanism and that this drug could also induce a heat shock-like response in human neutrophils [1
, 8
]. ROS, especially H2O2, have been found to be involved in signal transduction [27
] and could trigger the phosphorylation of different MAPK, including p38, which is involved in stress response. Therefore, we investigate whether or not ATO will activate p38. As illustrated in Fig. 1A
, ATO can strongly induce p38 phosphorylation. In these experiments, the stress-induced H2O2 molecule was used as a positive control. In contrast to H2O2, addition of catalase, an inhibitor of H2O2, did not reverse ATO-induced p38 phosphorylation (Fig. 1B)
. Using a flow cytometry approach, we found that ROS generation in ATO-induced neutrophils was detected only after at least 1 h of treatment and not after 5 or 30 min (Fig. 1C)
. This indicates that activation of p38 MAPK by ATO occurs before any ROS detection, including H2O2.
 View larger version (16K): [in a new window] |
Figure 1. ATO induces p38 phosphorylation in an H2O2-independent manner. (A) Freshly isolated human neutrophils (40x106/ml) were incubated for 5 min in the presence of buffer (Ctrl), 5 µM ATO, or increasing concentrations of H2O2 (0.5, 1, or 10 mM). (B) Cells were preincubated for 30 min with buffer or catalase (2000 U/ml) and then stimulated for 5 min in the presence (ATO) or absence (Ctrl) of 5 µM ATO or H2O2 as above. Immunoblotting was performed using a phosphospecific antibody directed against p38. Equal loading was verified using an antibody against the nonphosphorylated form of p38. Results are from one representative experiment out of three. (C) ATO-induced oxidative burst was detected by flow cytometry of cells labeled with H2DCFDA, as described in the Materials and Methods for the indicated period of time. Reactive oxygen species were detected after 3 h of treatment with 5 µM ATO (arrow). Results are from one representative experiment out of three.
|
 View larger version (31K): [in a new window] |
Figure 2. ATO induces p38 phosphorylation in a calcium-independent manner. (A) Intracellular calcium mobilization was assessed by flow cytometry using the Fluo-3-AM probe, as described in Materials and Methods. Cells were treated with buffer (Ctrl), ATO, or the calcium ionophore A23187 (10 µM). The Y mean amplitudes were plotted (n=3). (Inset) Representative data are plotted in the bar graph. (B) Neutrophils were pretreated for 30 min with the extracellular Ca2+ chelator EGTA (0.5 mM), intracellular Ca2+ chelator BAPTA-AM (50 µM), or buffer, and then treated for 5 min with buffer (Ctrl) or ATO (5 µM). p38 activation was investigated by immunoblotting. Results are from one representative experiment out of three. *, P  0.05 vs. Ctrl.
|
 View larger version (27K): [in a new window] |
Figure 3. ATO activation of MAPK is not involved in induction of neutrophil apoptosis. (A) Phosphorylation of Erk-1/2, JNK, and p38 MAPK was assessed by immunoblotting using specific antibody as described in Materials and Methods. Cells (40x106/ml) were incubated with ATO (5 µM) for 0.25–30 min, GM-CSF (65 ng/ml, for 1 min) or IL-4 (250 ng/ml), for 0.25 min. In some samples, a 30-min pretreatment with specific inhibitors of the MAPK (SB203580 10 µM, PD98059 10 µM, or JNKi II 10 µM) was performed. Shown here is one representative experiment of at least three separate experiments. (B) Neutrophils were pretreated with pharmacological specific inhibitors as above and then buffer, ATO (5 µM) or GM-CSF (65 ng/ml) was added and cells were further incubated for a period of 20 h. Apoptosis was evaluated by flow cytometry following CD16 staining by calculating the number of CD16 negative cells. *, P 0.05 vs. Ctrl. **, P 0.05.
|
Effects of ATO on neutrophil functions other than apoptosis: role of MAPKs in adhesion, migration, and phagocytosis
MAPKs have also been demonstrated to play key roles in the control of inflammation. For example, in a pulmonary inflammation model, p38 has been identified as the main kinase responsible for LPS-triggered neutrophil accumulation [31
]. Consequently, we hypothesized that the phosphorylation of MAPKs was implicated in an activation response. Adhesion of neutrophils to epithelial or endothelial cells is a major step in the recruitment of neutrophils to inflammatory sites. Therefore, we evaluated the ability of ATO-treated neutrophils to adhere to a confluent layer of human epithelial A549 cells in vitro in the presence or absence of MAPK inhibitors. Neutrophils treated with ATO showed an increase in adhesion to A549 cells as observed in Fig. 4A
. Of note, adhesion was significantly decreased with the use of p38 and JNK inhibitors, showing that both signaling pathways are involved in this biological process. Interestingly, the proadhesive effect of ATO was comparable to that of LPS, a potent neutrophil agonist.
