Targeted gene panel analysis of Japanese patients with maturity‐onset diabetes of the young‐like diabetes mellitus: Roles of inactivating variants in the ABCC8 and insulin resistance genes

Abstract

Aims/Introduction: To investigate the genetic background of Japanese patients with suspected maturity-onset diabetes of the young (MODY). Materials and Methods: On 340 proband patients referred from across Japan, genomic variants were analyzed using a targeted multigene panel analysis combined with the multiplex ligation probe amplification (MLPA) analysis, mitochondrial m.3243A > G analysis and methylation-specific polymerase chain reaction of the imprinted 6q24 locus. Pathogenic/likely pathogenic variants were listed according to the 2015 American College of Medical Genetics and Genomics and the Association for Molecular Pathology criteria. Additionally, variants with a population frequency <0.001 and Combined Annotation Dependent Depletion score >20 (CS >20) were listed as rare variants of uncertain significance-CS >20. Results: A total of 157 pathogenic/likely pathogenic variants and 44 rare variants of uncertain significance-CS >20 were identified. In the pathogenic/likely pathogenic variants, alterations in the GCK gene were the most common (82, 52.2%) followed by HNF1A (29, 18.5%), HNF4A (13, 8.3%) and HNF1B (13, 8.3%). One patient was a 29.5% mosaic with a truncating INSR variant. In the rare variants of uncertain significance-CS >20, 20 (45.5%) were in the genes coding for the adenosine triphosphate-sensitive potassium channel, KCNJ11 or ABCC8, and four were in the genes of the insulin-signaling pathway, INSR and PIK3R1. Four variants in ABCC8 were previously reported in patients with congenital hyperinsulinism, suggesting the inactivating nature of these variants, and at least two of our patients had a history of congenital hyperinsulinism evolving into diabetes. In two patients with INSR or PIK3R1 variants, insulin resistance was evident at diagnosis. Conclusions: Causative genomic variants could be identified in at least 46.2% of clinically suspected MODY patients. ABCC8-MODY with inactivating variants could represent a distinct category of MODY. Genes of insulin resistance should be included in the sequencing panel for MODY.

Clinical characteristics in children with maturity-onset diabetes of the young detected by urine glucose screening at schools in the Tokyo Metropolitan Area

“…In the present study, which was conducted on 55 non-obese children clinically diagnosed with type 2 diabetes, IGF, and IGT, 21 were identified as having MODY by genetic testing. This mutation identification rate (38.2%) was similar to that reported in previous studies conducted on Japanese children ( 22 , 23 ). Urine glucose screening in schools can detect not only a large number of children with type 2 diabetes but also MODY before the progression of diabetes.…”

“…The most common type of MODY is reported to be MODY3, as evidenced by its detection in Europe, North America, and Asia (24,25). In contrast, Yorifuji et al (22,23) reported that the most common type of MODY among Japanese children was MODY2, followed by MODY3. The present study also showed that the most common type was MODY2 with the same frequency as MODY3.…”

“…Patients with MODY2 do not usually require pharmacological interventions for glycemic control, as demonstrated in the present study. However, Kawakita et al (23) reported that 7 out of 55 patients with MODY2 were treated with oral hypoglycemic drugs: i.e., four with a SU, one with metformin, and two with an α-glucosidase inhibitor over time. In contrast, the present study demonstrated that 11 of 13 patients with MODY3, 1, and 5 required pharmacological treatment to control hyperglycemia within one month after diagnosis.…”

“…This study had several strengths. First, some Japanese children with type 2 diabetes are non-obese, which is a notable characteristic of the Japanese population, and the clinical features of non-obese type 2 diabetes are similar to those observed in MODY (13,14,22,23). It is difficult to diagnose non-obese type 2 diabetes based solely on clinical characteristics and laboratory data at diagnosis (22,23).…”

“…They were then analyzed using an ABI PRISM 3100xl automated sequencer (Applied Biosystems, Foster City, CA, USA). When negative, the sequencing of HNF4A , HNF1A and HNF1B was performed using targeted multigene panel analysis combined with multiplex ligation-dependent probe amplification (MLPA) analysis, as previously reported ( 22 , 23 , 24 ). Reactions were performed using the SALSA MLPA kit P241 (MRC-Holland, Amsterdam, Netherlands).…”