Scholarly article on topic 'The Herbicide Linuron Inhibits Cholesterol Biosynthesis and Induces Cellular Stress Responses in Brown Trout'

The Herbicide Linuron Inhibits Cholesterol Biosynthesis and Induces Cellular Stress Responses in Brown Trout Academic research paper on "Environmental engineering"

Share paper
Academic journal
Environmental Science & Technology
OECD Field of science

Academic research paper on topic "The Herbicide Linuron Inhibits Cholesterol Biosynthesis and Induces Cellular Stress Responses in Brown Trout"


Science sTecnno ogy

Subscriber access provided by SELCUK UNIV


The herbicide linuron inhibits cholesterol biosynthesis and induces cellular stress responses in brown trout

Tamsyn M Uren Webster, Mandy H Perry, and Eduarda M. Santos

Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/es505498u • Publication Date (Web): 29 Jan 2015

Downloaded from on February 11, 2015

Just Accepted

"Just Accepted" manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides "Just Accepted" as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. "Just Accepted" manuscripts appear in full in PDF format accompanied by an HTML abstract. "Just Accepted" manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). "Just Accepted" is an optional service offered to authors. Therefore, the "Just Accepted" Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the "Just Accepted" Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these "Just Accepted" manuscripts.

Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036

Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

The herbicide linuron inhibits cholesterol biosynthesis and induces cellular stress responses

in brown trout

Tamsyn M. Uren Webster1*, Mandy H. Perry2 and Eduarda M. Santos1*

1 Biosciences, College of Life & Environmental Sciences, Geoffrey Pope Building, University of Exeter, Exeter, United Kingdom

2 Blood Sciences Department, Royal Devon & Exeter Hospital, Exeter, United


KEYWORDS: Illumina, transcriptomics, anti-androgen, salmonid, sequencing


The herbicide linuron is used worldwide, and has been detected in surface waters, as well as in food and drinking water. Toxicological studies have reported that linuron acts as an anti-androgen, in vitro and in vivo, and disrupts mammalian male reproductive function. However, global mechanisms of linuron toxicity are poorly documented. We used RNA-seq to characterise hepatic transcriptional response of mature male brown trout exposed for four days to 1.7, 15.3 and 225.9 ^g linuron/L. We identified a striking decrease in the expression of transcripts encoding the majority of enzymes forming the cholesterol biosynthesis pathway. We also measured a very significant decrease in total hepatic cholesterol in fish exposed to 225.9 ^g/L, and a negative correlation between total cholesterol and linuron treatment concentration. We hypothesise that inhibition of cholesterol biosynthesis may result from the disruption of androgen signaling by linuron. Additionally, there was increased expression of a number of transcripts involved in cellular stress responses, including cypla (up to 560-fold), molecular chaperones and antioxidant enzymes. We found some evidence of similar patterns of transcriptional change in fish exposed to the environmentally-relevant concentration of linuron, and further research should investigate the potential for adverse effect to occur following chronic environmental exposure.


Linuron is a substituted phenylurea herbicide that disrupts photosynthesis by targeting protein D1, a central component of photosystem II, and inhibiting photo-dependent electron transport, leading to accumulation of reactive oxygen species (ROS) in plant cells1. This compound is used worldwide to control a number of broadleaf and grass weeds in the cultivation of a variety of crop plants, particularly vegetables and cereals. Linuron is known to enter surface waters in agricultural runoff, particularly in association with sediment, where it is known to be moderately persistent2. Despite the widespread usage of this herbicide, measurements of surface water concentrations are very scarce, but have been reported to occur in the ng to low ^g/L range. Concentrations detected include 1.05 ^g/L in a Canadian river within an agricultural catchment3, and 4.42 ^g/L in a Florida stream receiving agricultural run-off 4 Modelling approaches have predicted peak concentrations of 31.3 ^g/L in surface waters associated with application on a nearby carrot crop, highlighting the potential for short-term peaks in contamination to occur2. Linuron has also been detected in drinking water and in food residues5' 6. The potential for environmental exposure to this chemical, therefore, raises concerns about the risk linuron may pose to both human and wildlife health.

The majority of existing research investigating the toxicological effects of linuron has focused on its activity as an anti-androgenic compound. In vitro studies have shown that linuron competitively inhibits androgen binding to the androgen receptor (AR) in mammals, fish and amphibians7-11. In mammalian studies, anti-androgenic activity has been demonstrated in vivo, using the Hershberger assay, where exposure to linuron was shown to reduce the weight and development of androgen-sensitive reproductive tissues12-14. Linuron was also shown to have adverse impacts on male reproductive health in rats, including

abnormal reproductive development following in utero exposure, Leydig cell tumourgenesis and reduced testosterone production in vitro and in vivo15-17.

In stickleback, the production of spiggin, an androgen-dependent glycoprotein normally only produced by nest-building males, can be induced in females by androgen treatment and its subsequent inhibition by anti-androgens is assumed to occur specifically through AR antagonism18-21. Linuron was reported to suppress the production of dihydrotestosterone (DHT)-induced spiggin production in cultured kidney cells and, in vivo, concentrations of 100 and 250 ^g/L reduced spiggin induction at both transcript and protein level18-21. Recent transcriptomic and proteomic approaches in fathead minnow ovarian cells and in zebrafish embryos have demonstrated that the molecular signatures following exposure to linuron are more similar to that of the model anti-androgen, flutamide, than a model androgen (DHT) or oestrogen (ethinyl-estradiol), supporting an anti-androgenic mechanism of action22' 23.

It is hypothesised that, in addition to its action as an AR antagonist, linuron may also disrupt androgen synthesis and/or metabolism, but the relative contribution of each mechanism to overall anti-androgenic activity is poorly understood17. Linuron has also been shown to induce anti-estrogenic effects; reducing plasma vitellogenin concentrations in female fathead minnows24, 25. Linuron has therefore been shown to have multiple mechanisms of toxicity, and it is possible that others have not yet been identified. Furthermore, the majority of studies investigating the molecular mechanisms of toxicity of linuron, as well as other anti-androgens, have focused on the gonads, but androgen signalling also has an important role in regulating a range of other biological processes in different tissues, including the liver26. Hepatic metabolism is also the major mechanism responsible for steroid hormone degradation, as well as detoxification and elimination of xenobiotics27. This makes the liver an important and sensitive tissue for investigating toxic effect of chemical exposure. This

study, therefore, aimed to investigate global hepatic mechanisms of toxicity of linuron in brown trout (Salmo trutta) using transcriptomic analysis. We hypothesised that linuron would result in toxicological effects in the liver mediated through a range of molecular pathways including via anti-androgenic mechanisms of action. In order to address this hypothesis, we conducted an exposure of sexually mature male brown trout to three treatments of linuron (2.5, 25 and 250 ^g/L), including an environmentally-relevant concentration and a high concentration, chosen to facilitate a mechanistic analysis of the effects seen. We then investigated the molecular mechanisms of toxicity of this herbicide in the liver of exposed fish using RNA-seq.


