Research in the Harlow laboratory focuses on new approaches for functional analysis in mammalian cells. Our primary interest is learning how to do high throughput and unbiased screens for genes that affect key phenotypes of cancer biology. The levels of specific proteins can be increased or decreased by expressing the protein itself from a cDNA copy or by the introduction of an inhibitory RNA for the mRNA. We use libraries of individually cloned and sequenced full length coding regions and siRNAs to raise or lower protein levels in cells and study changes in cellular phenotypes. At present we have a complete proteome for several test organisms—bacteria Pseudomonas aeruginosa, the yeast Saccharomyces cerevisiae, and libraries for several viruses. For mammalian studies we have isolated full-length cDNAs for 3000 human proteins and a nearly complete library of human. Our goal is to collect and use cDNA copies of the entire human coding potential. We anticipate that by the end of 2004, we should have over 1/2 of the human proteome available for screening studies.
Cells use signaling pathways to respond to extracellular and intracellular cues and regulate important physiological events. It is clear that many regulatory processes integrate biochemical readouts from multiple pathways to control important cellular events. In simple genetic model systems, synergism between different pathways can be identified and studied by use of mutation screens that can ask, in an unbiased way, which pathways contribute to key decision-making steps. Until recently, similar approaches have been unavailable in mammalian species, particularly in humans. The growing availability of comprehensive repositories of all coding regions and siRNAs for each gene provides a technical approach to studies the role of each protein individually.
One of our first efforts is to use the response to anti-cancer drugs as simple model for pathway interaction. Cancer drugs as established and well-studied perturbants of known components of cellular physiology. We raise kinase protein levels using expression of cloned cDNAs or lower them using siRNAs in cells treated with the anti-cancer drug, and test for changes in the IC50 of drug action. The kinases were chosen because of their well-characterized roles as biochemical switches in signaling pathways and critical involvement in cancer. All known protein kinases are being tested in this manner. The interactions between the various positive or negative kinase effects are then used to search for interactions that confer hyper-sensitivity or hyper-resistance. Components of any pathway discovered in this way are good targets for drug discovery efforts designed to identify potential new agents that augment cancer chemotherapeutics, but fundamentally they provide a foundation to examine how different pathways work together to co-regulate quantitative phenotypes. In the near future we hope to be able to screen large collections of human proteins for changes in these responses and then to expand these studies to examine other cellular phenotypes of interest.
Kennedy, B., D. Barbie, M. Classon, A. Lai, P. Branton, N. Dyson, and E. Harlow. 2000. Nuclear organization of DNA replication in primary mammalian cells. Genes and Development, 14:2855-2868.
Stevenson, L., B. Kennedy, and E. Harlow. 2001. A large scale screen for the identification of new cell cycle genes, Proc. Natl. Acad. Sci, U.S.A. , 98:3946-51.
Braun, P., H. Yanhui, S. Binghua, A. Halleck, M. Koundinya, E. Harlow, and J. LaBaer. 2002. Proteome-scale purification of human proteins from bacteria. Proc. Natl. Acad. Sci. 99:2654-2659.