Lipid Chemistry and Mechanical State of the Membrane Modulate Ion Channel Function

My research focused on voltage-dependent K+ (Kv) channels. Kv channels serve many di erent functions in di erent cells, but most notably underlie action potentials in electrically excitable cells, such as neurons and muscle (Hodgkin and Huxley, 1952, 1945). Kv channel gating is governed by the transmembrane voltage, they are therefore voltage-dependent switches for ionic current (Hille, 2001). Changes in the transmembrane voltage are sensed by the channel\u27s voltage sensor domains, which contain charged amino acids (most often arginines) called gating charges. Shortly before I started to work on my PhD project, the crystal structure of the eukaryotic Kv1.2 channel had been solved. This structure reinforced the idea that the voltage sensors are arranged as independent domains at the perimeter of a Kv channel facing the lipid membrane, thus exposing some of the gating charges to the lipid. The obvious question to ask at that time was, given the energetic penalty for placing charged amino acids inside the hydrophobic core of the membrane, how does the lipid membrane stabilize the arginine residues? By studying the recombinantlyexpressed arch al Kv channel KvAP in an arti cial membrane system that allowed me to create a de ned lipid environment, I could show that the lipid membrane provides an environment that is suitable for voltage sensors because the lipid\u27s phosphate groups serve as countercharges for the voltage sensor\u27s arginine residues. I came to the conclusion that a direct interaction between the arginine side chains and lipid phosphodiesters stabilizes the voltage sensor through multidentate hydrogen bonding. I suggested that the usage of positively charged amino acids in voltage sensors is an adaptation to the phospholipid composition of the cell membrane. Prompted by these results, I studied the gating properties of KvAP in di erent lipid systems and was able to derive the rst quantitative kinetic gating model for KvAP. I found that, unlike the well studied eukaryotic Shaker Kv channel, KvAP possesses an inactivated state that is accessible from the pre-open state of the channel. Changing the lipid composition of the membrane in uences multiple gating transitions in the model, but most dramatically the rate of recovery from this inactivated state. I also showed that inhibition by the spider toxin VSTx1 is most easily explained if VSTx1 binds only to the depolarized conformation of the voltage sensor. By delaying the voltage sensor\u27s return to the hyperpolarized conformation VSTx1 favors the inactivated state of KvAP. Aside from varying the chemical composition, I also studied how the mechanical state of lipid membranes in uences Kv channel gating. I found that Kv channels are mechanosensitive proteins and that a model in which membrane tension in uences a single parameter (the equilibrium constant governing pore-opening after the voltage sensors have moved) can account quantitatively for complex changes in voltagedependent gating, that are caused by the formation of tight lipid/glass seal in patch clamp recordings. The mechanical state of the membrane also governs the apparent a nity of spider toxins for Kv channels. This unexpected relationship between voltage sensor toxin a nity and the mechanical state of the membrane suggests that the toxin modi es the membrane mechanical forces experienced by the Kv channel. In summary, my thesis research describes how both the chemical and mechanical properties of lipid membranes regulate Kv channel function and pharmacology. These results demonstrate that the lipid membrane is not solely a passive solvent for membrane proteins, but that its composition and structure might be considered a source for functional diversity, enabling a membrane protein\u27s function to be tuned to the requirements of a particular cell type