Skip to main content
NCBI home page
iScience logoLink to iScience
. 2021 Jan 20;24(2):102074. doi: 10.1016/j.isci.2021.102074

Neurod4 converts endogenous neural stem cells to neurons with synaptic formation after spinal cord injury

Toshiki Fukuoka 1,5, Akira Kato 1,5, Masaki Hirano 1, Fumiharu Ohka 1, Kosuke Aoki 1, Takayuki Awaya 1, Alimu Adilijiang 1, Maeda Sachi 1, Kuniaki Tanahashi 1, Junya Yamaguchi 1, Kazuya Motomura 1, Hiroyuki Shimizu 1, Yoshitaka Nagashima 1, Ryo Ando 1, Toshihiko Wakabayashi 1, Dasfne Lee-Liu 2,3, Juan Larrain 4, Yusuke Nishimura 1,∗, Atsushi Natsume 1,6,∗∗
PMCID: PMC7889987  PMID: 33644710

Summary

The transcriptome analysis of injured Xenopus laevis tadpole and mice suggested that Neurod4L.S., a basic-helix-loop-helix transcription factor, was the most promising transcription factor to exert neuroregeneration after spinal cord injury (SCI) in mammals. We generated a pseudotyped retroviral vector with the neurotropic lymphocytic choriomeningitis virus (LCMV) envelope to deliver murine Neurod4 to mice undergoing SCI. SCI induced ependymal cells to neural stem cells (NSCs) in the central canal. The LCMV envelope-based pseudotypedvector preferentially introduced Neurod4 into activated NSCs, which converted to neurons with axonal regrowth and suppressed the scar-forming glial lineage. Neurod4-induced inhibitory neurons predominantly projected to the subsynaptic domains of motor neurons at the epicenter, and Neurod4-induced excitatory neurons predominantly projected to subsynaptic domains of motor neurons caudal to the injury site suggesting the formation of functional synapses. Thus, Neurod4 is a potential therapeutic factor that can improve anatomical and functional recovery after SCI.

Subject areas: Biological Sciences, Neuroscience, Behavioral Neuroscience, Cellular Neuroscience

Graphical abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Neurod4 is predominantly expressed in injured Xenopus laevis tadpole

  • •

    An LCMV-based pseudotyped retroviral vector has tropism to neural stem cells

  • •

    Neurod4 converts endogenous neural stem cells to neurons after spinal cord injury

  • •

    The new excitatory and inhibitory synaptic formation leads to functional recovery

Biological Sciences; Neuroscience; Behavioral Neuroscience; Cellular Neuroscience

Introduction

Treating traumatic spinal cord injury (SCI) is difficult, and individuals often have permanent and severe disabilities. These disabilities partially result from the human body's limited ability to repair and regenerate neural tissue in the spinal cord.

Our group previously utilized exogenous neural stem cells (NSCs) as carriers to deliver the genes of a neurotrophic factor and an anti-inflammatory cytokine to promote neural regeneration and inhibit reactive astrocytic proliferation and glial scar formation (Ito et al., 2009) (Nishimura et al., 2013) (Takeuchi et al., 2007). These studies utilized the unidirectional conception of differentiation (i.e., from undifferentiated to differentiated). However, after SCI, the ependymal cells lining the central canal (CC) of the spinal cord can dedifferentiate between the acute and subacute phases. Indeed, the ependymal cells can revert into radial glia and NSCs, and redifferentiate into glia and neurons as they migrate to the injury site (Barnabe-Heider et al., 2010) (Lacroix et al., 2014) (McDonough and Martinez-Cerdeno, 2012) (Meletis et al., 2008) (Stenudd et al., 2015). Therefore, we hypothesized that nerve regeneration may be achievable if endogenous injury-induced dedifferentiated NSCs can be reprogrammed into neurons.

The African clawed frog (Xenopus laevis) (X. laevis) has an unusually high capacity for neuroregeneration and is an indispensable animal model for regenerative research (Lee-Liu et al., 2017). During the larval stage, X. laevis undergoes a near-complete recovery after SCI (Lee-Liu et al., 2014). In this study, we screened X. laevis larvae for potential transgenes that, based on a comprehensive assessment of expression profiles, would exert a regenerative effect in mammals. Our transcriptomic analysis revealed that Neurod4 is a potential neuroregenerative transcription factor. Moreover, we used a pseudotyped retroviral vector to introduce the Neurod4 gene into the dedifferentiated NSCs, which were derived from ependymal cells after SCI, and aimed to switch their lineage toward neurons and thereby improve neural function.

Results

Comparison of bHLH gene expression after SCI in Xenopus laevis tadpoles and mice

X. laevis is an excellent model to research neuroregeneration and exhibits regenerative and nonregenerative (NR) stages. The X. laevis tadpole functionally recovers after SCI, a capacity it loses once it metamorphosizes into its juvenile froglet form. Lee-Liu et al. (Lee-Liu et al., 2014) conducted a whole transcriptome analysis (i.e., RNA-seq) between the regenerative and NR stages at 1, 2, and 6 days post injury (DPI). Our team compared the expression of neural regeneration-associated genes and glial scar formation-associated genes between the regenerative (R) stage as a tadpole and the NR stage as a froglet, respectively.

Basic-helix-loop-helix (bHLH) transcription factors are functionally critical proteins that regulate cell proliferation; cell differentiation; cell lineage determination; the formation of muscle, neurons, gut, and blood; sex determination; and other essential developmental and genetic processes (Dennis et al., 2019). Therefore, we hypothesized that one or more key bHLH factors may exist for neural regeneration in the X. laevis RNA-seq data. Thus, we analyzed the whole transcriptome data of 107 bHLH transcriptional factor genes (Liu and Li, 2015) after SCI in X. laevis. We found that the upregulation of Neurod4 mRNA was pronounced at 2 DPI of the regeneration stage (Figure 1A). Moreover, this time point corresponded to the time when the stem cell marker Sox-2 significantly increases (Gaete et al., 2012).

Figure 1.

Figure 1

Open in a new tab

Comparison of the candidate gene expressions in the acute phase between Xenopus laevis tadpoles and aged mice after spinal cord injury

(A) The nonclustering heatmap shows the differential expression of basic-helix-loop-helix (bHLH) transcription factors during the regenerative (R) and nonregenerative (NR) stages in the spinal cord after injury in Xenopus laevis (X. laevis).

(B) Quantification of mRNA expression levels for candidate genes (Neurod4, Neurod1, Atoh1, Neurog2, Ascl1) in the mouse SCI model. Gene expression levels were compared to those of the Sham (n = 3 mice per group). Statistical analysis was performed using student's t-test: ∗p < 0.05, ∗∗p < 0.005, ∗∗∗p < 0.001 and N.S. = not significant. P = 0.1770 to 1 DPI, 0.7074 to 3 DPI in Neurod4; P = 0.0225 to 1 DPI, 0.2071 to 3 DPI in Neurod1; P = 0.0022 to 1 DPI, 0.0328 to 3 DPI in Atoh1; P = 0.0073 to 1 DPI, 0.0329 to 3 DPI in Neurog2; P < 0.0001 to 1 DPI, 0.1935 to 3 DPI in Ascl1. Data represent the mean ± S.D.

(C) Superior expression of genes of X. laevis tadpole (in both L and S chromosomes) to aged mice in the stage for nerve regeneration. The bars indicate that the sum of FPKM ratios in the injured group over the sham group in X. laevis after SCI at 1 and 2 DPIs in the R stage (R1 +R2) was divided by the sum of qPCR values in the SCI group over the sham group at 1 and 3 DPIs (1 DPI +3 DPI). FPKM, fragments per kilobase of transcript per million. DPI, days post injury.

See also Table S1.

