Metabolomics PDF

METABOLOMICS PIM LEONARDS Department of Environment & Health 1 AFTER THIS LECTURE, YOU Know what metabolomics is Understand how metabolomics can be used Can explain the basic workflows of metabolomics Know what the challenges are in metabolomics...

METABOLOMICS PIM LEONARDS Department of Environment & Health 1 AFTER THIS LECTURE, YOU Know what metabolomics is Understand how metabolomics can be used Can explain the basic workflows of metabolomics Know what the challenges are in metabolomics Vrije Universiteit Amsterdam CONTENT What is metabolomics? Why do we use metabolomics? Workflows of metabolomics Sampling Extraction Detection techniques (targeted vs. untargeted) Data analysis Challenges Applications Pesticide rat exposure study Vrije Universiteit Amsterdam WHAT IS METABOLOMICS? Metabolomics Systematic study of small organic molecules (5000 compounds transcripts proteins metabolites 11 Vrije Universiteit Amsterdam Primary and secondary metabolites Primary metabolites Involved in essential life processes (e.g. sugars, amino acids, organic acids) Used for growth and development Primary metabolic processes Glycolysis Respiration Photosynthesis Secondary metabolites Restricted distribution in cells and organisms Differences between tissues and species Do not participate in growth and development Nonessential for life Important to influence ecological interactions (e.g. protection) Secondary metabolites are produced by pathways derived from primary metabolic routes Vrije Universiteit Amsterdam Example of secondary metabolite Tetrodotoxin Strong neurotoxic compound Found in Tetraodontiformes fish But also in amphibian (e.g. rough-skinned newt) Inhalation of 0.15 milligram can kill a human LD50 mouse is 334 μg/kg Tetrodotoxin can be made by Pseudomonas bacteria https://upload.wikimedia.org/wikipedia/commons/1/1b/Rough -Skinned_Newt.JPG https://upload.wikimedia.org/wikipedia/comm ons/4/4e/Pufferfish_%28Butete%29.jpg Vrije Universiteit Amsterdam Why metabolomics? APPLICATION AREAS METABOLOMICS Health and medical “Hielprik” at birth: association between a specific metabolite and a disease: Biomarker discovery Example carnitine as marker for a Diagnostic metabolic disease Personalised medicine Carnitine transporter (OCTN2) deficiency Pharmaceutical Agriculture Development of pesticides Toxicology Foto: ANP Environment Environmental metabolomics Metabolomics analysis in biological systems that are exposed to environmental stress, such as the exposure to environmental contaminants Vrije Universiteit Amsterdam  APPLICATIONS IN ENVIRONMENTAL METABOLOMICS Deriving points of departure via benchmark dosing X X Dose-response X X X X Mechanism of toxicity/ mode(s) of action or molecular key events Molecular toxicity pathways and molecular key events (KEs) -> AOPs Linking metabolic pathways with organism’s functioning and MIE AO phenotype Biomarker discovery Specific metabolite as marker for exposure or effect disease Diagnostic markers of disease Chemical grouping for biological read-across Regulatory field read-across approach of chemicals Systems toxicology Combination of omics to further understand relationship between different omics and chemical exposure and MoA Targeted and untargeted metabolomics Targeted metabolomics To focus on changes in the abundances of specific metabolites in a biological system after environmental contaminant exposure Untargeted metabolomics To provide a ”complete” overview of changes in abundances of all detectable metabolites in a biological system after environmental contaminant exposure: Vrije Universiteit Amsterdam  BASIC METABOLOMICS WORKFLOW Research question Experimental Interpretation design Metabolite annotation targeted untargeted Statistical Sampling & analysis extraction Data Data processing acquisition 2.5 2.4 2.3 2.2 2.1 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5 0.4 19 0.3 0.2 0.1 0 5 6 7 8 9 10 11 12 13 Vrije Universiteit Amsterdam UNTARGETED METABOLITE EXTRACTION Tissue Quenching of sample with cold solvents Quenching Solvent extraction Polar fraction Polar compounds Apolar fraction Apolar compounds Add internal Extraction techniques standard Beads beating using ceramic beads Ultrasound assisted extraction Tissue disruption / …. metabolite extraction Extract concentration