Main

Transcription factors are among the most influential of proteins; they regulate gene expression that in turn determines the phenotype of cells. Harnessing this power, scientists have engineered these proteins to modulate gene expression.

The zinc finger family of transcription factors is especially amenable to modification because these proteins consist of defined domains, or zinc fingers, which specifically bind particular DNA sequences, thus selecting the genes they regulate. By shuffling three or four of these domains, a library of artificial transcription factors can easily be generated. Jin-Soo Kim from ToolGen in South Korea has a long history of working with such libraries, and he has demonstrated in eukaryotic systems that artificial transcription factors can modulate gene expression and change phenotypes. One open question, however, is how to achieve the direct identification of the target gene that causes a given phenotype. In a recent paper in the Journal of Bacteriology, Kim and his team present a method to directly link artificial transcription factor, target gene and phenotype.

To stack the cards in their favor, the scientists started with the less complex genome of bacteria rather than eukaryotic cells, expressing their zinc finger library in Escherichia coli. They then analyzed thermostable bacteria that managed to grow at 50 °C and characterized one of the artificial transcription factors that induced this desired phenotype. To find the target gene they cross-linked transcription factor and DNA in vivo, immunoprecipitated the protein and sequenced the bound DNA. By comparing their results with in silico predictions of target sites for a protein with this particular configuration of zinc fingers, they identified the target gene and thus demonstrated that these artificial transcription factors could be used to directly identify a gene responsible for a phenotype of interest.

The applications of this screen are numerous—Kim foresees many different phenotypes that could be generated with this library approach. He sees his work as bridging the gap between industry and academia. He says, “People in industry usually focus on interesting phenotypes,... in academia they look for genes [that] can induce the phenotypic change. We address the question of how to improve the phenotype and then we can identify the target genes.”

All that remains to be shown now, to get a large number of people in industry and academia really excited, is proof that this direct linking between gene and phenotype also works in eukaryotic systems, something Kim may well deliver in the near future.