Oncogene (2014), 1–11 © 2014 Macmillan Publishers Limited All rights reserved 0950-9232/14 www.nature.com/onc

ORIGINAL ARTICLE

A mutant p53/let-7i-axis-regulated gene network drives cell migration, invasion and metastasis M Subramanian1, P Francis1, S Bilke1, XL Li1, T Hara1, X Lu2, MF Jones1, RL Walker1, Y Zhu1, M Pineda1, C Lee3, L Varanasi4, Y Yang5, LA Martinez4, J Luo3, S Ambs6, S Sharma2, LM Wakefield5, PS Meltzer1 and A Lal1 Most p53 mutations in human cancers are missense mutations resulting in a full-length mutant p53 protein. Besides losing tumor suppressor activity, some hotspot p53 mutants gain oncogenic functions. This effect is mediated in part, through gene expression changes due to inhibition of p63 and p73 by mutant p53 at their target gene promoters. Here, we report that the tumor suppressor microRNA let-7i is downregulated by mutant p53 in multiple cell lines expressing endogenous mutant p53. In breast cancer patients, significantly decreased let-7i levels were associated with missense mutations in p53. Chromatin immunoprecipitation and promoter luciferase assays established let-7i as a transcriptional target of mutant p53 through p63. Introduction of let-7i to mutant p53 cells significantly inhibited migration, invasion and metastasis by repressing a network of oncogenes including E2F5, LIN28B, MYC and NRAS. Our findings demonstrate that repression of let-7i expression by mutant p53 has a key role in enhancing migration, invasion and metastasis. Oncogene advance online publication, 24 March 2014; doi:10.1038/onc.2014.46

INTRODUCTION The p53 tumor suppressor is a transcription factor that regulates the expression of hundreds of genes1 and inhibits neoplastic transformation by inducing cell cycle arrest, senescence and apoptosis.2,3 The TP53 gene is frequently (>50%) mutated in human cancers through missense mutations that result in the expression of a full-length mutant p53 protein that accumulates at high levels in cancer cells. Most missense mutations in the TP53 gene occur in the DNA-binding domain of p53, and among these are six ‘hotspot’ mutations at residues R175, G245, R248, R249, R273 and R282 that occur at a markedly high frequency.4 These diverse mutations result in a p53 protein with weakened or abrogated transactivation function resulting in loss of tumor suppressor activity. However, it is becoming increasingly clear that several p53 mutants, including the hotspot mutants R175H and R273H often acquire novel oncogenic functions and promote invasion and metastasis when introduced into p53-null cells, suggesting gain-of-function activities of mutant p53.5–9 Mutant p53 exerts its gain of function, in part, through interactions with several proteins including PIN1 and TOPBP1.10,11 Although mutant p53 proteins, in general, exhibit diminished DNA-binding activity, they can drive gene expression by binding to other transcription factors including NF-Y, VDR, E2F1, ETS2 and the p53 family members p63 and p73.12–15 Among these, the transcription factors p63 and p73 are the most widely studied mutant p53-interacting proteins. Mutant p53 binds to p63 and p73 at their target gene promoters to antagonize their activities.3,7,16–18 Despite the growing list of protein-coding genes whose transcription is inhibited by mutant p53, little is known about the regulation of microRNAs (miRNAs) by mutant p53. MiRNAs are an abundant

class of small non-coding RNAs (~22 nucleotides (nt)) transcribed by RNA polymerase II as long primary transcripts (pri-miRNAs). Mammalian miRNAs bind to the 3ʹ untranslated region (UTR) of target mRNAs via partial complementarity to inhibit translation and/or promote mRNA decay,19–21 with far-reaching regulatory effects. Target recognition involves base pairing between the miRNA seed region (nt 2–7 at the 5ʹ end of the miRNA) and the target mRNA.19,21 However, a perfect seed match is not necessary for gene suppression by miRNAs.22,23 Global downregulation of miRNAs has been often observed in human cancers, suggesting that a majority of miRNAs function as tumor suppressors. Although most miRNAs act to fine-tune target gene expression, some, including let-7 and miR-34a, act as master regulators of important biological processes.24,25 Recent studies have demonstrated a close relationship between p53 and miRNAs.26 In response to DNA damage, wild-type p53 directly induces the transcription of some miRNAs including miR-34a.27,28 However, not much is known about the involvement of miRNAs in mutant p53 gain of function. In the present study we have investigated the effects of mutant p53 on miRNA expression. Our findings suggest that a key mechanism by which mutant p53 promotes migration, invasion and metastasis is through downregulation of let-7i, resulting in derepression of a network of oncogenic let-7i target genes. RESULTS Mutant p53 regulates the expression of several miRNAs including let-7i To identify mutant p53-regulated miRNAs, we sequenced small RNAs from the p53-null H1299 cells (lung cancer) stably

1 Genetics Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA; 2Department of Biochemistry and Molecular Biology, College of Medicine, Howard University, Washington, DC, USA; 3Medical Oncology Branch, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA; 4Department of Biochemistry, University of Mississippi Cancer Institute, University of Mississippi Medical Center, Jackson, MS, USA; 5Laboratory of Cancer Biology and Genetics, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA and 6Laboratory of Human Carcinogenesis, National Cancer Institute, National Institutes of Health, Bethesda, MD, USA. Correspondence: Dr A Lal, Regulatory RNAs and Cancer Section, Genetics Branch, National Cancer Institute, National Institutes of Health (NIH), Building 37, Room 6134, 37 Convent Drive, Bethesda, MD 20892, USA. E-mail: [email protected] Received 18 July 2013; revised 18 November 2013; accepted 24 December 2013

