Macrophage Transcriptional Profile Identifies Lipid Catabolic Pathways That Can Be Therapeutically Targeted after Spinal Cord Injury

Abstract

Although infiltrating macrophages influence many pathological processes after spinal cord injury (SCI), the intrinsic molecular mechanisms that regulate their function are poorly understood. A major hurdle has been dissecting macrophage-specific functions from those in other cell types as well as understanding how their functions change over time. Therefore, we used the RiboTag method to obtain macrophage-specific mRNA directly from the injured spinal cord in mice and performed RNA sequencing to investigate their transcriptional profile. Our data show that at 7 d after SCI, macrophages are best described as foam cells, with lipid catabolism representing the main biological process, and canonical nuclear receptor pathways as their potential mediators. Genetic deletion of a lipoprotein receptor, CD36, reduces macrophage lipid content and improves lesion size and locomotor recovery. Therefore, we report the first macrophage-specific transcriptional profile after SCI and highlight the lipid catabolic pathway as an important macrophage function that can be therapeutically targeted after SCI.SIGNIFICANCE STATEMENT The intrinsic molecular mechanisms that regulate macrophage function after spinal cord injury (SCI) are poorly understood. We obtained macrophage-specific mRNA directly from the injured spinal cord and performed RNA sequencing to investigate their transcriptional profile. Our data show that at 7 d after SCI, macrophages are best described as foam cells, with lipid catabolism representing the main biological process and canonical nuclear receptor pathways as their potential mediators. Genetic deletion of a lipoprotein receptor, CD36, reduces macrophage lipid content and improves lesion size and locomotor recovery. Therefore, we report the first macrophage-specific transcriptional profile after SCI and highlight the lipid catabolic pathway as an important macrophage function that can be therapeutically targeted after SCI.

Keywords: axon regeneration; fibrotic scar; foamy macrophages; glial scar; myelin laden macrophages; neuroinflammation.

Figure 1.

Isolation of macrophage-specific mRNA from the spinal cord injury site. A–C, In uninjured spinal cord, hemagglutinin (green) is expressed in a minor population of Iba1⁺ microglia and dendritic cells (red). D–F, At 7 d after SCI, HA is expressed in the majority of myeloid cells at the injury core. Immunohistochemistry was performed on lysM^Rpl22(HA) mice. All images are sagittal sections of the midthoracic spinal cord. Scale bar, 500 μm. n = 3 biological replicates per group.

Figure 2.

Myeloid cells labeled in lysM-Cre mice are representative of the general macrophage population after SCI. A–T, We performed SCI on lysM^tdTom→WT chimera mice and assessed the injury site using flow cytometry to determine whether macrophages labeled in lysM^tdTom mice (CD11b⁺/tdTom⁺; B, red) were representative of the general macrophage population in the injury site (CD11b⁺/CD45^hi; A, circled region). Except for a slight difference in Ly6C⁺ cells at 14 d after SCI, macrophages labeled in lysM^tdTom (red bars) mice did not differ from the general macrophage population (gray bars) in the percentage of cells that expressed Ly6G, CD11c, CD86, CD206, or CD43 at both 7 and 14 d after SCI. *p < 0.05 using Student's unpaired t test. n = 5 biological replicates per group. Error bars indicate SEM.

Figure 3.

Macrophages at 3 and 7 d after SCI display distinct clustering and different expression of highly enriched genes. A, The 3 d (n = 3) and 7 d (n = 3) samples showed a range of 80–107 million 100 base pair sequence reads with 84–90% alignment. B, C, Both principal component analysis of all genes (B) and hierarchical clustering of differentially expressed genes (C) showed small intragroup differences and distinct separation of the 3 d and 7 d macrophage data sets. D, Venn diagram displaying the number of same or different genes from the top 20% of the highest expressed genes (2416 genes with highest FPKM values) in 3 and 7 d macrophages. From the top 20% expressed genes represented in D, the 25 genes with the highest FPKM values are listed in E and F. n is the number of biological replicates. Error bars indicate SEM.

Figure 4.

The most differentially expressed genes between macrophages at 3 and 7 d after SCI. A rank order list of 100 genes that were expressed higher in 7 or 3 d macrophages is shown. Note that although ctss, ctsd, ctsb, apoe, psap, and gpnmb are also very highly expressed at 3 d, they are not included in the graph because their differential expression could not be mathematically calculated (results in infinity).

Figure 5.

GO analysis shows enrichment of genes related to cell migration and cytokine signaling in 3 d macrophages. A, B, Many of the enriched biological processes are related to cell migration (A), while cytokine signaling represent the top two enriched pathways (B). C, Consistent with cell migration in response to cytokine signaling, many of the enriched cellular components are related to the cell membrane and extracellular matrix. D, E, Top differentially expressed genes (compared to 7 d) that are related to cytokine signaling (D) or cell adhesion (E). ARVC, Arrhythmogenic right ventricular cardiomyopathy. Differentially expressed genes (fold change > 2, FDR-adjusted p < 0.05) were analyzed using the DAVID bioinformatics tools with the following statistical testing cutoffs: p (Expression Analysis Systematic Explorer [EASE] score) < 0.05; FDR < 0.1; percentage of genes involved in a given term, >1%. *Migratory behavior refers to the GO term “locomotory behavior” (GO:0007626). “Migratory behavior” was used instead to avoid confusion with locomotor recovery after SCI.

