- Email: [email protected]
Calcium signaling and protein kinase C for TNF-α secretion in a rat mast cell line
Life Sriences, Vol. 62, Nos. 17f 18, pp. 165~1657,1998 CoWright @ 1998 Eltier sciina! Inc. Printed in the USA. All rights mclved 0024-3205/98 $19.00 + .oo PII
SOO24-3205(98)OOl234
CALCIUM SIGNALING AND PROTEIN KINASE C FOR TNF-a SECRETION IN A RAT MAST CELL LINE Yoshihiro Nakata and Izumi Hide Department of Pharmacology, Institute of Pharmaceutical Sciences, Hiroshima University School of Medicine, Hiroshima, JAPAN.
Summary In mast cells, like other nonexcitable cells, receptor activation produces C$‘mobilizing second messengers such as inositol 1,4,Qriphosphate or sphingosine-lphosphate, which induce Ca2’ release from internal stores. The resulting depletion of Ca” stores activates Ca2’ channels in plasma membranes designated as Ca2’ release-activated Ca” (CRAC) channels. Ionomycin appears to cause activation of CRAC channels by depleting intracellular Ca2’ stores rather than by acting as an ionophore. We compared the effects of azelastine, an anti-allergic drug, on TNF-a secretion 9 on Ca2’ signal, and on degmnulation in an antigen- or ionomycinstimulated rat mast RBL-2H3 cell line. Azelastine inhibited TNF-a release at concentrations lower than those needed for the inhibition of degranulation. In antigen-stimulated cells, azelastine also inhibited equipotently TNF-a mRNA expression/protein synthesis, TNF-a release and Ca” influx. In ionomycinstimulated cells, however, azelastine inhibited TNF-a release to a greater extent than TNF-a mRNA expression/ protein synthesis and Ca2’ influx, indicating that azelastine inhibits the release process more potently than transcription or production of TNF-a by interfering with a signal other than C$‘. Pretreatment with 1 pM azelastine inhibited ionomycin-induced, but not antigen-induced, protein kinase C translocation to the membranes. These results suggest that TNF-a transcription/production is mainly regulated by Ca2’ influx, but the release process of TNF-a is regulated by additional mechanism(s) possibly involving activation of protein kinase C.
Key Words:
calcium, PKC, TNF-a,
antigen, ionomycin
Mast cells express receptors with high affinity for IgE (FcERI), which trigger the release of preformed inflammatory mediators such as histamine from granules (degranulation) and synthesis of arachidonic acid metabolites and cytokines (1). Among mast cell-derived cytokines, TNF-a is of particular importance because it causes allergic inflammation by inducing the expression of adhesion molecules such as ICAM-I in endothelial cells, which is responsible for the local infiltration of inflammatory leukocytes into tissues. Correspondence to: Yoshihiro Nakata, Department of Pharmacology, Institute of Pharmaceutical Sciences, Hiroshima University School of Medicine, Hiroshima, 734 Japan, Phone:8 1-82-2575310, Fax: 81-82-257-5314, E-mail address: [email protected]
1654
Calcium and PKC in TNF-a Secretion
Vol.
