Skip to main content

Some NLM-NCBI services and products are experiencing heavy traffic, which may affect performance and availability. We apologize for the inconvenience and appreciate your patience. For assistance, please contact our Help Desk at info@ncbi.nlm.nih.gov.

BMJ Open logoLink to BMJ Open
. 2015 Apr 28;5(4):e007670. doi: 10.1136/bmjopen-2015-007670

Influence of lifestyle on the FAIM2 promoter methylation between obese and lean children: a cohort study

Lijun Wu 1, Xiaoyuan Zhao 1, Yue Shen 2, Guimin Huang 1, Meixian Zhang 1, Yinkun Yan 1, Dongqing Hou 1, Linghui Meng 1, Junting Liu 1, Hong Cheng 1, Jie Mi 1
PMCID: PMC4420961  PMID: 25922107

Abstract

Objective

An obesity-related gene, Fas apoptotic inhibitory molecule 2 (FAIM2), is regulated by nutritional state and the methylation levels of the FAIM2 promoter are significantly associated with obesity. Lifestyle factors, such as sedentary behaviour and physical activity, might modify epigenetic patterns that have been related to obesity. Whether the molecular mechanisms by which FAIM2 affects obesity are involved in lifestyle is unclear. This study investigates the potential differences of the FAIM2 promoter methylation with sedentary behaviour and physical activity in obese and lean children.

Design

Cohort study.

Setting

Institute of Pediatrics in China.

Participants

59 obese cases and 39 lean controls aged 8–18 years recruited from a cross-sectional survey of children from Beijing in 2013.

Primary and secondary outcome measures

The FAIM2 promoter methylation was quantified using the Sequenom MassARRAY platform. Sedentary behaviour and physical activity were investigated using a questionnaire. The influences of different lifestyles on methylation variations in obese and lean children were examined by multiple linear regression.

Results

The methylation levels at seven CpG sites of the FAIM2 promoter were significantly associated with sedentary behaviour, especially the methylation levels at site −975, site −413, sites −362 and −360, and sites −353 and −349 (p=0.00004, 0.00009, 0.0006 and 0.00005, respectively). There were significant differences between the methylation levels at four CpG sites in obese and lean participants with high or moderate physical activity level <150 min/week.

Conclusions

This study provides the first evidence that there are significant differences in the associations of the FAIM2 promoter methylation with sedentary behaviour and physical activity between obese and lean children. Our results suggest that lifestyle may possibly be mediating the process of the FAIM2 involved in obesity.

Keywords: GENETICS, EPIDEMIOLOGY


Strengths and limitations of this study.

  • This is the first study to evaluate the methylation levels of the Fas apoptotic inhibitory molecule 2 (FAIM2) promoter and lifestyle factors.

  • This study may not provide direct evidence that the expression of FAIM2 influences obesity.

  • This study investigated the methylation levels in peripheral blood leucocytes, but not in the hypothalamus or adipocytes.

Introduction

In recent years, the rising incidence of obesity has been a major threat to public health worldwide.1 Obesity most likely results from the complex interaction between genes and lifestyles such as physical activity, diet and behaviour.2 Current evidence indicated that lifestyle factors might modify epigenetic patterns,3 and DNA methylation has been suggested to be associated with the development of obesity.4

Fas apoptotic inhibitory molecule 2 (FAIM2), a 35.1 kDa membrane protein highly expressed in the hippocampus, regulates Fas ligand-mediated apoptosis in neurons.5 The single nucleotide polymorphism (SNP) rs7138803 near FAIM2 was reported to be associated with obesity based on genome-wide association (GWA) studies in Caucasians,6 and replicated studies showed that this SNP was significantly associated with obesity in Asian populations.7 8 FAIM2 was regulated by nutritional state9 and the methylation levels of the FAIM2 promoter were significantly associated with obesity,10 but whether the molecular mechanism by which FAIM2 affects obesity is involved in lifestyle has not been clarified.

Because obesity-related lifestyle factors might modify epigenetic patterns and the methylation levels of the FAIM2 promoter are significantly associated with obesity, we hypothesise that lifestyle factors play a role in the methylation of the FAIM2 promoter. The aim of this study was to investigate the differences of the methylation levels of the FAIM2 promoter between obese and lean participants according to different sedentary behaviour and physical activity.

Methods

Population

The cohort included 59 obese and 39 lean participants randomly recruited from a cross-sectional survey of children from Beijing aged 8–18 years in 2013. The survey was a physical fitness and health surveillance of Beijing school students, and included a questionnaire, medical examination, anthropometric measurement and collected venipuncture blood samples (n=3143; boys 50%). Obesity was diagnosed by the Chinese age-specific and sex-specific body mass index (BMI) cut-offs (see online supplementary table S1) and fat mass percentage (FMP) >40. Leanness was diagnosed by WHO BMI cut-offs and FMP <15. The research budget limited the sample size in the study. In future studies, we hope to examine the methylation levels of the FAIM2 promoter in greater sample sizes.

We obtained written informed consent from all participants (or their parents/guardians).