We also conducted a quantification of mRNA expression levels for candidate genes (Neurod4, Neurod1, Atoh1, Neurog2, Ascl1) (n = 3 mice per group), which are known as critical genes for nerve regeneration in the mouse SCI model (Pataskar et al., 2016) (Matsuda et al., 2019) (Lu et al., 2015) (Sayyid et al., 2019) (Smith et al., 2016) (Pollak et al., 2013) (Voronova et al., 2011). Among the candidate genes, Neurod4 showed no increase in expression after SCI compared to the Sham; therefore it was assumed that the exogenous introduction of Neurod4 most likely had a complementary effect to promote nerve regeneration (Figure 1B).

We further confirmed that the acute-phase expression of Neurod4 in X. laevis tadpole at 1 and 2 DPIs at the regenerative stage was substantially superior to injured mice at 1 and 3 DPIs among the candidate genes (Figure 1C). Based on these results, we focused on Neurod4 for a deeper analyses.

Preferential transduction of pseudotyped retrovirus with envelope of Lymphocytic choriomeningitis virus tropic to NSCs

Acutely after SCI in mice, the ependymal cells lining the CC begin to dedifferentiate into multipotent stem cells (Llorens-Bobadilla et al., 2015), which express the NSC marker: Nestin and Sox-2. Unlike lentiviruses or adeno-associated viruses (AAVs), retroviruses only infect dividing cells such as activated NSCs and progenitor cells, and do not infect nondividing cells such as neurons (Zhao et al., 2006). Lymphocytic choriomeningitis virus (LCMV) is a neurotropic RNA virus, which preferentially infects neural stem and progenitor cells; they can cause meningitis and encephalitis by establishing a persistent infection (Kang and McGavern, 2008) (Puccini et al., 2014) (Stein et al., 2005). We generated a pseudotyped retroviral vector with envelope glycoproteins derived from LCMV. The vector was designed to transduce replicating NSCs and therein transfer a transgene (Figure S1A). Three days before inducing an SCI, we injected the LCMV pseudovirus with an aqueous green fluorescence protein (AcGFP1 or tauAcGFP1) into the cisterna magna. Indeed, most of BrdU-positive dividing cells around the CC displayed AcGFP1 at 3 and 5 DPI (Figure 2A). Furthermore, Nestin-positive cells steadily increase after SCI, and peaks at approximately 3 DPI (Mao et al., 2016; Stenudd et al., 2015). We observed that approximately 70% cells around the CC were Nestin-positive and Sox-2-positive at 7 DPI, whereas no Nestin-positive cells were in the Sham-operated spinal cord (Figures 2B and S2D). These findings indicated that the LCMV pseudotyped virus could not infect the ependymal cells, but was restricted to dividing activated NSCs surrounding the CC, which are derived from dedifferentiated ependymal cells after SCI.

Figure 2.

Figure 2

Open in a new tab

Preferential transduction of pseudotyped retrovirus with envelope of Lymphocytic choriomeningitis virus tropic to neural stem cells

(A) Representative images of AcGFP1 (an aqueous green fluorescent protein) (green), BrdU staining (red), and DAPI (blue) of cells around the ependymal cells lining the central canal (CC) in the sham and the SCI at 3 and 5 DPI. High-magnification images of the center panels are shown in the right panels. Arrowheads indicate cells that are both virus infected (AcGFP1-positive) and dividing (BrdU-positive), these can be observed around the CC of SCI at 3 and 5 DPI; however these cannot be observed in the sham.

(B) Representative images of tauAcGFP1 (green), Nestin (red), and DAPI (blue) in the sham and the SCI at 7 DPI. A high-magnification image of the center panel is shown in the right panel. Arrowheads indicate cells that are virus infected (tauAcGFP1-positive) and Nestin-positive neural stem cells. The graphs show the percentage of Nestin-positive cells around the CC in the sham and the SCI groups at 7 DPI. The number of Nestin-positive cells in four separated fields were counted in the sham and the SCI groups at 7 DPI (Figure S2C). The efficiency of the virus in infecting activated neural stem cells is calculated as the ratio of green fluorescent protein (GFP)-positive cells to Nestin-positive cells. Statistical analysis was performed using student's-t test: ∗∗∗∗p < 0.00005. Data represent the mean ± S.D. CC, central canal; BrdU, Bromodeoxyuridine, 5-bromo-2′-deoxyuridine; DAPI, 4′,6-diamidino-2-phenylindole.

See also Figures S2 and S3.

Differentiation of activated NSCs into NeuN- and DCX-positive neurons by introducing Neurod4 into mice

We next evaluated the regenerative capacity of Neruod4 in mice. We subcloned murine Neurod4, orthologs of neurod4.L and neurod4.S, and constructed a murine Neurod4-expressing retroviral vector (Figure S1A). We investigated whether introducing murine Neurod4 into activated NSCs after SCI would result in neuronal regeneration. We chose DCX as a marker for immature neurons and NeuN as a marker for mature neurons. At 5 DPI, there was an obvious increase in the number of DCX-positive cells among Neurod4-introduced cells (i.e., AcGFP1-positive) (Figure 3A). Moreover, from 7 DPI to 42 DPI, we observed a progressive increase in the number of NeuN-positive cells among Neurod4-introduced (i.e., tauAcGFP1-positive) cells (Figures 3B and S3A). This increase was more pronounced than the tauAcGFP1-only introduced control cells (Neurod4 versus the control: 32.6% ± 6.7% vs. 8.6% ± 2.2% on 7 DPI, P = 0.0143; 46.7% ± 2.2% vs. 20.7% ± 2.0% on 42 DPI, p < 0.0001) (Figure 3B). A quantitative polymerase chain reaction (qPCR) confirmed that the expression of NeuN (i.e., Rbfox3 mRNA) was significantly increased in the Neurod4 group, compared to the control group (Figure S3B). Thus, SCI may induce ependymal cells to dedifferentiate into activated NSCs and introducing Neurod4 into these cells may facilitate their differentiation into neurons.

Figure 3.

Figure 3

Open in a new tab

Differentiation of activated neural stem cells into NeuN- and DCX-positive neurons by introducing Neurod4 into mice

(A) Neuronal differentiation from Neurod4-introduced neural stem cells at 5, 7, and 42 DPI. Representative images of AcGFP1 (green) and DCX (red) in the AcGFP1 and Neurod4-introduced groups at 5 DPI. High-magnification images of left panels are shown in the right panels. Arrowheads indicate Neurod4-introduced (AcGFP-positive) and immature (DCX-positive) neurons.

(B) Representative images of tauAcGFP1 (green), NeuN (red), and DAPI (blue) in the tauAcGFP1 and Neurod4-introduced groups at 7 and 42 DPI. High-magnification images of left panels are shown in right panels. Arrowheads indicate Neurod4-introduced (AcGFP-positive) and mature (NeuN-positive) neurons. The graphs show the percentage of NeuN-positive cells in GFP-positive cells around the CC in the tauAcGFP1 and Neurod4-introduced group at 7 and 42 DPI. Statistical analysis was performed using student's-t test: ∗p < 0.05 and ∗∗∗p < 0.0005. Data represent the mean ± S.D. DCX, doublecortin.

Promotion of the differentiation of excitatory and inhibitory neurons by Neurod4 expression in activated NSCs

Excitatory and inhibitory neurons and their complex neuronal networks in the spinal cord coordinate a wide variety of motor functions (Ramirez-Jarquin and Tapia, 2018) (Zholudeva et al., 2018). To identify the lineage of neurons promoted by introduction of Neurod4, we quantified gene expression related to neurotransmitter biogenesis and release. We assessed the expression of Slc17a6 (i.e., VGlut2, an excitatory neuronal marker) and Slc6a5 (i.e., glycine transporter 2 [GlyT2], an inhibitory neuronal marker), and choline acetyltransferase (ChAT), a transferase enzyme that synthesizes the neurotransmitter acetylcholine in motor neurons, in Neurod4-introduced cells after SCI.

We visualized Slc17a6 and Slc6a5 using in situ hybridization and ChAT using immunohistochemistry. The expression of Slc17a6 and Slc6a5 were elevated in the Neurod4-introduced group at 42 DPI (Figure 4A). Based on qPCR, we likewise confirmed that at 42 DPI the expression of Slc17a6 and Slc6a5 are elevated in the Neurod4 group, compared to the AcGFP1-introduced control ([Neurod4 vs. control] Slc17a6/AcGFP1: P = 0.0128; Slc6a5/AcGFP1: P = 0.0003, Figure 4B). Among AcGFP1-positive Neurod4-expressing cells, 49% were differentiated into Slc17a6-expressing excitatory neurons, 34% were differentiated into Slc6a5-expressing inhibitory neurons, and 17% were differentiated into other typed neurons such as motor neurons (Figure 4C). Moreover, we observed ChAT-positive motor neurons derived from Neurod4-introduced cells were dispersed around the CC (Figure 4D). Thus, Neurod4-expressing NSCs may differentiate into excitatory neurons, inhibitory neurons, or motor neurons, which may form neuronal networks and contribute to functional improvement after SCI.

Figure 4.

Figure 4

Open in a new tab

Promotion of the differentiation of excitatory and inhibitory neurons by Neurod4 expression in activated neural stem cells

(A) Representative images depicting in situ hybridization (ISH) analysis of Slc17a6 (purple), the mRNA for the excitatory neuronal marker VGlut2, and Slc6a5 (red), the mRNA for the inhibitory neuronal marker GlyT2, in AcGFP1-positive cells expressing Neurod4 at 42 DPI.

(B) Relative quantification of mRNA expression levels of neuronal subtype marker genes, Slc17a6 and Slc6a5 in AcGFP1 and Neurod4-expressing spinal cords at 42 DPI. Statistical analysis was performed using student's-t test: ∗p < 0.05 and ∗∗p < 0.005. Data represent the mean ± S.D.

(C) The percentage of neuronal subtypes differentiated from Neurod4-introduced GFP-positive cells. The number of positive cells was calculated using the H-score of RNAscope as the semi-quantitative expression level.

(D) Representative images of ChAT (red) in Neurod4-introduced AcGFP1-positive cells (green) in the spinal cord of the Neurod4-introduced group mice at 42 DPI. Images are zoomed up to right panels in rectangles of left panels. Arrowheads indicate the expression of ChAT in Neurod4-introduced AcGFP1-positive cells around CC. ChAT, choline acetyltransferase.

Neurod4-induced excitatory and inhibitory neurons form functional synapses

Next, we investigated the ability of Neurod4-induced excitatory and inhibitory neurons to form functional synapses. We constructed an LCMV pseudoretroviral vector with Neurod4 and Syp-AcGFP1 transgenes (i.e., AcGFP1-tagged murine synaptophysin) (Figure S1A). We observed that Neurod4-induced excitatory (i.e., VGlut2-positive) neurons projected predominantly to the postsynaptic density (PSD)-95-positive subsynaptic domains of the motor neurons at levels L2–L5 of the spinal cord from excitatory neurons in the dorsal portion of the gray matter at the epicenter, laminae IV, V (Figures 5A and 5B). We semi-quantitatively observed approximately 4 excitatory synapses and 12 excitatory synapses in the ventral horn at the epicenter and at levels L2-L5 of the spinal cord, respectively. Conversely, Neurod4-induced inhibitory (i.e., GlyT2-positive) neurons predominantly projected to the GlyR-positive subsynaptic domains of the motor neurons at the epicenter from inhibitory neurons in the intermediate portion of the gray matter at epicenter, laminae V, VI (Figures 5C and 5D). We also semi-quantitatively observed approximately 25 inhibitory synapses and 4 inhibitory synapses in the ventral horn at epicenter and levels L2–L5 of the spinal cord, respectively.

Figure 5.

Figure 5

Open in a new tab

Projections into motor neurons of Neurod4-introduced excitatory and inhibitory neurons

(A) Representative images of cell soma of Slc17a6-expressing (excitatory) Neurod4-introduced Syp-AcGFP1-expressing neurons at the dorsal portion of the gray matter, laminae (IV–V) and excitatory synapses in the motor neurons at the epicenter and at levels L2-L5 of the spinal cord. Arrows indicate the cell soma labeled by Slc17a6 (excitatory) (purple) in Neurod4-introduced cells (Syp-AcGFP1, green), and arrowheads indicate synapses labeled by presynaptic markers from Neurod4-introduced cells (Syp-AcGFP1, green), and the excitatory postsynaptic marker (PSD-95, red)-positive in the motor neurons at the epicenter and at levels L2-L5 of the spinal cord.

(B) Schematic cartoon of projection of newly formed relay neurons from the epicenter to motor neurons at the levels L2-L5 of the spinal cord.

(C) Representative images of the cell soma of Slc6a5-expressing (inhibitory) Neurod4-introduced Syp-AcGFP1-expressing neurons at the intermediate portion of the gray matter, (laminae V–VI) and inhibitory synapses in the motor neurons at the epicenter and levels L2-L5 of the spinal cord. Arrows indicate the cell soma labeled by Slc6a5 (inhibitory) (red) from Neurod4-introduced cells (Syp-AcGFP1, green), and arrowheads indicate synapses labeled by presynaptic markers from Neurod4-introduced cells (Syp-AcGFP1, green), and the inhibitory postsynaptic marker (GlyR, red)-positive in the motor neurons at the epicenter and levels L2-L5 of the spinal cord.

(D) Schematic cartoon of projection of newly formed inhibitory neuron from epicenter to motor neurons at the epicenter. MN, motor neuron.

See also Figure S4.

Furthermore, a structured illuminated microscope on the sections identical to those in Figures 5A and 5C was used to measure the distance between the pre- and postsynapses in the excitatory and inhibitory neuronal formations (Figure S4A). The mean distances were 202 nm and 135 nm for the excitatory and inhibitory synapses, respectively (Figure S4B). These findings are consistent with recent studies (Crosby et al., 2019) (Wang et al., 2016). Thus, Neurod4-regenerated excitatory and inhibitory neurons may form functional regulatory circuits that relay to and modulate motor neurons, respectively.

Suppression of GFAP-positive astrocytes and improvement of spinal cord regeneration by Neurod4 expression after SCI

Previous work has established that astrocytes derived from ependymal cells form glial scars after SCI and comprises more than one-half of all glial scarring (Stenudd et al., 2015). Reactive astrocytes derived from naive astrocytes also contribute to glial scarring (Hara et al., 2017). Since glial scars impair the regrowth of axons and inhibit neuronal regeneration, we investigated the formation of astrogliosis at 7 and 42 DPI. Control (tauAcGFP1-introduced) astrocyte underwent a typical change of hypertrophy, process extension at 7 DPI, while Neurod4-introduced astrocyte didn't. Furthermore, the number of astrocytes decreased in Neurod4-introduced mice compared to control mice at 42 DPI (Figure 6A). Approximately 20% of cells around the CC were GFAP-positive in the Neurod4 and tauAcGFP1-control groups. Unlike the progressive increase occurring in NeuN-positive cells, GFAP-positive cells decreased to approximately 10% in the Neurod4 group but increased to approximately 40% in the control group by 42 DPI (Figure 6B). A qPCR test confirmed that the GFAP expression was significantly reduced in the Neurod4 group, compared to the control group, at 42 DPI. However, no difference was observed at 7 DPI (Figure S3C). Thus, Neurod4 may potentially reduce astrocyte differentiation by diverting ependymal cells toward a neuronal lineage instead of an astrocytic lineage.

Figure 6.

Figure 6

Open in a new tab

Glial scar suppression and axonal tracing from projection neurons of M1 cortex beyond injured region of spinal cord after recovery

(A) Representative images of GFAP (red) and DAPI (blue) in tauAcGFP1 and Neurod4-introduced cells at 7 and 42 DPI. Arrowheads indicate the expression of GFAP protein.

