Preinjury Oligodendrocyte Precursor Cells Mediate Remyelination Following miR-219 and miR-338 Delivery After Spinal Cord Injury
Description
Oligodendrocyte myelination is essential for central nervous system function, enabling rapid electrical conduction and sustaining neuronal metabolism. Spinal cord injury (SCI) disrupts this process through oligodendrocyte loss and axonal demyelination, leading to profound neurological deficits. Although enhanced oligodendrocyte-mediated remyelination through the delivery of miR-219 and miR-338 contributes to improved functional recovery after SCI, the cellular origins through which miR-219 and miR-338 regulate remyelination remain unclear. To determine whether miR-219 and miR-338 promote remyelination through oligodendrocyte precursor cells that existed prior to SCI (i.e. preinjury OPCs), two transgenic lineage-tracing mouse models (PDGFRα-CreER:RCE and PDGFRα-CreER:mTmG) were used. Using a complete transection SCI model, sustained delivery of miR-219/miR-338 via a fiber–hydrogel scaffold significantly increased the density of GFP-positive preinjury OPCs in PDGFRα-CreER:RCE mice at the lesion site by Week 2. By Week 4, a greater numeber of these lineage-traced preinjury OPCs differentiated into CC1-positive mature oligodendrocytes, and came to represent a disproportionately larger share of the mature oligodendrocyte pool, compared to animals treated with scrambled microRNA controls (Neg miR). Immunostaining for myelin basic protein (MBP), neurofilament (NF200), and membrane-tethered GFP (mGFP) in PDGFRα-CreER:mTmG mice confirmed that cells derived from preinjury OPCs significantly contributed to remyelination, although additional remyelination also originated from other cellular sources. Further analysis showed that miR-219/miR-338 suppressed transdifferentiation of preinjury OPCs into Schwann cell–like phenotypes without this suppressed lineage itself contributing to Schwann cell–mediated myelination. These findings demonstrate that at 4 weeks post SCI treatment, miR-219/miR-338-mediated remyelination is driven primarily by activation and differentiation of OPCs that existed before the injury.
Files
Steps to reproduce
1. Animal models and lineage tracing. All experiments used 8–10-week-old female C57BL/6J mice (In Vivos Pte Ltd, Singapore) under NTU IACUC protocols A21005/A20049. Two tamoxifen-inducible lineage-tracing lines were generated: PDGFRα-CreERT2 crossed with Rosa26-EGFP (RCE; IMSR_JAX:006148) or Rosa26-mGFP (mTmG; IMSR_JAX:007576). Recombination was induced by five consecutive daily intraperitoneal tamoxifen injections (3 mg/day for RCE; 2 mg/day for mTmG) followed by a 2-week washout, which we validated as the optimal labeling protocol (Fig. S2/S4). No-TA controls confirmed absence of leaky recombination (Fig. S3/S5). 2. Scaffold and miRNA formulation. Aligned PCL fiber sheets were produced by two-pole air-gap electrospinning, coated with poly-D-lysine on alternating layers, rolled into a 1.6 mm cylindrical bundle, and infiltrated with collagen (3.5 mg/mL) containing GDNF plus either miR-219-5p/miR-338-3p/miR-338-5p mimics or scrambled negative miRNA (Thermo Fisher AM17100 series). Constructs were lyophilized and cut to 1.2 mm. 3. Surgery. Animals underwent complete T9–T10 spinal cord transection; a 1.2 mm scaffold was implanted to bridge the stumps and covered with an electrospun PCL membrane. Analgesia (meloxicam, buprenorphine) and antibiotic prophylaxis (Baytril) were given; bladders were manually expressed twice daily for 14 days. 4. Tissue processing and immunostaining. At 2 or 4 weeks, mice were perfused with 4% PFA; the cord was cryosectioned longitudinally at 16 μm. Sections were immunostained with validated primary antibodies (PDGFRα, Olig2, APC/CC1, GFP/mGFP, MBP, NF200, P0, Krox20, Caspr, GFAP; full list in Table 4) plus Alexa Fluor secondaries and DAPI. 5. Imaging and quantification. Regional images (rostral/caudal perilesional interface zones and scaffold) were acquired as 10× tile-scan collages on a Leica DMi8 and 40× confocal z-stacks on a Zeiss LSM 800. Cell densities (preinjury- vs non-preinjury-derived OPCs, mature oligodendrocytes, Krox20⁺ Schwann cells) and myelination indices (MBP⁺/NF200⁺ colocalization; mGFP⁺ preinjury-derived myelin; P0⁺ Schwann-cell myelin) were measured in ImageJ with custom macros, on three sections per animal spaced 160 μm apart. Raw cell counts and percentage data were recorded in GraphPad Prism. 6. Statistics and software. Normality/homogeneity were assessed by Shapiro–Wilk and Levene's tests; groups were compared by one-/two-way ANOVA with Tukey's post hoc or Student's t-test (Kruskal–Wallis/Mann–Whitney for non-parametric data). Values are mean ± SD (n = 4–7/group). Software: ImageJ (macros), GraphPad Prism 10, Microsoft Excel. 7. Reproducibility. Analysis macros, Prism analysis files, and representative raw images are available from the corresponding author on reasonable request and will be deposited in a public repository (e.g., Mendeley Data) upon acceptance; exported 400-dpi TIFFs of all figures accompany this submission.
Institutions
- Nanyang Technological UniversitySingapore