Role of let-7i in mutant p53 driven cancer M Subramanian et al

2 transfected with empty vector (EV) or the hot-spot aggressive mutant p53R273H (Supplementary Figure S1A). With an arbitrary cutoff of 1.5-fold (⩾100 reads), 38 miRNAs were upregulated and 3 were downregulated in H1299-p53R273H cells (Figure 1a, Supplementary Tables S1 and S2). The oncogenic miR-155 was the most highly upregulated (~27-fold) and the tumor suppressor let7i was significantly downregulated (~1.6-fold). To validate these results, we performed TaqMan miRNA qRT-PCR and observed consistent upregulation (~1.5- and ~ 10-fold) of miR-20b and miR-155, whereas let-7i levels significantly declined (approximately twofold) in H1299-p53R273H cells (Figure 1b). The abundance of let-7i primary transcript (pri-let-7i) significantly decreased (approximately twofold), whereas the transcript encoding the miR-155 host gene (miR-155-HG) was upregulated (approximately twofold) by p53R273H (Figure 1c), indicating transcriptional regulation. The observed upregulation of miR-155 by mutant p53 has also been reported recently.29 Although miR-155 was upregulated >10-fold, the number of reads of miR-155 (13 in H1299-EV and 356 in H1299-p53R273H cells) suggested that it was not abundant in these cells. On the other hand, the number of reads of let-7i (8623 in H1299-EV and 5323 in H1299-p53R273H cells) indicated robust let-7i expression. Indeed, let-7 is highly expressed in many tissues including lung.30 As a single miRNA can regulate hundreds of genes, abundant miRNAs are likely to be more potent regulators of gene expression. We therefore hypothesized that let-7i has a role in mutant p53 gain of function and focused on understanding the regulation and function of let-7i downstream of mutant p53. To further validate let-7i as a mutant p53 target, we transiently transfected H1299 cells with EV, wild-type p53 or p53R273Hexpressing constructs for 48 h and observed consistent downregulation of let-7i (approximately twofold) by p53R273H (Figure 1d). Although, miR-34a, a well-established direct target of wild-type p53,27 was upregulated by wild-type p53, let-7i was downregulated only by p53R273H, indicating that downregulation of let-7i was not due to residual wild-type p53 activity (Figure 1d). The expression levels of wild-type p53 and p53R273H were similar in these experiments (Figure 1e). Moreover, let-7i was repressed by mutant p53 in isogenic DLD-1 cells (colorectal cancer) (Figures 1f and g, Supplementary Figure S1B) and in the NCI-60 panel of cell lines (Supplementary Figure S1E, Supplementary Table S3). In addition to let-7i, another let-7 family member, miR-98, was also downregulated by mutant p53 (Supplementary Table S1, Supplementary Figures S1C, D). We next examined whether let-7i expression was associated with mutant p53 in a clinically relevant setting. Indeed, let-7i was less abundant (approximately threefold, P = 0.0028) in breast cancer patient samples with mutant p53 (N = 15) as compared with patients that expressed wild-type p53 (N = 15) (Figure 1h, Supplementary Table S4), further establishing let-7i as a target of mutant p53. Mutant p53 and p63 associate with the let-7i promoter to regulate let-7i transcription Human let-7i is transcribed from chromosome 12 as a monocistronic primary transcript (~750 nt).31 Our recent study32 suggested a potential mutant p53-binding site in a 2461-bp region upstream of transcription start site (TSS). We therefore inserted this let-7i promoter region upstream of firefly luciferase gene of pGL4 (Figure 2a) and performed luciferase assays. As compared with H1299-EV cells let-7i promoter activity was significantly reduced (~40%) when the let-7i promoter construct (pGL4-let-7i) was transfected for 36 h into H1299-p53R273H cells (Figure 2b, left panel). Silencing mutant p53 in H1299-p53R273H cells with p53 siRNAs for 48 h increased luciferase expression (~1.5-fold) (Figure 2b, right panel). Immunoblotting verified efficient knockdown of mutant p53 (Supplementary Figure S2A). To complement these results, we next examined the effect of Oncogene (2014), 1 – 11

transiently over-expressing p53R273H in H1299 cells on let-7i promoter activity. As compared with EV-transfected cells, let-7i promoter activity was significantly reduced by exogenous mutant p53 (Figure 2c), suggesting that mutant p53 represses let-7i transcription. To further establish let-7i as a direct target of mutant p53, we stably knocked down mutant p53 in the breast cancer cell line MDA-MB-231 (p53R280K) and the pancreatic cancer cell line MIA-PaCa-2 (p53R248W) using two different p53-shRNAs (Supplementary Figures S2B, C). Stable depletion of mutant p53 significantly elevated let-7i and pri-let-7i levels (Figures 2d and e) in both cell lines, suggesting that endogenous mutant p53 represses let-7i transcription. Mutant p53 drives migration, invasion and metastasis by inhibiting the transcriptional activity of p63 or p73.10,17,18,33 We therefore tested the interaction between p63, p73 and mutant p53 by immunoprecipitation (IP) from MDA-MB-231and A431 (p53R273H) cells. Consistent with previous reports,10,16,34 mutant p53 co-immunoprecipitated with p63 and p73 in both cell lines (Figure 2f, Supplementary Figures S2D, E). To determine whether mutant p53, p63 and/or p73 bind to the let-7i promoter, we performed chromatin immunoprecipitation (ChIP) assays from MDA-MB-231cells. We observed specific enrichment (greater than threefold) of the let-7i promoter with anti-p53 or anti-p63 (Figure 2g) but not with anti-p73 IP (data not shown). Two control regions located 2756–2868 bp (CTL#1) and 2644–2750 bp (CTL#2) upstream of the let-7i TSS did not bind to p63 or mutant p53, demonstrating the specificity. Moreover, transfection of MDA-MB-231 or A431 cells with CTL or p63 siRNA for 48 h significantly downregulated p63 mRNA (Supplementary Figure S2F), pri-let-7i and let-7i (Figure 2h, Supplementary Figure S2G), suggesting that p63 enhances let-7i transcription. A mutant p53–p63 complex regulates let-7i transcription through p63 Next, we inserted into pGL4, a 2071 bp let-7i promoter region lacking the ~ 400-bp region that was bound by mutant p53 and p63 and designated this construct as Δ400 (Figure 3a). Bioinformatic analysis (TRANSFAC) of this 400 bp region indicated a 25-nt p63-binding element located 116 nt upstream of the let-7i TSS (Supplementary Figure S3A). In order to further dissect the transcriptional regulation of let-7i by mutant p53/p63, we used site-directed mutagenesis to delete this putative p63-binding element from the full-length let-7i promoter construct (Δp63RE, Figure 3a). Luciferase assays were performed after cotransfecting pGL4, pGL4-let-7i (FL), Δ400 or Δp63RE constructs with CTL, p53 or p63 siRNAs in MDA-MB-231 cells for 48 h. Silencing mutant p53 significantly increased (~3.5-fold) luciferase activity of FL but to a lesser extent with Δ400 and Δp63RE (Figure 3b). Depletion of p63 showed less-pronounced luciferase activity (approximately twofold) of FL, whereas the luciferase activity from Δ400 and Δp63RE was comparable to the control. This result indicates that mutant p53 inhibits p63 in a ~ 400-bp region upstream of the let-7i TSS and mutant p53/p63-binding site is required for transrepression. To further confirm this result, we inserted the ~ 400-bp fragment into pGL3 and performed luciferase assays after transfecting MDA-MB-231 cells with CTL, p63 or p53 siRNAs. Knockdown of mutant p53 significantly increased, whereas silencing p63 significantly decreased luciferase expression (Supplementary Figures S3A, B), suggesting that the 400-bp region requires p63 to drive let-7i expression To assess whether mutant p53 and p63 bind to the let-7i promoter individually or as a complex, we performed ChIP assays after transfecting MDA-MB-231 cells with CTL, p63 or p53 siRNAs. Silencing p63 not only abolished the binding of p63 to the let-7i promoter region but also reduced the binding of mutant p53 (Figure 3c). As expected, silencing mutant p53 abolished the © 2014 Macmillan Publishers Limited

