c.s#cakdlm(19Qo)11,

501-506

QLongmanGrcupUKLtdlQOO

Regulation by diacylglycerol of calcium-evoked amylase secretion from intact and permeabilized pancreatic acinar cells T. KOMABAYASHI’, J.S. McKINNEY and R.P. RUBIN Department of Pharmacology, Medicat College of Virginia, Richmond, Virginia, USA The role of diacylglycerol in the mechanism of amylase release was Abstract investigated in isolated rat pancreatic acinar cells. Carbachol produced a time-dependent and dose-related increase in diacylglycerol production which paralleled the time course of The addition of atropine to acinar cells pretreated with 100 PM amylase secretion. carbachol produced a lag in the fall in diacylglycerol levels, which was preceded by a prompt fall in cytosolic Ca2+ and amylase secretion. A threshold concentration of ionomycin amplified the modest action of dioctanoylglycerol on amylase secretion. Ca2+-evoked amylase release elicited by saponin permeabilized acinar cells was markedly These collective findings support the hypothesis that enhanced by dioctanoylglycerol. diacylglycerol alone is not an adequate messenger to mediate pancreatic amylase release, but does serve to modulate the actions of Ca2+. breakdown Stimulus-induced phosphoinositide appears to be a key biochemical event in the activation of amylase secretion in exocrine pancreas Ill. Thus exposure to Ca2+ mobilizing agonists such as darbachol and caerulein leads to the breakdown of phospholipase C-mediated phosphatidyl inositol 4,5-bisphosphate and the very rapid accumulation of inositol trisphosphate (InsPs) r2, 31. Moreover, the addition of InsPs to permeabilized pancreatic acinar cells causes a selective release of cellular Ca2+ from a Presentaddresses ‘Departments of Physiology & Pharmacology, Tokyo College of Pharmacy, Tokyo 192-03, Japan 2Department of Ph srmacology 8z Therapeutics, School of Medicine & Biomedical Sciences, State University of New York at Buffalo, Buffalo, NY 142 14, USA

non-mitochondrial pool, and alterations in stimulated [3Hl-InsPs levels in permeabilized acinar cells are accompanied by parallel changes in Ca2’ release f41. Such findings taken together support the view that hSP3 serves as a cellular messenger for Ca2+ mobilization in pancreatic acinar cells, as it does in other cell types that utilize Ca2’ as a cellular messenger. In addition to the generation of inositol phosphates, Ca2’ mobilizing agonists also stimulate diacylglycerol (DA@ production in pancreatic acini [5-Q. DAG acts as an endogenous activator of protein kinase C, which also has been identified and characterized in exocrine pancreas [g-11]. Evidence that protein kinase C is involved in agonist stimulated amylase release is suggested by the finding that the secretory activity of the phorbol ester phorbol lZJ3dibutyrate (PDBu), a 501

502

pharmacological activator of protein kinase C, is potentiated by the Ca2’ ionophore ionomycin [12]. Although them are several lines of evidence suggesting that Ca2+ may not be the sole cellular messenger for amylase release [13-151, evidence supporting a role for endogenous DAG as a positive modulator of pancreatic secretion elicited by Ca2’ mobilizing receptor agonists is still fragmentary. The present study evaluates further the role of DAG in pancreatic exocrine secretion by comparing agonist-induced DAG accumulation with changes in cytosolic Ca2+ and amylase secretion and by examining the action of a permeant diacylglycerol analogue on Ca2+-activated secretion in intact and permeable acinar cells. These results provide evidence that DAG serves to modulate the stimulatory action of Ca2+ in exocrine pancreas. However, in the absence of elevated cytosolic Ca2+, DAG is not a sufficient mediator of amylase secretion.

CELL cALcITm4

cells were resugended at a density of 3 x lo6 cells/ml in a Ca -f&e medium whose composition resembled that of cytosol [161. Cells were permeabilized by exposure to saponin (50 pg/ml for 10 min). In these studies, the permeabilized cells were treated with oligomycin, antimycin and 2,4dinitrophenol followed by the addition of ATP and au ATP-regenerating system to abolish responses to any residual intact cells and to ensure that the responses observed were entirely those of permeabilized cells [ 161. Amylaserelease Amylase secretion was measured as described previously [17] and is expressed as a percentage of total cell content In each experiment, control samples were taken at each time point, from which basal secretion was determined and subtracted from drug-stimulated values. Diacylglycerolassay

Materials and Methods Materials [JJ-~~]-ATP (4500 Ci/mmole) was purchased from Radiochemicals (Irvine, CA, USA). ICN Collagenase (0.27 units/mg) was obtained from Boehringer Mannheim (Indianapolis, IN, USA). Dioctanoylglycerol was purchased from Avanti Polar Lipids (Birmingham, AL, USA). Purified purchased from octyl-@-D-glucoside was CaIbiochem (San Diego, CA, USA). Cell preparations Pancreatic acinar cells were prepared from Sprague-Dawley rats (125-150 g) as described previously [16] and incubated in a Krebs-Henseleit medium containing (mM): NaCl 98; KC1 5; KFkPO4 1.2; CaC12 1.3; MgSO4 1.2; NaHC03 2.4; HEPES 10; and dextrose 11. The medium also contained soyabean trypsin inhibitor (0.1 mg/ml) and essential amino acids and was maintained at pH 7.4 under an atmosphere of 95% oxygen and 5% carbon dioxide. For experiments with permeabihzed cells, acinar

Endogenous DAG levels in pancreatic acinar cells were determined by the method originally developed by Preiss et al. [IS], as modified by Rider et al. [19] and Wright et al. [20]. The amount of DAG in acinar cells was determined from the conversion of endogenously generated DAG to phosphatidic acid by membranes prepared from Escherichia coli containing DAG kinase; the membranes were generously supplied by Dr Robert Dougherty of Duke University Medical Center. Mixed micelles were prepared by solubilizing an aliquot of dried lipid extract of acinar cells in 20 pl of 7.5% octyl-fl-D-glucoside/cardiohpin solution (7.5% octyl-fi-D-glucoside, 5 mM cardiolipin in 1 mM diethylenetriaminepentaacetic acid). To the solubilized lipid/octylglucoside solution was added 10 pl of 20 mM dithiothreitol, 10 pl of diluted E. coli membranes (1:lOOO; 5 ug protein), and buffer to a total volume of 90 u.l. The reaction was initiated by the addition of 10 p.l of 10 mM [y-‘?P]-ATP. After extraction, au aliquot of the lipid phase was subjected to thin layer chromatography on a Silica gel 60 plate and developed with chloroform/pyrid acid (50/30/7 : v/v/v). Following autoradiography, the

503

DIACYLGLYCEROL & Ca’+-EVOKED AMYLASE SECRETION

acinar cells which had been preloaded with the Ca2+ indicator dye Fura-2, as previously described [17]. The fluorescence of Furaloaded cells was Shimadzu (RF-5000) determined in a spectrofluorometer. Cytosolic Ca2’ was calculated from the ratio of the fluorescent intensity at the two excitation wavelengths, 340 and 380 nm.

130-

Results 0

5

10

16 Time

20

26

30

hid

Concentration-response and time course of Fig. 1 carbachol-stimulated DAG levels in pancreatic acinar cells. Acinar cells were preincubaled for 5 min and then exposed to 5 pM (filled squares), 10 pM (open circles) or 100 j&I (filled triangles) carbachol. Samples were taken and processed at the indicated time points. Results are means f SEM of duplicate determinations for 3-5 different experiments and are expressed as percent of basal values at ZIXUtime. Mean basal value of DAG was 1.9 f 0.21 nmol/106 cells

