Aw

LRa

b

a

ARRAA

KAMHR

1

edcimtbo

rmp

(

HT

0h

Clinical Neurology and Neurosurgery 115 (2013) 1693– 1696

Contents lists available at ScienceDirect

Clinical Neurology and Neurosurgery

jou rn al h om epage: www.elsev ier .com/ locate /c l ineuro

ssociation of methylenetetrahydrofolate reductase polymorphismsith susceptibility to Alzheimer’s disease

eila Mansouria, Najiba Fekih-Mrissaa,∗, Sarra Klaia, Malek Mansourb, Nasreddine Gritli a,idha Mrissab

Laboratory of Molecular Biology, Department of Hematology, Military Hospital, Tunis, TunisiaDepartment of Neurology, Military Hospital, Tunis, Tunisia

r t i c l e i n f o

rticle history:eceived 3 July 2012eceived in revised form 15 March 2013ccepted 25 March 2013vailable online 6 May 2013

eywords:lzheimer’s diseaseTHFR protein

a b s t r a c t

Background: Genetic risk factors play an important role in the pathogenesis of Alzheimer’s disease (AD).In this case-control study, we examined the C677T and A1298C polymorphisms in the methylenetetrahy-drofolate reductase (MTHFR) gene and their correlation with this pathology.Objective: To verify the association between MTHFR C677T and A1298C polymorphisms and Alzheimer’sdisease.Method: This work was conducted as a case–control study. Cases consisted of thirty-eight patients and100 individuals without dementia constituted the control group. Genotyping of MTHFR polymorphismswas performed on patients and controls.

umanisk factor

Result: Genetic analyses did not indicate a significant association between the MTHFR C677T mutationand AD (C/T: 63.15% versus 39%, p = 0.087). However, the genotype prevalence of the missense variantMTHFR A1298C was significantly different between patients and controls (A/C: 55% versus 7%, p < 10−3).Our data suggest an association between the MTHFR A1298C mutation and AD; however, the MTHFRC677T mutation did not contribute to susceptibility for AD.Conclusion: The MTHFR A1298C polymorphism is a possible risk factor for Alzheimer’s disease.

. Introduction

The number of elderly has risen with the increase in lifexpectancy and, as a consequence, the prevalence of age-relatedisease has risen as well. Alzheimer’s disease (AD) is the leadingause of dementia in the elderly [1]. This type of dementia is a genet-cally complex disorder that causes progressive memory loss in

id-to-late adult life. Some individuals inherit this form of demen-ia before the age of 65 (referred to as early-onset or familial AD);ut most often, AD occurs late in life [2]. Many genes are responsibler associated with AD [3–5].

Methylene-tetrahydrofolate reductase (MTHFR) is a key enzymeegulating folate metabolism and it is thought to influence DNAethylation and nucleic acid synthesis [6,7]. The MTHFR polymor-

hisms are associated with reduced MTHFR activity [6].MTHFR irreversibly converts 5,10-methylenetetrahydrofolate

5,10-MTHF) to 5-methyltetrahydrofolate (5-MTHF) which is

∗ Corresponding author at: Laboratory of Molecular Biology, Department ofematology, Military Hospital, 1008 Mont Fleury, Tunis, Tunisia.el.: +216 22510488; fax: +216 70726084.

E-mail address: fnajiba@yahoo.fr (N. Fekih-Mrissa).

303-8467/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.clineuro.2013.03.015

© 2013 Elsevier B.V. All rights reserved.

necessary to remethylate the neurotoxic intermediate homocys-teine to methionine. Methionine, in turn, serves as precursor ofS-adenosylmethionine (SAM) that is essential for CNS myelination.SAM has anti-inflammatory properties and is necessary for CNS(re)myelination [8].

The reduced MTHFR activity can result in higher levels of plasmahomocysteine and lower levels of S-adenosylmethionine (SAM)which is necessary for CNS (re)myelination [8]; consequently, neu-rotoxicity is thought to increase while CNS (re)myelination isdiminished. High levels of plasma homocysteine can compromisenormal cell function in many different tissues; for example, thevascular endothelium can be damaged thereby causing neural celldegeneration and so further the development of dementia [8,9,10].

Mutations in the MTHFR gene (MTHFR C677T and MTHFRA1298C) have been factors implicated in several diseases; such as,arterial and venous thrombosis [3,11], obstetrical pathologies [12],ischemic stroke [13], Parkinson’s disease [10], and multiple sclero-sis [14]. The human MTHFR gene (MIM *607093) has been localizedto chromosome 1p36.3 and is composed of 11 exons [15,16]. Two

common polymorphisms of the MTHFR gene, the thermolabileC677T (ALA222VAL) and the A1298C (GLU429ALA) polymorphismhave been investigated in several diseases [3,10–14,17]. The muta-tion in the heterozygous or homozygous state is correlated with

1 and N

rcs[

2

2

Pwaedtsfhapttta

tTaicneadda

waeTnfwtrw

pib

2

−DsbTrt[

694 L. Mansouri et al. / Clinical Neurology

educed enzyme activity and increased thermolability in lympho-yte extracts [18]. Individuals homozygous for the mutation hadignificantly elevated plasma levels of total homocysteine (tHcy)3].

. Patients and methods

.1. Patients

We investigated 38 patients with a diagnosis of probable AD.atients were recruited consecutively and diagnoses for all patientsere established according to the National Institute of Neurological

nd Communicative Disorders and the Stroke and Alzheimer’s dis-ase and Related Disorders Association (NINCDS-ADRDA) clinicaliagnostic criteria [19] and by CT-scan and/or MRI. Exclusion cri-eria included any clinical, biological, or radiological evidence thatuggested a diagnosis other than Alzheimer’s disease. Patients suf-ering from any serious pathologies (e.g., cardiac, pulmonary, renal,epatic, thyroiditis, and immune deficiency) as well as those withn alcohol dependence or drug addiction were also excluded. Thoseatients, not initially excluded, were required to complete at leastwo years of follow-up before inclusion in the study. During thesewo follow-up years, all patients who experienced any event otherhan those strictly associated with AD progression were excludedlso.

The studied cohort consisted of 10 women and 28 men (male-o-female ratio = 2.8) with a mean age ± SD of 75 ± 10.22 years.heir ages ranged between 64 and 85 years and were referred fromll regions of Tunisia. The size of this study was somewhat lim-ted due to the stringent selection criteria. Many potential patientandidates were excluded when any one of inclusion criteria wasot satisfied or if the tests were not completed. Each patient wasvaluated to determine an approximate date of AD onset. Thisssessment was based on careful review of medical records andetailed interviews with primary caregivers. The date of onset wasefined as the date at which the “earliest definite AD symptom”ppeared. Patients were recruited after two years of follow-up.

The control group consisted of 100 healthy individuals (37omen and 63 men, male-to-female ratio = 1.7) with a mean

ge ± SD of 73 ± 9.6 years (range = 52–84 years). Controls werevaluated as subjects with good medical health for their age.he evaluation included medical history and both physical andeurological examinations. Only those with integrity of cognitive

unction and with a mini mental status (MMS) score of 29 or 30 andho had completed at least 2 years of follow-up were included in

he analysis. A family history of dementia was considered as crite-ion for exclusion. All the exclusion criteria for the patient groupere also applied to the control group.

Controls were of diverse Tunisian origin similar to that of theatients. All participants gave their informed consent before partic-

pation in the study and following the approval of the study protocoly the ethics committee of the Military Hospital of Tunis.

.2. Genotyping of MTHFR

Blood samples were collected in EDTA tubes and stored at20 ◦C until use. DNA was isolated with a commercial kit (QIAampNA Blood Kit, Qiagen, [Hilden, Germany]) using the protocol

uggested by the manufacturer. A multiplex amplification withiotinylated primers was used to amplify the genomic DNA.

he polymerase chain reaction (PCR) products then underwent aeverse hybridization by the use of the GenoType® MTHFR testhat is based on the DNA.STRIP technology (Hain Lifescience GmbHNehren, Germany]) and permits the combined molecular genetic

eurosurgery 115 (2013) 1693– 1696

characterization of the C677T and A1298C mutations of the humanMTHFR gene.

2.3. Statistical analyses

The Student’s t-test was used to compare differences in meansand differences between cases and controls were evaluated byusing the chi-square test. The odds ratios (OR) and 95% confidenceintervals (CI) were also calculated. Probability values P < 0.05 wereconsidered statistically significant.

3. Results

Mean data on patients and controls were reported in Table 1. Alldata was recorded after these two years of follow-up. Clinical infor-mation regarding the patients and controls revealed that tobaccoand alcohol use, diabetes, and hypertension did not appear to berisk factors for AD as was reflected by their p values (Table 1).

The relative frequencies between the patients and controls,respectively, were the following: 34.21% (13/38) vs. 52% for theCC genotype, 63.15% (24/38) vs. 39% for the CT genotype, and 2.63%(1/38) vs. 9% for the TT genotype. Our result did not indicate a sig-nificant association between the MTHFR C677T mutation and AD.In contrast, the A1298C mutation was significantly more frequentin patients with this pathology (60.52% (23/38) for the AC genotypevs. 7%; OR = 21.8 (7.95–60.3), p < 10−3). The CC genotype was absentin both the patient and control groups (Table 2).

A comparison of alleles showed that the frequency of the T allelein the MTHFR C677T polymorphism did not differ significantlybetween subjects and controls (34.2% (26/76) vs. 28.5% (57/200),OR = 1.3 (0.74–2.3), p = 0.35); however, in the A1298C polymor-phism, the C allele frequency was significantly more prevalent inthe pathologic group when compared to the controls (30.3% (23/76)vs. 3.5% (7/200), OR = 12 (4.8–29.4), p < 10−3).Mutations were alsoanalyzed according to gender and age of onset. The presence ofthe MTHFR 677 and 1298 mutations was found to be indepen-dent of sex in the patient group. Eighteen males were noted witha MTHFR 677CT mutation versus 7 females and 10 males versus 3females were wild-type (OR = 1.30 (0.27–6.2), p = 1.0). Similar anal-ysis revealed 18 males with a MTHFR 1298AC mutation versus5 females and 10 males and 5 females were wild-type (OR = 1.8(0.42–7.8), p = 0.47). There were also no significant differencesbetween the mean ages of those with the MTHFR C677T muta-tion (mean = 74.42 years, SD = 6.6) and those without (mean = 76years, SD = 6.5), or between those with the A1298C mutation(mean = 74.09 years, SD = 7.2) and those without (mean = 76.4 years,SD = 5.1). The respective p values for each mutation were p = 0.47and p = 0.3.

4. Discussion

According to Spiroski et al., the 677TT genotype was alwaysassociated with the normal 1298AA genotype in all individu-als; and, the 1298CC genotype was always associated with thenormal 677CC genotype [3]. In Spiroski’s study, double mutanthomozygous individuals were not observed [3]. Similar data andobservations were reported by others as well [20]. The presentstudy also contained no double mutant homozygous individualsas there were no carriers of the MTHFR 1298CC genotype.

In the case of the MTHFR C677T genotype, the mutation wasnot significantly associated with AD. Religa et al. also failed to

find an association between AD and the C/T polymorphism in theMTHFR gene [21]. Our results are similar to a study by Reglandet al. in which they found 40% of the population was heterozy-gous (CT), 11% were homozygous (TT), and 49% were wild type (CC)

L. Mansouri et al. / Clinical Neurology and Neurosurgery 115 (2013) 1693– 1696 1695

Table 1Comparison of variables in Alzheimer patients and controls after 2 follow-up years.

Clinical variable Patients (N = 38) Controls (N = 100) OR (95% CI) P-value

Gender 10 f/28 m 37 f/63 mMean age ± S.D. 75 ± 10.22a 73 ± 9.6b

Tobacco use 15 (39.5%) 30 (30%) 0.65 (0.30–1.43) 0.31Alcohol use 13 (34.2%) 22 (22%) 0.54 (0.24–1.23) 0.18Diabetes 12 (31.5%) 27 (27%) 0.80 (0.35–1.80) 0.67Hypertension 20 (52.6%) 43 (43%) 0.67 (0.32–1.44) 0.34

CI, confidence interval; OR, odds ratio; P-value, probability value.a Mean age of onset of the disease.b Age at examination for controls.

