- Email: [email protected]
Ca exchange by Phe-Met-Arg-Phe-NH2 (FMRFa)-related peptides in intact rat pancreatic B-cells
ELSEVIER
Molecular
and Cellular Endocrinology
106 (1994) Rl-RS
Inhibition of Na/Ca exchange by Phe-Met-Arg-Phe-NH2 (FMRFa)-related peptides in intact rat pancreatic B-cells F. Van Eylena, P. Gourletb, A. Vandermeersb, P. Lebruna, A. Herchuelza-* “Laboratory of Pharmacology, bL.uboratory of Biochemistry, Brussels University School of Medicine, Bat. GE, 808 route de Lennik, B-1070, Brussels, Belgium Received 20 July 1994; accepted
1 August 1994
Abstract Phe-Met-Arg-Phe-NH2 (FMRFa)-related peptides were recently shown to inhibit Na/Ca exchange in cardiac sarcolemmal vesicles. In the present study, we examined the effects of FMRFa-related peptides on Na/Ca exchange in intact (pancreatic B) cells. At 2.8 mM glucose, FMRFa-related peptides only weakly inhibited Na/Ca exchange although their effect was more marked under depolarizing conditions. The peptides blocked neither the Na/K-ATPase nor Ca *+ channels but slightly reduced membrane K+ permeability. Our data indicate that PMRFa-related peptides are weak and non-specific inhibitors of Na/Ca exchange in intact B cells. The data do not confirm the view that the peptides may exert some of their physiological modulatory role by inhibiting Na/Ca exchange. Keywor&:
Na/Ca exchange; FMRFamide-related peptides; Pancreatic B-cell
1. Introduction Two processes located at the plasma membrane appear to mediate Ca*+ extrusion from cells: the Ca*+-ATPase and Na/Ca exchange (Carafoli, 1988; Blaustein, 1988). In the heart, Na/Ca exchange appears to be the predominant mechanism for Ca*+ extrusion, being able to restore and control basal Ca*+ level on a beat-to-beat basis (Bers, 1991). However, the role played by the exchanger in many other cell types remains to be elucidated. One major factor that has hindered rapid progress in the knowledge of the contribution of NaKa exchange to the control of Ca*+ homeostasis, is the lack of specific inhibitors of the exchanger. Recently, opioid antagonists (e.g. naloxone) were shown to inhibit Na/Ca exchange in cardiac sarcolemmal vesicles (Khananshvili and Sarne, 1992). Similarly, Phe-Met-Arg-
Abbreviations: FLRFa, Phe-Leu-Arg-Phe-NH2; FMRFa, Phe-Met-ArgPhe-NH2; HMRFa, His-Met-Arg-Phe-NH2; TEA, tetraethylammonium cl$oride; VMRFa, Val-Met-Arg-Phe-NH2. Corresponding author. Laboratoire de Pharmacodynamie et de therapeutique, Universit6 Libre de Bruxelles, Faculte de Mtdecine, Route de Lennik, 808-Bfitiment G.E., B-1070 Bruxelles, Belgium. Tel. (32) 2 555 62 01. Fax (32) 2 55.5 63 70. E-mail: [email protected].
0 1994 Elsevier Science Ireland Ltd. All rights reserved 0303-7207/94/$07.00 SSDI 0303-7207(94)03394-Z
Phe-NH2 (FMRFa)-related peptides with naloxone-like activity were observed to provoke complete inhibition of Na/Ca exchange in cardiac sarcolemmal vesicles (Khananshvili et al., 1993). The two most potent inhibitors were the derivatives with NH,-Phe substituted by either Val or His (VMRFa and HMRFa). In addition, it was suggested that endogenous analogs of FMRFa may exert some of their physiological modulatory role by inhibiting Na/Ca exchange (Khananshvili et al., 1993). FMRFa-related peptides have been localized in several mammalian tissues and appear to exert various actions including antagonism to opioid analgesia (Price and Greenberg, 1989; Raffa, 1988). Because the effect of FMRFa-related peptides was studied in sarcolemmal vesicles, namely in a mixture of inside-out and outside-out vesicles, the view that such peptides may inhibit Na/Ca exchange in intact cells remains to be demonstrated. In the present study, we have examined the effects of FMRFa-related peptides on Na/Ca exchange in intact rat pancreatic B-cells. 2. Materials and methods
2.1. Islet-cell preparations Pancreatic islets were isolated by the collagenase technique from the pancreas of fed albino rats. The method
R2
F. Van Eylen et al. I Molecular and Cellular Endocrinology 106 (1994) RI-R5
used to isolate pancreatic islet cells has been described elsewhere (Gobbe et al., 1989). 2.2. Solutions The media used to incubate the islet cells consisted of a Krebs-Ringer bicarbonate-buffered solution (pH 7.4, 37°C) with the following composition (in mM): NaCl 115, CaC12 1, MgCl* 1, NaHC03 24. The media were equilibrated against a mixture of O2 (95%) and COZ (5%). In some media, NaHC03 was replaced by Hepes/NaOH (10 mM, gassed with ambient air) to avoid precipitation in the presence of La3+ (45Ca uptake experiments). In some experiments, NaCl was iso-osmotically replaced by sucrose (241 mM, Merck, Darmstadt, Germany) and Hepes/NaOH by Hepes/KOH. The different media also contained, when required, glucose (Merck), nifedipine (Calbiochem, La Jolla, USA), tetraethylammonium chloride (TEA Merck), BAY K 8644 (Calbiochem), ouabain (Sigma Chemical Co, St Louis, USA) and FMRFamide-related peptides. 2.3. Peptides The peptide FMRFa was obtained from Sigma. The other three peptides VMRFa, HMRFa and FLRFa (PheLeu-Arg-Phe-NHz) were synthesized by solid phase methodology with an automated 43 1A Applied Biosystem apparatus (Foster City, CA, USA) using the Fmoc strategy: a 4(2’,4’-dimethoxyphenylfluoren-9-yl-methoxyc~bonyl~inomethyl)phenoxy resin and fluoren-9-yl-methoxycarbonyllabelled amino acids activated with N-hydroxybenzotriazole and benzotriazol- 1-yloxytris (dimethyl amino) phosphonium hexafluorophosphate were used (Fournier et al., 1989). The peptides were purified by HPLC and their conformity established by both total amino acid composition and Edman degradation in a 477A sequencer (Applied Biosystems). All reagents for peptide synthesis were purchased from Novabiochem (Laiifelfingen, Switzerland). 2.4. 45Ca uptake The method used for the measurement of 45Ca uptake in isolated pancreatic islet cells has been described previously (Plasman et al., 1990). In brief, the islet cells were preincubated in 1 ml of a non-radioactive solution during 30 min and then incubated for 5 min in 1 ml of the same medium containing in addition 45Ca (lOyCilm1). FMRFa and related peptides were added to both pre-incubation and incubation media. At the end of the incubation, the cells were separated from the incubation medium by using a combined lanthanum and oil technique (Plasman et al., 1990). Na/Ca exchange was evaluated by measuring Nq+dependant 45Ca0 uptake. After pre-incubation, the islet cells were exposed to Na+-depleted media containing 45Ca. 2.5. @Rb uptake The method used for the measurement of 86Rb uptake in isolated pancreatic islet cells has been described previously (Lebrun et al., 1983). In brief, groups of 10 islets each were
placed in polyethylene microcentrifuge tubes and preincubated for 30 min in 0.06 ml of a non-radioactive medium placed on the top of a layer of oil (Versilube F50, General Electric, Waterford, USA). The islets were then incubated for a further 5 min in 0.120 ml of the same medium containing in addition [6,6’(n)-3H]sucrose (1 .O mM; 2O~Cilml) and 86Rb (0.11 mM; 10pCilml). At the end of the incubation, the islets were separated from the incubation medium by centrifugation through the oil layer. 2 .6 . 86Rb outflow The method used for the measurement of 86Rb outflow has been described elsewhere (Lebrun et al., 1990). Briefly, groups of 100 islets were incubated for 60 min at 37°C in the presence of 16.7 mM glucose and 86Rb (0.15-0.25 mM; SO@i/ml). After incubation, the islets were washed three times and then placed in a perifusion chamber. The efflux of @‘Rb (cpm/min) was expressed as a fractional outflow rate (FOR, % of instantaneous islet content per min (Herchuelz et al., 1980; Lebrun et al., 1990). The results are expressed as means f SEM. The statistical significance of differences between data was assessed by using a non-paired Student’s &test for two and analysis of variance for multisample comparison. 3. Results In the presence of 2.8 mM glucose, FMRFa failed to inhibit Na/Ca exchange in rat pancreatic islet cells (P> 0.05; Fig. 1A). Related peptides such as HMRFa, VMRFa and FLRFa were more active, a weak inhibition (32 + 5%, 39 + 3%, 22 2 3 %, respectively) being observed for the three compounds at the highest concentration tested (lOA M, P < 0.01; Fig. 1A). Since membrane potential is known to affect Na/Ca exchange activity, the effects of the two most active peptides (HMRFa and VMRFa) were examined under depolarizing conditions. The cells were depolarized by exposure to glucose (11 .l mM) + TEA (20 mM) or to glucose + TEA + BAY K 8644 (1 PM). Glucose is the physiological stimulus of the B-cell, TEA is a K+ channel blocker (Atwater et al., 1979) and BAY K 8644 a Ca*+ channel opener (Lebrun and Atwater, 1985). Glucose is known to depolarize the pancreatic B cell and to induce electrical activity. In the presence of glucose, TEA and BAY K 8644 increase the mean open time of voltagesensitive Ca*+ channels. Under such experimental conditions, the islet cells were exposed to the peptides during the pre-incubation, and Na/Ca exchange was measured during the incubation period under non-depolarizing conditions in the presence of the peptides only. Nifedipine (5-20pM) was also added to the incubation medium to block any residual action of the depolarizing agents. In the presence of glucose + TEA, the inhibitory effect of the two most active peptides (HMRFa and VMRFa) was increased, a significant inhibition (27 f 3% and 23 + 5%, respectively) being observed at 1 PM (Fig. lB, P < 0.001). The effect was dose-
F. Van Eylen et (11.I Molecular and Cellular Endocrinology 106 (1994) RI-R5 -. o---o o---o l
12or
A
R3
HMRFa VMRFa FMRFa FLRFa
o--a
VMRFa
-
HMRFa
--_-_________
0
I
10-e peptides
(M)
peptldes
C -
10-5
1o-4
1o-3
(M)
D HMRFa
*--.o
VMRFa
-
HMRFa
60
10-S Peptides
’
10-4
’
1 10-3
(M)
Peptides
(M)
Fig. 1. (A) Effect of FMRFa, HMRFa, VMRFa and FLRFa on Nai+-dependent 45Ca uptake in intact cells (reverse Na/Ca exchange). Data am expressed in % of control values found at 0 mM extracellular Na+ in the absence of the peptides, after subtraction of basal uptake measured at 139 mM extracellular Na+. Mean f SEM refer to at least 20 individual samples in each case. Basal uptake (139 mM Na+) and control uptake (0 mM Na+) averaged (n = 91) 379 f 12 fmol WI000 cells and (n = 90) 1010 f 30 fmol WlOOO cells, respectively. (B) Effect of HMRFa and VMRFa on reverse Na/Ca exchange in intact islet cells. At variance with (A), the islet cells were pm-incubated in the presence of glucose (16.7 mM) and TEA (20 mM). F’resenWiona in (A). Number of individual samples was at least 20 in each case. Basal and control uptake averaged (n = 42) 311 f 10 fmol CaIlOOO cells and (n = 42) 850 f 19 fmol G/l000 cells, respectively. (C) Effect of HMRFa on reverse Na/Ca exchange in intact islet cells. At variance with (A). the islet cells were pre-incubated in the presence of glucose (16.7 mM), TEA (20 mM) and BAY K 8644 (I PM). Presentation as in (A). Number of individual samples was at least 22 in each case. Basal uptake and control uptake averaged (n = 22) 183 i 12 fmol WlOOO cells and (n = 24) 310 f 6 fmol CaIlOOOcells, respectively. (D) Effect of HMRFa and VMRFa on 45Ca uptake through voltage-sensitive Ca2+ channels. Presentation as in (A). Control values were measured in the presence of KCI 25 mM. Mean * SEM refer to at least 21 individual samples in each case. Basal uptake and control uptake averaged (n = 44) 300 f 8 fmol CaIlOOOcells and (n = 44) 405 f 7 fmol G/l000 cells, respectively.
related but the inhibition was not complete even at the highest concentration (1OOpM). The exposure of the islet cells to glucose + TEA + BAY K 8644 further increased the inhibitory action of one of the two latter peptides (VMRFa), an almost complete inhibition being observed at 100pM (Fig. lC, P c 0.001). To determine the specificity of the action of the peptides on Na/Ca exchange, we examined their effects on voltagesensitive Ca2+ channels, on the Na+-K+-ATPase and on K+ channels. Fig. 1D shows that HMRFa and VMRFa did not affect 45Ca uptake induced by 25 mM extracellular K+ (P > 0.05), suggesting that the peptides do not interfere with voltage-sensitive Ca2+ channels. Similarly, the peptides did not inhibit the Na-K-ATPase. Indeed, VMRFa did not affect 86Rb uptake from islets exposed to 8.3 mM glu-
cose (P > 0.05), at variance with ouabain which inhibited 86Rb uptake by about 54% (P < 0.001, Fig. 2). At a concentration of 10 PM and 25 PM, VMRFa did not affect 86Rb outflow from islets perifused in the presence of 2.8 mM glucose (Fig. 3). However, in the presence of 16.7 mM glucose, VMRFa exerted a modest and rapidly reversible inhibitory effect. A similar modest inhibitory effect was also seen at a lower concentration of VMRFa (1 PM). 4. Discussion In cardiac sarcolemmal vesicles, FMRFa inhibited Na/Ca exchange with an ICsa of 750pM (Khananshvili et al., 1993). FLRFa inhibited Na/Ca exchange more potently (I& = 60pM), whilst derivatives with the NH2 terminal
R4
F. Van Eylen et al. I Molecular Effect
of VMRFa
Ctrl Ouab
1
on
10
86Rb
25
utdake
50
VMRFa Fig. 2. Effect of VMRFa
and Cellular Endocrinology
100
75
(uM)
and ouabbain (1 mM; ouab) on @Rb uptake by
isolated islet cells (Ctrl = control).
Phe substituted by Val or His (VMRFa and HMRFa) were 300-SOO-fold more potent showing an ICso of 1.5pM and 2.5pM, respectively (Khananshvili et al., 1993). In the present study, carried out in intact cells, HMRFa, VMRFa and FLRFa peptides were far less potent than in cardiac sarcolemmal vesicles. However, it is interesting to notice that, as in cardiac sarcolemmal vesicles, HMRFa, VMRFa and FLRFa were more active than FMRFa. The difference in potency between cardiac vesicles and intact B cells may result from the poor access of the peptides to the inhibitory binding site that would not be exposed extracellularly. Indeed, the cardiac sarcolemmal vesicle preparation represents a mixture of inside-out and outside-out vesicles and
Glu 2.8 jGlu 2.8; .-.-. f.-.-.*-.-.-.-.
