The transcriptional response of cortical neurons to concussion reveals divergent fates after injury

Abstract

Traumatic brain injury (TBI) is a risk factor for neurodegeneration, however little is known about how this kind of injury alters neuron subtypes. In this study, we follow neuronal populations over time after a single mild TBI (mTBI) to assess long ranging consequences of injury at the level of single, transcriptionally defined neuronal classes. We find that the stress-responsive Activating Transcription Factor 3 (ATF3) defines a population of cortical neurons after mTBI. Using an inducible reporter linked to ATF3, we genetically mark these damaged cells to track them over time. We find that a population in layer V undergoes cell death acutely after injury, while another in layer II/III survives long term and remains electrically active. To investigate the mechanism controlling layer V neuron death, we genetically silenced candidate stress response pathways. We found that the axon injury responsive dual leucine zipper kinase (DLK) is required for the layer V neuron death. This work provides a rationale for targeting the DLK signaling pathway as a therapeutic intervention for traumatic brain injury. Beyond this, our approach to track neurons after a mild, subclinical injury can inform our understanding of neuronal susceptibility to repeated impacts.

Fig. 1. Closed-skull mTBI induces layer V neuron degeneration and death.

a Schematic of injury model/location. Coronal view (top) shows tip positioned over left cortex. Horizontal view (bottom) shows impact site relative to bregma. b Example of extent of injury, representative of >50 samples in which immunostaining for neuroinflammation was performed. GFAP staining is shown across sections from anterior to posterior. Approximate mm from Bregma on the anterior-posterior axis shown on the right. c Quantification of righting times (time to wake from anesthesia) demonstrating loss of righting reflex in TBI animals consistent with mild TBI. N = 25 Sham, 218 mTBI. All wildtype mice in the study included, excluding any without accurately recorded righting times. d Longitudinal measurement of neurofilament light in serum of mTBI animals. Each shape and line represents measurements for one mouse and the average of two replicates per mouse for n = 3 mice. *p = 0.0421 (1 dpi), p = 0.0352 (9 dpi), and p = 0.0381 (14 dpi) by Tukey’s multiple comparisons test for each timepoint compared to baseline. e Low magnification image of ipsilateral and contralateral cortex in Thy1-YFP-h mice. Layer V is outlined. High magnification images of f dendrites, g cell bodies, and h axons in the contralateral and ipsilateral cortices. For ipsilateral cell bodies and axons, insets show DAPI expression in cell bodies, and lack of DAPI expression in axon swellings. i Quantification of dendrite degeneration at 7 dpi in sham (n = 5), contralateral (n = 8), and ipsilateral (n = 8, *p = 0.0302, ***p = 0.0007 by Kruskal–Wallis test). j Quantifications of axon beading (fragments with area < 10 μm², ***p = 0.0009 by Kruskal–Wallis test) and axon swellings (fragments with area > 10 μm², **p = 0.007 by Kruskal–Wallis test) at 7 dpi in sham (n = 5), contralateral (n = 7), and ipsilateral (n = 7). k Quantification of YFP+ neurons in ipsilateral compared to contralateral cortex in sham animals and at 7 (n = 7) and 14 (n = 6) dpi compared to sham (n = 5, *p = 0.0378, **p = 0.0073 by Kruskal–Wallis test). For i–k points represent the average of 3–4 sections per mouse. Error bars represent SEM. ns not significant, by unpaired t-test. Scale bars: b 1 mm, e 200 μm, f–h (shown in h) 50 μm. Created in BioRender. Le Pichon, C. (2025) https://BioRender.com/y05i143.

Fig. 2. mTBI activates an Atf3-associated transcriptional response in different subclasses of cortical neurons.

a Immunostaining of ATF3 (red) in a Thy1-YFP (white) mouse showing specific expression in the ipsilateral cortex, with higher expression in layer V. Inset highlights YFP+ ATF3+ neurons. ATF3+ neurons are outlined. b Quantification of percent of YFP+ neurons expressing ATF3 at 7 dpi in the ipsilateral cortex. Points represent the average of 3–4 sections per animal across n = 3 animals. Error bars represent SEM. c Immunostaining showing that CTIP2 (cyan) is not expressed in layer V ATF3+ (red) nuclei. ATF3+ nuclei are outlined. d A schematic representation of the single nucleus RNA sequencing workflow, including unilateral closed-skull CCI, tamoxifen dosing, nuclear isolation, and FACS, following by 10X Genomics sequencing. e UMAP showing neurons collected by snRNAseq of ipsilateral cortex from pooled Atf3-CreER animals, annotated by mapping to a reference atlas. For subsequent analyses, cell types with fewer than 20 nuclei are excluded. f Dotplot of marker genes for layer-specific excitatory neurons and subclasses of inhibitory neurons, showing downregulation of some markers in Atf3-CreER animals compared to the reference dataset. g Dotplot of a panel of known stress response genes involved in axon growth and regeneration, cell stress and transcription factor regulation, apoptosis, and ER stress highlighting different responses between Atf3-expressing neuron subclasses. Genes were selected based on altered expression compared to reference dataset. Low magnification scale bars, 500 µm. High magnification scale bars, 50 µm. Created in BioRender. Le Pichon, C. (2025) https://BioRender.com/g45x590.

Fig. 3. Atf3-expressing neurons in layer V die, while those in layer II/III survive following mTBI.

a Quantification of the average count per section of Atf3-GFP neurons by cortical layer at 3, 5, 7, 10, 14, 21, 42, and 70 dpi. N = 6 per timepoint, 3–4 sections counted per animal. b Examples of Atf3-GFP endogenous labeling (left) at 7 and 21 dpi. Layer V is outlined. High magnification examples of GFP-immunolabeled neurons (right) in layer II/III (top) and layer V (bottom) at 7 and 21 dpi. Panels a and b use the Atf3-Cre::Snap25-eGFP mouse, which labels all neurons in which Atf3 has ever been expressed. c Schematic representation of tamoxifen dosing and tissue collection. The Atf3-CreER mouse was crossed to the Ai14 RFP reporter (not neuron-specific). d Representative images of Ai14 signal in ipsilateral cortex at 7 dpi and 21 dpi. Layer V is outlined. Insets highlight neurons and other labeled cells in layer II/III and layer V. Insets for layer II/III suggest an earlier loss of projection complexity that is regained by 21 dpi. Inset for layer V at 21 dpi includes a neuron (left) and a glial cell (right). Panels c and d use the Atf3-CreER::Ai14 mouse, which labels any cell in which Atf3 is activated in the presence of tamoxifen. For a error bars represent standard error of the mean (SEM). Scale bars: b Low magnification, 50 µm, high magnification, 10 µm; d. Low magnification, 500 µm, high magnification, 50 µm.

Fig. 4. Layer V Atf3-GFP neurons are unable to fire and downregulate ion channels while layer II/III Atf3-expressing neurons are electrophysiologically functional following mTBI.

a Examples of electrophysiological traces from 7 dpi layer V ipsilateral GFP+ and GFP- neurons and contralateral GFP- neurons. Quantifications of (b) IO curve, (c) rheobase, (d) max spike count (*** p = 0.0002, ****p < 0.0001), (e). resting membrane potential (****p = <0.0001), and f. capacitance (*p = 0.0307) in 7 dpi layer V neurons. g Dot plot of select ion channels in the reference dataset compared to the Atf3-CreER dataset collected in this study showing dysregulation of ion channels in layer II/III and layer V neurons. Genes in bold were validated by in situ hybridization. h In situ hybridization validating downregulation of ion channels. Low magnification image of bilateral cortices from an Atf3-Cre::Snap25-GFP mouse at 7 dpi showing mRNA expression of Gfp (cyan) and Scn1a (yellow). Layer V is outlined. Inset shows mRNA of Gfp (cyan), Tubb3 (white), Scn1a (yellow), and Kcnq5 (red). Gfp+ neurons (outlined) lack expression of Scn1a and Kcnq5 and have little to no expression of Tubb3. Arrowheads highlight Tubb3+ Gfp− neurons with high expression of Scn1a and Kcnq5. A single Z-plane is shown in the insets. i Examples of electrophysiological traces from 21 dpi layer II/III surviving ipsilateral GFP+ and GFP− neurons and contralateral GFP− neurons. Quantifications of (j) IO curve, (k) rheobase (**p = 0.0050), (l) max spike count, (m) resting membrane potential (*p = 0.0191, **p = 0.0041), and (n) capacitance (*p = 0.0217, **p = 0.0034) in 21 dpi layer II/III neurons. For b and j points represent the average of all neurons per group. Error bars represent SEM. For c–f, GFP+ ipsi n = 14 cells, GFP− ipsi n = 9, GFP- contra n = 9, each point represents one neuron recorded from N = 3 animals. For k–n, GFP+ ipsi n = 18 cells, GFP− ipsi n = 14, GFP- contra n = 9; each point represents one neuron recorded from N = 2 animals. ns: not significant. Significance was determined by Tukey’s multiple comparisons test. Low magnification scale bar, 500 µm. High magnification scale bar, 50 µm.

Fig. 5. Layer V neurons can be rescued from mTBI-induced death and degeneration by deletion of DLK.

a P-c-Jun immunolabeling in the ipsilateral cortex of WT mice at 7 dpi (left). Layer V is outlined. Insets show high magnification images of immunolabeling in layer II/III and layer V. b Quantification of p-c-Jun intensity in layer II/III and layer V at 7 dpi (**p = 0.0016 by two-tailed paired t-test). Only p-c-Jun+ cells are included based on a threshold of 1.2-fold expression compared to background. Average per animal and value per cell is displayed. Each shape represents one animal. N = 4, 2 sections per animal, 11–53 nuclei per animal, II/III nuclei n = 89, V nuclei n = 190. c Overview of ipsilateral and contralateral cortices in DLK WT and DLK cKO mice showing layer V GFP+ nuclei. Insets shown on the right. d Quantification of GFP+ neurons in ipsilateral compared to contralateral cortex in WT (n = 5) and cKO (n = 7) mice at 14 dpi based on Sun1-GFP expression shown in c (** p = 0.0025 by two-tailed Mann–Whitney test). e High magnification images of YFP+ dendrites (top) and axons (bottom) in WT contralateral and ipsilateral cortex, and cKO ipsilateral cortex. f Quantification of dendrite degeneration at 7 dpi in WT (ipsi n = 5, contra n = 6) and cKO mice (ipsi n = 5, contra n = 6, **p = 0.0087 by two-tailed Mann–Whitney test). g. Quantification of axon beading (**p = 0.0079) and swelling (**p = 0.0079 by two-tailed Mann–Whitney test) at 7 dpi in WT (n = 5) and cKO (n = 5) mice. For d, f, g, each point represents the average of 3–4 sections per animal. Error bars represent SEM. Low magnification scale bars, 500 µm. High magnification scale bars, 50 µm.