C57BL/6 mice aged 6–8 weeks were obtained from the Experimental Animal Center of Guizhou Medical University (license no. SCXK [Qian] 2018-0001). All animal experiments were approved by the Animal Care and Use Committee of Guizhou Medical University.

All mice were anesthetized using 1.25% Avertin. The clip compression model of SCI was conducted as previously described (Cheriyan et al., 2014). Briefly, this model of SCI involves a laminectomy at the tenth thoracic vertebra of the spine. The clip is closed around the spinal cord at the tenth thoracic vertebra level, which then compresses the cord for 15 s. In the sham group, the cord was not compressed by the clip. Each group was done in triplicate at least.

Injection of vascular endothelial growth factor (VEGF) + basic fibroblast growth factor (bFGF) into the right ventricle was predominantly carried out as previously described (Shah et al., 2018a). The animals were anesthetized by 1.25% Avertin and, subsequently, mounted into a stereotaxic frame. The right lateral ventricle was injected with 3 ul PBS as a control or VEGF (1 mg/mL; Thermo Fisher Scientific, Waltham, MA, United States) and bFGF (1 mg/mL; Bio-Techne/R&D systems, United States) dissolved in PBS for growth factor stimulation. The needle was kept in place for 20 min following the injection. The spinal cord tissues were collected 1 week following VEGF/bFGF treatment. Each group was done in triplicate at least.

The animals were euthanized by Carbon Dioxide Euthanasia. The spinal cord tissue was fixed with 4% PFA at 4°C overnight, and the tissues were subsequently dehydrated with 30% sucrose gradients and embedded in OCT. The samples were penetrated with 0.25% Triton X-100 for 10 min and blocked with 1% BSA for 1 h. The sections were incubated with primary antibody at 4°C overnight. The primary antibodies included Nestin (sc23927,1:200, Santa Cruz Biotechnology, Dallas, TX, United States), Pkd2l1 (AB9084, 1;700, Merck Millipore, MA, United States), Sox2 (1:400, CST, Danvers, MA, United States), Dcx (1:200, Santa Cruz Biotechnology, Dallas, TX, United States), Gfap (1:300, CST, Danvers, MA, United States), Ki-67 (1:1000, Abcam, Cambridge, United Kingdom). After rinsing three times with PBS, the sections were incubated with secondary antibodies (goat anti-rabbit Alexa Fluor 488,1:500, CST, Danvers, MA, United States or goat anti-mouse Alexa Fluor 594, 1:500, CST, Danvers, MA, United States) in the dark for 1 h and then counterstained with DAPI.

The procedure for the isolation of CSF-cNs has been described in our previous studies (Wang et al., 2021). In brief, lentiviral particles harboring pLVX-UbC-rtTANgn 2:2 A: PKD2L1 (titers ranged 1_106 to 1_107 colony-forming unit, 10_L) were prepared and applied to CSF-cNs. Between 72 and 74 h after the application of lentiviral particles, the CSF-cNs were treated twice with puromycin (0.5–1 g/mL, Gibco, Thermo Fisher Scientific, Waltham, MA, United States). We selected CSF-cNs with green fluorescence as NSCs. Single-cell suspension was maintained in 24 well ultra-low adhesion culture plates at a volume of 400 ul fresh serum-free neural culture solution per well at a temperature of 37°C in a 5% CO₂ incubator. Every 3 days, 200 ul of the serum-free medium was added to each well until the cell reached 80% confluent, the cells were passaged. The neurospheres formed by the CSF-cNs of passage 3 were seeded in climbing slices coated with PDL and grown naturally for 12 h. Cell adhesion was observed and recorded. For cell differentiation, CSF-cNs were cultured in fresh neural culture solution with 2% fetal bovine serum (FBS) (Gibco, Thermo Fisher Scientific, Waltham, MA, United States), 1% GlutaMAX (Gibco, Thermo Fisher Scientific, Waltham, MA, United States), and 1% penicillin-streptomycin (Gibco, Thermo Fisher Scientific, Waltham, MA, United States) for 7 days.

