Ablation of microglia following infection of the central nervous system with a neurotropic murine coronavirus infection leads to increased demyelination and impaired remyelination

Abstract

Intracranial inoculation of susceptible mice with a glial-tropic strain of mouse hepatitis virus (JHMV), a murine coronavirus, results in an acute encephalomyelitis followed by viral persistence in white matter tracts accompanied by chronic neuroinflammation and demyelination. Microglia are the resident immune cell of the central nervous system (CNS) and are considered important in regulating events associated with neuroinflammation as well as influencing both white matter damage and remyelination. To better understand mechanisms by which microglia contribute to these immune-mediated events, JHMV-infected mice with established demyelination were treated with the small molecular inhibitor of colony stimulating factor 1 receptor (CSF1R), PLX5622, to deplete microglia. Treatment with PLX5622 did not affect viral replication within the CNS yet the severity of demyelination was increased and remyelination impaired compared to control mice. Gene expression analysis revealed that targeting microglia resulted in altered expression of genes associated with immune cell activation and phagocytosis of myelin debris. These findings indicate that microglia are not critical in viral surveillance in persistently JHMV-infected mice yet restrict white matter damage and remyelination, in part, by influencing phagocytosis of myelin debris.

Keywords: Coronavirus; Demyelination; Microglia; Neuroinflammation; Remyelination.

Figure 1.. Treatment of JHMV-infected mice with PLX5622 does not affect survival or control of viral replication.

JHMV-infected mice were fed either PLX5622 or control chow beginning at day 14 p.i. through day 28 p.i. PLX5622 treatment did not affect either (A) survival or (B) clinical disease severity (n=44 per group) compared to control mice. The red dash line indicates the time point that PLX5622 treatment was initiated. (C) PLX5622-treatment did not affect control of viral replication within the brains at days 21 and 28 p.i. (control n=4 at day 21 p.i., n=9 at day 28 p.i.; PLX5622 n=5 at day 21p.i., n=9 at day 28 p.i.) compared to control mice as determined by plaque assay. (D) Representative flow cytometric data from JHMV-infected mice treated with either PLX5622 or control chow and gating on microglia (CD45^loCD11b⁺ cells) or myeloid cells (CD45^hi CD11b⁺ cells). Quantification of microglia and myeloid cells in (E) brains at day 21 and 28 p.i. (n=6 per group at day 21 p.i.; n=4 per group at day 28 p.i.) and (F) spinal cords at day 21 p.i. (n=5 per group). PLX5622-treatment resulted in reduced numbers of microglia in brains at day 21 p.i. and day 28 p.i. compared to control mice. Data in C, E, and F are presented as average ± SEM; *p<0.05 and **p<0.01; ns, not significant.

Figure 2.. PLX5622 treatment does not affect T cell infiltration into the CNS.

Representative flow plots show (A) total CD4+ and virus-specific CD4+ T cells from control and PLX5622-treated mice at day 21 p.i. Quantification of (B) total and (C) virus-specific CD4+ T cells at days 21 and 28 indicate no differences between control mice and PLX5622-treated mice. Representative flow plots show (D) total CD8+ and virus-specific CD8+ T cells from control and PLX5622-treated mice at day 21 p.i. Quantification of (E) total and (F) virus-specific CD8+ T cells at days 21 and 28 indicate no differences between control mice and PLX5622-treated mice. PE-conjugated I-A^b/M133–147 tetramers and PE-conjugated D^b/S510–518 were used to stain virus-specific CD4⁺ T cells and virus-specific CD8⁺ T cells, respectively. Data is derived from two independent experiments: n=6 per group at day 21 p.i.; n=4 per group at day 28 p.i. Data in B-D are presented as average ± SEM.

Figure 3.. Targeting microglia impacts immune response gene expression.

RNA sequencing was performed on spinal cords from JHMV-infected mice given either control chow (n=7) or PLX5622 (n=8) beginning at day 14 p.i. and subsequently sacrificed at day 28 p.i. (A) Volcano plot displaying fold change of genes (log₂ scale) and P values (−log₁₀ scale) between JHMV-infected control and PLX5622-treated mice. (B) Heatmap analysis showing differential gene expression in spinal cords from control versus PLX5622-treated mice. Biological pathways (C) and (D) examined via GO enrichment that are downregulated in PLX5622-treated mice compared to controls.

Figure 4.. The severity of spinal cord demyelination is increased in PLX5622-treated mice compared to control mice.

JHMV-infected mice treated with either control chow or PLX5622 from day 14 p.i. were sacrificed at day 21 or 28 p.i. and spinal cords removed to evaluate the severity of demyelination. Representative images of H&E/LFB-stained spinal cord sections showing an increase in severity of demyelination (yellow line) in JHMV-infected mice treated with PLX5622 compared to control treated mice at day (A) 21 p.i. and (B) 28 p.i. (C) Quantification of spinal cord demyelination (control n=4 at day 21 p.i., n=9 at day 28 p.i.; PLX5622 n=5 at day 21p.i., n=12 at day 28 p.i.) reveals a significant increase after PLX5622 treatment compared to control-treated animals at 28 p.i. (D) qPCR for viral Matrix protein RNA transcripts in spinal cords of control and PLX5622-treated mice at day 28 p.i. (E) Representative EM images with magnification x1200 from control and PLX5622 mice spinal cord with demyelinated and remyelinated axon, vacuoles and damaged axons depicted. scatter plot (F) depicting individual g-ratio’s and (G) overall calculation of g-ratio of control and PLX5622, from lateral and ventral white matter columns of control and PLX5622 treated mice as a function of axon diameter. Representative images of H&E/LFB stained spinal cords of control and PLX5622-treated mice at day 28 p.i. with vacuoles indicated (H) and quantified (I). Data in C, G, and I are presented as average ± SEM. **p<0.01 and ***p<0.001.

Figure 5.. PLX5622 treatment alters expression of genes associated with remyelination and host defense.

JHMV-infected mice treated with either control chow or PLX5622 from day 14 p.i. were sacrificed at day 28 p.i. and spinal cords removed. Analysis of expression of transcripts Lpl and Cst7 within spinal cords of control and PLX5622-treated mice was evaluated by (A) heatmap analysis and (B) quantification of expression levels of transcripts. Expression of MHC class I (H2-M3, H2-Q7) and MHC class II (H2-Aa, H2Ab1, H2-Dmb1) transcripts was assessed by (C) heatmap analysis as well aa quantification of transcript levels for (D) MHC class I and (E) MHC class II genes. Representative immunofluorescent staining for (F) MHC class I and (G) MHC class II proteins in spinal cords of control and PLX5622-treated mice. (H) quantification of MHC class I and MHC class II positive cells in control (n=3) and PLX5622 (n=3) treated mice. Data in panels B, D, E, and H are presented as average±SEM. *p<0.05, **p<0.01, ***p<0.001 and ****p<0.0001.