Autosomal Dominant Hypocalcemia in Monozygotic Twins Caused by a De Novo Germline Mutation Near the Amino-Terminus of the Human Calcium Receptor

Authors


  • The authors have no conflict of interest.

Abstract

To define the molecular pathogenesis of severe postnatal hypocalcemia in monozygotic twin sisters, we sequenced their CaR gene and identified a missense mutation, K29E. Expression of the mutant receptor in vitro showed a marked increase in Ca2+ sensitivity explaining the observed phenotype. Additional mutagenesis studies lead us to speculate concerning a novel mechanism whereby the K29E mutation may lead to receptor activation.

Introduction: Activating mutations of the Ca2+-sensing receptor (CaR) gene have been identified in subjects with autosomal dominant hypocalcemia. Study of such mutations has provided insight into the mechanism of activation of the CaR.

Materials and Methods: We performed biochemical and molecular genetic studies on monozygotic twin sisters who presented with early postnatal hypocalcemia and on their unaffected sister and parents. Functional characterization of mutant CaRs transfected in HEK-293 cells included immunoblots to monitor protein expression and Ca2+ stimulation of phosphoinositide hydrolysis to measure Ca2+ sensitivity.

Results: We identified a K29E missense mutation in the twin sisters but not in their parents or unaffected sister. The K29E mutant CaR showed a marked increase in Ca2+ sensitivity, including when it was co-transfected with wildtype CaR cDNA, consistent with a dominant effect. Substitution of K29 by aspartate equivalently increased CaR sensitivity, whereas conservative substitution by arginine did not.

Conclusions: Severe postnatal hypocalcemia in the twin sisters was caused by a de novo germline activating mutation. In a model of the Venus flytrap-like domain of the extracellular amino-terminus of the CaR, K29 is located close to a peptide loop, “loop 2,” that forms part of the dimer interface and is the site of 10 of the previously reported naturally occurring activating CaR mutations. We speculate that K29E increases Ca2+ sensitivity of the CaR by disrupting a salt bridge between K29 and an acidic residue in loop 2 and thereby changes the normal structure of loop 2 that maintains the CaR in its inactive conformation.

INTRODUCTION

THE CRITICAL ROLE of the extracellular Ca2+-sensing receptor (CaR) in Ca2+ regulation of parathyroid hormone (PTH) secretion and of renal Ca2+ reabsorption was demonstrated by the identification of loss-of-function and gain-of-function mutations of the human CaR (hCaR) gene in subjects with genetic disorders of extracellular Ca2+ homeostasis, familial hypocalciuric hypercalcemia (FHH) and autosomal dominant hypocalcemia (ADH), respectively.(1) Naturally occurring missense mutations identified in subjects with FHH and ADH have provided important insights into key structure-function relationships of the CaR.(2)

In this report, we describe the clinical findings in monozygotic twin sisters who presented in the neonatal period with signs and symptoms of severe hypocalcemia and in whom a heterozygous, de novo germline mutation, K29E, of the CaR was identified. Functional characterization of the mutation showed that it increases sensitivity to Ca2+, as found for other activating CaR mutations that cause ADH. K29E is the most amino-terminal activating CaR mutation reported to date.(2) Our in vitro functional studies of mutations of K29 and its adjacent residues K28 and Q27 lead us to speculate concerning a novel mechanism for activation of the hCaR by the K29E mutation.

MATERIALS AND METHODS

Patients

Biochemical and clinical studies were conducted on monozygotic twin sisters of an Italian family (Fig. 1A). The findings are described in detail in the Results section. Informed consent for the molecular studies was obtained from the parents of the sisters.

Figure FIG. 1..

(A) Pedigree of the Italian family. Monozygotic twin sisters (arrows) presented shortly after birth with severe hypocalcemia, low serum concentration of PTH, and high phosphate levels. Neither parents nor elder sister were hypocalcemic. (B) Identification of the CaR mutation in monozygotic twin sisters. Automated sequence analysis of PCR-amplified genomic DNA from the affected subjects shows a heterozygous A to G transition at position 85 in exon 2 of the CaR (right), changing Lys to Glu (K29E). The normal sequence of the mother is shown at left.

