Uncoupling of Phospholipase C from Receptor Regulation of [Ca2+]i in Tg4 Colonic Cells by Prolonged Exposure to Phorbol Dibutyrate*

The Tg4 colonic cell line, a cultured C1- secretory cell, elevates intracellular free Ca2+ ([Ca2+li) in a concentra- tion-dependent manner when exposed to carbachol or histamine. As determined with a fluorescence microscope imaging system, exposure of Tg4 cells to 100 I.LM carbachol or histamine resulted in an immediate [CaZ+li rise of approximately 50-80 nM in all cells.

The Tg4 colonic cell line, a cultured C1-secretory cell, elevates intracellular free Ca2+ ([Ca2+li) in a concentration-dependent manner when exposed to carbachol or histamine. As determined with a fluorescence microscope imaging system, exposure of Tg4 cells to 100 I.LM carbachol or histamine resulted in an immediate [CaZ+li rise of approximately 50-80 nM in all cells. Preincubation of monolayers for 1 h or longer with 0.4 I.LM phorbol 12,13-dibutyrate (PDB) reduced the number of cells which responded to histamine or carbachol and reduced the magnitude of the increase in the responding cells. This effect reached its maximum after 2 h and persisted for at least 24 h of PDB incubation. Binding of quinuclidinyl benzilate, a cholinergic receptor antagonist, indicated that down-regulation of external receptors was not an explanation for this effect. Examination of phospholipase C activity in TS4 cell membranes showed increased basal activity in PDBtreated compared with control cells. Measurement of inositol phosphates generated by intact cells using rny~- [~H]inositol incorporation or receptor binding assays showed that 2 h of incubation with PDB elevated basal levels of inositol 1,4,5-trisphosphate and prevented any further carbachol-induced generation of inositol trisphosphate. Probably as a consequence, both total cell calcium and Ca2+ ionophore-releasable calcium were decreased after 2 h of PDB incubation. Membrane-associated protein kinase C activity was elevated after a 2 h exposure to PDB but was below the level of detection after 24 h with PDB. Protein kinase C antagonists neither duplicated nor blocked the uncoupling of carbachol receptors induced by long term treatment with PDB. The results suggest that prolonged PDB incubation caused uncoupling and elevation of phospholipase C activity from cholinergic and histaminergic receptor regulation resulting in in-The T,, human colonic cancer cell line is a model epithelial C1-secretory cell. Monolayers respond to physiologic secretagogues (Beuerlein et al., 1987;Cartwright et al., 1985;Dharmsathaphorn et al., 1984Dharmsathaphorn et al., , 1985Dharmsathaphorn et al., , 1989Madara and Dharmsathaphorn, 1985;Reinlib et al., 1989), including carbachol and histamine, which act through elevation of intracellular free Ca2+ ([Ca'+],) ' (Dharmasathaphorn et al., 1989;Merritt and Rink, 1987;Wollheim and Biden, 1986), with an increase in short-circuit current and transcellular C1-secretion. In TR4 cells the carbachol-induced [Ca2+], rise is dependent on both calcium stores and extracellular Ca2+ (Reinlib et al., 1989). Histamine also shows a biphasic Ca2+ response although a larger part of the [Ca2+Ir response appears to be from an extracellular source (Dharmsathaphorn et al., 1989). A role in TS4 cell C1-secretion for protein kinase C, which can be stimulated by phorbol esters, has recently been suggested (Beuerlein et al., 1987;Vongkovit et al., 1989). Since the simultaneous addition of phorbol 12-myristate 13-acetate and Ca2+-dependent secretagogues depressed C1-secretion in Tx, cells without affecting the normal rise of [Ca"],, protein kinase C was postulated to inhibit agonist-induced secretion at a step distal to [Ca*+']; elevation (Vongkovit et al., 1989).
Prolonged incubation of fibroblasts with phorbol esters has been shown to down-regulate protein kinase C (Issandou and Rozengurt, 1989;Rodriguez-Pena and Rozengurt, 1984). Prolonged phorbol ester exposure has now been applied to T,, cells to investigate the relationships among [Ca2+],, phospholipase C, and protein kinase C. In this study the effects of PDB and the role of protein kinase C on [Ca"], in T,, cells were investigated at the single cell level using Fura-2 and a fluorescence microscope imaging system. In addition, measurements were made of phospholipase C activity and the levels of cytosolic inositol phosphates under the same conditions that [Ca2+Ii was monitored. Phospholipase C was activated by prolonged PDB incubation while protein kinase C decreased Cell Culture-T, cells, used to passage number 69, were cultured on glutaraldehyde-cross-linked rat tail collagen-coated glass coverslips as described (Dharmsathaphorn and Pandol, 1984;Reinlib et al., 1989). Monolayers were maintained in 1:l Dulbecco-Vogt modified Eagle's and Ham's F-12 media with 15 mM HEPES, pH 7.5, 1.2 g/ liter NaHCO:], 40 mg/liter penicillin, 8 mg/liter streptomycin and 5% newborn calf serum and studied between 4 and 14 days after confluence. The cells were a complete monolayer, as examined by thin histologic sections. In rare areas, cells appeared stratified (Reinlib et al., 1989).
