Cyclic nucleotide-gated (CNG) channels of olfactory neurons are heterotetramers composed of three homologue subunits, CNGA2, CNGA4, CNGB1b, in a stoichiometry 2:1:1. CNGA4 and CNGB1b subunits produce functional channels only in heterotetramers with CNGA2. Therefore, the contribution of each of these subunits to channel activation has not been elucidated so far. Up to now it is even not known whether the ligand binding to the CNGA2 subunits alone or also to the CNGA4 and CNGB1b subunits leads to the opening of the olfactory channels. The fact that the Hill-coefficient for activation of heterotetrameric channels is close to 1 whereas it is near 2 in homotetrameric channels might be interpreted in the sense that not all subunits bind ligands.

Activation of heterotetrameric CNGA2A4B1b channels was studied in excised patches by measuring ligand binding and gating under both steady-state and non-steady state conditions. Ligand binding was measured by patch-clamp fluorometry using a fluorescent cGMP analogue 8-DY547-cGMP (fcGMP). Similar to the homotetrameric CNGA2 channels (Biskup et al., Nature 446, 440–443, 2007), we observed also in heterotetrameric channels a crossover of steady-state binding and steady-state activation. Moreover, the binding curve of the CNGA2A4B1b channels closely approximates the binding curve of the homotetrameric CNGA2 channels. The fluorescence estimated for a heterotetrameric channel was less than half of that estimated for a homotetrameric channel.

In an approach to study the binding to CNGA4 and CNGB1b subunits only, we coexpressed either CNGA4 or CNGB1b with a mutated CNGA2 subunit (R538E; c.f. Tibbs et al., J. Biol. Chem. 273, 4497–505, 1998) that has a ~300 times decreased sensitivity to cGMP. Using patch-clamp fluorometry with fcGMP up to 20 mM, no significant binding was observed to the CNGA4 and the CNGB1b subunits. In order to gain additional information, activation time courses following [cGMP] jumps were considered. Activation of CNGA2A4B1b channels was up to 10 times slower than that of homotetrameric CNGA2 channels over a wide concentration range.

Our results support the notion that heterotetrameric CNGA2A4B1b channels are activated by the binding of the cGMP to the two CNGA2 subunits only. Hence, the CNGA4 and CNGB1b subunits play only modulatory roles, e.g. to slow cGMP-induced activation in the heteromultimer.