Introduction: Traumatic brain injury (TBI) afflicts over 3 million Americans every year. Patients over 65 years of age suffer increased mortality as well as greater long-term neurocognitive and neuropsychiatric morbidity compared to younger adults. Microglia, the resident macrophages of the brain, are complicit in both. Our published and preliminary data have demonstrated a significant age-effect in which aged microglia are more prone to adopt a constitutively activated state associated with worse neurocognitive and neuropsychiatric outcomes. Therefore, we hypothesized that aged microglia would fail to return to a homeostatic state after TBI but instead adopt a long-term injury-associated state within the brain of aged mice as compared to young-adult mice after TBI.
Methods: Young-adult (14-weeks) and aged (80-weeks) C57BL/6 mice underwent TBI via controlled cortical impact vs. sham injury. We utilized single-cell RNA sequencing to examine age-associated cellular responses after TBI. Four months post-TBI or sham injury, brains were harvested, and CD45+ cells (N=4,000 cells) were isolated via florescence-activated cell sorting. cDNA libraries were prepared via the 10x Genomics Chromium Single Cell 3' Reagent Kit, followed by sequencing on a HiSeq 4000 instrument. The raw data were processed using the Cell Ranger pipeline mapped to the mm10 mouse reference genome and Seurat following standard workflow. Seurat and GOrilla were used for downstream clustering, differential gene expression, and pathway analysis. All cell types were annotated using canonical markers and top expressed genes. ProjecTILs was additionally used to interpret T cell states.
Results: Microglia from young-adult and aged mice have distinct transcriptional profiles pre-injury and markedly different transcriptional responses post-injury compared to young-adult mice. Pre-injury, aged mice demonstrated a disproportionate immune cell infiltration, including T cells, as compared to young-adult mice (aged versus young: 45.5% vs. 14.5%). Post-injury, the disparity was amplified with a proportional decrease in homeostatic microglia and greater increased infiltrating T cells compared to young-adult mice (Microglia: 27.5% vs. 71%; T cell: 45.5% vs. 4.5%). Of note, aged mice post-injury had a subpopulation of unique, age-specific, immune-inflammatory microglia resembling gene profiles of neurodegenerative disease-associated microglia (DAM) with enriched pathways involved in leukocyte recruitment and Alzheimer’s disease pathogenesis (FDR < 0.05). Contrastingly, post-injury, aged mice demonstrate a heterogenous T-cell infiltration with gene profiles corresponding to CD8 effector memory, CD8 native-like, CD4, and double-negative T cells (75.9%, 2.5%, 12.9%, and 8.6%, respectively) and enriched pathways including tau protein binding, macromolecule synthesis, and cytokine-mediated signaling pathways (FDR < 0.05).
Conclusion: We hypothesized that aged microglia would fail to return to a homeostatic state after TBI and adopt a long-term, injury-associated state within the brain of aged mice as compared to young-adult mice after TBI. In particular, our data suggest an age-dependent reduction of homeostatic microglia post-injury yet an upregulation in a unique microglial subpopulation with a distinct immuno-inflammatory profile. Furthermore, aged subjects demonstrated a markedly disproportionate inflammatory infiltrate after TBI predominated by the presence of CD8+ T cells. In addition, post-injury, brain trauma reorganized the T cell milieu, especially CD8 effector memory T cells, via upregulating genes associated with macromolecule biosynthesis process and negative regulation of neuronal death, possibly linking TBI with its long-term sequelae and complications. Taken together, our data showed that age-specific gene signature changes in the T-cell infiltrates and the microglial subpopulation contributes to increased vulnerability of the aged brain to TBI. Age should be an a priori consideration in future TBI clinical trials.