Prolonged activation of inhibitory somatostatin receptors increases adenylate cyclase activity in wild-type and Gsα-deficient (cyc−) S49 mouse lymphoma cells

Cellular Signalling Vol. 4, No. 5, pp. 571-581,1992.

089g-6568/92$5.00 + 0.00 PergamonPress Ltd

Printedin Great Britain.

PROLONGED ACTIVATION OF INHIBITORY SOMATOSTATIN RECEPTORS INCREASES ADENYLATE CYCLASE ACTIVITY IN WILD-TYPE AND Gs~-DEFICIENT (cyc-) $49 MOUSE LYMPHOMA CELLS JOHN M. THOMAS, SUSAN R. MEmR-DAVIS and BRIAN B. HO~MAN* Departments of Medicine and Pharmacology, Stanford University School of Medicine and Geriatric Research, Education and Clinical Center, Veterans Affairs Medical Center, Palo Alto, CA 94304, U.S.A.

(Received 20 February 1992; and accepted 19 March 1992) Abstract--Many cells develop enhanced adenylate cyclase activity after prolonged exposure to drugs that acutely inhibit the enzyme and it has been suggested that this adaptation may be due to an increase in G~. We have treated wild-type and G~-deficient cyc- $49 mouse lymphoma cells with a stable analogue (SMS 201-995) of the inhibitory agonist somatostatin. After incubation with SMS for 24 h, the forskolin-stimulated cAMP synthetic rate in intact cyc- cells was increased by 76%, similar to the increase found in the wild-type cells. Forskolin-stimulated adenylate cyclase activity in the presence of Mn 2+ was also increased in membranes prepared from SMS-treated cyc-cells; however, guanine nucleotide-mediated inhibition of adenylate cyclase activity was not changed despite a small decrease in inhibitory Gi~ subunits detected by immunobiotting. Pretreatment of cyc- cells with pertussis toxin prevented SMS from inducing the enhancement of forskolinstimulated cAMP accumulation in intact cells. After chronic incubation of cyc- cells with SMS, exposure to N-ethylmaleimide, which abolished receptor-mediated inhibition of cAMP accumulation, did not attenuate the enhanced rate of forskolin-stimulated cAMP synthesis compared to N-ethylmaleimide-treated controls. These results with cyc- cells demonstrate that an adaptive increase in adenylate cyclase activity induced by chronic treatment with an inhibitory drug can occur in the absence of expression of Gs~.

Key words: Adenylate cyclase, cAMP, inhibitory receptors, $49 cells, G protein.

INTRODUCTION

increased cAMP accumulation in cells after withdrawal of the inhibitory drug. This phenomenon of adenylate cyclase sensitization was originally observed in neuroblastoma x glioma hybrid cells (NG 108-15); pretreatment of N G 108-15 cells with morphine for at least 12h resulted in an enhanced basal and PGEcstimulated adenylate cyclase activity in membranes and an increased PGEj-stimulated cAMP accumulation in intact cells [2, 3]. These adaptive changes have been proposed to represent a biochemical correlation of narcotic tolerance and dependence [2]. Other inhibitory agonists (muscarinic cholinergic and alpha 2 adrenergic) also induce a similar sensitization of adenylate cyclase in N G 108-15 cells [4, 5]. Furthermore, this sensitization is observed in several different cells, such as pituitary cells,

A VARIETYof different types of cells exhibit an adaptive change in cAMP responsiveness after chronic treatment with drugs and hormones that acutely inhibit adenylate cyclase [1]. This adaptation is typified by enhanced stimulation of adenylate cyclase in membranes or an

* Author to whom correspondence should be addressed VA Medical Center, 182B, 3801 Miranda Avenue, Palo Alto, CA 94304, U.S.A. Abbreviations: BCIP--5-bromo-4-chloro-3-indolyl phospilate; BSA--bovine serum albumin; CAPS--3-(cyciohexylamino)-l-propanesulphonic acid; EGTA--ethylene-glycolbis(aminoethylether)tetra-acetate; GTP-~,- S---guanosine-5'O-(3-thiotriphosphate); HEPES ~ (2-hydroxyethyl)-lpiperazine-ethanesuiphonic acid; NBT nitroblue tetrazolium; NEM--N-ethylmaleimide; SMS--somatostatin analogue Sandostatin (SMS 201-995). at:

571

572

J.M. TxouAset al.

