Single-Cell Sequencing of iPSC-Dopamine Neurons Reconstructs Disease Progression and Identifies HDAC4 as a Regulator of Parkinson Cell Phenotypes

Abstract

Induced pluripotent stem cell (iPSC)-derived dopamine neurons provide an opportunity to model Parkinson's disease (PD), but neuronal cultures are confounded by asynchronous and heterogeneous appearance of disease phenotypes in vitro. Using high-resolution, single-cell transcriptomic analyses of iPSC-derived dopamine neurons carrying the GBA-N370S PD risk variant, we identified a progressive axis of gene expression variation leading to endoplasmic reticulum stress. Pseudotime analysis of genes differentially expressed (DE) along this axis identified the transcriptional repressor histone deacetylase 4 (HDAC4) as an upstream regulator of disease progression. HDAC4 was mislocalized to the nucleus in PD iPSC-derived dopamine neurons and repressed genes early in the disease axis, leading to late deficits in protein homeostasis. Treatment of iPSC-derived dopamine neurons with HDAC4-modulating compounds upregulated genes early in the DE axis and corrected PD-related cellular phenotypes. Our study demonstrates how single-cell transcriptomics can exploit cellular heterogeneity to reveal disease mechanisms and identify therapeutic targets.

Keywords: Parkinson’s disease; histone deacetylase 4; induced pluripotent stem cells; single-cell RNA sequencing.

Graphical abstract

Figure 1

Bulk RNA-Seq Analysis Confirms Purification of iPSC-Derived Dopamine Neurons and Identifies 247 DE Genes between Control and PD GBA-N370S Patients Enriched for Genes in Pathways of Neuronal Function (A) Schematic of sorting the tyrosine hydroxylase-positive (TH+) iPSC-derived dopamine neurons from three controls and three PD GBA-N370S patients displaying a FACS plot identifying live/TH+ cells for sorting into bulk collection and into 96-well plates for single-cell RNA sequencing (gray wells indicate blank wells). Bulk and single cells went through RNA extraction, cDNA synthesis, and amplification before undergoing sequencing and bioinformatic analysis. (B and C) Expression of dopamine neuron-specific markers (B) and the absence of glutamatergic markers (C) in the purified bulk iPSC-derived dopamine neurons. (D) Volcano plot showing 247 genes DE between GBA-N370S PD versus control identified by DESeq2 (FDR 1%). (E) GO enrichment analysis of the upregulated and downregulated genes in PD GBA-N370S patients highlights DE of genes involved in neuronal development and synaptic activity.

Figure 2

Single-Cell RNA-Seq Stratification Identifies iPSC-Derived Dopamine Neurons from GBA3 as Significantly Different from Both PD Patient and Control Neurons (A) Transcriptome PCA analysis resolves GBA3 neurons (yellow) from the remaining two PD GBA-N370S patients and three controls. (B) Over-dispersion analysis identifies a subset of genes that vary more than expected due to technical fluctuations in the dataset alone. (C) Heatmap of the single-cell RNA-seq samples identifies an enrichment in the endoplasmic reticulum (ER) signal recognition particle (SRP) pathway in GBA3. (D) Expression in log₂ (TPM+1) of three genes (RPS12, RPS17, and RPS6) prioritized from those significantly DE within the SRP pathway between GBA3 and controls 1, 2, and 3 and GBA1 and 2. DE analysis was performed using a two-sided Wilcoxon signed-rank test on all genes in the SRP pathway. (E) The upregulation of the three selected genes involved in this pathway was confirmed in iPSC-derived dopamine neurons differentiated from three iPSC clones of GBA3 compared to the three original controls and two PD GBA-N370S patients (GBA1 and 2), plus a fourth PD GBA-N370S patient (GBA4). Data are represented as mean ± SD (^∗∗p < 0.01; ^∗∗∗p < 0.001; ^∗∗∗∗p < 0.0001).

Figure 3

Pseudotime Analysis Temporally Orders the Core Set of 60 Functionally Similar Genes DE in Both the Bulk and Single-Cell RNA-Seq between Control and PD GBA-N370S Patients (A) Refined transcriptomic disease axis analysis of the core gene set of 60 genes. The control-disease single-cell transcriptomic axis was re-inferred with the 60 genes alone using a parametric factor analysis model that associated each gene with a point along the axis at which it was upregulated or downregulated. (B) The phenotypic linkage network demonstrates a higher functional similarity of the 60-gene set with each other compared to a background control set (p < 2.2e−16). This higher functional similarity was also identified between the 60-gene set and a group of known PD loci, compared to a background control set (p = 8.52e−08). The high functional similarity of PD genes to each other is used as a positive control. (C) Along the axis of disease, the downregulation of HDAC4-controlled genes (PRKCB, RTN1, ATP1A3, and TSPAN7) at 22 DIV precedes the upregulation of ER stress genes (ERO1A, FKBP9, and PDI) at 38 DIV. Data are represented as mean ± SD (^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001). The locations of the HDAC4 and ER genes analyzed from the core 60 set are marked on (A).

