International Union of Basic and Clinical Pharmacology. CVIII. Calcium-Sensing Receptor Nomenclature, Pharmacology, and Function

A. Endogenous and Exogenous Agonists

B. Endogenous and Exogenous Allosteric Modulators

Allosteric modulators bind to sites that are topographically distinct from the orthosteric binding site and act to either potentiate [positive allosteric modulators (PAMs)], inhibit [negative allosteric modulators (NAMs)], or have no effect on (neutral allosteric ligands) the binding or efficacy of the orthosteric agonist. Allosteric modulators may also be agonists (or inverse agonists) in the absence of orthosteric agonists and can simultaneously act as agonists and PAMs (PAM-agonists). CaSR PAMs have been termed type II calcimimetics and CaSR NAMs calcilytics.

C. Biased Agonism and Biased Allosteric Modulation

Given that the CaSR responds to a diverse array of different ligands, it is unsurprising that the CaSR is subject to biased agonism and biased modulation. Biased agonism is the phenomenon by which distinct ligands stabilize preferred GPCR signaling states, with each state having the potential to stimulate or inhibit discrete subsets of the full repertoire of intracellular signaling pathways that couple to a given receptor (Kenakin and Christopoulos, 2013). This is in contrast to the earlier dogma that all agonists activate the same subsets of GPCR signaling pathways to greater (e.g., full agonists) or lesser (e.g., partial agonists) extents. Similarly, biased modulation arises when an allosteric ligand differentially modulates different agonist-mediated signaling pathways.

For instance, in CaSR-HEK293 cells, Ca²⁺_o preferentially mediates stimulation of Ca²⁺_i mobilization over pERK1/2, whereas spermine preferentially activates pERK1/2 (Thomsen et al., 2012a). Similarly, L-amino acids activate Ca²⁺_i mobilization and ERK phosphorylation (Lee et al., 2007) and also inhibit cAMP synthesis. However, they are inactive in stimulating phosphatidylinositol-PLC and various other signaling events, including Rho-dependent actin stress fiber formation (Davies et al., 2006) and cAMP responsive element-binding protein phosphorylation (Avlani et al., 2013), and appear to promote Ca²⁺_i mobilization via a G_12/13/transient receptor potential cation 1–dependent Ca²⁺_o influx pathway (Rey et al., 2005, 2006).

Evidence from patients with FHH suggests CaSR bias may arise in part from spatial and temporal CaSR-signaling patterns. Loss-of-function germline mutations of the adaptor-related protein complex-2 (AP2)-S1 gene, which encodes the sigma subunit of the heterotetrameric AP2σ, cause FHH3 (Nesbit et al., 2013b; Hannan et al., 2015a). AP2σ forms part of the heterotetrameric AP2 that plays a critical role in clathrin-mediated endocytosis. AP2σ mutations increase CaSR cell surface expression yet reduce CaSR signaling because CaSR residency time in clathrin-coated pits is increased, consequently impairing CaSR G_q/11 signaling from endosomes (Gorvin et al., 2018c). In contrast, G_i/o-mediated signaling is less sensitive to AP2σ mutations. Thus, whereas the plasma membrane localized CaSR signals via G_q/11 and G_i/o, endosomal CaSRs signal predominantly via G_q/11 (Gorvin et al., 2018c).

It must be noted that many of the studies reporting differential CaSR-mediated pathway activation have not been performed in a systematic manner using identical conditions across assays (e.g., buffers, duration of agonist stimulation, etc.) or the same cellular background. Furthermore, bias has not been quantified in these studies. Therefore, it remains to be definitively proven whether biased agonism is truly operative at the CaSR or whether previous observations were due to observational bias (e.g., different assay conditions, different cell types) or system bias (e.g., the relative efficiency with which the receptor couples to different pathways).

Nonetheless, small-molecule allosteric modulators do appear to exhibit true biased modulation at the CaSR. Evidence of biased modulation comes from reversals in the magnitude of cooperativity in different pathways between distinct PAMs or NAMs or from differences in PAM or NAM affinity for receptor states that couple to different signal transducers. For instance, although cinacalcet and NPS 2143 preferentially potentiate or inhibit, respectively, Ca²⁺_o-mediated Ca²⁺_i mobilization over pERK1/2, AC265347 and R,R-calcimimetic B show reversed bias for CaSR-mediated pERK1/2 over Ca²⁺_i mobilization (Cook et al., 2015; Leach et al., 2016; Diepenhorst et al., 2018). Similarly, AC265347, NPS R-568, and calindol, but not cinacalcet or R,R-calcimimetic B, have a higher functional affinity (i.e., an affinity quantified in a functional assay using an operational model of allosterism (Leach et al., 2007)) for the CaSR state that signals to IP₁ accumulation versus Ca²⁺_i mobilization (Cook et al., 2015; Diepenhorst et al., 2018), whereas cinacalcet, NPS R-568, and NPS 2143 all have a higher functional affinity for the CaSR state that couples to membrane ruffling (Davey et al., 2012).

Evidence for small-molecule PAM and NAM bias also comes from pharmacochaperone studies, which reveal that although cinacalcet, AC265347, and BTU compound 13 are all PAMs in multiple CaSR-mediated signaling assays, only cinacalcet positively modulates the trafficking of an endosomally-trapped, naturally occurring mutant CaSR, rescuing its cell surface expression back to levels comparable to wild-type CaSR (Leach et al., 2013; Cook et al., 2015; Diepenhorst et al., 2018). In contrast, although NPS 2143 is a NAM of CaSR signaling, it is a PAM of loss-of-expression mutant receptor trafficking (Leach et al., 2013). This is in contrast to the actions of NPS 2143 at the wild-type CaSR, wherein it reduces CaSR surface expression (Huang and Breitwieser, 2007), suggesting naturally occurring mutations (which cause Ca²⁺_o homeostasis disorders; see VI. Calcium-Sensing Receptor–Related Genetic Diseases and Therapeutic Interventions) may engender bias in CaSR function. Indeed, Ca²⁺_o-mediated bias toward Ca²⁺_i mobilization is abolished by some naturally occurring mutations (Leach et al., 2012).

Although the physiological relevance of biased agonism and biased modulation at the CaSR is not at present known, differences in the propensity of CaSR PAMs to cause hypocalcemia could be linked to this phenomenon. For instance, as already mentioned, R,R-calcimimetic B and AC265347 are effective suppressors of PTH release. However, in comparison with cinacalcet, R,R-calcimimetic B and AC265347 demonstrate reduced propensity to cause hypocalcemia in rats successfully treated for severe hyperparathyroidism induced by CKD (R,R-calcimimetic B) or in normal rats (AC265347). The reduced incidence of hypocalcemia with R,R-calcimimetic B and AC265347 is presumably linked, in part, to their lower potency and efficacy for the stimulation of calcitonin secretion versus suppression of PTH release (Henley et al., 2011; Ma et al., 2011). Importantly, although suppression of PTH release has been associated with pERK1/2, calcitonin release is independent of pERK1/2 in rat medullary thyroid carcinoma cells (Thomsen et al., 2012b). This highlights differences in the coupling specificity of the CaSR in distinct tissues and is consistent with observations that when compared with cinacalcet, AC265347 and R,R-calcimimetic B show reversed bias for CaSR-mediated pERK1/2 over Ca²⁺_i mobilization.

Another apparent difference between CaSR PAMs points toward putative clinical advantages for cinacalcet. The CaSR agonist Sr²⁺ reduces the differentiation of bone-resorbing osteoclasts (Bonnelye et al., 2008) and stimulates osteoclast apoptosis (Hurtel-Lemaire et al., 2009) (described in V. (Patho)physiology of the Calcium-Sensing Receptor and Its Ligands). In cultured osteoclasts differentiated from human CD14+ monocytes, although cinacalcet potentiated Sr²⁺-mediated tartrate-resistant acid phosphatase expression (a marker of osteoclast activity) and robustly inhibited osteoclast-mediated hydroxyapartite artificial bone resorption, AC265347 and BTU compound 13 were without effect in these two assays (Diepenhorst et al., 2018). Although it is not clear whether differences in the biased profile of AC265347 and BTU compound 13 versus cinacalcet are responsible for their distinct PAM activities in osteoclasts, it is interesting that only cinacalcet, and not AC265347 or BTU compound 13, can pharmacochaperone loss-of-function mutant CaSRs potentially via differential stabilization of different conformations of the CaSR. A more detailed understanding of the signaling and trafficking pathways that couple the CaSR to its many physiological responses will aid our understanding of why the CaSR responds to so many endogenous activators and may facilitate the development of biased compounds with improved, tissue-specific effects.

In addition to bias engendered by small-molecule allosteric modulators, CaSR autoantibodies that cause acquired hypocalciuric hypercalcemia can act as biased allosteric modulators. Biased autoantibodies directed against the CaSR VFT can potentiate IP accumulation while inhibiting pERK1/2 generation (Makita et al., 2007; Makita and Iiri, 2014), whereas others inhibit pERK1/2 generation but have no effect on IP accumulation (Pallais et al., 2011). Importantly, cinacalcet corrected the severe hypercalcemia associated with acquired hypocalciuric hypercalcemia caused by a biased autoantibody (Makita et al., 2019). Taken together, these findings once again highlight how bias and allostery are key features of CaSR (patho)physiology and drug actions.

A. Calcium-Sensing Receptor Extracellular Domain

3. Cation Binding Sites

In both crystal structures of the CaSR VFT domain, cation binding sites were identified, but these sites differed in their number, location (with the exception of cation binding site 1), and the cation that was bound to each site (Fig. 3).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3.

Binding sites within the CaSR crystal structures. ECD conformations of the (A) active (PDB 5K5S) and (B) inactive (PDB 5K5S) calcium-bound structures and the VFT conformations of the (C) active Mg²⁺-bound (PDB 5FBK) and (D) active Mg²⁺- and Gd³⁺-bound (PDB 5FBH) structures. Crystal structures are shown as cartoon ribbon within the transparent molecular surface and colored as in Fig. 2. Hydrogen bond interactions (dashed lines) of calcium (red spheres), Mg²⁺ (green spheres), and Gd³⁺ (yellow spheres) with key residues or water molecules (black spheres) are shown for each proposed binding site.

Anomalous difference mapping indicated four Ca²⁺_o binding sites in the VFT structure solved by Geng et al. (2016) (Fig. 3). In lobe 1 of the active (L-Trp bound and closed) VFT conformation, backbone carbonyl oxygen atoms of Ile81, Ser84, Leu87, and Leu88 coordinate Ca²⁺_o binding at cation binding site 1 (PDB: 5K5S). There was no Ca²⁺_o coordinated at cation binding site 1 in the inactive structure (PDB: 5K5T), even though this site is not significantly different in the active versus inactive structures (Geng et al., 2016). As such, it is possible that Ca²⁺_o, which was used at a lower concentration in the crystallization conditions for the inactive structure, could bind to this site without the need for the VFT domain to be closed.

Cation binding site 2 is located adjacent to the L-Trp binding site above the interdomain cleft in lobe 1 of the VFT. Cation binding site 2 is occupied by Ca²⁺_o in both the inactive and active structures, in which Ca²⁺_o is coordinated by the hydroxyl group of Thr100 in both states and by the carbonyl of Asn102 via a water molecule in the active structure. Thr145 also lines cation binding site 2 and forms part of the L-Trp binding cleft in the active state (Geng et al., 2016).

