The atomistic Mechanism Underlying Regulation of the GPA1 G Protein Signaling Pathway Mediated by Abscisic Acid (ABA) Phytohormone Binding to the GCR1 Plant G Protein Coupled Receptor

Abstract

We propose an atomistic mechanism by which key plant processes, including seed dormancy, root elongation, secondary root proliferation, and flower and fruit produc-tion, are regulated. This regulation occurs through binding of the phytohormone abscisic acid (ABA) to the plant G protein-coupled receptor (GPCR) GCR1. This mirrors the central role of GPCRs in animal systems, where they mediate vision, taste, olfaction, pain perception, and neurotransmission. Establishing GCR1 as a bona fide GPCR in plants would represent a transformative advance in plant biology and agriculture. In particular, GCR1 would be shown to transduce ABA signals through interaction with the Gα subunit (GPA1). However, direct experimental evidence for this interaction and conformation that ABA binding to GCR1 modulates GPA1 inactivation, remains elusive. A major obstacle in testing these hypotheses is the lack of structural data on GPA1 interactions within the ABA-GCR1 complex. To address this gap, we employ molecular dynamics (MD) and metadynamics simulations based on the AMBER and CHARM31 force fields to characterize atomistically the ABA-GCR1-GPA1 ternary complex. Our MD simulations reveal an allosteric mechanism whereby GCR1-ABA binding induces a rigid-body closure of the GPA1 Ras and α–helical domains, creating a steric blockade that traps GDP in the nucleotide-binding pocket. This con-formation prevents GTP exchange and maintains GPA1 in an inactive state, effectively terminating the signaling cascade. Free energy landscape analysis further demonstrates that this closed state represents a deep energy minimum, suggesting biological relevance as a regulatory mechanism. We propose specific mutations in the ABA-binding site of GCR1 and at the GCR1-GPA1 interface that could experimentally validate (or refute) our proposed mechanism. Confirmation of this model would pave the way for designing novel agonists and inverse agonists to precisely manipulate critical plant processes.

1. Introduction

Plants dynamically perceive and adapt to environmental changes through sophisticated signaling networks. A key regulator in these pathways is G-protein-coupled-receptor 1 (GCR1), which mediates plant responses to diverse stimuli by integrating abscisic acid (ABA) signaling via GPA1 inactivation, thereby regulating stress adaptation, seed dormancy, root growth, as well as other aspects of plant development. − In analogy to animal GPCR systems, signaling outcomes in plant G-protein pathways depend on ligand binding. Specifically, ligands stabilize either active or inactive receptor–G-protein conformations. GCR1’s pleiotropic roles are well-documented: it governs seed germination through ABA and gibberellin (GA1) crosstalk, − modulates root growth, ,− and participates in blue light responses. − Notably, GCR1 functions as a signaling node, coordinating hormonal (ABA, GA1) and environmental (light) cues to optimize plant growth and stress responses.

Recently, we made significant progress in understanding the molecular structure of GCR1 and its interaction with hormones such as ABA and GA1 in Arabidopsis thaliana. We used first-principles molecular docking, homology modeling, and molecular dynamics (MD) simulations to gain insights into the structure of GCR1 and the functional consequences of hormone binding. We predicted the three-dimensional structures of GCR1 bound to either ABA or GA1 in order to evaluate how these ligands modulate the conformational dynamics of the transmembrane helices. Through advanced MD and metadynamics sim-ulations, we demonstrated that GA1 binding to GCR1 induces a GPCR-like activation of GPA1. This agonist-like behavior mirrors the canonical GPCR paradigm in which ligand binding promotes conformational rearrangements that enable productive G-protein coupling.

Our simulations revealed that GA1 binding proceeds through disruption of the conserved LYS288–ASP162 salt bridge (SB) in GPA1, driving a domain separation (34 → 42 Å) that facilitates GDP/GTP exchange.

In contrast, the free energy landscape for the ABA–GCR1–GPA1–GDP–Mg²⁺ complex exhibits four distinct metastable states separated by barriers of approximately 10 kcal/mol, (Figure ). Thus, we can see how, within this landscape, ABA preferentially stabilizes inactive conformations. This behavior is directly analogous to antagonist or inverse agonist action in animal GPCRs. In such systems, ligand binding stabilizes inactive GPCR–G-protein assemblies that are incapable of nucleotide exchange or downstream signaling. These find-ings provide the first dynamical evidence for plant GPCR-like signaling while identifying residue–level interactions suitable for experimental validation. More broadly, our atomic-level characterization of this inactivation pathway enables rational strategies for modulating plant signaling responses through ligand or receptor engineering.

1.

Panel 1 shows the potential of mean force (PMF) profiles for the ABA-GCR1-GPA1-GDP-Mg²⁺ complex. Panels 2a to 2d show the free energy for a closed, a partially open, and two fully open structures of the GPA1 structure as a function of Ras-Helical distance as the collective variable in the ABA-GCR1-GPA1 complex.

To investigate the stability and behavior of GCR1–ligand complexes under near-native conditions, we performed atomistic molecular dynamics (MD) simulations. These simulations involved docked ABA–GCR1 complexes coupled to GPA1 within an explicit POPC lipid bilayer and solvated environment representative of plant membranes. This approach allows us to assess both structural integrity and thermodynamic stability of the inactive signaling complex. We identify that ABA-bound GCR1 adopts a conformation with improved binding energy relative to apo-GCR1, reflecting stable hydrogen-bonding and electrostatic interactions within the ligand-binding pocket.

Consistent with prior observations, ABA and GA1 induce qualitatively different con-formational responses in GCR1. GA1 promotes pronounced rearrangements of the seventh transmembrane helix (TM7) and intracellular coupling regions, facilitating Ras–helical domain opening in GPA1 and productive G-protein activation. In contrast, ABA binding prevents this opening and locks GPA1 in an inactive conformation. In the plant G-protein α subunit GPA1, the Ras-like domain comprises residues 37–63 and 197–372, whereas the helical domain spans residues 68–188, forming a helical bundle that occludes the nucleotide-binding pocket in the inactive state.

By integrating experimental observations with extended simulations, our work provides a cohesive mechanistic picture of how GCR1 regulates GPA1 signaling in plants. This study specifically elucidates the structural basis of ABA-mediated inactivation of the GPA1–GCR1 signaling complex. Our metadynamics results demonstrate that ABA binding stabilizes a closed GPA1 conformation that traps GDP through persistent interdomain contacts. This interaction is critical for maintaining GPA1 in its inactive conformation following the inactiva-tion mediated by the A. thaliana regulator of G-protein signaling 1 (AtRGS1). ,

Three salient features emerge from this inactivation mechanism.

• Ligand-induced stabilization of the GPA1–GCR1 interface,

• Formation of key SBs that lock GPA1 in the GDP-bound state,

• A free energy landscape that strongly favors the closed, inactive conformation.

These predictions suggest clear targets for experimental validation via site-directed mutagenesis at the receptor–G-protein interface. Experimental confirmation would establish ABA as a functional antagonist or inverse agonist of GCR1 signaling, providing a structural framework for rational modulation of plant G-protein pathways.

The following section (Section ) describes the computational methodology, followed by detailed results and discussion in Section , and concluding remarks in Section .

2. Methodology

The structural models of GCR1 and GPA1 used in our simulations are described in refs ,, Notably, we did not use the AlphaFold-predicted GCR1 structure available in the Protein Data Bank; instead, the GCR1 model was generated using the GEnSeMBLE method, which we developed and have previously validated across multiple GPCR proteins. For GPA1, we employed the experimentally determined structure reported by Jones et al., corresponding to PDB entry 2XTZ.

2.1. Coupling of the ABA-GCR1 Complex to the GPA1 with Ther-Modynamic Analysis

We employed the DarwinDock complete sampling method to predict the possible binding sites of abscisic acid (ABA) as a potential antagonist for signal inactivation of GCR1. , The details and resulting structures of these calculations are shown in our previous works. , We docked the Gα subunit (GPA1) (containing the inactivated GPA1-GDP-Mg²⁺ com-plex) of the G protein (GP) into our recently predicted ABA-GCR1 ligand-GPCR structure. We then inserted the fully constructed inactive ABA-GCR1-GPA1-GDP-Mg²⁺ into a periodic simulation box. This box contained the 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC) lipid bilayer with explicit water solvent (TIP3) and salt (Na⁺ and Cl^– ions), at the physiological salinity of 154 mM. POPC is frequently used in research and model membrane systems due to its well-characterized properties. , The counterions balance any net charge from the protein–ligand systems. Next, we carried out MD simulations, using the NAMD , program with the AMBER and CHARM31 force fields through the topology and parameter files “par_all27_prot_lipid” for the biomolecules and gaff for the small molecules. Prior to binding GPA1 to GCR1, we found that LYS49–ASP115 (2.37–4.37, Balles-teros–Weinstein numbering) formed an ionic lock in the ABA–GCR1 complex. This ionic lock is the analog to the highly conserved ASP/GLU3.39 to H6.40 ionic lock found in human class A GPCRs, occurring with 94–98% conservation among the human Opioid Receptors, hORs. For hOR, the C-terminal COO^– of the Gα subunit of the G-protein binds to break the highly conserved ionic lock of the animal GPCR, opening it to form the precou-pled GPCR-GP complex that is subsequently activated upon binding of hOR agonists. For GCR1, we found an analogous coupling between the terminal carboxylate group on LEU383 of GPA1 to open the LYS49-ASP115 ionic lock of GCR1.

This equilibrated system served as the initial configuration for metadynamics simulations aimed at characterizing the free energy landscape associated with the closing of GPA1 within the ABA–GCR1–GPA1–GDP–Mg²⁺ complex.

2.2. Free Energy Landscape from Metadynamics Simulations

Metadynamics simulations on the ABA-GCR1-GPA1-GDP-Mg²⁺ complex were performed to study the energetics of the GDP/GTP exchange.

We used, as collective variable, the distance between the centers of mass of the Ras and α-helical domains of GPA1. The residues conforming to each of these domains are listed in the Supporting Information and the Data and Software Availability. The upper and lower limits of the collective coordinate (26 and 50 Å) were chosen in agreement with previous works on GPCR-G protein complexes in animal cells. We found that the guanosine nucleotides of GDP inactive GPA1 are held tightly between the GPA1 Ras and α–helical domains. The metadynamics simulations led to the potential of mean force landscape (PMF) of the complex ABA-GCR1-GPA1-GDP-Mg²⁺, as a function of the GPA1 Ras-α-helical domain collective variable.

We used the same set of metadynamics parameters employed previously to simulate the Ras and α-helical domain opening in animal GPCR-GP systems, namely: a Gaussian width of 1.00 Å and height of 0.01 kcal mol^–1 with a Gaussian hill deposition frequency of 500 fs. The minimum and maximum values allowed for the collective variable were set to 26 Å and 50 Å, respectively, over the total time of 2.01 μs for the metadynamics simulation.

3. Results and Discussions

In our first paper, we reported the ligand-induced conformational changes observed in the ligand-bound GCR1 structures compared to the apo-GCR1 structure for two phytohormones: Gibberellin A1 (GA1) and Abscisic acid (ABA). We demonstrated that each hormone induces unique conformational changes in the GCR1 structure through specific interactions within the predicted ligand binding sites at the external region of GCR1, and also we confirmed that GA1 activates the GPA1 signaling pathway. Then, we described how the GA1 bound to GCR1 activates the GPA1, by promoting the opening of GPA1 to allow GDP/GTP exchange.

These results encouraged us to build a model for the Gα-subunit (GPA1) bound to the ABA-GCR1 complex. Specifically, we targeted the ionic lock between residues LYS49 and ASP115 located on helices 2 and 4 of GCR1, respectively. To describe the binding, we established a SB between the positively charged nitrogen of LYS49 in GCR1 and the terminal carboxylate of LEU383 in GPA1, thereby anchoring GPA1 to the GCR1 intracellular region. Then, we equilibrated this structure under physiological salt conditions. Finally, we performed metadynamics simulations. The metadynamics free energy profile revealed four distinct structures, in each of the four potential wells, which correspond to the different minimum energies of the ABA-GCR1-GPA1 complex. We equilibrated all four to assess their stability and map a pathway for the GPA1 inactivation, since the lowest energy ABA-GCR1 complex keeps GPA1 in its closed conformation. Thus, the role of the ABA ligand bound to the GCR1-GPA1 complex, keeping it inactive, was exposed by using metadynamics and MD simulations. We obtained the free energy profile as a function of the Ras and α-helical center of mass (CM) distance for a time of 1.13 μs. The initial GPA1 structure (Figure S1) was obtained from X-ray crystallography. , Our Detailed atomistic methods are provided below.

To validate the proposed hypothesis and investigate the role of the ABA–GCR1–GPA1 complex in the G protein inactivation mechanism we performed metadynamics simulations to locate potential conformations to be equilibrated by MD simulations. The free energy profile and the conformations initially taken for MD equilibration can be seen in Figure .

To analyze these results, we performed a series of MD simulations, on each of the four minimal free energy conformations for the ABA-GCR1-GPA1-GDP-Mg²⁺ complexes, after 176 cycles to reach convergence of the metadynamics simulations as shown in Figure S2, which we examined to determine their stable conformations and thermodynamic stabilities.

In addition, we also analyze the effect of ABA, within the GCR1 binding sites, on the conformational changes undergone by the GPA1 structures, bound to each complex.

First, we analyzed the MD results for the two complexes obtained from metadynamics: one with a Ras α-helical distance of approximately 27–28 Å, referred to as complex A, and another with a distance of about 34–35.3 Å, referred to as complex B. After equilibration, complex A maintained its original distance of 27–28 Å on average (Figure S4A), consistent with the metadynamics results. In contrast, complex B, which started at 34–35.3 Å, shifted to a Ras α-helical distance of approximately 30.0–31.5 Å on average (Figure S5B). Given the significance of this refined conformation, we designate it as B′.

These results show that GPA1 in the inactive B′ configuration, bound to the ABA-GCR1 complex, with a Ras and α-helical distance of ≈30.0–31.5 Å is the most thermodynamically stable conformation. This can be seen by comparing the energetic profiles obtained from the last 50 ns of the simulation time, of both equilibrated complexes, Figure S6 and in Table .

