Conformational Recognition of Effector Proteins as a Common Determinant of Signaling Bias Across Receptor Classes

Within pharmacology we recognize several classes of receptor that are defined by their structure, function and signaling mechanisms: G protein-coupled receptors (GPCRs), nuclear hormone receptors (NHRs), receptor tyrosine kinases (RTKs) and ligand-gated ion channels (LGICs) (for review see Arey Biased Signaling in Physiology, Pharmacology and Therapeutics). Each receptor class has its own signaling effectors that act as a means of signaling activation of the receptor. In the case of GPCRs, they are characterized by their seven transmembrane spanning domains, are located on the cell surface and signal through interaction with the classical heterotrimeric G-proteins and scaffolding proteins such as β-arrestins. Likewise, RTKs are located at the cell surface but require dimerization for activation and possess a single transmembrane spanning domain and an intracellular carboxy-terminus that contains a tyrosine kinase domain. RTKs signal through recruitment of scaffolding proteins and stimulation of phosphorylation of tyrosines on enzyme effectors such as Stat, Src and MAPK proteins. The NHRs are cytoplasmic and nuclear in localization, act as dimers but interact with associated cellular proteins known as co-activators or co-repressors that form a complex that recognize key DNA sequences within gene promoters to directly stimulate or inhibit transcription. LGICs are also membrane bound and are a complex of multiple subunits that have a single membrane spanning domain but when together form a channel or pore within the cell that has specific permeability to certain ions and primarily act to regulate ion flow across the cell membrane.

Historically speaking, receptors have been viewed as singular sensors perceiving the arrival of extracellular stimuli to the cell and translating this information into biochemical signals that can affect numerous biochemical processes within cells. The classical understanding of how extracellular ligands interact with their receptors is best described by the “lock and key” analogy which holds that receptors possess a binding cavity for their extracellular ligands that is specifically shaped to accommodate its ligand(s). Once bound to the ligand, the receptor is “unlocked” to associate with key signaling effectors that act as transducers of the signal of ligand binding to the receptor. For many years the mechanism of how the act of ligand binding unlocked the receptor was largely unknown. However, with the advent of improved methodologies for developing crystal structures of receptors, we now have a much better understanding of the biochemical and structural changes that occur following ligand binding. Similarly, for many years following the elucidation of receptor signal transduction it was thought that each receptor could activate only a single effector. However, as researchers studied receptors more completely and especially as new ligands were synthesized in the course of the drug discovery process, this view was found to be a bit short-sighted. It was found that receptors could activate multiple signaling pathways to varying degrees even in response to their physiological ligands. Thus, we now understand that endogenous agonists stimulate a number of signaling effectors to varying degrees to generate an intracellular signaling pattern, or fingerprint, which leads to the full breadth of impact on the phenotype of the target cell.

Simultaneously, we have realized and modeled the biochemical interactions that take place between ligand and receptor and we now recognize that these key interactions are truly ligand specific such that each ligand is capable of stabilizing a unique conformation of the receptor through its individual chemical associations within the binding pocket found on its receptor. Furthermore, over the last 25 years it has become increasingly apparent that receptors have evolved the ability to activate signaling pathways in a ligand-specific way and that this effect is driven by the ligand-receptor conformation. In recent years, this has perhaps been best studied for the GPCRs where we have a multitude of examples of ligand-specific activation of receptor signaling. This phenomenon has been referred to with many names over time but presently it is most frequently referred to as functional selectivity or biased signaling.

Biased signaling was first thought to be an artefact of overexpression of receptors in recombinant cell systems that were being used in research. However, over time it became apparent that these effects occurred not only at more physiological receptor expression levels but also that this phenomenon was actually physiologically relevant.^1,2 It is now apparent that most, if not all, GPCRs are capable of signaling bias and that provides a new opportunity for this receptor class in the development of new drugs with more refined therapeutic efficacy and safety profiles. Indeed, there are several examples of biased agonists of GPCRs that have advanced to clinical trials for a variety of indications.³ But is signaling bias a phenomenon unique to GPCRs?

