Computational design and molecular modeling of morphine derivatives for preferential binding in inflamed tissue

The opioid epidemic has impacted over 10 million Americans in 2019. Opioids, like morphine, bind non-selectively in both peripheral tissue, leading to effective pain relief, and central tissue, resulting in dangerous side effects and addiction. The inflamed conditions of injured tissues have a lower pH (pH = 6-6.5) environment than healthy tissue (pH = 7.4). We aim to design a morphine derivative that binds selectively within inflamed tissue using molecular extension and dissection techniques. Morphine binds to the μ-opioid receptor (MOR) when the biochemically active amine group is protonated. Fluorination of a β-carbon from the tertiary amine group led to a reduced pKa of the derivative through induction. Through a decrease in the pKa, protonation is still statistically favored in lower pH environments of inflamed tissue but primarily deprotonated in healthy tissue. The cyclohexenol and N-methyl-piperidine rings of morphine are removed to increase conformational flexibility when binding while still maintaining the interactions required for analgesia. Electronic structure calculations were performed with Gaussian16 using the Keck Computational Research Cluster at Chapman University to determine the pKa. The theoretical pKa values are determined at the M06-2X(SMD)/aug-cc-pVDZ level of theory to calculate the ΔG°aq values for the amine deprotonation reactions. Fluoromorphine β-C2 was designed computationally and modeled within the MOR using Maestro: Schrödinger. This derivative exhibits a pKa reduction and enhanced ligand-protein interactions within the MOR. β-fluorination decreased the overall pKa values of the morphine derivatives (pKa: 6.1-7.83) relative to morphine, reducing binding within healthy, central tissue.