Amino Acid-stimulated Ca2+ Oscillations Produced by the Ca2+-sensing Receptor Are Mediated by a Phospholipase C/Inositol 1,4,5-Trisphosphate-independent Pathway That Requires G12, Rho, Filamin-A, and the Actin Cytoskeleton*

The G protein-coupled Ca2+-sensing receptor (CaR) is an allosteric protein that responds to two different agonists, Ca2+ and aromatic amino acids, with the production of sinusoidal or transient oscillations in intracellular Ca2+ concentration ([Ca2+]i). Here, we examined whether these differing patterns of [Ca2+]i oscillations produced by the CaR are mediated by separate signal transduction pathways. Using real time imaging of changes in phosphatidylinositol 4,5-biphosphate hydrolysis and generation of inositol 1,4,5-trisphosphate in single cells, we found that stimulation of CaR by an increase in the extracellular Ca2+ concentration ([Ca2+]o) leads to periodic synthesis of inositol 1,4,5-trisphosphate, whereas l-phenylalanine stimulation of the CaR does not induce any detectable change in the level this second messenger. Furthermore, we identified a novel pathway that mediates transient [Ca2+]i oscillations produced by the CaR in response to l-phenylalanine, which requires the organization of the actin cytoskeleton and involves the small GTPase Rho, heterotrimeric proteins of the G12 subfamily, the C-terminal region of the CaR, and the scaffolding protein filamin-A. Our model envisages that Ca2+ or amino acids stabilize unique CaR conformations that favor coupling to different G proteins and subsequent activation of distinct downstream signaling pathways.

leads to periodic synthesis of Ins(1,4,5)P 3 , whereas L-phenylalanine stimulation of the CaR does not induce any detectable changes in the level of this second messenger. Furthermore, we identified a novel pathway that mediates transient [Ca 2ϩ ] i oscillations produced by the CaR in response to amino acids, which requires the organization of the actin cytoskeleton and involves the small GTPase Rho, heterotrimeric proteins of the G 12 subfamily, the C-terminal distal region of the receptor, and the scaffolding protein filamin-A. Our model envisages that Ca 2ϩ or amino acids induce distinct conformational states of the CaR, providing for the possibility of differential coupling to downstream signaling pathways.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfections-Human Embryonic Kidney (HEK-293) cells and murine embryonic fibroblasts (MEF), generated from doubled knock-out mice for rhodopsin kinase (RK) or G 12/13 Ϫ/Ϫ were maintained as described previously (11,13). Primary renal proximal tubule epithelial cells (RPTEC) of human origin were obtained and maintained as suggested by Clonetics, CA. A cell line constitutively expressing the CaR was established by transfecting HEK-293 with an expression vector encoding the human CaR (pCR3.1-CaR), kindly provided by Dr. Allen Spiegel, NIDDK, National Institutes of Health. Transfections of the different plasmids were performed using Lipofectin or Lipofectamine Plus (Invitrogen) as described previously (14). Analysis of the cells transiently transfected were performed 24 or 72 (in vitro kinase assay) h posttransfection.
Single Cell [Ca 2ϩ ] i Imaging- [Ca 2ϩ ] i was measured in cells loaded with the calcium indicator fura-2 as described previously (11). Briefly, cells were incubated in saline solution (Hanks' balanced salt solution (Invitrogen) without phenol red, containing with 138 mM NaCl, 4 mM NaHCO 3 , 0.3 mM Na 2 HPO 4 , 5 mM KCl, 0.3 mM KH 2 PO 4 , 1.5 mM CaCl 2 , 0.5 mM MgCl 2 , 0.4 mM MgSO 4 , 5.6 mM D-glucose, 20 mM HEPES, pH 7.4) containing 5 M fura-2/AM for 45-60 min at 37°C. The cells were then washed and placed in an experimental chamber, which was perfused with saline solution at 1.5 ml/min at 37°C. The chamber in turn was placed on the stage of an inverted microscope connected to a digital imaging system. Ratios of images (340 nm excitation/380 nm excitation, emission filter 520 nm) were obtained at 1.5-s intervals. A region of interest covering 15 ϫ 15 m was defined over each cell, and the average ratio intensity over the region was converted to [Ca 2ϩ ] i using a calibration curve constructed with a series of calibrated buffered calcium solutions (Calcium Calibration Buffer Kit #2, Molecular Probes, OR). Identification of cells transiently expressing Clostridium botulinum C3 exoenzyme, Rho G14V, Fil-A14/15, CaRstop907, or CaR (see "cDNAs") was achieved by cotransfection with a plasmid encoding wild type green (GFP) or red fluorescent protein (RFP) (BD Biosciences). A minimum of 100 cells/experiment, each experiment done in at least duplicate or triplicate were used to measured [Ca 2ϩ ] i .
[Ca 2ϩ ] i , Ins (1,4,5)P 3 , and Diacylglycerol Imaging in a Single Live Cell-Simultaneous single live cell imaging of the fluorescent biosensors for [Ca 2ϩ ] i (PKC␣-yellow fluorescent protein (YFP)), Ins(1,4,5)P 3 (GFP-PHD), and diacylglycerol (PKD-RFP) was achieved by emission fingerprinting with a confocal LSM 510 Meta microscope (Carl Zeiss, Germany) (nm excitation/nm emission PKC␣-YFP: 514/527; GFP-PHD: 488/507; PKD-RFP: 543/583). Culture conditions for live cell imaging and quantitative analysis of the relative change in plasma membrane and cytosol fluorescence intensity of individual cells were performed as described previously (14), analyzing 50 cells/experiment, with each experiment done in at least duplicate or triplicate. The selected cells displayed in the figures were representative of 90% of the population of positive cells.
Western Blot, PKD Kinase Assays, and Exogenous Substrate Phosphorylation-Western blot analysis was performed as previously described (15). PKD autophosphorylation was determined in an in vitro kinase assay followed by SDS-PAGE analysis and quantification as described previously (13). Exogenous substrate syntide-2 phosphorylation by PKD was assayed and quantified by Cerenkov counting as previously published (16).
cDNAs-Primers for the synthesis of the cDNA encoding the pleckstrin homology domain (PHD) of human phospholipase C (PLC)-␦1 and human PKC␣ were designed with the Primer3 program (17). The sense primers for PHD and PKC␣ were 5Ј-CCGAATTCACGGCCTACAGGA-TGATGAG-3Ј and 5Ј-CCGAATTCGGGGGGGGACCATGGCTGACGT-3Ј, respectively, whereas the corresponding antisense primers were 5Ј-CCGAATTCGGAAGTTCTGCAGCTCCTTG-3Ј and 5Ј-CGGAATTC-GCGCTGGTGAGTTTGCTACTGCACTCTG-3Ј. Total RNA extracted from human pancreatic cancer Panc-1 cells was employed as template for the synthesis of the cDNAs encoding the PHD and PKC␣ using reverse transcription-PCR (18). PHD and PKC␣ were cloned into the EcoRI site of vectors encoding the GFP or YFP, pEGFP-C1 or pEYFP-N1. A cDNA encoding human filamin-A, containing amino acids 1530 through 1875 (14 and 15 domains) (Fil-A14/15), was synthesized by reverse transcription-PCR using total RNA extracted form Panc-1 cells (see above) and the forward/reverse primers 5Ј-CTTAAGCTTCCATG-GTACCCCGGAGCCCC-3Ј/5Ј-CGGAATTCCGAGGCCAGGCCCATAG-GC-3Ј. The obtained cDNA was cloned into the EcoRI/HindIII sites of pcDNA3.1 myc-His C (Invitrogen). Site-directed mutagenesis to introduce a stop codon after amino acid 906 of the CaSR (CaR stop907) was performed as previously described (14) with the forward and reverse primers 5Ј-GGAGGCTCCACGTGATCAACCCCCTCCTCC-3Ј and 5Ј-G-GAGGAGGGGGTTGATCACGTGGAGCCTCC-3Ј, respectively. All of the constructs were verified by DNA sequencing and Western blot analysis. The vectors encoding the fusion protein between protein kinase D (PKD) and a red fluorescent protein (pPKD-RFP), untagged PKD and C. botulinum C3 exoenzyme were described previously (14,16). The cDNA for the constitutively active human Rho G14V was obtained from the University of Missouri-Rolla cDNA Resource Center, MO.
Materials-[␥-32 P]ATP (370 MBq/ml) and horseradish peroxidaseconjugated donkey anti-rabbit IgG were from Amersham Biosciences. The anti-Ser(P) 744 /Ser(P) 748 , which specifically recognizes the phosphorylated state of those serines within the activation loop of protein kinase D (PKD), was obtained from Cell Signaling Technology. The anti-PKC antibody (C-20), which recognizes PKD, was obtained from Santa Cruz Biotechnology. The anti-RFP antibody was obtained from BD Biosciences. The anti-CaR was obtained from Affinity BioReagents. Fura-2/AM was purchased from Molecular Probes. All of the other reagents were the highest grade commercially available.

Differential Production of Ins (1,4,5)P 3 and Diacylglycerol (DAG) Mediated by the CaR in HEK-293 Cells Stimulated by L-Phenylalanine or [Ca 2ϩ
] e -Several models have been proposed to explain the generation of [Ca 2ϩ ] i oscillations in response to GPCR activation, but definitive evidence identifying the mechanism(s) involved is available in very few instances. A major advance in this field has been the development of probes that allow the monitoring of the intracellular concentration of the two second messengers generated by PLC activation, Ins(1,4,5)P 3 and DAG. To determine whether [Ca 2ϩ ] o and aromatic amino acids differ in their ability to trigger the synthesis of these second messengers via the CaR, we used a set of biosensors to monitor, in real time and in the same cells, the synthesis of Ins(1,4,5)P 3 and DAG simultaneously with [Ca 2ϩ ] i oscillations. The synthesis of Ins(1,4,5)P 3 was monitored by examining the dynamic distribution of a fusion protein between GFP and the PHD of PLC-␦1 (GFP-PHD). This PHD binds phosphatidylinositol 4,5-biphosphate in the plasma membrane but translocates to the cytosol in response to PLC-mediated Ins(1,4,5)P 3 synthesis (19 -22). DAG synthesis was monitored by imaging a fusion protein consisting of PKD and a red fluorescent protein (PKD-RFP) (14). PKD is a Ca 2ϩ -insensitive serine/threonine kinase that rapidly translocates from the cytosol to the plasma membrane in response to PLC-mediated DAG production in response to GPCR stimulation (14,23). As an indication of [Ca 2ϩ ] i oscillations, we monitored the cytosol to plasma membrane translocation of a chimeric protein between PKC␣, a Ca 2ϩ -and phospholipid-dependent PKC, and a yellow fluorescent protein (PKC␣-YFP) (24).
To examine CaR-mediated changes in second messenger generation in single cells, HEK-293 cells stably expressing the human CaR, a cell system widely used to study CaR regulation (6), were cotransfected with plasmids encoding GFP-PHD, PKD-RFP, and PKC␣-YFP. The cells were subsequently stimulated by increasing the [Ca 2ϩ ] o to 5 mM (Fig. 1A) or by addition of 5 mM L-phenylalanine (Fig. 1B). The intracellular distribution of the biosensors was monitored simultaneously in live cells using emission fingerprinting. In non-stimulated cells, the Ins(1,4,5)P 3 sensor GFP-PHD was predominantly localized at the plasma membrane, whereas the DAG sensor PKD-RFP was distributed throughout the cytoplasm but excluded from the nucleus. The distribution of PKC␣-YFP was similar to PKD-RFP, i.e. throughout the cytoplasm, although in few cells PKC␣-YFP was also present in the nucleus. GFP, RFP, or YFP alone localized both in the cytoplasm and nucleus (data not shown). [Ca 2ϩ ] o -elicited CaR stimulation induced an oscillatory translocation of GFP-PHD from the plasma membrane to the cytoplasm, reflecting periodic production of Ins(1,4,5)P 3 . These changes coincided with the oscillatory translocation of PKC␣-YFP from the cytoplasm to the plasma membrane. DAG synthesis, monitored by the translocation of its sensor from the cytosol to the plasma membrane, peaked after 3 min. Within 5-6 min of decreasing the [Ca 2ϩ ] o to the basal level (1.5 mM), the sensors returned to the subcellular compartments they occupied before stimulation (data not shown).