What is obvious from a general understanding of the receptor classes is that they are all quite distinct from each other in terms of localization, structure and function. However, despite very distinct differences between the receptor classes, there are some general principles that they hold in common especially as it relates to activation.² One of the key common principles to all receptor classes is their reliance on conformational changes that occur in response to agonist binding that transforms the signal of ligand binding into recruitment and activation of signal transducers.^4-8 This has several implications in terms of pharmacology, including providing a mechanistic basis for allosteric modulation of receptor function. Indeed, all receptor classes currently recognized have not only been shown to rely on conformational changes to drive activation of signaling effectors but as a result, also have numerous examples of allosteric modulators that can either act to promote or inhibit activation.⁹ This has been utilized to great effect in terms of drug discovery especially as it relates to LGICs.

 Biased signaling has been rationally exploited in the development of pharmaceuticals most clearly in the case of NHRs where tissueselective drugs have been marketed for two decades.¹⁰ Understanding of key movements of helices within NHRs via crystal structures has been obtained and similarities can be drawn between the movements of these helices and those within the transmembrane domains of GPCRs. Most notably, helix 12 of NHRs is affected by NHR ligands and has been shown to determine the pharmacology of the ligand by modulating the conformation of the binding pocket for co-activators and co-repressors. Thus, cell/tissue-selectivity of action induced by given NHR agonists is dependent upon the key movements of helix 12 to create a binding pocket for co-activators and co-repressors11 just as key movements of the intracellular regions of the transmembrane helices of GPCRs create a binding pocket for associated G-proteins.¹² For both of these receptor classes, the shape and biochemical interactions of the binding pocket for associated proteins determines the biological effect of the receptor (G-protein signaling for GPCRs and promoter sequences for NHRs).

Similarly, key conformational rearrangements of RTKs upon ligand binding lead to the potential for signaling bias in this class of receptor as well.^6,7 Although, there have been few reports of signaling bias in this receptor class despite the obvious potential for small molecule antagonists of these receptors in treatment of inflammatory diseases, neurological disorders and oncology. This may be due to the fact that a common strategy for inhibiting RTKs is to block the ATP- binding site of the cytoplasmic kinase domain, thus functionally blocking the receptor from signaling. Alternatively, this receptor class is often targeted using specific antibodies that block dimerization of the receptor which is required for receptor activation. There are examples, however, of purposeful design of allosteric modulators to RTKs.¹³ Similar to both GPCRs and NHRs, examples of partial agonism in this receptor class have also been reported14 suggesting ligand- specific alteration in signaling efficiency in the presence of certain ligands. In addition, there are is also evidence for selective antagonism of some signaling effectors but not others in the presence of a small molecule antagonist to the fibroblast growth factor receptor.¹⁵ Taken together, these data suggest the potential for biased agonism at RTKs.

It is difficult to imagine how signaling bias might apply to LGICs. However, if one accepts that ion flow through a given ion channel represents the signaling effector of the channel then we can find examples of signaling bias even within this receptor class. Upon ligand binding (often allosteric) to LGICs, conformational changes within the relative positioning of the channel subunits occurs thereby opening the pore at the center of the channel. The channel pore is exquisitely selective to the nature of the ion (anion or cation) as well as its chemical nature (sodium, calcium, chloride etc.) due to the biochemical interactions that take place in the channel pore that allow passage of the ion through the membrane. However, there are reports of promiscuity in ion flow through ion channels such as in the purinergic LGIC, P2XR, that upon prolonged stimulation by agonists develops permeability to even large macromolecules (reviewed in16).

In addition, some channels can undergo changes in ion selectivity depending upon the stimulus/agonist such as in potassium channels and some TRPV channels.¹⁷ Indeed, the TRPV1 channel has shown to alter its ratio of cation permeability depending upon the agonist bound, thereby clearly demonstrating agonist-specific signaling (signaling bias) for this receptor.¹⁷

In conclusion, conformational changes are a universal mechanistic determinant of receptor activation across all receptor classes. We now understand that this is at the heart of activation of signaling effectors following agonist binding and gives rise to allosteric modulation of receptor function. Looking closely across receptor pharmacology, we can find evidence to support that signaling bias is not restricted only to GPCRs but is actually a common potential for all receptor classes. This provides a new opportunity for the development of improved therapeutics with more refined efficacy and safety profiles regardless of the receptor class being targeted.

