Abstract
The kidney is one of the main targets of xenobiotic-induced toxicity, but early detection of renal damage is difficult. Recently, several novel biomarkers of nephrotoxicity have been identified by transcription profiling, including kidney injury molecule-1 (Kim-1), lipocalin-2, tissue inhibitor of metalloproteinases-1 (Timp-1), clusterin, osteopontin (OPN), and vimentin, and suggested as sensitive endpoints for acute kidney injury in vivo. However, it is not known if these cellular marker molecules may also be useful to predict chronic nephrotoxicity or to detect nephrotoxic effects in vitro. In this study, a panel of new biomarkers of renal toxicity was assessed via quantitative real-time PCR, immunohistochemistry, and immunoblotting in rats treated with the nephrotoxin ochratoxin A (OTA) for up to 90 days and in rat proximal tubule cells (NRK-52E) treated with OTA in vitro. Repeated administration of OTA to male F344/N rats for 14, 28, or 90 days resulted in a dose- and time-dependent increase in the expression of Kim-1, Timp-1, lipocalin-2, OPN, clusterin, and vimentin. Changes in gene expression were found to correlate with the progressive histopathological alterations and preceded effects on traditional clinical parameters indicative of impaired kidney function. Induction of Kim-1 messenger RNA expression was the earliest and most prominent response observed, supporting the use of this marker as sensitive indicator of chronic kidney injury. In contrast, no significant increase in the expression of putative marker genes and proteins were evident in NRK-52E cells after exposure to OTA for up to 48 h, suggesting that they may not be suitable endpoints for sensitive detection of nephrotoxic effects in vitro.
The kidney is one of the primary sites of xenobiotic-induced toxicity (Hart and Kinter, 2005). The organ-specific toxicity is in part attributable to the high blood flow rate, which causes delivery of high concentrations of xenobiotics to the kidneys. The proximal tubule epithelium is particularly vulnerable to nephrotoxic injury, as these cells express a variety of transporters, which enable active uptake and intracellular accumulation of toxic parent compounds or metabolites. In addition, proximal tubular epithelial cells are highly metabolically active and can bioactivate relatively nontoxic compounds into reactive intermediates, which may cause toxicity via damage to cellular macromolecules (Hart and Kinter, 2005).
Chemical injury may induce cell death within the renal proximal tubule epithelium, subsequently leading to cell dedifferentiation, migration, proliferation, and finally redifferentiation to restore a fully functional epithelial barrier (Bonventre, 2003). However, due to its functional reserve, minor effects on kidney function are difficult to detect. Thus, traditional markers of nephrotoxicity such as increased blood urea nitrogen (BUN) or serum creatinine are rather insensitive and only indicate damage when 70−80% of the renal epithelial mass has been lost. Release of the brush border enzyme γ-glutamyl transferase (γ-GT) or the lysosomal enzyme N-acetyl-β-D-glucosaminidase (NAG) from damaged renal cells into the urine is presently regarded as sensitive indicators of tubule injury. In addition, the urinary concentration of some proteins normally excluded from filtration or reabsorbed by the tubules may serve as indicator of kidney function. For example, increased excretion of albumin and β2-microglobulin is indicative of dysfunction of the proximal tubules due to inhibition of endocytotic protein uptake or cellular damage (Hart and Kinter, 2005). However, detection of enzymes and other proteins can be difficult due to their instability and highly variable levels in urine.
Therefore, there is a strong need for the identification and validation of more sensitive and reliable biomarkers, which may be used for improved prediction of kidney injury during drug development and chemical safety testing. Recently, several candidate gene-based markers for kidney injury have been identified using toxicogenomics approaches. Because changes in messemger RNA (mRNA) expression are considered to be one of the earliest events, which may occur in response to cellular stress and/or tissue damage, it has been speculated that these biomarkers might help to predict adverse effects before damage is indicated by the current gold standard, that is, clinical chemistry and histopathology. In various models of acute kidney injury, elevations in the expression of kidney injury molecule-1 (Kim-1), osteopontin (OPN), lipocalin-2/NGAL, tissue inhibitor of metalloproteinases-1 (Timp-1), clusterin, vimentin and heme oxygenase 1 (HO-1) have been reported to detect injury before the onset of major histopathological changes (Amin et al., 2004; Davis et al., 2004; Thukral et al., 2005). From these studies, it appears that Kim-1 and lipocalin-2/NGAL may be the most promising candidates for new biomarkers, because their mRNA expression has been shown to increase dramatically following acute kidney injury (Kharasch et al., 2006; Mishra et al., 2003). Furthermore, it was demonstrated that Kim-1 and lipocalin-2/NGAL are specifically induced at the target site of toxicity in both animal models and various human renal diseases involving acute injury of the proximal tubule epithelium (Ding et al., 2007; Han et al., 2002; Ichimura et al., 2004; Mishra et al., 2003, 2004). Similarly, target site specific changes in the expression of clusterin and OPN have been observed in response to a variety of models of renal injury (Gobe et al., 1995; Iguchi et al., 2004; Rosenberg and Silkensen, 1995; Witzgall et al., 1994; Xie et al., 2001b). Interestingly, several of these candidate nephrotoxicity markers have been shown to be secreted into urine, suggesting that they may also serve as early, noninvasive biomarkers of nephrotoxicity (Hidaka et al., 2002; Ichimura et al., 2004; Khan et al., 2002; Mishra et al., 2003; Vaidya et al., 2006; Xie et al., 2001b).
However, the ability of these putative biomarkers of nephrotoxicity to indicate or even predict mild histopathological changes following long-term exposure to nephrotoxicants still needs to be evaluated. In this study, we investigated the expression of both new and traditional biomarkers of nephrotoxicity in rats treated with low doses of ochratoxin A (OTA) for up to 90 days. The principal goal was to determine if changes in the expression of the newly identified biomarkers correlate with the progressive damage to the kidney as evidenced by histopathology and might even precede traditional endpoints of nephrotoxicity.
Because development of reliable in vitro test systems may contribute to reduction of animal experiments required for drug and chemical safety assessment, an ancillary aim of this study was to determine if these novel kidney biomarkers might also serve as sensitive endpoints for in vitro toxicity testing. Although primary kidney epithelial cells may more adequately reflect the in vivo situation than cell lines, they are more difficult to handle and are therefore not as convenient for routine use in large screening assays. We therefore investigated the effects of subtoxic concentrations of OTA on novel kidney biomarkers in a stable, well-characterized rat cell line derived from the proximal tubule epithelium (NRK-52E) (de Larco and Todaro, 1978).
MATERIAL AND METHODS
Material.
OTA was purchased from Axxora (Grünberg, Germany) and from Prof. Peter Mantle, Imperial College of Sciences, London, UK. Primary antibodies used were goat anti-albumin (M-13; Santa Cruz, Heidelberg, Germany), mouse anti-β2-microglobulin (MorphoSys AbD, Düsseldorf, Germany), goat anti-clusterin β (M-18; Santa Cruz), goat anti-Kim-1 (Immunology Consultants Laboratory, Newberg, OR), goat anti-lipocalin-2/NGAL (R&D Systems, Wiesbaden-Nordenstadt, Germany), goat anti-OPN (P-18; Santa Cruz), mouse anti-OPN (AKm2A1; Santa Cruz), rabbit anti-Timp-1 (Acris Antibodies, Hiddenhausen, Germany), and mouse anti-vimentin (V9; Santa Cruz). Unless otherwise indicated, all other chemicals were from Roth (Karlsruhe, Germany).