Fura-2 Loading and Single Cell Analysis-T, cells were loaded with Fura-2, and [Ca"], was studied in single cells as described previously (Grynciewicz et al., 1985;Reinlib et al., 1989). Coverslips of cells were incubated at 20 "C for 60 min in 1 ml of culture medium containing 10 p~ Fura-2 AM followed by a 30-min incubation a t 37 "C. Before study, the monolayer was washed three times with 1 ml of Ringer's-HCO:j, supplemented with 10 mM HEPES, pH 7.40, and 10 mM glucose (Buffer A). The coverslip was mounted in a watertight Dvorak perfusion chamber, with a total volume of 0.25 ml, and was maintained a t 30 "C. Cells were visualized through a 63X Zeiss Planeofluor lens (numerical aperture 1.3), and emitted fluorescence (480-520 nm) was collected via a DAGE 66 SIT camera (DAGE, Indianapolis, IN). The excitation light was controlled by a MicroVax I1 computer (Digital Electronics Corp., Maynard, MA) which rotated bandpass filters of 350 and 380 nm in a motorized wheel in front of the 75-watt xenon source. The time required for exposure to a pair of excitation filters and data acquisition was approximately 1 s. Data were determined for 8-frame averaged images a t 10-s intervals although some experiments were performed using l-s, nonaveraged intervals. Data were acquired and stored as the average fluorescence intensity within 7.8-pm2 areas within each cell, and calculations were made by a MicroVax I1 computer with 100 megabytes of memory and IP-512 imaging boards (Imaging Technology, Inc., Woburn, MA). Approximately eight cells/field were measured simultaneously by these methods, using previously defined selection criteria (Reinlib et al., 1989). Calculations of [Ca"], were made following Grynciewicz et al. (1985). The K,, for Fura-2 was confirmed as 224 nM at 30 "C in the fluorescence microscope. The average R, (F1 350/F1 385) values for 1 p~ Fura-2 dye standards in the absence (RmJ or presence (Rmax) of 2 mM Ca2+ were 0.73 & 0.01 and 7.50 k 0.47, respectively ( n = 16).
[Ca'+], changes of less than 10 nM were considered indistinguishable from system noise and calculated as zero change. All values were corrected for background and cellular autofluorescence, measured separately. Cells containing punctate fluorescence were rare and not studied when found. Thus, as shown previously (Reinlib et al., 1989), the dye did not appear to accumulate subcellularly. As single cells were studied and the peak [Ca"], occurred a t variable times after secretagogue exposure, the mean peak [Ca"], in the data tables differed slightly from the peak responses illustrated in the figures.
For fluorometry of whole monolayers, T,, cells were seeded onto glass coverslips coated with rat tail collagen. The coverslips were glued to plastic supports with silicone rubber adhesive (General Electric Co., Waterford, NY) and studied 7-15 days after seeding. Cells were loaded with Fura-2 by incubation of the cells a t 23 "C for 60 min in "Na medium" containing (in mM) 130 NaCl, 5 KC1, 2 CaCL, 1 MgS04, 1 NaH2P0,, 25 glucose, 20 HEPES, pH 7.30, plus 5 p M Fura-2 AM followed by a 30-min incubation at 37 "C. After washing three times with Na medium, the cells on the coverslips were mounted a t 45 in a glass cuvette. Fluorescence was measured in an SLM spectrofluorometer (model S P F 500C, SLM, Urbana, IL) equipped with a cuvette stirrer and kept at 37 "C. Wavelengths were computer controlled a t excitations 340 k 1 nm and 380 & 1 nm, and emission was monitored a t 505 k 20 nm. Autofluorescence, determined from a filter seeded with cells but not loaded with dye, was subtracted automatically from the experimental reading. Ro was calculated at 3-s intervals. After each experiment Rmax was determined for each coverslip by the addition of 10 p~ ionomycin in the presence of 2 mM Ca2+, and Rmi, was determined by addition of 8.3 mM EGTA.
Calculation of [Ca"], was made using the method described above.
Treatment of Cells with Phorbol 12,13-Dibutyrate-TR4 monolayers were incubated in the presence or absence of 0.4 p~ PDB for 0.5-24 h a t 37 "C in 1 ml of culture medium (ethanol concentration = 0.02%). The monolayer was washed three times with 1 ml of Buffer A a t 20 "C prior to further use. When [Ca"]i measurements were performed, Fura-2 AM was included in the medium for the final 90 min of PDB incubation, 60 min of which was carried out at 20 "C.
