Estradiol Selectively Regulates A,,-Noradrenergic Receptors in Hypothalamus and Preoptic Area

The Journal of Neuroscience, October 1992, 12(10): 3889-3876

Estradiol Selectively Regulates a,,-Noradrenergic Receptors in the Hypothalamus and Preoptic Area

Nicolas Petitti, George B. Karkanias, and Anne M. Etgen Departments of Psychiatry and Neuroscience, Albert Einstein College of Medicine, Bronx, New York 10461

We previously demonstrated that estradiol administered in of CAMP synthesis is correlated with a modest (20-30%) ele- viva elevates the number of a,-adrenoceptors in preoptic vation in the total number of specific3H-prazosin binding sites area (POA) and hypothalamic membranes from ovariecto- in hypothalamic and POA membranes(Etgen and Karkanias, mized female rats and potentiates a, receptor augmentation 1990). Interestingly, progesterone(P) abolishes(Y, receptor po- of ,&adrenoceptor-stimulated CAMP formation in slices from tentiation of CAMP synthesis in slicesfrom E,-primed female these brain regions. Present studies examined (1) if estradiol rats (Petitti and Etgen, 1989, 1990, 1992) without affecting the selectively regulates any cY,-adrenoceptor subtype, and (2) number or affinity of 3H-prazosinbinding sites(Etgen and Kar- which ,x, receptor subtype mediates the augmentation of kanias, 1990). CAMP synthesis. Hypothalamic and POA membranes from Norepinephrine (NE) receptors have long been divided into estradiol-treated rats, when compared to ovariectomized rats, p, o(,, and c+ subtypes.However, it is now clear that thesethree had modestly (3040%) but significantly elevated numbers classesof adrenergic receptors each can be subdivided further of 3H-prazosin ((Y,) binding sites. Estradiol affected neither basedupon both pharmacologicaland molecular biological cri- the number of a, receptor sites in frontal cortex nor the teria (for recent reviews, seeLomasney et al., 1991; Szabadiand affinity of 3H-prazosin binding in any brain region examined. Bradshaw, 1991). The /3 receptorscan be subdivided into 0, and Results of binding studies conducted in the presence of p2 receptor subtypes (Minneman and Molinoff, 1980), the 01, chlorethylclonidine, a selective, irreversible inactivator of the receptors can be subdivided into alA and ollB(and perhapsa,,-) (Y,~ receptor subtype, indicated that the estrogen-dependent receptor subtypes(McGrath, 1982; Morrow and Creese,1986; increase in total a, binding sites in POA and hypothalamic Minneman et al., 1988; Terman et al., 1990; Harrison et al., membranes was attributable to a selective, five- to sixfold 199 l), and the LYEreceptors can be subdivided into at least aZA, increase in (Y,~ receptor number. Progesterone had no mea- Q, and 01~~receptor subtypes(Bylund, 1988; Bylund et al., 1988; surable effects on a, receptor binding. Blockade of alB re- Harrison et al., 1991). Morrow and Creese (1986) originally ceptors with chlorethylclonidine eliminated phenylephrine reported that two classesof (Y, receptor binding sitesin rat ce- augmentation of isoproterenol-stimulated CAMP formation rebral cortex could be differentiated by their affinities for the in slices, whereas the a,A antagonist 5methyl-urapadil did competitive antagonistsWB4 101 and phentolamine, and des- not. This suggests that the a,B receptor subtype potentiates ignated them a,* and CY,~.Han et al. (1987a,b) subsequently CAMP formation. Thus, the increased (Y, receptor augmen- demonstratedthat thesebinding sitescould be distinguished by tation of CAMP formation seen in slices from estradiol-treat- their different sensitivities to inactivation by the site-directed ed rats is correlated with increased (Y,~ receptor number. alkylating agent chlorethylclonidine (CEC). They demonstrated that the o(,~site was relatively CEC insensitive whereasthe (Y,~ In some tissues,including brain slices,activation of a,-adren- site was sensitive to irreversible inactivation by CEC. ergic receptors potentiates agonist- and forskolin-stimulated The identity of the (Y, receptor subtype mediating CAMP po- CAMP accumulation (Perkins and Moore, 1973; Daly et al., tentiation remains unresolved. For example, Johnson and 1980; Duman et al., 1985; Sugdenet al., 1985; Etgen and Petitti, Minneman (1986, 1987) suggestthat the a), receptor subtype 1987; Petitti and Etgen, 1990, 1991). Ovarian steroidsmodulate that augmentsCAMP accumulation is similar to the 01,receptor a,-adrenoceptor augmentation of CAMP formation in brain ar- that activates inositol phospholipid hydrolysis, most likely the easthat regulate female reproductive function. When compared (Y,~subtype, whereasRobinson and Kendall (1989) suggestthat to hypothalamic and preoptic area (POA) slices from ovariec- it is not. Recent studiesemploying both CEC and antagonists tomized (OVX) rats, slices from estradiol (E,)-treated females with high selectivity for the o(,* subtype concluded that the exhibit enhancedo(, receptor augmentation of CAMP formation receptors mediating CAMP potentiation and phosphoinositol (Petitti and Etgen, 1990). This changein LY,receptor facilitation hydrolysis in brain slices fit neither the current (Y,* nor the (Y,~ classification (Minneman and Atkinson, 1991). Therefore, the purposesof the present studieswere (1) to evaluate the possi- Received Feb. 5, 1992; revised Apr. 7, 1992; accepted May 5, 1992. bility that the small E,-induced increase in total 01,receptor This work was supported by DHHS Grants MH41414 and RSDA MH00636 to A.M.E., by BRSG Grant RR05397, and by the Department of Psychiatry, binding in hypothalamus and POA is attributable to selective Albert Einstein Colleee of Medicine. alterations in a specific cu,receptor subtype, and (2) to identify Correspondence sh&ld be addressed to Dr. A. M. E&en, Department of Psy- chiatry, F113, Albert Einstein College of Medicine, 1300 Morris Park Avenue, which oc,-adrenoceptorsubtype mediates the augmentation of Bronx, NY 1046 1. CAMP formation in hypothalamic and POA slices.The selective Copyright 0 1992 Society for Neuroscience 0270-6474/92/123869-08$05.00/O aYIAreceptor antagonist 5-methyl-urapadil(5-MU; Gross et al., 3870 Petitti et al. * Estradiol increases a,.-Adrenergic Receptors

