Skip to main content
NCBI home page
The Yale Journal of Biology and Medicine logoLink to The Yale Journal of Biology and Medicine
. 2025 Jun 30;98(2):147–157. doi: 10.59249/YYQN6894

Cell-Derived Epigenomic Markers of Chronic Stress Distinguish Breast Cancer Phenotypes

Amrita Acharya a, Anthony S Zannas a,b,c,*
PMCID: PMC12204029  PMID: 40589942

Abstract

Breast cancer (BC) is a highly prevalent malignancy in women and is often resistant to available therapies, calling for urgent investigation of the molecular mechanisms underlying its pathogenesis and progression. BC is thought to result from a complex interplay between genetic and environmental factors. Among key factors, chronic stress has been associated with worse cancer outcomes and can profoundly impact the epigenome. However, both stress and BC phenotypes are complex and heterogeneous, making studies that examine their molecular links challenging. Despite their heterogeneity, stressors trigger a neuroendocrine response that in humans culminates in the release of cortisol, a highly lipophilic hormone that traverses essentially every cell and induces widespread genomic effects. Modeling such effects at the epigenetic level, here we examine whether cellular DNA methylation (DNAm) markers of chronic stress – derived from human fibroblasts undergoing prolonged exposure to physiological stress cortisol levels – distinguish BC phenotypes in two independent human cohorts. Our results show that methylomic signatures of stress are consistently higher in tumor samples as compared to normal samples and in more advanced tumor stages and grades. Follow-up analyses further identify specific DNAm sites driving these associations, which are significantly enriched for cell adhesion pathways in both cohorts. These findings provide insights into the molecular mechanisms linking stress with BC and a proof-of-concept for utilizing cell model-derived disease biomarkers in environmental epigenetics.

Keywords: cell adhesion, DNA methylation, environmental stress, glucocorticoid, methylation risk score

Introduction

Breast cancer (BC) is a highly prevalent malignancy in women worldwide and the second leading cause of cancer-related deaths in women in the United States [1,2]. BC is thought to result from a complex genome-environment interplay [3]. Various efforts have been made to understand the role of genetics in BC, thereby aiding in the development of treatments specific to molecular types and subtypes of the disease [4-7]. Despite these advancements, BC is often resistant to available therapies, calling for urgent investigation of the molecular mechanisms underlying its pathogenesis and progression. In particular, epigenetic mechanisms have been proposed as highly relevant to advancing BC prevention and treatment [8].

Epigenetic modifications involve chemical changes that regulate the gene expression profile of a cell in accordance with the environment, ultimately contributing to the cell’s phenotype [9]. Among key factors, stress exposure – especially when chronic – has been associated with worse cancer outcomes [8,10-16] and can profoundly impact the epigenome [17,18]. Stress can result from a host of heterogeneous stimuli, including physical, mental, and social stressors. Despite their heterogeneity, these stressors commonly trigger a neuroendocrine response that in humans culminates in the release of cortisol, a highly lipophilic hormone that traverses essentially every cell and induces widespread genomic effects by activating the glucocorticoid receptor (GR) [19,20]. Importantly, the stress and glucocorticoid-driven GR activation have been shown to induce epigenetic effects relevant to cell function and disease phenotypes [21-25]. While the responses to stress are adaptive when of limited magnitude and duration, they can have detrimental consequences for health when excessive or chronic. In particular, the epigenetic and cellular changes that occur during excessive or prolonged responses to stress and cortisol exposure can promote tumorigenesis, angiogenesis, resistance to apoptosis, and metastasis of certain cancers, including BC [8,10-12,26].

Both stress and BC phenotypes are complex and heterogeneous, making studies that examine their molecular links challenging. Previous studies in BC and other fields have addressed such complexity by developing cell-derived disease biomarkers [27,28], but there is a paucity of such markers in environmental epigenetics. Using for the first time a human cell (fibroblast) model of chronic stress, we recently showed that prolonged exposure to physiological stress levels of cortisol drives widespread but genomic context-dependent changes in DNA methylation (DNAm) – a critical epigenetic modification whereby a methyl group is commonly added onto cytosine-guanine (CpG) sites [21]. Here, we leveraged these cell model-derived DNAm signatures to develop epigenomic markers of chronic stress and quantify their ability to distinguish BC phenotypes in two independent human cohorts.

