Binding of an extracellular agonist to a G protein–coupled receptor (GPCR) usually stimulates a number of intracellular signaling pathways by inflicting the GPCR to couple to each G proteins and arrestins. GPCRs signify the targets of roughly one-third of all medication (1), and, in lots of instances, the specified results of a drug stem from arrestin signaling and the undesired ones from G protein signaling, or vice versa. Intriguingly, sure GPCR ligands preferentially stimulate both arrestin or G protein signaling, a phenomenon referred to as biased signaling (2).
The molecular mechanism of biased signaling stays unknown, hindering the invention and optimization of biased ligands which might be more practical and have fewer unintended effects than typical medication. Biophysical research point out that ligands with distinct bias profiles stabilize distinct receptor conformations (3, 4), however these research don’t determine what these conformations are or how ligands induce them. The receptor adopts related conformations in experimental buildings of GPCR–G protein (5–9) and GPCR-arrestin (10–12) complexes, leaving unclear what intracellular conformations are answerable for biased signaling and the way ligands within the extracellular binding pocket choose among the many related signaling conformations.
The angiotensin II (AngII) sort 1 receptor (AT1R) is a mannequin system for research of biased signaling (13–15). It stimulates each G protein–mediated and arrestin-mediated signaling pathways upon binding of its native ligand, the octapeptide AngII. Small modifications to AngII may end up in both arrestin-biased or G protein–biased ligands, which induce a better or decrease ratio of arrestin signaling to G protein signaling than AngII, respectively (13, 14). AT1R can also be a significant drug goal, and there’s curiosity in creating arrestin-biased AT1R ligands as medication for coronary heart failure as a result of such ligands can enhance cardiac contractility with out undesired hypertensive results (16–18).
The companion manuscript (19) presents crystal buildings of AT1R certain to AngII and to arrestin-biased agonists. The intracellular conformations in these buildings and within the beforehand printed active-state construction (20) are primarily equivalent, most probably as a result of they’re all stabilized for crystallography by binding to the identical high-affinity nanobody on the intracellular aspect. To find out how variations in intracellular conformation relate to the bias profile of the certain ligand, we carried out intensive molecular dynamics (MD) simulations of AT1R with out the nanobody.
Selecting the drug to suit the protein
Many accredited medication bind to G protein–coupled receptors (GPCRs). A problem in focusing on GPCRs is that completely different ligands preferentially activate completely different signaling pathways. Two papers present how biased signaling arises for the angiotensin II sort 1 receptor that {couples} to 2 signaling companions (G proteins and arrestins). Suomivuori et al. used large-scale atomistic simulations to indicate that coupling to the 2 pathways is thru two distinct GPCR conformations and that extracellular ligands favor one or the opposite conformation. Wingler et al. current crystal buildings of the identical receptor certain to ligands with completely different bias profiles. These buildings present conformational adjustments in and across the binding pocket that match these noticed in simulations. This work may present a framework for the rational design of medicine which might be more practical and have fewer unintended effects.
Science, this challenge p. 881, p. 888
Summary
Biased signaling, wherein completely different ligands that bind to the identical G protein–coupled receptor preferentially set off distinct signaling pathways, holds nice promise for the design of safer and more practical medication. Its structural mechanism stays unclear, nonetheless, hampering efforts to design medication with desired signaling profiles. Right here, we use intensive atomic-level molecular dynamics simulations to find out how arrestin bias and G protein bias come up on the angiotensin II sort 1 receptor. The receptor adopts two main signaling conformations, one in all which {couples} nearly solely to arrestin, whereas the opposite additionally {couples} successfully to a G protein. An extended-range allosteric community permits ligands within the extracellular binding pocket to favor both of the 2 intracellular conformations. Guided by this computationally decided mechanism, we designed ligands with desired signaling profiles.
AT1R transitions between two energetic intracellular conformations
Our simulations have been initiated from the beforehand printed active-state AT1R construction (20), with the nanobody eliminated. In most of those simulations, the cocrystallized ligand was eliminated, and completely different ligands have been modeled as a substitute, together with AngII, 4 arrestin-biased ligands, and two G protein–biased ligands (fig. S1 and desk S1). We additionally carried out management simulations from the buildings reported within the companion manuscript (19), and these simulations converged to the identical conduct.
In simulations with the nanobody eliminated, agonist-bound AT1R transitions between two intracellular conformations that differ primarily in transmembrane helix 7 (TM7) (Fig. 1 and fig. S2). One among these, the “canonical active” conformation, intently resembles beforehand decided buildings of GPCRs in complicated with G proteins (5–9) in addition to the active-state construction of the AngII sort 2 receptor (AT2R) (21). The opposite—the “alternative” conformation—differs in a number of regards. Considered from the extracellular aspect, TM7 is twisted counterclockwise above its proline kink, inflicting N1.50 (Asn46) to modify its most well-liked hydrogen-bond acceptor from N7.46 (Asn295) to C7.47 (Cys296). [We use the Ballesteros-Weinstein numbering scheme, where the digit before the decimal point specifies the TM helix (22).] Because of this twist, the intracellular portion of TM7 shifts towards TM3, inflicting the aspect chains of Y7.53 (Tyr302) of the NPXXY motif (N, Asn; P, Professional; X, any residue; Y, Tyr) and R3.50 (Arg126) of the DRY motif (D, Asp; R, Arg) to undertake “downward” rotamers pointing towards the intracellular aspect. Curiously, the nanobody-bound AT1R buildings resemble the choice intracellular conformation, besides that Y7.53 adopts the upward rotamer of the canonical energetic conformation as a result of its downward rotamer would conflict with the nanobody. Each the choice and canonical energetic conformations retain the TM6 conformation noticed within the nanobody-bound AT1R buildings (19, 20), which is attribute of active-state GPCR buildings in that it’s shifted outward relative to inactive-state buildings (23, 24).
In contrast to the canonical energetic conformation, the choice conformation has not been noticed in experimental GPCR buildings. The choice conformation of TM7 intently resembles a number of GPCR buildings (fig. S2) (25–30), together with the serotonin 2B receptor (5-HT2BR) certain to arrestin-biased ligands. Nevertheless, these buildings exhibit extra inactive-like positions of TM6, probably owing to the absence of intracellular binding companions. In earlier simulations of the β2-adrenergic receptor (β2AR) transitioning from its canonical energetic to inactive conformation, we noticed a uncommon intermediate wherein each TM6 and TM7 match the choice conformation, however we didn’t recommend a connection to biased signaling (31).
The choice conformation seems to accommodate β-arrestins however not Gq
To find out whether or not these two conformations couple otherwise to G proteins and arrestins, we ready structural fashions of AT1R in complicated with its most well-liked companions, Gq and β-arrestins 1 and a pair of (Fig. 2). These fashions recommend that, though the canonical energetic conformation {couples} properly to each Gq and β-arrestins, the choice conformation {couples} properly to β-arrestins however not Gq. The choice conformation stabilizes R3.50 in a downward rotamer, which clashes with the α5 helix of Gq. R3.50 shifts upward on transition to the canonical energetic conformation, accommodating insertion of the Gα subunit. Each the choice and canonical energetic conformations seem to readily accommodate the β-arrestin finger loop.
In simulations of rhodopsin certain to visible arrestin, rhodopsin sometimes transitions spontaneously from the canonical energetic conformation to the choice conformation, with each R3.50 and Y7.53 forming interactions with spine atoms on the finger loop (fig. S4). This means that each canonical energetic and various conformations of rhodopsin couple to arrestin.
Intracellular TM7 conformation is allosterically coupled to the ligand-binding pocket – “g protein receptor”
In simulations that transition between the choice and canonical energetic conformations, we noticed rearrangements in a number of residues that type an allosteric community between the ligand-binding pocket and the intracellular aspect of the receptor (Fig. 3A and fig. S5). The intracellular TM7 conformation is intently coupled to the place of Y7.43 (Tyr292) larger on TM7. The choice conformation is favored when Y7.43 factors towards TM3, and the canonical energetic conformation is favored when Y7.43 factors towards TM2 (Fig. 3B).
Y7.43 is coupled to the ligand by way of two adjoining residues on TM3, N3.35 (Asn111) and L3.36 (Leu112) (Fig. 3, A and B). When N3.35 factors inward (towards the middle of the helical bundle), Y7.43 practically all the time factors towards TM3. N3.35 should level outward for Y7.43 to level towards TM2, which requires that the aspect chains of Y7.43 and F2.53 (Phe77) swap positions (Fig. 3B). L3.36, which is commonly in direct contact with ligands, influences the place of neighboring residue N3.35: TM5- and TM2-proximal positions of L3.36 favor the inward- and outward-pointing positions of N3.35, respectively. L3.36 and Y7.43 can even work together instantly, so repositioning of L3.36 additionally has some direct impact on Y7.43 conformation.
Arrestin-biased ligands favor the choice conformation, and Gq-biased ligands favor the canonical energetic conformation
In replica-exchange MD (REMD) simulations designed to pattern effectively the conformational ensemble of AT1R certain to every ligand, we discovered that TM7 adopted the choice conformation extra incessantly with arrestin-biased ligands certain than with AngII certain and extra incessantly with AngII certain than with Gq-biased ligands certain (Fig. 3C). Mixed with our remark that the choice conformation preferentially binds arrestin, this implies that ligands obtain arrestin bias by favoring the choice conformation and G protein bias by favoring the canonical energetic conformation.
Ligands choose among the many various and canonical energetic conformations by means of the allosteric community described above. In simulation, L3.36 of the ligand-binding pocket was shifted towards TM2 extra incessantly with AngII than with arrestin-biased ligands and much more incessantly with Gq-biased ligands (Fig. 4B). The arrestin-biased ligands prolong a lot much less deeply into the binding pocket (Fig. 4A and fig. S1), so they can’t readily push L3.36 towards TM2. Against this, AngII and the Gq-biased ligands possess a cumbersome phenylalanine at place 8 (F8) and thus are likely to push L3.36 towards TM2.
Our simulations point out that the F8 residue of AngII and the Gq-biased ligands adopts distinct orientations with distinct results on L3.36 (Fig. 4B and figs. S6 and S7). When L3.36 is within the TM5-proximal place that favors the choice conformation, F8 tends to be vertical (i.e., the ring airplane is perpendicular to the membrane airplane), packing tightly above L3.36 (Fig. 4B and fig. S6). When F8 as a substitute adopts a horizontal orientation, it forces L3.36 towards TM2, which in flip favors the canonical energetic conformation as described above.
In simulation, Gq-biased ligands adopted the horizontal F8 orientation extra incessantly than AngII (fig. S7). That is probably as a result of, at AngII and arrestin-biased ligands, a positively charged arginine at place 2 (R2) engages negatively charged binding pocket residues D6.58 (Asp263) and D7.32 (Asp281), pulling the extracellular finish of TM6 inward. The Gq-biased ligands lack a positively charged residue at place 2, and because of this, the extracellular finish of TM6 tends to maneuver outward, creating more room inside the binding pocket for F8 to undertake a horizontal orientation (Fig. 4C).
The crystal buildings within the companion manuscript (19) help this biased signaling mechanism, which we recognized utilizing simulations initiated from the beforehand printed active-state construction (20). These crystal buildings are locked right into a single intracellular conformation, however within the buildings with arrestin-biased ligands certain, residues close to the binding pocket—N3.35, L3.36, and Y7.43—undertake positions that favor the choice conformation in simulation. Within the AngII-bound construction, then again, N3.35 adopts a place (fig. S8) that favors the canonical energetic conformation in simulation. The density for L3.36 on this construction is weak, suggesting that this residue adopts a number of conformations, and Y7.43 is so cellular that it can’t be resolved in any respect.
In settlement with our computational outcomes, our current double electron-electron resonance (DEER) spectroscopy examine of AT1R (4) recommended that arrestin-biased, balanced, and G protein–biased ligands stabilize subtly completely different intracellular TM7 conformations. The DEER information additionally present variations in helix 8 place, that are probably resulting from these conformational adjustments in TM7. Variations in TM6 place may be resulting from adoption of the TM6-bent conformation mentioned beneath. The DEER information recommend that the receptor undergoes extra ligand-dependent conformational adjustments on time scales longer than these of our simulations, though adjustments in different receptor areas will not be related to biased signaling. For instance, though the assorted arrestin-biased ligands stabilize various conformations in DEER experiments, these ligands have very related pharmacological bias profiles. Their pharmacological profiles are per our simulations, which point out that the assorted arrestin-biased ligands stabilize the choice TM7 conformation to an analogous diploma.
Computational design of ligands with desired biased signaling profiles
To additional validate our computationally decided mechanism, we used it to design ligands with desired biased signaling profiles (Fig. 5), a long-standing problem in GPCR drug discovery.
Lowering the dimensions of the C-terminal AngII residue is understood to lead to arrestin bias, however our simulations point out that the conformation of the C-terminal peptide residue—not simply its measurement—is a key determinant for bias. We predicted {that a} variant of AngII with the C-terminal fragrant ring constrained in a vertical orientation would favor the choice intracellular conformation, resulting in arrestin bias. We thus ready an AngII analog with a 2-aminoindan-2-carboxylic acid substitution at F8 (Ind8-AngII, Fig. 5A). Ind8-AngII is structurally equivalent to AngII besides that the C-terminal phenyl moiety is tied again to the Cα atom by the addition of a single connecting methylene group. Our simulations present that this modification restricts the phenyl ring to stay vertical. Certainly, experimental characterization of Ind8-AngII exhibits that it’s strongly arrestin-biased regardless of having a C-terminal residue even bigger than that of AngII and the Gq-biased ligands (Fig. 5A and desk S2).
On the idea of our discovering that outward movement of TM6 close to the binding pocket is related to the elevated Gq signaling of Gq-biased ligands, we hypothesized that an alanine substitution at R2 would get better Gq exercise for the partial agonist S1I8, which lacks a C-terminal phenylalanine however has one other comparatively massive residue, isoleucine, at this place. Certainly, mutating R2 of S1I8 to alanine will increase Gq exercise with out rising β-arrestin exercise (Fig. 5B and desk S2).
“g protein receptor”