Alterations in the expression or function of histone deacetylases (HDAC) contribute to the development and progression of hematologic malignancies. Consequently, the development and implementation of HDAC inhibitors has proven to be therapeutically beneficial, particularly for hematologic malignancies. However, the molecular mechanisms by which HDAC inhibition (HDACi) induces tumor cell death remain unresolved. Here, we investigated the effects of HDACi in Myc-driven B-cell lymphoma and five other hematopoietic malignancies. We determined that Myc-mediated transcriptional repression of the miR-15 and let-7 families in malignant cells was relieved upon HDACi, and Myc was required for their upregulation. The miR-15 and let-7 families then targeted and downregulated the antiapoptotic genes Bcl-2 and Bcl-xL, respectively, to induce HDACi-mediated apoptosis. Notably, Myc also transcriptionally upregulated these miRNA in untransformed cells, indicating that this Myc-induced miRNA-mediated apoptotic pathway is suppressed in malignant cells, but becomes reactivated upon HDACi. Taken together, our results reveal a previously unknown mechanism by which Myc induces apoptosis independent of the p53 pathway and as a response to HDACi in malignant hematopoietic cells. Cancer Res; 76(3); 736–48. ©2015 AACR.

Aberrant expression or function of histone deacetylases (HDAC) has been implicated in hematopoietic malignancies (1). Although HDAC inhibition (HDACi) induces tumor cell death while leaving normal cells relatively unaffected, the underlying mechanisms behind this remain unclear. Changes in expression of survival genes were observed following HDACi (1); however, whether these alterations resulted from hyper-acetylation of promoters or altered expression of transcriptional mediators was not determined. HDACi also affects DNA replication, likely from the inability to deacetylate histones at replication forks (2, 3).

Myc, an oncogenic transcription factor, is dysregulated in most hematopoietic malignancies (4). However, untransformed cells undergo apoptosis to counter hyper-proliferative signals from Myc dysregulation. Specifically, Myc overexpression activates the p53 tumor-suppressor pathway, eliciting apoptosis (5, 6). Myc-induced apoptosis also occurs independent of p53 through downregulation of anti-apoptotic Bcl-2 and Bcl-xL proteins, by an indirect and unclear mechanism (7–9). Both pathways become inactivated during tumorigenesis (10).

Myc transcriptionally activates or represses numerous genes, regulating many cellular processes (11). Myc represses protein-coding genes by recruiting HDACs and by binding and inhibiting the transcriptional activator Miz-1 (11, 12). Myc also regulates the expression of noncoding RNA, including miRNA that bind mRNA, typically inhibiting translation (13). In malignant cells, Myc represses many miRNA while specifically upregulating others (13, 14). Although Myc-mediated transcriptional activation has been extensively studied, mechanisms of Myc-mediated repression and their contribution to tumorigenesis are less understood.

Here, we describe a previously unknown miRNA-mediated mechanism of Myc-induced apoptosis. We determined that cellular transformation status dictates whether Myc transcriptionally activates or represses the miR-15 and let-7 families that target antiapoptotic Bcl-2 and Bcl-xL, respectively. This apoptotic mechanism was inactivated in transformed hematopoietic cells, but reactivated by HDACi. Our data reveal a general mechanism underlying HDACi-mediated malignant hematopoietic cell death and provide new insight into Myc-induced apoptosis.

Cell lines, transfection, infection, and vectors

Daudi, Ramos, Raji, Su-DHL-6, and NIH3T3 cells were cultured as described by the American Type Culture Collection. OCI-Ly-19 and OCI-Ly-3 cells were cultured in RPMI-1640 containing 10% FBS. P493-6 cells from Dr. Dirk Eick (Helmholtz-Zentrum-Muenchen, Munich, Germany) were cultured as described (15). Tetracycline (0.1 μg/mL; Sigma) was added to cultures of P493-6 cells for 24 hours to turn off MYC expression. Cell lines were obtained between 2001 and 2007 and immediately frozen, so cells were cultured for less than 6 months. Primary murine pre-B cultures were generated and infected with retrovirus as previously described (5; details in Supplementary Experimental Procedures). Eμ-myc lymphoma cells were previously isolated and maintained as published (16). Wild-type or p53−/− murine embryonic fibroblasts (MEF) were generated and cultured as described (6). Fibroblasts were transfected using Lipofectamine 2000 (Invitrogen). Retroviral infections of fibroblasts were performed as previously reported (6). Vector/retrovirus details are in Supplementary Experimental Procedures.

HDAC inhibition and cell growth and apoptosis assays

In vitro experiments utilized 1 μmol/L 4-hydroxytamoxifen (4-OHT) or vehicle (EtOH), and 5 nmol/L depsipeptide (Celgene), 10 μmol/L RGFP966 (Repligen), or vehicle (DMSO). Cell number and/or viability were determined by Trypan-Blue Dye exclusion (triplicate) and proliferation by MTT (Sigma; 570 nm), MTS (Promega; 490 nm), or Alamar Blue (Invitrogen) assays (quadruplicate). Apoptosis was evaluated by flow cytometry following propidium iodide (sub-G1 DNA) or AnnexinV/7-AAD staining.

Mice

For the in vivo lymphoma experiments, 10- to 12-week-old C57Bl/6 (The Jackson Laboratory) mice were subcutaneously injected (one flank) with 4 × 106 Eμ-myc lymphoma cells as previously described (17). Once tumors reached 200 mm3, Depsipeptide (2 mg/kg) or vehicle (DMSO) was intraperitoneally injected. Mice were sacrificed at intervals for tumor evaluation. Studies complied with state and federal guidelines and were approved by the Vanderbilt Institutional Animal Care and Use Committee. Normal B cells were purified from spleens of mice with the IMag Mouse B-Lymphocyte Enrichment Set (BD Biosciences).

Human tissue

Normal human B cells were purified from leukoreduction filters (Red Cross) and deidentified fresh spleens using the IMag Human B-Lymphocyte Enrichment Set (BD Biosciences). Deidentified fresh spleen and frozen lymph nodes were obtained from the Cooperative Human Tissue Network, following an Institutional Review Board approval as nonhuman subject research (#150139).

Western blotting

Whole-cell protein lysates were prepared and Western blotted as reported (6, 18). Fibroblasts were lysed 48 hours after transfection.

Antibodies

Western blotting antibodies were as follows: Bcl-2, Bcl-xL (BD Biosciences); Mcl-1 (Rockland); Bim (22–40; Calbiochem); cleaved caspase-3 (Cell Signaling Technology); Myc, H3K9K14ac (Millipore); H3K56ac, H4K5ac, H3, H4 (Abcam); Bax (N-20), Miz-1 (H-190, Santa Cruz Biotechnology); and β-actin (Sigma). qChIP antibodies: Myc (N-262) and isotype controls (Santa Cruz Biotechnology), RNA polymerase-II (Ser2-phosphorylated; Abcam), and H3K9K14ac (Millipore).

Quantitative chromatin immunoprecipitation

Quantitative chromatin immunoprecipitation (qChIP) was performed as previously described (19). Primer sequences and antibodies in Supplementary Experimental Procedures.

Quantitative real-time PCR

RNA was isolated, cDNA was generated, and Sybr Green (SA-Biosciences) and TaqMan MicroRNA Assays (Applied Biosciences) were used for quantitative real-time PCR (qRT-PCR; triplicate) to measure mRNA and miRNA, respectively, as previously described (16, 20). mRNA and miRNA expressions were normalized to β-actin and RNU6b levels, respectively, and presented as 2−ΔΔCt. Primer sequences in Supplementary Experimental Procedures.

Luciferase assays

NIH3T3 cells were transfected with luciferase reporters, β-galactosidase control plasmid, 50 nmol/L miR-15 or miR-195 miRIDIAN miRNA mimics or control RNA (Dharmacon/ThermoScientificBio), and/or 200 nmol/L miScript Target Protectors (Qiagen). Luciferase and β-galactosidase activity was measured as previously described (21).

Statistical analysis

Student t tests were used to statistically evaluate the data in Figs. 1A and C, 2D–G, 3D and E, 4B–G, 5A–E, 6E, 7A, C, E, and F. Wilcoxon rank-sum tests determined statistical significance for Figs. 2B and C, 3B, and 4A.

Figure 1.

