p53 acts as a safeguard of translational control by regulating fibrillarin and rRNA methylation in cancer

Abstract

Ribosomes are specialized entities that participate in regulation of gene expression through their rRNAs carrying ribozyme activity. Ribosome biogenesis is overactivated in p53-inactivated cancer cells, although involvement of p53 on ribosome quality is unknown. Here, we show that p53 represses expression of the rRNA methyl-transferase fibrillarin (FBL) by binding directly to FBL. High levels of FBL are accompanied by modifications of the rRNA methylation pattern, impairment of translational fidelity, and an increase of internal ribosome entry site (IRES)-dependent translation initiation of key cancer genes. FBL overexpression contributes to tumorigenesis and is associated with poor survival in patients with breast cancer. Thus, p53 acts as a safeguard of protein synthesis by regulating FBL and the subsequent quality and intrinsic activity of ribosomes.

Copyright © 2013 Elsevier Inc. All rights reserved.

Figures

Graphical abstract

Figure 1

p53 Regulates FBL Expression at both the mRNA and Protein Levels The expression of endogenous FBL in the indicated cell lines was analyzed at the mRNA level by RT-qPCR (A, C, and E) and at the protein level by western blot (B, D, and F). All graphs represent mean and SD of at least three experiments. ∗p < 0.05 and ∗∗p < 0.01 according to Student’s t test. See also Figure S1.

Figure 2

Modulation of p53 Expression Alters FBL Expression Endogenous FBL expression was analyzed in HME cells at the mRNA level by RT-qPCR (A and C) and at the protein level by western blot (B and D). The p53 expression is modulated by using an siRNA (A and B) or by treating or not (NT) with 2 μg/ml doxorubicin or 1 nM camptothecin (C and D). The p21 lanes in (D) were spliced together from discontinuous lanes of the same blot as indicated by dotted lines. All graphs represent mean and SD of at least three experiments. ∗p < 0.05 and ∗∗∗p < 0.001 according to Student’s t test. See also Figure S2.

Figure 3

p53 Regulates FBL Expression in Human Breast Cell Lines and Tumors (A) Quantification of FBL mRNA expression analyzed by RT-qPCR and normalized to RNA18S. (B) Quantification of FBL protein expression analyzed by western blot. All graphs represent mean and SD of at least three experiments. (C) Box-and-whisker plots of FBL mRNA expression quantification in wild-type p53 (n = 59) versus mutant p53 (n = 21) primary breast tumor samples. The bottom and the top of the boxes represent the 25th and 75th percentiles, respectively. The median values are visible as a line, the mean as a cross in the box and SD as error bars. The p value has been determined by a Mann-Whitney W test. See also Figure S3 and Table S1.

Figure 4

p53 Represses FBL Promoter Activity by Directly Binding to DNA (A) Alignment of the two putative p53 response elements (p53RE-1 and p53RE-2, black arrows) located within the intron 1 of FBL with the p53RE consensus (R, G/A; W, A/T; Y, C/T). n, spacer within p53RE consensus; dotted box, nucleotide region of the FBL gene cloned in the pFBL-Luc reporter vector; P1 and P2, primers pairs used in ChIP assays. (B and C) Luciferase reporter assays were performed in the absence of p53 (−) and in the presence of the wild-type (WT) or the indicated mutant p53 protein in HCT-116-p53−/− (B) and in HME-shp53 cells (C). Firefly luciferase activity is normalized to the renilla luciferase activity. Basic, luciferase reporter vector with no FBL sequence. (D and E) ChIP using an anti-p53 antibody and primer pairs P1 (D) or P2 (E) were performed in nontreated (NT) or 1 nM camptothecin-treated HME cells, or in HME-shp53 cells. All graphs represent mean and SD of at least three experiments. ∗p < 0.05 and ∗∗∗p < 0.001 according to Student’s t test. See also Figure S4.

Figure 5

p53 Regulates the rRNA Methylation Pattern and the Translational Fidelity of Ribosomes (A) The fold difference in rRNA methylation at 18 sites distributed throughout the 5.8S, 18S, and 28S rRNAs between HME-shp53 and HME cells were analyzed by RT-qPCR. (B–J) Translational fidelity was analyzed by transfecting cells with the pGL3mut1 vector (premature stop mutant, B–D), the pGL3mut2 vector (amino acid substitution mutant, E–G), or the SARS-CoV −1 programmed ribosome frameshift vector (H–J) in the indicated cells. (G) Translational fidelity was analyzed in nontransfected cells (NT) and after transfection of siRNA control (sc) or siRNA targeting FBL (si-FBL). FBL expression levels were verified by western blot (G, lower panel). All graphs represent mean and SD of at least three experiments. ∗p < 0.05 and ∗∗∗p < 0.001 according to Student’s t test. See also Figure S5.

Figure 6

p53 Regulates the IGF1R IRES-Dependent Translation (A and B) The IGF1R IRES-dependent translation initiation was determined by using luciferase bi-cistonic vectors in the indicated cells (A) and in cells after the downregulation of FBL by siRNA approach (B). (C) Analysis of the IGF1R IRES-dependent translation initiation in a panel of breast cell lines expressing either wild-type or mutant p53 proteins. (D) Typical polysomal profiles after fractionation of postmitochondrial supernatants from HME and HME-shp53 cells in a 10%–40% sucrose gradient are shown (upper). The distribution of the IGF1R mRNA within polysomes was determined by RT-qPCR (lower). (E and F) Endogenous IGF1R expression at mRNA (E, black bars) and protein levels (E, white bars; and F) was analyzed in HME and HMLE cells. All graphs represent mean and SD of at least three experiments. ∗p < 0.05, ∗∗p < 0.01 and ∗∗∗p < 0.001 according to Student’s t test. See also Figure S6.

Figure 7

Contribution of FBL Overexpression to Cancer Phenotype (A) Cell proliferation of three each independent stable MCF7 clones expressing FBL-GFP (F1, F2, and F3) or GFP (G1, G2, and G3). (B) Proliferation of the indicated cell clones not treated (NT) or treated with 1 μM Osi-906 for 72 hr. (C and D) Anchorage-independent growth of MCF7 clones using soft agar assay. Representative images are shown in (C) and the numbers of colonies determined in three experiments are shown in (D). (E) Impact of FBL overexpression on drug response was investigated by determining the IC50 of doxorubicin using MTS assays. Error bars represent SD. The p values have been determined by a Mann-Whitney W test. (F and G) Kaplan-Meier analysis of relapse-free survival rates (event = relapse) (F) and of breast cancer-specific survival rates (event = death related to breast cancer disease) (G) according to FBL mRNA level in primary breast tumor samples. The data are dichotomized at the upper quartile value into high and low expression groups. See also Figure S7 and Table S2.

Figure 8

Model of the Implication of p53 in the Control of Ribosomes Quantity and Ribosomes Quality, and Their Consequences on Translation In cells expressing functional p53 (top), p53 negatively regulates RNA Pol I activity to control ribosome quantity and FBL expression levels to control ribosome quality. This regulation would aim to coordinate the methylation of ribosomes and the rate of ribosome production according to the cell needs. These quality-controlled ribosomes allow a high translational fidelity together with a correct control of the balance between CAP- and IRES-dependent initiation of translation. In cells expressing a mutant or nonfunctional p53 (bottom), loss of the repression of RNA Pol I activity leads to an increase in rRNA synthesis. In parallel, p53 inactivation leads to an increase in FBL expression levels, resulting in a modification of the rRNA methylation patterns. Ribosomes with modified rRNA methylated translate mRNA with a lower fidelity (bypass of stop codon, amino acid misincorporation) and are more likely to initiate translation through IRES of mRNA coding for pro-oncogenic, anti-apoptotic, and survival proteins.