Phorbol Esters Inhibit al-Adrenergic Effects and Decrease the Affinity ...

VOl. 261, No. 2, Issue of January 15, pp. 520-526.1986 Printed in U.S.A.

THE.JOURNAL OF BIOLOGICAL CHEMISTRY (c> 1986 by The American Society of Biological Chemists, Inc

Phorbol Esters Inhibit al-Adrenergic Effects and Decrease the Affinity of Liver Cell al-Adrenergic Receptors for (-)-Epinephrine* Silvia Corvera$$,Kurt R. Schwarzll, Robert M. GrahamllII and J. Adolfo Garcia-S%nz$$$ From the $Departamento de Bioenergetica, Instituto de Investigaciones en Fisiologia Celular, Uniuersidad Nacionnl Autonoma de Mexico, 04510 Mexico, D. F. and the lCellular and Molecular Research Laboratory, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts 021 14

4&Phorboll2-myristate 13-acetate(PMA) modified the metabolic actions of three calcium-dependent hormones in different ways.The stimulations of glycogenolysis ureogenesis and phosphatidylinositol labeling produced by al-adrenergic agonist was blocked by the phorbol ester. In contrast, P M A slightly increased the stimulation of ureogenesis produced by low concentration of angiotensin I1 without modifying the maximal response. No effect of P M A was observed on the stimulation of ureogenesis induced by vasopressin. The stimulation of phosphatidylinositol labeling inducedby vasopressin was decreased by PMA, whereas that induced byangiotensin I1 was not affected. In intact freshly isolated hepatocytes, [‘Hlprazosin binds with high affinity to a site which displays the characteristics of a,-adrenergic receptor. Competitive inhibition studies with (-)-epinephrine reveal two different sites for this agonist: a high affinity site (& 9 nM) and a low affinity site (& 2 WM). In the presence of phorbol esters, (-)-epinephrine binding data now show the presence of a single classof low affinity sites, with similar affinity to those present in control cells. Thus, the inhibition of hepatocyte al-adrenergic action by PMA may be related to the loss of high affinity binding sites caused by the tumor promoter.

Phorbol esters are a group of tetracyclic &terpenes whose effects on tumor promotion and cell differentiation have been widely studied (reviewed in Ref. 1).These compounds bind to and directly activate calcium phospholipid-dependent protein kinase C (2, 3). This enzyme can also be activated by physiological concentrations of diglycerides. Thus, protein kinase C may have an important role in mediating the actions of hormones that increase the concentration of diglyceride within the cell. Amongsuch hormones are thosewhich rapidly stimulate the hydrolysis of membrane phosphoinositides and increase the level of cytosolic free calcium, such as vasopres*This study was supported in part by grants from the Consejo Nacional de Ciencia y Tecnologia (PCCBBNA 020747),the National Institutes of Health (NS-19583), the American Heart Association (83-1242; and in part with funds from the Massachusetts Affiliate) and the R. J . Reynolds Company. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solelyto indicate this fact. § Present address: Dept. of Biochemistry, University of Massachusetts Medical School, Worcester, MA 01605. 11 Established Investigator of the American Heart Association (AHA 82-240). $$ T o whom all correspondence should be addressed Instituto de Investigaciones en Fisiologia Celular, Universidad Nacional Aut6noma de Mexico, Apartado Postal 70-600,04510 Mbxico D. F.

sin, angiotensin 11, and a,-adrenergic amines (4-7). In fact, it has been observed that phorbol esters can partiallymimic the effects of these hormones on the phosphorylation of cytosolic proteins (6), and on the activity of liver glycogen synthase (8).An unexpected effect of phorbol esters is that they potently inhibit the stimulation of glycogenolysis produced by a,-adrenergic agents in rat liver cells (9). This inhibition seems specific for al-adrenergic actionsbecause the effects of other glycogenolytic hormones such as vasopressin, angiotensin 11, and glucagon, or the calcium ionophore A-23187, are not similarly affected by the tumor promoters (9). In this study, a detailed characterization of the effects of phorbol esters on the actions of al-adrenergic agents, aswell as on the actions of vasopressin and angiotensin 11, was undertaken in order to gain further insights into the mechanism by which these tumor promoters perturb al-adrenergic actions. Recently it has been shown that phorbol esters can potently inhibit the binding of insulin (10) and epidermal growth factor (EGF’) (11)by decreasing the affinity of their respective receptors. We, therefore, investigated whether a similar change in the affinity of the al-adrenergic receptor for (-)-epinephrine might be involved in the effects of the tumor promoters. The data presented in this paper indicate that phorbol esters specifically inhibit al-adrenergic actions and that this inhibition is associated and probably due to a decrease in the receptor affinity for (-)-epinephrine. MATERIALSANDMETHODS

