Allosteric Modulation of 7 Transmembrane Spanning Receptors: Theory, Practice and Opportunities for CNS Drug Discovery¹

1. INTRODUCTION

2. MODE OF ACTION OF ALLOSTERIC MODUALTORS

7TMRs are highly flexible proteins capable of assuming multiple conformations, of which some are active, some are inactive (pharmacologically silent), and some are partially active. In fact, 7TMRs should be thought of as ensembles of tertiary conformations, randomly sampled by the receptor, and very subtle changes (as small as 1Å) can engender profound effects on receptor activity.^{1,4–8,10,11} Thus, when an allosteric modulator binds to a site topographically distinct from the orthosteric site, a change in receptor conformation occurs that can modify receptor activity in either a positive, negative or neutral direction. As mentioned before, the allosterically bound receptor is a ‘new’ receptor type, with novel behavior and activity potential. Oligomerization of GPCRs, as either hetero- or homodimers adds additional opportunities for modulation and probe dependence.¹⁸⁹ Operationally, 7TMR allosteric modulators exhibit affinity modulation, efficacy modulation or varying degrees of both modes of modulation (Figure 4A). With affinity modulation, the conformational change in the 7TMR upon allosteric ligand binding can affect either the association or dissociation rate (or both) of the orthosteric ligand. ^{1,4–8,10,11} For example, a PAM that displays affinity modulation will result in a more potent orthosteric ligand (agonist). For efficacy modulation, the conformational change in the 7TMR upon allosteric ligand binding leads to a change in signaling capacity (also termed intrinsic efficacy) and thereby either facilitates or inhibits receptor coupling to downstream effectors. There are two models proposed to account for the interactions between the allosteric and orthosteric ligand (Figures 4B and C).^1,4,7 In one model, termed the ‘allosteric hot wire’, the allosteric modulation site is directly linked to the orthosteric site through specific pathways.^1,190,191 In a more recent model, termed ‘global allosteric modulation’, allosteric communication to the orthosteric site relies on long range interaction through order/disorder transitions from multiple receptor conformations (i.e., population dynamics).^192,193 A number of mass-action schemes, based on variants of the ternary complex model, have been presented to describe the molecular effects of allosteric ligands on orthosteric pharmacology in terms of one or more “cooperativity factors”, which indicate the magnitude and direction of an allosteric modulator-mediated stabilization of different 7TMR states.⁴ From these models, it can be appreciated that the functional potency of an allosteric modulator will depend not only on its affinity for the allosteric site, but also on the degree of cooperativity with the orthosteric ligand. Thus, a PAM (e.g., 11) may bind to the receptor with weak affinity but possess potent functional activity due to a high cooperativity factor.^4,10,90 In contrast, another PAM (e.g, benzodiazepines) may bind with very high affinity but low positive cooperativity, thus also displaying potent functional activity. The low affinity observed with some PAMs can preclude them from serving as radioligands and PET tracers, and the affinity/functional activity balance must be carefully assessed when considering receptor occupancy, for example, as a potential biomarker strategy.

Unfortunately, most mass-action based molecular models of allosteric modulation contain too many parameters to be fitted to real experimental data, and thus cannot be used to rationalize structure-activity studies in a manner that can inform drug candidate selection matrices. A useful means for placing these issues on a more practical quantitative level is through the use of an “operational” model of allosterism and agonism, which has been developed to describe allosteric effects in terms of a minimum number of experimentally accessible parameters.¹⁹⁴ This model is illustrated conceptually in (Figure 5), and the equation describing the signaling of an orthosteric agonist in the presence of an allosteric modulator according to the model is as follows:

where E is the effect, [A] and [B] are the concentrations and K_A, and K_B are the equilibrium dissociation constants of the orthosteric and allosteric ligand, respectively, α is the cooperativity factor describing the allosteric effect of each ligand on the other’s binding affinity, β is a scaling factor (from zero to infinity) that quantifies the magnitude by which the allosteric modulator modifies the efficacy of the orthosteric agonist at a given signal pathway, and the parameters τ_A and τ_B relate to the ability of the orthosteric and allosteric ligands, respectively, to promote receptor activation (direct agonism); these latter parameters incorporate the intrinsic efficacy of each ligand, the total density of receptors, and the efficiency of stimulus-response coupling. The parameters E_m and n denote the maximal possible system response and the slope factor of the transducer function that links occupancy to response, respectively.^4,10,194

