Astroglial toxicity promotes synaptic degeneration in the thalamocortical circuit in frontotemporal dementia with GRN mutations
Elise Marsan, … , Arnold R. Kriegstein, Eric J. Huang
Elise Marsan, … , Arnold R. Kriegstein, Eric J. Huang
Published January 5, 2023
Citation Information: J Clin Invest. 2023;133(6):e164919. https://doi.org/10.1172/JCI164919.
View: Text | PDF
Research Article Neuroscience

Astroglial toxicity promotes synaptic degeneration in the thalamocortical circuit in frontotemporal dementia with GRN mutations

Abstract

Mutations in the human progranulin (GRN) gene are a leading cause of frontotemporal lobar degeneration (FTLD). While previous studies implicate aberrant microglial activation as a disease-driving factor in neurodegeneration in the thalamocortical circuit in Grn–/– mice, the exact mechanism for neurodegeneration in FTLD-GRN remains unclear. By performing comparative single-cell transcriptomics in the thalamus and frontal cortex of Grn–/– mice and patients with FTLD-GRN, we have uncovered a highly conserved astroglial pathology characterized by upregulation of gap junction protein GJA1, water channel AQP4, and lipid-binding protein APOE, and downregulation of glutamate transporter SLC1A2 that promoted profound synaptic degeneration across the two species. This astroglial toxicity could be recapitulated in mouse astrocyte-neuron cocultures and by transplanting induced pluripotent stem cell–derived astrocytes to cortical organoids, where progranulin-deficient astrocytes promoted synaptic degeneration, neuronal stress, and TDP-43 proteinopathy. Together, these results reveal a previously unappreciated astroglial pathology as a potential key mechanism in neurodegeneration in FTLD-GRN.

Authors

Elise Marsan, Dmitry Velmeshev, Arren Ramsey, Ravi K. Patel, Jiasheng Zhang, Mark Koontz, Madeline G. Andrews, Martina de Majo, Cristina Mora, Jessica Blumenfeld, Alissa N. Li, Salvatore Spina, Lea T. Grinberg, William W. Seeley, Bruce L. Miller, Erik M. Ullian, Matthew F. Krummel, Arnold R. Kriegstein, Eric J. Huang

×

Figure 3

snRNA-Seq analyses in the thalamus and frontal cortex of individuals with and without FTLD-GRN, and Grn+/+ and Grn–/– mice.

Options: View larger image (or click on image) Download as PowerPoint
snRNA-Seq analyses in the thalamus and frontal cortex of individuals wit...
(A) A schematic diagram for single-cell transcriptomics using the thalamus and frontal cortex from control and FTLD-GRN cases and 19-month-old Grn+/+ and Grn–/– mice. (B and C) tSNE plots highlighting the distribution of cell clusters from hTH, hFCX from control and FTLD-GRN cases (B), mTH and mFCX from 19-month-old Grn+/+ and Grn–/– mice (C). (D and E) Violin plots showing GRN RNA levels in all the cell types in hTH and hFCX from individuals with and without FTLD-GRN. Statistics uses multiple unpaired Student’s 2-tailed t tests. (F) Immunostains for PGRN in hTH and hFCX of patients with FTLD-GRN. n, neuron; b, blood vessel; g, glia cell; L3, layer 3; L4, layer 4. Scale bars: 100 μm; 20 μm (inset). (G) Quttantification of PGRN intensity in neurons and glia in hTH and hFCX in individuals with and without FTLD-GRN. Statistics uses 2-tailed Student’s t test, n represents the number of independent samples used. (H) Confocal images of PGRN in IBA1+ microglia, CD34+ endothelial cells, MAP2+ neurons, and GFAP+ astrocytes, in the hFCX of control and FTLD-GRN patients. Scale bars: 20 μm. (I and J) Gene burden score for each cell cluster in the hTH, hFCX, mTH, and mFCX. Statistics use parametric (mTH, mFCX) or nonparametric (hTH, hFCX) 1-way ANOVA. (K) Bar graphs showing the ratio of the number of nuclei captured in the hTH and hFCX in control and FTLD-GRN patients. Statistics use nonparametric 2-tailed Student’s t test (hTH: ExNeu-1, ExNeu-2, and MG; hFCX: AST-REAC) or parametric 2-tailed Student’s t test (for remaining clusters). ExNeu, excitatiry neurons; InNeu, inhibitory neurons; AST, astrocytes; MG, microglia; OL, oligodendroglia; OPC, oligodendroglial precursors; ENDO, endothelial cells; PER, pericytes. All data represent mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. NS, not significant.