In striking contrast to the results obtained with [Ca 2ϩ ] o stimulation, L-phenylalanine-elicited CaR activation did not induce any detectable synthesis of either Ins(1,4,5)P 3 or DAG, as revealed by the lack of redistribution of their corresponding sensors (Fig. 1B). The lack of translocation of the DAG sensor to the plasma membrane was not because of a low level of DAG accumulation. Inhibition of DAG conversion to phosphatidic acid by preventing its phosphorylation with 10 M diacylglycerol kinase inhibitor I (IC 50 ϭ 2.8 M) or II (IC 50 ϭ 120 nM) for 10 min, concurrently with 5 mM L-phenylalanine stimulation, failed to promote any plasma membrane translocation of the DAG sensor (data not shown). However, L-phenylalanine stimulation of the CaR induced a distinct and oscillatory translocation of PKC␣-YFP from the cytosol to the plasma membrane ( Differential Regulation of PKD Activation Loop Phosphorylation and Kinase Activity in Response to CaR Activation by [Ca 2ϩ ] e or L-Phenylalanine-Recently, we demonstrated that the plasma membrane translocation of PKD, in response to GPCR-induced DAG synthesis, is necessary for PKC⑀-mediated phosphorylation of the activation loop of PKD, a critical step in the catalytic activation of this enzyme (23). In view of the results presented in Fig. 1, we predicted that [Ca 2ϩ ] o -elicited CaR activation should lead to the PKC-mediated activation loop phosphorylation of PKD at Ser 744 and Ser 748 , leading to the catalytic activation of this kinase (25,26). Fig. 2A shows that [Ca 2ϩ ] o -elicited CaR activation induced a striking increase in the phosphorylation of Ser 744 and Ser 748 of PKD. In contrast, L-phenylalanine did not promote any detectable increase in the activation loop phosphorylation of PKD ( Fig. 2A), even in the presence of diacylglycerol kinase inhibitor I or II at a 10 M final concentration (data not shown). Further support for this conclusion was obtained by measuring PKD autophosphorylation. HEK-293 cells expressing the CaR stimulated with [Ca 2ϩ ] o or L-phenylalanine were lysed and PKD immunoprecipitated from the extracts. The immune complexes were incubated with [␥-32 P]ATP, subjected to SDS-PAGE, and analyzed by autoradiography to detect the prominent 110-kDa band corresponding to autophosphorylated PKD. The results presented in Fig. 2B show that [Ca 2ϩ ] o , but not L-phenylalanine, induced a marked increase in PKD autophosphorylation activity. As illustrated in Fig. 2C, similar results were obtained when PKD activity in the immunocomplexes was determined by phosphorylation of syntide-2 (27, 28), a synthetic peptide previously demonstrated to be an excellent substrate for PKD (29). These results support the notion that an increase in [Ca 2ϩ ] o and amino In addition to the parathyroid gland, the CaR has also been cloned from the kidneys of humans, rabbits, and rats (30 -32). Within the kidney, several regions such as the proximal tubule express the CaR (33). Consequently, we obtained primary RPTEC of human origin to examine whether L-phenylalanine and [Ca 2ϩ ] o also induced different patterns of [Ca 2ϩ ] i oscillations in cells endogenously expressing the CaR. Western blot analysis of RPTEC showed that these cells express similar levels of CaR to HEK-293 ectopically expressing the same receptor (Fig. 3C, inset, lanes 3 and 2, respectively). No signal was detected in wild type non-transfected HEK-293 cells (Fig.  3C, inset, lane 1) or when the reactivity of the human CaR antibody was eliminated by preincubating the antiserum with the immunizing peptide (data not shown). Subsequently, RPTEC were loaded with the fluorescent Ca 2ϩ indicator fura-2 and imaged as described under "Experimental Procedures." As shown in Fig. 3C (Fig. 3C). In contrast, L-phenylalanine did not promote any detectable increase in the activation loop phosphorylation of PKD (Fig. 3C). These results demonstrated that the differential signaling evoked by L-phenylalanine and [Ca 2ϩ ] o are not restricted to cells transfected with constructs encoding CaR but can be also obtained in cells that endogenously express this receptor.
