Disentangling autoproteolytic cleavage from tethered agonist-dependent activation of the adhesion receptor ADGRL3

Abstract

Adhesion G protein-coupled receptor latrophilin 3 (ADGRL3), a cell adhesion molecule highly expressed in the central nervous system, acts in synapse formation through trans interactions with its ligands. It is largely unknown if these interactions serve a purely adhesive function or can modulate G protein signaling. To assess how different structural elements of ADGRL3 (e.g., the adhesive domains, autoproteolytic cleavage site, or tethered agonist (TA)) impact receptor function, we require constructs that disrupt specific receptor features without impacting others. While we showed previously that mutating conserved Phe and Met residues in the TA of ADGRL3-C-terminal fragment (CTF), a CTF truncated to the G protein-coupled receptor proteolysis site, abolishes receptor-mediated G protein activation, we now find that autoproteolytic cleavage is disrupted in the full-length version of this construct. To identify a construct that disrupts TA-dependent activity without impacting proteolysis, we explored other mutations in the TA. We found that mutating the sixth and seventh residues of the TA, Leu and Met, to Ala impaired activity in a serum response element activity assay for both full-length and CTF constructs. We confirmed this activity loss results from impaired G protein coupling using an assay that acutely exposes the TA through controlled proteolysis. The ADGRL3 mutant expresses normally at the cell surface, and immunoblotting shows that it undergoes normal autoproteolysis. Thus, we found a construct that disrupts tethered agonism while retaining autoproteolytic cleavage, providing a tool to disentangle these functions in vivo. Our approach and specific findings are likely to be broadly applicable to other adhesion receptors.

Keywords: G protein; adhesion G protein-coupled receptor latrophilin; adhesion G protein-coupled receptors; autoproteolysis; cell signaling; membrane protein; tethered agonist.

Figure 1

The GPCR proteolytic site undergoes autoproteolytic cleavage to release the tethered agonist and facilitate receptor activation.A, cartoon representation of full-length (FL) ADGRL3. The N-terminal fragment (NTF) of the receptor is comprised of rhamnose-binding lectin (RBL) and olfactomedin (OLF) domains, a serine/threonine-rich region, and a hormone receptor motif (HRM). Proteolysis occurs within the GPCR autoproteolysis-inducing domain (GAIN) at the GPCR proteolytic site (GPS). Cleavage occurs between HL^P1 and T^P1’, resulting in exposure of the TA peptide. The C-terminal fragment (CTF) of the receptor is composed of a transmembrane GPCR fold (7 TM) that signals through heterotrimeric G proteins. B–C, construct design for the ADGRL3 mutants tested in this study. D, anticipated functional outcomes for the ADGRL3 mutants tested in this study. The upper denotation represents the receptor’s ability to undergo autoproteolytic cleavage, whereas the lower denotation represents TA-mediated receptor activation. ADGRL3, adhesion G protein-coupled receptor latrophilin 3; GPCR, G protein-coupled receptor; TA, tethered agonist; 7 TM, seven transmembrane.

Figure 2

SRE-luciferase activity is disrupted in ADGRL3-L^P6’A/M^P7’A.A, schematic of the dual-glo serum response element (SRE)-luciferase reporter assay. To activate ADGRL3, the tethered agonist (TA) binds to the orthosteric pocket of the receptor, resulting in the release of Gα and Gβᵧ subunits from the heterotrimer. This release initiates a cascade of downstream second messenger pathways and eventual gene transcription of firefly luciferase by SRE. As an internal control, Renilla luciferase is expressed downstream of the constitutive promoter CMV. B, SRE-luciferase assay for full-length ADGRL3 and mutant constructs ADGRL3-F^P3’A/M^P7’A, ADGRL3-T^P1’G, ADGRL3-L^P6’A, and ADGRL3-L^P6’A/M^P7’A. C, SRE-luciferase assay for ADGRL3-CTF and mutant constructs ADGRL3-F^P3’A/M^P7’A-CTF, ADGRL3-T^P1’G-CTF, ADGRL3-L^P6’A-CTF, and ADGRL3-L^P6’A/M^P7’A-CTF. Data in (B) and (C) are expressed relative to the absence of receptor (control). Statistics were calculated using the two-way ANOVA with Tukey’s multiple comparison test. The mean of the WT receptor control at 200 ng or 600 ng was compared to the mean of the corresponding mutant receptors (∗p < 0.1; ∗∗∗∗p < 0.0001). ADGRL3, adhesion G protein-coupled receptor latrophilin 3; CTF, C-terminal fragment.

