Long-term effects of adverse childhood experiences on the microrna profile of women and their newborns

Exposure to adverse childhood experiences (ACEs), including physical, sexual, and emotional abuse, as well as physical and emotional neglect, confers risk for molecular alterations, such as changes in the profile of miRNAs, which are small non-coding RNAs responsible for regulating the expression of target genes. The presence of these molecular changes may serve as a potential link between ACEs exposure and neuropsychiatric outcomes in later life. However, accumulating evidence suggests that the effects of ACEs extend beyond the directly exposed individuals and can impact subsequent generations. Our hypothesis is grounded in the understanding that ACEs can lead to a chronic inflammatory state that persists into adulthood. During pregnancy, we propose that this altered inflammatory state may have an impact on the miRNA profile of mother and potentially of their newborns. However, mounting evidence points to ACEs effect not only to those directly exposed, but also to their next generation. Our hypothesis is based on the knowledge that ACEs leads to a chronic inflammatory state that may persist until adulthood. During pregnancy, we hypothesize this altered state could impact the miRNA profile of mother and potentially of their newborns. We investigated the miRNA expression profile of maternal peripheral blood and umbilical cord blood in a sample of 43 mother-infant dyads. Mothers’ exposure to ACEs were assessed with the Childhood Trauma Questionnaire (CTQ). miRNAs were isolated, their expression was measured by Next Generation Sequencing and the raw read counts of the expressed miRNAs were obtained by the miRge 3.0 pipeline (which includes the alignment to miRBase v22). Further analyses were performed in R. Read counts were normalized using the method provided by DESeq2 and then were fit into a generalized linear model for Poisson distribution [read counts ∼ maternal ACEs + covariate]. Mother age and infant biological sex were used as a covariate for maternal and cord blood samples, respectively. A total of 533 miRNAs were identified in maternal blood and 677 miRNAs in cord blood. We found hsa-miR-326-3p (p=0.016; pseudo-R²= 0.09) and hsa-miR-6868-3p (p=0.025; pseudo-R²=0.27) in maternal blood and hsa-miR-4443 (p=0.040; pseudo-R²=0.13) in umbilical cord blood predicted by maternal ACEs. However, no association was found after adjusting p-values for false discovery rate (FDR). The three miRNAs predicted by maternal ACEs did not survive correction for multiple comparisons, probably due to the small statistical power of sample. This is an on-going study and further investigation of the model parameters and assumptions are being carried out to assure the best goodness-of-fit. Given the unprecedented nature of our design, our results still contribute to future studies in this line of research.