Importantly, the operational model can be fitted to experimentally derived data to provide estimates of some, or all, of its parameters.^{47,52,90,195–197} At a minimum, there are three key parameters that can be routinely derived from application of this model to most functional screening data, as long as full concentration-response and curve-shift relationships are determined. These three parameters are: the allosteric modulator K_B, which provides information on the interaction of the allosteric ligand with the allosteric binding pocket on the free receptor, the composite cooperativity parameter, αβ, which provides information on the overall allosteric effect on the orthosteric agonist in the chosen functional assay, and the modulator efficacy parameter, τ_B, which provides information on the ability of the allosteric ligand to promote agonism in its own right in the absence of orthosteric ligand. Table 4 illustrates an example of such allosteric modulator SAR determined through analysis of the functional effects of a series of 2-amino-3-benzoylthiophenes (2A3BT) on A₁ adenosine receptor-mediated ERK1/2 phosphorylation.⁴⁶ From this analysis, it can be seen that, for instance, the increase in trifluoromethylphenyl substitutions to the R₂ group of the 2A3BT scaffold can increase positive cooperativity while having a detrimental effect on modulator affinity (compare 16b to 16d, and 16g to 16h), whereas conformational constraint in the R₂/R₃ regions tends to convert positive modulators into highly negative modulators (e.g., 16n, 16r). It should also be noted that the reference compound, 16a, progressed into Phase IIB clinical trials for the treatment of neuropathic pain (King Pharmaceuticals) prior to failing due to lack of efficacy; it is possible that this failure may be attributed to the rather low degree of positive cooperativity, as revealed by the application of the operational model to the in vitro data. Although speculative, this finding highlights some of the advantages of operational modeling when applied to allosteric SAR, namely, the ability to link the chemistry to the key, measured, biological parameters, and the ability to facilitate hypothesis generation to understand biological mechanisms and their relevance to the desired therapeutic profile. For instance, one can ask questions such as: how much cooperativity (αβ) or allosteric agonism (τ_B) is required to achieve in vivo efficacy? How do structural modifications affect compound affinity (K_B) versus cooperativity (αβ)? The latter is important as these properties are not correlated and thus different structural manipulations can change them in different directions. Probe dependence will manifest as different αβ values depending on the orthosteric agonist used and/or the signal pathway being assessed as a readout of receptor activation. In terms of drug discovery programs, these insights can be used to more rationally inform the design of candidate selection matrices for drugs acting allosterically. ^{46,47,52,90,195–197}

Table 4.

Allosteric Operational Model Parameters Describing the Functional Effect of Various 2-Amino-3-Benzoylthiophenes on ERK1/2 Phosphorylation mediated by the Orthosteric Agonist, R-PIA, at Adenosine A₁ Receptors.

Compound	Structure	pK3	logαβ(αβ)	logτB(τB)
16a(T62)a		5.49 ± 0.09	0.58 ± 0.08 (3.8)	−0.32 ± 0.05(0.5)
16b 9aa		6.37 ± 0.16	0.38 ±0.07 (2.4)	0.33 ± 0.04 (0.5)
16c(VCP333)a		5.23 ± 0.25	0.64 ± 0.12(4.4)	−1.34 ± 0.75 (0.05)
16d(9o)a		5.22 ± 0.07	1.06 ± 0.08 (11)	0.62 ± 0.24 (4.1)
16e(13c)a		5.10 ± 0.18	1.04 ± 0.20(11)	0.13 ± 0.21(2.7)
16f(13d)a		5.88 ± 0.16	0.67 ± 0.10(4.7)	−0.30 ± 0.08 (0.5)
16g(13a)a		5.74 ± 0.12	0.55 ± 0.06(3.5)	−0.48 ± 0.08 (0.3)
16h(13o)a		5.46 ± 0.32	1.31 ± 0.40(20)	0.50 ± 0.20(3.2)
16i(12j)a		5.08 ± 0.17	0.46 ± 0.07(2.9)	−0.41 ± 0.09(0.4)
16j(25b)a		6.01 ± 0.16	0.44 ± 0.07(2.7)	−0.48 ± 0.08 (0.3)
16k(25f)a		5.11 ± 0.41	0.58 ± 0.07(3.8)	−1000(ĩ0)
16l(25d)a		5.12 ± 0.32	0.82 ± 0.16 (6.6)	−1000(ĩ0)
16m(12o)a		5.10 ± 0.14	0.98 ± 0.14 (10)	0.31 ± 0.08 (2.0)
16n(17d)a		6.67 ± 0.13	−1000(~0)	−1000(~0)
16o(22d)a		5.82 ± 0.10	−1000(~0)	−1000(~0)
16p(22a)a		6.36 ± 0.13	−1000(~0)	−1000(~0)
16q(2o)a		6.07 ± 0.05	−1000(~0)	−1000(~0)
16r(13b)a		8.30 ± 0.24	−1000(~0)	−1000(~0)

