- Email: [email protected]
α2-adrenergic binding and action in human adipocytes. Comparison between binding to plasma membrane preparations and to intact adipocytes
European Journal of Pharmacology, 119 (1985) 101-112
101
Elsevier
a2-ADRENERGIC B I N D I N G AND A C T I O N IN H U M A N A D I P O C Y T E S . C O M P A R I S O N B E T W E E N B I N D I N G T O P L A S M A M E M B R A N E P R E P A R A T I O N S AND T O I N T A C T A D I P O C Y T E S B J ~ R N RICHELSEN * and O L U F PEDERSEN
Medical Department I11, Division of Endocrinology and Metabolism, Aarhus County Hospital, Aarhus, Denmark Received 23 August 1985, accepted 1 October 1985
B. RICHELSEN and O. PEDERSEN, a2-Adrenergic binding and action in human adipocytes. Comparison between binding to plasma membrane preparations and to intact adipocytes, European J. Pharmacol. 119 (1985) 101-112. Binding of the a/-adrenoceptor antagonist [3H]yohimbine was demonstrated on intact human adipocytes and on human adipocyte membranes. Specific binding was rapid, reversible, saturable and of high affinity in both preparations. [3 H]Yohimbine binding was inhibited by various adrenergic agents in a manner which suggests that the labeled sites probably represent the a2-receptor both in intact adipocytes and in the membrane fraction. In adipocyte membranes the maximal binding capacity (Bmax) was 463 +_ 38 fmol/mg protein and the binding was of high affinity with a K d of 2.1 _+0.4 nM. In intact human adipocytes, Bmax was 903 + 139 fmol/106 cells (or 342 + 21 fmol/100 cm2) and the binding affinity was 6.6 + 0.8 nM. Adrenergic antagonists bound to a homogeneous class of receptors (linear Scatchard plots) both in isolated membranes and in intact adipocytes. However, agonist binding was heterogeneous in both preparations. The affinity of agonist binding was 5-10 times higher in membranes than in intact adipocytes. The physiological relevance of the binding data was evaluated by correlating the binding of yohimbine with the antilipolytic effect of clonidine. A positive and significant correlation was found between BmaX and maximal antilipolytic effect of clonidine ( r = 0.70) in membranes. Furthermore, the binding affinity (Kd) was positively correlated to the sensitivity (ICs0) of the clonidine-induced antilipolysis (r=0.65). A positive and significant correlation was also found in intact adipocytes between Bmax and the maximal antilipolytic effect of clonidine (r = 0.79). However, there was no significant correlation between K d in intact adipocytes and the ICs0 of clonidine. It is concluded that both intact human adipocytes and membrane fractions are useful models to investigate the properties and regulation of a/-adrenoceptors. However, it appears from our correlation studies that the binding data obtained with the membranes are the best related to the physiological effects mediated by a2-receptors in human adipocytes. a2-Adrenoceptors
Intact human adipocytes
Adipocyte membranes
1. Introduction It is now accepted that human adipocytes possess both fl- and a2-adrenoceptors in the plasma membrane (Burns et al., 1981). Catecholamines stimulate both kinds of receptors which cause activation of the adenylate cyclase complex through the fl-receptor, and inhibition of the same complex through the az-receptor (Burns et al., 1981; Fain and Garcia-Sainz, 1983; Kather and * To whom all correspondence should be addressed: Medical Department III, Tage Hansens Gade, Aarhus amtssygehus, DK-8000 Aarhus C, Denmark. 0014-2999/85/$03.30 © 1985 Elsevier Science Publishers B.V.
[3H]Yohimbine
Antilipolyse
Simon, 1981). This a2-receptor-induced inhibition of adenylate cyclase in human adipocytes is presumably mediated through the guanine nucleotide binding regulatory component (N i) (Rodbell, 1980; Fain and Garcia-Sainz, 1983). In adipocytes, an increased concentration of cyclic A M P stimulates lipolysis whereas a decreased concentration of cyclic A M P inhibits lipolysis (Bums et al., 1980; Kather et al., 1980). Thus, it seems to be the balance between the fl- and az-adrenoceptors that controls the final lipolytic response induced by catecholamines in human adipocytes. Furthermore, human adipocytes also possess al-adrenoceptors. These receptors seem to be independent
102
of the adenylate cyclase system. Activation of these a~-receptors results in increased turnover of phosphatidylinositoland phosphatidic acid in human adipocytes (Burns et al., 1981; Fain and GarciaSainz, 1983). Recently, c~2-adrenoceptors have been demonstrated directly on human adipocyte membranes by using the radioactive, t~2-receptor antagonist, yohimbine ([3H]YOH) (Tharp et al., 1981; Lafontan et al., 1983; Berlan and Lafontan, 1982). However, no radioligand studies concerning the c~2-receptors have been performed using intact human adipocytes *, even though cell disruption could induce changes in receptor properties and regulation. There have been detailed analyses of the properties of /3-adrenoceptors in intact cells (Mukherjee and Lefkowitz, 1976; Pittman and Molinoff, 1980; Dulis and Wilson, 1980). The main findings from these studies were that the properties of agonist binding to the fl-receptor were different from the binding characteristics of membrane preparations. Generally, the affinities of agonists for the receptors were much weaker in intact cells than in membranes, whereas affinities of antagonist binding were similar in these two preparations. We have now characterized the binding of [-~H]YOH to intact human adipocytes and compared these data with results obtained from adipocyte membrane preparations of the same system. Furthermore, these binding data were correlated with the biological effect (antilipolysis) mediated by these receptors.
2. Materials and methods
2.1. Isolation and adipocytes Subcutaneous adipose tissue was obtained from patients undergoing abdominal surgery. The adipose tissue was collected at the beginning of the surgical procedure. Patients fasted overnight before tissue removal and had no identified metabolic or endocrine disorders. However, some were * A report by Engfeldt et al. has appeared (Scand. J. Clin. Invest. 1983, 43, 207).
obese ( > 20% of ideal body weight). There was no selection according to sex or age of the donors. Isolated adipocytes were prepared and cell concentration calculated as previously described (Pedersen et al., 1981; Richelsen et al., 1984). Briefly, the adipocytes were isolated by collagenase digestion of fat tissue fragments for 60 min at 37°C in 10 mM Hepes buffer containing 0.5 m g / m l collagenase and 2.5% bovine serum albumin (BSA). The adipocytes were then filtered on a silk screen and washed three times. The cell concentration in a given suspension was calculated as follows: the diameters of adipocytes were measured (SjOstr0m et al., 1971; Pedersen et al., 1981). Surface area (~rD 2) and volume (7rD3/6) were calculated for each cell diameter measured (about 200 individual cells per fat tissue preparation). The volume fraction of adipocytes was measured in a hematocrit capillary tube. The concentration of cells in the adipocyte suspensions was obtained as the adipocyte volume fraction divided by the mean adipocyte volume,
2.2. Preparation of human adipocyte membranes Isolated adipocytes were washed three times in a lysing medium (5 mM Hepes, 2.5 mM MgCI 2, 1 mM KHCO3, pH 7.6) then centrifuged of 3000 × g for 4 min. All the adipocytes were broken after this procedure. The adipocyte ghosts and crude broken membranes were then centrifuged at 24 000 × g for 20 min at 4°C. The pellet was washed in 10 mM Hepes buffer containing 1 mM EDTA and again centrifuged at the same speed. Finally, the membrane preparation was resuspended in 10 mM Hepes buffer containing 10 mM MgCI 2 and used immediately in the binding assay or frozen at - 8 0 ° C . The protein concentration of the membrane was measured (Lowry et al., 1951) with BSA as the standard.
2.3. Binding of [3H]YOH to intact human adipocytes Isolated adipocytes were washed four times by centrifugation to remove albumin from the cells. The adipocytes were then resuspended in a buffer containing the following solute concentrations in
103 mM: Hepes 10, NaC1 135, KC1 4.8, MgSO 4 1.7, CaC12 2.5, NaH2PO 4 0.2, N a 2 H P O 4 1.0, glucose 5.0 and pH 7.4. The binding study was performed in a final volume of 325 /tl. The incubation mixture consisted of 250 ~1 adipocytes (about 105 cells/ml), 25 #1 buffer or unlabeled ligand for competition, and 25 btl [3H]YOH (0.3-15 nM). There was also added 25/~1 of a mixture of ascorbic acid (0.5 mM) and catechol (0.3 mM) to diminish the oxidation of adrenoceptor agonists. For competition studies [3H]YOH was usually added to reach a final concentration of 1-2 nM. The binding incubation was carried out for 25 min at 20°C without shaking. The incubation was terminated by adding 5 ml of ice-cold buffer to each tube. The incubation mixture was then immediately vacuum-filtered through Whatman G F / C glass fiber filters placed on a Millipore manifold. The filters were washed twice with 10 ml ice-cold buffer. After the filters had been dried for 3 h they were placed in 5 ml scintillation fluid (Lumagel, Lumac BV, Netherlands), and counted in a Rackbeta counter (LKB, Finland). Non-specific binding was determined by adding unlabeled yohimbine (1/~M) or clonidine (10/~M) to the incubates. At a tracer concentration of 1-2 nM non-specific binding was 20-40% of the total binding to the adipocytes. Yohimbine and clonidine gave equivalent results at these concentrations. However, unlabeled yohimbine at even higher concentrations further inhibited some [3H]YOH binding in intact adipocytes. Presumably, this further inhibition of binding might have been at non-specific binding sites since clonidine from higher concentrations was unable to reproduce the effect. Specific binding was taken as the difference between total and non-specific binding. It was shown microscopically that the adipocytes remained intact after incubation for 50 min in this buffer, with unaltered mean fat cell volume before and after the incubation (422 _+ 22 pl and 406 + 24 pl, respectively, P > 0.05).
