Figure 5.
Enhanced FLT3ITDsignaling cooperates with NHD13 to create an emergent, isolatable pre-AML population. (A) Diagram showing removal of PGK-neo cassette with Vav1-Cre. (B) Western blot showing elevated STAT5 and ERK phosphorylation in Flt3ITDΔ MPPs. (C) Representative cytospin of Flt3ITDΔ/NHD13 AML. The scale bar indicates 20 μM. (D) Flow plot showing ectopic B220 expression in Flt3ITDΔ/NHD13 AML. (E) Normalized Bst2/CD317 expression in NHD13 and Flt3ITDΔ/NHD13 LK cells based on CITE-seq experiments in Figure 4. The NHD13 and Flt3ITDΔ/NHD13 columns reflect cells in cluster 13. “ALL” reflects cells in all other genotypes. ∗∗∗P < .0001 based on the Wilcoxon rank sum test. (F) Overview of limiting dilution experiment after selection of CD317+LK cells. (G) Flow cytometry plot showing CD317 expression in Flt3ITD/NHD13 LK cells. (H) Limiting dilution curves with comparison by extreme limiting dilution analysis.36 (I) UMAP plots demonstrating clusters of cells from ICGS2 for the indicated genotypes. The heatmap is shown in supplemental Figure 8. Runx1DEL clusters are also shown in supplemental Figure 8. (J) Bar graph showing percentages of cells for each genotype that populate clusters 15 and 33. Cluster 33 emerges with NHD13 and is amplified by Flt3ITD and Flt3ITDΔ (cluster 15 only emerges in Flt3ITDΔ/NHD13 mice). ∗∗∗P < .0001 by χ2 test relative to wild-type. Cluster 15 and 33 sizes were also significantly larger in Flt3ITDΔ/NHD13 relative to single mutants (P < .0001). (K) Reactome pathway analysis of cluster 33 and 15 marker genes. Significant gene ontology biological processes gene sets (FDR < 0.05) were ranked on the y-axis based on log(P values) shown on the x-axis. IFN-, tyrosine kinase- and RAS pathway-related gene sets are color coded as indicated. (L) GSVA scores for the IFN-1 target gene set in individual MPP3 or MPP4, based on surface barcode expression by CITE-seq. ∗∗∗P < .001 relative to wild-type. ### IFN-1 signature enrichment is elevated relative to all other genotypes by one-way ANOVA and Holm-Sidak posthoc test, P < .001. ˆˆˆIFN-1 signature enrichment is elevated relative to all other genotypes by one-way ANOVA and Holm-Sidak posthoc test, P < .001. (M) Cluster 15 and 33 markers were highly expressed in Flt3ITDΔ/NHD13 but not Flt3ITDΔ/Runx1DEL/DEL AML by GSVA. (N) GSVA analysis of cluster 33 markers in human NUP98-t AML relative to KMT2A-translocated or NPM1-mutated AML from combined TARGET and St. Jude pediatric data sets. (O) GSVA analysis of cluster 33 markers in human FLT3ITD-positive or negative AML from TARGET (pediatric) and TCGA (adult) data sets. For panels M-O, ∗P < .05, ∗∗∗P < .001 by Student t test. Box and whisker plots are shown by Tukey’s method. ANOVA, analysis of variance; ERK, extracellular signal-regulated kinase; ICGS2, Iterative Clustering and Guide Gene Selection version 2; TCGA, The Cancer Genome Atlas.

Enhanced FLT3ITDsignaling cooperates with NHD13 to create an emergent, isolatable pre-AML population. (A) Diagram showing removal of PGK-neo cassette with Vav1-Cre. (B) Western blot showing elevated STAT5 and ERK phosphorylation in Flt3ITDΔ MPPs. (C) Representative cytospin of Flt3ITDΔ/NHD13 AML. The scale bar indicates 20 μM. (D) Flow plot showing ectopic B220 expression in Flt3ITDΔ/NHD13 AML. (E) Normalized Bst2/CD317 expression in NHD13 and Flt3ITDΔ/NHD13 LK cells based on CITE-seq experiments in Figure 4. The NHD13 and Flt3ITDΔ/NHD13 columns reflect cells in cluster 13. “ALL” reflects cells in all other genotypes. ∗∗∗P < .0001 based on the Wilcoxon rank sum test. (F) Overview of limiting dilution experiment after selection of CD317+LK cells. (G) Flow cytometry plot showing CD317 expression in Flt3ITD/NHD13 LK cells. (H) Limiting dilution curves with comparison by extreme limiting dilution analysis.36 (I) UMAP plots demonstrating clusters of cells from ICGS2 for the indicated genotypes. The heatmap is shown in supplemental Figure 8. Runx1DEL clusters are also shown in supplemental Figure 8. (J) Bar graph showing percentages of cells for each genotype that populate clusters 15 and 33. Cluster 33 emerges with NHD13 and is amplified by Flt3ITD and Flt3ITDΔ (cluster 15 only emerges in Flt3ITDΔ/NHD13 mice). ∗∗∗P < .0001 by χ2 test relative to wild-type. Cluster 15 and 33 sizes were also significantly larger in Flt3ITDΔ/NHD13 relative to single mutants (P < .0001). (K) Reactome pathway analysis of cluster 33 and 15 marker genes. Significant gene ontology biological processes gene sets (FDR < 0.05) were ranked on the y-axis based on log(P values) shown on the x-axis. IFN-, tyrosine kinase- and RAS pathway-related gene sets are color coded as indicated. (L) GSVA scores for the IFN-1 target gene set in individual MPP3 or MPP4, based on surface barcode expression by CITE-seq. ∗∗∗P < .001 relative to wild-type. ### IFN-1 signature enrichment is elevated relative to all other genotypes by one-way ANOVA and Holm-Sidak posthoc test, P < .001. ˆˆˆIFN-1 signature enrichment is elevated relative to all other genotypes by one-way ANOVA and Holm-Sidak posthoc test, P < .001. (M) Cluster 15 and 33 markers were highly expressed in Flt3ITDΔ/NHD13 but not Flt3ITDΔ/Runx1DEL/DEL AML by GSVA. (N) GSVA analysis of cluster 33 markers in human NUP98-t AML relative to KMT2A-translocated or NPM1-mutated AML from combined TARGET and St. Jude pediatric data sets. (O) GSVA analysis of cluster 33 markers in human FLT3ITD-positive or negative AML from TARGET (pediatric) and TCGA (adult) data sets. For panels M-O, ∗P < .05, ∗∗∗P < .001 by Student t test. Box and whisker plots are shown by Tukey’s method. ANOVA, analysis of variance; ERK, extracellular signal-regulated kinase; ICGS2, Iterative Clustering and Guide Gene Selection version 2; TCGA, The Cancer Genome Atlas.

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal