Persistent hypoglycaemia in infancy is most commonly caused by hyperinsulinism. A case is reported of the somatic loss of the maternal 11p in an insulin secreting focal adenoma in association with a germline SUR-1 mutation on the paternal allele in a baby boy with hyperinsulinism diagnosed at 49 days old. A reduction to homozygosity of an SUR-1 mutation is proposed as a critical part of the cause of focal hyperinsulinism.
- sulphonylurea receptor
- mutational analysis
Statistics from Altmetric.com
Persistent hypoglycaemia in infancy is most commonly caused by hyperinsulinism. It is characterised by profound hypoglycaemia caused by inappropriate secretion of insulin. Hyperinsulinism may be inherited in both autosomal recessive1 and dominant2 forms. At the molecular level, mutations in the sulphonylurea receptor (SUR-1)3 and the potassium channel subunit (KIR6.2)4 have been identified in autosomal recessive forms of the disease. Somatic loss of maternal chromosome 11p has been demonstrated in non-familial cases with areas of focal adenomatosis5 and in adenomas6 of the pancreas. In these reports, the minimum region of chromosome 11p lost included the SUR-1 and KIR6.2 genes, as well as at least two maternally expressed imprinted tumour suppresser genes (H19 and P57KIP2).
We report the somatic loss of the maternal 11p in an insulin secreting focal adenoma in association with a germline SUR-1 mutation on the paternal allele in a patient with hyperinsulinism. We propose a reduction to homozygosity of an SUR-1 mutation as a critical part of the cause of focal hyperinsulinism.
A baby boy, with no family history of hypoglycaemia, presented with hypoglycaemia on day 42 of life. Hyperinsulinism was diagnosed at 49 days based on the finding of increased glucose requirement (15 mg/kg/min) to maintain euglycaemia. At a plasma glucose of 1.7 mmol/l the patient’s insulin concentration was 297 pmol/l (normal < 14 pmol/l); at a plasma glucose of 0.4 mmol/l, insulin was 75 pmol/l, 3-hydroxybutyrate was suppressed at 0.02 mmol/l (normal > 1.5 mmol/l), and the patient had a glycaemic response to glucagon of > 1.7 mmol/l. Diazoxide and octreotide treatment was unsuccessful and surgery was performed on day 84.
At surgery an adenoma was palpated and partial pancreatectomy (approximately 40%) performed. Histology confirmed an insulin staining adenoma with normal pancreatic tissue in the surrounding area. The patient recovered well, tolerated a 13 hour fasting interval at age 4 months, and has had no subsequent hypoglycaemia.
Genomic DNA was extracted from patient and parental peripheral blood leucocytes, and from patient paraffin embedded tissue using standard techniques. SUR-1 exon 10 sequences were amplified using standard polymerase chain reaction (PCR) conditions (GibcoBRL; Life Technologies, Paisley, UK) using exon 10 primers SUR-10F (5′–GTGGAGACGACCCGCA GGAAGGAG–3′) and SUR-10R (5′–CCTT GCATGTACGCAGCAGCACCC–3′). Equal quantities of PCR product was digested for at least two hours with Bsr1 using conditions as described by the manufacturer. Bsr1 digestion of wild-type product yields two fragments of 66 and 50 base pairs. The mutated allele remains uncut. For GATA23FO6 and TH7 microsatellite analysis, PCR was performed as above and analysed on a 7% Long Ranger gel run on a Pharmacia ALF automatic sequencing system (Pharmacia BioFech, St Albans, UK).
Patient leucocyte DNA was screened for 24 previously identified SUR-1 mutations8 and 2 KIR6.2 mutations; heterozygosity for an SUR 1 intron 10 splice site mutation (1630+1 G→T) was discovered. This mutation had previously been identified in three other patients with hyperinsulinism. One was a compound heterozygote for this and another SUR-1 mutation, whereas in the other two the mutation was identified on the paternal allele with no mutation identified on the maternal allele. As this mutation destroys a Bsr1 restriction site, enzyme digestion of exon 10 PCR product was used to identify the mutation in patient and parental samples (fig 1), demonstrating that the mutation was paternally derived. Analysis of DNA from the patient’s normal pancreatic tissue demonstrated a heterozygous pattern of normal and mutant allele. In contrast, DNA extracted from adenoma tissue contained mostly mutant allele.
To investigate if this was the result of loss of the maternal allele we performed microsatellite analysis using 11p specific markers GATA23FO6 and TH. TH lies approximately 30 centimorgans telomeric to the SUR-1 gene with the GATA23FO6 between the TH and SUR-1. The results demonstrated a Mendelian inheritance pattern in the patient’s leucocyte and normal pancreatic tissue, and a loss of the maternal 11p in the adenoma (fig 2). Our results indicate that the hyperinsulinism in this case is as a result of a reduction to homozygosity of the SUR-1 mutation in the adenoma tissue.
de Lonlay and colleagues5 demonstrated loss of heterozygosity (LOH) in focal hyperplasia of the pancreas. They proposed that the hyperplasia was either caused by a loss of the imprinted tumour suppresser genes H19 or P57KIP2, or an expression imbalance between H19 and IGF II. Their studies explain the hyperplasia in focal disease but do not adequately address the hypersecretion of insulin in the focal tissue. Beckwith-Wiedemann syndrome, a condition in which 20% have LOH of chromosome 11p (the region containing the SUR-1 locus) is associated with hyperinsulinism in 30% of cases, but the occurrence of hyperinsulinism in these patients does not correlate with the occurrence of the LOH.7 We propose a model where the somatic loss of the maternal 11p is the causative factor for focal hyperplasia but that the hyperplasia itself is not sufficient to cause insulin hypersecretion. With the additional presence of a paternally derived SUR-1 mutation, there is a reduction to homozygosity for SUR-1 in focal cells resulting in dysregulation of insulin secretion and ultimately to the hyperinsulinism phenotype. Resection of the affected tissue resulted in a cure in a similar fashion to the partial pancreatectomy performed by de Lonlay et al in focal hyperplasia, unlike autosomal recessive hyperinsulinism in which a 95% pancreatectomy is often insufficient. In conclusion, this case demonstrates a novel mechanism for the apparent sporadic inheritance of disease considered an autosomal recessive disorder.
The authors thank Professor Andrew Green for helpful discussion. MAP is the recipient of an NIH grant.