Accepted Manuscript Title: Brain-derived neurotrophic factor (BDNF) Val66Met polymorphism affects sympathetic tone in a gender-specific way Author: Chuan-Chia Chang Hsin-An Chang Tien-Yu Chen Wen-Hui Fang San-Yuan Huang PII: DOI: Reference:
S0306-4530(14)00152-8 http://dx.doi.org/doi:10.1016/j.psyneuen.2014.04.019 PNEC 2684
To appear in: Received date: Revised date: Accepted date:
12-2-2014 24-4-2014 25-4-2014
Please cite this article as: Chang, C.-C., Chang, H.-A., Chen, T.-Y., Fang, W.-H., Huang, S.-Y.,Brain-derived neurotrophic factor (BDNF) Val66Met polymorphism affects sympathetic tone in a gender-specific way, Psychoneuroendocrinology (2014), http://dx.doi.org/10.1016/j.psyneuen.2014.04.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Brain-derived neurotrophic factor (BDNF) Val66Met polymorphism affects sympathetic tone in a gender-specific way
ip t
Chuan-Chia Chang, M.D.a,b, Hsin-An Chang, M.D.a, Tien-Yu Chen, M.D.a,
us
cr
Wen-Hui Fang M.D.c*, and San-Yuan Huang, M.D., Ph.D.a,b*
Running title: BDNF gene and sympathetic tone
an
Word count: Abstract-214; Text-3531
M
Numbers: Figure-1; Table-3; Supplementary Table-1; References-62
*
Ac ce p
te
d
Corresponding authors: (1) San-Yuan Huang, M.D., Ph.D. Professor and Attending Psychiatrist, Department of Psychiatry, Tri-Service General Hospital, Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan (2) Wen-Hui Fang, M.D. Attending Physician, Department of Family and Community Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan
Address: No. 325, Sec. 2, Cheng-Kung Road, Nei-Hu District, Taipei, 114, Taiwan, ROC Tel: +886-2-8792-7220; Fax: +886-2-8792-7221 E-mail: [email protected] (SY Huang) ________________________ a Department of Psychiatry, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan c Department of Family and Community Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan
Page 1 of 34
Summary The Val/Val genotype of the brain-derived neurotrophic factor (BDNF) polymorphism (Val66Met) has been reported to affect human anxiety-related phenotypes. Substantial
ip t
research has demonstrated that anxiety is associated with sympathetic activation, while sex steroid hormones have been shown to exert differential actions in regulating
cr
BDNF expression. Thus, we examined whether the BDNF variant modulates
us
autonomic function in a gender-dependent manner. From 708 adults initially screened for medical and psychiatric illnesses, a final cohort of 583 drug-free healthy Han
an
Chinese (355 males, 228 females; age 34.43±8.42 years) was recruited for BDNF genotyping (Val/Val: 136, 23.3%, Val/Met: 294, 50.4%, and Met/Met: 153, 26.2%).
M
Time- and frequency-domain analyses of heart rate variability (HRV) were used to assess autonomic outflow to the heart. Significant genotype-by-gender interaction
d
effects were found on HRV indices. Even after adjusting for possible confounders,
te
male participants bearing the Val/Val genotype had significant increases in low frequency (LF), LF% and LF/high frequency (HF) ratio, indicating altered
Ac ce p
sympathovagal balance with increased sympathetic modulation, compared to male Met/Met homozygotes. Females, however, showed an opposite but non-significant pattern. These results suggest that the studied BDNF polymorphism is associated with sympathetic control in a gender-specific way. The findings here support the view that male subjects with the Val/Val genotype have increased risk of anxiety by association with sympathetic activation.
Keywords: Brain-derived neurotrophic factor (BDNF); polymorphism; sympathetic control; autonomic nervous system (ANS); heart rate variability (HRV)
2
Page 2 of 34
Introduction Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, influences almost all aspects of survival, differentiation, and function of the neurons
ip t
in the peripheral and central nervous systems (Lindsay et al., 1994). Several lines of evidence indicate that BDNF may not only have an effect on serotonergic
cr
neurotransmission (Benmansour et al., 2008), but also modulate the autonomic
us
nervous system (ANS) (Slonimsky et al., 2003; Clark et al., 2011). The BDNF gene, like other peptide growth factors, encodes a precursor peptide (pro-BDNF) that is
an
subsequently cleaved to form the mature protein. A naturally occurring variation in humans in the form of a common single nucleotide polymorphism (SNP) at nucleotide
M
196 (G/A) producing an amino acid substitution (valine to methionine) at codon 66 (Val66Met) affects the processing of the pro-BDNF polypeptide and its
d
activation-dependent release (Egan et al., 2003). More importantly, the BDNF
te
Val66Met polymorphism (rs6265) has been reported to be associated with human anxiety-related phenotypes. Despite the conflicting associations of the BDNF
Ac ce p
polymorphism with human anxiety-related phenotypes (Jiang et al., 2005; Lang et al., 2005; Kim et al., 2010; Yang et al., 2010; Arias et al., 2012), a meta-analysis has reported that the studied BDNF Val/Val genotype is associated with trait anxiety/neuroticism (Frustaci et al., 2008), which is an important risk factor for anxiety disorders.
Substantial research that analyzes heart rate variability (HRV), a non-invasive
electrocardiographic (ECG) index enabling quantitative assessment of ANS activity (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996), has shown that physically healthy patients with anxiety disorders, such as panic disorder and post-traumatic stress disorder, are
3
Page 3 of 34
associated with sympathetic activation and/or parasympathetic (vagal) withdrawal (Friedman and Thayer, 1998; Yeragani et al., 1998; Cohen et al., 2000; Mellman et al., 2004; Martinez et al., 2010). Indeed, autonomic dysregulation, particularly increased
ip t
sympathetic tone, has been considered to play a critical role in the development of anxiety disorders (for a review, see Cohen and Benjamin, 2006). Furthermore,
cr
patients with anxiety disorders have increased risk of cardiovascular vulnerability,
including ventricular arrhythmias, coronary heart disease, myocardial infarction, and
us
sudden cardiac death (Gorman and Sloan, 2000; Cohen and Benjamin, 2006). A recent
an
meta-analysis has shown that anxiety is associated with a 26% increased risk of coronary heart disease and a 48% increased risk of cardiac mortality (Roest et al.,
M
2010). Since changes in cardiac ANS activity are known risk factors for cardiovascular diseases (Meredith et al., 1991), autonomic dysregulation is assumed
d
to be one of the most plausible explanations that link anxiety with adverse
te
cardiovascular events (Cohen et al., 1999; Martens et al., 2008; Miu et al., 2009). However, little is known about the underlying molecular mechanisms responsible for
Ac ce p
ANS dysregulation. Heritability studies have demonstrated that genetic determinants contribute to a substantial proportion of the variance in HRV (Singh et al., 1999). Therefore, investigating the underlying genetic influences of BDNF on HRV is crucial for understanding the modulation of ANS and providing possible insights into the complex pathophysiology of anxiety and cardiovascular co-morbidities. Yang et al. (2010) investigated the association between BDNF Val66Met
polymorphism and autonomic function in an apparently healthy human sample (n=211; 58 males and 153 females) by employing an analysis of HRV. Their subjects with the Met/Met genotype manifested with altered sympathovagal balance and relatively decreased parasympathetic tone. However, ANS activity can be modulated
4
Page 4 of 34
by gonadal hormones, androgen (Mirabella et al., 2006) and estrogen (Krizsan-Agbas et al., 2003), in a BDNF-dependent manner. Furthermore, growing evidence indicates that sex steroid hormones are involved in the regulation of BDNF gene expression
ip t
(Solum and Handa, 2002; Cavus and Duman, 2003; Mirabella et al., 2006). All of these suggest that autonomic modulation in men and women may differ according to
cr
distinct sex hormonal influences on BDNF availability. Indeed, two recent human
studies have demonstrated that male subjects with the BDNF Val/Val genotype have a
us
greater rise in autonomic-mediated heart rate than Met-allele carriers while exposed to
an
acute psychological stress (Shalev et al., 2009; Alexander et al., 2010), and that females have an opposite response (Shalev et al., 2009). Thus, studying the effect of
M
BDNF-variation on ANS modulation by HRV analysis in a large sample and in a gender-stratified manner is imperative. In addition, HRV is influenced by many
d
factors like age, cigarette smoking, physical activity levels, position, hour of the day,
te
medications, serum metabolic measurements, and morbidities (Cohen et al., 1999; Chang et al., 2012; Chang et al., 2013). Efforts to control these non-genetic
Ac ce p
confounding effects are also needed.
The present study aimed to determine whether the studied BDNF Val66Met
polymorphism is associated with gender-specific autonomic modulation. This study also tested whether the gender-stratified effects of BDNF Val66Met variant on HRV are apparent and remain after adjusting for known confounders in a large sample of drug-free healthy Han Chinese subjects.
Methods Subjects The study cohort was composed of volunteers who underwent annual health
5
Page 5 of 34
examinations at the Tri-Service General Hospital, a medical teaching hospital of the National Defense Medical Center in Taipei, Taiwan. All were of Han Chinese ancestry. After detailed questionnaire screening, subjects taking any medication for at least one
ip t
month prior to the start of the study or those with a personal history of medical diseases, psychiatric illnesses, substance dependence, or pregnancy were excluded.
cr
After initial screening, 708 adult Han Chinese were recruited. All provided informed consent before participating. The hospital’s institutional review board approved the
us
study protocol, which adhered to the guidelines of the Declaration of Helsinki. The
smoking status, and weekly exercise level.
M
Assessment of psychiatric morbidity
an
demographic data collected included age, gender, body mass index (BMI: kg/m2),
Each enrolled subject was further evaluated by a trained research assistant using
d
the Chinese version of the Mini-International Neuropsychiatric Interview (MINI)
te
(Sheehan et al., 1998), a structured diagnostic instrument designed to yield psychiatric diagnoses according to the criteria of the Diagnostic and Statistical Manual of Mental
Ac ce p
Disorders, Fourth Edition (DSM-IV) (APA, 1994). Participants diagnosed with mental illness were excluded.
Assessment of medical conditions All of the study participants received health check-ups, which included physical
examination, biochemical analysis (blood, urine, and stool specimens), and chest X-ray and ECG examinations. Blood pressure (systolic [SBP] and diastolic [DBP]) was measured. Fasting plasma glucose was determined by the glucose oxidase method, while triglyceride and total cholesterol levels were measured using the dry, multi-layer analytical slide method. Participants with organic diseases like cardiovascular diseases (e.g., hypertension,
6
Page 6 of 34
arrhythmia), metabolic disorders (e.g., hypercholesterolemia, hypertriglyceridemia, diabetes mellitus), liver or kidney diseases, malignancy, neuropathy, or obesity (BMI ≥30 kg/m2) were excluded. A cohort of 583 healthy subjects (355 males, 228 females;
ip t
age 34.43±8.42 years) composed the final study sample. Measurements of heart rate variability
cr
An electrocardiogram analyzer (SA-3000P; Medicore Co., Ltd., Korea) was used to acquire, store, and process ECG signals for measuring HRV (Chang et al., 2012).
us
All of the subjects were examined in a quiet room with standard temperature, while
an
ECG monitoring was conducted in the morning (08:00-12:00 a.m.) to accommodate diurnal fluctuations. After 15 min of sitting at rest, each participant underwent ECG
M
recording with normal breathing in a sitting position for 5 min. The average heart rate (beats per minute) was derived from the R waves of the ECG.
d
The system automatically analyzed changes in heart rate using time-domain and
te
frequency-domain analyses (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The root mean
Ac ce p
square of successive R-R interval differences (RMSSD) was used to compare the time-domain index (set as milliseconds) and reflected cardiac parasympathetic activity (Goldberger et al., 2001). Power spectral analysis of changes in heart rate, using the fast-Fourier transformation, was then quantified into standard frequency-domain measurements, which consisted of total power (TP: 0-0.40 Hz), very low frequency power (VLF: 0-0.04 Hz), low frequency (LF: 0.04-0.15 Hz), high frequency (HF: 0.15-0.4 Hz), ratio of LF to HF (LF/HF), and normalized LF (LF%) (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The LF% was calculated from LF/(TP-VLF) × 100, while the LF, HF and LF/HF
7
Page 7 of 34
measurements were logarithmically transformed to correct skewed distribution. The HF represented vagal control of HRV, whereas LF represented both vagal and sympathetic control of HRV. The LF% and LF/HF ratio reflected the global
Society of Cardiology and the North American Society of Pacing and
cr
Electrophysiology, 1996).
ip t
sympathovagal balance or sympathetic modulations (Task Force of the European
Laboratory SNP genotyping
us
A QIAamp DNA Blood kit (Qiagen®, Valencia, CA) isolated DNA from blood samples. Agarose gel electrophoresis analysis assessed the quality of the isolated
an
genomic DNA for each sample, while spectro-photometry (NanoDrop®, Wilmington,
M
USA) determined the quantity of DNA in each sample. The BDNF SNP rs6265 was genotyped by the TaqMan® 5'-exonuclease allelic discrimination assay as described
Statistical analysis
te
instrument.
d
previously (Livak, 1999) using an Applied Biosystems (ABI®, CA, USA) Prism 7900
Ac ce p
The allele and genotype frequencies of the BDNF gene were calculated and
genotypic distribution was compared to the predicted value from the Hardy-Weinberg equilibrium. χ2 statistics compared categorical variables and ANOVA compared
continuous variables among the BDNF genotype groups. Bonferroni post hoc test was used for corrections of multiple comparisons. ANCOVA was employed to investigate the main and interaction effects of
genotype and gender on the HRV indices. To control for non-genetic confounders, the association between HRV indices and demographic/clinical variables, including age, BMI, smoking status, physical exercise levels, and serum metabolic parameters were tested by gender (Cohen et al., 1999; Chang et al., 2012). Pearson’s correlation was
8
Page 8 of 34
used to evaluate the relations between variables with normal distribution. Spearman’s correlation test was applied for non-normal distribution. Variables associated with HRV were used as covariates in separate ANCOVA models testing the effects of
ip t
genotype on each of the HRV indices. Power calculations were carried out using Quanto software version 1.2.4 (Gauderman and Morrison, 2006). The total sample
cr
(n=583) had a power of 99% to detect a proportion of variance, R2=.05, in HRV
indices explained by gene x gender effects. Furthermore, the powers of the HRV
us
indices explained by genetic effects in males (n=355) and females (n=228) were 99%
an
and 93%, respectively. Statistical significance was set at p
S0306-4530(14)00152-8 http://dx.doi.org/doi:10.1016/j.psyneuen.2014.04.019 PNEC 2684
To appear in: Received date: Revised date: Accepted date:
12-2-2014 24-4-2014 25-4-2014
Please cite this article as: Chang, C.-C., Chang, H.-A., Chen, T.-Y., Fang, W.-H., Huang, S.-Y.,Brain-derived neurotrophic factor (BDNF) Val66Met polymorphism affects sympathetic tone in a gender-specific way, Psychoneuroendocrinology (2014), http://dx.doi.org/10.1016/j.psyneuen.2014.04.019 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Brain-derived neurotrophic factor (BDNF) Val66Met polymorphism affects sympathetic tone in a gender-specific way
ip t
Chuan-Chia Chang, M.D.a,b, Hsin-An Chang, M.D.a, Tien-Yu Chen, M.D.a,
us
cr
Wen-Hui Fang M.D.c*, and San-Yuan Huang, M.D., Ph.D.a,b*
Running title: BDNF gene and sympathetic tone
an
Word count: Abstract-214; Text-3531
M
Numbers: Figure-1; Table-3; Supplementary Table-1; References-62
*
Ac ce p
te
d
Corresponding authors: (1) San-Yuan Huang, M.D., Ph.D. Professor and Attending Psychiatrist, Department of Psychiatry, Tri-Service General Hospital, Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan (2) Wen-Hui Fang, M.D. Attending Physician, Department of Family and Community Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan
Address: No. 325, Sec. 2, Cheng-Kung Road, Nei-Hu District, Taipei, 114, Taiwan, ROC Tel: +886-2-8792-7220; Fax: +886-2-8792-7221 E-mail: [email protected] (SY Huang) ________________________ a Department of Psychiatry, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan b Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei, Taiwan c Department of Family and Community Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan
Page 1 of 34
Summary The Val/Val genotype of the brain-derived neurotrophic factor (BDNF) polymorphism (Val66Met) has been reported to affect human anxiety-related phenotypes. Substantial
ip t
research has demonstrated that anxiety is associated with sympathetic activation, while sex steroid hormones have been shown to exert differential actions in regulating
cr
BDNF expression. Thus, we examined whether the BDNF variant modulates
us
autonomic function in a gender-dependent manner. From 708 adults initially screened for medical and psychiatric illnesses, a final cohort of 583 drug-free healthy Han
an
Chinese (355 males, 228 females; age 34.43±8.42 years) was recruited for BDNF genotyping (Val/Val: 136, 23.3%, Val/Met: 294, 50.4%, and Met/Met: 153, 26.2%).
M
Time- and frequency-domain analyses of heart rate variability (HRV) were used to assess autonomic outflow to the heart. Significant genotype-by-gender interaction
d
effects were found on HRV indices. Even after adjusting for possible confounders,
te
male participants bearing the Val/Val genotype had significant increases in low frequency (LF), LF% and LF/high frequency (HF) ratio, indicating altered
Ac ce p
sympathovagal balance with increased sympathetic modulation, compared to male Met/Met homozygotes. Females, however, showed an opposite but non-significant pattern. These results suggest that the studied BDNF polymorphism is associated with sympathetic control in a gender-specific way. The findings here support the view that male subjects with the Val/Val genotype have increased risk of anxiety by association with sympathetic activation.
Keywords: Brain-derived neurotrophic factor (BDNF); polymorphism; sympathetic control; autonomic nervous system (ANS); heart rate variability (HRV)
2
Page 2 of 34
Introduction Brain-derived neurotrophic factor (BDNF), a member of the neurotrophin family, influences almost all aspects of survival, differentiation, and function of the neurons
ip t
in the peripheral and central nervous systems (Lindsay et al., 1994). Several lines of evidence indicate that BDNF may not only have an effect on serotonergic
cr
neurotransmission (Benmansour et al., 2008), but also modulate the autonomic
us
nervous system (ANS) (Slonimsky et al., 2003; Clark et al., 2011). The BDNF gene, like other peptide growth factors, encodes a precursor peptide (pro-BDNF) that is
an
subsequently cleaved to form the mature protein. A naturally occurring variation in humans in the form of a common single nucleotide polymorphism (SNP) at nucleotide
M
196 (G/A) producing an amino acid substitution (valine to methionine) at codon 66 (Val66Met) affects the processing of the pro-BDNF polypeptide and its
d
activation-dependent release (Egan et al., 2003). More importantly, the BDNF
te
Val66Met polymorphism (rs6265) has been reported to be associated with human anxiety-related phenotypes. Despite the conflicting associations of the BDNF
Ac ce p
polymorphism with human anxiety-related phenotypes (Jiang et al., 2005; Lang et al., 2005; Kim et al., 2010; Yang et al., 2010; Arias et al., 2012), a meta-analysis has reported that the studied BDNF Val/Val genotype is associated with trait anxiety/neuroticism (Frustaci et al., 2008), which is an important risk factor for anxiety disorders.
Substantial research that analyzes heart rate variability (HRV), a non-invasive
electrocardiographic (ECG) index enabling quantitative assessment of ANS activity (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996), has shown that physically healthy patients with anxiety disorders, such as panic disorder and post-traumatic stress disorder, are
3
Page 3 of 34
associated with sympathetic activation and/or parasympathetic (vagal) withdrawal (Friedman and Thayer, 1998; Yeragani et al., 1998; Cohen et al., 2000; Mellman et al., 2004; Martinez et al., 2010). Indeed, autonomic dysregulation, particularly increased
ip t
sympathetic tone, has been considered to play a critical role in the development of anxiety disorders (for a review, see Cohen and Benjamin, 2006). Furthermore,
cr
patients with anxiety disorders have increased risk of cardiovascular vulnerability,
including ventricular arrhythmias, coronary heart disease, myocardial infarction, and
us
sudden cardiac death (Gorman and Sloan, 2000; Cohen and Benjamin, 2006). A recent
an
meta-analysis has shown that anxiety is associated with a 26% increased risk of coronary heart disease and a 48% increased risk of cardiac mortality (Roest et al.,
M
2010). Since changes in cardiac ANS activity are known risk factors for cardiovascular diseases (Meredith et al., 1991), autonomic dysregulation is assumed
d
to be one of the most plausible explanations that link anxiety with adverse
te
cardiovascular events (Cohen et al., 1999; Martens et al., 2008; Miu et al., 2009). However, little is known about the underlying molecular mechanisms responsible for
Ac ce p
ANS dysregulation. Heritability studies have demonstrated that genetic determinants contribute to a substantial proportion of the variance in HRV (Singh et al., 1999). Therefore, investigating the underlying genetic influences of BDNF on HRV is crucial for understanding the modulation of ANS and providing possible insights into the complex pathophysiology of anxiety and cardiovascular co-morbidities. Yang et al. (2010) investigated the association between BDNF Val66Met
polymorphism and autonomic function in an apparently healthy human sample (n=211; 58 males and 153 females) by employing an analysis of HRV. Their subjects with the Met/Met genotype manifested with altered sympathovagal balance and relatively decreased parasympathetic tone. However, ANS activity can be modulated
4
Page 4 of 34
by gonadal hormones, androgen (Mirabella et al., 2006) and estrogen (Krizsan-Agbas et al., 2003), in a BDNF-dependent manner. Furthermore, growing evidence indicates that sex steroid hormones are involved in the regulation of BDNF gene expression
ip t
(Solum and Handa, 2002; Cavus and Duman, 2003; Mirabella et al., 2006). All of these suggest that autonomic modulation in men and women may differ according to
cr
distinct sex hormonal influences on BDNF availability. Indeed, two recent human
studies have demonstrated that male subjects with the BDNF Val/Val genotype have a
us
greater rise in autonomic-mediated heart rate than Met-allele carriers while exposed to
an
acute psychological stress (Shalev et al., 2009; Alexander et al., 2010), and that females have an opposite response (Shalev et al., 2009). Thus, studying the effect of
M
BDNF-variation on ANS modulation by HRV analysis in a large sample and in a gender-stratified manner is imperative. In addition, HRV is influenced by many
d
factors like age, cigarette smoking, physical activity levels, position, hour of the day,
te
medications, serum metabolic measurements, and morbidities (Cohen et al., 1999; Chang et al., 2012; Chang et al., 2013). Efforts to control these non-genetic
Ac ce p
confounding effects are also needed.
The present study aimed to determine whether the studied BDNF Val66Met
polymorphism is associated with gender-specific autonomic modulation. This study also tested whether the gender-stratified effects of BDNF Val66Met variant on HRV are apparent and remain after adjusting for known confounders in a large sample of drug-free healthy Han Chinese subjects.
Methods Subjects The study cohort was composed of volunteers who underwent annual health
5
Page 5 of 34
examinations at the Tri-Service General Hospital, a medical teaching hospital of the National Defense Medical Center in Taipei, Taiwan. All were of Han Chinese ancestry. After detailed questionnaire screening, subjects taking any medication for at least one
ip t
month prior to the start of the study or those with a personal history of medical diseases, psychiatric illnesses, substance dependence, or pregnancy were excluded.
cr
After initial screening, 708 adult Han Chinese were recruited. All provided informed consent before participating. The hospital’s institutional review board approved the
us
study protocol, which adhered to the guidelines of the Declaration of Helsinki. The
smoking status, and weekly exercise level.
M
Assessment of psychiatric morbidity
an
demographic data collected included age, gender, body mass index (BMI: kg/m2),
Each enrolled subject was further evaluated by a trained research assistant using
d
the Chinese version of the Mini-International Neuropsychiatric Interview (MINI)
te
(Sheehan et al., 1998), a structured diagnostic instrument designed to yield psychiatric diagnoses according to the criteria of the Diagnostic and Statistical Manual of Mental
Ac ce p
Disorders, Fourth Edition (DSM-IV) (APA, 1994). Participants diagnosed with mental illness were excluded.
Assessment of medical conditions All of the study participants received health check-ups, which included physical
examination, biochemical analysis (blood, urine, and stool specimens), and chest X-ray and ECG examinations. Blood pressure (systolic [SBP] and diastolic [DBP]) was measured. Fasting plasma glucose was determined by the glucose oxidase method, while triglyceride and total cholesterol levels were measured using the dry, multi-layer analytical slide method. Participants with organic diseases like cardiovascular diseases (e.g., hypertension,
6
Page 6 of 34
arrhythmia), metabolic disorders (e.g., hypercholesterolemia, hypertriglyceridemia, diabetes mellitus), liver or kidney diseases, malignancy, neuropathy, or obesity (BMI ≥30 kg/m2) were excluded. A cohort of 583 healthy subjects (355 males, 228 females;
ip t
age 34.43±8.42 years) composed the final study sample. Measurements of heart rate variability
cr
An electrocardiogram analyzer (SA-3000P; Medicore Co., Ltd., Korea) was used to acquire, store, and process ECG signals for measuring HRV (Chang et al., 2012).
us
All of the subjects were examined in a quiet room with standard temperature, while
an
ECG monitoring was conducted in the morning (08:00-12:00 a.m.) to accommodate diurnal fluctuations. After 15 min of sitting at rest, each participant underwent ECG
M
recording with normal breathing in a sitting position for 5 min. The average heart rate (beats per minute) was derived from the R waves of the ECG.
d
The system automatically analyzed changes in heart rate using time-domain and
te
frequency-domain analyses (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The root mean
Ac ce p
square of successive R-R interval differences (RMSSD) was used to compare the time-domain index (set as milliseconds) and reflected cardiac parasympathetic activity (Goldberger et al., 2001). Power spectral analysis of changes in heart rate, using the fast-Fourier transformation, was then quantified into standard frequency-domain measurements, which consisted of total power (TP: 0-0.40 Hz), very low frequency power (VLF: 0-0.04 Hz), low frequency (LF: 0.04-0.15 Hz), high frequency (HF: 0.15-0.4 Hz), ratio of LF to HF (LF/HF), and normalized LF (LF%) (Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology, 1996). The LF% was calculated from LF/(TP-VLF) × 100, while the LF, HF and LF/HF
7
Page 7 of 34
measurements were logarithmically transformed to correct skewed distribution. The HF represented vagal control of HRV, whereas LF represented both vagal and sympathetic control of HRV. The LF% and LF/HF ratio reflected the global
Society of Cardiology and the North American Society of Pacing and
cr
Electrophysiology, 1996).
ip t
sympathovagal balance or sympathetic modulations (Task Force of the European
Laboratory SNP genotyping
us
A QIAamp DNA Blood kit (Qiagen®, Valencia, CA) isolated DNA from blood samples. Agarose gel electrophoresis analysis assessed the quality of the isolated
an
genomic DNA for each sample, while spectro-photometry (NanoDrop®, Wilmington,
M
USA) determined the quantity of DNA in each sample. The BDNF SNP rs6265 was genotyped by the TaqMan® 5'-exonuclease allelic discrimination assay as described
Statistical analysis
te
instrument.
d
previously (Livak, 1999) using an Applied Biosystems (ABI®, CA, USA) Prism 7900
Ac ce p
The allele and genotype frequencies of the BDNF gene were calculated and
genotypic distribution was compared to the predicted value from the Hardy-Weinberg equilibrium. χ2 statistics compared categorical variables and ANOVA compared
continuous variables among the BDNF genotype groups. Bonferroni post hoc test was used for corrections of multiple comparisons. ANCOVA was employed to investigate the main and interaction effects of
genotype and gender on the HRV indices. To control for non-genetic confounders, the association between HRV indices and demographic/clinical variables, including age, BMI, smoking status, physical exercise levels, and serum metabolic parameters were tested by gender (Cohen et al., 1999; Chang et al., 2012). Pearson’s correlation was
8
Page 8 of 34
used to evaluate the relations between variables with normal distribution. Spearman’s correlation test was applied for non-normal distribution. Variables associated with HRV were used as covariates in separate ANCOVA models testing the effects of
ip t
genotype on each of the HRV indices. Power calculations were carried out using Quanto software version 1.2.4 (Gauderman and Morrison, 2006). The total sample
cr
(n=583) had a power of 99% to detect a proportion of variance, R2=.05, in HRV
indices explained by gene x gender effects. Furthermore, the powers of the HRV
us
indices explained by genetic effects in males (n=355) and females (n=228) were 99%
an
and 93%, respectively. Statistical significance was set at p
