Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Rotating night work, lifestyle factors, obesity and promoter methylation in BRCA1 and BRCA2 genes among nurses and midwives

Rotating night work, lifestyle factors, obesity and promoter methylation in BRCA1 and BRCA2 genes among nurses and midwives

  • Beata Peplonska, 
  • Agnieszka Bukowska, 
  • Edyta Wieczorek, 
  • Monika Przybek, 
  • Shanbeh Zienolddiny, 
  • Edyta Reszka
PLOS
x

Abstract

Some recent evidence suggests that environmental and lifestyle factors may modify DNA methylation. We hypothesized that rotating night work and several modifiable factors may be associated with the methylation of the promoter regions within two tumor suppressor and DNA repair genes: BRCA1 and BRCA2. The methylation status of BRCA1 and BRCA2 was determined via qMSP reactions using DNA samples derived from blood leucocytes of 347 nurses and midwives working rotating nights and 363 working during the days. The subjects were classified into unmethylated vs methylated BRCA1 and BRCA2 when the methylation index was 0% or >0%, respectively. The adjusted odds ratios with 95% confidence intervals were calculated for night work status, smoking, obesity, physical activity and alcohol drinking. Current night shift work or night work history was not associated with methylation status of the promoter sites within BRCA1 and BRCA2 genes. We observed weak associations between smoking and the methylation status of BRCA1 with OR = 1.50 (95%CI: 0.98–2.29) for current smoking, OR = 1.83, 95CI: 1.08–3.13 for smoking longer than 31 years, and 0.1>p>0.05 for trends for the number of cigarettes per day, smoking duration and packyears. In conclusion, no links between night shift work and methylation of the promoter region within the BRCA1, and BRCA2 genes were observed in this exploratory analysis. The findings of our study weakly support the hypothesis that smoking may contribute to epigenetic events.

Introduction

Breast cancer is the most prevalent cancer among women worldwide. The etiology of breast cancer has not been fully explained; however, both the genetic and environmental factors are postulated to be playing a role. Of the genetic factors, the inherited germline mutations in two genes: BRCA1 and BRCA2, are the well-established risk factors of breast cancer, explaining approximately two percent of breast cancer cases [1] (up to 7 percent among Polish women diagnosed before age 50[2]). These two tumor suppressor genes are involved in genome stability, DNA repair and cell cycle control [3].

Of the modifiable risk factors, obesity after menopause [4] and alcohol drinking [5] have been recognized as potentially contributing to breast cancer, while physical activity has been proven to play some part in reducing breast cancer risk [6]. Epidemiological evidence has also implied a potentially causative role of active smoking, particularly long-term heavy smoking [7], and when initiated before first birth [8].

One of the occupational exposures that have been classified as probably carcinogenic is the shift work leading to circadian rhythm disruption [9]. A number of epidemiological studies were focused on a possible association between night shift work and breast cancer risk [1013]. The meta-analysis based on 12 case-control studies reported a 9% increase in the risk of breast cancer per 5 years of night work [10]. However, no increased risk of breast cancer was observed in the some recent prospective cohort studies [14]. Thus the link between night shift work and breast cancer risk still remains disputable.

One of the proposed mechanisms underlying the increased risk of breast cancer among night shift workers is a decrease in melatonin synthesis [1517] and an increase in reproductive hormones synthesis in response to light-at-night [18]. It has also been postulated that shift work may contribute to unhealthy changes in lifestyle [19], such as more common smoking [2034] and alcohol consumption [21,23,24,26,30,33,3538] as well as lower physical activity [27,33,35,39] and poor dietary habits [40], which may increase cancer risk. Moreover, night workers were found to be more likely to become obese than day workers [41,42].

Although the idea of shift work as being probably carcinogenic to humans is gaining widespread attention, there is a lack of experimental mechanistic evidence linking shift work to breast cancer.

Moreover, the biological and molecular mechanisms of shift work exposure have not been fully explored. Only recently have some in vivo animal models of human shift work been used to investigate the possible association between shift work and adverse health effects [43]. Epigenetic changes in the genome have been proposed as one of the biologically plausible mechanisms [4446]. A major epigenetic mechanism that may affect gene expression in the disease is the changes in the methylation level of 5-methylcytosine (5meC) in the regulatory regions (i.e. promoter, enhancer) of the genes. This may lead to an altered transcription of the regulatory genes, which can result in malignant cellular transformation.

Several studies have already investigated genome-wide CpG island methylation of the protein-coding genes and the regulatory microRNA (miRNA) genes [44,4749]Wee EJ, Peters K, Nair SS, Hulf T, Stein S, Wagner S, Bailey P, Lee SY, Qu WJ, Brewster B, et al. Oncogene. 2012 Sep 20; 31(38):4182–95 according to the shift work status. Their findings suggest that a long-term night shift work exposure may lead to methylation-dependent downregulation of the protein-coding genes and miRNA genes, which in turn may lead to the downregulation of several biological pathways that are important in breast carcinogenesis, such as DNA damage response and repair, oxidative stress response, inflammation, and immunomediated antitumor activity. Changes in promoter methylation in the CLOCK gene were also reported among long-term shift workers [44], which was consistent with the findings on a low level of the CLOCK methylation in breast cancer patients [50]. In vitro assays displayed a network of transcripts with an altered expression following CLOCK or CRY2 gene knockdown, including the genes involved in breast carcinogenesis [50,51].

The evidence for the role of environmental and lifestyle factors in DNA methylation has been mounting. Changes in the methylation pattern of the promoter within tumor suppressor genes have also been observed, with suggestive evidence supporting the hypothesis that not only the inherited mutations in the genes, but also their functional inactivation caused by the epigenetic events such as hypermethylation i.e. aberrant methylation of the CpG islands in the promoter region, may play a role in the etiology of breast cancer [52]. Some evidence points out that the epigenetic regulation of the BRCA1 gene may contribute to breast cancer risk [52,53].

Therefore, we hypothesized that the rotating night work and such modifiable factors as cigarette smoking, alcohol drinking, physical activity, and obesity, may be associated with the methylation of the promoter regions within two tumor suppressor and DNA repair genes: BRCA1 and BRCA2. To examine this hypothesis, we investigated associations between the methylation status of the CpG island in the promoter of the BRCA1 and BRCA2 genes and the lifestyle factors, using data obtained from a cross-sectional study on nurses and midwives.

