Small-molecule modulators of TRMT2A decrease PolyQ aggregation and PolyQ-induced cell death

Abstract

Polyglutamine (polyQ) diseases, including Huntington's disease, are characterized by an expansion of cytosine-adenine-guanine (CAG) trinucleotide repeats encoding for an uninterrupted prolonged polyQ tract. We previously identified TRMT2A as a strong modifier of polyQ-induced toxicity in an unbiased large-scale screen in Drosophila melanogaster. RNAi-mediated silencing of TRMT2A ameliorated polyQ-induced toxicity and polyQ aggregation in flies. This work aimed at identifying and validating pharmacological TRMT2A inhibitors as treatment opportunities for polyQ diseases. An in silico structure- and ligand-based lead discovery approach and computer-aided drug discovery (CADD) was used to identify TRMT2A inhibitors. Additionally, the crystal structure of one protein domain, the RNA recognition motif (RRM), was determined and Biacore experiments with the RRM were performed. The identified inhibitors were functionally validated for their potency to reduce polyQ aggregation and polyQ-induced cell death in human HEK293T cells and patient derived fibroblasts. Several candidate molecules were able to decrease cell death and ameliorate the aggregation of polyQ peptides in cultured cells comparable to the TRMT2A knockdown experiments. Among these, spermidine was identified as able to cause a decrease in the abundance of polyQ aggregates in SCA3-patient derived fibroblasts. Our work provides a first step towards a pharmacological inhibition of this enzyme and indicates TRMT2A as a viable drug target for polyQ diseases.

Human TRMT2A methylates tRNA and contributes to translation fidelity

“…In some instances, these amino acids also crosslinked to incompletely digested dinucleotides such as UU. Mapping of both crosslinked amino acids C111 and H83 onto the previously published X-ray structure of the RBD from hTRMT2A FL (Margreiter et al, 2022) shows tRNA interaction with the ß-sheet surface (C111) and with one of the flexible loops (H83) ( Figure 3C ). For RBDs, the ß-sheet surface has been identified as the major RNA binding platform, although also loop-mediated RBD-RNA interactions have been reported (Cléry et al, 2008; Cléry et al, 2013).…”

“…Based on the results from binding and methylation experiments we choose tRNA Gln to dissect the RNA-interacting domains of hTRMT2A. For this purpose we generated several hTRMT2A fragments, whose domain borders were chosen based on experimental evidence (Margreiter et al, 2022), sequence alignments and homology models. We purified hTRMT2A FL, the RNA-binding domain (RBD, aa 69-147) and the central domain (Central, aa 237-414).…”

“…Thus, we hypothesized that hTRMT2A might be the dedicated m 5 U methyltransferase of cytosolic rRNA. Two hTRMT2A knockdown HEK293 cell lines (KD1, KD2) were generated using stable transfection of shRNA and were validated previously (Margreiter et al, 2022) ( Supplementary Figure 6B ). For experimental assessment of rRNA methylation status, we isolated cytosolic ribosomes from hTRMT2A WT and K D cells.…”

“…Stable RNAi-mediated knockdown of hTRMT2A was achieved as described earlier (Margreiter et al, 2022). HEK293t cells were infected with commercially available Lentiviral particles for hTRMT2A K D and scrambled control cell line (MISSION ® shRNA Lentviral Transduction Particles: scrambled control WT = SHC002V, KD1 = NM_182984.2- 856 s1c1, KD2 =NM_182984.2- 1574 s1c1; Sigma-Aldrich).…”

Towards design of drugs and delivery systems with the Martini coarse-grained model

“…The Martini 3 CG FF, with its increased accuracy and expanded coverage of the chemical space (Souza et al, 2021a;Alessandri et al, 2022), represents a competitive alternative for extracting and targeting druggable structures on such time-scales and/or predicting ligand-target interactions (Souza et al, 2020;Souza et al, 2021b). So far, one of the limitations of the Martini models is the description of proteins' conformational flexibility (Souza et al, 2019;Poma et al, 2017), which is fundamentally linked to biological function (Henzler-Wildman & Kern, 2007;Luo, 2012;Veesler & Johnson, 2012;Campaner et al, 2017;Hadden et al, 2018;Matthes et al, 2018;Maggi, Carloni, & Rossetti, 2020;Bolnykh et al, 2021;Noreng et al, 2021;;Jackson et al, 2022), and pivotal to design new therapeutics (Hammes, 2002;Campaner et al, 2017;Sengupta & Udgaonkar, 2019;Schulz-Schaeffer, Wemheuer, & Wrede, 2020;Xiao et al, 2021;Gossen et al, 2021;Zhao et al, 2021;Margreiter et al, 2022).…”