Fig 2: Exo70 is redistributed into PSD in the hippocampus of mTBI mice. a Schematic representation of subcellular fractionation proteins. b Example of subcellular fractionation. Hippocampus from two-month-old male mice was fractionated and microsome, PSD, and nonPSD fractions were obtained. 20 µg of protein samples were resolved in a 10% SDS-PAGE and transferred to PVDF membranes. Membranes were incubated with the respective antibodies shown in the figure. Membranes were stripped and tested again with the indicated antibodies. c Proteins distribution were analyzed with densitometric analysis by comparing signal intensity from each fraction with homogenized. Mean values ± SEM are shown. d Cortex and Hippocampus (f) from Sham and mTBI mice were fractionated and analyzed by western blot using Exo70, PDI, Actin, and Tubulin antibodies. PDI/Actin/Tubulin was used as loading controls. 30 µg of protein samples were used. e, g The graph shows the Exo70 densitometric analysis normalized with loading controls. Values represent means ± SEM, n = 3 mice per experimental group. Statistical differences were determined by an unpaired t-test comparing Sham and mTBI. *p < 0.05, **p < 0.01

Fig 3: Interaction of Cdc42-palm and TC10 with Exo70.(A) Schematic representation of the Exo70 fragments that were used for interaction studies. The numbers indicate the position of amino acids in the coding sequence. (B, C) HEK293T cells were co-transfected with vectors for sfGFP-Exo70-NT (aa 1–84), sfGFP-Exo70-NT + H1 (aa 1–106), sfGFP-Exo70-H1-H2 (aa 85–106), sfGFP-Exo70-H2-H6 (aa 107–240), sfGFP-Exo70-H3-H6 (aa 133–240), sfGFP-Exo70-H7-H10 (aa 241–404), or sfGFP-Exo70-H11-H19 (aa 405–653) and myc-TC10-Q67L (B) or myc-Cdc42b-G12V (C) and immunoprecipitated with an anti-GFP nanobody (IP), and bound proteins were analyzed by Western blot (WB). The molecular weight is indicated in kD. (D) Validation of the Exo70 knockdown vector. HEK293T cells were co-transfected with vectors for mouse Exo70-sfGFP (mmExo70-sfGFP) or human Exo70 tagged at the N-terminus (sfGFP-hsExo70) or C-terminus (hsExo70-sfGFP) and control or knockdown vectors for Exo70 expressing EmGFP. The efficiency of the Exo70 knockdown was analyzed by Western blot using an anti-GFP antibody. Human Exo70 is resistant to the knockdown vector because of mismatches in the nucleotide sequence. (E) Quantification of the knockdown efficiency from (D). (F) Knockdown of Exo70 blocks Igf1-induced exocytosis. Hippocampal neurons from E18 rat embryos were transfected with control vector or knockdown vectors for Exo70 expressing H2B-mRFP. Neurons were incubated with BODIPY FL C5-ceramide at 1 d.i.v. to label post-Golgi vesicles and cultured in medium without (-Igf1) or with (+Igf1) 20 nM Igf1 for 10 min. Igf1-stimulated exocytosis results in a decrease in the fluorescence signal for BODIPY-labeled vesicles in growth cones at 620 nm. Scale bar: 2 µm. (G) Igf1-stimulated exocytosis was quantified as the decrease in the fluorescence intensity for BODIPY-labeled vesicles in growth cones at 620 nm normalized to the signal at 515 nm. The relative change in fluorescence intensities for BODIPY-labeled vesicles is shown for (F). The value for control neurons (-Igf1) was set to 100% (one-way ANOVA followed by Tukey’s test, N = 27 neurons from each condition from n = 3 independent experiments that are biological replicates). (H) Expression of hsExo70-sfGFP but not sfGFP-hsExo70 rescues the knockdown of Exo70 in neurons. Hippocampal neurons from E18 rat embryos were co-transfected with control or knockdown vectors for Exo70 expressing H2B-RFP (yellow) and vectors for sfGFP-hsExo70 or hsExo70-sfGFP (cyan). Neurons were analyzed at 3 d.i.v. with an anti-acetylated tubulin antibody (magenta). Scale bar: 20 µm. (I) Percentage of neurons without an axon (0), with a single axon (1), and with multiple axons (>1) was quantified for (H) (an unpaired t test, n = 3 independent experiments that are biological replicates, N= 250 neurons). Interestingly, a fusion of sfGFP to the N-terminus of Exo70 (sfGFP-Exo70), which is similar to N-terminally tagged constructs that were used in some previous studies (Zuo et al, 2006; Liu et al, 2007; Lira et al, 2018), did not rescue the knockdown. Values are means ± SEM. ns, P > 0.05; *P = 0.05; **P = 0.01; ***P = 0.001; and ****P = 0.0001.Source data are available for this figure.

Fig 4: Loss of Arhgef7 impairs the exocytosis of post-Golgi vesicles.(A) Cortical neurons from E17 Arhgef7flox/+;Emx1Cre/+ (Arhgef7+/-) and Arhgef7flox/flox;Emx1Cre/+ (Arhgef7-/-) embryos were incubated with BODIPY FL C5-ceramide at 1 d.i.v. to label post-Golgi vesicles and cultured in medium without (-Igf1) or with (+Igf1) 20 nM Igf1 for 10 min. A decrease in the fluorescence intensity at 620 nm indicates the exocytosis of vesicles in growth cones. Scale bar: 2 µm. (B) Igf1-stimulated exocytosis was quantified for (A) as the relative change in the fluorescence intensity for BODIPY-labeled vesicles in growth cones at 620 nm, normalized to the signal at 515 nm. The value for control neurons (-Igf1) was set as 100% (one-way ANOVA followed by Tukey’s test, N = 45 neurons from each condition from n = 3 independent experiments that are biological replicates). (C) Cortical neurons from Arhgef7+/- and Arhgef7-/- E17 embryos were analyzed at 1 d.i.v. by staining with an anti-Exo70 antibody. Epifluorescence and TIRF signals indicate total and plasma membrane-proximal Exo70, respectively. The neurite with the highest Exo70 signal (1) and a second neurite (2) are displayed at a higher magnification. Scale bar: 10 and 2 µm, respectively. (D, E) Membrane-proximal Exo70 was quantified for (C) at 1 d.i.v. in cortical neurons from Arhgef7+/- and Arhgef7-/- embryos. The intensity of the total internal reflection fluorescence microscopy signal in neurites was normalized to epifluorescence signals (total Exo70) and area. The average relative intensity for membrane-proximal Exo70 in all neurites (D) and the brightest neurites (E) was quantified (an unpaired t test, N = 42 neurons from n = 3 independent experiments that are biological replicates). Values are means ± SEM. ns, P > 0.05; *P = 0.05; **P = 0.01; ***P = 0.001; and ****P = 0.0001.Source data are available for this figure.

Fig 5: mTBI increases exocyst complex assembly and interaction between Exo70 and GluN2B in the PSD fraction. Two-month-old male mice were subjected to mTBI and hippocampal PSD/nonPSD fractions were obtained. 50 and 300 µg of PSD (a) and nonPSD (c) protein samples, respectively, were immunoprecipitated using a specific Exo70 polyclonal antibody. Beads were recovered by brief centrifugation. Samples were resolved in 6% SDS-PAGE and transferred to PVDF membranes. Signal intensity was normalized with Exo70 band intensity. Both hippocampi of each mouse were used. Membranes were stripped and tested again with the indicated antibodies. b, d GluN2A, GluN2B, GluR1, GluR2, Sec6, and Sec10 densitometric analysis from PSD and nonPSD immunoprecipitations. The analysis shows the assembly level of the exocyst complex and Exo70-GluN2B increased interaction after mTBI. Values represent means ± SEM, n = 3 mice per experimental group. Statistical differences were determined by an unpaired t-test comparing Sham and mTBI. *p < 0.05