G-Protein-Coupled Receptors in Drug Discovery

GPCR membrane bound receptor ligand

G-protein-coupled receptors (GPCRs) play a big role in the human body. As they maintain many of our key functions, much research is done to better understand the structure and function of GPCRs. In addition, GPCRs are potential starting points in many drug discovery programs. We look at how X-ray crystallography can provide a method to probe the structure of GPCRs, with a case study of the μ-opioid receptor.

G-Protein-Coupled Receptors

GPCRs are receptors that respond to hormones, neurotransmitters, ions and even photons of light. They then respond to these stimuli by regulating signal transduction pathways, which exert an effect on our bodies. ‘Activators’ of receptors are known as agonists, while ‘deactivators’ are known as antagonists. Collectively, molecules that bind to a receptor are called ligands.

As their name suggests, GPCRs contain G-proteins, also known as guanine-nucleotide-binding proteins. They act as molecular ‘switches’ inside the cell, turning components on or off. This, in turn, leads to a variety of responses. Understanding the structure-function relationships of GPCRs can, therefore, act as useful starting points in drug design and development.

X-Ray Crystallography

Protein X-ray crystallography and related techniques provide high resolution 3D structural information. A single pure crystal of pure protein scatters X-rays onto a detector, and rotation of the crystal provides a 3D scattering profile. Computational techniques then convert this scattering data into positions of the atoms, creating a 3D model of the protein.

Drug discovery and development programs rely heavily on X-ray crystallography to identify the structure of drug targets. The X-ray structure of human immunodeficiency virus (HIV) protease – an important enzyme in the virus’ lifecycle – was elucidated in the late 1980s. The 3D structure allowed for analysis of its active sites, enabling us to predict and synthesize compounds that might block them. This led to the development of HIV protease inhibitors, potent drugs against HIV still in use today.

Related: The first HIV therapy was discovered even before HIV

It might sound simple to ‘take a picture’ of a crystal, but the formation of a viable crystal of pure protein is another matter altogether. A sample of such purity can take months of optimization to obtain, following which crystallization must occur. Crystallization itself is largely trial and error because the conditions with favorable enthalpy are not immediately clear. Depending on the solvent system and the relative concentration of protein, it may take months to years before crystals form, if one even forms at all.

Following this, data collection and processing is a tedious and time-consuming task, even with the help of supercomputers. Overall, the end-to-end process can take years, with high risks of failure throughout the entire project. Hence, the cost to benefit analysis of crystal structure elucidation is still a matter of much debate.

An X-ray diffractometer; a single tiny crystal is suspended by the goniometer – arms which move to vary the rotational angles during data collection (Source)

Case Study: The μ-Opioid Receptor

The μ-opioid receptor (pronounced ‘mew’!) can produce a range of effects in humans when activated by chemicals known as opioid alkaloids. Activation of the μ-opioid receptor (MOR) leads to an analgesic effect, making it an effective target for painkillers such as morphine. On the other hand, it also triggers mechanisms of tolerance and addiction.

Both endogenous and synthetic opioids exist; however, there is yet to be an agonist that activates the MOR in an ideal manner. The discovery of such a ligand will have potentially far-reaching healthcare, social and economic benefits1.

The mouse MOR (which has 94% similarity to the human form) is a good starting point, and solving its structure might open up the possibility to create such an agonist. By conducting structure analysis, one can design a drug that provides therapeutic benefit, while suppressing unwanted adverse effects.

Crystal Structure Elucidation

The poorly structured third intracellular loop on the μ-opioid receptor was identified as a hindrance to its crystallization due to its instability in solution. Hence, it was truncated and replaced with a more stable protein (T4 lysozome) through GPCR engineering2. Protein fusion techniques such as this is a common strategy in crystal engineering, serving to restrict the movement of transmembrane (TM) helices. Stabilizing these regions improves the chances of crystallization.

It was found that the T4 lysozome also improved polar (charged) interactions between the molecules, allowing for better crystallization. The protein crystals were produced within a lipid mesophase (a liquid-solid intermediate) solvent system, improving its solubility3. Compared to traditional detergent solvents, the lipid mesophase also required less solvent, reducing the time taken to form a crystal.

