Skip to main content
NCBI home page
Scientific Reports logoLink to Scientific Reports
. 2025 Feb 15;15:5641. doi: 10.1038/s41598-025-90247-1

Genome-wide DNA methylation regulation analysis provides novel insights on post-radiation breast cancer

Jianghuai Ji 1,2, Dongguo Li 3, Xiaoxiao Zhao 4, Yajuan Wang 1,2, Binbing Wang 1,2,5,✉
PMCID: PMC11830005  PMID: 39955415

Abstract

Breast cancer (BC) is the most common malignancy with a poor prognosis. Radiotherapy is one of the leading traditional treatments for BC. However, radiotherapy-associated secondary diseases are severe issues for the treatment of BC. The present study integrated multi-omics data to investigate the molecular and epigenetic mechanisms involved in post-radiation BC. The differences in the expression of radiation-associated genes between post-radiation and pre-radiation BC samples were determined. Enrichment analysis revealed that these radiation-associated genes involved diverse biological functions and pathways in BC. Combining epigenetic data, we identified radiation-associated genes whose transcriptional changes might be associated with aberrant methylation. Then, we identified potential therapeutic targets and chemical drugs for post-radiation BC patient treatment by constructing a drug-target association network. Specifically, four radiation-associated genes (CD248, CCDC80, GADD45B, and MMP2) whose increased expression might be regulated by hypomethylation of the corresponding enhancer region were found to have excellent diagnostic effects and clinical prognostic value. Finally, we further used independent samples to verify CD248 expression and established a simple epigenetic regulatory model. In summary, this study provides novel insights for understanding the regulation of target genes mediated by DNA methylation and developing potential biomarkers for radiation-associated secondary diseases in BC.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-90247-1.

Keywords: Breast cancer, Radiotherapy, DNA methylation, Epigenetic regulation, Biomarker

Subject terms: Cancer, Biomarkers, Oncology

Introduction

Breast cancer (BC) is the most common and gradually becoming the leading cause of cancer-related premature death among women1. Its incidence increases with age and other risk factors, such as family history of cancer and genetic variants2. At present, radiotherapy, one of the standard treatment methods for BC, has effectively reduced the recurrence and mortality of patients3. However, since ionizing radiation can damage nearly half of the DNA in a cell, it can lead to many unintended consequences, such as cell mutation, carcinogenesis, or even death4. Furthermore, enough evidence also confirms that radiotherapy for BC can cause secondary cancers and cardiomyopathy decades later5,6. Moreover, heterogeneous differences in the effects of radiotherapy on BC patients indicated that relevant genetic and epigenetic variations among individuals seriously affect clinical outcomes and quality of life7,8. Therefore, exploring the molecular and epigenetic mechanisms underlying the response to breast cancer radiation (BCRT) can help patients effectively avoid overtreatment or undertreatment, which is still an urgent problem to be solved in clinical practice.

In recent years, studies have shown that aberrant gene expression is the leading cause of the cell response to radiation after radiotherapy for BC. For example, three genes were highly expressed in BC tissues after radiation, and the upregulation of their expression might be a genetic risk factor for radiotherapy-associated secondary cancer9. Minafra et al. showed that gene expression characteristics activated by radiation at different doses in primary breast cancer cells could predict the response to radiotherapy and help to determine personalized biologically driven treatment programs10. Although previous studies have focused on the regulatory activity of some genes after radiotherapy for BC, the exact molecular mechanism of most target genes, especially the regulatory effect of DNA methylation on the expression of target genes after BCRT, is still unknown.

DNA methylation, a vital epigenetic modification, frequently occurs in the promoter and enhancer regions of genes and can mediate the dysregulation of gene expression11. Generally, hypermethylation of promoter or enhancer regions can downregulate or even silence gene expression. In contrast, hypomethylation of promoter or enhancer regions tends to upregulate or even activate gene expression12,13. So far, the effect of some specific gene methylation on breast cancer after radiation has been reported. For instance, studies have shown that methylation of BRCA1 by PRMT1 may activate DNA repair of breast cancer cells exposed to ionizing radiation14. However, the association between methylation events in specific regions of target genes and alterations in transcriptional levels in tumors following BCRT remains unclear.

Using large-scale DNA methylation data and RNA-sequencing data produced by the Gene Expression Omnibus (GEO), we investigated aberrant gene expression and DNA methylation patterns after BCRT in this study (Fig. 1A). First, we reannotated gene expression profiles and detected differentially expressed genes (DEGs) to obtain radiation-associated genes after BCRT. Functional enrichment analysis was used to predict the biological functions of these target genes and further evaluate the regulatory mechanisms they play after BCRT. Importantly, we used a reannotation strategy to construct an integrated model combining multi-omics data to identify radiation-associated genes that might be regulated by promoter or enhancer region methylation. Then, by constructing a drug-target association network and performing a prognostic analysis, the present study identified potential drugs for patient treatment and prognostic targets after BCRT. Finally, we further used independent samples to verify CD248 expression and established a simple epigenetic regulatory model. The present study enhances the understanding of the BC cell response to radiation and might greatly aid in studying how DNA methylation regulates gene expression after BCRT.

Fig. 1.

Fig. 1

Open in a new tab

Characterization of radiation-associated genes in breast cancer (Drawing by PowerPoint and R 4.1.1 software). (A) Schematic description of the main steps used to identify radiation-associated genes regulated by aberrant methylation in breast cancer and analyze their function. (B, C) Volcano plots of DEGs in GSE65505 and GSE59733, respectively. (D) UpSet plot showing the number of radiation-associated genes.