Biochemical analyses

Blood was drawn in the morning after an overnight fast. Serum calcium, phosphate, and magnesium concentrations were determined by standard automated methods. Creatinine and calcium levels were also measured in 24-h urine collections by automated methods.

DNA amplification, sequence analysis, subcloning, and allele-specific PCR

Genomic DNA was isolated from white blood cells by a modified simple salting out procedure,(3) and the coding region of the CaR gene was PCR-amplified using primers and PCR conditions as previously described.(4)

PCR products were purified (GFX PCR DNA and Gel Band Purification Kit, Amersham Biosciences, Piscataway, NJ, USA) and directly sequenced with either forward or reverse primers used for PCR using the DYEnamic ET Dye Terminator Kit (Amersham Biosciences) and resolved by capillary electrophoresis on MegaBACE 1000 DNA Analysis System (Amersham Biosciences). The sequence of the PCR products in which mutations were identified were repeated twice from two different PCR products.

Exon 2 of CaR was amplified from leukocyte DNA of individuals I-1, I-2, and II-2, and subcloned into the TA cloning vector pCR2.1- TOPO (Invitrogen Life Technologies). Ninety-six subclones from each individual were examined by direct sequencing analysis and scored as wildtype (WT) or mutant sequence. In addition, we performed allele-specific PCR in individuals I-1, I-2, II-2, and II-3, according to the method of Kaltenböck and Schneider.(5)

Zygosity determination

Zygosity determination was carried out in samples from individuals II-2 and II-3 using 10 microsatellite markers (D2S305, D2S347, D3S1767, D3S3582, D3S3640, D7S657, D7S798, D14S258, D15S165, and D21S1252).

Site-directed mutagenesis of the hCaR

The hCaR cDNA cloned in the pCR3.1 expression vector was described previously.(6) Site-directed mutagenesis was performed using the Quickchange site-directed mutagenesis kit (Stratagene, La Jolla, CA, USA), according to the manufacturer's instructions. Parental hCaR cDNA in pCR3.1 vector was amplified using pfu Turbo DNA polymerase with mutagenic oligonucleotide primers (sequences available on request) for 16 cycles in a DNA thermal cycler (Perkin-Elmer, Norwalk, CT, USA). After digestion of the parental DNA with DpnI for 1 h, the amplified DNA with incorporated nucleotide substitution was transformed into E. coli (DH-5α strain). The mutations were confirmed by automated DNA sequencing using a dRhodamine Terminator Cycle Sequencing kit and ABI PRISM-373A DNA sequencer (PE Applied Biosystems, Foster City, CA, USA).

Transient transfection of WT and mutant receptors in HEK-293 cells

Transfections were performed using 12 μg (unless otherwise indicated) plasmid DNA for each transfection in a 75-cm2 flask of HEK-293 cells. DNA was diluted in serum-free DMEM (BioFluids, Rockville, MD, USA) and mixed with diluted Lipofectamine (GIBCO/BRL, Grand Island, NY, USA), and the mixture was incubated at room temperature for 30 minutes. The DNA-Lipofectamine complex was further diluted in 6 ml serum-free DMEM and was added to 80% confluent HEK-293 cells plated in 75-cm2 flasks. After 5 h of incubation, 15 ml complete DMEM containing 10% FBS (BioFluids) was added. Twenty-four hours after transfection, transfected cells were split and cultured in complete DMEM.

Phosphoinositide hydrolysis assay

Phosphoinositide (PI) hydrolysis assay has been described previously.(6) Briefly, 24 h after transfection, transfected cells from a confluent 75-cm2 flask were split. Typically one-eighth of cells were plated in one well in a 6-well plate, and whole cell lysate was prepared 48 h after transfection for Western blot assay. The remaining cells were plated in two 12-well plates in complete DMEM medium containing 3.0 μCi/ml of3H-myoinositol (New England Nuclear, Beverly, MA, USA) and cultured for another 24 h. Culture medium was replaced by 1× PI buffer (120 mM NaCl, 5 mM KCl, 5.6 mM glucose, 0.4 mM MgCl2, and 20 mM LiCl in 25 mM PIPES buffer, pH 7.2) and incubated for 1 h at 37°C. After removal of PI buffer, cells were incubated for an additional 1 h with different concentrations of Ca2+ in 1× PI buffer. The reactions were terminated by addition of 1 ml of acid-methanol (1:1000 vol/vol) per well. Total inositol phosphates were purified by chromatography on Dowex 1-X8 columns, and radioactivity for each sample was counted with a liquid scintillation counter. Graphs of concentration dependence for stimulation of PI hydrolysis by [Ca2+]0 for each transfection were drawn by using GraphPad Prism version 2.0 software. Each value on a curve is the mean of duplicate determinations. Graphs shown in this paper were representative ones from at least three independent experiments.