HDACi reduces Bcl-2 and Bcl-xL expression and induces apoptosis. Cells remained untreated (UT) or received depsipeptide (Depsi), RGFP966 (966), or vehicle (DMSO) control. A, following drug administration, proliferation (Alamar Blue; quadruplicate), cell number (triplicate), and viability (triplicate) were determined at the indicated intervals in murine (Eμ-myc, EM330) and human (Daudi) lymphoma cells. B and C, protein and mRNA levels were evaluated by Western blot (B) and qRT-PCR (triplicate; C), respectively. CC3, cleaved caspase-3. mRNA expression was normalized to β-Actin. Error bars are SD for A (*, P < 0.01) and SEM for C (*, P < 0.03); P values determined by comparison to DMSO.

Figure 1.

HDACi reduces Bcl-2 and Bcl-xL expression and induces apoptosis. Cells remained untreated (UT) or received depsipeptide (Depsi), RGFP966 (966), or vehicle (DMSO) control. A, following drug administration, proliferation (Alamar Blue; quadruplicate), cell number (triplicate), and viability (triplicate) were determined at the indicated intervals in murine (Eμ-myc, EM330) and human (Daudi) lymphoma cells. B and C, protein and mRNA levels were evaluated by Western blot (B) and qRT-PCR (triplicate; C), respectively. CC3, cleaved caspase-3. mRNA expression was normalized to β-Actin. Error bars are SD for A (*, P < 0.01) and SEM for C (*, P < 0.03); P values determined by comparison to DMSO.

Close modal
Figure 2.

HDACi reverses Myc-mediated repression of the miR-15 family and let-7a. Lymphoma cells were treated with depsipeptide (Depsi), RGFP966 (966), or vehicle (DMSO) control. A, Western blots in murine (EM330) and human (Daudi) lymphoma cells were performed. B and C, mature miRNA levels were determined by qRT-PCR (triplicate) in human diffuse large B-cell lymphoma (DLBCL), Burkitt's lymphoma, two normal human lymph nodes, and purified B cells from human peripheral blood (PB) and spleen (Sp; B), and six murine Eμ-myc lymphomas, two precancerous Eμ-myc spleens, and purified murine splenic B cells (C). D and E, lymphoma cells were treated with depsipeptide, RGFP966 (966), or vehicle (DMSO) for the time indicated. Mature miRNA (D) or pri-miRNA (E) levels were determined by qRT-PCR (triplicate) in Eμ-myc (EM330) and Daudi lymphoma cells. Small RNA, RNU6b, was used for qRT-PCR normalization for B–E. F, after treatment with depsipeptide or vehicle (DMSO) for 4 hours, ChIP with anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2) or isotype control (IgG) was performed followed by qRT-PCR (triplicate) for the indicated promoters or upstream regions (up; negative controls) in murine (EM330) and human (Daudi) lymphoma cells. Values are relative to their respective IgG control and input DNA. G, EM330 lymphoma cells were infected with either empty retrovirus (vector) or retrovirus encoding the miR-15a/16-1 or let-7a/7f miRNA clusters (14). Western blotting (left) was performed; U, uninfected; CC3, cleaved caspase-3. Viability (center; triplicate) and the percentage of cells with sub-G1 DNA content (right; triplicate) were determined at the indicated intervals. Error bars are SEM for B–F. B and C, *, P < 0.01, lymphoma versus mean of all normals; D–F, *, P < 0.001, depsipeptide or 966 versus DMSO. Error bars are SD for G, *, P < 0.03, compared with empty vector.

Figure 2.

HDACi reverses Myc-mediated repression of the miR-15 family and let-7a. Lymphoma cells were treated with depsipeptide (Depsi), RGFP966 (966), or vehicle (DMSO) control. A, Western blots in murine (EM330) and human (Daudi) lymphoma cells were performed. B and C, mature miRNA levels were determined by qRT-PCR (triplicate) in human diffuse large B-cell lymphoma (DLBCL), Burkitt's lymphoma, two normal human lymph nodes, and purified B cells from human peripheral blood (PB) and spleen (Sp; B), and six murine Eμ-myc lymphomas, two precancerous Eμ-myc spleens, and purified murine splenic B cells (C). D and E, lymphoma cells were treated with depsipeptide, RGFP966 (966), or vehicle (DMSO) for the time indicated. Mature miRNA (D) or pri-miRNA (E) levels were determined by qRT-PCR (triplicate) in Eμ-myc (EM330) and Daudi lymphoma cells. Small RNA, RNU6b, was used for qRT-PCR normalization for B–E. F, after treatment with depsipeptide or vehicle (DMSO) for 4 hours, ChIP with anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2) or isotype control (IgG) was performed followed by qRT-PCR (triplicate) for the indicated promoters or upstream regions (up; negative controls) in murine (EM330) and human (Daudi) lymphoma cells. Values are relative to their respective IgG control and input DNA. G, EM330 lymphoma cells were infected with either empty retrovirus (vector) or retrovirus encoding the miR-15a/16-1 or let-7a/7f miRNA clusters (14). Western blotting (left) was performed; U, uninfected; CC3, cleaved caspase-3. Viability (center; triplicate) and the percentage of cells with sub-G1 DNA content (right; triplicate) were determined at the indicated intervals. Error bars are SEM for B–F. B and C, *, P < 0.01, lymphoma versus mean of all normals; D–F, *, P < 0.001, depsipeptide or 966 versus DMSO. Error bars are SD for G, *, P < 0.03, compared with empty vector.

Close modal
Figure 3.

In vivo, HDACi increases miR-15 family and let-7a levels, inducing lymphoma cell death. C57Bl/6 mice with subcutaneous Eμ-myc lymphoma tumors (EM330) that reached 200 mm3 were treated with depsipeptide (Depsi) or vehicle (DMSO) control (n = 4/group). Tumors were harvested 24 hours later, and levels of the histone acetylation marks (A) and indicated proteins (C) were determined by Western blot. B, miRNA levels were assessed by qRT-PCR (triplicate), and RNU6b was used for qRT-PCR normalization. As a positive control, cultured EM330 lymphoma cells (in vitro) were treated with vehicle (DMSO; −) or depsipeptide (+). Apoptosis was measured by cleaved caspase-3 (CC3), Annexin V positivity (triplicate; D), and propidium iodide staining of sub-G1 (apoptotic) DNA (triplicate; E). Error bars are SEM for B (*, P < 0.001, depsipeptide versus mean of all DMSO controls) and SD for D and E (*, P < 0.03, depsipeptide versus DMSO).

Figure 3.

In vivo, HDACi increases miR-15 family and let-7a levels, inducing lymphoma cell death. C57Bl/6 mice with subcutaneous Eμ-myc lymphoma tumors (EM330) that reached 200 mm3 were treated with depsipeptide (Depsi) or vehicle (DMSO) control (n = 4/group). Tumors were harvested 24 hours later, and levels of the histone acetylation marks (A) and indicated proteins (C) were determined by Western blot. B, miRNA levels were assessed by qRT-PCR (triplicate), and RNU6b was used for qRT-PCR normalization. As a positive control, cultured EM330 lymphoma cells (in vitro) were treated with vehicle (DMSO; −) or depsipeptide (+). Apoptosis was measured by cleaved caspase-3 (CC3), Annexin V positivity (triplicate; D), and propidium iodide staining of sub-G1 (apoptotic) DNA (triplicate; E). Error bars are SEM for B (*, P < 0.001, depsipeptide versus mean of all DMSO controls) and SD for D and E (*, P < 0.03, depsipeptide versus DMSO).

Close modal
Figure 4.

Myc transcriptional activity is necessary to induce the miR-15 family and let-7a in nontransformed cells. Mature miRNA (A–C) and pri-miRNA (D and E) levels were determined by qRT-PCR (triplicate) and are normalized to RNU6b levels. A, precancerous splenocytes from Eμ-myc mice and wild-type (WT) nontransgenic littermates were evaluated (n = 4/group). B–E, MycER or MycΔMBII-ER was activated with 4-OHT or vehicle (EtOH) control at the indicated intervals in primary pre-B cells (B and E) and MEFs (C and D). F and G, following ChIP with anti-Myc, anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2), anti-H3K9K14ac, or isotype controls (IgG), qRT-PCR for the indicated promoters or regions upstream (up) that Myc does not bind (negative controls) was performed (triplicate). F, MycER-expressing p53−/− MEFs received vehicle (EtOH; −) or 4-OHT (+) for 4 hours to induce MycER. G, splenocytes from wild-type (nontransgenic; Tg−) or precancerous Eμ-myc transgenic (Tg+) littermate mice. Values for qChIP are relative to their respective IgG control and input DNA. Error bars, SEM; *, P < 0.001; Eμ-myc versus mean of all WT spleens (A); 4-OHT versus EtOH (B, E, and F); MycER versus MycΔMBII-ER, and (G) Eμ-myc (+) versus wild-type (−) (C and D).

