Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Dec 3.
Published in final edited form as: Structure. 2013 Nov 7;21(12):10.1016/j.str.2013.09.020. doi: 10.1016/j.str.2013.09.020

The role of a sodium ion binding site in the allosteric modulation of the A2A adenosine G protein-coupled receptor

Hugo Gutiérrez-de-Terán a,b,c,§, Arnault Massink d,§, David Rodríguez a,e, Wei Liu b, Gye Won Han b, Jeremiah S Joseph b, Ilia Katritch b, Laura H Heitman d, Lizi Xia d, Adriaan P IJzerman d, Vadim Cherezov b, Vsevolod Katritch b, Raymond C Stevens b,*
PMCID: PMC3858454  NIHMSID: NIHMS531997  PMID: 24210756

SUMMARY

The function of G protein-coupled receptors (GPCRs) can be modulated by a number of endogenous allosteric molecules. In this study, we used molecular dynamics, radioligand binding and thermostability experiments to elucidate the role of the recently discovered sodium ion binding site in the allosteric modulation of the human A2A adenosine receptor, conserved among class A GPCRs. While the binding of antagonists and sodium ions to the receptor was non-competitive in nature, the binding of agonists and sodium ions appears to require mutually exclusive conformational states of the receptor. Amiloride analogs can also bind to the sodium binding pocket showing distinct patterns of agonist and antagonist modulation. These findings suggest that physiological concentrations of sodium ions affect functionally relevant conformational states of GPCRs, and can help to design novel synthetic allosteric modulators or bitopic ligands exploiting the sodium ion binding pocket.

INTRODUCTION

Cellular responses to a wide variety of extracellular signals are mediated by the superfamily of seven transmembrane helical receptors coupled to intracellular G proteins (G protein-coupled receptors, GPCRs). It is now well recognized that many GPCRs function as “allosteric machines” with the orthosteric binding pocket representing just one of the many sites for possible signal modulation and pharmacological intervention. Thus, molecules targeting other (allosteric) sites can modulate binding of native orthosteric ligands and shift the delicate equilibrium between active and inactive states of GPCRs (Christopoulos, 2002). Potential therapeutic advantages include a gain in target selectivity, the “ceiling effect”, and preservation of the spatiotemporal profile of intercellular signaling (Conn et al., 2009; Gao and Jacobson, 2006; Goblyos and Ijzerman, 2011; Jacobson et al., 2011). Some endogenous chemical entities, such as ions or lipids, have also been demonstrated to act as allosteric modulators of GPCRs (Christopoulos, 2002; Gao and Ijzerman, 2000; Neve et al., 2001), but the structural basis and functional importance of these interactions are not well understood.

Recent advances in protein engineering and membrane protein crystallography (Cherezov, 2011; Rosenbaum et al., 2007; Tate and Schertler, 2009) have led to the elucidation of a growing number of experimental GPCR structures (Katritch et al., 2013), contributing to the alluring perspective of structure-based drug design (Congreve et al., 2011), and the deciphering of molecular mechanisms underlying conformational equilibrium (Dror et al., 2009). Several receptors, including one of the best-characterized GPCRs, the A2A adenosine receptor (A2AAR), have been crystallized in inactive (Congreve et al., 2012; Dore et al., 2011; Hino et al., 2012; Jaakola et al., 2008) and active-like (Lebon et al., 2011; Xu et al., 2011) conformations. The conformational changes associated with A2AAR activation mirrored similar structural changes observed in other receptors, namely the β2-adrenergic receptor (β2AR) (Rasmussen et al., 2011) and rhodopsin (Park et al., 2008). Recently, the 1.8 Å resolution structure of inactive A2AAR in complex with ZM241385 (Liu et al., 2012) revealed the presence of a sodium ion bound to the core of the transmembrane (TM) bundle, coordinated by Asp2.50 and other side chains highly conserved in class A GPCRs (Mirzadegan et al., 2003) and by a cluster of structural water molecules (Angel et al., 2009; Pardo et al., 2007). The allosteric effect of sodium ions has been described previously in A2A (Gao and Ijzerman, 2000) and A1 adenosine receptors (Barbhaiya et al., 1996), as well as in GPCRs from other subfamilies such as dopamine D2 (Neve et al., 2001; Selent et al., 2010), opioid (Pert et al., 1973; Snyder and Pasternak, 2003) or α-adrenergic receptors (Horstman et al., 1990; Tsai and Lefkowitz, 1978), among others. Similarly, the positively charged, synthetic small molecule amiloride and its analogs have also been found as allosteric modulators of agonist and antagonist binding in a number of receptors, including A2AAR, and were shown to compete with sodium ions for the same binding site (Gao and Ijzerman, 2000). Moreover, mutation of Asp2.50 to either asparagine or alanine has been shown to reduce or abrogate the allosteric effects of sodium ions or amiloride in many GPCRs (Gao et al., 2003a; Neve et al., 2001; Nie and Lewis, 2001; Proulx et al., 2008). While the 1.8 Å structure of A2AAR provides a static picture of sodium interactions with the receptor, the dynamic nature of the sodium ion-water cluster, its effect on binding of orthosteric agonists and antagonists, and its functional role are poorly understood. Molecular dynamics studies, supported by biochemical and biophysical characterization of the complexes here provide a molecular framework for the allosteric effects of sodium and amilorides, which can ultimately aid in the discovery of new compounds targeting this site (Goblyos and Ijzerman, 2011).

RESULTS

Molecular dynamics studies

A series of molecular dynamics (MD) simulations were designed to evaluate the functional role of the sodium ion in the A2AAR, as summarized in Table 1. Three MD replicates of 40–100 ns length each were run for each setup, in order to increase the statistics of the sampling, with a total simulation time exceeding 2.8 μs. The effect of the sodium ion on the conformational equilibrium of the receptor was examined considering both the inactive and active-like conformation of the A2AAR (simulations MDS1-MDS7). The influence of the orthosteric ligands in this process was examined by comparing the MD simulations with and without the antagonist ZM241385 (MDS1 and MDS2) or the agonists UK432097 and NECA (MDS5 through MDS7). In addition, the allosteric effect of amiloride and its derivative HMA [5-(N,N-hexamethylene)amiloride] in the antagonist-bound conformation was examined in MDS8 through MDS12, with two distinct chemotypes of antagonists (i.e., ZM241385 and caffeine).

Table 1.

Setup of the different MD simulations reported in this work.

Receptor Conformation Ligand Allosteric Modulator Replicas, length
MDS1a Inactive ZM241385 Na+ 3×100 ns
MDS1ba Inactive ZM241385 Na+ 3×100 ns
MDS2 Inactive --- Na+ 3×100 ns
MDS3 Inactive ZM241385 --- 3×100 ns
MDS4 Inactive --- --- 3×100 ns
MDS5 Active UK432097 Na+ 3×40 ns
MDS6 Active NECA Na+ 3×40 ns
MDS7 Active --- Na+ 3×40 ns
MDS8 Inactive ZM241385 Amiloride 3×100 ns
MDS9 Inactive Caffeine Amiloride 3×40 ns
MDS10 Inactive Caffeine --- 3×40 ns
MDS11 Inactive ZM241385 HMA 3×100 ns
MDS12 Inactive Caffeine HMA 3×40 ns
Open in a new tab
a

MDS1b corresponds to the same system as MDS1, but considering the physiological saline concentration of 150mM.

