Cinacalcet corrects hypercalcemia in mice with an inactivating Gα11 mutation

Loss-of-function mutations of GNA11, which encodes G-protein subunit α11 (Gα11), a signaling partner for the calcium-sensing receptor (CaSR), result in familial hypocalciuric hypercalcemia type 2 (FHH2). FHH2 is characterized by hypercalcemia, inappropriately normal or raised parathyroid hormone (PTH) concentrations, and normal or low urinary calcium excretion. A mouse model for FHH2 that would facilitate investigations of the in vivo role of Gα11 and the evaluation of calcimimetic drugs, which are CaSR allosteric activators, is not available. We therefore screened DNA from > 10,000 mice treated with the chemical mutagen N-ethyl-N-nitrosourea (ENU) for GNA11 mutations and identified a Gα11 variant, Asp195Gly (D195G), which downregulated CaSR-mediated intracellular calcium signaling in vitro, consistent with it being a loss-of-function mutation. Treatment with the calcimimetic cinacalcet rectified these signaling responses. In vivo studies showed mutant heterozygous (Gna11+/195G) and homozygous (Gna11195G/195G) mice to be hypercalcemic with normal or increased plasma PTH concentrations and normal urinary calcium excretion. Cinacalcet (30mg/kg orally) significantly reduced plasma albumin–adjusted calcium and PTH concentrations in Gna11+/195G and Gna11195G/195G mice. Thus, our studies have established a mouse model with a germline loss-of-function Gα11 mutation that is representative for FHH2 in humans and demonstrated that cinacalcet can correct the associated abnormalities of plasma calcium and PTH.


Introduction
Familial hypocalciuric hypercalcemia (FHH) is an autosomal dominant disorder of extracellular calcium (Ca 2+ o ) homeostasis characterized by lifelong elevations in serum calcium concentrations in association with normal or mildly elevated serum parathyroid hormone (PTH) concentrations and normal or low fractional excretion of calcium (1)(2)(3)(4). FHH is caused by a reduction in the sensitivity of the Ca 2+ o -sensing receptor (CaSR) signaling pathway to alterations in the prevailing Ca 2+ o concentration ([Ca 2+ ] o ) (1)(2)(3)(4). The CaSR is a widely expressed family C GPCR that regulates PTH secretion and urinary calcium excretion by transducing elevations in [Ca 2+ ] o into multiple intracellular signaling cascades in the parathyroid glands and kidneys, respectively (5,6). In the parathyroid glands, the CaSR has been shown to couple to the G q/11 protein family (7), which activates phospholipase C (PLC), thereby increasing intracellular calcium (Ca 2+ i ) and MAPK signaling responses (8,9), which in turn leads to decreased parathyroid PTH secretion.
Loss-of-function mutations of GNA11, which encodes G-protein subunit α 11 (Gα 11 ), a signaling partner for the calcium-sensing receptor (CaSR), result in familial hypocalciuric hypercalcemia type 2 (FHH2). FHH2 is characterized by hypercalcemia, inappropriately normal or raised parathyroid hormone (PTH) concentrations, and normal or low urinary calcium excretion. A mouse model for FHH2 that would facilitate investigations of the in vivo role of Gα 11 and the evaluation of calcimimetic drugs, which are CaSR allosteric activators, is not available. We therefore screened DNA from > 10,000 mice treated with the chemical mutagen N-ethyl-N-nitrosourea (ENU) for GNA11 mutations and identified a Gα 11 variant, Asp195Gly (D195G), which downregulated CaSR-mediated intracellular calcium signaling in vitro, consistent with it being a loss-of-function mutation. Treatment with the calcimimetic cinacalcet rectified these signaling responses. In vivo studies showed mutant heterozygous (Gna11 +/195G ) and homozygous (Gna11 195G/195G ) mice to be hypercalcemic with normal or increased plasma PTH concentrations and normal urinary calcium excretion. Cinacalcet (30mg/kg orally) significantly reduced plasma albumin-adjusted calcium and PTH concentrations in Gna11 +/195G and Gna11 195G/195G mice. Thus, our studies have established a mouse model with a germline loss-offunction Gα 11 mutation that is representative for FHH2 in humans and demonstrated that cinacalcet can correct the associated abnormalities of plasma calcium and PTH.
