The effect of extracellular ionic contents on the Ca²⁺-impermeable AMPA receptors of hippocampal pyramidal neurons has been studied using the patch-clamp technique. Neurons were isolated from rats decapitated under urethane anaesthesia. Divalent cations produced fast concentration-dependent inhibition of currents evoked by 100 µM kainate. The IC₅₀ (mean ± S.D.) values were: Ni²⁺, 0.42 ± 0.02; Zn²⁺, 1.1 ± 0.2; Co²⁺, 6.1 ± 0.1; Ca²⁺, 8.1 ± 0.3; Mn²⁺ > 20 and Mg²⁺ > 40 mM. The blocking effects were voltage-independent and can be prevented by 100 µM of cyclothiazide, which allosterically potentiates AMPA receptors. This suggests that divalent cations modulate activation or desensitization properties of the AMPA receptor rather than affect the ion channel. Acidification of the external media produces similar effects (Ihle & Patneau, 2000); the IC₅₀ value for proton inhibition of AMPA receptors in our experiments was 0.01 ± 0.006 (pH₅₀ = 6.2). The action of Ni²⁺ and Ca²⁺ at pH = 6.0 and pH = 7.3 have been compared (Fig. 1). Significant decrease of the slope of concentration- response curve at lower pH suggests that inhibition by protons and divalent cations is interrelated. We propose cooperative action of divalent ions and protons at the homologous sites.

Divalent cations induce a parallel shift of the kainate concentration-response curve; at high kainate concentrations (3 mM) IC₅₀ values of divalent cations become much higher. Therefore, cyclothiazide, which is known to increase affinity of agonists to AMPA receptors (Partin et al. 1996), is able to prevent inhibition by inorganic cations. Presently it is not clear how competitive-like binding of kainic acid and inorganic cations occurs. Possibly, cations have specific binding sites in the cleft between agonist-binding domains of the AMPA receptor, where binding sites for agonists and classical competitive antagonists are located (Armstrong & Gouaux, 2000).

These results show that activation of the AMPA receptor is not determined by intrinsic features of the receptor only, but depend on the ionic environment. It is unlikely that variations of extracellular ionic content required for significant change of AMPA receptor properties occur under physiological conditions. However, in some experiments the ionic content is substantially changed, and possible effect of such substitutions on AMPA receptors should be taken into account. It would also be interesting to compare determinants and mechanism of action of cations on AMPA and NMDA receptors, for which modulations by Zn²⁺ and protons is thought to be of physiological importance.