Open in a new tab

^a

Compound nomendature refers to compound identifier in the manuscript in which it originally appeared

3. IN VITRO PHARMACOLOGY OF ALLOSTERIC MODULATORS

The development of high-throughput functional (kinetic) assays, have enabled scientists to perform screens of large compound collections and identify small molecules capable of modulating the activity of a receptor through novel, allosteric mechanisms.^{4,160,198–201} While there are multiple approaches and technologies to accomplish this for 7TMRs, one of the most common approaches measures receptor-induced mobilization of intracellular calcium using an imaging-based plate reader that makes simultaneous measurements of calcium levels in each well of a multi-well-well microplate containing cells transfected with the receptor of interest and loaded with calcium-sensitive fluorescent dye. In the early stages of drug discovery for allosteric modulators of 7TMRs, HTS campaigns targeted the identification of either PAMs or NAMs, running single-point screens with either an EC₂₀ or EC₈₀ concentration of the orthosteric agonist, respectively. More recently, ‘triple add’ protocols have supplanted ‘single add’ screens to allow). for, in a single assay, the identification of PAMs, NAMs, agonists and antagonists (Figure 6A).^{4,160,198–201} Due to the emerging concept of ‘molecular switches’ (see below),²⁰² this screening paradigm is also optimal as for use as a program’s primary assay, to identify subtle structural changes leading to opposing modes of pharmacology.

As shown in Figure 6B, a prototypical PAM has no effect on the transfected cells in the absence of the orthosteric agonist, but in the presence of a sub-threshold concentration of the orthosteric agonist, increasing PAM concentrations provide a classical concentration-response-curve (CRC) from which an EC₅₀ for potentiation can be quantified as an empirical measure of modulator potency under a defined set of assay conditions.^4,5 This also affords a %Max value, the maximum increase in activity of the orthosteric agonist above the EC₂₀, i.e., from an EC₂₀ to an EC₉₀. In this type of experimental paradigm, the observed EC₅₀ reflects both the affinity of the modulator for the allosteric site, as well as the degree of cooperativity that it exerts on the orthosteric ligand. In contrast, the %Max value only reflects the cooperativity of the interaction. Each of these parameters must be optimized, as it is possible to have allosteric modulators with high potency and low %Max (reflecting high affinity and low cooperativity), weak potency and high %Max (low affinity and high cooperativity) and various combinations thereof. If the assay is run in an alternative mode, whereby the complete orthosteric agonist CRC is determined in the absence or presence of a fixed concentration of allosteric modulator, then another property that can be evaluated and optimized is the degree of the agonist curve ‘fold-shift’, either to the left (potentiation) or right/down (antagonism) of an agonist CRC (Figure 6C).^4,5 This is typically reported as a fold-shift at a single concentration, such as a 30-fold shift at 10 μM. Optimally, if multiple modulator concentrations are utilized in this latter type of experiment, then the data can be fitted to the operational model (above) and direct estimates of affinity and cooperativity can be obtained. Irrespective, each of the properties described in the preceding section (% Max, potency, fold-shift or operational model affinity, cooperativity and efficacy) must be simultaneously optimized for optimal in vivo efficacy.^4,5 A common question posed in small molecule allosteric modulators programs is: which of these parameters is most important for in vivo efficacy? Unfortunately, the answer is not clear and may vary by 7TMR, allosteric binding site, allosteric ligand chemotype and therapeutic indication. For now, a general empirical guideline followed by most researchers in the field is to optimize for potency (EC₅₀), and % Max and aim for at least a 5-fold shift of the agonist CRC; however, it should be notes that estimates of fold-shift for some Family A and C 7TMR allosteric ligands may approach >70-fold, depending on the assay.^4,5 It should be noted that other properties of a CNS compound, for example, the ability of a compound to penetrate the blood brain barrier, the clearance of a compound from the brain and systemic compartments, the amount of compound that is non-protein bound and able to interact with the target, and the in vivo situation being assessed (for example, how long does a receptor need to be occupied with drug to maintain efficacy), certainly must be balanced with in vitro pharmacology profiles when assessing in vivo efficacy and potency of allosteric modulators.

