421 (MOR-1117-842080) Patterns of gene expression using a cDNA Microarray in trout collected from high alpine lakes in Washington State.
Start time: 8:00 AM
Moran, P¹, Aluru, N², Vijayan, M.², ¹ US Geological Survey, Tacoma, WA, USA² University of Waterloo, Waterloo, ON, CAN
A growing number of studies have been investigating the use of cDNA libraries and microarray platforms to screen for xenobiotic responses in organisms at the transcriptional level. Fish tissue concentrations of xenobiotic chemicals are often observed at or below levels where other measures of physiological response, ie enzyme biomarkers, can be detected. It is hypothesized that even at very low levels of xenobiotics in fish tissue, the altered expression of certain genes will be indicative of tissue exposure and impact. To this end, liver tissue from rainbow (Oncorhynchus mykiss) and cutthroat (O. clarkii) trout collected from 5 high alpine lakes were analyzed with a targeted rainbow trout cDNA microarray. The 150 genes included on the microarray represent different aspects of teleost physiology; including endocrine, metabolic, immune, and stress-responsive pathways. Multivariate approaches were used to relate measured tissue and sediment concentrations of mercury and organochlorines from these lakes to patterns of gene expression. The variation in gene expression seen across all sites and the use of appropriate standard tissue will be discussed.

422 (RIS-1117-837421) Effects of chlorpyrifos exposures on juvenile rainbow trout gene expression and behavioral endpoint analyses.
Start time: 8:20 AM
Rise, M.¹, Rise, M.¹, Miller, S.¹, Schmoldt, A.¹, ¹ Great Lakes WATER Institute, University of Wisconsin-Milwaukee, Milwaukee, Wisconsin, USA
We develop tools to study biologically significant impacts of low levels of organophosphate pesticides (OPs) in water, and to resolve molecular mechanisms by which OPs can cause neurological and behavioral defects in vertebrates. OPs are widely used in U.S. and international agriculture. OPs exert their acutely toxic effects by inhibiting acetylcholinesterase, which is critical to the normal functioning of cholinergic synapses. A toxicogenomic approach was used to characterize response of the juvenile rainbow trout transcriptome to 48 hour sub-acute (1-100 ppb) waterborne chlorpyrifos exposure relative to vehicle control. A reference design microarray experiment, using 3600 gene GRASP cDNA microarrays with Genisphere Array 50 reagents and methods, identified candidate molecular biomarkers of chlorpyrifos exposure, including a novel HSP. Results were validated using quantitative RT-PCR (QPCR) on pooled samples. Subsequent QPCR of individual fish samples will provide an assessment of biological variability of biomarker expression. We created a suppression subtractive hybridization (SSH) cDNA library from chlorpyrifos-exposed and control tissues to identify novel trout genes responsive to chlorpyrifos. Rainbow trout sequences in this SSH library had significant (E-value < 1e-10) BLASTX hits against known vertebrate genes, including spinal cord-derived growth factor, annexin A5, translation factor sui1, activated protein kinase C receptor, beta crystalline B, lung cancer oncogene 7, and a novel HSP. Select SSH-identified genes will be validated by QPCR. In addition, we are developing methods for studying changes in behavior that are associated with chlorpyrifos exposure. We employ EthoVision, a video tracking system that enables automatic behavior recognition and analysis of locomotory tracks. Early results demonstrate that fish activity can be evaluated with this software. Behavioral profiling, and functional annotation of chlorpyrifos-responsive genes, will add detail to our understanding of potentially deleterious changes associated with OP exposure in fish, and provide biomarkers for assessing biological impacts of OPs on feral fish populations.

423 (BLA-1117-052413) Copper homeostasis in fish: Is regulation at the transcriptional level?
Start time: 8:40 AM
Blanchard, J¹, Kasper, A¹, Grosell, M, ¹ University of Miami, Miami, FL, USA
Copper is an essential nutrient for all organisms yet is potentially toxic. As such, organisms have evolved genes to regulate the uptake, internal transport, cellular trafficking, and excretion of copper. Fish have been demonstrated to regulate copper uptake and excretion when provided with copper above or below their physiological needs. This study aims to examine if regulation of copper uptake and excretion is at the level of gene expression in killifish, Fundulus heteroclitus. Currently, partial sequences of the killifish Cu-ATPase 7A (an exporter), two importers (CTR1 and the divalent metal transporter DMT1), the copper chaperone CCS, and the plasma Cu-carrying protein ceruloplasmin have been obtained. The expression patterns of this set of Cu specific genes are being examined in fish exposed to elevated waterborne copper as well as in copper deficient fish. Copper deficiency will be induced by injections with tetrathiomolybdate. By examining the expression patterns of genes involved in copper homeostasis we hope that a gene expression signature of copper exposure can be established for use in environmental monitoring.

424 (REY-1117-799737) Integration of a custom-made microarray into a multiple effect assessment for cadmium toxicity in Cyprinus carpio.
Start time: 9:00 AM
Reynders, H¹, Van der Ven, K¹, Moens, L¹, Van Campenhout, K¹, Bervoets, L¹, Van Leemput, K¹, De Coen, W², Blust, R¹, ¹ Laboratory of Ecophysiology, Biochemistry and Toxicology, Department of Biology, University of Antwerp, Antwerp, Belgium² Intelligent System Lab, Department of Mathematics and Informatics, University of Antwerp, Antwerp, Belgium
For a better understanding and characterisation of the toxic responses to cadmium and other contaminants in fish it is important to consider sensitive biomarkers that indicate early responses to toxicant exposure. Since reactions at the gene expression level form the basis of most toxicological responses, gene expression analysis provides an excellent method to unravel mechanisms of toxicity and to detect early and sensitive responses to toxicants. The combination of SSH-PCR and microarray technology enables the identification and evaluation of a large set of differentially expressed genes after toxicant exposure in non-model species, such as carp, thereby providing insight in the molecular reactions to toxicant exposure. Integrating gene expression information into a multi-level effect assessment can help to understand and possibly predict effects on higher levels of response. In this study carp (Cyprinus carpio) were simultaneously exposed to different cadmium concentrations via water and food (Tubifex tubifex) to induce genes reflecting both exposure routes. Fish were sampled after 3 hours, 1, 7 and 28 days to detect acute as well as sub-chronic responses. Up- and down regulated genes were isolated using suppression subtractive hybridization-PCR and spotted onto glass-slides to construct a first custom cadmium-related microarray for hepatopancreas tissue in cyprinid fish. Using these microarrays gene expression profiles were obtained for the different concentrations and time-intervals. Sequencing and classification of the differentially expressed genes into functional categories revealed an important role for metal-stress, energy and immune responsive genes. The expression level of a selected subset of the isolated genes was confirmed using semi-quantitative as well as real-time PCR. Finally, the obtained microarray information was integrated in understanding effects at the tissue, physiological and organismal level by measuring critical adverse responses such as decreased alanine transaminase (liver damage) and ion levels in plasma, growth and survival.

(58062) COFFEE BREAK.
Start time: 9:20 AM

425 (TIL-1117-832028) Possible mechanism for hepatic tumor promotion by perfluorooctanoic acid in rainbow trout: A toxicogenomic approach.
Start time: 10:00 AM
Tilton, S^{1, 2}, Orner, G^{1, 2}, Benninghoff, A¹, Hendricks, J¹, Williams, D^{1, 2}, ¹ Environmental and Molecular Toxicology, Marine and Freshwater Biomedical Sciences Center² Linus Pauling Institute, Oregon State University, Corvallis, OR, USA
Perfluorooctanoic acid (PFOA), a perfluorinated carboxylate, is considered a peroxisome proliferator (PP) in rat hepatocarcinogenesis studies. A marked species difference to peroxisome proliferation has been documented such that rodents are highly responsive and primates are relatively resistant. Previous studies with rainbow trout have shown that they are also insensitive to peroxisome proliferation by the PP, dehydroepiandrosterone (DHEA), but are still susceptible to enhanced hepatocarcinogenesis after chronic exposure. In this study, we determined whether PFOA is also a tumor promotor in trout and then examined hepatic gene expression profiles using a rainbow trout oligonucleotide microarray to further investigate possible mechanisms of action. Fish were initiated as fry to the hepatocarcinogen, aflatoxin B₁, and then fed 200-1800 ppm PFOA in the diet for 30 weeks. Two structurally diverse PPs, 1800 ppm clofibrate and DHEA, were included for comparison. PFOA (1800 ppm) and DHEA treatments resulted in enhanced liver tumor incidence and multiplicity while clofibrate showed no effect. Carcinogenesis seemed independent of peroxisome proliferation as no induction of peroxisomal -oxidation and catalase activity were observed. Alternately, plasma VTG was elevated in fish fed PFOA and DHEA suggesting that estrogenic mechanisms may play a role. We exposed juvenile trout to similar doses of PFOA, DHEA and clofibrate along with 5 ppm 17-estradiol (E2; a known tumor promotor) in the diet for 14 days to compare hepatic gene expression patterns. Both tumor promotors, PFOA and DHEA, showed strong correlation with gene expression patterns by E2 with Pearson correlation (r) values of 0.79 and 0.71, respectively. Of interest were upregulated genes known to be important for estrogen receptor signaling and vitellogenesis. In comparison, clofibrate regulated no genes in common with E2 and had a Pearson correlation value of 0.11. Overall, these data suggest a possible alternative mechanism for tumor enhancement by PFOA in a model that is relatively resistant to peroxisomal proliferation. Supported by NIH grants ES03850 and ES07060.

426 (VOL-1117-677203) Dynamic gene expression changes precede dioxin-induced liver pathogenesis in medaka.
Start time: 10:20 AM
Volz, D¹, Hinton, D¹, Law, J², Kullman, S¹, ¹ Duke University, Durham, North Carolina, USA² North Carolina State University, Raleigh, North Carolina, USA
A major challenge for toxicogenomics is predicting chemically induced morphological alterations from wide-scale gene expression changes. In this study, we exploited the aryl hydrocarbon receptor agonist dioxin (2,3,7,8-TCDD) as a model compound with well-defined toxicities and biomarkers of response (CYP1A). Adult orange-red male medaka were ip-injected once with vehicle (DMSO) or dioxin (0.1, 1.0, or 10 g/kg) and recovered for 13-d post-exposure. Subsets of fish within each treatment were sampled at days 1, 5, 9, and 13 for histologic evaluation, real-time PCR (CYP1A), and liver-specific gene expression changes using 175-gene arrays. Hepatic gene expression patterns were analyzed using mixed linear models and principal component analysis, and described relative to morphological alterations in the liver. Fish survival was >85% across all treatments during the 13-d exposure period. No histological changes were observed in livers from vehicle-treated fish at all time-points and day-1 dioxin-treated fish. However, overall hepatic transcript induction was most pronounced on day 1 in dioxin-treated fish. Relative to day 1 controls, there were 78, 161, and 158 significant transcript responses in fish treated with 0.1, 1.0, and 10.0 g/kg respectively. Following this strong day 1 response, fish exposed to 0.1 g/kg exhibited mild hepatocellular atrophy on day 5 with no adverse histological changes and minimal gene expression changes at the remaining time-points. However, moderate fatty liver and atrophy was observed on day 9 in fish exposed to 1.0 g/kg, and severe fatty liver (day 5), atrophy (day 9), and perivascular inflammation (day 13) was observed in fish exposed to 10 g/kg dioxin. While gene expression patterns approached controls after day 1 in fish exposed to 0.1 and 1.0 g/kg dioxin, transcript profiles in the highest dose were dynamic and preceded atrophy and inflammation at later time-points. CYP1A transcript levels measured by real-time PCR corroborated these dose- and time-dependent trends. In the highest dose, we also identified a gene (ependymin) that may be involved in hepatic injury repair. Overall, this study addresses complexities related to chemically induced hepatic gene expression responses, and illustrated an initial stress response followed by cytologic and adaptive changes.

427 (GRI-1116-947239) Identification of differentially expressed genes in Palaemonetes pugio following xenobiotic exposure using Serial Analysis of Gene Expression (SAGE).
Start time: 10:40 AM
Griffitt, R¹, Greig, T³, Chandler, G¹, Quattro, J^{2, 3}, ¹ University of South Carolina, Environmental Health Sciences Department, Columbia, SC, USA³ Center for Coastal Environmental Health and Biomolecular Research, Charleston, SC, USA² University of South Carolina, Biological Sciences Department, Columbia, SC, USA
We used the novel transcriptomics technique, Serial Analysis of Gene Expression (SAGE) to investigate effects of three xenobiotics on the transcriptional state of the estuarine sentinel species, Palaemonetes pugio. Adult male P. pugio were field collected, acclimated to laboratory conditions, and exposed to experimentally derived, sex-specific 96-hour LC50 concentrations of either the phenylpyrazole insecticide fipronil, the cyclodiene insecticide endosulfan, or the metal cadmium as well as an appropriate control exposure. Each exposure consisted of three replicates of 10 adult males each (30 total per treatment). At the termination of the exposure, surviving P. pugio were removed from the test solution and immediately sacrificed for RNA extraction. Equal amounts of RNA from three randomly chosen survivors were pooled for SAGE library construction. A total of four SAGE libraries have been constructed (CC, FP, ES, Cd exposures). The SAGE technique excises 17-bp ′tags′ from a defined location within mRNA molecules and concatenates them. The concatamers are cloned, sequenced, and the individual tags enumerated. Comparison of tag frequencies across treatments identify which genes are differentially expressed, as well as the magnitude of the effect. As an ′Open′ system, SAGE has certain advantages over more traditional ′Closed′approaches such as microarrays, in that SAGE requires no prior sequence knowledge, and no a priori selection of genes. The ′Control′ and ′Fipronil′ libraries were prioritized for sequencing, and initial results confirm the utility of SAGE as an ecotoxicogenomics tool. To date, 8764 tags representing 863 unique genes from the ′CC′ library, and 4844 tags representing 466 unique genes from the ′FP-exposure′ library have been sequenced, of a targeted 10,000 tags per library. Statistical comparison via the Audic-Claverie test between the two libraries identified 21 genes as being differentially expressed following application of the Bonferroni correction, and also estimated the strength of the response. For example, SAGE identified chaperonin subunit 3 as being upregulated (p-value = 0.02 ), and gamma-glutamyltransferase 3 as strongly downregulated (p-value < 0.001) following Fp exposure.

428 (DEN-1117-862896) Dietary exposure of largemouth bass to OCPs changes gene expression throughout the reproductive cycle.
Start time: 11:00 AM
Reyero, N¹, Barber, D¹, Johnson, K², Sepúlveda, M³, Gross, T^{1, 2}, Denslow, N¹, ¹ University of Florida, Gainesville, FL, USA² USGS-Center for Aquatic Resource Studies, Gainesville, FL, USA³ Purdue University, West Lafayette, IN, USA
Largemouth bass were exposed subchronically through the diet to either p,p DDE (7, 37 and 185 ppm) or dieldrin (0.6 and 3 ppm) for up to four months during their prime reproductive season. Our aim was to investigate the temporal expression patterns of several genes important for reproduction, including the three estrogen receptors (ERs), androgen receptor (AR), vitellogenin (Vtg), StAR, Cyp1A, Cyp2, Cyp3A and aromatase, among others. We also wanted to compare how different doses of p,p DDE and dieldrin would affect the expression pattern of these genes. The gene expression was measured and quantified by real-time PCR, and plasma E2 and 11KT concentrations by RIA. The predominantly expressed ER isotypes in the liver were ERs alpha and beta2, however, only ER alpha levels changed significantly over time, increasing highly during the reproductive season, in consonance with the levels of Vtg and E2. The levels of ER beta2 remain relatively constant through this time frame in the liver. ER beta1 is the least expressed isotype in the liver, but its expression peaks with the highest plasma E2 value. The expression patterns of the three ERs in gonad were distinct from the liver profile. In the gonad, ERs beta 1 and 2 are highly expressed early in the cycle and appear to be repressed with increasing plasma E2 levels. ER alpha is the least expressed. Exposure to p,&ptilde;DDE increased ER alpha and Vtg in the liver of both males and females, however dieldrin had no effect on these genes in males but did increase Vtg in females. In addition, there was a marked increase in Cyp3A, suggesting the fact that metabolism of endogenous hormones was increased. Our data shows that p,p DDE is acting mostly as a weak estrogen mimic and dieldrin affects mainly metabolism and detoxification pathways.

429 (WIN-1117-565967) The Use DNA Microarray Technology to Identify Biomarkers For Exposure of Fathead Minnow to 2,4-Dinitrotoluene.
Start time: 11:20 AM
Wintz, H.¹, Yoo, J.L.², Gibson, A.², Steevens, J.², Perkins, E.², Holman, P.¹, Loguinov, A.¹, Wong, E.¹, Vulpe, C.¹, ¹ UC-Berkeley, Berkeley, CA, USA² USACE Engineer Research and Development Center, Vicksburg, MS, USA
The production of explosive 2,4-DNT and its use in military training activities has resulted in its release to the aquatic environment through various surface water pathways. The aquatic toxicology of the DNT homologues is poorly understood and the potential risks involved in exposure to DNT are virtually unknown. The purpose of this study is to characterize the effects of 2,4-dinitrotoluene and the associated mechanisms of toxicity in the freshwater fish species, fathead minnow (Pimephales promelas), through the use of DNA microarray technology. We have constructed a fathead minnow microarray containing 5,000 anonymous cDNAs. A cDNA library was made using RNA extracted from four different populations of fish at different developmental stages between embryo and adult. 5,000 cDNA were picked from a primary library containing a total of 2 million individual clones, PCR-amplified and spotted in duplicate on glass slides. RNA extracted from the liver of fish exposed to 2,4-DNT (11, 22, 44, 88 mole/L) and from unexposed control fish were converted into cDNA, labeled using Cy3 and Cy5 and hybridized to the cDNA microarrays. Statistical analysis of hybridization signals was used to identify 72 cDNAs that are affected by exposure to different concentrations of 2,4-DNT. An analysis of the expression profiles and the identified genes and their potential role in the response to 2,4-DNT will be presented.