ToxSci Advance Access originally published online on August 13, 2007
Toxicological Sciences 2007 99(2):470-487; doi:10.1093/toxsci/kfm189
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
microRNAs in Adult Rodent Liver Are Refractory to Dioxin Treatment
,
* Department of Pharmacology, University of Toronto, Toronto, Canada M5S 1A8
Department of Food and Environmental Hygiene, Faculty of Veterinary Medicine, University of Helsinki, Helsinki, Finland
Finnish Food Safety Authority EVIRA, Kuopio Research Unit, Kuopio, Finland
1 To whom correspondence should be addressed at Department of Pharmacology, Room 4302 Medical Sciences Building, University of Toronto, 1 King's College Circle, Toronto, Ontario Canada M5S 1A8. Fax: (416) 978-6395. E-mail: allan.okey{at}utoronto.ca.
Received June 4, 2007; accepted July 17, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Dioxin-like chemicals are well known for their ability to upregulate expression of numerous genes via the AH receptor (AHR). However, recent transcriptomic analyses in several laboratories indicate that dioxin-like chemicals or AHR genotype itself also can downregulate levels of mRNAs encoded by numerous genes. The mechanism responsible for such downregulation is unknown. We hypothesized that microRNAs (miRNAs), which have emerged as powerful negative regulators of mRNA levels in several systems, might be responsible for mRNA downregulation in dioxin/AHR pathways. We used two miRNA array platforms as well as quantitative reverse transcriptase–polymerase chain reaction to measure miRNA levels in wild-type (WT) versus Ahr-null mice, in dioxin-sensitive Long-Evans (L-E; Turku/AB) rats versus dioxin-resistant Han/Wistar (H/W; Kuopio) rats and in rat 5L and mouse Hepa-1 hepatoma cells in culture. Treatment with 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in vivo caused few changes in miRNA levels in mouse or rat livers, and those changes that were statistically significant were of modest magnitude. Hepatoma cells in culture also exhibited few changes in miRNA levels in response to TCDD. AHR genotype had little effect on hepatic miRNA levels, either in constitutive expression or in response to TCDD—only a few miRNAs differed in expression between Ahr-null mice compared to mice with WT AHR or between L-E rats (that have WT AHR) compared to H/W rats (whose AHR has a large deletion in the transactivation domain). It is unlikely that mRNA downregulation by dioxins is mediated by miRNAs, nor are miRNAs likely to play a significant role in dioxin toxicity in adult rodent liver.
Key Words: aryl hydrocarbon receptor; dioxin; 2,3,7,8-tetrachlorodibenzo-p-dioxin; microRNA; microarray.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a potent environmental toxicant whose effects include hepatotoxicity, teratogenicity, cancer, severe anorexia-like wasting, and death (Bunger et al., 2003
; Nebert et al., 2000; Okey et al., 2005; Peters et al., 1999; Pohjanvirta and Tuomisto, 1994). Dioxin toxicity is mediated by the AH receptor (AHR) (reviewed in Okey, 2007) and appears to require altered transcription of target genes (Bunger et al., 2003). Dioxins exert powerful effects on expression of many genes, but the identities of the specific genes whose dysregulation by dioxins leads to major forms of dioxin toxicity remain unknown (Okey et al., 2005).
The vast majority of studies on dioxins in relation to gene expression have focused on AHR-mediated enzyme induction and upregulation of gene expression (Hankinson, 2005; Okey et al., 2005). However, large-scale transcriptomic studies reveal that levels of the mRNA products of many genes are decreased by TCDD rather than increased (Boverhof et al., 2006; Fletcher et al., 2005; Ovando et al., 2006; Tijet et al., 2006). Our previous gene array studies of dioxin-altered hepatic mRNA expression in two dioxin-resistant rodent models revealed numerous transcripts whose levels are suppressed. Ahr-null mice, which lack the AHR, are highly resistant to TCDD toxicity (Mimura et al., 1997; Peters et al., 1999). By contrasting transcript profiles from Ahr-null mice with those from mice that have wild-type (WT) AHR, we identified 218 mRNA ProbeSets whose levels were suppressed by presence of the AHR per se and 118 ProbeSets whose levels were suppressed by TCDD acting via an AHR-dependent mechanism (Tijet et al., 2006). The second model involves Han/Wistar (H/W) (Kuopio) rats which lack a portion of the AHR transactivation domain and consequently are exceptionally resistant to lethal effects of TCDD when compared with rats that express WT AHR (reviewed in Okey et al. [2005]; Pohjanvirta and Tuomisto [1994]). We identified 97 transcripts whose levels are suppressed by TCDD in dioxin-sensitive rats but not in dioxin-resistant rats (Moffat et al., in preparation).
The AHR-mediated mechanism of transcriptional upregulation is well understood; for example, direct binding of the AHR to AH-responsive elements (AHREs) within regulatory regions of the CYP1A1 gene upregulates its expression (Hankinson, 2005). However, very little is known about mechanisms by which dioxin-like chemicals suppress mRNA levels (Riddick et al., 2004). Altered transcription rates alone may not explain all the alterations in transcript levels. One plausible new mechanism by which mRNA levels might be reduced is by the action of microRNAs (miRNAs). miRNAs are evolutionarily conserved, small (
22 nucleotide [nt]), non-coding transcripts that suppress levels or activity of target mRNAs by multiple mechanisms including triggering mRNA degradation, blocking translation, or modifying chromatin structure to silence transcription. miRNAs regulate diverse biological processes ranging from embryonic development to fat storage, insulin secretion, drug metabolism, apoptosis, cell growth, tumorigenesis, and death (Gaur et al., 2007; Grimm et al., 2006; Tsuchiya et al., 2006). These same processes also are altered by dioxin exposure. Hypothetically, dioxins could upregulate or downregulate specific miRNAs which, in turn, could alter degradation, efficiency of translation, or transcription of target mRNAs.
Therefore, we conducted a thorough investigation to address the following questions: (1) Does AHR genotype itself affect constitutive expression of miRNAs? (2) Does TCDD affect miRNA levels and, if so, is this response dependent on the AHR? and (3) Does TCDD affect miRNA levels differently in animals that are sensitive to dioxin toxicity versus those that are dioxin resistant? We assessed the in vivo effect of TCDD on miRNA levels in liver at multiple time points after TCDD treatment using two different miRNA array platforms along with quantitative reverse transcriptase (qRT)–PCR. In addition, we used qRT–PCR to test the effect of TCDD on miRNA levels in thymus, lung, and kidney in vivo as well as the effect in hepatoma cells from mouse and rat in culture. We focused on hepatic miRNAs because liver displays a broad spectrum of mRNAs that are downregulated by dioxins or by AHR genotype (Tijet et al., 2006) and because liver is a prime site of dioxin toxicity (Niittynen et al., 2007). As an adjunct to laboratory measurements, we used bioinformatic techniques to predict which mRNAs are likely to be targets for specific miRNAs, to determine if specific miRNA sequences contain AHR-binding sites, and to assess whether mRNA for the AHR itself might be a target for any miRNA.
The cumulative results of our experiments indicate that downregulation of mRNA levels by dioxins in adult rodent livers is very unlikely to involve miRNAs.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Models In Vivo and In Vitro
Dioxin-Resistant Ahr-Null Mouse Model
Liver tissues were from mice in which we previously mapped AHR-dependent and dioxin-dependent gene batteries by transcriptomic analysis (Tijet et al., 2006
). Briefly, male Ahr-null (Ahr–/–) mice in a C57BL/6J background (10 weeks old) and WT (Ahr+/+) C57BL/6J mice (15 weeks old) were given a single dose of 1000 µg/kg TCDD or corn oil vehicle by gavage. Liver was harvested 19 h after treatment. We tested three TCDD-treated and three control mice in the Ahr–/– groups and four TCDD-treated and four control mice in the Ahr+/+ groups.
