miR-21 has a positive effect on SCI in vivo. To determine whether miR-21 could regulate astrocytes in vivo, expression of miR-21 was interrupted in a SCI mouse model. Mice were divided into four groups: sham, agomir-21, antagomir-21 and NC. (A) BMS score for each group (n = 8). Unpaired t-test was used for comparison of agomir-21 and antagomir-21 groups with the antagomir NC group. Univariate analysis of variance also used to analyze this result showed that miR-21 had significant correlation with BMS scores (P < 0.001). (B) The expression of miR-21 was detected by qRT-PCR. BDNF (D, F), and NGF (C, E) were detected by qRT-PCR and Elisa (n = 3). Immunohistochemistry was performed to examine GFAP (scale bar: low magnification, 100μm; high magnification, 20μm) (G) and CSPGs (scale bar: low magnification, 50μm; high magnification, 20μm) (H) and the results were analyzed by ImageJ and SPSS (I and J). Data are expressed as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001 compared with the NC group. BDNF, brain-derived neurotrophic factor; BMS, Basso Motor Score, CSPGs, chondroitin sulfate proteoglycans; GFAP, glial fibrillary acidic protein; miR, microRNA; NGF, nerve growth factor; NC, negative control; qRT-PCR, quantitative real-time polymerase chain reaction; Elisa, enzyme-linked immunosorbent assay; SCI, spinal cord injury.