62, Nos. 17/l& 1998
In rat basophilic leukemia (RBL-2H3) cells, a cognate mast cell line, tyrosine kinase Syk recruited by aggregated FcERI phosphorylates phospholipase Cy, which leads to the generation of inositol 1,4,5triphosphate (IP3). IP3 causes the release of Ca2’ from intracellular Ca2’ stores and activates capacitative Ca2+ entry via Ca2+ release-activated Ca2’ (CRAC) channels to replenish the depleted Ca2’ stores (2). Recently it has been revealed that antigen stimulates sphingosine kinase to generate another Ca2* mobilizing second messenger, sphingosine-1 -phosphate (3). Ca*+ ionophores such as ionomycin are widely used with the aim of increasing intracellular Ca” concentrations. It has been demonstrated that anti-allergic drugs, such as cromoglycate and/or some of the new lgeneration histamine Hi-antagonists including azelastine, 4-(p-chlorobenzyl)-2-(hexahydromethyl-lH-azepin-Cyl)-I-(2H)-phthalazinone, inhibit the release of histamine from mast cells and biosynthesis of leukotrienes in leukocytes, and these effects have been explained on the basis of their inhibitory effects on Ca2’ influx (4, 5), which is essential for both degranulation and cytokine production. However, the effect of azelastine on TNF-a production and/or release in mast cells is not known. In this study, we compared the effects of azelastine on TNF-a production and release, on degranulation, and on Ca2’ signals in RBL-2H3 cells. Our results indicate that azelastine is a more potent inhibitor of TNF-a release than it is of the degranulation process. Furthermore, in ionomycin-stimulated cells, it more potently inhibited the release process rather than Ca2’ influx, mRNA expression and protein synthesis of TNF-a, while in antigen-stimulated cells, TNF-a secretion was most likely inhibited by blocking Ca 2+ influx, resulting in suppression of TNF-a mRNA expression/protein synthesis. In order to clarify how azelastine regulates TNF-a release, we further examined the effects of azelastine on PKC translocation in RBL-2H3 cells. Methods Cell culture. RBL-2H3 cells were grown in RPMI-1640 supplemented with 10 % fetal calf serum, 100 U/ml penicillin and lOOpg/ml streptomycin (cRPM1). The cultures were incubated overnight with 0.5 &ml DNP-specific IgE. Following a washing step, the cells were stimulated with antigen, 2,4-dinitrophenylated-bovine serum albumin (DNP-BSA). Ionomycin stimulation was performed in the same way except that the IgE sensitization step was omitted. Measurement of intracellzdar Ca2’ concentration f[Ca”]i) in single cells. The cells with 2 pM fiua3/AM at 25 “C for 45 min. Fura-2-loaded cells were placed on a imaging microscope and perfused at 1.5 ml/min with medium at 37 ‘C. The furawas measured with excitation at 340 nrn and 380 nm and at an emission wavelength The video image output was digitized by an Argus 50 color image processor Photonics, Shizuoka, Japan).
were loaded fluorescence fluorescence of 510 run. (Hamamatsu
Measurement of release of P_hexosaminidase and TNF-a. Degranulation was determined by measuring the release of the granule marker, P-hexosaminidase (6). TNF-a was assayed in 50-pl samples of medium or cell lysate using an ELISA kit (Biosource International, Camarillo, CA ). Cell lysates were prepared in 0.1 % Triton X-l 00 in cRPM1, which had no effect on measurement by ELISA. Samples of both medium and cell lysate were stored at -80 ‘C for assay. PKC translocution. Cells were stimulated with 10 r&ml DNP-BSA for 5 min in medium that contained 119 mM NaCl, 5 mM KCI, 0.4 mM glucose, 1 mg/ml BSA, 25 mM Pipes, and pH adjusted to 7.4 medium, the cells were scraped and sonicated in 1 ml of an ice-cold
for 1 min or 1 pM ionomycin mM MgC12, 1 mM CaC12, 5.6 with NaOH. After removal of extraction buffer consisting of
Vol. 62, Nos. 17/18,1998
Calcium and PKC in TNF-a Secretion
1655
20 mM Tris-HCl, pH 7.5, 5 mM EDTA, 10 mM EGTA, 2 mM dithiothreitol, 100 l&ml PMSF and 100 ug/ml leupeptine. The homogenate was centrifuged at 100,000 x g for 10 min and the pellet was resuspended with sonication in 1 ml of extraction buffer and centrifuged at 100,000 x g for 10 min. The membrane fraction was solubilized by sonication in 200 ul of buffer containing 0.1 % Triton X-l 00. PKC activity was measured using a PKC enzyme assay system (Amersham). Results In RBL-2H3 cells, antigen stimulation caused a transient increase in [Ca*‘]i in the absence of external Ca2’ and a sustained increase in [Ca2’]i in the presence of external Ca*’ (Fig. 1A). The first component represented Ca*’ release from internal Ca*’ stores, and the following component was comprised of Ca*’ influx via Ca*’ channels, designated as Ca*’ release-activated Ca*’ (CRAC) channels, which are activated by the depletion of internal Ca2’ stores. The CRAC channel blocker, SK&F 96365 and La3’, blocked both Ca*’ influx and P-hexosaminidase release stimulated not only by antigen, but also by ionomycin, suggesting that ionomycin may cause Ca*’ influx by activating CRAC channels by a mechanism similar to that produced by antigen, rather than by transporting Ca*’ across plasma membranes as an ionophore. A
B BW
C 803
lonomycin
Antigen
COlltlUl I-.
800
-y -ix..
403
AZlOOpM
r
Time (min)
Ok-+TT-G
._. -..___. I..h I.
IO
Time (min)
Time (min)
Fig. 1 Antigen-induced Ca*’ signal in the presence or absence of external Ca*’ (A) and effects of azelastine (AZ) on antigen (B)- or ionomycin (C)-induced Ca*’ influx in RBL-2H3 cells.
KBL-2H3 cells released TNF-o as well as g-hexosaminidase in response to antigen, DNP-BSA, or ionomycin in a concentration-dependent manner. External Ca2’ was an absolute requirement for
Table I. Comparisons of the inhibitory effects (ICSO)of azelastine on the release of P-hexosaminidase or TNF-a in response to antigen or ionomycin in RBL-2H3 cells. ICSOvalues (PM)
g-Hexosaminidase
TNF-a
Antigen
Ionomycin
128 + 17
13.9 k 2.0
25.7 * 3.4
1.66 f 0.45
1656
Calcium and PKC in TNF-a Secretion
Vol. 62, Nos. 17/l& 1998
both g-hexosaminidase and TNF-a release because both responses were abolished by EGTA. Azelastine inhibited more potently TNF-a secretion than /3-hexosaminidase release in response to antigen or ionomycin and inhibited more potently ionomycin-induced release than antigen-induced release (Table I). Azelastine did not affect the release from intracellular Ca*’ stores, but significantly inhibited Ca*’ influx by either antigen or ionomycin (Fig. lB, C). In antigenstimulated cells, the inhibition of Ca*’ influx was most likely related to the inhibition of TNF-a release which resulted from suppression of mRNA expression. On the other hand, in ionomycinstimulated cells, azelastine inhibited TNF-a release more effectively than against Ca*’ influx or TNF-a mRNA expression. These results suggest that Ca*’ influx mainly regulates TNF-a transcription, while the TNF-a release process requires another signal regulation besides Ca*‘. In order to test whether low concentrations (-1 pM) of azelastine could inhibit PKC activation, we examined the effects of azelastine on antigen- or ionomycin-induced increase in membraneassociated PKC activity. Antigen and ionomycin induced a maximal increase in membrane-bound PKC in 1 min (by 91 %) and 5 min (by 33 %), respectively. Pretreatment with 1 PM azelastine at 37 “C for 10 min inhibited the membrane translocation of PKC induced by ionomycin, but not by antigen, to basal levels (Table II), suggesting that PKC is a potential target of azelastine in the TNF-a releasing process.
Table II. Effects of azelastine on antigen or ionomycin-induced PKC translocation to the membranes in RBL-2H3 cells.
PKC activity in membranes Antigen
(% of basal)
Ionomycin
Basal
100
100
Control
190+3
133 t 1
1
192 F 15
1oot3**
10
167 F 4*
Azelastine
100
(uM)
150+ 12**
92 k 3** 109 + 6*
*P
Discussion
Ca*’ ionophores including ionomycin have been widely used with the aim of increasing [Ca*‘]i by forming a lipid-soluble complex with Ca*’ and transporting Ca*’ across plasma membranes and bypassing receptor-mediated early events. The present study using CRAC channel blockers, SK&F 96365 or La3’ (7, 8), suggests that ionomycin might activate Ca*’ influx via CRAC channels by depleting Ca*’ stores rather than acting as an ionophore in RBL-2H3 cells, namely, ionomycin may cause Ca*’ influx in a similar way as does antigen. The anti-allergic drug, azelastine, which is known to inhibit degranulation (5), also blocked Ca *+ influx stimulated by both antigen and ionomycin, indicating that azelastine acts as a CRAC channel blocker in RBL-2H3 cells.