The hydroxyl groups of Ser302 and Ser303 coordinate cation binding site 3, either directly or indirectly through water molecules, at the edge of the interdomain cleft of lobe 2. The closing of lobe 1 and lobe 2 of the VFT is facilitated by Ca²⁺_o stabilization of a conformation that permits an interdomain hydrogen bond interaction between lobe 1 residue Arg66 and lobe 2 residue Ser301 (Geng et al., 2016).

Finally, upon agonist binding, cation binding site 4 forms part of the homodimer interface bridging the lobe 2 domain of one subunit and the CR domain of the second subunit. Three interfacial residues, the carboxylate group of Asp234, and carbonyl oxygen of Glu231 and Gly557, coordinate Ca²⁺_o binding to site 4 (Geng et al., 2016).

The anomalous difference map intensities varied at each of the Ca²⁺_o binding sites, where intensity was ranked as Ca²⁺_o binding site 1 = 2 > 3 > 4. The lower anomalous signal for Ca²⁺_o in sites 3 and 4 indicates incomplete occupancy or higher flexibility at these positions in the crystal lattice. The authors suggested the lower signal reflects a lower Ca²⁺_o affinity at these sites. In support of a lower Ca²⁺_o binding affinity for cation binding site 4, the authors proposed that Ca²⁺_o binding at site 4 stabilizes the active homodimer conformation, and thus the site is occupied only at elevated concentrations required for receptor activation (Geng et al., 2016).

In contrast to the structures by Geng et al. (2016), Zhang et al. (2016) identified two cation binding sites in their active VFT structures. Electron density and geometric restraints were used to identify Mg²⁺ occupying these cation binding sites, one of which overlapped with cation binding site 1 in the structure by Geng et al. (2016). However, in contrast to the unoccupied cation binding site 1 in the inactive structure by Geng et al. (2016), cation binding site 1 was occupied by Mg²⁺ in the inactive structure by Zhang et al. (2016). The Mg²⁺ is coordinated by Ser84 and backbone interactions with Ile81, Ile87, and Leu88, in addition to two water molecules. This site is similarly occupied by a Mg²⁺ cation in the rat mGlu₁ VFT structure (Kunishima et al., 2000).

The second Mg²⁺ binding site (cation-binding site 5) is located at the dimerization interface of lobe 2 and is coordinated through Ser240 and four water molecules (Zhang et al., 2016). The highly conserved residues Glu228 and Glu231 from one protomer and Glu241 from the other protomer surround this site.

Anomalous difference maps identified a Gd³⁺ binding site (cation binding site 6) coordinated by Glu232, Glu228, and Glu229 adjacent to cation binding site 5 on the lobe 2 dimerization interface (PDB: 5FBN) (Zhang et al., 2016). The Glu228Ile and the double mutant Glu228Ile/Glu229Ile have previously been shown to reduce Mg²⁺-induced Ca²⁺_i mobilization; therefore, other cations could bind here (Huang et al., 2009).

The crystal structures of the ECD suggest that Ca²⁺_o and other cations play a role in: 1) local stabilization of the CaSR ECD; and 2) activation of the receptor via stabilization of the homodimer through cation binding at sites 4-6 (Jensen et al., 2002; Geng et al., 2016; Zhang et al., 2016). It is unknown whether Ca²⁺_o alone can activate the receptor or whether it requires the presence of the cleft-binding ligands. Although Geng et al. (2016) obtained an active (closed) structure in the absence of amino acids, an unidentified continuous stretch of density in the conserved interdomain cleft was observed, which could be attributed to an endogenous ligand or a ligand acquired during the crystallization process. If ligands that bind the conserved interdomain cleft are difficult to remove during crystallography studies, it is likely that these same ligands are present during in vitro assays that measure CaSR activation. Furthermore, cations identified in the crystal structures could be artifacts of the crystallization conditions and merely stabilize the crystal contacts required for structure determination. Although mutagenesis was used to corroborate the observed cation binding sites (Geng et al., 2016; Zhang et al., 2016), these mutational studies neither accounted for mutation-induced changes in receptor expression nor quantified changes in cation affinity and efficacy. Therefore, a reduction in cation binding upon mutation of these sites has not been validated. Furthermore, analysis of Ca²⁺_o-binding proteins to predict the CaSR’s Ca²⁺_o sites, coupled with mutagenesis and spectroscopic techniques to validate the predictions, confirmed multiple VFT Ca²⁺_o binding sites, but they differed to those identified in the VFT crystal structures (Huang et al., 2007, 2009; Kirberger et al., 2008; Wang et al., 2009, 2010; Zhao et al., 2012). Moreover, analyses of a “headless” CaSR, in which the ECD has been removed, has shown that Ca²⁺_o can also activate the CaSR via binding sites in the 7TM domain (see IIIB. Calcium-Sensing Receptor 7-Transmembrane Domain) (Ray and Northup, 2002; Leach et al., 2016). Accordingly, the cooperative binding of Ca²⁺_o at multiple binding sites likely maximizes the CaSR’s ability to respond to Ca²⁺_o over a narrow physiological range. Additional active-state structures, biophysical studies, and mutagenesis work are required to fully understand how these sites interact.

Site-directed mutagenesis and functional studies show that Ca²⁺_o and L-amino acids potentiate each other’s activity in a positively cooperative manner (Conigrave et al., 2000; Zhang et al., 2002b, 2014a,b; Mun et al., 2005). Under physiological conditions, L-amino acids potentiate Ca²⁺_o potency for evoking intracellular responses (Conigrave et al., 2007), and mutating residues important for L-amino acid binding eliminated L-Phe potentiation of Ca²⁺_i mobilization (Conigrave et al., 2000; Zhang et al., 2002a). The ability of Ca²⁺_o and L-amino acids to cooperatively activate the CaSR was further demonstrated using saturation transfer difference NMR (Zhang et al., 2014b). Using saturation transfer difference NMR, L-Phe was estimated to bind to the CaSR with an affinity of ∼10 mM in the absence of Ca²⁺_o, whereas in the presence of Ca²⁺_o, L-Phe affinity was increased. Similarly, and as expected for reciprocal cooperativity, the binding affinity of Ca²⁺_o in the presence of 10 mM L-Phe was increased. Therefore, dual binding of Ca²⁺_o and amino acids enhances the sensitivity of the CaSR to changes in concentrations of these ligands.

B. Calcium-Sensing Receptor 7-Transmembrane Domain

C. Calcium-Sensing Receptor Dimerization

Like all class C GPCRs, CaSR dimerization is a key feature governing receptor function. The dominant interaction underpinning the CaSR dimer is two covalent disulfide bonds formed at the top of lobe 1 of the VFT domains between Cys129 and Cys131 (Ray et al., 1999). However, the CaSR is not dependent on the disulfide links for activity, as is evidenced by mutation of these residues to Ser, which does not alter surface expression or Ca²⁺_o potency in vitro (Fan et al., 1998; Zhang et al., 2001).

Dimerization influences allosteric modulation at the CaSR. For instance, negative allosteric modulators must bind both protomers to block signaling, whereas PAMs only need occupy one protomer to exert their full modulatory effect (Hauache et al., 2000; Jacobsen et al., 2017; Gregory et al., 2018). This feature likely reflects agonist-mediated signal transmission through the CaSR, which occurs across the dimer rather than propagating through a single protomer (Hauache et al., 2000). Consequently, transactivation across the dimer can result in unique pharmacology for CaSR allosteric modulators. An example is calhex 231, which shows positive allosteric activity when bound to the allosteric site in only one protomer but shows negative allosteric activity when occupying both the allosteric sites of the dimer (Gregory et al., 2018).

Immunoprecipitation data have demonstrated that the CaSR forms heterodimers in vitro with mGlu_1/5 or the GABA_B receptor, with heterodimers detected in bovine and mouse brain lysates, respectively (Gama et al., 2001; Chang et al., 2007). On the other hand, fluorescence resonance energy transfer studies have revealed that the CaSR does not heterodimerize with its closest receptor homolog, the GPRC6A receptor (Jacobsen et al., 2017). Heterodimerization may facilitate the varied functional roles of the CaSR in different tissues, particularly in the brain, wherein the expression of the GABA_B receptor regulates CaSR expression and vice versa (discussed in VK. Calcium-Sensing Receptor in the Brain and Nervous System).

D. Calcium-Sensing Receptor Glycosylation

The CaSR VFT domain contains 11 potential N-linked glycosylation sites; however, not all of these sites have been experimentally verified. The CaSR is glycosylated in the endoplasmic reticulum with mannose (immature) carbohydrate prior to mature complex glycosylation processing in the Golgi. Disruption of at least three glycosylation sites can impair receptor processing and cell surface expression (Ray et al., 1998). Eight glycosylation sites (Asn90, Asn130, Asn261, Asn287, Asn446, Asn468, Asn488, and Asn541) have been experimentally validated, whereas questions remain over the three remaining sites (Asn386, Asn400, and Asn594). Notably, Asn594 was glycosylated in the solved X-ray crystal structure, whereas Asn386 was mutated to Gln to prevent glycosylation and aid crystallization. However, it is unclear whether this observation at a truncated CaSR sample reflects the glycosylation arrangement of the full-length CaSR. Importantly, the functional role of glycosylation beyond controlling surface expression needs further investigation.

A. Phosphorylation and Dephosphorylation

PKC-mediated phosphorylation of the CaSR provides a rapid and quickly reversible mechanism for inhibiting receptor activity. Indeed, treatment of parathyroid cells with PKC-activating phorbol esters overcomes the inhibitory effect of Ca²⁺_o on PTH release (Brown et al., 1984; Nemeth et al., 1986). When first cloned, the CaSR was predicted to contain five PKC consensus motifs, although 54 serine and threonine residues reside in the receptor’s C-terminal tail and ICLs (Garrett et al., 1995a). However, the key inhibitory phosphorylation site is Thr888 in the C-terminal tail (Bai et al., 1998b). Thr888 is most likely phosphorylated by PKCα (Young et al., 2014) and dephosphorylated by a calyculin A–sensitive protein phosphatase (McCormick et al., 2010).

The functional importance of inhibitory Thr888 phosphorylation is most apparent with the clinical mutant, Thr888Met, which cannot be phosphorylated. In vitro Thr888Met is a gain-of-function mutant, whereas it suppresses PTH secretion in vivo, resulting clinically in ADH (see VIC. Autosomal Dominant Hypocalcemia and Bartter Syndrome Type V) (Lazarus et al., 2011). Therefore, CaSR phosphorylation contributes significantly to CaSR activity in vivo and thus to the overall control of PTH secretion and Ca²⁺_o homeostasis [reviewed in Conigrave and Ward (2013)].

The nonphosphorylatable mutant, Thr888Val, also produced a significant gain of function, which was not further enhanced by comutating the other four predicted PKC sites (Bai et al., 1998b). However, PKC inhibition at the wild-type CaSR resulted in a greater gain of function than produced at the Thr888Val mutant, thus it appeared likely that another unknown site may be phosphorylated in tandem with Thr888. However, the identity of this site has remained elusive. In mGlu₅, the key PKC phosphorylation site, Ser839 (Kim et al., 2005), aligns not with Thr888 in the CaSR but with Ser875, a residue not originally predicted to be phosphorylated by PKC (Garrett et al., 1995a). Intriguingly, current data indicate removal of this putative phosphorylation site from the CaSR (Ser875Ala) also produces a gain of function, similar to that of Thr888Ala, whereas a phosphomimetic mutation at this site (Ser875Asp) produces a loss of function (Binmahfouz et al., 2019). The double Ser875Ala plus Thr888Ala mutant exhibits a greater gain of function than Ser875Ala alone, and concomitant PKC inhibition exerts no further signal enhancement. Thus, Ser875 is most likely the second major inhibitory PKC site in the CaSR (Binmahfouz et al., 2019).

Ca²⁺_o induces biphasic concentration-dependent phosphorylation of Thr888 in CaSR-HEK cells, with 0.5-2.5 mM Ca²⁺_o eliciting increased Thr888 phosphorylation after 10 minutes, whereas 2.5-5 mM Ca²⁺_o decreases phosphorylation apparently by activating a calyculin A-sensitive protein phosphatase (McCormick et al., 2010). The decrease in Thr888 phosphorylation mediated by 2.5-5 mM Ca²⁺_o occurs at the same Ca²⁺_o concentrations that elicit sustained, as opposed to oscillatory, Ca²⁺_i mobilization. Consistent with this, the Thr888Ala mutant is not only gain-of-function but also exhibits less oscillatory and more sustained Ca²⁺_i mobilization, as does the wild-type CaSR when cotreated with a PKC inhibitor (Davies et al., 2007). Furthermore, PKC-dependent phosphorylation of Thr888 attenuates L-amino acid–dependent signaling in a manner similar to its effect on Ca²⁺_o (Bai et al., 1998b; McCormick et al., 2010). Since PKC-mediated Thr888 phosphorylation is thus a critical regulator of CaSR function, differential CaSR phosphorylation could provide a mechanism to permit biased signaling in different cells or in response to various agonists.

CaSR signaling is also modulated by the GPCR kinases (GRKs). Specifically, overexpression of GRK2 and GRK3 decreases CaSR-induced IP formation in a HEK-derived cell line by >70% (Lorenz et al., 2007). Mutating GRK2 so that it could no longer bind G_q overcame the inhibitory effect of GRK2 on CaSR signaling, indicating that GRK2 inhibition of CaSR signaling might be caused by sequestering of G_q rather than by phosphorylation of the CaSR. Overexpression of either β-arrestin 1 or β-arrestin 2 partly inhibits CaSR-induced IP production, and this effect was abolished by deleting all five of the predicted PKC sites as identified by Bai et al. (1998b) (Lorenz et al., 2007).

B. Internalization and Agonist-Driven Insertional Signaling

In heterologous expression systems, the CaSR undergoes constitutive internalization (Reyes-Ibarra et al., 2007; Gorvin et al., 2018c; Mos et al., 2019) and agonist-induced internalization (Lorenz et al., 2007; Zhuang et al., 2012; Nesbit et al., 2013b). Furthermore, CaSR internalization is increased by the CaSR PAM, NPS R-568, and agonist-induced but not constitutive internalization is inhibited by the NAM, NPS 2143 (Mos et al., 2019). GPCR internalization usually involves desensitization by kinase phosphorylation and subsequent β-arrestin binding followed by recruitment to clathrin-coated pits by β-arrestin and the clathrin-binding AP2 heterotetramer (Hanyaloglu and von Zastrow, 2008). As described above, phosphorylation by PKC and GRKs, or β-arrestin recruitment, is involved in CaSR desensitization (Pi et al., 2005b; Lorenz et al., 2007). Similarly, agonist-induced CaSR internalization requires β-arrestin, which is in contrast with the GABA_B and mGluRs, which function independently of β-arrestin (Pin and Bettler, 2016). However, constitutive and agonist-induced CaSR internalization is largely independent of G_q/11 and G_i/o in HEK293 cells (Mos et al., 2019), thus indicating that G protein–mediated activation of PKC is not involved in this cell line. Studies of the internalization mechanisms of endogenously expressed CaSRs in nonrecombinant cells are still lacking because of the technical difficulty of performing such experiments.

The CaSR is also predicted to couple directly to AP2σ through a dileucine motif in the CaSR C-terminal tail (Nesbit et al., 2013b). Similar to loss-of-function mutations in the CASR or GNA11 genes (Pollak et al., 1993; Nesbit et al., 2013a), mutations that disrupt the CaSR interaction with AP2σ reduce the sensitivity of CaSR-expressing cells to Ca²⁺_o (Nesbit et al., 2013b). Similarly, germline mutations of the AP2S1 gene that lead to alteration of Arg15 in AP2σ cause FHH3 (Nesbit et al., 2013b), which is clinically the most severe of the three FHH types (Hannan et al., 2015a) (VI. Calcium-Sensing Receptor–Related Genetic Diseases and Therapeutic Interventions). AP2σ Arg-15 mutations inhibit CaSR internalization (Nesbit et al., 2013b; Gorvin et al., 2018c), and the functional similarity between loss-of-function mutations in the CASR, GNA11, and AP2S1 genes shows a close relationship between internalization and CaSR signaling. This relationship could be explained by reduced resensitization and/or intracellular signaling when internalization is inhibited (Reyes-Ibarra et al., 2007; Zhuang et al., 2012; Gorvin et al., 2018c).

After internalization, GPCRs are either resensitized and recycled to the cell surface or degraded (Hanyaloglu and von Zastrow, 2008). Cell surface expression of CaSR is constant under basal conditions (Reyes-Ibarra et al., 2007; Zhuang et al., 2012), which means constitutively internalized receptors are replaced. In heterologous cells, internalized CaSR is recycled through Rab11a-dependent slow-recycling endosomes (Reyes-Ibarra et al., 2007) to be sorted to lysosomes for degradation (Grant et al., 2011; Zhuang et al., 2012). The CaSR’s C-terminal tail is involved in postendocytic sorting, as deletion of residues 920–970 increased cell surface expression and reduced colocalization with a lysosomal marker (Zhuang et al., 2012). Similarly, overexpression of associated molecule with the SH3 domain of signal-transducing adapter molecule, which interacts with the CaSR C-terminal tail, downregulated cell surface CaSR (Herrera-Vigenor et al., 2006; Reyes-Ibarra et al., 2007).

The interaction of 14-3-3 proteins with an arginine-rich motif in the CaSR C-terminal tail partly retains an intracellular CaSR pool (regulated by Ser899 phosphorylation) (Stepanchick et al., 2010; Grant et al., 2011, 2015), but the CaSR is upregulated at the cell surface upon agonist stimulation via a G_q/11-dependent mechanism called agonist-driven insertional signaling (Grant et al., 2011; Gorvin et al., 2018c). This process involves rapid mobilization of the intracellular pool of receptors to the cell surface and initiation of receptor synthesis to support prolonged upregulation (Grant et al., 2011, 2015). The rapid increase in cell surface receptors is proposed to support the high sensitivity of the CaSR to increases in Ca²⁺_o.

A. Calcium-Sensing Receptor in the Parathyroid Glands

CaSR expression appears during parathyroid development in response to key parathyroid-determining genes, including GCM2 (encoding Gcmb) and SHH (encoding the inhibitory controller, Sonic Hedgehog) (Grevellec et al., 2011). Consistent with a direct connection between Gcmb and CaSR expression, GCM2 control elements have been identified in the CASR promoters (Canaff et al., 2009), and short hairpin RNA directed against GCM2 in parathyroid cell cultures suppressed the protein levels of Gcmb and the CaSR (Mizobuchi et al., 2009).

The CaSR’s nonredundant roles in Ca²⁺_o metabolism have been clearly established by the hypercalcemic disorders, neonatal severe hyperparathyroidism (NSHPT) and FHH, and the hypocalcemic disorders, ADH and Bartter syndrome type V. These are discussed together with animal models of loss- and gain-of-function mutations of the CaSR in VI. Calcium-Sensing Receptor–Related Genetic Diseases and Therapeutic Interventions.

The CaSR negatively controls parathyroid function by suppressing acute PTH secretion primarily from chief cells [review: (Conigrave, 2016)], inhibiting cell proliferation and thus cell number and gland size (Fan et al., 2018), and reducing PTH gene transcription [review: (Chen and Goodman, 2004)]. It also activates the local synthesis, particularly in parathyroid oxyphil cells, of 1,25(OH)₂D₃ (Ritter et al., 2012), a recognized inhibitor of PTH synthesis. The CaSR’s effects on cell proliferation are particularly noticeable in the context of primary hyperparathyroidism (e.g., due to adenomatous disease) or hyperplasia in the context of CKD. Interestingly, in parathyroid adenoma and CKD, cellular CaSR expression is reduced (Kifor et al., 1996). Nonetheless, in patients with CKD and in rat models of secondary hyperparathyroidism, sustained treatment with cinacalcet suppresses parathyroid gland size as well as serum PTH levels (Colloton et al., 2005; Yamada et al., 2015). Similarly, exposure of parathyroid cells to cinacalcet in vitro suppresses proliferation and promotes apoptosis (Tatsumi et al., 2013).

The parathyroid CaSR continuously monitors the Ca²⁺_o concentration as well as various other stimuli that affect CaSR function, including the plasma levels of L-amino acids (Conigrave et al., 2004), pH (Campion et al., 2015), ionic strength (Quinn et al., 1998), and, perhaps, locally generated polyamines (Quinn et al., 1997) (see II. Agonists and Allosteric Modulators). CaSR activity in the parathyroid glands is resistant to desensitization, in part because of efficient receptor recycling as well as a large intracellular receptor pool that undergoes a high rate of trafficking from the endoplasmic reticulum and Golgi to the plasma membrane [reviews: (Breitwieser, 2013; Ray, 2015)]. Whether agonist-driven insertional signaling operates in parathyroid glands is unknown, but the CaSR interacts with a signaling assembly dependent on caveolin-1 (Kifor et al., 1998) and undergoes AP2σ-regulated endocytosis (Nesbit et al., 2013b; Gorvin et al., 2018c).

B. Calcium-Sensing Receptor in the Thyroid Gland

In the thyroid the CaSR is expressed at high levels in a relatively small subpopulation of cells, the parafollicular C cells (Garrett et al., 1995b). In C cells, the CaSR acts to promote secretion of the peptide hormone calcitonin. Evidence that the CaSR stimulates calcitonin secretion is supported by studies in CaSR knockout mice, in which plasma calcitonin levels were suppressed (Fudge and Kovacs, 2004; Kantham et al., 2009). Thus, elevated Ca²⁺_o stimulates calcitonin release, which, in turn, lowers the plasma calcium level, primarily by suppressing bone resorption. Both the CaSR and calcitonin (or calcitonin gene–related peptide) genes are under inhibitory regulation by thyroid transcription factor-1 in C cells, and CaSR activation promotes calcitonin synthesis, at least in part by suppressing the levels of thyroid transcription factor-1 (Suzuki et al., 1998). CaSR coupling to G proteins in C cells is similar to its coupling in parathyroid cells and various other cell types, and the C-cell CaSR thereby activates plasma membrane phospholipases and cellular protein kinases (McGehee et al., 1997). Activation of the CaSR also stimulates acute elevations in Ca²⁺_i in various C-cell models. In some cell types, this occurs via Ca²⁺_o entry through plasma membrane L-type voltage-gated Ca²⁺ channels (Muff et al., 1988; Fajtova et al., 1991; McGehee et al., 1997) and, in others, via Ca²⁺_i mobilization (Freichel et al., 1996). Recently, the amino acid–activated CaSR was shown to stimulate calcitonin release from human C cells (Mun et al., 2019). Despite these insights, the molecular mechanisms by which the CaSR stimulates calcitonin secretion are largely unknown.

