Differential Activation of Notch1 and Notch2 by Delta1 and Jagged1 Determines Hematopoietic Cell Fate.

Spatially restricted in vivo expression of the Notch ligands Delta1 and Jagged1 suggests their differential roles in inducing hematopoietic cell fates, and studies have shown that each induces alternative fates during in vitro culture. We hypothesize that the ligands induce alternative fates via differential activation of Notch1 and/or Notch2. To address this, we assessed fate outcomes of Notch1- and Notch2-deficient murine bone marrow derived lin−Sca-1+c-kit+ Hoescht side population progenitors (LSKSP) after 14-days incubation with either purified Delta1 or Jagged1 in serum containing medium. Ligands consisted of purified Delta1 and Jagged1 extracellular domains fused to HumanIgG1 (Delta1ext-IgG or Jagged1ext-IgG). Equal amounts of ligand, as determined by ELISA, were immobilized to plastic surfaces of culture wells along with fibronectin. Notch deficient cells were generated by infecting either Notch1fl/fl or Notch2fl/fl LSKSP with lentivirus encoding cre recombinase. We show here that both ligands inhibit myeloid differentiation, since after 14 days, LSKSP incubated with either ligand generated multi-log increased numbers of immature progeny with significantly reduced percentages of GR1+ and/or F480+ cells compared to LSKSP incubated with control-IgG. However, only Delta1ext-IgG promotes T-cell progenitor differentiation, since a higher percentage of progeny cultured with Delta1ext-IgG expressed CD25 (37.0+/−0.6%) compared to Jagged1ext-IgG (1.7+/−1.0%; p=0.01). In contrast, Jagged1ext-IgG is less effective at inhibiting myeloid differentiation, since a higher percentage of progeny cultured with Jagged1ext-IgG (2.1+/−0.9%) expressed GR1 and F4/80 compared to Delta1ext-IgG (56.5+/−6.3; p=0.02). Furthermore, Delta1ext-IgG is more effective than Jagged1ext-IgG at inducing Notch activation as measured by increased expression of Notch target Hes1, since we found 3.3-fold more expression of Hes1 mRNA following incubation of LSKSP cells with Delta1ext-IgG compared to Jagged1ext-IgG. We further show that Notch2 is required to prevent myeloid differentiation, since Notch2 deficient LSKSP incubated with either Delta1ext-IgG or Jagged1ext-IgG generated cultures containing fewer numbers of cells and a higher percentage of GR1+ and/or F480+ myeloid progeny (83% with Delta1ext-IgGor 86% with Jagged1ext-IgG) similarly to those generated with control-IgG. Likewise, we found a reduced percentage of immature Sca-1+c-kit+ cells (19.3+/−4.4 or 13.0+/−5.2) than wild-type cells incubating with Delta1ext-IgG or Jagged1ext-IgG (92.2+/−3.0 or 71.0+/−15.0: p=0.004 or p=0.05). We found that Notch1 is required to induce T-cell differentiation, since Notch1 deficient LSKSP incubated with Delta1ext-IgG had a reduced percentage of CD25+ cells compared to wild-type cells (4.4+/−1.9% to 30.3+/− 3.3) even though myeloid differentiation was inhibited. In summary, we show that both Delta1ext-IgG and Jagged1ext-IgG induce signaling via Notch2 to prevent myeloid differentiation, whereas only Delta1ext-IgG induces signaling via Notch1 to promote generation of T-cell progenitors. Our results indicate unique Notch ligands differentially activate Notch1 or Notch2, resulting in alternative cell fate choices and lay a framework for investigating the mechanisms underlying differential activation, including determining the role of Notch modifiers such as Fringe.