(-)-Epinephrine, (-)-propranolol, paragyline, arginine vasopressin, anglotensin 11, and phorbol esters, were obtained from Sigma. [”PIPi (carrier free) and [3H]prazosin (17.1 and 80.9 Ci/mmol) were from New England Nuclear. Prazosin was a gift from Pfizer. Collagenase and bovine serum albumin (fraction V) were obtained from Worthington and Reheis, respectively. Other substances were analytical grade. Hepatocyte Isolation and Metabolic Studies-Adult female Wistar rats (-200 g) fed ad libitum with Purina rat chow were used. Hepatocytes were isolated by the method of Berry and Friend (12), as modified by Tolbert et al. (13). Isolation, washing, and incubation of the cells were done in Krebs-Ringer bicarbonate buffer, saturated with Oz/C02 (95%:5%), pH 7.4, a t 37 “C. To determine the rates of ureogenesis, cells (=30 mg, wet weight) were incubated for 60 min in 1 ml of Krebs-Ringer bicarbonate buffer supplemented with 10 mM glutamine, 2 mM ornithine, and 1%albumin (14). It has been shown that the rate of urea synthesis is linear under these conditions. The rate of glycogenolysis wasdetermined as previously described (9). Phorbol esters were dissolved in dimethyl formamide as M stock solutions, and further diluted in water. The final concentration of the solvent in the cell suspensions did not exceed 0.001% and was added at an equal concentration in control tubes. In all metabolic The abbreviations used are: EGF, epidermal growth factor; PMA, 4@-phorbol12-myristate 13-acetate; PI, phosphatidylinositol; Hepes, 4-(2-hydroxyethyl)-l-piperazineethanesulfonic acid.

520

al-Adrenergic Phorbol Actions Inhibit Esters studies, IO-' M propranolol was present with (-)-epinephrine in order to block any possible &adrenergic component in the actions of the hormone. Incubations were ended by cooling the cell suspensions in iced water. After centrifugation for 5 min, urea was determined in the cell supernatant (15). Phospholipid Labeling-For the study of phospholipid labeling, cells were incubated in Krebs-Ringer buffer supplemented with 10 pCi/ml [32P]Pifor 60 min in the presence or absence of the agents tested or vehicle. Incubations were ended by the addition of chloroform/methanol(2:1, v/v).The lipids extracted were separated by onedimensional thin-layer chromatography as previously described (16). Incorporated radioactivity was counted in silica gel scrapping of each phospholipid. p HIPrazosin Binding Studies-Binding assays were performed as described by Hoffman et al. (17). Nonspecific binding, defined as the amount of [3H]prazosinbound in the presence of 10 p~ phentolamine, was less than 10% of the total binding a t the Kd of the radioligand. Protein was determined according to Lowry et al. (18) using bovine serum albumin as standard. For the study of [3H]prazosin binding to intact cells, hepatocytes were isolated and washed as described above. Hepatocytes (=lo6cells) were incubated in 1-ml final volumes of a medium consisting of Krebs-Ringer bicarbonate buffer supplemented with 20 mM Hepes, 2 mg/ml albumin, and 100 p~ pargyline, saturated with 02/C02 (95%:5%),pH 7.4, at 37°C. Cells were added in 100-pl volumes to initiate the assay; incubations were continued for the times indicated in kinetic studies, and for 30 min in saturation and competition experiments. The agentstested or vehicles were added in 20-4 volumes and were present from the beginning of the incubations. The concentrations of [3H]prazosin used were 0.025-2.0 nM for saturation experiments and 0.25 nM for competition experiments, except for competition with prazosin and yohimbine in which 1 nM [3H]prazosin was employed. Separation of unbound ligand was conducted by two different procedures. In one, 9 ml of ice-cold binding buffer was added to each tube. The tubes were then immediately centrifuged for 8 s a t top speed in a clinical centrifuge (Sorvall R T 6000). The supernatantwas then decanted, another 9 ml of buffer added, and the centrifugation repeated. Two ml of scintillation fluid was then added to theresulting cell pellet, and the contentsof each tube were transferred to scintillation vials for counting. In the second procedure, incubations were terminated by filtration on a Millipore apparatus and test tubes washed with 4 X 4 ml of binding buffer (25°C). With both methods, similar results were obtained. Data Analysis of Radioligand Binding-The untransformed binding data were fitted by a weighted nonlinear iterative computer-assisted curve-fitting program (LIGAND). For saturation studies, nonspecific M binding was also evaluated as binding in the presence of phentolamine. Both methods of determining nonspecific binding yielded virtually identical results. The underlying model of ligandreceptor interaction (one site, two or more sites) was determined with the F-test. Experiments were first evaluated individually and then pooledin groups and calculated simultaneously in one calculation step using correction factors, as described by Munson and Rodbard (19). All data are reported as the means -t S.E. of duplicate incuhations from at least three different cell preparations. Statistical comparisons were performed using the unpaired Student's t test; p < 0.05 was considered as significant. RESULTS

