In vivo mapping of notch pathway activity in normal and stress hematopoiesis

Abstract

Accumulating evidence suggests that Notch signaling is active at multiple points during hematopoiesis. Until recently, the majority of such studies focused on Notch signaling in lymphocyte differentiation and knowledge of individual Notch receptor roles has been limited due to a paucity of genetic tools available. In this manuscript we generate and describe animal models to identify and fate-map stem and progenitor cells expressing each Notch receptor, delineate Notch pathway activation, and perform in vivo gain- and loss-of-function studies dissecting Notch signaling in early hematopoiesis. These models provide comprehensive genetic maps of lineage-specific Notch receptor expression and activation in hematopoietic stem and progenitor cells. Moreover, they establish a previously unknown role for Notch signaling in the commitment of blood progenitors toward the erythrocytic lineage and link Notch signaling to optimal organismal response to stress erythropoiesis.

Figure 1. Notch(1-4)^CreER in vivo lineage tracing suggests a division of labor between Notch1&2 during early adult hematopoiesis

(A) Schematic depiction of the Notch(1-4)^CreER × ROSA26-tdRFP mouse strain. (B) Analysis of RFP reporter expression 72hrs after a single injection of tamoxifen focusing at defined stem and progenitors populations. Frequency of reporter labeling is represented as mean ± standard deviation (SD) (n=5 for each cohort). (C-D) Analysis of RFP reporter expression in LT-HSC and lymphoid progenitors and granulocyte/monocyte (G/M) and erythrocyte-megakaryocyte (Ery/Mk) progenitor subsets one week after a single tamoxifen injection (n=5). Bars denote mean ± SD. (E) Antibody staining with APC conjugated anti-Notch1 and PE conjugated anti-Notch2. Notch1 expression (blue histograms) and Notch2 (red histograms) are over-layed on isotype controls (grey filled histograms) for the same population. *p<0.05, **p<0.005 HSC: Lin^neg/cKit⁺/Sca1⁺/Flt3⁻/CD48⁻/CD150⁺, 48DP: Lin^neg/cKit⁺/Sca1⁺/Flt3⁻/CD48⁺/CD150⁺, 48SP: Lin^neg/cKit⁺/Sca1⁺/Flt3⁻/CD48⁺/CD150⁻, LMPP: Lin^neg/cKit⁺/Sca1⁺/Flt3⁺/IL7rα⁻, CLP: Lin^neg/cKit^lo/Sca1^lo/Flt3⁺/IL7rα⁺, GMP: Lin^neg/cKit⁺/Sca1⁻/CD41⁻/CD150⁻/FcgRII/III⁺, pre-MegE: Lin^neg/cKit⁺/Sca1⁻/CD41⁻/FcγRII/III⁻/CD150⁺/CD105⁻, CFU-E: Lin^neg/cKit⁺/Sca1⁻/CD41⁻/FcγRII/III⁻/CD150⁻/CD105⁺, MkP: Lin^neg/cKit⁺/Sca1⁻/CD41⁺/CD150⁺, DN3: CD4⁻/CD8⁻/CD44⁻/cKit⁻/CD25⁺, pro/preB: B220⁺/IgM⁻. See also Figure S1.

Figure 2. The Hes1^GFP reporter identifies novel points of Notch pathway activity

(A) Targeting strategy used to express GFP from the endogenous Hes1 promoter. (B) Analysis of LSK stem and progenitor cells in bone marrow of Hes1-GFP reporter mice. (C) High resolution separation of Lin^neg/cKit⁺/Sca1⁻ progenitors on the basis of CD105, CD150, CD16/32 (FcγRII/III) and CD41 expression and expression of Hes1^GFP in erythroid progenitors in bone marrow and (D) spleen (A representative of more than five experiments is shown). (E) Immunofluorescence staining of spleen from Hes1^GFP reporter mouse showing co-localization of GFP and CD71 staining in the red pulp. (F) Representative histograms showing GFP levels in erythroid progenitor in bone marrow (upper panel) and spleen (lower panel) of Ncstn^wt/wt Hes1^wt/GFP and Ncstn^−/− Hes1^wt/GFP. (G) Bar graphs representing average percentage of GFP positive cells in displayed populations from bone marrow (upper panel) and spleen (lower panel). See also Figure S2.

