For maximal oncogenic activity, cellular MYC protein levels need to be tightly controlled so that they do not induce apoptosis. Here, we show how ubiquitin ligase UBR5 functions as a molecular rheostat to prevent excess accumulation of MYC protein. UBR5 ubiquitinates MYC and its effects on MYC protein stability are independent of FBXW7. Silencing of endogenous UBR5 induced MYC protein expression and regulated MYC target genes. Consistent with the tumor suppressor function of UBR5 (HYD) in Drosophila, HYD suppressed dMYC-dependent overgrowth of wing imaginal discs. In contrast, in cancer cells, UBR5 suppressed MYC-dependent priming to therapy-induced apoptosis. Of direct cancer relevance, MYC and UBR5 genes were coamplified in MYC-driven human cancers. Functionally, UBR5 suppressed MYC-mediated apoptosis in p53-mutant breast cancer cells with UBR5/MYC coamplification. Furthermore, single-cell immunofluorescence analysis demonstrated reciprocal expression of UBR5 and MYC in human basal-type breast cancer tissues. In summary, UBR5 is a novel MYC ubiquitin ligase and an endogenous rheostat for MYC activity. In MYC-amplified, and p53-mutant breast cancer cells, UBR5 has an important role in suppressing MYC-mediated apoptosis priming and in protection from drug-induced apoptosis.

Significance:

These findings identify UBR5 as a novel MYC regulator, the inactivation of which could be very important for understanding of MYC dysregulation on cancer cells.

Emerging notion of overall poor concordance between gene amplifications and corresponding protein expression levels in cancer (1) highlights the need for better understanding of how protein expression levels are regulated at posttranslational levels. The transcription factor MYC regulates numerous physiologic and pathologic processes. Critically, distinct protein expression levels dictate MYC's biological output in vitro and in vivo (2–5). MYC protein levels are tightly regulated posttranslationally by phosphorylation, and by ubiquitin-mediated degradation (6–10). Especially in cancers, MYC is a classic example of an oncoprotein, whose overexpression at protein level do not match neither with the extent of mRNA overexpression nor with the frequency of genetic amplifications (7, 11).

In addition to cell culture and mouse models (3, 4, 8, 12, 13), the role of MYC has been widely studied in Drosophila, where the conserved homolog of MYC, dMYC regulates tissue growth and animal size (14, 15). In particular, overexpression of dMYC induces growth of cells of the imaginal discs, which are the organs giving rise to wings in adult fly (14). Importantly, dMYC can complement mammalian MYC in vivo (13) implying high degree of functional conservation. In humans, MYC is a master regulator of malignant growth. However, MYC protein levels exceeding the optimal proliferation promoting levels primes both normal and cancer cells to apoptosis (2, 3, 5, 12). The principle of MYC-mediated apoptosis priming was originally revealed by findings showing that highly overexpressed transgenic MYC initiates both proliferation and apoptosis programs in pancreatic islet cells. However, MYC overexpression was capable to drive tumorigenesis only in the context of efficient apoptosis suppression (12). More recently, the in vivo importance of MYC expression levels in defining the balance between proliferation and apoptosis was validated by allelic series of inducible MYC expression (3). In addition to spontaneous apoptosis, high MYC levels prime tumor cells to apoptosis induction by drugs that interfere with replication and cell division, such as topoisomerase inhibitors or taxanes (2, 16, 17). Indeed, MYC is an important determinant of in vivo cell survival both during development and in response to cancer therapy (4). However, how MYC protein levels are controlled endogenously to maintain an optimal MYC balance is incompletely understood.

UBR5 (Ubiquitin protein ligase E3 component n-recognin 5) is an evolutionary conserved E3 ubiquitin ligase that destabilizes proteins with N-terminal recognition sequences exposed by proteolytic cleavage (18, 19). Recently, UBR5 has been shown to target substrates also through other recognition mechanisms than N-terminal recognition (20). UBR5 is essential for mammalian development (21), and has been linked to both protumorigenic and tumor-suppressing activities (19, 21–23). However, the mechanism by which UBR5 would promote tumor growth are unclear. It is also unclear how UBR5 may have protumorigenic role in human cancer cells (19, 22–24), whereas it has tumor suppressor activity in Drosophila (25, 26).

Here, we have identified UBR5 as an FBXW7-independent ubiquitin ligase and rheostat for MYC. Consistent with tumor-suppressive activity of UBR5 in Drosophila, loss of UBR5 induced overgrowth of Drosophila wing imaginal disc epithelium in a dMyc-dependent manner. On the contrary, UBR5-mediated MYC suppression was found to protect the cancer cells from apoptosis priming. Furthermore, our data reveal genetic coamplification of UBR5 and MYC in several solid human cancers and demonstrate functional relevance of reciprocal protein-level regulation between these two cancer genes in defining the apoptosis sensitivity of p53-mutant breast cancer cells.

Cell culture and transfection

HCC38, HCC1937, and HCC1395 cell lines were obtained from ATCC. Osteosarcoma-MYC-off cell line (27) was a generous gift from Professor Dean Felsher (Stanford University, Stanford, CA). HeLa, U2OS, T98G, HEK293, Osteosarcoma-MYC-off, MDA-MB-468, and MDA-MB-231 cell lines were cultured in DMEM (Sigma). HCC38, HCC1937, and IMR-32 cell lines were cultured in RPMI (ATCC-modified vision, Thermo Fisher Scientific). MCF10A cells were cultured as described previously (8). The cell lines were authenticated by ATCC, European Collection of Authenticated Cell Cultures, or at Institut für Rechtsmedizin, Universität Würzburg (Würzburg, Germany). All the cell lines were tested negative for Mycoplasma during the period of this study. Drosophila S2 cells were cultured in Schneider Drosophila Medium (Thermo Fisher Scientific). All growth mediums were supplemented with 10% heat-inactivated FBS (Gibco), 2 mmol/L l-glutamine, and penicillin (50 U/mL)/streptomycin (50 mg/mL).

GFP-UBR5, GFP-UBR5-ΔHECT, Flag-UBR5, and Flag-UBR5-ΔHECT were kind gifts from Darren Saunders & Charles Watts (28, 29). V5-MYC, V5-MYCT58A and have been described previously (10). Flag-MYC plasmids and HA-MYC (1–262) were from Prof. Bruno Amati (European Institute of Oncology, Milan, Italy). HA-MYC-WT and HA-MYC mutants were kind gifts from Prof. William P. Tansey (Vanderbilt University School of Medicine, Nashville, TN). HA-ubiqiuitin was a kind gift from Prof. Lea Sistonen (Åbo Akademi University, Turku, Finland). Plasmids were transfected with Lipofectamine 2000 Transfection Reagent (Thermo Fisher Scientific) according to the manufacturer's instructions. After 48-hour transfection, cell lysate was collected. Small interfering RNA (siRNA) transfections were performed with Oligofectamine Transfection Reagent (Thermo Fisher Scientific) following to the manufacturer's protocol. Three days after transfections, cells were harvested for analysis. siRNA target sequences are in supplementary table 1. For double-stranded RNA (dsRNA)-mediated RNAi in Drosophila S2 cells, the DNA fragment of hyd was amplified with primers flanked by T7 promoter, and dsRNA was produced using the TranscriptAid T7 High Yield Transcription Kit (Thermo Fisher Scientific). To knock down Hyd, S2 cells were cultured with 5 μg/mL dsRNA for 120 hours.

Immunoblotting and immunoprecipitation

The following antibodies were used for Western blot analysis: UBR5 (sc-515494, Santa Cruz Biotechnology), MYC (ab32072, Abcam), cleaved PARP (ab32064, Abcam), BIM (2933, Cell Signaling Technology), vinculin (sc-25336, Santa Cruz Biotechnology), GAPDH (5G4-6C5, HyTest Ltd), HA (3724, Cell Signaling Technology), Flag (F3165, Sigma), GFP (sc-9996, Santa Cruz Biotechnology), GST (sc-138, Santa Cruz Biotechnology), V5 (R960-25, Invitrogen), Histone H3 (ab1791, Abcam), and Lys48-specific antibody (05-1307, Millipore). Secondary antibodies are from Dako (P0447 and P0399). Densitometric analysis of the blots was performed using ImageJ. For immunoprecipitation, the cells were lysed in IP buffer (150 mmol/L NaCl, 1% NP-40, 50 mmol/L Tris pH 8.0) supplemented with protease and phosphatase inhibitors. Cell lysates were centrifuged and immunoprecipitated with the indicated antibodies for 2 hours at 4°C, following by adding Protein A-Sepharose or Protein G-Sepharose beads (Sigma) overnight. The beads were washed with IP buffer, boiled in SDS sample buffer, and analyzed by immunoblotting. MYC (N-262, sc-764) and V5 agarose (A7345, Sigma) were used for immunoprecipitation.

To immunoprecipitate dMYC, lysates were incubated for 2 hours with anti-dMYC beads. Anti-dMyc beads were made from rabbit anti-dMyc antibody (Santa Cruz Biotechnology) and 50% protein A Sepharose beads (Amersham). Immunoprecipitates were washed five times with NP-40 lysis buffer, boiled in 2× SDS sample buffer, and resolved on 8% SDS-PAGE for analyzing via Western blotting. Immunoprecipitated Myc was detected using mouse anti-Myc antibody (a kind gift from Dr. Peter Gallant, Universität Zürich, Zürich, Switzerland).

Ubiquitination assays

For ubiquitination assay in denaturating condition, HEK293 cells transfected with relevant plasmids were lysed in a buffer containing 6 mol/L guanidinium-HCl, 0.1 mol/L Na2HPO4, 0.1 mol/L NaH2PO4, 0.01 mol/L Tris-HCl (pH 8), 5 mmol/L imidazole, and 10 mmol/L β-mercaptoethanol (β-ME). The lysates were incubated with Ni-NTA agarose beads (Qiagen) at 4°C overnight. The beads were washed once with a buffer containing 6 mol/L guanidinium-HCl, 0.1 mol/L Na2HPO4, 0.1 mol/L NaH2PO4, 0.01 mol/L Tris-HCl (pH 8), and 10 mmol/L β-ME, and twice with a buffer containing 8 mol/L urea, 0.1 mol/L Na2HPO4, 0.1 mol/L NaH2PO4, 0.01 mol/L Tris-HCl (pH 8), 10 mmol/L β-ME, and 0.1% Triton X-100. His-Ub–conjugated proteins were eluted using a buffer containing 200 mmol/L imidazole, 0.15 mol/L Tris (pH 6.7), 30% glycerol, 0.72 mol/L β-ME, and 5% SDS. In vitro ubiquitination assays were performed in reactions containing 2 μg ubiquitin (Boston Biochem), 2 mmol/L ATP, 80 ng hUBE1 (Boston Biochem), 2 μg UbcH5c (Boston Biochem), recombinant GST-MYC, and GFP-UBR5 purified from HEK293 cells, in a buffer with 800 mmol/L Tris-HCl (pH 7.5), 200 mmol/L MgCl2, and 12 mmol/L DTT. The reactions were incubated at 37°C for 90 minutes, followed by separation in Mini-PROTEAN TGX precast gels (Bio-Rad). The samples were analyzed by Western blotting using GFP and GST antibodies.

