Mapping partial agonism of mitragynine at the µ-opioid receptor through molecular dynamics and Markov state modelling analysis

Abstract

Mitragynine, a major indole alkaloid from Mitragyna speciosa (kratom), acts as a partial agonist at the µ-opioid receptor (µOR), yet the structural basis for its submaximal efficacy remains unclear. Here, we integrate microsecond-scale all-atom molecular dynamics (MD) simulations with Markov State Modelling (MSM) to probe how mitragynine modulates µOR conformational landscapes and kinetics versus a morphine-bound control. MD in explicit POPC bilayers quantified backbone stability, residue-level flexibility, global compactness, and hydrogen bonding. MMPBSA calculations indicated favourable binding for both ligands, with a more negative ΔG_bind for mitragynine (–16.3 ± 5.1 kcal mol⁻¹) than morphine (–10.8 ± 4.2 kcal mol⁻¹). MSMs built on TM3–TM6 separations, DRY/NPxxY χ₁ torsions, and ICL distances revealed distinct energy landscapes: morphine stabilised a deep active-like basin, whereas mitragynine broadened sampling of intermediate basins and reduced occupancy of fully active conformations. Kinetic analysis showed shorter intermediate→open transition times for morphine (hundreds of nanoseconds) but markedly longer, microsecond-scale transitions for mitragynine, yielding macrostate populations enriched in intermediates for mitragynine and in closed/open states for morphine. Together, these data provide a mechanistic explanation for mitragynine’s partial, G-protein-biased agonism at µOR and a quantitative framework to guide the design of biased µOR ligands.