Regulation of colony-stimulating factor production by normal and leukemic human cells.

Considerable progress has been made over the last 5 years in defining the exact factors which make up 'colony-stimulating activity', the cells that produce individual CSFs, and determining some of the stimuli that lead to secretion of specific CSFs. There is much to learn however about the mechanisms of CSF action, and also much to learn about the role these factors play in hematopoietic regulation in vivo. The role, if any, of marrow stromal cells in the production of CSFs is particularly important and needs much clearer definition. Much of our understanding of CSF activity has been previously dependent on in vitro bioassays which were sensitive but frequently imprecise. The availability of purified recombinant protein has alleviated the reliance on conditioned media. Previously used conditioned media frequently contained multiple growth factors and inhibitory factors. The cloning of the CSFs has revealed both structural homology and diversity. The conserved genomic structural schema between the majority of the CSFs suggest a common ancestral gene. However, M-CSF diverges from this schema. Conserved also is the 3' untranslated motif of AUUUA in the majority of CSFs. M-CSF is again divergent in this respect. However, where regulation of the mRNA transcript level has been characterized carefully, normal cells appear to regulate CSF mRNAs primarily in a post transcriptional manner. The regulation of CSF transcription in leukemia is complex. In retrovirally mediated leukemia, CSF production is due to increased transcription mediated by the retrovirus. In the few cases of human leukemias making CSFs which have been studied, evidence for both post-transcriptional regulation and structural rearrangements in the CSF genes has been presented. Due to the extreme rarity of normal hematopoietic progenitor cells that correspond to the same state of differentiation as that of the leukemic blast forms, several questions remain. Do normal progenitor cells also make CSFs at some stages of differentiation? What role, if any, do CSFs play in leukemogenesis? The rapid development of our understanding of CSFs over the past several years has led to a much better understanding of hematopoiesis. As we understand more of normal hematopoiesis we also begin to understand the complexities involved in the abnormal regulation as in myelogenous leukemias. With the powerful tools currently available we can be much more precise in our understanding of the intricacies involved.