Vol. 62, Nos. 17/l& 1998
Calcium and PKC in TNF-a Secretion
1657
In the present study, we investigated whether azelastine could also regulate TNF-a secretion in response to antigen or ionomycin in RBL-2H3 cells. Azelastine inhibited TNF-a secretion at lower concentrations than those needed for the inhibition of degranulation, suggesting a differential regulation of the degranulation process and of TNF-a secretion. It has been shown that RBL-2H3 cells secrete TNF-a via a brefeldin A-sensitive Golgi-dependent mechanism, which is not involved in the release of histamine or other preformed mediators (9). Therefore, there may be an intrinsic regulatory mechanism for cytokine secretion distinct from that for degmnulation. In antigen stimulation, azelastine inhibited transcription, protein synthesis and release of TNF-a to the same extent as was associated with inhibition of Ca*’ influx. In ionomycin-stimulated cells, however, azelastine inhibited TNF-a release more effectively than against Ca*’ influx and transcription of TNF-a. These results suggest that TNF-a production is mainly regulated by Ca*’ influx via CRAC channels, and the TNF-a release process is regulated by additional crucial signals distinct from Ca*‘, which is sensitive to low concentrations of azelastine in ionomycin-stimulated cells. It has been reported that, unlike degranulation, TNF-a production was induced by either ionomycin or PMA individually, although the effects were synergistic when used in combination. However, release of TNF-a occurred only in the presence of both ionomycin and PMA and was inhibited by either a PKC inhibitor (Ro31-7549) or EGTA without abolishing TNF-a production, indicating that TNF-a release is dependent on both PKC and Ca*’ (9). In the present study, we demonstrated that low concentrations of azelastine inhibited the translocation of PKC to the membranes in RBL-2H3 cells activated by ionomycin, but not by antigen. This result indicates that PKC activation generated by ionomycin stimulation specifically regulates the TNF-a release process and might be a target for azelastine. We are now investigating which PKC isozyme translocation is inhibited by azelastine. Acknowledgements We are grateful to Prof. S. Yamawaki, Dr. A. Kagaya and Dr. Neurology and Psychiatry, Hiroshima University School of fluorescence imaging microscope, and the Research Center for University School of Medicine for the use of its facilities. We Benveniste, University of Alabama at Birmingham, for providing Eisai Co. for supplying with azelastine hydrochloride.
M. Takebayashi, Department of Medicine, for the use of the Molecular Medicine, Hiroshima should like to thank Prof. E.N. a partial rat TNF-a cDNA, and
References 1. 2. 3. 4. 5. 6. 7.
8. 9.
M.A. BEAVEN and R.A. BAUMGARTNER. Curr. Opin. Immunol. 8 766-772 (1996). M.A. BEAVEN. Curr. Biol. 6: 798-801 (1996). O.H. CHOI, J.-H. KIM and J.-P. KINNET, Nature 380 634-636 (1996). N. CHAND, J. PILLAR, W. DIAMANTIS, J. L. PERHACH, Jr. and R.D. SOFIA. Eur. J. Pharrnacol. 96 227-233 (1983). N. CHAND and R.D. SOFIA. J. Asthma 32 227-234 (1995). I. HIDE, J.P. BENNETT, A. PIZZEY, G. BOONEN, D. BAR-SAGI, B.D. GOMPERTS and P.E.R.TATHAM, J. Cell. Biol. 123 585-593 (1993). J.E. MERRITT, W.P. ARMSTRONG, C.D. BENHAM, T.J. HALLAM, R. JACOB, A. JAXA-CHAMIEC, B.K. LEIGH, S.A. McCarthy, K.E. MOORES and T.J. RINK. Biochem. J. 271 515-522 (1990). M. HIDE and M.A. BEAVEN, J. Biol. Chem. 266 15221-15229 (1991). R.A. BAUMGARTNER, K. YAMADA, V.A. DERAMO and M.A. BEAVEN. J. Immunol. 153 2609-2617 (1994).
