Vasopressin (VP) release from the hypothalamo-neurohypophyseal system (HNS) is stimulated by ATP activation of P2X purinergic receptors and by activation of α1-adrenergic receptors by phenylephrine (PE). These responses are potentiated by simultaneous exposure to ATP+PE. Potentiation was blocked by depleting intracellular calcium stores with thapsigargin. To test the hypothesis that the synergistic response to ATP+PE reflects alterations in the intracellular calcium concentration ([Ca2+]i), [Ca2+]i was monitored in supraoptic neurons in HNS explants loaded with fura 2-AM. Both ATP and PE induced rapid, but transient, elevations in [Ca2+]i. In 0.3 mM Ca2+, the peak response to ATP was greater than to PE but did not differ from the peak response to ATP+PE. A sustained elevation in [Ca2+]i was induced by ATP+PE, that was greater than ATP or PE alone. In 2 mM Ca2+, the peak response to ATP+PE was significantly greater than to either ATP or PE alone, and the sustained response to ATP+PE was greater than to either agent alone. Responses were comparable in the presence of TTX. The sustained elevation in [Ca2+]i was also observed when ATP+PE was removed after 1 min, but it was eliminated by either thapsigargin or removing external calcium, indicating that both calcium influx and calcium release from internal stores are required. Some cells were vasopressinergic based on a VP-induced increase in [Ca2+]i. These observations support the hypothesis that simultaneous exposure to ATP+PE induces a different pattern of [Ca2+]i than either agent alone that may initiate events leading to synergistic stimulation of VP release.
- posterior pituitary
vasopressin (vp) release from the posterior pituitary is stimulated by a decrease in blood volume or pressure (25). Although information about moderate decreases in blood volume and/or pressure is transmitted to the neurons in the supraoptic nucleus (SON) and paraventricular nucleus (PVN) via the A1 noradrenergic pathway (4, 12, 30, 31), adrenoreceptor antagonists did not block A1 activation of VP cells, suggesting that these neurons use agents in addition to norepinephrine (NE) for neurotransmission (5). The finding that application of the P2 purinergic receptor antagonist, suramin (10 mM) in the SON reversibly blocked excitation of VP neurons by A1 stimulation without preventing the excitatory effect of locally applied NE (6, 7) suggested that in addition to NE, this pathway also uses ATP as a neurotransmitter. Local suramin injections in SON also reduced Fos expression induced by moderate hemorrhage, but not by osmotic stimulation in SON neurons (3). In studies designed to determine the importance of corelease of ATP and NE from this A1 pathway on VP release, it was found that simultaneous exposure of explants of the hypothalamo-neurohypophyseal system (HNS) to ATP and phenylephrine (PE), an alpha-1-adrenergic receptor agonist, resulted in synergistic stimulation of VP release. The synergistic response was larger than that observed with ATP or PE independently, and in contrast to the transient responses observed in response to independent delivery of ATP or PE, was sustained for several hours (17). The sustained component was blocked by actinomycin, indicating that new gene transcription is required (17). Because additional experiments implicated the calcium permeable ionotropic P2x-purinergic receptors in the ATP response (17) and because activation of alpha-1-adrenergic receptors activates inositol triphosphate and releases calcium from intracellular stores (9), the current experiments tested the hypothesis that the synergistic response to ATP+PE reflected convergence of these signaling cascades to increase intracellular calcium.
To test the hypothesis, the effect of depleting intracellular calcium stores with thapsigargin (TG) treatment on the synergistic action of ATP+PE on VP release from HNS explants was examined. However, it was not possible to use VP release to evaluate the effect of calcium influx, because exocytosis is dependent on calcium influx. Therefore, changes in intracellular calcium ([Ca2+]i) were measured using the calcium sensitive dye, fura-2AM.
