**4. BRET principle and its application in the field of 7TMRs dimerization**

## **4.1 BRET principle**

84 Bioluminescence – Recent Advances in Oceanic Measurements and Laboratory Applications

In addition to widespread intra-family hetero-dimerization, inter-family hetero-dimerization has also been reported, at least between both of the family A members β2-AR and opsin and the family B member gastric inhibitory polypeptide receptor (GIP) (Vrecl et al., 2006), and between the family A serotonin 5-HT2A receptors and the family C mGluR2 (Gonzalez-Maeso et al., 2008). Both types of hetero-dimers were demonstrated to be functional, either by their ability to induce cAMP production upon agonist stimulation (family A/B hetero-dimer), or by

Growing experimental data support the view that 7TMRs exist and function as contact dimers or higher order oligomers with TM regions at the interfaces. In contact dimers/oligomers of 7TMRs, the original TM helical-bundle topology of each individual protomer is preserved and interaction interfaces are formed by lipid-exposed surfaces. Although domain-swap models, i.e. models in which domains TM1/TM5 and TM6/TM7 would exchange between protomers, have also been proposed in the literature, there is there is limited direct evidence that supports these assumptions. On the other hand, compelling experimental evidence exists for the involvement of lipid exposed surfaces of TM1, TM4 and/or TM5 at the dimerization/oligomerization interfaces of several 7TMRs. Besides, the interface may depend on additional stabilizing factors such as the coiled-coil interactions reported in the GABAB receptor and the disulfide bridge interactions in the muscarinic and the other class C receptors (reviewed by (Filizola)). A web service, named G-protein coupled Receptors Interaction Partners (GRIP) that predicts the interfaces for 7TMRs oligomerization is also available at http://grip.cbrc.jp/GRIP/index.html (Nemoto et al., 2009). G protein coupled Receptor Interaction Partners DataBase (GRIPDB) has also been developed, which provides information about 7TMRs oligomerization i.e. experimentalaly indentified 7TMRs oligomers, as well as suggested interfaces for the

7TMRs are one of the most important drug targets in the pharmaceutical industry; approximately 40% of the prescription drugs on the market target 7TMRs, but only 5% of the known 7TMR targets are utilized. Agonists and antagonists of 7TMRs are used in the treatment of diseases of every major organ system including the central nervous system, cardiovascular, respiratory, metabolic and urogenital systems. The most exploited 7TMR drug targets include AT1 angiotensin, adrenergic, dopamine and serotonin (5 hydroxytryptamine, 5-HT) receptor subtypes (Schoneberg et al., 2004). For instance, antagonists of AT1 angiotensin II receptors are used to prevent diabetes mellitus-induced renal damage and to treat essential hypertension and congestive heart failure. β-adrenergic receptor antagonists, acting on β1- and/or β2-adrenergic receptors, are used in patients with congestive heart failure and to treat hypertension and coronary heart disease, while β2 adrenergic receptor agonists are used in the treatment of asthma, chronic obstructive pulmonary disease and to delay preterm labor. Dopamine receptor antagonists, primarily acting on D2 receptors, are utilized in the treatment of schizophrenia, while dopamine receptor agonists (e.g. precursor for dopamine levodopa (L-dopa)) remain the standard for treating Parkinson's disease. Inhibitors of 5-HT uptake, which act as indirect agonists at

their ability to modulate G-protein coupling (family A/C hetero-dimer).

**3.1 Dimerization interface** 

oligomerization (Nemoto et al., 2011).

**3.2 Therapeutic application and drug discovery** 

BRET is a biophysical method that enables monitoring of physical interactions between two proteins fused to BRET donor and acceptor moieties, respectively, dependent on their intermolecular distance (10 – 100 Å) and on relative orientation due to the dipole-dipole nature of the resonance energy transfer mechanism (Zacharias et al., 2000). BRET is a nonradiative energy transfer, occurring between a bioluminescent donor that emits light in the presence of its corresponding substrate and a complementary fluorescent acceptor, which absorbs light at a given wavelength and re-emits light at longer wavelengths. To fulfill the condition for energy transfer, the emission spectrum of the donor must overlap with the excitation spectrum of the acceptor molecule (Zacharias et al., 2000). BRET occurs naturally in some marine species (e.g. in the sea pansy *Renilla reniformis*) and in 1999, Xu et al. (Xu et al., 1999) utilized this approach to study dimerization of the bacterial Kai B clock protein. Since then, several versions of BRET assays have been developed that use different substrates and/or energy donor/acceptor couples. The original BRET1 technology used the pairing of *Renilla luciferase* (Rluc) as the donor and yellow fluorescent protein (YFP) as the acceptor (Xu et al., 1999; Xu et al., 2003). The addition of coelenterazine h, the natural substrate of *Renilla luciferase* (Rluc), leads to a donor emission of blue light (peak at ~480 nm). When the YFP-tagged acceptor molecule, adapted to this emission wavelength, is in close proximity to the Rluc-tagged donor molecule, excitation of YFP occurs by resonance energy transfer resulting in an acceptor emission of green light (peak at ~530 nm). The substantial overlap in the emission spectra of Rluc and YFP acceptor emission (Stokes shift

Quantitative Assessment of Seven Transmembrane Receptors

form functionally relevant inter-family oligomers (Table 2).