2.4. Binding of [3HI YOH to human adipocyte membranes Adipocyte membranes were used directly, or frozen membranes were thawed and rehomoge-
nized with ten pestle strokes in a Potter homogenizer before use. Binding was determined in 10 mM Hepes buffer containing 10 mM MgCI 2. The incubations were carried out much as described for intact adipocytes, except that no ascorbic acid or catechol were added to the membranes. The incubation was usually for 25 min at 20°C under constant shaking. The binding reaction was terminated as described for intact cells on Whatman G F / C filters. The protein concentration of the membrane preparations was usually about 0.2 mg/ml. Non-specific binding was determined as described for intact cells and was 4-10% of the total binding of [3H]YOH.
2.5. Measurements of lipolysis Adipocytes (200 ~ti fat cell suspension, containing about 6 × 104 cells) were incubated in 10 mM Hepes buffer containing the same substance concentrations as used in the binding study to intact adipocytes, including 5% human albumin. Adipocytes were preincubated for 30 rain at 37°C with clonidine or epinephrine. Lipolysis was stimulated by adding theophylline (1 mM) and adenosine deaminase (5/~g/ml) to the incubates and continuing the incubation for a further 60 min at 37°C. Glycerol release was taken as index of lipolysis and was measured by an enzymatic method (Chernick, 1969). In incubations concerning the antilipolytic effect of epinephrine, the fl-adrenoceptor antagonist, propranolol (50 /xM), was added to block the effect of epinephrine on the fl-receptor.
2.6. Data analysis Data are given as mean + 1 S.E.M. All determinations were performed in duplicate or triplicate. In saturation experiments, the number (Bmax) and the affinity (Kd) of the binding sites were analyzed according to the methods of Scatchard (1949) by linear regression. In competition experiments, half-maximal inhibitory concentrations (ECs0) and slope factors (pseudo Hill coefficients) were determined for each competitor by analyzing the data by a non-linear least squares curve-fitting procedure (DeLean et al., 1978). Correlation was calculated by linear regression analysis using the method of least squares.
104
2. 7. Chemicals [3H]Yohimbi'ne (YOH) (spec. radioact. 70-90 Ci/mmol) was purchased from the Radiochemical Center (Amersham, England). Yohimbine, epinephrine, norepinephrine, theophylline and human albumin were from Sigma Chemical Co. (St. Louis, MO, U.S.A.); collagenase from Worthington Biochemical Corp. (Freehold, N J, U.S.A.); phentolamine from Ciba-Geigy (Basel, Switzerland) and clonidine from Boehringer Ingelheim (Mannheim,
West Germany). Prazosin was kindly donated by Pfizer (Brussel, Belgium).
3. Results
3.1. Kinetics of [~H] YOH binding The binding of [3H]YOH (1-2 nM) to intact adipocytes and to adipocyte membranes was rapid at 20°C. Equilibrium was reached within the first
1OO. w
E
O
50'
>-
m
,b
2'0
:;o
,o
TIME I min I
B
A
~,
,oo.
50-
25
0 :i
1o TIME (rain)
Fig. 1. Kinetics of [3H]YOH binding to human adipocytes and adipocyte membranes at 20°C. (A) Time course of [3H]YOH binding to human adipocytes (O) and adipocyte membranes (O). Non-specific binding in presence of clonidine (10 # M) was also determined (D and I1). The curve is based on three independent experiments performed in duplicate. (B) Dissociation curve. Adipocytes or membranes were incubated with [3 H]YOH for 25 min at 20°C. Clonidine (10 /~M) was then added (zero time). Specific [3H]YOH binding was determined at the indi'cated subsequent times of incubation. Binding is expressed as the percentage of the binding before clonidine was added.
105 A
400'
K0 : 2.1nM :
MEMBRANES
"n
0.1, -6
B/F 20C
0.05'
-T>,-
,S Ib [3H] YOHIMBINE (nM)
O
l'S
250
51)0
B (fmol/mcj protein)
B
K0: 6.6nM Bmax : 903fmol 110e cells
INTACT CELLS 800 ¸
o
Z
0.1' 600"
B/F 400" 0.05'
0 )I 200'
(;
1'2
,'8
2'4
[3HI YOHIMBINE (nM)
s6o
,oob
B (fmol / 10e cells )
Fig. 2. Specific binding of [3H]YOH to intact human adipocytes and to adipocyte membranes. Both intact cells and membranes were incubated with increasing concentrations of [3H]YOH at 20°C for 25 rain. The experiments shown are representative of 5 experiments (duplicate) with membranes (A) and 5 experiments with intact cells (B). Right panel: Scatchard plot of the specific [3H]YOH binding derived from the saturation experiments. The slope of the plot (--1/Kd) was determined by linear regression analysis. The total binding capacity (Bma~) was obtained as the intercepts with the x-axis. Hill plots (not shown) gave a slope (n H, determined by linear regression analysis) of 1.03 for intact cells and 1.10 for membranes.
10 min in intact cells a n d within 15 m i n in m e m b r a n e p r e p a r a t i o n s (fig. 1A). Binding was c o n s t a n t up to 45 rain for b o t h p r e p a r a t i o n s at 20°C. N o n - s p e c i f i c b i n d i n g , d e t e r m i n e d with c l o n i d i n e (10 /zM) o r y o h i m b i n e (1 /~M), was c o n s i d e r a b l y
higher in intact a d i p o c y t e s t h a n in isolated m e m b r a n e s (34 + 2% a n d 7 + 0.6% of total binding, respectively). T o t a l a n d non-specific b i n d i n g of [ 3 H ] Y O H were linear with the n u m b e r of i n t a c t a d i p o c y t e s (0.1-1 × 106) a n d with the concentra-
106 100.
•MEMBRANES
E
~
50.
>-
0
loo INIACTC ED ~E
E
~
~
~
~
50"
e yohimbine o I~hentolarnine
\
\
\
• prazosinmine ~. "~ propranoIol bfol 0 10 9 8 7 6 5 4 -log [ADRENERGIC ANTAGONIST]( t,AI ,
//
,
,
,
,
,
,
,
Fig. 3. Adrenoceptor antagonist competition of specifically bound [3H]YOH to adipocyte membranes (A) and intact adipocytes (B). Intact adipocytes and membranes were incubated with [3HIYOH (1.5-2 nM) and various concentrations of adrenoceptor antagonists at 20°C for 25 min. Results are expressed as percent of [3H]YOH specifically bound in the absence of any competing agents. These results are mean values of different numbers of experiments. The number (n), ECs0 value and slope factors are all given in table 2.
tion of m e m b r a n e p r o t e i n s (0.1-1 mg protein), D i s s o c i a t i o n of b i n d i n g at 20°C in the presence of c l o n i d i n e (10 # M ) a d d e d after the e q u i l i b r i u m h a d been reached (25 min) was r a p i d a n d seemed to represent a first-order process for b o t h intact a d i p o c y t e s a n d m e m b r a n e s (fig. 1B).
3.2. Affinity and number of binding sites T h e specific b i n d i n g of [ 3 H ] Y O H was a saturab l e process b o t h in a d i p o c y t e m e m b r a n e s (fig. 2A) a n d in intact a d i p o c y t e s (fig. 2B). In the memb r a n e p r e p a r a t i o n the K d value was 2.1 _+ 0.4 n M a n d the total b i n d i n g c a p a c i t y ( B ~ x ) was 463 + 38 f m o l / m g protein, b o t h values c a l c u l a t e d from the S c a t c h a r d plot. T h e s a m e values in intact a d i p o cytes were 6.6 + 0.8 n M a n d 903 + 139 f m o l / 1 0 6
cells, respectively. The Scatchard plot yielded straight lines (fig. 2) a n d the Hill coefficient ( n u ) was a p p r o x i m a t e l y 1 ( d a t a not shown) for both p r e p a r a t i o n s . If the m e a n fat cell volume was very different in various fat cell suspension, the total b i n d i n g c a p a c i t y expressed per cell n u m b e r (10 n cells) had a large s t a n d a r d deviation. However, if b i n d i n g was expressed in relation to a d i p o c y t e surface, which w o u l d be r e a s o n a b l e since the b i n d ing reaction is m a i n l y a p l a s m a m e m b r a n e event, Bm,,x had a much smaller s t a n d a r d deviation. Thus, Bma, expressed in relation to a d i p o c y t e surface was 342 + 21 f m o l / 1 0 0 cm 2 (mean ___ S.E.M., n = 5). In table 1 the specific b i n d i n g of [-~H]YOH to intact a d i p o c y t e s is c o m p a r e d with the b i n d i n g to m e m b r a n e s o b t a i n e d from an equal a m o u n t of fat cells. Binding in the m e m b r a n e p r e p a r a t i o n was c o n s i d e r a b l y r e d u c e d as c o m p a r e d with that in intact cells. However, these b i n d i n g values are not d i r e c t l y c o m p a r a b l e since K d is different in the two p r e p a r a t i o n s , If these differences in affinity are taken into account, the variation in total b i n d ing would be even m o r e p r o n o u n c e d . Thus, it could be stated that m a n y of the b i n d i n g sites were lost during the m e m b r a n e isolation procedure. These differences in b i n d i n g were i n d e p e n d e n t of the buffer c o m p o s i t i o n since the total and nonspecific b i n d i n g for m e m b r a n e s were similar using the two different i n c u b a t i o n buffers ( d a t a not shown).
TABLE 1 Comparison of specific binding of [3H]YOH to intact adipocytes and to adipocyte membranes. The adipocyte suspension was divided in two parts containing the same amount of fat cells. In one part, binding was determined directly on the intact adipocytes. Adipocyte membranes were obtained from the other part. Binding of [3H]YOH (1.75 nM) to intact adipocytes and membranes were assessed as described under Methods. Total and non-specific binding were determined from each fat cell sample in triplicate. Results are expressed as mean-+ 1 S.E.M. o f 10 6 cells from 4 different fat samples,
[ 3H]YOH, specifically bound (dpm)
Intact adipocytes (106 cells)
Membranes (from 106 cells)
19 700 _+1 702 10056
7856 +_901 39.8 + 1.78%
107
1°°]A
for an c~2-adrenoceptor. The relative order of potency for antagonists was: yohimbine > phentolamine > prazosin > propranolol (fig. 3), and for agonists: clonidine > epinephrine > norepinephrine > isoproterenol (fig. 4). The affinity of agonists was 5-10-fold lower in intact adipocytes than in membranes whereas there was only a 2-3-fold difference between the affinity of antagonists (table 2). The slope factor was about 1 for the antagonist competition in membranes and about 0.95 in intact adipocytes, however, this difference was not statistically significant (P > 0.05, table 2). The slope factor of clonidine was about 0.8 both in membranes and in intact adipocytes, whereas the slope factors for the other agonists were considerably lower in both preparations (table 2). Addition of Gpp(NH)p (100 /~M), a nonhydrolyzable analogue of GTP, and NaC1 (150 mM) to the membrane preparation decreased the affinity of epinephrine but increased the slope factor (fig. 5). The same additions had much less effect on the competition of clonidine in the membranes (fig. 5). Addition of NaC1 (up to 300 mM) and Gpp(NH)p to intact adipocytes did not change the slope factor or the affinity of agonists (data not shown).