Figure 4.

Myc transcriptional activity is necessary to induce the miR-15 family and let-7a in nontransformed cells. Mature miRNA (A–C) and pri-miRNA (D and E) levels were determined by qRT-PCR (triplicate) and are normalized to RNU6b levels. A, precancerous splenocytes from Eμ-myc mice and wild-type (WT) nontransgenic littermates were evaluated (n = 4/group). B–E, MycER or MycΔMBII-ER was activated with 4-OHT or vehicle (EtOH) control at the indicated intervals in primary pre-B cells (B and E) and MEFs (C and D). F and G, following ChIP with anti-Myc, anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2), anti-H3K9K14ac, or isotype controls (IgG), qRT-PCR for the indicated promoters or regions upstream (up) that Myc does not bind (negative controls) was performed (triplicate). F, MycER-expressing p53−/− MEFs received vehicle (EtOH; −) or 4-OHT (+) for 4 hours to induce MycER. G, splenocytes from wild-type (nontransgenic; Tg−) or precancerous Eμ-myc transgenic (Tg+) littermate mice. Values for qChIP are relative to their respective IgG control and input DNA. Error bars, SEM; *, P < 0.001; Eμ-myc versus mean of all WT spleens (A); 4-OHT versus EtOH (B, E, and F); MycER versus MycΔMBII-ER, and (G) Eμ-myc (+) versus wild-type (−) (C and D).

Close modal
Figure 5.

Myc is required for HDACi-induced transcriptional upregulation of the miR-15 family and let-7a. A, B, D, and E, following ChIP with anti-Myc, anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2), anti-H3K9K14ac, or isotype controls (IgG), qRT-PCR for the indicated promoters or regions upstream (up) that Myc does not bind (negative controls) was performed (triplicate). A, murine (EM330) and human (Daudi) lymphoma cells treated with depsipeptide (Depsi) or vehicle (DMSO) for 4 hours. B, p21 (Myc repression target) and CAD (Myc activation target) were controls (11). Values for qChIP are relative to their respective IgG control and input DNA. C, human P493-6 lymphoma cells containing tetracycline-regulatable MYC exposed to tetracycline (+; MYC Off) for 24 hours or not (−; MYC On) were Western blotted. These cells were also treated for 12 hours with depsipeptide or vehicle (DMSO) control. miRNA were measured by qRT-PCR (triplicate) and normalized to RNU6b levels. D and E, before qChIP analyses, P493-6 cells received tetracycline for 24 hours (MYC Off) or not (MYC On) prior to treatment with depsipeptide or vehicle (DMSO) for 4 hours. Values for qChIP are relative to their respective IgG control and input DNA. Error bars, SEM; *, P < 0.0002 (B), *, P < 0.01 (E), both RNAPII-p-Ser2 and H3K9K14ac versus IgG (B), depsipeptide versus DMSO (C).

Figure 5.

Myc is required for HDACi-induced transcriptional upregulation of the miR-15 family and let-7a. A, B, D, and E, following ChIP with anti-Myc, anti–RNAPII-phosphorylated-Serine2 (RNAPII-p-Ser2), anti-H3K9K14ac, or isotype controls (IgG), qRT-PCR for the indicated promoters or regions upstream (up) that Myc does not bind (negative controls) was performed (triplicate). A, murine (EM330) and human (Daudi) lymphoma cells treated with depsipeptide (Depsi) or vehicle (DMSO) for 4 hours. B, p21 (Myc repression target) and CAD (Myc activation target) were controls (11). Values for qChIP are relative to their respective IgG control and input DNA. C, human P493-6 lymphoma cells containing tetracycline-regulatable MYC exposed to tetracycline (+; MYC Off) for 24 hours or not (−; MYC On) were Western blotted. These cells were also treated for 12 hours with depsipeptide or vehicle (DMSO) control. miRNA were measured by qRT-PCR (triplicate) and normalized to RNU6b levels. D and E, before qChIP analyses, P493-6 cells received tetracycline for 24 hours (MYC Off) or not (MYC On) prior to treatment with depsipeptide or vehicle (DMSO) for 4 hours. Values for qChIP are relative to their respective IgG control and input DNA. Error bars, SEM; *, P < 0.0002 (B), *, P < 0.01 (E), both RNAPII-p-Ser2 and H3K9K14ac versus IgG (B), depsipeptide versus DMSO (C).

Close modal
Figure 6.

Myc transcriptional activity regulates Bcl-2 and Bcl-xL expression independent of Miz-1 and p53. A, Eμ-myc lymphomas (n = 11) and precancerous Eμ-myc purified B cells (n = 2) and spleens (n = 3) were Western blotted. B, precancerous splenocytes from Eμ-myc mice (n = 5) and wild-type (WT) nontransgenic littermates (n = 5) were Western blotted. C, D, and F, at intervals following addition of 4-OHT, WT MycER-expressing MEFs and primary pre-B cells (C) and MycV394D-ER- (D) and MycΔMBII-ER–expressing (F) WT MEFs were harvested and Western blotted. E, miRNA levels were determined by qRT-PCR (triplicate; normalized to RNU6b levels) following MycV394D-ER activation with 4-OHT or vehicle (EtOH). Error bars, SEM; *, P < 0.001, 4-OHT versus EtOH.

Figure 6.

Myc transcriptional activity regulates Bcl-2 and Bcl-xL expression independent of Miz-1 and p53. A, Eμ-myc lymphomas (n = 11) and precancerous Eμ-myc purified B cells (n = 2) and spleens (n = 3) were Western blotted. B, precancerous splenocytes from Eμ-myc mice (n = 5) and wild-type (WT) nontransgenic littermates (n = 5) were Western blotted. C, D, and F, at intervals following addition of 4-OHT, WT MycER-expressing MEFs and primary pre-B cells (C) and MycV394D-ER- (D) and MycΔMBII-ER–expressing (F) WT MEFs were harvested and Western blotted. E, miRNA levels were determined by qRT-PCR (triplicate; normalized to RNU6b levels) following MycV394D-ER activation with 4-OHT or vehicle (EtOH). Error bars, SEM; *, P < 0.001, 4-OHT versus EtOH.

Close modal
Figure 7.

Myc induces the miR-15 family and let-7a that then target Bcl-2 and Bcl-xL. A and C, luciferase expression vectors containing the 3′-UTR of Bcl-2 or Bcl-xL with the wild-type (WT) or a mutated (Mut) miRNA-binding site were transfected into fibroblasts expressing the 4-OHT–inducible MycER or MycΔMBII-ER. An expression vector containing the miR-17 family binding site in the wild-type p21 3′-UTR was a positive control (13). Luciferase activity was measured (triplicate) 48 hours following vehicle (EtOH) control or 4-OHT addition to activate MycER. A β-galactosidase reporter plasmid was cotransfected for normalization. B–F, miR-15 family and let-7a miRNA-binding sites in the Bcl-2 and Bcl-xL 3′-UTR were blocked with site-specific small molecules (TP; Target Protectors). B, wild-type MEFs transfected with either Bcl-2 or Bcl-xL Target Protectors and/or miR-15a mimic were Western blotted. UT, untransfected cells. D–F, p53−/− MEFs, with or without Bcl-2 and/or Bcl-xL Target Protectors, expressing the 4-OHT–inducible MycV394D-ER were Western blotted (D), subjected to MTT assay (E; quadruplicate), or analyzed for Annexin V positivity (triplicate) by flow cytometry at intervals following addition of 4-OHT. CC3, cleaved caspase-3. Error bars, SEM (A and C, *, P < 0.009, 4-OHT vs. EtOH) and SD (E, *, P < 0.02; F, *, P < 0.0001; both TP vs. control).

Figure 7.