A mouse model for FHH1 has previously been generated by targeted germline disruption of the Casr gene, and heterozygous (Casr +/-) mice were shown to have a phenotype resembling that of FHH1 patients with elevated serum concentrations of calcium and PTH, and low urinary calcium excretion (12). In addition, homozygous (Casr -/-) mice had features of neonatal severe hyperparathyroidism (NSHPT), which is caused by biallelic inactivating CaSR mutations (1), and exhibited growth retardation and died within the first 30 days of life (12). An in vivo model is not available for FHH2, although mice with parathyroid-specific combined ablations of both the Gna11 and Gnaq (encoding Gα q ) genes have previously been reported to develop marked hypercalcemia and hyperparathyroidism (7). We therefore sought to establish a mouse model for FHH2 to define the in vivo role of Gα 11 in Ca 2+ o homeostasis and to undertake a more detailed characterization of the phenotype of this disorder, as limited information is available from the few FHH2 patients reported, to date (3,10). In addition, a mouse model for FHH2 would facilitate evaluation of therapeutic drugs such as CaSR allosteric activators, also known as calcimimetics (13). To establish a mouse model for FHH2, due to a germline lossof-function GNA11 point mutation (3,10), we screened a DNA archive of > 10,000 samples from male mice that had mutations induced by treating them with N-ethyl-N-nitrosourea (ENU), a chemical mutagen. ENU is an alkylating agent that introduces point mutations via transfer of an alkyl group from ENU to a DNA base, thus leading to mispairing and bp substitution during subsequent DNA replication (14,15). ENU mutagenesis programs utilize two complementary approaches that are phenotype-driven and genotype-driven screens. In phenotype-driven screens, offspring of mutagenized mice are assessed for abnormalities in a hypothesis-generating strategy, which may elucidate new genes, pathways, and mechanisms for disease phenotypes (14,15). Genotype-driven screens in which mutations in the gene of interest are sought are hypothesis driven and are feasible by available parallel archives of tissue-DNA and sperm samples from mutagenized male mice (14,15). The archived tissue-DNA samples from the mutagenized male mice are used to search for the mutations in the gene of interest, and once these mutations are found, a sperm sample from the male mouse with the mutation is used for in vitro fertilization (IVF) of normal female mice to establish progeny with the mutation (14,15). The probability of finding 3 or more variant alleles in an archive of tissue-DNA samples from > 5,000 ENU-mutagenized mice is > 90% (14). We sought for ENU-induced Gna11 variants in tissue-DNA samples from > 10,000 male mice treated with ENU, with the aim of establishing a mouse model for FHH2.

Results
Identification and analysis of 5 Gna11 variants in ENU-mutagenized mice. An analysis using melting curve analysis (16) of tissue-DNA samples from > 10,000 ENU-mutagenized male mice of the 7 exons and 12 intron-exon boundaries of the Gna11 gene revealed the presence of 5 Gna11 variants, comprising c.379C>T, c.395T>A, c.440G>A, c.584A>G, and c.806T>C (numbering starts from ATG; Supplemental Table 1; supplemental material available online with this article; https://doi.org/10.1172/jci.insight.96540DS1; numbering in Supplemental Table 1 starts from 5′UTR.). These 5 Gna11 variants predicted the occurrence of 4 missense variants (Ile132Asn, Arg147His, Asp195Gly, and Val269Ala) and 1 nonsense variant (Gln127Stop) ( Figure 1A and Supplemental Figure 1). Bioinformatic analysis predicted all the Gα 11 variants to be damaging and likely disease-causing (Supplemental Table 1). FHH2 has been reported to be caused by either an in-frame deletion or missense substitutions affecting Gα 11 (3,10), and we therefore further characterized only the 4 missense Gα 11 variants identified in ENU-mutagenized mice. All of these 4 missense variants affected evolutionary-conserved residues ( Figure 1B, Supplemental Table 1, and Supplemental Figure 1), and 2 variants (Asp195Gly and Val269Ala) were located in the Gα 11 GTPase domain, which mediates GPCR binding, guanosine triphosphate (GTP) hydrolysis, and effector coupling. The other 2 variants (Ile132Asn and Arg147His) were located in the Gα 11 helical domain, which stabilizes guanine nucleotide binding ( Figure 1A and Supplemental Figure 1) (17). Three-dimensional (3-D) modeling using the reported crystal structure of the related Gα q protein (18) predicted the Asp195Gly variant to disrupt polar contacts within the Gα 11 GTPase domain (Figure 1, C and D), whereas the other missense variants were not predicted to alter intramolecular interactions within the Gα 11 protein (Supplemental Figure 1). We therefore selected the Asp195Gly (D195G) variant for functional characterization for the following 4 reasons. First, this variant is located within the switch regions of the Gα 11 GTPase domain ( Figure 1, B and C), which are critical for mediating Gα-subunit conformational changes upon GTP binding and also for coupling to downstream effector proteins such as PLC (19,20). Second, the Asp195Gly variant is situated within a 13 amino acid region (residues 193-205), which links switches I and II (Figure 1, B and C) and is the location of a reported FHH2-causing Gα 11 mutation (Ile200del) (3). Third, this 13-amino acid linker region also contains the tetrapeptide β2-β3 loop (residues 196-199), which mediates G-protein-GPCR inter-actions (21) (Figure 1C), and our reported mutagenesis studies have shown that disruption of the Gα 11 β2-β3 loop impairs signaling in CaSR-expressing cells (3). Fourth, 3-D modeling of the Asp195Gly Gα 11 variant predicted that substitution of the WT Asp195 residue with the variant Gly195 residue would lead to a loss of a polar contact within the Gα 11 β2-β3 loop, which would likely disrupt this tetrapeptide loop ( Figure 1D) and thereby impair GPCR binding and Gα 11 activation (3,19,20). These combined observations indicated that the Asp195Gly variant was highly likely to be a pathogenic mutation.
In vitro functional characterization of the Asp195Gly Gα 11 mutation. To investigate the effects of these predicted Gα 11 structural changes due to the Asp195Gly mutation on CaSR-mediated signaling, human embryonic kidney 293 (HEK293) cells stably expressing the CaSR (HEK-CaSR) were transiently transfected with pBI-CMV2-GNA11-GFP constructs expressing either the WT (Asp195) or variant (Gly195) Gα 11 proteins, as reported (3). This bidirectional pBI-CMV2 vector allows for coexpression of Gα 11 and GFP at equivalent levels (3). Expression of the CaSR, Gα 11 , and GFP was confirmed by fluorescence microscopy and/or Western blot analyses (Figure 2, A and B). The expression of Gα 11 was shown to be similar in cells transiently transfected with WT or mutant proteins and to be greater than that observed in untransfected cells ( Figure 2B). Moreover, the expression of mutant Gα 11 in cells that endogenously express WT Gα 11 ( Figure 2B) corresponded to the heterozygous situation reported in FHH2 patients (3,10). The Ca 2+ i responses to alterations in [Ca 2+ ] o of cells expressing the different GNA11 vectors were assessed using a multiwell assay that utilized the Fluo-4 Ca 2+ -binding dye, as reported (22). The Ca 2+ i responses were shown to increase in a dose-dependent manner following stimulation with increasing [Ca 2+ ] o ( Figure 2C). However, responses in mutant Gly195-expressing . These results demonstrated that the Gα 11 Asp195Gly mutation is a loss-of-function mutation, similar to mutations that lead to FHH2 (3,10). We next investigated the ability of the CaSR allosteric activator, cinacalcet, to rectify this loss of function associated with the Asp195Gly Gα 11 mutation. Cinacalcet was added to Gly195 mutant cells at a 10 nM concentration, as this dose has previously been reported to normalize the altered signaling responses associated with FHH2-causing Gα 11 mutations