[Effects of bone marrow mesenchymal stem cells on learning and memory functional recovery in neonatal rats with...

Neonatal hypoxic-ischemic brain damage (HIBD) causes acute death and chronic nervous system sequelae in newborn infants and children. Whereas there have been no specific treatment towards it up to now. Studies have shown that bone marrow mesenchymal stem cells (MSCs) have the therapeutic potential in many nervous system diseases and the authors previously found that retinoid acid (RA), which plays an important role in brain development, could enhance the neural differentiation of rat MSCs (rMSCs) in vitro. This study aimed to examine effects of rMSCs and RA-preinduced rMSC on learning and memory functional recovery after HIBD in neonatal rats in order to explore a new treatment strategy for clinical application, and explore the mechanism of action of rMSCs.

Rat MSCs were isolated and purified from the whole bone marrow of juvenile Wistar rats by removing the non-adherent cells in primary and passage cultures. Neonatal hypoxic-ischemic brain damage rat models were built according to the methods described by Rice: the right carotid artery of 7-day-postnatal Wistar rats was ligated under anesthesia, and then the rats were exposed to 8% - 9% O2 in a container. At 5 days after hypoxia-ischemia, the HIBD neonatal rats were randomly divided into 3 groups and respectively transplanted with saline, BrdU marked rMSCs (1 - 2 x 10(5)) or RA-preinduced rMSCs (1 - 2 x 10(5)) into their lateral cerebral ventricle. Immunohistochemistry for nestin, neuron-specific enolase (NSE), neurofilament protein-heavy chain (NF-H) and glial fibrillary acidic protein (GFAP) were used to identify cells derived from rMSCs at 14 days and 42 days after transplantation. Shuttle box test was performed to evaluate the condition of learning and memory functional recovery when animals were 7 weeks old. Neurotrophin and receptors cDNA microarray were also employed at 14 days after transplantation to investigate the underlying action mechanisms of rMSCs treatment. Real-time PCR was used to confirm some of the remarkably changed genes.

(1) The neonatal rat model of HIBD was successfully established. (2) Immunohistochemistry showed rMSCs-derived cells survived, migrated into the hypoxic-ischemic brain tissue and a few of them expressed protein characteristic of neurons and astrocytes (NF-H and GFAP) in RA-preinduced group 14 days and 42 days after transplantation, while no positive expression of nestin and NSE were detected. (3) The shuttle box test showed that the average learning times in rats transplanted with saline, rMSC and RA-preinduced rMSCs were (94.10 +/- 38.18), (74.60 +/- 29.21) and (47.90 +/- 21.13), respectively. The difference between the former two was not significant (P > 0.05), while the latter one exhibited significant improvement (P < 0.05). (4) The cDNA microarray analysis showed that compared with normal control group, IL-6, Fas and BDNF genes of the saline control group significantly up-regulated (the ratios of the three genes were 11.4, 2.4 and 6.6 respectively). Compared with saline group, the three genes in rMSC group were down-regulated (the ratios were all 0.1), while the levels of IL-6 and Fas genes (the ratios were 0.3 and 0.4 respectively) in RA-preinduced rMSCs group were higher than rMSCs group after down-regulating, but the level of BDNF remained at the saline group level. Real-time PCR analysis suggested that the results of IL-6 and Fas genes were at equal level with microarray results on the whole, while the level of BDNF gene in RA-preinduced rMSC group was significantly down-regulated (with ratio of 0.34), but higher than rMSCs group (the ratio was 0.25) as well.

Transplantation of rMSC and RA-preinduced rMSCs into lateral cerebral ventricle can improve learning and memory functional recovery after HIBD in neonatal rats, especially RA-preinduced rMSCs. Regulating the levels of IL-6, Fas and BDNF in the brain to maintain at reasonable levels may be the mechanism.