Molecular and Cellular Endocrinology, 88 (1992) 129-133 0 1992 Elsevier Scientific Publishers Ireland, Ltd. 0303.7207/92/$05.00

MOLCEL

129

02836

Pancreastatin

increases cytosolic Ca2+ in insulin secreting RINmSF cells

Victor Sk-xhez-Margalet,

Miguel Lucas and Raimundo Goberna

Department of Medical Biochemistry and Molecular Biology, Medical School of Seoilla, Hospital Virgen Macarena, Unioersity of Secilla, Sepilla 41009, Spain (Received

Key words: Pancreastatin;

Calcium;

3 June 1992; accepted

ATP; cY-Adrenergic

receptor

blocker;

29 June 1992)

Insulin

secreting

cell line (RINm5F)

Summary We have investigated the effect of pancreastatin on cytosolic Ca2+ concentration in the insulin secreting cell line RINmSF. Changes in [Ca*+], induced by pancreastatin were detected by Fluo-3 fluorescence using both flow cytometry and batch analysis measurements, and turned out to be from 90 to 315 nM equivalent to 80% of that caused by ATP, which increased [Ca2+li from 90 nM to 400 nM. This effect of pancreastatin did not depend on extracellular calcium and was not mediated by a-adrenergic receptors since it was not prevented by the a-blocker yohimbine. It is concluded that pancreastatin has a role in the homeostasis of free cytosolic calcium in the insulin secreting cell line RinmSF.

Introduction Pancreastatin, a 49 amino acid peptide was first isolated from the porcine pancreas by Tatemoto et al. (1986). The better known biological effect of pancreastatin is the inhibition of insulin release in vivo (Ahren et al., 1988; Funakoshi et al., 1988) or in vitro (Tatemoto et al., 1986; Efendie et al., 1987; Silvestre et al. 1988); however, extrapancreatic effects have been described such as inhibition of parathormone secretion in vitro

Correspondence to: V. Sanchez-Margalet, Department of Medical Biochemistry and Molecular Biology, Hospital Virgen Macarena, Medical School, University of Sevilla, Avda Sanchez Pizjuan 4, Sevilla 41009, Spain. Abbreviations: EGTA, ethylene glycol-his@-aminoethyl ether) N,N,N’,N’-tetraacetic acid; ATP, adenosine 5’-triphosphate.

(Fasciotto et al., 1989) and the stimulation of hepatic glycogenolysis in vivo (Sanchez et al., 19901 and in vitro (Sanchez et al., 1992). Moreover, pancreastatin is widely distributed throughout the central nervous system and in endocrine cells of the gastrointestinal tract and islets of pig and human pancreas (Ravazzola et al., 1988). With regard to the mechanism of action of pancreastatin, it is still incompletely understood. However, the inhibition of insulin secretion by pancreastatin from the insulin secreting cell line RINmSF has been shown to be reversed by pertussis toxin (Lorinet et al., 1989) in a similar way to that which occurred with a,-adrenergic inhibition of insulin release in islets and RINmSF cells (Wollheim and Sharp, 1981). We have studied the effect of pancreastatin on [Ca2+li from the RINmSF cells in the presence and absence of extracellular Ca*+ and the (Yblocker yohimbine.

130 Materials

and methods

Research Software. Additional scribed in the legends to figures.

Cell culture Clonal insulin-producing RINm5F cells were cultured and cell suspensions were prepared as described (Amiranoff et al., 1988). The cell suspensions were incubated at least 2 h at 37°C before any of the experimental procedures. Measurement of /Ca2 ‘I, RINmSF cells were loaded with the calcium sensitive fluorescein derived chromophore Fluo-3 (Minta et al., 1989) as described (Lucas and Solano, 19911, in phosphate-buffered saline (PBS) containing 1.3 mM CaCl,, 1.3 mM MgCl, and 2.7 mM glucose. Cells, 10 X 106/ml, were incubated for 40 min with 4 PM Fluo-3, from a stock of 2 mM in dimethylsulfoxide. Excitation and emission wavelengths were 488 and 530 nm respectively. The fluorescence recordings were performed either under batch conditions in a spectrofluorometer (Hitachi F-2000), which allowed constant stirring of the cell suspension at 37°C or by individual cell analysis in a flow cytometer at room temperature. The traces shown by the spectrofluorometer are representative for at least three different cell preparations. The calibration procedure includes the obtainment of F,,, and Fmin,which was achieved by disruption of the cells with digitonin and the ulterior addition of 2 mM EGTA respectively. Intracellular free calcium, [Ca’+],, was calculated by the equation:

details

are

de-

Chemicals Fluo-3 and Br-A23187 were purchased from Molecular Probes (Eugene, OR, USA). Porcine pancreastatin (33-49 C-terminal fragment) was obtained from Peninsula Laboratories Europe (Merseyside, UK) and ATP, clonidine, and yohimbine from Sigma Chemical (St. Louis, MO, USA). Results Effect of pancreastatin and ATP on [Ca”], Pancreastatin (10e7 M final concentration) caused a rapid increase in [Ca’+], of Fluo-3loaded RINm5F cells in the presence of 1.3 mM CaCl,, followed by a slow decrease to near basal levels. The ulterior addition of 10e4 M ATP caused a higher rise in [Ca2+li, and finally, the calcium ionophore Br-A23187 (1O-6 M) produced a sustained increase in [Ca2+li; (see changes in units of absorbance in the cytometer scale of Fig. 11. None of the agents used in the above described experiments altered the perme-

210. 200 190180170-

where K,, the dissociation constant bound Fluo-3 is 400 and F represents mental fluorescence.

for Ca2+the experi-

160I

0 Cytometry Flow cytometry measurements were performed on a FACScan (Becton Dickinson, San Jose, CA, USA). Excitation was from an argon laser at 488 nm, and emission at either 530 or 585 nm was measured on a linear scale. Forward and orthogonal light scattering and fluorescence signals were collected in list mode. Acquisition and analysis were performed with the FACScan and Chronys

50

100 Time

150

200

(s)

Fig. 1. Flow cytometry recording of fluorescence of Flue-3 loaded RINmSF cells. The excitation and emission wavelengths were 488 and 525 nm respectively. Cell (105/ml) suspensions were in PBS containing 1.3 mM CaCI,, 1.3 mM MgCl, and 2.7 mM glucose. At the indicated times (arrows), 1OV’ M 33-49 pancreastatin (PST), 10m4 M ATP and lo-’ M Br-A231 87 (Ionop) were added. Fluorescence intensity is given as the mean channel values obtained in the linear scale of the Chronys software. Results are the mean f SEM of four separate cell preparations.

131

ability of the plasma membrane of RINmSF cells to the vital dye propidium iodide. This was ascertained under the assay conditions of the flow cytometer experiments. Upon the addition of digitonin, a rapid increase of fluorescence was observed, as should be expected (see Fig. 2). We determined in batch analysis experiments performed at 37°C that lo-’ M pancreastatin increased [Ca2+li from 90 to 315 nM in RINmSF cells incubated at 37°C in the presence of 1.3 mM CaCl,. Further challenge with pancreastatin, following the recovering of basal [Ca2+], by cells, reproduced the same peak of [Ca2+li (not shown). The addition of 1O-4 M ATP was able to produce an additional calcium wave, reaching up to 400 nM (see Fig. 3A), the same value obtained with ATP alone (not shown). When calcium was omitted from the medium (Fig. 3B), the effect of pancreastatin was not modified, but the effect of ATP was almost completely abolished. Although the effect of ATP has been demonstrated to be due to release of Ca 2+ from internal stores (Arkhammar et al., 1990), it is worth to point out the low glucose concentration employed in our experiments which may explain our results, since glucose is necessary to maintain high Ca2+ concentrations within internal store sites which are sensitive to ATP. To further study the changes in [Ca2+li caused

P:T

A]P

Ionpp

Dig

100

I

I

B [Caa*li

I

lo-‘M 4oo

- 300

I

10 M ATP

I

PST

L

1

2oo 8Old I

I

I

1i4M

I

I

ATP

+

4

r

L 0

PST

M

nh I

60

I

1

I

120

I

180

I

I

240

Time (set) Fig. 3. Effect of ATP and pancreastatin on [CaZ+li. Dependence on extracellular calcium. RINmSF cells (10 X 106/ml) were loaded with Flue-3 in the presence (A) and in the absence (Bl of calcium. Pancreastatin (PST, 33-49 C-terminal fragment) and ATP were added from stocks to give the indicated concentrations. Batch analysis of fluorescence intensity was performed at 37°C in the Hitachi F-2000 spectrofluorometer. [Ca*+ Ii was calculated as indicated in Materials and methods. A single experiment representative of four others is shown.

