- Email: [email protected]
Characteristics of the β1- and β2-adrenergic-sensitive adenylate cyclases in glial cell primary cultures and their comparison with β2-adrenergic-sensitive adenylate cyclase of meningeal cells
Brain Research, 213 (1981) 151-161 © Elsevier/North-Holland Biomedical Press
151
CHARACTERISTICS OF THE ill- AND fle-ADRENERGIC-SENSITIVE ADENYLATE CYCLASES IN GLIAL CELL PRIMARY CULTURES AND THEIR COMPARISON WITH fl2-ADRENERGIC-SENSITIVE ADENYLATE CYCLASE OF MENINGEAL CELLS
C. EBERSOLT, M. PEREZ, G. VASSENT and J. B O C K A E R T
Laboratoire de Physiologie Cellulaire, Coll~ge de France, 11 place Marcelin Berthelot, 75231 Paris, Cedex 05 (France) (Accepted October 9th, 1980)
Key words: fll-adrenergic receptor - - fl2-adrenergic receptor - - adenylate cyclase - - glial c e l l - meningeal cell
SUMMARY
The agonist specificity pattern of the fl-adrenergic adenylate cyclase in glial primary cultures was not typical of either ill- or fl2-adrenergic receptors. The dose-response curves for adrenaline did not correspond to simple mass action kinetics and their computer analysis suggests the presence of both fir and fl2-adrenergic-sensitive adenylate cyclase (58 4- 17 ~ and 42 4- 17 °/o respectively). Similar properties of ill- and fl2-adrenergic-sensitive adenylate cyclases were found by computer analysis of the dose-response curves for isoprenaline in the presence of a constant concentration of practolol (a selective fll antagonist) (55 10 ~ and 45 ± 10 ~ of ill- and fl2-sensitive adenylate cyclase respectively). The curves for displacement of [3H]dihydroalprenolol by practolol confirm these results. For purpose of comparison, the fl-adrenergic receptors of meningeal cells in cultures were subjected to similar analysis. The results dearly showed that these cells exclusively contained fl2-adrenergic receptors.
INTRODUCTION
Two hypotheses can be proposed to account for the complex drug affinity spectrum on fl-adrenergic receptors in various tissues. The first is that there is a large number of iso-fl-adrenergic receptors, and the second is that these tissues contain different ratios of the fit and f12 subtypes originally described by Lands et alA. The
152 results of recent direct binding experiments indicate that the second hypothesis is the correct one 9. In mammalian brain, the results of these experiments6,S,a0,14, as well as the characteristics observed for fl-adrenergic-sensitive adenylate cyclaseI indicated that both ill- and fl2-adrenergic receptors are present, and have different development patterns 12 and topographical distributions 1,s,13. The cerebral cortex contains mainly fll and only a few f12 receptors and the opposite proportions apply in the cerebeUar cortexS,~0,1.4. Minneman et al. 6 reported that in cerebral cortex, changes in receptor sensitivity following either 6-hydroxydopamine or desmethylimipramine treatment only occur in the/31 receptor subtype population. They therefore postulated that/31adrenergic receptors are true postsynaptic receptors, whereas/32 receptors are confined to glial cells and blood vessels. U'Prichard et al. ~4 confirmed these results for cerebral cortex and also demonstrated that in cerebellar cortex the sensitivity of fl2-adrenergic receptors could be altered after lesion of the ascending norepinephrine bundle. We recently obtained a homogeneous population of glial cells of the astrocyte type in primary culture 2 containing a highly potent fl-adrenergic-sensitive adenylate cyclase. The experiments reported here were designated to determine whether the /3adrenergic receptors involved could be classified as either fll and/32 subtypes. MATERIALS AND METHODS Cell cultures, homogenate preparations and adenylate cyclase assays were done as previously described z. Fluphenazine (t0-SM) was present in adenylate cyclase assays.
(3H]dihydroalprenolol ([3H]DHA) binding Cells from 10 dishes (diameter 10 cm) were rinsed 3 times with 9%0 NaC1, scraped off with a rubber policeman and centrifuged. The pellet was homogenized in 5 ml of 50 mM Tris.HC1 pH 7.8 using a teflon-glass homogenizer and centrifuged at 47,000 × g for 20 rain. The pellet was washed once more and finally resuspended in 1.6 ml of Tris.HCl pH 7.8. Aliquots of 10 #1 were incubated for 30 min at 30 °C with [~H]DHA, 50 mM Tris.HCl, pH 7.8 and other drugs to a total volume of 100/A. Non-specific binding was determined in the presence of (--) alprenolol (5 × 10-6 M). At the end of the incubation period, samples were diluted with l ml ice-cold 50 mM Tris-HC1, pH 7.8, filtered through a glass fiber filter (GF/B Whatman) and washed with 5 × 4 ml of the same medium.
Chemicals (--) Propranolol and (~_) practolol (ICI), ( - ) alprenolol (Hassle) ( ± ) butoxamine (Burrough), (--) isoprenaline, (--) adrenaline, (--) noradrenaline (Sigma), salbutamol (Allen and Hanbury's), hydroxybenzylisoproterenol was donated by J. Hanoune and was from Phillips-Durphar.
Computer fit of the data In our computer analysis we assumed like other authorsZ, 8 that there are only
153 two fl-adrenergic receptor subtypes and that interaction of drugs with receptors follows simple mass action kinetics. The formula used for fitting the adenylate cyclase dose-response curves and the [aH]DHA binding displacement curves was: F
F
Y -- N1
+ N2 F + Ks~ (1 + I/K~I or Kay)
F + Ks~ (1 + I/K~ or Ka2)
For the adenylate cyclase dose-response curves Y = adenylate minus basal activity, N1 and Nz adenylate cyclase activities due to the stimulation of ill- and fl2-adrenergic-sensitive adenylate cyclase respectively; F = agonist concentration, Ks~ and I ~ = apparent affinity of the agonist for fll and fl2-adrenergic-sensitive adenylate cyclases respectively, I = antagonist concentration, K~ and K~2 ---- apparent inhibition constants of the antagonist for ill- and fl2adrenergic receptors respectively. For the EaH]DHA binding displacement curves Y = specifically bound [aH]DHA, N1 and N2 -- maximal binding capacities of ill- and flz-adrenergic receptors respectively, F = free [aH]DHA concentration, Ks, ---Ks~ = affinities of [3H]DHA for both ill- and fl2-adrenergic receptors, I = displacing drug concentration, Kal and Ka~ = affinities of displacing drug for fll and f12 adrenergic receptors respectively. Minimization of E (Y e x p - Y theoretical) 2 is computerized by a non-linear regression analysis using a CDC 6600 computer (INs Pa Paris 6~me). RESULTS AND DISCUSSION
fll and fl2-adrenergic sensitive adenylate cyclase in glial cell primary cultures Several fl-adrenergic agonists stimulated the adenylate cyclase in primary cultures of glial cells (Fig. 1). When the potencies of the different drugs were compared Primary culture of Glial cells
._c E o 40(
[] Iso~renatine • HydroxybenzylIsoproterenol • Adrenaline 0 Nor -Adrenaline Salbutamol 10( u
=
0/_•/10I .8
1~.7
10 ~1 AGONIST (M)
I 10.5
10 "4[
Fig. 1. Adenylate cyclase activation by fl-adrenergicagonists in glial cell homogenates. Basal adenylate cyclase activity was 85 pmol/20 min/mg protein.
