Main

Several tried-and-tested methods can be used to detect single-stranded DNA, but there are no reliable techniques to detect specific DNA sequences in their double-stranded configuration. To meet this need, Indraneel Ghosh and David Segal at the University of Arizona put together their respective expertise in split protein assembly and zinc finger–DNA interactions.

Split versions of reporter proteins such as GFP, which are inactive until the proteins to which they are fused oligomerize, have been used to detect interactions between proteins. In their present work, described in the Journal of the American Chemical Society, Ghosh and Segal took this approach a step further by creating split protein–DNA binding domain fusions in which the two halves of the reporter reassemble only when the two proteins bind adjacent sites on DNA, thus allowing detection of specific DNA sequences.

The researchers fused two zinc fingers, the most widely used DNA-binding motif in the human genome, to GFP fragments that were known to be capable of functional reassembly. They used two well-characterized zinc fingers that recognize unique 9–base–pair sequences, and designed a double-stranded oligonucleotide with these recognition sites. In an in vitro reaction containing this template and the two purified fusion proteins, the researchers observed an increase in fluorescence, indicating the formation of a ternary complex.

This approach, called sequence-enabled reassembly, has the potential to be developed for use in many applications. “We want to make a kit that will allow us to detect, in a reagent-based fashion, any kind of viral DNA or any other double-stranded DNA,” says Ghosh. To make this a reality, however, the system has to be modified to increase signal intensity, because now the zinc fingers bind DNA tightly and only a single GFP reassembles. “So if we can get the zinc fingers to have lower affinity but the same selectivity, then they can bind DNA and rapidly turn over different GFPs or... other enzymes that we could detect,” says Ghosh, and he adds that it may even be possible to detect a single mutation within the cell. That will be possible, of course, only when the method is optimized in vivo, and will require a good signal-to-noise ratio of the reporter activity in a cell.

Ghosh hopes it will eventually be possible to use this method to selectively target cells that have a certain mutation. “What we hope to do is take this another step and actually reassemble toxins that can detect a [specific] mutation in cancerous cells, resulting in their elimination.”