4. STRUCTURE-ACTIVITY-RELATIONSHIPS (SAR) within mGluRs and mAChRs

Another common observation from numerous structure activity relationship (SAR) studies performed to date on 7TMR allosteric modulators is the finding of ‘flat’ or ‘shallow’ SAR for different classes, often making optimization of micromolar potency ligands either difficult or impossible.^1,4–12 The literature in this arena is filled with tales of heroic optimization campaigns wherein 100s or 1000s of compounds were synthesized and evaluated, affording only 5–10% active molecules. However, there are also cases wherein SAR is robust and tractable, though rare. Thus, for the chemical optimization of allosteric ligands,^4,197,202 focused, iterative library synthesis provides a distinct advantage over traditional singleton approaches; however, multiple dimensions of a scaffold must be surveyed to identify regions tolerant of modification. As SAR is often extremely ‘shallow’, the concept of walking fluorine atoms around an allosteric ligand, i.e., ‘the fluorine walk’ has achieved some success in identifying positions tolerant of change. For example, 6 displayed ‘shallow’ SAR, and multiple attempts at optimization, adding groups larger than fluorine (methyl, OR, Cl, Br, alkyl, etc…) to multiple positions (ie, R₁ and/or R₂) on the scaffold of 6 led to primarily inactive compounds 17.^33,203 The strategic installation of fluorine atoms to both the core R₁ and the benzyl side chain R₂ of 6 led to compounds with significant improvements in M₁ PAM activity (Figure 7), but only in the 5- (18), 8- (19) or 5,8-positions (20) – introduction of fluorine atoms in any other positions led to inactive compounds. Once identified, these new fluorinated cores opened up new avenues for diverse functionalization on the benzyl core that were not tolerated on the parent core, leading to potent M₁ PAMs such as 21 (M₁ EC₅₀ = 41 nM).²⁰³ A very similar ‘fluorine walk’ strategy has proven successful in developing sub-micromolar M₁ PAMs in other chemotypes.^83,84 It is important to note that in these cases, and many other published studies, only fluorine substitutions were tolerated, making this approach a potential first tier strategy in allosteric ligand optimization.

Adding to the ‘flat’ or ‘shallow’ allosteric structure activity relationships (SAR), the emerging concept of ‘molecular switches’, i.e., subtle structural changes that can change the mode of pharmacology (from PAM to NAM and/or SAM) and/or subtype selectivity within a family of receptors, threatens to diminish the application of rational drug design approaches to allosteric modulators.²⁰² Moreover, unexpected occurrences of ‘molecular switches’ require alterations to routine screening paradigms for optimization efforts. Here, the ‘triple add’ approach is ideal, ^{4,160,198–201} identifying allosteric agonists, PAMs, and NAMs in a single screen. Although examples of ‘molecular switches’ developed for mGluRs and mAChRs are presented herein, this subtle effect has been reported for allosteric kinase and phospholipase ligands as well.²⁰²

In the field of mGluRs, the first allosteric modulators disclosed were (−)-ethyl-(7E)-7-hydroxyimino-1,7α-dihydrocyclopropa[b]chromene-1α-carboxylate (CPCCOEt, mGlu₁ selective) and 10 (mGlu₅ selective), both NAMs, followed by the earliest mGlu₅ PAMs, 3,3′-difluorobenzaldazine (DFB), 14, and 3-cyano-N-(1,3-diphenyl-1H-pyrazol-5-yl)benzamide (CDPPB).^4,44, While various molecular switches on the difluorobenzaldazine scaffold had the ability to engender the full range of PAM, NAM and SAM pharmacology (structures not shown),¹⁶¹ an even more surprising example of pharmacological mode switching can be seen in Figure 8. A functional HTS with the mGlu₅ receptor identified the “partial antagonist” 22, which maximally inhibited 71% of the glutamate response with an mGlu₅ IC₅₀ = 486 nM.^{202,204–206} This simple and very low molecular weight hit was rapidly explored using a parallel iterative library approach to reveal numerous molecular switches. Introduction of a 3′-methyl group transformed the partial antagonist into a very potent full NAM 23 (mGlu₅ IC₅₀ = 7.5 nM). This methyl group exquisitely illustrates the idea of a “molecular switch” as it is moved just one position over to provide compound 24, which is now a weak PAM (mGlu₅ EC₅₀ = 3.3 μM) with the ability to potentiate a sub-maximal dose of glutamate up to a full glutamate response (99% Glu Max). Further optimization within this series of PAMs provided compound 25; a very potent mGlu₅ PAM which possessed robust in vivo efficacy in a rodent amphetamine-induced hyperlocomotion assay. ^{202,204–206}

