Understanding the Mechanism of PAQR-2 through Modeling and Simulations

Abstract

PAQR-2 is a transmembrane protein in C. elegans with seven transmembrane helices and a zinc site. As a homolog to the human ADIPOR receptors, which are involved in regulating membrane fluidity and are of importance when studying glucose toxicity in diabetes patients, the mechanism of PAQR-2 holds many undiscovered secrets to understanding membrane fluidity regulation. PAQR-2 has been shown to be vital for survival in cold temperatures and in the presence of high glucose levels. Other fluidity sensitizing proteins have been shown to change conformation in different membrane environments. In this study, molecular dynamics simulations of PAQR-2 were done with the purpose of observing the structural response of PAQR-2 to different membrane environments. Both simulations of only the transmembrane domain and of the full protein were made. In addition, two loss of function mutants (d282n and g533r) were simulated and compared with the wild-type. Furthermore, the IGLR-2 protein that has been shown to be vital for the function of PAQR-2 was simulated and docked with PAQR-2 yeilding a likely structure for the PAQR-2:IGLR-2 complex. Simulations of PAQR-2 in a thick and ordered DPPE membrane revealed an adaptation of the membrane thickness to accommodate PAQR-2, rather than a structural change within the protein itself. The g533r mutation introduced novel interaction sites between the helices. The d282n mutation resulted in a loss of hydrogen bonds of the residue, which sits close to the zinc site. Protein-protein docking and PMF calculations using umbrella sampling revealed four possible PAQR-2:IGLR-2 complexes. The interactions of the highest scoring complex were analyzed and classified as being primarily weak interactions. The full protein model of PAQR-2 which includes both the transmembrane domain and the cytosolic domain, shows most promise as a model of PAQR-2, as it captures dynamics surrounding the zinc site predicted by the homolog model. Moreover, the full protein model describes different behaviour between wild-type and the d282n mutant not found in the model of only the transmembrane domain. Further optimization of the full protein is required as an un-physical loss of secondary structure occurs in the cytosolic domain when simulated close to the membrane.