Bioluminescence resonance energy transfer to detect protein-protein interactions in live cells

Abstract

Bioluminescence resonance energy transfer (BRET) is a valuable tool to detect protein-protein interactions. BRET utilizes bioluminescent and fluorescent protein tags with compatible emission and excitation properties, making it possible to examine resonance energy transfer when the tags are in close proximity (<10 nm) as a typical result of protein-protein interactions. Here we describe a protocol for detecting BRET from two known protein binding partners (Gαi1 and RGS14) in HEK 293 cells using Renilla luciferase and yellow fluorescent protein tags. We discuss the calculation of the acceptor/donor ratio as well as net BRET and demonstrate that BRET can be used as a platform to investigate the regulation of protein-protein interactions in live cells in real time.

Fig. 1

BRET is dependent on the distance between the donor luciferase and the acceptor fluorophore. Addition of the cell permeant Renilla luciferase substrate coelenterazine (ctz) results in oxidation of the substrate to coelenteramide, which produces blue light at 482 nm. When protein-protein interactions between Protein X and Protein Y bring the donor luciferase (RLuc) and acceptor fluorophore (YFP) in close proximity (<10 nm), the energy from the donor can be transferred to the acceptor and light is produced at 527 nm. When the BRET tags are not in close enough proximity, light is only emitted at 482 nm

Fig. 2

The energy transfer between BRET pairs depends on the overlap of the donor emission spectrum with the excitation spectrum of the acceptor. For Renilla luciferase, oxidation of coelenterazine results in an emission peak at 482 nm. This emission overlaps well with the excitation spectrum of yellow fluorescent protein (excitation peak: 514 nm). The resulting energy transfer yields yellow light with an emission peak of 527 nm

Fig. 3

HEK 293 cells were transfected with increasing amounts of Gαi1-YFP (0, 10, 50, 100, 250, and 500 ng) and either 5 ng RGS14-WT-Luc or RGS14-Q515A/R516A-Luc. Wild type RGS14 shows a robust BRET signal with Gαi1. Conversely, the RGS14 mutant (Q515A/R516A) that can no longer bind Gαi1 shows a drastically reduced maximal BRET signal indicating a disruption in the protein-protein interaction. The above data is representative of three independent experiments. Curves were generated with GraphPad Prism 5 using the one-site binding curve fitting function. Additionally, Gαi1-YFP expression levels were verified by immunoblot analysis