Genetic deletion of nitric oxide synthase 2 ameliorates Parkinson’s disease pathology and neuroinflammation in a transgenic mouse model of synucleinopathy

Abstract

Studies of mouse models of Alzheimer's disease (AD) have demonstrated that nitric oxide synthase 2 (NOS2) is involved in AD pathology. However, the effects of NOS2 on the pathology of Parkinson’s disease (PD) are not well studied. To address this gap, we examined the impact of NOS2 on disease-associated phenotypes in a mouse model of PD. Transgenic mice carrying the A53T mutation of α-synuclein (Syn^A53T) and newly generated double transgenic mice with deletion of NOS2 (Syn^A53T/NOS2^−/−) were used. Compared with Syn^A53T mice, the loss of nos2 decreased α-synuclein phosphorylation at serine 129 and reduced α-synuclein-induced microglial and astrocyte activation in Syn^A53T/NOS^−/− mice. Additionally, neuroinflammation-related gene clusters in the deep mesencephalic nucleus (DpMe) were altered in Syn^A53T/NOS^−/− mice compared with Syn^A53T mice. Taken together, our results suggest that deletion of nos2 alleviates α-synuclein pathology and α-synuclein-associated neuroinflammatory responses in the brain.

Main text

Nitric oxide (NO) is a bioactive free radical that is involved in various physiological and pathological processes in several organ systems and the central nervous system (CNS) [1]. In the brain, nitric oxide synthase 2 (NOS2) plays an important role in neurotransmission, neural development, and the immune defense response [2]. Interestingly, several recent studies have reported that NOS2 differentially regulates Alzheimer’s disease (AD) pathology. For instance, deletion of nos2 in mice results in the expression of mutant amyloid precursor protein (APP) and hyperphosphorylation of tau in the brain [3]. Compared with APPSwDI mice, APPSwDI/NOS2^−/− mice exhibit spatial memory impairment and tau pathology [4]. However, the effects of NOS2 on α-synuclein-induced Parkinson’s disease (PD) pathology remain unclear.

To address this gap, we generated Syn^A53T/NOS2^−/− mice by hybridizing human Syn^A53T-expressing transgenic mice and nos2 knockout (NOS2^−/−) mice. Generation of the Syn^A53T/NOS2^−/− mice was confirmed by RT–PCR, which failed to detect nos2 mRNA (Additional file 1: Fig. S1).

We then examined whether genetic deletion of nos2 affects α-synuclein-induced PD pathology. The brains of 10- to 11-month-old non-transgenic (nTg), Syn^A53T, and Syn^A53T/NOS2^−/− mice were subjected to immunofluorescence staining with an anti-p-Syn^ser129 antibody. Compared with nTg mice, p-Syn^ser129 levels in the substantia nigra (SN), deep mesencephalic reticular nucleus (DpMe), and granular insular cortex (Gi) were significantly higher in Syn^A53T mice (Fig. 1A, B). Importantly, p-Syn^ser129 levels in the SN, DpMe, and Gi were significantly lower in Syn^A53T/NOS2^−/− mice than in Syn^A53T mice (Fig. 1A, B). Moreover, p-Syn^ser129 levels in the cortex, caudate and putamen (CPu) and hippocampus (Hippo) were significantly reduced in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice (Additional file 1: Fig. S2A, B). These data suggest that genetic deletion of nos2 alleviates synucleinopathy in the brain.

Fig. 1

Deletion of nos2 alleviates synuclein pathology and neuroinflammatory responses in Syn^A53T/NOS2^−/− mice. A Immunofluorescence staining of the substantia nigra (SN), deep mesencephalic nucleus (DpMe), and granular insular cortex (Gi) of 10- to 11-month-old nTg, Syn^A53T, and Syn^A53T/NOS2^−/− mice with an anti-p-Syn^ser129 antibody. B Quantification of the data in A (SN, DpMe, Gi region; nTg: n = 17–20 brain slices/5 mice; Syn^A53T: n = 20–22 brain slices/5 mice; Syn^A53T/NOS2^−/−: n = 14–16 brain slices/4 mice). C Immunofluorescence staining of the SN, DpMe, and Gi of 10- to 11-month-old nTg, Syn^A53T, and Syn^A53T/NOS2^−/− mice with an anti-Iba-1 antibody. D Quantification of the data in C (SN, DpMe, Gi region; nTg: n = 20 brain slices/5 mice; Syn^A53T: n = 17–20 brain slices/5 mice; Syn^A53T/NOS2^−/−: n = 11–15 brain slices/4 mice). E Immunofluorescence staining of the SN, DpMe, and Gi of 10- to 11-month-old nTg, Syn^A53T, and Syn^A53T/NOS2^−/− mice with an anti-GFAP antibody. F Quantification of the data in E (SN, DpMe, Gi region; nTg: n = 20 brain slices/5 mice; Syn^A53T: n = 17–20 brain slices/5 mice; Syn^A53T/NOS2^−/−: n = 12–16 brain slices/4 mice). G Differentially expressed genes (DEGs) were identified by comparing Syn^A53T/NOS2^−/− mice with Syn^A53T mice and Syn^A53T mice with nTg mice. The numbers of DEGs in each comparison and the number of overlapping DEGs are indicated. H Six clusters (C1–6) of DEGs were identified from the two comparisons. The color bar represents the gradient of log₂ fold changes. The number of DEGs in each cluster is denoted in parentheses. I Cellular processes represented by the DEGs in C2. The x-axis is the −log₁₀(P), where P is the enrichment P value from ConsensusPathDB software. J DEGs involved in the inflammatory response (n = 3 mice/group). *p < 0.05, **p < 0.01, ***p < 0.001, scale bar = 100 μm

Since genetic deletion of nos2 diminished α-synuclein aggregation in the brain, we further investigated the impact of nos2 deletion on α-synuclein-induced glial activation. The brains of 10- to 11-month-old nTg, Syn^A53T, and Syn^A53T/NOS2^−/− mice were subjected to immunofluorescence staining with anti-Iba-1 and anti-GFAP antibodies. Compared with nTg mice, microglial/astrocyte fluorescence intensity, the number of Iba-1-positive cells, and the Iba-1/GFAP % area were increased in the DpMe and Gi but not in the SN in Syn^A53T mice (Fig. 1C, D and Additional file 1: Fig. S3). Importantly, Iba-1 fluorescence intensity, the number of Iba-1-positive cells, and the Iba-1-positive % area in the SN, DpMe, and Gi were significantly lower in Syn^A53T/NOS2^−/− mice than in Syn^A53T mice (Fig. 1C, D and Additional file 1: Fig. S3). Moreover, GFAP fluorescence intensity in the SN, DpMe, and Gi was significantly reduced in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice (Fig. 1E, F). The α-synuclein-induced number of GFAP-positive cells and GFAP-positive % area in the DpMe and Gi were significantly diminished in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice (Additional file 1: Fig. S3). In addition, Iba-1/GFAP fluorescence intensity, the number of Iba-1/GFAP-positive cells and the Iba-1/GFAP % area in the cortex, CPu, and hippocampus were significantly reduced in Syn^A53T/NOS2^−/− mice compared with nTg and Syn^A53T mice (Figs. S4-S5). Taken together, these data suggest that deletion of nos2 diminishes α-synuclein-stimulated microglial and astrocyte activation and that NOS2 is required for α-synuclein-mediated neuroinflammation in the brain.

To investigate the effects of nos2 deletion on gene expression in the mouse model of PD, we isolated the DpMe region (which exhibited the greatest regulatory effects of nos2) from 10- to 11-month-old nTg, Syn^A53T, and Syn^A53T/NOS2^−/− mice and conducted RNA sequencing. A total of 1,339 differentially expressed genes (DEGs) were identified in Syn^A53T versus nTg mice and Syn^A53T/NOS2^−/− versus Syn^A53T mice (744 and 788 DEGs, respectively) (Fig. 1G and Additional file 2: Table S1). Among the 1339 DEGs, 193 overlapped between the two comparisons (Fig. 1G). These results indicate that nos2 deletion significantly alters gene expression in this mouse model of PD (Additional file 3).

To systematically investigate the cellular processes affected by nos2 deletion, we classified the 1,339 DEGs into 6 clusters (C1-6) based on their differential expression in the two comparisons (Fig. 1H). C2 was upregulated in Syn^A53T mice compared with nTg mice but downregulated in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice. Thus, we focused on this cluster because it likely includes genes associated with the effects of NOS2 on PD pathology. The cellular processes represented by the DEGs in C2 were identified by gene set enrichment analysis using Consensus Path DB [5]. Interestingly, the DEGs in C2 were mainly involved in neuroinflammatory responses, glial cell proliferation, oxidative stress, and apoptosis (Fig. 1I). Notably, genes involved in neuroinflammatory response-related processes were strongly downregulated in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice (Fig. 1J).

In summary, α-synuclein phosphorylation, α-synuclein-induced neuroinflammation, and the expression of related genes were significantly suppressed in the brains of Syn^A53T/NOS2^−/− mice. Overall, our results suggest that NOS2 is a crucial regulator of the synucleinopathy and neuroinflammatory response associated with PD pathology.

A recent study demonstrated that NOS2 overexpression induces NO production and α-synuclein aggregation in PC12 neurons [6]. In SH-SY5Y cells, NOS2 expression induces the formation of cytotoxic nitrated α-synuclein [7]. However, the effects of nos2 deletion on α-synuclein pathology have not been investigated. The significant reduction in p-Syn^ser129 levels in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice suggests that decreasing NOS2 expression may help alleviate α-synucleinopathy in the brain.

Interestingly, several recent studies have shown that NOS2 regulates neuroinflammatory responses in the brain. For instance, the lipopolysaccharide (LPS)-induced increase in TNF-α levels is significantly reduced in nos2 knockout mice [8], and deletion of nos2 decreases the number of Iba-1/GFAP-positive cells in the brain compared with wild-type mice [9]. In addition, GFAP expression is diminished by one-third in NOS2^−/− mice compared with nTG mice [10]. In the present study, microglial and astrocyte activation in the brain, which are associated with severe synuclein pathology, were dramatically reduced in Syn^A53T/NOS2^−/− mice compared with Syn^A53T mice. It is possible that brain region-specific synuclein aggregation and pathology contribute to Iba-1/GFAP expression when nos2 is knocked out. Another possibility is that unknown synuclein pathology/NOS2-associated molecular targets contribute to glial hypoactivity/degradation when nos2 is deleted in vivo. Future studies will focus on identifying the molecules that contribute to glial inactivation and the amelioration of synuclein pathology when nos2 is deleted. Overall, the available data suggest that NOS2 has critical functions in the modulation of glial homeostasis in this mouse model of PD.

In conclusion, we generated Syn^A53T/NOS2^−/− mice for the first time by crossing human α-synuclein A53T mutant mice and nos2 knockout mice and found that α-synuclein pathology, neuroinflammatory responses, and neuroinflammation-associated gene expression were reduced in the double transgenic mice compared with Syn^A53T mice. Our data indicate that NOS2 may be a therapeutic target for modulating PD pathology in the brain.