Headache ISSN 0017-8748C© 2006 by American Headache Society doi: 10.1111/j.1526-4610.2006.00614.xPublished by Blackwell Publishing

Brief Communication

Endothelial Nitric Oxide Synthase (Glu298Asp)Polymorphism is an Independent Risk Factor

for Migraine with Aura

Barbara Borroni, MD; Renata Rao, MD; Paolo Liberini, MD; Elisabetta Venturelli, MD; MichelaCossandi, MD; Silvana Archetti, PhD; Luigi Caimi, PhD; Alessandro Padovani, MD, PhD

Objective.—The aim of the present study was to evaluate whether the functional endothelial nitric oxidesynthase (eNOS) Glu298Asp polymorphism, which has been demonstrated to decrease the endothelial NOS activity,might be a risk factor for migraine.

Background.—It has widely demonstrated that nitric oxide (NO) is involved in migraine pathogenesis. Severalgenetic risk factors have been associated with migraine, but no study has unraveled a possible relationship betweenmigraine and eNOS Glu298Asp.

Methods.—One hundred fifty-six migraine patients and 125 healthy nonheadache volunteers entered the study.Demographic and clinical characteristics were carefully recorded, and a neurological workup was performed.

Results.—eNOS AspAsp homozygous patients had a 3-fold time risk for migraine with aura (MA) when com-pared to migraine without aura (MO) patients (OR-3.02, 95% CI-1.21 to 7.51), and more than 2-fold time increasedrisk when compared to control subjects (OR-2.21, 95% CI-1.00 to 5.04).

In migraine patients, no difference in age at onset, mean attack’s intensity, family history for any of the studiedcomorbidities, or the presence of comorbidities was found in eNOS AspAsp homozygous compared to eNOSGluGlu or eNOS GluAs carriers.

Conclusions.—Homozygous Asp298, a common variant of the eNOS gene, is an independent risk factor forMA in this study population.

Key words: migraine, aura, polymorphism, nitric oxide synthase, nitric oxide

(Headache 2006;46:1575-1579)

Migraine is a debilitating disorder affecting ap-proximately 12% of Caucasian populations. It hasbeen largely reported that the disease has a large ge-

From the Department of Medical Sciences, NeurologicalClinic, University of Brescia, Brescia, Italy (Drs. Borroni, Rao,Liberini, Venturelli, Cossandi, and Padovani) and Departmentof Biotechnology, III Laboratory of Analysis, University ofBrescia, Brescia, Italy (Drs. Archetti and Caimi).

Address all correspondence to Dr. Barbara Borroni, MD, Neu-rologic Clinic, University of Brescia, P.za Spedali Civili, 1, Bres-cia 25100, Italy.

Accepted for publication June 27, 2006.

netic component, although at present the type andnumber of genes involved is unclear.1,2 Candidate genestudies may be a useful strategy for identifying possibletherapeutic targets for migraine management.

Nitric oxide (NO), a potent endogenous vasodila-tor, is a key molecule affecting the pain associatedwith migraine, as NO causes an immediate headache inthose with migraine and less often in control subjectslikely by acting on the trigeminovascular system.3-5

NO is synthesized from L-arginine and molecularoxygen by a family of the NO synthase (NOS). In theendothelial cell, NO is synthesized by the endothelial

1575

1576 November/December 2006

NOS encoded by a 26 exon gene (NOS 3) located onchromosome 7.6

Several studies have speculated about migraineand NOS, suggesting that the latter could possibly bea cause, or a candidate gene, in migraine etiology, butin a previous study no association was found.7

Different common polymorphisms exist withinendothelial nitric oxide synthase (eNOS) gene.Among them, the substitution in nucleotide 894 (Gto T) results in the conversion of glutamate (Glu) toaspartate (Asp) for codon 298 and is a determinant ofthe decreased eNOS activity.8

The Asp298 variant has been implicated as a riskfactor for cardiovascular9,10 and cerebrovascular dis-eases.11 No data on the association between this poly-morphism and migraine are available.

This study was aimed to evaluate the role of NOSGlu298Asp genetic variation in migraine patients andin migraine subtypes.

PATIENTS AND METHODSOne hundred fifty-six consecutive migraine pa-

tients, and 125 nonheadache unrelated healthy vol-unteers were enrolled at the Headache Centre of theDepartment of Neurology, University of Brescia, andentered the study. All migraine patients and healthycontrols were interviewed by an experienced neurol-ogist. A standardized record of all demographic char-acteristics, the family history for migraine, cerebrovas-cular disease, cardiovascular disease, and neurologicaldisorders was obtained. The presence of comorbiditieswas also evaluated.

All subjects underwent a clinical and neurologicalworkup, and a blood drawing for eNOS genotyping.

The migraine patients were diagnosed as havingmigraine without aura (MO) or migraine with aura(MA) according to International Headache Society(IHS) criteria.12

The clinical features of migraine, that is disease in-tensity, and mean migraine attack duration, were care-fully recorded. Mean pain intensity during migraineattacks was scored as mild, moderate, or severe.

The study was conducted in accordance with localclinical research regulations and an informed consentwas required from all the subjects.

eNOS Glu298Asp genotyping.— Polymorphismanalyses were performed blinded to diagnosis and

genotype. Genomic DNA was extracted from sam-ples of whole blood, using a salting-out pro-cedure. Polymerase chain reaction (PCR) wasperformed using a forward primer 5′-CATGAGGCTCAGCCCCAGAAC-3′ and a reverse primer5′-AGTCAATCCCTTTGGTGCTCA-3′. PCR wasperformed as previously reported11 and genotyping ofthe eNOS single nucleotide polymorphism was doneby digesting the PCR products with MboI restrictionenzyme. In the presence of a T at nucleotide 894,which corresponds to Asp298, the 206bp PCR prod-uct is cleaved into 2 fragments of 119 and 87 bp.

Statistical Analysis.—To verify that our populationwas not selected for particular allele configuration,Hardy-Weinberg equilibrium was assessed with χ2 testanalysis. The differences in genotype frequencies andother risk factors among groups were analysed by theχ2 test. The strength of association between eNOSGlu298Asp polymorphism and migraine was quanti-fied as relative risks estimated with odds ratio, and95% confidence intervals were calculated from thestandard errors and coefficients of the logistic regres-sion. Demographic characteristics between the groupswere compared by Student’s t-test and χ2 test, as in-dicated. Results were expressed as mean ± standarddeviation (SD). The level of significance was taken atP < .05.

RESULTThe migraine patients and control subjects did not

differ for age (years mean ± SD, 31.4 ± 7.6 vs 32.7 ±9.2). As expected, in migraine patients compared tocontrols, female gender (78.8% vs 58.4, P = .001) andfamily history for migraine (76.8% vs 8.8%, P = .001)were more common.

No differences in either genotype distribution orallele frequency of Glu298Asp polymorphism was ob-served when controls and all migraine patients wereconsidered. Notwithstanding, AspAsp (Asp298 ho-mozygous) of the eNOS gene was more common inMA patients than in control subjects and in MO pa-tients (see Table 1).

As shown in Table 2, eNOS Asp298 homozygouspatients had a 3-fold time risk for MA when comparedto MO patients, and more than 2-fold time when com-pared to control subjects.

Headache 1577

Table 1.—Genotype Distribution and Allele Frequencies of eNOS Gene in Migraine Patients With and Without Aura and inControl Subjects

Groups Controls All Migraine MO MA P∗

Genotype distributionGluGlu 47.2 (59) 51.3 (80) 54.4 (56) 45.3 (24)GluAsp 40.0 (50) 34.0 (53) 35.9 (37) 30.2 (16)AspAsp 12.8 (16) 14.7 (23) 9.7 (10) 24.5 (13) .11

Allele frequencyGlu 67.2 (168) 68.3 (213) 72.3 (149) 60.4 (64)Asp 32.8 (82) 31.7 (99) 27.7 (57) 39.6 (42) .98

MO = migraine patients without aura; MA = migraine patients with aura.∗χ2 test among controls versus MO versus MA. Results are expressed as percentage (number of subjects between brackets).

Finally, migraine features were analyzed ac-cording to eNOS genotype. Migraine patients weregrouped as eNOS AspAsp homozygous vs eNOSGluGlu and GluAsp, and the demographic or clinicalcharacteristics were analyzed accordingly. As shownin Table 3, no difference in age at onset, family historyfor any of studied comorbidities, or the presence of co-morbidities was found in eNOS AspAsp homozygouscompared to eNOS GluGlu or eNOS GluAsp carriers.