Role of let-7i in mutant p53 driven cancer M Subramanian et al

3

Figure 1. Mutant p53 downregulates let-7i expression in cell lines and in patient samples. (a) Table shows the number of miRNAs up- or downregulated by mutant p53 according to deep sequencing results (Supplementary Table S1) from H1299 cells stably expressing EV or p53R273H. (b) Validation of deep sequencing results by TaqMan miRNA qRT-PCR normalized to U6. (c) qRT–PCR analysis of let-7i primary transcript (pri-let-7i) and the primary transcript for the miR-155 host gene (miR-155-HG) from H1299 cells transduced with EV or p53R273H. qRT–PCR was normalized to GAPDH mRNA and the housekeeping mRNA UBC served as an internal control. (d) H1299 cells were transiently transfected with empty vector (EV), wild-type p53 (p53-WT) or p53R273H-expressing constructs for 48 h and the levels of miR-34a and let-7i were measured by qRT–PCR normalized to U6. (e) Immunoblotting shows comparable expression of wild-type p53 and p53R273H. (f, g) Relative levels of let-7i and pri-let-7i were evaluated in three isogenic DLD-1 colon cancer cell lines by qRT–PCR normalized to U6 or UBC mRNA. The isogenic cell lines consisted of − /Sil DLD1 cells (no functional p53), wild-type p53-expressing +/Sil DLD1 cells and mutant p53-expressing DLD1 cells (S241F/Sil). (h) let-7i levels in breast cancer patient samples that were wild-type for p53 (WTp53) or had a missense mutation in p53 (MTp53) was measured by qRT–PCR relative to U6. Error bars represent mean ± s.d. from three independent experiments. *P o0.05; #P o0.01; **P o0.005; ##Po 0.001.

binding of mutant p53 to the let-7i promoter but we also observed increased binding of p63 to the let-7i promoter, indicating that mutant p53 inhibits binding of p63 to the let-7i promoter. Consistent with this result, the association of p63 and mutant p53 was not disrupted by DNase treatment (Figure 3d). We validated these results using a sequential IP strategy, in which we transfected MDA-MB-231 cells with EV or a HA-p63-expressing © 2014 Macmillan Publishers Limited

construct for 48 h and performed ChIP using anti-HA beads. The eluate from anti-HA ChIP was divided into two fractions. The first fraction was subjected to qPCR (after reversing the cross-links) for the let-7i promoter region. As expected, we observed significant enrichment of the let-7i promoter region in the HA-p63 ChIP (Figure 3e). The second fraction was diluted in IP buffer and subjected to mutant p53 ChIP. Significant enrichment of the let-7i Oncogene (2014), 1 – 11

Role of let-7i in mutant p53 driven cancer M Subramanian et al

4

Figure 2. Mutant p53 and p63 bind to the let-7i promoter to regulate let-7i expression. (a) Schematic representation of let-7i promoter region containing the putative mutant p53-binding site. (b) H1299 cells stably expressing EV or p53R273H were cotransfected with the luciferase promoter plasmid carrying the let-7i promoter (pGL4-let-7i) and pRL-TK (internal control). Luciferase activity was measured after 36 h (left panel). Repression of let-7i promoter by mutant p53 was also analyzed by cotransfecting H1299-p53R273H cells with control (CTL) or p53 siRNAs and pGL4-let-7i for 48 h and measuring luciferase activity (right panel). (c) Parental H1299 cells were cotransfected with the let-7i promoter reporter (pGL4-let-7i), EV or p53R273H plasmids and pRL-TK (internal control). Luciferase activity was measured after 48 h. The effect of stable knockdown of mutant p53 in MDA-MB-231 and MIA-PaCa-2 cells on mature let-7i (d) and primary let-7i (e) levels was assessed by qRT–PCR normalized to U6 or UBC. (f) Mutant p53 co-immunoprecipitates with p63 and p73 in whole-cell lysates prepared from MDA-MB-231 cells. (g) Chromatin immunoprecipitation (ChIP) assays were performed to pulldown mutant p53 or p63 from MDA-MB-231 cells. Fold enrichment of the let-7i promoter region in the mutant p53 or p63 ChIP was determined by qPCR. (h) MDA-MB-231 cells were transfected with CTL or p63 siRNAs for 48 h and the levels of primary let-7i transcript (left panel) or mature let-7i (right panel) was assessed by qRT–PCR normalized to UBC or U6.

promoter region was observed in the mutant p53 ChIP (Figure 3e), suggesting that a mutant p53-p63 complex binds to the let-7i promoter. As a control experiment, over-expression of HA-p63 was verified by immunoblotting (Figure 3f). Moreover, luciferase activity driven by the let-7i promoter increased upon Oncogene (2014), 1 – 11

over-expression of HA-p63 and this effect was reduced when mutant p53 was co-transfected with HA-p63 (Figure 3g). These results suggest that a mutant p53–p63 complex binds to the let-7i promoter through p63 and that mutant p53 inhibits the occupancy of p63 on the let-7i promoter. © 2014 Macmillan Publishers Limited

Role of let-7i in mutant p53 driven cancer M Subramanian et al

5

Figure 3. A mutant p53–p63 complex regulates let-7i transcription through p63. (a) Schematic representation of the let-7i promoter region containing the putative mutant p53/p63-binding site (FL), lacking the mutant p53/p63 binding region (Δ400) or lacking the p63 response element (Δp63RE). (b) MDA-MB-231 cells were cotransfected with pGL4, pGL4-let-7i or pGL4-let-7i-Δ400 or pGL4-let-7i-Δp63RE and CTL, p53 or p63 siRNAs. Luciferase activity was measured after 48 h. (c) ChIP assays were performed to pulldown mutant p53 or p63 from MDA-MB-231 cells transfected with CTL, p53 or p63 siRNAs. Fold enrichment of let-7i promoter region in the mutant p53 or p63 ChIP was determined by qPCR. (d) p63 was immunoprecipitated (IP) from MDA-MB-231 whole-cell lysates in the presence or absence of DNase, and immunoblotting for p53 was performed. (e) Sequential ChIP assays were performed to HA-p63 and then mutant p53 from MDA-MB-231. Enrichment of let-7i promoter region in ChIP reactions was measured by qPCR. (f) Parental H1299 cells were transfected with HA-p63 after 48 h and immunoblotting was performed for using anti-HA. (g) Parental H1299 cells were cotransfected with the let-7i promoter (pGL4-let-7i-FL), EV or HA-p63 or HA-p63 and p53R273H plasmids and pRL-TK (internal control). Luciferase activity was measured after 48 h. Error bars represent mean ± s.d. from three independent experiments. *P o0.05; #Po0.01; **P o0.005; ##P o0.001.