C. Calcium-Sensing Receptor in the Kidney

CaSR expression in the kidney is one of the highest in the body, and the renal CaSR plays a major role in the regulation of renal function in both a hormone-dependent and independent fashion [see Riccardi and Valenti (2016) and references therein]. A large body of functional, molecular, and genetic evidence indicates that the kidney CaSR plays a crucial role in mineral ion homeostasis. Indeed, the CaSR is widely expressed along the nephron at both the apical and basolateral sides of kidney cells, and thereby it is uniquely poised to monitor both urine and plasma and alter the final ultrafiltrate composition accordingly (Riccardi et al., 1998; Graca et al., 2016). Urinary calcium excretion mirrors serum calcium levels and is directly proportional to the filtered calcium load (Brown, 1991). Within the kidney, the TAL of Henle’s loop is the main site for active divalent cation movement, mostly via the paracellular route, and is coupled to NaCl reabsorption (Friedman, 1998). The latter occurs through a concerted action of an apical Na⁺:K⁺:2Cl⁻ cotransporter, NKCC2, and this is followed by basolateral exit via the voltage-gated Cl⁻ channel, chloride channel Kb, and the Na⁺:K⁺:ATPase. Overall, NaCl movement generates a favorable transepithelial electrochemical gradient for positively charged ions to move from the urine toward the basolateral side. In concert, the tight junctional proteins, claudins 14, 16, and 19, establish a divalent cation-selective permeable route, thereby allowing Ca²⁺_o (and Mg²⁺) reabsorption (Gong and Hou, 2014). The TAL has the highest CaSR expression, and here the CaSR is expressed basolaterally (Riccardi et al., 1998). In the event of hypercalcemia, CaSR activation dampens Ca²⁺_o reabsorption in two ways: firstly, it inhibits NaCl reabsorption, hence the driving force for divalent cation movement; secondly, it directly reduces Ca²⁺_o and Mg²⁺ junctional permeability through its actions on claudin 14 by activating microRNA-9 and -374 (Gong and Hou, 2014). If the hypercalcemic stimulus persists, hypercalciuria can occur with excess urinary calcium excretion in the terminal collecting duct.

In the presence of hypovolemia, the antidiuretic hormone, vasopressin, promotes water reabsorption through the insertion of aquaporin-2 water channels into the lumen of inner medullary collecting duct cells. However, excessive water reabsorption could lead to suprasaturating urinary calcium concentrations and attendant pathologic kidney stone formation, which could severely impair renal function. The CaSR is expressed at the luminal side of inner medullary collecting duct cells, where it monitors Ca²⁺_o concentration in the urine (Sands et al., 1997). Thus, CaSR activation inhibits the tubular response to vasopressin by limiting the number of apical aquaporin-2 water-channel insertions (Procino et al., 2012). In addition, CaSR activation stimulates the activity of the proton pump, V-ATPase, thereby evoking urine acidification and reducing the risk of precipitation (Renkema et al., 2009).

Further, the kidney proximal tubule is a major site of PTH action that promotes a phosphaturia by inhibiting the activity of the Na⁺:P_i cotransporters, Npt2a, and Npt2c (Murer et al., 2001). Excess phosphate in the urine could also exacerbate the risk of calcium-phosphorus stone formation by the distal nephron. In the proximal tubule, a luminal CaSR blunts the phosphaturic action of PTH and promotes acid secretion via stimulation of the Na⁺:H⁺ exchanger, Na⁺:H⁺ exchanger 3 (Capasso et al., 2013). Thus, by monitoring both urine and plasma composition, together with the integration of inputs deriving from urinary phosphate content, concentration, and acidification, the renal CaSR accomplishes divalent cation homeostasis while minimizing the risk of developing nephrolithiasis and nephrocalcinosis, which could arise as a consequence of enhanced urinary calcium excretion by the TAL (Hebert et al., 1997). The corollary is that altered CaSR expression or function due to CaSR mutations leads to FHH1, ADH1, or Bartter syndrome type V (see VI. Calcium-Sensing Receptor–Related Genetic Diseases and Therapeutic Interventions). In all circumstances, the aberrant calciuria is not the consequence of an impairment of renal function but rather the result of altered Ca²⁺_o sensing by the CaSR in the parathyroid glands and the kidney. In the context of CKD, hyperphosphatemia caused by decreased renal phosphate excretion and acidosis may both elicit CaSR underactivation, leading to secondary hyperparathyroidism. Therefore, similarly to the parathyroid CaSR, the kidney CaSR is a drug target and, indeed, pharmacological CaSR PAMs are employed to rectify abnormal Ca²⁺_o sensing by the kidney (Riccardi and Valenti, 2016). Furthermore, the use of NAMs for the treatment of nephrolithiasis and nephrocalcinosis could also be postulated (Riccardi and Valenti, 2016). Finally, it should be noted that CaSR PAMs increase urinary calcium excretion by means of their actions on both the parathyroid and kidney CaSR, and indeed, cinacalcet promotes calciuria in patients with secondary hyperparathyroidism, but this occurs in the absence of an increase in urine output (Courbebaisse et al., 2012). Given the clinical use of PAMs, the impact of their long-term use on urine production, acidification, and concentration, particularly in the context of kidney stone formation, remains to be fully understood (Riccardi and Valenti, 2016).

D. Calcium-Sensing Receptor in the Bone

The CaSR is expressed by several types of bone cells, including osteoblasts, osteocytes, osteoclasts, and some chondrocytes (Santa Maria et al., 2016). Although some controversies exist, there is good evidence that Ca²⁺_o and the CaSR contribute to skeletal development and maintenance (Chang et al., 2008; Goltzman and Hendy, 2015; Hannan et al., 2018a) and that bone CaSRs may even contribute to overall Ca²⁺_o homeostasis (Al-Dujaili et al., 2016). Elucidation of the CaSR’s roles in skeletal tissue was historically complicated by models that examined global deletion of the Casr gene (Kos et al., 2003). Global Casr deletion has numerous effects, partly through large alterations in PTH secretion and changes in serum calcium and phosphate concentrations (Hannan et al., 2018a), thus it is difficult to elucidate tissue-specific CaSR effects. To further complicate matters, early Casr knockouts involved deletion of Casr exon 5, which results in mice encoding a nonfunctional CaSR lacking a portion of its extracellular domain (Kos et al., 2003). When Casr exon 5–deleted mice were crossed with mice that had a deletion of Gcm2 (which results in no parathyroid gland development) or the Pth gene, the skeletal abnormalities seen in the global Casr knockout mice were largely abolished (Kos et al., 2003; Tu et al., 2003). Furthermore, studies of the Casr exon 5–deleted mice revealed an alternatively spliced Casr transcript in the growth plate and other organs, such as skin, that could compensate for the absence of full-length CaSR in cartilage and bone (Rodriguez et al., 2005). Nonetheless, alternative Casr knockout models and bone-specific Casr deletion has confirmed that the CaSR is critical to bone development and maintenance, as described below.

E. Calcium-Sensing Receptor in Keratinocytes

The CaSR is highly expressed in keratinocytes, the main epidermal cell type. Moreover, an ionic calcium gradient exists in the epidermis, which increases from the basal proliferative layer to reach a maximum in the stratum granulosum, wherein the keratinocytes are well-differentiated, decreasing again in the relatively water-deficient lipid-containing cells of the stratum corneum (Menon et al., 1985; Celli et al., 2010). The epidermal calcium gradient and the CaSR are critically important for various epidermal functions, including keratinocyte differentiation, water and xenobiotic barrier function, and wound healing (Tu and Bikle, 2013; Hannan et al., 2018a). Interestingly, the epidermal calcium gradient is predominantly present in intracellular organelles of keratinocytes, such as the endoplasmic reticulum and Golgi, although an extracellular gradient makes some contribution to the gradient (Celli et al., 2010).

Keratinocytes cultured in low-calcium media (<0.07 mM) proliferate well. Raising the Ca²⁺_o concentration above 0.1 mM promotes differentiation, as indicated by the appearance of E-cadherin/catenin complexes (adherens junctions) and desmosomes, upregulation of keratins 1 and 10, stratification of cells, and then formation of cornified envelope precursors (Braga et al., 1995). Disruption of the permeability barrier of the skin by tape stripping disrupts the epidermal calcium gradient, resulting in disorganization of the normally differentiated cell layers (Menon et al., 1994). When the calcium gradient is re-established over the next day or so, the permeability barrier also recovers. Skin diseases, such as psoriasis, characterized by abnormal barrier function also exhibit a loss of the calcium gradient (Menon and Elias, 1991).

CaSR expression increases in upper layers of the epidermis with the increase in differentiation, with high expression in the stratum granulosum but weak expression in the corneocytes (Komuves et al., 2002). There is some expression of the CaSR on the plasma membrane of keratinocytes, but its predominant localization in these cells is intracellular and in the cytoplasm around the nucleus (Komuves et al., 2002). This perinuclear localization is also seen, although not to the same extent, in rodent osteoblasts and chondrocytes (Chang et al., 1999). It is likely that Ca²⁺_o signals to the keratinocyte via the plasma membrane CaSR in a manner similar to that of more classic calcium targets, such as the parathyroid or kidney. The function of the intracellular CaSR is unclear at this time. Knockdown or inactivation of the CaSR in keratinocytes abrogates calcium-induced inhibition of proliferation and stimulation of differentiation of these cells (Tu et al., 2008). Not surprisingly, in mice in which there had been knockdown of the CaSR in the epidermis, skin barrier function was disrupted with impaired differentiation of keratinocytes, and these problems were exacerbated by a low-calcium diet (Tu et al., 2012). Keratinocytes from these epidermal Casr^−/− mice had blunted Ca²⁺_i mobilization in response to Ca²⁺_o, decreased Ca²⁺_i pools, defective cell-cell adhesion, and reduced expression of differentiation markers (Tu et al., 2012).

In contrast, mice engineered to constitutively overexpress the CaSR in basal keratinocytes displayed enhanced keratinocyte differentiation and barrier formation during development as well as accelerated hair growth at birth (Turksen and Troy, 2003). There was hypertrophy of the suprabasal keratinocyte layers with increased expression of early and late differentiation markers together with upregulation of epidermal growth factor and noncanonical Wnt-signaling pathways (Turksen and Troy, 2003).

In the epidermis, there are interactions between the CaSR and the vitamin D system in skin. The active vitamin D hormone, 1,25(OH)₂D, increases the calcium response in keratinocytes (Ratnam et al., 1999). Deletion of the epidermal CaSR reduces expression of both the vitamin D receptor (VDR) and CYP27B1, the enzyme that produces 1,25(OH)₂D from 25-hydroxyvitamin D (Tu et al., 2012). It is likely that these effects on the vitamin D system contribute to impaired differentiation of the epidermis in these mice and reduced function of the innate immune system (Schauber et al., 2007). Moreover, 1,25(OH)₂D increases transcription of the CASR (Canaff and Hendy, 2002).

