A case of familial isolated hypoparathyroidism type 2 with novel compound heterozygous variants in GCM2

Highlights

● Novel compound heterozygous GCM2 variants were identified in a boy with FIH type 2.

● Presymptomatic analysis showed hypoparathyroidism before seizure onset.

● p.Arg143Trp and p.Val382AsnfsTer8 GCM2 variants lead to GCM2 loss-of-function.

Introduction

Familial isolated hypoparathyroidism (FIH) is a form of primary hypoparathyroidism characterized by hypocalcemia and hyperphosphatemia resulting from insufficient parathyroid hormone (PTH) secretion (1). By definition, FIH occurs in the absence of abnormalities in other organs and is not associated with a recognized malformation syndrome. Based on the causative gene, FIH can be classified into FIH type 1 (PTH-related; OMIM #146200) and FIH type 2 (GCM2-related; OMIM #618883). Most reported cases of FIH type 2 follow an autosomal recessive inheritance pattern; however, autosomal dominant families have also been described, in which the GCM2 variants exert a dominant-negative effect (2). Clinically, the age at presentation and symptom severity are highly variable among individuals with FIH type 2.

GCM2 is located on chromosome 6p24.2 and comprises five exons encoding a 506–amino acid transcription factor. The protein contains several functional regions, including a DNA-binding domain (DBD), nuclear localization signal, transactivation domain 1 (TAD1), C-terminal conserved inhibitory domain (CCID), and TAD2 (Fig. 1A). Gcm2-deficient mice lack parathyroid glands and exhibit hypoparathyroidism, indicating Gcm2 as a master regulatory gene in parathyroid gland development (3). Biallelic loss-of-function GCM2 variants can result in failure of parathyroid gland development, leading to primary hypoparathyroidism (4). Previously reported FIH type 2-associated GCM2 variants show characteristic patterns, including frequent missense variants within the DBD and truncating variants lacking TAD2. Here, we report a Japanese boy with clinical features consistent with FIH type 2, in whom we identified two novel compound heterozygous GCM2 variants.

Fig. 1.

(A) Schematic representation of the GCM2 protein domains. GCM2 comprises a DNA-binding domain (DBD), nuclear localization signal (NLS), transactivation domain(s) (TAD), and a C-terminal conserved inhibitory domain (CCID). Previously reported FIH type2-associated missense variants are shown in black, and the missense variant identified in this case is highlighted in red. (B) Longitudinal changes in serum calcium, phosphate, and intact PTH levels from the presymptomatic period to the episode of afebrile seizures. Serial measurements from 2 to 14 mo of age were obtained by analyzing frozen residual specimens collected during a multivalent vaccine clinical trial (①–⑤). Ca, calcium; P, phosphate; PTH, parathyroid hormone. (C) Pedigree of the family. Both parents are clinically unaffected; the father is heterozygous for p.(Val382AsnfsTer8) and the mother is heterozygous for p.(Arg143Trp). The proband carries both variants in a compound heterozygous state. (D) Representative Sanger sequencing chromatograms of the relevant GCM2 regions in the proband and parents. The boxed region in the left panel indicates c.1139_1142dup, p.(Val382AsnfsTer8), and the arrow in the right panel indicates c.427C>T, p.(Arg143Trp).

Case Report

The patient was a Japanese boy, the first child of healthy nonconsanguineous parents, born at 41 wk of gestation via an uncomplicated vaginal delivery without fetal distress. At birth, his weight was 3110 g (−0.3 SD), length 48.1 cm (−1.0 SD), and head circumference 35.0 cm (+1.1 SD). At 14 mo of age, the boy experienced his first afebrile generalized tonic–clonic seizure, which lasted approximately 10 min. Laboratory testing performed immediately after the event demonstrated hypocalcemia and hyperphosphatemia with an inappropriately low intact PTH level: calcium 5.6 mg/dL (reference 9.4–10.8 mg/dL), phosphate 10.1 mg/dL (age-specific reference range: 4.5–6.4 mg/dL), intact PTH 5 pg/mL (reference range: 10–65 pg/mL), 25-hydroxyvitamin D 25.0 ng/mL (reference: > 20 ng/mL), and 1,25-dihydroxyvitamin D 126 pg/mL (reference range: 20–70 pg/mL). Based on these findings, the patient was diagnosed with primary hypoparathyroidism.

Notably, serial blood samples had been collected prior to the episode of afebrile seizures, as the patient had been enrolled in a clinical trial of a multivalent vaccine (KM Biologics Co., Ltd.) starting at 2 mo of age. Analysis of frozen residual specimens demonstrated persistent hyperphosphatemia and suppressed intact PTH levels from 2 to 14 mo of age, with hypocalcemia observed from 8 to 14 mo of age (Fig. 1B). Cranial CT revealed calcification in the right basal ganglia. After correcting hypocalcemia with intravenous calcium, maintenance therapy with oral alfacalcidol and calcium lactate powder was initiated. Chest radiography revealed a thymic shadow. No history of recurrent infections suggestive of immunodeficiency existed. Cardiac and renal ultrasonography revealed no structural abnormalities. Neonatal automated auditory brainstem response screening was normal bilaterally. Language and overall development were age-appropriate through 2 yr of age, with no evidence of hearing impairment. During follow-up, hypocalcemia recurred at 18 mo of age (lowest recorded calcium, 6.1 mg/dL); however, no further seizures occurred. Both parents denied symptoms suggestive of hypocalcemia (e.g., distal paresthesia, afebrile seizures, or tetany). At the time of genetic testing, parental serum calcium, inorganic phosphate, and intact PTH levels were within the reference ranges (data not shown).