in vitro (23). An assessment of Ca 2+ i responses showed 10 nM cinacalcet to induce a leftward shift of the concentration-response curve of cells expressing the Gly195 mutant Gα 11 protein ( Figure 2C) and decrease their mean EC 50 value to 2.70 mM (95% CI, 2.60-2.80 mM), a value that was indistinguishable from the EC 50 of untreated WT cells (Figure 2, C and D). Thus, cinacalcet normalized the signaling responses of Gly195 mutant cells. In vivo functional analysis in mice harboring the germline Gna11 Asp195Gly mutation. To investigate the in vivo effects of the Asp195Gly Gα 11 mutation on Ca 2+ o homeostasis, ENU mutagenesis-derived mice harboring this mutation were established on the C3H inbred genetic background (24). DNA sequence analysis confirmed the mutant mice to harbor a germline A-to-G transition at c.584A>G at codon 195 of the Gα 11 protein resulting in an Asp (D) to Gly (G) missense substitution ( Figure 3, A and B). This mutation led to a gain of a HaeIII restriction endonuclease site ( Figure 3C), which was used to confirm the presence of the mutation ( Figure 3D) and to genotype the subsequent generations of mice. Heterozygous-affected (Gna11 +/195G ) mice were healthy and fertile, and an analysis of offspring bred from crosses of Gna11 +/195G mice yielded homozygous-affected (Gna11 195G/195G ) mice and significant deviations from the Mendelian inheritance expected ratio of 1:2:1 for the WT (Gna11 +/+ ), Gna11 +/195G , and Gna11 195G/195G genotypes were not observed among the weaned mice, thereby indicating that the homozygous Gna11 195G/195G mice were viable and did not have embryonic or neonatal lethality (Table 1). Moreover, Gna11 195G/195G mice had a normal body weight compared with WT (Gna11 +/+ ) and Gna11 +/195G littermates (Table 2). Thus, Gna11 195G/195G mice did not have evidence of growth retardation or neonatal lethality to suggest an NSHPT phenotype. However, plasma biochemical analysis revealed Gna11 +/195G and Gna11 195G/195G mice to be significantly hypercalcemic compared with Gna11 +/+ mice ( Figure 4A). Moreover, Gna11 195G/195G mice had significantly reduced plasma phosphate concentrations and raised  The Mendelian inheritance expected ratio from heterozygous crosses is 1:2:1, and χ 2 analysis shows no significant differences in the expected vs. observed ratios of offspring genotypes at weaning (i.e., 19-21 days of age) (x 2 = 1.0, degrees of freedom = 2).
plasma PTH concentrations when compared with Gna11 +/+ mice, whereas Gna11 +/195G mice had plasma phosphate and PTH concentrations that were similar to those of Gna11 +/+ mice (Figure 4, B and C). Furthermore, the fractional excretion of calcium was not altered in Gna11 +/195G or Gna11 195G/195G mice compared with Gna11 +/+ mice ( Figure 4D and Table 3). However, there were sex differences in these calcitropic phenotypes, as follows. Female Gna11 195G/195G mice were significantly more hypercalcemic than male Gna11 195G/195G mice and female Gna11 +/195G mice (Table 2 and Supplemental Figure 2). In addition, female Gna11 195G/195G mice, but not the Gna11 mutant males, had significant hypophosphatemia, with a significant reduction in the tubular maximum reabsorption of phosphate (Table 3) and a raised alkaline phosphatase activity compared with female Gna11 +/+ mice (Table 2 and Supplemental Figure  2). Significant differences were not observed in plasma electrolytes, urea and creatinine concentrations, or 1,25-dihydroxyvitamin D or fibroblast growth factor-23 (FGF-23) concentrations in male or female Gna11 +/195G and Gna11 195G/195G mice, when compared with respective Gna11 +/+ mice ( Table 2). The fractional excretions of sodium and potassium were also not different between male and female mutant mice and respective Gna11 +/+ mice (Table 3). Finally, whole body dual-energy X-ray absorptiometry (DXA) did not reveal significant differences in the bone mineral content or bone mineral density (BMD) between male and female mutant mice and respective Gna11 +/+ mice (Table 4).