154 on the basis o f the K a a p p (agonist c o n c e n t r a t i o n giving half m a x i m a l stimulation), the o r d e r was ( - - ) isoprenaline :> h y d r o x y b e n z y l i s o p r o t e r e n o l :> adrenaline ~- salb u t a m o l ~ n o r a d r e n a l i n e (Fig. 1). The m e a n K a a p p values are given in Table I. S a l b u t a m o l was a partial agonist o f the fl-adrenergic-sensitive adenylate cyclase. The m a x i m a l stimulation p r o d u c e d by this ligand was 33 ~ 6 . 4 ° / (n =:= 3) o f that o f adrenaline. Simple analysis o f these results indicates t h a t a h o m o g e n e o u s p o p u l a t i o n o f either fl~- a n d fl2-adrenergic-sensitive a d e n y l a t e cyclase was n o t present. In the presence o f fl2-adrenergic-sensitive a d e n y l a t e cyclase only, we expected t h a t : (1) h y d r o x y b e n z y l i s o p r o t e r e n o l would be m o r e p o t e n t t h a n isoprenalinel,5,13 b u t this was not the case (Table I, Fig. l ) ; and (2) that the K~app for adrenaline w o u l d be 8-10 times lower than the the K a a p p for n o r a d r e n a l i n e s a n d not, as we observed, only a little lower (Table I). On the other hand, in the presence o f fl~-adrenergic receptors only isoprenaline was expected to be 8-10 times more p o t e n t than h y d r o x y b e n z y l i s o p r o t e r e n o l l , a3 and s a l b u t a m o l to be inactive 1,7 which was not the case. The shape o f the d o s e TABLE I Specificity pattern of the [J-adrenergic-sensitive adenylate cyclase in glial cells - - relative proportions of ill- and fl2-adrenergic receptors coupled with adenylate cyclase
Values are mean ± S.E.M. of 3-6 experiments. Drugs Agonists (--) Isoprenaline (_q_) Hydroxybenzylisoproterenol (--) Adrenaline on both fix and [32receptors on f12receptors on fll receptors ( ~ ) Salbutamol (--) Noradrenaline Antagonists (--) Propranolol ( ± ) Butoxamine C~) Practolol on f12receptors o n [~1 receptors
Kaapp or Ki (nM)
[h [?~ ( '% of the total)
43 ± 5* 75 -j 13" 656 160 1200 690 920
J 57* ± 117"* ± 340'* ± 300* -~ 300*
42 ~ 17"** 58 5 17"**
4± 1 4700 ± 1200§ 28,900 ± 3990§§ 556 -- 196§§
45 ± 10§§§ 55 ± 10§§§
* The Kaapp is the concentration of agonist giving half maximal activation. * * The Kaapp for ill- and fl2-adrenergic-sensitive adenylate cyclases is given by the complete analysis of the adrenaline dose-response curve described in Fig. 2. *** The percentage of ill- and fl2-adrenergic-sensitive adenylate cyclases is given by the computer analysis of the adrenaline dose-response curve described in Fig. 2. § The Ki are calculated from Hofstee plots similar to those in Fig. 3. §§ The K~of practolol for/51- and fl2-adrenergic-sensitive adehylate cyclases are given by the computer analysis of the practolol curve described in Fig. 3. §§§ The percentage of ill- and fl2-adrenergic-sensitive adehylate cyclases are given by the computer analysis of the practolol curve described in Fig. 3.
155 Primary culture of Glial cells .% • ©
• ___
Isoorenaline Adrenaline Nor-adrenaline
.~40C
40(
.c E
Adrenaline Theoretical I~2 Vmax = 118.5 _ KAapp= 5 x 10-t~ M I~1 Vmax = 276.5 KAapp =10-6 M
LJ
// 200f
0
Q,
/
/ / ,Q
•
/ill.-" l l - ~
/
E
?,"
>
0 ~)
10-5
5 xl~ 8 10x10.8 15x10~ V-b/[Adrenaline or Nor-Adrenaline] i
25 x10'8 V-b/[ISO]
I
50 x 10"8
~4-
Adrenaline [M] }IL
Fig. 2. Heterogeneity of the adrenaline dose-activation curve. Presence of fla- and fl2-adrenergicsensitive adenylate cyclase in glial cell homogenates. Left panel: Hofstee plots of isoprenaline, noradrenaline and adrenaline dose-response curves (data are from Fig. 1) V--b = adenylate cyclase activity above basal level. The slope gives the Kaapp of the drug. ISO = isoprenaline. Right panel: adenylate cyclase activation by adrenaline. Points are experimental data taken from Fig. 4. The dotted line is the theoretical line obtained by computer fitting (see Material and Methods). The amount of fll and fiesensitive adenylate cyclases and the Kaapp of adrenaline for these enzymes are given.
response curves for hydroxybenzylisoproterenol and adrenaline indicates that they do not correspond to simple mass-action kinetics, since they extend over more than two orders of magnitude (Fig. 1). The Hofstee plot of the dose-response curve for adrenaline confirmed this heterogeneity and showed two clear slopes (Fig. 2). Furthermore, non-linear regression analysis of the adrenaline dose-response curves was carried out by computer. The best fits were obtained with a model involving two sub-populations of fl-adrenergic receptors with different Kaapp and maximal velocities. An example of such fitting is given in Fig. 2. The mean K~app and percentages for the two sub-populations are given in Table I. We deduced from the above results that the fl-adrenergic sensitive adenylate cyclase with the highest affinity for adrenaline (Kaapp = 160 J: 117 nM) corresponded to a f12 sub-population since in meningeal cells which contained only fl2-adrenergic receptors (see below) the Kaapp for adrenaline was 213 ~ 79 nM (Table II) and the one with the lowest affinity (Kaapp = 1200 :k: 340 nM) to a fll sub-population. Computer analysis indicates that these two subpopulations represent 41 -+- 17 ~ and 58 :[: 17 700 of the total fl-adrenergic-sensitive adenylate cyclase respectively. The heterogeneity o f the hydroxybenzylisoproterenol dose-response curves substantiated this conclusion. The Hofstee plots of the curves for isoprenaline and noradrenaline were linear (Fig. 2). This is not surprising since isoprenaline is known to be equipotent in both ill- and fl2-adrenergic receptors, whereas noradrenaline has been reported to be only twice as potent on fll as on f12 receptors in the brain1, a3.