radioactive spot corresponding to phosphatidic acid was scraped off and counted by liquid scintillation spectrometry. In some experiments, chloroform / acetone / methanol / glacial acetic acid / water (10/4/3/2/l : v/v/v/v/v) was employed as the solvent s stem and gave essentially a similar separation of Y Known amounts of [ ?P]-phosphatidic acid. sn- 1,Zdioleoylglycerol similarly labelled and treated as experimental samples were converted in a linear manner to 13$]-phosphatidic acid in the range of 1.6-8 nmol. Thus, by comparing [3%]-phosphatidic from acid formed standard amounts of 1,2dioleoylglycerol and cellular lipid extracts, the DAG content of acinar cells was determined from the sample volume and specific activity of the ATP employed. [3%]-Phosphatidic acid formation was linear over the range of cell concentrations of 2-8 x lo6 cellsfml, and thus in all experiments cells were utilized at a concentration of 2-3 x lo6 cells/ml. All data points were measured at least in duplicate. Cytosolic pee Ca2+ measurements

Cytosolic ionic Ca2+ levels were determined in

Time course and concentration-dependence of carbachol-stimulated DAG levels 1 depicts the time course and Figure concentration-dependence of the net accumulation of DAG in carbachol-stimulated acinar cells. Carbachol concentrations as low as 5 pM produced a significant rise in DAG levels within 5 min after stimulation. In another series of experiments, 10 @I carbachol failed to elevate DAG levels after 10, 20 or 60 s. After a 60 s exposure to 10 p&l carbachol, DAG levels were 98 (+ 3) % of control (n = 4). Peak levels, which were approximately 25-30% above basal, were reached after 10 min. With 5’ pM carbachol, DAG levels began to decline after 10 min and fell to near basal levels by 30 min (Fig. 1). By contrast, with 10 p.M carbachol, DAG levels declined gradually after 15 min; and with 100 p.M carbachol, DAG levels were maintained above basal levels throughout the entire 30 min incubation period (Fig. 1). The addition of atropine to cells previously exposed to 10 p.lV carbachol produced a prompt fall in DAG content within 5 min. which was accompanied by a parallel termination of amylase release (data not shown). With 100 p.M carbachol, exposure to atropine elicited a fall in DAG content only after a 5 min lag period (Fig. 2A). In addition, the DAG level was maintained above basal levels throughout the duration of the experiment (Fig. 2A). Despite the lag in the fall in DAG levels, there was no lag in the decline in amylase secretion in response to 100 pM carbachol after atropine (Fig. 2B). Similarly, 100 @I carbachol elicited a marked rise in cytosolic Ca2+ which promptly decreased to near basal levels within 25 s after the addition of atropine (Fig. 2C).

CELL CALCIUM

504

150 r

140

10

A

-

r

\!

/:;,,:..;‘-

%,.,.. ...

‘. ._

1

x.,

8

“....,,_, ,..< ..’

.-I . ..--_........-..._.__.__ .____J i ! t .i

.I / ,./ ‘. I ...’

110 130: 120

v 3 -h I

..” loot’.’ 0

5

I

,

1

I

I

10

16

20

25

30

Time

0

5

15

10

Imid

Time Fig.

20

25

30

(mid

2 Time course of changes in (A) DAG levels, (II) amylase

release, and (C) free [Ca’+]i induced by carbachol and the subsequent addition of atmpine. Acinar cells were preincubated for 5 min and then exposed to 100 p.M carbachol (open symbols). Aver

5 min. atropine (5 @l)

(closed symbols) was added

(indicated by armw) A.

