The intramedullary control of marrow cell production has been a difficult area to approach experimentally. The introduction by Dr. Dexter and colleagues of long-term stromal dependent culture systems for murine marrow and the adaptation of these systems to human marrow growth have allowed for in-vitro studies of stromal dependent hemopoiesis. Despite some controversy in this area, most studies appear to show that adherent murine or human stromal cells are capable of producing a relatively large number of hemopoietic growth factors including G-CSF, GM-CSF, CSF-1, IL-6 and, at least by PCR analysis, IL-3. Other work indicates that the most primitive hemopoietic cells which appear to be multifactor responsive adhere directly to these stromal cells presumably through mediation of various adherence proteins.
An early acting, multilineage factor termed hemolymphopoietic growth factor-I (HLGF-1) has been isolated from a murine stromal cell line and may be identical to the recently described ligand for the c-kit receptor. This may represent an important early survivalhaintenance factor for stem cells in this system.
Studies on primitive stem cells, especially the high proliferative potential colony forming cell (HPP-CFC), indicate that they are responsive to varying combinations of growth factors and that with increasing numbers of growth factors, as studied in serum-free systems, decreasing concentrations of the factors may be biologically active.
These observations altogether suggest that intramedullary hemopoiesis may be regulated by the positioning of early multifactor responsive stem cells via adherent proteins in juxtaposition to synergistically acting combinations of growth factors attached to stromal cell surfaces or the extracellular matrix. In addition, selective production of different growth factors from different subsets of cells may create growth factor gradients and explain the spacial distribution of different cell types within the marrow cavity.