Abstract
Tissue hypoxia triggers the erythropoietic stress response, where high blood erythropoietin (Epo) stimulates increased red blood cell production rate. Stat5 is rapidly phosphorylated following ligation of the Epo receptor (EpoR) in erythroid cells in vitro and is required for normal erythropoiesis [Socolovsky et al. Blood, 2001; Cui et al., Mol Cel Biol, 2004]. Stat5-deficient mice, and mice expressing a truncated EpoR lacking Stat5 docking sites, are impaired in their response to erythropoietic stress, suggesting thatStat5 mediates EpoR signaling during stress [Socolovsky et al., 2001; Menon et al., J Clin Invest, 2006]. The identity of the erythroid progenitors in which Stat5 becomes active during stress, and the time-course of its activation, are not known. Recently, we developed flow-cytometric techniques that identify stress-responsive erythroblast subsets directly in freshly-explanted mouse hematopoietic tissue [Liu et al., Blood 2006]. Here we combined these techniques with intracellular flow-cytometry [Krutzik et al, J Immunol., 2005], to measure Stat5 activation within early erythroblasts in vivo.
We mimicked the effects of acute erythropoietic stress by injecting adult Balb/C mice with a single dose of Epo (10 IU/gram sub-cutaneously), and harvested spleen and bone marrow at different time points following Epo injection. These cells were labeled for the cell-surface markers Ter119 and CD71, and intracellularly with a specfic antibody against phospho-Stat5. Serum Epo was measured by ELISA.
Baseline Epo (10 to 50 mU/ml) increased to 600 mU/ml by 10 minutes post injection, peaked by 6 hours and remained high (over 5000 mU/ml) for 24 hours. Stat5 phosphorylation (=phospho-Stat5) was apparent by 15 minutes in both bone-marrow and spleen. In both tissues, it was highest in the least differentiated, ProE and Ery.A erythroblasts (Ter119-med CD71-high, and Ter119-high CD71-high FSC-high, respectively, Liu et al. 2006). In bone-marrow, the percentage of ProE that were positive for phospho-Stat5 (phospho-Stat5+) increased from a baseline of less than 1% to 65% by 30 minutes, but declined to 10% of ProE by 6 hours. This low-level of phospho-Stat5+ cells was maintained for the ensuing 10 hours. Of interest, in spite of the large variations in the percent of phospho-Stat5+ cells, the median phospho-Stat5 signal remained constant within the phospho-Stat5+ erythroblasts. This suggests that erythroblasts are either ‘on’ or ‘off’ with respect to Stat5 activation, and that the principal variant is the fraction of cells that are ‘on’ in the tissue. The decline in phospho-Stat5+ cells by 6 hours occurred in spite of persisting, high serum Epo, suggesting the activation of negative feedback mechanisms that limit EpoR signaling. We also noted a clear difference in the sensitivity of otherwise similar erythroblast subsets between spleen and bone-marrow: in spleen, a smaller percentage of erythroblasts became phospho-Stat5+, the signal was slower to develop and diminished sooner than in bone-marrow.
We conclude that Stat5 phosphorylation occurs rapidly upon an increase in serum Epo, but is likely to be damped from its peak by negative feedback meachanisms. Spleen erythroblasts are less sensitive than bone-marrow erythroblasts to Epo activation. Further, the principal regulation of the phospho-Stat5 signal appears to be at the level of the tissue, where the main variable is the fraction of cells expressing phospho-Stat5, rather than the level of phospho-Stat5 per cell. The molecular mechanisms responsible for this type of regulation remain to be elucidated.
A bacterial surface layer protein exploits multistep crystallization for rapid self-assembly