Introduction of let-7i in mutant p53 cells inhibits cell migration, invasion and metastasis Mutant p53 has been shown to promote cell migration and metastasis of MDA-MB-231 cells.7 Consistent with this report, we observed significantly decreased invasion upon stable knockdown of mutant p53 in MDA-MB-231 cells (Supplementary Figures S4A and C). To determine the downstream consequences of let-7i downregulation by mutant p53, we re-introduced let-7i in the highly invasive MDA-MB-231 cells and determined the effect on migration and invasion. MDA-MB-231 cells were transfected with a control (CTL) miRNA (cel-miR-67) mimic or let-7i mimics (10 nM) for 36 h and the effect on migration and invasion was monitored by transwell migration and matrigel invasion assays. Introduction of let-7i significantly impaired (approximately twofold) the migration and invasion of MDA-MB-231cells and also inhibited migration of MIA-PaCa-2 cells (Figures 4a–c). Consistent with the known antiproliferative function of let-7,30,35 let-7i inhibited proliferation of MDA-MB-231 cells (Supplementary Figure S4D) but this effect was observed after 3 days of transfection and not at earlier time points when the migration or invasion assays were performed, indicating that the anti-migratory or anti-invasive effect of let-7i was not due to inhibition of proliferation. © 2014 Macmillan Publishers Limited

To evaluate the anti-invasive effect of let-7i in mutant p53 cells in vivo, we transfected MDA-MB-231cells with CTL or let-7i mimics for 36 h and injected the cells into the tail vein of mice. After 8 weeks, the recipients were killed and the lungs were evaluated for the presence of metastatic tumors. Strikingly, MDAMB-231 cells expressing let-7i displayed a dramatic reduction (~15-fold, P o0.0001) in lung metastatic colonization (Figure 4d, Supplementary Figure S4E). These results established let-7i as an important downstream effector of mutant p53 gain of function in vitro and in vivo. Importantly, although we observed ~ 2000-fold increase in let-7i levels when MDA-MB-231 cells were transfected with let-7i mimics (10 nM) (Figure 4e), the abundance of let-7i in Ago2 IPs (Figures 4f and g) was near physiological levels, increasing only ~ 10-fold (~5-fold in CTL and ~ 50-fold in let-7i mimic-transfected cells). This result is consistent with a recent study36 and suggest that a majority of the transfected miRNA is not incorporated into the RNA-induced silencing complex (RISC). let-7i inhibits the expression of a highly connected network of genes centered on MYC To understand the mechanism(s) by which let-7i caused the observed phenotypic effects, we re-introduced let-7i in Oncogene (2014), 1 – 11

Role of let-7i in mutant p53 driven cancer M Subramanian et al

6

Figure 4. Introduction of let-7i in cell lines expressing endogenous mutant p53 inhibits cell migration, invasion and metastasis. MDA-MB-231 cells were reverse-transfected with CTL or let-7i mimics for 36 h and cell migration or invasion was assessed by transwell migration assays (a) and matrigel invasion assays (b). (c) MIA-PaCa-2 cells were transfected with CTL or let-7i mimics for 36 h and the effect of let-7i on cell migration was assessed by transwell migration assays. (d) MDA-MB-231 cells were reverse-transfected with CTL or let-7i mimics and metastasis to the lung was assessed 8 weeks after tail vein injection. Results are median ± interquartile range. Representative lung cross sections shows large metastases formed from CTL but not let-7i-transfected cells. (e) let-7i over-expression in MDA-MB-231 cells was determined by qRT–PCR from total RNA. (f) Specific enrichment of Ago2 in Ago2 IP from MDA-MB-231 cytoplasmic extract is shown. (g) Abundance of let-7i and miR-98 in IgG and Ago2 IPs from MDA-MB-231 cytoplasmic extracts prepared after transfecting the cells with CTL or let-7i mimic was measured by qRT–PCR from pulldown RNA. Error bars represent mean ± s.d. of three independent experiments. *P o0.05; ##P o0.001.

H1299-p53R273H cells and performed microarrays to identify the genome-wide targets of let-7i. Although this approach cannot identify let-7i targets that are translationally regulated, it is likely that this strategy will identify a large proportion of let-7i targets, based on recent studies.27,30,37 We therefore performed microarrays after transfecting H1299-p53R273H cells with CTL or let-7i mimics (10 nM) for 36 h. Using an arbitrary cutoff of twofold (P-value o0.05), 72 unique mRNAs were downregulated (Supplementary Table S5). LIN28B, a known target of the let-7 family,38 was the most significantly downregulated. Several known let-7 targets30,39,40 including DICER (2.4-fold), E2F5 (2.8-fold) and NRAS (2.3-fold) were also downregulated >2-fold, but some Oncogene (2014), 1 – 11

well-established let-7 targets30,41 including AURKB (1.9-fold) and MYC (1.6-fold) were not. With a lower threshold of 1.5-fold (P-value o0.05), 375 unique mRNAs were downregulated (Supplementary Table S5), of which 153 (~41%) were also predicted by TargetScan (Figure 5a), suggesting that a significant proportion of the down-regulated transcripts may be direct let-7i targets. Similar to MDA-MB-231 cells (Figures 4e–g), we observed ~ 2500-fold increase in let-7i levels upon transfection of let-7i mimics in H1299-p53R273H cells but association of let-7i in Ago2 IPs was not supraphysiological (Supplementary Figures S4F, G). Next, we performed gene ontology (GO) analysis on the 375 mRNAs downregulated by let-7i. Consistent with the © 2014 Macmillan Publishers Limited

Role of let-7i in mutant p53 driven cancer M Subramanian et al

7

Figure 5. let-7i inhibits a network of genes involved in proliferation and RNA post-transcriptional modification. (a) Venn diagram showing the overlap between the 375 mRNAs that were downregulated ⩾ 1.5-fold (Po 0.05) by let-7i (Supplementary Table S5) in H1299-p53R273H cells and let-7i targets predicted by TargetScan. (b) Gene ontology (GO) analysis (Ingenuity pathways) for the 375 mRNAs downregulated upon let-7i over-expression in H1299-p53R273H cells. (c) Direct gene interaction network (Ingenuity pathways) of the 375 let-7i target genes.

well-established anti-proliferative function of let-7,30,35 cell cycle, cell growth and proliferation were among the top two overrepresented processes (Figure 5b). Many genes that promote proliferation belonged to these two categories, including AURKB, CCND1, CDC25A, E2F2, E2F5, MYC and NRAS. Interestingly, RNA post-transcriptional modification was the third most significantly enriched GO process; 18 out of the 20 genes annotated for RNA post-transcriptional modification were downregulated by let-7i (Supplementary Table S7). Among these, DICER1 is a known target of let-7.39 The other 17 included genes implicated in the miRNA pathway, such as EIF2C2 (encoding Ago2) and TARBP2, DDX56 (encoding an RNA helicase) and CPSF1, which encodes a protein that has a central role in pre-mRNA splicing. The 375 genes downregulated by let-7i formed a network centered on MYC and its interacting gene products that are key factors in the control of cellular proliferation including AURKB, CCND1, CDC25A, E2F2, E2F4, © 2014 Macmillan Publishers Limited

