Abstract
Models of the human μ opioid receptor were constructed using available G-protein-coupled receptor (GPCR) structures using homology (comparative) modeling techniques. The recent publication of a high-resolution crystal structure of a construct based on the murine μ opioid receptor offers a unique opportunity to evaluate the reliability of the homology models and test the relevance of introducing more templates (known structures) to increase the accuracy of the comparative models. In the first model two templates were used: the β(2) adrenergic and bovine rhodopsin receptors. For the second model, four templates were utilized: the β(2) adrenergic, bovine rhodopsin, β(1) adrenergic, and A(2A) adenosine receptors. Including additional templates improved the accuracy of structural motifs and other features of the model when the same sequence alignment was used. The predicted structures were especially relevant in the case of important receptor regions such as the DRY motif, which has been associated with receptor activation. Additionally, this study showed that receptor sequence similarity is crucial in homology modeling, as indicated in the case of the highly diverse EC2 loop. This study demonstrates the reliability of the homology modeling technique in the case of the μ opioid receptor, a member of the rhodopsin-like family class of GPCRs. The addition of more templates improved the accuracy of the model. The findings regarding the modeling has significant implication to other GPCRs where the crystal structure is still unknown and suggest that homology modeling techniques can provide high quality structural models for interpreting experimental findings and formulating structurally based hypotheses regarding the activity of these important receptors.