It has previously been reported that wound healing is impaired in mice with epidermal deletion of the VDR (Oda et al., 2017). Very low dietary calcium or deletion of the Casr gene exacerbates this impairment in wound healing in epidermal VDR-deficient mice (Oda et al., 2017). There is a robust increase in Ca²⁺ in the bed of wounds within minutes of injury (Jungman et al., 2012), rapid increase in Ca²⁺_i in cells near the site of the wound, and spreading to surrounding cells (Tsutsumi et al., 2013), with all of these indicating that the CaSR may play an important role in wound healing. The CaSR is coexpressed with E-cadherin at the cell membranes of migratory keratinocytes. Blockade of either the CaSR or E-cadherin inhibited keratinocyte proliferation and migration after wound induction (Tu et al., 2019). Accordingly, the PAM, NPS R-568, accelerated wound healing in normal mice, potentially pointing to the epidermal CaSR as a therapeutic target to enhance repair of skin wound (Tu et al., 2019).

Mice with epidermal knockout of the VDR are more susceptible to UV- or chemically induced skin tumor formation (Zinser et al., 2002; Ellison et al., 2008). Neither the epidermal VDR knockout mice nor mice with epidermal Casr knockout develop skin tumors spontaneously, but mice null for both epidermal Vdr and Casr are reported to spontaneously develop squamous cell carcinomas (Bikle et al., 2015). In keratinocytes, stimulation of Wnt signaling results in β-catenin translocation to the nucleus and subsequent transcriptional activity, which may be important in skin tumorigenesis (Wei et al., 2007; Youssef et al., 2012). The VDR appears to suppress this β-catenin transcriptional activity in skin (Wei et al., 2007), in part by helping to keep β-catenin at the cell membrane as part of the E-cadherin/catenin complex (adherens junctions). As noted earlier, the CaSR is also important for the development of the E-cadherin/catenin complex, which helps to retain β-catenin at the cell membrane by promoting wound healing and differentiation of the skin cells (Oda et al., 2017) and inhibiting nuclear translocation and associated protumorigenic activities of β-catenin (Wei et al., 2007). This is in direct contrast with osteoblasts, wherein activation of the CaSR predominantly promotes β-catenin stabilization, subsequent β-catenin translocation to the nucleus, and increased transcriptional activity (Rybchyn et al., 2011). Some preliminary data indicate that both CaSR PAMs and NAMs enhance DNA repair after UV damage in cultured keratinocytes (Yang et al., 2016), although the mechanism and why PAMs and NAMs have a similar effect are unclear. How this observation fits with observed effects of CaSR knockdown in mice remains to be examined.

F. Calcium-Sensing Receptor in the Gastrointestinal Tract

The CaSR is expressed along the entire gastrointestinal tract in parietal and G cells of stomach gastric glands (Ray et al., 1997; Busque et al., 2005; Feng et al., 2010; Engelstoft et al., 2013), epithelial and enteroendocrine cells of the small and large intestine (Liou et al., 2011; Wang et al., 2011; Cheng et al., 2014; Alamshah et al., 2017), and neurons of the submucosal and myenteric plexuses of the enteric nervous system (ENS) (Geibel et al., 2006; Cheng, 2012; Tang et al., 2018). In the gastrointestinal tract, the CaSR functions as a nutrient sensor, binding not only Ca²⁺, Mg²⁺, and other cations but also L-amino acids and dipeptides and polypeptides (e.g., glutamyl dipeptides, poly-L-lysine). The CaSR is involved in regulation of gastric acid and hormone secretion, nutrient absorption, intestinal fluid homeostasis, energy metabolism, cellular differentiation and proliferation, motility and enteric nerve activity, maintenance of gut microbiota, immune homeostasis, and intestinal inflammation (Dufner et al., 2005; Ceglia et al., 2009; Geibel and Hebert, 2009; Feng et al., 2010; Brennan et al., 2014; Cheng et al., 2014; Tang et al., 2016b, 2018; Alamshah et al., 2017; Sun et al., 2018).

The CaSR responds to alterations in nutrient levels by regulating hormone secretion from enteroendocrine cells (Geibel and Hebert, 2009; Liou et al., 2011; Wang et al., 2011; Alamshah et al., 2017; Liu et al., 2018). In global Casr knockout mice, gastric G cell number was significantly reduced, suggesting the CaSR regulates G cell growth. Further, in wild-type but not knockout mice, NPS 2143 inhibited gastrin secretion after gavage of Ca²⁺_o, L-Phe, or cinacalcet (Feng et al., 2010). In rat whole-stomach preparations, ex vivo exposure to Ca²⁺_o increased acid production in parietal cells by enhancing H⁺-K⁺-ATPase activity. These effects were potentiated by L- but not D-amino acids, implicating the CaSR (Busque et al., 2005). The function of the recently identified acid secretory protein, vacuolar H⁺-ATPase, in parietal cells is also dependent on CaSR activity (Kitay et al., 2018).

The amino acid–stimulated CaSR may influence appetite and satiety via stimulatory effects on satiety hormones, cholecystokinin (CCK), glucagon-like peptide 1 (GLP-1) and protein YY (PYY) (Alamshah et al., 2017), and/or inhibitory effects on the release of the appetite-stimulating hormone ghrelin (Engelstoft et al., 2013). In the mouse enteroendocrine cell line, STC-1, L-Phe increased PYY and GLP-1 secretion, an effect inhibited by NPS 2143, suggesting involvement of the CaSR (Alamshah et al., 2017). In a transgenic mouse model expressing a CCK promoter–driven enhanced GFP, the CaSR was enriched in CCK-producing duodenal I cells, in which L-Phe and cinacalcet induced Ca²⁺_i changes and stimulated CCK in the presence of Ca²⁺_o (Liou et al., 2011). L-Phe- and Trp-stimulated CCK secretion was inhibited by calhex 231 (Wang et al., 2011). In intestinal L-cells, the CaSR was involved in peptone-stimulated GLP-1 release (Pais et al., 2016), whereas in a ghrelinoma cell line, the CaSR partially mediated the L-Phe, L-Ala, and peptone-induced secretion of octanoyl ghrelin (Vancleef et al., 2015). In swine duodenum, ex vivo L-Trp perfusion induced secretion of CCK and glucose-dependent insulinotropic peptide and upregulated CaSR expression. This effect was inhibited by NPS 2143 (Zhao et al., 2018). To conclude that the Trp-induced gut-hormone secretion is mediated by the CaSR, further proof is needed. In mice and rats, L-Phe reduced short-time food intake and plasma ghrelin release, affecting the appetite of the animals (Alamshah et al., 2017). In minks, CaSR-mediated secretion of CCK and PYY led to emesis (Wu et al., 2017). These findings might explain why cinacalcet and other CaSR PAMs cause gastrointestinal side effects (Block et al., 2017).

Specific knockout of intestinal epithelial cell Casr leads to epithelial cell hyperproliferation and changes in intestinal crypt structure driven by β-catenin signaling (Rey et al., 2012). Mice additionally have decreased intestinal transepithelial resistance and reduced levels of colonic tight-junction proteins, suggesting that the epithelial CaSR maintains intestinal barrier function (Cheng et al., 2014). The inadequate epithelial barrier function was associated with lower amounts of beneficial Lactobacilli bacteria and more Deferribacteraceae bacteria, which are linked to colitis (Cheng et al., 2014). This intestinal dysbiosis has been associated with more severe proinflammatory responses in the intestinal epithelium-specific Casr null mice compared with wild-type controls (Owen et al., 2016).

The amino acid-stimulated CaSR has recently been found to suppress intestinal inflammation in inflammatory bowel disease and other settings [reviews: (Owen et al., 2016; Sun et al., 2018)]. Inflammatory cytokine expression, including IL-1R, was higher in the distal colons of the CaSR knockout mice, in addition to a marked increase in nuclear factor-κB–dependent genes (Cheng et al., 2014). These mice developed more severe colitis with delayed recovery than the CaSR-expressing littermates when challenged with dextrane sulfate sodium (DSS) (Cheng et al., 2014). In a mouse model of colitis, poly-L-lysine (commonly used as a food preservative) and glutamyl dipeptides reduced DSS-induced inflammation, whereas intravenous administration of NPS 2143 inhibited this effect (Mine and Zhang, 2015; Zhang et al., 2015). Dietary supplementation of Trp, L-Phe, and Tyr also reduced the expression of intestinal inflammatory markers in piglets after short-term induction of inflammation by lipopolysaccharides (Liu et al., 2018). However, a recent study found that the CaSR PAMs, cinacalcet and GSK3004774 (an intestine-specific modulator), did not reduce the inflammatory effects of DSS, whereas NPS 2143 ameliorated the DSS-induced symptoms and reduced immune cell infiltration (Elajnaf et al., 2019).

The anti-inflammatory effect of the CaSR was also shown in vitro in cell lines. In a colonic myofibroblast cell line, activation of the CaSR inhibited tumor necrosis factor-α secretion (Kelly et al., 2011) and increased expression of bone morphogenetic protein-2, which is a promoter of colonic epithelial barrier maturation (Peiris et al., 2007). In colon cancer cell lines, amino acids and dipeptides inhibited proinflammatory cytokine secretion, and the effect was reversed by NPS 2143 (Mine and Zhang, 2015; Zhang et al., 2015). Inflammatory cytokines, such as tumor necrosis factor-α, IL-1β, and IL-6, increased the expression of the CaSR at the mRNA and protein level in some colon cancer cell lines (Fetahu et al., 2014), which could be a defense mechanism against inflammation in the intestines.

Bicarbonate (HCO₃⁻) secretion in the colon is fine-tuned by the CaSR (Tang et al., 2015). However, it seems that the neurogenic secretory responses in the intestinal epithelium are mediated mainly by the CaSR expressed in the ENS and not in the epithelium (Geibel et al., 2006; Cheng, 2012; Tang et al., 2018). Because abnormalities of the ENS affect the severity of intestinal inflammation and contribute to the pathogenesis of inflammatory bowel disease (Margolis et al., 2011), the CaSR could be a potential therapeutic target.

Increased dietary intake of calcium reduces the risk of several cancers. The inverse correlation between calcium intake and risk of colorectal cancer has been known for decades, although the mechanisms driving the protective effect of calcium were not clear. There is some evidence that the CaSR is one of the central mediators of the antitumorigenic effects of calcium and acts as a tumor suppressor (Kállay et al., 1997, 2003; Yang et al., 2018). In colon cancer cells, activation of the CaSR increased differentiation and reduced proliferation, epithelial-to-mesenchymal transition, and expression of stem cell markers (Aggarwal et al., 2015, 2017). The signaling pathways involved in these processes still need to be determined. Interestingly, in the upper intestinal tract, it seems that the CaSR functions as an oncogene, as it promoted gastric cancer cell proliferation (Xie et al., 2017).

G. Calcium-Sensing Receptor in the Pancreas

The CaSR is expressed in pancreatic acinar cells (Bruce et al., 1999), which promote digestion via nutrient-stimulated release of digestive enzymes and fluid. The CaSR is also expressed in the pancreatic islets on glucagon-secreting α cells and insulin-secreting β cells (Babinsky et al., 2017). Thus, the CaSR may influence not only protein metabolism but also carbohydrate and fat metabolism.