Discussion
We have established a mouse model for FHH2, and this will enable the calcitropic roles of Gα 11 to be further evaluated and also facilitate further pathophysiological studies that are difficult to pursue in the few reported patients with this condition. Our results revealed that heterozygous-affected (Gna11 +/195G ) mice had a similar plasma biochemical phenotype to that reported for FHH2 patients, who also harbor heterozygous loss-of-function Gα 11 mutations (Table 5) (3,10). Thus, Gna11 +/195G mice had mild hypercalcemia in association with normal plasma PTH concentrations; they also had no alterations in the plasma concentrations of phosphate and creatinine, or in alkaline phosphatase activity, which is consistent with   the reported phenotype of FHH2 patients (Table 5) (3,10). Gna11 +/195G mice additionally had normal plasma magnesium concentrations, which is consistent with one reported FHH2 proband (3) but which contrasts with the hypermagnesemia reported in a multigenerational FHH2 kindred (3). A key finding of this study is that Gna11 +/195G and Gna11 195G/195G mice had no alterations in urinary calcium excretion, and this would be consistent with studies of FHH2 patients, which have reported that not all FHH2 patients have a low fractional excretion of calcium (Table 5) (3,10). The absence of a urinary calcium phenotype in Gna11 +/195G and Gna11 195G/195G mice is also consistent with the reported findings in mice and humans harboring germline gain-of-function Gα 11 mutations that is associated with hypocalcemia and reduced plasma PTH concentrations but with mild or no alterations in urinary calcium excretion (25)(26)(27). These studies highlight a potential difference in the calcitropic phenotype of disorders caused by germline Gα 11 mutations and that of disorders caused by germline CaSR mutations, and they suggest that the Gα 11 protein may not play a major role in the renal handling of calcium. Thus, it remains to be established whether hypocalciuria represents a major component of the FHH2 disorder in humans. Furthermore, DXA analysis did not reveal any alterations in the BMD values of Gna11 195G/195G mice, which also suggests that the Gα 11 protein may not influence bone mass. Our studies of homozygous-affected (Gna11 195G/195G ) mice have highlighted the importance of Gα 11 for parathyroid gland function and PTH secretion, as Gna11 195G/195G mice had more pronounced hypercalcemia and hypophosphatemia, and significantly raised plasma PTH concentrations, consistent with primary hyperparathyroidism (28). Moreover, female Gna11 195G/195G mice also had significant elevations of plasma alkaline phosphatase activity, which is consistent with an elevated bone turnover associated with this likely primary hyperparathyroidism. However, the hypercalcemic phenotype of Gna11 195G/195G mice was, in general, milder than that observed in humans or mice harboring biallelic loss-of-function CaSR mutations, which typically lead to the life-threatening disorder of NSHPT (12,24). A possible explanation for the milder hypercalcemic phenotype observed in the Gna11 195G/195G mice is that the loss of Gα 11 function caused by the Asp195Gly mutation in vivo was partially compensated by the WT Gα q protein, which in the parathyroid glands continues to mediate signal transduction by the CaSR. Indeed, the importance of the Gα 11 and Gα q proteins for parathyroid gland function has been demonstrated by studies of mice with a parathyroid-specific ablation of both Gα 11 and Gα q , which have been reported to develop features of NSHPT such as severe hypercalcemia, skeletal demineralization, growth retardation, and early postnatal death (7). The hypercalcemia observed in Gna11 195G/195G mice was more severe in females compared with males, and such sex differences have not previously been reported in studies of FHH patients. However, sex differences have been noted in primary hyperparathyroidism patients, with females being more commonly affected than males (29). Moreover, estrogen may play a role in the pathogenesis and severity of primary hyperparathyroidism, as highlighted by a study that showed the potential involvement of estrogen signaling in parathyroid function and disease (30); such effects may have contributed to the more severe hypercalcemia of female Gna11 195G/195G mice.
