Development of selective Dyrk1a Inhibitors for Alzheimer's disease - PROJECT SUMMARY Alzheimer’s disease (AD) is a devastating neurodegenerative disease with limited treatment options. Most treatment approaches attack a single disease component. However, pleiotropic interventions will likely be needed and the dual-specificity tyrosine phosphorylation-regulated kinase 1A (Dyrk1a) protein represents one such opportunity, with activity in β-amyloid, tau, and neuroinflammatory pathways (E. Deboever et al., I. J. Mol. Sci., 2022). Thus, compounds that inhibit Dyrk1a offer a promising therapeutic approach through modulation of tau, amyloid, and neuroinflammatory pathways. To this end, recent efforts delivered a potent and brain penetrant molecule, Dyr533, which upon dosing in the 3xTg-AD model, improves behavioral and both tau and amyloid neuropathological features of AD, whilst lowering levels of the pro-inflammatory cytokine TNF in plasma and brain, all in a dose-dependent manner (Bartholomew et. al., Alzheimer’s & Dementia, 2023, 19, e076499). Similar results were observed in a PS19 tauopathy model of AD and Ts65Dn down syndrome model. We have also confirmed the original in vivo POC inhibitor Dyr219 (Branca et al., Aging Cell, 2017, Velazquez et. al., 2019, Mol. Neurobiology) and Dyr533 trigger Dyrk1a protein degradation, assigned to inhibition of the Dyrk1a autophosphorylation process. This effect reduces levels of active Dyrk1a in vivo, contributing to efficacy and pointing to a PK-PD relationship between Dyr533 brain exposure and robust neuropathological effects. However, mechanism-based safety studies revealed a significant liability (Pde3a IC50 330nM) underpinning MTD to lethality@35mg/kg through cardiovascular toxicity. Pde3a activity promotes inhibition of the coagulation cascade and induces arrhythmias. The therapeutic window for Dyr533 is also likely weakened by a sub-par Kpuu (0.1) and efflux (MDR1 & BCRP >> 3). This renewal proposes development of a second-generation inhibitor that recapitulates the efficacy of Dyr533 albeit with an improved therapeutic window by: 1. removal of Pde3a activity. 2. improvement of CNS exposure (Kpuu > 0.3) and efflux reduction (MDR1/BCRP < 3.0). 3. improvement of Dyrk1a affinity (KD < 1nM). These efforts will be conducted within the current lead series and relatively unexplored series (mono-cycles), supported by modeling and crystallography. Recent breakthrough co-crystallography (see Dyr574 & 518:Dyrk1a complexes) suggests lower MW targets, which maintain ‘Dyr533-like’ selectivity and negate Pde3a activity, are within reach to address therapeutic index limiting brain penetrance. In summary, with a modified chemistry plan, a reformatted screening paradigm to include a one-month in vivo study and an expanded group of collaborators, we aim to deliver an orally bioavailable, brain penetrant and selective ‘Dyr533 analog’ with an enhanced therapeutic window, deemed safe after mechanism-based safety studies, that recapitulates its’ in vivo effects in AD rodent models.
