OBJECTIVENeonatal hypoxic-ischemic encephalopathy (HIE) harms the lives and health of newborn infants and children severely. The prognosis is not satisfied, especially of the severe HIE. Mesenchymal stem cells (MSCs) can secrete a series of growth factors and neurotrophic factors. As well they have the potential ability to differentiate to the neural cells in vitro and in vivo. Therefore MSCs transplantation has been employed as a source of progenitor cells for cell therapy in patients with HIE in order to promote recovery of brain function and reduce the sequelae. Studies have shown that MSCs could enter the cerebral parenchyma and differentiate to neural cells through systemic infusion, but most of the researches applied adult stroke animal models. This study used neonatal HIE models to test the hypothesis that MSCs could enter the brain of newborn Wistar rats through the blood-brain barrier (BBB) by intraperitoneal infusion followed by observing the characteristics of the distribution and differentiation of MSCs in brain tissues, and exploring the effects of hypoxic-ischemic brain damage to the penetration and differentiation of MSCs.

METHODSIsolation and purification of MSCs were established from the whole bone marrow of juvenile Wistar rats by removing the nonadherent cells in primary and passage cultures. For cellular identification, MSCs of three to five passages were continuously pre-labeled with 5-bromo-2-deoxyuridine (BrdU) for 72 hours before transplantation. Animal models of HIE were built in 7-day-postnatal Wistar rats according to the method described by Rice. Two hours after hypoxia-ischemia, rats in HIE group (n = 8) were intraperitoneally infused with MSCs (4 x 10(6), 0.5 ml). In control group (n = 8), 7-day-postnatal normal Wistar rats were intraperitoneally infused with the same amount of MSCs. All rats were sacrificed and their cerebra were sectioned by cryomicrotome 14 days after transplantation. Immunohistochemical staining with chromogen diaminobenzidine (DAB) was used to detect and measure the cells derived from MSCs, and study the characteristics of distribution. To determine the differentiation of the BrdU positive cells entering the brains, immunofluorescence double labeling for BrdU and neural cells specific antigens was performed.

RESULTSMSCs were distributed throughout the cerebra in both groups at the 14th day after transplantation. The number of MSCs detected was 2415 +/- 226 in the control group, and 3626 +/- 461 in HIE group, respectively (t = 6.68, P < 0.05). More BrdU reactive cells were observed in the right ischemic hemisphere (1904 +/- 267) than in the contralateral hemisphere (1723 +/- 204), (t = 4.47, P < 0.05). No significant difference was found while comparing both cerebral hemispheres of the control group (t = 0.31, P > 0.05). In the HIE group, MSCs distributed more extensively, and some focal aggregations of MSCs were noticed. A few MSCs expressed Nestin-protein marker of neural progenitor cells, and almost none of the MSCs which expressed proteins characteristic of neuron (e.g. NSE) and astrocyte (e.g. GFAP) was detected at the 14th day after transplantation.

CONCLUSION1. MSCs could enter the cerebral parenchyma through BBB and migrate throughout the brain by intraperitoneal infusion. 2. More MSCs injected intraperitoneally were localized and directed to the sites of hypoxic-ischemic brain damage. 3. Transplanted MSCs could not differentiate to neuron and astrocyte without other interventions during 14 days after transplantation.