Working Package 5: Omics and epigenetics analyses

Home / Agenda / Working Package 5: Omics and epigenetics analyses
Working Package 5: Omics and epigenetics analyses

LEADER: TNO
PARTNERS: AUTH, UPD, ISS, TNO, FERA, CERETOX, CNR, UKR, SXS
CONTACT: Rob Stierum
START MONTH: 1
END MONTH: 42

Objectives:

  1. To develop a framework and best practices  for application of omics to cohort studies, in tandem  with the activities of WP4 (Human Biomonitoring).
  2. To adapt and optimize -omics protocols  for application on biological matrices  for existing and new cohorts identified in Stream 5.

Description of work and role of partners:

To a different extent, -omics technologies lend themselves to retrospective, cross-sectional  and prospective studies. This WP aims at addressing  their functional integration in the exposome  context and derive general guidance  from analyses of both existing and the prospective  twin population study envisaged  in Stream 5. The main goal of WP5 is to identify, integrate  and optimize the protocols of –omics technologies so that they can effectively support the environment-wide and health association  studies that are the subject of HEALS. The optimized protocols of the battery of –omics techniques exploited in WP5 will be used to re-analyse a small selection of biological samples of the populations in WP14-WP16 and to analyse analysis maternal or child samples selected from sub-groups of the population involved in the EXHES population survey in WP17.

The following tasks are foreseen in WP5:

 


Task 5.1 Selection and analysis of most suitable biomaterials from population cohorts (stream 5) TNO, CNR, AUTH, UPD, ISS, FERA, CERETOX, UKR, SXS).

In a consensus process and on the basis of state of the art knowledge,  criteria for applicability of omics to banked samples identified under Stream 5 will be established and applied to the biological samples collected in Stream 5. This task will select the samples that need to be studied  to ensure robustness and significance of the results. The measurements will then be carried  out with pilot samples from WP14-16 and performance of the different technologies  assessed  to refine applicability criteria. Based on these, the samples that qualify for extended analysis from the stream  5 case studies  will be selected. Key issues that will be addressed include availability of protocol development, ethics (including addressing  the need for informed consent  renewal), sample handling, biological matrix.

The following questions  will be addressed:

  1. Is the existing experimental  design  compatible with omics experimentation, in terms of subject n, confounding factors, stratification and matching?
  2. Is confirmation of omics derived cohort data possible,  with data generated using in vitro and systems biology models (WP7). In this context, is it possible  to use these data to design  in vitro human  cell-based assays and systems biology experiments,  to mechanistically  anchor  omics observations from cohort studies in light of the linkage between exposure  and health outcome data?
  3. Which extraction/omics procedures are suitable or require further development?

A workshop involving all WP5 partners  will be organized  jointly with Stream 5, to discuss and select cohort samples, back-to-back  with the biomonitoring workshop  mentioned under Task 4.1. The outcome of this task will be described in a working document  to direct the omics technology development and experimentation (Task 5.2) [as part of D5.2]. This working document  will also be synchronized  with the outcome described  in D4.2, since similar principles  to both the application of biomonitoring and omics to human cohort study may apply.


Task 5.2 Develop methodologies for linking omics technologies to population studies (AUTH, CNR, FERA, CERETOX, ISS, UM, SXS)

Based on the outcome of Task 5.2, standard operating procedures (SOP)  and quality assurance schemes will be developed for selected  -omics methodologies. The workflow would involve the following phases: explorative work to optimize protocols;  application to a larger number of cohort samples and finalizing Standard  Operating Procedures, best practices and guidance documents. The technologies are:

  1. Metabolomics [FERA,  AUTH, UPD], the holistic untargeted analysis of all metabolites in easy accessible human body fluids. Established  and robust sample preparation protocols (e.g. for plasma, urine) employing primarily NMR, UPLC-MS and GC-MS will be applied and optimised  to the task [FERA,  AUTH]. For novel matrices of particular relevance  to the EXHES twin study (e.g., meconium, umbilical cord) protocols  will be developed  to establish  metabolome profiles and detect exposures to metals,  dioxins, phthalates, PCBs, PAHs [FERA].  These approaches will be applied  to in vivo (and in vitro, Task 5.3) cohort samples [FERA],  together with metabolomics-specific  bioinformatics analysis that will be developed in WP7. In addition AUTH will assist with metabolome  analysis and develop bioinformatics approaches  for data mining and visualisation of pathways of toxicity (PoT) (work to be done in tandem  with WP7) inferred from the metabolome  to allow for mechanistic hypothesis  building using coupled UPLC-MS, MS-MS TOF  and NMR technologies.
  2. Adductomics of oxidation- and alkylation-induced damage, deals with the measurement  of adducts of (pro-oxidant derived)  electrophiles  with DNA, blood proteins (haemoglobin,  albumin), and glutathione. HPLC-ED, MS and DNA repair protein assays, will be used to monitor the levels of oxidatively damaged  DNA (8-hydroxyguanine) and alkylation damage (O6-alkylguanine)  and repair capacity in human populations [ISS, UM, AUTH]. Measurement of oxidative and nitrogen species stress markers (AOC  (total antioxidant capacity), isoprostanes (products of the free radical-catalyzed  peroxidation of essential  fatty acids), uric acid, cytokines) in blood will be performed by standard assays. [ISS]. Also, UPLC-MS/MS and fixed-step selected reaction monitoring technologies  (FS-SRM) for measuring  protein adducts,  will be further developed  for application in HBM settings  to contribute to the definition of the exposome  [AUTH]. These technologies  will be applied to cohorts selected in Stream 5.
  3. SNP  profiling in genes/loci of relevance  to toxicology in general and for contributing to the establishment  of the exposome  in particular  will be performed. In addition in vitro functional assays of proteins involved in repair of oxidatively generated DNA damage,  will be performed by partner ISS,  as read out for repair capacity.  [ISS, SXS]
  4. Genome-wide DNA methylation  profiling (ISS,  SXS, AUTH) is key to investigating the importance  of epigenetic effects on environmental health. Array and sequencing-based DNA methylation profiling technologies  are used to meet this goal. The new Illumina 450.000 CpG site platform for DNA methylation studies (Infinium HumanMethylation450 beadchip) appears to be a robust and reliable tool. This analysis is based on bisulfite DNA conversion and nucleotide level genotyping of CpG sites located in gene and miRNA promoter sequences, of CpG islands and flanking regions (CpG shores and CpG shelves). Using next-generation sequencing, genome-wide  bisulfite sequencing  technologies,  will be ran on the selected biosamples of the cohorts participating in the population studies  of Stream 5 by AUTH and ISS.  In a next step, the methylation at 48 selected CpG target sites will be characterized in a high-throughput setting using bisulfite conversion  and Taqman chemistry on a Biomark Fluidigm platform. This platform allows us to perform  9216 PCR reactions in one single experiment. SXS will design  and validate this targeted assay. In total, 2000 selected samples will we examined [SXS].  All case studies  will sample biomaterials in accordance with Task 5.2 A pilot study will be selected  for each of the different technologies  to analyze the feasibility and implementation. This will allow refinement of sample generation, measurement  and data analysis. In the final year, the applicability  to samples from other case studies  will be assessed  and measurements carried out.

Task 5.3 Generation of omics data in relevant in vitro models. (CERETOX, UPD)

The hypothesis is put forward that the correspondence  of omics signatures in cohort biomaterials  with those obtained from experimental  in vitro models  can demonstrate  the validity of chosen biomarkers. In fact this approach can be used to corroborate  if: (a) signatures are more closely related to the actual  compound  exposure (e.g. biotransformation) and thus represent a marker for exposure;  (b) more closely related to mechanisms  of disease, and consequently represent a marker for health impact. The causative  effect between exposure  and disease endpoint requires mechanistic  confirmation, which will be based on in vitro assays assessing complex  toxic endpoints relevant to the disease  endpoints of the case studies (such as migration and plasticity; differentiation, endogenous  metabolism; oxidative stress on cell models derived from liver (hepG2/  HuH7, HepaRG), adipocytes (hMADS) or cells differentiating into neurons  (C11) to establish  for the different technologies  omics compound specific hazard signatures. These will be compared to omics signatures in biomaterials for possible correlates.  This task will be supported by mining public data sources containing in vitro toxicogenomics data such as TG Gates.

For all of the above omics technologies first data preprocessing  involving, quality control, deconvolution, sequence analysis etc., will be performed in WP5, by those partners indicated above as responsible  for each of the different omics technologies. Processed data will be further integrated under WP7 towards systems toxicology models  linking human exposure  to environmental stressors to biological pathway alterations and eventually health outcomes.

 

In summary, WP5 will:

  1. Apply SNP,  miRNA analysis and next-generation sequencing  to define genetic susceptibility for chemicals at population level (e.g. DNA repair phenotypes, Phase II reaction genotypes) and identify differences  between epigenetically influenced and independent SNPs
  2. Develop sample handling and metabolomics workflows
  3. Generate metabolite expression  data; identify biomarkers  from Stream 5 cohorts and in vitro models (glutathione, s-adenosyl  methionine, bisphenol A).
  4. Develop methodologies (FS-SRM) and generate DNA and protein adduct data from cohorts.
  5. Determine methyl/hydroxymethylcytosine at specific genomic sequences (promoters, CpG islands, repeated sequences)
  6. Identify Illumina methylation at 20 genes, and apply Methylator phenotypic analysis for these
  7. Generate supportive in vitro omics data with chemicals of relevance  to health endpoints defined in stream 5, to anchor  human in vivo omics findings towards exposure  or mechanism of disease.

 

Methodological considerations

Omics approaches are resource‐intensive, analytically  demanding  and require the use of sophisticated statistical and modelling approaches  to analyse datasets consisting of hundreds  to thousands  of variables in order to minimize false positives (Type I error) and false negatives (Type II error). In HEALS principal biological matrices available  for biomarker discovery  through omics in population-based studies are mostly non-invasive: blood (cells, serum and plasma), urine, maternal milk, saliva, hair and eventually exhaled air, exhaled  breadth.  It is important that the potential constraints  of a given matrix are considered  when evaluating method development for any of the ‘omics approaches. Clearly, our matrix constituted of simple assessments is easily compelled. The diagnostic accuracy of potential identified biomarkers  will be examined and validated according  to international recommendations based on STARD guidelines (www.stard‐statement.org).