Effects of PMA on the Metabolic Actions of (-)-Epinephrine, Vasopressin, and Angiotensin 11-As seen in Fig. 1, (-)epinephrine, vasopressin, and angiotensin I1 produced concentration-dependentstimulation of urea synthesis in isolated rat liver cells (Fig. 1, panel A). PMA inhibited the stimulation produced by maximally effective concentrations of (-)-epinephrine in a dose-dependent fashion,but did not modify the effects of maximal concentrations of the vasopressor peptides (panel B ) . The action of other phorbol esters on the effect of a maximal concentration of (-)-epinephrine are presented in Fig. 2. It can be seen that the biologically active phorbol derivatives PMA and 4B-phorbol 12,13-diacetate were able to inhibit the

521

100

50

0

-Log [HORMONE] M

-Log [PMA]

M

FIG. 1. Selective inhibition of al-adrenergic stimulation of ureogenesis by PMA. Isolated hepatocytes were incubated for 60 min as described under "Materials and Methods," in the presence of increasing concentrations of vasopressin (0-0), angiotensin I1 (U or ) (-)-epinephrine , (M (panel ) A), or in the presence of10" M vasopressin, lo-' M angiotensin 11, M (-)-epinephrine, and increasing concentrations of PMA (panel B). Basal urea production was of 32 2 5 nmol/mg of cell, wet weight.Data were normalized to theper cent of maximal effect for each hormone, which was 126f 3, 128 -C 3, and 152 f 3% of basal urea production (mean f S.E.) for vasopressin, angiotensin I1 and, (-)-epinephrine, respectively. Data are the means -t S.E. of three experiments performed in duplicate. All points above 25% stimulation are significantly (p < 0.001) different from control values by non-paired Student's analysis. The inhibition by lo-' and lo" M PMA of the effect of epinephrine is also significant (p < 0.05 and p < 0.001, respectively).

effect of (-)-epinephrine, whereas the biologically inactive compound, 4p-phorbo1, was not. The relative potencies with which these derivatives inhibit the effect of (-)-epinephrine agree closely with their relative potencies as tumor promoters (20), and with their relative affinities for the phorbol ester receptor (3, 21). Fig. 2 also shows that the effect of (-)epinephrine can be totally blocked by prazosin, which demonstrates that isitproduced specificallythrough al-adrenergic receptor activation (14). Fig. 3 presents the action of lo-? M PMA on the effect of epinephrine on two metabolic parameters: ureogenesis (panel A ) and glycogenolysis (panel B). It can be observed that the phorbol ester practically abolished the action of epinephrine on these parameters. Concentration-response curves for vasopressin and angiotensin I1 in the absence or presence of M PMA are presented in Fig. 4. It canbe seen in panel A that thephorbol esterenhances the effects oflow doses of angiotensin 11, whereas the effect of low doses of vasopressin are not significantly altered (panel B). Maximal responses to these hormones were not affected by PMA. Effect of PMA on Phospholipid Labeling-Table I shows the effects of phorbol esters on the labeling of different phospholipids. The labeling of phosphatidylcholine was significantly increased ( p < 0.02) by PMA. On the contrary the labeling of phosphatidylinositol (PI) was significantly (p < 0.025)

522

Phorbol Esters InhibitLu,-Adrenergic Actions f l 16C

A

c

J

6 (0 4.

14C

m

LL

0

$

v

4. 120

W E

3

100

1 4

.

0 7 6 5 4

-LOG

FIG. 2. Inhibition of u,-adrenergic stimulation of ureogenesis by different phorbol esters. Hepatocytes were incubated M (-)-epinephrine and different concentrations of 40with 4p-phorbol 12,13-diacetate ( U - U ) , PMA phorbol (A-A), ( O - - O ) , or prazosin (A-A). Other experimental details are as for Fig. 1. All agents except 40-phorbol significantly and dose-dependently inhibited epinephrine-stimulated urea synthesis.

decreased by the phorbol ester. At M PMA,aconcentration at which the inhibition of cyl-adrenergic actions was maximal, the basal incorporation of [3'P]Pi into PI was decreased by ~ 2 8 % . The effect of PMA on the stimulation of PI labeling by vasopressin, angiotensin 11, and (-)-epinephrine is presented in Figs. 5 and 6. Although the basal incorporation of [32P]Pi into PI was decreased by PMA (Table I), thestimulation of PI labeling by angiotensin 11, taken as the per cent increase over the corresponding basal value, was not altered by this agent (Fig. 5, panel A ) . Interestingly, the stimulation of PI labeling by vasopressin was significantly inhibited by P M A i.e. the response to maximal and submaximal concentrations of the peptide was decreased by PMA. The ability of (-)epinephrine to stimulate PI labeling was blocked by PMA (Fig. 6). Fig. 7 shows the effect of different concentrations of PMA on the stimulation of PI labeling by maximally effective concentrations of vasopressin, angiotensin I1 and (-)-epinephrine. The effect of the hormones was calculated as the per cent increase over the basal incorporation observed with each of the concentrations of PMA employed (Table I). This eliminates the effect of PMA on the basal incorporation and limits the results to its effect on hormonally stimulated PI labeling. After this, theresults were normalized by taking the effect of the hormones in the absence of PMA as 100%.It can be seen that PMA did not alter theeffect of angiotensin 11on PI labeling, and decreased the effect of vasopressin to about 50%. In contrast, the effect of (-)-epinephrine was totally abolished by PMA, at concentrations similarto those required to block cyl-adrenergic-mediatedmetabolic actions (Fig. 1). Effect of PMA o n Radioligand Binding to 0,-Adrenergic Receptors i n LiverMembranes-The results shown in the

