Article Figures & Data
Figures
- Figure 1. Cebpb Deletion in MLL-ENL–Transformed Cells.
(A) Schematic illustration of murine tissue culture MLL-ENL leukemia model. Top to bottom: Murine fetal liver cells from WTFL animals were transduced with MLL-ENL and selected with G418 for 14 d in liquid culture. Subsequently, the cells were seeded in semi-solid methylcellulose medium. Single colonies were isolated, expanded in liquid culture, and assessed for MLL-ENL integration. Next, the cells were treated with TAT-Cre recombinase to remove floxed Cebp alleles. Gene excision was determined by PCR (see Fig S1B). (B) Top: representative microscopic scans of semi-solid methylcellulose cultures with WTFL and Cebpb KO colonies. Cells were seeded at a density of 5,000 cells per 35-mm well, and colonies were scored after 10 d. Bottom: colony size distribution (see also Fig S1C). (C) Top: WTFL and Cebpb KO colonies in semi-solid methylcellulose medium. Middle: Growth curves of WTFL and Cebpb KO colonies. Bottom: WST-1 assay showing the effect of Cebpb removal on MLL-ENL–transformed cells. Values are the mean ± SD (two-tailed Mann–Whitney U test, **P < 0.005).
- Figure S1. Myeloid Identity of MLL-ENL–Transformed Cells.
(A) Flow cytometry profiles (c-Kit, CD16/32, CD11b) of Cebpbfl/fl and Cebpafl/fl/Cebpbfl/fl MLL-ENL–transformed cells. (B) Removal of Cebpa and Cebpb genes from MLL-ENL–transformed cells, as indicated by representative genotyping results for the Cebpa and Cebpb deletion alleles detected by PCR. (C) Comparison of colony size distribution in semi-solid medium (in square pixel) indicating the calculated estimated cell number based on the colony size (right).
- Figure 2. Characterization and comparison of MLL-ENL–transformed WTFL, C/EBPβ KO, and C/EBPα/C/EBPβ dKO cells.
(A) Flow cytometric phenotyping of WTFL, Cebpb KO, and Cebpa/Cebpb dKO cells using CD11b, Ly6C, CD117, CD115, and Ly6G markers (left panel). Apoptosis was examined via annexin V staining in combination with 7-AAD (right panel). Two independent experiments were performed, yielding similar results. (B) Analysis of frequency of cell division using CFSE in flow cytometry. Cell types as in A and cells were analyzed 1, 3, and 4 d after CFSE loading. Two independent experiments were performed, yielding similar results. (C) Cytospins were prepared from exponentially growing cell types and stained with Giemsa/eosin. Micrographs were taken from small groups of cells (left) to show uniformity and from enlarged single cells (right) to show differences in subcellular features. (D) Serial replating in methylcellulose/Iscove’s DMEM supplemented with IL-3, stem cell factor, GM-CSF, and IL-6. Cells were seeded in triplicate, grown for 7 d, and colonies were counted before reseeding at 5,000 cells/well. Two independent experiments and for third and fourth replating three experiments were performed, yielding similar results. (E) Micrographs of colonies from fourth replating. (F) Cells from the third replating round were seeded in semi-solid medium supplemented with the complete cytokine cocktail or IL-3 only, as indicated. Colonies were counted after 7 d. Three independent experiments were performed, yielding similar results.
- Figure 3. MLL-ENL–Transformed dKO Clones Express C/EBPε.
(A) Data showing normalized RNA-seq read counts for indicated genes of the Cebp family in WTFL (gray) and two independent Cebpa/Cebpb dKO MLL-ENL–transformed cell clones (red) derived from different mice. RNA-seq was performed in triplicates. (B) Immunoblots show the expression of resident C/EBP family proteins from WTFL and dKO transformed MLL-ENL cells. The panel summarizes C/EBP family protein expression in all 14 dKO clones from two different mice (clone #1–12, mouse 1; clone #13 and #14, mouse 2).
- Figure 4. Complementation of Growth Defects by C/EBPα and C/EBPβ Isoforms in MLL-ENL–Transformed dKO Cells.
