BioWorld International Correspondent
BORNHEIM, Germany Cellzome AG and MDS Proteomics Inc. scientists each published last week in Nature drafts of maps recording protein interaction pathways inside yeast cells.
Cellzome, of Heidelberg, Germany, expects its maps to 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, Cellzome’s vice president of biology, Giulio Superti-Furga, said in a prepared statement.
MDS, of Toronto, said it developed and industrialized new ways to look at human cells in action, gaining novel insights into the disease process and potentially leading to faster drug development.
Cellzome researchers in their paper were joined by scientists from the Heidelberg-based European Molecular Biology Laboratory (EMBL) and the French National Research Centers’ molecular genetics center, of Gif sur Yvette.
Cellzome intends to find new drug targets by identifying complexes of proteins. The company recovers protein complexes inside the cell by attaching tags to bait proteins of interest, purifies them by chromatography, and then analyzes them by mass chromatography. The underlying technology was developed at nearby EMBL, which spun off Cellzome in May 2000.
The map created by Cellzome and its academic partners characterizes the function and interactions of 1,440 yeast proteins comprising 232 protein complexes, which directly affect biological activity, the company said.
“These maps 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,” Superti-Furga said.
Superti-Furga told BioWorld International that information from the yeast protein map is expected to be useful for understanding human cell physiology. “We have used proteins as baits for fishing [protein complexes], which were equivalents to human proteins. The complexes we found were actually enriched [with proteins showing equivalence] with human proteins.”
Thus, proteins were found that are conserved between human and yeast cells, Superti-Furga said.
“All basic cellular functions are conserved in human and yeast, [for example] the duplication of DNA, duplication of cellular organelles, growing of cells,” he said. “Using this yeast study as a scaffold, we already have a first idea of how basic cellular functions are organized in protein complexes in human cells.”
The researchers found 231 novel proteins of unknown function. They drew conclusions on the proteins’ functions from the proteins’ neighborhoods in the complexes they are associated with. If an unknown protein is found, for example, in “part of a complex of proteins, the majority of which are known to be involved in DNA repair, then by the concept of guilt by association’ the chance is very high that the novel protein has something to do with DNA repair,” Superti-Furga said.
For the company’s internal programs, the maps are of great value as guides for scouting the human proteome, Superti-Furga said.
MDS researchers in Toronto and Odense, Denmark, worked with academic partners from Toronto-based Mount Sinai Hospital.
Their findings included more than 3,617 interactions involving nearly 1,578 different proteins fished by 725 yeast bait proteins.
MDS said it plans to share its research capability with its partners. In its internal programs MDS plans to identify 1,000 new drug targets over the next five years. As a demonstration of its platform the company completed a pilot project to a yeast model organism and successfully identified 531 proteins of unknown function, the company said.
Anuj Kumar and Michael Snyder, of Yale University’s department of molecular, cellular and developmental biology, commented on the Cellzome and MDS papers in the Jan. 10, 2002, Nature, calling the companies’ approach “clearly powerful.” But they also said it had drawbacks as both groups found a significant number of false-positive interactions, while failing to identify many known associations.
They said that the whole set of protein interactions in yeast currently is far from being known. Up to 11,000 protein interactions in yeast are reported so far, but given that each protein in yeast interacts with five others, the set of protein interactions in yeast should encompass some 30,000, many of which change during the lifetime of the organism, they wrote.