Cytokine-induced activation of PKC isoenzymes in hemopoietic cell

Three cytokines which regulate the differentiation, mitogenesis, and activation of hemopoietic cells, interleukin (IL)-3, IL-5 and granulocyte-macrophage colony stimulating factor (GM-CSF), activate similar signalling pathways. Considerable knowledge regarding JAK-STAT and RAS-RAF pathways exists but, there is little information about the regulation of specific PKC isoforms by each of these cytokines. The first hypothesis was that unique functions of IL-3, IL-5 or GM-CSF result from activation of specific PKC isoforms. It was shown in human erythroleukemia cells, TF-1, that proliferation and tyrosine phosphorylation of similar substrates occurred in response to IL-3, IL-5, or GM-CSF. The results with IL-5 were unique. Growth of cells in IL-5 was necessary for the increased expression of the IL-5 receptor α subunit and responsiveness to IL-5. Another novel finding was that cells grown in IL-5 expressed increased levels of PKCα, δ and ε compared to cells grown in IL-3. Using SDS-PAGE and PKC isoform-specific antibodies, it was found that in cells stimulated with phorbol ester, PKCβ, α and ε translocated to the membrane from the cytosol. However, in response to IL-3, IL-5 or GM-CSF, PKCα and β signal translocation was not apparent. Interestingly, PKCδ and ε underwent band shifts in response to IL-3, IL-5, GM-CSF or phorbol ester. These novel results indicated that activation of PKCδ and ε may not require their translocation to the membrane. The recent findings, that PKCδ, ε and ς could be activated nonspecifically, by lipid products of phosphatidylinositol 3-kinase (PI3-kinase) led to a second hypothesis; that in response to IL-3, IL-5 or GM-CSF, PI3-kinase associates with specific PKC isoforms and somehow regulates their activity. It was found that both PKCδ and ε associated with PI3-kinase in unstimulated TF-1 cells. More importantly, PKCδ (but not PKCε) association with PI3-kinase was increased by GM-CSF treatment of the cells. Results using two inhibitors of PI3-kinase, wortmannin and LY294002, showed that the former inhibited the association while the latter did not, suggesting that PI3-kinase lipid products may not be required for the association. This association of PKCδ with PI3- kinase was also observed in platelets treated with platelet activating factor or thrombin. It was then hypothesized that perhaps the protein kinase activity of PI3-kinase regulates PKCδ activity. This was tested in platelets. Thrombin induced increased PKCδ kinase activity in the PKCδ:PI3-kinase immune complex. Also, within this complex, PKCδ underwent serine phosphorylation in a PI3-kinase protein kinase assay. The PKCδ activity and phosphorylation were inhibited by pre-treatment of platelets with LY294002 and wortmannin. Platelet aggregation by thrombin was also inhibited by LY294002 and wortmannin. In conclusion, these results support a role for an association of PI3- kinase and PKCδ that allows regulation of PKCδ activity by agonists in a novel fashion. Furthermore, this interaction occurs following hemopoietic receptor signalling or serpentine receptor signalling, suggesting one possible convergence of these pathways, at the point of PKCδ regulation.Medicine, Faculty ofMedicine, Department ofExperimental Medicine, Division ofGraduat