Abstract
We had reported that activation of presynaptic histamine H3-receptors inhibits norepinephrine exocytosis from depolarized cardiac sympathetic nerve endings, an action associated with a marked decrease in intraneuronal Ca2+ that we ascribed to a decreased Ca2+ influx. An H3-receptor-mediated inhibition of cAMP-dependent phosphorylation of Ca2+ channels could cause a sequential attenuation of Ca2+ influx, intraneuronal Ca2+ and norepinephrine exocytosis. We tested this hypothesis in sympathetic nerve endings (cardiac synaptosomes) expressing native H3-receptors and in human neuroblastoma SH-SY5Y cells transfected with H3-receptors. Norepinephrine exocytosis was elicited by K+ or by stimulation of adenylyl cyclase with forskolin. H3-receptor activation markedly attenuated the K+- and forskolin-induced norepinephrine exocytosis; pretreatment with pertussis toxin prevented this effect. Similar to forskolin, 8-bromo-cAMP elicited norepinephrine exocytosis but, unlike forskolin, it was unaffected by H3-receptor activation, demonstrating that inhibition of adenylyl cyclase is a pivotal step in the H3-receptor transductional cascade. Indeed, we found that H3-receptor activation attenuated norepinephrine exocytosis concomitantly with a decrease in intracellular cAMP and PKA activity in SH-SY5Y-H3 cells. Moreover, pharmacological PKA inhibition acted synergistically with H3-receptor activation to reduce K+-induced peak intracellular Ca2+ in SH-SY5Y-H3 cells and norepinephrine exocytosis in cardiac synaptosomes. Furthermore, H3-receptor activation synergized with N- and L-type Ca2+ channel blockers to reduce norepinephrine exocytosis in cardiac synaptosomes. Our findings suggest that the H3-receptor-mediated inhibition of norepinephrine exocytosis from cardiac sympathetic nerves results sequentially from H3-receptor-Gi/Go coupling, inhibition of adenylyl cyclase activity, and decreased cAMP formation, leading to diminished PKA activity, and thus, decreased Ca2+ influx through voltage-operated Ca2+ channels.
Histamine H3-receptors (H3Rs) were first recognized as inhibitory autoreceptors on histamine-containing nerve terminals (Arrang et al., 1983) and have since been shown to regulate the release of several neurotransmitters in the central and peripheral nervous systems (Hill et al., 1997; Levi and Smith, 2000). We previously reported that imetit, a selective H3R agonist (Garbarg et al., 1992), attenuates norepinephrine (NE) exocytosis evoked by depolarization of cardiac sympathetic nerve endings (Imamura et al., 1995), an action associated with a marked decrease in intraneuronal Ca2+ ([Ca2+]i) (Silver et al., 2002). Because ω-conotoxin (ω-CTX) and imetit each decreased [Ca2+]i and NE exocytosis (Silver et al., 2002), and since ω-CTX decreases [Ca2+]i by inhibiting Ca2+ influx through N-type Ca2+ channels (Sher et al., 1991), we speculated that, similar to ω-CTX, imetit-induced H3R activation might decrease [Ca2+]i by inhibiting Ca2+ influx through voltage-dependent Ca2+ channels in sympathetic nerve terminals (Silver et al., 2002). In fact, an H3R-mediated inhibition of N-type Ca2+ channel current has been claimed to occur in histaminergic neurons from the rat hypothalamus (Takeshita et al., 1998).
Cardiac H3Rs are possibly coupled to Gi/Go proteins, since we found that pertussis toxin, which inactivates Gi/Go (Bokoch et al., 1983; Codina et al., 1983), attenuates the H3R-mediated inhibition of adrenergic responses in the heart (Endou et al., 1994). Furthermore, H3R activation inhibits forskolin-stimulated cAMP formation in SKNMC neuroblastoma cells stably transfected with the human H3R (Lovenberg et al., 1999). Because cAMP-dependent phosphorylation of N-type Ca2+ channels increases their activity (Ahlijanian et al., 1991; Hell et al., 1995; Catterall, 2000), a decreased phosphorylation due to inhibition of the cAMP/PKA pathway could conceivably be involved in the H3R-mediated attenuation of N-type Ca2+ channel activity and NE exocytosis.
We have tested this hypothesis in human neuroblastoma cells transfected with human H3R (SH-SY5Y-H3) (Silver et al., 2002) and in sympathetic nerve endings (cardiac synaptosomes) expressing native H3R (Seyedi et al., 1997). We report that the H3R-mediated inhibition of NE exocytosis from cardiac sympathetic nerves results sequentially from H3R-Gi/Go coupling, inhibition of adenylyl cyclase activity, and cAMP formation, leading to diminished PKA activity, decreased Ca2+ influx through voltage-operated Ca2+ channels (VOCC), and thus, attenuation of NE exocytosis.
