Adenosine pre- and postsynaptic modulation of glutamate-dependent calcium activity in hypothalamic neurons.

Abstract

1. Within the hypothalamus, adenosine has been reported to influence temperature regulation, sleep homeostasis, and endocrine secretions. The effects of adenosine on hypothalamic neurons have not been studied at the cellular level. Adenosine (5 nM-30 microM) showed no influence on intracellular Ca2+ or electrical activity in the presence of glutamate receptor antagonists D-2-amino-5-phosphonovalerate and 6-cyano-7-nitroquinoxaline-2,3-dione; consequently, we examined the role of adenosine in modulating the activity of glutamate in cultured hypothalamic neurons (n > 1,700) with fura-2 Ca2+ digital imaging and whole cell patch-clamp electrophysiology in the absence of glutamate receptor block. 2. When glutamate receptors were not blocked, adenosine (1-30 microM) and the selective adenosine A1 receptor agonist N6-cyclopentyl adenosine (CPA; 5 nM-1 microM) caused a large reduction in intracellular Ca2+ and electrical activity, suggesting that glutamate neurotransmission was critical for an effect of adenosine to be detected. Neuronal Ca2+ levels were reversibly depressed by CPA (50 nM), with a maximum depression of 90%, and these effects were blocked by coadministration of the A1 receptor antagonist 8-cyclopentyl-1,3-dipropylxanthine (DPCPX). 3. Ca2+ levels in immature neurons before the time of synaptogenesis were not affected by adenosine. Adenosine A1 receptor activation suppressed glutamate-mediated Ca2+ activity in neurons in vitro 8 to 73 days. 4. Adenosine (1 or 10 microM) caused a hyperpolarization of membrane potential and a reduction of large postsynaptic potentials arising from endogenously released glutamate. The administration of low concentrations of CPA (5 nM) decreased the frequency of glutamate-mediated, neuronally synchronized Ca2+ transients and the frequency of postsynaptic potentials. 5. To compare the relative effects of adenosine on hypothalamic neurons with cells from other brain regions, we assayed the effects of CPA on glutamate-mediated Ca2+ in hippocampal and cortical cultures. CPA (50 nM) reversibly depressed glutamate-mediated Ca2+ rises in hypothalamic neurons by 35%, compared with 54% in hippocampal neurons and 46% in cortical neurons. 6. If it does play a functional role, adenosine should be released by hypothalamic cells. In some neurons the adenosine A1 receptor antagonists cyclopentyltheophylline or DPCPX caused an increase in intracellular Ca2+, suggesting that adenosine was secreted by hypothalamic cells, tonically depressing glutamate-enhanced neuronal Ca2+. 7. To determine whether adenosine could exert a postsynaptic effect, we coapplied it with glutamate agonists in the presence of tetrodotoxin. Within subpopulations of hypothalamic neurons, adenosine and CPA either inhibited (18% of total neurons) or potentiated (6% of total neurons) responses to glutamate, N-methyl-D-aspartate, and kainate by > or = 20%. 8. In contrast to the modest effects found in neurons, responses of hypothalamic astrocytes to the application of glutamate or the metabotropic glutamate receptor agonist (+/-)-trans-1-amino-1,3-cyclopentanedicarboxylic acid were strongly potentiated by adenosine (mean +225%) and CPA. 9. Together, these findings suggest that adenosine exerts a major presynaptic effect and a minor postsynaptic effect in the modulation of glutamate neurotransmission in the hypothalamus, where it can play a significant role in blocking a large part of the glutamate-induced Ca2+ rise. In the absence of glutamate transmission, adenosine has relatively little effect on either neuronal intracellular Ca2+ or electrical activity.