1988), WB4101, an antagonist with higher affinity for the allA tion was preincubated for 10 min at 37°C with vehicle or with 10 PM than for the LY,~subtype, and CEC, an irreversible, a,,-selective CEC. Reactions were stopped by addition of 6 ml of ice-cold Na-HEPES buffer and centrifugation for 10 min at 20,000 x g. The supematant alkylating agent, were usedto distinguish the (Y, receptor sub- was discarded and the pellet resuspended in 6 ml Tris-MgCl, buffer. types, Specific ‘H-prazosin binding after CEC inactivation reflects the (Y,~- Some of these findings have appeared in preliminary form adrenergic receptor population. The LY,~receptor population was deter- (Etgen et al., 1992). mined by subtracting the binding of 3H-prazosin after CEC inactivation from total specific ‘H-prazosin binding. Determinationof sliceCAMP accumulation. Each POA and MH slice Materials and Methods was maintained at 34-35°C in a shaking water bath (80 oscillations/ min) in an individual tissue culture well containing 300 ~1 of a modified Animals and hormonetreatments. Female Sprague-Dawley rats ob- Yamamoto’s medium (Yamamoto, 1972) in an O,/CO, (95:5)-saturated tained from Taconic Farm (Germantown, NY) and weighing 150-l 75 environment. The incubation conditions were identical to those used gm were ovariectomized bilaterally under Metofane anesthesia 4-7 d in our previous work except that NaHCO, was replaced by 10 mM prior to use. Estrogen treatment consisted of two subcutaneous injec- HEPES in the Yamamoto’s medium. Slices were preincubated for 75 tions of 2 pg of E, benzoate (EB) given 24 and 48 hr before death. P min prior to the addition of the NE agonists. CEC and 5-MU were treatment consisted of an injection of 500 fig of P given subcutaneously added directly to the incubation wells 30 min and 10 min, respectively, 3.5 hr before sacrifice. EB and P were dissolved in peanut oil and injected prior to phenylephrine (PHE) and isoproterenol (ISO). At the end of in a volume of 0.1 ml. the 20 min incubation period with the NE agonists, the slices were Tissuepreparation. Animals were killed by decapitation, and their transferred rapidly to 400 ~1 of ice cold 5% (w/v) trichloroacetic acid. brains were rapidly removed and placed on ice. The entire hypothalamus The slices were disrupted by sonication, and the supematant (containing and POA were removed, and 350 pm slices were cut on a McIlwain CAMP) and pellet (containing tissue protein) were separated by cen- tissue chopper beginning approximately 2 mm anterior to the optic trifugation. The pellet was dissolved in 2.0 M NaOH for determination chiasm and ending 1 mm anterior to the mammillary bodies. Based on of protein content. The supernatant was acidified with 1.O M HCl, and anatomical landmarks observed in comparable slices from fixed tissue, trichloroacetic acid was removed with 4 vol of washed ether. The re- four slices of POA and three slices of middle hypothalamus (MH) were sulting aqueous extracts were concentrated by lyophilization and ana- obtained as described earlier (Etgen and Petitti, 1986, 1987) and used lyzed for CAMP content using a modified Gilman protein binding assay for CAMP assays and for preparation of membranes for some radioli- (Brostrom and Kon, 1974). gand binding assays. The MH slices include the arcuate nucleus, the Chemicals. EB and P were purchased from Steraloids, Inc. (Wilton, ventromedial nucleus, the dorsomedial nucleus, and much of the lateral NH). Metofane was obtained from Pitman-Moore, Inc. (Atlanta, GA). hypothalamus. For binding competition studies, membranes from the Radiolabeled 3H-prazosin (87 Ci/mmol) was obtained from New En- entire hypothalamus plus POA were used (see below). A sample of gland Nuclear (Boston, MA). PHE, ISO, and phentolamine were pur- frontal cortex (CTX; 250-350 mg) was also removed from each brain chased from Sigma (St. Louis, MO). CEC and WB4101 were purchased for radioligand binding assays. from Research Biochemicals, Inc. (Natick, MA), and 5-MU was gen- Radioligandbinding assay. To provide sufficient material for Scat- erously provided by Byk Gulden (Konstanz, Germany). chard analysis, tissue from two rats given identical hormone injections Analysisofdata. Aliquots of each slice and membrane fraction were was combined for the POA and MH regions. For competition studies, analyzed for protein content by the method of Larson et al. (1986). For combined hypothalamus plus POA from three rats were used. Tissue binding assays, receptor number (i.e., B,,,) was expressed as fmol/mg samples were homogenized separately in 5 ml of ice-cold Tris-MgCl, protein. For CAMP experiments, data were converted to pmol CAMP/ buffer (50 mM Tris HCl; 10 mM MgCl,; pH 7.4) with a Polytron at mg protein. Values for the four POA or three MH slices were averaged speed 5-6 for 20 sec. The homogenates were centrifuged for 10 min at to give a single value for each brain region for each rat. Significant 20,000 x g, the supematant discarded, and the pellet containing the differences between means were determined using analysis of variance. membrane fraction frozen at -70°C until assay. Freezing of the crude Planned post hoc comparisons were made using a Newman-Keuls mul- membrane fraction does not result in a measurable loss of any NE tiple range test and/or t tests. Differences were considered statistically receptor subtype (Etgen and Karkanias, 1990). significant ifp < 0.05. For analysis of total 01~binding, frozen membranes were resuspended in 6 ml buffer (Tris-MgCl,) and preincubated for 10 min at 37°C. The preincubation was terminated by addition of 6 ml of ice-cold Tris-MgC1, Results buffer and centrifugation for 10 min at 20,000 x g. The supernatant Effects of steroids on (Y, receptor number and binding afinity was discarded and-the pellet resuspended in 6 ml-Tris-MgCl, buffer. Prior studies indicated that EB enhanced 01, receptor potentia- For Scatchard analvsis. trinlicate 200 ~1 aliauots were incubated for 20 min at 37°C with 0:05-5 nM 3H-prazdsin with and without a 2000-fold tion of CAMP formation in MH and POA slices (Etgen and excess of phentolamine to assessnonspecific binding. Competition stud- Petitti, 1987; Petitti and Etgen, 1990) and that this was corre- ies were conducted by incubating triplicate aliquots of membrane sus- lated with modest increasesin 3H-prazosin binding (Etgen and pensions with 2 nM 3H-prazosin in the presence of radioinert antagonists Karkanias, 1990). Thus, initial binding studies replicated the (WB4 10 1 or 5-MU). Antagonist concentrations were varied from lO-4 effects of ovarian steroids on total cu,receptor binding. Figure to lo- I I M in %-log increments. Bound and free >H-prazosin were separated by rapid filtration through glass fiber filters (FPB- 148 Whatman 1 and Table 1 show that EB significantly increased total 01, GF/B) on a Brandel (Gaithersburg, MD) cell harvester. Filters were receptor number in POA (33%) and MH (49%) membranes0, washed with 5 x 2 ml of ice cold buffer, and radioactivity was deter- < 0.05) with no change in 3H-prazosin binding affinity. Ap- mined. To reduce nonspecific adsorption of ligands, filters were soaked proximately 75% of the total 3H-prazosin binding represented in 1% polyethylenimine for 1 hr prior to use (Bruns et al., 1983). Ligand affinities (K,, and K,), apparent receptor numbers (B,,,,,), and Hill coef- specific binding. As reported previously, the administration of ficients were calculated using the LIGAND program of Munson and P to EB-primed rats did not alter the EB-induced increase in Rodbard (1980). )H-prazosin binding in POA and MH membranes,nor were the To distinguish aIA and aIB receptor subtypes, the effects of CEC in- binding parameters of cu, sites in CTX membranesinfluenced activation on 3H-prazosin binding in brain membranes were studied. by hormone treatment (Fig. 1, Table 1). The water solubility of CEC may prevent it from gaining access to binding sites enclosed in vesicles that may form following tissue ho- To determine whether alA or LY,~receptors are selectively al- mogenization in a Tris-MgCl, buffer (Minneman et al., 1988). In agree- tered by estrogen,we evaluated the effectsof CEC pretreatment ment with these observations, pilot experiments indicated that Tris- on 3H-prazosin binding in POA, MH, and CTX membranes. MgCl, buffer is unsuitable for assessing CEC inactivation of alB receptors Figure 2 shows representative protein-standardized Scatchard (N. Petitti and A. M. Etgen, unpublished observations). Therefore, the binding assays were performed in hypotonic, 10 mM Na-HEPES buffer plots of o(, receptor binding in MH membranesfrom OVX and (pH 7.6) to promote vesicle lysis. Frozen membranes were resuspended estrogen-treatedrats preincubated in the absenceor presenceof in 6 ml Na-HEPES buffer and split into two equal fractions. Each frac- CEC to inactivate cylBreceptors. In all tissuesfrom both OVX The Journal of Neuroscience, October 1992, 72(10) 3871