[EPINEPHRINE]

M

FIG. 3. Effects of PMA on the stimulations of ureogenesis (panel A ) and glycogenolysis (panel B ) by (-)-epinephrine. Hepatocytes were incubated in the presence of different concentrations of (-)-epinephrine alone ( O " - O ) or in the presence of M PMA (u). PMA alone had no effect on the production of glucose or urea. Data are presented as the percentage of basal productions which were 32 f 5 nmol/mg of cell, wet weight, for urea and 47 -t 5 nmol/mg of cell, wet weight, for glucose (mean & S.E. of four experiments performed in duplicate).

B

4

T

T 120

IIO

I

100

-Log[ANGIOTENSIN

lI]M

-LOO

LVASOPRESSINJ M

FIG. 4. Effects of PMA on the stimulation of ureogenesis by vasopressin and angiotensin 11. Hepatocytes were incubated in the absence (o"-o) or presence (+-e) of lo" M PMA and different concent,rations of angiotensin I1 (panel A ) or vasopressin @anel B). The stimulations by lo-" and M angiotensin II in the presence of PMA are significantly different than those in the absence of this agent (p < 0.05). Other differences between cells treated or not treated with PMA are not statistically significant.

Phorbol Esters Inhibit al-Adrenergic Actions

523

TABLEI Effects of PMA on the labeling of phospholipids by [32PlPi Hepatocytes were incubated as described under "Materials and Methods" in the presence of increasing concentrations of PMA. Results areexpressed as thepercentage of basal ["PIPi incorporation into each phospholipid, which is shown in the first column. Data are the means f S.E. of three experiments performed in duplicate. PE, phosphatidylethanolamine; PC, phosphatidylcholine; PA, phosphatidic acid. % of basal at [PMA] of:

Basal

Phospholipids

lo-* M

IO-~M

cpmlw

PE PC PI PA

4 7 0 f 38 1 O O f 5 102 f 6 1 9 9 f 4 51 1 0 f 51 7 5 f 2 2 129 f 15 1 0 2 f 5 89 f 6 117+ 13 1 1 4 f 7 1 1 4 f 9

10" M

10-6~

+

93 5 105 f 7 159 f 16 195+ 16" 5 9 f 6' 72 f 5 1 2 6 f 15 126f 7

J

Q:

CD a m LL

0

s

v

0 Z H -1 W

m

" p < 0.02. ' p < 0.025.

ANGIOTENSIN

h

6

-1

IK

VASOPRESSIN

H

a

I.

6 -LOG

[EPINEPHRINE]

M

FIG. 6. Effects of PMA on the stimulation of PI labeling by (-)-epinephrine. Hepatocytes were incubated in the absence ( G " 0 ) or presence (U of) M PMA and increasing concentrations of (-)-epinephrine. Results are expressed as the per cent of basal incorporation of["PIPi into PI, which was 115 f 10 and 88 f 10 cpm/mg of cell, wet weight, in the absence or presence of PMA, respectively. Plotted are the means f S.E. of three experiments performed in duplicate.

stimulation of P I labeling by epinephrine (Fig. 1). In order to determine whether the action of PMA could decrease the affinity of the al-adrenergic receptor for (-)epinephrine, the competition for [3H]prazosin binding sites by increasing concentrations of the hormone was studied. PMA did not decrease the affinity of the receptor for (-)-Log [AGONIST] M epinephrine (results not shown). In fact, in some experiments FIG. 5. Effects of PMA on the stimulation of PI labeling by PMA enhanced the competition of low concentrations of (-)vasopressin and angiotensin 11. Hepatocytes were incubated in epinephrine for [3H]prazosin binding.The significance of this or presence ( O " 0 ) of lo-' M PMA and the absence (W) increasing concentrations of angiotensin I1 (left) or vasopressin finding is presently unclear. Otherexperiments were per(right). Results are expressed as the per cent of basal incorporation formed inwhich membranes were isolated from livers perfused of [32P]Pi into PI,which was 129 f 15 and 93 f 6 cpm/mg of cell, for 10 min with Krebs-Ringer bicarbonate medium supplewet weight, in the absence or presence of PMA, respectively. Plotted mented with 10"j M PMA. The results of saturation curves are the means f S.E. of three experiments performed in duplicate. with [3H]prazosin, and of competition curves with (-)-epiThe stimulation of [32P]Piincorporation by and M vasonephrine with or without added PMA in these membrane pressin in cells treated with PMA are significantly (p < 0.001) in control memdifferent from the effects of these hormones in the absence of PMA. preparations were identical to those obtained branes obtained from livers perfused without PMA (data not previous section suggested that PMA may be acting at a step shown). Effect of PMA on Radioligand Binding to crl-Adrenergic prior to al-adrenergic receptor-mediated hydrolysis of membrane phosphoinositides. Thus, the effects of the phorbol ester Receptors inIntact Hepatocytes-Other investigators (10) on the bindingof [3H]prazosin to al-adrenergic receptors was have shown that phorbol estersdecrease the affinity of EGF studiedin rat liver membranes. Saturation curves of [3H] receptorsforEGF in cultured cells. This effect, however, prazosininmembranepreparations showed a n average of could only be observed when EGF binding was studied in cell disruption. 1019 f 126 fmol/mg protein (mean 2 S.E., n = 5 ) , and a Kd intact cell preparations, and disappeared upon we decided to studyradioliof0.2nM, which is in agreement with the results of other With these observations in mind, gand binding to al-adrenoceptors in whole isolated hepatoM PMAdidnot affect the investigators (17). Addition of saturation curve of [3H]prazosin. In addition, under condi- cytes, as well as the effects of PMA in this system. These tions in which unlabeled prazosin clearly competed with [3H] experiments were conducted both at 25 and at 37°C. The of [3H]prazosin prazosin for the al-adrenergic receptor binding site, PMA didsaturation curve and the binding kinetics not alter the binding of the radioligand (results not shown). were similar under both conditions, which suggests that exThere was no inhibition of [3H]prazosin binding a t concen- tensive metabolism of the ligands is not occurring in these trations of PMA (10-7-10-5 M ) that completely abolish the experiments. Moreover, after a 30-min incubation at 37"C,