COMMENTSThe molecular mechanisms of migraine pain have

yet to be clarified. The observation that headache maybe induced by infusions of glyceryl trinitrate (exoge-nous NO donor) and histamine (which liberates NOfrom the vascular endothelium) purports a crucial roleof NO as a likely candidate molecule for migrainepathogenesis.13

Table 2.—Risk of Migraine with Aura in AspAsp Homozygous

AspAsp vs GluGlu+GluAsp

Groups/Genotype OR 95% CI P

MA vs MO 3.02 1.21-7.51 .02MA vs controls 2.21 1.00-5.04 .05MO vs controls 1.16 0.77-1.78 .46

MO = migraine patients without aura; MA = migraine patientswith aura; OR = odds ratio; CI = confidence interval.

NO, constitutively produced by eNOS, plays crit-ical roles in vascular biology, including regulationof vascular tone. Here we reported that a commongenetic variation of eNOS, which results in replace-ment of an Asp residue instead of Glu at position 298

Table 3.—Demographic and Clinical Characteristics ofMigraine Patients According to eNOS Genotype

eNOS Genotype

Variable GluGlu+GluAsp AspAsp

N 133 23Age, years 33.1 ± 9.9 32.4 ± 11.1Gender, F % 78.1 (104) 82.6 (19)Age at onset, years 20.0 ± 9.3 20.3 ± 10.4FH migraine, % 67.7 (90) 69.6 (16)FH cerebrovascular disease, % 30.1 (40) 21.7 (5)FH cardiovascular disease, % 27.8 (37) 26.1 (6)FH neurological disease, % 4.5 (6) 8.7 (2)Smoking, % 21.0 (28) 26.1 (6)Hypertension, % 7.5 (10) 4.3 (1)Cardiomyopathy, % 6.0 (8) 4.3 (1)Diabetes, % 0.0 (0) 0.0 (0)Hypercholesterolemia, % 18.8 (25) 26.1 (6)Asthma, % 5.3 (7) 4.3 (1)Allergy, % 21.0 (28) 26.1 (6)Sinusitis, % 10.5 (14) 0.0 (0)Epilepsy, % 2.2 (3) 0.0 (0)Gastritis/gastric disease, % 21.8 (29) 17.4 (4)

F = female; FH = family history.Number of subjects between brackets.No significant differences between groups were found.

1578 November/December 2006

(Glu298Asp) of mature eNOS, is an independent riskfactor for MA.

In vitro studies have demonstrated a decreasedeNOS activity due to the presence of Asp genetic vari-ation;14,15 however, the functional significance of thispolymorphism in vivo is still not clear.

Clinical studies have reported an association be-tween Glu298Asp mutation and coronary artery dis-ease,9,16,17 essential hypertension,18,19 myocardial in-farction,20 and carotid atherosclerosis.11 These studiesall suggest a link between eNOS polymorphism andcerebrovascular disease.

The relationship between migraine and cere-brovascular disease, that is stroke, is one of the mostperplexing problems for neurologists. Although the2 diseases differ in terms of epidemiology, a num-ber of case-control and population-based studies havereported a significant association between ischemicstroke and a history of migraine, especially MA, whichappears to be an independent risk factor for ischemicstroke.21

The present study is the first to examine the re-lationship between migraine and a functional geneticvariation implicated in endothelial function and incerebrovascular pathology, ie, eNOS polymorphism.Different articles have addressed the concept that NOplays a major role in the vasodilation associated withthe headache phase of MA;22 the present findingsmight suggest that this genetic variation within eNOSgene might contribute to MA pathogenesis, affectingeNOS activity.

Some limits to this study should be acknowledged.This is a pilot study in a specific population, and therelevance of this polymorphism needs further inves-tigation in larger sample size as well as in other case-control studies. Finally, it should be of interest the eval-uation of the relationship between white matter lesionsdescribed in migraine patients23 and eNOS polymor-phism. Although this may be considered a pilot study,the present findings support a crucial role of NO inMA pathogenesis and might suggest new hints in ther-apeutic approaches.

Acknowledgment: The authors are particularly grate-

ful to Dr. Cristina Brambilla for excellent clinical assis-

tance.Conflicts of Interest: None

REFERENCES

1. Gardner K. The genetic basis of migraine: How muchdo we know? Can J Neurol Sci. 1999;26:S37-S43.

2. Montagna P. Molecular genetics of migraineheadaches: A review. Cephalalgia. 2000;20:3-14.

3. Moncada S, Higgs EA. Endogenous nitric oxide:Physiology, pathology and clinical relevance. Eur JClin Invest. 1991;21:361-374.

