Target Discovery and Validation Reviews and Protocols

2 Sioud
disease pathogenesis and perpetuation. Although the human project has made
available as many as 5000 potential new targets for drug intervention (http://
function.gnf.org/druggable) (1), yet the main challenge facing the pharmaceutical
industry is still to identify and validate novel disease-causing genes and to uncover
additional roles for genes of known functions. Molecular profiling, which includes
genomics and proteomics, has provided powerful tools with which to analyze gene
expression in diseased and normal cells and tissues (2–8). An obvious example is
mRNA expression profiling by using DNA microarray for large-scale analysis of
cellular transcripts by comparing expression levels of mRNAs. In all cases, gene
expression has been analyzed using clustering algorithms that group genes and
samples on the basis of expression profiles. Because the major challenge of
genomics studies is the detection of significant changes in gene expression levels,
the use of adequate controls and statistical tests is crucial for these technologies.
Based on the experimental design, appropriate algorithms should be used.
Microarray technology has been applied to the study of breast cancer to isolate
tumor-specific genes as markers for diagnosis and prognosis, monitor changes in
gene expression during chemotherapy to set predictors for outcome, classify ductal
tumors based on epithelial cell subtypes, and identify candidate molecular targets
(2,3). Also, the technology enabled the discovery of novel functional
pathways, prognostic and therapeutic targets in various cancer types. In addition to
their value for analyzing patient samples, gene profiling techniques also can provide
important insight into disease when applied to experimental models that can
be genetically or pharmacologically manipulated. Chapters 4 and 5, Volume 1,
review these technologies and their applications to the study of breast and pancreatic
cancers, providing a framework for appreciating the promise, while discussing
the challenges ahead in translating genomics into real diagnostic, prognostic, or
therapeutic products.
Although microarray technology has been successfully used to identify candidate
targets, it is limited to detection of whatever is spotted on a slide, making it a
“closed” approach for gene discovery. In contrast, other systems-based technologies
such as differential display, representational difference analysis (RDA), and
serial analysis of gene expression (SAGE) are capable of detecting both known
and novel genes for any sample (5,9) (see Chapter 6, Volume 1). By screening for
genes associated with high- but not low-grade human non-Hodgkin’s B-cell lymphoma
by RDA, we have identified a novel centrosome/microtubule-associated
coiled-coil protein (CSPP) involved in cell cycle progression and spindle organization
(10). Overexpression of CSPP is observed in human diffuse large B-cell
lymphomas and correlated to shorter survival (10). As illustrated in Fig. 1, overexpression
of green fluorescent protein-tagged CSPP in human embryonic kidney
(HEK)293 protein inhibited bipolar spindle formation and induced aberrant
mitotic spindle structures such as the depicted monopolar spindle.