Expression and characterization of inactivating and activating mutations in the human Ca2+o-sensing receptor.

Nearly 30 mutations have been identified to date in the coding region of the extracellular calcium-sensing receptor (CaR) that are associated with inherited human hypo- and hypercalcemic disorders. To understand the mechanisms by which the mutations alter the function of the receptor may help to discern the structure-function relationships in terms of ligand-binding and G protein coupling. In the present studies, we transiently expressed eight known CaR mutations in HEK293 cells. The effects of the mutations on extracellular calcium- and gadolinium-elicited increases in the cytosolic calcium concentration were then examined. Seven inactivating mutations, which cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism, show a reduced functional activity of the receptor because they may 1) reduce its affinity for agonists; 2) prevent conversion of the receptor from a putatively immature, high mannose form into the fully glycosylated and biologically active form of the CaR, in addition to lowering its affinity for agonists; or 3) fail to couple the receptor to and/or activate its respective G protein(s). Conversely, one activating mutation, which causes a form of autosomal dominant hypocalcemia, appears to increase the affinity of the receptor for its agonists.

The recently cloned extracellular Ca 2ϩ (Ca 2ϩ o )-sensing receptor (CaR) 1 (1) has provided key insights into the pathogenesis of inherited human hypo-and hypercalcemic disorders (2,3). The receptor, BoPCaR (bovine parathyroid Ca 2ϩ -sensing receptor) was first isolated from bovine parathyroid using expression cloning in Xenopus laevis oocytes and shows pharma-cological properties nearly identical to those of the native receptor in its responses to extracellular divalent cations (i.e. Ca 2ϩ o and Mg 2ϩ o ), trivalent cations (e.g. Gd 3ϩ o ), and polyamines (e.g. neomycin) (1). Subsequently, cDNAs encoding the human homolog of the same receptor have been cloned from human parathyroid (4) and kidney (5), using a homology-based strategy. The human and bovine receptors share a high degree of homology at the amino acid level (93% identity). Stimulation of the CaR by agonists activates phospholipase C, with resultant increases in inositol phosphates and the cytosolic calcium concentration (Ca 2ϩ i ) (1). The amino acid sequences of the human and bovine receptors deduced from their cDNAs predict proteins of 1078 and 1085 amino acids, respectively. Both contain a large (ϳ600 residues) extracellular amino terminus, a seven-transmembrane domain, similar to that in other members of the G protein-coupled receptor superfamily, and a ϳ200-amino acid carboxyl-terminal tail. The only other G protein-coupled receptors with which the CaR shares a significant degree of homology are the metabotropic glutamate receptors (mGluRs), which also feature large, putatively extracellular domains at their amino termini (about 600 amino acids) (6). The extracellular domains of both the mGluRs and the CaR have been proposed to bind their respective agonists (1,6).
The physiological importance of the receptor in determining the extracellular calcium concentration has been documented by the characterization of human syndromes resulting from activating or inactivating mutations of the CaR, which alter the function of parathyroid and kidney so as to produce hypo-or hypercalcemia, respectively (2,3). When present in the heterozygous state, inactivating mutations cause familial hypocalciuric hypercalcemia (FHH), whereas in the homozygous state, they cause neonatal severe hyperparathyroidism (NSHPT). In contrast, activating mutations are responsible for a form of autosomal dominant hypocalcemia (ADH). Both disorders show abnormal Ca 2ϩ o sensing and/or handling in kidney and parathyroid. However, the mechanisms by which these mutations alter the function of the CaR have not been studied in detail. In this report, we introduced some of known mutations causing either FHH/NSHPT or ADH into a human parathyroid Ca 2ϩ osensing receptor cDNA (called HuPCaR4.0) (2,3,7). We then studied the capacities of the resulting mutant receptor proteins to be properly biosynthesized and processed when expressed in human embryonic kidney cells (HEK293) and to mediate increases in Ca 2ϩ i in response to CaR agonists (e.g. Ca 2ϩ o and Gd 3ϩ o ).

