Microglia Adopt Longitudinal Transcriptional Changes After Traumatic Brain Injury

Abstract

Background: Traumatic brain injury (TBI) is an under-recognized public health threat. Even mild brain injuries can lead to long-term neurologic impairment. Microglia play a fundamental role in the development and progression of this ensuing neurologic impairment. Despite this, a microglia-specific injury signature has yet to be identified. We hypothesized that TBI would lead to long-term changes in the transcriptional profile of microglial pathways associated with the development of subsequent neurologic impairment.

Materials and methods: Male C57BL/6 mice underwent TBI via a controlled cortical impact and were followed longitudinally. FACSorted microglia from TBI mice were subjected to Quantiseq 3'-biased RNA sequencing at 7, 30, and 90 d after TBI. K-means clustering on 396 differentially expressed genes was performed, and gene ontology enrichment analysis was used to determine corresponding enriched processes.

Results: Differentially expressed genes in microglia exhibited four main patterns of expression over the course of TBI. In particular, we identified four gene clusters which corresponded to the host defense response, synaptic plasticity, lipid remodeling, and membrane polarization.

Conclusions: Transcriptional profiling within individual populations of microglia after TBI remains a critical unmet research need within the field of TBI. This focused study identified several physiologic processes within microglia that may be associated with development of long-term neurologic impairment after TBI. These data demonstrate the capability of longitudinal transcriptional profiling to uncover potential cell-specific targets for the treatment of TBI.

Keywords: FACS; Microglia; RNA Sequencing; TBI.

Figure 1:

Microglia from head-shielded bone marrow chimeric mice are distinct from infiltrating macrophages after TBI. Brains isolated from chimeric mice post TBI were analyzed by flow cytometry. CD45.1^neg ‘B, T, NK cells and Eosinophils’ acted as the gating control for CD45.2^hi cells, showing that resident microglia (CD45^lo) can be unambiguously differentiated from infiltrating monocyte-derived macrophages (CD45.1⁺). Arrows indicate the directionality of gating.

Figure 2:

Microglia from TBI mice show distinct time-dependent transcriptional profiles. K-means clustering (k=4): Four clusters representing 396 genes are shown with distinct time-specific expression patterns. Example genes from each cluster (left) and significant GO processes (right) are shown.

Figure 3:

Differentially expressed genes over the time course of injury. Volcano plots of differentially expressed genes in microglia (n=9) between 7D and 30D post-TBI (A), 30D and 90D post-TBI (B), and 7D and 90D post-TBI (C). Examples of genes with a p-value <0.05 (as calculated by DEseq model) are shown in red.

Figure 4:

Correlational analysis of change in gene expression microglia over time after TBI. Scatter plot to examine the relationship between the genes that are up-regulated or down-regulated in the microglia of mice at A) 7 vs 30 days post-TBI compared to 7 vs 90 days post-TBI and at B) 30 vs 90 days post-TBI compared to 7 vs 90 days. A log2 fold change of 1 is equal to a 2-fold change between time point mean expression (indicated in black dots) (n=9).

Figure 5.

Normalized expression of Microglial Ptpn5, Trem2 and APOE over the course of TBI. Ptpn5 expression from sequenced FACSorted microglia progressively increased from 7 days post-TBI to 90 days post-TBI; p=0.01, Trem2 did not significantly change over the course of brain injury while APOE expression progressively decreased over time following TBI p=0.005. Ordinary one way ANOVA with Bonferroni’s multiple comparison test (n=9).

Figure 6.

Pathway map analysis of differentially expressed microglial genes after TBI. Expressed genes were input into the Metacore software and pathway map analyses was performed. The red highlighted lines show inhibition or negative effect, the green lines show activation or positive effect, the grey lines show an unspecified relationship between the two linked elements. All highlighted interactions have been implicated in brain and neurodegenerative diseases as curated by Metacore analytics.