Animal experiment.
Tissue and urine samples for the study were taken from a previous animal experiment with OTA, in which male F344 rats had been treated with 0, 21, 70, or 210 μg/kg bw OTA for 14, 28, or 90 days (Rached et al., 2007). Briefly, animals were administered OTA dissolved in corn oil by gavage, 5 days/week. For urine collection, rats were transferred to metabolic cages 48 h prior to necropsy. During urine collection (20 h), animals were fasted but allowed free access to drinking water. An aliquot of urine was immediately used for clinical chemistry analyses. The remaining urine was stored at −20°C until further analysis. After urine collection, blood samples were drawn from the retro-orbital plexus under light isoflurane anesthesia. Blood and urine samples were transferred on ice to RCC Ltd. (Füllinsdorf, Switzerland) for clinical biochemistry and urine analyses. Animals were killed by CO2 asphyxiation and blood withdrawal by cardiac puncture. Kidneys were removed, weighed, and cut longitudinally. One half of the right kidney was fixed in 10% neutral buffered formalin and subsequently embedded in paraffin. The remaining parts of the kidneys were aliquoted, flash frozen in liquid nitrogen, and stored at −80°C.
Cells and treatment.
NRK-52E cells were acquired from the European Collection of Cell Cultures and cultured under standard cell culture conditions (37°C, 5% CO2) in Dulbecco's modified Eagle medium (high glucose) with 10% fetal calf serum, 2mM glutamine, nonessential amino acids and penicillin/streptomycin (PAA Laboratories, Cölbe, Germany). For subcultivation, 1.5 × 105 cells were seeded into 75-cm2 cell culture flasks and grown to subconfluency.
For treatment, cells were seeded into 21-cm2 cell culture dishes at a density of 1.5 × 105 and allowed to grow to 90–100% confluency. OTA (stored at −20°C) was freshly dissolved in ethanol (99%) and diluted with cell culture medium to reach the desired concentrations (final concentration of ethanol in medium did not exceed 0.36%).
Cytotoxicity (MTT assay).
Cells were seeded into 96-well plates at a density 3.5 × 103/well, allowed to grow for 48 h and subsequently treated with OTA for 24 and 48 h. Cytotoxicity was evaluated using the MTT-based in vitro toxicology assay kit (Sigma-Aldrich, Taufkirchen, Germany) according to the manufacturer's instructions. Each assay was performed in three independent experiments carried out in triplicate.
RNA isolation.
RNA was isolated from NRK cells using the RNeasy Mini Kit (Qiagen, Hilden, Germany) including DNase-digestion according to the manufacturer's instructions. For tissue samples, RNA was isolated from aliquots of frozen kidney using the TRIR total RNA isolation reagent (ABgene, Hamburg, Germany) and further purified using the RNeasy Mini Kit including DNase treatment. Isolated RNA was quantified by measuring the absorbance at 260 nm. RNA integrity was controlled by electrophoresis on a 1.2% formaldehyde agarose gel. Purified RNA samples were stored at −80°C until use.
Complimentary DNA synthesis and quantitative realtime PCR.
Complimentary DNA (cDNA) was synthesized from 1 μg RNA using the first Strand Synthesis Kit (ABgene, Hamburg, Germany), diluted 1:5 with H2O and stored at −20°C. Quantitative real-time PCR was performed using the ABI Prism 7000 Sequence Detection System (Applied Biosystems, Darmstadt, Germany) in 25-μl reactions containing 2× mastermix with SYBR Green I (Abgene, Hamburg, Germany), 2.5 μl of cDNA and 70nM of each primer. Amplification was carried out using the following temperature profile: 15 min enzyme activation at 95°C, followed by 40 cycles of 95°C for 15 s, 60°C for 30 s, and 72°C for 30 s. Primer sequences are shown in Table 1. PCR product formation was determined by measuring the fluorescence signal emitted by the incorporation of SYBR Green I into double stranded DNA. Using standard curves, amplification efficiencies of both endogenous reference gene and target genes were assessed. Product specificity was examined by melting curve analysis and electrophoresis on a 5% polyacrylamide gel.
Sequence of Forward and Reverse Primers for Real-Time PCR
| Gene symbol | Gene title | Accession no. | Forward primer 5′–3′ | Reverse primer 5′–3′ | Reference |
| Havr1/Kim-1 | Kidney injury molecule-1 | AF035963 | CGCAGAGAAACCCGACTAAG | CAAAGCTCAGAGAGCCCATC | (Amin et al., 2004) |
| Lcn2 | Lipocalin-2 | NM_130741 | TCTGGGCCTCAAGGATAACAAC | AGACAGGTGGGACCTGAACCA | (Chen et al., 2006) |
| Timp-1 | Tissue inhibitor of metallopeptidase 1 | NM_053819 | ATCAAGATGACTAAGATGCTCAAAGG | GGCCGCGATGAGAAACT | (Bolbrinker et al., 2006) |
| Spp1 | Secreted phosphoprotein 1 (osteopontin) | M99252 | CCAGCACACAAGCAGACGTTT | CAGTCCGTAAGCCAAGCTATCA | (Amin et al., 2004) |
| Vim | Vimentin | NM_031140 | GATGCTCCAGAGGGAGGAAG | AAGGTCAAGACGTGCCAGAG | — |
| Clu | Clusterin | M64723 | CACTACGGGCCTCTGAGCTT | ACGTCCATGGCCTGTTGAG | (Luhe et al., 2003) |
| Hmox | Heme oxygenase-1 | NM_012580 | CTAAGACCGCCTTCCTGCT | CCTCTGGCGAAGAAACTCTG | — |
| Actb | β-Actin | NM_031144 | GGGAAATCGTGCGTGACATT | GCGGCAGTGGCCATCTC | (Depreter et al., 2002) |
| Rplp1 | Ribosomal protein, large, P1 | NM_001007604 | AAAGCAGCTGGTGTCAATGT | GCAGATGAGGCTTCCAATGT | — |
| Gene symbol | Gene title | Accession no. | Forward primer 5′–3′ | Reverse primer 5′–3′ | Reference |
| Havr1/Kim-1 | Kidney injury molecule-1 | AF035963 | CGCAGAGAAACCCGACTAAG | CAAAGCTCAGAGAGCCCATC | (Amin et al., 2004) |
| Lcn2 | Lipocalin-2 | NM_130741 | TCTGGGCCTCAAGGATAACAAC | AGACAGGTGGGACCTGAACCA | (Chen et al., 2006) |
| Timp-1 | Tissue inhibitor of metallopeptidase 1 | NM_053819 | ATCAAGATGACTAAGATGCTCAAAGG | GGCCGCGATGAGAAACT | (Bolbrinker et al., 2006) |
| Spp1 | Secreted phosphoprotein 1 (osteopontin) | M99252 | CCAGCACACAAGCAGACGTTT | CAGTCCGTAAGCCAAGCTATCA | (Amin et al., 2004) |
| Vim | Vimentin | NM_031140 | GATGCTCCAGAGGGAGGAAG | AAGGTCAAGACGTGCCAGAG | — |
| Clu | Clusterin | M64723 | CACTACGGGCCTCTGAGCTT | ACGTCCATGGCCTGTTGAG | (Luhe et al., 2003) |
| Hmox | Heme oxygenase-1 | NM_012580 | CTAAGACCGCCTTCCTGCT | CCTCTGGCGAAGAAACTCTG | — |
| Actb | β-Actin | NM_031144 | GGGAAATCGTGCGTGACATT | GCGGCAGTGGCCATCTC | (Depreter et al., 2002) |
| Rplp1 | Ribosomal protein, large, P1 | NM_001007604 | AAAGCAGCTGGTGTCAATGT | GCAGATGAGGCTTCCAATGT | — |
Sequence of Forward and Reverse Primers for Real-Time PCR
| Gene symbol | Gene title | Accession no. | Forward primer 5′–3′ | Reverse primer 5′–3′ | Reference |
| Havr1/Kim-1 | Kidney injury molecule-1 | AF035963 | CGCAGAGAAACCCGACTAAG | CAAAGCTCAGAGAGCCCATC | (Amin et al., 2004) |
| Lcn2 | Lipocalin-2 | NM_130741 | TCTGGGCCTCAAGGATAACAAC | AGACAGGTGGGACCTGAACCA | (Chen et al., 2006) |
| Timp-1 | Tissue inhibitor of metallopeptidase 1 | NM_053819 | ATCAAGATGACTAAGATGCTCAAAGG | GGCCGCGATGAGAAACT | (Bolbrinker et al., 2006) |
| Spp1 | Secreted phosphoprotein 1 (osteopontin) | M99252 | CCAGCACACAAGCAGACGTTT | CAGTCCGTAAGCCAAGCTATCA | (Amin et al., 2004) |
| Vim | Vimentin | NM_031140 | GATGCTCCAGAGGGAGGAAG | AAGGTCAAGACGTGCCAGAG | — |
| Clu | Clusterin | M64723 | CACTACGGGCCTCTGAGCTT | ACGTCCATGGCCTGTTGAG | (Luhe et al., 2003) |
| Hmox | Heme oxygenase-1 | NM_012580 | CTAAGACCGCCTTCCTGCT | CCTCTGGCGAAGAAACTCTG | — |
| Actb | β-Actin | NM_031144 | GGGAAATCGTGCGTGACATT | GCGGCAGTGGCCATCTC | (Depreter et al., 2002) |
| Rplp1 | Ribosomal protein, large, P1 | NM_001007604 | AAAGCAGCTGGTGTCAATGT | GCAGATGAGGCTTCCAATGT | — |
| Gene symbol | Gene title | Accession no. | Forward primer 5′–3′ | Reverse primer 5′–3′ | Reference |
| Havr1/Kim-1 | Kidney injury molecule-1 | AF035963 | CGCAGAGAAACCCGACTAAG | CAAAGCTCAGAGAGCCCATC | (Amin et al., 2004) |
| Lcn2 | Lipocalin-2 | NM_130741 | TCTGGGCCTCAAGGATAACAAC | AGACAGGTGGGACCTGAACCA | (Chen et al., 2006) |
| Timp-1 | Tissue inhibitor of metallopeptidase 1 | NM_053819 | ATCAAGATGACTAAGATGCTCAAAGG | GGCCGCGATGAGAAACT | (Bolbrinker et al., 2006) |
| Spp1 | Secreted phosphoprotein 1 (osteopontin) | M99252 | CCAGCACACAAGCAGACGTTT | CAGTCCGTAAGCCAAGCTATCA | (Amin et al., 2004) |
| Vim | Vimentin | NM_031140 | GATGCTCCAGAGGGAGGAAG | AAGGTCAAGACGTGCCAGAG | — |
| Clu | Clusterin | M64723 | CACTACGGGCCTCTGAGCTT | ACGTCCATGGCCTGTTGAG | (Luhe et al., 2003) |
| Hmox | Heme oxygenase-1 | NM_012580 | CTAAGACCGCCTTCCTGCT | CCTCTGGCGAAGAAACTCTG | — |
| Actb | β-Actin | NM_031144 | GGGAAATCGTGCGTGACATT | GCGGCAGTGGCCATCTC | (Depreter et al., 2002) |
| Rplp1 | Ribosomal protein, large, P1 | NM_001007604 | AAAGCAGCTGGTGTCAATGT | GCAGATGAGGCTTCCAATGT | — |
Gene expression changes relative to untreated controls were determined by the 2−ΔΔCt method (see “Critical factors for Successful Real-time PCR,” Qiagen). Samples were amplified in duplicate and normalized against β-actin. Results are presented as mean fold change in mRNA expression of 5 animals per dose group compared with control animals respectively treated cells compared with untreated cells determined by three independent in vitro experiments.
Protein preparation.
To prepare urinary proteins, urine was centrifuged at 15,800 × g and 4°C for 15 min to pellet cells and cell debris. The supernatant was diluted with water to adjust for differences in creatinine concentrations. Urine was subsequently diluted in 6× sample buffer (12% sodium dodecyl sulfate [SDS], 50% glycerol, 25% 2-mercaptoethanol, 0.0125% bromphenol blue, 250mM Tris, pH 6.8). To concentrate urinary proteins, 300 μl of urine normalized on creatinine were mixed with an equal volume of 20% ice-cold trichloroacetic acid and incubated on ice for 15 min. Proteins were pelleted by centrifugation at 15,800 × g and 4°C for 15 min. The supernatant was removed and the pellet was washed two times with 300 μl of ice-cold ethanol (99%). After complete removal of ethanol, the pellet was dissolved in 50 μl of 1× sample buffer (2% SDS, 10% glycerol, 5% 2-mercaptoethanol, 0.002% bromphenol blue, 62.5mM Tris–HCl, pH 6.8), and pH was adjusted to 6.8 with 1M Tris, pH 8.8. Samples were stored at −20°C until use.
Immunoblotting.
