MYC and AP-1 oncogenes cooperatively bind enhancers to rewire transcription

Reshma Kalyan Sundaram; Kshitiz Parihar; Stephanie Monson; Ravi Radhakrishnan; Bomyi Lim

doi:10.1093/narcan/zcag010

. 2026 Apr 8;8(2):zcag010. doi: 10.1093/narcan/zcag010

MYC and AP-1 oncogenes cooperatively bind enhancers to rewire transcription

Reshma Kalyan Sundaram ¹, Kshitiz Parihar ², Stephanie Monson ³, Ravi Radhakrishnan ^4,^5,^✉, Bomyi Lim ^6,^✉

PMCID: PMC13069683 PMID: 41970279

Abstract

The transcription factor c-MYC (MYC) is deregulated in ~70% of human cancers. Through de novo motif discovery analysis on published MYC ChIP-seq datasets from cancer cell lines, we found cell-type-specific co-enrichment of the TRE motifs (AP-1 binding sites) alongside MYC’s canonical EBOX motif. MYC binds indirectly to TRE motifs in cooperation with AP-1 transcription factors, and these indirect interactions occur predominantly at enhancers rather than promoters. At elevated MYC levels, as seen in cancers, MYC’s indirect binding to TRE sites at enhancers increases. Integration of ChIP-seq and RNA sequencing data revealed that TRE enhancer-binding sites are frequently associated with MYC-mediated transcriptional repression. Gene Ontology analysis showed that MYC utilizes TRE sites to transcriptionally rewire cells, modulating cancer hallmarks like proliferation, apoptosis, and cell adhesion. These molecular insights into how increased MYC levels alter gene regulation could inform new therapeutic strategies targeting cancer-specific MYC functions and its co-regulators.

Graphical Abstract

For image description, please refer to the figure legend and surrounding text.

Introduction

The MYC family of proto-oncogenes comprises of c-MYC, N-MYC, and L-MYC [1–3]. These genes code for the transcription factors (TFs) c-MYC (MYC), N-MYC (MYCN), and L-MYC (MYCL), respectively [1, 2, 4]. Here, we focus on TF MYC, which was identified over 40 years ago [5–7]. MYC is expressed in various cell and tissue types and drives both general as well as cell-type-specific functions [1–3]. The MYC gene has high homology across vertebrates [8–11], with over 90% amino acid conservation between human and mouse MYC genes [9]. The genes regulated by MYC correspond to several important cellular processes, such as cell cycle, cell growth, metabolism, DNA replication, and apoptosis. For instance, MYC controls cell cycle by modulating the expression of cyclins and cell cycle inhibitors, such as Cyclin D, Cyclin E, p15, p21, and p27 [2, 4], and regulates energy metabolism by upregulating genes (such as LDHA, GLS, and TFAM) involved in glycolysis, glutamine metabolism, and mitochondrial biogenesis [3, 4]. MYC also regulates apoptosis through modulation of various pro-apoptotic and anti-apoptotic genes, including BAX, BIM, and BCL-2 [2, 4]. Additionally, MYC-regulated genes and pathways are known to indirectly regulate the global transcriptome through feedback mechanisms that modulate RNA synthesis [4]. Since MYC plays such a pivotal role as a master regulator of the transcriptome, MYC levels in cells are tightly controlled to ensure normal cellular function [3, 5].

MYC, however, is deregulated in up to 70% of human cancers (including Burkitt lymphoma, breast cancer, lung cancer, and hepatocellular carcinoma) [1–3] through mechanisms such as gene amplification, signal transduction, and protein stabilization [2]. These processes elevate MYC levels in cells, triggering the activation of new transcriptional programs by MYC and causing cellular transformation [4]. Such essential roles of MYC in cancer initiation and maintenance [2, 12] establish it as an important target of interest in cancer treatments [2, 13, 14]. Several approaches to target MYC have been explored, including silencing MYC gene expression, inhibiting MYC translation, promoting MYC protein degradation, and disrupting MYC protein–protein interactions [2, 13–16]. However, no MYC inhibitors are currently available for clinical use [15].

In cancers, MYC’s TF function is altered through changes in its binding patterns [1, 4, 17], regulated gene targets [1, 17–20], and interactions with co-regulator proteins [1, 14, 21]. Chromatin immunoprecipitation followed by sequencing (ChIP-seq) technology [22–25] enables genome-wide analysis of MYC’s DNA-binding patterns. Additionally, integrating ChIP-seq with RNA sequencing (RNA-seq) [20, 26] allows for the identification of MYC-regulated target genes. ChIP-seq can also be used to infer regulatory complex formation through MYC’s interactions with other proteins [27, 28], providing insights into its broader regulatory network.

Therefore, in this work, we analyzed publicly-available ChIP-seq datasets of various cancer cell lines to gain a comprehensive understanding of MYC’s oncogenic gene regulatory functions. Our analysis revealed co-enrichment of the TRE motif, a known AP-1 TF family binding site [29, 30], alongside the canonical EBOX motif [1, 4] in MYC ChIP-seq peaks. Through subsequent analysis of MYC and AP-1 ChIP-seq, we demonstrated that MYC occupies TRE sites in cooperation with AP-1. In this study, cooperativity refers to MYC indirectly binding to the TRE motif via interaction with AP-1 when AP-1 is bound to its consensus TRE site. Similar cooperative interactions between TFs have also been reported in previous studies [28, 31]. Additionally, we found that the TRE motif is an enhancer-specific MYC binding site, unlike the EBOX motif, which is both an enhancer and promoter MYC binding site. Analysis of MYC ChIP-seq datasets across different MYC expression levels [20, 32] confirmed that TRE remains an enhancer-specific, indirect MYC binding site, independent of MYC levels in cells. Gene Ontology (GO) on MYC target genes revealed that MYC utilizes TRE binding sites at enhancers to transcriptionally rewire cells by modulating several cancer hallmarks, such as proliferation, apoptosis, and cell adhesion [2, 33, 34]. Given the challenges in targeting MYC therapeutically [35, 36], our findings will potentially aid in development of new treatments targeting cancer-specific functions of MYC and its co-regulators.

Materials and methods

Motif discovery

The de novo motif discovery was performed on ChIP-seq datasets (in BED format) using the findMotifsGenome.pl program from the HOMER software suite (v 4.11.1) [37]. The algorithm employed by HOMER is based on identifying differentially enriched motifs in ChIP-seq peaks (“target” sequences) as compared to a set of randomly selected sequences from the genome (“background” sequences). For each de novo motif discovery analysis, we enabled the option for HOMER to generate the random background sequences, and also enabled the options for ensuring similar GC-content and avoiding an imbalance in 1-mers, 2-mers, and 3-mers between target and background sequences. The number of background sequences was set to be five times the total number of ChIP-seq peaks.

Reference genomes (hg19, hg38, or mm10) were specified according to the databases from which the ChIP-seq files were obtained. Enriched motifs were identified within a ±100 bp region around the center of ChIP-seq peaks. The motif length was set to 10 bp. For each de novo enriched motif, the HOMER tool provides the percentage of ChIP-seq peaks containing the motif (referred to as %target), the percentage of background sequences containing the motif (referred to as %background), and statistical significance (P-value) for enrichment of the motif in ChIP-seq peaks as compared to background sequences. Each motif discovery run was performed four times to ensure robustness of the results in lieu of a random set of sequences being used as “background.” The mean and standard deviation of enrichment values (%target) from the four runs are reported in the bar plot figures. The number of peaks containing an enriched motif (referred to as “occurrence”) was calculated by multiplying the mean enrichment value (in %) by the total number of ChIP-seq peaks. Statistical significance of the de novo enriched motif was calculated by the HOMER software using a binomial test. Only significantly enriched motifs (P-value < 1e-60) were used for analysis. The HOMER output files for all the de novo motif discovery runs containing %target, %background, and P-values are provided in Supplementary data S1.

ChIP-seq overlap

The mergePeaks.pl program in the HOMER software suite was used to identify overlapping peaks between two ChIP-seq datasets (in BED format). Peaks were considered overlapping if the distance between their centers was <200 bp. The genome size was set to the total length (in bp) of the reference genome of the ChIP-seq datasets. Statistical significance of overlap between ChIP-seq datasets was calculated by the mergePeaks program using a hypergeometric test.

ChIP-seq subtraction analysis

Requirement of MYC-AP-1 cooperation for MYC occupancy of TRE motif was demonstrated by comparing the change in occurrence of TRE (and EBOX) motifs in all MYC ChIP-seq peaks versus MYC ChIP-seq peaks that do not overlap with individual AP-1 TF ChIP-seq peaks. Overlapping peaks between MYC and various AP-1 family TFs were removed from MYC ChIP-seq datasets (in BED format) using the bedtools (v 2.31.0) [38] subtract utility with the -A option. De novo motif discovery was then performed on all MYC ChIP-seq peaks (or MYC ChIP-seq peaks that do not overlap with AP-1 ChIP-seq peaks) to determine the occurrence of enriched EBOX and TRE motifs, as described previously. The mean occurrences calculated from four replicates of motif discovery runs were used to generate Fig. 1E and F and Supplementary Fig. S1B and C. Within a cell line, motif occurrences in MYC ChIP-seq peaks that do not overlap with AP-1 ChIP-seq peaks were normalized to motif occurrences in all MYC ChIP-seq peaks, and the results were presented as a heatmap (Fig. 1E and F). The grayed-out boxes in the heatmaps (Fig. 1E and F) represent AP-1 TFs for which ChIP-seq data was not available for that particular cell line.

Multi-panel figure showing motif enrichment and overlap analyses of MYC ChIP-seq datasets across multiple cell lines, with subfigures labelled from A to F. A is an example of the output from the HOMER de novo motif enrichment analysis. B is a bar plot showing enrichment of EBOX and TRE motifs across multiple cell lines. C is an integrative genomic viewer track showing an example of a MYC ChIP-seq peak overlapping with a JUN ChIP-seq peak at a representative genomic locus. D is a venn diagram quantifying the overlap between MYC and JUN ChIP-seq peaks for the K562 cell line. E and F are heatmaps comparing the fraction of EBOX and TRE motif occurrences, respectively, remaining in MYC ChIP-seq peaks after removal of regions overlapping with AP-1 ChIP-seq peaks. Rows and columns in the heatmaps represent different cell lines and AP-1 transcription factors respectively. — Uncovering MYC and AP-1 cooperativity in DNA-binding in various cancer cell lines. (A) Enrichment of EBOX and TRE motifs in MYC ChIP-seq dataset of a representative cell line, K562, using the HOMER *de novo* motif enrichment analysis. P-value for motif enrichment was calculated using a binomial test. (B) Quantification of EBOX and TRE motif enrichment in MYC ChIP-seq datasets for various cell lines analyzed. Data represent mean and standard deviation of four replicates of motif discovery runs. (C) An example of overlapping MYC and an AP-1 family member JUN ChIP-seq peaks near WEE1 gene for a representative cell line, K562. (D) Quantification of the number of overlapping MYC and JUN ChIP-seq peaks, shown using Venn diagram, for a representative cell line, K562. The statistical significance (P-value) for the overlap of two sets was calculated using a hypergeometric test. Proportion of MYC ChIP-seq peaks containing (E) EBOX and (F) TRE motifs after the removal of overlapping peaks with AP-1 family TFs in various cell lines. (E, F) JUN, FOS, JUND, and JUNB are AP-1 family TFs included in the analysis. For each cell line (row), data is shown only for AP-1 TFs (column) for which ChIP-seq data was available. “MYC - JUN” column refers to removal of overlapping JUN peaks from MYC ChIP-seq data followed by *de novo* motif discovery to quantify the occurrence of EBOX and TRE motifs, and these occurrence values are then divided by EBOX and TRE motif occurrences in all MYC ChIP-seq peaks. Similar analysis was done for FOS, JUND, and JUNB.

IGV plotting

IGV tracks were generated by loading ChIP-seq files (in bigWig format) into the IGV genome browser (v 2.2.7) [39, 40]. Reference genomes (hg19, hg38, or mm10) were specified according to the databases from which the ChIP-seq files were obtained. Binding profiles of MYC, H3K4me3, or H3K27ac were visualized by zooming into specific genomic regions.

Promoter enhancer separation

Promoter identification [41]: Transcribed genes were defined as those with an H3K4me3 ChIP-seq peak within ±5 kb of the transcription start site (TSS). MYC-bound promoters were identified as MYC ChIP-seq peaks within ±1 kb of the TSS of transcribed genes. Enhancer identification [41]: Active enhancer regions were defined as H3K27ac ChIP-seq peaks located outside promoter regions (>5 kb away from the TSS). MYC-bound enhancers were identified as MYC ChIP-seq peaks within ±1 kb of the center of active enhancer regions.

For the identification of promoter and enhancer regions of MYC binding, the following tools were utilized. The annotatePeaks.pl program in the HOMER software suite was used to obtain the nearest gene corresponding to a ChIP-seq peak and the distance between the ChIP-seq peak and the TSS of the nearest gene. Information regarding the TSS of various genes was obtained from the genome.tss file included in the HOMER software suite. The bedtools window utility was used to select ChIP-seq peaks from one dataset that are within a specific window (in bp) of another ChIP-seq dataset. The bedtools subtract utility with -A option was used to remove overlapping peaks between two ChIP-seq datasets.

ChIP-seq signal heatmaps

Promoter and enhancer: Heatmaps for promoters [41] in Fig. 2C and Supplementary Fig. S3A were generated by plotting the MYC ChIP-seq signal intensity over MYC-bound transcriptionally active promoters, with each row representing a ±5 kb region centered on the TSS. The rows were arranged in descending order of MYC ChIP-seq signal intensity, and color bars indicate the intensity of the MYC ChIP-seq signal. Heatmaps for enhancers [41] in Fig. 2C and Supplementary Fig. S3B were generated by plotting the H3K27ac or MYC ChIP-seq signal intensity over MYC-bound active enhancer regions, with each row representing a ±5 kb region centered on the active enhancers. The rows were arranged in descending order of H3K27ac ChIP-seq signal intensity, and color bars indicate the intensity of the H3K27ac or MYC ChIP-seq signal. The computeMatrix tool from deepTools suite (v 3.5.3) [42] was used in reference-point mode to generate intermediate matrix files corresponding to each heatmap. These matrix files were then visualized as heatmaps using the plotHeatmap tool from deepTools suite.

Multi-panel figure showing MYC binding pattern at promoters and enhancers across multiple cell lines, with subfigures labelled from A to I. A and B are integrative genomic viewer tracks showing representative examples of MYC ChIP-seq peaks overlapping with H3K4me3 ChIP-seq peaks at promoter regions and H3K27ac ChIP-seq peaks at enhancer regions. C shows heatmaps of MYC ChIP-seq signal intensities centered around transcription start sites of active genes (promoters) and around active enhancer regions for the K562 cell line. D is a bar plot showing the distribution of MYC binding across promoters, enhancers, and other genomic regions across multiple cell lines. E and F are bar plots showing enrichment of EBOX and TRE motifs, respectively, at promoter and enhancer regions for multiple cell lines. G, H, and I are density plots showing enrichment of CACGTG (EBOX) and TGA(G/C)TCA (TRE) sequences genome-wide, at promoter regions, and at enhancer regions, respectively, for the K562 cell line. — Identifying TRE as an enhancer-specific MYC binding site in various cancer cell lines. (A) An example of overlapping MYC and H3K4me3 DNA-binding regions identified from ChIP-seq near KCNK5 gene for K562 cell line. Boxed region with MYC and H3K4me3 peaks shows the identified MYC promoter binding site. (B) An example of overlapping MYC and H3K27ac DNA-binding regions identified from ChIP-seq near MFSD1 gene for K562 cell line. Boxed region on left with MYC and H3K27ac peaks shows the identified MYC enhancer binding site. Boxed region on right with MYC and H3K4me3 peaks shows the identified MYC promoter binding site. (C) Heatmaps showing ChIP-seq signal intensities around promoter and enhancer regions of MYC for a representative cell line, K562. (Left) For promoters, MYC ChIP-seq signal intensities are plotted around the center of TSS of actively transcribed genes bound by MYC. Rows are sorted in decreasing order of MYC ChIP-seq signal intensities. (Right) For enhancers, MYC and H3K27ac ChIP-seq signal intensities are plotted around the center of active enhancers bound by MYC. Rows are sorted in decreasing order of H3K27ac ChIP-seq signal intensities. All heatmaps cover ±5 kb window around the reference chosen for centering. Color bar indicates ChIP-seq signal intensities for corresponding heatmaps. (D) Distribution of MYC binding to promoters, enhancers, and other genomic regions identified using MYC, H3K4me3, and H3K27ac ChIP-seq datasets of various cell lines. Enrichment of EBOX and TRE motifs in (E) promoter and (F) enhancer binding regions of MYC in various cell lines identified using the HOMER *de novo* motif enrichment analysis. Data represent mean and standard deviation of four replicates of motif discovery runs. (G–I) Density plot of EBOX and TRE motifs around the center of MYC binding regions identified from ChIP-seq experiment for K562 cell line. Dashed lines indicate ±100 bp window around the center of MYC ChIP-seq peak used for performing *de novo* motif discovery analysis. Plots correspond to (G) genome-wide, (H) promoter, and (I) enhancer regions of MYC binding.

MYC-AP-1 overlap and non-overlap regions: For cell lines K562 and MCF7, heatmaps were generated for the ChIP-seq signal intensity (of MYC and AP-1 TFs) at MYC ChIP-seq peaks overlapping with AP-1 ChIP-seq peaks, as well as at MYC ChIP-seq peaks non-overlapping with AP-1 ChIP-seq peaks (Supplementary Fig. S2). First, MYC ChIP-seq peaks either overlapping or non-overlapping with AP-1 were selected. Then, the selected MYC ChIP-seq were sorted in the decreasing order of MYC ChIP-seq signal intensity. MYC ChIP-seq signal and AP-1 ChIP-seq signal intensities were then plotted on the sorted list of peaks from previous step to generate their respective heatmaps. Within each heatmap, each row represents a ±0.5 kb region centered on a MYC peak. The color bars indicate the intensity of the MYC or AP-1 ChIP-seq signal. The computeMatrix tool from deepTools suite was used in reference-point mode to generate intermediate matrix files corresponding to each heatmap. These matrix files were then visualized as heatmaps using the plotHeatmap tool from deepTools suite.

Motif density plots

The annotatePeaks.pl program in the HOMER software suite was used to generate density plots of EBOX (CACGTG) and TRE (TGA(G/C)TCA) sequences in the MYC ChIP-seq dataset of K562 cell line. CACGTG and TGA(G/C)TCA sequences were scanned within ±300 bp around the center of the ChIP-seq peaks. The bin size of the generated density plot was set to 10 bp.

Relative occupancy

Relative occupancy (RO) was defined as the ratio of TRE motif enrichment to EBOX motif enrichment in the analyzed ChIP-seq dataset. RO was calculated for four replicates of motif discovery runs. The statistical significance of change in RO between promoter and enhancer for low-MYC and high-MYC ChIP-seq datasets was calculated using two-sided Welch’s unequal variances t-test.

Gene Ontology

GO analysis [43, 44] was performed on MYC-regulated genes to identify the biological significance of MYC binding to TRE sites at enhancers. MYC-regulated genes were identified using ChIP-seq and RNA-seq datasets from the U2OS cell line [20, 32]. In this analysis, genes were classified as MYC “target genes” only if they contained a MYC ChIP-seq peak at their promoter. Here, the gene sets and biological processes corresponding to the novel MYC-enhancer binding site, TRE, are studied in comparison to the known MYC binding site, EBOX, at enhancers. Two types of GO analysis were conducted in this study. For each analysis type, MYC target genes were defined as:

Analysis 1: Genes bound by MYC (Fig. 4A left), identified using low- and high-MYC ChIP-seq datasets.
Analysis 2: Genes bound by MYC and showing significant differential expression (Fig. 4A right), identified using high-MYC ChIP-seq and ± Dox RNA-seq datasets.

Multi-panel figure showing analysis of biological processes associated with MYC target genes classified based on enhancer binding patterns, with subfigures labelled from A to F. A is a schematic illustrating the classification of MYC target genes into subsets based on MYC binding at promoters and the presence of CACGTG (EBOX), TGA(G/C)TCA (TRE), or both sequences at enhancer regions. B is a volcano plot showing differentially expressed genes identified from RNA-seq data of the U2OS cell line following MYC induction. C and D are venn diagrams showing the number of upregulated and downregulated MYC target genes, respectively, in different subsets. E and F show the Gene Ontology analysis results in tabular form for the upregulated and downregulated gene subsets, respectively, associated with different enhancer element categories. — Identifying distinct biological processes regulated by indirect MYC occupancy of TRE sites at enhancers by performing GO. (A) Schematic depicting classification of MYC target genes into subsets based on the enhancer binding patterns of MYC in ChIP-seq of the U2OS cell line. The outermost gray circle in the Venn diagram represents MYC target genes set, i.e. genes containing MYC ChIP-seq peak at their promoters. Black circle represents a subset of MYC target genes that contain an additional MYC ChIP-seq peak at their enhancers. Green (or yellow) circle represents MYC target gene subsets, specifically containing a CACGTG (or TGA(G/C)TCA) sequence within their enhancer MYC ChIP-seq peak. In Analysis 1, MYC target genes were identified based on MYC binding to genes in low- and high-MYC ChIP-seq datasets of the U2OS cell line. In Analysis 2, strongly regulated MYC target genes were identified based on MYC binding to genes in high-MYC ChIP-seq, and differential expression of those genes in ±Dox RNA-seq of the U2OS cell line. Schematic created in BioRender. Kalyan Sundaram, R. (2025) https://BioRender.com/32blib1. (B–F) Results for Analysis 2. (B) Volcano plot of differentially expressed genes identified using RNA-seq in U2OS cell line. Genes that are significantly differentially expressed (|log₂(FC)|>0.5 and p-adj < 0.05) are highlighted in red. Fold change (FC) was calculated as the ratio of gene expression in +Dox condition over gene expression in −Dox condition. Number of genes bound by MYC in high-MYC condition, and genes significantly (C) up- or (D) downregulated upon Dox addition, identified using ChIP-seq and RNA-seq data of U2OS cell line. (E–F) GO analysis corresponding to gene lists identified from green and yellow subsets in panels (C) and (D), respectively.

Detailed description of steps used to generate the MYC target gene lists used in each analysis mentioned above is provided in supplementary methods. Previous studies [45–47] integrated gene expression datasets and protein–protein interaction networks to identify pathways regulated by differentially expressed genes. Following this methodology, the gene lists identified in Analysis 1 and 2 were expanded to include high-confidence, direct physical interaction protein partners, identified using the IntAct Molecular Interaction Database (v 1.0.4) [48], of proteins coded by the MYC target genes in each gene list. Functional enrichment analysis was then performed on the expanded gene lists from the previous step using the DAVID functional annotation tool (v 2023q4) [43, 44] to identify the biological processes mediated by these genes. Implementation details for IntAct and DAVID are provided in supplementary methods, along with the specific protocol followed for generating the GO tables shown in Fig. 4E and F, and Supplementary Fig. S5C and D. All the datasets generated at various stages for GO analyses are provided in Supplementary data S2.

Results

Genome-wide analysis reveals that MYC indirectly occupies TRE binding site in cooperation with AP-1

Genome-wide DNA binding patterns of MYC were identified through HOMER de novo motif discovery tool [37] using ENCODE MYC ChIP-seq datasets from 11 cell lines (Supplementary Table S1). The EBOX motif (CACGTG) emerged as the first-ranked significantly enriched motif in the MYC ChIP-seq datasets across all analyzed cell lines (Fig. 1A and B), consistent with its well-established role as a direct consensus binding site for MYC [1, 4]. Surprisingly, the TRE motif (TGA(G/C)TCA) also appeared among the top six ranked significantly enriched motifs in the MYC ChIP-seq datasets of 8 out of the 11 analyzed cell lines (Fig. 1A and B). The TRE motif is a well-known binding site for the AP-1 family of TFs [29, 30].

Since ChIP-seq captures genome-wide binding patterns of a TF of interest, including both direct and indirect binding regions [49–52], de novo motif discovery can reveal co-enriched motifs alongside the consensus motif, as observed in previous studies [28, 37, 50, 51]. Co-enriched motifs include DNA regions indirectly bound by the TF, typically due to the formation of regulatory complexes or cooperative interactions with other proteins [28, 37, 50, 51]. Given that the (i) TRE motif is a known consensus binding site for AP-1, (ii) TRE motif sequence significantly differs from MYC’s consensus binding motif (EBOX) sequence, and (iii) TRE enrichment was consistently lower than EBOX enrichment in the MYC ChIP-seq datasets across all analyzed cell lines (Fig. 1B), we hypothesized that TRE could represent an indirect binding site for MYC, occupied in cooperation with AP-1.

To test this hypothesis, we first examined whether MYC and AP-1 family TFs shared overlapping genomic binding regions. In K562 cells, ∼33% of MYC ChIP-seq peaks (9735 out of 30 158) overlapped with ChIP-seq peaks of the AP-1 family member JUN (Fig 1C and D). We subsequently analyzed the overlap of ChIP-seq datasets for MYC and four AP-1 family TFs (JUN, FOS, JUND, and JUNB) (Supplementary Table S1) across seven cell lines (Fig. 1D and Supplementary Fig. S1A). Remarkably, we found statistically significant overlap between ChIP-seq datasets for all combinations of MYC and AP-1 TFs across all analyzed cell lines (Fig. 1D and Supplementary Fig. S1A). Consistent with this result, we observed strong AP-1 ChIP-seq signal at the overlapping MYC binding regions (Supplementary Fig. S2). These findings confirmed that MYC and AP-1 indeed share overlapping genomic binding regions.

We then examined if the observed overlap between MYC and AP-1 TFs binding regions may explain the enrichment of the TRE motif in the MYC ChIP-seq datasets. We quantified the number of EBOX- and TRE-containing MYC ChIP-seq peaks (referred to as “occurrence”) that do not overlap with the AP-1 family member ChIP-seq peaks. Specifically, we first calculated the occurrence of EBOX and TRE motifs in the MYC ChIP-seq datasets for seven cell lines. Then, we recalculated the EBOX and TRE motif occurrence in the MYC ChIP-seq dataset of a cell line after removing peaks that overlapped with the ChIP-seq peaks for individual AP-1 TFs in the same cell line. This analysis was performed for four AP-1 family TFs (JUN, FOS, JUND, and JUNB) across the seven cell lines.

We found that the occurrence of TRE motif decreased drastically when overlapping regions with AP-1 TFs were individually removed from the MYC ChIP-seq datasets (Fig. 1F and Supplementary Fig. S1C). In most cell lines, this value even dropped to zero, depending on the specific AP-1 TFs available for analysis. These findings suggest that the enrichment of TRE motifs in MYC ChIP-seq datasets is predominantly due to cooperative interaction with AP-1 TFs, supporting the hypothesis that TRE motifs represent indirect binding sites for MYC mediated by interactions with AP-1.

The occurrence of the EBOX motif also decreased across all cell lines upon removal of the overlapping peaks with AP-1 TFs (Fig. 1E and Supplementary Fig. S1B). However, the magnitude of reduction in EBOX occurrence was lower than the reduction observed for TRE occurrence (Fig. 1E and F; Supplementary Fig. S1B and C). This observation suggests that a significant number of EBOX-containing MYC ChIP-seq peaks correspond to MYC regulation independent of AP-1. Therefore, AP-1 independent and AP-1 dependent regulation of transcriptional programs potentiated by MYC occur simultaneously. Altogether, these results suggest that MYC indirectly occupies TRE sites through cooperative interactions with AP-1 TFs, via one of two possible scenarios: MYC and AP-1 bind to neighboring sites at their respective consensus motifs (EBOX and TRE), or MYC interacts with AP-1, which in turn is directly bound to DNA at TRE sites.

Regulatory element analysis reveals that MYC binds indirectly to TRE sites specifically at enhancers

Previous studies have established that MYC regulates gene expression by binding to both promoters and enhancers [1, 20, 31, 41]. Given that MYC performs distinct gene regulatory functions at promoters and enhancers [1, 31, 41], we examined the differences in MYC binding patterns across these regions. This analysis was conducted for seven cell lines, for which MYC, H3K4me3, and H3K27ac ChIP-seq datasets were available in the ENCODE database (Supplementary Tables S1 and S2).

MYC-bound transcriptionally active promoters and MYC-bound active enhancer regions for seven cell lines were identified using the H3K4me3 and H3K27ac ChIP-seq datasets, respectively (see “Materials and methods” section). As expected, we observed MYC binding at both promoter (marked by H3K4me3) and enhancer (marked by H3K27ac) regions at individual gene locus (Fig. 2A and B), as well as on the genome-wide level (Fig. 2C and Supplementary Fig. S3). We examined the fraction of MYC ChIP-seq peaks corresponding to promoters and enhancers across the seven cell lines (Fig. 2D). We found that a significant percentage of peaks were associated with both promoters and enhancers in all analyzed cell lines, highlighting the critical role of MYC binding at these regulatory elements for its function [1, 31, 41].

We then performed de novo motif discovery on MYC ChIP-seq peaks at promoters and enhancers separately for the seven cell lines (Fig. 2E and F). MYC’s consensus binding site, the EBOX motif, emerged as the first-ranked significantly enriched motif in promoters, and among the top four ranked significantly enriched motifs in enhancers across all analyzed cell lines (Fig. 2E and F). However, the indirect MYC binding site identified in our study, the TRE motif, was among the top five ranked significantly enriched motifs only in enhancers but not in promoters across all analyzed cell lines (Fig. 2E and F). These findings suggest that while EBOX serves both as a promoter and enhancer binding site for MYC, TRE serves as an enhancer-specific (but not promoter-specific) MYC binding site. We generated density plots of EBOX and TRE canonical sequences (CACGTG and TGA(G/C)TCA, respectively) for the K562 cell line (Fig. 2G–I). In line with established patterns of EBOX motif enrichment (Figs 1B and 2E and F), the CACGTG sequence was prevalent in MYC’s genome-wide, promoter, and enhancer binding regions (Fig. 2G–I). Following TRE enrichment patterns in Figs 1B and 2E and F, the TGA(G/C)TCA sequence was prevalent in MYC’s genome-wide and enhancer binding regions, but not promoter regions (Fig. 2G–I). The central localization [17, 26, 28] of CACGTG sequence (Fig. 2H and I) confirms that MYC occupancy is driven by EBOX sequence at promoters and enhancers. Similarly, the central localization of the TGA(G/C)TCA sequence in Fig. 2I confirms that MYC occupancy is driven by TRE sequence at enhancers.

Enhancer-specific MYC association with TRE sites increases with increasing MYC levels

MYC oncogene levels are often elevated in various human cancer types [2, 14], leading to altered transcriptional properties of MYC in cancer cells [1, 4, 17–20]. Hence, we investigated whether MYC expression levels could also influence the extent of MYC’s indirect binding to TRE sites through cooperation with AP-1. We analyzed ChIP-seq datasets from the U2OS cell line with Doxycycline (Dox)-inducible MYC expression (Fig. 3A and Supplementary Table S3). In the absence of Dox (−Dox), U2OS cells maintain low endogenous MYC levels (low-MYC conditions) [20, 32]. However, upon Dox addition (+Dox), they exhibit overexpressed MYC levels (high-MYC conditions) [20, 32]. The number of peaks in MYC ChIP-seq of U2OS cells increased with the addition of Dox, transitioning from low-MYC to high-MYC conditions (Fig. 3B).

Multi-panel figure showing the effect of MYC expression levels on MYC DNA binding in the U2OS cell line, with subfigures labelled from A to F. A is a schematic illustration of doxycycline-inducible MYC expression in the U2OS cell line. B is a bar plot comparing the total number of MYC ChIP-seq peaks in low and high MYC conditions. C, D, and E are bar plots comparing the enrichment of EBOX and TRE motifs in low and high MYC conditions across genome-wide, promoter, and enhancer binding regions, respectively. F is a bar plot showing the ratio of TRE to EBOX motif enrichment at promoter and enhancer regions under low and high MYC conditions. — Using U2OS MYC ChIP-seq datasets to show increase in MYC occupancy of EBOX and TRE sites with increase in MYC levels. (A) Schematic depicting doxycycline-inducible MYC expression in the U2OS cell line. Schematic created in BioRender. Kalyan Sundaram, R. (2025) https://BioRender.com/vzrpvs1. (B) Number of MYC ChIP-seq peaks identified from MYC ChIP-seq datasets of U2OS cell line under low- (−Dox) and high- (+Dox) MYC conditions. Occurrence of EBOX and TRE motifs in (C) all genomic regions, (D) promoters, and (E) enhancers bound by MYC in the U2OS cell line as a function of MYC levels. P-value calculated using two-sided Mann–Whitney U test. (F) Relative occupancy (TRE/EBOX) for promoters and enhancers bound by MYC in low- (−Dox) and high- (+Dox) MYC conditions for the U2OS cell line. P-value calculated using two-sided Welch’s t-test. ***P < 0.001, **P < 0.01, *P < 0.05, and n.s. is not significant. (C–F)Data represents mean and standard deviation of four replicates of motif discovery runs.

When we performed de novo motif discovery on these low- and high-MYC ChIP-seq datasets, we found that the enrichment trends of EBOX and TRE motifs were consistent with previously established patterns of MYC binding observed in MYC ChIP-seq analysis from the ENCODE database (Figs 1 and 2). Specifically, both EBOX and TRE motifs were significantly enriched in low- and high-MYC conditions (Supplementary Fig. S4A). Furthermore, the EBOX motif was significantly enriched in both MYC-bound promoters and enhancers, whereas the TRE motif was highly abundant exclusively in MYC-bound enhancers but not in promoters (Supplementary Fig. S4B and C). These enrichment patterns were observed under both low- and high-MYC conditions. Previous analysis, using both MYC and AP-1 ChIP-seq datasets for multiple cell lines (Fig. 1F and Supplementary Fig. S1C), showed that MYC indirectly binds to TRE sites through cooperation with AP-1. Based on this, we inferred that MYC’s enhancer-specific occupancy of TRE sites in U2OS cell line, at both low and high MYC levels, is also mediated by cooperative interaction with AP-1.

To study how MYC binding patterns change with varying MYC expression levels, we calculated the occurrence of de novo enriched EBOX and TRE motifs in low- and high-MYC ChIP-seq datasets. There was a significant increase in the occurrence of EBOX motifs with increasing MYC levels across MYC’s genome-wide, promoter, and enhancer binding regions (Fig. 3C–E). In contrast, the occurrence of TRE motifs was significantly increased with increasing MYC levels across MYC’s genome-wide and enhancer binding regions, but not in promoters (Fig. 3C–E). This suggests that, with increasing MYC levels, MYC occupancy of TRE motifs in cooperation with AP-1 increases at enhancers, but not at promoters.

To quantify the relative enrichment of TRE motifs compared to EBOX motifs, we defined a metric called relative occupancy (RO), representing the ratio of TRE motif enrichment to EBOX motif enrichment (Fig. 3F). We examined the RO values across promoter and enhancer MYC binding regions. The RO value is significantly higher at enhancers than promoters in both low- and high-MYC cells. At promoters, the RO value was considerably less than one under both low- and high-MYC conditions. Thus, MYC primarily utilizes promoters to directly bind to DNA through EBOX sites rather than occupying TRE sites. In contrast, at enhancers, the RO value was around one under both low- and high-MYC conditions. This suggests that, at enhancers, MYC’s indirect binding to TRE sites through cooperation with AP-1 occurs as frequently as direct binding of MYC to DNA through EBOX sites.

MYC’s association with the TRE binding site at enhancers serves to regulate specific biological processes

Previous analyses revealed that MYC occupies TRE sites in cooperation with AP-1 specifically at enhancers, and that the number of such TRE sites occupied increases with increasing levels of MYC. However, the biological significance of MYC’s indirect binding to TRE sites at enhancers still remains to be determined. To help address this question, we performed two types of analyses. First, to gain a comprehensive understanding of MYC’s transcriptional network, we studied all genes directly bound by MYC in low- and high-MYC ChIP-seq datasets (referred to as “Analysis 1,” Fig. 4A left). Genes were classified as MYC “targets” only if they contained a MYC ChIP-seq peak at their promoters. MYC target genes identified through the analysis correspond to potential candidate genes that can be regulated upon MYC binding.

The identified MYC target genes were then divided into five gene sets (Fig. 4A), see “Materials and methods” section. The broadest set corresponds to all MYC target genes (gray). Within this set, genes that contained an additional MYC ChIP-seq peak at their enhancers were selected (black). These genes were further analyzed to identify exact matches for EBOX (CACGTG) (green) or TRE (TGA(G/C)TCA) (yellow) sequences within their enhancer MYC ChIP-seq peaks. Target genes with MYC-bound TRE at enhancers (yellow) and those with MYC-bound EBOX at enhancers (green) were selected for further analysis to characterize the role played by MYC occupancy of TRE and EBOX at enhancers. These two gene sets correspond to three types of MYC enhancer elements (EEs): EBOX-only, TRE-only, and both EBOX and TRE (referred to as EBOX-TRE) (Supplementary data S2). GO analysis was performed separately for genes associated with each of the EEs (as described in “Materials and methods” section) to identify their associated biological processes (Supplementary Fig. S5C and D; Supplementary data S2). GO terms corresponding to similar processes were grouped and labeled. Labeled groups include transcription, signaling, phosphorylation, apoptosis, cell cycle, cell proliferation, and cell migration (Supplementary data S2).

We observed that when MYC levels go from low to high, the number of genes bound to each subset increased (Supplementary Fig. S5A and B). Comparing the GO tables between low- (Supplementary Fig. S5C) and high-MYC conditions (Supplementary Fig. S5D) revealed distinct binding patterns. Under low-MYC conditions, biological processes were regulated by MYC binding to EBOX-only and TRE-only EEs (Supplementary Fig. S5C). In contrast, under high-MYC conditions, MYC also regulated processes by binding to EBOX-TRE EE (Supplementary Fig. S5D).

Several processes were regulated in both low- and high-MYC conditions but exhibited different binding patterns (Supplementary Fig. S5C and D). In low-MYC conditions (Supplementary Fig. S5C), transcription-related processes were regulated by MYC binding to TRE-only EE, whereas in high-MYC conditions (Supplementary Fig. S5D), these processes were also regulated by MYC binding to EBOX-only and EBOX-TRE EEs. Phosphorylation and signaling processes were regulated by MYC binding to EBOX-only and TRE-only EEs under low-MYC conditions (Supplementary Fig. S5C); however, in high-MYC conditions (Supplementary Fig. S5D), these processes were regulated by MYC binding to TRE-only and EBOX-TRE EEs. Cell cycle processes gained EBOX-only EE in addition to TRE-only EE when MYC levels increased from low to high (Supplementary Fig. S5C and D). However, cell proliferation, apoptosis, and cell migration remained regulated by TRE-only EE in both low- and high-MYC conditions (Supplementary Fig. S5C and D).

Distinct biological processes were also regulated by MYC binding to various EEs across low- and high-MYC conditions (Supplementary Fig. S5C and D). For instance, in low-MYC condition (Supplementary Fig. S5C), spliceosomal snRNP assembly and positive cyclin-dependent kinase activity were regulated by MYC binding to TRE-only EE. In contrast, under high-MYC conditions (Supplementary Fig. S5D), embryonic development was regulated by TRE-only EE. Additionally, under high-MYC conditions (Supplementary Fig. S5D), angiogenesis, receptor internalization, and Schwann cell development were regulated by MYC binding to EBOX-TRE EE.

Specifically, genes involved in extracellular matrix (ECM)-mediated signaling processes exhibited distinct MYC enhancer binding patterns between low- and high-MYC ChIP-seq (Supplementary Fig. S5C and D; Supplementary data S2). While cell migration was regulated by MYC binding to TRE-only EEs under both low- and high-MYC conditions (Supplementary Fig. S5C and D), cell–cell adhesion and actin cytoskeleton organization were additionally regulated through the same mechanism under high-MYC conditions (Supplementary Fig. S5D and Supplementary data S2). These findings are consistent with previous reports [17, 20, 53, 54] indicating MYC’s role in modulating ECM-related processes. Our analysis additionally reveals that these ECM-associated genes are regulated through MYC binding to TRE-only EEs, potentially in cooperativity with AP-1 (Supplementary Fig. S5C and D; Supplementary data S2). This suggests a mechanism by which MYC, via enhancer-mediated regulation at TRE sites, may modulate the metastatic ability of cells by directly regulating genes involved in cell migration, adhesion, and cytoskeletal dynamics [17, 20, 53, 54].

In order to study genes strongly regulated by MYC in cancerous conditions, we next analyzed RNA-seq data of U2OS Dox-inducible MYC cell line [20, 32] to identify genes that show significant differential expression upon MYC induction, focusing specifically on genes bound by MYC in the high-MYC ChIP-seq dataset (referred to as “Analysis 2,” Fig. 4A right). The up- and downregulated MYC target genes identified were then classified into different subsets as previously described (Fig. 4C and D; Supplementary data S2). GO analysis was performed on these target gene subsets associated with different EEs (as described in “Materials and methods” section) to identify the biological processes modulated by them (Fig. 4E and F; Supplementary data S2).

The number of genes in each subset is comparable between upregulated and downregulated genes, except for TRE-only EE, which MYC uses to downregulate more genes than to upregulate (Fig. 4C and D). In both upregulated and downregulated cases (Fig. 4E and F), transcription, signaling, cell proliferation, and apoptosis were identified as commonly regulated processes. A comparison of enhancer binding patterns for these processes between upregulated and downregulated cases revealed that GO terms for downregulated genes primarily contained TRE EE (Fig. 4F), whereas GO terms for upregulated genes contained all three EEs and exhibited distinct binding patterns (Fig. 4E). Interestingly, we observed that genes involved in endoplasmic reticulum (ER) stress related transcription were upregulated by MYC binding to EBOX-TRE EE (Fig. 4E), which may potentially represent a mechanism of stress response regulation by MYC [55].

The up- and downregulated cases also contained various distinct GO-enriched terms (Fig. 4E and F). Rhythmic process-related genes were upregulated through MYC binding to EBOX-only EE (Fig. 4E). Several other processes, including protein phosphorylation, cell cycle, DNA damage response, and neuron differentiation, were enriched for genes upregulated by MYC binding to TRE-only EE (Fig. 4E). Interestingly, actin-related processes were found to be enriched for genes downregulated through MYC binding to EBOX-only EE (Fig. 4F). Additionally, several more GO terms, including cell adhesion, angiogenesis, and embryonic development, were enriched for genes downregulated through MYC binding to TRE-only EE (Fig. 4F).

We also observed that cell migration emerged as a common GO term among both up- and downregulated gene sets specifically associated with MYC binding to TRE-only EEs (Supplementary data S2). However, other ECM-associated processes exhibited distinct patterns of regulation (Fig. 4E and F; Supplementary data S2). For instance, actin filament capping and actin cytoskeleton organization were enriched for genes downregulated via EBOX-only EEs (Fig. 4E and F), whereas processes such as integrin-mediated signaling, cell-cell adhesion, cell-matrix adhesion, cell-substrate junction assembly, and cell adhesion mediated by integrin were specifically enriched for genes downregulated through TRE-only EEs (Fig. 4E and F). These findings suggest a potential role for TRE-only EEs in mediating MYC-driven metastasis programs, notably through transcriptional downregulation, as reported previously [17, 20, 53, 54]. Consistent with our earlier analysis of MYC-bound genes (Supplementary Fig. S5C and D; Supplementary data S2), these results from differentially expressed MYC target genes (Fig. 4E and F; Supplementary data S2) further underscore the functional relevance of MYC’s indirect binding to TRE enhancer sites in controlling key ECM-associated genes and processes.

Discussion

In this work, we leverage publicly available next-generation sequencing datasets to uncover a novel mechanism that MYC may utilize to drive and sustain various types of human cancers. Using ChIP-seq, we demonstrate that TRE serves as an indirect enhancer binding site for MYC, enabling its cooperative interaction with AP-1 TFs and the modulation of its target gene expression. This cooperative interaction between major oncogenes, MYC and AP-1 families [1, 4, 29, 30], is relevant for both MYC-driven as well as AP-1 driven cancers. Prior studies have observed enrichment of the TRE motif in MYC ChIP-seq datasets [50, 56]. Separately, different studies have also observed co-association between MYC and AP-1 [27, 57, 58]. However, these observations were typically limited in scope (often confined to a single cell line), not integrated into a unifying framework, and their biological significance was also not investigated. In this study, we performed a comprehensive bioinformatic analysis of several cell lines, delineating the mechanistic framework for MYC-AP-1 interactions. We report complementary lines of evidence to demonstrate the functional relevance and biological significance of cooperation between MYC and AP-1 TFs at enhancer TRE sites. Specifically, we study the broad DNA binding patterns of MYC by analyzing MYC ChIP-seq datasets. Through de novo motif discovery on ENCODE MYC ChIP-seq datasets, we reveal the co-enrichment of the AP-1 family binding motif (TRE) [29, 30], in addition to the enrichment of the known MYC consensus binding motif (EBOX) [1, 4] in 8 out of 11 analyzed cell lines (Fig. 1A and B). Through analyses integrating MYC and various AP-1 family TF ChIP-seq datasets across seven cell lines, we demonstrate that the co-enriched TRE motif in MYC ChIP-seq datasets represents an indirect DNA binding site for MYC, occupied in cooperation with AP-1 TFs (Fig. 1C–F).

Our de novo motif enrichment analysis across seven cell lines reveals that the TRE motif is preferentially enriched at enhancers rather than promoters (Fig. 2E and F), suggesting that the TRE binding site plays an enhancer-specific role in MYC function. This finding indicates that MYC, in cooperation with AP-1, may utilize the TRE sites at enhancers to modulate gene expression from basal levels in cancer cells. Through de novo motif discovery on MYC ChIP-seq datasets at low- and high-MYC expression levels, we observe an increased MYC occupancy of TRE sites at enhancers (but not at promoters) with increasing MYC levels (Fig. 3D and E). This finding suggests that at high MYC levels, TRE sites at enhancers could become a key mechanism for MYC-mediated gene regulation in cancers. An important line of future research is to understand how an increase in MYC levels may affect expression levels of AP-1 TFs, which can in turn influence MYC’s indirect occupancy of TRE sites at enhancers.

Our genome-wide analysis reveals that TRE motif enrichment in MYC ChIP-seq occurs in a cell-type-specific manner, observed in 8 out of 11 analyzed cell lines (Fig. 1B). This cell-type specificity aligns with our finding that MYC-AP-1 interactions occur at TRE sites with enhancer functions, as enhancers are known to operate in a cell-type-specific manner [59, 60]. The availability of TRE binding sites at enhancers is determined by chromatin accessibility of these regulatory regions, which can vary considerably between cell types due to epigenetic and developmental influences [61, 62]. Furthermore, interactions between TFs and cofactors that mediate these interactions may also vary between cell types [63, 64], potentially contributing to the cell-type specificity of MYC and AP-1 cooperation. However, both MYC and AP-1 TFs are ubiquitously expressed [13, 65, 66] across most cell types, albeit to varying degrees. Despite the cell-type specificity of MYC-AP-1 cooperativity identified in this work, the widespread expression of these TFs suggests that this interaction could emerge when MYC and AP-1 expression levels are altered through various genomic [2, 5, 67, 68], epigenetic [69–71], and even microenvironment [72–76] potentiated changes during cancer initiation and progression. Hence, MYC-AP-1 cooperation has the potential to be co-opted in diverse cell types and cancers, extending MYC’s well-appreciated role as a cell-intrinsic driver [2, 12].

By integrating ChIP-seq and RNA-seq datasets of differential MYC expression (Fig. 4A and B), we identify several target genes regulated by MYC occupancy of TRE sites at enhancers (Fig. 4C and D). Our results reveal that when MYC levels increase, more of these target genes are downregulated than upregulated (Fig. 4C and D), suggesting that indirect MYC binding to TRE enhancer sites may play a more specific role in gene downregulation than upregulation. Additionally, we demonstrate that MYC utilizes TRE binding sites at enhancers to regulate a multitude of cellular processes, including transcription, signaling, proliferation, apoptosis, cell cycle, phosphorylation, and cell adhesion (Fig. 4E and F). Notably, several of these identified processes correspond to well-known cancer hallmarks [33, 34], indicating that gene regulation via indirect MYC binding to TRE sites at enhancers plays a key role in driving and sustaining cancers. Specifically, we found that several ECM-related processes regulated by MYC, such as integrin-mediated signaling, cell adhesion, and cell migration, are controlled through indirect MYC binding to TRE enhancer sites (Fig. 4E and F). These findings underscore a functional relevance of this novel binding mode, using which MYC may regulate gene programs associated with cancer metastasis [17, 20, 53, 54].

Although TRE is a well-established binding motif for the AP-1 TF family in mammalian cell lines, its role in MYC-mediated gene regulation has not been identified previously. This is likely because TRE functions as an enhancer binding site for MYC (Fig. 2E and F), which is often located at a considerable distance from its target genes [59, 60], making its identification inherently challenging. Our work reveals a novel finding that MYC cooperatively interacts with multiple (at least four) AP-1 TF family members to occupy DNA at TRE binding sites within enhancers (Figs 1C–F and 2). The AP-1 TF family consists of many members that dimerize with each other and bind to TRE sites [29, 30]. Further work is needed to determine the contribution of individual AP-1 family members and their respective dimers to this cooperativity with MYC. A previous study provided evidence of MYC forming a complex with the AP-1 TF member JUN through co-immunoprecipitation (co-IP) [77]. Co-IP between all AP-1 TFs and MYC can help identify individual AP-1 family members that form complexes with MYC in cells. Subsequent studies can then determine whether the interaction between MYC and AP-1 observed in this work is part of a regulatory complex or a direct protein–protein interaction. Future research is also needed to understand the higher-order genomic organization [78–80] potentiating the interaction between MYC-bound EBOX-containing promoters and AP-1-bound TRE-containing enhancers.

Supplementary Material

zcag010_Supplemental_Files

zcag010_supplemental_files.zip^{(7MB, zip)}

Acknowledgements

We acknowledge the ENCODE Consortium and the ENCODE production laboratory(ies) generating the particular dataset(s) utilized in this work. Computational resources were provided in part by the School of Engineering and Applied Sciences at the University of Pennsylvania, and the Pittsburgh Supercomputing Center through Access allocation MCB200101. This work also used the Galaxy web platform (https://usegalaxy.org/) for analysis. We thank members of Lim and Radhakrishnan labs for helpful discussions.

Author contributions: Reshma Kalyan Sundaram (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Investigation [equal], Methodology [equal], Resources [equal], Software [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), Kshitiz Parihar (Formal analysis [equal], Validation [equal], Writing—review & editing [equal]), Stephanie Monson (Formal analysis [equal], Validation [equal], Writing—review & editing [equal]), Ravi Radhakrishnan (Conceptualization [equal], Data curation [equal], Formal analysis [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Software [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]), and Bomyi Lim (Conceptualization [equal], Data curation [equal], Funding acquisition [equal], Investigation [equal], Methodology [equal], Project administration [equal], Resources [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—original draft [equal], Writing—review & editing [equal]).

Contributor Information

Reshma Kalyan Sundaram, Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, 19104, United States.

Kshitiz Parihar, Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, 19104, United States.

Stephanie Monson, Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, 19104, United States.

Ravi Radhakrishnan, Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, 19104, United States; Department of Bioengineering,University of Pennsylvania, Philadelphia, PA, 19104, United States.

Bomyi Lim, Department of Chemical and Biomolecular Engineering, University of Pennsylvania, Philadelphia, PA, 19104, United States.

Supplementary data

Supplementary data is available at NAR Cancer online.

Conflict of interest

None declared.

Funding

This work was supported in part by the National Institutes of Health (NIH) grants CA250044, EB01775309, and the Center for Precision Engineering for Health (CPE4H) at the University of Pennsylvania.

Data availability

The ChIP-seq datasets for the 11 cell lines (A549, HUVEC, HeLa-S3, K562, MCF7, NB4, CH12.LX, MEL, GM12878, H1, and HepG2) analyzed in this paper were accessed from ENCODE database (https://www.encodeproject.org/). The ENCODE accession IDs for all MYC, AP-1 (JUN, FOS, JUND, JUNB), H3K4me3, and H3K27ac ChIP-seq data are provided in Supplementary Table S1 (BED format files) and Supplementary Table S2 (BigWig format files). The ChIP-seq and RNA-seq datasets for U2OS cell line analyzed in this paper were accessed from Gene Expression Omnibus (GEO) database (https://www.ncbi.nlm.nih.gov/gds/). The GEO accession IDs for MYC, H3K4me3, and H3K27ac ChIP-seq data (BED format files) for U2OS cell line are provided in Supplementary Table S3. Differential expression data for ±Dox RNA-seq of the U2OS cell line was accessed from GEO (GSM1231609). The HOMER output files for all the de novo motif discovery runs showing the source data for Figs 1B, E, and F; 2E and F; and 3C–E, Supplementary Figs S1B and C, and S4 are provided in Supplementary data S1. Codes and figure-generation scripts used in this study are available in Zenodo (https://doi.org/10.5281/zenodo.18937945).

References

1. Jakobsen ST, Siersbæk R. Transcriptional regulation by MYC: an emerging new model. Oncogene. 2025;44:1–7. 10.1038/s41388-024-03174-2. [DOI] [PubMed] [Google Scholar]
2. Dhanasekaran R, Deutzmann A, Mahauad-Fernandez WD et al. The MYC oncogene—the grand orchestrator of cancer growth and immune evasion. Nat Rev Clin Oncol. 2022;19:23–36. 10.1038/s41571-021-00549-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Dang CV. MYC on the path to cancer. Cell. 2012;149:22–35. 10.1016/j.cell.2012.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Kress TR, Sabò A, Amati B. MYC: connecting selective transcriptional control to global RNA production. Nat Rev Cancer. 2015;15:593–607. 10.1038/nrc3984. [DOI] [PubMed] [Google Scholar]
5. Meyer N, Penn LZ. Reflecting on 25 years with MYC. Nat Rev Cancer. 2008;8:976–90. 10.1038/nrc2231. [DOI] [PubMed] [Google Scholar]
6. Vennstrom B, Sheiness D, Zabielski J et al. Isolation and characterization of c-myc, a cellular homolog of the oncogene (v-myc) of avian myelocytomatosis virus strain 29. J Virol. 1982;42:773–9. 10.1128/jvi.42.3.773-779.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Colby WW, Chen EY, Smith DH et al. Identification and nucleotide sequence of a human locus homologous to the v-myc oncogene of avian myelocytomatosis virus MC29. Nature. 1983;301:722–5. 10.1038/301722a0. [DOI] [PubMed] [Google Scholar]
8. Battey J, Moulding C, Taub R et al. The human c-myc oncogene: structural consequences of translocation into the IgH locus in Burkitt lymphoma. Cell. 1983;34:779–87. 10.1016/0092-8674(83)90534-2. [DOI] [PubMed] [Google Scholar]
9. Bernard O, Cory S, Gerondakis S et al. Sequence of the murine and human cellular myc oncogenes and two modes of myc transcription resulting from chromosome translocation in B lymphoid tumours. EMBO J. 1983;2:2375–83. 10.1002/j.1460-2075.1983.tb01749.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Watson DK, Psallidopoulos MC, Samuel KP et al. Nucleotide sequence analysis of human c-myc locus, chicken homologue, and myelocytomatosis virus MC29 transforming gene reveals a highly conserved gene product. Proc Natl Acad Sci USA. 1983;80:3642–5. 10.1073/pnas.80.12.3642. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Van Beneden RJ, Watson DK, Chen TT et al. Cellular myc (c-myc) in fish (rainbow trout): its relationship to other vertebrate myc genes and to the transforming genes of the MC29 family of viruses. Proc Natl Acad Sci USA. 1986;83:3698–702. 10.1073/pnas.83.11.3698. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Gabay M, Li Y, Felsher DW. MYC activation is a hallmark of cancer initiation and maintenance. Cold Spring Harb Perspect Med. 2014;4:a014241 10.1101/cshperspect.a014241. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Llombart V, Mansour MR. Therapeutic targeting of “undruggable” MYC. EBioMedicine. 2022;75:103756. 10.1016/j.ebiom.2021.103756. [DOI] [PMC free article] [PubMed] [Google Scholar]
14. Lourenco C, Resetca D, Redel C et al. MYC protein interactors in gene transcription and cancer. Nat Rev Cancer. 2021;21:579–91. 10.1038/s41568-021-00367-9. [DOI] [PubMed] [Google Scholar]
15. Whitfield JR, Soucek L. MYC in cancer: from undruggable target to clinical trials. Nat Rev Drug Discov. 2025;24:445–57. 10.1038/s41573-025-01143-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Wolf E, Eilers M. Targeting MYC proteins for tumor therapy. Annu Rev Cancer Biol. 2020;4:61–75. 10.1146/annurev-cancerbio-030518-055826. [DOI] [Google Scholar]
17. Lorenzin F, Benary U, Baluapuri A et al. Different promoter affinities account for specificity in MYC-dependent gene regulation. eLife. 2016;5:e15161. 10.7554/eLife.15161. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Kress TR, Pellanda P, Pellegrinet L et al. Identification of MYC-dependent transcriptional programs in oncogene-addicted liver tumors. Cancer Res. 2016;76:3463–72. 10.1158/0008-5472.CAN-16-0316. [DOI] [PubMed] [Google Scholar]
19. Sabò A, Kress TR, Pelizzola M et al. Selective transcriptional regulation by Myc in cellular growth control and lymphomagenesis. Nature. 2014;511:488–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. Walz S, Lorenzin F, Morton J et al. Activation and repression by oncogenic MYC shape tumour-specific gene expression profiles. Nature. 2014;511:483–7. 10.1038/nature13473. [DOI] [PMC free article] [PubMed] [Google Scholar]
21. Tu WB, Helander S, Pilstål R et al. Myc and its interactors take shape. Biochim Biophys Acta. 2015;1849:469–83. 10.1016/j.bbagrm.2014.06.002. [DOI] [PubMed] [Google Scholar]
22. Barski A, Cuddapah S, Cui K et al. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–37. 10.1016/j.cell.2007.05.009. [DOI] [PubMed] [Google Scholar]
23. Johnson DS, Mortazavi A, Myers RM et al. Genome-wide mapping of in vivo protein–DNA interactions. Science. 2007;316:1497–502. 10.1126/science.1141319. [DOI] [PubMed] [Google Scholar]
24. Robertson G, Hirst M, Bainbridge M et al. Genome-wide profiles of STAT1 DNA association using chromatin immunoprecipitation and massively parallel sequencing. Nat Methods. 2007;4:651–7. 10.1038/nmeth1068. [DOI] [PubMed] [Google Scholar]
25. Mikkelsen TS, Ku M, Jaffe DB et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007;448:553–60. 10.1038/nature06008. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Pellanda P, Dalsass M, Filipuzzi M et al. Integrated requirement of non-specific and sequence-specific DNA binding in Myc-driven transcription. EMBO J. 2021;40:e105464. 10.15252/embj.2020105464. [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Dunham I, Kundaje A, Aldred SF et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Feldker N, Ferrazzi F, Schuhwerk H et al. Genome-wide cooperation of EMT transcription factor ZEB1 with YAP and AP-1 in breast cancer. EMBO J. 2020;39:e103209. 10.15252/embj.2019103209. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer. 2003;3:859–68. 10.1038/nrc1209. [DOI] [PubMed] [Google Scholar]
30. Garces de los Fayos Alonso I, Liang H-C, Turner SD et al. The role of activator protein-1 (AP-1) family members in CD30-positive lymphomas. Cancers. 2018;10:93. 10.3390/cancers10040093. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Jakobsen ST, Jensen RAM, Madsen MS et al. MYC activity at enhancers drives prognostic transcriptional programs through an epigenetic switch. Nat Genet. 2024;56:663–74. 10.1038/s41588-024-01676-z. [DOI] [PubMed] [Google Scholar]
32. Muthalagu N, Junttila MR, Wiese KE et al. BIM is the primary mediator of MYC-induced apoptosis in multiple solid tissues. Cell Rep. 2014;8:1347–53. 10.1016/j.celrep.2014.07.057. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100:57–70. 10.1016/S0092-8674(00)81683-9. [DOI] [PubMed] [Google Scholar]
34. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–74. 10.1016/j.cell.2011.02.013. [DOI] [PubMed] [Google Scholar]
35. Duffy MJ, O’Grady S, Tang M et al. MYC as a target for cancer treatment. Cancer Treat Rev. 2021;94:102154. 10.1016/j.ctrv.2021.102154. [DOI] [PubMed] [Google Scholar]
36. Madden SK, de Araujo AD, Gerhardt M et al. Taking the Myc out of cancer: toward therapeutic strategies to directly inhibit c-Myc. Mol Cancer. 2021;20:3. 10.1186/s12943-020-01291-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
37. Heinz S, Benner C, Spann N et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38:576–89. 10.1016/j.molcel.2010.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26:841–2. 10.1093/bioinformatics/btq033. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Robinson JT, Thorvaldsdóttir H, Winckler W et al. Integrative genomics viewer. Nature Biotechnol. 2011; 29:24–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Thorvaldsdóttir H, Robinson JT, Mesirov JP. Integrative genomics viewer (IGV): high-performance genomics data visualization and exploration. Brief Bioinform. 2012;14:178–92. 10.1093/bib/bbs017. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Lin CY, Lovén J, Rahl PB et al. Transcriptional amplification in tumor cells with elevated c-Myc. Cell. 2012;151:56–67. 10.1016/j.cell.2012.08.026. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Ramírez F, Ryan DP, Grüning B et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 2016;44:W160–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
43. Sherman BT, Hao M, Qiu J et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022;50:W216–21. 10.1093/nar/gkac194. [DOI] [PMC free article] [PubMed] [Google Scholar]
44. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4:44–57. 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
45. Ma X, Lee H, Wang L et al. CGI: a new approach for prioritizing genes by combining gene expression and protein–protein interaction data. Bioinformatics. 2006;23:215–21. 10.1093/bioinformatics/btl569. [DOI] [PubMed] [Google Scholar]
46. Schaefer MH, Lopes TJS, Mah N et al. Adding protein context to the human protein-protein interaction network to reveal meaningful interactions. PLoS Comput Biol. 2013;9:e1002860. 10.1371/journal.pcbi.1002860. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Gitter A, Bar-Joseph Z. Identifying proteins controlling key disease signaling pathways. Bioinformatics. 2013;29:i227–36. 10.1093/bioinformatics/btt241. [DOI] [PMC free article] [PubMed] [Google Scholar]
48. del Toro N, Shrivastava A, Ragueneau E et al. The IntAct database: efficient access to fine-grained molecular interaction data. Nucleic Acids Res. 2021;50:D648–53. 10.1093/nar/gkab1006. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Neph S, Vierstra J, Stergachis AB et al. An expansive human regulatory lexicon encoded in transcription factor footprints. Nature. 2012;489:83–90. 10.1038/nature11212. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Wang J, Zhuang J, Iyer S et al. Sequence features and chromatin structure around the genomic regions bound by 119 human transcription factors. Genome Res. 2012;22:1798–812. 10.1101/gr.139105.112. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Ronzio M, Zambelli F, Dolfini D et al. Integrating peak colocalization and motif enrichment analysis for the discovery of genome-wide regulatory modules and transcription factor recruitment rules. Front Genet. 2020;11:72. 10.3389/fgene.2020.00072. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Liang J, Lacroix L, Gamot A et al. Chromatin immunoprecipitation indirect peaks highlight long-range interactions of insulator proteins and Pol II pausing. Mol Cell. 2014;53:672–81. 10.1016/j.molcel.2013.12.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
53. Liu H, Radisky DC, Yang D et al. MYC suppresses cancer metastasis by direct transcriptional silencing of αv and β3 integrin subunits. Nat Cell Biol. 2012;14:567–74. 10.1038/ncb2491. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Eilers M, Eisenman RN. Myc’s broad reach. Genes Dev. 2008;22:2755–66. 10.1101/gad.1712408. [DOI] [PMC free article] [PubMed] [Google Scholar]
55. Tameire F, Verginadis II, Leli NM et al. ATF4 couples MYC-dependent translational activity to bioenergetic demands during tumour progression. Nat Cell Biol. 2019;21:889–99. 10.1038/s41556-019-0347-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. Lu R, Mucaki EJ, Rogan PK. Discovery and validation of information theory-based transcription factor and cofactor binding site motifs. Nucleic Acids Res. 2016;45:e27. 10.1093/nar/gkw1036. [DOI] [PMC free article] [PubMed] [Google Scholar]
57. Guo J, Li T, Schipper J et al. Sequence specificity incompletely defines the genome-wide occupancy of Myc. Genome Biol. 2014;15:482. 10.1186/s13059-014-0482-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Roca H, Pande M, Huo JS et al. A bioinformatics approach reveals novel interactions of the OVOL transcription factors in the regulation of epithelial – mesenchymal cell reprogramming and cancer progression. BMC Syst Biol. 2014;8:29. 10.1186/1752-0509-8-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
59. Heinz S, Romanoski CE, Benner C et al. The selection and function of cell type-specific enhancers. Nat Rev Mol Cell Biol. 2015;16:144–54. 10.1038/nrm3949. [DOI] [PMC free article] [PubMed] [Google Scholar]
60. Schoenfelder S, Fraser P. Long-range enhancer–promoter contacts in gene expression control. Nat Rev Genet. 2019;20:437–55. 10.1038/s41576-019-0128-0. [DOI] [PubMed] [Google Scholar]
61. Calo E, Wysocka J. Modification of enhancer chromatin: what, how, and why?. Mol Cell. 2013;49:825–37. 10.1016/j.molcel.2013.01.038. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Hu Z, Tee WW. Enhancers and chromatin structures: regulatory hubs in gene expression and diseases. Biosci Rep. 2017;37:BSR20160183. [DOI] [PMC free article] [PubMed] [Google Scholar]
63. Wang Z, Wang P, Li Y et al. Interplay between cofactors and transcription factors in hematopoiesis and hematological malignancies. Sig Transduct Target Ther. 2021;6:24. 10.1038/s41392-020-00422-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
64. Inge MM, Miller R, Hook H et al. Rapid profiling of transcription factor-cofactor interaction networks reveals principles of epigenetic regulation. Nucleic Acids Res. 2024;52:10276–96. 10.1093/nar/gkae706. [DOI] [PMC free article] [PubMed] [Google Scholar]
65. Liu H, Deng X, Shyu YJ et al. Mutual regulation of c-Jun and ATF2 by transcriptional activation and subcellular localization. EMBO J. 2006;25:1058–69. 10.1038/sj.emboj.7601020. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Köster K-A, Duque Escobar J, Fietkau A et al. Increase of c-FOS promoter transcriptional activity by the dual leucine zipper kinase. Naunyn Schmiedebergs Arch Pharmacol. 2023;396:1223–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
67. Fittall MW, Mifsud W, Pillay N et al. Recurrent rearrangements of FOS and FOSB define osteoblastoma. Nat Commun. 2018;9:2150. 10.1038/s41467-018-04530-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
68. Panagopoulos I, Heim S. Neoplasia-associated chromosome translocations resulting in gene truncation. Cancer Genomics Proteomics. 2022;19:647–72. 10.21873/cgp.20349. [DOI] [PMC free article] [PubMed] [Google Scholar]
69. Fatma H, Maurya SK, Siddique HR. Epigenetic modifications of c-MYC: role in cancer cell reprogramming, progression and chemoresistance. Semin Cancer Biol. 2022;83:166–76. 10.1016/j.semcancer.2020.11.008. [DOI] [PubMed] [Google Scholar]
70. Schuijers J, Manteiga JC, Weintraub AS et al. Transcriptional dysregulation of MYC reveals common enhancer-docking mechanism. Cell Rep. 2018;23:349–60. 10.1016/j.celrep.2018.03.056. [DOI] [PMC free article] [PubMed] [Google Scholar]
71. Joo J-Y, Schaukowitch K, Farbiak L et al. Stimulus-specific combinatorial functionality of neuronal c-fos enhancers. Nat Neurosci. 2016;19:75–83. 10.1038/nn.4170. [DOI] [PMC free article] [PubMed] [Google Scholar]
72. Chiariello M, Marinissen MJ, Gutkind JS. Regulation of c-myc expression by PDGF through Rho GTPases. Nat Cell Biol. 2001;3:580–6. 10.1038/35078555. [DOI] [PubMed] [Google Scholar]
73. Benaud CM, Dickson RB. Regulation of the expression of c-Myc by β1 integrins in epithelial cells. Oncogene. 2001;20:759–68. 10.1038/sj.onc.1204152. [DOI] [PubMed] [Google Scholar]
74. Peake MA, Cooling LM, Magnay JL et al. Selected contribution: regulatory pathways involved in mechanical induction of c-fos gene expression in bone cells. J Appl Physiol. 2000;89:2498–507. 10.1152/jappl.2000.89.6.2498. [DOI] [PubMed] [Google Scholar]
75. Shynlova OP, Oldenhof AD, Liu M et al. Regulation of c-fos expression by static stretch in rat myometrial smooth muscle cells. Am J Obstet Gynecol. 2002;186:1358–65. 10.1067/mob.2002.122415. [DOI] [PubMed] [Google Scholar]
76. Cowles EA, Brailey LL, Gronowicz GA. Integrin-mediated signaling regulates AP-1 transcription factors and proliferation in osteoblasts. J Biomed Mater Res. 2000;52:725–37. 10.1002/1097-4636(20001215)52:4<725::AID-JBM18>3.0.CO;2-O. [DOI] [PubMed] [Google Scholar]
77. Levy S, Forman HJ. C-Myc is a Nrf2-interacting protein that negatively regulates phase II genes through their electrophile responsive elements. IUBMB Life. 2010;62:237–46. 10.1002/iub.314. [DOI] [PMC free article] [PubMed] [Google Scholar]
78. Uyehara CM, Apostolou E. 3D enhancer-promoter interactions and multi-connected hubs: organizational principles and functional roles. Cell Rep. 2023;42:112068. 10.1016/j.celrep.2023.112068. [DOI] [PMC free article] [PubMed] [Google Scholar]
79. Paldi F, Cavalli G. 3D genome folding in epigenetic regulation and cellular memory. Trends Cell Biol. 2026;36:28–41. 10.1016/j.tcb.2025.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
80. Buitrago N, Vidal JA, Lim B. Configuring the code: enhancer-promoter arrangement and transcriptional regulation. J Mol Biol. 2026;438:169417. 10.1016/j.jmb.2025.169417. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

zcag010_Supplemental_Files

zcag010_supplemental_files.zip^{(7MB, zip)}

Data Availability Statement

[B1] 1. Jakobsen ST, Siersbæk R. Transcriptional regulation by MYC: an emerging new model. Oncogene. 2025;44:1–7. 10.1038/s41388-024-03174-2. [DOI] [PubMed] [Google Scholar]

[B2] 2. Dhanasekaran R, Deutzmann A, Mahauad-Fernandez WD et al. The MYC oncogene—the grand orchestrator of cancer growth and immune evasion. Nat Rev Clin Oncol. 2022;19:23–36. 10.1038/s41571-021-00549-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B3] 3. Dang CV. MYC on the path to cancer. Cell. 2012;149:22–35. 10.1016/j.cell.2012.03.003. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B4] 4. Kress TR, Sabò A, Amati B. MYC: connecting selective transcriptional control to global RNA production. Nat Rev Cancer. 2015;15:593–607. 10.1038/nrc3984. [DOI] [PubMed] [Google Scholar]

[B5] 5. Meyer N, Penn LZ. Reflecting on 25 years with MYC. Nat Rev Cancer. 2008;8:976–90. 10.1038/nrc2231. [DOI] [PubMed] [Google Scholar]

[B6] 6. Vennstrom B, Sheiness D, Zabielski J et al. Isolation and characterization of c-myc, a cellular homolog of the oncogene (v-myc) of avian myelocytomatosis virus strain 29. J Virol. 1982;42:773–9. 10.1128/jvi.42.3.773-779.1982. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B7] 7. Colby WW, Chen EY, Smith DH et al. Identification and nucleotide sequence of a human locus homologous to the v-myc oncogene of avian myelocytomatosis virus MC29. Nature. 1983;301:722–5. 10.1038/301722a0. [DOI] [PubMed] [Google Scholar]

[B8] 8. Battey J, Moulding C, Taub R et al. The human c-myc oncogene: structural consequences of translocation into the IgH locus in Burkitt lymphoma. Cell. 1983;34:779–87. 10.1016/0092-8674(83)90534-2. [DOI] [PubMed] [Google Scholar]

[B9] 9. Bernard O, Cory S, Gerondakis S et al. Sequence of the murine and human cellular myc oncogenes and two modes of myc transcription resulting from chromosome translocation in B lymphoid tumours. EMBO J. 1983;2:2375–83. 10.1002/j.1460-2075.1983.tb01749.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B10] 10. Watson DK, Psallidopoulos MC, Samuel KP et al. Nucleotide sequence analysis of human c-myc locus, chicken homologue, and myelocytomatosis virus MC29 transforming gene reveals a highly conserved gene product. Proc Natl Acad Sci USA. 1983;80:3642–5. 10.1073/pnas.80.12.3642. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B11] 11. Van Beneden RJ, Watson DK, Chen TT et al. Cellular myc (c-myc) in fish (rainbow trout): its relationship to other vertebrate myc genes and to the transforming genes of the MC29 family of viruses. Proc Natl Acad Sci USA. 1986;83:3698–702. 10.1073/pnas.83.11.3698. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B12] 12. Gabay M, Li Y, Felsher DW. MYC activation is a hallmark of cancer initiation and maintenance. Cold Spring Harb Perspect Med. 2014;4:a014241 10.1101/cshperspect.a014241. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B13] 13. Llombart V, Mansour MR. Therapeutic targeting of “undruggable” MYC. EBioMedicine. 2022;75:103756. 10.1016/j.ebiom.2021.103756. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B14] 14. Lourenco C, Resetca D, Redel C et al. MYC protein interactors in gene transcription and cancer. Nat Rev Cancer. 2021;21:579–91. 10.1038/s41568-021-00367-9. [DOI] [PubMed] [Google Scholar]

[B15] 15. Whitfield JR, Soucek L. MYC in cancer: from undruggable target to clinical trials. Nat Rev Drug Discov. 2025;24:445–57. 10.1038/s41573-025-01143-2. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B16] 16. Wolf E, Eilers M. Targeting MYC proteins for tumor therapy. Annu Rev Cancer Biol. 2020;4:61–75. 10.1146/annurev-cancerbio-030518-055826. [DOI] [Google Scholar]

[B17] 17. Lorenzin F, Benary U, Baluapuri A et al. Different promoter affinities account for specificity in MYC-dependent gene regulation. eLife. 2016;5:e15161. 10.7554/eLife.15161. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B18] 18. Kress TR, Pellanda P, Pellegrinet L et al. Identification of MYC-dependent transcriptional programs in oncogene-addicted liver tumors. Cancer Res. 2016;76:3463–72. 10.1158/0008-5472.CAN-16-0316. [DOI] [PubMed] [Google Scholar]

[B19] 19. Sabò A, Kress TR, Pelizzola M et al. Selective transcriptional regulation by Myc in cellular growth control and lymphomagenesis. Nature. 2014;511:488–92. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B20] 20. Walz S, Lorenzin F, Morton J et al. Activation and repression by oncogenic MYC shape tumour-specific gene expression profiles. Nature. 2014;511:483–7. 10.1038/nature13473. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B21] 21. Tu WB, Helander S, Pilstål R et al. Myc and its interactors take shape. Biochim Biophys Acta. 2015;1849:469–83. 10.1016/j.bbagrm.2014.06.002. [DOI] [PubMed] [Google Scholar]

[B22] 22. Barski A, Cuddapah S, Cui K et al. High-resolution profiling of histone methylations in the human genome. Cell. 2007;129:823–37. 10.1016/j.cell.2007.05.009. [DOI] [PubMed] [Google Scholar]

[B23] 23. Johnson DS, Mortazavi A, Myers RM et al. Genome-wide mapping of in vivo protein–DNA interactions. Science. 2007;316:1497–502. 10.1126/science.1141319. [DOI] [PubMed] [Google Scholar]

[B24] 24. Robertson G, Hirst M, Bainbridge M et al. Genome-wide profiles of STAT1 DNA association using chromatin immunoprecipitation and massively parallel sequencing. Nat Methods. 2007;4:651–7. 10.1038/nmeth1068. [DOI] [PubMed] [Google Scholar]

[B25] 25. Mikkelsen TS, Ku M, Jaffe DB et al. Genome-wide maps of chromatin state in pluripotent and lineage-committed cells. Nature. 2007;448:553–60. 10.1038/nature06008. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B26] 26. Pellanda P, Dalsass M, Filipuzzi M et al. Integrated requirement of non-specific and sequence-specific DNA binding in Myc-driven transcription. EMBO J. 2021;40:e105464. 10.15252/embj.2020105464. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B27] 27. Dunham I, Kundaje A, Aldred SF et al. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57–74. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B28] 28. Feldker N, Ferrazzi F, Schuhwerk H et al. Genome-wide cooperation of EMT transcription factor ZEB1 with YAP and AP-1 in breast cancer. EMBO J. 2020;39:e103209. 10.15252/embj.2019103209. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B29] 29. Eferl R, Wagner EF. AP-1: a double-edged sword in tumorigenesis. Nat Rev Cancer. 2003;3:859–68. 10.1038/nrc1209. [DOI] [PubMed] [Google Scholar]

[B30] 30. Garces de los Fayos Alonso I, Liang H-C, Turner SD et al. The role of activator protein-1 (AP-1) family members in CD30-positive lymphomas. Cancers. 2018;10:93. 10.3390/cancers10040093. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B31] 31. Jakobsen ST, Jensen RAM, Madsen MS et al. MYC activity at enhancers drives prognostic transcriptional programs through an epigenetic switch. Nat Genet. 2024;56:663–74. 10.1038/s41588-024-01676-z. [DOI] [PubMed] [Google Scholar]

[B32] 32. Muthalagu N, Junttila MR, Wiese KE et al. BIM is the primary mediator of MYC-induced apoptosis in multiple solid tissues. Cell Rep. 2014;8:1347–53. 10.1016/j.celrep.2014.07.057. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B33] 33. Hanahan D, Weinberg RA. The hallmarks of cancer. Cell. 2000;100:57–70. 10.1016/S0092-8674(00)81683-9. [DOI] [PubMed] [Google Scholar]

[B34] 34. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144:646–74. 10.1016/j.cell.2011.02.013. [DOI] [PubMed] [Google Scholar]

[B35] 35. Duffy MJ, O’Grady S, Tang M et al. MYC as a target for cancer treatment. Cancer Treat Rev. 2021;94:102154. 10.1016/j.ctrv.2021.102154. [DOI] [PubMed] [Google Scholar]

[B36] 36. Madden SK, de Araujo AD, Gerhardt M et al. Taking the Myc out of cancer: toward therapeutic strategies to directly inhibit c-Myc. Mol Cancer. 2021;20:3. 10.1186/s12943-020-01291-6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B37] 37. Heinz S, Benner C, Spann N et al. Simple combinations of lineage-determining transcription factors prime cis-regulatory elements required for macrophage and B cell identities. Mol Cell. 2010;38:576–89. 10.1016/j.molcel.2010.05.004. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B38] 38. Quinlan AR, Hall IM. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinformatics. 2010;26:841–2. 10.1093/bioinformatics/btq033. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B39] 39. Robinson JT, Thorvaldsdóttir H, Winckler W et al. Integrative genomics viewer. Nature Biotechnol. 2011; 29:24–6. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B40] 40. Thorvaldsdóttir H, Robinson JT, Mesirov JP. Integrative genomics viewer (IGV): high-performance genomics data visualization and exploration. Brief Bioinform. 2012;14:178–92. 10.1093/bib/bbs017. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B41] 41. Lin CY, Lovén J, Rahl PB et al. Transcriptional amplification in tumor cells with elevated c-Myc. Cell. 2012;151:56–67. 10.1016/j.cell.2012.08.026. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B42] 42. Ramírez F, Ryan DP, Grüning B et al. deepTools2: a next generation web server for deep-sequencing data analysis. Nucleic Acids Res. 2016;44:W160–5. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B43] 43. Sherman BT, Hao M, Qiu J et al. DAVID: a web server for functional enrichment analysis and functional annotation of gene lists (2021 update). Nucleic Acids Res. 2022;50:W216–21. 10.1093/nar/gkac194. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B44] 44. Huang da W, Sherman BT, Lempicki RA. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 2009;4:44–57. 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]

[B45] 45. Ma X, Lee H, Wang L et al. CGI: a new approach for prioritizing genes by combining gene expression and protein–protein interaction data. Bioinformatics. 2006;23:215–21. 10.1093/bioinformatics/btl569. [DOI] [PubMed] [Google Scholar]

[B46] 46. Schaefer MH, Lopes TJS, Mah N et al. Adding protein context to the human protein-protein interaction network to reveal meaningful interactions. PLoS Comput Biol. 2013;9:e1002860. 10.1371/journal.pcbi.1002860. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B47] 47. Gitter A, Bar-Joseph Z. Identifying proteins controlling key disease signaling pathways. Bioinformatics. 2013;29:i227–36. 10.1093/bioinformatics/btt241. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B48] 48. del Toro N, Shrivastava A, Ragueneau E et al. The IntAct database: efficient access to fine-grained molecular interaction data. Nucleic Acids Res. 2021;50:D648–53. 10.1093/nar/gkab1006. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B49] 49. Neph S, Vierstra J, Stergachis AB et al. An expansive human regulatory lexicon encoded in transcription factor footprints. Nature. 2012;489:83–90. 10.1038/nature11212. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B50] 50. Wang J, Zhuang J, Iyer S et al. Sequence features and chromatin structure around the genomic regions bound by 119 human transcription factors. Genome Res. 2012;22:1798–812. 10.1101/gr.139105.112. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B51] 51. Ronzio M, Zambelli F, Dolfini D et al. Integrating peak colocalization and motif enrichment analysis for the discovery of genome-wide regulatory modules and transcription factor recruitment rules. Front Genet. 2020;11:72. 10.3389/fgene.2020.00072. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B52] 52. Liang J, Lacroix L, Gamot A et al. Chromatin immunoprecipitation indirect peaks highlight long-range interactions of insulator proteins and Pol II pausing. Mol Cell. 2014;53:672–81. 10.1016/j.molcel.2013.12.029. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B53] 53. Liu H, Radisky DC, Yang D et al. MYC suppresses cancer metastasis by direct transcriptional silencing of αv and β3 integrin subunits. Nat Cell Biol. 2012;14:567–74. 10.1038/ncb2491. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B54] 54. Eilers M, Eisenman RN. Myc’s broad reach. Genes Dev. 2008;22:2755–66. 10.1101/gad.1712408. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B55] 55. Tameire F, Verginadis II, Leli NM et al. ATF4 couples MYC-dependent translational activity to bioenergetic demands during tumour progression. Nat Cell Biol. 2019;21:889–99. 10.1038/s41556-019-0347-9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B56] 56. Lu R, Mucaki EJ, Rogan PK. Discovery and validation of information theory-based transcription factor and cofactor binding site motifs. Nucleic Acids Res. 2016;45:e27. 10.1093/nar/gkw1036. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B57] 57. Guo J, Li T, Schipper J et al. Sequence specificity incompletely defines the genome-wide occupancy of Myc. Genome Biol. 2014;15:482. 10.1186/s13059-014-0482-3. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B58] 58. Roca H, Pande M, Huo JS et al. A bioinformatics approach reveals novel interactions of the OVOL transcription factors in the regulation of epithelial – mesenchymal cell reprogramming and cancer progression. BMC Syst Biol. 2014;8:29. 10.1186/1752-0509-8-29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B59] 59. Heinz S, Romanoski CE, Benner C et al. The selection and function of cell type-specific enhancers. Nat Rev Mol Cell Biol. 2015;16:144–54. 10.1038/nrm3949. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B60] 60. Schoenfelder S, Fraser P. Long-range enhancer–promoter contacts in gene expression control. Nat Rev Genet. 2019;20:437–55. 10.1038/s41576-019-0128-0. [DOI] [PubMed] [Google Scholar]

[B61] 61. Calo E, Wysocka J. Modification of enhancer chromatin: what, how, and why?. Mol Cell. 2013;49:825–37. 10.1016/j.molcel.2013.01.038. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B62] 62. Hu Z, Tee WW. Enhancers and chromatin structures: regulatory hubs in gene expression and diseases. Biosci Rep. 2017;37:BSR20160183. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B63] 63. Wang Z, Wang P, Li Y et al. Interplay between cofactors and transcription factors in hematopoiesis and hematological malignancies. Sig Transduct Target Ther. 2021;6:24. 10.1038/s41392-020-00422-1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B64] 64. Inge MM, Miller R, Hook H et al. Rapid profiling of transcription factor-cofactor interaction networks reveals principles of epigenetic regulation. Nucleic Acids Res. 2024;52:10276–96. 10.1093/nar/gkae706. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B65] 65. Liu H, Deng X, Shyu YJ et al. Mutual regulation of c-Jun and ATF2 by transcriptional activation and subcellular localization. EMBO J. 2006;25:1058–69. 10.1038/sj.emboj.7601020. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B66] 66. Köster K-A, Duque Escobar J, Fietkau A et al. Increase of c-FOS promoter transcriptional activity by the dual leucine zipper kinase. Naunyn Schmiedebergs Arch Pharmacol. 2023;396:1223–33. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B67] 67. Fittall MW, Mifsud W, Pillay N et al. Recurrent rearrangements of FOS and FOSB define osteoblastoma. Nat Commun. 2018;9:2150. 10.1038/s41467-018-04530-z. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B68] 68. Panagopoulos I, Heim S. Neoplasia-associated chromosome translocations resulting in gene truncation. Cancer Genomics Proteomics. 2022;19:647–72. 10.21873/cgp.20349. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B69] 69. Fatma H, Maurya SK, Siddique HR. Epigenetic modifications of c-MYC: role in cancer cell reprogramming, progression and chemoresistance. Semin Cancer Biol. 2022;83:166–76. 10.1016/j.semcancer.2020.11.008. [DOI] [PubMed] [Google Scholar]

[B70] 70. Schuijers J, Manteiga JC, Weintraub AS et al. Transcriptional dysregulation of MYC reveals common enhancer-docking mechanism. Cell Rep. 2018;23:349–60. 10.1016/j.celrep.2018.03.056. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B71] 71. Joo J-Y, Schaukowitch K, Farbiak L et al. Stimulus-specific combinatorial functionality of neuronal c-fos enhancers. Nat Neurosci. 2016;19:75–83. 10.1038/nn.4170. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B72] 72. Chiariello M, Marinissen MJ, Gutkind JS. Regulation of c-myc expression by PDGF through Rho GTPases. Nat Cell Biol. 2001;3:580–6. 10.1038/35078555. [DOI] [PubMed] [Google Scholar]

[B73] 73. Benaud CM, Dickson RB. Regulation of the expression of c-Myc by β1 integrins in epithelial cells. Oncogene. 2001;20:759–68. 10.1038/sj.onc.1204152. [DOI] [PubMed] [Google Scholar]

[B74] 74. Peake MA, Cooling LM, Magnay JL et al. Selected contribution: regulatory pathways involved in mechanical induction of c-fos gene expression in bone cells. J Appl Physiol. 2000;89:2498–507. 10.1152/jappl.2000.89.6.2498. [DOI] [PubMed] [Google Scholar]

[B75] 75. Shynlova OP, Oldenhof AD, Liu M et al. Regulation of c-fos expression by static stretch in rat myometrial smooth muscle cells. Am J Obstet Gynecol. 2002;186:1358–65. 10.1067/mob.2002.122415. [DOI] [PubMed] [Google Scholar]

[B76] 76. Cowles EA, Brailey LL, Gronowicz GA. Integrin-mediated signaling regulates AP-1 transcription factors and proliferation in osteoblasts. J Biomed Mater Res. 2000;52:725–37. 10.1002/1097-4636(20001215)52:4<725::AID-JBM18>3.0.CO;2-O. [DOI] [PubMed] [Google Scholar]

[B77] 77. Levy S, Forman HJ. C-Myc is a Nrf2-interacting protein that negatively regulates phase II genes through their electrophile responsive elements. IUBMB Life. 2010;62:237–46. 10.1002/iub.314. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B78] 78. Uyehara CM, Apostolou E. 3D enhancer-promoter interactions and multi-connected hubs: organizational principles and functional roles. Cell Rep. 2023;42:112068. 10.1016/j.celrep.2023.112068. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B79] 79. Paldi F, Cavalli G. 3D genome folding in epigenetic regulation and cellular memory. Trends Cell Biol. 2026;36:28–41. 10.1016/j.tcb.2025.03.001. [DOI] [PMC free article] [PubMed] [Google Scholar]

[B80] 80. Buitrago N, Vidal JA, Lim B. Configuring the code: enhancer-promoter arrangement and transcriptional regulation. J Mol Biol. 2026;438:169417. 10.1016/j.jmb.2025.169417. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

MYC and AP-1 oncogenes cooperatively bind enhancers to rewire transcription

Reshma Kalyan Sundaram

Kshitiz Parihar

Stephanie Monson

Ravi Radhakrishnan

Bomyi Lim

Roles

Abstract

Graphical Abstract

Graphical Abstract.

Introduction

Materials and methods

Motif discovery

ChIP-seq overlap

ChIP-seq subtraction analysis

Figure 1.

IGV plotting

Promoter enhancer separation

ChIP-seq signal heatmaps

Figure 2.

Motif density plots

Relative occupancy

Gene Ontology

Figure 4.

Results

Genome-wide analysis reveals that MYC indirectly occupies TRE binding site in cooperation with AP-1

Regulatory element analysis reveals that MYC binds indirectly to TRE sites specifically at enhancers

Enhancer-specific MYC association with TRE sites increases with increasing MYC levels

Figure 3.

MYC’s association with the TRE binding site at enhancers serves to regulate specific biological processes

Discussion

Supplementary Material

Acknowledgements

Contributor Information

Supplementary data

Conflict of interest

Funding

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases