The mechanism underlying the normal glucose-induced suppression of glucagon release is, at least in part, thought to involve the action of GABA. In response to high glucose concentrations, GABA is released from pancreatic beta cells and activates GABA_A receptor (GABA_AR) chloride channels in alpha cells (Wendt et al. 2004; Rorsman et al. 1989). This hyperpolarizing action of GABA inhibits the secretion of glucagon. At present, very little is known about the composition or regulation of GABA_ARs in pancreatic alpha cells. Neuronal GABA_ARs show a high degree of plasticity in their subunit composition with the expression of individual subunits being altered in response to chronic drug treatments or ethanol withdrawal, as well as in disease states such as epilepsy or anxiety (e.g. Bailey et al. 2004). It is not known whether pancreatic GABA_ARs exhibit a similar degree of plasticity. Here we have examined the expression of GABA_AR subunit mRNAs in mouse and rat pancreatic tissues. In addition, we have tested whether the expression of GABA_ARs is influenced by depolarizing concentrations of potassium, GABA or insulin in the glucagon-releasing mouse alpha-TC1-9 cell line.

Trizol reagent was used to isolate total RNA from pancreatic islet preparations from adult CD1 mice (n=4) or adult Wistar rats (n=2). One step RT-PCR reactions were performed using published gene specific primers for GABA_AR α1-6, β1-3, γ1-3, and subunit mRNAs. Robust expression of α4 and β3 subunits, and somewhat weaker γ2 expression, was evident in both mouse and rat islet tissues, with no detectable expression of other α/β/γ subunits. The only species specific difference observed was in the weak expression of and subunits, in mouse and rat islets, respectively.

Cultured alpha-TC1-9 cell lines express only α4, β3 and γ2 GABA_AR subunits. Cells were cultured to 70% confluency and then incubated for 24 h to test the effects of high K⁺ (25 mM), GABA (25 m M) or insulin (25 nM) on GABA_AR gene expression. Quantitative real-time RT-PCR was used to assess the abundance of α4, β3 and γ2 mRNAs in treated samples, relative to control samples, using the “delta-delta threshold cycle method” (Bailey et al. 2004). High K + or GABA, but not insulin, significantly reduced the expression of α4 subunit to 47 ± 12% and 59 ± 5% of control levels (mean ± SD, n=3, paired Student’s t-test, P<0.05). GABA, alone of the tested stimuli, reduced the expression of the β3 subunit to 65 ± 10% of control levels. None of the treatments altered the expression of γ2 subunit. .

These results demonstrate that pancreatic islets and alphaTC1 cells express GABA_A R α4, β3 and γ2 subunit mRNAs suggesting that receptor assembly, with the stoichiometry of 2α , 2β and 1γ subunit proposed in the brain, is also likely to occur in pancreatic cells. Furthermore, we show that depolarizing concentrations of potassium or GABA can regulate the expression of individual subunits in pancreatic alpha cells, as they do in isolated neurones (Ives et al. 2002; Obrietan et al. 2001). This suggests that the local cellular environment in the pancreatic islet could be an important determinant of GABA_AR expression, in turn contributing to the control of hormone release and intra-islet feedback.

Bailey, SJ. et al. (2004). Journal of Neuroscience 24, 343-6351.
Ives, JH. et al. (2002). Neuropharmacology 43, 15-725.
Obrietan, K. et al. (2001). Journal of Neurophysiology 88, 1005-1015.
Rorsman, P. et al. (1989). Nature 341, 233-236.
Wendt, A. et al. (2004). Diabetes 53, 1038-1045.