Myc induces the miR-15 family and let-7a that then target Bcl-2 and Bcl-xL. A and C, luciferase expression vectors containing the 3′-UTR of Bcl-2 or Bcl-xL with the wild-type (WT) or a mutated (Mut) miRNA-binding site were transfected into fibroblasts expressing the 4-OHT–inducible MycER or MycΔMBII-ER. An expression vector containing the miR-17 family binding site in the wild-type p21 3′-UTR was a positive control (13). Luciferase activity was measured (triplicate) 48 hours following vehicle (EtOH) control or 4-OHT addition to activate MycER. A β-galactosidase reporter plasmid was cotransfected for normalization. B–F, miR-15 family and let-7a miRNA-binding sites in the Bcl-2 and Bcl-xL 3′-UTR were blocked with site-specific small molecules (TP; Target Protectors). B, wild-type MEFs transfected with either Bcl-2 or Bcl-xL Target Protectors and/or miR-15a mimic were Western blotted. UT, untransfected cells. D–F, p53−/− MEFs, with or without Bcl-2 and/or Bcl-xL Target Protectors, expressing the 4-OHT–inducible MycV394D-ER were Western blotted (D), subjected to MTT assay (E; quadruplicate), or analyzed for Annexin V positivity (triplicate) by flow cytometry at intervals following addition of 4-OHT. CC3, cleaved caspase-3. Error bars, SEM (A and C, *, P < 0.009, 4-OHT vs. EtOH) and SD (E, *, P < 0.02; F, *, P < 0.0001; both TP vs. control).

Close modal

HDAC inhibition decreases Bcl-2 and Bcl-xL expression, inducing apoptosis in multiple hematopoietic malignancies

To evaluate the molecular events following HDACi, B-cell lymphomas from Eμ-myc transgenic mice (Myc-driven B-cell lymphoma model; ref. 22) and human Burkitt lymphoma lines were treated with HDAC inhibitors. Depsipeptide (class-I HDACi), RGFP966 (HDAC3i; ref. 23), RGFP233 (HDAC1/2i), RGFP963 (HDAC1/2/3i), and Panobinostat (pan-HDACi) all decreased cell expansion and number (Fig. 1A; Supplementary Fig. S1A and S1B). Depsipeptide also reduced cell expansion in nine other malignant human hematopoietic lines, including acute myeloblastic leukemia (Kasumi), chronic myelogenous leukemia (K562), acute T-cell leukemia (Jurkat, Loucy), cutaneous T-cell lymphoma (Hut-78, MyLa), diffuse large B-cell lymphoma (Su-DHL-6, OCI-Ly-19), and multiple myeloma (H929; Supplementary Fig. S1C). Furthermore, HDACi decreased cell viability (Fig. 1A; Supplementary Fig. S1B) and increased caspase-3 cleavage (Fig. 1B), characteristics of apoptosis.

To identify the molecular determinants of HDACi-mediated apoptosis, we assessed expression of crucial prosurvival proteins. Compared with vehicle control-treated cells, HDACi of murine and human B-cell lymphoma cells decreased protein expression of antiapoptotic Bcl-2 and Bcl-xL, but not Mcl-1 (Fig. 1B; Supplementary Fig. S1D). Moreover, depsipeptide treatment also decreased Bcl-2 and Bcl-xL proteins in nine other malignant hematopoietic cell lines (Supplementary Fig. S1D). Decreased Bcl-2 and Bcl-xL mRNA (Fig. 1C) may explain the reduction in protein. Inhibiting Bcl-2 and/or Bcl-xL reportedly kills malignant hematopoietic cells, including lines we evaluated (24). Thus, HDACi specifically decreased Bcl-2 and Bcl-xL expression, inducing apoptosis.

HDAC inhibition reveals posttranscriptional regulation of Bcl-2 and Bcl-xL

HDACi-induced effects on histone acetylation were evaluated to gain insight into the mechanism responsible for decreasing Bcl-2 and Bcl-xL. Western blotting showed increased global histone acetylation marks associated with active transcription in murine and human lymphoma cells treated with depsipeptide or RGFP966 (Fig. 2A; Supplementary Fig. S2A). Analogous results were obtained in the other nine malignant hematopoietic cell lines tested (Supplementary Fig. S2B).

Histone acetylation is typically associated with gene activation, yet Bcl-2 and Bcl-xL mRNA decreased after HDACi. To investigate this, we examined precision global run-on transcription coupled with massively parallel sequencing (PRO-seq; ref. 25) data for BCL-2 and BCL-XL loci from Daudi cells treated for 4 hours with depsipeptide or vehicle control (S.W. Hiebert, unpublished data). This analysis showed no significant expression changes (increased or decreased) at these loci (Supplementary Fig. S2C), suggesting a posttranscriptional mechanism is likely responsible for the observed changes in Bcl-2 and Bcl-xL mRNA.

Therefore, we assessed expression of the miR-15 family (miR-15a, -16-1, -195, -497) and let-7a, as these miRNAs posttranscriptionally target and negatively regulate Bcl-2 and Bcl-xL expression, respectively, and induce apoptosis in multiple cell types (Supplementary Fig. S3; ref. 26). Consistent with previous reports (14, 27), miR-15 family and let-7a levels were decreased in human diffuse large B-cell lymphoma and Burkitt lymphoma cell lines compared with purified B cells and normal lymph nodes (Fig. 2B). Similarly, B-cell lymphomas from Eμ-myc transgenic mice also had reduced levels of these miRNA compared with levels in precancerous Eμ-myc splenocytes and purified B cells (Fig. 2C).

Notably, HDACi increased miR-15a and miR-195 (representative miR-15 family members) and let-7a in lymphoma cells (Fig. 2D; Supplementary Fig. S4A and S4B). miR-31 levels were unaffected, demonstrating not all miRNAs were upregulated with HDACi. All other malignant hematopoietic lines evaluated showed similar results (Supplementary Fig. S4C), indicating this HDACi-induced effect is not cell-type specific. Furthermore, with depsipeptide dose escalation, miRNA levels increased, indicating a dose response to HDACi (Supplementary Fig. S4D).

To evaluate whether the miR-15 family and let-7a were being transcribed upon HDACi, we measured primary miRNA transcripts of the miR-15a/16-1 and miR-195/497 clusters and let-7a. Following HDACi, these pri-miRNA transcripts increased (Fig. 2E). We next performed qChIP for RNA polymerase-II phosphorylated on serine 2 (RNAPII-p-Ser2), which is indicative of active transcriptional elongation. HDACi resulted in RNAPII-p-Ser2 enrichment at miR-15 family and let-7a promoters in lymphoma cells (Fig. 2F). Enrichment was not observed at regions upstream of these promoters or in vehicle control–treated cells. In addition, there was more RNAPII-p-Ser2 enrichment at miR-15 family and let-7a promoters in depsipeptide-treated lymphoma compared with normal lymphocytes (Supplementary Fig. S4E), indicating increased transcription following HDACi in lymphoma. Thus, HDACi in lymphoma activates transcription of the miR-15 family and let-7a.

To determine whether the HDACi-induced increase in the miR-15 family or let-7a was responsible for lymphoma cell death, we retrovirally expressed the miRNA in two lymphoma lines. Lymphomas expressing the miR-15a/16-1 or let-7a clusters showed reduced Bcl-2 or Bcl-xL protein, cell expansion, and viability, and had increased caspase-3 cleavage, sub-G1 DNA, and Annexin V positivity (Fig. 2G; Supplementary Fig. S4F–S4H). Thus, increased levels of the miR-15 family or let-7a are sufficient to induce apoptosis in lymphomas.

In vivo HDAC inhibition increases miR-15 family and let-7a levels, inducing lymphoma cell death

To extend our investigations in vivo, Eμ-myc lymphoma cells were subcutaneously injected into C57Bl/6 mice. Once tumors reached 200 mm3, mice were administered depsipeptide or vehicle control, and tumors were harvested at intervals. Analogous to our in vitro results, HDACi increased active histone acetylation marks (Fig. 3A) and levels of the miR-15 family and let-7a (Fig. 3B), and decreased Bcl-2 and Bcl-xL proteins (Fig. 3C). Apoptosis was evident by caspase-3 cleavage, Annexin V positivity, and sub-G1 DNA (Fig. 3C–E). These data confirm HDACi activates miR-15 family and let-7a transcription that adversely affect the expression of crucial prosurvival proteins, inducing lymphoma cell apoptosis in vivo.

Myc transcriptionally upregulates the miR-15 family and let-7a in untransformed cells

Myc transcriptionally activates specific miRNA while repressing others in cancer cells (13, 14). To determine whether Myc mediated the repression of the miR-15 family and let-7a and/or their induction following HDACi, we evaluated untransformed and transformed cells with altered Myc levels. Unexpectedly, in contrast with transformed cells, Myc-overexpressing precancerous Eμ-myc spleens had increased miR-15 family and let-7a transcripts compared with wild-type littermate spleens (Fig. 4A). miR-31 levels were analogous between genotypes, indicating Myc overexpression selectively increased specific miRNA in nontransformed cells.

We next assessed primary murine pre-B cells retrovirally expressing MycER, a 4-OHT–inducible Myc (28). Upon MycER activation with 4-OHT, miR-15 family and let-7a levels significantly increased compared with vehicle control-treated pre-B cells (Fig. 4B). This increase was analogous to that of miR-20a, a well-known Myc-induced miRNA (13). Levels of miR-31 were unaffected. Similar results were obtained following MycER activation in untransformed fibroblasts (Supplementary Fig. S5A), indicating this effect also occurs in nonhematopoietic cells. Addition of 4-OHT to non–MycER-expressing pre-B cells or fibroblasts had no effect on miRNA levels (Supplementary Fig. S5B). Therefore, Myc does not repress, but instead induces the miR-15 and let-7 families in untransformed cells.

To determine whether Myc was transcriptionally activating the miR-15 and let-7 families in untransformed cells, we utilized a MycER mutant lacking the Myc-Box-II domain (MycΔMBII-ER) essential for Myc-mediated transcriptional activation (11). Levels of the miR-15 family and let-7a were not induced following MycΔMBII-ER activation in wild-type MEFs, but were when transcriptionally competent MycER was activated (Fig. 4C). When primary transcripts of the miR-15a/16-1 and miR-195/497 clusters and let-7a were evaluated, MycER, but not MycΔMBII-ER, induced their expression in MEFs and pre-B cells (Fig. 4D and E), indicating Myc-mediated transcription was necessary to upregulate these miRNA.

To assess whether Myc was at the miRNA promoters and whether active transcription was occurring, qChIP was performed. qChIP for Myc in MycER-expressing MEFs revealed enrichment at the promoter regions of the miR-15a/16-1 and miR-195/497 clusters and let-7a following 4-OHT–induced MycER activation (Fig. 4F). Importantly, RNAPII-p-Ser2 and H3K9K14ac, indicators of active transcription, were also enriched (Fig. 4F). No enrichment was observed at regions upstream of the miRNA promoters or in vehicle control–treated cells. Similar qChIP results were obtained in vivo when nontransformed precancerous Eμ-myc transgenic spleens were compared with nontransgenic littermate-matched spleens (Fig. 4G), further demonstrating that Myc transcriptionally upregulates these miRNA families in untransformed cells.

Myc is required for HDACi-induced miR-15 family and let-7a transcriptional upregulation

To determine the role of Myc in repressing the miR-15 and let-7 families in malignant cells, qChIP was performed on Myc-overexpressing murine and human lymphoma cells. Myc was enriched at the promoters of both miR-15 family clusters and let-7a in both cell lines, but not at upstream regions (Fig. 5A). ENCODE MYC ChIP-seq data (29) also showed MYC at these promoters in hematopoietic and nonhematopoietic malignancies and nontransformed human cells (Supplementary Fig. S6A). Importantly, Myc was enriched at these same loci in both the presence and absence of HDACi (Fig. 5A). Consistent with Myc-mediated repression of these loci in lymphoma, there was significantly more enrichment of Myc at the promoter regions of the miR-15a/16-1 and miR-195/497 clusters and let-7a in lymphoma cells compared with normal lymphocytes (Supplementary Fig. S6B).

Myc transcriptionally activates CAD and represses p21 (11), consistent with the increase in RNAPII-p-Ser2 and H3K9K14ac at CAD and the lack of enrichment at p21 we observed in the lymphoma cells (Fig. 5B). Neither RNAPII-p-Ser2 nor H3K9K14ac enrichment was detected at promoters of either miR-15 family cluster or let-7a in the lymphomas (Fig. 5B). Collectively, the data indicate Myc activates miR-15 family and let-7a transcription in untransformed cells, whereas it appears to repress their transcription in transformed cells. Moreover, HDACi of B-cell lymphoma induced the miR-15 family and let-7a to levels similar to those in nontransformed precancerous B cells overexpressing Myc (Supplementary Fig. S6C). Furthermore, the derepression of the miR-15 family and let-7a detected in lymphomas following 6 hours of HDACi did not appear to be due to changes in Myc protein, as Myc levels were similar for at least 12 hours after HDACi (Supplementary Fig. S6D). Together, our data suggest HDACi converts Myc from a transcriptional repressor into a transcriptional activator in lymphoma.

To test whether Myc was required for HDACi-induced upregulation of the miR-15 family and let-7a, we utilized the human B-cell lymphoma line, P493-6, that expresses a tetracycline-regulatable MYC (15). With tetracycline present, MYC levels were significantly reduced and HDACi failed to increase miR-15a, miR-195, or let-7a levels (Fig. 5C). Only when MYC was expressed did HDACi increase their expression. Therefore, Myc was required to mediate the HDACi-induced upregulation of the miR-15 family and let-7a. In addition, irrespective of HDACi, when MYC expression was off, levels of the miR-15 and let-7 families were slightly increased compared with when MYC was expressed, providing additional evidence that MYC represses these miRNA in lymphoma.

To further assess the requirements of MYC on miR-15 family and let-7a expression, we performed qChIP with P493-6 cells. When Myc was expressed, it was enriched at the miR-15 family and let-7a promoters (Fig. 5D). Upon HDACi, enrichment of RNAPII-p-Ser2 and H3K9K14ac (marks of active transcription) at miR-15 family and let-7a promoters was only observed in MYC-expressing cells (Fig. 5E). When MYC was not expressed, a slight enrichment of RNAPII-p-Ser2 and H3K9K14ac was detected at the miR-15 family and let-7a promoters, regardless of HDACi (Fig. 5E). These data show that MYC mediated the repression of the miR-15 family and let-7a and was required for their HDACi-induced transcriptional upregulation.

Myc transcriptional activity is required to suppress Bcl-2 and Bcl-xL expression

In untransformed cells, Myc suppressed Bcl-2 and Bcl-xL expression, inducing apoptosis through an indirect, unresolved mechanism (7, 8), purportedly through binding Miz-1 (9). However, Bcl-2 and Bcl-xL proteins are overexpressed in the majority of Eμ-myc B-cell lymphomas and human lymphomas (8). Evaluation of Eμ-myc B-cell lymphomas compared with normal B cells and spleens from precancerous Eμ-myc mice confirmed these results (Fig. 6A). In contrast, precancerous Eμ-myc spleens with increased Myc had reduced Bcl-2 and Bcl-xL proteins compared with nontransgenic littermate-matched spleens (Fig. 6B). Mcl-1 and Bax (proapoptotic Bcl-2 family member) were unaffected. Similarly, in MycER-expressing primary pre–B cells and MEFs, decreases in Bcl-2 and Bcl-xL were detected following MycER activation (Fig. 6C). No change in Mcl-1 or Bax expression was detected, whereas Bim, a proapoptotic Bcl-2 family member upregulated by Myc (30), was increased upon MycER activation (Fig. 6C). Therefore, with elevated Myc, nontransformed cells decrease Bcl-2 and Bcl-xL expression, whereas their expression is increased in transformed cells.

Myc associates with the transcription factor Miz-1 through a motif requiring valine 394, reportedly suppressing Bcl-2 (9). To investigate this, MycER harboring a point mutation (V394D) disrupting the Myc:Miz-1 interaction (MycV394D-ER; Supplementary Fig. S7A; ref. 31) was expressed in wild-type MEFs. Following MycV394D-ER activation with 4-OHT, Bcl-2 and Bcl-xL were suppressed equivalently to cells expressing wild-type MycER, ruling out a Miz-1–mediated mechanism (Fig. 6D). Analogous data were obtained using p53-null MEFs (Supplementary Fig. S7B), consistent with prior results in p53-null myeloid and B cells (7, 8). These data reveal a Miz-1– and p53-independent mechanism for Myc-mediated suppression of Bcl-2 and Bcl-xL expression. Moreover, MycV394D-ER effectively induced miR-15 family and let-7a expression (Fig. 6E), further supporting this miRNA-mediated mechanism of Myc-induced suppression of Bcl-2 and Bcl-xL. In addition, evaluation of ChIP-seq data for Miz-1 in four human and six murine cell lines/tissues (32–35) showed Miz-1 enrichment at promoters of known Miz-1–regulated genes (e.g., VAMP4; Supplementary Fig. S7C). However, Miz-1 was not enriched at the miR-15 family or let-7a promoters in any of the cell lines/tissues evaluated, indicating Miz-1 does not transcriptionally regulate these miRNAs.