Materials and Methods

Overview

An overview of the methodological approach used in this study is provided in Appendix A: Supplementary Figure 1.

Figure 1.

Figure 1

Open in a new tab

Comparison of methylation risk scores between normal and tumor samples in the MPBC cohort. A. Cortisol-related β = 1.3×10-3, SE = 4.2×10-4, z = 3.029; p = 2.5×10-3; GR-specific β = 2.5×10-3, SE = 8×10-4, z = 3.095, p = 2×10-3. B. Receiver operating characteristic curves depicting how MRSs distinguish normal from tumor samples: Cortisol-related area under the curve (AUC) = 0.790; GR-specific AUC = 0.793.

Datasets and Data Preprocessing

Cell Model-Derived Epigenomic Markers of Chronic Stress: Composite epigenomic markers were derived from a cell model of chronic stress, the experimental details of which have been previously described [21]. To dissect the epigenetic effects of stress, we used the well-characterized IMR-90 (human fetal lung fibroblast) cells, which can be cultivated for a prolonged time yet have a finite replicative potential that physiologically models the impact of chronic stress at the cellular level. IMR-90 cells were cultured and treated continuously with 100 nM cortisol (similar to levels seen in humans during in vivo stress) and/or 500 nM relacorilant (sufficient concentration to inhibit GR activation) for 81 days to observe any changes in cell phenotypes for as long as possible throughout the cellular lifespan [21]. Cells were kept in Dulbecco’s Modified Eagle Medium without phenol red and supplemented with 15% fetal bovine serum, high glucose, L-glutamine, sodium pyruvate, non-essential amino acids, and Antibiotic-Antimycotic [21]. Cell viability was assessed at every passage and remained high (>90%) throughout the experiment. Since the effects on cell proliferation and migration were abrogated when cells were also treated with relacorilant, we concluded that GR activation is likely the cause of stress-driven cell phenotypes. DNA was also extracted and DNAm levels were measured using the Illumina Infinium Human MethylationEPIC BeadChip, which assesses over 850 000 CpG sites across the genome. After standard quality control steps, a total of 709 105 CpG sites were included in downstream analyses. The raw DNAm data and related cell phenotypes have been deposited into NCBI GEO (GSE210304).

Breast Cancer Cohorts and Phenotypes: Two online-available BC datasets, TCGA-BRCA (referred as BRCA hereafter) and Molecular Profiles of Human Breast Cancer (MPBC), were downloaded from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO), respectively [29-34].

In the BRCA cohort, samples were collected from patients who were newly diagnosed with invasive breast carcinoma, had no prior treatment, and were undergoing surgery to treat the tumor. Clinical data about tumor stage and its fields were collected according to the American Joint Committee on Cancer (AJCC) standards.

In the MPBC cohort, all samples were collected from patients in hospitals in Baltimore, Maryland, between 1993 and 2003, and information about race/ethnicity, tumor stage (according to the AJCC and tumor–node–metastasis (TNM) system), and tumor grade (according to the Nottingham system) were collected.

BRCA: DNAm beta values were downloaded from the BRCA dataset using the GDC Data Transfer Tool [35]. All downloaded files were combined into a database using Python. Male patients and samples with incomplete staging or grading information were removed from the BRCA dataset, resulting in 878 samples from 775 patients that were included in downstream analyses. To maximize analytic power, subcategories for stages were grouped into larger categories – Stage I, II, III, and IV.

MPBC: The DNAm data files and clinical data for 72 samples were downloaded for the MPBC cohort, and to make datasets comparable across cohorts, M-values were converted to beta-values using the formula: Beta i = 2M i / (2M i + 1). To increase analytic power, the subcategories for stages were grouped into larger categories – Stage I, II, and III. Two samples that had missing ages were removed from the MPBC dataset, which included a total of 70 samples taken from 63 patients (seven patients provided both normal and tumor samples).

DNAm data in both cohorts was collected using the Illumina Infinium HumanMethylation450K BeadChip.

Preprocessing (CpG Filtering) and Methylation Risk Score Calculation: To capture the epigenomic effects of chronic stress, two methylation risk scores (MRSs) were generated: i) cortisol-related MRS, which is derived from all CpG sites modulated by prolonged cortisol exposure; and ii) GR-specific MRS, which is derived from CpGs specifically driven by GR activation, ie, DNAm changes induced by cortisol but abrogated by concomitant relacorilant treatment. The list of GR-specific CpGs (21 393 sites) has been previously published [21]. The cortisol-related CpG sites were identified from all available (709 105) CpG sites by filtering those with cortisol-driven DNAm change > 0.05 and FDR-adjusted p-value < 0.01, resulting in a total of 122 421 CpGs. For both MRSs, CpG weights were defined based on the magnitude (t-statistic) of glucocorticoid effects. These lists were joined with the 485 577 CpG sites available in each sample for both BRCA and MPBC, resulting in 55 464 cortisol-related sites and 9667 GR-specific sites for each cohort. Missing DNAm beta values in both datasets were imputed using the impute.knn method from the impute R package. Each MRS was calculated by multiplying the t-value associated with each CpG site by that site’s beta value and summing the resultant products for each sample.

Statistical Analysis

The R programming language (version 4.4.1) was used for all statistical analyses. Logistic regression was used for the MPBC dataset to test each MRS (cortisol-related, GR-specific) as the explanatory variable and tumor/normal status as the response variable. Polynomial regression models were created for both the BRCA and MPBC datasets with cortisol-related or GR-specific MRSs as the explanatory variable and tumor stage as the response variable. Lastly, polynomial regression models were created for the MPBC dataset with cortisol-related or GR-specific MRSs as the explanatory variable and tumor grade as the response variable. Age and race were included as covariates in all models. Additional analyses included covariates available in each dataset: ethnicity, radiation therapy, and pharmacotherapy status in BRCA; neoadjuvant therapy, chemotherapy, and hormone therapy status in MPBC.

Follow-up analyses aimed to discover and replicate any specific CpG sites underlying associations of each MRS with BC stage (available in both cohorts). Using the BRCA data, polynomial regression models were created for each CpG site for each MRS with BC stage as the response variable and the CpG methylation (beta value) as the explanatory variable, again with age and race as covariates. Any CpG sites that remained significant with stage after FDR correction were tested independently in the MPBC using the same regression model. Throughout these analyses, an FDR-adjusted p-value of 0.05 was set a priori as the cutoff of statistical significance.

Significant sites for BC stage were analyzed for any enriched pathways in WebGestalt (https://2019.webgestalt.org/). An over-representation analysis was conducted using the GLAD4U disease database. All the genes that were associated with the significant sites were run against a reference list of all the genes covered by the CpG sites on the Illumina Infinium HumanMethylation450K BeadChip, and significant pathways after FDR-correction (adjusted p < 0.05) were identified.

Results

Cohort Characteristics

All BRCA participants were females, with a median age of 58. The minimum, 1st quartile, mean, 3rd quartile, and maximum age of the BRCA cohort were 26, 48, 58, 67, and 90 respectively. Most were European American or White and many had a tumor stage II, with less individuals having stages I, III, and IV (Table 1).

Table 1.

Characteristics BRCA (n=775) MPBC (n=63)
Age
Minimum 26.0 30.0
1st Quartile 48.0 43.0
Mean 58.0 54.1
3rd Quartile 67.0 64.0
Maximum 90.0 93.0
Race
American Indian or Alaskan Native 1 NA
Asian 38 NA
Black or African American 154 30
White or European American 567 33
Missing 15 NA
Ethnicity
Hispanic or Latino 36 NA
Not Hispanic or Latino 668 NA
Missing 71
Normal or tumor samples Total: 878 Total: 70
Normal: 0 Normal: 10
Tumor: 878 Tumor: 60
Tumor stage*
I 142 6
II 499 39
III 219 15
IV 11 NA
Tumor grade
1: 7
2: 17
3: 28
NA: 18
Open in a new tab

*American Joint Committee on Cancer (AJCC) Pathologic Stage

All patients in MPBC were females, with a median age of 50.5. The minimum, 1st quartile, mean, 3rd quartile, and maximum age of the MPBC cohort were 30, 43, 54.1, 64, and 93 respectively. MPBC had balanced inclusion of European American or White and African American or Black participants. Most participants had tumor stage II and grade 2 or 3 (Table 1).

Cell Model-Derived MRSs Distinguish Breast Cancer Phenotypes

Our analyses in cultured fibroblasts showed that prolonged cortisol robustly increases values for both MRSs (Cortisol-related p = 5.3×10-8, GR-specific p = 7.8×10-6; Appendix A: Supplementary Figure 2), thus supporting higher MRSs as an epigenomic signature of chronic stress hormone exposure.

To determine the extent to which epigenomic markers of chronic stress can distinguish BC phenotypes, cortisol-related and GR-specific MRSs were then compared between normal and tumor samples and across tumor stages and grades in both cohorts. Following covariate adjustment (see Methods), tumor samples had significantly higher values for both MRSs as compared to normal samples (Cortisol-related p = 2.5×10-3, GR-specific p = 2×10-3; Figure 1A). Differences in MRS followed the same pattern when comparing only the seven MPBC participants that had both normal and tumor samples taken, though these comparisons did not reach statistical significance due to the small sample size (Cortisol-related p = 0.36, GR-specific p = 0.37; Appendix A: Supplementary Figure 3). Receiver operating characteristic curves demonstrated good MRS performance in distinguishing normal from tumor samples (Figure 1B).

Furthermore, both MRSs significantly differed across tumor stages in BRCA (with age and race: Cortisol-related p = 2.5×10-3, GR-specific p = 1.6×10-3; with age, race, ethnicity, and therapy: Cortisol-related p = 1.4×10-2, GR-specific p = 9.6×10-3) and showed a similar pattern that did not reach statistical significance in the smaller MPBC cohort (with age and race: Cortisol-related: p = 0.29, GR-specific: p = 0.35; with age, race, ethnicity, and therapy: Cortisol-related p = 0.62, GR-specific p = 0.84). In both cohorts, higher scores were observed at higher stages (Figure 2A,B), suggesting that epigenetic signatures of chronic stress are associated with higher BC severity. In line with these findings, both MRSs significantly differed across tumor grades in the MPBC cohort (with age and race: Cortisol-related p = 1.2×10-2, GR-specific p = 7.1×10-3; with age, race, ethnicity, and therapy: Cortisol-related p = 2.5×10-2, GR-specific p = 1.6×10-2). Again, higher scores were consistently observed at higher tumor grades (Figure 3).

Figure 2.

Figure 2

Open in a new tab

Comparison of methylation risk scores for tumor stages. A. BRCA cohort: Cortisol-related: β = 0.2079, SE = 6.9×10-2, t = 3.030, p = 2.5×10-3; GR-specific: β = 0.2179, SE = 6.9×10-2, t = 3.174, p = 1.6×10-3. B. MPBC cohort: Cortisol-related: β = 0.3102, SE = 0.2894, t = 1.072, p = 0.29; GR-specific: β = 0.2770, SE = 0.2944, t = 0.9408, p = 0.35.

Figure 3.

Figure 3

Open in a new tab

Methylation risk scores for tumor grade in the MPBC cohort. Cortisol-related: β = 1.033, SE = 0.3961, t = 2.607, p = 1.2×10-2; GR-specific: β = 1.153, SE = 0.4097, t = 2.815, p = 7.1×10-3

CpG-specific and Pathway Analyses

To discover and replicate specific gene loci that may underlie the MRS-BC associations, follow-up analyses tested whether each CpG comprising the MRSs was associated with BC stage (available in both cohorts). Following covariate adjustment and FDR correction for multiple testing (see Methods), 119 cortisol-related and 13 GR-specific CpGs were statistically significant in the BRCA. Among these associations, 32 cortisol-related and four GR-specific CpGs were replicated in the MPBC cohort (Appendix B: Supplementary File). Enrichment analysis for the cortisol-related sites identified “Adhesion” as the only significant pathway in both BRCA (9.9-fold enrichment, FDR-adjusted p = 1.2x10-12) and MPBC (5.5-fold enrichment, FDR-adjusted p = 1.1x10-8). This enrichment was driven by genes encoding proteins with critical roles in cell adhesion, such as protein kinase B (AKT1), adherens junctions associated protein 1 (AJAP1), beta-parvin (PARVB), and protocadherins (eg, PCDHGA1) [36-38]. The small number of significant CpGs did not allow us to perform similar pathway enrichment analysis for the GR-specific sites.

