From AlphaFold to site-selective inhibition of DNA methylation in glioblastoma multiforme

Abstract

Targeting specific protein–protein interactions, particularly those involving DNA methyltransferases (DNMTs), offers a promising strategy to resensitise glioblastoma multiforme cells to temozolomide. Glioblastoma multiforme is a highly aggressive brain cancer with a poor prognosis. The long-term efficacy of standard treatment, including temozolomide, is limited due to resistance development. Temozolomide resistance, often due to MGMT overexpression owing to global DNA hypomethylation and complex epigenetic regulation, significantly limits its efficacy. Thus, we developed site-specific DNMT1 inhibitors using peptide-based protein complex disruptors. Their design was guided by AlphaFold2 multimer, which provided insights into the interaction domains between DNMT1 and its protein partners. Peptides mimicking these interaction domains were synthesised and evaluated in a glioblastoma multiforme model. Our studies identified a peptide, namely PPDM2203, that specifically disrupts the DNMT1/DMAP1 interaction. This peptide did not induce global DNA hypomethylation but selectively inhibited DNMT1 activity at specific sites. Moreover, this peptide exhibited significant antitumoral properties, including sensibilisation to temozolomide, site-specific DNA methylation inhibition, cell proliferation, invasion and cytotoxicity. Therefore, the use of AlphaFold2 led to a peptide and offers a novel approach to selectively inhibit DNMT1 activity, enhancing the efficacy and specificity of existing DNMT1 inhibitors. This strategy underscores the potential for developing advanced therapeutic agents that minimize the risk of global DNA hypomethylation and its associated oncogenic effects.