Stephanie Q. Hutsell, Adam J. Kimple, David P. Siderovski, University of North Carolina at Chapel Hill, 120 Mason Farm Road, CB#7365, Chapel Hill, NC 27599-7365, 09/2009
Many extracellular signals, including hormones, neurotransmitters, and growth factors, are relayed intracellularly by binding the extracellular portion of a seven-transmembrane domain (7TM) receptor that is coupled to an intracellular heterotrimeric G protein. The stimulated G-protein coupled receptor (GPCR) propagates the extracellular signal through activation of the alpha subunit of heterotrimeric G proteins (Gα), which occurs when Gα exchanges GDP for GTP. The Gα subunit is also modulated by a family of proteins containing the GoLoco motif, which serves to inhibit the exchange of GDP for GTP by the Gα subunit. The G protein/GoLoco interaction is therefore an attractive pharmacological target because it serves as a point of regulation for downstream GPCR signaling.(1)
Presently, no small molecule inhibitors of the Gα/GoLoco interaction are available for study. Although there is currently a lack of inhibitors for this recently discovered class of proteins, the potential applications of a GoLoco inhibitor are significant.(2) Combining existing GPCR agonists with specific GoLoco motif inhibitors could synergistically increase the specificity of existing drugs and serve as useful chemical probes.
To identify and assess potential modulators of the Gα/GoLoco interaction, putative inhibitors were screened in a FRET (Fluorescence/Förster Resonance Energy Transfer) assay using BMG LABTECH’s POLARstar Omega.
Fig. 1: Assay Principle for the FRET-based inhibitor screening
The ground state Gαi1 binds GDP. The GoLoco motif selectively recognizes Gαi1 (GDP) and binds with nanomolar affinity.(3) In a FRET-based assay which measures the Gα/GoLoco interaction, the FRET partners CFP-Gαi1·GDP and YFP-GoLoco yield a significantly increased (530/485 nm) FRET ratio when bound to each other than when compared with the non-interacting state of CFP-Gαi1 (as formed using aluminum tetrafluoride which mimics the transition state for GTP hydrolysis).
An inhibitor of this interaction could possibly perturb or displace the nucleotide from the Gα binding pocket or it could occlude the GoLoco motif binding surface, thus preventing the protein/protein interaction.(4) When Gα and GoLoco are no longer interacting, the (530/485 nm) FRET ratio decreases (Figure 1).
Fig. 2: BMG LABTECH’s multidetection microplate reader POLARstar Omega
The RGS12 Gα·GDP binding region (GoLoco motif, aa 1187-1221) was produced as a fluorescent chimera with the yellow fluorescent protein (YFP). Compounds were titrated into an optimized concentration of YFP-RGS12GoLoco and cyan fluorescent protein labeled Gαi1(CFP Gαi1) to confirm inhibitor activity. Ground state Gαi1 is GDP bound and the GoLoco motif binds readily, thus producing a higher emission ratio in the FRET assay from the non-directly excited fluorophore. In the mimicked GTP hydrolysis transition state, created by the binding of aluminum tetrafluoride (denoted AMF), the Gαi1/GoLoco interaction does not occur. This is evident by the decrease in the observed FRET ratio of non-directly excited fluorophore emission versus directly excited fluorophore emission.
For FRET measurements, the POLARstar Omega (Figure 2) was used with filters optimized for CFP/YFP FRET. Excitation filter with wavelength at 422 nm and dual emission wavelength filters at 530 nm and 485 nm were used. Instrument settings can be found below.
|Positioning Delay||0.2 sec|
|Measurement start time||0.0 sec|
|No. of flashes per well||10|
|No. of multichromatics||1|
|Simultaneous dual emission activated|
|Excitation filter||422 nm|
|EmissionA filter||530 nm|
|EmissionB filter||485 nm|
The FRET pair (100 nM CFP- Gαi1 and 100 nM YFP-RGS12GoLoco) in either buffer (GDP or AMF buffer) was aliquotted in duplicate 60 µL volumes with increasing concentrations of non-labeled competitor. The plate was read immediately and the FRET ratio (530/485) determined using the MARS Data Analysis Software.
GDP Buffer: 10 mM Hepes pH 7.4, 150 mM NaCl, 0.0005 % NP40 alternative, 100 µM GDP, 50 µM EDTA
AMF Buffer: 10 mM Hepes pH 7.4, 150 mM NaCl, 0.0005 % NP40 alternative, 100 µM GDP, 50 µM EDTA, 10 mM MgCl2, 10 mM NaF, 30 µM AlCl3
The G-protein/GoLoco interaction is dependent on the nucleotide bound state of the G-protein. The interaction occurs with the GDP bound state of the G-protein, but not in the GTP hydrolysis-mimicking state (AMF). Figure 3 shows the FRET ratio obtained in both buffers for different concentrations of unlabeled competitor Gαi1.
Fig. 3: 530/485 nm emission ratio of CFP-Gαi1 and YFP-RGS12GoLoco measurements using different buffers (GDP-buffer and AMF buffer) and varied concentrations of unlabeled competitor G-protein.
A Z’ value of 0.853 was obtained when the reaction volume was at 60 µL. The assay system also demonstrated a robust tolerance to DMSO (up to 5%), a commonly used solvent for small molecule screening (Figure 4).
Fig. 4: DMSO tolerance results of CFP-Gαi1 and YFP-RGS12GoLoco measurements in GDP and AMF buffer with competitive inhibition of the interaction using varied concentrations of unlabeled GoLoco motif peptide.
The CFP/YFP FRET based assay is a robust assay and was used in a medium-throughput small molecule screen of 57 compounds (Figure 5).
Fig. 5: GoLoco Screen of 57 compounds in GDP and AMF buffer.
The results show that the CFP-YFP FRET based assay is a robust assay that tolerates DMSO in concentrations often used to dissolve small molecules. A very good Z’ value of 0.853 in a low volume of only 60 µL shows the potential to perform the assay in 384-well format.
The POLARstar Omega proved to be a reliable and robust instrument to perform these FRET based measurements. With the MARS Data Analysis Software, the FRET ratio and Z’ value are automatically calculated providing fast and efficient evaluation.