dovnm/o?
J. Neural Transmission 53, 49-57 (1982)
Netwal a~mnamlssfon (g) by Springer-Verlag 1982
Testosterone Decreases fl-Adrenoceptor Sites in Rat Pineal Gland and Brain I Maria I.Vacas 2, P.R. Lowenstein, and D . P . Cardinali 2 Centro de Estudios Farmacol6gicos y de Principios Naturales (CEFAPRIN), Buenos Aires, Argentina With 4 Figures Received July 7, 1981
Summary Testosterone administration to orchidectomized rats brought about a significant, 55% decrease offl-adrenoceptor sites in the pineal gland, assessed from the specific binding of radioactive dihydroalprenolol (DHA). The changes in density of binding sites were not accompanied by significant modifications of the Kd. FSH or LH treatment of acutely castrated animals did not affect pineal fl-adrenoceptor binding. The depressive effects of testosterone in fl-adrenergic receptors were also observed in crude membrane fractions of medial basal hypothalamus and cerebral cortex. Sympathetic denervation of the pineal gland by superior cervical ganglionectomy did not abolish the changes in pineal fl-adrenoceptor density caused by testosterone. Hormone effects did not depend on a direct effect of the hormone on fl-adrenoceptor sites because testosterone did not compete with [3H]-DHA for the binding sites, in vitro. These results suggest that testosterone depresses pineal fl-adrenergic sites by acting mainly on postsynaptic sites. Key words: Pineal, superior cervical ganglion, fl-adrenoceptors, testosterone, FSH, LH.
i These studies were supported by grant no. 6638 from Consejo Nacional de Investigaciones Cientificas y T&nicas de la Repfblica Argentina (CONICET). 2 Established Investigator, CONICET. 4 Journal of Neural Transmission 53/I 0 3 0 0 - 9 5 6 4 / 8 2 / 0 0 5 3 / 0 0 4 9 / $ 01.80
50
Maria I. Vacas, P. R. Lowenstein, and D. P. Cardinali:
Introduction In mammals the pineal gland is an endocrine organ with some of the properties of a neuroendocrine transducer. One of its major functions is to convert an input of neural signals, namely norepinephrine (NE) released from the sympathetic nerve fibers originating in the superior cervical ganglia (SCG) to a hormonal output, ke., melatonin and presumably polypeptides (Nit et al., 1978). In addition an increasing amount of evidence indicates that the pineal gland is a site of feedback regulation by hormonal signals carrying information from the internal milieu (Cardinali, 1981). A number of hormone receptors (for estrogens, androgens, progestagens, Cardinali et aL, 1975 b; Vacas et al., 1979) are present in the gland, and at least for estradiol and progesterone a direct effect on melatonin release was shown in pineal organ cultures (Mizobe and Kurokawa, 1976; Wilkinson and Arendt, 1978; Cardinali, 1981). Hormones also affect the activity of the sympathetic neurons innervating the pineal gland. Evidence in this respect includes the modification of NE turnover in pineal nerve endings after estradiol, testosterone, FSH or LH administration to castrated rats (Cardinali et al., 1975a; Cardinali and Vacas, 1979), and the changes in the synthesis and metabolism of NE in the SCG following estradiol, testosterone, gonadotrophin or prolactin treatment (Cardinali and Vacas, 1979; Cardinali, 1981). The present experiments were undertaken to examine the effect of testosterone, FSH or LH injection to orchidectomized rats on the first step of the action of NE on the pinealocytes, that is, its binding to fl-adrenergic receptors. The potent fl-adrenergic antagonist 3H-dihydroalprenolol (DHA) labeled to high specific activity was used as a radioligand for binding studies.
