p53-Dependent induction of PVT1 and miR-1204

Abstract

p53 is a tumor suppressor protein that acts as a transcription factor to regulate (either positively or negatively) a plethora of downstream target genes. Although its ability to induce protein coding genes is well documented, recent studies have implicated p53 in the regulation of non-coding RNAs, including both microRNAs (e.g. miR-34a) and long non-coding RNAs (e.g. lincRNA-p21). We have identified the non-protein coding locus PVT1 as a p53-inducible target gene. PVT1, a very large (>300 kb) locus located downstream of c-myc on chromosome 8q24, produces a wide variety of spliced non-coding RNAs as well as a cluster of six annotated microRNAs: miR-1204, miR-1205, miR-1206, miR-1207-5p, miR-1207-3p, and miR-1208. Chromatin immunoprecipitation (ChIP), electrophoretic mobility shift assay (EMSA), and luciferase assays reveal that p53 binds and activates a canonical response element within the vicinity of miR-1204. Consistently, we demonstrate the p53-dependent induction of endogenous PVT1 transcripts and consequent up-regulation of mature miR-1204. Finally, we have shown that ectopic expression of miR-1204 leads to increased p53 levels and causes cell death in a partially p53-dependent manner.

FIGURE 1.

p53 stimulates transcription at the PVT1 locus. A, HCT116 or HCT116 p53^−/− cells were treated for 24 h with either nutlin-3a (10 μm) or daunorubicin (0.22 μm). Levels of p53, p21, and actin (loading control) were detected by immunoblotting. B, quantification of PVT1 ncRNA by qRT-PCR of HCT116 cells (hatched bars) and HCT116 p53^−/− cells (black bars) as treated in A. *, significant induction (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated HCT116 cells. C, quantification of PVT1 alternative exons 1A (hatched bars) and 1B (black bars) by qRT-PCR in HCT116 and HCT116 p53^−/− cells treated as in A. *, significant induction (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated HCT116 cells. D, SK-HEP1, WI38, RKO, and EW36 cells were treated for 24 h with either nutlin-3a (10 μm) or daunorubicin (0.22 μm). Quantification of PVT1 ncRNA by qRT-PCR following treatment with DMSO (hatched bars), nutlin (gray bars), or daunorubicin (black bars). *, significant induction (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated cells. Error bars, S.D.

FIGURE 2.

p53 directly binds to a canonical response element within the PVT1 locus in a sequence-specific manner. A, schematic diagram of the p53 RE in both the human and mouse PVT1 loci. Gray nucleotides in the human sequence represent deviations from the canonical p53-response element. Italicized nucleotides in the murine sequence denote changes from the human sequence. B, HCT116 cells were untreated (hatched bars) or treated for 8 h with 0.22 μm daunorubicin (black bars) and subjected to ChIP. qPCR was used to quantify DNA fragments co-immunoprecipitated with p53 or with beads alone (gray bars). Three primer pairs within the PVT1 locus were used to assay for p53 binding, −398, PVT1 RE, and exon 8. Two additional primer pairs at the CDKN1A/p21 locus, the 3′ p53 RE (−1391) and +11 kb (downstream of the CDKN1A/p21 gene), were used as positive and negative control, respectively. The extent of relative p53-binding to each genomic region is graphed. C, EMSA. An increasing amount of purified, FLAG-tagged p53 was incubated with labeled PVT1 RE probes. Such reactions were loaded onto 4% native polyacrylamide gels. Only the up-shifted probe is displayed. FLAG antibody was added into one reaction to ensure the specificity of p53-DNA complexes. D, competition EMSA. Excess unlabeled PVT1 RE (either wild type or mutant) probe was added in increasing amounts to reactions described in C. Error bars, S.D.

FIGURE 3.

p53 facilitates transcriptional activity from the PVT1 RE. Reporter constructs were engineered to contain either wild-type or mutant PVT1 RE that drive the transcription of the Firefly luciferase gene. A, empty vector or ectopic p53 were co-transfected with the luciferase reporter constructs in p53-null H1299 cells. Relative luminescence is quantified: empty vector (hatched bars) and p53 (black bars). *, significant induction of luciferase activity (p < 0.05, two-sided paired Student's t test) compared with H1299 cells transfected with empty vector. B, HCT116 (hatched bars) or HCT116 p53^−/− cells (black bars) were transfected with either wild-type or mutant PVT1 RE-containing luciferase constructs. Relative luminescence is quantified. *, significant induction of luciferase activity (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated HCT116 cells. Error bars, S.D.

FIGURE 4.

p53 induces primary and mature miR-1204. A, schematic of the human PVT1 locus that depicts that location of primary microRNAs relative to PVT1 exons. B, HCT116 or HCT116 p53^−/− cells were treated for 24 h with either nutlin-3a (10 μm) or daunorubicin (0.22 μm). Relative levels of the primary microRNA transcripts were quantified by qRT-PCR. The relative quantification of primiR-1204 (hatched bars), primiR-1205 (black bars with white dots), primiR-1206 (horizontal striped bars), primiR-1207 (gray bars), and primiR-1208 (black bars) is displayed. *, significant induction (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated HCT116 cells. C, HCT116 cells were treated with 0.22 μm daunorubicin. The microRNA-enriched RNA fraction was subjected to Qiagen miScript cDNA synthesis and qRT-PCR. Relative quantification of mature miR-1204 and miR-1207-5p is displayed and represents the -fold induction of each RNA species after daunorubicin treatment. *, significant difference (p < 0.05, two-sided paired Student's t test) between the induction ratio of miR-1204 and that of miR-1207-5p. D, SK-HEP1, WI38, RKO, and EW36 cells were treated for 24 h with either nutlin-3a (10 μm) or daunorubicin (0.22 μm). Shown is quantification of primiR-1204 by qRT-PCR following treatment with DMSO (hatched bars), nutlin (gray bars), or daunorubicin (black bars). *, significant induction (p < 0.05, two-sided paired Student's t test) compared with DMSO-treated cells. E, EW36 cells were treated for 24 h with daunorubicin (0.22 μm). Quantification of mature miR-1204, mature miR-1207-3p, and mature miR-34a (miScript PCR) is displayed. Error bars, S.D.

FIGURE 5.

Ectopic miR-1204 leads to increased p53 levels and causes cell death. A, HCT116 cells were transfected with 20 nm microRNA mimics. Cells were harvested 72 h post-transfection and subjected to propidium iodide staining and FACS analysis. The cell cycle profile following transfection of each mimic is shown. The percentage cell death, as quantified by sub-G₁ content, is graphed. *, significant induction of cell death (p < 0.05, two-sided paired Student's t test) compared with negative control-transfected HCT116 cells. B, HCT116 cells were transfected as in A. The levels of p53 and actin (loading control) were detected by immunoblotting. Quantification using the LI-COR Odyssey system is provided and represents the average -fold induction of p53 protein levels. Error bars, S.D. of multiple experiments. *, significant induction of p53 protein levels (p < 0.05, two-sided paired Student's t test) compared with the control microRNA mimic. C, HCT116 cells (hatched bars) or HCT116 p53^−/− cells (black bars) were transfected and subjected to FACS analysis as in A. The percentage cell death, as quantified by sub-G₁ content, is graphed. *, significant difference (p < 0.05, two-sided paired Student's t test) in the level of cell death between HCT116 cells and HCT116 p53^−/− cells.