Materials and methods

Data sources

The various omics data used in this study were obtained from different platforms. The expression and DNA methylation data used in this study were downloaded from GEO for all samples. The experimental expression datasets (GSE65505) contained 32 post-radiation BC samples and 30 pre-radiation BC samples. GSE59733, which included ten post-radiation tumor samples and nine pre-radiation tumor samples, was used to cross-validate the experimental dataset. The DNA methylation data (GSE45958) generated from the Infinium 450k platform included 14 post-radiation BC samples and 10 pre-radiation BC samples. The validation expression data came from two cell lines (MCF-7 and MDA-MB-231). In this study, six post-radiation samples (GSE63667) and six pre-radiation samples (GSE138442) were used for MCF-7 cells, and six post-radiation samples (GSE127789) and ten pre-radiation samples (GSE178748) were utilized for MDA-MB-231 cells. These validation data were generated by the same technology platform (GPL4133). The number of samples for each profile is shown in Supplementary Table S1. Clinical data were downloaded from TCGA for survival analysis. The comprehensive human gene annotation data (release 19) were obtained from GENCODE15, consistent with the human genome version referenced by the technology platform for generating the experimental expression datasets and DNA methylation data. The drug target information was derived from the Comparative Toxicogenomics Database (CTD)16.

Construction of gene expression profiles and identification of radiation-associated genes

Since the raw expression matrix files were the expression values of the probes, the gene expression profiles needed to be constructed first. According to the annotation information provided by the GPL17586 (GSE65505), GPL18990 (GSE59733), and GPL4133 (validation data) platforms, probe ID and gene symbol were converted in downloaded matrices. The probes were annotated to multiple gene regions, and unidentified probes were removed. If numerous probes were annotated to the same gene region, the average expression value of the probes was used as the gene expression.

We identified radiation-associated genes between post-radiation and pre-radiation BC samples through the R package ‘limma,’ designed based on generalized linear models17. P ≤ 0.05 and fold change ≥ 1.5 were used as thresholds for screening.

Functional annotation of radiation-associated genes in BC

To elucidate the changes in molecular function induced by BCRT, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses were performed to identify the significantly enriched functional terms and pathways related to the radiation-associated genes18–20. Enrichment analysis was performed via the R package ‘clusterProfiler’21 and visualized by the R package ‘ggplot2’. Only the GO terms with corrected p ≤ 0.05 and pathways with p ≤ 0.05 were retained to assess the potential functional relevance of radiation-associated genes. The p-values were corrected using the Benjamini-Hochberg (BH) method.

Construction of methylation profiles for promoter and enhancer regions in post-radiation and pre-radiation BC

We constructed methylation profiles for each sample using methylation levels of the probes mapped to the promoter and enhancer regions. To obtain a relatively comprehensive range of promoters, we used 10 kb upstream from the transcriptional start site (TSS) of each gene as its promoter region22. Since DNA methylation of the proximal promoter is closely related to transcriptional silencing, only the probe closest to each TSS was used to determine the methylation status of each gene promoter12.

Meanwhile, we got relatively comprehensive probes localized in enhancer regions from the supplemental file of a previously published reference23 and the platform GPL13534 (GSE45958). After merging the two sets of probes and removing duplicate probes, the remaining 161,708 enhancer probes were retained for further study. Since previous studies revealed that the size of a specific enhancer region is approximately 1000 bp24,25, we constructed an enhancer region using a window of 500 bp directly upstream of and downstream from the enhancer probe coordinates, and overlapping intervals were joined to expand into a larger interval25. Then, we calculated the average methylation level of probes within the interval to represent the methylation status of an enhancer region.

Identification of radiation-associated genes regulated by aberrantly methylated promoter or enhancer regions

We used the ‘limma’ package to identify aberrantly methylated regions between the post-radiation and pre-radiation samples (p ≤ 0.05). The intersection of genes with opposite methylation and expression differences were considered to be radiation-associated genes regulated by promoter regions in BC.

To screen for radiation-associated genes regulated by enhancer regions, we removed the genes whose expression was regulated by promoter methylation. Similarly, genes and enhancer regions with opposite methylation and expression differences were retained for further analysis. A previous study showed that the maximum known distance between an enhancer region and a gene was about 1 Mbp26. We selected aberrantly methylated enhancer region (AMER)-gene pairs located on the same chromosome, with a maximal linear distance of 1 Mbp between the center of the enhancer region and the TSS of the gene.

Prediction of potential drugs for post-radiation BC treatment and identification of radiation-associated genes related to good prognosis

CDT is an integrated knowledgebase that can detect the modification of chemotherapeutic drugs on mRNA expression16. We filtered out and integrated 356,358 nonredundant mRNA-drug regulatory associations for drugs that could reduce mRNA expression. Then, we matched these mRNA-drug associations with the radiation-associated genes to construct a drug-target association network.

To identify the clinical effects of radiation-associated genes regulated by aberrantly methylated promoter or enhancer regions, patients from TCGA who had received radiotherapy were divided into two groups using optimal truncation values of gene expression calculated by the R package ‘maxstat.’ Kaplan-Meier survival analysis and the log-rank test (p ≤ 0.05) were performed to estimate the significant prognosis differences between the two patient groups. Moreover, receiver operating characteristic (ROC) curves and their area under the curve (AUC) values were utilized to appraise the specificity and sensitivity of the signature by using the ‘ROC’ function in the package ‘pROC’27.

All analyses were performed using R 4.1.1 software.