Add solvent(s) for analysis Final analysis EXTRACTION EFFICIENCY * p value < 0.05 Beads beating * * Ultrasound assisted Log10 Peak Intensity (n=5) * * * Tufi et al., Anal. Bioanal. Chem., 2015, 407 (7), 1901-1912 ANALYSIS: DETECTION AND SEPARATION TECHNIQUES Mass spectrometry NMR  More sensitive  Less sensitive  Large range of metabolites  Limited number of detected metabolites detected  Often combined with gas  Little sample handling (GC) or liquid  Quantification more easy chromatography (LC)  Very reproducible  More difficult in quantification Vrije Universiteit Amsterdam DETECTION TECHNIQUES Targeted metabolomics LC- GC-MS NMR MS/MS Untargeted metabolomics Direct LC MS GC-MS HRMS (DIMS) Vrije Universiteit Amsterdam DATA PROCESSING UNTARGETED METABOLOMICS LC-QToF Peak alignment of different chromatograms Peak aligning sample 3 sample 2 sample 1 Peak deconvolution Peak annotation / & peak picking Generating a peak identification of table metabolite structure Integration of the peak heights Vrije Universiteit Amsterdam Identification of unknown metabolites Retention time Low resolution High resolution MS MS LC Rt prediction Electron impact Exact mass elution order (EI) spectrum mono-isotopic Mainly GC mass GC Retention time Comparison with EI Isotope pattern index MS spectral Rt libraries MS/MS spectrum Monoisotopic mass EI spectrum Isotope pattern m/z m/z MS/MS Rt (min) spectrum m/z Vrije Universiteit Amsterdam Identification of unknown metabolites Two main approaches: Match MS(/MS) spectra with MS(/MS) libraries EI MS libraries (large libraries available) MS/MS HRMS libraries (MS/MS depends on settings and equipment used HRMS: Prediction of elemental composition based on exact mass and isotope pattern Exact mass Monoisotopic mass Predict elemental composition e.g. C6H12O6 Isotope pattern m/z Vrije Universiteit Amsterdam Example untargeted metabolomics Chromatogram LC-QToF untargeted Selecting of one m/z value, for metabolomics. QToF measures all ions example m/z 952.084 provides a together new chromatogram with only the peak(s) that contains this specific m/z value m/z 952.084 Peak number Retention time m/z value 1 1.2 102.2498 2 1.4 203.0900 Peak table 3 1.4 105.0908 4 2.4 156.7800 5 4.5 345.0609 6 5.6 344.5672 7 5.7 126.7842 8 8.9 254.6783 9 8.9 434.5631 10 10.1 536.0948 … … Vrije Universiteit … Amsterdam Example annotation of unknown metabolite based on exact mass Exact mass / potential elemental composition 952.071: C61H22N4OP2S2 Isotope No. Formula Mass 1 C41H28O27 952.082 952.089: C61H22N4O3P2S filter 2 C41H28N8O20 952.142 952.081: C61H23N4OP3S 3 C41H28N16O13 952.202 952.098: C61H23N4O3P3 4 C41H28N24O6 952.262 952.090: C61H24N4OP4 952.097: C61H29O4PS3 952.088: C61H30O2P2S3 952.080: C61H31P3S3 952.098: C61H31O2P3S2 Database comparison 952.090: C61H32P4S2 (MS/MS spectra) 951.945: C61H119N6O 951.914: C61H123O4S 951.932: C61H123O6 Potential 951.906: C61H124O2PS candidate 951.924: C61H124O4P 951.915: C61H125O2P2 951.907: C61H126P3 Data from Kind and Fiehn: BMC Bioinformatics 2006, 7:234 Vrije Universiteit Amsterdam Annotation and identification Annotation: Tentative identification of a metabolite Identification: Exact structure of the metabolite based on confirmation with an analytical standard Vrije Universiteit Amsterdam Identification confidence levels Confidence levels based on Schymanski et al., 2014, Environ. Sci. Technol. 2014, 48, 2097−2098 Vrije Universiteit Amsterdam Quality control (QC) and quality assurance (QA) Quality control (QC) to monitor the performance of metabolomics workflows Untargeted metabolomics complicated process Sample Data Analysis processing Failures processing Measurement Drift Missed peaks errors Poor integration Vrije Universiteit Amsterdam Quality control (QC) Randomization of sample analysis order Pooled samples (QC sample) Correct for intensity drifts Correct for inter-batches differences Individual samples Pooled sample of small amount of each sample (10 µl) Pooled sample Vrije Universiteit Amsterdam Pooled QC samples Sample analysis in 3 batches analysis order From Broadhurst et al., Metabolomics (2018) 14:72 Evaluation of coefficients of variation of each metabolite Correction for inter-batch systematic error Vrije Universiteit Amsterdam QA/QC: Internal standards targeted metabolomics What are internal standards? These are metabolites which do not naturally occur and are used to correct for the loss of metabolite extraction during the sample treatment/quenching/extraction, but can also correct for incorrect measurements Examples are deuterium or 13C labelled metabolites Deuterium and 13C labelled glucose Vrije Universiteit Amsterdam  STATISTICAL MULTIVARIATE DATA ANALYSIS Principal component analysis (PCA) Partial least squares regression (PLS) Partial least squares discrimination analysis (PLS-DA) Video explaining PCA https://www.youtube.com/watch?v=_UVHneBUBW0 https://www.youtube.com/watch?v=HMOI_lkzW08 35 https://www.youtube.com/watch?v=kw9R0nD69OU Vrije Universiteit Amsterdam INTERPRETATION: METABOLOMICS PATHWAY ANALYSIS Examples of pathway databases Kegg The Human Metabolome Database Lipidmaps Deoxyribonuclei Chlorinerigic c acids Glycerophospholipid Amino acids Nucleosides GABAergic Glutamine Dopamergic Serotonergic Short chain FA CHALLENGES IN METABOLOMICS Metabolome is very dynamic, therefore, very time sensitive Metabolites have a wide range of physico- chemical properties and vary widely in concentration Some metabolites are very unstable during sample collection and sample treatment Identification of chemical structure of unknown metabolites is still a big challenge Example of an application of environmental metabolomics Vrije Universiteit Amsterdam LINKING AFFECTED METABOLIC PATHWAYS WITH ORGANISM’S FUNCTIONING AND PHENOTYPE Chemical Metabolic Behaviour exposure pathways Cognition WINDOWS OF EXPOSURE Learning, cognitive, motor activity tests GESTATION WEANING Birth GD 0 GD7 GD21 PND21 PND60 PND75 PND90 PND110 Exposure Pups exposure only by mother milk ♂♀ adults exposed EXPOSURES Endosulfan (organochlorine) Global ban on the manufacture and use of endosulfan 2012, but still used in some countries Exposure 0.5 mg/kg/day Sex-specific effects Increased anxiety and impaired spatial learning in males, but not in females METABOLOMIC AND PROTEOMICS STUDIES OF BRAIN TISSUES Cerebellum  Motor control, motor learning  Cognitive functions (attention and language) Cerebellum Hippocampus Cortex Cortex  Sensory, motor, and association areas  Voluntary movements, especially fine fragmented movements Hippocampus  Spatial memory Striatum  Short-term memory to long-term memory and spatial navigation Striatum  Coordinate body movement (fine-motor functions), working memory NON-TARGET METABOLOMICS BRAIN TISSUES PLS-DA VH CB_CX_HP_ST data final.M1 (PLS-DA) Cortex Colored according to classes in M1 Hippocampus Striatum Cerebellum Cerebellum Hippocampus Cortex Striatum CEREBELLUM CYPERMETHRIN, ENDOSULFAN AND CONTROL Bucket CB CYP END IS correctie final.M8 (PLS-DA), PLS-DA CB CYP END Cypermethrin Colored according to classes in M8 Endosulfan Control END F CYP F CYP M Endosulfan CYP M Cypermethrin END F END F CYP F END M CYP F END F CYP M END M CYP F VH M VH FI CYP M VH F ENDM END F VH M VH MI END M VH F END F VH F Control METABOLIC PATHWAYS ENDOSULFAN HIPPOCAMPUS Male Female Amino acids GABAergic Glutamine Significant metabolites compared to control Down-regulated Up-regulated Non-significant metabolites GABA PATHWAY HIPPOCAMPUS VS BEHAVIOUR Males Females glutamine glutamine Effects of endosulfan on glutamic acid glutamic acid learning, behavior and motor activity GABA Males Female GABA s Y-maze  GABA levels highest in Learning Water maze hippocampus Radial maze  GABA has important role in Motor spatial learning Motor activity activity and coordination Rotarod Beam walking  SUMMARY  Metabolomics studies small molecules using a combination of analytical chemistry, biochemistry, bioinformatics  Different analytical techniques are needed to determine the metabolome  Sample handling is the most crucial factor in metabolomics analysis ‹#› 47 Het begint met een idee 47 Faculty / department / title presentation Het begint met een idee  SUMMARY  Metabolites have a wide range of physico-chemical properties and concentrations  Some metabolites are very unstable during sample collection and sample treatment  Untargeted metabolomics is a complicated process with many different steps  Crucial is the identification/annotation of the metabolites ‹#› 48 Het begint met een idee 48 Faculty / department / title presentation Het begint met een idee