To determine whether Myc transcriptional activity is required to decrease Bcl-2 and Bcl-xL expression in normal cells, wild-type MEFs expressing MycER or the transcriptionally inactive MycΔMBII-ER mutant were evaluated. Following 4-OHT addition, decreased Bcl-2 and Bcl-xL and increased Bim expression were only observed in cells expressing wild-type MycER and not MycΔMBII-ER (Fig. 6F). Again, Mcl-1 and Bax expression were unaffected. Therefore, Myc-mediated transcriptional activity is necessary for Bcl-2 and Bcl-xL downregulation.

Myc transcriptionally induces the miR-15 family and let-7a that then target Bcl-2 and Bcl-xL

We reasoned that defining the Myc-induced mechanism of Bcl-2 and Bcl-xL downregulation in normal cells would provide insight into the mechanism behind their downregulation following HDACi. To test whether the reduction in Bcl-2 and Bcl-xL was a direct consequence of Myc-induced upregulation of the miR-15 family and let-7a in untransformed cells, luciferase assays were performed in MycER-expressing fibroblasts with reporters harboring wild-type or mutated miRNA-binding sites in the Bcl-2 or Bcl-xL 3′-untranslated region (3′-UTR). Following MycER activation, luciferase activity decreased in cells containing the reporter with wild-type miR-15 family or let-7a–binding sites in the Bcl-2 or Bcl-xL 3′-UTR, respectively (Fig. 7A). Luciferase activity remained unchanged in cells containing reporters with mutated miR-15 family or let-7a–binding sites or cells expressing the transcriptionally impaired MycΔMBII-ER mutant (Fig. 7A). A reporter containing the miR-17 family binding site of the p21 3′-UTR served as a positive control, as p21 is a validated target of the Myc-regulated miR-17 family (13). These data indicate a novel mechanism where Myc transcriptionally upregulates the miR-15 family and let-7a, which then target the 3′-UTR of Bcl-2 and Bcl-xL, respectively, leading to their downregulation in untransformed cells.

To further validate this mechanism, wild-type MEFs were transfected with modified RNA molecules (Target Protectors) designed to block the miR-15 family or let-7a from binding their specific target sites in the Bcl-2 or Bcl-xL 3′-UTR, respectively. Bcl-2 and Bcl-xL proteins increased in wild-type MEFs transfected with Bcl-2 or Bcl-xL Target Protectors, respectively, as endogenous miR-15 family members and let-7a were unable to bind and inhibit their expression (Fig. 7B). Combining the Bcl-2 Target Protector with miR-15a overexpression, which alone decreased Bcl-2 protein, rescued Bcl-2 protein expression (Fig. 7B). Then, MycER-expressing fibroblasts were transfected with luciferase reporters, as described above, together with the Bcl-2 or Bcl-xL Target Protectors. In the presence of Bcl-2 or Bcl-xL Target Protectors, little, if any, decrease in luciferase activity was detected following MycER activation (Fig. 7C). Next, MEFs expressing MycV394D-ER (unable to interact with Miz-1) were transfected with Bcl-2 or Bcl-xL Target Protectors. MycV394D-ER activation decreased Bcl-2 and Bcl-xL protein expression in the absence of any Target Protector (Fig. 7D). However, levels of Bcl-2 or Bcl-xL were maintained when Target Protectors blocked the miR-15 or let-7 family binding sites, respectively (Fig. 7D). Therefore, downregulation of Bcl-2 and Bcl-xL upon Myc activation in untransformed cells was due to induction of the miR-15 family and let-7a that bind the 3′-UTR of Bcl-2 and Bcl-xL, respectively.

We then tested whether targeting of Bcl-2 and Bcl-xL by the miR-15 family and let-7a contributes to Myc-induced apoptosis, independent of p53. MycV394D-ER was activated in p53-null MEFs under reduced serum conditions with or without Target Protectors. MycV394D-ER–activated MEFs containing Target Protectors had increased cell expansion and reduced cleaved caspase-3 and Annexin V positivity (Fig. 7D–F). Thus, Myc induces the expression of miR-15 family and let-7a independent of p53, leading to apoptosis. Collectively, the data reveal a novel mechanism whereby Myc upregulates the miR-15 family and let-7a that target Bcl-2 and Bcl-xL in untransformed cells to trigger apoptosis, and that this mechanism is reactivated in lymphomas following HDACi.

Although selective killing of tumor cells by HDACi is being clinically tested and multiple effects have been noted, such as altered expression of apoptotic genes and DNA damage (1–3), the mechanism for its effects remains incompletely understood. Here, we show HDACi switches Myc from a repressor to an activator of miRNA that control the expression of Bcl-2 and Bcl-xL, significantly contributing to HDACi-mediated tumor cell death. We identified a mechanism of HDACi-induced apoptosis that occurs in Myc-driven B-cell malignancies and likely contributes to other human cancers. Our data provide evidence of a novel Myc-induced miRNA-mediated mechanism of apoptosis that is present in nontransformed cells, repressed in malignant cells, and reactivated in tumor cells upon HDACi.

It was previously postulated that HDACi kills myeloid leukemia cells through changes in expression of extrinsic apoptotic pathway proteins (36). Others have reported that HDACi causes global changes in gene expression that alter the apoptotic threshold in favor of cancer cell killing (37). Specifically, expression profiling showed that HDACi altered mRNA levels of pro- and antiapoptotic Bcl-2 family members, yielding a proapoptotic signature in malignant cells (37). However, measuring stable pools of individual or multiple mRNA as an indirect readout of transcription would miss the posttranscriptional regulation of gene expression. Our data reveal that increased miRNA transcription, rather than direct transcriptional repression, leads to downregulation of antiapoptotic Bcl-2 and Bcl-xL gene expression. Likewise, other groups have reported HDACi-mediated changes in miRNA expression (38–41). For example, HDACi of a breast cancer line changed the expression of 27 miRNA within 5 hours, including let-7a (38). However, they reported downregulation of let-7a, whereas we detected increased let-7a upon HDACi; this discrepancy may be due to differences in the cell types evaluated or the HDAC inhibitors used. Importantly, our results indicate that HDACi-induced changes in Bcl-2 and Bcl-xL were mediated by Myc. Myc is overexpressed and/or dysregulated in most human malignancies and is essential in cancers driven by other oncogenes, such as mutant Ras (4, 42). Thus, our data, providing a molecular link between Myc, miRNA, and Bcl-2 and Bcl-xL, yield new insights into the transforming action of Myc and suggest how this pathway can be targeted therapeutically. Moreover, our work suggests that the expression of these miRNA may be useful biomarkers for HDACi sensitivity.

Due to previous reports that Myc repressed miRNA in malignant cells (14, 27), we were initially surprised when Myc induced the expression of the miR-15 family and let-7a in untransformed cells. However, Myc overexpression drives cancer cell proliferation, but triggers apoptosis in untransformed cells (4). Myc induces apoptosis by activating the p53 pathway and simultaneously downregulating Bcl-2 and Bcl-xL mRNA expression through an indirect mechanism (7, 8), reportedly involving Miz-1 (9). Myc expression did not change the half-life of Bcl-2 (7); therefore, we suspected a transcriptional or posttranscriptional mechanism. The miR-15 family and let-7a were known to target Bcl-2 and Bcl-xL, respectively, contributing to apoptosis (26), so we investigated whether a connection between Myc, these miRNA, and Bcl-2 and Bcl-xL downregulation existed. Indeed, Myc suppressed the expression of Bcl-2 and Bcl-xL independent of its interaction with Miz-1 and of p53, which can itself transcriptionally repress Bcl-2 and Bcl-xL expression (43, 44). Transcriptionally competent Myc was required for the reduction in Bcl-2 and Bcl-xL and the induction of the miR-15 and let-7 families. Moreover, open chromatin and activated RNA polymerase-II were observed at the miRNA promoters in Myc-overexpressing untransformed cells. Luciferase reporter assays confirmed that Myc induced the miR-15 family and let-7a, which directly targeted Bcl-2 and Bcl-xL 3′-UTRs, respectively. By inhibiting the miR-15 family from binding Bcl-2 and let-7a from binding Bcl-xL, the decrease in Bcl-2 and Bcl-xL following Myc activation was blocked. Combined, our data provide strong evidence that in untransformed cells, Myc induces the miR-15 family and let-7a that then target Bcl-2 and Bcl-xL, respectively, triggering apoptosis. These results reveal a novel miRNA-mediated mechanism of tumor suppression that is activated in normal cells upon Myc dysregulation.