- 0 ovx ovx .: 4OOA ELI EB I 2 - EB+P 0.200 k 350-- ii t E" 300-- \ : 250-- z * * & 200-- T T T 150-- : m lOO-- 0 20 40 60 80 100 120 140 160

.; 50 -- 0 EB a POA MH CTX 0.200 Figure 1. Effects of in vivo treatments with ovarian steroids on 3H- prazosin binding. OVX female rats were primed with vehicle (OK%‘), with EB only (2 pg, 24 and 48 hr before death), or with EB + 500 wg ofP 3.5 hr before death. B,,, values were obtained by Scatchard analysis of 3H-prazosin binding in the presence or absence of excess phentolamine to assess nonspecific binding. Each value represents the mean of four to six independent replications and is expressed as fmol/mg protein (&SEM). *, significantly greater than OVX (p < 0.05); **, significantly greater than OVX @ < 0.01). and E,-treated females,3H-prazosin appeared to bind to a single Bound (fmole/mg) classof saturable,high-affinity binding sites,and Hill coefficients Figure 2. Representative Scatchard plots of total and CEC-insensitive were not significantly different from 1.0 (data not shown). E, ‘H-prazosin binding in hypothalamic membranes from OVX and EB- administration had little influence on the number of CEC-in- treatedfemale rats. OVX female rats were primed with EB only (2 fig, sensitive binding sites despite a substantial increase in total 24 and 48 hr before death) or with -vehicle before death (OKI). Hy- specific 3H-prazosin binding sites. This is illustrated for group pothalamic membranes were preincubated in Na-HEPES buffer for 10 min at 37°Cwith (solid circles) or without 10 PM CEC (opencircles). data by the finding that administration of neither EB nor EB+P Values were obtained by Scatchard analysis of 3H-prazosin binding in significantly influenced CEC-insensitive sites, that is, the num- the presenceor absenceof excessphentolamine to assessnonspecific ber of aYIAreceptors, in any brain region (Fig. 3B). In agreement binding. Similar binding results were found in several independent pools with Figure 1, EB did increasethe total number of 3H-prazosin of hypothalamic and POA tissue from OVX and EB-treated animals (see Fig. 3). binding sitesin both POA and MH membranesbut not in CTX membranes(Fig. 3A). Moreover, POA (p < 0.05) and MH (p < 0.01) membranes from E,-treated females demonstrated a binding affinity (data not shown). However, CEC caused an five- to sixfold increasein CEC-sensitive sites(i.e., allBreceptor approximately twofold increasein K, for 3H-prazosin binding number) when compared to OVX controls (Fig. 3C’).This effect in all three tissues(p < 0.000 1). A similar CEC-dependentchange of estrogenwas not observed in CTX membranes.The admin- in K, for lz51-BE-2254binding has been observed in the hy- istration of P to EB-primed rats did not modify the EB-induced pothalamus (Wilson and Minneman, 1989) and liver (Han et increasein 3H-prazosin binding in the presenceor absenceof al., 1987a; Minneman et al., 1988). This would suggestthat aIB CEC in POA or MH membranes.Thus, the failure of P to modify receptors have a slightly higher affinity for prazosin than alA total prazosin binding cannot be attributed to reciprocal changes receptors. We were unable to resolve total 3H-prazosin binding in the concentration of the allAand allBsubtypes. into two components by subjecting saturation isotherms to a In viva hormone treatment did not influence 3H-prazosin two-site model, presumably becausethe affinities of the CY,~and o(,~sites for prazosin differ by only twofold. Additional binding studies were carried out to confirm the Table 1. Effects of in viva hormone treatments on ‘H-prazosin selectivity of CEC for a,,-adrenoceptors and the specificity of binding affinity estrogenaction on that receptor subtype. Figure 4 showsthat the displacementof 3H-prazosinby WB4 101, an antagonistwith Kn (nM k SEM) relatively higher affinity for alA than for o(,~receptors, was sim- Region ovx EB EB + P ilar in hypothalamic-POA membranesfrom OVX rats prein- POA 0.35 k 0.1 0.43 k 0.8 0.43 + 0.3 cubated in the presenceor absenceof CEC. In both cases,the slopecoefficient of the competition curve was closeto 1.O (Table MH 0.41 k 0.3 0.45 2 0.7 0.49 + 0.6 CTX 0.60 zk 0.3 0.87 +- 0.1 0.69 + 0.2 2) indicative of a single receptor population. Hence, theseob- servations are consistent with the interpretation that in mem- OVX female rats were primed with vehicle (OVX), with EB only (2 pg, 24 and branes from OVX females, most (Y, receptors are of the aYIA 48 hr before death), or with EB + 500 pg of P 3.5 hr before death. K, valueswere obtained from Scatchard analysis of )H-prazosin binding. Each value represents subtype (i.e., high affinity for WB4101). By contrast, the slope the mean of four or five independent replications. B,,,, valuesare shown in Figure 1. coefficient of the competition curve for WB4 101 displacement 3872 Petitti et al. * Estradiol Increases aI,-Adrenergic Receptors

0 ovx Table 2. WB4101 and S-MU displacement of 3H-prazosin binding m EB sites in hypothalamic-POA membianes - EB+P Competitor Slope coefficient K, (nrd WB4101 ovx -CEC 0.90 + 0.18 3.9 k 0.85 +CEC 0.95 f 0.09 1.3 k 0.06 EB -CEC 0.70 + 0.26 2.5 k 0.99 +CEC 1.0 t 0.15 0.48 k 0.11 5-MU ovx -CEC 0.39 f 0.10 407 k 108 +CEC 1.1 k 0.05 85 k 36 OVX female rats were primed with vehicle (OVX) or with 2 ~8 of EB 24 and 48 hr before death. Membranes were preincubated in Na-HEPES buffer in the presence (+CEC) or absence(GCEC) of 10 YM CEC. Aliauots of the membranes were then kubaied with 2 n; )H-prkosin kd increasing concentrations of WB4101 or 5-MU. Slope coefficientsand K, values were obtained by computer modeling using the LIGANDprogram of Munson and Rodbard (1980). Representative competition curves are shown in Figures 4 and 5. Values represent the mean f SEM of three (WB4101) or two (5-MU) independent replications. CEC and EB significantly reduced the apparent K, ofWB4101 (p < 0.001 andp < 0.02, respectively).

servations confirm other reports that CEC selectively inacti- vated the (Y,~receptor subtype. They are also consistent with the interpretation that E, selectively increasedthe proportion of 3H-prazosin binding attributable to (Y,~receptors. Displacement of 3H-prazosin by the a!,,-selective antagonist 5-MU was also evaluated in hypothalamic-POA membranes from OVX rats (Fig. 5). Approximately 80% of 3H-prazosinwas displaced by 5-MU in the absenceof CEC preincubation, indicating that the predominant receptor is of the (Y,~ subtype. When membraneswere preincubated in the presenceof CEC, the 5-MU competition curve shifted to the left, the slopecoefficient was closer to 1.O, and the K, decreased(Table 3). Because there is some evidence that 5-MU is the antagonist of choice 25 for blocking q,-adrenoceptors (Han and Minneman, 199 l), and becauseit clearly competed for prazosin binding in hypothalamic-POA membranes, we chose 5-MU for use in our slice POA MH CTX studies.

