The preparation of samples in metabolomics studies

CATHERINE WINDER: As We have discussed the metabalome is very dynamic in nature. The metabalome can change in seconds and minutes. If samples of different metabalomes are not collected, processed, stored, and extracted appropriately then the metabolites present and more importantly the concentration of metabolites are likely to have changed and the sample analysed will not be directly representative of the biological sample at the time of sampling. Within the context of sample preparation we will consider the sample type and how this impacts on sample collection, how the sample is stored, and how biological samples are extracted. There are many different types of biological samples that can be investigated in metabolomics studies.

The choice of sample is dependent on many factors, including the biological question being asked and the availability of appropriate samples. If we think about the question and clinical studies, if we are searching for a biomarker of a disease to apply in clinical practise then we study a biofluid like blood or urine as the sample is easy to collect for clinical practise. However, if we were interested in the molecular mechanism of a specific disease then we would prefer to investigate a sample where the mechanism is in operation, the cell of tissue, rather than a biofluid. Metabolism operates in a constant flux where metabolite concentrations can change rapidly.

Many metabolomics experiments capture a snapshot of the metabolome at a given point in time. In samples that are metabolically active- those that contain enzymes- such as cells and tissues, metabolism should be rapidly inhibited when the sample is collected to prevent metabolic turnover. We call these metabolic quenching. If this is not performed then the sample analysed will not be representative of the metabolome sample at the point of sampling. Metabolic quenching is achieved by a rapid change in temperature of the sample to stop metabolic turnover. The change in temperature reduces or stops how enzymes operate and therefore stops metabolic activity. The methods applied to perform this step were discussed in the previous article.

In samples that are not metabolically active- those which do not contain enzymes- then metabolic quenching is not required. Examples of such samples include the growth media from microbial cultures and urine. These samples provide a cumulative picture of metabolism over time rather than a snapshot of metabolism as is case with samples where enzymatic activity is in operation. The collection and storage of samples prior to their preparation for analysis should be considered in the experimental design. The collection of samples my require specialised training and equipment that may not be available at the site of sampling.

For example, maintaining samples at the appropriate temperature may require dry ice or liquid nitrogen and it may be problematic to provide this in a clinical environment due to health and safety regulations. All samples for analysis of the metabolome should be stored in chilled conditions at minus 80 degrees centigrade or if this is not possible, then at minus 20 degrees centigrade to ensure that enzyme activity does not return and to minimise degradation of unstable metabolites. In the preparation of samples for analysis of the metabolome, samples from all classes should be treated in the same way throughout the collection and preparation procedure. If samples from different classes are treated differently it is likely that bias will be introduced into the study.

Experimental procedures should be standardised and optimised to quench, extract, and store the samples in an identical manner. Small differences in the processing of samples can have a large effect on the sample quality and validity of the data acquired. For example. If all the samples from one sample group are collected in one hospital and a different sample group in a different hospital, the difference in collection site may be identified in the downstream analysis of the data and initially highlighted as of biological significance, whereas the observed metabolic differences are actually a combined result of biology and the collection site. Materials used throughout the procedure should be collected and tested.

For example, it is advisable to use sample collection tubes and other plastic consumables from the same supplier and production batch for the duration of the study. Otherwise the lengthy analysis of metabolomics data may result in the identification of a fantastic biomarker, that is actually an artefact introduced by using different batches sample collection tubes rather than metabolites of biological significance. The method necessary to extract the metabolites from the sample will vary depending on the sample type. This may be relatively simple, to analyse urine by liquid chromatography mass spectrometry the sample is simply diluted and analysed.

If gas chromatography mass spectrometry used to analyse urine, urease treatment can be applied to degrade urea which is in a large excess compared to the metabolites and without urea being degraded would not allow the metabolites to be detected. Serum or plasma requires the removal of proteins prior to mass spectrometric analysis so to stop blockages in the chromatography system. This is typically performed by the addition of solvents- methanol or acetonitrile, which precipitates the proteins- the precipitate is removed by centrifugation and the liquid phase can be analysed. Tissue and cell samples require additional procedures that may include mechanical approaches to assist with the liberation of metabolites from the cells.

Homogenizers and tissue lysers may be employed to tissue samples to homogenise and break open cells whereas multiple freeze-thaw cycles may be sufficient to permeabilise the cell walls of microbial cells. Finally, the extraction solvents applied are broadly separated into water-soluable and lipid-soluable methods, the solvent is selected to solubilise the metabolites of interest in the study. For example, chloroform is applied to extract lipid-soluable metabolites or methanol and water are applied to extract water-soluable metabolites.