INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

THE CALCIUM-SENSING RECEPTOR (CaR) was hypothesized to exist based on the sensitivity of parathyroid hormone secretion to serum calcium (5, 24). Expression cloning utilizing RNA isolated from parathyroid cells resulted in identification of the CaR (6), which is activated by extracellular Ca²⁺ in the physiological range (0.5-5 mM). A unique feature of CaR is the steep cooperativity of Ca²⁺-dependent activation, which is observed both in vivo (26) and when CaR is heterologously expressed (6, 15). In the parathyroid, CaR is chronically exposed to serum Ca²⁺ concentrations that are within the steeply cooperative range of its dose-response relation (EC₅₀ of 2-3 mM, and Hill coefficient n  2-3), providing tight control over parathyroid hormone secretion. Although these qualitative details of CaR-mediated modulation of parathyroid hormone secretion are well characterized, the kinetics of single-cell responses to tonic, physiological changes in extracellular Ca²⁺ have not been studied.

CaR couples changes in extracellular Ca²⁺ to a variety of intracellular responses, including activation of phospholipases, generation of inositol trisphosphate (IP₃) and diacylglycerol, increases in intracellular Ca²⁺, changes in protein phosphorylation, activation of ion channels, regulation of hormone secretion, and modulation of gene expression (32). Chronic activation of CaR resulting from constant exposure to serum Ca²⁺ would necessarily contribute to regulation of cellular functions through many of these pathways. Under conditions of constant agonist exposure, however, many G protein-coupled receptors undergo desensitization and, ultimately, downregulation of receptor and/or signal transduction pathway protein expression (7, 13). A key question with respect to CaR function is therefore whether CaR desensitizes at a fixed extracellular Ca²⁺ concentration and responds only to acute changes in serum Ca²⁺ or whether CaR is chronically activated under physiological conditions.

We have utilized heterologous expression of human CaR in HEK-293 cells to address two fundamental questions with respect to the in vivo contributions of CaR to parathyroid cell function, namely, How do cells expressing CaR adapt to chronic exposure to the concentrations of extracellular Ca²⁺ that are present in normal serum? and What are the kinetic features of CaR-mediated changes in intracellular Ca²⁺ in response to small, acute perturbations of extracellular Ca²⁺? We find that cells expressing CaR are uniquely sensitive to alterations in extracellular Ca²⁺ and that small perturbations in extracellular Ca²⁺ induce low-frequency intracellular Ca²⁺ oscillations. These results have implications for the mechanism(s) by which CaR activation regulates parathyroid hormone secretion as well as diverse cellular signal transduction pathways in many other cell types.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Cell culture and transfection. HEK-293 cells (American Type Culture Collection, Rockville, MD) were grown in high-glucose DMEM (Life Technologies, Grand Island, NY) supplemented with 10% BSA (Sigma, St. Louis, MO), 50 U/ml penicillin, and 50 µg/ml streptomycin (37°C, 5% CO₂). Cells were transiently or stably transfected with the human CaR (obtained from Dr. Klaus Seuwen, Novartis Pharma, Basel, Switzerland), which was subcloned into pEGFPN3 (Clontech Laboratories, Palo Alto, CA) as previously described (15). HEK-293 cells were transfected with 1 µg of each CaR construct using Effectene (Quiagen, Valencia, CA) or Novafector (VennNova, Pompano, FL), plated on collagen-coated coverslips, and kept in a 5% CO₂ incubator at 37°C until use (generally 72 h after transfection). HEK-293 cells stably expressing human CaR were produced by linearization of the hCaR-pEGFPN3 plasmid, transfected as described above, and selected with Geneticin (Life Technologies). Colonies were selected after 3 wk, and clones were confirmed by sequencing, CaR activity, and Western blotting with anti-green fluorescent protein (GFP) antibody (Molecular Probes, Eugene, OR).