Results

Characterization of differences and functional prediction of radiation-associated genes in BC

Expression profiles and DNA methylation profiles were used in this study to assess the pattern of radiation-associated genes after BCRT. To construct the expression profiles related to radiation in BC, we annotated 37,090 probes to 27,727 genes and 77,839 probes to 22,477 genes in GSE65505 and GSE59733, respectively. After preprocessing the profiles, we focused on genes with significantly differential expression before and after radiation in BC. A total of 218 DEGs were identified in GSE65505, including 183 highly expressed genes and 35 genes with low expression. Meanwhile, 448 DEGs, including 377 upregulated genes and 71 downregulated genes, were identified in post-radiation tumor samples compared to the corresponding pre-radiation tumor samples (GSE59733). The volcano plot was applied to the DEGs distribution in GSE65505 (Fig. 1B) and GSE59733 (Fig. 1C). To ensure the quality of the research, we used the intersection of DEGs in the two gene chips for further investigation. We identified 80 commonly upregulated genes via the Venn diagram (Fig. 1D). We performed multiple testing corrections for p-values, and the p-values were corrected using the false discovery rate (FDR) method. The results showed that the corrected p-values of these 80 genes were all ≤ 0.0138 (Supplementary Table S2). Hierarchical clustering analysis using expression level indicated that these 80 genes could obviously distinguish the post-radiation and pre-radiation tumor samples (Fig. 2A and B). We reasoned that these genes might become radiation-associated genes after BCRT.

Fig. 2.

Fig. 2

Open in a new tab

Radiation-associated genes revealing different biological functions and pathways after BCRT (Drawing by R 4.1.1 software). (A, B) Heatmaps of radiation-associated genes in GSE65505 and GSE59733, respectively. On the x-axis, red represents the post-radiation samples, and purple represents the pre-radiation controls. The y-axis represents the radiation-associated genes. (C) Significantly enriched GO terms of radiation-associated genes. (D) GO terms related to immune processes and inflammatory responses. (E) Significantly enriched KEGG pathways of radiation-associated genes.

Although radiotherapy for cancer is an intricate and dynamic biological process, recent studies have confirmed that genetic dysregulation after radiation plays complex and critical roles in cancer progression10,28. We then performed functional enrichment analysis on radiation-associated genes in BC to evaluate their biological characteristics. The results indicated these genes were enriched in 646 biological processes and 22 KEGG pathways. The results revealed that cell cycle activity-related processes (such as cell activation, cell growth, cellular metabolic process, cell differentiation, cell proliferation, cell adhesion, cell migration, apoptosis, and cell death) were enriched in these radiation-associated genes (Fig. 2C). Radiation initiates and modulates the immune and inflammatory responses through a range of cell-derived factors in the tumor microenvironment29. The immune system can modulate the expression of radiation-induced normal and tumor tissue damage30. While it can help cure cancer, it can also cause acute and late radiation side effects, often manifesting as acute and chronic inflammatory disease states. Notably, the enriched GO terms in this study were involved in many immune processes and inflammatory responses (Fig. 2D), which were closely related to the progression of the organism after radiotherapy. In addition, we also found that the significantly enriched GO terms in these target genes included some biological processes of immune cells (such as GO:2000516, 0042129, 0002661, and 2000448), which might be related to the regulation of tumor radiosensitivity. Interestingly, some radiation-associated genes were found to enrich biological processes of ionizing radiation (GO:0010212 and 0009314). According to the KEGG pathway analysis, the significantly enriched pathways included the p53 signaling pathway, the FoxO signaling pathway, the mineral absorption pathway, and transcriptional misregulation in cancer and tuberculosis (Fig. 2E).

Exploration of radiation-associated genes regulated by aberrant methylation in BC

Previous studies have revealed that differential methylation of gene promoters or distal enhancer regions can silence or activate the expression of corresponding genes31,32. Strikingly, enhancer methylation was significantly more strongly associated with distal gene dysregulation than promoter methylation in cancer and might even occur when the promoter is constantly unmethylated33. We inferred that the upregulation of radiation-associated genes after BCRT might be affected by hypomethylation in their corresponding promoter or enhancer regions.

To construct methylation profiles of promoter and enhancer regions for post-radiation and pre-radiation BC, this study adopted a computational strategy to reannotate data from Infinium 450k arrays into radiation-associated gene-related promoter and enhancer regions (Supplementary Fig S1 and described in the methods section for details). In total, 233 probes were located in the promoter regions of most radiation-associated genes (Fig. 3A). If one gene had several probes mapping to the corresponding promoter region, only the probe closest to each TSS was retained to determine the DNA methylation status of this promoter. Meanwhile, we constructed 114,499 non-overlapping enhancer regions using 161,708 nonredundant enhancer probes (Fig. 3B). The results showed that most enhancer regions were extended by one enhancer probe. For multiple probes forming the enhancer regions, the average methylation value of probes within one enhancer region was computed as the DNA methylation level of this region.

Fig. 3.

Fig. 3

Open in a new tab

Exploration of radiation-associated genes regulated by aberrant methylation in breast cancer (Drawing by Cytoscape software v3.7.2 and R 4.1.1 software). (A) Pie chart of the proportion of radiation-associated genes located in the promoter region. (B) Histogram of the proportion of enhancer region length in post-radiation breast cancer. (C, D) Bar plots of the differences in the expression and methylation of radiation-associated genes and the enhancer regions (or promoter regions) that regulate them, respectively. (E) Interaction network of radiation-associated genes regulated by aberrant methylation. (F) Aberrantly methylated enhancer regions and radiation-associated genes regulated by aberrant methylation distribution in human chromosomes. The first circle shows the chromosomal location and ordering information. The second circle represents the distribution of aberrantly methylated regions. The third and fourth circles represent the distribution of radiation-associated genes regulated by aberrant methylation. The innermost circle represents the interacting genes.

