Long-term effects of DHA and/or 5-MTHF supplementation in pregnant women on their offspring’s fatty acids status, neurodevelopment and behaviour

Resumen

The maternal intake of polyunsaturated fatty acids (PUFA) and their metabolism will determine the bioavailability of long-chain polyunsaturated fatty acids (LC-PUFAs) during the first months of life. However, the importance of the impact of altered maternal PUFA status on the scheduling and imprinting of PUFA metabolism in the foetus is unknown, and the long-term consequences on the functionality of the metabolic pathways involved in PUFA elongation and desaturation of the PUFAs that allow the endogenous synthesis of LC-PUFAs. The role of the genetic polymorphisms of fatty acid desaturases (FADS) in modulating the conversion of essential fatty acid (EFA) precursors of the n-3 and n-6 series into their consequent derivatives has been highlighted in the last 10 years. This may explain the heterogeneity of responses found in various studies on the effect of prenatal PUFA supplementation on growth and neurodevelopment. Genetic polymorphisms of FADS1, FADS2 and FADS3 (which encode the synthesis of δ5 and δ6 desaturases) have been identified as important determinants of plasma circulating LC-PUFAs, although due to the significant genetic variation in different populations, there is a lack of evidence of the effects of FADS polymorphisms on the metabolism of LC-PUFAs. There has been increasing interest in the study of the fatty acid (FA) concentrations in the cells of the cheek mucosa in recent years, since it is considered that it could be a plausible alternative for following the nutritional status of the LC -PUFAs in early stages of life, thus avoiding blood extraction with the problems that this entails in young children. Currently, there is no information about a possible association between the genetic polymorphisms of FADS1 and FADS2 and the concentrations of these FAs in cheek cells. Moreover, in Europe, saturated FAs and the n-6 series have replaced the intake of n-3 FAs, in a different way in each country. Although the n-6 are also essential for health during growth and development, an appropriate balance is required with respect to n-3 PUFAs. It has been found that relative deficiencies of n-3 PUFAs are associated with a wide range of pathologies in relation to physical and mental health, which pose increasing problems in developed countries. Nutrition plays an important role in the structural and functional growth of the human brain, from conception and during childhood and adolescence to adulthood. There is scientific evidence that early nutrition may influence cognitive development and subsequent behaviour. It is known that an optimal contribution of macro and micronutrients is important for the synthesis of neurotransmitters (and their receptors and transporters), for the renewal and maintenance of the cytoskeleton of axons and myelin sheaths, for the growth of synaptic spines and, therefore, for neuronal plasticity and neuronal survival. During the 9 months of gestation, from the creation of the initial stem cells the brain reaches more than 100 trillion nerve cells at birth and weighs about 400g. During the first 4 years of life, the brain continues to grow, reaching the size of 1,200 g, which is only 200g less than the brain of an adult. In the next 10-15 years, the brain continues to grow, involving different compartments differently. For example, the thickness of the cerebral cortex varies in different regions between the ages of 5 and 18 and at different times, and the regions of reasoning, planning and social communication do so later. Changes in brain volume are also accompanied by changes in the size of the cephalic perimeter. Preliminary studies have shown that the circumference of the head during the first years of life can predict brain size and volume and neurocognitive development. However, it is unknown whether this measurement has a predictive value in terms of regional and sub-cortical cerebral volumes. LC-PUFAs appear to play an essential role in these processes. LC-PUFA deficiencies during the first 1,000 days of life, together with an unfavourable genetic load, may contribute to the development of alterations in brain morphology, structure and function, and alterations in cognitive development and behaviour in childhood. Therefore, it is necessary to monitor the intake of these FAs, before, during gestation and lactation, and in the first 2 years of life to prevent pathologies in the longer term. Animal models have shown that maternal diets deficient in n-3 and n-6 LC-PUFAs alter the deposition of these FAs in the neonatal cortex, which determine changes in neurotransmitter metabolism and learning failures. Docosahexaenoic acid (DHA) deficiency during gestation is associated with a reduction in dendritic arborisation, a failure in gene expression involved in the regulation of neurogenesis, neurotransmission and connectivity. It has been demonstrated that FADS1 and FADS2 polymorphisms are related to important changes in FA metabolism and brain development. Koletzko et al. have shown a consistent association of