Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 26 (6), 2360-72

Targeting of C-terminal Binding Protein (CtBP) by ARF Results in p53-independent Apoptosis

Affiliations

Targeting of C-terminal Binding Protein (CtBP) by ARF Results in p53-independent Apoptosis

Seema Paliwal et al. Mol Cell Biol.

Abstract

ARF encodes a potent tumor suppressor that antagonizes MDM2, a negative regulator of p53. ARF also suppresses the proliferation of cells lacking p53, and loss of ARF in p53-null mice, compared with ARF or p53 singly null mice, results in a broadened tumor spectrum and decreased tumor latency. To investigate the mechanism of p53-independent tumor suppression by ARF, potential interacting proteins were identified by yeast two-hybrid screen. The antiapoptotic transcriptional corepressor C-terminal binding protein 2 (CtBP2) was identified, and ARF interactions with both CtBP1 and CtBP2 were confirmed in vitro and in vivo. Interaction with ARF resulted in proteasome-dependent CtBP degradation. Both ARF-induced CtBP degradation and CtBP small interfering RNA led to p53-independent apoptosis in colon cancer cells. ARF induction of apoptosis was dependent on its ability to interact with CtBP, and reversal of ARF-induced CtBP depletion by CtBP overexpression abrogated ARF-induced apoptosis. CtBP proteins represent putative targets for p53-independent tumor suppression by ARF.