Figure 3. Effects of in vivo treatments with ovarian steroids on putative Effects of CEC and 5-MU on (Y, receptor augmentation of qA- and a,,-adrenergic receptors. OVX female rats were primed with CAMP formation vehicle (OVX), with EB only (2 pg, 24 and 48 hr before death), or with EB + 500 ue. of P 3.5 hr before death. To distinguish air and a,, In our previous studies (Etgen and Petitti, 1987; Petitti and receptors, P%, MH, and CTX tissues were preincubated in I%-HEPki Etgen, 1990), o(, receptor potentiation of the CAMP responseto buffer for 10 min at 37°C with or without 10 PM CEC. Total 3H-prazosin P-adrenergicreceptor stimulation was more robust in POA and binding (A), corrected for nonspecific binding, reflects both (Y, receptor MH slicesfrom E,-treated than from OVX females. Thus, ex- subtypes. B shows the LY,~-adrenergic receptor population that is measurable after CEC inactivation. The qB receptor population (c) was periments to determine whether o(,~- or a,,-adrenoceptors me- determined by subtracting the binding of )H-prazosin after CEC inac- diate augmentation of ISO-stimulated CAMP formation were tivation from total 3H-prazosin binding. II,,,,, values were obtained by conducted in slices from EB-treated animals (Table 3). The Scatchard analysis. Each value represents the mean of three to five combination of LY,and p receptor activation (PHE+ISO) in- independent replications and is expressed in fmol/mg protein (k SEM). *, significantly greater than OVX @ < 0.05); **, significantly greater creasedCAMP accumulation significantly more than IS0 alone than OVX (p < 0.01). (p < 0.01). PHE alone had no effect on CAMP synthesisin these brain slices(Petitti and Etgen, 1990, 1991). When sliceswere preincubated with CEC to inactivate CY,~receptors, PHE poten- of 3H-prazosin binding in membranesfrom EB-treated females tiation of ISO-stimulated CAMP accumulation was eliminated. was less than 1.0 in the absenceof CEC (Table 2), indicating In contrast, the qA receptor antagonist 5-MU at a concentration the presenceof a heterogeneousreceptor population. In addi- that should block all or most qA sites(Fig. 5) did not alter PHE tion, the WB4 10 1 competition curve was shifted to left in the augmentation of CAMP synthesisin either POA or MH slices. presenceof CEC (Fig. 4), an observation consistent with the Neither CEC nor 5-MU significantly altered basalCAMP levels removal of the lower-affinity (Y,~receptor subtype. Moreover, in POA or MH slices.Although CEC appearedto have a modest CEC significantly reducesthe K, of WB4101 (p < 0.05) in both suppressiveeffect on ISO-stimulated CAMP accumulation, this OVX and E,-treated tissue(Table 2). Taken together, theseob- effect was not statistically significant. The Journal of Neuroscience, October 1992, 72(10) 3873 EB

0 80 80 Z 2 60 + CEC 60 M X 40 \ 40

20 20

0 I 0 “‘/’ I I 00 11 10 9 8 7 6 CxJ 11 10 9 8 7 6

- LOG [WB4101]

Figure 4. Representative competition curves of WB4 10 1 displacement of ‘H-prazosin binding in control and CEC-pretreated membranes from combined hypothalamus and POA of OVX and EB-treated female rats. OVX female rats were primed with vehicle (OKY) or with EB (2 fig) 24 and 48 hr before death. Membranes from combined hypothalamus and POA were preincubated in Na-HEPES buffer for 10 min at 37°C with (open circles) or without (solid circles) 10 /IM CEC prior to incubation with 2 nM 3H-prazosin and increasing concentrations of WB4 10 1. Similar results were observed in several independent pools of hypothalamic and POA membranes from OVX and EB-treated animals. Binding parameters (see Table 2) were obtained by computer modeling using the LIGAND program of Munson and Rodbard (1980).

Discussion 120 This is the first demonstration that an endogenousfactor, that is, the steroid hormone E,, regulatesa specificol,-adrenoceptor subtype in the CNS. The conclusionthat E, selectively increases 100 the number of or,,-adrenergic binding sitesin female rat hypothalamus and POA is supported by two findings. First, in vivo 80 administration of E, increasesthe number of CEC-sensitive,but n not CEC-insensitive 3H-prazosinbinding sites.Second, the com- Z petition curve for displacementof prazosin by the a,,-preferring 2 60 antagonist WB410 1 is influenced more markedly by preincubation with the a,,-preferring alkylating agent CEC in mem- m t39 40

Table 3. Effects of CEC and 5MU on PHE augmentation of ISO- I- stimulated CAMP formation in slices from EB-primed rats 20 0 CAMP (pmol/mg protein k SEM) 9 00 0 - POA MH 4 /’ I I I I I I I Basal 5.88 + 0.56 6.36 + 0.74 I; all 10 9 8 7 6 5 4 Iso (1 PM) 11.8 + 0.95* 12.4 k 1.35* IS0 + PHE (10 /.LM) 21.7 + 2.20** 26.0 + 2.82** -LOG [5-MU] CEC (100 PM) 5.79 + 0.56 6.85 f 0.74 + IS0 8.53 + 0.96* 9.92 k 0.41* Figure 5. Representative competition curves of 5-MU displacement of )H-prazosin binding in control and CEC-pretreated hypothalamic- + IS0 + PHE 8.28 + 0.30* 12.1 IL 0.63* POA membranes from OVX female rats. Combined hypothalamic- 5-MU (1 PM) 6.87 + 0.56 6.54 + 0.15 POA membranes from OVX rats were preincubated in Na-HEPES buff- + IS0 13.9 k 0.78* 15.0 + 0.93* er for 10 min at 37°C with (open circles) or without (solid circles) 10 pM + IS0 + PHE 25.3 1- 1.51** 25.5 + 1.46** CEC prior to incubation with 2 nM 3H-prazosin and increasing concentrations of 5-MU. Similar results were observed in a second preparation Each value represents the mean of four to five independent replications. of hypothalamic-POA membranes from OVX female rats. Binding pa- * Significantly greater than basal f.p < 0.01). rameters (see Table 2) were obtained by computer modeling using the ** Significantly greater than IS0 alone (p < 0.05). LIGAND program of Munson and Rodbard (1980). 3874 Petitti et al. * Estradiol Increases a,,-Adrenergic Receptors branes from estrogen-treated than from OVX females. Taken hances PHE potentiation of ISO-stimulated CAMP formation, together, these observations suggest that in OVX females, o(,~ our results suggest that estrogen enhances signaling via phos- receptors comprise only a small fraction of the total a), binding pholipase C. This interpretation is congruent with the recent sites. An earlier biochemical study (Johnson and Minneman, proposal by Mobbs et al. (1991) that phospholipase C-oc may 1987) reported that in male rat hypothalamus, only one-third be a widespread mediator of E, action in many tissues. More- to one-half of o(, binding sites are CEC sensitive. A more recent over, estrogen has been reported to increase cu,-adrenoceptor- report that utilized receptor autoradiography to distinguish be- mediated phosphoinositol hydrolysis in rabbit myometrium tween (Yereceptor subtypes (Blendy et al., 1990) likewise found (Riemer et al., 1987) and in rat cortical slices (Favit et al., 199 1j. a 3:l ratio of aIA:ale sites in male rat hypothalamus. We now The finding that (Y,~ receptors mediate adenylyl cyclase poten- find that estrogen priming increases the ratio of ale:aIA sites in tiation in hypothalamus and POA may also explain the inability female rat hypothalamus and POA, although the proportion of of the potent 01, receptor agonist methoxamine to augment IS0 aIB receptors is never more than 50% of the total 3H-prazosin stimulation of CAMP formation in slices from these brain areas binding in either brain region. (Petitti and Etgen, unpublished observations). A recent study In agreement with other reports (Wilson and Minneman, 1989; in rat hepatocytes reported that methoxamine preferentially Blendy et al., 1990; Han and Minneman, 1991), frontal CTX stimulates the alA receptor subtype to increase calcium influx membranes have relatively more ollB than o(,~ receptors. Estro- (Tsujimoto et al., 1989). gen administration does not significantly affect prazosin binding However, there is still limited information regarding which to either cy, receptor subtype in this relatively estrogen receptor- o(, receptor subtypes are linked to specific intracellular signaling poor brain region (Pfaff and Keiner, 1973). This finding is con- mechanisms. Moreover, the intracellular mediators of a,-ad- sistent with the interpretation that E,-induced increases in hy- renoceptor potentiation of adenylyl cyclase activity may be cell pothalamic-POA (Y,~ receptors may be mediated by genomic type specific. For example, in pinealocytes, the arachidonic acid actions of intracellular estrogen receptors. The long time lag cascade rather than phosphoinositol hydrolysis is likely to un- between hormone administration and increases in prazosin derlie ol,-adrenergic enhancement of agonist-stimulated cyclic binding observed in our earlier study (Etgen and Karkanias, nucleotide accumulation (Chik et al., 199 1). There is some ev- 1990) also supports a genomic mechanism of E, action. idence that a,,-mediated responses require calcium influx Although previous reports have shown that estrogen influ- whereas responses to (Y,~ receptors involve mobilization of in- ences oc,-adrenoceptor function and number (Weiland and Wise, tracellular calcium stores via phosphoinositol hydrolysis (see 1987; Condon et al., 1989; Roberts et al., 1989; Etgen and Han et al., 1987b, 1990; Tsujimoto et al., 1989; Wilson and Karkanias, 1990; Petitti and Etgen, 1990; Favit et al., 199 l), Minneman, 1990). However, stimulation of either allA or (Y,~ the methods used in those studies did not distinguish between receptors can lead to phosphoinositol hydrolysis (Robinson and the two or, receptor subtypes. The finding that E, selectively Kendall, 1990; Wilson and Minneman, 1990). Recent studies upregulates ol,” receptors, which account for the minority oftotal utilizing 5-MU, CEC, and niguldipine, another a,,-preferring 3H-prazosin binding in the POA and hypothalamus of OVX antagonist, to examine the signal transduction pathways linked females, may explain some ofthe inconsistencies in the literature to the two receptor subtypes in brain slices and primary glial on estrogen regulation of hypothalamic oc,-adrenoceptors. Like- cultures failed to resolve this issue (Minneman