Potential useful information to review and add to the readthedocs page - from learning unit 11

[timosachsenberg]

Context: this is some text and images from an older lecture we gave. Maybe something can be reused.

Chemical properties of metabolites
The metabolism is the sum of all chemical processes occurring in an organism at one time. It is concerned with the management of material and energy resources within the cell. Two types of metabolic processes can be distinguished:
Anabolic processes build larger molecules from smaller ones. These processes usually require energy.
Catabolic processes break down larger molecules into smaller units (e.g. nucleotides, amino acids, fatty acids). Theses processes usually release energy.
Metabolites are both educts and products of metabolic processes, the reactions are catalized by proteins. A series of metabolic processes is called a metabolic pathway.

The size of the human metabolome or other metabolomes is difficult to determine because in addition to endogenous compounds which are synthesized by the metabolism it also consists of exogenous compounds which depend on the environment (e.g. food consumption). As of 2014 the Human Metabolome Database (HMDB), the most comprehensive collection of the human metabolome, contains more than 40,000 metabolites of which about 20,000 have been measured or experimentally confirmed while the remaining metabolites are expected to occur in the human body but have not been measured so far. These numbers are likely to increase over time as more sensitive methods become available. In comparison, it has been estimated that the metabolome of the plant kingdom consists of 200,000 - 1,000,000 different metabolites.

The metabolome is much more diverse than the proteome. Proteins are composed of only 20 different amino acids which are connected by peptide bonds whereas the metabolome is composed of very diverse compound classes such as amino acids, sugars, sugar phosphates, lipids, phospholipids, fatty acids, organic acids, and alcohols. Consequently the chemistry and the fragmentation behavior of metabolites is equally diverse. In addition metabolites are in general not linear polymers, thus there is no ion series in the fragment spectrum. The fragmentation pattern of metabolites is difficult to predict and depends on the instrument and the fragmentation energy. As a result the identification of metabolites is much more challenging than the identification of peptides.

Due to the chemical and structural diversity of the metabolome it is also impossible to perform quantification using a simple in vitro labeling strategy (such as iTRAQ in proteomics) for arbirary metabolites. Instead a metabolic labeling has to be done in vivo. Feature detection is complicated by the fact that S, Cl, Br frequently occur which leads to complex isotope patterns. Due to the different composition of metabolites the averagine hypothesis cannot be applied.

Similar to targeted and non-targeted proteomics two fundementally different approaches to measure the metabolome can be distinguished:

Targeted metabolomics tries to identify only a well-defined subset of known metabolites. The chosen subset of metabolites can be reliably identified and quantified, but

Non-targeted metabolomics tries to identify as many compounds as possible in an unbiased manner. Whereas the majority of the metabolites can be seen only a small fraction of them can be identified in a typical experiment since metabolite identification is much more challenging than peptide identification.

---

Metabolite ID via Spectral Matching and Accurate Mass
Metabolites are commonly identified using their accurate mass and their fragmentation spectras. Additional information such as the retention time can be used to increase the cofidence of the identification or to exclude unlikely identifications.

Given a sufficient mass resolution the mass alone contains valuable information about the metabolite. Modern mass spectrometers such as the Orbitrap achieve a mass accuracy around 1 ppm which reduces the number of potential matches considerably. However even accuracies below 1 ppm would not be sufficient to yield unique results using just the accurate mass. In addition structural isomers, i.e. molecules with the same molecular formular but a different structural formular cannot be resolved.

D-Alanine and Sarcosine are structural isomers. Both have the molecular formula C3H7NO2 but a different structural formula. Both molecules cannot be distinguished using just their accurate mass.

More confident identifications are possible through the measurement of fragmentation spectras (MS2 or MSn). The experimental spectras can be compared with fragmentation spectras of known compounds that have been measured previously. However most spectrum databases are commerial and the fragmentation pattern depends both on the instrument and the fragmentation energy, so that suitable spectras are often not available. Alternatively the experimental spectras can be compared to theoretical fragmentation patterns that have been predicted using methods such as fragmentation trees.