Recent advances have revealed that hyperinsulinism in infancy is a disorder caused by loss of function of K-ATP channels (OMIM 600937, 600509). By contrast, little is known about the pathophysiology of human insulinomas, which are the most common cause of hypoglycaemia in adulthood.

Insulinomas were surgically removed from the pancreata of five patients and a controlled collagenase digestion procedure used to liberate single β-cells. Cells were maintained under standard tissue culture conditions in supplemented RPMI 1640 medium at 37 °C. Control human β-cells and islets were obtained from 56 cadaver organ donors. The function of K-ATP channels, voltage-gated Ca²⁺ channels (VGCC), and voltage-gated K⁺ channels (VGKC) was assessed using patch clamp electrophysiology. Intracellular Ca²⁺ levels [Ca²⁺]_i were estimated using fura-2-loaded cells or islets with microfluorimetry or digital imaging techniques. Results are expressed as means ± S.E.M. and Mann-Whitney rank sum tests were used to test for statistically significant differences.

Total RNA was extracted from cells or islets and used for RT-PCR to determine expression of glucose transporter and ion channel subunit genes using specific oligonucleotide primers. Unlike control islets, insulinoma β-cells were found to exclusively express the GLUT1 facilitated glucose transporter and not a combination of both GLUT-1 and -2 mRNAs. SUR1 and Kir6.2 mRNAs were expressed in insulinoma β-cells, and K-ATP channel currents were consistently recorded from all five patient tissues (n = 4-46). Although peak currents following patch excision were reduced by ~50 % (P < 0.001) when compared with control β-cells (11.6 ± 1.0 pA (n = 106, 5 patients) vs. 25.5 ± 1.5 pA (n = 269, 56 donors)), K-ATP channels responded normally to internally applied nucleotides (ATP 10 µM to 5 mM, ADP 500 µM, GDP 500 µM) and the K-ATP channel agonists diazoxide (10 µM to 500 µM), BPDZ 154 (10 µM) and NNC55-0118 (10 nM to 200 µM). Control human islets and insulinoma β-cells expressed VGCC subunit mRNAs for Cav1.2, 1.3, 2.1, 2.2, β2 and β3. Functional studies revealed that whilst there were no major differences in the magnitude of VGCCs (-11.8 ± 3.3 pA pF^-1 (n = 7) vs. -9.2 ± 0.7 pA pF^-1 (n = 56 from 10 donors, n.s.) at -10 mV from a holding potential of -80 mV), in one out of three patients there were marked increases in the basal cytosolic Ca²⁺ concentrations: 154 ± 12 nM (n = 74) vs. 80 ± 9 nM (n = 677 from 53 control donors, P < 0.001). We found no differences in the mRNA expression profiles of Kv α- and β-subunits, although VGKC currents were markedly increased in magnitude when compared to controls: 392 ± 50 pA pF^-1 (n = 6, 1 patient) vs. 132 ± 9 pA pF^-1 (n = 40 from 6 donors) at +80 mV from a holding potential of -70 mV (P < 0.001).

In summary, insulinoma β-cells selectively expressed the GLUT-1 glucose transporter, but there were no major alterations in the expression profiles or functions of K-ATP channels, VGKC or VGCC. Collectively, these data reveal that defects in glucose-regulated ion channels are unlikely to explain the general pathophysiology of inappropriate insulin release in human insulinomas.