Membrane mimetic-dependence of GPCR energy landscapes

Abstract

We leveraged variable-temperature ¹⁹F-NMR spectroscopy to compare the conformational equilibria of the human A_2A adenosine receptor (A_2AAR), a class A G protein-coupled receptor (GPCR), across a range of temperatures ranging from lower temperatures typically employed in ¹⁹F-NMR experiments to physiological temperature. A_2AAR complexes with partial agonists and full agonists showed large increases in the population of a fully active conformation with increasing temperature. NMR data measured at physiological temperature were more in line with functional data. This was pronounced for complexes with partial agonists, where the population of active A_2AAR was nearly undetectable at lower temperature but became evident at physiological temperature. Temperature-dependent behavior of complexes with either full or partial agonists exhibited a pronounced sensitivity to the specific membrane mimetic employed. Cellular signaling experiments correlated with the temperature-dependent conformational equilibria of A_2AAR in lipid nanodiscs but not in some detergents, underscoring the importance of the membrane environment in studies of GPCR function.

Keywords: A(2A) adenosine receptor; GPCR; NMR; cellular signaling; membrane mimetics.

Figure 1.. ¹⁹F-NMR observed conformational ensemble of A_2AAR-ligand complexes in lipid nanodiscs compared over variable temperatures.

(A – C) The 1-dimensional ¹⁹F-NMR spectra of A_2AAR[A289C^TET] reconstituted into lipid nanodiscs containing POPC:POPS (70:30 molar ratio) recorded at 280 K, 298 K and 310 K for (A) apo A_2AAR, and complexes with (B) the antagonist ZM241385, and (C) the full agonist NECA. The NMR spectra are interpreted by Lorentzian deconvolutions with the minimal number of components that provided a good fit, labeled P1 to P4, superimposed on the frequency-domain data (black lines). Summation of the individual fits is shown by the superimposed grey lines. The green fitted line at 310 K is from an NMR signal from free ¹⁹F-TET. (D – F) Relative populations of each state over the range of studied temperatures determined from the fitted data shown in panels A – C represented in a bar chart format for A_2AAR in complex with the (D) no ligand (apo), (E) antagonist ZM241385, and (F) the full agonist NECA. See also Figures S1 – S3 and Table S2 – S4.

Figure 2.. Variable temperature ¹⁹F-NMR spectra of agonist-bound A_2AAR reconstituted in nanodiscs of varied lipid composition as a response to temperature and membrane fluidity measurements of nanodiscs of varied lipid compositions.

(A – D) 1-dimensional ¹⁹F NMR spectra of A_2AAR-NECA reconstituted in (A) POPC:PI(4,5)P₂ (95:5, molar ratio), (B) POPC:POPG (70:30, molar ratio), (C) POPC:POPS:Cholesterol (70:30:5 mol%), and (D) POPC:POPG:Cholesterol (70:30:5 mol%) recorded at 280 K and 298 K. Other presentation details are the same as in Figure 1. (E) Laurdan-based fluidity measurements of empty nanodiscs containing lipids POPC, POPE:POPS (70:30), POPC:POPS(70:30), and POPS, across a range of temperature from 4 °C to 37 °C. Generalized polarization values are plotted as the mean ± s.e.m. for three independent trials. (F) Laurdan-based fluidity measurements of A_2AAR reconstituted in nanodiscs containing lipids POPC and POPC:POPS(70:30) across a range of temperature from 4 °C to 37 °C. Generalized polarization values are plotted as the mean ± s.e.m. for three independent trials. See also Figures S1 and Table S2 and S4.

Figure 3.. Variable temperature ¹⁹F-NMR observed ensembles for complexes with partial agonists of differing efficacy.

(A and B) ¹⁹F-NMR spectra of A_2AAR[A289C^TET] in lipid nanodiscs containing POPC:POPS (70:30 molar ratio) recorded at three temperatures for complexes with the partial agonists (A) Regadenoson and (B) LUF5834. The fit for the unique population observed for the partial agonist complexes is shown by the cyan line. Other figure presentation details the same as in Figure 1. (C) cAMP accumulation experiments upon stimulation of A_2AAR with full and partial agonists. Data are shown as the means ± standard error of mean (s.e.m.) for three independent experiments. The EC₅₀ values are 1.52 nM, 1.14 nM, 0.52 nM, 5.14 nM, and 2.26 nM for CGS21680, NECA, UK432097, LUF5834, and Regadenoson, respectively. The E_max values were 101%, 100%, 103%, 57%, and 87% for CGS21680, NECA, UK432097, LUF5834, and Regadenoson, respectively. (D and E) Relative populations of each state over the range of studied temperatures determined from the fitted data shown in A and B in a bar chart format for A_2AAR in complex with the partial agonist (D) Regadenoson and (E) LUF5834. See also Figures S1 – S4 and Tables S1 – S4.

Figure 4.. Comparison of the variation in temperature observed for the conformational equilibria of A_2AAR[A289C^TET] in three different membrane mimetics.

(A and B) Variable temperature ¹⁹F-NMR spectra of A_2AAR[A289C^TET] reconstituted in DDM-CHS micelles in complex with the (A) agonist NECA and (B) partial agonist LUF5834. Other figure presentation details are same as in Figures 1 and 3. The green fitted line at 298 K and 310 K is from free TET. (C) Variable temperature ¹⁹F-NMR spectra of A_2AAR[A289C^TET] in complex with the agonist NECA reconstituted in LMNG-CHS micelles. Other figure presentation details are same as in Figures 1 and 2. The green fitted line at 298 K and 310 K is from free TET. (D) Variable temperature ¹⁹F-NMR spectra of A_2AAR[A289C^TET] in complex with the agonist NECA reconstituted in POPE:POPS (70:30 molar ratio). Other figure presentation details are same as in Figures 1 and 3. (E – H) Relative populations of each state over the range of studied temperatures determined from the fitted data shown in A – D represented in a bar chart format for A_2AAR reconstituted in DDM-CHS micelles, in complex with the (E) agonist NECA and (F) partial agonist LUF5834, and A_2AAR in complex with the agonist NECA reconstituted in (G) LMNG-CHS micelles and (H) lipid nanodiscs containing POPE:POPS (70:30 molar ratio). See also Figures S1 – S4 and Table S1.