Epigenomics of Metabolic AGEing in Severe Obesity (iMAGE)


Between 1950 and 2000, the life expectancy in Western countries increased by 11.4 and 13.6 years for men and women, respectively and during the 10-year period 1995-2005, it increased by 3 years for men and 2 years for women in the EU. It is therefore expected that the number of persons older than 65 years will represent a third of the general population in 2050, compared to a sixth in 2000. An important contributor to this increase is the decline in deaths from cardiovascular disease, which will probably be counterbalanced by the dramatic increase of obesity-linked conditions, including metabolic syndrome and two of its components, hypertension and type 2 diabetes (Yusuf and Anand, Lancet 2011, Anand and Yusuf, Lancet2011).

The metabolic syndrome is the term used to describe a cluster of risk factors for cardiovascular disease (CVD), which includes abdominal obesity, high triglycerides, low high-density lipoprotein (HDL), hypertension, hyperglycemia/insulin resistance. It is closely associated with type 2 diabete. Individuals are defined as having a metabolic syndrome if at least three of the five tested components are present. The outcomes of diseased obesity, including steatohepatitis (NASH), cardiomyopathy and chronic arthritis (CA) share similarities in the complexity of their physiopathology, with metabolic dysregulations, physical stress and intestinal dysbiosis (Fontana et al., 2007) triggering inflammation, cellular stress and remodelling in these target tissues, these responses being modulated by genetic and epigenomic predictors. Indeed, heritability estimates are between 30 and 70% for diabetes, obesity and hypertension. An abundance of GWAS over the past five years have sought to identify genetic factors associated with chronic diseases. Despite these successes, geneticists have reached the important conclusion that common genetic variability cannot account for the full genetic contribution to disease risk. Recent theoretical models suggest that a large number of common variants with small to modest effect on complex traits may at best explain 50% of the genetic component of complex traits (Stahl et al, Nat Genet 2012). The crucial role of the epigenome in this scenario is only beginning to be evaluated as powerful methods to analyze the whole DNA methylome emerge (Rakyan et al, Nat Rev Genet, 2011). DNA methylation is closely associated with the ageing process itself and very few EWAS have been performed for evaluating the association between the DNA methylome and age-related outcomes. Epigenome-wide scans have identified age-associated changes in DNA methylation at no more than one hundred CpG sites in blood and skin (Koch CM et al, Aging, 2011, Florath et al, Hum Mol Genet, 2014; Koch CM et al, Aging, 2011). A recent EWAS with BMI in 479 individuals of European origin showed an association of increased BMI with increased methylation at the HIF3A locus in blood cells and in adipose tissue (Dick et al, The Lancet, 2014). The one carbon metabolism (1-CM) plays a central role in the influence of metabolic and nutritional factors on DNA methylation and regulation of gene expression.

This complex metabolic network is regulated by a number of genes and requires micronutrients such as folate, vitamins B12, B6, B2 among others to function correctly. Some of them, folate, vitamin B12 and choline are methyl donors, which are involved in the synthesis of the precursor of S-adenosylmethionine, the universal donor of methyl groups needed for DNA methylation. Thus dysregulation in any of the regulatory components (nutritional, metabolic or genetic determinants) of 1-CM can alter the epigenomic regulation of gene expression. Over the past 2 decades, many epidemiological studies and meta-analyses have clearly demonstrated an association between markers of the 1-CM (methyl donors and homocysteine) and manifestations of outcomes of metabolic syndrome. A study conducted in the OASI population of elderly ambulatory volunteers showed that plasma total homocysteine correlated negatively with ApoA-I and HDL-Cholesterol. The associations of homocysteine with increased small size HDL3c suggested mechanisms related with abnormal maturation of HDL particles (Guéant-Rodriguez et al, Atheroclerosis, 2011). Very recently, a double-blind placebo-controlled clinical trial of 5mg folate supplementation was conducted in obese women. It concluded that the supplementation resulted in a reduced homeostasis model of assessment-insulin resistance score (HOMA-IR) (Asemi et al, Mol Nutr Food Res, 2014). Dietary and genetic determinants of 1-CM are associated with non-alcoholic fatty liver disease NAFLD severity, in operated and non-operated obeses (Gulsen et al., 2005; Hirsch et al., 2005; Koplay et al., 2011). Hepatic steatosis is also observed in patients with genetic disorders of 1-CM (Christensen and Rosenblatt, 1995; Russo et al., 1992). The teams of supervisors showed that a deficiency in folate and vitamin B12 yields micro-vesicular liver steatosis in rat pups from dams with methyl donor deficiency during pregnancy and lactation (Blaise et al., 2007). The steatosis results from a deficit in carnitine synthesis and epigenomic deregulations, with hypomethylation and decreased expression of PPAR-α, ERR-α, ER-α and liver specific nuclear receptor HNF-4α, and hypo-methylation of PGC1-α co-activator through decreased activity of PRMT1 (Pooya et al., 2012). As shown in heart of rat pups, the hypomethylation of PGC1-α in the liver participates to the epigenomic deregulation of fatty acid beta-oxidation by reducing its partnership with PPAR-α, ERR-α and HNF-4α (Garcia et al., 2011; Pooya et al., 2012).


Hypotheses addressed by the project

The above-mentioned data suggest that common genome variation, epigenome patterns, microbiota and 1-CM are part of a same scenario in outcomes of diseased obesity. Despite recent strong experimental-based evidence, there is no population-based study of this risk prediction in relation to the interplay between epigenome, genome and 1-CM. The need for an integrated approach is illustrated by the association of genetic polymorphisms of the FTO gene with obesity risk and metabolic syndrome and of the modulation of the risk by FTO methylation (Meyre D et al, Nat Genetics, 2009; Kilpelainen et al, PLOS Med 2011). Interestingly, a recent study has shown that the methylation level of the FTO gene in the adipose tissue of Humans was modified by six months exercise intervention, providing a mechanistic explanation for the gene x physical activity interaction on obesity (Ronn et al, PLOS Genet 2013). The FTO story illustrates the urgent need for integrative approaches to better decipher the complex interplay between environmental exposures, genome, epigenome, transcriptome and proteome to prevent unhealthy ageing related to diseased obesity.


Aims and methods

iMAGE aims to investigate the interacting influence of genome wide methylome and genetic variants, determinants of 1-CM on risk and components of diseased obesity and metabolic syndrome, including lipodystrophy with increased abdominal fat, insulin resistance, NASH and altered bone density, through integrated analyses of obese patients before bariatric surgery (Cohort EOS ALDEPI). Replication studies will be performed in 2 ambulatory cohorts of elderly subjects with high frequency of metabolic syndrome outcomes, namely the TUDA (Dublin) and OASI (Sicily) cohorts. The project will also study the dysregulation related to interactions between epigenome, genome, and CM-1 in adipose tissue and liver. It will therefore implement integrative bioinformatics analyzes, epigenomic data, genomic, metabolomic and clinics to identify predictive biomarkers of morbid obesity and personalized medicine and surgery.

To address this objective, we propose to develop the following tasks:

Task 1: Analyses of epigenome and genome variations (M1-M18).
The methylation profile of genomic DNA from each individual will be by processing bisulphite-converted DNA on an Illumina HumanMethylation450 bead arrays in the Genomic plateform of U 954. The arrays will be scanned on Illumina iScan and raw methylation data extracted using Illumina’s Genome Studio methylation module. Genotyping of genome wide variants will be performed with the HumanOmni2.5-8 BeadChip array from Illumina.

Task 2: Metabolomics and biomakers of components of metabolic syndrome (M1-M12).
All classical and innovative markers of 1-CM, all the biological markers of metabolic syndrome and insulin resistance, including homocysteine, MMA, choline, betaine, trimethylamine N-oxide, isoforms of folate, as well as noninvasive biomarkers of NASH, NAFLD Fibrosis Score and Enhanced Liver Fibrosis panel will be measured in the ‘laboratoire de Biochimie-Biologie Moléculaire-Nutrition-Métabolisme’ of CHU of Nancy.

Task 3: Transcriptome and methylome in adipose and liver tissues (M1-M12).
RNA will be reverse-transcribed, amplified and hybridized on Agilent GE Microarray 4x 44K (Agilent Protocol G4140-90050_GeneExpression_ Two_Color_v6.5) at the Integrative Genomic platform of U954, FR CNRS/Inserm 3209

Task 4: Biostats analyses of genome and epigenome data (M18-M30).
Association between methylation and relevant phenotypes will be carried out at the Integrative Genomic platform of U954, FR CNRS/Inserm 3209 by fitting a separate linear model for each CpG site, using MethLAB software for incorporating both continuous and categorical phenotypes and covariates, as well as fixed or random batch or chip effects and accounts for multiple testing by controlling the false discovery rate (FDR) (Kilaru V et al., 2012, Epigenetics).

Task 5: Bioinformatics analyses of genomic, epigenomic and proteomic data in tissues (M28-M32)
The task will perform integrated analyses associating population characteristics, GWAS, EWAS, 1-CM outcomes at baseline and follow-up in discovery study at baseline and replication studies, using blind procedure. The data integration will use cutting-edge statistical methodologies such as canonical or cluster analyses for the analysis of omics and clinical data (Beyene Genet Epidemiol 2007, Hamid Hum Genomics Proteomics 2009, Parkhomenko Stat Appl Genet Mol Biol 2009, Fallah Stat Appl Genet Mol Biol 2008, Beyene Genet EPidemiol 2014). We will use our considerable expertise in data integration and methodological guidelines (Li & Meyre Int J Obes 2013, Li & Meyre, CRP2014a, Li & Meyre CPRb, Li & Meyre CPRc) to apply novel statistical techniques to integrate the disparate types of data produced by the various assessment technologies.
PhD supervisors:

Dr David Meyre (DR Inserm, Prof Invité UL lauréat du programme AGIR),
Prs Rosa-Maria Guéant-Rodriguez (PU-PH Nutrition)
Laurent Brunaud (PU-PH chirurgie digestive) (INSERM U954 NGERE).
Partners: Département d’épidémiologie clinique & biostatistiques de l’Université McMaster (Hamilton, Canada), UMR_S954 INSERM-UL “Nutrition, Génomique, Exposition aux Risques Environnementaux”.

Doctoral school for PhD recruitment/follow-up: BioSE (Biology, Health, Environment – ED 266)

Dr David Meyre (meyred@mcmaster.ca),
Prs Rosa-Maria Guéant-Rodriguez (rm.rodriguez@chru-nancy.fr or rosa-maria.gueant-rodriguez@univ-lorraine.fr)
Laurent Brunaud (l.brunaud@chu-nancy.fr)


How to apply

In order to prepare a PhD thesis within the Lorraine Université d’Excellence Program, the interested candidate should consult the PhD topics offered in each social and economic challenges.
These PhD thesis topics are proposed by faculty members or researchers accredited to supervise research.

Candidate application period: according to graduate school schedule (visit each topic)
Each candidate may submit an application on up to three separate research topics.

Application analysis period by each graduate school
The graduate school reviews the applicants for a doctoral contract in the relevant disciplines. They check the level of supervision for each supervisor and the situation of trained doctors. Each candidate will meet the laboratory director, supervisor and a representative from the graduate school. This interview is to identify the candidate’s motivations and suitability as a candidate for the PhD project proposed by the supervisor. A recommendation will be made to the graduate school. This will summarize the strengths and/or weaknesses of the application.

PhD grants will include monthly income for the PhD student (roughly 1700 € for research only, complement can be provided for teaching missions) and environment for research in the research unit.

Please be aware that in order to offer a variety of subjects, more positions are posted here than available funding. The LUE executive committee will make the final choice on the granted funding (up to 12 positions), based on the recommendations by the doctoral schools.