Pathogenesis of Migraine – Role of Neuromodulators

The pathogenesis of migraine is still, today, a hotly debated issue. Recent biochemical studies report the occurrence in migraine of metabolic abnormalities in the synthesis of neurotransmitters and neuromodulators. These include a metabolic shift directing tyrosine metabolism toward the decarboxylation pathway, therein resulting in an unphysiological production of noradrenaline and dopamine along with increased synthesis of traces amines such as tyramine, octopamine, and synephrine. This biochemical alteration is possibly favored by impaired mitochondrial function and high levels of glutamate in the central nervous system (CNS) of migraine patients.

The unbalanced levels of the neurotransmitters (dopamine and noradrenaline) and neuromodulators (eg, tyramine, octopamine, and synephrine) in the synaptic dopaminergic and noradrenergic clefts of the pain matrix pathways may activate, downstream, the trigeminal system that releases calcitonin gene-related peptide. This induces the formation of an inflammatory soup, the sensitization of first trigeminal neuron, and the migraine attack. In view of this, we propose that migraine attacks derive from a top-down dysfunctional process that initiates in the frontal lobe in a hyperexcitable and hypoenergetic brain, thereafter progressing downstream resulting in abnormally activated nuclei of the pain matrix.


Migraine is a disabling condition characterized by unilateral headache pain, pulsating in quality and lasting 4–70 hours, accompanied by photo-, phono-, osmo-phobia, nausea, and vomiting. Aura may precede the migraine attacks in about 30% of patients and, in some patients, occurs as an isolated symptom.[1] The etiology of migraine is still not completely understood. This is because of the multiple complex symptomatology characteristic of migraine (headache attacks and psychiatric, neurologic, and sympathetic symptoms) and difficulty in unifying these characteristics into one or more interelated pathophysiological processes.

A pathophysiological hypothesis that may reconcile with the proteiform symptomatology of migraine has been proposed by Welch. According to this hypothesis, migraine is a multifactorial (ie, biological and psychological) biobehavioral disorder[2] in which the crisis is a response to stressful agents within an hyperexcitable brain.[3] Genetic mutations and/or polymorphisms of genes, yet to be determined, that regulate neuronal mitochondrial energy, neurotransmitter metabolism, and ion channels in the central nervous system (CNS) are considered the main biological factors.[4] Menstrual cycle, pregnancy, lifestyle, diet, anxiety, chronic stress, etc, are among the main psychological factors.[5] Once the migraine threshold is primed, the frequency of the attacks depends on the type of stress and anomalies in the metabolism of neurotransmitters and neuromodulators that regulate the synapses of cortical, antinociceptive (antinociceptic system [ANS]), and sympathetic neurons.[6]

In the CNS, glutamic acid and aspartic acid are the main excitatory neurotransmitters, whereas gamma aminobutyric acid (GABA) is the inhibitory neurtransmitter.[7] The balance between these 2 systems constitutes the frame in which the other circuitries regulate the functions of the human brain. It has been hypothesized that anomalies in the metabolism of glutamate and GABA, together with those that govern pain and vegetative functions, such as serotonin (5-HT), dopamine (DA), and noradrenaline (NE), constitute the phenotypical biochemical causes of the migraine.[8]

Recent evidence also supports the old notion that elusive amines, such as tyramine (tyr) and octopamine (Oct), play a role in migraine pathogensis.[9] These amines, together with DA and NE, are products of two different metabolic pathways of tyrosine. Tyrosine hydroxylase generates 3,4-dihydroxyphenylalanine (DOPA), DA, and NE, with the last 2 compounds requiring the action of DOPA decarboxylase and dopamine β-hydroxylase (Dβ-H) enzymes, respectively. The alternative pathway, tyrosine decarboxylase, synthetizes tyr, Oct, and synephrine, with Oct and synephrine requiring in addition Dβ-H and phenylethanolamine-N-methyltransferase (PNMT) enzyme activities[10] (see the Figure). Although the hypothesis that tyr and Oct may contribute to pathogenesis of migraine was proposed several decades ago,[11] the recent discovery of a new class of G-protein-coupled receptors with high affinity for these amines in rodents and humans has fuelled ever-increasing scientific interest. The trace amine receptors (TAARs) are found in various tissues and organs, including specific brain areas such as the rhinencephalon, limbic system, amygdala, hypothalamus, extrapyramidal system, and locus coeruleus.[12] This and other evidence has led to the suggestion that tyr and/or Oct behave as neurotransmitters and neuromodulators via TAARs and other receptors (eg, catecholaminergic receptors), respectively, contributing to physiology of noradrenergic and dopaminergic synaptic transmission in the ANS.[13]

We hereby summarize briefly the results, mainly generated from our laboratory, that support a role for biochemical alterations of different neurotransmitters and neuromodulators in the pathophysiology of migraine. Based on this evidence, we propose that future research efforts aiming to comprehend the pathophysiological relevance of neuromodulators, such as trace amines, in the CNS, have the potential to provide for new, more effective, treatment options for migraine.

Excitatory Amino Acids and Aura

The aura constitutes the clinical phenotype of migraine with aura. The hypothesis that aura is a clinical counterpart of a cortical phenomena derives from the observation of Lashley’s own visual phenomena.[14] The speed of the scotomata on the visual field, 3 mm/minute, and the fundamental studies of Olesen and his group on cerebral blood flow (CBF), in patients during their auras, have substantially demonstrated that the spreading depression (SD) of Leao is the neurophysiological cortical event of the aura symptomology. In these human experiments, the positive scotomata (visual scintillations and/or paresthesia) is concomitant to a brief increase in blood flow, and the negative symptoms to a reduced flow (oligemia) that propagates at the speed of 3 mm/minute on the occipital cortex.[15] Studies with functional magnetic resonance imaging (MRI), during spontaneous aura, have confirmed that the first phase of the aura is accompanied with a brief focal occipital increase in the brain oxygen-level dependent (BOLD) signal that propagates at the same velocity of SD on the occipital cortex, followed by a longer lasting decrease and impaired BOLD response to functional activation.[3] The clinical picture and the features derived from CBF and functional MRI (fMRI) studies suggest that the positive and negative signs of the aura may be caused, as in SD, by a burst of activity followed by suppression of neuronal cortical activity.

We hypothesized that neuronal hyperexcitability constitutes the functional prerequisite of the occurrence of the aura.[3] An array of biochemical, neurophysiological, and pharmacological data in humans support this hypothesis. Glutamate, released from neurons and glia, is the main excitatory neurotransmitter in the CNS. Anomalous levels of glutamic acid determine SD in animal experiments and ingestion of glutamate-rich food provokes, in predisposed subjects, migraine attacks.[16,17] To date, it is almost impossible to directly measure glutamate levels in the human brain. Platelets, however, have constituted a valid model to study glutamate metabolism because these cells display glutamate-related components similar to those of the neurons. A number of studies, conducted in the last 2 decades, have demonstrated that the levels of glutamic and aspartic acids are significantly more elevated in platelets, plasma, and CSF of migraine patients, particularly in those with aura.[18–20]