Ten microliters of concentrated urinary proteins in sample buffer or 20 μl of urine sample normalized on creatinine were separated by SDS–polyacrylamide gel electrophoresis (PAGE) and transferred onto nitrocellulose membranes at 100 V and 4°C for 1 h using transfer buffer (25mM Tris, 192mM glycine, 20% methanol). For the detection of albumin, clusterin, Kim-1, and lipocalin-2/NGAL, membranes were blocked in 5% nonfat, dry milk in TBST buffer (50mM Tris–HCl, 150mM NaCl, pH 7.4, 0.1% Tween-20) for 2 h at room temperature (RT) and subsequently incubated with primary antibody diluted in blocking buffer overnight at 4°C. After extensive washing with TBST, membranes were incubated with horse radish peroxidase (HRP)-conjugated secondary antibody (donkey anti-goat; Santa Cruz) at a dilution of 1:5000 in blocking buffer for 1 h at RT. Following repeated wash steps, membranes were developed using the enhanced chemiluminescence (ECL) western blotting detection system (GE Healthcare, Freiburg, Germany) and exposed to hyperfilm (GE Healthcare). For detection of β2-microglobulin, membranes were blocked in 5% milk in phosphate-buffered saline (PBS) for 2 h at RT and incubated with primary antibody in blocking buffer with 0.1% Tween-20 overnight at 4°C. After washing with 5% milk/PBST, membranes were incubated with secondary antibody (goat anti-mouse, Santa Cruz; 1:5000) for 1 h at RT. Membranes were extensively washed with 5% milk/PBST, PBST, and PBS before proteins were detected using ECL detection system (see above). To detect urinary OPN, proteins were blotted onto either nitrocellulose or polyvinylidene fluoride membrane, blocked in 5% milk in TBST, and incubated with a goat anti-OPN antibody in 1% milk in TBST. Densitometry was performed using the Bio-Rad Gel Doc 2000 system. Changes in protein expression in response to treatment are presented as relative to controls.
Immunohistochemistry.
Kidney sections (5 μm) were prepared from formalin fixed, paraffin embedded tissue blocks and mounted onto glass slides. Sections were deparaffinized, rehydrated, and washed in PBS. Heat-induced antigen retrieval was achieved by 4 min autoclaving in 10mM citrate buffer, pH 6.0. For detection of Timp-1 and lipocalin-2/NGAL, sections were treated with 0.1% trypsin for 3 min at 37°C. After washing the sections in PBS and H2O, endogenous peroxidase was blocked by incubation with 3% H2O2 in PBS for 15 min. Sections were subsequently blocked with 10% goat serum or 5% donkey serum for 1.5 h, followed by 0.001% avidin in PBS for 15 min and 0.001% biotin for 15 min, with several washes in PBS in between. Then, sections were incubated with 2.5–5 μg/ml primary antibody diluted in 1% BSA in PBS for 1 h at RT. After three wash steps, sections were incubated with biotinylated secondary antibody (goat anti-mouse, goat anti-rabbit, or donkey anti-goat, Santa Cruz), diluted 1:100 in PBS for 1 h at RT and subsequently washed with PBS. Following 30 min incubation with Streptavidin–HRP (Becton Dickinson, Heidelberg, Germany), peroxidase activity was visualized using 3,3′-diaminobenzidine tetrahydrochloride (Becton Dickinson) for 1–5 min. The sections were counterstained with hematoxylin, dehydrated and mounted in Eukitt mounting medium (Sigma-Aldrich, Taufkirchen, Germany).
Statistical analysis.
Data are expressed as mean ± SD. Statistical analysis was performed by ANOVA and Dunnett's test. Values significantly different from control are indicated as *p < 0.05, **p < 0.01, and ***p < 0.001.
RESULTS
In Vivo Study
General observations.
As previously reported, OTA treatment resulted in histopathological changes, consisting of cell degeneration, basophilia, and karyomegaly in the S3 segment of the proximal tubule epithelium (PST), depending on dose and treatment time (Rached et al., 2007). Routine analysis of clinical chemistry parameters in blood and urine showed no effects on urinary volume, protein, glucose, or creatinine. In addition, no changes in serum urea (BUN) levels were observed throughout the study. In high-dose animals, urinary NAG activity was significantly increased after 28 and 90 days (1.7-fold and 1.3-fold, respectively), and elevated serum creatinine levels (1.2-fold) were observed after 90 days. At the same time, the activity of γ-GT was significantly decreased in urine of both mid-dose and high-dose animals to 60% and 20% of controls, respectively, possibly indicating loss of brush border or functional changes within the proximal tubule epithelium. Based on the histopathological observations (Table 2), a no-observed-effect-level (NOEL) of 21 μg/kg bw OTA was established.
Summary of Histopathological Observations in Kidneys of Male F344/N Rats Treated with 0, 21, 70, or 210 μg/kg bw OTA for 14, 28, or 90 Days
| Histopathological change | Interval (days) | OTA (μg/kg bw) | |||
| 0 | 21 | 70 | 210 | ||
| Tubular basophilia | 14 | + (3/5) | + (4/5) | + (2/5) | ++ (5/5) |
| 28 | + (1/5) | + (2/5) | + (2/5) | +++ (5/5) | |
| 90 | + (2/5) | + (5/5) | ++ (5/5) | +++ (5/5) | |
| Tubular degeneration | 14 | − | − | −/+ (1/5) | ++++ (5/5) |
| 28 | − | − | + (5/5) | ++ (5/5) | |
| 90 | − | − | ++(+) (5/5) | + (5/5) | |
| Nuclear enlargement | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | ++ (5/5) | ++++ (5/5) | |
| Cell proliferation | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | +++ (5/5) | ++++ (5/5) | |
| Histopathological change | Interval (days) | OTA (μg/kg bw) | |||
| 0 | 21 | 70 | 210 | ||
| Tubular basophilia | 14 | + (3/5) | + (4/5) | + (2/5) | ++ (5/5) |
| 28 | + (1/5) | + (2/5) | + (2/5) | +++ (5/5) | |
| 90 | + (2/5) | + (5/5) | ++ (5/5) | +++ (5/5) | |
| Tubular degeneration | 14 | − | − | −/+ (1/5) | ++++ (5/5) |
| 28 | − | − | + (5/5) | ++ (5/5) | |
| 90 | − | − | ++(+) (5/5) | + (5/5) | |
| Nuclear enlargement | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | ++ (5/5) | ++++ (5/5) | |
| Cell proliferation | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | +++ (5/5) | ++++ (5/5) | |
Note. Abbrevations: −, lesion not observed; +, minimal, ++ mild, +++ moderate, and ++++ high severity of lesion (Rached et al., 2007).
Summary of Histopathological Observations in Kidneys of Male F344/N Rats Treated with 0, 21, 70, or 210 μg/kg bw OTA for 14, 28, or 90 Days
| Histopathological change | Interval (days) | OTA (μg/kg bw) | |||
| 0 | 21 | 70 | 210 | ||
| Tubular basophilia | 14 | + (3/5) | + (4/5) | + (2/5) | ++ (5/5) |
| 28 | + (1/5) | + (2/5) | + (2/5) | +++ (5/5) | |
| 90 | + (2/5) | + (5/5) | ++ (5/5) | +++ (5/5) | |
| Tubular degeneration | 14 | − | − | −/+ (1/5) | ++++ (5/5) |
| 28 | − | − | + (5/5) | ++ (5/5) | |
| 90 | − | − | ++(+) (5/5) | + (5/5) | |
| Nuclear enlargement | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | ++ (5/5) | ++++ (5/5) | |
| Cell proliferation | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | +++ (5/5) | ++++ (5/5) | |
| Histopathological change | Interval (days) | OTA (μg/kg bw) | |||
| 0 | 21 | 70 | 210 | ||
| Tubular basophilia | 14 | + (3/5) | + (4/5) | + (2/5) | ++ (5/5) |
| 28 | + (1/5) | + (2/5) | + (2/5) | +++ (5/5) | |
| 90 | + (2/5) | + (5/5) | ++ (5/5) | +++ (5/5) | |
| Tubular degeneration | 14 | − | − | −/+ (1/5) | ++++ (5/5) |
| 28 | − | − | + (5/5) | ++ (5/5) | |
| 90 | − | − | ++(+) (5/5) | + (5/5) | |
| Nuclear enlargement | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | ++ (5/5) | ++++ (5/5) | |
| Cell proliferation | 14 | − | − | − | + (5/5) |
| 28 | − | − | + (5/5) | +++ (5/5) | |
| 90 | − | − | +++ (5/5) | ++++ (5/5) | |
Note. Abbrevations: −, lesion not observed; +, minimal, ++ mild, +++ moderate, and ++++ high severity of lesion (Rached et al., 2007).
