Objective To investigate the effects of hypoxia-preconditioned human dental pulp stem cell (H-hDPSCs) transplantation on cerebral white matter injury and endogenous neural stem cells (NSCs) in neonatal rats with hypoxic-ischemic brain damage (HIBD).Methods Healthy Sprague-Dawley rats, aged 7 days, were randomly divided into sham group, HIBD group, normoxia-cultured human dental pulp stem cell (N-hDPSCs) transplantation group and H-hDPSCs transplantation group, with 11 rats in each group. The HIBD model was established using the classical Rice-Vannucci method. The proliferation of endogenous NSCs in the subventricular zone (SVZ) of the lateral ventricle of rats in each group was detected by immunofluorescence double-labeling of the proliferating cell nuclear antigen (PCNA) and Nestin 7 days after modeling. The motor and spatial memory abilities of rats in each group were detected by the pole test and the Y-maze test 3 weeks after modeling and the Morris water maze test 4 weeks after modeling. Following the behavioral tests, the expression of neurons in the cerebral cortex and the CA1 region of hippocampus was detected via immunofluorescence staining with neuron-specific nuclear protein (NeuN) as the marker, while the expressions of myelin and nerve fibers in the striatum and corpus callosum was detected via immunofluorescence double-labeling of the myelin basic protein (MBP) and neurofilament 200 (NF200).Results The number of proliferating NSCs in SVZ of the lateral ventricle in the HIBD group was lower than that in the sham group 7 days after modeling (P < 0.05), while that in the N-hDPSCs transplantation group and that in the H-hDPSCs transplantation group were higher compared with the HIBD group but lower compared with the sham group (P < 0.05). Besides, the number of proliferating NSCs in SVZ of the lateral ventricle in the N-hDPSCs transplantation group was lower than that in the H-hDPSCs transplantation group (P < 0.05). The total time spent in the pole test 3 weeks after modeling in the N-hDPSCs transplantation group and the H-hDPSCs transplantation group was longer than that in the sham group (P < 0.05) but was shorter than that in the HIBD group (P < 0.05), while that in the H-hDPSCs transplantation group was shorter compared with the N-hDPSCs transplantation group (P < 0.05). In the Y-maze test, the time spent in the novel arm in the HIBD group was shorter than that in the sham group (P < 0.05), while that in the N-hDPSCs transplantation group and the H-hDPSCs transplantation group was longer compared with the HIBD group (P < 0.05) but was shorter compared with the sham group (P < 0.05). Furthermore, the time spent in the novel arm in the Y-maze test in the N-hDPSCs transplantation group was even shorter than that in the H-hDPSCs transplantation group (P < 0.05). The escape latency in the Morris water maze test on the first day, the second day, the third day and the fourth day was compared via repeated measures ANOVA, which revealed that the escape latency was different among the time points (F = 662.825, P = 0.000) and among the groups (F = 109.286, P = 0.000). Specifically, the escape latency in the N-hDPSCs transplantation group and the H-hDPSCs transplantation group was longer than that in the sham group (P < 0.05) but was shorter than that in the HIBD group (P < 0.05), and that in the H-hDPSCs transplantation group was even shorter relative to that in the N-hDPSCs transplantation group (P < 0.05). In addition, the change trend of the escape latency was different among the groups (F = 20.543, P = 0.000). The number of platform crossings in the probe test on the fifth day in the N-hDPSCs transplantation group and that in the H-hDPSCs transplantation group were lower compared with the sham group (P < 0.05) but were higher compared with the HIBD group (P < 0.05), and that in the H-hDPSCs transplantation group was even higher than that in the N-hDPSCs transplantation group (P < 0.05). The percentages of time spent in the target quadrant in the N-hDPSCs transplantation group and the H-hDPSCs transplantation group were lower compared with the sham group (P < 0.05) but were higher compared with the HIBD group (P < 0.05), and the percentage of time spent in the target quadrant in the H-hDPSCs transplantation group was even higher than that in the N-hDPSCs transplantation group (P < 0.05). The number of NeuN-positive cells in the N-hDPSCs transplantation group and that in the H-hDPSCs transplantation group were higher than that in the HIBD group (P < 0.05) but were lower than that in the sham group (P < 0.05), and that in the N-hDPSCs transplantation group was even lower than that in the H-hDPSCs transplantation group (P < 0.05). The mean fluorescent intensity of MBP and NF200 in the striatum and corpus callosum of rats in the N-hDPSCs transplantation group and the H-hDPSCs transplantation group was higher compared with the HIBD group (P < 0.05) but was lower compared with the sham group (P < 0.05), while that in the H-hDPSCs transplantation group was even higher than that in the N-hDPSCs transplantation group (P < 0.05).Conclusions Transplantation of both H-hDPSCs and N-hDPSCs could effectively improve the long-term behavioral function, enhance the proliferation of endogenous NSCs, reduce cerebral white matter injury and thus promote neural recovery in HIBD neonatal rats. Besides, the efficacy of H-hDPSCs transplantation is superior to that of N-hDPSCs transplantation.