Dioxin-Resistant H/W Rat Model
Tissues were from Long-Evans (L-E) and H/W rats in which we have determined mRNA levels in response to TCDD on Affymetrix GeneChips (Moffat et al., in preparation). Briefly, TCDD-sensitive L-E rats (which express WT AHR) and TCDD-resistant H/W rats (with a deletion in the AHR transactivation domain [Pohjanvirta et al., 1998]) were from breeding colonies of the National Public Health Institute, Division of Environmental Health, Kuopio, Finland. All animals were males 10–12 weeks old. Liver was harvested from rats treated by gavage with a single dose of 100 µg/kg TCDD or corn oil vehicle for 3, 19, or 96 h. Kidney, lung, and thymus were harvested from rats treated with a single dose of 50 µg/kg TCDD or corn oil vehicle by gavage for 19 h. Each group contained four animals.
Rat and Mouse Cell Models In Vitro
Rat hepatoma cells (5L) and mouse hepatoma cells (Hepa-1c1c7), obtained from the American Type Culture Collection (Manassas, VA), were cultured in 
-minimal essential medium supplemented with 10% heat-inactivated fetal bovine serum (HyClone, Logan, UT) and penicillin-streptomycin (100 units/ml and 100 µg/ml, respectively). Cells were seeded in 6-well plates (2.0 x 105 cells/well) and cultured at 37°C in a 5% CO2/air incubator with 90% humidity for 24 h prior to treatment. Cells were treated with the dimethyl sulfoxide vehicle (control cells; n = 6), 10nM TCDD (n = 6), or 100nM TCDD (n = 2) for 24 h.
RNA Isolation
Total RNA was isolated from rat livers using the mirVana miRNA Isolation Kit (Ambion, Austin, TX) according to the manufacturer's instructions except flashPAGE fractionation (Ambion) was not performed due to high technical variability. For the Ambion bioarray experiments, small population RNA (<200 nt) was enriched from mirVana-isolated total RNA according to manufacturer's instructions. Total RNA was isolated from all other rat tissues, mouse liver, and cell cultures using TRIzol reagent (Invitrogen, La Jolla, CA) according to the manufacturer's instructions. Total RNA yield was quantified by UV spectrophotometry, and RNA integrity was verified using an Agilent 2100 Bioanalyser (Agilent Technologies, Santa Clara, CA).
Ambion mirVana miRNA Bioarrays
Our initial miRNA array studies were conducted with Ambion mirVana Bioarrays. See Supplementary Table S1 for methods and results.
Exiqon MiRCury LNA miRNA Array
Labeling and hybridization of 2.5 µg of total RNA from rat and mouse models were performed using Exiqon's MiRCury LNA array labeling kit (version 1.2) and hybridization kit (version 1.2) (Exiqon, Vedbaek, Denmark) according to the manufacturer's instructions. Each total RNA sample was separately labeled either with a Hy3 or a Hy5 fluorophore (Exiqon). Hy3- and Hy5-labeled samples were cohybridized to an array as shown in Figure 1. Dye reversal was performed to eliminate dye bias. The array slides contain capture probes complementary to mature miRNAs registered in the miRBase release 8.0, May 2006 (Griffiths-Jones et al., 2006). The capture probes are melting temperature–normalized locked nucleic acid (LNA)–enhanced oligonucleotides (Vester and Wengel, 2004) that all hybridize optimally under a common set of conditions. Exiqon MiRCury LNA miRNA arrays contain 540 sequences representing 454 known miRNAs (human, mouse, or rat), 10 unknown miRNAs, and 76 control miRNAs and probes, all of which are spotted in quadruplicate. Slides were hybridized at 60°C for 17 h.
|
Arrays were scanned at 10-µm resolution on a GenePix 4000B Array Scanner (Axon Instruments, Inc., Union City, CA) using auto-photomultiplier tube calculation settings. Each element was located and analyzed with the GenePix Pro 6.0.1.26 software package. Using LimmaGUI (version 1.8.1; limma, version 2.9.1 [Wettenhall and Smyth, 2004
]), the array data were loaded into the R statistical package (version 2.4.0), investigated for spatial and distributional homogeneity, background subtracted (minimum method), normalized within arrays (robust spline), normalized between arrays (scale), and fit to a general linear model with empirical Bayes moderation of the standard error (Smyth, 2004). To adjust for multiple testing, the Benjamini and Hochberg false discovery rate correction was applied (Reiner et al., 2003). For each effect, we extracted estimates of differential expression (log2), expression level (log2), and significance (adjusted p < 0.05, B-statistic > 0). The B-statistic is the log odds that the miRNA is differentially expressed; a B-statistic of zero corresponds to a 50% chance that the gene is differentially expressed.
Linear Modeling of Exiqon Array Data
Mouse Model
Expression = AHR + TCDD + AHR:TCDD + batch.
Terms are as defined previously (Tijet et al., 2006). Briefly, the AHR effect represents TCDD-independent effects of the AHR. These effects are estimated both from the difference between wild-type control (WTC) mice and Ahr-null control (KOC) mice and from the difference between wild-type TCDD-treated (WTT) and Ahr-null TCDD-treated (KOT) mice. The TCDD effect represents AHR-independent effects of TCDD, i.e., differences between KOT and KOC that also are found between WTT and WTC mice. The AHR-TCDD interaction represents AHR-dependent effects of TCDD. They are differences between WTT and WTC that are not found between KOT and KOC animals. "Batch" represents the expression difference between batches of arrays.
Rat Models
Expression = 3-h TCDD + 19-h TCDD + Q (H/W only).
The array data from each rat strain were separately fit to the above linear model. This allowed us to estimate the effect of 3-h exposure to TCDD and the effect of 19-h exposure to TCDD on miRNA levels for each rat strain. For the H/W strain of rat only, the Q-term represents changes in miRNA expression at 96 h after TCDD administration. The Q term is an estimate of levels since miRNA levels 96-h post-TCDD treatment were compared to the H/W 19-h control levels instead of the H/W 96-h control levels.
Expression (L-E) = 96-h TCDD + 96-h feed restricted + batch.
Array data from L-E rats exposed to a single dose of TCDD, corn oil, or whose feed was restricted for 96 h were fit to the above model. Feed-restricted control L-E rats were included to ensure that changes in miRNA levels were due to TCDD treatment per se and not the result of the decreased feed intake which occurs in dioxin-sensitive strains within 96 h after TCDD exposure. The 96-h TCDD effect captures effects resulting from 96-h TCDD treatment only while the 96-h feed-restricted effect identifies those changes due to feed restriction in combination with TCDD treatment. This model does not identify those changes in expression that result from feed restriction alone and not found in TCDD-treated animals. Batch is as defined above.
Expression = strain + T3 + T19 + strain:T3 + strain:T19 + batch.