 View larger version (26K): [in a new window] |
Figure 4. Identification of ATO as an important neutrophil agonist: role of MAPKs in adhesion, migration, and phagocytosis. The ability of ATO to modulate adhesion, migration, and phagocytosis was investigated as described in Materials and Methods. Cells were pretreated with pharmacological inhibitors for 30 min as described in the legend of Fig. 3
. LPS (1 µg/ml), IL-8 (250 ng/ml), and toxaphene (10 µg/ml) were used as positive controls for adhesion, migration, and phagocytosis, respectively. (Inset, upper right panel) Representative pictures of fluorescent neutrophils adhering onto A549 cells. Results are from at least three different experiments. **, P 0.05 vs. Ctrl. **, P 0.05.
|
Phagocytosis is a crucial biological process in innate immunity, and impairment of phagocytosis leads to recurrent infections. Consequently, we next looked at the effect of ATO on neutrophil phagocytosis. As illustrated in Fig. 4C , ATO can enhance the ability of neutrophils to ingest opsonized sheep red blood cells. The percentage of phagocytosis increases from 38.8 ± 1.6 to 73 ± 5.2 (mean±SE, n=3). Using pharmacological inhibitors, we found that the effect of ATO was significantly reduced only when cells were pretreated with SB203580, indicating that p38 plays a major role. Although we noted a slight decrease with JNKi, this was not statistically significant. The ability of ATO to induce phagocytosis was at potency similar to that of the environmental pollutant toxaphene, used routinely as a positive control when investigating chemical-induced phagocytosis [24 ]. These results suggest that ATO stimulates neutrophil phagocytosis via a p38-dependent pathway.
ATO induces degranulation of gelatinase, specific, and secretory but not azurophilic granules in neutrophils
We have recently demonstrated that ATO induced the release of annexin-1, a protein that colocalizes with gelatinase within the specific and gelatinase granules. Therefore, we hypothesized that ATO could induce degranulation and the release of gelatinase. To verify this, we evaluated gelatinase (or MMP-9) release using a gelatin zymography assay, since this protease is found in the specific or gelatinase granules. As illustrated in Fig. 5 A and B
, ATO induces the release of gelatinase into the external milieu, peaking after 30 min of treatment. Since only the use of the p38 pharmacological inhibitor was suitable for investigating the role of MAPK in this assay (the other inhibitors were found to alter the basal level by themselves), we can only conclude that at least p38 is involved. In parallel, we systematically verified, by trypan blue exclusion assay, that no cell necrosis occurred during the ATO treatments (data not shown).
 View larger version (27K): [in a new window] |
Figure 5. ATO induces the release and activity of gelatinase by a p38 MAPK-dependent mechanism. (A) Extracellular release of gelatinase (or MMP-9) was evaluated using gelatin zymography, as described in Materials and Methods after the indicated period of time following the treatment with agonists. (Inset) Representative data obtained after 30 min of treatment with buffer (Ctrl), 5 µM ATO or 100 ng/ml tumor necrosis factor- (TNF-). Gelatinase activity was measured in arbitrary units by quantitative analysis of negatively stained bands through computerized image analysis (FluorS-MultiImager, Multi Analyst ver. 1.1 program (Bio-Rad). (B) Neutrophils were pretreated with the SB203580 p38 specific inhibitor and incubated for 15 min with buffer, ATO, or TNF-. Supernatants were then separated as described in Materials and Methods. Results are from one representative experiment out of four. Densitometric analysis was performed, and results are expressed as fold increase vs. Ctrl.
|
 View larger version (27K): [in a new window] |
Figure 6. ATO induces degranulation of gelatinase, specific and secretory, but not azurophilic granules in neutrophils by a p38-dependent mechanism. (A) Release of subset types of granules was evaluated by flow cytometry by monitoring cell surface expression of CD35, CD63, or CD66b, as described in Materials and Methods. Cells were incubated in presence of buffer (Ctrl), ATO (5 µM) for 30 min or fMLP (0.1 µM) for 5 min. Legend is as follows: isotypic antibody (filled), Ctrl (black), ATO (light gray), and fMLP (dotted). Results are from one representative experiment out of at least three. (B) Cells were preincubated with SB203580 and then for 30 min with ATO or fMLP (as above). *, P 0.05 vs. Ctrl. **, P 0.05 vs. ATO.