The sodium binding site in the A2AAR inactive conformation

In the high resolution crystal structure of inactive A2AAR (Liu et al., 2012), the sodium ion is directly coordinated by Asp522.50 and Ser913.39 (first shell residues) and three structured water molecules (Figure 1A). Water molecules also bridge interactions with residues in the second (Trp2466.48 and Asn2807.45), and the third shells (Thr883.36 and Ser2817.46). Overall, the sodium ion/water binding pocket is formed by 15 out of the 34 amino acids that are conserved in the majority of non-olfactory class A GPCRs (Mirzadegan et al., 2003) and their conformation is similar in most GPCRs crystallized in the inactive state (Figure 1B–D and Supplemental Figure S1).

Figure 1.

Figure 1

Open in a new tab

Structure and conservation of the central sodium ion-binding allosteric pocket in Class A GPCRs. (A) The Na+ distorted octahedral coordination as in the A2AAR crystal structure: The first shell is occupied by two conserved polar residues (green) and three water molecules, which contact with a second shell of residues (cyan), or with a second layer of water molecules connecting with the third shell of residues (magenta). (B) Sequence conservation of the 15 residues lining the binding pocket among inactive GPCR crystal structures. (C) Structure of the A2AAR complex with ZM241385, showing residues with higher than 50% conservation in all Class A receptors as sticks with green carbons. (D) A close-up of the central allosteric pocket (transparent blue surface), showing the side chains located within 5 Å from the ten waters of the sodium ion-water cluster (green sticks: A2AAR; gray thin lines: the corresponding side chains of the overlaid GPCR crystal structures depicted in panel B). See also Supplementary Figure S1.

Analysis of the sodium ion’s mobility and its coordination state reveals a high level of stability for the ion when bound to the receptor’s inactive conformation. The dominant charge-charge interaction with Asp522.50 is clearly maintained along the simulation runs in the different MD trajectories, while the side chain oxygen atoms of Ser913.39 and Asn2807.45 alternate direct interactions with the ion (Figure 2A). More precisely, the ion fluctuates between a coordination state as seen in the crystal structure, which we will refer to as position c1, and a second state that we will refer to as position c2. This fluctuation involves an exchange between the sodium ion and the water molecule W52, initially linked to the OD1 of Asn2807.45, as shown in Figure 2A by the overlay of the electron density of the crystal structure with the volumetric density map calculated from the MD simulations. In position c2, the ion is still coordinated by Asp522.50 (OD1), while it is the OD1 of Asn2807.45 (occasionally replaced by a new water molecule) that participates in the first coordination shell, which is completed with three other water molecules. The radial distribution function indicates the average sodium—oxygen distance is 2.4 Å, with the first coordination shell predominantly formed by 4 or 5 oxygen atoms (see supplemental Figure S2), in agreement with the geometric analysis of sodium ion binding sites found in the Protein Data Bank (PDB) (Harding, 2006). The consistent observation of these two dominant coordination modes suggests a possibility of fast exchange between the sodium ion and the water molecule, supported by the significantly shorter distance (2.2 Å) observed in the crystal structure for the Na+—O(W52) pair. Interestingly, this fast exchange occurs regardless of the presence of the antagonist ZM241385 in the orthosteric binding site (Table S1). In the setup MDS1b, we evaluated the effects of a physiological saline concentration in the simulations, and observed no difference in results (see supplemental Figure S2 and Table S1). Moreover, none of the sodium ions from the extracellular solvent environment could cross the narrow channel into the allosteric pocket during 100 ns simulations, suggesting that exchange of sodium ions in A2AAR may occur on a longer time scale.

Figure 2.

Figure 2

Open in a new tab

The two coordination states of the sodium ion as observed in MD simulations. (A) Volumetric density map (isosurface contoured at 0.3 Å−3 value, blue) corresponding to the sodium occupancy as calculated from simulations MDS1 (see Table 1). The starting crystal structure is displayed, together with the electron density (contoured at 1 σ level, black). (B) The time-evolution of the distances between the sodium ion and Asp2.50 (blue), Ser3.39 (red) and Asn7.45 (green), shown for the 3 independent replicates (R1–R3) of MDS1. The corresponding distances are denoted as dashed lines in panel (A) with the same color code, and are the source of the data in Table 2. The RMSD of the ion with respect to its crystallographic position is indicated with a black line, while the horizontal bar at 1.8 Å (the resolution of the parent crystal structure) denotes the limit for the crystallographic coordination state (position c1). See also Table S1 and Figure S2.

To investigate a possible effect of the sodium ion on the stability of the inactive conformation of the receptor, the MD simulations described above (MDS1 and MDS2) were compared with the corresponding simulations of the inactive A2AAR in the absence of the ion (MDS3 and MDS4). The most pronounced difference in local conformational dynamics was located in the region of the sodium ion: the highly conserved residue Trp2466.48 in the second sphere of solvation experienced a rotameric transition from the initial g+ rotamer to the trans (t) conformation, observed in two out of three independent simulations of the apo receptor without sodium ion (MDS4, see Figure 3). This rotameric change appears connected to a rotation of residue Asn2807.45 from the initial g- rotamer to either trans or g+ (Figure 3A and C). In contrast, rotamer changes in Tpr2466.48 and to a lesser extent Asn2807.45 were not observed when the sodium ion was bound to the allosteric site (MDS2), suggesting that the sodium ion contributes to a stabilization of these two residues (Figure 3A and B). Similarly, the furyl moiety of the antagonist ZM241385 stabilized the g+ rotamer of Trp2466.48 through Van der Waals interactions (simulations MDS1 and MDS3). Consequently, no movements in this microenvironment occured with ZM241385 bound, regardless of the presence of the sodium ion.

Figure 3.

Figure 3

Open in a new tab

Rotameric transitions of Trp6.48 and Asn7.45 in the apo simulations of the inactive A2AAR. (A) Populations of the initial conformational states in the simulations with (MDS2) and without (MDS4) the sodium ion, and the number of waters in the ion binding site (each data is an average of the 3 MD replicas). Note that Trp6.48 only finds the trans conformation when not in the initial g+ conformation, while Asn7.45 is more flexible and can be found in either trans, g+ or the initial g− conformations. (B) Representative snapshot (magenta) of the conformation of these two residues in MDS2 and (C) MDS4, with the reference crystal structure overlaid in light gray.

The sodium ion binding site in the A2AAR active-like conformation

Analysis of the agonist-bound structures of A2AAR (Lebon et al., 2011; Xu et al., 2011) indicates that the activation–related changes in helix VII partially collapse the sodium ion pocket, making it incompatible with ion binding (Liu et al., 2012). In order to further evaluate this effect, the sodium ion-water cluster was simulated in the context of the active-like conformation of A2AAR with agonists UK432097 (MDS5) or NECA (MDS6), or without any agonist (MDS7, see Table 1). The simulations show that in the active-like state the receptor cannot bind the sodium ion, as evidenced by two alternative events that occur in the early stages of production runs. In the first event, (observed in 1 out of 3 replicas in MDS5, in 2 out of 3 replicas in MDS6, and in all 3 replicas in MDS7) the ion escaped from the proposed binding site. In the alternative event, observed in the remaining simulations, the ion remained in the allosteric binding site, but a conformational change of helix VII occurred, resembling the inactive-like conformation of helix VII observed in all antagonist-bound A2AAR structures. In particular, the region between His2787.43 and Asn2847.49 undergoes an outward movement driving helix VII apart from helix III and expanding the pocket cavity (Figure 4). In the MDS5 and MDS6 simulations, this rearrangement was also accompanied by a loss of contact between His2787.43 and the O2′ of the ribose moiety of the agonist, suggesting that sodium ion binding destabilizes activation-related movements and agonist binding. Note that all conformational events described here occurred within the first 5–10 ns of the simulation, justifying that a total simulation time of 40 ns was sufficient to properly sample the sodium binding site in the active-like system. These findings indicate that the binding of sodium ions and agonists each require a different conformational state of the receptor and, therefore, are mutually exclusive.

