PACAP Signaling: A New Mechanism of Medication Overuse Headache from Opioids?

A recent animal study shows that inhibiting pituitary adenylate cyclase-activating polypeptide alleviates medication overuse headache following morphine. The findings have interesting translational potential.

Despite the serious adverse effects of opioids, these drugs are still widely used to treat pain and headache. One survey study even estimated that over half of patients presenting to a tertiary care headache center were prescribed opioids at some point for headache.

But chronic opioid use comes with an additional problem: It can cause or exacerbate headaches, especially in people already suffering from migraine or other headache disorders. This phenomenon is known as medication overuse headache (MOH).

Now, a new study has identified a novel mechanism of MOH involving signaling between the neuropeptide pituitary adenylate cyclase-activating polypeptide (PACAP) and one of its receptors, PAC1. Researchers led by Amynah Pradhan, Washington University School of Medicine in St. Louis, US, used two different migraine models – low-dose nitroglycerin or cortical spreading depression – to show that inhibiting PACAP-PAC1 signaling mitigated headache exacerbated by morphine overuse in mice.

“This is an exciting finding that reports a mechanistic link between opiates and migraine triggers,” said Andy Russo, a migraine researcher at the University of Iowa, US.

“In Europe, MOH is not a problem since guidelines are clear that opioids should not be used, but it continues to be a problem in America. This study is a good first step to work around the problem,” according to Russo, who was not involved with the study.

The research appeared in iScience on February 17, 2023.

A preclinical model of MOH
Pradhan said that the fundamental goal of the paper was to see if she and her collaborators could develop novel models of opioid-induced MOH.

“It brings us closer to what the clinical picture might look like,” she told Migraine Science Collaborative in an interview.

The team began by treating mice with morphine for 11 days, pairing this with an intermittent dose of nitroglycerine, a known migraine trigger in people, on day 3. Then they applied von Frey hair filaments to the head to test the animals for cephalic allodynia.

“We showed that low doses of opioids by themselves, every day, do not produce chronic cephalic allodynia,” said first author Zach Bertels. “But when we paired opioids with a low dose of nitroglycerin, this caused cephalic allodynia.”

The authors said this is reminiscent of the MOH that people with pre-existing migraine often experience after receiving opioids, whether for headache or other conditions.

“Over 50% of migraine patients in the United States still get prescribed opioids, and about 30% to 35% of them still use them for treatment of headache. It’s important to remember that migraine patients often have comorbidities with other types of pain disorders, so they might be prescribed opiates for something else,” said Pradhan.

The team also tested whether opioid-induced MOH could be observed in the cortical spreading depression (CSD) model of migraine. CSD is a wave of altered neural activity in the brain that is considered a physiological correlate of migraine aura.

So, the investigators again treated mice with morphine, for 4 days, followed by application of potassium chloride, a substance that can initiate CSD, directly onto the dura. Based on electrophysiological measurements, the team saw increased CSD in the animals receiving the opioid, compared to control mice, along with increased cephalic allodynia. The results thus demonstrated a key role for opioids in another migraine model.

PACAP inhibition prevents opioid-induced headache
The investigators next turned their attention to possible mechanisms underlying the effects of opioid overuse on migraine, setting their sights on PACAP.

Pradhan’s group was the first to identify a role for PACAP, and one of its receptors, PAC1, in opioid-induced MOH, in a study published in 2019.

“We previously found that if animals were given chronic opioids, or if they were treated in the chronic nitroglycerin migraine model, there was an upregulation of PACAP. It made us think PACAP could be a bridge between opioids and migraine,” said Pradhan.

The team tested whether inhibiting PACAP signaling could block cephalic allodynia in their migraine models. First, they administered a PAC1 antagonist, called M65, after cephalic allodynia had been established in their morphine-nitroglycerine model and measured the behavioral effects of this intervention.

The group found that blocking PACAP signaling completely inhibited the cephalic allodynia caused by the combination of morphine and nitroglycerin. It also reduced CSD in their model of opioid-exacerbated migraine aura.

“This showed allodynia can be blocked with M65, the PAC1 receptor antagonist. It’s therefore likely that opioids work through some sort of PACAP or PAC1 receptor cell population to cause headache,” said Bertels.

Russo thinks the translational potential of these findings is significant, especially considering ongoing testing of anti-PACAP therapy in people.

“There is currently a PACAP-blocking antibody in clinical trials for migraine. Both PACAP antibodies and PAC1 receptor antibodies had safe profiles in previous Phase I clinical trials and appear safe in ongoing Phase II trials,” said Russo.

Graphical abstract from Bertels et al. MOR: Mu opioid receptor; SCx: somatosensory cortex; PAG: periaqueductal gray; TNC: trigeminal nucleus caudalis; TG: trigeminal ganglia. M65: PAC1 antagonist.

Where is PACAP?
The next question was where in the nervous system the PACAP system was likely to interact with the opioid system. To address that issue, the investigators looked for co-expression of PACAP/PAC1 RNA with mu opioid receptor (one of the main classes of opioid receptors) RNA in several regions of the brain known to be involved with migraine and pain processing. That included the trigeminal nucleus caudalis, somatosensory cortex, and lateral-ventrolateral periaqueductal gray (PAG). They also measured co-expression in sensory neurons in the trigeminal ganglia.

Results showed a strong overlap of PACAP/PAC1 and mu opioid receptor expression in all three brain regions. The PAG showed the strongest overlap, with almost 100% co-expression between the mu opioid receptor and PAC1, and 50% co-expression with PACAP.

In the trigeminal ganglia, 42% of mu opioid receptor-expressing cells also expressed PACAP, whereas 23% expressed PAC1.

“Overall, we saw that these PACAP/PAC1 cell populations and opioid receptor populations are linked in some of the key migraine pain regions,” said Bertels.

However, Russo said this co-expression data should be interpreted with caution, since it doesn’t prove a mechanistic link between PACAP and opioid signaling when it comes to facilitating opioid-induced MOH.

“As I see it, they only have a connection. But now we have the evidence to start looking for the mechanism in more detail. How, really, are PACAP and PAC1 potentiating the effects of morphine on headache?” he said.

Bertels agreed, saying an important next step is to investigate whether PACAP antibodies have a predominantly peripheral or central mechanism of action.

“It isn’t well known if PACAP antibodies cross the blood-brain barrier. It would be interesting to focus on the trigeminal ganglia, specifically, how these receptor populations interact outside of the brain, and how this signal would feed forward into pain mechanisms we see in migraine,” Bertels said.

“The biggest gap moving forward is site of action; this needs to be addressed,” Russo said, along similar lines. “I don’t know if the PACAP antibody would be useful if it has a central target. It could be make-or-break for the antibody, as it would open up the chance for more side effects,” Russo explained.

Russo also pointed to an additional drawback of the co-expression experiments.

“One other limitation, as noted by the authors, was that localization of MOR [mu opioid receptor], PACAP, and PAC1 was based on RNA expression, which may not reflect protein. While RNA is a good indicator of protein, the two do not always match,” he added.

What about CGRP?
One curious finding was a lack of a role for calcitonin gene-related peptide (CGRP) in MOH. The team found that administration of olcegepant, a CGRP receptor antagonist, did not inhibit cephalic allodynia in the morphine-nitroglycerin model.

“This is a very puzzling finding,” said Russo. “It was shown way back in the 1990s by Rémi Quirion’s group that there was a link between CGRP and the mu opioid receptor. So why do we not see olcegepant working?”

And, a study published in 2001 found that CGRP knockout mice had decreased signs of morphine withdrawal, again suggestive of links between the CGRP and opioid pathways.

Bertels and Pradhan suspect that PACAP may have a unique role in opioid-induced MOH.

“The fact that CGRP antagonists did not affect opioid-induced MOH, but PACAP inhibitors did, suggests that PACAP is a special kind of bridge between these two issues,” said Pradhan.

Pradhan and her team are now delving deeper into the molecular pathways of the PACAP system, to learn more about its potential contributions to MOH.

Fred Schwaller, PhD, is a freelance science writer based in Germany. Follow him on Twitter @SchwallerFred

Reference
PACAP-PAC1 receptor inhibition is effective in opioid induced hyperalgesia and medication overuse headache models.
Bertels et al.
iScience. 2023 Feb 17;26(2):105950.