In our approach using BRET

In our approach using BRET and FRET, there is a substantial increase in BRET and FRET signals between Giα2 or Giα3-YFP and CFP- Gγ2/Rluc-Gγ2 upon apelin-13 activation, indicating that Gαi2, Gαi3 and N-terminus of Gβ1γ2 subunits become closer without subunit dissociation. In contrast, both the FRET and BRET signals decrease when the fluorescent protein is tagged at C-terminal of Gγ2, clearly indicating there is a heterotrimeric composition between Gαi2, Gai3 and Gβ1γ2 during apelin-13 simulated APJ activation. However, we only observed a substantial decrease in both BRET ratio and FRET signal between fluorescent Gαo and Gαq subunit and Gβγ subunits independent the position of fluorescent tags on Gγ2, indicating Gαo and Gαq subunits dissociate with Gβγ subunits in a classical manner during APJ activation. These data are consistent with previous findings that APJ activates intracellular effectors via Gαi2, Gαo and Gαq subunits. Meanwhile, to the best of our knowledge, this evidence is novel in showing APJ couple to Gαi3 upon apelin-13 activation. Previous findings that APJ couples to Gαq upon activation were conducted in mouse models. Our results also provide the direct evidence to support the human APJ is also coupled to Gαq after apelin-13 stimulation for the first time. The ability of coupling to variety kinds of G-protein has been suggested to be important for APJ differentially couple to downstream effectors (e.g. MAPK signalling). Recently, Kang et al. also demonstrated in human umbilical vein endothelial KN-93 Phosphate (HUVECs), APJ activates Gα13 resulting in translocation of HDAC4 and HDAC5 and activation of transcription factor MEF2, indicating the complexity of APJ signals [29]. Besides, using dominant-negative Gαi2, Bo Bai et al. also found that apelin-13 induce ERK1/2 activation through coupling APJ to Gαi2 pathway in HEK293 cells [30]. Moreover, the activation of ERK1/2 by apelin is mediated via PKC in HEK293 cells expressing mouse APJ, indicative of coupling to either Go or Gq/11[31]. Although it was reported that Gαi1 was activated during APJ activation, however, here we did not observe changes of FRET and BRET signals between fluorescent Gαi1 and Gβγ subunits. This discrepancy could be due to the difference of G-protein species used in the assays and using difference cell lines. Combined with our findings, it suggests the apelin signalling pathway shows a great diversity and might exhibit functional selectivity or biased signalling. Meanwhile, our data also provide a real-time analysis of dynamics of G-protein coupling to apelin receptor in living cells. It has revealed that upon activation, Gαi2, Gαi3, Gαo, and Gαq undergo very rapid conformational changes (less than 5s) demonstrated by both FRET and BRET. APJ binds numerous apelin isoforms and signals through various G-proteins to a variety of signalling pathways to culminate in different patterns of activation and desensitization that may be tissue- and cell type-specific. Therefore, our assays could provide a powerful tool to analyze crowded intracellular signalling associated with APJ. In the last decade, there has been a substantial re­evaluation of the assumption that GPCRs exist primarily as monomeric polypeptides, with support increasing for a model in which GPCRs can exist as homo­ or hetero­dimers or even high­order oligomers [32], [33], [34]. Recent reports suggest that homo­ or hetero­dimers exhibit specific functional properties which are distinct from monomeric receptors, involving signalling, trafficking amongst other pathways [35], [36], [37]. We have also employed these technologies which suggest that human APJ forms a heterodimer with KOR and leads to increased PKC and decreased protein kinase A (PKA) signal transduction evoked by apelin-13 or DynA(1-13)[20]. Our work here could help us clearly understand the G-protein coupling profiling of APJ and lays a solid foundation to identify whether APJ homo­ or hetero­dimers affect G-protein coupling in future studies. In addition, it is strongly convincing that Gβγ could be a promising pharmaceutical target in the treatment of some diseases. Turning on or off all of the signalling pathways downstream through targeting specific GPCRs has been very successful in the treatment of many diseases. As described above, Gβγ undergoing dissociation or rearrangement is a critical prerequisite for some signal transduction. Therefore, measuring distinct Gβγ signalling upon apelin-13 activation can be useful to decipher salutary effects in a number of pathological models.