Figure 4

Modulation of PP2A Activity Corrects HDAC4 Nuclear Mislocalization in PD GBA-N370S iPSC-Derived Dopamine Neurons (A) Cytoplasmic and nuclear localization of HDAC4 in control and PD GBA-N370S dopamine neurons shown by immunofluorescence at 45 DIV—TH, green; HDAC4, red; DAPI, blue; HDAC4/DAPI nuclear colocalization, purple. The HDAC4 nuclear/cytoplasmic ratio is significantly increased in PD GBA-N370S patients. Data are represented as mean ± SD (^∗∗p < 0.01). (B) HDAC4 cellular localization in the presence or absence of tasquinimod (HDAC4 allosteric inhibitor) or okadaic acid, cantharidin, and LB-100 (PP2A inhibitors) at 45 DIV—TH, green; HDAC, red; DAPI, blue; and HDAC4/DAPI nuclear colocalization, purple. The three PP2A inhibitors correct HDAC4 nuclear mislocalization in PD GBA-N370S patient-derived dopamine neurons compared to no treatment. In contrast, tasquinimod, a HDAC4 allosteric inhibitor, has no effect on HDAC4 localization. Data are represented as mean ± SD (^∗∗∗∗p < 0.0001).

Figure 5

Modulation of HDAC4 Activity or Localization Corrects the Downregulation of HDAC4-Controlled Genes in PD GBA-N370S iPSC-Derived Dopamine Neuron Cultures and Ameliorates PD GBA-N370S ER Stress Phenotypes Expression of four HDAC4-regulated genes (TSPAN7, ATP1A3, RTN1, and PRKCB; bottom) and three ER stress genes (PDIA6, FKBP9, and ERO1A; top) at the RNA (left) and protein (right) levels in the presence and absence of HDAC4-modifying drugs tasquinimod, okadaic acid, cantharidin, and LB-100 in PD GBA-N370S and control patient-derived neurons at 45 DIV. The upregulation of HDAC4-repressed genes in PD GBA-N370S iPSC-derived dopamine neurons by all four compounds was accompanied by a decrease in ER stress. Data are represented as mean ± SEM (^∗p < 0.05, ^∗∗p < 0.01, ^∗∗∗p < 0.001, and ^∗∗∗∗p < 0.0001).

Figure 6

Modulation of HDAC4 Activity or Localization Rescues Deficits in the Autophagic and Lysosomal Pathway and Reduces α-Synuclein Release in PD GBA-N370S iPSC-Derived Dopamine Neurons (A and B) Modulation of HDAC4 activity by allosteric inhibition of HDAC4 (tasquinimod) or inhibition of PP2A (cantharidin) rescues the increase in autophagosomal (LC3-II; A) and lysosomal (LAMP1; B) compartments seen by western blot in PD GBA-N370S patient iPSC-derived neurons compared to controls. (C) The reduction of lysosomes in PD GBA-N370S iPSC-derived dopamine neurons treated with tasquinimod or cantharidin was confirmed by a decrease in lysosome punctae by immunofluorescence. (D) Modulation of HDAC4 increases lysosomal activity in PD GBA-N370S iPSC-derived neurons measured by DQ-BSA cleavage. (E) Tasquinimod or cantharidin reduces the increase in α-synuclein release seen in PD GBA-N370S patient-derived neurons compared to controls. Data are represented as mean ± SEM (^∗p < 0.05, ^∗∗p < 0.01, and ^∗∗∗p < 0.001).

Figure 7

Nuclear Mislocalization of HDAC4 and Related Perturbations in Gene Expression Are Observed in Idiopathic PD Cases (A) Cytoplasmic and nuclear localization of HDAC4 in control and idiopathic PD iPSC-derived dopamine neurons shown by immunofluorescence at 45 DIV—TH, green; HDAC4, yellow; DAPI, blue. (B) The HDAC4 nuclear/cytoplasmic ratio is significantly increased in two of the four idiopathic PD patients. Data are represented as mean ± SEM (^∗p < 0.05). (C and D) A (C) decrease in the expression of HDAC4-controlled genes: TSPAN7; ATP1A3; RTN1; and PRKCBI and an (D) increase in the expression of ER stress genes: ERO1A; PDIA6; and FKBP9 is observed in the same two idiopathic PD cases that display HDAC4 mislocalization.