adipocytes, and cardiomyocytes, in response to chronic treatment with various drugs that acutely inhibit adenylate cyclase [1]. This phenomenon may represent a general cellular adaptation to chronic inhibition of adenylate cyclase. The molecular alteration(s) responsible for the expression of adenylate cyclase sensitization is uncertain. Several possibilities include: (i) an increase in the quantity or activity of the catalytic unit of adenylate cyclase; (ii) an increase in the quantity or activity of the stimulatory guanine nucleotide regulatory protein subunit(s) Gs~; (iii) a decrease in the quantity or activity of the inhibitory guanine nucleotide regulatory protein subunit(s) Gi~; or (iv) a decrease in the quantity or activity of the fl subunit associated with Gs. Several reports have described changes in the quantity of G protein subunits following chronic treatment of cells or animals with inhibitory drugs [6-21]. Since increases, decreases, or no changes in G protein subunits have been reported in these studies, there may be multiple adaptive mechanisms to account for the increase in adenylate cyclase activity. Also, it is uncertain to what degree the observed changes in quantity of G protein subunits contribute to the final expression of activity since the stoichiometric relationship between expression of Gs and adenylate cyclase activity is unclear. We have previously reported that chronic treatment of wild-type $49 mouse lymphoma cells with the inhibitory agonist somatostatin induces an increase in the forskolin-stimulated cAMP synthetic rate [22]. A mutant of the $49 cell, cyc-, does not express the G~ subunit or its mRNA, presumably due to a mutation in this gene [23, 24]. If chronic somatostatin treatment induces the sensitization response in cyccells, the molecular alteration in this cell would involve some mechanism other than an increase in expression of G~ or presumably a decrease in expression of the p subunit associated with G~. The present studies were designed to investigate the effects of chronic exposure to a somatostatin analogue on cAMP responses in eye- $49 cells.

MATERIALS AND METHODS Materials

Dulbecco's modified Eagle's medium (4500mg D-glucose/l) was from GIBCO (Grand Island, NY), and equine serum was from Hyclone Laboratories, Inc. (Logan, UT). cAMP antiserum was purchased from Research Products International (Mount Prospect, IL). Pertussis toxin was obtained from List Biological Laboratories, Inc. (Campbell, CA). N-Ethylmaleimide (NEM) was from CALBIOCHEM Corp. (San Diego, CA). Forskolin, cAMP, ATP, phosphocreatine, and creatine phosphokinase, Triton X-100, Tween 20, and bovine serum albumin were purchased from Sigma Chemical Co. (St Louis, MO). Guanosine-5'-O(thiotriphosphate) (GTP-~,-S) was obtained from Boehringer Mannheim Corp. (Indianapolis, IN). (~t'32p)ATP (10-50 Ci/mmol) and (8-3H)cAMP (20-30Ci/mmol) were purchased from Amersham Corp. (Arlington Heights, IL). Primary antibody AS/7 was a generous gift from Dr Alan Spiegel (Bethesda, MD) and antibody EC/2 was purchased from Du Pont NEN Research Products (Boston, MA). Biotinylated goat anti-rabbit IgG, alkaline phosphatase-streptavidin conjugate, 5-bromo-4chloro-3-indolyl phosphate (BCIP), and nitroblue tetrazolium (NBT) were purchased from Zymed Laboratories Inc. (South San Francisco, CA). RO 20-1724 was a gift from Hoffman-La Roche, Inc. (Nutley, N J) or purchased from BIOMOL Research Laboratories Inc. (Plymouth Meeting, PA). The somatostatin analogue Sandostatin (SMS 201-995) was kindly provided by Sandoz Research Institute (East Hanover, N J). Ce//s $49 wild-type and cyc- mutant cells (obtained from the University of California Cell Culture Facility, San Francisco, CA) were cultured in Dulbecco's modified Eagle's medium supplemented with NaHCO3 (l.0g/1) and heat-inactivated equine serum (10%), in a humidified atmosphere of 5% CO2 at 37°C. Cells were maintained at a density of 0.5-2.0 x 10 6 cells/ml by the daily addition of fresh medium. c A M P studies

$49 wild-type and eye- mutant cells were harvested in the logarithmic phase of growth and seeded at 0.6 and 0.8 x 106 cells/ml culture medium, respectively. Cells were incubated in the absence and presence of SMS (5 x 10-7 M) for 24 h. The cells were then washed three times at 22°C in culture medium (without NaHCO 3 or serum; supplemented with

