Year XVI -Issue 06 - 2000

2/8

Each CSF has a 4-a-helical three-dimensional configuration, with M-CSF being a dimer of two such molecules. Each CSF is likely to have two small regions that engage the membrane receptors, the bulk of the molecule serving merely to ensure the correct configuration of these two binding regions. Each CSF is encoded by a single unique gene, the genes for GM-CSF and IL-3 being adjacent and clearly related ancestrally, although in fact the CSF proteins share no significant amino acid sequence homology.

Work on the CSFs documented three peculiar features regarding the CSFs that were not anticipated from existing ideas regarding specific cell regulators. However, subsequent events have shown that the features are typical enough of the more than twenty similar hematopoietic regulators now recognized.

First, none of the CSFs exhibits strict lineage specificity of action because each has actions stimulating the proliferation of cells in more than one lineage. For example, GM-CSF stimulates the proliferation of granulocytes, macrophages and eosinophils and at higher concentrations the proliferation of some megakaryocyte and erythroid precursors (7). M-CSF stimulates predominantly macrophage formation but also some granulocyte formation (8). G-CSF has the converse action - primarily being a stimulating factor for granulocyte formation but also able to stimulate some macrophage formation (9). Multi-CSF (IL-3), at unremarkable concentrations, stimulates the production of granulocytes, macrophages, eosinophils, megakaryocytes, mast cells, erythroid and stem cells (10). This pattern of polylineage action means, as evident in part from Figure 1, that multiple regulators exist that are able to stimulate cell proliferation in any one lineage - the sole possible exception being the erythroid lineage.

Because receptors for multiple regulators are coexpressed simultaneously on individual cells, the design pattern predicts that significant interactions should occur when two or more regulators are used simultaneously. Indeed, this is observable as the important process of superadditive synergy, where combinations of two regulators can elicit more cell proliferation than is achievable by the use of twice the concentration of either regulator alone (11). This important phenomenon allows economy of production of required concentrations of regulators but has yet to be properly used in clinical medicine.

Second, the CSFs were discovered and purified as mandatory proliferative stimuli for granulocytic and macrophage cells. The CSFs force non-cycling cells into cell cycle then exert a concentration-dependent action in shortening subsequent cell cycle times (4, 12). Without CSF stimulation there is no proliferation and, as CSF concentrations are increased, so the number of progeny produced by individual progenitor cells is increased. However, subsequent work has shown that the CSFs are not simply proliferative stimuli. They are necessary to maintain the functional activity of the cell membrane, and CSF withdrawal leads to death by apoptosis of responsive cells (13-15). Furthermore, the CSFs have an influence on lineage commitment decisions made by bipotential precursors (16) and can initiate the onset of maturation in responding immature cells (17). Finally, and importantly, the CSFs can enhance the functional activity exhibited by mature granulocytes and macrophages (18, 19). The polyfunctionality of the CSFs was initially greeted with some skepticism but is now recognized to be typical of other regulators. Indeed, there is probably no such thing as a "simple" proliferative regulator in any tissue system.

Third, the CSFs were found to differ radically from classical hormones in not having a single cellular source that was restricted to a single organ. For example, GM-CSF can be produced by a wide range of cell types that include stromal, endothelial, fibroblasts, lymphoid, macrophage and at least some epithelial cells. Indeed, it may be that all cells in the body have a capacity to produce one or other CSF if suitably induced. As a consequence of these diverse cellular sources, all organs have the capacity to produce one or more CSFs (20) and individual cells have been documented to be able to produce simultaneously more than one CSF and indeed other regulators. This pattern of CSF production means that at times CSFs can behave as classical humoral factors while at other times, CSF production might be quite localized with the CSF acting basically in a paracrine fashion.

Production Rates of the CSFs

Under basal conditions, levels of CSF production, with the possible exception of M-CSF, are very low and levels of CSF in organ extracts are barely detectible and often CSF is undetectible in the circulation. CSF production is however highly inducible and transcription and production of CSF can be elevated within hours to levels 1000-fold higher than basal levels (4). Initial work suggested that CSF production might be regulated by target cell consumption - for example, granulocyte numbers determining the resulting levels of G-CSF production. Certainly, consumption of CSF by receptor-mediated endocytosis and degradation is well documented, and

Figure 1: The major regulators controlling the in vitro production by stem cells of committed progenitor cells and the production by these cells of maturing granulocytes and monocyte-macrophages. Regulators listed in bold lettering are those shown by gene deletion experiments to be quantitatively the most important. The two regulators of granulocyte production in the lower box have only relatively weak actions in vitro (IL = interleukin, SCF = stem cell factor, FL = Flt3-ligand, G-CSF = granulocyte colony stimulating factor, GM-CSF = granulocyte-macrophage colony stimulating factor, M-CSF = macrophage colony stimulating factor, Multi-CSF = multipotential colony stimulating factor, TPO = thrombopoietin).

Figure 2: The major current clinical uses of G-CSF and GM-CSF are to stimulate the regeneration of granulocytes and macrophages following chemotherapy or transplantation and to elicit peripheral blood stem cells for transplantation. Less common are the use of G-CSF to generate neutrophils for infusion into leukopenic patients and the use of GM-CSF to enhance chronic ulcer and wound healing. Both agents are used to enhance resistance to infections in a variety of patients.