*Homo sapiens, Rattus norvegicus*

*Homo sapiens, Rattus norvegicus*

*Homo sapiens, Mus musculus* 

**Oligomers (***in vitro***)** 

okb.org).

**Family A 7TMRs**  Adenosine A1 - Adenosine A2A oligomer (**A1 - A2A**)

Adenosine A2A - Cannabinoid CB1 oligomer (**A2A - CB1**)

Adenosine A2A - Dopamine D2 oligomer (**A2A - D2**)

Adrenergic 1B - Adrenergic 1D receptor oligomer (**1B - 1D adrenoreceptor**)

(7TMRs) Oligomerization by Bioluminescence Resonance Energy Transfer (BRET) Technology 87

evidence was provided *in vivo* by BRET are summarized in Table 2. It should be emphasized that besides the intra-family hetero-dimers, the members from different 7TMR families also

**Oligomers (***in vivo***) 7TMR Family A Family B Family C Family A/C Other BRET** 18 13 0 1 4 0 *Mus musculus* 7 5 0 1 1 0 *Rattus norvegicus* 9 5 0 1 3 0 *Homo sapiens* 9 8 0 0 1 0 **Other methods** 11 7 0 1 2 1

**BRET** 50 40 2 1 6 1 **Other methods** 192 160 4 13 13 2

**relevance** 

Implicated in Parkinson's disease.

Implicated in Parkinson's desease, schizophrenia. Level of adenosine is increased

in the striatal extracellular fluid in Parkinson's disease.

The study demonstrated that when the 1B-KO and 1D-KO strains of mice are used in conjunction with antagonists, a different pharmacological situation emerges relative to control (sensitivity to Phenylephrine).

Table 1. Comparisons of 7TMRs oligomers identified by BRET *vs.* others methods in different 7TMR families in *in vivo* and *in vitro.* Data source GPCR-OKB (http://www.gpcr-

**Oligomer name Organism** *In vivo* **evidence Potential clinical** 

technology

*Rattus norvegicus* evidence for physical association in

native tissue or primary cells

evidence for physical association in native tissue or primary cells, identification of a specific functional property in native tissue (brain)

evidence for physical association in native tissue or primary cells, identification of a specific functional property in native tissue (rat striatum, human striatum)

evidence for physical association in native tissue or primary cells, identification of a specific functional property in native tissue (brain), use of knockout animals or RNAi

only ~50 nm) creates a significant problem that has been overcome in a second generation of BRET assay (BRET2). In BRET2 assays, *Renilla luciferase* (Rluc) is used as the donor, the green fluorescent protein (GFP) variant GFP2 as the acceptor molecule (excitation ~400 nm, emission peak at 510 nm) and the proprietary coelenterazine DeepBlueCTM (also known as coelenterazine 400A) as a substrate. In the presence of DeepBlueCTM, Rluc emits light peaking at 395 nm, a wavelength that excites GFP2 resulting in the emission of green light at 510 nm. This modified BRET pair results in a broader Stokes shift of 115 nm, thus enabling superior separation of donor and acceptor peaks, as well as efficient filtration of the excitation light that it does not come to the detector, thereby enabling detection of the weak fluorescence signal. However, the disadvantage of BRET2, compared to BRET1 is the 100-300 times lower intensity of emitted light and a very fast decay of emitted light (Heding, 2004). BRET2 sensitivity can be improved by the development of suitably sensitive instruments (Heding, 2004) and the use of Rluc mutants with improved quantum efficiency and/or stability (e.g. Rluc8 and Rluc-M) as a donor (De et al., 2007). A third generation BRET assay (BRET3) has been developed recently and combines Rluc8 with the mutant red fluorescent protein (DsRed2) variant mOrange and the coelenterazine or EnduRen™ as a substrate (De et al., 2007). EnduRen™ is a very stable coelenterazine analogue that enables luminescence measurement for at least 24 hours after substrate addition and was utilized in the extended BRET (eBRET) technology (Pfleger et al., 2006). Therefore, in BRET3, donor spectrum is the same as in BRET1, and the red shifted mOrange acceptor signal (emission peak at 564 nm) improves spectral resolution to 85 nm, thereby reducing bleedthrough in the acceptor window. Improved spectral resolution and increased photon intensity allow imaging of protein-protein interactions from intact living cells to small living subjects. Additional optimized donor/acceptor BRET couples that combine Rluc/Rluc8 variant with the yellow fluorescent protein, the YPet variant and the Renilla green fluorescent protein (RGFP) has also been developed (Kamal et al., 2009).