Occurrence of CNS hyperexicitability in migraine is also supported by neurophysiological studies employing transcranial magnetic stimulation (TMS). Stimulation with TMS of the occipital lobe in migraine with aura patients determines the appearance of a higher number of phosphenes in comparison with control and migraine without aura sufferers.[21] These same results were also found stimulating the motor areas and measuring the resting motor threshold and the silent period.[22] Although the authors suggest that these response abnormalities are due to a reduced inhibitory cortical tone,[23] the possibility that an increase in glutamate levels in the CNS plays a major role in pathogenesis of the aura is supported by [31]pNMR studies and pharmacological evidences employing lamotrigine. In comparison with control subjects, the levels of magnesium, measured with [31]pNMR, are significantly lower in the brain of migraine patients, particularly during the painful attacks.[24] Magnesium is a unique compound known to block glutamate-dependent SD.[25] Lamotrigine is an antiepileptic drug useful in the prevention of partial and generalized seizures as well. It acts by blocking voltage-sensitive channels leading to an inhibition of neuronal glutamate release.[26] Based on these pharmacodynamic characteristics, we conducted an open pilot study aiming to assess the effects of lamotrigine in migraine with aura. Twenty-five migraine patients with high frequency of aura attacks (at least 2 auras/month) were treated for 2 months. There was a dramatic reduction in the aura frequency and duration in the treated patients. Thereafter, a number of studies have confirmed the efficacy of lamotrigine in the prevention of auras.[27,28] The specificity of this drug for the prophylaxis of the migraine aura is stressed by the inefficacy of lamotrigine in the reduction of migraine without aura attacks[29] and the capacity of lamotrigine, in contrast to valproate, to block the SD-induced by potassium cloride (KCl) on the rat occipital cortex.[30]

Elusive Amines, Premonitory Symptoms, and Migraine Attack

The modalities by which stressful agents within the brain may cause the painful attacks in migraine is not known; however, according to the theory put forth by Welch, the first pathophysiological event may occur in the orbitofrontal part of the frontal lobe and, thereafter, downstream to the limbic, amygdale, and hypothalamic-connected areas of the CNS. Although still partly speculative, an increasing body of clinical, biochemical, and functional studies now support this theory. Migraine attacks are, very often, preceded by premonitory symptoms, such as hyperosmia, yawning, mood changes, anxiety, food craving, sexual excitement, fatigue, and emotional lability, which last from minutes to days.[31,32] These symptoms are considered markers of activation of the above-mentioned brain areas, and, therefore, it is logical to attribute the first phase of the migraine attacks within these areas.[33] Further support also derives from evidences showing that, in these same brain areas, TAARs and dopamine receptors are abundantly distributed. The activation of these receptors is likely reflected in the high levels of dopamine and elusive amines found in plasma and platelets of migraine without aura sufferers during headache-free periods.[34,35]

Catecholamines, Elusive Amines, and the Migraine Attack

After the premonitory symptoms in migraine, the painful phase occurs. One current hypothesis considers the head pain a consequence of trigeminal activation. This determines release of the neuropeptides, calcitonin gene-related peptide (CGRP) and substance P (SP), in the trigemino-vascular system.[36] Both peptides stimulate platelets, leukocytes, and endothelium to secrete an inflammatory soup (5-HT, adenosine diphosphate [ADP], platelet-activating factor, nitric oxide [NO], interleukins, etc) that sensitizes the first-order neuron of the trigeminal system and, after minutes or hours, the second, by an early gene-related process in the nucleus.[37]

The pathophysiological process that underlies trigeminal activation is a debated question. One hypothesis, derived from studies performed in animal models, suggests that the wave of the SD on the occipital cortex stimulates the nerve endings of the trigeminal system surrounding the pial vessels. This stimulation determines a trigeminal antidromic reflex resulting in the release of CGRP and SP in the dura mater head circulation. CGRP has been long suggested to be important in the occurrence of headache attacks because of its capability to interact with circulating cells and to determine neurogenic inflammation. CGRP stimulates platelets, white cells, and endothelium to synthesize NO. NO is considered a diffusible neurotransmitter in the CNS and, as such, may play a role in the release of glutamate and diffusion of SD on the cortex. In the circulation, it is a potent vasodilatator. NO, while dilatating the vessel wall, stretches the trigeminal endings innervating the wall, already sensitized by the inflammatory soup, and determines the migraine attack.[38]

Another hypothesis, and not mutually exclusive, is that the activation of the trigeminal system is a result of abnormal pain processing initiating in the frontal lobe and, thereafter, progressing downstream to the connected pain centers.[2,39] In support of this hypothesis is a functional fMRI study showing that inhibition of the orbito-frontal cortex, an important pain inhibitory cortex area, occurs in chronic migraine when the pain subsides.[40] Also, evidence from fMRI and PET studies have shown activation of the red nucleus, extrapyramidal system, and nuclei behind the aqueduct of the brain stem before and during migraine attacks,[41,42] all parts of the descending centers of the pain matrix. The modalities and the characteristics of the activation of these nuclei, however, are not known.

