LRRK2 Gly2019Ser Mutation Promotes ER Stress via Interacting with THBS1/TGF-β1 in Parkinson's Disease

Abstract

The gene mutations of LRRK2, which encodes leucine-rich repeat kinase 2 (LRRK2), are associated with one of the most prevalent monogenic forms of Parkinson's disease (PD). However, the potential effectors of the Gly2019Ser (G2019S) mutation remain unknown. In this study, the authors investigate the effects of LRRK2 G2019S on endoplasmic reticulum (ER) stress in induced pluripotent stem cell (iPSC)-induced dopamine neurons and explore potential therapeutic targets in mice model. These findings demonstrate that LRRK2 G2019S significantly promotes ER stress in neurons and mice. Interestingly, inhibiting LRRK2 activity can ameliorate ER stress induced by the mutation. Moreover, LRRK2 mutation can induce ER stress by directly interacting with thrombospondin-1/transforming growth factor beta1 (THBS1/TGF-β1). Inhibition of LRRK2 kinase activity can effectively suppress ER stress and the expression of THBS1/TGF-β1. Knocking down THBS1 can rescue ER stress by interacting with TGF-β1 and behavior burden caused by the LRRK2 mutation, while suppression of TGF-β1 has a similar effect. Overall, it is demonstrated that the LRRK2 mutation promotes ER stress by directly interacting with THBS1/TGF-β1, leading to neural death in PD. These findings provide valuable insights into the pathogenesis of PD, highlighting potential diagnostic markers and therapeutic targets.

Keywords: LRRK2 G2019S; Parkinson's disease; TGF-β1; THBS1; endoplasmic reticulum (ER) stress.

Figure 1

LRRK2 G2019S mutation promotes ER stress in iPSC‐induced DA neurons. A) Genomic deoxyribonucleic acid (DNA) sequencing reveals the presence of heterozygous G2019S mutation in two G2019S patient‐derived DA neurons. B) A confocal image confirmed the expression of TH supported by immunofluorescence. Green: anti‐TH (antibody labeling DA). Blue: DAPI. Scale bar, 100 µm. The protein expression of arginine‐rich, mutated in early‐stage tumors (ARMET), phospho‐protein kinase‐like ER kinase (p‐PERK), PERK, phospho‐inositol requiring enzyme 1 alpha (p‐IRE1α), and IRE1α were determined in triplicate by using C) western blot (WB) analysis and D) their relative expression was calculated. The intensity of the protein bands was normalized to glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH). Reverse transcription‐quantitative real‐time polymerase chain reaction (RT‐qPCR) analysis of E) ATF4, F) CHOP, G) RTN1A, and H) GRP78 mRNA expression and normalized to the expression of GAPDH. I,J) The DA cells were subjected to pre‐treatment with Tunicamycin at a concentration of 1 µm for various time intervals (0, 3, 6, 9, 12, 15, 18, 24, and 48 h), or with Tunicamycin for 24 h at different concentrations (0, 0.01, 0.1, 1, and 10 µm). Fluorescent spectrophotometer traces were recorded, displaying the changes in Mag‐Fluo‐4 acetoxymethyl (AM) fluorescence intensity in triglyceride (TG)‐treated and CDN 1163 (CDN)‐treated DA cells in comparison to control cells. K) Mitochondrial membrane depolarization in WT and LRRK2 G2019S DA cells was assessed by measuring the fluorescence intensity of tetramethylrhodamine ethyl ester (TMRE) dye (250 nm). L) Cell viability was assessed using the cell counting kit‐8 (CCK‐8) assay. M) Cell death was determined by lactate dehydrogenase (LDH) assay. The level of ROS was studied using N) flow cytometry analysis, and O) relative intensity of ROS was calculated. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way analysis of variance (ANOVA) followed by Tukey's multiple comparison tests in (D, E, F, G, H, K, L, M, O). The difference in folds is statistically significant. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001.

