DLK1-expressing neural progenitor cells promote tissue repair and functional recovery after cervical spinal cord injury

Abstract

Spinal cord injury (SCI) elicits a hostile microenvironment characterized by inflammation, gliosis, and disrupted signaling pathways that collectively impede neural repair. Neural progenitor cells (NPCs) represent a promising regenerative approach, yet their survival and differentiation are often compromised in this setting. Here, we investigated whether engineering NPCs to overexpress the Notch pathway modulator Delta-like non-canonical Notch ligand 1 (DLK1) could overcome these limitations and improve functional outcomes after cervical SCI in rats. NPCs were engineered to express DLK1 under a Pax6 promoter-driven expression system, ensuring elevated DLK1 levels during the progenitor state. Following transplantation of DLK1-overexpressing NPCs or control NPCs, we assessed graft survival, lineage differentiation, behavioral performance, and electrophysiological integration over 12 weeks. DLK1-expressing NPCs exhibited significantly greater retention in the injured spinal cord and showed enhanced neuronal differentiation alongside reduced astrocytic commitment compared to controls. Behavioral tests-including forelimb grip strength and CatWalk gait assessments-demonstrated that DLK1-modified NPCs conferred robust improvements in forelimb motor coordination and overall locomotion. Concordantly, electrophysiological recordings revealed increased motor-evoked potential amplitudes and area-under-the-curve values in animals receiving DLK1-transduced NPC grafts, indicative of strengthened synaptic integration within the host motor circuitry.

Keywords: DLK1; axons; delta-like non-canonical Notch ligand 1; functional recovery; neural progenitor cells; spinal cord injury; tissue repair.

Graphical Abstract

Figure 1.

DLK1 expression enhances the retention of transplanted NPCs in the injured spinal cord. (A) Representative longitudinal sections of the injured spinal cord at 12 weeks post-transplantation, illustrating the distribution of transplanted cells labeled with the human cell–specific marker STEM121. (B) Quantification of surviving STEM121⁺ cells reveals a significantly higher number of retained cells in the DLK1-NPC group compared to control NPCs. Data are presented as mean ± SEM (n = 10). Statistical analysis was performed using an unpaired t-test (*P < .05). Scale bar: 2 mm.

Figure 2.

DLK1 expression modulates the in vivo differentiation of transplanted NPCs. (A) Representative immunohistochemical images of injured spinal cord sections transplanted with control NPCs or DLK1-NPC, stained for hNSE (human neuron-specific enolase, a neuronal marker), STEM123 (human GFAP, an astrocyte marker), and hGSTπ (GSTπ, an oligodendrocyte marker). Positive cells appear violet/purple as a result of the Vector VIP HRP substrate. Scale bar: 40 μm. (B) Violin plots showing the average number of hNSE⁺, STEM123⁺, and hGSTπ⁺ cells (cells/mm²) in the injured spinal cord from both NPC and DLK1-NPC groups. The central white line indicates the median, and the upper and lower white lines represent the first and third quartiles, respectively. Statistical analysis was performed using unpaired two-tailed t-tests (*P < .05).

Figure 3.

Longitudinal analysis of forelimb grip strength recovery and inclined plane assessment. Animals received NPC transplants or vehicle. Sham group (horizontal black dashed line) underwent laminectomy only (no SCI). Transplantation occurred at week 0 (black arrow). Measurements were taken from 4 weeks before transplantation (W–4) to 12 weeks afterward (W12). (A) Forelimb grip strength, measured biweekly using a grip-strength meter. Repeated-measures 2-way ANOVA showed a significant main effect of time (F(5.529, 138.2) = 26.50, P < .001) and treatment (F(25, 200) = 6.539, P < .001), but no significant time × group interaction (F(16, 200) = 1.351, P = .170). Post-hoc Dunnett’s tests revealed that NPC-DLK1-treated rats had significantly greater grip strength than vehicle-treated rats at W10 (adjusted P = .043) and W12 (adjusted P = .048). Data represent mean ± SEM. (B) Inclined plane performance, measured biweekly using a standardized inclined plane test. Both the NPC-DLK1 and NPC groups demonstrated a trend toward improved performance relative to vehicle, but these differences were not statistically significant. Error bars represent mean ± SEM. Statistical analyses were performed using repeated-measures 2-way ANOVA (Geisser-Greenhouse correction).

Figure 4.

Gait analysis in rats with cervical SCI using the CatWalk system. (A) PCA of gait parameters demonstrates distinct clustering of experimental groups based on locomotor function at 12 weeks post-transplantation. Each data point represents an individual rat. 95% confidence ellipses indicate within-group variance, with centroids (◉) representing the geometric center of each ellipse. DLK1-NPC-treated animals cluster closer to the sham group, suggesting improved locomotor function. PERMANOVA analysis confirmed significant between-group differences (P < .001). (B) PCA loading plot depicting the contributions of individual gait parameters to the first 2 principal components (PC1: 45.5%, PC2: 23.9%). Arrows represent individual gait parameters, with warmer colors (orange/red) indicating stronger contributions to group separation. Stride length, swing speed, regularity index, body speed, and print area were the dominant contributors to PC1 and were prioritized for further analysis. Quantitative comparisons of key gait parameters: (C) print area (cm²), (D) body speed (cm/s), (E) regularity index (%), (F) stride length (cm), and (G) swing speed (cm/s). Bars represent mean ± SEM. Statistical comparisons were performed using one-way ANOVA followed by Tukey’s post-hoc test. (*P < .05, P < .01).

Figure 5.

DLK1 expression in transplanted NPCs augments motor-evoked potentials (MEPs). (A) Representative MEP waveforms recorded from the Triceps brachii muscle in each experimental group: Sham, Vehicle, NPC, and DLK1-NPC. MEPs were elicited by intracranial electrical stimulation of the motor cortex using two 31G stainless-steel needle electrodes inserted through the scalp. (B) Quantification of the peak-to-peak amplitude (P_N Amplitude) demonstrates a significant increase in the NPC-DLK group relative to the Vehicle group, indicating enhanced neural connectivity. Data are presented as mean ± SEM (n = 6 per group). One-way ANOVA followed by Dunnett’s post-hoc test was used to assess statistical significance (*P < .05; F(2, 15) = 6.514). (C) Analysis of the MEP area under the curve (AUC) reveals a similar trend, with the DLK1-NPC group exhibiting a significantly higher AUC than the vehicle group. These findings further support the notion that DLK1 expression promotes improved functional integration of transplanted NPCs. Data are presented as mean ± SEM (n = 6 per group). One-way ANOVA followed by Dunnett’s post-hoc test was used to assess statistical significance (*P < .05; F(2, 15) = 3.985).