Grantee Research Project Results
Final Report: Estrogen Elicited Gene Expression Network Elucidation in the Rat Uterus
EPA Grant Number: R831847Title: Estrogen Elicited Gene Expression Network Elucidation in the Rat Uterus
Investigators: Zacharewski, Timothy , Chan, Christina , Harkema, Jack
Institution: Michigan State University
EPA Project Officer: Aja, Hayley
Project Period: September 1, 2004 through August 31, 2007
Project Amount: $747,960
RFA: Computational Toxicology and Endocrine Disruptors: Use of Systems Biology in Hazard Identification and Risk Assessment (2004) RFA Text | Recipients Lists
Research Category: Environmental Justice , Endocrine Disruptors , Human Health , Computational Toxicology , Safer Chemicals
Objective:
Systems biology involves the iterative development of strategies that integrate disparate physiological and biochemical data into computational models that are capable of predicting the biology of a cell or organism. In order to be predictive for hazard identification and risk assessment purposes, a comprehensive and quantitative understanding of the molecular, cellular, physiological, and toxicological effects that are elicited following acute and chronic exposure to synthetic and natural chemicals is required within the context of the whole organism. This proposal will develop a computational model that will identify critical estrogenic endocrine disruptor elicited changes in gene expression which play a central role in the observed physiological/toxic effects based on systematic and quantitative data obtained from comparative in silico, genomic, molecular and histopathological approaches.
Objective 1 Establish Estrogenic Endocrine Disruptor (EED) Elicited Dose- and Time-Dependant Changes in Rat Uterine Gene Expression
Objective 2 Investigate the Role of the ER in Mediating Changes in Gene Expression
Objective 3 Phenotypically Anchor Changes in Gene Expression to Histopathological Outcomes
Objective 4 Develop a Model that Describes the EED Elicited Uterine Gene Expression Network
Summary/Accomplishments (Outputs/Outcomes):
Objectives 1, 2 and 3
In-life time course and dose response studies for EE in immature ovariectomized Sprague Dawley rats were performed. EE elicited the characteristic uterotrophic effects (up to 7-fold induction in uterine wet weight) in a dose responsive manner (Figure 1), allowing for selection of a dose for the temporal studies that would result in optimal physiological response.
The 100 μg/kg dose of EE was thus selected and implemented in a comprehensive time course study (2, 4, 8, 12, 18, 24, and 72, 84, 96, 120, 144, 168 hr post treatment with one or three, daily doses, respectively) wherein a broad spectrum of physiological, histopathological and molecular endpoints were measured. This time course design captured the induction and subsequent regression in uterotrophy which occurred at 72 h. Temporal changes in uterine wet and blotted weights (5- and 7- fold, respectively, peaking at 72 h) were used to assess gross physiological endpoints including water imbibition, proliferation and regression. Histopathological samples were processed, mounted and stained (H&E) and evaluated by a veterinary pathologist (Co-Investigator, Dr. Jack Harkema). Subsequent findings of timed stromal edema, immune cell accumulation, basal lamina thickening, luminal epithelial hypertrophy, proliferation, and apoptosis were used to characterize the morphological changes preceding and overlapping with the uterotrophic response at 72 h. Immunohistochemical staining for BrdU nuclear incorporation into luminal epithelial cells at each time point allowed for morphological quantification of this marker of DNA synthesis and cell cycle progression. Notable induction was observed only at 18 and 24 hours with minimal to no staining at all other time points. Maximal BrdU staining (15-fold) was seen at 24 h indicating a synchronous and concerted progression of these cells through the cell cycle approximately 18-24 hours after treatment with EE. Luminal epithelial cell height was also significantly increased 1.6- and 2.7-fold at 24 and 72 h, respectively.
Clinical chemistry assessments of rat serum for BUN, creatine phosphokinase, total cholesterol, phosphorus, ALT, AST, triglycerides, magnesium, calcium, alkaline phosphatase and glucose and showed no treatment related changes in these parameters.