Chronic activation of somatostatin receptors in cyc- cells 20 mM NaHEPES, pH 7.4) with intermediate centrifugations at 500g. Cells were resuspended in this medium and equilibrated to 37°C in a gyratory shaking bath. The initial forskolin-stimulated cAMP synthetic rate was measured as described previously [22], except that cells were pretreated with RO 201724 (5 x 10-4M) for 2min prior to addition of forskolin. The presence of phosphodiesterase inhibitor was required to achieve a linear forskolin-stimulated cAMP accumulation in cyc- cells, in agreement with previous results [25]. Cells were stimulated with forskolin (10 -4 M) by adding a 100-fold concentrated solution to the cell suspension. At 30-s intervals over a 5-min period, aliquots of cells were withdrawn into 10-fold concentrated HCI (final concentration, 0.1 N HCI) to terminate the reaction. Aliquots taken at the beginning and end of the time-course were used for cell counts. To determine the effect of pertussis toxin on the SMS-induced increase in forskolin-stimulated cAMP accumulation, cyc- cells were first incubated with and without pertussis toxin (25 ng/ml) for 24 h. To determine whether the inhibitory response to SMS was abolished by pertussis toxin, cells were washed, resuspended as above, and stimulated with forskolin (10-4M) and RO 20-1724 ( 5 x 1 0 - 4 M ) , in the absence or presence of SMS (5 x 10 -7 M ) at 37°C for 10min. The reaction was stopped by adding a 10fold concentrated solution of HCI to yield a final concentration of 0.1 N HCI. After the first 24-h incubation with pertussis toxin, each control and pertussis toxin-pretreated group was incubated with or without SMS (5 x 10 -7 M) for 24 h; pertussis toxin treatment was continued or not during this time interval. After pretreatments cells were washed, resuspended, and stimulated with forskolin as described above. The effect of NEM was tested after chronic incubation with SMS. Cyc- cells were incubated in the presence or absence of SMS (5 × 10-7M) for 24h as described above. After the second wash, each group was incubated in the presence or absence of NEM (5x 10-SM) for 10min at 22°C; this reaction was quenched with dithiothreitol (10-4M). After an additional wash, cells were resuspended and stimulated with forskolin plus RO 20-1724 in the absence or presence of SMS for 10 min as above.

Preparation of plasma membranes Cyc- cells were washed in medium as described above, resuspended in buffer (20mM NaHEPES, 150mM NaCI, I mM EGTA, 2 mM MgCI2, pH 7.5 at 4°C, and equilibrated in a Parr cell disruption bomb (Parr Instrument Co., Moline, IL) at 500 psi for 20 min. Cells were broken by abrupt release of

573

pressure, and membranes were collected by centrifugation (35,000g, 10min) and washed once with buffer. The crude membrane preparation was used in adenylate cyclase assays. Protein concentration was assayed by a modified Lowry method with bovine serum albumin (BSA) as standard [26].

Adenylate cyclase assay The assay of adenylate cyclase activity was based on a recently described protocol [27]. Enzyme activity was measured at 37°C in an assay volume of 0.05 ml. The assay buffer contained 50 mM Na-HEPES (pH 7.4), 0.1 mM EGTA, 0.5 mM ATP, 2 x 106 c.p.m. (~-32p)ATP, I mM cAMP, 5 mM phosphocreatine, and 50 U/ml creatine phosphokinase. Forskolin (10-4M), Mn 2÷ ( l m M ) , and GTP-),-S (10 -6 M) were added as indicated. Membrane protein (10-15/~g) was added to initiate the reaction, which was stopped after 10 min by the addition of 0.325N HCI containing 10,000 c.p.m. [3H]cAMP. cAMP was separated from ATP by a single column chromatography procedure [27].

lmmunoblotting Crude membranes as prepared above were extracted with Triton X-100 (1%) for I h at 0°C. Triton-extracted membrane proteins (15/~g/lane) were separated by S D S - P A G E (12% acrylamide gel) [28]. Protein bands were electrophoretically transferred to Immobilon P membrane (Millipore Corp., Bedford, MA) in CAPS [3-(cyclohexylamino)-lpropanesulphonic acid] buffer, pH 11.0, containing 10% methanol for 3h at 300mA in a Transblot apparatus (Bio-Rad Laboratories, Richmond, CA) with > 80% efficiency [29]. Immobilon blots were blocked with 3% BSA in Tris-buffered saline (25 mM Tris-HC1, pH7.5, 500mM NaCI, 0.05% NAN3) overnight at 37°C and then washed three times with 1% BSA in Tris-buffered saline. Membrane blots were incubated with primary antibodies AS/7 or EC/2, diluted 1:2000 in 1% BSA, for 12h at 22°C. Following three washes as above, blots were incubated with secondary antibody, biotinylated goat anti-rabbit IgG, diluted 1:250 in 20mM Tris-HC1, pH7.6, 137mM NaCI, 0.05% Tween 20, for l h at 22°C. After additional washes, detection was performed using an alkaline phosphatase-streptavidin conjugate and the substrates BCIP and NBT. Immunodetected bands were quantitated by scanning laser densitometry (Molecular Dynamics, Sunnyvale, CA). Primary antibody recognition of Gi protein subunits was linear within the range of membrane protein tested.

