July 2006 ⅐ Vol. 8 ⅐ No. 7 article Analysis of seven maternal polymorphisms of genes involved in homocysteine/folate metabolism and risk of Down syndrome offspring Iris Scala, MD1, Barbara Granese, PhD1, Maria Sellitto, MD1, Serena Salome`, MD1, Annalidia Sammartino, MD2, Antonio Pepe, PhD1, Pierpaolo Mastroiacovo, MD3, Gianfranco Sebastio, MD1, and Generoso Andria, MD1 PURPOSE: We present a case-control study of seven polymorphisms of six genes involved in homocysteine/folate pathway as risk factors for Down syndrome. Gene-gene/allele-allele interactions, haplotype analysis and the association with age at conception were also evaluated. METHODS: We investigated 94 Down syndrome-mothers and 264 control-women from Campania, Italy. RESULTS: Increased risk of Down syndrome was associated with the methylenetetrahydrofolate reductase (MTHFR) 1298C allele (OR 1.46; 95% CI 1.02–2.10), the MTHFR 1298CC genotype (OR 2.29; 95% CI 1.06–4.96), the reduced-folate-carrier1 (RFC1) 80G allele (1.48; 95% CI 1.05–2.10) and the RFC1 80 GG genotype (OR 2.05; 95% CI 1.03–4.07). Significant associations were found between maternal age at conception Ն34 years and either the MTHFR 1298C or the RFC 180G alleles. Positive interactions were found for the following genotype-pairs: MTHFR 677TT and 1298CC/CA, 1298CC/CA and RFC1 80 GG/GA, RFC1 80 GG and methylenetetrahydrofolate-dehydrogenase 1958 AA. The 677–1298 T-C haplotype at the MTHFR locus was also a risk factor for Down syndrome (P ϭ 0.0022). The methionine-synthase-reductase A66G, the methionine-synthase A2756G and the cystathionine-beta-synthase 844ins68 polymorphisms were not associated with increased risk of Down syndrome. CONCLUSION: These results point to a role of maternal polymorphisms of homocysteine/folate pathway as risk factors for Down syndrome. Genet Med 2006:8(7):409–416. Key Words: Down syndrome, polymorphism, homocysteine, folates, gene-gene interaction Down syndrome (DS) is caused by trisomy of either the cellular methylations or both. Methylenetetrahydrofolate re- entire amount, or a critical portion of chromosome 21. Birth ductase (MTHFR) is a critical regulator of cellular methyla- prevalence increases with maternal age from 0.6 in 1,000 live tions by catalyzing the conversion of 5,10 methylenetetrahy- births at 20 years up to 11 in 1,000 live births at 40 years.1 In drofolate to 5-methyltetrahydrofolate, the methyl donor for 95% of cases, DS is caused by a meiotic error occurring at the remethylation of homocysteine (Hcy) to methionine meiosis I or II, mainly of maternal origin.2 (Met)8 (Fig. 1). This reaction is important for the synthesis of Despite huge efforts, the mechanisms underlying meiotic S-adenosylmethionine (SAM), the major intracellular methyl nondisjunction and the contribution of maternal age on tri- donor for DNA, protein and lipid methylation. The MTHFR somy 21 are poorly understood. Some studies suggest that 677T variant causes a reduced MTHFR activity and an in- genomic DNA hypomethylation may be associated with chro- creased need of folic acid to allow the remethylation of Hcy to mosomal instability and abnormal segregation.3–5 On the Met. A second common MTHFR mutation, the A1298C trans- other hand, impairment of folate metabolism has been causally version, causes decreased MTHFR activity. Both the 1298C related to both DNA hypomethylation and abnormal purine and the 677T polymorphisms may affect DNA methylation.9 synthesis.6 This combination leads to chromosomal breakage The remethylation of Hcy to Met is catalyzed by the methio- and deficient DNA repair surveillance that increase the degree nine synthase (MTR) enzyme, maintained in its functional of genomic instability.7 Polymorphisms of genes of the homo- state by the methionine synthase reductase (MTRR) reducing cysteine/folate metabolism may interfere with folate homeostasis, system. Two polymorphisms have been described in the MTR and the MTRR genes, A2756G and A66G, respectively, the former being associated with aberrant methylation in the ho- From the 1Department of Pediatrics, the 2Department of Gynecology and Obstetrics, Federico II 10 University of Naples, via S. Pansini 5, 80131 Naples, Italy; and the 3International Centre of Birth mozygous state. Cystathionine-beta-synthase (CBS) is a key Defects, via P. Albertelli 9, 00195 Rome, Italy. enzyme in the transsulfuration of Hcy to cystathionine. The Professor Generoso Andria, Department of Pediatrics, Federico II University of Naples, Via S. 844ins68 polymorphism, carried by nearly 10% of the general Pansini 5, 80131, Naples, Italy. population, may cause lower tHcy levels compared to noncar- Submitted for publication December 24, 2005. riers, in particular after methionine loading.11 A common mu- Accepted for publication April 11, 2006. tation (A80G) of the reduced-folate-carrier gene (RFC1), lead- DOI: 10.1097/01.gim.0000228206.21793.82 ing to nonsignificantly decreased plasma folate levels, has been Genetics IN Medicine 409 Scala et al. Fig. 1. The homocysteine and folates metabolic pathway. The enzymes encoded by genes investigated in the present study are indicated. MTHFR: methylenetetrahydrofolate reductase; MTRR: methionine synthase reductase; MTR: methionine synthase; CBS: cystathionine-beta-synthase; RFC1: reduced folate carrier 1; MTHFD: methylenetetrahydrofolate dehydrogenase; B12: vitamin B12;B2: vitamin B2;B6: vitamin B6. recently reported.12 Finally, a G1958A polymorphism was variants of Hcy/folate metabolism. Gene-gene/allele-allele in- identified in the methylenetetrahydrofolate dehydrogenase teractions and MTHFR haplotype analysis are also reported. (MTHFD) gene, which encodes for a trifunctional enzyme that Finally, all gene polymorphisms were analyzed according to catalyzes the conversion of tetrahydrofolate to 10-formyl, maternal age at conception to assess whether specific genetic 5,10-methenyl, and 5,10-methylenetetrahydrofolate, donor factors could contribute to the maternal age-effect on chromo- cofactors for nucleotide and DNA biosynthesis.13 somal nondisjunction. Based on this evidence, a number of studies from distinct geo- graphic regions evaluated the role of polymorphisms of genes of SUBJECTS AND METHODS Hcy/folate metabolism as maternal risk factors for DS with incon- sistent results. Specifically, the maternal MTHFR C677T poly- Study population morphism was associated with increased risk of DS in the USA- Blood samples were collected from a total of 94 women who Canadian and Brazilian populations14–17 but not in reports from had given birth to children with kariotypically confirmed full tri- Ireland, France, Turkey, Japan, and two Italian regions.18–24 Fur- somy 21. Mothers were recruited from the Department of Pedi- ther, two studies found an association between the MTRR 66GG atrics, Federico II University of Naples, the regional referral center genotype and increased risk of DS, particularly when associated for DS. A total of 264 control-women were enrolled at the same with the MTHFR C677T variant,15,18 while a study in a Sicilian University Hospital as follows: healthy women from the Blood population failed to confirm such association.23 On the other Donor Center, women referring to the Department of Gynecol- hand, the Sicilian study first reported the effect of the MTR ogy and Obstetrics for routine pregnancy checks and DNA sam- A2756G variant on the risk of DS that was stronger when associ- ples from healthy women collected at the Department of Pediat- ated with the MTRR A66G polymorphism. The MTHFR A1298C rics. Among the control group, no history of abnormal pregnancy variant was associated with increased risk of DS in one study16 outcome was reported. All case and control women belong to the while four other groups reported negative results.17,21,23,24 The same ethnicity (Caucasian). Moreover, as a geographical variation RFC1 A80G and the CBS 844ins68 polymorphic variants were not among different Italian regions has been described for the MTHFR associated with DS in one24 and in two studies,17,24 respectively. C677T polymorphism,26 only case and control women born in The MTHFD G1958A polymorphism, previously associated with the Campania region, Southern Italy, were included in the analy- increased risk of neural tube defects,25 has not been investigated as sis. To calculate the age at conception, all case-mothers were in- a risk factor for DS, yet. terviewed for date of birth, age at delivery and gestational age of Here we present a case-control study in a homogeneous the offspring. Among the control group, these data were available population from Campania, Italy, of seven polymorphic gene and/or complete for a total of 134 subjects out of 264. The mean 410 Genetics IN Medicine Maternal polymorphisms and risk of Down syndrome age at conception of case-mothers was 32.4 Ϯ 6.3 years. The mean annealing 60°C (1 minute), elongation 72°C (1 minute) for a age at conception of control-mothers was 30 Ϯ 5.6 years as calcu- total of 39 cycles. When present, the 844ins68 insertion caused lated at the time of the last pregnancy. The distribution of geno- a ϩ68bp shifted band of the PCR product (242 bp instead of type frequencies of the 134 control-mothers was comparable to 174 bp). those of the total control group. A potential bias deriving from MTR A2756G: The mutation creates a HaeIII site. To rely on hospital-based case-control studies was minimized according to an internal control, we introduced a HaeIII site by replacing an the recommendations by Wacholder et al. and Schulz et al.27,28 All A with a G (underlined)