Figures

FIG. 1.
FIG. 1.
ARF interacts with CtBP. (A) Schematic representation of CtBP functional domains and the ARF binding region at the C terminus, as identified by a two-hybrid screen. The HIPK2 phosphorylation (P) site (S428) is noted. (B) CtBP interacts with ARF in vitro. GST or GST-mArf fusion proteins were conjugated to glutathione-agarose beads and incubated with U2OS cell lysates. Bound, endogenous CtBP2 was assayed by Western blotting. Input lane shows 10% of the cell lysate. GST and GST-mArf migration positions are indicated by arrows. (C) The CtBP C terminus interacts with ARF. V5-tagged full-length (Fl), N-terminal (1-321), and C-terminal (322-445) hCtBP2 proteins were expressed in U2OS cells. Binding of V5-hCtBP2 proteins to GST versus GST-mArf was assayed by immunoblotting of GST pulldowns. Arrows indicate the migration positions of the wild type and deletion mutants of V5-hCtBP2 as well as GST-mArf. (D) The CtBP C terminus interacts with ARF. S-tagged C-terminal (224-440) or N-terminal (1-350) hCtBP2 fusion proteins or unfused S-tag protein were expressed in E. coli, purified with S-protein agarose, and incubated with a lysate of mArf-expressing U2OS cells. Bound mArf and the presence of S-tag fusion proteins were determined by immunoblotting.
FIG. 2.
FIG. 2.
In vivo ARF interaction with CtBP. (A) Exogenous CtBP interacts with exogenous ARF in transfected cells. U2OS cells were transfected with V5-hCtBP2 and mArf expression plasmids and UV or mock irradiated (10 J/m2) 24 h after transfection. Six hours after UV exposure, cell lysates were immunoprecipitated with anti-ras (control), anti-ARF, or anti-V5 antibody, followed by Western blot analysis with anti-V5, anti-CtBP1, or anti-ARF antibody. Input lane depicts 10% of the transfected cell lysate. (B) hARF interacts with CtBP. U2OS cells were transfected with V5-hCtBP2 and myc-hARF expression plasmids. Immunoprecipitation was performed with V5 or control antibody (Ab), followed by immunoblotting for hARF and V5. (C) (Middle and right panels) In vivo coimmunoprecipitation of mArf and mCtBP2 from lysates of wt, p53-null, or p53/mArf-null MEFs. Immunocomplexes were analyzed for the presence of endogenous ARF and mCtBP2 by Western blotting. (Left panel) 10% of input protein lysates. (D) CtBP2 and hARF colocalize. U2OS cells were transfected with hARF, V5-CtBP2, or both expression plasmids. Twenty-four hours after transfection, cells were immunostained with rabbit anti-hARF and mouse anti-hCtBP2 antibodies, followed by detection with fluorescein isothiocyanate-labeled (anti-rabbit) or rhodamine-labeled (anti-mouse) secondary antibody. Arrows indicate cells that were transfected, as noted by a CtBP2 level increased above that seen with adjacent untransfected cells. The merged image highlights relocalization of CtBP2 to nuclear bodies consistent with nucleoli, where hARF is found. Asterisks indicate untransfected cells.
FIG. 3.
FIG. 3.
The CtBP and MDM2 interaction domains of ARF are distinct. (A) (Left) ARF schematic showing MDM2 binding and nucleolar localization (NoLS) domains and a third conserved ARF domain of unknown function with human/mouse ARF amino acid sequence alignment (top). Conserved residues within this domain are set in boldface type. (Right) Lysates of U2OS cells transfected with wt or mutant mArf expression plasmids were immunoprecipitated with CtBP2 antibody, followed by analysis of the IPs by anti-ARF immunoblotting. Control IP was performed with anti-ras antibody using lysates of cells transfected with full-length mArf. (B) Mutations within the CtBP-binding region of mArf do not disrupt MDM2 interaction. U2OS cells were transfected with the indicated mArf expression plasmids and lysates immunoprecipitated with anti-ARF antibody (AEC40), followed by MDM2 and ARF immunoblotting. Numbers at left are molecular size markers (in kDa). vec, vector. (C) Mutations within the CtBP-binding region of mArf do not disrupt nucleolar localization. U2OS cells were transfected with the indicated mArf expression plasmids, and transfected cells were immunostained with ARF antibody (AEC40). All mArf alleles, except Δ8-32, which lacks the NoLS (amino acids 26 to 37), properly localize to nuclear bodies consistent with the appearance of nucleoli. (D) mArf mutations within the CtBP-binding region do not disrupt inhibition of MDM2-mediated p53 degradation. U2OS cells were transfected with HA-p53, MDM2, and the indicated mArf expression plasmids, followed by immunoblotting of transfected cell lysates with HA and ARF antibodies. (E) Mutations within the CtBP-binding region of mArf do not grossly disrupt induction of G1 arrest. Mouse 3T3-D1 cells (p53 wt) were transfected with the indicated mArf expression plasmids and GFP, and cell cycle profiles of GFP-gated cells were analyzed by propidium iodide staining and FACS analysis 48 h after transfection. A representative experiment is shown; the experiment was repeated three times with similar results.
FIG. 4.
FIG. 4.
UV induced CtBP degradation by the proteasome correlates with cellular ARF status. (A to D) MEFs of various mArf and p53 statuses were exposed to increasing doses of UV-C (0 to 30 J/m2). Levels of endogenous mCtBP2 6 h after UV treatment were determined by Western blotting. CtBP2 levels were quantitated by densitometry and normalized to a GAPDH loading control. (E) mCtBP2 is degraded by the proteasome in response to UV. MEFs (mArf + p53+) were incubated with or without proteasome inhibitor (MG132) for 24 h after mock or UV (10 J/m2) treatment. Cell lysates were analyzed for changes in CtBP2 level by Western blotting, followed by densitometry normalized to a GAPDH loading control. Experiments were repeated a minimum of three times with similar results, and data from representative experiments are shown in panels A to E. Numbers below blots represent GAPDH-normalized relative levels of CtBP2.
FIG. 5.
FIG. 5.
ARF causes CtBP degradation. (A) U2OS (hARF-null) cells were transfected with vector control or full-length mArf expression plasmids, followed by mock or UV treatment (UV-C at 0 to 30 J/m2), and CtBP2, hARF, and GAPDH levels were determined by immunoblotting 6 h after UV treatment. CtBP2 levels were quantitated by densitometry and normalized to GAPDH. (B) Degradation of endogenous CtBP by mArf mutants. Indicated ARF expression plasmids were transfected into HCT116 p53−/− cells along with pCD-GFP and treated with UV-C at 10 J/m2 24 h after transfection. Transfected cells were sorted for GFP 6 h after UV treatment, lysed, and analyzed by immunoblotting for CtBP2. CtBP2 levels were quantitated by densitometry and normalized to GAPDH. (C) hARF causes CtBP loss without stress. Lysates of HCT116 and HCT116 p53−/− cells obtained 24 h postinfection with Ad-LacZ or Ad-hARF were immunoblotted with ARF, CtBP2, or GAPDH specific antibody. (D) ARF causes CtBP1 loss. Lysates of HCT116 p53−/− cells infected with Ad-LacZ or Ad-hARF, as described for panel C, were assayed for ARF, CtBP1, and GAPDH expression by immunoblotting. (E) hARF-induced CtBP2 degradation requires an intact CtBP recognition domain. HCT116 p53−/− cells were transfected with vector, hARF, or hARF(L50D) mutant expression plasmids, followed by sorting for GFP 24 h after transfection. Cell lysates of GFP-expressing cells were analyzed by hCtBP2, GAPDH, and hARF immunoblotting. (F) hARF does not affect hCtBP2 mRNA level. Semiquantitative (18 cycles) reverse transcriptase PCR of mRNA prepared from HCT116 p53−/− cells infected with Ad-LacZ or Ad-hARF was carried out using CtBP2 and GAPDH specific primers. Numbers below blots in panels A and B represent GAPDH-normalized relative levels of CtBP2.
FIG. 6.
FIG. 6.
ARF destabilizes CtBP2 without altering its ubiquitination. (A) hARF causes hCtBP2 destabilization. Twenty-four hours after Ad-LacZ or Ad-hARF infection, HCT116 p53−/− cells were pulse-labeled with [35S]methionine, followed by chase in unlabeled DMEM. At indicated times during the chase, aliquots of cells were lysed and run directly on SDS-PAGE (not shown) or immunoprecipitated with anti-hCtBP2, followed by SDS-PAGE and autoradiography. Relative hCtBP2 levels (numbers below autoradiogram) were quantitated by densitometry and normalized against [35S]methionine incorporation as assayed for each sample by SDS-PAGE and autoradiography of total [35S]methionine-labeled proteins. (B) Effect of hARF on CtBP2 ubiquitination. HCT116 p53−/− cells were transfected with CtBP2 and HA-ubiquitin (HA-Ub) expression plasmids, and 16 h later, transfected cells were infected with Ad-LacZ or Ad-hARF. Twenty-four hours after infection, lysates were immunoprecipitated with anti-HA agarose, followed by CtBP2 immunoblotting. Ub-CtBP2 indicates the migration position of ubiquitinated CtBP2 species. Numbers at left are molecular size markers (in kDa).
FIG. 7.
FIG. 7.
ARF expression or CtBP depletion causes p53-independent apoptosis. (A) ARF induces p53-independent apoptosis. Twenty-four hours after infection with Ad-LacZ or Ad-hARF, HCT116 wt or p53-null cells were labeled with fluorescent caspase 3/7 substrate and propidium iodide and assayed by FACS analysis. The percentages of live cells with active caspase 3 in each sample were plotted. A representative experiment is shown, and similar results were seen with four separate repetitions. (B) CtBP depletion induces apoptosis. HCT116 wt or p53-null cells were treated with siRNA duplexes complementary to either hCtBP2 or hCtBP1. Specific depletion of hCtBP1 or hCtBP2 and identification of PARP and caspase 3 cleavage products were determined by immunoblotting. (C) hARF expression or CtBP siRNA reduces cell viability in the absence of p53. Percent viable cells, as determined by trypan blue exclusion (mean from three independent experiments), was plotted for CtBP knockdown compared to control siRNA treatment and for Ad-hARF infection compared to Ad-LacZ infection in HCT116 p53−/− cells. Error bars indicate 1 standard deviation. scr, scrambled.
FIG. 8.
FIG. 8.
hARF-induced p53-independent apoptosis requires CtBP2 depletion and interaction. (A) CtBP2 overexpression can revert hARF-induced apoptosis in HCT116 cells. (Left) HCT116 p53−/− cells were transfected with control vector or CtBP2 expression plasmids, and 24 h later, transfected cells were infected with Ad-LacZ or Ad-hARF. Twenty-four hours after infection, the cells were incubated with annexin V-PE along with 7-amino-actinomycin (7-AAD) and analyzed by flow cytometry. Percentages of annexin-positive, viable (7-AAD-negative) cells are displayed. A representative experiment is shown, and similar results were seen in three separate repetitions. (Right) Immunoblots for CtBP2, hARF, and GAPDH expression in the transfected cells. Arrows indicate migration positions of endogenous CtBP2 and exogenous V5-CtBP2. (B) hARF/CtBP interaction is required for the induction of apoptosis in p53-null HCT116 cells. (Left) HCT116 p53−/− cells were transfected with control vector, hARF, or L50D mutant expression plasmids, and 72 h after transfection, cells were incubated with annexin V-PE along with 7-AAD and analyzed by flow cytometry. Percentages of annexin-positive, viable (7-AAD-negative) cells are displayed. A representative experiment is shown, and similar results were seen in three separate repetitions. (Right) Immunoblots demonstrating CtBP2, hARF, and GAPDH expression in the transfected cells.

Similar articles

See all similar articles

Cited by 39 PubMed Central articles

See all "Cited by" articles

Publication types

MeSH terms

LinkOut - more resources

Feedback