Changes in the expression of putative biomarkers of nephrotoxicity in kidneys of OTA-treated rats.
Repeated administration of 70 or 210 μg/kg bw OTA to male rats resulted in a dose- and time-dependent increase in the expression of genes suggested as candidate biomarkers of renal injury (Fig. 1). Changes in the expression of Kim-1, lipocalin-2, Timp-1, OPN, and clusterin were found to correlate with the progressive histopathological alterations observed in kidneys of OTA-treated animals (Table 2) and preceded changes of clinical parameters indicative of impaired kidney function. In contrast to the mid and high dose, no alterations in gene expression were evident in kidneys of rats treated with 21 μg/kg bw OTA, consistent with the NOEL in this study.
Changes in mRNA expression of putative biomarkers of nephrotoxicity in kidneys of male rats repeatedly dosed with 0, 21, 70, or 210 μg/kg bw OTA for 14(a), 28(b), or 90(b) days. Data are presented as mean fold change ± SD compared with control animals (n = 5). Statistical analysis was performed by ANOVA and Dunnett's test (*p < 0.05, **p < 0.01, ***p < 0.001).
Kim-1, lipocalin-2/NGAL, and Timp-1 were among the most upregulated genes (about 10-fold increase in both mid and high dose). Induction of Kim-1 gene expression was one of the earliest responses of the kidney following OTA treatment, evident as early as 28 days in four of five animals exposed to 70 μg/kg bw OTA and after 14 days in all high-dose animals, where only subtle histopathological changes were observed. Furthermore, the degree of Kim-1 mRNA increase in the individual animals correlated well with the extent of tissue damage in the kidney. Expression levels of OPN and clusterin were also markedly increased (≥ fivefold) following treatment with 210 μg/kg bw OTA for 28 days. However, alterations in lipocalin-2/NGAL, OPN, Timp-1, and clusterin gene expression were only detectable after 90 days in mid-dose animals. In this study, only minor changes (> twofold) in HO-1 and vimentin gene expression were detected. Moreover, HO-1 was not consistently upregulated throughout the study and no changes in vimentin gene expression were evident in mid-dose animals (70 μg/kg bw OTA) as compared with controls.
Results obtained by quantitative realtime PCR were subsequently confirmed by immunohistochemistry (Fig. 2). Consistent with gene expression analyses, Kim-1 was detected as early as after 14 days in kidneys of high-dose animals (no shown) and after 90 days in both mid- and high-dose animals (Fig. 2a). Kim-1 was undetectable in proximal tubules located in the outer stripe of the outer medulla (OSOM) and medullary rays of control kidneys, but it occurred in epithelial cells of S3 tubules that were affected by OTA toxicity as seen by cell degeneration and regeneration. Kim-1 was located in the cytoplasm or at the apical side of flattened, dedifferentiated cells.
Expression of Kim-1 (a), vimentin (b), clusterin (c), OPN (d), Timp-1 (e), and lipocalin-2/NGAL (f) in kidneys of male F344/N rats treated with 0, 21, 70, or 210 μg/kg bw OTA for 90 days. (a) Treatment with up to 210 μg/kg bw OTA resulted in marked Kim-1 induction in tubules with cellular degeneration (thin arrow, →) and regeneration (thick arrow, fx1). Kim-1 was detected at the apical side or in the cytoplasm of flattened, dedifferentiated cells. (b) Vimentin is not expressed in tubules of control animals and rats exposed to 21 μg/kg bw OTA. Treatment with the two higher doses resulted in a dose-dependent increase in protein expression, predominantly in flattened epithelial cells of proximal tubules located in the OSOM. (c) In controls and rats exposed to 21 μg/kg bw OTA, clusterin is not detected in proximal tubules located in the OSOM. In contrast, clusterin expression is observed after 90 days of treatment with 70 or 210 μg/kg bw OTA in proximal straight tubules showing marked signs of OTA toxicity, like cell degeneration and regeneration. (d) OPN is constitutively expressed in epithelia of proximal tubules, where it is found in small vesicular structures and along the brush border. Repeated administration of 70 or 210 μg/kg bw OTA resulted in increased expression of OPN in regenerating proximal tubule epithelial cells. In these cells, OPN was found in vesicles or at the apical (luminal) side. (e) In proximal tubules in the OSOM, Timp-1 is normally expressed at cell-cell contacts and at the apical side of the cells. OTA treatment did not have a marked effect on Timp-1 expression, except for some vesicular staining in single epithelial cells of affected proximal straight tubules (arrowhead). (f) Lipocalin-2/NGAL is found at the apical side of proximal tubules in the deep OSOM. No OTA-mediated effects were observed at the target site of OTA toxicity. Original magnification 200×.
Vimentin was not detected in renal tubular epithelial cells of control animals or rats exposed to 21 μg/kg bw OTA. In contrast, positive staining of proximal straight tubules, the target sites of OTA toxicity, was observed in kidneys of rats exposed to 70 and 210 μg/kg bw OTA for 90 days, with a clear increase relative to dose. Interestingly, changes in vimentin protein expression were detectable at lower doses than required to induce significant changes in gene expression. Vimentin expression was predominantly observed in flattened cells that lacked a brush border—characteristics of dedifferentiated cells (Fig. 2b). Clusterin was constitutively expressed in distal convoluted tubules, localized at the apical side of the cells (not shown), whereas proximal tubules located within the OSOM were not labeled by the antibody. However, repeated administration of 70 or 210 μg/kg bw OTA for 90 days resulted in a dose-dependent increase in clusterin expression in proximal tubule cells located in the OSOM and medullary rays. Affected tubules showed clear signs of degeneration and regeneration (Fig. 2c). Expression of OPN was evident in proximal tubular epithelial cells of both treated and untreated animals, where positive OPN staining was observed both in small vesicular structures and at the brush border. Treatment with 70 or 210 μg/kg bw OTA for 90 days resulted in strong apical staining of OPN in epithelial cells within the OSOM that showed signs of regeneration, such as flattened epithelium and loss of brush border. In addition, several strongly affected tubules showed high amounts of vesicular OPN, indicating increased OPN expression at this site, consistent with the increase in OPN mRNA levels (Fig. 2d).
