| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Departments of Clinical Biochemistry (P.H.N., L.H.) and Endocrinology and Metabolism (S.E.C., L.M.), Aarhus University Hospital, Aarhus Sygehus, DK-8000 Aarhus C, Denmark; and Department of Endocrinology (K.B.), Odense University Hospital, DK-5000 Odense, Denmark
Address all correspondence and requests for reprints to: Peter H. Nissen, Department of Clinical Biochemistry, Aarhus University Hospital, Aarhus Sygehus, Tage Hansens gade 2, DK-8000 Aarhus C, Denmark. E-mail: sci08phn{at}as.aaa.dk.
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Objective: We sought to define the mutation spectrum of the CASR gene in a Danish FHH population and to establish genotype-phenotype relationships regarding the different mutations.
Design and Participants: A total of 213 subjects clinically suspected to have FHH, and 121 subjects enrolled as part of a family-screening program were studied. Genotype-phenotype relationships were established in 66 mutation-positive index patients and family members.
Main Outcome Measures: We determined CASR gene mutations, and correlating levels of plasma calcium (albumin corrected), ionized calcium (pH 7.4), and PTH were measured.
Results: We identified 22 different mutations in 39 FHH families. We evaluated data on circulating calcium and PTH for 11 different mutations, representing a spectrum of clinical phenotypes, ranging from calcium concentrations moderately above the upper reference limit, to calcium levels more than 20% above the upper reference limit. Furthermore, the mean plasma PTH concentration was within the normal range in eight of 11 studied mutations, but mild to moderately elevated in families with the mutations p.C582Y, p.C582F, and p.G553R.
Conclusions: The present data add 19 novel mutations to the catalog of inactivating CASR mutations and illustrate a variety of biochemical phenotypes in patients with the molecular genetic diagnosis FHH.
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Heterozygous inactivating mutations in the gene encoding the calcium sensing receptor (CASR) are responsible for FHH, whereas homozygous inactivating mutations in the CASR gene result in neonatal severe hyperparathyroidism (NSHPT) (2). Furthermore, activating mutations in the same gene cause autosomal dominant hypocalcaemia (3). The CASR is a member of the G protein-coupled receptor family and regulates blood calcium levels within a narrow range. It is located on the cell membrane of many different cells but regulates plasma calcium homeostasis mainly through its expression in the parathyroid glands and kidney tubules (4). It consists of a large amino terminal extracellular domain, a membrane spanning part containing seven transmembrane regions and the carboxyl-terminal intracellular domain (5).
The CASR is present on the cell surface as dimers, mediated by intermolecular disulfide bonds (6). It has been demonstrated in a cell culture system that mutant and wild-type CASRs form heterodimers (7). This could explain a proposed dominant negative effect in FHH, in which the mutant receptor interferes with the function of the normal receptor (8). The ligand-binding site is localized to the extracellular domain, whereas the transmembrane domain may contribute to the receptor activation (9). Furthermore, it has been shown that two intracellular loops are involved in the receptor’s signaling by activation of phospholipase C (10). Evidence from a naturally occurring mutation containing an ALU repeat insertion, resulting in a truncated receptor lacking a large part of the intracellular domain, shows that this type of defect reduces the number of normal receptors on the cell surface (11).
At present, at least 64 inactivating mutations, of which 17 have been observed in several families or studies, have been published (12). The majority of these are missense mutations, but a few nonsense mutations, a splice site mutation (13), and a large rearrangement caused by an ALU element insertion (14) have been described.
Here, we further add 19 novel mutations to the catalog of inactivating changes in the CASR gene causing FHH and describe the variation in the biochemical phenotype seen as a result of the mutations.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We studied genetic variations in the CASR gene in a total of 334 patients (Fig. 1
). A total of 213 patients were referred by one of three physicians (S.E.C., L.M., or K.B.) on the basis of a clinical suspicion of FHH, and 121 were referred as part of a family-screening program.
|
A total of 100 mutation-positive patients (39 index cases and 61 family members) were identified (Fig. 1
). In 66 mutation-positive patients (21 index patients and 45 family members), sufficient biochemistry was available to evaluate the biochemical characteristics of the different mutations identified. The age range of these subjects was 18–83 yr.
To do a preliminary evaluation of the clinical criteria for requiring a molecular genetic analysis of the CASR gene, the referring physician was asked to pass on brief clinical information. This information was available in 98 cases and included the presence of family history, previous parathyroid surgery, and PTH values.
All patients gave their informed consent to the molecular genetic analysis. The study was approved by the regional ethical committee of Aarhus County (no. 20030195).
DNA amplification, and sequence and microsatellite analysis
We purified genomic DNA from whole blood using the Puregene genomic DNA purification kit (Gentra Systems, Plymouth, MN) and performed PCR using 14 sets of oligonucleotide primers (primer sequences are available on request from the authors). The primers were designed to amplify exons 2–7, including a minimum of 10 nucleotides flanking the intron-exon border. Sense and antisense primers were modified with M13 sequences (M13 forward: GTAAAACGACGGCCAG and M13 reverse: CAGGAAACAGCTATGAC) to use one primer set for the sequencing reaction of all amplicons. We assessed the quality of the PCR products by electrophoresis in 2% agarose gels, using standard procedures. We purified PCR products with Microspin S-400 columns (GE Healthcare, Hilleroed, Denmark), sequenced the amplicons in both directions using BigDye terminator version 1.1 (Applied Biosystems, Foster City, CA), and separated the ethanol-precipitated fragments on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems) or an Applied Biosystems 3130 Genetic Analyzer. We aligned sequence traces to the CASR reference sequence (NM_000388.2) using Sequence Navigator (version 1.01; Applied Biosystems) or SeqScape (version 2.5; Applied Biosystems). We verified all mutations by bidirectional sequencing, using a second independent blood sample. To reconstruct haplotypes in families sharing the same mutation, three microsatellite markers were identified using the Human Genome Browser (http://genome.ucsc.edu/). One marker was located in the CASR gene, one was located 167 kilobases upstream in the neighboring CD86 gene, and one was located 224 kilobases downstream in the KPNA1 gene. Primers were fluorescently labeled, and PCR products were separated on an ABI PRISM 310 Genetic Analyzer (Applied Biosystems) and analyzed using GeneScan software (version 3.1.2; Applied Biosystems).
Controls
We obtained control samples from the Nordic Reference Interval Project Bio-bank and Database (15). Genomic DNA from 94 samples, representing normal Danish individuals, was isolated using the Capture Plate Kit (Gentra Systems). We performed PCR as previously described and purified PCR products with the GFX 96 PCR Purification kit (GE Healthcare). Unidirectional sequencing was performed using BigDye terminator version 1.1 and separated on an Applied Biosystems 3130 Genetic Analyzer. Sequence traces were aligned to the CASR reference sequence (NM_000388.2) using SeqScape (version 2.5).
Biochemistry
Measurements of plasma calcium (albumin corrected), ionized calcium, and PTH were available for a subset of 66 patients and representing 11 different mutations. Albumin corrected calcium and ionized calcium (pH 7.4) were measured on standard automated analyzers for clinical biochemistry. Intact PTH was analyzed using the Elecsys Intact PTH assay on an Elecsys 2010 analyzer (Roche, Basel, Switzerland). Biochemical values for each individual were based on mean values of three measurements (two measurements for ionized calcium) in the majority of the cases.
Bioinformatics analysis
Mutations resulting in an amino acid substitution were evaluated using a multiple sequence alignment of the CASR amino acid sequences from Homo sapiens (human), Canis familiaris (dog), Rattus norvegicus (rat), Mus musculus (mouse), Bos taurus (cow), and Salmo salar (salmon). The alignment was made using the ClustalW service at http://www.ebi.ac.uk/Tools/clustalw/. Further evaluation was made with PolyPhen (16) (http://genetics.bwh.harvard.edu/pph/). This application performs an analysis of multiple sequence alignments and protein three-dimensional structures. When a mutation did not result in an amino acid change, the mutation was analyzed for potential introduction of cryptic splice sites using the splice site prediction tool (17) (http://www.fruitfly.org/seq_tools/splice.html), or for disruption or creation of potential exonic splicing enhancers (ESEs) using the ESE finder (18) (http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi?process=home). Information on single nucleotide polymorphisms (SNPs) was extracted from the Ensembl database (http://www.ensembl.org/index.html) and the National Center for Biotechnology Information SNP database (dbSNP) (http://www.ncbi.nlm.nih.gov/sites/entrez?db=snp). The mutation nomenclature used follows the current guidelines (19, 20). The numbering of mutations is based on the cDNA sequence, where +1 is the A of the ATG start codon.
Statistical analysis
The association between PTH and calcium (ionized and albumin corrected), and PTH and age, was analyzed using linear regression. The dependency of the calcium and PTH values on sex and the comparison of groups were analyzed using t tests. PTH values were log transformed. Statistical analyses were performed with SPSS version 13.0 software (SPSS, Inc., Chicago, IL).
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In the studied population, we identified 30 different sequence variants. Of the 22 variants presumed to be functional mutations (Table 1), 19 were to our knowledge novel, whereas the p.C562Y, p.C582Y, and p.G670R mutations have been published previously (21, 22). Co-segregation with hypercalcemia was verified for 13 of 22 mutations, whereas family members were unavailable for analysis in the remaining nine families. We did not observe any of the presumed 22 mutations in 94 control individuals, representing 188 normal alleles. The three sequence variants c.114T>C, c.2610G>A, and c.2838G>A were synonymous substitutions, i.e. did not cause an amino acid change, and were present at a low frequency in the study sample and control group (Table 2).
|
|
A mutation was identified in 39 (18.3%) of the 213 patients referred on a clinical suspicion of FHH. There were 17 mutations observed in one index patient only, whereas five mutations were observed in two to eight index patients (Table 1). We analyzed microsatellite data in a number of families carrying the three recurrent mutations p.G397R, p.V728F, and p.R886W. For each mutation, samples from CASR heterozygous index patients and family members, and samples from CASR normal family members, were used to reconstruct haplotypes. These data reveal that all patients carrying the p.G397R mutation share the same alleles in all three microsatellite markers and the frequent variant p.R990G, indicating that the mutation is present on a shared haplotype block. All analyzed patients with the p.V728F mutation also shared the same microsatellite alleles, whereas analysis of allele sharing in patients carrying the p.R886W mutation revealed three different haplotype blocks associated with the mutation.
Biochemical characteristics of CASR mutations
Statistical evaluation of the 66 patients revealed no significant (P > 0.05) association between PTH and age, and no significant association (P > 0.05) between sex and each biochemical parameter. We found significant association between PTH and calcium (albumin corrected) (P < 0.001) (Fig. 2), and PTH and calcium (ionized) (P < 0.01).
|
data on plasma calcium and PTH are summarized for each of these mutations. The p.C582Y and p.C582F resulted in a similar phenotype with albumin corrected plasma calcium values above 3 mmol/liter. Similar phenotypes were observed with the p.P221Q, p.S171N, p.V728F, and p.G553R mutations. More moderate phenotypes were observed in patients with the p.G397R, p.G509R, p.R886W, p.W352X, and fs.L666:X707 mutations. Importantly, the two latter mutations that are predicted to result in premature stop codons and, consequently, a clearly truncated CASR only had a moderate biochemical phenotype compared with, for example, the p.C582F mutation.
|
In the 213 patients referred on clinical suspicion, we identified five variants in the CASR gene (p.A986S, p.R990G, p.Q1011E, c.1732 + 16T>C, and c.3237 + 60T>A) that occurred frequently. The same variants were identified in a control population. All five frequent variants are present in the public dbSNP. The frequencies of these variants are given in Table 2.
Clinical information
Data on clinical information are summarized in Table 3. It appears that a family history of hypercalcemia is the most important predictor of a CASR mutation because 42% (eight of 19) of patients with a positive family history had a CASR mutation. Of the 13 patients that underwent unsuccessful parathyroid surgery, 23% (three of 13) were found to have a CASR mutation. A total of 66 patients had a PTH value above the reference limit. Of these, 14% (nine of 66) were mutation positive.
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We identified three rare synonymous sequence variants. To evaluate whether these variants might introduce cryptic splice sites, or interfere with ESEs, further bioinformatics analyses were undertaken. This revealed that the variant c.2838G>A in theory interrupts a sequence motif that could serve as a potential ESE. Further analysis of the patient’s family will indicate if this variant has any influence on the phenotype in this patient.
The analysis of biochemical data of 11 different mutations and 66 patients revealed distinct differences in the biochemical manifestation of the different mutations. Data on calcium levels in plasma indicate that mutations resulting in truncated receptors have a modest effect on plasma calcium values, whereas missense mutations such as the previously described p.C582Y and the novel p.P221Q and p.G553R result in a more severe elevation in circulating calcium levels (Fig. 3). In a recent study, Ward et al. (23) presented similar data from four FHH families, showing that truncating mutations resulted in more moderate calcium levels than mutations resulting in amino acid substitutions. It was previously suggested that this phenomenon could be due to the fact that missense mutations could result in a dominant negative effect because the abnormal receptors dimerize with normal receptors (8). This results in abnormal functional domains within a heterodimeric CASR affecting the function of normal domains, probably via intermolecular interactions, as shown by Bai et al. (7).
In our series, the PTH levels depended on the mutation, as observed with the codon 582 mutations and the p.G553R mutation, which had mean plasma PTH levels above the upper reference limit (6.9 pmol/liter) (Fig. 3).
We identified a p.P221Q mutation in four individuals from the same family with high levels of plasma calcium. Although this mutation is novel, others have described two different mutations implicating codon 221. A proline to serine substitution (p.P221S) resulting in a family presenting with FHH was described by Pearce et al. (22), whereas, in contrast, a previous report described a p.P221L mutation in a family with autosomal dominant hypocalcaemia (24). The fact that mutations involving the same codon can result in inactivation as well as activation of the CASR is puzzling. One explanation could be that amino acid 221 is located in the region implicated in ligand binding, as suggested by Silve et al. (25). The proline residue should then maintain normal ligand binding, whereas serine and glutamine amino acids weaken the binding of calcium resulting in inactivation, and leucine in this position results in a stronger ligand binding effect.
Three of the identified mutations were observed in five to eight families, raising the question if each of these mutations occurred independently several times in different families or occurred in a common ancestor. Analysis of microsatellite markers in and around the CASR showed that the patients carrying the p.G397R and p.V728F mutations shared haplotypes specific for each mutation, indicating that these mutations most likely occurred in a common ancestor. In contrast to this, microsatellite data show that the p.R886W mutation co-segregates with three different haplotypes, suggesting that the mutational event happened several times independently.
The coding variants p.A986S, p.R990G, and p.Q1011E, observed frequently in this study, have been studied intensively by others. The p.A986S variant, particularly, has previously been observed to play a modest but significant role in modulating plasma calcium concentrations (26, 27, 28). Recently, these three variants were subjected to functional assays, which indicate that the variants do not alter the set point of the receptor compared with wild-type receptors (29). In this study we found five frequent variants with allele frequencies similar in the group of 213 suspected FHH individuals and in 94 normal controls. There was no difference in plasma calcium values for the three p.A986S genotypes (AA, AS, and SS) in these 94 controls (data not shown).
Clinical and biochemical parameters do not reliably distinguish between PHPT and FHH (30). In contrast, DNA testing may secure the diagnosis and, hence, the correct treatment. We used a molecular diagnostic strategy to confirm whether patients with a clinical diagnosis of FHH, or with a borderline condition indicating either FHH or PHPT, had a genetic defect in the CASR gene. In the 213 patients, we found that approximately 18% had FHH, i.e. the clinical diagnosis could be confirmed using the molecular genetic test. Does this mean that the remaining 82% do not have FHH? We suspect that a large proportion of the patients in which we did not find a CASR mutation suffer from PHPT. However, due to the inherent limitations of the method and the fact that locus heterogeneity may exist (31), there is still a possibility that some patients harbor mutations outside the exons and immediate exon flanking regions, which might result in a defective gene, or large deletions or insertions spanning one or several exons. However, only a single report describing a large complex rearrangement of the CASR gene in FHH families has been published so far (14). Based on the clinical information summarized in Table 3, we show that the most important predictor of a CASR mutation is a family history of hypercalcemia. However, we also observed that 23% of patients that underwent unsuccessful parathyroid surgery with persistent hypercalcemia were found to have a CASR mutation. This indicates that unsuccessful parathyroid surgery is an important predictor of molecularly verified FHH. Previously, it was found that 20% of FHH patients have elevated PTH levels (32). We find that 14% of the patients with plasma PTH above the reference limit were positive for a mutation in the CASR gene. Together, these preliminary data support that selection of patients for a molecular genetic analysis should be based on hypercalcemia, a positive family history, and unsuccessful parathyroid surgery, keeping in mind that patients with PTH above the reference limit could still represent FHH patients. The discriminative power of the calcium to creatinine clearance rate in this patient group is under way.
In conclusion, we identified 22 different mutations in the CASR gene, representing a spectrum of mutations in Danish FHH subjects. Moreover, we observed marked differences in plasma calcium and PTH levels depending on the mutation. Based on the present data, we find that a molecular genetic analysis of the CASR offers a valuable supplement to the clinical diagnosis of FHH in hypercalcemic patients.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Disclosure Information: The authors have nothing to declare.
First Published Online August 14, 2007
Abbreviations: CASR, Calcium sensing receptor; ESE, exonic splicing enhancer; FHH, familial hypocalciuric hypercalcemia; NSHPT, neonatal severe hyperparathyroidism; PHPT, primary hyperparathyroidism; SNP, single nucleotide polymorphism.
Received February 12, 2007.
Accepted August 7, 2007.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