CaR-mediated Transient [Ca 2ϩ ] i Oscillations Require the Organization of the Actin Cytoskeleton-A number of studies have shown that the organization of the actin cytoskeleton plays a role in the generation of [Ca 2ϩ ] i oscillations in at least some cell types (34,35). As a first step to elucidate the mechanism(s) involved in L-amino acid-induced [Ca 2ϩ ] i oscillations through the CaR, we determined whether disruption of the actin cytoskeleton differentially affects this type of [Ca 2ϩ ] i oscillations. We utilized the structurally unrelated agents cytochalasin D and latrunculin A, which induce actin cytoskeleton depolymerization through different mechanisms (36). Cytochalasin D binds to the growing end of actin filaments, leading to disrup-   These results suggest that the interaction of the C terminus of the CaR with filamin-A plays a role in the generation of differential Ca 2ϩ signaling by this receptor. Further support for this conclusion was obtained by expressing a truncated CaR (CaR-stop907) that does not interact with filamin-A (41,42) in HEK-293 cells. As shown in Fig. 5B, L-phenylalanine failed to evoke transient [Ca 2ϩ ] i oscillations in HEK-293 cells transfected with the truncated CaR. This mutated form of CaR was functional, because an elevation in [Ca 2ϩ ] o induced a marked increase in [Ca 2ϩ ] i characterized by a peak and plateau, a response similar to that obtained with another CaR mutant in which threonine 888, the major site of PKC phosphorylation, was converted to non-phosphorylatable alanine (11).

CaR-mediated Transient [Ca 2ϩ ] i Oscillations Require Functional Rho and the Interaction of Filamin-A with the CaR-The
G␣ 12/13 Mediate CaR-induced Transient [Ca 2ϩ ] i Oscillations-Recently, the G 12 subfamily has been implicated in pathways leading to activation of the low molecular weight G proteins of the Rho subfamily (44 -49). In view of the results presented here implicating the actin cytoskeleton and Rho in the generation of transient [Ca 2ϩ ] i oscillations via the CaR, we hypothesized that the G 12 subfamily mediates this type of oscillation. Most cell types express both G␣ 12 and G␣ 13 with overlapping functions, thus rendering it difficult to analyze the contribution of these G proteins to GPCR signaling. To circumvent this problem, we examined the [Ca 2ϩ ] i oscillations elicited by L-phenylalanine or an increase in [Ca 2ϩ ] o in MEFs generated from doubled knock-out mice for both G␣ 12 and G␣ 13 (referred as G 12/13 Ϫ/Ϫ) (13) transiently expressing the CaR. MEF lacking rhodopsin kinase (RKϪ/Ϫ) served as a control. In agreement with the results obtained with RPTEC and HEK-293 cells, L-phenylalanine or [Ca 2ϩ ] o stimulation of MEF RKϪ/Ϫ evoked transient and sinusoidal [Ca 2ϩ ] i oscillations, respectively (Fig. 6). These results further demonstrate that these agonists elicit a different pattern of [Ca 2ϩ ] i oscillations in a variety of cell types.
Next, we determined the effect of L-phenylalanine or [Ca 2ϩ ] o on [Ca 2ϩ ] i oscillations in MEF G 12/13 Ϫ/Ϫ cells. As shown in Fig.  6, [Ca 2ϩ ] o stimulation of MEF G 12/13 Ϫ/Ϫ elicited sinusoidal [Ca 2ϩ ] i oscillations, in agreement with the notion that these types of oscillations are generated by the CaR via G q and involve periodic production of Ins(1,4,5)P 3 and feedback inhibition by PKC (11). In striking contrast, the addition of Lphenylalanine to MEF G 12/13 Ϫ/Ϫ failed to promote any [Ca 2ϩ ] i oscillations (Fig. 6). These results demonstrate that the tran- Most models proposed to explain the mechanism by which [Ca 2ϩ ] i oscillations are generated in response to GPCR activation via the intracellular second messenger Ins(1,4,5)P 3 can be broadly divided in two major classes, depending on negative feedback effects of PKC on the production of Ins(1,4,5)P 3 or on the regulatory effects of [Ca 2ϩ ] i on the Ins(1,4,5)P 3 receptor (12). In the first case, the levels of Ins(1,4,5)P 3 change in a cyclical fashion thereby driving the [Ca 2ϩ ] i oscillations, whereas in the second case, [Ca 2ϩ ] i oscillations take place in cells with a steady state increased level of Ins(1,4,5)P 3 via a Ca 2ϩinduced Ca 2ϩ release process mediated by the Ins(1,4,5)P 3 receptor. The recent development of techniques to image real time changes in the generation of Ins(1,4,5)P 3 and DAG in single cells has provided a valuable approach to distinguish between these models (24).