Results are means f SBM for 3-4

different experiments

performed in duplicate and expressed as percent of basal values B ; : 100

-

_..... .._ ____ ._.Q C. Results are means (rt SEM) of 4-5 determinations derived

‘&, _. ._._....._......_._

f o

at zem time. Basal DAG levels were 1.8 f 0.1 nmoU106 cells B. Results am means j: SEM for 4 different experiments from two independentexperiments

4 1

2

3

4

6

5 Time

7

6

6

10

(mid

Interaction between dioctanoylglyceroland Ca2’ in intactand permeable cells In intact acinar cells, the cell permeable DAG aualogue dioctauoylglycerol (diCs) caused a modest enhancement of amylase reIease in the concentration range 1004000 p.M; secretion peaked at 500 p.M (Table 1). A threshold concentration of ionomycin (0.3 @f) with regard to amylase release produced a further enhancement of the secretory response to diC8 (Table 1). At 500 and 1000 @VI diCs, secretion elicited by both stimuli was significantly greater than the additive effect of each stimulus alone. In permeabilized acinar cells, the secretory machinery is made responsive to the direct actions of Ca2+ without the involvement of receptor-linked processes. In such situations, the release of amylase is elevated as cytosolic Ca2’ is raised within the

Table 1 The synergisticeffect of ionomycin on amylase release induced by increasingconcentrations of dioctanoylglycerol(dies). Acinar cells were incubated for 30 min with varied concentrations of diC8 in the presence and absence of 0.3 JJ,Mionomycin. The n~ults are means f SEM of 3 experiments hmylase release (% diCa WI

Alone

of total)

+ lonomycin

100

0.9 f 0.2

0.7 f 0.2 1.6 f 0.2

250 500

2.3 f 0.3 3.8 f 0.5 2.5 F 0.5

3.6 f 0.5 6.7 f 0.8” 9.2 f 1.2*

0

1000

-

* Significantly&&rent fromthe sum of the values obtainedwith diC!sand ionomycinalone (P < 0.05)

micromolar range (0.1-10 p.M; Fig. 3). Further, Ca2+-evoked amylase release was markedly enhanced by diCs over the range of Ca2+ concentrations (Fig. 3).

DIACYLGLYCEROL

& Ca2+-EVOKED

Ah4YLASE SECRETION

Discussion The present findings demonstrate, by direct measurement of the net accumulation of cellular DAG in pancreatic acinar cells, a concentration- and time-dependent rise in cellular DAG levels in response to carbachol. Significant increases in DAG content were observed 5 min after exposure to 10 p.M carbachol, with peak levels attained after 10 min. A stimulation of DAG accumulation by a maximal stimulatory concentration of carbachol with regard to amylase release was not detected within the first minute of stimulation. Ca2’ mobilizing agonists elicit an early transient stimulation of DAG accumulation only when employed at supramaximal stimulatory concentrations [gl. Our inability to detect an early rise in DAG accumulation may be ascribed to an action of DAG kinase to catalyze the rapid conversion of DAG to phosphatidatc. Favoring this view are the findings that carbachol causes a very rapid and marked increase in the kinase activity in exocrine cells [21] and a rise in phosphatidate precedes the increment in DAG levels (R.P. Rubin & J.S. McKinney, unpublished observations). In cells previously treated with 100 pM carbachol and then exposed to atropine, DAG accumulation was temporarily maintained, while cytosolic Ca2+ and amylase release fell abruptly. I 18

Control

I-

800

1000

Ca” fnM) Fig. 3

Enhancement

permeabilized

cells.