Although the effect of ATP on [Ca2+]i in dissociated SON neurons has been reported previously (24), the current studies extend these observations to evaluate the interaction between ATP and PE and use an organotypic preparation, the HNS explant, rather than dissociated neurons for calcium imaging. The latter is important, because in the HNS explant, neuronal/glial relationships are preserved, as well as local neuronal circuitry. The explant is also advantageous, because it is the same preparation in which the synergistic effect of ATP+PE on VP and oxytocin (OT) release was observed. Therefore, observed changes in [Ca2+]i could induce calcium-dependent mechanisms that drive new gene transcription and protein synthesis that sustains the effect of ATP+PE on hormone release. The effect of ATP+PE on [Ca2+]i was examined under four distinct conditions: extended (4 min) exposure to ATP+PE in low extracellular calcium ([Ca2+]o; 0.3 mM), normal [Ca2+]o (2 mM), and in the presence of TTX, and after short (1 min) exposure to ATP+PE. Low [Ca2+]o was used to replicate the conditions of the hormone release studies. However, because low [Ca2+]o increases the excitability of SON neurons by increasing the amplitude of the persistent Na+ current (INaP) (19), it was important to determine whether the effects of ATP+PE were dependent on [Ca2+]o. TTX was used to determine whether the effects of ATP+PE were dependent on INaP or on transient Na+ current (e.g., action potential generation). The latter could reflect activation of local neuronal circuitry by ATP and/or PE.
MATERIALS AND METHODS
Male Sprague-Dawley rats [Crl:CD(SD)Br; Charles Rivers Laboratories, Wilmington, MA] were used in all experiments. All protocols used were performed in accordance with the National Institutes of Health Guidelines for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee of the University of Colorado at Denver and Health Sciences Center.
HNS Explant Perifusion Experiments
Explants of the HNS were obtained from decapitated male Sprague-Dawley rats (125–150 g) as described previously (28). They were dissected from the base of the hypothalamus and included the VP neurons of the SON with their axonal projections extending through the median eminence and terminating in the neural lobe. They also included the organum vasculosum of the lamina terminalis, and the arcuate nucleus, but not the PVN. Explants were perifused individually in closed culture chambers (0.5 ml) at 37°C with oxygenated culture medium at a rate of 2.0 ml/h. The medium consisted of F12 nutrient mixture (Sigma, St. Louis, MO) fortified with 20% fetal bovine serum, 1 mg/ml glucose, 50 mU/ml penicillin, 50 mg/ml streptomycin. Bacitracin (0.28 mM) was added to the medium to prevent degradation of VP (26). The explants were equilibrated for 4 h and then maintained under different treatment conditions for an additional 4 to 5 h. Outflow was collected individually at 20-min intervals using a refrigerated fraction collector maintained at 4°C.
To evaluate the effect of TG alone on basal VP release, explants were treated with TG (200 nM) or vehicle (DMSO) for 0.5 h after 4 -h equilibration. TG was dissolved in 50 μl DMSO, and this was diluted to 50 ml.
To evaluate the effect of depleting the intracellular Ca2+ stores on the response to ATP+PE, explants were treated with TG (200 nM) or vehicle (DMSO) for 0.5 h after 4 -h equilibration and after 0.5 -h wash, both groups were exposed to ATP/PE (100 μM each) for another 4 h. The medium for both groups contained 0.03% ascorbic acid to prevent PE oxidation.
VP concentration in the perifusate was determined by RIA, as described previously (29), using antisera against VP (Arnel Products, Brooklyn, NY) at a final dilution of 1:100,000. The buffer used was 0.1 M PBS, pH 7.6, with 1 mg/ml BSA and 1 mg/ml sodium azide. Both standards and samples were incubated for 72 h at 4°C in the presence of 5,000 cpm of 125I-labeled arginine VP (NEN, DuPont). Antibody-bound VP was separated from free peptide with dextran-coated charcoal, and the amount of 125I-labeled VP in the pellet was determined with a γ-counter. The assay has a minimum sensitivity of 1.0 pg/tube and a useful range up to 100 pg/tube.
Basal hormone release was determined for each explant as the hormone release at the end of the 4-h equilibration period. Subsequent hormone release is expressed as a percentage of this basal value. In each experiment, two-way ANOVA with repeated measures and post hoc test (Student-Newman-Keuls) was used to determine the specific group differences at individual time points. The level of significance was set to P < 0.05. Results are expressed as the means ± SE.
The same HNS explant preparation, prepared from 125–150 g male Sprague-Dawley rats was used for live cell Ca2+ imaging. The acetoxymethyl (AM) ester form of fura-2 (Molecular Probes, Carlsbad, CA; 50 μg) was dissolved in 50 μl of anhydrous DMSO to make a 1 mM stock solution. Explants were incubated in 100 μM fura-2AM for 10 min followed by a 50 min incubation in 20 μM fura-2AM in medium containing 0.02% pluronic acid F-127 (Molecular Probes). Dye loading and imaging were performed in F12 nutrient medium (Sigma) fortified with 1 mg/ml glucose, (and 1.7 mM CaCl2 in some experiments), and gassed with 95%O2-5%CO2.
Perifusion and drug delivery.
Explants were placed in a chamber fitted onto the stage of an upright fluorescent microscope. The chamber was formed by an adaptor (Bioscience Tools, San Diego, CA) placed in a 35 mm Corning petri dish whose base was coated with Sylgard (1 mm). The explant was stabilized with a weighted, ring-shaped stainless steel net, which was pinned onto the Sylgard base. Gassed medium was delivered by a gravity-driven 8-valve system (Warner Instruments, Hamden, CT) to the perifusion port at a rate of ∼3 ml/min. and was removed by a vacuum port at the opposite side.
Fluorescence ratio imaging.
The explant was positioned in the recording chamber with its ventral side up, stabilized with a weighted nylon mesh, and allowed to equilibrate for 1 h. Thus the optic chiasm is easily visualized, and the SON can be located anatomically in the tissue immediately rostral and parallel to optic chiasm. Fura-2 loaded magnocellular neurons were alternatively excited with 340 nm and 380 nm UV light from a Xenon source (Sutter Instruments, Novato, CA). Exposure time for 380-nm wavelength was between 200 and 500 ms and was tripled for 340 nm wavelength accordingly. Emitted light was passed through a 60× fluor water immersion lens attached to an Olympus upright microscope and collected at 510 nm by an intensified charged-couple device camera (Hamamatsu, Japan). Paired 340/380 excitation images were acquired every 3 s using SlideBook software (Intelligent Imaging, Denver, CO) for a period of 100 frames. Explants treated with 10 μM ionomycin to allow [Ca2+]i to equilibrate with extracellular Ca2+ gave an apparent Rmax of 1.0684 ± 0.001 (n = 15) in 0.3 mM Ca2+ and 2.0643 ± 0.0028 (n = 16) in 2 mM Ca2+. As both values far exceed the highest ratio achieved by all agents studied in all experimental preparations, the peak responses reported were not limited by the maximum detectable change in the 340/380 ratio. The ratio data are presented as a percentage of the basal 340/380 ratio for each cell (see below).
Magnocellular neurons were identified by the size of the cell body (>25 μm in diameter), their location, and for some, their response to VP. Data from the identified neurons were collected as the average ratio of fluorescence from 340 nm excitation over fluorescence from 380 nm excitation from a region of interest that included the complete area of the cell body, including the nucleus if it were (rarely) evident at the focal plane of the image. This ratio is a direct, linear measurement of [Ca2+]i within physiological ranges (14). A baseline ratio value for each neuron was determined by averaging ratios of the 10 frames preceding drug application. Responses were then expressed as percentage changes in 340/380 ratio relative to basal. The end ratio was calculated by averaging the percentage ratios for frames 90–100 for each neuron. The mean ± SE of the percentage values from individual neurons were calculated and plotted. Parametric one-way ANOVA (F ratio) or Kruskal-Wallis one-way ANOVA on ranks (H value) were used to determine whether there were significant group differences in the peak response and the end ratio for ATP alone, PE alone, or ATP+PE. Paired Student's t-test analysis was used to determine whether the end-ratio was different from the initial basal ratio for any treatment group.
Perifusion Experiments: Effect of TG.
Effect of TG treatment on basal VP release.
VP release from HNS explants is stimulated by increases in osmolality (27, 35). Therefore, the increase in osmolality induced by the use of DMSO as the vehicle for TG, caused a transient increase in VP release that was observed in both the control and TG groups (Fig. 1A). TG treatment did not consistently alter the DMSO-induced VP release (Fig. 1A and B). Depletion of the intracellular Ca2+ store by TG is irreversible. Therefore, it and the vehicle DMSO were washed out after a 30-min exposure. With the return of osmolality to the basal level, subsequent VP release was not significantly different between the control and TG groups (Fig. 1A).