(B) The percentage of GFAP-expressing astrocytes relative to DAPI-labeled cells in the injured spinal cord of tauAcGFP1 and Neurod4-introduced mice at 7 and 42 DPI. Statistical analysis was performed using student's-t test: ∗∗p < 0.005; and N.S. = not significant. Data represent the mean ± S.D.

(C) Tissue clearing images of the axons in the corticospinal tract of the post-recovery spinal cord show pal-mKate2-labeled axons (red) in the control group (left, AcGFP1-introduced) and the Neurod4-introduced group (right). White broken line indicates the border of the injury site.

(D) Axial images of the axons in the corticospinal tract at the rostral, epicenter and caudal regions are represented as a red dot. High-magnification images of upper panels are shown in lower panels in each sample. A yellow broken line indicates the border between white matter and gray matter. Arrows indicate the axons through corticospinal tract in the caudal region.

See also Figures S3 and S5.

To examine an axonal regrowth after SCI, we constructed a palmitoylated red fluorescent protein, mKate2, with a palmitoylation signal derived from GAP-43 (pal-mKate2) (Figure S1B). AAV-Syn-pal-mKate2 infected the M1 cortex pyramidal neurons (Figure S5B) and labeled axons in the corticospinal tract (Figure S5A). We observed axons in the samples collected from the rostral, epicenter, and caudal regions (Figure S5C). In the Neurod4 group, the tissue clearing illustrated more axons visible at the epicenter while in the control (AcGFP1), the axons stopped at rostral to the lesion (Figure 6C), and observed a red signal in the corticospinal tract as indicated in Figure S2C at rostral, epicenter, and also caudal region beyond the injury site (Figure 6D). Therefore, Neurod4 may promote axonal regrowth by suppressing glial scar formation.

Synaptic formation and functional recovery by Neurod4 after SCI

We next aimed to visualize the corticospinal tract connection between the M1 cortex and Neurod4-induced neurons in the dorsal portion of the gray matter at epicenter or motor neurons at levels L2-L5 of the spinal cord. We injected the AAV expressing the murine synaptophysin tagged with mKate2 (Syp-mKate2) into the M1 cortex according to schedule (Figures S1B and S5A). In this manner, Syp-mKate2 was transferred to the presynaptic terminal. At the epicenter or levels L2–L5 of the spinal cord, the PSD was double-stained with anti-GFP or anti-ChAT antibody and with anti-PSD-95, respectively (Ippolito and Eroglu, 2010). In the dorsal portion of the gray matter at the epicenter, excitatory synapses were found around the AcGFP1-positive cells only in the Neurod4-introduced group (Figure 7A). Moreover, the Neurod4 group had a substantial number of Syp-mKate2 puncta with ChAT-positive neurons and PSD-95 puncta in the motor neurons at levels L2-L5 of the spinal cord, compared to the AcGFP1-introduced control group or sham group (Figure 7B). The number of newly formed synapses in the ventral horn were markedly increased in the Neurod4 group, compared to the AcGFP1-introduced control group (Figure 7C). Thus, synapse formations at the motor neurons were achieved beyond the injury site in the Neurod4 group.

Figure 7.

Figure 7

Open in a new tab

Synaptic formation and functional recovery by Neurod4 after SCI

(A) Excitatory synapses detected with PSD-95 antibody (light blue) and presynaptic marker (Syp-mKate2) (red) around AcGFP1-expressing cells at the dorsal portion of gray matter, laminae IV–V, at the epicenter of AcGFP1 or Neurod4-introduced group at 42 DPI. Arrowheads indicate the excitatory synapses labeled by a postsynaptic marker (PSD-95) (light blue) and presynaptic marker (Syp-mKate2) (red).

(B) Excitatory synapses detected with PSD-95 antibody and presynaptic marker (Syp-mKate2) around ChAT-expressing cells at levels L2-L5 of the spinal cord of AcGFP1, Neurod4-introduced or sham-operated group at 42 DPI. The left panels show the motor neurons stained with anti-ChAT (light blue) antibody in the ventral horn at levels L2-L5 of the spinal cord beyond the injury site in the AcGFP1 and Neurod4-introduced mice at 42 DPI, respectively. Yellow arrowheads indicate ChAT-neurons surrounded with presynaptic markers, Syp-mKate2 (red). The right panels show excitatory synapses detected with PSD-95 (light blue) and Syp-mKate2 (red) in the ventral horn at levels L2-L5 of the spinal cord beyond the injury site in the AcGFP1 and Neurod4-transduced mice at 42 DPI. The lower panel shows excitatory synapses detected with PSD-95 antibody and presynaptic marker (Syp-mKate2) at levels L2-L5 of the spinal cord of sham-operated mice. Arrowheads and arrows indicate functional excitatory synapses and clusters of synapses, respectively.

(C) The number of functional synapses in an area (1013 mm2) in the ventral horn beyond the injury site at levels L2-L5 of the spinal cord of mice at 42 DPI (3 mice per group). Statistical analysis was performed using an unpaired t test: ∗∗∗∗p < 0.00005. Data represent the mean ± S.D.

(D) Improvement in hindlimb locomotor function was evaluated using the Basso Mouse Scale (BMS) in Neurod4 and AcGFP1-introduced mice for 6 weeks (5 mice per group). Statistical analysis was performed using student's t-test: ∗p < 0.05, ∗∗p < 0.01. Data represent the mean ± S.E.M.

(E) Summarized schematic cartoon. Neurod4 introduction differentiates the ependymal-derived neural stem cells into VGlut2-positive-excitatory, GlyT2-positive inhibitory, and ChAT-positive motor neurons. Descending fibers from the M1 cortex synapse with these neurons and motor neurons at levels L2-L5 of the spinal cord, respectively. The VGlut2-positive excitatory neurons relay to motor neurons at levels L2-L5 of the spinal cord.

To further demonstrate functional rescue, we assessed hindlimb locomotor function by using the Basso Mouse Scale (BMS). For 6 weeks post-SCI, mice were evaluated for locomotor recovery. The Neurod4-introdeced mice showed a significant improvement in locomotor function, compared to the control mice (Neurod4 group vs. control group: at 1 week postinjury [WPI], 2.4 ± 1.1 vs. 0.30 ± 0.20, P = 0.056; at 6 WPI: 4.3 ± 1.1 vs. 0.60 ± 0.40, P = 0.0080) (Figure 7D).

Discussion

In this study, we used pseudotyped retroviral vectors to introduce the Neurod4 gene into the dedifferentiated NSCs, derived from the ependymal cells after SCI, and aimed to switch their lineage toward neurons and thereby improve neural function. We utilized two innovative research approaches. First, we conducted transcriptome analysis of X. laevis to identify transcription factors that could potentially regulate nerve regeneration. Our team comprehensively analyzed the highly expressed bHLH transcriptomes from tadpoles during a nerve regeneration after SCI (Lee-Liu et al., 2014). The bHLH transcription factor Neurod4 was dramatically expressed at 2 DPI during the regenerative stage. Furthermore, this time point corresponded to the time when the stem cell marker Sox-2 significantly increases (Gaete et al., 2012) (Figure 1A). In addition, we compared the expression levels of the candidate genes in the acute phase after mouse SCI. Among the candidate genes, the introduction of Neurod4 may be the most effective in improving the neurological function. Second, we developed a unique pseudotyped retrovirus that uses an LCMV envelope and that can therefore selectively infect mitotic (activated) NSCs and maintain constant expression of a transgene. We used this vector to deliver Neurod4 to the activated NSCs that were formed from dedifferentiated ependymal cells after SCI. To the best of our knowledge, there has been no study that used an LCMV-based retroviral vector to introduce Neurod4 into the activated NSCs to treat SCI.

