2-D protein gel electrophoresis: an old method with future potential

Abstract

The technique of two-dimensional protein electrophoresis (2-DE), with its ability to resolve several thousand pro- teins, is potentially extremely power- ful. For example, a single 2-D gel can separate all the proteins of E. coil or yeast cells and the result can be a beautiful image of the cell's protein machinery. Mammalian cells are likely to produce more proteins than a single 2-D gel can routinely resolve. However, subfractionation of cells prior to 2-D gel analysis should make it possible to identify all the proteins of individual cell types. This approach has been used to document the pro- teins present in human keratinocytes and has resulted in a 2-D gel database that contains over 4000 proteins ~ (Julio Cells, Aarhus, Denmark). No less impressive is the fact that over 1000 of these proteins have oeen identified by using a range of complementary techniques. Such results are not easy to reproduce and there are arguably three major obstacles to being able to make use of the power of 2-D gel technology. First, we need to be able to produce 2-D gels that can resolve thousands of proteins with quantitat- ive and spatial reproducibility. Next, it becomes important to be able to convert the beautiful (or perhaps not so beautiful) 2-D pattern of proteins into the Identification of the individual proteins on the gels, And finally, the vast amount of information generated by such analyses needs to be cata- logued, shared and, perhaps more importantly, Integrated with other protein and DNA sequence databases so that It can be used to good effect. These and other Issues were addressed at a recent conference on 2-DE*. Towards better 2.D gels There are two predominant appli- cations of 2-DE. A number of investi- gators use the method to undertake comprehensive analyses of changes in the production or covalent modifi. cation of proteins in disease states or in response to 'activation' of cells and tissues by a wide variety of stimuli. Others have begun the task of gener- ating 2-D gel databases in which all the proteins present on a 2.D gel of specific cells (human liver, keratin. ocytes), tissues (heart) or organisms (E. coil, yeast) are identified. Both groups depend on the ability to pro- duce well-resolved and quantitatively reproducible 2-D gels. As those who have attempted to run 2.D gels will testify, this in itself is no mean feat. It may come as no surprise to those who have tried and failed that, with one