Neuroscience

In vivo virus-free lineage tracing and live imaging reveal that NeuroD1 does not reprogram microglia into neurons

Xiaoyu Li
Yuxin Li
Yue Cao
Niewen Hu
Baozhi Yang
Pei Ouyang
Yuxiao Jin
Shuai Gao
Li Cao
Bo Peng author has email address
Yanxia Rao author has email address

National Children’s Medical Center, Children’s Hospital, Institute for Translational Brain Research, State Key Laboratory of Medical Neurobiology, MOE Frontiers Center for Brain Science, Fudan SANS Neuroscience Center, Fudan University, Shanghai, China
Department of Neurology, Zhongshan Hospital, Laboratory Animal Center, Fudan University, Shanghai, China
Key Laboratory of Molecular Neurobiology, Ministry of Education, Navy Military Medical University, Shanghai, China
MOE Innovative Center for New Drug Development of Immune Inflammatory Diseases, Shanghai Key Laboratory of Gene Editing and Cell Therapy for Rare Diseases, Fudan University, Shanghai, China
Department of Neurosurgery, Huashan Hospital, Fudan University, Shanghai, China

https://doi.org/10.7554/eLife.112083.2

Open access
Copyright information

Figures and data

Virus-free genetic induction can successfully induce NeuroD1 expression in microglia.
(A) Schematic of the specific expression of NeuroD1, lineage tracing of microglia and experimental timeline. D, day. i.g. oral gavage. (B) Representative confocal images at D4, D10, D30 and D45 showing that NeuroD1 is robustly co-expressed in GFP-positive cells after Cre recombination. (C) Representative confocal images at D18 showing that NeuroD1 is robustly co-expressed in GFP-positive cells after Cre recombination. (D) Quantification of the percentage of NeuroD1⁺GFP⁺ cells among GFP⁺ cells and NeuroD1⁺GFP^- cells among NeuroD1⁺, revealing NeuroD1 is highly co-expressed with GFP. N = 4-6 mice for each group. Data are presented as mean ± SD.

Sustained NeuroD1 expression in microglia does not induce microglia-to-neuron conversion across time.
(A) Experimental timeline. D, day. i.g. oral gavage. (B) Representative confocal images at D4, D10, D18, D30 and D45 showing that GFP is co-labeled with IBA1 but not co-labeled with NeuN. (C) Quantification of IBA1⁺GFP⁺ double-positive cells among GFP⁺ cells or NeuN⁺GFP⁺ double-positive cells among NeuN⁺ in cortex at D4, D10, D18, D30 and D45. N = 4 or 5 mice for each group. (D) Quantification of IBA1⁺GFP⁺ double-positive cells among GFP⁺ cells or NeuN⁺GFP⁺ double-positive cells among NeuN⁺ in striatum at D4, D10, D18, D30 and D45. N = 4 or 5 mice for each group.

NeuroD1-expressing microglia retain microglial their characteristic morphology, but undergo progressively cell loss over time.
(A) Experimental timeline. D, day. i.g. oral gavage. (B) Representative images showing GFP-positive cells with highly ramified morphologies characteristic of microglia rather than neurons. The white rectangular indicate no morphological change, the yellow arrow indicating the GFP⁺ cell loss over time. (C) Representative images illustrating changes in the number of GFP-positive cells over time (green dots indicate GFP-positive cell). (D) Quantification of GFP-positive cells density. N = 4 - 6 mice per time point. Data are presented as mean ± SD. One Way ANOVA with Turkey’s multiple comparison test. (E) Experimental timeline. (F) Representative confocal image shows TUNEL⁺ Reporter⁺ cells are increased in CX3CR1-ND1 animal. (G) Quantification of TUNEL⁺ Reporter⁺ cells at D4 and D18. Unpaired t test. Data are presented as mean ± SD. (H) Relative gene expression level at D18, Unpaired t test. Data are presented as mean ± SD. (I) Representative confocal image shows the LC3 and P62 level in CX3CR1-Ai14 and CX3CR1-ND1mice. Scale bar, 50μm. (J) Quantification of P62⁺ and LC3⁺ area per cell. Unpaired t test. Data are presented as mean ± SD.

NeuroD1 does not induce microglia-to-neuron conversion even in an injury conditioned microenvironment.
(A) Experimental design. D, day. i.g. oral gavage. (B) Quantification of grip strength showing no improvement in NeuroD1-expressing TBI mice. Two-way ANOVA with multiple comparisons. (C) Quantification of rotarod performance showing no improvement in NeuroD1-expressing TBI mice. Two-way ANOVA with multiple comparisons. (D) Representative confocal images of DAPI, GFP and NeuN co-staining results in CX3CR1-ND1 TBI mice, showing no co-localization of GFP with NeuN in either the lesion core or distal regions. Scale bar, 50 μm. (E) Quantification of the percentage of GFP⁺NeuN⁺ cells among total NeuN⁺ cells in both the lesion area and distal area. n.d., not detected. (F) Representative confocal images of DAPI, GFP, and MAP2 in CX3CR1-ND1 mice, showing no co-localization of GFP with MAP2 in either the lesion core or distal regions. Scale bar, 50 μm. (G) Quantification of the percentage of GFP⁺MAP2⁺ cells among total MAP2⁺ cells in the lesion area and distal region. n.d., not detected. (H) UMAP reduction shows the cell types of the FACS-sorted reporter⁺ cells in each group; note that there are no neurons. (I) Feature plot showing that the microglia-specific marker P2ry12 was highly expressed even under TBI conditions, but the neuron-specific marker Rbfox3 was barely expressed. (J) Violin plot showing the expression of microglia-specific genes (Hexb and Tmem119) and neuron-specific genes (Tubb3 and Map2) in 4 groups. (K) Experiment timeline of TMEM119-ND1 mice, TBI was performed before tamoxifen induction. (L) Representative confocal images of GFP, IBA1 and NeuN in TMEM119-ND1 mice, showing no co-localization of GFP with NeuN in either the lesion core or distal regions. Scale bar, 50 μm. (M) Quantification of the percentage of GFP⁺NeuN⁺ cells among total NeuN⁺ cells in the lesion area and distal region. n.d., not detected. (N) Experiment timeline of TMEM119-ND1 mice, TBI was performed after tamoxifen induction. (O) Representative confocal images of GFP, IBA1 and NeuN in TMEM119-ND1 mice, showing no co-localization of GFP with NeuN in either the lesion core or distal regions. Scale bar, 50 μm. (P) Quantification of the percentage of GFP⁺NeuN⁺ cells among total NeuN⁺ cells in the lesion area and distal region. n.d., not detected.

Lentivirus- or AAV-based infection caused leaky neuronal expression in vivo.
(A) The study design shows the different viruses used here. (B) Representative confocal images showing the expression of GFP, IBA1 and NEUN in the brains of the lentivirus- or AAV-infected mice. (C) The efficacy of infection was quantified, and the data are presented as the ratio of IBA1⁺GFP⁺/GFP⁺ cells. N = 3 mice for each group.

NeuroD1-expressing microglia is co-labeled with microglial marker IBA1 but not neuronal marker NeuN.
(A) Experimental timeline. (B) Representative confocal images at D4, D10, D18 and D30 showing that GFP is robustly co-expressed in IBA1⁺ cells but not NeuN⁺ cells after Cre recombination. (C) Quantification of IBA1⁺GFP⁺ double-positive cells in GFP⁺ cells and NeuN⁺GFP⁺ in GFP⁺ cells at D4, D10, D18 and D30. N = 5 mice for each group. (D) Quantification of GFP⁺ cells per brain slide in TMEM119-ND1 mice. N = 5 mice for each group.

Ectopic NeuroD1 expression in microglia does not induce microglia-to-neuron conversion, as GFP does not co-localize with TUJ1 and MAP2.
(A) Experimental timeline. (B) Representative confocal images at D4, D10, D30 and D45 showing that GFP is not co-labeled with TUJ1 or MAP2 in the brain. (C) Quantification of TUJ1⁺GFP⁺ double-positive cells among TUJ1⁺ cells at D4, D10, D30 and D45. N = 4 or 5 mice for each group. (D) Quantification of MAP2⁺GFP⁺ double-positive cells among MAP2⁺ cells at D4, D10, D30 and D45. N = 4 or 5 mice for each group. (E) Relative gene expression level at D18. N = 4 mice for each group.

FACS gating strategy and supplemental quality control information of scRNA-seq data.
(A) Gating strategy of CX3CR1-Ai14 and CX3CR1-ND1 mice. (B) Number of cells (after quality control) per group. (C) Violin plot shows the number of genes detected in each cell (nFeature_RNA), total number of molecules detected within a cell (nCount_RNA) and the distribution of mitochondrial genes in each cell (percent.mt) among the 4 groups. (D) UMAP plot show all cells were divided into 22 unsupervised cluster with resolution = 1. (E) Dotplot show microglia, T cell and NK cell specific markers expression over 22 clusters.

Sign up for email alerts