Immunoblotting analysis with detergent-solubilized whole cell extracts

Confluent cells in 6-well plates were rinsed with ice-cold PBS and scraped on ice in lysis buffer containing 20 mM Tris-HCl (pH 6.8), 150 mM NaCl, 10 mM EDTA, 1 mM EGTA, 1% Triton X-100, and freshly added protease inhibitors cocktail (Boehringer Mannheim, Indianapolis, IN, USA). Fifty micrograms of protein per lane was separated on a 5% SDS-PAGE gel under reducing conditions with 300 mM β-mercaptoethanol added. The proteins on the gel were electrotransferred onto nitrocellulose membrane and incubated with 0.1 μg/ml of protein A-purified mouse monoclonal anti-hCaR antibody ADD (raised against a synthetic peptide corresponding to residues 214-235 of hCaR protein). Subsequently, the membrane was incubated with a secondary sheep anti-mouse antibody conjugated to horseradish peroxidase (Amersham Biosciences) at a dilution of 1:2000. The hCaR protein was detected with an enhanced chemiluminescence system (ECL; Amersham Biosciences). For the detection of expression of β-tubulin as an internal control for equal loading of protein, samples were separated on 7.5% SDS-PAGE. The nitrocellulose membrane was stripped after blotting with ADD and followed by incubation with rabbit polyclonal anti-β-tubulin (Santa Cruz Biotech, Santa Cruz, CA, USA) at a dilution of 1:2000 and subsequent incubation with a secondary donkey anti-rabbit antibody conjugated to horseradish peroxidase (Amersham Biosciences) at a dilution of 1:2000.

Immunoprecipitation of WT and mutant hCaR receptors

Six hundred micrograms total protein of the whole cell lysate was incubated with 5 μg affinity-purified rabbit polyclonal hCaR-specific antibody GGD (raised against a synthetic peptide corresponding to amino acids 1037-1050 of the hCaR) for 1-2 h at 4°C. Subsequently, 25 μl of protein A agarose (Santa Cruz Biotech) was added, and the incubation was continued for an additional 1-2 h at 4°C. The protein A agarose was washed three times with lysis buffer containing 0.1% SDS, and the immunoreactive proteins bound on agarose were eluted in 100 μl of 1× sample buffer containing 300 mM β-mercaptoethanol at room temperature for 5 minutes. Fifty microliters of sample was loaded per lane, and immunoblotting was performed as described in the last section.

RESULTS

Clinical and biochemical studies

The probands are monozygotic twins (II-2 and II-3 in Fig. 1A) who were born at 34 weeks of gestation. At 7 days of life, apnea crisis, tremors, and tonic contractions of the limbs were noted in subject II-2. She was hospitalized, and a serum total calcium level of 1.25 mM (5 mg/dl) was found. At 8 days of life, the other twin sister (II-3) showed generalized seizure and cyanosis. Her serum total calcium concentration was 0.99 mM (4 mg/dl). Concomitantly, hyperphosphatemia and undetectable PTH serum concentrations were found in both sisters. The diagnosis of congenital hypoparathyroidism was made, and treatment with oral calcium and dihydrotachysterol (0.4-0.6 mg/day) was initiated.