Effect of TG treatment on VP response to ATP+PE.
In the absence of TG treatment, ATP+PE induced a sustained increase in VP release that characterizes the synergistic response to ATP+PE (Ftime = 3.277, P < 0.001). Depletion of intracellular Ca2+ stores by TG treatment completely prevented this sustained stimulation of VP release by ATP+PE (F=25.268, P < 0.001) (Fig. 1B). This result supports the hypothesis that mobilization of intracellular Ca2+ is required for synergistic stimulation of VP release by ATP+PE.
Identification of SON neurons for calcium imaging.
Two approaches were useful in identifying fura-2 loaded cells as SON neurons.
Location in explant.
Imaging was performed from the ventral surface of the HNS explant. SON is located immediately rostral and parallel to the optic chiasm (see Fig. 2A) which is easily visualized as a dark shadow when the explant is viewed under either bright light or UV illumination and serves as an excellent landmark for localizing SON. Fig. 2B shows the position of one imaged area and its relation to the optic chiasm. Neuronal size (>25 μm in diameter) and shape were also used to confirm that the fura-2-loaded cells were the magnocellular SON neurons. At high power (Fig. 2C), the characteristic bipolar/tripolar shapes of the large SON neurons are evident and neurites are seen to emanate from the cell body.
Response to VP.
Vasopressinergic neurons express V1a VP receptors and respond to VP with an increase in [Ca2+]i, whereas oxytocinergic neurons express oxytocin receptors and respond to oxytocin with an increase in [Ca2+]i. These responses have been used to identify VP and OT neurons in dissociated cell preparations (8, 16). As shown in Fig. 3, in a single imaged field in one HNS explant, five neurons showed an increase in [Ca2+]i when exposed to VP, identifying these neurons as vasopressinergic and confirming that the localization procedure used above identifies SON neurons. Because of the expense of using VP or OT to identify SON neurons, this approach was not routinely used in all explants.
Calcium Responses to ATP, PE, or ATP+PE.
The 340/380 ratio was used as an index of [Ca2+]i. Under basal conditions, the 340/380 ratio for SON neurons ranged from 0.45 to 0.77. To analyze changes in [Ca2+]i, the basal ratio was determined for each cell by averaging the 10 ratio measurements (frames) immediately preceding addition of the test agent, and responses to the agent were expressed as a percentage of this average basal ratio (% of basal).
As shown in Figs. 3 and 4, exposure of explants to ATP (200 μM) resulted in a rapid increase in [Ca2+]i that reached a peak within 25–60 s and then quickly declined to a plateau level that was sustained throughout the period of exposure to ATP (3–4 min). Fig. 4 shows the response to ATP in one preparation. Pseudocolor images depicting the 340/380 ratio indicative of the basal [Ca2+]i and the peak change in [Ca2+]i are shown in Fig. 4A and B, respectively. Fig. 4C shows the time course of the change in the 340/380 ratio for individual cells imaged in a preparation treated with ATP, and Fig. 4D shows the 340 and 380 fluorescence records for three representative cells. It is evident that individual cells show slight variation in the time to reach the peak. This may reflect differences in the diffusion distance to cells that are deeper in the preparation. Variability in the peak response is also evident, and spontaneous elevations in [Ca2+]i are present both before and after ATP exposure in some cells. During the period preceding drug application, spontaneous peaks in [Ca2+]i were observed in 19% of 114 neurons imaged in six different explants. Treatment of the explants with TTX before imaging reduced the occurrence of spontaneous spikes during the basal period to 8.6% (10 out of 116 cells in six different explants.)
Figure 5 shows the time course of responses to ATP, PE, and ATP+PE (200 μM each) in 0.3 mM Ca2+. The responses of multiple neurons in a single visual field were analyzed in 3 or 4 separate explants per drug treatment group. Because subsequent drug exposures generally yielded smaller responses, the responses shown in Fig. 5 and analyzed in Fig. 6A represent the first drug exposure for each explant. As seen in Fig. 5, in 0.3 mM Ca2+, the peak response to ATP and ATP+PE was similar in magnitude, but the peak response to PE was smaller (Fig. 6A, F = 7.483, P < 0.001). The plateau response to ATP and PE was similar and remained higher than basal (P < 0.001), but combined exposure to ATP+PE resulted in a greater sustained elevation in [Ca2+]i. This sustained elevation in [Ca2+]i was not due to loss of Ca2+ homeostasis and cell death, because as seen in Fig. 3, cells were able to mount subsequent Ca2+ responses. To analyze the sustained response, the “end ratio” was determined as the average of the last 10 ratios collected from each cell. The end ratio in response to ATP+PE was significantly larger than that to either ATP or PE alone (Fig. 6A; H = 35.264, P < 0.001). The larger sustained elevation in [Ca2+]i may drive mechanisms responsible for the synergistic stimulation of hormone release by ATP+PE.
The data presented in Figs. 5 and 6A were obtained from experiments in which F-12 nutrient mixture was used as the perifusion medium to duplicate the conditions used in earlier hormone release experiments (17). However, the Ca2+ concentration in this medium was only 0.3 mM Ca2+. Therefore, to determine whether the responses to ATP, PE, and ATP+PE would be similar in 2 mM Ca2+ (the physiological extracellular Ca2+), the experiments were repeated in F12 nutrient mixture supplemented with CaCl2. The peak ratios and end ratios for experiments performed in 2 mM Ca2+ are shown in Fig. 6B. In 2 mM Ca2+, all elevations in [Ca2+]i were larger than those in 0.3 mM Ca2+ medium (peak ratios: P < 0.001 for ATP, PE, and ATP+PE; end ratios: P < 0.001 for ATP, PE, and ATP+PE). The peak ratio in response to ATP+PE was larger than that to PE alone (P < 0.001), as observed in low Ca2+. The peak response to ATP+PE was also larger than the peak response to ATP, but this was not observed in subsequent experiments (see below and Figs. 6C and D). In 2 mM Ca2+, extended exposure to ATP or PE alone resulted in sustained elevations in [Ca2+]i compared with basal (P < 0.001), but the sustained elevation in [Ca2+]i was still larger in response to ATP+PE compared with ATP or PE alone. The end ratios followed this order: ATP+PE>PE>ATP (H = 48.108, P < 0.001). Thus in both 0.3 mM and 2 mM external Ca2+, the larger sustained elevation in [Ca2+]i in response to ATP+PE could initiate events leading to synergistic stimulation of hormone release by ATP+PE.
Effect of TTX on responses to ATP, PE, and ATP+PE.
To evaluate the role of efferent activity endogenous to the HNS explant in the responses to ATP and PE, the above experiments were repeated in the presence of TTX. Explants were pretreated with TTX (3 μM) for 5 min, and TTX was maintained in the perifusate throughout exposure to ATP, PE, or ATP+PE. Modified F-12 nutrient mixture with 2 mM Ca2+ was used as the perifusate. The results are shown in Fig. 6C. Neither the peak nor end responses to ATP and PE were significantly different than those observed in 2 mM Ca2+ without TTX (Fig. 6B). Both the peak and end responses to ATP+PE were smaller in the TTX treated preparations (F=9.34, P = 0.003 and F=11.83, P < 0.001, respectively), but this may reflect unusually large responses to ATP+PE in the explants analyzed in Fig. 6B, because when the peak response data in Fig. 6B and D were combined, the difference did not reach significance (F= 3.012, P = 0.086). In the TTX-treated explants, the peak response to ATP+PE was greater than that to PE, but not significantly different from that induced by ATP (H = 18.249. P < 0.001). The end ratio followed the same order as in the prior experiments in 2 mM Ca2+ without TTX treatment: ATP+PE>PE>ATP (H=28.984, P < 0.001). Thus TTX treatment did not change the nature of the sustained responses to ATP, PE and ATP+PE.
Calcium responses to short application of ATP, PE, and ATP+PE.
To determine whether the sustained elevation in the end ratio induced by ATP+PE required continued exposure to ATP+PE, the response to a short exposure to ATP, PE, or ATP+PE was evaluated. In this set of experiments, washout (3 ml/min) of drugs was initiated 1 min after the peak of the response. With a perifusion chamber volume of 1 ml, the concentration of the drugs would be reduced 20-fold within 1 min and reduced to 1 nM by 4 min. As seen in Fig. 6D, the end ratios were significantly lower in all groups than observed when the drugs were present throughout the experiment (compared to Fig. 6B: ATP, P < 0.001; PE, P < 0.001; ATP+PE, P = 0.004), but they still followed the same order as observed with continued exposure: ATP+PE>PE>ATP (H = 40.928, P < 0.001).
Source of the [Ca2+]i signal induced by ATP+PE.
To determine the source of the peak and sustained elevation in [Ca2+]i induced by ATP+PE, the responses to ATP+PE were examined after pretreatment of the fura-2-loaded preparation with TG (200 nM) for 30 min, in perifusion medium containing zero [Ca2+]o (plus 1 mM EGTA), or with both treatments (TG in zero [Ca2+]o; Fig. 7). Fig. 7A shows the time course of responses to ATP+PE of cells in a single preparation under these conditions. As shown in Fig. 7B, although removing external Ca2+ only caused a slight, nonsignificant decrease in the initial peak, it prevented the extended elevation in [Ca2+]i response. TG significantly reduced both the initial peak response and the sustained elevation in [Ca2+]i (Fig. 7B). The delay to reach the peak response after TG treatment (Fig. 7A) may reflect more efficient Ca2+ buffering as a result of TG treatment reducing basal [Ca2+]i. Finally, combined TG and zero [Ca2+]o completely eliminated both the peak response and the extended elevation in [Ca2+]i. Thus both Ca2+ influx and release of Ca2+ from internal stores contribute to the initial and the sustained Ca2+ responses to ATP+PE.
Although previous studies have used live cell Ca2+ imaging techniques to study the effect of glutamate, ATP, and neuropeptides on cytoplasmic [Ca2+]i in SON neurons (8, 13, 15, 18, 22–24), to identify the types of Ca2+ channels used (22), and to study the role of Ca2+ in regulating the ionic currents in SON neurons (20, 21, 34), this is the first study to measure changes in [Ca2+]i in response to ATP and an α1-adrenergic agonist in SON neurons maintained in an organotypic preparation. In this preparation, basal [Ca2+]i was similar to that reported previously in isolated dispersed SON neurons (8) and in individual neurons in slice preparations containing SON (21, 34). The occurrence of spontaneous fluctuations in [Ca2+]i was also similar to that observed in isolated SON neurons (8). These observations, in addition to the location, size, and responsiveness of a subpopulation of neurons to VP, supports the identification of the cells studied as SON neurons. In the preparations tested for responses to VP, 70% of the imaged neurons responded to VP. This probably reflects the selection of more ventrally oriented SON neurons for imaging, because VP neurons are preferentially located in the ventral portion of the nucleus. However, the 30% that were not VP responsive are probably oxytocinergic. Therefore, both OT and VP neurons are represented in the populations of neurons imaged. However, since the purpose of these studies was to examine the [Ca2+]i response to combined exposure to ATP and PE, and this treatment induced similar synergistic stimulation of both VP and OT release from HNS explants (17), we did not feel it was necessary to identify the VP or OT phenotype of all neurons in this study. In addition, virtually 100% of the cells responded to ATP and/or ATP+PE without evidence of distinct subpopulations suggesting similar effects of ATP and ATP+PE on VP and OT neurons in SON.
The observation that combined exposure to ATP+PE resulted in an extended elevation in [Ca2+]i that was significantly greater than that observed with either ATP or PE alone supports the hypothesis that the intracellular signaling cascades activated by ATP and PE converge to yield a greater sustained elevation in [Ca2+]i than is achieved with either agent independently. This was observed under several different conditions, including low (0.3 mM) and normal (2 mM) extracellular Ca2+ both with and without TTX. Although smaller, it was even observed several minutes later when washout of ATP and PE was initiated one min after the peak response, and in at least one preparation (Fig. 3), it remained slightly elevated following a 30-min washout. The sustained increase in [Ca2+]i induced by exposure to ATP+PE, although significantly greater than that observed with either agent alone, was not greater than the sum of the elevation observed to ATP or PE alone. Thus it probably reflects the additive effects of ATP+PE on [Ca2+]i. This and the observation that continued exposure to ATP+PE was not necessary to induce a sustained elevation in [Ca2+]i indicate that an altered [Ca2+]i signal persists that could initiate the events that lead to new gene transcription and synergistic stimulation of VP release.