156 TABLE II Specific of the fl-adrenergic-sensitive adenylate cyclase in meningeal cells The Kaapp are calculated from Hofstee plots similar to those in Fig. 4. Values are the mean ± S.E.M. of 3-5 experiments. Drugs Agonists (2) Hydroxybenzylisoproterenol (--) Isoprenaline (--) Adrenaline (2) Salbutamol (--) Noradrenaline Antagonists (--) Propranolol (~:) Butoxamine (A) Practolol
Kaapp (riM) or K~ 28 ± 8 58 ~ 3 213 ~: 79 765 -~ 189 3200 :t:: 300 2.2 ~: 0.6 1900 zk 170 43,000 5:860
Several fl-adrenergic antagonists have been shown to have a higher affinity for either fll and flz receptors, whereas others have the same affinity for both. Examples of the former class are practolol and butoxamine 8 reported to be ill- and fl2-selective antagonists respectively, whereas propranolol is an example of the latter class. Fig. 3 shows that the dose-activation curve for isoprenaline was monophasic (the Hofstee plot gave one slope). In the presence of constant propranolol or butoxamine concentrations, the dose-response curves for isoprenaline shifted to the right. The Hofstee plots of these shifted curves also gave one slope, indicating that the/31- and fl2adrenergic receptors coupled with an adenylate cyclase in glial cells showed no selectivity for propranolol or butoxamine. Butoxamine which has been reported to be more potent on flz than on fll adrenergic receptors in vivo 8 showed no such selectivity in our experiments. However, there are reports of poor fl2-selectivity by this drug at receptor level 3,7. Note that a 2-3-fold difference in the affinities of butoxamine for the two subpopulations would be difficult to detect. The dose-response curve for isoprenaline in the presence of a constant concentration ofpractolol was not parallel to the other curves and gave a curvilinear line when plotted according to Hofstee (Fig. 3). Regression analysis of this curve was done by computer. The best fit was obtained with a model assuming that isoprenaline has the same affinity for ill- and fle-adrenergic sensitive adenylate cyclases and in which practolol discriminated between them. The computer-determined Ki values for illand fl2-adrenergic-sensitive adenylate cyclases were 6.5 x 10-7 M and 1.2 × 10-~ M and the percentage of the two populations were respectively 48 % and 58 ~o of the total (Fig. 4). Mean values are given in Table I. fl2-adrenergic-sensitive adenylate cyclase in meningeal cell primary cultures The agonist specificity pattern of the fl-adrenergic-sensitive adenylate cyclase in meningeal cells was clearly different from that in glial cells. The order of potency was
157 Primary culture of glial cells 10£
~'~ /
o5£
•
ISO+Butoxamine 12 xlO-5M)
0
ISO+Practolol (2 xlO-5M)
i
!5x10-8M)
"5 ae ISO (MI
~C~'~ 13. '5
\~t,~,~ \.
",,~
~
[]
ISO KAapp = 6x10-8M
•
Butoxamine KI= 3xlO-6M
©
Praetolol (exl3erimental)
.... (Theoretical)
O 5£ .6;
N,~.~/.~
I
5 x 10"8
,1.42%
I
~
(al
I
I=
(O,@,A)
10 x 10~8
I
l x 10+8
2xl(f8
A/[,s0] Fig. 3. Competitive inhibition by butoxamine, practolol and propranolol of adenylate cyclase stimulation by isoprenaline in glial cell homogenates. Upper panel shows the dose-response curves. Adenylate cyclase activities are expressed in percentage of the maximal increase in cyclic AMP due to the presence of saturating concentrations of isoprenaline (140 pmol/20 min/mg protein). Lower panel shows the Hofstee plots of the dose-response curves. Note that the plots of the isoprenaline dose-activation curves obtained in the absence ([]) or presence of either butoxamine (&) or propranolol (@) were linear, but the plot of the isoprenaline dose-activation curve obtained in the presence of practolol (O) was curvilinear. The points are experimental data whereas the dotted line is the theoretical line obtained by computer fitting. For the latter, the apparent affinity of isoprenaline was taken from the Hofstee plot and was equal to 6 x 10-a M. The percentage of ill- and flz-sensitive adenylate cyclases and the K~ of the enzymes for practolol are given.
158 Primary culture of meningeal layer J
,<
2. 0 c_
, O/ 5£
/
.,~/
•
ISO + Butoxamine (2x10-5M)
~
O
ISO + Practolol
-
•
ISO + Propranolol (5xlO-8M)
(10-4M)
o
<~ ~6 ~e I 4"
~
ld 8
10-7
1~ ISO (M)
10-5
t
b--
10-4
10C :
Q..
ISO KAapp=6,4xlO8M
0
5O
t
t 5 x 10*8
I t
1 10 x 10*8
J
),
(: J)
15 x 10÷8
~
i
,
1 ~ I 0 ÷8
2×10"8
3x10 ~
J 2 x 10``.8
l 4 ~ 10`..8
l 6 × 10*8
~ (e,A) ~,
(C,)
% of 2L ~'lsO} Fig. 4. Competitive inhibition by butoxamine, practolol and propranolol of adenylate cyclase stimulation by isoprenaline in meningeal cell homogenates. Upper panel shows the dose-response curves. Adenylate cyclase activities are expressed in percentage of the maximal increase in cyclic A M P due to the presence of saturating concentrations of isoprenaline (100 pmol/20 min/mg protein). Lower panel shows the Hofstee plots of the dose-response curves. Note that the 4 plots were linear.
hydroxybenzylisoproterenol > isoprenaline > adrenaline > salbutamol > noradrenaline. Mean Kaapp values are given in Table II. Four observations suggested the presence of a homogeneous population of fl2-adrenergic receptors coupled with adenylate cyclase : (1) all the dose-response curves were monophasic (the Hofstee plots
159 gave only one slope, data not shown); (2) hydroxybenzylisoproterenol was more potent than isoprenaline; (3) adrenaline was 10 times more potent than noradrenaline; and (4) salbutamol was an agonist. Since the agonist specificity pattern indicated the presence of a homogeneous population of flz-adrenergic receptors in meningeal cells, we examined the effects of the antagonists. As expected (Fig. 4) the dose-response curves for isoprenaline on the presence of constant concentrations of propranolol, butoxamine or practolol were all parallel. The Hofstee representations gave only one slope for each curve (Fig. 4). Mean K~ values are given in Table II.