Since aberrant promoter and enhancer region methylation is generally negatively correlated with corresponding gene expression, we recognized hypomethylated promoter genes and hypomethylated enhancer regions (hypoERs) by comparing the methylation levels between post-radiation and pre-radiation samples. Only one promoter gene (CYP1B1) that exhibited both hypomethylation and high expression was selected for subsequent analysis. To further mine radiation-associated genes that distal hypoERs might regulate, we established a model linking hypoERs with target genes (described in the methods section for details). Finally, we obtained 40 hypoER-gene pairs, of which there were 40 hypoERs and 27 radiation-associated genes (Fig. 3C and D). Studies have shown that distal expression-related methylation sites are abundant in the human genome, co-localizing with enhancer chromatin marks, and are more predictive of expression levels than promoter methylations33. Interestingly, we found that far more radiation-associated genes might be regulated by enhancer methylation than promoter methylation (Supplementary Table S3). The above results might indicate that alterations in enhancer methylation play a significant role in BCRT progression.

Previous studies have confirmed that interactions between genes can lead to the occurrence and development of diseases34,35. To further explore the influence of the interaction between these radiation-associated genes regulated by aberrant methylation on organism progression after BCRT, a gene-gene interaction regulatory network was constructed using the STRING database in this study (Fig. 3E and F). The results showed that the regulatory network contained 32 relationship pairs and 19 genes. Finally, the regulatory network was visualized by Cytoscape software (v3.7.2).

Identification of drug targets and radiation-associated genes associated with BC prognosis

With the increasing importance of precision medicine in healthcare, pharmacogenomics has continued to gain prominence in the clinical field36. Finally, based on the radiation-associated genes in the regulatory network and the information in CDT, we inferred that some potential drugs could be used to treat patients with BC after radiotherapy by constructing a drug-target association network targeting mRNAs (Fig. 4A and Supplementary Table S4). We obtained 413 candidate drugs and 19 mRNAs in the drug-target association network. In this network, these potential drugs could achieve the purpose of treatment by decreasing the expression of the corresponding radiation-associated genes. For example, MMP2 is a gelatinase subfamily member crucial in extracellular matrix turnover. Previous reports have described that ionizing radiation (IR) increases MMP2 activity in multiple cell types, including BC cells37,38. In addition, MMP2 silencing increases the cytotoxicity and DNA damage of IR-induced BC cells and reduces the viability of BC cells39. Significantly, these drugs associated with MMP2 in the drug-target association network have been documented to downregulate MMP2 expression. For instance, Taiwanin E inhibited cell migration in human colon cancer cells by suppressing MMP2 expression via the p38 MAPK pathway40, and Fatostatin could attenuate BC-induced osteolysis41. In summary, these drugs might improve the curative effect of radiotherapy for BC by suppressing MMP2 expression (Supplementary Table S4).

Fig. 4.

Fig. 4

Open in a new tab

Construction of the drug-target association network and identification of four radiation-associated genes associated with good prognosis (Drawing by Cytoscape software v3.7.2 and R 4.1.1 software). (A) Drug-target network based on the mRNAs regulated by hypomethylated enhancer regions. The green arrows are drugs, and the red circles are mRNAs. (B–E) Survival analysis curves of CD248, CCDC80, GADD45B and MMP2. The numbers on the x-axis represent the number of living patients at that time. (F, G) Receiver operating characteristic analysis of the expression of the four radiation-associated genes.

To further evaluate the potential of radiation-associated genes as prognostic factors for BCRT, we integrated the expression profile with patient clinical information. This analysis enabled to identify target genes that were significantly associated with overall survival in individuals undergoing BCRT treatment. The results revealed that four radiation-associated genes (CD248, CCDC80, GADD45B, and MMP2) significantly affected survival. As shown in Fig. 4B-E, patients in the high-expression group had significantly lower survival rates compared to the low-expression group. To assess the diagnostic value of these four genes, we plotted ROC curves of the four genes according to two sets of expression data (Fig. 4F and G). The results showed that the AUC values for the diagnostic potential of the four gene expressions in BCRT were > 0.75, which suggested that they could distinguish post-radiation tumor samples from pre-radiation samples and might become diagnostic biomarkers for BCRT.

A hypomethylated enhancer region is negatively correlated with CD248 in post-radiation BC

To ensure the accuracy of our results, we further verified prognosis-related target gene expression in two BC cell lines by combining critical nodes in the regulatory network. To develop the expression profiles from the validated dataset, this study annotated 32,696 probes to 19,749 genes utilizing the GPL4133 platform. Exception for unverifiable target genes, CD248 expression in post-radiation samples was significantly greater than in pre-radiation samples (Fig. 5A and B). Numerous studies have shown that CD248 is a therapeutic target for fibrotic diseases and is closely associated with pulmonary fibrosis. Overexpressed CD248 might be involved in the pathogenesis of pulmonary fibrosis, and CD248 silencing reduces the proliferation of lung fibroblasts42,43. Interestingly, pulmonary fibrosis is known to be a common complication of thoracic radiotherapy in BC patients44. We speculated that the upregulation of CD248 expression after BCRT might be closely related to the pathogenesis of radiation-induced pulmonary fibrosis.

Fig. 5.

Fig. 5

Open in a new tab

A hypomethylated enhancer region is negatively correlated with CD248 in post-radiation breast cancer (Drawing by PowerPoint and R 4.1.1 software). (A, B) Violin plots are presented with comparisons of expression levels between post-radiation and pre-radiation breast cancer samples of CD248. (C) The correlation coefficient for the expression of CD248 with the methylation of cg06794612 in this enhancer region. (D) A model of this enhancer region regulating its target genes. (E) Potential chemical drugs targeting CD248 for the treatment of post-radiation breast cancer. (F) Heatmap of the correlation coefficients between immune factors and CD248.

