Skip to content
Publicly Available Published by De Gruyter November 11, 2017

Could a combination of heterozygous ABCC8 and KCNJ11 mutations cause congenital hyperinsulinism?

  • Klara Rozenkova , Azizun Nessa , Barbora Obermannova , Lenka Elblova , Petra Dusatkova EMAIL logo , Zdenek Sumnik , Jan Lebl , Khalid Hussain and Stepanka Pruhova

Abstract

Background:

Congenital hyperinsulinism (CHI) is frequently caused by mutations in one of the KATP channel subunits encoded by the genes ABCC8 and KCNJ11. The effect of simultaneous mutations in both of these genes on the pancreatic β-cell function is not known and patients with CHI carrying both ABCC8 and KCNJ11 mutations have not yet been reported. We questioned if a combination of heterozygous mutations in the ABCC8 and KCNJ11 genes could also lead to β-cell dysfunction presenting as CHI.

Methods:

As a model, we used a patient with transient CHI that paternally inherited novel heterozygous mutations in ABCC8 (p.Tyr1293Asp) and KCNJ11 (p.Arg50Trp) genes. The pathogenic effects on the pancreatic β-cells function were examined in an in vitro functional study using radioactive rubidium efflux assay.

Results:

We showed that the activation of the mutated KATP channels by diazoxide was decreased by 60.9% in the channels with the heterozygous combination of both mutations compared to the wild type channels. This could indicate the pathogenic effect on the pancreatic β-cell function leading to CHI although conclusive evidence is needed to be added.

Conclusions:

Our findings may widen the spectrum of genetic causes of CHI and suggest a novel pathogenic mechanism of CHI that must however, be further investigated.

Introduction

Congenital hyperinsulinism (CHI) is a common cause of persistent neonatal hypoglycemia. CHI is caused by mutations in the genes that are crucial for the regulation of insulin secretion from pancreatic β-cells [1]. In particular, mutations in sulfonylurea receptor 1 (SUR1), one of the adenosine triphosphate-sensitive K+ channel (KATP) subunits, and the pore-forming subunit (Kir6.2), which are encoded by the genes ABCC8 and KCNJ11 [2], respectively, are responsible for the development of CHI. A wide spectrum of mutations in these genes has been described, ranging from severe homozygous and compound heterozygous to milder heterozygous variants [3], [4], [5]. The majority of CHI cases caused by mutations in the ABCC8 or KCNJ11 genes require intensive treatment during the first years of life. A part of the most severe cases of diffuse CHI caused by homozygous mutations in the respective genes that previously had to be treated with near-total pancreatectomy, can now be successfully managed with sirolimus [6]. In contrast, focal lesions (paternally inherited heterozygous mutations) can be localized by 18F 3,4-dihydroxyphenylalanine positron emission tomography/computer tomography and surgically removed. Alternatively, some of these children can be successfully treated with octreotide. Generally, patients with CHI show gradual improvement of their condition over time allowing for less intensive treatment. However, spontaneous remission within the first 3 years of life is not common [3], [7].

Identification of one patient with CHI carrying heterozygous variants in the ABCC8 and KCNJ11 genes simultaneously led us to question if a combination of two mutations in different genes alter the pancreatic β-cell function.

Subjects and methods

Patient – clinical information

The male infant was the second child born to non-consanguineous, healthy Caucasian Czech parents with no family history of hypoglycemia or diabetes. He was born at the 36th week of gestation by normal delivery with an Apgar score of 9–8–8. He was large for his gestational age with a birth weight of 4.0 kg (+2.76 standard deviation score [SDS]) and a length of 51 cm (+1.31 SDS). Gestational diabetes in the mother was excluded by an oral glucose tolerance test (oGTT) in the 24th week of gestation. After birth, the patient was diagnosed with transposition of the great arteries and underwent cardiothoracic surgery at the age of 7 days. Since the first day of life, he presented with persistent non-ketotic hypoglycemia and unsuppressed insulin levels (blood glucose 1.3 mmol/L, β-hydroxybutyrate 0.2 mmol/L, insulin 29.9 mIU/L, C-peptide 1 811.0 pmol/L, cortisol 441 nmol/L). The patient was managed with parenteral administration of dextrose (above 10 mg/kg/min) and fortification of oral feeding with maltodextrin. On the 6th day of life, treatment with s.c. octreotide (5 μg/kg/day) instead of diazoxide was initiated because it was not clear when and how the patient would be able to tolerate oral medication due to cardiothoracic surgery. The dose was gradually increased with a simultaneous reduction in the parenteral administration of dextrose so that it could be stopped once the dose of octreotide reached 25 μg/kg/day on the 14th day of life. Because of the occasionally low blood glucose levels, the octreotide dose was subsequently increased to 35 μg/kg/day. On the 19th day of life, therapy with diazoxide (9.1 mg/kg/day) together with thiazide diuretics was introduced, maintaining blood glucose levels within the normal range and allowing for the reduction of the octreotide dose until it could be stopped. The patient’s blood glucose levels were stable on the diazoxide therapy. However, since the fourth week of life, the patient was noted to be exhausted with feeding and tachycardic. Despite the intensified therapy with diuretics in the fifth week of life, he presented with episodes of desaturation, lethargy and hypotonia. Echocardiography revealed dilated cardiomyopathy affecting both atria and leading to mitral and tricuspid insufficiency. His levels of N-terminal prohormone of brain natriuretic peptide (NT-proBNP) were significantly elevated (16,248 ng/L, normal range 37–646 ng/L) in accordance with the echocardiographic findings. Due to a suspicion of diazoxide cardiotoxicity, the treatment with diazoxide was stopped, and octreotide was re-introduced.

Within several days after stopping the diazoxide treatment, the clinical state of the patient improved, the episodes of desaturation and tachycardia resolved, the echocardiographic findings normalized and the NT-proBNP levels decreased (4388 ng/L). The patient was discharged at the age of 7 weeks on monotherapy with octreotide (35 μg/kg/day), which were tolerated well. At the age of 7 months, the therapy was switched to a more convenient therapy with a long-acting octreotide analog (Sandostatin-LAR) administered i.m. (5 mg) once every 4 weeks. The dose of short-acting octreotide was gradually decreased and eventually stopped 2 months thereafter. To our knowledge, a Sandostatin-LAR has not been used in children with CHI previously. The recommended single dose for an adult is 30 mg. We therefore opted for 4 mg i.m. every 4 weeks for an infant, which proved to be sufficient as the patient’s glycemic control was satisfactory until the age of 2.5 years when the treatment could be stopped with no further episodes of hypoglycemia (able to fast for 11 h, home blood glucose levels between 3.6 and 9.1 mmol/L) and a satisfactory growth trajectory (weight −0.28 SDS, height −1.38 SDS). The patient’s psychomotor development was normal.

Genetic testing

The patient’s genomic DNA was extracted from peripheral blood using the QIAamp DNA Blood Mini Kit (Qiagen, Hilden, Germany). The patient was tested for the genes associated with CHI, as published previously [8]. We tested the complete coding region and the intron-exon boundaries of the ABCC8 (GenBank accession number NG_008867), KCNJ11 (GenBank accession number NG_012446), GCK, HNF4A, HNF1A, GLUD1, HADH, UCP2 and SLC16A1 genes by Sanger sequencing. We also analyzed the ABCC8 gene using multiplex ligation-dependent probe amplification (MRC-Holland, Amsterdam, Netherlands) in order to rule out possible deletions or duplications. To assess the origin of the mutations, sequencing of relevant exons was also carried out in DNA samples from the patient’s parents. All primer sequences and protocols are available upon request.

Written informed consent was obtained from the parents of the patient. The protocol of the study was approved by the Institutional Ethics Committee of the 2nd Faculty of Medicine, Charles University in Prague, the Czech Republic and fulfills the Declaration of Helsinki regarding ethical conduct of research involving human subjects.

Functional study

To determine the pathogenicity of the novel mutations in the KATP channel subunits, we performed an in vitro functional analysis as described previously [8], [9]. Plasmid vectors expressing the mutant SUR1 and Kir6.2 subunits were created by Quick Change Site Directed Mutagenesis Kit (Stratagene, San Diego, USA), the ABCC8 and KCNJ11 gene mutation specific primers used are available upon request. Vectors cDNA3 (Invitrogen, Carlsbad, USA) for SUR1 (accession number L40623) and pcDNA3.1/Zeo (accession number D50581) (Invitrogen) for Kir6.2 were used in this study. A highly conserved and previously exploited [10] cDNA sequences of hamster (ABCC8 cDNA) and mouse (KCNJ11 cDNA) were used as templates. Both of them have 95% sequence homology to the respective human proteins. The online Clustal program was used to align these sequences with the human cDNA in order to locate the mutations found in our patient.

To assess the pathogenicity of the novel variants in genes ABCC8 and KCNJ11, an analysis using radioactive rubidium (Rb86+) efflux assay was performed in human embryonic kidney 293 (HEK293) cell cultures that were transfected with combinations of the mutations in both the KAPT channel subunits.

The mutant and wild type (WT) channels were exposed to five different drug conditions: (1) control – dimethyl sulfoxide (DMSO), (2) 100 μM diazoxide, (3) 100 μM diazoxide and 10 μM glibenclamide, (4) metabolic inhibition (MI) using 2.5 mM NaCN and 20 mM 2-deoxy-D-glucose, and (5) MI and 10 μM glibenclamide. For more details on the experimental conditions see the Figure 2 legend.

Figure 1: Pedigree of the patient’s family.Arrow, proband; black color, clinical symptoms of CHI; white color, healthy; BW, birth weight; LGA, large for gestational age; WT, wild type (mutation absent).
Figure 1:

Pedigree of the patient’s family.

Arrow, proband; black color, clinical symptoms of CHI; white color, healthy; BW, birth weight; LGA, large for gestational age; WT, wild type (mutation absent).

To better understand the effects of the novel mutations on the β-cell function, we first tested the mutations separately in the homozygous and heterozygous states. Then we tested combinations of these mutations in homozygous and heterozygous state, which reflected the situation in our patient.

Results

Genetic testing

The genetic analyses revealed a unique combination of novel heterozygous mutations in both of the KATP channel subunits, namely the novel p.Tyr1293Asp (c.3877T>G) mutation in the ABCC8 gene and the novel p.Arg50Trp (c.148C>T) mutation in the KCNJ11 gene. No mutations were identified in the other common genes associated with CHI that were tested. Both of the identified mutations were not listed in the 1000 Genomes Project, however, the p.Tyr1293Asp in the heterozygous state was reported in four subjects in the ExAc database. On the other hand, none of the reported mutations were detected in 100 chromosomes of healthy control subjects of Czech ancestry, and both were described as pathogenic by the in-silico prediction programs SIFT and PolyPhen-2. Paternal inheritance was observed. However, the father did not have any documented episodes of hypoglycemia in infancy, and his actual glycemic response at the age of 35 years was normal (normal oGTT and a glycated hemoglobin of 32 mmol/mol and 5.1%, respectively). He was significantly macrosomic at birth with a weight of 4.75 kg (+3.15 SDS), which may reflect a defect in insulin secretion in utero and supports the suggested dominant mode of action of the mutations (Figure 1) and a probable mild, transient hyperinsulinism in infancy. Because the paternal mode of inheritance was suggestive of a focal form of CHI, a 18F 3,4-dihydroxyphenylalanine positron emission tomography/computer tomography scan was performed, but no focal lesions were detected.

