Diazoxide-unresponsive congenital hyperinsulinism in children with dominant mutations of the β-cell sulfonylurea receptor SUR1

Abstract

Objective: Congenital hyperinsulinemic hypoglycemia is a group of genetic disorders of insulin secretion most commonly associated with inactivating mutations of the β-cell ATP-sensitive K(+) channel (K(ATP) channel) genes ABCC8 (SUR1) and KCNJ11 (Kir6.2). Recessive mutations of these genes cause hyperinsulinism that is unresponsive to treatment with diazoxide, a channel agonist. Dominant K(ATP) mutations have been associated with diazoxide-responsive disease. We hypothesized that some medically uncontrollable cases with only one K(ATP) mutation might have dominant, diazoxide-unresponsive disease.

Research design and methods: Mutations of the K(ATP) genes were identified by sequencing genomic DNA. Effects of mutations on K(ATP) channel function in vitro were studied by expression in COSm6 cells.

Results: In 15 families with diazoxide-unresponsive diffuse hyperinsulism, we found 17 patients with a monoallelic missense mutation of SUR1. Nine probands had de novo mutations, two had an affected sibling or parent, and four had an asymptomatic carrier parent. Of the 13 different mutations, 12 were novel. Expression of mutations revealed normal trafficking of channels but severely impaired responses to diazoxide or MgADP. Responses were significantly lower compared with nine SUR1 mutations associated with dominant, diazoxide-responsive hyperinsulinism.

Conclusions: These results demonstrate that some dominant mutations of SUR1 can cause diazoxide-unresponsive hyperinsulinism. In vitro expression studies may be helpful in distinguishing such mutations from dominant mutations of SUR1 associated with diazoxide-responsive disease.

FIG. 1.

Pedigrees of families with dominant, diazoxide-unresponsive mutations of SUR1. The pedigrees are listed in ascending order of SUR1 codons. Fraternal twins are indicated by diagonal lines connecting the offspring to the parents. Arrows indicate probands. Black shapes, hypoglycemia diagnosed; diagonal lines, asymptomatic carrier; and filled gray, suspected hypoglycemia. n/M, mutation positive; n/n, mutation negative.

FIG. 2.

Location of dominant, diazoxide-unresponsive (black circles) and dominant, diazoxide-responsive (gray circles) mutations of SUR1 (7). Mutation locations are approximate based on current SUR1 topology models (17,18). Secondary structures of SUR1 are shown, including the amino (NH₂-) and carboxy (COOH-) termini of the protein, glycosylation sites, and nucleotide binding fold 1 (NBF1) and nucleotide binding fold 2 (NBF2), which contain Walker A and B motifs (A and B) associated with regulatory nucleotide binding.

FIG. 3.

Functional analysis of mutant SUR1 expressed in COSm6 cells. A: Western blot analysis of mutant flag-SUR1 proteins. COSm6 cells were cotransfected with cDNAs for Kir6.2 and wild-type (WT) FLAG-SUR1, the diazoxide (Diaz)-responsive mutant R1539E FLAG-SUR1 or the diazoxide-unresponsive mutant S1387Y FLAG-SUR1. Both mutants exhibit the lower core glycosylated band (solid arrow) and the upper complex glycosylated band (open arrow) as seen in wild-type, indicating that the mutant proteins are processed correctly like the wild-type protein. B: Immunofluorescence staining of surface channels in COSm6 cells transfected with wild-type, R1539E, or S1387Y FLAG-SUR1 and wild-type Kir6.2. Both mutants were well expressed at the cell surface like wild-type channels. C: Representative inside-out patch-clamp recordings of wild-type, the diazoxide-responsive mutant R1539E, and the diazoxide-unresponsive mutant S1387Y show differences in MgADP response. Currents were measured at –50 mV in symmetrical K-INT solution, and inward currents are shown as upward deflections. Patches were exposed to differing concentrations of ATP and ADP, as indicated by the bars above the records. Free Mg²⁺ concentration was maintained at 1 mmol/L in all ATP-containing solutions. D: The same as in C except that channel response to diazoxide was compared. E: Quantification of MgADP response (left panel) and diazoxide response (right panel) using recordings shown in C and D. Currents were normalized to that seen in K-INT and expressed as percentage of currents. Responses in both homozygous and simulated heterozygous expression conditions are shown. In the MgADP response graph, each bar represents means ± SEM of 16 wild-type and 4–6 mutant patches. In the MgADP response graph, each bar represents means ± SEM of 15 wild-type and 4–6 mutant patches. (A high-quality digital representation of this figure is available in the online issue.)

FIG. 4.

Responsiveness of dominant diazoxide-unresponsive and dominant diazoxide-responsive SUR1 mutations. Responses to diazoxide and MgADP of expressed K_ATP channels containing diazoxide-unresponsive SUR1 mutations (black circles) and diazoxide-responsive SUR1 mutations (gray squares) are compared. Diazoxide-unresponsive mutations are labeled according to numbers in Table 2. Diazoxide-responsive mutations are as follows: A, D310N; B, R370G; C, R1353H; D, K1374R; E, G1478V; F, G1479R; G, S1386P; H, R1539Q; I, I1512T; and J, E1507K.