E2F5 and E2F6 (Figure 5c). In addition, many genes involved in RNA metabolism, such as CPSF1, DICER1, DDX18, DDX56, EIF2C2, LIN28B and XPO5 also interact with each other or with MYC. These analyses suggest that let-7i inhibits the expression of a network of genes that have pivotal roles in control of proliferation and RNA metabolism. We next looked at the overlap of select genes known to be transcriptionally activated by mutant p533 and the 375 mRNAs downregulated by let-7i. Out of the 62 genes (Supplementary Table S8) whose transcription is upregulated by mutant p53,3 E2F542 and MYC43 were also repressed by let-7i, suggesting that these mRNAs may be derepressed due to downregulation of let-7i. Next, we searched for known MYC and E2F targets in the genes downregulated by let-7i by performing Gene set enrichment analysis (GSEA) for the 222 genes downregulated by let-7i that were not predicted let-7i targets Oncogene (2014), 1 – 11

Role of let-7i in mutant p53 driven cancer M Subramanian et al

8

Figure 6. let-7i inhibits the 3′ UTR of multiple oncogenes and RNA post-transcriptional modification genes. MDA-MB-231 cells were reversetransfected with CTL or let-7i mimics for 36 h and changes in mRNA expression of known let-7 targets (a) and novel targets (b) identified in the microarrays were validated by qRT–PCR normalized to UBC. (c) MDA-MB-231 cells were reverse-transfected with CTL or let-7i mimics for 36 h and the protein levels of CPSF1 and MYC were analyzed by immunoblotting. β-actin was used as loading control. (d) Luciferase assays were performed from H1299-R273H cells cotransfected with psiCHECK2 (Vector), psiCHECK2 containing the 3′UTR of known let-7 targets (E2F5 or NRAS) or psiCHECK2 containing the 3′ UTR of nine RNA post-transcriptional modification genes and CTL or let-7i mimics. (e) Effect of silencing let-7i in H1299 cells with a let-7i sponge (let-7i-Spg) on the expression of let-7i target mRNAs was assessed by qRT–PCR. (f) Expression of CPSF1 and MYC protein from H1299 stably transfected with CTL-spg or let-7i sponge (let-7i-Spg) was measured by immunoblotting. β-actin was used as loading control. (g) mRNA levels of E2F5, LIN28B, MYC and NRAS were measured in breast cancer patient samples that were wild-type for p53 (WTp53) or had a missense mutation in p53 (MTp53) by qRT–PCR relative to GAPDH. (h) Schematic model suggesting a mechanism by which mutant p53 drives cell cycle progression, migration and invasion. Mutant p53 (MTp53) inhibits let-7i expression by inhibiting p63 at the let-7i promoter to upregulate let-7i targets including NRAS, MYC, E2F5, LIN28B, DDX18 and CPSF1 to enhance cellular proliferation, migration and invasion. Error bars represent mean ± s.d. of three independent experiments. *Po0.05; #P o0.01; **Po 0.005; ## P o0.001.

Oncogene (2014), 1 – 11

© 2014 Macmillan Publishers Limited

Role of let-7i in mutant p53 driven cancer M Subramanian et al

(Figure 5a). Notably, we found significant enrichment of MYC (P = 0.0013, 29 genes) and E2F (P = 0.0033, 10 genes) motifs in the promoters of these genes (Supplementary Table S9), suggesting that decreased MYC and E2F levels upon let-7i over-expression may have a role in downregulating some indirect let-7i targets. The 3′ UTR of many cell cycle and RNA metabolism genes is repressed by let-7i To validate the microarray results, MDA-MB-231 and H1299p53R273H cells were transfected with CTL or let-7i mimics and 36 h later qRT-PCR was performed. The levels of known let-7 targets E2F5, HMGA1, LIN28B, MYC and NRAS declined substantially (greater than twofold) (Figure 6a, Supplementary Figure S5A). We next validated 11 out of 18 RNA post-transcriptional modification genes. Among these 11 genes, DICER1 is a known let-7 target39 and was downregulated more than fourfold (Figure 6b, Supplementary Figure S5B). Expression of the other 10 let-7i targets (CPSF1, DDX18, DDX56, EIF4A1, EIF2C2, LSM6, PABPC4, RBM38, TARBP2 and ZC3H3) declined more than twofold (Figure 6b, Supplementary Figure S5B). Among these 10 mRNAs, 4 are predicted let-7i targets and 6 contain at least one 6-mer seed match in their 3′ UTR (Supplementary Table S10). Downregulation of CPSF1 and MYC protein was also validated by immunoblotting (Figure 6c). These results suggest that let-7i inhibits the expression of many oncoproteins that promote proliferation, migration and invasion.44,45 To investigate whether the let-7i targets were downregulated through their 3′ UTRs, we inserted their 3′ UTRs in psiCHECK2. When luciferase assays were performed from H1299-p53R273H cells cotransfected for 48 h with luciferase reporters and CTL or let-7i mimics, the 3′ UTR of eight out of nine genes (CPSF1, DDX18, EIF4A1, LSM6, PABPC4, RBM38, TARBP2 and ZC3H3) was significantly repressed by let-7i (Figure 6d); E2F5 3′ UTR was used a positive control. To determine whether a subset of these genes was also regulated by endogenous let-7i, we measured their mRNA levels by qRT-PCR after silencing let-7i in parental H1299 cells with a let-7i sponge. Silencing let-7i elevated the levels of 8 out of 10 let-7i target mRNAs, further indicating that a majority of these genes are regulated by let-7i (Figure 6e). Significant increase in CPSF1 and MYC protein levels was also observed with the let-7i sponge (Figure 6f). Moreover, elevated let-7i levels upon stable knockdown of mutant p53 in MDA-MB-231 and MIA-PaCa-2 cells was associated with downregulation of some of these let-7i targets (Supplementary Figures S5C, D). Next, we performed qRT–PCR to measure the expression of select let-7i targets in breast cancer patient samples where we observed decreased let-7i levels in patient samples that had mutant p53 (Figure 1h). In mutant p53 patient samples, the mRNA levels of E2F5, LIN28B and MYC were upregulated, whereas NRAS mRNA levels didn't change significantly (Figure 6g). Based on the results of this study we propose a model in which a mutant p53–p63 complex inhibits let-7i transcription to elevate the levels of transcripts encoding oncoproteins such as E2F5, LIN28B, MYC and NRAS and specific RNA metabolism genes, including CPSF1, DDX18 and LIN28B (Figure 6h). Deregulation of these genes by the mutant p53/let-7i axis may promote cellular proliferation, migration and invasion driven by mutant p53. DISCUSSION Despite the expanding list of mutant p53-regulated genes, only a handful of recent studies29,46–48 have investigated miRNAs downstream of mutant p53. Here we identified mutant p53-regulated miRNAs and demonstrate that the tumor suppressor let-7i is a direct target and a key downstream mediator of mutant p53. Our findings are consistent with recent studies suggesting the role of distinct p53 mutants in regulating miRNAs including miR-128,46 © 2014 Macmillan Publishers Limited

