We repopulated 'primitive' cells by replating late-passage MSCs at low density (17 cells/cm(2) ) regardless of donor age. Repopulated MSCs from low-density culture were smaller cells with spindle shaped morphology compared to MSCs from high-density culture. The latter had enhanced colony-forming ability, proliferation rate, and adipogenic and chondrogenic potential. Strong expression of osteogenic-related genes (Cbfa1, Dlx5, alkaline phosphatase and type Ι collagen) in late-passage MSCs was reduced by replating at low density, whereas expression of three pluripotency markers (Sox2, Nanog and Oct-4), Osterix and Msx2 reverted to levels of early-passage MSCs. Knockdown of Sox2 and Msx2 but not Nanog, using RNA interference, showed significant decrease in colony-forming ability. Specifically, knockdown of Sox2 significantly inhibited multipotentiality and cell proliferation. Our data suggest that plating density should be considered to be a critical factor for enrichment of 'primitive' cells from heterogeneous BM and that replicative senescence and multipotentiality of MSCs during in vitro expansion may be predominantly regulated through Sox2.

Transient/chronic microenvironmental hypoxia that exists within a majority of solid tumors has been suggested to have a profound influence on tumor growth and therapeutic outcome. Since the functions of novel antioxidant proteins, peroxiredoxin I (Prx I) and II, have been implicated in regulating cell proliferation, differentiation, and apoptosis, it was of our special interest to probe a possible role of Prx I and II in the context of hypoxic tumor microenvironment. Since both Prx I and II use thioredoxin (Trx) as an electron donor and Trx is a substrate for thioredoxin reductase (TrxR), we investigated the regulation of Trx and TrxR as well as Prx expression following hypoxia. Here we show a dynamic change of glutathione homeostasis in lung cancer A549 cells and an up-regulation of Prx I and Trx following hypoxia. Western blot analysis of 10 human lung cancer and paired normal lung tissues also revealed an elevated expression of Prx I and Trx proteins in lung cancer tissues. Immunohistochemical analysis of the lung cancer tissues confirmed an augmented Prx I and Trx expression in cancer cells with respect to the parenchymal cells in adjacent normal lung tissue. Based on these results, we suggest that the redox changes in lung tumor microenvironment could have acted as a trigger for the up-regulation of Prx I and Trx in lung cancer cells. Although the clinical significance of our finding awaits more rigorous future study, preferential augmentation of the Prx I and Trx in lung cancer cells may well represent an attempt of cancer cells to manipulate a dynamic redox change in tumor microenvironment in a manner that is beneficial for their proliferation and malignant progression.