(A) Transcripts of Cebpa and Cebpb with alternative translation initiation sites (arrowheads; cyan) giving rise to different protein isoforms (indicated underneath, purple, and on the right) as examined in Fig 4B. Below: C/EBP expression controls of retrovirally expressed FLAG-tagged C/EBP isoforms expressed in MLL-ENL–transformed dKO cells. Cell lysate immunoblots (IB) were probed with anti-FLAG. GFP: negative control retrovirus; C/EBPα isoforms p42, p30; C/EBPβ isoforms LAP*, LAP, and LIP. (B, C, D) C/EBPα complementation: (B, C, D). (E, F, G) C/EBPβ complementation: (E, F, G). (B, C, D, E, F, G) Left, bottom insert: Color code of complementation assays, as shown in (B, C, D, E, F, G). (B) Number of MLL-ENL–transformed colonies derived from WTFL or dKO cells with and without complementation by retrovirally encoded Cebpa p30. Cells were seeded at 500 cells per 35-mm well. The significance of the change in colony number between the dKO and other groups was evaluated with the Kruskal–Wallis test, followed by the post hoc Dunn test. (C) Growth curves of WTFL, dKO, and dKO cells complemented with p30 C/EBPα. (D) WST-1 assay of WTFL, dKO, and dKO cells complemented with p30 C/EBPα. Values are the mean ± SD. The significance of the change in metabolism level between the dKO and other groups was evaluated with the Kruskal–Wallis test, followed by the post hoc Dunn test. (B, C, D, E) Colony formation of MLL-ENL cells as in (B, C, D). Panel shows the number of MLL-ENL–transformed colonies formed in methylcellulose medium by WTFL, dKO, and complementation by retrovirally encoded C/EBPβ LAP*, LAP, and LIP isoforms. Cells were seeded at 500 cells per 35-mm well. The significance of the change in colony number between the dKO and other groups was evaluated with the Kruskal–Wallis test, followed by the post hoc Dunn test. (F) Growth curves of MLL-ENL–transformed WTFL, dKO, and LAP*-, LAP-, LIP-C/EBPβ complementation dKO cells. (G) WST-1 assay of MLL-ENL–transformed WTFL, dKO, and LAP*-, LAP-, and LIP-C/EBPβ complementation dKO cells. Values are the mean ± SD. The significance of the change in metabolism level between the dKO and other groups was evaluated with the Kruskal–Wallis test, followed by the post hoc Dunn test.
- Figure S2. Colony Formation of MLL-ENL–Transformed dKO Cells.
(A) Size frequency distribution of MLL-ENL–transformed colonies in WTFL, dKO, and dKO cells complemented with C/EBPα p30. (B) Size frequency distribution of MLL-ENL transformed colonies in methylcellulose medium by WTFL, dKO, and dKO complemented with LAP*/LAP/LIP cells. (C) The significance between the populations’ average colony size (as shown in Fig S2A) was evaluated with the Kruskal–Wallis test with the post hoc Dunn test (****P ≤ 0.0001). (D) The significance between the populations’ average colony size (as shown in Fig S2B) was evaluated with the Kruskal–Wallis test with the post hoc Dunn test (***P ≤ 0.0002, ****P ≤ 0.0001).
- Figure 5. Examination of C/EBPε dependency of MLL-ENL–transformed Cebpa/Cebpb dKO C/EBPε+ cells via genome editing.
