Traumatic Brain Injury in Aged Mice Induces Chronic Microglia Activation, Synapse Loss, and Complement-Dependent Memory Deficits

Abstract

Traumatic brain injury (TBI) is of particular concern for the aging community since there is both increased incidence of TBI and decreased functional recovery in this population. In addition, TBI is the strongest environmental risk factor for development of Alzheimer's disease and other dementia-related neurodegenerative disorders. Critical changes that affect cognition take place over time following the initial insult. Our previous work identified immune system activation as a key contributor to cognitive deficits observed in aged animals. Using a focal contusion model in the current study, we demonstrate a brain lesion and cavitation formation, as well as prolonged blood⁻brain barrier breakdown. These changes were associated with a prolonged inflammatory response, characterized by increased microglial cell number and phagocytic activity 30 days post injury, corresponding to significant memory deficits. We next aimed to identify the injury-induced cellular and molecular changes that lead to chronic cognitive deficits in aged animals, and measured increases in complement initiation components C1q, C3, and CR3, which are known to regulate microglial⁻synapse interactions. Specifically, we found significant accumulation of C1q on synapses within the hippocampus, which was paralleled by synapse loss 30 days post injury. We used genetic and pharmacological approaches to determine the mechanistic role of complement initiation on cognitive loss in aging animals after TBI. Notably, both genetic and pharmacological blockade of the complement pathway prevented memory deficits in aged injured animals. Thus, therapeutically targeting early components of the complement cascade represents a significant avenue for possible clinical intervention following TBI in the aging population.

Keywords: C1q; complement; microglia; synapse; traumatic brain injury.

Figure 1

T₂-weighted magnetic resonance imaging (MRI) monitors longitudinal changes in lesion size and cavitation after contusion injury in aged animals. (A) Axial T₂-weighted magnetic resonance (MR) representative images of an injured animal, acquired a day prior to injury (baseline), and on days 1, 7, and 28 post injury on a 14.1-tesla MRI system using the following parameters: echo time (TE)/repetition time (TR) = 12/2000 ms; slice thickness = 0.5 mm; number of averages = 8; matrix = 256 × 256; field of view (FOV) = 30 × 30 mm²; acquisition time = 8 min. (B) Corresponding three-dimensional (3D) rendering of the lesion (orange) and cavitation (red) over time for the same injured animal. (C) Lesion size (expressed in mm³, defined as a mix of hyper- and hypo-intense contrasts) was measured starting at one day post injury. One-way ANOVA revealed significant differences (F = 47.77; p < 0.0001). Tukey post hoc test revealed differences between groups; n = 5–6/time point. (D) Cavitation (expressed in mm³, defined as hyperintense contrasts) was measured starting at seven days, and peaked at 28 days post injury. One-way ANOVA was used for non-parametric analysis (Kruskal–Wallis test), and Dunn’s multiple comparison test was used post hoc. Error bars represent group means and standard errors of the mean (SEM); *** p < 0.01.

Figure 2

Post-contrast T₁-weighted MRI detects persistent blood–brain barrier (BBB) damage after contusion injury in aged animals. (A) Enhancement ratio map acquired at 28 days post injury (white arrow), as calculated from T₁-weighted magnetic resonance (MR) images acquired pre- and post-injection of 1 mmol/kg Magnevist^® on a 14.1-tesla MRI system using the corresponding parameters: TE/TR = 4.6/112 ms; slice thickness = 0.5 mm; number of averages = 30; flip angle = 10; matrix = 256 × 192; field of view (FOV) = 25 × 25 mm²; acquisition time = 10 min 41 s. Regions of interest (ROIs) used for quantification of ipsilateral (red) and contralateral (blue) signals are superimposed. (B) Quantification of the mean intensity ratio, where S_pre and S_post represent the signal intensity before and after contrast injection, in ipsilateral (closed black squares) and contralateral (open black squares) ROIs shows a significant increase in enhancement in the ipsilateral ROI at every time point post injury. Error bars represent means and SEM. Two-way repeated measure ANOVA found significant differences in the time effect (F = 4.575, p < 0.05) and brain region effect (F = 32.24, p < 0.0001), as well as their interaction (F = 3.77, p < 0.05), with Sidak correction. * p < 0.05; ** p < 0.01.

Figure 3

Increased microglia numbers and phagocytic activity chronically after contusion injury. (A) Cluster of differentiation 11b (CD11b) gene-expression changes in the hippocampus of aged animals were measured by qPCR analysis, comparing expression levels between sham uninjured animals, and those 7 and 30 days post injury. CD11b, a marker for microglia and macrophages, significantly increases post injury. One-way ANOVA revealed significant differences (F = 21.12; p < 0.0001). Tukey post hoc test revealed differences between groups; n = 4–6/group (B) Brain microglia composition increases at 10 and 30 days post injury. Flow cytometry result revealed that percentage of microglia in the brain significantly increased at both 10 and 30 days after traumatic brain injury (TBI). Two-way ANOVA revealed significant effects of time (F(1,8) = 21.93, p < 0.01) and injury (F(1,8) = 53.59, p < 0.0001) without significant interaction (F(1,8) = 2.341, p = 0.1645). (C) Changes in synaptosomes (containing post-synaptic density protein PSD95) phagocytosis by microglia. There was no change in microglia phagocytosis activity at 10 days after TBI. However, at 30 days after TBI, there was a significant increase of exogenous synaptosome phagocytosis activity in microglia. Two-way ANOVA revealed significant differences in the effects of time (F = 5.333, p < 0.05) and injury (F = 8.456, p < 0.05), with significant interaction (F = 6.111, p < 0.05). Bars depict group means and SEM; * p < 0.05, ** p < 0.01, *** p < 0.0001.

Figure 4

Contusion injury induces robust complement initiation in the aged brain. Complement initiation components (A) C1q and (B) C3 gene-expression changes in the hippocampus of aged animals were measured by qPCR analysis, comparing expression levels between sham animals, and those 17, 14, and 30 days post injury. Both C1q and C3 expression levels were increased after injury. One-way ANOVA revealed significant differences (C1q-F = 15.89; p < 0.0001; C3-F = 50.58; p < 0.0001). Bonferroni post hoc test revealed differences between groups; n = 4–6/group. (C) C1q protein expression was measured by immunohistochemical staining in the dorsal hippocampus. Representative images from sham and TBI (30 dpi) animals (inset, in red C1q staining and in blue cell body counterstaining). C1q protein expression increased chronically after injury. No reactivity was observed in secondary alone. White scale bar = 50 µm; 200× magnification; n = 4–6/group. (D) Hippocampi were collected and synaptosomes were isolated by sucrose gradient followed by size calibration beads. C1q co-localization of synapse was measured by antibody staining. TBI (30 dpi) significantly increases C1q accumulation at synapse when compared with sham animals. Student’s t-test with Welch’s correction was used to measure differences between groups; n = 6/group. Bars depict group means and SEM; * p < 0.05, ** p < 0.01, *** p < 0.01.

Figure 5

Contusive injury results in chronic synapse loss in the aged brain. (A) Hippocampi were collected and synaptosomes were isolated by sucrose gradient followed by size calibration beads, before the co-expression of pre- and post-synaptic markers. The pre-synaptic marker was Synapsin-1 (Syn-1) and the post synaptic marker was post-synaptic density protein 95 (PSD-95). Significant decreases were found in total synaptosome numbers in the TBI (30 dpi) group when compared to sham group; n = 8–9/group. (B) Synaptic levels were measured by PSD-95 staining in the dorsal hippocampus. Representative images are shown in the inset. Aged TBI (30 dpi) animals had reduced PSD-95 expression. No reactivity was observed in secondary alone. White scale bar = 30 µm; 200× magnification; i = 4/group (PSD95 in green and cell bodies in blue). Student’s t-test was used to measure differences. Bars depict group means and SEM; * p < 0.05.

Figure 6

Complement blockade prevents memory deficits in aged animals after contusion injury. Trauma-induced memory deficits were measured by novel object recognition. Animals were exposed to two identical objects; 24 h later, the animals were exposed to one familiar object and one novel object. Memory deficits were calculated by a deficit in distinguishing the new (novel) object. (A) Both male and female animals displayed memory deficits following TBI (30 dpi). Two-way ANOVA revealed a significant injury effect (F = 4.42, p < 0.05); n (males) = 15–17/group; n (females) = 6–7/group. (B) Genetic C3 knock-out (C3^−/−) or pharmacological (C1q-inhibiting antibody (Ab)) blockade of complement initiation factors prevents trauma-induced deficits; n (TBI wild-type) = 27 mice; n (TBI C3^−/−) = 4 mice; n (TBI + C1q Ab) = 7 mice. One-way ANOVA revealed significant differences (F = 6.35; p < 0.01). Tukey post hoc analysis was used for differences between groups. * p < 0.05.

Figure 7

Schematic diagram of the working model. Normal memory function in mice corresponds with numerous synapses and surveying microglia in the hippocampus, the brain region responsible for forming new memories. Traumatic brain injury causes activation of microglia and long-lasting increases in the expressions of C1q and C3 at the synapses in the hippocampus. Microglia expressing the C3R receptor move in and bind the C3-tagged synapses, resulting in phagocytosis of the synapse. The combination of activated microglia and complement-mediated neuronal tagging results in synapse loss in the hippocampus and consequent memory loss. Genetic or pharmacological intervention of the complement pathway prevents trauma-induced memory deficits.