Methods Wistar male and female rats (150-180g) were kept under light from 7 a.m. to 9 p.m. daily and were given access to Purina chow and water ad libitum. Bilateral superior cervical ganglionectomy (SCGx) or sham-operation were performed 7 days before treatment. For testosterone experiments male rats castrated 72 hours earlier were treated for 3days with 2 daily s.c. injections (at 9 a.m. and 5 p,m.) of 400/ag of testosterone dissolved in 0.2 ml of saline : ethanol (1 : 1). An additional injection of 400/ag was given on the day of sacrifice, 3 hours before killing. Control animals received only the vehicle. The effect of FSH or LH were examined in acutely orchidectomized rats. Two injections of 100/ag of ovine FSH or LH (kindly provided by the
Testosterone Decreases fl-Adrenoceptor Sites
51
National Pituitary Agency, NIH, Bethesda, MD) were given 3 hours and 18 hours after castration and the animals were killed 3 hours later. Controls received the vehicle alone (lml of saline). At the time of sacrifice (10 a.m. to 12 a.m.) the rats were killed by neck fracture, the brains were quickly removed and the pineal glands were dissected out. In some experiments a portion of the parietal cerebral cortex and the medial basal hypothalamus (MBH) were dissected out as previously described (Vacas and Cardinali, 1980). Binding studies were carried out according to standard procedures (Williams and Lefkowitz, 1978). fl-adrenergic receptors in pineal gland were assayed by [3H]DHA (dihydroalprenolol hydrochloride, 1-propyl-l,2,3,-3H -, 51.1Ci/mmol, New England Nuclear) binding to nuclei-free homogenates prepared in 80 mM Tris-HC1 buffer containing 6 mM Mg S04 and 0.6 mM EGTA. Pineal homogenates were centrifuged at 900 g for 10 min at 0 ~ and the supernatant was used for binding studies, fl-adrenergic receptor binding in MBH and cerebral cortex was assessed by [3H]DHA binding to P2 membrane fractions (Williams and Lefkowitz, 1978). The final pellet was resuspended in 75 mM Tris-HC1 buffer, pH 7.4, containing 25 mM MgC12. The typical binding reaction in pineal gland and MBH was carried out by triplicate in a total volume of 110/al containing 100/al of tissue suspension (70-120/ag of protein), 2--30nM of [3H]DHA (dissolved in 5% ethanol) plus 10/aM of propranolol (dissolved in 5/al of buffer) or 5/al of buffer. For parietal cerebral cortex, 300/al^of tissue suspension (150-250/ag of protein) was incubated with 2-30 nM [~H]DHA in the presence or absence ofl0/aM of propranolol in a total volume of 310/al. In each experiment a parallel series of reaction mixtures containing no tissue fraction was run as controls. Incubations were carried out at optimal conditions (10min at 37~ Separation of the "free" from "bound" ligand was achieved by rapid filtration through Whatman GFB glass fiber filters which were presoaked in cold buffer. After washing with 10 ml of ice cold buffer the radioactivity left on the filter was measured by liquid scintillation spectrometry. A toluene phosphor solution containing 30% Triton X--114 was used as a scintillator fluid. Quenching was corrected by automatic external standardization; each sample was counted long enough to yield a counting error less than 3 %. In all the experiments the difference between the total radioligand bound and the radioligand bound in the presence of competitor (non specific binding) was taken as the amount of specific radioligand binding. Specific binding ranged between 50 and 85%. Binding of radioligand to the filters in the absence of tissue was less than 1%. Repeated estimates of specific binding capacity of pineal and brain membranes yielded coefficients of variation of 6 to 10%. Equilibrium binding constants were calculated by Scatchard analysis after transformation of data; line slope and intercepts were determined by regression analysis. Differences in slopes and intercepts between groups were analyzed by an analysis of covariance (Snedecorand Cocbran, 1967). Protein concentration was determined by Lowry et al.'s procedure (1951) by using bovine serum albumin as a standard. Each experiment was repeated at least twice. 4 ~ 84
52
Maria I.Vacas, P. R. Lowenstein, and D. P. Cardinali: PINEAL
GLAND
400
g
5C
o
30C
"~.
Kd=8,2 nM
-6 E s
\o>
20C
0 m
g
I OC
:]z i
I0
20
[3H-DHA]
30
I00
nM
200
~H-DHA
300
400
500
BOUND(fmol/mg prot)
Fig. I. Effect of testosterone on ,8-adrenoceptors in the pineal gland of orchidectomized rats measured by the binding of [3H]-DHA as described in Methods. Data are shown in Scatchard plots; slopes and intercepts were calculated by regression analysis and analyzed statistically by means of an analysis of Covariance. Testosterone decreased significantly Bmax (p<0.05) without changing the Kd
PINEAL
GLAND
60
c
xN~NNx, ~ e
0 vehicle 9 LH
5c
9
g
4c
%.
3o Kd = 6.1 nM LH
o
Fs.
o \\
~'NQ,. N~\
/
2s
Kd = 6.0 nM voh
I0 I to
\'%,
I
0
I00 ~H- DHA
200
300
Bound (fmol /mg ,prot )
Fig. 2. Effect of LH and FSH on ,8-adrenoceptors in pineal gland oforchidectomized rats. Experimental design and analysis of data as described in Methods. There were no differences in Kd's or Bmax between groups
Testosterone Decreases fl-Adrenoceptor Sites
53
Results The effects of different hormonal treatments on pineal fl-adrenergic receptors are shown in Fig. 1 and 2. A m o n g the 3 hormones tested (testosterone, FSH and LH) only testosterone affected [3H]DHA binding to pineal 900 g supernatants. Testosterone resulted in a significant 55% depression of the maximal number of binding sites without changing their affinity for the radioligand (Fig. 1). A second experiment was carried out to examine comparatively the effect of testosterone on fl-adrenergic receptors in 2 other brain regions, the MBH and parietal cerebral cortex (Fig. 3). In both tissues testosterone treatment depressed to the extent of 34-37% the number of fl-adrenoceptor sites without affecting their Kd's. The addition of testosterone or of its 5 a-reduced metabolite 5 a-dihydrotestosterone (DHT) (up to 10-5 M) to the incubation medium of P2 cerebral cortex membranes did not affect the binding reaction of [3H]DHA (results not shown). Administration of testosterone to rats whose pineals had been priorly denervated by SCGx still affected pineal fl-adrenoceptors (Fig.4). A significant 32%-diminution of binding sites was found, Kd's values being not changed by hormone treatment.
HMB
E
o
~-
bo
Cerebral
Cortex
o Vehicle
S ~ o~ o
o
4C
9 testosterone
30
9
Kd = 34nM 0 <~
g
o
o
2C
~o
o o d= 11.4nM
Kd=2-5nM Kd= I0
i
E I00
200
300
3 H - DHA
i 400
0
BOUND ( f m o l / m g
I00
200
300
400
500
pro1}
fl-adrenoceptor in medial basal hypothalamus (MBH) and parietal cerebral cortex of orchidectomized rats. Data are shown in Scatchard plots. Testosterone decreased significantlyBmaxin both tissues (p <0.05) without changing the Kd's Fig. 3. Effect o f testosterone on
54
Maria I. Vacas, P. R. Lowenstein, and D. P. Cardinali: Denervated Pineol Gland 40
3o
~~~176 ~176o" ~ / Kd: 5,lnM O0~ I0
o
Kd:43nM
I~ I00
~N~I 200
SH-DHA BOUND(frnol/mg prot}
Fig.4. Effectof testosterone on pineal fl-adrenoceptorsin superior cervicalganglionectomized(Gx), orchidectomizedrats. Gx was performed one weekbeforedrug or vehicle injection. Experimentaldesignand analysisof data as describedin Methods. Testosterone decreasedsignificantlyBmax(p<0.05) but there was no difference in Kd's between groups
Discussion The pineal gland can be considered a target organ for testosterone. In mammals and non-mammals various relevant metabolic activities of the pineal gland (including melatonin synthesis) are affected by androgen administration (for references see Cardinali, 1981). Pineal androphilic receptors are detectable by biochemical and autoradiographic procedures ( Cardinali et al., 1975 b; Stumpf and Sar, 1977); as the estrophilic receptors, they are under partial control by the sympathetic nerves. The foregoing results indicate that testosterone, but not FSH or LH, decreased pineal fl-adrenoceptor sites in orchidectomized rats. These data together with the prior demonstration that estradiol treatment did not affect pineal fl-adrenoceptors in spayed rats (Vacas and Cardinali, 1980) suggest that testosterone may be unique among the reproductive hormones in bringing about changes in pineal fl-adrenergic receptor density. The activity of testosterone thus appears to be independent on the aromatization to estradiol or on altered plasma gonadotrophin titers. Testosterone increased (Cardinali et al., 1975 a) whereas LH or FSH decreased [3H]NE turnover in rat pineal gland (Cardinali and Vacas, 1979). Therefore a possible interpretation for the present
Testosterone Decreases fl-Adrenoceptor Sites
55