Figure 3

ADGRL3-L^P6’A/M^P7’A cannot activate Gα₁₃.A, schematic of the Gβγ release bioluminescence energy resonance transfer (BRET) assay. HEK293 cells with targeted deletion of all G proteins (12) were transfected with ADGRL3 cDNA, Gα13, Gβ1, Gγ2-Venus, membrane-anchored GRK3ct-Rluc8, and empty vector pCDNA5/FRT to balance. The protease activatable ADGRL3 constructs contain an ADGRL3 signal peptide, followed by a SNAP-tag, flexible linker, Flag tag, and the ADGRL3-CTF. Upon addition of 5.5 units of enterokinase, the construct is cleaved to acutely expose the tethered agonist and activate G protein. B, Gβγ release testing ADGRL3-CTF, ADGRL3-L^P6’A-CTF, and ADGRL3-L^P6’A/M^P7’A-CTF activation of Gα₁₃. Statistics were calculated using ordinary one-way ANOVA with Sidak’s multiple comparison test to compare each cell mean with the other cell mean in that column. This analysis was then followed by multiple unpaired t tests to compare the WT receptor to the two mutant constructs (∗∗∗p < 0.001; ∗∗∗∗p < 0.0001). C, cartoon representation of the full-length (FL) ADGRL3 receptor labeled with impermeant Janelia Fluor 646. D, fluorescent counts for FL ADGRL3, SNAP-ADGRL3, and SNAP-ADGRL3-L^P6’A/M^P7’A. E, fluorescent counts for FL ADGRL3, SNAP-ADGRL3-CTF, and SNAP- ADGRL3-L^P6’A/M^P7’A-CTF. The same negative control values were used for FL and CTF expression comparisons. Statistics for SNAP-tag labeling were calculated using ordinary one-way ANOVA with Tukey’s multiple comparison test (ns, not significant; ∗∗p < 0.01; ∗∗∗p < 0.001; ∗∗∗∗p < 0.0001). ADGRL3, adhesion G protein-coupled receptor latrophilin 3; CTF, C-terminal fragment.

Figure 4

ADGRL3-L^P6’A/M^P7’A retains normal autoproteolytic cleavage.A, expected molecular weights for the uncleaved and cleaved receptors are 172.8 kDa and 70.5 kDa, respectively. B, representative immunoblot against primary Flag (1:500, Thermo Fisher Scientific, PA1-984B) and secondary anti-rabbit IgG-HRP (1:10,000, Thermo Fisher Scientific, Cat #31458). ADGRL3, adhesion G protein-coupled receptor latrophilin 3.

Figure 5

The L^P6’A/M^P7’A mutation weakens the interactions of the TA in the binding pocket.A, the molecular dynamics (MD) simulation system of the human ADGRL3/G protein complex embedded in a lipid bilayer. The receptor, G protein, and lipid bilayer are colored in gray, gold, and cyan, respectively. The tethered agonist (TA) is highlighted with dark gray. B–C, the resulting conformations of the TA and several key residues in the binding pocket of ADGRL3 (cyan) and ADGRL3-L^P6’A/M^P7’A (magenta), respectively. The arrows in (C) indicate the rearrangements of these residues. D–E, scatter plots of the distances between residues P7′-F995 and P6′-F1092, and the χ1 and χ2 dihedral angles of W1158, respectively, showing the upward movement of the TA in the binding pocket. F, SRE-luciferase assay for ADGRL3-CTF-nluc and mutant construct ADGRL3-W1158A-CTF-nluc. Statistics were calculated using an unpaired t test (∗∗∗p < 0.001). G, Gβγ release testing ADGRL3-CTF and ADGRL3-W1158A-CTF activation of Gα₁₃. Statistics were calculated using an unpaired t test (∗∗∗p < 0.001; ∗∗∗∗p < 0.0001). H, the structure of the ADGRL1 GAIN domain (PDB 4DLQ) (5), with the TA highlighted with dark gray. I, a zoom-in view of the autoproteolysis site indicated by the dotted box in panel H. F^P3’ located at the turn between the last β-strand of the GPS and the TA is expected to stabilize the conformation necessary for the autoproteolysis occurring between L^P1 and T^P1’; T^P1’ also forms a hydrophobic-aromatic interaction with the F^P3’, indicated by the cyan space-filling representation. The “∗” indicates the location of autoproteolysis. Orange and green residues are those within 5 Å of F^P3’ and L^P6’, respectively. ADGRL3, adhesion G protein-coupled receptor latrophilin 3; CTF, C-terminal fragment; GAIN, GPCR autoproteolysis-inducing domain; GPS, GPCR proteolysis site; SRE, serum response element.