El Sendero: Pathways to Health Study

This project will continue to follow two birth cohorts of mother-infant Latino dyads through a series of new assessments at age 6y, with an emphasis on examining the the role early nutritional exposures, exposures to environmental toxins, and social determinants of health have on adiposity, eating behaviors, brain structure and function, cognitive outcomes, and chronic disease risk.

This project will continue our work in two birth cohorts of mother-infant Latino dyads (NCT04434027 & NCT03141346) through a series of new assessments at age 6y, with a focus on examining the role of early nutritional exposures, exposures to environmental toxins, and social determinants of health (SDOH) on adiposity, eating behaviors, brain structure and function, cognitive outcomes, and chronic disease risk. This collection of assessments at 6 years of age is funded by two grants (R01 DK110793 01A1 & P50 MD017344). Breast milk has been shown to contain certain macronutrients (human milk oligosaccharides; HMOs) that vary greatly among women. Research from our lab recently characterized associations of HMOs with infant obesity, eating behaviors, and brain development as well as dynamic changes in HMOs over the course of breastfeeding and suggests that these factors are significant predictors of infant weight gain and adiposity. Additionally, we already obtained detailed individual measures of ambient and near-roadway air pollution exposure from pregnancy to 2y of age. Our previous work in a subset of participants from the proposed cohort found that increased prenatal exposure to ambient air pollutants was associated with increased infant growth and adiposity after adjusting for infant sex and age, pre-pregnancy BMI, breastfeeding, maternal age, season of birth, and SES. By conducting longer-term follow-up with more rigorous outcomes at age 6y, we will be able to more definitively determine if and how early nutrition, particularly specific HMOs, impact chronic disease risk in Latino children as well as how environmental exposure to toxins and the food environment may exacerbate these health outcomes. Here, we will explore:

1. The impact of infant exposure to environmental toxins on subclinical markers of chronic disease risk at age 6y. We hypothesize that exposure to environmental toxins (air pollutants, CalEnviroScreen) are uniquely associated with longitudinal measures of child weight and subclinical markers of disease risk (fat distribution, liver fat, poor glycemic control, and serum lipid profile) at age 6y.
2. The impact of infant exposure to environmental toxins and nutrition, especially the HMOs 2'FL, LNFPI, LNFPII, LNnT & LNH, in early life on adiposity and chronic disease risk at age 6y. We hypothesize that infant feeding (formula feeding/type, exposure to HMOs in breast milk) and childhood diet (SSB/J) will be associated with adiposity (subcutaneous, visceral, and liver fat) at age 6y, and that this will be mediated by reduced eating in the absence of hunger as well as other subclinical markers of disease risk.
3. The impact of infant exposure to HMOs containing fucose and sialic acid on brain development at age 6y using image-based measures of brain structure (anatomical MRI), function (resting state fMRI), blood flow (arterial spin labeling), myelination and tissue microstructure (diffusion tensor imaging) as well as cognitive outcomes. We hypothesize that limited food access and other negative SDOH will a) independently be associated with child obesity, higher liver fat, poor glycemic control and/or lipids; and b) interact with infant feeding, childhood diet, and exposure to toxins to have synergistic effects on disease markers at 6y.
4. The impact of breast feeding (at breast vs delivered via a bottle) and the changing HMO profile on eating in the absence of hunger at age 6y, and structural and functional differences in key areas of the brain involved with appetite regulation (frontal cortex, basal ganglia, hippocampus, hypothalamus). We hypothesize that HMOs 2'FL, 3FL, and 3'SL, will be associated with improved brain outcomes, such as thicker cortical mantle and reduced fractional anisotropy (FA) of cortical gray matter, reduced resting perfusion of cortical gray matter, reduced diffusivity in white matter tracts, stronger measures of resting functional connectivity, and enhanced cognitive functioning. We further hypothesize that changes in cortical gray matter measures will mediate the association of specific HMOs with cognitive performance in early childhood.
5. The impact of the food environment and broader SDOH on subclinical markers of chronic disease and how these relationships may exacerbate the effects of poor nutrition and environmental toxins. We hypothesize that longer-term duration breastfeeding, and therefore greater exposure to HMOs, will be associated with reduced eating in the absence of hunger, and this association will be mediated by a thicker cortical mantle and reduced FA in aforementioned brain regions associated with appetite regulation.

Exposures:

Ambient Air pollution: Addresses will be used to generate x,y coordinates of latitude and longitude, geoIDs, and census tracts. These spatial elements will be used to query several databases to generate new geocoded social and environmental exposure variables. In addition, child dates of birth will be used to assign air pollution exposure estimates and community contextual variables for relevant time windows in early life through 6 years of age. Ambient air pollution will be measured by assigning PM2.5, NO2, and O3 exposures using hybrid model outputs developed by a team lead by Dr. Joel Schwartz (Harvard) over many years.

Near-Roadway Air Pollution (NRAP) : Addresses will be used to generate x,y coordinates of latitude and longitude, geoIDs, and census tracts. These spatial elements will be used to query several databases to generate new geocoded social and environmental exposure variables. In addition, child dates of birth will be used to assign air pollution exposure estimates and community contextual variables for relevant time windows in early life through 6 years of age. NRAP is a complex mixture of particles and gases, including particulate matter, organic compounds, elemental carbon, and polycyclic aromatic hydrocarbons. In this study, NRAP exposure will be characterized using the CALINE4 air quality dispersion model that incorporate HERE (www.here.com) detailed roadway geometry, traffic volumes from Streetlytics™ by Citilabs, Inc. (www.citilabs.com), vehicle emission rates from CARB's EMFAC2017 model, and atmospheric transport and dispersion (using wind speed, wind direction, atmospheric stability, and height of the mixing layer).

Social Determinants of Health (SDH): Addresses will be used to generate x,y coordinates of latitude and longitude, geoIDs, and census tracts. These spatial elements will be used to query several databases to generate new geocoded social and environmental exposure variables. In addition, child dates of birth will be used to assign air pollution exposure estimates and community contextual variables for relevant time windows in early life through 6 years of age. To capture SHD, we will use both individual-level data obtained from self-reported instruments as well context-level data obtained through geocoding and enrichment of participant data. For our geospatial analysis, using patient home addresses, we will assign (x,y) geographical coordinates using GIS software, which are spatially joined or associated with census tract polygons into which they fall. An 11-digit "Spatial GeoID" is appended to each point. The GeoID is then used to match the relevant data point to data from several state and national databases, including the US Census Bureau, the American Communities Survey, the US Department of Agriculture Economic Research Service, the Environmental Protection Agency, the California Environmental Health Screen, and the Healthy Places Index. PhenX SDOH core constructs will be used to capture sex, marital status, household composition, race/ethnicity, age, gender identity, birthplace, English proficiency, health literacy employment, food insecurity, and access to health services. Maternal Social Health will be assessed using validated measures from the Patient-Reported Outcomes Measurement Information System (PROMIS®). Two key domains relevant to diet and chronic disease will be measured: social isolation and social support. Acculturation to the US will be assessed by the Marin Short Acculturation Scale.

Child Dietary Intake: 24-hour dietary and physical activity recalls will be conducted in triplicate (1 weekend day and 2 weekdays). Dietary intake will be analyzed using the most current version of NDSR, which has been used extensively in the Goran lab and elsewhere for studies in Hispanics. Diet will be assessed at 6y of age.

Child Physical Activity: To assess physical activity, we will use the early years physical activity questionnaire (EY-PAQ). Mothers will be asked to report the frequency and duration of different Moderate to vigorous physical activity and sedentary time in which their child engaged during a typical week in the previous month. Physical activity will be assessed at 6y of age.

Child Sleep: To assess sleep, we selected the Children's Sleep Habits Questionnaire172, because it captures all domains of sleep health and was designed for children aged 2-10 years. Child sleep will be assessed at 6y of age.

Previously collected exposures: Early nutrition including breastfeeding versus formula feeding, breastfeeding exclusivity, breastfeeding duration, Breastmilk composition including HMO concentrations, and dietary intake were assessed at the timepoints 1 month (baseline), 6 months, 12 months, 24 months and 36 months. All previously collected variables can be found at the ClinicalTrials.gov pages for the original two cohorts from which this project stems: The Effects of Natural Sugars in Breast Milk on Healthy Infant Growth and Development (NCT04434027) and Improving the Eating Habits of Mother and Her Infant Via Sugar Reduction (MAMITA) (NCT03141346).