Greg Dussor, PhD, is an associate professor in the School of Behavioral and Brain Sciences at the University of Texas at Dallas. Dussor uses animal models to study migraine pain. Freelance journalist Allison Marin recently spoke with Dussor about the biological mechanisms underlying migraine, how the new understanding of this condition differs from previous ideas, and the work going on now in his lab. Below is an edited transcript of their conversation.
Historically, what have researchers thought causes migraine?
For almost as long as we’ve been thinking about it—going back many centuries—migraine has been thought of as a vascular disorder, meaning that blood vessels are the primary cause. Anyone who has had a migraine, and other headaches as well, can tell you that they feel blood pulsing during the headache. Several scientific findings support this idea. Vasodilating drugs, which enlarge blood vessels [and allow more blood to pass through], trigger migraines, while vasoconstricting drugs, which cause blood vessels to shrink, are effective at treating migraines.
The vasodilation story has been the leading hypothesis up until the last 10 years. Prior to that, researchers were beginning to realize there are some holes in that story, one of which is that not all drugs that dilate blood vessels are capable of triggering migraines. If it were as simple as dilated vessels causing migraines, you would expect that anything capable of enlarging blood vessels would be capable of causing a migraine, and that’s just not true. It’s also the case that not every vasoconstrictor helps treat migraines. Further, studies have shown that the rate of the throbbing pain of migraine does not match the heart rate. So it can’t simply be that blood pulsing through dilated vessels is the only problem.
How has the field responded to these doubts about the vascular theory of migraine?
In the last 10 years, researchers around the world have started using advanced imaging techniques to measure the diameter of blood vessels in patients who are having migraines (also called migraineurs). In one study, researchers used a vasodilating drug to trigger migraines and found that blood vessel diameter did indeed increase during the migraine, but another study with a different vasodilator reported that the diameter was unchanged. So, in the latter study, the patients got migraines from the drug, but it didn’t actually dilate vessels at the time they were having the migraines.
The most recent nail in the coffin for the vascular theory was a Danish imaging study in which researchers studied patients with naturally occurring migraines—not those induced by a vasodilating drug, as in previous studies. In the Danish study, there was very little vasodilation observed—certainly not a large enough amount to conclusively say that dilating vessels is the cause of migraine. That makes the vasodilation story problematic, because in naturally occurring migraines, there is little to no vasodilation.
It sounds like the field has really shifted directions. What is the latest thinking about the biological mechanisms of migraine?
The field has largely moved in the direction that migraine is a neurological disorder, meaning that there is something about the nervous system that is initiating and perpetuating a migraine attack. There are parts of the nervous system that we think can contribute to each of the phases of migraine and play a critical role in the pathology of migraine.
That does not mean that we have to completely ignore vessels as a whole because they might still contribute something to the pathology of migraine, but they might do that in the absence of vessel dilation. So, the field hasn’t necessarily completely eliminated the possibility that vessels contribute to migraine; it’s just probable that dilated vessels don’t play a role.
What parts of the nervous system contribute to the various phases of migraine?
Migraines start with a prodrome, or premonitory, phase in which patients have some signal that the migraine is coming. If you look across the symptoms in the premonitory phase—excessive yawning, and changes in appetite, mood and/or energy level—they don’t seem to have a whole lot to do with each other. However, a brain area called the hypothalamus is in charge of regulating a lot of these behaviors. Imaging studies show an increase in brain activity in the hypothalamus before the migraine really takes off, and so the hypothalamus is an area of interest in the brain thought to contribute something to the very early stages of migraine.
After the prodromal phase, migraines progress to sensory disturbances, called auras, the most common being anomalies in the visual system, such as blind spots or geometric patterns that move across the visual field. Sensory auras are almost certainly due to changes in neuronal activity in the outermost region of the brain known as the cortex.