100"1B
N >T
• clonidine 0 epinephtine • n0repinephrine
'?-" o u 0
~
~
~:~,~ " ~
~
s
o
p
~
9 6 7 6 5 -log [ADRENERGICAGONISTS] (M)
4
3
Fig. 4. Adrenoceptor agonist competition of specifically bound [3H]YOH to adipocyte membranes (A) and intact h u m a n adipocytes (B). The incubations were performed as described in fig. 3.
3.3. Specificity for agonists and antagonist
3.4. Antilipolytic effects of ae-agonists
Adrenoceptor antagonists and agonists competed for [3H]YOH binding sites in both membranes and intact adipocytes in the order expected
Clonidine and epinephrine (in the presence of propranolol) had only slight inhibitory effects on basal lipolysis in human adipocytes (data not
TABLE 2 Adrenoceptor antagonists and agonists competition constants in adipocyte membranes and in intact adipocytes. Comparison with the antilipolytic effect in intact cells, Intact adipocytes and adipocyte membranes were incubated at 20°C for 25 rain with [3H]YOH in the absence and presence of various competing agents. Competitor
Membranes
Intact adipocytes
N ~
ECso b
(-)Epinephrine (-)Norepinephrine Clonidine
5 3 8
659 1403 40
Yohimbine Phentolamine Prazosin
6 4 3
2.3 +_ 0.36 5.2+_ 0.90 2339 _+846
a Values
+ 79 +_261 +- 3.8
n H~
N "'
ECso b
0.53+-0.08 0.56+_0.10 0.86+_0.03
7 3 7
5997 15588 198
1.12 + 0.03 0.99+_0.10 1.10+0.16
6 3 3
+- 542 +_ 889 +_ 31
7.5 +_ 1.1 17.1+_ 4.5 9983 +__2225
Antilipolytic effect n H~
N ~
iCso d
0.49+-0.04 0.53_+0.13 0.79+_0.06
6
661+_51
6
52+_ 5,1
0.96 +_0.06 0.93+0.14 0.98+-0.16
are the mean + S.E.M. of (N) experiments, b ECs0 ' concentration of competing agent producing 50% inhibition of specific [3H]YOH binding (nM). c n n ' slope factor (pseudo Hill coefficient). Determined as described in Methods. a iC5o ' concentration of agonists producing 50% inhibition of the stimulated lipolysis (nM).
108 300 I00'
g
c
ff
~
200
50.
I00 c
:z:
A Bepinephrfne
+ Nacl+ Gpp(NH)p
~
~:~
..,.( -log [AGONISTS]
(M )
Fig. 5. Effect of Gpp(NH)p and NaCI on competition of epinephrine and clonidine for [ 3H]YOH binding sites in human adipocyte membranes. Adipocyte membranes were incubated with 2 nM [3H]YOH with increasing concentrations of epinephrine and clonidine in both the absence and presence of Gpp(NH)p (100 /~M) and NaC1 (150 mM). ECs0 of epinephrine and clonidine were 668 nM and 39 nM, respectively, In the presence of Gpp(NH)p and NaC1 these values were 4228 nM and 72 nM. The slope factors of epinephrine and clonidine were 0.54 and 0.83 respectively, and in the presence of Gpp(NH)p and NaCI these values were 0.86 and 0.92, respectively. The data are the means of 3 (clonidine) and 4 (epinephrine) different experiments performed in duplicate and run in parallel.
shown). The action of these a2-receptor agonists was pronounced when lipolysis was stimulated by a combination of theophylline (1 mM) and adenosine deaminase (5 #g/ml). This combination increased glycerol release from 260 + 21 nmol/106 cells per 90 min to 515 + 56 nmol/106 cells per 90 min. At maximally effective concentrations, clonidine (10 /~M) and epinephrine (10 #M) inhibited lipolysis by about 60% (fig. 6). However, the sensitivity expressed as IC50 was much higher for clonidine than for epinephrine (57 + 9 nM and 663 + 59 nM, respectively).
3.5. Correlation between binding and antilipolysis To evaluate if the binding data were related to the biological effect (antilipolysis), multiple correlations were performed between [ 3H]YOH binding data in both intact adipocytes and membrane fractions and the antilipolytic effect of clonidine (fig. 7). The total binding capacity (Bma x) for [3H]YOH binding in membranes was positively correlated to
~ 100
o]
-log [AGONISTS] (MI
Fig. 6. Antilipolytic effects of epinephrine and clonidine in human adipocytes. (A) Adipocytes were incubated with epinephrine (e) and clonidine (O) in the indicated concentrations at 37°C for 90 rain. For the last 60 rain lipolysis was stimulated with theophylline (1 mM) and adenosine deaminase (5/~g/ml). In experiments with epinephrine, propranolol (50 /LM) was included in the medium. Lipolysis is expressed as the stimulated glycerol release minus basal glycerol release, mean_+ S.E.M. for 7 independent experiments for both clonidine and epinephrine. (B) Data for the antilipolytic effect of epinephrine and clonidine are given as percent of the response to these agonists in maximally effective concentrations.
the maximal antilipolytic effect of clonidine (r = 0.70, P < 0.001, n = 20, fig. 7A). Furthermore, the binding affinity (Kd) of YOH in membranes was positively correlated to the sensitivity (IC50) of clonidine in inhibiting the stimulated lipolysis (r = 0.65, P < 0.01, n = 20, fig 7C). There was no significant correlation for intact adipocytes between Bmax and the maximal antilipolytic effect of clonidine when Bmax w a s expressed in relation to adipocyte number (106 cells). However, a significant and positive correlation was found (r = 0.79, P <0.001, n = 15, fig. 7B) if YOH binding to adipocytes was expressed in relation to the surface of the adipocytes (100 cm2). No correlation was demonstrated in intact adipocytes between the binding affinity and the IC50 value of the antilipolytic effect of clonidine (r = - 0.04, fig. 7D).
109 100-
100- B
7-
q-
~g
~
,w,~
~g
50,
so-
~.=_. CELLS r = 0.79 p < O.001
MEMBRANES r = 0.70 p < 0.001
~
o
0
IO~O
56o
t60
2~0
360
460
s~o
B max ( fmol / 100 cna=)
Bma x ( fnaol / nag protein )
150"! D
7., o,. ._%
100
..o •
~
~-=
• eee
so
•
~
so
>_ •
MEMBRANES r : o.65 p < 0.01
== CELLS
t-
r=
-0.04
NS 0
AFFINITY OF BINDING (K d} (nM)
AFFINITY OF BINDING ( Kd) (riM)
Fig. 7. Relationship between YOH binding and antilipolytic effect of clonidine. (A) Relationship between BmaX in membranes and maximal antilipolytic effect of clonidine. (B) Relationship as in A but obtained with intact adipocytes. (C) Relationship between K d in membranes and ICs0 of clonidine. (D) Relationship as in C but obtained with intact adipocytes. Total binding capacity (Bmax) and binding affinity (Kd) of YOH binding to membranes and intact adipocytes were calculated from Scatchard plots. The antilipolytic effect of clonidine is expressed as percent inhibition of the stimulated lipolysis without clonidine. The relationships were examined by regression analysis (intact adipocytes, n = 15; membrane preparations, n = 20).
4. D i s c u s s i o n
We have now demonstrated that az-adrenoceptors can be measured readily in both intact h u m a n adipocytes and adipocyte membranes. The binding of [ 3 H ] Y O H was saturable, specific and reversible both in intact cells and in membranes. The order of potency for agonists and antagonists for competition for [ 3 H ] Y O H binding was similar in intact adipocytes and in m e m b r a n e preparations and this order of potency was typical for the a2-receptor (Berthelsen and Pettinger, 1977). The total binding capacity of [ 3 H ] Y O H (B,,ax) in intact adipocytes
was 903 f m o l / 1 0 6 cells and the equilibrium binding constant (K d) was 6.6 nM. The values were 463 f m o l / m g protein and 2.1 nM, respectively, in the m e m b r a n e preparation. The Scatchard plot yielded straight lines for both preparations, indicating that [3H]YOH binds to a homogenous class of receptors in both preparations. All these data suggest that [3H]YOH binds to the same a2-adrenoceptors in intact adipocytes and in adipocyte membranes. However, some minor differences exist between the binding characteristics in the two preparations. [ 3 H ] Y O H (adrenoceptor antagonist) binds with lower affinity in intact adipocytes (Kd:
110
6.6 nM vs. 2.1 nM) and adrenoceptor agonists bind with 5-10 times lower affinity in intact cells. These differences could be explained by an enhanced degradation of the agonists/antagonists by the intact cells a n d / o r the fact that intact cells contain many different ions and other substances as nucleotides (see below) which could interfere with the hormone-receptor interaction. Nonspecific binding was considerably higher in intact adipocytes (34%) than in membranes (7%). These differences indicate that [~H]YOH presumably may bind to some sites of a non-receptor nature in intact adipocytes. These findings are in accordance with previous results obtained with intact cells using [3 H]dihydroalprenolol to identify ~8-adrenoceptors (Staehelin et al., 1982; Insel and Michele, 1979). Presumably, the lipophilic nature of yohimbine results in binding, especially intracellularly, to lipid structures which are not 'true' receptors. The finding that yohimbine ( > l >M) could inhibit more [3H]YOH binding in intact adipocytes than clonidine is also consistent with this hypothesis. The existence of a2-adrenoceptors in human adipocyte membranes had previously been shown with the ~e-receptor agonist, [)H]clonidine (Berlan and Lafontan, 1980) or the c~2-receptor antagonist, [)H]YOH (Tharp et al., 1981; Lafontan et al., 1983). The K d values obtained in these latter studies (0.8-3 nM) are in good agreement with the K d values obtained with membranes in the present study (2.1 nM). Furthermore, the total binding capacity was in the range described by others (400-600 f m o l / m g protein). Many [3H]YOH binding sites seemed to be lost during the membrane isolation procedure (table 1). Destruction of the receptors during the isolation procedure is unlikely since the characteristics of the receptor binding (affinity for agonists and antagonists) in membrane preparations were rather similar with the data obtained with intact adipocytes (table 2) and similar with data obtained from other studies concerning human adipocyte membranes (Tharp et al., 1981; Lafontan et al., 1983). Thus, the reduced binding in membranes may have been due to incomplete recovery of plasma membrane proteins a n d / o r loss of internalized receptors which may be labeled by [3H]YOH in intact adipocytes.