As to the NSCs of ventricular-subventricular zone, we micro-dissected the lateral wall of lateral ventricles from C57BL/6 mice (within 24 h of birth). Then the tissue was enzymatically dissociated with papain at 37°C for 18 min. We obtained the single cell suspension through a 70 mm Nylon cell strainer. Single-cell suspension was maintained in 24 well ultra-low adhesion culture plates at a volume of 400 ul fresh serum-free NSC culture solution per well at a temperature of 37°C in a 5% CO₂ incubator. Every 3 days, half of the serum-free medium was added to each well until the cell reached 80% confluent, the cells were passaged.

NE-4C (CL-0660, Procell, Wuhan, China) were maintained in minimum essential medium (MEM) with 10% FBS (Gibco, Thermo Fisher Scientific, Waltham, MA, United States), 1% penicillin-streptomycin (Gibco, Thermo Fisher Scientific, Waltham, MA, United States) at 37°C in a 5% CO₂ incubator. For cell passage, we used Accutase (Innovative cell technologies, CA, United States) to detach them every 4 days and transferred into new poly-D-lysine (PDL)-coated dishes.

The supernatant was discarded, and the CSF-cNs were then washed in PBS three times. Following this, the cells were fixed with 4% PFA for 15 min at room temperature. The cells were washed in PBS three times again before blocking (10% goat serum and 0.3% TritonX-100 in PBS) for 1 h. All antibodies were diluted using blocking buffer. For double immunofluorescence, two primary antibodies were used sequentially, and the cells were incubated overnight with primary antibodies at 4°C; the following day, the cells were warmed for 30 min at room temperature and washed three times for 5 min in PBS. The cells were incubated for 2 h with secondary antibodies After three further washes in PBS, the cell nuclei were stained with DAPI. Finally, we observed the results using a fluorescence microscope.

Similarly, three different cell types (CSF-cNs, SVZ-NSCs, and NE-4C) were washed with cold PBS for two times and centrifuged at 1,000 rpm for 5 min to obtain cell pellet. Then these cells were lysed in RIPA lysis buffer (RIPA, Beyotime, Shanghai, China). The protein concentration was measured with BCA assays (Thermo Fisher Scientific, Waltham, MA, United States). The proteins were separated by SDS-polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride (PVDF) membrane. We used 5% milk to block the membrane at room temperature for 90 min. PVDF membrane and the primary antibody were incubated overnight at 4°C. After washing with Tris-buffer saline tween (TBST) with three times, then the PVDF membrane was incubated with horseradish peroxidase (HRP) conjugated anti-mouse IgG and anti-rabbit IgG (1: 5000, Proteintech, Wuhan, China) for 1 h. Subsequently, we visualized these immunolabeled bands with an enhanced chemiluminescence reagent (Thermo Fisher Scientific, Waltham, MA, United States).

The primary antibodies included Nestin (66259-1-Ig,1:500, Proteintech, Wuhan, China), Pkd2l1 (13117-2-AP, 1: 500, Proteintech, Wuhan, China), Sox2 (66411-1-Ig,1: 1000, Proteintech, Wuhan, China), β-Actin (20536-1-AP, 1-10000, Proteintech, Wuhan, China).

In this study, three different cells (CSF-cNs, SVZ-NSCs, and NE-4C) and spinal cord of mice were used for qPCR. CSF-cNs, NE-4C, and SVZ-NSCs were washed with PBS and centrifuged (1,000 rpm, 5 min) to obtain cell precipitation. Briefly, the total RNA was extracted using MolPure^® Cell/Tissue Total RNA Kit (Yeasen Biotechnology, Shanghai, China). Then, we polyadenylated and reverse-transcribed the RNA into complementary DNA using a poly(T) adapter following manufacturer instructions. Real-time PCR was done using a thermal cycler under the following parameters: a 5-min initial denaturation step at 95°C; 44 cycles at 95°C for 15 s; 55°C for 30 s; 72°C for 20 s. We subjected each sample to the entire experimental procedure in triplicate. Table 1 lists the primers specific for mRNA.

To label proliferating cells following SCI or growth factor infusions, 5-ethnyl-2′-deoxyuridine (EdU) (Riobio, Guangzhou, China) was injected intraperitoneally (50 mg/kg). The tissue processing is similar to Immunofluorescence section. After Immunofluorescence staining, these sections were stained for EdU using cell-light EdU Apollo567 kit (Riobio, Guangzhou, China) according to the manufacturer’s instructions. We observed the results using a Leica confocal microscope.

The spinal cord sections were washed with distilled water and incubated in Nissl staining solution (Solarbio, Beijing, China) for 1 h at 37°C. Then sections were incubated Nissl Differentiation for a few seconds and dehydrated with 100% ethanol twice for 2 min. Unbiased stereology was used to estimate the number of Nissl body numbers in the spinal cord using the optical dissector method. Furthermore, the region of gray matter surrounding the central canal (lamina X) was defined as the region of interest (ROI). The counting frame was 50 × 50 um, and the sampling grid size was 100 × 100 um.

All experimental results were conducted in triplicate. ImageJ pro was used to count the cell numbers. For comparisons of groups across multiple groups, a one-way analysis of variance (ANOVA) was used. Mann–Whitney test was used for comparisons of groups across two groups. Data were represented and analyzed in Prism 8.4. A level of p < 0.05 was considered statistically significant.

There is a cluster of CSF-cNs located in the subependymal layer of the central canal. These cells were found to express the specific marker Pkd2l1 (Figure 1A), even though CSF-cNs had been discovered approximately a century ago. Certain physiological functions of CSF-cNs are gradually being revealed owning to the identification of Pkd2l1 in 2006. In our previous study, CSF-cNs were confirmed to have the properties of NSCs in vitro. In this study, we performed these studies in adult mouse spinal cord to explore whether this unique characteristic of CSF-cNs also existed in vivo. We observed that these Pkd2l1⁺ cells expressed NSC markers such as Sox2 and Nestin (Figures 1B,C).

Interestingly, these co-labeled cells were located away from the central canal of the spinal cord. In addition, we also found that some Pkd2l1⁺ cells were co-stained for Dcx, which is a neuronal migration marker (Figure 1D) and is considered by some researchers a well-established marker for immature neurons.

Generally, NSCs can form clonal colonies under growth factor conditions. To test the colony-forming capacities of CSF-cNs in vitro, we collected CSF-cNs from the microdissected cervical spinal cord and plated them under NSC culturing conditions. We further purified them using the screening of lentiviral particles harboring pLVX-UbC-rtTANgn2:2A: Pkd2l1 (Figure 2A). In this case, these target CSF-cNs showed green fluorescence under a fluorescent microscope (Figure 2B). Furthermore, within 1 week, we observed robust neurosphere formation. Moreover, these neurospheres can be expanded and passaged at least four generations (Figure 2C). In addition, to test whether these CSF-cNs have the unique characteristic compared to these widely accepted NSCs [NE-4C, ventricular subventricular NSCs (V-SVZ-NSCs)]. We validated their difference though western blot and quantitative real-time PCR (qRT-PCR). These three types of cells had the similar expression pattern of NSC genes (Sox2, Nestin). CSF-cNs had the high expression of Pkd2l1, while others did not (Figures 2D,E). To examine the multipotent capability of CSF-cNs, we induced cell differentiation by retracting EGF and FGF-2 from the medium. After 7 days, these dissociated CSF-cNs could differentiate into NeuN-expressing neurons (Figure 3A), S100β-expressing astrocytes (Figure 3B), and Olig2-expressing oligodendrocytes (Figure 3C).

To explore whether these Pkd2l1⁺ cells could be activated under growth factor conditions in vivo, either VEGF + bFGF or PBS were syringed into the right lateral ventricle in equal volumes (Figure 4A). After 1 week, we observed there was an obvious increase in Pkd2l1 expression following infusion of VEGF and bFGF compared to PBS (p < 0.05, n = 3, Figure 4B). We also checked the cellular reactivity of Pkd2l1⁺ cells. Unsurprisingly, we observed a relatively significant increase in Pkd2l1⁺ Nestin⁺ co-expressed cells following the VEGF + bFGF infusion compared with the PBS group (p < 0.0001, n = 6, Figure 4C). qRT-PCR analysis displayed that a significant increase in Nestin expression following infusion of VEGF and bFGF (p < 0.05, n = 3, Figure 4D). Additionally, the Pkd2l1⁺ Gfap⁺ co-labeled cells also noticeably increased after VEGF + bFGF administration (p < 0.0001, PBS vs. VEGF + bFGF, n = 6, Figure 4E). However, as reported in previous literature, CSF-cNs do not express Gfap (Petracca et al., 2016). Correspondingly, qRT-PCR analysis represented that a significant rise in Gfap expression following infusion of VEGF and bFGF (p < 0.05, n = 3, Figure 4F). Notably, VEGF + bFGF treatment elicited not only relatively higher percentages of proliferative Pkd2l1⁺ Ki-67⁺ co-expressed cells but also of Pkd2l1 Dcx⁺ immature neurons (p < 0.001, PBS vs. VEGF + bFGF, n = 6, Figures 4G,I). Moreover, Pkd2l1⁺ EdU⁺ co-expressed cells also confirmed proliferation, and Nissl staining showed a visible increase in cell number (Figures 4K,L). qRT-PCR analysis showed a significant difference in Ki-67 and Dcx expression following infusion of VEGF and bFGF. At the end, we analyzed these data and qualified these changes (p < 0.05, n = 3, Figures 4H,J).

There is a possibility that SCI can induce cell proliferation and cause NSCs at the lesion site to enter neurogenic reprogramming again. Some dormant NSCs will be activated after SCI. First, we performed the SCI mouse model at the T10 level (Figure 5A). Next, we validated the Nestin expression at different time point (1, 3, 7, and 14 days) following SCI using qRT-PCR. We observed that the Nestin expression reached its peak at 3 days after SCI (p < 0.05, n = 3, Figure 5B). Consequently, we also verified the expression of Pkd2l1 at 3 days of post-injury, qRT-PCR result displayed there was an obvious increase in Pkd2l1 expression (p < 0.05, n = 3, Figure 5C). To test whether these Pkd2l1⁺ cells could be activated by an SCI, we examined the molecular response of Pkd2l1⁺ cells to the NSC markers Nestin and Gfap at 3 days post-injury (DPI). Predictably, we observed that the percentage of Pkd2l1⁺ Nestin⁺ co-expressed cells significantly increased following SCI treatment (p < 0.05, n = 6, Figure 5D). qRT-PCR analysis displayed a significant rise in Nestin expression following SCI (p < 0.0001, n = 3, Figure 5E).

Similarly, the density of Pkd2l1⁺ Gfap⁺ co-expressed cells in the SCI group clearly exceeded the density in the sham group, for which few co-labeled cells could be detected (p < 0.01, n = 6, Figure 5F). qRT-PCR result showed a substantial increase in Gfap expression following SCI (p < 0.0001, n = 3, Figure 5G). In the SCI group, we also found that the percentage of Pkd2l1⁺ Ki-67⁺ and Pkd2l1⁺ Dcx⁺ co-stained cells were visibly higher than in the sham group (p < 0.05, n = 6, Figures 5H; p < 0.01, n = 6, Figure 5J). EdU incorporation assay indicated that Pkd2l1 positive cells proliferated after SCI, and Nissl staining also demonstrated an increase in cell number (Figures 5L,M). qRT-PCR analysis exhibited a significant augmentation in Ki-67 and Dcx expression following SCI (p < 0.01, n = 3, Figures 5I,K). This data indicates that SCI enhanced the proliferation of CSF-cNs and activated the potential of NSCs in the adult spinal cord.