Background

Conclusive evidence indicates that brain-derived neurotrophic factor (BDNF) plays a pivotal role in the pathophysiology of Huntington disease (HD). As the protein huntingtin (htt) directly modulates the expression of neuron-restrictive silencer factor (NRSF)-controlled genes, wild type (wt) htt stimulates the production of BDNF, whereas mutant htt causes the opposite effect [1]. It has been shown recently in transgenic mice that BDNF has an impact on the age at onset (AO) and the severity of motor dysfunction by controlling survival of striatal projection neurons [2].

The BDNF gene consists of five alternatively spliced 5' exons and one major 3' exon producing at least six BDNF transcripts leading to three pre-proprotein isoforms which differ in the lengths of the signal peptides. Sequence variations in BDNF may therefore lead to variations in gene expression or protein metabolism causing selective neuronal vulnerability. The single nucleotide polymorphism (SNP) rs6265, producing a valine-to-methionine substitution at codon 66 (Val66Met) in the human BDNF gene appears to exert an effect on intracellular trafficking and activity-dependent secretion of BDNF [3]. Furthermore, Met-BDNF carriers demonstrate substantial relative decreases in hippocampal volume, and Val/Met-BDNF affects the volume of gray matter in the cerebral neocortex of normal humans. Finally Met-BDNF is associated with volume reductions primarily in the lateral convexity of the prefrontal cortex [4]. Thus, the Val66Met polymorphism may be a modifying genetic factor in the expression of a number of normal and abnormal brain conditions, and therefore represents a good candidate gene for modifying AO in HD. In this context, several associations between BDNF polymorphisms and neurological and psychiatric disorders have been reported [5]. In addition, a recent study demonstrated that HD patients heterozygous for the Val66Met polymorphism present a later AO than homozygous carriers of Val-BDNF [6].

In this study, we investigated the relation between the BDNF gene and the AO of HD using genetic markers that represent the overall variability at this locus.

Methods

We selected tagging polymorphisms from the BDNF gene (rs6265, rs11030104, rs7103873, rs2049046 and rs12273363) based on HapMap data. We examined the associations between BDNF polymorphisms and motor AO of HD in 250 unrelated patients with clinical diagnosis of HD as recruited from the Huntington Center (HZ) NRW, Bochum (Germany) [7]. Informed consent was obtained and the institutional Ethics Committee of Bochum Ruhr-University approved this study.

The expanded CAG repeats explained 50.9% of the variance in AO in this cohort. The potential influence of certain genotypes on AO was calculated by linear regression, in which R2 illustrates the relative improvement of the regression model when the various genotypes are considered in addition to the expanded CAG block in the huntingtin gene.

Results and Discussion

LD analysis revealed strong LD between all neighboring markers. Patients carrying different genotypes and haplotypes, respectively, showed no differences in AO as evidenced in the box plot of Val66Met genotypes and AO (figure 1). Addition of BDNF genotype variations or one of the five marker haplotypes to the effect of CAG repeat lengths resulted in no significant increase of the R2 value. The same effect was evident in a sub-group of patients (n = 194) with higher variance in AO (CAG repeat range 40–45; R2 = 0.36). Observed frequencies for all five SNPs (table 1) were in Hardy-Weinberg equilibrium. Our results clearly indicate that genetic variations in the BDNF gene do not influence the variance of AO in HD in German patients in contrast to the results of Alberch and collaborators [6]. Several reasons may explain the discrepancy between these results. The previously reported association could have resulted from a type I error. Possible selection bias due to admixture may have been due to inclusion of DNA samples of relatives, the exclusion of which has not fully been detailed in the paper. Further evidence for such an assumption is implied by the very low R2 (0.11) in the interval between 42 and 49 CAG repeats. Alternatively, true association could exist only in groups of a specific ethnic origin and, presumably, a similar genetic background. While this manuscript was under review two other independent reports have been published by two different groups leading to the same conclusion indicating no association between the Val66Met genotype and variation in AO [8, 9], nor three other BDNF SNPs [9].

Figure 1
figure 1

Relationship between BDNF Val66Met (rs6265) genotypes and age at onset (AO) for 250 Huntington disease patients. For each genotype, the median AO is represented as a black bar, the quartile is shown as a solid box, and the range is indicated by the margins.

Full size image
Table 1 Allele and genotype frequencies of the BDNF polymorphisms
Full size table

Conclusion

In our association study between common polymorphisms that represent the entire variability of the BDNF gene and the variance in motor AO in HD, we were unable to verify the reported association between a single polymorphism at the BDNF gene and AO in HD.