Furthermore, we noticed that the increased expression of CD248 in this study might be influenced by hypomethylation of the corresponding enhancer region (cg06794612). We further evaluated the correlation between the expression levels of CD248 and the methylation levels of cg06794612. The results indicated significant negative correlations between the two variables (Fig. 5C). Therefore, this enhancer region might be a potential epigenetic therapeutic target for BCRT. In summary, these findings possibly supported a simple model (Fig. 5D) wherein the enhancer region (cg06794612) exhibited hypermethylation in preirradiated cells, leading to the repression of CD248. Following radiation exposure, this region became hypomethylated and activated CD248 expression, promoting pulmonary fibrosis proliferation after BCRT. Therefore, selecting appropriate drugs to inhibit CD248 expression effectively might be beneficial for hindering the progression of pulmonary fibrosis in BC patients after radiation exposure. In this study, we found nine drugs that might play a role in reducing CD248 expression (Fig. 5E). In addition, studies have shown that tumor-infiltrating immune cells are vital for BC treatment and patient prognosis. To compare differences in the immune microenvironment associated with CD248, abundance tumor-infiltrating immune cells data were estimated and downloaded from the TISIDB and TIMER databases45,46. The results showed that immunomodulator TGFB1 was most strongly correlated with CD248 (Fig. 5F). Furthermore, the CD4 + T-cell infiltration level was most strongly correlated with CD248 (Supplementary Fig S2).

Discussion

Radiotherapy has been one of the primary modalities of BC therapy47. However, a significant problem with radiotherapy for cancer patients is that many patients may not benefit from radiotherapy but may suffer from radiation-induced toxicity due to tumor molecular heterogeneity48. In the era of precision medicine, the exploration of tumor radiosensitivity at the genomic level and personalized radiotherapy based on cancer biology has attracted much attention and become increasingly important49.

In this study, we comprehensively investigated changes in gene expression after BCRT. We ultimately identified 80 radiation-associated genes with upregulated expression in post-radiation BC through differential expression analysis. These radiation-associated genes might be closely related to the radiosensitivity of BC and could become important targets of personalized radiotherapy for BC. For example, highly expressed MDM2 and FOSB were the two most significant genes in the two experimental datasets, respectively. Multiple studies have confirmed that the overexpression of MDM2 decreases the extent to which radiation induces an increase in the activity of the p53 tumor suppressor50,51. MDM2 is a potential target for therapeutic intervention, and inhibiting its expression could sensitize cells to ionizing radiation, thus increasing the effectiveness of radiotherapy. Nie et al.52 reported that FOSB was co-upregulated in cervical, nasopharynx, and tongue cancers by studying the peripheral blood of patients with three cancers after the first fraction of 2 Gy irradiation, suggesting that it might be a helpful predictor of hematological toxicity in patients after radiotherapy. Combined with our findings, overexpressed FOSB in post-radiation BC samples might be a potential predictor for the early diagnosis of radiation toxicity.

Genes play essential regulatory roles in cancer pathogenesis through different biological functions and signaling pathway networks. We investigated the importance of these radiation-associated genes in the progression of BC to radiotherapy. Through enrichment analysis, we found that the target genes were involved in many biological processes related to immune and inflammatory responses. The tumor immune microenvironment is a determinant of the response to cancer immunotherapy. Radiotherapy has direct cytotoxic effects on malignant cells, but there is increasing evidence that it reprograms the tumor’s immune microenvironment53. Furthermore, some cancer-related pathways were also enriched in these radiation-associated genes. These results help to understand the development of BC after radiation to some extent.

Numerous studies have shown that DNA methylation can serve as a predictive factor for radiotherapy response, and manipulating DNA methylation patterns can enhance tumor radiosensitization54,55. Therefore, we developed strategies to speculate which highly expressed radiation-associated genes were regulated by hypomethylated promoter or enhancer regions. Interestingly, only one gene with high expression in our study might be regulated by promoter region hypomethylation. A study revealed that high-dose radiation increased the expression of CYP1B1 in rats, zebrafish, and humans56. Notably, Tokizane et al. reported that CYP1B1 is overexpressed in prostate cancer and is regulated by hypomethylation of its promoter region57. Moreover, other studies have also reported significantly reduced methylation of CYP1B1 in some diseases58,59. We concluded that the increased CYP1B1 expression after BCRT might be influenced by hypomethylation of the corresponding promoter (cg26682499), indicating that hypomethylation of CYP1B1 might play an essential role in post-radiation breast cancer. We also believe that other radiation-associated genes and corresponding hypomethylated enhancer regions play irreplaceable roles after BCRT.

Importantly, we concluded that a radiation-associated gene (CD248) regulated by hypomethylated enhancers after BCRT might be strongly associated with pulmonary fibrosis. In carcinomas, CD248 protein was detected in tumor capillaries and fibroblasts. Meanwhile, Animal experiments have shown that the expression of CD248 in peripheral whole blood significantly changes after radiation60. We have reason to believe that CD248 might be an important therapeutic target after BCRT. Finally, in addition to evaluating the prognostic value of target genes regulated by DNA methylation, we also assessed the prognostic effects of the remaining radiation-associated genes in BRCT. The results showed that six radiation-associated genes significantly affected survival (Supplementary Figure S3). Patients in the low-expression group had significantly longer survival time than those in the high‐expression group.

There is no denying that the present research has some shortcomings that should be addressed. Since sample acquisition of post-radiation BC is a problem, this deficiency will likely affect the final results to some degree. In the future, studies of aberrant methylation and expression in post-radiation BC should be replicated in a larger cohort. Moreover, due to technical and time constraints, it was impossible to verify the radiation-associated genes in BC and the relationships between the expression changes of these genes and methylation events. Further functional investigations and molecular experiments are still required to explore the mechanisms underlying the roles of novel biomarkers.

Conclusion

This study aimed to investigate the regulatory mechanism of expression and DNA methylation after BCRT by integrating multi-omics data. The identification of radiation-associated genes and corresponding aberrantly methylated regions will aid in understanding the mechanism of many adverse events after BCRT. They might play essential roles in the progression of breast cancer, radiation resistance, or secondary disease. The identified clinically relevant radiation-associated genes could be further evaluated as post-radiation BC biomarkers. Moreover, this study provides insight into potential drug targets for post-radiation BC treatment.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1 (16.2KB, docx)
Supplementary Material 2 (26.7KB, docx)
Supplementary Material 3 (23.4KB, docx)
Supplementary Material 4 (40.6KB, xlsx)
Supplementary Material 5 (702.9KB, docx)

Abbreviations

BC

Breast cancer

BCRT

Breast cancer radiation

GEO

Gene Expression Omnibus

DEGs

Differentially expressed genes

GO

Gene Ontology

KEGG

Kyoto Encyclopedia of Genes and Genomes

TSS

Transcriptional start site

AMER

Aberrantly methylated enhancer region

Author contributions

JJ and BW conceived and designed the study. JJ wrote the manuscript. JJ, XZ and YW made substantial contributions to the acquisition of data. JJ and XZ analyzed and interpreted the data. BW and DL reviewed and revised the manuscript. All authors read and approved the final manuscript.

Funding

This study was financially supported by Zhejiang Province Key Research and Development Program (2019C03003), and the Medical Science and Technology Program of Zhejiang Province (2021PY039), National Natural Science Foundation of China (12005190).

Data availability

The datasets used and/or analyzed during the present study are available from the corresponding author or GitHub (https://github.com/huai-jj/methylation.git) upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