The patients were referred to the Pediatric Unit of the University of Milan at 20 months of age and followed thereafter. Both sisters showed low serum calcium concentrations, high serum phosphorus levels, and high urinary calcium excretion. Vitamin D treatment was then changed, and 1,25-dihydroxyvitamin D was given at a daily dose of 30 ng/kg, along with low doses of oral calcium. Since then, therapeutic regimen has been modified to keep serum calcium levels in the low-normal range and to minimize renal calcium excretion. Several attempts have been made to lower calcium excretion with thiazide diuretics. A good initial response to treatment was followed by a period of unresponsiveness and finally by the appearance of severe side effects (hypokalemia, arrhythmia), which required cessation of treatment. Despite these efforts, nephrocalcinosis was first documented by renal ultrasound at the age of 3 and 4 years in both subjects (II-2 and II-3, respectively). Small renal stones were seen by ultrasound in subject II-2 at the age of 15.5 years. Similarly, subject II-3 developed nephrolithiasis at 12.6 years. Renal function was normal throughout the observation period. Neither of the patients had signs of cataract. Both sisters had normal growth and a physiological menarche.

Concomitant serum calcium, phosphate and magnesium levels, and urinary calcium excretion (expressed as urinary calcium/urinary creatinine ratio) during treatment are shown in Fig. 2. A period of recurrent hypercalciuria coincided with the first years of primary school (age, 4-8 years), during which the patients had frequent episodes of fever caused by infections of the upper respiratory tract. Low-normal magnesium concentrations have been a constant finding in these patients.

Figure FIG. 2..

Concomitant changes of serum calcium, phosphate and magnesium, and urinary calcium excretion of subjects (A) II-2 and (B) II-3 during treatment. Appearance of nephrocalcinosis and treatment with thiazide diuretics are shown. The shaded areas indicate normal ranges.

The parents and the elder sister had normal serum calcium, phosphate, and PTH levels and never experienced seizures or tremors.

Detection of a point mutation in the CaR gene

The sequence analysis of PCR amplified genomic DNA revealed a heterozygous A to G transition at nucleotide 85 (with nucleotide “A” of the initiator ATG counted as +1) in exon 2 of the CaR gene in the patients (Fig. 1B). This substitution causes a missense mutation, K29E, in the extracellular domain of the receptor (Figs. 3A and 3B). Both parents and an elder sister showed a normal CaR gene sequence (Fig. 1). Comparison of 10 microsatellite markers confirmed the monozygosity of the twin sisters. Sequencing of multiple subclones of amplified exon 2 of the CaR, as well as allele-specific amplification, performed to search for somatic mosaicism showed only a normal sequence of the CaR gene in both parents.

Figure FIG. 3..

(A) Schematic diagram showing amino acid sequence of the hCaR ECD. The location of signal peptide, glycosylation sites, and the sequence of synthetic polypeptide used to raise monoclonal antibody ADD are indicated. All 19 cysteines are shown in black. The beginning and end of the VFT domain and the four loops in lobe 1 of the VFT are indicated. Naturally occurring activating mutations previously identified in the ECD, as well as the inactivating Q27R mutation (boxed) and the K29E mutation reported herein (bold), are also indicated. (B) The model of the VFT of the human Ca2+ receptor (left). α-Helices are red, and β sheets are yellow. Loops and turns are green, except for the loops labeled I-IV, which are dark red, orange, blue, and purple, respectively. The side chains of cysteine residues are shown as space-filling, and the corresponding residue numbers are given. Lobes I and II and the amino (N) and carboxyl (C) termini are labeled. On the right, the portion of the model containing loop 2 and residue K29 is shown in an expanded view with acidic residue D121 in loop 2 indicated. The model was rendered using the programs MolScript and Raster3D.

Functional characterization of the K29E mutant hCaR

By site-directed mutagenesis, we constructed a mutant hCaR with K29E mutation using WT hCaR cDNA as a template. We transfected hCaR cDNA into HEK-293 cells and analyzed receptor expression on immunoblots stained with anti-hCaR monoclonal antibody ADD to detect total CaR immunoreactive species. Under reducing conditions, ADD antibody detected two major bands of about 130 and 150 kDa for the WT hCaR (Fig. 4A). Previous studies have shown that the monomeric ∼150-kDa band represents hCaR forms expressed at the cell surface and modified with N-linked, complex carbohydrates; the ∼130-kDa band represents high mannose-modified forms, trapped intracellularly and sensitive to Endo-H digestion. We tested the function of the mutant receptor by using the intact cell [Ca2+]0-stimulated PI hydrolysis assay. Figure 4A shows that HEK-293 cells transfected with vector only had no calcium response. The K29E mutant hCaR not only retained responsiveness to [Ca2+]0, but its Ca2+ sensitivity was increased compared with WT hCaR. The effective concentration for 50% maximal response (EC50) value for the K29E mutant hCaR was 1.45 ± 0.07 (SE) mM (n = 5) compared with a value of 3.16 ± 0.04 mM (n = 5) for the WT hCaR. Compared with the WT hCaR, the K29E mutant shows slightly greater expression seen on immunoblot (Fig. 4A). Our previous study showed that different cell surface expression levels of hCaRs affect the maximal response of the receptor to [Ca2+]0 but not EC50 value of the receptor.(7)

Figure FIG. 4..

(A) Concentration dependence for [Ca2+]0 stimulation of PI hydrolysis (left) and immunoblot of CaR in transiently transfected HEK-293 cells expressing WT hCaR (12 μg DNA transfected), vector only (12 μg DNA transfected), K29E mutant hCaR (12 μg DNA transfected), and WT hCaR + K29E mutant hCaR (6 μg DNA each transfected; right). Transfection, PI assay, SDS-PAGE, and immunoblot with monoclonal anti-hCaR ADD were performed as described in the Materials and Methods section. Molecular mass standards are indicated at the right of the blots. Results of PI assay are expressed as percent of maximal response (WT hCaR at 12 mM). The immunoblots shown here and in Figs. 5-7 were done using cells from the same transfection as the cells used for PI hydrolysis assay. (B) Heterodimerization of K29E mutant CaR with TM1 CaR. (Left) Immunoblot of CaR in transiently transfected HEK-293 cells expressing K29E mutant CaR (12 μg DNA transfected), TM1 CaR (12 μg DNA transfected), and K29E mutant CaR + TM1 CaR (6 μg DNA each transfected). (Right) Cell lysate from the same experiment as shown at left was immunoprecipitated with anti-hCaR polyclonal antibody GGD as described in the Materials and Methods section. Lysates were from cells transfected with K29E alone, with TM1 alone, with both K29E and TM1 or (in the lane labeled “lysate mixture”), with a mixture of equal amounts of cell lysate from cells transfected separately with K29E alone and with TM1 alone as an additional negative control. Immunoblot was developed with anti-hCaR monoclonal antibody ADD. Molecular mass standards are indicated at the left of the blots.

The fact that the two patients we report here are heterozygotes with both WT and K29E mutant hCaR alleles suggests that the K29E mutation has a dominant effect in cells expressing both WT and K29E mutant hCaRs. Our in vitro study confirmed this. HEK-293 cells were co-transfected with equal amounts of WT and K29E mutant hCaR cDNA, and function was determined by [Ca2+]0-stimulated PI hydrolysis assay. Figure 4A shows that [Ca2+]0 sensitivity of cells transfected with equal amounts of K29E mutant and WT hCaR cDNA was intermediate between that of cells transfected with either the K29E mutant or WT hCaR cDNA alone. Both K29E mutant and WT hCaR homodimers, as well as heterodimers of the WT and K29E mutant, are presumed to form in the cells co-transfected with both forms of CaR cDNA. To assess the ability of the K29E mutant CaR to form heterodimers with the WT hCaR, the K29E mutant and TM1, a cDNA encoding a form of the hCaR truncated at lysine 644 in the first intracellular loop, were co-expressed in HEK-293 cells. Immunoprecipitation was performed with GGD, an antibody raised against a carboxy-terminal epitope of the hCaR, which will therefore not react with TM1. Figure 4B shows that TM1 CaR was not immunoprecipitated by GGD when expressed alone, but was immunoprecipitated when co-expressed with K29E mutant, indicating that K29E heterodimerizes with TM1 through its ECD.

