This study represents the first evidence of an effect of the BDNF Val66Met polymorphism on protein levels during the 15-17 weeks of pregnancy.
Egan and colleagues  have shown in their seminal paper the functional relevance of the Val66Met polymorphism, demonstrating that BDNF genotype affects the intracellular trafficking of the neurotrophin, suggesting that there may be a specific 'trafficking' signal in the pro-domain of BDNF, which is required for an efficient sorting of the neurotrophin to the regulated secretory pathway. Moreover, they demonstrated that the genotype can also affect the activity-dependent release of BDNF  as recently confirmed by Chen and colleagues . The decreased intracellular BDNF trafficking and release due to the presence of the Met allele are thought to underlie several dysfunctions in humans. In fact, individuals carrying the Met allele show reduced grey matter volumes within brain regions known to participate in verbal memory and visuospatial abilities  as well as altered hippocampal volume and gray matter density in the frontal and prefrontal cortex [11, 12]. Furthermore, Stern and colleagues  have recently demonstrated a brain deficiency in N-acetyl-aspartate levels, an indicator of neuronal integrity, in subjects carrying the Met allele, suggesting that this allele could play an important role on hippocampal functionality and vitality and may have putative implications for several psychiatric disorders. To this support, a marked atrophy in several limbic regions, was reported in patients with Major Depression, Schizophrenia, Bipolar Disorder and other psychiatric disorders , suggesting the presence of a deficiency in neurogenesis and in synaptic plasticity, mechanisms highly regulated by BDNF. Moreover, the Val66Met polymorphism was also associated with alterations in amygdala volume and in its activity in response to emotional stimuli, suggesting an association between the Met allele and an increased vulnerability for anxiety disorders . This polymorphism has been extensively studied in relation to several psychiatric disorders reporting often conflicting results. In particular, both alleles have been associated with different mental disorders and to date it is unknown how the two variants may exert their influence on disease susceptibility. Probably, a mental disorders or a cognitive or brain imaging endophenotype may be associated with the presence of the Met allele, while a distinct condition may have been associated with the Val variant. At this regard a recent case-control study in psychiatric disorders  reported that the Met allele significantly increases the risk for schizophrenia and eating disorders whereas exerts a protective effect for substance-related disorders.
Furthermore, studies analyzing gene-environment interactions [30, 31] reported that BDNF Met carriers exposed to early life stress events (ELS) have smaller hippocampal and amygdala volumes, heart rate elevations, a decline in working memory and higher depression symptoms. In contrast, the combination of the BDNF Val/Val genotype and ELS is associated with increases in amygdala and prefrontal cortex grey matter and with higher anxiety symptoms.
It is well known that human brain undergoes complex morphological and functional changes during fetal development. In particular, during the first and second trimester of pregnancy, cortical subplate reaches its peak of development, and cortical neurogenesis, characterized by proliferation, migration and organization of neuronal connections, is predominantly taking place [22, 23]. Therefore, during this period the immature brain may be particularly sensitive and vulnerable to a range of intrauterine influences like stress exposure and inflammation. During the first weeks of pregnancy Amniotic Fluid derives principally from maternal plasma, getting through to fetal membranes. Later, from tenth to twentieth week, bidirectional diffusion between placenta, umbilical cord and fetal skin occurs and in this period the AF composition becomes very similar to fetal plasma. Therefore, it may be assumed that the AF BDNF content in this period reflects the total circulating neurotrophin content of fetus .
In parallel, circulating BDNF levels in fetus could reflect those in the brain since a correlation between blood BDNF levels and the cortical protein expression has been observed during neurodevelopment in rats . In this context, alterations of AF BDNF levels induced by the presence of the BDNF polymorphism in a critical period for brain development (15-17 weeks of pregnancy), in concomitance with other susceptibility genes or adverse intrauterine events, may represent a vulnerability risk factor for brain development and maturation. As previously reported, BDNF plays an important role in neural proliferation, survival and differentiation; therefore, reduced BDNF levels during CNS development could influence the correct morphology of specific cerebral regions as well as the cognitive and memory abilities that are widely reported to be affected in healthy subjects carrying Met allele. It is interesting to notice that in the adulthood the Val66Met polymorphism does not affect the levels of the neurotrophin in serum and plasma [34, 35] and to date only Lang and colleagues  reported a significant serum BDNF increase in healthy subjects carrying Met allele. However, in this study the authors did not take in account the influence of additional factors, such as drugs, life style, environment and stress-life events, which interfere with BDNF levels. Moreover, another possible explanation for the discrepancy with our results could be the specific impact of the Val66Met polymorphism on BDNF levels during brain development and not in adult life, where this effect could be hidden by confounding factors putatively associated with the life style.
Some limitations of our study have to be mentioned. The data about AF BDNF levels are from one point in time, and the relationship between BDNF genotype and AF BDNF levels may be different at other times during gestation. Additionally, this study is correlational in nature and we cannot draw any definitive statements about the consequences of BDNF Val66Met genotype on AF protein levels. Larger studies in wider and better characterized samples are needed to confirm these findings.