Conception of DYRK1A kinase inhibitors via metal-catalyzed C–H arylation, inspired by fragment-growing studies

The search for therapeutic inhibitors of specific kinases has been developed in the last three decades as a major approach to discover new drugs . Our group is focused on the regulation of dual-specificity tyrosine phosphorylation-regulated kinase 1A (DYRK1A), a conserved eukaryotic kinase that belongs to the DYRK family and the CMGC group, which includes cyclin-dependent kinases (CDKs), mitogen-activated protein kinases (MAP kinases), glycogen synthase kinases (GSK), and Ccd2-like kinases (CLKs). Five years ago, a series of tricyclic aminopyrimidine derivatives was synthesized and evaluated on DYRK1A and DYRK1B. 
A fragment-growing approach was performed using a novel in silico tool that drills down through, to evaluate hundreds of thousands fragments extracted from co-crystallized kinase/inhibitor complexes. Addition of aromatic fragments on C2 seemed to increase the interaction with the hinge region. 
Efficient metal catalyzed C–H arylation of 8-alkyl-thiazolo[5,4-f]-quinazolin-9-ones was explored for SAR studies. Application of this powerful chemical tool at the last stage of the synthesis of kinase inhibitors allowed the synthesis of arrays of molecules inspired by fragment-growing studies generated by molecular modeling calculations. Among the potentially active compounds designed through this strategy, FC162 (Cc) exhibits nanomolar IC50 values against some kinases, and is the best candidate for development as a DYRK kinase inhibitor.


Introduction
Kinases catalyse protein phosphorylation, a key cellular regulatory mechanism, which is frequently dysregulated in human diseases. Protein kinases have consequently been linked to the progress of a variety of diseases including cancer and neurodegenerative disorders. Therefore, the search for therapeutic inhibitors of specific kinases has been developed in the last three decades as a major approach to discover new drugs [1,2].
Our group is focused on the regulation of dual-specificity tyrosine phosphorylationregulated kinase 1A (DYRK1A), a conserved eukaryotic kinase that belongs to the DYRK family and the CMGC group, which includes cyclin-dependent kinases (CDKs), mitogen-activated protein kinases (MAP kinases), glycogen synthase kinases (GSK), and Ccd2-like kinases (CLKs) [3] Introduction Introduction Introduction A fragment-growing approach was performed using a novel in silico tool that drills down through, to evaluate hundreds of thousands fragments extracted from co-crystallized kinase/inhibitor complexes. Addition of aromatic fragments on C2 seemed to increase the interaction with the hinge region. A library of novel C2-arylated N8-alkyl thiazolo [5,4f]quinazolin-9(8H)-ones was envisioned by addition of (hetero)-aromatic fragments.
In silico fragmentgrowing calculations Introduction 8 As a result in our recent experience in carbon-carbon bond formation [8-10], a regioselective C-H bond activation was performed to provide corresponding C2-arylated valuable compounds. Most of the syntheses were achieved under microwave irradiation as a powerful alternative to traditional heating with economic and environmental benefits. The inhibitory potency of N8-benzylated thiazolo[5,4-f]quinazolin-9(8H)-ones obtained was evaluated according to standard methods [11,12] on a panel of kinases (for details see kinase profiling paragraph). Among the tested molecules, only two (A and B) exhibited micromolar IC 50 values against kinases CLK1 and GSK3, and nanomolar range inhibition against DYRK1A . Compound A was the most active. Taking these preliminary results into account, a new series C was designed by keeping the 3-pyridinyl moiety in position C2, and modifying the alkyl substituents in position N8 of the thiazolo[5,4f]quinazolin-9(8H)-ones. Retrosynthetic route of series C products using compound 1 as intermediate. All compounds were first tested at a final concentration of 10 μM. Compounds showing less than 50% inhibition were considered as inactive (IC 50 >10 μM). Compounds displaying more than 50% inhibition at 10 μM were next tested over a wide range of concentrations (usually from 0.01 to 10 μM), and IC 50 values were determined from the dose-response curves (Sigma-Plot). Harmine, a b-carboline alkaloid known to inhibit DYRK1A , was used as a positive control. Docking calculations were next performed in order to predict the molecular interactions of FC162 with DYRK1A. Two main binding modes were obtained. Left, first predicted binding mode (green), with the same orientation of the skeleton, but slightly shifted. Right, second predicted binding mode, in which the skeleton is flipped compared to its initial placement (in brown).
The docking score of the two poses was quite similar, thus, both binding modes are equally possible for this compound.
The SAR study revealed that FC162 with a 3-pyridinyl group in position 2 had a higher activity than the series of phenylated derivatives A and B. These results are notably in agreement with the fragment-growing experiments, which suggested replacement of the imidate group by a more stable heteroaromatic substituent.
This work demonstrates the efficacy of synthetic methodologies, such as C-H arylation of arenes and hetero-arenes for SAR studies. The application of this powerful tool at the last stage of the synthesis of kinase inhibitors allowed the synthesis of arrays of molecules inspired by fragment-growing studies generated by molecular modeling calculations. Among the potential active compounds generated through this strategy, FC162 (Cc) was found to be the best candidate for development as a DYRK inhibitor.