the minor alleles of FADS1 and FADS2 in pregnant women with lower concentrations of DHA in the phospholipids (PL) of erythrocyte membranes. This doctoral thesis aims to: 1) to evaluate the influence of maternal supplementation during pregnancy and the genetic polymorphisms of FADS1, FADS2 and FADS3 on the composition of FAs in glycerophospholipid (GPL) cheek cells of children up to 9.5 years of age; 2) to analyse the relationship between the composition of FAs in the GPL of cheek cells in children up to 9.5 years, dietary intake, age, sex and country of origin; 3) to study the association between FA profile in the oral mucosa and behavioural problems in children; 4) to establish the extent to which the behaviour and the psychological characteristics of the mothers influence behaviour, internalizing and externalizing problems in their children at 8 years of age; and 5) to explore the relationship between cranial perimeter evolution from birth to 10 years, sub-cortical volumes and brain distribution of grey and white matter and cognitive development. 147 Spanish (n=105) and German (n=42) children participating in the NUHEAL study have been included. Mothers were supplemented from week 20 of gestation until delivery with fish oil (FO) (500 g DHA + 150 mg eicosapentaenoic acid (EPA)), 5-methyltetrahydrofolate (5-MTHF), placebo or both. The children of these pregnant women were re-evaluated in different visits until 10 years of age. At 8, 9 and 9.5 years of age, a profile of 21 FAs was obtained from GPL of children cheek cell samples, following the technique described by Klinger et al. (2011). 17 Single Nucleotide Peptides (SNPs) were also analysed in these children in umbilical cord blood samples by iPLEX (Sequenom, San Diego, CA, USA) and mass spectrometry (MALDI-TOF MS, Mass Array; Sequenom). These children also did the Child Behaviour Check List Test (CBCL) at 7.5 and 9 years old and the Eating Disorder Inventory (EDI-2) questionnaire at 8 and 9.5 years old. At the age of 10, neuro-imaging (brain magnetic resonance imaging - MRI) of 74 Spanish children was obtained, studying morphometry and volumetry based on voxels, superficial extraction of the cerebral cortex and sub-cortical segmentation. The evaluation of dietary intake was obtained from food frequency questionnaires (FFQ) completed by the parents and a record of dietary intake 24 hours for 3 days, (2 on workdays and 1 on weekends). Data from the food frequency questionnaire and the 3-day 24-hour dietary recall were processed using the PCN-CESNID and “Young Adolescents' Nutrition Assessment on Computer (YANA-C)” international software. The food composition tables from DIAL (Spain) and the German Food Code and Nutrient Data Base (BLS) (Germany) were used. To perform the analyses of FAs, in all cases, a descriptive and normal analysis was performed using the Shapiro-Wilk test. When the variables did not follow a normal distribution, transformations were made using logarithms (ln) and square root (sqrt). If normality was not reached, data analysis was performed using the non-parametric U Mann-Whitney test. The analysis of the continuous variables was performed using ANOVA and the Kruskal-Wallis test. In the case of categorical variables, χ2 was used. For the analysis of the genotypes, the Hardy-Weinberg equilibrium (H-W) was calculated using the statistical software R. The deviations of the H-W equilibrium were studied using Fisher's exact test. In addition, the Lewontin's D' test and quadratic (r2) parity correlations between genes were used to assess imbalance. Linear regression and a mixed linear model were also performed to establish the association between FADS polymorphisms, type of supplementation received during pregnancy, age, sex and country of birth, and GPL-FA concentrations in the different moments in which they were analysed. The multivariate analysis, Principal Component Analysis (PCA) and Partial Least Squares (PLS) were performed to verify the distribution of the results according to different criteria and to represent them in a more accurate way. To obtain the possible relationships between the eating behaviour and the psychological status of the mothers on the behaviour of their children, a binary logistic regression was performed. We used an analysis of covariance (ANCOVA) approach to evaluate early supplementation (four groups: FO, 5-MTHF, FO+5-MTHF, and Placebo) effects on head circumference measurements and cognitive abilities (MPI score), as well as on total and regional brain volumes and inner cortical surface area. Dependent variables were submitted to a 2 (FO) x 2 (5-MTHF) between subjects ANCOVA. The analyses were controlled by maternal age, maternal height and weight, maternal body mass index (BMI), family economic status, smoking (week 20 and 30 of pregnancy) and placental weight, sex, laterality and total intracranial volume (TIV), being included in the different models as confounding variables. In general, the SPSS statistical program, version 20.0 (SPSS Inc. Chicago, USA), R statistical software (3.2.2 Version, "genetics" package), and STATA (StataCorp 2011; Stata Statistical Software: Release 12). In all cases, the level of significance was set when p <0.05. The first study showed that supplementation with FO and/or 5-MTHF during pregnancy has long-term effects on the GPL-FA profile of cheek cells in children up to 9.5 years. In addition, it could be seen that the polymorphisms of FADS1 and FADS2 genotypes of children influence the concentrations of the GLP-PUFAs obtained in cheek cells. Most associations were established between different SNPs FADS1 and FADS2 with linoleic (LA) and α-Linolenic (ALA) acids. It was also verified that the different supplements that the mothers received during pregnancy along with the genetic background of their children affect the composition of the FAs of the children. In a subsequent analysis, it was possible to demonstrate that the FA profile in the GPL of the cheek cells analysed in school-age children, was strongly related to the country of origin, suggesting an effect of diet, lifestyle, etc. Likewise, statistical differences in the composition of GPL-FAs of cheek cells between 8 and 9.5 years old were verified in both Spanish and German children. The daily intake of AA showed a high impact on the percentage of FAs found in the cheek cells of children. The intake of lipids and fibre determines a decrease in araquidonic acid (AA). The intake of folic acid and EPA positively influences the concentrations of AA, EPA and docosapentaenoic acid (DPA) in cheek cells. Intake of DHA has a negative influence on the concentrations of DPA in the GPL of the cells of the buccal mucosa. Associations were found between ALA concentrations in GPL and outsourcing and care problems. LA was associated with attention problems. EPA, DHA and DPA were associated with better ability of games and sports competitions. Other analyses have shown that the type of maternal supplementation during pregnancy did not influence the behaviour of the children at 8 years old (30 children had internalization problems and 24 had externalization problems). An association was established between the internalizing problems of children and drive for thinness (DT) of their mothers. The mothers classified as pathological in the EDI-2 DT sub-scale were 1,138 times more likely to be influential in their children's internalizing problems. Finally, it was found that prenatal supplementation was not associated with different cranial circumference sizes at any age, nor did it influence cognitive abilities or total brain volumes. At 4 years old, the most significant association of the cranial perimeter with the total volume of grey matter (GMV), the total volume of white matter (WMV) and the surface area of the brain could be demonstrated. Also, the size of the cranial perimeter at 4 years old was associated with a higher GMV content in the frontal, temporal and occipital areas, as well as the volume of the caudate nucleus, globus pallidum, putamen and thalamus. In conclusion: 1. FO and/or 5-MTHF supplementation during pregnancy have long-term effects on the GPL-FA profiles in children’s cheek cells at school age; a programming effect of fatty acids metabolic pathways is suggested, which may cause differences in children’s FA status. 2. Children’s FADS1 and FADS2 gene cluster polymorphisms influence PUFA concentrations in children’s cheek cells. The majority of the associations were established between FADS1 and FADS2 SNPs with LA and ALA. 3. The interaction between early life nutrition and children’s FADS1, FADS2 and FADS3 genotype, demonstrate long-term effects on GPL concentrations of the oral mucosa in children, suggesting an early programming of the metabolic pathways involved in the status of these FAs, driven by the genetic background. 4. Differences amongst the FA profiles from cheek-cell samples analysed in the NUHEAL children are strongly related to the country of origin, and therefore diet, lifestyle, etc. 5. Dietary intake also appears to influence the fatty acid concentrations present in the cheek cells of the studied children, the daily intake of AA especially shows a high impact on the percentage of fatty acids found in the cheek cells of the children. 6. There is a relationship between certain characteristics of maternal eating disorders, such as the "obsession with thinness", and the subsequent development of internalizing behaviour problems in their children at 8 years of age. This relationship determines that those mothers, who have problems related to the obsession with being thin, increase the likelihood of their children having 1,138 times more risk of internalizing behaviour problems compared to those children whose mothers do not show this obsession. 7. We have verified that the measurement of the cephalic perimeter at 4 years of age serve as a marker of cognitive development at 10 years of age. In addition, this anthropometric measurement is very useful from birth to 4 years, as this period is considered the most sensitive for brain growth. Certain postnatal factors, such as diet or LC-PUFAs status, may have a more significant impact on the structural maturation of certain cortical areas and sub-cortical nuclei, independent of the effects determined by prenatal supplementation.