Figure 2

Inhibition the kinase of LRRK2 suppresses ER stress. The MLi‐2 compound was employed as a pharmacological inhibitor to target and inhibit the kinase activity of LRRK2. The protein levels of ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed in triplicate through A) WB analysis and B) their relative expression was calculated. The intensity of the protein bands was normalized to GAPDH. RT‐qPCR was performed to measure the levels of C) ATF4, D) CHOP, E) GRP78, and F) RTN1A. The resulting mRNA expression values were normalized to the expression of GAPDH. G,H) The DA cells were exposed to Tunicamycin at a concentration of 1 µm for varying time intervals ranging from 0 to 48 h. Additionally, another set of cells was treated with Tunicamycin for 24 h at different concentrations, including 0, 0.01, 0.1, 1, and 10 µm. Fluorescent spectrophotometer traces were recorded, displaying the changes in Mag‐Fluo‐4 AM fluorescence intensity in TG‐treated and CDN‐treated DA cells in comparison to control cells. I) Mitochondrial membrane depolarization in WT and LRRK2 G2019S DA cells was assessed by measuring the fluorescence intensity of TMRE dye at a concentration of 250 nm. J) Cell viability was assessed using the CCK‐8 assay. K) Cell death was determined by LDH assay. Flow cytometry analysis was utilized to study the level of L) ROS (L), and relative intensity of M) ROS was calculated. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison tests in (B, C, D, E, F, I, J, K, M). The difference in folds is statistically significant. ^*P < 0.05, ^**P < 0.01, ^*** P < 0.001.

Figure 3

LRRK2 interacts with THBS1. A) Volcano plots were utilized to visually present the DEGs. B) The heat maps of module correlations display samples and genes in each cell, along with P values and correlation coefficients. C) The BP category identifies gene list enrichments. As a background enrichment, all genes in the genome were employed. The enrichment factor is the ratio from observed counts to predicted counts. Terms were having a p‐value of <0.01, a minimum count of three, and an enrichment factor of >1.5 are aggregated and categorized into clusters based on membership commonality. D) KEGG biological pathways analysis of 607 involved in the mutation of LRRK2. E) The STRING web program created a PPI network for the hub genes to show the interaction between the proteins with the top hub genes. F) The relative hub genes levels of CD44, CTGF, THBS1, VEGFA, SPP1, epidermal growth factor (EGF), vascular cell adhesion molecule 1 (VCAM1), matrix metalloproteinase‐3 (MMP3), C‐X‐C chemokine receptor type 4 (CXCR4), and lysyl oxidase (LOX) in LRRK2 wild and mutation group. G) The relative gene expression levels of CD44, CTGF, THBS1, VEGFA, SPP1, EGF, VCAM1, MMP3, CXCR4, and LOX in LRRK2 were determined using TR‐qPCR in the wild and LRRK2 G2019S groups. The resulting mRNA expression values were normalized to the expression of GAPDH. H. The MLi‐2 compound was employed as a pharmacological inhibitor to target and inhibit the kinase activity of LRRK2. Then the relative gene expression levels of CD44, CTGF, THBS1, VEGFA, SPP1, EGF, VCAM1, MMP3, CXCR4, and LOX in LRRK2 were determined using TR‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. The protein levels of THBS1 and TGF‐β1 proteins were analyzed in triplicate through I) WB analysis and J) their relative expression was calculated. The MLi‐2 compound was employed as a pharmacological inhibitor to target and inhibit the kinase activity of LRRK2. The protein levels of THBS1 and TGF‐β1 proteins were analyzed in triplicate through K) WB analysis and L) their relative expression was calculated. M) Molecular docking was applied to verify the binding activity between LRRK2 and THBS1. The blue color represents the LRRK2 chain, and the green‐blue color represents the THBS1 chain. The blue dashed lines indicate hydrogen bonds, while the red dashed lines represent salt bridge interactions. The negatively charged ASP1274 in LRRK2 forms a salt bridge interaction with the positively charged LYS561 in the THBS1 protein. Additionally, several hydrogen bond interactions are formed between specific amino acids in LRRK2, including ASP1202, GLU1224, LYS1249, SER1228, ASP1274, GLN1182, GLU1882, GLN1879, TYR1894, LYS1906, ASN1909, and HIS1911, and specific amino acids in THBS1, including LYS571, TYR565, SER564, LYS561, GLU635, ASN582, GLN585, CYS556, SER553, and GLY559, which contribute to the stable binding of the two proteins. N–Q) Antibodies specific for LRRK2 and THBS1 were used to immunoprecipitation (IP) and reverse‐IP lysates from DA cells. Western blot analysis of immunoprecipitated proteins was performed using antibodies specific for LRRK2 and THBS1. Data were normalized to GAPDH. R) The protein structure of the wild‐type and G2019S mutant is depicted, displaying the binding interface of THBS1 and LRRK2. Blue corresponds to LRRK2, green represents THBS1, and the sticks indicate the amino acid positions within the proteins. The left side is the wild‐type protein structure, while the right side is the structure of G2019S mutation. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison test in (G, H, J, L, O, Q). The difference in folds is statistically significant. ^*P < 0.05, ^**P < 0.01, ^*** P < 0.001.