Chemical exposure

Sexually mature male brown trout were exposed to linuron via a flow through system for a period of 4 days. The treatment groups consisted of three nominal concentrations of linuron; 2.5, 25 and 250 ^g/L (Pestanol Analytical Standard, Sigma) or dilution water control alone. These included a low concentration within the range that has been measured in the most contaminated surface waters and an intermediate concentration representing concentrations that may potentially occur during peak contamination events, according to modelling scenarios. Both of these concentrations were lower than the chronic NOEC of <42 ^g/L established for rainbow trout using the early life cycle test2. The high concentration was included in this study to facilitate a mechanistic investigation of molecular pathways disrupted by this chemical. This concentration is known to cause toxicological effects in fish, including anti-androgenic responses, but is considerably lower than the acute LC50 of 3 mg linuron/ L established for rainbow trout2.

Each linuron treatment group consisted of one tank containing 8 individual fish (males and sexually immature males and females, of which only sexually mature males were taken forward for analysis in order to limit variation due to sex and reproductive stage of maturation). Sexually mature females, which secrete high levels of natural oestrogens, were excluded to limit potential interference of excreted sex steroids with the exposure (see supporting information). Due to the size of the fish used in this experiment it was unfortunately not feasible to employ more than one tank replicate for each linuron treatment, however two duplicate tanks were included for the control treatment to provide us with information on inter-tank variation. Water samples were collected from each tank on day 3 of the exposure period and were analysed using LC-MS by an accredited laboratory (South West Water, Exeter Laboratories). Fish were not fed during the exposure. Full details on fish husbandry and the exposure experiment are given in the supporting information.


Fish were humanely sacrificed on day four of the exposure period by a lethal dose of benzocaine (0.5 g/L; Sigma-Aldrich) followed by destruction of the brain, in accordance with UK Home Office regulations. For each individual fish, wet weight and fork length were recorded and the condition factor (k= (weight (g) x 100)/ (fork length (cm)3)) was calculated. Sex and maturity of individuals was confirmed by observation of the gonads, and mature males were selected for analysis (n=3-6 per treatment group). Livers were dissected and weighed, and the hepatosomatic index (HSI) (liver weight (mg)/ total weight (mg)) x 100)) and gonadsomatic index (GSI) (gonad weight (mg)/ total weight (mg)) x 100)) were determined. Portions of the liver were randomly selected, snap frozen in liquid nitrogen and stored at -80°C prior to transcript profiling and cholesterol quantification. Statistical analysis of morphological parameters was conducted in SigmaStat (version 12.0). All morphometric

Page 7 of 33

Environmental Science & Technology

140 data met assumptions of normality and equal variance and was analysed using single factor

141 one way analysis of variance (ANOVA). All values presented are mean ± SEM.

142 Illumina sequencing and transcriptomic analysis

143 Transcript profiling was conducted in the liver of three randomly selected individual

144 mature males per treatment group, including for each duplicate control tank. RNA was

145 extracted and spiked with External RNA Controls Consortium (ERCC) spike-in control

146 mixes (Ambion) then prepared for sequencing using the Illumina TruSeq RNA Sample

147 Preparation kit. The 15 individual libraries were multiplexed and sequenced on an Illumina

148 HiSeq 2500 platform to generate 100 bp paired-end reads. Full details of the transcriptome

149 assembly are given in the supporting information. Briefly, raw sequence reads were subject to

150 quality-related processing, filtering and digital normalisation, then a de novo transcriptome

151 assembly was conducted using Trinity (version r2013-02-25;28), specifying default

152 parameters. Transcripts were annotated using Blastx against Ensembl peptide databases using

153 an e-value cut off < 1e-15, and additional annotation of previously un-annotated differentially

154 expressed transcripts was performed using Blast (< 1e-15) against refseq, nr and nt databases.

155 Full details of the transcriptomic analysis are described in the supporting information.

156 Briefly, 83.2 % of raw sequence reads from each of the 15 individual libraries were re-

157 mapped to the newly assembled brown trout liver transcriptome using Bowtie2 (version

158 2.1.0,29). Raw count data for each transcript was extracted using idxstats in samtools (version

159 0.1.18,30) and input into edgeR31 for differential expression analysis. Initially, we performed

160 expression analysis between the two independent control groups and identified only 3

161 differentially-expressed transcripts. This suggests that, for these groups at least, there was

162 minimal inter-tank variation in our experimental conditions which only employed one

163 replicate tank for each of the linuron treatments.

Subsequently, we conducted transcript expression analysis between the three individuals in each linuron treatment group against the combined control groups (six individuals). Transcripts were considered differentially expressed with a FDR < 0.1 (Benjamini-Hochberg correction). Heatmaps were generated using Euclidean cluster analysis, using the pheatmap package in R/Bioconductor32 based on the expression levels of all transcripts that were differentially-expressed in at least one treatment group. Functional analysis was performed for differentially expressed genes from each treatment using the Database for Annotation, Visualisation and Integrated Discovery (DAVID v6.7;33), and Ingenuity Pathways Analysis (IPA; Ingenuity Systems, Gene ontology (GO) terms and Kegg pathways were considered significantly enriched with an adjusted P value < 0.05 (Benjamini-Hochberg correction).

The raw sequence data, and processed results from the expression analysis have been deposited in NCBI's Gene Expression Omnibus (, and are available via the GEO series accession number GSE57490.

Liver cholesterol measurements

Total lipids were extracted from liver samples using chloroform/methanol according to previously published protocols34, and is described in detail in the supporting information. Liver from all of the mature males from each treatment group were analysed (n=6, 4, 4 and 3, respectively, for the control, 2.5, 25 and 250 ^g/L treatment groups). Briefly, liver samples (~50 mg) were ground under liquid nitrogen and extracted with a 20x (v/w) solution of chloroform/methanol (2:1). A 0.9% NaCl solution was used to isolate the lipid containing (lower) phase and this was washed three times with a methanol/water (1:1) solution. Lipid extracts were evaporated under a stream of nitrogen, then re-dissolved in 100% ethanol (0.5 mg tissue/^l) before measurement of cholesterol. Total cholesterol was measured using the

enzymatic colorimetric test originally described by Allain et al.35 on a Roche Modular P800 analyser (Roche Diagnostics, USA) at the Blood Sciences Department, Royal Devon & Exeter Hospital, Exeter. All data met assumptions of normality and equal variance, and was analysed using ANOVA, followed by the Tukey post hoc test. The association between the measured concentration of linuron in the water and the concentration of cholesterol in liver samples was tested using Pearson Correlation. All values presented are mean ± SEM.


