Class A seven-transmembrane receptor of the melatonin-receptor subfamily, expressed mainly in hypothalamus (including tanycytes) and pituitary, and distinguished by the absence of N-glycosylation sites and by an exceptionally long (>300 residue) cytoplasmic C-terminal tail. Despite its family membership GPR50 does not bind melatonin: the receptor failed to bind radiolabelled melatonin when first expressed, and reciprocal swaps of the second extracellular loop between GPR50 and MT1 show that loop divergence is what makes GPR50 melatonin-unresponsive. A cryo-EM structure of the ligand-free receptor shows an inactive-like state with a putative pocket and ligand access channels unlike those of the melatonin receptors, together with moderate constitutive coupling to Galpha12 and weaker coupling to Galphai/o. Constitutive, agonist-independent signalling through G12/13 and RhoA restrains neurite outgrowth and cell migration in neural progenitor cells and tanycytes. GPR50 also heterodimerizes constitutively with the MT1 melatonin receptor and abolishes high-affinity agonist binding and G-protein coupling by the MT1 protomer, and it interacts with NOGO-A/RTN4, KAT5/TIP60 and ADAM17. Calpain-1 cleaves the receptor between Phe-408 and Ser-409, releasing a C-terminal domain that translocates to the nucleus and acts with the transcription factor GTF2I on FOS transcription. Whether GPR50 has an endogenous agonist is disputed: the neuropeptide Little-LEN, derived from the ProSAAS precursor, has been proposed as its ligand in a single 2026 study reporting Galphai coupling and a role in energy expenditure and thermogenesis, while a cryo-EM study published weeks later states that no endogenous agonist has been characterized. Genetic variants at the locus have been associated with mood disorders and other neuropsychiatric phenotypes.
| GO Term | Evidence | Action | Reason |
|---|---|---|---|
| GO:0004930 G protein-coupled receptor activity | IBA GO_REF:0000033 | ACCEPT | Summary: Phylogenetic assertion that GPR50 is a functional G protein-coupled receptor. This is supported by direct evidence, although the activity observed so far is constitutive rather than agonist-driven. Reason: Confirmed experimentally. A cryo-EM structure plus TRUPATH BRET2 profiling of the ligand-free receptor shows productive engagement of Galpha12 with weaker Galphai/o coupling, and GPR50 knockout mice show a G12/13-RhoA output. The term is agnostic as to whether an agonist is known, so no endogenous ligand needs to be established for it to be correct. Supporting Evidence: PMID:41666959 We showed that GPR50 exhibits moderate constitutive activity through interaction with GΞ±12. PMID:41666959 By contrast to MT1 and MT2, which primarily interact with GΞ±i/o and inhibits cAMP pathway, GPR50 activates GΞ±12 but also moderately activates GΞ±i/o, raising the possibility that it is involved in the multisignaling pathways. |
| GO:0005886 plasma membrane | IBA GO_REF:0000033 | ACCEPT | Summary: GPR50 is a multi-pass plasma membrane receptor; this is where its G-protein coupling occurs. Reason: Correct and concordant with the IDA, EXP and IEA localisation annotations carried by this gene. |
| GO:0007186 G protein-coupled receptor signaling pathway | IBA GO_REF:0000033 | ACCEPT | Summary: GPR50 transduces signal through heterotrimeric G proteins, constitutively in the assays reported so far. Reason: Supported by BRET2 G-protein activation from the ligand-free receptor and by the G12/13-RhoA phenotype of GPR50 knockout cells. Supporting Evidence: PMID:40091563 Collectively, we show that GPR50 acts as an inhibitor of neurite growth and cell migration in the brain by activating the G12/13 protein-RhoA pathway. |
| GO:0004930 G protein-coupled receptor activity | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro signature-based assignment of generic GPCR activity. Reason: Correct in substance and concordant with the experimental evidence for constitutive G-protein coupling. Same action as the IBA and TAS rows for this term. |
| GO:0005634 nucleus | IEA GO_REF:0000044 | ACCEPT | Summary: Nuclear localisation, projected from the UniProt subcellular-location annotation, which itself refers specifically to the cleaved C-terminal domain. Reason: Correct but needs to be read carefully. The intact 7TM receptor is not nuclear; calpain-1 cleaves GPR50 between Phe-408 and Ser-409 and the released C-terminal domain translocates into the nucleus, where it engages the general transcription factor TFII-I/GTF2I. Same action as the EXP row for this term. |