Glu
2 8
[ VMRFa j j 10 UM ; o-o-o-fo-o-c-p-o-o-0-0
5.0
Glu 16 7;Glu 16.7: Glu 16 7
45
106 (1994) RI-R5
only inside-out vesicles contribute to most, if not all, of the Na/Ca exchange activity (Li et al., 1991). The reason why the inhibitory activity of the peptides was increased under depolarizing conditions is unclear and the present study provides no further insight into the nature of this effect. Conceivably, it could result from an increased binding of the peptide to its receptor under depolarizing conditions. Indeed, Na/Ca exchange is thought to work according to a consecutive reaction mechanism (Hilgeman et al., 1991). In such a model, there is only one set of binding sites that bind either Na+ or Ca*+ and that is exposed at only one surface of the membrane. After binding, the ion crosses the membrane, is released at the other side and the active site may then bind the other cation. If it is postulated that the peptides bind to this cationic (Na/Ca) binding site or to a site located in its close vicinity and that this site is exposed intracellularly under resting conditions, then membrane depolarization during the period of pre-incubation would increase the probability of peptide binding to its site. Indeed, membrane depolarization would increase [Ca2+li, enhance Na/Ca exchange activity and hence increase the time of exposure of the binding site to the extracellular surface of the membrane. The study of the effect of the peptides after intracellular injection may help to confirm this view. Neither voltage-sensitive Ca2+-channels nor the Na/KATPase were blocked by one of the two most active peptides that nevertheless moderately inhibited K+ permeability. Since this effect of the peptide was seen only in the presence of a high concentration of glucose (16.7 mM), our data indicate that the peptide acted more probably on Ca*+sensitive K+ channels than on ATP-sensitive K+ channels. Indeed, in the presence of a high glucose concentration, ATP-sensitive K+ channels are inhibited while Ca2+sensitive K+ channels are activated (Atwater et al., 1989). Our finding is in agreement with previous work showing an inhibitory activity of FMRFa-related peptides on Ca*+sensitive K+ channels in the Helix giant serotonin neurons (Cottrell, 1982). Whatever the exact identity of the K+ channel affected, the present study shows that FMRFarelated peptides are weak and non-specific inhibitors of Na/Ca exchange in intact (B) cells. Therefore, their interest as tools for the study of Na/Ca exchange in intact cells appears limited. In addition, our data do not favour the view that FMRFa-related peptides may exert some of their physiological modulatory role by inhibiting N&a exchange. Acknowledgments
30
Fig. 3. Effect of VMRFu throughout the experiment cose.
40
(IOpM)
50
60
Time
(mln)
70
80
90
on 86Rb outflow from islets perifused
in the presence of 2.8 mM or 16.7 mM glu-
The authors are indebted to A. Van Praet and M. Hermann for technical assistance and to P. Surardt for secretarial help. This work was supported by the Belgian Fund for Medical Scientific Research (No 3.4547.9 1, 9.4584. 90 and 9.45 14.93) from which F. Van Eylen holds a bursary and P. Lebrun is Senior Research Associate.
F. Van Eylen et al, I M~le~u~~rand Cellular E~o~r~no~~gy 106 (1994) RI-R5
References Atwater, I., Carrel, P. and Li, M.X. (1989) in Insulin Secretion (Draznin, B., Shlomo, M. and Derek, V., eds.), pp. 49-68, Alan R. Liss, New York. Atwater, I., Ribdet, B. and Rojas, E. (1979) J. Physiol. (London) 288, 561-574. Blaustein, M.P. (1988) J. Cardiovasc. Pharmacol. 12 (suppl. S), S56568. Bern, D.M. (1991) Ann. N. Y. Acad. Sci. 639,375-385. Carafoli, E. (1988) Methods Bnzymol. 157,3-l 1. Cottrell. G.A. (1982) Nature 2%,87-89. Foumier, A.. Dahno, W. and Felix, A.M. (1989) Int. J. Pept. Protein Res. 33,133-139. Gobbe, P. and Herchuelz, A. (1989) Res. Commun. Chem. Pathol. Pharmacoi. 63.23 f-247. Herchuelz, A., Sener, A. and Ma&se, W.J. (1980) J. Membr. Biol. 57, I-12.
R5
Hilgemann, D.W., Nicoil, D.A. and Philipson. K.D. (1991) Nature 352, 715-718. Khananshvili, D. and Same, Y. (1992) Life Sci. 51,275-283. Khananshvili. D.. Price, DC., Greenberg, M.J. and Same, Y. (1993) J. Biol. Chem. 268,200-205. Lebrun, P., Antoine, M.H., Devreux, V., Hermann, M. and Herchuelz, A. (1990) J. Pharmacol. Exp. Ther. 255,948-954. Lebrun, P. and Atwater 1. (1985) Biophys. J. 48.9 19-930. Lebrun, P., Malaisse, W.J. and Herchuelz, A. (1983) 3. Membr. Biol. 74, 67-73. Li, 2.. Nicoil, D.A., Collins, A., Hilgemann, D.W., Elioteo, A.G., Penniston, J.T., Weiss, J.N., Tom&h, J.M. and Philipson, K.D. (1991) J. Bioi. Chem. 266.10161020. Plasmaa, P.O., kbrun, P. and Herchuelz, A. (1990) Am. J. Physiot. 259, E84dE850. Price, DC. and Greenberg, M.J. (1989) Biol. Bull. 177, 198-205. Raffa, R. (1988) Peptides 9.915-922.