Materials and Methods
Preparation of Cardiac Synaptosomes. Male Hartley guinea pigs (Charles River Laboratories, Inc., Wilmington, MA) weighing 250 to 300 g were anesthetized with CO2 vapor and exsanguinated. The ribcage was rapidly opened and the heart dissected away. A cannula was inserted in the aorta and the heart was perfused for 5 min at constant pressure (40 cm of H2O) in a Langendorff apparatus (Seyedi et al., 1997) with Ringer's solution (containing 154 mM NaCl, 5.61 mM KCl, 2.16 mM CaCl2, 5.95 mM , and 5.55 mM glucose, pH 7.4) equilibrated with 100% O2 at 37°C. This procedure ensured that no blood traces remained in the coronary circulation. Hearts were then freed from fat and connective tissue and minced in ice-cold 0.32 M sucrose containing 1 mM EGTA, pH 7.4. Synaptosomes were isolated as described previously (Imamura et al., 1995; Seyedi et al., 1997), with the following modifications. Minced tissue was digested with 40 to 75 mg of collagenase (type II, Worthington Biochemicals; Freehold, NJ) per 10 ml of HEPES-buffered saline solution (HBS; containing 50 mM HEPES, pH 7.4, 144 mM NaCl, 5 mM KCl, 1.2 mM CaCl2, 1.2 mM MgCl2, 10 mM glucose, and 1 mM pargyline hydrochloride to prevent enzymatic destruction of synaptosomal NE) per gram wet heart weight for 1 h at 37°C. After low-speed centrifugation (10 min at 120g, 4°C), the resulting pellet was suspended in 10 vol of 0.32 M sucrose and homogenized with a Teflon/glass homogenizer. The homogenate was spun at 650g for 10 min, 4°C and the pellet rehomogenized and respun. The pellet, which contained cellular debris, was discarded, and the supernatants from the last two spins were combined and equally subdivided into 10 to 12 tubes. Each tube was centrifuged at 20,000g for 20 min, 4°C. Each pellet containing cardiac synaptosomes was resuspended in HBS to a final volume of 500 μl and incubated with KCl (10-100 mM) or forskolin (0.1-10 μM) in the presence or absence of pharmacological agents in a water bath at 37°C. Each suspension functioned as an independent sample and was used only once. In every experiment, one sample was untreated (control, basal release) and the others were treated with high K+ or forskolin, high K+ or forskolin plus drugs, or with drugs alone. When high K+ was used, osmolarity was maintained constant by adjusting the NaCl concentration. Treated samples were incubated with a given agent for 10 min and then with high K+ for 5 min or forskolin for 20 min. When antagonists were used, samples were incubated with the antagonist for 10 min before incubation with the agonist. Controls were incubated for an equivalent length of time without drugs. At the end of the incubation period each sample was centrifuged again at 20,000g for 20 min, 4°C. The supernatant was assayed for NE content by high-pressure liquid chromatography with electrochemical detection (Seyedi et al., 1997). The pellet was assayed for protein content, by a modified Lowry procedure (Markwell et al., 1978).
Treatment of Synaptosomes with Pertussis Toxin (PTX). When PTX was used, isolated synaptosomes were incubated with PTX for 60 min, and then synaptosomes were washed free of PTX as follows. Tubes containing synaptosomes and 0.3 μg/ml PTX were centrifuged at 20,000g for 5 min, 4°C; the supernatant was discarded and the pellet was suspended in 2 ml of HBS for 10 min. The pellet was recentrifuged at 20,000g for 5 min, 4°C and the supernatant was discarded. The resulting pellet was resuspended in 1 ml of HBS and incubated with KCl (30 and 100 mM) or forskolin (10 μM) in the presence or absence of pharmacological agents for a total of 20 to 25 min in a water bath at 37°C, and recentrifuged at 20,000g for 20 min, 4°C. The supernatant and pellet were assayed for NE and protein content, as mentioned above, respectively.
SH-SY5Y-H3 Cells. A human neuroblastoma cell line stably transfected with the H3R (SH-SY5Y-H3) was kindly supplied by Dr. T. Lovenberg (Silver et al., 2002). Cells were maintained in a 1:1 ratio of Eagle's and Ham's F-12 minimal essential medium mixture, supplemented with 10% fetal bovine serum, 2 mM l-glutamine, 450 μg/ml geneticin, 50 units/ml penicillin, and 50 μg/ml streptomycin at 37°C, 5% CO2. Cells were grown to confluence in 6-, 24-, or 96-well plates for PKA activity, [3H]NE release experiments, and cAMP assay, respectively, or for 4 to 5 days on 22-mm2 standard glass coverslips (no. 1) for [Ca2+]i measurements. Cell culture media and supplements were purchased from Mediatech (Herndon, VA).
