Genomic profiling of the transcription factor Zfp148 and its impact on the p53 pathway

Zou, Zhiyuan V.; Gul, Nadia; Lindberg, Markus; Bokhari, Abdulmalik A.; Eklund, Ella M.; Garellick, Viktor; Patel, Angana A. H.; Dzanan, Jozefina J.; Titmuss, Ben O.; Le Gal, Kristell; Johansson, Inger; Tivesten, Åsa; Forssell-Aronsson, Eva; Bergö, Martin O.; Staffas, Anna; Larsson, Erik; Sayin, Volkan I.; Lindahl, Per

doi:10.1038/s41598-020-70824-2

Download PDF

Article
Open access
Published: 25 August 2020

Genomic profiling of the transcription factor Zfp148 and its impact on the p53 pathway

Zhiyuan V. Zou¹,
Nadia Gul^1,2,3,
Markus Lindberg⁴,
Abdulmalik A. Bokhari⁴,
Ella M. Eklund^2,3,
Viktor Garellick¹,
Angana A. H. Patel^2,3,
Jozefina J. Dzanan^2,3,
Ben O. Titmuss^2,3,
Kristell Le Gal^2,3,
Inger Johansson¹,
Åsa Tivesten¹,
Eva Forssell-Aronsson^5,6,
Martin O. Bergö^7,8,
Anna Staffas⁹,
Erik Larsson⁴,
Volkan I. Sayin^2,3 &
…
Per Lindahl^1,4

Scientific Reports volume 10, Article number: 14156 (2020) Cite this article

2011 Accesses
4 Citations
1 Altmetric
Metrics details

Subjects

Abstract

Recent data suggest that the transcription factor Zfp148 represses activation of the tumor suppressor p53 in mice and that therapeutic targeting of the human orthologue ZNF148 could activate the p53 pathway without causing detrimental side effects. We have previously shown that Zfp148 deficiency promotes p53-dependent proliferation arrest of mouse embryonic fibroblasts (MEFs), but the underlying mechanism is not clear. Here, we showed that Zfp148 deficiency downregulated cell cycle genes in MEFs in a p53-dependent manner. Proliferation arrest of Zfp148-deficient cells required increased expression of ARF, a potent activator of the p53 pathway. Chromatin immunoprecipitation showed that Zfp148 bound to the ARF promoter, suggesting that Zfp148 represses ARF transcription. However, Zfp148 preferentially bound to promoters of other transcription factors, indicating that deletion of Zfp148 may have pleiotropic effects that activate ARF and p53 indirectly. In line with this, we found no evidence of genetic interaction between TP53 and ZNF148 in CRISPR and siRNA screen data from hundreds of human cancer cell lines. We conclude that Zfp148 deficiency, by increasing ARF transcription, downregulates cell cycle genes and cell proliferation in a p53-dependent manner. However, the lack of genetic interaction between ZNF148 and TP53 in human cancer cells suggests that therapeutic targeting of ZNF148 may not increase p53 activity in humans.

Transient loss of Polycomb components induces an epigenetic cancer fate

Article Open access 24 April 2024

Targeting DCAF5 suppresses SMARCB1-mutant cancer by stabilizing SWI/SNF

Article 27 March 2024

Massively parallel screen uncovers many rare 3′ UTR variants regulating mRNA abundance of cancer driver genes

Article Open access 18 April 2024

Introduction

Activation of the tumor suppressor p53 may have beneficial effects on tumors with wild-type p53¹. The first drugs targeting the p53 repressor murine double minute 2 (MDM2) have entered clinical trials, but their clinical benefit is still under evaluation². Because knockout of Mdm2 in mice causes widespread and lethal activation of p53 in normal tissues³, efficient pharmacological inhibition of MDM2 may have adverse effects. Thus, identifying alternative ways to activate the p53 pathway with less adverse effects is warranted.

Zinc finger protein 148 (Zfp148, Zbp-89, BFCOL, BERF1, htβ) is a transcription factor that binds to GC-rich DNA sequences, thus activating or repressing transcription of target genes^{4,5,6,7,8,9,10,11}. Earlier studies have shown that Zfp148 is a potent repressor of p53 activity in mice. Firstly, deletion of Zfp148 caused p53-dependent proliferation arrest of cultured mouse embryonic fibroblasts (MEFs) and prenatal lung tissue that was rescued by reducing oxidative stress¹². Moreover, deletion of one copy of Zfp148 in the Apc^Min/+ model of intestinal adenomas reduced tumor numbers and increased survival by increasing p53 activity¹³. In line with this, conditional deletion of one or both alleles of Zfp148 in the gut epithelium of Apc^FL/+ mice reduced tumor formation through a mechanism involving β-catenin¹⁴. Finally, deletion of one copy of Zfp148 reduced proliferation of macrophages and atherosclerosis in Apoe^−/− mice by increasing p53 activity¹⁵. These results indicate that deletion of one copy of Zfp148 is sufficient to induce therapeutically meaningful p53 activation. This is important because mice lacking one copy of Zfp148 have a normal life span and appear to be healthy. Therapeutic targeting of Zfp148 could therefore protect against cancer or atherosclerosis by increasing p53 activity without causing detrimental side effects.

The mechanism leading to increased p53 activity is not well understood. Zfp148 interacts physically with p53¹⁶ raising the possibility that Zfp148 regulates p53 by protein–protein interaction. However, indirect mechanisms are equally possible. Here, we investigate mechanisms behind cell cycle arrest of Zfp148-deficient (Zfp148^gt/gt) MEFs by using global gene expression analysis, targeted inactivation of Cdkn2a genes, genome-wide chromatin immunoprecipitation, and public CRISPR and siRNA screen data.