by dicS

of @-evoked

Permeabilized

for 30 min in the presence

secretion from

acinar cells were incubated

of varied amounts of Ca2’ ad&d to a

constant EGTA concentration,

with or without 1000 pM diC8

505

These latter findings indicate that DAG alone is not a sufficient mediator of amylase secretion. In this context, carbachol elicits a rapid, transient increase in cytosolic Ca” in exocrine pancreas which is dependent on agonist concentration and bears a close dose-response relationship to that of carbachol-stimulated amylase release [12, 223. These findings suggest that Ca2’ plays a key role in initiating the secretory response. It is well established that phorbol esters, which activate protein kinase C, are not complete secretagogues when added alone to exocrine pancreas, but they do act synergistically with Ca2’ ionophoms which are effective secrctagogues [12]. Our study utilizing permeable cells substantiates the hypothesis that elevated levels of both Ca2’ and DAG produce a complete secretory response, presumably due to the ability of DAG to sensitize protein kinase C to physiologically relevant concentrations of cellular Ca . Our findings support previous studies utilizing permeabilized acinar cells, which demonstrated that 12-0-tetradecanoylphorbol 13-acetate (TPA), a potent activator of protein kinase C, was able to enhance both the Ca2’ sensitivity and the extent of amylase release in response to micromolar concentrations of Ca2’ [23, 241. Previous studies conducted on exocrine pancreas have demonstrated that the modulation of cytosolic Ca2+ is not the sole factor in mediating amylase secretion [13, 141, but that activation of amylase secretion requires stimulation of both Ca2+-dependent and protein kinase C-activated pathways [15]. In the present study, the permeable analogue dic8 was able to elicit a vigorous, dose-dependent secretory response in intact cells only in combination with a threshold concentration of ionomycin (see Table 1). This result, taken together with the fact that amylase release is reduced by less than 50% when protein kinase C is down-regulated or inhibited by drugs [15, 251, provides further support for the notion that the DAG-dependent protein kinase C pathway plays only a modulatory role in the secretory process. In conclusion, the main outcome of this work is the recognition that DAG synthesized in exocrine pancreas as a result of muscarinic receptor activation serves to modulate Ca2+-evoked amylase

506

Further experimentation is required to define the relative roles of the various phospholipase mediated pathways in DAG metabolism, which in secretion.

turn should provide new insights into cellular regulatory mechanisms that govern pancreatic amylase secretion.

Acknowledgements Thiswork

was supported by Grant AM28029 from the National Institutes of Health. We also thank Maureen Adolf for her expert assistance during the latter phase of this study.

References 1. Rubin RP. (1986) Inositol lipids and cell secretion. In Putney JW. Jr (ed.) Phosphoinositides and Receptor Mechanisms. New York, Alan R. Liss, pp. 149-162. 2. Rubin RP. Godfrey PP. Chapman DA. Putney JW. Jr (1984) Secretagogue-induced formation of inositol phosphates in rat exccrine pancreas. B&hem. J., 219, 655-659. 3. Dixon JF. Hokin LE. (1985) The formation of inositol 19cyclic phosphate on agonist stimulation of phosphoinositidc breakdown in mouse pancreatic lobules. J. Biol. Chem., 260, 16068-16071. 4. Strcb H. Hcslop JP. Irvine RF. Schultz I. Benidge MJ. (1985) Relationship between sccretagogue-induced Ca” release and inositol phosphate production in permeabilized pancreatic acinar cells. J. Biol. Chem., 260,7309-7315. 5. Banschbach MW. Geison I&. Hokin-Neaverson M. (1981) Effects of cholinergic stimulation on levels and fatty acid composition of diacylglycerols in mouse pancreas. B&him. Biophys. Acta, 663,34-45. 6. Pandol SJ. Schceffield MS. (1986) 1,2-Diacylglycetols, protein kinase C, and pancreatic enzyme secretion. J. Biol. Chem., 261,4438-4444. 7. Trimble ER. Bruzzone R. Biden TJ. Famse RV. (1986) Secretin induces rapid increases in inositol phosphate, cytosolic Ca” and diacylglycerol as well as cyclic AMP in rat pancreatic a&i. Biochem. J., 239,257-261. 8. Matozaki T. Williams JA. (1989) Multiple sources of 1,2-diacylglycerol in isolated rat pancreatic a&i stimulated by cholecystokinin. J. Biol. Chem., 264, 14729-14734. 9. Wooten MW. Wrcnn RW. (1984) Phorbol ester induces intracellular translocation of phospholipid/Ca2+-dependent protein kinase and stimulates amylase secretion in isolated pancreatic acini. FEB.8 Lett., 171, 183-186. 10. Noguchi M. Ada&i H. Gardner JD. Jensen RT. (1985) Calcium-activated, phospholipid dependent protein kinase in pancreatic acinar cells. Am. J. Physiol., 248, G692-G701. 11. Bumham DB. Munowitz P. Hootman SR. Williams JA. (1986) Regulation of protein phosphorylation in pancreatic acink distinct effects of Ca2+-ionophore A23 187 and 12-Gtetradccanoylphorbol 13-acetate. Biochem. J., 235, 125-131. 12. Men-in JE. Rubin RP. (1985) Pancreatic amylase secretion

CEJLCALCIUM and cytoplasmic flee calcium B&hem. J., 230.151-159. 13. Ochs DL. Korenbrot JI. Williams JA. (1985) Relation between free cytosolic calcium and amylase release by pancreatic acini. Am. J. Physiol., 249, G389-G398. 14. Bruzzone R. Regazzi R. Wollheim CB. (1988) Caemlein causes translccation of protein kinase C in rat acini without increasing cytosolic free Ca2+. Am. J. Physiol., 255, G33-G39. 15. Verme TB. Velarde RT. Cunningham RM. Hootman SR (1989) Effects of stautosporine on protein kinase C and amylase secretion from pancreatic a&i. Am. J. Physiol., 257, G548-G553. 16. Merritt JE. Bradford PG. Rubin RP. (1987) Penneabilized pancreatic acinar cells and neutrophils as models for studying the molecubu mechanism of action of secmtagogues. In: Poisner AM. Ttifaro JM (cds) In Vitro Methods for Studying Sect&on. Amsterdam, Elsevier, pp. 209-222. 17. Chaudhty A. Thompson RH. Rubin RP. Laychcck SG. (1988) Relationship between delta-9-tetmhydtocannabinolinduced arachidonic acid release and secretagogue-evoked phosphoinositide breakdown and Ca2’ mobilization of exocrine pancreas. Mol. Phannacol., 34,543-548. 18. Preiss J. Loomis CR. Bishop WR. Stein R Niedel JE. Bell RM. (1986) Quantitative measurement of sn-1,2diacylglycerol present in platelets, hepatocytes, and ras- and sis-transformed normal rat kidney cells. J. Biol. Chem., 261.8597-8600. 19. Rider LG. Dougherty RW. Niedel JE. (1988) Phorbol diesters and dioctanoylglycerols stimulate accumulation of both diacylglycetols and alkylacylglycerols in human neutrophils. J. Immunol., 140,200-207. 20. Wright TM. Rangan LA. Shin HS. Raben DM. (1988) Kinetic analysis of 1,2diacylglycerol mass levels in cultured fibroblasts. J. Biol. Chem., 263, 93749380. 21. Si%ing H-D. Fest W. Schmidt T. Essehnann H. Bachmann V. (1989) Signal transmission in exocrine cells is associated with rapid activity changes of acyltransferases and diacylglycerol kinase due to reversible protein phosphorylation. J. Biol. Chem., 264, 10643-10648. 22. Muallem S. (1989) Calcium transport pathways of pancreatic acinar cells. Annu. Rev. Physiol., 51, 83-105. 23. Knight DE. Koh E. (1984) Ca2’ and cyclic nucleotide dependence of amylase release from isolated rat pancreatic acinar cells rendered permeable by intense electric fields. Cell Calcium, 5, 401-418. 24. Kimura T. Imamura K. Eckhardt L. Schulz I. (1986) Ca2’-, phorbol ester-, and CAMP-stimulated enzyme secretion from permeabilized rat pancreatic acini. Am. J. Physiol., 250, G698-G708. 25. Sung CK. Hootman SR. Stuenkel EL. Kuroiwa C. Williams JA. (1988) Downregulation of protein kinase C in guinea pig pancreatic a&i, effects on secretion. Am. J. Physiol., 254, G242-G248. Please send reprint requests to : Dr RP. Rubin, Department of Pharmacology & Therapeutics, School of Medicine & Biomedical Sciences, State University of New York at Buffalo, Buffalo, NY 14214, USA Received : 19 June 1990 Revised : 6 July 1990 Accepted : 18 July 1990