To investigate β-arrestin 2 (ARRB2) interaction with and regulation of Relaxin Family Peptide Receptor 1 (RXFP1) in support of finding small molecule biased agonists of RXFP1 to ameliorate fibrosis and inflammation of the lower urinary tract.  Relaxin-2 (also referred to as serelaxin) signaling is antifibrotic and anti-inflammatory, with several pathologies showing benefit from relaxin treatment in preclinical models.  Relaxin has passed safety requirements in multiple clinical trials.  Relaxin signaling at RXFP1 can also promote angiogenesis via cyclic adenosine monophosphate (cAMP), signaling that would be counterindicated in patients with increased risk or history of cancer.  This suggests that the development of biased agonists of RXFP1 could be beneficial.  Towards that end, we have generated data in cellular assays evaulating subcellular localization and ARRB2 recruitment of RXFP1 following treatment with relaxin.

To determine protein subcellular localization, plasmids encoding fusion proteins were designed and cloned.  ARRB2-mCerulean3-N1 and RXFP1-mCherry2-N1 were transfected into HEK-293T cells either singly or in combination.  HEK-293T cells were seeded onto collagen coated glass coverslips at a density of 26,000 cells per square centimeter and then transfected with a total of 7.4*10^-13 g DNA per cell using Lipofectamine 3000 (Thermo Fisher Scientific) according to manufacturer instructions.  Cells were fixed in 4% paraformaldehyde and mounted in SlowFade Diamond Antifade Mountant (Thermo Fisher Scientific).  
To evaluate activation dependent ARRB2 recruitment to RXFP1 and RXFP2, we performed the TANGO assay, which generates signal based on reporter gene expression downstream of ARRB2 recruitment to a target GPCR transiently transfected into the HTLA cell line.  TANGO plasmids for human RXFP1 and RXFP2 were obtained from addgene (plasmids 66492 and 66493).  HTLA cells were a generous gift from Bryan Roth and Wesley K Kroeze.  HTLA cells were maintained in in DMEM supplemented with 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin, 2 μg/ml puromycin and 100 μg/ml hygromycin B in a humidified atmosphere at 37 °C in 5% CO2.  Cells were seeded into 96-well plates at 114,000 cells per well.  Cells were transfected with TANGO-RXFP1 or TANGO-RXFP2 at 42 ng DNA per well using Lipofectamine 3000 (Thermo Fisher Scientific) according to manufacturer instructions, with five replicate wells per condition.  Luminescence was measured on a SpectraFluor plate reader (Tecan) using the Bright-Glo Luciferase Assay System (Promega) according to manufacturer instructions.

In HEK-293T cells transfected with RXFP1-mCherry2-N1 alone, RXFP1 signal was consistent with membrane localization of the receptor.  In contrast, in cells transfected with both RXFP1-mCherry2-N1 and ARRB2-mCerulean3-N1, RXFP1 signal was lost from the membrane and was instead found in large and dense perinuclear deposits.  The distribution of ARRB2-mCerulean3-N1 signal was consistent with cytoplasmic localization whether or not RXFP1 was co-transfected.
A relaxin-dose-dependent increase in luminescence was seen in HTLA cells transfected with TANGO-RXFP1.  In HTLA cells transfected with RXFP2, dose-dependent increases in luminescence were seen for treatment with INSL3 and to a lesser extent, relaxin.  The dose-response curves show a bell shape consistent with receptor oligomerization and internalization, as reported elsewhere.  Background signal was higher in RXFP1 cells than in RXFP2 cells.

In cells derived from HEK 293, ARRB2 regulates RXFP1 protein subcellular localization.  Transfection of both RXFP1 and ARRB2 led to the formation of large deposits of RXFP1 at the rim of the nucleus, yet when the same amount of RXFP1 DNA was transfected into cells without exogenous ARRB2, RXFP1 did not form these deposits and was instead distributed in a fashion consistent with localization on the cell membrane.  This also provides a mechanistic explanation for the lack of observed ARRB2 relocalization after relaxin treatment seen in [1].  Working in HEK-293 cells, their work fluorescently labeled only ARRB2 and not RXFP1, so the loss of membrane expression of RXFP1 would not have been detected.  In experiments adding relaxin to cells lacking membrane expression of RXFP1, no response would be predicted.
The TANGO assay provides evidence for the relaxin driven association of RXFP1 and ARRB2.  The TANGO assay is a three part system.  HTLA cells are stably modified for ARRB2 fused to the Tobacco Etch Virus (TEV) protease and for tTA-driven expression of luciferase.  The TANGO plasmids encode a GPCR of interest C-terminally fused to the tTA transcription factor, so that when ARRB2 binds the TANGO receptor, cleavage of the transcription factor ensues, promoting expression of luciferase.  The assay is specific to the target receptor and independent of the G protein the receptor signals through.
RXFP1 TANGO assay showed a clear and strong dose dependent response to relaxin with the bell-shaped curve reported previously elsewhere.  The RXFP1 TANGO assay also showed higher baseline signal than the RXFP2 assay, which is consistent with the constitutive interaction between ARRB2 and RXFP1 deemed the “signalosome” [2].  RXFP2 TANGO assay results show receptor activation and ARRB2 recruitment in response to INSL3 and to a lesser extent, relaxin, as well as a bell-shaped response curve for INSL3.  All of these data are consistent with previous reports in the literature, serving to validate the TANGO assay and these results.

These data show that in at least two contexts, ARRB2 interacts with RXFP1/2.  It remains to be determined whether RXFP1 signals through ARRB2 to effect its biological phenotypes.  The implementation of a selective and sensitive assay for activation of RXFP1 and RXFP2 opens the door to the development of biased small molecule agonists of these receptors.

Callander, G. E.; Thomas, W. G.; Bathgate, R. A. D. Prolonged RXFP1 and RXFP2 Signaling Can Be Explained by Poor Internalization and a Lack of ß-Arrestin Recruitment. Am. J. Physiol. - Cell Physiol. 2009, 296 (5), C1058–C1066.
Halls, M. L.; Cooper, D. M. F. Sub-Picomolar Relaxin Signalling by a Pre-Assembled RXFP1, AKAP79, AC2, ß-Arrestin 2, PDE4D3 Complex. EMBO J. 2010, 29 (16), 2772–2787.