In the present study, we focused on the fact that NSCs derived from ependymal cells lining the CC are activated and proliferate after injury (Lacroix et al., 2014) (Stenudd et al., 2015). This research concept is completely different from the previously described stem cell transplantation (Li and Lepski, 2013) (Nakamura and Okano, 2013) (Yousefifard et al., 2016): in our case, a neuroregenerative gene was transferred directly into endogenous ependymal-derived NSCs that are already present in the spinal cord. From the acute phase to the subacute phase after an SCI, before the formation of a glial scar, the ependymal cells along the CC dedifferentiate into activated NSCs, and then predominantly differentiate into astrocytes, oligodendrocytes, and, to a lesser degree, migrating neurons (Barnabe-Heider et al., 2010) (McDonough and Martinez-Cerdeno, 2012) (Stenudd et al., 2015). Thus, the number of putative NSCs peaks at 3 DPI, and then gradually declines (Mao et al., 2016) (Mothe and Tator, 2005). We demonstrated that the cell fate of these activated NSCs could be reprogrammed through genetic engineering.

We also demonstrated that, after SCI, transducing spinal cord-based NSCs with Neurod4 facilitated their differentiation into diverse types of neural networks (i.e., excitatory and inhibitory neurons and motor neurons), and thereby facilitated a functional recovery. Several treatment approaches for SCI have been reported; however, directly reprogramming endogenous cells into neurons by using neuronal factors has been recently attempted (Gascon et al., 2016) (Masserdotti et al., 2016). Through these experiments, the key signaling pathways and transcriptional programs that instruct neuronal diversity during development have largely been identified.

In this study, we introduced Neurod4 as a transgene to differentiate endogenous NSCs into 3 or more types of neurons. Transcription factors of the bHLH family have emerged as key determinants of neural cell fate specification and differentiation. The NeuroD family comprises four closely related neuronal bHLH transcription factors (i.e., Neurod1/2/4/6), which have important roles in pyramidal neuron differentiation and embryonic cortex development (Lee et al., 1995). The patterning of these expressions have been reported in many studies. During mammalian development, neural progenitors express specific transcription factors in 11 compartments of the ventricular zone and in 13 compartments of the mantle zone. Neurog1, Ascl1, and Olig2 are expressed in neural progenitors at dp2, dp3, and pMN in the ventricular zone, respectively, while Neurog2, Neurod1, and Neurod4 are expressed in differentiating neurons at dl2, dl3, V1, V2a, and MN in the mantle zone. Extensive expression of Neurog3, Neurod1, and Neurod4 is observed at V3 in the mantle zone (Lai et al., 2016) (Delile et al., 2019). Expression of Atoh1 is followed by Neurog1/2, which is followed by Ascl1 (Bermingham et al., 2001). Neurog1/2 upregulates Neurod1, followed by Neurod4 (Seo et al., 2007) (Rea et al., 2020). Neurod4 is located far downstream in the bHLH family cascade. As a downstream target of Neurog2 and Ascl1, NeuroD4 induces late-stage neuronal reprogramming (Delile et al., 2019) (Masserdotti et al., 2015) (Sugimori et al., 2008) (Wapinski et al., 2013).

Excitatory and inhibitory neurons and their complex neuronal networks of the spinal cord coordinate a wide variety of motor functions (Ramirez-Jarquin and Tapia, 2018) (Zholudeva et al., 2018). Excitatory neurons, inhibitory neurons, and motor neurons reside in the compartments of dl2, dl3, V2a, and V3, in V1, and in MN, respectively. Therefore, VGlut2-positive excitatory neurons, GlyT2-positive inhibitory neurons, and ChAT-positive motor neurons could be finally converted by Neurod4.

Among these neuronal subtype cells, we observed that axons of excitatory neurons predominantly projected to motor neurons caudal to the injury site at the L2-L5 level of the mice spinal cord from VGlut2-positive cell soma, whereas inhibitory axons projected to motor neurons at the SCI epicenter from GlyT2-positive cell soma in the same level with axonal terminals (Figures 5A and 5B). These results indicated that the vGlut2-positive cells observed in our study may have a role as relay neurons and transmit excitation from the M1 cortex to caudal motor neurons. In addition, GlyT2-positive cells may modulate glutamatergic input from the M1 cortex to motor neurons at the epicenter.

The reticulospinal tract (Kuypers, 1981) (Holstege and Kuypers, 1982) (Holstege and Kuypers, 1987) (Jones and Yang, 1985) (Martin et al., 1985), vestibular spinal tract (Chen et al., 2012), tectospinal tract (Kuypers, 1981), corticospinal tract (Fink et al., 2015), and rubrospinal tract (Wang et al., 2011) (Ueno et al., 2018) (Bareyre et al., 2005); (Soderblom et al., 2015) control coordinated whole-body posture or balance, orienting movements, and locomotion. Among these, the corticospinal tract is the main pathway controlling voluntary movements from the primary motor cortex to motor neurons. To investigate the projections from the M1 cortex to the ventral horn in L2-L5, we conducted tissue clearing to trace the axons with AAV-mediated fluorescent protein labeling.

In Figure 7A, we see that, after SCI, only in the Neurod4-introduced group do axons of pyramidal neurons of the M1 cortex project to the dorsal gray matter portion at the epicenter level of the spinal cord. The corticospinal fibers terminate in all laminae but most heavily in laminae IV–V (Casale et al., 1988) (Liang et al., 1991). Input from the M1 cortex would then be relayed to the newly formed excitatory (i.e., VGlut2-positive) or inhibitory (i.e., Gly-T2-postive) cells. Moreover, axons of pyramidal neurons in the M1 cortex projected directly to motor neurons at levels L2–L5 of the spinal cord following SCI in the Neurod4-introduced group. Furthermore, there was a larger amount of synapse in the Neurod4 group than the control (AcGFP1 group) and the sham group (Figure 7B). In intact rodent spinal cord, the connections from the M1 cortex to the motor neurons in L2-5 are disynaptic and little monosynaptic (Liang et al., 1991) (Alstermark et al., 2004). The increased number of synapses in Figure 7B were newly formed by Neurod4 introduction although they could not be proven to be electrophysiologically functional (Alstermark et al., 2004).

We observed that Neurod4 introduction interestingly suppressed glial scar formation (Figure 6A). The glial scar formed during the subacute phase after SCI is a physical barrier that prevents axonal regrowth after SCI. Therefore, suppressing glial scar formation represents an important strategy for treating SCI. A glial scar comprises (1) astrocytes, which differentiate from NSCs derived from ependymal cells and are located in the inner portion of the scar, and (2) reactive astrocytes that migrate to the injury site and are located in the outer scar (Hara et al., 2017). In the present study (Figure 6), introducing Neurod4 into activated NSCs promoted their differentiation into neurons and thereby decreased GFAP-positive cells in inner portion of a glial scar (Hara et al., 2017). The conversion of NSCs into neurons instead of glial cells has been reported elsewhere (Hara et al., 2017; (Matsuda et al., 2019). Thus, transducing activated NSCs with Neurod4 promotes neuronal differentiation and, as a secondary effect, suppress the formation of glial scars.

Suppressing glial scar formation could also lead to axonal outgrowth from differentiated neurons and regrowth of axons from pyramidal neurons in the M1 cortex beyond the damaged region. This process may lead to a functional recovery because the BMS score significantly improved in the Neurod4 group mice, compared to the control group (Figure 7D). Improvement commenced at 2 WPI and gradually occurred from 3 to 6 WPI. This result indicated an incomplete restoration of hindlimb movement. We speculate this result occurred because the precise neuronal network governing movement remains unrefined. Nevertheless, our study showed synaptic connections from the M1 cortex through Neurod4-introduced relay neurons or direct monosynaptical to motor neurons in the lumbar region. Thus, some function was restored.

