Identification Of The C-Linker And Cnbd Residues Accounting For The High Efficacy Of Camp Activation In Hcn2 Channels

Hyperpolarization-activated, Cyclic Nucleotide–gated (HCN) channels are major determinants of the firing rate of pacemaker centers in the heart and brain. Direct binding of cAMP to the cyclic nucleotide binding domain (CNBD) activates HCN channels by increasing the maximum open probability and shifting the voltage-dependence of activation to less hyperpolarized potentials. HCN isoforms differ greatly in kinetics, voltage gating, and response to cyclic nucleotides. HCN2 responds strongly to cAMP despite having similar ligand affinity to HCN1, an isoform that responds very poorly to cAMP. To identify the key molecular determinants responsible for cAMP activation, we progressively mutated sites in the C-linker and CNBD of the mHCN2 to HCN1. Our studies identified two clusters of mutations that determine the differences in voltage dependent activation between these two isoforms. One mapped to the C-linker region (M485F, G497D, and S514T), far from the cAMP-binding site. Another was found in proximity to the binding site (V562A/S563G, L565I, and S575T). Concurrent substitutions of five sites (M485F, G497D, S514T, and V562A/S563G) is sufficient to confer HCN1 phenotype on the mutated HCN2 channels. To understand the role of these residues on various allosteric parameters that determine the gating of these channels, we built an allosteric model consisting of four separate modules: pore gates, four voltage-sensors, ligand binding domain, and C-linker. Our models suggest that these substitutions must alter at least the oligomerization state of the linker and two coupling parameters: one between the linker and the pore and the other between the linker and the voltage sensors. These findings will be discussed in light of the high-resolution structures of HCN channels.