Metformin normalizes mitochondrial function to delay astrocyte senescence in a mouse model of Parkinson's disease through Mfn2-cGAS signaling

Abstract

Background: Senescent astrocytes play crucial roles in age-associated neurodegenerative diseases, including Parkinson's disease (PD). Metformin, a drug widely used for treating diabetes, exerts longevity effects and neuroprotective activities. However, its effect on astrocyte senescence in PD remains to be defined.

Methods: Long culture-induced replicative senescence model and 1-methyl-4-phenylpyridinium/α-synuclein aggregate-induced premature senescence model, and a mouse model of PD were used to investigate the effect of metformin on astrocyte senescence in vivo and in vitro. Immunofluorescence staining and flow cytometric analyses were performed to evaluate the mitochondrial function. We stereotactically injected AAV carrying GFAP-promoter-cGAS-shRNA to mouse substantia nigra pars compacta regions to specifically reduce astrocytic cGAS expression to clarify the potential molecular mechanism by which metformin inhibited the astrocyte senescence in PD.

Results: We showed that metformin inhibited the astrocyte senescence in vitro and in PD mice. Mechanistically, metformin normalized mitochondrial function to reduce mitochondrial DNA release through mitofusin 2 (Mfn2), leading to inactivation of cGAS-STING, which delayed astrocyte senescence and prevented neurodegeneration. Mfn2 overexpression in astrocytes reversed the inhibitory role of metformin in cGAS-STING activation and astrocyte senescence. More importantly, metformin ameliorated dopamine neuron injury and behavioral deficits in mice by reducing the accumulation of senescent astrocytes via inhibition of astrocytic cGAS activation. Deletion of astrocytic cGAS abolished the suppressive effects of metformin on astrocyte senescence and neurodegeneration.

Conclusions: This work reveals that metformin delays astrocyte senescence via inhibiting astrocytic Mfn2-cGAS activation and suggest that metformin is a promising therapeutic agent for age-associated neurodegenerative diseases.

Keywords: Astrocyte senescence; Metformin; Mitofusin 2; Parkinson’s disease; cGAS-STING.

Fig. 1

Metformin delays astrocyte senescence in MPTP-induced PD model. A, B Movement distance within 5 min was recorded by open field test (n = 8 animals for each group). C the time taken to descend a pole (Time-total) was recorded in pole test (n = 8 animals for each group). D Time on the rod was measured by the rotarod test (n = 9 animals for each group). E Heatmap of relative mRNA levels of SASP as indicated in SNpc (n = 3 animals for each group). F–K qPCR measurement of the levels of IL-1α (F), IL-1β (G), MMP3 (H), MMP9 (I), IL-6 (J) and p16^Ink4a (K) in the SNpc (n = 6 animals for each group). L Representative double-immunostaining for lamin B1 and astrocytic marker GFAP, DA neuron marker TH or microglia marker IBA-1 in the SNpc. DAPI stains nucleus (blue). White arrow: high level of lamin B1, yellow arrow: low level of lamin B1. M–O Quantification of lamin B1 immunofluorescence intensity in GFAP⁺ astrocytes (M), TH⁺ DA neuron (N), and IBA-1⁺ microglia (O) in the SNpc (n = 6 animals for each group). P Quantification of p16 immunofluorescence intensity in GFAP⁺ astrocytes in the SNpc (n = 6 animals for each group). Q Representative double-immunostaining for p16 (red) and astrocytic marker GFAP (green) in the SNpc. DAPI stains nucleus (blue). The data shown are the mean ± SEM. One-way ANOVA with Tukey’s post-hoc tests were used. *p < 0.05, ***p < 0.001, NS: no significant