Water chemistry and morphological parameters

The measured concentrations of linuron on day 3 of the exposure were 1.7, 15.3 and 225.9 ^g/L for the tanks exposed to nominal concentrations of 2.5, 25 and 250 ^g/L, respectively. Throughout this paper we refer to these measured concentrations of linuron. Gonadal examination revealed that the number of mature males was 4, 4 and 3, respectively, in the 1.7, 15.3 and 225.9 ^g/L treatment groups, and 3 in each of the duplicate control treatments. The mean mass and length of these mature males across treatment groups was 452.7 ± 14.3 g and 33.9 ± 0.3 cm, respectively. There were no significant differences in size and condition factor (mean 1.15 ± 0.01), HSI (mean 1.04 ± 0.04) or GSI (mean 3.82 ± 0.27) for mature males between treatment groups and we observed no alteration in the general health or behaviour of the exposed fish during the experiment.

Sequencing, transcriptome assembly and quality control

The de novo transcriptome was assembled using a total of 225.3 million paired 100 bp reads from male brown trout liver, and consisted of 172,688 transcripts (107,095 loci) with a mean length of 767.5 bp and a N50 of 1292 bp. 62,236 transcripts were annotated using

blastx against Ensembl peptide databases, and these included representation of 16,121 unique zebrafish transcripts. Sequencing of the liver samples of fish exposed to linuron and associated controls generated a total of 137.9 million reads, averaging 9.2 million reads per sample, and 83.3 % of these re-mapped against the assembled transcriptome.

The ERCC-spike in controls were analysed to provide technical validation of the quantitative expression profiling conducted in this study. Only ERCC control transcripts that were detected in a minimum of three individual libraries (at least one count) were included in the analysis. There was a strong correlation between the calculated FPKM values and the expected concentration of control transcripts for all individual libraries (mean R2 =0.901 ±0.005 (SEM); Figure S3). The dynamic range was calculated individually for all samples (Table S1), and the mean dynamic range in expression levels across all libraries was 21,062 FPKM. Additionally, there was a good correlation between the calculated and expected fold changes in expression between samples spiked with ERCC mix 1 and mix 2 (R2=0.58). Together, these results provide strong technical validation for the quantitative expression profiling conducted in this study.

Transcript expression and functional analysis

Expression analysis using edgeR identified a total of 822 transcripts that were differentially expressed in one or more treatment group compared to the controls, 435 of which increased and 387 which decreased in expression (FDR < 0.1). The number and the overlaps between the lists of differentially expressed transcripts for each treatment group are shown in Figure 1. A full list of differentially expressed transcripts and associated FDR values are presented in Table S3. Clustering of all individual samples based on expression levels is presented in Figure 2, and on mean fold change for each treatment group compared to the control is shown in Figure S1. Additionally, multidimensional scaling (MDS) plots

illustrating the similarity of individuals within each treatment group compared to the controls are shown in Figure S2.


(2) /\(27) /

/ 0 \ \

35 5 628

(11) (3) (420)

15.3 \xg/l 225.9 ng/L

Figure 1

Cluster analysis indicated that the expression profiles of all individual fish exposed to 225.9 |ig linuron/L were distinct from those in the control and in treatment groups exposed to lower concentrations of linuron, corresponding to the highest number of differentially expressed transcripts in this treatment group (684). For the lowest concentration of linuron (1.7 |ig/L) there was a similar clustering of individuals, although less distinctly from the control group, reflecting the degree of transcriptional change detected following exposure to this environmentally-relevant concentration (154 differentially-expressed transcripts). In contrast, the individuals exposed to (15.3 |ig linuron/L) did not cluster together, and it was clear that there was a more variable transcript expression profile in this group (Figure S2). In particular, one individual displayed a distinct expression profile from the other two individuals in this group. This increased degree of biological variation between individuals

250 reduced the statistical power for expression analysis, and subsequently resulted in the

251 identification of fewer differentially expressed transcripts in this group (42).

The list of over-represented GO terms and Kegg pathways in fish exposed to 225.9 ^g linuron/L is shown in Table S2. The most significantly enriched terms were protein folding and unfolded protein binding, reflecting an increased expression of transcripts encoding a suite of molecular chaperones involved in cellular stress response. Transcript expression of CYP1A was increased by the greatest extent (340-560 fold) in fish exposed to 225.9 ^g linuron/L, and was also significantly increased in fish exposed to 1.7 ^g linuron/L (4.1-6.7 fold). In addition, there were increasing trends in the expression of cypla in fish exposed to 15.3 ^g linuron/L (4.6-8.5 fold), albeit not significantly.

Lipid biosynthetic process and steroid biosynthesis were also amongst the most over-represented processes in fish exposed to the high treatment concentration. Ingenuity pathway analysis identified a very significant over-representation of the cholesterol biosynthesis pathway in the lists of differentially expressed transcripts. This included reduced expression of transcripts encoding 14 individual enzymes involved in this pathway in fish exposed to 225.9 ^g linuron/L. One of these transcripts (encoding lanosterol synthase) was also significantly supressed in fish exposed to 1.7 ^g linuron/L. A schematic illustrating the down-regulation of this pathway is shown in Figure 3a.

Liver cholesterol measurement

The mean concentration of total cholesterol in the liver of mature male fish included in this experiment was 2.07 ± 0.05, 1.97 ± 0.05, 1.82 ± 0.09 and 1.30 ± 0.20 mg/g of liver tissue in fish from the 0, 1.7, 15.3 and 225.9 ^g linuron/L treatment groups, respectively. There was a significant decrease in the concentration of total cholesterol measured in the livers of fish exposed to 225.9 ^g linuron /L compared to those in the control group (Figure 3b). Additionally, there was a strong negative correlation between the measured concentration of total cholesterol in the liver and the treatment concentration of linuron (p = 5.88 x10-5).

Figure 3


RNA-seq analysis revealed a considerable degree of transcriptional change in fish exposed to 225.9 ^g/L of linuron during this short-term (4 day) exposure. This high concentration is in a range where a number of toxicological effects have previously been reported for fish exposed to linuron, including anti-androgenic effects in stickleback following 21 day exposure to 100 and 250 ^g linuron/L18-21 and histopathological changes in the liver and kidney of rainbow trout exposed to concentrations of 30 ^g/L and above for 5 weeks36, and is well above the established chronic LOEC for trout (<42 ^g/L)2 There were also marked transcriptional changes in fish exposed to the environmentally-relevant concentration of linuron (1.7 ^g/L), and we observed some broadly similar changes in the genes and pathways that were affected across treatment groups. This highlights the potential for transcriptomic analysis to identify sensitive signatures of chemical exposure, when compared to higher mechanistic doses, which may be useful for assessing exposure in the environment. Previously, some evidence of anti-estrogenic effects in female fathead minnows exposed to concentrations as low as 1 ^g linuron/L have also been reported, although this was following a longer-term (21 day) exposure24. Together, these datasets suggest that environmental exposure to linuron may, potentially, be a cause for concern and this should be further investigated.

