Fig. 6
ACTN4, ELAVL4 and USP9X overexpression reversed miR-142-5p- and plasma_EVCCI-induced dendritic spine damage. (A) The potential binding sites of miR-142-5p in ACTN4, ELAVL4 and USP9X. (B) The direct effects of miR-142-5p on ACTN4, ELAVL4 and USP9X were identified by reporter gene analysis. miR-142-5p has three potential binding sites for ELAVL4; therefore, two mutant plasmids were designed separately for ELAVL4 (MUT1), one of which included two relatively similar binding sites (MUT2), as presented in Supplementary Fig. 8F. After transfecting HEK293T cells with miR-142-5p and reporters carrying the 3′UTRs (including mutated binding sites) of ACTN4, ELAVL4 and USP9X, dual-luciferase reporter activities were detected. Compared to the WT + agomir group, the miR-142-5p agomir group exhibited reversed Rluc/luc (Renilla luciferase activity) activity in response to the mutated reporter plasmids ACTN4, ELAVL4 (MUT1) and USP9X. ordinary one-way followed by Dunnett’s multiple comparisons test was used for ACTN4 and USP9X; ordinary one-way followed by Tukey’s multiple comparisons test was used for ELAVL4. (C–H) After transfecting primary hippocampal neurons with AAVs overexpressing ACTN4, ELAVL4 or USP9X, plasma EVs and the miR-142-5p agomir were coadministered, and DIL staining and immunofluorescence staining were performed. Overexpression of these genes significantly ameliorated the impairments in the dendritic spine density of neurons and the protein expression of PSD95 and SYN in the plasma_EVCCI and miR-142-5p agomir groups compared with those in the plasma_EVCCI and miR-142-5p agomir groups. n = 3–4 images from three wells of each group. ordinary one-way followed by Dunnett’s multiple comparisons test was used. The data are shown as the means ± SDs. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001