Protein Kinase C Assay-After incubation of T,, monolayers in the absence or presence of 0.4 p~ PDB, protein kinase C was assayed in either whole cell homogenates or membrane and cytosol fractions using an activity assay based on phosphorylation of histone IIIS, as modified from the method of Cohen et al.' Each coverslip of cells was washed three times in 1 ml of Buffer A a t 20 "C and scraped into 0.5 ml of 5 mM Tris, pH 7.4, 5 mM EDTA, 0.01% leupeptin, 0.28 mM phenylmethylsulfonyl fluoride, 26 pg/ml aprotinin, 1 pg/ml phosphoramidon. The suspension of broken cells was homogenized five times with a Teflon pestle drill. When homogenate was used, the solution also contained 0.3% Triton X-100 and was centrifuged in a Beckman Microfuge for 3 min a t top speed to remove insoluble particulate. In other studies the homogenate included 2 mM dithiothreitol and 0.3% Triton X-100 and after 30 min was centrifuged a t 100,000 X g for 30 min in a Beckman Airfuge and separated into cytosol (supernatant) and membrane (particulate) fractions. A 0.45ml aliquot of each fraction was mixed for 10 min on ice with 0.5 ml of a slurry of packed DE52 (Whatman) equilibrated previously with 20 mM HCl and washed repeatedly in 20 mM Tris, pH 7.4, 0.5 mM EGTA, 5 mM dithiothreitol (Buffer B). The slurry was centrifuged for 4 min in a Microfuge a t approximately 12,000 X g and washed three times in Buffer B. Protein kinase C was released from the gel by a 10-min incubation in 0.4 ml of Buffer B plus 200 mM NaCl. DE52 was removed by centrifugation and the supernatant immediately assayed for phosphorylation of histone IIIS as follows. 10 pl of sample was incubated for 10 min a t 30 "C in a 100-pl volume containing 40 pg of histone IIIS, 2 mM Tris, p H 7.4, 10 mM magnesium acetate, 0.5 mM EGTA, 0.55 mM CaCI2 (51 p~ free Ca"), 5 p~ [y-=P]ATP (4.5 pCi) in the absence or presence of 1 pg/ml phosphatidylserine and 10 p M PDB. The assay was stopped by the addition of 1 ml of ice-cold 10% trichloroacetate and 0.2% NaPPi followed by 10 min of boiling. The sample was filtered through a presoaked GF/A filter (Whatman) and was washed twice with 5 ml of trichloroacetate-NaPPi. The filters were air dried overnight and assayed in a scintillation counter. Protein kinase C activity was considered as phosphatidylserine plus PDB-stimulatable activity above that occurring in the presence of Ca'+ alone and was expressed per mg of protein.
QNB Binding Assay-After exposure to control or PDB-containing solutions of 37 "C, particulate fractions of T,, cell homogenates were prepared for carbachol receptor binding, using the cholinergic receptor antagonist QNB, by scraping the washed monolayers into 5 mM HEPES, pH 7.5, 5 mM EDTA, 0.28 mM phenylmethylsulfonyl fluoride, 1 pg/ml phosphoramidon. The suspension was homogenized 10 times in a Teflon pestle drill homogenizer. The suspension was centrifuged a t 8,000 X g for 10 min and the pellet discarded. The membrane fraction was pelleted by centrifugation at 150,000 X g for 60 min and resuspended in 50 mM Tris, p H 7.4. Membranes (0.3-0.5 mg of protein/tube) were incubated in 50 mM Tris, pH 7.4, with 0.05-3 nM [3H]QNB, in a total volume of 0.5 ml, for 2 h at room temperature. Reactions were terminated by filtration through GF/B glass filters (Whatman) using a Brandel cell harvester (Brandel Inc., Gaithersburg, MD), washed three times with 5 ml of 50 mM Tris, pH 7.4, and radioactivity was determined in a liquid scintillation counter. Nonspecific binding was defined as binding in the presence of excess atropine (1 pM), and specific binding was determined by subtracting nonspecific from total binding. The KI, and the maximum density of binding sites (Bmax) for ['HIQNB binding were calculated by nonlinear least square regression analysis as described (Bevington, 1969).
Determination of Total Cell Calcium-To evaluate the effect of PDB on total cell calcium, plasma emission spectroscopy was carried out on Te4 monolayers. Monolayers were incubated for 2 h a t 37 "C in the presence or absence of 0.4 p M PDB and washed three times in Buffer A and then once in buffer A that contained no added Ca2+ M. E. Cohen, J. Wesolek, J. McCullen, S. Pandol, R. P. Rood, G. W. G. Sharp, and M. Donowitz, manuscript in preparation. (Muscholl, 1970). The cells were then scraped into 2 ml of ice-cold 0 . 3 mM LaCI:', 5 M HCl and drill homogenized 10 times. The suspension was centrifuged at 100,000 X g for 5 min, and the supernatant was frozen in liquid N, until the Ca'+ determination. In parallel, cells were treated identically except that 0.5 mg/ml ["HJpolyethylene glycol (10 pCi/ml; molecular weight 900) was included in the incubation solution for 1 h to assess the volume of, and correct for Cay+ in, the extracellular space. Data are expressed as mg of Ca"+/mg of cell protein.
Determination of Inositol Phosphate Levels and Phospholipase C Activity-Inositol phosphates were assayed in cultures incubated in 1.5 ml of inositol-free medium (Raben et al., 1987) supplemented with 5 pCi/ml myo-[:'H]inositol for 50-60 h a t 37 "C. In half of the cultures, 0.4 p~ PDB was included in the medium for the final 2 h of incubation. LiCl was also present in the wash and carbachol-containing solutions. The monolayers were washed once in Ringer's-HC03 containing 10 mM glucose, 10 mM Lic1, and 10 mM HEPES, pH 7.4. The solution was drawn off, and monolayers were incubated a t 30 "C in the same buffer in either the absence or presence of 100 p~ carbachol. After 30 min the solution was withdrawn and the reaction stopped by the addition of 2 ml of acidified methanol (MeOH/ concentrated HC1,50:0.3, v/v). InsP, Imp2, and InsP,' were separated and assayed as described (Downes and Mitchell, 1981;Raben et al., 1987) by anion-exchange column chromatography on Bio-Rad AG 1-X8 resin (formate form). The 0.1 M:0.2 M (formate/ammonium formate) eluate was defined as InsP1; 0.1 ~: 0 . 4 M as Imp,; 0.1 M:l M as InsP,,; and 0.1 ~: 1 . 5 M as InsP4. Two-ml fractions were collected, 10 ml of scintillation mixture added, and radioactivity determined in a liquid scintillation counter.
Total free Ins(1,4,5)Pa a t specified times after carbachol addition in nonradiolabeled T H 4 monolayers was also measured with a receptor binding assay. This assay was also used to measure phospholipase C activity in crude membranes prepared from Ta monolayers. TH, crude membranes were prepared by washing two T75 flasks of cells with 1 mM EDTA, 1 mM P-mercaptoethanol, 20 mM HEPES, pH 7.50,l pg/ ml phosphoramidone, 40 pg/ml phenylmethylsulfonyl fluoride. The cells were scraped into 10 ml of the same buffer, homogenized three times for 10 s in an Ultra-turrax tissue disrupter, and centrifuged for 45 min at 100,000 X g. The pellet was resuspended in a minimal volume of the homogenizing buffer and used for phospholipase C generation assays. For the assay, rat cerebellar microsomal membranes, which contain the specific Ins(1,4,5)P2 receptor (5), were prepared from adult, male Sprague-Dawley rats killed by decapitation. The cerebellum was homogenized with an Ultra-turrax tissue disrupter (Tekmar Co., Cincinnati, OH) at top setting, 10 s, in 30 volumes of ice-cold 50 mM Tris, pH 7.7, 1 mM EDTA, 1 mM P-mercaptoethano], pelleted by centrifugation at 20,000 x g, and resuspended in 30 volumes of the same buffer. The latter step was repeated three times, and the final pellet was resuspended in the same buffer a t 1.2 mg of protein/ml. TX4 monolayers were preincubated in the absence or presence of PDB, washed with 1 ml of Buffer A, and exposed to Buffer A plus 100 p~ carbachol a t 30 "C for specified time intervals. Then the solution was drawn off and the reaction stopped by addition of 0.5 ml of ice-cold 1 M trichloroacetate. In the case of T x~ crude membranes, the generation of was stopped by the addition of a n equal volume of 2 N trichloroacetate. The scraped cells or membrane pellets were homogenized three times and centrifuged a t 100,000 X g for 10 min, and the supernatants were extracted four times with water-saturated ether. The aqueous layers were used undiluted to quantify the amount of Ins(1,4,5)Pn present as follows. 100 p1 of sample was incubated in a total volume of 500 p1 of 50 mM Tris, pH 9.0, 1 mM EDTA, 1 mM P-mercaptoethanol with 1 nM ['HI Ins(1,4,5)P:, (5 pCi) and 60 pg of rat cerebellar membranes. Displacement of [:'H]In~(1,4,5)P:~ from the rat brain membrane was determined after 10 min by centrifuging the assay mixture at top speed in a Beckman Microfuge and determining the radioactivity in the pellet. Computation of cytosolic Ins(1,4,5)P3 was made by comparison with a standard curve after correcting for nonspecific binding not removed by 2 p M cold In~(l,4,5)P:~.

Effect of PDB Preincubation on the [Ca"], Response to
Carbachol and Histamine-Carbachol, a muscarinic secretagogue that operates through calcium-dependent pathways, was studied for its effects on [Ca"], in single TS4 cells. Single Tx, cells were monitored using a fluorescence microscope video camera system. After the addition of 100 p~ carbachol to control cells, [Ca2+Ii rose quickly to a peak and then declined slowly over the next few minutes (Fig. l A ) , as reported previously (Reinlib et al., 1989). All the T,, cells studied under control conditions in Ringer's-HC03, 10 mM glucose, 10 mM HEPES, pH 7.4, responded in this manner ( Table I). The effects of incubation with PDB on basal [Ca2+]i in TSq cells and on the elevation in [Ca"Ii caused by carbachol were studied at intervals up to 24 h (Fig. 1B). One min of phorbol ester exposure had no effect on basal [Ca'+Ji (data not shown) or on the changes caused by carbachol in terms of peak [Ca2+Ii rises, plateau level, and duration of [Ca2+]; elevation (Fig. lB), as earlier suggested (Beuerlein et al., 1987). In the absence or presence of PDB (0.4 p M ) for 1 min, 100% of the monitored cells responded to carbachol with a rise in [Ca"];. Prolonged PDB exposure decreased the percentage of cells which responded to carbachol and also decreased the magnitude of the response in those cells which did increase [Ca2+Ji (Table I and Fig. 1B). After     [Cay+]; after 24-h exposure to PDB was 90 f 20 nM.
Results similar to the carbachol experiments were found using 100 PM histamine, an agonist that also stimulates C1secretion through calcium-dependent mechanisms. In control cells, histamine elevated [Ca'+], immediately, with a peak increase of 47 nM within 30 s, with 95% of the cells responding ( Fig. 2 and Table I). As with carbachol, the [Ca2+Ii response after histamine was inhibited by preincubating with 0.4 FM PDB (Fig. 2). One h of PDB exposure significantly inhibited the peak increase in [Ca"], caused by histamine by 23%, with the inhibitory effect becoming maximum at 2 h of PDB exposure, at which time the peak A [Ca2+], was inhibited by 64% ( Fig. 2 and Table I). When pretreated with PDB for 2 h, only 19% of TX4 cells elevated [Ca"], in response to histamine (Table I), and with 24-h PDB exposure only 16% of cells responded (data not shown).
Effect of PDB on Cholinergic Receptors-As long term incubation with phorbol esters has been reported to downregulate cholinergic receptors, membrane fractions prepared from TR4 monolayers incubated in the absence or presence of PDB for 2 h were assayed for specific binding of the cholinergic receptor antagonist QNB. As shown in Fig. 3, the QNB binding characteristics in untreated and PDB-treated cells were similar. Equilibrium binding of ['HIQNB was found to be saturable and of high affinity. In one experiment it was calculated that control membranes had a Kr, of 1.2 nM and B,,,, of 18 fmol/mg, and PDB-treated membranes had a KD of 0.8 nM and B,,, of 17 fmol/mg. Results in a second preparation were similar, with control and experimental membranes both having a KI, of 0.4 nM, and B,,,, values were 12.5 Free pH] PNB (nM)

FIG. 3. Binding of [3H]QNB to T , , membranes was unchanged by pretreatment with PDB. TR1 cells, control (A) or
pretreated for 2 h with 0.4 M PDB (A), were washed three times in Buffer A (see "Experimental Procedures"), and the 100,000 X g pellet was assayed for equilibrium binding of ["'HIQNB. Binding was determined a t room temperature and was saturable and of high affinity (KT,, 1.2 and 0.8 nM for control and PDB-treated membranes, respectively). The results shown are representative of two experiments. and 13.0 fmol/mg, respectively. PDB Incubation Stimulates Membrane-associated Phospholipase C Activity in TR4 Cells-As carbachol receptors are functionally linked to phospholipase C in multiple cell types (Berridge, 1987;Berridge et al., 19871, the effect of prolonged PDB incubation on the activity of this enzyme was examined. Phospholipase C activity was first measured in membranes prepared from control and 2-h PDB-preincubated TX4 cells. The generation of In~(1,4,5)P:~ by the membranes was monitored by use of a recently developed competitive binding assay (Bredt et al., 1989). The membranes were suspended in a Ca2+-free medium (1 mM EGTA), and activity was considered as the appearance of Ins(1,4,5)P, when free Ca2+ was raised to 2 mM (Connolly et al., 1986). The medium was nominally Mg2+ free and contained P-glycerolphosphate to inhibit the action of residual 5'-phosphomonoesterase that may have been present. The basal level of Ins(1,4,5)P,, which accumulated during the membrane preparation in a low Caz+ solution, was 15-fold higher in cells incubated with PDB (Table 11,853 pmol/mg protein versus 12,980; p < 0.005). Compared with Ins(1,4,5)P3 accumulation in the absence of Cay+, elevating Ca2+ increased Ins(1,4,5)P3 accumulation approximately 2.5fold in control and 5-fold in PDB-treated membranes. As shown in Table 11, the PDB treatment increased membrane phospholipase C activity (Ca2+ stimulated activity) to 2.5 times the control value.
PDB Incubation Elevates Basal Ins(1,4,5)P:, Levels in Tx4 Cells-As 2-h PDB treatment appeared to stimulate phospho- Phospholipase C activity in TS, cell membranes is elevated by PDB incubation Phospholipase C activity was assayed by monitoring Ins(1,4,5)P3 generation in a crude membrane preparation. Immediately after preparation, the membranes (15-30 pg of protein) were incubated 30 min at 37 "C in 200 pl (final volume) containing 1 mM EGTA, 1 mM pmercaptoethanol, 1 mM Na,ATP, 20 mM HEPES, pH 7.50,20 mM 0glycerophosphate or in an identical solution except with the free Ca2+ raised to 2 mM. Ins(1,4,5)P2 was assayed by a competitive binding assay (Bredt et al., 1989). Phospholipase C activity was defined as the Ca'+-dependent amount of Ins(1,4,5)P3. The zero-time levels of Ins(1,4,5)P3 (generated during the 2-h preparation of membranes) were subtracted to obtain the values shown below and were as follows:  was determined in control (U) and 2-h PDB-pretreated (e) cultures by a competitive binding assay for the Ins(l,4,5)P3-specific receptor freshly prepared from rat cerebellum. Stars denote significant differences between PDB-treated monolayers and untreated controls; the plus signs show significant changes caused by carbachol. Data are mean & S.E. of three experiments. lipase C, it was of interest to determine whether the enzyme was still responsive to receptor activation. Ins(1,4,5)P3 levels were measured in Ts4 monolayers after incubation in the absence or presence of PDB (Fig. 4). Control cells had a significantly increased amount of Ins( 1,4,5)P3 10 s after carbachol addition. The Ins(1,4,5)P3 level peaked at 20 s, at which point it had doubled, and returned to base line by 30 s. When cells were first exposed for 2 h to PDB the resting level of Ins( 1,4,5)P3 was significantly higher than in the untreated cells, and in fact the basal value in this group was not significantly different from the carbachol-induced peak of the controls (83 & 8 uersus 104 f 14 pmol/monolayer, respectively; Fig. 4). Carbachol, however, had no further effect on Ins(1,4,5)P3 in PDB-treated cells. 30 s after carbachol exposure, Ins(1,4,5)P3 was again significantly greater in PDBtreated cells than in the corresponding control cells. A similar pattern of Ins(1,4,5)P3 elevation was found when Ts4 monolayers were preincubated for 24 h with PDB before carbachol exposure, with PDB-treated cells having a basal Ins(1,4,5)P3 level that was increased compared with untreated cells and no further elevation of Ins(1,4,5)P3 in response to carbachol (data not shown).
The effect of PDB treatment on carbachol-evoked changes in Ins(1,4,5)P3 was studied further using cells labeled with my~- [~H]inositol. T, monolayers were radiolabeled with myo- [3H]inositol, the inositol phosphates were extracted, and the fractions were separated by anion-exchange chromatography into InsP1, InsPg, InsP3, and InsPl (Table 111). Total incorporation of the radiolabel was not changed significantly by PDB incubation. The inclusion of 10 mM LiCl in all the aqueous solutions, which inhibits the further metabolism of inositol phosphates, was necessary to achieve radioactivity. As shown in Table 111, 100 p~ carbachol caused significant increases in untreated Tu cells in the levels of InsPg, InsP3, and InsP4 of 76,86, and 55%, respectively. After preincubation for 2 h with 0.4 p~ PDB, generation of InsPz, InsP3, and InsP4 was not elevated significantly by carbachol. Although increased 5'-phosphomonoesterase activity by PDB could explain the lack of increase in InsPz, InsP3, and InsP4, the level of InsPl would be expected to be comparatively high, which was not the case. These data indicate uncoupling of carbachol receptors from phospholipase C.

PDB Incubation Decreases Stored Calcium in T, Cells-
The assays of InsP:, used in this study measured total InsP3 (Fig. 4) or generation of ["H]InsP3 derived from incorporated my~- [~H]inositol. In these Ta4 cells, abnormally activated phospholipase C could potentially be compensated by raised activities of 5'-phosphomonoesterase or InsP3 kinase. The actions of these enzymes would tend to keep basal Ins( 1,4,5)P3 at close to normal levels. However, if the basal level of Ins(1,4,5)P3 was indeed elevated after PDB preincubation, an effect on cellular calcium stores would be expected. Therefore, the amount of stored calcium after PDB treatment was estimated. Two techniques were used plasma emission spectroscopy of whole monolayers, to measure total cell calcium; and Ca2+ ionophore-releasable calcium, measured in single cells using Fura-2, which reflects intracellular stored calcium. After incubation for 2 h in the absence or presence of PDB, monolayers were prepared for plasma emission spectroscopy. Parallel sets of cells were used to correct for Ca2+ present in the extracellular spaces by use of [3H]polyethylene glycol. PDB exposure did not significantly alter the extracellular space (32 and 38 pl/mg protein for untreated and PDBincubated monolayers, respectively). In two experiments, after PDB treatment total cell Ca2+ decreased 28% (control values were 4.0 and 3.8 mg of Ca2+/mg of protein, and PDB-treated cells contained 2.9 and 2.7 mg of Ca2+/mg of protein).
Intracellular calcium stores were also evaluated by studies of [Ca2+Ii in T, cells at the single cell level using ionomycin to permeabilize the endoplasmic reticulum and CCCP to release mitochondrial calcium. After control or PDB incubation for 2 h, cells were perfused with Ca2+-free Ringer's-HCOa (see "Experimental Procedures") for 5 min and then treated with 10 p~ ionomycin plus 10 NM CCCP. Peak changes in [Caz+li were taken as an indication of the amount of calcium available in the internal stores. As shown in Fig. 5, the total stored Ca2+ was significantly reduced by preexposure to PDB, with controls able to elevate [Ca2+]i 548 f 84 nM above the basal level whereas PDB-treated cells raised [Ca2+], 301 & 54 nM above basal ( p < 0.05). The cells were perfused at the end of the study with Ringer's-HC03 solution containing 10 pM ionomycin, 10 PM CCCP, 4 mM ca2+ to ensure that under all conditions the intracellular dye responded in a manner similar to free dye. The latter treatment caused elevation of [Ca2+], in both control and PDB-treated cells to similar values. In summary, prolonged PDB treatment decreased total calcium

TABLE 111
Generation of inositol phosphate with carbachol is inhibited by 0.4 p~ PDB preincubation T,, monolayers were radiolabeled with rny~-[~H]inositol for 50-60 h and either left untreated or exposed to 0.4 PM PDB for the final 2 h of incubation at 37 "C. The formation of inositol phosphates after the addition of 100 PM carbachol in Ringer's-HC03, 10 mM glucose, 10 mM HEPES, pH 7.40, plus 10 mM LiCl a t 30 "C was compared after 30 min with controls that received no carbachol. Inositol phosphates were determined using AG 1-X8 anionexchange chromatography as under "Experimental Procedures." Values are mean f S.E. from three cultures and represent the percent change relative to paired controls. Student's paired t tests compare the following: p* inositol phosphate levels after carbachol with basal levels; p+ PDB-treated with untreated monolayers. Basal values (cpm/ monolayer) for control and PDB-incubated monolayers, respectively, were: InsP1, 11,451 f 2, 763   cells. Intracellularly stored Ca2+ was determined in control and 2-h PDB-treated cells by single cell analysis using Fura-2. Cells were perfused for 5 min with Ca"-free Ringer's-HCOa, 10 mM glucose, 10 mM HEPES, pH 7.40, which had been treated with 10 g/liter Chelex 100 overnight to lower free Ca". [Ca"Ii increase in each cell was quantitated when 10 p~ ionomycin plus 10 p~ CCCP were added to the perfusate. Cells in which the latter treatment saturated the intracellular Fura-2 calcium signal were assigned a value for [Ca'+]i of 1 pM. All cells elevated R, maximally in the presence of 4 mM CaCL, performed at the end of each experiment. Basal [Ca"+li: control, 128 t 8 nM ( n = 10); PDB treated, 108 t 14 nM ( n = 10).
stores, consistent with the effective level of Ins(1,4,5)P3 in the intact cells being elevated.

Effect of Prolonged P D B Incubation on Tx4
Cell Protein Kinase C-PDB is known to activate protein kinase C. To determine whether the demonstrated uncoupling of phospholipase C from its receptor(s) was through protein kinase C, the kinase inhibitors staurosporine and H-7 were used, as well as the "nonactivating" phorbol ester, a-PDB, as a control. These agents were first applied to control and PDB-incubated T n 4 monolayers, and [Ca2+Ii was measured in a fluorometer after carbachol addition. In a population of control TB4 cells grown on glass coverslips, carbachol elevated [Ca2+], by 171 +-10 nM (Table IVA) a-PDB, did not significantly reduce the response of [Ca'+Ii to 100 p~ carbachol (Table IVA).
That the inhibitory effect of 2-h PDB exposure on the carbachol-induced elevation of [Ca2+Ii was not exerted directly by protein kinase C was suggested by studies with the protein kinase C inhibitors H-7 and staurosporine. Neither H-7 nor staurosporine altered basal [Ca2+Ii (data not shown), nor did they inhibit the rise in [Ca2++li following exposure to 100 PM carbachol (Table IV, A and B). The carbachol-induced rise in [Ca2+], after incubation with H-7 (60 or 100 p~) for either 15 min or 2 h before carbachol addition was not significantly Phospholipase C Uncoupling in TS4 Cells different from the control carbachol response (Table IV, A and B). Similar results were obtained when the cells were incubated for 15 min or 2 h with staurosporine (1 p~) ; again the carbachol stimulation of [Ca2+Ii was no different from the control response (Table IVA). Not only did H-7 and staurosporine not duplicate the effect of 2-h treatment with PDB, but a 2-h exposure to H-7 did not significantly reverse the effect of 2-h PDB exposure to inhibit the carbachol-induced elevation of [Ca2+Ii (Table IVB). As shown in Table IVB, PDB continued to inhibit the carbachol-induced [Ca2+]i increase in the presence of the protein kinase C blocker H-7. Although a marginal increase in [Ca2+Ii was observed under these conditions, only 31% of the cells responded. Thus the [Ca2+Ii elevation that occurred with carbachol in cells preincubated with PDB plus H-7 was significantly less than in control cells.
The effects of prolonged PDB exposure were studied to determine whether any correlation could be found between protein kinase C activity and [Ca"], response to carbachol. Specific activity of protein kinase C in Ts4 cell homogenate, cytosol, and a crude membrane preparation was determined after 2-and 24-h of PDB exposure and compared with untreated control cells. Protein kinase C activity in cell homogenates decreased in a time-dependent manner after PDB addition to the medium ( Fig. 6 and Table V). Homogenate activity of Ts4 monolayers fell measurably after 30 min of exposure to PDB and continued to a low of approximately 18% of control after 24 h. Membrane and cytosolic fractions, prepared from Ta monolayers after incubation with or without PDB, were assayed for protein kinase C activity. Table V shows that after 2 h with PDB, membrane-associated protein kinase C rose by 175%, constituting 44% of the total cellular activity, as compared with 9% of the total activity found in the membrane of control cells. Cytosolic protein kinase C activity was decreased by 2 h of PDB treatment and was nearly absent after 24 h with PDB (5% of control). After 24 h of PDB incubation the membrane-associated level was undetectable (Table V). Therefore, total protein kinase C activity decreased over a 24-h incubation period while membrane-associated activity increased at 2 h and was below detection after 24-h exposure to PDB. No correlation was found between membrane-associatedprotein kinase C activity and the inhibition of receptor-stimulated increases in [Caz+],.
-._ c a -/-. FIG. 6. Total cellular protein kinase C activity decreases in a time-dependent manner upon PDB exposure. Following the specified times of preincubation with 0.4 p~ PDB, 0.5 ml of 1 M trichloroacetate was added to each monolayer, the cells were scraped and homogenized, and total cell protein kinase C activity was assayed as under "Experimental Procedures." Protein kinase C activity was defined as the rate of transfer of "'Pi from ATP to histone 111s stimulated by phosphatidylserine plus PDB corrected for the activity in the presence of Ca", alone. Data shown are normalized to simultaneously studied control cells, unexposed to PDB. Control activity was 0.2 pmol/mg protein-min. Results are the means of two to four cultures.

DISCUSSION
Carbachol and histamine, which have been shown to stimulate C1-secretion in colonic epithelia and Ts4 cells, act at least partially through mobilization of intracellular Ca2+. As shown previously for Ts4 cells, the peak increase in [Ca2+Ii in response to carbachol was highly dependent upon stored calcium and could be eliminated by the addition of dantrolene, which immobilizes stored calcium, or by prolonged preincubation of cells with 100 pM EGTA to deplete the cellular calcium stores (Renlib et al., 1989). The latter phase of [Ca2+], elevation required the presence of extracellular Ca2+ and could be inhibited by the Ca2+ channel blockers, verapamil or nifedipine (Renlib et al., 1989). These studies further probe the mechanism of the carbachol-induced rise in [Ca2+Ii.
Recent investigations suggest a role for protein kinase C (Beuerlein et al., 1987;Dharmsathaphorn et al., 1989;Weymer et al., 1985) in C1-secretion in Ts4 cells. Specifically, exposure to phorbol 12-myristate 13-acetate for 1-2 min before stimulating T,, monolayers with carbachol blocked C1-secretion without inhibiting the rise in [Ca2+Ii. However, incubations of greater than 15 min with phorbol 12-myristate 13-acetate also inhibited [Ca"]i rises stimulated by carbachol addition. Previous work did not go beyond measurements of C1-fluxes and [Ca2+Ii elevation of whole monolayers. The present work was initially intended to investigate receptor-mediated [Ca2+Ii increases under conditions that either activated (2-h exposure to PDB; see Table V) or decreased membrane levels of protein kinase C (24-h exposure to PDB). However, after both 2-and 24-h exposure to PDB there was increased basal phospholipase C activity compared with controls, based either on measurement of membrane bound phospholipase C activity or of total Ins(1,4,5)P3 (Tables I1 and I11 and Fig. 4).
Overall, the results demonstrate that prolonged PDB incubation induces an uncoupling of phospholipase C from receptor regulation and results in prolonged activation of phospholipase C. The mechanism by which uncoupling from the receptor activates the phospholipase C is unknown, although possibilities include an effect on a guanine nucleotide binding protein. Nishizuka (1988) suggested a negative feedback role for protein kinase C in signal transduction in which the C kinase lowers Imp3 levels, either by blocking receptormediated hydrolysis of inositol phosphates or increasing hydrolysis of InsP3. An example of the latter has been demonstrated in human platelets in which protein kinase C action regulates inositol trisphosphate 5'-phosphomonoesterate ac-tivity (Connolly et al., 1986). These negative feedback aspects of protein kinase C action occur rapidly and are distinct from the more prolonged effect of phorbol ester exposure demonstrated here. The uncoupling of phospholipase C resulting in prolonged phospholipase C activation and inhibition of further stimulation of phospholipase C via plasma membrane receptors occurs distal to the plasma membrane receptor and initially does not produce a negative feedback effect on the entire system. In fact, the consequence of such a mechanism would be predicted to be an initial, short lived activation of InsP:{ generation, which would persist until the intracellular calcium stores were emptied, followed by inhibition of signal transduction. Thus, this would ultimately complement the above described negative feedback on the system by protein kinase C. However, in interpreting the physiologic significance of these results it must be remembered that a phorbol ester, not a physiologic stimulus, was being studied.
The results of studying generation of inositol phosphates using my~-[:~H]inositol incorporation were consistent with results of the competitive binding assay, showing that carbachol could activate inositol phosphate turnover in cells but not in PDB-incubated cells ( Fig. 4 and Table 111). Basal levels of inositol phosphates are not easily determined using myo-[:'H] inositol, and high pressure liquid chromatography was not performed. The fact that basal levels of [3H]InsP3 were not changed by PDB pretreatment whereas basal total Ins(1,4,5)P3, measured by a competitive binding assay, was increased suggests that separate inositol pools were involved in generation of carbachol-stimulated [3H]In~Pa (Table 111) and total Ins(1,4,5)P3. Table I11 also shows no significant difference in PDB-treated uersus control cells in the amounts of InsP, and InsPz produced following carbachol, which suggests that the 5'-phosphomonoesterase was similarly active under both conditions. Additionally, in the studies, PDB did not increase basal or carbachol-induced InsP, levels, indicating that kinase(s) was not activated. Finally, it was of interest to determine whether the effects of prolonged PDB incubation described here involved protein kinase C. Long term PDB exposure lowers protein kinase C in many systems (Rodriguez-Pena and Rozengurt, 1984). Assays of protein kinase C indicated that the enzyme was gradually decreased in both cytosol and whole cell homogenate with prolonged PDB exposure, but the membrane fraction had elevated protein kinase C at 2 h and undetectable levels at 24 h after PDB. In contrast, the membrane fraction at both 2 and 24 h had highly activated phospholipase C (Table I and Fig. 4). Thus, phospholipase C activation after PDB incubation did not correlate with membrane-bound activity of protein kinase C. In addition, as further indication that the results described were not a result of a direct effect of protein kinase C is that antagonists of protein kinase C, H-7 and staurosporine, neither prevented nor duplicated the uncoupling of carbachol receptors induced by long term treatment with PDB (Table IV). The possibility cannot be eliminated, however, that the effects demonstrated here could be caused by a nuclear effect of protein kinase C or a protein kinase C isoform that is not detected by the histone 1 1 1 s assay and/or is insensitive to the protein kinase C antagonists used in these studies. Thus our results suggest that these effects of prolonged PDB exposure are ultimately carried out by some mediator, which is initially altered by PDB, but that protein kinase C is not the direct cause of the prolongedphospholipase C activation shown here.