The [)H]YOH binding sites were homogeneous towards antagonists, as indicated by the Hill coefficients (near unity), in both membranes and intact adipocytes (table 2). However, agonist binding had different characteristics on membranes and on intact cells. The slope factor of agonist binding in these two preparations was considerably below 1, indicating that agonists bound to the c~2-receptor with different affinities. The agonist binding data could be explained by binding of agonists to the receptor in a high affinity state and a low affinity state as shown in membranes from platelets and liver cells (Hoffman et al., 1980a). In the presence of Gpp(NH)p and Na + the high affinity state of the receptor in membranes is converted to a low affinity state (Hoffman et al., 1980a,b; Gonzalez et al., 1981). These findings are in accordance with our data from membranes since the affinity of epinephrine decreased and the slope factor increased in the presence of Gpp(NH)p and Na + (fig. 6). However, even though the slope factor increased significantly, it was still less than 1, suggesting incomplete disappearance of the binding heterogeneity. The same additions bad much less effect on the clonidine competition curve (fig. 6), presumably because clonidine is only a partial agonist. Addition of Gpp(NH)p and Na" to the membrane preparation increased the ECs0 values of epinephrine about 10 times (to 4120 nM) which is close to the EC50 value of epinephrine in intact adipocytes (5997 nM). These findings suggest that nearly all the az-receptors in intact adipocytes exist in the low affinity state for agonist binding, which also would seem likely in view of the high content of Na + and G T P in intact cells. However, the agonist competition curves in intact cells are, as was mentioned, shallow suggesting that the binding of agonists to the a2-receptor in intact cells could not be explained only by binding of agonists to a homogeneous class of low affinity receptors but other mechanisms must also be acting, e.g. different affinities of the receptor or negative cooperativity. One can only speculate about the exact relationship between radioligand binding studies, the adenylate cyclase complex and the biological effect. However, we have now found that the ECs0 values of clonidine and epinephrine obtained from competition curves in adipocyte membranes corre-
111
lated very well, and were in the same range as the ICs0 values of the same agonists on the biological effect (antilipolysis, table 2). Direct correlation studies between [3H]YOH binding (at 20°C) and antilipolytic effect of clonidine (at 37°C) showed a positive relationship between Bmax in adipocyte membranes and the maximal antilipolytic effect of clonidine (fig. 7). In membranes, a positive correlation was also demonstrated between the binding affinity and the sensitivity of the clonidine-induced antilipolysis. No correlation was detected between the binding affinity and the ICs0 values of clonidine with intact adipocytes. This latter finding indicates that the affinity of antagonist binding in intact adipocytes is not directly related to the sensitivity of the antilipolytic effect of agonists. This discrepancy between c~2-receptor antagonist binding and the biological response is unexplainable at the present moment. However, the maximal receptor number identified by [3H]YOH in intact adipocytes was positively correlated to the maximal antilipolytic effect of clonidine. In conclusion it is demonstrated that a2-rece ptors could easily be studied in intact adipocytes and in adipocyte membranes. Adrenoceptor antagonists, which are without intrinsic activity, bound to a homogeneous class of receptors with high affinity in intact cells and in membranes. In the membrane preparation, agonists bound to the t~z-receptor with different affinities and binding was dependent on the presence of GTP and Na +. In intact adipocytes, agonists bound to the receptor in a heterogeneous manner with shallow competition curves. The study demonstrated that the [3H]YOH binding data obtained with the membrane fractions correlated best with the biological effect mediated by these receptors.
Acknowledgements We appreciate the skillful technical assistance of J. Soholt, T. Skrumsager, L. Blak and P. Jorgensen. The study was supported by grants from Aarhus University, Landsforeningen for Sukkersyge, Nordisk Insulinfond, P. Carl Petersens Fond and the Danish Medical Research Council.
References Berlan, M. and M. Lafontan, 1980, Identification of t~2-adrenergic receptors in human fat cell membranes by [3H]clonidine binding, European J. Pharmacol. 67, 481. Berlan, M. and M. Lafontan, 1982, The a2-adrenergic receptor of human fat cells: comparative study of a2-adrenergic radioligand binding and biological response, J. Physiol. (Paris) 78, 279. Berthelsen, S. and W.A. Pettinger, 1977, A functional basis for classification of c~-adrenergic receptors, Life Sci. 21,596. Burns, T.W., P.E. Langley, B.E. Terry, D.B. Byland, B.B. Hoffman, M.D. Tharp, R.J. Lefkowitz, J.A. Garcia-Sainz and J.N. Fain, 1981, Pharmacological characterizations of adrenergic receptors in human adipocytes, J. Clin. Invest. 67, 467. Burns, T.W., B.E. Terry, P.E. Langly and G.A. Robinson, 1980, Role of cyclic AMP in human adipose tissue lipolysis, Adv. Cycl. Nucl. Res. 12, 329. Chernick, S.S., 1969, Determination of glycerol in acyl glycerols, Methods Enzymol. 14, 627. DeLean, A., P.J. Munson and D. Rodbard, 1978, Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay and physiological dose-response curves, Am. J. Physiol. 235, E 97. Dulis, B.H. and I.B. Wilson, 1980, The fl-adrenergic receptor of live human polymorphonuclear leukocytes, J. Biol. Chem. 255, 1043. Fain, J.N. and J.A. Garcia-Sainz, 1983, Adrenergic regulation of adipocytes metabolism, J. Lipid Res. 24, 945. Gonzalez, J.:L., C. Carpene, M. Berlan and M. Lafontan, 1981, Discrepancy in the regulatory effects of sodium and guanine nucleotides on adrenaline and clonidine binding to a 2adrenoceptors in human fat cell membranes, European J. Pharmacol. 76, 289. Hoffman, B.B., T. Michel, D. Mullikin-Kilpatrick, R.J. Lefkowitz, M.E.M. Tolbert, H. Gilman and J.N. Fain, 1980a, Agonist versus antagonist binding to alpha-adrenergic receptors, Proc. Natl. Acad. Sci. U.S.A. 77, 4569. Hoffman, B.B., D. Mullikin-Kilpatrick and R.J. Lefkowitz, 1980b, Heterogeneity of radioligand binding to a-adrenergic receptors: analysis of guanine nucleotide regulation of agonist binding in relation to receptor subtypes, J. Biol. Chem. 255, 4645. Insel, P.A. and S. Michele, 1979, Temperature-dependent changes in binding to fl-adrenergic receptors in intact $49 lymphoma cells, J. Biol. Chem. 254, 6554. Kather, H., J. Pries, V. Schneider and B. Simon, 1980, Inhibition of human fat cell adenylate cyclase mediated via alpha-adrenoceptors, European J. Clin. Invest. 10, 345. Kather, H. and S. Simon, 1981, Adrenoceptor of the alpha 2-subtype mediating inhibition of the human fat cell adenylate cyclase, European J. Clin. Invest. 11, 111. Lafontan, M., M. Berlan and A. Villeneuve, 1983, Prependance of a z - over fll-adrenergic receptor sites in human fat cells is not predictive of the lipolytic effect of physiological catecholamines, J. Lipid Res. 24, 429.
1i2 Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265. Mukherjee, C, and R.J. Lefkowitz, 1976, Direct studies of /3-adrenergic receptors in intact frog erythrocytes, Life Sci. 19, 1897. Pedersen, O., E. Hjollund, H. Beck-Nielsen, O. Sonne and J. Gliemann, 1981, Insulin receptor binding and receptor mediated insulin degradation in human adipocytes, Diabetologia 20, 636. Pittman, R.N. and P.B. Molinoff, 1980, Interactions of agonists and antagonists with ,8-adrenergic receptors on intact L6 muscle cells, J. Cycl. Nucl. Res. 6, 421. Richelsen, B., E.F. Eriksen, H. Beck-Nielsen and O. Pedersen, 1984, Prostaglandin E 2 receptor binding and action in human fat cells, J. Clin. Endocrinol. Metab. 59, 7. Rodbell, M., 1980, The role of hormone receptors and GTPregulatory proteins in membrane transduction, Nature 284, 17.
Scatchard, G., 1949, The attractions of protein for small molecules and ions, Ann. N.Y. Acad. Sci. 51,660. Sj6str6m, L., P. Bj6rntorp and J. Vrhna, 1971. Microscopic fat ceils size measurements of frozen-cut adipose tissue in comparison with automatic determination of osiumfixed fat cells, J. Lipid Res. 12, 521. Staehelin, M., P. Simons, K. Jaeggi and N. Wagger, 1982, CGP-12177, a hydrophilic fl-adrenergic receptor radioligand reveals high affinity binding of agonists to intact cells, J. Biol. Chem. 258, 3496. Tharp, M.D., B.B. Hoffman and R.J. Lefkowitz, 1981, aAdrenergic receptors in human adipocyte membranes: direct determination by (3H)-yohimbine binding, J. Clin, Endocrinol. Metab. 52, 709.