Figure 3. Differential patterns of Notch activity between fetal and adult hematopoiesis

(A) Quantitative RT-PCR analysis of Notch receptor genes in Lin^neg/cKit⁺/Sca1⁺/CD48⁻/CD150⁺ HSC, Lin^neg/cKit⁺/Sca1⁻/CD34^+/FcgRII/III^lo CMP, Lin^neg/cKit⁺/Sca1⁻/CD34⁻/FcgRII/III⁻ MEP, Lin^neg/cKit⁺/Sca1⁻/CD34^+/FcgRII/III^hi GMP and CD71+/Ter119⁺ RBC from E13.5 Fetal Liver. Data represent mean ± SD of 3 biological replicates (B) Expression of Hes1-GFP in Fetal Liver Lin^neg/cKit⁺/Sca1⁺ stem and multipotential progenitor populations separated into LT-HSC (CD48⁻/CD150⁺), CD48⁺/CD150⁺ double positive cells (CD48DP) and CD48⁺/CD150⁻ single positive cells (CD48SP) (upper panel); in Lin^neg/ cKit⁺/Sca1^−/CD41⁻/FcgRII/III⁻ Pre-MegE (CD150+/CD105−), PreCFU-E (CD150+/CD105+) and CFU-E (CD150−/CD105+) (middle panel) and late erythrocyte progenitors (lower panel). (C) Quantitative RT-PCR analysis of Notch signaling pathway target genes in Lin^neg/cKit⁺/Sca1⁻ progenitors (MP) and Lin^neg/cKit⁺/Sca1⁺ (LSK) GFP+ and GFP− from E13.5 Fetal Liver. Data represent mean ± SD of 3 biological replicates (D) Gene set enrichment plots of fractionated Lin^neg/cKit⁺/Sca1⁺ GFP⁺ versus GFP⁻ from E13.5 Fetal liver of Hes1-GFP mice for the indicated genesets. (E) Heat map of lymphoid differentiation genes and Notch signaling pathway target genes expressed in Lin^neg/cKit⁺/Sca1⁺ GFP⁺ versus GFP⁻ from E13.5 Fetal Liver. See also Figure S3.

Figure 4. Hes1 expression and Notch activity are predictive of commitment to the erythrocytic lineage

(A) One week after methylcellulose culture in a complete cocktail of cytokines, colonies from Hes1 expressing (upper panels) and Hes1 negative Lin^neg/cKit⁺/Sca1⁻ progenitors (lower panels) were analyzed by FACS for expression of Gr1 vs CD11b and Ter119 vs CD41. Representative images of colonies are shown. (B) CFU-E in methylcellulose supplemented with EPO and (C) CFU-Mk assays in collagen gel with TPO and IL-3 of sorted Lin^neg/cKit⁺/Sca1⁻ progenitors from Hes1^GFP bone marrow. (D) Hierarchical clustering of fractionated Hes1^GFP Lin^neg/cKit^+/Sca1⁻ progenitors with other progenitor populations. (E) Heat map of Notch target genes and key genes expressed in the erythroid, granulocyte/monocyte (G/M), and megakaryocyte (MegaK) lineages. Bars denote standard deviation. (F) Lineage potential of sorted pre-MegE progenitors was evaluated in complete methylcellulose seven days after sorting (G) CFU-E assay and (H) CFU-Mk assay of pre-MegE sorted based on Hes1 expression (Hes1⁺ dark grey bar, Hes1⁻ light grey bar). For (B,C) and (F-H) data are representative of average ± SD of 3 independent experiments. *p<0.05, **p<0.005. See also Figure S4.

Figure 5. Notch2 gain-of-function enhances erythroid differentiation

(A) Generation of ROSA26-Notch(1-4)-IC mice for conditional expression of intracellular Notch and IRES-YFP driven by the ROSA26 promoter (B) Representative FACS plot of Erythroid progenitors. CFU-E progenitors were increased in both the spleen and bone marrow upon conditional expression of ICN2. (C) Representative FACS plot of erythroblasts. CD71⁺ erythroblasts were increased in the spleen and bone marrow with ICN2, (D) Bar graph showing absolute myelo-erythroid progenitor counts per femur (n=4). (E) Platelet counts from peripheral blood of WT (white bar), ICN1+ (blue bar) or ICN2+ (red bar) mice. (F) H&E stained sections of spleen and bone marrow from control and ICN2 mice. Images in are at 10x magnification with 63x magnification inset in lower right. Bone marrow megakaryocytes were counted in 5 bone marrow (10X) high-powered fields (HPF) (n=3). For (D-F) data are representative of mean ± SD, *p<0.05, **p<0.005 (G) Lin^neg/cKit^+/Sca1⁻ progenitors expressing ICN2 were sorted for gene expression arrays and a heat map of genes involved in lineage specific differentiation was generated. (H) GSEA of erythroid gene signatures (d-ery) and myeloid-lymphoid genes (r-myly) in ICN2+ versus WT littermates Lin^neg/cKit^+/Sca1⁻ progenitors. See also Figure S5 and Table S1.

Figure 6. Notch loss-of-function affects early erythroid differentiation and recovery from erythroid stress

(A) Adult mice were analyzed one week after 3 injections of polyI:polyC to induce compound deletion of Notch1 and Notch2 using the Mx1-Cre strain (Mx1cre⁺Notch1^−/−2^−/−). The frequency of CFU-E progenitors and CD45⁻CD71⁺ erythroblasts is shown. (B) Absolute numbers of early erythroid progenitors in control and Notch1^−/−2^−/− mice (n=5) from bone marrow and (C) spleen. Data represent mean ± SD. (D) H&E stained paraffin sections of spleens from control, Notch1^−/−2^−/− and Ncstn^−/− mice. Megakaryocytes are indicated with black arrows. (E) Erythropoietic response to acute hemolytic anemia in control (Ncstn^f/f) and Ncstn^−/− (Ncstn^f/f Vav1-cre⁺) (n=5) after PHZ-induced hemolysis. Data represent mean ± SD. (F) Representative FACS plot (left panel) and quantification of proportion of GFP⁺ cells in Hes1^GFP/wt mice before (blue) and 4 days after (green) sublethal 4Gy irradiation. Data represent mean ± SD of 3 biological replicates (G) Representative FACS plots and (H) absolute quantification of erythroid progenitors from bone marrow of control (Ncstn^f/f) and Ncstn^−/− (Ncstn^f/f Vav1-cre⁺) (n=3) littermates 4 days after sublethal 4Gy irradiation. Data represent mean ± SD. *p<0.05, **p<0.005, ***p<0.001. See also Figure S6 and Figure S7

Figure 7. “Road map” of Notch signaling pathway in hematopoiesis

General overview of level of Notch1 receptor expression (blue gradient), Notch2 receptor expression (red gradient), Notch signaling pathway activation reported by Hes1^GFP expression (green gradient) as well as known niches and Notch ligands involved in adult hematopoiesis. LT-HSC: Long Term Hematopoietic Stem Cell; MPP: Multi-Potential Progenitor; CMP: Common Myeloid Progenitor; GMP: Granulocyte/Monocyte Progenitor; Pre-MegE: Pre-Megakaryocyte/Erythrocyte Progenitor; MkP: Megakaryocyte Progenitor; CFU-E: Colony Forming Unit Erythrocyte; ProE: Pro-Erythroblast; LMPP: Lymphocyte-Primed Multi-Potential Progenitor; CLP: Common Lymphocyte Progenitor; ProB: Pro B-cell; FoB: Follicular B-cell; MzB: Marginal zone B-cell; ETP: Early Thymic Progenitor; DN: Double Negative (CD4⁻8⁻) T cell progenitor 3.