Infant C677T MTHFR polymorphism and severe mental retardation
BACKGROUND: We investigated whether infants with homozygous genotype TT of the MTHFR gene were at increased risk of severe mental retardation. METHODS: One hundred children with severe mental retardation (cases) were investigated from a large geographic-based study of infants born in California in 1992–1993. Cases were compared to 743 randomly selected nonmalformed control infants born in California during 1987–1991. DNA was extracted from newborn screening filter papers. Cases and controls were genotyped TT if homozygous for the MTHFR C677T allele, CT if heterozygous for the C677T allele, and CC if homozygous for the C677 (wild type) allele. RESULTS: Overall, case and control infants had similar percentages of TT and CT genotypes. Percentages between cases and controls differed somewhat across race/ethnic groups. Elevated ORs of 1.9 (95% CI: 0.7–5.0) and 2.6 (95% CI: 1.1–5.8) were observed for the TT and CT genotypes, respectively, among Hispanic children. Observed results were not substantially altered for analyses that removed 41 case children who also had structural birth defects. CONCLUSIONS: Folate-related mechanisms are important to investigate for etiologies of birth defects, and such lines of inquiry may be revealing for mental retardation given the relationships between mental retardation and birth defects and potential relationships between folate, DNA methylation, and mental retardation. Birth Defects Research (Part A), 2007. © 2006 Wiley-Liss, Inc.
Folate is key to the development of the central nervous system (Rozen, 1996; Kluijtmans et al., 1998; Botto et al., 1999), with rare congenital errors of folate metabolism known to be precursors to mental retardation (Zittoun, 1995). Down syndrome, an important condition leading to mental retardation, has been associated with polymorphisms in genes involved in folate metabolism (James et al., 1999; Hobbs et al., 2000). One gene variant in particular, the C677T/C677T genotype, of the 5,10 methylenetetrahydrofolate reductase (MTHFR) gene, has been associated with Down syndrome as well as being a risk factor for other complex congenital anomalies (Botto and Yang, 2000). Moreover, a recent study reported an association between MTHFR and intelligence quotient in patients with Down syndrome (Gueant et al., 2005). The variant form of MTHFR codes for a thermolabile enzyme with reduced activity and has been linked to elevated plasma homocysteine levels in C677T individuals under conditions of folate deficiency (Frosst et al., 1995). Because of MTHFR's involvement with the metabolism of folate and observed associations with Down syndrome and intelligence quotient, we hypothesized that infants homozygous for the C677T genotype would be at increased risk for severe mental retardation. Thus, we investigated whether those infants with homozygous genotype TT of the MTHFR gene were at increased risk of severe mental retardation.
MATERIALS AND METHODS
Cases of children with mental retardation (cases) were derived from a large geographic-based study of infants born in nonmilitary hospitals in selected areas of California in 1992–1993 and who survived to 1 year of age. Details about the ascertainment of these infants can be found elsewhere (Jelliffe-Pawloski et al., 2003). Briefly, information about mental retardation was obtained from the California Department of Developmental Services. Individuals with developmental disabilities including mental retardation receive services from birth until death from this agency. The agency's service delivery charts were reviewed for all 893 children born in 1992–1993 and who were receiving services in March 2001. Considered to have mental retardation were 635 children (American Psychiatric Association, 2000). Excluded were 22 children who had normal intellectual functioning prior to severe head trauma, near drowning, or stroke. Of the 613, 285 we categorized as having “severe” mental retardation and 318 as having “mild” mental retardation. Also excluded were 10 children who were found to have mental retardation based on maladaptive functioning, but who did not have specific diagnoses for adequate information to determine the severity of the mental retardation. Meeting our definition of “severe” were 213 children who were tested as having an intelligence quotient <50 and an additional 72 children who did not have standardized testing within the past 3 years but were consistent with adaptive functioning criteria for severe mental retardation (American Association of Mental Deficiency, 1983; American Association on Mental Retardation, 2002) and had a diagnosis of moderate, severe, or profound mental retardation from a licensed psychologist. The severe group was the focus of the current study. These 285 children were further linked to information about birth defects derived from the California Birth Defects Monitoring Program (Croen et al., 1991). This population-based active surveillance system collects diagnostic information from multiple sources of medical records on all stillborn and liveborn infants with structural birth defects diagnosed within 1 year of delivery in California counties.
Control infants were derived from two previously conducted studies described in detail elsewhere (Shaw et al., 1998a, b; Rozen et al., 1999). These infants were randomly selected from among nonmalformed births from each surveillance area hospital in proportion to the hospital's estimated contribution to the total population of infants born alive in a given month from June 1989 to May 1991 or were randomly selected from all nonmalformed infants born alive between 1987 and 1989.