EXPERIMENTAL PROCEDURES
Reconstruction of the Receptor-Because the FHH/NSHPT and activating mutations discovered to date are distributed throughout the CaR protein, it would greatly facilitate studies on the expression of these mutations to have a construct of the receptor that is subdivided into cassettes of convenient sizes for site-directed mutagenesis. We have reconstructed the coding region of HuPCaR4.0 and divided it into five segments of about 450 base pairs each as well as one 900-base pair segment by creating five unique restriction sites in the order of BspEI, NheI, AflII, HpaI, and XhoI (see Fig. 2 in "Results") without changing the predicted amino acid sequence. In the reconstructed receptor (rH-uPCaR4.0), the altered codons were incorporated into the 5Ј ends of the respective primers for polymerase chain reaction, and the appropriate segments of the reconstructed receptor were amplified using the human Ca 2ϩ o -sensing receptor (HuPCaR4.0) as a template. The modified codons for each of the unique restriction sites in their respective primers were flanked at their 3Ј ends by the wild type sequence. Moreover, each forward primer carries a KpnI site preceding the unique restriction sites with the respective modified codon, except that the KpnI site in the first primer (i.e. that nearest the 5Ј end of the open reading frame) precedes the translational start codon. Each reverse primer carries an XbaI site preceding the unique restriction sites, except that the XbaI site in the last primer precedes the complement sequence of the translational stop codon. The polymerase chain reaction products for each segment of the reconstructed receptor were then cloned into the KpnI and XbaI sites of the cloning vector, pBluescript SK Ϫ (Stratagene), and these latter constructs served as cassettes for site-directed mutagenesis. The sequences of the six cassettes, which together encode the full coding sequence of the receptor, were confirmed using an Applied Biosystems 373A Automated DNA Sequencer. Finally, the wild type receptor was reconstructed by fusing the six clones via stepwise subcloning.
Site-directed Mutagenesis-Site-directed mutagenesis to produce specific FHH/NSHPT or ADH mutations was performed using the approach described by Kunkel (8). The dut-1 ung-1 strain of Escherichia coli, CJ236, was transformed separately with each relevant mutagenesis cassette. Uracil-containing, single stranded DNA was produced by infecting the cells with the helper phage, VCSM13. The single stranded DNA was then annealed to a mutagenesis primer, which contained the desired nucleotide change encoding a single point mutation flanked on both sides by wild type sequence. The primer was then extended around the entire single stranded DNA and ligated to generate closed circular heteroduplex DNA. DH5␣ competent cells were transformed with these DNA heteroduplexes, and incorporation of the desired mutation was confirmed in all cases by sequencing the entire cassette.
Transient Expression of the Reconstructed HuPCaR4.0 (rHuP-CaR4.0) and Mutated Receptors Carrying FHH/NSHPT or ADH Mutations in a Human Embryonic Kidney Cell Line (HEK293)-The fulllength rHuPCaR4.0 and mutant receptors were cloned into the KpnI and XbaI sites of the mammalian expression vector, pcDNA3 (Invitrogen). The DNA for transfection was prepared using the Midi Plasmid Kit (Qiagen). LipofectAMINE (Life Technologies, Inc.) was employed as a DNA carrier for transfection (9,10). The HEK293 cells used for transient transfection were provided by NPS Pharmaceuticals, Inc. (Salt Lake City, UT) and were cultured in Dulbecco's modified Eagle's medium (Life Technologies, Inc.) with 10% fetal bovine serum (Hyclone). The DNA-liposome complex was prepared by mixing DNA and LipofectAMINE in OPTI-MEM I reduced serum medium (Life Technologies, Inc.) and incubating the mixture at room temperature for 30 min. The DNA-lipofectAMINE mixture was then diluted with OPTI-MEM I reduced serum medium and added to 90% confluent HEK293 cells plated on 13.5 ϫ 20.1-mm glass coverslips using 2.5 g of DNA (for measurement of Ca 2ϩ i ) or in 100-mm Petri dishes using 15 g of DNA (for obtaining membrane protein for Western analysis). After 5 h of incubation at 37°C, equivalent amounts of OPTI-MEM I reduced serum medium with 20% fetal bovine serum were added to the medium overlying the transfected cells, and the latter was replaced with fresh Dulbecco's modified Eagle's medium with 10% fetal bovine serum at 24 h after the transfection. The expressed Ca 2ϩ o -sensing receptor protein was assayed 48 h after the start of transfection. To perform coexpression of wild type and mutant receptors, 0.625 g each of the wild type and a mutant receptor cDNA were mixed and used to transfect HEK293 cells as described above.
Determination of Transfection Efficiency by Immunocytochemistry-The transfected cells and nontransfected cells on coverslips were first fixed with 4% formaldehyde in phosphate-buffered saline (PBS) for 5 min at room temperature. Endogenous peroxidases were then inhibited by incubating the cells in DAKO peroxidase blocking reagent (DAKO Corp.) for 10 min, and the cells were further treated with DAKO protein block serum-free solution (DAKO Corp.) for 1 h (11,12). Cells were then incubated overnight at 4°C with a protein A-purified primary antiCaR antibody (4641, kindly provided by Drs. Forrest Fuller and Rachel Simin at NPS Pharmaceuticals, Inc.) with or without preincubation with the synthetic peptide of BoPCaRl (amino acids 215-237) against which it was raised (as a control for nonspecific staining). The final concentration of antibody was 10 g/ml. After washing the cells three times with 0.5% BSA in PBS for 20 min each, a peroxidase-coupled, goat anti-rabbit secondary antiserum diluted 1:200 (Sigma) was added and incubated for 1 h at room temperature. The cells were then washed with PBS three times for 20 min each and stained with the DAKO AEC substrate system (DAKO Corp.) for about 5 min. The color reaction was stopped by washing three times with water. The stained coverslips were mounted on slides with AQUA-MOUNT (Lerner Laboratories). Photomicrographs were taken with a 10ϫ objective.
Tunicamycin Treatment of Transiently Transfected Cells-Tunicamycin (5 g/ml) (Boehringer Mannheim) was added to HEK293 cells 5 h after transfection with rHuPCaR4.0 to block N-glycosylation of the newly synthesized receptor protein (13). The cells were then incubated for 48 h before harvesting of plasma membranes for Western analysis of the CaR as described below.
Crude Plasma Membrane Preparations from Transfected HEK293 Cells and Bovine Parathyroid Glands-Crude plasma membranes were isolated from HEK293 cells transiently transfected with rHuPCaR4.0 or mutant receptors by differential speed centrifugation as described by Sun et al. (14). Confluent cultured cells in 100-mm culture plates were rinsed twice with PBS and treated with 0.02% EDTA in PBS at 37°C for 5 min. The detached cells were pelleted and suspended in 300 l of homogenization buffer (50 mM Tris-HCl, pH 7.4, containing 0.32 M sucrose, 1 mM EDTA, and protease inhibitors, including 83 g/ml aprotinin, 30 g/ml leupeptin, 1 mg/ml Pefabloc SC, 50 g/ml calpain inhibitor, 50 g/ml bestatin, 5 g/ml pepstatin, and 1 mM EDTA). Then the cells were homogenized with 15 strokes of a motor-driven Teflon pestle in a tightly fitted glass tube. The homogenate was sedimented at 18,800 ϫ g for 20 min to remove nuclei and mitochondria. The supernatant was subsequently sedimented at 43,500 ϫ g for 20 min to pellet the plasma membranes, and the resultant pellet was solubilized with 1% Triton X-100. All steps were carried out at 4°C. One bovine parathyroid gland was first cut into small pieces and placed in 600 l of the above homogenization buffer and then homogenized with 45 strokes in a motor-driven Teflon pestle in a tightly fitted glass tube. Crude plasma membranes were isolated from the homogenate as described above.
Digestion of CaR with Endoglycosidase H or with Peptide-N-Glycosidase F (PNGase F)-For digestion with endoglycosidase H, crude plasma membrane protein, 1.6 g, was first denatured at 37°C for 15 min in the presence of 0.05% SDS and 50 mM 2-mercaptoethanol and then incubated with or without 2 milliunits of endoglycosidase H (Boehringer Mannheim) in a buffer containing 70 mM sodium acetate, pH 5.2, and 0.8% Triton X-100 at 30°C overnight (15). For digestion with peptide-N-glycosidase F, crude membrane protein, 1.6 g, was denatured at 37°C for 15 min in the presence of 0.375% SDS and 75 mM 2-mercaptoethanol and then incubated with or without 0.5 unit of PNGase F (Boehringer Mannheim) in a buffer containing 150 mM Tris-HCl, pH 8, and 1.3% Triton X-100 at 30°C overnight (15).
