在蛋白质组学研究策略的能力和局限的电泳分离.doc

在蛋白质组学研究策略的能力和局限的电泳分离Power and limitations of electrophoretic separations in proteomics strategiesThierry. Rabilloud 1,2, Ali R.Vaezzadeh 3 , Noelle Potier 4, Cécile Lelong1,5, Emmanuelle Leize-Wagner 4, Mireille Chevallet 1,21: CEA, IRTSV, LBBSI, 38054 GRENOBLE, France.2: CNRS, UMR 5092, Biochimie et Biophysique des Systèmes Intégrés, Grenoble France3: Biomedical Proteomics Research Group, Central Clinical Chemistry Laboratory, Geneva University Hospitals, Geneva, Switzerland4: CNRS, UMR 7177. Institut de Chime de Strasbourg, Strasbourg, France5: Université Joseph Fourier, Grenoble FranceCorrespondence : Thierry Rabilloud, iRTSV/LBBSI, UMR CNRS 5092,CEA-Grenoble, 17 rue des martyrs, F-38054 GRENOBLE CEDEX 9Tel (33)-4-38-78-32-12Fax (33)-4-38-78-44-99e-mail: Thierry.Rabilloud cea.fr Abstract: Proteomics can be defined as the large-scale analysis of proteins. Due to the complexity of biological systems, it is required to concatenate various separation techniques prior to mass spectrometry. These techniques, dealing with proteins or peptides, can rely on chromatography or electrophoresis. In this review, the electrophoretic techniques are under scrutiny. Their principles are recalled, and their applications for peptide and protein separations are presented and critically discussed. In addition, the features that are specific to gel electrophoresis and that interplay with mass spectrometry( i.e., protein detection after electrophoresis, and the process leading from a gel piece to a solution of peptides) are also discussed. Keywords: electrophoresis, two-dimensional electrophoresis, isoelectric focusing, immobilized pH gradients, peptides, proteins, proteomics. Table of contentsI. IntroductionII. The principles at playIII. How to use electrophoresis in a proteomics strategyIII.A. Electrophoretic separation of peptides.III.A.1. Capillary IEFIII.A.2. Solution IEF III.A.3. IPG-IEFIII.A.4. Off-Gel III.A.5. A new dimension for data validationIII.B. Electrophoretic separation of proteins. III.B.1. Electrophoretic separations of native proteinsIII.B.2. Electrophoretic separations of denatured proteins-rationaleIII.B.3. The implementation of denaturing protein electrophoresis in proteomicsIII.B.4. The workhorse of protein separation in proteomics: SDS gel electrophoresisIII.B.5. Classical high-resolution 2D electrophoresis: the source of proteomicsIII.B.6. Specialized, IEF-free, 2D electrophoretic systemsIV. From gels to peptides: additional features of electrophoresis in a proteomics strategyIV.A. Detection of proteins after gel electrophoresisIV.B. Production of peptides from gel-separated proteinsV. Concluding remarksVI. ReferencesI. IntroductionAmong the variety of available spectroscopic techniques, the combination of speed, sensitivity, and accuracy makes mass spectrometry an obvious choice in proteomics strategies; i.e., strategies that aim at the wide-scale characterization of proteins and protein variants in biological samples.However, mass spectrometry alone cannot solve this analytical problem of the wide-scale characterization of proteins, for several major reasons. First of all, biological samples often contain several hundreds to several tens of thousands of different protein forms. The word of “protein forms” encompass, of course, gene products with all of their post-translational modifications (PTM); e.g., glycosylation, phosphorylation, or protein cleavage. On top of the diversity problem, the dynamic range of protein presence ( i.e., the mass ratio between the rarest protein form and the most abundant ones) is often far beyond the quantitative dynamic range of a mass spectrometer. For example, this dynamic range covers 4 orders of magnitude in prokaryotic samples, 6 in eukaryotic cells (Lu et al., 2007), and 12 in complex biological fluids such as plasma (Anderson and Anderson, 2002). Still going on this trend of complexity, the analyte itself (i.e, the protein) is very complex, so that even the most precise mass measurement of a complete, modified protein often does not give an unequivocal answer, in the sense that several modified proteins can fit within the same mass measurement window.Last but not least, it is quite frequent in biology to have to address quantitation problems. A good example is provided by blood tests, in which normal and pathological values are defined, with the normal values frequently not zero. Thus, a quantitative approach is often needed, and this goal is not always easy to implement for mass spectrometry. However, the discussion of the quantitation issues in mass spectrometry for proteomics is clearly outside the scope of this review. Because of the complexity of the analytes, the mass spectrometric measurement is not carried out on intact proteins, but on smaller peptides that are produced from the proteins by a controlled proteolytic process. The multiple measurements made on peptide masses, either on a few peptides by MS/MS or on a combination of peptides that arise from the same protein (e.g., for peptide mass fingerprinting), allow an unequivocal characterization of the proteins of interest and sometimes of some post-translational modifications.Because of the complexity of biological samples, it is absolutely necessary to separate (and sometimes to quantify) the analytes prior to their measurement with mass spectrometry, so that the complexity of what is introduced into the mass spectrometer is compatible with the performance of the instrument in the chosen measurement mode. This separation can be made on the proteins of the sample or on the peptides that arise from protein digestion (or on both). Among the various separation modes available for proteins and peptides, chromatography and electrophoresis are almost exclusively used at the current time. Thus the purpose of this review is to provide to the reader a critical review of the input of electrophoretic techniques in a proteomics strategy. II. The principles at playBy definition, electrophoresis consists of the separation of analytes (here peptides or proteins) as ions that are driven differently (in order to achieve separation) under an electric field. Two modes of electrophoresis are mostly used for peptide and protein separation; zone electrophoresis and isoelectric focusing. In zone electrophoresis, a speed race is made in the separation medium, usually at constant pH, and the speed of each analyte is dictated by its charge, which drives it under the electric field, and by the friction forces, which slow it down. In order to minimize the impact of diffusion, which always tends to broaden the peaks of analytes, and also to minimize the impact of the volume of the initial sample on the final resolution, a special trick, called discontinuous electrophoresis, is used (Davis, 1964), schematized on figure 1. In this system, a composite separation medium with two phases is used, (figure 1, left panel). The sample is loaded on top of the first phase, called the concentration phase, where an isotachophoresis is carried out. Isotachophoresis (same speed electrophoresis) uses the rules of transport of ions in an electric field to build what is called a moving boundary, in which the ions are ranked in the order of their mobility. Under proper conditions, all analyte ions can be included in this moving boundary, whose dimensions are dictated by the ionic strength of the medium and by the concentration of each ion (for detailed calculations, see Jovin (Jovin, 1973a, Jovin, 1973b). By sweeping this moving boundary through the sample and then through the concentration phase of the separation medium, all analytes can be concentrated in the moving boundary (figure 1, center panel). Although this produces a spectacular increase in resolution, the concentration zone reflects its name very adequately, because the analytes are extremely concentrated in the moving boundary, with the risk that they exceed their solubility and start to precipitate. But, by definition, no macroscopic separation of the analytes takes place in the concentration phase. Separation takes place in another phase, called the separation phase, in which the speed of the moving boundary is increased to exceed the mobility of the analytes, which are left behind and separated by zone electrophoresis in the trailing phase (figure1, right panel). Isoelectric focusing exploits special properties of some complex ions, which bear both weak acid and weak bases, which is the case of peptides and proteins via the amino and carboxy termini and by the side chains of some amino acids. Such complex ions have an isoelectric point, also noted pI (i.e. a pH at which the positive and negative charges are equal) so that the resulting electric charge of the analyte is zero. Below the isoelectric point (i.e. at a more-acidic pH) the analyte will be a cation, and above the isoelectric point (i.e. more-basic pH) it will be an anion.The basis of isoelectric focusing is to use a separation medium in which a smooth pH gradient that encompasses all of the pIs of the analytes of interest (figure 2A). If the acidic end of the gradient is connected to the + electrode and the basic end to the - electrode, then the analytes will be separated according to their pIs, in the sense that each analyte, wherever it is applied in the pH gradient, will migrate to its steady-state position represented by its pI (figure 2B). The method is called isoelectric focusing because each analyte is concentrated at its pI by the counteracting forces induced by diffusion (which tends to broaden the analyte zone) and by the electric field (which tend to drive back the analyte at its exact pI) (see figure 2C). More detailed explanations on isoelectric focusing can be found in dedicated books (Righetti, 1983). Opposite to the situation in chromatography, the separation in electrophoresis is not strictly dependent on the presence of a solid phase. However, the choice of electrophoresis configurations is indirectly limited by the Joule effect. The heat induced by an electric current provokes important convection flows, which limit in turn the resolution by blurring the separations produced upon electrophoresis. Thus, adequate resolutions are obtained only when convection flows are limited, and this limitation is achieved mostly by two very different setups.The most obvious setup, but not the most extensively-used one, is to use a capillary fluid column for separation. Because of the minimal cross-section, minimal electric power is used, and thus minimal heating is produced. Although such a setup could be, in principle, a real alternative to chromatographic columns for the separation of proteins or peptides, it has never gained widespread use, although there are some publications in the field. This limited use is due to a variety of reasons, among which two prominent ones are worth mentioning.First, the capillary electrophoresis separation of complete proteins has always been plagued by adsorption on the walls, which has prevented any robustness in separationSecond and more importantly, the loading achievable in capillary electrophoresis is too low to be of practical interest in proteomics studies. Indeed, proteomics studies imply that one can analyze at the same time rather abundant analytes and rather rare ones in the same sample. That requirement implies in turn that one can load rather high amounts of sample. Such loadings are achievable in column chromatography by recirculating the sample in a loading cartridge ,which is pulsed into the capillary column, but this preconcentration process is not so easy to achieve with a capillary electrophoresis setup, despite the use of stacking processes (Locke and Figeys, 2000), solid-phase extraction (Figeys et al., 1998) or isoelectric focusing and its built-in concentrating effect (Simpson and Smith, 2005, Yu et al., 2006, Wang et al., 2007). All these reasons explain why the second setup for limiting convection ( i.e. gel electrophoresis) has gained the most widespread use. As its name says, it consists of conducting the electrophoretic separation in a conductive gel; i.e,. a mixed buffer-crosslinked polymer system that has pores of a more or less defined size. The polymer chains that delimit the pores break the convection flows, and ensure adequate resolution. Moreover, they also limit the diffusion, which thereby further enhances resolution. Last but not least, the friction of the analytes on the gel material adds another separation parameter that can be used to enhance the resolution of the complete technique, especially when the size of the analyte is the separating parameter. In addition to these advantages in separation, working in a gel medium has other important advantages related to the use of solid phase chemistry, but these features are detailed later in this paper (section IV.B). However, the price to pay for the use of gel media is the loss of automation, but this loss is compensated, at least in part by the fact that gel separations are continuous separations and can be parallel ones, either handling multiple samples on one gel, or providing a continuous two-dimensional process. This latter aspect is worth discussing in more detail:When a liquid separation is carried out, and when the separated analytes must be further analyzed in another liquid (or gel) separation process, fractions must be collected. When a few analytes are separated, this fraction collection process can be data-dependent, thereby isolating each analyte peak in a limited number of fractions. However, when very complex samples are separated, as is the case in proteomics, data-dependent fraction collection becomes pointless, because it is obvious that any fraction collection scheme will split at least some peaks in two. This peak splitting means in turn that the fraction-collection process that is mandatory in liquid separations introduces artefactual discontinuities in the separations, and splits some analytes into multiple fractions. A good visual illustration of this phenomenon can be found in schemes that couple liquid chromatography to gel electrophoresis (Szponarski et al., 2004, Szponarski et al., 2007, Schluesener et al., 2007).Conversely, two-dimensional gels, whatever the separation principles used in both dimensions, represent a happy exception to this rule, because the first-dimension gel, which represents a continuous separation of the analytes, can be loaded as a whole on the top of second dimension gel, to provide a completely continuous two-dimensional separation, therefore without artefactual discontinuity. This continuity in separation, results in higher resolution and higher reproducibility of the two-dimensional gel-based separations.Of course, there are constraints on