524

IO0

-

Phorbol Esters Inhibit al-Adrenergic Actions T

A

50

MINUTES OF INCUBATION

o EPI

K o : 0.31 nM

10-5 M

0 VASO IO-B M

AN010

M

0

-Log

[PMA]

M

FIG. 7. Effects of PMA on the stimulation of PI labeling by (-)-epinephrine, vasopressin, and angiotensin 11. Hepatocytes M, angiotensin were incubated with lo-' M vasopressin (U) 11 (U or), M (-)-epinephrine (M), and increasing concentrations of PMA. The stimulation by the hormones was calculated as thepercentage of the basal incorporation of ["PIPi into PI at each of the concentrations of PMA employed (see Table I), thus correcting for the decrease in PI labeling produced by PMA itself. Results were then normalized to themaximal effects of the hormones, which were 598 f 18, 324 f 9, and 512 f 75% of basal incorporation for vasopressin, angiotensin 11, and (-)-epinephrine, respectively. Data are the means 9 S.E. of five experiments performed in duplicate.

more than 90% of the radioactivity in the incubation buffer was recovered as [3H]prazosin, as determined by thin-layer chromatography of the medium before and after incubation with the cells. The competition curves with (-)-epinephrine for bound [3H]prazosinwere also similar under both temperatures. In order to make the binding studies comparable to the metabolic studies in which the stimulation of urea synthesis was measured, all the results shown are from experiments conducted at 37°C. The binding kinetics of [3H]prazosin to a suspension of isolated hepatocytes is shown in Fig. 8A. The radioligand bound to thecells rapidly and reversibly, reaching equilibrium within 5 min, and it was stable for at least 30 min (Fig. 8A). The observed rate constant(Kobs) was calculated from a plot - B,)]uersus time for association, and was 0.34 of In [Beq/(Beq min-'. [ 3 H ] P r a ~ ~dissociated ~in from the binding sites with a half-time of dissociation of 12 min following addition of M phentolamine. The dissociation rate constant ( k 2 ) ,taken as the slope of a plot of In [Bt/Beq]uersus time, was 0.07 min-'. From these two values, the bimolecular association rate constant ( k l ) and thekinetically derived equilibrium dissociation constant (&) were calculated, and were found to be 0.27 X 10' min" M" and 0.25 nM, respectively. Fig. 8B shows that [3H]prazosin bindingto intact cells was saturable. Computer analysis of the data indicated that the binding was best fitted to a one-site model. A K d of 0.21 & 0.03 nM was determined from these data, which is in agree-

BOUND h o N m g l

0

I

I

0.5

1

I

I

2 3 f3tf1-PRAZ0SlN, nM

4

FIG. 8. Kinetics of [3H]prazosin binding to intact hepatocytes. Isolated hepatocytes were incubated at 37°C with 1 nM [3H] prazosin. At the times indicated, cells were centrifuged and washed as described under "Materials and Methods." Results were normalized to the percentage of maximal specific binding, which was 7.1 f 0.5 fmol/mg of cell, wetweight, and represented 70% of the totalradioligand bound. Reversal of binding was initiated with the addition of M phentolamine (arrow). Plotted are the means 9 S.E. of three experiments performed in duplicate. B, binding isotherm and Scatchard analysis of [3H]prazosin binding to intact hepatocytes. Cells were incubated at 37°C for 30 min in the presence of increasing concentrations of 13H]prazosin.The inset shows the Scatchard analysis of the binding isotherm. Data are the means -t S.E. of three experiments performed in duplicate.