574

J . M . THOMAS et aL

(A)

Statistical analysis 250

Paired Student's t-tests were used for comparing groups. RESULTS Wild-type and cyc- $49 mouse lymphoma cells were incubated in the presence or absence of SMS for 24 h. After washing the cells free of SMS, the forskolin-stimulated cAMP synthetic rate was measured (Fig. 1, Table 1). Chronic SMS treatment of cyc- mutant cells, as well as wild-type cells, resulted in a significant increase in the forskolin-stimulated synthetic rate in intact cells. The absolute value for the synthetic rate is much less in cyc- cells than in wild-type cells, in agreement with other previous reports [30-32]. Nevertheless, the magnitude of the SMS-induced increase in cyc- is similar to that in wild-type. We also tested for the possibility of any induction of Gs~ activity in cyc- cells after chronic treatment with SMS by checking for isoproterenol stimulation of cAMP synthesis, which is dependent on G~ [30]; isoproterenol did not stimulate cAMP accumulation in cyc- cells with or without prior incubation with SMS (data not shown). Stimulation of adenylate cyclase by forskolin in the presence of Mn 2+ was compared in membranes from control cyc- cells and cells that had been incubated with SMS for 24 h. The activity (pmol/min/mg protein) was greater in membranes from SMS-pretreated cells (1117_+60) than in membranes from untreated control cells (964_+73; n = 13, P < 0.05). In parallel experiments with intact cells, i.e. when cells were subjected to longer washes and resuspension in buffer at 4°C in parallel with cells designated for membrane preparation, the SMS-induced increase in cAMP responsiveness was attenuated in magnitude to only a 36_+ 4% increase (n = 13), substantially less than the 76% increase observed under our routine conditions (Table 1). Consequently, the change induced by SMS is labile under conditions used for membrane preparation. Nevertheless, a small but significant increase in adenylate cyclase activity was still retained in membranes

200 ~o o

o WT $49 control / • WT S49 StlS p r e t r e e t e ( : l e J •

150'

100'

o. 50'

Time (minutes)

(B) 10 8

O cyc- $49 control • cyc- $49 SMS treated

•

S

2

0

1

2 3 4 Time (minutes)

5

6

FIG. 1. Forskolin-stimulated initial cAMP synthetic rate in (A) wild-type (WT) $49 cells and (B) eyemutant cells: control vs 24 h incubation with SMS. Wild-type or cyc- cells were either untreated controls (C)) or incubated with SMS for 24 h (0). After this incubation, cells were washed free of SMS and stimulated with forskolin in the presence of the phosphodiesteras¢ inhibitor RO 20-1724. Aliquots were withdrawn over a 5-rain time-course to estimate the initial cAMP synthetic rate. These experiments were repeated four times for wild-type cells and eight times for cyc- cells (see Table 1 for summary of rate constants). from cells chronically treated with SMS. Guanine nucleotide-mediated inhibition of adenylate cyclase activity was compared in the two groups using the analogue GTP-~-S. GTP-~-S (10-6M) inhibited forskolin stimulation by 52.1 + 2.3% in membranes from control

A

B

66-

4536-

2924-

2014......

C

ii

S

C S

FIG. 2. Immunoblot of membrane proteins to quantitate Gi~ subunits in cyc- cells. Cyc- cells were either (C) untreated controls or (S) incubated with SMS for 24 h, Blots were prepared as described in Materials and Methods with incubation with (A) AS/7 antibody to detect Gi~, 2, and (B) EC/2 antibody to detect G~3. Molecular weight standards are indicated on the left (mol.wt× 10-3). This experiment was repeated six times, with the following result: chronic SMS pretreatment caused a 19'/o decrease (P < 0.01) in the quantity of G~j,2 and a 24% decrease in (P < 0.01) in the quantity of Gi~3. 575

Chronic activation o f sornatostatin receptors in cyc- cells

577

TABLE I. FORSKOLIN-STIMULATED c A M P SYNTHETXC RATES IN INTACT $ 4 9 WILD-TYPE AND c y c - MUTANT CELLS: CONTROL VS S i S PRETREATMENT

Initial synthetic rate (pmol cAMP/106 cells/min) Control SMS-pretreated WT (n = 4) cyc- (n = 8)

20.22+2.02 1.069+0.104

37.66+3.77 1.874+0.183

SMS-induced increase (%) 87+ 11 (P < 0.001) 76+5 (P < 0.001)

$49 wild-type (WT) and cyc- mutant cells were either untreated control or pretreated with SMS and stimulated with forskolin as in Fig. 1. Data are means_S.E.M. Statistical comparisons are by paired t-tests.

