This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Copyright Permission Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Kreda, S. M. Articles by Wetsel, W. C. Search for Related Content PubMed PubMed Citation Articles by Kreda, S. M. Articles by Wetsel, W. C.

₁-Adrenergic Receptors Mediate LH-Releasing Hormone Secretion through Phospholipases C and A₂ in Immortalized Hypothalamic Neurons

Laboratory of Signal Transduction, National Institute of Environmental Health Sciences (S.M.K., M.S., S.F., C.M.R., S.S.), Research Triangle Park, North Carolina 27709; and Department of Psychiatry and Behavioral Sciences, Duke University Medical Center (G.X.L., W.X., K.W.D., S.C., W.C.W.), Durham, North Carolina 27710

Address all correspondence and requests for reprints to: Dr. William C. Wetsel, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center, Box 3497, 028 CARL Building, Durham, North Carolina 27710. E-mail: wetse001{at}mc.duke.edu

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Norepinephrine has long been known to stimulate the pulsatile and preovulatory release of LH-releasing hormone (LHRH). In vivo and in vitro studies indicate that these effects are mediated primarily through ₁-adrenergic receptors (₁-ARs). With the immortalized hypothalamic LHRH neurons, we have found that ₁-adrenergic agents directly stimulate the secretion of LHRH in a dose-dependent manner. Ligand binding and RNA studies demonstrate that the GT1 cells contain both _1A- and _1B-ARs. Competition binding experiments show that approximately 75% of the binding is due to _1B-ARs; the remainder is made up of _1A-ARs. Receptor activation leads to stimulation of PLC. PLCß1 and PLCß3 are expressed in GT1 neurons, and these PLCs are probably responsible for the release of diacylglycerol and IP as well as the increase in intracellular calcium. The mobilization of cytoplasmic calcium is sufficient to stimulate cytosolic PLA₂ (cPLA₂) and release arachidonic acid. A dissection of the contributions of the phospholipases to LHRH secretion suggests that cPLA₂ acts downstream of PLC and that it significantly augments the PLC-stimulated LHRH secretory response. Inasmuch as the ₁-ARs are known to play a critical role in LHRH physiology, we propose that both PLC and cPLA₂ are critical in regulating and amplifying LHRH release.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
ADRENERGIC RECEPTORS HAVE long been known to play a critical regulatory role in the neuroendocrine control of ovulation in mammals (1, 2). The effect appears to be mediated through hypothalamic LH-releasing hormone (LHRH) neurons, because this peptide represents the final common pathway for regulation of LH and FSH secretion from the anterior pituitary (3). Norepinephrine (NE) was one of the first neurotransmitters that has been shown to be involved in the regulation of reproduction (1). Subsequent in vivo studies revealed that blockade of -adrenoceptors (-ARs), but not ß-ARs, suppressed pulsatile LH release in intact females (4, 5). Consistent with these observations, NE has been reported to stimulate LHRH release from median eminence tissue fragments in a dose-dependent manner, and ₁-AR antagonists can block this secretory response (6, 7).

Although activation of ₁-ARs has been shown by a number of investigators to be a major regulator of the proestrous surge (8, 9), it is unclear whether these effects are mediated by direct NE innervation of LHRH neurons or whether the response is indirect and mediated by interneurons. Morphology studies have revealed that NE-containing nerve terminals have a widespread, but uneven, distribution throughout the preoptic area and hypothalamus (10, 11) where LHRH neuronal cell bodies primarily reside (12, 13). Immunocytochemical investigations have demonstrated that catecholaminergic axons terminate in close proximity to LHRH perikaryia, and these neurons may contain _1B-ARs (14, 15, 16). More recently, results from in situ hybridization studies have revealed that all three ₁-AR subtypes are expressed in the forebrain and hypothalamus (17, 18, 19, 20, 21). Despite this fact, there is substantial controversy in the literature with regard to the different ₁-AR subtypes. Presently, at least three different native ₁-ARs have been pharmacologically identified, and three distinct cDNAs have been cloned (22). Unfortunately, the pharmacological findings with the native and recombinant receptors have not always been consistent. As a result, the subtype names and descriptions became confusing. On the basis of pharmacological, biochemical, and molecular biology criteria, the International Union of Pharmacology has proposed that the ₁-ARs be classified into three different subtypes: the _1A-, _1B-, and _1D-ARs (22).

Besides difficulties in defining the characteristics of the different ₁-ARs, a more significant problem encountered in describing the role of the ₁-ARs in LHRH physiology pertains to these neurons themselves. The numbers of LHRH neurons in mammalian brain are relatively few, and they are scattered along the base of the brain from the preoptic area to the anterior hypothalamic region (12, 13). This topology is quite challenging for any molecular or cellular analysis of this neuronal system. For this purpose, we have used an immortalized hypothalamic neuronal cell line that secretes LHRH as its transmitter (23, 24). This LHRH neuronal cell line mimics very closely the morphological, biochemical, and functional features of LHRH neurons in situ (23, 24, 25, 26, 27, 28, 29). In our present studies we have determined that the immortalized LHRH neurons express at least two different ₁-AR subtypes. Activation of these receptors leads to a stimulation of LHRH secretion and this response is mediated by PLC and cytosolic PLA₂ (cPLA₂).

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
Reagents
Alprenolol, ascorbic acid, and NE (e.g. arterenol) were purchased from Sigma (St. Louis, MO); phenylephrine (PHE), prazosin, rauwolscine, yohimbine, 5-methyl urapidil, (±)-niguldipine, WB-4101, and BMY 7378 were obtained from Research Biochemicals International (Natick, MA). SKF 105854 was a gift from Dr. J. Paul Hieble (SmithKline Beecham, King of Prussia, PA). The arachidonyl trifluoromethyl ketone (AACOCF₃) was purchased from Biomol (Plymouth Meeting, PA), A23187 was obtained from Calbiochem (La Jolla, CA), neomycin sulfate was purchased from Life Technologies, Inc. (Gaithersburg, MD), and thapsigargin (THA) was obtained from Alexis Corp. (San Diego, CA). [5,6,8,9,11,12,14,15-³H(N)]-arachidonic acid (250 Ci/mmol), [-³²P]deoxy-CTP (5000 Ci/mmol), [¹²⁵I]-2-ß-4-hydroxy-3-iodophenylethylamino-methyltetralone (HEAT) (2200 Ci/mmol), [¹²⁵I]Na for iodination of LHRH, 2-[³H]arachidonyl-phosphatidylcholine (55 mCi/mmol), and [³H]myo-inositol (22.3 Ci/mmol) were obtained from NEN Life Science Products (Boston, MA).

Cell culture
GT1-1 and GT1-7 cells were cultured and maintained in DMEM (Life Technologies, Inc.) as previously described (24). RAW 264.7 and DDT₁-MF2 cells were cultured and maintained in DMEM supplemented with 4.5 mg/ml glucose, 10% FCS (HyClone Laboratories, Inc., Logan, UT), 100 U/ml penicillin-G sodium, and 100 µg/ml streptomycin sulfate (Life Technologies, Inc.).