All array information from both rat strains at 3 h and 19 h after treatment were fit to the above linear model. This model allowed us to use all array information to estimate constitutive differences in miRNA levels between L-E versus H/W rats in addition to expression changes caused by the administration of TCDD (3 or 19 h) that are dependent on the strain of rat. Specifically, the strain effect represents differences in miRNA levels between strains (L-E vs. H/W) independent of TCDD. The T3 effect represents expression changes 3 h after TCDD administration alone, independent of rat strain. The T19 effect represents expression changes 19 h after TCDD administration alone, independent of rat strain. The strain:T3 and strain:T19 interaction terms represent expression changes caused by administration of TCDD (3 or 19 h) that are different between L-E versus H/W rat strains.
Real-Time Quantitative PCR
Reverse transcription and PCR reactions used looped primers (Chen et al., 2005) and a probe that was specific for mature miRNA (TaqMan miRNA assays, Applied Biosystems, Foster City, CA). See Table 1 for primer and probe details. Ten nanograms of total RNA were reverse transcribed in a 7.5-µl reaction using MuLV reverse transcriptase and a TaqMan miRNA RT primer (Applied Biosystems). The reaction mixture was incubated at 16°C for 30 min, 42°C for 30 min, then 85°C for 5 min. Five microlitres of the reverse transcribed product was assayed using TaqMan Universal PCR Master Mix, no AmpErase, and 1 µl of TaqMan miRNA PCR primers/probe mix in a 15-µl reaction (Applied Biosystems). Real-time PCR was performed on a Stratagene MX4000 real-time PCR system using the following conditions: after 10 min at 95°C, 40 cycles were performed at 95°C for 15 s, and 60°C for 1 min. Normalized expression (NE) was calculated using NE = 2–
Ct, where Ct is the threshold cycle to detect fluorescence. The data were normalized to miRNAs: miR-23b and/or let-7b. To define significant differences in miRNA levels in the mouse in vivo model and the cell culture models, ANOVA followed by Bonferroni post hoc tests were performed using GraphPad version 4.0 (GraphPad Software, Inc., San Diego, CA). In the rat model, significant differences in miRNA levels were determined using a t-test (two-sided, equal variance as supported by F-test). Differences between treatment groups were considered significant when: *p < 0.05, **p < 0.01, ***p < 0.001.
|
Potential miRNAs for RT–PCR Normalization
To identify potential miRNAs to use as a normalization standard for RT–PCR studies, we examined all miRNA array profiles. miRNAs whose expression was not significantly altered by AHR genotype or by TCDD treatment in both rat and mouse models (p > 0.05, B-statistic < 0) and whose average log2 expression levels were greater than 7.6 were selected for further evaluation by RT–PCR (Supplementary Table S2). RT–PCR analysis confirmed that levels of let-7b and miR-23b were unaltered by AHR genotype or by TCDD treatment in either rat or mouse in vivo models; however, miR-23b levels in the in vivo rat model were variable and, consequently, were not suitable as a normalization standard. miR-181b was not detected in rat liver by RT–PCR. Therefore, miRNA levels were normalized to let-7b and miR-23b miRNAs in the mouse model and to let-7b only in the rat model.
Prediction of miRNA-mRNA Target Interactions
miRNA targets can be difficult to identify due to lack of strict base pairing between miRNA and mRNA target sequences. Computational algorithms aid this task by examining base-pairing rules between miRNA and mRNA target sites, location of binding sites within the target's 3'-UTR, and conservation of target binding sequences within related genomes. We searched for rat and mouse mRNAs that are predicted to be targeted by candidate miRNAs within the miRBase database (version 9.1, MIRANDA predicted targets [Griffiths-Jones et al., 2006]) using miRGEN (version 3; Megraw et al., 2007). In addition, we searched miRBase for miRNAs that are predicted to target the 3'-UTR of the rat AHR (ENSRNOT00000006618, miRBase Targets version 4; Griffiths-Jones et al., 2006).
Search for AHREs Upstream of miRNA Coding Regions
We searched for AHREs that potentially regulate miRNA levels by first downloading transcription start site (TSS) annotation for all mRNAs and miRNAs from both mouse and rat genomes (mRNA: University of California Santa Cruz genome builds mm8 and rn4, 2007-04-07 [Karolchik et al., 2003]; miRNA: 2007-05-13). Second, we extracted regions of three different lengths surrounding the TSS of each gene: – 10,000 to + 5000 bp; – 4000 to + 1000 bp; and – 1000 to + 250 bp. Third, we searched for AHRE-I (core, extended, and full sequences) and for AHRE-II motifs within each of the three regions extracted; the specific sequences and search method are as described previously (Boutros et al., 2004). Subsequently, a PhyloHMM conservation score was associated with each nucleotide base within each identified AHRE motif and then the average PhyloHMM score for each motif was calculated. PhyloHMM scores provide a measure of conservation that accounts for phylogenetic relationships across the different species considered and range between 0 (minimal conservation) and 1 (strong conservation) (Siepel and Haussler, 2004). Both the number of motifs and the maximal PhyloHMM score were extracted for each AHRE motif. Last, Gaussian kernel densities were fit separately to the miRNA and mRNA PhyloHMM scores and number of AHRE motifs for each rodent species, region of sequence, and AHRE motif.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
As detailed in the following sections, we detected some statistically significant changes in miRNA levels related to AHR genotype or to TCDD treatment. However, in virtually all instances the magnitude of change was small. As a point of reference, the magnitude of changes in miRNA levels during development or during tumorigenesis can be as much as a 1000-fold (Gaur et al., 2007
). The effect of TCDD or of AHR genotype on miRNA levels has never been examined before; therefore, no known miRNA was available that could be used as a positive control for AHR-dependent dioxin effects in our experiments. Nevertheless, we previously found that classic AHR-regulated mRNAs in exactly the same tissue samples used for these miRNA studies responded normally to TCDD and AHR status, thereby confirming the validity of the overall in vivo treatment, sample preparation, and responsiveness to TCDD (Moffat et al., in preparation; Tijet et al., 2006). Furthermore, we measured the effects of TCDD and AHR genotype on miRNA levels using two substantially different array platforms supplemented with real-time RT–PCR. Two specific miRNAs, miR-122a and let-7b, are most universally known to be constitutively expressed at high levels in rodent livers (Babak et al., 2004; Lagos-Quintana et al., 2002); both miRNAs produced a strong signal on both array platforms, validating the ability of these arrays to detect miRNAs in our liver samples. All raw array data have been deposited in the GEO database (accession number GSE8474) available at http://www.ncbi.nlm.nih.gov/projects/geo/query/acc.cgi?acc=GSE8474. The results of the overall study are summarized in Supplementary Table S3.
Ahr-Null Mouse Model
Presence or absence of the AHR has modest effects on constitutive expression of adult hepatic miRNA.