and Atkinson, wise, the data suggest that the modest (30-50%) estrogen-de- 199 1). Moreover, in contrast with our results, those investiga- pendent increase in the total number of hypothalamic and POA tors found that oc,-adrenoceptor potentiation ofcAMP responses o(, binding sites (Etgen and Karkanias, 1990; Fig. 1) actually in cortical and hippocampal slices was insensitive to CEC and represents a nearly 600% increase in o(,~ receptors. A large in- was inhibited in a noncompetitive fashion by 5-MU, leading crease m LY,~receptor number is difficult to detect when total them to conclude that the oc, receptors mediating CAMP accu- 3H-prazosin binding is measured. However, it must also be mulation do not easily fall into the traditional ala/ale subclas- noted that the present data do not rule out the possibility that sification and may not be linked to phosphoinositol hydrolysis. increases m cylBreceptors are also limited to specific hypotha- The discrepancies between our findings and the work of other lamic nuclei, further compounding the difficulty of detecting laboratories may reflect tissue and/or sex differences in the cou- hormone-dependent changes when large tissue blocks are ana- pling of o(, receptor subtypes to specific membrane effecters. lyzed. Alternatively, as postulated by Minneman and Atkinson (199 l), Present data also suggest that a,-adrenoceptor augmentation cy, receptor subtypes in addition to those fitting in with the of CAMP generation in hypothalamic and POA slices is medi- current o(,~/LY,~scheme may exist. In support of this possibility ated by the alB subtype. This conclusion is supported by the is the finding that the properties of cloned, putative (Y,~ and allB observation that inactivating (Y,~ receptors by preincubation with receptors, when expressed in mammalian cells, are not entirely CEC abolishes PHE potentiation of ISO-stimulated CAMP syn- consistent with the o(,~/cY,~classification (see Harrison et al., thesis in slices from estrogen-primed females whereas blocking 1991). (Y,~ receptors with 5-MU does not. Because we have also shown The present analysis of 3H-prazosin binding also demon- that a,-adrenoceptor potentiation of CAMP accumulation in strates that administration of P to estrogen-primed female rats hypothalamic slices is blocked by inhibitors of protein kinase does not modify the number of either LY,~or a,* binding sites C and phospholipase C (Petitti and Etgen, 199 l), it is reasonable in hypothalamus or POA. This is an interesting finding in light to propose that a,,-adrenoceptor modulation of adenylyl cyclase of our consistent observation that when hypothalamic and POA may be mediated via activation of phosphoinositol hydrolysis. tissue from estrogen-primed rats is exposed to P either in vivo This hypothesis is consistent with data from other laboratories (Petitti and Etgen, 1990) or in vitro (Petitti and Etgen, 1992), demonstrating that ayIB receptor activation stimulates the for- o(, receptor augmentation of &receptor-stimulated CAMP formation of inositol phosphates (Han et al., 1987b; Michel et al., mation is abolished. Thus, the P-induced reduction in 01, re- 1990; Terman et al., 1990). Furthermore, since E, priming en- ceptor coupling to the CAMP second messenger system does not The Journal of Neuroscience, October 1992, 72(10) 3875 correlate with the density or affinity of oIB receptors, the subtype Favit A, Fiore L, Nicoletti F, Canonico PL (199 1) Estrogen modulates that apparently mediates CAMP potentiation in these tissues. stimulation of inositol phospholipid hydrolysis by norepinephrine in This suggests that P may act on factors controlling al,* receptor rat brain slices. Brain Res 555:65-69. Gross G, Hanft G, Rugevics C (I 988) 5-Methyl-urapadil discriminates coupling to its effector system (e.g., G-protein coupling, avail- between subtypes of the or,-adrenoceptor. Eur J Pharmacol 151:333- ability of substrates, enzymes, or G-proteins). 335. In summary, these results demonstrate (1) that activation of Han C, Minneman KP (199 1) Interaction of subtype-selective antag- the a,,-adrenergic receptor subtype mediates potentiation of onists with a,-adrenergic receptor binding sites in rat tissues. Mol Pharmacol40:531-538. ISO-stimulated CAMP formation in hypothalamic and POA Han C, Abel PW, Minneman KP (1987a) Heterogeneity of a,-adren- slices and (2) that estrogen selectively increases a,,-adrenoceptor ergic receptors revealed by chlorethylclonidine. Mol Pharmacol 32: number in these brain regions. Furthermore, P does not influ- 505-5 10. ence 3H-prazosin binding in estrogen-primed rats. Thus, the loss Han C, Abel PW, Minneman KP (1987b) ol,-Adrenoceptor subtypes of o(, receptor potentiation of CAMP synthesis in EB+P-treated linked to different mechanisms for increasing intracellular Ca*+ in smooth muscle. Nature 329:333-335. rats is not attributable to downregulation of CX,or o,a receptors Han C, Wilson KM, Minneman KP (1990) oc,-Adrenergic receptor but may reelect a modification of the coupling of the olB receptor subtypes and formation of inositol phosphates in dispersed hepato- to various components of its signal transduction system. Be- cytes and renal cells. Mol Pharmacdl 37;903-910. cause behavioral data from this laboratory suggest that acti- Harrison JK. Pearson WR. Lvnch KR (1991) Molecular characterization of a,- and oc,-adrenoceptors. Trends Pharmacol Sci 12:62-67. vation of hypothalamic ol,-adrenoceptors by endogenously re- Johnson RD, Minneman KP (1986) Characterization of or,-adreno- leased NE may participate in the hormonal facilitation of female ceptors which increase cyclic AMP accumulation in rat cerebral cor- reproductive behavior (see Etgen et al., 1992), an important tex. Eur J Pharmacol 129:293-305. direction for future study will be to determine which 01, receptor Johnson RD, Minneman KP (1987) Differentiation of cY,-adrenergic subtypes and signal transduction mechanisms underlie the mod- receptors linked to phosphatidylinositol turnover and cyclic AMP accumulation in rat brain. Mol Pharmacol 31:239-246. ulatory actions of NE on female reproductive function. Larson E, Howlett B, JagendorfA (1986) Artificial reductant enhancement of the Lowry method for protein determination. Anal Biochem References 155:243-248. Lomasney JW, Cotecchia S, Lelkowitz RJ, Caron MG (199 1) Molec- Blendy JA, Grimm LJ, Perry DC, West-Johnsrud L, Kellar KJ (1990) ular biology of a-adrenergic receptors: implications for receptor clas- Electroconvulsive shock differentially increases binding to alpha- 1 sification and for structure-function relationships. Biochim Biophys adrenergic receptor subtypes in discrete regions of rat brain. J Neu- Acta 1095:127-139. rosci 10:2580-2586. McGrath JC (1982) Evidence for more than one type of post-junctional Brostrom C, Kon C (1974) An improved protein binding assay for ol-adrenoceptor. Biochem Pharmacol 3 1:467-4&l. cyclic AMP. Anal Biochem 58:459-468. Michel MC. Hanft G. Gross G (1990) ala- but not a,,-adrenocevtors Bruns RF. Lawson-Wendina K. Puaslev TA (1983) A ravid filtration mediate inositol phosphate generation. Naunyn Schmiedebergs krch assay for soluble receptor&sing polyeihylenimine:treated filters. Anal Pharmacol341:385-387. Biochem 132:74-8 1. Minneman KP, Atkinson B (199 1) Interaction of subtype-selective Bylund DB (1988) Subtypes of cu,-adrenoceptors: pharmacological and antagonists with cy,-adrenergic receptor-mediated second messenger molecular biological evidence converge. Trends Pharmacol Sci 9:356- responses in rat brain. Mol Pharmacol 40:523-530. 361. Bylund DB, Ray-Prenger C, Murphy TJ (1988) Alpha-2A and alpha- Minneman KP, Molinoff PB (1980) Classification and quantitation of 2B adrenergic receptor subtypes: antagonist binding in tissues and P-adrenergic receptor subtypes. Biochem Pharmacol29: 13 17-l 323. cell lines containing only one subtype. J Pharmacol Exp Ther 245: Minneman KP, Han C, Abel PW (1988) Comparison of ol,-adrenergic 600-607. receptor subtypes distinguished by chlorethylclonidine and WB 4 10 1. Chik CL, Young I, Ho AK (1991) Differential involvement of the Mol Pharmacol 33:509-5 14. arachidonic acid cascade on the oc,-adrenergic potentiation of vaso- Mobbs C, Kaplitt M, Kow L-M, Pfaff DW (199 1) PLC-a: a common active intestinal peptide- versus P-adrenergic-stimulated cyclic AMP mediator of the action of estrogen and other hormones? Mol Cell and cyclic GMP accumulation in rat pinealocytes. J Neurochem 57: Endocrinol8O:C 187-Cl9 1. 1534-1539. Morrow AL, Creese I (1986) Characterization of oc,-adrenergic recep- Condon TP, Ronnekleiv OK, Kelly MJ (1989) Estrogen modulation tor subtypes in rat brain: a reevaluation of [3H]WB4104 and of the ol,-adrenergic response of hypothalamic neurons. Neuroendo- [‘Hlprazosin binding. Mol Pharmacol 29:321-330. crinology 50:51-58. Munson PJ, Rodbard D (1980) LIGAND: a versatile computerized Daly JW, Padgett W, Nimitkitpaisan Y, Creveling CR, Cantacuzene D, approach for characterization of ligand-binding systems. Anal Bio- Kirk KL (1980) Fluoronorepinephrine: specific agonists for the ac- them 107:220-239. tivation of (Yand p adrenergic sensitive cyclic AMP-generating sys- Perkins JP, Moore MM (1973) Characterization of the adrenergic tems in brain slices. J Pharmacol Exp Ther 212:382-389. receptors mediating a rise in cyclic 3’,5’-adenosine monophosphate Duman RS, Strada SJ, Enna SJ (1985) Effects of imipramine and in rat cerebral cortex. J Pharmacol Exp Ther 185:37 l-378. adrenocorticotropin administration on the rat brain norepinephrine- Petitti N, Etgen AM (1989) Progesterone depression of norepineph- coupled cyclic nucleotide generating system: alterations in a and p tine-stimulated CAMP accumulation in hypothalamic slices. Mol Brain adrenergic components. J Pharmacol Exp Ther 234:382-389. Res 5:109-l 19. Etgen AM, Karkanias GB (1990) Estradiol regulates the number of oll but not p or 01~noradrenergic receptors in hypothalamus of female Petitti N, Etgen AM (1990) o(! adrenoceptor augmentation of @-stim- rats. Neurochem Int 16: l-9. ulated CAMP formation is enhanced by estrogen and reduced by Etgen AM, Petitti N (1986) Norepinephrine-stimulated cyclic AMP progesterone in rat hypothalamic slices. J Neurosci 10:2842-2849. accumulation in rat hypothalamic slices: effects of estrous cycle and Petitti N, Etgen AM (199 1) Protein kinase C and phospholipase C ovarian steroids. Brain Res 375:385-390. mediate 01!-and fl-adrenoceptor intercommunication in rat hypotha- Etgen AM, Petitti N (1987) Mediation of norepinephrine-stimulated lamic slices. J Neurochem 56:628-635. cyclic AMP accumulation by adrenergic receptors in hypothalamic Petitti N, Etgen AM (1992) Progesterone promotes rapid desensiti- and preoptic area slices: effects of estradiol. J Neurochem 49: 1732- zation of a,-adrenergic receptor augmentation of CAMP formation in 1739. rat hypothalamic slices. Neuroendocrinology 55:1-8. Etgen AM, Ungar S, Petitti N (1992) Estradiol and progesterone mod- Pfaff DW, Keiner M (1973) Atlas of estradiol-concentrating cells in ulation of norepinephrine neurotransmission: implications for the the central nervous system of the female rat. J Comp Neurol 15 1: regulation of female reproductive behavior. J Neuroendocrinol, in 121-158. press. Riemer RK, Goldfien A, Roberts JM (1987) Estrogen increases ad- 3878 Petitti et al. l Estradiol Increases a,,-Adrenergic Receptors