Material and methods

This cross-sectional study was described elsewhere [34]. In short, the study included 725 nurses and midwives (84% of those eligible), aged 40–60 years, who were employed as a nurse or midwife in public health care settings in Lodz, Poland. The study participants were recruited based on the data derived from the Regional Registry of the Chamber of Nurses and Midwives in Lodz. The nurses currently working night shifts (n = 347) were employed according to the fast rotating system (12 h long shifts), with a night shift followed by a day off. Night duties started at 7 p.m. and ended at 7:00 a.m. while day shifts started at 7 a.m. and ended at 7 p.m., with no evening shifts. There were no permanent night workers in the study population.

Day workers (n = 363) worked mostly in the outpatient clinics and the shift usually lasted 7.5 hours, between 7 a.m. and 4 p.m. Among the subjects currently working day shifts, the majority had some history of night shift work, but most of them (83%) had resigned from night shifts more than five years before the onset of the present study.

A structured questionnaire was administered during in-person interviews that were carried out during the years 2008–2010, to elicit information on occupational history, demographics, medical and reproductive history, hormone use, physical activity, smoking, alcohol use, and diet; the latter based on the food frequency questionnaire (FFQ). Questions were asked about current cigarette smoking, and the amounts smoked within 5 years prior to the project; with the age categories starting at age <15 years. With regard to alcohol consumption, we asked questions about the number of glasses of beer (0.5 l), wine (100ml) or vodka (50ml) drunk on average per day (week, month or year) over the period of the preceding year, and we also asked about the number of alcoholic beverages drunk on average, by the age categories starting from the age of 15–17 years. Anthropometric measurements of body weight, height, waist and hip circumference were performed by trained nurses, and body mass index (BMI in kg/m2), and waist to hip ratio (WHR) were calculated.

Physical activity was assessed according to the International Physical Activity Questionnaire (IPAQ) [54], that included questions about four physical activity domains: leisure-time, occupational, transport-related and household activities. The details were described previously [55]. Based on the IPAQ guidelines, we calculated the MET-hours per week (MET—metabolic equivalent ratio of the metabolic rate to a standard resting metabolic rate of 1) for each category of physical activity. The total physical activity was calculated by summing the scores for all the four domains. Blood samples were collected from 710 women.

Between the years 2014 and 2015, the first follow-up was carried out among 632 women participants (87.2% of the original study population). At that time we collected information about the subjects’ diet, using the FFQ. Trained interviewers asked the participants about the frequency and amount of a usual intake of 151 nutritional items as well as about micronutrient and vitamin intake. To help the respondents assess the weight of the consumed foods, a photo book of food products and dishes was used, with pictures showing food items of different size and estimated weight attributed to each item in each size.

Both the original study and the follow-up were approved by the ethical institutional review board at the Nofer Institute of Occupational Medicine. A signed informed consent was obtained from each participant for each phase of the study.

Laboratory methods

Blood samples were collected into S-Monovette® heparinized test tubes in the morning hours (06:00–10:00 a.m.). The samples were preserved immediately after being delivered to the lab and stored at -20°C until DNA isolation. Genomic DNA was isolated from 710 whole blood samples using QIAamp DNA Blood Mini Kit (Qiagen), according to the manufacturer’s instructions.

The promoter region with the transcriptional start site of BRCA1 and BRCA2 was analyzed using DBTSS database (http://dbtss.hgc.jp) for further CpG island identification. Chemical modification of 500 ng of genomic DNA isolated from whole blood was performed with the use of Cells-to-CpG™ Bisulfite Conversion Kit (Thermo Fisher Scientific, Waltham, MA, USA). DNA methylation was analyzed using quantitative methylation-specific real-time PCR assay (qMSP) with FastStart Essential DNA Probes Master (Roche, Basel, Switzerland). Primers and MethyLight dual-labelled probe were designed using Beacon Designer 7.01 (PREMIER Biosoft Int., Palo Alto, CA, USA). BRCA1 and BRCA2 amplicons covered 11 and 12 CpG dinucleotides, respectively. The qMSPBRCA1 oligonucleotides were as follows: qMSP: 5’-GTATTTTGAGAGGTTGTTGTTTAGC-3’ (sense), 5’-CGTCCAAAAAATCTCAACGA-3’ (antisense) and 5’FAM-ACGCCGCGCAATCGCAATTTTAAT-3’BHQ1 (probe); qMSPBRCA2–5’-TTGAGAAATATTCGTAGCGGTTTATTTAGG-3’ (sense), 5’-CTAACCACGTAACGCCGTAACG-3’ (antisense) and 5’FAM-CACGCAACACACGCACCACCCGAAA-3’BHQ1 (probe). The qMSP was performed with internal reference genes ACTB and MYOD [56]. The Primers and the probe of the internal reference gene were located in the area without CpG nucleotides, thus amplifying the modified ACTB and MYOD gene independently of the methylation status of CpG nucleotides.

Water blanks were included in each plate to check possible contamination. Positive (fully CpG-methylated human genomic DNA), (NEB, Ipswich, MA, USA) and negative (5-Azadc treated human genomic DNA) (NEB, Ipswich, MA, USA) DNA controls were included in the methylation analysis. All the samples of 2μl of converted DNA were amplified with 0.6μM of primers and 0.2μM of probe in a 20μl reaction assay. To determine inter-assay variability, the qMSP for randomly selected samples was repeated. Inter-assay coefficients of variability (CV) were below 18% and intra-assay CV was below 8%. qMSP reactions were carried out on a 96-well plate in LC96 (Roche, Basel, Switzerland).

To calculate the methylation status of BRCA1 and BRCA2, reference gene-normalized relative methylation quantification, expressed as fully methylated reference (PMR—%), was used [56]. Briefly, the Ct values of the gene of interest were compared with the Ct values of the internal reference gene: PMR = 100% x 2-ΔΔCt (ΔΔCt = (Ct of target gene in sample–Ct of control gene in sample)—(Ct of target gene in methylated control–Ct of control gene in methylated control).

Statistical methods

Arithmetic means, with standard deviations and frequencies of the basic characteristics were calculated in the total population and according to the BRCA1 and BRCA2 methylation status.

To explore the association between lifestyle factors, current rotating night shift work and its duration and the BRCA1 or BRCA2 methylation status, a logistic regression model was used. The population was dichotomized into unmethylated vs methylated BRCA1 and BRCA2 when the methylation index was 0% or >0% respectively. Apriori list of the potential confounders was specified, and the subjects’ age and folate intake were included in all models as covariates. Folate intake was considered as a potentially important confounder given the key role of folate in the methyl metabolism pathway and demonstrated DNA hypomethylation in response to folate deficiency [57].

The following categories of descriptive variables were used: current rotating night work (yes, no), duration of night work in years (≤ 10, >10-≤20, >20), smoking (never, current, past); number of cigarettes smoked per day among current smokers (1–4, 5–14, >15); smoking duration (tertiles: ≤22.6, >22.6-≤31, >31 years); packyears (tertiles: ≤10.4, >10.4 - ≤17, >17); average alcohol drinking, in drinks per week (≤0.25, >0.25-≤1, >1); lifetime duration of alcohol drinking, in years (≤20, >20-≤30, >30); drinkyears—cumulative measure of alcohol drinking, calculated as the sum of the average number of drinks multiplied by the duration of drinking for each age category from the age of 15 years up to the age at the interview, expressed as drinks per week*years (≤5, >5-≤10, >10); current alcohol abstinence (yes vs no); BMI (<25, ≥25-<30, ≥30kg/m2); WHR (≤0.85, >0.85); total physical activity in MET*hr/wk (≤155, >155- ≤220, >220); and recreational physical activity (none vs any).

The odds ratios with 95% confidence intervals were calculated for each level of the lifestyle characteristics in the total study population. The basic model included adjustment for age and folate intake, while the extended one included mutual adjustment for all the other lifestyle characteristics, to gain information about the importance of the covariates. Trends were tested for continuous descriptive variables.

We also examined ORs for the subgroups of women characterized by the system of work and the category of lifestyle factors in order to determine the combined effect of these two factors. Unfortunately the analyses testing interactions between lifestyle factors and current work system were generally underpowered and thus their results are not reported.

Considering the missing data for folate intake for 91 women (13% of total) we imputed the values based on estimates derived from the FFQ at baseline study (FFQ1) and at follow-up (FFQ2). To this end, the linear regression model was fitted with the total amount of folate intake (in grams) as the dependent variable, and the frequency of food item intake, according to the FFQ1, and the average folate intake per a given food item (in grams per 100 g of a given food item), determined via FFQ2, as the independent variables.

Statistical analyses were performed using R version 3.1.1 (Vienna, Austria) and STATA 11 (StataCorp LP).

Results

The general characteristics of the study population by BRCA1 and BRCA2 methylation status are displayed in Table 1. The average age of the study participants was 49 years, and age showed inverse and significant association with BRCA2 methylation (β-coef. = -0.039, 95%CI:-.0.075–-0.003). The study included roughly equal proportions of nurses currently working according to rotating night system or on day shifts (49% vs 51%). There was a high rate (49%) of subjects with a long lifetime duration of night work (>20 years) in the total population (80.4% among night shift workers and 18.5% among current day workers). As much as 30.6% of the subjects reported current smoking, while 42.5% had never smoked, and 27% were ex-smokers. Of the current smokers, roughly 41% smoked more than 15 cigarettes per day, and the majority (78%) were long-term smokers (>20 years). The subjects generally reported drinking small amounts of alcohol, with an average of 0.6 drinks per week, and 5% reported current abstinence. As much as 63% of the subjects were overweight or obese (BMI>25), and abdominal obesity (WHR.0.85) was observed among 25% of the subjects. About 28% reported no leisure-time physical activity, while their total physical activity was high (221 MET*hr per week). In most of the women studied, the promoter regions of the BRCA1 and BRCA2 genes were unmethylated (78% and 82%, respectively). The subjects’ characteristics according to the current work system are provided in S1 Table. Current night workers were on average two years younger than current day workers (p<0.001). As many as 80% of current night workers reported more than 20 years of night work (vs. 19% among current day workers), while 49% of day workers reported night work of up to 10 years (vs 3% of current night workers). More current night workers than day workers reported current smoking (34.6% vs. 26.7% respectively), with higher frequency of smokers, who smoked for 20 years or longer among ever smokers (82.6% vs 73.0%), and higher number of the packyears (9.9 vs 8.5, p = 0.007). The total physical activity was higher among night workers (241 vs. 202 MET*hr per week, p<0.001), and recreational physical inactivity was more common among night workers(p = 0.02).

thumbnail
Table 1. Selected characteristics of the studied population of nurses and midwives in the cross-sectional study.

https://doi.org/10.1371/journal.pone.0178792.t001

Neither the current night work status nor night work duration showed any association with the methylation status of the promoter site in the BRCA1 gene (Table 2). Of the lifestyle factors, smoking was found to be significantly associated with BRCA1 methylation, with OR = 1.52, 95%CI: 1.00–2.30 among current smokers compared to never-smokers. This results was slightly attenuated in the model with the other lifestyle factors included as covariates. No clear pattern was found for the relationship between the number of cigarettes smoked per day among current smokers and the methylation status of BRCA1. The OR was increased significantly among the subjects smoking 5–14 cigarettes per day (OR = 1.69, 95%CI:1.03–2.80), but not among those who reported smoking more than 15 cigarettes per day (p- trend = 0.053). The OR of methylated BRCA1 was significantly increased among those who reported smoking longer than 31 years (OR = 1.83, 95CI: 1.08–3.13), with a trend of borderline significance (p = 0.070). The relationship between the the number of packyears and the methylation of the promoter of BRCA1 was slightly weaker, with the OR = 1.55, 95%CI:0.97–2.47 for the highest tertile and p for trend of 0.082. BMI and other anthropometric measures were not associated with the methylation status of the BRCA1 gene. Neither night work nor smoking or alcohol drinking showed any association with the methylation status of the BRCA2 gene (Table 3). Likewise, no effect for the methylation status of BRCA2 could be noted for obesity or physical inactivity in the total study population.

thumbnail
Table 2. The association between methylation of BRCA1 gene and lifestyle factors among nurses and midwives in the cross-sectional study.

https://doi.org/10.1371/journal.pone.0178792.t002

thumbnail
Table 3. The association between methylation of BRCA2 gene and lifestyle factors among nurses and midwives in the cross-sectional study.

https://doi.org/10.1371/journal.pone.0178792.t003

Discussion

In the reported cross-sectional study of nurses and midwives, we examined several determinants of the methylation status in two suppressor genes, namely BRCA1 and BRCA2, whose silencing predisposes to breast cancer. We sought associations for the rotating night work, cigarette smoking, alcohol consumption, physical activity and two anthropometric measures: BMI and WHR. The study found no associations for rotating night shift while suggested a positive association between the methylation status of the BRCA1 gene and current smoking, particularly among moderate smokers (5–14 cigarettes per day), and long term smokers (>31 years). No other significant associations were noted.

