GPCR-β-arrestin interactions are an emerging field in GPCR drug discovery. Accurate, precise and easy to set up methods are necessary to monitor such interactions in living systems. We show a structural complementation assay to monitor GPCR-β-arrestin interactions in real time living cells, and it can be extended to any GPCR.
Interactions between G-protein coupled receptors (GPCRs) and β-arrestins are vital processes with physiological implications of great importance. Currently, the characterization of novel drugs towards their interactions with β-arrestins and other cytosolic proteins is extremely valuable in the field of GPCR drug discovery particularly during the study of GPCR biased agonism. Here, we show the application of a novel structural complementation assay to accurately monitor receptor-β-arrestin interactions in real time living systems. This method is simple, accurate and can be easily extended to any GPCR of interest and also it has the advantage that it overcomes unspecific interactions due to the presence of a low expression promoter present in each vector system. This structural complementation assay provides key features that allow an accurate and precise monitoring of receptor-β-arrestin interactions, making it suitable in the study of biased agonism of any GPCR system as well as GPCR c-terminus ‘phosphorylation codes’ written by different GPCR-kinases (GRKs) and post-translational modifications of arrestins that stabilize or destabilize the receptor-β-arrestin complex.
GPCRs represent the target of nearly 35% of current drugs in the market1,2 and a clear understanding of their pharmacology is crucial in the development of novel therapeutic drugs3. One of the key aspects in GPCR drug discovery, particularly during the development of biased agonists is the characterization of novel ligands towards receptor-β-arrestin interactions4 and β-arrestin interactions with other cytosolic proteins such as clathrin5.
It has been documented that β-arrestin dependent signaling plays a key role in neurological disorders such as bipolar disorder, major depression, and schizophrenia6 and also severe side effects in some medications such as morphine7.
Current methods used to monitor these interactions usually do not represent actual endogenous levels of the proteins in study, in some cases they show weak signal, photobleaching and depending of the GPCR it might be technically challenging to set up8. This novel structural complementation assay uses low expression promoter vectors in order to mimic endogenous physiological levels and provides high sensitivity compared to current methods9. Using this approach, it was possible to easily characterize Galanin receptor-β-arrestin1/2 and also β-arrestin2-clathrin interactions10. This methodology can be widely used to any GPCR of particular interest where β-arrestins play a key physiological function or their signaling is relevant in some diseases.
1. Primer design strategy
2. PCR
3. DNA digestion
4. DNA agarose gel purification and cloning
5. Transformation of clones
6. Isolation of the finished plasmid
7. Transfection and protein expression
8. Monitoring receptor-β-arrestin1/2 interactions in HEK293 cells
Using the procedure presented here, interactions between a prototypical GPCR and two β-arrestin isoforms were monitored. Glucagon like peptide receptor (GLP-1r) constructs were made using primers containing NheI and EcoRI enzyme restriction sites and cloned into the vectors pBiT1.1-C [TK/LgBiT] and pBiT2.1-C [TK/SmBiT] while in the case of β-arrestins, two additional vectors were used pBiT1.1-N [TK/LgBiT] and pBiT2.1-N [TK/SmBiT] using enzyme restriction sites BgIII and EcoRI in the case of β-arrestin2 and NheI and XhoI in the case of β-arrestin1. HEK293 cells were transfected using 50 ng of GLP-1r-LgBiT/SmBiT and 50 ng of β-arrestin tagged with LgBiT or SmBiT at the N- or C-terminal. Four different plasmid combinations were screened (Figure 3) and the one with the highest luminescent signal was chosen for further experiments (Figure 4). In order to determine the EC50 values for each β-arrestin isoform recruitment, dose response curves were performed using 10 μM, 1 μM, 100 nM, 10 nM and 1 nM of GLP-1 ligand concentration (Figure 4c). Dose response curves were obtained from the maximum response of each concentration from the kinetic studies (Figure 4a, 4b).
Figure 1. Nucleotide sequences of the multicloning sites of the vectors used in the design of GLP-1r-β-arrestin1/2 structural complementation assay.
In order to develop the structural complementation assay for GLP-1r-β-arrestin1/2 system, it was necessary to tag at the C-terminal the GLP-1r with LgBiT and SmBiT using the enzyme restrictions NheI and EcoRI at the pBiT1.1-C [TK/LgBiT] and pBiT2.1-C [TK/SmBiT] vector. In the case of β-arrestin1/2 they also were tagged with the LgBiT and SmBiT at the C- and N-terminal using the enzyme restrictions BglII/EcoRI for β-arrestin2 and NheI/XhoI for β-arrestin1 at the four vectors. All vectors use the HSV-TK promoter to minimize nonspecific association and reduce experimental artifacts and each vector contains an expression cassette for ampicillin resistance in bacteria. Image adapted from reference 11. Please click here to view a larger version of this figure.
Figure 2. Schematic representation of the GPCR:β-arrestin1/2 structural complementation assay.
(a) How the GPCR:β-arrestin1/2 structural complementation assay works in the presence of ligand. (b) Structural representation of the different plasmid combinations for the GPCR and β-arrestin isoforms tagged with LgBiT or SmBiT. Please click here to view a larger version of this figure.
Figure 3. GLP-1r/β-arrestins orientation screening.
In order to obtain the highest sensitivity 4 different plasmid combinations were expressed during 24 h after transfection. Luminescent signals were detected in almost all different plasmid combinations (a-c) except for only one GLP-1r:β-arrestin orientation (d). The results are expressed as mean ± S.E.M. of two experiments performed in duplicate; each duplicate was averaged before calculating the S.E.M. The arrows indicate the time at which the cells were treated with GLP-1 at 10 μM final concentration. Please click here to view a larger version of this figure.
Figure 4. GLP-1r-β-arrestin1/2 interactions are dose dependent manner.
Dose dependent ligand relationship of β-arrestin2 (a) and β-arrestin1 recruitment (b). Dose response curves showing differential recruitment between β-arrestin1 and β-arrestin2 by GLP-1r (c). The results are expressed as mean ± S.E.M. of two experiments performed in duplicate; each duplicate was averaged before calculating the S.E.M. The arrows indicate the time at which the cells were treated with GLP-1 at the corresponding concentrations (10 μM, 1 μM, 100 nM, 10 nM and 1 nM, final concentrations). Please click here to view a larger version of this figure.
Vector | Enzyme restriction used | Primer sequence |
pBiT1.1-C [TK/LgBiT] / pBiT2.1-C [TK/SmBiT] | SacI | 5 ́-XXXXXXXXGAGCTCC(Rev SI)-3 ́ |
EcoRI | 5 ́-XXXXXXXXGAATTCCC(Rev SI)-3 ́ | |
XhoI | 5 ́-XXXXXXXXCTCGAGCC(Rev SI)-3 ́ | |
pBiT1.1-N [TK/LgBiT] / pBiT2.1-N [TK/SmBiT] | Xho | 5 ́-XXXXXXXXCTCGAGCGGT (SI)-3 ́ |
SacI | 5 ́-XXXXXXXXGAGCTCAG(SI)-3 ́ | |
EcoRI | 5 ́-XXXXXXXXGAATTCA(SI)-3 ́ |
Table 1. Sequences of primers for the different restriction enzyme sites in the coding sequence of the linker for the pBiT1.1 and pBiT2.1 Vectors.
SI = Sequence of interest; Rev SI = reverse complementary of the sequence of interest. Table adapted from reference 11.
Vector | Linker sequence | Enzyme restriction used | ||
pBiT1.1-C [TK/LgBiT] / pBiT2.1-C [TK/SmBiT] | SI-GlyAlaGlnGlyAsnSerGlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGly-(LgBiT/SmBiT) | SacI | ||
SI-GlyAsnSerGlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGly-(LgBiT/SmBiT) | EcoRI | |||
SI-GlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGly-(LgBiT/SmBiT) | XhoI | |||
pBiT1.1-N [TK/LgBiT] / pBiT2.1-N [TK/SmBiT] | (LgBiT/SmBiT)-GlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGly-SI | XhoI | ||
(LgBiT/SmBiT)-GlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGlyGlyAlaGln-SI | SacI | |||
(LgBiT/SmBiT)-GlySerSerGlyGlyGlyGlySerGlyGlyGlyGlySerSerGlyGlyAlaGlnGlyAsnSer-SI | EcoRI |
Table 2. Linker amino acid sequences related with SacI, EcoRI or XhoI restriction sites in the pBiT1.1 and pBiT2.1 Vectors.
Red residues have to be encoded by PCR primers. Table adapted from reference 11.
Using the method presented here, interactions between any GPCR and β-arrestin1/2 can be monitored in real time living systems using this GPCR-β-arrestin structural complementation assay. In this regard, we were able to observe differential β-arrestin recruitment between the two β-arrestin isoforms by the GLP-1r (A prototypical Class B GPCR), we also observed a dissociation of the receptor-β-arrestin complex a few minutes after reaching the maximum luminescent signal.