1. Conformational States of GPA1 Bound to ABA-GCR1 Complex: Distances and Potential Energies .

GPA1 structure	starting distance (Å)	final distance (Å)	potential energy (kcal/mol)
high E inactive (A)	27.3	27.8	–1475.6
inactive (B′)	34.8	30.8	–1636.2
partially open (C)	41.1	36.3	–912.9
fully open (D)	45.9	46.2	–1101.0

^aThe criteria used to define each state were the ras-helical distances after MD equilibration and the self-potential energy for each complex.

Thus, after equilibration, the B conformation returned to a closed and inactive conformation (B′), showing a much lower self-potential energy than the final self-potential energy obtained after equilibration from the high-energy inactive conformation (A). In addition, in Figure S6 we observe that both the electrostatic and VDW energies are also lower for the inactive B′ conformation than for the high energy inactive A conformation, which also is explained by the stability gained for the B conformation, by evolving to the inactive conformation (B′). In conclusion, for these two initial complexes, we confirm that the GPA1-GDP-Mg²⁺ in its inactive conformation, bound to the ABA-GCR1 complex, is the most stable conformation, which can be understood by analyzing the ABA-GCR1 interaction and how these interactions modify the interactions between GCR1 and GPA1 to promoting inactivation of GPA1, by keeping the GPA1 in its inactive conformation. This prevents the GDP/GTP exchange.

Interestingly, although the free energy barrier separating the 30.3 Å and 27.9 Å conformations is minimal (≈ 0.4 kcal/mol), as shown in Figure (A and B), the system relaxes and remains trapped at ≈30.3 Å during unbiased molecular dynamics. The conformational state centered at ≈30 Å, B′, appears to be kinetically and potentially stabilized. This behavior suggests the presence of a stabilizing network of interactions Figure A,B. These may include persistent Hydrogen Bonds (HBs) within the ABA binding site (Figure ) and intermolec-ular contact between GPA1 and GCR1 (Table ). Such interactions can effectively reduce the conformational flexibility of the system and slow down the transition toward the more thermodynamically stable closed state.

2.

RMSD, for the last 50 ns of the simulation time, shows the equilibration (below 2 Å) of the high-energy inactive complex (A) and the inactive complex (B′) for the ABA-GCR1-GPA1-GDP-Mg²⁺ complex. The last point is the reference for the RMSD calculation.

3.

Equilibrated GCR1 binding site views for both ABA–GCR1–GPA1–GDP–Mg²⁺ complexes with the corresponding GPA1 in its high-energy inactive conformation (A) (Panels A and B) and inactive structure (B′) (Panels C and D). Each complex is shown in its lowest-energy conformation after equilibration. The 3D views in Panels (A) and (C) illustrate a series of water-mediated HBs between ABA and amino acids within the GCR1 binding sites. Panels (B) and (D) display the pharmacophore views for these conformations. In these views, R denotes a protein residue and W denotes a water molecule, following Maestro-Schrödinger notation. Orientation note: the molecular orientation is consistent across both figures: the CO group is positioned on the right side, the OH group on the left side, and the COO^– group is oriented backward. This description is provided to ensure clarity in interpreting the structures.

2. SB Interactions in GPA1 Conformational States at Varying Ras and α–helical Domain .

GPA1 Ras-hel distance (Å)	SBs ICL1 and GPA1	SBs ICL2 and GPA1	SBs H6 and GPA1
high energy inactive A (≈ 28.0 Å)	ARG48-ASP345 and ARG48-GLU379	ASP115-ARG374	ARG209-ASP342
inactive B’ (≈30.5 Å)	LYS45-GLU278	LYS113-GLU379 and ASP115-ARG374	ARG209-ASP342 and LYS213-ASP342
ppen C (≈36.1 Å)		LYS113-GLU379 and ASP115-ARG374	ARG209-ASP342
fully open D (≈ 46.0 Å)		LYS113-GLU379 and ASP115-ARG374	ARG209-ASP342 and LYS213-ASP342

^aThe criteria used to define each state were the ras-helical distances after MD equilibration and the stable sb for each complex.

The root-mean-square deviations (RMSDs) for both the high-energy inactive complex (A) and the inactive complex (B′) demonstrate convergence of each simulation. As shown in, both systems reach stable RMSD values. In addition, Figures S3 and S2 show that, over the final 50 ns of the trajectories, the complexes remain equilibrated following total simulation times of 500 ns for complex A and 350 ns for complex B’.

To confirm this interpretation, we compared the GCR1 binding sites of the two inactive complexes to understand why the complex initially adopting the GPA1-B structure relaxed back to the inactive GPA1-B′ conformation. We further extended this analysis by comparing these results with those obtained for the partially open (C) and fully open (D) structures, which are described below.

