Figure 1.
Gα as a regulated molecular signaling nexus.
This graphic of the Gα signaling nexus delineates functional elements within the molecule such as nucleotide binding (e.g. NKxD motif, TCAT motif, P-loop) and GPCR-driven nucleotide exchange (α-helix 5, β-strand 6), the different conformations of Gα (i.e. transition and GDP-, GTP-bound states), along with mammalian macromolecules that have been reported to directly interact with Gα. Reported interactions are classified by nucleotide dependence and by functional outcome (GDI, GEF, etc.). The single-headed arrow represents an interaction leading to signaling, the ‘X’ represents an interaction that does not lead to signaling, a blunt arrow represents interactions leading to signal termination, while the double-headed arrow represents a neutral physical interaction. While the list of reported interactions is intended to be extensive, it is not intended to be exhaustive, particularly in regard to the GPCRs. An expanded figure legend with additional references is in Text S1.
Figure 2.
Evolution of the molecular signaling nexus Gα.
A heuristic tree that captures commonly accepted characteristics between taxa and Gα and Gβ subunits from representative genomes is shown to the left. Plants have a single Gα subunit and the two fungi have either 2 or 3 Gα that do not fit any of the 4 animal subtypes. Thus, the evolution of the Gα subunit from a single gene to the multigene family evident in mammals occurs within eukaryotes. Divergent Gα subunits found in some genomes are not included in this diagram. Homologs of well-studied proteins that interact with mammalian Gα subunits are indicated to the right. For incomplete genomes, the presence of the interactor may be indeterminate and is indicated by a ‘?’. Homology indicates the presence of a protein in the given organism, but not all interactions have been verified in lower metazoans. As is evident from the chart, plant Gα subunits contain a single known interacting protein, the Gβγ heterodimer. The number of interacting proteins grew steadily throughout evolution of the GPCR signaling system. No bootstrap or other credibility scores are shown for the heuristic tree as this is not intended to be a definitive phylogeny. Marine sponge = Geodia cydonium; Freshwater sponge = Ephydatia fluviatilis; Worm = Caenorhabditis elegans; Fruitfly = Drosophila melanogaster; Human = Homo sapiens.
Figure 3.
Alignment of select human Gα subtypes highlighting invariant and class-distinctive sites.
Invariant residues are conserved across all 4 Gα classes (INV: colored dark gray) while class-distinctive sites are conserved across 3 of the 4 Gα classes to a non-distinctive (η: colored light gray) amino acid value. At class-distinctive sites, distinctive (∂) amino acid values are allowed in the remaining class but are not required to be absolutely conserved within all sequences in the distinctive Gα class, thus allowing for subclass variation at that site. Some sites are identified as class-distinctive based on variation in a single non-human sequence. See Table S2 to identify sequences where ∂ occurs. d sites lie within 5 Å of a distinctive site but are conserved in 2 classes (see Table S2 and Table S3 for summaries). ∂ amino acid values are colored according to Gα class and noted above the alignment: ‘I’ = G(io) site (green); ‘Q’ = G(q) site (magenta); ‘S’ = G(s) site (blue); ‘2’ = G(12) site (yellow orange) and ‘d’ = d site. Functional regions are indicated below the alignment, including regions important for coupling to the receptor (GPCR), guanine-nucleotide-dependent conformational change (switches I, II, III) and nucleotide binding (P-loop, NKxD, TCAT). Also noted below the alignment (‘*’) are distinctive sites discussed in more detail in results. Distinctive sites for all 4 Gα classes were defined using 58 mammalian Gα sequences from 14 subtypes and a reduced amino acid alphabet (Materials and Methods).
Figure 4.
G(q) class-distinctive sites in structural context.
