Roseti C, Palma E, Martinello K, Fucile S, Morace R, Esposito V, Cantore G, Arcella A, Giangaspero F, Aronica E, Mascia A, Di Gennaro G, Quarato PP, Manfredi M, Cristalli G, Lambertucci C, Marucci G, Volpini R, Limatola C, Eusebi F. (2009) Blockage of A2A and A3 adenosine receptors decreases the desensitisation of human GABA(A) receptors microtransplanted to Xenopus oocytes. Proc Natl Acad Sci U S A. 106:15927–1531
Roseti C, Martinello K, Fucile S, Piccari V, Mascia A, Di Gennaro G, Quarato PP, Manfredi M, Esposito V, Cantore G, Arcella A, Simonato M, Fredholm BB, Limatola C, Miledi R, Eusebi F. (2008) Adenosine receptor antagonists alter the stability of human epileptic GABA A receptors. Proc Natl Acad Sci U S A. 105:15118–15123
In these two tightly related papers, the authors analysed the influence of adenosine upon human GABAA receptor instability, assessed by the rundown of the GABAA currents (IGABA) that occurs upon repeated application of GABA. IGABA were recorded from oocytes injected with membranes extracted from human epileptic brain, including pharmacoresistant mesial temporal lobe epilepsy, focal cortical dysplasia and periglioma epileptic tissues. For comparison, data were also obtained from epileptic slices obtained from neurosurgical resection in patients suffering from some of the aforementioned diseases, as well as from cortical slices of pilocarpine-treated epileptic rats and A1 or A2A receptor-deficient mice.
Rundown (desensitisation) of GABAergic currents is pro-excitatory in mature neurons since it reduces the efficacy of the GABAergic signal. Accordingly, exacerbated IGABA rundown together with slow recovery is characteristic of epileptic tissue. The well-established antiepileptic action of adenosine could allow the anticipation that if any, adenosine would attenuate IGABA rundown and that A1 receptors would be involved. Surprisingly and, in my view, refreshing in the field, it was found that adenosine A1 receptors are not involved in the effect of this endogenous purine receptor agonist. Indeed, the most consistent finding throughout the two papers is that adenosine A1 receptors do not modulate IGABA rundown, irrespectively of the model (human vs. rats or mice, type of epilepsy, injected oocytes vs. brain slices) or the approach (pharmacological vs. receptor knockout) used. All sets of data are fully consistent in what concerns the absence of modulation by A1 receptors of GABAA receptor instability.
Contrasting with the absence of effect of a selective A1 receptor antagonist, a blanket adenosine receptor antagonist and adenosine deaminase reduced IGABA rundown, indicating that adenosine present in the extracellular space is exacerbating GABAA receptor desensitisation, therefore highlighting a putative proconvulsant action of adenosine. Whilst trying to dissect out the types of receptors involved, no clear-cut conclusions could be drawn. The most striking case was with A2A receptor antagonists or A2A knockout mice. Indeed, not all A2A receptor antagonists tested had effects in all models used, and when they did, opposite actions were often observed in subsets of cell populations. No clear association with disease type or experimental model could be identified to explain discrepant results. Within the most extensively used model, injected oocytes, the same A2A receptor antagonist had opposite effects in different cell subsets. In a majority of cells and models tested, A2A receptor antagonism reduced IGABA rundown, suggesting that A2A receptor activation by endogenous adenosine exacerbates GABAA receptor desensitisation. This could be expected since receptor desensitisation often occurs through phosphorylation (frequently mediated by cAMP and PKA, the well-known transduction pathways of A2A receptors). However, in a minority of cells, the selective A2A receptor antagonists exacerbated IGABA rundown and, surprisingly, data from A2A receptor knockout mice corroborated the results obtained in this minority of cells—IGABA rundown was more pronounced in mice lacking A2A receptors. The A3 receptor antagonists mostly reduced the IGABA rundown but increased it if tissue was from dysplasic neocortices. The A2B receptor was not extensively studied, but its antagonism also reduced IGABA rundown.
Summarising, in what concerns a relevant dysfunctional characteristic of epileptic tissue, GABAA receptor instability, adenosine tonically exacerbated it. For sure, this action does not involve A1 receptors. It may, however, involve any other adenosine receptor. The trend is that selective antagonists of non-A1 receptors attenuate GABAA receptor desensitisation. One should, however, keep in mind that the same antagonist may have opposite effects in different cell populations within the same pathological model. Furthermore, it still remains to be understood why data from A2A receptor knockout mice and data obtained with the A2A receptor antagonists in the majority of cells lead to opposite conclusions. Indeed, data from A2A receptor knockout mice would fit to the idea that A2A receptors could have an anticonvulsant role. Irrespectively of the adenosine receptor subtype(s) involved in the modulation of GABAA receptor desensitisation in epileptic tissue and the mechanism(s) operated by them, these two papers highlight the need of thinking beyond A1 receptors whilst aiming at a full understanding of the role of adenosine in epilepsy.
About the author
Ana M. Sebastião is at present Associate Professor at the Faculty of Medicine, University of Lisbon. Her research interests relate to purines since 1983. She has been interested in the characterisation of adenosine receptors as well as in the role played by endogenous adenosine in the nervous system, especially at the neuromuscular junction and hippocampus. More recently, she has been concentrating on the mechanisms by which adenosine receptors interact with other neuromodulatory systems, including neuropeptides, cannabinoids and neurotrophic factors.
About this article
Cite this article
Sebastião, A.M. Adenosine and epilepsy—thinking beyond A1 receptors. Purinergic Signalling 6, 1–2 (2010). https://doi.org/10.1007/s11302-010-9179-6