Proteomics: An Infinite Problem with Infinite Potential

Proteomics is an exciting discipline in its infancy that means different things to different people. It is also a field limited by the technologies currently available to its practitioners. Important questions arise: How can we usefully define proteomics? Which research questions offer the greatest promise for the field? What new tools will be needed to pursue those questions? Perhaps the most appropriate definition of proteomics is "any large-scale or systematic characterization of the proteins present in a cell, tissue, or organism." It is a different paradigm from conventional reductionistic scientific investigations that typically focus on a single gene or protein. Proteome research asks unbiased biological questions such as what protein levels or activities change between two experimental conditions. Although sequencing the human genome was initially daunting, it was a finite problem with a specific endpoint. In contrast, the proteome of any organism is highly dynamic with an endless number of possible variations. Even the proteomes of prokaryotes and the single-cell eukaryote yeast are essentially infinite because, in addition to changes during normal development, growth, and cell division, their proteomes respond in biologically important ways to environmental changes. Hence, there is no discrete proteome, per se, even for simple single-cell organisms. The Complexity of It All It is readily apparent that the scope of the problem for multicellular organisms, especially humans, is far more complex, with a virtually boundless number of possible proteomes. For example, current estimates are that the human genome has about 50,000 genes. Many of those genes are alternatively spliced, and most proteins are post-translationally modified, frequently at multiple sites and with several moieties. At least one million different protein components that are functionally distinct are likely to be produced by the human genome. For most of us engaged in proteome research, the most important goal is not simply to catalog protein components and define their macromolecular interactions under static conditions. Instead, the more interesting questions include understanding how changes in proteomes contribute to biological function or disease states, and to define components of cellular machines, pathways, and networks. The scientific literature from the past several years amply illustrates the unprecedented insights that proteome approaches are already providing into biology and human diseases. A few examples that illustrate the diverse utility of proteomic studies are analysis of the composition, architecture and transport mechanism of the nuclear pore complex;1 systematic analyses of yeast protein complexes;2,3 identification of putative ovarian cancer serological markers using a novel bioinformatics approach;4 identification of breast cancer markers;5 and identification of lung disease markers in bronchoalveolar lavage fluid.6 So, given the multi-dimensionality, complexity, and infinite variations of proteomes, together with limitations of current technologies, how can we best realize the great potential of proteomics in biology and medicine? In particular, how should available funding be distributed among the competing needs of purchasing currently available proteomics instrumentation, developing new and improved proteome technologies, and applying current proteomic tools to biological and biomedical problems? The Need for Adequate Funding Most proteome approaches require access to expensive instrumentation and software. At the same time, far better methods are urgently needed to realize the full potential of proteomics, and new tools are rapidly evolving. But currently available technologies are already capable of critically important biological and clinical insights. It is therefore critical that all three areas be strongly supported, probably at roughly equal levels for the near term. Later, as technologies mature, a larger proportion of funding could then be shifted to pursuing proteome analyses. Most existing funding mechanisms, however, inadequately address all three areas of funding needs. The only major instrument funding mechanism, the National Institutes of Health shared instrumentation program, is dramatically underfunded and only covers single instruments that will be shared by multiple users. Similarly, a substantial portion of proteomics research applications does not readily fit into the current RO1 or PO1 grant mechanisms at the NIH. And few funding mechanisms exist for technology development. It is encouraging that a few funding institutions have recently initiated specific funding mechanisms for technology development such as the National Cancer Institute's Innovative Molecular Analysis Technologies program and a recent proteome centers initiative by the National Heart, Lung, and Blood Institute. But expanded use of funding mechanisms specifically targeted to proteomics technology development, purchase of instrumentation, and pursuit of biological studies is needed to fully realize the great potential of proteomics for providing insights into biological processes from an unbiased, global or systems perspective. Finally, it is interesting to speculate concerning which technologies will dominate during the next decade. Certainly mass spectrometry plays such an important role for multiple proteome approaches that it will almost certainly continue to be a critical major technology. And protein arrays including antibodies or antibody mimics and expressed proteins with affinity tags can be expected to play expanded roles. One technology that most laboratories, including ours, would like to replace is two-dimensional (2-D) gels. This older technology has a number of limitations, including variability and poor detection of certain groups of proteins, and it is difficult to automate. However, unlike nucleic acids which generally behave similarly and are readily analyzed by a single technique, proteins are highly diverse and complete characterization of proteomes requires detection of many types of protein changes. For this reason, I predict proteome analysis will continue to require use of a number of different analytical tools. One or two methods will not come to dominate the field. It is likely that 2-D gels will remain as one of several tools that will sometimes be integrated into strategies for comprehensive quantitative comparisons of protein profiles. This view is in part based upon the observation that some protein changes can be easily detected by 2-D gels but would be difficult to detect by non-gel methods and vise versa. In summary, we have unprecedented opportunities to unlock the mysteries of biological processes and to develop new diagnostics and therapeutics for human diseases using global and targeted proteomic approaches. As a child, I watched "Star Trek" and often regretted that I was born too soon and would not have the opportunity to explore the galaxy at warp 8. But as a protein biochemist, I find that the first decade of the 21st century is a great time to be pursuing biomedical research due to the exciting and virtually unlimited potential of proteomics. David W. Speicher, Ph.D., is a professor and chair of the Structural Biology Program at The Wistar Institute, Philadelphia, Pa. References 1. M.P. Rout et al., "The yeast nuclear pore complex: composition, architecture, and transport mechanism," Journal of Cell Biology, 148:635-51, 2000. 2. A.C. Gavin et al., "Functional organization of the yeast proteome by systematic analysis of protein complexes," Nature, 415:141-7, Jan. 10, 2002. 3. Y. Ho et al., "Systematic identification of protein complexes in Saccharomyces cerevisiase by mass spectrometry," Nature, 415:180-3, Jan.10, 2002. 4. E.F. Petricoin et al., "Use of proteomic patterns in serum to identify ovarian cancer," Lancet, 359:572-7, Feb. 16, 2002. 5. H. Hondermarck et al., "Proteomics of breast cancer for marker discovery and signal pathway profiling," Proteomics, 1:1216-32, Oct. 1, 2001. 6. R. Wattiez et al., "Human bronchoalveolar lavage fluid protein two-dimensional database: study of interstitial lung diseases," Electrophoresis, 21:2703-12, 2000.

Proteomics: An Infinite Problem with Infinite Potential

Interested in reading more?

Become a Member of