Figure 2: Determining the functional impact of the SUR1 mutation p.Tyr1293Asp and the Kir6.2 mutation p.Arg50Trp on KATP channel function using the Rb86+ efflux assay.HEK293 cells were transiently transfected with KATP channel subunits at a final concentration of 2 μg for SUR1 (WT/mutant) and 500 ng for Kir6.2 (WT/mutant). The mutant constructs were studied in three different states: homozygous, heterozygous and a combination. The transfected HEK293 cells were then loaded with 0.037 MBq/mL of Rb86+ for 24 h. The cells were later exposed to five different drug conditions for 5 min and, in the case of metabolic inhibition (MI), for 15 min. The drug conditions included the following: control cells treated with DMSO to determine the baseline efflux, 100 μM diazoxide (DZX) to activate channels, 100 μM diazoxide and 10 μM glibenclamide (DZX+GLB) to inhibit channels, 2.5 mM NaCN and 20 mM 2-deoxy-D-glucose (MI) to activate channels, and MI plus 10 μM glibenclamide (MI+GLB) to inhibit channels. The Rb86+ efflux was measured as a percentage of the total Rb86+ in the medium. The graphs show the % mean Rb86+ efflux for WT and individual mutations. Data were analyzed using one-way analysis of variance (ANOVA), followed by Bonferroni’s post t-test, n=3 (in triplicates), ***p<0.001. The first graph (A) shows response to MI and MI+GLB of the untransfected cells (UTF), cells transfected with Kir6.2 only (Kir6.2), WT cells and the respective mutations in the homozygous (designated as p.Tyr1293Asp or p.Arg50Trp) or heterozygous (p.Tyr1293Asp/WT or p.Arg50Trp/WT) state, respectively. The second graph (B) shows response to DMSO, DZX and DZX+GLB of the respective mutations in comparison to WT cells.
Figure 2:

Determining the functional impact of the SUR1 mutation p.Tyr1293Asp and the Kir6.2 mutation p.Arg50Trp on KATP channel function using the Rb86+ efflux assay.

HEK293 cells were transiently transfected with KATP channel subunits at a final concentration of 2 μg for SUR1 (WT/mutant) and 500 ng for Kir6.2 (WT/mutant). The mutant constructs were studied in three different states: homozygous, heterozygous and a combination. The transfected HEK293 cells were then loaded with 0.037 MBq/mL of Rb86+ for 24 h. The cells were later exposed to five different drug conditions for 5 min and, in the case of metabolic inhibition (MI), for 15 min. The drug conditions included the following: control cells treated with DMSO to determine the baseline efflux, 100 μM diazoxide (DZX) to activate channels, 100 μM diazoxide and 10 μM glibenclamide (DZX+GLB) to inhibit channels, 2.5 mM NaCN and 20 mM 2-deoxy-D-glucose (MI) to activate channels, and MI plus 10 μM glibenclamide (MI+GLB) to inhibit channels. The Rb86+ efflux was measured as a percentage of the total Rb86+ in the medium. The graphs show the % mean Rb86+ efflux for WT and individual mutations. Data were analyzed using one-way analysis of variance (ANOVA), followed by Bonferroni’s post t-test, n=3 (in triplicates), ***p<0.001. The first graph (A) shows response to MI and MI+GLB of the untransfected cells (UTF), cells transfected with Kir6.2 only (Kir6.2), WT cells and the respective mutations in the homozygous (designated as p.Tyr1293Asp or p.Arg50Trp) or heterozygous (p.Tyr1293Asp/WT or p.Arg50Trp/WT) state, respectively. The second graph (B) shows response to DMSO, DZX and DZX+GLB of the respective mutations in comparison to WT cells.

Functional study

The β-cell function was assessed as KATP channel activity when exposed to diazoxide. Both novel mutations significantly decreased the activity of the KATP channel in comparison to WT channels when expressed separately in a homozygous state – by 84.3% for mutation p.Tyr1293Asp in gene ABCC8 and by 75% for mutation p.Arg50Trp in the gene KCNJ11 (see Figure 2). On the other hand, when expressed separately in a heterozygous state the effect on the KATP channel activity was smaller (38.5% and 45.4%, respectively, see Figure 2).

The most deteriorating effect on the β-cell function represented by KATP channel activation by diazoxide was observed when a combination of the homozygous mutations was expressed simultaneously – channel activity decreased by 86.3% (Figure 2).