Immunofluorescence staining

Cells plated on chambered coverslip (80826, Ibidi) were transfected with scrambled siRNA or UBR5 siRNA. After 72-hour transfection, the cells were fixed with 4% paraformaldehyde 15 minutes under room temperature, and then cells were permeabilized with 0.5% Triton X-100 in PBS on ice for 5 minutes. Next, the cells were blocked by 10% normal goat serum (ab7481, Abcam) diluted in PBS for 30 minutes, and followed by incubating the primary antibodies anti-UBR5 (sc-515494, Santa Cruz Biotechnology) and anti-MYC (ab32072, Abcam) overnight at 4°C. Subsequently, cells were washed with PBS and incubated with secondary antibodies, Alexa Fluor 594 goat anti-mouse IgG (A-11005, Invitrogen), and Alex Fluor 488 goat anti-rabbit IgG (A-11008, Invitrogen) for 1 hour under room temperature. After secondary antibody incubation, the cells were washed with PBS and nuclei were stained with DAPI (D1306, Invitrogen) in PBS at room temperature for 10 minutes. Images were acquired with confocal microscope (LSM780, Carl Zeiss).

Drosophila genetics

The following fly stocks were used in this study: Myc RNAi (VDRC 2947), hyd-mutant allele K3.5 (30), and MARCM82b (a kind gift from Dr. Osamu Shimmi, University of Helsinki, Helsinki, Finland). To induce MARCM clones in developing wing discs (31), larvae were heat-shocked for 1 hour at 37°C at 72 hours after egg laying. Wing imaginal discs were dissected from late third instar larvae, fixed in 4% formaldehyde for 30 minutes, after washing in PBT (0.3% TritonX100 in PBS), samples were mounted with VectaShield Mounting Medium with DAPI (Mediq), and imaged using a Zeiss LSM 700 microscope. Clone roundness was analyzed as described previously (32). Volume per cell and pH3-positive cells were counted using Imaris software. To induce MARCM clones in larvae fat body, larvae were heat-shocked for 1 hour at 37°C at 24 hours after egg laying. Fat bodies were dissected from third instar larvae, fixed in 4% formaldehyde for 30 minutes, and washed in PBT (0.3% TritonX100 in PBS). After blocking in 5% BSA in PBT for 3 hours at room temperature, primary antibody (anti-FBL, 1:500, Abcam) was incubated at 4°C overnight. Primary antibodies were washed with PBT and secondary antibody (anti-mouse Alexa Fluor 647, Life Technologies) was incubated for 4 hours at room temperature. After three washes, samples were mounted with VectaShield Mounting Medium with DAPI (Mediq) and imaged using a Zeiss LSM 700 microscope.

Full genotypes in Fig. 3D: yw, hsFLP/+; Tub-G4, UAS-mCD8 GFP/+; FRT82b, Tub-Gal80/FRT82, HydK3.5. Full genotypes in Fig. 3E and F:

Ctrl: yw, hsFLP/+; Tub-G4, UAS-mCD8 GFP/+; FRT82b, Tub-Gal80/FRT82b

Myc RNAi: yw, hsFLP/+; Tub-G4, UAS-mCD8 GFP/Myc RNAi; FRT82b, Tub-Gal80/FRT82b

Hyd mutant: yw, hsFLP/+; Tub-G4, UAS-mCD8 GFP/+; FRT82b, Tub-Gal80/FRT82, Hyd K3.5

Hyd mutant, Myc RNAi: yw, hsFLP/+; Tub-G4, UAS-mCD8 GFP/Myc RNAi; FRT82b, Tub-Gal80/FRT82b, Hyd K3.5.

Breast cancer IHC and immunofluorescence analysis

Tissue material was treated according to standard histology practice, that is, fixed in buffered formalin (pH 7.0) and embedded into paraffin blocks. Tissue microarrays (TMA) were prepared by collecting two 1.5 mm diameter tissue cores from the representative tumor area, defined by an experienced breast cancer pathologist (P. Kronqvist), of each patient with breast cancer. Both IHC and double immunofluorescence were performed on sections cut at 3.5 μm. For MYC (ab32072, Abcam, Y69), IHC was performed with Lab Vision Autostainer 480 (Thermo-Fisher Scientific) and detected with PowerVision+ Polymer Kit (DPVB+110HRP; Immunovision Technologies, Vision Biosystems) according to standard protocol with diaminobenzidine as chromogen. Before staining, tissue sections were deparaffinized and treated twice for 7 minutes each in Target Retrieval Solution, pH 9 (S2367, Dako) in a microwave oven for antigen retrieval. MYC antibody was applied at a dilution of 1:250. An automated immunostaining machine Discovery XT (Roche Diagnostics/Ventana Medical Systems) was used for UBR5 (sc-515494, Santa Cruz Biotechnology) IHC and UBR5/MYC double immunofluorescence. Deparaffinization, epitope retrieval (standard option, Cell Conditioning 1 reagent, 950-124, Roche/Ventana), and primary antibody incubation (40 minutes at 37°C, dilution 1:500 for UBR5-IHC and 1:1,000 and 1:50 for UBR5 and MYC, respectively, for double immunofluorescence) were done on the platform. OmniMap HRP (760-4310, Roche/Ventana) and ChromoMap DAB Kit (760-159, Roche/Ventana) were applied for detection of UBR5 IHC. For double immunofluorescence, OmniMap HRP (Roche/Ventana, 760-4310 and 760-4311 for anti-mouse and anti-rabbit, respectively) together with Rhodamine and FAM fluorescent substrates (Roche/Ventana, 760-233 and 760-243, respectively) were used. Finally, immunofluorescence slides were mounted applying ProLong Gold antifade reagent with DAPI (P36935, Molecular Probes by Life Technologies). Two slides were stained as controls for double immunofluorescence. One, by replacing UBR5 primary antibody, and the other, by replacing MYC primary antibody with antibody diluent.

UBR5 suppress MYC protein expression

We conducted a screen for ubiquitin ligases that regulate MYC protein levels using a library of siRNAs targeting 591 ubiquitin ligases (Fig. 1A). To focus on ubiquitin ligases that would not regulate MYC stability via the important MYC ubiquitin ligase FBXW7 (9, 10, 33), the siRNA library was screened against U2OS cells stably expressing MYCT58A mutant (threonine 58 mutated to alanine) that is resistant to FBXW7-mediated destabilization (9, 10). Forty-eight hours after siRNA transfection, cells were treated for 3.5 hours with cycloheximide to emphasize the impact of protein stability, and immunofluorescence detection of MYC was thereafter used as a read-out in a high-content imaging-based assay (Fig. 1A). Among a small group of siRNAs consistently affecting MYCT58A levels (Supplementary Table S2), the HECT domain–containing E3 ligase UBR5 (alias EDD) was the only E3 ligase affecting only protein levels but not mRNA levels (Supplementary Fig. S1A) and was therefore selected for further validation experiments.

Figure 1.

UBR5 negatively regulates MYC protein levels. A, High-content imaging-based screen was performed to identify E3 ubiquitin ligases that regulate MYCT58A. An example of immunofluorescence (IF) staining of MYC and a secondary antibody control only is shown. B, Proximity ligation assay analysis of the UBR5 and MYC association in HeLa cells transfected with scrambled (Scr), MYC siRNA, or UBR5 siRNA. Shown is a quantification of the average number of UBR5-MYC PLA signals per cell in cytoplasm and nucleus. C, UBR5, wild-type MYC, or MYCT58A mutants were overexpressed in HEK293 cells by transient transfection. Cell lysates were immunoprecipitated with GFP or V5 agarose and interaction was detected with relevant antibodies. The numbers show fold change of relevant proteins. Shown is a representative experiment; n = 3. D, HEK293 cells were cotransfected with empty vector, GFP-tagged UBR5, UBR5 mutant, V5-tagged MYC, and MYCT58A-mutant plasmids. Shown is a representative experiment; n = 2. E, UBR5, wild-type MYC, or MYC mutants were overexpressed in HEK293 cells by transient transfection. Cell lysates were immunoprecipitated with HA antibody and interaction was detected with GFP antibody. Shown is a representative experiment; n = 3. F, HeLa cells were transfected with scrambled or 6 different UBR5 siRNAs. After 72 hours, cell lysates were collected to detect MYC expression by Western blot analysis. G, Effect of UBR5 siRNA on MYC in T98G, MDA-MB-231, and MCF10A cells. H, Gene set enrichment analysis of RNA-seq data generated in HeLa cells transfected with Scr, UBR5, or MYC siRNAs. NES, normalized enrichment score. I, Heatmap illustrating the expression-level changes (>2 folds by siUBR5) of genes coregulated by UBR5 and MYC in H.

Figure 1.