There is currently no effective treatment for FHH2, and we therefore evaluated the therapeutic potential of cinacalcet, which is a licensed CaSR-positive allosteric modulator (13), for this condition. In vitro studies have previously reported that nanomolar concentrations of cinacalcet can successfully rectify the altered signaling responses of HEK-CaSR cells expressing FHH2-associated Gα 11 mutant proteins (23). Consistent with these findings, our study showed a 10 nM concentration of cinacalcet to normalize the Ca 2+ i responses of HEK-CaSR cells expressing the mutant Gly195 Gα 11 protein. Moreover, oral administration of a single 30 mg/kg cinacalcet dose led to a transient suppression of PTH secretion in Gna11 +/195G and Gna11 195G/195G mice, and this was associated with a sustained reduction in plasma calcium concentrations, which lasted for ≥ 6 hours. This dose of cinacalcet was well tolerated in the mice and did not lead to hypocalcemia, with mean plasma calcium concentrations remaining at > 2.0 mmol/l. However, transient hyperphosphatemia was noted in cinacalcet-treated mice, which was likely to be a consequence of suppressed PTH secretion (31). These results suggest that calcimimetics such as cinacalcet will likely be of benefit for FHH2 patients, who also harbor loss-of-function Gα 11 mutations (23).
In summary, we have established a mouse model for FHH2 and have shown the in vivo efficacy of cinacalcet in reducing plasma calcium and PTH concentrations, thereby illustrating the potential utility of this CaSR allosteric modulator for the treatment of hypercalcemia in patients with FHH2. A Kruskal-Wallis test followed by Dunn's test for nonparametric pairwise multiple comparisons were used for analysis of A-I. *P < 0.05, **P < 0.01 compared with respective untreated mice. $ Untreated Gna11 195G/195G mice were significantly (P < 0.05) hypercalcemic compared with untreated Gna11 +/+ mice.

Methods
Animals. ENU-treated G0 C57BL/6J male mice (The Jackson Laboratory) were mated to C3H/HeH (C3H) mice (MRC Harwell) to produce G1 progeny, and tissue-DNA samples from > 10,000 G1 ENU mutagenized male mice -together with their sperm -was archived, as reported (14). These tissue-DNA samples were used to identify Gna11 variants by melt curve analysis of PCR products utilizing a Lightscanner and gene-specific primers (BioFire Diagnostics Inc.), and sperm from mice with Gna11 variants was used for IVF to generate G2 progeny on a C3H background strain, as reported (16,24). Heterozygous-affected (Gna11 +/195G ) mutant male and female mice were intercrossed to generate homozygous (Gna11 195G/195G ) mice, which were studied along with their Gna11 +/195G and WT (Gna11 +/+ ) littermates. All study mice were housed in a controlled environment at the MRC Harwell Institute in accordance with UK Home Office and MRC Welfare guidance. Mice were fed on a standard diet (Rat and Mouse number 3, Special Diet Services) that contained 1.15% calcium, 0.58% phosphate, and 4089 IU/kg of vitamin D, and they were provided with water ad libitum (25,32).
Compounds. Cinacalcet (AMG-073 HCL) was obtained from Cambridge Bioscience (catalog CAY16042) and dissolved in a 20% aqueous solution of 2-hydroxypropyl-β-cyclodextrin (MilliporeSigma, catalog H107) prior to use in in vitro and in vivo studies.
Protein sequence alignment and 3-D modeling. Protein sequences of Gα 11 orthologs and paralogs were aligned with Clustal Omega (35). The PyMOL Molecular Graphics System (Version 1.2r3pre, Schrödinger LLC) was used for structural modeling based on the complexed crystal structure of Gα q , which has 90% Skeletal imaging. Bone mineral content and density were assessed by whole body DXA scanning, which was performed on mice anesthetized by inhaled isoflurane and using a Lunar Piximus densitometer (GE Medical Systems), as reported (25). DXA images were analyzed using Piximus software, as reported (25).
Statistics. All in vitro studies involved 8 biological replicates. Statistical comparisons of the Ca 2+ i EC 50 responses were undertaken using the F-test, as reported (3). For the in vivo studies, a Kruskal-Wallis test was undertaken for multiple comparisons, and any significant differences identified were further assessed using the Dunn's test for nonparametric pairwise multiple comparisons (25). All analyses were performed using GraphPad Prism (GraphPad), and a value of P < 0.05 was considered significant for all analyses.
Study approval. Animal studies were approved by the MRC Harwell Institute Ethical Review Committee and were licensed under the Animal (Scientific Procedures) Act 1986, issued by the UK Government Home Office Department (PPL30/2433 and PPL30/3271).