[3H]NE Release Assay. The [3H]NE release method was adapted from that described by Murphy et al. (1991). The culture medium was removed and cells were washed once with HEPES-buffered Na+ Ringer's solution (140 mM NaCl, 5 mM KCl, 10 mM HEPES, 2 mM CaCl2, 1 mM MgCl2, pH 7.4), and then incubated in 230 μl/well of Na+ Ringer's buffer containing 50 nM [3H]NE for 60 min at 37°C. This was followed by three washes with 450 μl/well Na+ Ringer's buffer (containing 1 μM desipramine). Release buffer (Na+ Ringer's buffer with 100 mM K+, adjusted to maintain osmolarity) was then added to each well (450 μl/well) for 5 min at room temperature. A 300-μl aliquot of the supernatant was taken from each well for counting, and the remaining solution was discarded. Then 0.3% Triton X-100 was added to the cells (450 μl/well) for 30 min, and 300 μl of lysate was taken for counting. Samples taken for counting were each added to 4 ml of Bio-Safe II scintillation cocktail and counted on a Beckman Coulter LS6000 scintillation counter. For drug experiments, after the three washes cells were incubated in 450 μl of Na+ Ringer's buffer containing the given drug for 5 min at 37°C, followed by release as described above. [3H]NE release was expressed as a percentage of the total [3H]NE content.
cAMP Assay. SH-SY5Y-H3 cells were grown to confluence in 96-well plates. After a 20-min treatment with the cAMP phosphodiesterase inhibitor 3-isobutyl-1-methylxanthine (2 mM), the cells were incubated for 5 min with or without the H3R agonist imetit (100 nM), either alone or in combination with the H3R antagonist clobenpropit (CBP; 25 nM). Cells were incubated with CBP for 5 min before the addition of imetit. Intracellular cAMP levels were then stimulated by either forskolin (10 μM) for 20 min, or high K+ (100 mM) for 5 min. The cells were immediately aspirated and the intracellular cAMP levels determined using a cAMP Biotrak EIA kit (Amersham Biosciences Inc., Piscataway, NJ) following the manufacturer's protocol.
Treatment of SH-SY5Y-H3 Cells with PTX. SH-SY5Y-H3 cells were grown to confluence in six-well plates. Cells were incubated in PTX (200 ng/ml) for 24 h before experimentation.
Detection of PKA Activation. When SH-SY5Y-H3 cells were confluent, cells were washed with serum-free medium and then maintained in Eagle's minimal essential medium/Ham's F-12 with 0.1% bovine serum albumin for 48 h. PKA phosphorylation (i.e., an indication of PKA activation) was elicited by incubating SH-SY5Y-H3 cells with forskolin (10 μM) or K+ (100 mM) for 5 min, in the absence or presence of imetit (100 nM), either with or without CBP (25 nM). SH-SY5Y-H3 cells were lysed (lysis buffer composition: 1% Triton X-100, 0.5% deoxycholic acid, 50 mM Tris-HCl, 0.1% SDS, 1 mM EDTA, 50 mM NaCl). Samples of lysate (10 μg/lane) were prepared with 5× Tris-glycine SDS sample buffer and boiled for 5 min before separation on 8% Tris-glycine SDS-polyacrylamide minigels (Gradipore, French's Forest, NSW, Australia). Electrophoresis was carried out at 200 V, 40 mA/gel for 1 h. Gels were soaked in transfer buffer (25 mM Tris-base, 0.2 M glycine, and 10% methanol, pH 8.5) and electrotransferred to polyvinylidine difluoride (PVDF) membranes (Immobilon-P; Millipore, Billerica, MA) for 90 min at 25 V, 100 mA, room temperature. Membranes were blocked for at least 2 h in blocking buffer [Tris-buffered saline (TBS) containing 0.1% Tween 20, 5% (w/v) nonfat dry milk]. p-PKAα antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA) was used as primary antibody. It was incubated with the PVDF membrane overnight at 4°C, diluted appropriately in primary antibody dilution buffer (TBS containing 0.1% Tween 20, 5% bovine serum albumin). The PVDF membrane was washed three times with TBS and then horseradish peroxidase-coupled anti-rabbit IgG (Cell Signaling Technology Inc., Beverly, MA) was added at a 1:2000 dilution in blocking buffer for 1 h. After three further TBS washes, the protein of interest was detected using enhanced chemiluminescence (LumiGLO; Cell Signaling Technology Inc.) and exposure to X-ray film (BIomax MR; Eastman Kodak, Rochester, NY). In the immunoblot, phosphorylated PKA was visualized, as expected, as a single band at 54 kDa (Tasken et al., 1995). Bands were analyzed by densitometry using NIH Image, version 1.61.