[aH]DHA binding to glial cell membranes confirms the presence of both fl~- and fl~adrenergic receptors Scatchard analysis showed that [aH]DHA bound to a single class of high affinity receptors in glial cell membranes. The calculated Ka values was 0.5 nM. Binding site density (Bmnx) was 145 fmol/mg protein. Displacement curves of [aH]DHA binding by propranolol and practolol on glial cell membranes were not parallel. They extended over two orders of magnitude for propral~olol and more than 3 for practolol (Fig. 5). These data fitted with computer models involving one or two classes of sites. In glial cell membranes, the best fit was obtained with a one-site model for propranolol and with a two-sites model for Propranolol - glial cells
_'~
• --
100
--&~
--'l~
o
E
Practolol ~ glial cells • L meningeal cells
O
~,L
"6
i
Glial cells I
"10 r-
x~ 50
•
B2 35%
B1 65%
\o
\
-r cl i
'''* 10-8
.... 10-7
'
10 -6 10-5 ANTAGONIST (M)
10 4
10-3
Fig. 5. Curves for displacement of [aH]DHA binding by propranolol and practolol. [3H]DHA binding was 116 and 40 fmol/mg protein in glial and meningeal cell particulate fractions respectively. [a H ] D H A concentrations in glial and meningeal cell particulate fractions were 1.6 a n d 1.3 nM respectively. Points are experimental data and dotted lines represent computer fitting. The curve for displacment of [aH]D H A binding by practolol in glial cell particulate fractions fitted the model involving two classes of sites: 65 ~ high-affinity (Ki = 1.2 × 10 -7 M) and 35 ~ of low-affinity (K~ = 8 x 10 -6 M). Curves for [aHIDHA binding displacement by propranolol in glial cells and by practolol in meningeal cells fitted the model involving only one class of sites. K~ were 1.2 x 10 -9 M and 5 x 10 -5 M respectively.
160 practolol. The Ka of practolol for the two classes of sites were 1.2 × 10-7 M and 8 × 10-6 M respectively, suggesting that they represent fit- and flz-adrenergic receptors respectivelyT,s. fll-adrenergic receptors represented 65%o of the total fl-adrenergic receptor population. For purpose of comparison [3H]DHA binding on meningeal cell membranes was done. Only one single class of sites was detected )Ka := 1.3 nM, Bmax = 80 fmol/mg protein). The curve for [3H]DHA displacement by practolol indicated the presence of only one class of sites having a Ka of 5 x 10-2 M (fl2-adrenergic receptors). Two arguments may be advanced as proof that the presence of fl~,-adrenergic receptors in glial cell primary cultures is not due to contamination by meningeal cells. Firstly, 96 ~ of the cells present in our primary cultures contained the glial fibrillary acid protein 2, meningeal cells did not contain this proteinL Secondly, the possible contamination of 4-5 ~ of these glial cells by meningeal cells, which of course cannot be excluded, would not suffice to account for the presence of 40-50 ~i of fl2-adrenergic receptors, since meningeal cells contained fewer receptors than glial cells (see above). It is evident that our cell cultures do not contain all types of differentiated glial cells which could exist in the brain. However, the finding that both fit- and fl2adrenergic receptors are present in the glial cells described here does not support the proposal of Minneman et al. 6 that glial cells contained only fl2-adrenergic receptors. Thus, two questions remain open: (1) what is the nature of the fl-adrenergic receptors present in neurons ? Previous binding 9A°,14 and adenylate cyclase experiments t in the cerebral cortex show that fl-adrenergic receptors (more than 75~.(,) are of the fll subtype, whereas our present data indicate an almost equal concentration of fit- and fl2-adrenergic receptors in glial cells. Since cerebral cortex also contains fl2-adrenergic receptors from blood vessels 11, it is tempting to speculate that most neurons' adrenergic receptors are of the fll class, however, a direct proof of this is still lacking; and (2) are fll-adrenergic receptors in glial cells capable of adaptive responsiveness ? Our experiments do not of course show whether/31- and flz-adrenergic receptors are present in a single cell or if we are dealing with two different populations of glial cells. From a physiological and biochemical point of view it is important to clarify this question. Preliminary experiments indicates that Ca glioma cell lines also have both fitand fl2-adrenergic receptors. ACKNOWLEDGEMENTS This work was supported by LA 219 from CNRS and grants from INSERM (ATP 58 78 90) and CNRS (ATP 4149). We thank Anne du Parc for manuscript preparation. REFERENCES 1 Dolphin, A., Hamon, M. and Bockaert, J., The resolution of dopamine and [-¢1-and fl2-adrenergicsensitive adenylate cyclase activities in homogenates of cat cerebellum, hippocampus and cerebral cortex, Brain Research, 179 (1979) 305-317. 2 Ebersolt, C., Perez, M. and Bockaert, J., Neuronal, glial and meningeal localizations of neurotransmitter sensitive adenylate cyclases in cerebral cortex of mice, Brain Research, 213 0981) 139-150.
161 3 Hancok, A. A., De Lean, A. L. and Lefkowitz, E. J., Quantitative resolution of beta-adrenergic receptor subtypes by selective iigand binding:application of a computerized model fitting technique, Molec. Pharmacol., 16 (1979) 1-9. 4 Lands, A. M., Arnold, A., McAuliff, J. P., Luduean, F. P. and Brown, T. G., Differentiation of receptor systems activated by sympathomimetic amines, Nature (Lond.), 214 (1967) 597-598. 5 Lefkowitz, R. J. and Williams, L. T., Catecholamine binding to fl-adrenergic receptor, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 515-519. 6 Minneman, K. P., Dibner, M. D., Wolfe, B. B. and Molinoff, P. B., 81- and 82-adrenergic receptors in rat cerebral cortex are independently regulated, Science, 204 (1979) 866-868. 7 Minneman, K. P., Hegstrand, L. R. and Molinoff, P. B., The pharmacological specificity of beta-1 and beta-2 adrenergic receptors in rat heart and lung in vitro, Molec. Pharmacol., 16 (1979) 21-33. 8 Minneman, K., Hegstrand, L. R. and Molinoff, P. B., Simultaneous determination of beta-1 and beta-2 adrenergic receptors in tissues containing both receptor subtypes, Molec. Pharmacol., 16 (1979) 34-46. 9 Minneman, K. P. and Molinoff, P. B., Classification and quantification of 8-adrenergic receptor subtypes, Biochem. Pharmacol., 29 (1980) 1317-1322. 10 Nahorski, S., Heterogeneity of cerebral 8-adrenoreceptor bindingsites in various vertebrate species, Europ. J. PharmacoL, 51 (1978) 199-209. 11 Palmer, G. C., Beta-adrenergic receptors mediate adenylate cyclase responses in rat cerebral capillaries, Neuropharmacology, 19 (1980) 17-23. 12 Pittman, R. N., Minneman, K. P. and Molinoff, P. B., Ontogeny of 81- and 82-adrenergic receptors in rat cerebellum and cerebral cortex, Brain Research, 188 (1980) 357-368. 13 U'Prichard, D. C., Bylund, D. B. and Snyder, S. H., (~:) [all]epinephrineand (--) [aH]dihydroalprenolol binding to 81 and 82 noradrenergic receptors in brain, heart and lung membranes, J. biol. Chem., 25 (1978) 5090-5102. 14 U'Prichard, D. C., Reisine, T. D., Yamamura, S., Mason, S. T., Fibiger, H. C., Ehlert, F. and Yamamura, H. 1., Differential supersensitivity of 8-receptor subtypes in rat cortex and cerebellum after central noradrenergic denervation, Life Sci., 26 (1980) 355-364.