results is that the increased interaction of NE with pineal receptors triggers a decrease in the number of binding sites as a part of the complex mechanism leading to desensitization of the synaptic pathway. That this decrease in [3H]DHA binding was due to true decreases of binding sites rather than to competition of the radioligand with endogenous NE remaining at binding sites is indicated by the lack of effect of testosterone treatment on receptor Kd values; moreover the binding technique used allowed total exchange of occupied binding sites (Williams and Lefkowitz, 1978). It should be stressed however that the explanation that testosterone depressed pineal fl-adrenoceptor density as a consequence of increased NE release from nerve terminals is not entirely satisfactory because estradiol, another hormone which accelerated [3H]NE turnover from pineal nerve endings in oophorectomized rats (Cardinali et al., 1975 a), was unable to affect pineal [3H]DHA binding. Testosterone could affect the post-synapsis directly either by binding to the fl-adrenergic receptors themselves or by triggering a post-synaptic mechanism leading to receptor regulation. The first of these two possibilities did not receive experimental support since neither testosterone nor DHT (up to 0.05 mM) caused any alteration in [3H]DHA binding to brain membranes in vitro. In order to examine whether the effect of testosterone in vivo has a post-synaptic component we studied the denervated pineal gland of SCGx animals. Even after a complete degeneration of the sympathetic nerve terminals, testosterone injection depressed fl-adrenergic receptor sites. Therefore the changes induced by testosterone in pineal fl-adrenoceptors probably depend upon a post-synaptic site of action of the hormone. Other pineal effects of testosterone also occur in the absence of intact sympathetic nerves; for example, treatment of SCGx rats with the hormone increased pineal serotonin levels (Vacas and Cardinali, 1979). There are controversial reports as to whether a pharmacological antagonism between testosterone and fl-adrenergic agonists occurs at the level of the pineal gland. Isoproterenol injection to SCGx on sham-operated rats increased pineal protein synthesis, an effect blocked by prior administration of testosterone proprionate (Cardinali et al., 1976). On the other hand neither orchidectomy nor testosterone injection affected the responsiveness of pineal adenylate cyclase to NE (Weiss and Crayton, 1970); orchidectomy but not testosterone was found to affect the nocturnal rise of pineal N-acetyltransferase activity in rats (Rudeen and Reiter, 1980). How this occurs in the presence of diminished number of fl-adrenoceptor sites awaits further investigation. That an enhanced pineal response to fl-adre-
56
Marfa I. Vacas, P.R. Lowenstein, and D. P. Cardinali:
nergic agonist may co-exist with normal or low fl-adrenoceptor density is indicated by the comparison of binding data of control and SCGx animals in Fig. 1 and 4. Denervated pineal glands, which are unequivocally supersensitive to fl-adrenergic agonists, exhibited fewer fl-adrenoceptor sites. In partial agreement with this Zatz (1977) and Cantor and co-workers (1980) failed to detect significant increases in pineal fl-adrenoceptor sites after ganglionectomy in rats. Our present results also indicate that testosterone treatment diminished fi-adrenergic receptor density in MBH and cerebral cortex. Previous observations from this and other Laboratories (Wilkinson et al., 1979; Vacas and Cardinali, 1980) indicated that following estradiol injection to spayed rats an increase in MBH, but not in cerebral cortex fl-adrenoceptors is found. Therefore from the two aspects of testosterone metabolism significant for steroid receptor occupation in brain, namely aromatization to estrogens or d4--5-reduction to 5 u-reduced metabolites (McEvaen, 1980) the former can be ruled out. To what extent the changes in fl-adrenoceptors density brought about by testosterone play a role in hormone mechanism of action in brain awaits further elucidation.
References Cardinali, D. P.: Hormone effects on the pineal gland. In: The pineal gland, Vol. 1: Anatomy and biochemistry (Reiter, R.J., ed.), pp. 243-272. Boca Raton, Fla.: CRC Press. 1981. Cardinati, D. P., Gomez, E., Rosner,J. M.." Changes in 3H leucine incorporation into pineal proteins following estradiol or testosterone administration. Involvement of the sympathetic superior cervical ganglion. Endocrinology 98, 849--858 (1976). Cardinali, D. P., Nagle, C. A., Gomez, E., Rosner, J. M.: Norepinephrine turnover in the rat pineal gland. Acceleration by estradiol and testosterone. Life Sci. 16, t717--1724 (1975 a). Cardinali, D. P., Nag& C. A., Rosner, J. M.: Control of estrogen and androgen receptors in the rat pineal gland by catecholamine transmitter. Life Sci. 16, 93--106 (1975 b). Cardinali, D.P., Vacas, M. I.: Norepinephrine turnover in pineal gland and superior cervical ganglia. Changes after gonadotrophin administration to castrated rats. J. Neural Transm. 45, 273-284 (1979). Cantor, E.H., Greenberg~ L.H., Weiss, B.: Effect of long-term changes in sympathetic nervous activity on the beta-adrenergic receptor-adenylate cyclase complex of rat pineal gland. Molec. Pharm. 19, 21-26 (1981). Lowry, 0., Rosebrough, N., Farr, A., Randall, R.: Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-275 (1951).
Testosterone Decreases ~-Adrenoceptor Sites
57
McEwen, B. S.: Binding and metabolism of sex steroids by the hypothalamispituitary unit: physiological implications. Ann. Rev. Physiol. 42, 97-110 (1980). Mizobe, F., Kurokawa, M.: Enhancement of hydroxyindole-0-methyltransferase and DNA dependent RNA polymerase activities induced by oestradiol in rat pineals in culture. Eur. J. Biochem. 66, 193-199 (1976). Nir, L, Reiter, R.J., Wurtman, R.J. (eds.): The pineal gland. (Journal of Neural Transmission, Suppl. 13.) Wien-New York: Springer. 1978. Rudeen, P. K., Reiter, R.J.: Depression of nocturnal pineal serotonin N-acetyltransferase activity in castrate male rats. J. Neural Transm. 48, 1--8 (1980). Snedecor, G.W., Cochran, W.G. (eds.): Statistical methods, 6th ed. Ames, Iowa: The Iowa State University Press. 1967. Stumpf, W.E., Sar, M.: Steroid hormone target cells in the periventricular brain: relationship to peptide hormone producing cells. Fed. Proc. 36, 1973-1983 (1977). Vacas, M.L, Cardinali, D.P.: Effects of castration and reproductive hormones on pineal serotonin metabolism in rats. Neuroendocrinology 28, 187-195 (1979). Vacas, M.I., Cardinali, D.P.: Effect of estradiol on ce- and fl-adrenoceptor density in medial basal hypothalamus, cerebral cortex and pineal gland of ovariectomized rats. Neuroscience Lett. 17, 73--77 (1980). Vacas, M.I., Lowenstein, P., Cardinali, D.P.: Characterization of a cytosol progesterone receptor in bovine pineal gland. Neuroendocrinology 29, 84-89 (1979). Weiss, B., Crayton, B.: Gonadal hormones as regulators of pineal adenyl cyclase activity. Endocrinology 87, 527-533 (1970). Wilkinson, M., Arendt, J.: Effects ofoestrogen and progesterone on rat pineal N-acetyl transferase activity and melatonin production. Experientia 34, 667--669 (1978). Wilkinson, ill., Herdon, H., Pearce, M., Wilson, C.: Radioligand binding studies on hypothalamic noradrenergic receptors during the estrous cycle or after steroid injection in ovariectomized rats. Brain Res. 168, 652-655 (1979). Williams, L.T., Lefkowitz, R.J.: Receptor binding studies in adrenergic pharmacology (Costa, E., Gessa, G.L., Sandier, M., eds.). New York: Raven Press. 1978. Zatz, M.: Effects of cholera toxin on supersensitive and subsensitive rat pineal glands: regulation of sensitivity at multiple sites. Life Sci. 21, 1267--1276 (1977). Authors' address: Dr. D.P. Cardinali, CEFAPRIN, Serrano 665/669, 1414 Buenos Aires, Argentina.