Results

Zfp148 deficiency downregulates expression of E2F-responsive cell cycle genes

To define mechanisms that cause proliferation arrest of Zfp148^gt/gt MEFs, we used transcript profiling to identify genes that were differentially expressed between Zfp148^gt/gt MEFs and wild-type controls. We also compared Zfp148^gt/gt MEFs infected with adenovirus carrying Zfp148 cDNA or empty vector to verify that differentially expressed genes depended on Zfp148. The analysis revealed a set of more than 300 genes that were downregulated in Zfp148^gt/gt MEFs compared to wild-type controls, and rescued in Zfp148^gt/gt MEFs after adenoviral expression of Zfp148 cDNA compared to empty vector (Fig. 1A and Supplementary Data S1). Overrepresentation analysis showed strong enrichment of cell cycle-related genes (Fig. 1B). Since this category of genes is regulated by E2F transcription factors, we identified known E2F targets from earlier publications^17,19 and investigated their rank in the gene list. The majority of known E2F targets clustered at the top of the list confirming that Zfp148 deficiency downregulated E2F-dependent cell cycle genes (Fig. 1A).

To investigate the impact of p53 on gene expression, we repeated the analysis with RNA extracted from MEFs generated on a Trp53-null genetic background. Downregulation of the cell cycle-enriched gene set was markedly attenuated on this background (Fig. 1A), confirming that p53 plays a dominant role. However, the rank of differentially expressed genes was similar to the rank obtained on Trp53 wild-type background (Spearman correlation, r = 0.56, P < 0.0001) with strong enrichment of cell cycle-related gene ontology (GO) terms among the downregulated genes (Fig. 1B). We therefore conclude that Zfp148 deficiency downregulates cell cycle genes by both p53-dependent and p53-independent mechanisms.

The previous data panels show qualitative changes. To obtain quantitative data, we determined mRNA levels with real-time RT PCR for 17 genes arbitrarily selected from the downregulated gene set. Expression of genes related to the GO term cell cycle was reduced by half in Zfp148^gt/gt MEFs compared to controls (Fig. 1C). Deletion of one copy of Trp53 is sufficient to rescue cell proliferation of Zfp148-deficient MEFs¹². In line with this, expression of 11 genes was restored to near normal levels in Zfp148^gt/gtTrp53^+/− compared to Zfp148^gt/gt MEFs, and 9 of those genes were linked to the GO-term cell cycle (Fig. 1C). Only 1 of the 6 remaining genes that were not restored on Trp53^+/− background was linked to the cell cycle. These results suggest that p53-mediated repression of cell cycle genes causes the proliferative arrest of Zfp148^gt/gt MEFs.

To assess whether downregulation of the gene set observed in response to Zfp148 deficiency in MEFs also occurred in vivo, we isolated RNA from neonatal lungs and brains and from bone marrow-derived macrophages of Zfp148^gt/gt and wild-type mice. We determined gene expression by microarrays and showed that the gene set was markedly downregulated compared to wild-type controls in Zfp148^gt/gt lungs but not in brains or macrophages (Fig. 1D). Previous studies showed that Zfp148 deficiency induces cell cycle arrest in prenatal lungs in a p53-dependent manner¹². In contrast, brains and bone marrow derived macrophages from Zfp148 deficient mice displayed no overt phenotypes^12,20,21. Thus, the observed downregulation of cell cycle genes in lungs but not brain or bone marrow derived macrophages from Zfp148 deficient mice compared to controls makes sense (Fig. 1D).

The Cdkn2a transcript ARF is required for cell proliferation arrest of Zfp148^gt/gt MEFs

The downregulation of cell cycle genes in Zfp148^gt/gt MEFs by p53-dependent and -independent mechanisms prompted us to investigate mRNA levels of the cyclin-dependent kinase inhibitor Cdkn2a. Cdkn2a produces two gene products, ARF and p16, which downregulate cell cycle genes by activating the p53 pathway and inactivating the CDK4/6 complex, respectively (Fig. 2A). ARF and p16 mRNA were markedly increased in Zfp148^gt/gt MEFs compared to controls (Fig. 2B), supporting a role for Cdkn2a in the downregulation of cell cycle genes.

We have previously demonstrated that Zfp148 deficiency induces cell proliferation arrest in MEFs after a few passages, and that p53 is required for the arrest¹². To further investigate the role of Cdkn2a in the proliferation arrest, Zfp148^gt/+ mice were bred on a Cdkn2a^+/− background to generate Zfp148^gt/gtCdkn2a^+/− MEFs. Deletion of one copy of Cdkn2a restored normal proliferation of Zfp148^gt/gt MEFs (Fig. 2C), demonstrating that Cdkn2a is required for proliferation arrest of these cells. To determine the relative importance of ARF and p16 for the proliferation arrest, we used lentiviral guide-RNAs and CRISPR to generate Zfp148^gt/gt MEFs lacking either ARF or p16 (Fig. 2D, E and Supplementary Fig. S1). Selective ARF knockout rescued growth of Zfp148^gt/gt cells (Fig. 2F) despite increased expression of p16, which is an expected consequence of ARF inactivation²². Knockout of p16 had no effect on cell growth (Fig. 2F). Collectively, these data show that ARF but not p16 is required for p53 activation and proliferation arrest of Zfp148^gt/gt MEFs.

Zfp148 binds to clustered cytosine-rich DNA elements in basal promoters of other transcription factors

To understand the primary basis of ARF and p53 activation in Zfp148^gt/gt MEFs, we used chromatin immunoprecipitation-on-chip (ChIP-chip) assays to identify target sites of FLAG-tagged Zfp148 (AdZfp148FLAG) in Zfp148^gt/gtTrp53^−/− MEFs. The experiment was performed on a Trp53-null background to overcome senescence and obtain sufficient amounts of starting material. The CMV-driven adenovirus vector used produces Zfp148 protein amounts that are similar to those found in wild-type MEFs¹².

Zfp148 target sites were recorded in four independent experiments with a strong preference for binding close to transcription starts (Fig. 3A and Supplementary Data S2). To determine the DNA-binding preferences of Zfp148, we constructed a position weight matrix (PMW) model from 16 known Zfp148 binding sites (Table 1). As expected, the model predicted a cytosine-rich binding motif (Fig. 3B). We next applied the model to DNA regions surrounding known transcription starts. While there was no correlation between peak positions of ChIP-chip data and PWM-predicted single motifs, Zfp148 showed a strong preference for binding to regions with clustered PWM-predicted motifs (Fig. 3C,D). These data indicate that Zfp148 binds to clustered motifs in cytosine-rich basal promoters.

Table 1 Zfp148 binding sites used to construct a position weight matrix (PMW) model.

Full size table

Overrepresentation analysis of 3,488 genes that bound to Zfp148 at 1% false discovery rate in all experiments showed strong enrichment of functional terms. Binning genes according to their rank in the ChIP-chip experiment revealed a striking correlation between binding scores and the enrichment of functional terms; the enrichment declined abruptly after 600 genes (Spearman correlation r = 0.74, P < 0.001, Fig. 3E). Since functional connectivity is an expected property of co-regulated genes, the number of Zfp148 target genes is likely much lower than the number of binding genes. The top-ranked genes (top 600) were enriched for functional terms related to transcriptional regulation, suggesting that Zfp148 mainly operates upstream of other transcription factors (Fig. 3F). There was no overrepresentation of cell cycle-related terms.

By integrating ChIP-chip and gene expression data, we confirmed a small but significant upregulation of Zfp148 target genes in Zfp148^gt/gt MEFs after infection with AdZfp148 (Fig. 3G). However, there was no significant overlap between Zfp148-binding genes and the set of cell cycle-enriched genes that was downregulated in Zfp148^gt/gt MEFs (Fig. 3H). Notably, Zfp148 bound to the ARF promoter in all experiments with binding regions centered at 800 base pairs upstream of the transcription start (Fig. 3I) suggesting that Zfp148 may repress ARF transcription. However, the preferential binding of Zfp148 to promoters of other transcription factors indicates that deletion of Zfp148 could have pleiotropic effects that activate ARF and p53 indirectly.

No genetic interaction between ZNF148 and TP53 in human cancer cell lines

Since the integrated transcriptomic and promoter occupancy analysis suggested pleiotropic and indirect mechanisms of p53 and ARF activation in Zfp148^gt/gt MEFs, we decided to explore the genetic interaction between ZNF148 (the human orthologue of Zfp148) and TP53 in CRISPR and siRNA data from human cancer cell lines.

We downloaded CRISPR and siRNA scores for all genes in 487 human cancer cell lines from the Cancer Dependency Map database (https://depmap.org). A negative score indicates that knockout (CRISPR) or knockdown (siRNA) of that gene has a negative impact on cell growth or survival. Genes with CRISPR scores below − 0.6 are considered to be essential²³. The median CRISPR and siRNA scores for ZNF148 were 0.05 (± 0.01 95% CI) and − 0.12 (± 0.02 95% CI), suggesting that cancer cells are marginally affected by ZNF148 inactivation (Fig. 4A). ZNF148 was not essential for the growth or survival of any of the tested cell lines (Fig. 4B).

We next divided the cell lines into two groups depending on their TP53 genotype. As expected, inactivation of TP53 increased the growth of cells with wild-type TP53 but had no impact on cells with TP53 mutations (Fig. 4C). The p53 repressor MDM2 showed a reverse pattern with low scores in TP53 wild-type cells and scores near zero in cells with TP53 mutations (Fig. 4D). In contrast, there was no marked difference in ZNF148 scores between TP53 wild-type and mutant cell lines (Fig. 4E).

We finally ranked the genes according to their similarity with TP53 as judged by the Pearson correlation of CRISPR or siRNA scores across the set of human cancer cell lines. Canonical activators of the p53 pathway such as USP28, CDKN1A, and TP53BP1 showed strong correlation with TP53 and were positioned at the top of the list, while repressors including MDM2, PPM1D, and MDM4 were positioned at the end (Fig. 4F). In contrast, ZNF148 did not show strong correlation or anti-correlation with TP53 (Fig. 4F), suggesting that there is no genetic interaction between ZNF148 and TP53 in human cancer cell lines under standard culture conditions.

To exclude the possibility that ZNF148 exerts mild regulation of p53 that is compensated under standard conditions, we investigated the role of ZNF148 under DNA damaging conditions. We used CRISPR-Cas9 to delete ZNF148 in H838, A549, and H1299 (TP53 null) human lung cancer cells (Fig. 5A and Supplementary Fig. S3), and confirmed that p53-induction by etoposide was intact (Fig. 5B and Supplementary Fig. S3). To uncover quantitative differences in the DNA damage response, we established dose–response curves for cisplatin and etoposide that activates p53 through ATM/ATR, and for TH588 that activates p53 though USP28²⁴. There was no difference in IC50 values between ZNF148 knockout cells and controls in any of the tested cell lines (Fig. 5C).

To assess the role of Zfp148 under DNA damaging conditions in vivo, we exposed Zfp148 heterozygous mice and controls to ionizing radiation and quantified the depletion of immature thymocytes, a process that is entirely p53-dependent²⁵. As expected, ionizing radiation reduced the number of thymocytes in a dose dependent manner (Fig. 5D). However, there was no difference in the total number of thymocytes, or the distribution of different thymocyte subsets, between Zfp148^gt/+ and control mice 72 h after exposure to ionizing radiation (Fig. 5D). We conclude that Zfp148 does not regulate the p53-pathway under DNA damaging conditions.

Discussion

We have previously shown that Zfp148 deficiency arrests proliferation in MEFs by activating p53¹², and here we investigated the underlying mechanism by analyzing the transcriptional output and promoter occupancy of Zfp148 in MEFs. Our results suggest that Zfp148 deficiency, by derepressing ARF transcription, downregulates cell cycle genes and cell proliferation in a p53-dependent manner. The model is supported by three lines of evidence. Firstly, Zfp148 deficiency downregulated cell cycle genes in a p53-dependent and -independent manner, with p53 playing a dominant role. Secondly, proliferation arrest of these cells required increased expression of the Cdkn2a transcript ARF, which is a major regulator of the p53 pathway. Finally, Zfp148 interacted physically with the ARF promoter, indicating that Zfp148 may regulate ARF transcription.

Zfp148 did not bind to promoters of cell cycle genes more often than predicted by chance, suggesting that regulation of cell cycle genes is indirect. p53 can downregulate cell cycle genes by inducing the cyclin-dependent kinase inhibitor p21, which is also regulated by ZNF148^8,9. Although regulation of the p21 promoter by ZNF148 could theoretically explain the p53-independent repression of cell cycle genes, the promoter was negatively regulated by ZNF148 (in response to butyrate) in most published experiments^8,9, which would not explain our current finding. Given that the second transcript of the Cdkn2a locus, p16, was markedly increased in Zfp148^gt/gt MEFs compared to controls, a more likely explanation is that p53-independent downregulation of cell cycle genes is mediated by p16.

The cell cycle arrest in Zfp148^gt/gt MEFs was rescued by heterozygous deletion of Cdkn2a or selective knockout of the ARF transcript, indicating a plausible mechanism for p53 activation by ARF. In contrast to homozygous deletion of Cdkn2a, which causes hyper-proliferation of MEFs and spontaneous escape from senescence, heterozygous deletion does not increase cell proliferation on its own²⁶. Hence, increased expression of ARF is likely part of the mechanism underlying proliferation arrest in Zfp148^gt/gt MEFs. We have previously shown that Zfp148^gt/gt MEFs are rescued by exogenous antioxidants or culture at reduced oxygen concentration showing that oxidative stress contributes to the proliferation arrest in these cells¹². Because ARF suppresses the transcriptional activity of NRF2 (NFE2L2)²⁷—the master regulator of endogenous antioxidants—increased ARF transcription may explain the oxidative stress weakness of Zfp148^gt/gt MEFs.

Zfp148 interacted physically with the ARF promoter, raising the possibility that ARF is a direct target of Zfp148. Thus, deletion of Zfp148 may activate p53 by derepressing ARF transcription, similar to knockouts of polycomb members Bmi1, M33, Mel18, and Phc2^28,29,30,31. In line with this, inhibition of ZNF148 potentiates butyrate-induced senescence of human HT116 colon cancer cells by derepressing the p16 promoter³². However, increased expression of ARF could have many underpinnings. The strong enrichment of transcription factors among Zfp148-binding genes suggests that Zfp148 operates at a high level of the transcription factor hierarchy³³. Small orchestrated alterations of many transcripts may therefore contribute to the activation of ARF and p53 in Zfp148^gt/gt MEFs, possibly by affecting cellular homeostasis. Collectively, our data suggest that Zfp148 deficiency activates p53 indirectly, by derepressing ARF and by altering the activity of other transcription factors.

A previous publication showed that ZNF148 interacts physically with p53¹⁶, thus identifying another possible mechanism behind the activation of p53 in Zfp148^gt/gt MEFs. However, our current data on ZNF148 and TP53 dependencies in human cancer cells strongly argue against a physical interaction between ZNF148 and p53 playing a significant role. There was no evidence of genetic interaction between ZNF148 and TP53 in CRISPR and siRNA screens of hundreds of human cancer cell lines, performed under conditions that revealed strong dependencies between TP53 and known repressors of p53 activity including MDM2, MDM4, and PPM1D. Moreover, the was no evidence of genetic interaction between ZNF148 and TP53 under DNA damaging conditions in human lung cancer cells or mice. However, it remains possible that regulation of ARF is important. For reasons not well understood, ARF appears to play a more prominent role as a tumor suppressor in mice than in humans³⁴. Moreover, accumulating evidence indicate that ARF has important roles that are independent of p53³⁴. In line with this, we observed that the correlation between CDKN2A and TP53 was weak in the CRISPR screen and absent in the siRNA screen (data not shown).

In the light of these results, we need to reevaluate the role of Zfp148 and its impact on the p53 pathway. Without evidence of genetic interaction between ZNF148 and TP53 in human cancer cells, the incentive for therapeutic targeting of ZNF148 becomes weak. The recent finding that inhibition of ZNF148 potentiates butyrate-induced senescence of human HT116 colon cancer cells³² suggests that core functions of Zfp148 are evolutionarily conserved. Further investigation of the interaction between Zfp148 and Cdkn2a promoters may provide valuable mechanistic insights. However, transcriptomes are shaped by combinatorial interactions of transcription factors forming complex network motifs, including feedforward and feedbackward loops, that cooperatively govern gene expression³⁵. Our current data suggest that Zfp148 operates at a high level of the transcription factor hierarchy. Thus, future studies of Zfp148 warrant a more integrated and comprehensive approach.

Material and methods

Mice

Cdkn2a^tm1rdb (Cdkn2a^−/) and Trp53^tm1Tyj (Trp53^−/) mice were obtained from The Jackson Laboratory and Zfp148^gt/+ mice were produced by us¹². The mice were kept on a 129/Bl6 mixed genetic background and all experiments were performed with littermate controls. Genotyping was performed by PCR amplification of genomic DNA from mouse biopsies obtained during earmarking. PCR primers used for genotyping are listed in Supplementary Table S1. Mice were fed on a regular diet and had unlimited supply of food and water. All animal procedures used in this study were approved by The Animal Research Ethics Committee in Gothenburg and performed in accordance with relevant guidelines and regulations.

Cell culture

Primary MEFs were isolated from E13.5–15.5 embryos as described¹². Cells were cultured in DMEM low glucose medium with 10% fetal bovine serum, 100 µg/ml penicillin and streptomycin, 1% non-essential amino acids, 1–4 mM glutamine, and 0.5 mM β-mercaptoethanol. Growth curves were established using a modified 3T3 protocol (50,000 cells/well in 6-well plates).

Adenoviral transduction

Cells were incubated with 15 multiplicities of infection of empty control adenoviruses (AdNull) or adenoviruses encoding Zfp148 (AdZfp148) or Zfp148FLAG (AdZfp148FLAG) (Vector Biolabs) for 36–48 h before analyses, as described in¹².

Gene expression analysis

Total RNA from MEFs, lungs, brains and macrophages was isolated using the Gene Elute kit (Sigma) and hybridized to Affymetrix Mouse Gene 1.0 ST chips. Data were normalized with the Robust Multi-Array Analysis (RMA) method³⁶. The heat-map was constructed using the Hierarchical Clustering Explorer 3.0 software. Gene ontology statistics was calculated using the DAVID software^37,38.

Real-time quantitative PCR

TaqMan and SYBR Green assays were performed as described³⁹ using TaqMan/SYBR Green universal PCR mastermix (A25741, Applied-biosystem) and the pre-designed TaqMan assays (Applied Biosystems) or RT-PCR primers listed in Supplementary Table S1.

CRISPR-Cas9 knockouts

Pre-designed guide-RNA (gRNA) sequences targeting p16 or ARF or ZNF148 or control (does not target a sequence in the genome) were cloned into the Lenticrisprv2 vector, and co-transfected with pCMV-dR8.2 (Addgene # 8455) and pCMV-VSV-G (Addgene # 8454) vectors into HEK293T cells to produce complete Cas9 and gRNA coding lentivirus. Virus infected Zfp148^gt/gt MEFs or human lung cancer cells (H838, A549, or H1299) were selected with puromycin for 3 days to obtain batch clones that were used for experiments. The gRNA sequences used are listed in Supplementary Table S1. Lenticrisprv2 was a kind gift from Feng Zhang (Addgene # 52961)⁴⁰.

Western blot analyses

Protein levels was determined as previously described⁴¹ with antibodies against p16ink4a (sc-1207, santa cruz), p19^arf (sc-3278, santa cruz), ZNF148 (HPA001656, SigmaAldrich Atlas), p53 (DO-1 (sc126), Santa Cruz), and actin (A2066, Sigma Aldrich Atlas). Secondary antibodies were anti-mouse IRDye 680RD (926-68072, LI-COR Biosciences), anti-rabbit 680RD (926-68071, LI-COR Biosciences), anti-rabbit IgG Conformation Specific (L27A9) HRP Conjugate (5,127, Cell Signaling), and anti-mouse IgG-hl-HRP (ab6728, abcam). Protein bands were detected with the immubilon western chemiluminescent HRP substrate (Millipore) using a Chemi Doc Touch Imaging system (Bio Rad) or the Li-Cor Odyssey Imager. Exposure times were adjusted to the signal intensity of each blot. Images were adjusted for brightness and contrast with the Photoshop tool levels. All changes were applied equally across the entire image.

ChIP-on-chip analysis

MEFs (1 × 10⁸) transduced with AdZfp148FLAG were fixed in 1% formaldehyde. 200–500-bp DNA fragments were precipitated using the anti-FLAG M2 antibody (Sigma) conjugated to sheep anti-mouse IgG Dynabeads (Invitrogen), purified (QIAquick, Qiagen), amplified (WGA2 kit, Sigma), and hybridized to MM8_Deluxe_Promoter_HX1 chips at NimblGen (www.nimblegen.com). The analysis was done on Zfp148^gt/gt background to avoid competition with endogenous Zfp148 and Trp53^−/− background to overcome senescence and obtain sufficient amounts of starting material. Gene ontology statistics was calculated using the DAVID software with “functional annotation clustering” setting for DAVID enrichment scores and “functional annotation chart” setting for ranking of functional terms^37,38.

Motif prediction

A position weight matrix (PWM) model based on 16 known binding sites of ZNF148 (Table 1)^{4,5,6,8,42,43,44,45,46,47,48,49,50} was applied on regions around known transcription starts (TS), following the approach described in Wasserman and Sandelin (2004). The PWM model slides across the region one nucleotide at a time, generating a score between 0 and 1 at each position for both strands. The best score (of the two strands) for each position was then averaged over a 200 bp window. A region was called as a potential binding site (PWM-peak positive) when the average score of the window exceeded 0.6.

Meta-analysis of CRISPR and siRNA data

CRISPR-scores (Achilles_gene_effect), siRNA-scores (D2_combined_gene_dep_scores), and somatic mutation calls (CCLE_mutations) for human cancer cells (cancer cell line encyclopaedia) were downloaded from the Cancer Dependency Map portal at the Broad Institute (https://depmap.org/portal/download/).

Drug response curves

15,000 cells were plated in 12-well plates (per well), incubated with cisplatin (Sigma), etoposide (Sigma), or TH588 (TH588 hydrochloride, Axon Medchem) for three days, and stained for 20 min with a fixing/staining solution containing 0.025% crystal violet (Sigma), 1% formaldehyde, and 1% methanol in PBS. After drying overnight, the crystal violet stained cells were solubilized in 10% acetic acid and the absorbance determined at 590 nm with a Spectramax II (Molecular Devices).

Radiosensitivity

Zfp148^gt/+ mice and wild-type controls were radiated using an RS-2000 X-ray irradiator (Rad Source Technologies) at the indicated doses. The thymus was collected after sacrifice on day 4 after radiation. Thymocytes were disseminated into single cell suspension and stained with CD45-BUV737 (clone 104, cat. no. 612778, BD Biosciences), CD4-BV785 (clone RM4-5, cat. no. 1102760, Sony Biotechnology), CD8-FITC (clone 53-6.7, cat. no. 1103525, Sony Biotechnology) and propidium iodide (cat. no. P3566, Life Technologies). Samples were run on a Fortessa X20 (BD Biosciences) and data was analyzed using FlowJo version 10.6.1.

Statistical analyses

Values are mean ± SEM. Statistics were performed with Student’s t-test for comparisons between two groups; one-way ANOVA for multiple groups; Fisher exact test for contingency tables; two-sample Kolgomorov-Smirnov test for cumulative distribution function plots; Spearman correlation coefficient for ranked values; and Pearson correlation coefficient for continuous variables. Differences between groups were considered significant when P < 0.05.

References

Wade, M., Li, Y. C. & Wahl, G. M. MDM2, MDMX and p53 in oncogenesis and cancer therapy. Nat. Rev. Cancer 13, 83–96. https://doi.org/10.1038/nrc3430 (2013).
Article CAS PubMed PubMed Central Google Scholar
Tisato, V., Voltan, R., Gonelli, A., Secchiero, P. & Zauli, G. MDM2/X inhibitors under clinical evaluation: perspectives for the management of hematological malignancies and pediatric cancer. J. Hematol. Oncol. 10, 133. https://doi.org/10.1186/s13045-017-0500-5 (2017).
Article CAS PubMed PubMed Central Google Scholar
Ringshausen, I., O’Shea, C. C., Finch, A. J., Swigart, L. B. & Evan, G. I. Mdm2 is critically and continuously required to suppress lethal p53 activity in vivo. Cancer Cell 10, 501–514. https://doi.org/10.1016/j.ccr.2006.10.010 (2006).
Article CAS PubMed Google Scholar
Merchant, J. L. et al. ZBP-89, a Kruppel-like zinc finger protein, inhibits epidermal growth factor induction of the gastrin promoter. Mol. Cell. Biol. 16, 6644–6653. https://doi.org/10.1128/mcb.16.12.6644 (1996).
Article CAS PubMed PubMed Central Google Scholar
Wang, Y., Kobori, J. A. & Hood, L. The ht beta gene encodes a novel CACCC box-binding protein that regulates T-cell receptor gene expression. Mol. Cell. Biol. 13, 5691–5701 (1993).
Article CAS PubMed PubMed Central Google Scholar
Hasegawa, T., Takeuchi, A., Miyaishi, O., Isobe, K. & de Crombrugghe, B. Cloning and characterization of a transcription factor that binds to the proximal promoters of the two mouse type I collagen genes. J. Biol. Chem. 272, 4915–4923 (1997).
Article CAS PubMed Google Scholar
Passantino, R. et al. Negative regulation of beta enolase gene transcription in embryonic muscle is dependent upon a zinc finger factor that binds to the G-rich box within the muscle-specific enhancer. J. Biol. Chem. 273, 484–494 (1998).
Article CAS PubMed Google Scholar
Bai, L. & Merchant, J. L. Transcription factor ZBP-89 cooperates with histone acetyltransferase p300 during butyrate activation of p21waf1 transcription in human cells. J. Biol. Chem. 275, 30725–30733 (2000).
Article CAS PubMed Google Scholar
Bai, L., Kao, J. Y., Law, D. J. & Merchant, J. L. Recruitment of ataxia-telangiectasia mutated to the p21(waf1) promoter by ZBP-89 plays a role in mucosal protection. Gastroenterology 131, 841–852 (2006).
Article CAS PubMed Google Scholar
Wu, Y., Zhang, X., Salmon, M. & Zehner, Z. E. The zinc finger repressor, ZBP-89, recruits histone deacetylase 1 to repress vimentin gene expression. Genes Cells 12, 905–918. https://doi.org/10.1111/j.1365-2443.2007.01104.x (2007).
Article CAS PubMed Google Scholar
Feng, Y. et al. The transcription factor ZBP-89 suppresses p16 expression through a histone modification mechanism to affect cell senescence. FEBS J. 276, 4197–4206. https://doi.org/10.1111/j.1742-4658.2009.07128.x (2009).
Article CAS PubMed Google Scholar
Sayin, V. I. et al. Zfp148 deficiency causes lung maturation defects and lethality in newborn mice that are rescued by deletion of p53 or antioxidant treatment. PLoS ONE 8, e55720. https://doi.org/10.1371/journal.pone.0055720 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Nilton, A. et al. Targeting Zfp148 activates p53 and reduces tumor initiation in the gut. Oncotarget https://doi.org/10.18632/oncotarget.10899 (2016).
Article PubMed PubMed Central Google Scholar
Essien, B. E. et al. Transcription factor ZBP-89 drives a feedforward loop of beta-catenin expression in colorectal cancer. Cancer Res. https://doi.org/10.1158/0008-5472.CAN-15-3150 (2016).
Article PubMed PubMed Central Google Scholar
Sayin, V. I. et al. Loss of one copy of Zfp148 reduces lesional macrophage proliferation and atherosclerosis in mice by activating p53. Circ. Res. https://doi.org/10.1161/CIRCRESAHA.115.304992 (2014).
Article PubMed Google Scholar
Bai, L. & Merchant, J. L. ZBP-89 promotes growth arrest through stabilization of p53. Mol. Cell. Biol. 21, 4670–4683 (2001).
Article CAS PubMed PubMed Central Google Scholar
Kong, L. J., Chang, J. T., Bild, A. H. & Nevins, J. R. Compensation and specificity of function within the E2F family. Oncogene 26, 321–327. https://doi.org/10.1038/sj.onc.1209817 (2007).
Article CAS PubMed Google Scholar
Chong, J. L. et al. E2f1-3 switch from activators in progenitor cells to repressors in differentiating cells. Nature 462, 930–934. https://doi.org/10.1038/nature08677 (2009).
Article ADS CAS PubMed PubMed Central Google Scholar
Xu, X. et al. A comprehensive ChIP-chip analysis of E2F1, E2F4, and E2F6 in normal and tumor cells reveals interchangeable roles of E2F family members. Genome Res. 17, 1550–1561. https://doi.org/10.1101/gr.6783507 (2007).
Article CAS PubMed PubMed Central Google Scholar
Sayin, V. I. et al. Loss of one copy of Zfp148 reduces lesional macrophage proliferation and atherosclerosis in mice by activating p53. Circ. Res. 115, 781–789. https://doi.org/10.1161/CIRCRESAHA.115.304992 (2014).
Article CAS PubMed Google Scholar
Nilton, A. et al. Zinc finger protein 148 is dispensable for primitive and definitive hematopoiesis in mice. PLoS ONE 8, e70022. https://doi.org/10.1371/journal.pone.0070022 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Kobayashi, T., Wang, J., Al-Ahmadie, H. & Abate-Shen, C. ARF regulates the stability of p16 protein via REGgamma-dependent proteasome degradation. Mol. Cancer Res. 11, 828–833. https://doi.org/10.1158/1541-7786.MCR-13-0207 (2013).
Article CAS PubMed PubMed Central Google Scholar
Meyers, R. M. et al. Computational correction of copy number effect improves specificity of CRISPR-Cas9 essentiality screens in cancer cells. Nat. Genet. 49, 1779–1784. https://doi.org/10.1038/ng.3984 (2017).
Article CAS PubMed PubMed Central Google Scholar
Gul, N. et al. The MTH1 inhibitor TH588 is a microtubule-modulating agent that eliminates cancer cells by activating the mitotic surveillance pathway. Sci. Rep. 9, 14667. https://doi.org/10.1038/s41598-019-51205-w (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Lowe, S. W., Schmitt, E. M., Smith, S. W., Osborne, B. A. & Jacks, T. p53 is required for radiation-induced apoptosis in mouse thymocytes. Nature 362, 847–849. https://doi.org/10.1038/362847a0 (1993).
Article ADS CAS PubMed Google Scholar
Serrano, M. et al. Role of the INK4a locus in tumor suppression and cell mortality. Cell 85, 27–37 (1996).
Article CAS PubMed Google Scholar
Chen, D. et al. NRF2 is a major target of ARF in p53-independent tumor suppression. Mol. Cell. 68, 224–232. https://doi.org/10.1016/j.molcel.2017.09.009 (2017).
Article CAS PubMed PubMed Central Google Scholar
Jacobs, J. J., Kieboom, K., Marino, S., DePinho, R. A. & van Lohuizen, M. The oncogene and Polycomb-group gene bmi-1 regulates cell proliferation and senescence through the ink4a locus. Nature 397, 164–168. https://doi.org/10.1038/16476 (1999).
Article ADS CAS PubMed Google Scholar
Core, N., Joly, F., Boned, A. & Djabali, M. Disruption of E2F signaling suppresses the INK4a-induced proliferative defect in M33-deficient mice. Oncogene 23, 7660–7668. https://doi.org/10.1038/sj.onc.1207998 (2004).
Article CAS PubMed Google Scholar
Isono, K. et al. Mammalian polyhomeotic homologues Phc2 and Phc1 act in synergy to mediate polycomb repression of Hox genes. Mol. Cell. Biol. 25, 6694–6706. https://doi.org/10.1128/MCB.25.15.6694-6706.2005 (2005).
Article CAS PubMed PubMed Central Google Scholar
Miki, J., Fujimura, Y., Koseki, H. & Kamijo, T. Polycomb complexes regulate cellular senescence by repression of ARF in cooperation with E2F3. Genes Cells 12, 1371–1382. https://doi.org/10.1111/j.1365-2443.2007.01135.x (2007).
Article CAS PubMed Google Scholar
Ocadiz-Ruiz, R. et al. ZBP-89 function in colonic stem cells and during butyrate-induced senescence. Oncotarget 8, 94330–94344. https://doi.org/10.18632/oncotarget.21698 (2017).
Article PubMed PubMed Central Google Scholar
Gerstein, M. B. et al. Architecture of the human regulatory network derived from ENCODE data. Nature 489, 91–100. https://doi.org/10.1038/nature11245 (2012).
Article ADS CAS PubMed PubMed Central Google Scholar
Fontana, R., Ranieri, M., La Mantia, G. & Vivo, M. Dual role of the alternative reading frame ARF protein in cancer. Biomolecules. https://doi.org/10.3390/biom9030087 (2019).
Article PubMed PubMed Central Google Scholar
Alon, U. Network motifs: theory and experimental approaches. Nat. Rev. Genet. 8, 450–461. https://doi.org/10.1038/nrg2102 (2007).
Article CAS PubMed Google Scholar
Irizarry, R. A. et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Biostatistics 4, 249–264. https://doi.org/10.1093/biostatistics/4.2.249 (2003).
Article PubMed MATH Google Scholar
da Huang, W., Sherman, B. T. & Lempicki, R. A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 4, 44–57. https://doi.org/10.1038/nprot.2008.211 (2009).
Article CAS Google Scholar
da Huang, W., Sherman, B. T. & Lempicki, R. A. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 37, 1–13. https://doi.org/10.1093/nar/gkn923 (2009).
Article CAS Google Scholar
Larsson, E. et al. Discovery of microvascular miRNAs using public gene expression data: miR-145 is expressed in pericytes and is a regulator of Fli1. Genome Med. 1, 108. https://doi.org/10.1186/gm108 (2009).
Article CAS PubMed PubMed Central Google Scholar
Sanjana, N. E., Shalem, O. & Zhang, F. Improved vectors and genome-wide libraries for CRISPR screening. Nat. Methods 11, 783–784. https://doi.org/10.1038/nmeth.3047 (2014).
Article CAS PubMed PubMed Central Google Scholar
Ibrahim, M. X. et al. Targeting isoprenylcysteine methylation ameliorates disease in a mouse model of progeria. Science 340, 1330–1333. https://doi.org/10.1126/science.1238880 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Reizis, B. & Leder, P. Expression of the mouse pre-T cell receptor alpha gene is controlled by an upstream region containing a transcriptional enhancer. J. Exp. Med. 189, 1669–1678. https://doi.org/10.1084/jem.189.10.1669 (1999).
Article CAS PubMed PubMed Central Google Scholar
Yamada, A. et al. Identification and characterization of a transcriptional regulator for the lck proximal promoter. J. Biol. Chem. 276, 18082–18089 (2001).
Article CAS PubMed Google Scholar
Ye, S., Whatling, C., Watkins, H. & Henney, A. Human stromelysin gene promoter activity is modulated by transcription factor ZBP-89. FEBS Lett. 450, 268–272 (1999).
Article CAS PubMed Google Scholar
De Bustos, C. et al. A PDGFRA promoter polymorphism, which disrupts the binding of ZNF148, is associated with primitive neuroectodermal tumours and ependymomas. J. Med. Genet. 42, 31–37. https://doi.org/10.1136/jmg.2004.024034 (2005).
Article CAS PubMed PubMed Central Google Scholar
Law, G. L. et al. Transcription factor ZBP-89 regulates the activity of the ornithine decarboxylase promoter. J. Biol. Chem. 273, 19955–19964. https://doi.org/10.1074/jbc.273.32.19955 (1998).
Article CAS PubMed Google Scholar
Bai, L. & Merchant, J. L. Transcription factor ZBP-89 is required for STAT1 constitutive expression. Nucleic Acids Res. 31, 7264–7270 (2003).
Article CAS PubMed PubMed Central Google Scholar
Keates, A. C. et al. ZBP-89, Sp1, and nuclear factor-kappa B regulate epithelial neutrophil-activating peptide-78 gene expression in Caco-2 human colonic epithelial cells. J. Biol. Chem. 276, 43713–43722. https://doi.org/10.1074/jbc.M107838200 (2001).
Article CAS PubMed Google Scholar
Feo, S. et al. Transcription of the human beta enolase gene (ENO-3) is regulated by an intronic muscle-specific enhancer that binds myocyte-specific enhancer factor 2 proteins and ubiquitous G-rich-box binding factors. Mol. Cell. Biol. 15, 5991–6002. https://doi.org/10.1128/mcb.15.11.5991 (1995).
Article CAS PubMed PubMed Central Google Scholar
Cheng, P. Y., Kagawa, N., Takahashi, Y. & Waterman, M. R. Three zinc finger nuclear proteins, Sp1, Sp3, and a ZBP-89 homologue, bind to the cyclic adenosine monophosphate-responsive sequence of the bovine adrenodoxin gene and regulate transcription. Biochemistry 39, 4347–4357. https://doi.org/10.1021/bi992298f (2000).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