151
CHARACTERISTICS OF THE ill- AND fle-ADRENERGIC-SENSITIVE ADENYLATE CYCLASES IN GLIAL CELL PRIMARY CULTURES AND THEIR COMPARISON WITH fl2-ADRENERGIC-SENSITIVE ADENYLATE CYCLASE OF MENINGEAL CELLS
C. EBERSOLT, M. PEREZ, G. VASSENT and J. B O C K A E R T
Laboratoire de Physiologie Cellulaire, Coll~ge de France, 11 place Marcelin Berthelot, 75231 Paris, Cedex 05 (France) (Accepted October 9th, 1980)
Key words: fll-adrenergic receptor - - fl2-adrenergic receptor - - adenylate cyclase - - glial c e l l - meningeal cell
SUMMARY
The agonist specificity pattern of the fl-adrenergic adenylate cyclase in glial primary cultures was not typical of either ill- or fl2-adrenergic receptors. The dose-response curves for adrenaline did not correspond to simple mass action kinetics and their computer analysis suggests the presence of both fir and fl2-adrenergic-sensitive adenylate cyclase (58 4- 17 ~ and 42 4- 17 °/o respectively). Similar properties of ill- and fl2-adrenergic-sensitive adenylate cyclases were found by computer analysis of the dose-response curves for isoprenaline in the presence of a constant concentration of practolol (a selective fll antagonist) (55 10 ~ and 45 ± 10 ~ of ill- and fl2-sensitive adenylate cyclase respectively). The curves for displacement of [3H]dihydroalprenolol by practolol confirm these results. For purpose of comparison, the fl-adrenergic receptors of meningeal cells in cultures were subjected to similar analysis. The results dearly showed that these cells exclusively contained fl2-adrenergic receptors.
INTRODUCTION
Two hypotheses can be proposed to account for the complex drug affinity spectrum on fl-adrenergic receptors in various tissues. The first is that there is a large number of iso-fl-adrenergic receptors, and the second is that these tissues contain different ratios of the fit and f12 subtypes originally described by Lands et alA. The
152 results of recent direct binding experiments indicate that the second hypothesis is the correct one 9. In mammalian brain, the results of these experiments6,S,a0,14, as well as the characteristics observed for fl-adrenergic-sensitive adenylate cyclaseI indicated that both ill- and fl2-adrenergic receptors are present, and have different development patterns 12 and topographical distributions 1,s,13. The cerebral cortex contains mainly fll and only a few f12 receptors and the opposite proportions apply in the cerebeUar cortexS,~0,1.4. Minneman et al. 6 reported that in cerebral cortex, changes in receptor sensitivity following either 6-hydroxydopamine or desmethylimipramine treatment only occur in the/31 receptor subtype population. They therefore postulated that/31adrenergic receptors are true postsynaptic receptors, whereas/32 receptors are confined to glial cells and blood vessels. U'Prichard et al. ~4 confirmed these results for cerebral cortex and also demonstrated that in cerebellar cortex the sensitivity of fl2-adrenergic receptors could be altered after lesion of the ascending norepinephrine bundle. We recently obtained a homogeneous population of glial cells of the astrocyte type in primary culture 2 containing a highly potent fl-adrenergic-sensitive adenylate cyclase. The experiments reported here were designated to determine whether the /3adrenergic receptors involved could be classified as either fll and/32 subtypes. MATERIALS AND METHODS Cell cultures, homogenate preparations and adenylate cyclase assays were done as previously described z. Fluphenazine (t0-SM) was present in adenylate cyclase assays.
(3H]dihydroalprenolol ([3H]DHA) binding Cells from 10 dishes (diameter 10 cm) were rinsed 3 times with 9%0 NaC1, scraped off with a rubber policeman and centrifuged. The pellet was homogenized in 5 ml of 50 mM Tris.HC1 pH 7.8 using a teflon-glass homogenizer and centrifuged at 47,000 × g for 20 rain. The pellet was washed once more and finally resuspended in 1.6 ml of Tris.HCl pH 7.8. Aliquots of 10 #1 were incubated for 30 min at 30 °C with [~H]DHA, 50 mM Tris.HCl, pH 7.8 and other drugs to a total volume of 100/A. Non-specific binding was determined in the presence of (--) alprenolol (5 × 10-6 M). At the end of the incubation period, samples were diluted with l ml ice-cold 50 mM Tris-HC1, pH 7.8, filtered through a glass fiber filter (GF/B Whatman) and washed with 5 × 4 ml of the same medium.
Chemicals (--) Propranolol and (~_) practolol (ICI), ( - ) alprenolol (Hassle) ( ± ) butoxamine (Burrough), (--) isoprenaline, (--) adrenaline, (--) noradrenaline (Sigma), salbutamol (Allen and Hanbury's), hydroxybenzylisoproterenol was donated by J. Hanoune and was from Phillips-Durphar.
Computer fit of the data In our computer analysis we assumed like other authorsZ, 8 that there are only
153 two fl-adrenergic receptor subtypes and that interaction of drugs with receptors follows simple mass action kinetics. The formula used for fitting the adenylate cyclase dose-response curves and the [aH]DHA binding displacement curves was: F
F
Y -- N1
+ N2 F + Ks~ (1 + I/K~I or Kay)
F + Ks~ (1 + I/K~ or Ka2)
For the adenylate cyclase dose-response curves Y = adenylate minus basal activity, N1 and Nz adenylate cyclase activities due to the stimulation of ill- and fl2-adrenergic-sensitive adenylate cyclase respectively; F = agonist concentration, Ks~ and I ~ = apparent affinity of the agonist for fll and fl2-adrenergic-sensitive adenylate cyclases respectively, I = antagonist concentration, K~ and K~2 ---- apparent inhibition constants of the antagonist for ill- and fl2adrenergic receptors respectively. For the EaH]DHA binding displacement curves Y = specifically bound [aH]DHA, N1 and N2 -- maximal binding capacities of ill- and flz-adrenergic receptors respectively, F = free [aH]DHA concentration, Ks, ---Ks~ = affinities of [3H]DHA for both ill- and fl2-adrenergic receptors, I = displacing drug concentration, Kal and Ka~ = affinities of displacing drug for fll and f12 adrenergic receptors respectively. Minimization of E (Y e x p - Y theoretical) 2 is computerized by a non-linear regression analysis using a CDC 6600 computer (INs Pa Paris 6~me). RESULTS AND DISCUSSION
fll and fl2-adrenergic sensitive adenylate cyclase in glial cell primary cultures Several fl-adrenergic agonists stimulated the adenylate cyclase in primary cultures of glial cells (Fig. 1). When the potencies of the different drugs were compared Primary culture of Glial cells
._c E o 40(
[] Iso~renatine • HydroxybenzylIsoproterenol • Adrenaline 0 Nor -Adrenaline Salbutamol 10( u
=
0/_•/10I .8
1~.7
10 ~1 AGONIST (M)
I 10.5
10 "4[
Fig. 1. Adenylate cyclase activation by fl-adrenergicagonists in glial cell homogenates. Basal adenylate cyclase activity was 85 pmol/20 min/mg protein.