Figure 6.

A, GO analysis shows enrichment of genes related to lipid catabolism and immune response in 7 d macrophages. Most of the enriched biological processes are related to lipid catabolism with some representation of immune response processes. B, C, Enrichment of lipid catabolic processes is consistent with enrichment of lysosome pathway (B) as well as lysosomal cellular components (C). D, E, Top differentially expressed genes (compared to 3 d) that are related to lipid metabolism processes (D) or immune response processes (E). Differentially expressed genes (fold change > 2, FDR-adjusted p < 0.05) were analyzed using the DAVID bioinformatics tools with the following statistical testing cutoffs: p (EASE score) < 0.05; FDR < 0.1; percentage of genes involved in a given term, >1%.

Figure 7.

Pathway analysis identifies canonical lipid metabolism pathways with distinct network hubs. Molecular network generated by ingenuity pathway analysis using 7 d macrophage genes represented in lipid catabolism GO biological processes. Direct interactions are represented by continuous lines, while indirect interactions are represented by dashed lines. The color intensity of a molecule indicates the degree of increased (red) or decreased (green) enrichment of the gene compared to 3 d macrophages. Other colors indicate the presence (gray) or absence (white) of a given gene in the data set. The network forms hubs around cd36, lpl, pparg, abca1, and tnf (blue dotted circles). LXR/RXR and PPARα/RXRα canonical pathways (CPs) are the most highly enriched pathways identified by the network analysis.

Figure 8.

qPCR validation of select differentially expressed genes in 3 and 7 d macrophages. Total RNA isolated from the entire injury site at 3 or 7 d after SCI in C57BL/6 wild-type mice. Values are normalized to average 3 d expression levels. Error bars indicate SEM. *p < 0.05 compared to 7 d (unpaired two-tailed Student's t test). n = 5 per group (biological replicates).

Figure 9.

Foam cell genes are enriched in 7 d macrophages. GSEA comparing all genes in our data set to previously published data set on non-foam cells (614 genes) and foam cells (210 genes; Spann et al., 2012) is shown. A–H, Non–foam cell genes were overrepresented in our 3 d macrophages (A), while foam cell genes were overrepresented in our 7 d macrophages (B). This was consistent with the absence of macrophage (Iba1) lipid droplets (BODIPY; green) at 3 d (C–E; n = 7) and their presence at 7 d after SCI (F–H; n = 7). Each black vertical line in A and B represents a macrophage gene in our data set and shows a gene's relative location in the ranking list generated according to FPKM values. Statistical tests are shown in Table 1. Insets in D and G are magnified images of dotted regions. Bright green dots are lipid droplets stained by BODIPY. Scale bars: 500 μm. n indicates the number of biological replicates.

Figure 10.

ssGSEA analysis comparing 7 d macrophages to other macrophage treatment conditions. The 7 d macrophages are most enriched in foam cell genes compared to genes expressed under 18 other macrophage treatment conditions. The foam cell data set here is from Thomas et al. (2015), whereas the data set in Figure 9 is from Spann et al. (2012). ¹Thomas et al. (2015); ²Haribhai et al. (2016); ³El Chartouni and Rehli (2010); ⁴Riquelme et al. (2013); ⁵Li et al. (2015); ⁶Goo et al. (2016); ⁷Robblee et al. (2016); ⁸Suzuki et al. (2010); ⁹MacKenzie et al. (2013).

Figure 11.

CD36 expression is increased in 7 d macrophages and regulates the amount of lipid droplets in macrophages after SCI. A–O, While there is very little CD36 expression at 3 d after SCI (A–D; n = 7), it is significantly increased in macrophages (Iba1) at 7 d (E–H; n = 7). Genetic deletion of CD36 (L–O; n = 5) resulted in less large lipid droplets (BODIPY) compared to WTs (I–K, O; n = 6). B–D represent dotted region in A, and F–H represent dotted region in E. *p < 0.05 compared to WT (two-way ANOVA with Bonferroni posttest). DAPI in blue in D, H. Scale bars: 50 μm. Error bars indicate SEM. n is the number of biological replicates.

Figure 12.

Genetic deletion of CD36 results in smaller lesion size and improved locomotion. A–C, At 4 weeks after SCI, CD36 KOs display significantly smaller lesion size (B, C, n = 10) compared to WT mice (A, C, n = 9). Lesion size was defined as GFAP-negative area (GFAP in green; A, B). D, CD36 KO mice (n = 10) also displayed better locomotor recovery compared to WT mice (n = 9) as assessed by the BMS score. Data are combined results from two independent experiments that each showed similar results. *p < 0.05 compared to WT (C, unpaired two-tailed Student's t test; D, two-way repeated measures ANOVA with Bonferroni posttest). Scale bar, 500 μm. Error bars indicate SEM. n is the number of biological replicates.