Research Project 3
PI: Timothy York, Ph.D.
There is no clear understanding of how social and environmental determinants of health, shown to largely correlate with racial categories and birth outcomes, exert their effect on biologically mediated mechanisms of preterm birth. Substantial evidence points to epigenetic remodeling as a consequence of environmental exposures, i.e., functionally relevant modifications of the genome that do not involve changes in nucleotide sequence. The integration of epigenetic studies into preterm birth research provides a description at the level of biological mechanism of the interaction between the environment and DNA sequence and promises to elucidate the factors associated with the observed racial differential in the rate of preterm birth.
Pregnant African American and European American women will be recruited through VCU Hospital clinics and prospectively tracked at defined intervals to measure a broad array of social, economic, behavioral and stress-related measures and to obtain blood, decidual and placental tissue samples for methylation, gene expression and genotype studies. The analytic strategy will take advantage of a longitudinal, repeated measures design to identify robust environmental-biological mechanisms that both influence preterm birth and account for racial differences in preterm birth rates. There are three specific aims.
Specific Aim 1: Employ data reduction and prioritization methods to develop a set of optimal environmental and biological constructs and identify those that correlate with preterm birth. Data collection will result in a rich and expansive set of environmental measures and multiple levels of genome-wide variation (epigenetic, gene expression, DNA sequence). From this large set of variables the most informative environmental and biological risk factors will be identified using a series of psychometric and bioinformatic approaches. The set of variables meeting sufficient criteria for informativeness will be carried forward to Aim 2 to develop environmental-biological models of causal mechanisms.
- Indices of environmental risk and the collection of birth outcome measures will be summarized using traditional psychometric approaches including factor analysis, auto-regressive methods and sum scores to test their strength of association with preterm birth phenotypes.
- Longitudinal models will summarize both the mean level and rate of change across pregnancy for both epigenetic and gene expression measures at each locus.
- The derived biological measures of change will be further reduced and prioritized based on assessment of association with preterm birth, correcting for multiple testing, and bioinformatic methods to summarize known candidate systems and exploratory analysis using data reduction techniques
Specific Aim 2: Identify causal mechanisms of preterm birth that integrate environmental and biological risk factors through epigenetic processes. A theory driven structural equation modeling approach will be utilized to identify etiologically related indices of environmental and biological risk identified in Aim 1. Although the modeling structure is flexible to accommodate testing a large number of potential analyses the following relationships will be of primary concern.
- The effect of social and environmental indices of health and pregnancy-specific environments will causally influence preterm birth through direct pathways and/or indirectly mediated through epigenetic and gene expression processes.
- Examine extent that indices derived from longitudinal models of epigenetic and gene expression measures can co-vary with respect to mean level and rate of change.
- Determine the influence attributable to allelic differences in DNA sequence versus direct environmental influences on epigenetic and gene expression measures.
- Examine the correspondence between environmental-biological relationships identified from maternal blood samples during pregnancy with maternal decidual and fetal placental tissue collected at birth.
Specific Aim 3: Determine if the influence of race on preterm birth persists over and above the mechanisms identified in Aim 2. An important goal of this study is to identify those social and environmental determinants of health that account for the observed racial disparity in preterm birth. In the present aim we will evaluate the extent to which any remaining effect of race, after accounting for the mechanisms identified in Aim 2, accounts for additional variance in preterm birth. We will test whether the residual effect of race is directly associated with preterm birth or mediated through the same or different biological processes identified in Aim 2.