Western Analysis of Plasma Membrane Proteins-After determination of the protein concentration in the crude plasma membrane preparations using the Pierce BCA protein assay (Pierce), an appropriate amount of membrane protein (4 g) was subjected to SDS-polyacrylamide gel electrophoresis (PAGE) (16) using a linear gradient of polyacrylamide (4 -12%). The proteins on the gel were subsequently electrotransferred to a nitrocellulose membrane. After blocking with 5% milk, the blot was incubated with the same primary antibody (4641) used for immunocytochemistry with or without preincubation with the peptide against which it was raised (as a control for nonspecific binding) and then with a secondary, goat anti-rabbit antibody conjugated to horseradish peroxidase (Sigma; diluted 1:500). The Ca 2ϩ o -sensing receptor protein was detected with an Enhanced Chemiluminescence system (Amersham Corp.).

Measurement of Ca 2ϩ i by Fluorimetry in Cell
Populations-Coverslips coated with HEK293 cells that had been transfected with wild type or mutant CaR cDNAs were loaded for 2 h at room temperature with fura-2/AM in 20 mM HEPES, pH 7.4, containing 125 mM NaCl, 4 mM KCl, 1.25 mM CaCl 2 , 1 mM MgSO 4 , 1 mM NaH 2 PO 4 , 0.1% BSA, and 0.1% dextrose and washed once with a bath solution (20 mM HEPES, pH 7.4, containing 125 mM NaCl, 4 mM KCl, 0.5 mM CaCl 2 , 0.5 mM MgCl 2 , 0.1% dextrose, and 0.1% BSA) at 37°C for 20 min. The coverslips were then placed diagonally in a thermostatted quartz cuvette containing the bath solution, using a modified technique employed previously in this laboratory (17). CaR agonists were added to give desired concentrations. Excitation monochrometers were centered at 340 and 380 nm with emission light collected at 510 Ϯ 40 nm through a wide band emission

Expression of Ca 2ϩ o Receptor Mutations
filter. The 340/380 excitation ratio of emitted light was used to calculate Ca 2ϩ i as described previously (17). The addition of Ca 2ϩ o with stepwise 1 mM increments was employed to characterize the wild type, the inactivating mutations (such as R62M and T138M, or the activating mutation (E127A). Increments of 5 mM were used for mutations with attenuated responses in order to study the effects of higher levels of

FHH/NSHPT and ADH Mutations Introduced into the Human CaR and the Restriction Map of the Reconstructed CaR
Used for Site-directed Mutagenesis- Fig. 1 shows the activating and inactivating mutations engineered into the reconstructed human CaR as well as their positions relative to the predicted topology of the CaR. Fig. 2 shows the restriction maps for both the wild type and reconstructed human CaRs, illustrating the silent mutations introduced into the wild type CaR to create five unique restriction sites defining six mutagenesis cassettes encompassing the complete coding sequence of the receptor.
Expression of the CaR in Transiently Transfected HEK293 Cells-We immunostained the HEK293 cells transiently transfected with rHuPCaR4.0 using pcDNA3 as a vector and Lipo-fectAMINE as a carrier. The transfection efficiency of the receptor was estimated to be about 25%, as shown in Fig. 3. The immunostaining of the cells was specific, because it could be completely blocked by preincubation of the CaR antibody with the peptide against which it was raised (data not shown). Moreover, there was no staining of nontransfected HEK293 cells with the antiCaR antibody (data not shown).
Western Analysis of the CaR Expressed in HEK293 Cells-On Western analysis of crude plasma membrane proteins isolated from transiently transfected or wild type HEK293 cells using the same antiCaR antibody, there were three immunoreactive bands between 120 and 200 kDa as well as additional bands of

Expression of Ca 2ϩ o Receptor Mutations
higher molecular masses (ϳ350 kDa) that were found only in the transfected cells and were ablated following preabsorption of the antibody with specific peptide (Fig. 4, A and B, lanes 1 and 2). The pattern of immunoreactive bands observed in crude plasma membranes prepared from transfected HEK293 cells was almost identical to that in membranes prepared from bovine parathyroid glands (Fig. 4A, lane 3). Inhibition of Glycosylation of the CaR with Tunicamycin-The immunoreactive bands present as a doublet between 140 and 200 kDa in transfected HEK293 cells were absent in the cells treated with tunicamycin for 48 h, beginning 5 h after transfection with rHuPCaR4.0 (Fig. 5). The band around 120 kDa in the crude plasma membranes isolated from the tunicamycin-treated cells co-migrated with the lowest band of the triplet between 120 and 200 kDa found in tunicamycin mocktreated cells in the same experiment.
Characterization of Multiple Forms of the CaR Expressed in HEK293 Cells-The lower band of the doublet between 140 and 200 kDa was sensitive to endoglycosidase H as well as to PNGase F, but the upper band was only sensitive to PNGase F (Fig. 5). The species generated by endoglycosidase H co-migrated with the nonglycosylated form of the receptor at around 120 kDa (see tunicamycin-treated cells, Fig. 5). However, following treatment with PNGase F, the upper band migrated between 120 and 140 kDa in untreated cells. This suggests that the mature receptor may carry posttranslational modification(s) in addition to N-glycosylation.
Western Analysis of Transfected CaRs Bearing FHH/ NSHPT and ADH Mutations-Crude plasma membranes were isolated from HEK293 cells transiently transfected with either rHuPCaR4.0 or receptors containing the mutations indicated in Fig. 1. On Western analysis the intensity of the upper band relative to the lower band of the doublet between 140 and 200 kDa was reduced in cells transfected with cDNAs carrying the R66C, G143E, and E297K mutations (Fig. 6). Nevertheless, some protein at the position of the upper band was detected for each of these three mutant receptors. The remaining mutant receptors exhibited an expression pattern on Western analysis similar to that of rHuPCaR4.0 (Fig. 6). The ADH mutation (E127A) also exhibited a pattern of immunoreactive bands similar to that of the wild type receptor (data not shown).  o that was similar to that of rHuP-CaR4.0 but had a left-shifted dose response curve, with an EC 50 of 3.3 Ϯ 0.1 mM (n ϭ 11) (Fig. 8).

Coexpression of the Wild Type Receptor with Mutant
Receptors-R185Q caused a substantially delayed initial response of the coexpressed wild type receptor in HEK293 cells to Ca 2ϩ o ( Fig. 9C) with the most prominently increased EC 50 among the cotransfections (Table I). R795W also shifted the initial response to the right but to a lesser extent when coexpressed with the wild type receptor (data not shown). In contrast, the other FHH mutant receptors did not substantially alter the initial response of the coexpressed wild type receptor to Ca 2ϩ o (T138M as an example is shown in Fig. 9B), although some of them may produce modest increases in the EC 50 for the coexpressed wild type receptor (Table I). The coexpression of a mutant receptor bearing E127A with the wild type receptor caused a ϳ13% shift to the left in the dose response to Ca 2ϩ o (Fig. 8).

DISCUSSION
Multiple forms of the CaR were detected by Western blotting of crude plasma membrane proteins prepared from HEK293 cells transiently transfected with the receptor that were similar to those seen in membranes prepared from bovine parathyroid gland. Three bands between 120 and 200 kDa correspond to three distinct forms of the monomeric receptor varying in their states of glycosylation. The lowest band, around 120 kDa, is a nonglycosylated form which co-migrated with the band isolated from tunicamycin-treated cells. The middle band is a high mannose form of the receptor that was sensitive to endoglycosidase H as well as to peptide-N-glycosidase F. The uppermost of the three bands most likely represents the fully glycosylated, mature form of the receptor, which was only sensitive to peptide-N-glycosidase F. In addition to the monomeric forms, high molecular mass species (around 350 kDa) were observed as well, with varied sensitivities to endoglycosidases. The nature of these high molecular mass species is not clear. They may represent an oligomeric form of the receptor that is present in the transfected cells and is of some physiological relevance or might simply be nonspecific aggregates that are artifacts of protein extraction and preparation for electrophoresis.
Western analysis of expressed receptors containing FHH/ NSHPT mutations provides some indirect evidence that the mature, glycosylated form of the receptor is important for full biological activity. For example, R66C, G143E, and E297K each had a substantially reduced amount of the mature, glycosylated form relative to the high mannose forms in comparison with those of the wild type receptor and the remaining mutants. Correspondingly, the maximal Ca 2ϩ i responses to CaR Crude plasma membrane proteins were isolated from HEK293 cells (grown in 100-mm Petri dishes) that were transiently transfected with 15 g of the wild type rHuPCaR4.0 or with cDNAs containing each of the inactivating mutations shown in Fig. 1. The protein samples, 4 g each, were subjected to SDS-PAGE on a linear gradient running gel of 4 -12% in the order (from left to right): mock-transfected (mock), wild type, R62M, R66C, T138M, G143E, R185Q, E297K, and R795W. The CaR proteins were stained with antireceptor antibody, 4641, as described in legends to the previous figures. The blot shown is representative of the pattern seen in three runs of each of two protein preparations from two independent transfections.

Expression of Ca 2ϩ o Receptor Mutations
agonists for these three mutants was severely attenuated. The reduced amount of the mutant receptors containing higher molecular weight, complex carbohydrates may have resulted from unsuccessful cellular translocation of the mutant receptor from endoplasmic reticulum to Golgi where the high mannose forms of glycoproteins including receptors are thought to be processed to more mature forms. In contrast, receptors containing the mutations R62M, E127A, T138M, R185Q, and R795W all showed expression patterns on Western blotting similar to that of the wild type receptor. Among these, R62M, E127A, T138M, and R185Q had maximal responses to Ca 2ϩ o , Gd 3ϩ o , or both that were similar to those of the wild type receptor. All inactivating mutations had varying shifts to the right in their dose response curves to agonists. In contrast, the activating mutation, E127A, caused an apparent increase in the affinity of the receptor for Ca 2ϩ o , with a shift to the left in its dose response curve.
These observations suggest that the mutations studied here within the extracellular domain, which reside exclusively within its first, amino-terminal half, modulate the affinity of the CaR for Ca 2ϩ o and other polycationic agonists without markedly affecting the capacity of the receptor to couple to intracellular signal transduction systems (in this instance phospholipase C). It is of interest in this regard that Hammerland et al. (20) have recently shown that chimeric receptors in which the extracellular domains of the CaR and the homologous mGluRs have been exchanged show specificity for activation of intracellular signaling that is determined by the extracellular amino terminus. That is, a receptor with the CaR extracellular domain and mGluR transmembrane and intracellular domains is activated by agonists of the CaR but not of the mGluRs. Moreover, the first half of the extracellular domain of the mGluRs has been shown to determine the specificity of the receptor for agonists that interact differentially with different forms of mGluRs (21). The mechanisms by which the amino acid substitutions in FHH/NSHPT and ADH affect binding of agonists to the CaR remain unclear. The mutations may indirectly affect the secondary structure of the CaR or its overall structure by replacing arginines or glutamates otherwise engaged in salt bridges. Alternatively, these mutations may disrupt a direct interaction of electron-rich atoms such as oxygen or nitrogen on their respective amino acid residues with divalent or trivalent cations.
