A Boost For Biosensors

FRET (fluorescence resonance energy transfer) biosensing is a popular technique for monitoring the activity of kinases and other signaling molecules in living cells. Researchers have now developed a revised form of the technique that makes it possible to use it to identify small-molecule kinase-inhibiting and -activating agents in an efficient, convenient, and inexpensive high-throughput screening format. Potential applications of the new approach include drug discovery and mechanistic studies.

FRET biosensors were developed a few years ago to allow visualization of the activities of specific kinase enzymes and other signaling molecules within living cells at high spatial and time resolution. In a kinase FRET biosensor, two fluorescent proteins are combined into a fusion protein, which also includes an intervening peptide substrate for a specific kinase. The fusion protein is produced by expression in cells.

Kinase phosphorylation of the peptide substrate causes it to either change its conformation or interact with an adjacent domain in a way that changes the fusion protein's overall conformation. In both cases, the distance or relative orientation of the two fluorescent domains changes, and the resulting fluorescence signal change can be detected. The bottom line is that a FRET biosensor generates a signal when the activity of a specific kinase in a cell increases or decreases in response to a variety of cellular and environmental signals.

FRET biosensor studies are typically carried out in a one-at-a-time mode by analyzing fluorescence changes in single modified cells under a microscope. High-throughput imaging techniques have been developed to carry out such studies more rapidly, but they require specialized equipment and are expensive.

Johns Hopkins University assistant professor of pharmacology and molecular sciences Jin Zhang and coworkers have now developed a way to run FRET biosensor assays on small molecules by high-throughput screening on convenient and inexpensive microtiter plates (ACS Chem. Biol. 2006, 1, 371). They achieved this by developing improved sensors that double the sensitivity of the assays.

Asked to comment on the work, assistant professor of pharmacology Benjamin E. Turk of Yale University School of Medicine, explains that "the problem with FRET biosensors, for kinases in particular, is that they are not incredibly sensitive; that is, phosphorylation of the reporter provides a very modest change in signal. For studies where you're continuously monitoring cells by microscopy, this is not a problem because you can detect small changes in fluorescence intensity, but this is really not practical for small-molecule screening."

What the Zhang group has done, he says, "is put substantial effort into reengineering these FRET biosensors so they provide a much more robust change in signal, to the point where activity can be monitored in a multiwell plate reader. As far as I know, this is the first time anyone has managed to do this with a kinase biosensor."

Zhang and coworkers demonstrated the technique by using it to analyze 160 small molecules from the Johns Hopkins clinical compound library for their effects on protein kinase A (PKA) and a cell-signaling pathway in which PKA participates, the G- protein-coupled receptor/cyclic adenosine monophosphate (cAMP)/PKA pathway. Among the compounds, they found three activating agents and a couple of antagonists of PKA and the pathway.

One of the antagonists, surprisingly, was the endogenous hemoglobin-breakdown product bilirubin, which accumulates excessively in patients with jaundice. "We found that bilirubin can inhibit cyclic AMP production," Zhang says, a finding "that may provide further insights into the mechanism of bilirubin toxicity." She notes that her group is currently analyzing about 3,000 other compounds from the library "to see if there are any other interesting hits."

The high-throughput format for FRET biosensing could provide advantages over existing cell-free methods for identifying kinase inhibitors and activators, Zhang says. "Compared with in vitro assays, living cells are reaction vessels with targets of interest, cofactors, and regulators present at endogenous levels in their natural cellular environment." Thus they permit effects on signaling pathways to be monitored in subcellular locations and as a function of time. That kind of information could aid the discovery of compounds with unique mechanisms of action, such as agents that specifically activate or inhibit kinases localized at subcellular sites.

The high-throughput format could also permit RNA interference (RNAi), a technique for reducing the expression of specific genes, to be used for new sets of mechanistic studies. "We imagine that instead of adding different compounds to individual wells, we could add different RNAi constructs for functional genomics studies," Zhang says. An RNAi library could be used to inhibit specific genes, making it possible to study the influence on PKA activity of proteins expressed by those genes.

Turk notes that "there's been a lot of excitement in the signal transduction field about the sort of fluorescent reporters described in this paper, because they allow you to look at spatially restricted pools of particular signaling molecules-cAMP and cAMP-dependent PKA, in the case of Jin's work. Using these tools for small-molecule screening is a logical application, and I think it's fair to say that this sort of work has been anticipated for some time."