Our results demonstrated a dramatic difference in the ability of the CaR agonists to induce phosphatidylinositol 4,5-biphosphate hydrolysis and the corresponding generation of Ins(1,4,5)P 3 and DAG in single cells ectopically (HEK-293) or endogenously The notion that the CaR acts trough PLC-dependent and PLC-independent pathways was further substantiated by the striking difference in the ability of the CaR agonists to induce translocation of PKD-RFP to the plasma membrane, indicative of DAG production. Furthermore, we found that [Ca 2ϩ ] o -elicited CaR activation induced a prominent increase in the phosphorylation of PKD at Ser 744 and Ser 748 , the key residues located in the activation loop of this enzyme that are phosphorylated by novel PKCs and promote the catalytic activation of PKD (25,26,51 The results presented in this study demonstrate, for the first time, that L-phenylalanine-elicited CaR-mediated base-line [Ca 2ϩ ] i spiking was selectively abolished by multiple approaches. These include pharmacological disruption of the actin cytoskeleton using the structurally unrelated agents cytochalasin D and latrunculin A, inactivation of Rho GTPases with toxin B and C3, interference with the binding of filamin-A to the CaR, expression of a CaR lacking the binding region of filamin-A, and by expressing the CaR in cells derived from genetically modified mice lacking the ␣ subunits of both G 12 and G 13 . In each case, we demonstrated suppression of [Ca 2ϩ ] i signaling in response to amino acid stimulation but retention of ] i base-line in cells expressing the CaR, suggesting a novel role for this receptor in the generation of intracellular Ca 2ϩ signaling in response to Rho activation. We concluded that L-phenylalanine-elicited CaR activation promotes baseline [Ca 2ϩ ] i spiking through a novel signal transduction pathway involving the heterotrimeric proteins of the G 12 subfamily, the scaffolding protein filamin-A, the small molecular weight GTPase Rho, and the intactness of the actin cytoskeleton. Although the precise mechanism and components mediating the generation of transient [Ca 2ϩ ] i oscillations requires further investigation, it is tempting to speculate that the cytoplasmic tail of the CaR acts as a "nucleation" component around which Rho, filamin A, microfilaments, and possibly other proteins are recruited in response to L-phenylalanine stimulation of this GPCR. In this regard, it was recently demonstrated that thrombin-mediated activation of RhoA regulates Ca 2ϩ entry by a mechanism involving a complex among activated RhoA, the transient receptor potential channel 1, the inositol triphosphate receptor, and an intact cytoskeleton (52). RhoA activation not only mediated the formation of this complex but also its translocation to the plasma membrane (52).
Upon agonist binding, a change in the GPCR conformation facilitates the activation of heterotrimeric G proteins, which in turn activate downstream signaling pathways depending on the specific type of G protein to which the receptor is coupled. This simple GPCR active state model is being reviewed in the context of recent theoretical and experimental evidence showing that the binding of different agonists to the same receptor can lead to the activation of different G proteins by inducing distinct GPCR conformations (reviewed in Ref. 53). Although evidence supporting this model of agonist-specific trafficking of receptor signaling, i.e. agonist trafficking (54,55), has been found for a limited number of GPCR using mostly synthetic agonists, the importance of this effect under physiological conditions has not been clearly established (53). Our model is that the physiologically relevant agonists, Ca 2ϩ and amino acids, induce distinct conformational states of the CaR, providing for the possibility of differential coupling to downstream signaling pathways.
In conclusion, our results demonstrate that the CaR mediates sinusoidal or transient patterns of [Ca 2ϩ ] i oscillations in response to different agonists through different mechanisms. Transient oscillations are produced via a PLC/Ins(1,4,5)P 3independent pathway that involves Rho, filamin-A, and the organization of the actin cytoskeleton. Therefore, a conceptual model for the CaR could be that Ca 2ϩ or amino acids induce distinct conformational states of this allosteric GPCR, providing for the possibility of differential coupling to G proteins and thus to downstream signaling pathways.