Animals
Tissues from adult male C57BL/6J mice (The Jackson Laboratory, Bar Harbor, ME) were used for mRNA analyses. Adult male and female New Zealand rabbits (Hazelton Laboratories, Vienna, VA) were used to generate antiserum to cPLA₂. All animal studies were conducted with approved protocols and in accordance with NIH Guidelines for the Care and Use of Animals and the NIEHS and Duke University Medical Center animal care and use committees.

LHRH secretion experiments
Approximately 6 x 10⁵ GT1-1 cells were grown for 48 h in 24-well culture dishes that had been previously coated with Matrigel (Collaborative Biomedical Products, Bedford, MA). At the end of this culture period, cells were washed with PBS and preincubated in Krebs-Ringer bicarbonate glucose (KRBG) buffer for 1 h, followed by two 30-min incubation periods in the same buffer. GT1 neurons were then exposed to the various pharmacological agents for 30 min. The media were collected and analyzed for LHRH content by RIA as previously described (30). The intra- and interassay variabilities were approximately 5% and 10%, respectively.

Radioligand binding assays
GT1 cells were cultured in 150-mm dishes and harvested at 90% confluence. A crude membrane fraction was prepared. Briefly, cells were harvested in an ice-cold hypotonic lysis buffer containing 5 mM Tris (pH 7.4) and 2 mM EDTA with the following proteinase inhibitors: 10 µg/ml soybean trypsin inhibitor, 10 µg/ml benzamidine, 2 µg/ml aprotinin, and 0.1 mM phenylmethylsulfonylfluoride. The cells were disrupted by Polytron homogenization (three 15-sec bursts at 70% power output). A low speed centrifugation (1,000 x g) was performed for 5 min at 4 C to remove nuclei and unbroken cells. The resulting supernatant was centrifuged at 49,000 x g for 20 min at 4 C. The pellet was resuspended in the same buffer and centrifuged a second time. This final pellet was resuspended in the homogenization buffer, and an aliquot was taken for protein determination (31). For radioligand binding assays, 50–100 µg membrane proteins were incubated in a 50 mM Tris (pH 7.4)-5 mM EDTA-150 mM NaCl buffer containing 80 pM [¹²⁵I]HEAT in the presence of increasing concentrations of various unlabeled competing ₁-AR antagonists as indicated in the respective figures. Incubations were performed for 1 h at room temperature with gentle shaking. The reactions were terminated by adding ice-cold 0.5x PBS, and the samples were rapidly collected onto Whatman GF/C filters (Whatman, Hillsboro, OR) with a Brandel cell harvester (Gaithersburg, MD). Competition binding curves were analyzed by least squares nonlinear regression and assessed for best fit to a one- or two-component model with Prism (GraphPad Software, Inc., San Diego, CA). A test for the adequacy of the one- vs. two-component models was performed by ANOVA. P < 0.05 was considered significant.

Expression of ₁-AR subtypes
The expression of individual ₁-ARs was examined in the three clones of GT1 cells by Northern blot. For these analyses, total RNA (30 µg) from the DDT₁-MF2 cells or total (30 µg) and polyadenylated [poly(A)⁺] RNA (10 µg) from the GT1 cells were fractionated on 1.2% agarose gels and transferred to Biodyne nylon membranes (ICN Biomedicals, Inc., Costa Mesa, CA). A cDNA probe to hamster _1B-AR (32) was radiolabeled with [-³²P]deoxy-CTP by nick translation to a specific activity of 8.3 x 10⁸ dpm/µg DNA. Blots were hybridized and washed as previously outlined (30) and exposed to X-OMAT film (Kodak, Rochester, NY).

For the RT-PCR analyses, poly(A)⁺ RNA was isolated from GT1-1 and GT1-7 cells and from different murine tissues using an mRNA purification kit (DynAl, Lake Success, NY). After first strand synthesis using the Superscript kit (Life Technologies, Inc.), PCR analysis was performed. The murine sequences for the _1A- and _1D-AR subtypes were used to select specific oligonucleotide primers for the PCR reactions described below. For the _1A-AR PCR reaction, an upper (5'-TCTTCCATGCCCCAGGGAT-3') and a lower (5'-CTAGACTTCCTCCCCGTTTTCACC-3') primer yielded a reaction product of approximately 201 bp that spanned the region of the transcript from nucleotides 1201–1378 of the mouse sequence (33). The _1D-AR reaction was run with an upper (5'-TTGGGCCGCTACAGAGACC-3') and a lower (5'-TTTGGATCCGAAGGCAGAATC-3') primer that produced a product of approximately 297 bp that included nucleotides 1586–1862 of the mouse sequence (34). Mouse kidney, heart, vas deferens, cerebral cortex, and hypothalamus were used as positive control tissues for these reactions. The negative controls for the PCR reaction included samples run with primers but no template or with RNA from GT1 cells that had not been subjected to first strand synthesis. The reaction conditions for both PCR reactions consisted of an initial denaturation step at 94 C for 2 min, followed by 35 cycles of 94 C for 30 sec, 58 C (_1A-AR reaction) or 60 C (_1D-AR reaction) for 40 sec, and 72 C for 90 sec. The PCR products were separated in a NuSieve 3:1 agarose gel (FMC Corp., Rockport, ME) and verified by sequencing at the Duke University Medical Center facility.

IP analysis
PLC activity was determined by measuring formation of soluble IP as previously reported (35). Briefly, GT1-1 cells (8 x 10⁵ cell/well) were grown for 48 h in 24-well culture plates previously coated with Matrigel. Cells were labeled with 5 µCi/well [³H]myo-inositol (NEN Life Science Products) for 24 h. The next day medium was aspirated, cells were washed twice with PBS (without magnesium or calcium), and preincubated for 10 min with 10 mM LiCl in DMEM or together with various receptor antagonists or enzyme inhibitors. The medium was aspirated, and the cells were incubated with various agents in the presence of LiCl at the concentrations and for the periods indicated in the figures. Incubations were terminated by adding 0.25 ml of an ice-cold solution of 0.6 M perchloric acid-0.2 mg/ml IP6 to the samples. The samples were neutralized with a 1.2 M KOH-75 mM HEPES-60 mM EDTA buffer and loaded onto ion exchange columns (AG 1-X8 resin, 200–400 mesh, formate form; Bio-Rad Laboratories, Inc., Hercules, CA) to separate the IPs. Production of IPs was quantitated by liquid scintillation counting.

