A New Approach to Studying Ochratoxin A (OTA)-Induced Nephrotoxicity: Expression Profiling in Vivo and in Vitro Employing cDNA Microarrays

Anke Lühe*,1, Heinz Hildebrand{dagger}, Ute Bach{ddagger}, Theodor Dingermann§ and Hans-Jürgen Ahr{dagger}

* Department of Molecular and Genetic Toxicology, {dagger} Department of Molecular and Genetic Toxicology, and {ddagger} Department of Pathology, Bayer AG, Aprather Weg 18a, 42096 Wuppertal, Germany; and § Institute of Pharmaceutical Biology, Johann Wolfgang Goethe Universität, Marie-Curie-Str. 9, 60439 Frankfurt/Main, Germany

Received November 12, 2002; accepted February 20, 2003

ABSTRACT

Ochratoxin A (OTA) is a mycotoxin often found in cereals as a contaminant, and it is known to cause severe nephrotoxicity in animals and humans. There have been several investigations studying the mode of action of this toxicant, suggesting inhibition of protein synthesis, formation of DNA adducts, and provocation of DNA single-strand breaks as a result of oxidative stress, but little is known about the transcriptional alterations underlying OTA-derived nephrotoxicity so far. We carried out DNA microarray analyses to assess OTA-specific expression profiles in vivo and in vitro. Cultures of primary rat proximal tubular cells and male Wistar rats were treated with a low dose (5 µM and 1 mg/kg, respectively) or a high dose (12.5 µM and 10 mg/kg, respectively) of OTA for 24 or 72 h. Microarray experiments were carried out after dual fluorescent labeling of sample cDNA, and data analysis was performed utilizing different statistical methods. Validity of selected microarray data was confirmed by quantitative real-time PCR. We were able to demonstrate that microarray data derived from our proximal tubule cell (PTC) culture model were highly comparable to the in vivo situation. Marked treatment-specific transcriptional changes were detected for genes involved in DNA damage response and apoptosis (upregulation of GADD 153, GADD 45, annexin V), response to oxidative stress (differential expression of hypoxia-inducible factor 1 and catalase), and inflammatory reactions (upregulation of alpha 2 macroglobulin, ceruloplasmin, and cathepsin S). We conclude that our results provide a molecular basis for interpretation of OTA-induced nephrotoxicity.

Key Words: toxicogenomics; microarray; expression profiling; Ochratoxin A; nephrotoxicity; kidney; proximal tubule; cell culture.

Gene expression analysis employing DNA microarrays is considered a powerful tool for studying gene expressin of thousands of genes simultaneously in one sample. Techniques, equipment, and data analysis software tools have been developed rapidly during the last few years. Microarray analyses are utilized in a variety of research areas such as investigation of gene networks in physiological pathways, analysis of genetic variation, or identification of new therapeutic drug targets. The term toxicogenomics describes another exciting field of microarray application, which offers the possibility to characterize transcriptional responses towards toxic compounds. Toxicogenomic approaches comprise the assessment of gene expression profiles of animal tissues or cell cultures after administration of known toxicants, which may lead to the establishment of toxicant-specific "fingerprints." Databases containing these fingerprint profiles may then serve as a basis for the prediction of the potential toxicity of new compounds (Thomas et al., 2001Go). Another challenge provided by toxicogenomic-based investigations is the possibility to assess the toxic mode of action of a compound. Interpretation of transcriptional alterations and assignment of differentially expressed genes to biological pathways might culminate in the generation of hypotheses about the toxicological mechanism of compounds (Burczynski et al., 2000Go).

Ochratoxin A (OTA) is the major component of a group of secondary metabolites produced by several fungi such as aspergillus ochraeus or penicillium verrucosum. The mycotoxin is often found as a contaminant in grains or in other plant products such as red wine, coffee beans, nuts, and several spices (Walker, 2002Go). OTA is regarded as a major causal determinant of mycotoxin porcine nephropathy as well as endemic balkan nephropathy in humans (Hald, 1991Go; Stoev, 1998Go). Both diseases are characterized by degeneration of epithelial cells of the proximal tubules and interstitial fibrosis resulting in polyuria and various changes in haematological and biochemical parameters (Stoev et al., 2001Go). With regards to the chemical structure, the OTA molecule represents a phenylalanine-dihydroisocoumarine derivative, which is very stable to both temperature and hydrolysis. Due to its structural analogy to the amino acid phenylalanine, the toxin is able to competitively inhibit tRNA phenylalanine synthetases and as a consequence protein synthesis is interrupted (Dirheimer and Creppy, 1991Go). Involvement of OTA in the development of different types of cancers such as renal adenocarcinomas, bladder carcinomas, testicular cancers, and liver tumors have been reported in rats, mice, and also in humans (Castegnaro et al., 1998Go; Huff, 1991Go; Schwartz, 2002Go). Moreover, there is plenty of evidence for genotoxic effects of OTA (Dopp et al., 1999Go; Pfohl-Leszkowicz et al., 1991Go). Teratotoxicity and immunotoxicity have also been reported after exposure to OTA (Thuvander et al., 1996Go; Wei and Sulik, 1993Go). The mechanisms underlying these various toxic effects of OTA still have not been uncovered in detail. DNA adduct formation and single strand breaks observed in several studies are considered the most predominant cause of OTA-derived genotoxicity and carcinogenesis (Obrecht-Pflumio and Dirheimer, 2000Go; Pfohl-Leszkowicz et al., 1991Go). It has been discussed that these effects might be due to the generation of reactive oxygen species (ROS; Gillman et al., 1999Go). Observation of lipid peroxidation and prevention of OTA toxicity after sc injection of superoxide dismutase also provide evidence for the induction of oxidative stress by OTA (Baudrimont et al., 1994Go; Omar et al., 1990Go). It is a known fact that these findings contribute predominantly to the toxicity of OTA, but there is still little knowledge about the underlying transcriptional deregulations. Meisner and Cimbala (1986)Go studied changes in the abundance of phosphoenolpyruvate carboxykinase (PEPCK) mRNA after administration of OTA to rats and found a 50% decrease in PEPCK mRNA abundance compared to control rats. They also observed changes in the expression of several other mRNAs, but these clones have not been restriction mapped and the function of the encoded proteins was unknown.

The kidney is the main target organ for OTA toxicity. The mammalian kidney plays an important role as an elimination system. A vast majority of metabolic products of the organism as well as many pharmacological compounds are excreted via the urine. Within the complex and heterogeneous structure of the kidney the proximal tubule is one of the main targets for nephrotoxic compounds such as OTA. This is due to the fact that most of the active transport and biotransformation processes are located in the proximal tubule. We utilized primary rat proximal tubular cells as an in vitro model for our studies. Especially under the aspect of the "3Rs-rule" of animal welfare (reduction, replacement, and refinement) it is necessary to have a reliable and reproducible cell culture model that appropriately reflects the in vivo situation. In vitro models offer the opportunity to perform gene expression studies with small amounts of the investigated compound and furthermore, in vitro experiments can be extended to human proximal tubule cells. It is of outstanding importance to investigate the capacities and the limitations of a cell culture model for proper gene expression analyses in toxicology to get an idea of how well the in vivo situation could be reflected in vitro.

These considerations lead to the aims of our study. We wanted to assess the suitability of a proximal tubular cell culture model for performing gene expression analyses after application of nephrotoxicants and compare the results with expression profiles obtained from animal studies. In the current study we investigated the changes in gene expression in rat kidney cortices and in rat proximal tubular cells in vitro after short- and long-term exposure to low and high doses of OTA. We employed cDNA microarrays containing approximately 450 rat gene sequences and 250 EST sequences with unknown function in order to generate expression profiles in vivo and in vitro, which might serve as a basis for the mechanistic understanding of OTA toxicity on the transcriptional level. By employing innovative bioinformatic tools a huge amount of data has been analyzed and will be available for future comparisons.

MATERIALS AND METHODS

Proximal tubular cell culture.
Isolation of rat proximal tubule cells (PTC) has been described in detail elsewhere (Boom et al., 1992Go), so the procedure will only be mentioned briefly. Male Wistar rats (strain HsdCpb:Wu; ~180 g) obtained from Harlan-Winkelmann (Borchen, Germany) were anesthetized by an ip injection of sodium pentobarbital (50 mg/kg body weight [bw]), and the kidneys were in situ perfused with Ca2+-free HBSS (Hank’s balanced salt solution, Invitrogen, Karlsruhe, Germany) containing 0.5 mM EGTA (Sigma, Deisenhofen, Germany). After digestion in a 0.08% collagenase IV-HBSS solution containing 4 mM CaCl2 and 1% penicillin/streptomycin (all purchased from Sigma), cortical sections of the kidneys were separated and minced, and proximal tubular cells were isolated mechanically by sequential filtration through gauzes of 135 µm and 63 µm, respectively. The isolated PTCs were washed, pelleted, and resuspended in culture medium (DMEM/F12 1:1, Invitrogen), supplemented with 5% FCS (PAA, Linz, Austria), 1% penicillin/streptomycin, insulin (5 µg/ml), hydrocortisone (0.5 µg/ml), transferrin (10 µg/ml), human epidermal growth factor (0.01 µg/ml), epinephrine (0.5 µg/ml), and triiodothyronine (6.5 ng/ml; all part of REGM singlequots, Clonetics, San Diego, CA), and seeded into collagen I-coated 24-well culture plates (Beckton Dickinson, Heidelberg, Germany) at a density of approximately 0.5 x 105 cells per well. The cells remained undisturbed at 37°C in a 5% CO2 atmosphere for 120 h and were then washed with phosphate-buffered saline (PBS; Invitrogen). Cells reached confluency after 168 h of cultivation. Morphology of the cells was examined every second day by light microscopy. Identity of the proximal tubular cells has been confirmed by staining with antibodies against specific proximal tubular antigens such as Na+/H+ exchanger NHE3 purchased from Chemicon (Temecula, CA; Biemesderfer et al., 1998Go) and N-cadherin (A-CAM) purchased from BD Transduction Laboratories (Lexington, KY; Nouwen et al., 1993Go). Staining with E-cadherin (L-CAM) antibody was carried out to confirm absense of other kidney cell types, such as distal tubular cells, collecting duct cells and glomerulum-derived cells (Nouwen et al., 1993Go). Treatment of the cells with 5 µM and 12.5 µM of OTA dissolved in DMSO (Sigma) and diluted in culture medium started after 168 h of cultivation. OTA-containing medium was exchanged daily. Cells were harvested after 24 and 72 h of OTA treatment by lysis with RLT-Buffer (part of RNeasy Midi Kit for RNA Isolation, Qiagen, Hilden, Germany) containing 1% ß-Mercaptoethanol (Sigma) and stored at -80°C until isolation of total RNA.

Animal treatment.
Male Wistar rats (strain Hsd Cpb:Wu; ~200 g, Harlan-Winkelmann) were housed and treated according to OECD guideline for the testing of chemicals No. 407 with full access to food and drinking water. OTA was dissolved in corn oil and administered once daily by gavage. There were four different dosage groups (1 mg/kg bw and 10 mg/kg, bw sacrificed after 24 h; 1 mg/kg bw and 10 mg/kg bw, sacrificed after 72 h) containing five animals each, and two control groups (administration of pure corn oil once daily, sacrificed after 24 or 72 h, respectively) containing four animals each. Animals were sacrificed by exsanguination and kidneys were removed and cut longitudinally resulting in four half kidneys. One half was fixed in 10% formalin solution for histopathological examination and the remaining three kidney pieces were immediately placed in RNA-Later (Qiagen) and stored at 4°C to prevent degradation of RNA (Florell et al., 2001Go). Within four weeks kidneys were separated into cortex and medulla and both tissue sections were stored separately at -80°C until RNA was isolated.

Microarray hybridization.
Total RNA was isolated from kidney cortex or PTC culture samples using the RNeasy Midi Kit (Qiagen) according to the manufacturer’s protocol. RNA was quantified with Ribo Green (Molecular Probes, Leiden, the Netherlands) and quality was checked with a Bioanalyzer 2100 (Agilent Technologies, Waldbronn, Germany). RNA was reverse transcribed into aminoallyl-dUTP containing cDNA with Superscript II. cDNA was labelled by chemical condensation with Cy-3-N-hydroxysuccinimid (NHS) monofunctional dye (treated) and Cy-5-NHS monofunctional dye (control) (both purchased from Amersham Biosciences, Freiburg, Germany) following an indirect labelling protocol described in detail at the following Web site: www.pangloss.com/seidel/Protocols/amino-allylRT.html. Labeled cDNA from treated and pooled control samples was mixed and hybridized at 42°C overnight onto spotted cDNA microarrays (Phase-1, Santa-Fe, NM) containing 696 different toxicological relevant rat genes in four replicates as well as negative controls and several spots with Arabidopsis thalliana sequences serving as hybridization controls. Microarrays were washed with decreasing stringency, dried by centrifugation for 5 min at 1000 x g and stored at 4°C in darkness until scanning. Intensity of fluorescence was measured within 24 h after washing at 532 nm (Cy-3) and 635 nm (Cy-5) using a GenePix 4000A scanner (Axon Instruments, Foster City, CA).

Analysis of microarray data.
Fluorescence intensities in both channels were background subtracted for each spot by GenePix 4000A Software (Axon Instruments). Matrix express software (Phase-1) was employed for calculating the mean fluorescence intensities of the four replicates of every gene. Spots were considered as valid if fluorescence intensity was higher than 50 for each dye. Genes were only taken into further calculations if at least two of the four replicate spots were valid and the coefficient of variation between the replicates was below 25%. To assess the fold changes in gene expression the ratios of fluorescence intensity of Cy-3 (treated) against Cy-5 (control) were calculated for every gene passing the quality settings. Comparability of different chips was ensured by normalizing these ratios to the mean ratio of the chip. Statistical analysis, including t-test and different clustering methods like hierarchical clustering was performed with the Expressionist software of Gene Data (Basel, Switzerland). P-values < 0.01 were considered as statistically significant. Two analyses were carried out with p-values of 0.005 and 0.001 in order to focus on the most relevant genes. It is indicated in the different results sections which p-value was applied in each case. Changes in gene expression were only considered as biologically relevant if they exceeded twofold in comparison to control.

Histopathology.
Formalin fixed kidney sections were processed and embedded in paraffin. Sections of approximately 5 µm were taken and stained with haematoxylin and eosin (H&E). The sections were microscopically examined by a veterinary pathologist with experience in evaluation of laboratory animal tissues and peer reviewed.