Of the inactivating mutations that we characterized in this report, only R795W is predicted to reside inside the plasma membrane, whereas the rest are within the first half of the large amino-terminal, putatively extracellular domain. Two mutations, R185Q and R795W, had substantially attenuated responses to Ca 2ϩ o , although they exhibited normal patterns of receptor protein expression. However the dose response curves for these two mutants were different. R795W, despite having a markedly reduced maximal response to Ca 2ϩ o that was only 10 -15% of the maximal wild type receptor, nevertheless showed a dose-dependent increase in Ca 2ϩ i that plateaued at about 20 mM Ca 2ϩ o , at which concentration the response of the wild type receptor reaches its maximum (see Fig. 7A). These results suggest that this mutation produces a receptor with only a modest reduction in affinity for Ca 2ϩ o but with a marked impairment in its ability to elicit intracellular responses (such as G protein-coupling, activation of phospholipase C, etc.). In contrast, Ca 2ϩ o -elicited increases in Ca 2ϩ i for the receptor with the mutation R185Q did not reach its maximum even at 50 mM Ca 2ϩ o , which may indicate that the primary abnormality in receptor function is in ligand binding rather than signal transduction. This notion is supported by its nearly normal maximal response to Gd 3ϩ o , another CaR agonist. Interestingly mutant receptors, like those containing R66C, G143E, or E297K that had little or no response to Ca 2ϩ o , had some responses to Gd 3ϩ o , which are proportional to the amounts of their mature, fully glycosylated forms. This observation raises the possibility that the Ca 2ϩ o -sensing receptor may have low affinity sites for interaction with Gd 3ϩ o that differ from those for Ca 2ϩ o or that disruption of the binding of Ca 2ϩ o to a common binding site(s) for both ions does not necessarily interfere as much with the binding affinity for Gd 3ϩ o . Because the Gd 3ϩ o dose response curves for all of the FHH/ NSHPT mutants were right-shifted in proportion to the rightshift in their Ca 2ϩ o dose response curves, this would support the idea that Ca 2ϩ o and Gd 3ϩ o may share the same binding site(s). However, R62M and T138M had very similar EC 50 values for Gd 3ϩ o , which were only slightly higher than that of the wild type receptor, despite the 2-3-fold difference in the EC 50 values of R62M and T138M for Ca 2ϩ o . The latter observation may favor the hypothesis that discrete but interrelated sites exist in the receptor for the two ligands. Indeed, more direct support for this hypothesis has come from the preliminary studies of Hammerland et al. (20). These workers produced a truncated form of the CaR lacking essentially all of its amino-terminal extracellular domain. Although this receptor failed to respond to high levels of Ca 2ϩ o , it still showed a reasonably robust response to Gd 3ϩ o , consistent with the presence of a binding site for Gd 3ϩ o in some part of the receptor other than its extracellular amino terminus. Although the number of binding sites both within and outside of the amino i response of the wild type CaR to Gd 3ϩ o was around 1, suggesting that there is no cooperativity although there may still be multiple binding sites.
The coexpression of inactivating mutant receptors with the wild type receptor could provide a useful model that mimics the phenotype of FHH (2,(22)(23)(24)(25), assuming that the expression levels of both the wild type and mutant allele as well as the respective G proteins to which they couple are not too dissimilar from those that are present in affected tissues (i.e. parathyroid and kidney) in FHH patients in vivo. FHH is characterized by modest elevation in the serum calcium con-centration, ranging from 0.3 to 0.76 mM (Table I)  o at all, they might not compete with the wild type receptor for G proteins and other elements of the signaling systems activated by the CaR (i.e. phospholipase C). In these cases, the mild elevation of serum calcium may simply result from the decrease in the dose of the normal CaR gene. It is of note that mice heterozygous for targeted disruption of the CaR gene also show a modest increase in serum calcium concentration (26) as do affected members of a FHH family with a mutation that introduces a premature stop codon in the extracellular domain of the CaR just proximal to the first membrane spanning segment (22). Therefore, in both of the two latter cases, the complete lack of functional receptor contributed by one allele is associated with a mild phenotype, supporting a role for gene dosage in determining the level of the CaR on the parathyroid cell surface in vivo. The very small or absent shift in most of the dose response to Ca 2ϩ o in our coexpression studies is most likely due to the fact that the number of CaRs on the cell surface of the HEK293 cells is not a limiting factor for activation of phospholipase C.
FHH patients harboring the mutations with patterns for protein expression similar to that of the wild type receptor (such as R62M, T138M, and R185Q but with the exception of R795W), on the other hand, have elevations of serum calcium that vary in proportion to their EC 50 values when expressed transiently in HEK293 cells (see Table I and Fig. 7A). Among these, R185Q showed a prominent "dominant negative" effect on the coexpressed wild type receptor, which may contribute to the unusual elevation in the serum calcium concentration in o were performed in a solution without BSA as described under "Experimental Procedures." All responses are normalized to the maximum response of the wild type receptor (rHuPCaR4.0). Each data point is the mean value of 3-12 measurements (see Table I). The standard error of the mean (S.E.) is indicated with a vertical bar through each point. Some error bars are smaller than the symbol. In this experiment, 2.5 g of each cDNA was used to transfect HEK293 cells plated on a rectangular coverslip (in an individual well of 12-well plates).