PDB 4DKL structure GPCR transmembrane domain bound to T4 lysozome
X-ray crystal structure of a single μ-opioid receptor bound to T4 lysozome, the bundle of alpha helices form the transmembrane core characteristic of GPCRs (Source: PDB ID 4DKL, created with RCSB PDB and associated publication)

The factors listed above provided an ideal environment for the crystallization of membrane proteins – including GPCRs – as they are generally unstable when removed from their native membrane environment. X-ray crystallography was then used to generate a scattering profile of pure mouse MOR protein fused with the T4 lysozome, bound to an antagonist. Elucidation of the first 3D crystal structure of MOR was reported by Kobilka’s group in 2012. Brian Kobilka won the Nobel Prize that same year for his work on G-protein-coupled receptors.

3D Structural Analysis

The μ-opioid receptor – like all GPCRs – are membrane proteins that consist of 7 linked alpha helices with an extracellular N terminus and an intracellular C terminus. The topology of 7 linked alpha helices is conserved across all GPCRs but may differ in the lengths of individual TM loops. The extracellular domains and ligand binding regions are variable regions that differ between classes of GPCRs4.

The aforementioned alpha helices are situated within the lipid bilayer of the cell membrane, meaning they consist mainly of hydrophobic residues. They have to be aligned such that they all face outward while the remaining hydrophilic residues, if any, can form a polar core toward the center of the protein5. This polar region allows for charge interactions with incoming ligands – which is the case for MORs.

Studies on GPCRs show that when a ligand is bound, conformational changes to the transmembrane domains causes the entire three-dimensional structure of the receptor to change. This is what activates the G-protein signaling mechanism4.

Top-down view of β-FNA (space-filling balls) bound to residues (stick models) on the transmembrane alpha helices (Source: PyMol software)

Crystal structure of the MOR shows the binding interactions with a ligand, beta-funaltrexamine (β-FNA). It binds through at least 9 of its residues, with a total of 14 residues within 4 Å of the ligand. The nature of binding consists of charge-charge interactions, hydrogen bonding, hydrophobic interactions, and a covalent bond also being formed between the K233 residue and a carbon atom on β-FNA. Also, 2 water molecules form a binding network with the H297 residue to interact with phenol group on β-FNA1.

Potential Drug Design

The binding region of the μ-opioid receptor is situated close to the extracellular surface of the cell (as shown in Fig. 1a), forming a shallow groove. This suggests a loose binding interaction, with faster dissociation rates. Hence there is a higher possibility of the ligand being displaced by solvent molecules, deactivating the receptor. In contrast, some receptor ligands are ‘buried’ deeper within the helical bundles and with their exit hindered by certain residues. Some muscarinic receptor ligands, for example, exhibit very slow dissociation kinetics1.

Due to the shallow binding region in MORs, ligands with few and/or weak interactions do not remain bound to the transmembrane helices for long. This also shortens the amount of time that the receptor can signal. Using structure-based design to synthesize a ligand that binds more tightly, a more effective agonist can be produced to offset the ‘shallow’ binding region. This can be done, in theory, by increasing the strength or number of non-covalent interactions.

Other binding sites on MOR can also be exploited by interactions with ligands. Adverse effects such as addiction is a common consequence of MOR activation, and there appear to be multiple sites on the receptor that can influence this6. It is therefore apparent that understanding the structure and function of the μ-opioid receptor, as well as other G-protein-coupled receptors, can be of great benefit.

Cover graphic: artist’s rendition of a μ-opioid receptor on a cell membrane by Melanie (@nanoclustering)


  1. Manglik, A., Kruse, A. C., Kobilka, T. S., Thian, F. S., Mathiesen, J. M., Sunahara, R. K., … & Granier, S. (2012). Crystal structure of the µ-opioid receptor bound to a morphinan antagonist. Nature485(7398), 321.
  2. Rosenbaum, D. M., Cherezov, V., Hanson, M. A., Rasmussen, S. G., Thian, F. S., Kobilka, T. S., … & Kobilka, B. K. (2007). GPCR engineering yields high-resolution structural insights into β2-adrenergic receptor function. science318(5854), 1266-1273.
  3. Liu, W., & Cherezov, V. (2011). Crystallization of membrane proteins in lipidic mesophases. JoVE (Journal of Visualized Experiments), (49), e2501.
  4. Rosenbaum, D. M., Rasmussen, S. G., & Kobilka, B. K. (2009). The structure and function of G-protein-coupled receptors. Nature459(7245), 356.
  5. Crasto, C. J. (2010). Hydrophobicity profiles in G protein-coupled receptor transmembrane helical domains. Journal of Receptor, Ligand and Channel Research2010(3), 123.
  6. Contet, C., Kieffer, B. L., & Befort, K. (2004). Mu opioid receptor: a gateway to drug addiction. Current Opinion in Neurobiology14(3), 370-378.

You may also like...

Leave a Reply