154 on the basis o f the K a a p p (agonist c o n c e n t r a t i o n giving half m a x i m a l stimulation), the o r d e r was ( - - ) isoprenaline :> h y d r o x y b e n z y l i s o p r o t e r e n o l :> adrenaline ~- salb u t a m o l ~ n o r a d r e n a l i n e (Fig. 1). The m e a n K a a p p values are given in Table I. S a l b u t a m o l was a partial agonist o f the fl-adrenergic-sensitive adenylate cyclase. The m a x i m a l stimulation p r o d u c e d by this ligand was 33 ~ 6 . 4 ° / (n =:= 3) o f that o f adrenaline. Simple analysis o f these results indicates t h a t a h o m o g e n e o u s p o p u l a t i o n o f either fl~- a n d fl2-adrenergic-sensitive a d e n y l a t e cyclase was n o t present. In the presence o f fl2-adrenergic-sensitive a d e n y l a t e cyclase only, we expected t h a t : (1) h y d r o x y b e n z y l i s o p r o t e r e n o l would be m o r e p o t e n t t h a n isoprenalinel,5,13 b u t this was not the case (Table I, Fig. l ) ; and (2) that the K~app for adrenaline w o u l d be 8-10 times lower than the the K a a p p for n o r a d r e n a l i n e s a n d not, as we observed, only a little lower (Table I). On the other hand, in the presence o f fl~-adrenergic receptors only isoprenaline was expected to be 8-10 times more p o t e n t than h y d r o x y b e n z y l i s o p r o t e r e n o l l , a3 and s a l b u t a m o l to be inactive 1,7 which was not the case. The shape o f the d o s e TABLE I Specificity pattern of the [J-adrenergic-sensitive adenylate cyclase in glial cells - - relative proportions of ill- and fl2-adrenergic receptors coupled with adenylate cyclase
Values are mean ± S.E.M. of 3-6 experiments. Drugs Agonists (--) Isoprenaline (_q_) Hydroxybenzylisoproterenol (--) Adrenaline on both fix and [32receptors on f12receptors on fll receptors ( ~ ) Salbutamol (--) Noradrenaline Antagonists (--) Propranolol ( ± ) Butoxamine C~) Practolol on f12receptors o n [~1 receptors
Kaapp or Ki (nM)
[h [?~ ( '% of the total)
43 ± 5* 75 -j 13" 656 160 1200 690 920
J 57* ± 117"* ± 340'* ± 300* -~ 300*
42 ~ 17"** 58 5 17"**
4± 1 4700 ± 1200§ 28,900 ± 3990§§ 556 -- 196§§
45 ± 10§§§ 55 ± 10§§§
* The Kaapp is the concentration of agonist giving half maximal activation. * * The Kaapp for ill- and fl2-adrenergic-sensitive adenylate cyclases is given by the complete analysis of the adrenaline dose-response curve described in Fig. 2. *** The percentage of ill- and fl2-adrenergic-sensitive adenylate cyclases is given by the computer analysis of the adrenaline dose-response curve described in Fig. 2. § The Ki are calculated from Hofstee plots similar to those in Fig. 3. §§ The K~of practolol for/51- and fl2-adrenergic-sensitive adehylate cyclases are given by the computer analysis of the practolol curve described in Fig. 3. §§§ The percentage of ill- and fl2-adrenergic-sensitive adehylate cyclases are given by the computer analysis of the practolol curve described in Fig. 3.
155 Primary culture of Glial cells .% • ©
• ___
Isoorenaline Adrenaline Nor-adrenaline
.~40C
40(
.c E
Adrenaline Theoretical I~2 Vmax = 118.5 _ KAapp= 5 x 10-t~ M I~1 Vmax = 276.5 KAapp =10-6 M
LJ
// 200f
0
Q,
/
/ / ,Q
•
/ill.-" l l - ~
/
E
?,"
>
0 ~)
10-5
5 xl~ 8 10x10.8 15x10~ V-b/[Adrenaline or Nor-Adrenaline] i
25 x10'8 V-b/[ISO]
I
50 x 10"8
~4-
Adrenaline [M] }IL
Fig. 2. Heterogeneity of the adrenaline dose-activation curve. Presence of fla- and fl2-adrenergicsensitive adenylate cyclase in glial cell homogenates. Left panel: Hofstee plots of isoprenaline, noradrenaline and adrenaline dose-response curves (data are from Fig. 1) V--b = adenylate cyclase activity above basal level. The slope gives the Kaapp of the drug. ISO = isoprenaline. Right panel: adenylate cyclase activation by adrenaline. Points are experimental data taken from Fig. 4. The dotted line is the theoretical line obtained by computer fitting (see Material and Methods). The amount of fll and fiesensitive adenylate cyclases and the Kaapp of adrenaline for these enzymes are given.
response curves for hydroxybenzylisoproterenol and adrenaline indicates that they do not correspond to simple mass-action kinetics, since they extend over more than two orders of magnitude (Fig. 1). The Hofstee plot of the dose-response curve for adrenaline confirmed this heterogeneity and showed two clear slopes (Fig. 2). Furthermore, non-linear regression analysis of the adrenaline dose-response curves was carried out by computer. The best fits were obtained with a model involving two sub-populations of fl-adrenergic receptors with different Kaapp and maximal velocities. An example of such fitting is given in Fig. 2. The mean K~app and percentages for the two sub-populations are given in Table I. We deduced from the above results that the fl-adrenergic sensitive adenylate cyclase with the highest affinity for adrenaline (Kaapp = 160 J: 117 nM) corresponded to a f12 sub-population since in meningeal cells which contained only fl2-adrenergic receptors (see below) the Kaapp for adrenaline was 213 ~ 79 nM (Table II) and the one with the lowest affinity (Kaapp = 1200 :k: 340 nM) to a fll sub-population. Computer analysis indicates that these two subpopulations represent 41 -+- 17 ~ and 58 :[: 17 700 of the total fl-adrenergic-sensitive adenylate cyclase respectively. The heterogeneity o f the hydroxybenzylisoproterenol dose-response curves substantiated this conclusion. The Hofstee plots of the curves for isoprenaline and noradrenaline were linear (Fig. 2). This is not surprising since isoprenaline is known to be equipotent in both ill- and fl2-adrenergic receptors, whereas noradrenaline has been reported to be only twice as potent on fll as on f12 receptors in the brain1, a3.