Single-cell fluorescence measurements of intracellular Ca²+. Cells were loaded with 0.5 µM fura 2-AM (Calbiochem-Novabiochem, San Diego, CA) for 30 min at 37°C in a solution containing variable CaCl₂ (0.5, 2, or 3 mM), 130 mM NaCl, 5 mM KCl, 1 mM MgSO₄, 20 mM HEPES, 0.83 mM Na₂HPO₄, 0.17 mM NaH₂PO₄, 25 mM mannose, and 1 mg/ml BSA, pH 7.4. After the loading period, the coverslip was mounted in an imaging chamber (Warner Instrument, Hamden, CT) having both a top and bottom coverslip and washed with a bath solution having a CaCl₂ concentration matched to the loading solution plus 140 mM NaCl, 5 mM KCl, 0.55 mM MgCl₂, 10 mM HEPES, pH 7.4. Selected regions of the field were excited at 340/380 nm (emission wavelength 510 nm) at 8- or 10-s intervals (10 frame averaging) on a Universal imaging system (Universal Imaging, West Chester, PA) based on the MetaFluor software package. Background images at the same settings used during a particular experiment were obtained on regions of the coverslip devoid of cells. Calibration of intracellular Ca²⁺ utilized a series of buffered Ca²⁺ standards (Molecular Probes), assuming a dissociation constant of 145 nM (determined in vitro at 22°C; Molecular Probes Handbook, 7th edition). All solutions were osmolality matched (290-300 mosM), measured on a Wescor 5500 vapor pressure osmometer (Wescor, Salt Lake City, UT). Variations in extracellular Ca²⁺ concentration were produced by isosmolar substitution for NaCl. Thapsigargin and the protein kinase inhibitors bisindolylmaleimide I, KN-93, and staurosporine (Calbiochem-Novabiochem) were prepared as stock solutions in DMSO and added to the appropriate extracellular solution. DMSO controls were performed under identical experimental conditions and had no effect on measured parameters. All experiments were performed at room temperature (22-24°C). Multiple cells were analyzed from at least three independent transfections or cell passages (for stably transfected cell lines). Data obtained from stably and transiently transfected HEK-293 cells were comparable, and each result was verified for both conditions. Data were normalized/averaged as described in the text and are presented as means ± SE. Curves were fitted by least-squares minimization using the Marquardt-Levenberg algorithm (NFIT; Island Products, Galveston, TX). Statistical significance was determined by unpaired t-test (SigmaPlot 2000 for Windows, version 6), with significance ascribed at P < 0.01.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Contribution(s) of CaR expression to resting intracellular Ca²+ concentration. Activation of CaR is steeply dependent on extracellular Ca²⁺. Figure 1A illustrates a typical experiment in which extracellular Ca²⁺ is progressively increased from a baseline value of 0.5 to 30 mM, and intracellular Ca²⁺ is monitored with fura 2 fluorescence. Data compiled from 20 cells in this representative experiment were fitted with the Hill equation (Fig. 1B) to obtain the EC₅₀ for extracellular Ca²⁺ (3.4 ± 0.4 mM) and the Hill coefficient (1.8 ± 0.4), as we have previously reported (15). These results suggest that exposure of cells expressing CaR to normal serum-containing media (Ca²⁺ of ~1.8-2 mM) should partially activate the receptor.

View larger version (26K): [in this window] [in a new window]   Fig. 1.   Dose-dependent increase in intracellular Ca2+ on Ca2+-sensing receptor (CaR) activation. A: HEK-293 cells expressing CaR-green fluorescent protein (GFP) were exposed to incremental increases in extracellular Ca2+, from a baseline of 0.5 mM. Responses from 20 cells (randomly chosen, GFP positive) are illustrated. B: peak responses at each extracellular Ca2+ concentration for cells in A were averaged and fitted with the Hill equation (fit parameters: maximum intracellular Ca2+ = 153 ± 10.7 nM; EC50 = 3.39 ± 0.4 nM; n = 1.8 ± 0.4).