Although the research concerning the epigenetic effects of smoking, mostly in the context of cancer (e.g. lung cancer for which smoking is a strong risk factor), has been relatively abundant, we were able to identify only a few studies focused on associations between smoking and the methylation status [58] within the promoter regions of BRCA1, and in fact none for BRCA2. In a recent study conducted among monozygotic twins (21 couples) with discordant smoking habits, no significant difference in the methylation status of the promoter site of BRCA1 was noted between smokers and nonsmokers [59]. The odds ratio of BRCA1 promoter hypermethylation was increased among the subjects who smoked for more than 10 years (OR = 1.71, 95%CI: 0.09–31.9) but the result was insignificant. The study was limited by its relatively small population size, including mostly light smokers (<15 cigarettes per day) with a short duration of the smoking habit (mean 10 years). In another study, neither active smoking (current or ever) nor passive smoking was related to the methylation status within the promoter site of BRCA1 when measured in tumor tissue from breast cancer cases [60]. A number of epigenome-wide studies reported differential methylation of various loci across genome between smokers and non-smokers, although none of them identified statistically significant variability in the methylation pattern by the smoking status for the BRCA1 or BRCA2 genes [6166]. The analyses of data from the genome-wide studies imply strong thresholds for significance; thus smaller effects may be excluded from being quantified. For instance, these studies did not find associations for smoking and methylation in the promoter region of another suppressor gene–CDKN2A (p16), while a meta-analysis of 19 epidemiological studies using the candidate genes approach demonstrated a higher frequency of the p16 gene hypermethylation in smoking vs non-smoking patients in tumor tissues of patients with diagnosed non-small cell lung carcinoma (OR = 2.25, 95%CI:1.81–2.80) [67].

The biological mechanism explaining the relations between smoking and epigenetic effects in particular gene-specific hypo- or hypermethylation remains unclear. Several mechanisms were postulated, mostly related to the activity of DNA methyltransferases, the enzymes responsible for DNA methylation. The carcinogens contained in cigarette smoke, such as polycyclic aromatic hydrocarbons, chromium and formaldehyde, are well known to generate DNA damage including double-strand breaks [6870]. It was also shown that the DNA repair sites recruit DNA methyltransferase 1 (DNMT1), resulting in the methylation of the repaired DNA fragments [7174]. Also cadmium, contained in cigarette smoke, was demonstrated to alter DNA methylation pattern through cadmium-mediated overexpression of DNMT3B with concomitant hypermethylation of the tumor suppressor genes p16 and RASSF1A [75]. Changes in DNA methylation were related to histone modification caused by cigarette smoking [76], which correlated with decreased DNMT1 and increased DNMT3B expression along with hypomethylation of repetitive DNA sequences and hypermethylation of tumor suppressor genes RASSF1A and RARB [76].

The epidemiological data regarding other environmental determinants of the epigenetic events in BRCA1 and BRCA2 have been sparse and inconclusive thus far. To our knowledge, no previous study that focused on these two genes examined the possible associations with night work. Two broad studies on epigenome investigated associations for the night work [44,47], but only the latter reported findings for the BRCA1 gene [47]. In this study, as much as 65% of the investigated sites showed hypomethylation in night workers compared to day workers. The hypomethylation referred to the sites (cytosines) within the BRCA1 gene (both within the promoter region and the gene body) [47]. This was the only study that showed that night shift work is associated with hypomethylation in the investigated genomic DNA, while in other studies in this field, both hypo- and hypermethylation was observed, depending on the gene or the gene region studied [44,45,77].

We found no significant associations between BMI or WHR and the BRCA1 or BRCA2 methylation status in the total study population. In one previous report, BMI was significantly associated with methylation in the BRCA1 gene determined in DNA extracted from breast tumor tissue [78]. This observation has not been supported by two other studies that investigated the methylation pattern of BRCA1 besides other cancer-related genes in breast tumor tissue [79]. The inconsistency of the findings may be explained by a possible mismatch of loci within the BRCA1 gene that were examined in each of these studies, as suggested by McCullough et al. [79]. An epigenome-wide association study of the methylation pattern of DNA deriving from peripheral white blood cells did not report inferences between BMI and the methylation pattern in the BRCA1 or BRCA2 gene [80].

We also did not find associations between the methylation status of both the examined genes and the physical activity in the total study population. Previous research demonstrated that physical activity may be associated with a “healthier methylation profile”[81,82]. As far as we know, there was only one study addressing this specific association for the genes we investigated. The intervention trial examined methylation level across 45 sites within a panel of 21 breast cancer-related genes (including 4 sites in BRCA1 and 1 site in BRCA2) [81]. This study demonstrated lower levels of methylation among individuals who exercised more minutes per week and also among those who increased their physical activity during 12 months of the trial. The most recent review evaluated data from 25 studies (both observational and intervention based) addressing association between physical activity and methylation pattern [83]. The authors concluded that both acute and chronic exercising significantly influence DNA methylation. The biological mechanisms of the process remain to be elucidated [83].

In our study we analyzed the main effect of the factors on BRCA1 and BRCA2 genes methylation with two models. The comparison of results from these two models showed no substantial differences, hence no strong confounding was detected. The present project was a pioneer study in the field of epigenetics within two important breast cancer-related genes as analyzed by night shift work, thus contributing to the sparse epidemiological data about inferences for modifiable lifestyle factors, and this can be regarded as one of its strong points. The size of the study population was relatively large when compared to other cross-sectional studies searching for mechanistic explanation of increased breast cancer risk among night shift workers. We also obtained detailed exposure data through in-person interviews and anthropometric measurements. The analyses were focused and respective hypotheses were formulated. The qMSP used for determining the BRCA1 and BRCA2 methylation patterns bears a high degree of sensitivity. This highly quantitative technique can accurately determine the relative prevalence of a particular pattern of CpG dinucleotides methylation.

There are also several limitations to consider. The conclusions regarding occupational exposure are confined to the rotating night shift work, therefore the results cannot be generalized to other shift work systems. The part of analysis referring to alcohol consumption was limited by the fact that in the population we studied, the level of alcohol consumption was generally low. Only 6 women reported having more than 5 drinks per week. Thus we were not able to examine associations for heavy drinkers. Even though the total study population was relatively large, the number of subjects in specific categories in some instances was rather small (below 10). This resulted in wide confidence intervals, which was the case e.g. for the shortest duration of night work history among current night workers (<10 years), light current smokers (1–4 cigarettes per day), or ever-smokers with a shorter smoking duration (<10 years).

In the analysis, we included the age and folate intake and other lifestyle characteristics that were considered in this study. While we could not rule out some confounding by other factors, we are not aware of any strong determinants of de novo methylation in BRCA1 or BRCA2.

Apart from the limitations mentioned above, the utility of peripheral blood for the assessment of associations between the risk factors and methylation alterations in genomic DNA may have had impact, given that the sample does not directly reflect the target tissue, as it has been indicated for human blood and brain using 450K Human Methylation array[84]. On the other hand, 4 of 8 investigated CpG loci showed a good correlation between blood and buccal cells in DNA methylation patterns [85].

There are only a few studies that focus on the potential reflection of the epigenetic status in DNA from the surrogate tissues compared with DNA from breast tumor or normal tissue. Fu et al and Sharma et al noted a concordance between tumor DNA methylation in breast cancer patients and paired serum or plasma DNA methylation of multiple cancer genes [86,87]. These findings indicate that particular DNA methylation patterns in peripheral blood cells may be useful for predicting breast cancer risk. However, no correlation was found between white blood cell DNA and normal breast epithelial cell DNA methylation of RASSF1 tumor suppressor gene and repetitive elements in healthy women undergoing reduction mammoplasty [88]. It has also been postulated that breast cancer treatment may potentially affect DNA methylation which would imply that the epigenetic analyses should ideally be conducted in untreated patients [50,51].

Nevertheless, blood-derived DNA methylation seems to be a promising epigenetic indicator in breast cancer investigations. In a recent systematic review, a possibility of using blood-derived DNA methylation features (global methylation or gene-specific methylation) for breast cancer stratification was suggested. The most frequently investigated gene in whole blood was BRCA1, which presented a higher methylation status in the promoter region in patients compared to controls [89]. Of concern is also that DNA methylation may vary depending on the white blood cell type [84]. Tissue-specific methylation patterns are established through a combination of demethylation and de novo methylation reactions [90]. Tissue specificity may be due to the differences in specific DNA binding proteins that bind to methylated and unmethylated DNA, and the availability of DNA methylase, all of which may play a role in the extent of de novo methylation or demethylation. Using the total population of mono- and polymorphonuclear white blood cells for DNA isolation may affect the methylation pattern of the genes. However, it has been shown that the majority of CpG loci tested presented stable methylation over a long time (11–20 years) in DNA isolated from blood taken at two time points, despite the possible changes in the cellular morphology of blood [85]. Thus considering the absence of temporal variation in the methylation pattern, we are of the opinion that specific gene-methylation can be reliably assessed using one blood sample.

Another limitation refers to the methodology for determining the methylation status. We did not examine all the epigenetic events in the CpG island within the whole promoter region, but only the 11CpG dinucleotides in BRCA1 and 10 in BRCA2.

The question also arises whether the storage time may affect the methylation pattern. It has recently been shown that most of the storage conditions for blood specimens have no effect on DNA integrity and methylation [91]. Thus it seems reasonable that the covalent binding of a methyl group to cytosines in gDNA samples obtained from venous blood and preserved for 6–7 years, since 2008 were not affected by the storage period. One of the mechanisms that may have influence on the stability of the methylation patterns is the formation of oxidative DNA bases such as 8-oxo-guanine, which is present in the CpG dinucleotides [91]. However, the rate of formation of this modified base in sub-freezing temperatures used for the storage of blood samples is negligible.

In conclusion, our study did not detect associations between the methylation index within the promoter sites of BRCA1 and BRCA2 genes and the rotating night shift work. The results of our research indicate that smoking may be one of the environmental determinants of the methylation status of the promoter region in the BRCA1 gene. Given the novelty of these findings, and limitations resulting from the relatively small number of subjects in subgroup analyses further validation research is warranted.

Supporting information

S1 Table. Selected characteristics of the studied population of nurses and midwives in the cross-sectional study by current rotating night work status.

https://doi.org/10.1371/journal.pone.0178792.s001

(PDF)

Acknowledgments

We thank Regional Registry of the Nurses and Midwives in Lodz for help in organisation of the study. We thank interviewers. We also thank all nurses and midwives for their participation in this project.

Author Contributions

  1. Conceptualization: BP ER AB SZ.
  2. Data curation: AB.
  3. Formal analysis: BP ER EW MP.
  4. Funding acquisition: BP ER SZ.
  5. Investigation: BP AB.
  6. Methodology: BP ER.
  7. Project administration: BP.
  8. Supervision: BP ER SZ.
  9. Validation: ER EW MP.
  10. Writing – original draft: BP.
  11. Writing – review & editing: BP AB EW MP SZ ER.