SOO24-3205(98)OOl234
CALCIUM SIGNALING AND PROTEIN KINASE C FOR TNF-a SECRETION IN A RAT MAST CELL LINE Yoshihiro Nakata and Izumi Hide Department of Pharmacology, Institute of Pharmaceutical Sciences, Hiroshima University School of Medicine, Hiroshima, JAPAN.
Summary In mast cells, like other nonexcitable cells, receptor activation produces C$‘mobilizing second messengers such as inositol 1,4,Qriphosphate or sphingosine-lphosphate, which induce Ca2’ release from internal stores. The resulting depletion of Ca” stores activates Ca2’ channels in plasma membranes designated as Ca2’ release-activated Ca” (CRAC) channels. Ionomycin appears to cause activation of CRAC channels by depleting intracellular Ca2’ stores rather than by acting as an ionophore. We compared the effects of azelastine, an anti-allergic drug, on TNF-a secretion 9 on Ca2’ signal, and on degmnulation in an antigen- or ionomycinstimulated rat mast RBL-2H3 cell line. Azelastine inhibited TNF-a release at concentrations lower than those needed for the inhibition of degranulation. In antigen-stimulated cells, azelastine also inhibited equipotently TNF-a mRNA expression/protein synthesis, TNF-a release and Ca” influx. In ionomycinstimulated cells, however, azelastine inhibited TNF-a release to a greater extent than TNF-a mRNA expression/ protein synthesis and Ca2’ influx, indicating that azelastine inhibits the release process more potently than transcription or production of TNF-a by interfering with a signal other than C$‘. Pretreatment with 1 pM azelastine inhibited ionomycin-induced, but not antigen-induced, protein kinase C translocation to the membranes. These results suggest that TNF-a transcription/production is mainly regulated by Ca2’ influx, but the release process of TNF-a is regulated by additional mechanism(s) possibly involving activation of protein kinase C.
Key Words:
calcium, PKC, TNF-a,
antigen, ionomycin
Mast cells express receptors with high affinity for IgE (FcERI), which trigger the release of preformed inflammatory mediators such as histamine from granules (degranulation) and synthesis of arachidonic acid metabolites and cytokines (1). Among mast cell-derived cytokines, TNF-a is of particular importance because it causes allergic inflammation by inducing the expression of adhesion molecules such as ICAM-I in endothelial cells, which is responsible for the local infiltration of inflammatory leukocytes into tissues. Correspondence to: Yoshihiro Nakata, Department of Pharmacology, Institute of Pharmaceutical Sciences, Hiroshima University School of Medicine, Hiroshima, 734 Japan, Phone:8 1-82-2575310, Fax: 81-82-257-5314, E-mail address: [email protected]
1654
Calcium and PKC in TNF-a Secretion
Vol.