View larger version (33K): [in this window] [in a new window]   Fig. 2.   Steady-state intracellular Ca2+ in cells expressing CaR. A: HEK-293 cells transiently expressing CaR-GFP were loaded with fura 2-AM in various Ca2+ concentrations (0.5, 2, and 3 mM), and experiments were begun in the same Ca2+ concentration. Intracellular Ca2+ concentrations were measured during the first minute of recording. Data represent average of >60 cells/condition from at least 3 independent transfections. Significance was determined relative to 0.5 mM at P < 0.01 (*) or 2 mM at P < 0.01 (**). B: untransfected HEK-293 cells were subjected to the same protocol as in A, with significance determined relative to 0.5 mM at P < 0.01 (*). C: 20 cells were exposed to 2 mM extracellular Ca2+, then to 0.5 mM, and then returned to 2 mM. Average intracellular Ca2+ at steady state in each condition was measured. Significance is relative to 2 mM at P < 0.01 (*). Data are representative of 3 independent experiments. D: protocol similar to that of C performed in untransfected HEK-293 cells.

Contributions of CaR expression to Ca²+ content of thapsigargin-sensitive stores. Expression of CaR has a significant effect on resting intracellular Ca²⁺ concentrations when cells are exposed to extracellular Ca²⁺ concentrations in the physiological range, i.e., 0.5-3 mM (Fig. 2). A second compartment of intracellular Ca²⁺ that might be affected by CaR expression (and chronic activation) is the thapsigargin-sensitive store, which releases Ca²⁺ on CaR activation (32). To assess the content of thapsigargin-sensitive stores after exposure of CaR-expressing HEK-293 cells to 0.5, 2, or 3 mM extracellular Ca²⁺, experiments were begun in the loading Ca²⁺ concentration to establish the resting intracellular Ca²⁺ level, switched to a nominally zero Ca²⁺ bath solution, and then exposed to 1 µM thapsigargin in the continued absence of bath Ca²⁺. Figure 3, A-C, illustrates this type of experiment in HEK-293 cells preincubated in 0.5, 2, and 3 mM Ca²⁺, respectively, and Fig. 3, D-F, illustrates the results obtained with CaR-expressing cells.

View larger version (38K): [in this window] [in a new window]   Fig. 3.   Thapsigargin-sensitive Ca2+ stores in cells expressing CaR. Intracellular Ca2+ was monitored in cells loaded with fura 2-AM in 0.5 (A, D), 2 (B, E), or 3 (C, F) mM extracellular Ca2+. Experiments were begun in the same Ca2+ concentration. Extracellular Ca2+ was reduced to nominally zero, and then cells were exposed to 1 µM thapsigargin. The entire time course of the thapsigargin-induced intracellular Ca2+ transient was monitored. Experiments illustrate the results from 20 untransfected HEK-293 cells (A-C) or HEK-293 cells stably transfected with human CaR (D-F). Experiments are representative of 3-5 independent experiments under each condition.

View larger version (32K): [in this window] [in a new window]   Fig. 4.   Content but not kinetics of thapsigargin-sensitive Ca2+ stores is altered by CaR expression. A: experiments such as those illustrated in Fig. 3 were used to estimate size of the thapsigargin-sensitive Ca2+ stores. Averages of the thapsigargin-induced intracellular Ca2+ transients for 3-4 independent experiments from each condition (60-80 cells) were transferred to Adobe Photoshop, and the pixel areas [in arbitrary units (AU)] under the curves were calculated. There was a significant difference in the average intracellular store content between HEK-293 cells and CaR-expressing cells (*P < 0.01). B: half times for the rising (tR) or declining (tD) phases of the thapsigargin-induced intracellular Ca2+ transients for the cells analyzed in A. There were no significant differences between HEK-293 cells and CaR-expressing cells.