Figure 4

Knocking down of THBS1 reduces ER stress via interacting TGF‐β1. A) Schematic representation of human THBS1 CRIPSRi. B) THBS1 CRIPSRi was transferred into the cells. Knocking down efficiency was confirmed by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. The protein levels of TGF‐β1, ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed in triplicate through C) WB analysis and D) their relative expression was calculated. The intensity of the protein bands was normalized to GAPDH. RT‐qPCR was performed to measure the levels of E) ATF4, F) CHOP, G) GRP78, and H) RTN1A. The resulting mRNA expression values were normalized to the expression of GAPDH. I,J) The DA cells were exposed to Tunicamycin at a concentration of 1 µm for varying time intervals ranging from 0 to 48 h. Additionally, another set of cells was treated with Tunicamycin for 24 h at different concentrations, including 0, 0.01, 0.1, 1, and 10 µm. Fluorescent spectrophotometer traces were recorded, displaying the changes in Mag‐Fluo‐4 AM fluorescence intensity in TG‐treated and CDN‐treated DA cells in comparison to control cells. K) Mitochondrial membrane depolarization in WT and LRRK2 G2019S DA cells was assessed by measuring the fluorescence intensity of TMRE dye at a concentration of 250 nm. L) Cell viability was assessed using the CCK‐8 assay. M) LDH assay was carried out to measure the cell apoptosis. N) Molecular docking was applied to verify the binding activity between THBS1 and TGF‐β1. The purple color represents the TGF‐β1 chain, and the green color represents the THBS1 chain. The blue dashed lines indicate hydrogen bonds. Several hydrogen bond interactions are formed between specific amino acids in THBS1, including ASN601, ASP652, ASN648, THR651, TYR665, CYS663, ASP1134, SER1135, PRO1137, TYR1139, HIS1095, THR1092, ASN1085, PRO1052, GLU1074, and HIS1075, and specific amino acids in TGF‐β1, including LYS280, ARG343, TYR340, HIS283, LYS286, HIS289, GLN349, ARG356, ARG249, ALA250, GLU69, ARG181, TYR1454, and ARG151. These hydrogen bond interactions are crucial for the stable binding of the two proteins. O–R) Antibodies specific for LRRK2 and THBS1 were used to IP and reverse‐IP lysates from DA cells. Western blot analysis of immunoprecipitated proteins was performed using antibodies specific for THBS1 and TGF‐β1. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison test in (B, D, E, F, G, H, K, L, M, P, R). The difference in folds is statistically significant. ^*P < 0.05, ^**P < 0.01, ^*** P < 0.001.

Figure 5

Knocking down of TGF‐β1 reduces ER stress. A) Schematic representation of human TGF‐β1 CRIPSRi. B) The CRISPRi system targeting TGF‐β1 was introduced into the cells to achieve knockdown of TGF‐β1 expression. The knockdown efficiency was confirmed by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. The protein levels of ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed in triplicate through C) WB analysis and D) their relative expression was calculated. The intensity of the protein bands was normalized to GAPDH. RT‐qPCR was performed to measure the levels of E) ATF4, F) CHOP, G) GRP78, and H) RTN1A. The resulting mRNA expression values were normalized to the expression of GAPDH. I,J) The DA cells were exposed to Tunicamycin at a concentration of 1 µm for varying time intervals ranging from 0 to 48 h. Additionally, another set of cells was treated with Tunicamycin for 24 h at different concentrations, including 0, 0.01, 0.1, 1, and 10 µm. Fluorescent spectrophotometer traces were recorded, displaying the changes in Mag‐Fluo‐4 AM fluorescence intensity in TG‐treated and CDN‐treated DA cells in comparison to control cells. K) Mitochondrial membrane depolarization in WT and LRRK2 G2019S DA cells was assessed by measuring the fluorescence intensity of TMRE dye at a concentration of 250 nm. L) Cell viability was assessed using the CCK‐8 assay. M) LDH assay was carried out to measure the cell apoptosis. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison test in (B, D, E, F, G, H, K, L, M). The difference in folds is statistically significant. ^*P < 0.05, ^**P < 0.01, ^*** P < 0.001.

