BioWorld International Correspondent

LONDON - The first analysis of all the proteins made by a cell, and the ways in which they interact with each other, has just become available. The researchers in Germany who carried out the study liken their achievement to the first time that scientists sequenced the genome of an entire organism.

The analysis of the proteome of the yeast cell, Saccharomyces cerevisiae, showed that proteins associate with each other to form "molecular machines." Those complexes often share proteins with each other, and can change their functions by locating to a different site in the cell.

About 500 of the molecular machines are present in S. cerevisiae, the study showed, but the researchers predicted that human cells probably have about 3,000 such machines, give or take 500.

Scientists at Cellzome AG, a biotechnology company based in Heidelberg, Germany, and at the European Molecular Biology Laboratory (EMBL), also in Heidelberg, carried out the research, which was published in Nature, available online Jan. 22, "Proteome Survey Reveals Modularity of the Yeast Cell Machinery."

Cellzome already has started to apply the technology to human cells. Gitte Neubauer, vice president at Cellzome, told BioWorld International that the group published data on the tumor necrosis factor alpha pathway in 2004.

Cellzome also has a collaboration with the Novartis Institute of Biomedical Research Inc., of Boston, working on mapping several different disease pathways. A second collaboration with Johnson & Johnson PRD, of Springhouse, Pa., already has identified several novel proteins involved in processing amyloid precursor protein, which is known to play a role in Alzheimer's disease.

Giulio Superti-Furga, who launched the project to analyze the proteins of yeast at Cellzome four years ago, but who now works at the Centre for Molecular Medicine of the Austrian Academy of Sciences in Vienna, told BioWorld International: "This work is ground breaking because this is the first time that the entire molecular machinery of a proteome has been examined in any type of cell. Furthermore, for each and every protein, we have described its propensity to associate with each and every other protein."

The study found, he said, that most proteins in the yeast cell bind to other proteins. "So the functional unit of biological material is not the single gene product but rather groups of proteins, or protein complexes," he added.

Superti-Furga likened the tendency of proteins to be shared between different "molecular machines" to the roles played by individual people in human society. "Individuals are also participants in several teams or functional units," he said. "Someone may belong to his or her family, but also to a soccer team, an orchestra and a group at work, and has a different propensity to associate with each of them."

Anne-Claude Gavin, former director of molecular and cell biology at Cellzome, and currently a team leader at EMBL, said the information available to scientists before this study was complete - the list of genes in the yeast genome - had been like having the phone book for an organization, but not knowing who worked in which department. "Now we know how everything is organized in functional units," she said.

In order to make that leap, the Cellzome/EMBL team had combined several methods of investigating the entire "protein household" of yeast. They used a method of extracting complete protein complexes from cells (called tandem affinity purification, which was developed in 2001 at EMBL), mass spectrometry and bioinformatics.

That approach allowed them to identify 257 molecular machines that had never been observed before, as well as new components of nearly every complex already known. The team found that cells dismantle and reassemble their machines at different stages of the cell cycle, and in response to environmental challenges, such as infections, as needed.

Gavin said, "We found that cells prefabricate core elements of machines and then synthesize additional, snap-on molecules that give each machine a precise function."

If the cell needs to respond quickly, such as in a disease, it may only need to produce a few parts to switch on the machine. Conversely, if it wants to shut down a process, it may need to block the production of only a few molecules.

Superti-Furga said the breakthrough had important consequences for those working in drug discovery. "You have to understand molecular organization in detail in order to be informed about a pharmacological intervention, to know whether you want to target the core of the machine, or one of its molecular variants," he said.