A Japanese school urine screening program led to the diagnosis of KCNJ11-MODY: A case report

Abstract.

Although KCNJ11 mutation is the main cause of neonatal diabetes mellitus, reports of maturity-onset diabetes in the young (MODY) related to KCNJ11 are rare. Here, we report a case of KCNJ11-MODY in a 12-yr-old Japanese female. Hyperglycemia was initially detected during a school urine screening program. Subsequent laboratory examinations revealed impaired insulin secretion; however, no islet autoantibodies were detected. Genetic testing of KCNJ11 revealed a novel heterozygous variant, c.153G>C, p.Glu51Asp. The patient’s father had the same mutation and was diagnosed with diabetes at 46 yr of age. KCNJ11-MODY was suspected, and sulfonylurea administration resulted in adequate glycemic control in the patient. The American College of Medical Genetics and Genomics guidelines classify this variant as likely pathogenic, and the effectiveness of sulfonylureas supports its pathogenicity. The patient could be treated with 0.02–0.03 mg/kg/d of glibenclamide, as this mutation may be responsive to only a small amount of sulfonylurea. A detailed family history and sequencing of causative genes, including KCNJ11, may help diagnose diabetes in school-aged patients.

Highlights

● A novel KCNJ11 pathogenic variant was detected in a school-aged female.

● A school urine screening program contributed to early detection and diagnosis of MODY.

● Treatment with low-dose sulfonylurea was effective in this KCNJ11-MODY case.

Introduction

Maturity-onset diabetes of the young (MODY) is characterized by the onset of hyperglycemia and impaired insulin secretion without insulin intolerance or obesity in younger patients (1). KCNJ11-MODY is caused by pathogenic variants of the KCNJ11 gene and can often be successfully treated with sulfonylureas (SU) (2). Neonatal diabetes mellitus (NDM) caused by KCNJ11 variants has been extensively reported; however, reports of KCNJ11-MODY are rare (3, 4) and KCNJ11-MODY has not been previously reported in school-aged Japanese children. Herein, we report a case of KCNJ11-MODY in a 12-yr-old, school-aged Japanese female in whom glucosuria was initially detected during a school urine screening program. Subsequently, the patient was switched from insulin to SU therapy.

Case Report

A 12-yr-old Japanese female presented to our hospital after the detection of glucosuria during a school urine screening program. The patient had not experience thirst, polydipsia, or weight loss. Her height, weight, and body mass index were 153.1 cm (+ 0.5 SD), 50.15 kg (+ 0.9 SD), and 21.4 kg/m² (+ 0.97 SD), respectively. Physical examination results were unremarkable and A. nigricans was not present. Laboratory tests revealed a hemoglobin A1c (HbA1c) level of 7.8%, and screenings for anti-glutamic acid decarboxylase (GAD) antibody, anti-insulin autoantibody, and anti-insulinoma-associated antigen-2 (IA-2) antibody were all negative. Other biochemical markers were within normal limits, with no accompanying hyperlipidemia or fatty liver observed on ultrasonography. The oral glucose tolerance test led to a diagnosis of diabetes, where the homeostasis model assessment for insulin resistance was 4.1, the homeostasis model assessment for β-cell function was 45.9%, and the insulinogenic index was 0.057, indicating a decreased insulin secretion capacity (Table 1). Her paternal grandfather had been diagnosed with diabetes in his 30s or 40s. Her 43-yr-old father did not have diagnosed diabetes; however, his HbA1c level was slightly elevated (6.0%) during a routine health checkup. Although MODY was suspected, HNF4A, GCK-, HNF1A-, and HFN1B-MODY were ruled out, because no mutations or deletions were found in HNF4A, GCK, HNF1A, or HFN1B per results of PCR-direct sequencing and multiplex ligation-dependent probe amplification. Slowly progressive type 1 diabetes was suspected and treatment with self-injected insulin was initiated. However, anti-GAD and anti-IA-2 antibodies remained negative six months and one year after the initial visit, respectively. Eight months later, an oral glucose tolerance test revealed a fasting C-peptide level of 1.5 ng/dL, suggesting that insulin secretion was maintained. The patient’s HbA1c level remained stable, at approximately 7%, with a basal-bolus insulin regimen and carbohydrate count (5 units of insulin degludec and a carbohydrate insulin ratio of 33 g/U).

Table 1. The oral glucose tolerance test at onset revealed obviously decreased insulin secretion capacity.

The patient’s clinical course indicated the presence of other types of MODY, prompting us to perform additional genetic sequencing of KCNJ11 and INS. Whole-exon Sanger sequencing revealed a novel heterozygous variant, c.153G>C (p.Glu51Asp), in KCNJ11. Therefore, KCNJ11-MODY was suspected and the initial insulin treatment was stopped, in favor of SU for treatment. The glucagon tolerance test after SU initiation revealed sufficient insulin secretion (Fig. 1), and her insulinogenic index, calculated using the oral glucose tolerance test, improved from 0.057 at onset to 0.47 after SU initiation (Table 2). Glibenclamide was initially started at a dosage of 0.02 mg/kg/d, and later increased to 0.03 mg/kg/d. Adequate glycemic control was maintained with HbA1c levels primarily below 7% with SU treatment for one year. Genetic analysis of the patient’s father revealed a similar variant. Subsequently, his HbA1c level increased to 6.5%, and he was diagnosed with diabetes at 46 yr of age.

Fig. 1.

Glucagon tests were performed consecutively before administration, and at 30, 60, and 240 min after administering 0.02 mg/kg of glibenclamide. Insulin secretion increased over time, and an adequate response was observed at 240 min after administration. Some data before administration is missing due to a technical error.

Table 2. The oral glucose tolerance test after administrating glibenclamide showed improved insulin secretion.

Informed consent was obtained from the patient and her parents for the publication of this case report.

Discussion

In pancreatic beta cells, insulin secretion is initiated when ATP-sensitive potassium (K_ATP) channels are closed and inhibited when they are open. This channel is an octameric complex of four pore-forming, inwardly rectifying potassium channel subunits (Kir6.2) encoded by KCNJ11 and four regulatory sulfonylurea-receptor subunits (SUR1). Pathogenic variants of KCNJ11 cause decreased insulin secretion from pancreatic beta cells by conferring reduced ATP sensitivity, resulting in a gain of channel function (5). The cell membrane remains hyperpolarized, leading to varying degrees of glucose metabolic abnormalities, including NDM and KCNJ11-MODY (6).

Here, we report a novel KCNJ11 variant, c.153G>C (p.Glu51Asp), in a 12-yr-old patient with diabetes. The variant was located in the well-established functional domain of KCNJ11, but not found in the control database (gnomAD); the variant co-segregates with MODY in multiple affected family members. Another missense variant of the fifty-first amino acid residue of KCNJ11 (p.Glu51Gly) has been reported as a pathogenic variant that can lead to NDM (7). According to the American College of Medical Genetics and Genomics (ACMG), this variant is “likely to be pathogenic” (PM1+PM2+PM5+PP1) (8). Additionally, in silico analysis evaluated this variant as “probably damaging” (score of 0.993 by Polymorphism Phenotyping version 2) (9). The 51st amino acid residue in KCNJ11 (Glu51) is well-conserved across various species and consists of the slide helix of Kir 6.2, which plays an important role in translating conformational changes induced by ATP binding (10). Variants in the slide helix (p.Arg50Gly, p.Arg50Pro, p.Arg50Gln, p.Gly53Arg, and p.Tur62Arg) were considered pathogenic or likely pathogenic, suggesting the importance of this domain. Moreover, the remarkable response to SU therapy in this case suggests that KCNJ11 c.153G>C (p.Glu51Asp) is a pathogenic variant.

In this report, KCNJ11-MODY was diagnosed in a 12-yr-old patient with diabetes, and her father, with the same KCNJ11 variant, was later diagnosed with diabetes at the age of 46 yr. Her paternal grandfather, who developed diabetes in his 30s or 40s, potentially had KCNJ11-MODY. KCNJ11 mutations are associated with various phenotypes on the diabetes spectrum. Among patients with KCNJ11-MODY, the age onset of diabetes varies widely from the teens to the 50s, even when affected by the same pathogenic variants (2, 4). This phenotypic spectrum of diabetes has also been reported among NDM patients with KCNJ11 pathogenic variants, and non-genetic factors have been suggested to explain these differences (2). Although family history is an important predictor of KCNJ11-MODY, family members or relatives may not develop diabetes symptoms, even if they possess KCNJ11 pathogenic variants. Regardless, when KCNJ11-MODY is suspected, any degree of glucose intolerance in relatives should be considered.

Many cases of diabetes caused by KCNJ11 mutations have been effectively treated with SU (11, 12). SU binds to the SUR1 subunit of the mutant K_ATP channel, causing it to close and leading to membrane depolarization, which also allows insulin secretagogues, such as incretins, to augment insulin secretion (13). In patients with NDM resulting from KATP channel mutations, the required dosage also depends on the type of mutation (14, 15), and the transition to SU from insulin at an earlier age is associated with a higher success rate (15) and a lower required dosage of SU (16). In this case, the required SU dose was 0.02–0.03 mg/kg/d, which is close to the smallest SU dosage reported to date (13). This suggests that KCNJ11 p.Glu51Asp could be controlled with low-dose SU or SU initiation in early life, allowing for a lower SU dosage for treatment.

The known side effects of SU include hypoglycemia and digestive disorders (14). However, a previous study reported that no patients discontinued SU treatment because of its side effects (11). Treatment with SU is considered safe and can improve the quality of life of patients with KCNJ11-MODY. Therefore, early identification and diagnosis of MODY is important and can be achieved through school urine screening programs.

Conclusion

We report a case of KCNJ11-MODY that was initially detected during a screening for glucosuria during a school urine screening program. The patient was successfully treated with SU. In school-aged patients with diabetes, obtaining a detailed family history and sequencing of the causative gene, including KCNJ11, may aid in specific diagnosis and lead to a better treatment strategy.

Conflict of interests

The authors have nothing to declare.