Measurement of anthropometric parameters and biochemical analyses

The anthropometric measurements included FMP, weight and height. All instruments were validated according to the methods of the manufacturers. FMP was measured using a body composition analyser (InBody 720, Biospace Co, Ltd, Seoul, Korea). BMI was calculated as weight in kilograms divided by square of height in metres. Blood pressure was measured via auscultation using a standard clinical sphygmomanometer.

Levels of triglycerides, high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, total cholesterol and fasting plasma glucose were examined using an automatic biochemical analyser (Olympus chemistry analyser AU640, Olympus Optical Co, Ltd, Tokyo, Japan) using commercial kits (Olympus Optical Co, Ltd, Tokyo, Japan).

Measurement of sedentary behaviour and physical activity

A validated questionnaire was used to investigate sedentary behaviour and physical activity in children. The questionnaire referred to the questionnaire used in the study by Meng et al.11 The questionnaires were completed by parents or guardians. The questions were retrospective and the questionnaire collected the data in the past 6 months. Sedentary behaviour was determined by the time spent either watching television or playing video/computer games per day in a week. Moderate physical activity (MPA) was determined by the time spent jogging or playing table tennis, or practising Tai Chi, etc, per week. High physical activity (HPA) was determined by the time spent playing football, basketball or badminton, etc, per week. HPA or MPA level was determined by the time spent on MPA or HPA per week. According to sedentary behaviour and physical activity, we stratified the participants into six groups, including sedentary behaviour <60 min/day, sedentary behaviour ≥60 min/day, HPA level <30 min/day, HPA level ≥30 min/day, HPA or MPA level <150 min/week and HPA or MPA level ≥150 min/week.

DNA methylation analysis

Genomic DNA was extracted from peripheral white blood cells using a QIAamp DNA Blood Mini Kit (QIAGEN Inc, Valencia, California, USA).

The Sequenom MassARRAY platform (BioMiao Biological Technology, Beijing, China) was used to quantitatively examine the methylation of FAIM2 promoter. The genomic DNA was bisulfite-treated using the EpiTect Bisulfite kit (Qiagen) according to the manufacturer's instructions. The target DNA region was amplified using the bisulfite-modified DNA. The specific primers are listed in online supplementary table S2. The primers were designed using epidesigner (http://www.epidesigner.com), and the methylation ratios were calculated using Epityper software V.1.0 (Sequenom, San Diego, California, USA).

Statistical analysis

The continuous variables are presented as the mean±SD, and the categorical variables are presented as the percentages. Multiple linear regressions were used to investigate the difference in methylation levels between the different groups after adjusting for age and gender, or age, gender and BMI. A logistic regression model was used to investigate the interaction between the methylation levels of the FAIM2 promoter and physical activity regarding the obesity status of the children (the model: obesity=CpG+PA+CpG*PA). The p value <0.05 was used to indicate statistically significant differences. We applied multiple testing to correct for multiple comparisons, the false discovery rate (FDR)12 approach was used; FDR analysis (0.05 as criteria) was applied for six groups (sedentary behaviour <60 min/day, sedentary behaviour ≥60 min/day, HPA level <30 min/day, HPA level ≥30 min/day, HPA or MPA level <150 min/week and HPA or MPA level ≥150 min/week), and 36 CpG sites simultaneously (number of test: 36×6=216). In brief, if the original p value was less than p value for FDR, then it represented statistical significance; otherwise, it suggested non-statistical significance. Data were analysed using SPSS, V.13.0 (SPSS Inc, Chicago, Illinois, USA).

Results

The characteristics of the study participants are shown in online supplementary table S3. We analysed the associations between the methylation levels of the FAIM2 promoter and the categories of sedentary behaviour or physical activity (see online supplementary tables S4–S6). There was no statistically significant difference after multiple testing. We analysed the associations between the methylation levels of the FAIM2 promoter and sedentary behaviour in the cohort (table 1). The methylation levels at seven CpG sites showed significant differences between obese and lean participants, with sedentary behaviour <60 min/day, especially the methylation levels at site −975, site −413, sites −362 and −360, and sites −353 and −349 (p=0.00004, 0.00009, 0.0006 and 0.00005, respectively), after adjusting for age and gender. We conducted multiple testing and these four CpG sites reached significance. The methylation levels at two CpG sites (site −500 and site −371) showed significant differences between obese and lean participants with sedentary behaviour ≥60 min/day, but these two CpG sites did not reach significance after multiple testing.

Table 1.