Discussion

Prior studies have linked chronic stress with heightened cancer risk and worse outcomes [8,10-16], but the underlying molecular mechanisms are unclear. While stress impacts the epigenome [17,18], both stress and BC phenotypes are highly complex and heterogeneous, making studies that examine their molecular links challenging. To address this challenge, here we generated cell model-derived MRSs of chronic stress and applied them in two independent human cohorts with available DNAm data and BC phenotypes.

More specifically, we generated two MRSs, the first reflecting cell exposure to physiological stress levels of cortisol and the second capturing epigenomic signatures driven more specifically by GR activation. Both MRSs were significantly higher in chronically stressed cells and in tumor vs normal samples, indicating that stress-driven epigenetic changes are associated with BC risk. Additionally, our analyses identified and replicated a positive dose-response relationship between MRSs and BC stage and grade, irrespective of covariates, suggesting that stress-driven epigenetic changes are associated with worse BC phenotypes. Prior work has suggested that chronic stress and glucocorticoid exposure can promote tumor development and progression [39,40]. Accordingly, we previously showed that chronically stressed cells exhibit cancer-related phenotypes, including robustly increased proliferation and migration, which are accompanied by DNAm and mRNA changes at functionally relevant gene loci [21,41]. No significant difference was seen in the MRSs between different histological types of BC, suggesting that chronic stress is associated with BC outcomes irrespective of the exact cell type from which the cancer originates. Together, these findings support that stress-driven epigenomic signatures may hold promise as markers of BC risk and outcomes. Moreover, due to their reversible nature and amenability to environmental input, epigenetic modifications may be more tractable targets than genetic mutations [11], offering opportunities to supplement established hormone and chemotherapy strategies for BC [8,10,11].

Follow-up analyses of the individual CpG sites comprising each MRS identified 32 cortisol-related and four GR-specific CpG sites significantly associated with BC stage in both the BRCA and MPBC cohorts (Appendix B: Supplementary File). Pathway analysis further identified and replicated in both cohorts that the cortisol-related CpGs are enriched for cell adhesion, a process that plays a critical role in maintaining healthy interactions between cells and their extracellular matrix [42]. More specifically, this enrichment was driven by genes encoding proteins with critical roles in cell adhesion, such as protein kinase B (AKT1), adherens junctions associated protein 1 (AJAP1), beta-parvin (PARVB), and protocadherins (eg, PCDHGA1) [36-38]. Alterations in the expression of cell adhesion proteins have been shown to promote anchorage-independent cell survival and growth, contributing to cancer [42,43]. Together with our findings, these observations point to cell adhesion as a potentially critical biological pathway that can be epigenetically dysregulated by chronic stress and contribute to worse BC outcomes. However, this hypothesis will need to be mechanistically addressed by future cell culture and in vivo studies.

Strengths and limitations of this work should be highlighted. An innovation of this study is that it uses a well-characterized human cell (fibroblast) line to derive physiologically relevant epigenetic markers of chronic stress associated with BC phenotypes in two independent human cohorts. The strength of this approach is that it supports chronic stress-driven epigenetic signatures as promising markers that can be applied across tissues and conditions; however, its limitation is the lack of potentially more robust markers that could have been derived from a disease-relevant cancer cell line. Importantly, our cell model-derived markers consistently distinguished BC phenotypes in both discovery and replication cohorts. The enrichment of the adhesion pathway was replicated in both cohorts as well. However, causation cannot be inferred due to the lack of experimental in vivo data and unavailable information on potential confounder variables and longitudinal BC outcomes, thereby not allowing us to dissect the role of stress-driven epigenetic changes in cell adhesion and BC initiation vs progression. Furthermore, the small sample size of the MPBC cohort and the lack of data on detailed phenotypes or potential confounders, such as lifestyle parameters and stressful life events, are additional limitations. Future human cohort studies may address these limitations by collecting more detailed information on stress exposure and potential confounders, ideally in a longitudinal observation setting with a large sample size.

