Current Proteomics, 2005, 2, 1-13 1
1570-1646/05 $50.00+.00 ©2005 Bentham Science Publishers Ltd.
Drug Discovery Using Yeast as a Model System: A Functional Genomic
and Proteomic View
Daniel Auerbach
1
, Anthony Arnoldo
2
, Boris Bogdan
2
, Michael Fetchko
2
and Igor Stagljar
2,
*
1
Dualsystems Biotech Inc., Winterthurerstrasse 190, CH-8057 Zurich, Switzerland and
2
Institute of Veterinary
Biochemistry and Molecular Biology, University of Zurich-Irchel, Winterthurerstrasse, 190, CH-8057 Zurich,
Switzerland
Abstract: Drug discovery is a complex process that includes the identification of biological targets as well as the
identification of leads that aim at altering or inhibiting the function of a particular target. The budding yeast
Saccharomyces cerevisiae has long been recognized as a valuable model organism for studies of eukaryotic cells since
many of the basic cellular processes between yeast and humans are highly conserved. In this review, we highlight
emerging yeast-based functional genomic and proteomic technologies that are advancing the utility of yeast as a model
organism in the drug-discovery process. These approaches include the utilization of yeast deletion strain collection,
synthetic genetic array combined with chemical genomics, variations of the yeast two-hybrid system, yeast biosensor
assay, and protein microarrays. Although still at an early stage, these technologies show promise as novel and useful
methods for development of target-specific therapeutic approaches.
Key Words: Drug discovery, yeast two hybrid system, genomics, proteomics.
INTRODUCTION
Drug discovery is a lengthy and costly process which
involves target identification and validation, drug screening
and safety assays, development of animal models and
ultimately, testing of potential drug candidates in clinical
trials. The development of a novel drug may take 10-15
years, with cost estimates of ~800 million US$ (DiMasi et
al., 2003). Any technology that is able to increase the speed
of the screening process and deliver more reliable predic-
tions on the efficacy or potential toxicity of drug candidates,
and thereby effectively help to cut cost and time in this
process, would be helpful. Both genomics and proteomics
have been considered to be effective for such purpose.
Whereas genomics can speed up the target identification
process by making complete genome sequences of a wide
range of organisms available (http://www.genomenews
network.org/resources/sequenced _ genomes / genome _ guide _
p1.shtml), proteomics holds the promise of identifying the
complete set of proteins expressed in a particular cell at a
given stage, as well as uncovering the myriad of regulatory
networks that are created by protein-protein interactions.
Ultimately, using systems biology one should be able to
receive a “snap-shot” of a cell with all information regarding
expressed proteins, their particular localization within the
cell, their post-translational modification status and their
interactions with various other proteins expressed at the
same time.
To date, the major impact of proteomics has been on
processes in the early stages of drug development, namely
target identification and validation. In particular, the use of
*Address correspondence to this author at the Institute of Veterinary
Biochemistry and Molecular Biology, University of Zurich-Irchel,
Winterthurerstrasse, 190, CH-8057 Zurich, Switzerland; Tel: +41-1-635 54
74; Fax: +41-1-635 68 40; Email: stagljar@vetbio.unizh.ch
model organisms such as the bakers yeast Saccharomyces
cerevisiae, the fruit fly Drosophila melanogaster and the
worm Caenorhabditis elegans has been very helpful since
many human genes have orthologues in lower eukaryotes
(consortium, 2001; Venter et al., 2001). In this review, we
focus on the use of the model organism S. cerevisiae and its
applications in finding and characterizing disease pathways,
as well as on attempts at converting yeast to an in vivo
screening platform to identify novel compounds in the early
stage of drug development. These include the use of
engineered yeast strains in assessing the toxicity of drug
candidates, synthetic genetic array combined with chemical
genomics, the establishment of large scale protein interac-
tions maps from several organisms and attempts to use these
maps to predict pathways implicated with human diseases,
the direct use of yeast two-hybrid strategies to identify novel
drug candidates that alter or inhibit protein-protein interac-
tions, and the application of protein microarrays for drug
discovery. The reader is referred to several excellent recent
reviews covering additional yeast based functional genomics
technologies that are not covered here (Hughes, 2002; Melese
and Hieter, 2002; Parsons et al., 2004; Tucker, 2002).
YEAST AS A MODEL SYSTEM FOR DRUG
DISCOVERY
The unicellular eukaryote S. cerevisiae is an excellent
model system for drug discovery: it is inexpensive to
maintain and grow, it is classified as a GRAS (generally
recognized as safe) microorganism, its entire genome has
been sequenced (Goffeau et al., 1996), its ~ 6, 200 open
reading frames (ORFs) exist in a readily usable form
(Hudson et al ., 1997), it is well suited for the expression of
heterologous proteins, and it contains a multitude of
selective markers, including markers for nutritional selection
(e.g. HIS3, URA3) , drug resistance (e.g. kanMX, patMX,