Organ-specific function of adhesion G protein-coupled receptor GPR126 is domain-dependent

Despite their abundance and multiple functions in a variety of organ systems, the function and signaling mechanisms of adhesion G protein-coupled receptors (GPCRs) are poorly understood. Adhesion GPCRs possess large N termini containing various functional domains. In addition, many of them are autoproteolytically cleaved at their GPS sites into an N-terminal fragment (NTF) and C-terminal fragment. Here we demonstrate that Gpr126 is expressed in the endocardium during early mouse heart development. Gpr126 knockout in mice and knockdown in zebrafish caused hypotrabeculation and affected mitochondrial function. Ectopic expression of Gpr126-NTF that lacks the GPS motif (NTF^Δ”GPS) in zebrafish rescued the trabeculation but not the previously described myelination phenotype in the peripheral nervous system. These data support a model in which the NTF of Gpr126, in contrast to the C-terminal fragment, plays an important role in heart development. Collectively, our analysis provides a unique example of the versatile function and signaling properties of adhesion GPCRs in vertebrates.

Members of the adhesion G protein-coupled receptor (GPCR) family are expressed in many developing organs (e.g., nervous system and reproductive organs), immune cells, and cancer cells, suggesting that they might play an important role in physiological and also in pathological functions (1). Compared with their potential importance, the function and signaling mechanisms of adhesion GPCRs are poorly understood. Gpr126 (DREG) is expressed in mice heart and somites during embryogenesis and in the adult lung (2). It has been shown that both human and mouse Gpr126 can be cleaved at the GPS site by an endogenous proteolytic process resulting in a membrane-bound 35-kDa protein containing a seven-transmembrane domain [C-terminal fragment (CTF)] and an extracellular soluble protein [N-terminal fragment (NTF)], which might be further cleaved resulting in a 70-kDa soluble protein containing a complement, Uegf, Bmp1 (CUB) and a pentraxin (PTX) domain (2). These data indicated that the membrane-bound CTF can act as an independent receptor and the soluble NTF can act as a ligand or coreceptor for unknown receptors. This hypothesis is supported by the recent finding that the GAIN/GPS structure of latrophilin, another adhesion GPCR, operates independently of the CTF in Caenorhabditis elegans fertility (3). It has also been reported that the NTF of the brain angiogenesis inhibitor 1 (BAI1) inhibits endothelial cell proliferation and angiogenesis in mice (4, 5) and that the NTFs of EMR2 and GPR56 can cross-react with the CTF of latrophilin (6).

Studies on Gpr126 knockout mice have shown that disruption of the Gpr126 gene leads to fully penetrant embryonic lethality with cardiac abnormality (7). In another Gpr126 knockout mouse line from Taconic (T-Gpr126^/) (8), most mutants die in utero, although a few mice survive to postnatal stages. T-Gpr126^/ mice are characterized both by a lack of myelination in the peripheral nervous system (PNS) and by multiple defects in peripheral nerves. A myelination phenotype has also been observed in the zebrafish mutant line gpr126^st49 (9). Several experiments including rescue experiments with forskolin indicate that cAMP and PKA are involved in the GPR126-mediated pathway initiating myelination (9, 10). However, although mouse and zebrafish Gpr126 are true orthologs (7), no heart phenotype has been described in the zebrafish mutant line. The gpr126^st49 mutation introduces a premature stop codon before the GPS motif. This raises the possibility that gpr126^st49 mutant fish still express a functional fragment of NTF (amino acid 1–783, hereafter referred to as NTF^”GPS) but no CTF. Therefore, we hypothesize that the NTF of Gpr126 functions independently of its CTF during heart development in contrast to the role of Gpr126 in the PNS.