Our data indicate that Myc transcriptionally activates the miR-15 family and let-7a in untransformed cells, while transcriptionally repressing them in transformed cells. In lymphoma cells, Myc was present at miR-15 family and let-7a promoters, which were closed and transcriptionally inactive, indicating that Myc was likely mediating this repression, as had been reported (14). Our data from multiple hematopoietic malignancies indicate HDACs contributed to the repression of both miR-15 family clusters and let-7a, as repression was relieved following HDACi, resulting in transcriptional upregulation of these miRNA, which required Myc. The HDAC3-selective inhibitor RGFP966 induced miR-15a and let-7a to an equal extent as the class-I HDAC inhibitor depsipeptide and the more specific HDAC1/2/3 inhibitor RGFP963 did, suggesting HDAC3 may be the primary HDAC involved in mediating the repression. Consistent with these results, others have reported HDAC3 is specifically involved with repression of miR-29a/b/c and miR-15a/16-1 in B-cell and mantle-cell lymphoma lines, respectively (41, 45). Moreover, Myc is reported to recruit HDAC3 to the promoters of protein-coding genes and miRNA to repress their expression (12, 41, 45). However, RGFP233, the HDAC1/2 inhibitor, also increased levels of miR-15a and let-7a, indicating that they may also contribute to the repression. Although how Myc and HDAC interactions contribute to tumorigenesis remains unresolved, our data suggest that Myc, together with HDACs, alters the epigenome, leading to repression of miRNA and possibly other genes whose expression results in cancer cell apoptosis. HDACi relieves the transcriptional repression of the miR-15 and let-7 families in malignant hematopoietic cells, resulting in transcription of these miRNA, which targeted Bcl-2 and Bcl-xL, killing the cancer cells.

Furthermore, hematopoietic cell lines with either mutant or wild-type p53 showed analogous results, indicating HDACi-induced effects are independent of p53 status. Given the p53 pathway is inactivated and Myc is dysregulated in most human cancers (4, 46), our results identify a new potentially therapeutic avenue to induce apoptosis that capitalizes on Myc and is independent of p53. Of note, overexpression of Bcl-2 and/or Bcl-xL protected from HDACi-induced cell death (47–50), supporting our conclusion that miRNA targeting these genes leads to apoptosis. Our studies have identified a novel mechanism of Myc-induced apoptosis that capitalizes on miRNA to suppress the expression of crucial prosurvival proteins. While this mechanism is inactivated in malignancies through epigenetic alterations involving HDACs, we have shown that it can be reactivated by HDACi. Our current study provides a novel mechanism that underlies HDACi-mediated cell death and offers new insights that should aid in improving cancer therapies.

No potential conflicts of interest were disclosed.

Conception and design: C.M. Adams, S.W. Hiebert, C.M. Eischen

Development of methodology: C.M. Adams

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): C.M. Adams

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): C.M. Adams, S.W. Hiebert, C.M. Eischen

Writing, review, and/or revision of the manuscript: C.M. Adams, S.W. Hiebert, C.M. Eischen

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): C.M. Eischen

Study supervision: C.M. Eischen

The authors thank Pankaj Acharya, Pia Arrate, Robert Duszynski, and Christina Wells for technical assistance, Repligen for RGFP compounds, Dr. Josh Mendell for miRNA retroviruses, the ENCODE Consortium and production laboratories, Eischen lab members for helpful discussions, and Dr. Bill Tansey for critically reading the article.