101
Elsevier
a2-ADRENERGIC B I N D I N G AND A C T I O N IN H U M A N A D I P O C Y T E S . C O M P A R I S O N B E T W E E N B I N D I N G T O P L A S M A M E M B R A N E P R E P A R A T I O N S AND T O I N T A C T A D I P O C Y T E S B J ~ R N RICHELSEN * and O L U F PEDERSEN
Medical Department I11, Division of Endocrinology and Metabolism, Aarhus County Hospital, Aarhus, Denmark Received 23 August 1985, accepted 1 October 1985
B. RICHELSEN and O. PEDERSEN, a2-Adrenergic binding and action in human adipocytes. Comparison between binding to plasma membrane preparations and to intact adipocytes, European J. Pharmacol. 119 (1985) 101-112. Binding of the a/-adrenoceptor antagonist [3H]yohimbine was demonstrated on intact human adipocytes and on human adipocyte membranes. Specific binding was rapid, reversible, saturable and of high affinity in both preparations. [3 H]Yohimbine binding was inhibited by various adrenergic agents in a manner which suggests that the labeled sites probably represent the a2-receptor both in intact adipocytes and in the membrane fraction. In adipocyte membranes the maximal binding capacity (Bmax) was 463 +_ 38 fmol/mg protein and the binding was of high affinity with a K d of 2.1 _+0.4 nM. In intact human adipocytes, Bmax was 903 + 139 fmol/106 cells (or 342 + 21 fmol/100 cm2) and the binding affinity was 6.6 + 0.8 nM. Adrenergic antagonists bound to a homogeneous class of receptors (linear Scatchard plots) both in isolated membranes and in intact adipocytes. However, agonist binding was heterogeneous in both preparations. The affinity of agonist binding was 5-10 times higher in membranes than in intact adipocytes. The physiological relevance of the binding data was evaluated by correlating the binding of yohimbine with the antilipolytic effect of clonidine. A positive and significant correlation was found between BmaX and maximal antilipolytic effect of clonidine ( r = 0.70) in membranes. Furthermore, the binding affinity (Kd) was positively correlated to the sensitivity (ICs0) of the clonidine-induced antilipolysis (r=0.65). A positive and significant correlation was also found in intact adipocytes between Bmax and the maximal antilipolytic effect of clonidine (r = 0.79). However, there was no significant correlation between K d in intact adipocytes and the ICs0 of clonidine. It is concluded that both intact human adipocytes and membrane fractions are useful models to investigate the properties and regulation of a/-adrenoceptors. However, it appears from our correlation studies that the binding data obtained with the membranes are the best related to the physiological effects mediated by a2-receptors in human adipocytes. a2-Adrenoceptors
Intact human adipocytes
Adipocyte membranes
1. Introduction It is now accepted that human adipocytes possess both fl- and a2-adrenoceptors in the plasma membrane (Burns et al., 1981). Catecholamines stimulate both kinds of receptors which cause activation of the adenylate cyclase complex through the fl-receptor, and inhibition of the same complex through the az-receptor (Burns et al., 1981; Fain and Garcia-Sainz, 1983; Kather and * To whom all correspondence should be addressed: Medical Department III, Tage Hansens Gade, Aarhus amtssygehus, DK-8000 Aarhus C, Denmark. 0014-2999/85/$03.30 © 1985 Elsevier Science Publishers B.V.
[3H]Yohimbine
Antilipolyse
Simon, 1981). This a2-receptor-induced inhibition of adenylate cyclase in human adipocytes is presumably mediated through the guanine nucleotide binding regulatory component (N i) (Rodbell, 1980; Fain and Garcia-Sainz, 1983). In adipocytes, an increased concentration of cyclic A M P stimulates lipolysis whereas a decreased concentration of cyclic A M P inhibits lipolysis (Bums et al., 1980; Kather et al., 1980). Thus, it seems to be the balance between the fl- and az-adrenoceptors that controls the final lipolytic response induced by catecholamines in human adipocytes. Furthermore, human adipocytes also possess al-adrenoceptors. These receptors seem to be independent
102
of the adenylate cyclase system. Activation of these a~-receptors results in increased turnover of phosphatidylinositoland phosphatidic acid in human adipocytes (Burns et al., 1981; Fain and GarciaSainz, 1983). Recently, c~2-adrenoceptors have been demonstrated directly on human adipocyte membranes by using the radioactive, t~2-receptor antagonist, yohimbine ([3H]YOH) (Tharp et al., 1981; Lafontan et al., 1983; Berlan and Lafontan, 1982). However, no radioligand studies concerning the c~2-receptors have been performed using intact human adipocytes *, even though cell disruption could induce changes in receptor properties and regulation. There have been detailed analyses of the properties of /3-adrenoceptors in intact cells (Mukherjee and Lefkowitz, 1976; Pittman and Molinoff, 1980; Dulis and Wilson, 1980). The main findings from these studies were that the properties of agonist binding to the fl-receptor were different from the binding characteristics of membrane preparations. Generally, the affinities of agonists for the receptors were much weaker in intact cells than in membranes, whereas affinities of antagonist binding were similar in these two preparations. We have now characterized the binding of [-~H]YOH to intact human adipocytes and compared these data with results obtained from adipocyte membrane preparations of the same system. Furthermore, these binding data were correlated with the biological effect (antilipolysis) mediated by these receptors.
2. Materials and methods
2.1. Isolation and adipocytes Subcutaneous adipose tissue was obtained from patients undergoing abdominal surgery. The adipose tissue was collected at the beginning of the surgical procedure. Patients fasted overnight before tissue removal and had no identified metabolic or endocrine disorders. However, some were * A report by Engfeldt et al. has appeared (Scand. J. Clin. Invest. 1983, 43, 207).
obese ( > 20% of ideal body weight). There was no selection according to sex or age of the donors. Isolated adipocytes were prepared and cell concentration calculated as previously described (Pedersen et al., 1981; Richelsen et al., 1984). Briefly, the adipocytes were isolated by collagenase digestion of fat tissue fragments for 60 min at 37°C in 10 mM Hepes buffer containing 0.5 m g / m l collagenase and 2.5% bovine serum albumin (BSA). The adipocytes were then filtered on a silk screen and washed three times. The cell concentration in a given suspension was calculated as follows: the diameters of adipocytes were measured (SjOstr0m et al., 1971; Pedersen et al., 1981). Surface area (~rD 2) and volume (7rD3/6) were calculated for each cell diameter measured (about 200 individual cells per fat tissue preparation). The volume fraction of adipocytes was measured in a hematocrit capillary tube. The concentration of cells in the adipocyte suspensions was obtained as the adipocyte volume fraction divided by the mean adipocyte volume,
2.2. Preparation of human adipocyte membranes Isolated adipocytes were washed three times in a lysing medium (5 mM Hepes, 2.5 mM MgCI 2, 1 mM KHCO3, pH 7.6) then centrifuged of 3000 × g for 4 min. All the adipocytes were broken after this procedure. The adipocyte ghosts and crude broken membranes were then centrifuged at 24 000 × g for 20 min at 4°C. The pellet was washed in 10 mM Hepes buffer containing 1 mM EDTA and again centrifuged at the same speed. Finally, the membrane preparation was resuspended in 10 mM Hepes buffer containing 10 mM MgCI 2 and used immediately in the binding assay or frozen at - 8 0 ° C . The protein concentration of the membrane was measured (Lowry et al., 1951) with BSA as the standard.
2.3. Binding of [3H]YOH to intact human adipocytes Isolated adipocytes were washed four times by centrifugation to remove albumin from the cells. The adipocytes were then resuspended in a buffer containing the following solute concentrations in
103 mM: Hepes 10, NaC1 135, KC1 4.8, MgSO 4 1.7, CaC12 2.5, NaH2PO 4 0.2, N a 2 H P O 4 1.0, glucose 5.0 and pH 7.4. The binding study was performed in a final volume of 325 /tl. The incubation mixture consisted of 250 ~1 adipocytes (about 105 cells/ml), 25 #1 buffer or unlabeled ligand for competition, and 25 btl [3H]YOH (0.3-15 nM). There was also added 25/~1 of a mixture of ascorbic acid (0.5 mM) and catechol (0.3 mM) to diminish the oxidation of adrenoceptor agonists. For competition studies [3H]YOH was usually added to reach a final concentration of 1-2 nM. The binding incubation was carried out for 25 min at 20°C without shaking. The incubation was terminated by adding 5 ml of ice-cold buffer to each tube. The incubation mixture was then immediately vacuum-filtered through Whatman G F / C glass fiber filters placed on a Millipore manifold. The filters were washed twice with 10 ml ice-cold buffer. After the filters had been dried for 3 h they were placed in 5 ml scintillation fluid (Lumagel, Lumac BV, Netherlands), and counted in a Rackbeta counter (LKB, Finland). Non-specific binding was determined by adding unlabeled yohimbine (1/~M) or clonidine (10/~M) to the incubates. At a tracer concentration of 1-2 nM non-specific binding was 20-40% of the total binding to the adipocytes. Yohimbine and clonidine gave equivalent results at these concentrations. However, unlabeled yohimbine at even higher concentrations further inhibited some [3H]YOH binding in intact adipocytes. Presumably, this further inhibition of binding might have been at non-specific binding sites since clonidine from higher concentrations was unable to reproduce the effect. Specific binding was taken as the difference between total and non-specific binding. It was shown microscopically that the adipocytes remained intact after incubation for 50 min in this buffer, with unaltered mean fat cell volume before and after the incubation (422 _+ 22 pl and 406 + 24 pl, respectively, P > 0.05).
2.4. Binding of [3HI YOH to human adipocyte membranes Adipocyte membranes were used directly, or frozen membranes were thawed and rehomoge-
nized with ten pestle strokes in a Potter homogenizer before use. Binding was determined in 10 mM Hepes buffer containing 10 mM MgCI 2. The incubations were carried out much as described for intact adipocytes, except that no ascorbic acid or catechol were added to the membranes. The incubation was usually for 25 min at 20°C under constant shaking. The binding reaction was terminated as described for intact cells on Whatman G F / C filters. The protein concentration of the membrane preparations was usually about 0.2 mg/ml. Non-specific binding was determined as described for intact cells and was 4-10% of the total binding of [3H]YOH.