Then we get to the headache phase. There certainly can be a vascular contribution during this time, but there also has to be some role for the nervous system in the headache phase because ultimately a headache is painful, and pain requires the nervous system. It may be that the neurons responsible for the pain of a migraine have become sensitized to things that the blood vessels are doing, not necessarily to dilating vessels, but rather just to the normal activity of the vessels. Here the pathology is actually not in the vessels; the vessels are doing what they do on a daily basis and are responding normally to what the rest of the body is telling them to do. The pathology lies in the nervous system, which is now responding in a pathological way to what the vessels are doing. In people who do not get migraines, they may be having identical vascular events, but their nervous system is not responding to those events with pain.
What else do researchers think about what causes the pain of migraine?
Much of the field agrees that the pain phase of migraine is mediated by pain-sensing neurons that send information into the nervous system from the meninges, the layers of membranes that cover the brain and spinal cord. The meninges seem to be the important site of the pain of migraine.
There were a couple of very well-known papers in the field, published in 1940, describing experiments in which neurologists probed the meninges of humans who were undergoing open-skull surgery. The patients were conscious, like open-skull surgery patients often are, which allowed the neurologists to ask the patients, “if I probe here what do you feel?” The only sensation those patients reported was pain. In fact, the neurologists could recreate the common locations of migraine pain—behind the eye and into the temple—if they stimulated specific locations on the meninges.
Those studies led to the long-standing idea that the meninges send painful signals if you stimulate them. The meninges are the only pain-sensitive structure inside the skull. You cannot feel pain from the brain itself because there are no pain-sensing neurons that project from it. We also know that an infection of meninges, known as meningitis, causes very painful headaches. Together, these observations implicate the pain-sensing neurons that are projecting from the meninges to the brain in migraine pain.
What actually happens to these pain-sensing neurons during migraine remains a huge question in the field. Some have proposed that the events in the cortex that cause aura can ultimately turn into pain by activating neurons in the meninges. Others propose sterile inflammation of the meninges, in other words, inflammation not due to any bacterial or viral infection but from the body’s own natural inflammatory processes. Unfortunately, we just don’t know the answer yet. Is it a matter of some pathological event occurring in the meninges during a migraine that is not normally occurring, and that is what drives the pain input from the meninges? Or are those neurons sensitized to something that’s happening all of the time? Those are some questions we need to know the answer to.
Since the different phases of migraine seem to be different neurological events, are there any theories about what might tie them all together or lead to progression of the condition?
Yes—that’s where the hypothalamus comes in, because it is the master regulator of almost everything the body does, including many of the things that the nervous system does. The hypothalamus can modulate pain, and it can also produce the premonitory symptoms of migraine. The hypothalamus could also potentially cause some events that may be occurring in the cortex to change neuronal activity and produce an aura. The hypothalamus is one of the few places in the brain where you can tie a lot of the nervous system together.
The question, though, is what is wrong with the hypothalamus and how did it happen? The hypothalamus plays a critical role in the body’s response to stress, and stress is the most commonly reported trigger for migraine. So it could be that the hypothalamus changes its properties in response to repetitive stress exposure. Once this happens, the hypothalamus alters the way it regulates many neuronal functions, and not necessarily in a good or productive way. So new exposures to stress can now begin to produce the events within the nervous system that culminate in a migraine attack.
In your lab you study the mechanisms of migraine using animal models. How do you create and study migraine in animals when the causes in humans are still largely unknown?
This is really one of the big challenges in the field. What researchers have done is create models for some of the various phases of migraine. It’s hard to model changes in things like irritability and mood that might mimic the premonitory phase, because they are hard to detect and measure in animals. But there are pretty good models of the events in the cortex that we think are causing aura; we can induce changes in neuronal activity in the cortex pretty reliably. We can then study how these changes propagate, the pharmacology behind them, and how we can block them. What we can’t really do is tell if the animal is actually having an aura, as it is almost impossible to measure blind spots in the visual field of animals.
There are also ways to study the headache phase. We know that the nerve endings that send information from the meninges to the brain are critical for migraine headaches. We can stimulate those nerve endings in an animal and observe if that animal has a behavioral response that is consistent with a migraine. This is one of the main focuses of research in my laboratory, and we’ve come up with a variety of chemical methods to stimulate meninges.
When we look at the animal’s behavior, we typically measure hypersensitivity of the skin in the face of the animal, specifically on the forehead between the eyes. This comes from the observation in humans that during the headache phase of a migraine facial skin is hypersensitive.
We also look at a variety of different exploratory behaviors that animals will do when you put them in a chamber with lots of open space. Normally, rodents explore when placed in a new environment. One of the hallmarks of migraine in humans is a decrease in physical activity, because movement makes the headaches worse. If we can see a decrease in the physical activity of animals after we give them what we think causes a headache, then that fits along with the other pieces of data.
What have you been finding using these rodent models in your laboratory?
One of the first observations we made when we got into the field was that the nerve endings that project from the meninges to the brain are extremely sensitive to changes in pH—the level of acidity. Very small decreases in pH can start to activate these nerve endings and cause what looks like a headache in rats. Following up on that, we figured out that there’s a specific family of receptors, protein molecules appropriately named acid-sensing ion channels, that mediate this effect. We have spent a fair amount of time investigating how these acid-sensing ion channels might contribute to headache.
The next question became: what causes these pH shifts in migraine? We reasoned that the receptor for the pH change may be conveniently positioned near where the pH change originates, so we started looking at the anatomy of the meninges to see where these receptors are located. We found that the pH-sensing receptor is certainly present on the pain-sensing neurons that are critically important for headache. But, unexpectedly, the receptor is also found in cells that are not neurons, called fibroblasts, in the meninges. These findings have, to some extent, taken us away from the original question, which is where does the pH change come from, to looking at how the different cell types in the meninges contribute to the process together.
Fibroblasts play a critical role in the overall maintenance of the meninges as well as keeping their structure intact. Subsequently we’ve looked at other receptors besides the acid-sensing ion channels that are present on the meningeal fibroblasts. One of the interesting things we’ve found is that the fibroblasts can respond to norepinephrine, which is one of the classic hormones released during stress. As mentioned earlier, stress is one of the major triggers of migraine in humans. It has been known for quite some time that there are receptors for norepinephrine on blood vessels as well as on pain-sensing neurons. What we didn’t appreciate until now is that these receptors are present on fibroblasts in the meninges as well, so there’s another cell type that we need to think about. Together, these different cell types may play some role in the response to stress, and it takes the coordinated effort of all of them to produce the migraine headache.
Where is research on the mechanisms of migraines heading, and what are the remaining questions?
Migraine is a heterogeneous disorder—there are many ways to end up having them, and we need to understand all of those mechanisms. If we only understand one mechanism that’s capable of causing a migraine and we try to develop a therapeutic that is entirely based on that mechanism, it will work very well in patients who have migraines due to that particular mechanism, but not in all the other patients who have migraines for a whole host of other reasons.
One example of a specific mechanism-based therapy is antibodies that block the function of calcitonin gene-related peptide (CGRP), a neuropeptide [a molecule that neurons use to communicate with each other] that is released from pain-sensing neurons and is elevated during migraine attacks. CGRP antibodies are currently in clinical trials for the treatment of migraine, and early results show that they are very promising for both episodic and chronic migraines [see related RELIEF feature story]. I think CGRP antibodies are going to make a big impact on the treatment of migraine.
Unfortunately, there are migraineurs who don’t respond at all to CGRP antibodies. This implies that CGRP is not contributing to their migraines, and we don’t yet know what is happening in those non-responders. Some patients have migraines triggered by stress, some have migraines triggered by particular smells, and for others migraines are triggered by certain foods. Are all of those the same—for instance, is a stress-triggered migraine the same as a smell-triggered migraine, or are the underlying mechanisms different?
We need to better understand all of the different mechanisms that lead to a migraine. Some of them may be blood vessel-dependent and some of them may not; some may be CGRP-dependent, and some may not. The heterogeneity of migraine probably means we’ll need many different therapeutics, based on many different mechanisms, to effectively treat all migraine patients.
Home page image credit: zenina/123RF Stock Photo