156 TABLE II Specific of the fl-adrenergic-sensitive adenylate cyclase in meningeal cells The Kaapp are calculated from Hofstee plots similar to those in Fig. 4. Values are the mean ± S.E.M. of 3-5 experiments. Drugs Agonists (2) Hydroxybenzylisoproterenol (--) Isoprenaline (--) Adrenaline (2) Salbutamol (--) Noradrenaline Antagonists (--) Propranolol (~:) Butoxamine (A) Practolol
Kaapp (riM) or K~ 28 ± 8 58 ~ 3 213 ~: 79 765 -~ 189 3200 :t:: 300 2.2 ~: 0.6 1900 zk 170 43,000 5:860
Several fl-adrenergic antagonists have been shown to have a higher affinity for either fll and flz receptors, whereas others have the same affinity for both. Examples of the former class are practolol and butoxamine 8 reported to be ill- and fl2-selective antagonists respectively, whereas propranolol is an example of the latter class. Fig. 3 shows that the dose-activation curve for isoprenaline was monophasic (the Hofstee plot gave one slope). In the presence of constant propranolol or butoxamine concentrations, the dose-response curves for isoprenaline shifted to the right. The Hofstee plots of these shifted curves also gave one slope, indicating that the/31- and fl2adrenergic receptors coupled with an adenylate cyclase in glial cells showed no selectivity for propranolol or butoxamine. Butoxamine which has been reported to be more potent on flz than on fll adrenergic receptors in vivo 8 showed no such selectivity in our experiments. However, there are reports of poor fl2-selectivity by this drug at receptor level 3,7. Note that a 2-3-fold difference in the affinities of butoxamine for the two subpopulations would be difficult to detect. The dose-response curve for isoprenaline in the presence of a constant concentration ofpractolol was not parallel to the other curves and gave a curvilinear line when plotted according to Hofstee (Fig. 3). Regression analysis of this curve was done by computer. The best fit was obtained with a model assuming that isoprenaline has the same affinity for ill- and fle-adrenergic sensitive adenylate cyclases and in which practolol discriminated between them. The computer-determined Ki values for illand fl2-adrenergic-sensitive adenylate cyclases were 6.5 x 10-7 M and 1.2 × 10-~ M and the percentage of the two populations were respectively 48 % and 58 ~o of the total (Fig. 4). Mean values are given in Table I. fl2-adrenergic-sensitive adenylate cyclase in meningeal cell primary cultures The agonist specificity pattern of the fl-adrenergic-sensitive adenylate cyclase in meningeal cells was clearly different from that in glial cells. The order of potency was
157 Primary culture of glial cells 10£
~'~ /
o5£
•
ISO+Butoxamine 12 xlO-5M)
0
ISO+Practolol (2 xlO-5M)
i
!5x10-8M)
"5 ae ISO (MI
~C~'~ 13. '5
\~t,~,~ \.
",,~
~
[]
ISO KAapp = 6x10-8M
•
Butoxamine KI= 3xlO-6M
©
Praetolol (exl3erimental)
.... (Theoretical)
O 5£ .6;
N,~.~/.~
I
5 x 10"8
,1.42%
I
~
(al
I
I=
(O,@,A)
10 x 10~8
I
l x 10+8
2xl(f8
A/[,s0] Fig. 3. Competitive inhibition by butoxamine, practolol and propranolol of adenylate cyclase stimulation by isoprenaline in glial cell homogenates. Upper panel shows the dose-response curves. Adenylate cyclase activities are expressed in percentage of the maximal increase in cyclic AMP due to the presence of saturating concentrations of isoprenaline (140 pmol/20 min/mg protein). Lower panel shows the Hofstee plots of the dose-response curves. Note that the plots of the isoprenaline dose-activation curves obtained in the absence ([]) or presence of either butoxamine (&) or propranolol (@) were linear, but the plot of the isoprenaline dose-activation curve obtained in the presence of practolol (O) was curvilinear. The points are experimental data whereas the dotted line is the theoretical line obtained by computer fitting. For the latter, the apparent affinity of isoprenaline was taken from the Hofstee plot and was equal to 6 x 10-a M. The percentage of ill- and flz-sensitive adenylate cyclases and the K~ of the enzymes for practolol are given.
158 Primary culture of meningeal layer J
,<
2. 0 c_
, O/ 5£
/
.,~/
•
ISO + Butoxamine (2x10-5M)
~
O
ISO + Practolol
-
•
ISO + Propranolol (5xlO-8M)
(10-4M)
o
<~ ~6 ~e I 4"
~
ld 8
10-7
1~ ISO (M)
10-5
t
b--
10-4
10C :
Q..
ISO KAapp=6,4xlO8M
0
5O
t
t 5 x 10*8
I t
1 10 x 10*8
J
),
(: J)
15 x 10÷8
~
i
,
1 ~ I 0 ÷8
2×10"8
3x10 ~
J 2 x 10``.8
l 4 ~ 10`..8
l 6 × 10*8
~ (e,A) ~,
(C,)
% of 2L ~'lsO} Fig. 4. Competitive inhibition by butoxamine, practolol and propranolol of adenylate cyclase stimulation by isoprenaline in meningeal cell homogenates. Upper panel shows the dose-response curves. Adenylate cyclase activities are expressed in percentage of the maximal increase in cyclic A M P due to the presence of saturating concentrations of isoprenaline (100 pmol/20 min/mg protein). Lower panel shows the Hofstee plots of the dose-response curves. Note that the 4 plots were linear.
hydroxybenzylisoproterenol > isoprenaline > adrenaline > salbutamol > noradrenaline. Mean Kaapp values are given in Table II. Four observations suggested the presence of a homogeneous population of fl2-adrenergic receptors coupled with adenylate cyclase : (1) all the dose-response curves were monophasic (the Hofstee plots
159 gave only one slope, data not shown); (2) hydroxybenzylisoproterenol was more potent than isoprenaline; (3) adrenaline was 10 times more potent than noradrenaline; and (4) salbutamol was an agonist. Since the agonist specificity pattern indicated the presence of a homogeneous population of flz-adrenergic receptors in meningeal cells, we examined the effects of the antagonists. As expected (Fig. 4) the dose-response curves for isoprenaline on the presence of constant concentrations of propranolol, butoxamine or practolol were all parallel. The Hofstee representations gave only one slope for each curve (Fig. 4). Mean K~ values are given in Table II.