[Ca2+]i Measurements. SH-SY5Y-H3 cells grown on coverslips were loaded with the membrane-permeant form of the [Ca2+]i indicator Fura-2 (10 μM) for 40 min at 37°C. After loading with the dye, the cells were rinsed with Na+ Ringer's solution. The coverslip was attached to the bottom of a flow-through superfusion chamber and mounted on the stage of an inverted epifluorescence microscope (Nikon Diaphot). The cells in the chamber were superfused and maintained at 37°C as described previously (Cardone et al., 1996). Cells were first visualized under transmitted light with a Nikon CF Fluor (40×/1.3 numerical aperture oil immersion objective) before starting fluorescence measurements. Cells were depolarized with a high-K+ solution (based on the Na+ Ringer's composition described above with 100 mM KCl replacing 100 mM NaCl). Calibration of the emitted Fura-2 signal from each cell in the field was carried out in the presence of the Ca2+ ionophore ionomycin (10 μM) in the presence of HEPES buffer containing either 2.6 mM Ca2+ or 10 mM EGTA titrated to pH 7.4. [Ca2+]i levels were calculated as described previously (Grynkiewicz et al., 1985). Cells in the experimental field of view were analyzed singularly and independently from their neighbors.
Reagents. Ionomycin was prepared in dimethyl sulfoxide to a concentration of 10 mM; it was subsequently diluted as mentioned above to a 10 μM solution. Individual vials (50 μg) of the acetoxymethyl derivative of Fura-2 were stored dry at 0°C and reconstituted in dimethyl sulfoxide, at a concentration of 5 mM, for each experiment.
Equipment. The basic components of the experimental apparatus have been described previously (Cardone et al., 1996; Silver et al., 2001). The imaging workstation was controlled using the Metafluor software package (Universal Imaging Corporation, Downingtown, PA). Quantitative image pairs at 340- and 380-nm excitation with emission at 510 nm were obtained either every 15 s before K+ depolarization or every second immediately preceding and during depolarization. The fluorescence excitation was shuttered off except during the brief periods required to record an image. To check for interference from intrinsic autofluorescence and background, images were obtained on cells by using the same exposure time and filter combination used for the experiments, and found to be minor compared with the fluorescence signal.
Drugs and Chemicals. Desipramine hydrochloride (DMI), atropine sulfate, imetit dihydrobromide, clobenpropit dihydrobromide (CBP), PTX, carbamyl choline chloride (carbachol), N6-cyclopentyl adenosine (CPA), 3-isobutyl-1-methylxanthine, forskolin, ω-CTX, nifedipine, ionomycin, pargyline, and Triton X-100 were purchased from Sigma-Aldrich (St. Louis, MO). 8-Bromo-cAMP (8-bromoadenosine-3′,5′-cyclic monophosphate Na), 2′,5′-dideoxyadenosine 9-(2′,5′-dideoxy-erythro-pento furanosyl) adenine (P site ligand), and myristoylated PKI(14-22) amide (PKA inhibitor) were purchased from BIOMOL Research Laboratories (Plymouth Meeting, PA). 8-Cyclopentyl-1,3-Dipropylxanthine (DPCPX) was purchased from Sigma/RBI (Natick, MA). [3H]NE (28.0 Ci/mmol) was purchased from Amersham Biosciences, Inc. Fura-2 AM was obtained from Molecular Probes (Eugene, OR). Forskolin, DPCPX, nifedipine, and ω-CTX were dissolved in dimethyl sulfoxide. Further dilutions were made with HBS buffer; at the concentration used, dimethyl sulfoxide did not affect NE release.
Statistics. NE release values are expressed as mean percentage of increases above basal NE release ± S.E.M. cAMP levels are expressed as mean absolute values ± S.E.M. PKA phosphorylation is expressed as mean percentage of change from the effect of forskolin (10 μM). K+-induced peak [Ca2+]i levels are expressed as mean values (nanomolar) ± S.E.M. Decreases in K+-induced peak [Ca2+]i are expressed as percentage of inhibition ± S.E.M. Statistical analysis was performed by unpaired t test or by one-way ANOVA followed by post hoc testing (Dunnett's test) as indicated in the legend to each figure. A P value of <0.05 was considered statistically significant.