|
ATO induction of MAPK signaling in APL and non-APL cancer cell lines has been well characterized. However, little is known regarding the effect of ATO on signaling in cells of the immune system, especially in neutrophils. We have recently demonstrated that ATO induces human neutrophil apoptosis by a mechanism that is dependent upon de novo protein synthesis and H2O2 production [1 , 8 ]. Considering that H2O2 could activate cell stress in neutrophils, we hypothesized that ATO might activate members of the MAPKs and that this could be related to its proapoptotic activity. However, the results obtained in this study reveal that the general picture of neutrophil activation by ATO is not that simplistic. We provide the first evidence that ATO effectively activates p38 in human neutrophils. We demonstrated that ATO-induced p38 phosphorylation occurs by a H2O2-independent mechanism, based on the fact that 1) activation of p38 by ATO, in contrast with H2O2 itself, is not reversed by catalase treatment; and 2) ATO activates p38 far before an increase of ROS (including H2O2) production can be observed. In addition to p38, we are the first to demonstrate that ATO recruits JNK, but not Erk-1/2 in human neutrophils. ATO has been found to be a proapoptotic agent in many cancer cell lines. The implication of activation of MAPK in the apoptosis effect has been shown to be dependent on cell type. For example, in human myelomonocytic leukemia U937 cells, the activation of p38 was shown to be essential for ATO proapoptotic effect, as well as for production of ROS and calcium burst [33 ]. This is in sharp contrast to our results, which show that the proapoptotic activity of ATO is not linked to MAPK activation. Moreover, we did not observe any calcium burst in response to ATO. However, there is a discrepancy regarding the precise role of the p38 stress kinase for ATO effects. The addition of various p38 inhibitors, including SB203580, SD-282, and SCIO-469 or siRNA against p38, strongly enhanced the induction of apoptosis in chronic myelogenous leukemia KT-1 cells [34 ]. In this case, JNK activation was found to be the kinase responsible for the transduction of proapoptotic signals inside the cells as has been reported in other studies [28 ].
In addition to the new observation that ATO activates p38 and JNK MAPKs in neutrophils, we provide the first evidence that ATO is a neutrophil agonist with a larger spectrum of activity than previously reported, since it modulates several important neutrophil functions other than apoptosis, including adhesion, migration, phagocytosis, and degranulation, in which p38 and/or JNK are involved.
Accumulation of ROS is also a common feature observed in response to ATO cell activation. Indeed, many genes associated with the NADPH oxidase subunits, such as p67phox, gp91phox or rac-2 are up-regulated upon ATO treatment in NB4 cells, and this could lead to production of ROS and cell death [35 ]. Although we demonstrated that the addition of catalase, which converts H2O2 to water and oxygen, could reverse the proapoptotic effect of ATO in neutrophils [8 ], one can imagine that at least two separate phenomena occur in ATO-induced neutrophils. First, ATO could trigger p38 phosphorylation without need for calcium burst or ROS production (particularly H2O2). Second, this drug could simultaneously lead to increased accumulation and release of H2O2, which could trigger apoptosis. Therefore, activation of the p38 and JNK kinases would be a fast response, and this would occur before any apparent signs of apoptosis. Indeed, ATO-induced phosphorylation of p38 and JNK was found to be maximal at 5 min and 15 s, respectively (Fig. 3A) . Interestingly, the modes of entry of ATO in a cell, as well as the release of H2O2, appear to involve the aquaglyceroprotein 9 (AQP9) that was found to be expressed in human neutrophils [36 ]. Moreover, a clear correlation between the expression of AQP9 with arsenic uptake and sensitivity in leukemia cells has been recently reported [37 ].
One feature of classic activators of neutrophils, such as fMLP or platelet-activating factor (PAF) is the stimulation of calcium release. Curiously, we did not observe any calcium burst with ATO treatment in neutrophils, even at concentrations up to 20 µM, which is four times the concentration needed to induce apoptosis in neutrophils. This is in contrast with other reports that have demonstrated a transient increase of calcium in various models with the use of ATO [38
]. However, neutrophils are nondividing primary cells and do not respond in the same way as cancer cells in the presence of ATO. In fact, no production of ROS from the NADPH oxidase or mitochondria has been detected with ATO in human neutrophils [8
] in contrast with APL cells, for example [39
]. It is notable that arsenic has also been shown to alter calcium homeostasis in certain models. Perturbation of calcium signaling could be the result of initiation of an endoplasmic reticulum (ER) stress response [40
]. Interestingly, ATO was reported to up-regulate genes encoding proteins of the metallothionein family or protein-folding and chaperone proteins [40
], which are indicative of an ER stress response. We have recently demonstrated that in addition to annexin-1, ATO increases de novo protein synthesis of heat shock protein 89 (Hsp89), as well as gene expression of Hsp27 and Hsp70 in human neutrophils [1
], supporting the notion that ATO is implicated in an ER stress response in these cells. This remains to be investigated further.