Comparison of function of mutant hCaRs with K29 substituted by A, D, E, N, or R

To determine if CaR activation is solely a function of substitution of K29 or if activation depends on substitution of K29 by specific amino acids, we constructed four additional mutant hCaRs with residue K29 replaced individually by A, D, N, or R by site-directed mutagenesis. Figure 5 shows the results of expression and functional assay of these mutants. All were expressed at comparable levels in the same pattern of two major bands as WT, except for K29R, which showed a modest decrease in expression of the 150-kDa form. As a control for equal loading, the immunoblot of CaR was stripped and re-probed with anti-β-tubulin, and the result is shown in Fig. 5B.

Figure FIG. 5..

(A) Concentration dependence for [Ca2+]0 stimulation of PI hydrolysis and (B) immunoblot of CaR (top) and β-tubulin (bottom) in transiently transfected HEK-293 cells expressing WT hCaR, K29A, K29D, K29E, K29N, K29R, and K28E mutant hCaRs. Methods and format for presentation of results are as Fig. 4A except that the samples were separated on 7.5% SDS-PAGE and the maximal response is WT hCaR at 30 mM. Twelve micrograms DNA of each of the CaRs was used in transfection. The nitrocellulose membrane was stripped after blotting with ADD for expression of CaR and re-probed with anti-β-tubulin as an internal control for equal loading.

Mutation of the basic residue K29 to another basic residue R did not notably affect Ca2+ sensitivity of the receptor, whereas mutation of K29 to either neutral (A and N) or acidic (E and D) residues significantly increased the Ca2+ sensitivity of the receptor. The EC50 values for K29A, K29N, and K29D mutant CaRs are 2.22 ± 0.05 (n = 3), 2.23 ± 0.07 (n = 3), and 1.28 ± 0.05 mM (n = 3), respectively. These results suggest that a basic-to-basic residue substitution at residue 29 has minimal functional effect, whereas substitutions that increase net negative charge, particularly lysine substitution by acidic glutamate or aspartate, increase Ca2+ sensitivity of the receptor. We noted that adjacent residue 28 is also a basic residue, K. To see if substitution of K28 by acidic glutamate would mimic the effects of the naturally occurring K29E mutation, we constructed and studied the K28E mutant CaR. The K28E mutant CaR was expressed at a level modestly reduced compared with WT CaR. It showed no increase in sensitivity to [Ca2+]0 and a slight reduction in maximal response compared with WT CaR (Fig. 5).

Effects of K29E mutation on a receptor with a naturally occurring inactivating mutation Q27R

Among the naturally occurring missense mutations of the CaR identified to date, the activating K29E mutation described herein and the Q27R inactivating mutation identified in a subject with familial hypocalciuric hypercalcemia(8) are closest to the N terminus of the CaR and are two residues apart. Given this proximity and their opposing functional effects on the CaR, we elected to study the effects of combining these mutations. Our in vitro study shows that the Q27R mutation causes a drastic right-shift in concentration dependence for [Ca2+]0 activation of the receptor, with a EC50 value larger than 6 mM, and on immunoblot, expression of the upper form (∼150 kDa) of the mutant receptor is sharply reduced (Fig. 6), indicating that both the cell surface expression and calcium sensitivity of the Q27R mutant receptor are severely impaired. We created a new mutant hCaR with K29E/Q27R double mutation, and measured [Ca2+]0-stimulated PI hydrolysis of cells transfected with the double mutant. Figure 6A shows that the double mutant has nearly normal calcium responsiveness, with an EC50 value of 3.87 ± 0.07 mM (n = 3). In addition, compared with the Q27R mutant CaR, the cell surface expression of the K29E/Q27R double mutant was also increased reflected by a darker ∼150 kDa band seen on immunoblot (Fig. 6B).

Figure FIG. 6..

(A) Concentration dependence for [Ca2+]0 stimulation of PI hydrolysis and (B) immunoblot of CaR in transiently transfected HEK-293 cells expressing WT hCaR (12 μg DNA transfected), K29E (12 μg DNA transfected), Q27R (12 μg DNA transfected), K29E/Q27R (12 μg DNA transfected), and K29E + Q27R (6 μg DNA each transfected). Methods and format for presentation of results are as in Fig. 4A except that maximal response is WT hCaR at 30 mM.

We also compared [Ca2+]0 responsiveness of cells co-transfected with equal amounts of K29E and Q27R mutant cDNA with cells transfected with either the K29E mutant or the Q27R mutant alone. Cells transfected with the two mutant CaR cDNAs also showed calcium sensitivity and cell surface expression intermediate between that of either mutant CaR alone.