(A) The RH domain of GRK2, shown as a sand colored cartoon display, in complex with activated Gαi/q•GDP•Mg2+•AlF4− (PDB ID 2BCJ). In all structural panels in this figure, Gαi/q is shown as spheres with core residues colored gray if the residues are conserved between Gα subunits (either INV (invariant) or η (non-distinct) amino acids) while G(q)-distinctive sites are colored hot pink only if they contain a ∂ (distinct) amino acid. Non-core residues and d sites are colored white. G(q)-distinctive sites are numbered according to their position in the signature sequence (see panel (D)). (B) The DH and PH domains of p63RhoGEF, in a teal colored cartoon and surface display, binds to activated Gαi/q•GDP•Mg2+•AlF4− (PDB ID 2RGN). Gαi/q is in the same orientation as panel A. (C) Homology model of Gαq•GDP (sphere display) bound to Gβ•Gγ (deepblue/copper cartoon) heterodimer. Two orientations related by a 180° rotation about the vertical axis are shown. (D) Signature sequences are formed by grouping all distinctive sites for a given class together, removing all residues between individual distinctive sites of the noted class. The distinctive sites for each class are presented in order from the N-terminus to the C-terminus and numbered accordingly. Amino acids that correspond to the ∂ values at the G(q) site are colored hot pink. Sites that interact with GRK2 are denoted by ‘G’ above the site, while sites that are buried and not visible are denoted by ‘b’ above the site.
Figure 5.
G(12) class-distinctive sites in structural context.
(A) The structure of the p115RhoGEF RGS-like box domain (dark teal cartoon) and a βN-αN hairpin element (cyan loop cartoon) bound to an activated Gα13/i1 chimera (Gα13/i1•GDP•Mg2+•AlF4−) (PDB ID 1SHZ). In all structural panels in this figure, Gα is shown as spheres with core residues colored gray if they are conserved between Gα subunits (either INV (invariant) or η (non-distinct) amino acids) while G(12)-distinctive sites are colored yellow orange only if they contain a ∂ (distinct) amino acid. The chimeric Gα subunit in this structure also contained ∂ (distinct) amino acids at several G(io)-distinctive sites (green spheres). Non-core residues and d sites are colored white. G(12)-distinctive sites are numbered according to their position in the signature sequence (see panel (D)). (B) Model of Gα12/i1 in complex with p115RhoGEF. Gα is in the same orientation as panel A. (C) Homology model of Gα12•GDP (sphere display) bound to Gβ•Gγ (deep blue/copper cartoon) heterodimer. Two orientations related by a 180° rotation about the vertical axis are shown. The inset is a close up view of the Gβγ binding region in the right view. (D) Signature sequences are formed by grouping all distinctive sites for a given class together, removing all residues between individual distinctive sites of the noted class. The distinctive sites for each class are presented in order from the N-terminus to the C-terminus and numbered accordingly. Amino acids that correspond to the ∂ values at the G(12) site are colored yellow orange. Sites that have direct interactions with p115RhoGEF are denoted by ‘R’ above the site, while additional sites in switches I or II are denoted by ‘1’ or ‘2’, respectively, above the site.
Figure 6.
G(s) class-distinctive sites in structural context.
(A) The structure of the catalytic domains of adenylyl cyclase (VC1 in sand cartoon, IIC2 in purple cartoon) bound to an activated Gαs (Gαs•GTPγS) (PDB ID 1AZS). In all structural panels in this figure, Gαs is shown as spheres or cartoon with core residues colored gray if they are conserved between Gα subunits (either INV (invariant) or η (non-distinct) amino acids) while G(s)-distinctive sites are colored blue only if they contain a ∂ amino acid. Non-core residues and d sites are colored white. G(s)-distinctive sites are numbered according to their position in the signature sequence (see panel (D)). (B) Superimposition of Gαs (light gray cartoon with ∂ amino acids at G(s) sites of interest rendered as blue sticks) and Gαi1 (PDB ID 1GIA in light green cartoon with corresponding η amino acids at G(s) sites in sticks and colored gray) highlighting sequence and backbone conformational changes in switch II (“sw II”) and loops near the adenylyl cyclase interface. The two views are related by a 90° rotation about the vertical axis. VC1 is in sand spheres and IIC2 is in purple spheres. G(s) site 11 lies in a loop neighboring switch II that forms part of the binding interface (“neigh”) while G(s) site 13 is in a loop that abuts the binding interface (“abut”). (C) Homology model of Gαs•GDP (sphere display) bound to Gβ•Gγ (deepblue/copper cartoon) heterodimer. Two orientations related by a 180° rotation about the vertical axis are shown. (D) Signature sequences are formed by grouping all distinctive sites for a given class together, removing all residues between individual distinctive sites of the noted class. The distinctive sites for each class are presented in order from the N-terminus to the C-terminus and numbered accordingly. Amino acids that correspond to the ∂ values at the G(s) site are colored blue. Sites that have been proposed to be important to the interaction with adenylyl cyclase are denoted by ‘A’ above the site, while additional sites in switches II or III are denoted by ‘2’ or ‘3’, respectively, above the site.
Figure 7.
G(io) class-distinctive sites in structural context.
(A) The structure of the GoLoco domain of RGS14 (blue cartoon) bound to an inactive Gαi1 (Gαi1• GDP) (PDB ID 1KJY). In all structural panels in this figure, Gαi1 is shown as spheres or cartoon with core residues colored gray if they are conserved between Gα subunits (either INV (invariant) or η (non-distinct) amino acids) while G(io)-distinctive sites are colored green only if they contain a ∂ (distinct) amino acid. Non-core residues and d sites are colored white. G(io)-distinctive sites are numbered according to their position in the signature sequence (see panel (D)). Panel A shows two views of Gαi1 related by a 180° rotation about the horizontal axis. The left view is of the switch region of Gαi1 while the right view is of the top of the subunit. (B) Closeup view of β-strand 6 and α-helix 5 from Gαi1 with the side chains of G(io)-distinctive residues in a stick rendering. The orientation is the same as the right-hand view in panel A. Helix 5 rotates and translates toward β-strand 6 (arrow) during GPCR-mediated activation of the Gα subunit. Sites 12 and 13 in Gαi1 are the ∂ residue but show subtype variation within the G(io) class. G(io) site 11 (colored lime green) lies in β-strand 6 and also shows subtype variation; Gαi1 possesses the η residue at that site and is, therefore, colored gray in (A). (C) Structure of Gαi1•GDP (sphere display) bound to Gβ•Gγ (deepblue/copper cartoon) heterodimer (PDB ID 1GP2). Two orientations related by a 180° rotation about the vertical axis are shown. (D) Signature sequences are formed by grouping all distinctive sites for a given class together, removing all residues between individual distinctive sites of the noted class. The distinctive sites for each class are presented in order from the N-terminus to the C-terminus and numbered accordingly. Amino acids that correspond to the ∂ values at the G(io) site are colored green. Sites in helix 5 that have been proposed to be important for coupling to the GPCR are denoted by ‘5’ above the site, while site 11 in strand 6 is denoted by ‘b’ above the site.
Figure 8.
Class-distinctive signature sequences of Gα family members from select organisms.
Signature sequences from select organisms are used to follow the evolution of class-distinctive sites (also see Figure 2). Each panel reflects the evolutionary history of a subtype: (A) Gαi1, (B) Gαq, (C) Gαs, (D) both Gα12 and Gα13. Signature sequences are formed by grouping all distinctive sites for a given class together, removing all residues between individual distinctive sites of the noted class. The distinctive sites for each class are presented in order from the N-terminus to the C-terminus and numbered accordingly. Amino acids that correspond to the ∂ values at that site are colored according to distinctive class: green = G(io); magenta = G(q); blue = G(s); and yellow orange = G(12). Class-distinctive sites were determined using only mammalian sequences. Occasionally a non-mammalian subunit will contain a ∂ or variable (white) amino acid where only η amino acids were observed in the mammalian sequences (for comparison see Figures 4D, 5D, 6D, 7D). (A) Class-distinctive signature sequences of G(io) family members from select organisms. Sites in helix 5 that have been proposed to be important for coupling to the GPCR are denoted by ‘5’ above the site, while site 11 in strand 6 is denoted by ‘b’ above the site. (B) Class-distinctive signature sequences of G(q) family members from select organisms. Sites that interact with GRK2 are denoted by ‘G’ above the site, while sites that are buried and not visible are denoted by ‘b’ above the site. (C) Class-distinctive signature sequences of G(s) family members from select organisms. Sites that have been proposed to be important for the interaction with adenylyl cyclase are denoted by ‘A’ above the site, while additional sites in switches II or III are denoted by ‘2’ or ‘3’, respectively, above the site. (D) Class-distinctive signature sequences of G(12) family members from select organisms. Sites that have direct interactions with p115RhoGEF are denoted by ‘R’ above the site, while additional sites in switches I or II are denoted by ‘1’ or ‘2’, respectively, above the site. Marine sponge = Geodia cydonium; Freshwater sponge = Ephydatia fluviatilis; Worm = Caenorhabditis elegans; Fruitfly = Drosophila melanogaster; Sea urchin = Strongylocentrotus purpuratus; Frog = Xenopus laevis; Human = Homo sapiens.