The single-channel dose-response relation is consistently steep for rod cyclic nucleotide-gated channels: implications for the interpretation of macroscopic dose-response relations.

Cyclic nucleotide-gated channels contain four subunits, each with a C-terminal binding site for cGMP or cAMP. The dose-response relation for activation is usually fit with the Hill equation, I/I(max) = [cGMP]n/([cGMP]n + K(1/2)n, where I/I(max) is the fraction of maximal current, K(1/2) is the concentration of cGMP that gives a half-maximal current, and n is the Hill coefficient, taken as the minimum number of ligands required for significant activation. The dose-response relations in multichannel patches are often fit with Hill coefficients of </=2.0, even though other lines of evidence indicate that these channels contain four binding sites and that the binding of three or four ligands is required for significant opening. We have measured dose-response relations for a large number of single cyclic nucleotide-gated channels composed of the bovine rod alpha subunit. We find that the single-channel Hill coefficient is consistently higher than 2.5, with an average of 3.0 +/- 0.37 over 16 patches. In both multichannel and single-channel patches, large variations in K(1/2) have been observed, and are thought to arise from modifications such as phosphorylation. Here we show that mixtures of single channels with high Hill coefficients and variable K(1/2) values will give rise to shallow macroscopic dose-response relations with anomalously low Hill coefficients. This is because activation occurs over a broad range of cGMP concentrations. Thus, dose-response relations from multichannel patches should be interpreted with caution, particularly when detailed mechanistic issues such as cooperativity are being investigated.