Results
Activation of H3R in Cardiac Synaptosomes Attenuates NE Exocytosis Elicited by K+ and Forskolin. As shown in Fig. 1A, exposure of sympathetic nerve endings isolated from guinea pig hearts (cardiac synaptosomes) to increasing extracellular K+ concentrations (10-100 mM) resulted in an ∼12-33% increase in endogenous NE release. In the presence of the selective H3R agonist imetit (100 nM) the K+-induced NE exocytosis was markedly attenuated. Imetit-induced attenuation of NE exocytosis was abolished when synaptosomes were preincubated with the selective H3R antagonist CBP (25 nM) (Van der Goot et al., 1992) (Fig. 1A). Since NE exocytosis can also be elicited when the intraneuronal cAMP level is increased (Markstein et al., 1984; May et al., 1995), we incubated cardiac synaptosomes with increasing concentrations of the adenylyl cyclase activator forskolin (0.1-10 μM). This resulted in an ∼10 to 35% increase in endogenous NE release (Fig. 1B). In the presence of imetit (100 nM) the forskolin-induced NE release was markedly attenuated. CBP (25 nM) abolished the effect of imetit.
The H3R-Induced Attenuation of NE Exocytosis in Cardiac Synaptosomes Is Mediated by a Gi/Go-Coupled Decrease in Adenylyl Cyclase Activity. These findings revealed that activation of H3R attenuates NE exocytosis whether it is elicited by membrane depolarization with K+ or by stimulation of adenylyl cyclase with forskolin. As shown in Fig. 2, NE exocytosis from cardiac synaptosomes was enhanced not only by forskolin but also by administration of the permeant form of cAMP, 8-bromo-cAMP (1 mM) (Meyer and Miller, 1974). Similar to imetit, the adenylyl cyclase inhibitor P-site ligand (Desaubry et al., 1996) markedly attenuated the forskolin-induced NE release from cardiac synaptosomes (Fig. 2). In contrast, when NE release was induced by 8-bromo-cAMP, neither imetit nor the P-site ligand had any effect (Fig. 2). This suggested that the release of NE elicited by forskolin results from an increase in intraneuronal cAMP and that a reduction in adenylyl cyclase activity is the pivotal mechanism by which imetit attenuates NE exocytosis.
Since inhibition of adenylyl cyclase could be due to coupling of H3R to Gi/Go protein (Lovenberg et al., 1999), we pretreated cardiac synaptosomes with the inhibitor of Gi/Go PTX. As shown in Fig. 3A, in cardiac synaptosomes pretreated with PTX, imetit failed to attenuate NE exocytosis elicited by depolarization with K+. PTX pretreatment also prevented the imetit-induced attenuation of NE release elicited by forskolin (Fig. 3B). As a positive control of a Gi/Go involvement in the imetit-induced attenuation of NE exocytosis, we activated muscarinic (i.e., M2R and M4R) and adenosine A1-receptors, which are both known to be coupled to Gi/Go (Caulfield and Birdsall, 1998; Fredholm et al., 2001). As shown in Fig. 4, the forskolin-induced NE release was greatly attenuated by the muscarinic agonist carbachol (100 nM; Fig. 4A) and by the selective adenosine A1-receptor agonist CPA (100 nM; Fig. 4B). The effects of carbachol and CPA were prevented by the respective muscarinic- and A1-receptor antagonist atropine (100 nM) and DPCPX (300 nM) (Fig. 4, A and B). Similar to imetit, the carbachol- and CPA-induced attenuation of NE release was abolished by PTX pretreatment (Fig. 3B).
The H3R-Mediated Attenuation of NE Exocytosis in SH-SY5Y-H3 Cells Is Associated with a Decrease in Intracellular cAMP. Since these findings implied an H3R-mediated, Gi/Go-coupled, decrease in adenylyl cyclase activity, we next assessed whether this would result in a decrease in intracellular cAMP levels. For this, we used the H3R-transfected neuroblastoma cell line SH-SY5Y (SH-SY5Y-H3) (Silver et al., 2002). As shown in Fig. 5A, forskolin (0.1-10 μM), elicited an ∼10-35% increase in [3H]NE release. At 10 μM, forskolin also caused a large increase in cAMP (Fig. 5D). Imetit (100 nM) markedly attenuated the increase in cAMP and associated NE release in response to 10 μM forskolin (Fig. 5, B and D). Both effects were inhibited by the H3R-antagonist CBP (25 nM) (Fig. 5, B and D). Similarly, imetit antagonized the increase in cAMP and associated NE release elicited by K+ depolarization; CBP blocked both of these effects (Fig. 5, C and E).