Cytoplasmic Ca²⁺ measurements
For measurements of cytoplasmic calcium, 2.5 x 10⁶ cells were grown on Matrigel-coated glass coverslips for 72 h. At the end of this period, coverslips were mounted into a Teflon chamber (Bionique, Denver, CO) and incubated in a HEPES-buffered physiological saline solution containing 116 mM NaCl, 5.4 mM KCl, 0.8 mM MgSO₄, 1.8 mM CaCl₂, 20 mM HEPES, and 10 mM glucose with 1 µM fura-2/AM (Molecular Probes, Inc., Eugene, OR) for 30 min at room temperature. The cells were then washed and incubated in the same buffer at room temperature for at least 20 min before calcium measurements were made. The fluorescence of the cells was monitored with a photomultiplier-based system, mounted on a Nikon Diaphot microscope equipped with a 40x (1.3 N.A.) Neofluor objective (Nikon, Melville, NY). The Deltascan D101 light source (Photon Technology International Ltd., Princeton, NJ) was equipped with a light path chopper that enabled rapid interchange between two excitation wavelengths (340 and 380 nm). Emission fluorescence was monitored at 510 nm with a barrier filter. All experiments were conducted at 25 C in a field of four to eight cells. Calibration and calculation of cytoplasmic calcium were performed as previously described (36).

Arachidonic acid release and lipid analyses
GT1 cells were grown in 24-well plates for 48 h to approximately 70–75% confluence. The cells were labeled for 22 h with 1 µCi/ml [³H]arachidonic acid in DMEM containing 0.01% fatty acid-free BSA (Sigma) and 1% N-2 neuronal supplement (Life Technologies, Inc.). Cells were washed twice for 3 min each time with DMEM containing 0.1% fatty acid-free BSA or, for the thapsigargin experiments, with KRBG buffer containing 0 or 1.8 mM calcium with 0.1% fatty acid-free BSA. Pharmacological agents were administered in this same solution at the concentrations and for the periods indicated in the respective figures. Incubations were terminated by collection of the medium on ice. The medium was centrifuged at low speed to remove cells and was measured for [³H]arachidonic acid content by liquid scintillation counting. In some experiments medium was extracted with a chloroform-methanol (2:1, vol/vol) solution. The extract was evaporated and analyzed by TLC on LHPKDF plates (Whatman, Clifton, NJ) with a solvent system consisting of n-heptane-isopropyl ether-acetic acid (10:13:0.66, vol/vol/vol) confirming that most of the radioactivity released into the medium was free arachidonic acid. The lipid positions were identified by comigration with authentic lipid standards (Avanti Polar Lipids, Birmingham, AL) and were visualized by exposure to iodine. These lipid samples were scraped from the plates, mixed with liquid scintillation fluid, and quantitated in a liquid scintillation counter. Alternatively, plates were analyzed with a Image 200 densitometer (Bioscan, Inc., Washington, DC).

Lipid contents of the cells were examined at the same time as medium analyses. Cells were quickly lysed and scraped into either 1 M NaCl or 0.1% SDS and then extracted with a chloroform-methanol (2:1, vol/vol) solution. In some experiments before lipid extraction an aliquot was taken for protein measurement (31). The lipid extracts were submitted to both liquid scintillation counting and analyses by TLC on LHPKDF or LK6DF plates (Whatman). Arachidonyl-containing diacylglycerol and other neutral lipids were analyzed by the solvent system described above. Polar lipids were analyzed by TLC with a chloroform-methanol-ammonium hydroxide-water (70:25:3.5:1.5, vol/vol/vol/vol) solvent system for the first run and a chloroform-methanol-acetic acid-water (80:10:2:0.75, vol/vol/vol/vol) system for the second run. Identification and quantitation of lipids were performed as described above.

cPLA₂ activity measurements
GT1-1 cells and RAW 264.7 were grown in 60-mm dishes until they reached approximately 90% confluence. RAW 264.7 cells were used as a positive control for cPLA₂ enzymatic activity determinations and Western blot analyses. GT1-1 and RAW 264.7 cells were washed twice with warm PBS and preincubated for 2 h in the KRBG buffer in the absence of calcium chloride but containing 100 µM EGTA. After this period, cells were incubated for 15 or 30 min in KRBG buffer (with 1.8 mM CaCl₂) containing various pharmacological agents. Controls included incubation of the different agents in the presence or absence of calcium. After the stimulation period, cells were quickly scraped, centrifuged at low speed, and resuspended in 0.5 ml ice-cold lysis buffer [20 mM HEPES (pH 7.4), 032 M sucrose, 2 mM EGTA, 5 mM dithiothreitol, 2 mM phenylmethylsulfonylfluoride, 1 µg/ml pepstatin, and 100 µg/ml leupeptin]. Samples were sonicated on ice for 20 sec and centrifuged at 6,000 rpm for 4 min at 4 C. This supernatant was centrifuged at 100,000 x g for 1 h. The supernatant (soluble fraction) was collected, and the pellet (membrane pellet fraction) was resuspended in 0.3 ml ice-cold lysis buffer and sonicated on ice for 20 sec. Aliquots from these soluble and membrane pellet samples were taken for protein measurement, Western blot analysis, and cPLA₂ activity determinations.

The cPLA₂ enzymatic activity was determined using a modification of an assay (37). Briefly, substrate was prepared by resuspending lyophilized 1-palmitoyl, 2-[¹⁴C]arachidonyl phosphatidylcholine in 2 µl dimethylsulfoxide to a final concentration of 15 µM. The assay conditions used in this study were optimized for GT1 cells. Enzymatic reactions were initiated by the addition of 34 µl from the soluble or membrane pellet samples (same amount of protein) and by bringing the solution to 3 mM calcium chloride and 4 mM dithiothreitol. Incubations were performed at 37 C for 30 min, and they were terminated by the addition of 40 µl of an ice-cold chloroform-methanol (1:1, vol/vol) solution containing 2% acetic acid. A 30-µl aliquot was analyzed by TLC (LK6DF plates) using the solvent system for neutral lipids described above. Arachidonic acid, diacylglycerol, and phospholipid positions were identified and analyzed as indicated above. Enzymatic activity was calculated as the percentage of hydrolyzed arachidonic acid from 2-arachidonyl-phosphatidylcholine. Results were expressed as the ratio of the enzymatic activities in the membrane pellet with respect to the soluble fractions, where the activity ratio for the basal condition was set equal to 1.

Western blot analysis
Soluble and membrane pellet fractions were obtained from GT1-1 (300 µg total protein) and RAW 264.7 cells (100 µg total protein) as described above. Samples were fractionated on an 8% SDS-PAGE gel (38) and transferred to an Immobilon-P membrane (Millipore Corp., Bedford, MA). For the immunodetection of cPLA₂, two antisera were used: a monoclonal antibody raised against mouse cPLA₂ (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) and a polyclonal antiserum produced against mouse cPLA₂ and generated in our laboratory. Briefly, a peptide was synthesized that included amino acids 152–179 of the murine cPLA₂ sequence (39). This peptide was conjugated to bovine thyroglobulin by carbodiimide (Sigma), the protein was dialyzed, and antisera were generated as previously described (38). The specificity of the antiserum was verified by Western blotting and preabsorption studies. Secondary antimouse or antirabbit antibodies coupled to horseradish peroxidase (Kirkegaard & Perry Laboratories, Gaithersburg, MD) were used, and Western blots were developed by chemiluminescence (Pierce Chemical Co., Rockford, IL).

For the PLC Western blots, antisera to PLCß1, -ß2, -ß3, and ß4 (Santa Cruz Biotechnology, Inc.) were used. The immunizing peptide used to generate each antiserum was used to demonstrate the specificity of the immunostaining. The procedures for these analyses are described above.