Also within the mGluR field, molecular switches have been reported that profoundly affect the receptor subtype selectivity of a given allosteric ligand, and also the mode of pharmacology. Figure 9 shows 26 ((−)-PHCCC), which possess both mGlu₄ PAM and mGlu₁ NAM activity.¹⁵⁵ Introduction of a weakly basic nitrogen to arrive at 27 (VU0359516) effectively abolished all mGlu₁ NAM activity to provide a pure mGlu₄ PAM (EC₅₀ > 30 μM versus mGlu_1–3,5–8) with good potency and efficacy (mGlu₄ EC₅₀ = 380 nM, Glu Max = 121%, 20-fold shift).¹⁵⁵ Subsequent to this work, an mGlu₂ FRET-based binding assay identified the fluorinated analog 28 which displayed an mGlu₂ K_i = 6.6 μM, while at the same time, lacked all functional activity at mGlu₂ and mGlu₃.²⁰⁷ Importantly, 28 could also block (silence) the activities of related mGlu₂ and mGlu₃ allosteric modulators 29–31: the definition of an mGlu_2/3 SAM. As alluded to above, the position of the fluorine in 28 represented a productive location for the introduction of alternate molecular switches (Figure 9, X = Cl, Me, OMe) which could transform an mGlu₁ NAM/mGlu₄ PAM 26 or an mGlu_2/3 SAM 28 into a series of mGlu₂ NAM/mGlu₃ PAMs 29–31.²⁰⁷

These types of molecular switches that govern receptor subtype selectivity have similarly been reported in the mAChR literature. Figure 10 shows 32, which was an attractive, low-molecular weight PAM hit from a functional HTS assay with the M₁ mAChR.^{82,84,94–96} Although originally identified as only an M₁ PAM, subsequent characterization revealed 32 to be a non-selective G_q-coupled mAChR PAM displaying activity at the M_1,3,5 receptor subtypes, while being devoid of activity at the G_i-coupled M₂ and M₄ mAChRs.⁸² This interesting selectivity for the different G protein-coupled signaling pathways spurred medicinal chemistry efforts to determine if subtle molecular switches could be discovered that might provide selectivity for a single mAChR subtype. The first break through came with the discovery that placing a 5-trifluoromethoxy substituent on the isatin core lead to a profound preference for M₅ PAM activity. This molecular switch employed during an exploration of substituents about the benzyl group ultimately provided 33, (X = Me, ML129) and 34 (R = Ph, ML172) the first and most highly selective M₅ PAMs, respectively, reported to date.^94–96 Further structural modifications around the non-selective lead 32 revealed that replacing the bromine with various aryl groups identified the N-methyl pyrazole as a powerful molecular switch for establishing high levels of M₁ selectivity. Fine tuning this M₁ PAM activity through the strategic introduction of a fluorine atom, ie, the ‘fluorine walk’ provided the highly selective M₁ PAM 35 (ML137).⁸⁴ Ongoing studies exploring the utility of these and other allosteric ligands promise to reveal numerous additional molecular switches and as these compounds progress into more animal models and detailed DMPK evaluations it will only be a matter of time before metabolism-induced molecular switches are reported.²⁰¹ It should be noted that such molecular switches may not be observed within all allosteric ligand scaffolds, or at all allosteric binding sites; in fact, some allosteric ligands and binding sites are considered ‘molecular locks’, with robust tractable SAR.

