Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Aug;33(16):3343-53.
doi: 10.1128/MCB.01213-12. Epub 2013 Jun 17.

The Corepressor CTBP2 Is a Coactivator of Retinoic Acid Receptor/Retinoid X Receptor in Retinoic Acid Signaling

Affiliations
Free PMC article

The Corepressor CTBP2 Is a Coactivator of Retinoic Acid Receptor/Retinoid X Receptor in Retinoic Acid Signaling

Prashanth Kumar Bajpe et al. Mol Cell Biol. .
Free PMC article

Abstract

Retinoids play key roles in development, differentiation, and homeostasis through regulation of specific target genes by the retinoic acid receptor/retinoid X receptor (RAR/RXR) nuclear receptor complex. Corepressors and coactivators contribute to its transcriptional control by creating the appropriate chromatin environment, but the precise composition of these nuclear receptor complexes remains to be elucidated. Using an RNA interference-based genetic screen in mouse F9 cells, we identified the transcriptional corepressor CTBP2 (C-terminal binding protein 2) as a coactivator critically required for retinoic acid (RA)-induced transcription. CTBP2 suppression by RNA interference confers resistance to RA-induced differentiation in diverse murine and human cells. Mechanistically, we find that CTBP2 associates with RAR/RXR at RA target gene promoters and is essential for their transactivation in response to RA. We show that CTBP2 is indispensable to create a chromatin environment conducive for RAR/RXR-mediated transcription by recruiting the histone acetyltransferase p300. Our data reveal an unexpected function of the corepressor CTBP2 as a coactivator for RAR/RXR in RA signaling.

Figures

Fig 1
Fig 1
A genome-wide RNAi screen identifies Ctbp2 as a component of the RA signaling pathway. (A) Schematic outline of the RA resistance screen performed in mouse F9 cells. Mouse shRNA library polyclonal virus was produced to infect F9 cells. The infected cells were treated with 1 μM all-trans-retinoic acid (RA) for 4 weeks, and shRNA inserts were recovered from the resistant colonies as described previously (25). (B) Analysis of the relative abundance of the recovered shRNA cassettes from the RA bar code experiment. Averaged data from three independent experiments were normalized and log2 transformed. Among the 24 top shRNA candidates (M > 4.0, A > 10.0), shCtbp2 and shRxrα were identified. shZfp423 (M = 9.3, A = 2.1) was also positively selected. (C) Individual shRNAs from the library targeting Rxrα, Ctbp2, and Zfp423 conferred RA resistance. F9 cells expressing shCtbp2, shRxrα, shZfp423, or the control, shGFP, were treated with 1 μM RA for 3 weeks, after which the cells were photographed, fixed, and stained. (D) Validation of independent shRNAs targeting Ctbp2. The functional phenotypes of nonoverlapping shRNAs targeting each gene are indicated by the colony formation assay in 1 μM RA. Lib, library. (E and F) The knockdown ability of each of the shRNAs was measured by examining the mRNA levels of the intended target gene by qRT-PCR (E) and the protein levels by Western blotting (F). Error bars represent standard deviations (SD) of triplicate independent experiments. (G) Proliferation curves according to the 3T3 protocol of F9 cells expressing shRNAs targeting Ctbp2 or GFP in the absence and presence of RA (1 μM). Error bars represent SD of triplicate independent experiments.
Fig 2
Fig 2
Ctbp2 is also required for RA-induced differentiation in mouse ES cells. (A) Downregulation of Ctbp2 by RNAi in E14T mouse ES cells confers resistance to RA-induced differentiation. E14T cells expressing shRNAs against Ctbp2, Rxrα, or GFP (control) were grown in the absence or presence of RA (1 μM) for 1 week, after which cells were fixed, stained for alkaline phosphatase (AP), and photographed. Bars, 50 mm. (B) mRNA levels of Ctbp2and Rxrα in E14T mouse ES cells expressing shRNAs targeting Ctbp2, Rxrα, or GFP. Error bars denote SD. (Control images reprinted from reference with permission of the publisher.)
Fig 3
Fig 3
Ctbp1 is not required for RA-induced differentiation. (A) Two different nonoverlapping shRNAs against Ctbp1 did not confer resistance in the presence of 1 μM RA in a colony formation assay. Two independent shRNAs against Ctbp2 conferred resistance. The cells were plated at low density and grown for 3 weeks, after which the cells were photographed, fixed, and stained. Ctrl, control. (B) The knockdown ability of each of the shRNAs against Ctbp1 and Ctbp2 was measured by examining the mRNA levels of the intended target gene by qRT-PCR. Error bars denote SD.
Fig 4
Fig 4
CTBP2 is a transcriptional cofactor for RXRα/RARα. (A to C) Activation of a RARE-luciferase (RARE-Luc) reporter gene by RXRα/RARα requires CTBP2. The normalized luciferase activities shown represent ratios between luciferase values and Renilla internal control values and are the averages ± SD from three independent transfections. (A) Downregulation of Ctbp2 by RNAi inhibited the transcriptional induction of the RARE-Luc reporter gene in mouse F9 cells in response to 24 h of 1 μM RA stimulation. (B) Suppression of CTBP2 by RNAi inhibited the activation of the RARE-Luc reporter gene in human NTERA2 cells in response to 1 μM RA treatment. (C) Overexpression of human CTBP2, mouse Ctbp2, and human p300 histone acetyltransferase was able to hyperactivate the luciferase expression induced by 1 μM RA in the NTERA2 cell line. (D to F) Ctbp2 is required for transcriptional regulation of endogenous RA target genes in response to RA. Shown are the results from mRNA expression analysis of RA target genes Crabp2 (D), Cyp26a1 (E), and Hoxb5 (F) in F9 cells expressing shRNAs targeting Ctbp2, Rxrα, or GFP cultured in the absence or presence of 1 μM RA for 48 h. Error bars represent SD of triplicate independent experiments. (G) Pie chart showing the results from the RNA sequencing analysis in F9 shGFP control cells cultured in the absence or presence of RA for 48 h. Twenty-five percent (2,754) of the genes are upregulated in an RA-dependent manner, 14% (1,518) of the genes are downregulated in an RA-dependent manner, and 61% (6,568) of the genes remain unchanged. (H and I) Effect of Ctbp2 and Rxr knockdown on RA-upregulated genes by whole-genome transcriptome analysis in F9 cells. A bar chart shows that 52% of the upregulated genes are dependent on Ctbp2 (H) and 55% of the genes are dependent on Rxr (I) for transcription. (J) Ctbp2-dependent genes for activation and repression in the absence of RA (J). (K) Representation of the major bona fide RA target genes and their dependence on Ctbp2 and Rxrα for activation. (See Materials and Methods for details.) (L and M) Venn diagram representation of the overlap of genes between Ctbp2 and Rxr dependency (L). Shown is the overlap (228 genes) of Ctbp2-dependent genes for activation (711 genes) with the RA-dependent Ctbp2-dependent genes for activation (1,423 genes) (M).
Fig 5
Fig 5
CTBP2 is also involved in RA-induced differentiation of human promyelocytic leukemia cells. (A) mRNA expression of CTBP2 and RXRα in HL-60 cells expressing pRS control or vectors targeting CTBP2 or RXRα was examined by qRT-PCR. Error bars denote SD. (B and C) CTBP2 is required for transcriptional induction of RA target genes in HL-60 cells in response to RA. Shown are the results from mRNA expression analysis of CYP26A1 (B) and RARβ (C) in HL-60 cells expressing the pRS control or vectors targeting CTBP2 or RXRα cultured in the absence or presence of 1 μM RA for 48 h. Error bars denote SD. (D) CTBP2 is involved in RA-induced differentiation in human promyelocytic leukemia cells. Shown are results from HL-60 cells expressing the pRS control or vectors targeting CTBP2 or RXRα cultured in the absence or presence of 1 μM RA for 48 h. Flow cytometric analysis of CD11b expression was used to assess differentiation.
Fig 6
Fig 6
CTBP2 physically associates with RARα/RXRα and the RARE region of RA target gene promoters. (A) In vitro interaction of CTBP2 and RXRα. A GST pulldown assay was used to determine the binding between GST fusions of RARα or RXRα or PPARγ with 35S-labeled CTBP2. 35S-labeled CTBP2 or RXRα was incubated with equal amounts of GST fusion proteins with and without all-trans-RA at 4°C for 2 h in buffer containing 25 mM HEPES-KOH (pH 7.6), 0.1 mM EDTA, 12.5 mM MgCl2, 10% glycerol, 0.1% NP-40, and 150 mM KCl. After 5 washes, bound proteins were subjected to SDS-PAGE followed by autoradiography. Note that CTBP2 interacts significantly more with GST-RXRα. (B) In vivo interaction of CTBP2 and RXRα complex in mouse F9 cells. RXRα was immunoprecipitated (46) from F9 total cell lysates using RXR-specific antibodies, and coimmunoprecipitated proteins were immunoblotted for RXRα and CTBP2. Note that RXRα coimmunoprecipitated CTBP2 and normal rabbit IgG did not. (C) In vivo interaction of CTBP2 and RXRα was further confirmed by reverse immunoprecipitation. CTBP2 was immunoprecipitated using CTBP2 antibodies, and coimmunoprecipitated proteins were immunoblotted for CTBP2 and RXRα. (D and E) CTBP2 is associated with the RAREs of two different RA target gene promoters. Chromatin immunoprecipitation (ChIP) was followed by qPCR analysis of CTBP2 binding to RAREs of Cyp26a1 and Hoxb5. Note the loss of CTBP2 binding in Ctbp2 knockdown cells. The data presented here are shown after subtraction of IgG background and are the average of four PCRs from two independent chips. Error bars denote SD.
Fig 7
Fig 7
CTBP2 recruits p300 histone acetyltransferase to RA target genes. (A to C) Chromatin immunoprecipitation (ChIP) followed by qPCR analysis of p300 binding to RAREs of Crabp2, Hoxb5, and Cyp26a1 in control knockdown (shCtr) and Ctbp2 knockdown (shCtbp2) cells. (D to F) ChIP followed by qPCR analysis of H3 acetylation on RAREs of Crabp2, Hoxb5, and Cyp26a1 in control knockdown (shCtr) and Ctbp2 knockdown (shCtbp2) cells. Note that loss of Ctbp2 leads to concomitant loss of p300 (A to C) and H3 (K9 and K14) acetylation (D to F). The data presented here are shown after subtraction of IgG background and are the average of four PCRs from two independent chips, and error bars denote SD. (G to J) mRNA expression analysis of Cbp (G) and RA target genes Cyp26a1 (H) and Crabp2 (I) and RARβ (J) in F9 cells expressing shRNAs targeting Cbp, Ctbp2, or empty vector cultured in the absence or presence of 1 μM RA for 48 h. Error bars represent SD of triplicate independent experiments.

Similar articles

See all similar articles

Cited by 6 articles

See all "Cited by" articles

Publication types

MeSH terms

Feedback