Discovery of Stable and Permeable <i>N</i> -Methylated Cyclic Peptides to Block the Endocytosis of GluA2 AMPAR for Ischemic Stroke Therapy

The membrane-permeable peptide <b>Tat-GluA2-3Y</b> exhibits therapeutic potential by competitively binding BRAG2 to disrupt GluA2 endocytosis. However, their poor <i>in vivo</i> stability has become a barrier to clinical application. To address this limitation, we systematically designed a series of <i>N</i>-methylated cyclic peptides derived from <b>GluA2-3Y</b> and identified the optimal peptide, <b>c10c-G1V3</b>, which exhibited high binding affinity for BRAG2 (<i>K</i>_d = 7.37 μM) and potent neuroprotective activity in both the OGD-induced cell model (cell viability: 85.97%) and the Glu-induced cell model (cell viability: 76.34%) at 10 μM. Mechanistically, <b>c10c-G1V3</b> markedly attenuated Glu-induced intracellular reactive oxygen species accumulation. Notably, it displayed superior plasma stability and favorable <i>in vivo</i> pharmacokinetic properties compared to <b>Tat-GluA2-3Y</b>. In SD rat tMCAO models, 8 mg/kg <b>c10c-G1V3</b> reduced infarct volume comparably to that of <b>Tat-GluA2-3Y</b>. Collectively, these findings establish <b>c10c-G1V3</b> as a promising candidate for the development of novel therapeutics against an ischemic stroke.