Small changes in extracellular Ca²+ induce intracellular Ca²⁺ oscillations. While decreases in extracellular Ca²⁺ over the range from 3 to 0.5 mM cause CaR-mediated changes in the steady-state concentration of intracellular Ca²⁺ (Figs. 2 and 3), small incremental increases in extracellular Ca²⁺ (0.5 mM) induce oscillations of intracellular Ca²⁺. Illustrated in Fig. 5 is a typical experiment, in which HEK-293 cells transiently transfected with CaR were loaded with fura 2-AM and the experiment was begun in 2 mM extracellular Ca²⁺. Increases (0.5 mM increments) in extracellular Ca²⁺ from 2 to 3.5 mM initiate oscillatory intracellular Ca²⁺ changes, which are most pronounced at 3.5 mM extracellular Ca²⁺. The oscillatory change in intracellular Ca²⁺ ceased abruptly when extracellular Ca²⁺ was decreased to 2 mM. At the end of the experiment, the cells were exposed to a concentration of extracellular Ca²⁺ that maximally activates CaR (20 mM). This concentration of extracellular Ca²⁺ generated a rapidly activating peak of intracellular Ca²⁺ (up to 200 nM), which decayed to an elevated plateau for the duration of agonist exposure. This representative experiment illustrates the range of responses to alterations in extracellular Ca²⁺, from oscillatory changes in intracellular Ca²⁺ at physiological levels of extracellular Ca²⁺ to a maximal response that is typical of many G protein-coupled receptors that activate the phosphatidylinositol-phospholipase C (PI-PLC) pathway and increase intracellular Ca²⁺.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Small increases in extracellular Ca2+ induce intracellular Ca2+ oscillations. HEK-293 cells transiently transfected with CaR-GFP were loaded and the experiment was initiated in 2 mM extracellular Ca2+. Extracellular Ca2+ was then increased in 0.5 mM increments up to 3.5 mM. Reducing extracellular Ca2+ caused an immediate cessation of oscillations. The experiment was ended by a step change in extracellular Ca2+ to 20 mM, eliciting peak intracellular Ca2+ release, which decayed to an elevated plateau.

The threshold for CaR-mediated intracellular Ca²+ oscillations. The experiment illustrated in Fig. 5 suggests that the threshold for intracellular Ca²⁺ oscillations is in the range of 2.5-3 mM extracellular Ca²⁺. To determine whether this threshold is influenced by the starting extracellular Ca²⁺ concentration, HEK-293 cells expressing CaR were loaded with fura 2-AM at fixed Ca²⁺ concentrations (0.5, 2, or 3 mM), and the subsequent experiment was begun in the same extracellular Ca²⁺ concentration. Extracellular Ca²⁺ was then increased in 0.5 mM increments. Representative experiments begun in 0.5, 2, or 3 mM extracellular Ca²⁺ are illustrated in Fig. 6, A, B, and C, respectively. Inspection of the plots reveals that cells do not oscillate in the extracellular Ca²⁺ concentration in which they have been loaded with fura 2-AM (although there are differences in resting intracellular Ca²⁺ concentrations, as documented in Figs. 2 and 3). Cells loaded in 2 or 3 mM extracellular Ca²⁺ begin to oscillate after the first 0.5 mM increment in extracellular Ca²⁺, with peak numbers of cells oscillating after a 1 mM change from the initial Ca²⁺ concentration. Cells loaded in 0.5 mM Ca²⁺ do not oscillate until extracellular Ca²⁺ reaches 2 mM, with additional recruitment of cells at concentrations up to 5 mM, the highest concentration tested. The extracellular Ca²⁺ concentration dependence of oscillation threshold, characterized as the percent of cells oscillating in a given condition, was determined for three independent experiments initiated at 0.5, 2, and 3 mM Ca²⁺, and the results are illustrated in Fig. 6D.

View larger version (43K): [in this window] [in a new window]   Fig. 6.   Threshold for intracellular Ca2+ oscillations. A-C: HEK-293 cells transiently transfected with CaR-GFP were loaded and experiments initiated in 0.5, 2, and 3 mM Ca2+, respectively, and then exposed to 0.5 mM incremental increases in extracellular Ca2+. Experiments were ended with exposure to maximally activating concentrations of extracellular Ca2+ (10-20 mM). D: threshold for intracellular Ca2+ oscillations as a function of the initial extracellular Ca2+ concentration. A total of 60 cells under each condition (20 cells/experiment, randomly chosen from GFP-positive cells) were analyzed. E: peak to peak intervals were measured for the cells illustrated in A-C and plotted relative to the tested extracellular Ca2+ concentrations. Bars: solid, loaded in 0.5 mM Ca2+; hatched, loaded in 2 mM; shaded, loaded in 3 mM. The dashed line indicates the average of all data, independent of loading or test Ca2+ concentration.

