Figure 1.
ApoE is required for GW3965 to improve NOR performance after mrTBI.
NOR memory was evaluated in untreated (V) and GW3965-treated (G) WT (A) and apoE −/− (B) mice at 2, 7, and 14 days post-mrTBI. Bars represent discrimination index (DI) scores. Cohorts were: sham (WT, n = 23, pooled; apoE−/−, n = 18, pooled) and 2 day (WT: untreated, n = 18, GW3965-treated, n = 15; apoE−/−: untreated, n = 10, GW3965-treated, n = 10), 7 day (WT: untreated, n = 13, GW3965-treated, n = 10; apoE−/−: untreated, n = 11, GW3965-treated, n = 11), and 14 day (WT: untreated, n = 14, GW3965-treated, n = 14; apoE−/−: untreated, n = 9, GW3965-treated, n = 9) post-mrTBI. DI scores were analyzed using two-way ANOVA and Bonferroni post hoc test. **: p<0.01, ***: p<0.001, ###: p<0.001.
Figure 2.
mrTBI-induced motor impairment recovers spontaneously independent of GW3965 and apoE.
Motor performance of WT (n = 38) and apoE−/− (n = 30) mice was evaluated by the accelerating rotarod task (0 to 30 rpm in 210 s). (A) Rotarod latencies of untreated (V, open squares) and GW3965-treated (G, black filled squares) WT mice before and following mrTBI. (B) Rotarod latencies of untreated (V, open circles) and GW3965-treated (G, black filled circles) apoE−/− mice before and following mrTBI. Asterisks in A and B denote significant differences between baseline and post-mrTBI latencies within each group. For both genotypes, the treatment effect was not significant at any post-TBI time point (curly brackets). (C) Rotarod latencies of untreated WT (open squares) and apoE−/− (open circles) mice before and following mrTBI. (D) Rotarod latencies of GW3965-treated WT (black filled squares) and apoE−/− (black filled circles) mice before and following mrTBI. # and § in C and D denote significant differences between the latencies of WT and apoE−/− mice at the respective time points. Cohorts were: Untreated WT mice: (1 d = 34, 2 d = 34, 7 d = 24; 14 d = 14); GW3965-treated WT mice: (1 d = 35, 2 d = 35, 7 d = 25, 14 d = 14); untreated apoE−/− mice: (1 d = 24, 2 d = 24, 7 d = 20, 14 d = 9), and GW3965-treated apoE−/− mice: (1 d = 27, 2 d = 22, 7 d = 20, 14 d = 9). Data were analyzed using two-way repeated measures ANOVA followed by a Bonferroni post hoc test. *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001, #: p<0.05, ##: p<0.01, §§: p<0.01.
Figure 3.
GW3965 prevents mrTBI-induced accumulation of endogenous Aβ in WT and apoE−/− mice.
Total, soluble murine Aβ40 and Aβ42 levels were measured from ipsilateral half brains of WT (A, B) and apoE−/− (C, D) mice, respectively, at 2, 7, and 14 d post-mrTBI. Data from sham animals within each genotype were pooled (grey bars). Numbers inside the bars indicate sample size. Asterisks above individual bars indicate significant difference compared to the respective sham levels. *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001. # indicates a significant difference (p<0.05) between Aβ40 levels of untreated WT and apoE−/− brains at 7 d post-mrTBI. § indicates a significant difference (p<0.05) between Aβ40 levels of GW3965-treated WT and apoE−/− brains at 7 d post-mrTBI. Data were analyzed by two-way ANOVA followed by a Bonferroni post hoc test. Legend: S: sham-operated mice, gray bars, V: untreated mice, open bars, G: GW3965-treated mice, black bars.
Figure 4.
GW3965 augments ABCA1 levels in WT and apoE−/− mice following mrTBI.
ABCA1 protein levels were determined in ipsilateral half brains of WT (A) and apoE−/− (B) mice following mrTBI using Western blots, with representative blots shown below the graphs. Data are expressed as fold difference normalized to sham values. Data from sham animals within each genotype were pooled (grey bars). Numbers inside the bars indicate sample size. Asterisks above individual bars indicate significant difference compared to the respective sham levels. *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001. Data were analyzed by two-way ANOVA followed by a Bonferroni post hoc test. Legend: S: sham V: untreated mice, open bars, G: GW3965-treated mice, black bars.
Figure 5.
Loss of apoE exacerbates axonal injury after mrTBI.
Axonal damage following mrTBI was assessed with silver staining (arrows). The left panel depicts representative images of silver-stained coronal sections at approximately −1.58 mm from bregma [66] of untreated WT (A) and apoE−/− (B) mouse brains harvested at 7 d post-mrTBI. White matter areas with prominent silver staining are indicated by the black squares. Injury location is indicated by the arrowhead. The right panel (C–L) depicts representative 40x-magnified images of silver staining in five white matter areas in the brains of untreated WT (C–G) and apoE−/− (H–L) mice harvested at 7 d post-mrTBI.
Figure 6.
ApoE is required for GW3965 to suppress axonal damage after mrTBI.
Silver-stained images of white matter areas in WT and apoE−/− brains were analyzed semiquantitatively using an arbitrary silver staining scale extending from 0 (<10% argyrophilic structures covering the image field) to 3 (>70% argyrophilic structures covering the image field). The bar graphs represent mean ± SEM silver stain intensity score (arbitrary value) of WT (A–E) and apoE−/− (F–J) brains in the corpus callosum (A, F), cingulum (B, G), external capsule (C, H), internal capsule (D, I), and optic tracts (E, J). Sample sizes were: sham, n = 6/genotype (pooled); untreated mrTBI, n = 5/time point/genotype; GW3965-treated mrTBI, n = 5/time point/genotype. Asterisks above individual bars indicate significant differences compared to the respective sham values (grey bars). *: p<0.05, **: p<0.01, ***: p<0.001, ****: p<0.0001. ### (p<0.001) and #### (p<0.0001) represent significant differences in silver stain scores between untreated WT and apoE−/− mice. § (p<0.05), §§ (p<0.01), §§§ (p<0.001), and §§§§ (p<0.0001) represent significant differences in silver stain scores between GW3965-treated WT and apoE−/− mice. Data were analyzed by two-way ANOVA followed by a Bonferroni post hoc test. Legend: V- untreated mice, open bars, G- GW3965-treated mice, black bars.
Figure 7.
Microglia are not significantly activated by mrTBI in either WT or apoE−/− mice.
Microglial activation in sham and injured WT and apoE−/− mice was assessed with Iba1 immunohistochemistry. The top panel depicts representative images of Iba1-stained coronal sections at approximately −1.82 mm from bregma [66]. The bottom panel depicts 10X-magnified images of ipsilateral cortex underlying the injury site (indicated by the black rectangle). Legend: V- untreated mice, G- GW3965-treated mice.
Figure 8.
Pronounced microglial activation is localized only around contused areas.
In this study, approximately 4–8% of brains subjected to mrTBI showed micro contusions in the absence of gross skull fracture. The left panel shows a Iba1-stained coronal section at approximately −1.82 mm from bregma [66] with a micro contusion in the cortex below impact site (black square). The right panel shows representative 10X-magnified images of Iba1-stained untreated WT and apoE−/− contused cortices. Microcontusions are denoted by black arrows. Note the pronounced localized activation of microglia around the contusion.