Several reports imply that bone marrow hematopoietic stem cells transdifferentiate into tissue-specific stem cells; however, the possibility of committed tissue-specific stem cells pre-existing in the bone marrow has not been dealt with adequately. We present here an alternative explanation of the so-called phenomenon of stem cell transdifferentiation. First, we postulate that tissue-committed stem/progenitor cells circulate in the peripheral blood and compete for tissue-specific niches. The circulation of these cells plays an important physiological role in maintaining a pool of stem cells in distant parts of the body and the number of these cells in peripheral blood can be increased by the administration of agents similar to those used for mobilization of hematopoietic stem cells. Second, we postulate that bone marrow tissue is a source of various stem-cell chemoattractants and survival factors and provides an environment that chemoattracts tissue-specific circulating stem/progenitor cells. In this context, we envision bone marrow as a "home" or "hide-out place" not only of hematopoietic stem cells but also of already differentiated circulating tissue-specific stem/progenitors. In support of this concept, we report here that mRNA of several early markers for muscle (Myf-5, Myo-D), neural (GFAP, nestin) and liver (CK19, fetoprotein) is detectable in circulating (adherent cell-depleted) peripheral blood mononuclear cells. Moreover, using real-time RT-PCR, we found that the level of expression of these markers increases in the peripheral blood of humans and mice after mobilization by G-CSF. Furthermore, using stromal-derived factor-1 (SDF-1) chemotaxis and real-time RT-PCR analysis, we present evidence that early tissue-specific stem cells reside in normal human and murine bone marrow, express the CXCR4 receptor on their surface and can be highly enriched (in humans and mice) after chemotaxis to SDF-1 gradient. All our experiments were performed on freshly isolated cells to exclude the potential contribution of transdifferentiated hematopoietic stem or mesenchymal cells in the culture. We maintain that any transdifferentiation studies employing populations of bone marrow cells should rule out the possibility that the apparently pure hematopoietic stem cell population could in fact contain pre-existing tissue-specific stem/progenitors.