This work was supported by F31CA165728 (to C.M. Adams), T32GM08554 (C.M. Adams), R01CA178030 (S.W. Hiebert), R01CA148950 (C.M. Eischen), and the NCI Cancer Center Support Grant P30CA068485 utilizing the VANTAGE core.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
West
AC
,
Johnstone
RW
. 
New and emerging HDAC inhibitors for cancer treatment
.
J Clin Invest
2014
;
124
:
30
9
.
2.
Conti
C
,
Leo
E
,
Eichler
GS
,
Sordet
O
,
Martin
MM
,
Fan
A
, et al
Inhibition of histone deacetylase in cancer cells slows down replication forks, activates dormant origins, and induces DNA damage
.
Cancer Res
2010
;
70
:
4470
80
.
3.
Stengel
KR
,
Hiebert
SW
. 
Class I HDACs affect DNA replication, repair, and chromatin structure: implications for cancer therapy
.
Antioxid Redox Signal
2015
;
23
:
51
65
.
4.
Dang
CV
. 
MYC on the path to cancer
.
Cell
2012
;
149
:
22
35
.
5.
Eischen
CM
,
Weber
JD
,
Roussel
MF
,
Sherr
CJ
,
Cleveland
JL
. 
Disruption of the ARF-Mdm2-p53 tumor suppressor pathway in Myc-induced lymphomagenesis
.
Genes Dev
1999
;
13
:
2658
69
.
6.
Zindy
F
,
Eischen
CM
,
Randle
DH
,
Kamijo
T
,
Cleveland
JL
,
Sherr
CJ
, et al
Myc signaling via the ARF tumor suppressor regulates p53-dependent apoptosis and immortalization
.
Genes Dev
1998
;
12
:
2424
33
.
7.
Eischen
CM
,
Packham
G
,
Nip
J
,
Fee
BE
,
Hiebert
SW
,
Zambetti
GP
, et al
Bcl-2 is an apoptotic target suppressed by both c-Myc and E2F-1
.
Oncogene
2001
;
20
:
6983
93
.
8.
Eischen
CM
,
Woo
D
,
Roussel
MF
,
Cleveland
JL
. 
Apoptosis triggered by Myc-induced suppression of Bcl-X(L) or Bcl-2 is bypassed during lymphomagenesis
.
Mol Cell Biol
2001
;
21
:
5063
70
.
9.
Patel
JH
,
McMahon
SB
. 
BCL2 is a downstream effector of MIZ-1 essential for blocking c-MYC-induced apoptosis
.
J Biol Chem
2007
;
282
:
5
13
.
10.
Dang
CV
,
O'Donnell
K A
,
Juopperi
T
. 
The great MYC escape in tumorigenesis
.
Cancer Cell
2005
;
8
:
177
8
.
11.
Eilers
M
,
Eisenman
RN
. 
Myc's broad reach
.
Genes Dev
2008
;
22
:
2755
66
.
12.
Kurland
JF
,
Tansey
WP
. 
Myc-mediated transcriptional repression by recruitment of histone deacetylase
.
Cancer Res
2008
;
68
:
3624
9
.
13.
Bui
TV
,
Mendell
JT
. 
Myc: maestro of MicroRNAs
.
Genes Cancer
2010
;
1
:
568
75
.
14.
Chang
TC
,
Yu
D
,
Lee
YS
,
Wentzel
EA
,
Arking
DE
,
West
KM
, et al
Widespread microRNA repression by Myc contributes to tumorigenesis
.
Nat Genet
2008
;
40
:
43
50
.
15.
Pajic
A
,
Spitkovsky
D
,
Christoph
B
,
Kempkes
B
,
Schuhmacher
M
,
Staege
MS
, et al
Cell cycle activation by c-myc in a Burkitt lymphoma model cell line
.
Int J Cancer
2000
;
87
:
787
93
.
16.
Arrate
PM
,
Vincent
T
,
Odvody
J
,
Kar
R
,
Jones
SN
,
Eischen
CM
. 
MicroRNA biogenesis is required for Myc-induced B cell lymphoma development and survival
.
Cancer Res
2010
;
70
:
6083
92
.
17.
Adams
CM
,
Eischen
CM
. 
Inactivation of p53 is insufficient to allow B cells and B-cell lymphomas to survive without Dicer
.
Cancer Res
2014
;
74
:
3923
34
.
18.
Alt
JR
,
Bouska
A
,
Fernandez
MR
,
Cerny
RL
,
Xiao
H
,
Eischen
CM
. 
Mdm2 binds to Nbs1 at sites of DNA damage and regulates double strand break repair
.
J Biol Chem
2005
;
280
:
18771
81
.
19.
Grieb
BC
,
Gramling
MW
,
Arrate
MP
,
Chen
X
,
Beauparlant
SL
,
Haines
DS
, et al
Oncogenic protein MTBP interacts with MYC to promote tumorigenesis
.
Cancer Res
2014
;
74
:
3591
602
.
20.
Wang
P
,
Lushnikova
T
,
Odvody
J
,
Greiner
TC
,
Jones
SN
,
Eischen
CM
. 
Elevated Mdm2 expression induces chromosomal instability and confers a survival and growth advantage to B cells
.
Oncogene
2008
;
27
:
1590
8
.
21.
McGirt
LY
,
Adams
CM
,
Baerenwald
DA
,
Zwerner
JP
,
Zic
JA
,
Eischen
CM
. 
miR-223 regulates cell growth and targets proto-oncogenes in mycosis fungoides/cutaneous T-cell lymphoma
.
J Invest Dermatol
2014
;
134
:
1101
7
.
22.
Adams
JM
,
Harris
AW
,
Pinkert
CA
,
Corcoran
LM
,
Alexander
WS
,
Cory
S
, et al
The c-myc oncogene driven by immunoglobulin enhancers induces lymphoid malignancy in transgenic mice
.
Nature
1985
;
318
:
533
8
.
23.
Wells
CE
,
Bhaskara
S
,
Stengel
KR
,
Zhao
Y
,
Sirbu
B
,
Chagot
B
, et al
Inhibition of histone deacetylase 3 causes replication stress in cutaneous T cell lymphoma
.
PLoS One
2013
;
8
:
e68915
.
24.
Czabotar
PE
,
Lessene
G
,
Strasser
A
,
Adams
JM
. 
Control of apoptosis by the BCL-2 protein family: implications for physiology and therapy
.
Nat Rev Mol Cell Biol
2014
;
15
:
49
63
.
25.
Kwak
H
,
Fuda
NJ
,
Core
LJ
,
Lis
JT
. 
Precise maps of RNA polymerase reveal how promoters direct initiation and pausing
.
Science
2013
;
339
:
950
3
.
26.
Lima
RT
,
Busacca
S
,
Almeida
GM
,
Gaudino
G
,
Fennell
DA
,
Vasconcelos
MH
. 
MicroRNA regulation of core apoptosis pathways in cancer
.
Eur J Cancer
2011
;
47
:
163
74
.
27.
Lu
J
,
Getz
G
,
Miska
EA
,
Alvarez-Saavedra
E
,
Lamb
J
,
Peck
D
, et al
MicroRNA expression profiles classify human cancers
.
Nature
2005
;
435
:
834
8
.
28.
Littlewood
TD
,
Hancock
DC
,
Danielian
PS
,
Parker
MG
,
Evan
GI
. 
A modified oestrogen receptor ligand-binding domain as an improved switch for the regulation of heterologous proteins
.
Nucleic Acids Res
1995
;
23
:
1686
90
.
29.
Consortium
EP
. 
An integrated encyclopedia of DNA elements in the human genome
.
Nature
2012
;
489
:
57
74
.
30.
Egle
A
,
Harris
AW
,
Bouillet
P
,
Cory
S
. 
Bim is a suppressor of Myc-induced mouse B cell leukemia
.
Proc Natl Acad Sci U S A
2004
;
101
:
6164
9
.
31.
Herold
S
,
Wanzel
M
,
Beuger
V
,
Frohme
C
,
Beul
D
,
Hillukkala
T
, et al
Negative regulation of the mammalian UV response by Myc through association with Miz-1
.
Mol Cell
2002
;
10
:
509
21
.
32.
Peter
S
,
Bultinck
J
,
Myant
K
,
Jaenicke
LA
,
Walz
S
,
Muller
J
, et al
Tumor cell-specific inhibition of MYC function using small molecule inhibitors of the HUWE1 ubiquitin ligase
.
EMBO Mol Med
2014
;
6
:
1525
41
.
33.
Rashkovan
M
,
Vadnais
C
,
Ross
J
,
Gigoux
M
,
Suh
WK
,
Gu
W
, et al
Miz-1 regulates translation of Trp53 via ribosomal protein L22 in cells undergoing V(D)J recombination
.
Proc Natl Acad Sci U S A
2014
;
111
:
E5411
9
.
34.
Walz
S
,
Lorenzin
F
,
Morton
J
,
Wiese
KE
,
von Eyss
B
,
Herold
S
, et al
Activation and repression by oncogenic MYC shape tumour-specific gene expression profiles
.
Nature
2014
;
511
:
483
7
.
35.
Wolf
E
,
Gebhardt
A
,
Kawauchi
D
,
Walz
S
,
von Eyss
B
,
Wagner
N
, et al
Miz1 is required to maintain autophagic flux
.
Nat Commun
2013
;
4
:
2535
.
36.
Insinga
A
,
Monestiroli
S
,
Ronzoni
S
,
Gelmetti
V
,
Marchesi
F
,
Viale
A
, et al
Inhibitors of histone deacetylases induce tumor-selective apoptosis through activation of the death receptor pathway
.
Nat Med
2005
;
11
:
71
6
.
37.
Bolden
JE
,
Shi
W
,
Jankowski
K
,
Kan
CY
,
Cluse
L
,
Martin
BP
, et al
HDAC inhibitors induce tumor-cell-selective pro-apoptotic transcriptional responses
.
Cell Death Dis
2013
;
4
:
e519
.
38.
Scott
GK
,
Mattie
MD
,
Berger
CE
,
Benz
SC
,
Benz
CC
. 
Rapid alteration of microRNA levels by histone deacetylase inhibition
.
Cancer Res
2006
;
66
:
1277
81
.
39.
Cho
JH
,
Dimri
M
,
Dimri
GP
. 
MicroRNA-31 is a transcriptional target of histone deacetylase inhibitors and a regulator of cellular senescence
.
J Biol Chem
2015
;
290
:
10555
67
.
40.
Majid
S
,
Dar
AA
,
Saini
S
,
Shahryari
V
,
Arora
S
,
Zaman
MS
, et al
miRNA-34b inhibits prostate cancer through demethylation, active chromatin modifications, and AKT pathways
.
Clin Cancer Res
2013
;
19
:
73
84
.
41.
Zhang
X
,
Zhao
X
,
Fiskus
W
,
Lin
J
,
Lwin
T
,
Rao
R
, et al
Coordinated silencing of MYC-mediated miR-29 by HDAC3 and EZH2 as a therapeutic target of histone modification in aggressive B-Cell lymphomas
.
Cancer Cell
2012
;
22
:
506
23
.
42.
Soucek
L
,
Whitfield
J
,
Martins
CP
,
Finch
AJ
,
Murphy
DJ
,
Sodir
NM
, et al
Modelling Myc inhibition as a cancer therapy
.
Nature
2008
;
455
:
679
83
.
43.
Haldar
S
,
Negrini
M
,
Monne
M
,
Sabbioni
S
,
Croce
CM
. 
Down-regulation of bcl-2 by p53 in breast cancer cells
.
Cancer Res
1994
;
54
:
2095
7
.
44.
Sugars
KL
,
Budhram-Mahadeo
V
,
Packham
G
,
Latchman
DS
. 
A minimal Bcl-x promoter is activated by Brn-3a and repressed by p53
.
Nucleic Acids Res
2001
;
29
:
4530
40
.
45.
Zhang
X
,
Chen
X
,
Lin
J
,
Lwin
T
,
Wright
G
,
Moscinski
LC
, et al
Myc represses miR-15a/miR-16-1 expression through recruitment of HDAC3 in mantle cell and other non-Hodgkin B-cell lymphomas
.
Oncogene
2012
;
31
:
3002
8
.
46.
Olivier
M
,
Hollstein
M
,
Hainaut
P
. 
TP53 mutations in human cancers: origins, consequences, and clinical use
.
Cold Spring Harb Perspect Biol
2010
;
2
:
a001008
.
47.
Ellis
L
,
Bots
M
,
Lindemann
RK
,
Bolden
JE
,
Newbold
A
,
Cluse
LA
, et al
The histone deacetylase inhibitors LAQ824 and LBH589 do not require death receptor signaling or a functional apoptosome to mediate tumor cell death or therapeutic efficacy
.
Blood
2009
;
114
:
380
93
.
48.
Lindemann
RK
,
Newbold
A
,
Whitecross
KF
,
Cluse
LA
,
Frew
AJ
,
Ellis
L
, et al
Analysis of the apoptotic and therapeutic activities of histone deacetylase inhibitors by using a mouse model of B cell lymphoma
.
Proc Natl Acad Sci U S A
2007
;
104
:
8071
6
.
49.
Thompson
RC
,
Vardinogiannis
I
,
Gilmore
TD
. 
The sensitivity of diffuse large B-cell lymphoma cell lines to histone deacetylase inhibitor-induced apoptosis is modulated by BCL-2 family protein activity
.
PLoS One
2013
;
8
:
e62822
.
50.
Whitecross
KF
,
Alsop
AE
,
Cluse
LA
,
Wiegmans
A
,
Banks
KM
,
Coomans
C
, et al
Defining the target specificity of ABT-737 and synergistic antitumor activities in combination with histone deacetylase inhibitors
.
Blood
2009
;
113
:
1982
91
.