SRA

The problem of traumatic brain injury (TBI) heterogeneity has been a critical barrier to successful translation of therapies in the field. TBI heterogeneity exists in the patient substrate pre-injury (genetics, sex, comorbidities), external injury characteristics (severity, mechanism), and resultant post-injury host response that is responsible for deleterious secondary injury processes (seizures, neuroinflammation, neurodegeneration) and repair/regeneration. Identification of final common molecular pathways and signatures that integrate this vast heterogeneity could be valuable for guiding biomarkers, therapeutic targets, and predictive enrichment. In this study, we present the first large-scale searchable murine single-cell atlas of the transcriptomic response to TBI in 339,357 cells as a foundational step in molecularly deconstructing TBI heterogeneity. We identify 23 cell types with massive heterogeneity in the single-cell response across extrinsicand intrinsic factors, that has been underestimated. Majority of response to TBI was unique to individual cell populations with minimal overlap even within a single injury-model thus highlighting the importance of cell-level resolution. Through this effort, we report novel cell-specific targets and a previously unrecognized role for specific microglial and ependymal subtypes in post-TBI pathophysiology that is highly variable depending on the extrinsic and intrinsic factors studied. One ependymal subtype was a hub of neuroinflammatory signaling after contusional-TBI, particularly related to Il-1b. A single microglial-lineage along pseudotime (comprising 3 microglial subtypes) was a key mediator of host-response after TBI, and shared features with disease associated microglia noted in Alzheimer's disease and other neurodegenerative disorders, potentially providing a link between TBI and accelerated neurodegeneration. One microglial subtype within this lineage emerged as a key target – it was the only cell type of all 23 that retained persistent and marked gene expression changes 6 months post contusional-TBI. We identify sexually dimorphic gene expression and pathway vulnerabilities with cell-specific differences in both immune and non-immune biological processes. These likely contribute to sex-based outcome and warrant further study to facilitate discovery of cell- and sex-specific druggable targets. Active changes in brain regions distal from the site of primary TBI impact included infiltration of specific microglial populations as well as cell-specific transcriptomic changes in several genes and inflammatory processes distinct from both the peri-contusional and naïve signatures. This atlas validates several known contributors in TBI pathophysiology, and also identifies previously unrecognized targets and avenues for further research. Beyond our presented exemplar analyses (including pathways of clinical interest like sulfonylurea-receptor-1), the companion searchable atlas serves as a foundation for countless future efforts to understand cell-specific heterogeneity after TBI (https://shiny.crc.pitt.edu/cerebri/) as well as numerous other neurological diseases with overlapping pathophysiology. Overall design: We isolated and dissociated fresh tissue from naïve or injured cortex post-TBI in three clinically-relevant mouse models across a spectrum of TBI. Experiments were performed individually without hashing. Our established translational models included repetitive-Closed-Head-Injury TBI (rCHI, mimicking sports injuries), controlled cortical impact (CCI, mimicking cerebral contusion), and CCI with hemorrhagic-shock (devastating polytrauma frequently seen in critically ill TBI). Workflow of injury models/conditions and sampling locations from naïve and injured brain are described in the manuscript- briefly tissue was isolated from left pericontusional or naïve brain 24h post-injury (rCHI, CCI or CCI+HS), and 7d and 6mo post-CCI (Allen-Mouse Brain bregma values -0.755 to -2.75). Tissue was also obtained from females (ipsilateral) and male contralateral mirror-image brain-tissue 24h post-CCI in separate experiments. Tissue was micro-dissected, single-cells were isolated, sequenced and computationally analyzed. There are extensive publications on outcome implications in these models as cited in the paper. In CCI (the most common preclinical TBI-model), cells were isolated from both sexes, the ipsilateral/peri-contusional and mirror-image contralateral cortex, and multiple timepoints. After generating a single-cell gene-expression matrix that integrated naïve and injured cells we defined cell-cluster specific markers and identified ten umbrella cell-types with 23 transcriptomically distinct cell populations.