Objective To observe the repair effect of catalpol on human neuroblastoma cells (SH-SY5Y) after oxygen-glucose deprivation/reoxygenation injury, and to investigate its mechanism. Methods SH-SY5Y cells were divided into the control group, all-trans retinoic acid (ATRA) group, model group, and catapol group. Control group： SH-SY5Y cells did not receive any treatment; ATRA group： the naive SH-SY5Y cells were induced by ATRA for 72 h to differentiate into mature neurons. Model group： after ATRA induction, the complete medium was replaced by sugar-free EBSS solution, and these cells were then placed in a three-gas incubator( oxygen 1%, carbon dioxide 5%, nitrogen 94%) for 4 h; after the deprivation of oxygen and sugar was over, SH-SY5Y cells were cultured in original condition with oxygen and sugar for 12 h. Catalpol group： after receiving the same treatment as the model group, cells were placed in the complete medium containing different concentrations of catalpol (50, 100, and 200 μmol/L) for 72 h, respectively. The cell viability in each group was detected by CCK-8. β-III-tubulin immunofluorescence staining was used to label the axons, and we measured their length. Western blotting was used to detect the protein levels of growth associated protein 43(GAP 43), osteopontin (OPN), phosphorylation-insulin-like growth factor 1 receptor (p-IGFR), phosphatase and tensin homolog deleted on chromosome ten (PTEN), mammalian target of rapamycin (mTOR), and phosphorylated ribosomal protein S6(p-S6). Results The survival rate was higher in the ATRA group than in the control group (P<0. 0001). The survival rate of the model group was lower than that of the ATRA group(P<0. 0001). The survival rates in different concentrations of catalpol groups were higher than that of the model group (all P<0. 05), and there was no statistical difference between the different concentration groups (P>0. 05). The cell bodies of the control group were round and the axons were short. The cell axons in the ATRA group were long and connected to form neural networks. Compared with the ATRA group, the number of cells in model group decreased, the cell body was flat, and the axon was retracted. Compared with model group, the number of cells in the catalpol group increased and the length of axons was longer. Axon length was longer in the induction group than in the blank control group. Compared with the ATRA group, the axon length in the model group was shorter (P< 0. 001), and the length of axon in different concentrations of catalpol groups was extended (P<0. 01). The scratch area of different concentrations of catalpol groups was smaller than that of the model group after 48 h(P<0. 05), and there was no statistical difference between the different concentrations of catalpol groups (P>0. 05). Compared with ATRA group, the relative expression level of PTEN increased, while the OPN decreased in the model group (both P<0. 05). In catalpol groups with different concentrations, the relative expression levels of GAP43, OPN, p-IGFR, mTOR, and p-S6 protein increased, while the relative expression level of PTEN protein decreased in comparison to those of the model group (all P< 0. 05). Conclusions SH-SY5Y cells after oxygen-glucose deprivation/reoxygenation injury can be repaired by catalpol. The underlying mechanism may be that catalpol induces the phosphorylation of IGFR receptor on the cell surface to p-IGFR by up-regulating the expression of OPN, and then activate the mTOR pathway; at the same time, it decreases the expression of PTEN to reduce its inhibition on the mTOR pathway. The intrinsic growth ability of cells is fully activated to promote neuronal cell migration, and axon outgrowth. Different concentrations of catalpol (50, 100, and 200 μmol/L) have similar effects on the repair of SH-SY5Y cells injuried by oxygen-glucose deprivation/reoxygenation. [ABSTRACT FROM AUTHOR]