Figure 4.

Figure 4

Open in a new tab

The movement of helix VII (orange) in order to accommodate the sodium ion. (A) Starting (grey) and ending (rainbow) conformations of the agonist-bound A2AAR in the presence of sodium ion (MDS5), with the inactive crystal structure denoted in anthracite. (B) The distance between helices III and VII (X axis, Cα of residues Ile3.40 and Asn7.45 as indicated by a dashed line in the 3D-structure) is plotted against the backbone RMSD of the motif His7.43-Asn7.49, (Y axis), using as a reference the active conformation of A2AAR. Each dot is a snapshot extracted every 0.5 ns, with the time evolution depicted by the shading code (light grey -> black). Starting and ending conformations indicated with an asterisk and a triangle, respectively.

Amiloride and HMA as A2AAR allosteric modulators

Amiloride and its derivatives are known to be non-specific GPCR modulators (Garritsen et al., 1991). A binding mode of amiloride and its bulkier analog HMA to the A2AAR was determined by flexible side chain docking where the charged guanidinium group of amiloride interacted with Asp522.50 (Liu et al., 2012). An additional hydrogen bond was also predicted between amiloride and the Trp2466.48 side chain (see Figure 5A), which is shifted ~1.5 Å towards the orthosteric pocket as a result of induced fit. This tight binding of amilorides is clearly not compatible with the collapsed allosteric pocket observed in the active-like conformation of the A2AAR, in an even more pronounced way than in the case of sodium ions. Therefore, we explored this docking hypothesis with a series of MD simulations of the inactive A2AAR in the presence of different orthosteric antagonists, i.e. the ternary complexes A2AAR—ZM241385—amiloride (MDS8) and A2AAR—caffeine—amiloride (MDS9). The effect of amiloride on the binding of antagonists turned out to be complex and varied between the different antagonist chemotypes tested. Figure 5D shows a comparison of the RMSF (root mean square fluctuation) of compound ZM241385 with no allosteric modulator present (MDS3), and with either sodium ion (MDS1), amiloride (MDS8) or HMA (MDS11) present in the proposed allosteric site. A significant increase in the mobility of ZM241385 (p<0.05) was observed in the presence of both amilorides. This effect is likely due to the influence of the side chain of Trp2466.48, which is the only residue that interacts with ZM241385 and amilorides simultaneously (in particular with the N-hexamethylene substituent of HMA, see Figure 5C), leading to the hypothesis that this highly conserved residue acts as an important link between the two sites. Consequently, binding of antagonists that do not directly interact with Trp2466.48, for example, caffeine (Dore et al., 2011), should be less affected by the presence of amiloride in the allosteric site (Figure 5B). Indeed, we found no statistically significant difference for the mobility of caffeine as a function of the presence of amilorides (Figure 5D). Due to the high mobility of caffeine, which is small in size, has few receptor contacts, and displays low affinity, the simulation time in this particular system was limited to 40 ns time scale.

Figure 5.

Figure 5

Open in a new tab

Impact of amiloride and HMA in the binding of antagonists. (A) Amiloride docking (magenta carbons) induces a shifted position of Trp2466.48 side chain, revealing potential steric clashes with the orthosteric ligand ZM241385 (yellow carbons, superimposed from the crystal structure with PDB code 4EIY). (B) Same conformation of the amiloride- bound A2AAR, with the caffeine pose (green carbons) superimposed from the crystal structure of A2AAR/caffeine complex (PDB code 3RFM). (C) Flexible docking of HMA (magenta carbons) is predicted to further shift Trp2466.48 and interfere with ZM241385 binding. (D) Mobility of the antagonists in the presence or absence of amiloride, HMA and sodium in the allosteric pocket, calculated as the RMSF from the MD simulations (dark shaded bars for ZM241385; light shaded bars for caffeine). The error bars indicate the standard deviation estimated from three MD replicas (n=3); ami=amiloride; ZM= ZM241385. Significantly different from the control simulation (i.e. absence of any orthosteric ligand) in a student t-test with *p < 0.05.

Biochemical studies

Radioligand binding experiments were performed to examine the effects of sodium ions and amiloride derivatives on antagonist, [3H]ZM241385, and agonist, [3H]NECA, binding to the A2AAR. In order to experimentally assess the dependence of ligand binding on the presence of a sodium ion in the allosteric site, we performed equilibrium displacement studies with increasing concentrations of NaCl (Figure 6A). These experiments show a full displacement of [3H]NECA by sodium ions, with an IC50 value of 49 ± 7 mM (Table S2 and Figure 6A). In contrast, there is an enhancement of antagonist [3H]ZM241385 binding, especially at higher sodium ion concentrations. All these results correlate with the MD conclusions that the sodium ion selectively stabilizes the antagonist-bound receptor state.

Figure 6.

Figure 6

Open in a new tab

Equilibrium displacement of [3H]ZM241385 and [3H]NECA by allosteric modulators. (A) NaCl, (B) amiloride, and (C) HMA. Experiments performed in duplicate on human A2AARs transiently expressed in HEK293T cell membranes. Associated IC50 values listed in Table S2.

Saturation binding experiments performed with [3H]NECA also show that the presence of NaCl significantly reduces [3H]NECA binding to the A2AAR (Table 2 and Figure S3). Interestingly, this reduction in agonist binding was due to an increase of the KD value while the radioligand’s Bmax value remained at a control level, within experimental error. This profile of pharmacological parameters usually implies a competitive interaction between the two ligands. However, in this case the binding sites of the two ligands are not overlapping, therefore the observed “mutually exclusive binding” suggests that the sodium ion-bound conformation of A2AAR is not compatible with agonist binding, and vice-versa. In contrast, no significant effect of sodium ions was observed on the binding of the antagonist radioligand [3H]ZM241385 to the A2AAR (Table 2).

Table 2.

Saturation of [3H]ZM241385 and NECA spiked with 25% [3H]NECA binding to human A2AARs transiently expressed in HEK293T cell membranes in the absence and presence of ZM241385, NaCl, amiloride, and HMA. See associated experiments in Figure S3.

[3H]ZM241385 [3H]NECA
KD (nM) Bmaxa (%) KD (nM) Bmaxa (%)
Control 1.3 ± 0.4 100 ± 10 84 ± 11 100 ± 2
+ 10 nM ZM 8.3 ± 2.2** 85 ± 7 123 ± 9* 95 ± 3
+ 30 mM NaCl 1.2 ± 0.3 120 ± 11 213 ± 10*** 106 ± 4
+ 100 mM NaCl 0.8 ± 0.1 87 ± 3 471 ± 53*** 114 ± 9
+ 30 μM Amiloride 1.9 ± 0.8 61 ± 4* 290 ± 52** 119 ± 13
+ 3 or 4 μM HMA† 4.1 ± 0.9* 78 ± 9 246 ± 2*** 110 ± 1*
Open in a new tab
a

% of Bmax of control (= 100%)

†

3 μM HMA for [3H]ZM241385 and 4 μM HMA for [3H]NECA

Significantly different from control in a student t-test with

*

p < 0.05,

**

p < 0.01, or

***

p < 0.001.

Values are means ± S.E.M. of 2–5 separate assays performed in duplicate.

