p53 transcriptionally regulates SQLE to repress cholesterol synthesis and tumor growth

Article30 August 2021free access Transparent process p53 transcriptionally regulates SQLE to repress cholesterol synthesis and tumor growth Huishan Sun Huishan Sun State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Li Li Li Li State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Wei Li Wei Li State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Fan Yang Fan Yang State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Zhenxi Zhang Zhenxi Zhang State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Zizhao Liu Zizhao Liu State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Wenjing Du Corresponding Author Wenjing Du [email protected] orcid.org/0000-0002-9761-0774 State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Huishan Sun Huishan Sun State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Li Li Li Li State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Wei Li Wei Li State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China These authors contributed equally to this work Search for more papers by this author Fan Yang Fan Yang State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Zhenxi Zhang Zhenxi Zhang State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Zizhao Liu Zizhao Liu State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Wenjing Du Corresponding Author Wenjing Du [email protected] orcid.org/0000-0002-9761-0774 State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China Search for more papers by this author Author Information Huishan Sun1, Li Li1, Wei Li1, Fan Yang1, Zhenxi Zhang1, Zizhao Liu1 and Wenjing Du *,1 1State Key Laboratory of Medical Molecular Biology, Department of Cell Biology, Institute of Basic Medical Sciences Chinese Academy of Medical Sciences, School of Basic Medicine Peking Union Medical College, Beijing, China *Corresponding author. Tel: +86 010 69156953; E-mail: [email protected] EMBO Reports (2021)22:e52537https://doi.org/10.15252/embr.202152537 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions Figures & Info Abstract Cholesterol is essential for membrane biogenesis, cell proliferation, and differentiation. The role of cholesterol in cancer development and the regulation of cholesterol synthesis are still under active investigation. Here we show that under normal-sterol conditions, p53 directly represses the expression of SQLE, a rate-limiting and the first oxygenation enzyme in cholesterol synthesis, in a SREBP2-independent manner. Through transcriptional downregulation of SQLE, p53 represses cholesterol production in vivo and in vitro, leading to tumor growth suppression. Inhibition of SQLE using small interfering RNA (siRNA) or terbinafine (a SQLE inhibitor) reverses the increased cell proliferation caused by p53 deficiency. Conversely, SQLE overexpression or cholesterol addition promotes cell proliferation, particularly in p53 wild-type cells. More importantly, pharmacological inhibition or shRNA-mediated silencing of SQLE restricts nonalcoholic fatty liver disease (NAFLD)-induced liver tumorigenesis in p53 knockout mice. Therefore, our findings reveal a role for p53 in regulating SQLE and cholesterol biosynthesis, and further demonstrate that downregulation of SQLE is critical for p53-mediated tumor suppression. Synopsis This study reveals a SREBP2-independent role for p53 in regulating SQLE and cholesterol biosynthesis under normal sterol conditions, and further demonstrate that downregulation of SQLE is critical for p53-mediated tumor suppression. Under normal-sterol conditions, p53 directly represses the expression of SQLE in a SREBP2-independent manner. Through transcriptional downregulation of SQLE, p53 represses cholesterol production in vivo and in vitro, leading to tumor growth suppression. Inhibition of SQLE using small interfering RNA (siRNA) or terbinafine (a SQLE inhibitor) reverses the increased cell proliferation caused by p53 deficiency. Pharmacological inhibition of SQLE by terbinafine represses NAFLD-HCC tumorigenesis caused by p53 loss in vivo. Introduction Cholesterol is an essential component of cell membrane and important for cell proliferation and differentiation (Silvente-Poirot & Poirot, 2012). In mammalian cells, cholesterol is synthesized through multiple steps catalyzed by different metabolic enzymes. Squalene epoxidase (SQLE), one of the two rate-limiting