Folate gene polymorphisms and the risk of Down syndrome pregnancies in young Italian women

ß 2006 Wiley-Liss, Inc. American Journal of Medical Genetics Part A 140A:1083 – 1091 (2006) Folate Gene Polymorphisms and the Risk of Down Syndrome Pregnancies in Young Italian Women Fabio Coppedè,1 Giulia Marini,1 Stefania Bargagna,2 Liborio Stuppia,3,4 Fabrizio Minichilli,5 Ilaria Fontana,1 Renato Colognato,1 Guia Astrea,2 Giandomenico Palka,3,6 and Lucia Migliore1* 1 Department of Human and Environmental Sciences, University of Pisa, Pisa, Italy 2 Scientific Institute ‘‘Stella Maris,’’ Calambrone, Pisa, Italy 3 Department of Biomedical Sciences, ‘‘G. D’Annunzio’’ University Foundation, Chieti-Pescara, Italy 4 I.T.O.I. CNR, c/o IOR, Bologna, Italy 5 Department of Epidemiology, Institute of clinical Physiology, National Council of Research (C.N.R), Pisa, Italy 6 Human Genetic Division, Pescara Hospital, Pescara, Italy Received 4 November 2005; Accepted 17 February 2006 Maternal impairments in folate metabolism and elevated homocysteinemia are known risk factors for having a child with Down syndrome (DS) at a young age. The 80G>A polymorphism of the reduced folate carrier gene (RFC-1) has been recently demonstrated to affect plasma folate and homocysteine levels, alone or in combination with the 677C>T polymorphism in the methylenetetrahydrofolate reductase (MTHFR) gene. We performed the present study on 80 Italian mothers of DS individuals, aged less than 35 at conception, and 111 Italian control mothers, to study the role of the RFC-1 80G>A, MTHFR 677C>T, and MTHFR 1298A>C genotypes to the risk of a DS offspring at a young maternal age. When polymorphisms were considered alone, both allele and genotype frequencies did not significantly differ between DS mothers and control mothers. However, the combined MTHFR677TT/RFC-1 80GG genotype was borderline associated with an increased risk (OR 6 (CI 95%: 1.0– 35.9), P ¼ 0.05), and to be MTHF1298AA/RFC-1 80(GA or AA) was inversely associated with the risk (OR 0.36 (CI 95%: 0.14–0.96), P ¼ 0.04). Present results seem to indicate that none of the RFC-1 80G>A, MTHFR 677C>T, and MTHFR 1298A>C polymorphisms is an independent risk factor for a DS offspring at a young maternal age; however, a role for the combined MTHFR/RFC-1 genotypes in the risk of DS pregnancies among young Italian women cannot be excluded. ß 2006 Wiley-Liss, Inc. Key words: Down syndrome; MTHFR; RFC-1; folate metabolism; folate gene polymorphisms; risk of DS How to cite this article: Coppedè F, Marini G, Bargagna S, Stuppia L, Minichilli F, Fontana I, Colognato R, Astrea G, Palka G, Migliore L. 2006. Folate gene polymorphisms and the risk of Down syndrome pregnancies in young Italian women. Am J Med Genet Part A 140A:1083–1091. INTRODUCTION Down syndrome (DS) is a genetic disease resulting from the presence and the expression of three copies of the genes located on chromosome 21. Since the discovery of Lejeune et al. [1959], the phenotype of DS has been associated with trisomy for chromosome 21. For the majority of DS cases (92%), the extra chromosome stems from the failure of a normal chromosome segregation during meiosis (meiotic nondisjunction) [Epstein, 1995], and the nondisjunction is maternal in 95% of cases, occurring primarily during meiosis I in the maturing oocyte, before conception [Antonarakis et al., 1998]. Only 5% of DS cases are due to translocations and the remaining 3% of cases to mosaicism [Antonarakis, 1998]. The prevalence of DS births has decreased over the past two decades from 1 in 700 to about 1 in 1000, likely due to the increased application of prenatal diagnosis [Olsen and Cross, 1997]. The molecular mechanisms underlying meiotic nondisjunction leading to trisomy 21 are still poorly understood and the major risk factor for trisomy 21 is advanced maternal age at conception [Antonarakis, 1998], while paternal age seems to be insignificant in the etiology of DS [Gaulden, 1992]. After age 35 the risk of bearing a child with DS increases substantially with increasing maternal age; however, several mothers of DS individuals (DS mothers) are *Correspondence to: Prof. Lucia Migliore, Department of Human and Environmental Sciences, University of Pisa, Via S. Giuseppe 22, 56126 Pisa, Italy. E-mail: l.migliore@geog.unipi.it DOI 10.1002/ajmg.a.31217 American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1084 COPPEDÈ ET AL. 35 years or younger, suggesting a genetic susceptibility to early nondisjunction for chromosome 21 in such women [Schupf et al., 1994]. Moreover, we recently observed that young mothers of DS children, aged less than 35 at conception, are more prone than control mothers to chromosome malsegregation events for both chromosomes 13 and 21 in peripheral blood cells, suggesting a genetic susceptibility to chromosome malsegregation in both somatic cells and gametes [Migliore et al., 2005]. Impairments in folate and methyl metabolism can result in DNA hypomethylation and abnormal chromosomal segregation [Fenech, 2001; Wang et al., 2004], and folate deficiency has been associated with chromosomes 17 and 21 aneuploidy in human lymphocytes [Wang et al., 2004]. Methylenetetrahydrofolate reductase (MTHFR) plays a crucial role in regulating cellular methylation, through the reduction of 5,10-methylentetrahydrofolate (5,10-MTHF) to 5-methyltetrahydrofolate (5-MeTHF) the main circulatory form of folate and one carbon donor for the remethylation of homocysteine to methionine, then converted to the methylating agent S-adenosylmethionine (SAM) [Bailey