Development of an antibody fragment that stabilizes GPCR/G-protein complexes

Abstract

Single-particle cryo-electron microscopy (cryo-EM) has recently enabled high-resolution structure determination of numerous biological macromolecular complexes. Despite this progress, the application of high-resolution cryo-EM to G protein coupled receptors (GPCRs) in complex with heterotrimeric G proteins remains challenging, owning to both the relative small size and the limited stability of these assemblies. Here we describe the development of antibody fragments that bind and stabilize GPCR-G protein complexes for the application of high-resolution cryo-EM. One antibody in particular, mAb16, stabilizes GPCR/G-protein complexes by recognizing an interface between Gα and Gβγ subunits in the heterotrimer, and confers resistance to GTPγS-triggered dissociation. The unique recognition mode of this antibody makes it possible to transfer its binding and stabilizing effect to other G-protein subtypes through minimal protein engineering. This antibody fragment is thus a broadly applicable tool for structural studies of GPCR/G-protein complexes.

Fig. 1

Isolation of mAb16 and its binding profile to each component. a Analytical SEC of rhodopsin/G_i1 with each antibody. Rhodopsin/G_i1 runs at 8.2 mL and each mAb alone runs 8.4–9 mL (c). Rhodopsin/G_i1 bound to mAb makes higher molecular weight product and migrates at the elution volume of 6–7 mL depending on the mAb. b Analytical SEC of rhodopsin/G_i1 with each antibody following to GTPγS treatment. Intact complex remains at 6–7 mL only in the mAb16 condition. c Analytical SEC of individual component of rhodopsin/G_i1 or heterotrimeric G_i1 with each antibody. Top left: Binding experiment with Gαi1 subunit and each mAb. The peak of Gαi1 at 11 mL stays intact indicating there is no binding with each mAb. Top right: Binding experiment with opsin. Both mAb peaks and Opsin peak (at 9.5 mL) stays intact. Bottom left: Binding experiment with Gβγ subunit. The peak of Gβγ at 11.2 mL disappears upon incubating with mAbs except mAb16 and each mAb peak shifts towards left compared to the ones with Gαi1 or opsin indicating those mAbs recognize Gβγ subunit as an epitope. Bottom right: Binding experiment with heterotrimeric Gi1

Fig. 2

Crystal structure of G_i1/scFv16 and characterization of Fab16. a Overall structure of Gi1/scFv16 complex. Cartoon representation with Gα_i1 in gold, Gβ in cyan, Gγ in magenta, scFv-heavy chain in light grey and scFv-light chain in light blue. b Superposition of G_i1/scFv16 structure onto G_i1 (PDB: 1GP2) based on alignment of Gβγ subunits. Gα_i1 (1GP2) in grey and G_i1/scFv16 in the same colour code as in a. For clarity, Gβγ subunits and scFv16 is shown as transparent cartoon. Arrows show a slight rotational displacement of Gα_i1 towards Gβ₁ compared to Gi1 alone. Additional interactions are formed between switch I and switch II of Gα_i1 and Gβ₁. c Interaction between G_i1 and scFv16. The residues participating in the interactions are depicted with stick models in the expanded panels. Residue numbers are shown with Common Gα Numbering (CGN) code for Gα_i1

Fig. 3

Sequence alignment of G-protein family members and binding profile of Fab16. a Multiple sequence alignment of amino-termini of representative Gα subunits from human. UniProt numbers are provided after each G-protein subtype name. Secondary structures are shown as cylinder (helix) and arrow (strand). The asterisks indicate the residues in contact with scFv16 in Gα_i1 and those corresponding residues are coloured according to their property: Positive in blue, negative in red, hydrophobic in green, polar in purple, cysteine in yellow. b Fluorescent SEC analysis of binding of the fluorescently labelled Fab16 with G-protein family members. c Analytical tryptophane fluorescent SEC of μOR/G_i1 and M₂R/G_oA with GTPγS in the presence or absence of Fab16. Each complex alone runs around 12.2 mL. Upon binding to Fab16, they run at 11.4 mL or 11.6 mL indicating the binding of Fab16 to these GPCR/G-protein complexes. Excess free Fab16 runs at 16.1 mL. Dissociated components upon incubating with GTPγS show smaller peaks at 13.5–16 mL

Fig. 4

Nucleotide-binding kinetics. a Influence of Fab16 on the nucleotide-binding kinetics of the purified M₂R/G_oA complex. Nucleotide binding was monitored by using BODIPY-FL-GTPγS or BODIPY-FL-GDP. b Influence of Fab16 on the nucleotide release kinetics of G_i1. GDP release was monitored by BODIPY-FL-GTPγS- or BODIPY-FL-GDP-binding kinetics under conditions where GDP release is the rate-limiting step. c Nucleotide binding to β2AR/G_s complex in the presence or absence of Nb35. The curves represent the mean ± standard error of three experiments

Fig. 5

Generation of chimeric G-proteins. a Alignment of the αN helix of the G-protein subfamilies and the sequence of the chimeric Gα subunits. Transferred region from Gα_i1 in each chimera is colored in orange. b, c Analytical SEC of β2AR/G_siN and M₁R/G_11iN complexes incubated with GTPγS in the presence or absence of Fab16. Protein elution profiles were monitored by the intrinsic tryptophan fluorescence. d Negative stain electron microscopy image of purified the M₁R/G_11iN/scFv16 complex