Real-time polymerase chain reaction (PCR; TaqMan®) experiments.
Total RNA was isolated from samples as described above, but this time including DNase-digestion with Qiagen’s DNase-Set according to the manufacturer’s protocol. RNA was quantified and quality-checked as described above. RNA was reverse transcribed into cDNA following the manufacturer’s advice using the "Superscript First-Strand Synthesis System for RT-PCR" Kit (Invitrogen). Real-time PCR was performed with an ABI Prism® 7900HT Sequence Detection System (Applied Biosystems, Weiterstadt, Germany) using 5 ng/µl of template cDNA and SYBR Green PCR Core Reagents (Applied Biosystems) as recommended in the manufacturer’s protocol. Primers were designed for calpactin I heavy chain, ceruloplasmin, clusterin, GADD153, GADD45, glutathion-S-transferase (GST) alpha and sulfotransferase K2 with Primer Express Software v2.0 (Applied Biosystems) following Applied Biosystem’s instructions for optimal primer design. Sequences of forward and reverse primers are listed in Table 1Go. Primers were purchased from Invitrogen and checked for specificity by blast and by performing melting curves after PCR. To confirm their identity the PCR-amplicons were additionally analysed by gel electrophoresis and in part by DNA sequencing using the DNA Sequencing Kit "Big-Dye Terminator Cycle Sequencing v2.0" and an ABI Prism 310 Genetic Analyzer (both from Applied Biosystems). Samples were amplified in triplicate and normalized against 18S RNA amplified in each sample using primers from Quantum RNA 18S Internal Standards Kit (Ambion, Huntingdon, U.K.) and following the {Delta}{Delta}Ct calculation method described in detail in Applied Biosystems User Bulletin #2: "Relative Quantitation of gene expression". Results represent average fold changes of five animals per group.


View this table:
[in this window]
[in a new window]
 
TABLE 1 Sequences of Forward and Reverse Primers for TaqMan® Experiments
 
RESULTS

Histopathology
Rats dosed with OTA (1 mg/kg bw or 10 mg/kg bw) for 24 h did not exhibit remarkable histopathological changes except a slight accumulation of hyaline material in the tubular lumen, indicating an altered glomerular filtration rate (data not shown). Rats treated with both doses of OTA for 72 h showed clear degenerative lesions mainly located in the inner part of the cortex and in the outer stripe of the medulla (Figs. 1CGo and 1DGo). No such changes were seen in the subcapsular region. Single cell necrosis of the tubular epithelium with frequent exfoliation of cells into the tubular lumen were detectable. Necrosis was accompanied by scattered apoptotic bodies, as evidenced by diminished cell size and condensed nuclear chromatin.



View larger version (170K):
[in this window]
[in a new window]
 
FIG. 1. Light micrographs of paraffin-embedded kidneys from control rats and kidneys of rats treated with 10mg/kg OTA for 72 h (H&E stain). (A) Control kidney of a rat treated with pure cornoil for 72 h. The inner part of the cortex exhibits normal morphology. Original magnification: x250. (B) Same section at higher magnification (x400). (C) Kidney section from a treated rat. Single cell necroses, indicating apoptotic cell death are scattered throughout the section (arrows) as well as exfoliation of single epithelial cells into the tubular lumen (stars). Original magnification x250. (D) Same section at higher magnification (x400). Bars indicate 40 µm (A and C) or 25 µm (B and D), respectively.

 
Gene Expression Analyses
Proximal tubular cells were treated with a low (5 µM) and a high (12.5 µM) concentration of OTA over a short (24 h) and a long (72 h) period of time. Experiments were carried out in duplicate. In vivo experiments were also performed with a low (1 mg/kg bw) and a high (10 mg/kg bw) concentration of OTA over a short (24 h) and a long (72 h) period of time. Each dose group consisted of five animals, which were analyzed individually on cDNA chips. Quality of microarrays was confirmed by box plot analysis and microarrays that did not match all of the quality criteria were withdrawn from further analyses. The box plot analysis provides a visual summary of data within all samples, next to each other. It enables identification of samples that contain a high percentage of outliers and do not follow a normal data distribution. For all remaining microarrays fold changes of the genes were calculated by taking the ratio between treated and untreated signal for every gene. Only fold changes exceeding twofold were considered as biologically relevant. Our analyses revealed 254 genes that were differentially regulated more than twofold in at least one of the experiments. Eighty-nine were upregulated and 165 were downregulated. In the following we will concentrate on those differentially regulated genes that exhibited characteristic model-, dose-, or time-dependent expression profiles (84 genes) according to our statistical criteria.

Effects of OTA in Vivo and in Vitro: Similarities and Differences
Principal components analysis suggested discrepancies between in vivo and in vitro experiments as shown in Figures 2aGo and 2bGo. To extract genes that behave similarly or significantly different between in vitro and in vivo experiments, we compared both groups by t-tests (p = 0.001). The statistical tests revealed model-dependent and model-independent gene expression changes induced by OTA (Table 2Go). We identified 215 model-independent differentially regulated genes ("common genes"), which are discussed later according to their dose- or time-dependent expression profiles. The common genes represent 84.6% of all transcriptional changes, indicating a high correspondence of in vitro and in vivo data. We also found 39 model-dependent differentially regulated genes ("discriminator genes") and evaluated these genes in the discussion section with respect to the different characteristics of the cell culture test model and the in vivo test model.



View larger version (11K):
[in this window]
[in a new window]
 
FIG. 2. (a) Principal Components Analysis (PCA) employing 254 twofold differentially regulated genes; in vivo (dark) and in vitro (light) experiments cluster in different areas. (b) PCA employing 215 twofold differentially regulated genes (model-discriminating genes are excluded). In vivo and in vitro experiments cluster in one cloud.

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Genes Behaving Similarly or Differently in Vivo and in Vitro
 
Of these 39 "discriminator genes," four genes were upregulated and 35 were downregulated more than twofold in at least one experiment either in vivo or in vitro. The "discriminator genes" were clustered employing a hierarchical clustering method (Fig. 3Go). This yielded three different groups of genes according to their expression profiles in vivo and in vitro. The first group comprised genes that are upregulated more than twofold in vivo and hardly deregulated in vitro (Table 3Go, section A). The second group consisted of genes that are not deregulated in vivo, but downregulated more than twofold in vitro (Table 3Go, section B). Genes that were downregulated in vivo and hardly differentially regulated in vitro were taken together in the third group (Table 3Go, section C). Within the three groups the genes were sorted according to their physiological functions. Accession numbers of ESTs with unknown function were listed at the end of the tables. Interestingly there were no genes that were upregulated in vitro, but downregulated in vivo implying that in vitro upregulated genes showed the same direction of gene expression change in vivo.



View larger version (55K):
[in this window]
[in a new window]
 
FIG. 3. Hierarchical clustering of 39 twofold differentially regulated "discriminator genes." There are three groups of genes exhibiting different expression profiles. The scale on the left side represents the color code according to the fold changes of the genes in each experiment. Further explanations in the text.

 

View this table:
[in this window]
[in a new window]
 
TABLE 3 "Discriminator Genes," Grouped in Clusters 1–3 According to Their Expression Profiles
 
"Discriminator genes" have been excluded from the following analyses to guarantee that the investigated changes in gene expression are not model-dependent but only dose- or time-dependent effects of OTA treatment.

Dose-Dependent Effects after OTA Administration
Dose-dependent profiles were only investigated in the 72 h experiments and time-dependent profiles were only investigated in high-dose experiments in order to identify the most predominant biological effects of OTA. Performing the t-test with all 72 h low-dose groups (5 µM in vitro and 1 mg/kg bw in vivo) against all 72 h high-dose groups (12.5 µM in vitro and 10 mg/kg bw in vivo) revealed 31 genes that are discriminating significantly (p = 0.005) between high-dose and low-dose OTA treatment. In this group 10 genes were upregulated more than twofold and 21 genes were downregulated more than twofold. All dose-dependent differentially regulated genes are listed in Tables 4Go–7Go according to the physiological function of their encoded proteins. The genes were grouped into four main functional categories: acute phase response and inflammation (Table 4Go), responses to DNA damage (like DNA repair processes, apoptosis, and cancerogenesis; Table 5Go), cellular metabolism and detoxification processes (Table 6Go), and oxidative stress (Table 7Go). Tendencies of the directions of gene expression changes from low dose to high dose are indicated in the tables by black arrows. In Tables 4Go–7Go only average fold changes of all 72 h low-dose or all 72 h high-dose experiments, respectively, are shown. We also found 11 dose-dependent deregulated ESTs, which are not further discussed, because their function has not been described so far.


View this table:
[in this window]
[in a new window]
 
TABLE 4 Acute Phase Response and Inflammation among Dose-Dependent Regulated Genes
 

View this table:
[in this window]
[in a new window]
 
TABLE 7 Oxidative Stress among Dose-Dependent Regulated Genes
 

View this table:
[in this window]
[in a new window]
 
TABLE 5 DNA Damage, DNA Repair, Apoptosis, and Cancer among Dose-Dependent Regulated Genes
 

View this table:
[in this window]
[in a new window]
 
TABLE 6 Cellular Metabolism and Detoxification Processes among Dose-Dependent Regulated Genes
 
Time-Dependent Effects after OTA Administration
To extract genes that are differentially regulated in a time-dependent manner after OTA treatment, we performed a t-test with all high-dose, early time-point experiments (24 h) against all high-dose, late time-point experiments (72 h). Fourteen genes were found to be discriminating significantly (p = 0.01) between the timepoints. Twelve genes are upregulated and two genes are downregulated. In Tables 8Go–10Go average fold changes of these genes are listed according to their physiological function. Time-dependent differentially regulated genes appeared to belong to the same functional categories as were described above for the dose-dependent differentially regulated genes, except that there were no time-dependent regulated oxidative stress response genes. Arrows correspond to the tendency of direction of the expression changes from 24 to 72 h of OTA treatment. Two ESTs showed time-dependent expression changes, but due to missing knowledge about their function, they also remain undiscussed.


View this table:
[in this window]
[in a new window]
 
TABLE 8 Acute Phase Response and Inflammation among Time-Dependent Regulated Genes
 

View this table:
[in this window]
[in a new window]
 
TABLE 10 Cellular Metabolism and Detoxification Processes among Time-Dependent Regulated Genes
 
Dose- and Time-Dependent Effects after OTA Administration
We found one gene, GADD 45, which exhibited both dose- and time-dependent expression patterns. GADD 45 belongs to the DNA damage category (Tables 5Go and 9Go).


View this table:
[in this window]
[in a new window]
 
TABLE 9 DNA Damage, DNA Repair, Apoptosis, and Cancer among Time-Dependent Regulated Genes
 
Other Effects after OTA Administration
We also found two genes, Ref-1 and stem cell factor, which are downregulated in all of the experiments. This seems to be a general effect of OTA treatment, which is neither dose- nor time-dependent. The remaining 129 genes that show transcriptional changes over twofold in at least one experiment did not exhibit any dose-, time- or otherwise related expression profiles according to our statistical criteria (t-test with p-value < 0.01) and were hence not discussed.

Quantitative Real-Time PCR Analyses
To confirm some selected microarray data we also performed quantitative real-time PCR (TaqMan®-PCR) for several genes with samples of animals treated with the high dose of OTA for 72 h. In Figure 4Go results are shown as fold changes of calpactin I heavy chain, ceruloplasmin, clusterin, GADD 153, GADD 45, GST alpha, and sulfotransferase K2 in comparison to the fold changes measured in microarray experiments. The shown real-time PCR and microarray data represent average fold changes of five animals per group.



View larger version (23K):
[in this window]
[in a new window]
 
FIG. 4. Average fold changes of Taqman® (TQ) and microarray (MA) experiments are shown. Error bars represent SDs. Upper line indicates +twofold change; lower line indicates -twofold change.

 
DISCUSSION

As shown in Table 2Go most of the changes in gene expression induced by OTA are similar in vitro and in vivo. This demonstrates that the employed in vitro model is able to reflect most of the expression changes due to OTA toxicity observed in vivo. Interestingly, all of the in vitro upregulated genes are also upregulated in vivo, implying that all upregulations in vitro reflect the in vivo situation. Nevertheless there are a few effects which cannot be explained as general reactions to OTA treatment, but must be explained under consideration of the employed test model.

Model-Dependent Gene Expression Changes after OTA Treatment
The twofold differentially regulated genes found to be discriminating between the in vitro and the in vivo test model could be clustered into three groups according to their expression profiles. 60S ribosomal protein L6 and activin receptor II are upregulated in vivo but hardly regulated in vitro (Table 3Go, section A). 60SL6 is part of the large 60S unit of the ribosomes and hence involved in translational processes and ACRII plays an important role in cell growth and proliferation after activation by its ligand activin. We speculate about the upregulation of ACRII being a consequence of a possible downregulation of activin, which is known to be essential for tissue regeneration after renal injury (Maeshima et al., 2001Go), but this has to be confirmed in future investigations.

Genes that show significant downregulation in vitro but are hardly regulated in vivo are grouped together in cluster 2 (Table 3Go, section B). Alpha prothymosin is known to play an important role in the negative regulation of cell differentiation (Rodriguez et al., 1998Go) and ID-1 has been described as a negative regulator of differentiation of tubular cells in the regenerating kidney (Matejka et al., 1998Go). In PTC culture the basal proliferation rate of the cells is higher than in vivo, because after isolation from the organ, a confluent cell layer has to be formed. Due to this "preconditioning," downregulation of alpha prothymosin and ID-1 after OTA treatment in order to regenerate the destroyed cell formation might be more distinct in cell culture than in vivo. Summarizing the above differential regulation of these two genes also might serve to regenerate damaged tissue after OTA treatment. Moreover, our results indicate that tissue regeneration is mediated, at least in part, differently in vitro an in vivo.

Collagen II and protein kinase C alpha are also downregulated in vitro but do not show any expression changes in vivo (Table 3Go, section B). Collagen II is of major importance for the structure of the extracellular matrix and contributes to the development of fibrosis. Our recent studies (data not shown) revealed that primary PTC culture tends to show upregulation in fibrosis markers such as collagen II and Lysyl oxidase after a certain period of cultivation. An apparent downregulation of collagen II after OTA treatment in PTC culture might therefore be due to an overexpression of this gene in control cultures, which is not accompanied by an overexpression in treated cultures. PKC{alpha} is part of intracellular signaling and it has recently been suggested that downregulation of PKC{alpha} expression might have a suppressing effect on tumor cell growth in melanoma cell cultures (Krasagakis et al., 2002Go). Our findings implicate that effects mediated by downregulation of PKC{alpha} are supported after OTA treatment in vitro but not in vivo.