We also characterized the influence of amiloride and HMA on radioligand binding to the A2AAR. The two compounds inhibited the binding of both agonist and antagonist radioligands in displacement assays, albeit with different potencies (Figure 6B and 6C and Table S2). HMA proved to be more active than amiloride in both cases, and displayed the highest potency (2.4 μM) with the agonist [3H]NECA as the radiolabel. Radioligand saturation experiments were performed in the presence of the amiloride analogs, revealing distinct differences between the two radioligands (Table 2 and Figure S3). The interaction between amiloride and [3H]ZM241385 was non-competitive in nature as the radioligand’s Bmax value was significantly reduced with little effect on the KD value, whereas unlabeled ZM241385, serving as a control orthosteric ligand, showed all traits of a competitive ligand. In this experimental setup, HMA behaved somewhat in between ZM241385 and amiloride, showing a small but significant shift in KD and a non-significant change in Bmax value. Similar experiments with the radiolabeled agonist [3H]NECA produced a very different outcome. While unlabeled ZM241385 appeared competitive with [3H]NECA, as expected for two orthosterically binding compounds, the same was true for the interaction between the amilorides and [3H]NECA. Hardly any effect was observed on NECA’s Bmax value, whereas all KD values were significantly increased, which we attribute to the mechanism of “mutually exclusive binding” discussed above for the case of agonist and sodium ion binding.

Biophysical studies

The effects of sodium ions, orthosteric ligands and amiloride analogs on A2AAR stability were analyzed with a series of thermal stability assays (Alexandrov et al., 2008). Increasing sodium ion concentrations induced a significant increase in the thermostability of the unliganded A2AAR-BRIL complex, as shown in Figure 7A. A two-phase response was observed, with a modest increase in thermostability from 52 to 57 °C at sodium ion concentrations below the physiological concentration of 150 mM, and a further more substantial increase to 65 °C upon addition of higher concentrations up to 500 mM NaCl.

Figure 7.

Figure 7

Open in a new tab

Effect of allosteric binders on A2AAR thermostability measured by CPM assays. (A) Titration of NaCl effect on A2AAR thermostability, mean ± S.E.M. shown for measurements performed in triplicate. (B) Effect of NaCl (150 mM), amiloride (100 μM), caffeine (500 μM), ZM241385 (50 μM), UK432097 (50 μM) and combinations thereof on A2AAR thermostability.

In order to evaluate the effects of allosteric modulators on various receptor-ligand complexes, we measured the thermal stability of the A2AAR-BRIL construct in the presence or absence of sodium ions and/or amiloride, and their combinations with the orthosteric ligands caffeine (antagonist), ZM241385 (antagonist), or UK432097 (full agonist) (Figure 7B). Sodium ions and amiloride each increased A2AAR thermostability by 5–6 °C, but their effect when combined was non-additive, corroborating the suggested competition of these two charged molecules for the same binding site. In contrast, the addition of caffeine in the presence of saturating concentrations of amiloride or sodium ions caused a further 6 °C increase in the thermostability of the complex, suggesting an additive stabilizing effect of the orthosteric caffeine and allosteric ligands. Also, while we observed an additive effect of sodium ions and ZM241385, amiloride did not contribute to the stability of A2AAR in saturating concentrations of ZM241385, in agreement with unfavorable indirect interactions between amiloride and ZM241385. Finally, neither sodium ions nor amiloride had a stabilizing effect on the A2AAR saturated with UK432097, likely because this agonist precluded binding of these allosteric modulators.

DISCUSSION

The 1.8 Å resolution structure of A2AAR in complex with the antagonist ZM241385 (Liu et al., 2012) revealed a new highly conserved sodium ion binding site and provided a unique opportunity to examine the molecular mechanism of the allosteric modulation of sodium ions in this receptor. Our MD simulations suggest that the electron density assigned to the water molecule closest to sodium ion in the crystal structure (W52) could correspond to a second resonance position of the ion, which would involve direct interaction with another conserved residue, Asn2807.45. Although we could not find a clear evidence of such dynamic rearrangements in the A2AAR crystal structure, this may reflect the “frozen” state of the sodium ion-water cluster at the low temperatures (~100K) used in cryo-crystallography. While further crystallographic studies at room temperature may help to validate such subtle effects experimentally (Fraser et al., 2011), the thermal fluctuations explored by the MD simulations under physiological conditions are particularly suited to explore this phenomenon (Ulmschneider et al., 2013). The potential dynamic nature of the sodium ion and water network in the allosteric pocket could partially explain why the ion was not observed previously in lower resolution GPCR structures. A closer look at this region in two recently published, antagonist-bound GPCR structures, i.e. carvedilol—β1-adrenergic receptor (Warne et al., 2012) and β FNA—μ opioid receptor (Manglik et al., 2012), reveals that they are indeed compatible with the presence of a sodium ion (Figure S1). Most recently, a similar configuration of the sodium ion/water network was also identified in other ligand complexes of the β1-adrenergic receptor (Christopher et al., 2013).

The communication between the highly conserved class A GPCR residues Asp2.50, Trp6.48 and to a lesser extent Asn7.45, had been suggested to occur through a cluster of water molecules (Pardo et al., 2007). The current MD results strongly support the preference of the sodium ion for the inactive conformation of this micro-environment, at least in the A2AAR, suggesting that it contributes to its stabilization. The increased dynamic flexibility of Trp2466.48 and Asn2807.45 in the absence of the sodium ion is consistent with previous MD simulations of the A2AAR (Rodriguez et al., 2011), and agrees well with similar MD simulations in the dopamine D2 receptor with explicit consideration of a sodium ion (Selent et al., 2010). The conformational flexibility of Trp6.48 has been associated with the initial steps of the activation mechanism of GPCRs (Schwartz et al., 2006; Shi and Javitch, 2002), probably by facilitating the higher order conformational changes observed in helix VI between inactive and active states. Conversely, the presence of the sodium ion and coordinating water molecules in this pocket hampers an activation-related inward movement of helix VII towards helix III (Lebon et al., 2011; Xu et al., 2011), as indicated by MD simulations. Instead, we observed an “inactivation” movement of helix VII in the agonist-bound conformation that led to the loss of key agonist—receptor interactions, suggesting that the simultaneous binding of the allosteric ion and the orthosteric agonist to the same receptor molecule is unlikely. Such a structural mechanism explains the negative allosteric effect of sodium ions on agonist binding to the A2AAR observed here (Figure 6A and Table 2) and in earlier studies (Gao and Ijzerman, 2000; Liu et al., 2012).

In radioligand binding experiments performed with both agonist and antagonist radioligands, sodium ions differentially affect radioligand binding to the A2AAR, providing further insights into its allosteric effect: they induced an increase in [3H]ZM241385 binding, but abrogated [3H]NECA binding in a concentration-dependent manner with an IC50 value of approximately 50 mM. This potency is in the same range of sodium ion concentrations that cause an increase in receptor thermostability (Figure 7), indicating that sodium ion binding could mediate both effects. Moreover, this IC50 is about 1/3 of the extracellular physiological concentration of sodium, which suggests that about 75% of the A2AARs are in a sodium ion occupied state, provided unrestricted access to the sodium binding site from the outside of the cell as recently proposed for the dopamine D2 receptor (Selent et al., 2010). Thus, it is likely that sodium is significantly involved in modulating the physiological state of the A2AAR and possibly other class A GPCRs. Interestingly, the saturation binding experiments (Table 2) demonstrate that the presence of sodium ions (30 and 100 mM) reduces [3H]NECA affinity (KD) rather than its Bmax value, while no affinity reduction was observed for [3H]ZM241385. This fact, together with the observed increase of [3H]ZM241385 binding at high sodium ion concentrations (Figure 6A), is in good agreement with an earlier study (Gao et al., 2000) where the slight affinity increase of this radioligand at even higher [1 M] sodium concentrations was due to a decrease in its dissociation rate (Gao and Ijzerman, 2000). The structural information and the computational results provided in this study suggest that binding of agonists and sodium ions can be considered as “mutually exclusive” (Neubig et al., 2003). This mechanism is further supported by thermal stability assays, showing that antagonists ZM241385 and caffeine display additive stabilizing effects with sodium ions, whereas the agonist UK432097 stabilizes the receptor but shows no additive effect when any allosteric ligand is added.

Similar to sodium ions, the diuretic drug amiloride is an allosteric modulator of several GPCRs, including such diverse subfamilies as adenosine, aminergic and gonadotropin-releasing hormone receptors (Gao and Ijzerman, 2000; Gao et al., 2003b; Heitman et al., 2008; Hoare et al., 2000; Howard et al., 1987; Pauwels, 1997). The present study in the human A2AAR shows that both amiloride and its derivative HMA negatively modulate agonist binding (Table 2), with HMA being more potent than amiloride in radioligand displacement studies (Figure 6) as previously suggested with kinetic studies in the rat A2AAR (Gao and Ijzerman, 2000). The saturation studies with [3H]ZM241385 as the radioligand (Table 2) also point to a non-competitive interaction between amiloride and the antagonist binding site. Moreover, the additive increase in A2AAR thermostability in the presence of both amiloride and the orthosteric antagonist caffeine (Figure 7) suggests that both allosteric and orthosteric ligands can bind simultaneously. These findings are in concert with docking and MD simulations (Figure 5), which suggest that amiloride and HMA bind in the allosteric sodium pocket, with the charged guanidinium group anchored by the carboxyl of Asp2.50. Although the proposed amiloride binding does not directly overlap with the orthosteric ligand binding site, our simulations indicate that amiloride derivatives can impact orthosteric ligand binding indirectly, primarily via modulation of Trp2466.48 conformation. This explains a more pronounced negative allosteric effect of HMA, which has extensive steric interactions with the Trp2466.48 side chain. Overall, the distinct allosteric effects of amilorides are probably the result of a delicate balance between an improved stability of the inactive conformation and an indirect (non-competitive) interference with the orthosteric site for antagonists.

In summary, a unique combination of biochemical and thermal stability data with MD simulations, based on both inactive and the active-like crystal structures of the A2AAR, provided new mechanistic insights into the role of the allosteric sodium ion in the conformational equilibrium of the receptor. Our findings suggest that the binding of either the sodium ion or amilorides to the allosteric pocket selectively stabilizes the inactive conformation of the receptor, and this allosteric effect is responsible for the observed reduction in orthosteric agonist binding. Comprehensive experimental and theoretical analyses of A2AAR in simultaneous complex with both allosteric modulators and orthosteric ligands also explain, on a molecular basis, the distinct interaction profiles observed between allosteric amiloride derivatives and different orthosteric antagonists. These observations pave the way for a better understanding of GPCR allosteric control, which may be used in the design of novel allosteric modulators to more precisely tune receptor function.

EXPERIMENTAL PROCEDURES

Computational Simulations

The standard amino acid sequence numbering for the human A2AAR is used in the text, with the Ballesteros and Weinstein residue numbering for GPCRs (Ballesteros and Weinstein, 1995) shown in superscript if the residue belongs to a transmembrane helix. The inactive structure of the A2AAR in complex with ZM241385 and a sodium ion [PDB code 4EIY (Liu et al., 2012)], was refined in order to model the missing loops and add protons as detailed in the Supplemental Experimental Procedures section, prior to MD simulations. When the sodium ion was not considered, manual replacement with a water molecule, also maintaining the surrounding water molecules, was followed by energy minimization to fully optimize the H-bond network in the allosteric site. In the simulations with amiloride, the initial conformation of the A2AAR—amiloride complex was used as proposed previously by flexible docking (Liu et al., 2012). The starting coordinates of the antagonist caffeine were obtained by superimposing the A2AAR—caffeine complex [PDB code 3RFM (Dore et al., 2011)] with the A2AAR structure described above using PyMOL (The PyMOL Molecular Graphics System version 1.4, Schrödinger). The MD simulations of the active-like conformation were performed starting from the A2AAR in complex with the stabilizing agonist UK432097 [PDB code 3QAK (Xu et al., 2011)]. The starting coordinates of the sodium ion and coordinating water molecules were transposed from the inactive structure by structural superimposition with 4EIY. Finally, the A2AAR in complex with the agonist NECA was obtained by “morphing” the A2AAR-ZM241385 structure (PDB code 4EIY) to the active-state A2AAR-UK432097 conformation (PDB code 3QAK), and subsequent reconstruction of the intracellular loop 3 as detailed in the Supplemental Experimental Procedures. Note that the available crystal structure of NECA in complex with thermostabilized A2AAR [PDB code 2YDV, (Lebon et al., 2011)] was not suitable for this analysis because of a highly distorted conformation of the allosteric site, which includes a deformed helix VII backbone due to a cis-Proline in the NPxxY motif.

Membrane insertion and all MD simulations were performed with the GROMACS software (Hess et al., 2008), using our original protocol for the MD simulations of GPCRs (Rodriguez et al., 2011) as adapted in the PyMemDyn program (Gutierrez-de-Teran et al., 2013). The final systems, consisting of approximately 50,000 atoms (~74% belong to solvent molecules, ~15% to lipids and ~11% to protein and ligand atoms), were energy minimized and equilibrated for a total of 5 ns, with specific details provided in the Supplemental Experimental Procedures. The production phase of unrestrained MD simulations followed for 100 ns simulation time (shorter production times were considered in certain cases, see Table 1 and explanation in main text). MD simulations were performed under the OPLSAA force field (Kaminski et al., 2001), with ligand parameters obtained with Macromodel (Schrödinger, 2009), lipid parameters adapted from Berger (Berger et al., 1997) together with the use of the half-ε double-pairlist method (Chakrabarti et al., 2010) and the SPC water model (Berendsen et al., 1981). The periodic boundary conditions (PBC) were implemented with hexagonal prism-shaped boxes in the isobaric NPT ensemble, using a Nose-Hoover thermostat (Nose and Klein, 1983) with a target temperature of 310 K using. Electrostatic interactions beyond a cutoff of 12 Å were estimated with the particle mesh Ewald (PME) method. All MD analyses were conducted with several GROMACS and VMD (Humphrey et al., 1996) utilities. Molecular superimpositions, trajectory visualizations and molecular images were performed with PyMOL.

Cell growth and transfection

HEK293T cells were grown in culture medium consisting of Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% newborn calf serum (NCS), 50 μg/ml streptomycin and 50 IU/ml penicillin at 37 °C and 7% CO2. Cells were subcultured twice a week at a ratio of 1:15 on 10 cm ø plates. Cells were transfected with the wild type A2AAR-plasmid (pcDNA3.1, 1 μg) using the calcium phosphate precipitation method (Sambrook and Fritsch, 1989).

Membrane preparation

Cells were detached from plates 48 h after transfection by scraping them into 5 ml phosphate buffered saline (PBS), collected and centrifuged at 700 ×g (3000 r.p.m.) for 5 min. Pellets derived from 50 plates (10 cm ø) were pooled and resuspended in 40 ml of ice-cold assay buffer (50 mM Tris-HCl supplemented with 5 mM MgCl2, pH 7.4). An UltraThurrax was used to homogenize the cell suspension. Membranes and the cytosolic fraction were separated by centrifugation at 100,000 ×g (31,000 r.p.m.) in a Beckman Optima LE-80K ultracentrifuge at 4 °C for 20 min. The pellet was resuspended in 20 ml of Tris buffer and the homogenization and centrifugation step was repeated. Assay buffer (10 ml) was used to resuspend the pellet and adenosine deaminase (ADA) was added (0.8 IU/ml) to break down endogenous adenosine. Membranes were stored in 250 μL aliquots at −80 °C. Membrane protein concentrations were measured using the BCA (bicinchoninic acid) method (Smith et al., 1985).