In conclusion, we demonstrated that introducing Neurod4 into activated NSCs after SCI facilitates the production of relay neurons, which project to motor neurons of the hindlimbs. As a secondary effect, glial scar formation was suppressed after the subacute phase of SCI. This effect allowed an environment that was conducive for axons to elongate beyond the injury site and form synapses to motor neurons monosynaptically and thereby improve motor function in the hindlimbs.

Limitations of the study

We have to develop a treatment protocol to treat patients suffering from loss of function, such as mobility and/or feeling.

In this study, we only used a mouse model because it is easy to induce SCI and is accessible to gene delivery by using a virus. LCMV has tropism in NSCs. The name is based on the tendency of an individual to have abnormally high levels of lymphocytes during infection. Common symptoms by LCMV infection include fever, lack of appetite, headache, muscle aches, malaise, nausea, and/or vomiting. The onset of the second phase occurs several days after recovery and consists of symptoms of meningitis or encephalitis. Furthermore, retrovirus is integrated into host genome; this integration would cause insertional mutation and induce tumors. In our study, we only used LCMV's envelope and constructed a pseudotyped retroviral vector. Although it is safe to use a pseudotyped retrovirus, since a retrovirus is unable to self-replicate, further experiments are needed to confirm the safety and efficacy of using a retroviral vector.

Neural circuits in the brain and spinal cord are quite different between primates and rodents. Connections from the M1 cortex to the motor neurons governing the locomotion of the hindlimb are mainly disynapse in rodents, while the connection is monosynapse in primates. In this study, after SCI, newly formed relay neurons connected directly to motor neurons, and axons from the M1 cortex innervated monosynaptically. However, we only determined the newly formed neural network histologically. We were unable to know whether the relay neurons or the monosynaptic connections from the M1 cortex to the motor neurons preferentially transmit input from the brain to the motor neurons. Therefore, we need to electrophysiologically confirm the presence of an efficient neural circuit after recovering from the injury.

Despite these limitations, our study demonstrated that regeneration from endogenous NSCs to neurons occur. Furthermore, the establishment of new neural circuit leads to an improvement of the hindlimb locomotion after SCI.

Resource availablity

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Atsushi Natsume (anatsume@med.nagoya-u.ac.jp).

Material availability

All unique/stable reagents generated in this study are available from the lead contact without restriction.

Data and code availability

The raw sequencing data for the RNA-seq are deposited at the DNA Data Bank of Japan (Mishima, Shizuoka Prefecture, Japan; accession number DRA009306).

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

This work was supported by Grant-in-Aid for Scientific Research on Innovative Areas “Chemistry for Multimolecular Crowding Biosystems” Grant Number 17H06356 (AN), the JSPS Grant-in-Aid for Scientific Research (C) 19K09453 (Y Nishimura).

We thank K. Yamada, T. Nagai, and N. Ito from the Department of Neuropsychopharmacology and Hospital Pharmacy (Nagoya University School of Medicine, Nagoya, Japan) and E. Yorifuji from the Division for Medical Research Engineering at Nagoya University School of Medicine for their technical assistance.

Author contributions

A.N., A.K., and Y. Nishimura conceived and designed the project. T.F. and A.K. conducted all animal experiments. R.A., K.M., and T.W. analyzed all animal data. K.A., T.A., A.A., and S.M. conducted the histology and reverse transcriptase polymerase chain reaction (RT-PCR) experiments. Y. Nagashima performed the immunohistochemistry and analyzed the data. H. S., and J. Y. conducted the RT-PCR experiments. T.F. conducted the animal behavioral tests and analyzed the data. M.H., K.A., and F.O. analyzed the RNA-seq data. D.L. and L.L. conducted the whole transcriptomic experiments and analyzed the data. T.F., A.N., and A.K., assisted by M.H., wrote the paper.

Declaration of interests

The authors declare no competing interests.

Published: February 19, 2021

Footnotes

Supplemental information can be found online at https://doi.org/10.1016/j.isci.2021.102074.

Contributor Information

Yusuke Nishimura, Email: yusuken0411@med.nagoya-u.ac.jp.

Atsushi Natsume, Email: anatsume@med.nagoya-u.ac.jp.

Supplemental information

Document S1. Transparent methods, Figures S1–S5, and Tables S1–S4
mmc1.pdf (6.9MB, pdf)

References

  1. Alstermark B., Ogawa J., Isa T. Lack of monosynaptic corticomotoneuronal EPSPs in rats: disynaptic EPSPs mediated via reticulospinal neurons and polysynaptic EPSPs via segmental interneurons. J. Neurophysiol. 2004;91:1832–1839. doi: 10.1152/jn.00820.2003. [DOI] [PubMed] [Google Scholar]
  2. Bareyre F.M., Kerschensteiner M., Misgeld T., Sanes J.R. Transgenic labeling of the corticospinal tract for monitoring axonal responses to spinal cord injury. Nat. Med. 2005;11:1355–1360. doi: 10.1038/nm1331. [DOI] [PubMed] [Google Scholar]
  3. Barnabe-Heider F., Goritz C., Sabelstrom H., Takebayashi H., Pfrieger F.W., Meletis K., Frisen J. Origin of new glial cells in intact and injured adult spinal cord. Cell Stem Cell. 2010;7:470–482. doi: 10.1016/j.stem.2010.07.014. [DOI] [PubMed] [Google Scholar]
  4. Bermingham N.A., Hassan B.A., Wang V.Y., Fernandez M., Banfi S., Bellen H.J., Fritzsch B., Zoghbi H.Y. Proprioceptor pathway development is dependent on Math1. Neuron. 2001;30:411–422. doi: 10.1016/s0896-6273(01)00305-1. [DOI] [PubMed] [Google Scholar]
  5. Casale E.J., Light A.R., Rustioni A. Direct projection of the corticospinal tract to the superficial laminae of the spinal cord in the rat. J. Comp. Neurol. 1988;278:275–286. doi: 10.1002/cne.902780210. [DOI] [PubMed] [Google Scholar]
  6. Chen Y., Takano-Maruyama M., Fritzsch B., Gaufo G.O. Hoxb1 controls anteroposterior identity of vestibular projection neurons. PLoS One. 2012;7:e34762. doi: 10.1371/journal.pone.0034762. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Crosby K.C., Gookin S.E., Garcia J.D., Hahm K.M., Dell'Acqua M.L., Smith K.R. Nanoscale subsynaptic domains underlie the organization of the inhibitory synapse. Cell Rep. 2019;26:3284–3297.e3. doi: 10.1016/j.celrep.2019.02.070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Delile J., Rayon T., Melchionda M., Edwards A., Briscoe J., Sagner A. Single cell transcriptomics reveals spatial and temporal dynamics of gene expression in the developing mouse spinal cord. Development. 2019;146:dev173807. doi: 10.1242/dev.173807. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dennis D.J., Han S., Schuurmans C. bHLH transcription factors in neural development, disease, and reprogramming. Brain Res. 2019;1705:48–65. doi: 10.1016/j.brainres.2018.03.013. [DOI] [PubMed] [Google Scholar]
  10. Fink K.L., Strittmatter S.M., Cafferty W.B. Comprehensive corticospinal labeling with mu-crystallin transgene reveals axon regeneration after spinal cord trauma in ngr1-/- mice. J. Neurosci. 2015;35:15403–15418. doi: 10.1523/JNEUROSCI.3165-15.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Gaete M., Munoz R., Sanchez N., Tampe R., Moreno M., Contreras E.G., Lee-Liu D., Larrain J. Spinal cord regeneration in Xenopus tadpoles proceeds through activation of Sox2-positive cells. Neural Dev. 2012;7:13. doi: 10.1186/1749-8104-7-13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Gascon S., Murenu E., Masserdotti G., Ortega F., Russo G.L., Petrik D., Deshpande A., Heinrich C., Karow M., Robertson S.P. Identification and successful negotiation of a metabolic checkpoint in direct neuronal reprogramming. Cell Stem Cell. 2016;18:396–409. doi: 10.1016/j.stem.2015.12.003. [DOI] [PubMed] [Google Scholar]
  13. Hara M., Kobayakawa K., Ohkawa Y., Kumamaru H., Yokota K., Saito T., Kijima K., Yoshizaki S., Harimaya K., Nakashima Y., Okada S. Interaction of reactive astrocytes with type I collagen induces astrocytic scar formation through the integrin-N-cadherin pathway after spinal cord injury. Nat. Med. 2017;23:818–828. doi: 10.1038/nm.4354. [DOI] [PubMed] [Google Scholar]
  14. Holstege G., Kuypers H.G. The anatomy of brain stem pathways to the spinal cord in cat. A labeled amino acid tracing study. Prog. Brain Res. 1982;57:145–175. doi: 10.1016/S0079-6123(08)64128-X. [DOI] [PubMed] [Google Scholar]
  15. Holstege J.C., Kuypers H.G. Brainstem projections to spinal motoneurons: an update. Neuroscience. 1987;23:809–821. doi: 10.1016/0306-4522(87)90160-6. [DOI] [PubMed] [Google Scholar]
  16. Ippolito D.M., Eroglu C. Quantifying synapses: an immunocytochemistry-based assay to quantify synapse number. J. Vis. Exp. 2010:2270. doi: 10.3791/2270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ito M., Natsume A., Takeuchi H., Shimato S., Ohno M., Wakabayashi T., Yoshida J. Type I interferon inhibits astrocytic gliosis and promotes functional recovery after spinal cord injury by deactivation of the MEK/ERK pathway. J. Neurotrauma. 2009;26:41–53. doi: 10.1089/neu.2008.0646. [DOI] [PubMed] [Google Scholar]
  18. Jones B.E., Yang T.Z. The efferent projections from the reticular formation and the locus coeruleus studied by anterograde and retrograde axonal transport in the rat. J. Comp. Neurol. 1985;242:56–92. doi: 10.1002/cne.902420105. [DOI] [PubMed] [Google Scholar]
  19. Kang S.S., McGavern D.B. Lymphocytic choriomeningitis infection of the central nervous system. Front. Biosci. 2008;13:4529–4543. doi: 10.2741/3021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kuypers H.G. Anatomy of the descending pathways. In: Brookhart J.M., Mountcastle V.B., editors. Handbook of Physiology—The Nervous System II. American Physiological Society; 1981. pp. 597–666. [Google Scholar]
  21. Lacroix S., Hamilton L.K., Vaugeois A., Beaudoin S., Breault-Dugas C., Pineau I., Levesque S.A., Gregoire C.A., Fernandes K.J. Central canal ependymal cells proliferate extensively in response to traumatic spinal cord injury but not demyelinating lesions. PLoS One. 2014;9:e85916. doi: 10.1371/journal.pone.0085916. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Lai H.C., Seal R.P., Johnson J.E. Making sense out of spinal cord somatosensory development. Development. 2016;143:3434–3448. doi: 10.1242/dev.139592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee J.E., Hollenberg S.M., Snider L., Turner D.L., Lipnick N., Weintraub H. Conversion of Xenopus ectoderm into neurons by NeuroD, a basic helix-loop-helix protein. Science. 1995;268:836–844. doi: 10.1126/science.7754368. [DOI] [PubMed] [Google Scholar]
  24. Lee-Liu D., Mendez-Olivos E.E., Munoz R., Larrain J. The African clawed frog Xenopus laevis: a model organism to study regeneration of the central nervous system. Neurosci. Lett. 2017;652:82–93. doi: 10.1016/j.neulet.2016.09.054. [DOI] [PubMed] [Google Scholar]
  25. Lee-Liu D., Moreno M., Almonacid L.I., Tapia V.S., Munoz R., von Marees J., Gaete M., Melo F., Larrain J. Genome-wide expression profile of the response to spinal cord injury in Xenopus laevis reveals extensive differences between regenerative and non-regenerative stages. Neural Dev. 2014;9:12. doi: 10.1186/1749-8104-9-12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Li J., Lepski G. Cell transplantation for spinal cord injury: a systematic review. Biomed. Res. Int. 2013;2013:786475. doi: 10.1155/2013/786475. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Liang F.Y., Moret V., Wiesendanger M., Rouiller E.M. Corticomotoneuronal connections in the rat: evidence from double-labeling of motoneurons and corticospinal axon arborizations. J. Comp. Neurol. 1991;311:356–366. doi: 10.1002/cne.903110306. [DOI] [PubMed] [Google Scholar]
  28. Liu W., Li F. Identification and bioinformatics analyses of the basic helix-loop-helix transcription factors in Xenopus laevis. Pakistan J. Biol. Sci. 2015;18:149–165. doi: 10.3923/pjbs.2015.149.165. [DOI] [PubMed] [Google Scholar]
  29. Llorens-Bobadilla E., Zhao S., Baser A., Saiz-Castro G., Zwadlo K., Martin-Villalba A. Single-cell transcriptomics reveals a population of dormant neural stem cells that become activated upon brain injury. Cell Stem Cell. 2015;17:329–340. doi: 10.1016/j.stem.2015.07.002. [DOI] [PubMed] [Google Scholar]
  30. Lu F., Kishida S., Mu P., Huang P., Cao D., Tsubota S., Kadomatsu K. NeuroD1 promotes neuroblastoma cell growth by inducing the expression of ALK. Cancer Sci. 2015;106:390–396. doi: 10.1111/cas.12628. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mao Y., Mathews K., Gorrie C.A. Temporal response of endogenous neural progenitor cells following injury to the adult rat spinal cord. Front. Cell. Neurosci. 2016;10:58. doi: 10.3389/fncel.2016.00058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Martin G.F., Vertes R.P., Waltzer R. Spinal projections of the gigantocellular reticular formation in the rat. Evidence for projections from different areas to laminae I and II and lamina IX. Exp. Brain Res. 1985;58:154–162. doi: 10.1007/BF00238963. [DOI] [PubMed] [Google Scholar]
  33. Masserdotti G., Gascon S., Gotz M. Direct neuronal reprogramming: learning from and for development. Development. 2016;143:2494–2510. doi: 10.1242/dev.092163. [DOI] [PubMed] [Google Scholar]
  34. Masserdotti G., Gillotin S., Sutor B., Drechsel D., Irmler M., Jorgensen H.F., Sass S., Theis F.J., Beckers J., Berninger B. Transcriptional mechanisms of proneural factors and REST in regulating neuronal reprogramming of astrocytes. Cell Stem Cell. 2015;17:74–88. doi: 10.1016/j.stem.2015.05.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Matsuda T., Irie T., Katsurabayashi S., Hayashi Y., Nagai T., Hamazaki N., Adefuin A.M.D., Miura F., Ito T., Kimura H. Pioneer factor NeuroD1 rearranges transcriptional and epigenetic profiles to execute microglia-neuron conversion. Neuron. 2019;101:472–485.e7. doi: 10.1016/j.neuron.2018.12.010. [DOI] [PubMed] [Google Scholar]
  36. McDonough A., Martinez-Cerdeno V. Endogenous proliferation after spinal cord injury in animal models. Stem Cells Int. 2012;2012:387513. doi: 10.1155/2012/387513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Meletis K., Barnabe-Heider F., Carlen M., Evergren E., Tomilin N., Shupliakov O., Frisen J. Spinal cord injury reveals multilineage differentiation of ependymal cells. PLoS Biol. 2008;6:e182. doi: 10.1371/journal.pbio.0060182. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Mothe A.J., Tator C.H. Proliferation, migration, and differentiation of endogenous ependymal region stem/progenitor cells following minimal spinal cord injury in the adult rat. Neuroscience. 2005;131:177–187. doi: 10.1016/j.neuroscience.2004.10.011. [DOI] [PubMed] [Google Scholar]
  39. Nakamura M., Okano H. Cell transplantation therapies for spinal cord injury focusing on induced pluripotent stem cells. Cell Res. 2013;23:70–80. doi: 10.1038/cr.2012.171. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Nishimura Y., Natsume A., Ito M., Hara M., Motomura K., Fukuyama R., Sumiyoshi N., Aoki I., Saga T., Lee H.J. Interferon-β delivery via human neural stem cell abates glial scar formation in spinal cord injury. Cell Transplant. 2013;22:2187–2201. doi: 10.3727/096368912X657882. [DOI] [PubMed] [Google Scholar]
  41. Pataskar A., Jung J., Smialowski P., Noack F., Calegari F., Straub T., Tiwari V.K. NeuroD1 reprograms chromatin and transcription factor landscapes to induce the neuronal program. EMBO J. 2016;35:24–45. doi: 10.15252/embj.201591206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Pollak J., Wilken M.S., Ueki Y., Cox K.E., Sullivan J.M., Taylor R.J., Levine E.M., Reh T.A. ASCL1 reprograms mouse Muller glia into neurogenic retinal progenitors. Development. 2013;140:2619–2631. doi: 10.1242/dev.091355. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Puccini J.M., Ruller C.M., Robinson S.M., Knopp K.A., Buchmeier M.J., Doran K.S., Feuer R. Distinct neural stem cell tropism, early immune activation, and choroid plexus pathology following coxsackievirus infection in the neonatal central nervous system. Lab. Invest. 2014;94:161–181. doi: 10.1038/labinvest.2013.138. [DOI] [PubMed] [Google Scholar]
  44. Ramirez-Jarquin U.N., Tapia R. Excitatory and inhibitory neuronal circuits in the spinal cord and their role in the control of motor neuron function and degeneration. ACS Chem. Neurosci. 2018;9:211–216. doi: 10.1021/acschemneuro.7b00503. [DOI] [PubMed] [Google Scholar]
  45. Rea J., Menci V., Tollis P., Santini T., Armaos A., Garone M.G., Iberite F., Cipriano A., Tartaglia G.G., Rosa A. HOTAIRM1 regulates neuronal differentiation by modulating NEUROGENIN 2 and the downstream neurogenic cascade. Cell Death Dis. 2020;11:527. doi: 10.1038/s41419-020-02738-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Sayyid Z.N., Wang T., Chen L., Jones S.M., Cheng A.G. Atoh1 directs regeneration and functional recovery of the mature mouse vestibular system. Cell Rep. 2019;28:312–324.e4. doi: 10.1016/j.celrep.2019.06.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Seo S., Lim J.W., Yellajoshyula D., Chang L.W., Kroll K.L. Neurogenin and NeuroD direct transcriptional targets and their regulatory enhancers. EMBO J. 2007;26:5093–5108. doi: 10.1038/sj.emboj.7601923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Smith D.K., Yang J., Liu M.L., Zhang C.L. Small molecules modulate chromatin accessibility to promote NEUROG2-mediated fibroblast-to-neuron reprogramming. Stem Cell Reports. 2016;7:955–969. doi: 10.1016/j.stemcr.2016.09.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Soderblom C., Lee D.H., Dawood A., Carballosa M., Jimena Santamaria A., Benavides F.D., Jergova S., Grumbles R.M., Thomas C.K., Park K.K. 3D imaging of axons in transparent spinal cords from rodents and nonhuman primates. eNeuro. 2015;2 doi: 10.1523/ENEURO.0001-15.2015. ENEURO.0001-15.2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Stein C.S., Martins I., Davidson B.L. The lymphocytic choriomeningitis virus envelope glycoprotein targets lentiviral gene transfer vector to neural progenitors in the murine brain. Mol. Ther. 2005;11:382–389. doi: 10.1016/j.ymthe.2004.11.008. [DOI] [PubMed] [Google Scholar]
  51. Stenudd M., Sabelstrom H., Frisen J. Role of endogenous neural stem cells in spinal cord injury and repair. JAMA Neurol. 2015;72:235–237. doi: 10.1001/jamaneurol.2014.2927. [DOI] [PubMed] [Google Scholar]
  52. Sugimori M., Nagao M., Parras C.M., Nakatani H., Lebel M., Guillemot F., Nakafuku M. Ascl1 is required for oligodendrocyte development in the spinal cord. Development. 2008;135:1271–1281. doi: 10.1242/dev.015370. [DOI] [PubMed] [Google Scholar]
  53. Takeuchi H., Natsume A., Wakabayashi T., Aoshima C., Shimato S., Ito M., Ishii J., Maeda Y., Hara M., Kim S.U., Yoshida J. Intravenously transplanted human neural stem cells migrate to the injured spinal cord in adult mice in an SDF-1- and HGF-dependent manner. Neurosci. Lett. 2007;426:69–74. doi: 10.1016/j.neulet.2007.08.048. [DOI] [PubMed] [Google Scholar]
  54. Ueno M., Nakamura Y., Li J., Gu Z., Niehaus J., Maezawa M., Crone S.A., Goulding M., Baccei M.L., Yoshida Y. Corticospinal circuits from the sensory and motor cortices differentially regulate skilled movements through distinct spinal interneurons. Cell Rep. 2018;23:1286–1300.e7. doi: 10.1016/j.celrep.2018.03.137. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Voronova A., Fischer A., Ryan T., Al Madhoun A., Skerjanc I.S. Ascl1/Mash1 is a novel target of Gli2 during Gli2-induced neurogenesis in P19 EC cells. PLoS One. 2011;6:e19174. doi: 10.1371/journal.pone.0019174. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Wang G.X., Smith S.J., Mourrain P. Sub-synaptic, multiplexed analysis of proteins reveals Fragile X related protein 2 is mislocalized in Fmr1 KO synapses. Elife. 2016;5:e20560. doi: 10.7554/eLife.20560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Wang X., Smith G.M., Xu X.M. Preferential and bidirectional labeling of the rubrospinal tract with adenovirus-GFP for monitoring normal and injured axons. J. Neurotrauma. 2011;28:635–647. doi: 10.1089/neu.2010.1566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wapinski O.L., Vierbuchen T., Qu K., Lee Q.Y., Chanda S., Fuentes D.R., Giresi P.G., Ng Y.H., Marro S., Neff N.F. Hierarchical mechanisms for direct reprogramming of fibroblasts to neurons. Cell. 2013;155:621–635. doi: 10.1016/j.cell.2013.09.028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  59. Yousefifard M., Rahimi-Movaghar V., Nasirinezhad F., Baikpour M., Safari S., Saadat S., Moghadas Jafari A., Asady H., Razavi Tousi S.M., Hosseini M. Neural stem/progenitor cell transplantation for spinal cord injury treatment; A systematic review and meta-analysis. Neuroscience. 2016;322:377–397. doi: 10.1016/j.neuroscience.2016.02.034. [DOI] [PubMed] [Google Scholar]
  60. Zhao C., Teng E.M., Summers R.G., Jr., Ming G.L., Gage F.H. Distinct morphological stages of dentate granule neuron maturation in the adult mouse hippocampus. J. Neurosci. 2006;26:3–11. doi: 10.1523/JNEUROSCI.3648-05.2006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Zholudeva L.V., Qiang L., Marchenko V., Dougherty K.J., Sakiyama-Elbert S.E., Lane M.A. The neuroplastic and therapeutic potential of spinal interneurons in the injured spinal cord. Trends Neurosci. 2018;41:625–639. doi: 10.1016/j.tins.2018.06.004. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent methods, Figures S1–S5, and Tables S1–S4
mmc1.pdf (6.9MB, pdf)

Data Availability Statement

The raw sequencing data for the RNA-seq are deposited at the DNA Data Bank of Japan (Mishima, Shizuoka Prefecture, Japan; accession number DRA009306).

Articles from iScience are provided here courtesy of Elsevier

RESOURCES