Conformationally selective RNA aptamers allosterically modulate the β2-adrenoceptor

Abstract

G-protein-coupled receptor (GPCR) ligands function by stabilizing multiple, functionally distinct receptor conformations. This property underlies the ability of 'biased agonists' to activate specific subsets of a given receptor's signaling profile. However, stabilizing distinct active GPCR conformations to enable structural characterization of mechanisms underlying GPCR activation remains difficult. These challenges have accentuated the need for receptor tools that allosterically stabilize and regulate receptor function through unique, previously unappreciated mechanisms. Here, using a highly diverse RNA library combined with advanced selection strategies involving state-of-the-art next-generation sequencing and bioinformatics analyses, we identify RNA aptamers that bind a prototypical GPCR, the β2-adrenoceptor (β2AR). Using biochemical, pharmacological, and biophysical approaches, we demonstrate that these aptamers bind with nanomolar affinity at defined surfaces of the receptor, allosterically stabilizing active, inactive, and ligand-specific receptor conformations. The discovery of RNA aptamers as allosteric GPCR modulators significantly expands the diversity of ligands available to study the structural and functional regulation of GPCRs.

Figure 1. Generation of conformation-specific RNA aptamers against the β₂AR

(a) Schematic overview of the selection strategy, NGS, bioinformatics analysis, and characterization of candidate aptamers. Ribbon diagram representation for selection against inactive β₂AR (colored in blue; PDB: 2RH1) or active β₂AR bound to high affinity agonist BI167107 (colored in red; PDB: 3SN6). MNG detergent micelles are represented in gray. Encircled orange areas show potential binding regions for aptamers to different β₂AR conformations. (b) Scatter plot from NGS analysis, comparing fold enrichment-ratios (R4 to R9) for the top 20-aptamer sequences from selection on unliganded β₂AR (x-axis) versus BI167107-bound β₂AR (y-axis). Each point in the plot represents a unique aptamer (the top seven candidate binders are color-coded in red, blue, or purple) according to their enrichment and selectivity to a selection target. (c) Bar graph shows top seven aptamers and their capacity to bind unliganded β₂AR or BI167107-bound β₂AR as assessed by pull-down assay. Boxed bars denote the four aptamers, selected for further characterization. Data shown represent the mean ± s.e.m. (*P < 0.05; **P < 0.01; ***P < 0.001) of three independent experiments, analyzed by one-way ANOVA followed by a Fisher’s LSD post-test.

Figure 2. Aptamers distinguish between inactive and active conformations of the β₂AR

Binding kinetic profiles for the interactions of four biotinylated aptamers with BI167107-bound (active) β₂AR or ICI-118,551-bound (inactive) β₂AR as analyzed using biolayer interferometry (BLI). (a–d) Representative sensorgrams for the interactions of four biotinylated aptamers with BI167107-bound β₂AR (left panels) or ICI-118,551-bound β₂AR (right panels): A1 (a), A2 (b), A13 (c), and A16 (d). Data was globally fit to 1:1 binding model as described in methods. K_d (dissociation constant) is shown as the ratio of k_off (dissociation) to k_on (association) rate constants. Each K_d value represents the mean affinity values ± s.e.m. of three independent experiments. Blue, gray, green, and light blue curves represent the measured responses for each tested concentration of β₂AR (BI167107 or ICI-118,551-bound β₂AR). Whereas overlay the curves in red show the global fitting results of the binding data.

Figure 3. Selectivity of aptamers for specific β₂AR conformations correlates with receptor ligand efficacy or ligand specificity

(a) Dose-response curves for the competition of the radioiodinated antagonist cyanopindolol ([¹²⁵I]-CYP) binding with isoproterenol in β₂AR reconstituted within HDL particles in the presence or absence of aptamers (A1, A2, A13, A16 or CNT-Apt). The ordinates represent the specific binding of 60 pmol/L [¹²⁵I]-CYP in the presence of different concentrations of ISO. Curves were obtained from three independent experiments. Error bars represent standard errors. (b–e) Western blotting analyses from binding experiments using specified biotinylated aptamers (A1, A2, A13 or A16) interacting with β₂AR in the absence or presence of various ligands. Agonists are shown (bars, royal blue) and antagonists (bars, cyan). Structures and functional properties of the nine β-adrenoceptor ligands are shown in Supplementary Figure 6. Binding of four aptamers to various forms of β₂AR: A1 (b) and A13 (d) preference for agonist-bound β₂AR; A2 (c) preference for ligand-specific form of β₂AR; and A16 (e) preference for antagonist-bound β₂AR. Aptamer input is indicated by ethidium bromide staining of the eluted RNA. Data correspond to the mean ± standard error of at least three independent experiments.

Figure 4. Influence of aptamers on conformational changes conferred by ligands

(a–d) Bimane fluorescence quenching measurement shows that aptamers A1, A2, and A13 stabilize active forms of β₂AR, while aptamer A16 stabilizes an inactive conformation. Bimane fluorescence quenching measurement detects conformational changes of the β₂AR via movement of a bimane probe on TM6 (at C265) upon the binding of agonists, (BI167107 [BI] or isoproterenol [ISO]) and/or aptamer: A1 (a), A2 (b), A13 (c) or A16 (d). Fluorescence emission spectra showing ligand-induced conformational changes of bimane-labeled β₂AR in the absence (black dashed line) or presence of full agonist (ISO, blue dashed line, or BI, red dashed line), inverse agonist ICI-118,551 (ICI, green dashed line), aptamer A1, A2, A13, or A16 (black solid line), or a combination of aptamer (A1, A2, A13, or A16) with ISO (blue solid line), BI (red solid line), or ICI (green solid line).

Figure 5. Functional effect of aptamers on β₂AR–mediated Gα_s and AC activation

β₂AR-dependent stimulation of AC activity and accumulation of cAMP was measured in HEK-293 membrane homogenates stably expressing β₂AR (expression level: ~2.5 pmoles/mg) in the presence of 100 nM of isoproterenol (ISO) or combination of 100 nM ISO with aptamer (A1, A2, A13, A16 or control aptamer). Data represents the means ± s.e.m. of at least four independent experiments. Asterisks in the bar graphs denote significant differences (*P < 0.05; **P < 0.01) by one-way ANOVA (with Tukey’s multiple comparisons test) from results for AC activity performed with control aptamer.

Figure 6. EM analysis and molecular architecture of β₂AR–aptamer complexes

(a–d) EM characterization of purified β₂AR-aptamer-Fab complexes. Shown on the left side of each panel are two-dimensional (2D) particle class average particles of the β₂AR–aptamer-Fab complexes. Scale bar is 10 nm. The left side of each panel shows 2D particle averages of anti-FLAG Fab labeled β₂AR in complex with aptamer A1 (a), A2 (b), A13 (c) or A16 (d). The right side of each panel shows a cartoon representation of the class average; the various components in the 2D–image maps (β₂AR in red; detergent micelle labeled with “m” in light-gray; anti-FLAG Fab antibody in dark-red; the aptamers in lime-green [A1], medium-purple [A2], pink [A13] and gray [A16]).