ment with the kinetically derived dissociation constant (see above). The concentration of binding sites of isolated cells was of 10.4 rfr 0.5 fmol/mg of cell, wet weight, ( n = 12), which is equivalent to lo5 sites/cell. Both the dissociation constant and thenumber of specific [3H]prazosinsites/cell are in close agreement with the data obtained in liver membranes (17) and in whole cell preparations (22, 23). Comparison of the abilities of adrenergic agonists and antagonists to compete with (3H]prazosin indicated that, as we have previously demonstrated (23),binding was stereoselective and had the characteristics expected for al-adrenergic receptors (datanot M did not alter the binding of [3H]prazosin shown). PMA to intactcells (data not shown). In order to determine whether phorbol esters could alter the affinity of the al-adrenergic receptor for agonists, cells were incubated with [3H]prazosin and varying concentrations M PMA. of (-)-epinephrine in the presence or absence of It can be seen in Fig. 9 that the competition by (-)-epinephrine for [3H]prazosin binding to intact cells was biphasic. Computer analysis of these data revealed that (-)-epinephrine

al-Adrenergic Phorbol Actions Inhibit Esters

525

by these hormones is differentially altered by phorbol esters. Labeling induced by angiotensin I1 is unaltered by phorbol esters, that due to vasopressin is decreased, and that stimulated by (-)-epinephrine is completely blocked by PMA. The inhibition of a,-adrenergic actionsby phorbol esters is striking, because neither the metabolic effects nor the stimulation of phosphatidylinositol labeling by vasopressin or angiotensin I1 is inhibited in a similar way. It appears that the phorbol ester acts on the a,-adrenergic receptor molecule. One possible mechanism of inhibition could be that PMA is directly competing with (-)-epinephrine for the binding to the al-adrenergic receptor. To address this possibility, radioligand binding studies with [3H]prazosin were performed. PMA was unable to compete with [3H]prazosin forbinding to a,-adrenergic receptors in liver membranes. Thus, it appears that phorbol esters do not interactdirectly with al-adrenergic -Log r(-)EPINEPHRINEJ, M receptors. The specific receptor for phorbol esters has been shown to be protein kinase C. Thus, it is likely that the FIG.9. Effects of PMA on the competition by (-)-epinephinhibitory effects of PMA on a,-adrenergic actions aremedirine for [’Hlprazosin binding to intact hepatocytes. Cells were incubated with 0.25 nM [3H]prazosin a t 37°C for 30 min as described ated through this proteinkinase. It has been demonstrated that phorbol esters can inhibit under “Materials and Methods,” and increasing concentrations of the binding of insulin and EGF to their receptors in several (U or ) presence (W) of (-)-epinephrine in the absence IO-? PMA. Data from four experiments were calculated simultane- cell types (9, 10). It has also been shown that these agents ously with LIGAND, using computer-derived correction-factors for can increase the phosphorylation state of these receptors(24). the differences in binding-site concentrations. The depicted curves The possibility that PMA could be inhibiting the binding of are the “best fits”from these pooled experiments. (-)-epinephrinethroughalterations(perhapsphosphorylation) of thea,-adrenergic receptor was considered. Initial bound to two sites 0, < 0.001), which exhibiteddifferent experiments performed with ratliver membranes did not show affinities for the agonist. The high affinity sites represented any modification of the affinity of the a,-adrenergic receptor 30%of the total binding sites and bound (-)-epinephrine with for (-)-epinephrine under the influence of PMA. Also, the a dissociation constant (Kd)of 9 1 nM. The low affinity site affinity for (-)-epinephrine of membranespreparedfrom represented 70% of the total sites and had a Kd value of 2.0 livers perfused with or without PMAwas identical. However, f 0.2 p ~In.the presence of PMA,computeranalysis of it has been reported that thedecrease in affinity of the EGF the inhibition curve now revealed only a single class of binding receptors following treatment by PMA can only be observed sites for (-)-epinephrine. This site had an identical Kd (2 p M ) in preparations of intact cells, and disappears upon cell disto the low affinity site observed in control cells (Fig. 9). ruption (10).Considering that thispossibility could also apply to the effects of PMA on a,-adrenergic receptors, we develDISCUSSION oped a system to study radioligand binding in intact isolated Stimulation of liver cells by a,-adrenergic agonists, vaso- hepatocytes. The [3H]prazosin binding siteidentified in intact pressin, or angiotensin I1 results in an increase in cytosolic cells had characteristics similar to binding sitesidentified as free calcium,accompanied by a rapid breakdownof membrane a,-adrenergic receptors in membrane preparations from rat polyphosphoinositides and formationof diacylglycerol (4, 5). liver. The dissociation constant for [3H]prazosin observed in It is currently hypothesized that calcium phospholipid-de- intact cells (0.12-0.4nM) was practically identical to that pendent protein kinase C is activated by diacylglycerols and observed in membrane preparations(0.20 nM) and similar to thus forms part of the mechanism whichcouples receptor values published by a number of investigators using memactivation to intracellular responses (4-7). brane preparations (17) and whole cells (21-23). The inhibiEvidence in support of this hypothesis is provided by the tory constants of several agonists and antagonists were also increased phosphorylation by phorbol esters of some of the similar to those previously reported, and had the characterproteins that are phosphorylated in response to a,-adrenergic istics expected for binding to a,-adrenergic receptors. Thus, agonists, vasopressin,or angiotensin I1 in rat hepatocytes(6). the [3H]prazosin binding sites characterized as a,-adrenergic PMA binds to protein kinaseC and activates thisenzyme by receptors in intact hepatocytes were indistinguishable from substituting for diacylglycerols (3). Accordingly, phorbol es- those of membrane preparations. Computer analyses of the ters shouldbe able tomimic or magnify the metabolic actions competition curves of (-)-epinephrine for [‘Hlprazosin bindof hormones which produce phosphoinositidebreakdown. The ing to intact isolated hepatocytes reveal the presence of two results from this study show, however, that PMA affects the different sites for the agonist. Of the total [3H]prazosin bindaction of three of such hormones in different ways. Whereas ing sites, 30% show a high affinity (9 nM) for (-)-epinephrine, the stimulation of urea synthesis by angiotensin I1 is indeed whereas the remaining 70% are sites of lower affinity (2 p ~ ) . slightly enhancedby PMA, the stimulationof urea synthesis This finding is in agreement with the data of Goodhardt et by vasopressin is not affected by this agent. The effects of al. (25) and El-Refaiet al. (26), who have shown the presence (-)-epinephrine, mediated through a,-adrenergic receptors, of two affinitystates for (-)-epinephrinein[3H]prazosin are practically abolished by PMA. The stimulation of phos- binding and [3H]catecholamine binding studieswith rat liver phatidylinositol labeling by these hormones was investigated plasma membranes and with previous data using primary to determine where phorbol esters are producing their effects culture of rat hepatocytes( 2 3 ) .The additionof PMA produces in the sequenceof events leading from the hormone-receptor a marked change in the competitioncurve of (-)-epinephrine interaction to the activation of specific metabolic pathways. for [3H]prazosin binding to isolated hepatocytes. Computer These studies demonstrate that phosphatidylinositol labeling analysis of these data shows the presence of a single class of

