Insulin Resistance Is a Modifying Factor for Parkinson's Disease

Abstract

Background: Parkinson's disease (PD) is the second most common, and the fastest-growing neurodegenerative disorder with unclear etiology in most cases. Therefore, the identification of non-genetic risk factors for PD pathology is crucial to develop effective preventative or therapeutic strategies. An increasing number of evidence suggests that central insulin resistance might have an essential role in PD pathology. Nevertheless, it is not clear whether insulin resistance arises from external factors/lifestyle, comorbidities such as type 2 diabetes or it can occur in a PD patient's brain independently from peripheral insulin resistance.

Objective: We aimed to investigate insulin resistance and its role in GBA1 mutation-associated PD pathogenesis and phenotype severity.

Methods: Midbrain organoids, generated from induced pluripotent stem cells (iPSCs) of PD patients carrying the GBA1-N370S heterozygous mutation (GBA-PD) and healthy donors, were exposed to different insulin concentrations to modify insulin signaling function. Transcriptomics analysis was performed to explore insulin signaling gene expression patterns in GBA-PD and to find a potential target for GBA-PD-associated phenotype rescue.

Results: The insulin signaling pathway genes show dysregulation in GBA-PD. Particularly, we highlight that a knockdown of FOXO1 mitigates the loss of dopaminergic neurons and cellular death in GBA-PD. Additionally, our findings suggest a promising therapeutic potential of the anti-diabetic drug Pioglitazone in decreasing dopaminergic neuron loss associated with GBA-PD.

Conclusion: Local insulin signaling dysfunction plays a substantial role in GBA-PD pathogenesis, exacerbating dopaminergic neuron death. © 2024 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Keywords: GBA; Parkinson's disease; comorbitity; diabetes; pioglitazone.

FIG. 1

GBA1‐N370S mutation (GBA‐PD) organoids show dopaminergic neuron loss and altered lipid profile. (A) Representative immunofluorescence images of MAP2 and tyrosine hydroxylase (TH) staining in sections of 60 days old GBA‐PD and healthy control (WT) midbrain organoids cultured in insulin resistant (IR) or insulin sensitive (IS) conditions. Scale bars: 100 μm. (B) Quantification of TH‐positive area normalized to the Hoechst area in 60 days old midbrain organoids. (C) Extracellular dopamine levels measured in spent media of 60 days old midbrain organoids. (D) Representative immunofluorescence images of Tunel staining in sections of 60 days old midbrain organoids. Scale bars: 100 μm. (E) Quantification of Tunel‐positive area normalized to the Hoechst area in 60 days old midbrain organoids. (F) Quantification of cholesterol esters. Concentration normalized to the metagenomic DNA. (G) Quantification of 1‐alkenyl, 2‐acylphosphatidylcholine. Concentration normalized to the metagenomic DNA. (H) The relative concentration of the sum of ceramide and hexosylceramide species sharing the same carbon number expressed as log2 fold change (FC) in GBA‐PD samples against insulin‐resistant (IR) healthy control (WT) samples. Statistical significance was tested using the non‐parametric Kruskal‐Wallis test for multiple comparisons, followed by Dunn's post hoc test. P values are adjusted using Benjamini‐Hochberg method (*P < 0.05, **P < 0.01, ***P < 0.001). [Color figure can be viewed at wileyonlinelibrary.com]

FIG. 2

Transcriptome analysis of GBA1‐N370S mutation (GBA‐PD) midbrain organoids. (A) Unsupervised clustering of insulin signaling genes between insulin‐resistant (IR), GBA, and IR, healthy control (WT) midbrain organoids. Gene expression was normalized using Z‐score transformation. (B) UpSet plot demonstrating the number of unique differentially expressed genes (P.adjust < 0.05) found in each comparison. (C) Gene enrichment analysis of the selected specific genes. (D) Volcano plot demonstrating expression of a set of FOXO1 target genes in IR, GBA versus IR, WT midbrain organoids. (E) A probabilistic Boolean model with defined perturbations: AKT1‐OFF and FOXO1‐ON, demonstrating dynamic behavior of downstream effects associated with Parkinson's disease (axis‐Y) over the number of iterations (axis‐X). [Color figure can be viewed at wileyonlinelibrary.com]

FIG. 3

Rescue experiments of GBA1‐N370S mutation (GBA‐PD) associated phenotypes using antisense oligonucleotides (ASOs) and anti‐diabetic drugs. (A) Organoid treatment strategy with ASOs, pioglitazone, and metformin. (B) Representative immunofluorescence images of MAP2, tyrosine hydroxylase (TH) and Tunel staining in sections of 60 days old GBA‐PD midbrain organoids treated with negative control (NC), and ASO targeting FOXO1 transcript (FOXO1‐ASO). Scale bars: 100 μm. (C) Quantification of TH‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with NC, and ASO targeting FOXO1 transcript. (D) Extracellular dopamine levels measured in spent media of 60 days old GBA‐PD midbrain organoids treated with NC, and ASO targeting FOXO1 transcript. (E) Quantification of Tunel‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with NC, and ASO targeting FOXO1 transcript. (F) Quantification of TH‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with vehicle (dimethylsulfoxide [DMSO]) and pioglitazone (Piogl). (G) Extracellular dopamine levels measured in spent media of 60 days old GBA‐PD midbrain organoids treated with vehicle (DMSO) and Piogl. (H) Quantification of Tunel‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with vehicle (DMSO) and Piogl. (I) Quantification of TH‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with vehicle (phosphate buffered saline [PBS]) and metformin (Metf). (J) Extracellular dopamine levels measured in spent media of 60 days old GBA‐PD midbrain organoids treated with vehicle (PBS) and Metf. (K) Quantification of Tunel‐positive area normalized to the Hoechst area in 60 days old GBA‐PD midbrain organoids treated with vehicle (PBS) and Metf. Statistical significance was tested using the non‐parametric Kruskal‐Wallis test for multiple comparisons, followed by Dunn's post hoc test. P values are adjusted using Benjamini‐Hochberg method (*P < 0.05, **P < 0.01, ***P < 0.001). [Color figure can be viewed at wileyonlinelibrary.com]