Skip to main content

Effect of Physical Activity in Pregnancy on Maternal and Fetal Human Milk Oligosaccharides

Final Report Summary - HMO (Effect of Physical Activity in Pregnancy on Maternal and Fetal Human Milk Oligosaccharides)

This IIF project focused on investigating the role of Human Milk Oligosaccharides (HMO) as potential novel biomarkers for metabolic diseases in pregnancy.

HMO are a highly bioactive and diverse group of glycans in breast milk, known for their many beneficial effects in the breast-fed infant. New evidence, detecting HMO already in the systemic circulation of pregnant women (maternal HMO) and in the umbilical cord blood (fetal HMO), raised the question, whether prenatal HMO can influence pregnancy outcomes and developmental programming of the fetus.

In breast milk, HMO concentration and composition are known to be highly individual, depending on the woman’s genetics, but potentially also on environmental factors, such as nutritional status. Nothing was known on biological variability and influencing factors of prenatal HMO composition and concentration in maternal serum. We here hypothesized that prenatal HMO are associated with life-style induced metabolic changes, and that HMO can have an impact on maternal and fetal health. To test this hypothesis, the following main questions were asked:

1. When in pregnancy, do HMO appear in maternal serum? What is their biological variation? Do concentration and/or composition change over the course of pregnancy?
2. How do HMO concentration and composition in cord blood compare to maternal HMO? Are arterial and venous cord blood HMO different in composition/concentration?
3. Are prenatal HMO concentration and/or composition associated with maternal factors such as body mass index, body composition, or physical activity? Are prenatal HMO associated with fetal outcomes such as fetal growth and body composition?

In the first phase of the project we established the high performance liquid chromatography (HPLC) method with fluorescence detection for HMO identification and quantification at the host institution. This method is highly sensitive and allowed for quantification of up to 18 HMO using an internal standard. Moreover, the method allowed for typing according to Secretor blood group status, by determining the presence or absence of blood type characteristic, fucosylated HMO structures (2’Fucosyllactose and Lactodifucotetraose), indicative of Secretor status. HMO were annotated using retention times of standards, and confirmed by mass spectrometry and enzymatic digest. To establish variability and temporal patterns of maternal HMO concentration in serum, we used an observatory, longitudinal cohort study over the course of gestation. In total, serum samples from 27 women at 4 time points were analyzed (Visit 1: 10-14 weeks, Visit 2: 18-22 weeks, Visit 3: 33-35 weeks and at delivery). Total HMO concentrations were determined as sum of the 18 identified oligosaccharides in the profiles. The relative proportion of each HMO per summed total HMO concentration was calculated. We could establish that HMO are detectable in maternal serum already at the end of the first trimester/beginning of the second trimester (Visit 1). Total HMO concentrations increased significantly from Visit 1 to delivery. HMO composition in maternal serum changed during pregnancy, with the largest changes in profiles seen between Visit 1 and Visit 2, mostly due to a strong increase in fucosylated HMO, shifting the profile from a predominantly sialylated to a more balanced ratio of sialylated to fucosylated HMO. Presence or absence of 2’Fucosyllactose and Lactodifucotetraose in profiles allowed for annotation of the Secretor blood group status at Visit 2.

Next part of the project was to compare fetal (cord blood) and maternal HMO profiles at delivery in these 27 mother-infant dyads. Overall, HMO profiles in cord blood serum strongly resembled HMO composition in maternal serum, containing the same individual HMO structures, and similar total concentrations. Comparing venous and arterial cord blood HMO revealed higher total HMO concentrations in arterial than in venous cord blood. Fucosylated HMO were strongly positively correlated, whereas sialylated HMO were less, but still significantly correlated between venous and arterial cord blood. When comparing maternal and cord blood serum, this difference in correlation of fucosylated and sialylated HMO was even more pronounced: Secretor-active, fucosylated HMO were strongly correlated, but no correlations were found for individual sialylated HMO (absolute and relative concentrations). These findings strongly speak for a common source and trans-placental transport of fucosylated HMO, and leave room for speculations on potential additional sources or different transport mechanisms for some sialylated HMO.

Finally, statistical analyses were performed to investigate associations of maternal and fetal HMO with maternal factors and fetal outcomes. Remarkably, in this small cohort of healthy, normal weight women, we found significant association of maternal serum HMO (total and individual) with maternal factors, such as pre-pregnancy BMI, body composition, glucose and lipid parameters, inflammatory factors, and physical activity, indicating that maternal metabolic status indeed influences prenatal HMO concentration and composition in maternal serum. We also found associations of maternal HMO with some inflammatory markers and metabolic parameters in cord blood, indicating that maternal HMO could also affect fetal outcome. However, we did not detect any significant associations of maternal or fetal HMO with fetal growth or body composition.

HMO research currently aims to identify protective structures and potential biomarkers in breast milk for diseases most relevant to the neonate and more recently, also to the lactating mother. Biological variability in HMO concentration and composition in breast milk may lead to varying extent of protection for the infant. Understanding these correlations may lead to identification of biomarkers for disorders, potential pharmaceutical targets, or therapeutics to supplement a particular structure. The same concept is also conceivable for prenatal/perinatal HMO in the maternal/fetal circulation. By characterizing for the first time, prenatal HMO profiles in a healthy pregnancy cohort and reporting on associations with maternal metabolic factors and fetal outcomes, the foundation has been laid for larger studies in the future, investing prenatal HMO against the background of pregnancy pathologies, such as gestational diabetes or preeclampsia.

The fellow has successfully transferred the HMO methodology and knowledge to the host lab at the Medical University of Graz, now being an integral part of the perinatology research lab. By training students and personnel, the technology transfer already proved sustainable in further establishing the emerging field of HMO research in the European research area. This project has bridged 'traditional' HMO research and topics of pediatrics and neonatology, with perinatology and feto-maternal medicine, laying the foundation for a new research field, the role of HMO in pregnancy. In the course of the project a mini-symposium on HMO in pregnancy has been organized. Targeting both, experts from the HMO field, and perinatologists and feto-maternal specialists, the mini-symposium introduced this novel concept. This mini-symposium also targeted medical personnel and medical students to educate them on the benefits on human milk.
The novel concept that HMO composition may affect not only the health of the breast-fed newborn, but already have implications on the health of mother and fetus will set the stage for future research. The future outcomes of this research have the potential of being of large clinical, economic and social impact, as identifying non-invasive biomarkers in pregnancy related diseases is hugely needed to inhibit aberrant fetal programming. Furthermore, if maternal lifestyle factors and modifiable metabolic factors identified here to influence maternal and fetal HMO will be confirmed in larger studies, this should be relevant for policy makers, public health agents and to civil societies for implementing effective life-style interventions.