Grantee Research Project Results
Final Report: Integrating Innovative Biomarkers of Environmentally Induced Disease for Children in Agricultural Communities
EPA Grant Number: R832733Title: Integrating Innovative Biomarkers of Environmentally Induced Disease for Children in Agricultural Communities
Investigators: Faustman, Elaine , Griffith, William C. , Yu, Xiaozhong
Institution: University of Washington
EPA Project Officer: Callan, Richard
Project Period: October 1, 2005 through September 30, 2008 (Extended to September 30, 2010)
Project Amount: $749,997
RFA: Early Indicators of Environmentally Induced Disease (2004) RFA Text | Recipients Lists
Research Category: Biology/Life Sciences , Children's Health
Objective:
The purpose of this U.S. Environmental Protection Agency (EPA) Science to Achieve Results (STAR) award (RD832733) was to develop an integrative tool for evaluating genomic biomarkers of susceptibility and early response for public health. Our hypothesis was that an understanding of these biomarkers and their relationship could allow us to identify the potential for pesticide exposure and effect and to design effective intervention and prevention approaches. The integrative tool was designed for identifying and characterizing the potential for exposure and response relationships in adults and children in agricultural and non-agricultural communities using the public health biomarker paradigm.
We collaborated with the Community Based Participatory Research (CBPR) project in the Yakima Valley, which is part of the Center for Child Environmental Health Risks Research (CHC) funded by the U.S. Environmental Protection Agency (EPA) and the National Institute of Environmental Health Sciences (NIEHS). Of particular utility was the availability of a biorepository from their study of farmworker and non-farmworker families. The biorepository included samples of blood, urine, and dust collected from households and families of Hispanic farmworkers (N = 100) and their children (N=100) and Hispanic non-farmworkers (N = 100) and their children (N = 100). Samples were taken frequently over the agricultural season including samples over a 5-day period during times of pesticide application (thinning season) and non-application (non-spray season). Organophosphate (OP) pesticides present in the biospecimens were characterized. The biospecimens were collected with informed consent in accordance with the Fred Hutchinson Cancer Research Center (FHCRC) Institutional Review Board (IRB) #5946, and access to the samples is governed by the University of Washington (UW) Human Subjects protocol #27533. The biospecimens were de-identified. We evaluated the potential for exposures using the following biomarkers:
Integrated Tool: The models used in this study are linked through the study design by the collection of environmental and biological samples from adult/child pairs living in the same household, and the collection of samples from these households across seasons. These samples then were analyzed in multiple assays so that it was possible to link measurements to households and seasons. To better understand the potential for OP exposures to affect neurodevelopment, these exposures were linked to studies in laboratory animals. We used biologically based dose response models to understand the dynamics of response in laboratory animals to estimate the potential for human responses. Acetylcholinesterase inhibitioni was used as a common biomarker of effect linking potential for animal and humal responses to OP exposure. This allowed for cross species comparisons for biomarkers of early biological response.
The four types of models used were:
1. Environmental and Exposure Biomarkers Multivariate Correlation Model for estimating
geometric means θk of measurement and Xiik and within - and between - person variability and
correlation by Σω and Σb
log Xiik ~ MVNm(θk+ MVNm(Ο,Σb)j, Σω) (1)
for households j = l,...,J; households' descriptions k = i,...,K; samples in household i = l,...,njk and number of means to be estimated m.
2. Biomarkers of early Biological Response Model for relating early response fj to biomarker/environmental exposure Xj in terms of their slope ßfx
fj = afM + ßfx Xj + N(0, σfx ) for individuals j = ...,j (2)
3. Biomarkers of Suscptibility Model for determining whether the genotypes of a polymorphism alter the slopes of the Biomarkers of Early Biological Response MOdel (2)
fj = af + Σi ßfx Xj + N (0, σfx ) for genotypes i = 1,.I; individualsj = 1,...J.
4. Biomarkers of Effect Showing Altered Function Model used a GO-quaint index GI = Z Σjm1 INi/n where INi are the intensities of the statistically significant genes i-1,...,n in a Gene Ontology (GO) pathway and Z is a normallized score for the GO pathway based upon the fraction of genes on the array that were statitistically significant in relation to dose.
Environmental and Exposure biomarkers: We modeled six urinary dialkyl phosphate (DAP) metabolites (dimethylphosphate [DMP], dimethylthiophosphate [DMTP], dimethyldithiophosphate [DMDTP], diethylphosphate [DEP], diethylthiophosphate [DETP] and diethyldithiophosphate [DEDTP]). We also measured several parent compounds of OP in blood, including chlorpyrifos (CP) and azinphos-methyl (AZ) in blood. We characterized environmental sources of exposure by measuring six parent compounds in home and vehicle dust (AZ, phosmet, malathion, methyl-parathion, CP, and diazinon).
Biomarkers of Early Biological Response: We evaluated activities for two cholinesterase (ChE) enzymes from blood samples collected from farmworkers and non-farmworkers, acetylcholinesterase (AChE) and butylcholinesterase (BuChE). We explored the feasibility of using buccal swabs to examine gene-expression analyses.
Biomarkers of Susceptibility: We modeled the following three classes of biomarkers of susceptibility for OP to characterize the potential for individual differences in response to OP exposure: 1) genotype of CYP450 metabolism genes; 2) geno/phenotype for paraoxonase 1 (PON1); and 3) genotype for oxidative response pathways.
Biomarkers of Effect Showing Altered Function: To evaluate biomarkers of effect showing altered function we also developed models from CHC studies using laboratory animals, mice, exposed during pregnancy to CP. This system-based GO-Quant approach identified a diversity of functional gene pathways known to be disrupted by CP and highlighted possible additional consequences of CP neurotoxicity, such as disturbance of the ubiquitin proteasome system.
Summary/Accomplishments (Outputs/Outcomes):
We have developed an integrative tool that is organized around the Public Health paradigm V- diagram, which models the occurrence of disease in response to original sources of exposure (see Figure below). The diagram identifies intermediate processes (which may be subject to public health intervention) and conditions (which may be observable for public health monitoring and hypothesis testing) along the pathway from exposure source to health effect. While the CBPR Project focused on identifying the potential for a take-home pathway of OP transfer we focused on modeling biomarkers of exposure, susceptibility, and effect as shown in the figure below.
Figure 1. Integrative tool containing models we developed in boxes A-G organized around the Public Health paradigm V‐diagram.
Environmental and Exposure biomarkers: We have developed Bayesian-based mixed effects Markov Chain Monte Carlo (MCMC) methodologies to model the potential for exposure to OPs in farmworkers and their children during the thinning, harvest and non-spray seasons. The MCMC methods allowed us to develop methods for treating the many observations below the limit of detection as censored observations. Our findings demonstrated that the day-to-day variability within an individual is frequently larger than between individuals. Thus, our multiple sampling methods allowed us to more accurately estimate the exposures within study groups by accounting for between and within person variability. Urine from farmworker adults and children had considerably higher concentrations of OP metabolites, particularly DMTP, during the thinning season compared to non-farmworkers and to nation-wide estimates from the National Health and Nutrition Examination Survey IV (NHANES IV). Furthermore, farmworker adults and children had significant correlations with house dust residues and with each other's DMTP concentration indicating the importance of the take-home pathway. DMTP was the OP metabolite with the highest concentration in the urine of the farmworkers and their children. We observed that during the thinning season, about half of farmworkers had detectable levels of AZ in blood whereas no one in the non-farmworkers group had levels above the limit of detection. In contrast, farmworkers' samples did not have detectable levels of AZ during the non-spray season, whereas only one non-farmworker had detectable levels of AZ. We observed a significant correlation between the levels of AZ in blood and concentrations of urinary metabolites, suggesting that AZ is a significant source of these metabolites.
Biomarkers of Early Biological Response: We observed a higher number of farmworkers compared to non-farmworkers having > 20% inhibition (the action level for monitoring worker exposure under the Washington State Department of Labor and Industry) for both AChE and BuChE. We observed statistically significant associations between biomarkers of exposure AZ and dimethyl DAPs and ChE inhibition indicating a direct link between pesticide exposure and these biomarkers of response. In order to characterize buccal gene expression patterns, we optimized a ribonucleic acid (RNA) isolation protocol and began exploring reverse transcription polymerase chain reaction (RT-PCR) and microarray assays.
Biomarkers of Susceptibility: We observed that polymorphism of the genes encoding cytochrome P450s (CYP450s) impacts the production of oxon, which is the active metabolite that causes AChE inhibition. For example, participants who had the A-allele in the CYP3A5 position 6986 had lower AChE inhibition than those with the G-allele. This suggests that these two groups differ in their ability to metabolize OP and may indicate different susceptibility for developing OP-related effects. Our findings suggest that PON1 status does not modify the relationship between the biomarkers of exposure and effect in our populations. This is in agreement with the idea that AZ, the most frequently detected OP among study participants, is not a substrate for PON1. We currently are evaluating the oxidative response pathways. Preliminary results show marginally significant associations between glutathione S-transferase (GST) genotype and AChE inhibition. More analysis is in progress to determine the impact of genetic variation in these oxidative response pathways.
Biomarkers of Effect Showing Altered Function: To evaluate biomarkers of effect showing altered function we also developed models from CHC studies using laboratory animals, mice, exposed during pregnancy to multiple doses of CP. We have applied a GO-Quant approach to examine the toxicogenomic responses in mice after in utero exposure to CP. This system-based GO-Quant approach identified a diversity of functional gene pathways known to be disrupted by CP and highlighted possible additional consequences of CP neurotoxicity, such as disturbance of the ubiquitin proteasome system. We explored dose-dependent alterations in toxicogenomic response in the fetal and maternal C57BL/6 mouse brain after daily gestational exposure (days 6 to 17) to CP (2, 4, 10, 12 or 15 mg/kg by subcutaneous injection to the dam). The dose-effect relationship of CP on gene expression, both at the gene and pathway levels was non-monotonic and had a different pattern than brain AChE inhibition. In the maternal brain, lower doses (4 mg/kg) influenced gene ontology (GO) categories and pathways such as cell adhesion, behavior, lipid metabolism, long-term potentiation, nervous system development, neurogenesis, and synaptic transmission. In the fetal brain, lower doses (2 and/or 4 mg/kg) significantly altered gene responses in pathways for cell division, translation, transmission of nerve impulse, chromatin modification, and long-term potentiation. In addition, some genes involved in nervous system development and signaling were shown to be specifically influenced by these lower CP doses. Other genes were shown to be mainly affected at 10 mg/kg indicating that the genes in the brain with altered function depend upon the dose of CP. We demonstrated that toxicogenomic changes in gene expression showing altered function were more sensitive than the early biological response of AChE inhibition in demonstrating the effect of CP in the fetal brain. The altered gene expression analyzed with the GO-Quant tool reflected the diversity of responses known to be disrupted by CP and highlighted possible additional consequences of CP neurotoxicity.
The models used to link biomarkers of exposure to biomarkers of effect in our integrated tool provided links to describe the cascade of events that are part of the continuum from exposure to adverse outcome described in the Computational Toxicology Research Program of EPA. Because our models used observations in a farmworker population occupationally exposed to OPs it was possible to directly link many of the events along the continuum. For example, we showed direct links between biomarkers of exposure and biomarkers of early biological response and biomarkers of susceptibility in the farmworkers and their children. We have begun the process of filling in these steps by developing PBPK models incorporating genotypes of polymorphisms of metabolizing genes for OPs. As additional data become available to provide details about these steps, it will be possible to more accurately predict effects of pesticide exposures using the models in our integrated tool.
Conclusions:
We have conceptualized and developed an integrative tool that allowed us to pose and answer important questions along stages of the pathway from exposure to disease. We determined that farmworkers and their children are exposed to OPs to a greater degree than non-farmworkers. We have identified a critical path of take-home exposure for both adults and children living in an agricultural region and we have demonstrated that we can decrease these exposures by using a public health intervention model (see Presentations Vigoren, et al., 2009). We also have identified that exposed farmworkers and non-farmworkers have differential susceptibility based on differences in genotype for metabolizing enzymes. We have shown using laboratory animal studies that at the lowest doses tested, we could observe changes in function measured by gene expression pathways in the fetal brain even when we could not detect changes in AChE inhibition, a biomarker of early biological response. Thus, the public health implications of these findings suggest the need for looking at additional ways to decrease exposure, addressing the take-home exposure by implementing evidence-based interventions and further investigating other sources of exposure for workers in the field, additional characterization of susceptibility markers, and development of screening tools to identify at risk individuals.
Journal Articles on this Report : 1 Displayed | Download in RIS Format
Other project views: | All 20 publications | 4 publications in selected types | All 3 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Stanaway I, Wallace J, Hong S, WIlder C, Green F, Tsai J, Knight M, Workman T, Vigoren E, Smith M, Griffith W, Thompson B, Shojaie A, Faustman E. Alteration of oral microbiome composition in children living with pesticide-exposed farm workers. INTERNATIONAL JOURNAL OF HYGIENE AND ENVIRONMENTAL HEALTH 2023;248(114090). |
R832733 (Final) R834514 (Final) |
Exit |
Supplemental Keywords:
Organophosphate pesticides, cholinesterase inhibition, genomic biomarkers, biomonitoring, farmworkers, children;, RFA, Health, Scientific Discipline, ENVIRONMENTAL MANAGEMENT, Health Risk Assessment, Biochemistry, Children's Health, Risk Assessment, pesticide exposure, Human Health Risk Assessment, assessment of exposure, children's vulnerablity, susceptibility, children's environmental health, biological markers, agricultural community, diseaseProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.