#### **4.2 BRET and 7TMRs dimerization**

The use of energy-based techniques such as FRET and BRET has been fundamental for taking the theme of 7TMRs dimerization/oligomerization at the front of 7TMRs research. In 2000, BRET was introduced in the 7TMR field demonstrating β2-adrenergic receptor (β2-AR) dimerization (Angers et al., 2000) and since then BRET-based information about 7TMRs homo-/hetero-dimerization is rapidly accumulating (for a recent reviews see (Achour et al., 2011; Ayoub & Pfleger, 2010; Ferré et al., 2009; Ferré & Franco, 2010; Gurevich & Gurevich, 2008a; Gurevich & Gurevich, 2008b; Palczewski, 2010)). As a consequence, knowledge databases have been developed to gather and organize these scattered data and provide researchers with the comprehensive collection of information about 7TMR oligomerization. Existing databases are G protein-coupled receptor oligomer knowledge base (GPCR-OKB) (Skrabanek et al., 2007; Khelashvili et al., 2010) that is freely available at http://www.gpcrokb.org and G protein-coupled receptor interaction partners database (GRIPDB) (Nemoto et al., 2011) available at http://grip.cbrc.jp/GDB/index.html. By analyzing the data in the GPCR-OKB, we can see that BRET-based approaches were used more often than other experimental approaches such as co-immunoprecipitation, cross-linking, co-expression of fragments or modified protomers, use of dimer specific antibodies, fluorescence resonance energy transfer (FRET) and time resolved FRET to detect oligomerization *in vivo* while in *in vitro* systems others methods still prevail (Table 1). The 7TMR pairs for which functional

only ~50 nm) creates a significant problem that has been overcome in a second generation of BRET assay (BRET2). In BRET2 assays, *Renilla luciferase* (Rluc) is used as the donor, the green fluorescent protein (GFP) variant GFP2 as the acceptor molecule (excitation ~400 nm, emission peak at 510 nm) and the proprietary coelenterazine DeepBlueCTM (also known as coelenterazine 400A) as a substrate. In the presence of DeepBlueCTM, Rluc emits light peaking at 395 nm, a wavelength that excites GFP2 resulting in the emission of green light at 510 nm. This modified BRET pair results in a broader Stokes shift of 115 nm, thus enabling superior separation of donor and acceptor peaks, as well as efficient filtration of the excitation light that it does not come to the detector, thereby enabling detection of the weak fluorescence signal. However, the disadvantage of BRET2, compared to BRET1 is the 100-300 times lower intensity of emitted light and a very fast decay of emitted light (Heding, 2004). BRET2 sensitivity can be improved by the development of suitably sensitive instruments (Heding, 2004) and the use of Rluc mutants with improved quantum efficiency and/or stability (e.g. Rluc8 and Rluc-M) as a donor (De et al., 2007). A third generation BRET assay (BRET3) has been developed recently and combines Rluc8 with the mutant red fluorescent protein (DsRed2) variant mOrange and the coelenterazine or EnduRen™ as a substrate (De et al., 2007). EnduRen™ is a very stable coelenterazine analogue that enables luminescence measurement for at least 24 hours after substrate addition and was utilized in the extended BRET (eBRET) technology (Pfleger et al., 2006). Therefore, in BRET3, donor spectrum is the same as in BRET1, and the red shifted mOrange acceptor signal (emission peak at 564 nm) improves spectral resolution to 85 nm, thereby reducing bleedthrough in the acceptor window. Improved spectral resolution and increased photon intensity allow imaging of protein-protein interactions from intact living cells to small living subjects. Additional optimized donor/acceptor BRET couples that combine Rluc/Rluc8 variant with the yellow fluorescent protein, the YPet variant and the Renilla green fluorescent protein (RGFP) has

The use of energy-based techniques such as FRET and BRET has been fundamental for taking the theme of 7TMRs dimerization/oligomerization at the front of 7TMRs research. In 2000, BRET was introduced in the 7TMR field demonstrating β2-adrenergic receptor (β2-AR) dimerization (Angers et al., 2000) and since then BRET-based information about 7TMRs homo-/hetero-dimerization is rapidly accumulating (for a recent reviews see (Achour et al., 2011; Ayoub & Pfleger, 2010; Ferré et al., 2009; Ferré & Franco, 2010; Gurevich & Gurevich, 2008a; Gurevich & Gurevich, 2008b; Palczewski, 2010)). As a consequence, knowledge databases have been developed to gather and organize these scattered data and provide researchers with the comprehensive collection of information about 7TMR oligomerization. Existing databases are G protein-coupled receptor oligomer knowledge base (GPCR-OKB) (Skrabanek et al., 2007; Khelashvili et al., 2010) that is freely available at http://www.gpcrokb.org and G protein-coupled receptor interaction partners database (GRIPDB) (Nemoto et al., 2011) available at http://grip.cbrc.jp/GDB/index.html. By analyzing the data in the GPCR-OKB, we can see that BRET-based approaches were used more often than other experimental approaches such as co-immunoprecipitation, cross-linking, co-expression of fragments or modified protomers, use of dimer specific antibodies, fluorescence resonance energy transfer (FRET) and time resolved FRET to detect oligomerization *in vivo* while in *in vitro* systems others methods still prevail (Table 1). The 7TMR pairs for which functional

also been developed (Kamal et al., 2009).