2.5. Measurements of lipolysis Adipocytes (200 ~ti fat cell suspension, containing about 6 × 104 cells) were incubated in 10 mM Hepes buffer containing the same substance concentrations as used in the binding study to intact adipocytes, including 5% human albumin. Adipocytes were preincubated for 30 rain at 37°C with clonidine or epinephrine. Lipolysis was stimulated by adding theophylline (1 mM) and adenosine deaminase (5/~g/ml) to the incubates and continuing the incubation for a further 60 min at 37°C. Glycerol release was taken as index of lipolysis and was measured by an enzymatic method (Chernick, 1969). In incubations concerning the antilipolytic effect of epinephrine, the fl-adrenoceptor antagonist, propranolol (50 /xM), was added to block the effect of epinephrine on the fl-receptor.
2.6. Data analysis Data are given as mean + 1 S.E.M. All determinations were performed in duplicate or triplicate. In saturation experiments, the number (Bmax) and the affinity (Kd) of the binding sites were analyzed according to the methods of Scatchard (1949) by linear regression. In competition experiments, half-maximal inhibitory concentrations (ECs0) and slope factors (pseudo Hill coefficients) were determined for each competitor by analyzing the data by a non-linear least squares curve-fitting procedure (DeLean et al., 1978). Correlation was calculated by linear regression analysis using the method of least squares.
104
2. 7. Chemicals [3H]Yohimbi'ne (YOH) (spec. radioact. 70-90 Ci/mmol) was purchased from the Radiochemical Center (Amersham, England). Yohimbine, epinephrine, norepinephrine, theophylline and human albumin were from Sigma Chemical Co. (St. Louis, MO, U.S.A.); collagenase from Worthington Biochemical Corp. (Freehold, N J, U.S.A.); phentolamine from Ciba-Geigy (Basel, Switzerland) and clonidine from Boehringer Ingelheim (Mannheim,
West Germany). Prazosin was kindly donated by Pfizer (Brussel, Belgium).
3. Results
3.1. Kinetics of [~H] YOH binding The binding of [3H]YOH (1-2 nM) to intact adipocytes and to adipocyte membranes was rapid at 20°C. Equilibrium was reached within the first
1OO. w
E
O
50'
>-
m
,b
2'0
:;o
,o
TIME I min I
B
A
~,
,oo.
50-
25
0 :i
1o TIME (rain)
Fig. 1. Kinetics of [3H]YOH binding to human adipocytes and adipocyte membranes at 20°C. (A) Time course of [3H]YOH binding to human adipocytes (O) and adipocyte membranes (O). Non-specific binding in presence of clonidine (10 # M) was also determined (D and I1). The curve is based on three independent experiments performed in duplicate. (B) Dissociation curve. Adipocytes or membranes were incubated with [3 H]YOH for 25 min at 20°C. Clonidine (10 /~M) was then added (zero time). Specific [3H]YOH binding was determined at the indi'cated subsequent times of incubation. Binding is expressed as the percentage of the binding before clonidine was added.
105 A
400'
K0 : 2.1nM :
MEMBRANES
"n
0.1, -6
B/F 20C
0.05'
-T>,-
,S Ib [3H] YOHIMBINE (nM)
O
l'S
250
51)0
B (fmol/mcj protein)
B
K0: 6.6nM Bmax : 903fmol 110e cells
INTACT CELLS 800 ¸
o
Z
0.1' 600"
B/F 400" 0.05'
0 )I 200'
(;
1'2
,'8
2'4
[3HI YOHIMBINE (nM)
s6o
,oob
B (fmol / 10e cells )
Fig. 2. Specific binding of [3H]YOH to intact human adipocytes and to adipocyte membranes. Both intact cells and membranes were incubated with increasing concentrations of [3H]YOH at 20°C for 25 rain. The experiments shown are representative of 5 experiments (duplicate) with membranes (A) and 5 experiments with intact cells (B). Right panel: Scatchard plot of the specific [3H]YOH binding derived from the saturation experiments. The slope of the plot (--1/Kd) was determined by linear regression analysis. The total binding capacity (Bma~) was obtained as the intercepts with the x-axis. Hill plots (not shown) gave a slope (n H, determined by linear regression analysis) of 1.03 for intact cells and 1.10 for membranes.
10 min in intact cells a n d within 15 m i n in m e m b r a n e p r e p a r a t i o n s (fig. 1A). Binding was c o n s t a n t up to 45 rain for b o t h p r e p a r a t i o n s at 20°C. N o n - s p e c i f i c b i n d i n g , d e t e r m i n e d with c l o n i d i n e (10 /zM) o r y o h i m b i n e (1 /~M), was c o n s i d e r a b l y
higher in intact a d i p o c y t e s t h a n in isolated m e m b r a n e s (34 + 2% a n d 7 + 0.6% of total binding, respectively). T o t a l a n d non-specific b i n d i n g of [ 3 H ] Y O H were linear with the n u m b e r of i n t a c t a d i p o c y t e s (0.1-1 × 106) a n d with the concentra-
106 100.
•MEMBRANES
E
~
50.
>-
0
loo INIACTC ED ~E
E
~
~
~
~
50"
e yohimbine o I~hentolarnine
\
\
\
• prazosinmine ~. "~ propranoIol bfol 0 10 9 8 7 6 5 4 -log [ADRENERGIC ANTAGONIST]( t,AI ,
//
,
,
,
,
,
,
,
Fig. 3. Adrenoceptor antagonist competition of specifically bound [3H]YOH to adipocyte membranes (A) and intact adipocytes (B). Intact adipocytes and membranes were incubated with [3HIYOH (1.5-2 nM) and various concentrations of adrenoceptor antagonists at 20°C for 25 min. Results are expressed as percent of [3H]YOH specifically bound in the absence of any competing agents. These results are mean values of different numbers of experiments. The number (n), ECs0 value and slope factors are all given in table 2.
tion of m e m b r a n e p r o t e i n s (0.1-1 mg protein), D i s s o c i a t i o n of b i n d i n g at 20°C in the presence of c l o n i d i n e (10 # M ) a d d e d after the e q u i l i b r i u m h a d been reached (25 min) was r a p i d a n d seemed to represent a first-order process for b o t h intact a d i p o c y t e s a n d m e m b r a n e s (fig. 1B).
3.2. Affinity and number of binding sites T h e specific b i n d i n g of [ 3 H ] Y O H was a saturab l e process b o t h in a d i p o c y t e m e m b r a n e s (fig. 2A) a n d in intact a d i p o c y t e s (fig. 2B). In the memb r a n e p r e p a r a t i o n the K d value was 2.1 _+ 0.4 n M a n d the total b i n d i n g c a p a c i t y ( B ~ x ) was 463 + 38 f m o l / m g protein, b o t h values c a l c u l a t e d from the S c a t c h a r d plot. T h e s a m e values in intact a d i p o cytes were 6.6 + 0.8 n M a n d 903 + 139 f m o l / 1 0 6
cells, respectively. The Scatchard plot yielded straight lines (fig. 2) a n d the Hill coefficient ( n u ) was a p p r o x i m a t e l y 1 ( d a t a not shown) for both p r e p a r a t i o n s . If the m e a n fat cell volume was very different in various fat cell suspension, the total b i n d i n g c a p a c i t y expressed per cell n u m b e r (10 n cells) had a large s t a n d a r d deviation. However, if b i n d i n g was expressed in relation to a d i p o c y t e surface, which w o u l d be r e a s o n a b l e since the b i n d ing reaction is m a i n l y a p l a s m a m e m b r a n e event, Bm,,x had a much smaller s t a n d a r d deviation. Thus, Bma, expressed in relation to a d i p o c y t e surface was 342 + 21 f m o l / 1 0 0 cm 2 (mean ___ S.E.M., n = 5). In table 1 the specific b i n d i n g of [-~H]YOH to intact a d i p o c y t e s is c o m p a r e d with the b i n d i n g to m e m b r a n e s o b t a i n e d from an equal a m o u n t of fat cells. Binding in the m e m b r a n e p r e p a r a t i o n was c o n s i d e r a b l y r e d u c e d as c o m p a r e d with that in intact cells. However, these b i n d i n g values are not d i r e c t l y c o m p a r a b l e since K d is different in the two p r e p a r a t i o n s , If these differences in affinity are taken into account, the variation in total b i n d ing would be even m o r e p r o n o u n c e d . Thus, it could be stated that m a n y of the b i n d i n g sites were lost during the m e m b r a n e isolation procedure. These differences in b i n d i n g were i n d e p e n d e n t of the buffer c o m p o s i t i o n since the total and nonspecific b i n d i n g for m e m b r a n e s were similar using the two different i n c u b a t i o n buffers ( d a t a not shown).
TABLE 1 Comparison of specific binding of [3H]YOH to intact adipocytes and to adipocyte membranes. The adipocyte suspension was divided in two parts containing the same amount of fat cells. In one part, binding was determined directly on the intact adipocytes. Adipocyte membranes were obtained from the other part. Binding of [3H]YOH (1.75 nM) to intact adipocytes and membranes were assessed as described under Methods. Total and non-specific binding were determined from each fat cell sample in triplicate. Results are expressed as mean-+ 1 S.E.M. o f 10 6 cells from 4 different fat samples,
[ 3H]YOH, specifically bound (dpm)
Intact adipocytes (106 cells)
Membranes (from 106 cells)
19 700 _+1 702 10056
7856 +_901 39.8 + 1.78%
107
1°°]A
for an c~2-adrenoceptor. The relative order of potency for antagonists was: yohimbine > phentolamine > prazosin > propranolol (fig. 3), and for agonists: clonidine > epinephrine > norepinephrine > isoproterenol (fig. 4). The affinity of agonists was 5-10-fold lower in intact adipocytes than in membranes whereas there was only a 2-3-fold difference between the affinity of antagonists (table 2). The slope factor was about 1 for the antagonist competition in membranes and about 0.95 in intact adipocytes, however, this difference was not statistically significant (P > 0.05, table 2). The slope factor of clonidine was about 0.8 both in membranes and in intact adipocytes, whereas the slope factors for the other agonists were considerably lower in both preparations (table 2). Addition of Gpp(NH)p (100 /~M), a nonhydrolyzable analogue of GTP, and NaC1 (150 mM) to the membrane preparation decreased the affinity of epinephrine but increased the slope factor (fig. 5). The same additions had much less effect on the competition of clonidine in the membranes (fig. 5). Addition of NaC1 (up to 300 mM) and Gpp(NH)p to intact adipocytes did not change the slope factor or the affinity of agonists (data not shown).