526

al-Adrenergic Actions Inhibit Esters Phorbol

binding sites for (-)-epinephrine. Thesesitesare of low affinity and are indistinguishable from the low affinity sites observed in untreated cells. Because the totalnumber of [3H] prazosin binding sites is similar in control and phorbol estertreated cells, it therefore appears that the tumor promoter induces a transformation of the high affinity sites to a low affinity form. This transformation by PMA is remarkably similar to the effect of guanine nucleotides observed on al-adrenergicreceptors in rat liver plasma membranes (23, 25, 26). The involvement of guanine nucleotides in the regulation of the affinity state of al-adrenoceptors has raised the possibility that a guanine nucleotide-binding type protein may be involved in the a,-adrenergic transduction system (27). It is conceivable that phorbol esters, through the stimulation of protein kinase C, decrease the high affinity state of the al-adrenergic receptor by their interaction with such a protein. This interaction could be altered by a protein kinase C-mediatedmodification of the a,-adrenergic receptor-coupled guanine nucleotidebinding protein. However, PMA totallyblocks the stimulation of phosphatidylinositolturnover by al-adrenergicagents, whereas the effects of vasopressin and angiotensin I1 are not similarly affected. Therefore, it seems likely that the PMAinduced modification is at the level of the al-adrenergic receptor itself. Other hormone receptors, such as those for insulin, insulinlike growth factor I, and EGF also seem to be substrates for protein kinase C. In most cases, phosphorylation of these receptors is associated with adecrease in their affinity for the hormones (10,11, 24, 28). The &-adrenergic receptors may also be a substrate for protein kinase C ; the consequence of -such putative al-adrenergic receptor phosphorylation seems to be a decrease in affinity for (-)-epinephrine, which results in the inhibition of the effects of this hormone on hepatocyte metabolism. A similar inhibition of adrenergic effects as a consequence of adrenergic receptor phosphorylation is seen in the case of desensitization of the P-adrenergic receptor. Phosphorylation of these receptors, which act through elevations in cyclic AMP levels, seems to be driven by the cyclic AMP-dependent protein kinase(29). It seems possible, therefore, that protein kinase Ccould mediate the phosphorylation of the al-adrenergic receptor and produce desensitization to al-adrenergic stimulation. It remains to be established if a physiological activation of proteinkinaseC could lead to inhibition of a,-adrenergic effects. Insulin can decrease aladrenergic stimulation of hepatocyte metabolism (30,31). However, this effect is clearly different from the effect of PMA, since insulin does not block the effects of (-)-epinephrine on phosphoinositide metabolism (31). Studies todirectly determine whether the al-adrenoceptor can be phosphorylated by protein kinase C are presently being conducted. During the revision of this paper two papers have appeared describing effects of phorbol esters on al-adrenergic action in liver cells (32, 33). Acknowledgments-We thank Laureen Sena for technical assistance and Guadalupe Ramirez for typing the manuscript.

