Physical Activity-Associated Differences in Placental Lipid Profiles and Docosahexaenoic Acid Uptake: Insights from Lipidomics and an In Vitro Exercise Model

Abstract

Background: Gestational physical activity (PA) has been associated with a multitude of positive outcomes for both the gestational parent and the offspring. Given the placenta’s central role in supporting pregnancy, it is likely involved in mediating these benefits, though the mechanistic explanations have yet to be elucidated. Placental lipid composition and uptake can provide a wealth of information on placental and fetal lipid availability, and their downstream pathways. Docosahexaenoic acid (DHA) is a fatty acid of particular importance for fetal brain and retinal development, and its proper delivery to the fetus is critical. The influence of gestational PA on these factors has not been explored. Methods: The first objective of this thesis was to characterize the lipid profile of placentas from pregnancies classified as physically active or inactive. Thirty-six term placenta samples were analyzed using liquid chromatography mass spectrometry (LC-MS/MS). PA levels were objectively measured using accelerometry. Major facilitator superfamily domain containing 2a (MFSD2a) protein expression in placental tissue was measured semi-quantitatively using Western blotting. The second objective was to evaluate DHA uptake in vitro. Primary trophoblast cells were exposed to bouts of intermittent hypoxia (IH) and high shear stress as proxies for PA. Hypoxia was considered 3% and normoxia 10% O2. Low shear stress was defined as 1 dyn/cm2 and high shear stress as 3 dyn/cm2. DHA was added to the cell culture media, and uptake into cells was quantified by gas chromatography - mass spectrometry (GC-MS). Results: A total of 192 lipid species were annotated in placenta samples, of which 34 differed significantly between active and inactive groups (p < 0.05; |log2FC| ≥ 0.58). Seven of the 17 lipids containing DHA were higher in the active group. Multivariate analyses demonstrated modest separation between active and inactive cohorts. MFSD2a expression did not differ between groups, however, infant weight-to-length percentile was negatively correlated with MFSD2a (r = -0.40; p = 0.019). Contrary to our hypothesis, DHA uptake was significantly greater in cells exposed to normoxia + static conditions compared to IH + high shear stress (p = 0.008), and IH + low shear stress (p = 0.003). Conclusions: There is a modest relationship between gestational PA and the placental lipidome, with several lipids differentially expressed between groups. Although MFSD2a (a main DHA transporter) expression did not differ, DHA-containing lipids were higher in placentas from physically active pregnancies. Our in vitro model of PA impacted intracellular DHA uptake, though not in the expected direction. This unanticipated finding may reflect limitations in the application of IH and SS to proxy PA, or an inability of the experiment to fully model fetal DHA delivery in vivo. Collectively, these findings provide a basis for targeted lipid analyses, and help guide future in vitro modeling of PA.