Scientists the world over celebrated the millennial 2001 A.D. as the year of the human genome. Now 2002 is following on fast as the year of the proteome. A proteome is an organism’s total protein set.
This week’s issue of Nature, dated Jan. 10, 2002, reports two of the first attempts to map the ways in which proteins work together in the yeast cell. In many useful ways, yeast cells model the behavior of human cells.
Two major microbiology companies, one European, the other Canadian, share Nature’s joint coverage.
Cellzome GmbH in Heidelberg, Germany, fielded 39 co-authors reporting: “Functional organization of the yeast proteome by systematic analysis of protein complexes.” Its senior author is Cellzome’s vice president, biology, Giullio Superti-Furga. The company was spun off in May 2000 from EMBO — the European Molecular Biology Laboratory — to isolate protein assemblies directly from cells under near-physiological conditions. It included pioneers in the fields of protein mass spectrometry, bioinformatics, structural biology and signal transduction.
MDS Proteomics Inc., a subsidiary of MDS Inc., both based in Toronto, weighs in (47 co-authors) with a paper titled, “Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry.”
Cellzome’s paper presents the first draft of a functional map of the yeast proteome, which visualizes an entire network of protein complexes and their interaction in S. cerevisiae. “These maps,” Superti-Furga said, “will enable researchers to more fully assess the roles of individual proteins in biology, and provide for a more comprehensive approach in choosing targets for drug discovery. By knowing the molecular context in which targets act,” he added, “investigators can better predict the effect of drug candidates on the parameters that influence safety and efficacy. This information can be used to improve the efficiency of the drug discovery process.”
Cut To The Chase For Drug Payoffs
In the same vein, MDS Proteomics’ president and CEO, Frank Gleeson (a co-author), announced that his corporate team has “developed and industrialized new ways to look at human cells in action, gaining novel insights into disease processes, and leading to faster drug development. This research capability will be shared with MDS Proteomics’ partners and used internally by the company to identify 1,000 new drug targets over the next five years. In practical terms,” he pointed out, “this promises to help accelerate the drug discovery process, reduce failure rates of drugs in clinical development and lead to improvements in the productivity of the pharmaceutical industry.
“Determining protein function is essential to mastering human biology and treating illness, as proteins do most of the work in the body,” Gleeson said. “Even with the knowledge of the human genome, the structure and activity of an entire cell remain hidden, providing neither a comprehensive understanding of the cause of a disease, nor identifying specific targets for treatment.”
To which the firm’s chief scientific officer, Michael Moran (also a co-author), added, “To stop the spread of a disease like cancer, the key is to identify, understand and manipulate the interactions among proteins in those pathways involved in that disease process.” Beginning with 10 percent of predicted yeast protein as baits, MDS detected 3,617 associated proteins covering 25 percent of the yeast proteome.
Both Nature entries rely on snaring individual protein tags of relevance, and baiting them with DNA to fish for, and catch, multiprotein complexes. In an accompanying “News & Views” commentary, developmental biologist Michael Snyder at Yale University in New Haven, Conn., paraphrased John Donne’s sonnet: “no protein is an island entire of itself,” but hedged, “at least, very few are.” The two studies, Snyder observed, “exemplify an emerging paradigm in protein biology: the systematic analysis of an organism’s complete complement of proteins — its proteome.” Snyder’s own group at Yale “has developed a microarray technology in which purified, active proteins from almost the entire yeast proteome are printed onto a microscope slide at high density, such that thousands of protein interactions can be assayed simultaneously.”
Cellzome processed 1,739 genes, including 1,143 human orthologues of relevance to human biology, and purified 589 protein assemblies. Its analysis defined 232 distinct multiprotein complexes, and proposed new cellular roles for 344 proteins, including 231 with no previous functional assignment. Comparison of yeast and human complexes showed that conservation across species extends from single proteins to their molecular environment.
“The study provides a wealth of new information on 231 previously uncharacterized yeast proteins, and on a further 113 proteins to which the authors attach a previously unknown cellular role,” Snyder commented in Nature. Underlining yeast’s counterparts in more complex species, the researchers analyzed three equivalent multiproteins from yeast and human cells that share comparable functions.
Free Public Access To Firms’ Proteomic Data
MDS co-authors constructed an initial bait set of 725 yeast proteins, from which they identified 3,617 interactions involving 1,578 different proteins.
Cellzome’s Superti-Furga made the point that, “After decades of deconstructing biological complexity through the process of creating catalogues of molecular biological information, the new aim is to reconstruct and integrate the information so as to uncover a real depiction of cultural behavior. The yeast genome,” he continued, “was sequenced in 1996, yet as many as half its genes still await functional assignment. Our goal was to decipher the functional architecture of the proteosome so that gene function could be interpreted within the molecular environment of the corresponding gene product, and that the cellular processes can be understood in terms of the concerted action on molecular machines.”
While granting the MDS and Cellzome approaches as “clearly powerful,” Snyder’s commentary pointed out “a significant number of false-positive interactions.” He cited Cellzome’s “estimate that 30 percent of the interactions they detect may be spurious.” As for MDS, “that group did not detect nucleotide excision repair factor-2, a tight complex. So as in most large-scale studies, these results are imperfect.” He summed up by observing that “the yeast proteome should encompass some 30,000 protein interactions. So far, existing reports have collectively identified at most 11,000 different protein associations. Although feasible, the characterization of all remaining interactions will almost certainly be labor-intensive. But,” he concluded, “the resulting data will be more than worth the effort.”