Stem cell plasticity revisited: CXCR4-positive cells expressing mRNA for early muscle, liver and neural cells ‘hide out’ in the bone marrow MZ Ratajczak 1,2 , M Kucia 1,2 , R Reca 1 , M Majka 2 , A Janowska-Wieczorek 3 and J Ratajczak 1 1 Stem Cell Biology Program at James Graham Brown Cancer Center and Department of Medicine, University of Louisville, Louisville, KY, USA; 2 European Stem Cell Therapeutic Excellence Center, Medical College, Jagiellonian University, Cracow, Poland; and 3 University of Alberta and Canadian Blood Services, Edmonton, Canada It has been suggested that bone marrow (BM)-derived hemato- poietic stem cells transdifferentiate into tissue-specific stem cells (the so-called phenomenon of stem cell plasticity), but the possibility of committed tissue-specific stem cells pre-existing in BM has not been given sufficient consideration. We hypothesized that (i) tissue-committed stem cells circulate at a low level in the peripheral blood (PB) under normal steady- state conditions, maintaining a pool of stem cells in peripheral tissues, and their levels increase in PB during stress/tissue injury, and (ii) they could be chemoattracted to the BM where they find a supportive environment and that the SDF-1–CXCR4 axis plays a prominent role in the homing/retention of these cells to BM niches. We performed all experiments using freshly isolated cells to exclude the potential for ‘transdifferentiation’ of hematopoietic stem or mesenchymal cells associated with in vitro culture systems. We detected mRNA for various early markers for muscle (Myf-5, Myo-D), neural (GFAP, nestin) and liver (CK19, fetoprotein) cells in circulating (adherent cell- depleted) PB mononuclear cells (MNC) and increased levels of expression of these markers in PB after mobilization by G-CSF (as measured using real-time RT-PCR). Furthermore, SDF-1 chemotaxis combined with real-time RT-PCR analysis revealed that (i) these early tissue-specific cells reside in normal murine BM, (ii) express CXCR4 on their surface and (iii) can be enriched (up to 60  ) after chemotaxis to an SDF-1 gradient. These cells were also highly enriched within purified populations of murine Sca-1 þ BM MNC as well as of human CD34 þ -, AC133 þ - and CXCR4-positive cells. We also found that the expression of mRNA for SDF-1 is upregulated in damaged heart, kidney and liver. Hence our data provide a new perspective on BM not only as a home for hematopoietic stem cells but also a ‘hideout’ for already differentiated CXCR4-positive tissue-committed stem/ progenitor cells that follow an SDF-1 gradient, could be mobilized into PB, and subsequently take part in organ/tissue regeneration. Leukemia (2004) 18, 29–40. doi:10.1038/sj.leu.2403184 Published online 30 October 2003 Keywords: CXCR4; SDF-1; stem cell plasticity; stem cell mobilization; stem cell homing Introduction Bone marrow (BM) has been traditionally envisioned as a ‘home’ of hematopoietic stem cells. It is also known to contain mesenchymal stem cells that give rise to the various mesodermal tissues (eg smooth muscles, chondrocytes, osteocytes). 1–3 Surprisingly, BM hematopoietic stem cells were also recently reported to be able to ‘transdifferentiate’ into cells that express early heart, 4,5 skeletal muscle, 6,7 neural, 8 liver 9 or pancreatic cell 10,11 markers. This was supported in humans by the observations that transplantation of peripheral blood (PB) stem cells expressing early hematopoietic CD34 þ antigen led to the appearance of donor-derived hepatocytes, 12 epithelial cells 12 and neurons. 13 Similarly, human BM-derived cells contributed to the regeneration of infarcted myocardium. 14 These findings have been interpreted as evidence for the existence of the phenomenon of transdifferentiation or plasticity of adult stem cells. However, the concept of transdifferentiation of adult tissue- specific stem cells has recently been called into question. 15,16 Studies aimed at reproducing these experiments with neural stem cells have failed. 17 Similarly, it has been shown that hematopoietic stem/progenitor cells (HSPC) found in muscle tissue are in fact of BM origin. 18–20 Recent studies on chimeric animals involving the transplantation of a single hematopoietic stem cell marked with a green fluorescent protein into lethally irradiated nontransgenic mice demonstrated that so-called transdifferentiation or plasticity of circulating HPSC and/or their progeny is an extremely rare event, if it occurs at all. 21 Several potential alternative explanations for the phenomen- on of apparent stem cell plasticity should be taken into consideration when evaluating reported data. First, some of the results could be explained by the phenomenon of cell fusion. 22–24 The significant role of cell fusion has recently been demonstrated during liver regeneration in mice transplanted with BM cells. 25 Next, it is likely that in certain circumstances cells may undergo epigenetic changes caused by external stimuli that affect/change gene expression. 26 Supporting this are observations showing that, during animal cloning employing nuclear transfer, the nuclei isolated from differentiated somatic cells may be reprogrammed and de-differentiated when injected into the cytoplasm of enucleated embryonic stem cells or oocytes. 27,28 It is likely that a similar ‘reprogramming effect’ may also occur in cells isolated from their physiological environment and exposed in vitro to stress factors related to culture conditions. Finally, in our opinion, there is the explanation, which we believe to be the most likely, that committed tissue-specific stem cells pre-exist in various organs/ tissues. 29 We present evidence for this explanation and discuss it with respect to BM tissue. There is compelling evidence that the a-chemokine stromal derived factor (SDF)-1, which is secreted by BM stroma, plays an essential role not only in the homing of HSPC 30 but also in chemoattracting to the BM other CXCR4-positive cells. 31–35 Several pediatric tumors metastasizing to the BM such as rhabdomyosarcoma, neuroblastoma, nephroblastoma, hepato- blastoma and retinoblastoma express functional CXCR4 on their surface and follow an SDF-1 gradient. 31–35 All these tumors derive from early muscle, neural, kidney, liver and retina pigment epithelial cells, respectively, which are also CXCR4- positive. 31–46 Hence it seems likely that the CXCR4–SDF-1 axis plays an essential role in the chemoattraction/retention in BM Received 13 August 2003; accepted 24 September 2003; Published online 30 October 2003 Correspondence: Dr MZ Ratajczak, Stem Cell Biology Program at James Graham Brown Cancer Center, University of Louisville, 529 South Jackson Street, Louisville, KY 40202, USA This paper was presented as an oral presentation at the 32nd Annual Meeting of the International Society for Experimental Hematology, Paris, France, July 5–8, 2003 Leukemia (2004) 18, 29–40 & 2004 Nature Publishing Group All rights reserved 0887-6924/04 $25.00 www.nature.com/leu