The environmental carcinogen benzo[a]pyrene regulates epigenetic reprogramming and metabolic rewiring in a two-stage mouse skin carcinogenesis model

Md Shahid Sarwar; Christina N Ramirez; Hsiao-Chen Dina Kuo; Pochung Chou; Renyi Wu; Davit Sargsyan; Yuqing Yang; Ahmad Shannar; Rebecca Mary Peter; Ran Yin; Yujue Wang; Xiaoyang Su; Ah-Ng Kong

doi:10.1093/carcin/bgad024

. 2023 Apr 26;44(5):436–449. doi: 10.1093/carcin/bgad024

The environmental carcinogen benzo[a]pyrene regulates epigenetic reprogramming and metabolic rewiring in a two-stage mouse skin carcinogenesis model

Md Shahid Sarwar ^1,^#, Christina N Ramirez ^2,^#, Hsiao-Chen Dina Kuo ^3,^#, Pochung Chou ⁴, Renyi Wu ⁵, Davit Sargsyan ⁶, Yuqing Yang ⁷, Ahmad Shannar ⁸, Rebecca Mary Peter ⁹, Ran Yin ¹⁰, Yujue Wang ¹¹, Xiaoyang Su ¹², Ah-Ng Kong ^13,^✉

PMCID: PMC10414144 PMID: 37100755

Abstract

Non-melanoma skin cancer (NMSC) is the most common cancer in the world. Environmental exposure to carcinogens is one of the major causes of NMSC initiation and progression. In the current study, we utilized a two-stage skin carcinogenesis mouse model generated by sequential exposure to cancer-initiating agent benzo[a]pyrene (BaP) and promoting agent 12-O-tetradecanoylphorbol-13-acetate (TPA), to study epigenetic, transcriptomic and metabolic changes at different stages during the development of NMSC. BaP/TPA caused significant alterations in DNA methylation and gene expression profiles in skin carcinogenesis, as evidenced by DNA-seq and RNA-seq analysis. Correlation analysis between differentially expressed genes and differentially methylated regions found that the mRNA expression of oncogenes leucine rich repeat LGI family member 2 (Lgi2), kallikrein-related peptidase 13 (Klk13) and SRY-Box transcription factor (Sox5) are correlated with the promoter CpG methylation status, indicating BaP/TPA regulates these oncogenes through regulating their promoter methylation at different stages of NMSC. Pathway analysis identified that the modulation of macrophage-stimulating protein-recepteur d’origine nantais and high-mobility group box 1 signaling pathways, superpathway of melatonin degradation, melatonin degradation 1, sirtuin signaling and actin cytoskeleton signaling pathways are associated with the development of NMSC. The metabolomic study showed BaP/TPA regulated cancer-associated metabolisms like pyrimidine and amino acid metabolisms/metabolites and epigenetic-associated metabolites, such as S-adenosylmethionine, methionine and 5-methylcytosine, indicating a critical role in carcinogen-mediated metabolic reprogramming and its consequences on cancer development. Altogether, this study provides novel insights integrating methylomic, transcriptomic and metabolic-signaling pathways that could benefit future skin cancer treatment and interception studies.

Exposure to environmental carcinogen benzo[a]pyrene causes alterations in DNA methylation, gene expression and mitochondrial metabolism in skin carcinogenesis, as evidenced by DNA-seq, RNA-seq and metabolomic analysis.

Graphical abstract

Introduction

Non-melanoma skin cancer (NMSC) is one of the most common cancers in the world, and its prevalence has been increasing over the last decades (1). NMSC consists of basal cell carcinoma and squamous cell carcinoma which is the most common skin cancer. Basal cell carcinoma and squamous cell carcinoma are the major NMSCs consisting of 99% of all NMSC cases, while the remaining NMSCs are Merkel cell carcinoma, Kaposi’s sarcoma, apocrine adenocarcinoma, sebaceous carcinoma and other rare cancers (2). The most common causes of NMSC include ultraviolet (UV) radiation and environmental chemicals/pollutants. Occupational exposure to the polycyclic aromatic hydrocarbon, organophosphate compounds and arsenic are associated with the development of NMSC (3). Benzo[a]pyrene (BaP) is a polycyclic aromatic hydrocarbon produced by inadequate combustion of carbon-containing substances, such as petroleum products, coal, wood and tobacco which are familiar sources of BaP exposure in human (4). The International Agency for Research on Cancer classified BaP as a Group-1 human carcinogen. Overexposure to BaP causes lung, bladder and skin cancers (5).

The development of skin cancer involves multiple stages, including initiation, promotion and progression. The two-stage skin carcinogenesis mouse model is a well-established experimental model utilizing chemical/physical agents as initiating and promoting agents for understanding sequential and stepwise mechanisms (6). The initiation stage is triggered by polycyclic aromatic hydrocarbons, including 7,12-dimethylbenz[a]-anthracene and BaP (5). Then, the progression stage is elicited by the repeated topical application of chemical agents such as 12-O-tetradecanoylphorbol-13-acetate (TPA) to maintain epidermal hyperplasia (6). Previous studies show that BaP and TPA exert their effect through formation of covalent DNA adduct, reactive oxygen species and pro-inflammatory cytokines to regulate the tumor suppressor genes and proto-oncogenes (7). The SKH-1 hairless mice used in this study are highly susceptible to chemical-induced skin carcinogenesis (8) with several advantages in performing experiments, including saving time for depilation, avoiding the effects of the hair cycle, large hairless area for drug application and easily identifying tumors.

Skin carcinogenesis primarily starts with DNA damage, followed by chronic inflammation, immunosuppression, photoaging and/or gene mutations. The transition processes involve genetic and epigenetic alterations, which lead to a cascade of cellular and molecular events for carcinogenesis (9). For over a decade now, non-mutational epigenetic reprogramming has been increasingly studied. The tumor microenvironment’s aberrant properties and the genome’s chronic instability are associated with epigenetic reprogramming. Growing evidence reports that cellular metabolism plays a vital role in carcinogen/oncogene/UV-driven cancers (10,11) and that cellular metabolism/metabolites are firmly related to the fundamental epigenetic machinery (12). Epigenetic changes, including histone acetylation and DNA methylation are sensitive to cellular metabolic conditions. Strong molecular association between epigenetic reprogramming and metabolic rewiring exists, which mutually regulate each other in the event of oncogenesis (13). Nucleosides/nucleosides metabolism plays an important role in cancer development by supplying the precursors necessary for DNA and RNA synthesis and is connected to epigenetic gene regulation (14). Numerous studies reported that dysregulation of pyrimidine metabolism is closely linked to tumorigenesis, and several drugs, including 5-fluorouracil targeting pyrimidine metabolism have been approved for the treatment of cancers (15). Recently, we reported that UV exposure regulates multiple metabolites and metabolism-related signaling pathways, epigenetic CpG methylation and transcriptomic gene expression in human skin keratinocytes (11). Our previous studies found that exposure to UVB irradiation drives DNA methylome and transcriptome changes during the development and progression of NMSC (16,17). However, how the environmental carcinogen BaP, one of the major causes of NMSC, regulates epigenetic and metabolic pathways during different stages of skin cancer progression is unknown.

This study investigated the alterations of the genome-wide gene expression, DNA methylation signatures and metabolic homeostasis in the two-stage skin carcinogenesis model using multi-omic liquid chromatography–mass spectrometry (LC–MS) metabolomic, RNA-seq and methyl-seq analysis. The results provide insight into metabolome, transcriptome, epigenome and signaling networks during different stages of skin carcinogenesis to identify novel therapeutic targets.

Materials and methods

Chemicals and reagents

Acetone (HPLC grade), BaP and TPA were obtained from Sigma (St. Louis, MO). Phosphate-buffered formalin (10%) was purchased from Fisher Scientific (Hampton, NH).

Animal studies

All animal studies were conducted following the guidelines for the Care and Use of Laboratory Animals of Rutgers, The State University of New Jersey. The study protocols were reviewed and approved by the Animal Care and Use Committee of Rutgers University (Protocol number: PROTO999900171). A total of 51 mice were included in the exposure group and 12 mice were in the control group. Cancer initiation was done by two topical applications of freshly prepared BaP in 200 μl of acetone to the dorsal skin in 6 to 8-week old female SKH-1 mice (18,19). The two applications of 200 and 100 nmol BaP were separated by a week. The initiation process was followed by 2 weeks of waiting time for BaP to exert its effect. Then, animals were treated with 6.8 nmol of TPA in 200 μl acetone. Before conducting the main study, we conducted pilot studies in SKH-1 mice to determine the initiating dose of BaP following the previously reported protocols (5,20). The topical application of TPA was to maintain the progression stage of skin carcinogenesis and was performed twice weekly until animal sacrifice. The mice in the negative control group were treated with 200 μl acetone. Experimental animals with ≥10% weight loss compared to control animals of a similar age were considered as the humane endpoint. At least 6 mice from the exposure groups were sacrificed at 5, 20 and 26 weeks, while 4 mice from the control group were sacrificed at 5 and 20 weeks by CO₂ asphyxiation to study the initiation, promotion and progression stages of skin carcinogenesis (6,21,22). The number of tumors, tumor volume and tumor multiplicity in the carcinogen-treated mice were measured. The dorsal skins (25 mm length and 5 mm width) of the mice were excised and the skin epidermis and dermal sections were segregated by heat treatment. The epidermis was then gently removed using a scalpel on ice. The experimental study design is shown in Figure 1A. For RNA-seq and metabolomic experiments, 3 biological replicates of these tissues at 5, 20 and 26 weeks were used. For DNA-seq analysis, 3 biological replicates at 5 and 20 weeks and 2 biological replicates at 26 weeks were used.

Figure 1. — Schematic presentation of experimental design and tumor incidence. (A) The generation of skin carcinogenesis model in female SKH-1 mice after administration of BaP/TPA. Number of tumors, tumor volume and tumor multiplicity were measured every two weeks. (B) The percentage of mice producing skin tumor between control and BaP/TPA groups. (C) Tumor volume per mouse in mm³. Tumor volume was measured using the digital slide caliper. (D) The number of tumors per mouse.

Nucleic acids isolation and next-generation sequencing

RNA and DNA samples were isolated using an AllPrep DNA/RNA Mini Kit (QIAGEN, Germantown, MD) following the method described in our previously published papers (16,17). The concentration and quality of extracted nucleic acids were measured using an Infinite M200 PRO (Tecan, Männedorf, Switzerland) and Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA). RNA samples were subjected to RNA-seq with library preparation and sequencing performed by Genewiz (South Plainfield, NJ). The RNA library was sequenced on an Illumina HiSeq instrument generating ~30–40 million reads per sample. DNA samples were prepared by SureSelect Methyl-seq Target Enrichment System (Agilent Technologies), and bisulfite conversion was performed by Genewiz. The DNA library was sequenced on an Illumina HiSeq instrument (2 × 150 bp configuration, paired end), generating 30–40 million reads per sample.

Bioinformatic analyses of RNA-seq

The RNA-seq reads were preprocessed by Cutadapt for Illumina Universal Adapter sequence recognition and removal (23). The reads were then aligned to the in silico mouse genome (mm10) by HISAT2 (version 2.1.0) (24), and aligned reads were sorted by SAMtools (version 1.8) (25). PCR duplicates were removed by the Picard tool (version 2.23.8). Next, we quantified the reads from different treatment groups, which mapped to predetermined genomic features using the FeatureCounts (version 1.5.1) program (26). Finally, 3 replicates from 5, 20 to 26 weeks time points were inputted for differential gene expression analysis with 20 weeks control used for 26 weeks BaP/TPA versus Control comparison. We analyzed the differential gene expression with DESeq2 (version 1.36.0) in R (version 4.2.1) with false discovery rate (FDR) adjusted P-values (P_adj) ≤ 0.05 as the cutoff to obtain differential expressed genes (DEGs). The genes with 0 reads in one of the samples and <10 reads on average across samples were removed. The raw counts were normalized by the sequencing depth and RNA composition using the median of ratios method built in DESeq2. The normalized counts were calculated from dividing the raw counts by sample-specific size factors calculated by median ratio of counts relative to geometric mean per gene. The differential expression was tested measuring the effect of the treatment (Control, BAP/TPA) and controlling for batch differences (replicate1, replicate2 and replicate3). The results of DEGs analysis were subjected to Approximate Posterior Estimation for generalized linear model (apeglm) to provide the inference on the maximum likelihood log₂ fold change (log₂FC) and to reduce the large estimates for the log₂FC caused by the high variance. The raw reads can be converted to transcript per million by normalization for gene length, followed by normalization for sequencing depth BaP/TPA versus Control comparison. The raw reads from RNA-seq and the data analysis pipeline can be found on the GitHub repositories (https://github.com/dina567/RNA-seq-Analysis-Workflow_Skin-Project, https://github.com/KongLabRUSP/shiny.ngs).

Bioinformatic analyses of SureSelect DNA Methyl-seq

The DNA reads were aligned to the bisulfite-converted mouse genome (mm10) with the Bismark (version 0.22.3) alignment algorithm (27). After alignment, DMRfinder (version 0.1) (28) was used to extract CpG counts, and cluster CpG sites into differential methylation regions (DMRs), with each DMR containing at least three CpG sites (28). Genomic annotation was performed using ChIPseeker (version 1.32.0) in R (29). Finally, 3 replicates from 5 to 20 weeks time points and 2 replicates from 26 weeks time point were inputted for DMR analysis and 20 weeks control was used for 26 weeks BaP/TPA versus Control comparison. The differential analysis of methylation levels was performed using DSS (version 2.44.0). The raw CpG reads from methyl-seq and the data analysis pipeline can be found on the GitHub repositories (https://github.com/KongLabRUSP/shiny.ngs, https://github.com/dina567/Methyl-seq-Analysis-Workflow_Skin-Project).

Correlation between DNA methylation and RNA expression

An inverse correlation between DNA methylation in gene promoters and gene expression is found in various tissues and species (30). The current study profiled the correlation between the methylation difference of DMRs in methylome and the corresponding DEG expression fold change in the transcriptome in BaP/TPA versus control comparisons. The correlations were visualized in Starburst plots with DNA methylation difference on x-axis and log₂FC on y-axis. The gene regions obtained from ChIPseeker annotation were further classified into promoter, body and downstream region for visualization on the Starburst plots. The gene body region covers exon, intron, 3ʹ-UTR and 5ʹ-UTR. The downstream region covers the downstream flanking region and the distal intergenic region. The length of promoter is ±3 kb. The DMRs with an inverse relationship between the difference of CpG methylation and log₂FC of gene expression that falls in the promoter region were labeled by their gene names. P_adj ≤ 0.05 for gene expression analysis and FDR ≤ 0.05 for methylation level analysis were applied for DNA methylation and RNA expression correlation. The genes of promoter DNA hypermethylation/RNA downregulation or promoter DNA hypomethylation/RNA upregulation were considered genes of interest.

LC–MS metabolomic analysis

We performed the LC–MS metabolomic study in the Metabolomics Shared Resources of Rutgers Cancer Institute of New Jersey, according to our recently published method (31). Skin samples were obtained from the dorsal area of the mouse from 2 mouse groups (Control versus BaP/TPA) at 5, 20 and 26 weeks (n = 3 per group), followed by organic extraction of tissue metabolites for metabolomic analysis. The metabolites were analyzed and 3 replicates from 5 to 20 weeks time points were compared to their corresponding control mice with 20 weeks control used for 26 weeks BaP/TPA versus Control comparison. In brief, 20–30 mg tissue was pulverized with Yttria Grinding ball using CryoMill at 20 Hz for 2 min to ensure complete homogenization of the tissue samples. The metabolites were extracted using 1 ml ice-cold lysis buffer consisting of methanol:acetonitrile:water (40:40:20) with 0.5% formic acid, followed by incubation on ice for 5 min. Then 50 μl 15% NH₄HCO₃ was added to neutralize the acetic acid. The metabolite solutions were collected into the Eppendorf tube and centrifuged at 15000 × g for 10 min at 4°C. The clear supernatant solution was collected into new tubes and stored at −80°C until further analysis by LC–MS. The MS spectra were recorded in positive and negative ionization modes with a nominal mass resolution of 70 000 (defined at m/z 200), in addition to an automatic gain control target of 3 × 10⁶ and m/z scan range 72–1000. MAVEN software was used to obtain the metabolite data, while the statistical and pathway analyses were performed by the web-based tool MetaboAnalyst (V5.0) (https://www.metaboanalyst.ca/).

Statistical analysis

The results are presented as the mean ± SD from three independent replicates, unless otherwise stated. Unpaired t-test was used to determine the significant differences between the control and BaP/TPA groups. P ≤ 0.05 was considered statistically significant. P-values obtained from the differential expression and differential methylation analyses were adjusted by FDR using Benjamini-Hochberg method. We used adjusted P-value of ≤0.05 and absolute value of log₂FC ≥ 2 as cutoffs for the RNA-seq analysis and DEGs visualization. Thresholds of methylation ratio difference ≥10% or ≤−10% and FDR adjusted P ≤ 0.05 were used as the threshold for methyl-seq analysis. P_adj ≤ 0.05 for gene expression changes and FDR ≤ 0.05 for methylation level changes were applied for correlation analysis.

Results

Development of skin carcinogenesis in SKH-1 mice

In this study, we developed a two-stage skin cancer model by combining BaP with tumor promoter TPA. To determine the optimal dose of TPA in skin cancer progression, we performed ear edema assay with three doses of TPA such as 0.65, 1.3 and 2.6 nmol/20 µl. We found that all three concentrations of TPA significantly increased the ear edema in comparison to the control (acetone) group (P < 0.001) (Supplementary Figure 1A is available at Carcinogenesis Online). Considering the possible toxic effects of high doses of TPA and no significant difference for edema effect between low and higher doses, 0.65 nmol/20 μl was used for the downstream experiments. We then performed the two-stage carcinogenesis model by administering BaP and TPA. We demonstrated that squamous papillomas are generated in SKH1 mice after administration of BaP/TPA. The percentage of mice bearing the tumors was 7.88, 9.74, 16.34 and 56.57% at 20, 22, 24 and 26 weeks, respectively (Figure 1B). The average tumor volume per mouse were 0.14 ± 0.24, 0.24 ± 0.19, 1.28 ± 0.44 and 2.54 ± 0.79 mm³ (Figure 1C), while the tumor multiplicity were 0.08 ± 0.06, 0.09 ± 0.06, 0.20 ± 0.08 and 0.79 ± 0.16 at 20, 22, 24 and 26 weeks, respectively (Figure 1D). We collected whole skin samples from the mice and processed them for histopathological analysis. Two samples were randomly selected from each group. Corresponding H&E images and pathologist consultation confirmed squamous papilloma formation in BaP/TPA administered mice (Supplementary Figure 1B is available at Carcinogenesis Online).

BaP/TPA mediates DNA methylation changes in skin carcinogenesis

Next, we performed methyl-seq analysis to evaluate the DNA methylation profile between BaP/TPA and control groups. We observed that the majority of the DMRs were found in the promoters and the distal intergenic regions [≤1 and 1–2 kb upstream of the transcription start site or downstream of the 3ʹ untranslated region (UTR)] (Supplementary Figure 2A is available at Carcinogenesis Online). The distribution pattern of the DMRs according to the number of CpGs and region also displayed that the number of CpG sites in the promoter, gene body and downstream regions was higher in comparison to other regions (Supplementary Figure 2B is available at Carcinogenesis Online). Then, the DNA methylation levels between control and BaP/TPA groups were compared. No significant methylation difference was found among different groups (Supplementary Figure 2C is available at Carcinogenesis Online). However, we observed lower CpG methylation in the promoters compared with other regions in these groups. There was a clear separation between control mice and BaP/TPA groups for each time point, as shown by the principal component analysis (PCA) (Supplementary Figure 2D is available at Carcinogenesis Online).

