Epigenetic roles of MLL oncoproteins are dependent on NF-κB - PubMed (original) (raw)

Epigenetic roles of MLL oncoproteins are dependent on NF-κB

Hsu-Ping Kuo et al. Cancer Cell. 2013.

Abstract

MLL fusion proteins in leukemia induce aberrant transcriptional elongation and associated chromatin perturbations; however, the upstream signaling pathways and activators that recruit or retain MLL oncoproteins at initiated promoters are unknown. Through functional and comparative genomic studies, we identified an essential role for NF-κB signaling in MLL leukemia. Suppression of NF-κB led to robust antileukemia effects that phenocopied loss of functional MLL oncoprotein or associated epigenetic cofactors. The NF-κB subunit RELA occupies promoter regions of crucial MLL target genes and sustains the MLL-dependent leukemia stem cell program. IKK/NF-κB signaling is required for wild-type and fusion MLL protein retention and maintenance of associated histone modifications, providing a molecular rationale for enhanced efficacy in therapeutic targeting of this pathway in MLL leukemias.

Copyright © 2013 Elsevier Inc. All rights reserved.

PubMed Disclaimer

Figures

Figure 1

Figure 1. Enrichment of NF- κB signaling in MLL leukemia cells

(A) Single cell-based shRNA screening was used to identify the effects of kinase andphosphatase knockdowns on the growth of mouse myeloid cells transduced by MLL-AF9. Combined results from two independent replicates are expressed as the relativecell number compared to cells transduced with control shRNA. Potential candidatesassociated with NF-κB signaling are indicated. (B) GSEA analyses demonstrate that expression of genes associated with the Toll-likereceptor signaling pathway is enriched in leukemic BM cells from mice with AMLinduced by MLL-AF1p (left) and MLL-AF10 (right) compared to normal BM. Thenormalized enrichment scores (NES) are based on analysis of a public dataset (GSE13796). (C) Aberrantly methylated NF-κB gene networks in two MLL-associated epigeneticallydefined human AML clusters. Genes with DNA hypomethylation compared with normalCD34+ cells are shown in green, whereas hypermethylated genes appear in red. (D) Comparison of _HOXA9_-correlated and _NF-κB_-correlated (correlation > 0.5) gene expression in human MLL acute myeloid leukemias showed 121 unique genes whose expression overlapped (p = 1×10-16, calculated using R package: SAGx_1.32.0 under R version 2.15.3). A similar analysis in NPM1c acute myeloid leukemias showed no overlap. See also Figure S1 and Tables S1 and S2.

Figure 2

Figure 2. Differential sensitivity of human MLL leukemia cells to IKK/NF- κB signaling

(A) The growth of human myeloid leukemia cell lines was assessed after 4 days culture in the absence or presence of 2 μM IKK inhibitor VII. Results are expressed as the relative cell number compared to vehicle treated cells. (B and C) MV4;11 human leukemia cells were transduced with lentiviral vectors expressing the indicated shRNAs. Relative RELA transcript and protein levels were measured by qRT-PCR and western blot analysis, respectively (B). Cell numbers (C) were enumerated at 3 days, and expressed relative to cells transduced with control shRNA. (D) Human myeloid leukemia cell lines were serum starved (0.1% FBS) overnight, then stimulated with 10 μg/ml LPS for 30 min with or without IKK inhibitor VII pretreatment, and analyzed by phospho-flow cytometry with antibodies specific to p65 RELA phosphorylated at S529 or S536. Representative results are shown from three independent experiments. (E) Human leukemia cells were treated as (D). Nuclear (N) and cytoplasmic (C) fractions were collected and detected by antibodies specific to RELA, histone H3 (nuclear control), and α-Tubulin (cytoplasmic control). All error bars represent SD of triplicate analyses. See also Figure S2.

Figure 3

Figure 3. Inhibition of IKK signaling suppresses cell growth and colony formation of mouse MLL leukemia cells

(A and B) Mouse myeloid progenitors transformed by the indicated oncogenes (bone marrow c-kit+ cells transduced with MLL oncogene) were plated in methylcellulose medium with different concentrations of IKK inhibitor VII (A) or IV (B). Colonies were enumerated after 5 days and expressed relative to cells treated with vehicle (DMSO) alone. (C and D) Mouse myeloid progenitors transformed by MLL-AF9 were stably transduced with lentiviral vectors expressing control shRNA or shRNAs targeting Ikkα, Ikkβ or Ikkγ. Protein (C) and relative mRNA (D) levels were determined by western blot analysis and qRT-PCR, respectively. (E and F) Mouse MLL-AF9 transduced cells were treated as in panels C and D. Viable cell numbers at day 2 (E) and colony numbers at day 5 (F) were enumerated and expressed relative to the number obtained with control shRNA transduced cells. (G) Mouse MLL-AF10 leukemia cells and bone marrow c-kit+ cells (normal hematopoietic progenitors) were plated in methylcellulose medium with different concentrations of IKK inhibitor VII. Colonies were enumerated after 5 days and expressed relative to cells treated with vehicle (DMSO) alone. (H) The growth of mouse MLL-AF10 leukemia cells and c-kit+ cells was assessed after 2 days culture in the absence or presence of the indicated concentrations of IKK inhibitor VII. Results are expressed as relative cell number compared to vehicle treated cells. All error bars represent SD of triplicate analyses. See also Figure S3.

