Migraine preventive drugs differentially affect cortical spreading depression in rat

Abstract

Cortical spreading depression (CSD) is the most likely cause of the migraine aura. Drugs with distinct pharmacological properties are effective in the preventive treatment of migraine. To test the hypothesis that their common denominator might be suppression of CSD we studied in rats the effect of three drugs used in migraine prevention: lamotrigine which is selectively effective on the aura but not on the headache, valproate and riboflavin which have a non-selective effect. Rats received for 4 weeks daily intraperitoneal injections of one of the three drugs. For valproate and riboflavin we used saline as control, for lamotrigine its vehicle dimethyl sulfoxide. After treatment, cortical spreading depressions were elicited for 2 h by occipital KCl application. We measured CSD frequency, its propagation between a posterior (parieto-occipital) and an anterior (frontal) electrode, and number of Fos-immunoreactive nuclei in frontal cortex. Lamotrigine suppressed CSDs by 37% and 60% at posterior and anterior electrodes. Valproate had no effect on posterior CSDs, but reduced anterior ones by 32% and slowed propagation velocity. Riboflavin had no significant effect at neither recording site. Frontal Fos expression was decreased after lamotrigine and valproate, but not after riboflavin. Serum levels of administered drugs were within the range of those usually effective in patients. Our study shows that preventive anti-migraine drugs have differential effects on CSD. Lamotrigine has a marked suppressive effect which correlates with its rather selective action on the migraine aura. Valproate and riboflavin have no effect on the triggering of CSD, although they are effective in migraine without aura. Taken together, these results are compatible with a causal role of CSD in migraine with aura, but not in migraine without aura.

Introduction

Migraine is the most frequent neurological disorder with a mean world-wide 1 year prevalence of 11.2% (Stovner et al., 2007) and ranks second after the dementias in global costs it causes to most developed societies (Andlin-Sobocki et al., 2005). In patients with frequent and/or disabling attacks, preventive treatments are recommended to reduce attack frequency and intensity (Silberstein, 2000). Most preventive anti-migraine drugs were found effective by serependity. Their therapeutic score does not exceed 50-60% and many have cumbersome adverse effects (Fumal and Schoenen, 2008). There is thus a need for better preventive treatments which implies a better understanding of the cellular and molecular mechanisms underlying the disorder as well as more precise knowledge on the pharmacological actions of anti-migraine drugs, including of presently available ones.

In 20% of cases (Russell and Olesen, 1996), the migraine headache is preceded by an aura of neurological symptoms, typically a scintillating scotoma, which develops gradually over several minutes and lasts for no more than 1 hour (Headache Classification Committee of The International Headache Society, 2004). There is strong evidence from clinical observations (Milner, 1958) and functional brain imaging studies (Hadjikhani et al., 2001, Lauritzen and Olesen, 1984) that the migraine aura is due to cortical spreading depression (CSD) originating in the occipital cortex. CSD is a slowly progressing wave (3-5 mm/min) of neurono-glial depolarization followed by a long-lasting suppression of neuronal activity and excitability (Leao, 1944). It can be elicited by a number of triggers including high concentrations of K⁺ and is considered a valid experimental model for the migraine aura (Bergerot et al., 2006). CSD in animals is able to activate the trigeminovascular nociceptive system thought to be responsible for the headache in migraineurs (Bolay et al., 2002, Moskowitz, 1984, Moskowitz et al., 1993).

Chronic administration of various established preventive anti-migraine drugs, including valproate, was found to suppresses CSD in rats (Ayata et al., 2006). This led to the hypothesis that CSD suppression may be a common mechanism explaining the therapeutic action of anti-migraine drugs and hence that CSD or CSD-like events play a causative role in all migraine types, including migraine without aura. This hypothesis cannot be taken without reservation for several reasons (Schoenen, 2006, Wolthausen et al., 2009). Besides the fact that there is no proof for the occurrence of CSD in migraine without aura, several drugs which are able to suppress CSD in the animal model, like phenytoin (Chebabo and Do Carmo, 1991), carbamazepine (Chen et al., 2005) and clonidine (Richter et al., 2005) have no effect in migraine (Evers et al., 2009). High dose magnesium (van der Hel et al., 1998) or fluoxetine (dos Santos et al., 2006) which also inhibit CSD in animal experiments, have little or no efficacy (Pfaffenrath et al., 1996); (Evers et al., 2009). The NMDA receptor antagonist ketamine is a proven blocker of CSD in animals (Gorelova et al., 1987). Concordantly, it stopped the neurological aura symptoms in some patients suffering from familial hemiplegic migraine, but it had no effect on the headache (Kaube et al., 2000). Several other drugs used in migraine prevention, such as lamotrigine and riboflavin, were not studied by Ayata et al. Lamotrigine, in particular, is of major interest. It induces distinct neurophysiological changes in the human brain compared to other anticonvulsants (Li et al., 2009, Smith et al., 2006). More importantly, there is evidence, albeit only from open label clinical trials and experience, that it is effective in migraine with aura (D'Andrea et al., 1999, Fumal and Schoenen, 2008, Lampl et al., 2005, Pascual et al., 2004) while it had no effect in migraine without aura in placebo-controlled trials (Pascual et al., 2004, Steiner et al., 1997). Lamotrigine would thus be expected to have a potent suppressive action on CSD. By contrast, riboflavin, a nutriceutical enhancing mitochondrial energy mechanism and proven effective at high doses for the prevention of adult migraine in one placebo-controlled trial (Schoenen et al., 1998) and one randomized open trial (Boehnke et al., 2004, Schoenen et al., 1998), does not modify cortical excitability as assessed by evoked potentials (Sandor et al., 2000) and would thus not be expected to have an effect on CSD.

We have explored the effect of chronic administrations of lamotrigine or riboflavin in the KCl-induced CSD model, using valproate as a positive control. As Fos expression is durably activated by CSD in the ipsilateral cortex and varies with therapeutic interventions (Herrera and Robertson, 1990, Moskowitz, 1993, Supornsilpchai et al., 2010), we have complemented the electrophysiological experiments with counts of Fos-immunoreactive nuclei in the frontal cortex to assess lasting neuronal changes.