cells and by 51.2-t-2.8% in membranes from cells chronically treated with SMS (n = 6, not significant). We next addressed whether there might be a change in the quantity of Gi= subunits in membranes from eye- cells chronically treated with SMS. Immunoblots were performed with AS/7 antibody, to detect Gi=l,2 [33], and EC/2 antibody to detect Gi=3 [34] (Fig. 2). Although EC/2 antibody also recognizes Go= [34], Go= was not detected with a specific antibody in cyccells [35]. Chronic SMS treatment of cyc- cells resulted in a 19.2-t-4.3% decrease (n = 6, P < 0.01) in the quantity of G~,~.2 and a 24.4+ 5.8% decrease (n = 6, P < 0.01) in the quantity of

pretreatment in control cells (Fig. 3). When the acute inhibitory response to SMS was abolished after 24 h of pertussis toxin pretreatment, subsequent chronic treatment of the cells with SMS did not result in an additional increase in forskolin-stimulated cAMP synthesis. This result suggests that a pertussis toxin-sensitive G protein transduces the signal for chronic inhibitorg Pertussis Toxin

200o¢u

Gi= 3•

To characterize further the chronic SMS-induced sensitization of adenylate cyclase in cyc- cells, we tested whether inactivation of G i by pertussis toxin would also prevent the chronic SMS-induced increase in forskolinstimulated cAMP accumulation. Cyc- cells were incubated with and without pertussis toxin for 24 h and subsequently incubated with and without SMS for 24 h in the continued presence or absence of pertussis toxin. Forskolin-stimulated cAMP was measured in intact cells. In control cyc- cells, SMS inhibited forskolin-stimulated cAMP accumulation by about 34%. However, after 24 and 48h of pretreatment with pertussis toxin, the inhibition was completely prevented (n = 4), in agreement with other previous results [36]. Pertussis toxin pretreatment alone caused an increase in forskolin stimulation to the same extent as SMS

Control

-5 E

o

E

0 U.

0

C SMS

1

C SMS

FXG. 3. Forskolin-stimulated cAMP accumulation in cyc- cells: control vs prolonged incubation with SMS, with and without prior pretrcatment with pertussis toxin. Cyc- cells were either untreated controls or pretrcated with pertussis toxin for 24 h. Each of these groups was then either untreated control (C) or incubated with SMS for 24 h (SMS). After these pretrcatments, cells were washed free of

SMS and stimulated with forskolin in the presence of RO 20-1724 for 10min. Data are expressed as percentage of response of control cells to forskolin (no pertussis toxin or SMS pretreatments) and are means_+ S.E.M. of four experiments, each performed in duplicate.

J . M . THOMAS et al.

578

(A) Control

SMS pretreatea

60 O-

~te v

40

20 ¸ ffl u~

C NEM

(B)

Control

C NEM

SMS pretroated

450.

4ooo

z5 0

o_ ~

300

u !

250

~ S v E

2oo

= 'S

loo

•

g

1so

~

~o

u.

0

C NEM

C NEM

FIG. 4. (A, B) SMS-mediated inhibition and forskolin-stimulated cAMP accumulation in cyc- cells: control vs prolonged incubation with SMS, with and without subsequent acute treatment with NEM. Cyc- cells were either untreated control or incubated with SMS for 24 h. Cells were washed free of SMS; during the second wash each group was either not treated (C) or treated with NEM for 10rain, followed by dithiothreitol (NEM). After another wash, cells were stimulated for 10 rain with forskolin in the presence of RO 20-1724, and in the absence or presence of SMS to measure the capacity of SMS to acutely inhibit forskolin-stimulated cAMP a c c u m u lation. Data are expressed as (A) percentage SMS-mediated inhibition of forskolin stimulation for each group or (B) percentage of control forskolin stimulation (without chronic SMS pretreatrnent or acute NEM treatment); data are means-I-S.E.M, of nine experiments, each performed in duplicate.

induced sensitization of adenylate cyclase in cyc- cells; in addition, this result raises the possibility that the explanation for enhanced cAMP responses after prolonged exposure to SMS relates to diminished function of G~. N E M has been shown to rapidly inactivate receptor-mediated inhibition of adenylate cyclase in many different cells [37, 38]. Also, N E M has been reported to eliminate essentially GTP-~,-S-mediated inhibition of adenylate cyclase in membranes from cyc- cells [39]. These effects of NEM probably result from alkylation of a G protein at specific sites [40, 41]. We tested the effect of N E M on intact cyccells after 24 h incubation with SMS, as illustrated in Fig. 4. As expected, treatment of intact cyc- cells with NEM substantially impaired the SMS-mediated inhibition o f forskolin-stimulated cAMP accumulation. Also, N E M enhanced forskolin-stimulated cAMP; the effect o f N E M on forskolin stimulation of cAMP accumulation was similar to the effect of pertussis toxin as shown in Fig. 3. In cells incubated with SMS for 24h, the inhibitory response to SMS was desensitized, and an additional treatment with NEM completely abolished the SMS-mediated inhibition of cAMP accumulation. Interestingly, cells preincubated with SMS and then exposed to NEM maintained an augmented response to forskolin compared to cells exposed only to N E M (Fig. 4). This result suggests that the SMS-induced increase in forskolin-stimulated cAMP accumulation is not simply due to a deficit in function of G~. DISCUSSION This study demonstrates that an increase in adenylate cyclase activity occurs after chronic incubation with an inhibitory agonist in $49 eye- cells, which do not express G~. The increase in activity, assessed by cAMP synthesis in intact cells, is prevented by pretreatment of the cells with pertussis toxin. However, the increase in activity in SMS-treated cells persists after exposure to NEM, which inactivates receptor-mediated inhibition of adenylate