The third group consists of genes, which are downregulated in vivo, but hardly regulated in vitro (Table 3Go, section C). Most of these genes, including betaine homocysteine methyltransferase, cytochrome P450 2D18, D-dopachrome tautomerase, gamma-glutamyl-transpeptidase, paraoxonase, peroxisomal-3-ketoacyl-CoA thiolase 2, and retinol dehydrogenase, encode for proteins with known functions in metabolism, biotransformation, or other detoxification processes. Others such as bile salt export pump, organic anion transporter K1, organic cation transporter 2, renal organic anion transporter, and senescence marker protein 30 play important roles in ion or transcellular transport processes. It is a known fact that in contrast to in vivo, cultured cells sometimes show declines in special transport capacities and oxidative metabolism (Dickmann and Mandel, 1989Go). With regards to the raw data (fluorescence intensities not shown) some of the genes mentioned above were indeed expressed at very low rate in our PTC culture in comparison to their fluorescence intensities in vivo and a decrease in expression after OTA treatment might not be detectable in the cell culture anymore. Moreover, transport proteins usually make use of a concentration gradient between basolateral and luminal plasma membrane. In our in vitro experiments PTCs were cultured on collagen-coated plastic surfaces, resulting in the loss of a basolateral compartment and hence loss of a concentration gradient. This might be, at least in part, a reason for the low abundance of transporter encoding mRNAs in PTC culture. In contrast to these in vitro results, the metabolic and transport activities of the proximal tubule in vivo are severely affected by OTA. It has been described before that OTA treatment causes dramatic decreases in transport capacities of proximal tubular cells leading to prolongation of OTA toxicity (Gekle and Silbernagl, 1994Go), although it has not been suggested so far that these changes might yet occur on the transcriptional (mRNA) level of the transporter encoding genes.

Apolipoprotein C1 is known to be expressed mainly in the liver and only to a very low extent in the kidney (Simonet et al., 1991Go). This correlates with the low fluorescence intensity we observed for ApoC1 in all of our microarray experiments (raw data not shown). For this reason the apparent downregulation of this low abundance gene must be considered carefully.

Differences in gene expression changes after OTA administration in vivo and in vitro might also result from the fact that the kidney consists of a great variety of different cell types. Thus, genes like aspartoacylase, which is known to be specifically expressed in neural cells (Baslow et al., 1999Go), are likely to be detected in renal tissue samples, but not in our PTC monoculture.

The CDK 108 gene encodes for a NADP-regulated thyroid hormone-binding protein in kidney (Vie et al., 1997Go) and transcription of the neutral endopeptidase 24.11 gene has been described to be androgen-regulated (Shen et al., 2000Go). Obviously, transcription of these two genes is at least in part hormonally regulated. In vitro hormonal interactions that might lead to changes in gene expression can hardly be imitated.

Functions of the ESTs belonging to clusters 1, 2, and 3 remain unclear and are therefore not discussed in this article.

Dose- and Time-Dependent OTA Effects
In this part of the discussion most emphasis is laid on upregulated genes, because downregulations are difficult to interpret and in most cases they can be traced back to the occurrence of cell destruction after OTA treatment as observed in the histopathological examinations. The dose- and time-dependent gene expression changes we observed after OTA treatment mainly affect four different functional categories (Tables 4Go–10Go), which will be discussed in detail.

It is apparent from our studies that changes in cellular metabolism and oxidative stress are more dose-dependent, while inflammation and acute phase responses as well as responses to DNA damage like repair processes, apoptosis, and carcinogenesis occur predominantly in a time-dependent manner.

Acute Phase Response and Inflammation
The first functional category comprises genes, which encode for proteins with known roles in acute phase response (APR) and inflammation. Most of the APR-related genes encode for protease inhibitors. Alpha-2-macroglobulin is a unique protease inhibitor that inhibits both foreign and own proteases in the plasma and is well known as a positive acute phase protein (Schreiber et al., 1989Go). In correspondence with this report a2MG expression is significantly upregulated after 72 h of OTA treatment.

Ceruloplasmin is a copper-dependent positive acute phase protein, expressed in the kidney during inflammation (Kalmovarin et al., 1991Go) and in connection with oxidative stress (DeSilva and Aust, 1993Go). Under hypoxic conditions its expression is regulated, at least in part, by binding of hypoxia-inducible factor 1 to a hypoxia-responsive element (HRE) in the promoterregion of the ceruloplasmin gene (Mukhopadhyay et al., 2000Go). In our experiments HIF1 is upregulated at high doses of OTA and will be further discussed in the section "Oxidative Stress."

Contrapsin-like protease inhibitor is known as a negative acute phase protein (Pages et al., 1990Go) and has also been found to be downregulated in mouse lung adenocarcinomas (Lin et al., 2001Go). In our experiments CPI-21 gene expression decreased at high concentrations of OTA suggesting a potential support of APR and cancer development.

Alpha-fibrinogen exhibits significant time-dependent upregulation and the tissue factor gene shows significant dose-dependent upregulation. Both genes play important roles in coagulation after tissue lesions. By binding to PAR-1 thrombin mediates the conversion of fibrinogen into fibrin fibres to form blood clots (Macfarlane et al., 2001Go). TF is known to start the intrinsic pathway of blood clotting in traumatised tissue and is also discussed to be involved in the PAR-1 mediated blood clotting pathway (Riewald and Ruf, 2001Go). Our results suggest severe tissue lesions leading to activation of several mechanisms of blood clotting after OTA treatment.

Ciliary neurotrophic factor shows dose-dependent upregulation in our experiments. It is known as a tissue repair factor, which is upregulated in areas surrounding necrosis (Lin et al., 1998Go). Zinc finger protein is induced as an immediate-early response gene in acute liver injury and has also been suggested to play an important role in repair of tissue lesions (Kojima et al., 2000Go). This correlates with the increasing occurrence of necrosis in our histopathological examinations.

DNA Damage, DNA Repair, Apoptosis, and Cancer
Some of our results confirm the formerly described DNA damaging effect of OTA (Obrecht-Pflumio and Dirheimer, 2000Go). For example, upregulation of GADD 153 (growth arrest and DNA damage-inducible gene) and GADD 45 is often reported after treatment of cells or laboratory animals with DNA damaging agents such as UV radiation or methylmethanesulfonate (Beard et al., 1996Go). Furthermore, an upregulation of dynein light chain 1 has been interrelated with the transport of p53 to the nucleus after DNA damage by Giannakakou et al.(2000)Go.

Upregulation of calpactin I heavy chain (annexin II) has formerly been observed in renal cell carcinomas of rats induced by oxidative damaging agents (Tanaka et al., 2000Go). There is evidence for calp1 being a cofactor for DNA polymerase alpha (Kumble et al., 1992Go) and it has also been proved as a substrate for an oncogene growth factor-associated protein tyrosine kinase (Skouteris and Schroder, 1996Go) suggesting induction of calp1 as a common molecular event in DNA repair and development of different cancers.

Nucleosome assembly protein is essential for unpacking of chromatin in order to replicate or repair DNA (Rodriguez et al., 1997Go). The upregulation of NAP in our experiments might therefore be due to repair processes after DNA damage.

Our results also provide evidence for induction of apoptosis after OTA treatment. Annexin V upregulation is known to be involved in marking apoptotic cells for recognition by phagocytes (Bacso et al., 2000Go). Clusterin has been reported to be upregulated in human renal cell carcinoma cells and a function of clusterin as a protector of apoptosis has been suggested (Miyake et al., 2002Go). The expression changes of these genes provide evidence for the induction of apoptotic processes after longer administration of OTA.

Cathepsin S is discussed to play a crucial role in tumor progression due to its matrix degrading properties even at a very early stage of tumor development (Fernandez et al., 2001Go). Stem cell factor (SCF), also known as "kit ligand," is downregulated in almost all of the experiments. It plays a role in cell-matrix adhesion and has also been described as a rescue factor in mast cell apoptosis (Ashman, 1999Go; Mekori et al., 1995Go). The upregulation of CatS and the downregulation of SCF might support loss of cell–cell adhesion and hence tumor development might be facilitated.

Cellular Metabolism and Detoxification Processes
Liver fatty acid binding protein was upregulated after 72 h of OTA treatment. It is known to be involved in fatty acid metabolism and peroxisome proliferation (Stremmel et al., 2001Go). This indicates a possible influence of OTA on fat metabolism.

The downregulated genes in this group mostly belong to two main subcategories: energy metabolism (NADP-dependent isocitrate dehydrogenase, NADPH quinone oxidoreductase 1, beta-alanine synthetase, ATP-stimulated glucocorticoid-receptor translocation promoter) and biotransformation and detoxification (alcohol dehydrogenase 1, argininosuccinate lyase, glutamine synthetase, GSTs alpha, P1 and Ya, sulfotransferase K2). These dramatic losses of basic cell functions are mainly dose-dependent and might be a consequence of the widespread occurrence of necrotic alterations after OTA treatment, which have been observed in the histopathological examinations.

Oxidative Stress
HIF1 is known to transcriptionally activate several genes under hypoxic conditions via binding to HREs in their promoter regions (Caro, 2001Go). Such HREs are definitely present in the ceruloplasmin gene (Mukhopadhyay et al., 2000Go). HIF1 is also discussed as a prognostic marker for various cancers (L’Allemain, 2002Go).

Genes encoding for catalase and glutathione synthetase are downregulated at high doses of OTA and have both been mentioned before in connection with oxidative stress. Downregulation of catalase mRNA is considered as an early marker for oxidative stress (Dobashi et al., 2000Go). GlutS catalyzes the synthesis of Glutathione, which is an important antioxidant and it has been reported that due to this function GlutS is upregulated during hypoxia (Anderson, 1998Go). In contrast, after OTA treatment GlutS shows downregulation in a dose-dependent manner. Ref-1 (Redox factor-1) has multiple functions in DNA repair, oxidative signalling, transcription factor activation, and cancer (Evans et al., 2000Go). To our knowledge there is no literature that reports the role of permanently downregulated ref-1.

Conclusion
From our experiments we can conclude that there are only small differences between the two investigated models, which could be almost completely traced back to the following circumstances:

  1. Genes, whose expression is restricted to cell types in renal cortex samples other than PTC, are not supposed to be detected in a PTC monoculture.
  2. Lack of hormonal or immunological influences in vitro.
  3. Some of the transport and biotransformation processes appear to be expressed to a lesser extent in isolated PTC compared to in vivo.

Nevertheless, the majority of transcriptional alterations we found in the kidney following OTA treatment was comparable to the alterations observed in PTC cultures. This indicates that PTCs are able to maintain most of their physiological responsiveness in vitro. Therefore considerable extent of agreement can be concluded between the in vitro and the in vivo model.

Our second aim was to investigate the mode of action of OTA-induced nephrotoxicity in more detail employing DNA-microarrays as powerful research tools. There have been several approaches before showing that microarray-based analyses of nephrotoxic events can provide new insights into development of nephrotoxicity on a molecular basis (Katsuma et al., 2002Go). We were able to confirm most of the literature-described effects of OTA such as DNA damage possibly mediated by oxidative stress on the transcriptional level. We also found evidence for acute phase responses, for severe tissue lesions, and for massive deterioration of cellular metabolism. We were able to detect transcriptional alterations and gene expression networks such as upregulation of calpactin I heavy chain, Annexin V, Gadd153, clusterin, and hypoxia-inducible factor I, which have not been described before to be involved in mediation of OTA-induced nephrotoxicity. This provides a challenging basis for detailed future research. Finally, we were able to confirm selected microarray data by real-time quantitative PCR. Although this method relies on amplification while microarrays reflect original, unamplified values, the PCR data corresponded well with the microarray data in their direction of change, thus supporting the validity of the conclusions drawn from microarray-based interpretations.

ACKNOWLEDGMENTS

We thank Dr. F. Krötlinger and A. Brockes for supervision of the animal studies, Dr. M. Rinke for crosscheck of the histopathological examinations, and Dr. J. Franz and T. Boldt for assistance in sequencing of PCR amplicons.

NOTES

1 To whom correspondence should be addressed. Fax: + 49 (0) 202 36 4137. E-mail: anke.luehe.al{at}bayer-ag.de. Back

REFERENCES

Anderson, M. E. (1998). Glutathione: An overview of biosynthesis and modulation. Chem. Biol. Interact. 111–112, 1–14.[CrossRef]

Ashman, L. K. (1999). The biology of stem cell factor and its receptor C-kit. Int. J. Biochem. Cell Biol. 31(10), 1037–1051.[CrossRef][ISI][Medline]

Bacso, Z., Everson, R. B., and Eliason, J. F. (2000). The DNA of annexin V-binding apoptotic cells is highly fragmented. Cancer Res. 60(16), 4623–4628.[Abstract/Free Full Text]

Baslow, M. H., Suckow, R. F., Sapirstein, V., and Hungund, B. L. (1999). Expression of aspartoacylase activity in cultured rat macroglial cells is limited to oligodendrocytes. J. Mol. Neurosci. 13(1–2), 47–43.[CrossRef][ISI][Medline]

Baudrimont, I., Betbeder, A. M., Gharbi, A., Pfohl-Leszkowicz, A., Dirheimer, G., and Creppy, E. E. (1994). Effect of superoxiddismutase and catalase on the nephrotoxicity induced by subchronical administration of ochratoxin A in rats. Toxicology 89(2), 101–111.[CrossRef][ISI][Medline]

Beard, S. E., Capaldi, S. R., and Gee, P. (1996). Stress responses to DNA damaging agents in the human colon carcinoma cell line RKO. Mutat. Res. 371(1–2), 1–13.[CrossRef][ISI][Medline]

Biemesderfer, D., DeGray, B., and Aronson, P. S. (1998). Membrane topology of NHE3. Epitopes within the carboxyl-terminal hydrophilic domain are exoplasmic. J. Biol. Chem. 273(20), 12391–12396.[Abstract/Free Full Text]

Boom, S. P., Gribnau, F. W., and Russel, F. G. (1992). Organic cation transport and cationic drug interactions in freshly isolated proximal tubular cells of the rat. J. Pharmacol. Exp. Ther. 263(2), 445–450.[Abstract]

Burczynski, M. E., McMillian, M., Ciervo, J., Li, L., Parker, J. B., Dunn, R. T., II, Hicken, S., Farr, S., and Johnson, M. D. (2000). Toxicogenomics-based discrimination of toxic mechanism in HepG2 human hepatoma cells. Toxicol. Sci. 58, 399–415.[Abstract/Free Full Text]

Caro, J. (2001). Hypoxia regulation of gene transcription. High Alt. Med. Biol. 2(2), 145–154.[CrossRef][ISI][Medline]

Castegnaro, M., Mohr, U., Pfohl-Leszkowicz, A., Esteve, J., Steinmann, J., Tillmann, T., Michelon, J., and Bartsch, H. (1998). Sex- and strain-specific induction of renal tumors by ochratoxin A in rats correlates with DNA adduction. Int. J. Cancer 77(1), 70–75.[CrossRef][ISI][Medline]

DeSilva, D. M., and Aust, S. D. (1993). Ferritin and ceruloplasmin in oxidative damage: Review and recent findings. Can. J. Physiol. Pharmacol. 71(9), 715–720.[ISI][Medline]

Dickmann, K. G., and Mandel, L. J. (1989). Glycolytic and oxidative metabolism in primary renal proximal tubule cultures. Am. J. Physiol. 257(2 Pt.1), C333–C340.[ISI][Medline]

Dirheimer, G., and Creppy, E. E. (1991). Mechanism of action of ochratoxin A. IARC Sci. Publ. 115, 171–186.[Medline]

Dobashi, K., Ghosh, B., Orak, J. K., Singh, I., and Singh, A. K. (2000). Kidney ischemia-reperfusion: modulation of antioxidant defenses. Mol. Cell Biochem. 205(1–2), 1–11.[CrossRef][ISI][Medline]

Dopp, E., Muller, J., Hahnel, C., and Schiffmann, D. (1999). Induction of genotoxic effects and modulation of the intracellular calcium level in syrian hamster embryo (SHE) fibroblasts caused by ochratoxin A. Food Chem. Toxicol. 37(7), 713–721.[CrossRef][ISI][Medline]

Evans, A. R., Limp-Foster, M., and Kelley, M. R. (2000). Going APE over ref-1. Mutat. Res. 461(2), 83–108.[ISI][Medline]

Fernandez, P. L., Farre, X., Nadal, A., Fernandez, E., Peiro, N., Sloane, B. F., Shi, G. P., Chapman, H. A., Campo, E., and Cardesa, A. (2001). Expression of cathepsins B and S in the progression of prostate carcinoma. Int. J. Cancer 95(1), 51–55.[CrossRef][ISI][Medline]

Florell, S. R., Coffin, C. M., Holden, J. A., Zimmermann, J. W., Gerwels, J. W., Summers, B. K., Jones, D. A., and Leachman, S. A. (2001). Preservation of RNA for functional genomic studies: A multidisciplinary tumor bank control. Mod. Pathol. 14(2), 116–128.[ISI][Medline]

Gekle, M., and Silbernagl, S. (1994). The role of the proximal tubule in ochratoxin A nephrotoxicity in vivo: Toxodynamic and toxokinetic aspects. Ren. Physiol. Biochem. 17(1), 40–49.[ISI][Medline]

Giannakakou, P., Sackett, D. L., Ward, Y., Webster, K. R., Blagosklonny, M. V., and Fojo, T. (2000). P53 is associated with cellular microtubules and is transported to the nucleus by dynein. Nat. Cell. Biol. 2(10), 709–717.[CrossRef][ISI][Medline]

Gillman, I. G., Clark, T. N., and Manderville, R. A. (1999). Oxidation of ochratoxin A by an Fe-porphyrin system: Model for enzymatic activation and DNA cleavage. Chem. Res. Toxicol. 12(11), 1066–1076.[CrossRef][ISI][Medline]

Hald, B. (1991). Porcine nephropathy in Europe. IARC Sci. Publ. 115, 49–56.[Medline]

Huff, J. E. (1991). Carcinogenicity of ochratoxin A in experimental animals. IARC Sci. Publ. 115, 229–244.[Medline]

Kalmovarin, N., Friedrichs, W. E., O’Brien, H. V., Linehan, L. A., Bowman, B. H., and Yang, F. (1991). Extrahepatic expression of plasma protein genes during inflammation. Inflammation 15(5), 369–379.[ISI][Medline]

Katsuma, S., Shiojima, S., Hirasawa, A., Takagaki, K., Kaminishi, Y., Koba, M., Hagidai, Y., Murai, M., Ohgi, T., Yano, J., and Tsujimoto, G. (2002). Global analysis of differentially expressed genes during progression of calcium oxalate nephrolithiasis. Biochem. Biophys. Res. Com. 296, 544–552.[CrossRef][ISI][Medline]

Kojima, S., Hayashi, S., Shimokado, K., Suzuki, Y., Shimada, J., Crippa, M. P., and Friedmann, S. L. (2000). Transcriptional activation of urokinase by the kruppel-like factor ZF9/COPEB activates latent TGF-ß1 in vascular endothelial cells. Blood 95(4), 1309–1316.[Abstract/Free Full Text]

Krasagakis, K., Fimmel, S., Genten, D., Eberle, J., Quas, P., Ziegler, W., Haller, H., and Orfanos, C. E. (2002). Lack of protein kinase C (PKC)-beta and low PKC-alpha, -delta, -epsilon, and -zeta isozyme levels in proliferating human melanoma cells. Int. J. Oncol. 20(4), 865–871[ISI][Medline]

Kumble, K. D., Iversen, P. L., and Vishwanatha, J. K. (1992). The reole of primer recognition proteins in DNA replication: Inhibition of cellular proliferation by antisense oligodeoxyribonucleotides. J. Cell Sci. 101(Pt. 1), 35–41.[Abstract]

L’Allemain, G. (2002). The hypoxia-inducible factor HIF as a new target in cancer research. Bull. Cancer 89(3), 257–260.[ISI][Medline]

Lin, L., Wang, Y., Bergmann, G., Kelloff, G. J., Lubet, R. A., and You, M. (2001). Detection of differentially expressed genes in mouse lung adenocarcinomas. Exp. Lung Res. 27(3), 217–229.[CrossRef][ISI][Medline]

Lin, T. N., Wang, P. Y., Chi, S. I., and Kuo, J. S. (1998). Differential regulation of ciliary neurotrophic factor (CNTF) and CNTF receptor alpha (CNTFR alpha) expression following focal cerebral ischemia. Mol. Brain Res. 55(1), 71–80.[ISI][Medline]

Macfarlane, S. R., Seatter, M. J., Kanke, T., Hunter, G. D., and Plevin, R. (2001). Proteinase-activated receptors. Pharmacol. Rev. 53(2), 245–282.[Abstract/Free Full Text]

Maeshima, A., Zhang, Y., Nojima, Y., Naruse, T., and Kojima, I. (2001). Involvement of the activin-follistatin system in tubular regeneration after renal ischemia in rats. J. Am. Soc. Nephrol. 12, 1685–1695.[Abstract/Free Full Text]

Matejka, G. L., Thornemo, M., Kernholt, A., and Lindahl, A. (1998). Expression of Id-1 mRNA and protein in the post-ischemic regenerating rat kidney. Exp. Nephrol. 6(3), 253–264.[CrossRef][ISI][Medline]

Meisner, H., and Cimbala, M. (1986). Effect of ochratoxin A on gene expression in rat kidneys. Dev. Toxicol. Environ. Sci. 12, 261–271.[Medline]

Mekori, Y. A., Oh, C. K., and Metcalfe, D. D. (1995). The role of c-Kit and its ligand, stem cell factor, in mast cell apoptosis. Int. Arch. Allergy Immunol. 107(1–3), 136–138.[ISI][Medline]

Miyake, H., Gleave, M. E., Arakawa, S., Kamidono, S., and Hara, I. (2002). Introducing the clusterin gene into human renal cell carcinoma cells enhances btheir metastatic potential. J. Urol. 167(5), 2203–2208.[CrossRef][ISI][Medline]

Mukhopadhyay, C. K., Mazumder, B., and Fox, P. L. (2000). Role of hypoxia-inducible factor-1 in transcriptional activation of ceruloplasmin by iron deficiency. J. Biol. Chem. 275(28), 21048–21054.[Abstract/Free Full Text]

Nouwen, E. J., Dauwe, S., van der Biest, I., and De Broe, M. E. (1993). Stage- and segment-specific expression of cell-adhesion molecules N-CAM, A-CAM, and L-CAM in the kidney. Kidney Int. 44(1), 147–158.[ISI][Medline]

Obrecht-Pflumio, S., and Dirheimer, G. (2000). In vitro DNA and dGMP adducts formation caused by ochratoxin A. Chem. Biol. Interact. 127(1), 29–44.[CrossRef][ISI][Medline]

Omar, R. F., Hasinoff, B. B., Mejilla, F., and Rahimtula, A. D. (1990). Mechanism of Ochratoxin A stimulated lipid peroxidation. Biochem. Pharmacol. 40(6), 1183–1191.[CrossRef][ISI][Medline]

Pages, G., Rouayrenc, J. F., Le Cam, G., Mariller, M., and Le Cam, A. (1990). Molecular characterization of three rat liver serine-protease inhibitors affected by inflammation and hypophysectomy. Protein and mRNA analysis and cDNA cloning. Eur. J. Biochem. 190(2), 385–391.[Abstract]

Pfohl-Leszkowicz, A., Chakor, K., Creppy, E. E., and Dirheimer, G. (1991). DNA adduct formation in mice treated with ochratoxin A. IARC Sci. Publ. 115, 245–253.[Medline]

Riewald, M., and Ruf, W. (2001). Mechanistic coupling of protease signaling and initiation of coagulation by tissue factor. PNAS 98(14), 7742–7747.[Abstract/Free Full Text]

Rodriguez, P., Munroe, D., Prawitt, D., Chu, L. L., Bric, E., Kim, J., Reid, L. H., Davies, C., Nakagama, H., Loebbert, R., et al. (1997). Functional characterization of human nucleosome assembly protein-2 (NAP1L4) suggests a role as a histone chaperone. Genomics 44(3), 253–265.[CrossRef][ISI][Medline]

Rodriguez, P., Vinuela, J. E., Alvarez-Fernandez, L., Buceta, M., Vidal, A., Dominguez, F., and Gomez-Marquez, J. (1998). Overexpression of prothymosin alpha accelerates proliferation and retards differentiation in HL-60 cells. Biochem. J. 331(Pt. 3), 753–761.[ISI][Medline]

Schreiber, G., Tsykin, A., Aldred, A. R., Thomas, T., Fung, W. P., Dickson, P. W., Cole, T., Birch, H., De Jong, F. A., and Milland, J. (1989). The acute phase response in the rodent. Ann. N.Y. Acad. Sci. 557, 61–85.[Abstract]

Schwartz, G. G. (2002). Hypothesis: Does ochraroxin A cause testicular cancer? Cancer Causes Control. 13(1), 91–100.[CrossRef][ISI][Medline]

Shen, R., Sumimoto, M., Dai, J., Hardy, D. O., Navarro, D., Usmani, B., Papandreou, C. N., Hersh, L. B., Shipp, M. A., Freedman, L. P., and Nanus, D. M. (2000). Identification and characterization of two androgen response regions in the human neutral endopeptidase gene. Mol. Cell. Endocrinol. 170(1–2), 131–142.[CrossRef][ISI][Medline]

Simonet, W. S., Bucay, N., Pitas, R. E., Lauer, S. J., and Taylor, J. M. (1991). Multiple tissue-specific elements control the apolipoprotein E/C-I gene locus in transgenic mice. J. Biol. Chem. 266(14), 8651–8654.[Abstract/Free Full Text]

Skouteris, G. G., and Schroder, C. H. (1996). The hepatocyte growth factor receptor kinase-mediated phosphorylation of lipocortin-1 transduces the proliferating signal of the hepatocyte growth factor. J. Biol. Chem. 271(44), 27266–27273.[Abstract/Free Full Text]

Stoev, S. D. (1998). The role of ochratoxin A as a possible cause of Balkan endemic nephropathy and its risk evaluation. Vet. Hum. Toxicol. 40(6), 352–360.[ISI][Medline]

Stoev, S. D., Vitanov, S., Anguelov, G., Petkova-Bocharova, T., and Creppy, E. E. (2001). Experimental mycotoxic nephropathy in pigs provoked by a diet containing ochratoxin A and penicillic acid. Vet. Res. Commun. 25(3), 205–223.[CrossRef][ISI][Medline]

Stremmel, W., Pohl, L., Ring, A., and Herrmann, T. (2001). A new concept of cellular uptake and intracellular trafficking of long-chain fatty acids. Lipids 36(9), 981–989.[ISI][Medline]

Tanaka, T., Kondo, S., Iwasa, Y., Hiai, H., and Toyokuni, S. (2000). Expression of stress-response and cell proliferation genes in renal cell carcinoma induced by oxidative stress. Am. J. Pathol. 156(6), 2149–2157.[Abstract/Free Full Text]

Thomas, R. S., Rank, D. R., Penn, S. G., Zastrow, G. M., Hayes, K. R., Pande, K., Glover, E., Silander, T., Craven, M. W., Reddy, J. K., et al. (2001). Identification of toxicologically predictive gene sets using cDNA microarrays. Mol. Pharmacol. 60(6), 1189–1194.[Abstract/Free Full Text]

Thuvander, A., Dahl, P., and Breitholtz-Emanuelsson, A. (1996). Influence of perinatal ochratoxin A exposure on the immune system in mice. Nat. Toxins 4(4), 174–180.[Medline]

Vie, M. P., Evrard, C., Osty, J., Breton-Gilet, A., Blanchet, P., Pomerance, M., Rouget, P., Francon, J., and Blondeau, J. P. (1997). Purification, molecular cloning, and functional expression of the human nicodinamide-adenine dinucleotide phosphate-regulated thyroid hormone-binding protein. Mol. Endocrinol. 11(11), 1728–1736.[Abstract/Free Full Text]

Walker, R. (2002). Risk assessment of ochratoxin: Current views of the European Scientific Committee on Food, the JECFA and the Codex Committee on Food Additives and Contaminants. Adv. Exp. Med. Biol. 504, 249–255.[ISI][Medline]

Wei, X., and Sulik, K. K. (1993). Pathogenesis of craniofacial and body wall malformations induced by ochratoxin A in mice. Am. J. Med. Genet. 47(6), 862–871.[ISI][Medline]