FRETting for a more detailed interactome

signaling update home

SOURCE:

By combining fluorescence-activated cell sorting (FACS) and fluorescence resonance energy transfer (FRET) researchers have established a powerful high-throughput method to screen for protein—protein interactions in yeast and mammalian cells.

A quantitative high-throughput FRET-based method of screening fluorescent protein fusion libraries brings the promise of more detailed interactome maps.

Fluorescence resonance energy transfer (FRET) is a powerful method to quantitatively measure protein-protein interactions in cells, but it has not been applied to high-throughput projects, where it has the potential to provide more detailed quantitative information than traditional methods such as the yeast two-hybrid (Y2H) assay. "Despite the clear utility of library screening, it has been too challenging to try it because of the low efficiency of FRET," explains Patrick Daugherty of the University of California, Santa Barbara.

Using YPet and CyPet fluorescent proteins (Nguyen & Daugherty, 2005), which had been optimized for intracellular FRET, Daugherty and colleagues now developed a FRET-based assay for screening protein interactions in virtually all types of cells (You et al., 2006). They coexpressed a target protein-YPet fusion and a library of peptide ligands as fusions with CyPet, and then used fluorescence-activated cell sorting (FACS) to detect cells exhibiting FRET. In bacterial cells, the partly overlapping FRET signals necessitated several rounds of enrichment by FACS, but in yeast and mammalian cells, the large dynamic range in FRET signals between interacting and noninteracting pairs is sufficient for one-step screening of libraries for interaction partners. The researchers were also able to determine the interaction affinities in selected clones by measuring fluorescence intensity as a function of YPet concentration in a lysate dilution series.

One notable result was that the protein-protein interaction affinities measured in vitro for purified proteins were different from those measured in cell lysates. "This is interesting because it begs the question of whether or not in vitro data are sufficient when we develop models and maps for a systems biology perspective. Context-dependent data will probably be required," says Daugherty. He believes that protein crowding is one contributor to this discrepancy and that no existing theory for protein crowding can describe true binding affinity in a crowded environment. "The affinities that we measure in solution using typical methods for purified proteins are just one way to assess binding affinity, but they do not really reflect what is going on inside of the cell." The new FRET-based method is unique because the interaction occurs in the crowded environment of the cell—which has the additional benefit of increasing the specificity of the selection.

Although there are many advantages of using this new method over Y2H assays, including the ability to detect interactions outside the cell nucleus and in different cell types as well as the ability to quantify binding affinities, it has inherent limitations, being a FRET-based method. The fluorescent proteins must be brought to within about 6 nm of each other for FRET to occur, and thus the system is limited to studying smaller proteins or parts of proteins.

Daugherty points out that an exciting application of this method would be domain-level protein interaction screening, for example, to build a higher-resolution interaction map after a Y2H screen. The information obtained in a FRET assay—such as which domains interact and their binding affinities—could be used to assemble more detailed protein interaction maps with enough information build models about the system. But to build such a map, a huge effort would be required to examine each of the domains of a given system: first determining which proteins are potentially involved in the system using expression data, then cloning individual domains of the identified proteins and finally screening individual domains against a cDNA fragment library. "That would yield interesting data that are not showing up in the current Y2H screens," adds Daugherty. "We are contemplating whether to jump in or not."

Irene Kaganman

References

Nguyen, A.W. & Daugherty, P.S. Evolutionary optimization of fluorescent proteins for intracellular FRET. Nat. Biotechnol. 23, 355–360 (2005). | Article | PubMed |
You, X. et al. Intracellular protein interaction mapping with FRET hybrids. Proc. Natl. Acad. Sci. USA 103, 18458–18463 (2006). | Article | PubMed |