By comparing the binding sites and pharmacophore views for both the high energy inac-tive complex (A) (≈27.0–28.0 Å), Figure Panels (A and B) and the inactive conformation (B′) (≈30.0–31.5 Å), Figure Panels (C and D), we found better water-mediated HBs net-work, panels C and D which correspond to the inactive conformation (B′), between amino acids in the GCR1 binding site with functional groups of ABA. These interactions modify the way that the GCR1 intracellular region interacts with the GPA1 and generates, in turn, changes in the Ras and α-helical distance due to interactions between amino acids of Ras and α-helical domains and with the GDP and Mg²⁺.

Specifically, we found that the differences in the water-mediated HBs between polar amino acids, on the GCR1 binding site, of each complex (A) and (B′), with the carbonyl and carboxylate groups on the ABA hormone, shown in Figure B,D, strongly influence the interactions of amino acids in the intracellular region of GCR1 with amino acids on the Ras domain of GPA1, as shown in Figures S9 and S10. Thus, by comparing Figures S9 and S10, we see that configuration (B) (Figure S10) has a better SB network, four stable and one transient SB, than configuration (A) (Figure S9), three stable and one transient SBs.

To determine how interactions of the ABA ligand with amino acids in the GCR1 binding site, affect the interactions between GCR1 and GPA1, we analyzed differences in the SB interactions, between both complexes, the one with the GPA1 in its high energy inactive A conformation (≈ 27–28 Å) and the GPA1, in its inactive (B′) conformation (≈ 30.0–31.5 Å). Specifically, between amino acids close to or belonging to the intracellular loop 1 (ICL1), loop 2 (ICL2), and loop 3 (ICL3) with amino acids in the Ras domain of GPA1, see Figures S7 and S8. Specifically, we observe, for conformation (A) (≈ 27–28 Å), two SBs between ARG48 in ICL1 of GCR1 with APS345 and GLU379 on GPA1 and one SB between ASP115 on ICL2 of GCR1 with ARG374 of GPA1, while for the conformation (B′) (≈ 30.0–31.5 Å) we found one SB between LYS45 in ICL1 of GCR1 with GLU278 of GPA1 and two SBs in conformation (B′), the first one between LYS113 on ICL2 with GLU379 on GPA1 and a second between ASP115 on ICL2 with ARG374 of GPA1, as can be seen in Figure S7 and in Figures S9 and S10.

Finally, while we found one SB between ARG209 in H6 of GCR1 with ASP342 on the GPA1-Ras domain of the high energy inactive conformation (A), we found two SBs in the inactive conformation (B′), the first between ARG209 of H6 of GCR1 with ASP342 in the GPA1-Ras domain, and another between LYS213 of H6 of GCR1 with ASP342 of the GPA1-Ras domain, as can be seen in Figure S8. Thus, we propose that the ABA-GCR1 interactions, shown in Figure , promote specific conformational changes in the structure of GPA1, through the GCR1-GPA1 interaction, affecting the Ras and α-helical distance, ulti-mately impacting the inactivation process of GPA1, by preventing the GDP/GTP exchange, in both complexes studied, as can be seen in Figures S4 and S5.

By comparing the potential energies, for all four equilibrated ABA-GCR1-GPA1-GDP-Mg²⁺ complexes, shown in Table , along with the analysis of SB networks, between amino acids of the intracellular region of GCR1 with amino acids in the Ras domain of GPA1, we observed that the most stable configuration is that where GPA1 obtained the inactive ABA- GCR1-GPA1-GDP-Mg²⁺ configuration (B′) followed by the high energy inactive complex (A). On the other hand, the equilibrated complexes with GPA1 are partially open (C) and fully open structures (D), Figures S14 and S15, have much higher potential energies, (−912.9 kcal/mol) and (−1101.0 kcal/mol), respectively, and the poorest SB networks among them.

This analysis of the SB networks shows that these open and fully open structures have the poorest GCR1-GPA1 SB interactions compared with those found in the closed conformation at ≈30.0–31.5 Å, as can be seen in Figures S11 and S12. Interestingly neither of them (the open and fully open structures) makes SB interactions between amino acids in ICL1 of GCR1 with amino acids in GPA1, as shown Table .

To explain the specific conformational changes observed in each complex, we analyzed the differences in GCR1–GPA1 interactions across the metadynamics-derived states. In par-ticular, we sought to understand how distinct SB networks may influence the GDP/GTP exchange process and the inactivation of GPA1. Because the closed GPA1 conformation represents the most stable configuration among the metadynamics-obtained complexes, we focused on comparing the ABA interactions within the GCR1 binding site for the inactive and open states. Specifically, we examined how differences in ABA binding to GCR1 alter its intracellular interactions with GPA1. As shown in Figure , comparison of the pharmacophore views for the equilibrated inactive complexes reveals that all interactions between ABA and amino acids in the GCR1 binding site are mediated by water molecules. Among the water-mediated hydrogen bonds, one of the most relevant is the interaction that links the carboxylate group of ABA to a key NH group in the binding site through a single water molecule. Figure S13 illustrates the distance between the ABA carboxylate and the mediating water molecule as a function of simulation time. The trajectory contains 200 frames corresponding to a 50 ns simulation, so the last 50 frames represent approximately 12.5 ns, a meaningful time scale for evaluating the stability of this interaction. As shown in Figure S13, the interaction distance stabilizes over these final frames, indicating that this water-mediated hydrogen bond is persistent and likely contributes significantly to the binding mechanism. Notably, this stabilization during the last 12.5 ns is consistent with the behavior observed in Figure S3, where the RMSD of the proteins forming the complex also reaches a stable regime over the same time interval. This agreement between the two analyses further supports the robustness of the binding configuration in the final portion of the trajectory. Moreover, the persistence of water-mediated hydrogen bonds over tens of nanoseconds, together with the stable Ras-like helical locking observed over hundreds of nanoseconds, indicates that these interactions are mechanistically meaningful for GPCR–G-protein regulation. Such long-lived structural features are unlikely to arise from transient fluctuations alone and instead sug-gest coordinated stabilization within the ABA–GCR1–GPA1 complex. These observations therefore strengthen the plausibility of the proposed inactivation mechanism and provide concrete molecular contacts that can be directly tested experimentally. In contrast, for the partially open and fully open GPA1 conformations, the HBs between ABA and GCR1 are not water-mediated, as illustrated in Figure . This difference indicates that, in the inactive complexes, the polar functional groups of ABA remain relatively distant from polar amino acids or backbone functional groups within the GCR1 binding pocket. Conversely, in the open and fully open conformations, the carboxylate, carbonyl, and hydroxyl groups of ABA are positioned closer to polar residues, enabling direct hydrogen-bond formation. These dis-tinct interaction patterns correlate with the different intracellular responses of GCR1 and, consequently, with the conformational state of GPA1.

4.

Equilibrated GCR1 binding site views for both the ABA–GCR1–GPA1–GDP–Mg²⁺ complexes with the corresponding GPA1 in its Open (Panels A and B) and fully open configurations (Panels C and D). Each complex is shown in its lowest-energy conformation after equilibration. Panels (A) and (C) show the 3D view, and Panels (B) and (D) show the 2D view, illustrating a series of HB interactions between ABA and amino acids within the GCR1 pocket. Orientation note: The molecular orientation is consistent across both figures: the CO group is positioned on the right side, the OH group on the left side, and the COO^– group is oriented backward. This description is provided to ensure clarity in interpreting the structures.

Thus, we conclude that HBs and/or SB of functional groups on the hormone with charged or polar amino acids within the GCR1 binding site are responsible for GPA1 activation. Here, we compare these results with those published in our previous article, where we described the activation of GPA1 mediated by the GA1 hormone binding to GCR1.

Then, the importance of this GCR1-ABA-mediated negative regulation is amplified when considering the reported capacity for GPA1 self-activation in plants. Various studies suggest that the plant Gα subunit (GPA1) can perform rapid and autonomous GDP/GTP exchange, unlike their mammalian counterparts. ,,, If GPA1 were to self-activate, the transient inactivation provided by Regulator of G protein Signaling (RGS1) proteins, which merely accelerate the intrinsic GTPase activity of the Gα subunit to return GPA1 to its GDP-bound state, would be insufficient for sustained control. Indeed, without an additional locking mechanism, the release of GPA1 from its RGS1 regulator could lead to a futile cycle of rapid self-activation and subsequent inactivation, resulting in inefficient and uncontrolled signaling. It is in this context that the ABA-mediated function of GCR1 proves indispensable. ABA binding to GCR1 not only induces GPA1 inactivation but also actively maintains it in an inactive state. It does so by forcing domain closure, stably trapping GDP-Mg²⁺, and reinforcing the ASP162–LYS288 SB. This represents a “hard lock” mechanism that effectively counteracts the tendency of GPA1 for self-activation, ensuring that the Gα subunit remains in a low-energy, inactive state after RGS1 action. Only the subsequent binding of GA1 to GCR1, by breaking this LYS288-ASP162 “ionic lock”, would then promote its controlled and specific activation.

In summary, GCR1 not only facilitates GPA1 activation with GA1 but is an indispensable element for the sustained and robust inactivation of GPA1 via ABA, a crucial mechanism for preventing uncontrolled self-activation within the intricate balance of plant hormonal signaling. This understanding positions us to design agronomic modulators precisely that, by influencing the interaction of GCR1 with GPA1 through either GA1 or ABA, enable unprecedented control over vital processes such as germination, growth, and stress adaptation, ushering in a new era in agricultural biotechnology. Therefore, the controlled activation of GPA1, indispensable for the transition from dormancy to growth, fundamentally depends on GCR1 as the sole control point that allows the exchange of ABA for GA1, when opti-mal environmental conditions prevail. This mechanism of fine, atomistically precise GPA1 switching, orchestrated by GCR1 to integrate antagonistic hormonal signals, underscores the profound sophistication of plant signaling and opens revolutionary avenues for the design of advanced agronomic modulators. Experimental validation of these predictions will not only confirm GCR1’s GPCR function but also pave the way for manipulating vital processes such as germination, growth, and stress adaptation with unprecedented precision, promising a new era in agricultural biotechnology.

3.1. Prediction of Residue Mutations to Validate the Signaling Pathway of the GPA1 Inactivation the ABA-GCR1

Although there are no Cryo-EM or X-ray structures for apo-GCR1, there is experimental evidence for a GPA1-GCR1 interaction. On the other hand, there is neither theoretical nor experimental evidence for the GPA1 inactivation mediated by ABA GA1-GCR1 interaction. To provide a means to validate the predicted GPA1 inactivation mediated by ABA-GCR1, we have identified mutations expected to decrease the stability in the ABA-GCR1-GPA1 structure, which may stabilize the activated GPA1.

The proposed mutations aim to disrupt electrostatic and hydrogen bonding interac-tions between GCR1 and GPA1. Specifically, by substituting charged residues (e.g., ASP/GLU/LYS/ARG or protonated His) with neutral residues (e.g., ASN/GLN/ALA), or removing donor/acceptor groups, the SBs and HBs predicted at the interface should be abol-ished. This disruption is expected to impair GCR1-mediated regulation of GPA1. Specifically, Table enumerates the most important mutations to modify the interactions between ABA and amino acids in the GCR1 binding site, along with the predicted deleterious and destabilizing mutations in the interaction between GPA1 and GCR1 to modify the GPA1 inactivation mediated by the ABA-GCR1 interaction.

3. Predicted Mutations on HB and SB Interactions to Validate the Inactivation of GPA1 Mediated by the Interaction of ABA to GCR1,

interaction between ABA–GCR1	destabilizing mutation	interaction between GCR1–GPA1	destabilizing mutation
SB HSP88–ABA(COO)	HSP88 → ALA	SB LYS45–GLU278	LYS45 → PHE
HB HSP163(N-backbone)–ABA(−OH)	ABA(C–OH) → ABA(C–CH)	ARG48–GLU379	ARG48 → ALA
		LYS113–GLU379	LYS113 → PRO
		ASP115–ARG374	ASP115 → ALA
		LYS213–ASP342	LYS213 → GLY
		ARG209–ASP342	ARG209 → ALA

^aColumn 1 of Table lists the stabilizing interactions between ABA and GCR1 within the orthosteric binding site, whereas column 3 lists the stabilizing interactions between GCR1 and GPA1 in the intracellular region. Based on these interactions, we propose targeted mutations in which the polar or charged residues listed in columns 1 and 3 are replaced by nonpolar residues, as shown in columns 2 and 4. These mutations are expected to reduce the stability of the ABA–GCR1 complex or to disrupt the GCR1–GPA1 interface, thereby impairing GPA1 inactivation. Specifically, in the ABA–GCR1 binding site (column 1), we identified one key salt bridge (SB) and one hydrogen bond (HB) that contribute to ligand sta-bilization. In addition, column 3 identifies several salt bridges that stabilize the GCR1–GPA1 interaction. We depict that mutating these residues to nonpolar amino acids such as alanine or glycine will strongly destabilize the GCR1–GPA1 complex, potentially leading to loss of function of GCR1 in maintaining GPA1 in its inactive state.

^bThe abbreviation HSP denotes protonated histidine (His⁺).

4. Conclusions

To confirm the GPCR-like role of GCR1, we examined the possibility of the ABA-GCR1-GPA1 complex being inactivated to prevent the GDP-GTP exchange and signaling. We did this by using metadynamics and MD simulations to predict the free energy as a function of the distance between the centers of mass of the Ras and α-helical subdomains of GPA1 (known to be involved in the G-protein regulation for plants). Our simulations show that decreasing the Ras and α-helical distance for the ABA-GCR1-GPA1 complex closed GPA1 preventing GDP/GTP exchange. We find that the closed inactive form of GPA1 is stabilized by the formation of several SBs between the Ras and helical domains, especially LYS288-ASP162, which is analogous to the ASP150-LYS270 SB found in the animal Gαi subunit to be essential in keeping the GP inactive. We find that the free energy surface as a function of Ras and α-helical distance favors closing the GPA1 to reach a stable minimum at 30 Å, which is sufficient to prevent GDP-GTP exchange and signaling. Thus, these calculations associate ABA-GCR1 interactions with inactivation of GPA1. Our metadynamics simulation provide insightful free energy surface profiles for the ABA-GCR1-GPA1-GDP-Mg²⁺ complexes, showing the interplay between the Ras and α-helical domains of GPA1. These free energy surfaces exhibit distinctive features, including four relative minima with the closed form exhibiting much lower free energy than the open and fully open configurations. Moreover, the energy barrier demarcating these relative wells is 10 kcal/mol.

This comparative analysis of free-energy landscapes shows that ABA promotes the tran-sition of GPA1 to its inactive conformations. These findings on ABA, GCR1, and GPA1 indicate that GCR1 mediates an ABA-induced inverse-agonist mechanism that inactivates GPA1. These results extend the inverse-agonist paradigm of animal GPCR signaling to plant G-protein regulation at atomistic resolution. The proposed ABA–GCR1–GPA1 inactivation mechanism is directly falsifiable through targeted disruption of the predicted salt-bridge and hydrogen-bond networks listed in Table . From the equilibrated complex, we propose residue mutations that should strongly affect the signaling pathway of GP regulation. These mutations can enable experimental testsusing gene-editing toolsof the predicted struc-ture of GCR1. Such tests could provide new evidence and insights into the role of GCR1 in plant physiology and its relationship to GPCRs. This knowledge may ultimately guide the design of improved inverse agonists of GPA1, offering greater experimental control over these processes.

Our research deepens the atomistic understanding of the GCR1 receptor in plants as a pivotal molecular switch, capable of interpreting and transducing antagonistic hormonal signals through the differential modulation of the Gα subunit, GPA1. We have elucidated that Gibberellin A1 (GA1) acts as an agonist, which, upon binding to GCR1, induces con-formational changes that, in turn, cause a crucial opening of the Ras and α-helical domains of GPA1, facilitating GDP-GTP exchange and the initiation of signaling. In fundamental contrast, Abscisic Acid (ABA) operates as a inverse agonist, which, by binding to GCR1, induces a rigid closure of the GPA1 Ras and α-helical domains. This compact confor-mation traps GDP in the nucleotide-binding pocket, effectively preventing GTP exchange and, therefore, maintaining GPA1 in an inactive state. The free energy landscape from metadynamics demonstrates that this closed and inactive state of GPA1 represents a deep, thermodynamically favored energy minimum, with an energy barrier of approximately 10 kcal/mol separating it from more open conformations.

The proposed mechanism, while supported by extensive in silico analysis, remains to be experimentally validated. Future studies should focus on biochemical and structural approaches, such as site-directed mutagenesis, protein–protein interaction assays, and crystallography or cryo-EM, to confirm the predicted HB and SB disruptions. Additionally, functional assays in planta or in representative cellular systems will be essential to deter-mine whether the proposed mutations indeed impair GPA1 regulation mediated by GCR1. Integrating these experimental findings with computational modeling will not only validate our hypothesis but also refine the mechanistic understanding of ABA–GCR1–GPA1 signal-ing.

This repository contains all the files required to perform metadynamics calculations and molecular dynamics simulations using the NAMD software. The information collected here is organized to facilitate the reproduction of simulations and the analysis of results related to the ABA-GCR1-GPA1-GDP-Mg²⁺ complex.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.jcim.5c02308.

• Additional figures are provided in the to this article, including a detailed description of the results of the metadynamics and MD simulations on newly con-structed ABA-GCR1-GPA1 complexes, and the thermodynamic analysis of the high-energy inactive, inactive, open and fully open structures of GPA1 bound to ABA-GCR1 complex. Included files.Input files for metadynamics calculations Configurations, and input files needed to run metadynamics simulations.Parameter Files:Include specific parameters for simulations of the complex studied. Structural files (.pdb and.psf) Initial structures of the ABA-GCR1-GPA1-GDP-Mg²⁺ complex, required for setting up the simulations.Metadynamics results pdb files corresponding to the configurations with the lowest free energies obtained from the metadynamics calculations. Files for molecular dynamics simulations:Inputs, parameters, and selected structures (pdb and psf), as the lowest-energy con-figurations from the metadynamics simulations to be used in molecular dynamics sim-ulations. The goal of this repository is to provide a comprehensive and accessible resource for those interested in replicating or building upon this work for future molecular simulation studies (PDF)

Pedro Hernández was responsible for all calculations, the creation of graphs and tables, and the writing of the initial draft. He also participated in editing the manuscript. Additionally, Pedro conducted the literature review, data analysis, and coordinated the research activities. William A. Goddard III, Andŕes Jaramillo-Botero, Carlos A. Arango, and Soo-Kyung Kim contributed to the conceptualization of this paper and served as advisors to Pedro Hernández. They provided critical feedback and helped shape the research, analysis, and manuscript. They also assisted in editing the manuscript.

The authors declare no competing financial interest.