FIG. 8. High Ca 2؉
o -evoked increases in Ca 2؉ i in fura-2-loaded HEK293 cells transiently transfected with the rHuPCaR4.0 (wild type) cDNA or a receptor cDNA containing the activating mutation, E127A, as well as in HEK293 cells cotransfected with the wild type and E127A cDNAs (E127A/wt) in a solution containing 0.1% BSA (see "Experimental Procedures" for details). Each data point is the mean value of 10 or 11 measurements. All responses are normalized to the maximum response of the wild type receptor (rHuPCaR4.0). The standard error of the mean (S.E.) is marked with a vertical bar through each point. The EC 50 for each curve is presented as mean Ϯ S.E. The means with different superscripts are significantly (p Յ 0.05) different. In this experiment, 0.625 g of each cDNA was employed to transfect HEK293 cells plated on a rectangular coverslip (within individual wells of 12-well plates).

Expression of Ca 2ϩ o Receptor Mutations
affected members of the family harboring this mutation (Table  I). However, it is not clear how R185Q exerts this effect on the wild type receptor. A possible explanation is that the presence of relatively inactive, mutant receptors on the cell surface reduces the number of G proteins available to bind to the wild type receptor prior to activation. Analogously, R795W produced a relatively mild dominant negative effect on the coexpressed wild type receptor, causing about a 0.6 mM elevation in the serum calcium concentration of affected family members. Because the receptor with the R795W mutation could potentially still bind to its respective G protein(s) but fail to activate them, the availability of G proteins for the wild type allele of CaR in FHH patients with this mutation may be less than that in those with other mutations, whereas the number of normal receptors and/or G proteins might not be as limited in HEK293 cells, which show less increase in EC 50 than expected. The ADH mutation, E127A, on the other hand, increases the apparent affinity of coexpressed wild type receptor by about 13%, similar in magnitude to the reduction (0.41 Ϯ 0.04 mM, n ϭ 13) in the serum calcium concentration (2.1 Ϯ 0.1 mM, n ϭ 8) in affected ADH family members who harbor this mutation (3). Thus the reciprocal changes in the EC 50 values for Ca 2ϩ oevoked increases in Ca 2ϩ i in the expressed mutant receptors containing FHH/NSPTH or ADH mutations, which parallel the alterations observed for affected individuals in vivo, provided additional evidence for both the physiological relevance of the receptor under normal circumstances and in the pathogenesis of these disorders, although the EC 50 of the wild type receptor for Ca 2ϩ o -mediated increases in Ca 2ϩ i is 3-4-fold higher than that for Ca 2ϩ o -regulated PTH release in vivo or in vitro (27). Likewise, the EC 50 for the activation of intracellular signaling pathways by a variety of hormones is usually severalfold to manyfold higher than that for the regulation of the biological response of a given tissue (27). Finally, further characterization and classification of FHH and ADH mutations may provide useful insights into the structure-function relationships of CaR. i . At each arrowhead, the concentration of Ca 2ϩ o was increased to the indicated millimolar value by adding a small volume of concentrated CaCl 2 stock solutions to the bath solution containing 0.1% BSA. Each of the first five additions resulted in a 1% change in the buffer volume in the cuvette, and each subsequent addition produced an additional 0.5% change. The results shown are representative of the patterns seen in 12 transfections with three different DNA preparations for R185Q. In this experiment, 0.625 g of each DNA was used to transfect HEK293 cells plated on a rectangular coverslip (within individual wells of 12-well plates). A, transfection with rHuPCaR4.0 (wild type) alone. B, cotransfection of a CaR cDNA containing the inactivating mutation, T138M, with wild type receptor; the results of this cotransfection experiment are representative of those obtained with mutant receptors having little or no effect on the response of the wild type receptor during cotransfection. C, cotransfection of a CaR cDNA containing the inactivating mutation, R185Q, with wild type receptor. The mean EC 50 for each cotransfection is reported in Table I.