Ca²⁺_o is critical for pancreatic islet function and acts via voltage-gated Ca²⁺ channels to trigger the exocytosis of insulin- and glucagon-containing secretory granules from β- and α-cells, respectively (Rorsman et al., 2012). Ca²⁺_o also activates the pancreatic islet CaSR, with ex vivo and in vitro studies demonstrating a role for the CaSR in mediating islet hormone secretion. Thus, stimulation of isolated human islets and an insulin-secreting mouse cell line (MIN6) with the PAM, NPS R-568, potentiated Ca²⁺_o-mediated insulin secretion (Gray et al., 2006), whereas knockdown of the CaSR through RNA interference diminished glucose-induced insulin secretion in MIN6 cells that were cultured as pseudoislets (Kitsou-Mylona et al., 2008). Studies involving MIN6 pseudoislets have also revealed CaSR-stimulated insulin secretion to be mediated by PLC and the MAPK pathway (Gray et al., 2006). Furthermore, CaSR-mediated MAPK activation in MIN6 cells induces β-cell proliferation, thus highlighting a potential role for the CaSR in the regulation of β-cell mass (Kitsou-Mylona et al., 2008). The CaSR also upregulates the expression of E-cadherin in MIN6 cells, which is associated with increased adherence between neighboring β-cells (Hills et al., 2012). Thus, the CaSR may facilitate cell-to-cell communication within an individual pancreatic islet to coordinate insulin secretion from β-cells (Hodgkin et al., 2008). In addition, the level of islet CaSR expression correlates with insulin secretion from isolated wild-type mouse pancreatic islets (Oh et al., 2016). Studies in wild-type mice have shown that islet CaSR expression increases with age, which may compensate for the insulin resistance in aged mice by increasing insulin secretion (Oh et al., 2016). Transient stimulation of isolated human islets with Ca²⁺_o and NPS R-568 also promoted glucagon secretion, thereby indicating a role for the CaSR in α-cells (Gray et al., 2006). The intestinal CaSR, which is activated by dietary amino acids and peptides, may also influence pancreatic islet function by regulating the secretion of incretin hormones. In support of this, studies involving isolated mouse intestine have shown that the CaSR is expressed in GLP-1-secreting L-cells and also that oligopeptides enhance GLP-1 secretion by activation of the CaSR (Diakogiannaki et al., 2013).

The role of the CaSR in systemic glucose homeostasis has been investigated in studies involving human subjects. An association study reported a common coding-region CASR gene variant to be an independent determinant of plasma glucose concentrations in renal transplant recipients (Babinsky et al., 2015). However, another study involving patients with FHH1 (caused by germline loss-of-function CASR mutations) did not reveal any alterations in insulin secretion or glucose tolerance (Wolf et al., 2014). The effect of altered CaSR function on glucose tolerance has also been evaluated in an ADH1 mouse model, which is referred to as Nuclear flecks (Nuf) because the mutant mouse was initially identified to have nuclear cataracts (Babinsky et al., 2017). Nuf mice, which harbor a germline gain-of-function CaSR mutation (Leu723^ICL2Gln) causing hypocalcemia, have impaired glucose tolerance and hypoinsulinemia in association with reductions in pancreatic islet mass and β-cell proliferation (Babinsky et al., 2017). Nuf mice also lack glucose-mediated suppression of glucagon secretion, which was associated with an increase in α-cell proliferation and an impairment of α-cell membrane depolarization (Babinsky et al., 2017). Administration of the NAM, ronacalceret, ameliorated the hypocalcemia and glucose intolerance of Nuf mice, and these findings highlight the potential utility of targeted CaSR compounds for modulating glucose metabolism (Babinsky et al., 2017).

H. Calcium-Sensing Receptor in Mammary Glands

The CaSR is expressed in breast epithelial cells, in which its main role is to fine tune maternal Ca²⁺ metabolism by balancing Ca²⁺_o mobilization and usage: It ensures the supply of Ca²⁺ for milk while protecting against maternal hypocalcemia (Cheng et al., 1998; VanHouten et al., 2004, 2007; Kim and Wysolmerski, 2016). The expression of the CaSR is increased during lactation when it regulates Ca²⁺_o transport into milk. In parallel, it inhibits synthesis of PTHrP by coupling with G_i to inhibit adenylyl cyclase activity and cAMP production (VanHouten et al., 2004). During milk production, the CaSR enables the lactating breast to participate in the regulation of systemic Ca²⁺_o and bone metabolism. VanHouten and Wysolmerski (2013) suggested a negative feedback between systemic Ca²⁺_o delivered to the lactating breast and PTHrP synthesis and secretion by mammary epithelial cells. When the mother’s serum calcium level is adequate, the CaSR in breast epithelial cells stimulates calcium secretion into milk, but reduces Ca²⁺_o usage when the mother’s calcium supply becomes limited (VanHouten and Wysolmerski, 2013). In mammary epithelial cells, the CaSR regulates Ca²⁺_o transport by altering the activity of the plasma membrane Ca²⁺ ATPase 2; however, the detailed molecular mechanism is not yet known (VanHouten et al., 2007; VanHouten and Wysolmerski, 2007).

The CaSR is expressed also in the neoplastic mammary gland. In contrast with normal breast cells, in breast cancer cells the CaSR stimulates PTHrP secretion. This is possible because the CaSR switches coupling from G_i/o to G_s, leading to stimulation of cAMP and PTHrP synthesis (Mamillapalli et al., 2008). The higher PTHrP levels, secreted because of activation of the CaSR, inhibit the cell cycle inhibitor p27^kip1 and the apoptosis-inducing factor, stimulating cell proliferation and reducing apoptosis (Kim et al., 2016). Moreover, PTHrP is an activator of osteoclasts and often stimulates osteolytic bone destruction when secreted from cancer cells that metastasize to bone (Wysolmerski, 2012).

The CaSR is highly expressed by metastatic breast cancer cells and potentiates their osteolytic ability, promoting a more aggressive behavior. In vitro, the CaSR promoted breast cancer cell migration only in cells capable of forming bone metastases (e.g., MDA-MB-231, MCF-7) but not in BT474 cells that have no bone-metastatic potential, even though CaSR levels were similar in all cell types (Saidak et al., 2009). It has been shown recently that breast cancer cells overexpressing the wild-type CaSR injected intratibially into BALB/c-Nude mice led to osteolytic lesions through an epiregulin-mediated mechanism (Boudot et al., 2017). The oncogenic potential of the CaSR in breast cancer cells was also suggested by the fact that activation of the CaSR by NPS R-568 or Ca²⁺_o increased secretion of proangiogenic and chemotactic cytokines and growth factors from the highly invasive MDA-MB-231 breast cancer cells (Hernández-Bedolla et al., 2015). Another group, however, found that activating the CaSR with Ca²⁺_o induced sensitivity of MCF-7 and MDA-MB-435 cells to the chemotherapeutic drug paclitaxel and reduced malignant behavior. The paclitaxel-resistant cells expressed no CaSR (Liu et al., 2009). This group suggested a positive link between the tumor suppressive functions of BRCA1 and the CaSR (Promkan et al., 2011).

I. Calcium-Sensing Receptor in Airway Smooth Muscle and Epithelium

Asthma is characterized by airway hyperresponsiveness, inflammation, and remodeling of the conducting airways. A number of mechanisms, many driven by inflammation, have been hypothesized to contribute to airway hyperresponsiveness and/or remodeling. Among these, local increases of polycations are seen in the airways of patients who are asthmatic (Kurosawa et al., 1992) and, vice versa, increased inflammation increases the local concentration of polycations. Furthermore, the polycations, eosinophil cationic protein and major basic protein, are markers for asthma severity and stability. Elevated arginase activity increases the consumption of L-arginine to enhance production of the polycations, spermine, spermidine, and putrescine (North et al., 2013). Indeed, arginase inhibitors have been proposed to have therapeutic potential for allergic asthma (van den Berg et al., 2018). Recent evidence suggests that the CaSR is expressed in the airway epithelium, smooth muscle, and inflammatory cells and that polycations act at the CaSR and are directly implicated in the pathogenesis of asthma (Yarova et al., 2015). Yarova et al. (2015) have also shown that inhaled CaSR NAMs delivered topically reverse-airway hyperresponsiveness, inflammation, and remodeling in in vivo models of allergic asthma and other inflammatory lung diseases, such as chronic obstructive pulmonary disease. Inhaled NAMs also show efficacy in nonallergic asthma, which is often associated with poor response to steroids, and for which currently there is no treatment (Riccardi unpublished observations). Four CaSR NAMs have been studied in humans; NPSP795, ronacaleret, AXT914, and JTT-305 (IIB. Endogenous and Exogenous Allosteric Modulators, Table 1), which could be repurposed, via the inhaled route, as novel asthma treatments. Crucially, delivery of CaSR NAMs directly to the lung does not significantly affect serum calcium levels up to 24 hours post-treatment, suggesting absence of any significant systemic overspill and possible effects on whole-body mineral ion homeostasis in vivo. Thus, CaSR NAMs could provide a new therapeutic approach to treating inflammatory lung disease in humans.

J. Calcium-Sensing Receptor in the Vasculature

The CaSR is expressed in the intima, media, and adventitia of the blood vessels in endothelial, smooth muscle cells and in the perivascular neurons. Although consumption of dietary calcium reduces blood pressure (Nakamura et al., 2019; Rietsema et al., 2019), direct actions of Ca²⁺_o on isolated blood vessels have yielded contrasting effects, with both relaxation and constriction reported (Bohr, 1963). Furthermore, the molecular mechanisms underlying these actions are elusive. Studies carried out over the last two decades indicate that the CaSR could mediate at least some of the effects of Ca²⁺_o on vascular function, with opposing effects in the endothelium and smooth muscle cell layers of the blood vessels. Specifically, CaSR activation by the CaSR PAM, cinacalcet, in the vascular endothelium leads to hyperpolarization and attendant nitric oxide release and vasodilatation (Smajilovic et al., 2007). In contrast, studies of Casr gene ablation in the vascular smooth muscle cells show that activation of the CaSR in these cells leads to contraction, as evidenced by the fact that Casr knockout mice exhibit impaired vascular reactivity, hypotension, and reduced contractile response to Ca²⁺_o (Schepelmann et al., 2016). Thus, the CaSR sets blood vessel tone by integrating prorelaxing (endothelium-mediated) actions with procontractile (smooth muscle–mediated) effects (Schepelmann et al., 2016). Therefore, altered CaSR expression within either the endothelium or smooth muscle could account for the abnormal vascular reactivity seen in advanced CKD or in type 2 diabetes. Indeed, systemic administration of the CaSR PAM, NPS R-568, initially evokes an increase in blood pressure in control and in uremic rats (a model for advanced CKD), which is followed by a reduction in blood pressure, but only in uremic animals (Odenwald et al., 2006), suggesting partial loss of the CaSR-dependent contractile component of the vasculature.

However, there is some controversy regarding the role of the CaSR in regulating blood pressure. In ex vivo studies in rat mesenteric arteries, relaxant responses to cinacalcet and calindol were not blocked by calhex 231 (Thakore and Ho, 2011), which at the time was believed to be a CaSR NAM (but has now been shown to have mixed PAM and NAM activity, see II.B. Endogenous and Exogenous Allosteric Modulators). Nonetheless, Ca²⁺_o influx into these vessels stimulated by the α1 adrenergic receptor agonist, methoxamine, was inhibited, not potentiated, by cinacalcet and calindol, as were contractions in response to an L-type calcium channel activator (Thakore and Ho, 2011). Given that arylalkylamine PAMs are structurally related to the nonselective calcium channel blocker, fendiline, and have low affinity for calcium channels, the relaxing effects of CaSR PAMs in arteries may in part arise from off-target calcium channel effects. Further support for a non-CaSR-mediated effect of PAMs in the vasculature comes from findings that, although the S-enantiomers of arylalkylamine PAMs have little activity at the CaSR, the effects of NPS R-568 on vascular tone, blood pressure, and heart rate are not stereoselective and only occur at concentrations in excess of those required to inhibit PTH secretion (Nakagawa et al., 2009).

End-stage CKD is associated with impaired mineral ion metabolism, which can lead to pathologic vascular calcification of the medial layer of the blood vessel, left ventricular hypertrophy, and increased cardiovascular mortality (Locatelli et al., 2002). CaSR expression is significantly reduced in the medial layer of calcifying blood vessels and is completely absent in areas of extensive medial calcification, suggesting an involvement of the CaSR in the vascular calcification process (Alam et al., 2009). Human and bovine vascular smooth muscle cells exposed to Ca²⁺ and phosphate concentrations mimicking those seen during pathologic CKD exhibit marked calcification in vitro, an effect that is exacerbated by CaSR downregulation and that is reversed by the CaSR PAM, NPS R-568 (Alam et al., 2009). In addition, NPS R-568 reduces blood pressure and ameliorates cardiac remodeling in animal models of advanced CKD in vivo (Ogata et al., 2003). Taken together, these results suggest that loss of CaSR expression by the medial layer of the blood vessels in advanced CKD leads to vascular calcification and that CaSR PAMs might be vasculo-protective by directly restoring normal CaSR expression levels within the vasculature. However, CaSR PAMs reduce systemic levels of serum P_i and PTH through their actions on the parathyroid CaSR, and parathyroidectomy suppresses vascular calcification (Kawata et al., 2008); therefore, PAM-mediated reduction of vascular calcification may be dependent on activation of parathyroid CaSRs. Although in vitro and in vivo studies support a direct role for the vascular CaSR in protecting vascular function, human observational studies of clinical evaluation of the CaSR PAM, cinacalcet, assessed by the Evaluation of Cinacalcet Hydrochloride Therapy to Lower Cardiovascular Events randomized controlled trial failed to reach its endpoints (reduction of all-cause and cardiovascular mortality in patients with advanced CKD) (Chertow et al., 2012). However, a recent Bayesian meta-analysis combined with a systematic literature review concluded that once subject ages and high drop-out rates throughout the trial are accounted for, cinacalcet treatment does reduce mortality rates in patients with secondary hyperparathyroidism on hemodialysis (Lozano-Ortega et al., 2018). Therefore, further clinical studies are needed to fully evaluate the effects of CaSR PAMs on cardiovascular and all-cause mortality in patients with advanced CKD.

Finally, it should be pointed out that the CaSR is also expressed in arterial smooth muscle cells of the pulmonary vasculature, wherein receptor activation leads to pulmonary vasoconstriction and proliferation. Here, CaSR NAMs prevent the development and progression of pulmonary hypertension in mouse and rat models in vivo (Tang et al., 2016a). Thus, targeting the CaSR in the pulmonary arteries with inhaled NAMs might provide a novel treatment of patients with idiopathic pulmonary hypertension.

K. Calcium-Sensing Receptor in the Brain and Nervous System

For a comprehensive review of all evidence for CaSR function in the brain, readers are directed to a recent review (Giudice et al., 2019).

Although the role of the CaSR in human brain requires validation, the CaSR is expressed throughout the rat brain, with particular abundance in the hippocampus, striatum, cerebellum, pituitary, and olfactory bulb (Ruat et al., 1995). However, CaSR expression can change with developmental age (Vizard et al., 2008), which supports a role for the CaSR in brain development. For instance, rat CaSR expression increases in fetal oligodendrocyte precursor cells and postnatal immature oligodendrocytes during myelination of nerve axons, but expression declines in mature oligodendrocytes (Chattopadhyay et al., 1998, 2008; Ferry et al., 2000).

Although hyperparathyroidism and consequent early lethality resulting from global Casr ablation preclude determination of the role of the CaSR in brain development, concomitant Casr and PTH ablation prevents hyperparathyroidism, and mice survive to adulthood (Kos et al., 2003). In brains of Casr^−/−/Pth^−/− mice, neuron and glial cell differentiation markers were reduced after birth, whereas differentiation of neural stem cells from Casr^−/− mice was delayed (Liu et al., 2013). These mice also had reduced numbers of gonadotropin-releasing hormone-positive neurons in the hypothalamus. These findings suggest a role for the CaSR in neuron and glial cell differentiation.

To elucidate region-specific CaSR brain functions, hippocampus-specific Casr ablation 3 weeks postbirth has been undertaken (Kim et al., 2014). Although mice did not display an obvious phenotype under normal conditions, they were protected from hippocampal neuronal damage in response to ischemia-induced injury, which mimics injury sustained during cardiac arrest or stroke. In line with these findings, hypoxia increases CaSR expression in rat hippocampal neurons in vivo and in vitro (Bai et al., 2015), but neuroprotection from ischemia is blocked when the related class C GPCR, GABA_BR1, is also ablated (Kim et al., 2014). This may be explained by the discover of an increase in CaSR expression in hippocampal neurons in culture upon suppression of GABA_BR1 levels (Chang et al., 2007), which is also observed in cortical neurons of mice who have experienced controlled cortical impact as a model for traumatic brain injury (Kim et al., 2013). In support of a role for the CaSR in the hippocampus, in rat hippocampal neurons from wild-type but not Casr^−/− mice, CaSR activation opens nonselective cation channels (Ye et al., 1997b). Similarly, transfection of a dominant negative Arg185Gln mutant CaSR into pyramidal neurons of hippocampal brain slice cultures resulted in significantly shorter and less complex dendritic branching (Vizard et al., 2008). Taken together, these studies suggest that CaSR NAMs could be neuroprotective.

In addition to neuroprotective effects upon brain injury, inhibition of brain CaSRs may afford neuroprotection in Alzheimer disease. The first evidence for a possible role of the CaSR in Alzheimer disease came from a study that demonstrated activation of nonselective cation channels in cultured hippocampal pyramidal neurons from wild-type rats and mice but not from Casr^−/− mice (Ye et al., 1997a). Although additional studies have since suggested β-amyloid proteins activate the CaSR (Conley et al., 2009; Dal Pra et al., 2014), these findings warrant further validation. Nonetheless, CaSR expression is increased in the hippocampus of an Alzheimer disease mouse model (Gardenal et al., 2017), and there is a positive association between CaSR SNPs and Alzheimer disease, although this is only in patients who do not harbor the Alzheimer risk allele encoding apolipoprotein E4 (Conley et al., 2009). Furthermore, in human cortical astrocytes and neurons in culture, neurotoxic β-amyloid_25–35 stimulates full-length β-amyloid₄₂ secretion, an effect that is blocked by the CaSR NAM, NPS 2143 (Armato et al., 2013; Chiarini et al., 2017b). NPS 2143 also blocked β-amyloid_25–35–mediated GSK-3β activation and subsequent phosphorylation of τ in cultured human astrocytes (Chiarini et al., 2017a).

The CaSR has also been implicated in the etiology of neuroblastomas, tumors originating from precursor nerve cells of the sympathetic nervous system. Approximately 98% of neuroblastomas are associated with spontaneous mutations in a variety of genes (Aygun, 2018). Analysis of mRNA from neuroblastoma tumors indicates that although the CaSR is expressed in benign differentiated tumors, epigenetic hypermethylation of the CASR P2 promoter region silences CASR transcription in some aggressive neuroblastomas (de Torres et al., 2009; Casalà et al., 2013). Similarly, two noncoding CASR SNPs (rs7652579 and rs1501899) that reduce CaSR expression are present in homozygous or heterozygous form in 58% of neuroblastoma tumors but in only 47% of the general population (Masvidal et al., 2017). In a subset of ganglioneuromas, CASR expression was absent in four out of six tumors harboring rs7652579 and rs1501899. However, neuroblastoma patients with rs7652579 and rs1501899 SNPs did not have poorer outcomes or survival (Masvidal et al., 2017). In contrast, neuroblastomas with a haplotype SNP in the CASR gene-coding region were associated with poorer outcomes, including increased risk of death (Masvidal et al., 2013). Importantly, cinacalcet reduced neuroblastoma tumor growth in immunocompromised mice carrying neuroblastoma xenografts by inducing endoplasmic reticulum stress, tumor differentiation, and fibrosis as well as upregulation of cancer-testis antigens (Rodríguez-Hernández et al., 2016).

Finally, approximately 40% of patients harboring ADH1 gain-of-function CASR mutations present with seizures (Gorvin, 2019). Although this could be associated with the consequent reduction in serum calcium concentrations, a gain-of-expression CaSR mutation, Arg898Gln, was identified in a patient with idiopathic epilepsy who did not have low serum concentrations of calcium or PTH (Kapoor et al., 2008). These findings suggest a possible role for CaSR in neurotransmission, which is supported by numerous in vitro studies suggesting the CaSR regulates synaptic transmission and neuronal activity via activation of nonselective cation channels on presynaptic terminals [reviewed in Jones and Smith (2016)].

A. Loss- and Gain-of-Function Mutations in the Calcium-Sensing Receptor and Its Signaling Partners

Alterations in CaSR signaling, which lead to derangements of mineral homeostasis, can result from loss-of-function germline mutations of the CASR gene on chromosome 3q21.1, which cause FHH1 and NSHPT, or gain-of-function germline CASR mutations, which lead to ADH1 and Bartter syndrome type V (Fig. 4; Table 2) (Hannan et al., 2012; Hannan and Thakker, 2013). In addition, loss- and gain-of-function germline mutations of the GNA11 gene on chromosome 19p13.3, which encodes Gα₁₁, are associated with FHH2 and ADH2, respectively (Table 2) (Nesbit et al., 2013a; Hannan et al., 2016). Furthermore, loss-of-function germline mutations of the AP2S1 gene on chromosome 19q13.3, which encodes AP2σ, cause FHH3 (Table 2) (Nesbit et al., 2013b; Hannan et al., 2015a).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4.

CaSR snakeplot with residues linked to loss- and gain-of-function germline mutations. Snakeplot of the CaSR showing the location of the ECD, 7TM, ICLs, ECLs, and carboxy terminus. Sites of loss- and gain-of-function germline mutations causing FHH1/NSHPT (red), ADH1/Bartter syndrome type V (green), or both FHH1/NSHPT and ADH1/Bartter syndrome type V (yellow), respectively. Snakeplot generated by GPCRdb (Munk et al., 2016) with data from the Human Gene Mutation Database (Stenson et al., 2012). C-term, C terminal; N-term, N terminal.

B. Familial Hypocalciuric Hypercalcemia and Neonatal Severe Hyperparathyroidism

FHH is a genetically heterogeneous autosomal dominant disorder characterized by lifelong nonprogressive elevations of serum calcium concentrations, mild hypermagnesemia, normal or mildly raised serum PTH concentrations, and low urinary calcium excretion (Table 2) (Hannan and Thakker, 2013). FHH1 (OMIM 145980) accounts for ∼65% of all FHH cases and is usually an asymptomatic disorder. It has been associated with >150 different CASR mutations (Hannan et al., 2018a). The majority (>85%) of these loss-of-function CASR mutations are heterozygous missense substitutions, which are predominantly located in the VFT of the CaSR ECD and also in the 7TM (Hannan et al., 2012). These FHH1-associated missense mutations cause a loss of function by diminishing the signaling responses of CaSR-expressing cells (Leach et al., 2012) or by reducing CaSR anterograde trafficking and cell surface expression (Huang and Breitwieser, 2007; White et al., 2009). In addition, FHH1-causing missense mutations may induce biased agonism by switching from a wild-type CaSR that preferentially increases Ca²⁺_i mobilization to mutant receptors that demonstrate equal preference for Ca²⁺_i and MAPK pathways or that preferentially act via MAPK (Leach et al., 2012, 2013; Leach and Gregory, 2017). Between 10% and 15% of FHH1 cases are caused by heterozygous deletion, insertion, nonsense, and splice-site mutations that lead to nonsense-mediated decay of mRNA and CaSR haploinsufficiency or truncate the CaSR protein (Hannan et al., 2012). The offspring of two parents with FHH1 can harbor biallelic loss-of-function CASR mutations that cause NSHPT (OMIM 239200), which is associated with marked hyperparathyroidism that leads to hypercalcemia and bone demineralization causing fractures and respiratory distress (Hannan and Thakker, 2013). Occasionally, biallelic loss-of-function CASR mutations can lead to primary hyperparathyroidism, which presents in adulthood (Table 2) (Hannan et al., 2010). Furthermore, some heterozygous mutations (e.g., Arg185Gln) can cause NSHPT due to dominant negative effects on the wild-type CaSR (Bai et al., 1997).

FHH2 (OMIM 145981) is the least common form of FHH and has been reported in four probands to date (Nesbit et al., 2013a; Gorvin et al., 2016, 2018b). FHH2 appears to have a mild clinical presentation with serum-adjusted total calcium concentrations usually between 2.55 and 2.80 mM (normal range 2.10–2.55 mM). Urinary calcium excretion may be normal or low (Table 2) (Nesbit et al., 2013a; Gorvin et al., 2016, 2018b). The GNA11 mutations reported in FHH2 probands consist of three missense substitutions (Thr54Met, Leu135Gln, Phe220Ser) and an in-frame isoleucine deletion (Ile200del) (Nesbit et al., 2013a; Gorvin et al., 2016, 2018b). All of these mutations impair CaSR-signaling responses and are located within key domains of the Gα₁₁ protein (Nesbit et al., 2013a; Gorvin et al., 2016, 2018b). Thus, the Ile200del and Phe220Ser mutations are located within the Gα₁₁ GTPase domain and are predicted to diminish the interaction of Gα₁₁ with the CaSR or PLC, respectively (Nesbit et al., 2013a; Gorvin et al., 2018b). In contrast, the Leu135Gln mutation is situated within the PLC-interacting portion of the Gα₁₁ helical domain, and the Thr54Met mutation is located at the interface between the helical and GTPase domains and may potentially affect GTP binding (Nesbit et al., 2013a; Gorvin et al., 2016).

FHH3 (OMIM 600740) has been reported in >60 FHH probands and has a more marked clinical phenotype than FHH1. Thus, FHH3 is associated with significant elevations of serum calcium and magnesium and also a significantly reduced urinary calcium excretion compared with FHH1 (Table 2) (Hannan et al., 2015a; Vargas-Poussou et al., 2016). In addition, hypercalcemic symptoms, low bone mineral density, and alterations in cognitive function have been described in some patients with FHH3 (McMurtry et al., 1992; Hannan et al., 2015a). Nearly all FHH3 cases are caused by a missense mutation of the AP2σ Arg15 residue (Arg15Cys, Arg15His, or Arg15Leu) (Fujisawa et al., 2013; Nesbit et al., 2013b; Hendy et al., 2014; Hannan et al., 2015a; Howles et al., 2016; Vargas-Poussou et al., 2016; Hovden et al., 2017). In addition, a genotype-phenotype correlation has been observed at the AP2σ Arg15 residue with the Arg15Leu mutation being associated with significant increases in serum calcium and an earlier age of presentation compared with patients harboring the Arg15Cys or Arg15His AP2σ mutations (Hannan et al., 2015a; Hovden et al., 2017). The AP2σ subunit forms part of the heterotetrameric AP2 complex, which is involved in clathrin-mediated endocytosis (Kelly et al., 2008), and the FHH3-causing AP2σ Arg15 mutations have been shown to reduce CaSR endocytosis and impair endosomal signaling from the internalized CaSR (Gorvin et al., 2018c). However, given the role of AP2 in clathrin-mediated endocytosis, it remains to be established whether phenotypic observations, such as cognitive deficits in FHH3, are attributable to CaSR dysregulation or potentially due to alterations in the endocytosis of other plasma membrane proteins.

C. Autosomal Dominant Hypocalcemia and Bartter Syndrome Type V

ADH is comprised of two genetically distinct variants, designated ADH1 and 2, which are caused by germline gain-of-function mutations of the CaSR and Gα₁₁, respectively (Table 2) (Hannan et al., 2016). ADH1 (OMIM 601198) is characterized by mild-to-moderate hypocalcemia in association with mild hypomagnesemia, hyperphosphatemia, and serum PTH concentrations that are usually detectable but within the lower half of the reference range (Nesbit et al., 2013a). Patients with ADH1 have significantly increased urinary calcium excretion compared with patients with hypoparathyroid (Yamamoto et al., 2000), and ∼10% of patients with ADH1 have an absolute hypercalciuria (Nesbit et al., 2013a). Some patients with ADH1 may have ectopic calcifications and/or elevations in bone mineral density (Pearce et al., 1996), and patients with a severe gain-of-function CaSR mutation may also develop a Bartter syndrome (referred to as Bartter syndrome type V) (Table 2), which is characterized by renal salt wasting leading to volume depletion, hyper-reninemic hyperaldosteronism, and hypokalemic alkalosis (Watanabe et al., 2002). Over 90 different ADH1-causing CaSR mutations have been reported (Hannan and Thakker, 2013; Hannan et al., 2016), and around 95% of these are heterozygous missense substitutions, whereas ∼5% are in-frame or frameshift insertion or deletion mutations (Hannan et al., 2012). ADH1 mutations cluster within the second loop of the VFT domain (residues 116–136), which contributes to the dimeric CaSR interface (Geng et al., 2016) (Fig. 4). A second ADH1 mutational hotspot is located in a region that encompasses transmembrane helices 6 and 7 and the intervening third ECL of the CaSR (residues 819–837) (Hannan et al., 2016). This transmembrane region may participate in a network of interactions with other transmembrane helices (Dore et al., 2014), thereby causing the CaSR to adopt an inactive conformational state.

ADH2 (OMIM 615361) (Table 2) has been reported in seven probands (Mannstadt et al., 2013; Nesbit et al., 2013a; Li et al., 2014a; Piret et al., 2016; Tenhola et al., 2016). Patients with ADH2 generally have mild-to-moderate hypocalcemia, in keeping with the serum biochemical phenotype of ADH1 (Hannan et al., 2016). However, ADH2 is associated with a milder urinary phenotype, with significantly reduced urinary calcium excretion compared with ADH1 (Li et al., 2014a). Moreover, short stature caused by postnatal growth insufficiency has been reported in two ADH2 kindreds (Li et al., 2014a; Tenhola et al., 2016). ADH2-causing mutations all comprise missense substitutions (Arg60Cys, Arg60Leu, Arg181Gln, Ser211Trp, Val340Met, and Phe341Leu), which enhance CaSR-mediated signaling responses, consistent with a gain of function (Nesbit et al., 2013a; Li et al., 2014a; Piret et al., 2016). ADH2-causing mutations cluster at the interface between the Gα₁₁ helical and GTPase domains (Piret et al., 2016) and may enhance the exchange of GDP and GTP, thereby leading to G protein activation. ADH2 mutations also affect the C-terminal portion of the Gα₁₁ protein, which facilitates G protein–GPCR coupling (Piret et al., 2016).

D. Animal Models of Genetic Diseases

Mouse models for FHH, NSHPT, and ADH have been generated using gene knockout and knock-in techniques and also by using chemical mutagenesis (Piret and Thakker, 2011).

E. Therapeutic Interventions—Successes and Failures

CaSR PAMs represent a targeted therapy for symptomatic forms of FHH (Hannan et al., 2018b) and potentiate the signaling responses of cells expressing FHH-associated CaSR, Gα₁₁, or AP2σ mutant proteins in vitro (Table 3) (Rus et al., 2008; Leach et al., 2013; Babinsky et al., 2016; Howles et al., 2016; Gorvin et al., 2018b). Furthermore, cinacalcet treatment is effective at decreasing serum calcium concentrations in patients with FHH1 and has been reported to improve hypercalcemic symptoms occasionally associated with FHH1, such as anorexia, polydipsia, and constipation (Table 3) (Alon and VandeVoorde, 2010; Rasmussen et al., 2011; Sethi et al., 2017). However, the response of NSHPT to cinacalcet is variable and appears to depend on the underlying CASR mutation. Indeed, cinacalcet rectifies the hypercalcemia and hyperparathyroidism in patients with NSHPT harboring a heterozygous Arg185Gln CaSR mutation (Reh et al., 2011; Gannon et al., 2014; Fisher et al., 2015) but is less effective for patients with NSHPT with biallelic truncating CASR mutations (Table 3) (García Soblechero et al., 2013; Atay et al., 2014), which would be a consequence of the truncated mutant receptor being unable to bind cinacalcet or couple with intracellular signaling proteins (Hannan et al., 2018b). Cinacalcet has also rectified the hypercalcemia in a mouse model for FHH2 (Howles et al., 2017) and ameliorated the hypercalcemia in a patient with symptomatic FHH2 (Table 3) (Gorvin et al., 2018b). Furthermore, cinacalcet is an effective therapy for symptomatic hypercalcemia caused by all three types of FHH3-causing Arg15 AP2σ mutations (Table 3) (Howles et al., 2016). However, hypocalcemic symptoms have occurred in a cinacalcet-treated child affected by FHH3 and the chromosome 22q11.2 deletion syndrome (Tenhola et al., 2015). Thus, long-term surveillance is required to detect hypocalcemia in cinacalcet-treated patients with FHH (Howles et al., 2016).

CaSR NAMs have been evaluated as a potential targeted therapy for ADH. In vitro studies have demonstrated that NPS 2143 normalizes the signaling responses associated with gain-of-function CASR and GNA11 mutations, which cause ADH1 and ADH2, respectively (Table 3) (Letz et al., 2010; Leach et al., 2013; Hannan et al., 2015b; Babinsky et al., 2016). However, NPS 2143 is less effective for gain-of-function mutations causing Bartter syndrome type V (Letz et al., 2010; Leach et al., 2013). In contrast, the quinazolinone-derived NAMs rectify gain-of-function CASR mutations that cause Bartter syndrome V (Table 3) (Letz et al., 2014). CaSR NAMs have also been characterized in vivo, and single-dose studies have demonstrated that NPS 2143 significantly increases circulating concentrations of calcium and PTH in ADH1 and ADH2 mouse models (Table 3) (Hannan et al., 2015b; Gorvin et al., 2017; Roszko et al., 2017). In addition, repetitive dosing studies have shown that the NAM, JTT-305/MK-5442, prevents the occurrence of nephrocalcinosis in mouse models of ADH1 (Dong et al., 2015). Furthermore, the NAM, NPSP795, has been administered to five ADH1 patients in a phase IIa clinical trial and increased plasma PTH concentrations and reduced urinary calcium excretion (Table 3) (Roberts et al., 2019). However, circulating calcium concentrations were not altered in these patients, and the optimal dosing regimen of NPSP795 for ADH remains to be established.