The functional study of the unique combination of heterozygous mutations, which accurately mimics the situation in our patient, revealed that the activation of mutated KATP channels by diazoxide was decreased by 60.9% when compared to WT channels (Figure 2).

To assess the activation of endogenous pathways via MI, additional control experiments including un-transfected cells (UTF) and cells transfected with Kir6.2 only were performed. The WT KATP channels had the strongest response to MI with an Rb86+ efflux of 65.5%±15.5% after a 15-min incubation, and there was a response to MI in some of the mutants.

Discussion

The p.Arg50Trp mutation in the KCNJ11 gene acts as a dominant mutation. It was non-functional when expressed in a homozygous state, but its activity improved in the presence of WT subunits. The arginine residue at position 50 may play an indirect role in the process of ATP binding, as well as affecting ATP sensitivity within the Kir6.2 protein [11], [12]. Therefore, we hypothesize that the ATP sensitivity was enhanced, which explains why the channel remained closed even in the presence of diazoxide.

The p.Tyr1293Asp mutation in the ABCC8 gene is located within transmembrane domain 2 of the SUR1 protein. This channel was activated by MI even when expressed in the homozygous state. It is likely that the MgADP-binding site was still intact, which was why the channel was responsive. However, the diazoxide-binding site had clearly been damaged.

A combination of the mutations and the WT subunits showed that functional KATP channels remained, which were responsive to channel activators. This explains why the patient was responsive to medical treatment. Moreover, the mild effect of these mutations explains the tendency for spontaneous remission of CHI.

The results of this functional study could indicate that the presence of single heterozygous mutations alters the KATP channel slightly, and the combination of these mutations intensifies their pathogenic effects. However, against this hypothesis are presence (albeit in single cases) of the ABCC8 variant p.Tyr1293Asp in databases of polymorphisms, inconsistent phenotype of the proband and his father and maintained activation of the channel by MI in the experiment. Moreover, due to financial and technical reasons we were not able to perform surface expression assays that could assess the protein expression of the mutant constructs in HEK293 cells. Therefore, further investigation using additional functional methods and/or additional detection of patients carrying combination of variants in the genes encoding KATP channel subunits would be needed to confirm this newly described phenomenon.

From the clinical point of view it is important to note that the mild phenotype could be missed during the neonatal and infant periods. This could explain that the same combination of mutations was detected in the patient’s healthy father, who was born large for his gestational age, but had no documented episodes of hypoglycemia. Alternatively, as the phenotypical variabilities in identical mutations detected even in members from same families are very common, it is definitely possible that the functional characteristics of a mutation evaluated in vitro could be different from the in vivo effects.

In conclusion, we have reported for the first time a case involving a patient with transient CHI that might be caused by a combination of novel heterozygous mutations in the genes encoding the KATP channel subunits. Although our results indicate a pathogenic effect of this combination of mutations on pancreatic β-cell function through an in vitro functional study, further investigation is needed to confirm it. Our findings will then widen the spectrum of genetic causes of CHI and suggested a novel pathogenic mechanism of CHI.


Corresponding author: Petra Dusatkova, MSc, PhD, Department of Pediatrics, 2nd Faculty of Medicine, Charles University in Prague, University Hospital Motol, V Uvalu 84, 150 06 Prague 5, Czech Republic, Phone: +420 224 432 026, Fax: +420 224 432 221

  1. Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.

  2. Research funding: The work was supported by the Grant Agency of Charles University (GAUK 248 213), the Project for Conceptual Development of Research Organization 00064203/6001 (Ministry of Health, Czech Republic) and the ESPE Short-term Research Fellowship, which allowed Klara Rozenkova to perform functional studies of the ABCC8 and KCNJ11 gene mutations at the Institute of Child Health, University College London, UK.