Competition and saturation binding assays using HEK293T cell membranes

For competition binding experiments with [3H]ZM241385 (46.6 Ci/mmol, ARC Inc, St. Louis, MO, USA), between 6 and 8 μg of membranes were used for the experiments to ensure that total binding was less than 10% of the total radioactivity added to prevent radioligand depletion. For [3H]NECA (16.3 Ci/mmol, Perkin Elmer, Groningen, The Netherlands) competition binding experiments between 18 and 25 μg of membranes were used for the experiments. Membrane aliquots were incubated in a total volume of 100 μL of assay buffer at 25 °C for 2 h. Radioligand displacement experiments were performed using five concentrations of competing ligand (NaCl, amiloride or HMA, all from Sigma Aldrich, Zwijndrecht, The Netherlands). [3H]ZM241385 and [3H]NECA were used at concentrations of ~ 4.0 nM and 15–20 nM, respectively. Nonspecific binding was determined in the presence of 100 μM CGS21680 (Ascent Scientific, Bristol, UK, for experiments with [3H]ZM241385) or 10 μM ZM241385 (Ascent Scientific, Bristol, UK, for experiments with [3H]NECA) and represented less than 15% of the total binding. For saturation experiments, total binding was determined at increasing concentrations of [3H]ZM241385 (0.10–45 nM), in the absence or presence of ZM241385 (10 nM), NaCl (30 or 100 mM), amiloride (30 μM), or HMA (3 μM). In addition, unlabeled NECA (Ascent Scientific, Bristol, UK) was spiked with 25% [3H]NECA resulting in final concentrations of 8.0 to 400 nM, in the absence or presence of ZM241385 (10 nM), NaCl (30 or 100 mM), amiloride (30 μM), or HMA (4 μM). Nonspecific binding was determined at three concentrations of radioligand and analyzed by linear regression. Incubations were terminated by rapid vacuum filtration to separate the bound and free radioligand through 96-well GF/B filter plates using a Filtermate-harvester (PerkinElmer Life Sciences). Filters were subsequently washed three times with ice-cold assay buffer. The filter-bound radioactivity was determined by scintillation spectrometry using the PE 1450 Microbeta Wallac Trilux scintillation counter (PerkinElmer Life Sciences).

Thermostability assays

The A2AAR-BRIL-ΔC receptor construct was purified from Spodoptera frugiperda (Sf9) insect cells in the apo form as described previously (Liu et al., 2012), except that KCl was used throughout purification instead of NaCl. N-[4-(7-diethylamino-4-methyl-3-coumarinyl)phenyl]maleimide (CPM) dye (Invitrogen) was dissolved in DMSO (Sigma) at 4 mg/mL and stored at −80 °C. Before use, the CPM stock solution was thawed and diluted 1:40 in dye dilution buffer (10 mM HEPES pH 7.50, 10% glycerol, 0.05% dodecyl maltoside (DDM) (Anatrace)). The thermal denaturation assay was performed with a total volume of 200 μL per sample in a quartz fluorimeter cuvette (Starna Cells, Inc., Atascadero, CA) and an apparent relative Tm was obtained (Alexandrov et al., 2008). Receptor (4 μg) was diluted in assay buffer (10 mM HEPES pH 7.5, 0.05% DDM, 0.01% cholesterol hemisuccinate (CHS) (Sigma)) with and without different concentrations and combinations of NaCl, amiloride, and ZM241385 to a final volume of 200 μL. 5 μL of the diluted dye was added to the protein-containing assay solution and incubated for 30 min at 4 °C. The mixed solutions were transferred into cuvettes and fluorescence data were collected by a Cary Eclipse spectrofluorometer (Varian, USA) with a temperature ramping rate of 2 °C/min. The excitation wavelength was 387 nm and the emission wavelength was 463 nm. All assays were performed over a temperature range starting from 20 °C and ramping to 90 °C.

Data analysis

The radioligand binding and thermostability data were processed with Prism 5 (GraphPad Software, San Diego, CA, USA). Statistical significance was assessed with a Student t-test.

Supplementary Material

01

HIGHLIGHTS.

  • A multidisciplinary study on the role of the sodium ion binding site in the A2AAR

  • Mechanism for the allosteric modulation by sodium ions and amilorides on A2AAR

  • Sodium ions selectively bind and stabilize the inactive conformation of the A2AAR