100"1B
N >T
• clonidine 0 epinephtine • n0repinephrine
'?-" o u 0
~
~
~:~,~ " ~
~
s
o
p
~
9 6 7 6 5 -log [ADRENERGICAGONISTS] (M)
4
3
Fig. 4. Adrenoceptor agonist competition of specifically bound [3H]YOH to adipocyte membranes (A) and intact h u m a n adipocytes (B). The incubations were performed as described in fig. 3.
3.3. Specificity for agonists and antagonist
3.4. Antilipolytic effects of ae-agonists
Adrenoceptor antagonists and agonists competed for [3H]YOH binding sites in both membranes and intact adipocytes in the order expected
Clonidine and epinephrine (in the presence of propranolol) had only slight inhibitory effects on basal lipolysis in human adipocytes (data not
TABLE 2 Adrenoceptor antagonists and agonists competition constants in adipocyte membranes and in intact adipocytes. Comparison with the antilipolytic effect in intact cells, Intact adipocytes and adipocyte membranes were incubated at 20°C for 25 rain with [3H]YOH in the absence and presence of various competing agents. Competitor
Membranes
Intact adipocytes
N ~
ECso b
(-)Epinephrine (-)Norepinephrine Clonidine
5 3 8
659 1403 40
Yohimbine Phentolamine Prazosin
6 4 3
2.3 +_ 0.36 5.2+_ 0.90 2339 _+846
a Values
+ 79 +_261 +- 3.8
n H~
N "'
ECso b
0.53+-0.08 0.56+_0.10 0.86+_0.03
7 3 7
5997 15588 198
1.12 + 0.03 0.99+_0.10 1.10+0.16
6 3 3
+- 542 +_ 889 +_ 31
7.5 +_ 1.1 17.1+_ 4.5 9983 +__2225
Antilipolytic effect n H~
N ~
iCso d
0.49+-0.04 0.53_+0.13 0.79+_0.06
6
661+_51
6
52+_ 5,1
0.96 +_0.06 0.93+0.14 0.98+-0.16
are the mean + S.E.M. of (N) experiments, b ECs0 ' concentration of competing agent producing 50% inhibition of specific [3H]YOH binding (nM). c n n ' slope factor (pseudo Hill coefficient). Determined as described in Methods. a iC5o ' concentration of agonists producing 50% inhibition of the stimulated lipolysis (nM).
108 300 I00'
g
c
ff
~
200
50.
I00 c
:z:
A Bepinephrfne
+ Nacl+ Gpp(NH)p
~
~:~
..,.( -log [AGONISTS]
(M )
Fig. 5. Effect of Gpp(NH)p and NaCI on competition of epinephrine and clonidine for [ 3H]YOH binding sites in human adipocyte membranes. Adipocyte membranes were incubated with 2 nM [3H]YOH with increasing concentrations of epinephrine and clonidine in both the absence and presence of Gpp(NH)p (100 /~M) and NaC1 (150 mM). ECs0 of epinephrine and clonidine were 668 nM and 39 nM, respectively, In the presence of Gpp(NH)p and NaC1 these values were 4228 nM and 72 nM. The slope factors of epinephrine and clonidine were 0.54 and 0.83 respectively, and in the presence of Gpp(NH)p and NaCI these values were 0.86 and 0.92, respectively. The data are the means of 3 (clonidine) and 4 (epinephrine) different experiments performed in duplicate and run in parallel.
shown). The action of these a2-receptor agonists was pronounced when lipolysis was stimulated by a combination of theophylline (1 mM) and adenosine deaminase (5 #g/ml). This combination increased glycerol release from 260 + 21 nmol/106 cells per 90 min to 515 + 56 nmol/106 cells per 90 min. At maximally effective concentrations, clonidine (10 /~M) and epinephrine (10 #M) inhibited lipolysis by about 60% (fig. 6). However, the sensitivity expressed as IC50 was much higher for clonidine than for epinephrine (57 + 9 nM and 663 + 59 nM, respectively).
3.5. Correlation between binding and antilipolysis To evaluate if the binding data were related to the biological effect (antilipolysis), multiple correlations were performed between [ 3H]YOH binding data in both intact adipocytes and membrane fractions and the antilipolytic effect of clonidine (fig. 7). The total binding capacity (Bma x) for [3H]YOH binding in membranes was positively correlated to
~ 100
o]
-log [AGONISTS] (MI
Fig. 6. Antilipolytic effects of epinephrine and clonidine in human adipocytes. (A) Adipocytes were incubated with epinephrine (e) and clonidine (O) in the indicated concentrations at 37°C for 90 rain. For the last 60 rain lipolysis was stimulated with theophylline (1 mM) and adenosine deaminase (5/~g/ml). In experiments with epinephrine, propranolol (50 /LM) was included in the medium. Lipolysis is expressed as the stimulated glycerol release minus basal glycerol release, mean_+ S.E.M. for 7 independent experiments for both clonidine and epinephrine. (B) Data for the antilipolytic effect of epinephrine and clonidine are given as percent of the response to these agonists in maximally effective concentrations.
the maximal antilipolytic effect of clonidine (r = 0.70, P < 0.001, n = 20, fig. 7A). Furthermore, the binding affinity (Kd) of YOH in membranes was positively correlated to the sensitivity (IC50) of clonidine in inhibiting the stimulated lipolysis (r = 0.65, P < 0.01, n = 20, fig 7C). There was no significant correlation for intact adipocytes between Bmax and the maximal antilipolytic effect of clonidine when Bmax w a s expressed in relation to adipocyte number (106 cells). However, a significant and positive correlation was found (r = 0.79, P <0.001, n = 15, fig. 7B) if YOH binding to adipocytes was expressed in relation to the surface of the adipocytes (100 cm2). No correlation was demonstrated in intact adipocytes between the binding affinity and the IC50 value of the antilipolytic effect of clonidine (r = - 0.04, fig. 7D).
109 100-
100- B
7-
q-
~g
~
,w,~
~g
50,
so-
~.=_. CELLS r = 0.79 p < O.001
MEMBRANES r = 0.70 p < 0.001
~
o
0
IO~O
56o
t60
2~0
360
460
s~o
B max ( fmol / 100 cna=)
Bma x ( fnaol / nag protein )
150"! D
7., o,. ._%
100
..o •
~
~-=
• eee
so
•
~
so
>_ •
MEMBRANES r : o.65 p < 0.01
== CELLS
t-
r=
-0.04
NS 0
AFFINITY OF BINDING (K d} (nM)
AFFINITY OF BINDING ( Kd) (riM)
Fig. 7. Relationship between YOH binding and antilipolytic effect of clonidine. (A) Relationship between BmaX in membranes and maximal antilipolytic effect of clonidine. (B) Relationship as in A but obtained with intact adipocytes. (C) Relationship between K d in membranes and ICs0 of clonidine. (D) Relationship as in C but obtained with intact adipocytes. Total binding capacity (Bmax) and binding affinity (Kd) of YOH binding to membranes and intact adipocytes were calculated from Scatchard plots. The antilipolytic effect of clonidine is expressed as percent inhibition of the stimulated lipolysis without clonidine. The relationships were examined by regression analysis (intact adipocytes, n = 15; membrane preparations, n = 20).
4. D i s c u s s i o n
We have now demonstrated that az-adrenoceptors can be measured readily in both intact h u m a n adipocytes and adipocyte membranes. The binding of [ 3 H ] Y O H was saturable, specific and reversible both in intact cells and in membranes. The order of potency for agonists and antagonists for competition for [ 3 H ] Y O H binding was similar in intact adipocytes and in m e m b r a n e preparations and this order of potency was typical for the a2-receptor (Berthelsen and Pettinger, 1977). The total binding capacity of [ 3 H ] Y O H (B,,ax) in intact adipocytes
was 903 f m o l / 1 0 6 cells and the equilibrium binding constant (K d) was 6.6 nM. The values were 463 f m o l / m g protein and 2.1 nM, respectively, in the m e m b r a n e preparation. The Scatchard plot yielded straight lines for both preparations, indicating that [3H]YOH binds to a homogenous class of receptors in both preparations. All these data suggest that [3H]YOH binds to the same a2-adrenoceptors in intact adipocytes and in adipocyte membranes. However, some minor differences exist between the binding characteristics in the two preparations. [ 3 H ] Y O H (adrenoceptor antagonist) binds with lower affinity in intact adipocytes (Kd:
110
6.6 nM vs. 2.1 nM) and adrenoceptor agonists bind with 5-10 times lower affinity in intact cells. These differences could be explained by an enhanced degradation of the agonists/antagonists by the intact cells a n d / o r the fact that intact cells contain many different ions and other substances as nucleotides (see below) which could interfere with the hormone-receptor interaction. Nonspecific binding was considerably higher in intact adipocytes (34%) than in membranes (7%). These differences indicate that [~H]YOH presumably may bind to some sites of a non-receptor nature in intact adipocytes. These findings are in accordance with previous results obtained with intact cells using [3 H]dihydroalprenolol to identify ~8-adrenoceptors (Staehelin et al., 1982; Insel and Michele, 1979). Presumably, the lipophilic nature of yohimbine results in binding, especially intracellularly, to lipid structures which are not 'true' receptors. The finding that yohimbine ( > l >M) could inhibit more [3H]YOH binding in intact adipocytes than clonidine is also consistent with this hypothesis. The existence of a2-adrenoceptors in human adipocyte membranes had previously been shown with the ~e-receptor agonist, [)H]clonidine (Berlan and Lafontan, 1980) or the c~2-receptor antagonist, [)H]YOH (Tharp et al., 1981; Lafontan et al., 1983). The K d values obtained in these latter studies (0.8-3 nM) are in good agreement with the K d values obtained with membranes in the present study (2.1 nM). Furthermore, the total binding capacity was in the range described by others (400-600 f m o l / m g protein). Many [3H]YOH binding sites seemed to be lost during the membrane isolation procedure (table 1). Destruction of the receptors during the isolation procedure is unlikely since the characteristics of the receptor binding (affinity for agonists and antagonists) in membrane preparations were rather similar with the data obtained with intact adipocytes (table 2) and similar with data obtained from other studies concerning human adipocyte membranes (Tharp et al., 1981; Lafontan et al., 1983). Thus, the reduced binding in membranes may have been due to incomplete recovery of plasma membrane proteins a n d / o r loss of internalized receptors which may be labeled by [3H]YOH in intact adipocytes.