by pancreastatin, we determined the dose response of pancreastatin in Fluo-3 loaded RINmSF cells. Fig. 4 shows changes in [Ca2+li at lo-15 s in response to increasing concentrations of pancreastatin (from lo- ” to 10p6). Maximal effect was achieved at lo-’ M, with an ED,, approaching nanomolar range.

i-~-I-~-I-*-I-*-I-*-i 50

-4

-7

10

150

Time (s) Fig. 2. Plasma membrane permeability to propidium iodide. Experiments were conducted by flow cytometry analysis of the fluorescence intensity at 585 nm. Arrows indicate the addition of lo-’ M PST (33-49 pancreastatinl, 10m4 M ATP, 10m6 M Br-A23187 (Ionopl and 5 pg/ml digitonin (Dig). Values, expressed as percent of maximal, are the mean f SEM of four separate experiments.

Effect of pancreastatin and clonidine on [Ca2 ‘Ii

The addition of clonidine to the suspension of Fluo-3-loaded cells resulted in an increase in [Ca2+li (Fig. 5A). Unlike pancreastatin, the effect of clonidine on [Ca2+li was prevented by the presence of the a-blocker yohimbine, which by contrast failed to modify the increase in [Ca2+li caused by pancreastatin (Fig. 5B).

132

Discussion

5 - .h

-s

8

-10

I

-9

-8

-7

-6

Fig. 4. Dose response of the pancreastatin-induced increase in cytosolic free Ca2+. Cytosolic free Ca2+ changes were determined as described in the legend to Fig. 3. The maximal cytosolic free Ca*+ concentration achieved for each dose of pancreastatin is shown. Values are the mean f SEM from four separate experiments.

I

0

I

I

I

60

I

I

120 Time

I

160

I

I

240

(set)

Fig. 5. Effect of the a-blocker yohimbine on pancreastatinand clonidine-induced calcium fluxes. [Ca2+ 1, of Fluo-3 loaded RINm5F cells measured in the spectrofluorometer, see Fig. 3. Cells were suspended in medium without (A) or with (B) lo-” M yohimbine. Arrows indicate the addition of pancreastatin (PST, 33-49 C-terminal fragment) and clonidine (CLON). A single experiment representative of three others is shown.

Changes in intracellular Cazf concentrations have a major role in the regulation of insulin secretion (Wollheim and Sharp, 19811, although Ca2+ uptake can exist in the absence of insulin release (Cook and Perara, 1982). We took advantage of the establishment of the clonal tumor-derived p-cell line RINmSF which permits generation of a sufficient number of cells for studies with Cazf indicators, such as Fluo-3. This study describes for the first time the effect of pancreastatin on cytosolic Ca2+. We have used the Cterminal fragment of porcine pancreastatin (3349) because of its biological activity in vivo and in vitro (Tatemoto et al., 1986; Funakoshi et al., 1988) as well as its homology with the corresponding region of rat chromogranin A (76%), the precursor of pancreastatin Oancangelo, 1988). Pancreastatin has been shown to act similarly to epinephrine in RINmSF cells (Lorinet et al., 1989) as well as in hepatocytes (Sanchez et al., 1992), since they both inhibit insulin release and produce hepatic glycogenolysis. Moreover, epinephrine has been described to increase cytosolic Ca2+ in RINmSF cells (Ullrich and Wollheim, 1985) in a similar fashion as we describe in the present work. To investigate if the effect of pancreastatin increasing cytosolic calcium in the insulin secreting RINmSF cells was mediated by a+receptors (Ullrich and Wollheim, 1989), we tested the effect of the a-blocker yohimbine, which suppresses the inhibition of insulin secretion as well as the rise of cytosolic Ca*+ induced by epinephrine in these cells (Ullrich and Wollheim, 1988). The resistance of this effect of pancreastatin on [Ca*+l, to yohimbine contrasts with the sensitivity of clonidine to the a-blocker, indicating that the peptide acts on other receptors, although the presence of specific pancreastatin receptors remains to be identified. The raised cytosolic Ca*+ induced by pancreastatin in calcium-free solutions is probably due to intracellular Ca2’ release from an inositol 1,4,5-triphosphate (InsPJ-sensitive calcium pool. However, whether this effect is mediated by InsP, remains to be studied. Pancreastatin increased [Ca2+li of RINmSF cells in a dose-dependent manner within a range