enzymes in cholesterol synthesis, catalyzes the first oxygenation step converting squalene to 2,3(S)-monooxidosqualene (MOS). SQLE is relatively unstable and can be degraded through cholesterol-dependent proteasomal turnover (Gill et al, 2011; Foresti et al, 2013; Zelcer et al, 2014). Under lipid-depleted conditions, the expression of SQLE can be transcriptionally regulated by mature form of SREBP2 (Hidaka et al, 1990; Nakamura et al, 1996; Nagai et al, 2002). Among the multiple enzymes in cholesterol synthesis, SQLE appears to be the key one that is crucial for tumor development. High expression of SQLE is frequently observed in many human cancers and is associated with poor patient outcomes (Helms et al, 2008; Brown et al, 2016; Stopsack et al, 2016; Liu et al, 2018). Moreover, abnormal elevation of SQLE is responsible for hepatic cholesterol accumulation and accelerates the nonalcoholic fatty liver disease (NAFLD)-associated hepatocellular carcinoma (HCC) development (Liu et al, 2018). However, how tumor cells augment SQLE expression to reprogram cholesterol metabolism still remains unclear. Tumor suppressor p53, the most frequent mutant gene in human cancer, controls a wide variety of biological processes, including apoptosis, cell-cycle arrest, and senescence (Vousden & Prives, 2009). However, it appears that manipulation of antioxidant function and metabolism regulation is more critical for the tumor-suppressive function of p53 (Li et al, 2012; Valente et al, 2013; Kastenhuber & Lowe, 2017). Numerous studies suggest that p53 plays an important role in regulating glucose, lipid, amino acid, as well as other metabolic pathways (Vousden & Ryan, 2009; Floter et al, 2017; Liu et al, 2019; Lahalle et al, 2021; Liu & Gu, 2021). Hepatic p53 has been recognized as an important regulator of different liver diseases, such as NAFLD development, hepatic insulin resistance, nonalcoholic steatohepatitis, HCC development, and liver regeneration (Krstic et al, 2018). Interestingly, recent study reveals that under low-sterol conditions p53 represses cellular mevalonate pathway to mediate liver tumor suppression through inhibition of SREBP2 maturation (Moon et al, 2019). Intriguingly, p53 unexpectedly functions in promoting hepatocellular carcinoma (HCC) tumorigenesis through inducing PUMA-dependent suppression of oxidative phosphorylation (Kim et al, 2019). Thus, the role of p53 in HCC is contentious and needs further investigation. Here we report that, under normal-sterol conditions, p53 has a role in repressing cholesterol accumulation and liver tumor growth through transcriptional repression of SQLE, a key metabolic enzyme in cholesterol synthesis. We also provide an evidence that p53 is capable to directly bind to SQLE gene in a SREBP2-independent manner. Thus, our findings, together with others (Moon et al, 2019), uncover a strong surveillance capability of p53 in guarding cholesterol synthesis pathway under both low-sterol and normal-sterol conditions. Results p53 suppresses cholesterol synthesis To assess the role of p53 in NAFLD-induced HCC, we firstly evaluated the role of p53 in NAFLD. p53 wild-type (p53+/+) and knockout (p53−/−) mice were fed with normal diet (Normal) or high-fat diet (HFD) starting at the age of 8 weeks. All mice were weighed every week. Mice were sacrificed at the age of 16 weeks, and the livers were analyzed. p53−/− mice showed increased liver weight and body weight with HFD (Fig 1A–C and Fig EV1A and B). We also valued the effect of p53 on the lipid droplets formation in liver. p53−/− mice liver tissues had more lipid compared with the liver tissues from p53+/+ mice fed with HFD (Fig 1D). p53−/− mice showed an increase in both serum and hepatic cholesterol concentrations compared with p53+/+ mice with either normal diet or HFD (Fig 1E and F). Similarly, higher levels of hepatic triglyceride were observed in p53−/− mice (Fig 1G). Moreover, HFD-treated mice showed increased hepatic cholesterol accumulation and triglyceride concentrations compared to normal-treated mice (Fig 1F and G). These data indicate that p53 restricts cholesterol accumulation. To further determine the function of p53 in cholesterol metabolism, we knocked out p53 using CRISPR/Cas9 system in human hepatocellular carcinoma cell line HepG2. p53 knockout augmented cholesterol accumulation (Fig 1H). We also knocked down p53 using two different sets of small interfering RNA (siRNA) in other two HCC cell lines SK-HEP-1 and BEL-7402. p53 deficiency led to increased cholesterol concentration compared with their wild-type counterparts (Fig EV1C and D). Next, we examined the cholesterol levels in isogenic p53+/+ and p53−/− human colon cancer HCT116 cells (Bunz et al, 1998). Cholesterol concentration increased in p53−/− cells compared to p53+/+ cells (Fig 1I and J). These data suggest p53 suppresses cholesterol accumulation both in vivo and in vitro. Figure 1. p53 suppresses cholesterol biosynthesis A–G. p53+/+ and p53−/− C57BL/6N male mice were treated as in (A). Data are means ± s.d. (n = 6 to 8). (B) Representative liver photos of mice livers. (C) Representative changes in liver weight. (D) H&E (hematoxylin-eosin) staining (left) and oil red O staining (right) of livers. Scale bar, 50 μm. (E) Serum cholesterol concentrations of each group. Liver cholesterol concentrations (F) and triglyceride levels (G) of each group were examined. H, I. Cholesterol levels in p53+/+ and p53−/− HepG2 cells (H) or HCT116 cells (I) were examined. Protein expression was shown by Western blotting (bottom panel). J. Cholesterol concentrations of p53+/+ and p53−/− HCT116 cells were determined by Filipin III staining. K. Heat map analysis of 17 sterol biosynthesis genes and 3 other SREBP2 target genes from RNA-seq data using mice livers of p53+/+ and p53−/− mice fed with normal or HFD diet. Expression levels were normalized to the mean level of each gene among all samples and compared to p53+/+ normal diet mice. Color scale indicates the expression fold change of target gene. L. Enrichment of cholesterol biosynthesis genes in the liver of normal diet (left) and HFD diet (right). FDR: false discovery rate; NES: normalized enrichment score. Data information: In (C, E, F, G, H, I, J), bars represent mean ± s.d., *P < 0.05; **P < 0.01; ***P < 0.001; for (B, D), n = 6–8 biologically independent samples; for (F, G, H, I, J) n = 3 biologically independent samples; statistical significance was determined by two-tailed unpaired t-test. Download figure Download PowerPoint Click here to expand this figure. Figure EV1. p53 restricts cholesterol synthesis A. p53 wild-type and knockout mice were examined by genotyping. p53+/+ product size: 281 bp. p53−/− product size: 441 bp (left). Representative mice photos are shown (right). B. Weekly body weight gain of mice is shown. C, D. Cholesterol concentration of SK-HEP-1 (C) and BEL-7402 (D) cells treated with control or two sets of p53 siRNA for 72 h. E, F. Differential expression genes from RNA-seq of mice livers from p53+/+ and p53−/− mice fed with normal (E) or HFD (F) diet were analyzed by REACTOME Database Enrichment analysis. Dot plots show the 20 most significantly enriched gene biological process. Dot sizes represent counts of enriched differential expressed genes. Dot color scale changes from red to blue, with blue indicating lower adjusted P-value (Padj) for the category. Data information: (C, D) Bars represent mean ± s.d., **P < 0.01; ***P < 0.001; (B) n = 6-8 biologically independent samples; (C, D) n = 3 biologically independent samples; statistical significance was determined by two-tailed unpaired t-test. Download figure Download PowerPoint To investigate the mechanism by which p53 regulates the cholesterol metabolism, we performed RNA-seq using the liver mice samples in Fig 1A. Reactome pathway enrichment analysis showed different genes could be targeted by p53, and notably, steroids metabolism and cholesterol biosynthesis were significantly enriched under both normal diet and HFD (Fig EV1E and F). Mevalonate pathway is a route to produce sterols in mammalian cells. As shown in Fig 1K, expression of genes in mevalonate pathway mostly increased in p53−/− mice. Moreover, gene set enrichment analysis of the genome-wide dataset revealed that p53 expression correlated with decreased gene signature of cholesterol biosynthesis (Fig 1L). These data suggest that p53 may have a role in repressing expression of genes involved in cholesterol biosynthesis. p53 represses SQLE expression in a SREBP2-independent manner under normal-sterol conditions To investigate the mechanism(s) for p53 in regulating cholesterol biosynthesis, we performed RNA-seq using isogenic p53+/+ and p53−/− human colon cancer HCT116 cells. Consistent with previous data, p53 deficiency increased expression of genes in mevalonate pathway. Of note, SQLE expression was vastly