PerspectivesAre you interested in submitting a Perspective Article? Be sure to read The Science Advisory Board's Editorial Guides for Perspective Articles. Click here. Proteome Analysis - Limits and Expectations by Simone König Simone König Medical Faculty University of Muenster koenigs@uni-muenster.de Analyzing a proteome means to investigate a highly complex mixture of thousands of different proteins caught at one particular point in time and space for a good scientific reason. Depending on the project layout and the questions asked, one has to deal with the protein extracts of a specific number of cells or of tissue homogenates. While cell culture has the potential to supply enough material for exhaustive analysis, obtaining sufficient animal or human tissue may be another story. Keeping and raising animals requires facilities managed in an extremely careful way and is, therefore, expensive. Asking a human for a piece of one of their organs won't exactly meet open ears. Nevertheless, there are human brain proteome studies published. So, obtaining the object of your scientific desire is one difficulty and it brings us right to the next: standardization of sample collection. It is simply not good enough to cut a piece of tumor and put it in the freezer until you are ready to analyze it. Proteins have a life of their own and respond to environmental changes. Handling and transport steps may introduce unwanted modifications depending on the time span the sample was exposed to open air or room temperature. It has been found that proteins change when stored at –20 °C or –80 °C. Studying a sample which was in the freezer for several years will raise eyebrows. So, how do you store valuable samples? The answer lies in careful planning before the project is started and in checking the current technology. There is much knowledge readily available if one thinks of blood or sperma banks and the set-up of biobanks is increasingly funded. Nevertheless, well maintained storage and documentation is associated with cost and not necessarily accessible to the average scientist. Compromises in that area need to be added to the list of “ifs” when it comes to answering the question of how much the data is worth. The next step in the protocol involves the generation of protein extracts. Since proteins have different solubility in hydrophobic or hydrophilic buffers, the proteome is already separated at this point mainly into water-soluble/insoluble proteins. The addition of detergents or other tricks may expand the soluble proteome but the fact remains that the solubilization step is associated with information loss. Take a critical look at any laboratory procedure, and it's clear that there is no such thing as a whole proteome analysis. Now that the proteome can be handled, it needs to be visualized and gel electrophoresis is an established technique for that purpose. It creates two-dimensional maps of proteome mixtures with respect to protein isoelectric points and molecular weights. The concentration of the proteins is reflected in their staining intensity or spot size, but the dynamic range of the visualization method leaves proteins present in lower amounts invisible. The general observation is that the abundant proteins are seldom of interest and that the relevant proteins are hard to catch. The problem has been discussed by several authors. Corthals et al. (1) stated that one needs at least 109 cells to perform comprehensive proteome analysis and suspected up to 12 orders of magnitude in dynamic range. Anderson & Anderson (2) demonstrated that the two clinically useful proteins albumin and interleukin 6 differ in plasma abundance by a factor of 1010 and that a technology allowing analysis of the single molecule of interest would require a dynamic range of 1017. In addition, gel electrophoresis has its own limits. It excludes proteins of extreme properties such as the very basic, very acidic or very hydrophobic, which decreases the content of the proteome under investigation. Moreover, handling might introduce unwanted modifications. For instance, reduction and alkylation is a necessary procedure to unfold the proteins for gel electrophoresis and separate them according to their molecular weight. However, incomplete carbamidomethylation causes perl strings of isoforms to appear on the gel for a single protein diluting and overlaying the biological relevant modifications. The latter are complicated in their own right, because a protein of one predicted sequence might be present in a number of different forms due to splice variants, isoforms, mutations, or amino acid residue modifications essential for functionality. The protein form of biological importance might only be a single one of those and is often present at low percentages compared to the bulk protein with this particular sequence. Not unexpectedly, the correlation between mRNA and protein levels was shown to be insufficient to predict protein expression levels from quantitative mRNA data (3) so that one is forced to study proteins in order to get to the essence of all efforts - function. Technology platforms provide protein separation and identification techniques and teach the importance of sample pre-fractionation and sub-proteome analysis. However, in daily practice they are hardly heard. Most of the time, scientists want a general proteome scan and settle for a 2D gel of the soluble proteome for two simple reasons: cost and sample availability. As a result, housekeeping proteins and metabolic enzymes characterize many gels. Nevertheless, a general proteome scan is a valid starting point for any analysis in order to define the experimental problem. Extensive sub-proteome analysis is time consuming and also analyte-consuming and this is often underestimated. Modern analytical technologies such as mass spectrometry provide fast access to large quantities of data in proteomics, which needs to be analyzed and stored. The infrastructure in that respect requires considerable planning as well. Most importantly, detailed single-protein studies can not be replaced by high-throughput experiments for the elucidation of function Therefore, it is not surprising that proteomics is most successful in bundled efforts with a solid financial basis. References (1) Corthals, G.L., Wasinger, V.C., Hochstrasser, D. F., Sanchez, J.-C. 2000 Electrophoresis 21: 1104-1115. (2) Anderson, N.L. and Anderson N.G. 2002 Molecular & Cellular Proteomics 1.11: 845-867. (3) Gygi, S.P., Rochon, Y., Franza, B.R., and Aebersold, R. 1999 Molecular and Cellular Biology 19, 3: 1720-1730. ### << Previous Next >> [ View All Perspectives ] |
|