REFERENCES 1. Blumberg, P. M. (1980) Crit. Reu. Toxicol. 8, 53-234 2. Castagna, M.,Takai, Y., Kaibuchi, K., Sano, K., Kikkawa, U., and Nishisuka, Y. (1982) J. Biol. Chem. 257,7847-7851 3. Kikkawa, U., Takai, Y., Tanaka, Y., Miyake, R., and Nishizuka, Y. (1983) J. Biol. Chem. 258, 11442-11445 4. Fain, J . N., and Garcia-Sainz, J . A. (1980) Life Sci. 26, 11831194 5. Creba, J. A., Downes, C. P., Hawkins,P. T., Brewster, G., Michell, R. H., and Kirk, C. H. (1983) Biochem. J. 212, 733-749 6. Garrison, J. C., Johnsen, D. E., and Companile, C. P. (1984) J . Biol. Chem. 259,3283-3292 7. Cooper, R. H., Kobayashi, K., and Williamson, J. R. (1984) FEBS Lett. 166, 125-130 8. Roach, P. J., and Goldman, M. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 7170-7172 9. Corvera, S., and Garcia-Sainz, J. A. (1984) Biochem. Biophys. Res. Commun. 119,1128-1133 10. Thompoulos, P., Testa, U., Gourdin, M. F., Hervy, C., Titeux, M., and Vainchenker, W. (1982) Eur. J. Biochem. 129, 389393 11. Hollenberg, M. D., Nex, E., Berhanu, P., and Hock, R. (1981)

Receptor-mediated Binding and Internalization of Toxins and Hormones, pp. 181-196, Academic Press, Inc., New York 12. Berry, M. N., and Friend, D. S. (1969) J . Cell Biol. 43, 506-520 13. Tolbert, M. E. M., White, A. C., Aspry, K., Cutts, J., and Fain, J. N. (1980) J. Biol. Chem. 255, 1938-1944 14. Corvera, S., and Garcia-Sainz, J. A. (1981) Eur. J. Pharmacol. 72,387-390 15. Gutman, I., and Bergmeyer, H. V. (1974) in Methods ofEnzymatic Analysis (Bergmeyer, H. U., ed) Vol.4, pp. 1791-1794, Academic Press, New York 16. Garcia-Saini, J. A., and Fain, J. N. (1980) Biochem. J. 186,781789 17. Hoffman, B. B., Dikes, D. F., and Lefkowitz, R. J. (1981) Life Sci. 28,265-272 18. Lowry, 0. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275 19. Munson, P. J., and Rodbard, D. (1980) Anal. Biochem. 107,220239 20. Dreidger, P. E., and Blumberg, P. M. (1977) Cancer Res. 37, 3257-3265 21. Leach, K. L., James, M. L., and Blumberg, P. M. (1983) Proc. Natl. Acad. Sci. U. S. A. 80, 4208-4212 22. Mohel, N., Svartengren, J., and Cannon, B. (1983) Eur. J. Pharmacol. 92, 15-25 23. Schwarz, K., Lanier, S. M., Carter, E. A., Homcy, C. J., and Graham, R. M. (1985) Mol. Pharmacol. 27, 200-209 24. Jacobs, S., Sahyorin, N. E., Saltiel, A., and Cuatrecasas, P. (1983) Proc. Natl. Acad. Sci. U. S. A . 80, 6211-6213 25. Goodhardt, M., Ferry, N., Geynet, P., and Hanoune, J. (1982) J . Biol. Chem. 257, 11577-11583 26. El-Refai, M. F., Blackmore, P. F., and Exton, J.H. (1979) J . Bid. Chem. 254,4375-4386 27. Boyer, J. L., Garcia, A., Posadas, C . , and Garcia-Sainz, J. A. (1984) J . Biol. Chem. 259, 8076-8079 28. Brown, K., Dicher, P., and Rozengurt, E. (1979) Biochem. Biophys. Res. Commun. 86, 1037-1043 29. Stadel, J. M., Nambi, P., Shorr, R. G. L., Sawyer, D., Caron, M. G., and Lefkowitz, R. J. (1983) Proc. Natl. Acad. Sci. U. S. A. 80,3173-3177 30. Dehaye, J. P., Hughes, B. P., Blackmore, P. F., and Exton, J. H. (1981) Biochem. J. 194,949-956 31. Pushpendran, C. K., Corvera, S., and Garcia-Sainz, J. A. (1984) Biochem. Biophys. Res. Commun. 118,451-459 32. Lynch, C. J., Charest,R., Bocckino, S. B., Exton, J. H., and Blackmore, P. F. (1985) J . Biol. Chem. 260, 2844-2851 33. Cooper, R. H., Coll, K. E., and Williamson, J. R. (1985) J. Biol. Chem. 260,3281-3288