Genomics

Identifying Drug Candidates through Gene Expression Screening

Todd Golub, MD 

A decade ago, Dana-Farber pediatric oncologist and researcher Todd Golub, MD, pioneered the use of DNA microarrays to classify cancers based on their gene expression signatures, a technique used worldwide today. Recently, with colleague Kimberly Stegmaier, MD, a former postdoc in his lab, Golub finessed this technique into an unconventional drug discovery tool. Known as gene expression-based high-throughput screening (GE-HTS), this approach makes ingenious use of genomics to identify anti-tumor compounds even when their molecular targets are unknown or considered un-druggable.

In GE-HTS, investigators first define the gene expression signature of a desired biological state, such as differentiation, and then screen a library of small molecules for compounds that induce this target signature and corresponding phenotype.

In collaboration with the Broad Institute, where Golub is director of the cancer program, Stegmaier and colleagues demonstrated the feasibility of GE-HTS in a landmark study of acute myeloid leukemia (AML), a cancer in which white blood cells fail to mature.

Kimberly Stegmaier, MD 

After comparing gene expression profiles of undifferentiated AML cells with their mature myeloid counterparts, the investigators selected a core signature of five genes that collectively acted as a surrogate for the differentiation phenotype. "The tricky part," says Stegmaier, "was figuring out how to quantify the expression of this collection of genes affordably in 384 wells," the format of small-molecule libraries. Of the technologies available at the time, Stegmaier and Golub chose mass spectrometry, building on an assay used by Levi Garraway, MD, PhD, to measure mRNA levels after cells were exposed to 1,700 small molecules provided by the Broad. This work led to identification of candidate compounds that reliably reproduced the target signature. Based upon the results of this screen, Dana-Farber investigators are now collaborating with industry partners to test a new drug for patients with AML.

With recent enhancements to bead-based Luminex technology, investigators gain the capacity to analyze expression of up to 500 genes simultaneously, increasing both the sensitivity and specificity of screening and enabling the study of increasingly intricate gene expression patterns induced by small molecules, says Stegmaier, of the Department of Pediatric Oncology. "This opens up new possibilities in predicting synergy between compounds, screening for very complex biological signatures, and analyzing multiple phenotypes in a single screen."