We hypothesized that abnormal levels of elusive amines and catecholaminergic neurotransmitters such as DA and NE, all products of tyrosine metabolism, play an important role in the pathophysiology of migraine attacks.[34] As mentioned previously, TAARs are located in the rhinencephalon, limbic system, amygdala, hypothalamus, extrapyramidal system, and locus coeruleus. These areas are important parts of the pain matrix that modulates the pain threshold.[39] The functions of the pain matrix neurons are mainly regulated by synapses that utilize DA and NE as neurotransmitters. Intriguingly, the highest levels of Oct are found in the same brain regions.[43] Oct acts, in the same synaptic clefts, as a neuromodulator. A neuromodulator is a chemical released from a neuron that causes no change in the excitability of the postsynaptic cells in the absence of a neurotransmitter. The released neuromodulator acts to modify the action (increase or decrease) of a coexisting neurotransmitter.[14] Thus, one possible physiologic role Oct is to regulate, together with other neurotransmitters, DA and NE synapses in the centers that regulate the pain threshold. As hypothesized by Welch, it is possible that, in the particular metabolic circumstances such as that associated with migraine, there occurs an abnormality in the metabolism of tyrosine toward increased synthesis of products of the decarboxylase pathway, resulting in increased synthesis of tyr, Oct, and synephrine in association with a decrease of NE. In support of this, we recently demonstrated that there occur higher levels of circulating Oct and synephrine along with increased levels of DA in migraine without aura (MwoA) patients, in comparison with healthy controls subjects.[44] The higher levels of DA are suggestive of a reduction in dopamine Dβ-H enzyme activity. Reduced Dβ-H enzyme activity[45] and reduced NE levels have been reported in migraine without aura patients.[46] Also, more recently, a polymorphism in the gene that encodes for Dβ-H protein has been identified.[47] All these results support the possibility that complex abnormalities in the metabolism of tyrosine-related pathways occurs in migraine patients, resulting in possible derangement in neurotransmitters and neuromodulators. If the same biochemical anomalies found in the circulation of migraine sufferers are present in the synaptic clefts innervating the pain matrix, an unphysiologic balance between neuromodulators (Oct and synephrine) and neurotransmitters (DA and NE) the intra-synaptic milieu should be expected. Possible pathological consequences include abnormal function of the hypothalamus,[33] the sympathetic system with related autonomic symptoms, reported in migraine patients (eg, orthostatic blood pressure changes, anomalies of pupillary control, and vertigo),[48] and the ANS nuclei, with downstream activation of the trigeminal nucleus, CGRP release in the encephalic circulation and head pain. Another possibility is that cortical SD may direct activate second-order trigeminovascular neurons, as recently suggested by Lambert et al[49] employing animal models. However, the biochemical pathway(s) involved in this process is (are) still unknown.

Biochemical Tryptophan Anomalies and Painful Attacks

The main product of tryptophan hydroxylase is serotonin (5-HT), whereas tryptamine is the neuromodulator that derives from the decarboxylase product of tryptophan. The involvement of 5-HT in migraine was hypothesized more than 50 years ago when F. Sicuteri demonstrated the occurrence of significantly elevated levels in urine of 5-hydroxyindoleacetic acid, stable metabolite of 5-HT, during migraine attacks.[50] Since this, numerous studies have attempted to clarify the possible biochemical anomalies of 5-HT in migraine, mainly utilizing platelets as a model of serotonergic neurons. Studies from our laboratory have shown that the levels of platelet 5-HT fluctuate in female migraine sufferers differently from those in healthy woman in the different phases of the menses and, more importantly, the levels of the indole decrease significantly in the luteal phase in menstrual migraine before the painful attacks.[51,52] The reason why the levels of 5-HT drop before the attack is not known. However, it is possible to conceive that there may occur a biochemical shift of tryptophan metabolism toward decarboxylation rather than hydroxylation, therein favoring an increase in the synthesis of tryptamine and a reduction in the synthesis 5-HT, respectively, in the synaptic clefts of neurons of the ANS nuclei of the brain stem.

