Histological analyses of the treatment effects on microvasculature and remyelination in the spinal cord. (A) Visualization of blood vessels using an anti-RECA-1 antibody. Images in the upper row are low-magnification views of the gray matter areas of sagittal sections immunostained for RECA-1 at 1 mm caudal to the transected site. Scale bars = 50 μm. Images in the lower row are high-magnification views that correspond to the boxed areas in the upper row images. Scale bars = 50 μm. (B) Representative image of a blood vessel with a lumen with a diameter larger than 20 μm (arrow), which indicated newly formed blood vessels following injury. Scale bars = 50 μm. Arrows in (A) also represent blood vessels with lumen diameters larger than 20 μm. The left side is rostral (A,B). (C) Quantitative analysis of RECA-1-positive areas in each group. (D) Quantitative analysis of the total areas of RECA-1-positive blood vessels with lumen diameters larger than 20 μm. *P < 0.05, **P < 0.01. Statistical analyses were based on one-way ANOVA and Bonferroni post hoc analyses. (E-M) Analyses of remyelination performed using immunohistochemistry against MBP or electron microscopy 12 weeks post-injury. (E,F,H,I,K,L) Reconstructed confocal images showing double staining (sagittal sections) for MBP (green) and GAP43 (red) in the control group (E,F), SM-345431 treatment group (H,I) and combined group (K, L). F, I and L show magnified images of the boxed areas in E, H and K, respectively. Scale bars = 100 μm. The arrow in F shows a non-myelinated (MBP-negative) GAP-43-positive axon, and the arrows in I and L show myelinated (MBP-positive) GAP-43-positive axons. The left side is rostral. (G,J) Electron microscopic images of transverse sections from the control group (G) and SM-345431 treatment group (J) at the lesion site. Scale bars = 2 μm. (N) Statistical analysis of the number of myelinated (MBP-positive) GAP-43-positive axons in each group, which were analyzed by immunohistochemistry. *P < 0.05, **P < 0.01. Statistical analyses were performed using one-way ANOVA and Bonferroni post hoc analyses. All the data are represented as the mean ± S.E.M.