Ng GenomeStudio software (ver. 2011.0, Illumina Inc.). One sample, including its matched pair, was omitted from subsequent analyses due to incomplete bisulfite conversion.then incubated at 37uC for 16 hours followed by 20 min at 65uC. 5-HTTLPR genotypes were assessed on a 2 gel.Statistical AnalysisDifferences in mean MBI-GS and BDI scores between highand low work stress groups were tested by the Student’s t-test. Correlations between methylation levels in all five CpG residues were calculated using the Pearson’s Correlation. Structural equation measurement model was analyzed in order to test the hypothesis that these correlations could be explained by a single latent factor. Principal component factor analysis was used to obtain the single sum methylation value METsum. Statistical significance (2-tailed p,0.05) between methylation of each five CpG residue and work stress was assessed using the Student’s ttest. Associations between work stress, MBI-GS, METsum and 5HTTLPR were assessed by analysis of covariance (ANCOVA). In the initial multivariable model, we analyzed the main effects of 6 variables. Non-significant variables were then systematically eliminated using a stepwise inhibitor backward elimination method until all significant variables were left in the final model (see Table S1). Goodness of fit by chi-squared test was used for examining differences between direct sequencing and 450 k BeadChip methylation results in the age-matched pairs. The tools used in all analysis were SPSS Statistics 18, PLINK (http://pngu.mgh.harvard.edu/purcell/plink/) [43] and SAS analytics.Supporting InformationFigure S1 Flow chart of sample selection. A large cohort of 5615 health care professionals was divided into lowest and highest quartile on workplace aggregated stress as 18055761 per Karasek Model. 422 nurses were invited for further analysis. 99 nurses attended to laboratory examination and successful measurements were Epigenetics available for 95 individuals Laboratory measurements were available from 95 individuals. After exclusion criteria, the promoter region of SLC6A4 of 49 individuals was successfully amplified and bisulfite sequenced. Success of sequencing was based on DNA yield after bisulfite conversion and PCR amplification. (TIF) Figure S2 Distribution of methylation levels in the high and work stress environments. Work stress is defined according to Karasek’s Model by the difference between averages of work demand and control. (TIF) Table S1 A. METsum in high and low work stress groups and5-HTTLPR Genotyping5-HTTLPR genotype was assayed in two stages. Amplification of 5-HTTLPR was carried out using PCR using the primers HTTLPR_F (gttgccgctctgaatgccag) and HTTLPR_R (ggataatgggggttgcaggg) that amplify either the long, L allele (280 bp product) or the short, S allele (236 bp product). The reaction was run in a total volume of 20 ml containing 60 ng of genomic DNA, 500 nM primers, 10 mM dNTP, 2.5 mM MgCl2 and 0.7U Dynazyme II Hotstart polymerase (Finnzymes). Thermocycling conditions were 94uC initial denaturing for 10 min followed by 35 cycles of the following: denaturing at 94uC for 15 s, annealing at 65uC for 20 s and extension at 72uC for 25 s. This was followed by a final extension at 72uC for 10 min. 5 ml of pcr product was then mixed with 8 units of Msp I restriction enzyme (Pharmacia Biotech) in One-Phor-All assay buffer. The reaction mixture waswork stress environment as a whole. The effect size can be calculated using Cohen’s d, defined as the difference b.Ng GenomeStudio software (ver. 2011.0, Illumina Inc.). One sample, including its matched pair, was omitted from subsequent analyses due to incomplete bisulfite conversion.then incubated at 37uC for 16 hours followed by 20 min at 65uC. 5-HTTLPR genotypes were assessed on a 2 gel.Statistical AnalysisDifferences in mean MBI-GS and BDI scores between highand low work stress groups were tested by the Student’s t-test. Correlations between methylation levels in all five CpG residues were calculated using the Pearson’s Correlation. Structural equation measurement model was analyzed in order to test the hypothesis that these correlations could be explained by a single latent factor. Principal component factor analysis was used to obtain the single sum methylation value METsum. Statistical significance (2-tailed p,0.05) between methylation of each five CpG residue and work stress was assessed using the Student’s ttest. Associations between work stress, MBI-GS, METsum and 5HTTLPR were assessed by analysis of covariance (ANCOVA). In the initial multivariable model, we analyzed the main effects of 6 variables. Non-significant variables were then systematically eliminated using a stepwise backward elimination method until all significant variables were left in the final model (see Table S1). Goodness of fit by chi-squared test was used for examining differences between direct sequencing and 450 k BeadChip methylation results in the age-matched pairs. The tools used in all analysis were SPSS Statistics 18, PLINK (http://pngu.mgh.harvard.edu/purcell/plink/) [43] and SAS analytics.Supporting InformationFigure S1 Flow chart of sample selection. A large cohort of 5615 health care professionals was divided into lowest and highest quartile on workplace aggregated stress as 18055761 per Karasek Model. 422 nurses were invited for further analysis. 99 nurses attended to laboratory examination and successful measurements were available for 95 individuals Laboratory measurements were available from 95 individuals. After exclusion criteria, the promoter region of SLC6A4 of 49 individuals was successfully amplified and bisulfite sequenced. Success of sequencing was based on DNA yield after bisulfite conversion and PCR amplification. (TIF) Figure S2 Distribution of methylation levels in the high and work stress environments. Work stress is defined according to Karasek’s Model by the difference between averages of work demand and control. (TIF) Table S1 A. METsum in high and low work stress groups and5-HTTLPR Genotyping5-HTTLPR genotype was assayed in two stages. Amplification of 5-HTTLPR was carried out using PCR using the primers HTTLPR_F (gttgccgctctgaatgccag) and HTTLPR_R (ggataatgggggttgcaggg) that amplify either the long, L allele (280 bp product) or the short, S allele (236 bp product). The reaction was run in a total volume of 20 ml containing 60 ng of genomic DNA, 500 nM primers, 10 mM dNTP, 2.5 mM MgCl2 and 0.7U Dynazyme II Hotstart polymerase (Finnzymes). Thermocycling conditions were 94uC initial denaturing for 10 min followed by 35 cycles of the following: denaturing at 94uC for 15 s, annealing at 65uC for 20 s and extension at 72uC for 25 s. This was followed by a final extension at 72uC for 10 min. 5 ml of pcr product was then mixed with 8 units of Msp I restriction enzyme (Pharmacia Biotech) in One-Phor-All assay buffer. The reaction mixture waswork stress environment as a whole. The effect size can be calculated using Cohen’s d, defined as the difference b.