Custom in-house rat cDNA microarrays were used to characterize the temporal changes in gene expression involved in uterotrophic induction after a single dose of EE as well as monitor the changes at and subsequent to the uterotrophic time point (72 h) after three doses. Activated genes were categorized into distinct temporal profiles using k-means clustering which distinguished the time, duration and direction of gene expression. Differentially expressed genes were analyzed and functionally annotated using a variety of computational and literature search-based methods. The resulting functional categories represented by the EE expression profiles included transcription factors, water and ion transporters, angiogenic factors, pro- and anti- apoptotic genes, redox regulators, xenobiotic metabolism enzymes, heat shock proteins, protein synthesis genes and proteasome family members, DNA replication and cell cycle control genes and cellular energetics control genes. The expression profiles of several (20) representative genes (Figure 2) covering multiple categories of response were verified by QRT-PCR with an average temporal correlation coefficient of 0.83 between the microarray and real time expression data indicating excellent concordance between the two methods.
The temporal regulation of these families of gene responses mapped well to the time points preceding or overlapping with the histopathological and physiological endpoints which were observed in the uterus. Phenotypic linkages between gene responses and cellular or structural endpoints have been established providing a robust baseline data set for comparison to subsequent studies of estrogenic compounds.
Identical in life studies were performed examining the expanded uterotrophic effects of tamoxifen (Tam) in the rat. Dose response studies indicated a comparable EC50 oral dose for induction in uterine wet weight although with lower efficacy (Figure 1). The 100 μg/kg dose was subsequently used for the time course study. Physiological and histomorphological endpoints (wet weight, water content, luminal cell height, histopathological assessments) were captured at each time point and used to interpret temporal gene expression profiles generated by microarray analysis. A statistically significant difference was observed in the level of induction in luminal epithelial cell height after three doses of TAM observed at 72 hours compared to EE (Figure 5). This endpoint along with a decrease in water imbibition suggests a differential effect in water or solute transport that would account for cell volume and transport of fluid into the lumen. Comparative analysis of the active genes demonstrated high overlap in the number and nature of genes responding to both EE and Tam. However, a temporal delay in the modulation of all genes, especially immediate-early EE-responsive genes (Fos, Vegf, etc.), suggested either decreased absorption rate or pharmacokinetic delay due to bioactivation to the active 4-hydroxy- form. Coactivity analysis (pair-wise comparison of the actual time and direction of gene expression change) of the temporal expression profiles indicated otherwise comparable, although in most cases less efficacious (lower fold change), gene expression profiles (Figure 3).
This is consistent with comparable Tam-induction in uterotrophy in an immature, ovariectomized rat model. Thus there is no clear indication from the gene expression data to suggest a qualitative difference (mechanistic difference) as opposed to merely a quantitative difference (efficacy and pharmacokinetic shift) in TAM’s uterotrophic effects, relative to EE.
In-life dose response and time course studies for ortho,para’-dichlorodiphenyltrichloroethane (DDT) in immature ovariectomized Sprague Dawley rats were completed to evaluate the estrogenicity of DDT in the rat uterus. DDT induced uterotrophy in a dose responsive manner after treatment with 1, 3, 10, 30, 100 or 300 mg/kg b.w., with a peak of 3.4-fold induction in uterine wet weight, allowing for selection of a dose for the temporal studies that would result in optimal physiological response. The dose response study revealed DDT to behave in a manner less potent (EE and DDT elicited uterotrophic EC50s of 16.2 μg/kg and >64.6 mg/kg, respectively) and with less efficacy than EE in inducing wet weight. However, induction in blotted weights had comparable efficacy (Figure 1). DDT only induced a 3.4-fold increase in wet weight with a marked decrease in water imbibition compared to the 7-fold induction observed previously with EE.). The difference in induction in water content suggests a differential effect in water or ion transport. Several genes involved in related processes were regulated by EE including Vegf, aquaporins, solute carrier family proteins and sodium channel proteins. Subsequent examination of DDT expression profiles will provide evidence to determine if these mechanisms are involved in the EE but not DDT induction in water content among other endpoints.
DDT enantiomers have also been shown to activate MAPK pathways which are also involved in proliferative responses. So in order to demonstrate the ER-dependancy of DDT’s uterotrophic effects, the pure ER antagonist ICI 182,780 was co-administered with DDT during the dosing regimen. The uterotrophic response of 30 mg/kg DDT alone was completely blocked by ICI co-treatment (Figure 4).
Further studies are ongoing to examine the gene expression responses specifically to determine if DDT still mediates changes in gene expression despite ICI’s antagonism of the uterotrophic endpoint via the ER.
The 300 mg/kg dose of DDT was thus selected and used in a comprehensive time course study (2, 4, 8, 12, 18, 24, and 72 hr post treatment with one or three, daily doses, respectively) wherein a broad spectrum of physiological, histopathological and molecular endpoints were captured for measurement. Temporal changes in uterine wet and blotted weights (approx 3.5- fold for both at 72 h) were used to assess gross physiological endpoints including water imbibition and proliferation. Histopathological samples were processed and stained (H&E) and evaluated by a veterinary pathologist (Co-Investigator, Dr. Kurt Williams). Subsequent findings of stromal and epithelial apoptosis, immune cell accumulation, luminal epithelial hypertrophy, and proliferation characterized the morphological changes during the uterotrophic response at 72 h. Morphological measurements of luminal epithelial cell height revealed statistically lower induction in cell height when compared to ethynylestradiol positive control treated rats (Figure 5).
Custom in-house rat cDNA microarrays were used to characterize the dose-responsive changes in gene expression involved in uterotrophic induction after a single dose of DDT as well as monitor the changes at the uterotrophic time point (72 h) after three doses. Differentially expressed genes were analyzed and functionally annotated using a variety of computational and literature search-based methods. The resulting functional categories represented by the DDT expression profiles included water and ion transporters, angiogenic factors, pro- and anti-apoptotic genes, redox regulators, xenobiotic metabolism enzymes, cell cycle control genes and cellular energetics control genes. The expression profiles of several representative genes covering multiple categories of response were verified by QRT-PCR and showed excellent concordance between the two methods.
A number of genes in particular indicate a larger response in genes regulating cell death and water imbibition (Pycard and MIP/Aqp0) while classically estrogen regulated genes (C3 and Calb3) exhibited equivalent or less efficacious responses in response to DDT (Fig). These results suggest that DDT and EE elicit responses in similar genes but quantitatively differ in induction levels which can result in differences in uterotrophic and pathological effects. Several genes involved in related processes, regulated by EE including Vegf, aquaporins, solute carrier family proteins and sodium channel proteins will be further examined for their role in the differential water imbibition between the two compounds. Subsequent examination of DDT expression profiles will provide evidence to determine if these mechanisms are involved in the EE but not DDT induction in water content among other endpoints.
Following individual analyses of TAM and DDT in comparison to EE, a combined comparison of the gene expression activity was performed in which the number of unique genes activated by each compound was assessed for overlapping and uniquely expressed genes. A majority of regulated genes (~80%) were found to overlap between all three ligands. Coactivity analysis of the 1,984 overlapping genes exhibited high correlation in the direction and time of regulation between all three ligands. Those genes falling outside of the 1,984 conserved genes were typically changed less than 1.5 at isolated time points, indicating genes with lower impact in the over all response that minimally passed our screening criteria. Further analysis and data mining is currently in progress to extend these findings.
The Chan group has been actively engaged in developing techniques and approaches to integrate multi-level data have also been developed initially based on the effects of lipotoxicity. These techniques are being evaluated in light of the uterotrophic data for applicability. This includes:
- Developing a Three-Stage Integrative Pathway Search (TIPS©) framework to identify toxicity relevant genes and pathways. The ability to obtain profiles of gene expressions, proteins and metabolites with the advent of high throughput technologies has advanced the study of pathway and network reconstruction. Genome-wide network reconstruction requires either interaction measurements or large amount of perturbation data, often not available for mammalian cell systems. To overcome these shortcomings, the Chan group developed a TIPS© approach to reconstruct context-specific active pathways involved in conferring a specific phenotype, from limited amount of perturbation data. The TIPS© approach was able to reconstruct active pathways that confer a particular phenotype by integrating gene expression and phenotypic profiles. The pathways identified are specific to the cytotoxicity induced by the environmental factors. A web-based version of TIPS© is available.
- Developing a hierarchical approach to integrate a prior knowledge and gene expression with phenotype profiles to identify phenotype relevant genes and pathways. A hierarchical framework consisting of three stages was developed to integrate a priori knowledge and gene expression with metabolite profiles to identify phenotype relevant genes and pathways. First, discriminant analysis identified the metabolites that were the most relevant in differentiating various phenotypes. Second, gene set enrichment analysis (GSEA) was applied to the cDNA microarray data to identify the transcriptionally altered pathways and processes. Finally, the genes and gene sets that regulate the metabolic responses identified in step 1 were obtained by integrating the expression of the enriched gene sets and the metabolic profiles with a multi-block partial least squares (MBPLS) regression model. The hierarchical approach suggested potential mechanisms involved in mediating the cytotoxic and cytoprotective pathways, and identified novel targets as potential modulator of the toxic phenotypes. The hierarchical approach integrated gene expression data, phenotypic profile with a prior knowledge and identified new target genes to alleviate toxicity.
- Extending the genetic algorithm coupled partial least squares (GA/PLS) analysis to identify a subset of relevant genes involved in regulating multiple metabolites. In order to devise efficient treatments for complex, multi-factorial diseases, it is important to identify the genes which regulate multiple cellular processes. Thus, they developed a framework to identify such genes in the context of lipotoxicity. The genetic algorithm coupled partial least squares (GA/PLS) analysis was extended to identify a subset of relevant genes whose expression levels can predict the level of multiple metabolites. The aim was to identify the genes that could be altered to treat or ameliorate the cellular responses affected by a complex disease or environmental factors. These results demonstrate the applicability of GA/PLS in identifying the genes that regulate multiple cellular responses of interest and that genes regulating multiple cellular responses may be better candidates for countering complex diseases.
- Dynamic gene module map analysis to identify targets that modulate phenotypes.A framework was developed that integrated dynamic gene expression and metabolite profiles and applied dynamic Bayesian network analysis to reconstruct phenotype-specific dynamic network. The framework identifies pathways that regulate phenotypic responses in cells. Dynamic module map analysis is used to uncover the transcriptional regulation exerted by an environmental factor, and suggests the causal factors that triggered the phenotypic response. They further integrated gene expression and metabolites profiles with discriminant analysis and PLS regression model to identify these casual factors. Finally, dynamic Bayesian network analysis was applied to identify the dynamic network relevant to inducing the observed phenotype.
Conclusions:
These results demonstrate that EE, TAM and DDT elicit the uterotrophic response in the rat uterus with varying efficacy and potency via a common mode of action through binding and activation of ER. However, despite notable differences in both physiological and histopathological endpoints (i.e. uterine size, water content, cytoarchitecture, epithelial cell height, cell death, etc.) there is a conserved overall proliferative response as observed in the induction in uterine weight. Even so, there is little evidence to suggest that TAM or DDT’s differential binding or ER-activation events (activities through ER’s AF-1, AF-2, or both) are qualitatively different in the mRNA expression profiles that are observed. The results are consistent with other in vivo results (Korach et al. J Biol Chem 2006) suggesting non-ERE mediated proliferative pathways responsible for the uterotrophic effect. Although more comparative studies are need to further elucidate and validate the divergent or compound specific pathways knowledge of the conserved uterotrophic transcriptome is of benefit for assessing further SERM-like drugs and environmental contaminants and their evaluation across species to determine human relevancy.
The modeling approaches provide the computational power to assess phenotypic linkages between differentially active genes and physiologically and histomorphologically based endpoints. In order to devise efficient treatments for complex, multi-factorial diseases, it is important to identify the genes which regulate multiple cellular or physiological processes. We extended our GA/PLS methodology to evaluate more than one physiological endpoint and demonstrated the ability of the method to identify genes that regulate multiple phenotypic responses and that genes regulating multiple physiological responses may be better candidates for countering the negative effects of treatment. We are currently applying the approaches developed to uterotrophic data. Finally, we developed an approach based upon gene module map analysis and applied the approach to dynamic gene expression profiles to identify patterns of transcriptional regulation in cells exposed to different environmental factors. The analysis revealed potential mechanisms by which the environmental factors induced cellular toxicity as well as identified potential targets that could alleviate the toxicity. This approach may be easily applied to other systems and conditions.
Journal Articles on this Report : 14 Displayed | Download in RIS Format
Other project views: | All 90 publications | 15 publications in selected types | All 14 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Kidambi S, Udpa N, Schroeder SA, Findlan R, Lee I, Chan C. Cell adhesion on polyelectrolyte multilayer coated polydimethylsiloxane surfaces with varying topographies. Tissue Engineering 2007;13(8):2105-2117. |
R831847 (Final) |
|
|
Kidambi S, Sheng L, Yarmush ML, Toner M, Lee I, Chan C. Patterned co-culture of primary hepatocytes and fibroblasts using polyelectrolyte multilayer templates. Macromolecular Bioscience 2007;7(3):344-353. |
R831847 (Final) |
Exit |
|
Kiyosawa N, Kwekel JC, Burgoon LD, Williams KJ, Tashiro C, Chittim B, Zacharewski TR. o,p'-DDT elicits PXR/CAR-, not ER-, mediated responses in the immature, ovariectomized rat liver. Toxicological Sciences 2008;101(2)350-363. |
R831847 (Final) |
Exit Exit Exit |
|
Kwekel JC, Burgoon LD, Burt JW, Harkema JR, Zacharewski TR. A cross-species analysis of the rodent uterotrophic program: elucidation of conserved responses and targets of estrogen signaling. Physiological Genomics 2005;23(3):327-342. |
R831847 (2006) R831847 (Final) |
Exit Exit Exit |
|
Kwekel J, Forgacs A, Williams K, Zacharewski T. o-p'-DDT-mediated uterotrophy and gene expression in immature C57BL/6 mice and Sprague-Dawley rats. TOXICOLOGY AND APPLIED PHARMACOLOGY 2013;273(3):532-541 |
R831847 (Final) |
Exit |
|
Li Z, Shaw SM, Yedwabnick MJ, Chan C. Using a state-space model with hidden variables to infer transcription factor activities. Bioinformatics 2006;22(6):747-754. |
R831847 (Final) |
Exit Exit Exit |
|
Li Z, Srivastava S, Mittal S, Yang X, Sheng L, Chan C. A three stage integrative pathway search (TIPS©) framework to identify toxicity relevant genes and pathways. BMC Bioinformatics 2007;8:202. |
R831847 (Final) |
Exit Exit Exit |
|
Li Z, Srivastava S, Yang X, Mittal S, Norton P, Resau J, Haab B, Chan C. A hierarchical approach employing metabolic and gene expression profiles to identify the pathways that confer cytotoxicity in HepG2 cells. BMC Systems Biology 2007;1:21. |
R831847 (Final) |
Exit Exit Exit |
|
Li Z, Srivastava S, Findlan R, Chan C. Using dynamic gene module map analysis to identify targets that modulate free fatty acid induced cytotoxicity. Biotechnology Progress 2008;24(1):29-37. |
R831847 (Final) |
Exit Exit Exit |
|
Patil S, Melrose J, Chan C. Involvement of astroglial ceramide in palmitic acid-induced Alzheimer-like changes in primary neurons. European Journal of Neuroscience 2007;26(8):2131-2141. |
R831847 (Final) |
Exit |
|
Srivastava S, Chan C. Hydrogen peroxide and hydroxyl radicals mediate palmitate-induced cytotoxicity to hepatoma cells:relation to mitochondrial permeability transition. Free Radical Research 2007;41(1):38-49. |
R831847 (Final) |
Exit |
|
Srivastava S, Li Z, Yang X, Yedwabnick M, Shaw S, Chan C. Identification of genes that regulate multiple cellular processes/responses in the context of lipotoxicity to hepatoma cells. BMC Genomics 2007;8:364. |
R831847 (Final) |
Exit Exit |
|
Srivastava S, Chan C. Application of metabolic flux analysis to identify the mechanisms of free fatty acid toxicity to human hepatoma cell line. Biotechnology and Bioengineering 2008;99(2):399-410. |
R831847 (Final) |
Exit |
|
Yang X, Chan C. Repression of PKR mediates palmitate-induced apoptosis in HepG2 cells through regulation of Bcl-2. Cell Research 2009;19(4):469-486. |
R831847 (Final) |
Exit Exit Exit |
Supplemental Keywords:
RFA, Health, PHYSICAL ASPECTS, Scientific Discipline, POLLUTANTS/TOXICS, Environmental Chemistry, Chemicals, Endocrine Disruptors - Environmental Exposure & Risk, Risk Assessments, endocrine disruptors, Environmental Microbiology, Physical Processes, Biochemistry, Biology, Endocrine Disruptors - Human Health, altered gene expression, germ cell vulnerability, molecular mechanisms, endocrine disrupting chemicals, exposure, altered sexual development, EDCs, exposure studies, developmental biology, gestational exposure, animal models, fetal development, mice, reproductive processes, fetal genocyte degeneration, human health riskProgress 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.