Age increases reactive oxygen species production in macrophages and potentiates oxidative damage after spinal cord injury

Abstract

Age potentiates neurodegeneration and impairs recovery from spinal cord injury (SCI). Previously, we observed that age alters the balance of destructive (M1) and protective (M2) macrophages; however, the age-related pathophysiology in SCI is poorly understood. Nicotinamide adenine dinucleotide phosphate oxidase (NOX) contributes to reactive oxygen species (ROS)-mediated damage and macrophage activation in neurotrauma. Further, NOX and ROS increase with central nervous system age. Here, we found significantly higher ROS generation in 14 versus 4-month-old (MO) mice after contusion SCI. Notably, NOX2 increased in 14 MO ROS-producing macrophages suggesting that macrophages and NOX contribute to SCI oxidative stress. Indicators of lipid peroxidation, a downstream cytotoxic effect of ROS accumulation, were significantly higher in 14 versus 4 MO SCI mice. We also detected a higher percentage of ROS-producing M2 (Arginase-1-positive) macrophages in 14 versus 4 MO mice, a previously unreported SCI phenotype, and increased M1 (CD16/32-positive) macrophages with age. Thus, NOX and ROS are age-related mediators of SCI pathophysiology and normally protective M2 macrophages may potentiate secondary injury through ROS generation in the aged injured spinal cord.

Keywords: Aging; Arginase-1; Dihydroethidium; Macrophage polarization; Microglia; gp91(phox).

Fig. 1. Reactive oxygen species (ROS) production is higher after 14 vs. 4 MO SCI

Representative images of spinal cord sections at the lesion epicenter stained with the superoxide-sensitive dye DHE from 4 and 14 MO (month old) mice at 3 (A-B), 7 (C-D), and 14 (E-F) days post injury (dpi). Quantification of oxidized DHE fluorescence labeling (red) reveals significantly higher ROS production in 14 vs. 4 MO injured spinal cords at 3 (G) and 7 (H) dpi. Scale bar= 100 μm. Results are mean +/− SEM, n=4-5/group. *p<0.05.

Fig. 2. Lipid peroxidation is increased after 14 vs. 4 MO SCI

Representative images and densitometry quantification of immunoblotting of 4-HNE (A-A’). ELISA quantification of 4-HNE adducts in injured spinal cords of 4 and 14 MO mice (B). Cross sections at the lesion epicenter were labeled with anti-4-HNE antibody. Representative images and quantification of 4-HNE immunoreactivity of 4 MO and 14 MO at 3 (C-C”), 7 (D-D”) and 14 (E-E”) dpi. 4-HNE immunoreactivity significantly increased in lesion epicenter of 14 vs. 4 MO SCI mice at 7 dpi. Results are mean +/− SEM, n=4-5/group. *p<0.05. scale bar= 100 μm.

Fig. 3. ROS are primarily detected in the macrophages/microglia following SCI

Superoxide generation was detected by oxidized-DHE (ox-DHE; A, C & E;). Representative confocal images show colocalization of ox-DHE with TomL-positive (A-B; arrowheads), NeuN-positive (C-D; arrowheads), and GFAP-positive (E-F; arrowheads) cells in the lesion epicenter from 4 MO mice at 3 dpi. Notice the high degree of red-blue overlap in B vs. D and F. The percentage of ROS production by different cell types at 3 dpi (G) and 7 dpi (H) was quantified at the three continuous sections at the lesion epicenter. (G) Macrophage/microglia (TomL-positive cells) accounts for =80% of ROS production in the lesion epicenter at 3 dpi, while astrocytes (GFAP-positive cells) = 5%, and neurons (NeuN-positive cells)=10%. Other ROS-producing cells (4%) were not phenotyped. See Supplementary Fig. 2 for the specificity of ox-DHE double labeling with cellular markers. (H) Macrophage/microglia account ~50% of ROS production at 7 dpi, while astrocytes (GFAP-positive cells) = 23%, neurons (NeuN-positive cells)=6.3%, and other non-phenotyped=23%. There was no observable difference in the cellular distribution of DHE between 4 and 14 MO after SCI, n=4-5/group. Scale bar= 10 μm.

Fig. 4. NOX2 activity is increased in 14 vs. 4 MO spinal cords after SCI

Representative images of the lesion epicenter from 4 and 14 MO mice at 3 dpi stained with gp91^phox (A&E, green), DHE (B&F, red), and TomL (C&G, blue). NOX2 activation was confirmed through co-labeling of gp91^phox and DHE. The majority of activated NOX2 was detected in TomL-positive macrophages (D&H, arrowheads). (I) NOX2 activation is significantly upregulated in 14 MO as compared to 4 MO SCI. Results are mean +/− SEM, n=4-5/group. *p<0.05. Scale bar= 20 μm.

Fig. 5. Age increases ROS-production in M2 SCI macrophages

Sections from the lesion epicenter were immunolabeled with anti-Arg-1 antibody (green; D, G, J&M) and DHE (red; E, H, K&N). (A) Arg-1-positive staining is not significant different between 4 and 14 MO mice at either 3 or 7 dpi. High-powered confocal images reveal significantly more ARG-1-positive macrophages (green) producing ROS after 14 vs. 4 MO SCI at both 3 (B, F&I, arrowheads) and 7 (C, L&O, arrowheads) dpi. The percentage of ARG-1-positive macrophages expressing ROS was quantified by co-labeling of DHE and ARG-1. Results are mean +/− SEM, n=4-5/group. *p<0.05. Scale bar= 10 μm.

Fig. 6. Age enhances NOX2-mediated oxidative stress following SCI

(1) Spinal cord injury triggers macrophage NOX2 enzyme activity, a primary cellular and subcellular source of reactive oxygen species (ROS). We detected significantly increased NOX2 activation with age after SCI. (2) This further gives rise to lipid peroxidation and resulting aldehyde formation, such as 4-hydroxynonenal (4-HNE), and subsequent secondary injury. Our results of molecular and histochemical analyses of 4-HNE indicate that age potentiates this ROS-induced oxidative damage in injured spinal cord. (3) Age also potentiated M1 activation (CD16/32) and increased ROS production in normally protective Arginase-1 (Arg-1)-positive M2 macrophages. (4) Lipid peroxidation may also facilitate an M2 to M1 conversion thereby further increasing age-related, macrophage-mediated SCI tissue damage.