With these strengths and limitations in mind, the findings of the present study provide insights into the molecular mechanisms linking stress with BC and a proof-of-concept for utilizing cell model-derived disease biomarkers in environmental epigenetics.

Supplementary Material

Appendix A

Supplementary figures

yjbm_98_2_147_s01.pdf (186.6KB, pdf)
Appendix B

Supplementary file

yjbm_98_2_147_s02.xls (124KB, xls)

Acknowledgments

The results shown here are in part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga. The following cell line was obtained from the NIGMS Human Genetic Cell Repository at the Coriell Institute for Medical Research: I90-83.

Glossary

BC

Breast cancer

GR

glucocorticoid receptor

DNAm

DNA methylation

CpG

cytosine-guanine

TCGA

The Cancer Genome Atlas

GEO

Gene Expression Omnibus

BRCA

TCGA-BRCA

MPBC

Molecular Profiles of Human Breast Cancer

AJCC

American Joint Committee on Cancer

TNM

tumor–node–metastasis

MRSs

methylation risk scores

Author Contributions

Conceptualization and funding acquisition: ASZ; formal analysis: AA; writing—original draft preparation, review, and editing: AA and ASZ; All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Funding Statement

This research received no external funding.

Data Availability Statement

The BRCA cohort data was obtained from the Genomic Data Commons (GDC) Data Portal from the National Cancer Institute at https://portal.gdc.cancer.gov/. The MPBC cohort data was obtained from the Gene Expression Omnibus (GEO) under the accession number GSE37754, and the fibroblast data was obtained from the data under the accession number GSE210304.