| GO:0005886 plasma membrane | IEA GO_REF:0000120 | ACCEPT | Summary: Automated assignment of plasma membrane localisation. Reason: Correct primary location for the intact receptor. |
| GO:0007186 G protein-coupled receptor signaling pathway | IEA GO_REF:0000002 | ACCEPT | Summary: InterPro signature-based assignment of GPCR signalling. Reason: Correct; same action as the IBA and TAS rows for this term. |
| GO:0008502 melatonin receptor activity | IEA GO_REF:0000002 | REMOVE | Summary: An electronic assignment of melatonin receptor activity derived from the melatonin-receptor family signature InterPro:IPR000025. GPR50 is melatonin-related by sequence but does not bind melatonin, and this has been shown repeatedly and never contradicted. Reason: This is a family-signature mis-mapping that the primary literature directly contradicts. Three independent lines of evidence: (i) the paper that first cloned and expressed the receptor reported that it did not bind either iodinated or tritiated melatonin in COS-1 cells; (ii) Clement et al. showed by reciprocal chimeras that divergence of the second extracellular loop is the structural reason GPR50 is melatonin-unresponsive, and that restoring MT1 E2-loop residues confers melatonin responsiveness on GPR50 - i.e. wild-type GPR50 is melatonin-blind by design; (iii) the 2026 cryo-EM structure restates that melatonin does not bind GPR50 and shows that its putative pocket and ligand access channels differ from those of the melatonin receptors. GPR50 matches IPR000025 by descent, not by pharmacology. Because this is an unsupported electronic inference that can be argued against on direct biological grounds, REMOVE is the appropriate action rather than demotion. Note the irony that GPR50 does have a documented relationship to melatonin signalling, but an inhibitory one: in the GPR50-MT1 heterodimer it abolishes high-affinity agonist binding and G-protein coupling by MT1. Supporting Evidence: PMID:8647286 H9 transiently expressed in COS-1 cells did not bind [125I]melatonin or [3H]melatonin. PMID:28898928 showing how evolutionary processes appear to have selected for modifications in the E2 loop in order to make GPR50 unresponsive to melatonin. PMID:41666959 However, melatonin does not bind to GPR50 (Clement et al., 2017), and GPR50 is not directly involved in the melatonin-related pathways. PMID:40091563 GPR50 is an orphan GPCRs belonging to the melatonin receptor subfamily and represents 45% sequence homology with melatonin MT1 and MT2 receptors in humans but does not bind melatonin or any other known ligand |
| GO:0014069 postsynaptic density | IEA GO_REF:0000120 | ACCEPT | Summary: Automated transfer of the postsynaptic-density localisation that UniProt curates from the NOGO-A interaction study. Reason: Consistent with the experimental report of Gpr50 and neuronal Nogo-A enrichment at the synapse. Same action as the EXP row for this term. |
| GO:0016020 membrane | IEA GO_REF:0000002 | ACCEPT | Summary: Generic membrane localisation from the 7TM signatures. Reason: Correct, though uninformative next to the plasma membrane annotations this gene already carries. |
| GO:0005515 protein binding | IPI PMID:16778767 The orphan GPR50 receptor specifically inhibits MT1 melatoni... | MODIFY | Summary: Bare protein binding recording the GPR50-MT1 (MTNR1A) interaction. Reason: The interaction itself is real and important - GPR50 heterodimerizes constitutively with MT1 and abolishes high-affinity agonist binding and G-protein coupling to the MT1 protomer - but the bare protein binding term carries no functional information. The content is better captured as a signalling receptor inhibitor activity acting in a GPCR heterodimeric complex, which is recorded in core_functions. Replaced by GO:0030547 signaling receptor inhibitor activity, which states what the interaction accomplishes. Proposed replacements: signaling receptor inhibitor activity Supporting Evidence: PMID:16778767 Whereas the association between GPR50 and MT(2) did not modify MT(2) function, GPR50 abolished high-affinity agonist binding and G protein coupling to the MT(1) protomer engaged in the heterodimer. |
