Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

The 5-HT_1A receptor is a member of the 5-HT₁ family of serotonin receptors, which are seven-transmembrane spanning receptors characterized as having high affinity for 5-HT and which are negatively coupled to adenylyl cyclase (Hoyer et al., 1994). In addition to inhibition of adenylyl cyclase activity (De Vivo and Maayani, 1986), activation of 5-HT_1A receptors also leads to decreased calcium conductance (Penington et al., 1991) and increased K⁺ conductance (Andrade et al., 1986; Colino and Halliwell, 1987) via pertussis toxin-sensitive G proteins (i.e., G_i/G_o). Additionally, the 5-HT_1A receptor may couple to the pertussis toxin-insensitive G protein G_z to increase the secretion of some neuroendocrine hormones (Serres et al., 2000).

The 5-HT_1A receptor system has been implicated in a variety of physiological functions and behaviors, including mood (anxiety and depression), temperature regulation, learning and memory, sexual behavior, and feeding (Zifa and Fillion, 1992). Moreover, 5-HT_1A receptors are important targets for various psychotherapeutic drugs, such as the selective serotonin reuptake inhibitors (SSRIs) (Blier and de Montigny, 1999), one of the most frequently prescribed antidepressant drugs. Although SSRIs inhibit 5-HT reuptake by blockade of transporter function, this pharmacological effect alone is unlikely to be responsible for their antidepressant effects, because inhibition of 5-HT reuptake occurs within minutes of acute administration of SSRIs, whereas maximal clinical therapeutic effects generally require weeks of continuous treatment. It is currently thought that a progressive decrease in the responsiveness of the 5-HT_1A receptor system, secondary to prolonged 5-HT reuptake inhibition, is necessary for SSRIs to produce clinical therapeutic effects (Blier and de Montigny, 1999).

Although we have learned a considerable amount about the cellular signal transduction pathways coupled to the 5-HT_1A receptor, we know very little about how the function of this important receptor system can be regulated. Recently, we reported that activation of phospholipid-coupled receptor systems (e.g., 5-HT_2C, 5-HT_2A, and purinergic P₂) can reduce the responsiveness of the 5-HT_1B receptor system (Berg et al., 1994b, 1996, 1998a), a system that shares a high degree of homology with the 5-HT_1A receptor system. The mechanism for the reduction in responsiveness of the 5-HT_1B receptor system was due to the action of a cyclooxygenase-dependent metabolite of the arachidonic acid (AA) signaling cascade. The goal of this study was to examine the effect of activation of phospholipid-coupled receptors on the responsiveness of the 5-HT_1A receptor system and to explore the role of the AA signaling cascade in the regulation of 5-HT_1A receptor function.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. Forskolin (FSK), n,n-dipropyl 5-carboxamidotryptamine (dp-5-CT), (±)-2,5-dimethoxy-4-iodoamphetamine hydrochloride (DOI), 5-carboxamidotryptamine (5-CT), and thapsigargin were purchased from RBI/Sigma (Natick, MA); [³H]8-hydroxy-dipropylaminotetralin (8-OH-DPAT), [³H]arachidonic acid, [¹⁴C]arachidonic acid, and [¹²⁵I]-cAMP tracer were from PerkinElmer Life Science Products (Boston, MA); anti-cAMP antibody was from ICN Biomedicals (Costa Mesa, CA); 1,2-bis-(o-aminophenoxy)-ethane-N,N,N',N'-tetraacetic acid tetra(acetoxymethyl) ester (BAPTA-AM) was from Calbiochem (La Jolla, CA); and fura-2 AM was from Molecular Probes (Eugene, OR). Rolipram was a generous gift from Berlex Laboratories (Cedar Knolls, NJ) and 4-(2'-methoxyphenyl)-1-[2'-(N-2"-pyridyl)-p-fluorobenzamido] ethylpiperazine (p-MPPF) was a generous gift from Dr. Hank Kung (University of Pennsylvania, Philadelphia, PA). All tissue culture reagents and Hanks' balanced salt solution (HBSS) were purchased from Life Technologies (Grand Island, NY). All other drugs and chemicals (reagent grade) were purchased from Sigma Chemical Co. (St. Louis, MO).

Transfection and Cell Culture. The human (h)5-HT_1A (G21) receptor cDNA and CHO-K1 cells stably expressing h5-HT_1A receptors were kindly provided by Dr. Alan Saltzman (Rhone-Poulenc Rorer Central Research, Collegeville, PA). To establish stable lines coexpressing h5-HT_1A and h5-HT_2C receptors, CHO-1C19 cells, which express the h5-HT_2C receptor at a density of ~250 fmol/mg (Berg et al., 1994b, 1996), were transfected with pZeoSV (Invitrogen, San Diego, CA) containing the coding region of the h5-HT_1A receptor using lipofectAMINE (15 µl/2 µg of DNA). Clones resistant to either G418 (CHO-1A; 500 µg/ml) or zeocin (CHO-2C/1A, 250 µg/ml) were screened for 5-HT_1A receptor expression using BMY-7378-sensitive [³H]8-OH-DPAT binding as well as for the ability of the 5-HT_1A agonist dp-5-CT to inhibit forskolin-stimulated cAMP accumulation (FScA). A cell line exhibiting low expression (~140 fmol/mg protein, CHO-1A_low) and a cell line with high receptor expression (~1 pmol/mg protein, CHO-1A_high) levels were chosen for this study. In addition, a cell line expressing both 5-HT_1A and 5-HT_2C receptors (CHO-2C/1A), which express 5-HT_1A receptors at a density of ~1 pmol/mg, were also used in this study. Cells were maintained in minimal essential medium- formulation supplemented with 5% fetal bovine serum and 50 µg/ml G418 (CHO-1A_low, CHO-1A_high) or 125 µg/ml zeocin + 300 µg/ml hygromycin (CHO-2C/1A). For all experiments, cells were seeded into 24-well, 15-cm or T175 tissue culture vessels at a density of 4 × 10⁴ cells/cm². Following a 24-h plating period, cells were washed with HBSS and placed into Dulbecco's modified Eagle medium/F-12 (1:1) with 5 µg/ml insulin, 5 µg/ml transferrin, 30 nM selenium, 20 nM progesterone, and 100 µM putrescine (serum-free media) and grown for an additional 24 h prior to experimentation.

Inhibition of FScA. Receptor-mediated inhibition of FScA was measured as described previously (Berg et al., 1994b, 1996). Cells, washed twice with HBSS containing calcium and magnesium supplemented with 10 mM HEPES (pH 7.4) (i.e., wash buffer), were preincubated in 500 µl of wash buffer/well for 15 min in a CO₂ incubator (5%, 37°C). For experiments using nominally extracellular calcium-free media, cells, washed twice with calcium-free HBSS containing magnesium supplemented with 10 mM HEPES (pH 7.4) (i.e., calcium-free buffer), were preincubated in 500 µl of calcium-free buffer/well for 15 min in a CO₂ incubator (5%, 37°C). Where indicated, drugs (e.g., enzyme inhibitors/activators or antagonists) were present during the preincubation period. 5-HT_1A receptor-mediated responses were determined by measuring agonist (dp-5-CT)-mediated inhibition of cAMP accumulated in response to 1 µM FSK (15 min, 37°C) in the presence of the phosphodiesterase inhibitor rolipram (0.1 mM). Cellular cAMP was extracted by the addition of 500 µl of ice-cold ethanol, measured by radioimmunoassay and normalized to protein content, which was measured according to the method of Lowry et al. (1951).

AA Release. AA release was measured as described previously (Berg et al., 1996, 1998b). Cells were labeled with 0.1 µCi/ml [³H]AA (180-240 Ci/mmol) or [¹⁴C]AA (57 mCi/mmol) for 4 h at 37°C. Following the labeling period, cells were incubated in 1.0 ml of HBSS wash buffer/0.1% bovine serum albumin (BSA) containing vehicle [distilled H₂O or 0.01% dimethyl sulfoxide (DMSO) as necessary] or the indicated drug concentrations. Aliquots (100 µl) of the media were taken after incubation for 10 min and added directly to scintillation vials and the ³H or ¹⁴C contents determined by liquid scintillation spectrometry in a Beckman LS7500 scintillation counter.

Intracellular Calcium Measurements. Increases in intracellular calcium levels ([Ca²⁺]_i) were determined essentially as described previously (Berg et al., 1994a, 1998b). Cells in suspension were loaded with fura-2 AM by incubating the cells in HBSS containing 0.1% BSA and 5 µM fura-2 AM at 37°C for 30 min in the dark followed by a hydrolysis period of 30 min at room temperature in the dark. Cells were washed once, resuspended in HBSS/BSA, and placed (2 × 10⁶ cells) in a stirred, temperature-controlled (37°C) cuvette in a fluorescence spectrometer (Photon Technology International, Monmouth Junction, NJ) equipped with automatic data collection/analysis software. After a 5-min equilibration period, data were collected using dual-wavelength excitation at 340 and 380 nm and an emission wavelength of 510 nm at a frequency of 1 Hz. Drugs were added to the cuvette after collection of baseline values for 60 s. [Ca²⁺]_i was calculated from the fluorescence ratios (F340/F380) after calibration with 15 µM digitonin (to obtain F_max values), followed by 10 mM EGTA, pH > 9 (to obtain F_min values), according to the equation described previously (Grynkiewicz et al., 1985). Basal [Ca²⁺]_i was calculated as the average value obtained during 30 s before drug addition. Maximal changes in [Ca²⁺]_i were calculated as the peak level of [Ca²⁺]_i reached after agonist administration following subtraction of the basal value.

5-HT_1A Receptor Binding. Binding assays were performed as previously described (Clarke and Maayani, 1990). After 24 h in serum-free media, cells in 15-cm plates (~320 µg of total protein) were treated with arachidonic acid (10 µM) as described above for cAMP studies, washed twice with ice-cold HBSS, scraped, and pelleted. Pellets were flash frozen and stored in liquid nitrogen until assay. Membranes were prepared (pellets pooled from 5 × 15-cm plates) by homogenization on ice, with a Polytron at setting 7, in 40 volumes of ice-cold HEPES buffer (50 mM HEPES, 2.5 mM MgCl₂, 2.0 mM EGTA, pH 7.4 at 23° C). The homogenate was centrifuged (39,000g, 4° C, 10 min) and the pellet washed two times by resuspension in 40 volumes of the same buffer and centrifugation. The homogenate was incubated in a shaking water bath (37°C, 10 min) in HEPES assay buffer (HEPES buffer; 0.1% ascorbic acid; 10 mM Na₄P₂O₇; 1 mM phenylmethylsulfonyl fluoride; 5 µg/ml each leupeptin, soybean trypsin inhibitor, and benzamidine; pH 7.4 at 23° C), centrifuged, and the pellet washed one additional time by resuspension in 40 volumes of ice-cold HEPES buffer and centrifugation. Following protein determination according to the method of Bradford (1976), aliquots (400 µl containing ~30 µg of protein) of membrane suspension were incubated (60 min, 23°C, total volume of 0.5 ml) with various concentrations (0.05-5 nM) of [³H]8-OH-DPAT in duplicate for saturation studies. For experiments measuring receptor-G protein-coupling efficiency, aliquots (400 µl containing ~30 µg of protein) of membrane suspension were incubated (60 min, 23°C, total volume of 0.5 ml) with various concentrations (0.01-100 µM) of the guanine nucleotide guanosine 5'-(,-imino) triphosphate [Gpp(NH)p], with a final concentration of 1 nM [³H]8-OH-DPAT. Samples were filtered through Whatman GF/C filters with a Brandell cell harvester. The filters were washed twice with 3 ml of ice-cold HEPES buffer and counted with a Beckman LS7500 liquid scintillation counter (efficiency of 48%). Nonspecific binding was determined in the presence of 1 µM BMY-7378 (100 × K_d for 5-HT_1A sites).

GTP[³⁵S] Binding. After 24 h in serum-free media, cells in 15-cm plates (~320 µg of total protein) were treated with arachidonic acid (10 µM) as described above for cAMP studies, washed twice with ice-cold HBSS, scraped, and pelleted. Pellets were flash frozen and stored in liquid nitrogen. Membranes were prepared by repeated trituration of thawed cell pellets through a 1-ml pipette in ice-cold wash buffer (20 mM HEPES, 3 mM MgCl₂, 0.2 mM EGTA, and 100 mM NaCl, pH 7.4 at 23°C). The homogenate was centrifuged (39,000g, 4° C, 10 min) and the pellet washed two times by resuspension in 40 volumes of the same buffer and centrifugation. Membranes were resuspended in assay buffer [wash buffer plus GDP (10 µM), okadaic acid (100 nM), and cypermethrin (10 nM)] at a protein concentration of 50 µg/ml. Aliquots (100 µl) of the membrane suspension were preincubated with dp-5-CT (100 nM final) or vehicle (assay buffer) in Millipore 96-well Multiscreen filtration plates for 30 min at 37°C in triplicate. The assay was initiated by the addition of GTP[³⁵S] (final concentration of 0.2 nM) at various time points such that incubation time with GTP[³⁵S] varied from 5 to 60 min. The assay was terminated by rapid filtration and subsequent washing of filters with ice-cold wash buffer 8 × 200 µl. Filters from the plates were removed, placed in scintillation vials, and counted with a Beckman LS7500 liquid scintillation counter. Nonspecific binding was determined in the presence of Gpp(NH)p (1 mM). Protein determination was according to the method of Bradford (1976).

Data Analysis. For cAMP accumulation experiments, concentration-response data were fit with nonlinear regression to eq. 1, to provide estimates of R_o, R_i, EC₅₀, and n:

(1)

where R is the measured response (pmol of cAMP/mg of protein) at a given agonist concentration ([a]), R_o is the response in the absence of agonist, R_i is the response after maximal inhibition by agonist, EC₅₀ is the concentration of agonist that produces a half-maximal response, and n is the slope factor. R_max (the maximal inhibition produced by the agonist) was calculated as R_o  R_i. Data were normalized for each experiment by defining the response to 1 µM FSK as 100%. Since in these experiments slope factors were not different from unity, data provided in the text were obtained by setting the slope factor to 1.

(2)

(3)

Statistical Analysis. Where indicated, statistical significance was determined using one-way ANOVA and the Newman-Keuls post hoc test. All other analyses were done using the Student's t test (paired). Asterisks denote statistically significant p values <0.05 (*), <0.01 (**), and <0.001 (***).

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Characterization of Human 5-HT_1A Receptors Stably Expressed in CHO Cells. CHO-K1 cells express endogenously a 5-HT_1B receptor (Berg et al., 1994b; Dickenson and Hill, 1995; Giles et al., 1996), therefore, we verified that the 5-HT_1A agonist dp-5-CT, which is reported to have 1000-fold selectivity for 5-HT_1A receptors over that of 5-HT_1B receptors (Zifa and Fillion, 1992), indeed did not alter FScA (1 µM) in parent CHO cells or in CHO cells expressing the h5-HT_2C receptor (CHO-1C19). No change in FScA was detected in response to dp-5-CT at concentrations up to 10 µM, the highest concentration tested (data not shown). In CHO-1A_low cells, dp-5-CT inhibited FScA to a maximum of 80% with an EC₅₀ value of 15 nM (Fig. 1). In the presence of the selective 5-HT_1A receptor antagonist p-MPPF (30 nM), the concentration curve for dp-5-CT was shifted to the right (25-fold) in a parallel and surmountable manner (EC₅₀ = 388 nM), indicating a single receptor population. The apparent K_B for p-MPPF was calculated to be 1.2 nM, which is consistent with the reported affinity for this antagonist at the 5-HT_1A receptor (Kung et al., 1996). In parent CHO-K1 cells, activation of the 5-HT_1B receptor with 5-CT (300 nM) produced maximal inhibition of FScA of 80%, which was unaffected in the presence of p-MPPF at concentrations up to 30 µM. Therefore, in cells coexpressing 5-HT_1A and 5-HT_1B receptors, 5-HT_1A receptors can be selectively activated with d5-CT, whereas 5-HT_1B receptors can be selectively activated with 5-CT in the presence of p-MPPF.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   AA reduces 5-HT1A receptor-mediated inhibition of FScA through production of an indomethacin-sensitive metabolite. A, cAMP accumulation was measured in CHO-1Alow cells incubated with FSK (1 µM) and dp-5-CT in the presence of vehicle (ethanol, 0.1%) or AA (10 µM) for 15 min at 37°C. Data are expressed as a percentage of forskolin stimulation and are the mean ± S.E.M. from five experiments. Individual concentration-response curve data were fit with nonlinear regression to eq. 1 to determine the mean EC50 and Emax values, which are provided in the text. FSK-stimulated cAMP accumulation was 45 ± 20 and 99 ± 21 pmol/mg of protein for vehicle and AA, respectively. B, effect of AA is blocked in the presence of the cyclooxygenase inhibitor indomethacin. CHO-1Alow cells were treated with vehicle (DMSO, 0.02%) or indomethacin (2 µM) for 15 min at 37°C, followed by incubation with FSK (1 µM) and an approximate EC50 concentration of dp-5-CT (20 nM) in the presence of either vehicle (EtOH, 0.1%) or AA (10 µM) for 15 min (37°C). Data shown are expressed as a percentage of forskolin stimulation and represent the mean ± S.E.M. from six experiments. FSK-stimulated cAMP accumulation was 82 ± 27 and 161 ± 58 pmol/mg of protein for vehicle and AA, respectively; and 75 ± 33 and 95 ± 40 pmol/mg of protein for vehicle and AA, respectively, in the presence of indomethacin. **p < 0.005.

Effects of Direct Activation of Signaling Components within the PLA₂ and PLC Effector Pathways on 5-HT_1A Receptor System Responsiveness. We have reported previously that the responsiveness of the 5-HT_1B receptor system in CHO cells is sensitive to consequences of PLA₂, but not PLC, activation (Berg et al., 1994b, 1996). To determine whether stimulation of PLC and PLA₂ could alter the responsiveness of the 5-HT_1A receptor system, we directly activated various signaling components known to be produced by activation of these effectors.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   A, phorbol ester-mediated activation of PKC reduces the responsiveness of the 5-HT1A receptor system. cAMP accumulation was measured in CHO-1Alow cells treated with vehicle (DMSO, 0.01%) or the phorbol ester PdBu (1 µM) for 15 min at 37°C, followed by incubation with FSK (1 µM) and the indicated concentrations of dp-5-CT for 15 min (37°C). Data shown are expressed as a percentage of forskolin stimulation and represent the mean ± S.E.M. from four experiments. Individual concentration-response curve data were fit with nonlinear regression to eq. 1 to determine the mean Emax and EC50 values, which are provided in the text. FSK-stimulated cAMP accumulation was 53 ± 11 and 47 ± 11 pmol/mg of protein for vehicle and PdBu, respectively. B, reduction in responsiveness of the 5-HT1A receptor system by PdBu is blocked in the presence of staurosporine (SSP). dp-5-CT (300 nM) inhibition of FScA was measured in cells pretreated with staurosporine (1 µM) or vehicle for 5 min prior to addition of PdBu for 15 min at 37°C. Data represent the mean ± S.E.M. of three experiments. FScA was not changed by either treatment with staurosporine or PdBu and was 70 ± 9 pmol/mg of protein. *p < 0.05.

View larger version (21K): [in this window] [in a new window]   Fig. 3.   Increases in [Ca2+]i enhance agonist efficacy at 5-HT1A receptors. The effect of the calcium ionophore, A23187, or thapsigargin in the presence or absence of extracellular calcium (i.e., nominally calcium free) on the inhibition of FScA by dp-5-CT (A) or FScA alone (B). cAMP accumulation was measured in CHO-1Alow cells treated with vehicle (DMSO, 0.01%), the calcium ionophore, A23187 (1 µM), or thapsigargin (200 nM) for 5 min at 37°C, followed by incubation (15 min, 37°C) with FSK (1 µM) and an approximate EC50 concentration of dp-5-CT (10 µM) (A) or FSK alone (B). Data are expressed as the percentage of inhibition of forskolin stimulation by dp-5-CT (A) or the percentage of forskolin (vehicle control) stimulation (B). Each bar represents the mean ± S.E.M. from four experiments. FSK-stimulated cAMP accumulation was 164 ± 58 and 209 ± 62 pmol/mg protein in calcium-containing and nominally calcium-free conditions, respectively. *p < 0.05 from corresponding control.

Effects of Receptor-Mediated Activation of PLA₂ and PLC on the Responsiveness of the 5-HT_1A Receptor System. When expressed in CHO cells, 5-HT_2C receptors couple independently to PLA₂, leading to activation of the AA signaling cascade, and to PLC, leading to activation of PKC and increases in [Ca²⁺]_i (Berg et al., 1996, 1998b). We have recently reported that activation of 5-HT_2C receptors with DOI completely blocks inhibition of FScA by the endogenous 5-HT_1B receptor, an effect that is mediated by an indomethacin-sensitive AA metabolite produced by PLA₂ activation (Berg et al., 1996). As shown in Fig. 4, in CHO-2C/1A cells, activation of 5-HT_2C receptors with DOI (1 µM) also reduced the inhibition of FScA by dp-5-CT (10 nM) by 50%. This effect of DOI was blocked by the PLA₂ inhibitor mepacrine (100 µM). The DOI-mediated reduction in dp-5-CT-mediated inhibition of FScA was 40 ± 3% and 12 ± 8% in the presence of vehicle (DMSO) or mepacrine, respectively (mean ± S.D., n = 2). The inhibitory effect of DOI on the 5-HT_1A response was blocked by indomethacin (2 µM, Fig. 4), suggesting that, similar to regulation of the 5-HT_1B receptor system, 5-HT_2C receptor-mediated reduction in agonist efficacy at the 5-HT_1A receptor is mediated by a cyclooxygenase-dependent metabolite of AA.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   5-HT2C receptor activation reduces the responsiveness of the 5-HT1A receptor system in an indomethacin-sensitive manner. CHO-2C/1A cells were treated with vehicle (DMSO, 0.02%) or indomethacin (2 µM) for 15 min at 37°C, followed by incubation with FSK (1 µM) and dp-5-CT (10 nM) in the presence or absence of the 5-HT2C receptor agonist DOI (1 µM). Data are expressed as the percentage of inhibition of FScA by dp-5-CT and represent the mean ± S.E.M. of three to four experiments. FSK-stimulated cAMP accumulation was 206 ± 63, 113 ± 70, and 160 ± 52 pmol/mg of protein for vehicle, indomethacin, and DOI, respectively. ***p < 0.001.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   A, P2-purinergic receptor activation does not alter the 5-HT1A agonist concentration-response curve. cAMP accumulation was measured in CHO-1Alow cells incubated with FSK (1 µM) and the indicated concentrations of dp-5-CT with or without ATP (1 mM) for 15 min at 37°C. B, P2-purinergic receptor activation reduces the responsiveness to the 5-HT1B agonist. 5-HT1B receptor system responsiveness was determined by measuring the inhibition of FScA by the agonist 5-CT in the presence of the 5-HT1A receptor antagonist p-MPPF (10 µM). Data are expressed as a percentage of forskolin stimulation and represent the mean ± S.E.M. from five (5-HT1A) or three (5-HT1B) experiments. Individual concentration-response curve data were fit with nonlinear regression to eq. 1 to determine the mean Emax and EC50, which are provided in the text. The presence of p-MPPF did not alter FSK-stimulated cAMP accumulation, which was 77 ± 26 and 82 ± 33 pmol/mg of protein for vehicle and ATP, respectively.

View larger version (14K): [in this window] [in a new window]   Fig. 6.   Comparison between P2 receptor- and 5-HT2C receptor-mediated AA release and increases in [Ca2+]i. A, CHO-2C/1A cells were labeled with [3H]arachidonic acid for 4 h, washed, and AA release was measured in response to incubation with DOI (1 µM) or ATP (1 mM) for 10 min at 37°C. Data shown are mean ± S.E.M. of three experiments. **p < 0.01. B, CHO-2C/1A cells were loaded with fura-2 AM and the peak increase in [Ca2+]i in response to stimulation with DOI (1 µM) or ATP (1 mM) was measured using spectrofluorimetry. Data are mean peak increases in [Ca2+]i ± S.E.M. of three experiments. Resting calcium levels were 339 ± 32 nM. **p < 0.01.

View larger version (20K): [in this window] [in a new window]   Fig. 7.   Effect of PLA2 blockade (extracellular calcium-free conditions) on P2 receptor-mediated AA release, increases in [Ca2+]i, and regulation of 5-HT1A receptor-mediated inhibition of FScA. A, ATP (1 mM)-mediated AA release was measured in CHO-1Alow cells in the presence (+[Ca2+]e) or absence ([Ca2+]e free) of extracellular calcium in the medium. Data shown are mean ± S.E.M. from four experiments. *p < 0.005. B, increases in [Ca2+]i were measured in CHO-1Alow cells loaded with fura-2 AM using spectrofluorimetry. Data are representative tracings, which show the typical changes in [Ca2+]i following ATP (1 mM) addition (arrows) in either normal (calcium containing) or calcium-free media. C, dp-5-CT (10 nM)-mediated inhibition of FScA was measured in CHO-1Alow cells in normal (+[Ca2+]e) or calcium free ([Ca2+]e-free) media in the presence of vehicle (DMSO, 0.01%), thapsigargin (200 µM), or ATP (1 mM). Data shown are expressed as the percentage of inhibition of FScA by dp-5-CT (10 nM) and are the mean ± S.E.M. of four experiments. This experiment was done in conjunction with that shown in Fig. 3 and the vehicle and thapsigargin data are reproduced for comparative purposes. FScA was 164 ± 58 pmol/mg of protein normal calcium-containing conditions and 209 ± 62 pmol/mg of protein under nominally extracellular calcium free conditions. *p < 0.05.

View larger version (15K): [in this window] [in a new window]   Fig. 8.   P2 receptor activation inhibits 5-HT1A receptor system responsiveness when effects of increased [Ca2+]i release are blocked. CHO-1Alow cells were incubated for 30 min at 37°C in the dark in the presence of vehicle (DMSO, 0.003%) or BAPTA-AM (30 µM), washed, and incubated for an additional 30 min in the dark at room temperature to allow for BAPTA-AM hydrolysis. A, dp-5-CT (10 nM)-mediated inhibition of FScA was measured in the presence or absence of ATP (1 mM). Data are expressed as the percentage of inhibition of FScA and are mean ± S.E.M. of six experiments. FSK-stimulated cAMP accumulation was: 66 ± 3, 55 ± 3, 49 ± 3, and 42 ± 2 pmol/mg of protein for vehicle, BAPTA-AM, ATP, and BAPTA-AM + ATP, respectively. **p < 0.005. B, changes in [Ca2+]i were measured in CHO-1Alow cells that had been loaded with both 5 µM fura-2 AM and 30 µM BAPTA-AM for 30 min at 37°C, washed, and incubated for an additional 30 min in the dark at room temperature to allow for hydrolysis of both methyl esters. Data shown represent typical changes in [Ca2+]i following ATP addition in the presence of vehicle (DMSO, 0.003%) or BAPTA-AM (30 µM). Graph shown is a representative tracing and arrows indicate the time of ATP (1 mM) addition.

View larger version (14K): [in this window] [in a new window]   Fig. 9.   P2 receptor activation has no effect on 5-HT1A receptor system responsiveness when PKC activity is blocked. dp-5-CT-mediated inhibition of FScA was measured in CHO-1Alow cells preincubated for 5 min (37°C) in the presence of vehicle (DMSO, 0.01%) or staurosporine (SSP,1 µM) followed by incubation with FSK (1 µM) and dp-5-CT (10 nM) in the presence and absence of ATP (1 mM) for 15 min at 37°C. Data are expressed as the percentage of inhibition of FScA by dp-5-CT and are mean ± S.E.M. of six experiments. FScA was 66 ± 10 and 60 ± 11 pmol/mg of protein for vehicle and ATP, respectively, in the presence of DMSO (0.01%) pretreatment; and 67 ± 8 and 59 ± 11 pmol/mg of protein for vehicle and ATP, respectively, in the presence of staurosporine.

View larger version (13K): [in this window] [in a new window]   Fig. 10.   5-HT2C receptor-mediated inhibition of 5-HT1A receptor system responsiveness is abolished by P2 receptor activation. After preincubation for 5 min (37°C) with either vehicle or ATP (1 mM), dp-5-CT-mediated inhibition of FScA was measured in CHO-2C/1A cells incubated with FSK (1 µM) with and without dp-5-CT (10 nM) for 15 min (37°C) with or without the 5-HT2C receptor agonist DOI (1 µM). Data are expressed as the percentage of inhibition of FScA by dp-5-CT and are the mean ± S.E.M. of six experiments. FSK-stimulated cAMP accumulation was 43 ± 5 and 45 ± 3 pmol/mg of protein for vehicle under control and ATP pretreatment conditions, respectively, and 32 ± 9 and 43 ± 7 pmol/mg of protein for DOI under control and ATP pretreatment conditions, respectively. *p < 0.05.

Mechanism for AA-Mediated Regulation of 5-HT_1A and 5-HT_1B Receptor Systems. Both 5-HT_1A (above) and 5-HT_1B (Berg et al., 1996) receptor system responsiveness is reduced by a cyclooxygenase-dependent AA metabolite. To determine whether the mechanism by which the AA metabolite reduces the function of these receptor systems is the result of changes at the level of the receptor, we measured the binding characteristics of [³H]8-OH-DPAT to the 5-HT_1A receptor after treatment of CHO-1A_high cells with AA. Binding studies for the endogenous 5-HT_1B receptor could not be performed due to low receptor expression levels in CHO cells (Giles et al., 1996). In cells expressing ~1 pmol/mg of protein [CHO-1A_high (not shown) and CHO-2C/1A (Fig. 4)], inhibition of FScA by dp-5-CT was reduced by the AA metabolite. As shown in Table 1, treatment with AA did not alter either the B_max or the K_d of [³H]8-OH-DPAT binding to the 5-HT_1A receptor. Nor did AA alter the capacity of Gpp(NH)p (E_max or IC₅₀) to reduce high-affinity [³H]8-OH-DPAT binding (Table 1), suggesting no effect of AA on receptor-G protein-coupling efficiency. Finally, AA treatment did not alter the capacity of the 5-HT_1A receptor to activate G protein(s) as measured by dp-5-CT stimulation of GTP[³⁵S] binding. Taken together, these data suggest that the effect of the AA metabolite is exerted at a point distal to the receptor.

View this table: [in this window] [in a new window]   TABLE 1 Effect of arachidonic acid on 5-HT1A receptor binding parameters and receptor-G protein-coupling and activation Cells were treated with AA, 10 µM or vehicle (EtOH, 0.1%). The effect of AA treatment on agonist affinity (Kd) and receptor density (Bmax) was measured using [3H]8-OH-DPAT binding in membrane homogenates. AA effects on receptor-G protein-coupling efficiency were studied by measuring the capacity of Gpp(NH)p (a nonhydrolyzable analog of GTP) to reduce high-affinity [3H]8-OH-DPAT binding (1 nM). To assess the capacity of the activated receptor to activate the G protein after AA treatment, the time course (5-60 min) of binding of GTP[35S] to G protein was measured in response to activation of 5-HT1A receptors with dp-5-CT (100 nM) and the rate constant (kobs) and maximal binding (Bmax) were calculated. AA did not alter any of the measured parameters (p > 0.05).

View larger version (16K): [in this window] [in a new window]   Fig. 11.   AA-metabolite-mediated inhibition of 5-HT1A receptor system responsiveness does not occur when adenylyl cyclase is stimulated with Gs. dp-5-CT-mediated inhibition of FScA was measured in CHO-2C/1A cells treated with PGE2 (1 µM), cholera toxin (CTx, 10 µg/ml, ), or FSK (1 µM) with and without dp-5-CT (10 nM) in the presence and absence of the 5-HT2C receptor agonist DOI (1 µM) for 15 min (37°C). Data are expressed as the percentage of inhibition by dp-5-CT and are mean ± S.E.M. of eight experiments. cAMP accumulation (pmol/mg of protein) was 102 ± 7, 136 ± 7, and 69 ± 7 for FSK, PGE2, and CTx, respectively, and in the presence of DOI was 83 ± 5, 69 ± 7, and 60 ± 8 for FSK, PGE2, and CTx, respectively. *p < 0.05.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Previously, we reported that the responsiveness of the 5-HT_1B receptor system is reduced by activation of phospholipid-coupled receptors and that the mediator of this effect is a cyclooxygenase-dependent metabolite of AA. Given the high sequence homology between the 5-HT_1A and 5-HT_1B receptor systems, we expected to find similar regulation of responsiveness of the 5-HT_1A receptor system. While we did find similarities between the regulation of 5-HT_1A and 5-HT_1B function, important differences were also observed.

Similar to the 5-HT_1B receptor system, the responsiveness of the 5-HT_1A receptor system was reduced by a cyclooxygenase-dependent metabolite of AA. Coactivation of 5-HT_2C receptors reduced the ability of the 5-HT_1A agonist dp-5-CT to inhibit FScA. This effect was blocked by the PLA₂ inhibitor mepacrine, suggesting mediation by the PLA₂-AA signaling cascade. Similarly, the efficacy of dp-5-CT was reduced when the 5-HT_1A receptor was activated in the presence of AA itself. Furthermore, both the AA effect and the effect of 5-HT_2C receptor activation were blocked by the cyclooxygenase inhibitor indomethacin, suggesting that, like for the 5-HT_1B receptor system, a metabolite of AA (e.g., a prostaglandin or thromboxane) is responsible for the reduction in the responsiveness of the 5-HT_1A receptor system. Finally, the site of action of the AA metabolite to reduce 5-HT_1A and 5-HT_1B receptor system function appears to be the enzyme adenylyl cyclase. AA had no effect on the binding characteristics or G protein-coupling/activation efficiency of the 5-HT_1A receptor, suggesting a postreceptor site of action. [Complementary studies could not be performed with the endogenous 5-HT_1B receptor due to low receptor density (Giles et al., 1996).] This postreceptor target appears to be adenylyl cyclase since the effect of the AA metabolite to reduce the responsiveness of both the 5-HT_1A and 5-HT_1B receptor systems was dependent upon the activation state of adenylyl cyclase: the AA metabolite was effective only when adenylyl cyclase was stimulated with forskolin, but not when stimulated with G_s or cholera toxin.

Albert and colleagues have reported that the activation state of adenylyl cyclase (forskolin versus G_s) profoundly influences which G_i protein couples the D2S receptor to adenylyl cyclase (Ghahremani et al., 1999). When adenylyl cyclase is activated with forskolin, D2S-mediated inhibition of adenylyl cyclase activity is mediated by G_i2. However, when G_s (liberated via PGE₁ stimulation) is used, inhibition of adenylyl cyclase is mediated by G_i3. Similarly, Taussig et al. (1993) showed that G_i1 could inhibit adenylyl cyclase type I only when forskolin or calmodulin were used as stimulators, not when stimulation was via G_s (Taussig et al., 1993). These data suggest that the conformation of adenylyl cyclase may differ depending upon whether forskolin, calmodulin, or G_s is bound to the enzyme and that different adenylyl cyclase conformations may render the enzyme differentially sensitive to other regulators of its activity. Perhaps the AA metabolite reduces 5-HT_1A/1B receptor-mediated inhibition of FScA by altering the conformation of adenylyl cyclase such that inhibition by G_i is reduced. Furthermore, it is possible that the G_s-dependent conformation of adenylyl cyclase is resistant to regulation by the AA metabolite.

It is quite likely that the capacity of the AA metabolite to regulate the responsiveness of the 5-HT_1A and 5-HT_1B receptor systems will be cell type-dependent. To date, nine mammalian isoforms of adenylyl cyclase (types I-IX) have been cloned, each with distinct and dynamic regulatory properties (for reviews, see Sunahara et al., 1996; Hanoune et al., 1997; Defer et al., 2000). Moreover, these adenylyl cyclase isoforms are heterogeneously expressed throughout the various tissues of the body and it is likely that this differential distribution extends to individual cell types as well. For example, mRNA for types I, VI, and IX can be found in AtT-20 cells (Paterson et al., 1995; Premont et al., 1996), while CHO cells appear to express only types VI and VII (Varga et al., 1998). If the AA metabolite reduces 5-HT_1A/1B signaling by impairing the capacity of adenylyl cyclase to be inhibited by G_i, then it would seem likely that, in our CHO cell system, the target for the AA metabolite would be adenylyl cyclase type VI. Furthermore, we would speculate that regulation of 5-HT_1A/1B receptor system responsiveness would occur in only those cells that express G_i inhibitable forms of adenylyl cyclase (types I, V, and VI; Sunahara et al., 1996). Such cellular specificity may allow for the development of drugs designed to selectively alter 5-HT_1A receptor function in particular cell types.

Although both the 5-HT_1A and 5-HT_1B receptor systems are regulated similarly by the PLA₂-AA signaling cascade, the responsiveness of the 5-HT_1A, but not the 5-HT_1B (Berg et al., 1994b), receptor system is regulated by the consequences of PLC activation. Interestingly, the 5-HT_1A system was regulated differentially by activation of PKC and by increases in [Ca²⁺]_i, two consequences of PLC activation. Activation of PKC with the phorbol ester PdBu reduced the efficacy of the 5-HT_1A agonist to inhibit FScA. This effect was blocked by the PKC inhibitor staurosporine. Our findings are consistent with other reports of PKC-mediated reductions in 5-HT_1A receptor system responsiveness (Raymond, 1991; Harrington et al., 1994; Lembo and Albert, 1995; Hensler et al., 1996). Although the 5-HT_1A receptor has four consensus sites for PKC phosphorylation and the receptor has been shown to be phosphorylated by PKC (Raymond, 1991), it is not yet clear whether receptor phosphorylation is responsible for the reduced efficacy of 5-HT_1A agonists to inhibit adenylyl cyclase activity. In contrast to the effects of PKC, increases in [Ca²⁺]_i, with either a calcium ionophore or with thapsigargin, enhanced the responsiveness of the 5-HT_1A receptor system. Thus, two consequences of PLC activation alter the responsiveness of the 5-HT_1A receptor system in opposing directions.

Given the opposing effects of calcium versus PKC and AA on the 5-HT_1A receptor system, it is difficult to predict the effect on 5-HT_1A responsiveness of activation by receptors that couple to both PLC and PLA₂. The effect of activation of a phospholipid-coupled receptor appears to depend upon the relative capacity of the receptor to activate the PLC and PLA₂ signaling systems. 5-HT_2C receptor activation reduced the ability of the 5-HT_1A agonist to inhibit FScA and this effect was completely sensitive to inhibition by the PLA₂ inhibitor mepacrine and the cyclooxygenase inhibitor indomethacin, suggesting the reduction in responsiveness is due to activation of the PLA₂-AA signaling cascade. However, activation of the P₂ purinergic receptor did not appear to alter the efficacy of the 5-HT_1A agonist. This lack of apparent effect of P₂ receptor activation on the 5-HT_1A receptor system was due to a balance between the opposing actions of positive ([Ca²⁺]_i) and negative (AA metabolite) regulators. However, even though the parameters of the 5-HT_1A agonist concentration-response curve were not changed in response to activation of the P₂ receptor, regulation of the 5-HT_1A receptor response by the AA metabolite was blocked, suggesting that P₂ receptor stimulation indeed altered the 5-HT_1A receptor system.

In conclusion, the responsiveness of the 5-HT_1A receptor system, like that of the 5-HT_1B system, is reduced by a cyclooxygenase-dependent AA metabolite, which can be derived from receptor-mediated activation of PLA₂. This AA metabolite appears to reduce the capacity of G_i to inhibit adenylyl cyclase activity when it is stimulated by forskolin, but not by G_s. We hypothesize that the AA metabolite may interact differentially with activated conformations of adenylyl cyclase (i.e., forskolin- and G_s-stimulated). We are currently attempting to determine the identity of this cyclooxygenase-dependent AA metabolite. Unlike the 5-HT_1B receptor system, the responsiveness of the 5-HT_1A system was also regulated by consequences of PLC activation. As has been reported previously (Raymond, 1991; Harrington et al., 1994; Lembo and Albert, 1995; Hensler et al., 1996), 5-HT_1A agonist efficacy was reduced by activation of PKC, however, 5-HT_1A responsiveness was enhanced by increases in [Ca²⁺]_i. This dual regulation of the 5-HT_1A receptor system provides for some interesting possible effects in response to activation of receptors that couple to PLC and PLA₂. The net effect on 5-HT_1A responsiveness appears to depend upon the relative capacity of the phospholipid-coupled receptor to produce these positive and negative regulators. Consequently, the effect of receptor-mediated activation of phospholipid signaling cascades on 5-HT_1A function will likely depend upon both the phenotype of the cell and the nature of the coupling between receptors and the phospholipid effectors.

The 5-HT_1A receptor system plays important roles in a variety of physiological functions and behaviors. Accordingly, knowledge of the cellular components that regulate the responsiveness of the 5-HT_1A receptor system is important for understanding the regulation of these physiological systems. Furthermore, reduction in responsiveness of the 5-HT_1A receptor system has been implicated in the therapeutic mechanism of action of the SSRI antidepressant drugs. Perhaps an understanding of the cellular mechanisms that regulate the responsiveness of the 5-HT_1A receptor system may provide new targets for development of drugs for the treatment of depression.