Molecular
and Cellular Endocrinology
106 (1994) Rl-RS
Inhibition of Na/Ca exchange by Phe-Met-Arg-Phe-NH2 (FMRFa)-related peptides in intact rat pancreatic B-cells F. Van Eylena, P. Gourletb, A. Vandermeersb, P. Lebruna, A. Herchuelza-* “Laboratory of Pharmacology, bL.uboratory of Biochemistry, Brussels University School of Medicine, Bat. GE, 808 route de Lennik, B-1070, Brussels, Belgium Received 20 July 1994; accepted
1 August 1994
Abstract Phe-Met-Arg-Phe-NH2 (FMRFa)-related peptides were recently shown to inhibit Na/Ca exchange in cardiac sarcolemmal vesicles. In the present study, we examined the effects of FMRFa-related peptides on Na/Ca exchange in intact (pancreatic B) cells. At 2.8 mM glucose, FMRFa-related peptides only weakly inhibited Na/Ca exchange although their effect was more marked under depolarizing conditions. The peptides blocked neither the Na/K-ATPase nor Ca *+ channels but slightly reduced membrane K+ permeability. Our data indicate that PMRFa-related peptides are weak and non-specific inhibitors of Na/Ca exchange in intact B cells. The data do not confirm the view that the peptides may exert some of their physiological modulatory role by inhibiting Na/Ca exchange. Keywor&:
Na/Ca exchange; FMRFamide-related peptides; Pancreatic B-cell
1. Introduction Two processes located at the plasma membrane appear to mediate Ca*+ extrusion from cells: the Ca*+-ATPase and Na/Ca exchange (Carafoli, 1988; Blaustein, 1988). In the heart, Na/Ca exchange appears to be the predominant mechanism for Ca*+ extrusion, being able to restore and control basal Ca*+ level on a beat-to-beat basis (Bers, 1991). However, the role played by the exchanger in many other cell types remains to be elucidated. One major factor that has hindered rapid progress in the knowledge of the contribution of NaKa exchange to the control of Ca*+ homeostasis, is the lack of specific inhibitors of the exchanger. Recently, opioid antagonists (e.g. naloxone) were shown to inhibit Na/Ca exchange in cardiac sarcolemmal vesicles (Khananshvili and Sarne, 1992). Similarly, Phe-Met-Arg-
Abbreviations: FLRFa, Phe-Leu-Arg-Phe-NH2; FMRFa, Phe-Met-ArgPhe-NH2; HMRFa, His-Met-Arg-Phe-NH2; TEA, tetraethylammonium cl$oride; VMRFa, Val-Met-Arg-Phe-NH2. Corresponding author. Laboratoire de Pharmacodynamie et de therapeutique, Universit6 Libre de Bruxelles, Faculte de Mtdecine, Route de Lennik, 808-Bfitiment G.E., B-1070 Bruxelles, Belgium. Tel. (32) 2 555 62 01. Fax (32) 2 55.5 63 70. E-mail: [email protected].
0 1994 Elsevier Science Ireland Ltd. All rights reserved 0303-7207/94/$07.00 SSDI 0303-7207(94)03394-Z
Phe-NH2 (FMRFa)-related peptides with naloxone-like activity were observed to provoke complete inhibition of Na/Ca exchange in cardiac sarcolemmal vesicles (Khananshvili et al., 1993). The two most potent inhibitors were the derivatives with NH,-Phe substituted by either Val or His (VMRFa and HMRFa). In addition, it was suggested that endogenous analogs of FMRFa may exert some of their physiological modulatory role by inhibiting Na/Ca exchange (Khananshvili et al., 1993). FMRFa-related peptides have been localized in several mammalian tissues and appear to exert various actions including antagonism to opioid analgesia (Price and Greenberg, 1989; Raffa, 1988). Because the effect of FMRFa-related peptides was studied in sarcolemmal vesicles, namely in a mixture of inside-out and outside-out vesicles, the view that such peptides may inhibit Na/Ca exchange in intact cells remains to be demonstrated. In the present study, we have examined the effects of FMRFa-related peptides on Na/Ca exchange in intact rat pancreatic B-cells. 2. Materials and methods
2.1. Islet-cell preparations Pancreatic islets were isolated by the collagenase technique from the pancreas of fed albino rats. The method
R2
F. Van Eylen et al. I Molecular and Cellular Endocrinology 106 (1994) RI-R5
used to isolate pancreatic islet cells has been described elsewhere (Gobbe et al., 1989). 2.2. Solutions The media used to incubate the islet cells consisted of a Krebs-Ringer bicarbonate-buffered solution (pH 7.4, 37°C) with the following composition (in mM): NaCl 115, CaC12 1, MgCl* 1, NaHC03 24. The media were equilibrated against a mixture of O2 (95%) and COZ (5%). In some media, NaHC03 was replaced by Hepes/NaOH (10 mM, gassed with ambient air) to avoid precipitation in the presence of La3+ (45Ca uptake experiments). In some experiments, NaCl was iso-osmotically replaced by sucrose (241 mM, Merck, Darmstadt, Germany) and Hepes/NaOH by Hepes/KOH. The different media also contained, when required, glucose (Merck), nifedipine (Calbiochem, La Jolla, USA), tetraethylammonium chloride (TEA Merck), BAY K 8644 (Calbiochem), ouabain (Sigma Chemical Co, St Louis, USA) and FMRFamide-related peptides. 