Effects of K29E mutation on a receptor with a mutation of an acidic residue in the loop 2 region

Of the 17 activating mutations clustering in the amino-terminal portion of the VFT, 10 are located within a region we have termed loop 2 (see Fig. 3). Loop 2 comprises a portion of the dimer interface and is the site of intermolecular disulfide linkage.(9, 10) The key role of loop 2 in receptor activation is demonstrated by a difference in the structure of the proximal portion of loop 2 in the free versus agonist-bound form.(9) Interestingly, K29E mutation occurs in close proximity to loop 2 (Fig. 3B, right, shows expanded view of location of loop 2 and K29). D42, the residue equivalent to K29 in rat mGluR1, seems to form a salt bridge with K154 at the end of loop 2.(9) A similar salt bridge may exist between K29 and an acidic residue, such as D121, D126, E127, or E133 in loop 2 of the hCaR. Such a salt bridge might be necessary to maintain loop 2 in an inactive conformation such that disruption of the salt bridge by the K29E mutation would mimic the activating effect of mutations directly involving residues within loop 2. To test this hypothesis, we mutated acidic residues D121, D126, E127, and E133 to lysine and superimposed these mutations on the K29E mutant receptor. We reasoned that substitution of these acidic residues by basic lysine might re-establish a salt bridge with the mutant E29 and thereby normalize Ca2+ sensitivity. Of these double mutants, only K29E/D121K showed a reduction in Ca2+ sensitivity but was not fully equivalent to WT (Fig. 7)

Figure FIG. 7..

(A) Concentration dependence for [Ca2+]0 stimulation of PI hydrolysis and (B) immunoblot of CaR in transiently transfected HEK-293 cells expressing WT hCaR, K29E, D121K, and K29E/D121K. Methods and format for presentation of results are as in Fig. 4A except that maximal response is WT hCaR at 30 mM.

DISCUSSION

The present cases are notable for the severity and early neonatal onset of hypocalcemia in both members of a pair of monozygotic twins. Inappropriately suppressed serum PTH and relative hypercalciuria, particularly after correction of hypocalcemia with vitamin D and oral calcium treatment, are consistent with the diagnosis of ADH caused by a heterozygous activating mutation of the hCaR.(11, 12) Attempts to reduce hypercalcuria with thiazides were not completely effective. The nephrocalcinosis observed in both sisters emphasizes the importance of not attempting to completely normalize serum calcium in patients with ADH but rather to raise the serum calcium to the minimum level needed to avoid symptoms.

Both parents of the twins were asymptomatic and showed normal serum calcium and normal CaR gene sequence, indicating that the K29E mutation identified in the twins occurred as a de novo germline mutation. A similar occurrence was reported in a Japanese patient presenting with early-onset, severe hypocalcemia and in whom a de novo heterozygous L125P CaR mutation was identified.(13) A recent report documented ADH in two siblings caused by an activating CaR mutation transmitted by the mother who was mosaic for the CaR mutation.(14) We found no evidence, however, for somatic mosaicism in either parent in the family reported herein, and the documentation of the twins' monozygosity makes a de novo germline mutation the most likely explanation.

How does the K29E mutation lead to increased Ca2+ sensitivity of the CaR? The CaR, like other family 3 G-protein-coupled receptors (GPCRs), is characterized by a large amino-terminal extracellular domain (ECD) and the seven transmembrane domain (7TM) characteristic of GPCRs.(15) The ECD itself consists of a VFT domain(9, 16) and a cysteine-rich domain (Fig. 3A). X-ray crystallographic studies of free and agonist-bound forms of the VFT domain of another family 3 GPCR, the rat metabotropic glutamate type 1 receptor (mGluR1), show that the mGluR1 VFT domain is an intermolecular disulfide-linked dimer and that agonist binding to a cleft between the two lobes of the VFT leads to two conformational changes: (1) closure of the VFT and (2) rotation of one protomer by 70° relative to the other about an axis perpendicular to the dimer interface.(9)

Naturally occurring activating mutations of the CaR identified in subjects with ADH cluster in two regions of the receptor: the amino-terminal portion of the VFT domain (17 reported to date including K29E) and the region just proximal to (3 reported), and including, the 7TM domain (10 reported; Fig. 3A.(2) Of the 17 activating mutations clustering in the amino-terminal portion of the VFT, 10 are located within a region we have termed loop 2 (see Fig. 3B). Loop 2 comprises a portion of the dimer interface and is the site of intermolecular disulfide linkage.(9, 10) Loop 2 is not visualized in the 3D structure of the mGluR1 VFT determined by X-ray crystallography, reflecting its likely flexibility and motion, but its key role in receptor activation is demonstrated by a difference in the structure of the proximal portion of loop 2 in the free versus agonist-bound form.(9) Based on these observations, we have proposed a model in which mutations causing ADH lead to increased Ca2+ sensitivity by facilitating the conformational change about the dimer interface occurring on agonist binding.(2, 7)

Unlike the ADH mutations clustered in loop 2 of lobe I and additional ADH mutations such as P221L, E228Q, and Q245R that are located in lobe II (Fig. 3) but are also positioned at the dimer interface, the K29E mutation is not located at the dimer interface. It occurs just after the first β sheet at the amino-terminus of the VFT domain, but interestingly, in close proximity to loop 2 (Fig. 3B, right). We hypothesized that K29E may activate the CaR by disrupting a salt bridge between K29 and an acidic residue in loop 2. Unfortunately, because loop 2 is not visualized in the X-ray crystal structure of the mGluR1 (note that the structure has a gap between R131 and K154), our homology model of the hCaR does not allow us to verify this speculation. Nonetheless, our functional data with the combined K29E/D121K mutant and with substitution of K29 with other residues that alter positive charge are consistent with, but certainly do not prove, such a model. Definitive explanation for the activating effect of the K29E mutation await further molecular and structural studies that will include loop 2 of the CaR.

Finally, we also studied a previously reported inactivating mutation, Q27R, adjacent to K29. This is one of three pairs of closely juxtaposed naturally occurring mutations of the hCaR with opposite effects on function, the others being R220W/Q with P221L and R227L/Q with E228Q.(2) These are, respectively, inactivating mutations identified in FHH and activating mutations identified in ADH, and in the case of P221, the same residue mutated to serine has been identified as an inactivating mutation in FHH. While these pairings may be merely coincidental, it is more likely that they identify these regions of the hCaR as particularly critical for structure and/or function.

Expression of the Q27R mutant showed markedly reduced sensitivity to Ca2+, similar to a previous report.(8) In addition, we documented reduced expression of the mature, fully glycosylated receptor. We suggest that substitution of Q27 at the end of a β sheet with arginine compromises normal folding and processing of the CaR and that even the cell-surfaced expressed mutant receptor is impaired in Ca2+ response. Interestingly, when we co-expressed the separate Q27R and K29E mutant CaR cDNAs, we observed a degree of sensitivity closer to that of the activating K29E mutant than to that of the inactivating Q27R mutant, presumably reflecting the preferential expression of the former. More strikingly, a double Q27R/K29E mutant behaved almost like the WT CaR. This implies that the K29E mutation can overcome both impaired expression and Ca2+ sensitivity caused by the Q27R mutation. These observations have interesting implications for the molecular genetics of the hCaR. Compound heterozygosity for different inactivating mutations of the hCaR, each of which individually leads to an FHH phenotype, has been reported as a cause of neonatal severe hyperparathyroidism.(17) In contrast, an activating mutation such as K29E and an inactivating mutation such as Q27R could theoretically lead to an apparently normal phenotype. Much less likely, but still of theoretical interest, two opposing mutations such as K29E and Q27R could occur in cis on one CaR allele and would be unlikely to be detected clinically if, as we predict based on our functional studies, the phenotype would be normal. As human genome sequencing becomes less expensive and more widely performed, the frequency of such functionally “mutually opposing” sequence variations may eventually be determined.

Acknowledgements

We thank Prof Luigi Rossi-Bernardi (CISI University of Milan) for the HTS platform. IZ was supported by Fondazione Carlo Erba.

Ancillary