Figure 6

LRRK2 G2019S could promote the expression of THBS1, TGF‐β1, and induce ER stress in vivo. A,B) Immunohistochemical analysis was performed to analyze the THBS1 expression. The protein levels of THBS1, TGF‐β1, ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed in triplicate through C) WB analysis and D) their relative expression was calculated. E‐H. RT‐qPCR was performed to measure the levels of E) ATF4, F) CHOP, G) RTN1A, and H) GRP78. The resulting mRNA expression values were normalized to the expression of GAPDH. I,J) A confocal image provided by immunofluorescence determined the expression levels of THBS1 and TGF‐β1. Green: anti‐THBS1; red: anti‐TGF‐β1. The scale bar represents 50 µm. K,L) A confocal image provided by immunofluorescence determined the expression levels of Iba1⁺ and ARMET. Green: anti‐Iba1⁺; red: anti‐ARMET. The scale bar represents 50 µm. The mRNA expression of M) IL‐1β, N) IL‐6, O) TGF‐β1, and P) TNF‐α were determined by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. Behavioral characterization of mice was analyzed with Q) open‐field, R) Y‐maze, S) Rotarod testing, and T) morris water maze. Data were presented as means ± SD. The experiments were carried out three times (n = 3). Unpaired student's t‐test in (B, D, E, F, G, H, J, L, M, N, O, P, Q, R, S, T). The difference in folds is statistically significant. ^*P < 0.05, ^**P < 0.01, ^*** P < 0.001.

Figure 7

Inhibition of THBS1 attenuates TGF‐β1 and ER stress induced by LRRK2 G2019S in vivo. A) Schematic diagram of stereotaxic injection of plasmids into the SNpc. B) Schematic representation of mice THBS1 CRIPSRi. C) THBS1 CRIPSRi was injected stereotaxically into the SNpc. Knocking down efficiency was confirmed by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. The protein levels of TGF‐β1, ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed in triplicate through D) WB analysis, and E) their relative expression was calculated. The intensity of the protein bands was normalized to GAPDH. F,G) A confocal image provided by immunofluorescence determined the expression levels of TGF‐β1 and ARMET. Green: anti‐ARMET; red: anti‐TGF‐β1. The scale bar represents 50 µm. The mRNA expression of H) TGF‐β1, I) IL‐1β, J) TNF‐α, and K) IL‐6 were determined by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. Behavioral characterization of mice was analysed with L) open‐field, M) Y‐maze, N) Rotarod testing, and O) Morris water maze (O). Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison test in (C, E, F, H, I, J, K, L, M, N, O). The difference in folds is statistically significant. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001.

Figure 8

TGF‐β1 suppression mitigates ER stress induced by LRRK2 G2019S in vivo. A) Schematic representation of mice TGF‐β1 CRIPSRi. B) TGF‐β1 CRIPSRi was injected stereotaxically into the SNpc. Knocking down efficiency was confirmed by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. The protein levels of C) ARMET, p‐PERK, PERK, p‐IRE1α, and IRE1α proteins were analyzed through WB analysis and D) their relative expression was calculated. The intensity of the protein bands was normalized to GAPDH. A confocal image provided by immunofluorescence determined the expression levels of E) p‐IRE1α and F) ARMET. Green: anti‐ARMET; red: anti‐p‐IRE1α. The scale bar represents 50 µm. RT‐qPCR was performed to measure the levels of G) ATF4, H) CHOP, I) GRP78, and J) RTN1A. The resulting mRNA expression values were normalized to the expression of GAPDH. The mRNA expression of K) IL‐1β, L) TNF‐α, and M) IL‐6 were determined by RT‐qPCR. The resulting mRNA expression values were normalized to the expression of GAPDH. N–Q) Behavioral characterization of mice was analyzed with L) open‐field, M) Y‐maze, N) Rotarod testing, and O) Morris water maze. Data were presented as means ± SD. The experiments were carried out three times (n = 3). One‐way ANOVA followed by Tukey's multiple comparison test in (B, D, F, G, H, I, J, K, L, M, N, O, P, Q). The difference in folds is statistically significant. ^* P < 0.05, ^** P < 0.01, ^*** P < 0.001.

Figure 9

Schematic showing the mechanisms underlying LRRK2 G2019S mutation promotes ER stress via interacting with THBS1/TGF‐β1 in PD.