G-Protein Coupled Receptors (GPCR) are involved in a myriad of pathways key for human physiology through the formation of complexes with intracellular partners such as G-proteins and Arrestins (Arr). However, the structural and dynamical determinants of these complexes is still largely unknown. Herein, we developed a computational big-data pipeline that enables the structural characterisation of G-protein coupled receptor (GPCR) complexes with no available structure. This pipeline was used to study a well-known group of catecholamine receptors, the human Dopamine Receptor (DXR) family and its complexes, producing novel insights into the physiological properties of these important drug targets. A detailed description of the protein interfaces of all members of the DXR family (D1R, D2R, D3R, D4R and D5R) and the corresponding protein interfaces of their binding partners (Arrs: Arr2, Arr3; G-proteins: Gq, Gz, Gt2, Gi1, Gi2, Gi3, Gs(sh), Go, Gs(lo), Gob) were generated. To produce reliable structures of the DXR family in complex with either G-proteins or ARR, we performed homology modelling using as templates the structures of the β2-Adrenergic Receptor (β2-AR) bound to Gs, the Rhodopsin bound to Gi and the recently acquired Neurotensin Receptor-1 (NTSR1) bound to Arrestin (Arr). The structural and dynamic analysis of all DXR complexes highlighted three major partners groups: Arrs, Gs- and Gi-proteins. Additionally, it was revealed the involvement of different structural motifs in G-protein selective coupling between D1- and D2-like receptors. Having constructed and analysed 40 models involving DXR, this work represents an unprecedented large-scale analysis of GPCR-intracellular partner interface determinants.
Preto et al, 2020 - Submitted
Both DXR-Arrs and DXR-G-protein complexes tends to lead to higher TM3-TM6 distances in comparison to DXR alone. Arrs coupling tends to promote lower TM3-TM6 distances, particularly observed at D1R-like-Arrs and D3R-Arrs complexes. Arrs and G-proteins tend to be clustered differently, depending on the DxR bound. In D1-like receptors, Gs is isolated from Gi and Gq subfamilies. Gz is the most isolated partner among Gi/o subfamily in all DxR. Indeed, Gs in complex with D1R-like receptors led to higher TM3-TM6 (around 17-18 Å) and lower TM3-TM7 distances (around 12-13 Å). Conversely, DXR-Gi complexes tend to promote lower TM3-TM6 (around 14-17 Å) and higher TM3-TM7 (around 13-15 Å) distances, especially if D2-like receptors are involved. Complexes with Arrs are clustered within 14-15 Å in TM3-TM6 distance, apart from D4R complexes which are closer to 16 Å. Regarding TM3-TM7 distance, DXR-Arrs complexes are more dispersed. Nevertheless D2R-Arrs complexes have the highest distance values, following the trend observed in the G-protein complexes.
All complexes were analysed using normal mode analysis (NMA) and fluctuation values were calculated for all relevant structural motifs. The average fluctuation fold changes were calculated as the fold change between the receptor in complex and as a monomeric template. Considering that fluctuation can be seen as a proxy for protein motility, an increase in the average fold change indicates that, upon binding to its partner, the DxR undergoes conformational changes that give it more motility for the relevant structural motif. These values allow us to see distinct changes between relevant complex groups, most relevantly between DxR-Arrestin complexes, for which the average fluctuation value increases the least for all TM segments and even decreases in H8. Also relevant is how DxR-Gs complexes show a lower fold change in fluctuation when compared to other DxR-Gprot complexes (in a few cases it is possible to see a decrease in fluctuation upon binding). What is also fairly easy to observe is that D1R-like receptors will show fairly lower increases in fluctuation when compared with D2R-like receptors, and within the latter group D3R is the receptor showing the highest fold change in general.
Structural motif flexibility is here measured as the Bhattacharyya coefficient (BC) for each structural motif between the receptor when in complex and its monomeric template. To interpret this, we can see this as a movement correlation between DxR in complex and its monomeric template - if this value is closer to one, this implies that a smaller change in flexibility took place. This allows us to draw similar conclusions to the fluctuations - DxR-Arr complexes show a fairly high BC, which is indicative of smaller changes in flexibility upon binding. However, while not true for all structural motifs, DxR-Arr complexes will have values that can be fairly close to DxR-Gprot complexes. Additionally, differences between D1R-like-Gi/Go/Gq complexes and D2R-like-Gi/Go/Gq complexes are fairly pronounced, with the former showing, in general, higher values.
Here, we can see the two-dimensional projection of the flexibility for each receptor obtained through multidimensional scaling. This helps us, in a more intuitive and visual way, reason the results presented in the previous heatmap. Particularly, pronounced differences between D1R- and D2R-like receptors are immediately visible. What is also evident is the very clear clustering of Arrestins, of DxR-Gi proteins and its colocation with DxR-Go/Gz/Gq/Gt2 proteins, and, finally, the distinctly different flexibility changes upon binding in D4R-Gob when compared with any other complex.