The H3R-Induced Attenuation of NE Exocytosis in SH-SY5Y-H3 Cells and Cardiac Synaptosomes Is Mediated by a Decrease in PKA Activity. Inasmuch as the attenuation of NE exocytosis elicited by H3R activation coincided with a decrease in intracellular cAMP, we next determined whether a diminished PKA activity may be the intermediate step between the reduction in cAMP and the decreased NE exocytosis. Since PKA phosphorylation is a measure of PKA activity (Erlichman et al., 1974), we determined PKA phosphorylation by Western blot analysis in SH-SY5Y-H3 cells. As shown in Fig. 6, imetit (100 nM) significantly reduced the level of PKA phosphorylation elicited by either forskolin (10 μM; left) or K+ (100 mM; right). Pretreatment with CBP (25 nM) prevented the imetit-induced reduction of PKA activity, independently of whether PKA phosphorylation was enhanced by forskolin or K+ (Fig. 6). Not shown in Fig. 6 is that when SH-SY5Y-H3 cells were preincubated with PTX, the effect of imetit on PKA was abolished. To further assess the role of a diminished PKA activity in the H3R-mediated attenuation of NE exocytosis, we next determined whether PKA inhibition would reduce NE exocytosis and whether a synergistic effect could be seen when H3R activation was combined with PKA inhibition. As shown in Fig. 7, A and B, the cell-permeable cAMP-dependent PKA inhibitor [PKI(14-22); 0.2-20 nM] (Glass et al., 1989) decreased both K+- and forskolin-induced NE exocytosis in cardiac synaptosomes by ∼5 to 60%. The effect of PKI(14-22) was similar to that of imetit (Fig. 7, C and D). When subthreshold concentrations of PKI(14-22) (0.2 nM) and imetit (0.2-3 nM) were used in combination, a marked synergistic effect was observed (Fig. 7, E and F). Collectively, these findings suggested that a decrease in PKA activity is likely to be involved in the H3R-mediated attenuation of NE exocytosis.
The H3R-Mediated Decrease in PKA Activity in SH-SY5Y-H3 Cells Leads to a Reduction in [Ca2+]i and an Attenuation of NE Exocytosis. We next questioned whether the H3R-mediated decrease in PKA activity would lead to a reduction in [Ca2+]i and thus, NE exocytosis. For this, we measured peak [Ca2+]i in SH-SY5Y-H3 cells depolarized with K+, in the absence and presence of imetit and PKI(14-22), both alone and in combination. As shown in Fig. 8 A, in response to 100 mM K+, peak [Ca2+]i rose to ∼310 nM. In the presence of imetit (3 and 100 nM), peak [Ca2+]i was reduced to ∼250 and ∼200 nM, respectively. Similarly, in the presence of PKI(14-22) (0.02 and 20 nM), peak [Ca2+]i decreased to ∼250 and ∼200 nM, respectively. When the two lower doses of imetit and PKI(14-22) were combined (3 and 0.02 nM, respectively) peak [Ca2+]i decreased to ∼160 nM. Thus, as shown in Fig. 8B, although an ∼20% inhibition of K+-induced peak [Ca2+]i occurred with each 3 nM imetit and 0.02 nM PKI(14-22), when imetit and PKI(14-22) were combined at these concentrations, the inhibition of K+-induced peak [Ca2+]i significantly increased to ∼50%.
A Decrease in Ca2+ Influx through N- and L-Type Ca2+ Channels Is Involved in the H3R-Mediated Attenuation of NE Exocytosis in Cardiac Synaptosomes. The finding that H3R activation and PKA inhibition acted synergistically to reduce peak [Ca2+]i in response to K+ implied that a decreased phosphorylation of VOCC and thus, Ca2+ influx, may be responsible for the H3R-mediated reduction in peak [Ca2+]i. We thus assessed the role of VOCC in the K+- and forskolin-induced NE exocytosis in cardiac synaptosomes. As shown in Fig. 9, the K+-induced NE release was also markedly diminished by the selective N-type and L-type Ca2+ channel inhibitors ω-CTX (100 nM) and nifedipine (5 μM), both alone and in combination (Fig. 9A). The forskolin-induced NE release was also markedly diminished by ω-CTX (100 nM) and nifedipine (5 μM), both alone and in combination (Fig. 9B). These findings revealed not only a similarity between K+-depolarization and forskolin administration but also a resemblance of H3R activation to N- and L-type Ca2+ channel inhibition. As this resemblance suggested a similarity of mechanisms of action, we next investigated the effects of imetit in combination with either Ca2+ channel blocker. As shown in Fig. 10, A and B, ω-CTX and nifedipine each inhibited as a function of its concentration NE exocytosis elicited by K+ depolarization of cardiac synaptosomes. When a subthreshold concentration of imetit (3 nM) was used in combination with a subthreshold concentration of either ω-CTX (0.1 nM) or nifedipine (0.1 μM), a marked synergistic effect was observed (Fig. 10C). These findings suggested that a decrease in Ca2+ influx through N- and L-type Ca2+ channels is likely to be involved in the H3R-mediated attenuation of NE exocytosis.
Discussion
We had reported that activation of presynaptic H3R inhibits the exocytotic release of NE elicited by depolarization of cardiac sympathetic nerve endings (cardiac synaptosomes) (Seyedi et al., 1997), atrial tissue (Imamura et al., 1995), and intact heart (Imamura et al., 1994). While exploring possible mechanisms of this modulatory action, we found that H3R activation is associated with a marked decrease in [Ca2+]i (Silver et al., 2002). We hypothesized that when the neuronal membrane is depolarized and H3R are activated, Ca2+ influx through VOCC is diminished. Conceivably, a decreased phosphorylation of VOCC could play a role in this decreased Ca2+ influx. We have thus explored whether H3R stimulation might be associated with a decreased cAMP-dependent phosphorylation, which would result in a sequential decrease in neuronal Ca2+ influx, [Ca2+]i and NE exocytosis.
Since NE exocytosis can be elicited when intraneuronal cAMP is enhanced (Markstein et al., 1984; May et al., 1995), and an increase in extracellular K+ is known to elevate intraneuronal cAMP (Cooper et al., 1998), we compared NE exocytosis initiated by K+-induced depolarization with the exocytosis elicited by adenylyl cyclase stimulation with forskolin. As a target, we used cardiac adrenergic nerve terminals isolated from the guinea pig heart (i.e., synaptosomes) (Seyedi et al., 1997) and SH-SY5Y-H3 neuroblastoma cells (Silver et al., 2002), endowed with native and stably transfected H3R, respectively. We found that H3R activation attenuated NE exocytosis independently of whether it was elicited by neuronal depolarization or adenylyl cyclase stimulation. Although this indicated an effect on a common downstream signal, most likely cAMP, we found that imetit effectively attenuated NE exocytosis initiated by forskolin, but not that initiated by the administration of the permeant form of cAMP, 8-bromo-cAMP. This clearly indicated that adenylyl cyclase, and not cAMP, is the pivotal initial site of H3R-induced attenuation of NE exocytosis. Indeed, a decreased generation of cAMP by forskolin had already been observed in SKNMC-H3 neuroblastoma cells as a result of H3R stimulation (Lovenberg et al., 1999).
The next question was how H3R activation results in a diminished adenylyl cyclase activity. Preliminary evidence from our laboratory had suggested that Gi/Go may be involved in the H3R-mediated attenuation of adrenergic inotropic responses in the heart (Endou et al., 1994). Other investigators had also suggested a Gi/Go involvement in the H3R-mediated inhibition of NE release from intestinal sympathetic nerves (Blandizzi et al., 2000). We therefore determined whether the H3R-mediated inhibition of NE exocytosis elicited by K+ or forskolin is attenuated by PTX, a toxin that inactivates Gi/Go via ADP-ribosylation (Bokoch et al., 1983; Codina et al., 1983). Inasmuch as muscarinic and adenosine A1-receptors are coupled to Gi/Go (Caulfield and Birdsall, 1998; Fredholm et al., 2001), we activated these receptors on cardiac synaptosomes (with carbachol and CPA, respectively), we verified that their activation would attenuate the forskolin-induced NE exocytosis, and then, as a positive control, ensured that pretreatment with PTX would prevent the attenuation of NE exocytosis by carbachol and CPA. Indeed, PTX pretreatment abolished equally well the inhibitory effects of carbachol, CPA, and imetit. In fact, PTX abolished also the imetit-induced attenuation of NE exocytosis elicited by depolarization with K+. Accordingly, our data compellingly demonstrate that H3R are negatively coupled to adenylyl cyclase via Gi/Go.
Given this H3R-mediated, Gi/Go-coupled decrease in adenylyl cyclase activity, we next assessed whether H3R activation would lower intracellular cAMP levels. For this, we used the SH-SY5Y-H3 neuroblastoma cell line, an optimal model of sympathetic nerve endings (Silver et al., 2002), better suited than cardiac synaptosomes for the measurement of cyclic nucleotides. We found that NE exocytosis from SH-SY5Y-H3 cells, elicited by either K+ or forskolin, was associated with an increase in cAMP. Notably, H3R activation inhibited NE exocytosis as well as the increase in cAMP, suggesting that the H3R-induced attenuation of NE exocytosis is mediated by a decrease in cAMP. Inasmuch as a fall in cAMP would be expected to result in a decreased cAMP-dependent phosphorylation, we next assessed whether H3R activation is associated with a decrease in PKA activity. Since PKA phosphorylation is a measure of PKA activity (Erlichman et al., 1974), we determined levels of PKA phosphorylation by Western blot analysis in SH-SY5Y-H3 cells. We found that H3R activation significantly reduced the stimulation of PKA activity in response to either forskolin or K+. Moreover, we found that the selective H3R agonist imetit (Garbarg et al., 1992) and the specific PKA inhibitor PKI(14-22) (Glass et al., 1989) acted synergistically to inhibit NE exocytosis from cardiac synaptosomes, independently of whether NE exocytosis was elicited by K+ or forskolin. Notably, 8-bromo-cAMP is a potent stimulator of PKA (Hei et al., 1991). We found that imetit, like the P-site ligand, inhibited the forskolin-induced NE exocytosis but not that elicited by 8-bromo-cAMP (see Fig. 2). Thus, all of our evidence indicates that the reduction in adenylyl cyclase activity which is the hallmark of H3R activation (Lovenberg et al., 1999) leads to a decrease in cAMP and PKA activity in cardiac sympathetic nerve terminals.
cAMP-dependent phosphorylation of VOCC increases their activity (Ahlijanian et al., 1991; Hell et al., 1995; Catterall, 2000); thus, a decreased phosphorylation due to inhibition of the cAMP/protein kinase A pathway could conceivably be involved in the H3R-mediated attenuation of Ca2+ influx, [Ca2+]i, and NE exocytosis. Indeed, we found that H3R activation and PKA inhibition each reduced peak [Ca2+]i in response to K+-induced depolarization in SH-SY5Y-H3 cells; moreover, imetit and PKI(14-22) acted synergistically to diminish peak [Ca2+]i. This implied that a decreased phosphorylation of VOCC and thus, Ca2+ influx, may be responsible for the H3R-mediated reduction in [Ca2+]i. In fact, we observed that ω-CTX (Sher et al., 1991) and nifedipine (Vater et al., 1972), both alone and in combination, markedly diminished NE exocytosis elicited by K+ and forskolin in cardiac synaptosomes. These findings revealed that both N- and L-type Ca2+ channels are involved in the exocytotic release of NE from cardiac sympathetic nerve endings. Furthermore, given the similarity between H3R activation and VOCC inhibition (compare Figs. 1 and 9), our findings implied that the H3R-induced decrease in neuronal [Ca2+]i is ultimately due to an inhibition of Ca2+ influx. In support of this concept, we found that imetit acted synergistically with each of the N- and L-type Ca2+ channel blockers, ω-CTX and nifedipine, in attenuating NE exocytosis. Although PKA is known to promote Ca2+ release from the endoplasmic reticulum (ER) (Bugrim, 1999), it is unlikely that H3R activation reduces neuronal [Ca2+]i by limiting Ca2+ release from the ER. Indeed, neither imetit nor PKI(14-22) affects the caffeine-induced exocytotic release of NE from cardiac synaptosomes; furthermore, ryanodine does not modify NE exocytosis elicited by membrane depolarization with K+ (N. Seyedi and R. Levi, unpublished).
Collectively, our findings suggest that the H3R-mediated attenuation of NE exocytosis from cardiac sympathetic nerves involves an H3R-Gi/Go coupling, adenylyl cyclase inhibition by Giα, decreased cAMP formation, diminished PKA activity, decreased Ca2+ influx through VOCCs, culminating in a decreased [Ca2+]i transient and thus, in an impaired exocytosis. The possibility should also be considered that the H3R-mediated attenuation of NE exocytosis may result in part from an inhibition of Ca2+ influx by direct coupling of the Gβγ subunit to VOCC (Herlitze et al., 1996; Ikeda, 1996; Catterall, 2000) or to a direct interaction between Gβγ and the exocytotic fusion machinery at the presynaptic terminal downstream of Ca2+ entry (Blackmer et al., 2001).
In conclusion, the elucidation of transductional mechanisms implicated in the H3R-induced modulation of NE exocytosis in cardiac sympathetic nerve terminals will help our understanding of neurotransmitter release in hyperadrenergic states characterized by enhanced NE exocytosis, such as myocardial ischemia, hypertension, and congestive heart failure.
Acknowledgments
Preliminary experiments were performed by Guy Partridge and Marcus McFerren.
Footnotes
-
This work was supported by National Institutes of Health Grants HL34125, HL46403, and DK60726 and Minority Access to Research Careers Grant F31GM64875.
-
doi:10.1124/jpet.104.072504.
-
ABBREVIATIONS: H3R, histamine H3-receptor; NE, norepinephrine; [Ca2+]i, intracellular Ca2+ concentration; ω-CTX, ω-conotoxin GVIA; VOCC, voltage-operated Ca2+ channel; HBS, HEPES-buffered saline; PTX, pertussis toxin; PKA, cAMP-dependent protein kinase; CBP, clobenpropit; PVDF, polyvinylidene difluoride; TBS, Tris-buffered saline; DMI, desipramine; CPA, N6-cyclopentyl adenosine; DPCPX, 8-cyclopentyl-1,3-dipropylxanthine; ANOVA, analysis of variance; PKI(14-22), PKA inhibitor.
- Received June 8, 2004.
- Accepted August 11, 2004.
- The American Society for Pharmacology and Experimental Therapeutics