62, Nos. 17/l& 1998
In rat basophilic leukemia (RBL-2H3) cells, a cognate mast cell line, tyrosine kinase Syk recruited by aggregated FcERI phosphorylates phospholipase Cy, which leads to the generation of inositol 1,4,5triphosphate (IP3). IP3 causes the release of Ca2’ from intracellular Ca2’ stores and activates capacitative Ca2+ entry via Ca2+ release-activated Ca2’ (CRAC) channels to replenish the depleted Ca2’ stores (2). Recently it has been revealed that antigen stimulates sphingosine kinase to generate another Ca2* mobilizing second messenger, sphingosine-1 -phosphate (3). Ca*+ ionophores such as ionomycin are widely used with the aim of increasing intracellular Ca” concentrations. It has been demonstrated that anti-allergic drugs, such as cromoglycate and/or some of the new lgeneration histamine Hi-antagonists including azelastine, 4-(p-chlorobenzyl)-2-(hexahydromethyl-lH-azepin-Cyl)-I-(2H)-phthalazinone, inhibit the release of histamine from mast cells and biosynthesis of leukotrienes in leukocytes, and these effects have been explained on the basis of their inhibitory effects on Ca2’ influx (4, 5), which is essential for both degranulation and cytokine production. However, the effect of azelastine on TNF-a production and/or release in mast cells is not known. In this study, we compared the effects of azelastine on TNF-a production and release, on degranulation, and on Ca2’ signals in RBL-2H3 cells. Our results indicate that azelastine is a more potent inhibitor of TNF-a release than it is of the degranulation process. Furthermore, in ionomycin-stimulated cells, it more potently inhibited the release process rather than Ca2’ influx, mRNA expression and protein synthesis of TNF-a, while in antigen-stimulated cells, TNF-a secretion was most likely inhibited by blocking Ca 2+ influx, resulting in suppression of TNF-a mRNA expression/protein synthesis. In order to clarify how azelastine regulates TNF-a release, we further examined the effects of azelastine on PKC translocation in RBL-2H3 cells. Methods Cell culture. RBL-2H3 cells were grown in RPMI-1640 supplemented with 10 % fetal calf serum, 100 U/ml penicillin and lOOpg/ml streptomycin (cRPM1). The cultures were incubated overnight with 0.5 &ml DNP-specific IgE. Following a washing step, the cells were stimulated with antigen, 2,4-dinitrophenylated-bovine serum albumin (DNP-BSA). Ionomycin stimulation was performed in the same way except that the IgE sensitization step was omitted. Measurement of intracellzdar Ca2’ concentration f[Ca”]i) in single cells. The cells with 2 pM fiua3/AM at 25 “C for 45 min. Fura-2-loaded cells were placed on a imaging microscope and perfused at 1.5 ml/min with medium at 37 ‘C. The furawas measured with excitation at 340 nrn and 380 nm and at an emission wavelength The video image output was digitized by an Argus 50 color image processor Photonics, Shizuoka, Japan).
were loaded fluorescence fluorescence of 510 run. (Hamamatsu
Measurement of release of P_hexosaminidase and TNF-a. Degranulation was determined by measuring the release of the granule marker, P-hexosaminidase (6). TNF-a was assayed in 50-pl samples of medium or cell lysate using an ELISA kit (Biosource International, Camarillo, CA ). Cell lysates were prepared in 0.1 % Triton X-l 00 in cRPM1, which had no effect on measurement by ELISA. Samples of both medium and cell lysate were stored at -80 ‘C for assay. PKC translocution. Cells were stimulated with 10 r&ml DNP-BSA for 5 min in medium that contained 119 mM NaCl, 5 mM KCI, 0.4 mM glucose, 1 mg/ml BSA, 25 mM Pipes, and pH adjusted to 7.4 medium, the cells were scraped and sonicated in 1 ml of an ice-cold
for 1 min or 1 pM ionomycin mM MgC12, 1 mM CaC12, 5.6 with NaOH. After removal of extraction buffer consisting of
Vol. 62, Nos. 17/18,1998
Calcium and PKC in TNF-a Secretion
1655
20 mM Tris-HCl, pH 7.5, 5 mM EDTA, 10 mM EGTA, 2 mM dithiothreitol, 100 l&ml PMSF and 100 ug/ml leupeptine. The homogenate was centrifuged at 100,000 x g for 10 min and the pellet was resuspended with sonication in 1 ml of extraction buffer and centrifuged at 100,000 x g for 10 min. The membrane fraction was solubilized by sonication in 200 ul of buffer containing 0.1 % Triton X-l 00. PKC activity was measured using a PKC enzyme assay system (Amersham). Results In RBL-2H3 cells, antigen stimulation caused a transient increase in [Ca*‘]i in the absence of external Ca2’ and a sustained increase in [Ca2’]i in the presence of external Ca*’ (Fig. 1A). The first component represented Ca*’ release from internal Ca*’ stores, and the following component was comprised of Ca*’ influx via Ca*’ channels, designated as Ca*’ release-activated Ca*’ (CRAC) channels, which are activated by the depletion of internal Ca2’ stores. The CRAC channel blocker, SK&F 96365 and La3’, blocked both Ca*’ influx and P-hexosaminidase release stimulated not only by antigen, but also by ionomycin, suggesting that ionomycin may cause Ca*’ influx by activating CRAC channels by a mechanism similar to that produced by antigen, rather than by transporting Ca*’ across plasma membranes as an ionophore. A
B BW
C 803
lonomycin
Antigen
COlltlUl I-.