The [)H]YOH binding sites were homogeneous towards antagonists, as indicated by the Hill coefficients (near unity), in both membranes and intact adipocytes (table 2). However, agonist binding had different characteristics on membranes and on intact cells. The slope factor of agonist binding in these two preparations was considerably below 1, indicating that agonists bound to the c~2-receptor with different affinities. The agonist binding data could be explained by binding of agonists to the receptor in a high affinity state and a low affinity state as shown in membranes from platelets and liver cells (Hoffman et al., 1980a). In the presence of Gpp(NH)p and Na + the high affinity state of the receptor in membranes is converted to a low affinity state (Hoffman et al., 1980a,b; Gonzalez et al., 1981). These findings are in accordance with our data from membranes since the affinity of epinephrine decreased and the slope factor increased in the presence of Gpp(NH)p and Na + (fig. 6). However, even though the slope factor increased significantly, it was still less than 1, suggesting incomplete disappearance of the binding heterogeneity. The same additions bad much less effect on the clonidine competition curve (fig. 6), presumably because clonidine is only a partial agonist. Addition of Gpp(NH)p and Na" to the membrane preparation increased the ECs0 values of epinephrine about 10 times (to 4120 nM) which is close to the EC50 value of epinephrine in intact adipocytes (5997 nM). These findings suggest that nearly all the az-receptors in intact adipocytes exist in the low affinity state for agonist binding, which also would seem likely in view of the high content of Na + and G T P in intact cells. However, the agonist competition curves in intact cells are, as was mentioned, shallow suggesting that the binding of agonists to the a2-receptor in intact cells could not be explained only by binding of agonists to a homogeneous class of low affinity receptors but other mechanisms must also be acting, e.g. different affinities of the receptor or negative cooperativity. One can only speculate about the exact relationship between radioligand binding studies, the adenylate cyclase complex and the biological effect. However, we have now found that the ECs0 values of clonidine and epinephrine obtained from competition curves in adipocyte membranes corre-
111
lated very well, and were in the same range as the ICs0 values of the same agonists on the biological effect (antilipolysis, table 2). Direct correlation studies between [3H]YOH binding (at 20°C) and antilipolytic effect of clonidine (at 37°C) showed a positive relationship between Bmax in adipocyte membranes and the maximal antilipolytic effect of clonidine (fig. 7). In membranes, a positive correlation was also demonstrated between the binding affinity and the sensitivity of the clonidine-induced antilipolysis. No correlation was detected between the binding affinity and the ICs0 values of clonidine with intact adipocytes. This latter finding indicates that the affinity of antagonist binding in intact adipocytes is not directly related to the sensitivity of the antilipolytic effect of agonists. This discrepancy between c~2-receptor antagonist binding and the biological response is unexplainable at the present moment. However, the maximal receptor number identified by [3H]YOH in intact adipocytes was positively correlated to the maximal antilipolytic effect of clonidine. In conclusion it is demonstrated that a2-rece ptors could easily be studied in intact adipocytes and in adipocyte membranes. Adrenoceptor antagonists, which are without intrinsic activity, bound to a homogeneous class of receptors with high affinity in intact cells and in membranes. In the membrane preparation, agonists bound to the t~z-receptor with different affinities and binding was dependent on the presence of GTP and Na +. In intact adipocytes, agonists bound to the receptor in a heterogeneous manner with shallow competition curves. The study demonstrated that the [3H]YOH binding data obtained with the membrane fractions correlated best with the biological effect mediated by these receptors.
Acknowledgements We appreciate the skillful technical assistance of J. Soholt, T. Skrumsager, L. Blak and P. Jorgensen. The study was supported by grants from Aarhus University, Landsforeningen for Sukkersyge, Nordisk Insulinfond, P. Carl Petersens Fond and the Danish Medical Research Council.
References Berlan, M. and M. Lafontan, 1980, Identification of t~2-adrenergic receptors in human fat cell membranes by [3H]clonidine binding, European J. Pharmacol. 67, 481. Berlan, M. and M. Lafontan, 1982, The a2-adrenergic receptor of human fat cells: comparative study of a2-adrenergic radioligand binding and biological response, J. Physiol. (Paris) 78, 279. Berthelsen, S. and W.A. Pettinger, 1977, A functional basis for classification of c~-adrenergic receptors, Life Sci. 21,596. Burns, T.W., P.E. Langley, B.E. Terry, D.B. Byland, B.B. Hoffman, M.D. Tharp, R.J. Lefkowitz, J.A. Garcia-Sainz and J.N. Fain, 1981, Pharmacological characterizations of adrenergic receptors in human adipocytes, J. Clin. Invest. 67, 467. Burns, T.W., B.E. Terry, P.E. Langly and G.A. Robinson, 1980, Role of cyclic AMP in human adipose tissue lipolysis, Adv. Cycl. Nucl. Res. 12, 329. Chernick, S.S., 1969, Determination of glycerol in acyl glycerols, Methods Enzymol. 14, 627. DeLean, A., P.J. Munson and D. Rodbard, 1978, Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay and physiological dose-response curves, Am. J. Physiol. 235, E 97. Dulis, B.H. and I.B. Wilson, 1980, The fl-adrenergic receptor of live human polymorphonuclear leukocytes, J. Biol. Chem. 255, 1043. Fain, J.N. and J.A. Garcia-Sainz, 1983, Adrenergic regulation of adipocytes metabolism, J. Lipid Res. 24, 945. Gonzalez, J.:L., C. Carpene, M. Berlan and M. Lafontan, 1981, Discrepancy in the regulatory effects of sodium and guanine nucleotides on adrenaline and clonidine binding to a 2adrenoceptors in human fat cell membranes, European J. Pharmacol. 76, 289. Hoffman, B.B., T. Michel, D. Mullikin-Kilpatrick, R.J. Lefkowitz, M.E.M. Tolbert, H. Gilman and J.N. Fain, 1980a, Agonist versus antagonist binding to alpha-adrenergic receptors, Proc. Natl. Acad. Sci. U.S.A. 77, 4569. Hoffman, B.B., D. Mullikin-Kilpatrick and R.J. Lefkowitz, 1980b, Heterogeneity of radioligand binding to a-adrenergic receptors: analysis of guanine nucleotide regulation of agonist binding in relation to receptor subtypes, J. Biol. Chem. 255, 4645. Insel, P.A. and S. Michele, 1979, Temperature-dependent changes in binding to fl-adrenergic receptors in intact $49 lymphoma cells, J. Biol. Chem. 254, 6554. Kather, H., J. Pries, V. Schneider and B. Simon, 1980, Inhibition of human fat cell adenylate cyclase mediated via alpha-adrenoceptors, European J. Clin. Invest. 10, 345. Kather, H. and S. Simon, 1981, Adrenoceptor of the alpha 2-subtype mediating inhibition of the human fat cell adenylate cyclase, European J. Clin. Invest. 11, 111. Lafontan, M., M. Berlan and A. Villeneuve, 1983, Prependance of a z - over fll-adrenergic receptor sites in human fat cells is not predictive of the lipolytic effect of physiological catecholamines, J. Lipid Res. 24, 429.
1i2 Lowry, O.H., N.J. Rosebrough, A.L. Farr and R.J. Randall, 1951, Protein measurement with the Folin phenol reagent, J. Biol. Chem. 193, 265. Mukherjee, C, and R.J. Lefkowitz, 1976, Direct studies of /3-adrenergic receptors in intact frog erythrocytes, Life Sci. 19, 1897. Pedersen, O., E. Hjollund, H. Beck-Nielsen, O. Sonne and J. Gliemann, 1981, Insulin receptor binding and receptor mediated insulin degradation in human adipocytes, Diabetologia 20, 636. Pittman, R.N. and P.B. Molinoff, 1980, Interactions of agonists and antagonists with ,8-adrenergic receptors on intact L6 muscle cells, J. Cycl. Nucl. Res. 6, 421. Richelsen, B., E.F. Eriksen, H. Beck-Nielsen and O. Pedersen, 1984, Prostaglandin E 2 receptor binding and action in human fat cells, J. Clin. Endocrinol. Metab. 59, 7. Rodbell, M., 1980, The role of hormone receptors and GTPregulatory proteins in membrane transduction, Nature 284, 17.
Scatchard, G., 1949, The attractions of protein for small molecules and ions, Ann. N.Y. Acad. Sci. 51,660. Sj6str6m, L., P. Bj6rntorp and J. Vrhna, 1971. Microscopic fat ceils size measurements of frozen-cut adipose tissue in comparison with automatic determination of osiumfixed fat cells, J. Lipid Res. 12, 521. Staehelin, M., P. Simons, K. Jaeggi and N. Wagger, 1982, CGP-12177, a hydrophilic fl-adrenergic receptor radioligand reveals high affinity binding of agonists to intact cells, J. Biol. Chem. 258, 3496. Tharp, M.D., B.B. Hoffman and R.J. Lefkowitz, 1981, aAdrenergic receptors in human adipocyte membranes: direct determination by (3H)-yohimbine binding, J. Clin, Endocrinol. Metab. 52, 709.