Fig. 2

Metformin suppresses senescence of astrocytes in vitro. A Astrocytes were pretreated with metformin at indicated concentrations for 30 min and then stimulated with α-Syn PFF. Representative immunoblots of relative expression of p16 in astrocytes. B Quantification of relative expression of p16 in A (Three independent experiments). C–F qPCR measurement of the levels of IL-1α (C), IL-1β (D), MMP3 (E), and MMP9 (F) in astrocytes treated with metformin (0.2 mM) and α-Syn PFF (Six independent experiments). G, H Representative images of SA-β-gal staining and quantification of the percentage of SA-β-gal⁺ astrocytes over total astrocytes in astrocytes treated with metformin (0.2 mM) and α-Syn PFF (Three independent experiments). I Astrocytes were pretreated with metformin (0.2 mM) for 30 min and then stimulated with MPP⁺. Representative immunoblots of relative expression of p16 in astrocytes. J Quantification of relative expression of p16 in I (Three independent experiments). K, L Representative images of SA-β-gal staining and quantification of the percentage of SA-β-gal⁺ astrocytes over total astrocytes in astrocytes treated with metformin (0.2 mM) and MPP⁺ (Three independent experiments). M Astrocytes were cultured for 7 days or 40 days with or without metformin (0.2 mM). Heatmap of relative mRNA levels of SASP as indicated in astrocytes. N–P qPCR measurement of IL-6 (N), MMP3 (O), and p16 (P) mRNA expression in astrocytes (Three independent experiments). Q, R Representative immunoblots and quantification of relative expression of p16 in astrocytes (Three independent experiments). S, T Representative images of SA-β-gal staining and quantification of the percentage of SA-β-gal⁺ astrocytes over total astrocytes in astrocytes (Three independent experiments). U, V Immunofluorescence and quantification of lamin B1 in GFAP⁺ astrocytes (Three independent experiments). DAPI stains nucleus (blue). The data shown are the mean ± SEM. One-way ANOVA with Tukey’s post-hoc tests were used. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 3

Metformin normalizes mitochondrial function in senescent astrocytes by Mfn2. A Astrocytes were pretreated with metformin (0.2 mM) for 30 min and then stimulated with MPP⁺. Representative images of JC-1 in astrocytes were analyzed using confocal microscopy (Three independent experiments). DAPI stains nucleus (blue). B, C Representative images and quantification of depolarized mitochondria analyzed by flow cytometry (Five independent experiments). D, E Representative images and quantification of mitochondrial ROS level analyzed by flow cytometry (Six independent experiments). F, G Representative images and quantification of mitochondrial ROS immunofluorescence intensity in immunofluorescence using ImageJ software (Six independent experiments). DAPI stains nucleus (blue). H–J qPCR measurement of mtDNA in cytosol of astrocytes pretreated with metformin (0.2 mM) and then stimulated with MPP⁺ (H), α-Syn PFF (I) or cultured for 40 days (J) (Three independent experiments). K, L Representative immunoblots and quantification of relative expression of Mfn2 in astrocytes treated with metformin (0.2 mM) and MPP⁺ (Three independent experiments). M, N Astrocytes were transfected with empty vector (pHBLV-CMV-MCS-3FLAG-EF1-ZsGreen-T2A-PURO) or Mfn2 plasmid for 48 h and then treated with metformin (0.2 mM) and MPP⁺. Representative images and quantification of mitochondrial ROS level analyzed by flow cytometry (Five independent experiments). O, P Representative images and quantification of dysfunctional mitochondria analyzed by flow cytometry (Four independent experiments). Q–S qPCR measurement of mtDNA in cytosol of astrocytes stimulated with MPP⁺ (Q), α-Syn PFF (R) or cultured for 40 days (S) (Three independent experiments). The data shown are the mean ± SEM. One-way ANOVA with Tukey’s post-hoc tests were used (C, E, F, H–J, L). Two-way ANOVA with Tukey’s post-hoc tests were used (N, O, Q–S). **p < 0.01, ***p < 0.001, NS: no significant