miR-15529 and miR-130b.48 Further studies are required to assess what proportion of the 41 mutant p53-regulated miRNAs are direct targets of p53R273H and/or p63–p53R273H complexes. In addition to p63, mutant p53 may also regulate miRNAs through other mutant p53-interacting proteins including NF-Y,12 ETS-1,14 ETS-232 and VDR.49 The human let-7 family consists of 10 distinct miRNAs25 with identical seed sequences. Our results indicate that mutant p53 downregulates let-7i, unlike MYC and LIN28B that inhibit the expression of a majority of let-7 family members.50,51 However, we have shown that inhibition of let-7i by mutant p53 has functional outcomes. Introduction of let-7i in MDA-MB-231 cells reduced migration and invasion and dramatically inhibited metastasis. A recent study showed that Twist suppresses let-7i levels and promotes a mesenchymal movement of head and neck cancer cells52 by upregulating NEDD9 and DOCK3, leading to RAC1 activation. However, these mRNAs were not altered by let-7i in our microarrays, suggesting that they may not be downstream of the mutant p53/let-7i axis. Our microarray analysis identified many established let-7 targets including E2F5, LIN28B, MYC and NRAS. Stable knockdown of mutant p53 elevated let-7i and downregulated its target mRNAs indicating that these genes are regulated by the mutant p53/let-7i axis. Moreover, the let-7i targets E2F5 and MYC have been shown to be transcriptionally upregulated by mutant p53,42,43 suggesting that the mutant p53/let-7i axis has a role in elevating MYC and E2F5 levels. In addition to these genes, we have shown that let-7i directly inhibits the 3′ UTR of several RNA post-transcriptional modification genes including CPSF1, DDX18, LIN28B and TARBP2. Among these genes, DDX18 has a role in cell cycle progression,53 whereas LIN28B promotes migration, invasion and transformation.44,54 A recent study suggested that CPSF6 (a homolog of CPSF1) is a direct target of mutant p53 and silencing CPSF6 inhibits migration of mutant p53 cells.10 We propose that downregulation of let-7i by mutant p53 increases the expression of CPSF1, DDX18 and LIN28B to promote cell cycle progression, migration and invasion. Collectively, our results suggest that in addition to transcriptional regulation, mutant p53 indirectly exerts its oncogenic functions post-transcriptionally, by downregulating let-7i. Although our findings establish let-7i as a direct and functionally important target of mutant p53, further studies are needed to determine the mechanism of biogenesis and function of other mutant p53-responsive miRNAs identified in our study.

MATERIALS AND METHODS Cell culture, transfections and plasmids H1299 cells stably expressing EV or p53R273H and HA-p63 construct were a kind gift from K Vousden. All cell lines were maintained in DMEM with 10% fetal calf serum (Invitrogen, Carlsbad, CA, USA) at 37 °C and 5% CO2. MiRNA mimics were purchased from Dharmacon (Lafayette, CO, USA) and reverse-transfected (10 nM) using RNAimax (Invitrogen). let-7i-luciferase promoter construct (pGL4-let-7i) was a gift from S O’Hara.31 Sponge constructs for let-7i were a kind gift from M Yang.52 pGL4-let-7i (Δ400) and (Δp63RE) deletion construct were generated by PCR or site-directed mutagenesis. The 3′ UTR constructs for luciferase assays were generated by PCR (Supplementary Table S12).

Stable transductions Retrovirus expressing shRNA-targeting luciferase (CTLsh), p53 shRNA3 p53 shRNA4 was generated in 293T cells. MDA-MB-231 and MIA-PaCa-2 cells were transduced with retrovirus and stably transfected cells were selected after 48 h with puromycin (1.0 μg/ml). H1299 cells were transfected with let-7i sponge (pmCherry-spg-let-7i) or a control vector (pmCherry-spg-ctrl) for 48 h and selected after 48 h using puromycin (1.0 μg/ml). Oncogene (2014), 1 – 11

9

Role of let-7i in mutant p53 driven cancer M Subramanian et al

10 Deep sequencing of small RNAs

p53 mutational analysis in breast cancer patient samples

Total RNA was isolated using the miRNeasy Mini Kit (Qiagen, Germantown, MD, USA) and quantified using the NanoDrop ND-1000 spectrophotometer (Wilmington, DE, USA). RNA integrity was evaluated with a 2100 BioAnalyzer (Agilent, Santa Clara, CA, USA). The small RNA fraction (18–30 nt) was enriched from 10 μg total RNA using 15% denaturing trisborate-EDTA (TBE) urea polyacrylamide gel electrophoresis (PAGE). Purified small RNAs were ligated to proprietary 3ʹ and 5ʹ RNA adapters (Illumina) and used as templates for cDNA synthesis. cDNA PCR amplification was performed using adapter-specific primers (12 cycles), purified using TBEPAGE and diluted to 8 pM for cluster generation and sequencing on an Illumina GAII. Adapter sequences were clipped and subsequently analyzed using BWA. Coordinates of mature miRNAs were downloaded from miRBase (v18) and tag densities were counted and averaged over the length of the mature miRNA.

Freshly frozen tumor specimens were obtained at the time of surgery from breast cancer patients and the p53 mutation status was determined as described previously.56

RNA isolation, qRT–PCR and microarrays Isolation of total RNA followed by qRT–PCR (Supplementary Table S11) and microarrays were performed as previously described.22

CONFLICT OF INTEREST The authors declare no conflict of interest.

ACKNOWLEDGEMENTS This work was supported by the Intramural Research Program of the National Institutes of Health, National Cancer Institute, Center for Cancer Research. Sudha Sharma was supported by the National Institute of General Medical Sciences of the National Institutes of Health under Award Number SC1GM093999. We thank K Prasanth, W Bodmer, B Hassel, M Thomas, J Lieberman and M Gorospe for their comments on this manuscript. We thank B Vogelstein for the DLD1 isogenic cell lines, K Vousden for the H1299-EV and H1299-p53R273H cells, S O’Hara for the pGL4-let-7i and Z Mourelators for the Ago2 (2A8) antibody.

Migration and invasion assays

REFERENCES

For transwell migration assays, MDA-MB-231 or MIA-PaCa-2 cells were transfected with CTL mimic or let-7i mimic for 36 h and 4 × 105 cells were plated in the top chamber with the non-coated membrane (24-well insert; pore size, 8 μm; BD Biosciences, San Jose, CA, USA). For invasion assays, matrigel (BD biosciences) was polymerized in transwell inserts for 45 min at 37 °C. Transfected cells or MDA-MB-231 (CTLsh, p53sh3 and p53sh4) were plated in the top chamber in medium without serum. In both assays, the lower chamber filled with 10% FCS and 25 ng/ml EGF (Sigma, St Louis, MO, USA) was used as a chemoattractant. Cells were incubated for 24 h and the cells that did not migrate or invade through the pores were removed by a cotton swab. Cells on the lower surface of the membrane were stained with crystal violet and counted.