800
-y -ix..
403
AZlOOpM
r
Time (min)
Ok-+TT-G
._. -..___. I..h I.
IO
Time (min)
Time (min)
Fig. 1 Antigen-induced Ca*’ signal in the presence or absence of external Ca*’ (A) and effects of azelastine (AZ) on antigen (B)- or ionomycin (C)-induced Ca*’ influx in RBL-2H3 cells.
KBL-2H3 cells released TNF-o as well as g-hexosaminidase in response to antigen, DNP-BSA, or ionomycin in a concentration-dependent manner. External Ca2’ was an absolute requirement for
Table I. Comparisons of the inhibitory effects (ICSO)of azelastine on the release of P-hexosaminidase or TNF-a in response to antigen or ionomycin in RBL-2H3 cells. ICSOvalues (PM)
g-Hexosaminidase
TNF-a
Antigen
Ionomycin
128 + 17
13.9 k 2.0
25.7 * 3.4
1.66 f 0.45
1656
Calcium and PKC in TNF-a Secretion
Vol. 62, Nos. 17/l& 1998
both g-hexosaminidase and TNF-a release because both responses were abolished by EGTA. Azelastine inhibited more potently TNF-a secretion than /3-hexosaminidase release in response to antigen or ionomycin and inhibited more potently ionomycin-induced release than antigen-induced release (Table I). Azelastine did not affect the release from intracellular Ca*’ stores, but significantly inhibited Ca*’ influx by either antigen or ionomycin (Fig. lB, C). In antigenstimulated cells, the inhibition of Ca*’ influx was most likely related to the inhibition of TNF-a release which resulted from suppression of mRNA expression. On the other hand, in ionomycinstimulated cells, azelastine inhibited TNF-a release more effectively than against Ca*’ influx or TNF-a mRNA expression. These results suggest that Ca*’ influx mainly regulates TNF-a transcription, while the TNF-a release process requires another signal regulation besides Ca*‘. In order to test whether low concentrations (-1 pM) of azelastine could inhibit PKC activation, we examined the effects of azelastine on antigen- or ionomycin-induced increase in membraneassociated PKC activity. Antigen and ionomycin induced a maximal increase in membrane-bound PKC in 1 min (by 91 %) and 5 min (by 33 %), respectively. Pretreatment with 1 PM azelastine at 37 “C for 10 min inhibited the membrane translocation of PKC induced by ionomycin, but not by antigen, to basal levels (Table II), suggesting that PKC is a potential target of azelastine in the TNF-a releasing process.
Table II. Effects of azelastine on antigen or ionomycin-induced PKC translocation to the membranes in RBL-2H3 cells.
PKC activity in membranes Antigen
(% of basal)
Ionomycin
Basal
100
100
Control
190+3
133 t 1
1
192 F 15
1oot3**
10
167 F 4*
Azelastine
100
(uM)
150+ 12**
92 k 3** 109 + 6*
*P
Discussion
Ca*’ ionophores including ionomycin have been widely used with the aim of increasing [Ca*‘]i by forming a lipid-soluble complex with Ca*’ and transporting Ca*’ across plasma membranes and bypassing receptor-mediated early events. The present study using CRAC channel blockers, SK&F 96365 or La3’ (7, 8), suggests that ionomycin might activate Ca*’ influx via CRAC channels by depleting Ca*’ stores rather than acting as an ionophore in RBL-2H3 cells, namely, ionomycin may cause Ca*’ influx in a similar way as does antigen. The anti-allergic drug, azelastine, which is known to inhibit degranulation (5), also blocked Ca *+ influx stimulated by both antigen and ionomycin, indicating that azelastine acts as a CRAC channel blocker in RBL-2H3 cells.
