Dissection of Microglial Heterogeneity and Identification of Pain-Related Subpopulations After Traumatic Brain Injury

Pain following traumatic brain injury (TBI) is a major challenge in clinical management. Neuroinflammatory responses, particularly those involving microglia, are increasingly implicated in the synaptic and circuit-level remodeling associated with post-TBI pain. However, the conventional M1/M2 framework does not adequately capture the temporal and functional diversity of microglial states after injury. This study aimed to characterize microglial heterogeneity and identify specific microglial subpopulations potentially associated with post-TBI pain-related neural remodeling.

We reanalyzed the publicly available single-cell RNA sequencing dataset GSE226207, which comprised intact cortical samples and cortical samples collected at 3 and 5 days after cortical stab-wound injury from control and inhibitor-treated mice. The inhibitor-treated groups received combined pharmacological inhibition of the CXCR3 and TLR1/2 pathways. Seurat was used for data processing, clustering, cell-type annotation, and microglial subclustering. AUCell was applied to evaluate inflammatory-, repair-, and pain-relevant gene signatures. CytoTRACE and Slingshot were used to assess relative transcriptional complexity and infer state-transition trajectories. CellChat analysis of pooled cells from all groups was performed to predict intercellular communication, and SCENIC was used to characterize subpopulation-associated regulon activity.

A total of 66,310 cells were retained, including 7648 microglia that were resolved into eight transcriptionally distinct subpopulations. Among them, the Lars2⁺ microglial subpopulation showed lower representation in the 3 days postinjury (dpi) groups and greater representation in the 5 dpi groups. This subpopulation exhibited relatively high TGF-β- and MAPK-associated signature scores and was enriched in biological processes related to synaptic organization, axonogenesis, dendrite development, and protein synthesis. Slingshot positioned Lars2⁺ microglia at the distal end of one inferred state-transition lineage. CellChat analysis predicted prominent neuron-to-microglia communication through the CX3CL1-CX3CR1 ligand-receptor pair, with Lars2⁺ microglia predominantly positioned as signal receivers. SCENIC further identified relatively high activity of regulatory module M1 in Lars2⁺ microglia, with Thra among the highly ranked regulators.

This study identifies the Lars2⁺ microglial subpopulation as a temporally dynamic and transcriptionally specialized component of the postinjury microglial landscape. The convergence of pain-relevant signaling signatures, neural remodeling programs, inferred state-transition characteristics, predicted neuron-microglia communication involving the CX3CL1-CX3CR1 pair, and Thra-associated regulon activity supports a potential role for Lars2⁺ microglia in posttraumatic neuroimmune and synaptic remodeling. These findings provide a cellular and molecular framework for future investigation of microglia-targeted strategies for post-TBI pain.