Oscillation frequency is constant. The dependence of the oscillation frequency on extracellular Ca²⁺, characterized by the peak to peak duration, was analyzed for the experiments in Fig. 6, A-C, plotted in Fig. 6E. Regardless of the preconditioning extracellular Ca²⁺, the peak to peak intervals of the oscillations elicited over the range of extracellular Ca²⁺ concentrations from 2.5 to 5 mM were comparable; the average of all conditions was 46.8 s (indicated by the dashed line in Fig. 6E).

Dependence of oscillations on intracellular Ca²+. Oscillations induced by small changes in extracellular Ca²⁺ were sustained for periods up to 30 min. An example is illustrated in Fig. 7. In this experiment, the oscillations were initiated by a step change in extracellular Ca²⁺ from 2 to 3 mM, and intracellular Ca²⁺ was monitored until oscillations dissipated. The insets in Fig. 7 illustrate the behavior of cells early (from 8 to 16 min) and late (from 33 to 42 min) in the exposure to 3 mM Ca²⁺. This experiment is representative of 12 experiments [3 independent transient transfections (n = 3 experiments each) plus 3 experiments in cells stably expressing CaR]. In all experiments, oscillatory behavior ceased between 20 and 40 min after initiation by a 1 mM increase in extracellular Ca²⁺. Despite the cessation of oscillations, the cells exhibited an elevated intracellular Ca²⁺ in the presence of 3 mM extracellular Ca²⁺, which decreased on return to 2 mM Ca²⁺, e.g., Fig. 7.

View larger version (31K): [in this window] [in a new window]   Fig. 7.   Intracellular Ca2+ oscillations are sustained for 30-40 min. Oscillations were initiated in cells stably expressing CaR by a step from 2 to 3 mM extracellular Ca2+ and followed until oscillations stopped. The experiment was ended by exposure to 10 mM Ca2+ to assess the responsiveness of all cells. Insets: the oscillatory behavior of the cells at early (500-1,000 s) and late (2,000-2,500 s) times in the experiment, depicting both the decrease in oscillating cells and the rise in baseline intracellular Ca2+. Representative of >12 experiments in stably or transiently CaR-transfected HEK-293 cells.

View larger version (37K): [in this window] [in a new window]   Fig. 8.   Dependence of oscillations on intracellular Ca2+. A: cells transiently expressing CaR-GFP were twice exposed to 3.5 mM extracellular Ca2+ from a baseline Ca2+ of 2 mM. B: oscillation intervals were calculated for the first and second exposure to 3.5 mM Ca2+ and were not significantly different. C: peak intracellular Ca2+ was estimated during the first and second exposures to 3.5 mM Ca2+, and baseline Ca2+ was determined before, between, and after the 2 exposures to 3.5 mM Ca2+. In both cases, there was a significant decrease in Ca2+ (*P < 0.01).

View larger version (28K): [in this window] [in a new window]   Fig. 9.   Effect of thapsigargin on intracellular Ca2+ oscillations. A: control experiment in which intracellular Ca2+ oscillations were initiated by an increase in extracellular Ca2+ from 2 to 3.5 mM. Experiment was ended by exposure of the cells to 10 mM Ca2+. B: experiment was begun as in A, but 1 µM thapsigargin was added after 4 min in 3.5 mM Ca2+. Experiment was ended by exposure of cells to 10 mM Ca2+. Note the reduced response to 10 mM Ca2+, indicating compromised thapsigargin-sensitive intracellular stores.

