Characterization of the functional role of alkaline phosphatase in the control of synaptic transmission through adenosine synthesis
Résumé
Tissue non-specific alkaline phosphatase (TNAP) is an extracellular enzyme that removes phosphate groups from a variety of substrates. It has previously been shown that TNAP activity is located on the neuronal membrane, especially in the synaptic cleft and on the node of Ranvier. Nevertheless TNAP function in the brain remains poorly understood. This enzyme might be involved in adenosine synthesis through the hydrolysis of extracellular ATP, ADP and AMP. It is known that adenosine inhibits synaptic transmission in cortex through A1 presynaptic receptors. Here, we examined the potential role of TNAP in extracellular adenosine synthesis by using metabolomic, histochemical and electrophysiological approaches. The metabolomic approach was used to compare wild-type mice and TNAP-Knockout mice. These experiments demonstrated that, in addition to GABA level, adenosine level was reduced in the brain of TNAP-/- and TNAP+/- mice. These results corroborate that TNAP is involved in adenosine synthesis in the mouse brain. The electrophysiological approach examined synaptic transmission and presynaptic inhibition in mouse piriform cortex (LFP recording in layer Ia). Our aim was to determine whether TNAP is involved in the synthesis of adenosine, which in turn is involved in presynaptic inhibition. We first verified that AMP and adenosine exerted an inhibition of postsynaptic responses and we confirmed that this inhibition was mediated through adenosine A1 receptors. We next tested whether AMP to adenosine conversion depended on TNAP by using MLS-0038949, a selective TNAP inhibitor. Our expectation was that AMP inhibitory effect should be suppressed in the presence of MLS-0038949. However, our results did not confirm this expectation. Yet our histochemical experiments confirmed that MLS-0038949 was effective at inhibiting TNAP activity in cortex. We then supposed that another enzyme was involved in the conversion of AMP into adenosine: the ecto-5’-nucleotidase (NT5E or CD73). We therefore tested an NT5E inhibitor (α,β-methylene ADP), alone or in combination with MLS-0038949. Again, we did not see abolition of AMP effect on presynaptic inhibition. These results led us to hypothesize that AMP could act directly on A1 receptors and would be an agonist of these receptors.