By examining the details of the CpG methylation alterations, we observed a gradual shift of methylation level difference from 5, 20 to 26 weeks, indicating exacerbation of the epigenetic modulation during the progression of NMSC. Total 1048, 2431 and 3005 genes related to DMRs were identified in the control versus BaP/TPA group at 5, 20 and 26 weeks, respectively, of which 867 genes were hypermethylated and 181 genes were hypomethylated at 5 weeks; 1959 genes were hypermethylated and 472 genes were hypomethylated at 20 weeks; and 2276 genes were hypermethylated and 729 genes were hypomethylated at 26 weeks (Figure 2A). Next, we visualized the DMRs regulated by the BaP/TPA and the top 10 regulated DMRs (5 hyper- and 5 hypo-methylated) are shown in volcano plots (Figure 2B). The most significantly dysregulated DMRs in the promoter region of the genes are Hlf, Gm29638, Bcorl1, Ptpn13, Hsh2d, LyzI4, Spg20, Gm30262, Gm36546 and SIc19a3 during the initiation phase, Apcdd1, F2r, Gas2I1, Plekhm2, Tcirg1, Mir1949, Gm36546 and Nedd9 during the promotion phase and Chst3, F2r, Gm39962, Cnbd2, Tcirg1, Hmgn1, Mir1949, Gm36546 and Isoc2b during the progression phase, which could be potential epigenetic targets in the interception and treatment of skin carcinogenesis.

Figure 2. — DNA methyl-seq profiles in control and BaP/TPA-treated mice at 5, 20 and 26 weeks. (A) Bar diagrams showing the hyper- and hypo-methylated genes in control and BaP/TPA-treated mice at 5, 20 and 26 weeks; (B) Volcano plot showing the DMRs in response to BaP/TPA versus control after 5, 20 and 26 weeks. The cutoff threshold of methylation ratio difference ≥10% or ≤−10%, P < 0.05.

BaP/TPA mediates transcriptomic changes in skin carcinogenesis

RNA-seq was analyzed with RNA samples extracted from the epidermis of control and BaP/TPA mice to detect gene expression changes. The counts and raw gene expression between the two groups were observed in a similar pattern (Supplementary Figure 3A is available at Carcinogenesis Online). Furthermore, analysis of PCA and Euclidean distance clustering displayed that the BaP/TPA groups were clearly separated from the control groups in 5, 20 and 26 weeks mice (Supplementary Figure 3B and C is available at Carcinogenesis Online). To examine the effect of BaP/TPA-regulated gene expression, we further analyzed the DEG profiles between control and BaP/TPA groups with a cutoff P ≤ 0.05 coupled to log₂FC ≥2.0 or ≤−2.0. The MA plot shows that 165 and 200 genes were significantly up- and down-regulated, respectively after exposure to BaP/TPA at 5 weeks; while 247 and 275 genes were significantly up- and down-regulated, respectively at 20 weeks; where 252 and 330 genes were significantly up- and down-regulated, respectively at 26 weeks (Figure 3A). The top-regulated DEGs of tumor suppressor and oncogenes are Micu1, Cdhr1, Tcf23, Mrgprb3, Spink7, Krt2, Krt77, Krt6a, Krt6b, Lgi2, Chil4, Kctd12, Krt25, Krt27, Rptn, Krt16, Lpo, Chil4 and Stfa2, indicating their critical role in carcinogen-mediated gene regulation and NMSC progression (Figure 3B). The normalized mRNA levels of the significantly up- and down-regulated skin cancer-relevant oncogenes/tumor suppressor genes are presented in Figure 4. Furthermore, the expression of 1183 miRNAs were measured in the RNA-seq profile containing 24 421 transcripts. The normalized reads of the miRNAs in different treatment groups at 5, 20 and 26 weeks have been summarized in Supplementary File 1 is available at Carcinogenesis Online.

Figure 3. — Transcriptomic profiles regulated by BaP/TPA in comparison to control. (A) MA plots showing DEGs in response to control versus BaP/TPA with P_adj ≤ 0.05 and absolute log₂FC ≥ 2 as cutoffs. The DEGs with P_adj ≤ 0.05 and absolute log₂FC ≥ 2 were marked in red and the number of DEGs upregulated or downregulated were labeled; (B) Volcano plots presenting statistical significance (P_adj) versus magnitude of change (log₂FC) obtained from differential expression analysis. Genes with P_adj ≤ 0.05 or absolute log₂FC ≥ 2 were marked in blue or green and those that meet both cutoffs were highlighted with red. The top 10 gene with largest expression (log₂FC) changes along with the top 5 smallest P_adj were labeled.

Figure 4. — Skin cancer-associated oncogenes found from RNA-seq analysis. The mRNA levels were presented as transcript per million calculated from raw count normalized by gene length and sequencing depth in different groups (BaP/TPA versus Control) at 5, 20 and 26 weeks. Data are shown as mean ± SD of three independent experiments. Welch’s unpaired t-test was used to determine the significant differences between the two groups. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 show significant differences between the control and BaP/TPA groups.

BaP/TPA regulates signaling pathways in skin carcinogenesis

The DEGs that were regulated after exposure to BaP/TPA shown in Figure 3A and B were filtered using a threshold P ≤ 0.05 coupled to log₂FC ≥2.0 or ≤−2.0 from the RNAseq datasets to perform pathway analysis using QIAGEN Ingenuity Pathway Analysis. Briefly, BaP/TPA significantly (−logP < 2 or P < 0.05) regulated 17, 30 and 23 signaling pathways compared to their corresponding control mice at 5, 20 and 26 weeks, respectively (Supplementary File 2 is available at Carcinogenesis Online), suggesting that BaP/TPA regulated higher number of signaling pathways in late stages (20 and 26 weeks) of NMSC compared to early stage (5 weeks). The most significantly regulated (up- and down-regulated) signaling pathways were presented with their ‘activation z scores’ (Figure 5A and Supplementary File 2 is available at Carcinogenesis Online). During the initiation phase (5 weeks), the most significantly regulated oncogenic pathways are interleukin-17 (IL-17) signaling, focal adhesion kinase signaling, macrophage-stimulating protein-recepteur d’origine nantais (MSP-RON) signaling and high-mobility group box 1 (HMGB1) signaling pathways; while Superpathway of melatonin degradation, Melatonin degradation I and Sirtuin signaling pathways are regulated in 20 weeks mice. Sirtuin signaling and actin cytoskeleton signaling pathways are significantly downregulated in the late-stage counterparts. These results indicate the regulation of different signaling pathways at different stages of NMSC, which could be targeted by chemotherapeutic agents.

Figure 5. — Pathway analysis and association study between DEGs and DMRs. (A) Top-regulated signaling pathways by BaP/TPA at 5, 20 and 26 weeks; (B) Correlation study between DEGs and DMRs in control versus BaP/TPA groups at different time points. DEGs with P_adj ≤ 0.05 and DMRs with FDR ≤ 0.05 were plotted. Negatively correlated DMRs with their corresponding gene expression with the cutoff of methylation difference ≥10% or ≤−10% coupled to gene expression log₂FC ≥2.0 or ≤−2.0 were selected as shown in the upper left and bottom right corners. The DMRs that fall in the promoter region were labeled by their gene names. The locations of the DMRs (Region, Downstream, Body and Promoter, etc.) are designated by different colors. The gene body denotes gene regions including exon, intron, 3ʹ-UTR and 5ʹ-UTR. The downstream region covers the downstream flanking region and the distal intergenic region; (C) The expression of mRNA levels of oncogenes that were correlated with their corresponding promoter CpG methylation in control and BaP/TPA mice. Data are shown as mean ± SD of three independent experiments. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 show significant differences between the control and BaP/TPA groups. Welch’s unpaired t-test was used to determine the significance level between the two groups.

Inverse correlation between DNA methylation and RNA expression shows potential epigenetic targets

Next, we integrated methylome with transcriptome by arranging DMRs with corresponding gene expression in BaP/TPA versus control group. We acquired lists of DMRs with their corresponding gene expression in control versus BaP/TPA groups. As shown in the correlation starburst plot (Figure 5B), the DMRs in specific genes were illustrated as dots with different colors representing annotated features (gene regions). We especially emphasized on the DMRs with an inverse relationship between the CpG methylation ratio and gene expression changes, indicating hypomethylation coupled with transcription promotion and hypermethylation coupled with transcription suppression. The corresponding genes of these DMRs could be potential targets that can be epigenetically activated or inhibited to exert effect during skin cancer development. In BaP/TPA versus control comparisons, we labeled the inverse correlated DMRs, which are in the promoter with >10% DNA methylation level difference and log₂FC of 2 (Supplementary File 3 is available at Carcinogenesis Online). The genes with methylation changes opposite to transcriptional changes are Lgi2 (Leucine Rich Repeat LGI Family Member 2), KLK13 (Kallikrein-Related Peptidase 13) and Sox 5 (SRY-Box Transcription Factor 5) (Figure 5C).

BaP/TPA rewires metabolites/metabolic pathways and epigenetic-associated metabolites

To establish the underlying molecular associations between metabolic rewiring and skin cancer, the epidermis tissue samples collected from control and Bap/TPA mice were used to conduct LC–MS based metabolomic analysis. There was a clear separation of metabolic profile between control and exposure groups (Supplementary Figure 4 is available at Carcinogenesis Online). The results identified that BaP/TPA regulated a total of 237 metabolites (Supplementary File 4 is available at Carcinogenesis Online). The top 25 modulated metabolites between two groups (n = 3 in each group) are further shown in heatmaps (Figure 6A, Supplementary File 5 is available at Carcinogenesis Online). We next conducted pathway analysis together with pathway enrichment analysis in BaP/TPA versus control groups and found that exposure to BaP/TPA significantly alters the cancer-associated metabolisms, such as pyrimidine metabolism and amino acids metabolism (Figure 6B). The most significantly regulated metabolites from pyrimidine to amino acid metabolisms were further quantified. As shown in Figure 6C, thymine, thymidine, cytidine, deoxycytidine, 2-deoxyuradine and uracil from pyrimidine metabolism were significantly regulated at 5 and 20 weeks, indicating pyrimidine metabolism is associated with the initiation and progression of BaP-driven NMSC. At the late stage (26 weeks), the top-regulated metabolites were from amino acid metabolism such as aspartate, histidine, proline, glutamate and guanidineacetic acid, suggesting amino acid metabolism is linked to the progression of NMSC. Furthermore, we also found BaP/TPA regulates epigenetic-related metabolites such as S-adenosylmethionine (SAM), 5-methylcytosine (5-mC), methionine sulfoxide, nicotinamide and methionine (Figure 6D). These results indicate that exposure to BaP/TPA rewires the mitochondrial metabolism/metabolites and linking the down-stream impact on epigenetic reprogramming in NMSC.

Figure 6. — Metabolic alterations in the epidermis samples of control and BaP/TPA-treated mice. (A) Top 25 modulated metabolites between the control and BaP/TPA mice (n = 3); (B) Metabolic pathway analysis between the control and BaP/TPA groups. The color codes indicate different levels of significance with yellow to red increased level of significance. Only the statistically significant pathways were marked in the figures. (C) Quantification of key metabolites from pyrimidine and amino acid metabolism in control and BaP/TPA-exposed mice. Data are shown as mean ± SD of three independent experiments. *P < 0.05 and **P < 0.01 show significant differences between the control and BaP/TPA groups. Unpaired t-test was used to determine the significance level between two groups. (D) Quantification of epigenetic-related metabolites in control and BaP/TPA-treated mice at 5, 20 and 26 weeks. Data are shown as mean ± SD of three independent experiments. *P < 0.05 and **P < 0.01 show significant differences between control and BaP/TPA groups. Unpaired t-test was used to determine the significance level between the two groups.

Discussion

Skin cancer, including malignant melanoma and NMSC, is the leading type of cancer in the world. Environmental exposure to polycyclic hydrocarbons, including BaP is a major cause of NMSC. It is important to understand how the carcinogen BaP regulates epigenetic DNA methylation, gene expression and metabolites/metabolic pathways in cancer initiation and progression.

Aberrant DNA methylation is closely connected to the event of oncogenesis (32). This study provides new understandings into the time course alteration of genome-wide DNA methylation and gene expression in response to the carcinogen BaP exposure. We identified several genes with altered CpG methylation, which could play important roles at different stages of skin carcinogenesis. For instance, hepatic leukemia factor (HLF) is an oncogenic transcript factor reported in various malignancies, including skin cancer, and promotes resistance to cell death in mouse epidermal cells and human keratinocytes (33). BCL6 corepressor-like 1 (BCORL1) is a transcriptional corepressor that has been reported as a prognostic factor in myeloid leukemia (34) and hepatocellular carcinoma (35). Protein tyrosine phosphatase non-receptor 13 (PTPN13) is a class I non-receptor protein tyrosine phosphatase, which has a dual role in cancer development and progression (36–38). Hematopoietic SH2 domain containing (HSH2D) has been reported as a differentially expressed gene in cutaneous melanoma (39). Spastic paraplegia-20 (Spg20) promoter hypermethylation has been identified as a biomarker for the detection of colorectal cancer (40). Adenomatosis polyposis coli down-regulated 1 (APCDD1) is a Wnt antagonist, which plays in role in skin development (41), and its promoter hypermethylation is linked to invasion and metastasis of osteosarcoma (42). Coagulation factor 2 thrombin receptor (F2R), also known as proteinase-activated receptor 1 (PAR1), is an essential molecule in thrombosis and has been reported to contribute to the metastatic phenotype of melanoma (43). Growth arrest-specific 2 like 1 (GAS2L1) is overexpressed in pancreatic cancer (44), and one of its family member gene GAS1 has been demonstrated to suppress melanoma metastasis (45). T-cell immune regulator 1 (TCIRG1) is a prognostic marker in diverse cancers, including hepatocellular carcinoma and glioblastoma multiforme (46,47). Neural precursor cell expressed, developmentally down-regulated 9 (Nedd9) has been reported as a melanoma metastasis gene (48). Furthermore, the RNA-seq data analysis shows significantly altered gene expression profile in response to BaP exposure. Mitochondrial calcium uptake 1 (MICU1) serves as a key component in regulating metabolic aberrations to inhibit tumor growth and overcome chemoresistance (49). Serine protease inhibitor Kazal type 7 (SPINK7) is expressed in human skin and plays a vital function in skin homeostasis and inflammatory skin diseases (50). Keratins are the intermediate filament-forming proteins of epithelial cells which are commonly used as prognostic biomarkers in cancers. The aberrant expression of keratins has been demonstrated in melanoma (51).

Correlation analysis between methylation level change of DMRs with their corresponding alteration in gene expression at different time points shows that the mRNA expression of oncogenes Klk13, Sox5 and Lgi2 were correlated with the promoter CpG methylation status. Klk genes are considered tumor-specific markers for diagnosing human malignancies. Aberrant expression of Klk genes is associated with the abrasiveness of human melanoma by inducing cell migration and invasion (52). High mRNA expression of Klk13 and Klk6 is a potential indicator of poor prognosis and shorter recurrence-free survival in ovarian cancer patients (53). In the current study, we showed that gene expression of Klk13 was upregulated coupled with hypomethylation in the promoter as early as 5 weeks, which continued to 26 weeks. Sox5, one of the Sox family genes, comprises 20 Sox genes which are categorized into A–H. Sox5 falls into the Sox D group of genes. Mounting evidence shows that Sox5 augments the development of various tumors, including prostate, hepatocellular, nasopharyngeal, breast and pituitary tumor cancer (54). It acts as a regulator of the microphthalmia-associated transcription factor in melanoma cells (55). We also observed upregulation of the Lgi2 gene coupled with hypomethylation in the promoter from the early-stage initiation to late-stage progression in SKH1 mice compared to the control mice. Lgi2 is a homolog gene of the well-studied Lgi1 gene (56).

In general, the challenge to a carcinogen can trigger the activation of several oncogenes/pathways and inhibition of tumor suppressor genes/pathways. Our Ingenuity Pathway Analysis shows up- and down-regulation of different signaling pathways in skin carcinogenesis. MSP-RON pathway significantly regulates the tissue microenvironment, specifically in the tumor immune microenvironment. A previous study reported that the knockout of RON receptors in mice inhibits metastatic breast cancer by increasing the number of cytotoxic CD8-positive T cells (57). Activation of RON leads to upregulation of Fos by stimulating MAPK, which in turn binds to the promoter region of the Arg1 gene to increase Arg1 expression. The dysregulation of RON-MSP signaling characterized in our two-stage skin carcinogenesis is also comparable to previous studies showing RON-MSP signaling promotes tumorigenesis of breast, colon, ovarian, lung and epithelial cancers in in vitro, in vivo and human cancer specimens (58). HMGB1 is a non-histone nuclear protein that plays a significant role in inflammation and tumorigenesis. Previous study shows that HMGB1 derived from keratinocytes interrupts the wound healing of the skin and promotes tumor formation regulating neutrophil extracellular trap formation (59). Increased expression of cutaneous HMGB1 level enhances wound-induced skin cancer (60). Focal adhesion kinase is one of the essential drivers of integrin- and growth factor receptor-mediated signals, regulating critical cellular functions in normal and cancer cells by controlling their kinase activity. Higher expression and activity of focal adhesion kinase are found in diverse localized and metastasized cancers and are linked to poor patient outcomes (61). Sirtuins are a group of signaling proteins that regulate various critical cellular pathways, such as senescence, DNA repair, metabolism, transcriptional regulation and aging. However, a previous study reported the dual function of Sirtuins and their signaling pathways in tumor suppression and promotion depending on their tissue-specific expression (62). Melatonin is an essential skin protective agent due to its free radical scavenging and DNA damage-repairing abilities. It also possesses antiproliferative, antitumor and oncostatic properties in melanoma cells (63). In the current study, we observed inactivation/inhibition of the superpathway of melatonin degradation and the melatonin degradation pathway I are associated with the progression or development of late-stage NMSC. Furthermore, epithelial-mesenchymal transition is related to loss/disruption of the cell–cell adhesion molecule E-cadherin and cell–cell junctions as well as with the gaining of migratory behaviors, including reorganization of the actin cytoskeleton. During the process of epithelial-mesenchymal transition, cells gradually downregulate their basolateral and apical epithelial-specific proteins, such as cytokeratins, catenins and E-cadherin, and re-express several mesenchymal components, such as N-cadherin, fibronectin and vimentin (64).

Metabolic reprogramming is a characteristic event within the tumor cells to support cell proliferation and survival (65). Numerous studies have reported that sugar, lipid and amino acid metabolism eventually impacts tumor growth regulating nucleotide metabolism (15). In the present study, we observed that BaP/TPA significantly regulates nucleotide metabolism, e.g. pyrimidine metabolism, with the highest impact score during the early initiation and late promotion stages of NMSC. Pyrimidine nucleotide synthesis is considered the pathway of choice to target tumors because pyrimidine nucleotides are the essential building block of DNA synthesis in cells and are progressively needed by cancer cells due to their rapid growth. 5-Fluorouracil is the most common pyrimidine analog, which is extensively used in cancer treatment. We also observed significant regulation of amino acid metabolism, e.g. histidine, arginine and proline, alanine, aspartate and glutamate metabolism, and phenylalanine, tyrosine and tryptophan biosynthesis during different phases of NMSC. Our findings are in line, at least partly, with the previously published paper reporting the role of alanine, aspartate and aspartate metabolism, and arginine metabolism in human melanoma cells (66). Alanine is increasingly used for protein synthesis during cell proliferation and is secreted as a glycolytic byproduct to carry extra carbon from glycolysis (67). An earlier study reported that reprogramming of proline metabolism contributes to the proliferative and metabolic responses regulated by oncogenic transcription factor c-MYC (68). The histidine catabolism/degradation pathway significantly affects the sensitivity of tumor cells to methotrexate. It could be exploited to improve methotrexate efficacy through a simple dietary intervention/starvation in NMSC therapy (69). Tyrosine, phenylalanine and tryptophan are aromatic amino acids, while the increased concentration is linked to gastroesophageal cancer (70). In addition, tyrosine and phenylalanine were found to be the main precursors of melanin in melanoma (71). Considering the interconnection between metabolic rewiring and epigenetic changes, we measured the levels of epigenetic-regulated metabolites in control and BaP/TPA groups. We found several vital metabolites, including SAM, methionine, methionine sulfoxide, nicotinamide and 5-mC, associated with the epigenetic regulation of gene expression, were significantly modulated in BaP/TPA groups compared to the controls. SAM acts as a substrate for methylation to preserve the epigenetic status, including DNA methylation, histone methylation, RNA methylation and post-translational modifications such as the methylation of arginine and lysine residues of non-histone proteins (72). 5-mC is a pyrimidine that is a cytosine analog, containing a methyl group at the 5-position. It also acts as a human metabolite. 5-mC plays a critical role in methylation-regulated gene expression in cancers (73).

In conclusion, this study provides unique insight into DNA methylation, gene expression and metabolic profiles at different stages of skin carcinogenesis. We found that modulation of MSP-RON and HMGB1 signaling pathways during the initiation phase; superpathway of melatonin degradation, melatonin degradation 1 and sirtuin signaling pathway during the promotion phase; and sirtuins signaling and actin cytoskeleton signaling pathways during tumor progression phase potentially contribute to the development of skin carcinogenesis. The metabolomic study showed BaP/TPA regulated cancer-associated metabolisms like pyrimidine and amino acid metabolism/metabolites as well as several epigenetic-related metabolites, such as methionine, SAM and 5-mC indicating their critical role in carcinogen-driven skin cancer. These findings could benefit future skin cancer interception and treatment research targeting the epigenetic- and metabolism-regulated metabolites/genes and pathways. However, further study is needed in the regulation of metabolic and epigenetic pathways to understand the sex differences between males and females.

Supplementary material

Supplementary data are available at Carcinogenesis online.

Supplementary Figure 1. Pilot study to develop BaP/TPA induced two-stage skin carcinogenesis. (A) SKH-1 mice (n = 3) were treated with or without three doses of TPA 0.65, 1.3 or 2.6 nmol/20 μl. Mice were euthanized 6 h after treatment. Ear punches were collected, and average edema weight was measured. #P < 0.00001 and ***P < 0.0001 shows significant differences compared to control group. (B) H&E staining of whole skin samples of animals at 26 weeks.

Supplementary Figure 2. DNA methylation profile in control and BaP/TPA treated mice at 5, 20 and 26 weeks. (A) Distribution pattern the of annotated DMRs such as promoter (<=1, 1–2 and 2–3 kb), distal intergenic, intron and other regions. Individual DMR includes at least three CpG sites; (B) Distribution pattern of the DMRs based on number of CpG sites and region; (C) Average methylation levels of DMRs based on gene regions for samples in control and BaP/TPA mice at 5, 20 and 26 weeks. (D) PCA of CpG methylation in control and BaP/TPA groups at 5, 20 and 26 weeks. There was a clear separation between control mice and BaP/TPA groups for each time point.

Supplementary Figure 3. (A) RNA-seq analysis revealed the distribution of DEGs by number of genes and normalized annotated data for the control and BaP/TPA groups at 5, 20 and 26 weeks. (B) PCA of transcriptomic profiles in control versus BaP/TPA mice at 5, 20 and 26 weeks. (C) Dendrogram of the gene expression profiles clustered by Euclidean distance of control and Bap/TPA groups at 5, 20 and 26 weeks.

Supplementary Figure 4. PCA of metabolic profiles in control versus BaP/TPA mice at 5, 20 and 26 weeks.

bgad024_suppl_Supplementary_Data_S1

Click here for additional data file.^{(111.2KB, xlsx)}

bgad024_suppl_Supplementary_Data_S2

Click here for additional data file.^{(12.1KB, xlsx)}

bgad024_suppl_Supplementary_Data_S3

Click here for additional data file.^{(24.5KB, xlsx)}

bgad024_suppl_Supplementary_Data_S4

Click here for additional data file.^{(55.6KB, xlsx)}

bgad024_suppl_Supplementary_Data_S5

Click here for additional data file.^{(17.4KB, xlsx)}

bgad024_suppl_Supplementary_Figures

Click here for additional data file.^{(1.6MB, pptx)}

Acknowledgements

This paper is dedicated to Dr Yaoping Lu, who passed away, for his contribution to the experimental design of this study. We thank all the members of Professor Ah-Ng Kong’s laboratory for their invaluable discussion and technical support in the preparation of this manuscript.

Conflict of Interest Statement: None declared.

Abbreviations

BaP: benzo[a]pyrene

FDR: false discovery rate

LC–MS: liquid chromatography–mass spectrometry
NMSC: non-melanoma skin cancer

Contributor Information

Md. Shahid Sarwar, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Christina N Ramirez, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Hsiao-Chen Dina Kuo, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Pochung Chou, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Renyi Wu, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Davit Sargsyan, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Yuqing Yang, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Ahmad Shannar, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Rebecca Mary Peter, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Ran Yin, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Yujue Wang, Metabolomics Shared Resource, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ 08901, USA.

Xiaoyang Su, Metabolomics Shared Resource, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ 08901, USA.

Ah-Ng Kong, Department of Pharmaceutics, Ernest Mario School of Pharmacy, Rutgers, The State University of New Jersey, Piscataway, NJ 08854, USA.

Funding

National Cancer Institute (R01 CA200129 to A.N.K.).

Authors’ contributions

Concept and design: C.N.R., R.W., A.N.K.; Experiments: M.S.S., C.N.R., H.C.D.K., R.W., Y.Y.; Data analysis and interpretation: P.C., M.S.S., H.C.D.K., D.S., A.S., R.M.P., R.Y., Y.W., X.S.; Drafting of the paper: M.S.S., H.C.D.K.; Supervision and funding acquisition: A.N.K.

Data availability

All data described in this study are available in the results, figures and supplementary files.

References

1. Shalhout, S.Z., et al. (2022) Immunotherapy for nonmelanoma skin cancer: facts and hopes. Clin. Cancer Res., 28, 2211–2220. [DOI] [PubMed] [Google Scholar]
2. Ciazynska, M., et al. (2021) The incidence and clinical analysis of non-melanoma skin cancer. Sci. Rep., 11, 4337. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. Kasper, M., et al. (2012) Basal cell carcinoma - molecular biology and potential new therapies. J. Clin. Invest., 122, 455–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Vermillion Maier, M.L., et al. (2022) Benzo[a]pyrene (BaP) metabolites predominant in human plasma following escalating oral micro-dosing with [(14)C]-BaP. Environ. Int., 159, 107045. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Siddens, L.K., et al. (2012) Polycyclic aromatic hydrocarbons as skin carcinogens: comparison of benzo [a]pyrene, dibenzo[def,p]chrysene and three environmental mixtures in the FVB/N mouse. Toxicol. Appl. Pharmacol., 264, 377–386. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Abel, E.L., et al. (2009) Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications. Nat. Protoc., 4, 1350–1362. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Courter, L.A., et al. (2008) The influence of diesel exhaust on polycyclic aromatic hydrocarbon-induced DNA damage, gene expression, and tumor initiation in Sencar mice in vivo. Cancer Lett., 265, 135–147. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Widyarini, S., et al. (2005) Protective effect of the isoflavonoid equol against hairless mouse skin carcinogenesis induced by UV radiation alone or with a chemical cocarcinogen. Photochem. Photobiol., 81, 32–37. [DOI] [PubMed] [Google Scholar]
9. Melnikova, V.O., et al. (2005) Cellular and molecular events leading to the development of skin cancer. Mutat. Res., 571, 91–106. [DOI] [PubMed] [Google Scholar]
10. Hudlikar, R.R., et al. (2022) Tobacco carcinogen 4-[methyl(nitroso)amino]-1-(3-pyridinyl)-1-butanone (NNK) drives metabolic rewiring and epigenetic reprogramming in A/J mice lung cancer model and prevention with diallyl sulphide (DAS). Carcinogenesis, 43, 140–149. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Li, S.Y., et al. (2022) UVB drives metabolic rewiring and epigenetic reprogramming and protection by sulforaphane in human skin keratinocytes. Chem. Res. Toxicol, 35, 1220–1233. [DOI] [PubMed] [Google Scholar]
12. Janke, R., et al. (2015) Metabolism and epigenetics. Annu. Rev. Cell Dev. Biol., 31, 473–496. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. Kinnaird, A., et al. (2016) Metabolic control of epigenetics in cancer. Nat. Rev. Cancer, 16, 694–707. [DOI] [PubMed] [Google Scholar]
14. Suganuma, T., et al. (2021) Nucleotide metabolism behind epigenetics. Front. Endocrinol. (Lausanne), 12, 731648. [DOI] [PMC free article] [PubMed] [Google Scholar]
15. Wang, W., et al. (2021) Targeting pyrimidine metabolism in the era of precision cancer medicine. Front. Oncol., 11, 684961. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Yang, Y., et al. (2019) UVB drives different stages of epigenome alterations during progression of skin cancer. Cancer Lett., 449, 20–30. [DOI] [PMC free article] [PubMed] [Google Scholar]
17. Yang, Y.Q., et al. (2019) DNA methylome and transcriptome alterations and cancer prevention by triterpenoid ursolic acid in UVB-induced skin tumor in mice. Mol. Carcinog., 58, 1738–1753. [DOI] [PMC free article] [PubMed] [Google Scholar]
18. Wang, Y., et al. (2005) Combined subcarcinogenic benzo[a]pyrene and UVA synergistically caused high tumor incidence and mutations in H-ras gene, but not p53, in SKH-1 hairless mouse skin. Int. J. Cancer, 116, 193–199. [DOI] [PubMed] [Google Scholar]
19. Kiraly, A.J., et al. (2016) Apigenin inhibits COX-2, PGE2, and EP1 and also initiates terminal differentiation in the epidermis of tumor bearing mice. Prostaglandins Leukot. Essent. Fatty Acids, 104, 44–53. [DOI] [PubMed] [Google Scholar]
20. Digiovanni, J., et al. (1980) Comparison of the tumor-initiating activity of 7,12-dimethylbenz[a]anthracene and benzo[a]pyrene in Female Sencar and Cd-1 mice. Carcinogenesis, 1, 381–389. [DOI] [PubMed] [Google Scholar]
21. Chan, K.S., et al. (2008) Forced expression of a constitutively active form of Stat3 in mouse epidermis enhances malignant progression of skin tumors induced by two-stage carcinogenesis. Oncogene, 27, 1087–1094. [DOI] [PubMed] [Google Scholar]
22. Steinel, H.H., et al. (1990) Effects of beta-carotene on chemically-induced skin tumors in Hra/Skh hairless mice. Cancer Lett., 51, 163–168. [DOI] [PubMed] [Google Scholar]
23. Martin, M. (2011) Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J., 17, 10–12. [Google Scholar]
24. Kim, D., et al. (2015) HISAT: a fast spliced aligner with low memory requirements. Nat. Methods, 12, 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Li, H., et al.; 1000 Genome Project Data Processing Subgroup. (2009) The sequence alignment/map format and SAMtools. Bioinformatics, 25, 2078–2079. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Liao, Y., et al. (2013) featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics, 30, 923–930. [DOI] [PubMed] [Google Scholar]
27. Krueger, F., et al. (2011) Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics, 27, 1571–1572. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Gaspar, J.M., et al. (2017) DMRfinder: efficiently identifying differentially methylated regions from MethylC-seq data. BMC Bioinf., 18, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
29. Yu, G.C., et al. (2015) ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics, 31, 2382–2383. [DOI] [PubMed] [Google Scholar]
30. Anastasiadi, D., et al. (2018) Consistent inverse correlation between DNA methylation of the first intron and gene expression across tissues and species. Epigenet. Chromatin., 11, 37. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Wang, L., et al. (2022) PTEN-knockout regulates metabolic rewiring and epigenetic reprogramming in prostate cancer and chemoprevention by triterpenoid ursolic acid. FASEB J., 36, e22626. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Saghafinia, S., et al. (2018) Pan-cancer landscape of aberrant DNA methylation across human tumors. Cell Rep., 25, 1066–1080.e8. [DOI] [PubMed] [Google Scholar]
33. Waters, K.M., et al. (2013) Hepatic leukemia factor promotes resistance to cell death: implications for therapeutics and chronotherapy. Toxicol. Appl. Pharmacol., 268, 141–148. [DOI] [PubMed] [Google Scholar]
34. Schaefer, E.J., et al. (2022) BCOR and BCORL1 mutations drive epigenetic reprogramming and oncogenic signaling by unlinking PRC1.1 from target genes. Blood Cancer Discov., 3, 116–135. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Yin, G.Z., et al. (2016) BCORL1 is an independent prognostic marker and contributes to cell migration and invasion in human hepatocellular carcinoma. BMC Cancer, 16, 103. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]
36. Dromard, M., et al. (2007) The putative tumor suppressor gene PTPN13/PTPL1 induces apoptosis through insulin receptor substrate-1 dephosphorylation. Cancer Res., 67, 6806–6813. [DOI] [PubMed] [Google Scholar]
37. Niu, J., et al. (2009) Genetic polymorphisms in the PTPN13 gene and risk of squamous cell carcinoma of head and neck. Carcinogenesis, 30, 2053–2058. [DOI] [PMC free article] [PubMed] [Google Scholar]
38. McHeik, S., et al. (2020) Dual role of the PTPN13 tyrosine phosphatase in cancer. Biomolecules, 10, 1659. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Ji, C., et al. (2020) Identification of four genes associated with cutaneous metastatic melanoma. Open Med. (Wars), 15, 531–539. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Lind, G.E., et al. (2011) SPG20, a novel biomarker for early detection of colorectal cancer, encodes a regulator of cytokinesis. Oncogene, 30, 3967–3978. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Vonica, A., et al. (2020) Apcdd1 is a dual BMP/Wnt inhibitor in the developing nervous system and skin. Dev. Biol., 464, 7172–7187. [DOI] [PMC free article] [PubMed] [Google Scholar]
42. Han, W., et al. (2017) Epigenetic silencing of the Wnt antagonist APCDD1 by promoter DNA hyper-methylation contributes to osteosarcoma cell invasion and metastasis. Biochem. Biophys. Res. Commun., 491, 91–97. [DOI] [PubMed] [Google Scholar]
43. Tellez, C., et al. (2003) Role and regulation of the thrombin receptor (PAR-1) in human melanoma. Oncogene, 22, 3130–3137. [DOI] [PubMed] [Google Scholar]
44. Zhu, L., et al. (2020) GAS2L1 is a potential biomarker of circulating tumor cells in pancreatic cancer. Cancers, 12, 3774. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. Gobeil, S., et al. (2008) A genome-wide shRNA screen identifies GAS1 as a novel melanoma metastasis suppressor gene. Genes Dev., 22, 2932–2940. [DOI] [PMC free article] [PubMed] [Google Scholar]
46. Yang, H.D., et al. (2018) T-cell immune regulator 1 enhances metastasis in hepatocellular carcinoma. Exp. Mol. Med., 50, e420. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Qi, C.X., et al. (2021) T cell immune regulator 1 is a prognostic marker associated with immune infiltration in glioblastoma multiforme. Oncol. Lett., 21, 252. [DOI] [PMC free article] [PubMed] [Google Scholar]
48. Kim, M., et al. (2006) Comparative oncogenomics identifies NEDD9 as a melanoma metastasis gene. Cell, 125, 1269–1281. [DOI] [PubMed] [Google Scholar]
49. Chakraborty, P.K., et al. (2017) MICU1 drives glycolysis and chemoresistance in ovarian cancer. Nat. Commun., 8, 14634. [DOI] [PMC free article] [PubMed] [Google Scholar]
50. Weber, C., et al. (2017) The serine protease inhibitor of Kazal-type 7 (SPINK7) is expressed in human skin. Arch. Dermatol. Res., 309, 767–771. [DOI] [PubMed] [Google Scholar]
51. Han, W., et al. (2021) Transcript levels of keratin 1/5/6/14/15/16/17 as potential prognostic indicators in melanoma patients. Sci. Rep., 11, 1023. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Delaunay, T., et al. (2017) Aberrant expression of kallikrein-related peptidase 7 is correlated with human melanoma aggressiveness by stimulating cell migration and invasion. Mol. Oncol., 11, 1330–1347. [DOI] [PMC free article] [PubMed] [Google Scholar]
53. White, N.M., et al. (2009) KLK6 and KLK13 predict tumor recurrence in epithelial ovarian carcinoma. Br. J. Cancer, 101, 1107–1113. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Hu, J., et al. (2018) Sox5 contributes to prostate cancer metastasis and is a master regulator of TGF-beta-induced epithelial mesenchymal transition through controlling Twist1 expression. Br. J. Cancer, 118, 88–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
55. Kordass, T., et al. (2016) SOX5 is involved in balanced MITF regulation in human melanoma cells. BMC Med. Genom., 9, 10. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. Gu, W., et al. (2002) The LGI1 gene involved in lateral temporal lobe epilepsy belongs to a new subfamily of leucine-rich repeat proteins. FEBS Lett., 519, 71–76. [DOI] [PubMed] [Google Scholar]
57. Eyob, H., et al. (2013) Inhibition of ron kinase blocks conversion of micrometastases to overt metastases by boosting antitumor immunity. Cancer Discov., 3, 751–760. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Huang, L., et al. (2020) MSP-RON pathway: potential regulator of inflammation and innate immunity. Front. Immunol., 11, 569082. [DOI] [PMC free article] [PubMed] [Google Scholar]
59. Hoste, E., et al. (2019) Epithelial HMGB1 delays skin wound healing and drives tumor initiation by priming neutrophils for NET formation. Cell Rep., 29, 2689–2701.e4. [DOI] [PubMed] [Google Scholar]
60. Hoste, E., et al. (2015) Innate sensing of microbial products promotes wound-induced skin cancer. Nat. Commun., 6, 5932. [DOI] [PMC free article] [PubMed] [Google Scholar]
61. Lee, B.Y., et al. (2015) FAK signaling in human cancer as a target for therapeutics. Pharmacol. Ther., 146, 132–149. [DOI] [PubMed] [Google Scholar]
62. Carafa, V., et al. (2019) Dual tumor suppressor and tumor promoter action of sirtuins in determining malignant phenotype. Front. Pharmacol., 10, 38. [DOI] [PMC free article] [PubMed] [Google Scholar]
63. Alvarez-Artime, A., et al. (2020) Melatonin-induced cytoskeleton reorganization leads to inhibition of melanoma cancer cell proliferation. Int. J. Mol. Sci., 21, 548. [DOI] [PMC free article] [PubMed] [Google Scholar]
64. Wicki, A., et al. (2006) Tumor invasion in the absence of epithelial-mesenchymal transition: podoplanin-mediated remodeling of the actin cytoskeleton. Cancer Cell, 9, 261–272. [DOI] [PubMed] [Google Scholar]
65. Ward, P.S., et al. (2012) Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell, 21, 297–308. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Kim, H.Y., et al. (2017) Discovery of potential biomarkers in human melanoma cells with different metastatic potential by metabolic and lipidomic profiling. Sci. Rep., 7, 8864. [DOI] [PMC free article] [PubMed] [Google Scholar]
67. DeBerardinis, R.J., et al. (2010) Q’s next: the diverse functions of glutamine in metabolism, cell biology and cancer. Oncogene, 29, 313–324. [DOI] [PMC free article] [PubMed] [Google Scholar]
68. Liu, W., et al. (2012) Reprogramming of proline and glutamine metabolism contributes to the proliferative and metabolic responses regulated by oncogenic transcription factor c-MYC. Proc. Natl. Acad. Sci. U. S. A., 109, 8983–8988. [DOI] [PMC free article] [PubMed] [Google Scholar]
69. Kanarek, N., et al. (2018) Histidine catabolism is a major determinant of methotrexate sensitivity. Nature, 559, 632–636. [DOI] [PMC free article] [PubMed] [Google Scholar]
70. Wiggins, T., et al. (2015) Tyrosine, phenylalanine, and tryptophan in gastroesophageal malignancy: a systematic review. Cancer Epidemiol. Biomarkers Prev., 24, 32–38. [DOI] [PubMed] [Google Scholar]
71. Blagoeva, P.M. (1984) Aminoacids--precursors of melanin synthesis in hamster melanoma. J. Cancer Res. Clin. Oncol., 108, 366–368. [DOI] [PMC free article] [PubMed] [Google Scholar]
72. Huo, M.M., et al. (2021) Interplay among metabolism, epigenetic modifications, and gene expression in cancer. Front. Cell Dev. Biol., 9, 793428. [DOI] [PMC free article] [PubMed] [Google Scholar]
73. Du, C., et al. (2022) Analysis of 5-methylcytosine regulators and DNA methylation-driven genes in colon cancer. Front. Cell Dev. Biol., 9, 657092. [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

bgad024_suppl_Supplementary_Data_S1

Click here for additional data file.^{(111.2KB, xlsx)}

bgad024_suppl_Supplementary_Data_S2

Click here for additional data file.^{(12.1KB, xlsx)}

bgad024_suppl_Supplementary_Data_S3

Click here for additional data file.^{(24.5KB, xlsx)}

bgad024_suppl_Supplementary_Data_S4

Click here for additional data file.^{(55.6KB, xlsx)}

bgad024_suppl_Supplementary_Data_S5

Click here for additional data file.^{(17.4KB, xlsx)}

bgad024_suppl_Supplementary_Figures

Click here for additional data file.^{(1.6MB, pptx)}

Data Availability Statement

All data described in this study are available in the results, figures and supplementary files.

[CIT0001] 1. Shalhout, S.Z., et al. (2022) Immunotherapy for nonmelanoma skin cancer: facts and hopes. Clin. Cancer Res., 28, 2211–2220. [DOI] [PubMed] [Google Scholar]

[CIT0002] 2. Ciazynska, M., et al. (2021) The incidence and clinical analysis of non-melanoma skin cancer. Sci. Rep., 11, 4337. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0003] 3. Kasper, M., et al. (2012) Basal cell carcinoma - molecular biology and potential new therapies. J. Clin. Invest., 122, 455–463. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0004] 4. Vermillion Maier, M.L., et al. (2022) Benzo[a]pyrene (BaP) metabolites predominant in human plasma following escalating oral micro-dosing with [(14)C]-BaP. Environ. Int., 159, 107045. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0005] 5. Siddens, L.K., et al. (2012) Polycyclic aromatic hydrocarbons as skin carcinogens: comparison of benzo [a]pyrene, dibenzo[def,p]chrysene and three environmental mixtures in the FVB/N mouse. Toxicol. Appl. Pharmacol., 264, 377–386. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0006] 6. Abel, E.L., et al. (2009) Multi-stage chemical carcinogenesis in mouse skin: fundamentals and applications. Nat. Protoc., 4, 1350–1362. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0007] 7. Courter, L.A., et al. (2008) The influence of diesel exhaust on polycyclic aromatic hydrocarbon-induced DNA damage, gene expression, and tumor initiation in Sencar mice in vivo. Cancer Lett., 265, 135–147. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0008] 8. Widyarini, S., et al. (2005) Protective effect of the isoflavonoid equol against hairless mouse skin carcinogenesis induced by UV radiation alone or with a chemical cocarcinogen. Photochem. Photobiol., 81, 32–37. [DOI] [PubMed] [Google Scholar]

[CIT0009] 9. Melnikova, V.O., et al. (2005) Cellular and molecular events leading to the development of skin cancer. Mutat. Res., 571, 91–106. [DOI] [PubMed] [Google Scholar]

[CIT0010] 10. Hudlikar, R.R., et al. (2022) Tobacco carcinogen 4-[methyl(nitroso)amino]-1-(3-pyridinyl)-1-butanone (NNK) drives metabolic rewiring and epigenetic reprogramming in A/J mice lung cancer model and prevention with diallyl sulphide (DAS). Carcinogenesis, 43, 140–149. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0011] 11. Li, S.Y., et al. (2022) UVB drives metabolic rewiring and epigenetic reprogramming and protection by sulforaphane in human skin keratinocytes. Chem. Res. Toxicol, 35, 1220–1233. [DOI] [PubMed] [Google Scholar]

[CIT0012] 12. Janke, R., et al. (2015) Metabolism and epigenetics. Annu. Rev. Cell Dev. Biol., 31, 473–496. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0013] 13. Kinnaird, A., et al. (2016) Metabolic control of epigenetics in cancer. Nat. Rev. Cancer, 16, 694–707. [DOI] [PubMed] [Google Scholar]

[CIT0014] 14. Suganuma, T., et al. (2021) Nucleotide metabolism behind epigenetics. Front. Endocrinol. (Lausanne), 12, 731648. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0015] 15. Wang, W., et al. (2021) Targeting pyrimidine metabolism in the era of precision cancer medicine. Front. Oncol., 11, 684961. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0016] 16. Yang, Y., et al. (2019) UVB drives different stages of epigenome alterations during progression of skin cancer. Cancer Lett., 449, 20–30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0017] 17. Yang, Y.Q., et al. (2019) DNA methylome and transcriptome alterations and cancer prevention by triterpenoid ursolic acid in UVB-induced skin tumor in mice. Mol. Carcinog., 58, 1738–1753. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0018] 18. Wang, Y., et al. (2005) Combined subcarcinogenic benzo[a]pyrene and UVA synergistically caused high tumor incidence and mutations in H-ras gene, but not p53, in SKH-1 hairless mouse skin. Int. J. Cancer, 116, 193–199. [DOI] [PubMed] [Google Scholar]

[CIT0019] 19. Kiraly, A.J., et al. (2016) Apigenin inhibits COX-2, PGE2, and EP1 and also initiates terminal differentiation in the epidermis of tumor bearing mice. Prostaglandins Leukot. Essent. Fatty Acids, 104, 44–53. [DOI] [PubMed] [Google Scholar]

[CIT0020] 20. Digiovanni, J., et al. (1980) Comparison of the tumor-initiating activity of 7,12-dimethylbenz[a]anthracene and benzo[a]pyrene in Female Sencar and Cd-1 mice. Carcinogenesis, 1, 381–389. [DOI] [PubMed] [Google Scholar]

[CIT0021] 21. Chan, K.S., et al. (2008) Forced expression of a constitutively active form of Stat3 in mouse epidermis enhances malignant progression of skin tumors induced by two-stage carcinogenesis. Oncogene, 27, 1087–1094. [DOI] [PubMed] [Google Scholar]

[CIT0022] 22. Steinel, H.H., et al. (1990) Effects of beta-carotene on chemically-induced skin tumors in Hra/Skh hairless mice. Cancer Lett., 51, 163–168. [DOI] [PubMed] [Google Scholar]

[CIT0023] 23. Martin, M. (2011) Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J., 17, 10–12. [Google Scholar]

[CIT0024] 24. Kim, D., et al. (2015) HISAT: a fast spliced aligner with low memory requirements. Nat. Methods, 12, 357–360. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0025] 25. Li, H., et al.; 1000 Genome Project Data Processing Subgroup. (2009) The sequence alignment/map format and SAMtools. Bioinformatics, 25, 2078–2079. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0026] 26. Liao, Y., et al. (2013) featureCounts: an efficient general purpose program for assigning sequence reads to genomic features. Bioinformatics, 30, 923–930. [DOI] [PubMed] [Google Scholar]

[CIT0027] 27. Krueger, F., et al. (2011) Bismark: a flexible aligner and methylation caller for Bisulfite-Seq applications. Bioinformatics, 27, 1571–1572. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0028] 28. Gaspar, J.M., et al. (2017) DMRfinder: efficiently identifying differentially methylated regions from MethylC-seq data. BMC Bioinf., 18, 1–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0029] 29. Yu, G.C., et al. (2015) ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics, 31, 2382–2383. [DOI] [PubMed] [Google Scholar]

[CIT0030] 30. Anastasiadi, D., et al. (2018) Consistent inverse correlation between DNA methylation of the first intron and gene expression across tissues and species. Epigenet. Chromatin., 11, 37. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0031] 31. Wang, L., et al. (2022) PTEN-knockout regulates metabolic rewiring and epigenetic reprogramming in prostate cancer and chemoprevention by triterpenoid ursolic acid. FASEB J., 36, e22626. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0032] 32. Saghafinia, S., et al. (2018) Pan-cancer landscape of aberrant DNA methylation across human tumors. Cell Rep., 25, 1066–1080.e8. [DOI] [PubMed] [Google Scholar]

[CIT0033] 33. Waters, K.M., et al. (2013) Hepatic leukemia factor promotes resistance to cell death: implications for therapeutics and chronotherapy. Toxicol. Appl. Pharmacol., 268, 141–148. [DOI] [PubMed] [Google Scholar]

[CIT0034] 34. Schaefer, E.J., et al. (2022) BCOR and BCORL1 mutations drive epigenetic reprogramming and oncogenic signaling by unlinking PRC1.1 from target genes. Blood Cancer Discov., 3, 116–135. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0035] 35. Yin, G.Z., et al. (2016) BCORL1 is an independent prognostic marker and contributes to cell migration and invasion in human hepatocellular carcinoma. BMC Cancer, 16, 103. [DOI] [PMC free article] [PubMed] [Google Scholar] [Retracted]

[CIT0036] 36. Dromard, M., et al. (2007) The putative tumor suppressor gene PTPN13/PTPL1 induces apoptosis through insulin receptor substrate-1 dephosphorylation. Cancer Res., 67, 6806–6813. [DOI] [PubMed] [Google Scholar]

[CIT0037] 37. Niu, J., et al. (2009) Genetic polymorphisms in the PTPN13 gene and risk of squamous cell carcinoma of head and neck. Carcinogenesis, 30, 2053–2058. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0038] 38. McHeik, S., et al. (2020) Dual role of the PTPN13 tyrosine phosphatase in cancer. Biomolecules, 10, 1659. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0039] 39. Ji, C., et al. (2020) Identification of four genes associated with cutaneous metastatic melanoma. Open Med. (Wars), 15, 531–539. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0040] 40. Lind, G.E., et al. (2011) SPG20, a novel biomarker for early detection of colorectal cancer, encodes a regulator of cytokinesis. Oncogene, 30, 3967–3978. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0041] 41. Vonica, A., et al. (2020) Apcdd1 is a dual BMP/Wnt inhibitor in the developing nervous system and skin. Dev. Biol., 464, 7172–7187. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0042] 42. Han, W., et al. (2017) Epigenetic silencing of the Wnt antagonist APCDD1 by promoter DNA hyper-methylation contributes to osteosarcoma cell invasion and metastasis. Biochem. Biophys. Res. Commun., 491, 91–97. [DOI] [PubMed] [Google Scholar]

[CIT0043] 43. Tellez, C., et al. (2003) Role and regulation of the thrombin receptor (PAR-1) in human melanoma. Oncogene, 22, 3130–3137. [DOI] [PubMed] [Google Scholar]

[CIT0044] 44. Zhu, L., et al. (2020) GAS2L1 is a potential biomarker of circulating tumor cells in pancreatic cancer. Cancers, 12, 3774. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0045] 45. Gobeil, S., et al. (2008) A genome-wide shRNA screen identifies GAS1 as a novel melanoma metastasis suppressor gene. Genes Dev., 22, 2932–2940. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0046] 46. Yang, H.D., et al. (2018) T-cell immune regulator 1 enhances metastasis in hepatocellular carcinoma. Exp. Mol. Med., 50, e420. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0047] 47. Qi, C.X., et al. (2021) T cell immune regulator 1 is a prognostic marker associated with immune infiltration in glioblastoma multiforme. Oncol. Lett., 21, 252. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0048] 48. Kim, M., et al. (2006) Comparative oncogenomics identifies NEDD9 as a melanoma metastasis gene. Cell, 125, 1269–1281. [DOI] [PubMed] [Google Scholar]

[CIT0049] 49. Chakraborty, P.K., et al. (2017) MICU1 drives glycolysis and chemoresistance in ovarian cancer. Nat. Commun., 8, 14634. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0050] 50. Weber, C., et al. (2017) The serine protease inhibitor of Kazal-type 7 (SPINK7) is expressed in human skin. Arch. Dermatol. Res., 309, 767–771. [DOI] [PubMed] [Google Scholar]

[CIT0051] 51. Han, W., et al. (2021) Transcript levels of keratin 1/5/6/14/15/16/17 as potential prognostic indicators in melanoma patients. Sci. Rep., 11, 1023. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0052] 52. Delaunay, T., et al. (2017) Aberrant expression of kallikrein-related peptidase 7 is correlated with human melanoma aggressiveness by stimulating cell migration and invasion. Mol. Oncol., 11, 1330–1347. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0053] 53. White, N.M., et al. (2009) KLK6 and KLK13 predict tumor recurrence in epithelial ovarian carcinoma. Br. J. Cancer, 101, 1107–1113. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0054] 54. Hu, J., et al. (2018) Sox5 contributes to prostate cancer metastasis and is a master regulator of TGF-beta-induced epithelial mesenchymal transition through controlling Twist1 expression. Br. J. Cancer, 118, 88–97. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0055] 55. Kordass, T., et al. (2016) SOX5 is involved in balanced MITF regulation in human melanoma cells. BMC Med. Genom., 9, 10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0056] 56. Gu, W., et al. (2002) The LGI1 gene involved in lateral temporal lobe epilepsy belongs to a new subfamily of leucine-rich repeat proteins. FEBS Lett., 519, 71–76. [DOI] [PubMed] [Google Scholar]

[CIT0057] 57. Eyob, H., et al. (2013) Inhibition of ron kinase blocks conversion of micrometastases to overt metastases by boosting antitumor immunity. Cancer Discov., 3, 751–760. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0058] 58. Huang, L., et al. (2020) MSP-RON pathway: potential regulator of inflammation and innate immunity. Front. Immunol., 11, 569082. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0059] 59. Hoste, E., et al. (2019) Epithelial HMGB1 delays skin wound healing and drives tumor initiation by priming neutrophils for NET formation. Cell Rep., 29, 2689–2701.e4. [DOI] [PubMed] [Google Scholar]

[CIT0060] 60. Hoste, E., et al. (2015) Innate sensing of microbial products promotes wound-induced skin cancer. Nat. Commun., 6, 5932. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0061] 61. Lee, B.Y., et al. (2015) FAK signaling in human cancer as a target for therapeutics. Pharmacol. Ther., 146, 132–149. [DOI] [PubMed] [Google Scholar]

[CIT0062] 62. Carafa, V., et al. (2019) Dual tumor suppressor and tumor promoter action of sirtuins in determining malignant phenotype. Front. Pharmacol., 10, 38. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0063] 63. Alvarez-Artime, A., et al. (2020) Melatonin-induced cytoskeleton reorganization leads to inhibition of melanoma cancer cell proliferation. Int. J. Mol. Sci., 21, 548. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0064] 64. Wicki, A., et al. (2006) Tumor invasion in the absence of epithelial-mesenchymal transition: podoplanin-mediated remodeling of the actin cytoskeleton. Cancer Cell, 9, 261–272. [DOI] [PubMed] [Google Scholar]

[CIT0065] 65. Ward, P.S., et al. (2012) Metabolic reprogramming: a cancer hallmark even warburg did not anticipate. Cancer Cell, 21, 297–308. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0066] 66. Kim, H.Y., et al. (2017) Discovery of potential biomarkers in human melanoma cells with different metastatic potential by metabolic and lipidomic profiling. Sci. Rep., 7, 8864. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0067] 67. DeBerardinis, R.J., et al. (2010) Q’s next: the diverse functions of glutamine in metabolism, cell biology and cancer. Oncogene, 29, 313–324. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0068] 68. Liu, W., et al. (2012) Reprogramming of proline and glutamine metabolism contributes to the proliferative and metabolic responses regulated by oncogenic transcription factor c-MYC. Proc. Natl. Acad. Sci. U. S. A., 109, 8983–8988. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0069] 69. Kanarek, N., et al. (2018) Histidine catabolism is a major determinant of methotrexate sensitivity. Nature, 559, 632–636. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0070] 70. Wiggins, T., et al. (2015) Tyrosine, phenylalanine, and tryptophan in gastroesophageal malignancy: a systematic review. Cancer Epidemiol. Biomarkers Prev., 24, 32–38. [DOI] [PubMed] [Google Scholar]

[CIT0071] 71. Blagoeva, P.M. (1984) Aminoacids--precursors of melanin synthesis in hamster melanoma. J. Cancer Res. Clin. Oncol., 108, 366–368. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0072] 72. Huo, M.M., et al. (2021) Interplay among metabolism, epigenetic modifications, and gene expression in cancer. Front. Cell Dev. Biol., 9, 793428. [DOI] [PMC free article] [PubMed] [Google Scholar]

[CIT0073] 73. Du, C., et al. (2022) Analysis of 5-methylcytosine regulators and DNA methylation-driven genes in colon cancer. Front. Cell Dev. Biol., 9, 657092. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

The environmental carcinogen benzo[a]pyrene regulates epigenetic reprogramming and metabolic rewiring in a two-stage mouse skin carcinogenesis model

Md Shahid Sarwar

Christina N Ramirez

Hsiao-Chen Dina Kuo

Pochung Chou

Renyi Wu

Davit Sargsyan

Yuqing Yang

Ahmad Shannar

Rebecca Mary Peter

Ran Yin

Yujue Wang

Xiaoyang Su

Ah-Ng Kong

Abstract

Graphical abstract

Graphical Abstract.

Introduction

Materials and methods

Chemicals and reagents

Animal studies

Figure 1.

Nucleic acids isolation and next-generation sequencing

Bioinformatic analyses of RNA-seq

Bioinformatic analyses of SureSelect DNA Methyl-seq

Correlation between DNA methylation and RNA expression

LC–MS metabolomic analysis

Statistical analysis

Results

Development of skin carcinogenesis in SKH-1 mice

BaP/TPA mediates DNA methylation changes in skin carcinogenesis

Figure 2.

BaP/TPA mediates transcriptomic changes in skin carcinogenesis

Figure 3.

Figure 4.

BaP/TPA regulates signaling pathways in skin carcinogenesis

Figure 5.

Inverse correlation between DNA methylation and RNA expression shows potential epigenetic targets

BaP/TPA rewires metabolites/metabolic pathways and epigenetic-associated metabolites

Figure 6.

Discussion

Supplementary material

Acknowledgements

Abbreviations

Contributor Information

Funding

Authors’ contributions

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases