Tampere University of Technology

TUTCRIS Research Portal

Protocol to Avoid Possible Artifacts in Atomistic Simulation of GPCR Proteins whose Crystal Structure is Heavily Engineered

Research output: Contribution to journalArticleScientific


Original languageEnglish
Pages (from-to)59a
Number of pages1
JournalBiophysical Journal
Issue number3, Supplement 1
Publication statusPublished - 16 Feb 2016
Publication typeB1 Article in a scientific magazine


G protein-coupled receptors (GPCRs) are versatile signaling proteins that mediate diverse cellular responses. Atomistic molecular dynamics (MD) simulations are widely used to elucidate the properties of GPCRs. These simulations are based on three-dimensional protein structures that in turn are often based on crystallography. However, to facilitate structure determination, typically the crystallized proteins are heavily engineered, including structural modifications (mutations, replacement of protein sequences by antibodies, bound ligands, etc.) whose impact on protein structure and dynamics is largely unknown. Here we address this issue through atomistic MD simulations, focusing on the β2-adrenergic receptor (β2AR), a well-characterized GPCR. Starting from an inactive-state crystal structure of β2AR, we reverted the numerous structural modifications done on β2AR in multiple consecutive steps, one at a time, each followed by extensive equilibration in a lipid membrane. The systematic step-by-step approach provides results that are superior in terms of maintaining protein structural stability, as compared to the usual and computationally less expensive approach of removing all modifications instantaneously at once. Another great advantage of the step-wise method is that it clarifies the effects of individual crystallization modifications on the native properties of the receptor, such as on the dynamics of the ligand and the G-protein binding sites and the packing at the transmembrane helix interface of β2AR in the present case. Our results emphasize that the preparation of membrane protein structure for atomistic MD simulations is a very delicate and sensitive process and can lead to significant artifacts if done in a straightforward manner by cutting corners. We consider that the protocol described here (step-by-step approach) may offer a useful strategy for simulating a variety of native state GPCRs, whose crystal structures suffer from similar structural engineering.

Field of science, Statistics Finland