augmented in p53−/− cells (Fig 2A). SQLE is the first monooxygenase and a rate-limited enzyme in cholesterol synthesis pathway. We next examined how p53 modulates SQLE expression. Hepatic SQLE mRNA levels were markedly higher in p53−/−mice in comparison to p53+/+ mice (Fig 2B). We also examined the SQLE expression in various tissues from p53−/− and p53+/+ mice. The tissues from p53−/− mice-including liver, brain, spleen, and colon had higher levels of SQLE expression, compared with those in the corresponding tissues from p53+/+ mice (Fig 2C). Similarly, p53 deficiency increased both protein and mRNA levels of SQLE in several cell lines (Figs 2D and E and EV2A–D). These data suggest p53 suppresses SQLE expression. Figure 2. A SREBP2-independent mechanism for p53-mediated SQLE inhibition A. Heat map analysis of 17 sterol biosynthesis genes and 3 other SREBP2 target genes from RNA-seq data using p53+/+ and p53−/− HCT116 cells. Expression levels were normalized to the mean level of each gene among all samples and compared to p53+/+ cells. Color scale indicates the expression fold change of target gene. B. mRNA expression of SQLE in the mouse liver of p53+/+ and p53−/− mice with normal or HFD diet. C. mRNA and protein levels of SQLE in different tissues of p53+/+ or p53−/− mice were determined by qRT–PCR and Western blot respectively. D, E. SQLE protein expression and mRNA levels were examined in p53+/+ and p53−/−HCT116 cells (D) and HepG2 cells (E). Actin was used as loading control. F, G. p53+/+ and p53−/− HepG2 cells were cultured for 48 h in medium containing fetal bovine serum (Serum) or lipoprotein-depleted FBS (LPDS). Protein expression was shown by Western blotting (F). mRNA levels of SQLE were examined by qRT–PCR (G). P, premature SREBP2; M, mature SREBP2. H–J. Protein expression and mRNA levels of p53+/+ and p53−/− HepG2 cells treated with control siRNA or three separated sets of SREBP2 siRNAs for 72 h as indicated. K. p53+/+ and p53−/− HepG2 cells were treated with control siRNA or SREBP2 siRNA, followed by ectopically expressed RNA-resistant SREBP2 for 48 h. mRNA and protein expression were examined by qRT–PCR and Western blotting. Data information: In (B, C, D, E, G, H, I, J, K), bars represent mean ± s.d., *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; for (B), n = 4 biologically independent samples; for (C, D, E, G, H, I, J, K), n = 3 biologically independent samples; statistical significance was determined by two-tailed unpaired t-test. Download figure Download PowerPoint Click here to expand this figure. Figure EV2. p53 inhibits SQLE expression A–C. mRNA and protein expression of SQLE in BEL-7402 (A) and SK-HEP-1 (B and C) treated with control or p53 siRNA for 48 h. D. SQLE protein expression and mRNA levels in MCF-7 cells stably expressing control or p53 shRNA. E. p53+/+ and p53−/− HCT116 cells were treated with Fatostatin (10 μM) for 24 h. mRNA and protein expression were analyzed by qRT–PCR and Western blotting respectively. F. H1299 p53-inducible cells treated with or without doxycycline (1 μg/ml) were transfected with control siRNA or SREBP2 siRNA for 48 h. mRNA and protein expression were examined as indicated. G, H. Control and SREBP2 knockout MCF-7 cells (G) or SK-HEP-1 cells (H) using sgRNA CRISPR/Cas9 were treated with control siRNA or p53 siRNA for 48 h. mRNA and protein expression were examined as indicated. I. mRNA and protein expression of SQLE and p53 in MCF-7 (wild-type p53), MDA-MB-468 (mutant p53 R273H), SK-BR-3 (mutant p53 R175H), and MDA-MB-435s (mutant p53 G266E) cells treated with p53 or control siRNA for 48 h. J. H1299 cells were transfected with increasing amounts of control plasmid or plasmid expressing wild-type (PRK5-flag p53) or mutant p53 (PRK5-flag-p53-G266E, PRK5-HA-p53-R273H, and PRK5-HA-p53-R175H) for 24 h as indicated. Cell lysates were analyzed by Western blotting. Relative SQLE/actin ratios are shown below. Data represent three independent experiments. K. Human HCC TCGA datasets were analyzed to determine whether tumors bearing wild-type p53 correlate with lower expression of SQLE. Patients were classified by p53 status (wild-type versus mutant). SQLE exhibited higher expression levels in p53 mutant (mut, n = 220) HCC tumors compared with wild-type (n = 508) p53 tumors. L. SQLE expression in p53 wild-type (n = 307) and p53 mutant (mut, n = 795) human breast tumors (BRCA) of the TCGA database. Data information: (A–I) Bars