4. Akerman S, Williamson DJ, Kaube H, Goadsby PJ.Nitric oxide synthase inhibitors can antagonize neu-rogenic and calcitonin gene-related peptide induceddilation of dural meningeal vessels. Br J Pharmacol.2002;137:62-68.

5. Lassen LH, Christiansen I, Iversen HK, Jansen-Olesen I, Olesen J. The effect of nitric oxide syn-thase inhibition on histamine induced headacheand arterial dilatation in migraineurs. Cephalalgia.2003;23:877-886.

6. Alderton WK, Cooper CE, Knowles RG. Nitricoxide synthases: Structure, function and inhibition.Biochem J. 2001;357:593-615.

7. Griffiths LR, Nyholt DR, Curtain RP, Goadsby PJ,Brimage PJ. Migraine association and linkage studiesof an endothelial nitric oxide synthase (NOS3) genepolymorphism. Neurology. 1997;49:614-617.

8. Veldman BA, Spiering W, Doevendans PA, et al. TheGlu298Asp polymorphism of the NOS 3 gene as a de-terminant of the baseline production of nitric oxide.J Hypertens. 2002;20:2023-2027.

9. Hingorani AD, Liang CF, Fatibene J, et al. A com-mon variant of the endothelial nitric oxide synthase(Glu298–>Asp) is a major risk factor for coronaryartery disease in the UK. Circulation. 1999;100:1515-1520.

10. Rose KM, Carson AP, Sanford CP, et al. Migraineand other headaches: Associations with Rose anginaand coronary heart disease. Neurology. 2004;63:2233-2239.

11. Lembo G, De Luca N, Battagli C, et al. A com-mon variant of endothelial nitric oxide synthase(Glu298Asp) is an independent risk factor for carotidatherosclerosis. Stroke. 2001;32:735-740.

12. Headache Classification Subcommittee of the Inter-national Headache Society. The international classi-fication of headache disorders. 2nd ed. Cephalalgia.2004;24(suppl 1):9-160.

13. Olesen J, Jansen-Olesen I. Nitric oxide mechanismsin migraine. Pathol Biol (Paris). 2000;48:648-657.

14. Tesauro M, Thompson WC, Rogliani P, Qi L,

Headache 1579

Chaudhary PP, Moss J. Intracellular processing ofendothelial nitric oxide synthase isoforms associatedwith differences in severity of cardiopulmonary dis-eases: Cleavage of proteins with aspartate vs gluta-mate at position 298. Proc Natl Acad Sci U S A.2000;97:2832-2835.

15. Fairchild TA, Fulton D, Fontana JT, Gratton JP, Mc-Cabe TJ, Sessa WC. Acidic hydrolysis as a mechanismfor the cleavage of the Glu(298)–>Asp variant of hu-man endothelial nitric-oxide synthase. J Biol Chem.2001;276:26674-26679.

16. Yoshimura M, Yasue H, Nakayama M, et al. A mis-sense Glu298Asp variant in the endothelial nitric ox-ide synthase gene is associated with coronary spasmin the Japanese. Hum Genet. 1998;103:65-69.

17. Liao YL, Saku K, Ou J, et al. A missense mutation ofthe nitric oxide synthase (eNOS) gene (Glu298Asp)in five patients with coronary artery disease—casereports. Angiology. 1999;50:671-676.

18. Shoji M, Tsutaya S, Saito R, Takamatu H, YasujimaM. Positive association of endothelial nitric oxide syn-

thase gene polymorphism with hypertension in north-ern Japan. Life Sci. 2000;66:2557-2562.

19. Jachymova M, Horky K, Bultas J, et al. Associationof the Glu298Asp polymorphism in the endothelialnitric oxide synthase gene with essential hypertensionresistant to conventional therapy. Biochem BiophysRes Commun. 2001;284:426-430.

20. Shimasaki Y, Yasue H, Yoshimura M, et al. Asso-ciation of the missense Glu298Asp variant of theendothelial nitric oxide synthase gene with myocar-dial infarction. J Am Coll Cardiol. 1998;31:1506-1510.

21. Milhaud D, Bogousslavsky J, van Melle G, LiotP. Ischemic stroke and active migraine. Neurology.2001;57:1805-1811.

22. Myers DE. Potential neurogenic and vascular rolesof nitric oxide in migraine headache and aura.Headache. 1999;39:118-124.

23. Kruit MC, van Buchem MA, Hofman PA, et al. Mi-graine as a risk factor for subclinical brain lesions.JAMA. 2004;291:427-434.