Identifying the biological effects of exposures

In the Johnson Lab, we use metabolomics to simultaneously analyze both exogenous chemicals, their metabolites, and changes to the endogenous metabolome to allow assessment of exposures and their biological impact. We also adjust for confounding using epidemiologic techniques, as shared factors could influence the metabolome and exposure measured. We are using this framework within various projects outlined below:

Conception is a critical window for female reproductive function, and environmental exposures may influence fertility. Using a metabolome-wide association study (MWAS) framework, we are investigating how parabens modulate the urinary metabolome of women attempting to conceive, to provide insight into mechanisms that may underlie the biological effects of these exposures.

We also discovered that changes to kidney function across the menstrual cycle can impact urine metabolite concentration, therefore we assessed different post-acquisition normalization approaches to identify the optimal method for urinary metabolomics. Our data published in the journal Metabolites shows that either specific gravity or probabilistic quotient normalization are reliable methods to adjust for urinary concentration.

Early-life exposures can have biological effects that lead to later-life adverse outcomes such as cancer. Dried blood spots (DBS) have been collected in large populations for decades, to screen infants for congenital metabolic disorders, however they represent an untapped resource for studies in epidemiology and population sciences. Metabolomics analysis of DBS from newborns provides the opportunity to measure both environmental exposures and metabolic perturbations simultaneously, offering both biomarkers of exposure and biological effect to gain insight into mechanisms of later life diseases.

No consensus exists for the optimal method for metabolite normalization in DBS untargeted LC-MS-based metabolomics. To address this issue and setup a robust experimental protocol for DBS metabolomics, we assessed the performance of various normalization methods. We identified that normalization using hemoglobin (Hb) is the best method. We also introduced a novel role of specific gravity as a predictor of Hb in DBS. This work is published in Science of The Total Environment.