represent mean ± s.d., *P < 0.05; **P < 0.01; ***P < 0.001; n = 3 biologically independent samples; statistical significance was determined by two-tailed unpaired t-test. Download figure Download PowerPoint To examine whether p53-mediated SQLE suppression is related to sterol conditions, we cultured p53+/+ and p53−/− HepG2 cells in medium containing normal-sterol fetal bovine serum (Serum) or lipoprotein-deficient serum (LPDS). As shown in Fig 2F, when cells were cultured in LPDS medium, more precursor SREBP2 (P) was cleaved to active mature form of SREBP2 (M). Consistent with previous findings (Moon et al, 2019), p53−/− cells showed more active mature form of SREBP2 (M) than p53+/+ cells, which led to increased SQLE expression under sterol-depleted conditions (Fig 2F and G), Interestingly, p53 loss also led to increased levels of SQLE when cells were cultured in normal serum medium, but had no effect on SREBP2 maturation (Fig 2F and G). These data indicate there is a SREBP2-independent mechanism for p53-mediated SQLE suppression under normal-sterol conditions. To evaluate whether SREBP2 is involved in the regulating SQLE by p53 under normal-sterol conditions, we knocked down SREBP2 using multiple sets of siRNAs. Surprisingly, upregulation of SQLE by p53 deficiency also occurred in SREBP2-depleted cells (Fig 2H–J). Enforced expression of RNAi-resistant SREBP2 cDNA in SREBP2-depleted cells augmented SQLE expression, while p53-mediated SQLE downregulation still existed in these cells (Fig 2K). Consistent with this, SREBP2 inhibitor fatostatin (Kamisuki et al, 2009) failed to sufficiently block SQLE expression induced by p53 loss, because both SQLE mRNA and protein levels were still higher in p53−/− cells than those in p53 WT cells (Fig EV2E). Similar results were obtained when we used a p53 Tet-on expression system in a p53-null lung cancer cell line H1299. Doxycycline-induced ectopic p53 expression still reduced SQLE levels even in the absence of SREBP2 (Fig EV2F). Moreover, we generated SREBP2 knockout cells using CRISPR/Cas9 system. Concordant with SREBP2 siRNA data, p53-mediated inhibition of SQLE expression still exhibited in SREBP2 knockout cells (Fig EV2G and H). These data indicate that p53 represses SQLE expression in a SREBP2-independent manner under normal-sterol conditions. Mutant p53 enhances mevalonate pathway through SREBP2 (Freed-Pastor et al, 2012). Next, we examined the effect of mutant p53 on SQLE expression using four different breast cancer cell lines which carry wild-type p53 or mutant p53. Conformably, wild-type p53 inhibited SQLE expression, whereas mutant p53 increased the expression of SQLE (Fig EV2I). Moreover, we overexpressed exogenous wild-type or mutant p53 in H1299 cells. Unlike wild-type p53, mutant p53 failed to suppress SQLE expression but enhanced the expression of SQLE in a dose-dependent manner (Fig EV2J). To further investigate the clinical relevance between SQLE expression and p53 status, we analyzed a human HCC database and a human BRCA (The Cancer Genome Atlas, TCGA). SQLE expression levels were higher in human carcinomas harboring p53 mutations (mut) than those with wild-type (wt) p53 (Fig EV2K and L). These findings support that wild-type p53 inhibits SQLE expression, while cancer-derived p53 mutants upregulate SQLE expression. SQLE is a transcriptional target of p53 Under lower-sterol conditions, SREBP2 is responsible for p53-mediated suppression of mevalonate pathway (Moon et al, 2019). Thus, we wanted to know how p53 inhibits SQLE expression under normal-sterol conditions. To investigate whether the transcriptional activity of p53 was required to p53 mediate SQLE suppression, we used an inhibitor of p53 transcriptional activity, pifithrin-α (PFTα) (Komarov et al, 1999). PFTα restored p53-inhibited SQLE expression. As a control, p53-induced expression of p21 was inhibited by PFTα (Fig EV3A). The mutant p53 (R175H) that lost DNA binding ability failed to repress SQLE expression (Fig EV3B). These data indicate p53 requires transcriptional activity to repress SQLE. Next, we examined whether SQLE is a transcriptional target of p53. We analyzed the human SQLE gene sequence for potential p53 response elements (Riley et al, 2008). We identified two putative p53 response elements (RE1 and RE2) in the first intron of human SQLE gene (Fig 3A). Chromatin immunoprecipitation assays in HCT116 cells revealed that p53 bound to the genomic