UBR5 negatively regulates MYC protein levels. A, High-content imaging-based screen was performed to identify E3 ubiquitin ligases that regulate MYCT58A. An example of immunofluorescence (IF) staining of MYC and a secondary antibody control only is shown. B, Proximity ligation assay analysis of the UBR5 and MYC association in HeLa cells transfected with scrambled (Scr), MYC siRNA, or UBR5 siRNA. Shown is a quantification of the average number of UBR5-MYC PLA signals per cell in cytoplasm and nucleus. C, UBR5, wild-type MYC, or MYCT58A mutants were overexpressed in HEK293 cells by transient transfection. Cell lysates were immunoprecipitated with GFP or V5 agarose and interaction was detected with relevant antibodies. The numbers show fold change of relevant proteins. Shown is a representative experiment; n = 3. D, HEK293 cells were cotransfected with empty vector, GFP-tagged UBR5, UBR5 mutant, V5-tagged MYC, and MYCT58A-mutant plasmids. Shown is a representative experiment; n = 2. E, UBR5, wild-type MYC, or MYC mutants were overexpressed in HEK293 cells by transient transfection. Cell lysates were immunoprecipitated with HA antibody and interaction was detected with GFP antibody. Shown is a representative experiment; n = 3. F, HeLa cells were transfected with scrambled or 6 different UBR5 siRNAs. After 72 hours, cell lysates were collected to detect MYC expression by Western blot analysis. G, Effect of UBR5 siRNA on MYC in T98G, MDA-MB-231, and MCF10A cells. H, Gene set enrichment analysis of RNA-seq data generated in HeLa cells transfected with Scr, UBR5, or MYC siRNAs. NES, normalized enrichment score. I, Heatmap illustrating the expression-level changes (>2 folds by siUBR5) of genes coregulated by UBR5 and MYC in H.

Close modal

Physical association between endogenous UBR5 and MYC in HeLa cell nuclei was confirmed by proximity ligation analysis (PLA; Fig. 1B). The specificity of the PLA reaction was confirmed by staining with PLA secondary antibodies alone (Supplementary Fig. S1B), and by a clear decrease of positive MYC-UBR5 PLA signals (Fig. 1B), and of UBR5 immunofluorescence signal (Supplementary Fig. S1C and S1D) by siRNA treatments. Whereas MYC and UBR5 siRNA treatments exclusively decreased the number of nuclear PLA signals, they did not affect cytoplasmic background signals (Fig. 1B).

To confirm FBXW7-independent function of UBR5, we studied whether MYC binding to UBR5 is affected by MYC T58A mutation, which abolishes recognition by FBXW7 (7, 9, 34). As a result, GFP-UBR5 associated equally efficiently with both forms of MYC (Fig. 1C). Lack of interaction between overexpressed GFP and MYC was controlled separately (Supplementary Fig. S1E). In addition, UBR5 overexpression reduced the levels of MYC T58A and WT MYC to a similar degree (Fig. 1D). Importantly, mutant UBR5 lacking the ubiquitin ligase HECT domain did not reduce the MYC levels (Fig. 1D).

By using different MYC fragments (35, 36), we delineated amino acids 181–189 of MYC as the minimal candidate region for MYC–UBR5 association (Fig. 1E; Supplementary Fig. S1F). Importantly, the MYC mutant deficient in UBR5 binding (d127–189) repeatedly showed higher protein expression than either WT MYC or the N-terminal MYC mutants (Fig. 1E; Supplementary Fig. S1G), and was resistant to inhibition by UBR5 overexpression (Supplementary Fig. S1G).

The effects of UBR5 depletion on endogenous MYC protein levels in HeLa cells was confirmed by immunoblot analysis using six independent UBR5 siRNA sequences (Fig. 1F). However, MYC did not regulate UBR5 mRNA or protein expression (Supplementary Fig. S1H and S1I). In addition to HeLa cells, endogenous MYC protein induction by UBR5 inhibition was confirmed in T98G glioblastoma, and MDA-MB-231 breast cancer cells, and in immortalized MCF-10A mammary epithelial cells (Fig. 1G). In addition, we performed RNA-sequencing analysis of HeLa cells depleted of MYC or UBR5 (Supplementary Fig. S1I). By gene-set enrichment analysis, we found that UBR5 depletion induces expression of same MYC target signature that is suppressed by MYC depletion (Fig. 1H). A heatmap of the individual genes regulated to opposite direction with UBR5 or MYC siRNA is shown in Fig. 1I and the genes are listed in Supplementary Table S3.

UBR5 is a MYC ubiquitin ligase

UBR5 depletion induced a robust increase in the levels of MYC protein but its impact on the MYC mRNA levels was minimal (Fig. 2A). Furthermore, UBR5 depletion significantly increased MYC protein expression when ectopic MYC mRNA expression was induced by gradual removal of doxycycline from mouse OS-Tet-Off-MYC cells (27), further supporting a regulation at posttranslational levels (Fig. 2B). Consistent with UBR5 regulating MYC protein stability, UBR5 depletion stabilized MYC in a cycloheximide chase experiment in HeLa (Fig. 2C; Supplementary Fig. S2A) and in U2OS cells (Supplementary Fig. S2B and S2C). Regarding other MYC family members, endogenous MYCN was also stabilized by UBR5 siRNA in IMR-32 neuroblastoma cells (Supplementary Fig. S2D and S2E). Further validating the UBR5 degron on MYC, MYC d129-189 mutant was insensitive to UBR5-mediated destabilization (Supplementary Fig. S2F and S2G). Notably, depletion of either FBXW7 or UBR5 increased endogenous MYC stability and codepletion of both ubiquitin ligases further increased MYC stability (Fig. 2D; Supplementary Fig. S2H and S2I). Additive effects of UBR5 and FBXW7 inhibition on MYC stability were confirmed by M. Eiler's laboratory by using independent UBR5 and FBXW7 siRNAs (Supplementary Fig. S2J). These results clearly support independent roles for UBR5 and FBXW7 in MYC protein regulation.

Figure 2.

Validation of UBR5 as a MYC ubiquitin ligase. A, MYC mRNA and protein expression in HeLa cells transfected with scrambled (Scr) or UBR5 (5#) siRNA. Error bars, SD (for mRNA expression, n = 3; for protein expression, n = 14). ***, P < 0.001, ****, P < 0.0001 by Student t test. B, Analysis of MYC expression in mouse osteosarcoma-MYC-off cells transfected with scrambled or UBR5 siRNA. C, HeLa cells transfected with scrambled or UBR5 (5#) siRNA for 72 hours were treated with cycloheximide (CHX; 60 μg/mL) for indicated time points. For Western blot analysis, see Supplementary Fig. S2A. Error bars, mean ± SD. *, P < 0.05 by Student t test (n = 3). D, HeLa cells transfected with indicated siRNAs for 72 hours were treated with cycloheximide (60 μg/mL) for different time points. For Western blot analysis, see Supplementary Fig. S2I. Error bars, mean ± SD. *, P < 0.05 by Student t test (n = 3). E, HEK293 cells were cotransfected with GFP-tagged wild-type UBR5, UBR5-HECT domain mutant, and V5-tagged MYC plasmid. After 48-hour transfection, the cells were treated with MG132 (20 μmol/L) for 6 hours. SE, short exposure time; LE, long exposure time. F, Quantification of MYC expression from E. Error bars show SD from four independent experiments. *, P < 0.05 by Student t test. GI, HEK293 (G) or HeLa (H and I) cells were cotransfected with indicated plasmids. After 48-hour transfection, the cells were treated with MG132 (20 μmol/L) for 6 hours. Immunoprecipitation was performed with anti-V5 agarose. Ubiquitination of MYC was detected by HA antibody (G and H) or ubiquitin K48–specific antibody. J, Recombinant GST-MYC (1–262) and GFP-UBR5 purified from HEK293 cells were incubated in in vitro–ubiquitination reactions. Samples were analyzed by Western blotting using GFP and GST antibodies. G–J, Shown is a representative experiment (n = 2–3). n.s., nonsignificant.

Figure 2.

Validation of UBR5 as a MYC ubiquitin ligase. A, MYC mRNA and protein expression in HeLa cells transfected with scrambled (Scr) or UBR5 (5#) siRNA. Error bars, SD (for mRNA expression, n = 3; for protein expression, n = 14). ***, P < 0.001, ****, P < 0.0001 by Student t test. B, Analysis of MYC expression in mouse osteosarcoma-MYC-off cells transfected with scrambled or UBR5 siRNA. C, HeLa cells transfected with scrambled or UBR5 (5#) siRNA for 72 hours were treated with cycloheximide (CHX; 60 μg/mL) for indicated time points. For Western blot analysis, see Supplementary Fig. S2A. Error bars, mean ± SD. *, P < 0.05 by Student t test (n = 3). D, HeLa cells transfected with indicated siRNAs for 72 hours were treated with cycloheximide (60 μg/mL) for different time points. For Western blot analysis, see Supplementary Fig. S2I. Error bars, mean ± SD. *, P < 0.05 by Student t test (n = 3). E, HEK293 cells were cotransfected with GFP-tagged wild-type UBR5, UBR5-HECT domain mutant, and V5-tagged MYC plasmid. After 48-hour transfection, the cells were treated with MG132 (20 μmol/L) for 6 hours. SE, short exposure time; LE, long exposure time. F, Quantification of MYC expression from E. Error bars show SD from four independent experiments. *, P < 0.05 by Student t test. GI, HEK293 (G) or HeLa (H and I) cells were cotransfected with indicated plasmids. After 48-hour transfection, the cells were treated with MG132 (20 μmol/L) for 6 hours. Immunoprecipitation was performed with anti-V5 agarose. Ubiquitination of MYC was detected by HA antibody (G and H) or ubiquitin K48–specific antibody. J, Recombinant GST-MYC (1–262) and GFP-UBR5 purified from HEK293 cells were incubated in in vitro–ubiquitination reactions. Samples were analyzed by Western blotting using GFP and GST antibodies. G–J, Shown is a representative experiment (n = 2–3). n.s., nonsignificant.

Close modal

Consistent with ubiquitination-mediated regulation of MYC by UBR5, MYC inhibition by overexpression of the wild-type UBR5 was rescued by cotreatment with proteasome inhibitor MG132 (Fig. 2E and F). Moreover, in ubiquitination assays, only overexpression of the wild-type UBR5, but not the HECT mutant, induced ubiquitination of V5-MYC (Fig. 2G). Reciprocally, siRNA-mediated depletion of UBR5 potently inhibited MYC ubiquitination (Fig. 2H). Moreover, we found that UBR5-induced ubiquitination of MYC is at least in part via K48-linked ubiquitin chains (Fig. 2I), which are mainly responsible for proteasomal degradation of proteins. In addition, increased covalent attachment of ubiquitin to MYC in UBR5-overexpressing cells was confirmed by performing the His-ubiquitin pull-down under denaturating conditions (Supplementary Fig. S2K). Finally, by using purified UBR5 and MYC proteins, we demonstrate that UBR5 can in vitro conjugate ubiquitin to the 1–262 fragment of recombinant MYC, spanning the UBR5 degron (Fig. 2J).

Together, these data identify UBR5 as a novel MYC ubiquitin ligase that regulates MYC stability independently of FBWX7.