Fig. 4

Metformin inactivates the cGAS-STING signal in senescent astrocytes in vitro and in vivo. A–C Astrocytes were pretreated with metformin at indicated concentrations for 30 min and then stimulated with α-Syn PFF. Representative immunoblots (A) and quantification of relative expression of cGAS (B) and p-STING (C) in astrocytes (Three independent experiments). D–F, Astrocytes were pretreated with metformin (0.2 mM) for 30 min and then stimulated with MPP⁺. Representative immunoblots (D) and quantification of relative expression of cGAS (E) and p-STING (F) in astrocytes (Three independent experiments). G–I, Astrocytes were pretreated with metformin (0.2 mM) and then cultured for 40 days. Representative immunoblots (G) and quantification of relative expression of cGAS (H) and p-STING (I) in astrocytes (Three independent experiments). J–M Representative immunoblots (J) and quantification of relative expression of cGAS (K), p-STING (L) and p16 (M) in the SNpc from MPTP-treated mice (n = 3 animals for each group). N Representative double-immunostaining for cGAS (red) and astrocytic marker GFAP (green) in the SNpc. DAPI stains nucleus (blue). O Quantification of the cGAS level in GFAP⁺ astrocytes from SNpc (n = 6 animals for each group). The data shown are the mean ± SEM. One-way ANOVA with Tukey’s post-hoc tests were used. *p < 0.05, **p < 0.01, ***p < 0.001

Fig. 5

Metformin delays astrocytes senescence by blocking astrocytic Mfn2-cGAS. A–D Astrocytes were transfected with vector or Mfn2 plasmid for 48 h and then treated with metformin (0.2 mM) and MPP⁺. qPCR measurement of the levels of IL-1α (A), IL-1β (B), IL-6 (C), and MMP3 (D) in astrocytes (Four independent experiments). E, F Representative images of SA-β-gal staining and quantification of the percentage of SA-β-gal⁺ astrocytes over total astrocytes in astrocytes (Three independent experiments). DAPI stains nucleus (blue). G–J Representative immunoblots (G) and quantification of relative expression of p16 (H), cGAS (I) and p-STING (J) in astrocytes (Three independent experiments). K Heatmap of relative mRNA levels of SASP as indicated in SNpc (n = 3 animals for each group). L–Q qPCR measurement of the levels of IL-1α (L), IL-1β (M), IL-6 (N), MMP3 (O), MMP9 (P), and p16^Ink4a (Q) in the SNpc (n = 6 animals for each group). R Representative double-immunostaining for lamin B1 and astrocytic marker GFAP in the SNpc. DAPI stains nucleus (blue). S Quantification of lamin B1 immunofluorescence intensity in GFAP⁺ astrocytes in the SNpc (n = 6 animals for each group). The data shown are the mean ± SEM. Two-way ANOVA with Tukey’s post-hoc tests were used. **p < 0.01, ***p < 0.001, NS: no significant

Fig. 6

Metformin ameliorates PD-like pathology in MPTP-induced PD model via delay of cGAS-mediated astrocytes senescence. A Mesencephalic primary neurons were treated with astrocytic conditioned medium (ACM) from astrocytes transfected with Mfn2 plasmid and then stimulated with metformin (0.2 mM) and MPP⁺. Representative pictures of MAP2 (green) immunostaining. DAPI stains nucleus (blue). B, C Quantification of relative the number of neurons (B) and mean total neuritis length (C, Three independent experiments). D Mesencephalic primary neurons were treated with astrocytic conditioned medium (ACM) from astrocytes transfected with cGAS siRNA (si-cGAS) and then stimulated with metformin (0.2 mM) and MPP⁺. Representative pictures of MAP2 (green) immunostaining. DAPI stains nucleus (blue). E, F Quantification of relative the number of neurons (E) and mean total neuritis length (F, four independent experiments). G, H the time taken to turn around (Time-turn) and descend a pole (Time-total) was recorded in pole test (n = 9–12 animals for each group). I Time on the rod was measured by the rotarod test (n = 10–13 animals for each group). J, K Movement distance within 5 min was recorded by open field test (n = 10–13 animals for each group). L, M Representative immunoblots (L) and quantification of relative expression of TH (M) in the SNpc (n = 5 animals for each group). N Microphotographs of TH-positive neurons in the SNpc. O Stereological counts of TH-positive neurons in the SNpc (n = 6 animals for each group). P Proposed model depicting metformin inhibits astrocyte senescence via Mfn2-mtDNA-cGAS-STUNG signal in PD. The data shown are the mean ± SEM. Two-way ANOVA with Tukey’s post-hoc tests were used. **p < 0.01, ***p < 0.001, NS: no significant