Statistics
All data are presented as the mean and SEM. The data were submitted to ANOVAs, with time or treatment as the independent variables. The a posteriori comparisons were made using Scheffé’s or Newman-Keuls’ tests (40).

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Activation of ₁-ARs stimulates LHRH secretion
To determine whether LHRH neurons can directly respond to ₁-AR stimulation, immortalized hypothalamic LHRH neurons were exposed to different concentrations of the ₁-AR agonists, PHE or NE. in the presence of ascorbate and a ß-AR blocker (alprenolol; Figs. 1, A and B). Compared with the unstimulated control (basal), both PHE and NE stimulated LHRH secretion in a dose-dependent manner. Neither alprenolol nor ascorbate had any effect on LHRH release when administered alone (data not shown).

View larger version (31K): [in this window] [in a new window]   Figure 1. Dose-dependency and pharmacological specificity of LHRH release in response to 1-AR agents. A and B, GT1 cells were preincubated with KRBG buffer and then stimulated for 30 min with KRGB buffer alone (for the 0-µM dose of phenylephrine), varying concentrations (5–100 µM) of PHE, KRGB buffer with 60 µM ascorbate and 100 µM alprenolol (for the 0 µM dose of norepinephrine), or varying concentrations (5–100 µM) of NE with the ascorbate and alprenolol. Media were collected and analyzed by RIA. Neither alprenolol nor ascorbate interfered with the RIA. C, GT1-1 cells were preincubated for 30 min in KRBG buffer containing various adrenergic antagonists. Neurons were then incubated for 30 min with KRBG buffer alone (basal secretion), with 100 µM PHE or 100 µM NE (plus 60 µM ascorbate) alone or in combination with propranolol, prazosin (PRA), rauwolscine (RAW), yohimbrine (YOH), or alprenolol (ALP). The data are presented as an average of four independent experiments. *, P < 0.05 vs. basal secretion in KRBG buffer.

Ligand binding studies for ₁-ARs
Competition binding studies were conducted with [¹²⁵I]HEAT to ascertain whether the GT1 cells contained binding sites for the ₁-ARs. Binding analyses with (±)-niguldipine (Fig. 2A), 5-methylurapidil (Fig. 2B), or WB-4104 (Fig. 2C) revealed that the GT1-1 cells contained at least two different populations of binding sites (Table 1). These sites consisted of a high affinity site, representing approximately 22–27% of the total number of sites, and a low affinity site, comprising approximately 73–78% of the sites. From findings in other published reports (22), our data suggest that the high affinity binding site corresponds to the _1A-AR, while the low affinity site represents the _1B- and/or _1D-ARs.

View larger version (23K): [in this window] [in a new window]   Figure 2. Ligand binding analyses of 1-ARs on GT1 cells. A–E, Competitive binding assays were performed on GT1 cell crude membrane preparations using [125I]HEAT and subtype-specific 1-AR antagonists: (±)-niguldipine (A), 5-methyl urapidil (B), WB-4101 (C), BMY7378 (D), and SKF105854 (E). The competition curves are representative of two or three independent experiments.

View this table: [in this window] [in a new window]   Table 1. Summary of antagonist competition binding studies for 1-AR subtypes in GT1 cells

RNA expression studies of ₁-AR subtypes
In addition to the radioligand binding studies, expression of the 1-AR subtypes in GT1 neurons was investigated by Northern blot and RT-PCR. In the Northern blot experiments, only the _1B-AR mRNA was detected in the LHRH neuronal cell line (DDT₁-MF-2 cells were used as a positive control; Fig. 3A); no other subtypes could be detected by this method using rat cDNA probes to the _1A- and _1D-ARs (data not shown). This result was anticipated because detection of some of the ₁-AR subtypes are difficult by Northern blot and ribonuclease protection analyses, and PCR analyses are normally required to verify their expression even in tissues known to contain high levels of these receptors (22).

View larger version (27K): [in this window] [in a new window]   Figure 3. Expression of 1-AR subtypes in the GT1 cells. A, Total (30 µg) RNA from DDT1-MF2 cells (lane 1) or poly(A)+ RNA (10 µg; lane 2) or total RNA (30 µg; lane 3) from GT1 cells were analyzed by Northern blot using a hamster 1B-AR cDNA probe. A single hybridizing band of the same mol wt (2.4 kb) was observed in samples from the DDT1-MF2 and GT1-1 cells. B, Poly(A)+ RNA was isolated from mouse kidney (lane 2), heart (lane 3), vas deferens (lane 4), cortex (lane 5), and hypothalamus (lane 6) as well as from the GT1-1 (lane 7) and GT1-7 (lane 8) cells and subjected to RT-PCR. Specific primers from the murine sequences for the 1A- and 1D-AR subtypes were used in the PCR reactions. Negative controls consisted of PCR reactions that used nonreverse transcribed GT1 poly(A)+ RNA (lane 9) or that contained only primers and no template (lane 10). The positions of the DNA mol wt markers are shown (lane 1). A specific PCR product is observed in the GT1 cells only for the 1A-AR (200 bp; see top panel).

₁-AR activation leads to stimulation of PLC
Although activation of the ₁-ARs is commonly associated with stimulation of PLC, in some tissues activation of these receptors can also stimulate PLA₂, phospholipase D, and voltage-activated calcium channels (44). To determine whether PLC was an effector for the ₁-ARs in the immortalized hypothalamic neurons, cells were analyzed for production of soluble IPs and diacylglycerol after PHE and NE treatments. Both ₁-AR agents stimulated the production of IPs within the first 30 sec of activation, and maximal levels were achieved after 5–10 min of stimulation (Fig. 4A). Diacylglycerol followed a similar time course, and the highest levels were reached within the first 5 min of activation (Fig. 4B).

View larger version (19K): [in this window] [in a new window]   Figure 4. 1-AR stimulation of GT1 cells leads to the production of IPs and diacylglycerol and the release of intracellular calcium. A, To study IP production in GT1 cells, the neurons were cultured for 48 h and labeled with 5 µCi [3H]myo-inositol for 24 h. Cells were preincubated with 10 mM LiCl for 10 min, then exposed to 100 µM PHE for the indicated periods of time, and IP production was measured. B, To examine diacylglycerol production, GT1 cells were cultured as described above and labeled with 1 µCi [3H]arachidonic acid/ml medium for 22 h. Cells were incubated with 100 µM PHE for the indicated periods of time, and cellular [3H]arachidonyl-diacylglycerol (DAG) was measured. C, To evaluate the dose-dependency of IP production, GT1 neurons were preincubated with 10 mM LiCl for 10 min, then exposed to varying concentrations of NE (in the presence of alprenolol and ascorbate) or PHE for 10 min in the continuous presence of LiCl; media were collected and analyzed for IP content. The data are expressed as the percentage of IPs produced, where 100% corresponds to IP production at the 100-µM concentration of agonist (6307 and 3039 dpm/well for NE and PHE, respectively), and 0% is the basal IP production (2217 and 1109 dpm/well for the same two respective agents). D, To examine the pharmacological specificity of the IP response, GT1 cells were exposed to 100 µM NE (with ascorbate) and several adrenergic receptor antagonists (see Fig. 1) in the continued presence of LiCl for 10 min. 100% IP production for 100 µM NE = 4713 dpm/well and 0% (BASAL) = 2561 dpm/well. *, P < 0.05 vs. basal. E, To ascertain which PLCß isoforms are expressed in GT1 cells, mouse cerebellum and GT1 neurons were homogenized, and proteins were extracted, separated by SDS-PAGE, and submitted to Western blotting using primary antiserum alone (-) or antiserum preabsorbed to the immunizing peptide (+). As PLCß4 is highly expressed in cerebellum, this murine tissue was used as a positive control for this isoform. Note that the immunoreactive band identified for PLCß2 is unlikely to represent this enzyme in GT1 cells, because the size obtained is much smaller than that reported for this enzyme and because this gene is primarily expressed in immune cells. F, To study cytoplasmic calcium responses to adrenergic agents, GT1 cells were preloaded with fura-2/AM. Calcium levels were monitored by fluorescence imaging for at least 300 sec. PHE was added to the cells at 50 sec in calcium-free medium. Extracellular calcium levels were restored to 1.8 mM at 150 sec. The data in all panels are representative of two or three independent experiments.

To determine the pharmacological specificity of the response, GT1 neurons were stimulated with NE in the presence of an ₁-AR antagonist (prazosin), ₂-AR antagonists (rauwolscine or yohimbine), or a ß-AR blocker (propranolol). Compared with the unstimulated control, production of IPs was augmented by NE, and this response was only inhibited by prazosin (Fig. 4D). These data suggest that NE can stimulate PLC through the ₁-AR.

Because in most mammalian cells PLCß is responsible for IP production after activation of G protein-coupled receptors (45), Western blots were used to investigate which PLCß isoenzymes were expressed in GT1 cells. Figure 4E reveals that GT1 cells contain immunoreactive materials that are recognized by PLCß1 and PLCß3 antisera. There was no evidence that PLCß4 is expressed in GT1 neurons. This result was expected, because this isoform has been reported to reside in the cerebellum (45). With respect to the PLCß2 isoform, it is unlikely that the immunoreactive band migrating at approximately 107,000 mol wt is authentic for two reasons. First, the size of the native enzyme in tissues is approximately 150,000 mol wt (45). Second, Northern blots using a rat PLCß2 probe failed to identify the expression of this gene in GT1 cells (data not shown). Hence, the PLCß1 and PLCß3 isoenzymes are probably responsible for generating the IPs and diacylglycerol in response to ₁-AR stimulation in the immortalized LHRH neurons.

Activation of ₁-ARs leads to a release of cytoplasmic calcium
One consequence of PLC stimulation is the production of IP3 and the release of calcium from intracellular stores, resulting in an increase in cytoplasmic calcium concentrations (46). Mobilization of cytoplasmic calcium was monitored with fura-2/AM in cells stimulated with PHE under nominally calcium-free conditions, followed by restoration of extracellular calcium concentrations to physiological levels (1.8 mM CaCl₂). Two different phases of cytoplasmic calcium increases were observed (Fig. 4F). The first phase resulted from the PHE-stimulated IP3-induced release of calcium from intracellular stores. A depletion of the intracellular calcium pool, in turn, signals the activation of a calcium influx pathway termed capacitative calcium entry (47). This second phase was detected by restoring intracellular calcium to physiological levels. Our data clearly demonstrate that ₁-AR stimulation in the GT1 cells involves activation of PLC signaling and the mobilization of intracellular calcium.

₁-AR stimulation induces arachidonic acid release
In some systems ₁-AR stimulation has been linked to activation of PLA₂ and arachidonic acid release (44). We examined whether stimulation of ₁-ARs in the immortalized LHRH neurons was also associated with PLA₂ responses. To study arachidonic acid release, the incorporation of [³H]arachidonic acid into lipids was first optimized for GT1 cells. A plateau of [³H]arachidonic acid incorporation was reached after 20 h of labeling, and approximately 95–98% of the total [³H]arachidonic acid was incorporated into lipids, with the remainder found in the medium (Fig. 5A, inset). In all experiments, at least 95% of the total arachidonic acid was esterified in the phospholipid fraction, whereas less than 5% was associated with neutral lipids (data not shown). After 20 h of incorporation, 24%, 40%, and 36% of the total [³H]arachidonic acid were esterified to phosphatidylinositol (PI), phosphatidylethanolamine (PE), and phosphatidylcholine (PC), respectively. The phospholipid composition (mass) in GT1-1 cells was 10% for PI, 30–33% for PE, and 40–44% for PC. Thus, the estimated specific incorporation into each phospholipid species suggested that the most rapid and highest incorporation occurred in the PI, followed by the PE, and then the PC fraction.

View larger version (29K): [in this window] [in a new window]   Figure 5. Arachidonic acid release elicited by 1-AR stimulation of GT1 cells. A, To examine the time course of [3H]arachidonic acid release, GT1 cells were prelabeled with 1 µCi [3H]arachidonic acid/ml for 22 h. Cells were stimulated with 50 µM NE (plus 50 µM ascorbic acid, 100 µM propranolol, and 100 µM yohimbine) or 100 µM PHE for the indicated periods of time. Medium was collected and analyzed for [3H]arachidonic acid content. Inset, cells were prelabeled with 1 µCi [3H]arachidonic acid/ml for the indicated periods of time, and the media and cells were collected and analyzed for [3H]arachidonic acid that was free or incorporated into lipids. The data are expressed as the percentage of [3H]arachidonic acid remaining in the medium (M; 100% represents the amount at 0 h) and that incorporated into cellular PC, PE, and PI at each time point with respect to 24 h (100% incorporation). B, To study the dose-response relationships, GT1-1 cells were labeled for 22 h and stimulated for 30 min with NE (in the presence of 60 µM ascorbate, 100 µM propranolol, and 100 µM yohimbine) or PHE at the specified concentrations. The data are expressed as the percentage of [3H]arachidonic acid release, where 100% refers to the release at the 100-µM dose (1091 and 1269 dpm/well for NE and PHE, respectively), and 0% represents basal conditions (471 and 399 dpm/well for the NE and PHE, respectively). The triangles to the right of the curves depict [3H]arachidonic acid release upon stimulation with 50 µM NE () or PHE () in the presence of 50 µM prazosin. The data are representative of two or three independent experiments. C, To analyze cPLA2 activity, GT1 and RAW 264.7 cells were preincubated in calcium-free KRBG buffer with 100 µM EGTA for 2 h and then incubated for 30 min in KRBG medium containing 1.8 mM CaCl2 in the absence (basal) or presence of 100 µM PHE or 10 µM A23187 (ION) alone or in combination with 100 µM EGTA. Soluble and membrane pellet fractions were prepared and assayed for PLA2 activity in vitro and expressed as the ratio of the activities in the membrane pellet to soluble fractions relative to basal conditions (activity ratio = 1). The results are presented as the average of two independent experiments. *, P < 0.05 vs. basal. D, Western blot analysis of the cPLA2 protein in soluble and membrane pellet fractions (pellet) from unstimulated cells (GT1 cells = 300 µg protein and RAW 264.7 cells = 100 µg protein) analyzed with two different antisera against the mouse cPLA2 [Santa Cruz Biotechnology, Inc. (a) and an antiserum from our laboratory (b and c)]. Preabsorption with the immunizing peptide is shown in C. The arrowhead indicates the position of the immunoreactive cPLA2 at approximately 110,000 mol wt.

Arachidonic acid release is mediated by cPLA₂
It is widely accepted that receptor-activated arachidonic acid release is mediated by PLA₂ (48); however, this eicosanoid can also be released through the actions of other lipases. Despite this fact, the results from our studies with GT1 cells suggest that the ₁-AR-mediated arachidonic acid release derives solely from the action of PLA₂. For instance, careful monitoring of the different arachidonate-containing lipid fractions during ₁-AR stimulation failed to reveal any detectable changes in arachidonate-containing neutral lipid fractions (e.g. mono- and triacylglycerol). By comparison, ₁-adrenergic stimulation was related to a decrease in the amount of arachidonate-labeled PI species and a concomitant increase in lyso-PI in a time- and dose-dependent manner (data not shown). No changes were observed in any other phospholipid or lysophospholipid factions (data not shown). Together, these findings support the idea that the release of arachidonic acid is due to PLA₂ activity rather than merely to the activity of alternative lipases.

To determine which PLA₂ species were recruited by ₁-adrenergic stimulation of the GT1 neurons, PLA₂ activity was examined by an in vitro assay. This procedure can distinguish the calcium-dependent cPLA₂, the secretory PLA₂ enzymes, and the calcium-independent PLA₂ forms from each other. To determine whether the calcium-dependent or -independent forms of PLA₂ were activated by ₁-adrenergic agents, cells were stimulated with PHE in the presence or absence of EGTA (calcium-free conditions). To study the activity classically attributed to cPLA₂, all enzyme assays were run in the presence of a high concentration of ß-mercaptoethanol that completely inactivates the secretory PLA₂ enzyme (37, 48). Additionally, because the cPLA₂ translocates from the cytosol to the plasma membrane upon activation (48, 49), PLA₂ activity was measured in both fractions after activation. The results indicate that upon ₁-adrenergic activation, PLA₂ activity increases in the membrane pellet fraction with a consequent decrease in the soluble fraction. Moreover, calcium-free conditions completely abolished the enzymatic activity (Fig. 5C, left), whereas the absence of ß-mercaptoethanol in the assay exerted no affect on PHE stimulation of arachidonic acid release (not shown). Additionally, the calcium ionophore ionomycin, a well known activator of cPLA₂ (48), exerted a similar effect in GT1 cells (Fig. 5C, left) as well as in a macrophage cell line that contains high levels of this enzyme (49) (Fig. 5C, right). These data indicate that ₁-AR stimulation promotes the translocation of a calcium-dependent PLA₂ activity from the cytosol to the membrane fraction, and these activities are resistant to thiol-reducing agents.

Western blot analysis using two different cPLA₂ antibodies revealed that the immortalized LHRH neurons contain cPLA₂ (Fig. 5D). In this case, immunoreactive bands of approximately 110,000 mol wt could be clearly discerned in the soluble and membrane pellet fractions from both GT1 cells and the RAW 264.7 macrophage cell line. Preabsorption of our antisera with the immunizing peptide resulted in the complete abolition of immunostaining in both cell lines.

To independently evaluate the role of cPLA₂ in the release of arachidonic acid, a specific inhibitor of cPLA₂ (AACOCF3) was used. Here, 10 µM AACOF3 completely abolished the PHE-stimulated release of arachidonic acid (Fig. 6C). Collectively, these results suggest that activation of ₁-ARs leads to stimulation of cPLA₂ and release of arachidonic acid from GT1 cells.

View larger version (19K): [in this window] [in a new window]   Figure 6. Role of PLC and cPLA2 in mediating 1-adrenergic stimulation of LHRH secretion. GT1 cells were assayed for LHRH secretion (A), IP production (B), and arachidonic acid release (C) as described in Figs. 1, 4, and 5 after incubation with several agents alone or in combination: 100 µM PHE, 3 mM neomycin sulfate (NEO), and 10 µM AACOCF3 (AAC). The data are representative of two or three experiments. *, P < 0.05 vs. the unstimulated control (basal); +, P < 0.05 vs. PHE treatment.

Contributions of PLC and PLA₂ to ₁-AR-mediated LHRH secretion

To demonstrate that neomycin was effective in inhibiting PLC activity in GT1 cells, neurons were exposed to this agent and stimulated with PHE. Compared with the PHE-treated group, 3 mM neomycin depressed the production of soluble IPs by at least 75% (Fig. 6B). Moreover, neomycin reduced the production of IPs in a dose-dependent manner (data not shown). As anticipated, neomycin treatment alone had no effect on IP production in unstimulated cells. These data imply that PLC plays a direct role in ₁-AR-mediated LHRH secretion.

To study the role of cPLA₂ in controlling PHE-stimulated LHRH secretion, GT1 cells were incubated with PHE in the presence of the inhibitor, AACOCF₃. AACOCF₃ (10 µM) reduced the PHE-stimulated LHRH secretory response by approximately 65% (Fig. 6A). AACOCF₃ alone had no effect on basal LHRH release. As noted in the previous section, this inhibitor completely blocked the PHE-stimulated release of arachidonic acid (Fig. 6C). Collectively, these data clearly implicate cPLA₂ in the ₁-AR signaling cascade.

We decided to further investigate the effects of neomycin on PLC and cPLA₂. Blockade of PLC with neomycin inhibited not only PHE-stimulated LHRH secretion and depressed IP production, but it also eliminated the release of arachidonic acid (Fig. 6). This effect suggests that neomycin may be able to inhibit both PLC and cPLA₂. To examine this latter possibility, in vitro PLA₂ activity was evaluated in the presence or absence of neomycin. Neomycin (3 mM) exerted no effect on arachidonic acid release (data not shown). These data suggest that neomycin can inhibit PLC, but not PLA₂, in GT1 cells.