We would like to warmly thank Rosie Perkins for constructive criticism on the manuscript. The project was supported by the Swedish Research Council, Polysackaridforskning AB, the Swedish Cancer Association, the Swedish Heart-Lung Foundation, The Swedish Child Cancer Foundation, the Sahlgrenska University Hospital ALF research grants, and the University of Gothenburg (to P. Lindahl) and Swedish Research Council, Swedish Society for Medicine, BioCARE and Knut and Alice Wallenberg foundation (to V.I. Sayin).

Funding

Open Access funding provided by University of Gothenburg.

Author information

Authors and Affiliations

Wallenberg Laboratory, Department of Molecular and Clinical Medicine, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden
Zhiyuan V. Zou, Nadia Gul, Viktor Garellick, Inger Johansson, Åsa Tivesten & Per Lindahl
Sahlgrenska Cancer Center, Department of Surgery, Institute of Clinical Sciences, University of Gothenburg, Gothenburg, Sweden
Nadia Gul, Ella M. Eklund, Angana A. H. Patel, Jozefina J. Dzanan, Ben O. Titmuss, Kristell Le Gal & Volkan I. Sayin
Wallenberg Centre for Molecular and Translational Medicine, University of Gothenburg, Gothenburg, Sweden
Nadia Gul, Ella M. Eklund, Angana A. H. Patel, Jozefina J. Dzanan, Ben O. Titmuss, Kristell Le Gal & Volkan I. Sayin
Department of Biochemistry, Institute of Biomedicine, University of Gothenburg, Gothenburg, Sweden
Markus Lindberg, Abdulmalik A. Bokhari, Erik Larsson & Per Lindahl
Sahlgrenska Cancer Center, Department of Radiation Physics, Institute of Clinical Sciences, University of Gothenburg, Gothenburg, Sweden
Eva Forssell-Aronsson
Department of Medical Physics and Biomedical Engineering, Sahlgrenska University Hospital, University of Gothenburg, Gothenburg, Sweden
Eva Forssell-Aronsson
Sahlgrenska Cancer Center, Department of Molecular and Clinical Medicine, Institute of Medicine, University of Gothenburg, Gothenburg, Sweden
Martin O. Bergö
Department of Biosciences and Nutrition, Karolinska Institute, Huddinge, Sweden
Martin O. Bergö
Department of Microbiologi and Immunology, Institute of Biomedicine, University of Gothenburg, Gothenburg, Sweden
Anna Staffas

Authors

Zhiyuan V. Zou
View author publications
You can also search for this author in PubMed Google Scholar
Nadia Gul
View author publications
You can also search for this author in PubMed Google Scholar
Markus Lindberg
View author publications
You can also search for this author in PubMed Google Scholar
Abdulmalik A. Bokhari
View author publications
You can also search for this author in PubMed Google Scholar
Ella M. Eklund
View author publications
You can also search for this author in PubMed Google Scholar
Viktor Garellick
View author publications
You can also search for this author in PubMed Google Scholar
Angana A. H. Patel
View author publications
You can also search for this author in PubMed Google Scholar
Jozefina J. Dzanan
View author publications
You can also search for this author in PubMed Google Scholar
Ben O. Titmuss
View author publications
You can also search for this author in PubMed Google Scholar
Kristell Le Gal
View author publications
You can also search for this author in PubMed Google Scholar
Inger Johansson
View author publications
You can also search for this author in PubMed Google Scholar
Åsa Tivesten
View author publications
You can also search for this author in PubMed Google Scholar
Eva Forssell-Aronsson
View author publications
You can also search for this author in PubMed Google Scholar
Martin O. Bergö
View author publications
You can also search for this author in PubMed Google Scholar
Anna Staffas
View author publications
You can also search for this author in PubMed Google Scholar
Erik Larsson
View author publications
You can also search for this author in PubMed Google Scholar
Volkan I. Sayin
View author publications
You can also search for this author in PubMed Google Scholar
Per Lindahl
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

Z.V.Z., V.I.S. and P.L. conceived the study. Z.V.Z., N.G., A.A.B., V.G., E.M.E., A.A.H.P., J.J.D., B.O.T., K.L.G., I.J., A.S., V.I.S. and P.L. performed experiments and analyzed data. M.L, and E.L. performed the transcription factor motif analyses. Å.T., E.F.A., and M.O.B. provided conceptual and technical advice. P.L. and V.I.S. wrote the manuscript. All authors reviewed the text.

Corresponding authors

Correspondence to Volkan I. Sayin or Per Lindahl.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information 1.

Supplementary information 2.

Supplementary information 3.

Supplementary information 4.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Cite this article

Zou, Z.V., Gul, N., Lindberg, M. et al. Genomic profiling of the transcription factor Zfp148 and its impact on the p53 pathway. Sci Rep 10, 14156 (2020). https://doi.org/10.1038/s41598-020-70824-2

Download citation

Received: 07 November 2019
Accepted: 04 August 2020
Published: 25 August 2020
DOI: https://doi.org/10.1038/s41598-020-70824-2

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.