Springer Nature
12915_2016_247_MOESM4_ESM.tif (3.06 MB)

Additional file 4: Figure S4. of Quaternary structure of a G-protein-coupled receptor heterotetramer in complex with Gi and Gs

Download (3.06 MB)
posted on 2016-04-05, 05:00 authored by Gemma Navarro, Arnau Cordomí, Monika Zelman-Femiak, Marc Brugarolas, Estefania Moreno, David Aguinaga, Laura Perez-Benito, Antoni Cortés, Vicent Casadó, Josefa Mallol, Enric Canela, Carme Lluís, Leonardo Pardo, Ana García-Sáez, Peter McCormick, Rafael Franco
Possible interfaces in A2AR homodimers in complex with Gs. In A–E, the A2AR homodimer was modeled through TM4 using the H1-receptor structure as template (A), through TM5 using the structure of squid rhodopsin (B), through TM4/5 using the β1-receptor structure (C), and via TM5/6 (D) and TM1 (E) using the μ-OR structure. TM helices 1, 4, and 5 involved in receptor dimerization are highlighted in dark blue, light blue, and gray, respectively. A2AR protomers bound to Gs (in gray) are shown in light green, whereas Gs-unbound A2AR protomers are shown in dark green. Rluc (blue) is attached to the N-terminal αN helix of Gs, and YFP (yellow) is attached to the C-terminal domain of the Gs-unbound A2AR protomer (light green). It is important to note that the position of YFP is highly dependent on the orientation of the long and highly flexible C-tail of A2AR (102 amino acids, Gln311–Ser412), which was modeled as described for the OXER [32] (see Additional file 9: Figure S9 for details). Despite these limitations, we can crudely estimate the approximate distances between the center of mass of Rluc and YFP as 4.6, 10.1, 6.5, 11.6, and 8.3 nm for panels A–E, respectively. Thus, among all these possible dimeric interfaces, only the molecular models depicted in panels A (TM4 interface) and C (TM4/5 interface) would favor the observed high-energy transfer between Gs-Rluc and A2AR-YFP (Fig. 4a in main paper). However, there is a steric clash between the N-terminal helix of Gs and the dark-green protomer in the TM4 interface. Accordingly, we have modeled A2AR homodimerization via the TM4/5 interface. Unfortunately, similar experiments with cells transfected with Gi-Rluc and A1R-YFP could not be accomplished because of a lack of receptor expression (not shown); it is likely that the shorter C-tail of A1R (16 amino acids, Pro311–Asp326) could not accommodate YFP in the presence of Gi in the right three-dimensional structure. The A1R homodimer was built using the same TM4/5 interface as for A2AR. (TIF 3135 kb)


Secretaría de Estado de Investigación, Desarrollo e Innovación