Chronic activation of somatostatin receptorsin cyc- cells cyclase. After the prolonged incubation with SMS, there is a small decrease in inhibitory G , subunits; however, guanine nucleotidemediated inhibition of adenylate cyclase activity in these membranes is unaltered. The magnitude of the SMS-induced increase in forskolin-stimulated cAMP synthetic rate in intact cyc- cells is similar to that observed in wild-type $49 cells. However, the absolute values for forskolin-stimulated synthetic rate in cyc- cells is about 5% of that in wild-type cells. This observation reflects the synergism of (3, and the catalytic unit of adenylate cyclase in response to stimulation by forskolin [30-32]. As in several other types of cells [42--44], pretreatment of cyc- cells with pertussis toxin prevents both the acute inhibition of cAMP accumulation as well as the inhibitor-induced enhancement in stimulation of cAMP accumulation. This suggests that a pertussis toxin-sensitive G protein functions in transducing the signal for induction of adenylate cyclase sensitization in cyc- cells. It is uncertain whether the lack of adenylate cyclase inhibition or loss of some other G protein-mediated response after pretreatment with pertussis toxin is the reason for the lack of development of sensitization after chronic treatment with SMS. Activation of receptors coupled to inhibition of adenylate cyclase via a pertussis toxin-sensitive G protein often has additional effects in other cells or tissues [45]. Defining the molecular alteration responsible for the expression of increased adenylate cyclase activity after chronic exposure to inhibitory drugs such as opiates has been a major goal. There has been much focus on the possibility that altered expression of Gs or Gi may explain this adaptation. Several studies have provided evidence for changes in the quantity of G protein subunits following chronic treatment of cells or animals with inhibitory drugs [6-21]. One possible alteration to explain the increase in adenylate cyclase activity would be an increase in the quantity of G=. An increase in the quantity of G~ as measured by cholera toxin-catalysed ADP-ribosylation [9] or immunoblotting [15] has been reported for CELLS 4:5-H

579

adipocytes from rats chronically treated in vivo with an agonist that activates inhibitory A~ adenosine receptors. In contrast, a decrease in cholera toxin labelling has been reported in isolated rat fat cells acutely treated with this adenosine receptor agonist, possibly due to an increase in endogenous ADP ribosylation of G~ [7]. An increase in adenylate cyclase activity •was found in each of these studies involving rat adipocytes [7, 9, 15]. Other studies using neuronal cells or brain tissues have not detected a change in the quantity of G= despite some increase in adenylate cyclase activity [12, 14, 17] or have not detected a change in G~ or adenylate cyclase activity [11]. Another possible alteration to explain the increase in adenylate cyclase activity is a decrease in the quantity of Gi~ subunits. Several reports have provided evidence for a decrease in Gi~ subunits after chronic treatment of animals or various cells with inhibitory agonists; these decreases were detected by either pertussis toxin-catalysed ADP ribosylation [8, 9, 21] or immunoblotting [11, 15, 18, 19]. A decrease in pertussis toxin labelling could possibly be due to an increase in endogenous ADP ribosylation rather than a quantitative decrease in mass of Gi+. Such a change could also explain an increase in adenylate cyclase activity. Decreases in G+ffisubunits were detected in parallel with increases in G , in rat adipocytes [9, 15]. In cultured neurons, a substantial decrease in G+~ was detected after chronic opiate treatment, but there was no change in adenylate cyclase activity or other G protein subunits [11]. On the other hand, a lack of alteration in the quantity of G~ subunits in brain tissue [13, 21] or cultured neuronal cells [6, 10, 14] after chronic inhibitor treatment has been reported. An apparently paradoxical increase in G,~ subunits in brain tissue [12, 17, 20, 21], cultured brain cells [19], or myenteric plexus [16] has also been reported, in some cases associated with an increase in adenylate cyclase activity [12, 17, 20, 21]. It was suggested that an increase in Gi~ might explain the increased enzyme activity via a mass action effect on the l/ subunit, which would decrease the quantity of 1/subunit avail-

580

J.M. THOMAS et al.