Associations of the FAIM2 promoter methylation with sedentary behaviour

Sedentary behaviour (<60 min/day) (mean±SD)
Sedentary behaviour (≥60 min/day) (mean±SD)
 CpG site Obese (n=8) Lean (n=24) p Value β (95% CI) Obese (n=38) Lean (n=8) p Value β (95% CI)
−1319 87.50±3.25 86.92±2.57 0.59 0.012 (−0.032 to 0.055) 86.81±2.30 85.13±2.75 0.05 0.045 (−0.001 to 0.091)
−1172 96.75±4.37 89.38±11.22 0.30 0.006 (−0.006 to 0.018) 93.42±8.51 93.75±7.61 0.36 0.007 (−0.008 to 0.022)
−1157 87.88±4.70 89.83±2.71 0.004 −0.046 (−0.076 to −0.016) 88.72±7.89 85.63±7.63 0.64 0.003 (−0.011 to 0.018)
−1106 90.25±1.98 89.46±2.67 0.95 0.002 (−0.047 to 0.05) 88.72±4.10 86.00±5.48 0.36 0.012 (−0.015 to 0.039)
−1061, −1059 79.38±15.01 86.50±8.70 0.27 −0.006 (−0.017 to 0.005) 78.97±18.97 69.13±18.59 0.48 0.002 (−0.004 to 0.008)
−975 81.25±9.66 84.92±2.41 0.00004 −0.040 (−0.057 to −0.023)) 84.83±2.25 83.50±2.56 0.41 0.023 (−0.033 to 0.078)
−931 95.13±1.89 96.88±2.17 0.14 −0.039 (−0.092 to 0.013) 96.20±3.74 94.38±6.12 0.69 0.006 (−0.023 to 0.034)
−878 67.63±7.46 67.33±4.38 0.15 0.017 (−0.006 to 0.039) 66.39±6.63 65.25±5.42 0.94 −0.001 (−0.019 to 0.018)
−859 75.13±6.36 78.29±3.65 0.12 −0.020 (−0.044 to 0.005) 74.75±9.33 72.50±5.81 0.93 −0.001 (−0.014 to 0.013)
−500 13.50±2.33 13.88±2.59 0.85 −0.005 (−0.05 to 0.043) 13.30±2.96 16.00±2.73 0.021 −0.041 (−0.075 to −0.006))
−413 5.38±5.68 3.00±0.88 0.00009 0.067 (0.037 to 0.097) 3.14±1.44 4.13±2.30 0.18 −0.046 (−0.113 to 0.021)
−399, −397, −390 6.63±4.27 6.46±3.65 0.56 −0.009 (−0.040 to 0.022) 7.68±3.54 9.50±3.85 0.57 −0.009 (−0.041 to 0.023)
−371 3.75±1.16 4.17±1.74 0.33 −0.035 (−0.107 to 0.038) 4.19±1.79 5.63±1.30 0.042 −0.062 (−0.121 to −0.002))
−362, −360 3.50±5.07 0.39±0.66 0.0006 0.065 (0.031 to 0.1) 1.40±2.01 1.29±1.60 0.72 −0.012 (−0.081 to 0.057)
−353, −349 6.25±6.45 1.96±0.86 0.00005 0.056 (0.032 to 0.08) 2.73±2.76 1.88±1.13 0.70 0.008 (−0.036 to 0.053)
−338, −336 12.25±3.81 10.08±2.02 0.028 0.046 (0.005 to 0.087) 11.43±3.27 14.50±4.24 0.16 −0.023 (−0.056 to 0.01)
−312, −307, −305 2.43±0.79 4.54±6.34 0.22 −0.010 (−0.025 to 0.006) 4.71±3.12 2.60±0.89 0.20 0.022 (−0.012 to 0.057)
−289 12.14±4.14 12.86±4.04 0.53 −0.007 (−0.029 to 0.015) 14.50±8.80 14.00±6.75 0.70 0.003 (−0.012 to 0.017)
−272, −269 11.43±2.51 18.25±13.03 0.17 −0.005 (−0.012 to 0.002) 21.29±14.82 16.17±16.85 0.28 0.004 (−0.003 to 0.011)
−259 11.43±3.36 16.08±5.69 0.044 −0.015 (−0.030 to 0) 16.75±7.52 11.00±3.74 0.08 0.012 (−0.002 to 0.026)
−225, −222, −219 1.57±1.62 1.75±5.14 0.56 0.006 (−0.014 to 0.026) 1.82±3.76 0.83±1.60 0.59 0.009 (−0.024 to 0.041)
−209 1.29±1.11 3.42±5.31 0.56 −0.005 (−0.024 to 0.013) 1.94±3.59 0.50±0.55 0.61 0.008 (−0.025 to 0.041)
−203, −201 7.57±4.96 10.25±13.26 0.10 −0.006 (−0.013 to 0.001) 13.42±11.42 10.83±8.82 0.76 0.002 (−0.009 to 0.012)
−195, −190 8.29±7.89 9.75±17.95 0.84 0.001 (−0.005 to 0.006) 20.31±24.14 9.17±12.78 0.26 0.003 (−0.002 to 0.007)
−183 5.14±1.21 5.42±4.25 0.86 0.002 (−0.021 to 0.025) 5.97±4.90 4.33±2.07 0.12 0.019 (−0.005 to 0.042)
−164 2.43±0.98 4.33±5.26 0.41 −0.007 (−0.026 to 0.011) 3.88±2.73 2.50±2.07 0.42 0.017 (−0.026 to 0.06)
−147, −145, −142 6.86±2.54 10.26±11.24 0.86 0.001 (−0.009 to 0.01) 14.94±13.90 6.20±5.07 0.25 0.004 (−0.003 to 0.012)
−134, −132, −129 2.57±1.40 5.21±9.36 0.60 0.003 (−0.008 to 0.014) 3.31±2.74 2.83±2.14 0.95 0.001 (−0.040 to 0.043)
−124 22.14±11.32 24.52±24.52 0.91 0.000 (−0.004 to 0.004) 30.56±22.11 12.17±16.41 0.24 0.003 (−0.002 to 0.008)
−115 5.14±1.21 5.42±4.25 0.86 0.002 (−0.021 to 0.025) 5.97±4.90 4.33±2.07 0.12 0.019 (−0.005 to 0.042)
−56 1.71±4.54 2.50±5.79 0.92 −0.001 (−0.018 to 0.016) 5.31±10.31 0.00±0.00 0.75 0.002 (−0.010 to 0.014)
−44, −36 8.86±1.77 9.08±8.16 0.93 0.001 (−0.012 to 0.013) 8.27±9.26 6.83±4.17 0.95 0.000 (−0.013 to 0.014)
−14, −7 5.86±3.63 4.92±5.96 0.12 −0.012 (−0.028 to 0.003) 4.00±3.29 6.00±6.96 0.60 −0.007 (−0.035 to 0.021)
19 3.86±1.35 4.83±4.22 0.55 −0.007 (−0.031 to 0.017) 4.36±2.04 4.33±2.07 0.63 −0.013 (−0.067 to 0.041)
54 32.29±9.39 27.83±15.59 0.41 0.003 (−0.004 to 0.009) 28.00±17.57 24.67±14.08 0.79 −0.001 (−0.008 to 0.006)
61 13.14±3.53 10.88±5.32 0.39 0.007 (−0.010 to 0.024) 13.56±5.03 15.17±5.27 0.16 −0.016 (−0.039 to 0.007)

p Values adjusted for age and gender. Those highlighted in bold indicate that the associations showed statistical significance.

Tables 2 and 3 show the associations between the methylation levels of the FAIM2 promoter and physical activity. There were significant differences between the methylation levels at four CpG sites in obese and lean participants with HPA or MPA level <150 min/week, but these four CpG sites did not reach significance after multiple testing. There was no statistically significant difference between obese and lean participants with HPA or MPA level ≥150 min/week.

Table 2.

Associations between the methylation levels of the FAIM2 promoter and high physical activity

High physical activity level (<30 min/day) (mean±SD)
High physical activity level (≥30 min/day) (mean±SD)
 CpG site Obese (n=32) Lean (n=10) p Value β (95% CI) Obese (n=12) Lean (n=3) p Value β (95% CI)
−1319 86.90±2.57 87.00±1.76 0.69 0.011 (−0.043 to 0.064) 86.91±2.43 86.67±1.53 0.72 0.013 (−0.064 to 0.09)
−1172 94.03±8.80 96.60±5.64 0.62 0.004 (−0.013 to 0.021) 94.73±6.02 85.00±9.54 0.022 0.021 (0.004 to 0.038)
−1157 88.16±8.45 89.80±3.68 0.15 −0.012 (−0.029 to 0.005) 89.27±4.10 86.00±2.00 0.20 0.025 (−0.016 to 0.066)
−1106 88.94±3.71 88.30±5.40 0.90 −0.002 (−0.035 to 0.031) 89.09±4.68 88.33±5.03 0.34 0.016 (−0.020 to 0.052)
−1061, −1059 79.29±18.56 82.00±11.26 0.16 −0.005 (−0.013 to 0.002) 76.91±19.12 77.33±19.50 0.88 0.001 (−0.009 to 0.01)
−975 83.94±5.34 85.00±2.83 0.22 −0.017 (−0.045 to 0.011) 84.73±2.33 87.00±3.61 0.36 −0.027 (−0.090 to 0.036)
−931 95.80±4.00 97.00±2.62 0.025 −0.040 (−0.074 to −0.005)) 96.36±1.96 95.33±4.51 0.72 0.011 (−0.056 to 0.079)
−878 66.58±7.00 67.10±4.15 0.56 −0.006 (−0.026 to 0.014) 66.36±5.32 72.00±5.00 0.12 −0.022 (−0.050 to 0.007)
−859 74.68±9.44 75.80±3.77 0.11 −0.013 (−0.028 to 0.003) 74.36±7.58 77.67±5.13 0.48 −0.008 (−0.032 to 0.016)
−500 13.10±3.06 12.80±3.12 0.77 0.006 (−0.035 to 0.047) 13.92±2.23 17.00±1.73 0.81 −0.009 (−0.088 to 0.07)
−413 3.52±3.16 3.50±1.43 0.42 0.018 (−0.027 to 0.064) 3.42±1.56 2.67±0.58 0.59 0.027 (−0.081 to 0.135)
−399, −397, −390 7.52±3.46 9.30±5.64 0.31 −0.016 (−0.048 to 0.016) 8.17±4.04 6.33±3.06 0.53 −0.013 (−0.056 to 0.03)
−371 4.03±1.56 4.80±1.99 0.16 −0.052 (−0.126 to 0.021) 4.08±2.02 4.67±2.08 0.55 −0.022 (−0.099 to 0.056)
−362, −360 1.46±2.85 0.44±1.01 0.26 0.031 (−0.024 to 0.086) 2.73±3.32 0.33±0.58 0.11 0.038 (−0.010 to 0.086)
−353, −349 2.87±2.95 1.80±1.14 0.11 0.037 (−0.009 to 0.083) 4.75±5.63 2.33±1.15 0.42 0.011 (−0.019 to 0.041)
−338, −336 11.35±3.17 12.30±3.89 0.76 0.006 (−0.036 to 0.049) 12.17±4.04 11.00±1.00 0.52 0.012 (−0.029 to 0.054)
−312, −307, −305 4.79±3.10 4.78±3.90 0.95 −0.001 (−0.041 to 0.039) 2.64±1.63 12.00±16.52 0.20 −0.013 (−0.036 to 0.009)
−289 14.90±9.24 11.75±4.80 0.39 0.006 (−0.008 to 0.021) 11.11±2.26 13.00±4.24 0.16 −0.050 (−0.126 to 0.026)
−272, −269 18.79±10.59 18.22±11.69 0.39 0.005 (−0.007 to 0.018) 22.00±20.94 30.67±17.56 0.38 −0.003 (−0.011 to 0.005)
−259 15.48±6.41 15.11±6.68 0.59 0.005 (−0.015 to 0.026) 14.50±6.56 21.33±6.66 0.37 −0.010 (−0.034 to 0.014)
−225, −222, −219 2.17±4.01 2.22±4.89 0.91 0.002 (−0.031 to 0.034) 0.73±1.27 0.00±0.00 0.19 0.087 (−0.051 to 0.225)
−209 1.97±3.18 3.44±6.19 0.10 −0.026 (−0.057 to 0.005) 1.83±3.95 0.00±0.00 0.80 0.005 (−0.039 to 0.049)
−203, −201 12.59±10.69 16.78±17.88 0.40 −0.004 (−0.015 to 0.006) 11.08±11.44 9.33±8.02 0.32 −0.007 (−0.022 to 0.008)
−195, −190 18.97±22.72 12.67±20.55 0.17 0.004 (−0.002 to 0.01) 16.33±24.26 4.67±2.52 0.48 −0.003 (−0.010 to 0.005)
−183 6.55±4.93 6.56±3.91 0.41 0.012 (−0.017 to 0.04) 4.75±3.14 4.67±2.52 0.59 0.013 (−0.039 to 0.065)
−164 4.07±2.70 3.89±3.79 0.70 −0.008 (−0.053 to 0.036) 2.64±2.01 8.67±13.28 0.14 −0.018 (−0.044 to 0.007)
−147, −145, −142 12.48±11.16 10.56±12.89 0.37 0.005 (−0.006 to 0.016) 15.08±17.58 3.50±0.71 0.70 −0.002 (−0.012 to 0.008)
−134, −132, −129 3.31±2.84 5.56±11.06 0.59 −0.006 (−0.030 to 0.017) 3.00±2.04 2.67±1.15 0.32 −0.047 (−0.144 to 0.051)
−124 30.07±23.72 20.13±12.47 0.52 0.002 (−0.004 to 0.008) 24.83±9.96 14.67±17.47 0.76 −0.002 (−0.017 to 0.013)
−115 6.55±4.93 6.56±3.91 0.41 0.012 (−0.017 to 0.04) 4.75±3.14 4.67±2.52 0.59 0.013 (−0.039 to 0.065)
−56 3.00±5.44 0.56±1.33 0.77 0.004 (−0.024 to 0.032) 4.67±7.82 0.67±1.15 0.94 0.001 (−0.022 to 0.024)
−44, −36 8.57±9.18 8.67±9.33 0.54 −0.004 (−0.019 to 0.01) 8.40±7.20 5.33±4.93 0.77 −0.004 (−0.032 to 0.025)
−14, −7 4.28±3.55 6.22±7.07 0.30 −0.014 (−0.042 to 0.013) 4.75±3.14 7.33±12.70 0.82 0.003 (−0.025 to 0.031)
19 4.34±2.04 3.33±1.87 0.80 −0.009 (−0.083 to 0.064) 4.33±1.83 9.00±7.94 0.25 −0.023 (−0.064 to 0.019)
54 26.93±13.52 32.11±11.62 0.14 −0.007 (−0.017 to 0.003) 33.00±21.88 21.00±27.71 0.88 −0.001 (−0.008 to 0.007)
61 12.93±5.04 12.22±7.77 0.71 −0.004 (−0.028 to 0.019) 14.82±4.12 8.67±4.93 0.037 0.031 (0.002 to 0.06)

p Values adjusted for age and gender. Those highlighted in bold indicate that the associations showed statistical significance.

Table 3.

Associations between the methylation levels of the FAIM2 promoter and high or moderate physical activity

High or moderate physical activity level (<150 min/week) (mean±SD)
High or moderate physical activity level (≥150 min/week) (mean±SD)
 CpG site Obese (n=28) Lean (n=15) p Value β (95% CI) Obese (n=15) Lean (n=13) p Value β (95% CI)
−1319 86.93±2.53 85.93±2.46 0.09 0.046 (−0.007 to 0.099) 86.79±2.67 85.62±1.66 0.15 0.044 (−0.017 to 0.104)
−1172 94.63±9.22 90.40±11.85 0.09 0.011 (−0.002 to 0.024) 94.00±5.57 89.69±10.56 0.10 0.013 (−0.003 to 0.029)
−1157 87.33±8.80 87.20±5.89 0.59 −0.005 (−0.022 to 0.013) 90.00±3.72 89.69±3.04 0.71 0.008 (−0.035 to 0.05)
−1106 88.74±3.81 87.53±4.41 0.92 0.002 (−0.034 to 0.038) 89.21±4.32 88.77±3.27 0.51 0.012 (−0.025 to 0.05)
−1061, −1059 78.89±19.44 78.27±16.77 0.70 −0.001 (−0.009 to 0.006) 78.29±18.10 82.00±13.31 0.53 −0.003 (−0.012 to 0.006)
−975 83.48±5.69 84.00±2.62 0.16 −0.021 (−0.051 to 0.008) 85.07±1.54 84.92±2.93 0.91 0.004 (−0.065 to 0.073)
−931 95.69±4.29 95.00±4.78 0.72 −0.006 (−0.038 to 0.027) 96.36±1.74 97.23±2.62 0.55 −0.019 (−0.085 to 0.047)
−878 66.30±7.29 65.40±4.39 0.64 0.005 (−0.017 to 0.026) 67.50±5.98 67.85±4.69 0.23 −0.016 (−0.043 to 0.011)
−859 74.48±9.98 73.87±5.59 0.34 −0.008 (−0.025 to 0.009) 74.71±7.17 78.38±3.57 0.36 −0.011 (−0.036 to 0.013)
−500 13.19±2.91 14.33±2.99 0.23 −0.027 (−0.073 to 0.018) 13.47±2.85 14.08±2.78 0.50 −0.016 (−0.065 to 0.032)
−413 3.59±3.32 3.07±1.03 0.11 0.041 (−0.009 to 0.091) 3.20±1.66 3.23±1.01 0.61 −0.025 (−0.126 to 0.075)
−399, −397, −390 7.44±3.49 7.00±3.98 0.47 0.014 (−0.025 to 0.052) 7.80±4.18 7.69±4.17 0.66 −0.007 (−0.041 to 0.026)
−371 4.19±1.36 5.07±2.09 0.12 −0.063 (−0.143 to 0.017) 3.87±2.13 4.31±0.95 0.16 −0.055 (−0.134 to 0.024)
−362, −360 1.71±2.90 0.62±0.96 0.23 0.036 (−0.024 to 0.095) 2.25±3.39 0.50±0.67 0.19 0.039 (−0.020 to 0.098)
−353, −349 2.96±3.11 1.93±1.03 0.06 0.048 (−0.002 to 0.099) 4.07±5.23 1.85±0.99 0.40 0.015 (−0.020 to 0.049)
−338, −336 11.37±3.47 12.07±3.69 0.44 0.017 (−0.027 to 0.06) 12.20±3.45 10.31±2.50 0.67 0.010 (−0.036 to 0.055)
−312, −307, −305 4.36±2.83 3.77±3.35 0.61 0.012 (−0.034 to 0.058) 3.00±1.66 5.42±8.35 0.53 −0.008 (−0.034 to 0.018)
−289 14.44±9.73 13.08±3.86 0.55 0.005 (−0.012 to 0.022) 12.83±4.45 13.70±5.74 0.81 0.004 (−0.030 to 0.038)
−272, −269 18.48±10.42 14.46±6.44 0.023 0.016 (0.002 to 0.03) 21.00±19.57 24.00±17.99 0.83 −0.001 (−0.008 to 0.007)
−259 15.12±6.78 16.15±5.65 0.87 0.002 (−0.020 to 0.024) 15.87±6.09 15.77±5.88 0.63 0.006 (−0.019 to 0.03)
−225, −222, −219 1.84±3.22 1.38±4.11 0.48 0.014 (−0.026 to 0.055) 0.57±0.85 2.38±5.75 0.62 −0.009 (−0.046 to 0.028)
−209 2.60±4.05 1.23±1.64 0.92 −0.002 (−0.046 to 0.042) 0.60±0.99 3.08±5.24 0.46 −0.014 (−0.051 to 0.024)
−203, −201 11.72±10.16 10.38±13.59 0.78 0.002 (−0.011 to 0.015) 10.53±10.51 11.85±12.54 0.35 −0.006 (−0.018 to 0.007)
−195, −190 18.16±22.58 9.46±17.36 0.036 0.007 (0 to 0.013) 13.73±22.16 11.69±19.14 0.58 −0.002 (−0.009 to 0.005)
−183 6.56±4.98 5.15±2.85 0.011 0.040 (0.01 to 0.07) 4.40±2.85 5.85±5.11 0.32 −0.017 (−0.050 to 0.017)
−164 3.88±2.47 2.69±1.70 0.29 0.032 (−0.028 to 0.092) 2.71±1.94 4.85±6.78 0.26 −0.016 (−0.045 to 0.013)
−147, −145, −142 12.16±11.49 9.54±10.49 0.18 0.008 (−0.004 to 0.021) 14.00±15.78 11.91±11.84 0.69 −0.002 (−0.013 to 0.009)
−134, −132, −129 2.76±1.69 5.54±9.16 0.40 −0.011 (−0.037 to 0.015) 3.00±1.81 5.15±9.12 0.77 −0.003 (−0.026 to 0.019)
−124 26.84±20.58 28.00±26.02 0.66 −0.001 (−0.008 to 0.005) 25.07±9.72 17.23±20.56 0.72 0.002 (−0.008 to 0.012)
−115 6.56±4.98 5.15±2.85 0.011 0.040 (0.01 to 0.07) 4.40±2.85 5.85±5.11 0.32 −0.017 (−0.050 to 0.017)
−56 3.60±5.68 0.85±1.41 0.86 0.003 (−0.029 to 0.035) 4.00±7.24 1.15±3.60 0.72 0.004 (−0.021 to 0.03)
−44, −36 8.92±9.77 5.92±4.31 0.65 0.004 (−0.013 to 0.021) 7.62±6.75 13.23±9.05 0.14 −0.013 (−0.030 to 0.004)
−14, −7 3.64±2.25 4.85±5.61 0.29 −0.020 (−0.057 to 0.017) 5.53±4.56 5.85±7.50 0.42 0.009 (−0.014 to 0.033)
19 4.52±2.06 4.54±2.76 0.13 −0.049 (−0.114 to 0.016) 4.07±1.87 4.92±4.82 0.50 −0.014 (−0.055 to 0.028)
54 28.56±17.76 25.23±14.50 0.52 −0.003 (−0.012 to 0.006) 29.43±14.96 25.75±16.68 0.70 0.002 (−0.008 to 0.012)
61 12.92±4.37 11.31±6.99 0.47 0.009 (−0.017 to 0.036) 14.43±5.65 11.92±4.77 0.59 −0.008 (−0.037 to 0.022)

p Values adjusted for age and gender. Those highlighted in bold indicate that the associations showed statistical significance.

There was no statistically significant interaction between the methylation levels of the FAIM2 promoter and physical activity regarding obesity status after multiple testing (see online supplementary table S7).

Discussion

Previous studies have shown that lifestyle factors might modify epigenetic patterns3 and the methylation levels of the FAIM2 promoter are significantly associated with obesity,10 but whether the molecular mechanisms by which FAIM2 affects obesity are involved in lifestyle is unclear. In this study, we investigated the potential associations of the FAIM2 promoter methylation with sedentary behaviour and physical activity in obese and lean children. The methylation levels of the FAIM2 promoter were significantly different in obese and lean children when the groups were stratified by sedentary behaviour and physical activity. There were significant differences in the associations between the FAIM2 promoter methylation with sedentary behaviour and physical activity between obese and lean participants. Our results suggest that lifestyle may possibly be mediating the process of the FAIM2 involved in obesity.

FAIM2 can bind directly to the Fas receptor to protect cells against apoptosis,5 13 which is a nervous system-specific inhibitor of Fas/CD95-mediated apoptosis. GWA studies in Caucasians and subsequently replicated studies in Asian populations showed that the SNP rs7138803 near FAIM2 was significantly associated with obesity.6–8 Recently, it was reported that the expression of FAIM2 was regulated by high fat feeding in the brain,14 and that the methylation levels of the FAIM2 promoter were significantly associated with obesity.10

Increasing evidence shows that lifestyle factors may affect epigenetic alterations, especially DNA methylation, which has been identified to be associated with a variety of human diseases including obesity.3 4 15 For example, physical activity is associated with changes in DNA methylation at many genes related to adipocyte metabolism and type 2 diabetes.16 17

This study may not provide direct evidence that the expression of FAIM2 influences obesity, because of the lack of gene expression data from leucocytes and tissues, but our study demonstrated the differences of the associations of the FAIM2 promoter methylation with sedentary behaviour and physical activity between obese and lean children.

There are a few limitations to this study. First, the age range of the obese and lean participants in this study is wide and the sample size is small. Studies with greater sample sizes are needed to examine the associations. Second, obese participants were on average 3.5 years older than lean participants. We want to eliminate the effect by adjusting for age. It is important to adjust the stage of puberty in the analyses, so we will collect data on the stage of puberty in future studies. Third, the data of sedentary behaviour and physical activity were collected by questionnaires. The reliability of questionnaires depends on good communication between the parents and children. Fourth, there were no gene expression data from leucocytes or tissues in the study. Previous studies showed the expression of FAIM2 in the hypothalamus or some other tissues, but there were no expression data of this gene in peripheral blood leucocytes. Fifth, our study investigated the methylation levels in peripheral blood leucocytes, but not in the hypothalamus or adipocytes.

Conclusions

In this study we explored, for the first time, the significant differences in the associations between the FAIM2 promoter methylation with sedentary behaviour and physical activity between obese and lean children. Although the function of FAIM2 in the brain remains to be fully researched, identification of methylation changes affected by lifestyle factors will provide important potential to develop new therapies to prevent obesity.

Acknowledgments

The authors thank the children and their parents for their participation in this study. They also thank Ping Yang, Zhongjian Su, Xinying Shan, Fangfang Chen and Wenpeng Wang, for their help with sample collection.

Footnotes

Contributors: LW designed the study and wrote the manuscript. LW and GH collected the data. YS performed the statistical analysis. XZ, MZ, YY, DH, LM, JL and HC collected the DNA samples. All the authors reviewed the manuscript. LW and XZ contributed equally to this work. JM directed the project.

Funding: This study was supported by the National Natural Science Foundation of China (grant no. 81200624) and the National Basic Research Program of China (973 Program, grant no. 2013CB530605).

Competing interests: None declared.

Patient consent: Obtained.

Ethics approval: The study was approved by the ethics committees of the Capital Institute of Pediatrics.

Provenance and peer review: Not commissioned; externally peer reviewed.

Data sharing statement: No additional data are available.

References

  • 1.Wang Y, Lobstein T. Worldwide trends in childhood overweight and obesity. Int J Pediatr Obes 2006;1:11–25. 10.1080/17477160600586747 [DOI] [PubMed] [Google Scholar]
  • 2.Xi B, Wang C, Wu L et al. Influence of physical inactivity on associations between single nucleotide polymorphisms and genetic predisposition to childhood obesity. Am J Epidemiol 2011;173:1256–62. 10.1093/aje/kwr008 [DOI] [PubMed] [Google Scholar]
  • 3.Slattery ML, Curtin K, Sweeney C et al. Diet and lifestyle factor associations with CpG island methylator phenotype and BRAF mutations in colon cancer. Int J Cancer 2007;120:656–63. 10.1002/ijc.22342 [DOI] [PubMed] [Google Scholar]
  • 4.Wang X, Zhu H, Snieder H et al. Obesity related methylation changes in DNA of peripheral blood leukocytes. BMC Med 2010;8:87 10.1186/1741-7015-8-87 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Fernandez M, Segura MF, Sole C et al. Lifeguard/neuronal membrane protein 35 regulates Fas ligand-mediated apoptosis in neurons via microdomain recruitment. J Neurochem 2007;103:190–203. [DOI] [PubMed] [Google Scholar]
  • 6.Thorleifsson G, Walters GB, Gudbjartsson DF et al. Genome-wide association yields new sequence variants at seven loci that associate with measures of obesity. Nat Genet 2009;41:18–24. 10.1038/ng.274 [DOI] [PubMed] [Google Scholar]
  • 7.Wu L, Xi B, Zhang M et al. Associations of six single nucleotide polymorphisms in obesity-related genes with BMI and risk of obesity in Chinese children. Diabetes 2010;59:3085–9. 10.2337/db10-0273 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hotta K, Nakamura M, Nakamura T et al. Association between obesity and polymorphisms in SEC16B, TMEM18, GNPDA2, BDNF, FAIM2 and MC4R in a Japanese population. J Hum Genet 2009;54:727–31. 10.1038/jhg.2009.106 [DOI] [PubMed] [Google Scholar]
  • 9.Boender AJ, van Rozen AJ, Adan RA. Nutritional state affects the expression of the obesity-associated genes Etv5, Faim2, Fto, and Negr1. Obesity 2012;20:2420–5. 10.1038/oby.2012.128 [DOI] [PubMed] [Google Scholar]
  • 10.Wu L, Zhao X, Shen Y et al. Promoter methylation of fas apoptotic inhibitory molecule 2 gene is associated with obesity and dyslipidaemia in Chinese children. Diab Vasc Dis Res 2015;12:217–20. 10.1177/1479164114565630 [DOI] [PubMed] [Google Scholar]
  • 11.Meng L, Liang Y, Liu J et al. Prevalence and risk factors of hypertension based on repeated measurements in Chinese children and adolescents. Blood Press 2013;22:59–64. 10.3109/08037051.2012.701790 [DOI] [PubMed] [Google Scholar]
  • 12.Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 1995;57:289–300. [Google Scholar]
  • 13.Somia NV, Schmitt MJ, Vetter DE et al. LFG: an anti-apoptotic gene that provides protection from Fas-mediated cell death. Proc Natl Acad Sci USA 1999;96:12667–72. 10.1073/pnas.96.22.12667 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Yoganathan P, Karunakaran S, Ho MM et al. Nutritional regulation of genome-wide association obesity genes in a tissue-dependent manner. Nutr Metab (Lond) 2012;9:65 10.1186/1743-7075-9-65 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Moleres A, Campion J, Milagro FI et al. Differential DNA methylation patterns between high and low responders to a weight loss intervention in overweight or obese adolescents: the EVASYON study. FASEB J 2013;27:2504–12. 10.1096/fj.12-215566 [DOI] [PubMed] [Google Scholar]
  • 16.Ronn T, Volkov P, Davegardh C et al. A six months exercise intervention influences the genome-wide DNA methylation pattern in human adipose tissue. PLoS Genet 2013;9:e1003572 10.1371/journal.pgen.1003572 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nitert MD, Dayeh T, Volkov P et al. Impact of an exercise intervention on DNA methylation in skeletal muscle from first-degree relatives of patients with type 2 diabetes. Diabetes 2012;61:3322–32. 10.2337/db11-1653 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from BMJ Open are provided here courtesy of BMJ Publishing Group

RESOURCES