Abstract
Studies using mice with germline or tissue-specific deletion of SOCS3 indicate that SOCS3 is a negative regulator of IL-6, LIF, Leptin and G-CSF-induced STAT3 phosphorylation. We have investigated the physiological importance of SOCS3 in blood cells by creating conditionally-targeted mice with SOCS3-deficient hematopoiesis (Immunity2004; 20:153–65). These mice develop a fatal inflammatory disease in adulthood characterized by tissue infiltration with neutrophils and macrophages, and display hyper-responsiveness to G-CSF in vitro and in vivo, with the development of pathological myeloid cell-mediated tissue damage. In hepatocytes and macrophages stimulated with IL-6, we and others (Nature Immunol2003; 4:540–5, and 4:546–60) have found that SOCS3 is not only important for modulating the intensity of signalling from cytokine receptors, but also for the maintenance of specificity of the signal. Whether these qualitative changes revealed by microarray experiments have physiological significance remains to be proven. To investigate this issue in a more plastic system, we analyzed the consequences of SOCS3-deficiency in hematopoietic progenitor cells. We hypothesised that if SOCS3 was required to maintain the specificity, as well as intensity of signals arising from cytokine receptors, then changes in differentiation as well as proliferation would be observed. To exclude confounding effects from cytokine production by mature cells, and to minimize potential selection bias within the starting populations, purified SOCS3-deficient lin- kit+ progenitor cells from healthy young mice with SOCS3-deficent hematopoiesis, were compared with both littermate control and C57BL/6 lin- kit+ progenitor cells. Proliferation was monitored in liquid and agar cultures stimulated with SCF, IL-3, GM-CSF, G-CSF and IL-6 alone, or in combination with SCF. At the completion of the experiments ( 1–7 days), proportions of neutrophil, macrophage and precursors were determined by microscopy. The number of divisions progenitors underwent was monitored by clone-mapping experiments in agar and CFSE-labelling in liquid cultures. No differences between SOCS3-deficient and wild-type (WT) cells were observed after stimulation with SCF, IL-3, GM-CSF or combinations of these, suggesting that progenitor cells of each genotype were similar in their developmental potential. However, marked differences were observed for G-CSF and IL-6. G-CSF induced a 2-3-fold increase in cellular output in both liquid and agar cultures, and the distribution of CFSE-intensity was consistent with an additional division occurring over a 4 day timespan in SOCS3-deficient cells. With IL-6 stimulation, while SOCS3-deficient progenitor cells initiated 1.7-fold more colonies, the overall cellular output was no greater than that of WT progenitors. With WT progenitors for both stimuli, the vast majority of clones were neutrophil colonies and >90% of emergent cells at 7 days were neutrophils or their precursors. In contrast, SOCS3-deficient progenitors stimulated with IL-6 generated 43% granulocyte-macrophage colonies and 13% pure macrophage colonies with >70% of emergent cells being macrophages. For G-CSF, a similar, but less pronounced shift towards macrophage development was observed. We conclude that SOCS3 is required to maintain normal cellular proliferative and differentiative responses to G-CSF and IL-6. The precise perturbations in signalling responsible for these aberrations are being defined through microarray and biochemical experiments.
Nuclear receptor RORα regulates pathologic retinal angiogenesis by modulating SOCS3-dependent inflammation