References

  1. King L, Flaus A, Holian E, Golden A. Survival outcomes are associated with genomic instability in luminal breast cancers. PLoS One. 2021. Feb;16(2):e0245042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Giaquinto AN, Sung H, Newman LA, Freedman RA, Smith RA, Star J, et al. Breast cancer statistics 2024. CA Cancer J Clin. 2024;74(6):477–95. [DOI] [PubMed] [Google Scholar]
  3. Mbemi A, Khanna S, Njiki S, Yedjou CG, Tchounwou PB. Impact of Gene-Environment Interactions on Cancer Development. Int J Environ Res Public Health. 2020. Nov;17(21):8089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Sanchez-Vega F, Mina M, Armenia J, Chatila WK, Luna A, La KC, et al. Cancer Genome Atlas Research Network . Oncogenic Signaling Pathways in The Cancer Genome Atlas. Cell. 2018. Apr;173(2):321–337.e10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Pereira B, Chin SF, Rueda OM, Vollan HK, Provenzano E, Bardwell HA, et al. The somatic mutation profiles of 2,433 breast cancers refines their genomic and transcriptomic landscapes. Nat Commun. 2016. May;7:11479. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Wang Y, Li Y, Liu B, Song A. Identifying breast cancer subtypes associated modules and biomarkers by integrated bioinformatics analysis. Biosci Rep. 2021. Jan;41(1):BSR20203200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Nik-Zainal S, Davies H, Staaf J, Ramakrishna M, Glodzik D, Zou X, et al. Landscape of somatic mutations in 560 breast cancer whole-genome sequences. Nature. 2016. Jun;534(7605):47–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Schröder R, Illert AL, Erbes T, Flotho C, Lübbert M, Duque-Afonso J. The epigenetics of breast cancer - Opportunities for diagnostics, risk stratification and therapy. Epigenetics. 2022. Jun;17(6):612–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cavalli G, Heard E. Advances in epigenetics link genetics to the environment and disease. Nature. 2019. Jul;571(7766):489–99. [DOI] [PubMed] [Google Scholar]
  10. Ramírez-Expósito MJ, Dueñas-Rodríguez B, Carrera-González MP, Navarro-Cecilia J, Martínez-Martos JM. Circulating levels of β-endorphin and cortisol in breast cancer. Compr Psychoneuroendocrinol. 2021. Jan;5:100028. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Kim A, Mo K, Kwon H, Choe S, Park M, Kwak W, et al. Epigenetic Regulation in Breast Cancer: insights on Epidrugs. Epigenomes. 2023. Feb;7(1):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lempesis IG, Georgakopoulou VE, Papalexis P, Chrousos GP, Spandidos DA. Role of stress in the pathogenesis of cancer (Review) [Review]. Int J Oncol. 2023. Nov;63(5):124. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chiriac VF, Baban A, Dumitrascu DL. Psychological stress and breast cancer incidence: a systematic review. Clujul Med. 2018;91(1):18–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Bahri N, Fathi Najafi T, Homaei Shandiz F, Tohidinik HR, Khajavi A. The relation between stressful life events and breast cancer: a systematic review and meta-analysis of cohort studies. Breast Cancer Res Treat. 2019. Jul;176(1):53–61. [DOI] [PubMed] [Google Scholar]
  15. He XY, Gao Y, Ng D, Michalopoulou E, George S, Adrover JM, et al. Chronic stress increases metastasis via neutrophil-mediated changes to the microenvironment. Cancer Cell. 2024. Mar;42(3):474–486.e12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Cao Q, Zhao M, Su Y, Liu S, Lin Y, Da H, et al. Chronic Stress Dampens Lactobacillus Johnsonii-Mediated Tumor Suppression to Enhance Colorectal Cancer Progression. Cancer Res. 2024. Mar;84(5):771–84. [DOI] [PubMed] [Google Scholar]
  17. Zannas AS. Epigenetics as a key link between psychosocial stress and aging: concepts, evidence, mechanisms. Dialogues Clin Neurosci. 2019. Dec;21(4):389–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Cunliffe VT. The epigenetic impacts of social stress: how does social adversity become biologically embedded? Epigenomics. 2016. Dec;8(12):1653–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Nicolaides NC, Galata Z, Kino T, Chrousos GP, Charmandari E. The human glucocorticoid receptor: molecular basis of biologic function. Steroids. 2010. Jan;75(1):1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Chrousos GP, Gold PW. The concepts of stress and stress system disorders. Overview of physical and behavioral homeostasis. JAMA. 1992. Mar;267(9):1244–52. [PubMed] [Google Scholar]
  21. Leung CS, Kosyk O, Welter EM, Dietrich N, Archer TK, Zannas AS. Chronic stress-driven glucocorticoid receptor activation programs key cell phenotypes and functional epigenomic patterns in human fibroblasts. iScience. 2022. Aug;25(9):104960. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Thomassin H, Flavin M, Espinás ML, Grange T. Glucocorticoid-induced DNA demethylation and gene memory during development. EMBO J. 2001. Apr;20(8):1974–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Seifuddin F, Wand G, Cox O, Pirooznia M, Moody L, Yang X, et al. Genome-wide Methyl-Seq analysis of blood-brain targets of glucocorticoid exposure. Epigenetics. 2017. Aug;12(8):637–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Wiechmann T, Röh S, Sauer S, Czamara D, Arloth J, Ködel M, et al. Identification of dynamic glucocorticoid-induced methylation changes at the FKBP5 locus. Clin Epigenetics. 2019. May;11(1):83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Zannas AS, Jia M, Hafner K, Baumert J, Wiechmann T, Pape JC, et al. Epigenetic upregulation of FKBP5 by aging and stress contributes to NF-κB-driven inflammation and cardiovascular risk. Proc Natl Acad Sci USA. 2019. Jun;116(23):11370–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Flaherty RL, Intabli H, Falcinelli M, Bucca G, Hesketh A, Patel BA, et al. Stress hormone-mediated acceleration of breast cancer metastasis is halted by inhibition of nitric oxide synthase. Cancer Lett. 2019. Sep;459:59–71. [DOI] [PubMed] [Google Scholar]
  27. Mertens J, Wang QW, Kim Y, Yu DX, Pham S, Yang B, et al. Pharmacogenomics of Bipolar Disorder Study . Differential responses to lithium in hyperexcitable neurons from patients with bipolar disorder. Nature. 2015. Nov;527(7576):95–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Castro-Guijarro AC, Sanchez AM, Flamini MI. Potential Biomarkers Associated with Prognosis and Trastuzumab Response in HER2+ Breast Cancer. Cancers (Basel). 2023. Sep;15(17):4374. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ciriello G, Gatza ML, Beck AH, Wilkerson MD, Rhie SK, Pastore A, et al. TCGA Research Network . Comprehensive Molecular Portraits of Invasive Lobular Breast Cancer. Cell. 2015. Oct;163(2):506–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Koboldt DC, Fulton RS, McLellan MD, Schmidt H, Kalicki-Veizer J, McMichael JF, et al. Cancer Genome Atlas Network . Comprehensive molecular portraits of human breast tumours. Nature. 2012. Oct;490(7418):61–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Tang W, Zhou M, Dorsey TH, Prieto DA, Wang XW, Ruppin E, et al. Integrated proteotranscriptomics of breast cancer reveals globally increased protein-mRNA concordance associated with subtypes and survival. Genome Med. 2018. Dec;10(1):94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Terunuma A, Putluri N, Mishra P, Mathé EA, Dorsey TH, Yi M, et al. MYC-driven accumulation of 2-hydroxyglutarate is associated with breast cancer prognosis. J Clin Invest. 2014. Jan;124(1):398–412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Putluri N, Maity S, Kommagani R, Creighton CJ, Putluri V, Chen F, et al. Pathway-centric integrative analysis identifies RRM2 as a prognostic marker in breast cancer associated with poor survival and tamoxifen resistance. Neoplasia. 2014. May;16(5):390–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Edgar R, Domrachev M, Lash AE. Gene Expression Omnibus: NCBI gene expression and hybridization array data repository. Nucleic Acids Res. 2002. Jan;30(1):207–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Grossman RL, Heath AP, Ferretti V, Varmus HE, Lowy DR, Kibbe WA, et al. Toward a Shared Vision for Cancer Genomic Data. N Engl J Med. 2016. Sep;375(12):1109–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Honig B, Shapiro L. Adhesion protein structure, molecular affinities, and principles of cell-cell recognition. Cell. 2020. Apr;181(3):520–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Khwaja A, Rodriguez-Viciana P, Wennström S, Warne PH, Downward J. Matrix adhesion and Ras transformation both activate a phosphoinositide 3-OH kinase and protein kinase B/Akt cellular survival pathway. EMBO J. 1997. May;16(10):2783–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Chen L, Wang Y, Zhang B. Hypermethylation in the promoter region inhibits AJAP1 expression and activates the JAK/STAT pathway to promote prostate cancer cell migration and stem cell sphere formation. Pathol Res Pract. 2023. Jan;241:154224. [DOI] [PubMed] [Google Scholar]
  39. Brücher BL, Jamall IS. Epistemology of the Origin of Cancer II: Fibroblasts Are the First Cells to Undergo Neoplastic Transformation. Cell Physiol Biochem. 2023. Dec;57(6):512–37. [DOI] [PubMed] [Google Scholar]
  40. Yan M, Wang J, Wang H, Zhou J, Qi H, Naji Y, et al. Knockdown of NR3C1 inhibits the proliferation and migration of clear cell renal cell carcinoma through activating endoplasmic reticulum stress-mitophagy. J Transl Med. 2023. Oct;21(1):701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zannas AS. Naturalistic Stress Hormone Levels Drive Cumulative Epigenomic Changes along the Cellular Lifespan. Int J Mol Sci. 2021. Aug;22(16):8778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Janiszewska M, Primi MC, Izard T. Cell adhesion in cancer: beyond the migration of single cells. J Biol Chem. 2020. Feb;295(8):2495–505. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sousa B, Pereira J, Paredes J. The Crosstalk Between Cell Adhesion and Cancer Metabolism. Int J Mol Sci. 2019. Apr;20(8):1933. [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

Appendix A

Supplementary figures

yjbm_98_2_147_s01.pdf (186.6KB, pdf)
Appendix B

Supplementary file

yjbm_98_2_147_s02.xls (124KB, xls)

Data Availability Statement

The BRCA cohort data was obtained from the Genomic Data Commons (GDC) Data Portal from the National Cancer Institute at https://portal.gdc.cancer.gov/. The MPBC cohort data was obtained from the Gene Expression Omnibus (GEO) under the accession number GSE37754, and the fibroblast data was obtained from the data under the accession number GSE210304.

Articles from The Yale Journal of Biology and Medicine are provided here courtesy of Yale Journal of Biology and Medicine

RESOURCES