Dr. Bianco began by noting that in 1868, it was observed that marrow continuously produces blood cells in post-natal life. Yet it was nearly 25 years later when uncooked marrow was first used to treat anemia. He cited this example to demonstrate that an incomplete understanding of biology often leads to an indirect translation into treatment. He observed that stem cell therapies are currently subject to paradigms that limit the full exploration of their therapeutic potential. He stressed that these paradigms must be supplanted by a new approach that focuses on identifying and understanding the ontogeny and function of postnatal progenitors, the use of stem cells as model systems, and the use of the cells as a target for intervention.

He observed that the transplantation of human stromal cells into mouse tissue promotes murine hematopoiesis in which the stromal cells are human and the endothelial compartment is murine. For example, CD146+ cells have the capability to self-renew into CD146+ adventitial reticular cells (Sacchetti B, et.al. Cell 2007;131:324-336). The cells express Ang-1 in vitro when undifferentiated and also in vivo. These cells have features of mural cells, “niche” cells, and osteoprogenitors. They direct and stabilize vasculogenesis in vivo, and their ability to create a niche suggests that they may be sub-endothelial, pre-osteoblastic cells (Moore KA and Lemischka IR. Science 2006;311:1880-1885). However, a proportion of the immunodeficient mice used in these transplantation experiments developed spontaneous leukemia, suggesting that the behavior of these cells can inform a model of human cancer metastasis in stem cell-generated human bone. Metastasis to bone is actually metastasis to bone marrow stroma. Markers such as CD146 and MCAM can identify pericytes in skeletal muscle and colony-forming units--fibroblasts (CFU-Fs) in muscle. CFU-F cells from muscle are highly myogenic but not osteogenic, and muscle CFU-F-derived cells share a phenotype with bone marrow stromal cells. CD146+ cells must be distinguished from satellite (CD56+) cells, endothelial (CD34+) cells, and CD34- and CD56- cells. A myogenic program can be activated in CD146+/CD34- cells in vitro. Moreover, human CD146+ cells contribute CD56+ cells and myofibers in immunodeficient mice with cardiotoxin injury.

Results from experiments with these cells support the existence of a system of committed progenitors in which plural potency results from plural origins. Committed myogenic cells (myoblasts) may give rise to pericytes, muscle tissue, and satellite cells, whereas committed osteogenic cells my give rise to pericytes and bone. This observation can be applied to the treatment of a disease such as fibrous dysplasia in which the causative mutation is carried within inner cell-mass cells; transplantation of stem cells that carry the mutation thereby causes the disease. Currently, models of the disease (e.g., transgenic mice and transgenic hSCs) are being developed. However, only 10% of mice that express the transgene develop the clinical phenotype. Introduction of the mechanism into the stem cell also introduces a series of compensatory mechanisms to counter the effects of the mutations. As such, these mechanisms decrease the likelihood that traditional pharmacologic interventions will be successful..

Discussion:

One attendee inquired whether CD146+ cells are committed and whether they differ between sites. Dr. Bianco noted that CD146 is marker of cell adhesion, and that these cells can be sorted from virtually all tissues. They express a similar surface phenotype regardless of source. However, when their potency is probed, the populations are remarkably different. Those in bone give rise to bone, whereas those in muscle give rise to muscle but not bone. Another participant noted that human bone marrow does not contain any immune reactivity for Ang-1. Dr. Bianco observed that marrow cells that produce Ang-1 are not mature osteoblasts.