able for interaction with G~ and thus render G,~ less inhibited by the fl subunit [12]. Since the Gi~: subunit has been implicated as the Gi~ subunit coupled to inhibition of adenylate cyclase [35, 46], changes in the other Gi~ subunits may not be relevant to an increase in adenylate cyclase activity. Another possible change in G protein subunits to explain an increase in adenylate cyclase activity would be a decrease in the fl subunit. A decrease in the quantity of fl subunit detected by immunoblotting was reported for cultured rat adipocytes chronically treated with an adenosine receptor agonist [18]. Studies using other experimental systems have suggested either an increase [16] or no change [11, 12, 17] in the quantity of fl subunit. The quantity of the subunit of G o has also been monitored by immunoblotting in several studies on the effects of chronic exposure to inhibitory drugs. An increase [12, 16, 17, 19], decrease [19], and no change [10, 11] have been reported. Although the ~to subunit is not coupled directly to adenylate cyclase, it is possible that this subunit might influence the enzyme activity via a cross-talk mechanism [47, 48]. In summary, various combinations of alterations in the quantity of G protein subunits have been described in several experimental systems after chronic treatment with drugs that inhibit adenylate cyclase activity. In some, but not all, cases these changes have been associated with an increase in adenylate cyclase activity. Consequently, it is uncertain to what extent changes in quantity of G protein subunits contribute to alterations in adenylate cyclase activity. In addition, it is not clear whether an increase in Gs would necessarily increase activation of adenylate cyclase since the amount of G s may not be rate-limiting. Using immunoblotting, we have detected small but significant decreases in the quantities of G~,: and Gia 3 in membranes from cyc- cells chronically treated with SMS. However, the guanine nucleotide-mediated inhibition of adenylate cyclase activity in membranes from cyc- cells chronically treated with SMS was not different from that in membranes from control

cells. Also, the chronic SMS-induced enhancement in forskolin-stimulated cAMP accumulation was evident in intact cells when the functional inhibitory response was abolished by NEM. We used NEM to treat the cells after the prolonged incubation with SMS because NEM has previously been shown to inactivate rapidly receptor-mediated inhibition of adenylate cyclase in many different cells [37, 38] and essentially to eliminate GTP-~-S inhibition of adenylate cyclase in membranes from cyc- cells [39]. Consequently, two separate pieces of functional experimental evidence suggest that the SMS-induced increase in forskolin stimulation is not due to an attenuation in the inhibitory pathway. Therefore, the decreased quantity of G~ subunits in membranes from chronically treated cyc- cells does not appear to be responsible for the enhanced stimulation of adenylate cyclase activity by forskolin. We have tested whether prolonged activation of an inhibitory receptor can induce an increase in adenylate cyclase activity in the cyc- mutant of the $49 mouse lymphoma cells which lack G~ stimulatory coupling proteins. These experiments were undertaken since an increase in quantity or activity of G~ is a possible explanation for the increase in enzyme activity and an increase in the quantity of Gs~ has been reported in some cases of chronic treatment with an inhibitory drug. The results with cyccells demonstrate that an adaptive increase in adenylate cyclase activity induced by chronic treatment with an inhibitory drug can occur in the absence of expression of G~. Also, although we have found a small decrease in quantity of inhibitory Gi~ subunits following prolonged exposure to inhibitory drug, there does not appear to be a loss in Gi~ activity which might otherwise explain the enhanced adenylate cyclase activity. Therefore, in the cyc- $49 cell, a change in quantity or activity of G proteins does not appear to be a major contributor to the adaptive increase in adenylate cyclase activity induced by chronic treatment with inhibitory drug. Acknowledgements--This work was supported by National Institutes of Health grants 5RO 1 HL 41315