region of RE2, but not RE1 (Fig 3B). Moreover, PFTα reduced the amount of p53 bound to SQLE-RE2 as well as p21-RE (Fig EV3C). To determine if SREBP2 is also involved in p53-mediated SQLE inhibition, we knocked down SREBP2 expression using siRNA. SREBP2 depletion had no effect on the recruitment of p53 to SQLE genomic region RE2 (Fig 3C). Similar result was obtained in SREBP2 knockout cells using CRISPR/Cas9 system (Fig 3D). Furthermore, to investigate whether p53 directly binds to SQLE genomic region RE2, we performed EMSA assay. A specific band was observed in RE2, but not RE2 mut in the presence of nuclear extracts. Super-shift band with anti-p53 antibodies identified p53 as the protein present in the EMSA band (Fig 3E). Taken together, these results demonstrate that p53 directly binds to the genomic region RE2. Click here to expand this figure. Figure EV3. Murine p53 transcriptionally regulates mouse SQLE p53+/+ and p53−/− HepG2 cells were treated with PFTα (20 μM) for 24 h. mRNA and protein expression were analyzed by qRT–PCR and Western blotting respectively. mRNA and protein expression of H1299 p53-inducible cells (wild-type versus R175H mutant) treated with or without doxycycline (1 μg/ml). p53+/+ and p53−/− HepG2 cells treated with or without PFTα (20 μM) for 24 h were analyzed by chromatin immunoprecipitation (ChIP) assay using normal IgG and anti-p53 antibody as indicated. Schematic representation of mouse SQLE genomic structure. Shown are the exon/intron organization and two potential p53 binding sites (BS1 and BS2) and the corresponding mutant binding sites. p53+/+ and p53−/− mice livers were analyzed by ChIP assay using normal IgG and anti-p53 antibody. p53+/+ mice livers from Normal or HFD mice were analyzed by ChIP assay using normal IgG and anti-p53 antibody. Luciferase reporter constructs containing mouse SQLE potential binding sites BS1 and BS2 were transfected into HEK293T cells together with control or mouse p53 expression vector for 48 h. Renilla vector pRL-CMV was used as a transfection internal control. Relative levels of luciferase are shown. Protein expression is shown. Luciferase reporter constructs containing mouse SQLE potential binding sites BS1, BS2, and their corresponding mutant binding sites (BS1 mut and BS2 mut) were transfected into HEK293T cells together with control or mouse p53 expression vector for 48 h. Renilla vector pRL-CMV was used as a transfection internal control. Relative levels of luciferase are shown. Protein expression is shown. Data information: (A, B, C, E, F, G, H) Bars represent mean ± s.d., *P < 0.05; ***P < 0.001; ****P < 0.0001; n = 3 biologically independent samples; statistical significance was determined by two-tailed unpaired t-test. Download figure Download PowerPoint Figure 3. SQLE is a transcriptional target of p53 Schematic representation of human SQLE genomic structure. Shown are the exon/intron organization and two potential p53 response elements (RE1 and RE2) and the corresponding mutant response elements. p53+/+ and p53−/− HCT116 cells were analyzed by chromatin immunoprecipitation (ChIP) assay using normal IgG and anti-p53 antibody as indicated. p53+/+ HCT116 cells transfected with control siRNA or SREBP2 siRNA for 72 h were analyzed by ChIP assay with the indicated antibodies. Control and SREBP2 knockout MCF-7 cells using sgRNA CRISPR/Cas9 were analyzed by ChIP assay using anti-p53 or normal mouse IgG antibodies. The electrophoretic mobility shift assay (EMSA) of SQLE-RE2 or RE2 mut in the presence or absence of p53 antibodies as indicated. Blue box (shift band) means the binding between nuclear extracts and RE2, red box (super-shift band) means p53 as the protein presented in the EMSA band. Luciferase reporter constructs containing RE1, RE2, RE1mut, or RE2mut were transfected into p53−/− HCT116 cells together with control (PRK5-flag vector) or p53 expression vector (PRK5-flag p53) for 48 h. Renilla vector pRL-CMV was used as a transfection internal control. Relative levels of luciferase are shown. Data represent three independent experiments. Protein expression is shown. Luciferase reporter constructs containing RE1, RE2, RE1mut, or RE2mut were transfected into HEK293T cells together with control (PRK5-flag vector) or p53 expression vector (PRK5-flag p53) for 48 h. Renilla vector pRL-CMV was used as a transfection internal control. Relative levels of luciferase are shown. Protein expression is shown.