The bleeding edge of technology.

In any given proteomics experiment, a cell or tissue can express hundreds or thousands of proteins at a time. Unlike the static genome, the proteome can change quickly, and key responses to toxicants and disease may involve small amounts of rare proteins. As an added complication, gene-protein and protein-protein interactions often are not linear. So what's a researcher to do? At the seminar "Proteomics and Systems Biology," held at the 2004 annual meeting of the American Association for the Advancement of Science in Seattle, Washington, presenters discussed advances at the "bleeding edge" of proteomics research and their use in the study of complex biochemical interactions within and among cells--advances that may help overcome some of the challenges posed by proteomics.

Seminar presenters discussed techniques to both measure protein signals and analyze the enormous amounts of data that result from such experiments. In most of the experiments reported, new techniques were tested on biological systems that had already been partially characterized, such as blood plasma. This allowed researchers to validate their systems with the bonus of potentially adding to knowledge about the biological systems in question.

In one analysis of blood plasma conducted at Pacific Northwest National Laboratory (PNNL), scientists claim to have identified about 3,700 proteins (not counting immunoglobulins) from human plasma, results that are an order of magnitude greater than those described only 18 months ago, according to Richard Smith, director of the NIH Proteomics Research Resource Center at PNNL. The newly detected proteins include many found at very low levels, some of which could be used as biomarkers of toxic exposure or disease progression, said Smith.

The plasma analysis used high-sensitivity, high-throughput instrumentation and techniques developed at PNNL, including Fourier transform ion cyclotron resonance (an advanced form of mass spectrometry). The PNNL researchers also separated out the most abundant proteins, allowing measurements to focus on less abundant proteins and increasing the number of proteins found in their plasma samples from about 1,000 to about 3,700. In addition, using electrospray ionization and low flow rates of solutions into the mass spectrometer facilitated detection of proteins in amounts as small as 10 zeptomoles, a level of sensitivity that makes it possible to analyze many proteins expressed by a single cell, said Smith.

Leroy Hood, president of the nonprofit Institute for Systems Biology (ISB), too, discussed the potential for analyzing single cells. Researchers at the ISB and the California Institute of Technology are developing nanochips measuring 100 microns on a side that will assess the behavior of individual cells and gauge the concentrations of the mRNAs and proteins from a single cell. The ISB researchers have applied microfluidics--the study of how fluids behave at the nano level--to successfully conduct biological assays on single cells. ISB and Caltech researchers are now working on nanochips that can analyze several cells simultaneously. "We'll be able to interrogate a T cell and then an antigen-presenting cell separately. Then we will let the cells interact and interrogate their combined behaviors," said Hood.

Some presenters noted cautions about the prospects for analysis of single, living cells. Smith, for example, said that much more progress is needed in areas such as the construction of nanochips and microfluidics in order to make these methods of measurement "truly useful and not just a stunt."

Presenter Matthias Mann, a professor of biochemistry and molecular biology at the University of Southern Denmark, emphasized the need to distinguish between at least three states to track changes in protein expression. He displayed preliminary data from cells that were treated with a growth factor and sampled at five points. Analysis detected changes in levels of about 400 phosphoproteins over time. "The activation of some proteins decayed faster, and some proteins were activated later," Mann said.

Beyond the proteome is the metabolome--the sugars, amino acids, and other molecules that are created by or combine to create proteins. Masaru Tomita and colleagues at Keio University and the bioventure firm Human Metabolome Technologies are combining capillary electrophoresis and mass spectrometry to analyze the metabolites from rice, Escherichia coli, and Bacillus subtilis. For the purposes of these experiments, the team defined metabolites as molecules with a molecular weight of less than 1,000. In one experiment, the team detected more than 1,700 possible metabolites. The team is now constructing a model of entire metabolic pathways with several thousand reactions.

It may never be possible to fully describe the proteome of any species. Unlike the genome, which Hood characterized as "digital" and so "ultimately knowable," the proteome is affected by a multitude of external factors. In its absolute sense, according to Mann, the proteome is "as unreachable as the horizon."