Structural determination of the mechanism of domain separation of G‐protein‐coupled receptor kinase 4g

Polymorphisms of G‐protein‐coupled receptor kinase 4γ (GRK4γ) are associated with human hypertension and induce hypertension in mice. Data suggest that three polymorphisms of GRK4γ (A142V, R65L, A486V) have increased activity compared to the wild‐type (wt) GRK4γ; however, the mechanism of activation of GRK4γ remains unknown. GRK5 and GRK6, which are members of the GRK4 family, are well studied and provide a model for GRK4. Electrostatic interactions hold together the regulators of G protein signaling homology (RH) domain and kinase domain (KD) of GRK5 in a closed (inactive) conformation, and the RH‐KD must separate (open) to display kinase activity. The opening of the RH‐KD may be a family‐wide biochemical mechanism. Molecular dynamics simulations (MD) of wt GRK4γ and A142V GRK4γ in a water water‐box indicate that GRK4γ shares similar electrostatic interactions as GRK5, but displays little spontaneous RH‐KD opening. When wt GRK4γ and A142V GRK4γ are simulated bound to a lipid membrane, domain separation statistically increases suggesting that GRK4γ is active when bound to a membrane matching initial biochemical characterization of GRK4. To investigate which GRK4γ residues interact with the membrane and the mechanism of membrane‐induced RH‐KD separation a frequency‐matrix of each residue's interaction with the membrane was clustered by K‐means and HDBSCAN. Both clustering techniques indicate that A142V GRK4γ samples a more diverse protein‐lipid interaction surface and the clusters consolidate into three different groups based on RH‐KD separation. As HDBSCAN does not force samples into a cluster, our analysis focused on the HDBSCAN clusters. The mean RH‐KD separation over 4.5 μs of simulation is 14.1% ± 11%. The large deviation is due to one cluster opening 31.4% of the time while another only 0.8% of the time; the rest of the clusters are near the mean. wt GRK4γ populates a unique and a shared cluster that is open 12.1% and 13.5% of the time, respectively. A142V GRK4γ populates three unique clusters and the shared cluster that open 12.7%, 31.4%, 0.8%, and 13.5% of the time, respectively. The 31.4% and 0.8% clusters have unique frequency‐matrixes suggesting that altered lipid interactions drive RH‐KD separation. Specifically, greater open probability associates with additional KD contacts with the membrane compared to the three average clusters, and loss of common interactions at the termini results in depressed open probability. The interaction data suggests that for GRK4γ to undergo domain separation it is more favorable to have the kinase domain anchored to the membrane. The KD attachment to the membrane allows for a hinge predicted to lie between the RH domain and KD to flex. Understanding the mechanism underlying GRK4γ domain separation allows for testing if domain separation is required for receptor phosphorylation and desensitization. Additionally, MD identified changes in orientation at the membrane resulting in increased RH‐KD separation provides a plausible and testable hypothesis to explain the observed increased activity of A142V GRK4γ. Support or Funding Information Support for this work is from WesternU MSPS program funds. This abstract is from the Experimental Biology 2019 Meeting. There is no full text article associated with this abstract published in The FASEB Journal .