UBR5 controls in vivo tissue growth in dMYC-dependent manner in Drosophila

To explore the possible functional conservation, and the role of the UBR5–MYC axis in normal tissue homeostasis in vivo, we utilized a Drosophila melanogaster model. The Drosophila UBR5 ortholog HYD (Hyperplastic discs) was originally identified as a putative tumor suppressor (19, 22, 26, 37). RNAi-mediated depletion of Hyd in Drosophila S2 cells (Supplementary Fig. S3A) led to elevated levels of Drosophila MYC (dMYC) protein (Fig. 3A), whereas the mRNA levels of dMyc did not correlate with dMYC protein induction (Fig. 3B). Importantly, the levels of Hyd RNA reduction by RNAi was directly reflected in a comparable 2-fold induction of dMYC protein levels (Supplementary Fig. S3A). Similar to regulation of MYC target genes by UBR5 in human cells (Fig. 1H and I), Hyd depletion induced mRNA expression of two dMYC targets eIF6 and Nop5 (Fig. 3C; Supplementary Fig. S3B), and protein expression of fibrillarin (Fig. 3D and E), which is a well-established MYC target in both human and Drosophila.

Figure 3.

Loss of UBR5 ortholog Hyd drives wing imaginal disc overgrowth in Myc-dependent manner. A, Immunoblot showing immunoprecipitated dMyc levels in S2 cells transfected with Scr or Hyd dsRNA. Kinesin was used as a control for the input. Shown is a representative blot of two experiments with similar results. B, Quantitative RT-PCR analysis of dMyc mRNA expression in control and Hyd-depleted S2 cells. RP49 was used as a reference gene. Error bars, SD; n = 3. C, eIF6 mRNA expression in control and Hyd-depleted S2 cells. Error bars, SD; n = 3. *, P < 0.05 by Student t test. D, Loss of Hyd in GFP-labeled fat body mutant clones leads to increase of fibrillarin expression. E, Quantification of D. The fibrillarin level of an adjacent GFP-negative cell was used as control. n = 10; **, P < 0.01 by Student t test. F, Induction of hyd-mutant clones leads to uneven wing morphology, which in suppressed by simultaneous knockdown of Myc within the clones. Arrows in magnified image show uneven wing margins and veins. G, Loss of Hyd in GFP-labeled mutant clones leads to MYC-dependent overgrowth, indicated by roundness and loss of normal epithelial morphology. Scale bar, 10 μm. H, Representative images show clone morphology in wing imaginal discs in control and hyd-mutant clones. Scale bar, 10 μm. Quantification of the clone roundness is shown. Error bars, SD. n = 30; **, P < 0.01 by Student t test. I, Loss of Hyd in GFP-labeled wing disc–mutant clones leads to increased cell volume. Simultaneous knockdown of dMyc suppresses this phenotype (n = 3; *, P < 0.05 by Student t test). n.s., nonsignificant.

Figure 3.

Loss of UBR5 ortholog Hyd drives wing imaginal disc overgrowth in Myc-dependent manner. A, Immunoblot showing immunoprecipitated dMyc levels in S2 cells transfected with Scr or Hyd dsRNA. Kinesin was used as a control for the input. Shown is a representative blot of two experiments with similar results. B, Quantitative RT-PCR analysis of dMyc mRNA expression in control and Hyd-depleted S2 cells. RP49 was used as a reference gene. Error bars, SD; n = 3. C, eIF6 mRNA expression in control and Hyd-depleted S2 cells. Error bars, SD; n = 3. *, P < 0.05 by Student t test. D, Loss of Hyd in GFP-labeled fat body mutant clones leads to increase of fibrillarin expression. E, Quantification of D. The fibrillarin level of an adjacent GFP-negative cell was used as control. n = 10; **, P < 0.01 by Student t test. F, Induction of hyd-mutant clones leads to uneven wing morphology, which in suppressed by simultaneous knockdown of Myc within the clones. Arrows in magnified image show uneven wing margins and veins. G, Loss of Hyd in GFP-labeled mutant clones leads to MYC-dependent overgrowth, indicated by roundness and loss of normal epithelial morphology. Scale bar, 10 μm. H, Representative images show clone morphology in wing imaginal discs in control and hyd-mutant clones. Scale bar, 10 μm. Quantification of the clone roundness is shown. Error bars, SD. n = 30; **, P < 0.01 by Student t test. I, Loss of Hyd in GFP-labeled wing disc–mutant clones leads to increased cell volume. Simultaneous knockdown of dMyc suppresses this phenotype (n = 3; *, P < 0.05 by Student t test). n.s., nonsignificant.

Close modal

The Drosophila models allows the use of somatic recombination to generate clones of mutant tissue in an otherwise heterozygous background. Clones of HydK3.5-mutant cells generated during mid larval development (72 houfrs after egg laying) showed a phenotype clearly visible in the adult wings (Supplementary Fig. S3C). Wings with Hyd-mutant clones were irregular, lacking the normal flat wing morphology, and displaying uneven wing margins and veins (Fig. 3F; Supplementary Fig. S3C, see arrows). Notably, this irregular wing phenotype was strongly suppressed by simultaneous knockdown of dMyc in the mutant clones (Fig. 3F; Supplementary Fig. S3C).

The use of the MARCM system to generate mutant clones allowed us to GFP label, and to visualize the tissue morphology of the wing imaginal discs, the larval wing precursors (31). Control and Myc RNAi clones appeared morphologically normal with GFP-positive clones having a normal but irregular shape (Fig. 3G and H). In contrast, Hyd-mutant clones had a round morphology with clusters of cells growing out from the epithelial plane (Fig. 3G and H), which was confirmed by quantification of the clone roundness (Fig. 3H). The roundness phenotype in wing clones has been earlier validated to mark activated Ras–MAPK–MYC signaling (32). Most importantly, simultaneous RNAi-mediated knockdown of dMyc suppressed the round morphology, thus rescuing the phenotype of the hyd-mutant clones, which was further confirmed by quantification (Fig. 3H).

To dissect the relevance of cell size and proliferation (15) in dMYC-dependent regulation of Drosophila tissue growth by Hyd, we first analyzed cell size in Hyd-mutant clones. Hyd-mutant cells were significantly larger than controls and this phenotype was fully suppressed by simultaneous knockdown of dMyc (Fig. 3I). We also observed higher percentage of phospho-histone H3, a marker commonly used as a proxy for cell proliferation, in the Hyd-mutant clones, but this phenotype was independent of dMYC expression (Supplementary Fig. S3D). Thus, we conclude that the increased cell size, and roundness (established mark of activated MYC signaling; ref. 32), but not proliferation, are dependent on dMYC in Hyd-mutant Drosophila tissues.

These results demonstrate that in Drosophila suppression of dMYC protein expression is a critical part of growth control by the ortholog of UBR5 in vivo.

UBR5 suppresses MYC-mediated apoptosis priming in cancer cells

Proliferation and apoptosis are known to be governed by different levels of MYC (2–5, 12, 17, 38, 39). In HeLa cells, UBR5 depletion decreased colony growth (Fig. 4A), suggesting that MYC levels may have reached the levels that prime them to apoptosis. This was confirmed by the increased PARP cleavage, and by the induction of caspase-3/7 activity in at least a fraction of UBR5-depleted cells (Fig. 4B and C). To evaluate the balance between MYC-regulated proliferation and apoptosis, we titrated UBR5 siRNA and studied the dose-dependent correlation between MYC levels, and induction of cleaved-PARP or inhibition of a directly MYC-regulated cell-cycle inhibitor p21 (40). Interestingly, while p21 suppression was almost maximal already with 10 nmol/L of UBR siRNA (Fig. 4D and E), cleaved PARP continued to increase with higher UBR5 siRNA concentrations (Fig. 4D and E). These results indicate that both proliferation and apoptosis priming are initiated upon increased MYC expression, and that loss of cells in colony growth assay (Fig. 4A) is a result of MYC expression exceeding the apoptosis-inducing levels. Importantly, apoptosis priming by UBR5 inhibition was highly dependent on MYC induction. First, both PARP cleavage, and caspase-3/7 activation were fully rescued by concurrent MYC depletion (Fig. 4F and G). Second, UBR5 depletion was unable to induce PARP cleavage in OS-Tet-Off-MYC cells with maximal doxycycline-elicited MYC suppression (Fig. 4H).

Figure 4.

UBR5 suppresses MYC-mediated apoptosis in cancer cells. A, HeLa cells transfected with three UBR5 siRNAs were cultured 7 days for colony growth assay. Error bars, SD. n = 3. **, P <0.01 by Student t test. B, HeLa cells were transfected with two UBR5 siRNAs for 72 hours. Total cell lysates were subjected to immunoblotting with indicated antibodies. C, Caspase-3/7 activity was examined in UBR5-depleted HeLa cells. Error bars show SD for seven technical repeats with similar results. D, HeLa cells were transfected with different concentrations of UBR5 siRNA for 72 hours. Cell lysates were analyzed by Western blot with indicated antibodies. E, Quantification of MYC, cleared PARP, and p21 expression from D. Error bars show SD from two independent experiments. F, HeLa cells were transfected with two UBR5 siRNAs or with combination of UBR5 siRNA and MYC siRNA for 72 hours. Cell lysates were collected for Western blot analysis with indicated antibodies. G, Caspase-3/7 activity was examined in HeLa cells transfected with indicated siRNAs. Error bars show SD of two independent experiments. H, The osteosarcoma-MYC-off cells treated with 20 ng/mL doxycycline were transfected with scrambled (Scr) or UBR5 siRNA and cultured in medium with 20 ng/mL doxycycline. After 48 hours, medium was changed with or without 20 ng/mL doxycycline for culture another 48 hours. I, Caspase-3/7 activity was examined in HeLa cells transfected with indicated siRNAs. Shown is mean ± SD of two independent experiments. J, HeLa cells were transfected with scrambled, UBR5 siRNAs, combination of UBR5 and MYC siRNA, and combination of UBR5 and BIM siRNAs. K, Colony growth assay was performed in HeLa cells transfected with scrambled, siUBR5, or combination of UBR5 and MYC siRNAs, followed by 24-hour treatment of Taxol (2 nmol/L) or CPT (20 nmol/L). After 7 days, cells were fixed and stained. L, Quantification of drug response fold changes relative to scrambled siRNA in K. Error bars show SEM of four independent experiments. *, P < 0.05; **, P < 0.01 by Student t test; n.s., nonsignificant.