Figure 4

Figure 4. RELA is required for MLL leukemia development

(A-C) Mouse MLL-AF9 cells were transduced with lentiviral vectors expressing control or Rela shRNAs. Protein levels of Rela were detected by western blot analysis (A). Cell numbers (B) and colony numbers (C) were enumerated after 2 and 5 days, respectively, and expressed relative to the numbers obtained with control shRNA transduced cells. (D and E) Mouse MLL-AF9 cells were transduced with Rela over-expression or control vectors. Cell numbers (D) and colony numbers (E) were enumerated after 3 and 5 days, respectively, and expressed relative to the numbers obtained with control vector transduced cells. (F) Hematopoietic progenitors obtained from mouse fetal livers (E13.5) of wild-type (Rela+/+) or Rela knockout (Rela-/-) embryos were transduced with the indicated oncogenes and used for serial myeloid replating assays. Representative results from two independent replicates through four rounds of serial methylcellulose culture are shown as colony number per 3000 cells. (G) Survival curves are shown for cohorts of mice transplanted with mouse MLL-AF10 leukemia cells (5 × 105) transduced with control or Rela shRNAs (n = 5 each cohort). Acute leukemia was confirmed by peripheral blood leukocyte count and necropsy. Log-rank Test was used for statistical analysis (p = 0.006). (H) Survival curves are shown for mice transplanted with mouse MLL-AF6 transformed cells (1 × 106) co-transduced with Rela or control vectors (n = 5 each cohort). Log-rank Test was used for statistical analysis (p = 0.002). All error bars represent SD of triplicate analyses. See also Figure S4.

Figure 5

Figure 5. IKK inhibition decreases proliferation, increases apoptosis and induces differentiation of MLL leukemia cells

(A) GSEA plot shows downregulation of cell cycle process related genes in mouse MLL-AF10 leukemia cells treated for 24 hr with IKK inhibitor versus vehicle treated cells. (B) Mouse MLL-AF10 leukemia cells were cultured in the presence of 2 μM IKK inhibitor IV or 0.5 μM IKK inhibitor VII for 2 days, and BrdU incorporation was quantified by flow cytometry analysis. (C) GSEA plot shows upregulation of apoptosis related genes in mouse MLL-AF10leukemia cells treated for 24 hr with IKK inhibitor versus vehicle treated cells. (D) Mouse MLL-AF10 leukemia cells were cultured in the presence of 1 μM IKK inhibitor IV or 0.5 μM IKK inhibitor VII for 3 days. The annexin-V positive and PI negative populations constitute early apoptotic cells. (E) GSEA plots show upregulation of monocyte or granulocyte fingerprint genes in mouse MLL-AF10 leukemia cells treated for 24 hr with IKK inhibitor versus vehicle treated cells. (F) Light microscopy of May-Grunwald/Giemsa-stained mouse MLL-AF10 leukemia cells after 2 days of IKK inhibitor IV treatment (2 μM). (G) Quantification of leukemia cell populations with indicated morphological features after 2 days of IKK inhibitor IV treatment (2 μM). (H) Flow cytometry analysis of Mac-1 and Gr-1 surface expression by mouse MLL-AF10 leukemia cells after 2 days of 2 μM IKK inhibitor IV treatment. See also Figure S5 and Table S3.

Figure 6

Figure 6. IKK inhibition reduces the LSC population

(A-C) GSEA plots show downregulation of MLL LSC maintenance signature genes (A), core ESC-like gene module (B), and MYC core module genes (C) in mouse MLL-AF10 leukemia cells treated for 24 hr with IKK inhibitor versus vehicle treated cells. (D) GSEA plots show downregulation of poor prognosis AML genes and upregulation of good prognosis AML genes in mouse MLL-AF10 leukemia cells treated with IKK inhibitor versus vehicle treated cells. (E) Limit-dilution analyses show the estimated cell number of transplanted mouse MLL-AF10 leukemia cells required to initiate AML in sublethally irradiated recipient mice (n = 3 for each cell dose). MLL-AF10 leukemia cells were pretreated with IKK inhibitor IV (2 μM) or control vehicle for 2 days. Viable cells used for transplantation were confirmed by Trypan Blue staining. Mice were followed for 190 days, with the longest disease latencies being 84 days. See also Figure S6 and Table S4.

Figure 7

Figure 7. IKK/NF- κB signaling regulates Meis1 and _Hoxa9_gene expression

(A) GSEA plot shows downregulation of MLL-AF9 regulated genes in mouse MLL-AF10leukemia cells treated with IKK inhibitor versus vehicle treated cells. (B) GSEA plots show downregulation of HOXA9 and MEIS1 upregulated target genes (left) and upregulation of HOXA9 and MEIS1 downregulated target genes (right) in mouse MLL-AF10 leukemia cells treated with IKK inhibitor versus vehicle treated cells. (C) Mouse MLL-AF10 leukemia cells were cultured in the absence or presence of IKK inhibitors (IV or VII) for 2 days. Meis1 or Hoxa9 transcripts were quantified by qRT-PCR, and expressed relative to vehicle treated cells. (D) Mouse MLL-AF9 transformed cells transduced with lentiviral vectors expressing control or Rela shRNAs were assessed for Rela, Meis1 or Hoxa9 transcript levels by qRT-PCR. Results are displayed relative to control shRNA-transduced cells. (E) Mouse MLL-AF9 transformed cells transduced with retroviral vector expressing Rela or control vector were assessed for Rela, Meis1 or Hoxa9 transcript levels by qRT-PCR. Results are displayed relative to control vector-transduced cells. (F) Mouse HOXA9/MEIS1 leukemia cells were plated in methylcellulose medium with different concentrations of IKK inhibitor VII. Colonies were enumerated after 5 days and expressed relative to cells treated with vehicle (DMSO) alone. All error bars represent SD of triplicate analyses. See also Figure S7 and Table S5.

Figure 8

Figure 8. Promoter occupancies of MLL wild-type and oncoprotein are dependent on IKK/NF- κB signaling

(A) ChIP was performed on mouse MLL-AF10 leukemia cells with antibodies against Rela, HA (MLL-AF10), and control IgG. Genomic regions amplified by qPCR are indicated relative to the transcription start site (TSS) on Meis1 and Hoxa9 genes. Ey-globin and H4 promoter primers were used for negative controls. (B) Promoter sequence of Meis1 and Hoxa9. NF-κB consensus sequence is indicated in capital letters. (C) Mouse MLL-AF10 leukemia cells were cultured in the absence or presence of IKK inhibitors (IV or VII) for 2 days. ChIP was performed using antibodies against Rela, HA (MLL-AF10), and control IgG. (D-F) Mouse MLL-AF10 leukemia cells were cultured in the absence or presence of IKK inhibitor VII for 2 days. ChIP was performed using antibodies against H3K79me2 and control IgG (D), Mll-c term (E), and H3K4me3 (F) and qPCR primers amplifying the indicated genomic regions. All error bars represent SD of triplicate analyses. See also Figure S8.

Comment in

Similar articles

Cited by

References

    1. Accordi B, Galla L, Milani G, Curtarello M, Serafin V, Lissandron V, Viola G, Te Kronnie G, De Maria R, Petricoin EF, et al. AMPK inhibition enhances apoptosis in MLL-rearranged pediatric B-acute lymphoblastic leukemia cells. Leukemia. 2013;27:1019–1027. - PubMed
    1. Ben-Porath I, Thomson MW, Carey VJ, Ge R, Bell GW, Regev A, Weinberg RA. An embryonic stem cell-like gene expression signature in poorly differentiated aggressive human tumors. Nat Genet. 2008;40:499–507. - PMC - PubMed
    1. Bernt KM, Zhu N, Sinha AU, Vempati S, Faber J, Krivtsov AV, Feng Z, Punt N, Daigle A, Bullinger L, et al. MLL-rearranged leukemia is dependent on aberrant H3K79 methylation by DOT1. L Cancer Cell. 2011;20:66–78. - PMC - PubMed
    1. Birkenkamp KU, Geugien M, Schepers H, Westra J, Lemmink HH, Vellenga E. Constitutive NF-kappaB DNA-binding activity in AML is frequently mediated by a Ras/PI3-K/PKB-dependent pathway. Leukemia. 2004;18:103–112. - PubMed
    1. Bitoun E, Oliver PL, Davies KE. The mixed-lineage leukemia fusion partner AF4 stimulates RNA polymerase II transcriptional elongation and mediates coordinated chromatin remodeling. Hum Mol Genet. 2007;16:92–106. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources