Proteomics methods

‘Global’ protein quantification

These experiments are aimed at the identification of many proteins and allow comparison of relative protein quantities between samples.  The changes in the proteome are assessed depending on an experimental variable e.g. diseased vs. healthy, sample +/- addition of some external factor or upon knockdown/transfection with a particular gene.

Within CADET, we primarily use iTRAQ tags from ABSciex (isobaric Tags for Relative and Absolute Quantitation) to achieve relative quantification [1]. These tags allow up to eight samples to be compared simultaneously, although for experiments with more than eight samples, each sample can be compared to a standard reference to allow comparisons between experimental runs. Our workflow consists of a reproducible protein extraction step, followed by trypsin digestion of proteins, and labelling with iTRAQ tags. Samples are pooled and then the peptides are fractionated, firstly by high pH reverse phase chromatography and then by low pH reverse phase nanoflow chromatography, from where they are directly eluted into the mass spectrometer. The mass spectrometer detects peptides and fragments them (tandem MS, or MS/MS) to provide structural (sequence) information and relative quantification by virtue of the release of iTRAQ tagged ‘reporter’ ions (Figure 1).

Figure 1 iTRAQ comparison of up to 8 samples by comparing specific reporter ions

In some cases, especially with clinical samples where patient-to-patient variability can be a key factor in experimental design, the restriction of the iTRAQ 8-plex on sample number is limiting. To this end, we are implementing label-free quantification approaches where samples are analysed sequentially in ‘MS-only’ mode. Peptide peak heights are extracted and compared across samples to identify peptides whose signals are consistently higher or lower in one experimental group.

‘Targeted’ protein quantification

Frequently researchers are interested in the behaviour or characteristics of only a specific protein. For this kind of experiment, we use a targeted proteomics approach.

Samples for targeted protein analysis are most often analysed using Selected Reaction Monitoring-Mass Spectrometry (SRM-MS; Figure 2). This method can be used for confirming the presence of a protein or investigating novel post-translational modifications on a target protein, using a workflow called MRM-Initiated Detection and Sequencing (MIDAS), where a specific SRM is used to trigger MS/MS on a peptide of interest [2]. More commonly, however, we use SRM analyses to quantify peptides, either relatively between samples or to gain an approximate concentration (Figure 3).  This can be useful, for example, to measure the levels of a post-translationally modified peptide in samples under various conditions, or to measure the levels of a particular protein in healthy individuals compared with those who have a particular disease.


Figure 2 Selected Reaction Monitoring (SRM) for specific detection and quantification of target peptides

Figure 3 SRM allows quantification of protein levels by mass spectrometry

(MDH = malate dehydrogenase)

Plasma proteomics

Plasma is a special case when it comes to analysis of its proteome, because of the huge variation in protein concentrations.  This variation can be up to 1010-fold between the most highly expressed protein (albumin) and lower abundant yet highly relevant proteins e.g.  interleukins. The dynamic range of most ‘discovery’ mass spectrometers is limited to 104-105. To alleviate this problem to an extent, we have established a semi-automated workflow based on the Agilent Multi-Affinity Removal System (MARS-14) where we can reproducibly deplete the top 14 most abundant proteins in the sample.  This allows us to identify and quantify lower abundance proteins, including non-plasma proteins that ‘leak’ from the tissues and where disease markers may reside [3].

[1] Unwin RD, Griffiths JR, Whetton AD. (2010) ‘Simultaneous analysis of relative protein expression levels across multiple samples using iTRAQ isobaric tags with 2D nano LC-MS/MS.’ Nature Protocols 5:1574-82. [ 21085123]

[2] Unwin RD, Griffiths JR, Whetton AD. (2009) ‘A sensitive mass spectrometric method for hypothesis-driven detection of peptide post-translational modifications: multiple reaction monitoring-initiated detection and sequencing (MIDAS).’ Nature Protocols 4:870-7. [ 19444244]

[3] Blankley RT, Fisher C, Westwood M, North R,  Baker PN, Walker MJ, Williamson AJ, Whetton AD, Lin W, McCowan L,  Roberts CT, Cooper GJS, Unwin RD and Myers JE. (2013) ‘A label-free SRM workflow identifies a subset of pregnancy specific glycoproteins as novel putative predictive markers of early-onset pre-eclampsia.’ Molecular and Cellular Proteomics. 12:3148-59. [ 23897580]