The first step to submitting samples is a discussion with the JGI Metabolomics group and program manager regarding experimental design and sample preparation for metabolomics analysis. All samples are different, and it is important to make sure that the correct controls are prepared, enough sample provided, experimental design and samples are compatible for MS analysis, and that extraction method and solvents are suitable to detect metabolites of interest and compatible with the sample and sample container. Measurements for normalization, if necessary, are also discussed (TOC analysis, dry weight, OD, etc).
Prior to submitting all experimental samples, a small subset of samples for a pilot project are determined. This is typically 3-20 samples in size, depending on the experiment. According to the methods and procedures discussed, these will be extracted, run on LC-MS and undergo a preliminary analysis.
Here we will test / determine / decide:
- Amount of sample, sample preparation
- Can we detect the metabolites of interest? Is there anything contaminating the MS signal?
- Sample container
- Is this compatible with extraction solvents, etc?
- Extraction method
- Additional measurements (e.g. dry weight, OD)
- Number of replicates
- How variable are replicates?
- Extraction blanks, conditioned media
- LC-MS method (w/ chromatography)
- HILIC, C18, etc
Once the pilot experiment has been completed and everything “goes well,” the final samples can be submitted. More than 1 pilot may be necessary based on results.
Sample Submission Form
Before shipping samples, a “Sample Submission Form” must be filled out and approved. This provides the relevant metadata necessary to catalog and appropriately analyze the submitted samples. Review of this sheet also ensures all the appropriate controls and samples are being included.
An example form can be viewed here (March 2017 version).
The form has 3 sheets:
- Sheet 1 – Sample information
- Sheet 2 – Media recipes
- Sheet 3 – Metabolites of interest
Sample information. The requested metadata is important for identifying each sample and the components of each sample. Much of this information is needed for generating high quality LC-MS data for subsequent analysis.
Media recipes. Submit detailed media recipes used in the experiment including a list of specific compounds and concentrations. When appropriate, include media preparation steps. For complex media or “environmental” media, include the source or catalog information.
Metabolites of interest. This information is used to ensure that if specific metabolites of interest are important to an experiment, especially in a “targeted” analysis, these will be considered in the analysis. Metabolic pathways or classes of metabolites may also be listed (e.g. amino acids) but please provide as much detail as possible and hyperlinks.
Considerations in Experimental Design
Sample amount. Usually >1 mg C for soil, media, plant exudates. Other typical amounts – soil (0.5-2 g), media (1 mL), wet biomass (~20 mg, 3 mm pellet – often pellet from 1 mL media), dry biomass (2-10 mg).
Number of replicates. Usually 4 replicates (up to 8 if highly variable between sample replicates). Note: For each type of MS analysis, a separate replicate sample needs to be prepared (different extractions used for each).
Sample processing / things to avoid. Some chemical are not compatible with sample extraction (e.g. glycerol, DMSO), or have strong signals that suppress and interfere with detecting metabolites of interest. This includes surfactants, PEG, dirty glass with soap residue, some buffers. Other sources of contamination include polymers leaching from plastic, especially if they contact organic solvent (e.g. filters, acrylate, pipette tips with low adhesive polymer coating).
Media. Defined and minimal media are preferred since competitive ionization and signal suppression is minimized. Also, since metabolites in these can be fully characterized, this makes it easier to detect new peaks / metabolites and to track increases or decreases in metabolites versus control media. Complex media is OK but analysis may also be more complex.
Salt. In general, high salt is not mass spectrometry friendly when using chromatography for primary metabolite analysis (HILIC). For these samples (e.g. ocean), a large amount of salt is removed during extraction but sometimes the remaining salt may still suppress signal during portions of the LC-MS run. For C18 analysis (secondary metabolites, non-polar), salt is not an issue.
Controls – Do not forget!!
Conditioned media control. Incubate media but do not inoculate. This media should be subject to the same processing conditions and environments as the inoculated samples (e.g. filtering, temperature, light, time, etc). This control is necessary to determine what is being made by the organism, or consumed, when compared to the control. Fresh media is not comparable because metabolite degradation or other changes may occur during incubation, especially in the case of high temperature samples.
Extraction blanks (just the containers). Please send the same containers that samples were shipped in but with no sample. This control is necessary to determine sample signal from “background” signal that arises from the container itself, extraction solvents, or somewhere else during LC-MS sample preparation or system itself.
Fresh media. This control is used to evaluate how much the experimental procedures have changed the media, and to determine the initial metabolite composition.
Positive / Negative. When possible, please include a positive and negative control. For instance, if you expect to see synthesis of a metabolite of interest in one sample and not in another.
Unlabeled controls. When performing an experiment with stable isotopic labeling, include replicate samples that do not have any labeling. This control is used to determine the relative isotopic levels of a specific metabolite when unlabeled, and necessary to calculate how much labeling has occurred.
Other. Additional controls may be necessary depending on your experimental design.