Regulation of intracellular Ca²+ oscillations. Intracellular Ca²⁺ oscillations are observed in many cell types, elicited by a wide variety of G protein-coupled receptors that activate the PI-PLC pathway, ultimately causing release of Ca²⁺ from internal stores by activating intracellular IP₃ receptors, and often inducing influx of Ca²⁺ from the extracellular medium (16). For CaR-activated oscillations, the data in Fig. 9 indicate an integral contribution of Ca²⁺ cycling into and out of intracellular stores. A second widely observed mechanism regulating G protein-coupled receptor activation is cyclical phosphorylation-dephosphorylation, which has been shown to be responsible for mGluR5-mediated intracellular Ca²⁺ oscillations in neurons (14, 22) and in heterologous expression systems (19). The effects of several protein kinase inhibitors were therefore tested for their ability to regulate CaR-mediated Ca²⁺ oscillations. Figure 10 illustrates the results of typical experiments with staurosporine, a broad-specificity protein kinase inhibitor that was applied at a concentration (100 nM) that should inhibit protein kinases A (IC₅₀  15 nM), C (IC₅₀  7 nM), G (IC₅₀  8.5 nM), myosin light chain kinase (IC₅₀  1.3 nM), and calmodulin kinase (IC₅₀  20 nM). In the control experiment, intracellular Ca²⁺ oscillations were elicited by a step change in extracellular Ca²⁺ from 2 to 3.5 mM and were sustained until a maximal response was elicited by 10 mM extracellular Ca²⁺ at the end of the experiment (Fig. 10A). The companion experiment utilized the same 2.0-3.5 mM step in extracellular Ca²⁺ to elicit oscillations, and then 100 nM staurosporine was applied during the oscillations (Fig. 10B). In this and all other experiments of similar protocol, there was no effect on oscillation frequency on addition of staurosporine, although, in some experiments, including that illustrated in Fig. 10B, there was a decrease in the maximum intracellular Ca²⁺ response during oscillations. To determine whether the lack of effect of staurosporine on oscillation frequency was due to slow diffusion of the blocker into the cells, a second protocol was used to assess the effect(s) of the protein kinase inhibitor (Fig. 10C): oscillations were elicited by a step from 2 to 3.5 mM Ca²⁺, followed by return to 2 mM Ca²⁺, application of 100 nM staurosporine for 2 min, and then a second step to 3.5 mM Ca²⁺ in the continued presence of staurosporine. As with the protocol of Fig. 10B, there was no effect of staurosporine on oscillation frequency despite the 2-min preexposure to the blocker. Experiments similar to that illustrated in Fig. 10B were also performed with 1 µM bisindolylmaleimide I and 5 µM KN-93, inhibitors of protein kinase C (IC₅₀  8-200 nM, depending on isoform) and calmodulin kinase II (IC₅₀  370 nM) , respectively. Although the protein kinase inhibitors did induce a decrease in baseline intracellular Ca²⁺ and the maximal peak Ca²⁺ of the oscillations in some experiments, none of the inhibitors blocked the oscillations or affected the peak to peak interval of the oscillations (Fig. 10D). These results suggest that protein kinases are not required for initiation or maintenance of intracellular Ca²⁺ oscillations elicited by CaR activation.

View larger version (34K): [in this window] [in a new window]   Fig. 10.   Effect of protein kinase inhibitors on intracellular Ca2+ oscillations. A: control experiment, as in Fig. 9A. B: effect of staurosporine on oscillations induced by 3.5 mM Ca2+. Cells were exposed to 3.5 mM Ca2+ for 4 min to permit determination of oscillation interval, then 100 nM staurosporine was added in the continued presence of 3.5 mM Ca2+. C: two-pulse protocol, similar to that illustrated in Fig. 8A. Immediately on washout of the first exposure to 3.5 mM, cells were exposed to 100 nM staurosporine, and then, after 4 min, cells were tested with a second exposure to 3.5 mM Ca2+ (in the continued presence of staurosporine). D: summary of protein kinase inhibitor studies on oscillation frequency. Data from 3 independent experiments (>60 cells) utilizing the protocol illustrated in B were averaged for each inhibitor. Solid bars, no blocker; open bars, protein kinase inhibitor. E: experiments such as those in Fig. 6, A-C, were performed in the continuous presence of 100 nM staurosporine, and the percent of oscillating cells was determined at each Ca2+ concentration >3 mM. Plotted are the average percent oscillating cells for 60 cells from 3 independent experiments. The solid line is control data (absence of staurosporine) redrawn from Fig. 6D. BIM, bisindolylmaleimide I.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

This report delineates the complex intracellular Ca²⁺ responses resulting from activation of CaR by extracellular Ca²⁺. Cells expressing CaR exhibit both modulation of resting intracellular Ca²⁺ concentrations at extracellular Ca²⁺ concentrations below the EC₅₀ for CaR activation and a decrease in net thapsigargin-sensitive Ca²⁺ stores. Alterations in the resting intracellular Ca²⁺ concentration may generate low-level activation of Ca²⁺-sensitive signaling pathways, in particular those with a high affinity for Ca²⁺. In addition, the low level of CaR activation under these conditions may "sensitize" cells expressing CaR to activation of other G protein-coupled receptors that also utilize the PI-PLC pathway for cellular signaling. This sensitizing effect has recently been seen in monocytes coactivated by extracellular Ca²⁺ and the chemokine MCP-1 (25).

Increasing extracellular Ca²⁺ by as little as 0.5 mM in the range around the EC₅₀ for CaR activation generates low-frequency (40- to 50-s peak to peak interval) intracellular Ca²⁺ oscillations. The oscillation interval is independent of the extracellular Ca²⁺ concentration, although the number of cells responding to the perturbation of extracellular Ca²⁺ peaks at ~1 mM above the ambient extracellular Ca²⁺. Cells expressing CaR acclimate to the ambient concentration of extracellular Ca²⁺ with a cessation of oscillations after 30-40 min (Fig. 7). The threshold for initiation of intracellular Ca²⁺ oscillations is thus reset by the concentration of extracellular Ca²⁺ to which the cells are chronically exposed. This "resetting" of the threshold for initiation of oscillations is not due to alterations in the Ca²⁺ content of intracellular stores (Fig. 4). In parathyroid cells in vivo, increases in extracellular Ca²⁺ >1.8 mM induce oscillations with an interval of 42 s (21). Our results suggest that this oscillatory behavior (oscillation interval, threshold) can be conferred on a heterologous cell type by expression of CaR and therefore must be inherent in either CaR or highly conserved elements of the PI-PLC signaling pathway.

Many G protein-coupled receptors that activate the PI-PLC pathway generate intracellular Ca²⁺ oscillations over a limited agonist concentration range. Oscillation frequencies range from very rapid "pacemaker" activity (1 Hz) that signifies elemental Ca²⁺ release events (4) to intermediate (3-4 min¹) (2), to slow (<1 min¹) (12). Agonist concentration-dependent alterations in oscillation frequency have also been noted (10, 12, 23, 27, 28). While we observed relatively slow intracellular Ca²⁺ oscillations, there was no systematic increase in oscillation frequency over the limited range of extracellular Ca²⁺ (agonist) concentrations tested. It should be noted that CaR is unique in that its agonist is Ca²⁺; therefore, increases in extracellular Ca²⁺ mediate receptor activation and increase the driving force for Ca²⁺ influx through a variety of permeability pathways. This may limit the range of agonist concentrations that can be explored, since increasing extracellular Ca²⁺ from 2 to 5 mM is sufficient to induce a peak Ca²⁺ response, which decays to a sustained plateau, largely eliminating intracellular Ca²⁺ oscillations. Nevertheless, small increases in extracellular Ca²⁺ "recruit" cells to oscillate, and the net effect at the level of the parathyroid may be to increase the fraction of cells in which parathyroid hormone secretion is modulated.