DNA used in genotyping experiments was extracted from newborn screening filter papers by standard laboratory procedures and was amplified using PCR. Genotyping for MTHFR was performed by restriction digestion of PCR products with HinfI (Frosst et al., 1995). Cases and controls were genotyped TT if homozygous for the C677T allele, CT if heterozygous for the C677T allele, and CC if homozygous for the C677 (wild-type) allele. Because of the exploratory nature of this investigation, we randomly selected 100 of the 285 cases so as to conserve DNA for subsequent investigations. Genotyping for 743 controls was performed previously and is described elsewhere (Shaw et al., 1998a, b).
Risks of severe mental retardation were estimated by ORs and their 95% CIs. Considered as covariates in some analyses were maternal race/ethnicity (Hispanic, White non-Hispanic, Black, other) and gender.
Overall, case and control infants had similar percentages of TT and CT genotypes (Table 1). ORs indicated no substantial increased risk for severe mental retardation associated with the CT or TT MTHFR genotype. Percentages between cases and controls differed somewhat across race/ethnic groups. Elevated ORs of 1.9- and 2.6-fold were observed for the TT and CT genotypes among Hispanic children (Table 1). These elevated risks, however, may have arisen as a result of too few case children with the CC genotype, rather than an excess of children with the T allele. The lack of CC individuals may have occurred because of random variation or may have unknown biologic relevance. The 19.6% of Hispanic case children with the CC genotype is, nevertheless, within the range of frequencies reported in native Mexican populations (Davalos et al., 2000).
Table 1. Infant Methylenetetrahydrofolate Reductase (MTHFR) Genotype and Risk (OR) for Severe Mental Retardation by Race/Ethnic Group
|All race/ethnic groups|
| CC||39 (39.0)||314 (42.3)||Ref|| |
| CT||48 (48.0)||330 (44.4)||1.2||0.7–1.8|
| TT||13 (13.0)||99 (13.3)||1.1||0.5–2.1|
| CC||9 (19.6)||83 (36.2)||Ref|| |
| CT||27 (58.7)||97 (42.4)||2.6||1.1–5.8|
| TT||10 (21.7)||49 (21.4)||1.9||0.7–5.0|
| CC||15 (45.5)||171 (40.6)||Ref|| |
| CT||15 (45.5)||206 (48.9)||0.8||0.4–1.7|
| TT||3 (9.1)||44 (10.5)||0.8||0.2–2.8|
| CC||9 (81.8)||14 (60.9)||Ref|| |
| CT||2 (18.2)||6 (26.1)||0.5||0.1–3.2|
| TT||0 (0)||3 (13.0)||—|| |
|Other race/ethnic group|
| CC||6 (60.0)||46 (65.7)||Ref|| |
| CT||4 (40.0)||21 (30.0)||1.5||0.4–5.7|
| TT||0 (0)||3 (4.3)||—|| |
The overall observed results were not substantially altered for analyses that removed 41 (22 with Down syndrome) case children who also had a structural birth defect (OR = 1.4 [CI: 0.8–2.7] for CT; and 0.90 [CI: 0.3–2.5] for TT) and were not substantially altered for analyses that specifically removed the 22 case children with Down syndrome (OR = 1.2 [CI: 0.7–2.1] for CT; and 1.0 [CI: 0.4–2.2] for TT).
Because of a substantial reduction in newborn female infants, those homozygous for TT has been observed previously (Rozen et al., 1999), we further explored analyses stratified by sex. Combining all race/ethnic groups did not reveal substantial elevations in estimated risks for severe mental retardation for males or females associated with the TT or CT genotypes. Analyses stratified by sex and maternal race/ethnicity (Hispanic and White non-Hispanic only) revealed that all 10 Hispanic case children with the TT genotype were males (OR = 5.4 [95% CI: 1.8–16.3]), whereas one in three White non-Hispanic case children with the TT genotype was male (OR = 0.2 [CI: 0.02–1.5]). Thus, numbers were too sparse to derive further inference.
Our study is the first to investigate variants of this folate-related gene as possible risk factors for severe mental retardation. The observed findings of no overall increased risk for severe mental retardation among children homozygous for the C677T genotype and a possible association specifically among Hispanic male children will require further study before firm inferences can be drawn.
The prevalence of severe mental retardation in the general population is ∼3.0 per 1,000 live births, and causes remain unknown for a large proportion of cases. A paucity of epidemiologic studies has investigated genetic associations likely to be involved in the etiologies of severe mental retardation. Mechanisms underlying risk reductions associated with prenatal folate intake are important in the etiologies of many congenital anomalies, and such lines of inquiry may be revealing for mental retardation as well, given the relationships between mental retardation and birth defects (Jelliffe-Pawlowski et al., 2003; Decoufle et al., 2001) and potential relationships between folate, DNA methylation, and mental retardation (Costello and Plass, 2001).