Opioids provide rapid and potent pain relief, but these drugs are associated with numerous side effects, including respiratory depression, tolerance, dependence, and addiction. Now, a new study led by Daniel Wacker and Bryan Roth, University of North Carolina at Chapel Hill, US, raises the possibility of developing new opioids with lower risk of such adverse consequences.
An international team of researchers successfully capture, for the first time, the X-ray crystal structure of the human kappa opioid receptor (KOR) by using a nanobody to stabilize the receptor in its active state. The same group had previously characterized the structure of KOR in its inactive state (Wu et al., 2012), and comparisons between the two forms reveal insights into KOR pharmacology and signaling that will aid the design of safer and more effective KOR therapeutics.
“This is a very nice study, building on previous approaches employed with the mu opioid receptor,” said John Streicher, University of Arizona, Tucson, US, who was not involved in the study. “After crystallizing the active state of the kappa opioid receptor, they go beyond that by combining molecular pharmacology with mutational studies to determine how specific ligands activate the receptor and what makes specific ligands selective for the different subtypes of the opioid receptors.”
The findings were published online January 4 in Cell.
KOR revisited
The KOR has long interested scientists as a potential alternative analgesic to the mu opioid receptor agonist morphine and its litany of unwanted side effects. But selective KOR agonists developed previously were quickly found to produce problems of their own, including dysphoria and hallucinations, thus limiting their therapeutic potential (Chavkin, 2011).
In recent years, interest in KOR ligands has re-emerged with the discovery that they might be useful in additional therapeutic areas, including depression and addiction. More relevant to the pain field was the recent demonstration of biased agonism at the KOR, that is, selective activation of beneficial signaling pathways over deleterious pathways that all come into play after binding of agonists to G protein-coupled receptors (GPCRs) like KOR. Exploiting the promise of biased agonism could potentially help in the design of safer KOR therapeutics devoid of the adverse effects of conventional KOR agonists (Brust et al., 2016; see PRF related news story here). KOR agonism could also have certain advantages over mu opioid receptor-targeted approaches.
“A lot of research has been conducted on the mu opioid receptor to try and develop safer medications, as morphine and other analgesics primarily act through this receptor. But it has been shown for a long time that its closely related sister receptor, the kappa opioid receptor, may have similar potential, if not more, because it does not mediate the rewarding effects or the respiratory depression that are the main cause of overdose deaths associated with opioids,” said co-senior author Daniel Wacker.
“Therefore, we started to conduct detailed mechanistic studies on what the structural determinants of activation of the kappa opioid receptor would be. This would help us to better understand how to design a drug that would activate it, and specifically activate pain-relieving pathways,” according to Wacker.
X-ray crystallography has traditionally been used to determine the structure of many different GPCRs. This approach involves condensing a protein into a crystal lattice, subjecting it to X-ray, and then studying the diffraction of the X-ray beams to identify receptor structure. However, this remains a challenge with opioid receptors, particularly in their transient active state.
First author Tao Che and colleagues set about addressing this issue by using nanobodies—small fragments of antibodies—that bind tightly to KOR and keep it in an active state. They injected KOR liposomes bound to the KOR agonist salvinorin A into a llama and then used a technique known as phage display to identify nanobody clones of interest. Next, using a cell-based bioluminescence resonance energy transfer (BRET) assay to measure nanobody binding to KOR, the investigators discovered a clone called Nb39 that could bind KOR with increasing concentrations of salvinorin A. This suggested that Nb39 recognized the active state of KOR.
To identify a ligand suitable for crystallization of the Nb39-stabilized active state of KOR, the researchers again turned to BRET and found that the experimental opioid agonist MP1104 displayed the highest potency and efficacy for recruitment of Nb39 to the receptor. Since Nb39 and MP1104 promoted a stable active state of KOR, the authors determined the X-ray crystal structure of the KOR-MP1104-Nb39 complex, which hitherto had been unknown.
Structure drives function
Next, the authors sought to compare a previously characterized structure of an inactive state of KOR (known as KOR-JDTic; Wu et al., 2012) with the newly characterized structure of the active state KOR-MP1104-Nb39 complex. Substantial rearrangements in the relative positions of transmembrane helices (TM) were observed between the two structures; GPCRs have seven of these helices. Many of the differences observed in the KOR active state structure were reminiscent of those seen in other active state GPCRs and are thought to facilitate the binding of signal transducers.
Interactions of agonists with the TM helices take place through a range of hydrogen bonds involving the binding pocket of the receptor. The structural interaction between a ligand and its receptor is crucial for determining the subsequent downstream signaling and eventual cellular response.
For example, the researchers demonstrated that MP1104, the opioid agonist, interacted with transmembrane helix 5 (TM5) of the active state KOR. Replacing a molecule from this interaction (a hydroxyl group) with another (a methoxy group) decreased the affinity of MP1104 for KOR. However, this alteration affects MP1104’s binding to the mu opioid receptor to a greater degree, suggesting that interactions of a ligand with TM5 could be exploited for designing selective KOR ligands. Further, substituting a cyclopropylmethyl group with a methyl group on MP1104 resulted in a 15-fold reduction in the potency of this agonist.
“The power of the structural approach is that it provides a blueprint for chemically modifying specific aspects of ligands. If we change these ligands, then we may be able to change the signaling behavior of the receptor,” explained Wacker.
The basis of selectivity
There are two primary approaches to the design of safer opioids. One is to design ligands that, upon binding to their receptors, are biased in the signaling pathways they activate. In particular, pain researchers are designing ligands that favor the G protein pathway, which is associated with opioid analgesia, over the b-arrestin pathway, which is associated with opioid side effects (see PRF related story here). Another approach involves creating ligands that favor one subtype of opioid receptor over another (i.e., kappa over mu) with the intention of avoiding the detrimental consequences of mu opioid receptor activation.
The authors were particularly interested in the latter approach, and so they performed structure-activity relationship experiments to characterize the molecular determinants of selectivity at KOR. IBNtxA, a close analog of MP1104 that displays G protein signaling bias at the mu opioid receptor, was docked into the active state structure of KOR and the mu opioid receptor using computer modeling. The modeling showed that aromatic phenol groups of both MP1104 and IBNtxA interacted with various amino acid residues (portions of the receptors) in differing ways, depending upon the specific opioid receptor subtype involved. The results suggested that removal of the phenol molecule may reduce mu opioid receptor binding without impacting KOR binding.
Armed with this knowledge, Roth and colleagues developed a novel KOR agonist containing an altered phenol group. Comparison of this compound with nalfurafine, an approved IBNtxA-related analog for chronic itch, revealed a large reduction in mu opioid receptor affinity while retaining high KOR affinity. Therefore, modification of the phenol group could be a viable path to generating future KOR selective ligands with improved efficacy.
“The next step for this newly developed compound is to conduct in vivo studies to see if it produces beneficial analgesic effects in animal models without any serious adverse effects,” said Wacker. “But the whole thing is an iterative process, and, depending on what we see with the in vivo studies, we may have to go back to the drawing board and make additional modifications based on our structural insights,” he said.
Ultimately, the current study may result in a shift in the way drug design is approached in the future, according to Streicher.
“In the past, the main way to do drug screening was to screen a massive number of compounds with ever-increasing sizes of compound libraries,” he said. “However, that really has not come to fruition, as the number of new drug approvals is decreasing and the costs keep climbing. Having the active state of this and other receptors in the future will enable us to become more effective at rational drug design and have a compound that has the characteristics we expect and want.”
Dara Bree is a postdoctoral fellow at Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, US.
Image credit: From Che et al., 2018, with permission from Elsevier.
