G-CSF Receptor Mutations Found in Patients with Severe Congenital Neutropenia Confer a Strong Competitive Growth Advantage at the Hematopoietic Stem Cell Level That Is Mediated by STAT5 Activation.

Patients with severe congenital neutropenia (SCN) have a markedly increased risk of developing myelodysplasia (MDS) or acute myeloid leukemia (AML). Though the genetic basis for this increased susceptibility is unknown, gain-of-function mutations of the G-CSF receptor (G-CSFR) have been found in the great majority of patients with SCN who develop MDS/AML. These mutations are somatic and produce a truncated G-CSFR that, though remaining ligand-dependent, transmits a hyperproliferative signal. We and others have shown that targeted transgenic mice expressing a representative G-CSFR mutation (d715) have markedly exaggerated neutrophil responses to G-CSF treatment. Based on these observations, it has been suggested that these gain-of-function G-CSFR mutations contribute to leukemogenesis. However, direct evidence supporting this hypothesis is scant. Moreover, it is unclear how hematopoietic cells expressing the mutant G-CSFR gain clonal dominance.

We previously showed that expression of the d715 G-CSFR results in a strong competitive advantage at the hematopoietic stem cell (HSC) level, but only in the presence of an increased concentration of G-CSF. Herein, we describe studies to characterize the cellular and molecular mechanisms responsible for the clonal dominance of HSC expressing the d715 G-CSFR. At baseline, the percentage of cycling c-Kit⁺ lineage⁻ Sca⁺ (KLS) cells was similar in WT (8.9±2.0%) and d715 G-CSFR mice (10±3.0%). However, 24 hours after a single injection of G-CSF, a significantly greater percentage of cycling KLS was observed in d715 G-CSFR compared with WT mice (34.4±2.4% versus 20±3.8%; p < .05). We next harvested KLS cells from WT or d715 G-CSFR mice 3 hours after treatment with a single injection of G-CSF or saline alone and performed RNA expression profiling. 14 genes were identified that were consistently differentially regulated by G-CSF in d715 G-CSFR versus WT KLS cells. A striking features shared by most of these genes is their regulation by STAT3 or STAT5. These data suggested the hypothesis that activation of STAT3 and/or STAT5 transduces the signal leading to HSC clonal dominance. To test this hypothesis, we first directly measured STAT activation by G-CSF in KLS cells using a flow cytometry-based method. These data showed that STAT3 and to a lesser extent STAT5 are activated by G-CSF in WT KLS cells. d715 G-CSFR KLS cells displayed significantly increased STAT5 activation by G-CSF, but STAT3 activation was slightly decreased, suggesting that STAT5 may play a key role in HSC activity. To further test this hypothesis, mice carrying a targeted mutation of their G-CSFR in which the sole remaining tyrosine in d715 G-CSFR is mutated to phenylalanine (termed d715F) were analyzed. Importantly, G-CSF induced STAT3 and STAT5 in KLS cells from d715F mice was markedly attenuated. Competitive repopulation experiments showed that the d715F G-CSFR did not confer a clonal advantage and may, in fact, confer a competitive disadvantage. Collectively, these data suggest that the d715 G-CSFR confers a clonal advantage at the HSC level that may be mediated by accentuated STAT5 activation.