[aH]DHA binding to glial cell membranes confirms the presence of both fl~- and fl~adrenergic receptors Scatchard analysis showed that [aH]DHA bound to a single class of high affinity receptors in glial cell membranes. The calculated Ka values was 0.5 nM. Binding site density (Bmnx) was 145 fmol/mg protein. Displacement curves of [aH]DHA binding by propranolol and practolol on glial cell membranes were not parallel. They extended over two orders of magnitude for propral~olol and more than 3 for practolol (Fig. 5). These data fitted with computer models involving one or two classes of sites. In glial cell membranes, the best fit was obtained with a one-site model for propranolol and with a two-sites model for Propranolol - glial cells
_'~
• --
100
--&~
--'l~
o
E
Practolol ~ glial cells • L meningeal cells
O
~,L
"6
i
Glial cells I
"10 r-
x~ 50
•
B2 35%
B1 65%
\o
\
-r cl i
'''* 10-8
.... 10-7
'
10 -6 10-5 ANTAGONIST (M)
10 4
10-3
Fig. 5. Curves for displacement of [aH]DHA binding by propranolol and practolol. [3H]DHA binding was 116 and 40 fmol/mg protein in glial and meningeal cell particulate fractions respectively. [a H ] D H A concentrations in glial and meningeal cell particulate fractions were 1.6 a n d 1.3 nM respectively. Points are experimental data and dotted lines represent computer fitting. The curve for displacment of [aH]D H A binding by practolol in glial cell particulate fractions fitted the model involving two classes of sites: 65 ~ high-affinity (Ki = 1.2 × 10 -7 M) and 35 ~ of low-affinity (K~ = 8 x 10 -6 M). Curves for [aHIDHA binding displacement by propranolol in glial cells and by practolol in meningeal cells fitted the model involving only one class of sites. K~ were 1.2 x 10 -9 M and 5 x 10 -5 M respectively.
160 practolol. The Ka of practolol for the two classes of sites were 1.2 × 10-7 M and 8 × 10-6 M respectively, suggesting that they represent fit- and flz-adrenergic receptors respectivelyT,s. fll-adrenergic receptors represented 65%o of the total fl-adrenergic receptor population. For purpose of comparison [3H]DHA binding on meningeal cell membranes was done. Only one single class of sites was detected )Ka := 1.3 nM, Bmax = 80 fmol/mg protein). The curve for [3H]DHA displacement by practolol indicated the presence of only one class of sites having a Ka of 5 x 10-2 M (fl2-adrenergic receptors). Two arguments may be advanced as proof that the presence of fl~,-adrenergic receptors in glial cell primary cultures is not due to contamination by meningeal cells. Firstly, 96 ~ of the cells present in our primary cultures contained the glial fibrillary acid protein 2, meningeal cells did not contain this proteinL Secondly, the possible contamination of 4-5 ~ of these glial cells by meningeal cells, which of course cannot be excluded, would not suffice to account for the presence of 40-50 ~i of fl2-adrenergic receptors, since meningeal cells contained fewer receptors than glial cells (see above). It is evident that our cell cultures do not contain all types of differentiated glial cells which could exist in the brain. However, the finding that both fit- and fl2adrenergic receptors are present in the glial cells described here does not support the proposal of Minneman et al. 6 that glial cells contained only fl2-adrenergic receptors. Thus, two questions remain open: (1) what is the nature of the fl-adrenergic receptors present in neurons ? Previous binding 9A°,14 and adenylate cyclase experiments t in the cerebral cortex show that fl-adrenergic receptors (more than 75~.(,) are of the fll subtype, whereas our present data indicate an almost equal concentration of fit- and fl2-adrenergic receptors in glial cells. Since cerebral cortex also contains fl2-adrenergic receptors from blood vessels 11, it is tempting to speculate that most neurons' adrenergic receptors are of the fll class, however, a direct proof of this is still lacking; and (2) are fll-adrenergic receptors in glial cells capable of adaptive responsiveness ? Our experiments do not of course show whether/31- and flz-adrenergic receptors are present in a single cell or if we are dealing with two different populations of glial cells. From a physiological and biochemical point of view it is important to clarify this question. Preliminary experiments indicates that Ca glioma cell lines also have both fitand fl2-adrenergic receptors. ACKNOWLEDGEMENTS This work was supported by LA 219 from CNRS and grants from INSERM (ATP 58 78 90) and CNRS (ATP 4149). We thank Anne du Parc for manuscript preparation. REFERENCES 1 Dolphin, A., Hamon, M. and Bockaert, J., The resolution of dopamine and [-¢1-and fl2-adrenergicsensitive adenylate cyclase activities in homogenates of cat cerebellum, hippocampus and cerebral cortex, Brain Research, 179 (1979) 305-317. 2 Ebersolt, C., Perez, M. and Bockaert, J., Neuronal, glial and meningeal localizations of neurotransmitter sensitive adenylate cyclases in cerebral cortex of mice, Brain Research, 213 0981) 139-150.
161 3 Hancok, A. A., De Lean, A. L. and Lefkowitz, E. J., Quantitative resolution of beta-adrenergic receptor subtypes by selective iigand binding:application of a computerized model fitting technique, Molec. Pharmacol., 16 (1979) 1-9. 4 Lands, A. M., Arnold, A., McAuliff, J. P., Luduean, F. P. and Brown, T. G., Differentiation of receptor systems activated by sympathomimetic amines, Nature (Lond.), 214 (1967) 597-598. 5 Lefkowitz, R. J. and Williams, L. T., Catecholamine binding to fl-adrenergic receptor, Proc. nat. Acad. Sci. (Wash.), 74 (1977) 515-519. 6 Minneman, K. P., Dibner, M. D., Wolfe, B. B. and Molinoff, P. B., 81- and 82-adrenergic receptors in rat cerebral cortex are independently regulated, Science, 204 (1979) 866-868. 7 Minneman, K. P., Hegstrand, L. R. and Molinoff, P. B., The pharmacological specificity of beta-1 and beta-2 adrenergic receptors in rat heart and lung in vitro, Molec. Pharmacol., 16 (1979) 21-33. 8 Minneman, K., Hegstrand, L. R. and Molinoff, P. B., Simultaneous determination of beta-1 and beta-2 adrenergic receptors in tissues containing both receptor subtypes, Molec. Pharmacol., 16 (1979) 34-46. 9 Minneman, K. P. and Molinoff, P. B., Classification and quantification of 8-adrenergic receptor subtypes, Biochem. Pharmacol., 29 (1980) 1317-1322. 10 Nahorski, S., Heterogeneity of cerebral 8-adrenoreceptor bindingsites in various vertebrate species, Europ. J. PharmacoL, 51 (1978) 199-209. 11 Palmer, G. C., Beta-adrenergic receptors mediate adenylate cyclase responses in rat cerebral capillaries, Neuropharmacology, 19 (1980) 17-23. 12 Pittman, R. N., Minneman, K. P. and Molinoff, P. B., Ontogeny of 81- and 82-adrenergic receptors in rat cerebellum and cerebral cortex, Brain Research, 188 (1980) 357-368. 13 U'Prichard, D. C., Bylund, D. B. and Snyder, S. H., (~:) [all]epinephrineand (--) [aH]dihydroalprenolol binding to 81 and 82 noradrenergic receptors in brain, heart and lung membranes, J. biol. Chem., 25 (1978) 5090-5102. 14 U'Prichard, D. C., Reisine, T. D., Yamamura, S., Mason, S. T., Fibiger, H. C., Ehlert, F. and Yamamura, H. 1., Differential supersensitivity of 8-receptor subtypes in rat cortex and cerebellum after central noradrenergic denervation, Life Sci., 26 (1980) 355-364.