Figure 4.

UBR5 suppresses MYC-mediated apoptosis in cancer cells. A, HeLa cells transfected with three UBR5 siRNAs were cultured 7 days for colony growth assay. Error bars, SD. n = 3. **, P <0.01 by Student t test. B, HeLa cells were transfected with two UBR5 siRNAs for 72 hours. Total cell lysates were subjected to immunoblotting with indicated antibodies. C, Caspase-3/7 activity was examined in UBR5-depleted HeLa cells. Error bars show SD for seven technical repeats with similar results. D, HeLa cells were transfected with different concentrations of UBR5 siRNA for 72 hours. Cell lysates were analyzed by Western blot with indicated antibodies. E, Quantification of MYC, cleared PARP, and p21 expression from D. Error bars show SD from two independent experiments. F, HeLa cells were transfected with two UBR5 siRNAs or with combination of UBR5 siRNA and MYC siRNA for 72 hours. Cell lysates were collected for Western blot analysis with indicated antibodies. G, Caspase-3/7 activity was examined in HeLa cells transfected with indicated siRNAs. Error bars show SD of two independent experiments. H, The osteosarcoma-MYC-off cells treated with 20 ng/mL doxycycline were transfected with scrambled (Scr) or UBR5 siRNA and cultured in medium with 20 ng/mL doxycycline. After 48 hours, medium was changed with or without 20 ng/mL doxycycline for culture another 48 hours. I, Caspase-3/7 activity was examined in HeLa cells transfected with indicated siRNAs. Shown is mean ± SD of two independent experiments. J, HeLa cells were transfected with scrambled, UBR5 siRNAs, combination of UBR5 and MYC siRNA, and combination of UBR5 and BIM siRNAs. K, Colony growth assay was performed in HeLa cells transfected with scrambled, siUBR5, or combination of UBR5 and MYC siRNAs, followed by 24-hour treatment of Taxol (2 nmol/L) or CPT (20 nmol/L). After 7 days, cells were fixed and stained. L, Quantification of drug response fold changes relative to scrambled siRNA in K. Error bars show SEM of four independent experiments. *, P < 0.05; **, P < 0.01 by Student t test; n.s., nonsignificant.

Close modal

Mechanistically, MYC overexpression can induce apoptosis either via ARF–MDM2–p53–dependent pathway (2, 41, 42), or by engaging proapoptotic BH3 protein BIM in a p53-independent manner (39, 40, 43). Supportive of a BIM-dependent mechanism, we found that codepletion of BIM abolished apoptosis induction in UBR5-depleted cells (Fig. 4I). Moreover, BIM induction by UBR5 siRNA was abolished by codepletion of MYC (Fig. 4J). On the other hand, cells depleted of p53 displayed equally potent colony growth reduction and PARP cleavage by UBR5 depletion than control cells (Supplementary Fig. S4A and S4B). Furthermore, endogenous inactivating p53 mutations did not prevent MYC-dependent PARP cleavage, or BIM induction, in UBR5 siRNA–transfected breast cancer cells (Supplementary Fig. S4C).

MYC-dependent apoptosis priming is particularly relevant for increased sensitivity of cancer cells to antimitotic agents such as taxanes (17), and to drugs that interfere with high replication activity, such as topoisomerase I inhibitors (camptothecins; ref. 16). Accordingly, growth inhibition in UBR5-depleted cells was greatly potentiated by treatment of cells with either taxol or camptothecin, at doses that alone do not induce massive loss of cells (Fig 4K; Supplementary Fig. S4D). Importantly, MYC expression was essential for hypersensitivity of UBR5-depleted cells to both taxol and camptothecin (Fig. 4K and L) because drug responses were indistinguishable between the control cells, and cells treated with UBR5 + MYC siRNA in a concentration-dependent manner (Supplementary Fig. S4E).

Collectively, these results demonstrate that UBR5 suppresses MYC-mediated, but p53-independent apoptosis priming in cancer cells.

Coamplification of UBR5 and MYC protects breast cancer cells from MYC-mediated priming to drug-induced apoptosis

The results above suggest that expression of UBR5 might provide a selective advantage for cancer cells by suppressing MYC-mediated apoptosis priming. The cancer types in which this biology is particularly relevant would be expected to exhibit coexpression of UBR5 and MYC at mRNA level, but UBR5high/MYClow status at protein level (see schematic in Fig. 5A). Across 672 cell lines and over 20 different cancer types (44), a weak but statistically significant correlation between UBR5 and MYC mRNA expression levels was detected (Pearson correlation 0.21, P < 0.01; Supplementary Fig. S5A). Significant positive correlation between MYC and UBR5 mRNA expression was observed in four cancer types, ovary, lymphoid, breast, and pancreas (Supplementary Fig. S5A). Importantly, MYC has been closely linked to progression of all these four cancer types, and existing data indicate a growth-promoting role of UBR5 in at least pancreatic, ovary, and breast cancer (22, 23, 45, 46).

Figure 5.

UBR5 and MYC are coamplified in human breast cancer. A, Schematic figure of UBR5 and MYC gene locus in the long arm of chromosome 8 and indicated relationship between their expression at mRNA and protein levels. Whereas mRNA expression is predicted to correlate in coamplified samples (upward arrows), UBR5 is predicted to suppress MYC protein expression in these samples (downward arrows). B, Scatter plot showing the relationship between UBR5/MYC coamplification and UBR5 gene essentiality in the breast cancer cell lines. The minimum of UBR5 and MYC amplification levels were correlated against the zGARP scores, a measure of UBR5 essentiality in a shRNA dropout screen. C, The volume of tumors isolated from mice on the final day of the experiment (day 28; n = 8). D,UBR5/MYC coamplified HCC1937 or HCC38 cells were transfected with indicated siRNAs. E, Colony growth assay of siRNA-transfected HCC38 cells was performed following 24-hour treatment with irinotecan (80 nmol/L) or topotecan (40 nmol/L). Error bars show SEM of two independent experiments. F, Colony growth assay of siRNA-transfected HCC38 cells was performed following 24-hour treatment with Taxol (8 nmol/L). Error bars show SEM of three independent experiments. *, P < 0.05; **, P < 0.01 by Student t test; n.s., nonsignificant.

Figure 5.

UBR5 and MYC are coamplified in human breast cancer. A, Schematic figure of UBR5 and MYC gene locus in the long arm of chromosome 8 and indicated relationship between their expression at mRNA and protein levels. Whereas mRNA expression is predicted to correlate in coamplified samples (upward arrows), UBR5 is predicted to suppress MYC protein expression in these samples (downward arrows). B, Scatter plot showing the relationship between UBR5/MYC coamplification and UBR5 gene essentiality in the breast cancer cell lines. The minimum of UBR5 and MYC amplification levels were correlated against the zGARP scores, a measure of UBR5 essentiality in a shRNA dropout screen. C, The volume of tumors isolated from mice on the final day of the experiment (day 28; n = 8). D,UBR5/MYC coamplified HCC1937 or HCC38 cells were transfected with indicated siRNAs. E, Colony growth assay of siRNA-transfected HCC38 cells was performed following 24-hour treatment with irinotecan (80 nmol/L) or topotecan (40 nmol/L). Error bars show SEM of two independent experiments. F, Colony growth assay of siRNA-transfected HCC38 cells was performed following 24-hour treatment with Taxol (8 nmol/L). Error bars show SEM of three independent experiments. *, P < 0.05; **, P < 0.01 by Student t test; n.s., nonsignificant.

Close modal

We noted that both UBR5 and MYC genes are located in the long arm of chromosome 8, suggesting a model where UBR5 coamplification with MYC might provide means to control excessive MYC protein levels, and thereby protect cells from apoptosis priming (Fig. 5A). Examination of The Cancer Genome Atlas cancer patient amplification data for UBR5 and MYC in the four cancer types with an evidence for mRNA coexpression (Supplementary Fig. S5A), revealed that the percentage of patient samples with single amplification of either UBR5 or MYC ranged from 2% in lymphoid cancers, to 44 % in ovary cancers (Table 1). Interestingly, coamplification of UBR5 and MYC appeared in 65% of breast, 42% of ovary, and 39% of pancreas cancers. On the other hand amplification of UBR5 alone was a very rare event in any of these cancer types. Breast cancers were the only cancer type in which the percentage of cancers with coamplification exceeded the percentage of cancers with MYC amplification alone (Table 1), indicating that UBR5 coamplification, and UBR5 protein expression, may provide a particular benefit for breast cancer cells with MYC amplification.

Table 1.

Distribution of amplification rates of MYC and UBR5 in The Cancer Genome Atlas patient samples in the indicated cancer types.

BreastPancreasOvaryLymphoid
Total number of samples 1,080 183 579 48 
No. of samples with either MYC and/or UBR5 amplifications (% of total) 247 (22.8%) 23 (12.5%) 254 (43.8%) 1 (2%) 
Cancers with MYC amplifications only (% of amplifications) 28% 61% 54% 1% 
Cancer with UBR5 amplifications only (% of amplifications) 7% 0% 4% 0% 
Cancers with MYC/UBR5 coamplifications (% of amplifications) 65% 39% 42% 0% 
BreastPancreasOvaryLymphoid
Total number of samples 1,080 183 579 48 
No. of samples with either MYC and/or UBR5 amplifications (% of total) 247 (22.8%) 23 (12.5%) 254 (43.8%) 1 (2%) 
Cancers with MYC amplifications only (% of amplifications) 28% 61% 54% 1% 
Cancer with UBR5 amplifications only (% of amplifications) 7% 0% 4% 0% 
Cancers with MYC/UBR5 coamplifications (% of amplifications) 65% 39% 42% 0% 

To assess functional relevance of UBR5 in coamplified breast cancer cells, we correlated breast cancer cell line UBR5 essentiality index (zGARP score; ref. 47), with MYC/UBR5 gene copy numbers in these same cells. Notably, we observed a statistically significant correlation between the zGARP score and MYC/UBR5 gene copy numbers in cell lines with zGARP score < −1.5 (Fig. 5B). Next, we tested the xenograft growth potential of UBR5-depleted HCC38 cells. Transient siRNA transfection of HCC38 cells resulted in at least 10 days suppression of UBR5 protein expression validating the approach for xenograft experiment (Supplementary Fig. S5B). Remarkably, UBR5 depletion from HCC38 even with transient siRNA transfection was sufficient to almost entirely inhibit tumor growth in mice (Fig. 5C). In culture, we observed that UBR5 depletion in both HCC38 and HCC1937 cells did induce MYC protein levels and MYC-dependent apoptosis priming (as observed by PARP cleavage; Fig. 5D; Supplementary Fig. S5C). As HCC38 harbors p53 R273L mutation in the DNA-binding domain, these results further strengthen the p53 independence of apoptosis priming induced by loss of UBR5.