2.3. Peptides The peptide FMRFa was obtained from Sigma. The other three peptides VMRFa, HMRFa and FLRFa (PheLeu-Arg-Phe-NHz) were synthesized by solid phase methodology with an automated 43 1A Applied Biosystem apparatus (Foster City, CA, USA) using the Fmoc strategy: a 4(2’,4’-dimethoxyphenylfluoren-9-yl-methoxyc~bonyl~inomethyl)phenoxy resin and fluoren-9-yl-methoxycarbonyllabelled amino acids activated with N-hydroxybenzotriazole and benzotriazol- 1-yloxytris (dimethyl amino) phosphonium hexafluorophosphate were used (Fournier et al., 1989). The peptides were purified by HPLC and their conformity established by both total amino acid composition and Edman degradation in a 477A sequencer (Applied Biosystems). All reagents for peptide synthesis were purchased from Novabiochem (Laiifelfingen, Switzerland). 2.4. 45Ca uptake The method used for the measurement of 45Ca uptake in isolated pancreatic islet cells has been described previously (Plasman et al., 1990). In brief, the islet cells were preincubated in 1 ml of a non-radioactive solution during 30 min and then incubated for 5 min in 1 ml of the same medium containing in addition 45Ca (lOyCilm1). FMRFa and related peptides were added to both pre-incubation and incubation media. At the end of the incubation, the cells were separated from the incubation medium by using a combined lanthanum and oil technique (Plasman et al., 1990). Na/Ca exchange was evaluated by measuring Nq+dependant 45Ca0 uptake. After pre-incubation, the islet cells were exposed to Na+-depleted media containing 45Ca. 2.5. @Rb uptake The method used for the measurement of 86Rb uptake in isolated pancreatic islet cells has been described previously (Lebrun et al., 1983). In brief, groups of 10 islets each were
placed in polyethylene microcentrifuge tubes and preincubated for 30 min in 0.06 ml of a non-radioactive medium placed on the top of a layer of oil (Versilube F50, General Electric, Waterford, USA). The islets were then incubated for a further 5 min in 0.120 ml of the same medium containing in addition [6,6’(n)-3H]sucrose (1 .O mM; 2O~Cilml) and 86Rb (0.11 mM; 10pCilml). At the end of the incubation, the islets were separated from the incubation medium by centrifugation through the oil layer. 2 .6 . 86Rb outflow The method used for the measurement of 86Rb outflow has been described elsewhere (Lebrun et al., 1990). Briefly, groups of 100 islets were incubated for 60 min at 37°C in the presence of 16.7 mM glucose and 86Rb (0.15-0.25 mM; SO@i/ml). After incubation, the islets were washed three times and then placed in a perifusion chamber. The efflux of @‘Rb (cpm/min) was expressed as a fractional outflow rate (FOR, % of instantaneous islet content per min (Herchuelz et al., 1980; Lebrun et al., 1990). The results are expressed as means f SEM. The statistical significance of differences between data was assessed by using a non-paired Student’s &test for two and analysis of variance for multisample comparison. 3. Results In the presence of 2.8 mM glucose, FMRFa failed to inhibit Na/Ca exchange in rat pancreatic islet cells (P> 0.05; Fig. 1A). Related peptides such as HMRFa, VMRFa and FLRFa were more active, a weak inhibition (32 + 5%, 39 + 3%, 22 2 3 %, respectively) being observed for the three compounds at the highest concentration tested (lOA M, P < 0.01; Fig. 1A). Since membrane potential is known to affect Na/Ca exchange activity, the effects of the two most active peptides (HMRFa and VMRFa) were examined under depolarizing conditions. The cells were depolarized by exposure to glucose (11 .l mM) + TEA (20 mM) or to glucose + TEA + BAY K 8644 (1 PM). Glucose is the physiological stimulus of the B-cell, TEA is a K+ channel blocker (Atwater et al., 1979) and BAY K 8644 a Ca*+ channel opener (Lebrun and Atwater, 1985). Glucose is known to depolarize the pancreatic B cell and to induce electrical activity. In the presence of glucose, TEA and BAY K 8644 increase the mean open time of voltagesensitive Ca*+ channels. Under such experimental conditions, the islet cells were exposed to the peptides during the pre-incubation, and Na/Ca exchange was measured during the incubation period under non-depolarizing conditions in the presence of the peptides only. Nifedipine (5-20pM) was also added to the incubation medium to block any residual action of the depolarizing agents. In the presence of glucose + TEA, the inhibitory effect of the two most active peptides (HMRFa and VMRFa) was increased, a significant inhibition (27 f 3% and 23 + 5%, respectively) being observed at 1 PM (Fig. lB, P < 0.001). The effect was dose-
F. Van Eylen et (11.I Molecular and Cellular Endocrinology 106 (1994) RI-R5 -. o---o o---o l
12or
A
R3
HMRFa VMRFa FMRFa FLRFa
o--a
VMRFa
-
HMRFa
--_-_________
0
I
10-e peptides
(M)
peptldes
C -
10-5
1o-4
1o-3
(M)
D HMRFa
*--.o
VMRFa
-
HMRFa
60
10-S Peptides
’
10-4
’
1 10-3
(M)
Peptides
(M)
Fig. 1. (A) Effect of FMRFa, HMRFa, VMRFa and FLRFa on Nai+-dependent 45Ca uptake in intact cells (reverse Na/Ca exchange). Data am expressed in % of control values found at 0 mM extracellular Na+ in the absence of the peptides, after subtraction of basal uptake measured at 139 mM extracellular Na+. Mean f SEM refer to at least 20 individual samples in each case. Basal uptake (139 mM Na+) and control uptake (0 mM Na+) averaged (n = 91) 379 f 12 fmol WI000 cells and (n = 90) 1010 f 30 fmol WlOOO cells, respectively. (B) Effect of HMRFa and VMRFa on reverse Na/Ca exchange in intact islet cells. At variance with (A), the islet cells were pm-incubated in the presence of glucose (16.7 mM) and TEA (20 mM). F’resenWiona in (A). Number of individual samples was at least 20 in each case. Basal and control uptake averaged (n = 42) 311 f 10 fmol CaIlOOO cells and (n = 42) 850 f 19 fmol G/l000 cells, respectively. (C) Effect of HMRFa on reverse Na/Ca exchange in intact islet cells. At variance with (A). the islet cells were pre-incubated in the presence of glucose (16.7 mM), TEA (20 mM) and BAY K 8644 (I PM). Presentation as in (A). Number of individual samples was at least 22 in each case. Basal uptake and control uptake averaged (n = 22) 183 i 12 fmol WlOOO cells and (n = 24) 310 f 6 fmol CaIlOOOcells, respectively. (D) Effect of HMRFa and VMRFa on 45Ca uptake through voltage-sensitive Ca2+ channels. Presentation as in (A). Control values were measured in the presence of KCI 25 mM. Mean * SEM refer to at least 21 individual samples in each case. Basal uptake and control uptake averaged (n = 44) 300 f 8 fmol CaIlOOOcells and (n = 44) 405 f 7 fmol G/l000 cells, respectively.
related but the inhibition was not complete even at the highest concentration (1OOpM). The exposure of the islet cells to glucose + TEA + BAY K 8644 further increased the inhibitory action of one of the two latter peptides (VMRFa), an almost complete inhibition being observed at 100pM (Fig. lC, P c 0.001). To determine the specificity of the action of the peptides on Na/Ca exchange, we examined their effects on voltagesensitive Ca2+ channels, on the Na+-K+-ATPase and on K+ channels. Fig. 1D shows that HMRFa and VMRFa did not affect 45Ca uptake induced by 25 mM extracellular K+ (P > 0.05), suggesting that the peptides do not interfere with voltage-sensitive Ca2+ channels. Similarly, the peptides did not inhibit the Na-K-ATPase. Indeed, VMRFa did not affect 86Rb uptake from islets exposed to 8.3 mM glu-
cose (P > 0.05), at variance with ouabain which inhibited 86Rb uptake by about 54% (P < 0.001, Fig. 2). At a concentration of 10 PM and 25 PM, VMRFa did not affect 86Rb outflow from islets perifused in the presence of 2.8 mM glucose (Fig. 3). However, in the presence of 16.7 mM glucose, VMRFa exerted a modest and rapidly reversible inhibitory effect. A similar modest inhibitory effect was also seen at a lower concentration of VMRFa (1 PM). 4. Discussion In cardiac sarcolemmal vesicles, FMRFa inhibited Na/Ca exchange with an ICsa of 750pM (Khananshvili et al., 1993). FLRFa inhibited Na/Ca exchange more potently (I& = 60pM), whilst derivatives with the NH2 terminal
R4
F. Van Eylen et al. I Molecular Effect
of VMRFa
Ctrl Ouab
1
on
10
86Rb
25
utdake
50
VMRFa Fig. 2. Effect of VMRFa
and Cellular Endocrinology
100
75
(uM)
and ouabbain (1 mM; ouab) on @Rb uptake by
isolated islet cells (Ctrl = control).
Phe substituted by Val or His (VMRFa and HMRFa) were 300-SOO-fold more potent showing an ICso of 1.5pM and 2.5pM, respectively (Khananshvili et al., 1993). In the present study, carried out in intact cells, HMRFa, VMRFa and FLRFa peptides were far less potent than in cardiac sarcolemmal vesicles. However, it is interesting to notice that, as in cardiac sarcolemmal vesicles, HMRFa, VMRFa and FLRFa were more active than FMRFa. The difference in potency between cardiac vesicles and intact B cells may result from the poor access of the peptides to the inhibitory binding site that would not be exposed extracellularly. Indeed, the cardiac sarcolemmal vesicle preparation represents a mixture of inside-out and outside-out vesicles and
Glu 2.8 jGlu 2.8; .-.-. f.-.-.*-.-.-.-.
Glu
2 8
[ VMRFa j j 10 UM ; o-o-o-fo-o-c-p-o-o-0-0
5.0
Glu 16 7;Glu 16.7: Glu 16 7
45
106 (1994) RI-R5
only inside-out vesicles contribute to most, if not all, of the Na/Ca exchange activity (Li et al., 1991). The reason why the inhibitory activity of the peptides was increased under depolarizing conditions is unclear and the present study provides no further insight into the nature of this effect. Conceivably, it could result from an increased binding of the peptide to its receptor under depolarizing conditions. Indeed, Na/Ca exchange is thought to work according to a consecutive reaction mechanism (Hilgeman et al., 1991). In such a model, there is only one set of binding sites that bind either Na+ or Ca*+ and that is exposed at only one surface of the membrane. After binding, the ion crosses the membrane, is released at the other side and the active site may then bind the other cation. If it is postulated that the peptides bind to this cationic (Na/Ca) binding site or to a site located in its close vicinity and that this site is exposed intracellularly under resting conditions, then membrane depolarization during the period of pre-incubation would increase the probability of peptide binding to its site. Indeed, membrane depolarization would increase [Ca2+li, enhance Na/Ca exchange activity and hence increase the time of exposure of the binding site to the extracellular surface of the membrane. The study of the effect of the peptides after intracellular injection may help to confirm this view. Neither voltage-sensitive Ca2+-channels nor the Na/KATPase were blocked by one of the two most active peptides that nevertheless moderately inhibited K+ permeability. Since this effect of the peptide was seen only in the presence of a high concentration of glucose (16.7 mM), our data indicate that the peptide acted more probably on Ca*+sensitive K+ channels than on ATP-sensitive K+ channels. Indeed, in the presence of a high glucose concentration, ATP-sensitive K+ channels are inhibited while Ca2+sensitive K+ channels are activated (Atwater et al., 1989). Our finding is in agreement with previous work showing an inhibitory activity of FMRFa-related peptides on Ca*+sensitive K+ channels in the Helix giant serotonin neurons (Cottrell, 1982). Whatever the exact identity of the K+ channel affected, the present study shows that FMRFarelated peptides are weak and non-specific inhibitors of Na/Ca exchange in intact (B) cells. Therefore, their interest as tools for the study of Na/Ca exchange in intact cells appears limited. In addition, our data do not favour the view that FMRFa-related peptides may exert some of their physiological modulatory role by inhibiting N&a exchange. Acknowledgments
30
Fig. 3. Effect of VMRFu throughout the experiment cose.
40
(IOpM)
50
60
Time
(mln)
70
80
90
on 86Rb outflow from islets perifused
in the presence of 2.8 mM or 16.7 mM glu-
The authors are indebted to A. Van Praet and M. Hermann for technical assistance and to P. Surardt for secretarial help. This work was supported by the Belgian Fund for Medical Scientific Research (No 3.4547.9 1, 9.4584. 90 and 9.45 14.93) from which F. Van Eylen holds a bursary and P. Lebrun is Senior Research Associate.
F. Van Eylen et al, I M~le~u~~rand Cellular E~o~r~no~~gy 106 (1994) RI-R5
References Atwater, I., Carrel, P. and Li, M.X. (1989) in Insulin Secretion (Draznin, B., Shlomo, M. and Derek, V., eds.), pp. 49-68, Alan R. Liss, New York. Atwater, I., Ribdet, B. and Rojas, E. (1979) J. Physiol. (London) 288, 561-574. Blaustein, M.P. (1988) J. Cardiovasc. Pharmacol. 12 (suppl. S), S56568. Bern, D.M. (1991) Ann. N. Y. Acad. Sci. 639,375-385. Carafoli, E. (1988) Methods Bnzymol. 157,3-l 1. Cottrell. G.A. (1982) Nature 2%,87-89. Foumier, A.. Dahno, W. and Felix, A.M. (1989) Int. J. Pept. Protein Res. 33,133-139. Gobbe, P. and Herchuelz, A. (1989) Res. Commun. Chem. Pathol. Pharmacoi. 63.23 f-247. Herchuelz, A., Sener, A. and Ma&se, W.J. (1980) J. Membr. Biol. 57, I-12.
R5
Hilgemann, D.W., Nicoil, D.A. and Philipson. K.D. (1991) Nature 352, 715-718. Khananshvili, D. and Same, Y. (1992) Life Sci. 51,275-283. Khananshvili. D.. Price, DC., Greenberg, M.J. and Same, Y. (1993) J. Biol. Chem. 268,200-205. Lebrun, P., Antoine, M.H., Devreux, V., Hermann, M. and Herchuelz, A. (1990) J. Pharmacol. Exp. Ther. 255,948-954. Lebrun, P. and Atwater 1. (1985) Biophys. J. 48.9 19-930. Lebrun, P., Malaisse, W.J. and Herchuelz, A. (1983) 3. Membr. Biol. 74, 67-73. Li, 2.. Nicoil, D.A., Collins, A., Hilgemann, D.W., Elioteo, A.G., Penniston, J.T., Weiss, J.N., Tom&h, J.M. and Philipson, K.D. (1991) J. Bioi. Chem. 266.10161020. Plasmaa, P.O., kbrun, P. and Herchuelz, A. (1990) Am. J. Physiot. 259, E84dE850. Price, DC. and Greenberg, M.J. (1989) Biol. Bull. 177, 198-205. Raffa, R. (1988) Peptides 9.915-922.