Timp-1 was detected in proximal tubules of both kidney cortex and OSOM. In cortical tubules, the protein was expressed in vesicular structures. In addition, protein droplets and debris in tubular lumina were strongly stained, in line with reports by Eddy and Giachelli (1995) obtained in proteinuric animals. Interestingly, Timp-1 was exclusively detected at cell-cell contacts and at the apical side of the cells located within the OSOM (Fig. 2e). Due to the high basal expression of Timp-1, clear changes of Timp-1 protein expression in response to OTA exposure were not evident, except for single epithelial cells of affected tubules, which contained many positively stained vesicles that did not occur in control kidneys (Fig. 2e). Using a specific antibody against rat-lipocalin-2/NGAL, the protein was exclusively detected at the apical side of some proximal tubules in the cortex and in the deep OSOM of all animals. However no treatment-related effects on lipocalin-2/NGAL expression were observed at the target site of OTA toxicity (Fig. 2f).
Urinary biomarkers for tubular toxicity.
Changes in the levels of urinary proteins were studied by SDS–PAGE and western blot as previously described (Hidaka et al., 2002; Ichimura et al., 1998; Khan et al., 2002; Mishra et al., 2004). Coomassie-staining of separated urinary proteins on polyacrylamide gels showed a clear dose-dependent increase in the amount of a 70 kDa protein that was subsequently confirmed to be albumin. Quantification of urinary albumin by immunoblotting showed a statistically significant increase (3-fold relative to control) in urine of rats treated with 210 μg/kg bw OTA as early as 14 days after start of treatment (Fig. 3a). Albuminuria can result from both impaired glomerular filtration and decreased protein uptake by the proximal tubule epithelium. Therefore, tubular uptake of low molecular weight proteins was studied by measuring the urinary concentration of β2-microglobulin (12 kDa). As expected from the histopathological observations, both microalbuminuria and an elevated excretion of β2-microglobulin were evident in high-dose animals (Fig. 3b), indicating impaired proximal tubular function rather than altered glomerular filtration (Hart and Kinter, 2005). Similar effects were observed at the later time-points, although some changes were not found to be statistically significant due to the high variability in urinary protein levels. Administration of 70 μg/kg bw OTA resulted in minor changes in albumin excretion and had no effect on urinary β2-microglobulin levels throughout the study.
Expression of albumin (a), β2-microglobulin (b), Kim-1 (c), and clusterin (d) in concentrated urine of male F344/N rats treated repeatedly with 0, 21, 70, or 210 μg/kg bw OTA for 14, 28, or 90 days. Densitometry data are presented as fold change of protein expression in control animals. Statistical analysis was performed by ANOVA and Dunnett's test (*p < 0.05, **p < 0.01, ***p < 0.001).
Several studies indicate that increased secretion of Kim-1, lipocalin-2/NGAL, clusterin, and OPN into urine may occur in response to tubule cellular injury (Hidaka et al., 2002; Ichimura et al., 1998; Khan et al., 2002; Mishra et al., 2004). Because all these marker genes were found to be upregulated in kidneys of OTA-treated rats as compared with controls, we were interested to determine if urinary concentrations were also altered in response to OTA treatment. Consistent with both the induction of mRNA expression and the positive immunolabeling of Kim-1 in kidneys of high-dose rats, a significant increase in Kim-1 excretion was detected as early as after 14 days in urine of these animals. After 90 days, elevated levels of Kim-1 were also found in mid-dose rats (Fig. 3c). In contrast to urine of male Wistar and Sprague Dawley rats, lipocalin-2/NGAL was undetectable in urine of male F344 rats (data not shown). In addition, OPN could not be detected in urine of different rat strains using commercially available antibodies. Clusterin was present in urine of both controls and OTA-treated animals. However, no treatment-related effect on urinary clusterin levels was observed (Fig. 3d).
Expression of putative biomarkers in OTA-treated NRK-52E cells.
A small decrease in cell viability (20–30%) was observed in NRK cells after treatment with OTA at concentrations up to 30 μM (Fig. 4). Based on these results, 3 treatment concentrations (10, 20, 30 μM), ranging from noncytotoxic to weakly cytotoxic were chosen for determination of gene expression of putative marker genes. In contrast to modulation of marker gene expression in vivo, treatment of NRK-52E cells with up to 30 μM OTA for 24 or 48h did not result in marked changes in the mRNA expression of candidate kidney biomarkers (Table 3) with the exception of HO-1, which was slightly induced in response to OTA treatment. Kim-1, which was found to be the most sensitive marker for nephrotoxicity in vivo, was expressed at low to undetectable levels in vitro, and no increase in Kim-1 mRNA was observed after treatment with OTA. Similarly, expression of lipocalin-2 and Timp-1 was not significantly affected by OTA exposure. Inconsistent changes in both mRNA and protein expression of putative biomarkers were also observed following treatment with a variety of other nephrotoxins that target the proximal tubule epithelium, including cisplatin, cadmium chloride, and potassium bromate (data not shown).
Changes in Gene Expression of Putative Biomarkers of Nephrotoxicity in NRK-52E Cells after Treatment with OTA for 24 or 48 h
| Fold deregulation | ||||
| OTA (μM) | ||||
| Gene title | Interval (h) | 10 | 20 | 30 |
| Lipocalin-2 | 24 | 0.8 ± 0.1 | 0.5 ± 0.1 | 0.5 ± 0.1 |
| 48 | 0.8 ± 0.1 | 0.4 ± 0.1 | 0.3 ± 0.1 | |
| Osteopontin | 24 | 0.8 ± 0.2 | 0.7 ± 0.2 | 0.7 ± 0.2 |
| 48 | 0.8 ± 0.4 | 1.3 ± 0.8 | 2.7 ± 1.3 | |
| Clusterin | 24 | 1.5 ± 0.5 | 2.0 ± 1.1 | 1.6 ± 0.9 |
| 48 | 1.7 ± 0.4 | 1.8 ± 0.5 | 0.8 ± 0.4 | |
| Vimentin | 24 | 1.1 ± 0.1 | 1.3 ± 0.1 | 1.0 ± 0.1 |
| 48 | 1.2 ± 0.1 | 1.3 ± 0.1 | 1.4 ± 0.1 | |
| Timp-1 | 24 | 0.9 ± 0.1 | 0.9 ± 0.5 | 0.5 ± 0.2 |
| 48 | 1.0 ± 0.1 | 1.1 ± 0.4 | 0.8 ± 0.2 | |
| Heme oxygenase-1 | 24 | 1.1 ± 0.1 | 1.7 ± 0.2 | 3.9 ± 1.0 |
| 48 | 1.4 ± 0.1 | 2.3 ± 0.5 | 5.0 ± 2.6 | |
| Fold deregulation | ||||
| OTA (μM) | ||||
| Gene title | Interval (h) | 10 | 20 | 30 |
| Lipocalin-2 | 24 | 0.8 ± 0.1 | 0.5 ± 0.1 | 0.5 ± 0.1 |
| 48 | 0.8 ± 0.1 | 0.4 ± 0.1 | 0.3 ± 0.1 | |
| Osteopontin | 24 | 0.8 ± 0.2 | 0.7 ± 0.2 | 0.7 ± 0.2 |
| 48 | 0.8 ± 0.4 | 1.3 ± 0.8 | 2.7 ± 1.3 | |
| Clusterin | 24 | 1.5 ± 0.5 | 2.0 ± 1.1 | 1.6 ± 0.9 |
| 48 | 1.7 ± 0.4 | 1.8 ± 0.5 | 0.8 ± 0.4 | |
| Vimentin | 24 | 1.1 ± 0.1 | 1.3 ± 0.1 | 1.0 ± 0.1 |
| 48 | 1.2 ± 0.1 | 1.3 ± 0.1 | 1.4 ± 0.1 | |
| Timp-1 | 24 | 0.9 ± 0.1 | 0.9 ± 0.5 | 0.5 ± 0.2 |
| 48 | 1.0 ± 0.1 | 1.1 ± 0.4 | 0.8 ± 0.2 | |
| Heme oxygenase-1 | 24 | 1.1 ± 0.1 | 1.7 ± 0.2 | 3.9 ± 1.0 |
| 48 | 1.4 ± 0.1 | 2.3 ± 0.5 | 5.0 ± 2.6 | |
Note. Data are presented as mean fold change ± SD compared with controls (3 independent experiments).