1 Wei CL, Wu Q, Vega VB, Chiu KP, Ng P, Zhang T et al. A global map of p53 transcription-factor binding sites in the human genome. Cell 2006; 124: 207–219. 2 Vousden KH, Lu X. Live or let die: the cell's response to p53. Nat Rev Cancer 2002; 2: 594–604. 3 Freed-Pastor WA, Prives C. Mutant p53: one name, many proteins. Genes Dev 2012; 26: 1268–1286. 4 Nigro JM, Baker SJ, Preisinger AC, Jessup JM, Hostetter R, Cleary K et al. Mutations in the p53 gene occur in diverse human tumour types. Nature 1989; 342: 705–708. 5 Muller PA, Vousden KH. p53 mutations in cancer. Nat Cell Biol 2013; 15: 2–8. 6 Muller PA, Caswell PT, Doyle B, Iwanicki MP, Tan EH, Karim S et al. Mutant p53 drives invasion by promoting integrin recycling. Cell 2009; 139: 1327–1341. 7 Adorno M, Cordenonsi M, Montagner M, Dupont S, Wong C, Hann B et al. A mutant-p53/Smad complex opposes p63 to empower TGFbeta-induced metastasis. Cell 2009; 137: 87–98. 8 Muller PA, Trinidad AG, Timpson P, Morton JP, Zanivan S, van den Berghe PV et al. Mutant p53 enhances MET trafficking and signalling to drive cell scattering and invasion. Oncogene 2012; 32: 1252–1265. 9 Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez-Barrueco R et al. Mutant p53 disrupts mammary tissue architecture via the mevalonate pathway. Cell 2012; 148: 244–258. 10 Girardini JE, Napoli M, Piazza S, Rustighi A, Marotta C, Radaelli E et al. A Pin1/mutant p53 axis promotes aggressiveness in breast cancer. Cancer Cell 2011; 20: 79–91. 11 Liu K, Ling S, Lin WC. TopBP1 mediates mutant p53 gain of function through NF-Y and p63/p73. Mol Cell Biol 2011; 31: 4464–4481. 12 Di Agostino S, Strano S, Emiliozzi V, Zerbini V, Mottolese M, Sacchi A et al. Gain of function of mutant p53: the mutant p53/NF-Y protein complex reveals an aberrant transcriptional mechanism of cell cycle regulation. Cancer Cell 2006; 10: 191–202. 13 Murphy M, Ahn J, Walker KK, Hoffman WH, Evans RM, Levine AJ et al. Transcriptional repression by wild-type p53 utilizes histone deacetylases, mediated by interaction with mSin3a. Genes Dev 1999; 13: 2490–2501. 14 Sampath J, Sun D, Kidd VJ, Grenet J, Gandhi A, Shapiro LH et al. Mutant p53 cooperates with ETS and selectively up-regulates human MDR1 not MRP1. J Biol Chem 2001; 276: 39359–39367. 15 Li Y, Prives C. Are interactions with p63 and p73 involved in mutant p53 gain of oncogenic function? Oncogene 2007; 26: 2220–2225. 16 Gaiddon C, Lokshin M, Ahn J, Zhang T, Prives C. A subset of tumor-derived mutant forms of p53 down-regulate p63 and p73 through a direct interaction with the p53 core domain. Mol Cell Biol 2001; 21: 1874–1887. 17 Marin MC, Jost CA, Brooks LA, Irwin MS, O'Nions J, Tidy JA et al. A common polymorphism acts as an intragenic modifier of mutant p53 behaviour. Nat Genet 2000; 25: 47–54. 18 Di Como CJ, Gaiddon C, Prives C. p73 function is inhibited by tumor-derived p53 mutants in mammalian cells. Mol Cell Biol 1999; 19: 1438–1449. 19 Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell 2009; 136: 215–233. 20 Djuranovic S, Nahvi A, Green R. miRNA-mediated gene silencing by translational repression followed by mRNA deadenylation and decay. Science 2012; 336: 237–240. 21 Thomas M, Lieberman J, Lal A. Desperately seeking microRNA targets. Nat Struct Mol Biol 2010; 17: 1169–1174.

In vivo metastasis assays Animal studies were performed under a protocols approved by the National Cancer Institute Animal Care and Use Committee following AALAAC guidelines and policies. MDA-MB-231 cells were reversetransfected with CTL or let-7i mimics (50 nM). Thirty-six hours after transfection, the cells were trypsinized, washed with PBS and suspended in PBS at a concentration of 5 × 106 cells/ml. A total of 0.1 ml cell suspension was injected into 6–8-week old female athymic nude mice (Animal Production Program, Frederick, MD, USA) through the tail vein (N = 10) each group. Eight weeks after inoculation, the mice were killed by CO2 narcosis. Histologically detectable metastases were enumerated on hematoxylin and eosin (H&E)-stained sections of inflated lungs.

Luciferase reporter assays Luciferase reporter assays were performed using the Dual-Luciferase Reporter Assay System (Promega, Milwaukee, WI, USA) as previously described.22

Immunoprecipitation and chromatin immunoprecipitation (ChIP) assay Immunoprecipitation from MDA-MB-231, A431 or H1299-R273H cytoplasmic extracts was performed as previously described.55 For each IP, 2.0 μg of anti-p53 (D0–1, Santa Cruz, Santa Cruz, CA, USA), anti-p63 (H-137, Santa Cruz), anti-p73 (Bethyl, Montgomery, TX, USA), anti-Ago2 (Mouse 2A8), rabbit IgG, mouse IgG and anti-HA beads was used. ChIP assays (Primer sequences in Supplementary Table S9) were performed as described previously.32

Immunoblotting Immunoblotting was performed using standard protocols using anti-p53 (D0–1, Santa Cruz, 1:5000), anti-p63 (H-137, Santa Cruz, 1:2500), anti-p73 (Bethyl, 1:5000), anti-HA High affinity (Roche, Indianapolis, IN, USA, 1:1000), anti-Ago2 (2A8 1:5000), anti-CPSF1 (Sigma, 1:2500), anti-Myc (9E10 Santa Cruz, 1:2500) and anti-β-actin (Sigma, 1:5000) antibodies. Oncogene (2014), 1 – 11