| GO:0005515 protein binding | IPI PMID:25416956 A proteome-scale map of the human interactome network. | REMOVE | Summary: Bare protein binding from a proteome-scale yeast two-hybrid interactome map (partner ZNF572). Reason: Uninformative as a molecular function, and derived from a high-throughput screen with no follow-up for this pair. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005515 protein binding | IPI PMID:32405532 GPR50 Promotes Hepatocellular Carcinoma Progression via the ... | REMOVE | Summary: Bare protein binding recording the GPR50-ADAM17 interaction in hepatocellular carcinoma. Reason: Uninformative as a molecular function. The underlying observation, that GPR50 supports ligand-independent Notch activation through ADAM17, would be better captured as a biological-process annotation if it is replicated. No more informative molecular function is supportable from this evidence, so the row is removed rather than rewritten. Removal does not imply the reported interaction is false. |
| GO:0005886 plasma membrane | IDA GO_REF:0000052 | ACCEPT | Summary: Human Protein Atlas immunofluorescence localisation to the plasma membrane. Reason: Consistent with all other localisation evidence for this receptor. |
| GO:0005634 nucleus | EXP PMID:31900622 GPR50-Ctail cleavage and nuclear translocation: a new signal... | ACCEPT | Summary: Experimental demonstration that the calpain-released C-terminal domain of GPR50 translocates to the nucleus and regulates transcription with TFII-I/GTF2I. Reason: Well supported and mechanistically specific. The annotation refers to the cleaved cytoplasmic fragment rather than to the intact seven-transmembrane receptor, which is a genuinely unusual signal-transduction mode for a GPCR. Supporting Evidence: PMID:31900622 Activation of the calcium-dependent calpain protease cleaves off the CTD from the transmembrane-bound GPR50 core domain between Phe-408 and Ser-409 as determined by MALDI-TOF-mass spectrometry. |
| GO:0005886 plasma membrane | EXP PMID:21858214 GPR50 interacts with TIP60 to modulate glucocorticoid recept... | ACCEPT | Summary: Experimental localisation of full-length GPR50 to the cell membrane. Reason: Correct primary location. |
| GO:0014069 postsynaptic density | EXP PMID:19699797 GPR50 interacts with neuronal NOGO-A and affects neurite out... | ACCEPT | Summary: Gpr50 and neuronal Nogo-A are both enriched at the synapse. Reason: Experimentally supported, and consistent with the receptor's role in restraining neurite outgrowth. Supporting Evidence: PMID:19699797 We confirmed the interaction in mammalian cells and found an enrichment of both Gpr50 and neuronal Nogo-A at the synapse. |
| GO:0004930 G protein-coupled receptor activity | TAS PMID:8647286 Cloning of a melatonin-related receptor from human pituitary... | ACCEPT | Summary: Author statement from the 1996 cloning paper that H9/GPR50 is a G protein-coupled receptor. Reason: Correct by sequence and now confirmed functionally. Note that the same paper explicitly reported the absence of melatonin binding, so it does not support the melatonin receptor activity annotation this gene also carries. |
| GO:0005886 plasma membrane | TAS PMID:8647286 Cloning of a melatonin-related receptor from human pituitary... | ACCEPT | Summary: Author statement of plasma membrane localisation for the cloned receptor. Reason: Correct, and independently confirmed by later experimental work. |
| GO:0007186 G protein-coupled receptor signaling pathway | TAS PMID:8647286 Cloning of a melatonin-related receptor from human pituitary... | ACCEPT | Summary: Author statement that the cloned receptor signals as a GPCR. Reason: Correct; same action as the IBA and IEA rows for this term. |
| GO:0007267 cell-cell signaling | TAS PMID:8647286 Cloning of a melatonin-related receptor from human pituitary... | MARK AS OVER ANNOTATED | Summary: Very general process term taken from the closing speculation of the 1996 cloning paper that the receptor and its unknown natural ligand are involved in neuroendocrine function. Reason: Not wrong, but it rests on a stated inference from expression pattern rather than on any assay, and it is far too general to be informative. Thirty years later the ligand referred to in that speculation is still not agreed on. |
| GO:0045744 negative regulation of G protein-coupled receptor signaling pathway | IDA PMID:16778767 The orphan GPR50 receptor specifically inhibits MT1 melatoni... | NEW | Summary: Proposed new biological process. By suppressing agonist binding and G-protein coupling of the MT1 protomer, GPR50 negatively regulates melatonin receptor signalling. Reason: The process consequence of the receptor-inhibitor activity proposed above. Supporting Evidence: PMID:16778767 Deletion of the large C-terminal tail of GPR50 suppressed the inhibitory effect of GPR50 on MT(1) without affecting heterodimerization, indicating that this domain regulates the interaction of regulatory proteins to MT(1). |
