120

CURRENT LITERATURE

CURRENT PREVENTIVE THERAPY: WHAT WE KNOW ABOUT

MECHANISM(S) OF ACTION?

Migraine Prophylactic Drugs Work via Ion Channels

Cohen GK

Medical Hypotheses 2005;65:114–122

In recent decades, the concept of a vascular origin of mi-

graine has been replaced by theories based on neuronal

pathophysiology. These theories involve rapid changes

in the functioning of the brain, particularly the brain-

stem, and the trigeminal nerves. While such paroxysmal

changes in function could be the result of altered synaptic

transmission, or other physiological changes, the author

hypothesizes that they could also be due to changes in

the function of voltage-regulated sodium and calcium ion

channels. Support for this view of migraine as a chan-

nelopathy comes from an examination of the likely mech-

anism of action of migraine prophylactic drugs, as most

of the widely used drugs for migraine prevention work by

inhibiting the function of one or both of these ion chan-

nels. This review of the laboratory research done on most

of the commonly used migraine prophylactic drugs, di-

vided into five classes, reveals that they all may work on

sodium channels, calcium channels, or both. If this is the

common mechanism of action of migraine prophylactics,

it should lead toward the development of more effective

prophylactic drugs.

COMMENTARY

T he efficacy of some β-blockers, flunarizine, some

antiepileptics, and amitriptyline in migraine prevention

has been demonstrated, both in controlled clinical trials and

in clinical practice. Their mechanisms of action, however, are

unknown. Curiously, none of these medications were devel-

oped specifically for the prophylaxis of migraine, but for the

treatment of other diseases. Recent genetic discoveries in the

familial hemiplegic variant of migraine, together with many

of its clinical characteristics, support the hypothesis that mi-

graine should be considered as one channelopathy. A gain of

function of sodium and calcium channels has been proposed

as the abnormality responsible. These two channels have much

in common, as they may be evolved from a common gene.

In this complete and highly recommendable paper, the au-

thor proposes the idea that migraine preventatives have as their

mechanism of action an inhibition of voltage-regulated sodium

and/or calcium channels and reviews the available experimental

evidence in this regard. Beginning with antidepressants, all tri-

cyclics block sodium channels, but with varying potencies. In

fact, amitriptyline is a very effective blocker, while imipramine

and other tricyclics are less effective. SSRIs were developed with

the idea of saving some of the adverse events of tricyclics by

avoiding any blockage of sodium channels. Interestingly, fluox-

etine, the only SSRI with some efficacy in migraine prevention,

appears to be the only compound in this group showing any in-

hibition on sodium channels. Finally, there is limited evidence

that tricyclics may also inhibit calcium channels. With regard

to β-blockers, propranolol not only inhibits sodium channels

(membrane-stabilizing effect), but does so more potently than

other β-blockers. Propranolol has also shown, in slices of cor-

tical synaptosomes, calcium-channel blocking properties. Flu-

narizine, and in a lesser degree, verapamil are the most effec-

tive “calcium channel blockers” for migraine prophylaxis. The

two drugs work nonspecifically on different calcium channels.

They also block voltage-regulated sodium channels, which may

distinguish them from other members of the class, such as ni-

modipine, nonefficacious for migraine prevention. The same

is true for classical antiepileptics: like verapamil and flunar-

izine, valproic acid might be more potent in migraine as com-

pared to other older anticonvulsants because it blocks both

sodium and calcium channels. Among the new antiepileptics,

topiramate is the only one with demonstrated efficacy in mi-

graine prevention. Topiramate’s ability to block both sodium

and calcium channels is well demonstrated. Gabapentin, an-

other “new” antiepileptic with low-grade efficacy in migraine

prevention, blocks calcium channels but not sodium channels.

The author sensibly concludes that, even though these drugs

could also work in migraine via multiple mechanisms, it seems

more than a coincidence that most of the commonly used drugs

in migraine prophylaxis act as blockers of sodium or calcium

channels, or both. It seems logical to propose that inhibition of

sodium channels would be especially useful in a migraine pro-

phylactic, as it is the ability to block both sodium and calcium

channels.

Clinical Science 121

The Role of Dopamine in a Model of Trigeminovascular

Nociception

Akerman S, Goadsby PJ

J Pharmacol Exp Therapeutics 2005;314:162–169

In animal studies, dopamine has been found to cause

vasodilation of cranial arteries at very low doses. Using

a model of intravital microscopy, the authors examined

the effect of dopamine receptor agonists on dural blood

vessel calibre and the effect of dopamine on trigemi-

novascular neurogenic dural vasodilation. Dopamine hy-

drochloride caused a significant vasoconstriction (p <

0.05) and increase in arterial blood pressure (p < 0.05) that

was reversed by an α2-adrenoceptor antagonist, yohim-

bine, rather than specific dopamine receptor antagonists.

