Usage
  • 39 views
  • 70 downloads

DNA methylation patterns derived from fetal vulnerability to maternal smoking relate to future child outcomes

  • Author / Creator
    Ng, Jane WY
  • Fetuses exposed to maternal smoking during pregnancy (MSP) are at higher risk of developing illnesses such as obesity, diabetes and heart, lung and mental health disease. These disorders are part of a group of non-fatal, non-communicable common complex diseases (CCDs) that form the greatest burdens on quality of life and longevity around the world. The problem is worsening - yet scientists remain baffled over the pathways that lead more and more children to this fate of chronic disability and premature death. A large body of research suggests that MSP alters mechanisms that regulate how human cells read genetic information. The field of epigenetics studies these mechanisms, one of the most well-known being DNA methylation. Starting around the 1980’s, the detection and treatment of various human cancers has dramatically improved by measuring and therapeutically modifying DNA methylation. Despite fervent interest in replicating this success in the debilitating diseases linked to MSP, this has yet to occur.This failure in part may be because vulnerability to MSP is not equal among individuals – it is dependent on the unique combination of genetic and environmental factors influencing a fetus. For instance, this contributes to the phenomenon where a given level and timing of MSP will be associated with greater than expected numbers of below average weight newborns - but most will still be average weight or above. If we neglect to account for these differences between fetuses, we risk either failing to find MSP-related DNA methylation changes culpable for poor health outcomes, or finding only “bystanders” related to MSP but irrelevant to outcome. We propose to triangulate familial, maternal and infant factors to better identify individual fetal vulnerability by accounting for the interplay of factors conferring resilience and susceptibility to MSP. In this way, we aim to improve sensitivity and specificity in detection of clinically relevant DNA methylation changes. This field encounters another problem by way of how DNA methylation is analysed. Previous studies sought DNA methylation differences in small and/or isolated sections of DNA. However, there now exist many studies that suggest DNA methylation affects cell fate by changing the shape and interaction of multiple domains of DNA concurrently. We suggest using pattern matching computational methods to find global configurations of DNA methylation. This may help visualize critical changes related to MSP that disrupt how cells read genetic information and thus lead to poor long-term health. We applied these two innovations to a population-based pregnancy cohort. Using DNA methylation data from blood samples at birth, we extracted a number of patterns related to how vulnerable the fetus is to MSP. We uncovered patterns that appear biologically reasonable because 1) they affect areas of DNA that are sensitive to environmental exposures, 2) certain patterns are stable within children as they age, rendering them viable candidates as long term actors on later health and 3) children sharing these patterns had similar physical and mental health outcomes, suggesting specific disruptions in DNA methylation in fetushood lead to common development trajectories. Last, we reproduced these vulnerability patterns in an independent cohort of newborns. In this entirely distinct population and without the use of any MSP related data, the DNA methylation patterns similarly related to later health. These findings offer proof of concept for vulnerability-based pattern finding of DNA methylation traits.This study strives to break new ground on how we view data - whether from a questionnaire or a blood sample – to generate a context-sensitive map of vulnerability to adverse exposures and its effect on clinical and molecular processes of the growing child. Today, laboratory technology enables the generation of millions of data points from thousands of individuals using merely micrograms of sample, all within a fraction of time. Unlike a decade ago, the greatest knowledge-limiting step is how to harness this immense amount of information. We challenge whether current methods best optimize data for knowledge translation. We propose that a dramatic paradigm shift in our analytic lens is the step change needed to successfully detect and possibly reverse molecular disruptions underlying CCDs.

  • Subjects / Keywords
  • Graduation date
    Spring 2021
  • Type of Item
    Thesis
  • Degree
    Doctor of Philosophy
  • DOI
    https://doi.org/10.7939/r3-p554-wm21
  • License
    This thesis is made available by the University of Alberta Libraries with permission of the copyright owner solely for non-commercial purposes. This thesis, or any portion thereof, may not otherwise be copied or reproduced without the written consent of the copyright owner, except to the extent permitted by Canadian copyright law.