renergic- but not cholinergic-mediated production of inositol phos- tification and structural characterization of or,-adrenergic receptor nhates in rabbit uterus. Mol Pharmacol 32:663-668. subtypes. Mol Pharmacol 37526-534. Roberts JM, Riemer RK, Bottari SP, Wu YY, Goldfien A (1989) Hor- Tsujimoto G, Tsujimoto A, Suzuki E, Hashimoto K (1989) Glycogen monal regulation of myometrial adrenergic response: the receptor and phosphorylase activation by two different a,-adrenergic receptor sub- beyond. J Dev Physiol 11: 125-l 34. types: methoxamine selectively stimulates a putative a,-adrenergic Robinson JP, Kendall DA (1989) Inositol phospholipid hydrolysis receptor subtype @,A that couples with Ca2+ influx. Mol Pharmacol and potentiation of cyclic AMP formation by noradrenaline in rat 36:166-176. cerebral cortex slices are not mediated by the same a-adrenoceptor Weiland NG, Wise PM (1987) Estrogen alters the diurnal rhythm of subtype. J Neurochem 52:690-698. a,-adrenergic receptor densities in selected brain regions. Endocri- Robinson JP, Kendall DA (1990) Niguldipine discriminates between nology~121:1751-1758. a,-adrenoceptor-mediated second messenger responses in rat cerebral Wilso&M, Minneman KP (1989) Regional variations in a,-adren- cortex slices. Br J Pharmacol 100:3-4. ereic recentor subtvnes in rat brain. J Neurochem 53: 1782-1786. Sugden D, Namboodiri MAA, Klein DC, Pierce JE, Grady R Jr, Mefford Wil& KM; Minneman KP (1990) Different pathways of [3H]inositol IN (1985) Ovine pineal a,-adrenoceptors: characterization and ev- phosphate formation mediated by qA- and q,-adrenergic receptors. idence for a functional role in the regulation of serum melatonin. J Biol Chem 265:17601-17606. Endocrinology 116:1960-1967. Yamamoto C (1972) Activation of hippocampal neurons by mossy Szabadi E, Bradshaw CM, eds (199 1) Adrenoceptors: structure, mech- fiber stimulation in thin brain sections in vitro. Exp Brain Res 14: anisms, function. Basel: Birkhauser. 423-435. Terman BI, Riek RP, Grodski A, Hess H-J, Graham RM (1990) Iden-