**4.2 BRET and 7TMRs dimerization** 

evidence was provided *in vivo* by BRET are summarized in Table 2. It should be emphasized that besides the intra-family hetero-dimers, the members from different 7TMR families also form functionally relevant inter-family oligomers (Table 2).


Table 1. Comparisons of 7TMRs oligomers identified by BRET *vs.* others methods in different 7TMR families in *in vivo* and *in vitro.* Data source GPCR-OKB (http://www.gpcrokb.org).


Quantitative Assessment of Seven Transmembrane Receptors

*norvegicus, Mus musculus* 

Data source GPCR-OKB (http://www.gpcr-okb.org).

**4.3 Interpretation of BRET results – Possible drawbacks** 

*Rattus* 

Adenosine A2A - Dopamine D2 - Metabotropic glutamate 5 (mGLU5) oligomer (**A2A - D2 -** 

Serotonin 5-HT2A receptor oligomer - Metabotropic

glutamate 2 (**5-HT2A –** 

**mGLU5**)

**mGLU2**)

(7TMRs) Oligomerization by Bioluminescence Resonance Energy Transfer (BRET) Technology 89

evidence for physical association in native tissue or primary cells

**relevance** 

5-HT2A levels increase and mGLU2 levels decrease in schizophrenia

**Oligomer name Organism** *In vivo* **evidence Potential clinical** 

*Homo sapiens* evidence for physical association in

Table 2. Intra- and inter-family oligomers with *in vivo* evidence discovered by BRET method.

BRET signal indicates that molecules of the same (or two different) receptors are at maximum distance of 100 Å (that equals 10 nm) or more accurately that the donor and acceptor moieties are within this distance. The efficiency of energy transfer depends on the relative orientation of the donor and acceptor and the distance between them (Zacharias et al., 2000), so that absolute distances can not be measured. Experimentally determined Förster distance R0 (distance at which the energy transfer efficiency is 50%) for BRET1 and BRET2 is 4.4 nm and 7.5 nm, respectively (Dacres et al., 2010). 7TMR transmembrane core spans ~40 Å across the intracellular surface (Palczewski et al., 2000), which makes BRET suitable to the study of dimerization. However, certain facts need to be considered when interpreting BRET results. Firstly, the size of 27 kDa fluorescent proteins and 34 kDa *Renilla luciferase* is comparable to that of the transmembrane core of 7TMRs (diameter ∼40 Å). These proteins are usually attached to the receptor C-terminus, which in different 7TMRs varies in length from 25 to 150 amino acids. Polypeptides of this length in extended conformation can cover 80−480 Å. Thus, a BRET signal indicates that the donor and acceptor moieties are at distance less than 10 nm, which may occur when receptors form structurally defined dimer or when they are far >500 Å apart (reviewed by (Gurevich & Gurevich, 2008a)). The use of acceptor and donor molecules genetically fused to 7TMRs can alter the functionality of the receptor; fusion proteins can also be expressed in the intracellular compartments, thus making difficult to demonstrate that the RET results from a direct interaction of proteins at the cell surface (Ferre & Franco, 2010). The use of fusion proteins can therefore be a major limitation for this application. Secondly, quantitative BRET measurements are limited by the quality of the signal and noise level. Fluorescent proteins and luciferase yield background signals arising from incompletely processed proteins inside the cell and high cell autofluorescence in the spectral region used (Gurevich & Gurevich, 2008a). Thirdly, so called bystander BRET results from frequent encounters between overexpressed receptors and has no physical meaning (Kenworthy & Edidin, 1998; Mercier et al., 2002). BRET assays should therefore be able to discriminate between genuine dimerization compared to random collision due to over-expression. To determine specify of BRET signal the following experiments has been proposed: negative control with a non-interacting receptor or protein, BRET saturation and competition assays and experiments that observe ligand-promoted changes in BRET (Achour et al., 2011; Ayoub

native tissue or primary cells, identification of a specific functional property in native tissue (brain)



Table 2. Intra- and inter-family oligomers with *in vivo* evidence discovered by BRET method. Data source GPCR-OKB (http://www.gpcr-okb.org).
