LINGO-1 regulates Wnt5a signaling during neural stem and progenitor cell differentiation by modulating miR-15b-3p levels

Abstract

Background: Manipulation of neural stem and progenitor cells (NSPCs) is critical for the successful treatment of spinal cord injury (SCI) by NSPC transplantation, since their differentiation into neurons and oligodendrocytes can be inhibited by factors present in inflamed myelin. In this study, we examined the effects of LINGO-1 on spinal cord-derived NSPC (sp-NSPC) differentiation, the underlying mechanisms of action, and the functional recovery of mice after transplantation of manipulated cells.

Methods: sp-NSPCs were harvested from female adult C57/BL6 mice after SCI induced with an NYU impactor. These cells were infected with lentiviral vectors containing LINGO-1 shRNA sequence or a scrambled control and transplanted into SCI mice. Tuj-1- and GFAP-positive cells were assessed by immunofluorescence staining. Wnt5a, p-JNK, JNK, and β-catenin expression was determined by Western blot and RT-qPCR. miRNAs were sequenced to detect changes in miRNA expression. Motor function was evaluated 0-35 days post-surgery by means of the Basso Mouse Scale (BMS) and by the rotarod performance test.

Results: We discovered that LINGO-1 shRNA increased neuronal differentiation of sp-NSPCs while decreasing astrocyte differentiation. These effects were accompanied by elevated Wnt5a protein expression, but unexpectedly, no changes in Wnt5a mRNA levels. miRNA-sequence analysis demonstrated that miR-15b-3p was a downstream mediator of LINGO-1 which suppressed Wnt5a expression. Transplantation of LINGO-1 shRNA-treated sp-NSPCs into SCI mice promoted neural differentiation, wound compaction, and motor function recovery.

Conclusions: LINGO-1 shRNA promotes neural differentiation of sp-NSPCs and Wnt5a expression, probably by downregulating miR-15b-3p. Transplantation of LINGO-1 shRNA-treated NSPCs promotes recovery of motor function after SCI, highlighting its potential as a target for SCI treatment.

Keywords: Differentiation; LINGO-1; Neural stem and progenitor cells; Spinal cord injury; Wnt5a; miR-15b-3p.

Fig. 1

Schematic illustration of sp-NSPC culture, differentiation, and experimental design. A sp-NSPCs, harvested from injured spinal cords, differentiated into neurons, astrocytes, and oligodendrocytes. B Transplantation of LINGO-1 shRNA-treated sp-NSPCs and timeline for motor function analysis

Fig. 2

Primary adult mouse sp-NSPCs cultured under different conditions show divergent differentiation. A Neurosphere formation in suspension cultures (bar = 50 μm) and sp-NSPCs in adherent cultures (bar = 100 μm). Cells were labeled with nestin-specific antibodies. B Immunofluorescent staining shows sp-NSPC differentiation into Tuj-1-positive neurons, GFAP-positive astrocytes, and O4-positive oligodendrocytes. Nuclei were stained with DAPI (blue) (bar = 100 μm). The analysis was calculated based on three biological replicates

Fig. 3

Expression of LINGO-1 by sp-NSPCs and differentiated cells. A LINGO-1 mRNA and B LINGO-1 protein expression levels increased significantly during differentiation, as revealed by quantitative real-time PCR and Western blot analyses. Double immunostaining shows C LINGO-1/Nestin-positive sp-NSPCs cultured in NSPC growth media and differentiated into D LINGO1/Tuj-1-positive neurons, LINGO-1/GFAP-positive astrocytes, and LINGO-1/O4-positive oligodendrocytes (arrows) when the medium was changed to the differentiation medium. Enlarged areas show cells at higher magnification. Nuclei were stained with DAPI (blue) (bar = 20 μm), ^*P < 0.05. Results are expressed as the mean ± SD. All of the analysis was calculated based on three biological replicates

Fig. 4

Effects of LINGO-1 shRNA on sp-NSPC differentiation. A sp-NSPCs were transfected with GFP-expressing LINGO-1shRNA lentiviral vectors. B LINGO-1 mRNA and C LINGO-1 protein expression levels decreased significantly. Immunofluorescence staining and quantification demonstrated that D, E Tuj1+/GFP+ neurons increased significantly, while F, G GFAP+/GFP+ astrocytes decreased in LINGO-1 shRNA-treated sp-NSPCs. Enlarged areas show that astrocytes in the LINGO-1shRNA group appeared more mature (arrowheads) than those in the control group (arrows) (bar = 100 μm), ^*P < 0.05. Results are expressed as the mean ± SD. All of the analysis was calculated based on three biological replicates

Fig. 5

miR-15b-3p is downregulated in LINGO-1 shRNA-treated sp-NSPCs, resulting in increased expression of Wnt5a. A Wnt5a protein levels increased after LINGO-1 shRNA treatment. However, B its mRNA levels did not change. C Volcano plot analysis of miRNA levels between LINGO-1 shRNA and control NSPC groups. D qPCR of miRNAs in the two sp-NSPC groups showing miR-15b-3p was downregulated after LINGO-1 RNA interference. E Levels of p-JNK, JNK, and F β-catenin in LINGO-1 shRNA-treated sp-NSPCs. ^*P < 0.05. Results are expressed as the mean ± SD. All of the analysis was calculated based on three biological replicates

Fig. 6

Wnt5a is the target of miR-15b-3p. A miR-15b-3p binding site in the 3′UTR of Wnt5a, as predicted by TargetScan. B, C Validation of miRNA binding to the 3′UTR of Wnt5a by luciferase activity assay. Reduction in luciferase activity demonstrated binding of miR-15-3p to the 3′UTR of Wnt5a. D–F Downregulation of Wnt5a protein levels by overexpression of miR-15b-3p in sp-NSPCs results in less neurons and more astrocytes, based on immunofluorescence staining and quantification. G–I Upregulation of Wnt5a protein levels after LNA inhibition of miR-15b-3p in sp-NSPCs results in more neurons but less astrocytes (bar = 100 μm), ^*P < 0.05. Results are expressed as the mean ± SD. All of the analysis was calculated based on three biological replicates

Fig. 7

Transplantation of LINGO-1 shRNA-treated sp-NSPCs after SCI improves neuronal differentiation, has a cavity-filling effect, and promotes tissue repair. Immunofluorescent staining of sagittal sections of thoracic spinal cords at 35 d.p.i. showing A, B more differentiated neurons, as revealed by DCX staining; C, D less differentiated astrocytes, as revealed by GFAP staining; E, F reduced lesion volume, as revealed by CSPG staining; and G, H less cavities, as revealed by GFAP staining. n = 10 animals per group, unpaired two-tailed Student’s t-test. ^*P < 0.05. Results are expressed as the mean ± SD

Fig. 8

LINGO-1 downregulation is required for functional recovery after SCI. Time course of locomotor recovery in the LINGO-1 shRNA and control groups evaluated by means of A the 9-point Basso Mouse Scale (BMS) and B the 11-point BMS subscore. Motor function measured with the rotarod test (C duration and D speed) at 35 d.p.i. n = 10 animals per group, repeated measures analysis of variance (rmANOVA) with Bonferroni post hoc correction for BMS and the BMS subscore, unpaired two-tailed Student’s t-test for the rotarod test. ^*P < 0.05. Results are expressed as the mean ± SD