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

References

  • 1.Sung, H. et al. Global Cancer statistics 2020: GLOBOCAN estimates of incidence and Mortality Worldwide for 36 cancers in 185 countries. CA Cancer J. Clin.71, 209–249. 10.3322/caac.21660 (2021). [DOI] [PubMed] [Google Scholar]
  • 2.Coughlin, S. S. Epidemiology of breast Cancer in women. Adv. Exp. Med. Biol.1152, 9–29. 10.1007/978-3-030-20301-6_2 (2019). [DOI] [PubMed] [Google Scholar]
  • 3.Klein, J. et al. Locally advanced breast cancer treated with neoadjuvant chemotherapy and adjuvant radiotherapy: a retrospective cohort analysis. BMC Cancer. 19, 306. 10.1186/s12885-019-5499-2 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Santivasi, W. L. & Xia, F. Ionizing radiation-induced DNA damage, response, and repair. Antioxid. Redox Signal.21, 251–259. 10.1089/ars.2013.5668 (2014). [DOI] [PubMed] [Google Scholar]
  • 5.Pignol, J. P., Keller, B. M. & Ravi, A. Doses to internal organs for various breast radiation techniques–implications on the risk of secondary cancers and cardiomyopathy. Radiat. Oncol.6, 5. 10.1186/1748-717X-6-5 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Darby, S. C., McGale, P., Taylor, C. W. & Peto, R. Long-term mortality from heart disease and lung cancer after radiotherapy for early breast cancer: prospective cohort study of about 300,000 women in US SEER cancer registries. Lancet Oncol.6, 557–565. 10.1016/S1470-2045(05)70251-5 (2005). [DOI] [PubMed] [Google Scholar]
  • 7.Speers, C. et al. Development and validation of a Novel Radiosensitivity signature in human breast Cancer. Clin. Cancer Res.21, 3667–3677. 10.1158/1078-0432.CCR-14-2898 (2015). [DOI] [PubMed] [Google Scholar]
  • 8.Bourguignon, M. H. et al. Genetic and epigenetic features in radiation sensitivity part I: cell signalling in radiation response. Eur. J. Nucl. Med. Mol. Imaging. 32, 229–246. 10.1007/s00259-004-1730-7 (2005). [DOI] [PubMed] [Google Scholar]
  • 9.Yao, G., Zhao, K., Bao, K. & Li, J. Radiation increases COL1A1, COL3A1, and COL1A2 expression in breast cancer. Open. Med. (Wars). 17, 329–340. 10.1515/med-2022-0436 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Minafra, L. et al. Radiation Gene-expression signatures in primary breast Cancer cells. Anticancer Res.38, 2707–2715. 10.21873/anticanres.12512 (2018). [DOI] [PubMed] [Google Scholar]
  • 11.Hernando-Herraez, I., Garcia-Perez, R., Sharp, A. J. & Marques-Bonet, T. DNA methylation: insights into human evolution. PLoS Genet.11, e1005661. 10.1371/journal.pgen.1005661 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Zhi, H. et al. A novel reannotation strategy for dissecting DNA methylation patterns of human long intergenic non-coding RNAs in cancers. Nucleic Acids Res.42, 8258–8270. 10.1093/nar/gku575 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Flam, E. L. et al. Differentially methylated super-enhancers regulate Target Gene expression in Human Cancer. Sci. Rep.9, 15034. 10.1038/s41598-019-51018-x (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Montenegro, M. F. et al. PRMT1-dependent methylation of BRCA1 contributes to the epigenetic defense of breast cancer cells against ionizing radiation. Sci. Rep.1010.1038/s41598-020-70289-3 (2020). [DOI] [PMC free article] [PubMed]
  • 15.Harrow, J. et al. GENCODE: the reference human genome annotation for the ENCODE Project. Genome Res.22, 1760–1774. 10.1101/gr.135350.111 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Davis, A. P. et al. Comparative toxicogenomics database (CTD): update 2023. Nucleic Acids Res.51, D1257–D1262. 10.1093/nar/gkac833 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ritchie, M. E. et al. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res.43, e47. 10.1093/nar/gkv007 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kanehisa, M. & Goto, S. KEGG: Kyoto Encyclopedia of genes and genomes. Nucleic Acids Res.28, 27–30. 10.1093/nar/28.1.27 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kanehisa, M. Toward understanding the origin and evolution of cellular organisms. Protein Sci.28, 1947–1951. 10.1002/pro.3715 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kanehisa, M., Furumichi, M., Sato, Y., Kawashima, M. & Ishiguro-Watanabe, M. KEGG for taxonomy-based analysis of pathways and genomes. Nucleic Acids Res.51, D587–D592. 10.1093/nar/gkac963 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yu, G., Wang, L. G., Han, Y. & He, Q. Y. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS16, 284–287. 10.1089/omi.2011.0118 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Xiao, W. et al. Genome-wide DNA methylation patterns analysis of noncoding RNAs in temporal lobe Epilepsy patients. Mol. Neurobiol.55, 793–803. 10.1007/s12035-016-0353-x (2018). [DOI] [PubMed] [Google Scholar]
  • 23.Yao, L., Shen, H., Laird, P. W., Farnham, P. J. & Berman, B. P. Inferring regulatory element landscapes and transcription factor networks from cancer methylomes. Genome Biol.16, 105. 10.1186/s13059-015-0668-3 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hnisz, D. et al. Super-enhancers in the control of cell identity and disease. Cell155, 934–947. 10.1016/j.cell.2013.09.053 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bell, R. E. et al. Enhancer methylation dynamics contribute to cancer plasticity and patient mortality. Genome Res.26, 601–611. 10.1101/gr.197194.115 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lettice, L. A. et al. A long-range shh enhancer regulates expression in the developing limb and fin and is associated with preaxial polydactyly. Hum. Mol. Genet.12, 1725–1735. 10.1093/hmg/ddg180 (2003). [DOI] [PubMed] [Google Scholar]
  • 27.Robin, X. et al. pROC: an open-source package for R and S + to analyze and compare ROC curves. BMC Bioinform.12, 77. 10.1186/1471-2105-12-77 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Shen, J. et al. An 11-Gene signature based on treatment responsiveness predicts Radiation Therapy Survival Benefit among breast Cancer patients. Front. Oncol.11, 816053. 10.3389/fonc.2021.816053 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.McKelvey, K. J., Hudson, A. L., Back, M., Eade, T. & Diakos, C. I. Radiation, inflammation and the immune response in cancer. Mamm. Genome. 29, 843–865. 10.1007/s00335-018-9777-0 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Schaue, D. et al. Radiation and inflammation. Semin Radiat. Oncol.25, 4–10. 10.1016/j.semradonc.2014.07.007 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Li, H., Wang, F., Guo, X. & Jiang, Y. Decreased MEF2A expression regulated by its enhancer methylation inhibits autophagy and may play an important role in the progression of Alzheimer’s Disease. Front. Neurosci.15, 682247. 10.3389/fnins.2021.682247 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Ji, J. et al. Genome–wide DNA methylation regulation analysis of long non–coding RNAs in glioblastoma. Int. J. Mol. Med.46, 224–238. 10.3892/ijmm.2020.4579 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Aran, D., Sabato, S. & Hellman, A. DNA methylation of distal regulatory sites characterizes dysregulation of cancer genes. Genome Biol.14 (R21). 10.1186/gb-2013-14-3-r21 (2013). [DOI] [PMC free article] [PubMed]
  • 34.Chen, G. et al. mRNA and lncRNA expression profiling of Radiation-Induced Gastric Injury reveals potential Radiation-Responsive transcription factors. Dose Response. 17, 1559325819886766. 10.1177/1559325819886766 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Cao, Y. et al. Identification of prognostic biomarkers in glioblastoma using a long non-coding RNA-mediated, competitive endogenous RNA network. Oncotarget7, 41737–41747. 10.18632/oncotarget.9569 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Barbarino, J. M., Whirl-Carrillo, M., Altman, R. B. & Klein, T. E. PharmGKB: a worldwide resource for pharmacogenomic information. Wiley Interdiscip Rev. Syst. Biol. Med.10, e1417. 10.1002/wsbm.1417 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Zhao, W., Goswami, P. C. & Robbins, M. E. Radiation-induced up-regulation of Mmp2 involves increased mRNA stability, redox modulation, and MAPK activation. Radiat. Res.161, 418–429. 10.1667/3155 (2004). [DOI] [PubMed] [Google Scholar]
  • 38.Sawaya, R. et al. Induction of tissue-type plasminogen activator and 72-kDa type-IV collagenase by ionizing radiation in rat astrocytes. Int. J. Cancer. 56, 214–218. 10.1002/ijc.2910560212 (1994). [DOI] [PubMed] [Google Scholar]
  • 39.Shailender, G., Kumari, S., Kiranmayi, P. & Malla, R. R. Effect of MMP-2 gene silencing on radiation-induced DNA damage in human normal dermal fibroblasts and breast cancer cells. Genes Environ.41, 16. 10.1186/s41021-019-0131-x (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hsu, H. H. et al. Taiwanin E inhibits cell migration in human LoVo colon cancer cells by suppressing MMP-2/9 expression via p38 MAPK pathway. Environ. Toxicol.32, 2021–2031. 10.1002/tox.22379 (2017). [DOI] [PubMed] [Google Scholar]
  • 41.Jie, Z. et al. SREBP-2 aggravates breast cancer associated osteolysis by promoting osteoclastogenesis and breast cancer metastasis. Biochim. Biophys. Acta Mol. Basis Dis.1865, 115–125. 10.1016/j.bbadis.2018.10.026 (2019). [DOI] [PubMed] [Google Scholar]
  • 42.Bartis, D. et al. Role of CD248 as a potential severity marker in idiopathic pulmonary fibrosis. BMC Pulm Med.1610.1186/s12890-016-0211-7 (2016). [DOI] [PMC free article] [PubMed]
  • 43.Matsushima, S. et al. CD248 and integrin alpha-8 are candidate markers for differentiating lung fibroblast subtypes. BMC Pulm Med.2010.1186/s12890-020-1054-9 (2020). [DOI] [PMC free article] [PubMed]
  • 44.Chen, Z., Wu, Z. & Ning, W. Advances in Molecular mechanisms and Treatment of Radiation-Induced Pulmonary Fibrosis. Transl Oncol.12, 162–169. 10.1016/j.tranon.2018.09.009 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Li, T. et al. A web server for Comprehensive Analysis of Tumor-infiltrating Immune cells. Cancer Res.77, e108–e110. 10.1158/0008-5472.CAN-17-0307 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Ru, B. et al. TISIDB: an integrated repository portal for tumor-immune system interactions. Bioinformatics35, 4200–4202. 10.1093/bioinformatics/btz210 (2019). [DOI] [PubMed] [Google Scholar]
  • 47.Azimian, H., Dayyani, M., Toossi, M. T. B. & Mahmoudi, M. Bax/Bcl-2 expression ratio in prediction of response to breast cancer radiotherapy. Iran. J. Basic. Med. Sci.21, 325–332. 10.22038/IJBMS.2018.26179.6429 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Arvold, N. D. et al. Age, breast cancer subtype approximation, and local recurrence after breast-conserving therapy. J. Clin. Oncol.29, 3885–3891. 10.1200/JCO.2011.36.1105 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Meehan, J. et al. Precision Medicine and the role of biomarkers of Radiotherapy response in breast Cancer. Front. Oncol.10, 628. 10.3389/fonc.2020.00628 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Impicciatore, G., Sancilio, S., Miscia, S. & Di Pietro, R. Nutlins and ionizing radiation in cancer therapy. Curr. Pharm. Des.16, 1427–1442. 10.2174/138161210791033932 (2010). [DOI] [PubMed] [Google Scholar]
  • 51.Perry, M. E. Mdm2 in the response to radiation. Mol. Cancer Res.2, 9–19 (2004). [PubMed] [Google Scholar]
  • 52.Nie, Y. H. et al. Analysis of mRNA expression patterns in Peripheral Blood cells of 3 patients with Cancer after the First Fraction of 2 gy irradiation: an Integrated Case Report and systematic review. Dose Response. 17, 1559325819833474. 10.1177/1559325819833474 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Charpentier, M., Spada, S., Van Nest, S. J. & Demaria, S. Radiation therapy-induced remodeling of the tumor immune microenvironment. Semin Cancer Biol.86, 737–747. 10.1016/j.semcancer.2022.04.003 (2022). [DOI] [PubMed] [Google Scholar]
  • 54.Miousse, I. R., Kutanzi, K. R. & Koturbash, I. Effects of ionizing radiation on DNA methylation: from experimental biology to clinical applications. Int. J. Radiat. Biol.93, 457–469. 10.1080/09553002.2017.1287454 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Chen, X. et al. Analysis of DNA methylation and gene expression in radiation-resistant head and neck tumors. Epigenetics10, 545–561. 10.1080/15592294.2015.1048953 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Li, X. et al. Effects of Radiation on Drug Metabolism: a review. Curr. Drug Metab.20, 350–360. 10.2174/1389200220666190405171303 (2019). [DOI] [PubMed] [Google Scholar]
  • 57.Tokizane, T. et al. Cytochrome P450 1B1 is overexpressed and regulated by hypomethylation in prostate cancer. Clin. Cancer Res.11, 5793–5801. 10.1158/1078-0432.CCR-04-2545 (2005). [DOI] [PubMed] [Google Scholar]
  • 58.Chen, X., Bai, G. & Scholl, T. O. Spontaneous Preterm Delivery, particularly with reduced fetal growth, is Associated with DNA hypomethylation of Tumor related genes. J. Pregnancy Child. Health. 310.4172/2376-127X.1000215 (2016). [DOI] [PMC free article] [PubMed]
  • 59.Xia, Y. et al. APC2 and CYP1B1 methylation changes in the bone marrow of acute myeloid leukemia patients during chemotherapy. Exp. Ther. Med.12, 3047–3052. 10.3892/etm.2016.3719 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 60.Tapio, S. et al. Gene expression changes in male and female rhesus macaque 60 days after irradiation. Plos One. 1610.1371/journal.pone.0254344 (2021). [DOI] [PMC free article] [PubMed]

Associated Data

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

Supplementary Materials

Supplementary Material 1 (16.2KB, docx)
Supplementary Material 2 (26.7KB, docx)
Supplementary Material 3 (23.4KB, docx)
Supplementary Material 4 (40.6KB, xlsx)
Supplementary Material 5 (702.9KB, docx)

Data Availability Statement

The datasets used and/or analyzed during the present study are available from the corresponding author or GitHub (https://github.com/huai-jj/methylation.git) upon reasonable request.

Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

RESOURCES