© 2014 Macmillan Publishers Limited

Role of let-7i in mutant p53 driven cancer M Subramanian et al

11 22 Lal A, Navarro F, Maher CA, Maliszewski LE, Yan N, O'Day E et al. miR-24 Inhibits cell proliferation by targeting E2F2, MYC, and other cell-cycle genes via binding to ‘seedless’ 3'UTR microRNA recognition elements. Mol Cell 2009; 35: 610–625. 23 Chi SW, Hannon GJ, Darnell RB. An alternative mode of microRNA target recognition. Nat Struct Mol Biol 2012; 19: 321–327. 24 He L, He X, Lim LP, de Stanchina E, Xuan Z, Liang Y et al. A microRNA component of the p53 tumour suppressor network. Nature 2007; 447: 1130–1134. 25 Roush S, Slack FJ. The let-7 family of microRNAs. Trends Cell Biol 2008; 18: 505–516. 26 Hermeking H. MicroRNAs in the p53 network: micromanagement of tumour suppression. Nat Rev Cancer 2012; 12: 613–626. 27 Chang TC, Wentzel EA, Kent OA, Ramachandran K, Mullendore M, Lee KH et al. Transactivation of miR-34a by p53 broadly influences gene expression and promotes apoptosis. Mol Cell 2007; 26: 745–752. 28 Raver-Shapira N, Marciano E, Meiri E, Spector Y, Rosenfeld N, Moskovits N et al. Transcriptional activation of miR-34a contributes to p53-mediated apoptosis. Mol Cell 2007; 26: 731–743. 29 Neilsen PM, Noll JE, Mattiske S, Bracken CP, Gregory PA, Schulz RB et al. Mutant p53 drives invasion in breast tumors through up-regulation of miR-155. Oncogene 2013; 32: 2992–3000. 30 Johnson CD, Esquela-Kerscher A, Stefani G, Byrom M, Kelnar K, Ovcharenko D et al. The let-7 microRNA represses cell proliferation pathways in human cells. Cancer Res 2007; 67: 7713–7722. 31 O'Hara SP, Splinter PL, Gajdos GB, Trussoni CE, Fernandez-Zapico ME, Chen XM et al. NFkappaB p50-CCAAT/enhancer-binding protein beta (C/EBPbeta)-mediated transcriptional repression of microRNA let-7i following microbial infection. J Biol Chem 2010; 285: 216–225. 32 Do PM, Varanasi L, Fan S, Li C, Kubacka I, Newman V et al. Mutant p53 cooperates with ETS2 to promote etoposide resistance. Genes Dev 2012; 26: 830–845. 33 Irwin MS, Kondo K, Marin MC, Cheng LS, Hahn WC, Kaelin Jr. WG. Chemosensitivity linked to p73 function. Cancer Cell 2003; 3: 403–410. 34 Strano S, Fontemaggi G, Costanzo A, Rizzo MG, Monti O, Baccarini A et al. Physical interaction with human tumor-derived p53 mutants inhibits p63 activities. J Biol Chem 2002; 277: 18817–18826. 35 Esquela-Kerscher A, Trang P, Wiggins JF, Patrawala L, Cheng A, Ford L et al. The let-7 microRNA reduces tumor growth in mouse models of lung cancer. Cell Cycle 2008; 7: 759–764. 36 Thomson DW, Bracken CP, Szubert JM, Goodall GJ. On measuring miRNAs after transient transfection of mimics or antisense inhibitors. PLoS One 2013; 8: e55214. 37 Baek D, Villen J, Shin C, Camargo FD, Gygi SP, Bartel DP. The impact of microRNAs on protein output. Nature 2008; 455: 64–71. 38 Guo Y, Chen Y, Ito H, Watanabe A, Ge X, Kodama T et al. Identification and characterization of lin-28 homolog B (LIN28B) in human hepatocellular carcinoma. Gene 2006; 384: 51–61. 39 Forman JJ, Legesse-Miller A, Coller HA. A search for conserved sequences in coding regions reveals that the let-7 microRNA targets Dicer within its coding sequence. Proc Natl Acad Sci USA. 2008; 105: 14879–14884.

40 Johnson SM, Grosshans H, Shingara J, Byrom M, Jarvis R, Cheng A et al. RAS is regulated by the let-7 microRNA family. Cell 2005; 120: 635–647. 41 Sampson VB, Rong NH, Han J, Yang Q, Aris V, Soteropoulos P et al. MicroRNA let-7a down-regulates MYC and reverts MYC-induced growth in Burkitt lymphoma cells. Cancer Res 2007; 67: 9762–9770. 42 Scian MJ, Stagliano KE, Anderson MA, Hassan S, Bowman M, Miles MF et al. Tumor-derived p53 mutants induce NF-kappaB2 gene expression. Mol Cell Biol 2005; 25: 10097–10110. 43 Frazier MW, He X, Wang J, Gu Z, Cleveland JL, Zambetti GP. Activation of c-myc gene expression by tumor-derived p53 mutants requires a discrete C-terminal domain. Mol Cell Biol 1998; 18: 3735–3743. 44 King CE, Wang L, Winograd R, Madison BB, Mongroo PS, Johnstone CN et al. LIN28B fosters colon cancer migration, invasion and transformation through let-7-dependent and -independent mechanisms. Oncogene 2011; 30: 4185–4193. 45 Wolfer A, Ramaswamy S. MYC and metastasis. Cancer Res 2011; 71: 2034–2037. 46 Donzelli S, Fontemaggi G, Fazi F, Di Agostino S, Padula F, Biagioni F et al. MicroRNA-128-2 targets the transcriptional repressor E2F5 enhancing mutant p53 gain of function. Cell Death Differ 2011; 19: 1038–1048. 47 Wang W, Cheng B, Miao L, Mei Y, Wu M. Mutant p53-R273H gains new function in sustained activation of EGFR signaling via suppressing miR-27a expression. Cell Death Dis 2013; 4: e574. 48 Dong P, Karaayvaz M, Jia N, Kaneuchi M, Hamada J, Watari H et al. Mutant p53 gain-of-function induces epithelial-mesenchymal transition through modulation of the miR-130b-ZEB1 axis. Oncogene 2012; 32: 3286–3295. 49 Stambolsky P, Tabach Y, Fontemaggi G, Weisz L, Maor-Aloni R, Siegfried Z et al. Modulation of the vitamin D3 response by cancer-associated mutant p53. Cancer Cell 2010; 17: 273–285. 50 Chang TC, Zeitels LR, Hwang HW, Chivukula RR, Wentzel EA, Dews M et al. Lin-28B transactivation is necessary for Myc-mediated let-7 repression and proliferation. Proc Natl Acad Sci USA 2009; 106: 3384–3389. 51 Zhu H, Shyh-Chang N, Segre AV, Shinoda G, Shah SP, Einhorn WS et al. The Lin28/ let-7 axis regulates glucose metabolism. Cell 2011; 147: 81–94. 52 Yang WH, Lan HY, Huang CH, Tai SK, Tzeng CH, Kao SY et al. RAC1 activation mediates Twist1-induced cancer cell migration. Nat Cell Biol 2012; 14: 366–374. 53 Payne EM, Bolli N, Rhodes J, Abdel-Wahab OI, Levine R, Hedvat CV et al. Ddx18 is essential for cell-cycle progression in zebrafish hematopoietic cells and is mutated in human AML. Blood 2011; 118: 903–915. 54 Hamano R, Miyata H, Yamasaki M, Sugimura K, Tanaka K, Kurokawa Y et al. High expression of Lin28 is associated with tumour aggressiveness and poor prognosis of patients in oesophagus cancer. Br J Cancer 2012; 106: 1415–1423. 55 Lal A, Thomas MP, Altschuler G, Navarro F, O'Day E, Li XL et al. Capture of microRNA-bound mRNAs identifies the tumor suppressor miR-34a as a regulator of growth factor signaling. PLoS Genet 2011; 7: e1002363. 56 Prueitt RL, Boersma BJ, Howe TM, Goodman JE, Thomas DD, Ying L et al. Inflammation and IGF-I activate the Akt pathway in breast cancer. Int Cancer J Int Cancer 2007; 120: 796–805.

Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)

© 2014 Macmillan Publishers Limited

Oncogene (2014), 1 – 11