TOXICOLOGICAL SCIENCES 89(1), 120–134 (2006)

doi:10.1093/toxsci/kfj017
Advance Access publication October 26, 2005

A Toxicogenomics Approach to Identify New Plausible Epigenetic

Mechanisms of Ochratoxin A Carcinogenicity in Rat
M. Marin-Kuan,*,1 S. Nestler,†,‡ C. Verguet,* C. Bezencxon,* D. Piguet,* R. Mansourian,* J. Holzwarth,* M. Grigorov,*
T. Delatour,* P. Mantle,‡ C. Cavin,* and B. Schilter*
*Nestlé Research Center, PO Box 44, Vers-chez-les-Blanc, CH-1000 Lausanne 26, Switzerland; †Nephro-Urology Unit, Institute of Child Health,
University College London, 30 Guilford Street, London WC1E 1EH, UK; and ‡Department of Environmental Science and Technology,
Imperial College London, London SW7 2AZ, UK

Received August 24, 2005; accepted October 7, 2005

Ochratoxin A (OTA) is a naturally occurring mycotoxin

Ochratoxin A (OTA) is a mycotoxin occurring naturally in

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

produced by several species of Aspergillus and Penicillium. As
a wide range of food commodities. In animals, it has been shown a consequence of its widespread occurrence in a variety of food
to cause a variety of adverse effects, nephrocarcinogenicity being
commodities such as cereals, green coffee, cocoa, dried fruits,
the most prominent. Because of its high toxic potency and the
continuous exposure of the human population, OTA has raised and meat products (WHO, 2001), the human population is
public health concerns. There is significant debate on how to use continuously exposed to OTA, although usually only in trace
the rat carcinogenicity data to assess the potential risk to humans. amounts.
In this context, the question of the mechanism of action of OTA causes a range of adverse effects in monogastric
OTA appears of key importance and was studied through the animals, with renal toxicity being the most prominent (NTP,
application of a toxicogenomics approach. Male Fischer rats were 1989; O’Brien and Dietrich, 2005; WHO, 2001). Other OTA-
fed OTA for up to 2 years. Renal tumors were discovered during mediated toxic effects observed in vitro or in vivo include
the last 6 months of the study. The total tumor incidence reached inhibition of protein synthesis (Dirheimer and Creppy, 1991),
25% at the end of the study. Gene expression profile was analyzed
impairment of energetic metabolism (Gekle et al., 2005; WHO,
in groups of animals taken in intervals from 7 days to 12 months.
Tissue-specific responses were observed in kidney versus liver. For 2001), induction of oxidative stress (Gautier et al., 2001a;
selected genes, microarray data were confirmed at both mRNA Kamp et al., 2005; Petrik et al., 2003; WHO, 2001), and
and protein levels. In kidney, several genes known as markers of apoptosis (Gekle et al., 2005; O’Brien and Dietrich, 2005;
kidney injury and cell regeneration were significantly modulated Petrik et al., 2003; Sauvant et al., 2005).
by OTA. The expression of genes known to be involved in DNA In a 2-year carcinogenicity study, OTA administration by
synthesis and repair, or genes induced as a result of DNA damage, oral gavage caused dose-dependent incidence of renal tumors
was only marginally modulated. Very little or no effect was found in male rats, while females were much less susceptible (NTP,
amongst genes associated with apoptosis. Alterations of gene 1989). Similar effects were observed in mice, although at much
expression indicating effects on calcium homeostasis and a dis-
higher doses (Bendele et al., 1985). OTA has been suspected to
ruption of pathways regulated by the transcription factors hepato-
cyte nuclear factor 4 alpha (HNF4a) and nuclear factor-erythroid be an etiological agent in the human disease Balkan endemic
2-related factor 2 (Nrf2) were observed in the kidney but not nephropathy (BEN) and in its association with an increased
in the liver. Previous data have suggested that a reduction in incidence of urinary tract tumors (Petkova-Bocharova et al.,
HNF4a may be associated with nephrocarcinogenicity. Many 1988; Plestina et al., 1990). However, causal involvement of
Nrf2-regulated genes are involved in chemical detoxication and the toxin in these two diseases lacks robust etiological
antioxidant defense. The depletion of these genes is likely to evidence, and the actual health significance of current natural
impair the defense potential of the cells, resulting in chronic OTA exposure in the human is still unclear.
elevation of oxidative stress in the kidney. The inhibition of At present, there is still insufficient mechanistic under-
defense mechanism appears as a highly plausible new mechanism,
standing of OTA toxicity, particularly in respect to the
which could contribute to OTA carcinogenicity.
Key Words: ochratoxin A; nephrocarcinogenicity. occurrence of renal carcinomas. OTA has, for instance, been
shown to alter several cell-signaling pathways known to be
involved in carcinogenesis. They include modulation of the
calcium-dependent signaling and the mobilization of two
1
To whom correspondence should be addressed at Nestle Research Center,
mitogen-activated protein kinases (MAPK), the extracellular
P.O. Box 44, Vers-chez-les Blanc, CH-1000 Lausanne 26, Switzerland. Fax: signal-regulated kinase (ERK1/2) and the c-jun amino terminal
þþ41 21 785 85 53. E-mail: [email protected]. kinase (JNK1/2) (Gekle et al., 2005; Sauvant et al., 2005).

The Author 2005. Published by Oxford University Press on behalf of the Society of Toxicology. All rights reserved.
For Permissions, please email: [email protected]
 OCHRATOXIN A CARCINOGENICITY 121

Since the carcinogenic mechanism remains obscure, there period of the study, animals were housed in cages on absorbent paper under
is still significant debate about the best model for human tightly controlled conditions (21 ± 1C, 55 ± 10% relative humidity, air-
exchange, 12-h light-dark cycle). Animal growth and welfare were monitored
risk assessment. Knowledge of the extent of a toxicant’s by regular weighing and daily surveillance. For the gene expression study, five
genotoxicity is vital in evaluating its potency as a carcinogen. time points were selected to represent early (7 and 21 days) and later (4, 7, and
At present, the genotoxic status of OTA is still unclear 12 months) responses. At each time point, four control and treated animals were
(Manderville, 2005; Turesky, 2005), mainly due to contradic- randomly chosen for tissue harvest. Kidneys and liver from these rats were im-
tory experimental evidence. (1) Applying 32P-postlabelling, mediately snap-frozen in liquid nitrogen. All handling and procedures were
carried out in accordance with the UK Animals (Scientific Procedures) Act 1986.
some authors have found low levels of DNA lesions in organs
of animals treated with OTA (Castegnaro et al., 1998; Faucet RNA isolation. Homogenates from liver and kidney tissue samples were
prepared with the FastPrep system (Q.BIOgene, Germany). Total RNA was
et al., 2004; Pfohl-Leszkowicz et al., 1993). These lesions have isolated using the Clontech Atlas Pure Total RNA Labeling System (BD
been interpreted as OTA-specific covalent DNA adducts, Biosciences, Switzerland) following the manufacturer’s protocol. Total RNAs
suggesting that OTA is a genotoxin sensu stricto. Such were treated with DNase I to remove genomic DNA contamination. RNA
a compound would require dose-response modeling for concentration was measured with the RiboGreen RNA quantification Kit
quantitative risk assessment. It has, however, also been (Molecular Probes, Eugene, OR) according to the manufacturer’s instruction.
Assessment of RNA quality and measurement of the 28S to 18S ratio were
suggested that OTA-mediated DNA lesions detected by 32P- performed by dynamic gel electrophoresis using the Agilent Bioanalyser
postlabelling may have arisen through indirect mechanisms (Agilent Biotechnologies, Germany).
rather than direct DNA-binding (Gautier et al., 2001b). (2) In

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

cRNA preparation. Five lg of total RNA from kidney and liver samples of
studies that used radiolabeled OTA, DNA-binding could not be treated and control rats was used for the synthesis of double-stranded cDNA
detected (Gautier et al., 2001b; Gross-Steinmeyer et al., 2002; (Superscript Choice System; Invitrogen Life Technologies, Carlsbad, CA) in
Mally et al., 2004), indicating that the compound may not be the presence of a T7-(dT) DNA oligonucleotide primer. After synthesis, cDNA
was purified by phenol/chloroform/isoamyl alcohol extraction and subsequent
directly genotoxic. If this is the case, an alternative safety-
ethanol precipitation. The purified cDNA was transcribed in vitro using
based approach would need to take into account uncertainty a MEGAscript kit (Ambion Diagnostics, Austin, TX) in the presence of
factors in order to evaluate the human health risk of OTA biotinylated ribonucleotides to form biotin-labeled cRNA (ENZO Life
exposure. Sciences, Farmingdale, NY). Labeled cRNA was purified using a nucleospin
The aim of the present study was to exploit the power of matrix (Macherey-Nagel, Germany), quantified by RiboGreen, and its quality
toxicogenomics to seek evidence for potential involvement of was assessed with Agilent Bioanalyzer before subsequent fragmentation.
epigenetic mechanisms in OTA carcinogenesis under exposure Hybridisation and labeling. Twenty lg of each fragmented cRNA was
conditions producing a significant incidence of renal tumors hybridized to a GeneChip cartridge, rotated at 60 rpm for 16 h at 40C. Sample
was analyzed independently using the Affymetrix platform RG-U34A expres-
without overt toxicity. Gene expression profiles were analyzed sion probe arrays (Affymetrix Inc., Santa Clara, CA) containing 8,799 probe
in kidney and liver of male Fischer (F-344) rats, given dietary sets. For each time point, at least three independent controls and treated animals
OTA over various periods ranging from 7 days to 12 months. were analyzed individually on a GeneChip cartridge. Quality control steps were
This duration spans the first half of the standard 2-year protocol performed during the microarray procedure and were collected according to the
for studying lifetime responses to carcinogens. laboratory information management system (L.I.M.S). After hybridization,
GeneChips were washed 10 times in 63 SSPE-T at 25C, followed by four
washes in 0.53 SSPE-T at 50C. GeneChips were stained with 10 lg/ml
streptavidin-P-phycoerythrin, SAPE (Molecular Probes Inc., Eugene, OR), in
a GeneChip Fluidics Workstation 400 (Affymetrix Inc., Santa Clara, CA). The
MATERIALS AND METHODS signal was amplified using biotinylated goat antistreptavidin antibody (Vector
Laboratories, Burlingame, CA), followed from a second staining with SAPE.
Ochratoxins. Standardized OTA production was performed by growing The array was washed and scanned at 560 nm using a confocal laser scanner
A. ochraceus isolate D2306 (Harris and Mantle, 2001) in shaken solid substrate (GeneArray Scanner 2500, Hewlett Packard).
fermentations at 28C for 2 weeks to yield a product containing 5–6 mg OTA/g.
Data analysis and clustering algorithms. Signals from scanned images
More specifically, 40 g of sterilized shredded wheat (Cereal Partners UK,
were quantified using the Affymetrix Gene Expression Analysis software
Welwyn, UK) in 500 ml Erlenmeyer flasks was inoculated with a concentrated
(MAS 5.0). Normalization of gene expression data was performed. Differences
spore suspension in water (16 ml), and the flasks were shaken at 200 rpm and 10
in gene expression occurring in treated rats compared to controls were
cm eccentric throw. An aliquot of each fermentation was assayed for OTA-
calculated by applying analysis of variance (ANOVA). This analysis was
concentration by HPLC with diode array detection (Harris and Mantle, 2001).
performed on log-transformed signal values for each gene individually (By-
Batches of fermentation product contained ochratoxin B (OTB) equivalent to
Gene test). Three factors were used for ANOVA: treatment, time, and the
5–10% of the amount of OTA. No other mycotoxins (e.g., penicillic acid,
interaction between treatment and time. ANOVA between the treatments was
citrinin) were biosynthesized in this fermentation. Each week, a weighed
performed applying the Global Error Assessment method (GEA) as previously
amount of standardized fermentation product was homogenized into powdered
described (Mansourian et al., 2004; Mutch et al., 2004). This approach
standard commercial rat feed (Special Diets Services, UK) to a final
increases the statistical power of the data. Complete data sets generated in this
concentration appropriate for the required OTA-intake per animal. Each animal
study are available at the NCBI Gene Expression Omnibus (http://
was given 20 g of contaminated feed daily, which was always fully consumed.
www.ncbi.nlm.nih.gov/geo/), access number GSE2852.
Animal treatment. Male Fischer 344 (F-344) rats (B. & K. Universal Ltd., Gene clustering dendrograms were obtained with the software Spotfire
Hull, UK) in groups of five were administered OTA in diet given daily over (Spotfire Inc.) following the UPGMA clustering and Tanimoto similarity
2 years. As from their initial weight of ~175 g, daily dietary intake was 300 lg measure (Matter, 1997). Only genes with a significant p
0.001 in at least two
OTA/kg bw, but was held at 100 lg/rat after animals reached 333 g. Over the time points were included in dendrograms of differential expression of genes in
122 MARIN-KUAN ET AL.

kidney and liver. The vertical axis of each dendrogram shows individually the gene expression study originated. The body weight of
selected probes, whereas the horizontal axis represents the clustering level at treated animals did not differ significantly from that of control
the time points studied (7 days, 21 days, 4 months, 7 months, and 12 months).
animals. All tumors that were found before the 2-year end point
Single value decomposition. The variability of gene modulation induced of the carcinogenicity study were from animals that had been
by OTA during the time was also investigated by a matrix factorization
technique equivalent to Principal Component Analysis (PCA), the singular
euthanized because of loss of condition. The first renal tumor
value decomposition (SVD) method (Alter et al., 2003). SVD method provides was found 75 weeks after OTA treatment commenced (Fig. 1).
a low-dimensional projection of the original dataset by its decomposition in The total incidence of animals with a renal tumor reached 25%
three quantities—the matrices containing the left- and right-singular vectors of (n ¼ 64) at the end of the study. At least 75% of these animals
the original data matrix, and the diagonal matrix containing the singular values. had unilateral renal carcinoma as determined by tumor size,
Clusters formed by the SVD method can be interpreted as the relative
frequencies with which a given gene is modulated at every one of the original
proliferating and disorganized histopathology, and abundance
single time points which, when combined, form the temporal modes of gene of metastasis (Mantle et al., in press).
modulation. Average from each SVD-generated cluster was performed in order
to evaluate the profile of the gene expression during the time. Gene Selection and Clustering
Real-time polymerase chain reaction (PCR). Several key genes, each of
which represents a category of genes found to be modulated by OTA, were
Applying the stringency criteria for selection of significantly
selected to validate the microarray data. RNA preparation, quantification, and differentiated genes, statistical analysis revealed that the
quality control were conducted as described above. RNA was converted to expression of 470 genes in the kidney and 233 genes in the

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

cDNA using the two-step TaqMan Gold RT-PCR system (Applied Biosystem, liver were differentially modulated (p
0.001 in at least two
Foster City, CA) containing the TaqMan Reverse Transcription Reagents and time points) by the OTA treatment. Of these, only 45 genes
the TaqMan PCR Core kit for relative quantification experiments. Reverse
transcription was performed in triplicate, according to the manufacturer’s
were modulated in both kidney and liver.
instructions (Applied Biosystems, Foster City, CA), using the random hexamers Hierarchical clustering analysis was performed for kidney
primer and 5 lg of pooled RNA samples. cDNA synthesis was performed in and liver mRNA expression profiles (Figs. 2A and 2B). The
a Thermal Cycler 9600 (Applied Biosystems) under the following conditions: vertical axis of the dendrograms represents each transcript, and
10 min at 25C, 30 min at 48C, and a final inactivation step of 5 min at 95C. the x-axis represents the various time points, and at the
Samples were amplified in a MicroAmp 96-well reaction plate with the ABI
Prism 7000 Sequence Detection System using 1 ll of cDNA template and the
intersection of these two values, the expression intensity as
TaqMan PCR kit (Applied Biosystems). The thermocycler conditions included compared to control (fold-expression) can be seen. The
two incubations of 2 min at 50C and 10 min at 95C, followed by 40 cycles, organization in both the horizontal and vertical axes represents
each consisting of a denaturing step for 15 sec at 95C and a second annealing the similarity of the expression values as detailed by the tree
and extension step for 1 min at 60C. Quantification of amplified PCR products connecting each category. The distance of the connecting lines
was performed using ABI Prism 7000 Sequence Detection System software
Version 1.1, normalized to the rat 18S gene as internal control. All rat PCR
in the tree represents the extent of similarity. Gene expression
probes and primers used in this study were obtained from the Assays on clusters were generated accordingly; interestingly, the 7-day
Demand (AoD) Service of Applied Biosystems. and 12-month profiles exhibited the most similar time points
Protein expression analysis by Western blot. Correlation between mRNA (in terms of the expression values of the affymetrix probes at
and protein expression was confirmed by Western blot. Briefly, 50 mg of liver these time points) in both organs. In the kidney, these two time
and kidney tissue samples were directly homogenized using RIPA (150 mM points taken together are then most similar to the 7-month time
PBS containing 1% (vol/vol) Igepal CA630, 0.5% (wt/vol) sodium deoxy- point; these three are most similar to the 21-day and, finally, the
cholate, 0.1% (wt/vol) SDS, and 5 lg/ll protease inhibitor mixture), pH 7.4
(SantaCruz Biotechnology, Santa Cruz, CA) following the manufacturer’s
4-month time point. In the liver, these two time points taken as
instructions. Protein concentration was determined with a BioRad Protein assay a cluster are most similar to the 12-month, 7-day, and finally,
kit (Bio-Rad, Richmond, CA). Using the MiniCell XCellSureLock system the 7-month time point. In both organs, down regulation was
(Invitrogene), 15 lg of protein was loaded in the NuPAGE Bis-Tris Pre-Cast
Gel 4–12% System (Invitrogene Ltd, Paisley, UK). Protein was transferred to
a nitrocellulose transfer membrane (Invitrogene) for 1 h at 30 V. Transferred Cumulative tumor incidence
membranes were probed with antibodies specific for the proteins Gstp, Gclc 18
Number of animals with a

(kindly offered by Dr. Leslie McLellan, Dundee University, Scotland, UK), 16

Oatp1 (kindly offered by Prof. B. Hagenbuch, University of Kansas Medical 14
Center, Kansas City, KS), HNF4a (Abcam, Cambridge, UK), Oct2 and Oat-k1 12
renal tumor

(Alpha Diagnostic, San Antonio, TX). Each blot was subsequently probed with
10
horseradish peroxidase linked to the specific secondary antibody. Chemilumi-
8
nescence of the immunoblots was detected using ECL solution (Pierce,
6
Rockford, IL). Membranes were wrapped in polythene film and exposed to
Kodak films. 4
2
0
70 75 80 85 90 95 100 105 110
RESULTS
Weeks of OTA-treatment
OTA treatment was generally well tolerated throughout the FIG. 1. Cumulative renal tumor incidence during the last quarter of live in
2-year carcinogenicity study from which the tissue samples for rats given OTA-contaminated diet continuously for up to 2 years.
 OCHRATOXIN A CARCINOGENICITY 123

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

FIG. 2. Hierarchical clustering representing the genes significantly modulated ( p
0.001) in at least two time points in the kidney (A) and the liver (B). The
y-axis of the dendrograms represents the genes and the corresponding expression level displayed in green for down-regulation, red for up-regulation, and black for
insignificant change in gene expression. Time points of analysis are plotted on the x-axis at 7 days (7D), 21 days (21D), 4 months (4M), and 12 months (12M), and
their corresponding time clustering is shown at the top of each dendrogram. Mean values of the fold-expression change for the eight clusters, as determined by the
single value decomposition (SVD) method, are plotted as a function of time points for kidney (C) and liver (D). The two main clusters identified in each organ are
highlighted in bold (red squares: up-regulation; grey triangles: down-regulation). They illustrate the gene expression pattern typical for the majority of significantly
modulated genes.

the predominant effect, with 60% and 56% of genes down- time (Figs. 2C and 2D). In both organs, eight clusters with
regulated in kidney and liver, respectively. differentiated time-dependent expression profile were identi-
Further analysis of the gene expression profile in kidney fied. The modulation of OTA-induced gene expression was
and liver was performed by application of a single value generally less than two-fold. The two main clusters include
decomposition (SVD) method along the time coordinate. The only genes that, over the five time points, are either continu-
mean fold-expression values were plotted as a function of ously up-regulated (31% in kidney, 25% in liver) or
124 MARIN-KUAN ET AL.

Glutathione metabolism

Cytochrome P450

Xenobiotics metabolism

Oxidoreductase activity

DNA (synthesis, repair, damage)

Oncogenes and cell growth

Apoptosis

Cell cycle
Biological function

Cell differentiation/development
Binding

Cell adhesion

Transporters

Regulation of transcription

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

Signaling

Lipids and Cholesterol metabolism

Androgen and estrogen metabolism

Immune response

Coagulation cascades and blood pressure

Muscle, bone, cytoskeletal and cell organization

Metabolism

Miscellaneous

0 5 10 15 20 25 30 35 40 45
Number of genes

FIG. 3. Biological function clustering. Genes exhibiting at least 1.5-fold up- or down-regulation in kidney (white columns) and liver (black columns), plotted
as a function of the total number of genes in each category.

continuously down-regulated (47% in kidney, 27% in liver). No as being under transcriptional control of HNF4a (Table 2). Of
other clear and easily interpretable time-dependent trends, such interest was the finding that the mRNA specific for HNF4a
as an early and/or late response, could be observed. itself was down-regulated.
Grouping of genes modulated by OTA was performed Expression of many transporter genes was down-regulated in
according to their biological functions. For this analysis, a gene the kidney (Table 3). Several genes known as markers of kidney
ontology approach was applied to genes exhibiting at least 1.5- injury, cell regeneration, and oncogenesis were significantly
fold expression values. The number of genes attributed to each modulated by OTA. For example, the kidney injury molecule
functional class is depicted in Figure 3, showing a wide (KIM-1, 4.4-fold) and the survival markers c-myc (3.7-fold),
diversity of highly tissue-selective biological responses. In the Akt-1 (1.7-fold), Polo-like kinase (Plk, 1.5-fold), and Cdkn1a
kidney, genes involved in xenobiotic metabolism and oxida- alias p21 (2.3-fold) were up-regulated. The regucalcin (REG),
tive stress response were generally down-regulated by OTA, also known as the senescence marker protein 30 (Smp30), was
whereas this group of genes was much less modulated in the down-regulated more than 10-fold by OTA.
liver. Interestingly, many of these genes share the antioxidant Only small changes occurred in expression of genes known
regulatory element (ARE) as a regulatory motif (Table 1). to be involved in DNA synthesis and repair or in genes induced
Other gene classes, such as several enzymes involved in fatty as a result of DNA damage. Few genes associated with DNA
acid metabolism and cytochrome P450, were also selectively repair were slightly down-regulated. Similarly, very little or no
down-regulated in the kidney of OTA-treated animals. Many of effects were found on expression of apoptosis-related genes.
these genes possess a common promoter targeted by the Three genes involved in the regulation of protein synthesis
transcription factor hepatocyte nuclear factor alpha (HNF4a). were modulated by OTA. The prostaglandin F2 receptor nega-
Several genes, which belong to different functional classes and tive regulator (Ptgfrn) and the eukaryotic translation initia-
were down-regulated following OTA-treatment, were identified tion factor 4E binding protein 1 (Eif4ebp1) were up-regulated
 OCHRATOXIN A CARCINOGENICITY 125

TABLE 1
Genes Modulated by OTA under Transcriptional Control of Nrf2

Accession no. Gene title Symbol Folda Fold 7D Fold 21D Fold 4M Fold 7M Fold 12M

Xenobiotic metabolism
X65296 carboxylesterase 3 Ces3 3.4* 1.5 9.8* 4.2* 3.3* 2.0*
M33747 UDP-glucuronosyltransferase 2 family, member 5 Ugt2b5 2.7* 2.5 1.3 5.2* 3.1 3.0
X81395 carboxylesterase 1 Ces1 1.9* 2.0 2.2 2.7 1.8 1.3
S81433 Heme-oxygenase 2/5# region HMOX2 1.6* 2.4 3.0* 1.1 1.0 1.6
AF045464 aflatoxin B1 aldehyde reductase Akr7a3/ Afar 1.5* 1.0 2.0* 2.1* 1.2 1.4
J02722cds heme oxygenase 1 Hmox1 2.1* 1.4 4.4 1.6* 1.3 3.2*
Carbon metabolism
AI008020 malic enzyme 1 Me1 1.8* 1.3 2.8* 3.2* 1.0 1.7
AJ005046 fructose bisphosphatase 2 Fbp2 2.6* 1.8 1.8 2.9 4.1 3.2
Glutathionerelated genes
J05181 glutamate-cysteine ligase catalytic subunit Gclc 4.8* 3.4* 8.0* 6.0* 2.7* 5.8*
S65555 glutamate cysteine ligase, modifier subunit Gclm 3.1* 3.5* 3.4* 4.1* 1.7 3.5*
L38615 glutathione synthetase Gss 1.9* 1.1 2.5* 2.7* 1.8* 1.8

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

AI138143 glutathione S-transferase, theta 2 Gstt2 1.8* 1.7 1.8* 2.6* 1.7* 1.6
X03518 gamma-glutamyl transpeptidase Ggtp 1.8* 1.0 1.8 2.6* 1.8 2.1*
X02904 glutathione S-transferase, pi 2 Gstp2 1.7* 1.4 1.5 2.0* 1.8 2.1*
S72506 glutathione-S-transferase, alpha type2 Gsta2 3.8* 1.6 2.9 21.3* 4.4 4.3
U76252 gamma-glutamyltransferase-like activity 1 Ggtla1 1.7* 2.8* 2.2 1.5 1.1 1.2
Amino acid metabolism
M93297 ornithine aminotransferase Oat 2.7* 1.6 6.2* 2.4* 2.8* 2.1*
Others
X06107 insulin-like growth factor 1 Igf1 1.9* 2.8 1.5 3.8 2.9 4.9*
L48060 prolactin receptor Prlr 1.7* 1.1 3.4* 1.6 1.4 1.8
X71127 complement component 1, q subcomponent, C1qb 2.0* 1.6 3.6* 1.7 2.0* 1.8
beta polypeptide

a
Average fold-change in gene expression from the five time points (7D and 21D days and 4M, 7M, and 12M months).
*Statistical significance represented by the alpha risk <0.001.

(3.2- and 1.5-fold, respectively), while the Elongation factor nephrotoxicity and nephrocarcinogenicity in rat (NTP, 1989).
1-alpha 1 (eEF1A-1) was down-regulated (3.3-fold). There is significant debate on the use of rat carcinogenicity data
to assess the potential carcinogenic risk to humans. A very
Confirmation of Microarray Expression Data relevant question with respect to OTA-mediated carcinogenic-
ity and risk assessment is whether the toxin acts through
In order to confirm the microarray data, selected gene
genotoxic or epigenetic mechanisms. For human safety
products assigned to various functional classes were analyzed
assessment, a nongenotoxic mechanism of OTA would trigger
by quantitative real-time PCR (TaqMan-PCR). As with the
the use of a health-based guidance value established by the
microarray data, TaqMan RT-PCR data is presented as fold-
application of uncertainty factors to the pivotal No Observed
expression value of OTA-treated samples as compared to
Adverse Effect Level obtained in animal studies (WHO, 2001).
controls. As shown in Figure 4, expression values obtained
The present work was aimed to generate data relevant to
with RT-PCR are highly comparable to the data generated by
understanding the mechanism of OTA action and should help to
microarray technology.
select the most appropriate risk assessment procedure. This
Protein expression of six genes was measured from samples
study was intended better to reflect the pattern of human
taken at the early and late time-points (21 days and 12 months)
exposure, and therefore, for the first time, oral feeding was
by Western blot analysis. A good correlation was observed be-
chosen as the route of administration. This allowed ingestion of
tween microarray data and measured protein expression (Fig. 5).
OTA in the most natural way for the experimental animals,
homogenized into feed and consumed as desired, mainly in the
DISCUSSION dark part of the diurnal cycle. The dose was chosen to produce
tumors, although avoiding overt nephrotoxicity, which could
OTA is a common trace contaminant of some foodstuffs. It interfere with the carcinogenic process. A previous acute study
produces various toxicological effects, the most relevant being had shown a marked difference between the toxicity of single
126 MARIN-KUAN ET AL.

TABLE 2
Genes Modulated by OTA under Transcriptional Control of HNF4a

Accession no. Gene title Gene symbol Folda Fold 7D Fold 21D Fold 4M Fold 7M Fold 12M

Glucose metabolism
X05684 pyruvate kinase, liver and RBC Pklr 1.6* 1.8 2.1* 1.8 1.6 1.1
Androgen and estrogen metabolism
L19998 sulfotransferase family 1A, Sult1a1 1.9* 1.3 2.0* 2.2* 2.0* 1.9*
phenol-preferring, member 1
Lipids and steroid metabolism
M73714 aldehyde dehydrogenase family 3, Aldh3a2 1.9* 1.6 1.2 4.4* 1.0 4.6*
subfamily A2
D90109 fatty acid Coenzyme A ligase Facl2 1.6* 2.2* 1.1 1.7 1.5 1.8*
Xenobiotic metabolism
J02657 cytochrome P450, subfamily IIC Cyp2c 10.0* 8.2* 7.5* 64.9* 1.8 13.5*
J02861 cytochrome P450 2c13 Cyp2c13 2.4* 2.5* 1.1 2.6* 4.8* 2.2*
M18335 cytochrome P450, 2c39 Cyp2c39 1.5* 1.3 1.2 1.1 3.0* 2.0*

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

Lipids and retinal transport
M10934 retinol binding protein RBP4 2.2* 1.3 4.9* 4.6* 1.7* 1.8*
S69874 fatty acid binding protein 5, Fabp5 2.1* 1.6 2.7* 1.8* 2.1* 2.5*
epidermal
M27440 apolipoprotein B Apob 1.6* 1.4 1.8* 2.7 1.7* 1.1*
D38380 transferrin Tf 2.5* 3.2* 1.9* 1.4 3.0* 3.9*
Transcription factors
X57133 hepatocyte nuclear factor 4, alpha Hnf4a 2.4* 1.9 4.3* 1.3 3.8* 2.0*
Coagulation cascades and blood
pressure
M12112 angiotensinogen Agt 3.5* 1.1 10.5* 5.8* 2.2* 4.4*
M81397 coagulation factor 2 F2 1.5* 1.6* 2.3* 1.7* 1.1 1.1
Transporters (See Table 3)

a
Average fold-change in gene expression from the five time points (7D and 21D days and 4M, 7M, and 12M months).
*Statistical significance represented by the alpha risk <0.001.

daily doses by gavage and the same amount homogenized into incidence of carcinomas was significantly lower than that
feed (Miljkovic et al., 2003). In the initial phase of the study observed in the NTP study at the highest dose (Mantle et al.,
(until rats reached 333 g), the chosen dose used (300 lg/kg bw) in press). However, the development of renal cancer in this
was twice that estimated as the highest average daily intake in study makes it a suitable source for studying carcinogenic
the NTP study (NTP, 1989), in which the toxin was adminis- processes. The difference of cancer incidence between the two
tered by oral gavage. After rats reached 333 g, the dose was studies highlights the importance of the route of administration
held to 100 lg/rat, aimed to gradually reduce the two-fold for tumor outcomes.
difference in intake between the NTP and the present regime as OTA-modulated gene expression profiles were obtained in
the body weight further increased. This protocol was consid- both kidney, the toxin’s main target organ, and liver. Overall,
ered to better mimic the human intake pattern (Mantle et al., the effects observed were relatively modest. Real-time PCR
in press). Dosing adult rats according to body weight (as in the
data confirmed the results. The effect on abundance of certain
NTP study) would have resulted in an increasing exposure to
mRNAs correlated with the change in expression of respective
OTA (due to weight gain from fat accumulation), which would
not have reflected exposure of human adults. At a given proteins, suggesting that such OTA-induced changes could be
occurrence level of OTA, human intake is likely to be relatively of biological significance. The various clustering approaches
stable because of the limited expected age-dependent varia- showed the distinctive difference between kidney and liver
tions in body weight and food intake. The various selected time responses, consistent with the known organ selectivity of OTA
points represented early (7 and 21 days) and late (up to 12 carcinogenicity in the male rat. It is also possible that the
months) exposure stages. The treatment was well tolerated, and apparently higher susceptibility of kidney is due to a dose-
cumulative incidence of animals with a renal tumor of 25% had response effect from the elevated content and the prolonged
occurred by the 2-year end point of the experiment. This residence of OTA in that organ. Indeed, it is well documented
 TABLE 3
Transporter-Related Genes Modulated by OTA

Renal proximal
Accession no. Gene title Gene symbol Folda Fold 7D Fold 21D Fold 4M Fold 7M Fold 12M tubule Function Regulation Ref.

U15176 ATPase, Naþ/Kþ Atp1a4 4.3* 3.3 4.3 6.4* 4.6 3.6 basolateral Intracellular sodium PKC, ERK1/2 Khundmiri et al.,
and potassium balance 2005
AB004559 solute carrier Slc22a6/rOAT1 2.7* 2.7* 4.4* 3.6* 1.8 1.9 basolateral Endogenous and PKC, ERK1/2 Koepsell and Endou,
family 22, exogenous anions 2004
member 6 Ochratoxin A Tsuda et al., 1999
D79981 solute carrier Slc21a4/OAT-K1 2.5* 2.7 5.6* 3.0 1.2 1.8 apical Organic anions, Takeuchi et al., 2001
family 21, xenobiotics,
member 4 Ochratoxin A
L19031 solute carrier Slc21a1/Oatp1 2.5* 1.6 3.8* 2.9 3.9* 1.3 apical Organic anions PKC, Terlouw et al., 2003
family 21, Ochratoxin A testosterone Hagenbuch and
member 1 Meier, 2003

OCHRATOXIN A CARCINOGENICITY
AB005547 aquaporin 8 Aqp8 2.5* 4.6* 1.0 2.1 2.1 4.4* apical/ H2O, Urine Vasopressin, Verkman et al., 1996
basolateral concentration adenylate
cyclase
D13871 solute carrier Slc2a5/GLUT5 2.2* 1.5 2.6* 2.7* 2.4* 2.2* apical fructose transport Swissprot
family 2,
member 5
D86086 ATP-binding Abcc2, Mrp-2 2.1* 2.1 1.9 2.5* 1.9 2.2 apical Organic anions, Lee and Kim, 2004
cassette glutathione conjugated
AB013455 solute carrier Slc34a1, NaPi-2 1.8* 1.5 3.0* 1.9 1.6* 1.3* apical Na/Pi cotransport Dietary Pi Magagnin et al., 1993
family 34, intake and
member 1 parathyroid
hormone
D83044 solute carrier Slc22a2/Oct2 1.7* 1.0 3.0* 2.0 1.8* 1.4* basolateral Endogenous and >Male Koepsell and Endou,
family 22, exogenous cations 2004
member 2 Lee and Kim, 2004
M80804 solute carrier Slc3a1/NBAT 1.7* 1.5 2.4* 2.0* 1.6 1.2 apical Amino acid transport Wells and Hediger,
family 3, 1992
member 1
X67948 aquaporin Aqp1 1.7* 1.8 2.5* 1.5 1.5 1.4 apical/ H2O, Urine Vasopressin, Ma et al., 1998
basolateral concentration adenylate cyclase
U76379 solute carrier Slc22a1/OCT1A 1.5* 1.9* 1.3 1.7* 1.5 1.1 basolateral Organic cations, PKC/PKA, Koepsell and Endou,
family 22, anions, weak bases TK cGMP 2004
member 1 Lee and Kim, 2004
U28504 solute carrier Slc17a1/ NaPi-1 1.5* 1.3 2.3* 1.8 2.3* 1.5 apical Na/Pi co-transport Swissprot
family 17,
member 1
M81855 P-glycoprotein/ Mdr1/Pgy1 4.6* 1.0* 6.9 21.8* 6.5* 2.2 apical Removal of Lee and Kim, 2004
multidrug xenobiotics

127
resistance 1

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

128 MARIN-KUAN ET AL.

that OTA is actively transported in kidney cells (Gekle et al.,

Database (RGD)
2005; O’Brien and Dietrich, 2005; WHO, 2001).

Rat Genome
Ref.
Hierarchical clustering analysis resulted in the identification

Swissprot
Swissprot
of relatively complex and unexpected time-related responses.

RGD
For example, a cluster was noted between the early response (at
7 days) and the late response (at 12 months). Currently, this
Regulation

finding is difficult to interpret. Time-related differences may

occur as a result of toxicokinetics, treatment duration, and/or
age-specific factors. The experimental design and the statistical
Fatty acylcarnitine approach applied in the present study did not allow thorough
addressing of these factors.
Function

carrier

General Markers of Nephrotoxicity and Carcinogenicity

fructose

calcium
cations

Up-regulation of the prototypical renal toxicity marker

kidney injury molecule (KIM-1) was observed in treated
Renal proximal

animals. KIM-1 is a type-1 membrane protein highly expressed

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

mitochondria
tubule

both in early stage renal tubular injury and in regenerating

membrane
membrane

proximal tubule epithelial cells (Ichimura et al., 1998). Early

apical

induction of KIM-1 expression with a maximum at 21 days (21-

fold) shows that the OTA-dose in the present study produced
Average fold-change in gene expression from the five time points (7D and 21D days and 4M, 7M, and 12M months).
Fold 12M

some nephrotoxicity, even if not clinically apparent. From

3.2*

1.5
1.6
1.4

a mechanistic perspective, this finding may be relevant. It is

acknowledged that toxicity might induce tissue regeneration
TABLE 3—Continued

and cell proliferation, which may then promote tumor de-

Fold 7M

velopment. OTA-induced proliferation has been observed

1.6

1.2
1.6

1.5

in vivo (NTP, 1989) and in vitro (Hong et al., 2000; Kamp

et al., 2005). However, there was no histological evidence of
Fold 4M

renal cell proliferation at any time point analyzed in this study,

3.9*

1.9
1.2
1.7

apart from the consistent presence of karyomegalic nuclei (data

not shown). In the present study, genes implicated in cell
Fold 21D

Note. Underlined genes correspond to the genes under the regulation of HNF4a.

survival and proliferation were induced by OTA, including c-

6.7*

1.6
1.7
1.7

myc (Amati et al., 1998), Akt1 (Vanhaesebroeck and Alessi,

2000), and Plk (Holtrich et al., 1994). In addition, a strong up-
regulation of P-glycoprotein (P-gp) was observed. P-gp
Fold 7D

2.0*

belongs to the ATP-binding cassette (ABC) multidrug trans-

2.2
1.8
1.6

porter superfamily overexpressed in tumor cells (Thomas and

*Statistical significance represented by the alpha risk <0.001.

Coley, 2003).
Folda

1.7*
1.6*
3.0*

1.5*

Inhibition of protein synthesis is thought to be a major mode

of OTA toxicity, although the exact molecular mechanism
Slc2a2/GLUT2

Slc25a20/Cact
Gene symbol

involved is not known (Creppy et al., 1983). However, only few

differentially expressed genes relate directly to a known
Atp2a2

mechanism of protein synthesis inhibition. The prostaglandin

Atp9a

F2 receptor negative regulator (Ptgfrn; Orlicky, 1996) and the

eukaryotic translation initiation factor 4E binding protein 1
ATPase, class II
ATPase, Caþþ

(Eif4ebp1; Gingras et al., 1999) were up-regulated. Both are

member 20
solute carrier

solute carrier
Gene title

family 25,
member 2
family 2,

documented to be involved in negative regulation of protein

synthesis. In contrast, elongation factor 1-alpha 1 (eEF1A-1)
was down-regulated. This protein is known to promote the
GTP-dependent binding of aminoacyl-tRNA to the A-site of
Accession no.

ribosomes during protein biosynthesis (Petroulakis and Wang,

AA800120

2002). Although our data imply inhibition of protein synthesis,

U78977
L28135

J04024

comprehensive data on protein expression and modification

a

would be required for a more definite conclusion. For example,

 OCHRATOXIN A CARCINOGENICITY 129

Taqman RT-PCR Affymetrix

A 85 B 85
70 70
45 45
25 25

Fold-expression

Fold-expression
20 20
15 15
10 10
5 5
REG REG
0 0
Pgy1 KIM-1 c-Myc Akt1 Pgy1 KIM-1 c-Myc Akt1
-5 -5
-10 -10
-15 Genes -15 Genes

C Taqman RT-PCR D Affymetrix

HNF4a Cyp2c Agt rOAT1 Glut 5 HNF4a Cyp2c Agt rOAT1 Glut 5
0 0
-5 -5

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

-10 -10

Fold-expression
Fold-expression

-15 -15

-20 -20

-25 -25

-50 -50

-100 -100

-150 Genes -150 Genes

E Taqman RT-PCR F Affymetrix

OAT-K1 Oatp1 NaPi-2 OAT-K1 Oatp1 NaPi-2
0 0
-3 -3
-5 -5
Fold-expression

Fold-expression

-8 -8
-10 -10
-13 -13
-15 -15
-30 -30
-45 -45
Genes Genes

G Taqman RT-PCR H Affymetrix

Gclm Gclc Gstm3 Gstt2 Gstp UDP-G Gclm Gclc Gstm3 Gstt2 Gstp UDP-G
0 0

-2 -2
Fold-expression

Fold-expression

-4 -4

-6 -6

-8 -8

-10 Genes -10 Genes

FIG. 4. Quantitation of gene expression changes for selected kidney genes as measured by TaqMan and Affymetrix methods at three time points: 21 days
(black), 4 months (white), and 12 months (grey). The relative quantities of TaqMan RT-PCR products were normalized to the 18S rat housekeeping gene. TaqMan
RT-PCR data are presented as the fold expression values as compared to those of control samples from technical replicates from pooled kidney mRNA (n ¼ 3).
(A–B) Nephrotoxicity markers. (C–D) HNF4a-regulated genes. (E–F) Transport-related genes. (G–H) Nrf2-regulated genes.
130 MARIN-KUAN ET AL.

of calcium homeostasis in the mechanism of OTA action,

which deserves further investigation.
OTA was found to significantly reduce the expression of
a number of genes encoding transport proteins, such as those
involved in organic anion transport. Similar trends were
observed in an acute and high-dose experiment (Luhe et al.,
2003). Importantly, these results support previous physiolog-
ical data in vitro and in vivo showing that chronic treatment
with OTA resulted in an impairment of the secretion of organic
anions in the proximal tubules (Gekle and Silbernagl, 1994;
Gekle et al., 2005; Sauvant et al., 2005). In this context it is
interesting to note that angiotensinogen mRNA expression was
down-regulated in our study. This may explain the effects of
OTA on renal hemodynamics, which is thought to be mediated
by angiotensin II (Gekle et al., 2005).
The transport proteins Oat1, Oat-k1, and Oatp1 are well
documented to selectively transport OTA (Gekle and Silbernagl,

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

1994; Gekle et al., 2005; O’Brien and Dietrich, 2005; Zepnik
et al., 2003). In our study they were markedly down-regulated
at RNA and protein levels, indicating that OTA is likely to affect
its own toxicokinetics. In addition, the down-regulation of
transporters involved in the excretion of xenobiotic metabolites
(e.g., multidrug resistance-associated protein 2, mrp2) may
possibly result in interactions between OTA and other toxic
chemicals.
Recently, it has been proposed that gene expression profiling
can identify distinctive fingerprints or signatures to discrimi-
FIG. 5. Correlation between changes in gene and protein expression. nate between compounds that act either as directly genotoxic or
Kidney protein lysates from control and OTA-treated animals were analyzed by are genotoxic through indirect mechanisms (Dickinson et al.,
Western blots and visualized by chemiluminescence. Transferred membranes
2004; Hu et al., 2004). Although this method of classification
were probed with antibodies specific to the following proteins: Gstp, Gclc,
HNF4a, Oct2, Oat-k1, Oatp1 (cortex only), and Oatp1 (medulla only). Three still requires validation, a trend has been observed in which
controls and three samples from OTA-treated animals were analyzed at 21 days direct-acting genotoxins induce toxicity-associated pathways
and 12 months. followed by early modulation of specific genes related to DNA-
damage. Interestingly, the genes reported as potentially specific
for direct-acting genotoxins were not modulated in our study,
it is known that the activity of the Eif4ebp1 protein is which may strongly suggest an indirect genotoxic mechanism
dependent on its phosphorylation status (Gingras et al., 1999). for OTA.
One of the most striking effects obtained in the present study
was a strong down-regulation of regucalcin, also known as the
Disruption of Biological Pathways
senescence marker protein30 (Smp30). The maximum effect
was observed after 21 days of treatment. In kidney tubule cells, Analysis of DNA-binding sites revealed that many genes
regucalcin is known to play a role in the regulation of intra- down-regulated by OTA share the motif 5#-AGGTCA-3# as
cellular Ca2þ homeostasis. The reduction of regucalcin is a DNA-binding domain. This sequence is the regulatory motif
compatible with documented effects of OTA on intracellular through which the transcription factor HNF4a acts (Ellrott
calcium and on calcium signaling homeostasis (Sauvant et al., et al., 2002). This strongly suggests that OTA may disrupt
2005). In rat renal cortex, regucalcin was shown to produce HNF4a-dependent regulatory pathways. The exact mechanism
a suppressive effect on DNA synthesis (Xue et al., 2000; involved in the inhibition of HNF4a-regulated gene expression
Morooka and Yamaguchi, 2002). Interestingly, recent studies is not clear, although it was found that down-regulation of
have suggested that regucalcin may possess tumor suppression HNF4a expression was observed at both mRNA and protein
activity in hepatoma cells (Tsurusaki and Yamaguchi, 2003, levels.
2004). Inhibition of regucalcin expression was also observed From the perspective of cancer development, the biological
with other nephrotoxicant and nephrocarcinogenic compounds significance of down-regulation of HNF4a-regulated genes is
like cisplatin (Huang et al., 2001; Misawa and Yamaguchi, difficult to define. Many of the HNF4a target genes play im-
2001). Taken together, these data confirm a potential key role portant roles in development, differentiation, and homeostasis
 OCHRATOXIN A CARCINOGENICITY 131

(Sladek, 1993), and the depletion of HNF4a-pathway was tion factors. Previous investigations have provided evidence
previously associated with cancer development (Sel et al., that phase II enzymes may not be regulated exclusively by
1996). HNF4a-expression and activity was analyzed in human Nrf2-mediated mechanisms (Kang et al., 2003). For example,
renal carcinomas (Sel et al., 1996). The data indicated that in recent studies indicated that Nrf2 might not significantly
tumor samples both the expression and activity of HNF4a were contribute to the expression of Gsta1/2 in the mouse small
reduced as compared to normal tissue, suggesting that this intestine (McMahon et al., 2001). Our results show that some
pathway might be an important molecular mechanism in renal Nrf2-regulated genes are not down- but up-regulated, suggest-
carcinogenesis. However, the present analyses occurred well ing that OTA may also alter other signaling pathways that have
before any renal tumors were evident in the OTA-treated rats. not yet been identified.
In the kidney, OTA treatment resulted in an average 10-fold The toxicological consequences of the OTA-mediated down-
down-regulation of the expression of the male-specific, HNF4a- regulation of many Nrf2-regulated genes are likely to be
regulated cytochrome P450 CYP 2C11. CYP 2C11 is involved biologically significant. It is well documented that chemicals
in phase I metabolism of endogenous compounds (i.e., steroids) inducing an oxidative stress response through activation of
and several xenobiotics (Riddick et al., 2004). A potential role of Nrf2 are associated with chemoprotective properties (Chen and
CYP 2C11 in bioactivation of OTA was suggested (Pfohl- Kong, 2004; Zhang and Gordon, 2004). With OTA, the
Leszkowicz et al., 1998), but data indicating a strong down- inhibition of expression of genes involved in the biosynthesis
regulation of this gene do not support this hypothesis. of glutathione (including the rate-limiting Gclc) is likely to

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

OTA has been reported to induce oxidative stress. For result in a reduction of the cellular glutathione content, as
example, an increased formation of malondialdehyde (MDA) already documented in OTA-treated cell cultures (Schaaf et al.,
was observed in the kidney of rats exposed to 120 lg/kg bw/day 2002). Because of its major importance in the cellular redox-
of OTA for 60 days (Petrik et al., 2003). Other studies balance, a reduction of intracellular glutathione content should
confirmed lipid peroxidation as a result of OTA exposure compromise the cellular defense against oxidative damage. 4-
(Omar et al., 1990; Sauvant et al., 2005). Contrary to this, in Hydoxynonenal (4-HNE) is a reactive metabolite resulting
another study, treatment of male rats with OTA (up to 2.0 mg/kg from lipid peroxidation that forms DNA adducts (Hartley et al.,
bw administered by oral gavage) did not increase the formation 1995). Because of the affinity to conjugate 4-HNE with
of biomarkers of oxidative damage such as the lipid perox- glutathione, Gstp is thought to play a role in the detoxication
idation marker MDA in plasma, kidney, and liver, or the DNA of this lipid peroxidation product (Hartley et al., 1995). It can
damage marker 8-oxo-7,8-dihydro-2#deoxyguanosine in kid- be hypothesized that the OTA-mediated down-regulation of
ney DNA (Gautier et al., 2001b). However, a significant Gstp may impair the detoxication of 4-HNE and thus result in
increase in the expression of the marker of oxidative stress increased oxidative damage.
response HSP32 protein was observed in kidney but not in liver. Many Nrf2-regulated gene products down-regulated by OTA
Surprisingly, in the present study, many genes normally induced are involved in xenobiotic detoxication processes. They in-
by oxidative stress were down-regulated during OTA treatment. clude several glutathione S-transferases and UDP-glucuronyl-
This response was only observed in kidney, and the mechanism transferases. Their reduced expression may possibly result in
of this down-regulation is still unknown. Interestingly, these a higher concentration of xenobiotics, normally detoxified by
genes share the antioxidant regulatory element (ARE) as these enzymes, which could act synergistically with OTA, as
a regulatory motif in their promoter region. The ARE-motif is suggested for OTA and other mycotoxins (Creppy et al., 2004).
recognized by the nuclear factor-erythroid 2-related factor 2 Finally, an attempt to link HNF4a and Nrf2-pathways
(Nrf2), a member of the ‘‘Cap-n-Collar’’ family of basic-region pointed to protein kinase C (PKC) as a common regulatory
leucine zipper transcription factors (Mathers et al., 2004). Nrf2 feature. PKC is the major kinase involved in ARE-mediated
is involved in both the basal expression and the induction of gene induction by Nrf2 (Bloom and Jaiswal, 2003; Numazawa
genes encoding mainly for detoxication, cytoprotective, and et al., 2003) and in activation of the HNF4apathway
antioxidant enzymes (Lee and Johnson, 2004; Nguyen et al., (Hashimoto et al., 2005; Roy et al., 2001). Notably, in our
2004). These proteins are responsible for cellular redox status study several transport-related genes, reported to be regulated
and for cellular defense against oxidative damage. We are through PKC-dependent processes, were also down-regulated
currently studying inhibition of Nrf2-activation by OTA as by OTA (see Table 2). These data strongly suggest that
a molecular mechanism involved in OTA toxicity. disruption of PKC-mediated processes may play a role in the
Although the expression of most of the renal Nrf2-regulated toxicity of OTA (Fig. 6). Further investigation is necessary to
genes seemed to be inhibited by OTA treatment, a limited confirm this hypothesis.
number of them appeared to be induced (e.g., Gsta2). It is
important to note that, for many genes, the 5# promoter region
Summary and Conclusion
may contain multiple binding sites specific for several tran-
scription factors, and therefore the observed effects on mRNA In the present study significantly different gene expression
expression may be the combined effects on different transcrip- profiles were observed in kidney and liver of male Fischer rats
132 MARIN-KUAN ET AL.

Nrf2
Detoxication enzymes
OTA GCS
Nrf2 GCLC Inhibition of oxidative
A/G TGA C/T NNN GC A/G GST stress response
PKC AFAR
? ARE
Growth factors? HNF4α Malignant
Toxicity
Fatty acids tumors
Cell metabolism ?
? REG AGGTCA
(Ca2+) Cyp2c11
Lipid metabolism Disruption of
Nucleolus Cell transporters cell homeostasis
?

FIG. 6. Role of PKC in the response of kidney cells to OTA. Putative scheme of plausible mechanism of OTA toxicity, assembled from the gene expression
data. Effect of OTA on potential signaling molecules (growth factors, fatty acids, and/or Ca2þ) could disrupt PKC-regulated pathways downstream. Down-
regulation of genes under transcriptional control of Nrf2 may lead to a reduced oxidative stress response. In addition, OTA-induced down-regulation of genes under
HNF4a-control may affect key metabolic processes. This could make kidney cells more vulnerable to OTA-induced toxicity leading to tumor development. Dashed
lines: hypothetical effects (? inhibition). Solid lines: OTA-effects demonstrated in this study. Down-regulation is denoted by Y.

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

following chronic dietary exposure to OTA in a concentration authors was supported by the EU-Grant # QLK1-CT-2001–011614. Conflict of
sufficient to cause renal carcinoma late in life. Gene expression interest: none declared.
data indicated occurrence of renal toxicity, which may involve
oxidative damage and eventually lead to cell proliferation and
cancer. Gene expression analysis pointed toward predominant
involvement of epigenetic mechanisms in OTA carcinogenic- REFERENCES
ity. These mechanisms may include alteration of calcium
Alter, O., Brown, P. O., and Botstein, D. (2003). Generalized singular value
homeostasis and disruption of pathways regulated by the decomposition for comparative analysis of genome-scale expression datasets
transcription factors HNF4a and Nrf2. The depletion of of two different organisms. Proc. Natl. Acad. Sci. U.S.A. 100, 3351–3356.
Nrf2-regulated enzymes is likely to impair the cellular defense Amati, B., Alevizopoulos, K., and Vlach, J. (1998). Myc and the cell cycle.
potential, resulting in chronic elevation of oxidative stress in Front. Biosci. 3, d250–d268.
the kidney. Sources of oxidants can be either endogenous Bendele, A. M., Carlton, W. W., Krogh, P., and Lillehoj, E. B. (1985).
(normal physiological processes) or exogenous (xenobiotics). Ochratoxin A carcinogenesis in the (C57BL/6J3C3H)F1 mouse. J. Natl.
Low concentration of antioxidants can result in cell pro- Cancer Inst. 75, 733–742.
liferation and, ultimately, cancer development (Klaunig and Bloom, D. A., and Jaiswal, A. K. (2003). Phosphorylation of Nrf2 at Ser40 by
protein kinase C in response to antioxidants leads to the release of Nrf2 from
Kamendulis, 2004). Therefore, inhibition of cellular defense INrf2, but is not required for Nrf2 stabilization/accumulation in the nucleus
appears to be a highly plausible novel mechanism, which could and transcriptional activation of antioxidant response element-mediated
contribute to OTA-mediated carcinogenicity. However, it has to NAD(P)H:quinone oxidoreductase-1 gene expression. J. Biol. Chem. 278,
be acknowledged that the present data do not provide any 44675–44682.
absolute demonstration of a direct link between the gene Castegnaro, M., Mohr, U., Pfohl-Leszkowicz, A., Esteve, J., Steinmann, J.,
expression effects observed and the late development of Tillmann, T., Michelon, J., and Bartsch, H. (1998). Sex- and strain-specific
tumors. Further investigations are necessary to further establish induction of renal tumors by ochratoxin A in rats correlates with DNA
adduction. Int. J. Cancer 77, 70–75.
the relevance of the biological effects to the carcinogenic
Chen, C., and Kong, A. N. (2004). Dietary chemopreventive compounds and
process. In addition, it has to be noted that the interpretation of ARE/EpRE signaling. Free Radic. Biol. Med. 36, 1505–1516.
gene expression data in this study has focused on significantly Creppy, E. E., Stormer, F. C., Kern, D., Roschenthaler, R., and Dirheimer, G.
modulated genes. It is, of course, feasible that subtle changes in (1983). Effects of ochratoxin A metabolites on yeast phenylalanyl-tRNA
the expression of other genes, together with cell signaling and synthetase and on the growth and in vivo protein synthesis of hepatoma cells.
post-translational modification processes, which were not Chem. Biol. Interact. 47, 239–247.
further analyzed in this study, are also implicated in OTA Creppy, E. E., Chiarappa, P., Baudrimont, I., Borracci, P., Moukha, S., and
carcinogenesis and could help to accomplish the comprehen- Carratu, M. R. (2004). Synergistic effects of fumonisin B1 and ochratoxin A:
Are in vitro cytotoxicity data predictive of in vivo acute toxicity? Toxicology
sion of the OTA mechanism of action.
201, 115–123.
Dickinson, D. A., Warnes, G. R., Quievryn, G., Messer, J., Zhitkovich, A.,
ACKNOWLEDGMENTS Rubitski, E., and Aubrecht, J. (2004). Differentiation of DNA reactive and
non-reactive genotoxic mechanisms using gene expression profile analysis.
We would like to thank the Central Biomedical Services at Imperial College Mutat. Res. 549, 29–41.
London for their technical help, Prof. B. Hagenbuch, Dr. Leslie McLellan, and Dirheimer, G., and Creppy, E. E. (1991). Mechanism of action of ochratoxin A.
Dr. S. Lariani for their advice and contributions to our studies. The work of the IARC Sci. Publ. 115, 171–186.
 OCHRATOXIN A CARCINOGENICITY 133

Ellrott, K., Yang, C., Sladek, F. M., and Jiang, T. (2002). Identifying and apoptosis in mammalian cell lines and primary cells. Toxicology 206,
transcription factor binding sites through Markov chain optimization. 413–425.
Bioinformatics. 18(Suppl. 2), S100–S109. Kang, K. W., Park, E. Y., and Kim, S. G. (2003). Activation of CCAAT/
Faucet, V., Pfohl-Leszkowicz, A., Dai, J., Castegnaro, M., and Manderville, enhancer-binding protein beta by 2#-amino-3#-methoxyflavone (PD98059)
R. A. (2004). Evidence for covalent DNA adduction by ochratoxin A leads to the induction of glutathione S-transferase A2. Carcinogenesis 24,
following chronic exposure to rat and subacute exposure to pig. Chem. Res. 475–482.
Toxicol. 17, 1289–1296. Khundmiri, S. J., Dean, W. L., McLeish, K. R., and Lederer, E. D. (2005).
Gautier, J. C., Holzhaeuser, D., Markovic, J., Gremaud, E., Schilter, B., and Parathyroid hormone-mediated regulation of Naþ-Kþ-ATPase requires
Turesky, R. J. (2001a). Oxidative damage and stress response from ERK-dependent translocation of protein kinase Calpha. J. Biol. Chem.
ochratoxin A exposure in rats. Free Radic. Biol. Med. 30, 1089–1098. 280, 8705–8713.
Gautier, J., Richoz, J., Welti, D. H., Markovic, J., Gremaud, E., Guengerich, Klaunig, J. E., and Kamendulis, L. M. (2004). The role of oxidative stress in
F. P., and Turesky, R. J. (2001b). Metabolism of ochratoxin A: Absence of carcinogenesis. Annu. Rev. Pharmacol. Toxicol. 44, 239–267.
formation of genotoxic derivatives by human and rat enzymes. Chem. Res. Koepsell, H., and Endou, H. (2004). The SLC22 drug transporter family.
Toxicol. 14, 34–45. Pflugers Arch. 447, 666–676.
Gekle, M., Sauvant, C., and Schwerdt, G. (2005). Ochratoxin A at nanomolar Lee, J. M., and Johnson, J. A. (2004). An important role of Nrf2-ARE pathway
concentrations: A signal modulator in renal cells. Mol. Nutr. Food Res. 49, in the cellular defense mechanism. J. Biochem. Mol. Biol. 37, 139–143.
118–130.
Lee, W., and Kim, R. B. (2004). Transporters and renal drug elimination. Annu.
Gekle, M., and Silbernagl, S. (1994). The role of the proximal tubule in
Rev. Pharmacol. Toxicol. 44, 137–166.
ochratoxin A nephrotoxicity in vivo: Toxodynamic and toxokinetic aspects.

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

Luhe, A., Hildebrand, H., Bach, U., Dingermann, T., and Ahr, H. J. (2003). A
Ren. Physiol. Biochem. 17, 40–49.
new approach to studying ochratoxin A (OTA)-induced nephrotoxicity:
Gingras, A. C., Gygi, S. P., Raught, B., Polakiewicz, R. D., Abraham, R. T.,
Expression profiling in vivo and in vitro employing cDNA microarrays.
Hoekstra, M. F., Aebersold, R., and Sonenberg, N. (1999). Regulation of
Toxicol. Sci. 73, 315–328.
4E-BP1 phosphorylation: A novel two-step mechanism. Genes Dev. 13,
1422–1437. Ma, T., Yang, B., Gillespie, A., Carlson, E. J., Epstein, C. J., and Verkman, A. S.
(1998). Severely impaired urinary concentrating ability in transgenic mice
Gross-Steinmeyer, K., Weymann, J., Hege, H. G., and Metzler, M. (2002).
lacking aquaporin-1 water channels. J. Biol. Chem. 273, 4296–4299.
Metabolism and lack of DNA reactivity of the mycotoxin ochratoxin A in
cultured rat and human primary hepatocytes. J. Agric. Food Chem. 50, 938–945. Magagnin, S., Werner, A., Markovich, D., Sorribas, V., Stange, G., Biber, J.,
and Murer, H. (1993). Expression cloning of human and rat renal cortex Na/
Hagenbuch, B., and Meier, P. J. (2003). The superfamily of organic anion
Pi cotransport. Proc. Natl. Acad. Sci. U.S.A 90, 5979–5983.
transporting polypeptides. Biochim. Biophys. Acta 1609, 1–18.
Manderville, R. A. (2005). A case for the genotoxicity of ochratoxin A by
Harris, J. P., and Mantle, P. G. (2001). Biosynthesis of ochratoxins by
bioactivation and covalent DNA adduction. Chem. Res. Toxicol. 18, 1091–1097.
Aspergillus ochraceus. Phytochemistry 58, 709–716.
Mally, A., Zepnik, H., Wanek, P., Eder, E., Dingley, K., Ihmels, H., Volkel, W.,
Hartley, D. P., Ruth, J. A., and Petersen, D. R. (1995). The hepatocellular
and Dekant, W. (2004). Ochratoxin A: Lack of formation of covalent DNA
metabolism of 4-hydroxynonenal by alcohol dehydrogenase, aldehyde
dehydrogenase, and glutathione S-transferase. Arch. Biochem. Biophys. adducts. Chem. Res. Toxicol. 17, 234–242.
316, 197–205. Mantle, P., Kulinskaya, E., and Nestler, S., (2005). Renal tumourigenesis in
Hashimoto, N., Kido, Y., Uchida, T., Matsuda, T., Suzuki, K., Inoue, H., male rats in response to chronic dietary ochratoxin A. Food Add. Contam.
Matsumoto, M., Ogawa, W., Maeda, S., Fujihara, H., et al. (2005). (in press).
PKClambda regulates glucose-induced insulin secretion through modulation Mansourian, R., Mutch, D. M., Antille, N., Aubert, J., Fogel, P., Le Goff, J. M.,
of gene expression in pancreatic beta cells. J. Clin. Invest 115, 138–145. Moulin, J., Petrov, A., Rytz, A., Voegel, J. J., et al. (2004). The Global Error
Holtrich, U., Wolf, G., Brauninger, A., Karn, T., Bohme, B., Rubsamen- Assessment (GEA) model for the selection of differentially expressed genes
Waigmann, H., and Strebhardt, K. (1994). Induction and down-regulation of in microarray data. Bioinformatics 20, 2726–2737.
PLK, a human serine/threonine kinase expressed in proliferating cells and Mathers, J., Fraser, J. A., McMahon, M., Saunders, R. D., Hayes, J. D., and
tumors. Proc. Natl. Acad. Sci. U.S.A 91, 1736–1740. McLellan, L. I. (2004). Antioxidant and cytoprotective responses to redox
Hong, J. T., Park, K. L., Han, S. Y., Park, K. S., Kim, H. S., Oh, S. D., Lee, stress. Biochem. Soc. Symp. 71, 157–176.
R. D., and Jang, S. J. (2000). Effects of ochratoxin A on cytotoxicity and cell Matter, H. (1997). Selecting optimally diverse compounds from structure
differentiation in cultured rat embryonic cells. J. Toxicol. Environ. Health A databases: A validation study of two-dimensional and three-dimensional
61, 609–621. molecular descriptors. J. Med. Chem. 40, 1219–1229.
Hu, T., Gibson, D. P., Carr, G. J., Torontali, S. M., Tiesman, J. P., Chaney, J. G., McMahon, M., Itoh, K., Yamamoto, M., Chanas, S. A., Henderson, C. J.,
and Aardema, M. J. (2004). Identification of a gene expression profile that McLellan, L. I., Wolf, C. R., Cavin, C., and Hayes, J. D. (2001). The
discriminates indirect-acting genotoxins from direct-acting genotoxins. Cap‘n’Collar basic leucine zipper transcription factor Nrf2 (NF-E2 p45-
Mutat. Res. 549, 5–27. related factor 2) controls both constitutive and inducible expression of
Huang, Q., Dunn, R. T., Jayadev, S., DiSorbo, O., Pack, F. D., Farr, S. B., Stoll, intestinal detoxification and glutathione biosynthetic enzymes. Cancer Res.
R. E., and Blanchard, K. T. (2001). Assessment of cisplatin-induced 61, 3299–3307.
nephrotoxicity by microarray technology. Toxicol. Sci. 63, 196–207. Miljkovic, A., Pfohl-Leszkowicz, A., Dobrota, M., and Mantle, P. G. (2003).
Ichimura, T., Bonventre, J. V., Bailly, V., Wei, H., Hession, C. A., Cate, R. L., Comparative responses to mode of oral administration and dose of
and Sanicola, M. (1998). Kidney injury molecule-1 (KIM-1), a putative ochratoxin A or nephrotoxic extract of Penicillium polonicum in rats, Exp.
epithelial cell adhesion molecule containing a novel immunoglobulin Toxicol. Pathol. 54, 305–312.
domain, is up-regulated in renal cells after injury. J. Biol. Chem. 273, Misawa, H., and Yamaguchi, M. (2001). Involvement of nuclear factor-1 (NF1)
4135–4142. binding motif in the regucalcin gene expression of rat kidney cortex: The
Kamp, H. G., Eisenbrand, G., Schlatter, J., Wurth, K., and Janzowski, C. expression is suppressed by cisplatin administration. Mol. Cell Biochem.
(2005). Ochratoxin A: Induction of (oxidative) DNA damage, cytotoxicity 219, 29–37.
134 MARIN-KUAN ET AL.

Morooka, Y., and Yamaguchi, M. (2002). Endogenous regucalcin suppresses Sauvant, C., Holzinger, H., and Gekle, M. (2005). The nephrotoxin ochratoxin
the enhancement of protein phosphatase activity in the cytosol and nucleus A induces key parameters of chronic interstitial nephropathy in renal
of kidney cortex in calcium-administered rats. J. Cell Biochem. 85, 553–560. proximal tubular cells. Cell. Physiol. Biochem. 15, 125–134.
Mutch, D. M., Anderle, P., Fiaux, M., Mansourian, R., Vidal, K., Wahli, W., Schaaf, G. J., Nijmeijer, S. M., Maas, R. F., Roestenberg, P., de Groene, E. M.,
Williamson, G., and Roberts, M. A. (2004). Regional variations in ABC and Fink-Gremmels, J. (2002). The role of oxidative stress in the ochratoxin
transporter expression along the mouse intestinal tract. Physiol. Genomics A-mediated toxicity in proximal tubular cells. Biochim. Biophys. Acta 1588,
17, 11–20. 149–158.
Nguyen, T., Yang, C. S., and Pickett, C. B. (2004). The pathways and molecular Sel, S., Ebert, T., Ryffel, G. U., and Drewes, T. (1996). Human renal cell
mechanisms regulating Nrf2 activation in response to chemical stress. Free carcinogenesis is accompanied by a coordinate loss of the tissue specific
Radic. Biol. Med. 37, 433–441. transcription factors HNF4 alpha and HNF1 alpha. Cancer Lett. 101,
205–210.
NTP (1989). Toxicology and Carcinogenesis Studies of Ochratoxin A (CAS
No. 303–47–9) in F344/N Rats (Gavage Studies). Natl. Toxicol. Program Sladek, F. M. (1993). Orphan receptor HNF-4 and liver-specific gene
Tech. Rep. Ser. 358, 1–142. expression. Receptor 3, 223–232.
Numazawa, S., Ishikawa, M., Yoshida, A., Tanaka, S., and Yoshida, T. (2003). Takeuchi, A., Masuda, S., Saito, H., Abe, T., and Inui, K. (2001). Multispecific
Atypical protein kinase C mediates activation of NF-E2-related factor 2 in substrate recognition of kidney-specific organic anion transporters OAT-K1
and OAT-K2. J. Pharmacol. Exp. Ther. 299, 261–267.
response to oxidative stress. Am. J. Physiol Cell Physiol. 285, C334–C342.
Terlouw, S. A., Masereeuw, R., and Russel, F. G. (2003). Modulatory effects of
O’Brien, E., and Dietrich, D. R. (2005). Ochratoxin A: The continuing enigma.
hormones, drugs, and toxic events on renal organic anion transport. Biochem.
Crit. Rev. Toxicol. 35, 33–60.
Pharmacol. 65, 1393–1405.

Downloaded from toxsci.oxfordjournals.org by guest on December 8, 2010

Orlicky, D. J. (1996). Negative regulatory activity of a prostaglandin F2 alpha
Thomas, H., and Coley, H. M. (2003). Overcoming multidrug resistance in
receptor associated protein (FPRP). Prostaglandins Leukot. Essent. Fatty
cancer: An update on the clinical strategy of inhibiting p-glycoprotein.
Acids 54, 247–259.
Cancer Control 10, 159–165.
Omar, R. F., Hasinoff, B. B., Mejilla, F., and Rahimtula, A. D. (1990).
Tsuda, M., Sekine, T., Takeda, M., Cha, S. H., Kanai, Y., Kimura, M., and
Mechanism of ochratoxin A stimulated lipid peroxidation. Biochem.
Endou, H. (1999). Transport of ochratoxin A by renal multispecific organic
Pharmacol. 40, 1183–1191.
anion transporter 1. J. Pharmacol. Exp. Ther. 289, 1301–1305.
Petkova-Bocharova, T., Chernozemsky, I. N., and Castegnaro, M. (1988).
Tsurusaki, Y., and Yamaguchi, M. (2003). Overexpression of regucalcin
Ochratoxin A in human blood in relation to Balkan endemic nephropathy and modulates tumor-related gene expression in cloned rat hepatoma H4-II-E
urinary system tumours in Bulgaria. Food Addit. Contam. 5, 299–301. cells. J. Cell. Biochem. 90, 619–626.
Petrik, J., Zanic-Grubisic, T., Barisic, K., Pepeljnjak, S., Radic, B., Ferencic, Z., Tsurusaki, Y., and Yamaguchi, M. (2004). Role of regucalcin in liver nuclear
and Cepelak, I. (2003). Apoptosis and oxidative stress induced by ochratoxin function: Binding of regucalcin to nuclear protein or DNA and modulation of
A in rat kidney. Arch. Toxicol. 77, 685–693. tumor-related gene expression. Int. J. Mol. Med. 14, 277–281.
Petroulakis, E., and Wang, E. (2002). Nerve growth factor specifically Turesky, R. J. (2005). Perspective: Ochratoxin A is not a genotoxic carcinogen.
stimulates translation of eukaryotic elongation factor 1A-1 (eEF1A-1) Chem. Res. Toxicol. 18, 1082–1090.
mRNA by recruitment to polyribosomes in PC12 cells. J. Biol. Chem. 277, Vanhaesebroeck, B., and Alessi, D. R. (2000). The PI3K-PDK1 connection:
18718–18727. More than just a road to PKB. Biochem. J. 346, 561–576.
Pfohl-Leszkowicz, A., Grosse, Y., Kane, A., Creppy, E. E., and Dirheimer, G. Verkman, A. S., van Hoek, A. N., Ma, T., Frigeri, A., Skach, W. R., Mitra, A.,
(1993). Differential DNA adduct formation and disappearance in three Tamarappoo, B. K., and Farinas, J. (1996). Water transport across
mouse tissues after treatment with the mycotoxin ochratoxin A. Mutat. Res. mammalian cell membranes. Am. J. Physiol. 270, C12–C30.
289, 265–273. Wells, R. G., and Hediger, M. A. (1992). Cloning of a rat kidney cDNA that
Pfohl-Leszkowicz, A., Pinelli, E., Bartsch, H., Mohr, U., and Castegnaro, M. stimulates dibasic and neutral amino acid transport and has sequence
(1998). Sex- and strain-specific expression of cytochrome P450s in ochratoxin similarity to glucosidases. Proc. Natl. Acad. Sci. U.S.A 89, 5596–5600.
A-induced genotoxicity and carcinogenicity in rats. Mol. Carcinog. 23, 76–85. WHO (2001). Safety Evaluation of certain mycotoxins in food. Fifty-sixth
Plestina, R., Ceovic, S., Gatenbeck, S., Habazin-Novak, V., Hult, K., Hokby, E., meetingt of the Joint FAO/WHO Expert Committee on Food Additives.
Krogh, P., and Radic, B. (1990). Human exposure to ochratoxin A in areas of WHO Food Additives Series 47, i-706.
Yugoslavia with endemic nephropathy. J. Environ. Pathol. Toxicol. Oncol. Xue, J. H., Takahashi, H., and Yamaguchi, M. (2000). Stimulatory effect of
10, 145–148. regucalcin on mitochondrial ATP-dependent calcium uptake activity in rat
Riddick, D. S., Lee, C., Bhathena, A., Timsit, Y. E., Cheng, P. Y., Morgan, E. T., kidney cortex. J. Cell Biochem. 80, 285–292.
Prough, R. A., Ripp, S. L., Miller, K. K., Jahan, A., et al. (2004). Zepnik, H., Volkel, W., and Dekant, W. (2003). Toxicokinetics of the
Transcriptional suppression of cytochrome P450 genes by endogenous and mycotoxin ochratoxin A in F 344 rats after oral administration. Toxicol.
exogenous chemicals. Drug Metab. Dispos. 32, 367–375. Appl. Pharmacol. 192, 36–44.
Roy, N., Guruprasad, M. R., Kondaiah, P., Mann, E. A., Giannella, R. A., and Zhang, Y., and Gordon, G. B. (2004). A strategy for cancer prevention:
Visweswariah, S. S. (2001). Protein kinase C regulates transcription of the Stimulation of the Nrf2-ARE signaling pathway. Mol. Cancer Ther. 3,
human guanylate cyclase C gene. Eur. J. Biochem. 268, 2160–2171. 885–893.