Vol. 62, Nos. 17/l& 1998
Calcium and PKC in TNF-a Secretion
1657
In the present study, we investigated whether azelastine could also regulate TNF-a secretion in response to antigen or ionomycin in RBL-2H3 cells. Azelastine inhibited TNF-a secretion at lower concentrations than those needed for the inhibition of degranulation, suggesting a differential regulation of the degranulation process and of TNF-a secretion. It has been shown that RBL-2H3 cells secrete TNF-a via a brefeldin A-sensitive Golgi-dependent mechanism, which is not involved in the release of histamine or other preformed mediators (9). Therefore, there may be an intrinsic regulatory mechanism for cytokine secretion distinct from that for degmnulation. In antigen stimulation, azelastine inhibited transcription, protein synthesis and release of TNF-a to the same extent as was associated with inhibition of Ca*’ influx. In ionomycin-stimulated cells, however, azelastine inhibited TNF-a release more effectively than against Ca*’ influx and transcription of TNF-a. These results suggest that TNF-a production is mainly regulated by Ca*’ influx via CRAC channels, and the TNF-a release process is regulated by additional crucial signals distinct from Ca*‘, which is sensitive to low concentrations of azelastine in ionomycin-stimulated cells. It has been reported that, unlike degranulation, TNF-a production was induced by either ionomycin or PMA individually, although the effects were synergistic when used in combination. However, release of TNF-a occurred only in the presence of both ionomycin and PMA and was inhibited by either a PKC inhibitor (Ro31-7549) or EGTA without abolishing TNF-a production, indicating that TNF-a release is dependent on both PKC and Ca*’ (9). In the present study, we demonstrated that low concentrations of azelastine inhibited the translocation of PKC to the membranes in RBL-2H3 cells activated by ionomycin, but not by antigen. This result indicates that PKC activation generated by ionomycin stimulation specifically regulates the TNF-a release process and might be a target for azelastine. We are now investigating which PKC isozyme translocation is inhibited by azelastine. Acknowledgements We are grateful to Prof. S. Yamawaki, Dr. A. Kagaya and Dr. Neurology and Psychiatry, Hiroshima University School of fluorescence imaging microscope, and the Research Center for University School of Medicine for the use of its facilities. We Benveniste, University of Alabama at Birmingham, for providing Eisai Co. for supplying with azelastine hydrochloride.
M. Takebayashi, Department of Medicine, for the use of the Molecular Medicine, Hiroshima should like to thank Prof. E.N. a partial rat TNF-a cDNA, and
References 1. 2. 3. 4. 5. 6. 7.
8. 9.
M.A. BEAVEN and R.A. BAUMGARTNER. Curr. Opin. Immunol. 8 766-772 (1996). M.A. BEAVEN. Curr. Biol. 6: 798-801 (1996). O.H. CHOI, J.-H. KIM and J.-P. KINNET, Nature 380 634-636 (1996). N. CHAND, J. PILLAR, W. DIAMANTIS, J. L. PERHACH, Jr. and R.D. SOFIA. Eur. J. Pharrnacol. 96 227-233 (1983). N. CHAND and R.D. SOFIA. J. Asthma 32 227-234 (1995). I. HIDE, J.P. BENNETT, A. PIZZEY, G. BOONEN, D. BAR-SAGI, B.D. GOMPERTS and P.E.R.TATHAM, J. Cell. Biol. 123 585-593 (1993). J.E. MERRITT, W.P. ARMSTRONG, C.D. BENHAM, T.J. HALLAM, R. JACOB, A. JAXA-CHAMIEC, B.K. LEIGH, S.A. McCarthy, K.E. MOORES and T.J. RINK. Biochem. J. 271 515-522 (1990). M. HIDE and M.A. BEAVEN, J. Biol. Chem. 266 15221-15229 (1991). R.A. BAUMGARTNER, K. YAMADA, V.A. DERAMO and M.A. BEAVEN. J. Immunol. 153 2609-2617 (1994).