Our previous array studies of mRNA expression revealed that the AHR affects constitutive expression of a large number of genes independently of exogenous ligands (Tijet et al., 2006). The AHR's effect on transcriptional regulation of miRNAs has not previously been studied. Our analyses using Exiqon arrays revealed that 28 of the 464 miRNA sequences on the arrays differ significantly in hepatic levels between Ahr+/+ versus Ahr–/– mice, independent of TCDD treatment ("AHR" column, Table 2; Fig. 2). Of the 28 miRNAs, nine were affected only by the AHR genotype alone, while 19 miRNAs were also affected by TCDD alone or by the AHR-TCDD interaction (Fig. 2). In fact, the AHR genotype influences hepatic levels of nearly as many miRNAs in the absence of exogenous ligands, 28 miRNAs, as it does following TCDD administration, 30 miRNAs (Fig. 2). Constitutive expression of 24 of the 28 miRNAs was lower in Ahr+/+ mice than in Ahr–/– mice; however, the magnitude of the AHR effect was very small. The maximum difference observed between Ahr+/+ versus Ahr–/– mice was a 1.5-fold suppression of miR-151 by the presence of the AHR (Table 2).
|
|
As an independent method to evaluate miRNA levels, we measured three candidate miRNAs by RT–PCR (Table 2; Fig. 3). RT–PCR did not detect any significant constitutive differences in miR-122a levels between Ahr–/– and WT Ahr+/+ mice (Fig. 3), although a slight statistically significant difference had been indicated by the array data (Table 2). In agreement with the array data, constitutive expression levels of miR-138 were lower in Ahr+/+ mice than in Ahr–/– mice as measured by RT–PCR analysis (Fig. 3). RT–PCR confirmed that AHR genotype alone alters miR-203 expression; however, miR-203 levels were higher in Ahr+/+ mice than in Ahr–/– mice (Fig. 3), rather than lower as suggested by the array. Similarly, RT–PCR analysis revealed that constitutive expression levels of miR-542-5p were lower in Ahr+/+ mice than in Ahr–/–, while the arrays indicated a TCDD effect but not an effect of AHR genotype alone.
|
TCDD has little effect on miRNA levels in the adult mouse model.
Nearly all effects of TCDD on mRNA expression require the AHR (Tijet et al., 2006
). We anticipated that TCDD effects on miRNA expression also would require the AHR. However, analyses on Exiqon arrays revealed that similar numbers of miRNAs were altered by TCDD exposure in an AHR-independent manner, 29 miRNAs, as are altered in an AHR-dependent manner, 30 miRNAs (Table 2; Fig. 2). Specifically, in response to TCDD, levels of 25 of the 30 miRNAs were increased in Ahr+/+ versus Ahr–/– mice ("AHR:TCDD" column, Table 2). These AHR-TCDD interactions included four miRNAs that were affected by TCDD via the AHR and 19 miRNAs that were affected constitutively by the AHR alone and also were responsive to TCDD via the AHR. Our previous examination of mRNA levels in this mouse model indicated that AHR effects and AHR:TCDD effects usually are in accord (Tijet et al., 2006), but we found no such agreement of effects on miRNA levels. Instead, we found that the AHR effect alone led to constitutive suppression of miRNA levels, whereas the AHR-TCDD interaction enhanced levels for 17 of the 19 miRNAs. The magnitude of effect was small; for instance, the level of miR-373* was suppressed by TCDD to the greatest extent (2.4-fold) in an AHR-dependent fashion. Of note, two separate probes on the array that each represent miR-20a both displayed an AHR:TCDD response (upregulation) and a TCDD response (downregulation); in addition, six members of the mir-17 family all displayed an AHR:TCDD response (upregulation). Six miRNA candidates, whose expression by array analyses appeared to be altered by TCDD via an AHR-dependent mechanism, were measured by RT–PCR ("AHR:TCDD" column, Table 2; Fig. 3). RT–PCR analysis failed to confirm AHR-dependent alteration of miR-122a, miR-138, miR-203, miR-373*, miR-498, or miR-542-5p levels following TCDD exposure.
AHR-independent responses to TCDD in the adult mouse model.
Exiqon array analysis revealed that levels of 29 miRNAs were altered by TCDD exposure via an AHR-independent mechanism ("TCDD" column, Table 2; Fig. 2). Fifteen of the 29 miRNAs showed statistically significant responses to TCDD (independent of the AHR); for example, the levels of two members of the mir-101 family—miR-101a and miR-101b—were suppressed by TCDD approximately twofold.
Six miRNA candidates, whose expression by array analyses appeared to be altered by TCDD independently of the AHR, were examined by RT–PCR ("TCDD" column, Table 2; Fig. 3). RT–PCR analysis confirmed array results for two miRNAs only—lower levels of miR-101a and miR-203 in TCDD-treated mice than in control mice independent of AHR status (Fig. 3).
In addition, two candidates from array analyses in the rat model—miR-361 and miR-510 (see below)—were examined in the mouse model by RT–PCR. In agreement with array analysis in the mouse model, RT–PCR did not detect a significant effect of AHR genotype and/or TCDD on miR-361 or miR-510 levels (Fig. 3). Of note, miR373*, whose levels were predicted by array analysis to be affected by AHR:TCDD and TCDD alone, was not detectable by RT–PCR in mouse liver.
In summary, RT–PCR confirmed that AHR genotype alone alters constitutive levels of at least three miRNAs and that TCDD alters levels of at least two miRNAs by an AHR-independent mechanism, but RT–PCR failed to confirm miRNA regulation by TCDD via an AHR-dependent mechanism in the mouse model.
Dioxin-Resistant Rat Model
Very few miRNAs differ in constitutive expression between dioxin-sensitive L-E rats (AHRWT/WT) versus dioxin-resistant H/W rats (AHRHW/HW).
In dioxin-resistant H/W rats (which bear an AHR with a deletion in the transactivation domain), constitutive expression levels of some mRNAs are selectively altered while levels of other mRNAs remain similar to that in rats with the WT AHR (Boutros et al., in preparation). We sought to determine if such strain-dependent differences also exist for miRNA regulation. Exiqon array analysis identified five miRNAs whose constitutive levels differ between H/W rats (AHRHW/HW) and L-E rats (AHRWT/WT) (Tables 3 and 4, "Strain" column in Table 4). Two of the five miRNAs—miR-373* and miR-498—were assayed by RT–PCR (Fig. 4), but miR-373* was not detectable in rat liver (data not shown). Exiqon array data indicated higher constitutive levels of miR-498 in H/W rats than in L-E rats (Table 4), and this was confirmed by RT–PCR analysis in a group of control animals at 96 h (Fig. 4). In 19-h control groups, miR-498 levels were higher in H/W than in L-E, but the difference was not statistically significant (Fig. 4). Also, we used RT–PCR to measure, in the rat model, levels of miR-138, whose constitutive levels differed between Ahr-null and WT mice (Fig. 3). The constitutive levels of miR-138 did not differ between H/W and L-E rats in 19-h control groups but were lower in H/W than in L-E 96-h control groups (Fig. 4). RT–PCR also detected significantly higher expression levels of miR-122a in H/W than L-E 19-h control rats (Fig. 4). Last, RT–PCR confirmed the lack of constitutive differences between L-E and H/W rats for miR-101a, miR-203, miR-361, miR-510, and miR-542-5p levels (Fig. 4), in accord with analyses by array (Table 4).
|
|
|
Since no information for miR-373* is available for rodent tissues, we also examined miR-373* levels in nonhepatic tissues. Similar to liver, we found that miR-373* was not detectable in lung (data not shown) but was expressed in thymus and kidney (Fig. 5); however, there was no significant difference between L-E versus H/W in constitutive levels in these nonhepatic tissues (Fig. 5).
|
In response to TCDD, the expression levels of very few hepatic miRNAs are altered and mainly in dioxin-sensitive L-E rats (AHRWT/WT) rather than in dioxin-resistant H/W rats (AHRHW/HW).
At 19 h post-TCDD, no miRNA levels were altered in either L-E or H/W rats as measured on Ambion miRNA arrays or on Exiqon miRNA arrays (Tables 3 and 4). The Exiqon array has broader coverage of rodent genomes.
Ambion and Exiqon arrays contain 322 miRNAs that are common to both arrays. However, Exiqon arrays contain 142 miRNA sequences that are not represented on Ambion arrays, while only 55 miRNAs are unique to the Ambion platform. Further, the Exiqon array platform uses LNA technology (Vester and Wengel, 2004) which claims to discriminate between miRNA family members with as little as 1 nt difference; hence, we conducted all remaining experiments using the Exiqon platform.
Subsequent to the initial examination of miRNA responses at 19 h, we examined miRNA expression profiles 3 h after TCDD treatment to focus on primary alterations in miRNA expression. After 3 h of exposure to TCDD, the levels of only nine miRNAs were affected by TCDD and only in dioxin-sensitive L-E rats (Table 4). Of these nine miRNAs, seven were suppressed by TCDD—including two sequences representing miR-361.
Next, to discover miRNAs whose expression levels might accompany the onset of toxicity, we examined miRNA expression profiles 96 h after TCDD treatment. After 96 h of exposure to TCDD, L-E rats display overt signs of dioxin toxicity (Pohjanvirta and Tuomisto, 1994). The arrays identified six miRNAs from L-E rats and three miRNAs from H/W rats (each different from those in L-E rats) whose levels differed from those in control animals 96 h after TCDD exposure. Levels of mir-373* only were altered 3 and 96 h after TCDD treatment. The maximum magnitude of response was observed in H/W rats 96 h post-TCDD where the levels of miR-485-3p were increased 2.5-fold compared to control levels. miR-485-3p is not likely to play a key role in an early AHR-mediated mechanism of dioxin toxicity since its response to TCDD was independent of the AHR in mouse and occurred only at a late time point in the dioxin-resistant H/W rat. To ensure that changes in miRNA levels were due to TCDD treatment rather than to the decreased feed intake, which occurs in dioxin-sensitive strains within 96 h after TCDD exposure, we included feed-restricted control L-E rats in our analysis at this 96-h exposure time. The arrays detected four miRNAs whose levels were altered as the consequence of feed restriction per se ("Feed effect" column, Table 4).
To evaluate the TCDD effect on miRNA levels in the rat model, nine miRNA candidates, whose levels by array analysis appeared to be altered by TCDD treatment in the mouse or rat model, were selected for independent quantitation by RT–PCR (Fig. 4). Levels of only one miRNA candidate—miR-122a—significantly differed between TCDD-treated (3 or 19 h) and control rats, in either H/W or L-E rats (Fig. 4). In H/W rats only, miR-122a exhibited increased levels at 19 h after TCDD treatment. In addition, levels of miR-122a were significantly higher in H/W rats than in L-E rats at 3 and 19 h post-TCDD (Fig. 4). Similarly, levels of miR-542-5p were significantly higher in H/W rats than in L-E rats 3 h post-TCDD, while levels of miR-138 were significantly lower in H/W than in L-E rats 19 h post-TCDD (Fig. 4).
After 96 h of exposure to TCDD, RT–PCR confirmed the array results which showed suppression of miR-203 levels in L-E rats only. RT–PCR also detected a small increase in the level of miR-498 at 96 h post-TCDD; again this was only significant in L-E rat liver. Since RT–PCR did not detect any significant differences in levels of the eight miRNAs between feed-restricted control versus corn oil control L-E rats (Fig. 4), the changes in miRNA levels due to 96 h TCDD treatment are not likely to be the result of decreased feed intake.
Since miR-122a is the only confirmed candidate for involvement in dioxin toxicities (see "Discussion" section) and since miR-373* appeared to respond at 3, 19, and 96 h after TCDD exposure by array (but could not be detected by RT–PCR in rat liver), we characterized their responses to TCDD in nonhepatic tissues. By RT–PCR, we found that levels of miR-122a were increased at 19 h post-TCDD in L-E rat thymus but rose only to a level similar to that which was constitutively expressed in the H/W rat (Fig. 5). In discord with liver, miR-122a levels in the H/W thymus were not altered by 19 h exposure to TCDD. Further, TCDD did not alter levels of miR-122a in kidney from either the L-E strain or the H/W strain of rats. RT–PCR detected increased miR-373* levels 19 h after TCDD treatment in thymus and kidney of L-E rats only (Fig. 5), but mir-122a and mir-373* were not detectable in lung tissue.
Cell Culture Models
The Hepa-1 cell model is widely used for studies of AHR-regulated gene expression (Dere et al., 2006). To determine if miRNAs are responsive to TCDD in vitro, we used RT–PCR to measure selected miRNA levels constitutively and after 24 h exposure to 10 or 100nM TCDD in both the rat 5L and the mouse Hepa-1 hepatoma cell lines. We found that each candidate miRNA was expressed constitutively both in the rat and in the mouse cell lines except miR-542-5p which was not detected in the rat 5L cell line (Fig. 6). With the exception of miR-122a and miR-203, candidate miRNAs were unresponsive to TCDD in either cell line (Fig. 6). TCDD treatment significantly increased miR-122a levels in the Hepa-1 mouse cells only, whereas TCDD increased miR-203 levels in the rat 5L cells only (Fig. 6).
|
Predicted mRNA Targets for Candidate miRNA Regulation
Incomplete knowledge of which genes and mRNAs are targets for a given miRNA hampers understanding miRNA function. We therefore analyzed the predicted miRNA target genes in order to gain a glimpse into the potential functions of miRNAs that were affected by TCDD or by AHR genotype in our experiments. For each candidate miRNA identified by array analysis, there hypothetically can be tens, hundreds, or even thousands of predicted mRNA targets. To focus on targets relevant to AHR-mediated dioxin toxicities, we report only those mRNA targets of candidate miRNAs that overlap with mRNAs previously shown to be altered by AHR genotype and/or TCDD treatment in the mouse model (Table 2, last column; Tijet et al., 2006
) or the rat model (Table 4, last column; Moffat et al., in preparation). See Supplementary Tables S4a (mouse) and S4b (rat) for a complete list of all predicted targets. We identified 287 mouse mRNAs, whose expression previously was reported to be affected by AHR genotype and/or TCDD, which contained binding sites for the candidate miRNAs—including, classic dioxin-responsive genes CYP1A1 and CYP1A2. Many of these mRNAs are predicted to be targeted by multiple miRNAs—for example, CYP27A1 has four potential miRNA binding sites. Only 30 dioxin-responsive rat mRNAs contain binding sites for candidate rat miRNAs. The predicted target mRNAs are those whose transcript levels potentially can be altered by the candidate miRNAs and do not include mRNAs whose translation may be blocked by the miRNA.
miRNAs Predicted to Affect the AHR
miRNAs could regulate the AHR directly, especially, since regulatory proteins such as transcription factors are preferentially regulated by miRNA-dependent regulation (Gaidatzis et al., 2007). Our search of miRBase (Griffiths-Jones et al., 2006) identified 15 potential miRNA target binding sites within the rat AHR mRNA sequence (Supplementary Table S5). However, at present, our experiments do not indicate that the AHR is, in fact, significantly affected by the predicted miRNAs in adult rodent livers.
AHREs Upstream of miRNA Coding Regions
The AHR regulates conventional gene transcription into mRNAs by binding to specific nucleotide sequences that constitute enhancer elements in the 5'-flanking sequence of the gene—for example, AHRE-I in the CYP1A1 gene (Tijet et al., 2006) or AHRE-II in the rat CYP1A2 gene (Boutros et al., 2004). Little is known about what regulates expression of miRNAs. Our microarray studies showed that miRNAs in adult rodent livers are largely refractory to dioxin treatment, whereas mRNAs are profoundly affected. Because many AHR-responsive genes harbor AHREs in their regulatory regions, we sought to determine if genes that encode miRNAs and genes that encode mRNAs differ in their promoter architecture. We searched the 5'-flanking sequences upstream of coding regions for each miRNA represented on the arrays to determine if AHREs are available to potentially regulate miRNA responses to TCDD or to the AHR genotype. The core AHRE-I sequence (GCGTG) is present within the region –4000 to +1000 bp from the predicted TSS for virtually all miRNAs on the arrays (Supplementary Tables S6 and S7). This pentanucleotide sequence occurs at random once every 256 bases throughout the genome (Tijet et al., 2006). Therefore, we computed a PhyloHMM score for each AHRE motif, a procedure which uses phylogenetic conservation of the motif across multiple species to infer which motifs are most likely to constitute functional regulatory sites. The PhyloHMM scores (Supplementary Tables S6 and S7) indicate that the 5'-flanking sequences for several miRNAs contain conserved full or extended AHRE-I sequences. When we plotted densities of AHRE-I and AHRE-II motifs in the 5'-flanking sequences upstream of genomic regions that encode miRNAs and densities of these same motifs in flanking sequences of regions that encode mRNAs, there were no substantive differences between mRNAs and miRNAs (Supplementary Figures 1–8). AHRE-I and AHRE-II sequences are neither overrepresented nor underrepresented in potential regulatory regions of miRNA genes. Thus, it is unlikely that the hepatic nonresponsiveness of miRNAs to TCDD or to AHR status is due to a general deficiency of AHREs in miRNA genes.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Dioxin toxicity often is assumed to be the result of upregulation or overexpression of particular genes. However, it is plausible that toxic effects of dioxin-like chemicals are due to downregulation of specific mRNAs and the proteins that they encode. The possibility that miRNAs might modulate mRNA levels and subsequent toxicity of xenobiotic chemicals has been virtually unexplored. Only one miRNA related to AHR-regulated genes has been reported; breast cancers exhibit low levels of miR-27b, leading to increased CYP1B1 levels, thereby fostering conversion of estrogens into genotoxic metabolites (Tsuchiya et al., 2006
).
Constitutive miRNA Expression Is Affected Only Modestly by AHR-Genotype in Adult Rodent Liver
Even in the absence of exogenous ligands, the AHR plays a significant role in physiological signaling, growth, and development (Bunger et al., 2003; Fernandez-Salguero et al., 1995; Lahvis et al., 2000; Walisser et al., 2004). Our array analyses detected 24 miRNAs whose levels are lower in mice that express a functional AHR than in mice whose AHR is knocked out. For example, levels of miR-138 are significantly lower in Ahr+/+ mice than in Ahr–/–, suggesting that under constitutive physiological conditions in adult mouse liver the AHR plays a small role in suppressing levels or function of miR-138. One predicted function of miR-138 is to target the classic dioxin-responsive gene, CYP1A1 (Table 2). Constitutive CYP1A1 levels are low in both Ahr+/+and Ahr–/– mice (Tijet et al., 2006), and CYP1A1 may be regulated, in part, by miR-138.
Whereas the AHR is completely absent in the Ahr-null mouse model, H/W rats express an AHR in which 38 or 43 amino acids are deleted from the transactivation domain (Pohjanvirta et al., 1998). miRNA levels in H/W rats are not very different from those in L-E rats which have WT AHR. Our array analysis identified five miRNAs whose constitutive expression differs between H/W rats and L-E rats. RT–PCR confirmed higher constitutive levels in H/W rats than L-E rats for miR-498 and mir-122a. However, neither miR-498 nor miR-122a was affected by AHR genotype in the mouse model; therefore, the differences in constitutive levels are unlikely to be AHR-mediated and may reflect differences at non-AHR loci between H/W and L-E strains. Differences in constitutive miRNA expression are unlikely to explain why certain mRNAs are constitutively suppressed in H/W rat nor the H/W rat's ability to avoid toxicities when exposed to dioxins.
miRNAs in Adult Rodent Livers Are Refractory to Dioxin Treatment
Dioxin exposure alters expression of numerous mRNA transcripts in rodent livers and rodent hepatoma cells (Boverhof et al., 2006; Fletcher et al., 2005; Ovando et al., 2006; Puga et al., 2000) through a mechanism that is largely AHR dependent (Tijet et al., 2006). We sought to determine if miRNA levels are modulated by TCDD and if any modulation is dependent on the AHR genotype, as is the case for mRNAs.
Using arrays that contain homologous miRNAs (human/mouse/rat), we generated hepatic miRNA expression profiles in response to TCDD in both rat and mouse. The magnitude of response to TCDD for any particular miRNA in either rodent species was small—only a maximum 2.5-fold TCDD effect. Of the 45 miRNAs indicated as TCDD responsive by array analysis in the mouse model (Table 2) and the 17 miRNAs from the rat model (Table 4), only eight miRNAs were altered by TCDD in both species (shaded miRNA names in Tables 2 and 4). For example, levels of miR-20a were increased 19 h post-TCDD via an AHR-dependent mechanism in mouse and 3 h post-TCDD in dioxin-sensitive L-E rats. Previous in vitro reporter gene assays showed that miR-20a directly regulates expression of Tgfbr2 (Volinia et al., 2006), an mRNA whose expression we previously found to be affected by TCDD (Tijet et al., 2006). RT–PCR analysis of four of the eight miRNAs, which responded to TCDD in both rat and mouse, revealed only miR-203 was affected by TCDD in both species in vivo; reducing the number of potentially conserved TCDD responses to five out of the 464 miRNAs represented on the array. Levels of miR-203 were suppressed 96 h post-TCDD in livers from dioxin-sensitive L-E rats only and were suppressed 19 h post-TCDD in mouse but independent of the AHR. Therefore, miR-203 is unlikely to constitute an early mechanism of AHR-dependent response to dioxin; however, miR-203 levels are altered at 96 h, perhaps, as a consequence of toxicity rather than TCDD per se. Furthermore, TCDD enhanced (rather than suppressed) levels of miR-203 in vitro in the rat 5L cell line, whereas miR-203 was not affected by TCDD in the mouse Hepa-1 cell line. This discrepancy warrants a cautionary note—miRNA expression levels and responses to xenobiotics in commonly used cell lines do not necessarily predict expression levels and responses in vivo.
miRNAs do not appear to be part of the universal mechanism of dioxin toxicity despite a high degree of conservation of miRNA structures and function as well as a considerable conservation in AHR biology across mammalian species. AHR regulons might modulate miRNAs differently in response to TCDD treatment between the two rodent species, in agreement with the highly dissimilar hepatic mRNA expression profiles in rats and mice after TCDD exposure (Boverhof et al., 2006). In the mouse model, our miRNA array analysis revealed that only 30 of the 464 miRNAs on the array were affected by TCDD exposure via an AHR-dependent mechanism, but RT–PCR did not confirm AHR-mediated regulation by TCDD for five of the miRNAs identified by array. While a few of the remaining 25 miRNAs may be regulated by TCDD via an AHR-dependent mechanism, this regulation appears not to be directly via AHR since the 19-h exposure time may also capture secondary effects in addition to primary transcriptional events. RT–PCR confirmed some, but not all, results from miRNA arrays. The discrepancies between array data and RT–PCR data may reflect the fact that effects of TCDD and/or AHR genotype on miRNA levels were of low magnitude and often near the threshold of statistical significance.
Our array analysis identified more miRNAs that were affected by TCDD treatment in dioxin-sensitive L-E rats (15 miRNAs) than in dioxin-resistant rats H/W (three miRNAs). To attempt to identify primary transcriptional events, we examined miRNA levels at short durations of TCDD exposure (3 or 19 h). In dioxin-sensitive L-E rats, eight of the nine miRNA levels measured by array were suppressed 3 h post-TCDD, but RT–PCR analysis did not support a TCDD effect for three of the nine miRNAs. Suppression of miRNA levels after TCDD treatment, if genuine, is unlikely to explain mRNA downregulation by dioxins.
In the dioxin-resistant H/W rat, miR-122a was the only miRNA whose level was significantly affected by TCDD, as measured both by array and by RT–PCR. Levels of miR-122a increased 19 h post-TCDD versus control levels. Furthermore, levels of miR-122a were significantly higher in liver of dioxin-resistant H/W rats than in dioxin-sensitive L-E rats at 3 and 19 h post-TCDD. Dioxin treatment produces biological effects similar to effects produced by silencing miR-122a in vivo with an antagomir or by limiting its biogenesis (Grimm et al., 2006; Krutzfeldt et al., 2005). Both dioxin treatment and silencing of miR-122a alter regulation of cholesterol and fatty acid metabolism, expression of mRNAs (e.g. CYP17a1, CYP7a1, Thrsp, Scd1 and Tgfbp1i4), toxicity, and morbidity. The lower levels of miR-122a in dioxin-sensitive L-E rats (in contrast to the enhanced levels of miR-122a in the dioxin-resistant H/W rat), both constitutively and in response to TCDD treatment, may fail to suppress mRNAs lying in the pathway to dioxin toxicities. However, our experiments in the Ahr-null mouse (Tijet et al., 2006) indicate that regulation of mir-122a is via a mechanism independent of the AHR, whereas TCDD toxicity itself is AHR dependent (Okey, 2007).
Alternative Potential AHR-Mediated Mechanisms for Downregulation of mRNAs
New miRNA species may be discovered whose actions do intersect with the AHR pathway and dioxin responses in specific tissues. However, our extensive interrogation of miRNA expression indicates that dioxin exposure or AHR genotype have remarkably little impact on miRNAs in adult rodent livers. Still the question remains: what mediates the suppression of greater than 100 rodent transcripts (Tijet et al., 2006) in response to TCDD? Answers may lie with epigenetic mechanisms, such as hypermethylation of gene promoters, which have been reported to be associated with gene suppression by dioxins in an AHR-dependent manner (Ray and Swanson, 2004) or with one of the mechanisms proposed for suppression of cytochrome P450 genes—such as steric hindrance, alteration of transcription factor/corepressor binding, or inhibitory dioxin response elements (Duan et al., 1999; discussed in Riddick et al., 2004).
In summary, constitutive levels of 27 miRNAs in mouse liver and four miRNAs in rat liver potentially are influenced by AHR genotype. Only five of the 464 miRNAs examined by array analysis were altered by TCDD treatment in both rodent species. The magnitude of the TCDD effect on any particular miRNA in mouse or rat liver was small. Differences in miRNA expression, either constitutively or following TCDD exposure, do not seem to explain why H/W rats are extraordinarily resistant to TCDD toxicity, whereas L-E rats are highly sensitive. The cumulative results of our experiments indicate that downregulation of mRNA levels by dioxins in adult rodent livers is very unlikely to involve miRNAs.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Supplementary Tables S1–S7 and Figures 1–8 are available online at http://toxsci.oxfordjournals.org/.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Canadian Institutes of Health Research to A.B.O.; the Academy of Finland (grant 211120) to R.P.
|
ACKNOWLEDGMENTS |
|---|
We are grateful to Ms Arja Tamminen and Ms Ulla Naukkarinen for excellent technical assistance.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONSUPPLEMENTARY DATAFUNDINGREFERENCES |
|---|
Babak T, Zhang W, Morris Q, Blencowe BJ, Hughes TR. Probing microRNAs with microarrays: tissue specificity and functional inference. Rna (2004) 10:1813–1819.
Boutros PC, Moffat ID, Franc MA, Tijet N, Tuomisto J, Pohjanvirta R, Okey AB. Dioxin-responsive AHRE-II gene battery: identification by phylogenetic footprinting. Biochem. Biophys. Res. Commun. (2004) 321:707–715.[CrossRef][Web of Science][Medline]
Boverhof DR, Burgoon LD, Tashiro C, Sharratt B, Chittim B, Harkema JR, Mendrick DL, Zacharewski TR. Comparative toxicogenomic analysis of the hepatotoxic effects of TCDD in Sprague Dawley rats and C57BL/6 mice. Toxicol. Sci. (2006) 94:398–416.
Bunger MK, Moran SM, Glover E, Thomae TL, Lahvis GP, Lin BC, Bradfield CA. Resistance to 2,3,7,8-tetrachlorodibenzo-p-dioxin toxicity and abnormal liver development in mice carrying a mutation in the nuclear localization sequence of the aryl hydrocarbon receptor. J. Biol. Chem. (2003) 278:17767–17774.
Chen C, Ridzon DA, Broomer AJ, Zhou Z, Lee DH, Nguyen JT, Barbisin M, Xu NL, Mahuvakar VR, Andersen MR, et al. Real-time quantification of microRNAs by stem-loop RT-PCR. Nucleic Acids Res. (2005) 33:e179.
Dere E, Boverhof DR, Burgoon LD, Zacharewski TR. In vivo-in vitro toxicogenomic comparison of TCDD-elicited gene expression in Hepa1c1c7 mouse hepatoma cells and C57BL/6 hepatic tissue. BMC Genomics (2006) 7:80.[CrossRef][Medline]
Duan R, Porter W, Samudio I, Vyhlidal C, Safe S. Transcriptional activation of c-fos protooncogene by 17beta-estradiol: mechanism of aryl hydrocarbon receptor-mediated inhibition. Mol. Endocrinol. (1999) 13:1511–1521.
Fernandez-Salguero P, Pineau T, Hilbert DM, McPhail T, Lee SS, Kimura S, Nebert DW, Rudikoff S, Ward JM, Gonzalez FJ. Immune system impairment and hepatic fibrosis in mice lacking the dioxin-binding Ah receptor. Science (1995) 268:722–726.
Fletcher N, Wahlstrom D, Lundberg R, Nilsson CB, Nilsson KC, Stockling K, Hellmold H, Hakansson H. 2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) alters the mRNA expression of critical genes associated with cholesterol metabolism, bile acid biosynthesis, and bile transport in rat liver: A microarray study. Toxicol. Appl. Pharmacol. (2005) 207:1–24.[Web of Science][Medline]
Gaidatzis D, van Nimwegen E, Hausser J, Zavolan M. Inference of miRNA targets using evolutionary conservation and pathway analysis. BMC Bioinformatics (2007) 8:69.[CrossRef][Medline]
Gaur A, Jewell DA, Liang Y, Ridzon D, Moore JH, Chen C, Ambros VR, Israel MA. Characterization of microRNA expression levels and their biological correlates in human cancer cell lines. Cancer Res. (2007) 67:2456–2468.
Griffiths-Jones S, Grocock RJ, van Dongen S, Bateman A, Enright AJ. miRBase: microRNA sequences, targets and gene nomenclature 10.1093/nar/gkj131. Nucleic Acids Res. (2006) 34:D140–D144.
Grimm D, Streetz KL, Jopling CL, Storm TA, Pandey K, Davis CR, Marion P, Salazar F, Kay MA. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature (2006) 441:537–541.[CrossRef][Medline]
Hankinson O. Role of coactivators in transcriptional activation by the aryl hydrocarbon receptor. Arch. Biochem. Biophys. (2005) 433:379–386.[CrossRef][Web of Science][Medline]
Karolchik D, Baertsch R, Diekhans M, Furey TS, Hinrichs A, Lu YT, Roskin KM, Schwartz M, Sugnet CW, Thomas DJ, et al. The UCSC genome browser database. Nucleic Acids Res. (2003) 31:51–54.
Krutzfeldt J, Rajewsky N, Braich R, Rajeev KG, Tuschl T, Manoharan M, Stoffel M. Silencing of microRNAs in vivo with ‘antagomirs’. Nature (2005) 438:685–689.[CrossRef][Medline]
Lagos-Quintana M, Rauhut R, Yalcin A, Meyer J, Lendeckel W, Tuschl T. Identification of tissue-specific microRNAs from mouse. Curr. Biol. (2002) 12:735–739.[CrossRef][Web of Science][Medline]
Lahvis GP, Lindell SL, Thomas RS, McCuskey RS, Murphy C, Glover E, Bentz M, Southard J, Bradfield CA. Portosystemic shunting and persistent fetal vascular structures in aryl hydrocarbon receptor-deficient mice. Proc. Natl. Acad. Sci. U.S.A. (2000) 97:10442–10447.
Megraw M, Sethupathy P, Corda B, Hatzigeorgiou AG. miRGen: A database for the study of animal microRNA genomic organization and function. Nucleic Acids Res. (2007) 35:D149–D155.
Mimura J, Yamashita K, Nakamura K, Morita M, Takagi TN, Nakao K, Ema M, Sogawa K, Yasuda M, Katsuki M, et al. Loss of teratogenic response to 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in mice lacking the Ah (dioxin) receptor. Genes Cells (1997) 2:645–654.[Abstract]
Nebert DW, Roe AL, Dieter MZ, Solis WA, Yang Y, Dalton TP. Role of the aromatic hydrocarbon receptor and [Ah] gene battery in the oxidative stress response, cell cycle control, and apoptosis. Biochem. Pharmacol. (2000) 59:65–85.[CrossRef][Web of Science][Medline]
Niittynen M, Simanainen U, Syrjala P, Pohjanvirta R, Viluksela M, Tuomisto J, Tuomisto JT. Differences in acute toxicity syndromes of 2,3,7,8-tetrachlorodibenzo-p-dioxin and 1,2,3,4,7,8-hexachlorodibenzo-p-dioxin in rats. Toxicology (2007) 235:39–51.[CrossRef][Web of Science][Medline]
Okey AB. An AH receptor odyssey to the shores of toxicology: The Deichmann lecture—International Congress of Toxicology-XI. Toxicol. Sci. (2007) 98:5–38.
Okey AB, Franc MA, Moffat ID, Tijet N, Boutros PC, Korkalainen M, Tuomisto J, Pohjanvirta R. Toxicological implications of polymorphisms in receptors for xenobiotic chemicals: The case of the aryl hydrocarbon receptor. Toxicol. Appl. Pharmacol. (2005) 207:S43–S51.[CrossRef]
Ovando BJ, Vezina CM, McGarrigle BP, Olson JR. Hepatic gene downregulation following acute and subchronic exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin. Toxicol. Sci. (2006) 94:428–438.
Peters JM, Narotsky MG, Elizondo G, Fernandez-Salguero PM, Gonzalez FJ, Abbott BD. Amelioration of TCDD-induced teratogenesis in aryl hydrocarbon receptor (AhR)-null mice. Toxicol. Sci. (1999) 47:86–92.
Pohjanvirta R, Tuomisto J. Short-term toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals: Effects, mechanisms, and animal models. Pharmacol. Rev. (1994) 46:483–549.[Web of Science][Medline]
Pohjanvirta R, Wong JM, Li W, Harper PA, Tuomisto J, Okey AB. Point mutation in intron sequence causes altered carboxyl-terminal structure in the aryl hydrocarbon receptor of the most 2,3,7,8-tetrachlorodibenzo-p-dioxin-resistant rat strain. Mol. Pharmacol. (1998) 54:86–93.
Puga A, Maier A, Medvedovic M. The transcriptional signature of dioxin in human hepatoma HepG2 cells. Biochem. Pharmacol. (2000) 60:1129–1142.[CrossRef][Web of Science][Medline]
Ray SS, Swanson HI. Dioxin-induced immortalization of normal human keratinocytes and silencing of p53 and p16INK4a. J. Biol. Chem. (2004) 279:27187–27193.
Reiner A, Yekutieli D, Benjamini Y. Identifying differentially expressed genes using false discovery rate controlling procedures. Bioinformatics (2003) 19:368–375.
Riddick DS, Lee C, Bhathena A, Timsit YE, Cheng PY, Morgan ET, Prough RA, Ripp SL, Miller KK, Jahan A, et al. Transcriptional suppression of cytochrome P450 genes by endogenous and exogenous chemicals. Drug Metab. Dispos. (2004) 32:367–375.
Siepel A, Haussler D. Combining phylogenetic and hidden Markov models in biosequence analysis. J. Comput. Biol. (2004) 11:413–428.[CrossRef][Web of Science][Medline]
Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat. Appl. Genet. Mol. Biol. (2004) 3:3.
Tijet N, Boutros PC, Moffat ID, Okey AB, Tuomisto J, Pohjanvirta R. Aryl hydrocarbon receptor regulates distinct dioxin-dependent and dioxin-independent gene batteries. Mol. Pharmacol. (2006) 69:140–153.
Tsuchiya Y, Nakajima M, Takagi S, Taniyam T, Yokoi T. MicroRNA regulates the expression of human cytochrome P450 1B1. Cancer Res. (2006) 66:9090–9098.
Vester B, Wengel J. LNA (locked nucleic acid): High-affinity targeting of complementary RNA and DNA. Biochemistry (2004) 43:13233–13241.[Medline]
Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, Visone R, Iorio M, Roldo C, Ferracin M, et al. A microRNA expression signature of human solid tumors defines cancer gene targets. Proc. Natl. Acad. Sci. U.S.A. (2006) 103:2257–2261.
Walisser JA, Bunger MK, Glover E, Bradfield CA. Gestational exposure of Ahr and Arnt hypomorphs to dioxin rescues vascular development. Proc. Natl. Acad. Sci. U.S.A. (2004) 101:16677–16682.
Wettenhall JM, Smyth GK. limmaGUI: A graphical user interface for linear modeling of microarray data. Bioinformatics (2004) 20:3705–3706.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  P. C. Boutros, K. A. Bielefeld, R. Pohjanvirta, and P. A. Harper Dioxin-Dependent and Dioxin-Independent Gene Batteries: Comparison of Liver and Kidney in AHR-Null Mice Toxicol. Sci., November 1, 2009; 112(1): 245 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Celius, S. Roblin, P. A. Harper, J. Matthews, P. C. Boutros, R. Pohjanvirta, and A. B. Okey Aryl Hydrocarbon Receptor-Dependent Induction of Flavin-Containing Monooxygenase mRNAs in Mouse Liver Drug Metab. Dispos., December 1, 2008; 36(12): 2499 - 2505. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hudder and R. F. Novak miRNAs: Effectors of Environmental Influences on Gene Expression and Disease Toxicol. Sci., June 1, 2008; 103(2): 228 - 240. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||