(469e) Kir-Cholesterol Interactions: Molecular Simulations Reveal a Dynamic Ensemble of Lipid Ligands and a Composite Binding Domain of Asymmetric Concentration Dependence

Cholesterol is a major regulator of multiple types of ion channels, but the specific mechanisms through which cholesterol modulates channel function are not well understood. The cholesterol sensitivity of one family of ion channels, inwardly rectifying potassium (Kir) channels, has been shown to depend on direct interactions with “cholesterol-sensitive” regions on the channel protein. However, it is unclear whether these regions correspond to a well-localized binding site or are representative of a broad and shallow thermodynamic landscape. Furthermore, as cholesterol is abundant in plasma membranes, it is unclear why an increase in membrane cholesterol content – one precursor of cardiovascular disease – leads to pronounced channel inhibition. Using MARTINI coarse-grained molecular simulations of Kir2.2 channels we simulated the spontaneous and unbiased migration of cholesterol molecules from a POPC/cholesterol bilayer to the protein surface of Kir2.2 and thus quantitatively discriminated residues responsible for favorable interactions.

Unlike small drug-like molecules which tightly bind to a single site, we found that multiple cholesterol molecules concurrently interact with the entire transmembrane region of the Kir channel. These concurrent interactions form a complex milieu of short- and long-term contacts that do not readily segregate into distinct phenomena, instead comprising a continuous spread of contact times ranging from nanoseconds to microseconds. This presented an interesting data challenge: given this broad range of non-incidental contacts, and given that these contacts do not readily segregate spatially into distinct sites, how can binding events be identified? To tackle this problem, we utilized principles from network theory to identify temporally correlated contact residues and then applied unbiased learning approaches to segregate the resulting interaction patterns into distinct binding domains. We thus identified two discrete cholesterol binding sites within a previously identified cholesterol-sensitive region, with one site capable of binding two sterol molecules in distinct sub-sites. Furthermore, we also discovered that a 2-fold decrease in the cholesterol level of the membrane, a perturbation previously observed to increase Kir2 activity, results in a site-specific decrease of cholesterol occupancy. That is, cholesterol occupancy at one of these discrete sites remains consistent regardless of the membrane cholesterol content, while occupancy of the other site is highly concentration dependent.