Cross-interaction of tau PET tracers with monoamine oxidase B: evidence from in silico modelling and in vivo imaging.

Abstract

Several tracers have been designed for tracking the abnormal accumulation of tau pathology in vivo. Recently, concerns have been raised about the sources of off-target binding for these tracers; inconclusive data propose binding for some tracers to monoamine oxidase B (MAO-B). Molecular docking and dynamics simulations were used to estimate the affinity and free energy for the binding of several tau tracers (FDDNP, THK523, THK5105, THK5317, THK5351, T807 [aka AV-1451, flortaucipir], T808, PBB3, RO-948, MK-6240, JNJ-311 and PI-2620) to MAO-B. These values were then compared with those for safinamide (MAO-B inhibitor). PET imaging was used with the tau tracer [¹⁸F]THK5317 and the MAO-B tracer [¹¹C]DED in five patients with Alzheimer's disease to investigate the MAO-B binding component of this first generation tau tracer in vivo. The computational modelling studies identified a binding site for all the tau tracers on MAO-B; this was the same site as that for safinamide. The binding affinity and free energy of binding for the tau tracers to MAO-B was substantial and in a similar range to those for safinamide. The most recently developed tau tracers MK-6240, JNJ-311 and PI-2620 appeared, in silico, to have the lowest relative affinity for MAO-B. The in vivo investigations found that the regional distribution of binding for [¹⁸F]THK5317 was different from that for [¹¹C]DED, although areas of suspected off-target [¹⁸F]THK5317 binding were detected. The binding relationship between [¹⁸F]THK5317 and [¹¹C]DED depended on the availability of the MAO-B enzyme. The developed tau tracers show in silico and in vivo evidence of cross-interaction with MAO-B; the MAO-B component of the tracer binding was dependent on the regional concentration of the enzyme.