The development of methods for the global analysis of the protein complement of the cell is still in its infancy. Since proteins are vastly more physico-chemically diverse than nucleic acids, a universal separation method is unlikely to be found and this is further compounded by the lack of an amplifying method analogous to PCR. There are only two partially satisfactory methods for analysing the state of expression of the majority of proteins in a cell are those based on two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) and the new non-gel based methods discussed later. In the former technique, proteins are separated in the first dimension according to their isoelectric point, i.e. by migrating to a point in the gel where the pH causes the net charge on the protein to become neutral. In the second dimension they are separated according to their mobility in a porous gel, which is proportional to the amount of detergent, sodium dodecyl-sulphate bound which is approximately mass dependent. However 2D-gel technology suffers many drawbacks. The separation on a single gel can show up to 10,000 species, however many of these are due to post-translational modifications and the number of gene products being visualised is probably only of the order of 1,000. The basic format of 2D-PAGE has not changed that drastically since its introduction by O´Farrel and Klose in 1975. The increase in reproducibility that has been brought about by the introduction of commercial immobilised pH gradient first dimension gels (IPG) allows very accurate and quantitative comparative 2D gel mapping. Detailed 'proteome maps' can be created with advanced computer imaging programs and then analysed by subtractive or cluster methods to find relationships between the protein spots. The weakness of 2D-PAGE lies in its inability to deal with certain classes of proteins, mostly highly hydrophobic ones (membrane and cytoskeletal especially) and those with isoelectric points at either extreme of the pH scale (such as acidic hyper-phosphorylated and alkaline DNA binding proteins). There are also problems with quantitation due to the low dynamic range of stains.
Schematic illustration of standard proteome analysis by 2DE-MS.
The basic scheme of 2D-based proteome analysis is shown above diagrammatically. The proteins are separated by 2DE and stained and the images analysed. Significant protein spots are excised, digested ‘in-gel’ with trypsin and the resulting peptides are analysed by MALDI mass spectrometry. The peptide masses can be used to search DNA or protein databases to identify the protein. The set of masses generated experimentally are compared to the theoretical digest patterns of all proteins in the database and the closest matching are displayed. This has been termed protein-fingerprinting. Alternatively the protein digest can be separated by HPLC and the eluent is analysed by electrospray ionisation, enters the mass spectrometer mass spectrometry. Each peptide is sequentially isolated and fragmented to generate a ‘peptide fingerprint’ (MS/MS) spectrum analogous to the protein fingerprint. The MS/MS spectrum from the selected peptide is compared to predicted tandem mass spectra from all peptides in the sequence database with a similar mass to identify the protein.