Regresión no lineal por mínimos cuadrados
aplicada a ensayos de transferencia de energía por resonancia de bioluminiscencia (BRET)

Nonlinear least squares regression
applied to Bioluminescence Resonance Energy Transfer (BRET) assays

Background information about BRET:

BRET is based on energy transfer between two molecules when they are close enough. The first molecule —the luminescent donor— emits light, while the second —fluorescent acceptor— absorbs that radiation and emits another of a longer wavelength, hence fluorescence.

Examples: donor may be luciferase from the Renilla genus (Rluc), and acceptor may be the green or yellow fluorescent proteins (GFP, YFP); both donor and acceptor will be attached to the respective proteins under study.

The signal (BRET) theoretically will increase hyperbolically as a function of increasing acceptor/donor ratio, in case of any specific interaction between the donor-coupled protein and the acceptor-coupled protein. The asymptotical maximum signal (BRETmax) corresponds to saturation when all donor molecules are close to acceptors.

The BRET saturation assay has been used to compare the relative affinity of membrane receptors for each other and the probability that they form a complex. A commonly used quantitative measure of this is the midpoint in the saturation curve: 50% of the maximum BRET signal is achieved for an acceptor/donor ratio designated BRET50 (note that, despite the name, this is not a value of signal —Y axis—, but an acceptor/donor ratio —X axis—)

Values of both BRETmax and BRET50 may change when a ligand binds to one of the interacting proteins and the conformation of the dimer is modified, thus leading to either a change in the distance between the donor and acceptor or a change in the efficiency of energy transfer, or both, therefore affecting the BRET signal measured. 

See e.g. D.O. Borroto-Escuela, M. Flajolet, L.F. Agnati, P. Greengard, K. Fuxe (2013) Methods Cell Biol. 117, 141-164. doi:10.1016/B978-0-12-408143-7.00008-6

I acknowledge Prof. Rafael Franco (Universidad de Barcelona) who suggested the application of the existing nonlinear regression algorithm to BRET and provided assay information as wel as test data.