References

  1. 1. Oesterreich S, Fuqua SA. Tumor suppressor genes in breast cancer. Endocr Relat Cancer 1999; 6: 405–419. pmid:10516854
  2. 2. Szwiec M, Jakubowska A, Gorski B, Huzarski T, Tomiczek-Szwiec J, Gronwald J, et al. Recurrent mutations of BRCA1 and BRCA2 in Poland: an update. Clin Genet 2015; 87: 288–292. pmid:24528374
  3. 3. Rosen EM, Fan S, Pestell RG, Goldberg ID. BRCA1 gene in breast cancer. J Cell Physiol 2003; 196: 19–41. pmid:12767038
  4. 4. James FR, Wootton S, Jackson A, Wiseman M, Copson ER, Cutress RI. Obesity in breast cancer—what is the risk factor? Eur J Cancer 2015; 51: 705–720. pmid:25747851
  5. 5. Shield KD, Soerjomataram I, Rehm J. Alcohol Use and Breast Cancer: A Critical Review. Alcohol Clin Exp Res 2016; 40: 1166–1181. pmid:27130687
  6. 6. Stewart B, Wild CP, (eds.) (2014) World Cancer Report 2014.
  7. 7. Reynolds P. Smoking and breast cancer. J Mammary Gland Biol Neoplasia 2013; 18: 15–23. pmid:23179580
  8. 8. Gaudet MM, Carter BD, Brinton LA, Falk RT, Gram IT, Luo J, et al. Pooled analysis of active cigarette smoking and invasive breast cancer risk in 14 cohort studies. Int J Epidemiol. 2016; pmid:28031315
  9. 9. Straif K, Baan R, Grosse Y, Secretan B, El GF, Bouvard V, et al. Carcinogenicity of shift-work, painting, and fire-fighting. Lancet Oncol 2007; 8: 1065–1066. pmid:19271347
  10. 10. Ijaz S, Verbeek J, Seidler A, Lindbohm ML, Ojajarvi A, Orsini N, et al. Night-shift work and breast cancer—a systematic review and meta-analysis. Scand J Work Environ Health 2013; 39: 431–447. pmid:23804277
  11. 11. Wang F, Yeung KL, Chan WC, Kwok CC, Leung SL, Wu C, et al. A meta-analysis on dose-response relationship between night shift work and the risk of breast cancer. Ann Oncol 2013; 24: 2724–2732. pmid:23975662
  12. 12. Papantoniou K, Castano-Vinyals G, Espinosa A, Aragones N, Perez-Gomez B, Ardanaz E, et al. Breast cancer risk and night shift work in a case-control study in a Spanish population. Eur J Epidemiol 2016; 31: 867–878. pmid:26205167
  13. 13. Cordina-Duverger E, Koudou Y, Truong T, Arveux P, Kerbrat P, Menegaux F, et al. Night work and breast cancer risk defined by human epidermal growth factor receptor-2 (HER2) and hormone receptor status: A population-based case-control study in France. Chronobiol Int 2016; 33: 783–787. pmid:27078711
  14. 14. Travis RC, Balkwill A, Fensom GK, Appleby PN, Reeves GK, Wang XS, et al. Night Shift Work and Breast Cancer Incidence: Three Prospective Studies and Meta-analysis of Published Studies. J Natl Cancer Inst 2016; 108. pmid:27758828
  15. 15. Hansen MA, Garde HA, Hansen J. Diurnal urinary 6-sulfatoxymelatonin levels among healthy Danish nurses during work and leisure time. Chronobiol Int 2006; 23: 1203–1215. pmid:17190706
  16. 16. Schernhammer ES, Rosner B, Willett WC, Laden F, Colditz GA, Hankinson SE. Epidemiology of urinary melatonin in women and its relation to other hormones and night work. Cancer Epidemiol Biomarkers Prev 2004; 13: 936–943. pmid:15184249
  17. 17. Schernhammer ES, Kroenke CH, Dowsett M, Folkerd E, Hankinson SE. Urinary 6-sulfatoxymelatonin levels and their correlations with lifestyle factors and steroid hormone levels. J Pineal Res 2006; 40: 116–124. pmid:16441548
  18. 18. Davis S, Mirick DK, Chen C, Stanczyk FZ. Night shift work and hormone levels in women. Cancer Epidemiol Biomarkers Prev 2012; 21: 609–618. pmid:22315366
  19. 19. Fritschi L, Glass DC, Heyworth JS, Aronson K, Girschik J, Boyle T, et al. Hypotheses for mechanisms linking shiftwork and cancer. Med Hypotheses 2011; 77: 430–436. pmid:21723672
  20. 20. Wang XS, Travis RC, Reeves G, Green J, Allen NE, Key TJ, et al. Characteristics of the Million Women Study participants who have and have not worked at night. Scand J Work Environ Health 2012; 38: 590–599. pmid:22772745
  21. 21. Saksvik-Lehouillier I, Bjorvatn B, Hetland H, Sandal GM, Moen BE, Mageroy N, et al. Individual, situational and lifestyle factors related to shift work tolerance among nurses who are new to and experienced in night work. J Adv Nurs 2013; 69: 1136–1146. pmid:22853193
  22. 22. Zhao I, Bogossian F, Turner C. Does maintaining or changing shift types affect BMI? A longitudinal study. J Occup Environ Med 2012; 54: 525–531. pmid:22576459
  23. 23. Zhao I, Bogossian F, Turner C. A cross-sectional analysis of the association between night-only or rotating shift work and overweight/obesity among female nurses and midwives. J Occup Environ Med 2012; 54: 834–840. pmid:22796928
  24. 24. Trinkoff AM, Storr CL. Work schedule characteristics and substance use in nurses. Am J Ind Med 1998; 34: 266–271. pmid:9698996
  25. 25. van Amelsvoort LG, Schouten EG, Kok FJ. Duration of shiftwork related to body mass index and waist to hip ratio. Int J Obes Relat Metab Disord 1999; 23: 973–978. pmid:10490804
  26. 26. Kivimaki M, Kuisma P, Virtanen M, Elovainio M. Does shift work lead to poorer health habits? A comparison between women who had always done shift work with those who had never done shift wok. Work & Stress 2001; 15: 3–13.
  27. 27. Geliebter A, Gluck ME, Tanowitz M, Aronoff NJ, Zammit GK. Work-shift period and weight change. Nutrition 2000; 16: 27–29. pmid:10674231
  28. 28. Knutsson A, Nilsson T. Tobacco use and exposure to environmental tobacco smoke in relation to certain work characteristics. Scand J Soc Med 1998; 26: 183–189. pmid:9768448
  29. 29. De Bacquer D, Van Risseghem M, Clays E, Kittel F, De BG, Braeckman L. Rotating shift work and the metabolic syndrome: a prospective study. Int J Epidemiol 2009; 38: 848–854. pmid:19129266
  30. 30. Kawachi I, Colditz GA, Stampfer MJ, Willett WC, Manson JE, Speizer FE, et al. Prospective study of shift work and risk of coronary heart disease in women. Circulation 1995; 92: 3178–3182. pmid:7586301
  31. 31. Hansen J, Stevens RG. Case-control study of shift-work and breast cancer risk in Danish nurses: impact of shift systems. Eur J Cancer 2012; 48: 1722–1729. pmid:21852111
  32. 32. Schernhammer ES, Kroenke CH, Laden F, Hankinson SE. Night work and risk of breast cancer. Epidemiology 2006; 17: 108–111. pmid:16357603
  33. 33. Bushnell PT, Colombi A, Caruso CC, Tak S. Work schedules and health behavior outcomes at a large manufacturer. Ind Health 2010; 48: 395–405. pmid:20720331
  34. 34. Peplonska B, Bukowska A, Gromadzinska J, Sobala W, Reszka E, Lie JA, et al. Night shift work characteristics and 6-sulfatoxymelatonin (MT6s) in rotating night shift nurses and midwives. Occup Environ Med 2012; 69: 339–346. pmid:22368032
  35. 35. van Amelsvoort LG, Schouten EG, Kok FJ. Impact of one year of shift work on cardiovascular disease risk factors. J Occup Environ Med 2004; 46: 699–706. pmid:15247809
  36. 36. de Assis MA, Kupek E, Nahas MV, Bellisle F. Food intake and circadian rhythms in shift workers with a high workload. Appetite 2003; 40: 175–183. pmid:12781167
  37. 37. Morikawa Y, Nakagawa H, Miura K, Soyama Y, Ishizaki M, Kido T, et al. Effect of shift work on body mass index and metabolic parameters. Scand J Work Environ Health 2007; 33: 45–50. pmid:17353964
  38. 38. Pietroiusti A, Neri A, Somma G, Coppeta L, Iavicoli I, Bergamaschi A, et al. Incidence of metabolic syndrome among night-shift healthcare workers. Occup Environ Med 2010; 67: 54–57. pmid:19737731
  39. 39. Nakamura K, Shimai S, Kikuchi S, Tominaga K, Takahashi H, Tanaka M, et al. Shift work and risk factors for coronary heart disease in Japanese blue-collar workers: serum lipids and anthropometric characteristics. Occup Med (Lond) 1997; 47: 142–146.
  40. 40. Amani R, Gill T. Shiftworking, nutrition and obesity: implications for workforce health- a systematic review. Asia Pac J Clin Nutr 2013; 22: 505–515. pmid:24231009
  41. 41. van Drongelen A, Boot C, Merkus S, Smid T, van der Beek A. The effects of shift work on body weight change—a systematic review of longitudinal studies. Scand J Work Environ Health. 2011; 37:263–75. pmid:21243319
  42. 42. Peplonska B, Bukowska A, Sobala W. Association of Rotating Night Shift Work with BMI and Abdominal Obesity among Nurses and Midwives. PLoS One 2015); 10: e0133761. pmid:26196859
  43. 43. Opperhuizen AL, van Kerkhof LW, Proper KI, Rodenburg W, Kalsbeek A. Rodent models to study the metabolic effects of shiftwork in humans. Front Pharmacol 2015; 6: 50. pmid:25852554
  44. 44. Zhu Y, Stevens RG, Hoffman AE, Tjonneland A, Vogel UB, Zheng T, et al. Epigenetic impact of long-term shiftwork: pilot evidence from circadian genes and whole-genome methylation analysis. Chronobiol Int 2011; 28: 852–861. pmid:22080730
  45. 45. Bollati V, Baccarelli A, Sartori S, Tarantini L, Motta V, Rota F, et al. Epigenetic effects of shiftwork on blood DNA methylation. Chronobiol Int 2010; 27: 1093–1104. pmid:20636218
  46. 46. Liu R, Jacobs DI, Hansen J, Fu A, Stevens RG, Zhu Y. Aberrant methylation of miR-34b is associated with long-term shiftwork: a potential mechanism for increased breast cancer susceptibility. Cancer Causes Control 2015; 26: 171–178. pmid:25398683
  47. 47. Bhatti P, Zhang Y, Song X, Makar KW, Sather CL, Kelsey KT, et al. Nightshift work and genome-wide DNA methylation. Chronobiol Int 2015; 32: 103–112. pmid:25187986
  48. 48. Kochan DZ, Kovalchuk O. Circadian disruption-induced breast cancer—knowns and unknowns. Cell Cycle 2016; 15: 613–614. pmid:26735780
  49. 49. Shi F, Chen X, Fu A, Hansen J, Stevens R, Tjonneland A, et al. Aberrant DNA methylation of miR-219 promoter in long-term night shiftworkers. Environ Mol Mutagen 2013; 54: 406–413. pmid:23813567
  50. 50. Hoffman AE, Yi CH, Zheng T, Stevens RG, Leaderer D, Zhang Y, et al. CLOCK in breast tumorigenesis: genetic, epigenetic, and transcriptional profiling analyses. Cancer Res 2010; 70: 1459–1468. pmid:20124474
  51. 51. Hoffman AE, Zheng T, Yi CH, Stevens RG, Ba Y, Zhang Y, et al. The core circadian gene Cryptochrome 2 influences breast cancer risk, possibly by mediating hormone signaling. Cancer Prev Res (Phila) 2010; 3: 539–548.
  52. 52. Rice JC, Massey-Brown KS, Futscher BW. Aberrant methylation of the BRCA1 CpG island promoter is associated with decreased BRCA1 mRNA in sporadic breast cancer cells. Oncogene 1998; 17: 1807–1812. pmid:9778046
  53. 53. Wong EM, Southey MC, Fox SB, Brown MA, Dowty JG, Jenkins MA, et al. Constitutional methylation of the BRCA1 promoter is specifically associated with BRCA1 mutation-associated pathology in early-onset breast cancer. Cancer Prev Res (Phila) 2011; 4: 23–33.
  54. 54. Craig CL, Marshall AL, Sjostrom M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003; 35: 1381–1395. pmid:12900694
  55. 55. Peplonska B, Bukowska A, Sobala W. Rotating night shift work and physical activity of nurses and midwives in the cross-sectional study in Lodz, Poland. Chronobiol Int 2014; 31: 1152–1159. pmid:25216072
  56. 56. Eads CA, Danenberg KD, Kawakami K, Saltz LB, Blake C, Shibata D, et al. MethyLight: a high-throughput assay to measure DNA methylation. Nucleic Acids Res 2000; 28: E32. pmid:10734209
  57. 57. Hardy TM, Tollefsbol TO. Epigenetic diet: impact on the epigenome and cancer. Epigenomics 2011; 3: 503–518. pmid:22022340
  58. 58. Gao X, Jia M, Zhang Y, Breitling LP, Brenner H. DNA methylation changes of whole blood cells in response to active smoking exposure in adults: a systematic review of DNA methylation studies. Clin Epigenetics 2015; 7: 113. pmid:26478754
  59. 59. Ottini L, Rizzolo P, Siniscalchi E, Zijno A, Silvestri V, Crebelli R, et al. Gene promoter methylation and DNA repair capacity in monozygotic twins with discordant smoking habits. Mutat Res Genet Toxicol Environ Mutagen 2015; 779: 57–64. pmid:25813726
  60. 60. White AJ, Chen J, Teitelbaum SL, McCullough LE, Xu X, Hee CY, et al. Sources of polycyclic aromatic hydrocarbons are associated with gene-specific promoter methylation in women with breast cancer. Environ Res 2016; 145: 93–100. pmid:26671626
  61. 61. Tsaprouni LG, Yang TP, Bell J, Dick KJ, Kanoni S, Nisbet J, et al. Cigarette smoking reduces DNA methylation levels at multiple genomic loci but the effect is partially reversible upon cessation. Epigenetics 2014; 9: 1382–1396. pmid:25424692
  62. 62. Harlid S, Xu Z, Panduri V, Sandler DP, Taylor JA. CpG sites associated with cigarette smoking: analysis of epigenome-wide data from the Sister Study. Environ Health Perspect 2014; 122: 673–678. pmid:24704585
