Activation of the kappa opioid receptor (KOR) relieves pain and itch. However, downstream recruitment of β-arrestin proteins and accompanying dysphoria—a feeling of discomfort or aversion—and sedation prevent drugs targeting KOR from being widely used in the clinic. Now, new research suggests that through biased signaling, it may be possible to avoid β-arrestin recruitment and exclusively activate the pain- and itch-relieving properties of KOR.
Using the biased KOR agonist, triazole 1.1, researchers demonstrate antinociception and itch relief comparable to other KOR agonists in preclinical models. But the drug did not decrease dopamine release in mice, which is thought to produce the dysphoria seen with classical KOR agonists, nor did it cause dysphoria in the rat model of intracranial self-stimulation.
“These findings are an important piece of evidence supporting the hypothesis that β-arrestin mediates aversion downstream of the KOR,” wrote John Streicher, University of Arizona, Tucson, US, in an email to PRF. “They also support the further development of KOR biased agonists for pain relief with reduced side effects,” continued Streicher, who was not involved in the study.
The study was published online November 29 in Science Signaling and led by senior authors Laura Bohn, Scripps Research Institute, Jupiter, US; Thomas Martin and Sara Jones, Wake Forest School of Medicine, Winston-Salem, US; and Jeffrey Aubé, University of North Carolina at Chapel Hill, US.
Biased agonism …
Like other opioid receptors, KOR is a G protein-coupled receptor (GPCR) with multiple downstream pathways. Evidence suggests that activation of GPCRs such as KOR may signal to different downstream pathways in different contexts. For example, activation of KOR on a particular neuron type may result in G protein-coupled signaling, leading to analgesia and itch relief, while activation of KOR on another neuron type may instead cause the recruitment of β-arrestin proteins, leading to unwanted effects. β-arrestins have been implicated in producing dysphoria, as mice lacking the ability to recruit β-arrestins do not demonstrate aversion in response to KOR agonists, suggesting that this pathway may be the “dysphoric context” of KOR activation (Bruchas et al., 2007).
Given new advances in pharmacology, it may be possible to target one signaling context over another by using a specific ligand—a concept known as biased agonism or functional selectivity. This idea has been explored recently with the mu-opioid receptor (see PRF related news story), as well as the KOR (Morgenweck et al., 2015; White et al., 2015), with exciting results.
“This is a success story from the Molecular Library Program, sponsored by the NIH [National Institutes of Health, Bethesda, US], to try to find ligands for KOR,” Bohn told PRF. “In the screening effort, we identified a potential kappa agonist. The irony was that the screen used was one that measured β-arrestin recruitment, but we inadvertently found a biased agonist.”
The ligand, triazole 1.1, is one of a series of ligands discovered through the screen. In a previous study, Bohn and colleagues showed that triazole 1.1 preferentially activated G protein signaling over β-arrestin recruitment (Zhou et al., 2013). However, the behavioral effects of this ligand had not been fully examined until now.
… shows promising behavioral results
In the current study, first author Tarsis Brust and colleagues found that triazole 1.1 induced antinociception comparable to a reference compound called U50,488H, a previously characterized KOR agonist that activates both G protein and β-arrestin signaling, in the tail-flick assay in mice. The antinociceptive effects of both drugs were reversed by the KOR antagonist norbinaltorphimine, and neither drug had efficacy in KOR knockout (KOR-KO) mice, together demonstrating KOR selectivity.
To test if triazole 1.1 could also prevent itch, mice were injected at the back of the neck with chloroquine phosphate, a drug that induces robust scratching. Triazole 1.1 and U50,488H similarly blocked scratching behaviors at all doses tested. Here, too, the antipruritic effects of both drugs were blocked with norbinaltorphimine.
Many KOR agonists decrease dopamine transmission, thus acting as sedatives and decreasing locomotion. Importantly, triazole 1.1 did not affect locomotion in mice in the open-field activity test relative to vehicle control. This suggests that the drug does not affect dopamine, unlike U50,488H, which significantly reduced locomotion in wild-type but not KOR-KO mice.
To test the hypothesis that triazole 1.1 was not acting at off-target sites, the researchers examined the presence of both triazole 1.1 and U50,488H in the brain using liquid chromatography-mass spectrometry. Using doses that were found to yield different behavioral effects (i.e., unchanged locomotion using triazole 1.1, and decreased locomotion using U50,488H), they found that both drugs were present in comparable amounts in the striatum, a primary site of KOR modulation of dopamine, 30 minutes after injection. In addition, both drugs similarly prevented KOR-specific radioligand binding, suggesting that they both occupy the same site on KOR. Importantly, both drugs were similarly potent in activating G protein signaling in the striatum, and no binding was observed for either drug in KOR-KO mice.
A direct test of the dopamine hypothesis
The open-field test and binding data suggested triazole 1.1 may not affect dopamine signaling as other KOR agonists do. To directly test this idea, the group examined dopamine release in slice preparations of the nucleus accumbens from mice, including both core and shell regions of this brain area, using cyclic voltammetry, an electrochemical technique. The nucleus accumbens is a key part of the brain’s reward circuit that uses dopamine as its essential neurotransmitter.
While U50,488H produced a dose-dependent decrease in dopamine release in both the core and shell regions, triazole 1.1 did not similarly alter dopamine in these regions. Comparable results were seen using microdialysis to measure dopamine levels in freely moving mice.
A rat model of aversion
For additional evidence that triazole 1.1 avoids the dysphoria seen with classical KOR agonists, the researchers used a rat model of intracranial self-stimulation (ICSS). Rats were implanted with stimulating electrodes in the ventral tegmental area (VTA), a site of dopamine release, and then trained to press a lever for brain stimulation. In the ICSS model, classical KOR agonists reduce the rewarding effects of brain stimulation, consistent with aversion, and result in decreased lever pressing. In the new study, while U50,488H dose-dependently decreased ICSS, triazole 1.1 did not, suggesting that the biased agonist does not share the same aversive effects as classical KOR agonists.
In the ICSS model, peripheral injection of lactic acid causes irritation and pain at the site of injection, lowering the frequency or threshold of lever presses. In a final experiment using the ICSS model, prior administration of U50,488H did not alter the effects of lactic acid on ICSS. However, pretreatment with triazole 1.1 reduced the effects of lactic acid relative to vehicle controls. Importantly, the analgesic effects of triazole 1.1 were blocked upon pretreatment with norbinaltorphimine, decreasing the number of stimulations and indicating that the effects of triazole 1.1 are KOR dependent. Together, the results show that triazole 1.1 produces analgesia without the aversion associated with classical KOR agonists.
“This is a really important step forward,” wrote Thomas Kash, University of North Carolina School of Medicine, Chapel Hill, US, in an email to PRF. “It would be really interesting to look at some follow-up studies that probe additional KOR agonist-sensitive behaviors, such as conditioned aversion, as well as some addiction-like 'drug-taking' phenotypes.”
Streicher agrees that more testing is necessary to further understand the dysphoric effects of biased KOR agonists. A similar biased KOR agonist, RB-64, produced conditioned place aversion but did not alter ICSS responses (White et al., 2015), suggesting that there may be subtle yet important differences among tests of aversion.
KOR agonists have never been used in the US for pain treatment. “They’re really nice analgesics, and they work well, but they have always caused dysphoria and even hallucinations,” Bohn said. However, the possibility of avoiding such negative side effects with biased agonists such as triazole 1.1 may greatly impact the future of drug treatment for pain and itch, especially since kappa agonists do not have the same rewarding properties as many mu-opioid agonists.
“The study is really exciting for us because triazole 1.1 is from a whole series of compounds that we’ve been studying. It’s encouraging because we know they are doing something fundamentally different,” Bohn said.
Hillary Doyle is a PhD candidate and science writer studying pain and analgesia at Georgia State University in Atlanta.
Image credit: everythingpossible/123RF Stock Photo