Chronic activation of somatostatin receptors in cyc- cells and AG09597 and the Research Service of the Department of Veterans Affairs. B. B. H. was an Established Investigator of the American Heart Association at the time this research was conducted. REFERENCES 1. Thomas J. M. and Hoffman B. B. (1987) Trends pharmac. Sci. 8, 308-311. 2. Sharma S. K., Klee W. A. and Nirenberg M. (1975) Proc. natn. Acad. Sci. U.S.A. 72, 3092-3096. 3. Traber J., Gullis R. and Hamprecht B. (1975) Life Sci. 16, 1863-1868. 4. Nathanson N. M., Klein W. L. and Nirenberg M. (1978) Proc. natn. Acad. Sci. U.S.A. 75, 1788-1791. 5. Sabol S. L. and Nirenberg M. (1979) J. Biol. Chem. 254, 1921-1926. 6. Law P.-Y., Louie A. K. and Loh H. H. (1985) J. biol. Chem. 260 14,818-14,823. 7. Jacquemin C., Thibout H., Lambert B. and Correze C. (1986) Nature 323, 182-184. 8. Green A. (1987) J. Biol. Chem. 262, 15,702-15,707. 9. Parsons W. J. and Stiles G. L. (1987) J. Biol. Chem. 262, 841-847. 10. Lang J. and Costa T. (1989) J. Neurochemistry 53, 1500-1506. 11. Attali B. and Vogel Z. (1989) J. Neurochemistry 53, 1636-1639. 12. Nestler E. J., Erdos J. J., Terwilliger R., Duman R. S. and Tallman J. F. (1989) Brain Res. 476, 230-239. 13. Nestler E. J., Terwilliger R. and Beitner D. (1989) Life Sci. 45, 1073-1080. 14. Convents A., de Backer J.-P., Andre C. and Vauquelin G. (1989) Biochem. J. 262, 245-251. 15. Longabaugh J. P., Didsbury J., Spiegel A. and Stiles G. L. (1989) Molec. Pharmac. 36, 681-688. 16. Lang J. and Sehulz R. (1989) Neuroscience 32, 503-510. 17. Nestler E. J. (1990) Prog. Cell Res. 1, 73-88. 18. Green A., Johnson J. L. and Milligan G. (1990) J. Biol. Chem. 265, 5206-5210. 19. Vogel Z., Barg J., Attali B. and Simantov R. (1990) J. Neurosci. Res. 27, 106-111. 20. Rasmussen K., Bcitner-Johnson D. B., Krystal J. H., Aghajanian G. K. and Nestler E. J. (1990) J. Neuroscience 10, 2308-2317. 21. Terwilliger R. Z., Beitner-Johnson D., Sevarino K. A., Crain S. M. and Nestler E. J. (1991) Brain Res. 548, 100-110. 22. Thomas J. M. and Hoffman B. B. (1988) Molec. Pharmac. 34, 116-124. 23. Johnson G. L., Kaslow H. R., Farfel Z. and Bourne H. R. (1980) J. Cyclic Nucl. Res. 13, 1-37.

581

24. Harris B. A., Robishaw J. D., Mumby S. M. and Gilman A. G. (1985) Science 229, 1274-1277. 25. Clark R. B., Goka T. J., Green D. A., Barber R. and Butcher R. W. (1982) Molec. Pharmac. 22, 609-613. 26. Peterson G. L. (1977) Analyt. Biochem. 83, 346-356. 27. Alvarez R. and Daniels D. V. (1990) Analyt. Biochem. 187, 98-103. 28. Laemmli U. K. (1970) Nature 227, 680-685. 29. Matsudaira P. (1987) J. Biol. Chem. 262, 10,035-10,038. 30. Daffier F. J., Mahan L. C., Koachman A. M. and Insel P. A. (1982) J. Biol. Chem. 257, 11,901-11,907. 31. Green D. A. and Clark R. B. (1982) J. Cyclic Nucl. Res. 8, 337-346. 32. Downs R. W. Jr and Aurbach G. D. (1982) J. Cyclic Nucl. Res. 8, 235-242. 33. Spiegel A. M. (1990) In G Proteins (Iyengar R. and Birnbaumer L., Eds), pp. 115-143. Academic Press, San Diego. 34. Simonds W. F., Goldsmith P. K., Codina J., Unson C. G. and Spiegel A. M. (1989) Proc. hath. Acad. Sci. U.S.A. 86, 7809-7813. 35. Lang J. and Costa T. (1989) J. Receptor Res. 9, 313-329. 36. Aktories K., Schultz G. and Jakobs K. H. (1983) FEBS Left. 158, 169-173. 37. Jakobs K. H., Lasch P., Minuth M., Aktories K. and Sehultz G. (1982) J. Biol. Chem. 257, 2829-2833. 38. Smith M. M. and Harden T. K. (1984) J. Pharmac. exp. Ther. 228, 425-433. 39. Jakobs K. H., Gehring U., Gaugler B., Pfeuffer T. and Sehultz G. (1983) Eur. J. Biochem. 130, 605-611. 40. Winslow J. W., Bradley J. D., Smith J. A. and Necr E. J. (1987) J. Biol. Chem. 262, 4501-4507. 41. Hoshino S., Kikkawa S., Takahashi K., Itoh H., Kaziro Y., Kawasaki H., Suzuki K., Katada T. and Ui M. (1990) FEBS Lett. 276, 227-231. 42. Thomas J. M. and Hoffman B. B. (1986) J. Cyclic Nucl. Res. Protein Phosphorylation 11, 317-325. 43. Linden J. (1987) FASEB J. 1, 119-124. 44. Jones S. B. and Bylund D. B. (1988) J. Biol. Chem. 263, 14,236-14,244. 45. Limbird L. E. (1988) FASEB J. 2, 2686-2695. 46. McKenzie F. R. and Milligan G. (1990) Biochem. J. 267, 391-398. 47. Houslay M. D. (1991) Eur. J. Biochem. 195, 9-27. 48. Moriarty T. M., Padrell E., Carry D. J., Omri G., Landau E. M. and Iyengar R. (1990) Nature 343, 79-82.