  • The binding of sodium ions and agonists is mutually exclusive

Acknowledgments

This work was supported by NIGMS PSI:Biology grant U54 GM094618 (R.C.S., V.K., V.C.). H.G.T. acknowledges the Spanish Ministry of Education for the mobility program (JC2011-0387) and financial support from the Spanish National Plan for R+D (SAF2011-30104). L.H.H., A.P.IJ. and A.M. thank the Dutch Research Council (NWO) for financial support (NWO-TOP #714.011.001 and NWO-Veni #11188).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Alexandrov AI, Mileni M, Chien EY, Hanson MA, Stevens RC. Microscale fluorescent thermal stability assay for membrane proteins. Structure. 2008;16:351–359. doi: 10.1016/j.str.2008.02.004. [DOI] [PubMed] [Google Scholar]
  2. Angel TE, Chance MR, Palczewski K. Conserved waters mediate structural and functional activation of family A (rhodopsin-like) G protein-coupled receptors. Proc Natl Acad Sci U S A. 2009;106:8555–8560. doi: 10.1073/pnas.0903545106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ballesteros JA, Weinstein H. Methods Neurosci. San Diego: Academic Press; 1995. Integrated methods for the construction of three dimensional models and computational probing of structure-function relations in G-protein coupled receptors; pp. 366–428. [Google Scholar]
  4. Barbhaiya H, McClain R, Ijzerman A, Rivkees SA. Site-directed mutagenesis of the human A1 adenosine receptor: influences of acidic and hydroxy residues in the first four transmembrane domains on ligand binding. Mol Pharmacol. 1996;50:1635–1642. [PubMed] [Google Scholar]
  5. Berendsen HJC, Postma JPM, van Gunsteren WF, Hermans J. Interaction models for Water in Relation to Protein Hydration. In: Pullman B, editor. Intermolecular Forces. Dordretch: D. Reidel Publishing Company; 1981. pp. 331–342. [Google Scholar]
  6. Berger O, Edholm O, Jähnig F. Molecular dynamics simulations of a fluid bilayer of dipalmitoylphosphatidylcholine at full hydration, constant pressure, and constant temperature. Biophys J. 1997;72:2002–2013. doi: 10.1016/S0006-3495(97)78845-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Chakrabarti N, Neale C, Payandeh J, Pai EF, Pomès R. An iris-like mechanism of pore dilation in the CorA magnesium transport system. Biophys J. 2010;98:784–792. doi: 10.1016/j.bpj.2009.11.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cherezov V. Lipidic cubic phase technologies for membrane protein structural studies. Curr Opin Struct Biol. 2011;21:559–566. doi: 10.1016/j.sbi.2011.06.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Christopher JA, Brown J, Dore AS, Errey JC, Koglin M, Marshall FH, Myszka DG, Rich RL, Tate CG, Tehan B, et al. Biophysical fragment screening of the beta1-adrenergic receptor: identification of high affinity arylpiperazine leads using structure-based drug design. J Med Chem. 2013;56:3446–3455. doi: 10.1021/jm400140q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Christopoulos A. Allosteric binding sites on cell-surface receptors: novel targets for drug discovery. Nat Rev Drug Discov. 2002;1:198–210. doi: 10.1038/nrd746. [DOI] [PubMed] [Google Scholar]
  11. Congreve M, Andrews SP, Dore AS, Hollenstein K, Hurrell E, Langmead CJ, Mason JS, Ng IW, Tehan B, Zhukov A, et al. Discovery of 1,2,4-triazine derivatives as adenosine A(2A) antagonists using structure based drug design. J Med Chem. 2012;55:1898–1903. doi: 10.1021/jm201376w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Congreve M, Langmead CJ, Mason JS, Marshall FH. Progress in structure based drug design for G protein-coupled receptors. J Med Chem. 2011;54:4283–4311. doi: 10.1021/jm200371q. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Conn PJ, Christopoulos A, Lindsley CW. Allosteric modulators of GPCRs: a novel approach for the treatment of CNS disorders. Nat Rev Drug Discov. 2009;8:41–54. doi: 10.1038/nrd2760. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dore AS, Robertson N, Errey JC, Ng I, Hollenstein K, Tehan B, Hurrell E, Bennett K, Congreve M, Magnani F, et al. Structure of the adenosine A(2A) receptor in complex with ZM241385 and the xanthines XAC and caffeine. Structure. 2011;19:1283–1293. doi: 10.1016/j.str.2011.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dror RO, Arlow DH, Borhani DW, Jensen MØ, Piana S, Shaw DE. Identification of two distinct inactive conformations of the beta2-adrenergic receptor reconciles structural and biochemical observations. Proc Natl Acad Sci U S A. 2009;106:4689–4694. doi: 10.1073/pnas.0811065106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Fraser JS, van den Bedem H, Samelson AJ, Lang PT, Holton JM, Echols N, Alber T. Accessing protein conformational ensembles using room-temperature X-ray crystallography. Proc Natl Acad Sci U S A. 2011;108:16247–16252. doi: 10.1073/pnas.1111325108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gao ZG, Ijzerman AP. Allosteric modulation of A(2A) adenosine receptors by amiloride analogues and sodium ions. Biochem Pharmacol. 2000;60:669–676. doi: 10.1016/s0006-2952(00)00360-9. [DOI] [PubMed] [Google Scholar]
  18. Gao ZG, Jacobson KA. Keynote review: allosterism in membrane receptors. Drug Discov Today. 2006;11:191–202. doi: 10.1016/S1359-6446(05)03689-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Gao ZG, Jiang Q, Jacobson KA, Ijzerman AP. Site-directed mutagenesis studies of human A(2A) adenosine receptors: involvement of glu(13) and his(278) in ligand binding and sodium modulation. Biochem Pharmacol. 2000;60:661–668. doi: 10.1016/s0006-2952(00)00357-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gao ZG, Kim SK, Gross AS, Chen A, Blaustein JB, Jacobson KA. Identification of essential residues involved in the allosteric modulation of the human A(3) adenosine receptor. Mol Pharmacol. 2003a;63:1021–1031. doi: 10.1124/mol.63.5.1021. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gao ZG, Melman N, Erdmann A, Kim SG, Muller CE, APIJ, Jacobson KA. Differential allosteric modulation by amiloride analogues of agonist and antagonist binding at A(1) and A(3) adenosine receptors. Biochem Pharmacol. 2003b;65:525–534. doi: 10.1016/s0006-2952(02)01556-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Garritsen A, Ijzerman AP, Tulp MT, Cragoe EJ, Jr, Soudijn W. Receptor binding profiles of amiloride analogues provide no evidence for a link between receptors and the Na+/H+ exchanger, but indicate a common structure on receptor proteins. J Recept Res. 1991;11:891–907. doi: 10.3109/10799899109064686. [DOI] [PubMed] [Google Scholar]
  23. Goblyos A, Ijzerman AP. Allosteric modulation of adenosine receptors. Biochim Biophys Acta. 2011;1808:1309–1318. doi: 10.1016/j.bbamem.2010.06.013. [DOI] [PubMed] [Google Scholar]
  24. Gutierrez-de-Teran H, Bello X, Rodriguez D. Characterization of the dynamic events of GPCRs by automated computational simulations. Biochem Soc Trans. 2013;41:205–212. doi: 10.1042/BST20120287. [DOI] [PubMed] [Google Scholar]
  25. Harding MM. Small revisions to predicted distances around metal sites in proteins. Acta Crystallogr D. 2006;62:678–682. doi: 10.1107/S0907444906014594. [DOI] [PubMed] [Google Scholar]
  26. Heitman LH, Ye K, Oosterom J, Ijzerman AP. Amiloride derivatives and a nonpeptidic antagonist bind at two distinct allosteric sites in the human gonadotropin-releasing hormone receptor. Mol Pharmacol. 2008;73:1808–1815. doi: 10.1124/mol.107.043521. [DOI] [PubMed] [Google Scholar]
  27. Hess B, Kutzner C, van der Spoel D, Lindahl E. GROMACS 4: Algorithms for Highly Efficient, Load-Balanced, and Scalable Molecular Simulation. J Chem Theory Comput. 2008;4:435–447. doi: 10.1021/ct700301q. [DOI] [PubMed] [Google Scholar]
  28. Hino T, Arakawa T, Iwanari H, Yurugi-Kobayashi T, Ikeda-Suno C, Nakada-Nakura Y, Kusano-Arai O, Weyand S, Shimamura T, Nomura N, et al. G-protein-coupled receptor inactivation by an allosteric inverse-agonist antibody. Nature. 2012;482:237–240. doi: 10.1038/nature10750. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Hoare SR, Coldwell MC, Armstrong D, Strange PG. Regulation of human D(1), d(2(long)), d(2(short)), D(3) and D(4) dopamine receptors by amiloride and amiloride analogues. Br J Pharmacol. 2000;130:1045–1059. doi: 10.1038/sj.bjp.0703370. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Horstman DA, Brandon S, Wilson AL, Guyer CA, Cragoe EJ, Jr, Limbird LE. An aspartate conserved among G-protein receptors confers allosteric regulation of alpha 2-adrenergic receptors by sodium. J Biol Chem. 1990;265:21590–21595. [PubMed] [Google Scholar]