The relatively low number of differentially expressed transcripts observed in the 15.3 ^g/L treatment group reflects the larger degree of biological variation between individuals in this group, which reduced the statistical power for identifying statistical differences in expression. In particular, one individual showed distinct transcriptional changes from the others, perhaps reflecting different threshold concentrations of response between individuals. In this experiment it was only feasible to perform RNA-seq analysis on three fish per treatment,

restricting our ability to detect differentially expressed transcripts due to the high level of biological variation in this group. In addition, all three individuals analysed were exposed within the same exposure tank, further limiting the power of our analysis. These results serve to highlight the importance of optimising the number of biological replicates in RNA-seq analysis, which is rapidly becoming more practical with developments in sequencing technology. However, the robustness of our data is assured by the strong clustering of individuals in the other treatment groups, and similarities in transcriptional changes between groups exposed to the three concentrations of linuron.

Cholesterol biosynthesis

Functional analysis revealed a striking reduction in expression of transcripts encoding for the majority of enzymes involved in the cholesterol biosynthesis pathway following exposure to the highest concentration of linuron (225.9 ^g/L). Although largely not statistically significant, there were also general trends towards reduced expression of transcripts encoding the other enzymes in this pathway, as well as in the lower treatment groups (Figure 3 a). Based on these results we hypothesised that liver cholesterol biosynthesis would be reduced following exposure to linuron, and we tested this by measuring the concentration of total cholesterol in the livers of exposed fish. There was a strong reduction in total cholesterol in the livers of fish exposed to the 225.9 ^g linuron/L, verifying this hypothesis. In addition, the hepatic cholesterol concentrations in all exposed fish were strongly negatively correlated with linuron concentrations in the exposure water. Together, these data demonstrate that the transcriptional changes observed resulted in effects at the biochemical level in liver cells of exposed fish.

Cholesterol biosynthesis is a well characterised pathway consisting of a series of complex reactions involving more than 20 enzymes. Briefly, a precursor molecule, Acetyl CoA, is

converted through the mevalonate pathway to lanosterol. This stage includes the synthesis of mevalonic acid by 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR), which is generally regarded as the major irreversible, rate-limiting step in the biosynthesis of cholesterol. Lanosterol is then converted to cholesterol via a series of successive demethylation and double bond reductions, through either the Bloch or Kandutsch-Russell pathways37, 38.

Members of the sterol regulatory element binding protein (SREBP) family of transcription factors are involved in controlling a number of aspects of lipid and sterol metabolism. Specifically, the isoform SREBP-2 is a major transcriptional regulator of cholesterol biosynthesis37. Highly regulated feedback mechanisms are responsible for controlling the activity of SREBPs. Sterol-sensing SREBP-cleavage-activating proteins (SCAPs) bind and retain inactive SREBP precursors. When cholesterol levels are depleted, SCAP dissociates from endoplasmic reticulum (ER) membranes and transports SREBP-2 to the Golgi complex where it is cleaved and activated by site 1 and site 2 proteases. Activated SREBP-2 then moves to the nucleus and induces the transcription of target genes by binding sterol response elements (SREs) in their promoter regions, together with associated cofactors. Conversely, elevated levels of cholesterol stimulate the inactivation of SREBP through its re-association with SCAP in the ER membrane39. Nearly all genes encoding cholesterol biosynthesis enzymes have been shown to be regulated by SREBP-2 in mammalian studies; Sharpe and Brown38 describe 22 enzymes involved in the cholesterol biosynthesis pathway, 21 of which are regulated by SREBP-2. These include all 14 transcripts that showed significantly reduced expression in fish exposed to 225.9 ^g linuron/L. Furthermore, the expression of SREBP-2 encoding transcript (srebf2) was also significantly lower in fish exposed to 225.9 ^g linuron/L, suggesting that linuron disrupts the regulation of cholesterol biosynthesis by reducing the transcription of srebf2.

353 We hypothesise that the anti-androgenic activity of linuron is a likely mechanism by which

354 it down-regulates SREBP-2, and cholesterol biosynthesis. In prostate cancer cell lines

355 androgens are known to stimulate the expression of cholesterol biosynthesis enzymes, by

356 increasing the transcription of SREBP and enhancing SCAP-mediated cleavage of precursor

357 SREBP into the mature form40, 41. In vivo, the expression of SREBP and cholesterol

358 biosynthesis enzymes were reduced following castration in male rats, and restored following

359 androgen treatment42. In fish, transcriptomic profiling of male fathead minnow exposed to

360 pulp and paper mill effluent, which has been shown to have androgenic activity, revealed an

361 increased expression of transcripts encoding cholesterol biosynthesis enzymes43. Potentially,

362 the inhibition of cholesterol biosynthesis may be a mechanism of toxicity shared with other

363 chemicals that antagonistically interact with the AR. Although some previous studies have

364 reported that anti-androgen exposures modulate the expression and activity of individual

365 enzymes involved in cholesterol metabolism, and alter serum cholesterol and lipid

366 concentrations23, 44, the down-regulation of cholesterol biosynthesis has not been generally

367 associated with anti-androgen toxicity in environmental studies.

368 Cholesterol is an essential component of cellular membranes, where it is involved in the

369 regulation of membrane fluidity and permeability, transmembrane transport and signalling.

370 Cholesterol is also the precursor of a number of other essential biological molecules

371 including bile acids and steroid hormones. Bile synthesis in the liver is critical for the

372 elimination of endogenous and xenobiotic metabolites37. We observed an increase in the

373 expression level of cyp7a1, which encodes the rate limiting enzyme responsible for biliary

374 acid formation from cholesterol, possibly suggesting a compensatory response to the

375 reduction in cholesterol biosynthesis. Disruption of cholesterol homeostasis has been

376 implicated in a number of human pathologies including prostate cancer, in which androgen

377 signalling is a principal factor, dementia, diabetes and Alzheimer's disease. Smith-Lemli-

Opitz syndrome, which is caused by a mutation of one of the final enzymes in the cholesterol biosynthesis pathway (DHCR7), results in depleted cholesterol and is characterised by a wide range of clinical effects including impaired cellular membrane function and steroid production, resulting in developmental and behavioural abnormalities45. The liver is the primary organ responsible for vertebrate cholesterol production, therefore prolonged inhibition of cholesterol biosynthesis would likely result in a number of adverse health effects. In order to more fully assess the long term health effects of exposure to linuron, it would be important to investigate if the down-regulation of cholesterol biosynthesis enzymes and the depletion of hepatic cholesterol observed here is maintained following chronic exposure, or restored via compensatory cellular responses.

Although the majority of cholesterol for sex steroid production is synthesised in the gonads and adrenal gland, disruption of cholesterol biosynthesis may potentially contribute to the known anti-androgenic effects of linuron in reducing androgen synthesis, in particular if similar down-regulation of the cholesterol biosynthesis pathway is also occurring in those organs. To our knowledge the impact of linuron on cholesterol biosynthesis in the testis has not been investigated, although Ornostay et al.23 report reduced expression of a number of transcripts encoding cholesterol biosynthesis enzymes in fathead minnow ovarian cell cultures exposed to linuron. It is possible that an inhibition of cholesterol biosynthesis may lead to a reduction in sex steroid biosynthesis, which in turn may have also contributed to the anti-oestrogenic effects reported for fathead minnow by these authors in this study and in others23-25, potentially linking the inhibition of cholesterol biosynthesis to the reproductive toxicity of linuron.

