Modulation of insulin release by adenosine A1 receptor agonists and antagonists in INS-1 cells: the possible contribution of 86Rb+ efflux and 45Ca2+ uptake.

Due to the lack of specific agonists and antagonists the role of adenosine receptor subtypes with respect to their effect on the insulin secretory system is not well investigated. The A1 receptor may be linked to different 2nd messenger systems, i.e. cAMP, K+- and 45Ca2+ channel activity. Partial A1 receptor agonists are going to be developed in order to improve diabetes (increase in insulin sensitivity, lowering of FFA and triglycerides). In this study newly synthesized selective A1 receptor agonists and antagonists were investigated thereby integrating three parameters, insulin release (RIA), 45Ca2+ uptake and 86Rb+ efflux (surrogate for K+ efflux) of INS-1 cells, an insulin secretory cell line. The presence of A1-receptors was demonstrated by Western blotting. The receptor nonselective adenosine analogue NECA (5-N-ethylcarboxyamidoadenosine) at high concentration (10 microM) had no effect on insulin release and 45Ca2+ uptake which could be interpreted as the sum of effects mediated by mutual antagonistic adenosine receptor subtypes. However, an inhibitory effect mediated by A1 receptor agonism was detected at 10 nM NECA and could be confirmed by adding the A1 receptor antagonist PSB-36 (1-butyl-8-(3-noradamantyl)-3-(3-hydroxy-propyl)xanthine). NECA inhibited 86Rb+ efflux which, however, did not fit with the simultaneous inhibition of insulin secretion. The selective A1 receptor agonist CHA (N6-cyclohexyladenosine) inhibited insulin release; the simultaneously increased Ca2+ uptake (nifedipine dependent) and inhibition of 86Rb+ efflux did not fit the insulin release data. The CHA effect (even the maximum effect at 50 microM) can be increased by 10 microM NECA indicating that CHA and NECA have nonspecific and physiologically non-relevant effects on 86Rb+ efflux in addition to their A1-receptor interaction. Since PSB-36 did not influence the NECA-induced inhibition of 86Rb+ efflux, the NECA effect is not mediated by potassium channel-linked A1 receptors. The nonselective adenosine receptor antagonist caffeine increased insulin release which was reversed by CHA as expected when hypothesizing that both act via A1 receptors in this case. In conclusion, stimulation of A1 receptors by receptor selective and nonselective compounds reduced insulin release which is not coupled to opening of potassium channels (86Rb+ efflux experiments) or inhibition of calcium channels (45Ca2+ uptake experiments). It may be expected that of all pleiotropic 2nd messengers, the cAMP system (not tested here) is predominant for A1 receptor effects and the channel systems (K+ and Ca2+) are of minor importance and do not contribute to insulin release though being coupled to the receptor in other tissues.