Table 2Genotype frequencies of MTHFR polymorphisms in Alzheimer patients and controls.

Genotypes Patients (N = 38) Controls (N = 100) OR (95% CI)a P-value

MTHFR A1298CAA 15 (39.48%) 93 (93%) 21.8 (7.95–60.3) < 10−3

AC 23 (60.52%) 7 (7%)CC 0% 0 (0%)

MTHFR C677TCC 13 (34.21%) 52 (52%) 2 (0.92–4.37) 0.087CT 24 (63.15%) 39 (39%)TT 1 (2.63%) 9 (9%)

Ces) ve

[pMeCCc

iad6(c

twgh

hmtaSmsel

mfpog(rfa

I, confidence interval; OR, odds ratio.a Patients versus controls, carriers of the polymorphism (hetero- and homozygot

22]. Nishiyama et al. revealed no association between the C677Tolymorphism and AD although the same authors had found thatTHFR was slightly associated with the onset of senile dementia

specially among men [6]. These results are similar to Japanese andanadian population data, in which they also found that the MTHFR677T allele frequencies were not significantly different betweenontrol and case groups [1].

The MTHFR 677C (677T) allele frequencies were 65.8% (34.2%)n the case group versus 71.5% (28.5%) in the control group. Thisssociation was not statistically significant. Our finding is in concor-ance with those of Fernandez and Scheibe who observed MTHFR77C (677T) allele frequencies of 65% (35%) of cases versus 71%29%) of controls. This difference was also not statistically signifi-ant [1].

On the contrary however, the distribution of AC and CC geno-ypes in MTHFR A1298C showed a highly significant associationith AD. There were no subjects of the study who were homozy-

ous (1298CC) for the mutation. Among the patients, 60.52% wereeterozygous versus only 7% of the controls.

Clark et al., Postiglione et al., Brunelli et al. and Prince et al.,owever, did not find a significant relation between MTHFR C677Tutation and AD [7,23–25]. Seripa et al. also found no difference in

he distribution of this MTHFR polymorphism between AD casesnd elderly controls in both American and Italian cohorts [26].ilva et al. reported that neither C677T nor A1298C MTHFR poly-orphisms contributed to genetic susceptibility for AD in Brazilian

ubjects [17]. By contrast, Wakutani et al. reported the protectiveffects of the (677C–1298C) haplotype of the MTHFR gene againstate-onset AD (LOAD) [27].

The present study, however, found that the MTHFR A1298Cutation was a significant risk factor for AD. The MTHFR 1298 allele

requencies were the following: allele A was present in 69.7% ofatients versus 96.5% of controls and the minor allele frequencyf MTHFR A1298C was not significantly present in the controlroup (3.5%); however, the C allele was well represented in patients

30.3%) and so resulted in a statistically significant p value. Thisesult indicates that this allele can be considered as a significant riskactor for AD. It should be noted, however, that given the exception-lly low frequency among the controls, this polymorphism should

rsus wild-type; P-value, probability value; Significant P-value is in bold.

be tested in larger cohorts recruited countrywide to confirm thisresult.

The literature is sparse with results concerning the effect of theC allele in the MTHFR 1298 gene on the risk of AD. One such studyby Bosco et al. failed to find a significant relation between this alleleand the pathology [28].

There are several limitations of this study. The number ofincluded patients was relatively low and requires validation ofthe results in larger cohorts to confirm our results. It shouldalso be noted that MTHFR polymorphisms are associated withreduced MTHFR activity that can result in the impairment of Hcymetabolism; therefore, homocysteine levels should be measured tocorrelate these results with MTHFR polymorphisms. Lastly, APOE isone of the most influential genes related to onset and evolution ofdementia due to Alzheimer’s disease. This marker should be testedin our patients to evaluate the relationship between the APOE geneand MTHFR polymorphisms. The future continuation of this workis to study homocysteine levels, APOE haplotypes, and to correlatethese findings with the MTHFR polymorphisms.

5. Conclusion

This study observed no significant association between theMTHFR C677T polymorphism and AD; however, a significant asso-ciation was demonstrated between the MTHFR A1298C mutationand AD. The limit of this study is the small number of patients. How-ever, further study of these polymorphisms in AD, throughout theworld, could help to understand the contributions of these markersin AD susceptibility.

