NADPH oxidase and aging drive microglial activation, oxidative stress, and dopaminergic neurodegeneration following systemic LPS administration

Abstract

Parkinson's disease is characterized by a progressive degeneration of substantia nigra (SN) dopaminergic neurons with age. We previously found that a single systemic lipopolysaccharide (LPS, 5 mg/kg, i.p.) injection caused a slow progressive loss of tyrosine hydroxylase immunoreactive (TH+IR) neurons in SN associated with increasing motor dysfunction. In this study, we investigated the role of NADPH oxidase (NOX) in inflammation-mediated SN neurotoxicity. A comparison of control (NOX2(+/+) ) mice with NOX subunit gp91(phox) -deficient (NOX2(-/-) ) mice 10 months after LPS administration (5 mg/kg, i.p.) resulted in a 39% (P < 0.01) loss of TH+IR neurons in NOX2(+/+) mice, whereas NOX2(-/-) mice did not show a significant decrease. Microglia (Iba1+IR) showed morphological activation in NOX2(+/+) mice, but not in NOX2(-/-) mice at 1 hr. Treatment of NOX2(+/+) mice with LPS resulted in a 12-fold increase in NOX2 mRNA in midbrain and 5.5-6.5-fold increases in NOX2 protein (+IR) in SN compared with the saline controls. Brain reactive oxygen species (ROS), determined using diphenyliodonium histochemistry, was increased by LPS in SN between 1 hr and 20 months. Diphenyliodonium (DPI), an NOX inhibitor, blocked LPS-induced activation of microglia and production of ROS, TNFα, IL-1β, and MCP-1. Although LPS increased microglial activation and ROS at all ages studied, saline control NOX2(+/+) mice showed age-related increases in microglial activation, NOX, and ROS levels at 12 and 22 months of age. Together, these results suggest that NOX contributes to persistent microglial activation, ROS production, and dopaminergic neurodegeneration that persist and continue to increase with age.

Fig. 1

NOX2-deficient mice are more resistant to systemic LPS-induced neurotoxicity. Eight-week-old male C57BL/6 (NOX2^+/+) and Cybb (NOX2^−/−) mice were intraperitoneally injected with a single dose of LPS (5 mg/kg, i.p.) or saline and then maintained under normal conditions. Mice were sacrificed and brain sections (35 μm) that encompass the entire substantia nigra (SN) were collected 10 months after LPS or saline injection. After immunostaining with TH antibody, the number of TH+IR neurons in the substantia nigra pars compacta (SNpc) was counted as described in methods. (A) Visualization of TH+IR neurons in the substantia nigra (SN) of saline and LPS-treated animals. Scale bar=200 μM. (B) Number of TH+IR neurons in the SN of saline and LPS-treated NOX2^+/+ and NOX2^−/− mice. Systemic LPS injection caused a greater loss of TH+IR neurons in the SN of NOX2^+/+ mice than that of NOX2^−/− mice, compared with the corresponding saline controls. ** P < 0.01, compared with the corresponding saline controls.

Fig. 2

Immunocytochemical analysis of microglia. NOX2^+/+ and NOX2^−/− mice were sacrificed 1 hr following saline or LPS (5 mg/kg, i.p.). (A) Iba1 mRNA in midbrain was determined by real-time PCR. Systemic LPS treatment significantly upregulated Iba1 gene expression in NOX2^+/+ mice, but not in NOX2^−/− mice. (B) Brain sections were immunostained with an antibody against Iba1, a specific microglial marker. Activated microglia in substantia nigra were shown by increased cell size, irregular shape and intensified Iba1 staining in LPS-treated NOX2^+/+ mouse brains. The images presented are representative of three independent experiments. Scale bar=50μM.

Fig. 3

Time course of microglial response to systemic LPS treatment. Male C57BL/6 mice were injected with saline or LPS (5 mg/kg, i.p.) and sacrificed at the time points indicated. Brain sections were immunostained with Iba1 microglial antibody. (A) Quantification of activated Iba1+IR cells in substantia nigra. LPS groups at all the time points studied increased the fraction of activated microglia with a peak at 3 hr. Age-related percent activated Iba1+IR cells were significantly increased at 2, 7 and 10 months after saline injection (4, 9 and 12 months of age). (B) Representative images from saline and LPS-treated mice. LPS groups showed persistent activation of microglia with the largest morphological changes at 3 hr. In the saline control groups, microglia at 1, 3, and 24 hr have a resting morphology: small cell bodies with thin, highly ramified processes. However, at 2, 7 and 10 months some microglia were activated, large cell bodies, irregular shape and intensified Iba1 staining. *P<0.05, **P<0.01, compared with the corresponding saline control group. ^##P<0.01, compared with 1h saline group. Scale bar=50μm.

Fig. 4

Systemic LPS increases brain NOX2 expression. Male C57BL/6 mice were injected with LPS (5 mg/kg, i.p.) or saline and sacrificed at 24 hr, 10 and 20 months following LPS treatment. (A) NOX2 gene expression in the midbrain was determined by real-time PCR. NOX2 mRNA was significantly increased 24 hr after LPS treatment. (B) Brain sections were immunostained with monoclonal mouse NOX2 antibody, which did not stain NOX2^−/− mouse brain. The level of NOX2 immunoreactivity in SN was quantified by BioQuant image analysis system. Systemic LPS significantly enhanced NOX2 immunoreactivity at 24 hr, 10 and 20 months. (C) The images shown are representative ofNOX2+IR cells from control and LPS groups. Age-related NOX2 up-regulation was observed at 10 and 20 months in saline control mice. Scale bar=30μm. **P < 0.01, compared with the corresponding saline control mice, ^#P<0.05, ^##P<0.01, compared with 24 hr saline control mice.

Fig. 5

Systemic LPS treatment induces ROS production at 1 and 24 hr and remains elevated at 10 and 20 months. Mice were injected with hydroethidine (10 mg/kg, i.p.) 0.5hr, 23.5 hr, 10 months and 20 months after a single i.p. injection of LPS. Brains were harvested 30 min later and frozen sections (15μm) were examined for hydroethidine oxidation product, ethidium accumulation, by fluorescence microscopy. (A) Level of fluorescence intensity of ethidium was quantified by BioQuant image analysis software. (B) Images of ethidium fluorescence. Systemic LPS treatment significantly induced O₂⁻ and O₂⁻-derived oxidant production (ROS) 1 and 24hr and remains elevated 10 and 20 months after LPS injection, compared with the corresponding saline controls. Age-related ROS production was enhanced at 10 and 20 months in saline control mice, ** P < 0.01, compared with the corresponding saline control group. ## P< 0.01, compared with 1 hr saline group. Scale bar=200 μM.

Fig. 6

Colocalization of NOX2 expression and ROS production. Mice were injected with hydroethidine (10 mg/kg, i.p.) 10 months after a single dose of LPS i.p. injection. Brains were harvested 30 min later and frozen sections (15 μm) were double-labeled for NOX2 (blue) and Iba1 (green), NOX2 (blue) and TH (green) as well as NOX2 (blue) and GFAP (green) to analyze triple labeling or colabeling by using the Leica SP2 LCS confocal software. Confocal microscopy shows that NOX2+IR cells and ROS are triple-labeled with TH (A-merged, the right lower panel and B-merged, the left panel) or Iba1 (B-merged, the middle panel) in white with arrows, indicating NADPH oxidase (NOX) subunit NOX2 (gp91^phox) activation and ROS production predominantly occurred in DA neurons and microglial cells. Conversely, NOX2+IR cells and ROS are little glial fibrillary acidic protein (GFAP) positive. Double-labeled representative images are shown in pink with arrow heads indicating the colabeling of NOX2 with ROS (A-merged, the right lower panel; B-merged, the right panel). In Fig. 6A, Scale bar=30μm. In Fig. 6B, Scale bar=20μm.

Fig. 7

Diphenyleneiodonium (DPI) blocks LPS-induced microglial activation. Male C57BL/6 mice were treated with LPS (5 mg/kg, i.p.) or saline. DPI (3 mg/kg) was injected intraperitoneally on two consecutive days 2 months after LPS treatment. Mice were sacrificed 3 hr after the last dose of DPI. Brain sections were stained with Iba1 antibody. Systemic LPS markedly caused microglial activation. In the saline or DPI treated mice, most of the microglia were in a resting morphological shape. Iba1+IR cells in LPS-treated mouse brains were activated as shown by increased cell size, irregular shape, and intensified Iba1 staining. DPI blocked LPS-induced microglial activation shown by a resting morphological shape. Scale bar=50μm.

Fig. 8

Diphenyleneiodonium (DPI) inhibits LPS-induced production of ROS, TNFα, IL-1β, and MCP-1. Male C57BL/6 mice were treated with LPS (5 mg/kg, i.p.) or saline. DPI (3 mg/kg) was applied intraperitoneally on two consecutive days 2 months after LPS treatment. Mice were sacrificed 3 hr after the last dose of DPI. Brain O₂⁻ and O₂⁻-derived oxidants (ROS), TNFα, IL-1β, and MCP-1 were measured as described in materials and methods. (A) Representative images of O₂⁻ and O₂⁻-derived oxidants in SN. DPI significantly reduced LPS-induced production of O₂⁻ and O₂⁻-derived oxidants. Scale bar=30μm. (B) Systemic LPS treatment significantly increased production of TNFα, IL-1β, and MCP-1 compared with saline controls. DPI significantly reduced LPS-induced increases in production of TNFα, IL-1β, and MCP-1. * P < 0.05, compared with the saline control mice, ^#P < 0.05, compared with the LPS-treated mice.

Fig. 9

NOX-ROS is a key signaling in systemic LPS-induced dopaminergic neurodegeneration. LPS as a pro-inflammatory trigger activates microglia to release neurotoxic factors, such as TNFα, IL-1β, MCP-1, and ROS (O₂⁻). Among these pro-inflammatory factors, ROS have been implicated as key mechanisms of LPS neurotoxicity. Further, damaged or dying neurons have the potential to prime microglia to become more sensitive to additional stimuli and result in an exaggerated and prolonged proinflammatory response that enhances neuronal damage (i.e. reactive microgliosis). NOX2-deficient (NOX2^−/−) mice showed reduced DA neurotoxicity and decreased microglial activation. Blockade of NOX with DPI inhibits activation of microglia and production of ROS, TNFα, IL-1β and MCP-1. These results suggest that NOX and aging contribute to systemic LPS-elicited microglial activation, oxidative stress and dopaminergic neurodegeneration.