Changes in Gene Expression of Putative Biomarkers of Nephrotoxicity in NRK-52E Cells after Treatment with OTA for 24 or 48 h
| Fold deregulation | ||||
| OTA (μM) | ||||
| Gene title | Interval (h) | 10 | 20 | 30 |
| Lipocalin-2 | 24 | 0.8 ± 0.1 | 0.5 ± 0.1 | 0.5 ± 0.1 |
| 48 | 0.8 ± 0.1 | 0.4 ± 0.1 | 0.3 ± 0.1 | |
| Osteopontin | 24 | 0.8 ± 0.2 | 0.7 ± 0.2 | 0.7 ± 0.2 |
| 48 | 0.8 ± 0.4 | 1.3 ± 0.8 | 2.7 ± 1.3 | |
| Clusterin | 24 | 1.5 ± 0.5 | 2.0 ± 1.1 | 1.6 ± 0.9 |
| 48 | 1.7 ± 0.4 | 1.8 ± 0.5 | 0.8 ± 0.4 | |
| Vimentin | 24 | 1.1 ± 0.1 | 1.3 ± 0.1 | 1.0 ± 0.1 |
| 48 | 1.2 ± 0.1 | 1.3 ± 0.1 | 1.4 ± 0.1 | |
| Timp-1 | 24 | 0.9 ± 0.1 | 0.9 ± 0.5 | 0.5 ± 0.2 |
| 48 | 1.0 ± 0.1 | 1.1 ± 0.4 | 0.8 ± 0.2 | |
| Heme oxygenase-1 | 24 | 1.1 ± 0.1 | 1.7 ± 0.2 | 3.9 ± 1.0 |
| 48 | 1.4 ± 0.1 | 2.3 ± 0.5 | 5.0 ± 2.6 | |
| Fold deregulation | ||||
| OTA (μM) | ||||
| Gene title | Interval (h) | 10 | 20 | 30 |
| Lipocalin-2 | 24 | 0.8 ± 0.1 | 0.5 ± 0.1 | 0.5 ± 0.1 |
| 48 | 0.8 ± 0.1 | 0.4 ± 0.1 | 0.3 ± 0.1 | |
| Osteopontin | 24 | 0.8 ± 0.2 | 0.7 ± 0.2 | 0.7 ± 0.2 |
| 48 | 0.8 ± 0.4 | 1.3 ± 0.8 | 2.7 ± 1.3 | |
| Clusterin | 24 | 1.5 ± 0.5 | 2.0 ± 1.1 | 1.6 ± 0.9 |
| 48 | 1.7 ± 0.4 | 1.8 ± 0.5 | 0.8 ± 0.4 | |
| Vimentin | 24 | 1.1 ± 0.1 | 1.3 ± 0.1 | 1.0 ± 0.1 |
| 48 | 1.2 ± 0.1 | 1.3 ± 0.1 | 1.4 ± 0.1 | |
| Timp-1 | 24 | 0.9 ± 0.1 | 0.9 ± 0.5 | 0.5 ± 0.2 |
| 48 | 1.0 ± 0.1 | 1.1 ± 0.4 | 0.8 ± 0.2 | |
| Heme oxygenase-1 | 24 | 1.1 ± 0.1 | 1.7 ± 0.2 | 3.9 ± 1.0 |
| 48 | 1.4 ± 0.1 | 2.3 ± 0.5 | 5.0 ± 2.6 | |
Note. Data are presented as mean fold change ± SD compared with controls (3 independent experiments).
Cell viability of NRK-52E cells treated 24 or 48 h with various concentrations of OTA. Statistical analysis was performed by ANOVA and Dunnett's test (*p < 0.05, **p < 0.01).
DISCUSSION
In recent years, several putative biomarkers have been identified by gene expression analysis in various models of drug-induced nephrotoxicity (Amin et al., 2004; Davis et al., 2004; Kharasch et al., 2006; Thukral et al., 2005). Although these novel markers have been demonstrated to be sensitive, specific markers of acute renal tubular injury, their potential to detect subtle changes in the kidney following long-term exposure to nephrotoxic agents has not been investigated so far. In this study, we observed upregulation of mRNA expression of the majority of the selected candidate biomarkers in kidneys of rats following treatment with OTA in the absence of effects on traditional clinical chemistry markers of nephrotoxicity. Moreover, both the time-course and the degree of the response correlated well with the severity of tissue injury.
Changes in the expression of Kim-1 were one of the earliest and most prominent effects observed in kidneys of animals treated with 70 or 210 μg/kg bw OTA, occurring at time-points at which only minor histopathological changes in the form of single tubular degeneration/regeneration were evident. Although the specific functions of Kim-1 are still unknown, upregulation of this gene is thought to be associated with proliferation/regeneration and repair in response to toxicity or disease (Ichimura et al., 1998), consistent with the histopathological alterations observed following treatment with OTA. Similarly, lipocalin-2/NGAL, OPN, vimentin, Timp-1, and clusterin are proposed to be involved in the process of regeneration (Gobe et al., 1995; Huang et al., 2001; Iguchi et al., 2004; Mishra et al., 2003; Wallin et al., 1992; Witzgall et al., 1994; Xie et al., 2001a). However, significant alterations in mRNA levels of these putative biomarkers occurred at later time-points as compared with Kim-1, indicating that they might not be as sensitive to detect minor kidney damage.
