Prenatal stress, the placenta and maternal microbial transmission; implications for health and disease

There is an extensive amount of epidemiological evidence showing that prenatal maternal distress (PNMD) is a risk factor for a wide range of poor obstetric and neonatal outcomes, as well as an increased risk for the development of metabolic, immune and nervous system disorders in affected children later in life. Whilst many epidemiology studies have supported these associations, the biological mechanisms linking maternal prenatal distress with adverse outcomes remains understudied, particularly in human cohorts. One potential mechanism, known as the glucocorticoid hypothesis proposes that fetal overexposure to stress-induced maternal cortisol during critical windows of development increases the risk of adverse outcomes in the offspring. At the core of this hypothesis is the placenta, which expresses the enzyme 11beta hydroxysteroid dehydrogenase type 2 (HSD11B2), which ultimately controls the amount of cortisol a fetus is exposed to. Prenatal stress has been shown to reduce the placental expression of this enzyme; however the molecular mechanisms through which this occurs have not been well examined. More recently, the transmission of a suboptimal stressed maternal microbiota is emerging as an alternative mechanism that may mediate the impact of prenatal stress on infant development. However this has not yet been examined in a clinical population. We first utilized an in vitro placenta model, JEG-3 cells, to examine the effects of stress on the placental expression of HSD11B2. JEG-3 cells were cultured with exogenous cortisol and interleukin-1 beta (IL-1β), two potential biological mediators of prenatal stress. This study showed both cortisol and IL-1β can reduce HSD11B2 expression, an effect that could be prevented by co-treatment with a histone deacetylase inhibitor. Having established that cortisol can directly affect the expression of HSD11B2, we moved on to our first clinical study to examine this question in a clinical population by examining the impact of prenatal distress on placental gene expression and infant outcomes. A cohort of 121 pregnant women receiving antenatal care at Cork University Maternity Hospital (CUMH) completed the Perceived stressed scale (PSS), State Trait Anxiety Inventory (STAI) and Edinburgh Postnatal Depression Scale (EPDS) in late pregnancy and donated placental biopsies at the time of birth. This study identified a significant reduction in HSD11B2 mRNA along with an increase in the glucocorticoid receptor (NR3C1) in placentae from high distressed pregnancies. Additionally prenatal distress was associated with a number of adverse outcomes including delivering via Caesarean section, reduced Apgar scores and reduced birth temperature, supporting a role for placental glucocorticoid signalling in the relationship between prenatal distress and adverse outcomes. Having reported that stress impacts molecular placental signals and birth outcomes, we moved on to complete the SMArTI (Stressed Microbial Transfer to the Infant) study, a more detailed pregnancy cohort to examine the impact prenatal distress on the maternal and infant microbiome. This study yielded a final cohort off 111 nulliparous pregnant women that were recruited from the IMPROvED consortium at CUMH. Women enrolled in SMArTI completed distress questionnaires and provided saliva and fecal samples in the second and/or third trimester of pregnancy. Vaginal swabs, placenta samples and newborn hair were acquired at birth and infant fecal samples were subsequently collected across the first 5 months of life. We first used this cohort to further examine and validate the relationship between prenatal distress, placental glucocorticoid genes and infant outcomes. We found this relationship to be dependent on the timing of distress, type of distress and infant sex. Most notably we observed second trimester maternal anxiety correlated with reduced birthweight in female infants, a relationship mediated by placental FK506-bind protein 51 (FKBP51) mRNA expression. We finally used the SMArTI cohort to examine, for the first time, the impact of PNMD on the maternal and infant microbiome, using 16S rRNA gene sequencing. Reduced diversity of the maternal gut microbiome in the second trimester was associated with second trimester distress, most substantially with maternal depressive symptoms, an effect that was no longer apparent by the third trimester. The third trimester gut microbiome appeared relatively resistant to change with only modest alterations observed in women who had high second trimester cortisol. Of interest, third trimester distress had no effect on the third trimester gut microbiome, highlighting the experience of distress specifically in the second trimester as an important window of vulnerability. Reduced diversity of the vaginal microbiome, just prior to delivery, was associated with second trimester cortisol, with no alterations linked third trimester distress. When examining the infant gut microbiome we found increased diversity across the first 5 months of life to be associated with second trimester stress with corresponding decreases to the important Bifidobacteriaceae and Lactobacillaceae family. In conclusion, this thesis indicates the experience of PNMD influences key placental genes involved in glucocorticoid signalling in the placentae. The timing of maternal distress and infant sex are important factors in this relationship. Of particular interest we find placental FKBP51 to mediate a relationship between maternal anxiety and infant birthweight, demonstrating a direct role for placental glucocorticoid signalling underlying the relationship between prenatal distress and infant outcomes. The work presented in this thesis is the first of its kind to prospectively examine the influence of PNMD on the maternal gut, vaginal and infant gut microbiome. Stress-induced alterations in the maternal gut microbiome may contribute to adverse obstetric and birth outcomes albeit via a mechanism other than transmission of a suboptimal maternal microbiota during birth. Taken together, our results identify the second trimester as an especially vulnerable period to stress exposures and implicate the placenta and microbiome in mediating these effects. Counteracting the impact of stress during this critical time window may have important obstetric implications. Additionally understanding the consequence of the altered infant gut microbiome as a result of prenatal distress warrants further investigation