The D1 receptor agonist caused vasoconstriction (p <

0.05) and blood pressure increase (p < 0.05), which were

reversed by yohimbine and therefore α2-adrenoceptor-

mediated. None of the specific dopamine receptor antag-

onists were able to attenuate neurogenic dural vasodila-

tion. Dopamine hydrochloride infusion (p < 0.05) and a D1

receptor agonist were able to attenuate vasodilation (p <

0.05), which maximal dilation returning after cessation of

the dopamine agonist infusion. The authors conclude that

this response may be due to the vasoconstrictor effects

of the α2-adrenoceptor and an action at the D1 receptor.

In addition, in the intravital model of trigeminal activa-

tion, it seems that dopamine receptors do not play a ma-

jor role and may not present an acute treatment option.

Their data, however, do not exclude a role for dopamine

receptor modulators in short- or long-term prevention.

COMMENTARY

D opaminergic mechanisms seem to be involved in mi-

graine pathophysiology, especially in its premonitory

phase, which is characterized by symptoms such as nausea, vom-

iting, drowsiness, tiredness, yawning, or mood changes. Given

that these changes may be a result of monoamine, and specifi-

cally dopaminergic neurotransmission, dopamine has been im-

plicated in migraine. Migraine patients seem to show hyper-

sensitivity to dopamine agonists and preliminary studies have

shown domperidone can prevent the occurrence of migraine if

taken during the premonitory phase. These observations have

led to a dopamine theory of migraine.

To examine whether dopamine is involved in the patho-

physiology of the headache phase of migraine, this work tests the

effect of dopamine on the dural vasculature. The authors make

use of the intravital microscopy model of trigeminovascular ac-

tivation, which has proven to be an excellent model in predict-

ing antimigraine efficacy. Dopamine receptors are present in the

trigeminovascular system. The infusion of dopamine attenuated

neurogenic dural vasodilation, possibly due to its vasoconstric-

tive effects. This attenuation was also seen with A68930, a D1

receptor agonist, but not with a D2 agonist and was partially

reversed by the D1 receptor antagonist SCH-23390. The data

obtained by Akerman and Goadsby support the fact that D1

receptors may be involved in the control of the central vascu-

lature on trigeminal nerve endings. All the effects of dopamine

and the D1 agonist were clearly antagonized by yohimbine,

an α2-adrenoceptor antagonist. The same authors, however,

have previously shown that the α2-adrenoceptor is not involved

in neurogenic vasodilation. Significant blood pressure changes

were caused by dopamine receptor antagonists, and these were

accompanied by a change in dural vessel diameter. Dopamine

acts as a precursor to the catecholamines noradrenaline and

adrenaline, which mediate vasoconstriction and blood pressure

increase via the α1- and α2-adrenoceptors. It is possible that the

effect of dopamine antagonists on blood pressure is a response

to inhibition of the precursor to adrenergic synthesis, namely

dopamine and that the dural blood vessel diameter changes are

a response to the change in blood pressure.

What would the theoretical implications of these findings

for migraine pathophysiology and treatment be? As only the D1

receptor was able to slightly tone down the activation of dural

blood vessels or trigeminal neurons, the role of dopamine and

D1 receptors in the acute phase of migraine seems small. There-

fore the involvement of dopamine and its receptors in migraine

may dominate other aspects of the attack, such as the initia-

tion, this providing a potential role for dopamine modulation

in prevention.

Noradrenergic Agonists and Antagonists Influence Mi-

gration of Cortical Spreading Depression in Rat—a Pos-

sible Mechanism of Migraine Prophylaxis and Preven-

tion of Postischemic Neuronal Damage

Richter F, Milulik O, Ebersberger A, Schaible HG

J Cereb Blood Flow Metab 2005;25:1225–1235

The authors analyze the effect of adrenergic agonists and

antagonist on cortical spreading depression (CSD). They

applied different drugs topically to an area of the exposed

cortex to anesthetized adult rats and observed the mi-

gration of CSD-related DC potential deflections across

the treated area. The adrenergic agonist norepinephrine

(1 mmol/L) and the α2-agonist clonidine (0.56 mmol/L)

blocked reversibly the migration of CSD. The β-blocker

122 Clinical Science

propranolol dose dependently diminished migration ve-

locity or even blocked CSD migration. The CSD blockade

by the α2-antagonist yohimbine (1.75 mmol/L) was be-

cause of its action on inhibitory 5-HT1A receptors. None

of the substances in the concentrations used had influ-

ence on regional cerebral blood flow or systemic arterial

blood pressure. The data suggest that the interference

of these compounds with CSD may contribute to their

beneficial therapeutic effect. The effect of β-receptor an-

tagonists in human migraine needs further exploration,

because these drugs also work in migraine without aura.

COMMENTARY

S urprisingly, both (α and β) adrenergic agonists and an-

tagonists have shown efficacy in migraine prevention. For

instance, tizanidine (an α2-adrenoceptor agonist), clonidine (an

α2-agonist) and mainly β-blockers have been reported as suc-

cessful preventatives. How these drugs, including β-blockers,

exert their antimigraine actions is not known. In this paper,

Ritcher et al. test the effect of adrenergic drugs on a CSD model

in rats. Even though the role of CSD in the pathophysiology

of migraine without aura is debatable, CSD epiphenomenon is

crucial in the pathophysiology of migraine aura.