Acknowledgements

We would like to thank Dr. Christian Winchell for his precioushelp in correcting this manuscript.

References

[1] Fernandez LL, Scheibe RM. Is MTHFR polymorphism a risk factor for Alzheimerdisease like APOE? Arquivos de Neuro-Psiquiatria 2005;63:1–6.

1 and N

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

696 L. Mansouri et al. / Clinical Neurology

[2] Wingo TS, Lah JJ, Levey AI, Cutler DJ. Autosomal recessive causeslikely in early-onset Alzheimer disease. Archives of Neurology 2012;69:59–64.

[3] Spiroski I, Kedev S, Antov S, Arsov T, Krstevska M, Dzhekova-Stojkova S,et al. Methylenetetrahydrofolate reductase (MTHFR-677 and MTHFR-1298)genotypes and haplotypes and plasma homocysteine levels in patients withocclusive artery disease and deep venous thrombosis. Acta Biochimica Polonica2008;55:587–94.

[4] Huerta C, Alvarez V, Mata IF, Coto E, Ribacoba R, Martinez C, et al. Chemokines(RANTES and MCP-1) and chemokine-receptors (CCR2 and CCR5) gene poly-morphisms in Alzheimer’s and Parkinson’s disease. Neuroscience Letters2004;370:151–4.

[5] Jicha GA, Parisi JE, Dickson DW, Cha RH, Johnson KA, Smith GE, et al. Age andapoE associations with complex pathologic features in Alzheimer’s disease.Journal of the Neurological Sciences 2008;273:34–9.

[6] Nishiyama M, Kato Y, Hashimoto M, Yukawa S, Omori K, Apolipoprotein E.methylenetetrahydrofolate reductase (MTHFR) mutation and the risk of seniledementia – an epidemiological study using the polymerase chain reaction (PCR)method. Epidemiology 2000;10:163–72.

[7] Clarke R, Smith AD, Jobst KA, Refsum H, Sutton L, Ueland PM. Folate, vita-min B12, and serum total homocysteine levels in confirmed Alzheimer disease.Archives of Neurology 1998;55:1449–55.

[8] Surtees R, Leonard J, Austin S. Association of demyelination with deficiency ofcerebrospinal-fluid S-adenosylmethionine in inborn errors of methyl-transferpathway. Lancet 1991;338:1550–4.

[9] Graham IM, Daly LE, Refsum HM, Robinson K, Brattström LE, Ueland PM, et al.Plasma homocysteine as a risk factor for vascular disease. The European Con-certed Action Project. JAMA 1997;277:1775–81.

10] Religa D, Czyzewski K, Styczynska M, Peplonska B, Lokk J, Chodakowska-Zebrowska M, et al. Hyperhomocysteinemia and methylenetetrahydrofolatereductase polymorphism in patients with Parkinson’s disease. NeuroscienceLetters 2006;404:56–60.

11] Fekih-Mrissa N, Klai S, Bafoun A, Nciri B, Hmida J, Gritli N. Role of thrombophiliain vascular access thrombosis among chronic hemodialysis patients in Tunisia.Therapeutic Apheresis and Dialysis 2011;15:40–3.

12] Klai S, Fekih-Mrissa N, El Housaini S, Kaabechi N, Nsiri B, Rachdi R, et al. Associa-tion of MTHFR A1298C polymorphism (but not of MTHFR C677T) with elevatedhomocysteine levels and placental vasculopathies. Blood Coagulation and Fib-rinolysis 2011;22:374–8.

13] Casas JP, Hingorani AD, Bautista LE, Sharma P. Meta-analysis of genetics studiesin ischemic stroke. Thirty-two genes involving approximately 18,000 cases and58,000 controls. Archives of Neurology 2004;61:1652–62.

14] Klotz L, Farkas M, Bain N, Keskitalo S, Semmler A, Ineichen B, et al. The vari-ant methylenetetrahydrofolate reductase c.1298A > C (p.E429A) is associatedwith multiple sclerosis in a German case-control study. Neuroscience Letters

2010;468:183–5.

15] Goyette P, Sumner JS, Milos R, Duncan AMV, Rosenblatt DS, MatthewsRG, et al. Human methylenetetrahydrofolate reductase: isolation ofcDNA, mapping and mutation identification. Nature Genetics 1994;7:195–200.

[

eurosurgery 115 (2013) 1693– 1696

16] Goyette P, Pai A, Milos R, Frosst P, Tran P, Chen Z, et al. Gene structureof human and mouse methylenetetrahydrofolate reductase (MTHFR). Mam-malian Genome 1998;9:652–6.

17] da Silva VC, Ramos FJ, Freitas EM, de Brito-Marques PR, Cavalcanti MN,D’Almeida V, et al. Alzheimer’s disease in Brazilian elderly has a relationwith homocysteine but not with MTHFR polymorphisms. Arquivos de Neuro-Psiquiatria 2006;64:941–5.

18] Strandhagen E, Zetterberg H, Aires N, Palmér M, Rymo L, Blennow K, et al.The methylenetetrahydrofolate reductase C677T polymorphism is a majordeterminant of coffee-induced increase of plasma homocysteine: a random-ized placebo controlled study. International Journal of Molecular Medicine2004;13:811–5.

19] McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinicaldiagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Groupunder the auspices of Department of Health and Human Services Task Force onAlzheimer’s Disease. Neurology 1984;34:939–44.

20] Freitas AI, Mendonc a I, Guerra G, Brión M, Reis RP, Carracedo A, et al.Methylenetetrahydrofolate reductase gene, homocysteine and coronary arterydisease: the A1298C polymorphism does matter. Inferences from a case study(Madeira, Portugal). Thrombosis Research 2008;122:648–56.

21] Religa D, Styczynska M, Peplonska B, Gabryelewicz T, Pfeffer A, Chodakowska M,et al. Homocysteine, apolipoprotein E and methylenetetrahydrofolate reduc-tase in Alzheimer1s disease and mild cognitive impairment. Dementia andGeriatric Cognitive Disorders 2003;16:64–70.

22] Regland B, Blennow K, Germgard T, Koch-Schmidt AC, Gottfries CG. The role ofpolymorphic genes apolipoprotein E and methylenotetrahidrofolate reductasein the development of dementia of Alzheimer type. Dementia and GeriatricCognitive Disorders 1999;10:245–51.

23] Postiglione A, Milan G, Ruoceo A, Gallotta G, Guiotto G, Di Minno G. Plasmafolate, vitamin B12, and total homocysteine and homozygosity for the C677Tmutation of the 5,10-methylene tetrahydrofolate reductase gene in patientswith Alzheimer’s dementia: a case-control study. Gerontology 2001;47:324–9.

24] Brunelli T, Bagnoli S, Giusti B, Nacmias B, Pepe G, Sorbi S, et al. TheC677T methylenetetrahydrofolate reductase mutation is not associated withAlzheimer’s disease. Neuroscience Letters 2001;315:103–5.

25] Prince JA, Feuk L, Sawyer SL, Gottfries J, Ricksten A, Nägga K, et al. Lack ofreplication of association findings in complex disease: an analysis of 15 poly-morphisms in prior candidate genes for sporadic Alzheimer’s disease. EuropeanJournal of Human Genetics 2001;9:437–44.

26] Seripa D, Forno GD, Matera MG, Gravina C, Margaglione M, Palermo MT,et al. Methylenetetrahydrofolate reductase, angiotensin converting enzymegene polymorphisms in two genetically, and diagnostically distinct cohort ofAlzheimer patients. Neurobiology of Aging 2003;24:933–9.

27] Wakutani Y, Kowa H, Kusumi M, Nakaso K, Yasui K, Isoe-Wada K, et al. A hap-lotype of the methylenetetrahydrofolate reductase gene is protective against

late-onset Alzheimer’s disease. Neurobiology of Aging 2004;25:291–4.

28] Bosco P, Guéant-Rodríguez RM, Anello G, Romano A, Namour B, Spada RS, et al.Association of IL-1 RN*2 allele and methionine synthase 2756 AA genotypewith dementia severity of sporadic Alzheimer’s disease. Journal of Neurology,Neurosurgery and Psychiatry 2004;75:1036–8.