Genetic Analysis

This study was approved by the Ethics Committee of the Keio University School of Medicine (No. 20150104). Written informed consent for genetic testing was obtained from the parents of the patient. Peripheral blood samples were collected from the proband and both parents, and genomic DNA was extracted. Genetic analyses were conducted at a commercial diagnostic laboratory. Next-generation sequencing was performed to analyze the protein-coding exons and flanking splice sites of genes associated with hereditary hypoparathyroidism (ACADM, AIRE, CASR, CHD7, CLDN16, CLDN19, DHCR7, FAM111A, GATA3, GCM2, GNA11, HADHA, HADHB, NEBL, NLRP5, PTH, SEMA3E, SOX3, TBCE, TBX1, TBX2, and TRPM6). Candidate variants in the proband and both parents were confirmed by Sanger sequencing.

This analysis identified compound heterozygous variants in GCM2 (NM_004752.4): c.427C>T, p.(Arg143Trp), transmitted from the mother, and c.1139_1142dup, p.(Val382AsnfsTer8), transmitted from the father (Figs. 1C, 1D). Neither variant was registered in ClinVar or the Human Gene Mutation Database. In the gnomAD v4.1 database (accessed on December 27, 2025), the p.(Arg143Trp) variant was observed in 2 of 74,908 African and 1 of 91,066 South Asian individuals, whereas p.(Val382AsnfsTer8) was observed in 3 of 44,900 East Asian individuals. In silico prediction tools supported the deleterious effect of p.(Arg143Trp) (PROVEAN: “deleterious”, score −7.73; SIFT: “damaging”, score 0.0; PolyPhen-2: “probably damaging”, score 1.00; CADD: “indeterminate”, score 23.7). Predicted structural data for GCM2 (AF_AFD3ZXW4F1) were obtained from the AlphaFold Protein Structure Database, and a three-dimensional model was generated using PyMOL version 0.99. In this model, Arg¹⁴³ formed hydrogen bonds with the main chains of Ile²³, Asp²⁵, and Pro²⁶; substitution of Arg¹⁴³ with tryptophan was predicted to abolish these hydrogen bonds (Fig. 2A). According to the American College of Medical Genetics and Genomics criteria, p.(Val382AsnfsTer8) was classified as pathogenic (PVS1 + PM1 + PM2 + PP4) and p.(Arg143Trp) as likely pathogenic (PM1 + PM2 + PP3 + PP4).

Fig. 2.

(A) Three-dimensional structural model of GCM2. Oxygen and hydrogen atoms are shown in red and blue, respectively. In the wild-type protein, Arg143 (blue arrow) forms hydrogen bonds with the main-chain atoms of Ile23, Asp25, and Pro26. Substitution of Arg143 with tryptophan (green arrow) removes the positively charged, polar side chain, and is predicted to abolish these hydrogen-bond interactions. (B) Cross-species sequence conservation analysis. Ile23, Asp25, Pro26, and Arg143 are highly conserved among multiple species, supporting the functional and structural importance of this region.

Discussion

We identified novel compound heterozygous GCM2 variants, p.(Val382AsnfsTer8) and p.(Arg143Trp), in a child with a clinical presentation consistent with FIH type 2. A unique aspect of this case was the availability of presymptomatic biochemical data, which demonstrated persistent hyperphosphatemia and inappropriately low PTH levels from early infancy, followed by the development of hypocalcemia after weaning, when dietary calcium intake likely decreased. Although functional assays were not performed, several lines of evidence support a loss-of-function effect of these GCM2 variants. Notably, p.(Val382AsnfsTer8) is a frameshift variant that introduces a premature stop codon within the C-terminal CCID domain, resulting in truncation of the C-terminus. As this variant is located in the final exon (exon 5), it is predicted to escape nonsense-mediated mRNA decay. A previous in vitro study showed that TAD2, located downstream of CCID, is required for GCM2 transcriptional activity (5). Therefore, this truncation is expected to substantially impair transcriptional function through loss-of-TAD2. Importantly, exon 5 truncating variants in GCM2 with experimentally confirmed dominant-negative effects have been reported exclusively within the TAD2 region (5). In contrast, our frameshift variant, p.(Val382AsnfsTer8), although located in exon 5, lies N-terminal to TAD2 and is predicted to completely abolish TAD2. Consistent with this interpretation, Mitsui et al. reported that p.(Arg367ThrfsTer15), another exon 5 frameshift variant located outside TAD2, did not exhibit a dominant-negative effect in functional assays (2). Collectively, these observations suggest that the variant position relative to TAD2—rather than localization to exon 5 per se—may influence the likelihood of a dominant-negative effect. Additionally, p.(Arg143Trp) is a missense variant within the DBD. Previously reported FIH type 2-associated missense variants are predominantly clustered in this domain (Fig. 1A). Structural modeling indicated that Arg¹⁴³ contributes to local stability through hydrogen bonding with Ile²³, Asp²⁵, and Pro²⁶; substitution with tryptophan is predicted to disrupt these interactions (Fig. 2A). Moreover, these residues are highly conserved across species (Fig. 2B), supporting the functional importance of these intramolecular interactions. Finally, both parents, each heterozygous for one of the variants, were clinically unaffected and had normal calcium–phosphate–PTH profiles, arguing against a dominant-negative effect of either variant alone.

In conclusion, we report a clinically isolated case of primary hypoparathyroidism caused by novel compound heterozygous GCM2 variants. Although functional experiments were not performed, the variant types, affected domains, structural considerations, and segregation data collectively support a loss-of-function mechanism consistent with autosomal recessive FIH type 2.

Conflict of interests

The authors declare no conflicts of interest.