Fragment Ligands of the m⁶A-RNA Reader YTHDF2

Abstract

We report 17 small-molecule ligands that compete with N6 -methyladenosine (m 6 A) for binding to the m 6 A-reader domain of YTHDF2 (YT521-B homology domain family 2). We determined their binding mode at high resolution by X-ray crystallography and quantified their affinity by a fluorescence-based binding assay. 6-Cyclopropyluracil and a pyrazolopyrimidine derivative have favorable ligand efficiencies of 0.47 and 0.38 kcal mol –1 per non-hydrogen atom, respectively. They represent useful starting points for hit optimization.

Atomic Mutagenesis of N6-Methyladenosine Reveals Distinct Recognition Modes by Human m6A Reader and Eraser Proteins

“…These experimental observations correlate with previously reported crystal structures of m 6 A bound to YTHDF2 and YTHDC1 (Figure ). , These structures show the m 6 A nucleobase located in an aromatic cage, which is stabilized by π–π-stacking with tryptophan side chains located parallel to the purine heterocycle (W432 and W491 in DF2, and W377 in DC1), and a methyl−π interaction with a tryptophan side chain that is located orthogonal to the N 6 -methyl group (W486 in DF2 and W428 in DC1). The RNA-protein complexes are further stabilized by hydrogen bonding interactions of the N1 nitrogen of m 6 A with a carboxyl group (D422) in YTHDF2, or with an amide (N367) in YTHDC1.…”

“…The design of chemical inhibitors requires knowledge of the specific interactions between RNA and proteins. Such insights can be derived from crystal structures in combination with molecular dynamics simulations , and are supported by biochemical data obtained with mutated proteins, in which key functional amino acids in the active site are exchanged, e.g., by replacement with alanine. Introducing smaller changes by exchanging single functional groups via incorporation of artificial amino acids is more challenging (e.g., via stop-codon suppression).…”

Pliability in the m6A-Binding Region Extends Druggability of YTH Domains

“…13−16 However, in crystallographic apo structures, only DF2 shows a correctly organized aromatic cage. 14,16,17 In apo DF1 and DF3, the β4−β5 loop is disordered, and the m 6 A pocket is largely distorted. 13,15 This observation suggests that while in DF2, the m 6 A-binding site could be preformed, in the other YTH domains, the pocket organizes itself around the ligand upon binding.…”

“…Moreover, a Tyr residue (Tyr397 in YTHDF1) closes the cage at the back of the bound m 6 A. Two of the above amino acids (Trp465 and Trp470 in DF1) are part of the extended β4–β5 loop, which assumes similar conformations in all YTH domains with bound m 6 A, methylated oligoribonucleotides, or small molecules mimicking the natural ligand. − However, in crystallographic apo structures, only DF2 shows a correctly organized aromatic cage. ,, In apo DF1 and DF3, the β4–β5 loop is disordered, and the m 6 A pocket is largely distorted. , This observation suggests that while in DF2, the m 6 A-binding site could be preformed, in the other YTH domains, the pocket organizes itself around the ligand upon binding. This is surprising, considering the high degree of sequence and structure conservation within the DF family.…”

Small-Molecule Inhibitors of the m7G-RNA Writer METTL1

“…The number of epitranscriptomic targets with an active program of drug discovery is very small. Inhibitors were disclosed for the m 6 A writer METTL3-METTL14 − and even more recent is the discovery of fragment binders of the m 6 A readers YTHDC1 and YTHDF2 . The innovative nature of the field and the role of METTL1 in oncogenesis make it a favorable target.…”