133

of concentrations which correlated with that previously shown for inhibition of insulin secretion (Lorinet et al., 1989). Concerning the inhibition of insulin secretion by pancreastatin it could be speculated to be due to the mobilization of Ca2+ from certain intracellular Ca*+ stores and, by emptying these pools of calcium, it may provoke the inhibition of insulin secretion. On the other hand, pancreastatin could act at a further step to second messengers, closer to the exocytotic event, as it has been proposed for epinephrine (Ullrich and Wollheim, 1988) and galanin (Sharp et al., 1989). However, further investigation is needed to clarify this point. We conclude that pancreastatin may have a role in the homeostasis of intracellular calcium in the insulin secreting cell line RINmSF. Acknowledgements

This work was supported Investigaciones Sanitarias de (92/390; 92/399X We thank Dr. B. Amiranoff and A.M. the RINmSF cells.

by the Fondo de la Seguridad Social Dr. M. Laburthe, Lorinet for gifts of

References Ahren, B., Lindskog, S., Tatemoto, K. and Efendic, S. (1988) Diabetes 37, 281-285. Amiranoff, B., Lorinet, A.M., Lagny-Pourmir, I. and Laburthe, M. (1988) Eur. J. Biochem. 177, 147-150. Arkhammar, P., Hallberg, A., Kindmark, H., Nilsson, T.,

Rorsman,

P. and Berggren,

P.O. (1990) Biochem.

J. 265,

203-211. Cook, D.L. and Perara, E. (1982) Diabetes 31, 985-990. Efendic, S., Tatemoto, K., Mutt, V., Quan, C., Chang, D. and &tenson, C.G. (1987) Proc. Natl. Acad. Sci. USA 84, 7257-7260. Fasciotto, B.H., Gorr, S.U., DeFranco, D.J., Levine, M.A. and Cohn, D.V. (1989) Endocrinology 125, 1617-1622. Funakoshi, A., Miyasaka, K., Kitani, K. et al., (1989) Regul. Pept. 24, 225-231. Iancangelo, A.L., Fischer-Colbrie, R., Keller, K.J., Brownstein, M.J. and Eiden, L.E. (1988) Endocrinology 122, 2339-2341. Lorinet, A.M., Tatemoto, K., Laburthe, M. and Amiranoff, B. (1989) Eur. J. Pharmacol. 160, 405-407. Lucas, M. and Solano, F. (1991) Biochem. Pharmacol. 42, 425-430. Minta, A., Kao, J.P.Y. and Tsien, R.Y. (1989) J. Biol. Chem. 264, 8171-8178. Ravazzola, M., Efendic, S., &tenson, CC., Tatemoto, K., Hutton, J.C. and Orci, L. (1988) Endocrinology 123, 227229. Sanchez, V., Calvo, J.R. and Goberna, R. (1990) Biosci. Rep. 10, 87-91. Sanchez, V., Lucas, M., Calve, J.R. and Goberna, R. (1992) Biochem. J. 284, 659-662. Sharp, G.W.G., Marchand-Brustel, Y.M., Yada, T. et al. (1989) J. Biol. Chem. 264, 7302-7309. Silvestre, R.A., Peiro, E., Miralles, P, Villanueva, M.L. and Marco, J. (1988) Life Sci. 42, 1361-1367. Tatemoto, K., Efendic, S., Mutt, V., Makk, G. and Barchas, J.D. (1986) Nature 324, 476-478. Ullrich, S. and Wollheim, C.B. (1985) Mol. Pharmacol. 28, 100-106. Ullrich, S. and Wollheim, C.B. (1988) J. Biol. Chem. 263, 8615-8620. Ullrich, S. and Wollheim, C.B. (1989) Acta Endocrinol. 121, 525-532. Wollheim, C.B. and Sharp, G.W.G. (1981) Physiol. Rev. 61, 914-973.