SEARCH

SEARCH BY CITATION

Keywords:

  • Genomics;
  • exercise;
  • cardiovascular;
  • bioinformatics;
  • responsiveness

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

The rates of obesity and sedentary lifestyle are on a dramatic incline, with associated detrimental health effects among women in particular. Although exercise prescriptions are useful for overcoming these problems, success can be hampered by differential responsiveness among individuals in cardiovascular fitness indices (i.e. improvements in strength, lipids, VO2max). Genomic factors appear to play an important role in determining this inter-individual variation. We performed microarray analyses on mRNA in whole blood from 60 sedentary women from a multi-ethnic cohort who underwent 12 weeks of exercise, to identify gene subsets that were differentially expressed between individuals who experienced the greatest and least improvements in fitness. We identified 43 transcripts in 39 unique genes (FDR<10%; FC>1.5) whose expression increased the most in “high” versus “low” pre-menopausal female responders. These 39 genes were enriched in six biological pathways, including oxidative phosphorylation (p = 8.08 × 10−3). Several of the 39 genes (i.e. TIGD7, UQCRH, PSMA6, WDR12, TFB2M, USP15) have previously reported associations with fitness-related phenotypes. In summary, we identified gene signatures based on mRNA analysis that define responsiveness to exercise in a largely minority-based female cohort. Importantly, this study validates several genes/pathways previously associated with exercise responsiveness and extends these findings with additional novel genes.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

One of the greatest, preventable challenges to human health is physical inactivity. Although increased exercise is considered a proven prescription for reducing the risk of cardiovascular disease, stroke, type 2 diabetes and certain types of cancers (Writing Group Members et al., 2010), the rates of physical inactivity in the United States have been on a steady incline. This trend is further exacerbated by a growing reliance on technology, a rise in sedentary jobs (83% since 1950) and longer work days (average 47 h a week). The rates of physical inactivity among Hispanics and African-Americans are even greater, requiring targeted intervention strategies for reducing health disparities in these populations. The current guidelines issued by the American College of Sports Medicine (ACSM) for prescribing individualized exercise to healthy adults recommend moderate-intensity cardiovascular exercise coupled with resistance training exercises (Garber et al., 2011). Following these recommendations, we designed the Genetics, Exercise And Research (GEAR) study to test the efficacy of a 12-week exercise intervention in a multi-ethnic cohort of healthy but sedentary individuals that reflects the growing demographics of the United States population. A secondary goal was to investigate genomic causes of variability in fitness responsiveness that can influence the success of exercise interventions.

Recent studies have shown that there is a great deal of inter-individual heterogeneity in gains in fitness indices (i.e. improvements in cardiorespiratory fitness or metabolic markers) even among individuals performing exercises of equal duration and intensity (Bouchard & Rankinen, 2001; Mori et al., 2009; Vollard et al., 2006; Timmons et al., 2010). There is accumulating evidence which suggests that this variability in exercise responsiveness has a significant genetic component (Bouchard et al., 1999). Several studies have attempted to define a “molecular mRNA signature” or “genomic profile” of exercise responsiveness by focusing on individual fitness measures (Radom-Aizik et al., 2009; Christiansen et al., 2010). For example, a 29-gene transcript signature from skeletal muscle tissue gene expression has been correlated with gains in VO2max, the maximal oxygen uptake during exercise (Timmons et al., 2010). In addition, resistance training in post-menopausal women resulted in significant changes in mRNA levels of 11 genes related to local inflammatory signalling in skeletal muscle, a contributing factor to the development of sarcopenia (Buford et al., 2009). However, given the widespread physiological effects of exercise on multiple organs and tissues in the human body, it is worth considering whether gene expression profiling of peripheral blood cells can be used to classify individuals based on a more global definition of fitness. To that end, we developed a composite index of fitness that more comprehensively captures improvements in multiple fitness measures including adiposity, inflammation, musculoskeletal strength, blood pressure and cardiorespiratory fitness. We used this composite index to categorize participants in the GEAR exercise intervention study as “high” and “low” responders to exercise and to identify a gene expression signature that differentiates these two groups.

The GEAR study participants represent a multi-ethnic cohort of healthy but sedentary individuals, representative of the general United States population. Whereas most studies have focused on identifying fitness genes in the context of endurance exercise-based interventions (i.e. using treadmill or exercise bike), the GEAR study utilized an exercise protocol which followed minimal intensity guidelines and therefore may be more practical and safely incorporated into strategies for the general population. Gene expression profiling was done on whole blood obtained prior to initiation of the exercise regime and at the completion of the 12-week intervention. Our ultimate goal is to integrate these measures into a comprehensive wellness programme that can be adapted for large numbers of individuals in diverse settings; thus, we wanted to identify genomic signatures associated with exercise responsiveness from whole blood which can be easily obtained in a non-invasive manner. Also, the majority of studies examining genomic markers of exercise responsiveness have focused on exercise interventions in men. Therefore, we have chosen to examine the impact of the 12-week exercise intervention on gene expression levels in women, including those from multiple ethnic groups. We identified a signature of 39 genes that showed differential expression levels between low and high responding pre-menopausal women. Pathway analysis on this set of differentially regulated genes showed a statistically significant enrichment for genes involved in several pathways, including the oxidative phosphorylation (OS) pathway which has been previously implicated to play a role in exercise responsiveness (Selivanov et al., 2008; Gordon et al., 2012). Two genes, LRRFIP1 and SNORD30, were also identified with lower expression in “high” responding post-menopausal women. The establishment of gene expression signatures associated with high responsiveness to exercise intervention among women who are at risk of cardiovascular disease may allow for development of lifestyle interventions tailored to a person's unique molecular background.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Study Population and Enrolment Process

We enrolled a multi-ethnic cohort of sedentary women aged 18–65 years old (mean age = 43.0 ± 11.9 years) to participate in a 12-week exercise study (GEAR). The Institutional Review Board at the University of Miami (UM) approved the protocol. Participants were recruited from the UM and surrounding Miami-Dade County area. All were required to be sedentary, defined as not having exercised for more than 20 min for three or more days per week in the 6 months prior to enrolment. Individuals were excluded from participating if they met any of the following criteria: pregnant, diagnosed with cancer, type 1 diabetes, autoimmune disorders such as multiple sclerosis, lupus, Hashimoto's or Crohn's disease, neuromuscular disorders such as Amyotrophic lateral sclerosis, Parkinson's disease, Alzheimer's disease, muscular dystrophy, or history of heart-related conditions such as myocardial infarction, high-risk unstable angina, uncontrolled cardiac arrhythmias, active endocarditis, acute pulmonary embolus or infarction, acute myocarditis or pericarditis. Participants on cholesterol-lowering, anti-hypertensive or diabetes medications were required to have a physician's note prior to study enrolment. After telephone screening, eligible participants were provided a detailed study overview and informed consent, and were scheduled for a baseline fitness assessment. All assessments were performed in the fitness laboratory of the UM Medical Wellness Center in the early morning following a 10 h fast. Assessments were performed in the following order for all participants: (1) fasting blood draw; (2) consumption of snack; (3) administration of the demographic and self-report assessment questionnaires; (4) clinical assessments: resting heart rate and blood pressure, height/weight, waist/hip circumference; (5) submaximal treadmill test to estimate maximal oxygen uptake during peak exercise (VO2max); (6) upper/lower body strength test. Assessments were performed in the same manner at baseline and after completion of the 12-week exercise training protocol. The demographic characteristics of 60 female GEAR participants (38 pre-menopausal and 22 post-menopausal) whose gene expression profiles were analysed in this report are presented in Table 1.

Table 1. Characteristics of study participants
 Women (n = 60)
VariablePre-menopausal (n = 38)Post-menopausal (n = 22)
  1. Percentages are calculated based on total sample in each group.

Age (years)35.8 ± 8.4456.6 ± 5.12
White N (%)28 (73.7)20 (90.9)
Hispanic22 (57.9)10 (45.5)
Non-Hispanic6 (15.8)10 (45.5)
Black N (%)10 (26.3)2 (9.1)
Hispanic00
Non-Hispanic10 (26.3)2 (9.1)

Exercise Training Programme

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Participants followed a supervised 12-week exercise programme consisting of both cardiovascular and progressive resistance training. Twice per week, participants performed combined exercise sessions (both cardiovascular and resistance training) while on the third day they performed cardiovascular exercise only. The cardiovascular exercise protocol progressed systematically from 50% estimated relative VO2max (rVO2max) for 20 min during the first week to 75% estimated rVO2max for 45 min during the 12th week. During the combined sessions, subjects also completed a 10-exercise resistance training circuit. Initial loads were set at approximately 60% of one-rep max (1-RM). Subjects were instructed to perform one set of 8–12 repetitions on each exercise. Once 12 repetitions could be performed in a slow-controlled manner, the load was increased by 5% for the next workout. If muscular failure occurred prior to the 12th repetition, then the load was maintained for the next workout. This exercise training protocol is in accordance with recommendations set forth by the American College of Sports Medicine (ACSM, 2011).

Fitness Measures

Quantitative measures of multiple fitness traits were collected on all participants. For each trait, we calculated the percentage change from baseline to 12 weeks post-exercise (i.e. (trait12weeks − traitbaseline)/traitbaseline) × 100). We classified individuals as “high” responders if their percentage change fell within the upper or lower quartile of the distribution of the trait, depending on the trait. For some traits such as BMI and pulse pressure, a percentage decrease (i.e. the lower quartile) indicated high responsiveness, whereas for other traits including rVO2max, a percentage increase (i.e. the upper quartile) indicated high responsiveness. Thus, when the upper quartile was used to categorize “high” responsiveness, all individuals falling below the upper quartile were considered “low” responders. However, when the lower quartile was used to categorize “high” responsiveness, all individuals falling above the lower quartile were considered “low” responders. For each trait, high responsiveness was assigned a score of 1 and low responsiveness a score of 0. An overall fitness composite score index was created by summing the responsiveness status across 20 individual fitness traits (see Table 2). Since none of the traits had ≥5 missing observations and 91% of traits had < 3 missing observations, this approach is reasonable. The maximum fitness composite score (FCS) among all participants was 13. Women with FCS ≥8 (based on the upper quartile of the FCS distribution) were thus categorized as overall high responders to the 12-week exercise protocol. Individuals with FCS < 8 were categorized as overall low responders. Figure 1 shows the percentage of overall “high” and “low” responders based on the FCS index who were also categorized as high responders for each of the individual traits. For example for BMI, 63% of overall composite score index “high” responders also had the greatest decrease in BMI following 12 weeks of exercise compared to only 10% for the overall low responders. Equally dramatic differences can be seen for weight, waist circumference, waist−hip ratio (WHR), low density lipoprotein (LDL), high-density lipoprotein (HDL), very low lipoprotein (VLDL), triglycerides (TG), C-reactive protein (CRP), insulin, glucose and pulse pressure. There was less discrepancy between the percentage of “high” and “low” overall responders who were also categorized as high responders for cardiorespiratory and muscular strength traits in this study cohort.

Table 2. Baseline and change in fitness traits (anthropomorphic, blood biomarkers, blood pressure, cardiorespiratory and muscular strength variables) in women (N = 60) participating in GEAR
VariableBaseline12-weekAvg%change*Avg Diff**P***
Age (years)43.6 ± 12.4
  1. *(12-week follow-up − baseline)/baseline × 100); **(12week follow-up − baseline). ***Based on paired t-test of difference between 12-week follow-up and baseline measures.

Anthropomorphic
BMI (kg/m2)29.8 ± 6.029.6 ± 5.9−0.4−0.140.19
Weight (lb)174.4 ± 37.1173.3 ± 36.5−0.5−1.130.06
Waist circumference (cm)34.1 ± 4.633.3 ± 4.5−2−0.76<.0001
Hip circumference (cm)43.5 ± 4.642.8 ± 4.4−2−0.69<.0001
Waist-hip ratio0.78 ± 0.060.78 ± 0.06−0.6−0.0050.19
Blood biomarkers (e.g. lipids, insulin)*
Total cholesterol (mg/dL)181.9 ± 31.1179.3 ± 28.6−2−4.60.10
HDL-cholesterol (mg/dL)59.9 ± 18.259.0 ± 17.0−3−0.120.94
LDL-cholesterol (mg/dL)112.8 ± 28.3106.2 ± 28.0−5−7.70.01
VLDL-cholesterol (mg/dL)22.3 ± 12.819.8 ± 10.2−11−1.980.17
Trigylcerides (mg/dL)106.8 ± 55.299.1 ± 50.8−7−4.690.41
C-reactive protein (mg/dL)2.4 ± 1.92.6 ± 2.380.330.28
Glucose (mg/dL)77.7 ± 11.476.5 ± 8.5−2−1.30.35
Insulin (mg/dL)12.0 ± 7.010.7 ± 5.4−10−0.890.29
Blood pressure
Diastolic BP (mmHg)81.2 ± 8.476.9 ± 8.4−5−4.28<.0001
Systolic BP (mmHg)122.8 ± 10.4114.2 ± 12.2−7−8.65<.0001
Pulse pressure (mmHg)41.7 ± 7.537.3 ± 8.5−9−4.370.0002
Cardiorespiratory and muscular strength
Relative VO2 max (ml/kg/min)28.3 ± 5.329.6 ± 4.681.330.004
Resting heart rate (bpm)72.5 ± 9.768.1 ± 9.5−7−5.04<.0001
1-RM bench press54.1 ± 15.279.0 ± 19.45125.54<.0001
1-RM leg press150.1 ± 54.6251.9 ± 64.783101.8<.0001
image

Figure 1. Percentage of individuals classified as “high” and “low” responders from a composite fitness index who were also “high” responders for individual measured traits.

Download figure to PowerPoint

Blood Collection and Measurement of Lipid Biomarker

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Fasting blood was collected by certified phlebotomists, following protocols to ensure patient safety and minimize risk. Up to 30 cc of blood was collected from participants by venipuncture at each visit (baseline and after 12 weeks). Pre-menopausal women were surveyed at the time of blood collection to determine menstruation status, since menstrual cycle hormones are known to influence stress and inflammatory responses (Smyth 2004) and possibly gene expression. Serum was isolated for the analysis of cardiovascular biomarkers. Total cholesterol (TC), high density lipoprotein (HDL-C), low density lipoprotein (LDL-C), triglyceride (Tg), glucose, insulin and C-reactive protein (CRP) levels were measured at baseline and after 12 weeks using standard methods. All biomarker assays were processed in the Immunoassay and Chemistry Core Laboratory of the University of Miami Diabetes Research Institute.

Total RNA Preparation and Microarray Processing

RNA isolation was performed using a PAXgene Blood RNA Kit (Qiagen, Venio, Netherlands) according to manufacturer instructions. RNA quality was assessed by Agilent 2100 BioAnalyzer (Agilent, Santa Clara, CA, USA). Samples were amplified and labelled using Ovation RNA Amplification kit (NuGEN, West Cumbria, UK). After sense-strand cDNA synthesis with WT-Ovation Exon Module (NuGEN), samples were fragmented and labelled using Encore Biotin Module (NuGEN) and then hybridized with the Affymetrix GeneChip Human Gene 1.0 ST microarray for 17 h at 65°C according to the manufacturing recommendation (Affymetrix, Santa Clara, CA, USA). This array represents 28,869 unique genes (or transcripts) which are captured by ∼26 probes per gene, and was designed to avoid limitations of 3′ bias from other arrays. All arrays were washed and stained using the Affymetrix Fluidic stations 450 and scanned using the Affymetrix GeneChip scanner 3000 7G. βeta-globin reduction was performed using the Nugen Ribo SPIA kit. Image analysis was initially performed using the Affymetrix Command Console Software (AGCC). The raw expression data (log2 values) were quantile normalized using the GCRMA algorithm using Robust Multi-array Average (RMA) with GC-content background correction with the Partek software (St. Louis, MO). The raw data set has been submitted to the Gene Expression Omnibus (GEO) database (GSE34788). All RNA and microarray processing was performed at the University of Miami's Microarray Core Facility.

Statistical Analyses

Student's paired t-tests were used to assess change in participant fitness and clinical phenotypic characteristics with 12-week exercise. Statistical analyses of phenotypic data were performed using SAS version 9.2 (SAS Institute Inc., Cary, NC). To identify differentially expressed genes, normalized log2 datafiles exported from Partek were analysed with the Significance Analysis of Microarray (SAM v2.0) statistical package in R. SAM identifies statistically significant genes by carrying out gene-specific t-tests and repeated permutations to account for between gene correlations and the non-parametric distribution of individual genes. For this study, 1000 permutations were performed to estimate the False Discovery Rate (FDR) (Benjamini & Hochberg, 1995). We used ±1.5-fold change and FDR<0.10 as thresholds for determining significance. We tested for differential gene expression patterns from mRNA collected at baseline and after 12 weeks of exercise between “high” and “low” responders, as previously defined by the FCS index. First, we identified individual genes whose change in levels of gene expression differed significantly between “high” and “low” exercise responders. The goal of this analysis was to better understand which genes are affected (i.e. which showed significant change in mRNA expression levels) by the exercise regimen. We then determined whether baseline mRNA expression differed for any genes among individuals we later classified into “high” and “low” response groups. All analyses were conducted separately in subgroups of pre- (n = 38) and post-menopausal (n = 22) women. Gene expression analysis was also conducted using linear models in the LIMMA Smyth (2004) package in R with the composite score included as the outcome variable, and age, menopausal status and self-reported ethnicity included as covariates.

Gene Ontology, Pathway and Gene Set Enrichment analysis

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Small differences in mRNA expression can be missed at the gene level, particularly when using stringent multiple-testing cutpoints. Thus, we imported lists of differentially expressed genes from SAM (FDR<0.10) into MetaCore™ (GeneGo, Inc., Hollywood, FL, USA) for further gene set enrichment analysis. MetaCore™ integrates data from multiple functional databases and calculates statistical significance (FDR p-value) of over-represented gene transcripts in defined gene ontologies (GO) or pathways using a classic hypergeometric distribution. We also analysed the entire unfiltered expression dataset with Gene Set Enrichment Analysis (GSEA) (http://www.broadinstitute.org/gsea) (Subramanian et al., 2005) using the Molecular Signatures Database, MSigDB (http://www.broad.mit.edu/gsea/msigdb/index.jsp). GSEA identified significant sets of genes that were over-represented at the top or bottom of the ranked set of genes that were differentially expressed between “high” and “low” responders. This approach avoided the need for pre-filtering of genes by p-values or fold change prior to enrichment analysis. For our analyses, we used the MSigDB c2 curated gene sets, which collect pathway information from online databases including BioCarta, Gene arrays, BioScience Corp, KEGG, Reactome, Sigma-Aldrich pathways, Signal transduction knowledge environment, and signalling gateway and MSigDB c5 GO gene sets which are based on GO terms and their associations to human genes. Gene sets with FDR p-value < 0.25 were considered significant.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

The mean age of all participants was 46.2 (range 18–65 years) (Table 1). The study population was 53% Hispanic White, 27% non-Hispanic White and 20% non-Hispanic Black. Table 2 shows the baseline and 12-week changes in anthropometric, metabolic marker, blood pressure, cardiorespiratory and muscular strength traits in the 60 women. Overall, the women who exercised for 12 weeks through GEAR experienced 2% reduction in both waist (p < 0.0001) and hip circumference (p < 0.0001), 5% reduction in LDL-cholesterol (p = 0.01), 5% reduction in diastolic (p < 0.0001) and 7% reduction in systolic (p < 0.0001) blood pressure, 9% reduction in pulse pressure (p < 0.0001), 8% increase in relative VO2max (p = 0.004), 7% reduction in resting heart rate (p < 0.0001), 51% increase in upper body strength measured by 1-RM bench press (p < 0.0001) and 83% increase in lower body strength measured by 1-RM leg press (p < 0.0001).

Genes differentially expressed in “high” and “low” responders to exercise based on a fitness composite score index. A total of 6240 unique transcripts that were expressed and had at least ±1.5 fold change (FC) in ≥25% of the participants were included in our analyses. Using significance analysis of microarrays (SAM), we identified 43 transcripts in 39 unique genes (FDR<10%; FC>1.5) that had significantly increased change in expression levels in “high” versus “low” pre-menopausal responders, following 12 weeks of exercise (Table S1). Further, we performed a regression-based analysis of the 39 genes in the entire dataset using LIMMA, and adjusted for age, self-reported ethnicity and menopausal status, which resulted in 31 genes that remained significant (Table S1).

Bioinformatics pathway analysis of these 39 genes in MetaCore™ identified enrichment in the following pathways: wtCFTR and deltaF508 traffic (p = 8.74 × 10−4), signal transduction PKA signalling (p = 1.96 × 10−3), development beta-adrenergic receptors signalling via cAMP (p = 2.04 × 10−3), muscle contraction regulation of eNOS activity in cardiomyocytes (p = 2.36 × 10−3), regulation of CFTR activity (p = 2.53 × 10−3) and oxidative phosphorylation (p = 8.08 × 10−3) (data not shown). Among the individual genes identified, several [TIGD7, tigger transposable element derived 7 (Berisha et al., 2011); UQCRH, ubiquinol-cytochrome c reductase hinge protein (Lu et al., 2008); PSMA6, proteasome (prosome, macropain) subunit, alpha type, 6 (Alsmadi et al., 2009; Bennett et al., 2008, Freilinger et al., 2009, Liu et al., 2009); WDR12, WD repeat domain 12 (Myocardial Infarction Genetics Consortium, 2009); TFB2M, nuclear encoded mitochondrial transcription factor B2] have previously established relationships with cardiovascular disease, lipid metabolism or related fitness traits (Norrbom et al., 2010) (Table S1). Two separate genes (CTTN, cortactin, PRKAR2B, protein kinase, cAMP-dependent, regulatory, type II, beta) were also significantly over-expressed in baseline mRNA levels among high responding pre-menopausal women, although change in expression levels of these genes was not significantly different between response groups (Table S1). Among post-menopausal women, LRRFIP1, leucine-rich repeat flightless-interacting protein 1 and SNORD30, small nuclear RNA 30, had significantly lower expression in high versus low responders (Table S1). No genes were identified with higher levels of expression in high post-menopausal responders. Also no single gene met ±1.5 fold change and FDR significance thresholds for baseline differences in gene expression among post-menopausal women.

Gene Set Enrichment Analysis

To determine whether any canonical pathways or biological processes were enriched among “high” versus “low” composite score index responders, we also carried out gene set enrichment analysis (GSEA) using the entire unfiltered set of 6424 transcripts. Seven GO processes were enriched among “high” pre-menopausal female responders at baseline (Table 3), including the lipid biosynthetic process. Three separate GO processes involved in regulation of RNA metabolic processes, DNA and RNA transcription were also enriched after 12-week exercise in pre-menopausal “high” responders.

Table 3. Gene set enrichment analysis for canonical pathways and gene ontology (GO) biological processes
Enriched pathways and GO processesGene SetSizeNESNOM p-valFDR q-valEnrichment group
Pre-menopausal women – Baseline gene expression      
 MSigDB     
  1. Size = gene set size; NES = normalized enrichment score; HR = enrichment was found among “high” responders.

Lipid biosynthetic processC5:GO381.650.0130.1795HR
 MSigDB     
Pyrophosphatase activityC5:GO801.730.0150.1919HR
 MSigDB     
GTPase activityC5:GO381.660.0180.2024HR
 MSigDB     
Nucleoside triphosphatase activityC5:GO751.750.0080.2238HR
 MSigDB     
Transferase activity transferrin ACYL groupsC5:GO251.620.0140.2276HR
 MSigDB     
Cytoskeleton organization and biogenesisC5:GO711.660.0130.2312HR
 MSigDB     
Establishment and/or maintenance of chromatin architectureC5:GO311.680.0420.2493HR
Pre-menopausal women – 12-week gene expression
 MSigDB     
Negative regulation of RNA metabolic processC5:GO421.770.0020.1259HR
 MSigDB     
Negative regulation of DNA-dependent transcriptionC5:GO421.770.0020.1889HR
 MSigDB     
Negative regulation of transcription from RNA polymerase II promoterC5:GO321.8100.2458HR

In silico miRNA Target Prediction

Over the past decade, there has been growing appreciation for the role that small regulatory RNAs, namely microRNAs (miRNAs), are playing in the maintenance of cellular homeostasis, disease progression and responsiveness to environmental stimuli including diet, smoking and response to physical activity (Dykxhoorn, 2010; Hou et al., 2011; Izzotti et al., 2011; Timmons, 2011). miRNAs exert their effect by recognizing specific target mRNA sequences and inducing translational silencing and/or mRNA degradation (Bartel, 2009). Therefore, we investigated in silico whether any genes identified from the mRNA expression analyses had predicted miRNA target binding sites with reported roles in cardiovascular disease or related traits using TargetScanHuman5.2 (http://www.targetScan.org). In total, 31 miRNAs were identified which were predicted to bind to at least one of the differentially expressed genes and, of those, four [ANKRD22, hsa-miR-637 (Wei et al., 2011); LRRFIP1, hsa-miR-132 (Strum et al., 2009; Anand et al., 2010); PRKAR2B, hsa-miR-92a (Taurino et al., 2010); RSAD2, hsa-miR-192 (Isumi et al., 2011)] have been previously reported to be associated with CVD (Table 4).

Table 4. In silico predicted miRNA targets for genes that were differentially expressed in pre-menopausal female responders and non-responders
GenemiRNA*CVD phenotype
  1. *miRNA predicted to bind to gene.

ANKRD22hsa-miR-637Hypertension (29)
LRRFIP1hsa-miR-132Angiogenesis (30), obesity (31)
PRKAR2Bhsa-miR-92aCoronary artery disease (32)
RSAD2hsa-miR-192Diabetes (33)

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

The goal of the current study was to examine the gene expression profiles of previously sedentary, overweight, but otherwise healthy women from a variety of ethnic backgrounds who participated in 12 weeks of exercise training in the GEAR study. Our primary analyses were performed separately in pre- and post-menopausal women since it is known that sex-specific hormones can influence gene expression and may bias results (Hornum et al., 1997). Among pre-menopausal women, we identified 43 transcripts in 39 unique genes (FDR<10%; FC>1.5) that were over-expressed in “high” versus “low” responders of exercise (i.e. 12 weeks after exercising) based on a composite fitness score index we developed. This score enabled us to categorize women as “high” responders if they showed the most improvements in at least eight individual fitness traits, suggesting an overall fitness gain. While individual traits indicative of improved fitness are increasingly being investigated for biomarker discovery, ours is among the first to use a composite score that summarizes fitness improvements across multiple traits, in order to determine a more global profile of exercise responsiveness. With this composite score, we reasoned that the inherent measurement errors of individual traits would be less likely to bias the grouping of participants into the high responder category.

Of the 39 genes identified, several had reported associations with cardiovascular or related fitness traits TIGD7 was among seven genes with expression levels that were highly correlated with change in weight, mean glycosylated haemoglobin (HbA1c) and fasting plasma glucose following bariatric surgery in obese individuals with type-2 diabetes (Berisha et al., 2011). UQCRH was found to be associated with global islet protein and gene expression changes in diabetic versus non-diabetic control mice (Lu et al., 2008). PSMA6 polymorphisms were associated with myocardial infarction (MI) and coronary artery disease (CAD), while a SNP in WDR12 was one of nine to achieve genome-wide significance in a large study (n = 6024) of early-onset MI (Myocardial Infarction Genetics Consortium 2009). mRNA levels of TFB2M were higher in skeletal muscle among elite athletes compared to moderately active individuals following a 10-day endurance training programme (Norrbom et al., 2010). Transgenic mice with cardiac-specific over-expression of human USP15 had increased heart−body weight ratios and gene expression in the heart, indicating that USP15 is involved in the regulation of hypertrophic responses in cardiac muscle (Mootha et al., 2003).

Overall, the 39 genes were enriched in six pathways, including the oxidative phosphorylation pathway. This pathway was also over-enriched for genes that differentiated post-menopausal women with the highest decreases in resting heart rate (data not shown). This is an encouraging finding since exercise is known to increase rates of oxidative phosphorylation, a crucial component of cellular metabolism that results in the production of adenosine triphosphate (ATP). Reduced oxidative metabolism and down-regulation of genes involved in oxidative phosphorylation have also been observed in muscle and adipose tissue of insulin-resistant (Toledo et al., 2006; Mustelin et al., 2008) and obese individuals (Timmons et al., 2010), respectively. In addition, Timmons (2011) reported enrichment of this pathway when examining molecular predictors of  VO2max responsiveness to aerobic endurance training in RNA from skeletal muscle tissue. DNAJB1, one of the genes they identified as over-expressed among high male responders following 6 weeks of aerobic training, was also significantly over-expressed in mRNA collected at baseline among high responders to the GEAR exercise protocol. DNAJB1 is an androgen protein co-chaperone in the glucocorticoid receptor signalling (Jones, 2011) that is under transcriptional regulation by insulin (Goodall et al., 2010). Two genes (CTTN, PRKAR2B) were also over-expressed (FDR < 0.10) in baseline mRNA levels among the women who were subsequently categorized as “high” versus “low” responders. Protein kinase, PRKAR2B, is involved in lipolysis and was previously shown to be down-regulated in subcutaneous adipose tissue in obese compared to lean individuals (Marrades et al., 2010). Among post-menopausal women, one of the two genes identified with lower expression in high compared to low responders was LRRFIP1, which has been implicated in regulating platelet responsiveness (Goodall et al., 2010).

To gain further insight into the genes we found in pre-menopausal women, bioinformatic analysis was used to identify predicted miRNA target binding sites (Drummond et al., 2008; Friedman et al., 2009; Malumbres & Lossos, 2010; Nielsen, 2010; Radom-Aizik et al., 2010 Davidsen et al., 2011). We identified four miRNAs that have been reported previously in the literature in relation to CVD. MicroRNA-637 (Bartel et al., 2009) in ANKRD22 was shown to alter processing of chromogranin A, a precursor of catecholamine release-inhibitor catestatin that is considered a sub-phenotype of hypertension. MicroRNA-132 in LRRFIP1 has been shown to act as an angiogenic switch by targeting p120RasGAP in the endothelium (Strum et al., 2009). Over-expression of miRNA-132 also induced production of IL-8 and MCP-1 in primary human pre-adipocytes and in vitro differentiated adipocytes (Wei et al., 2011). MicroRNA-92a (Anand et al., 2010) in PRKAR2B had significantly reduced expression among patients with CAD who completed a cardiac rehabilitation programme following surgical coronary revascularization. MicroRNA-192 (Taurino et al., 2010) has been linked to type-2 diabetes although its exact role and relationship with TGF-β1 signalling still remains to be fully understood.

A potential limitation of this study is that we chose to create a composite score index instead of using factor analysis, a statistical technique that identifies a latent underlying variable characterizing the common features of multiple variables. This approach is useful when the measured variables are correlated and there is missing data. In our analyses, only a handful of variables (waist hip ratio−waist, r = 0.87; blood pressure−pulse pressure, r = 0.89; body mass index−weight, r = 0.95; triglycerides−very low density lipoprotein, r = 0.99) were highly correlated and there was also very little missing data, thus, we considered the composite score calculation was appropriate. Also while we corrected for multiple testing at the level of the number of gene transcripts by using permutations and applying FDR significance thresholds, we opted against additional corrections to account for the three sets of analyses we conducted (i.e. baseline, 12-week and change in gene expression. The small sample size of the post-menopausal group was also a limitation of this study. For example, while we had ∼80% power to detect ±1.5-fold differences in pre-menopausal female response groups (13 high and 25 low responders), the post-menopausal female dataset (3 high and 19 low) was notably underpowered. Thus, the lack of individual genes achieving stringent FDR corrections among post-menopausal women is not surprising. Also owing to small sample sizes, analysis based on responder groups defined by extremes of the distribution (i.e. upper vs. lower quartiles), along with ethnic-specific group comparisons, was not possible. Our goal is to evaluate whether such heterogeneity exists at the level of mRNA expression in the context of exercise responsiveness in larger samples of Hispanic, African American and other ethnic groups we are currently enrolling in GEAR. We also recognize that certain medications can directly impact exercise response; however, in our formal comparison of cholesterol-lowering, anti-hypertensive and diabetes medications, we did not find any statistically significant differences between female high and low responders (data not shown). Lastly, we did not account for diet in our analysis; however, we are actively collecting nutritional data using a Food Frequency Questionnaire and 24-h diet recall which will allow for exploration of the impact of diet in future analyses.

The results of our study, while preliminary, suggest that genes in the oxidative phosphorylation, immunity and inflammation pathways are significantly perturbed among women who we categorized as high responders using the fitness composite score index. Our in silico analyses provide some support for continued investigation of miRNAs and their role in regulating gene expression in the context of environmental stimuli including exercise and nutrition (Davidsen et al. 2011). In summary, our analyses provide support for the growing hypothesis that exercise causes perturbations to the genomic landscape and that these gene changes can be used to differentiate cardiovascular responsiveness to exercise. Further work, particularly in multiple ethnic groups and under various exercise regimens, will be crucial for elucidating important fitness genes that can be translated to clinical practice.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

We thank Gail Haldeman and the staff of the UM Medical Wellness Center for assistance with the recruitment of GEAR study participants. We acknowledge the dedication of current and previous GEAR research team members, Maria A. Marin, Amanda L. Mageean, Kimberly Perez, Kristy Whyte, Ryan Dauer, Orentes L. Talavera and Jonathan Weislow. We thank Toumy Guettouche for his technical assistance. We are grateful for the participation and enthusiasm of our GEAR participants who continue to inspire our research.

Sources of Funding

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Research conducted in this paper was funded in part by a grant from the Forum on Women's Health and funds from Florida's Office of Tourism, Trade and Economic Development.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information
  • Alsmadi, O., Muiya, P., Khalak, H., Al-Saud, H., Meyer, B. F., Al-Mohanna, F., Alshahid, M., & Dzimiri, N. (2009) Haplotypes encompassing the KIAA0391 and PSMA6 gene cluster confer a genetic link for myocardial infarction and coronary artery disease. Ann Hum Genet 73, 475483.
  • Anand, S., Majeti, B. K., Acevedo, L. M., Murphy, E. A., Mukthavaram, R., Scheppke, L., Huang, M., Shields, D. J., Lindquist, J. N., Lapinski, P. E., King, P. D., Weis, S. M. & Cheresh, D. A. (2010) MicroRNA-132-mediated loss of p120RasGAP activates the endothelium to facilitate pathological angiogenesis. Nat Med 16, 909914.
  • Bartel, DP. (2009) MicroRNAs: target recognition and regulatory functions. Cell 136, 215233.
  • Benjamini, Y., & Hochberg,Y. (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B (Method) 57, 289300.
  • Bennett, DA., Xu, P., Clarke, R., Zondervan, K., Parish, S., Palmer, A., Cardon, L., Peto, R., Lathrop, M., & Collins, R. (2008) International Study of Infarct Survival Collaborators. The exon 1–8C/G SNP in the PSMA6 gene contributes only a small amount to the burden of myocardial infarction in 6946 cases and 2720 controls from a United Kingdom population. Eur J Hum Genet 16, 480486.
  • Berisha, S. Z., Serre, D., Schauer P., Kashyap, S. R., & Smith, J. D. (2011) Changes in whole blood gene expression in obese subjects with type 2 diabetes following bariatric surgery: a pilot study. PLoS One 6, e16729.
  • Bouchard, C., An, P., Rice, T., Skinner, J. S., Wilmore, J. H., Gagnon, J., Perusse, L., Leon, A. S., & Rao D. C. (1999) Familial aggregation of VO(2max) response to exercise training: results from the HERITAGE Family Study. J Appl Physiol 87, 10031008.
  • Bouchard, C., & Rankinen T. (2001) Individual differences in response to regular physical activity. Med Sci Sports Exerc 33, S446S451.
  • Buford, T. W., Cooke, M. B., & Willoughby, D. S. (2009) Resistance exercise-induced changes of inflammatory gene expression within human skeletal muscle. Eur J Appl Physiol 107, 463471.
  • Christiansen, T., Paulsen, S. K., Bruun, J. M., Pedersen, S. B., & Richelsen, B. (2010) Exercise-training versus diet-induced weight-loss on metabolic risk factors and inflammatory markers in obese subjects. A 12-week randomized intervention-study. Am J Physiol Endocrinol Metab 298, E824E831
  • Davidsen, P. K., Gallagher, I. J., Hartman, J. W., Tarnopolsky, M. A., Dela, F., Helge, J. W., Timmons, J. A., & Phillips, S. M. (2011) High responders to resistance exercise training demonstrate differential regulation of skeletal muscle microRNA expression. J Appl Physiol 110, 309317.
  • Drummond, M. J., McCarthy, J. J., Fry, C. S., Esser, K. A., & Rasmussen, B. B. (2008) Aging differentially affects human skeletal muscle microRNA expression at rest and after an anabolic stimulus of resistance exercise and essential amino acids. Am J Physiol Endocrinol Metab 295, E1333E1340.
  • Dykxhoorn, D. M. (2010) MicroRNAs and metastasis: little RNAs go a long way. Cancer Res 70, 64016406.
  • Freilinger, T., Bevan, S., Ripke, S., Gschwendtner, A., Lichtner, P., Muller-Myhsok, B., Wichmann, H. E., Markus, H. S., Meitinger, T., & Dichgans, M. (2009) Genetic variation in the lymphotoxin-alpha pathway and the risk of ischemic stroke in European populations. Stroke 40, 970972.
  • Friedman, R. C., Farh, K. K., Burge, C. B., & Bartel, D. P. (2009) Most mammalian mRNAs are conserved targets of microRNAs. Genome Res 19, 92105.
  • Garber, C. E., Blissmer, B., Deschenes, M. R., Franklin, B. A., Lamonte, M. J., Lee, I. M., Nieman, D. C., & Swain, D. P. (2011) American College of Sports Medicine position stand. Quantity and quality of exercise for developing and maintaining cardiorespiratory, musculoskeletal, and neuromotor fitness in apparently healthy adults: guidance for prescribing exercise. Med Sci Sports Exerc 43, 13341359.
  • Goodall, A. H., Burns, P., Salles, I., Macaulay, I. C., Jones, C. I., Ardissino, D., de Bono, B., Bray, SL., Deckmyn, H., Dudbridge, F., Fitzgerald, D. J., Garner, S. F., Gusnanto, A., Koch, K., Langford, C., O'Connor, M. N., Rice, C. M., Stemple, D., Stephens, J., Trip, M. D., Zwaginga, J. J., Samani, N. J., Watkins, N. A., Maguire, P. B., Ouwehand, W. H., & Bloodomics Consortium. (2010) Transcription profiling in human platelets reveals LRRFIP1 as a novel protein regulating platelet function. Blood 116, 46464656.
  • Gordon, P. M., Liu, D., Sartor, M. A., Iglayreger, H. B., Pistilli, E. E., Gutmann, L., Nader, G. A., & Hoffman, E. P. (2012) Resistance exercise training influences skeletal muscle immune activation: a microarray analysis. J Appl Physiol 112, 443453.
  • Hornum, M., Cooper, D. M., Brasel, J. A., Bueno, A., & Sietsema, KE. (1997) Exercise-induced changes in circulating growth factors with cyclic variation in plasma estradiol in women. J Appl Physiol 82, 19461951
  • Hou, L., Wang, D., & Baccarelli, A. (2011) Environmental chemicals and microRNAs. Mutat Res 714, 10512.
  • Isumi, Y., Hirata, T., Saitoh, H., Miyakawa, T., Murakami, K., Kudoh, G., Doi, H., Ishibashi, K. & Nakajima, H. (2011) Transgenic overexpression of USP15 in the heart induces cardiac remodeling in mice. Biochem Biophys Res Commun 405, 216221.
  • Izzotti, A., Larghero, P., Longobardi, M., Cartiglia, C., Camoirano, A., Steele, V. E., & De Flora S. (2011) Dose-responsiveness and persistence of microRNA expression alterations induced by cigarette smoke in mouse lung. Mutat Res 717, 916.
  • Jones, M. R., Chua, A., Chen, Y. D., Li, X., Krauss, R. M., Rotter, J. I., Legro, R. S., Azziz, R., & Goodarzi, M. O. (2011) Harnessing expression data to identify novel candidate genes in polycystic ovary syndrome. PLoS One 6, e20120.
  • Liu, X., Wang, X., Shen, Y., Wu, L., Ruan, X., Lindpaintner, K., Yusuf, S., Engert, J. C., Anand, S., Tan, X., & Liu, L. (2009) The functional variant rs1048990 in PSMA6 is associated with susceptibility to myocardial infarction in a Chinese population. Atherosclerosis 206, 199203.
  • Lu, H., Yang, Y., Allister, E. M., Wijesekara, N., & Wheeler, M. B. (2008) The identification of potential factors associated with the development of type 2 diabetes: a quantitative proteomics approach. Mol Cell Proteomics 7, 14341451.
  • Malumbres, R., & Lossos, I. S. (2010) Expression of miRNAs in lymphocytes: a review. Methods Mol Biol 667, 129143.
  • Marrades, M. P., Gonzalez-Muniesa, P., Martinez, J. A., & Moreno-Aliaga, M. J. (2010) A dysregulation in CES1, APOE and other lipid metabolism-related genes is associated to cardiovascular risk factors linked to obesity. Obes Facts 3, 312318.
  • Mootha, V. K., Lindgren, C. M., Eriksson, K. F., Subramanian, A., Sihag, S., Lehar, J., Puigserver, P., Carlsson, E., Ridderstrale, M., Laurila, E., Houstis, N., Daly, M. J., Patterson, N., Mesirov, J. P., Golub, T. R., Tamayo, P., Spiegelman, B., Lander, E. S., Hirschhorn, J. N., Altshuler, D., & Groop, L. C. (2003) PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 34, 267273.
  • Mori, M., Higuchi, K., Sakurai, A., Tabara, Y., Miki, T., & Nose, H. (2009) Genetic basis of inter-individual variability in the effects of exercise on the alleviation of lifestyle-related diseases. J Physiol 587, 55775584.
  • Mustelin, L., Pietilainen, K. H., Rissanen, A., Sovijarvi, A. R., Piirila, P., Naukkarinen, J., Peltonen, L., Kaprio, J., & Yki-Jarvinen, H. (2008) Acquired obesity and poor physical fitness impair expression of genes of mitochondrial oxidative phosphorylation in monozygotic twins discordant for obesity. Am J Physiol Endocrinol Metab 295, E148E154.
  • Myocardial Infarction Genetics Consortium, Kathiresan, S., Voight, B. F., Purcell, S., Musunuru, K., Ardissino, D., Mannucci, P. M., Anand, S., Engert, J. C., Samani, N. J., Schunkert, H., Erdmann, J., Reilly, M. P., Rader, D. J., Morgan, T., Spertus, J. A., Stoll, M., Girelli, D., McKeown, P. P., Patterson, C. C., Siscovick, D. S., O'Donnell, C. J., Elosua, R., Peltonen, L., Salomaa, V., Schwartz, S. M., Melander, O., Altshuler, D., Ardissino, D., Merlini, P. A., Berzuini, C., Bernardinelli, L., Peyvandi, F., Tubaro, M., Celli, P., Ferrario, M., Fetiveau, R., Marziliano, N., Casari, G., Galli, M., Ribichini, F., Rossi, M., Bernardi, F., Zonzin, P., Piazza, A., Mannucci, P. M., Schwartz, S. M., Siscovick, D. S., Yee, J., Friedlander, Y., Elosua, R., Marrugat, J., Lucas, G., Subirana, I., Sala, J., Ramos, R., Kathiresan, S., Meigs, J. B., Williams, G., Nathan, D. M., MacRae, C. A., O'Donnell, C. J., Salomaa, V., Havulinna, A. S., Peltonen, L., Melander, O., Berglund, G., Voight, B. F., Kathiresan, S., Hirschhorn, J. N., Asselta, R., Duga, S., Spreafico, M., Musunuru, K., Daly, M. J., Purcell, S., Voight, B. F., Purcell, S., Nemesh, J., Korn, J. M., McCarroll, S. A., Schwartz, S. M., Yee J., Kathiresan, S., Lucas, G., Subirana, I., Elosua, R., Surti, A., Guiducci, C., Gianniny, L., Mirel, D., Parkin, M., Burtt, N., Gabriel, S. B., Samani, N. J., Thompson, J. R., Braund, P. S., Wright, B. J., Balmforth, A. J., Ball, S. G., Hall A. S., Wellcome Trust Case Control Consortium, Schunkert, H., Erdmann, J., Linsel-Nitschke, P., Lieb, W., Ziegler, A., Konig, I., Hengstenberg, C., Fischer, M., Stark, K., Grosshennig, A., Preuss, M., Wichmann, H. E., Schreiber, S., Schunkert, H., Samani, N. J., Erdmann, J., Ouwehand, W., Hengstenberg, C., Deloukas, P., Scholz, M., Cambien, F., Reilly, M. P., Li, M., Chen, Z., Wilensky, R., Matthai, W., Qasim, A., Hakonarson, H. H., Devaney, J., Burnett, M. S., Pichard, A. D., Kent, K. M., Satler, L., Lindsay, J. M., Waksman, R., Knouff, C. W., Waterworth, D. M., Walker, M. C., Mooser, V., Epstein, S. E., Rader, D. J., Scheffold, T., Berger, K., Stoll, M., Huge, A., Girelli, D., Martinelli, N., Olivieri, O., Corrocher, R., Morgan, T., Spertus, J. A., McKeown, P., Patterson, C. C., Schunkert, H., Erdmann E., Linsel-Nitschke, P., Lieb, W., Ziegler, A., Konig, I. R., Hengstenberg, C., Fischer, M., Stark, K., Grosshennig, A., Preuss, M., Wichmann, H. E., Schreiber, S., Holm, H., Thorleifsson, G., Thorsteinsdottir, U., Stefansson, K., Engert, J. C., Do, R., Xie, C., Anand, S., Kathiresan, S., Ardissino, D., Mannucci, P. M., Siscovick, D., O'Donnell, C. J., Samani, N. J., Melander, O., Elosua, R., Peltonen, L., Salomaa, V., Schwartz, S. M., & Altshuler, D. (2009) Genome-wide association of early-onset myocardial infarction with single nucleotide polymorphisms and copy number variants. Nat Genet 41, 334341.
  • Nielsen, S., Scheele, C., Yfanti, C., Akerstrom, T., Nielsen, A. R., Pedersen, B. K., & Laye, M. J. (2010) Muscle specific microRNAs are regulated by endurance exercise in human skeletal muscle. J Physiol 588, 40294037.
  • Norrbom, J., Wallman, S. E, Gustafsson, T., Rundqvist, H., Jansson, E., & Sundberg, C. J. (2010) Training response of mitochondrial transcription factors in human skeletal muscle. Acta Physiol (Oxf) 198, 7179.
  • Radom-Aizik, S., Zaldivar, F.,Jr, Leu, S. Y., & Cooper, D. M. (2009) Brief bout of exercise alters gene expression in peripheral blood mononuclear cells of early- and late-pubertal males. Pediatr Res 65, 447452.
  • Radom-Aizik, S., Zaldivar, F.,Jr, Oliver, S., Galassetti, P., & Cooper D. M. (2010) Evidence for microRNA involvement in exercise-associated neutrophil gene expression changes. J Appl Physiol 109, 252261.
  • Selivanov, V. A., de Atauri, P., Centelles, J. J., Cadefau, J., Parra, J., Cusso, R., Carreras, J., & Cascante, M. (2008) The changes in the energy metabolism of human muscle induced by training. J Theor Biol 252, 402410.
  • Smyth, G. K. (2004) Linear models and empirical Bayes methods for assessing differential expression in microarray experiments. Stat Appl Gen Mol Biol 3, PMID: 16646809.
  • Strum, J. C., Johnson, J. H., Ward, J., Xie, H., Feild, J., Hester, A., Alford, A., & Waters, K. M. (2009) MicroRNA 132 regulates nutritional stress-induced chemokine production through repression of SirT1. Mol Endocrinol 23, 18761884.
  • Subramanian, A., Tamayo, P., Mootha, V. K., Mukherjee, S., Ebert, B. L., Gillette, M. A., Paulovich, A., Pomeroy, S. L., Golub, T. R., Lander, E. S., & Mesirov, J. P. (2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci USA 102, 1554515550.
  • Taurino, C., Miller, W. H., McBride, M. W., McClure, J. D., Khanin, R., Moreno, M. U., Dymott, J. A., Delles, C., & Dominiczak, A. F. (2010) Gene expression profiling in whole blood of patients with coronary artery disease. Clin Sci (Lond) 119, 335343.
  • Timmons, J. A., Knudsen, S., Rankinen, T., Koch, LG., Sarzynski, M., Jensen, T., Keller, P., Scheele, C., Vollaard, N. B., Nielsen, S., Akerstrom, T., MacDougald, O. A., Jansson, E., Greenhaff, P. L., Tarnopolsky, M. A., van Loon, L. J., Pedersen, B. K., Sundberg, C. J., Wahlestedt, C., Britton, S. L., & Bouchard, C. (2010) Using molecular classification to predict gains in maximal aerobic capacity following endurance exercise training in humans. J Appl Physiol 108, 14871496.
  • Timmons, J. A. (2011) Modulation of microRNAs during exercise and disease in human skeletal muscle. Exerc Sport Sci Rev 39, 218; author reply 219.
  • Toledo, F. G., Watkins, S., & Kelley, D. E. (2006) Changes induced by physical activity and weight loss in the morphology of intermyofibrillar mitochondria in obese men and women. J Clin Endocrinol Metab 91, 32243227.
  • Wei, Z., Biswas, N., Wang, L., Courel, M., Zhang, K., Soler-Jover, A., Taupenot, L., & O'Connor, D. T. (2011) A common genetic variant in the 3'-UTR of vacuolar H+-ATPase ATP6V0A1 creates a micro-RNA motif to alter chromogranin A processing and hypertension risk. Circ Cardiovasc Genet 4, 381389.
  • Writing Group Members, Lloyd-Jones, D., Adams, R. J., Brown, T. M., Carnethon, M., Dai, S., De Simone, G, Ferguson, T. B., Ford, E., Furie, K., Gillespie, C., Go, A., Greenlund, K., Haase, N., Hailpern, S., Ho P. M., Howard, V., Kissela, B., Kittner, S., Lackland, D., Lisabeth, L., Marelli, A., McDermott, M. M., Meigs, J., Mozaffarian, D., Mussolino, M., Nichol, G., Roger, V. L., Rosamond, W., Sacco, R., Sorlie, P., Roger, V. L., Thom, T., Wasserthiel-Smoller, S., Wong, N. D., & Wylie-Rosett, J. (2010) American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics—2010 update: a report from the American Heart Association. Circulation 121, e46e215.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Exercise Training Programme
  6. Blood Collection and Measurement of Lipid Biomarker
  7. Gene Ontology, Pathway and Gene Set Enrichment analysis
  8. Results
  9. Discussion
  10. Acknowledgements
  11. Sources of Funding
  12. References
  13. Supporting Information

Additional supporting information may be found in the online version of this article:

FilenameFormatSizeDescription
ahg12006-sup-0001-TableS1.doc79KTable S1 Genes with differential expression in “high” and “low” responders from 12 weeks of combined aerobic/resistance training.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.