Tom Sharpton and Emily Ho: Tracing Environmental Chemical Actions on the Gut Microbiome with Zebrafish

Investigators (alphabetical):

  • Emily Ho (Director Professor of Moore Family Center for Whole Grain Foods, Nutrition and Preventive Health, College of Public Health and Human Sciences)
  • Tom Sharpton (Assistant Professor in Department of Microbiology, College of Science)


The human gastrointestinal tract is home to a diverse array of microorganisms (i.e., the gut microbiome) that live in delicate balance with their host to perform essential functions. Disruption of this balance, known as dysbiosis, can associate with, result in, and contribute to various human diseases. Our insight into the mechanisms of dysbiosis is currently restricted, which limits the prevention and treatment of microbiome-mediated disease. For example, while humans consume and are surrounded by a plethora of microbial stimulators and inhibitors, we understand surprisingly little about how specific environmentally acquired metabolites or xenobiotics influence the gut microbiome. This is largely because the interaction between the host, environment, and microbiome is complex, and current experimental tools fail to provide the sample sizes needed to tease apart subtle effects.


We propose to develop a high-throughput experimental infrastructure for the study of gut microbiome-environmental health interactions and test the hypothesis that variation in environmental health perturbs the gut microbiome to induce dysbiosis. Our proposal includes proof-of-concept experiments that use a powerful zebrafish model to quantify how the gut microbiome changes as a function of perturbation dose, and explores several types of environmental health perturbations: (1) diet (i.e., fat content), (2) a toxicant (i.e., chlorine), and (3) an environmental xenobiotic (i.e., triclosan).  Our long-term objective is to use this experimental infrastructure to explore additional perturbing agents and host genotypes, and discover how the host, environment, and microbiome interact to influence health.


Our work will (1) clarify the ability of the microbiome to resist changes in environmental health, (2) discover thresholds of perturbation sensitivity, and (3) quantify how the type of perturbation (i.e., diet versus toxicant) impacts the magnitude of disruption. These findings will have profound implications for the consideration of consumer safety and the prevention, treatment, and management of dysbiotic diseases. Our results will be used as preliminary data in grant proposals designed to explore (1) the resiliency of the microbiome to these perturbations, (2) chronic exposure to the perturbing agent, and (3) the impact of microbiome perturbation on host physiology and health. Additionally, by validating our experimental infrastructure, we can subsequently explore the impact of virtually any dietary, pharmaceutical, or xenobiotic agent on the microbiome. Ultimately, the translational relevance of the most promising candidate agents will be solidified by follow-up studies in mice and, ultimately, humans.

Susan Carozza and Perry Hystad: A Spatial Assessment of Air Pollution from Oil and Gas Shale Extraction in Texas and Adverse Birth Outcomes, Birth Defects, and Childhood Cancers

Investigators (alphabetical):

  • Susan Carozza (Associate Professor in the College of Public Health and Human Sciences)
  • Perry Hystad (Assistant Professor in the College of Public Health and Human Sciences)


            Unconventional oil and gas production using hydraulic fracturing is nearly 60% of total production in the US, with over half of all unconventional wells located in the state of Texas.1 The rapid expansion of unconventional oil and gas extraction has enormous economic benefits, reduces reliance on foreign energy sources, and is a cleaner burning fuel than other fossil-fuels. However, shale gas development may have significant local environmental and public health implications. Approximately 15 million Americans now live within one mile of a natural gas well drilled since 2000.2

            Despite the scale of development and potential for environmental exposures, there is a dearth of population health studies examining potential impacts of oil and gas shale development. While the initial focus of concern was water contamination, closer attention is being paid to air pollution as a potential public health concern. Sources of air pollution include well flaring, compressor stations and truck traffic (e.g. 1500-2000 diesel trucks are required to bring in materials for the drilling process).  Air pollutants of concern include fine particulate matter, polycyclic aromatic hydrocarbons (PAHs), diesel exhaust, benzene, sulfur oxides and nitrogen oxides. For example, PAHs continuously measured for one year near a well pad (operating with closed loop best management practices) were substantially elevated within 1.1 km's.3 A risk assessment conducted in Colorado also estimated cancer risk within 0.5 miles of a well as 10 in a million (with benzene driving risk estimates).4

            While evidence is accumulating for increased emissions and exposures to air pollution associated with oil and gas shale development, there is extremely limited epidemiological evaluation of health impacts. McKenzie et al.5 examined 124,842 births and residential proximity to natural gas developments in rural Colorado and observed an increased odds ratio (OR) of 1.3 (95% CI: 1.2, 1.5) for congenital heart defects and 2.0 (1.0-3.9) for neural tube defects for the highest versus lowest exposure tertile (using an IDW measure of well proximity). Exposure was also negatively associated with preterm birth and positively associated with fetal growth, although the magnitude of association was small. Hill et al.6 also examined birth outcomes in Pennsylvania from 2003-2010 and observed increases in low birth weight and decreased term birth weight among mothers living within 2.5 km of a well compared to mothers living within 2.5 km of a future well. Currently, these are the only epidemiological studies of exposure to air pollution from oil and gas shale extraction, highlighting the urgent need for further research in this area.


Pilot Study Objectives:

            We propose to conduct a large epidemiological study of birth outcomes, birth defects and childhood cancers in Texas, examining spatial associations with oil and gas shale development activities. Dr. Carozza has secured access to a database of all births in Texas (n=~6 million) between 1996 and 2009. This time period corresponds to the rapid expansion of oil and gas development in Texas. A database of incident childhood cancers for this time period is also available.  This data file includes approximately 7,000 childhood cancer records which have been linked back to Texas birth certificates, and so include data on both the cancer diagnosis and factors related to the pregnancy and birth of the child. .  The study data also includes records of birth defects reported during that study period; these data are linked to Texas birth certificates as well, and to any reported childhood cancers during the study period.  All records include geocoded birth residence based on full residential addresses; cancer cases also have residence at time of diagnosis.

             Spatial exposure assessment methods will be used to estimate air pollution exposures metrics from oil and gas shale extraction activities at study participants' residential addresses. This will include well proximity measures as well as estimates of exposure to truck traffic emissions. The railroad commission of Texas holds a database of all oil and gas well locations and production information that will be used for spatial exposure assessments.

            We will examine associations between different spatial exposure assessment metrics and birth outcomes, birth defects and childhood cancers. In addition, spatial-temporal analyses will examine associations' pre/post well development, which will substantially reduce the potential for bias in this study. Information on maternal and paternal demographics, pregnancy conditions and birth factors are also available in these databases and can be used to control for potential confounding factors.