GABA and Migraine

GABA is the main inhibitory neurotransmitter in the CNS. GABA plays an important role in the modulation of pain threshold. The antiepileptic drugs valproate and topiramate, the most efficacious drugs in preventing migraine without aura attacks, are potent GABAergic agonists.[53] Other than this, however, direct evidence for a role of GABA-related abnormalities in migraine is very scarce. One study has shown that plasma levels of GABA are not detectable during migraine attacks, but after this phase, its plasma levels increase, suggesting that activation of the GABAergic pathways is necessary to end the pain crisis.[54]

Mitochondrial Energy, Metabolic Shifts, and Migraine

An increasing number of studies suggest that migraine sufferers display a reduction in the metabolism of cellular energy in different tissues, including brain. The first line of evidence in support of this is the platelet anomalies. Platelets of migraineurs display elevated free intracytosolic calcium levels[55] and abnormal high number of dense bodies together with increased levels of serotonin within these organelles.[56] These abnormalities are accompanied by a reduction in dense body secretion when platelets are stimulated by collagen.[57] The reason for the accumulation of dense bodies and the impaired secretion in response to agonists may be due to a decrease of multiple enzymatic activities found in the mitochondria of platelets of MWoA and migraine with aura (MWA) patients.[58] The same mitochondrial energy defect(s) has (have) been demonstrated in brain and muscle in [31]pNMR spectroscopy studies in different types of headache patients.[59]

The synthesis of neurotransmitters is energy dependent. A shift of tyrosine metabolism leading to increased levels of trace amines in migraine may depend on energy defects. This hypothesis is supported by a study of these amines in CSF of early postmortem subjects. In the first hours after death, the levels of tyr, Oct, and synephrine increase dramatically suggesting that when the brain energy fails, the activity of tyrosine decarboxylase prevails.[60] Also, deposition of free radicals in brain stem structures of migraine patients may contribute to mitochondrial energy decline in these patients. The accumulation of iron ions is proportional to the frequency of attacks, being greatest in chronic migraine wherein the attacks may occur every day.[61] It is known that deposition of iron ions deteriorate the surface of mitochondria and reduce the efficiency of the respiratory chain and the neuronal energy substrates. Thus, it is probable that the deposition of iron radicals may progressively favor tyrosine metabolism along the decarboxylation pathway and deteriorate the neurotransmission of the pain matrix.


All of the above results provide for a functional framework in which anomalies in different biochemical pathways together act in determining migraine. Although many steps remain speculative, future biochemical studies should be focused on the study of the functional role of the TAARs alone and together with other neuromodulators, in particular, those affecting serotoninergic (eg, tryptamine) and dopaminergic (phenylethylamine and phenylethanolamine) neurotransmission, in the CNS of patients with migraine and, possibly, in adequately stressed animal models. Also, eventual effects of elusive amines on ion channels present on sensory neurons, the activation of which are associated with allodynia and hyperalgesia, as well as their interaction with TAARs should be questioned. Some genes recently implicated in migraine (eg, TRPM8 and KCNK18) are, intriguingly, predominantly expressed in trigeminal and dorsal root ganglia, a finding suggestive of an important role in the initiation of the headache attack.[62]

Another important point in need of clarification is the nature of the energy failure in migraine and the relationship between this failure and the metabolic alterations of the strategic amino acid parents of the dopaminergic and serotoninergic CNS circuitries. Studies on possible mutations or polymorphisms in genes that regulate the decarboxylase enzyme activity in brain are also warranted. These studies may shed light on the physiological and pathological significance of these ancient enzymes, of evolutionary importance, in humans. Oct and synephrine, in fact, are the main noradrenergic neurotransmitters found in the lowest species of the evolutionary scale, such as worms and insects.[14] It is possible that in conditions of neuronal energy failure, as has been demonstrated in migraine, the metabolism regresses, under the push of an excitatory status of the cortex,[3] into an archaic modality of neurotransmission, leading to modifications in the biochemical milieu of synaptic clefts of the pain matrix, the end result of which produces the migraine attacks.


  1. Launer LJ, Terwindt GM, Ferrari MD. The prevalence and characteristics of migraine in a population based cohort: The GEM study. Neurology. 1999;53: 537–542.
  2. Welch KMA. Migraine a biobehavioral disorder. Arch Neurol. 1987;44:323–327.
  3. Welch KMA, D’Andrea G, Tepley N, Barkley G, Ramadan NM. The concept of migraine as a state of central neuronal hyperexcitability. Neurol Clin. 1990;8:817–828.
  4. Migraine P. Migraine:A genetic disease? Neurol Sci. 2008;29:S47-S51.
  5. Ziegler DK, Paolo AM. Headache symptoms and psychological profile of headache-prone individuals: A comparison of clinic patients and controls. Arch Neurol. 1995;52:602–606.
  6. D’Andrea G, Perini F, Terrazzino S, Nordera GP. Contributions of biochemistry to the pathogenesis of primary headaches. Neurol Sci. 2004;3(Suppl 3): 589–592.
  7. Millan MJ. Descending control of pain. Prog Neurobiol. 2002;66:355–474.
  8. D’Andrea G, Nordera GP, Perini F, Allais G, Granella F. Biochemistry of neuromodulation in primary headaches: Focus on tyrosine metabolism. Neurol Sci. 2007;28:S94-S96.
  9. D’Andrea G, Terrazzino S, Fortin D, Cocco P, Balbi T. Elusive amines and primary Headaches: Historical background and prospectives. Neurol Sci. 2003; 24:S65-S67.
  10. Axelrod J, Saavedra JM. Octopamine. Nature. 1977;265:501–504.
  11. Sever PS. False transmitters and migraine. Lancet. 1979;1:333.
  12. Borowsky B, Adham N, Jones KA, et al. Trace amines: Identification of a family of mammalian G protein-coupled receptors. Proc Natl Acad Sci U S A. 2001;98:933–941.
  13. Berry DB. Mammalian central nervous system trace amines. Pharmacologic amphetamines, physiologic neuromodulators. J Neurochem. 2004;90:257–271.
  14. Lashley KS. Patterns of cerebral integration indicated by the scotomatas of migraine. Arch Neurol Psychiatry. 1941;46:331–339.
  15. Lauritzen M, Skyhoj E, Olesen J, et al. The role of spreading depression in acute brain disorders. Ann Neurol. 1981;4:569–572.
  16. Van Gelder N. Calcium mobility and release associated with EEG abnormalities, migraine and epilepsy. In: Lugaresi AF, ed. Migraine and Epilepsy. Boston: Butterworths; 1987:367–378.
  17. Freeman M. Reconsidering the effect of monosodium glutamate: A literature review. J Am Acad Nurse Pract. 2006;18:482–486.
  18. D’Andrea G, Cananzi AR, Joseph R, et al. Platelet glycine, glutamate, and aspartate in primary headache. Cephalalgia. 1991;11:197–200.
  19. Ferrari MD, Odink J, Bos KD, Maleny MS, Bruin GW. Neuroexcitatory plasma amino acids are elevated in migraine. Neurology. 1990;40:1582–1586.
  20. Peres MF, Zukermann E, Senne Soares CA, Alonso EO, Santos BF, Faulhaber MH. Cerebrospinal fluid glutamate levels in chronic migraine. Cephalalgia. 2004;24:735–739.
  21. Aurora SK, Ahmad BK, Welch KM, Bhardhwaj P, Ramadan NM. Transcranial magnetic stimulation confirms hyperexcitability of occipital cortex in migraine. Neurology. 1998;50:1111–1114.
  22. Khedr EM, Ahmed MA, Mohamed KA. Motor and visual cortical excitability in migraineurs patients with and without aura:Transcranial magnetic stimulation. Neurophysiol Clin. 2006;36:13–18.
  23. Schoenen J. Neurophysiological features of the migraineous brain. Neurol Sci. 2006;27:S77-S81.