Many neutrophil agonists such as fMLP, TNF-, or LPS are known to activate MAPKs in neutrophils, especially p38 and JNK, which elicit functional responses, including degranulation, adherence, chemotaxis, or superoxide production [41
42
43
]. As previously mentioned, activation of these functions by ATO is under the control of two kinases, namely p38 and JNK. Other upstream kinases, including MKK4, have been associated with activation of p38 or JNK [44
]. Interestingly, Mkk4–/– cells were largely resistant to arsenic toxicity, and phosphorylation of JNK in the presence of arsenite was abrogated in these cells [45
]. Other kinases upstream of p38 or JNK that are phosphorylated with ATO include MKK3, MKK6, and MKK7 [34
, 46
]. Further studies in this area of research will elucidate the precise upstream target of ATO responsible for the initiation of cell signaling in neutrophils.
Our results clearly show that ATO, a chemotherapeutic drug, can stimulate many events implicated in the inflammatory cascade. In fact, adhesion of neutrophils to epithelial or endothelial cells and disruption of the subendothelial extracellular matrix (ECM) by metalloprotease are important steps leading to migration of neutrophils to inflammatory sites [47 ]. Release of MMP-9 is also implicated in a proangiogenic response [48 ], which could be detrimental to cancer patients. However, ATO also enhances the ability of neutrophils to exert phagocytosis. This is interesting from a clinical point of view, if we consider that patients with APL often suffer from neutropenia [49 ]. Indeed, APL is the result of a block in the differentiation of promyelocytic cells, which are precursors of the neutrophils. Therefore, ATO could stimulate innate immunity in patients receiving ATO treatment, in part, by increasing phagocytosis of neutrophils. Given that cancer patients could suffer from recurrent infections due to the low number of leukocytes, we propose that ATO might confer dual protection, by inducing apoptosis of cancerous promyelocytic cells and also by activating neutrophils, particularly with regards to their ability to exert phagocytosis.
Given the central role of neutrophils in various inflammatory disorders and knowing that ATO can regulate several important neutrophil functions, we propose that ATO might have broader clinical implications beyond cancer treatment. On the basis of the results presented in this study, we propose a first model of neutrophil activation in response to ATO, in which MAPKs, including p38 and JNK, but not Erk-1/2, are involved in neutrophil responses. The rapid activation of these MAPKs can regulate neutrophil functions, including migration, adhesion, phagocytosis, and degranulation, after short to moderate time periods of treatment with ATO. In contrast, a longer period of activation with ATO causes the release of accumulated intracellular H2O2 which, if released into the external milieu, leads to apoptosis by a MAPK-independent mechanism (Fig. 7 ).
 View larger version (27K): [in a new window] |
Figure 7. The first proposed model of ATO activation in human neutrophils. ATO enters the cells via aquaglyceroporin (AQP9) expressed in neutrophils (1) and then activates MAPKs (2), p38 and JNK, but not Erk-1/2; MAPKs are activated by ATO. For short to moderate periods of time after activation (3), ATO enhances neutrophil migration, adhesion phagocytosis, and degranulation (all granules but azurophilic) (4). For a long period of activation (5), ATO induces apoptosis by a MAPK-independent mechanism. In parallel, by an unknown mechanism (black box), ATO increases intracellular H2O2 content (a) that can lead directly to apoptosis (b, dotted line). H2O2 may be released into the external milieu via AQP9 (H2O2 is known to pass through such channels) (c) and can induce in a paracrine and/or autocrine fashion (d,e–g). This is supported by the fact that the addition of exogenous catalase (to which the cell is not permeable) is known to totally inhibit ATO-induced apoptosis.
|
ACKNOWLEDGEMENTS
This study was partly supported by the Canadian Institutes of Health Research (CIHR; MOP-89534) and Fonds de la Recherche en Santé du Québec (FRSQ). F. B. holds a FRSQ Ph.D. award and D. G. is a Scholar from FRSQ. We thank Mary Gregory for reading this manuscript.
Received July 16, 2008; revised July 30, 2008; accepted August 5, 2008.
REFERENCES
B-related survival pathway down-regulation J. Pharmacol. Exp. Ther. 316,304-314 induces tyrosine phosphorylation of mitogen-activated protein kinase in adherent human neutrophils J. Immunol. 154,4785-4792[Abstract]B kinase beta in metallothionein-1 expression and protection against arsenic toxicity J. Biol. Chem. 282,21487-21496This article has been cited by other articles:
 |
|
 |
|
  J. Ennaciri and D. Girard IL-4R{alpha}, a New Member that Associates with Syk Kinase: Implication in IL-4-Induced Human Neutrophil Functions J. Immunol., October 15, 2009; 183(8): 5261 - 5269. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