  3. Employment or leadership: None declared.

  4. Honorarium: None declared.

  5. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

References

1. Rozenkova K, Guemes M, Shah P, Hussain K. The diagnosis and management of hyperinsulinaemic hypoglycaemia. J Clin Res Pediatr Endocrinol 2015;7:86–97.10.4274/jcrpe.1891Search in Google Scholar PubMed PubMed Central

2. Flanagan SE, Kapoor RR, Hussain K. Genetics of congenital hyperinsulinemic hypoglycemia. Semin Pediatr Surg 2011;20:13–7.10.1053/j.sempedsurg.2010.10.004Search in Google Scholar PubMed

3. Pinney SE, MacMullen C, Becker S, Lin YW, Hanna C, et al. Clinical characteristics and biochemical mechanisms of congenital hyperinsulinism associated with dominant KATP channel mutations. J Clin Invest 2008;118:2877–86.10.1172/JCI35414Search in Google Scholar PubMed PubMed Central

4. Huopio H, Reimann F, Ashfield R, Komulainen J, Lenko HL, et al. Dominantly inherited hyperinsulinism caused by a mutation in the sulfonylurea receptor type 1. J Clin Invest 2000;106:897–906.10.1172/JCI9804Search in Google Scholar PubMed PubMed Central

5. Nessa A, Aziz QH, Thimas AM, Harmer SC, Tinker A, et al. Molecular mechanisms of congenital hyperinsulinism due to autosomal dominant mutations in ABCC8. Hum Mol Genet 2015;24:5142–53.10.1093/hmg/ddv233Search in Google Scholar PubMed

6. Senniappan S, Brown RE, Hussain K. Sirolimus in severe hyperinsulinemic hypoglycemia. N Engl J Med 2014;370:2448–9.10.1056/NEJMc1404716Search in Google Scholar PubMed

7. Thornton PS, MacMullen C, Ganguly A, Ruchelli E, Steinkrauss L, et al. Clinical and molecular characterization of a dominant form of congenital hyperinsulinism caused by a mutation in the high-affinity sulfonylurea receptor. Diabetes 2003;52:2403–10.10.2337/diabetes.52.9.2403Search in Google Scholar PubMed

8. Rozenkova K, Malikova J, Nessa A, Dusatkova L, Bjorkhaug L, et al. High incidence of heterozygous ABCC8 and HNF1A mutations in Czech patients with congenital hyperinsulinism. J Clin Endocrinol Metab 2015;100:E1540–9.10.1210/jc.2015-2763Search in Google Scholar PubMed

9. Muzyamba M, Farzaneh T, Behe P, Thomas A, Christesen HB, et al. Complex ABCC8 DNA variations in congenital hyperinsulinism: lessons from functional studies. Clin Endocrinol (Oxf) 2007;67:115–24.10.1111/j.1365-2265.2007.02847.xSearch in Google Scholar PubMed

10. Macmullen CM, Zhou Q, Snider KE, Tewson PH, Becker SA, et al. Diazoxide-unresponsive congenital hyperinsulinism in children with dominant mutations of the β-cell sulfonylurea receptor SUR1. Diabetes 2011;60:1797–804.10.2337/db10-1631Search in Google Scholar PubMed PubMed Central

11. Shimomura K, Girard CA, Proks P, Nazim J, Lippiat JD, et al. Mutations at the same residue (R50) of Kir6.2 (KCNJ11) that cause neonatal diabetes produce different functional effects. Diabetes 2006;55:1705–12.10.2337/db05-1640Search in Google Scholar PubMed

12. Proks P, Gribble FM, Adhikari R, Tucker SJ, Ashcroft FM. Involvement of the N-terminus of Kir6.2 in the inhibition of the KATP channel by ATP. J Physiol 1999;514:19–25.10.1111/j.1469-7793.1999.019af.xSearch in Google Scholar PubMed PubMed Central

Received: 2017-4-21
Accepted: 2017-9-28
Published Online: 2017-11-11
Published in Print: 2017-11-27

©2017 Walter de Gruyter GmbH, Berlin/Boston

Downloaded on 29.3.2024 from https://www.degruyter.com/document/doi/10.1515/jpem-2017-0163/html
Scroll to top button