Neomycin is known to inhibit PLC and to block the entry of extracellular calcium through plasma membrane Ca²⁺ channels (50). As PLC activation produces IP3, and this lipid can mobilize cytoplasmic calcium stores (46), leading to an influx of extracellular calcium through the capacitive calcium pathway (47), neomycin would be expected to significantly depress both calcium efflux and influx after ₁-AR stimulation. To examine the roles of these two sources of calcium on PLA₂ activity, GT1 cells were prelabeled with 1 µCi [³H]arachidonic acid for 22 h in DMEM. Cells were washed with KRGB medium containing 0 or 1.8 mM CaCl₂ and exposed for 30 min to vehicle (0.1% dimethylsulfoxide), 100 µM PHE, or 1 µM THA in the same medium. Under the 0 mM CaCl₂ condition, both PHE and THA significantly stimulated arachidonic acid release (Fig. 7, left). As no extracellular calcium was present in this experiment and because THA is a potent releaser of cytoplasmic calcium stores, these data suggest that PHE can also activate calcium efflux (see Fig. 4F). This response is adequate to stimulate arachidonic acid release. In the 1.8 mM extracellular CaCl₂ condition, basal arachidonic acid release was higher than in the former experiment, probably because the cPLA₂ is responsive to calcium (compare Fig. 7, left and right; see also Fig. 5C). PHE and THA efficiently stimulated arachidonic acid release over this baseline (Fig. 7, right). The response to THA was higher than that to PHE, probably because the former agent is more potent in activating the capacitive calcium current where calcium influx would enhance the response (49). In summary, these data indicate that mobilization of cytoplasmic calcium stores is sufficient to stimulate arachidonic acid release, and this response is further potentiated by extracellular calcium concentrations.

View larger version (19K): [in this window] [in a new window]   Figure 7. Role of intracellular calcium stores in mediating 1-adrenergic stimulation of arachidonic acid release. GT1 cells were prelabeled for 22 h [3H]arachidonic acid as described in Fig. 5. Neurons were incubated with 0.1% dimethylsulfoxide or vehicle (basal), 100 µM PHE, or 1 µM THA in the presence of 0 (left panel) or 1.8 mM CaCl2 (right panel) for 30 min. Medium was collected and assayed for [3H]arachidonic acid contents. The data are depicted as 0% arachidonic acid release (basal release in the presence of 0 mM extracellular calcium, 2987 dpm/well) to 100% arachidonic acid release (THA release in the presence of 1.8 mM extracellular calcium, 7077 dpm/well). *, P < 0.05 vs. each respective basal secretion.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Catecholamines have long been known to regulate LHRH secretion, and NE has been ascribed a prominent role in this regard (1, 2, 3, 4, 5, 6, 7, 8, 9). In fact, NE is secreted into the hypophyseal circulation in a pulsatile manner that is synchronous with LHRH and LH release (51). Prazosin can suppress LHRH secretion, whereas other ₂- and ß-AR antagonists are without effect (52). These results clearly support a role for ₁-ARs in regulating the preovulatory surge. Nonetheless, it has not been clear whether the noradrenergic effects are direct or whether they are indirect and are mediated by other neighboring neurons (10, 11, 12, 13). Electron microscopy studies have shown that although some noradrenergic nerve terminals synapse directly on LHRH dendrites, the numbers of these contacts are very low (15). In contrast to the perikaryal region, the status of the connections at the level of the LHRH nerve terminals is unknown. Although our present findings with the immortalized LHRH neurons do not directly address this latter point, they clearly illustrate that these cells are responsive to ₁-adrenergic agents and that ₁-AR activation directly stimulates LHRH secretion from GT1 cells.

To date, both ligand binding and cloning experiments have identified three different ₁-AR subtypes (22). Transcripts of all three receptor subtypes are found in the hypothalamus, with expression of the _1B-AR being the most prominent (17, 18, 19, 20, 21). In our studies we used both ligand binding and gene expression analyses to identify the receptor subtypes that are present in the immortalized LHRH neurons. These findings reveal that the GT1 cells contain not only _1B-AR, but also _1A-AR. The proportion of _1B-ARs is approximately 3-fold higher than that of the _1A subtype. By contrast, repeated attempts using ligand binding and PCR approaches failed to identify any _1D-ARs. Thus, if these receptors are present, then they must reside at extremely low concentrations. Several points should be made. First, our ligand binding data replicate a report by Al-Damluji and colleagues (53), who also demonstrated that the GT1-1 cells contain ₁-ARs; the subtypes of receptors were not identified. Second, our secretion findings are not consistent with those of Martínez de la Escalera and co-workers (42), who showed that NE only stimulated cAMP production and LHRH release in GT1 cell lines; prazosin was unable to modify NE-stimulated LHRH release. Hence, in their experiments NE appeared to only activate the ß-ARs. On the other hand, in a subsequent publication the same investigator reported that NE stimulated IP production, and this could be blocked with phentolamine (54). As ₁-ARs, but not ß-ARs, stimulate IP production, the reasons for the discrepancy between their results and ours are not clear. Finally, our findings that GT1 cells contain ₁-ARs may be physiologically relevant, as Hosny and Jennes (16) reported that LHRH perikarya in rats contain _1B-AR immunoreactivity (16). Whether LHRH neurons in vivo also contain other ₁-ARs is unknown, but it will probably be difficult to discern because these receptors are expressed at very low levels (17, 18, 19, 20, 21, 22).

The ₁-ARs belong to the family of G protein-coupled receptors. Activation of ₁-ARs has been reported to stimulate a number of different signal transduction pathways that include PLC, PLA₂, PLD, and voltage-activated calcium channels (44). Presently, there are no conclusive data indicating that a particular ₁-AR subtype is preferentially linked to a particular signal transduction pathway. Instead, it appears as though ₁-AR signaling may be tissue or cell specific. In the present study we found that activation of ₁-ARs in the immortalized LHRH neurons leads to a time- and dose-dependent production of diacylglycerol and IPs. Moreover, two independent PLC inhibitors were able to block this response. For instance, neomycin was found to block IP production in a dose-dependent manner up to at least 75%. The blockade was not complete, but this magnitude of inhibition was sufficient to reduce ₁-AR-stimulated LHRH secretion to basal levels. Additionally, we tested U-73122, a reputed inhibitor of PLC (55), and found it to be effective in blocking IP production in the GT1 cells. Despite this fact, it was a strong stimulator of LHRH release. Recent reports have also found the U-73122 to be nonspecific in its actions (56, 57).

Besides stimulating the production of IPs and diacylglycerol, ₁-adrenergic agents stimulated the release of intracellular calcium. This effect probably occurs as a result of IP3 binding to its endoplasmic receptor and stimulating the mobilization of cytoplasmic calcium (46, 47). The depletion of the intracellular calcium pool, in turn, resulted in a second increase in cytoplasmic calcium due to an influx of extracellular calcium. This biphasic response is typical of IP3-dependent mobilization of cytoplasmic calcium that is coupled to receptor activation (47). In PHE-stimulated GT1 cells, the increase in intracellular calcium appears to be sufficient for PLA₂ activation and arachidonic acid release, and it may also be sufficient on its own to activate additional signaling cascades and regulate the secretion of LHRH (28). For example, one might anticipate that the ₁-AR-stimulated PLC production of diacylglycerol and the mobilization of cytoplasmic calcium would also activate the calcium-dependent forms of PKC (46). Although this pathway was not examined in the present study, it is well known that stimulation of these enzymes by phorbol esters can lead to robust LHRH secretion from median eminence tissue fragments and from the immortalized LHRH neurons (28).

