Van Eyk, Dunn - Proteomic and Genomic Analysis of Cardiovascular Disease - 2003 (522919), страница 65
Текст из файла (страница 65)
This approach,although 25 years old, has been developed and optimised so that it remains unrivalled in terms of the number of proteins that may be ‘displayed’ on a single gel,up to 10,000 [50], the utility for generating protein maps [17, 51–56], and the analysis of relative levels of protein expression (see Proteomics 1 (10) and [57–62] forexamples). There have been developments in acrylamide and buffer formulationsand in electrophoresis apparatus that improve the reproducibility of 2DE gels.New stains have also been developed that allow for highly sensitive detection ofproteins over a wider dynamic range and that are compatible with subsequentidentification by mass spectrometry [45, 63, 64].
For instance, modified silverstains are available that exhibit sensitivity comparable to the standard silver stainbut that are more compatible with subsequent protein characterisation by matrixassisted laser desorption-ionisation (MALDI) mass spectrometry (MS) [65].
Similarly, Sypro Ruby (Molecular Probes) has been reported to match silver staining interms of sensitivity but it exhibits a more linear correlation between signal intensity and protein level, and interferes minimally with mass spectrometric analysis[63, 66]. Non-fluorescent dyes that do not modify the target proteins and that arecompatible with MS, such as India Ink staining of proteins electroblotted onto nitrocellulose, are also being evaluated [67].There are, of course, many disadvantages with employing 2DE as a platformtechnology for proteomics. The method is notorious for problems with reproducibility, it is labour intensive and it is relatively demanding with regards to theamount of material that is required for each analysis [45, 61, 68]. The quantitativecomparison of multiple gels is also somewhat of a bottleneck in the process fromsample to protein identification.
The development of image analysis software haseased the problem and semi-automated gel analysis, such as the Progenesis system from Non Linear Dynamics, is being explored as an option. The introductionof dual sample labelling, such as difference in-gel electrophoresis (DIGE, Amersham Biosciences), where a control and test sample are labelled with different fluorescent dyes prior to mixing and electrophoresis on the same 2DE gel, minimisesthe problems with gel to gel inconsistency and enables quantitative analysis basedon the relative amounts of the two dye colours within each protein spot [60, 69,70].
The combination of laser capture microdissection with 2DE, especially in se-14.3 Alternative Approaches to Protein Separationlective analysis of human tumour cells, has also shown much promise [71, 72],and sequential detergent extraction techniques have improved the ability to studymembrane-associated proteins by 2DE [42, 73]. Many of the subsequent steps inprotein characterisation have been automated so that spots may be robotically excised, digested and loaded onto MALDI targets and the acquisition of peptidemass fingerprints and searching of protein databases may be performed withminimal user input [74-78]. Despite these advances, the initial separation of proteins on 2D gels is the rate-limiting step in terms of sample throughput so thatthe development of alternatives to 2DE has become an increasingly high priority.14.3Alternative Approaches to Protein SeparationOrthogonal approaches to sample separation, such as multi-dimensional liquidchromatography (MDLC), capillary electrochromatography (CEC), and isoelectricfocussing size exclusion chromatography (IEF-SEC) have proven to be useful whencoupled with tandem mass spectrometry [46, 47, 79–83] and are the preferred methods of many research groups.
However, they suffer from distinct drawbacks, such asthe relatively large quantity of starting material required, particularly if low abundance proteins are of interest, and the lack of quantitative data generated. However, new methods are being developed that will allow quantitative analysis of samples by MS. Isotope coded affinity tagging (ICAT) involves chemical modification ofcysteine residues in the samples with either heavy or light reagents coupled to a biotin tag [84–87]. The samples are then mixed, digested with protease and the biotintagged peptides isolated by affinity chromatography. The simplified peptide mixturemay then be separated by gel electrophoresis or column chromatography and thefractions analysed by MS.
The ratio of heavy peptide to light peptide provides a measure of the relative quantities of the native proteins in the original samples [84–87].The process is relatively high throughput, open to automation and shows great promise as a complementary technology to 2DE.Miniaturisation and automation of sample delivery systems and multi-dimensional separation procedures reduces sample consumption, enables greater sample throughput and may be linked directly to tandem MS for protein detectionand identification [88–96]. Thus LC-MS/MS has been performed on complex mixtures of proteins by detecting and separating components on the basis of mass(by precursor ion scanning) and subsequently identifying each component (byhigh energy collision-induced dissociation) [89, 96–101]. Microfabricated modulesthat enable one- or two-dimensional separation by liquid chromatography or electrophoresis are being developed [88, 91, 94, 95, 102–106], initially based on etchedsilicon devices but more recently on soft-lithography of elastomers, a simpler andmore reproducible option [104].
These ‘lab on a chip’ devices have been used forserial and parallel fragmentation of samples prior to MS analysis.On-chip sample simplification may be performed using the SELDI chip thatcomprises a capture reagent such as cation exchange resin immobilised to a spe-23723814 Protein Chip Technologycialised MALDI target [107]. A subset of the proteins in a mixture bind to the capturesurface under a given set of conditions and can be assayed directly by MALDI-MS.This process enables rapid sample profiling and is particularly suited to determining biomarkers for disease where sample volumes are limited but where samplenumbers need to be high, thus ruling out 2DE as an appropriate method [108–111].
The recent introduction of an adaptor for the Q-TOF that will enable potential biomarkers to be identified by tandem MS has increased the value of the Ciphergen system, particularly to the clinical research environment.The above is not an exhaustive review of the methods available for proteomeanalysis but it provides an overview of separation-dependent approaches. Theseare providing a wealth of information on comparative protein expression profiles,protein-ligand interactions and differential post-translational modification, butthey are still largely dependent on a degree of sample fractionation and the serialanalysis of the protein components of complex mixtures. In order to achievehighly parallel analyses of protein mixtures, protein arrays are being developed.14.4Protein ArraysThe separation-dependent methods described above have developed rapidly to become automated and miniaturised, but they still require relatively large quantitiesof protein sample and operate in a sequential rather than a parallel fashion.
Interest in protein and antibody arrays as tools to complement and perhaps to replaceseparation-dependent methods of protein expression analysis has increased dramatically over the past two or three years. The ability to perform highly parallelproteomics studies in a way that is analogous to those made possible by DNA microarray technology would revolutionise the study of dynamic cellular and metabolic processes [14, 61, 112]. These analyses exploit molecular recognition to isolate, detect and/or identify multiple target molecules simultaneously. Nucleic acidmolecules are remarkably well-suited to such analyses as the specific base-pairingbetween an immobilised probe and its target is predictable and occurs with approximately the same kinetics regardless of the specific DNA/mRNA sequence.Also, labelling of the target sequences with radioactive or fluorescent tags is bothsimple and stoichiometric, and does not affect the interaction of the target withthe probe.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