  31. Howard MJ, Hughes RJ, Motulsky HJ, Mullen MD, Insel PA. Interactions of amiloride with alpha- and beta-adrenergic receptors: amiloride reveals an allosteric site on alpha 2-adrenergic receptors. Mol Pharmacol. 1987;32:53–58. [PubMed] [Google Scholar]
  32. Humphrey W, Dalke A, Schulten K. VMD - Visual Molecular Dynamics. J Mol Graphics. 1996;14:33–38. doi: 10.1016/0263-7855(96)00018-5. [DOI] [PubMed] [Google Scholar]
  33. Jaakola VP, Griffith MT, Hanson MA, Cherezov V, Chien EYT, Lane JR, Ijzerman AP, Stevens RC. The 2.6 angstrom crystal structure of a human A2A adenosine receptor bound to an antagonist. Science. 2008;322:1211–1217. doi: 10.1126/science.1164772. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Jacobson KA, Gao ZG, Goblyos A, Ijzerman AP. Allosteric modulation of purine and pyrimidine receptors. Adv Pharmacol. 2011;61:187–220. doi: 10.1016/B978-0-12-385526-8.00007-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Kaminski GA, Friesner RA, Tirado-Rives J, Jorgensen WL. Evaluation and reparametrization of the OPLS-AA force field for proteins via comparison with accurate quantum chemical calculations on peptides. J Phys Chem B. 2001;105:6474–6487. [Google Scholar]
  36. Katritch V, Cherezov V, Stevens RC. Structure-function of the G protein-coupled receptor superfamily. Annu Rev Pharmacol Toxicol. 2013;53:531–556. doi: 10.1146/annurev-pharmtox-032112-135923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lebon G, Warne T, Edwards PC, Bennett K, Langmead CJ, Leslie AG, Tate CG. Agonist-bound adenosine A2A receptor structures reveal common features of GPCR activation. Nature. 2011;474:521–525. doi: 10.1038/nature10136. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Liu W, Chun E, Thompson A, Chubukov P, Xu F, Katritch V, Han GW, Heitman L, Ijzerman A, Cherezov V, et al. Structural Basis for Allosteric Regulation of GPCRs by Sodium Ions. Science. 2012;337:232–236. doi: 10.1126/science.1219218. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Manglik A, Kruse AC, Kobilka TS, Thian FS, Mathiesen JM, Sunahara RK, Pardo L, Weis WI, Kobilka BK, Granier S. Crystal structure of the micro-opioid receptor bound to a morphinan antagonist. Nature. 2012;485:321–326. doi: 10.1038/nature10954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Mirzadegan T, Benko G, Filipek S, Palczewski K. Sequence analyses of G-protein-coupled receptors: Similarities to rhodopsin. Biochemistry. 2003;42:2759–2767. doi: 10.1021/bi027224+. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Neubig RR, Spedding M, Kenakin T, Christopoulos A. International Union of Pharmacology Committee on Receptor Nomenclature and Drug Classification. XXXVIII Update on terms and symbols in quantitative pharmacology. Pharmacol Rev. 2003;55:597–606. doi: 10.1124/pr.55.4.4. [DOI] [PubMed] [Google Scholar]
  42. Neve KA, Cumbay MG, Thompson KR, Yang R, Buck DC, Watts VJ, DuRand CJ, Teeter MM. Modeling and mutational analysis of a putative sodium-binding pocket on the dopamine D2 receptor. Mol Pharmacol. 2001;60:373–381. doi: 10.1124/mol.60.2.373. [DOI] [PubMed] [Google Scholar]
  43. Nie J, Lewis DL. Structural domains of the CB1 cannabinoid receptor that contribute to constitutive activity and G-protein sequestration. J Neurosci. 2001;21:8758–8764. doi: 10.1523/JNEUROSCI.21-22-08758.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Nose S, Klein ML. Constant pressure molecular-dynamics for molecular-systems. Mol Phys. 1983;50:1055–1076. [Google Scholar]
  45. Pardo L, Deupi X, Dölker N, López-Rodríguez ML, Campillo M. The role of internal water molecules in the structure and function of the rhodopsin family of G protein-coupled receptors. Chembiochem. 2007;8:19–24. doi: 10.1002/cbic.200600429. [DOI] [PubMed] [Google Scholar]
  46. Park JH, Scheerer P, Hofmann KP, Choe HW, Ernst OP. Crystal structure of the ligand-free G-protein-coupled receptor opsin. Nature. 2008;454:183–187. doi: 10.1038/nature07063. [DOI] [PubMed] [Google Scholar]
  47. Pauwels PJ. Competitive and silent antagonism of recombinant 5-HT1B receptors by amiloride. Gen Pharmacol. 1997;29:749–751. doi: 10.1016/s0306-3623(97)00008-6. [DOI] [PubMed] [Google Scholar]
  48. Pert CB, Pasternak G, Snyder SH. Opiate agonists and antagonists discriminated by receptor binding in brain. Science. 1973;182:1359–1361. doi: 10.1126/science.182.4119.1359. [DOI] [PubMed] [Google Scholar]
  49. Proulx CD, Holleran BJ, Boucard AA, Escher E, Guillemette G, Leduc R. Mutational analysis of the conserved Asp2.50 and ERY motif reveals signaling bias of the urotensin II receptor. Mol Pharmacol. 2008;74:552–561. doi: 10.1124/mol.108.045054. [DOI] [PubMed] [Google Scholar]
  50. Rasmussen SG, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, Thian FS, Chae PS, Pardon E, Calinski D, et al. Crystal structure of the beta2 adrenergic receptor-Gs protein complex. Nature. 2011;477:549–555. doi: 10.1038/nature10361. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Rodriguez D, Pineiro A, Gutierrez-de-Teran H. Molecular dynamics simulations reveal insights into key structural elements of adenosine receptors. Biochemistry. 2011;50:4194–4208. doi: 10.1021/bi200100t. [DOI] [PubMed] [Google Scholar]
  52. Rosenbaum DM, Cherezov V, Hanson MA, Rasmussen SGF, Thian FS, Kobilka TS, Choi HJ, Yao XJ, Weis WI, Stevens RC, et al. GPCR engineering yields high-resolution structural insights into beta2-adrenergic receptor function. Science. 2007;318:1266–1273. doi: 10.1126/science.1150609. [DOI] [PubMed] [Google Scholar]
  53. Sambrook J, Fritsch EF. Molecular Cloning: A Laboratory Manual. 2. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory; 1989. [Google Scholar]
  54. Schrödinger, L.L.C. Macromodel, version 9.7. New York, NY: 2009. [Google Scholar]
  55. Schwartz TW, Frimurer TM, Holst B, Rosenkilde MM, Elling CE. Molecular mechanism of 7TM receptor activation--a global toggle switch model. Annu Rev Pharmacol Toxicol. 2006;46:481–519. doi: 10.1146/annurev.pharmtox.46.120604.141218. [DOI] [PubMed] [Google Scholar]
  56. Selent J, Sanz F, Pastor M, De Fabritiis G. Induced effects of sodium ions on dopaminergic G-protein coupled receptors. PLoS Comput Biol. 2010;6:e1000884. doi: 10.1371/journal.pcbi.1000884. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Shi L, Javitch JA. The binding site of aminergic G protein-coupled receptors: the transmembrane segments and second extracellular loop. Annu Rev Pharmacol Toxicol. 2002;42:437–467. doi: 10.1146/annurev.pharmtox.42.091101.144224. [DOI] [PubMed] [Google Scholar]
  58. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC. Measurement of protein using bicinchoninic acid. Anal Biochem. 1985;150:76–85. doi: 10.1016/0003-2697(85)90442-7. [DOI] [PubMed] [Google Scholar]
  59. Snyder SH, Pasternak GW. Historical review: Opioid receptors. Trends Pharmacol Sci. 2003;24:198–205. doi: 10.1016/S0165-6147(03)00066-X. [DOI] [PubMed] [Google Scholar]
  60. Tate CG, Schertler GF. Engineering G protein-coupled receptors to facilitate their structure determination. Curr Opin Struct Biol. 2009;19:386–395. doi: 10.1016/j.sbi.2009.07.004. [DOI] [PubMed] [Google Scholar]
  61. Tsai BS, Lefkowitz RJ. Agonist-specific effects of monovalent and divalent cations on adenylate cyclase-coupled alpha adrenergic receptors in rabbit platelets. Mol Pharmacol. 1978;14:540–548. [PubMed] [Google Scholar]
  62. Ulmschneider MB, Bagneris C, McCusker EC, Decaen PG, Delling M, Clapham DE, Ulmschneider JP, Wallace BA. Molecular dynamics of ion transport through the open conformation of a bacterial voltage-gated sodium channel. Proc Natl Acad Sci U S A. 2013;110:6364–6369. doi: 10.1073/pnas.1214667110. [DOI] [PMC free article] [PubMed] [Google Scholar]
  63. Warne T, Edwards PC, Leslie AG, Tate CG. Crystal structures of a stabilized beta1-adrenoceptor bound to the biased agonists bucindolol and carvedilol. Structure. 2012;20:841–849. doi: 10.1016/j.str.2012.03.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Xu F, Wu H, Katritch V, Han GW, Jacobson KA, Gao ZG, Cherezov V, Stevens RC. Structure of an Agonist-Bound Human A2A Adenosine Receptor. Science. 2011;332:322–327. doi: 10.1126/science.1202793. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS531997-supplement-01.pdf (3.4MB, pdf)

RESOURCES