and Gregory, 1999]. Two common polymorphisms have been described which reduce MTHFR activity: the T variant at nucleotide 677 (MTHFR 677C>T) and the C variant at nucleotide 1298 (MTHFR 1298A>C), this latter to a lesser extent [Frosst et al., 1995; Weisberg et al., 1998]. Conflicting results have been obtained in subsequent studies of association between the MTHFR 677T and 1298C variants, and the risk of giving birth to a child with DS [Chadefaux-Vekemans et al., 2002; O’Leary et al., 2002; Stuppia et al., 2002; Bosco et al., 2003; Boduroglu et al., 2004; Acacio et al., 2005; Chango et al., 2005; da Silva et al., 2005], probably due to both the different size of the case-control groups considered and to geographic variations in allele frequencies among different populations. However, a role of an abnormal folate metabolism in the risk of DS offsprings cannot be excluded, and interactions between folate gene polymorphisms predisposing to elevated homocysteine levels have emerged as a risk factor for having a DS child at a young age [Bosco et al., 2003]. The reduced folate carrier (RFC-1) is responsible for the internalization of 5MeTHF within cells. A single nucleotide polymorphism 80G>A has been discovered in the RFC-1 gene [Chango et al., 2000], this polymorphism has an impact on folate status separately or in combination with the MTHFR 677C>T genotype, and a significant increase in plasma total homocysteine levels was detected in doubly homozygous 80GG/677TT individuals compared with 80GG/677CC subjects [Chango et al., 2000]. A link between DS and neural tube defects (NTD) in the same family has been observed [Barkai et al., 2003], and NTD and DS are influenced by the same genetic determinants of folate metabolism [Gueant et al., 2003]. The RFC-1 80GG genotype has been associated with the risk of NTD in the Italian population [De Marco et al., 2003], however, to the best of our knowledge, it has never been studied as a risk factor for DS pregnancies among young Italian women. In the present study we examined the role of the MTHFR 677C>T, MTHFR 1298A>C, and RFC-1 80G>A polymorphisms, separately or in combinations, in the risk of having a DS child in young Italian women aged less than 35 years at conception. MATERIALS AND METHODS Study Population Peripheral blood samples were collected from 80 DS mothers aging less than 35 years at conception (mean age 28.4
4.5), and from 111 control mothers of the same mean age at conception, who had at least one healthy child and no experience of miscarriages or abnormal pregnancies. All mothers were white Caucasians (Italians), residents of central Italy at interview, and filled-in a detailed questionnaire on their personal, occupational, and medical history and lifestyle. Informed consent for participation to the study was obtained from each subject, and the study was approved by the Scientific Institute ‘‘Stella Maris’’ Ethics Committee, according to the Helsinki declaration. Blood samples (5 ml) were obtained in EDTA tubes from all participants. Genotyping Genomic DNA was isolated from whole blood by means of the QIAamp1 Blood Mini Kit (Qiagen, Milan, Italy) following the manufacturer’s instructions. All genotype analyses were performed using PCR-RFLP technique. The genotyping protocol for the MTHFR 677C>T polymorphism was adapted from Frosst et al. [1995]: a 198-bp product was amplified using 1.25 Units of Taq DNA polymerase (Invitrogen, Milan, Italy), 20 pmol of each primer (Forward: 50 -TGA AGG AGA AGG TGT CTG CGG GA-30 and Reverse: 50 -AGG ACG GTG CGG TGA GAG TG-30 ), 0.2 mM of each dNTP, 2.5 mM MgCl2, and 30 ng of genomic DNA in a total volume of 25 ml. PCR conditions were 40 cycles of 30 sec at 948C, 30 sec at 628C, and 30 sec at 728C, preceded by an initial denaturation of 2 min at 948C and followed by a final extension of 7 min at 728C. Two hours digestion at 378C with Hinf I (Fermentas, Milan, Italy) resulted in 175- and 23-bp products for the 677T allele, and in a 198-bp undigested product for the 677C allele. Digestion products were visualized after electrophoresis on a 3% agarose gel with ethidium bromide. For the analysis of the MTHFR 1298A>C polymorphism we used a modification of a method from American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1085 FOLATE METABOLISM AND THE RISK OF DOWN SYNDROME Weisberg et al. [1998]: a 163-bp product was amplified using 1.25 Units of Taq DNA polymerase (Invitrogen), 10 pmol of each primer (Forward: 50 CTT TGG GGA GCT GAA GGA CTA CTA C-30 and Reverse: 50 -CAC TTT GTG ACC ATT CCG GTT TG30 ), 0.1 mM of each dNTP, 3 mM MgCl2, and 50 ng of genomic DNA in a total volume of 25 ml. PCR conditions were 38 cycles of 1 min at 928C, 1 min at 608C, and 30 sec at 728C, preceded by an initial denaturation of 2 min at 928C, and followed by a final extension of 7 min at 728C. Three hours digestion with Mbo II (Fermentas) resulted in 56, 31, 30, 28, and 18-bp fragments for the 1298A allele, and 84, 31, 30, and 18-bp fragments for the 1298C allele. Digestion products were separated through electrophoresis on a 3% agarose gel containing ethidium bromide, and the major visible bands were those corresponding to the 84-bp and the 56-bp fragments. The genotyping protocol for the RFC-1 80G>A polymorphism was adapted from Chango et al. [2000]: a 230-bp product was amplified using 1.25 Units of Taq DNA polymerase (Invitrogen), 10 pmol of each primer (Forward: 50 -AGTGTCACCTTCGTCCC-30 and Reverse: 50 -TCCCGCGTG AAGTTCTTG-30 ), 0.15 mM of each dNTP, 1.5 mM MgCl2, and 10 ng of genomic DNA in a total volume of 25 ml. PCR conditions were 44 cycles of 30 sec at 948C, 30 sec at 528C, and 45 sec at 728C, preceded by an initial denaturation of 2 min at 948C, and followed by a final extension of 7 min at 728C. Three hours digestion with Cfo I (Sigma, Milan, Italy) resulted in three fragments of 125, 68 and 37-bp, in the presence of the 80G allele. The 80A allele produced two fragments of 162- and 68-bp. Digestion products were visualized after electrophoresis on a 3% agarose gel containing ethidium bromide. To avoid genotyping errors, two replicates of the PCR/RFLP procedures followed by independent readings and comparison were performed for 20% of our samples randomly chosen. In addition, control samples from homozygous and heterozygous individuals, whose genotypes have been previously confirmed, were always included in the PCR/RFLP procedures and analyzed on each gel. Statistical Analyses Allele frequencies were calculated for each genotype, and the difference in allele frequencies between DS mothers and control mothers were determined using chi-square test. Unconditional logistic regression was performed to obtain odds ratios (ORs). We used ORs to quantify the association between each polymorphism and the risk of having a DS child at a young age, and to assess possible interactions between polymorphisms. The statistical significance of each association was tested by z-test. Analyses were performed using the software STATA 8.0 SE. RESULTS Allele and Genotype Frequencies The allele frequencies of MTHFR 677C>T, MTHFR 1298A>C, and RFC-1 80G>A in DS mothers and control mothers are listed in Table I and are consistent with previous reports in Caucasians [Rady et al., 2001; Shi et al., 2003]. The variant allele frequencies among DS mothers were 47.5% for the MTHFR 677T allele, 25.4% for the MTHFR 1298C allele, and 40% for the RFC-1 80A allele (Table I). The variant allele frequencies among control mothers were 40.5% for the MTHFR 677T allele, 30% for the MTHFR 1298C allele, and 44% for the RFC-1 80A allele (Table I). No statistically significant differences in allele frequencies have been observed between DS mothers and control mothers (Table I). Genotype frequencies are listed in Table II. All genotype frequencies among controls were consistent with Hardy–Weinberg equilibrium expectations. The frequencies of MTHFR 677C>T genotypes (CC, CT, and TT) among DS mothers were 25.3, 54.4, and 20.3%, respectively (Table II). The corresponding frequencies among controls were 35.1, 48.7, and 16.2% (Table II). There were no significant differences in genotype frequencies between the two groups. Combination of heterozygous and homozygous MTHFR 677 variant genotypes (CC or TT) did not show significant differences between the case and the control groups (Table II). The frequencies of MTHFR 1298A>C genotypes (AA, AC, and CC) among case mothers were 53.6, 42.0, and 4.4%, respectively (Table II). The corresponding frequencies among controls were 46.0, 48.0, and 6.0% (Table II). There were no significant differences in genotype frequencies between the two groups. Combination of heterozygous and homozygous MTHFR 1298 variant genotypes (AC or CC) did not show significant differences between the case and the control groups (Table II). The frequencies of RFC-1 80G>A genotypes (GG, GA, and AA) among case mothers were 39.1, 42.0, and 18.9%, respectively (Table II). The corresponding frequencies among controls were 33.3, 45.2, and 21.5% (Table II). There were no significant differences in genotype frequencies between the two TABLE I. Allele Frequencies of MTHFR 677C > T, 1298A > C, and RFC-1 80G > A in Mothers of Down Syndrome Children (DS Mothers) and Control Mothers Genotype MTHFR 677 MTHFR 1298 RFC-1 80 Allele C T A C G A DS mothers Control mothers alleles (%) alleles (%) 83 (52.5) 75 (47.5) 103 (74.6) 35 (25.4) 83 (60) 55 (40) 132 (59.5) 90 (40.5) 140 (70.0) 60 (30.0) 104 (56) 82 (44) w2 P 1.80 0.179 0.87 0.351 0.58 0.446 American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1086 COPPEDÈ ET AL. TABLE II. Genotype Frequencies of MTHFR 677C > T, 1298A > C, and RFC-1 80G > A in Mothers of Down Syndrome Children (DS Mothers) and in Controls Mothers Genotype MTHFR 677 CC CT TT CT or TT MTHFR 1298 AA AC CC AC or CC RFC-1 80 GG GA AA GA or AA Number (%) of DS mothers Number (%) of control mothers 79 (total) 20 (25.3) 43 (54.4) 16 (20.3) 59 (74.7) 69 (total) 37 (53.6) 29 (42.0) 3 (4.4) 32 (46.4) 69 (total) 27 (39.1) 29 (42.0) 13 (18.9) 42 (60.9) 111 (total) 39 (35.1) 54 (48.7) 18 (16.2) 72 (64.9) 100 (total) 46 (46.0) 48 (48.0) 6 (6.0) 54 (54.0) 93 (total) 31 (33.3) 42 (45.2) 20 (21.5) 72 (66.7) groups. Combination of heterozygous and homozygous RFC-1 80 variant genotypes (GA or AA) did not show significant differences between the case and the control groups (Table II). MTHFR 677C>T and MTHFR 1298A>C Genotype Interaction We compared the genotype frequencies between MTHFR 677CC, CT, TT and MTHFR 1298AA, AC, CC within the case mothers and with the control mothers. Results are shown in Table III. None of the combined genotypes was associated with the risk of having a child with DS (Table III). MTHFR 677C>T and RFC-1 80G>A Genotype Interaction We compared the genotype frequencies between MTHFR 677CC, CT, TT and RFC-1 80GG, GA, AA within the case mothers and with the control OR (95%CI) P 1 (referent) 1.55 (0.79–3.04) 1.73 (0.73–4.11) 1.59 (0.84–3.02) 0.199 0.211 0.151 1 (referent) 0.75 (0.40–1.41) 0.62 (0.14–2.65) 0.74 (0.40–1.36) 0.375 0.521 0.330 1 (referent) 0.79 (0.39–1.59) 0.74 (0.31–1.78) 0.78 (0.41–1.49) 0.516 0.509 0.447 mothers. Results are shown in Table IV. The 677TT/80GG genotype was associated with a borderline significant increased risk of a DS offspring compared with the 677CC/80GG genotype (OR 6 [1.00–35.9] P ¼ 0.05) (Table IV). For the combined 677(CT or TT)/80GG, compared with the 677CC/ 80GG, an increased OR was observed (OR 3.47 [0.96–12.6] P ¼ 0.058), even if not statistically significant (Table IV). MTHFR 1298A>C and RFC-1 80G>A Genotype Interaction We compared the genotype frequencies between MTHFR 1298AA, AC, CC and RFC-1 80GG, GA, AA within the case mothers and with the control mothers. Results are shown in Table V. A significant inverse association with the risk of having a child with DS was observed for the combined 1298AA/ 80(GA or AA) genotype compared with the 1298AA/ 80GG genotype (OR 0.36 [0.14–0.96] P ¼ 0.04) (Table V). TABLE III. Interaction Between MTHFR 677C > T and 1298A > C Genotypes in Mothers of Down Syndrome Children (DS Mothers) and Control Mothers MTHFR677/1298 genotype 677CC/1298AA 677CT/1298AA 677TT/1298AA CT or TT/AAa 677CC/1298AC 677CC/1298CC 677CC/AC or CCb 677CT/1298AC CT or TT/AC or CCc Number of DS mothers (%) Number of control mothers (%) OR 95% CI P 68 (total) 3 (4.4) 18 (26.5) 16 (23.5) 34 (50.0) 8 (11.8) 3 (4.4) 11 (16.2) 20 (29.4) 20 (29.4) 100 (total) 11 (11.0) 18 (18.0) 17 (17.0) 35 (35.0) 16 (16.0) 6 (6.0) 22 (22.0) 31 (31.0) 32 (32.0) 1.0 3.67 3.45 3.56 1.83 1.83 1.83 2.37 2.29 Referent 0.87–15.38 0.81–14.67 0.91–13.89 0.40–8.49 0.28–12.06 0.42–7.95 0.59–9.54 0.57–9.23 0.076 0.094 0.067 0.438 0.528 0.418 0.226 0.243 Missing combinations are due to the absence of DS mothers and/or control mothers with that particular genotype. a Combined 677(CT or TT)/1298AA. b Combined 677CC/1298(AC or CC). c Combined 677(CT or TT)/1298(AC or CC). American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1087 FOLATE METABOLISM AND THE RISK OF DOWN SYNDROME TABLE IV. Interaction Between MTHFR 677C > T and RFC-1 80G > A Genotypes in Mothers of Down Syndrome Children (DS Mothers) and Control Mothers MTHFR677/RFC-180 Genotype Number of DS mothers (%) Number of control mothers (%) OR 95% CI P 677CC/80GG 677CT/80GG 677TT/80GG CT or TT/GGa 677CC/80GA 677CC/80AA CC/GA or AAb 677CT/80GA 677CT/80AA 677TT/80GA 677TT/80AA CT or TT/GA or AAc 68 (total) 4 (5.9) 16 (23.5) 6 (8.8) 22 (32.3) 11 (16.2) 3 (4.4) 14 (20.6) 13 (19.1) 6 (8.8) 5 (7.4) 4 (5.9) 28 (41.2) 93 (total) 12 (12.9) 16 (17.2) 3 (3.2) 19 (20.4) 17 (18.3) 5 (5.4) 22 (23.7) 18 (19.4) 9 (9.7) 7 (7.5) 6 (6.5) 40 (43) 1.0 3.0 6.0 3.47 1.94 1.8 1.9 2.17 2.0 2.14 2.0 2.1 Referent 0.79–11.3 1.00–35.9 0.96–12.6 0.49–7.58 0.29–11.1 0.51–7.11 0.57–8.25 0.43–9.26 0.43–10.74 0.37–10.91 0.61–7.19 0.105 0.050 0.058 0.340 0.528 0.335 0.257 0.375 0.354 0.423 0.237 a Combined 677(CT or TT)/80GG. Combined 677CC/80(GA or AA). Combined 677(CT or TT)/80(GA or AA). b c DISCUSSION Although maternal age is the major risk factor for trisomy 21 and after age 35 the risk of bearing a child with DS increases substantially with increasing maternal age, many DS children are born to mothers aged <35 years [James et al., 1994], suggesting the existence of genetic or environmental factors responsible for the formation of disomic gametes in younger women. Our recent finding of an increased occurrence of chromosome 13 and 21 malsegregation events in peripheral lymphocytes of young mothers of DS children, aged less than 35 at conception, seems to indicate that chromosomal nondisjunction occurs also in somatic cells in such women, suggesting a generalized susceptibility to chromosomal malsegregation events taking place either in meiotic and mitotic processes [Migliore et al., 2005]. Impairments in folate metabolism resulting in pericentromeric DNA hypomethylation, which is associated with impaired segregation and aneuploidy [Fenech, 2002], have been largely studied as a risk factor for human nondisjunction and folate gene polymorphisms have been associated with an increased risk for trisomy 21 in several such studies [James et al., 1999; Grillo et al., 2002; Bosco et al., 2003]. Folate deficiency has been linked to chromosomal instability and chromosome 21 aneuploidy [Wang et al., 2004; Beetstra et al., 2005], and the genome damaging effect of folate deficiency in cultured lymphocytes is modulated by the MTHFR genotype [Kimura et al., 2004]; moreover, elevated homocysteinemia has been demonstrated to be a potent risk factor for having a child with DS in young Italian women [Bosco et al., 2003]. In our recent study, we also observed an increased frequency of binucleated cells with micronuclei (MNBN) in lymphocytes of mothers of DS children aged less TABLE V. Interaction Between MTHFR 1298A > C and RFC-1 80G > A Genotypes in Mothers of Down Syndrome Children (DS Mothers) and Control Mothers MTHFR1298/RFC1-80 Genotype Number of DS mothers (%) Number of control mothers (%) OR 95% CI P 1298AA/80GG 1298AC/80GG 1298CC/80GG AC or CC/GGa 1298AA/80GA 1298AA/80AA AA/GA or AAb 1298AC/80GA 1298AC/80AA 1298CC/80GA AC or CC/GA or AAc 64 (total) 16 (25) 10 (15.6) 1 (1.6) 11 (17.2) 11 (17.2) 6 (9.4) 17 (26.6) 13 (20.3) 5 (7.8) 2 (3.2) 20 (31.3) 87 (total) 11 (12.6) 12 (13.6) 3 (3.45) 15 (17.05) 20 (23) 12 (13.8) 32 (36.8) 20 (23) 6 (6.9) 1 (1.15) 29 (33.35) 1.0 0.57 0.23 0.50 0.38 0.34 0.36 0.45 0.57 1.37 0.47 Referent 0.18–1.79 0.02–2.50 0.17–1.50 0.13–1.09 0.09–1.19 0.14–0.96 0.16–1.26 0.14–2.35 0.11–17.09 0.18–1.23 0.337 0.227 0.219 0.073 0.093 0.041 0.128 0.440 0.804 0.126 Missing combinations are due to the absence of DS mothers and/or control mothers with that particular genotype. a Combined 1298(AA or CC)/80GG. b Combined 1298AA/80(AG or AA). c Combined1298(AC or CC)/80(GA or AA). American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1088 COPPEDÈ ET AL. than 35 at conception, compared to control mothers [Migliore et al., 2005], and folate deficiency has been recently demonstrated to increase sensitivity to radiation-induced micronuclei in human lymphocytes [Beetstra et al., 2005]. Results of the present study are in agreement with previous studies indicating that conditions affecting the folate status and/or homocysteine levels may underlie susceptibility to nondisjunction and therefore trisomy 21 [Bosco et al., 2003; Wang et al., 2004; Beetstra et al., 2005]. In agreement with previous studies by some of us and others performed in Italian and other European populations the MTHFR 677T allele alone was not associated with an increased risk for having a DS child [Chadefaux-Vekemans et al., 2002; Stuppia et al., 2002; Chango et al., 2005]; moreover, when the combined MTHFR 677/1298 genotypes were considered, we failed to find any significant association with the risk of a DS offspring, and again the present results are consistent with previous findings in European populations [Bosco et al., 2003; Chango et al., 2005]. The two polymorphisms 677C>T and 1298A>C in the MTHFR gene are known to be in linkage disequilibrium (LD), with a measured D0 value of 0.945 in Europe [Shi et al., 2003]; particularly the 677T allele has been nearly always observed in cis with the 1298A allele, and the 677C allele has been more often observed in cis with the 1298T allele [Shi et al., 2003]. However, LD is not complete, and the presence of some individuals with the 677TT/ 1298AC genotypes in large sample-size studies [Parle-McDermott et al., 2003; Shi et al., 2003], indicates that the frequency of the rare MTHFR 677T/1298C haplotype among Europeans is not zero, and that the two alleles have been separated during evolution. In the present study, LD is indicated by the deviation of the observed frequencies of MTHFR 677/1298 combinations (Table III) from those expected assuming independence between the two polymorphisms, derived by multiplying genotype frequencies (data not shown); moreover, we did not encounter individuals with the combined MTHFR 677TT/1298AC or MTHFR 677TT/1298CC genotype, and this is probably due to the small sample size of the study population that has not allowed to observe the rare 677T/1298C haplotype. Several studies have been published that aim to evaluate the role of the MTHFR gene polymorphisms in the risk of DS, and results are often conflicting [James et al., 1999; Chadefaux-Vekemans et al., 2002; O’Leary et al., 2002; Stuppia et al., 2002; Boduroglu et al., 2004; Chango et al., 2005; da Silva et al., 2005]. One limitation of all those studies, and of the present, is the small size of young DS mothers and control mothers analyzed, that reduces the power of the statistical analyses. This is mainly due to the low frequency of young mothers of DS individuals, and, sometimes, to socio-psychological or other difficulties in their recruitment for such a kind of study. In fact, it is not a coincidence that all the studies performed to date to test for association between folate gene polymorphisms and the risk of a DS offspring, have been conducted in a number of DS mothers ranging from 31 to 157 [Hobbs et al., 2000; Takamura et al., 2004], and that even in the largest studies several DS mothers were aged more than 35 years at conception [Boduroglu et al., 2004; Chango et al., 2005; da Silva et al., 2005]. Using data and results of the present study, obtained in a total of 190 subjects, we have estimated that an adequately powered case-control study to determine only the role of the MTHFR 677C>T polymorphism in the risk of a DS offspring should include more than 850 subjects, and the number of individuals should be more than doubled to include other polymorphisms and interactions; such a number could probably be obtained by pooling together DNA samples from different laboratories. To give one example, one of us recently participated in two of the largest studies performed to date in order to clarify the contribution of polymorphisms in five folate-metabolizing genes to the risk of nonHodgkin lymphoma (NHL): the first study was conducted on 1,068 individuals from Northern California [Skibola et al., 2004], the latter on 1,344 English subjects [Lightfoot et al., 2005]; however, despite the large number of subjects included, when the two studies were compared, results were still controversial for several of the studied polymorphisms [Lightfoot et al., 2005]; we concluded that, to better clarify the potential role of folate metabolism in NHL risk, there is the need for larger and more comprehensive international studies [Lightfoot et al., 2005]. We think that similar studies would be helpful also to clarify the role of folate gene polymorphisms as potential risk factors to have a child with DS. However, results obtained in small sample-size populations are useful, they can both underline particular associations in a defined population or be indicative of a more general potential association so that, when evaluating together results produced to date, some general information can be drawn. For example, the MTHFR gene polymorphisms, mainly the 677C > T, have been associated with an increased risk of having a DS child in the majority of the studies performed both in Northern [James et al., 1999; Hobbs et al., 2000] and Southern America [Grillo et al., 2002; Acacio et al., 2005; da Silva et al., 2005], while all the studies performed in Mediterranean populations [Chadefaux-Vekemans et al., 2002; Stuppia et al., 2002; Bosco et al., 2003; Boduroglu et al., 2004; Chango et al., 2005] and the one in a Japanese population [Takamura et al., 2004], failed to find such an association. An association with the risk was observed for the combined presence of the MTHFR 677T polymorphism and the methionine synthase reductase variant (MTRR 66G) in Northern American and Irish populations [Hobbs et al., 2000; O’Leary et al., 2002]. The contrast of these results American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a FOLATE METABOLISM AND THE RISK OF DOWN SYNDROME could be due to variations in allele frequencies among different populations or to different sizes of the case-control studies; however, it could be attributable to the contribution of geographic environmental factors that further requires clarification, for example, increased homocysteine levels have been observed in DS mothers compared to control mothers in several populations, including Americans and Europeans [Bosco et al., 2003; da Silva et al., 2005]. Several previous studies aimed to clarify the role of the combined MTHFR 677/1298 genotypes on plasma homocysteine levels reported a major role for the 677T allele in affecting homocysteine concentrations, whereas the 1298A>C polymorphism seemed not to influence plasma homocysteine levels [Dekou et al., 2001; Strandhagen et al., 2004]; however, a correlation between the MTHFR 677TT genotype and homocysteine levels has not been observed in the Italian study conducted by Bosco et al. [2003], and similar results have been obtained in a French population study [ChadefauxVekemans et al., 2002], suggesting that the effect of the Mediterranean diet could counteract the individual effect of the MTHFR 677C>T polymorphism. Similar results have been obtained in a Japanese population study, and the Japanese is another folaterich population [Takamura et al., 2004]. Even if the contribution of the MTHFR polymorphisms alone seems to be not relevant to the risk of DS in European populations [Chadefaux-Vekemans et al., 2002; O’Leary et al., 2002; Bosco et al., 2003; Boduroglu et al., 2004; Chango et al., 2005], other folate gene polymorphisms and interactions between polymorphisms in the folate metabolic pathway have emerged as risk factors for a DS pregnancy in different European populations [O’Leary et al., 2002; Bosco et al., 2003], so that the role of the folate metabolic pathway cannot be excluded in the risk of DS pregnancy. Further studies aimed to explore the contribution of known polymorphisms in other genes of the same metabolic pathway are required. We included the RFC-1 80G>A polymorphism in the present study because there is evidence that some mothers of infants with DS have abnormal folate and methyl metabolism similar to mothers of infants with NTD [Botto and Yang, 2000], and sometimes DS and NTD occur in the same family, suggesting that, at least in a proportion of cases, DS and NTD could have a common etiological pathway [Barkai et al., 2003]. The RFC-1 GG genotype was associated with the risk of NTD [De Marco et al., 2003]; moreover, increased total homocysteine levels have been observed in subjects with the MTHFR 677TT/RFC-1 80GG genotype when compared with those with the 677CC/80GG genotype, and higher plasma folate levels in individuals who were RFC-1 80AA compared with those who were GG [Chango et al., 2005]. To the best of our knowledge, the RFC-1 80G>A polymorphism has never been studied alone 1089 or in combination with MTHFR polymorphisms, as a risk factor for a DS pregnancy in young Italian women. Present results seem to indicate that the RFC1 80G>A polymorphism is not an independent risk factor for DS, and are consistent with those recently obtained in a French population study of similar size, the only one to date, evaluating the contribution of the RFC-1 80G>A polymorphism to the risk of DS [Chango et al., 2005]. In the present study we observed a borderline significant fivefold increased risk (P ¼ 0.05) for having a DS child in mothers bearing the 677TT/80GG genotype compared with mothers bearing the 677CC/80GG genotype. Although not statistically significant, there appeared to be a trend, in the present study, towards higher risk for having a DS child in individuals who were TT for the MTHFR 677 polymorphism as compared to those who were CT or CC in individuals, who were GG for the RFC-1 80 polymorphism (OR ¼ 6, 3, 1, respectively). Interestingly, even if not statistically significant, Chango et al. [2005] observed an increased frequency of certain MTHFR677C>T/ RFC-1 80G>A combined genotypes in French mothers of DS individuals compared to control mothers. Both the present and the French study seem to indicate that a role of the MTHFR677C>T/ RFC-1 80G>A combined genotypes in the risk of DS cannot be excluded and additional studies are required. Moreover, we observed a significant inverse association with the risk of a DS pregnancy for the combined MTHFR1298AA/RFC-180(GA or AA) genotypes compared with the 1298AA/80GG genotype (OR ¼ 0.36, P ¼ 0.04). The result is interesting, mainly if considered together with similar previous findings of an effect of the combined MTHFR1298/RFC-180 genotypes in the risk of NTD in the Italian population [De Marco et al., 2003]. The combined effect of these two polymorphisms on the folate status requires further clarification. However, because of the incomplete linkage between MTHFR 677 and 1298 polymorphisms, there is the need of a larger and adequately powered study to evaluate the contribution of combinations of the three MTHFR677/MTHFR1298/RFC-180 polymorphisms to the risk of a DS pregnancy. Data obtained both at interview and from medical records indicate that subjects included in the present study were not under particular dietary restrictions during pregnancy. Moreover, the Mediterranean diet in regions of central Italy is usually rich in fruit and green vegetables providing an adequate amount of folate and vitamins, so that folic acid fortification is not common. Despite these considerations, one limitation of the present study is that blood homocysteine and folate values were not available from the majority of DS mothers and control mothers at the birth of their children, in order to study whether or not any effect of the MTHFR/RFC-1 combined genotypes on their levels, similar to those already American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a 1090 COPPEDÈ ET AL. reported in literature [Chango et al., 2000], was observable in our study population. In conclusion, the present study is the largest performed in an Italian population to evaluate the role of folate gene polymorphisms to the risk of DS, and one of the largest performed to date in Europe taking into account only mothers aged less than 35 at the time of DS pregnancy outcome. Present results, in agreement with previous findings in Mediterranean populations, seem to indicate that none of the studied MTHFR 677C>T, MTHFR 1298A>C, and RFC-1 80G>A polymorphisms is an independent risk factor for a DS offspring at a young maternal age; however, they are also indicative for a possible role of combined MTHFR/RFC-1 genotypes in the risk of DS pregnancies in young Italian women. It is important to stress that results here provided have been obtained in a relatively small sample-size population, so that they must not be considered conclusive, but only indicative of possible associations that require confirmation in largest and adequately powered designed studies. ACKNOWLEDGMENTS Authors thank Francesca Migheli and Cosima Natola for their help in processing DNA samples. A special acknowledgement is due to both DS mothers and control mothers for donating their blood for the present study. REFERENCES Acacio GL, Barini R, Bertuzzo CS, Couto EC, Annichino-Bizzacchi JM, Junior WP. 2005. Methylenetetrahydrofolate reductase gene polymorphisms and their association with trisomy 21. Prenat Diagn 25:1196–1199. Antonarakis SE. 1998. Down Syndrome. In: Jameson JL, editor. Principle of molecular medicine. Totowa, NJ: Humana Press, Inc. p 1069–1078. Antonarakis SE, Petersen MB, McInnis MG. 1998. The meiotic stage of nondisjunction in trisomy 21: Determination by using DNA polymorphisms. Am J Hum Genet 50:544–550. Bailey LB, Gregory JF III. 1999. Folate metabolism and requirements. J Nutr 129:779–782. Barkai G, Abruzova S, Berkenstadt M, Heifetz S, Cuckle H. 2003. Frequency of Down syndrome and neural-tube defects in the same family. Lancet 361:1331–1335. Beetstra S, Thomas P, Salisbury C, Turner J, Fenech M. 2005. Folic acid deficiency increases chromosomal instability, chromosome 21 aneuploidy and sensitivity to radiation-induced micronuclei. Mutat Res 578:317–326. Boduroglu K, Alanay Y, Koldan B, Tuncbilek E. 2004. Methylenetetrahydrofolate reductase enzyme polymorphisms as maternal risk for Down syndrome among Turkish women. Am J Med Genet Part A 127A:5–10. Bosco P, Gueant-Rodriguez RM, Anello G, Barone C, Namour F, Caraci F, Roman A, Romano C, Gueant JL. 2003. Methionine synthase (MTR) 2756 (A ! G) polymorphism, double heterozygosity methionine synthase 2756 AG/methionine synthase reductase (MTRR) 66 AG, and elevated homocysteinemia are three risk factors for having a child with Down syndrome. Am J Med Genet Part A 121A:219–224. Botto LD, Yang Q. 2000. 5,10-Methylenetetrahydrofolate reductase gene variants and congenital anomalies: A HuGE review. Am J Epidemiol 151:862–877. Chadefaux-Vekemans B, Coudè M, Muller F, Oury JF, Chabli A, Jais JP, Kamoun P. 2002. Methylenetetrahydrofolate reductase polymorphism in the etiology of Down syndrome. Pedriat Res 51:766–767. Chango A, Emery-Fillon N, de Courcy GP, Lambert D, Pfister M, Rosenblatt DS, Nicolas JP. 2000. A polymorphism (80G ! A) in the reduced folate carrier gene and its associations with folate status and homocysteinemia. Mol Genet Metab 70:310– 315. Chango A, Fillon-Emery N, Mircher C, Blehaut H, Lambert D, Herbeth B, James SJ, Rethore MO, Nicolas JP. 2005. No association between common polymorphisms in genes of folate and homocysteine metabolism and the risk of Down’s syndrome among French mothers. Br J Nutr 94:166– 169. da Silva LR, Vergani N, Galdieri Lde C, Ribeiro Porto MP, Longhitano SB, Brunoni D, D’Almeida V, Alvarez Perez AB. 2005. Relationship between polymorphisms in genes involved in homocysteine metabolism and maternal risk for Down syndrome in Brazil. Am J Med Genet Part A 135A:263–267. De Marco P, Calevo MG, Moroni A, Merello E, Raso A, Finnell RH, Zhu H, Andreussi L, Cama A, Capra V. 2003. Reduced folate carrier polymorphism (80A ! G) and neural tube defects. Eur J Hum Genet 11:245–252. Dekou V, Whincup P, Papacosta O, Ebrahim S, Lennon L, Ueland PM, Refsum H, Humphries SE, Gudnason V. 2001. The effect of the C677T and A1298C polymorphisms in the methylenetetrahydrofolate reductase gene on homocysteine levels in elderly men and women from the British regional heart study. Atherosclerosis 154:659–666. Epstein CJ. 1995. Down syndrome (trisomy 21). In: Stansbury JB, Wyngarden JB, Fredrickson DS, editors. The metabolic and molecular bases of inherited disease, 7th edition. New York: McGraw-Hill. p 749–795. Fenech M. 2001. The role of folic acid and vitamin B12 in genomic stability of human cells. Mutat Res 475:56–67. Fenech M. 2002. Recommended dietary allowances (RDAs) for genomic stability. Mutat Res 480-481:51–54. Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, Boers GJ, den Heijer M, Kluijtmans LA, van den Heuvel LP, Rozen R. 1995. A candidate genetic risk factor for vascular disease: A common mutation in methylenetetrahydrofolate reductase. Nat Genet 10:111–113. Gaulden ME. 1992. Maternal age effect: The enigma of Down syndrome and other trisomic conditions. Mutat Res 296:69–88. Grillo LB, Acacio GL, Barini R, Pinto W Jr, Bertuzzo CS. 2002. Mutations in the methylene-tetrahydrofolate reductase gene and Down syndrome. Cad Saude Publica Rio de Janeiro 18:1795–1797. Gueant JL, Gueant-Rodriguez RM, Anello G, Bosco P, Brunaud L, Romano C, Ferri R, Romano A, Candito M, Namour B. 2003. Genetic determinants of folate and vitamin B12 metabolism: A common pathway in neural tube defect and Down syndrome? Clin Chem Lab Med 41:1473–1477. Hobbs CA, Sherman SL, Yi P, Hopkins SE, Torfs CP, Hine RJ, Pogribna M, Rozen R, James SJ. 2000. Polymorphisms in genes involved in folate metabolism as maternal risk factors for Down syndrome. Am J Hum Genet 67:623–630. James LS, Erickson D, McLearn A. 1994. Prevalence of birth defects. Birth outcomes: From data to action. CDC public health surveillance for women, infants, and children. Atlanta, GA: Centers for Disease Control and Prevention. p 203–216. James SJ, Pogribna M, Pogribny IP, Melnyk S, Hine RJ, Gibson JB, Yi P, Tafoya DL, Swenson DH, Wilson VL, Gaylor DW. 1999. Abnormal folate metabolism and mutation in the methylenetetrahydrofolate reductase gene may be maternal risk factors for Down syndrome. Am J Clin Nutr 70:495–501. Kimura M, Umegaki K, Higuchi M, Thomas P, Fenech M. 2004. Methylenetetrahydrofolate reductase C677T polymorphism, folic acid and riboflavin are important determinants of genome stability in cultured human lymphocytes. J Nutr 134:48–56. American Journal of Medical Genetics Part A: DOI 10.1002/ajmg.a FOLATE METABOLISM AND THE RISK OF DOWN SYNDROME Lejeune J, Gautier M, Turpin R. 1959. Etude des chromosomes somatiqaues de neuf enfants mongoliens. Compte Rendu d’Acad Sci 248:1721–1722. Lightfoot TJ, Skibola CF, Willet EV, Skibola DR, Allan JM, Coppedè F, Adamson PJ, Morgan GJ, Roman E, Smith MT. 2005. Risk of non-Hodgkin lymphoma associated with polymorphisms in folate-metabolizing genes. Cancer Epidemiol Biomarkers Prev 14:2999–3003. Migliore L, Boni G, Bernardini R, Trippi F, Colognato R, Fontana I, Coppedè F, Sbrana I. 2005. Susceptibility to chromosome malsegregation in lymphocytes of women who had a Down syndrome child in young age. Neurobiol Aging (in press). O’Leary VB, Parle-McDermott A, Molloy AM, Nirke PN, Johnson Z, Conley M, Scott JM, Mills JL. 2002. MTRR and MTHFR polymorphism: Link to Down syndrome? Am J Med Genet 15:151–155. Olsen CL, Cross PK. 1997. Trends in the use of prenatal diagnosis in New York State and the impact of biochemical screening on the detection of Down syndrome. Prenat Diagn 17:1113–1124. Parle-McDermott A, Mills JL, Kirke PN, O’Leary VB, Swanson DA, Pangilinan F, Conley M, Molloy AM, Cox C, Scott JM, Brody LC. 2003. Analysis of the MTHFR 1298A > C and 677C > T polymorphisms as risk factors for neural tube defects. J Hum Genet 48:190–193. Rady PL, Szucs S, Matalon RK, Grady J, Hudnall SD, Kellner LH, Nitowsky H. 2001. Genetic polymorphism (G80A) of reduced folate carrier gene in ethnic populations. Mol Genet Metab 73:285–286. Schupf N, Kapell D, Lee JH, Ottman R, Mayeux R. 1994. Increased risk of Alzheimer’s disease in mothers of adults with Down’s syndrome. Lancet 344:353–356. 1091 Shi M, Caprau D, Romitti P, Christensen K, Murray JC. 2003. Genotype frequencies and linkage disequilibrium in the CEPH human diversity panel for variants in folate pathway genes MTHFR, MTHFD, MTRR, RFC1 and GCP2. Birth Defect Res 67:545–549. Skibola CF, Forrest MS, Coppedè F, Agana L, Hubbard A, Smith MT, Bracci PM, Holly EA. 2004. Polymorphisms and haplotypes in folate-metabolizing genes and risk of non-Hodgkin lymphoma. Blood 104:2155–2162. Strandhagen E, Zetterberg H, Aires N, Palmer M, Rymo L, Blennow K, Landaas S, Thelle DS. 2004. The methylenetetrahydrofolate reductase C677T polymorphism is a major determinant of coffee-induced increase of plasma homocysteine: A randomized placebo controlled study. Int J Mol Med 13:811–815. Stuppia L, Gatta V, Gaspari AR, Antonucci I, Morizio E, Calabrese G, Palka G. 2002. C677T mutation in the 5,10-MTHFR gene and risk of Down syndrome in Italy. Eur J Hum Genet 10:388– 390. Takamura N, Kondoh T, Ohgi S, Arisawa K, Mine M, Yamashita S, Aoyagi K. 2004. Abnormal folic acid-homocysteine metabolism as maternal risk factors for Down syndrome in Japan. Eur J Nutr 43:285–287. Wang X, Thomas P, Xue J, Fenech M. 2004. Folate deficiency induces aneuploidy in human lymphocytes in vitro-evidence using cytokinesis-blocked cells and probes specific for chromosomes 17 and 21. Mutat Res 551:167–180. Weisberg I, Tran P, Christensen B, Sibani S, Rozen R. 1998. A second genetic polymorphism in methylenetetrahydrofolate reductase (MTHFR) associated with decreased enzyme activity. Mol Genet Metab 64:169–172.