An additional signaling pathway that we found to be activated by ₁-adrenergic stimulation was the cPLA₂-mediated release of arachidonic acid. In some systems cPLA₂ has been reported to be directly coupled to receptors by G proteins (58), whereas in others cytoplasmic calcium and MAPK activate cPLA₂ (59, 60). In the present study cPLA₂ did not appear to be directly coupled to the ₁-AR. Activation of these receptors stimulates PLC to produce IP3, and this lipid mobilizes cytoplasmic calcium. This increase in cytoplasmic calcium efflux is sufficient to stimulate PLA₂ and the release of arachidonic acid. This cascade of events suggests that cPLA₂ signaling is downstream of PLC activation and cytoplasmic calcium mobilization.

Although both PLC and cPLA₂ participate in the ₁-AR-mediated LHRH secretory response, our findings show that their contributions are not equal. For instance, in the GT1 cells neomycin sulfate completely inhibited PHE-stimulated LHRH secretion, it successfully depressed IP production, and it eliminated the release of arachidonic acid. By contrast, AACOCF₃ reduced LHRH secretion by approximately 65%, it exerted no effect on IP production, but it completely suppressed arachidonic acid release. As PHE-stimulated LHRH secretion in the presence of the specific cPLA₂ inhibitor, AACOCF₃, was only about 35% of maximal release, these data suggest that the contribution of PLC to LHRH secretion was much less than that associated with cPLA₂ activation. Thus, because cPLA₂ activity can be regulated by cytoplasmic calcium and extracellular calcium concentrations can potentiate this activity, it would appear that cPLA₂ can augment and modify the LHRH secretory response. It should be emphasized that although all of our studies were conducted using static cultures of the GT1 neurons, it is likely that the signal transduction and consequent LHRH secretory responses would be even more robust in a perifusion system or in vivo.

The binding of NE to ₁-ARs has long been known to play a critical role in reproduction (1, 2, 3, 4, 5, 6, 7, 8, 9). The ₁-AR signal transduction pathway that we have identified in the immortalized LHRH neurons includes IP and diacylglycerol production, the mobilization of cytoplasmic calcium, and the release of arachidonic acid. As each of these components can generate their own signaling and/or trafficking cascades, it is anticipated that additional second messengers contribute to the LHRH secretory response. It should be recalled that arachidonic acid can be metabolized to many different eicosanoid products and at least one of these lipids (e.g. PGE₂) has been reported to stimulate LHRH secretion in vitro (7, 28). The diversity of signaling cascades that are activated by ₁-AR stimulation may serve to ultimately control the frequency, amplitude, and time course of LHRH secretion in vivo. Inasmuch as the GT1 neurons represent a good approximation of LHRH neurons in vivo (28) and because noradrenergic stimulation plays a critical role in reproduction (1, 2, 3, 4, 5, 6, 7, 8, 9), analyses of the intracellular events following ₁-AR activation may provide some unique insights into the signaling mechanisms and molecular interactions that are necessary to mount a successful proestrous surge.

	Acknowledgments

 
We thank Dr. P. Mellon (University of California-San Diego, La Jolla, CA) for providing us with the GT1 cells, Dr. T. Eling for giving us the RAW 264.7 cells, the Protein Chemistry Laboratory (University of North Carolina, Chapel Hill, NC) for synthesizing the 28-amino acid peptide used to generate the cPLA₂ antiserum, and Dr. A. Arimura (Tulane University, New Orleans, LA) for providing the A772 LHRH antiserum. We also express our thanks to Dr. Susanna Cotecchia (University of Lausanne, Lausanne, Switzerland) for providing us with the unpublished nucleotide sequences for the ₁-ARs.

	Footnotes

 
This work was supported by the Intramural NIEHS program (to S.S. and W.C.W.) and NIH Grant HD-36015 (to W.C.W.).

¹ Current address: Cystic Fibrosis/Pulmonary Research and Treatment Center, University of North Carolina, Chapel Hill, North Carolina 27599.

Abbreviations: AACOCF₃, Arachidonyl trifluoromethyl ketone; ₁-AR, ₁-adrenergic receptors; cPLA₂, cytosolic PLA₂; HEAT, [¹²⁵I]-2-ß-4-hydroxy-3-iodophenylethylaminomethyltetralone; KRBG, Krebs-Ringer bicarbonate glucose; LHRH, LH-releasing hormone; NE, norepinephrine; PC, phosphatidylcholine; PE, phosphatidylethanolamine; PHE, phenylephrine; PI, phosphatidylinositol; poly(A)⁺, polyadenylated; THA, thapsigargin.

Received January 24, 2001.

Accepted for publication July 30, 2001.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Sawyer C, Markee J, Hollinshead W 1947 Inhibition of ovulation in the rabbit by the adrenergic-blocking agent dibenamine. Endocrinology 41:395–402
2. Barraclough CA, Wise PM, Selmanoff MK 1984 A role for hypothalamic catecholamines in the regulation of gonadotropin secretion. Recent Prog Horm Res 40:487–529
3. McCann SM 1982 Physiology and pharmacology of LHRH and somatostatin. Annu Rev Pharmacol Toxicol 22:491–515[CrossRef][Medline]
4. Weick R 1978 Acute effects of adrenergic receptor blocking drugs and neuroleptic agents in pulsatile discharges of luteinizing hormone in the ovariectomized rat. Neuroendocrinology 26:108–117[Medline]
5. Hancke J, Berk W, Baumgarten H, Hohn K, Kuttke W 1977 Modulatory effect of noradrenaline and serotonin on circiral LH release in adult ovariectomized rats. Acta Endocrinol (Copenh) 208(Suppl):22–23
6. Negro-Vilar A, Ojeda SA, McCann S 1979 Catecholamineric modulation of luteinizing hormone-releasing hormone release by median eminence terminals in vitro. Endocrinology 104:1749–1757[Medline]
7. Ojeda S, Nego-Vilar A, McCann S 1982 Evidence for involvement of -adrenergic receptors in norepinephrine induced prostaglandin E₂ and luteinizing hormone-releasing hormone release from the median eminence. Endocrinology 110:409–412[Abstract]
8. Douva S, Laplante E, Kordon C 1982 ₁-Adrenergic receptor involvement in the LH surge in variectomized estrogen-primed rats. Eur J Pharmacol 81:341–344[CrossRef][Medline]
9. Coen CW, Coombs MC 1983 Effects of manipulating catecholamines on the incidence of the preovulatory surge of luteinizing hormone and ovulation in the rat: evidence for a necessary involvement of hypothalamic adrenaline in the normal "midnight" surge. Neuroscience 10:187–206[CrossRef][Medline]
10. Palkovits M, Brownstein M, Saavedra J, Axelrod J 1974 Norepinephrine and dopamine content of hypothalamic nuclei of the rat. Brain Res 77:137–149[CrossRef][Medline]
11. Palkovits M 1981 Catecholamines in the hypothalamus: an anatomical review. Neuroendocrinology 33:123–128[CrossRef][Medline]
12. Silverman AJ, Krey LC 1978 The luteinizing hormone-releasing hormone (LHRH) neuronal network of the guinea pig brain. I. Intra- and extrahypothalamic projections. Brain Res 157:233–246