5. Opportunities for CNS Disorders and Therapeutics

Despite these challenges, allosteric ligands have enabled researchers to develop small molecule tools (Tables 1–3) ^{4–8, 10,11, 42–180} with exquisite selectivity for a particular target, not possible with orthosteric ligands, and achieve proof of concept in preclinical models of various CNS disorders. For example, mGlu₅ NAMs have provided preclinical target validation in models of anxiety, fragile X syndrome, chronic pain, migraine and GERD.¹⁷⁰ In fact, mGlu₅ NAMs are one of the most advanced with multiple compounds in clinical development, and displaying efficacy in Phase II (Table 5).^208–230 Diverse chemotypes of both mGlu₅¹⁷¹ and mGlu₂¹⁴⁸ PAMs have shown robust activity in preclinical models of schizophrenia and cognition, while selective mGlu₄ PAMs¹⁵⁷ have validated the target for both pain and Parkinson’s disease.⁴

Table 5.

mGlu₅ NAM Clinical Compounds

Drug Name	Structure	Organization	Clinical Trials
fenobam™ (NPL-2009)208–210		Neuropharm	anxiety, fragile X syndrome
raseglurant™ (ADX10059)211–214		Addex	anxiety, GERD, migraine
dipraglurant™ (ADX48621)211–216		Addex	PD-LID
AFQ056217–222		Novartis	fragile X syndrome, PD-LID, GERD, nicotine addiction
AZD2066223,224	Structure Unknown	AstraZeneca	GERD, chronic neuropathic pain, major depressive disorder, diabetic neuropathy
AZD2516223,225	Structure Unknown	AstraZeneca	GERD, chronic neuropathic pain, major depressive disorder
RG7090226	Structure Unknown	Roche	treatment resistant depression
RO4917523227	Structure Unknown	Roche	fragile X syndrome, treatment resistant depression
STX107228,229	Structure Unknown	Seaside Therapeutics	fragile X syndrome

Open in a new tab

The biogenic amine receptors, the prototypes for promiscuous pharmacology, have benefited greatly from allosteric approaches.^4,42–180 For example, the mAChRs have been targets of interest for multiple CNS disorders since the 1950s, but the highly conserved orthosteric (acethycholine) binding site led to unselective small molecules. Due to the therapeutic relevance of the mAChRs, multiple companies advanced orthosteric, pan-mAChR agonists into the clinic and noted efficacy in Phase II and Phase III trials in Alzheimer’s disease and schizophrenia patients; however, the adverse events from activation of peripheral M₂ and M₃ prevented further development. Recently, allosteric ligands, both allosteric/bitopic agonists and PAMs, have been developed for M₁, M₄ and M₅.^4,43 These tools are now dissecting the individual contributions of these three mACh receptor subtypes in the clinical efficacy of pan-mAChR agonists. Importantly, this is but one example against the backdrop of the allosteric ligands in Tables1–3, ^4,42–180 that are validating discrete targets within large families of receptors, and shedding light on the therapeutic potential of GPCRs, long obscured by the lack of selective tool compounds.

Finally, it is very important to point out that allosteric ligands have been advanced into marketed therapeutics, suggesting the approach of targeting allosteric mechanisms is sound and tractable (Figure 11). 36 is a PAM of the calcium sensing receptor (a Family C GPCR) and is used to treat hyperparathyroidism.¹¹⁵ 37 is a NAM of chemokine receptor 5 (a Family A GPCR) that inhibits HIV entry into cells and is used to treat HIV infections.⁷² Thus, both PAMs and NAMs have advanced to the market, with many other allosteric ligands in clinical trials and late preclinical development.⁴

6. Conclusion

The broad uptake of the concept of allosteric modulation has led to a renaissance in pharmacological approaches towards 7TMR pharmacology, promising new and exciting avenues to the pursuit of receptor-subtype and pathway-selective small molecules. Although the promises of this approach are apparent, there remain significant challenges with regards to optimal means for detecting, validating and quantifying allosteric ligand effects in a manner that routinely informs SAR and candidate selection matrices, and is also ultimately predictive of therapeutic efficacy. To meet these challenges, further convergence is required between the disciplines of pharmacology, which can shed insight into the nature and mechanisms underlying phenomena such as probe dependence and pathway-biased modulation, medicinal/synthetic chemistry, which can overcome the issue of ‘flat’ allosteric SAR and explore the full potential of molecular switches in creating new allosteric behaviors, and structural biology, which can identity the underlying structural basis of allosteric ligand binding and the associated receptor conformations that mediate allosteric effects. Fortunately, recent work in the field is illustrating how truly translational approaches towards 7TMR pharmacology promise to overcome many of these challenges and realize the therapeutic potential of allosteric modulators as novel medicines.

Biographies