  24. Ramadan NM, Halvorson H, Vande-Linde A, Levine SR, Helpern JA,Welch KM. Low brain magnesium and migraine. Headache. 1989;9:416–419.
  25. Van Harreveld A. The nature of chick’s magnesium sensitive retinal spreading depression. J Neurobiol. 1984;15:333–334.
  26. Leach MJ, Lees G, Riddall DR. Lamotrigine. Mechanisms of action. In: Levy RH, Mattson RH, Meldrum BS, eds. Antiepileptic Drugs, 4th edn. New York: Raven Press; 1995:851–859.
  27. D’Andrea G, Granella F, Cadaldini M, Manzoni GC. Effectiveness of lamotrigine in the prophylaxis. Cephalalgia. 1999;19:64–66.
  28. Lampl C, Katsarava Z, Diener HC, Limmroth V. Lamotrigine reduces migraine aura and migraine attacks in patients with migraine with aura. J Neurol Neurosurg Psychiatry. 2005;76:1730–1732.
  29. Steiner TJ, Findley LJ, Yuen AW. Lamotrigine versus placebo in the prophylaxis of migraine with and without aura. Cephalalgia. 1997;17:109–112.
  30. Bogdanov VB, Multon S, Chauvel V, Bogdanova OV, Makarchuk MY, Schoenen J. Cortical spreading depression and associated neuronal Fos expression in rats are affected differentially by chronic treatment with lamotrigine, valproate or riboflavin. Cephalalgia. 2009;29:10.
  31. Griffin NJ, Ruggiero L, Lipton RB, Silberstein SD, et al. Premonitory symptoms in migraine: An electronic diary study. Neurology. 2003;25:935–940.
  32. Quintela E, Castillo J, Mnoz P, Pascual J. Premonitory and resolution symptoms in migraine: A prospective study in 100 unselected patients. Cephalalgia. 2006;26:1051–1060.
  33. Denuelle M, Fabre N, Payoux P, Chollet F, Geroud G. Hypothalamic activation in spontaneous migraine attacks. Headache. 2007;47:1418–1426.
  34. D’Andrea G, Terrazzino S, Leon A, et al. Elevated levels of circulating trace amines in primary headaches. Neurology. 2004;62:1701–1705.
  35. D’Andrea G, Granella F, Leone M, Perini F, Farruggio A, Bussone G. Abnormal platelet trace amine profiles in migraine with and without aura. Cephalalgia. 2006;26:968–972.
  36. Villalon CM, Olesen J. The role of CGRP in the pathophysiology of migraine and efficacy of CGRP receptor antagonists as acute antimigraine drugs. Pharmacol Ther. 2009;124:309–323.
  37. Perini F, D’Andrea G, Galloni E, et al. Plasma cytokine levels in migraineurs and controls. Headache. 2005;5:926–931.
  38. Bolay N, Reuter U, Dunn AK, Huang Z, Boas DA, Moskowitz MA. Intrinsic brain activity triggers trigemional meningeal afferent in migraine model. Nat Med. 2002;8:136–142.
  39. Moulton EA, Burstein R, Tully S, Hargreaves R, Becerra L, Borsok D. Interictal dysfunction descending modulatory center in migraine patients. PLoS ONE. 2008;3:e3799.
  40. Fumal A, Laureys S, Di Clemente L, et al. Orbitofrontal cortex involvement in chronic analgesic-overuse headache evolving from episodic migraine. Brain. 2006;129:543–550.
  41. Welch KMA,Cao Y,Aurora SK,Wiggins G,Vikingstad EM. MRI of the occipital cortex, red nucleus and substantia nigra during visual aura of migraine. Neurology. 1998;51:1465–1469.
  42. Weiller C, May A, Limmroth V, et al. Brain stem activation in spontaneous human migraine attacks. Nat Med. 1995;1:658–660.
