Figure 1.
PML/RARα exerts both repressive and activating functions through direct binding. (A) Genome-wide identification of PML/RARα direct target genes from integrative assays of ChIP-seq and RNA-seq. Schematic illustration of the experimental design, including the criteria used for analysis (left). Heatmap showing the expression change of the identified direct target genes upon PML/RARα knockdown (right). Two classes of target genes were categorized: PML/RARα-repressed targets (in blue) and PML/RARα-activated targets (in red). The maximum distance of 50 kb was used to associate ChIP-seq peaks with the nearest differentially expressed genes identified by RNA-seq. (B) Validation of PML/RARα targets in primary APL samples. ChIP-qPCR was performed on primary blasts from 2 APL patients using the antibody against the fusion site of PML/RARα. Negative controls include 2 irrelevant genomic regions with no PML/RARα binding signals (NC-1 and NC-2). Validation of PML/RARα targets in NB4 cells can be found in supplemental Figure 3. (C) PML/RARα effects on transcriptional activities of the directly repressed gene CEBPE (bottom right) or the directly activated gene GFI1 (top right). The left panel shows the genome browser tracks of PML/RARα binding, with the regions cloned in the luciferase constructs highlighted in dotted box. The luciferase reporter plasmid of each detected region was cotransfected with the PML/RARα-expressing plasmid in U937 cells or small interfering RNA (siRNA) targeting the fusion site of PML/RARα in NB4 cells. Luciferase activity was detected at 24 hours after transfection. N.C. siRNA, nonspecific siRNA control. Data represent the mean of 3 replicates ± standard deviation (SD). Statistical significance was determined using the unpaired, 2-tailed Student t test. *P < .01, **P < .001. chr, chromosome. (D) GSEA in terms of differentially expressed genes identified comparing blasts from 16 APL patients vs 163 non-APL AML patients (left panel) and comparing APL patients vs normal promyelocytes (right panel). NES, normalized enrichment score. (E) Enrichment analysis of PML/RARα-activated and repressed targets using the gene sets signifying genes specifically expressed in each of 10 stages of myeloid differentiation. t-Distributed stochastic neighbor embedding (t-SNE) plots showing the 10 stages of myeloid differentiation constructed using the transcriptome data of sorted hematopoietic stem cells, myeloid progenitors, and their mature progeny (bottom panel). The statistical significance of the enrichment was determined by the hypergeometric test (top panel). BCs, band cells; CMPs, common myeloid progenitors; GMPs, granulocyte/monocyte progenitors; HSCs, hematopoietic stem cells; MEPs, megakaryocyte/erythroid progenitors; MMs, metamyelocytes; MPPs, multipotential progenitors; MYs, myelocytes; PMs, promyelocytes; PMNs, polymorphonuclear cells. Red bars represent PML/RARα-activated targets, and blue bars represent PML/RARα-repressed targets.

PML/RARα exerts both repressive and activating functions through direct binding. (A) Genome-wide identification of PML/RARα direct target genes from integrative assays of ChIP-seq and RNA-seq. Schematic illustration of the experimental design, including the criteria used for analysis (left). Heatmap showing the expression change of the identified direct target genes upon PML/RARα knockdown (right). Two classes of target genes were categorized: PML/RARα-repressed targets (in blue) and PML/RARα-activated targets (in red). The maximum distance of 50 kb was used to associate ChIP-seq peaks with the nearest differentially expressed genes identified by RNA-seq. (B) Validation of PML/RARα targets in primary APL samples. ChIP-qPCR was performed on primary blasts from 2 APL patients using the antibody against the fusion site of PML/RARα. Negative controls include 2 irrelevant genomic regions with no PML/RARα binding signals (NC-1 and NC-2). Validation of PML/RARα targets in NB4 cells can be found in supplemental Figure 3. (C) PML/RARα effects on transcriptional activities of the directly repressed gene CEBPE (bottom right) or the directly activated gene GFI1 (top right). The left panel shows the genome browser tracks of PML/RARα binding, with the regions cloned in the luciferase constructs highlighted in dotted box. The luciferase reporter plasmid of each detected region was cotransfected with the PML/RARα-expressing plasmid in U937 cells or small interfering RNA (siRNA) targeting the fusion site of PML/RARα in NB4 cells. Luciferase activity was detected at 24 hours after transfection. N.C. siRNA, nonspecific siRNA control. Data represent the mean of 3 replicates ± standard deviation (SD). Statistical significance was determined using the unpaired, 2-tailed Student t test. *P < .01, **P < .001. chr, chromosome. (D) GSEA in terms of differentially expressed genes identified comparing blasts from 16 APL patients vs 163 non-APL AML patients (left panel) and comparing APL patients vs normal promyelocytes (right panel). NES, normalized enrichment score. (E) Enrichment analysis of PML/RARα-activated and repressed targets using the gene sets signifying genes specifically expressed in each of 10 stages of myeloid differentiation. t-Distributed stochastic neighbor embedding (t-SNE) plots showing the 10 stages of myeloid differentiation constructed using the transcriptome data of sorted hematopoietic stem cells, myeloid progenitors, and their mature progeny (bottom panel). The statistical significance of the enrichment was determined by the hypergeometric test (top panel). BCs, band cells; CMPs, common myeloid progenitors; GMPs, granulocyte/monocyte progenitors; HSCs, hematopoietic stem cells; MEPs, megakaryocyte/erythroid progenitors; MMs, metamyelocytes; MPPs, multipotential progenitors; MYs, myelocytes; PMs, promyelocytes; PMNs, polymorphonuclear cells. Red bars represent PML/RARα-activated targets, and blue bars represent PML/RARα-repressed targets.

Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal