Share this article on:

Michael A. Moskowitz: Cortical Spreading Depression is Key to Migraine Genesis

MacReady, Norra

doi: 10.1097/01.NT.0000271246.56688.45
News From the Headache Research Summit
Back to Top | Article Outline


  • ✓ Cortical spreading depression is key to the genesis of migraine, particularly migraine with aura, and it may play a role in the etiology of other brain disorders, including subarachnoid hemorrhage and trauma, as well as stroke.

PALM SPRINGS—Evidence is mounting that cortical spreading depression is key to the genesis of migraine, particularly migraine with aura, and it may play a role in the etiology of other brain disorders, as well.

So said Michael A. Moskowitz, MD, professor of neurology at Harvard University, here in a lecture at the fourth annual Headache Research Summit in February.

Aristides Leão first studied cortical spreading depression (CSD) more than 60 years ago, he explained, but only in the past few years have investigators begun to tease out its role in migraine. In fact, many researchers originally doubted whether CSD occurred in humans, but now it is known to be “a fundamental property of electrically excitable tissue” in mammals of all species, said Dr. Moskowitz. “There are compelling data in subarachnoid hemorrhage and trauma, and we expect to find it in stroke, as well.” (Brain 2006;129(Pt3):778–790); (Adv Tech Stand Neurosurg 2005;30:3–49).

CSD involves a slow wave of propagation or discharge of neurons and glia across the surface of the brain, traveling at about 3 to 5 millimeters per minute, he explained. Brain imaging advances, especially functional MRI (fMRI), support the notion that CSD plays a role in human migraine, Dr. Moskowitz told Neurology Today in a telephone interview. “With fMRI you can really see it from beginning to end: there's nothing we see in humans that is discrepant from [our findings in] rats.”

What's more, research in his laboratory has shown that CSD is probably the final common pathway for prophylactic drugs that have different mechanisms of action (Ann Neurol 2006;59:652–661). This knowledge should help in developing agents that work even better than current products, Dr. Moskowitz said at the conference.



Dr. Moskowitz started this research more than 20 years ago, when he and his colleagues demonstrated that trigeminal nerve fibers innervated portions of the circle of Willis and meningeal blood vessels (Science 1981;213:228–230; Ann Neurol 1984;16:157–168). They began investigating the trigeminal triggers of migraine, and ultimately showed that CSD could produce vasodilation of the middle meningeal artery for as long as 40 minutes and that the effect depended on trigeminal and parasympathetic activity. Cutting the trigeminal pathway stopped the vasodilation.

“This mechanism explains vasodilation during headache and links intense neurometabolic brain activity with the transmission of headache pain by the trigeminal nerve,” they wrote (Nat Med 2002;8:136–142). These findings demonstrate that “CSD is noxious and sufficient to activate the trigeminal vascular system,” Dr. Moskowitz said in his lecture.

Back to Top | Article Outline


From these and similar studies, he and colleagues hypothesized that migraine prophylactic drugs worked by raising the patient's threshold to develop a CSD. They designed studies to test the effects of various agents on neural activity (Ann Neurol 2006;59:652–661). The work went slowly because it involved giving individual drugs to animals seven days a week for up to four months. The doses were based on published data on other animal models of neurological and psychiatric disease such as epilepsy, pain, and depression. They sacrificed a few animals each week and studied the effects of each drug on CSD susceptibility, evoked either by applying 1M KCl directly to the pial surface, or with an electrical current.

Topiramate was the first test drug. “We saw a slowing of the propagated wave at the third week, and it became more pronounced in the fourth and fifth weeks,” Dr. Moskowitz recalled at the conference. Valproate, amitriptyline, propranolol, and methysergide all yielded similar findings, usually within six- to eight weeks. In general, chronic treatment reduced CSD-frequency from 16 CSDs in vehicle-treated controls to eight CSDs recorded in two hours. Single doses of the drugs had no effect on CSD.

To confirm that the drugs were indeed raising the CSD threshold, the team compared the threshold required to evoke a CSD after chronic drug treatment to that of control animals. The median ranged from 350 to 1,000 microcoulombs (μC) for the drug-treated animals, compared to a median of 10 μC for the controls. These differences also were statistically significant.

These agents have different chemical properties, so they must act on different receptors or pathways, all with the common result of inhibiting susceptibility to CSD (Ann Neurol 2006;59:652–661), Dr. Moskowitz pointed out. “The need for chronic treatment suggests that these drugs may lead to long-term modulation of gene expression or their encoded proteins,” he and his coauthors wrote.

The longer the treatment lasted, the more it raised CSD threshold, Dr. Moskowitz said.

Back to Top | Article Outline


“These findings are powerful because the investigators used both positive and negative controls,” said Nabih Ramadan, MD, professor and chair of neurology at the Rosalind Franklin University in Chicago. “In addition to using a racemic, DL mixture of propranolol, they also used the D-isomer, which is known not to produce a response to migraines.”

Dr. Ramadan, who co-chaired the session where Dr. Moskowitz spoke, told Neurology Today he thinks more compounds should be tested in this fashion to confirm its value as a screen for migraine prophylactic drugs.

“Except for this, we currently have no preclinical model that helps us characterize compounds for migraine prevention.”



Back to Top | Article Outline


Dr. Moskowitz also described his laboratory's research on genetics and sex hormones in migraine susceptibility. Familial hemiplegic migraine (FHM) in humans involves mutations in ion pumps and channels in the brain. People with FHM appear normal between attacks, but have a heightened sensitivity to migraine triggers such as stress, alcohol, exercise, and some types of head injuries. Two-thirds of these patients develop migraines with severe and prolonged visual and somatosensory auras. Hemiplegia is also part of the syndrome and is “probably caused by CSD,” Dr. Moskowitz said.

Since migraine occurs more often in women than in men, he and his associates worked with “knock-in mice,” which had the gene for FHM incorporated into their genome, to examine the effect of sex hormones on CSD threshold.

In research that is not yet published, they ovariectomized one group of female mice, and allowed another group of females to live to about 13 months of age — in other words, mouse menopause. Prior to ovariectomy or aging, CSDs occurred in the female knock-in mice at a rate of 21 per hour, significantly higher than the 13.5 per hour in the male knock-in mice, and the 9-per hour in male wild-type mice and 9.5-per hour in female wild-type mice.

After gonadectomy, the CSD rate in the female knock-in mice dropped to the level recorded in the knock-in males, which is still higher than in the wild-type mice, but significantly lower than the pre-surgical levels. Postmenopausal mice had a comparable rate to that recorded in ovariectomized females (13 CSDs per hour).

This research “clearly shows the ovarian hormones are playing a critical role in changing the electrical properties of the brain,” Dr. Moskowitz said in the interview.

Back to Top | Article Outline


“No one knows the source of migraine pain,” he told Neurology Today. “Is there some linear relationship going from CSD to the headache? Those studies are yet to be done.” That research would involve placing electrodes directly into meningeal fibers and showing that they are activated by CSD, he explained.

His laboratory is now examining the relationship between CSD and other neurological disorders. Each episode of CSD causes the meningeal vessels to swell with blood and a massive redistribution of ions across neural membranes. This activity is a significant insult to the brain, and may increase the patient's risk for stroke. Migraine is known to be a risk factor for stroke, Dr. Moskowitz said.

They are also studying other mutations that increase susceptibility to migraines, including changes along the gene that results in FHM. “There seems to be a principle here that we're scratching away at, which is that CSD susceptibility is common to different genotypes.”

Back to Top | Article Outline


• Leão AAP. Spreading depression of activity in cerebral cortex. J Neurophysiol 1944;7:359–390.
    • Bolay H et al. Cortical spreading is linked to pain transmission by the trigeminal nerve. J Neurol 2002;249:643–645.
      • Pietrobon D. Migraine: new molecular mechanisms. The Neuroscientist 2005;11:373–386.
        • Lauritzen M. Pathophysiology of the migraine aura. Brain 1994;117(1):199–210.
          • Fabricius M, et al. Cortical spreading depression and per-infarct depolarization in acutely injured human cerebral cortex. Brain 2006;129(Pt3):778–790.
            • Strong AJ, Dardis R. Depliarisation phenomena in traumatic and ischaemic brain injury. Adv Tech Stand Neurosurg 2005;30:3–49.
              • Ayata C, et al. Suppression of cortical spreading depression in migraine prophylaxis. Ann Neurol 2006;59:652–661.
                • Mayberg M, et al. Perivascular meningeal projections from cat trigeminal ganglia: possible pathway for vascular headaches in man. Science 1981;213:228–230.
                  • Moskowitz MA. The neurobiology of vascular head pain. Ann Neurol 1984;16(2):157–168.
                    • Bolay H, et al. Intrinsic brain activity triggers trigeminal meningeal afferents in a migraine model. Nat Med 2002;8(2):136–142.
                      © 2007 American Academy of Neurology