| GO:0038039 G protein-coupled receptor heterodimeric complex | IDA PMID:16778767 The orphan GPR50 receptor specifically inhibits MT1 melatoni... | NEW | Summary: Proposed new cellular component. GPR50 heterodimerizes constitutively and specifically with the MT1 and MT2 melatonin receptors, shown by both biochemical and biophysical (BRET) approaches in intact cells. Reason: The complex in which the receptor-inhibitor activity is exercised; more informative than the bare protein binding annotation currently carried. Supporting Evidence: PMID:16778767 We show that GPR50, an orphan GPCR, heterodimerizes constitutively and specifically with MT(1) and MT(2) melatonin receptors, using biochemical and biophysical approaches in intact cells. |
| GO:0035025 positive regulation of Rho protein signal transduction | ISO PMID:40091563 Orphan GPR50 Restrains Neurite Outgrowth and Cell Migration ... | NEW | Summary: Proposed new biological process. GPR50 knockout mice show that the receptor acts through the G12/13 protein-RhoA cassette in neural progenitor cells and tanycytes, in a manner similar to but independent of Nogo-A and its receptors. Reason: This is the transducer arm that the existing generic GPCR-signalling annotations leave unspecified, and it is the output that the ligand-free structure also points to. The term is the positive-regulation child rather than GO:0007266 itself, because the receptor sits upstream of the cassette and activates it - the source's own title says GPR50 restrains neurite outgrowth "by Activating the G(12/13) Protein-RhoA Pathway". Evidence is ISO rather than IMP because the knockout is in mouse, so the transfer to the human gene is by orthology. Supporting Evidence: PMID:40091563 Inhibition of neurite growth by GPR50 occurs through activation of the G12/13 protein-RhoA pathway in a manner similar to, but independent of Nogo-A and its receptors. |
| GO:0010977 negative regulation of neuron projection development | ISO PMID:40091563 Orphan GPR50 Restrains Neurite Outgrowth and Cell Migration ... | NEW | Summary: Proposed new biological process. Loss of GPR50 increases neurite outgrowth and cell motility in isolated neural progenitor cells and hypothalamic tanycytes, so the receptor normally restrains neurite growth. Reason: Loss-of-function evidence in mice for the physiological consequence of constitutive G12/13-RhoA signalling. Note that an earlier overexpression study reported the opposite sign; the knockout direction is the more reliable one. Supporting Evidence: PMID:40091563 Here, we generated GPR50 knockout (GPR50-KO) mice and show that the absence of GPR50 increases neurite outgrowth, cell motility and migration of isolated neural progenitor cells (NPCs) and hypothalamic radial glial cells (tanycytes). |
| GO:0006357 regulation of transcription by RNA polymerase II | IDA PMID:31900622 GPR50-Ctail cleavage and nuclear translocation: a new signal... | NEW | Summary: Proposed new biological process. The calpain-released C-terminal domain of GPR50 enters the nucleus and acts with the general transcription factor TFII-I/GTF2I to regulate c-fos transcription, with RNA-seq indicating a broader transcriptional role. Reason: GOA already records the nuclear localisation of this fragment but not what it does there, leaving the most distinctive feature of the protein unannotated. Supporting Evidence: PMID:31900622 The cytosolic CTD then translocates into the nucleus assisted by its 'DPD' motif, where it interacts with the general transcription factor TFII-I to regulate c-fos gene transcription. |
Loading supporting contentβ¦
Download this section (compressed HTML)Q: Is Little-LEN (L-LEN), the ProSAAS-derived 10-mer proposed in 2026, the endogenous agonist of GPR50? The claim rests on a single laboratory using one method (genetically encoded photo-cross-linking from hypothalamic extract), and a cryo-EM study published five weeks later states that no endogenous agonist has been characterized without citing or engaging the L-LEN result. Independent replication in a second laboratory, ideally with a binding assay orthogonal to cross-linking, is needed before a ligand-specific molecular function term is justified.