A number of mechanisms have been proposed for initiation and maintenance of intracellular Ca²⁺ oscillations, most requiring cyclical alterations in the state of the Ca²⁺ release system (16), generally involving fluctuations in either the cellular levels of IP₃ or the activity of IP₃ receptors in the endoplasmic reticulum. Thapsigargin rapidly disrupts CaR-initiated intracellular Ca²⁺ oscillations, implying that intracellular stores are intimately involved in their initiation and maintenance. In systems in which the intracellular Ca²⁺ oscillations require plasma membrane level Ca²⁺ influx, thapsigargin has no effect (30). We cannot exclude a contribution of Ca²⁺ influx, since we cannot eliminate Ca²⁺ from the extracellular solution and maintain CaR activation. We attempted to address the question of Ca²⁺ influx during oscillations by utilizing Mn²⁺ uptake and quench (29), but found that this approach was not feasible since Mn²⁺ acted as an agonist for CaR in the range required for quench of fluorescence (data not shown). With respect to receptors having the closest homology with CaR, namely group I metabotropic glutamate receptors, mGluR5-induced intracellular Ca²⁺ oscillations have been noted both in vivo (14, 22) and in vitro (19) and result from the cyclical protein kinase C-mediated phosphorylation and dephosphorylation of the receptor itself, at a threonine residue within the carboxy terminus (19). Sequence alignment of CaR with mGluR5 suggests that the critical threonine of mGluR5 is present in CaR, although it is not part of a consensus sequence for protein kinase C phosphorylation in CaR (1, 8). Furthermore, we have generated truncations of the carboxy terminus in which all of the consensus sites for protein kinases A and C phosphorylation were removed (15). These truncated forms of human CaR were also observed to oscillate (data not shown).

In this report, we have tested the effects of a variety of protein kinase inhibitors, with a particular focus on those that are regulated by Ca²⁺. Neither staurosporine, bisindolylmaleimide I, nor KN-93 altered the frequency of oscillations elicited by extracellular Ca²⁺. Protein kinase C-mediated phosphorylation of CaR causes a reduction in Ca²⁺ affinity of the receptor (1, 8, 26). For the cholecystokinin receptor, a similar protein kinase C-mediated decrease in agonist affinity of the cholecystokinin 8 receptor is responsible for the cessation of intracellular Ca²⁺ oscillations (31). It is therefore possible that protein kinase C-mediated phosphorylation of CaR at low agonist concentrations is responsible for the shift in threshold for initiation of oscillations observed in the "Ca²⁺ pretreatment" protocol, since staurosporine, a broad-specificity protein kinase inhibitor, increased the number of cells oscillating at each extracellular Ca²⁺ concentration (Fig. 10E). Further work on protein kinase C-specific site mutants of CaR will be needed to explicitly address this possibility.

Parathyroid hormone secretion is inversely controlled by extracellular Ca²⁺, i.e., increases in extracellular Ca²⁺ result in decreases in parathyroid hormone secretion (32). Despite intensive efforts, a fundamental question with regard to CaR regulation of parathyroid hormone secretion remains the locus of the inversion signal, given the observation that CaR activation mediates an increase in intracellular Ca²⁺. A confocal study that followed subcellular changes in intracellular Ca²⁺ on a step change in extracellular Ca²⁺ from 0.5 to 2 mM noted significant time-dependent inhomogeneities in intracellular Ca²⁺ (9). In this study, we have shown that global intracellular Ca²⁺ also exhibits oscillations over this concentration range. It is possible that the global and local oscillatory changes in intracellular Ca²⁺ contribute to the inversion of the secretory response. At a minimum, any model for CaR-mediated regulation of parathyroid hormone secretion must incorporate intracellular Ca²⁺ oscillations.

CaR contributes to the regulation of diverse physiological functions, and there may be unique implications of CaR-mediated intracellular Ca²⁺ oscillations that depend on cell type. It is interesting to note, however, that intracellular Ca²⁺ oscillations have been recognized recently as an extremely potent and specific means of increasing gene expression (11, 17, 20). While direct contributions of CaR activation to changes in gene expression have not been explicitly determined in most systems, there is evidence in the Caco-2 intestinal epithelial cell line for rapid, CaR-mediated increases in c-myc expression (18). In addition, CaR activation potently mediates keratinocyte differentiation (3). Our results suggest that another important locus for CaR-mediated signaling that should be explored is the dynamic regulation of gene expression via intracellular Ca²⁺ oscillations.