Unique Subtype of Microglia in Degenerative Thalamus After Cortical Stroke

Abstract

Background and purpose: Stroke disrupts neuronal functions in both local and remotely connected regions, leading to network-wide deficits that can hinder recovery. The thalamus is particularly affected, with progressive development of neurodegeneration accompanied by inflammatory responses. However, the complexity of the involved inflammatory responses is poorly understood. Herein we investigated the spatiotemporal changes in the secondary degenerative thalamus after cortical stroke, using targeted transcriptome approach in conjunction with histology and flow cytometry.

Methods: Cortical ischemic stroke was generated by permanent occlusion of the left middle cerebral artery in male C57BL6J mice. Neurodegeneration, neuroinflammatory responses, and microglial activation were examined in naive and stroke mice at from poststroke days (PD) 1 to 84, in both ipsilesional somatosensory cortex and ipsilesional thalamus. NanoString neuropathology panel (780 genes) was used to examine transcriptome changes at PD7 and PD28. Fluorescence activated cell sorting was used to collect CD11c⁺ microglia from ipsilesional thalamus, and gene expressions were validated by quantitative real-time polymerase chain reaction.

Results: Neurodegeneration in the thalamus was detected at PD7 and progressively worsened by PD28. This was accompanied by rapid microglial activation detected as early as PD1, which preceded the neurodegenerative changes. Transcriptome analysis showed higher number of differentially expressed genes in ipsilesional thalamus at PD28. Notably, neuroinflammation was the top activated pathway, and microglia was the most enriched cell type. Itgax (CD11c) was the most significantly increased gene, and its expression was highly detected in microglia. Flow-sorted CD11c⁺ microglia from degenerative thalamus indicated molecular signatures similar to neurodegenerative disease-associated microglia; these included downregulated Tmem119 and CX3CR1 and upregulated ApoE, Axl, LpL, CSF1, and Cst7.

Conclusions: Our findings demonstrate the dynamic changes of microglia after stroke and highlight the importance of investigating stroke network-wide deficits. Importantly, we report the existence of a unique subtype of microglia (CD11c⁺) with neurodegenerative disease-associated microglia features in the degenerative thalamus after stroke.

Keywords: cerebral ischemia; microglia; neuroinflammation; transcriptome.

Figure 1. Delayed neuron loss in thalamus after primary cortical ischemic injury.

(A) Diagram shows the location of secondary thalamic injury (blue labeled area in bottom section) after primary cortical injury (blue labeled area in upper section). The thalamic degeneration results in the anterograde/retrograde degeneration (arrows) after disruption of cortico-thalamic connections. (B) Experimental design and timeline. A time course histology study was performed to detect spatiotemporal changes in somatosensory cortex and thalamus after stroke. PD, post-stroke day. Transcriptome analysis was performed on samples collected from naïve, PD7 (beginning of secondary thalamic injury) and PD28 (severe secondary thalamic injury). FACS was used to sort microglia and CD11c⁺ microglia from naïve and PD28 thalamic tissues respectively to detect molecular features of CD11c⁺ microglia. (C) Representative full coronal sections show NeuN (green) and MAP2 (orange) immunostaining in thalamus. Enlarged images represent the square labeled region in full section images. Scale bar = 250μm. The numbers of NeuN⁺ cell (D) and MAP2⁺ cell (E) in iTH were counted and compared to naive. N=4–5/time-point. **P < 0.01, ***P < 0.001, ****P < 0.0001. Data are expressed as mean ± SEM.

Figure 2. Targeted transcriptome analysis of somatosensory cortex and thalamus after cortical ischemic stroke.

(A) Heat map of DEGs in iS1 and iTH at PD7 and PD28. (B) Volcano plots show fold-change relative to naïve samples. Red, blue and grey dots indicate up-regulated, down-regulated and unchanged genes respectively. Top 10 DEGs are labeled. N=3 in naïve, N=4 in PD7 and N=4 in PD28. (C) Numbers of altered genes in iS1 and iTH (D) Venn diagram represents the number of common DEGs at two time points in iS1 and iTH (E) Bar graphs indicate enriched cell types of DEGs in iS1 at PD7 and in iTH at PD28, plotted with −log10(p-value) and number of genes in the cell type. OPC: oligodendrocyte progenitor cell.

Figure 3. Ingenuity pathway analysis highlighted the involvement of neuroinflammation signaling after stroke.

(A) Top-ranked diseases and molecular/cellular functions. (B) Top significant canonical pathways, plotted by −log10(p-value) and ratio. NO, nitric oxide; ROS, reactive oxygen species; PRRs, pattern recognition receptors. (C) qPCR validated DEGs in neuroinflammation signaling (CSF1R, CX3CR1 and TGFB) in iTH. N=3 in naïve, N=4 in PD7 and N=4 in PD28. Itgax in iTH was quantified with another cohort of samples. N=4 in naïve, N=6 in PD7 and N=5 in PD28. *P < 0.05, **P < 0.01, ***P < 0.001. Data are expressed as mean ± SEM.

Figure 4. Microglia dynamically changed during the development of secondary thalamic injury.

(A) Representative full coronal sections show Tmem119 (green) and Iba-1 (red) immunostaining in thalamus. Enlarged images represent the square labeled region in full section images. Scale bar = 250μm. The integrated density of Tmem119 (B) and the integrated density of Iba-1 (C) in iTH were quantified and compared to naïve mice. N=4–5/time-point. *P < 0.05, **P < 0.01, ****P < 0.0001. Data are expressed as mean ± SEM. (D) Representative 3D-images of Iba-1 show morphology of microglia/macrophages at iTH in naïve and stroke mice. Scale bar = 10μm.

Figure 5. CD11c⁺ microglia/macrophage presented in the secondary thalamic injury.

(A) Representative full coronal sections show CD11c (green) and Iba-1 (red) immunostaining in thalamus. Enlarged images represent the square labeled region in full section images. Scale bar = 250μm. The percentage of CD11c area (B) and the integrated density of CD11c (C) were quantified in iTH and compared to naive. N=4–5/time-point. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data are expressed as mean ± SEM.

Figure 6. CD11c⁺ microglia in degenerative thalamus exhibited molecular features of DAM.

(A) Representative dot plots from FACS show gating strategy used to analyze immune cells isolated from samples of thalamus in naïve, iTH and cTH at PD28. Microglia were gated by CD45^intCD11b⁺ from singlet/live cells. (B) The percentage of CD11C⁺ microglia in naïve, iTH and cTH at PD28. (C) The expression of Itgax (CD11c) was measured on iTH CD11c⁺, iTH CD11c⁻, cTH and naïve microglia. (D) qPCR measured DAM molecular signatures (ApoE, Axl, LpL, CSF1, Cst7, Tmem119 and CX3CR1) on CD11c⁺ microglia from iTH, and compared with CD45^intCD11b⁺ microglia from naïve, cTH and CD11c⁻ microglia from iTH. N=3/group, with each sample pooled from eight mice in PD28 group. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. Data are expressed in mean ± SEM. (E) Schematic figure represents the dynamics of microglia in the secondary thalamic injury. A secondary degenerative injury generates in ipsilesional thalamus after primary cortical stroke, resulting in the disruption of functional connections between cortex and thalamus. Morphological changes of microglia are shown in different shapes and colors. The key microglia-related genes are listed, and the trend of gene expression are pointed by arrowheads. A subpopulation of microglia (CD11c⁺) are present in the degenerative thalamus between PD14–28. Figure is created with BioRender.com.