  63. 63. Zeilinger S, Kuhnel B, Klopp N, Baurecht H, Kleinschmidt A, Gieger C, et al. Tobacco smoking leads to extensive genome-wide changes in DNA methylation. PLoS One 2013; 8: e63812. pmid:23691101
  64. 64. Sun YV, Smith AK, Conneely KN, Chang Q, Li W, Lazarus A, et al. Epigenomic association analysis identifies smoking-related DNA methylation sites in African Americans. Hum Genet 2013; 132: 1027–1037. pmid:23657504
  65. 65. Shenker NS, Polidoro S, van VK, Sacerdote C, Ricceri F, Birrell MA, et al. Epigenome-wide association study in the European Prospective Investigation into Cancer and Nutrition (EPIC-Turin) identifies novel genetic loci associated with smoking. Hum Mol Genet 2013; 22: 843–851. pmid:23175441
  66. 66. Breitling LP, Yang R, Korn B, Burwinkel B, Brenner H. Tobacco-smoking-related differential DNA methylation: 27K discovery and replication. Am J Hum Genet 2011; 88: 450–457. pmid:21457905
  67. 67. Zhang B, Zhu W, Yang P, Liu T, Jiang M, He ZN, et al. Cigarette smoking and p16INK4alpha gene promoter hypermethylation in non-small cell lung carcinoma patients: a meta-analysis. PLoS One 2011; 6: e28882. pmid:22174919
  68. 68. Lee KW, Pausova Z. Cigarette smoking and DNA methylation. Front Genet 2013; 4: 132. pmid:23882278
  69. 69. Smith CJ, Hansch C. The relative toxicity of compounds in mainstream cigarette smoke condensate. Food Chem Toxicol 2000; 38: 637–646. pmid:10942325
  70. 70. DeMarini DM. Genotoxicity of tobacco smoke and tobacco smoke condensate: a review. Mutat Res 2004; 567: 447–474. pmid:15572290
  71. 71. Mortusewicz O, Schermelleh L, Walter J, Cardoso MC, Leonhardt H. Recruitment of DNA methyltransferase I to DNA repair sites. Proc Natl Acad Sci U S A 2005; 102: 8905–8909. pmid:15956212
  72. 72. Cuozzo C, Porcellini A, Angrisano T, Morano A, Lee B, Di PA, et al. DNA damage, homology-directed repair, and DNA methylation. PLoS Genet 2007; 3: e110. pmid:17616978
  73. 73. Le Gac G, Esteve PO, Ferec C, Pradhan S. DNA damage-induced down-regulation of human Cdc25C and Cdc2 is mediated by cooperation between p53 and maintenance DNA (cytosine-5) methyltransferase 1. J Biol Chem 2006; 281: 24161–24170. pmid:16807237
  74. 74. Ha K, Lee GE, Palii SS, Brown KD, Takeda Y, Liu K, et al. Rapid and transient recruitment of DNMT1 to DNA double-strand breaks is mediated by its interaction with multiple components of the DNA damage response machinery. Hum Mol Genet 2011; 20: 126–140. pmid:20940144
  75. 75. Benbrahim-Tallaa L, Waterland RA, Dill AL, Webber MM, Waalkes MP. Tumor suppressor gene inactivation during cadmium-induced malignant transformation of human prostate cells correlates with overexpression of de novo DNA methyltransferase. Environ Health Perspect 2007; 115: 1454–1459. pmid:17938735
  76. 76. Liu F, Killian JK, Yang M, Walker RL, Hong JA, Zhang M, et al. Epigenomic alterations and gene expression profiles in respiratory epithelia exposed to cigarette smoke condensate. Oncogene 2010; 29: 3650–3664. pmid:20440268
  77. 77. Jacobs DI, Hansen J, Fu A, Stevens RG, Tjonneland A, Vogel UB, et al. Methylation alterations at imprinted genes detected among long-term shiftworkers. Environ Mol Mutagen 2013; 54: 141–146. pmid:23193016
  78. 78. Naushad SM, Hussain T, Al-Attas OS, Prayaga A, Digumarti RR, Gottumukkala SR, et al. Molecular insights into the association of obesity with breast cancer risk: relevance to xenobiotic metabolism and CpG island methylation of tumor suppressor genes. Mol Cell Biochem 2014; 392: 273–280. pmid:24676543
  79. 79. McCullough LE, Chen J, White AJ, Xu X, Cho YH, Bradshaw PT, et al. Gene-Specific Promoter Methylation Status in Hormone-Receptor-Positive Breast Cancer Associates with Postmenopausal Body Size and Recreational Physical Activity. Int J Cancer Clin Res 2015; 2: 013. pmid:26005715
  80. 80. Dick KJ, Nelson CP, Tsaprouni L, Sandling JK, Aissi D, Wahl S, et al. DNA methylation and body-mass index: a genome-wide analysis. Lancet 2014; 383: 1990–1998. pmid:24630777
  81. 81. Bryan AD, Magnan RE, Hooper AE, Harlaar N, Hutchison KE. Physical activity and differential methylation of breast cancer genes assayed from saliva: a preliminary investigation. Ann Behav Med 2013; 45: 89–98. pmid:23054940
  82. 82. Zhang FF, Cardarelli R, Carroll J, Zhang S, Fulda KG, Gonzalez K, et al. Physical activity and global genomic DNA methylation in a cancer-free population. Epigenetics 2011; 6: 293–299. pmid:21178401
  83. 83. Voisin S, Eynon N, Yan X, Bishop DJ. Exercise training and DNA methylation in humans. Acta Physiol (Oxf) 2015; 213: 39–59.
  84. 84. Farre P, Jones MJ, Meaney MJ, Emberly E, Turecki G, Kobor MS. Concordant and discordant DNA methylation signatures of aging in human blood and brain. Epigenetics Chromatin 2015; 8: 19. pmid:25977707
  85. 85. Talens RP, Boomsma DI, Tobi EW, Kremer D, Jukema JW, Willemsen G, et al. Variation, patterns, and temporal stability of DNA methylation: considerations for epigenetic epidemiology. FASEB J 2010; 24: 3135–3144. pmid:20385621
  86. 86. Sharma G, Mirza S, Prasad CP, Srivastava A, Gupta SD, Ralhan R. Promoter hypermethylation of p16INK4A, p14ARF, CyclinD2 and Slit2 in serum and tumor DNA from breast cancer patients. Life Sci. 2007; 80(20):1873–1881. pmid:17383681
  87. 87. Fu D, Ren C, Tan H, Wei J, Zhu Y, He C, et al. Sox 17 promoter methylation in plasma DNA is associated with poor survival and can be used as a prognostic factor in breast cancer. Medicine (Baltimore) 2015; 94(11):e637.
  88. 88. Sturgeon SR, Arcaro KF, Johnson MA, Balasubramanian R, Zorn M, Jerry DJ, et al. DNA methylation in paired breast epithelial and white blood cells from women undergoing reduction mammoplasty. Anticancer Res. 2014; 34(6):2985–2990. pmid:24922663
  89. 89. Tang Q, Cheng J, Cao X, Surowy H, Burwinkel B. Blood-based DNA methylation as biomarker for breast cancer: a systematic review. Clin Epigenetics 2016; 8: 115. pmid:27895805
  90. 90. Szyf M. DNA methylation patterns: an additional level of information? Biochem Cell Biol 1991; 69: 764–767. pmid:1818583
  91. 91. Cerda S, Weitzman SA. Influence of oxygen radical injury on DNA methylation. Mutat Res 1997; 386: 141–152. pmid:9113115