Background: Bone marrow mononuclear cells (MNCs) are an investigational, autologous cell-based therapy for acute ischemic stroke. These cells are easily obtained by bone marrow harvest and exert neuroprotective and immunomodulatory effects in the injured rodent brain after stroke. We have investigated a new mechanism in which MNCs exert neuroprotective effects by modulating microglia. Methods: Microglia and neurons were isolated and cultured from E20 pregnant rats. Microglia were activated by oxygen-glucose deprivation (OGD) and then were cultured with neurons (n=4). At the time microglia were placed into culture with neurons, bone marrow MNCs or media vehicle was injected directly into the co-cultures or placed in transwells with the neuron-microglia cultures. The number of NeuN positive cells was quantified at 24 hrs afterwards. In another experiment, adult rats were subjected to middle cerebral artery occlusion for 90 minutes, and then treated with autologous bone marrow MNCs or saline IV at 24 hours after stroke (N=5). Three days later, brain samples were analyzed by immunohistochemistry to investigate the microglial response. Results: Adding activated microglia to neuronal cultures resulted in 36.2%±1.8% neuronal survival while adding non-activated microglia led to 85.1±15.6% neuronal survival ( p =0.001). Treatment with MNCs improved survival of neurons cultured with activated microglia to 64.0%±19.9% ( p =0.037) compared with neuronal-microglia cultures treated with media vehicle alone (36% neuronal survival). However, MNCs in transwell did not alter neuronal death as there was 31.9%±11.7% neuronal survival in co-cultures with activated microglia. All data were normalized to controls. In the in vivo study, there were significantly more OX-42 positive microglial cells in the peri-infarct area from animals treated with MNCs compared with saline treatment (43±3.9 vs.29±3.1 per high power field, p <0.05). However, the percentage of iNOS positive cells among OX-42 positive cells was less in the same regions from animals treated with MNCs compared with saline treatment (9.1%±1.1% vs. 61.5%±7.8%, p <0.05). There were also significantly more cells co-stained with Iba-1 and SDF-1α (13.8±1.5 vs. 7.4±0.8, p <0.05) in the peri-infarct region of animals treated with MNCs compared with saline controls. Conclusion: MNCs administered 24h after stroke increase the number of microglia in the peri-infarct area. MNCs decrease the number of pro-inflammatory microglia and increase the number of SDF-1a expressing microglia. Furthermore, MNCs reduce microglial-mediated neurotoxicity in vitro . Thus, the results of these studies suggest that MNCs may exert cytoprotective effects in animal models of stroke by modulating microglia. MNCs may protect neurons not only by directly reducing microglial toxicity but also by upregulating neuroprotective microglia in the post-ischemic brain.
Bing Yang, Xiaopei Xi, S. Savitz