Elevations in [Ca2+]i alter a wide array of cellular functions that might initiate events leading to the extended elevation in VP release observed in response to exposure to ATP+PE. Specifically, in VP neurons, elevated [Ca2+]i inactivates a Ca2+-sensitive K+-current (IK,leak) causing membrane depolarization and excitation (19). [Ca2+]i also regulates the activity of Ca2+-sensitive kinases such as PKC and Ca2+-calmodulin kinase. This could lead to alterations in K+-channel activation (19), receptor trafficking, or gene expression by phosphorylation of transcription factors. The latter possibility is supported by the prior finding by this laboratory that actinomycin treatment blocked the extended elevation in VP release, which indicates that gene transcription is required for the response (17). Thus it is possible that the larger extended elevation in [Ca2+]i although not in itself synergistic, initiates cellular processes, including new protein synthesis that convert a transient response into a response that, by virtue of its extended nature, is clearly synergistic (17). Future experiments in preparations that allow precise calibration of the fura signals to [Ca2+]i (18, 21) will be useful for quantitatively assessing the potential for the observed changes in Ca2+ to alter channel or kinase activity.
A role for [Ca2+]i in driving the synergistic response is further supported by experiments that have eliminated several other possible mechanisms. Alterations in ATP metabolism to adenosine is not likely to participate, because inhibition of ATP conversion to adenosine does not convert the transient response to ATP to a sustained response (32). Modification of stimulus secretion coupling at the nerve terminal was eliminated in experiments using isolated neural lobes, and involvement of glutamatergic afferents was eliminated in experiments demonstrating that neither ionotropic nor metabotropic excitatory amino acid receptor antagonists block the synergistic stimulation of VP release by ATP+PE (33).
The source of Ca2+ for the extended elevation in [Ca2+]i in response to ATP+PE was evaluated by depleting internal Ca2+ stores with TG and by eliminating Ca2+ from the perifusate. Since P2X receptors are expressed in SON neurons (24), and our prior studies with the P2X receptor antagonist, PPADS, implicated P2X receptors in the synergistic response to ATP+PE (17), influx of Ca2+ through these nonselective cation channels is expected to contribute to the response. The requirement for an influx of external Ca2+ was confirmed by the absence of an extended elevation in [Ca2+]i when ATP+PE was applied in zero [Ca2+]o. Activation of α1-adrenergic receptors induces production of inositol triphosphate resulting in release of Ca2+ from intracellular stores. The sustained elevation in [Ca2+]i was eliminated by depleting the intracellular stores with TG, and the importance of intracellular Ca2+ stores to the synergistic response was demonstrated by the ability of TG to block the extended increase in VP release in response to ATP+PE. Thus the additive increase in [Ca2+]i after exposure to ATP+PE may reflect summation of these two sources of intracellular Ca2+. ATP and PE also depolarize SON neurons initiating action potential firing (1, 16). This will also initiate influx of Ca2+ through the high voltage-activated Ca2+ channels (L-, N-, and P-type) that have been functionally identified in SON neurons (10, 11) and have been shown to be primarily responsible for the rapid increases in [Ca2+]i in response to action potentials (21). The sphere of influence of these respective sources of Ca2+ may determine their relative contributions to induction of the extended increase in VP release. In elegant modeling studies, Roper et al. (21) have provided evidence supporting roles for independent microdomains of [Ca2+]i on Ca2+-sensitive potassium channels in SON neurons, and Li and Hatton (20) demonstrated that internal Ca2+ release induced by Ca2+ influx is essential for generation of the phasic pattern of electrical activity, which most efficiently drives VP release (2). In a similar fashion, the location of the kinases and transcription factors that are the postulated triggers for the synergistic hormone release induced by ATP+PE relative to the sources of increases in [Ca2+]i (e.g., P2X receptors, endoplasmic reticulum, and voltage-gated Ca2+ channels) will determine the relative influence of each respective Ca2+ source on ATP+PE-induced synergistic release of VP.
This study was supported by National Institutes of Health Grant RO1-NS27975 to C. Sladek. We thank Adam Zweifach for his help in developing these techniques.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society