Family-based association study of the MTHFR polymorphism C677T in the bladder-exstrophy-epispadias-complex^†

To the Editor:

Periconceptional folate supplementation has been shown to reduce the risk of several isolated midline defects, for example, neural tube defects (NTDs), cleft lip and palate (CLP), and omphalocele [Czeizel and Dudas, 1992; Botto et al., 2002; van Rooij et al., 2004]. There is considerable evidence that the impact of periconceptional folate intake on pregnancy outcome is modified by variants in both maternal and fetal genes coding for critical enzymes in the folate pathway [Botto and Yang, 2000]. Among these enzymes, special attention has been drawn to the 5,10-methyltetrahydrofolate reductase (MTHFR) enzyme, which catalyzes the synthesis of 5-methylenetetrahydrofolate [van der Put et al., 1995]. A common polymorphism in the MTHFR gene (C → T at nucleotide position 677, resulting in substitution of valine for alanine in position 222 of the protein), is associated with a thermolabile enzyme and decreased specific MTHFR activity (activity reduced by ∼35% [CT] and ∼70% [TT], respectively) [Frosst et al., 1995]. Previous studies have suggested association for the homozygous MTHFR 677TT genotype and the occurrence of non-syndromic cases with NTDs, CLP, and omphalocele [van der Put et al., 1995; Botto and Yang, 2000; Mills et al., 2005]. However, other studies could not confirm that the 677C → T polymorphism is a genetic risk factor for human NTDs and CLPs [Shaw et al., 1998; Botto and Yang, 2000].

In the present study, we investigated the MTHFR 677C → T polymorphism for association with the occurrence of the isolated non-syndromic bladder-exstrophy-epispadias complex (BEEC), an anterior midline defect with variable expression involving the anterior abdominal wall, the pelvis, urinary tract, and external genitalia [Ludwig et al., 2005]. Exstrophy of the bladder (EB) represents the most common form in the BEEC spectrum, ranging from epispadias as the mildest to exstrophy of the cloaca (EC) as the most severe form [Carey, 2001]. The latter is often also referred to as the OEIS complex, comprising omphalocele, exstrophy, imperforate anus, and spinal defects [Carey, 2001]. The incidence of isolated EB among European descendents varies between 1:20,000 and 1:80,000, with males being affected more often than females [ICBDMS, 1987; Ludwig et al., 2005]. Current published data suggest that a complex genetic mode of inheritance underlies the BEEC, with no genetic or environmental factors having been identified so far in humans [Boyadjiev et al., 2004]. To the best of our knowledge, only five molecular genetic studies investigating genetic factors in the development of BEEC have been carried out to date [Boyadjiev et al., 2004, 2005; Thauvin-Robinet et al., 2004; Reutter et al., 2006a,b] and only a few chromosomal aberrations have been reported [for review: Ludwig et al., 2005]. However, none of these molecular and cytogenetic investigations revealed any BEEC-specific candidate genes. The largest molecular genetic study to date included a total of 33 BEEC patients [Reutter et al., 2006a].

For the present investigation, 91 patients with BEEC and their families have been recruited from Germany, Austria, and Spain, comprising two different ethnicities, Central European (n = 63) and Spanish (n = 28), including 68 parent–offspring triads (mother, father, affected BEEC child) and 23 mother–child pairs. The phenotypic spectrum included isolated epispadias (7 patients), EB (72 patients), and EC (12 patients). For the case-control analysis of the maternal genotypes, 93 unrelated healthy individuals were chosen as a control group from the German population. For the Spanish group, genotype data of controls were taken from the literature [Guillen et al., 2001]. None of the 28 Spanish mothers and only 7 (11%) of the 63 German and Austrian mothers had taken any form of folate supplementation during their periconceptional period. The study design was approved by the ethics committee of the University of Bonn and informed consent was obtained from patients, parents, and control individuals. To analyze the 677C → T variant in the MTHFR gene, exon 4 of the gene was amplified by polymerase chain reaction (PCR) following standard conditions and with the use of modified primers (4F: 5′-TCTTCATCCCTCGCCTTGAAC-3′; 4R: 5′-AGGACGGTGCGGTGAGAGTG-3′) according to Frosst et al. 1995. The digested PCR products were analyzed on 1.5% agarose gels.

In order to test for association of the MTHFR 677C → T polymorphism with the BEEC, we applied the transmission/disequilibrium test (TDT) described by Spielman et al. 1993 on 68 parent–offspring triads and 23 mother–child pairs. The observed distributions of genotypes among patients, parents, and controls were consistent with Hardy–Weinberg equilibrium (data not shown). No significant deviation from random transmission to children with the BEEC was observed (P = 0.90) in the 58 trios with an informative genotype (Table I). Furthermore, the investigation of subgroups revealed no biased transmission of the MTHFR alleles to either EB (P = 0.79) or other forms of BEEC (P = 1.00) (Table I). To test if the maternal genotype might be associated with the child's risk of developing phenotypic expressions of the BEEC spectrum, a case-control study was performed in a second step, using 91 mothers selected from the triads and mother–child pairs. By grouping the genotypes as has been previously suggested [van der Put et al., 1995], the TT genotype was significantly overrepresented among Spanish mothers when tested against Spanish controls (P = 0.022). After applying the Bonferroni correction [Pocock et al., 1987] for multiple testing, the difference did not reach significance (P = 0.154). No significant differences were observed between German mothers and ethnically matched controls (P = 0.94). This result was also negative when we restricted the analysis to those German mothers who had not received periconceptional folate supplementation (P = 0.65) (Table II).

In a sample of 23 patients, Mills et al. 2005, applying a case-control method in the EC-related phenotype, omphalocele, were able to show association for the MTHFR 677CT and TT genotypes with this anomaly. Our results, however, in a comparatively large sample were not supportive of the tested hypothesis (Table I). If the original finding of Mills et al. 2005 was correct, this result might suggest that different genetic factors contribute to isolated omphalocele than to omphalocele occurring in association with the EC.

Although we did not find a significant association of the maternal genotype with the child's risk of developing phenotypic expressions of the BEEC spectrum, the results raise the question of possible differences between the Spanish and German populations that might have an impact on the occurrence of BEEC. If the maternal genotype confers risk only in combination with population specific environmental or genetic factors, one might expect to find association only in certain populations. Periconceptional folate supplementation is a well-established modifier of the child's risk for developing a midline defect. None of the Spanish mothers had reported use of periconceptional folate supplementation, which would be in accordance with this hypothesis. However, the majority of the German mothers received no periconceptional folate supplementation either, and if analysis is restricted to these cases, the results are similarly negative.

In conclusion, we found no effect of the children's MTHFR genotype on the development of BEEC. An effect of the maternal genotype, depending on the interaction with other population-specific risk factors, has not received firm statistical support in our study. However, we acknowledge that the limited size of our sample might have prevented detection of smaller effects and that investigation of larger samples is necessary in future research.