Q: Which G protein does GPR50 actually use in vivo? The L-LEN study reports dominant Galphai coupling with pertussis-toxin-sensitive inhibition of forskolin-evoked cAMP, whereas both the cryo-EM/BRET2 study and the GPR50 knockout mouse work place the principal output on Galpha12/13-RhoA. These have not been reconciled in a common system, and they imply different downstream biology (metabolic cAMP signalling versus cytoskeletal growth restraint).
Q: Should InterPro:IPR000025 (melatonin receptor family) continue to generate a GO:0008502 melatonin receptor activity mapping for GPR50? The signature is a family signature and GPR50 is a documented non-binder; the mapping should either be blocked for this protein or the InterPro2GO rule narrowed to the signatures that actually imply melatonin binding.
Q: Is the constitutive activity seen in heterologous cells a property of the receptor itself or of overexpression? Distinguishing true agonist-independent activity from mass-action coupling matters for whether GPR50 should be modelled as a constitutively active receptor or as a receptor whose agonist has simply not been found.
Q: Does the calpain-released C-terminal domain act independently of the membrane-bound core, and is the proportion of receptor that is cleaved under physiological calcium large enough for the nuclear route to be a major output?
Experiment: Replicate the L-LEN pairing in an independent laboratory using assays that do not depend on covalent capture - saturation and competition binding with labelled L-LEN on membranes from GPR50-expressing and GPR50-null cells, plus a full TRUPATH G-protein panel comparing basal and L-LEN-stimulated coupling side by side for Galphai1/2/3, Galpha12 and Galpha13 in the same cells.
Hypothesis: L-LEN is a genuine endogenous agonist of GPR50.
Type: pharmacology
Experiment: In a single cell background, measure GPR50 coupling to Galphai and to Galpha12/13 with and without L-LEN, with and without pertussis toxin, and with the ECL2/TM6 variants that the chimera work identified as activation-critical, to see whether the ligand switches the receptor between transducers.
Hypothesis: The Galphai (L-LEN) and Galpha12/13 (constitutive) readouts are both real and reflect ligand-directed coupling rather than a discrepancy between laboratories.
Type: signalling
Experiment: Repeat the negative melatonin binding result on the modern reagents - saturation binding with 2-iodomelatonin on purified GPR50 in nanodiscs, alongside MT1 as a positive control, and a thermal-shift or mass-photometry ligand-binding readout. A clean, quantitative modern negative would let the InterPro2GO mapping be retired with a citable reference rather than a chain of assertions.
Hypothesis: GPR50 does not bind melatonin under any condition relevant to the IEA annotation.
Type: biochemistry
Experiment: In GPR50 knockout mice, test whether the neurite-outgrowth and tanycyte-migration phenotypes are rescued by wild-type GPR50 but not by a Galpha12/13-uncoupled receptor variant, and whether they are insensitive to loss of L-LEN/ProSAAS.
Hypothesis: The G12/13-RhoA growth-restraining function is the physiological core function.
Type: mouse genetics
Loading supporting contentβ¦
Download this section (compressed HTML)Loading supporting contentβ¦
Download this section (compressed HTML)