Androgen-signalling is also important in regulating other aspects of lipid metabolism, including through SREBP-1 and the nuclear liver X receptor (LXR)40, 46. We found a

significant increase in expression of SREBP-1 in the 225.9 ^g linuron/L treatment group, but a less consistent response of known SREBP-1 regulated genes. A number of GO terms relating to wider lipid and fatty acid metabolic processes were also enriched. This is consistent with the results of previous studies that have reported alteration of the metabolism and transport of lipids to be among the most common processes regulated by androgens and anti-androgens in fish26.

Stress response

Functional analysis of the lists of differentially-expressed transcripts revealed that exposure to linuron caused a significant cellular stress response. Linuron induced a large and significant induction of transcripts encoding CYP1A, including at the environmentally-relevant concentration. CYP1A is a primary phase 1 biotransformation enzyme involved in the detoxification or metabolic activation of a number of xenobiotics, as well as many endogenous compounds47. It is amongst the most readily induced cellular proteins, and it is primarily regulated via aryl hydrocarbon receptor (AhR) signalling48. Planar aromatic hydrocarbons, including PCBs and PAHs, are known to be strong agonists of the AhR, and CYP1A induction (gene and protein) has been extensively used in ecotoxicology as a marker of exposure to these environmental contaminants49. Amongst commonly used pesticides, linuron was reported to be one of the most potent activators of AhR in vitro and also been shown to strongly induce CYP1A gene expression in mouse liver, and this was attributed to a structural feature, a dichlorophenyl residue50. In fish, CYP1A was shown to be induced by, and to metabolise, linuron in Japanese eel liver 51. In mammals, linuron is metabolised to 1-(3,4-dichlorophenyl)-3-methoxyurea, 3,4-dichloroanaline and 3,4-dichlorophenylurea, which are known to be weaker AR antagonists than the parent compound52, 53 , although little else is known about the toxicity of linuron metabolites in mammals, or in fish. Some compounds are

detoxified by CYP1A activity while others, including some PAHs, are bioactivated. The latter can generate highly reactive intermediates that subsequently induce cellular damage, including genotoxicity and carcinogenesis47. CYP1A transcription can also be regulated by hormones, directly or indirectly via upstream signalling pathways, which can affect the biological effects of xenobiotics. Androgens are known to inhibit CYP1A expression in mammals, and this is suggested to occur via AR-AhR interactions54. In addition to classical regulation via direct AhR activation, it is possible that the anti-androgenic effects of linuron contributed in part to the large induction of CYP1A, either through antagonism of the AR and/or the reduction in androgen production.

Protein folding was the most significantly enriched GO Biological Process in the list of differentially expressed transcripts following exposure to 225.9 ^g linuron/L, corresponding with a consistent increase in expression of a number of transcripts encoding molecular chaperones which can bind and stabilise damaged proteins. The chaperonin containing TCP-1 complex (CCT) is made up of eight primary subunits, six of which (cct2, cct3, cct4, cct5, cct6a, cct8), showed significantly increased expression following exposure to 225.9 ^g linuron/L (by 2.3-3.6 fold). Although not statistically significant, several of these (cct3 and cct5) also showed an increased expression by more than 2 fold in both of the lower treatment concentrations. CCT participates in normal protein metabolism by folding many proteins, particularly actin and tubulin55, but has also been shown to be induced in response to stress, aiding cellular recovery. Examples include induction by chemical stress in mammalian cells56 and temperature stress in fish57. There was a strong increase in expression of transcripts encoding several heat shock proteins (HSPs) in fish exposed to 225.9 ^g linuron/L (hsp90aa1.2, hsp90ab1, hspa4a, hspa8, dnaja1l, dnaja4, dnajb1a, dnajb1b). HSPs are probably the most well-known stress-inducible molecular chaperones, and have been extensively reported to respond to various environmental stressors in fish, including many

pesticides. HSP90 proteins are also known to have a role in chaperoning CYP1A49. This strong, and consistent, increase in expression of transcripts encoding molecular chaperones suggests than linuron induces protein-damaging cellular stress.

There was an increase in the expression of a suite of transcripts encoding glutathione-related antioxidant enzymes (gpx1b, gstal, gsto1, mgst1, mgst3) and glutathione reductase (gsr) in fish exposed to 225.9 ^g linuron/L (by 2.7-10 fold). Furthermore, there was a significant increase in expression of two members of the nuclear factor erythroid-derived 2-like family (nfe2l1b, nfe2l2a) following exposure to 225.9 ^g linuron/L. These transcription factors play a key role in regulating the response of the antioxidant system58. This suggests that linuron generates oxidative stress, through the reactivity of the parent compound and/or metabolites. This mechanism of toxicity is common to many chemicals, particularly at high concentrations. While evidence linking linuron with generation of oxidative stress in fish, and other species is scarce, other phenylurea-based pesticides with a similar chemical structure, including diuron, have been shown to induce oxidative stress, and cellular damage59, 60. Oxidative stress has been linked to pathological changes in the liver, including necrosis, apoptosis and carcinogenesis61. Linuron exposure was previously found to cause lesions and a range of adverse effects on cellular components in the liver of rainbow trout exposed to concentrations of 30 ^g/L and above for five weeks36. Although we only found significant differences in the regulation of these stress-responsive processes in fish exposed to 225.9 ^g/L, a concentration unlikely to occur in the environment, the possibility that chronic exposure to linuron may cause oxidative stress at environmentally-relevant concentrations cannot be excluded, and further research is required to investigate this.

Environmental implications

Overall, using RNA-seq we have demonstrated that a high concentration of linuron (225.9 ^g/L) induces considerable transcriptional changes in the liver of mature male brown trout. There was evidence of a striking decrease in expression of transcripts encoding the majority of the enzymes involved in the cholesterol biosynthesis pathway, likely via SREBP-2 interaction, which resulted in a reduction in the concentration of total cholesterol in the liver following exposure to 225.9 ^g/L of linuron. We hypothesise that the anti-androgenic activity of linuron is the likely mechanism responsible for this effect. Additionally, we found differential expression of a number of transcripts involved in cellular stress responses. In particular, there was a considerable increase in cypla expression, including at environmentally-relevant concentrations, and increased expression of molecular chaperones that bind and stabilise damaged proteins, as well as a number of enzymes involved in the antioxidant system.

It is important to note that the majority of the changes observed for these processes occurred only at a concentration far higher than those currently measured in water systems, but effects at environmentally relevant concentrations were also observed. It is possible that some of the transcriptional changes observed, as well as reduced liver cholesterol concentration, may reflect an acute, adaptive stress response to this short-term linuron exposure and may not lead to long term adverse effects. Toxicological responses to linuron may change over time, however effects mediated through interaction with key transcription factors, such as the AhR and SREBP-2 may be likely to continue. Future research should investigate whether similar transcriptional and biochemical changes occur after longer term exposure, in particular for environmentally relevant concentrations and, importantly, if these changes result in adverse health effects.


Figure 1. Venn diagrams displaying the numbers of differentially expressed transcripts (FDR<0.1 and FDR<0.05; the later presented in brackets) in each treatment group, obtained using edgeR.

Figure 2. Heatmap illustrating the expression level of all differentially-regulated transcripts in all individual samples. Individual fish are represented by the following codes: 0-a - 0-f represent the control individuals; 1.7-a, 1.7-b and 1.7-c represent individuals exposed to 1.7 ^g linuron/L; 15-a, 15-b and 15-c represent individuals exposed to 15.3 ^g linuron/L; 225-a, 225-b and 225-c represent individuals exposed to 225.9 ^g linuron/L. Data presented are log10 transformed read counts per transcript. The hierarchical clustering to generate gene and condition trees was conducted using an Euclidean distance metric using the pheatmap package in R 32

Figure 3. (A) Schematic illustration of the effects of exposure to linuron on the transcript profiles of genes encoding the cholesterol biosynthesis pathway enzymes. Each colour-coded bar represents mean transcript fold change, quantified using edgeR, for each treatment compared to the control (left to right: 1.7, 15.3, 225.9 ^g linuron/L) and asterisks indicate a significant decrease in expression (FDR <0.1). ACAT2 (acetyl-CoA acetyltransferase 2), HMGCS (3-hydroxy-3-methylglutaryl-Coenzyme A synthase), HMGCR (3-hydroxy-3-methylglutaryl-Coenzyme A reductase), MVK (mevalonate kinase), PMVK (phosphomevalonate kinase), MVD (mevalonate decarboxylase), IDI1 (isopentenyl-diphosphate delta isomerase 1), FDPS (farnesyl diphosphate synthase), GGPS1 (geranylgeranyl diphosphate synthase 1), FDFT1 (farnesyl-diphosphate farnesyltransferase 1), SQLE (squalene epoxidase), LSS (lanosterol synthase), CYP51A1 (cytochrome P450,family 51), TM7SF2 (transmembrane 7 superfamily member 2), SC4MOL

(methylsterol monooxygenase 1), NSDHL (NAD(P) dependent steroid dehydrogenase-like), HSD17B7 (hydroxysteroid (17-beta) dehydrogenase 7), EBP (sterol isomerase) , SC5D (sterol C5 desaturase), DHCR7 (7-dehydrocholesterol reductase), DHCR24 (24-dehydrocholesterol reductase).

(B) Total concentration of cholesterol in the livers of fish exposed to linuron or kept under control conditions (n=6, 4, 4 and 3 individuals for the 0, 1.7, 15.3 and 225.9 ^g linuron/L treatments, respectively). Values presented are mean ± SEM. Asterisks indicate significant differences between treatment groups (*P < 0.05;**P < 0.005; ***P < 0.001) identified using an ANOVA followed by Tukey's post hoc test.


Supporting Information Available: Supplemental experimental section, heatmap of the fold changes in transcript expression between treatments (Figure S1), multidimensional scaling plots illustrating expression profiles for all treatments (Figure S2), ERCC spike-in control analysis (Figure S3, Table S1), enriched Gene Ontology terms and Kegg pathways (Table S2), all differentially expressed transcripts (Table S3).

This material is available free of charge via the Internet at

AUTHOR INFORMATION Corresponding Authors

* Tamsyn M. Uren Webster

Biosciences, College of Life & Environmental Sciences, Geoffrey Pope Building, University of Exeter, Exeter, EX4 4QD tu202@exeter.

Phone: +44 (0)1392 724677, Fax: +44 (0)1392 263434

* Eduarda M. Santos

Biosciences, College of Life & Environmental Sciences, Geoffrey Pope Building, University

of Exeter, Exeter, EX4 4QD

Phone: +44 (0)1392 264607, Fax: +44 (0)1392 263434 ACKNOWLEDGMENT

We wish to thank Jan Shears for help with the fish husbandry, Ronny van Aerle for advice on the bioinformatics, and Audrey Farbos, Karen Moore and Konrad Paszkiewicz for facilitating the sequencing experiments. This work was funded by a Natural Environment Research Council CASE PhD studentship (Grant number NE/I528326/1) and the Salmon & Trout Association. The Exeter Sequencing facility was funded by a Wellcome Trust Institutional Strategic Support Award (WT097835MF).


1. Arnaud, L.; Taillandier, G.; Kaouadji, M.; Ravanel, P.; Tissut, M., Photosynthesis inhibition by phenylureas: A QSAR approach. Ecotoxicol. Environ. Saf. 1994, 28, (2), 121133.

2. Patterson, M., Linuron- Analysis of Risks to Endangered and Threatened Salmon and Steelhead US EPA Report 2004.

3. Woudneh, M. B.; Ou, Z.; Sekela, M.; Tuominen, T.; Gledhill, M., Pesticide multiresidues in waters of the Lower Fraser Valley, British Columbia, Canada. Part I. Surface Water. J Environ Qual. 2009, 38, 940-947.

4. Schuler, L. J.; Rand, G. M., Aquatic risk assessment of herbicides in freshwater ecosystems of south Florida. Arch. Environ. Contam. Toxicol. 2008, 54, (4), 571-583.

5. US EPA, Linuron R.E.D. Facts. Prevention, Pesticdes and Toxic Substances (7508W) 1995.

6. Health Canada, Proposed Re-evaluation Decision PRVD2012-02, Linuron. 2012.

7. Bauer, E. R.; Meyer, H. H.; Sauerwein, H.; Stahlschmidt-Allner, P., Application of an androgen receptor assay for the characterisation of the androgenic or antiandrogenic activity of various phenylurea herbicides and their derivatives. Analyst 1998, 123, (12), 2485-2487.

8. Wilson, V. S.; Cardon, M. C.; Gray, L. E.; Hartig, P. C., Competitive binding comparison of endocrine disrupting compounds to recombinant androgen receptor from fathead minnow, rainbow trout, and human. Environ. Toxicol. Chem. 2007, 26, (9), 17931802.

9. Orton, F.; Lutz, I.; Kloas, W.; Routledge, E. J., Endocrine disrupting effects of herbicides and pentachlorophenol: in vitro and in vivo evidence. Environ. Sci. Technol 2009, 43, (6), 2144-2150.

10. Freyberger, A.; Weimer, M.; Tran, H.-S.; Ahr, H.-J., Assessment of a recombinant androgen receptor binding assay: Initial steps towards validation. Reprod. Toxicol. 2010, 30, (1), 2-8.

11. Kojima, H.; Katsura, E.; Takeuchi, S.; Niiyama, K.; Kobayashi, K., Screening for estrogen and androgen receptor activities in 200 pesticides by in vitro reporter gene assays using Chinese hamster ovary cells. Environ. Health Perspect. 2004, 112, (5), 524-528.

12. Lambright, C.; Ostby, J.; Bobseine, K.; Wilson, V.; Hotchkiss, A.; Mann, P.; Gray, L., Cellular and molecular mechanisms of action of linuron: an antiandrogenic herbicide that produces reproductive malformations in male rats. Toxicol. Sci. 2000, 56, (2), 389-399.

13. Kang, I. H.; Sik Kim, H.; Shin, J.-H.; Kim, T. S.; Moon, H. J.; Kim, I. Y.; Choi, K S.; Kil, K. S.; Park, Y. I.; Dong, M. S., Comparison of anti-androgenic activity of flutamide, vinclozolin, procymidone, linuron, and DDE in rodent 10-day Hershberger assay. Toxicology 2004, 199, (2), 145-159.

14. Freyberger, A.; Schladt, L., Evaluation of the rodent Hershberger bioassay on intact juvenile male testing of coded chemicals and supplementary biochemical investigations. Toxicology 2009, 262, (2), 114-120.

15. Cook, J.; Mullin, L.; Frame, S.; Biegel, L., Investigation of a mechanism for Leydig cell tumorigenesis by linuron in rats. Toxicol. Appl. Pharmacol. 1993, 119, (2), 195-204.

16. Hotchkiss, A.; Parks-Saldutti, L.; Ostby, J.; Lambright, C.; Furr, J.; Vandenbergh, J.; Gray, L., A mixture of the "antiandrogens" linuron and butyl benzyl phthalate alters sexual differentiation of the male rat in a cumulative fashion. Biol. Reprod. 2004, 71, (6), 18521861.

17. Wilson, V. S.; Lambright, C. R.; Furr, J. R.; Howdeshell, K. L., The herbicide linuron reduces testosterone production from the fetal rat testis during both in utero and in vitro exposures. Toxicol. Lett. 2009, 186, (2), 73-77.

18. Pottinger, T.; Katsiadaki, I.; Jolly, C.; Sanders, M.; Mayer, I.; Scott, A.; Morris, S.; Kortenkamp, A.; Scholze, M., Anti-androgens act jointly in suppressing spiggin concentrations in androgen-primed female three-spined sticklebacks-prediction of combined effects by concentration addition. Aquat. Toxicol. 2013, 140, 145-156.

19. Jolly, C.; Katsiadaki, I.; Morris, S.; Le Belle, N.; Dufour, S.; Mayer, I.; Pottinger, T. G.; Scott, A. P., Detection of the anti-androgenic effect of endocrine disrupting environmental contaminants using in vivo and in vitro assays in the three-spined stickleback. Aquat. Toxicol. 2009, 92, (4), 228-239.

20. Hogan, N. S.; Gallant, M. J.; van den Heuvel, M. R., Exposure to the pesticide linuron affects androgen-dependent gene expression in the three-spined stickleback (Gasterosteus aculeatus). Environ. Toxicol. Chem. 2012, 31, (6), 1391-1395.

21. Katsiadaki, I.; Morris, S.; Squires, C.; Hurst, M. R.; James, J. D.; Scott, A. P., Use of the three-spined stickleback (Gasterosteus aculeatus) as a sensitive in vivo test for detection of environmental antiandrogens. Environ. Health Perspect. 2006, 114, (S-1), 115-121.

22. Schiller, V.; Wichmann, A.; Kriehuber, R.; Schafers, C.; Fischer, R.; Fenske, M., Transcriptome alterations in zebrafish embryos after exposure to environmental estrogens and anti-androgens can reveal endocrine disruption. Reprod. Toxicol. 2013, 42, 210-223.

Page 29 of 33

Environmental Science & Technology

635 23. Ornostay, A.; Cowie, A. M.; Hindle, M.; Baker, C. J.; Martyniuk, C. J., Classifying

636 chemical mode of action using gene networks and machine learning: A case study with the

637 herbicide linuron. Comp. Biochem. Physiol., Part D: Genomics Proteomics 2013, 8, (4), 263638 274.

639 24. Marlatt, V. L.; Lo, B. P.; Ornostay, A.; Hogan, N. S.; Kennedy, C. J.; Elphick, J. R.;

640 Martyniuk, C. J., The effects of the urea-based herbicide linuron on reproductive endpoints in

641 the fathead minnow Pimephalespromelas). Comp. Biochem. Physiol., Part C: Toxicol.

642 Pharmacol. 2012, 157, 24-32.

643 25. Martyniuk, C. J.; Alvarez, S.; Lo, B. P.; Elphick, J. R.; Marlatt, V. L., Hepatic protein

644 expression networks associated with masculinization in the female fathead minnow

645 (Pimephales promelas). J. Proteome Res. 2012, 11, (8), 4147-4161.

646 26. Martyniuk, C. J.; Denslow, N. D., Exploring androgen-regulated pathways in teleost

647 fish using transcriptomics and proteomics. Integr. Comp. Biol. 2012, 52, (5), 695-704.

648 27. Goks0yr, A.; Andersson, T.; Hansson, T.; Klungs0yr, J.; Zhang, Y.; Forlin, L.,

649 Species characteristics of the hepatic xenobiotic and steroid biotransformation systems of two

650 teleost fish, Atlantic cod (Gadus morhua) and rainbow trout (Salmo gairdneri). Toxicol. Appl.

651 Pharmacol. 1987, 89, (3), 347-360.

652 28. Grabherr, M. G.; Haas, B. J.; Yassour, M.; Levin, J. Z.; Thompson, D. A.; Amit, I.;

653 Adiconis, X.; Fan, L.; Raychowdhury, R.; Zeng, Q., Full-length transcriptome assembly from

654 RNA-Seq data without a reference genome. Nat. Biotechnol. 2011, 29, (7), 644-652.

655 29. Langmead, B.; Salzberg, S. L., Fast gapped-read alignment with Bowtie 2. Nat.

656 Methods 2012, 9, 357-359.

657 30. Li, H.; Handsaker, B.; Wysoker, A.; Fennell, T.; Ruan, J.; Homer, N.; Marth, G.;

658 Abecasis, G.; Durbin, R., The sequence alignment/map format and SAMtools. Bioinformatics

659 2009, 25, (16), 2078-2079.

660 31. Robinson, M. D.; McCarthy, D. J.; Smyth, G. K., edgeR: a Bioconductor package for

661 differential expression analysis of digital gene expression data. Bioinformatics 2010, 26, (1),

662 139-140.

663 32. Kolde, R., Pretty Heatmaps. Rpackage version 0.7.4, 2012 2012.

664 33. Huang, D. W.; Sherman, B. T.; Lempicki, R. A., Systematic and integrative analysis

665 of large gene lists using DAVID bioinformatics resources. Nat. Protocols 2008, 4, (1), 44-57.

666 34. Folch, J.; Lees, M.; Sloane-Stanley, G., A simple method for the isolation and

667 purification of total lipids from animal tissues. J. Biol. Chem. 1957, 226, (1), 497-509.

35. Allain, C. C.; Poon, L. S.; Chan, C. S.; Richmond, W.; Fu, P. C., Enzymatic determination of total serum cholesterol. Clin. Chem. 1974, 20, (4), 470-475.

36. Oulmi, Y.; Negele, R.; Braunbeck, T., Cytopathology of liver and kidney in rainbow trout Oncorhynchus mykiss after long-term exposure to sublethal concentrations of linuron. Dis. Aquat. Org. 1995, 21, 35-35.

37. Ikonen, E., Cellular cholesterol trafficking and compartmentalization. Nat. Rev. Mol. Cell Biol. 2008, 9, (2), 125-138.

38. Sharpe, L. J.; Brown, A. J., Controlling cholesterol synthesis beyond 3-hydroxy-3-methylglutaryl-CoA reductase (HMGCR). J. Biol. Chem. 2013, 288, (26), 18707-18715.

39. Bengoechea-Alonso, M. T.; Ericsson, J., SREBP in signal transduction: cholesterol metabolism and beyond. Curr. Opin. Cell Biol. 2007, 19, (2), 215-222.

40. Swinnen, J. V.; Ulrix, W.; Heyns, W.; Verhoeven, G., Coordinate regulation of lipogenic gene expression by androgens: evidence for a cascade mechanism involving sterol regulatory element binding proteins. PNAS 1997, 94, (24), 12975-12980.

41. Heemers, H.; Maes, B.; Foufelle, F.; Heyns, W.; Verhoeven, G.; Swinnen, J. V., Androgens stimulate lipogenic gene expression in prostate cancer cells by activation of the sterol regulatory element-binding protein cleavage activating protein/sterol regulatory element-binding protein pathway. Mol. Endocrinol. 2001, 15, (10), 1817-1828.

42. Heemers, H.; Vanderhoydonc, F.; Roskams, T.; Shechter, I.; Heyns, W.; Verhoeven, G.; Swinnen, J. V., Androgens stimulate coordinated lipogenic gene expression in normal target tissues in vivo. Mol. Cell. Endocrinol. 2003, 205, (1), 21-31.

43. Costigan, S. L.; Werner, J.; Ouellet, J. D.; Hill, L. G.; Law, R. D., Expression profiling and gene ontology analysis in fathead minnow (Pimephales promelas) liver following exposure to pulp and paper mill effluents. Aquat. Toxicol. 2012, 122, 44-55.

44. Johnson, K. J.; McDowell, E. N.; Viereck, M. P.; Xia, J. Q., Species-specific dibutyl phthalate fetal testis endocrine disruption correlates with inhibition of SREBP2-dependent gene expression pathways. Toxicol. Sci. 2011, 120, (2), 460-474.

45. Kelley, R. I.; Hennekam, R. C., The Smith-Lemli-Opitz syndrome. J. Med. Genet. 2000, 37, (5), 321-335.

46. Krycer, J. R.; Brown, A. J., Cross-talk between the Androgen Receptor and the Liver X Receptor- implications for cholesterol homeostasis. J. Biol. Chem. 2011, 286, (23), 2063720647.

47. Sarasquete, C.; Segner, H., Cytochrome P4501A (CYP1A) in teleostean fishes. A review of immunohistochemical studies. Sci. Total Environ. 2000, 247, (2-3), 313-332.

48. Tompkins, L. M.; Wallace, A. D., Mechanisms of cytochrome P450 induction. J.

Biochem. Mol. Toxicol. 2007, 21, (4), 176-181.

49. Bucheli, T. D.; Fent, K., Induction of cytochrome P450 as a biomarker for environmental contamination in aquatic ecosystems. Crit. Rev. Environ. Sci. Technol. 1995, 25, (3), 201-268.

50. Takeuchi, S.; Iida, M.; Yabushita, H.; Matsuda, T.; Kojima, H., In vitro screening for aryl hydrocarbon receptor agonistic activity in 200 pesticides using a highly sensitive reporter cell line, DR-EcoScreen cells, and in vivo mouse liver cytochrome P450-1A induction by propanil, diuron and linuron. Chemosphere 2008, 74, (1), 155-165.

51. Uno, T.; Kaji, S.; Goto, T.; Imaishi, H.; Nakamura, M.; Kanamaru, K.; Yamagata, H.; Kaminishi, Y.; Itakura, T., Metabolism of the herbicides chlorotoluron, diuron, linuron, simazine, and atrazine by CYP1A9 and CYP1C1 from Japanese eel (Anguilla japonica). Pestic. Biochem. Physiol. 2011, 101, (2), 93-102.

52. Cook, J. C.; Mullin, L. S.; Frame, S. R.; Biegel, L. B., Investigation of a Mechanism for Leydig Cell Tumorigenesis by Linuron in Rats. Toxicol. Appl. Pharmacol. 1993, 119, (2), 195-204.

53. Anfossi, P.; Roncada, P.; Stracciari, G.; Montana, M.; Pasqualucci, C.; Montesissa, C., Toxicokinetics and metabolism of linuron in rabbit: In vivo and in vitro studies. Xenobiotica 1993, 23, (10), 1113-1123.

54. Monostory, K.; Pascussi, J.-M.; Kabori, L.; Dvorak, Z., Hormonal regulation of CYP1A expression. DrugMetab. Rev. 2009, 41, (4), 547-572.

55. Brackley, K. I.; Grantham, J., Activities of the chaperonin containing TCP-1 (CCT): implications for cell cycle progression and cytoskeletal organisation. Cell Stress Chaperones 2009, 14, (1), 23-31.

56. Yokota, S.; Yanagi, H.; Yura, T.; Kubota, H., Upregulation of cytosolic chaperonin CCT subunits during recovery from chemical stress that causes accumulation of unfolded proteins. Eur. J. Biochem. 2000, 267, (6), 1658-1664.

57. Buckley, B. A.; Gracey, A. Y.; Somero, G. N., The cellular response to heat stress in the goby Gillichthys mirabilis: a cDNA microarray and protein-level analysis. J. Exp. Biol. 2006, 209, (14), 2660-2677.

58. Kobayashi, M.; Yamamoto, M., Molecular mechanisms activating the Nrf2-Keap1 pathway of antioxidant gene regulation. Antioxid. Redox Signaling 2005, 7, (3-4), 385-394.

59. Ihlaseh, S. M.; Bailey, K. A.; Hester, S. D.; Jones, C.; Ren, H.; Cardoso, A. P. F.; Oliveira, M. L. C.; Wolf, D. C.; de Camargo, J. L. V., Transcriptional profile of diuron-

induced toxicity on the urinary bladder of male Wistar rats to inform mode of action. Toxicol. Sci. 2011, 122, (2), 330-338.

60. Akcha, F.; Spagnol, C.; Rouxel, J., Genotoxicity of diuron and glyphosate in oyster spermatozoa and embryos. Aquat. Toxicol. 2012, 106, 104-113.

61. Matés, J.; Segura, J.; Alonso, F.; Márquez, J., Intracellular redox status and oxidative stress: implications for cell proliferation, apoptosis, and carcinogenesis. Arch Toxicol 2008, 82, (5), 273-299.

Page 33 of 33

Environmental Science & Technology