(A) C/EBPε protein levels after prolonged growth of WTFL and dKO cells infected with control virus (GFP) or vectors encoding different C/EBPβ isoforms. (B) Scheme of Cas9-mediated genome editing. C/EBPε+ dKO cells were infected with control virus or C/EBPβ LAP*-encoding virus, both expressing GFP. After infection, cells were grown for 3 d, sorted by flow cytometry, and GFP+ cells were expanded for 9 d. C/EBPε+ dKO vector control cells and C/EBPβ LAP*-complemented cells were infected with retrovirus encoding BFP, Cas9, and guide RNAs targeting Cebpe exon 1. After 3 d, the cells were sorted and expanded for 6 d, and seeded in semi-solid medium containing IL-3. Individual colonies were isolated and expanded in 96-well plates in liquid medium (plus IL-3). (C) Growth of isolated cells from single colonies was microscopically inspected over 1 wk and scored as fast, slow, or abortive growth. (D) Randomly chosen fast-growing clones from (D) were examined by protein blotting for expression of C/EBPε and C/EBPβ, and GAPDH as the loading control by protein-specific antibodies, as indicated on the left. (E) Cebpe exon 1 DNA derived from 27 vector control clones and 31 C/EBPβ LAP*-complemented clones was amplified by PCR and analyzed by Sanger sequencing and the Inference of CRISPR Edits (ICE) software tool (ice.synthego.com). (F) (Middle) Scaled schematic representation of C/EBPε showing the transactivation (TAD), regulatory (RD1 and RD2), and basic zipper (bZIP) regions. Top: examples of in-frame deletions in RD1 from two vector control clones (clones #49 and #56). Bottom: examples of in-frame deletions in RD2 (vector control clone #12) and one LAP*-complemented clone (#4) with an in-frame deletion in one C/EBPε allele and frameshift in the second allele.
- Figure 6. Molecular Genetic Profiling of MLL-ENL–Transformed dKO Cells.
(A) Venn diagrams showing the overlap between up-regulated (Benjamini–Hochberg corrected P-value < 0.05) and down-regulated (Benjamini–Hochberg corrected P-value < 0.05) genes in both dKO clones as determined by RNA-seq. N = 3 per genotype. (B) Scatter plot showing dysregulation of MLL-ENL target genes, which are colored according to P-value and fold change. All remaining genes are colored according to their P-value only (see figure key). Two genes, Igf1 and Bcl11a, are included in the plot, as they are known MLL translocation targets. (C) Gene set enrichment analysis showing significant enrichment of genes co-repressed by HOXA9/C/EBPα among the deregulated genes. Most of these genes were up-regulated in the dKO cells. (D) Comparison of expression patterns to that of ImmGen data shows a shift in overall gene expression patterns, with loss of myeloid characteristics and maintenance of stem cell patterns in dKO cells indicated by Spearman correlation. (E) Heatmap of inflammatory and TNFα/NF-κB hallmark genes (gene sets derived from MSigDB) in WTFL and dKO cells. Both terms are significantly enriched (adjusted P-value < 0.05, according to Gene set enrichment analysis). The heat maps show genes of the respective gene sets exhibiting an adjusted P-value < 0.05 and an absolute fold change > 2. Row z-score of normalized log2 counts. (F) Subset of the co-regulated transcriptome and genome structure as determined by RNA-seq and ATAC-seq. The whole list comprises genes that exhibit both the re-establishment of gene expression and restoration of the chromatin status in dKO-LAP* cells to WTFL status. Row z-score of normalized log2 counts. (F, G) Known motif prevalence in dKO-LAP* cells in the full set of genes shown in (F) as determined by HOMER. (H) Examples of replicate ATAC-seq data from WTFL, dKO, and dKO-LAP* MLL-ENL–transformed cells indicating gain of peaks (left, blue outline) in dKO at the Sox4 and Ikzf2 loci, and loss of peaks at the Bcl11a and Igf1 loci (right, red outline). The top three lanes show C/EBPα and C/EBPβ chromatin immunoprecipitation (ChIP) data from MLL-AF9–transformed cells (Roe et al, 2016); the bottom three lanes show Hoxa9 ChIP data from Zhong et al (2018). (I) Igf1 gene expression (top: RNA-seq data from triplicates) and IGF1 secretion into the growth medium (bottom: ELISA data; ELISA was repeated 3 times yielding similar results) by MLL-ENL–transformed cells. (J) Response of MLL-ENL–transformed dKO cells to the addition of IGF1 to the culture medium as determined by cell counts, viability, and metabolic activity. (K) Comparison of response to IGF1 of WTFL, dKO, and dKO-LAP*-complemented cells.
In this Issue
Subjects
Related Articles
Cited By...
- No citing articles found.