The authors elegantly showed that CSD can be inhibited by

the topical application of norepinephrine, the α2-adrenoceptor

agonist clonidine, the α2-adrenergic antagonist yohimbine and

propranolol, a β-adrenergic receptor antagonist, which suggests

a role of both α2 and β-adrenoceptors in CSD propagation.

How can drugs with opposite actions in the same receptors have

identical final effects in CSD? The inhibition of CSD induced

by noradrenaline and clonidine is probably due to an action on

presynaptic neuronal α2-adrenoceptors. These receptors inhibit

adenylate-cyclase via G-proteins, which in turn inhibits calcium

currents necessary for the release of glutamate. Inhibition of glu-

tamate release could thus explain why α2-adrenoceptor agonists

reverse CSD. The authors of this paper show that the inhibi-

tion of α2-adrenoceptor antagonists on CSD is mainly because

of their agonistic action at the inhibitory 5-HT1A receptors,

which are known to regulate cAMP levels and glutamate release.

Finally, the inhibition of CSD propagation by propranolol is

very likely due to a reduction of glutamate release via an an-

tagonism of adenylate cyclase that activates P/Q-type calcium

channels. These results suggest that CSD inhibition can be a

novel explanation for the preventive effects of β-blockers and

also of adrenergic agonists. These findings open new approaches

in the preventive treatment of migraine with future drugs reg-

ulating presynaptic cAMP concentration and its consequence,

glutamate release.

Cortical Spreading Depression Activates and Upregu-

lates MMP-9

Gursoy-Ozdemir Y, Chu J, Matsuoka N, Bolay H,

Bermpohl D, Jin H, Wang X, Rosenberg GA, Lo EH,

Moskowitz MA

J Clin Invest 2004;113:1447–1455

In this study, Gursoy-Ozdemir et al. demonstrate that

CSD alters blood-brain barrier (BBB) permeability by

activating brain metalloproteases (MMPs). Beginning at

3–6 hours, MMP-9 levels increased within cortex ipsi-

lateral to the CSD, reaching a maximum at 24 hours

and persisting for at least 48 hours. Gelatinolytic activity

was detected earliest within the matrix of cortical blood

vessels and later within neurons and pia-arachnoid (>3

hours), particularly within piriform cortex; this activity was

suppressed by injection of the metalloprotease inhibitor

GM6001 or in vitro by the addition of a zinc chelator (1,10-

phenanthroline). At 3–24 hours, immunoreactive laminin,

endothelial barrier antigen, and zona occludens-1 dimin-

ished in the ipsilateral cortex, suggesting that CSD al-

tered proteins critical to the integrity of BBB. At 3 hours

after CSD, plasma protein leakage and brain edema de-

veloped contemporaneously. Albumin leakage was sup-

pressed by the administration of GM6001. Protein leak-

age was not detected in MMP-9-null mice, implicating

the MMP-9 isoform in barrier disruption. They conclude

that intense neuronal and glial depolarization initiates a

cascade that disrupts the BBB via an MMP-9-dependent

mechanism.

COMMENTARY

I n this excellent and extensive work, Gursoy-Ozdemir et al.

study the relationship between CSD and a metalloprotease

protein, MMP-9. MMPs are important for BBB breakdown

and for edema formation and leakage in several neurological

disorders, such as stroke, multiple sclerosis or Guillain-Barre.

MMPs are able to disrupt the BBB, which comprises endothe-

lial tight junctions, astrocytic end feet and basal lamina. The

basal lamina contains extracellular matrix collagen, laminin and

fibronectin, which are substrates for MMPs, namely MMP-2

and MMP-9.

These authors were able to detect MMP-9 activation at as

early as 15–30 minutes, coincidental with changes in vascular

permeability. This MMP-9 activation was long-lasting, still ob-

served 48 hours after CSD induction. At 3 hours most MMP-9

activity was localized within neurons. Even though the method

Clinical Science 123

of activation of MMP-9 needs to be studied further, oxygen

radicals, nitric oxide and proteases, such as stromelysin-1 and

plasmin, have been implicated in MMP activation. CSD, in ad-

dition, stimulates molecules that bind to the promoter region of

MMP-9, that is, c-fos, tumor necrosis factor-α and interleukin-

1β. MMP activation cleaves a broad range of substrates, includ-

ing components of basal lamina, like laminin, type IV collagen,

and brain ECM components such as elastin, fibronectin, tenas-

tin, and collagen. Gursoy-Ozdemir et al.’s findings open a vari-

ety of new potential targets for the preventive treatment of mi-

graine. One obvious question is whether the MMP-9 increase

is a specific consequence of CSD (and not an epiphenomenon)

causing BBB dysfunction. By using several clever experimental

approaches, such as an MMP-9-null mice model, the authors

were able to demonstrate that BBB breakdown develops as a

true consequence of MMP-9 activation. It is clear that drugs

able to prevent MMP-9 upregulation are one of the future con-

ceptually new promises to test for the preventive treatment of

migraine.

by Julio Pascual, M.D., Ph.D.