Patients with coamplification of MYC and UBR5 had significantly poorer overall survival than patients without neither of the genes amplified (Supplementary Fig. S5D). However, as survival of patients with MYC amplification alone did not statistically differ from survival of patients with coamplification, we conclude that potential relevance of UBR5-mediated MYC suppression in coamplified breast cancers could rather associate with the resistance to therapy-induced apoptosis. Fully supporting the hypothesis, UBR5 depletion from HCC38 cells with MYC/UBR5 coamplification resulted in modest MYC-dependent inhibition of colony growth, but led to a very potent MYC-dependent sensitization of cell killing to both FDA-approved topoisomerase I inhibitors, irinotecan and topotecan (Fig. 5E; the full concentration-dependent dose–response curves are show in Supplementary Fig. S5E and S5F). We also confirmed a significant MYC-dependent priming to Taxol-induced cell killing by UBR5 inhibition in HCC38 cells (Fig. 5F).

Together, these results identify coamplification of MYC and UBR5 in large fraction of MYC-dependent solid cancer types. Results further indicate that UBR5 coamplification with MYC might be relevant to suppress drug-induced apoptosis in p53-mutant breast cancer cells.

UBR5 dominates MYC protein levels in individual breast cancer cells in vivo

Finally, we searched for in vivo evidence that UBR5 suppresses MYC protein expression in human breast cancer tissue. Consistent with published results (48), MYC amplification was most frequent in basal-type breast cancers, and this was also the subtype where coamplification was most pronounced (Supplementary Fig. S6A). On the basis of this information, we used tissue microarray of 345 samples from 197 human basal-type breast cancers to examine correlation between UBR5 and MYC protein expression in vivo. The IHC staining was optimized such that intensities from 0 to +++ were reliably observed with both UBR5 and MYC antibodies (Supplementary Fig. S6B). Consistently, with the hypothesis of UBR5 dominating MYC expression, the percentage of staining positive cells (+, ++, or +++) was greater for UBR5 than for MYC in 78% of the samples (Supplementary Fig. S6C). This finding was reconfirmed by using an independent set of human breast cancer samples (n = 74) and a different staining protocol in Klefström laboratory (Supplementary Fig. S6D).

When tumor samples were examined by using adjacent 3-μm tissue sections, we observed an apparent dichotomy of UBR5 and MYC protein levels in individual basal-type breast cancer cells (Supplementary Fig. S6E; regions 1 and 2). Importantly, while UBR5 and MYC positivity is almost exclusively confined to tumor cells (Supplementary Fig. S6E and S6F), we could identify some immune-like cells that also show very clear inverse correlation between UBR5 and MYC expression (Supplementary Fig. S6E). These results further emphasize the general importance of our results across different cell types.

To quantitatively validate these observations, we developed a double immunofluorescence staining protocol. Antibody specificity was demonstrated by siRNA-mediated depletion (Fig. 1B; Supplementary Fig. S1D), and by secondary antibody only control staining (Supplementary Fig. S6G). Furthermore, the double immunofluorescence was optimized so that maximal staining intensities were practically indistinguishable with both UBR5 and MYC antibodies (Supplementary Fig. S6H). Thus, individual cells could be categorized into MYChigh/UBR5low (green), UBR5high/MYClow (red), and MYChigh/UBR5high (yellow) phenotypes (Fig. 6A). Staining of TMA cores revealed that basal-type breast cancers could be broadly divided into either UBR5 dominant or MYC dominant, or to cancers that showed both UBR5 or MYC-dominant cells (Fig. 6B). Supporting the IHC results, UBR5 was overall more highly expressed than MYC across all quantified 18,667 cells (Fig. 6C). Pairwise comparison of UBR5 and MYC intensities in all quantified individual cells demonstrated that in 64% of cells UBR5 were more highly expressed than MYC, whereas 27% of cells were MYC dominant (Fig. 6D). Fully consistent with a dichotomy of UBR5 and MYC expression (Supplementary Fig. S6E), only 9% of the cells displayed equal UBR5 and MYC protein expression intensities (yellow; Fig. 6D).

Figure 6.

UBR5 controls MYC protein expression at single-cell level in basal type breast cancer in vivo. A, Intensity of MYC and UBR5 protein from dual immunofluorescence (IF) staining was used to categorize individual cells in breast cancer tissues into either MYC or UBR5 dominant. B, Dual immunofluorescence staining of UBR5 and MYC in TMA of human basal-type breast cancer. C, Quantification of UBR5 and MYC staining at single-cell level from 18,667 cells by software FIJI. Error bars, SD. ****, P < 0.0001. D, Pie chart analysis of distribution of cells based on whether either MYC (green) or UBR5 (red) was expressed at higher level. Cells in which there was < 10% intensity difference between UBR5 and MYC were classified as having equal protein expression (yellow). E, Pairwise density distribution analysis of 18 667 individual cells revealed bimodal distribution of UBR5 or MYC-dominant cells. F, Regression analysis between UBR5 and MYC immunofluorescence intensities in individual cells (n = 18,667). G, Schematic presentation of UBR5-mediated control of MYC levels critical for proliferation and apoptosis regulation. MYC protein levels accumulate differently between UBR5high- and UBR5low-expressing cells. In UBR5low–expressing cells (blue area), MYC protein accumulates efficiently due to high protein stability, and high MYC protein–expressing cells exceed the apoptosis priming levels. In UBR5high-expressing cells (green area), UBR5 restrains MYC protein accumulation to the levels that maximally support proliferation, but do not exceed the levels that would result in apoptosis priming even with maximal c-Myc mRNA transcription rates.

Figure 6.

UBR5 controls MYC protein expression at single-cell level in basal type breast cancer in vivo. A, Intensity of MYC and UBR5 protein from dual immunofluorescence (IF) staining was used to categorize individual cells in breast cancer tissues into either MYC or UBR5 dominant. B, Dual immunofluorescence staining of UBR5 and MYC in TMA of human basal-type breast cancer. C, Quantification of UBR5 and MYC staining at single-cell level from 18,667 cells by software FIJI. Error bars, SD. ****, P < 0.0001. D, Pie chart analysis of distribution of cells based on whether either MYC (green) or UBR5 (red) was expressed at higher level. Cells in which there was < 10% intensity difference between UBR5 and MYC were classified as having equal protein expression (yellow). E, Pairwise density distribution analysis of 18 667 individual cells revealed bimodal distribution of UBR5 or MYC-dominant cells. F, Regression analysis between UBR5 and MYC immunofluorescence intensities in individual cells (n = 18,667). G, Schematic presentation of UBR5-mediated control of MYC levels critical for proliferation and apoptosis regulation. MYC protein levels accumulate differently between UBR5high- and UBR5low-expressing cells. In UBR5low–expressing cells (blue area), MYC protein accumulates efficiently due to high protein stability, and high MYC protein–expressing cells exceed the apoptosis priming levels. In UBR5high-expressing cells (green area), UBR5 restrains MYC protein accumulation to the levels that maximally support proliferation, but do not exceed the levels that would result in apoptosis priming even with maximal c-Myc mRNA transcription rates.

Close modal

To demonstrate statistical significance of UBR5 dominance over MYC at a single-cell level, we performed a pairwise density distribution analysis of UBR5 dominance from all 18,667 individual cells. To this end, MYC intensity in each individual cell was subtracted from UBR5 intensity from the same cell, and distribution of the resulting pairwise intensity difference was blotted on a scale that corresponded to actual immunofluorescence intensity levels (Fig. 6E; Supplementary Fig. S6H). The null hypothesis was that UBR5 and MYC expression would have been equal, and thus pairwise UBR5-MYC intensity difference would have centered around zero. The results, however, show that cells with either UBR5 dominance (UBR5 intensity-MYC intensity > 0), or MYC dominance (UBR5 intensity-MYC intensity < 0), constitute two different populations (Fig. 6E). Moreover, the median of UBR5 intensity-MYC intensity was strongly positive, and the density of UBR5-dominant cells is significantly higher than with MYC-dominant cells. Therefore, these results confirm that UBR5 dominates MYC protein expression in a majority of individual basal type breast cancer cells in vivo. This conclusion was further supported by regression analysis between UBR5 and MYC immunofluorescence intensities in individual cells that was clearly indicative of negative correlation between these proteins at the single-cell level (Fig. 6F). Interestingly, all cells having lower than approximately 30% of maximal UBR5 intensity had strong MYC expression (Fig. 6F), indicating that other mechanisms are not capable of effectively suppressing MYC in basal type breast cancers in the absence of UBR5. On the basis of our other results (Figs. 4K and l and 5E and F; Supplementary Fig. S5E and S5F), these UBR5low/MYChigh cells would constitute the irinotecan- and topotecan-sensitive cell population in vivo.

This first single-cell analysis of protein expression of MYC and its ubiquitin ligase demonstrate that UBR5 and MYC are exclusive for high level protein expression in most of the basal type breast cancer cells in vivo. As high UBR5 expression dominated high MYC expression in most of the cells, these results provide in vivo relevance for the role of UBR5 in negative regulation of MYC.

UBR5 regulates various cellular processes (19, 28, 49, 50), but the role for UBR5 in both inhibiting and promoting organismal growth has been enigmatic (19, 22, 23, 25, 45, 49, 51). The observed overgrowth in Drosophila epithelial tissue in Hyd-mutant flies via dMYC is fully consistent with the reported growth suppressor role of HYD (19, 25). On the other hand, UBR5-mediated suppression of MYC-induced apoptosis observed in cancer cell lines is well in accordance with previously reported oncogenic activities of UBR5 (22, 23). In support of in vivo relevance of these observations, we demonstrate both that UBR5 suppresses dMYC activity in wing development (Fig. 3), and that in most of the basal type breast tumor cells UBR5 dominates MYC protein levels at the single-cell level (Fig. 6CF; Supplementary Fig. S6E). Directly supportive of oncogenic role of UBR5 in human cancer cells in vivo, UBR5-depleted basal-like breast cancer cells with UBR5/MYC amplification failed to grow as xenograft tumors (Fig. 5C). Collectively, we conclude that UBR5 finetunes MYC protein expression to the levels that balance cell growth and survival (Fig. 6G).

Although the conclusions that lower levels of MYC would be more beneficial for the tumor are somewhat counterintuitive, they are fully in accordance with previous results using transgenic MYC mouse models. These studies have demonstrated the importance of MYC protein levels in controlling the delicate balance between proliferation and apoptosis (2, 3, 12). Whereas these previous studies convincingly demonstrated that the dose of MYC protein is critical to sustain maximal fitness of the cell, identification of endogenous candidate rheostat proteins such as UBR5, that define MYC levels in vivo, may help in future to better understand how the balance between MYC-mediated regulation of cell growth and apoptosis is defined by cell-intrinsic mechanisms. Notably, in addition to demonstrating both the in vitro and the in vivo relevance of UBR5–MYC axis, we also show that opposite to some of the other UBR5 functions that are strictly restricted to certain cell types (21, 51), we failed to identify a cell type in which UBR5 inhibition would have not increased MYC protein expression. Therefore, we anticipate these findings to have wide-reaching implications in different biological systems.

Coamplification with MYC introduces UBR5 as so far unique MYC ubiquitin ligase. On the basis of amplification frequencies of UBR5 and MYC, it is clear that UBR5 amplification alone does not provide significant benefit for breast cancer cells. However, also MYC amplification without UBR5 amplification was only seen in 28% of the samples with any amplifications, whereas coamplification was observed in 70% of all cases in which MYC was amplified (Table 1). Together with results that MYC/UBR5 coamplification frequency correlates with essentiality of UBR5 for breast cancer cell survival (Fig. 5B), and with UBR5-mediated protection of MYC/UBR5 coamplified breast cancer cells drug-induced apoptosis (Fig. 5E and F), we postulate that coamplification of UBR5 gives a survival benefit to MYC-amplified cells, for example, under drug-induced stress conditions. An additional discovery with high cancer relevance was p53 independence of MYC-mediated apoptosis in UBR5-inhibited cells. As cancer cells with combination of MYC amplification and p53 mutation constitute the most malignant cell population in many cancers, identification of a ubiquitin ligase regulating MYC-mediated apoptosis in this type of cells constitutes a major advancement with profound future potential. Intriguingly, opposite to breast cancers, UBR5 amplifications were not observed in lymphoid cancers. The discrepancy between solid and lymphoid cancers in regards to their UBR5 status is illustrated also by the fact that frequent UBR5 mutations (which are predicted to disrupt E3 ligase activity) are observed in mantle cell lymphoma (24). Therefore, it is possible that whereas UBR5 promotes oncogenesis in solid cancers by decreasing MYC-mediated apoptosis, in hematologic cancers UBR5 functions as a tumor suppressor, and this activity is lost by frequent loss-of-function mutations.

Even though the study did not directly address potential role of UBR5 as a cancer therapy target, our results proposes for a possible strategy to unleash MYC's proapoptotic activity for cancer therapy. On the basis of our data, we envision that UBR5 inhibition could harmonize the tumor cells to more uniformly express MYC protein at the levels that would predispose the cells to cell killing by cancer drugs. Optimally this could prevent reappearance of those cancer cell populations that were protected from the initial drug-induced apoptosis due to their low MYC expression. In addition to the presented data, this model is supported by recent demonstration of MYC as a major in vivo determinant of taxane response (17), and the data that pretreatment mitochondrial priming correlates with clinical response to cytotoxic chemotherapy in cancer (52). Unfortunately, testing of this hypothesis is currently unfeasible because no small-molecule inhibitors for UBR5 are available. Another potential clinical application of the data is use of breast cancer tissue UBR5low/MYChigh IHC status for patient stratification to topoisomerase I inhibitor therapies.

In summary, this study characterizes UBR5 as a novel MYC ubiquitin ligase controlling MYC activity in development of epithelial tissue in vivo, and in p53-mutant cancer cells. Collectively the results provide a novel molecular explanation for both the tumor-suppressive, as well as tumor-promoting roles of UBR5. Coamplification of UBR5 with MYC indicates for a concept that at certain circumstances it is beneficial for cancer cells to gain strong oncogenic activity, but simultaneously assure that this activity is kept in leash by coamplification of the natural suppressor mechanism.

No potential conflicts of interest were disclosed.

Conception and design: X. Qiao, Y. Liu, M.L. Prada, A. Meinander, V. Hietakangas, M. Eilers, J. Westermarck

Development of methodology: X. Qiao, Y. Liu, M.L. Prada, A. Gupta, K. Talvinen

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): X. Qiao, M.L. Prada, A.K. Mohan, M. Sharma, J. Merisaari, H.M. Haikala, P. Kronqvist

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): X. Qiao, Y. Liu, M.L. Prada, A. Gupta, A. Jaiswal, M. Sharma, H.M. Haikala, L. Yetukuri, J.W. Pylvänäinen, J. Klefström, A. Meinander, T. Aittokallio, M. Eilers, J. Westermarck

Writing, review, and/or revision of the manuscript: X. Qiao, Y. Liu, M.L. Prada, A. Gupta, L. Yetukuri, J. Klefström, P. Kronqvist, A. Meinander, T. Aittokallio, V. Hietakangas, M. Eilers, J. Westermarck

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): J. Klefström

Study supervision: A. Meinander, V. Hietakangas, J. Westermarck

The authors thank Prof. Johanna Ivaska and Prof. Lea Sistonen for providing valuable comments to the manuscript. The authors thank Prof. Dean Felsher for providing osteosarcoma-MYC-off cell line. Prof. Saunders, Watts Amati, Sears, and Tansey are thanked for plasmids and Peter Gallant for the anti-dMyc antibody. Birgit Samans is thanked for the support with the statistical analysis of the high-content siRNA screen image assay and Karolina Pavic for purification of recombinant MYC protein. We thank greatly Taina Kalevo-Mattila for excellent technical help, Sinikka Collanus for help in immunofluorescence and IHC, Samu Kurki, Eliisa Löyttyniemi, Srikar Nagelli, and Teemu Daniel Laajala for statistical analysis of IHC data. Personnel of Cell Imaging Core facility, Finnish Functional Genomics Centre, and Medical Bioinformatics Centre of Turku Bioscience Centre (supported by Biocenter Finland and Elixir-Finland) are all thanked for their valuable contributions. This study was supported by funding from Sigrid Juselius Foundation, Finnish Cancer Associations, Deutsche Forschungsgemeinschaft grant (Ei222/12-1), Academy of Finland (286767 and 312439), Foundation Martin Escudero, Finnish Cultural Foundation South-Western Regional Fund, Instrumentarium Science Foundation, Maud Kuistila Memorial Foundation, and Ella and Georg Enhrooth Foundation, K. Albin Johanssons Foundation, Paolo Foundation, and South-Western Finland Cancer Society.

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.
Mertins
P
,
Mani
DR
,
Ruggles
KV
,
Gillette
MA
,
Clauser
KR
,
Wang
P
, et al
Proteogenomics connects somatic mutations to signalling in breast cancer
.
Nature
2016
;
534
:
55
62
.
2.
McMahon
SB
. 
MYC and the control of apoptosis
.
Cold Spring Harb Perspect Med
2014
;
4
:
a014407
.
3.
Murphy
DJ
,
Junttila
MR
,
Pouyet
L
,
Karnezis
A
,
Shchors
K
,
Bui
DA
, et al
Distinct thresholds govern Myc's biological output in vivo
.
Cancer Cell
2008
;
14
:
447
57
.
4.
Sarosiek
KA
,
Fraser
C
,
Muthalagu
N
,
Bhola
PD
,
Chang
W
,
McBrayer
SK
, et al
Developmental regulation of mitochondrial apoptosis by c-myc governs age- and tissue-specific sensitivity to cancer therapeutics
.
Cancer Cell
2017
;
31
:
142
56
.
5.
Nieminen
AI
,
Eskelinen
VM
,
Haikala
HM
,
Tervonen
TA
,
Yan
Y
,
Partanen
JI
, et al
Myc-induced AMPK-phospho p53 pathway activates Bak to sensitize mitochondrial apoptosis
.
Proc Natl Acad Sci U S A
2013
;
110
:
E1839
48
.
6.
Thomas
LR
,
Tansey
WP
. 
Proteolytic control of the oncoprotein transcription factor Myc
.
Adv Cancer Res
2011
;
110
:
77
106
.
7.
Farrell
AS
,
Sears
RC
. 
MYC degradation
.
Cold Spring Harb Perspect Med
2014
;
4
.
DOI: 10.1101/cshperspect.a014365
.
8.
Myant
K
,
Qiao
X
,
Halonen
T
,
Come
C
,
Laine
A
,
Janghorban
M
, et al
Serine 62-phosphorylated MYC associates with nuclear lamins and its regulation by CIP2A is essential for regenerative proliferation
.
Cell Rep
2015
;
12
:
1019
31
.
9.
Welcker
M
,
Orian
A
,
Jin
J
,
Grim
JE
,
Harper
JW
,
Eisenman
RN
, et al
The Fbw7 tumor suppressor regulates glycogen synthase kinase 3 phosphorylation-dependent c-Myc protein degradation
.
Proc Natl Acad Sci U S A
2004
;
101
:
9085
90
.
10.
Yeh
E
,
Cunningham
M
,
Arnold
H
,
Chasse
D
,
Monteith
T
,
Ivaldi
G
, et al
A signalling pathway controlling c-Myc degradation that impacts oncogenic transformation of human cells
.
Nat Cell Biol
2004
;
6
:
308
18
.
11.
Levens
D
. 
You don't Muck with MYC
.
Genes Cancer
2010
;
1
:
547
54
.
12.
Pelengaris
S
,
Khan
M
,
Evan
GI
. 
Suppression of Myc-induced apoptosis in beta cells exposes multiple oncogenic properties of Myc and triggers carcinogenic progression
.
Cell
2002
;
109
:
321
34
.
13.
Trumpp
A
,
Refaeli
Y
,
Oskarsson
T
,
Gasser
S
,
Murphy
M
,
Martin
GR
, et al
c-Myc regulates mammalian body size by controlling cell number but not cell size
.
Nature
2001
;
414
:
768
73
.
14.
Grifoni
D
,
Bellosta
P
. 
Drosophila Myc: A master regulator of cellular performance
.
Biochim Biophys Acta
2015
;
1849
:
570
81
.
15.
Johnston
LA
,
Prober
DA
,
Edgar
BA
,
Eisenman
RN
,
Gallant
P
. 
Drosophila myc regulates cellular growth during development
.
Cell
1999
;
98
:
779
90
.
16.
Rohban
S
,
Campaner
S
. 
Myc induced replicative stress response: How to cope with it and exploit it
.
Biochim Biophys Acta
2015
;
1849
:
517
24
.
17.
Topham
C
,
Tighe
A
,
Ly
P
,
Bennett
A
,
Sloss
O
,
Nelson
L
, et al
MYC is a major determinant of mitotic cell fate
.
Cancer Cell
2015
;
28
:
129
40
.
18.
Varshavsky
A
. 
The N-end rule pathway and regulation by proteolysis
.
Protein Sci
2011
;
20
:
1298
345
.
19.
Shearer
RF
,
Iconomou
M
,
Watts
CK
,
Saunders
DN
. 
Functional roles of the E3 ubiquitin ligase UBR5 in cancer
.
Mol Cancer Res
2015
;
13
:
1523
32
.
20.
Shen
Q
,
Qiu
Z
,
Wu
W
,
Zheng
J
,
Jia
Z
. 
Characterization of interaction and ubiquitination of phosphoenolpyruvate carboxykinase by E3 ligase UBR5
.
Biol Open
2018
;
7
.
DOI: 10.1242/bio.037366
.
21.
Saunders
DN
,
Hird
SL
,
Withington
SL
,
Dunwoodie
SL
,
Henderson
MJ
,
Biben
C
, et al
Edd, the murine hyperplastic disc gene, is essential for yolk sac vascularization and chorioallantoic fusion
.
Mol Cell Biol
2004
;
24
:
7225
34
.
22.
Clancy
JL
,
Henderson
MJ
,
Russell
AJ
,
Anderson
DW
,
Bova
RJ
,
Campbell
IG
, et al
EDD, the human orthologue of the hyperplastic discs tumour suppressor gene, is amplified and overexpressed in cancer
.
Oncogene
2003
;
22
:
5070
81
.
23.
Liao
L
,
Song
M
,
Li
X
,
Tang
L
,
Zhang
T
,
Zhang
L
, et al
E3 ubiquitin ligase UBR5 drives the growth and metastasis of triple-negative breast cancer
.
Cancer Res
2017
;
77
:
2090
101
.
24.
Meissner
B
,
Kridel
R
,
Lim
RS
,
Rogic
S
,
Tse
K
,
Scott
DW
, et al
The E3 ubiquitin ligase UBR5 is recurrently mutated in mantle cell lymphoma
.
Blood
2013
;
121
:
3161
4
.
25.
Callaghan
MJ
,
Russell
AJ
,
Woollatt
E
,
Sutherland
GR
,
Sutherland
RL
,
Watts
CK
. 
Identification of a human HECT family protein with homology to the drosophila tumor suppressor gene hyperplastic discs
.
Oncogene
1998
;
17
:
3479
91
.
26.
Mansfield
E
,
Hersperger
E
,
Biggs
J
,
Shearn
A
. 
Genetic and molecular analysis of hyperplastic discs, a gene whose product is required for regulation of cell proliferation in Drosophila melanogaster imaginal discs and germ cells
.
Dev Biol
1994
;
165
:
507
26
.
27.
Jain
M
,
Arvanitis
C
,
Chu
K
,
Dewey
W
,
Leonhardt
E
,
Trinh
M
, et al
Sustained loss of a neoplastic phenotype by brief inactivation of MYC
.
Science
2002
;
297
:
102
4
.
28.
Gudjonsson
T
,
Altmeyer
M
,
Savic
V
,
Toledo
L
,
Dinant
C
,
Grofte
M
, et al
TRIP12 and UBR5 suppress spreading of chromatin ubiquitylation at damaged chromosomes
.
Cell
2012
;
150
:
697
709
.
29.
Henderson
MJ
,
Russell
AJ
,
Hird
S
,
Munoz
M
,
Clancy
JL
,
Lehrbach
GM
, et al
EDD, the human hyperplastic discs protein, has a role in progesterone receptor coactivation and potential involvement in DNA damage response
.
J Biol Chem
2002
;
277
:
26468
78
.
30.
Lee
JD
,
Amanai
K
,
Shearn
A
,
Treisman
JE
. 
The ubiquitin ligase hyperplastic discs negatively regulates hedgehog and decapentaplegic expression by independent mechanisms
.
Development
2002
;
129
:
5697
706
.
31.
Lee
T
,
Luo
L
. 
Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis
.
Neuron
1999
;
22
:
451
61
.
32.
Prober
DA
,
Edgar
BA
. 
Interactions between Ras1, dMyc, and dPI3K signaling in the developing drosophila wing
.
Genes Dev
2002
;
16
:
2286
99
.
33.
Welcker
M
,
Clurman
BE
. 
FBW7 ubiquitin ligase: a tumour suppressor at the crossroads of cell division, growth and differentiation
.
Nat Rev Cancer
2008
;
8
:
83
93
.
34.
Gregory
MA
,
Qi
Y
,
Hann
SR
. 
Phosphorylation by glycogen synthase kinase-3 controls c-myc proteolysis and subnuclear localization
.
J Biol Chem
2003
;
278
:
51606
12
.
35.
Sanchez-Arevalo Lobo
VJ
,
Doni
M
,
Verrecchia
A
,
Sanulli
S
,
Faga
G
,
Piontini
A
, et al
Dual regulation of Myc by Abl
.
Oncogene
2013
;
32
:
5261
71
.
36.
Herbst
A
,
Salghetti
SE
,
Kim
SY
,
Tansey
WP
. 
Multiple cell-type-specific elements regulate Myc protein stability
.
Oncogene
2004
;
23
:
3863
71
.
37.
Martin
P
,
Martin
A
,
Shearn
A
. 
Studies of l(3)c43hs1 a polyphasic, temperature-sensitive mutant of drosophila melanogaster with a variety of imaginal disc defects
.
Dev Biol
1977
;
55
:
213
32
.
38.
Freie
BW
,
Eisenman
RN
. 
Ratcheting Myc
.
Cancer Cell
2008
;
14
:
425
6
.
39.
Muthalagu
N
,
Junttila
MR
,
Wiese
KE
,
Wolf
E
,
Morton
J
,
Bauer
B
, et al
BIM is the primary mediator of MYC-induced apoptosis in multiple solid tissues
.
Cell Rep
2014
;
8
:
1347
53
.
40.
Seoane
J
,
Le
HV
,
Massague
J
. 
Myc suppression of the p21(Cip1) Cdk inhibitor influences the outcome of the p53 response to DNA damage
.
Nature
2002
;
419
:
729
34
.
41.
Sherr
CJ
. 
Tumor surveillance via the ARF-p53 pathway
.
Genes Dev
1998
;
12
:
2984
91
.
42.
Finch
A
,
Prescott
J
,
Shchors
K
,
Hunt
A
,
Soucek
L
,
Dansen
TB
, et al
Bcl-xL gain of function and p19 ARF loss of function cooperate oncogenically with Myc in vivo by distinct mechanisms
.
Cancer Cell
2006
;
10
:
113
20
.
43.
Hemann
MT
,
Bric
A
,
Teruya-Feldstein
J
,
Herbst
A
,
Nilsson
JA
,
Cordon-Cardo
C
, et al
Evasion of the p53 tumour surveillance network by tumour-derived MYC mutants
.
Nature
2005
;
436
:
807
11
.
44.
Klijn
C
,
Durinck
S
,
Stawiski
EW
,
Haverty
PM
,
Jiang
Z
,
Liu
H
, et al
A comprehensive transcriptional portrait of human cancer cell lines
.
Nat Biotechnol
2015
;
33
:
306
12
.
45.
O'Brien
PM
,
Davies
MJ
,
Scurry
JP
,
Smith
AN
,
Barton
CA
,
Henderson
MJ
, et al
The E3 ubiquitin ligase EDD is an adverse prognostic factor for serous epithelial ovarian cancer and modulates cisplatin resistance in vitro
.
Br J Cancer
2008
;
98
:
1085
93
.
46.
Mann
KM
,
Ward
JM
,
Yew
CC
,
Kovochich
A
,
Dawson
DW
,
Black
MA
, et al
Sleeping beauty mutagenesis reveals cooperating mutations and pathways in pancreatic adenocarcinoma
.
Proc Natl Acad Sci U S A
2012
;
109
:
5934
41
.
47.
Marcotte
R
,
Sayad
A
,
Brown
KR
,
Sanchez-Garcia
F
,
Reimand
J
,
Haider
M
, et al
Functional genomic landscape of human breast cancer drivers, vulnerabilities, and resistance
.
Cell
2016
;
164
:
293
309
.
48.
Dillon
JL
,
Mockus
SM
,
Ananda
G
,
Spotlow
V
,
Wells
WA
,
Tsongalis
GJ
, et al
Somatic gene mutation analysis of triple negative breast cancers
.
Breast
2016
;
29
:
202
7
.
49.
Sanchez
A
,
De Vivo
A
,
Uprety
N
,
Kim
J
,
Stevens
SM
 Jr.
,
Kee
Y
. 
BMI1-UBR5 axis regulates transcriptional repression at damaged chromatin
.
Proc Natl Acad Sci U S A
2016
;
113
:
11243
8
.
50.
Su
H
,
Meng
S
,
Lu
Y
,
Trombly
MI
,
Chen
J
,
Lin
C
, et al
Mammalian hyperplastic discs homolog EDD regulates miRNA-mediated gene silencing
.
Mol Cell
2011
;
43
:
97
109
.
51.
Kinsella
E
,
Dora
N
,
Mellis
D
,
Lettice
L
,
Deveney
P
,
Hill
R
, et al
Use of a conditional Ubr5 mutant allele to investigate the role of an N-end rule ubiquitin-protein ligase in hedgehog signalling and embryonic limb development
.
PLoS One
2016
;
11
:
e0157079
.
52.
Ni Chonghaile
T
,
Sarosiek
KA
,
Vo
TT
,
Ryan
JA
,
Tammareddi
A
,
Moore Vdel
G
, et al
Pretreatment mitochondrial priming correlates with clinical response to cytotoxic chemotherapy
.
Science
2011
;
334
:
1129
33
.