References

1. Arey, B.J., et al., Induction of Promiscuous G Protein Coupling of the Follicle-Stimulating Hormone (FSH) Receptor: A Novel Mechanism for Transducing Pleiotropic Actions of FSH Isoforms. Molecular Endocrinology, 1997. 11(5): p. 517-526.
2. Arey, B.J., An Historical Introduction to Biased Signaling, in Biased Signaling in Physiology, Pharmacology and Therapeutics, B.J. Arey, Editor. 2014, Academic Press: San Diego. p. 1-39.
3. Correll, C.C. and B.A. McKittrick, Biased Ligand Modulation of Seven TransmembraneReceptors (7TMRs): Functional Implications for Drug Discovery. Journal of MedicinalChemistry, 2014. 57(16): p. 6887-6896.
4. Absalom, N.L., T.M. Lewis, and P.R. Schofield, Mechanisms of Channel Gating of the Ligandgated Ion Channel Superfamily Inferred from Protein Structure. Experimental Physiology, 2004. 89(2): p. 145-153.
5. Baron, V., et al., The Insulin Receptor Activation Process Involves Localized Conformational Change. Journal of Biological Sciences, 1992. 267: p. 23290-23294.
6. Dawson, J.P., et al., Epidermal Growth Factor Receptor Dimerization and Activation Require Ligand-Induced Conformational Changes in the Dimer Interface. Molecular and CellularBiology, 2005. 25(17): p. 7734-7742.
7. Frankel, M., et al., Conformational Changes in the Activation Loop of the InsulinReceptor’s Kinase Domain. Protein Science : A Publication of the Protein Society, 1999. 8(10): p. 2158- 2165.
8. Hoffmann, C., et al., Conformational Changes in G-protein-coupled receptors—the Quest for Functionally Selective Conformations is Open. British Journal of Pharmacology, 2008. 153(Suppl 1): p. S358-S366.
9. Changeux, J.-P. and A. Christopoulos, Allosteric Modulation as a Unifying Mechanism for Receptor Function and Regulation. Cell, 2016. 166(5): p. 1084-1102.
10. Flaveny, C.A., et al., Biased Signaling and Conformational Dynamics in Nuclear Hormone Receptors, in Biased Signaling in Physiology, Pharmacology and Therapeutics, B.J. Arey, Editor. 2014, Academic Press: San Diego. p. 103-135.
11. Huang, P., V. Chandra, and F. Rastinejad, Structural Overview of the Nuclear ReceptorSuperfamily: Insights into Physiology and Therapeutics. Annual review of physiology,2010. 72: p. 247-272.
12. Dror, R.O., et al., Activation Mechanism of the β2-adrenergic Receptor. Proceedings of the National Academy of Sciences, 2011. 108(46): p. 18684-18689.
13. De Smet, F., A. Christopoulos, and P. Carmeliet, Allosteric Targeting of Receptor Tyrosine Kinases. Nature Biotechnology, 2014. 32: p. 1113.
14. Riese, D.J., Ligand-based Receptor Tyrosine Kinase Partial Agonists: New Paradigm forCancer Drug Discovery? Expert opinion on drug discovery, 2011. 6(2): p. 185-193.
15. Bono, F., et al., Inhibition of Tumor Angiogenesis and Growth by a Small-Molecule MultiFGF Receptor Blocker with Allosteric Properties. Cancer Cell, 2013. 23(4): p. 477-488.
16. Herrington, J. and B.J. Arey, Conformational Mechanisms of Signaling Bias of Ion Channels, in Biased Signaling in Physiology, Pharmacology and Therapeutics, B.J. Arey, Editor. 2014, Academic Press: San Diego. p. 173-207.
17. Chung, M.-K., A.D. Güler, and M.J. Caterina, TRPV1 Shows Dynamic Ionic Selectivity During Agonist Stimulation. Nature Neuroscience, 2008. 11: p. 555.

Author Biography

Brian J. Arey, PhD is Director of Mechanistic Pharmacology within Leads Discovery and Optimization at Bristol-Myers Squibb Co. where he leads a team of scientists dedicated to understanding mechanism of action of drug candidates and bringing innovative solutions to complex biological problems. He has nearly 25 years of experience within pharmaceutical drug discovery having contributed to marketed medicines across many different disease areas including endocrine, metabolic and cardiovascular diseases. He was an early pioneer studying signaling bias as it relates to GPCRs where his laboratory demonstrated the physiological significance of signaling bias to the gonadotropin system and led drug discovery programs to rationally design biased agonists for GPCRs. In addition, he has championed the concept of signaling bias as a common mechanism across receptor classes.