  43. Danielson TJ, Boulton AA, Robertson HS. Moctopamine, p-octopamine and phenylethylamine in mammalian brain:A sensitive specific assay and the effect of drugs. J Neurochem. 1977;29:1131–1135.
  44. D’Andrea G, Granella F, Perini F, Farruggio A, Leone M, Bussone G. Platelet levels of dopamine are increased in migraine and cluster headache. Headache. 2006;46:585–591.
  45. Gallai V, Gaiti A, Sarchielli P, Coata G, Trequattrini A, Paciaroni M. Evidence for an altered dopamine β-hydroxylase activity in migraine and tension type headache. Acta Neurol Scand. 1992;86:403–446.
  46. Martignoni F, Blandini F, D’Andrea G, et al. Platelet and plasma catecholamines in migraine patients. Evidences of menstrual-related variability of the noradrenergic tone. Biogenic Amines. 1990;10:227–237.
  47. Fernandez F, Lea RA, Colson NJ, Bellis C, Quinlan S, Griffits LR. Association between a 19 bp deletion polymorphism at dopamine betahydroxylase (DBH) locus and migraine with aura. J Neurol Sci. 2006;251:118–123.
  48. Peroutka JP. Migraine: A chronic sympathetic nervous system disorder. Headache. 2004;44:53–64.
  49. Lambert GA,Truong L, Zagami A. Effect of cortical spreading depression on basal and evoked traffic in the trigeminovascular sensory system. Cephalalgia. 2011;31:1439–1451.
  50. Sicuteri F, Testi A, Anselmi B. Biochemical investigations in headache: Increase of hydroyindoleacetic acid excretion during migraine attack. Int Arch Allergy. 1961;19:265–271.
  51. D’Andrea G, Hasselmark L, Cananzi AR, et al. Metabolism and menstrual cycle rhythmicity of serotonin in primary headaches. Headache. 1995;35: 216–221.
  52. Fioroni L, D’Andrea G, Alecci M, Cananzi AR, Facchinetti F. Platelet serotonin pathway in menstrual migraine. Cephalalgia. 1996;16:427–430.
  53. Casucci G,Villani V, Frediani F. Central mechanism of action of antimigraine prophylactic drugs. Neurol Sci. 2008;29(Suppl. 1):S123-S126.
  54. Welch KMA, Chabi E, Nell JH, Bartosh K, Achar VS, Meyer JS. Cerebrospinal fluid gamma aminobutyric acid levels and migraine. Br Med J. 1975;3:516–517.
  55. Joseph R, D’Andrea G, Grunfeld S, Welch KMA. Cytosolic ionized calcium homeostasis in platelet: An abnormal sensitivity to PAF-activation in migraine. Headache. 1988;28:396–402.
  56. D’Andrea G, Welch KMA, Riddle JM, Grunfeld S, Joseph R. Platelet serotonin metabolism and ultra-structure in migraine. Arch Neurol. 1989;46:1187–1189.
  57. Joseph R,Welch KMA,Levine S, D’Andrea G.ATP hyposecretion from platelet dense bodies-evidence for the purinergic hypothesis and a marker of migraine. Headache. 1986;26:403–410.
  58. Sangiorgi S, Mochi M, Riva R, et al. Abnormal platelet mitochondrial function in patients affected by migraine with and without aura. Cephalalgia. 1994;14:31–23.
  59. Welch KMA, Ramadan MN. Mitochondria, magnesium and migraine. J Neurol Sci. 1995;134:9–14.
  60. Balbi T, Fusco M,Vasapollo D, et al.The presence of trace amines in postmortem cerebrospinal fluid in humans. J Forensic Sci. 2005;50:630–632.
  61. Welch KMA. Iron in the migraine brain:A resilient hypothesis. Cephalalgia. 2009;29:283–285.
  62. Su L,Wang C, Ren YY, Xie KL,Wang GL. Role of TRPM8 in dorsal root ganglion in nerve injury-induced chronic pain. BMC Neurosci. 2011;12:120–135.