Nowadays, metabolomics and lipidomics are a dynamically evolving parts of analytical biochemistry dealing with the study of metabolome/lipidome. A metabolome is defined as a set of low-molecular-weight compounds (metabolites) present in the biological matrices which are involved in all the metabolic processes necessary for the normal functioning of the cell and the whole organism. Lipidomics deals with a study of water-insoluble biomolecules (lipids) that are present in a given biological system.
Metabolomics appropriately complements “older” fields such as transcriptomics, proteomics, and genomics. Physiological changes that manifest themselves as a result of deviation in gene expression are characterized by a change of a metabolome as well as lipidome. These changes are more sensitive than alternation in transcriptome or a proteome. Even a small change in protein expression can have a huge effect on the activity of the biochemical pathway and the concentration of the relevant metabolites and lipids.
Targeted or untargeted approach is used for sample analysis. The first one is based on quantitative or semi-quantitative measurement of certain groups of metabolites/lipids (e.g. substrates and products of basic biochemical pathways; compounds belonging to a particular chemical group). The untargeted metabolomic/lipidomic analysis is characterized by semi-quantitative analysis of a broad spectrum of metabolites (thousands of features).
- 30 years’ experience with multicomponent analysis by mass spectrometry and 10 years’ experience with metabolomics.
- Robust and standardized metabolomic pipeline based on advanced mass spectrometry.
- Development of new analytical methods.
- Flexible data processing and thorough statistical analysis.
- Data evaluation and biochemical interpretation.
- Study of drug metabolization.
- Structural elucidation of unknown compounds.
Design of experiment
- One of the most critical steps in metabolomic/lipidomic project.
- It is crucial to plan experiments together with statisticians, analytical chemists, biochemists and (eventually) physicians to correctly adjust numbers and type of subjects/samples. Samples must be taken representatively and precisely.
- The targeted approach is usually chosen for a better understanding of pathobiochemical mechanisms of the disease. Untargeted methods are used to search for new potential biomarkers.
- For metabolomics, the stop metabolic processes and changes in metabolite profiles are required.
- Fast preparation by organic solvents or cell quenching are usually used.
- We are experienced with preparation of urine (dilution on creatinine concentration); plasma/serum (deproteination); cells (lysis & deproteination); dry blood spot (deproteination/extraction); sweat (deproteination); breath condensate (dilution); tissues (homogenisation & deproteination).
Methods of analysis
- Targeted metabolomics:
- HILIC UHPLC/QqQ (TripleQuad 6500, Sciex); aminopropyl column under alkaline conditions; relative quantitation of 365 compounds in one run (amino acids | organic acids | purine and pyrimidine bases, ribosides and nucleotides | sugar phosphates | acylcarnitines | fatty acids | carbohydrates | amines).
- FIA/QqQ (API 4000, Sciex); quantification of amino acids and acylcarnitines.
- Untargeted metabolomics:
UHPLC/HR-MS (Orbitrap Elite, Thermo); profile of 1000+ metabolites and phospholipids.
- Targeted lipidomics:
- UHPLC/QqQ (TripleQuad 6500, Sciex); C8 column; relative quantitation of 1000+ lipid compounds in positive and negative mode.
- Untargeted lipidomics:
- RP UHPLC/HR-MS (Orbitrap Elite, Thermo); lipid classes in biofluids, cells and tissues; profile of 1000+ compounds.
- Compound discoverer 3.0
- R software – LOESS for QC processing
- Data transformation (clr, ilr, log, PQR)
- Search for the most discriminant metabolites/lipids between groups
- Univariate statistical methods (Boxplots, Volcano plots)
- Multivariate data analysis (PCA, Hierarchical Cluster Analysis, PLS-DA, OPLS-DA, Heat-maps, Random forest)
- Correlation analysis
- ROC curve
Interpretation of findings
- HMDB, SMPDB, Metlin, mzCloud, ChemSpider, Kegg, PubChem, Lipids maps, Metagene, BRENDA, UniProt
- Biochemical interpretation
- Structure identification