In order to have the best sensitivity in the structural complementation assay system, it was screened with four different spatial orientations between the receptor and each β-arrestin isoform and the one with highest luminescent signal was used for posterior studies such as dose response stimulation curves (Figure 4). Using this methodology it was possible to characterize receptor-β-arrestin interactions using a GPCR of high therapeutic value in endocrinological diseases such as Diabetes mellitus15. In the same way this strategy can be easily adapted to any GPCR by simple tagging the GPCR of interest with LgBiT or SmBiT at the C-terminal and using the β-arrestin1/2 constructs described here and it emerges as a powerful alternative to current methodologies without the necessity of a complex set up where the overlapping between the donor and acceptor can be an obstacle as in some cases of BRET and FRET. Another significant advantage is that the vectors used in this system contain low expression promoters in an attempt to mimic endogenous expression levels. With this feature, we can rule out the possibility of non-specific associations due to high expression levels of the receptor and/or β-arrestin. In Figure 3 and Figure 4, there is a clear difference in the receptor-β-arrestin complex between β-arrestin1 versus β-arrestin2 and also a higher efficacy and intensity towards β-arrestin1 over β-arrestin2. The screening into four different orientations was proposed to increase the sensitivity of the assay making the system highly sensitive even at endogenous expression levels.
This methodology is very straight forward to perform. Perhaps the most critical step within this protocol is the primer design to amplify the receptor of interest. The user must be very careful in selecting what restriction enzyme to use according to Figure 1 and based on this to add the corresponding nucleotides to the primers (Table 1) to encode the red highlighted amino acids (Table 2).
One limitation of this methodology can be that the furimazine will degrade in an aqueous solution at or near physiological pH, leading to a gradual decrease in luminescence intensity independent of any change in GPCR-β-arrestin interactions16. To overcome this limitation the user should always include a normalization control (vehicle treated samples) when continuously monitoring luminescence for extended time periods. It is also important to use low levels of fetal bovine serum during the assay since its presence it might increase the rate of furimazine degradation16. One problem that may arise during the assay is that for luminescent values for a known GPCR-β-arrestin interaction can be no significant increase is registered compared to the base line values in all four different plasmid combinations. This can be due to the low expression from the HSV-TK promoter17. In that case, one alternative is to subclone the Open Reading Frames encoding LgBiT and SmBiT fusion proteins into expression vectors using the CMV promoter. When changing to a stronger promoter, optimization of the amount of transfected DNA should be done in order to obtain the best assay response.
Using this structural complementation assay we were able to observe with great accuracy the β-arrestin1/2 recruitment interactions by a prototypical class B GPCR. Using this method, it is possible to pharmacologically characterize novel drugs of particular interest targeting GPCRs.
The authors have nothing to disclose.
This work was supported by grants from the Research Program (NRF- 2015M3A9E7029172) of the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT, and Future Planning.
Antibiotics penicillin streptomycin | Welgene | LS202-02 | Penicillin/Streptomycin |
Bacterial Incubator | JEIO Tech | IB-05G | Incubator (Air-Jacket), Basic |
Cell culture medium | Welgene | LM 001-05 | DMEM Cell culture medium |
Cell culture transfection medium | Gibco | 31985-070 | Optimem 1X cell culture medium |
CO2 Incubator | NUAIRE | NU5720 | Direct Heat CO2 Incubator |
Digital water bath | Lab Tech | LWB-122D | Digital water bath lab tech |
DNA Polymerase proof reading | ELPIS Biotech | EBT-1011 | PfU DNA polymerase |
DNA purification kit | Cosmogenetech | CMP0112 | miniprepLaboPass Purificartion Kit Plasmid Mini |
DNA Taq Polymerase | Enzynomics | P750 | nTaq DNA polymerase |
Enzyme restriction BglII | New England Biolabs | R0144L | BglII |
Enzyme restriction buffer | New England Biolabs | B72045 | CutSmart 10X Buffer |
Enzyme restriction EcoRI | New England Biolabs | R3101L | EcoRI-HF |
Enzyme restriction NheI | New England Biolabs | R01315 | NheI |
Enzyme restriction XhoI | New England Biolabs | R0146L | XhoI |
Fetal Bovine Serum | Gibco Canada | 12483020 | Fetal Bovine Serum |
Gel/PCR DNA MiniKit | Real Biotech Corporation | KH23108 | HiYield Gel/PCR DNA MiniKit |
Ligase | ELPIS Biotech | EBT-1025 | T4 DNA Ligase |
Light microscope | Olympus | CKX53SF | CKX53 Microscope Olympus |
lipid transfection reagent | Invitrogen | 11668-019 | Lipofectamine 2000 |
Luminometer | Biotek/Fisher Scientific | 12504386 | Synergy 2 Multi-Mode Microplate Readers |
NanoBiT System | Promega | N2014 | NanoBiT PPI MCS Starter System |
Nanoluciferase substrate | Promega | N2012 | Nano-Glo Live Cell assay system |
PCR Thermal cycler | Eppendorf | 6336000015 | Master cycler Nexus SX1 |
Poly-L-lysine | Sigma Aldrich | P4707-50ML | Poly-L-lysine solution |
Trypsin EDTA | Gibco | 25200-056 | Trysin EDTA 10X |
White Cell culture 96 well plates | Corning | 3917 | Assay Plate 96 well plate |