Changes in marker gene expression were subsequently confirmed by immunohistochemistry. In accordance with other studies, Kim-1 occurred in dedifferentiated proximal tubule epithelial cells (Ichimura et al., 1998, 2004), which confirms a role for this protein in proliferation and tissue repair. Induction of vimentin protein was a sensitive indicator of cell dedifferentiation and tubular regeneration following OTA exposure. Similarly, immunolabeling of OPN demonstrated increased expression of OPN at sites of proximal tubular regeneration. OPN protein was found in intracellular vesicles and at the apical side of epithelial cells within affected tubules, supporting the hypothesis that this protein takes part in tissue repair (Persy et al., 1999). Consistent with other studies (Gobe et al., 1995; Witzgall et al., 1994), increased expression of clusterin occurred in tubules showing clear signs of cell death and regeneration. The strong clusterin immunolabeling in flattened, regenerating cells is in line with studies indicating a role for clusterin in promoting cell–cell and cell–substratum interactions, thereby facilitating tissue remodeling after damage (Rosenberg and Silkensen, 1995). In contrast, only small changes in protein expression of Timp-1 and no changes in lipocalin-2/NGAL protein expression were observed in kidneys of OTA-treated rats. Increased mRNA and protein expression of Timp-1 was reported during the development of interstitial fibrosis (Ahmed et al., 2007; Eddy and Giachelli, 1995). However, because Timp-1 protein was not detected in the tubulointerstitial space of OTA-treated rats, it might have another role in OTA-mediated progressive nephrotoxicity. Timp-1 possesses both anti-apoptotic and growth-stimulating properties (Lambert et al., 2004), and overexpression of this protein has been associated with a high proliferation rate and poor prognosis in malignant tissues (Kallakury et al., 2001; Miyata et al., 2004). Therefore, Timp-1 induction by OTA might be related to the induction of proximal tubular cell proliferation (Rached et al., 2007).
Although routine urinalysis after repeated administration of up to 210 μg/kg bw OTA for 90 days did not indicate an increase in total protein (Rached et al., 2007), western blot analysis of urinary proteins demonstrated a small, but statistically significant increase in urinary albumin and β2-microglobulin levels in high-dose animals. Microalbuminuria together with low molecular weight proteinuria is a characteristic sign of a disrupted protein uptake into the proximal tubules (Birn and Christensen, 2006; Hart and Kinter, 2005), which has been shown to occur in OK cells treated with OTA (Gekle et al., 1994). In high-dose animals, this effect preceded changes in parameters of routine urinalysis, like NAG and serum creatinine. However, in mid-dose rats, changes in gene expression of putative biomarkers were found to be more sensitive and reliable than effects on these urinary markers.
Both Kim-1 and lipocalin-2/NGAL have been demonstrated to be excreted into urine in response to kidney injury (Mishra et al., 2004; Vaidya et al., 2006). Kim-1 detection in urine is in line with other studies, in which a correlation between induction of Kim-1 gene and protein expression and an increased urinary Kim-1 excretion had been observed (Ichimura et al., 2004; Vaidya et al., 2006; Zhou et al., 2008). Surprisingly, lipocalin-2/NGAL, which was easily identified in urine of other rat strains, was not present in urine of untreated or OTA-treated Fischer rats. Similarly, OPN and clusterin have previously been used as urinary markers of nephrotoxicity (Aulitzky et al., 1992; Eti et al., 1993; Hidaka et al., 2002; Khan et al., 2002). However, in this study, OPN was not detectable in urine of both control and treated animals, and no treatment-related changes in urinary clusterin were evident by immunoblotting.
In summary, modulation of the expression of several candidate biomarkers of nephrotoxicity by low doses of OTA in vivo is consistent with findings with other nephrotoxins and supports their use as biomarkers of kidney injury. In this model, Kim-1 was the most sensitive marker of tissue injury and mirrored well the progression of kidney damage, which is also in accordance with a recent subchronic study with cadmium chloride (Prozialeck et al., 2007). However, it should be emphasized that none of the putative biomarkers was significantly altered before the onset of histopathological changes, indicating that they may represent diagnostic, but not necessarily predictive markers of nephrotoxicity.
A second aim of this study was to investigate the potential use of these markers as predictive tools for toxicity testing in vitro, which may allow rapid screening of large numbers of compounds. Therefore, mRNA expression of Kim-1, lipocalin-2/NGAL, OPN, clusterin, vimentin, Timp-1, and HO-1 was studied in NRK-52E cells, a well-established kidney cell line for in vitro toxicity assays (Lash et al., 2002; Leussink et al., 2002; Prozialeck et al., 2006), following exposure to OTA. However, treatment resulted in only minor alterations in gene expression. Consistent with the results from the animal experiment, a slight induction of HO-1 mRNA expression was observed in cell cultured treated with OTA. In contrast, Kim-1, lipocalin-2, and Timp-1, which were the most sensitive markers for toxicity in vivo, were not significantly affected by treatment. Similar effects were also observed with other model nephrotoxins (CdCl2, KBrO3, and cisplatin). Overall, determination of cell viability appeared to be the most sensitive endpoint of toxicity in our in vitro model.
A possible explanation for the weak effects in vitro compared with the situation in vivo might be that, in contrast to the kidney, some of the marker genes were already expressed at relatively high levels in NRK cells. This is supported by the fact that expression of lipocalin-2/NGAL, vimentin, and clusterin was recently reported to increase during isolation and cultivation of proximal tubular cells (Weiland et al., 2007). It is therefore plausible to speculate that an additional increase in response to injury is no longer possible. Similarly, exposure of NRK-52E cells to OTA and other nephrotoxins did not lead to upregulation of Kim-1. Like most of the other marker genes studied here, Kim-1 is thought to be induced in the course of tissue regeneration (Ichimura et al., 1998; van Timmeren et al., 2007). Results from a number of studies suggest that, in contrast to the situation in vivo, regenerative proliferation may not occur in monolayers of cultured kidney cells in response to a continuous nephrotoxic insult, but only after removal of the toxin (Huang et al., 2001; Kamp et al., 2005; Kays and Schnellmann, 1995; Kays et al., 1993; Wiegele et al., 1998; Witzgall et al., 1994). Therefore it appears likely that the signaling pathways resulting in the expression of Kim-1 and other markers for regeneration were not turned on under the experimental conditions of this study, which were chosen to mimic progressive nephrotoxic effects.
These results are in line with a recent comparative study on gene expression in rat liver and primary hepatocytes in response to peroxisome proliferators. In this study, changes in the expression of genes related to β-oxidation were found to be induced both in vivo and in vitro. In contrast, genes associated with cell proliferation and apoptosis were exclusively altered in vivo (Tamura et al., 2006), which may be due to the lack of nonparenchymal cells, for example, Kupffer cells, which are thought to play an important role in the toxicity of peroxisome proliferators. Thus, interactions of target cells with surrounding cells, matrix components, or extracellular factors, as well as toxicokinetics, which are critical determinants in the overall response to a toxic compound, may not be adequately reflected by cell culture models. Although complete replacement of animal studies, therefore, may not be feasible, in vitro assays may still be valuable tools for rapid screening of compounds for their potential to cause nephrotoxicity. This may serve to reduce the need for animal testing, particularly within pharmaceutical industries, because potent nephrotoxins may be excluded from further development and only those candidates showing no effects in vitro may need to be tested in vivo (Prieto et al., 2006). However, although the new -omics derived markers show good potential as sensitive indicators of nephrotoxicity in vivo, results from this study do not support their use as reliable endpoints of toxicity in vitro.
FUNDING
RCC, Ltd, Itingen, Switzerland.
We would like to thank Ursula Tatsch, Theresa Ehrlich, and Elisabeth Rüb-Spiegel for excellent technical assistance.
Comments