Up to 72% of spina bifida cystica (SB) is preventable by maternal periconceptual folic acid supplementation. The C677T allele of the methylenetetrahydrofolate reductase (MTHFR) gene and some other functional polymorphisms are risk factors for SB in some populations. However, despite extensive study, the genetic risk factors for SB are incompletely understood. Polymorphic alleles that diminish bioavailability of reduced folate in the mother during pregnancy could contribute to SB in her fetus, acting in the mother as teratogenic alleles. We recently discovered a polymorphic 19 bp deletion allele (frequency 0.45) within intron-1 of dihydrofolate reductase (DHFR) that is a good candidate for such a genetic factor. Since there is precedence for intron-1 regulatory elements and the deletion allele removes a potential Sp1 transcription factor binding site, we hypothesized that the deletion allele could be functional and act in SB mothers to increase the risk of SB in her fetus. We found that homozygosity for this deletion allele was significantly more frequent in SB mothers, but not in SB fathers or patients, compared with controls and was associated with a significantly increased odds ratio (OR) (2.035) of being an SB mother compared with other genotypes. Genotype distribution obeyed the constraints of Hardy-Weinberg equilibrium in controls, SB patients and fathers, but not in SB mothers. If confirmed, these findings could lead to improved forms of folate supplementation for pregnancy. About half of dietary folates and all of folic acid supplements must be reduced by DHFR to be available for mother and fetus. Reduced folates could be preferable for supplements during pregnancy to prevent SB.

Inheritance appeared to be autosomal recessive, suggesting that mutations in nuclear DNA are responsible for mtDNA depletion. mtDNA depletion should be considered in children with mitochondrial disorders of uncertain etiology, and criteria for diagnosis are proposed.

No abstract

Fifty families (491 individuals in 137 sibships) with more than one living case of isolated, nonsyndromic spina bifida (SB) were analyzed genetically. There were twice as many gene-carrier females (56) as gene-carrier males (28) (P < 0.005). This was not an artifact of ascertainment bias because the sex ratio of gene-carriers was the same whether the pedigree was obtained through the proband's father or mother. Also, this effect was not observed in other disorders analyzed by the same method. Neither was the effect due to differential fertility because the number and sex of affected and unaffected children per gene-carrier parent were not different for male or female gene-carrier parents. There was no evidence that the missing male gene-carriers were lost by selective spontaneous abortion. There was no deficit of male-to-male or male-to-female transmission, excluding simple X-linked or simple mitochondrial inheritance. If genomic imprinting plays a role in the unequal female and male carrier frequencies in SB, penetrance should differ with parental sex. Penetrance was higher for offspring of female parents than of male parents, but the difference was not statistically significant. In addition, both male and female gene-carriers were frequently found in the same pedigree. Thus, the present data suggest a possible role for imprinting in SB.

From studies in the mouse and from the clinical and molecular analysis of patients with type 1 Waardenburg syndrome, particular members of the PAX gene family are suspected factors in the aetiology of human neural tube defects (NTD). To investigate the role of PAXI, PAX3, PAX7, and PAX9, allelic association studies were performed in 79 sporadic and 38 familial NTD patients from the Dutch population. Sequence variation was studied by SSC analysis of the paired domain regions of the PAXI, PAX7, and PAX9 genes and of the complete PAX3 gene. In one patient with spina bifida, a mutation in the PAXI gene was detected changing the conserved amino acid Gln to His at position 42 in the paired domain of the protein. The mutation was inherited through the maternal line from the unaffected grandmother and was not detected in 300 controls. In the PAX3 gene, variation was detected at several sites including a Thr/Lys amino acid substitution in exon 6. All alleles were present among patients and controls in about the same frequencies. However, an increased frequency of the rare allele of a silent polymorphism in exon 2 was found in NTD patients, but no significant association was observed (p=0.06). No sequence variation was observed in the paired domain of the PAX7 and PAX9 genes. Our findings so far do not support a major role of the PAX genes examined in the aetiology of NTD. However, the detection of a mutation in PAXI suggests that, in principle, this gene can act as a risk factor for human NTD. (JMed Genet 1996;33:655-660)